Merge tag 'v3.16-rc5' into timers/core

Reason: Bring in upstream modifications, so the pending changes which
depend on them can be queued.

2164 files changed, 98870 insertions, 40932 deletions

diff --git a/drivers/acpi/ac.c b/drivers/acpi/ac.c
index c67f6f5ad611..36b0e61f9c09 100644
--- a/drivers/acpi/ac.c
+++ b/drivers/acpi/ac.c
@@ -30,6 +30,10 @@
 #include <linux/types.h>
 #include <linux/dmi.h>
 #include <linux/delay.h>
+#ifdef CONFIG_ACPI_PROCFS_POWER
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#endif
 #include <linux/platform_device.h>
 #include <linux/power_supply.h>
 #include <linux/acpi.h>
@@ -52,6 +56,7 @@ MODULE_AUTHOR("Paul Diefenbaugh");
 MODULE_DESCRIPTION("ACPI AC Adapter Driver");
 MODULE_LICENSE("GPL");
 
+
 static int acpi_ac_add(struct acpi_device *device);
 static int acpi_ac_remove(struct acpi_device *device);
 static void acpi_ac_notify(struct acpi_device *device, u32 event);
@@ -67,6 +72,13 @@ static int acpi_ac_resume(struct device *dev);
 #endif
 static SIMPLE_DEV_PM_OPS(acpi_ac_pm, NULL, acpi_ac_resume);
 
+#ifdef CONFIG_ACPI_PROCFS_POWER
+extern struct proc_dir_entry *acpi_lock_ac_dir(void);
+extern void *acpi_unlock_ac_dir(struct proc_dir_entry *acpi_ac_dir);
+static int acpi_ac_open_fs(struct inode *inode, struct file *file);
+#endif
+
+
 static int ac_sleep_before_get_state_ms;
 
 static struct acpi_driver acpi_ac_driver = {
@@ -91,6 +103,16 @@ struct acpi_ac {
 
 #define to_acpi_ac(x) container_of(x, struct acpi_ac, charger)
 
+#ifdef CONFIG_ACPI_PROCFS_POWER
+static const struct file_operations acpi_ac_fops = {
+	.owner = THIS_MODULE,
+	.open = acpi_ac_open_fs,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+#endif
+
 /* --------------------------------------------------------------------------
                                AC Adapter Management
    -------------------------------------------------------------------------- */
@@ -143,6 +165,83 @@ static enum power_supply_property ac_props[] = {
 	POWER_SUPPLY_PROP_ONLINE,
 };
 
+#ifdef CONFIG_ACPI_PROCFS_POWER
+/* --------------------------------------------------------------------------
+                              FS Interface (/proc)
+   -------------------------------------------------------------------------- */
+
+static struct proc_dir_entry *acpi_ac_dir;
+
+static int acpi_ac_seq_show(struct seq_file *seq, void *offset)
+{
+	struct acpi_ac *ac = seq->private;
+
+
+	if (!ac)
+		return 0;
+
+	if (acpi_ac_get_state(ac)) {
+		seq_puts(seq, "ERROR: Unable to read AC Adapter state\n");
+		return 0;
+	}
+
+	seq_puts(seq, "state:                   ");
+	switch (ac->state) {
+	case ACPI_AC_STATUS_OFFLINE:
+		seq_puts(seq, "off-line\n");
+		break;
+	case ACPI_AC_STATUS_ONLINE:
+		seq_puts(seq, "on-line\n");
+		break;
+	default:
+		seq_puts(seq, "unknown\n");
+		break;
+	}
+
+	return 0;
+}
+
+static int acpi_ac_open_fs(struct inode *inode, struct file *file)
+{
+	return single_open(file, acpi_ac_seq_show, PDE_DATA(inode));
+}
+
+static int acpi_ac_add_fs(struct acpi_ac *ac)
+{
+	struct proc_dir_entry *entry = NULL;
+
+	printk(KERN_WARNING PREFIX "Deprecated procfs I/F for AC is loaded,"
+			" please retry with CONFIG_ACPI_PROCFS_POWER cleared\n");
+	if (!acpi_device_dir(ac->device)) {
+		acpi_device_dir(ac->device) =
+			proc_mkdir(acpi_device_bid(ac->device), acpi_ac_dir);
+		if (!acpi_device_dir(ac->device))
+			return -ENODEV;
+	}
+
+	/* 'state' [R] */
+	entry = proc_create_data(ACPI_AC_FILE_STATE,
+				 S_IRUGO, acpi_device_dir(ac->device),
+				 &acpi_ac_fops, ac);
+	if (!entry)
+		return -ENODEV;
+	return 0;
+}
+
+static int acpi_ac_remove_fs(struct acpi_ac *ac)
+{
+
+	if (acpi_device_dir(ac->device)) {
+		remove_proc_entry(ACPI_AC_FILE_STATE,
+				  acpi_device_dir(ac->device));
+		remove_proc_entry(acpi_device_bid(ac->device), acpi_ac_dir);
+		acpi_device_dir(ac->device) = NULL;
+	}
+
+	return 0;
+}
+#endif
+
 /* --------------------------------------------------------------------------
                                    Driver Model
    -------------------------------------------------------------------------- */
@@ -243,6 +342,11 @@ static int acpi_ac_add(struct acpi_device *device)
 		goto end;
 
 	ac->charger.name = acpi_device_bid(device);
+#ifdef CONFIG_ACPI_PROCFS_POWER
+	result = acpi_ac_add_fs(ac);
+	if (result)
+		goto end;
+#endif
 	ac->charger.type = POWER_SUPPLY_TYPE_MAINS;
 	ac->charger.properties = ac_props;
 	ac->charger.num_properties = ARRAY_SIZE(ac_props);
@@ -258,8 +362,12 @@ static int acpi_ac_add(struct acpi_device *device)
 	ac->battery_nb.notifier_call = acpi_ac_battery_notify;
 	register_acpi_notifier(&ac->battery_nb);
 end:
-	if (result)
+	if (result) {
+#ifdef CONFIG_ACPI_PROCFS_POWER
+		acpi_ac_remove_fs(ac);
+#endif
 		kfree(ac);
+	}
 
 	dmi_check_system(ac_dmi_table);
 	return result;
@@ -303,6 +411,10 @@ static int acpi_ac_remove(struct acpi_device *device)
 		power_supply_unregister(&ac->charger);
 	unregister_acpi_notifier(&ac->battery_nb);
 
+#ifdef CONFIG_ACPI_PROCFS_POWER
+	acpi_ac_remove_fs(ac);
+#endif
+
 	kfree(ac);
 
 	return 0;
@@ -315,9 +427,20 @@ static int __init acpi_ac_init(void)
 	if (acpi_disabled)
 		return -ENODEV;
 
+#ifdef CONFIG_ACPI_PROCFS_POWER
+	acpi_ac_dir = acpi_lock_ac_dir();
+	if (!acpi_ac_dir)
+		return -ENODEV;
+#endif
+
+
 	result = acpi_bus_register_driver(&acpi_ac_driver);
-	if (result < 0)
+	if (result < 0) {
+#ifdef CONFIG_ACPI_PROCFS_POWER
+		acpi_unlock_ac_dir(acpi_ac_dir);
+#endif
 		return -ENODEV;
+	}
 
 	return 0;
 }
@@ -325,6 +448,9 @@ static int __init acpi_ac_init(void)
 static void __exit acpi_ac_exit(void)
 {
 	acpi_bus_unregister_driver(&acpi_ac_driver);
+#ifdef CONFIG_ACPI_PROCFS_POWER
+	acpi_unlock_ac_dir(acpi_ac_dir);
+#endif
 }
 module_init(acpi_ac_init);
 module_exit(acpi_ac_exit);
diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c
index 63407d264885..9cb65b0e7597 100644
--- a/drivers/acpi/acpi_lpss.c
+++ b/drivers/acpi/acpi_lpss.c
@@ -34,6 +34,9 @@ ACPI_MODULE_NAME("acpi_lpss");
 
 /* Offsets relative to LPSS_PRIVATE_OFFSET */
 #define LPSS_CLK_DIVIDER_DEF_MASK	(BIT(1) | BIT(16))
+#define LPSS_RESETS			0x04
+#define LPSS_RESETS_RESET_FUNC		BIT(0)
+#define LPSS_RESETS_RESET_APB		BIT(1)
 #define LPSS_GENERAL			0x08
 #define LPSS_GENERAL_LTR_MODE_SW	BIT(2)
 #define LPSS_GENERAL_UART_RTS_OVRD	BIT(3)
@@ -99,6 +102,17 @@ static void lpss_uart_setup(struct lpss_private_data *pdata)
 	writel(reg | LPSS_GENERAL_UART_RTS_OVRD, pdata->mmio_base + offset);
 }
 
+static void lpss_i2c_setup(struct lpss_private_data *pdata)
+{
+	unsigned int offset;
+	u32 val;
+
+	offset = pdata->dev_desc->prv_offset + LPSS_RESETS;
+	val = readl(pdata->mmio_base + offset);
+	val |= LPSS_RESETS_RESET_APB | LPSS_RESETS_RESET_FUNC;
+	writel(val, pdata->mmio_base + offset);
+}
+
 static struct lpss_device_desc lpt_dev_desc = {
 	.clk_required = true,
 	.prv_offset = 0x800,
@@ -171,6 +185,7 @@ static struct lpss_device_desc byt_i2c_dev_desc = {
 	.prv_offset = 0x800,
 	.save_ctx = true,
 	.shared_clock = &i2c_clock,
+	.setup = lpss_i2c_setup,
 };
 
 #else
diff --git a/drivers/acpi/acpi_pnp.c b/drivers/acpi/acpi_pnp.c
index 6703c1fd993a..4ddb0dca56f6 100644
--- a/drivers/acpi/acpi_pnp.c
+++ b/drivers/acpi/acpi_pnp.c
@@ -14,6 +14,8 @@
 #include <linux/module.h>
 
 static const struct acpi_device_id acpi_pnp_device_ids[] = {
+	/* soc_button_array */
+	{"PNP0C40"},
 	/* pata_isapnp */
 	{"PNP0600"},		/* Generic ESDI/IDE/ATA compatible hard disk controller */
 	/* floppy */
diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c
index e48fc98e71c4..130f513e08c9 100644
--- a/drivers/acpi/battery.c
+++ b/drivers/acpi/battery.c
@@ -32,8 +32,10 @@
 #include <linux/jiffies.h>
 #include <linux/async.h>
 #include <linux/dmi.h>
+#include <linux/delay.h>
 #include <linux/slab.h>
 #include <linux/suspend.h>
+#include <linux/delay.h>
 #include <asm/unaligned.h>
 
 #ifdef CONFIG_ACPI_PROCFS_POWER
@@ -70,6 +72,7 @@ MODULE_DESCRIPTION("ACPI Battery Driver");
 MODULE_LICENSE("GPL");
 
 static int battery_bix_broken_package;
+static int battery_notification_delay_ms;
 static unsigned int cache_time = 1000;
 module_param(cache_time, uint, 0644);
 MODULE_PARM_DESC(cache_time, "cache time in milliseconds");
@@ -532,6 +535,20 @@ static int acpi_battery_get_state(struct acpi_battery *battery)
 			" invalid.\n");
 	}
 
+	/*
+	 * When fully charged, some batteries wrongly report
+	 * capacity_now = design_capacity instead of = full_charge_capacity
+	 */
+	if (battery->capacity_now > battery->full_charge_capacity
+	    && battery->full_charge_capacity != ACPI_BATTERY_VALUE_UNKNOWN) {
+		battery->capacity_now = battery->full_charge_capacity;
+		if (battery->capacity_now != battery->design_capacity)
+			printk_once(KERN_WARNING FW_BUG
+				"battery: reported current charge level (%d) "
+				"is higher than reported maximum charge level (%d).\n",
+				battery->capacity_now, battery->full_charge_capacity);
+	}
+
 	if (test_bit(ACPI_BATTERY_QUIRK_PERCENTAGE_CAPACITY, &battery->flags)
 	    && battery->capacity_now >= 0 && battery->capacity_now <= 100)
 		battery->capacity_now = (battery->capacity_now *
@@ -930,7 +947,10 @@ static ssize_t acpi_battery_write_alarm(struct file *file,
 		goto end;
 	}
 	alarm_string[count] = '\0';
-	battery->alarm = simple_strtol(alarm_string, NULL, 0);
+	if (kstrtoint(alarm_string, 0, &battery->alarm)) {
+		result = -EINVAL;
+		goto end;
+	}
 	result = acpi_battery_set_alarm(battery);
       end:
 	if (!result)
@@ -1062,6 +1082,14 @@ static void acpi_battery_notify(struct acpi_device *device, u32 event)
 	if (!battery)
 		return;
 	old = battery->bat.dev;
+	/*
+	* On Acer Aspire V5-573G notifications are sometimes triggered too
+	* early. For example, when AC is unplugged and notification is
+	* triggered, battery state is still reported as "Full", and changes to
+	* "Discharging" only after short delay, without any notification.
+	*/
+	if (battery_notification_delay_ms > 0)
+		msleep(battery_notification_delay_ms);
 	if (event == ACPI_BATTERY_NOTIFY_INFO)
 		acpi_battery_refresh(battery);
 	acpi_battery_update(battery, false);
@@ -1106,17 +1134,60 @@ static int battery_notify(struct notifier_block *nb,
 	return 0;
 }
 
+static int battery_bix_broken_package_quirk(const struct dmi_system_id *d)
+{
+	battery_bix_broken_package = 1;
+	return 0;
+}
+
+static int battery_notification_delay_quirk(const struct dmi_system_id *d)
+{
+	battery_notification_delay_ms = 1000;
+	return 0;
+}
+
 static struct dmi_system_id bat_dmi_table[] = {
 	{
+		.callback = battery_bix_broken_package_quirk,
 		.ident = "NEC LZ750/LS",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "NEC"),
 			DMI_MATCH(DMI_PRODUCT_NAME, "PC-LZ750LS"),
 		},
 	},
+	{
+		.callback = battery_notification_delay_quirk,
+		.ident = "Acer Aspire V5-573G",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Aspire V5-573G"),
+		},
+	},
 	{},
 };
 
+/*
+ * Some machines'(E,G Lenovo Z480) ECs are not stable
+ * during boot up and this causes battery driver fails to be
+ * probed due to failure of getting battery information
+ * from EC sometimes. After several retries, the operation
+ * may work. So add retry code here and 20ms sleep between
+ * every retries.
+ */
+static int acpi_battery_update_retry(struct acpi_battery *battery)
+{
+	int retry, ret;
+
+	for (retry = 5; retry; retry--) {
+		ret = acpi_battery_update(battery, false);
+		if (!ret)
+			break;
+
+		msleep(20);
+	}
+	return ret;
+}
+
 static int acpi_battery_add(struct acpi_device *device)
 {
 	int result = 0;
@@ -1135,9 +1206,11 @@ static int acpi_battery_add(struct acpi_device *device)
 	mutex_init(&battery->sysfs_lock);
 	if (acpi_has_method(battery->device->handle, "_BIX"))
 		set_bit(ACPI_BATTERY_XINFO_PRESENT, &battery->flags);
-	result = acpi_battery_update(battery, false);
+
+	result = acpi_battery_update_retry(battery);
 	if (result)
 		goto fail;
+
 #ifdef CONFIG_ACPI_PROCFS_POWER
 	result = acpi_battery_add_fs(device);
 #endif
@@ -1227,8 +1300,7 @@ static void __init acpi_battery_init_async(void *unused, async_cookie_t cookie)
 	if (acpi_disabled)
 		return;
 
-	if (dmi_check_system(bat_dmi_table))
-		battery_bix_broken_package = 1;
+	dmi_check_system(bat_dmi_table);
 	
 #ifdef CONFIG_ACPI_PROCFS_POWER
 	acpi_battery_dir = acpi_lock_battery_dir();
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
index ad11ba4a412d..a66ab658abbc 100644
--- a/drivers/acpi/ec.c
+++ b/drivers/acpi/ec.c
@@ -1,11 +1,14 @@
 /*
- *  ec.c - ACPI Embedded Controller Driver (v2.1)
+ *  ec.c - ACPI Embedded Controller Driver (v2.2)
  *
- *  Copyright (C) 2006-2008 Alexey Starikovskiy <astarikovskiy@suse.de>
- *  Copyright (C) 2006 Denis Sadykov <denis.m.sadykov@intel.com>
- *  Copyright (C) 2004 Luming Yu <luming.yu@intel.com>
- *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
- *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *  Copyright (C) 2001-2014 Intel Corporation
+ *    Author: 2014       Lv Zheng <lv.zheng@intel.com>
+ *            2006, 2007 Alexey Starikovskiy <alexey.y.starikovskiy@intel.com>
+ *            2006       Denis Sadykov <denis.m.sadykov@intel.com>
+ *            2004       Luming Yu <luming.yu@intel.com>
+ *            2001, 2002 Andy Grover <andrew.grover@intel.com>
+ *            2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *  Copyright (C) 2008      Alexey Starikovskiy <astarikovskiy@suse.de>
  *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
@@ -52,6 +55,7 @@
 /* EC status register */
 #define ACPI_EC_FLAG_OBF	0x01	/* Output buffer full */
 #define ACPI_EC_FLAG_IBF	0x02	/* Input buffer full */
+#define ACPI_EC_FLAG_CMD	0x08	/* Input buffer contains a command */
 #define ACPI_EC_FLAG_BURST	0x10	/* burst mode */
 #define ACPI_EC_FLAG_SCI	0x20	/* EC-SCI occurred */
 
@@ -78,6 +82,9 @@ enum {
 	EC_FLAGS_BLOCKED,		/* Transactions are blocked */
 };
 
+#define ACPI_EC_COMMAND_POLL		0x01 /* Available for command byte */
+#define ACPI_EC_COMMAND_COMPLETE	0x02 /* Completed last byte */
+
 /* ec.c is compiled in acpi namespace so this shows up as acpi.ec_delay param */
 static unsigned int ec_delay __read_mostly = ACPI_EC_DELAY;
 module_param(ec_delay, uint, 0644);
@@ -109,7 +116,7 @@ struct transaction {
 	u8 ri;
 	u8 wlen;
 	u8 rlen;
-	bool done;
+	u8 flags;
 };
 
 struct acpi_ec *boot_ec, *first_ec;
@@ -127,83 +134,104 @@ static int EC_FLAGS_CLEAR_ON_RESUME; /* Needs acpi_ec_clear() on boot/resume */
 static inline u8 acpi_ec_read_status(struct acpi_ec *ec)
 {
 	u8 x = inb(ec->command_addr);
-	pr_debug("---> status = 0x%2.2x\n", x);
+	pr_debug("EC_SC(R) = 0x%2.2x "
+		 "SCI_EVT=%d BURST=%d CMD=%d IBF=%d OBF=%d\n",
+		 x,
+		 !!(x & ACPI_EC_FLAG_SCI),
+		 !!(x & ACPI_EC_FLAG_BURST),
+		 !!(x & ACPI_EC_FLAG_CMD),
+		 !!(x & ACPI_EC_FLAG_IBF),
+		 !!(x & ACPI_EC_FLAG_OBF));
 	return x;
 }
 
 static inline u8 acpi_ec_read_data(struct acpi_ec *ec)
 {
 	u8 x = inb(ec->data_addr);
-	pr_debug("---> data = 0x%2.2x\n", x);
+	pr_debug("EC_DATA(R) = 0x%2.2x\n", x);
 	return x;
 }
 
 static inline void acpi_ec_write_cmd(struct acpi_ec *ec, u8 command)
 {
-	pr_debug("<--- command = 0x%2.2x\n", command);
+	pr_debug("EC_SC(W) = 0x%2.2x\n", command);
 	outb(command, ec->command_addr);
 }
 
 static inline void acpi_ec_write_data(struct acpi_ec *ec, u8 data)
 {
-	pr_debug("<--- data = 0x%2.2x\n", data);
+	pr_debug("EC_DATA(W) = 0x%2.2x\n", data);
 	outb(data, ec->data_addr);
 }
 
-static int ec_transaction_done(struct acpi_ec *ec)
+static int ec_transaction_completed(struct acpi_ec *ec)
 {
 	unsigned long flags;
 	int ret = 0;
 	spin_lock_irqsave(&ec->lock, flags);
-	if (!ec->curr || ec->curr->done)
+	if (ec->curr && (ec->curr->flags & ACPI_EC_COMMAND_COMPLETE))
 		ret = 1;
 	spin_unlock_irqrestore(&ec->lock, flags);
 	return ret;
 }
 
-static void start_transaction(struct acpi_ec *ec)
+static bool advance_transaction(struct acpi_ec *ec)
 {
-	ec->curr->irq_count = ec->curr->wi = ec->curr->ri = 0;
-	ec->curr->done = false;
-	acpi_ec_write_cmd(ec, ec->curr->command);
-}
-
-static void advance_transaction(struct acpi_ec *ec, u8 status)
-{
-	unsigned long flags;
 	struct transaction *t;
+	u8 status;
+	bool wakeup = false;
 
-	spin_lock_irqsave(&ec->lock, flags);
+	pr_debug("===== %s =====\n", in_interrupt() ? "IRQ" : "TASK");
+	status = acpi_ec_read_status(ec);
 	t = ec->curr;
 	if (!t)
-		goto unlock;
-	if (t->wlen > t->wi) {
-		if ((status & ACPI_EC_FLAG_IBF) == 0)
-			acpi_ec_write_data(ec,
-				t->wdata[t->wi++]);
-		else
-			goto err;
-	} else if (t->rlen > t->ri) {
-		if ((status & ACPI_EC_FLAG_OBF) == 1) {
-			t->rdata[t->ri++] = acpi_ec_read_data(ec);
-			if (t->rlen == t->ri)
-				t->done = true;
+		goto err;
+	if (t->flags & ACPI_EC_COMMAND_POLL) {
+		if (t->wlen > t->wi) {
+			if ((status & ACPI_EC_FLAG_IBF) == 0)
+				acpi_ec_write_data(ec, t->wdata[t->wi++]);
+			else
+				goto err;
+		} else if (t->rlen > t->ri) {
+			if ((status & ACPI_EC_FLAG_OBF) == 1) {
+				t->rdata[t->ri++] = acpi_ec_read_data(ec);
+				if (t->rlen == t->ri) {
+					t->flags |= ACPI_EC_COMMAND_COMPLETE;
+					wakeup = true;
+				}
+			} else
+				goto err;
+		} else if (t->wlen == t->wi &&
+			   (status & ACPI_EC_FLAG_IBF) == 0) {
+			t->flags |= ACPI_EC_COMMAND_COMPLETE;
+			wakeup = true;
+		}
+		return wakeup;
+	} else {
+		if ((status & ACPI_EC_FLAG_IBF) == 0) {
+			acpi_ec_write_cmd(ec, t->command);
+			t->flags |= ACPI_EC_COMMAND_POLL;
 		} else
 			goto err;
-	} else if (t->wlen == t->wi &&
-		   (status & ACPI_EC_FLAG_IBF) == 0)
-		t->done = true;
-	goto unlock;
+		return wakeup;
+	}
 err:
 	/*
 	 * If SCI bit is set, then don't think it's a false IRQ
 	 * otherwise will take a not handled IRQ as a false one.
 	 */
-	if (in_interrupt() && !(status & ACPI_EC_FLAG_SCI))
-		++t->irq_count;
+	if (!(status & ACPI_EC_FLAG_SCI)) {
+		if (in_interrupt() && t)
+			++t->irq_count;
+	}
+	return wakeup;
+}
 
-unlock:
-	spin_unlock_irqrestore(&ec->lock, flags);
+static void start_transaction(struct acpi_ec *ec)
+{
+	ec->curr->irq_count = ec->curr->wi = ec->curr->ri = 0;
+	ec->curr->flags = 0;
+	(void)advance_transaction(ec);
 }
 
 static int acpi_ec_sync_query(struct acpi_ec *ec, u8 *data);
@@ -228,15 +256,17 @@ static int ec_poll(struct acpi_ec *ec)
 			/* don't sleep with disabled interrupts */
 			if (EC_FLAGS_MSI || irqs_disabled()) {
 				udelay(ACPI_EC_MSI_UDELAY);
-				if (ec_transaction_done(ec))
+				if (ec_transaction_completed(ec))
 					return 0;
 			} else {
 				if (wait_event_timeout(ec->wait,
-						ec_transaction_done(ec),
+						ec_transaction_completed(ec),
 						msecs_to_jiffies(1)))
 					return 0;
 			}
-			advance_transaction(ec, acpi_ec_read_status(ec));
+			spin_lock_irqsave(&ec->lock, flags);
+			(void)advance_transaction(ec);
+			spin_unlock_irqrestore(&ec->lock, flags);
 		} while (time_before(jiffies, delay));
 		pr_debug("controller reset, restart transaction\n");
 		spin_lock_irqsave(&ec->lock, flags);
@@ -268,23 +298,6 @@ static int acpi_ec_transaction_unlocked(struct acpi_ec *ec,
 	return ret;
 }
 
-static int ec_check_ibf0(struct acpi_ec *ec)
-{
-	u8 status = acpi_ec_read_status(ec);
-	return (status & ACPI_EC_FLAG_IBF) == 0;
-}
-
-static int ec_wait_ibf0(struct acpi_ec *ec)
-{
-	unsigned long delay = jiffies + msecs_to_jiffies(ec_delay);
-	/* interrupt wait manually if GPE mode is not active */
-	while (time_before(jiffies, delay))
-		if (wait_event_timeout(ec->wait, ec_check_ibf0(ec),
-					msecs_to_jiffies(1)))
-			return 0;
-	return -ETIME;
-}
-
 static int acpi_ec_transaction(struct acpi_ec *ec, struct transaction *t)
 {
 	int status;
@@ -305,12 +318,6 @@ static int acpi_ec_transaction(struct acpi_ec *ec, struct transaction *t)
 			goto unlock;
 		}
 	}
-	if (ec_wait_ibf0(ec)) {
-		pr_err("input buffer is not empty, "
-				"aborting transaction\n");
-		status = -ETIME;
-		goto end;
-	}
 	pr_debug("transaction start (cmd=0x%02x, addr=0x%02x)\n",
 			t->command, t->wdata ? t->wdata[0] : 0);
 	/* disable GPE during transaction if storm is detected */
@@ -334,7 +341,6 @@ static int acpi_ec_transaction(struct acpi_ec *ec, struct transaction *t)
 		set_bit(EC_FLAGS_GPE_STORM, &ec->flags);
 	}
 	pr_debug("transaction end\n");
-end:
 	if (ec->global_lock)
 		acpi_release_global_lock(glk);
 unlock:
@@ -634,17 +640,14 @@ static int ec_check_sci(struct acpi_ec *ec, u8 state)
 static u32 acpi_ec_gpe_handler(acpi_handle gpe_device,
 	u32 gpe_number, void *data)
 {
+	unsigned long flags;
 	struct acpi_ec *ec = data;
-	u8 status = acpi_ec_read_status(ec);
 
-	pr_debug("~~~> interrupt, status:0x%02x\n", status);
-
-	advance_transaction(ec, status);
-	if (ec_transaction_done(ec) &&
-	    (acpi_ec_read_status(ec) & ACPI_EC_FLAG_IBF) == 0) {
+	spin_lock_irqsave(&ec->lock, flags);
+	if (advance_transaction(ec))
 		wake_up(&ec->wait);
-		ec_check_sci(ec, acpi_ec_read_status(ec));
-	}
+	spin_unlock_irqrestore(&ec->lock, flags);
+	ec_check_sci(ec, acpi_ec_read_status(ec));
 	return ACPI_INTERRUPT_HANDLED | ACPI_REENABLE_GPE;
 }
 
@@ -1066,8 +1069,10 @@ int __init acpi_ec_ecdt_probe(void)
 	/* fall through */
 	}
 
-	if (EC_FLAGS_SKIP_DSDT_SCAN)
+	if (EC_FLAGS_SKIP_DSDT_SCAN) {
+		kfree(saved_ec);
 		return -ENODEV;
+	}
 
 	/* This workaround is needed only on some broken machines,
 	 * which require early EC, but fail to provide ECDT */
@@ -1105,6 +1110,7 @@ install:
 	}
 error:
 	kfree(boot_ec);
+	kfree(saved_ec);
 	boot_ec = NULL;
 	return -ENODEV;
 }
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index 147bc1b91b42..bad25b070fe0 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -235,7 +235,8 @@ void acpi_os_vprintf(const char *fmt, va_list args)
 static unsigned long acpi_rsdp;
 static int __init setup_acpi_rsdp(char *arg)
 {
-	acpi_rsdp = simple_strtoul(arg, NULL, 16);
+	if (kstrtoul(arg, 16, &acpi_rsdp))
+		return -EINVAL;
 	return 0;
 }
 early_param("acpi_rsdp", setup_acpi_rsdp);
@@ -1810,6 +1811,16 @@ acpi_status __init acpi_os_initialize(void)
 	acpi_os_map_generic_address(&acpi_gbl_FADT.xpm1b_event_block);
 	acpi_os_map_generic_address(&acpi_gbl_FADT.xgpe0_block);
 	acpi_os_map_generic_address(&acpi_gbl_FADT.xgpe1_block);
+	if (acpi_gbl_FADT.flags & ACPI_FADT_RESET_REGISTER) {
+		/*
+		 * Use acpi_os_map_generic_address to pre-map the reset
+		 * register if it's in system memory.
+		 */
+		int rv;
+
+		rv = acpi_os_map_generic_address(&acpi_gbl_FADT.reset_register);
+		pr_debug(PREFIX "%s: map reset_reg status %d\n", __func__, rv);
+	}
 
 	return AE_OK;
 }
@@ -1838,6 +1849,8 @@ acpi_status acpi_os_terminate(void)
 	acpi_os_unmap_generic_address(&acpi_gbl_FADT.xgpe0_block);
 	acpi_os_unmap_generic_address(&acpi_gbl_FADT.xpm1b_event_block);
 	acpi_os_unmap_generic_address(&acpi_gbl_FADT.xpm1a_event_block);
+	if (acpi_gbl_FADT.flags & ACPI_FADT_RESET_REGISTER)
+		acpi_os_unmap_generic_address(&acpi_gbl_FADT.reset_register);
 
 	destroy_workqueue(kacpid_wq);
 	destroy_workqueue(kacpi_notify_wq);
diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c
index 0bdacc5e26a3..2ba8f02ced36 100644
--- a/drivers/acpi/resource.c
+++ b/drivers/acpi/resource.c
@@ -77,7 +77,7 @@ bool acpi_dev_resource_memory(struct acpi_resource *ares, struct resource *res)
 	switch (ares->type) {
 	case ACPI_RESOURCE_TYPE_MEMORY24:
 		memory24 = &ares->data.memory24;
-		if (!memory24->address_length)
+		if (!memory24->minimum && !memory24->address_length)
 			return false;
 		acpi_dev_get_memresource(res, memory24->minimum,
 					 memory24->address_length,
@@ -85,7 +85,7 @@ bool acpi_dev_resource_memory(struct acpi_resource *ares, struct resource *res)
 		break;
 	case ACPI_RESOURCE_TYPE_MEMORY32:
 		memory32 = &ares->data.memory32;
-		if (!memory32->address_length)
+		if (!memory32->minimum && !memory32->address_length)
 			return false;
 		acpi_dev_get_memresource(res, memory32->minimum,
 					 memory32->address_length,
@@ -93,7 +93,7 @@ bool acpi_dev_resource_memory(struct acpi_resource *ares, struct resource *res)
 		break;
 	case ACPI_RESOURCE_TYPE_FIXED_MEMORY32:
 		fixed_memory32 = &ares->data.fixed_memory32;
-		if (!fixed_memory32->address_length)
+		if (!fixed_memory32->address && !fixed_memory32->address_length)
 			return false;
 		acpi_dev_get_memresource(res, fixed_memory32->address,
 					 fixed_memory32->address_length,
@@ -150,7 +150,7 @@ bool acpi_dev_resource_io(struct acpi_resource *ares, struct resource *res)
 	switch (ares->type) {
 	case ACPI_RESOURCE_TYPE_IO:
 		io = &ares->data.io;
-		if (!io->address_length)
+		if (!io->minimum && !io->address_length)
 			return false;
 		acpi_dev_get_ioresource(res, io->minimum,
 					io->address_length,
@@ -158,7 +158,7 @@ bool acpi_dev_resource_io(struct acpi_resource *ares, struct resource *res)
 		break;
 	case ACPI_RESOURCE_TYPE_FIXED_IO:
 		fixed_io = &ares->data.fixed_io;
-		if (!fixed_io->address_length)
+		if (!fixed_io->address && !fixed_io->address_length)
 			return false;
 		acpi_dev_get_ioresource(res, fixed_io->address,
 					fixed_io->address_length,
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index c11e3795431b..b3e3cc73ba79 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -19,6 +19,7 @@
 #include <linux/acpi.h>
 #include <linux/module.h>
 #include <asm/io.h>
+#include <trace/events/power.h>
 
 #include "internal.h"
 #include "sleep.h"
@@ -501,6 +502,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
 
 	ACPI_FLUSH_CPU_CACHE();
 
+	trace_suspend_resume(TPS("acpi_suspend"), acpi_state, true);
 	switch (acpi_state) {
 	case ACPI_STATE_S1:
 		barrier();
@@ -516,6 +518,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
 		pr_info(PREFIX "Low-level resume complete\n");
 		break;
 	}
+	trace_suspend_resume(TPS("acpi_suspend"), acpi_state, false);
 
 	/* This violates the spec but is required for bug compatibility. */
 	acpi_write_bit_register(ACPI_BITREG_SCI_ENABLE, 1);
diff --git a/drivers/acpi/tables.c b/drivers/acpi/tables.c
index 05550ba44d32..6d5a6cda0734 100644
--- a/drivers/acpi/tables.c
+++ b/drivers/acpi/tables.c
@@ -360,7 +360,8 @@ static int __init acpi_parse_apic_instance(char *str)
 	if (!str)
 		return -EINVAL;
 
-	acpi_apic_instance = simple_strtoul(str, NULL, 0);
+	if (kstrtoint(str, 0, &acpi_apic_instance))
+		return -EINVAL;
 
 	pr_notice("Shall use APIC/MADT table %d\n", acpi_apic_instance);
 
diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 101fb090dcb9..071c1dfb93f3 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -82,7 +82,7 @@ module_param(allow_duplicates, bool, 0644);
  * For Windows 8 systems: used to decide if video module
  * should skip registering backlight interface of its own.
  */
-static int use_native_backlight_param = -1;
+static int use_native_backlight_param = 1;
 module_param_named(use_native_backlight, use_native_backlight_param, int, 0444);
 static bool use_native_backlight_dmi = false;
 
@@ -241,13 +241,14 @@ static bool acpi_video_use_native_backlight(void)
 		return use_native_backlight_dmi;
 }
 
-static bool acpi_video_verify_backlight_support(void)
+bool acpi_video_verify_backlight_support(void)
 {
 	if (acpi_osi_is_win8() && acpi_video_use_native_backlight() &&
 	    backlight_device_registered(BACKLIGHT_RAW))
 		return false;
 	return acpi_video_backlight_support();
 }
+EXPORT_SYMBOL_GPL(acpi_video_verify_backlight_support);
 
 /* backlight device sysfs support */
 static int acpi_video_get_brightness(struct backlight_device *bd)
@@ -563,6 +564,14 @@ static struct dmi_system_id video_dmi_table[] __initdata = {
 		},
 	},
 	{
+	 .callback = video_set_use_native_backlight,
+	 .ident = "Acer TravelMate B113",
+	 .matches = {
+		DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
+		DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate B113"),
+		},
+	},
+	{
 	.callback = video_set_use_native_backlight,
 	.ident = "HP ProBook 4340s",
 	.matches = {
diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c
index 33e3db548a29..c42feb2bacd0 100644
--- a/drivers/acpi/video_detect.c
+++ b/drivers/acpi/video_detect.c
@@ -166,6 +166,14 @@ static struct dmi_system_id video_detect_dmi_table[] = {
 		DMI_MATCH(DMI_PRODUCT_NAME, "UL30A"),
 		},
 	},
+	{
+	.callback = video_detect_force_vendor,
+	.ident = "Dell Inspiron 5737",
+	.matches = {
+		DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+		DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 5737"),
+		},
+	},
 	{ },
 };
 
diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index 0033fafc470b..7671dbac6015 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -123,6 +123,15 @@ config AHCI_IMX
 
 	  If unsure, say N.
 
+config AHCI_MVEBU
+	tristate "Marvell EBU AHCI SATA support"
+	depends on ARCH_MVEBU
+	help
+	  This option enables support for the Marvebu EBU SoC's
+	  onboard AHCI SATA.
+
+	  If unsure, say N.
+
 config AHCI_SUNXI
 	tristate "Allwinner sunxi AHCI SATA support"
 	depends on ARCH_SUNXI
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index 44c8016e565c..5a02aeecef5b 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -12,6 +12,7 @@ obj-$(CONFIG_SATA_DWC)		+= sata_dwc_460ex.o
 obj-$(CONFIG_SATA_HIGHBANK)	+= sata_highbank.o libahci.o
 obj-$(CONFIG_AHCI_DA850)	+= ahci_da850.o libahci.o libahci_platform.o
 obj-$(CONFIG_AHCI_IMX)		+= ahci_imx.o libahci.o libahci_platform.o
+obj-$(CONFIG_AHCI_MVEBU)	+= ahci_mvebu.o libahci.o libahci_platform.o
 obj-$(CONFIG_AHCI_SUNXI)	+= ahci_sunxi.o libahci.o libahci_platform.o
 obj-$(CONFIG_AHCI_ST)		+= ahci_st.o libahci.o libahci_platform.o
 obj-$(CONFIG_AHCI_XGENE)	+= ahci_xgene.o libahci.o libahci_platform.o
diff --git a/drivers/ata/acard-ahci.c b/drivers/ata/acard-ahci.c
index b51605ac5974..0cd7c7a39e5b 100644
--- a/drivers/ata/acard-ahci.c
+++ b/drivers/ata/acard-ahci.c
@@ -77,7 +77,7 @@ static bool acard_ahci_qc_fill_rtf(struct ata_queued_cmd *qc);
 static int acard_ahci_port_start(struct ata_port *ap);
 static int acard_ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int acard_ahci_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
 static int acard_ahci_pci_device_resume(struct pci_dev *pdev);
 #endif
@@ -118,13 +118,13 @@ static struct pci_driver acard_ahci_pci_driver = {
 	.id_table		= acard_ahci_pci_tbl,
 	.probe			= acard_ahci_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= acard_ahci_pci_device_suspend,
 	.resume			= acard_ahci_pci_device_resume,
 #endif
 };
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int acard_ahci_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 60707814a84b..dae5607e1115 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -445,10 +445,14 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	  .driver_data = board_ahci_yes_fbs },			/* 88se9172 */
 	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x9192),
 	  .driver_data = board_ahci_yes_fbs },			/* 88se9172 on some Gigabyte */
+	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x91a0),
+	  .driver_data = board_ahci_yes_fbs },
 	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x91a3),
 	  .driver_data = board_ahci_yes_fbs },
 	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, 0x9230),
 	  .driver_data = board_ahci_yes_fbs },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, 0x0642),
+	  .driver_data = board_ahci_yes_fbs },
 
 	/* Promise */
 	{ PCI_VDEVICE(PROMISE, 0x3f20), board_ahci },	/* PDC42819 */
diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
index af63c75c2001..5513296e5e2e 100644
--- a/drivers/ata/ahci.h
+++ b/drivers/ata/ahci.h
@@ -237,6 +237,7 @@ enum {
 						        error-handling stage) */
 	AHCI_HFLAG_MULTI_MSI		= (1 << 16), /* multiple PCI MSIs */
 	AHCI_HFLAG_NO_DEVSLP		= (1 << 17), /* no device sleep */
+	AHCI_HFLAG_NO_FBS		= (1 << 18), /* no FBS */
 
 	/* ap->flags bits */
 
@@ -370,7 +371,9 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 		      int pmp, unsigned long deadline,
 		      int (*check_ready)(struct ata_link *link));
 
+unsigned int ahci_qc_issue(struct ata_queued_cmd *qc);
 int ahci_stop_engine(struct ata_port *ap);
+void ahci_start_fis_rx(struct ata_port *ap);
 void ahci_start_engine(struct ata_port *ap);
 int ahci_check_ready(struct ata_link *link);
 int ahci_kick_engine(struct ata_port *ap);
diff --git a/drivers/ata/ahci_da850.c b/drivers/ata/ahci_da850.c
index 2c83613ce2db..2b77d53bccf8 100644
--- a/drivers/ata/ahci_da850.c
+++ b/drivers/ata/ahci_da850.c
@@ -85,7 +85,8 @@ static int ahci_da850_probe(struct platform_device *pdev)
 
 	da850_sata_init(dev, pwrdn_reg, hpriv->mmio);
 
-	rc = ahci_platform_init_host(pdev, hpriv, &ahci_da850_port_info, 0, 0);
+	rc = ahci_platform_init_host(pdev, hpriv, &ahci_da850_port_info,
+				     0, 0, 0);
 	if (rc)
 		goto disable_resources;
 
diff --git a/drivers/ata/ahci_imx.c b/drivers/ata/ahci_imx.c
index 8befeb69eeb1..cac4360f272a 100644
--- a/drivers/ata/ahci_imx.c
+++ b/drivers/ata/ahci_imx.c
@@ -58,6 +58,8 @@ enum ahci_imx_type {
 struct imx_ahci_priv {
 	struct platform_device *ahci_pdev;
 	enum ahci_imx_type type;
+	struct clk *sata_clk;
+	struct clk *sata_ref_clk;
 	struct clk *ahb_clk;
 	struct regmap *gpr;
 	bool no_device;
@@ -224,7 +226,7 @@ static int imx_sata_enable(struct ahci_host_priv *hpriv)
 			return ret;
 	}
 
-	ret = ahci_platform_enable_clks(hpriv);
+	ret = clk_prepare_enable(imxpriv->sata_ref_clk);
 	if (ret < 0)
 		goto disable_regulator;
 
@@ -291,7 +293,7 @@ static void imx_sata_disable(struct ahci_host_priv *hpriv)
 				   !IMX6Q_GPR13_SATA_MPLL_CLK_EN);
 	}
 
-	ahci_platform_disable_clks(hpriv);
+	clk_disable_unprepare(imxpriv->sata_ref_clk);
 
 	if (hpriv->target_pwr)
 		regulator_disable(hpriv->target_pwr);
@@ -324,6 +326,9 @@ static void ahci_imx_error_handler(struct ata_port *ap)
 	writel(reg_val | IMX_P0PHYCR_TEST_PDDQ, mmio + IMX_P0PHYCR);
 	imx_sata_disable(hpriv);
 	imxpriv->no_device = true;
+
+	dev_info(ap->dev, "no device found, disabling link.\n");
+	dev_info(ap->dev, "pass " MODULE_PARAM_PREFIX ".hotplug=1 to enable hotplug\n");
 }
 
 static int ahci_imx_softreset(struct ata_link *link, unsigned int *class,
@@ -385,6 +390,19 @@ static int imx_ahci_probe(struct platform_device *pdev)
 	imxpriv->no_device = false;
 	imxpriv->first_time = true;
 	imxpriv->type = (enum ahci_imx_type)of_id->data;
+
+	imxpriv->sata_clk = devm_clk_get(dev, "sata");
+	if (IS_ERR(imxpriv->sata_clk)) {
+		dev_err(dev, "can't get sata clock.\n");
+		return PTR_ERR(imxpriv->sata_clk);
+	}
+
+	imxpriv->sata_ref_clk = devm_clk_get(dev, "sata_ref");
+	if (IS_ERR(imxpriv->sata_ref_clk)) {
+		dev_err(dev, "can't get sata_ref clock.\n");
+		return PTR_ERR(imxpriv->sata_ref_clk);
+	}
+
 	imxpriv->ahb_clk = devm_clk_get(dev, "ahb");
 	if (IS_ERR(imxpriv->ahb_clk)) {
 		dev_err(dev, "can't get ahb clock.\n");
@@ -407,10 +425,14 @@ static int imx_ahci_probe(struct platform_device *pdev)
 
 	hpriv->plat_data = imxpriv;
 
-	ret = imx_sata_enable(hpriv);
+	ret = clk_prepare_enable(imxpriv->sata_clk);
 	if (ret)
 		return ret;
 
+	ret = imx_sata_enable(hpriv);
+	if (ret)
+		goto disable_clk;
+
 	/*
 	 * Configure the HWINIT bits of the HOST_CAP and HOST_PORTS_IMPL,
 	 * and IP vendor specific register IMX_TIMER1MS.
@@ -432,18 +454,27 @@ static int imx_ahci_probe(struct platform_device *pdev)
 	reg_val = clk_get_rate(imxpriv->ahb_clk) / 1000;
 	writel(reg_val, hpriv->mmio + IMX_TIMER1MS);
 
-	ret = ahci_platform_init_host(pdev, hpriv, &ahci_imx_port_info, 0, 0);
+	ret = ahci_platform_init_host(pdev, hpriv, &ahci_imx_port_info,
+				      0, 0, 0);
 	if (ret)
-		imx_sata_disable(hpriv);
+		goto disable_sata;
 
+	return 0;
+
+disable_sata:
+	imx_sata_disable(hpriv);
+disable_clk:
+	clk_disable_unprepare(imxpriv->sata_clk);
 	return ret;
 }
 
 static void ahci_imx_host_stop(struct ata_host *host)
 {
 	struct ahci_host_priv *hpriv = host->private_data;
+	struct imx_ahci_priv *imxpriv = hpriv->plat_data;
 
 	imx_sata_disable(hpriv);
+	clk_disable_unprepare(imxpriv->sata_clk);
 }
 
 #ifdef CONFIG_PM_SLEEP
diff --git a/drivers/ata/ahci_mvebu.c b/drivers/ata/ahci_mvebu.c
new file mode 100644
index 000000000000..fd3dfd733b84
--- /dev/null
+++ b/drivers/ata/ahci_mvebu.c
@@ -0,0 +1,128 @@
+/*
+ * AHCI glue platform driver for Marvell EBU SOCs
+ *
+ * Copyright (C) 2014 Marvell
+ *
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+ * Marcin Wojtas <mw@semihalf.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/ahci_platform.h>
+#include <linux/kernel.h>
+#include <linux/mbus.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include "ahci.h"
+
+#define AHCI_VENDOR_SPECIFIC_0_ADDR  0xa0
+#define AHCI_VENDOR_SPECIFIC_0_DATA  0xa4
+
+#define AHCI_WINDOW_CTRL(win)	(0x60 + ((win) << 4))
+#define AHCI_WINDOW_BASE(win)	(0x64 + ((win) << 4))
+#define AHCI_WINDOW_SIZE(win)	(0x68 + ((win) << 4))
+
+static void ahci_mvebu_mbus_config(struct ahci_host_priv *hpriv,
+				   const struct mbus_dram_target_info *dram)
+{
+	int i;
+
+	for (i = 0; i < 4; i++) {
+		writel(0, hpriv->mmio + AHCI_WINDOW_CTRL(i));
+		writel(0, hpriv->mmio + AHCI_WINDOW_BASE(i));
+		writel(0, hpriv->mmio + AHCI_WINDOW_SIZE(i));
+	}
+
+	for (i = 0; i < dram->num_cs; i++) {
+		const struct mbus_dram_window *cs = dram->cs + i;
+
+		writel((cs->mbus_attr << 8) |
+		       (dram->mbus_dram_target_id << 4) | 1,
+		       hpriv->mmio + AHCI_WINDOW_CTRL(i));
+		writel(cs->base, hpriv->mmio + AHCI_WINDOW_BASE(i));
+		writel(((cs->size - 1) & 0xffff0000),
+		       hpriv->mmio + AHCI_WINDOW_SIZE(i));
+	}
+}
+
+static void ahci_mvebu_regret_option(struct ahci_host_priv *hpriv)
+{
+	/*
+	 * Enable the regret bit to allow the SATA unit to regret a
+	 * request that didn't receive an acknowlegde and avoid a
+	 * deadlock
+	 */
+	writel(0x4, hpriv->mmio + AHCI_VENDOR_SPECIFIC_0_ADDR);
+	writel(0x80, hpriv->mmio + AHCI_VENDOR_SPECIFIC_0_DATA);
+}
+
+static const struct ata_port_info ahci_mvebu_port_info = {
+	.flags	   = AHCI_FLAG_COMMON,
+	.pio_mask  = ATA_PIO4,
+	.udma_mask = ATA_UDMA6,
+	.port_ops  = &ahci_platform_ops,
+};
+
+static int ahci_mvebu_probe(struct platform_device *pdev)
+{
+	struct ahci_host_priv *hpriv;
+	const struct mbus_dram_target_info *dram;
+	int rc;
+
+	hpriv = ahci_platform_get_resources(pdev);
+	if (IS_ERR(hpriv))
+		return PTR_ERR(hpriv);
+
+	rc = ahci_platform_enable_resources(hpriv);
+	if (rc)
+		return rc;
+
+	dram = mv_mbus_dram_info();
+	if (!dram)
+		return -ENODEV;
+
+	ahci_mvebu_mbus_config(hpriv, dram);
+	ahci_mvebu_regret_option(hpriv);
+
+	rc = ahci_platform_init_host(pdev, hpriv, &ahci_mvebu_port_info,
+				     0, 0, 0);
+	if (rc)
+		goto disable_resources;
+
+	return 0;
+
+disable_resources:
+	ahci_platform_disable_resources(hpriv);
+	return rc;
+}
+
+static const struct of_device_id ahci_mvebu_of_match[] = {
+	{ .compatible = "marvell,armada-380-ahci", },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, ahci_mvebu_of_match);
+
+/*
+ * We currently don't provide power management related operations,
+ * since there is no suspend/resume support at the platform level for
+ * Armada 38x for the moment.
+ */
+static struct platform_driver ahci_mvebu_driver = {
+	.probe = ahci_mvebu_probe,
+	.remove = ata_platform_remove_one,
+	.driver = {
+		.name = "ahci-mvebu",
+		.owner = THIS_MODULE,
+		.of_match_table = ahci_mvebu_of_match,
+	},
+};
+module_platform_driver(ahci_mvebu_driver);
+
+MODULE_DESCRIPTION("Marvell EBU AHCI SATA driver");
+MODULE_AUTHOR("Thomas Petazzoni <thomas.petazzoni@free-electrons.com>, Marcin Wojtas <mw@semihalf.com>");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:ahci_mvebu");
diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c
index ef67e79944f9..b10d81ddb528 100644
--- a/drivers/ata/ahci_platform.c
+++ b/drivers/ata/ahci_platform.c
@@ -16,6 +16,7 @@
 #include <linux/module.h>
 #include <linux/pm.h>
 #include <linux/device.h>
+#include <linux/of_device.h>
 #include <linux/platform_device.h>
 #include <linux/libata.h>
 #include <linux/ahci_platform.h>
@@ -33,6 +34,7 @@ static int ahci_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct ahci_platform_data *pdata = dev_get_platdata(dev);
 	struct ahci_host_priv *hpriv;
+	unsigned long hflags = 0;
 	int rc;
 
 	hpriv = ahci_platform_get_resources(pdev);
@@ -55,7 +57,11 @@ static int ahci_probe(struct platform_device *pdev)
 			goto disable_resources;
 	}
 
-	rc = ahci_platform_init_host(pdev, hpriv, &ahci_port_info, 0, 0);
+	if (of_device_is_compatible(dev->of_node, "hisilicon,hisi-ahci"))
+		hflags |= AHCI_HFLAG_NO_FBS | AHCI_HFLAG_NO_NCQ;
+
+	rc = ahci_platform_init_host(pdev, hpriv, &ahci_port_info,
+				     hflags, 0, 0);
 	if (rc)
 		goto pdata_exit;
 
@@ -76,6 +82,7 @@ static const struct of_device_id ahci_of_match[] = {
 	{ .compatible = "snps,exynos5440-ahci", },
 	{ .compatible = "ibm,476gtr-ahci", },
 	{ .compatible = "snps,dwc-ahci", },
+	{ .compatible = "hisilicon,hisi-ahci", },
 	{},
 };
 MODULE_DEVICE_TABLE(of, ahci_of_match);
diff --git a/drivers/ata/ahci_st.c b/drivers/ata/ahci_st.c
index 633222226c19..2595598df9ce 100644
--- a/drivers/ata/ahci_st.c
+++ b/drivers/ata/ahci_st.c
@@ -166,7 +166,7 @@ static int st_ahci_probe(struct platform_device *pdev)
 	if (err)
 		return err;
 
-	err = ahci_platform_init_host(pdev, hpriv, &st_ahci_port_info, 0, 0);
+	err = ahci_platform_init_host(pdev, hpriv, &st_ahci_port_info, 0, 0, 0);
 	if (err) {
 		ahci_platform_disable_resources(hpriv);
 		return err;
diff --git a/drivers/ata/ahci_sunxi.c b/drivers/ata/ahci_sunxi.c
index 42d3f64e74b3..02002f125bd4 100644
--- a/drivers/ata/ahci_sunxi.c
+++ b/drivers/ata/ahci_sunxi.c
@@ -157,8 +157,6 @@ static void ahci_sunxi_start_engine(struct ata_port *ap)
 }
 
 static const struct ata_port_info ahci_sunxi_port_info = {
-	AHCI_HFLAGS(AHCI_HFLAG_32BIT_ONLY | AHCI_HFLAG_NO_MSI |
-			  AHCI_HFLAG_NO_PMP | AHCI_HFLAG_YES_NCQ),
 	.flags		= AHCI_FLAG_COMMON | ATA_FLAG_NCQ,
 	.pio_mask	= ATA_PIO4,
 	.udma_mask	= ATA_UDMA6,
@@ -169,6 +167,7 @@ static int ahci_sunxi_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct ahci_host_priv *hpriv;
+	unsigned long hflags;
 	int rc;
 
 	hpriv = ahci_platform_get_resources(pdev);
@@ -185,7 +184,11 @@ static int ahci_sunxi_probe(struct platform_device *pdev)
 	if (rc)
 		goto disable_resources;
 
-	rc = ahci_platform_init_host(pdev, hpriv, &ahci_sunxi_port_info, 0, 0);
+	hflags = AHCI_HFLAG_32BIT_ONLY | AHCI_HFLAG_NO_MSI |
+		 AHCI_HFLAG_NO_PMP | AHCI_HFLAG_YES_NCQ;
+
+	rc = ahci_platform_init_host(pdev, hpriv, &ahci_sunxi_port_info,
+				     hflags, 0, 0);
 	if (rc)
 		goto disable_resources;
 
diff --git a/drivers/ata/ahci_xgene.c b/drivers/ata/ahci_xgene.c
index 77c89bf171f1..ee3a3659bd9e 100644
--- a/drivers/ata/ahci_xgene.c
+++ b/drivers/ata/ahci_xgene.c
@@ -78,6 +78,7 @@
 struct xgene_ahci_context {
 	struct ahci_host_priv *hpriv;
 	struct device *dev;
+	u8 last_cmd[MAX_AHCI_CHN_PERCTR]; /* tracking the last command issued*/
 	void __iomem *csr_core;		/* Core CSR address of IP */
 	void __iomem *csr_diag;		/* Diag CSR address of IP */
 	void __iomem *csr_axi;		/* AXI CSR address of IP */
@@ -98,20 +99,62 @@ static int xgene_ahci_init_memram(struct xgene_ahci_context *ctx)
 }
 
 /**
+ * xgene_ahci_restart_engine - Restart the dma engine.
+ * @ap : ATA port of interest
+ *
+ * Restarts the dma engine inside the controller.
+ */
+static int xgene_ahci_restart_engine(struct ata_port *ap)
+{
+	struct ahci_host_priv *hpriv = ap->host->private_data;
+
+	ahci_stop_engine(ap);
+	ahci_start_fis_rx(ap);
+	hpriv->start_engine(ap);
+
+	return 0;
+}
+
+/**
+ * xgene_ahci_qc_issue - Issue commands to the device
+ * @qc: Command to issue
+ *
+ * Due to Hardware errata for IDENTIFY DEVICE command, the controller cannot
+ * clear the BSY bit after receiving the PIO setup FIS. This results in the dma
+ * state machine goes into the CMFatalErrorUpdate state and locks up. By
+ * restarting the dma engine, it removes the controller out of lock up state.
+ */
+static unsigned int xgene_ahci_qc_issue(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
+	struct xgene_ahci_context *ctx = hpriv->plat_data;
+	int rc = 0;
+
+	if (unlikely(ctx->last_cmd[ap->port_no] == ATA_CMD_ID_ATA))
+		xgene_ahci_restart_engine(ap);
+
+	rc = ahci_qc_issue(qc);
+
+	/* Save the last command issued */
+	ctx->last_cmd[ap->port_no] = qc->tf.command;
+
+	return rc;
+}
+
+/**
  * xgene_ahci_read_id - Read ID data from the specified device
  * @dev: device
  * @tf: proposed taskfile
  * @id: data buffer
  *
  * This custom read ID function is required due to the fact that the HW
- * does not support DEVSLP and the controller state machine may get stuck
- * after processing the ID query command.
+ * does not support DEVSLP.
  */
 static unsigned int xgene_ahci_read_id(struct ata_device *dev,
 				       struct ata_taskfile *tf, u16 *id)
 {
 	u32 err_mask;
-	void __iomem *port_mmio = ahci_port_base(dev->link->ap);
 
 	err_mask = ata_do_dev_read_id(dev, tf, id);
 	if (err_mask)
@@ -133,16 +176,6 @@ static unsigned int xgene_ahci_read_id(struct ata_device *dev,
 	 */
 	id[ATA_ID_FEATURE_SUPP] &= ~(1 << 8);
 
-	/*
-	 * Due to HW errata, restart the port if no other command active.
-	 * Otherwise the controller may get stuck.
-	 */
-	if (!readl(port_mmio + PORT_CMD_ISSUE)) {
-		writel(PORT_CMD_FIS_RX, port_mmio + PORT_CMD);
-		readl(port_mmio + PORT_CMD);	/* Force a barrier */
-		writel(PORT_CMD_FIS_RX | PORT_CMD_START, port_mmio + PORT_CMD);
-		readl(port_mmio + PORT_CMD);	/* Force a barrier */
-	}
 	return 0;
 }
 
@@ -300,10 +333,10 @@ static struct ata_port_operations xgene_ahci_ops = {
 	.host_stop = xgene_ahci_host_stop,
 	.hardreset = xgene_ahci_hardreset,
 	.read_id = xgene_ahci_read_id,
+	.qc_issue = xgene_ahci_qc_issue,
 };
 
 static const struct ata_port_info xgene_ahci_port_info = {
-	AHCI_HFLAGS(AHCI_HFLAG_NO_PMP | AHCI_HFLAG_YES_NCQ),
 	.flags = AHCI_FLAG_COMMON | ATA_FLAG_NCQ,
 	.pio_mask = ATA_PIO4,
 	.udma_mask = ATA_UDMA6,
@@ -382,6 +415,7 @@ static int xgene_ahci_probe(struct platform_device *pdev)
 	struct ahci_host_priv *hpriv;
 	struct xgene_ahci_context *ctx;
 	struct resource *res;
+	unsigned long hflags;
 	int rc;
 
 	hpriv = ahci_platform_get_resources(pdev);
@@ -450,7 +484,10 @@ static int xgene_ahci_probe(struct platform_device *pdev)
 		goto disable_resources;
 	}
 
-	rc = ahci_platform_init_host(pdev, hpriv, &xgene_ahci_port_info, 0, 0);
+	hflags = AHCI_HFLAG_NO_PMP | AHCI_HFLAG_YES_NCQ;
+
+	rc = ahci_platform_init_host(pdev, hpriv, &xgene_ahci_port_info,
+				     hflags, 0, 0);
 	if (rc)
 		goto disable_resources;
 
diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index 9498a7d3846f..9ff545ce8da3 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -241,7 +241,7 @@ static struct pci_driver ata_generic_pci_driver = {
 	.id_table	= ata_generic,
 	.probe 		= ata_generic_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 6334c8d7c3f1..893e30e9a9ef 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -830,7 +830,7 @@ static bool piix_irq_check(struct ata_port *ap)
 	return ap->ops->bmdma_status(ap) & ATA_DMA_INTR;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int piix_broken_suspend(void)
 {
 	static const struct dmi_system_id sysids[] = {
@@ -1767,7 +1767,7 @@ static struct pci_driver piix_pci_driver = {
 	.id_table		= piix_pci_tbl,
 	.probe			= piix_init_one,
 	.remove			= piix_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= piix_pci_device_suspend,
 	.resume			= piix_pci_device_resume,
 #endif
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index b9861453fc81..d72ce0470309 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -68,7 +68,6 @@ static ssize_t ahci_transmit_led_message(struct ata_port *ap, u32 state,
 
 static int ahci_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
 static int ahci_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
-static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc);
 static bool ahci_qc_fill_rtf(struct ata_queued_cmd *qc);
 static int ahci_port_start(struct ata_port *ap);
 static void ahci_port_stop(struct ata_port *ap);
@@ -464,6 +463,11 @@ void ahci_save_initial_config(struct device *dev,
 		cap |= HOST_CAP_FBS;
 	}
 
+	if ((cap & HOST_CAP_FBS) && (hpriv->flags & AHCI_HFLAG_NO_FBS)) {
+		dev_info(dev, "controller can't do FBS, turning off CAP_FBS\n");
+		cap &= ~HOST_CAP_FBS;
+	}
+
 	if (force_port_map && port_map != force_port_map) {
 		dev_info(dev, "forcing port_map 0x%x -> 0x%x\n",
 			 port_map, force_port_map);
@@ -615,7 +619,7 @@ int ahci_stop_engine(struct ata_port *ap)
 }
 EXPORT_SYMBOL_GPL(ahci_stop_engine);
 
-static void ahci_start_fis_rx(struct ata_port *ap)
+void ahci_start_fis_rx(struct ata_port *ap)
 {
 	void __iomem *port_mmio = ahci_port_base(ap);
 	struct ahci_host_priv *hpriv = ap->host->private_data;
@@ -641,6 +645,7 @@ static void ahci_start_fis_rx(struct ata_port *ap)
 	/* flush */
 	readl(port_mmio + PORT_CMD);
 }
+EXPORT_SYMBOL_GPL(ahci_start_fis_rx);
 
 static int ahci_stop_fis_rx(struct ata_port *ap)
 {
@@ -1940,7 +1945,7 @@ irqreturn_t ahci_interrupt(int irq, void *dev_instance)
 }
 EXPORT_SYMBOL_GPL(ahci_interrupt);
 
-static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc)
+unsigned int ahci_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	void __iomem *port_mmio = ahci_port_base(ap);
@@ -1969,6 +1974,7 @@ static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(ahci_qc_issue);
 
 static bool ahci_qc_fill_rtf(struct ata_queued_cmd *qc)
 {
diff --git a/drivers/ata/libahci_platform.c b/drivers/ata/libahci_platform.c
index 7cb3a85719c0..b0077589f065 100644
--- a/drivers/ata/libahci_platform.c
+++ b/drivers/ata/libahci_platform.c
@@ -250,8 +250,13 @@ struct ahci_host_priv *ahci_platform_get_resources(struct platform_device *pdev)
 	if (IS_ERR(hpriv->phy)) {
 		rc = PTR_ERR(hpriv->phy);
 		switch (rc) {
-		case -ENODEV:
 		case -ENOSYS:
+			/* No PHY support. Check if PHY is required. */
+			if (of_find_property(dev->of_node, "phys", NULL)) {
+				dev_err(dev, "couldn't get sata-phy: ENOSYS\n");
+				goto err_out;
+			}
+		case -ENODEV:
 			/* continue normally */
 			hpriv->phy = NULL;
 			break;
@@ -283,6 +288,7 @@ EXPORT_SYMBOL_GPL(ahci_platform_get_resources);
  * @pdev: platform device pointer for the host
  * @hpriv: ahci-host private data for the host
  * @pi_template: template for the ata_port_info to use
+ * @host_flags: ahci host flags used in ahci_host_priv
  * @force_port_map: param passed to ahci_save_initial_config
  * @mask_port_map: param passed to ahci_save_initial_config
  *
@@ -296,6 +302,7 @@ EXPORT_SYMBOL_GPL(ahci_platform_get_resources);
 int ahci_platform_init_host(struct platform_device *pdev,
 			    struct ahci_host_priv *hpriv,
 			    const struct ata_port_info *pi_template,
+			    unsigned long host_flags,
 			    unsigned int force_port_map,
 			    unsigned int mask_port_map)
 {
@@ -312,7 +319,8 @@ int ahci_platform_init_host(struct platform_device *pdev,
 	}
 
 	/* prepare host */
-	hpriv->flags |= (unsigned long)pi.private_data;
+	pi.private_data = (void *)host_flags;
+	hpriv->flags |= host_flags;
 
 	ahci_save_initial_config(dev, hpriv, force_port_map, mask_port_map);
 
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index ef8567de6a75..72691fd93948 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -1993,7 +1993,11 @@ static unsigned int ata_scsiop_inq_std(struct ata_scsi_args *args, u8 *rbuf)
 	memcpy(rbuf, hdr, sizeof(hdr));
 	memcpy(&rbuf[8], "ATA     ", 8);
 	ata_id_string(args->id, &rbuf[16], ATA_ID_PROD, 16);
-	ata_id_string(args->id, &rbuf[32], ATA_ID_FW_REV, 4);
+
+	/* From SAT, use last 2 words from fw rev unless they are spaces */
+	ata_id_string(args->id, &rbuf[32], ATA_ID_FW_REV + 2, 4);
+	if (strncmp(&rbuf[32], "    ", 4) == 0)
+		ata_id_string(args->id, &rbuf[32], ATA_ID_FW_REV, 4);
 
 	if (rbuf[32] == 0 || rbuf[32] == ' ')
 		memcpy(&rbuf[32], "n/a ", 4);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index b603720b877d..1121153f1ecd 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -2433,15 +2433,6 @@ int ata_pci_sff_activate_host(struct ata_host *host,
 		mask = (1 << 2) | (1 << 0);
 		if ((tmp8 & mask) != mask)
 			legacy_mode = 1;
-#if defined(CONFIG_NO_ATA_LEGACY)
-		/* Some platforms with PCI limits cannot address compat
-		   port space. In that case we punt if their firmware has
-		   left a device in compatibility mode */
-		if (legacy_mode) {
-			printk(KERN_ERR "ata: Compatibility mode ATA is not supported on this platform, skipping.\n");
-			return -EOPNOTSUPP;
-		}
-#endif
 	}
 
 	if (!devres_open_group(dev, NULL, GFP_KERNEL))
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index 5108b8744dce..b70fce2a38eb 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -265,7 +265,7 @@ static struct pci_driver pacpi_pci_driver = {
 	.id_table		= pacpi_pci_tbl,
 	.probe			= pacpi_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index 1b7b2ccabcff..d19cd88ed2d3 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -589,7 +589,7 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		return ata_pci_bmdma_init_one(pdev, ppi, &ali_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int ali_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -616,7 +616,7 @@ static struct pci_driver ali_pci_driver = {
 	.id_table	= ali,
 	.probe 		= ali_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ali_reinit_one,
 #endif
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 1206fa6b62ca..8d4d959a821c 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -574,7 +574,7 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_pci_bmdma_init_one(pdev, ppi, &amd_sht, hpriv, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int amd_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -625,7 +625,7 @@ static struct pci_driver amd_pci_driver = {
 	.id_table	= amd,
 	.probe 		= amd_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= amd_reinit_one,
 #endif
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index 3aa4e655e3c6..96c05c9494fa 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -422,7 +422,7 @@ static const struct pci_device_id artop_pci_tbl[] = {
 	{ }	/* terminate list */
 };
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int atp8xx_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -444,7 +444,7 @@ static struct pci_driver artop_pci_driver = {
 	.id_table		= artop_pci_tbl,
 	.probe			= artop_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= atp8xx_reinit_one,
 #endif
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 30fa4ca4cef6..970f7767e5fd 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -298,7 +298,7 @@ static struct pci_driver atiixp_pci_driver = {
 	.id_table	= atiixp,
 	.probe 		= atiixp_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.resume		= ata_pci_device_resume,
 	.suspend	= ata_pci_device_suspend,
 #endif
diff --git a/drivers/ata/pata_atp867x.c b/drivers/ata/pata_atp867x.c
index 7e73a0f1e323..a705cfca90f7 100644
--- a/drivers/ata/pata_atp867x.c
+++ b/drivers/ata/pata_atp867x.c
@@ -530,7 +530,7 @@ err_out:
 	return rc;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int atp867x_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -558,7 +558,7 @@ static struct pci_driver atp867x_driver = {
 	.id_table 	= atp867x_pci_tbl,
 	.probe 		= atp867x_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= atp867x_reinit_one,
 #endif
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index ba0d8a29dc23..03f2f2bc83bd 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1619,7 +1619,7 @@ static int bfin_atapi_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int bfin_atapi_suspend(struct platform_device *pdev, pm_message_t state)
 {
 	struct ata_host *host = platform_get_drvdata(pdev);
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index 57f1be64dbf2..c47caa807fa9 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -231,7 +231,7 @@ static int cmd640_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_pci_sff_init_one(pdev, ppi, &cmd640_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int cmd640_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -256,7 +256,7 @@ static struct pci_driver cmd640_pci_driver = {
 	.id_table	= cmd640,
 	.probe 		= cmd640_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= cmd640_reinit_one,
 #endif
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index 6bca3505b9e9..13ca5883285b 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -487,7 +487,7 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_pci_bmdma_init_one(pdev, ppi, &cmd64x_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int cmd64x_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -518,7 +518,7 @@ static struct pci_driver cmd64x_pci_driver = {
 	.id_table	= cmd64x,
 	.probe 		= cmd64x_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= cmd64x_reinit_one,
 #endif
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
index bcde4b786807..d65cb9d2fa8c 100644
--- a/drivers/ata/pata_cs5520.c
+++ b/drivers/ata/pata_cs5520.c
@@ -229,7 +229,7 @@ static int cs5520_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_host_register(host, &cs5520_sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 /**
  *	cs5520_reinit_one	-	device resume
  *	@pdev: PCI device
@@ -278,7 +278,7 @@ static int cs5520_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 	pci_save_state(pdev);
 	return 0;
 }
-#endif /* CONFIG_PM */
+#endif /* CONFIG_PM_SLEEP */
 
 /* For now keep DMA off. We can set it for all but A rev CS5510 once the
    core ATA code can handle it */
@@ -295,7 +295,7 @@ static struct pci_driver cs5520_pci_driver = {
 	.id_table	= pata_cs5520,
 	.probe 		= cs5520_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= cs5520_pci_device_suspend,
 	.resume		= cs5520_reinit_one,
 #endif
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index 8afe854a5a50..48ae4b434474 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -326,7 +326,7 @@ static int cs5530_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_pci_bmdma_init_one(pdev, ppi, &cs5530_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int cs5530_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -343,7 +343,7 @@ static int cs5530_reinit_one(struct pci_dev *pdev)
 	ata_host_resume(host);
 	return 0;
 }
-#endif /* CONFIG_PM */
+#endif /* CONFIG_PM_SLEEP */
 
 static const struct pci_device_id cs5530[] = {
 	{ PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5530_IDE), },
@@ -356,7 +356,7 @@ static struct pci_driver cs5530_pci_driver = {
 	.id_table	= cs5530,
 	.probe 		= cs5530_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= cs5530_reinit_one,
 #endif
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index 2c0986fa4bb2..97584e8456d9 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -200,7 +200,7 @@ static struct pci_driver cs5535_pci_driver = {
 	.id_table	= cs5535,
 	.probe 		= cs5535_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index 32ddcae5a360..6c15a554efbe 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -297,7 +297,7 @@ static struct pci_driver cs5536_pci_driver = {
 	.id_table	= cs5536,
 	.probe		= cs5536_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index 3435bd6a5cc9..793018460d82 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -151,7 +151,7 @@ static struct pci_driver cy82c693_pci_driver = {
 	.id_table	= cy82c693,
 	.probe 		= cy82c693_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index f440892225f4..4a57a6f032d9 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -288,7 +288,7 @@ static struct pci_driver efar_pci_driver = {
 	.id_table		= efar_pci_tbl,
 	.probe			= efar_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_ep93xx.c b/drivers/ata/pata_ep93xx.c
index cad9d45749c4..6ad5c072ce34 100644
--- a/drivers/ata/pata_ep93xx.c
+++ b/drivers/ata/pata_ep93xx.c
@@ -708,8 +708,8 @@ static void ep93xx_pata_dma_start(struct ata_queued_cmd *qc)
 	struct dma_chan *channel = qc->dma_dir == DMA_TO_DEVICE
 		? drv_data->dma_tx_channel : drv_data->dma_rx_channel;
 
-	txd = channel->device->device_prep_slave_sg(channel, qc->sg,
-		 qc->n_elem, qc->dma_dir, DMA_CTRL_ACK, NULL);
+	txd = dmaengine_prep_slave_sg(channel, qc->sg, qc->n_elem, qc->dma_dir,
+		DMA_CTRL_ACK);
 	if (!txd) {
 		dev_err(qc->ap->dev, "failed to prepare slave for sg dma\n");
 		return;
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index 8e76f79689d3..cbc3de793d1d 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -386,7 +386,7 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	return ata_pci_bmdma_init_one(dev, ppi, &hpt36x_sht, hpriv, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int hpt36x_reinit_one(struct pci_dev *dev)
 {
 	struct ata_host *host = pci_get_drvdata(dev);
@@ -411,7 +411,7 @@ static struct pci_driver hpt36x_pci_driver = {
 	.id_table	= hpt36x,
 	.probe		= hpt36x_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= hpt36x_reinit_one,
 #endif
diff --git a/drivers/ata/pata_hpt3x3.c b/drivers/ata/pata_hpt3x3.c
index 255c5aaff3a8..d019cdd5bc9f 100644
--- a/drivers/ata/pata_hpt3x3.c
+++ b/drivers/ata/pata_hpt3x3.c
@@ -249,7 +249,7 @@ static int hpt3x3_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 				 IRQF_SHARED, &hpt3x3_sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int hpt3x3_reinit_one(struct pci_dev *dev)
 {
 	struct ata_host *host = pci_get_drvdata(dev);
@@ -277,7 +277,7 @@ static struct pci_driver hpt3x3_pci_driver = {
 	.id_table	= hpt3x3,
 	.probe 		= hpt3x3_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= hpt3x3_reinit_one,
 #endif
diff --git a/drivers/ata/pata_imx.c b/drivers/ata/pata_imx.c
index e0872db913d6..af424573c2ff 100644
--- a/drivers/ata/pata_imx.c
+++ b/drivers/ata/pata_imx.c
@@ -185,7 +185,7 @@ static int pata_imx_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int pata_imx_suspend(struct device *dev)
 {
 	struct ata_host *host = dev_get_drvdata(dev);
@@ -244,7 +244,7 @@ static struct platform_driver pata_imx_driver = {
 		.name		= DRV_NAME,
 		.of_match_table	= imx_pata_dt_ids,
 		.owner		= THIS_MODULE,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 		.pm		= &pata_imx_pm_ops,
 #endif
 	},
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index 81369d187a5c..4f97d1e52f85 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -283,7 +283,7 @@ static struct pci_driver it8213_pci_driver = {
 	.id_table		= it8213_pci_tbl,
 	.probe			= it8213_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index dc3d7877f29d..a5088ecb349f 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -935,7 +935,7 @@ static int it821x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_pci_bmdma_init_one(pdev, ppi, &it821x_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int it821x_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -965,7 +965,7 @@ static struct pci_driver it821x_pci_driver = {
 	.id_table	= it821x,
 	.probe 		= it821x_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= it821x_reinit_one,
 #endif
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index b1cfa0258fd3..4d1a5d2c4287 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -157,7 +157,7 @@ static struct pci_driver jmicron_pci_driver = {
 	.id_table		= jmicron_pci_tbl,
 	.probe			= jmicron_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_macio.c b/drivers/ata/pata_macio.c
index c28d0645e851..a02f76fdcfcd 100644
--- a/drivers/ata/pata_macio.c
+++ b/drivers/ata/pata_macio.c
@@ -845,8 +845,7 @@ static int pata_macio_slave_config(struct scsi_device *sdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-
+#ifdef CONFIG_PM_SLEEP
 static int pata_macio_do_suspend(struct pata_macio_priv *priv, pm_message_t mesg)
 {
 	int rc;
@@ -907,8 +906,7 @@ static int pata_macio_do_resume(struct pata_macio_priv *priv)
 
 	return 0;
 }
-
-#endif /* CONFIG_PM */
+#endif /* CONFIG_PM_SLEEP */
 
 static struct scsi_host_template pata_macio_sht = {
 	ATA_BASE_SHT(DRV_NAME),
@@ -1208,8 +1206,7 @@ static int pata_macio_detach(struct macio_dev *mdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-
+#ifdef CONFIG_PM_SLEEP
 static int pata_macio_suspend(struct macio_dev *mdev, pm_message_t mesg)
 {
 	struct ata_host *host = macio_get_drvdata(mdev);
@@ -1223,8 +1220,7 @@ static int pata_macio_resume(struct macio_dev *mdev)
 
 	return pata_macio_do_resume(host->private_data);
 }
-
-#endif /* CONFIG_PM */
+#endif /* CONFIG_PM_SLEEP */
 
 #ifdef CONFIG_PMAC_MEDIABAY
 static void pata_macio_mb_event(struct macio_dev* mdev, int mb_state)
@@ -1316,8 +1312,7 @@ static void pata_macio_pci_detach(struct pci_dev *pdev)
 	ata_host_detach(host);
 }
 
-#ifdef CONFIG_PM
-
+#ifdef CONFIG_PM_SLEEP
 static int pata_macio_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -1331,8 +1326,7 @@ static int pata_macio_pci_resume(struct pci_dev *pdev)
 
 	return pata_macio_do_resume(host->private_data);
 }
-
-#endif /* CONFIG_PM */
+#endif /* CONFIG_PM_SLEEP */
 
 static struct of_device_id pata_macio_match[] =
 {
@@ -1360,7 +1354,7 @@ static struct macio_driver pata_macio_driver =
 	},
 	.probe		= pata_macio_attach,
 	.remove		= pata_macio_detach,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= pata_macio_suspend,
 	.resume		= pata_macio_resume,
 #endif
@@ -1383,7 +1377,7 @@ static struct pci_driver pata_macio_pci_driver = {
 	.id_table	= pata_macio_pci_match,
 	.probe		= pata_macio_pci_attach,
 	.remove		= pata_macio_pci_detach,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= pata_macio_pci_suspend,
 	.resume		= pata_macio_pci_resume,
 #endif
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index 6bad3df3a13c..ae9feb1ba8f7 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -171,7 +171,7 @@ static struct pci_driver marvell_pci_driver = {
 	.id_table		= marvell_pci_tbl,
 	.probe			= marvell_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index 0024ced3e200..ccd1c83a05cc 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -819,9 +819,7 @@ mpc52xx_ata_remove(struct platform_device *op)
 	return 0;
 }
 
-
-#ifdef CONFIG_PM
-
+#ifdef CONFIG_PM_SLEEP
 static int
 mpc52xx_ata_suspend(struct platform_device *op, pm_message_t state)
 {
@@ -847,10 +845,8 @@ mpc52xx_ata_resume(struct platform_device *op)
 
 	return 0;
 }
-
 #endif
 
-
 static struct of_device_id mpc52xx_ata_of_match[] = {
 	{ .compatible = "fsl,mpc5200-ata", },
 	{ .compatible = "mpc5200-ata", },
@@ -861,7 +857,7 @@ static struct of_device_id mpc52xx_ata_of_match[] = {
 static struct platform_driver mpc52xx_ata_of_platform_driver = {
 	.probe		= mpc52xx_ata_probe,
 	.remove		= mpc52xx_ata_remove,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= mpc52xx_ata_suspend,
 	.resume		= mpc52xx_ata_resume,
 #endif
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index f39a5379e816..202b4d601393 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -223,7 +223,7 @@ static struct pci_driver mpiix_pci_driver = {
 	.id_table	= mpiix,
 	.probe 		= mpiix_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index e3b97093ef9a..0ea18331d466 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -92,7 +92,7 @@ static struct pci_driver netcell_pci_driver = {
 	.id_table		= netcell_pci_tbl,
 	.probe			= netcell_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_ninja32.c b/drivers/ata/pata_ninja32.c
index 56201a69af12..efb272da8567 100644
--- a/drivers/ata/pata_ninja32.c
+++ b/drivers/ata/pata_ninja32.c
@@ -152,8 +152,7 @@ static int ninja32_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 				 IRQF_SHARED, &ninja32_sht);
 }
 
-#ifdef CONFIG_PM
-
+#ifdef CONFIG_PM_SLEEP
 static int ninja32_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -183,7 +182,7 @@ static struct pci_driver ninja32_pci_driver = {
 	.id_table	= ninja32,
 	.probe 		= ninja32_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ninja32_reinit_one,
 #endif
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 6154c3ee11a5..200e1eb23a20 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -161,7 +161,7 @@ static struct pci_driver ns87410_pci_driver = {
 	.id_table	= ns87410,
 	.probe 		= ns87410_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index d44df7ccfe43..84c6b225b56e 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -385,7 +385,7 @@ static const struct pci_device_id ns87415_pci_tbl[] = {
 	{ }	/* terminate list */
 };
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int ns87415_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -407,7 +407,7 @@ static struct pci_driver ns87415_pci_driver = {
 	.id_table		= ns87415_pci_tbl,
 	.probe			= ns87415_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ns87415_reinit_one,
 #endif
diff --git a/drivers/ata/pata_octeon_cf.c b/drivers/ata/pata_octeon_cf.c
index 83c4ddb1bc7f..2a97d3a531ec 100644
--- a/drivers/ata/pata_octeon_cf.c
+++ b/drivers/ata/pata_octeon_cf.c
@@ -865,7 +865,7 @@ static int octeon_cf_probe(struct platform_device *pdev)
 	if (node == NULL)
 		return -EINVAL;
 
-	cf_port = kzalloc(sizeof(*cf_port), GFP_KERNEL);
+	cf_port = devm_kzalloc(&pdev->dev, sizeof(*cf_port), GFP_KERNEL);
 	if (!cf_port)
 		return -ENOMEM;
 
@@ -881,10 +881,9 @@ static int octeon_cf_probe(struct platform_device *pdev)
 	n_size = of_n_size_cells(node);
 
 	reg_prop = of_find_property(node, "reg", &reg_len);
-	if (!reg_prop || reg_len < sizeof(__be32)) {
-		rv = -EINVAL;
-		goto free_cf_port;
-	}
+	if (!reg_prop || reg_len < sizeof(__be32))
+		return -EINVAL;
+
 	cs_num = reg_prop->value;
 	cf_port->cs0 = be32_to_cpup(cs_num);
 
@@ -901,16 +900,13 @@ static int octeon_cf_probe(struct platform_device *pdev)
 				res_dma = platform_get_resource(dma_dev, IORESOURCE_MEM, 0);
 				if (!res_dma) {
 					of_node_put(dma_node);
-					rv = -EINVAL;
-					goto free_cf_port;
+					return -EINVAL;
 				}
 				cf_port->dma_base = (u64)devm_ioremap_nocache(&pdev->dev, res_dma->start,
 									 resource_size(res_dma));
-
 				if (!cf_port->dma_base) {
 					of_node_put(dma_node);
-					rv = -EINVAL;
-					goto free_cf_port;
+					return -EINVAL;
 				}
 
 				irq_handler = octeon_cf_interrupt;
@@ -921,41 +917,34 @@ static int octeon_cf_probe(struct platform_device *pdev)
 			of_node_put(dma_node);
 		}
 		res_cs1 = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-		if (!res_cs1) {
-			rv = -EINVAL;
-			goto free_cf_port;
-		}
+		if (!res_cs1)
+			return -EINVAL;
+
 		cs1 = devm_ioremap_nocache(&pdev->dev, res_cs1->start,
 					   resource_size(res_cs1));
-
 		if (!cs1)
-			goto free_cf_port;
+			return rv;
+
+		if (reg_len < (n_addr + n_size + 1) * sizeof(__be32))
+			return -EINVAL;
 
-		if (reg_len < (n_addr + n_size + 1) * sizeof(__be32)) {
-			rv = -EINVAL;
-			goto free_cf_port;
-		}
 		cs_num += n_addr + n_size;
 		cf_port->cs1 = be32_to_cpup(cs_num);
 	}
 
 	res_cs0 = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-
-	if (!res_cs0) {
-		rv = -EINVAL;
-		goto free_cf_port;
-	}
+	if (!res_cs0)
+		return -EINVAL;
 
 	cs0 = devm_ioremap_nocache(&pdev->dev, res_cs0->start,
 				   resource_size(res_cs0));
-
 	if (!cs0)
-		goto free_cf_port;
+		return rv;
 
 	/* allocate host */
 	host = ata_host_alloc(&pdev->dev, 1);
 	if (!host)
-		goto free_cf_port;
+		return rv;
 
 	ap = host->ports[0];
 	ap->private_data = cf_port;
@@ -1020,17 +1009,12 @@ static int octeon_cf_probe(struct platform_device *pdev)
 
 	ata_port_desc(ap, "cmd %p ctl %p", base, ap->ioaddr.ctl_addr);
 
-
 	dev_info(&pdev->dev, "version " DRV_VERSION" %d bit%s.\n",
 		 is_16bit ? 16 : 8,
 		 cf_port->is_true_ide ? ", True IDE" : "");
 
 	return ata_host_activate(host, irq, irq_handler,
 				 IRQF_SHARED, &octeon_cf_sht);
-
-free_cf_port:
-	kfree(cf_port);
-	return rv;
 }
 
 static void octeon_cf_shutdown(struct device *dev)
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index 319b64491b7b..b9bf78b7d48d 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -258,7 +258,7 @@ static struct pci_driver oldpiix_pci_driver = {
 	.id_table		= oldpiix_pci_tbl,
 	.probe			= oldpiix_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index fb042e0519d0..3a944a029264 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -184,7 +184,7 @@ static struct pci_driver opti_pci_driver = {
 	.id_table	= opti,
 	.probe 		= opti_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index bb71ea214b99..bdec7efa4643 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -440,7 +440,7 @@ static struct pci_driver optidma_pci_driver = {
 	.id_table	= optidma,
 	.probe 		= optidma_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index 1151f23177bb..4d06a5cda987 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -62,7 +62,7 @@ enum {
 };
 
 static int pdc2027x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int pdc2027x_reinit_one(struct pci_dev *pdev);
 #endif
 static int pdc2027x_prereset(struct ata_link *link, unsigned long deadline);
@@ -128,7 +128,7 @@ static struct pci_driver pdc2027x_pci_driver = {
 	.id_table		= pdc2027x_pci_tbl,
 	.probe			= pdc2027x_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= pdc2027x_reinit_one,
 #endif
@@ -761,7 +761,7 @@ static int pdc2027x_init_one(struct pci_dev *pdev,
 				 IRQF_SHARED, &pdc2027x_sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int pdc2027x_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index defa050e1784..9001991d2830 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -377,7 +377,7 @@ static struct pci_driver pdc202xx_pci_driver = {
 	.id_table	= pdc202xx,
 	.probe 		= pdc202xx_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_piccolo.c b/drivers/ata/pata_piccolo.c
index 0b46be117051..35cb0e263237 100644
--- a/drivers/ata/pata_piccolo.c
+++ b/drivers/ata/pata_piccolo.c
@@ -110,7 +110,7 @@ static struct pci_driver ata_tosh_pci_driver = {
 	.id_table	= ata_tosh,
 	.probe 		= ata_tosh_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index be3f10240dca..a3f1123d17aa 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -237,7 +237,7 @@ static struct pci_driver radisys_pci_driver = {
 	.id_table		= radisys_pci_tbl,
 	.probe			= radisys_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_rdc.c b/drivers/ata/pata_rdc.c
index 521b2137ea3e..9ce5952216bc 100644
--- a/drivers/ata/pata_rdc.c
+++ b/drivers/ata/pata_rdc.c
@@ -382,7 +382,7 @@ static struct pci_driver rdc_pci_driver = {
 	.id_table		= rdc_pci_tbl,
 	.probe			= rdc_init_one,
 	.remove			= rdc_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index caedc90855b2..b3ec18c6bcc6 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -101,7 +101,7 @@ static int rz1000_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 	return -ENODEV;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int rz1000_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -133,7 +133,7 @@ static struct pci_driver rz1000_pci_driver = {
 	.id_table	= pata_rz1000,
 	.probe 		= rz1000_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= rz1000_reinit_one,
 #endif
diff --git a/drivers/ata/pata_samsung_cf.c b/drivers/ata/pata_samsung_cf.c
index 0610e78c8a2a..fb528831fb92 100644
--- a/drivers/ata/pata_samsung_cf.c
+++ b/drivers/ata/pata_samsung_cf.c
@@ -619,7 +619,7 @@ static int __exit pata_s3c_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int pata_s3c_suspend(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -670,7 +670,7 @@ static struct platform_driver pata_s3c_driver = {
 	.driver		= {
 		.name	= DRV_NAME,
 		.owner	= THIS_MODULE,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 		.pm	= &pata_s3c_pm_ops,
 #endif
 	},
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 96a232fffae6..c71de5d680d1 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -254,7 +254,7 @@ static struct pci_driver sc1200_pci_driver = {
 	.id_table	= sc1200,
 	.probe 		= sc1200_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index f1f5b5ae3382..4e006d74bef8 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -1096,7 +1096,7 @@ static struct pci_driver scc_pci_driver = {
 	.id_table		= scc_pci_tbl,
 	.probe			= scc_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_sch.c b/drivers/ata/pata_sch.c
index 5a1cde0ea360..b920c3407f8b 100644
--- a/drivers/ata/pata_sch.c
+++ b/drivers/ata/pata_sch.c
@@ -64,7 +64,7 @@ static struct pci_driver sch_pci_driver = {
 	.id_table		= sch_pci_tbl,
 	.probe			= sch_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index e27f31fe1b67..fc5f31d4828e 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -436,7 +436,7 @@ static int serverworks_init_one(struct pci_dev *pdev, const struct pci_device_id
 	return ata_pci_bmdma_init_one(pdev, ppi, &serverworks_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int serverworks_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -468,7 +468,7 @@ static struct pci_driver serverworks_pci_driver = {
 	.id_table	= serverworks,
 	.probe 		= serverworks_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= serverworks_reinit_one,
 #endif
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 73fe362d9716..f597edccedec 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -403,7 +403,7 @@ use_ioports:
 	return ata_pci_bmdma_init_one(pdev, ppi, &sil680_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sil680_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -429,7 +429,7 @@ static struct pci_driver sil680_pci_driver = {
 	.id_table	= sil680,
 	.probe 		= sil680_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= sil680_reinit_one,
 #endif
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index 78d913aa93c8..626f989d5c6a 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -869,7 +869,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	return ata_pci_bmdma_init_one(pdev, ppi, &sis_sht, chipset, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sis_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -899,7 +899,7 @@ static struct pci_driver sis_pci_driver = {
 	.id_table		= sis_pci_tbl,
 	.probe			= sis_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= sis_reinit_one,
 #endif
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index 900f0e4a1faf..4935f61f629c 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -337,7 +337,7 @@ static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id
 	return ata_pci_bmdma_init_one(dev, ppi, &sl82c105_sht, NULL, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sl82c105_reinit_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -365,7 +365,7 @@ static struct pci_driver sl82c105_pci_driver = {
 	.id_table	= sl82c105,
 	.probe 		= sl82c105_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= sl82c105_reinit_one,
 #endif
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index 7bc78e264f9e..d9364af8eb63 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -207,7 +207,7 @@ static const struct pci_device_id triflex[] = {
 	{ },
 };
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int triflex_ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -233,7 +233,7 @@ static struct pci_driver triflex_pci_driver = {
 	.id_table	= triflex,
 	.probe 		= triflex_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= triflex_ata_pci_device_suspend,
 	.resume		= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index f6c9632bdff6..1ca6bcab369f 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -659,7 +659,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return ata_pci_bmdma_init_one(pdev, ppi, &via_sht, (void *)config, 0);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 /**
  *	via_reinit_one		-	reinit after resume
  *	@pdev; PCI device
@@ -704,7 +704,7 @@ static struct pci_driver via_pci_driver = {
 	.id_table	= via,
 	.probe 		= via_init_one,
 	.remove		= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= via_reinit_one,
 #endif
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index fb0b40a191c2..616a6d2ac20c 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -774,20 +774,6 @@ static int sata_fsl_port_start(struct ata_port *ap)
 	VPRINTK("HControl = 0x%x\n", ioread32(hcr_base + HCONTROL));
 	VPRINTK("CHBA  = 0x%x\n", ioread32(hcr_base + CHBA));
 
-#ifdef CONFIG_MPC8315_DS
-	/*
-	 * Workaround for 8315DS board 3gbps link-up issue,
-	 * currently limit SATA port to GEN1 speed
-	 */
-	sata_fsl_scr_read(&ap->link, SCR_CONTROL, &temp);
-	temp &= ~(0xF << 4);
-	temp |= (0x1 << 4);
-	sata_fsl_scr_write(&ap->link, SCR_CONTROL, temp);
-
-	sata_fsl_scr_read(&ap->link, SCR_CONTROL, &temp);
-	dev_warn(dev, "scr_control, speed limited to %x\n", temp);
-#endif
-
 	return 0;
 }
 
@@ -1588,7 +1574,7 @@ static int sata_fsl_remove(struct platform_device *ofdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sata_fsl_suspend(struct platform_device *op, pm_message_t state)
 {
 	struct ata_host *host = platform_get_drvdata(op);
@@ -1644,7 +1630,7 @@ static struct platform_driver fsl_sata_driver = {
 	},
 	.probe		= sata_fsl_probe,
 	.remove		= sata_fsl_remove,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= sata_fsl_suspend,
 	.resume		= sata_fsl_resume,
 #endif
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 5c54d957370a..069827826b20 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -785,7 +785,7 @@ static int init_controller(void __iomem *mmio_base, u16 hctl)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int inic_pci_device_resume(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -898,7 +898,7 @@ static const struct pci_device_id inic_pci_tbl[] = {
 static struct pci_driver inic_pci_driver = {
 	.name 		= DRV_NAME,
 	.id_table	= inic_pci_tbl,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend	= ata_pci_device_suspend,
 	.resume		= inic_pci_device_resume,
 #endif
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index 05c8a44adf8e..391cfda1e83f 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -4222,7 +4222,7 @@ static int mv_platform_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int mv_platform_suspend(struct platform_device *pdev, pm_message_t state)
 {
 	struct ata_host *host = platform_get_drvdata(pdev);
@@ -4289,7 +4289,7 @@ static struct platform_driver mv_platform_driver = {
 #ifdef CONFIG_PCI
 static int mv_pci_init_one(struct pci_dev *pdev,
 			   const struct pci_device_id *ent);
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int mv_pci_device_resume(struct pci_dev *pdev);
 #endif
 
@@ -4299,7 +4299,7 @@ static struct pci_driver mv_pci_driver = {
 	.id_table		= mv_pci_tbl,
 	.probe			= mv_pci_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= mv_pci_device_resume,
 #endif
@@ -4457,7 +4457,7 @@ static int mv_pci_init_one(struct pci_dev *pdev,
 				 IS_GEN_I(hpriv) ? &mv5_sht : &mv6_sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int mv_pci_device_resume(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index ba5f27120332..cdf99fac139a 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -295,7 +295,7 @@ struct nv_swncq_port_priv {
 #define NV_ADMA_CHECK_INTR(GCTL, PORT) ((GCTL) & (1 << (19 + (12 * (PORT)))))
 
 static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int nv_pci_device_resume(struct pci_dev *pdev);
 #endif
 static void nv_ck804_host_stop(struct ata_host *host);
@@ -379,7 +379,7 @@ static struct pci_driver nv_pci_driver = {
 	.name			= DRV_NAME,
 	.id_table		= nv_pci_tbl,
 	.probe			= nv_init_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= nv_pci_device_resume,
 #endif
@@ -2431,7 +2431,7 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	return ata_pci_sff_activate_host(host, ipriv->irq_handler, ipriv->sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int nv_pci_device_resume(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
diff --git a/drivers/ata/sata_rcar.c b/drivers/ata/sata_rcar.c
index 2b25bd83fc9d..61eb6d77dac7 100644
--- a/drivers/ata/sata_rcar.c
+++ b/drivers/ata/sata_rcar.c
@@ -937,7 +937,7 @@ static int sata_rcar_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sata_rcar_suspend(struct device *dev)
 {
 	struct ata_host *host = dev_get_drvdata(dev);
@@ -991,7 +991,7 @@ static struct platform_driver sata_rcar_driver = {
 		.name		= DRV_NAME,
 		.owner		= THIS_MODULE,
 		.of_match_table	= sata_rcar_match,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 		.pm		= &sata_rcar_pm_ops,
 #endif
 	},
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 3062f8605b29..40b76b2d18c6 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -112,7 +112,7 @@ enum {
 };
 
 static int sil_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sil_pci_device_resume(struct pci_dev *pdev);
 #endif
 static void sil_dev_config(struct ata_device *dev);
@@ -166,7 +166,7 @@ static struct pci_driver sil_pci_driver = {
 	.id_table		= sil_pci_tbl,
 	.probe			= sil_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= sil_pci_device_resume,
 #endif
@@ -802,7 +802,7 @@ static int sil_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 				 &sil_sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sil_pci_device_resume(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index aa1051ba6d13..0534890f118a 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -353,8 +353,10 @@ static void sil24_error_handler(struct ata_port *ap);
 static void sil24_post_internal_cmd(struct ata_queued_cmd *qc);
 static int sil24_port_start(struct ata_port *ap);
 static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sil24_pci_device_resume(struct pci_dev *pdev);
+#endif
+#ifdef CONFIG_PM
 static int sil24_port_resume(struct ata_port *ap);
 #endif
 
@@ -375,7 +377,7 @@ static struct pci_driver sil24_pci_driver = {
 	.id_table		= sil24_pci_tbl,
 	.probe			= sil24_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= sil24_pci_device_resume,
 #endif
@@ -1350,7 +1352,7 @@ static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 				 &sil24_sht);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sil24_pci_device_resume(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
@@ -1370,7 +1372,9 @@ static int sil24_pci_device_resume(struct pci_dev *pdev)
 
 	return 0;
 }
+#endif
 
+#ifdef CONFIG_PM
 static int sil24_port_resume(struct ata_port *ap)
 {
 	sil24_config_pmp(ap, ap->nr_pmp_links);
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index b513428171b3..d1637ac40a73 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -82,7 +82,7 @@ static struct pci_driver sis_pci_driver = {
 	.id_table		= sis_pci_tbl,
 	.probe			= sis_init_one,
 	.remove			= ata_pci_remove_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index f72e84228c5c..47bf89464cef 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -103,7 +103,7 @@ static struct pci_driver svia_pci_driver = {
 	.name			= DRV_NAME,
 	.id_table		= svia_pci_tbl,
 	.probe			= svia_init_one,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.suspend		= ata_pci_device_suspend,
 	.resume			= ata_pci_device_resume,
 #endif
diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c
index 204814e88e46..d4725fc0395d 100644
--- a/drivers/atm/fore200e.c
+++ b/drivers/atm/fore200e.c
@@ -2780,7 +2780,7 @@ static struct pci_driver fore200e_pca_driver = {
 
 static int __init fore200e_module_init(void)
 {
-	int err;
+	int err = 0;
 
 	printk(FORE200E "FORE Systems 200E-series ATM driver - version " FORE200E_VERSION "\n");
 
diff --git a/drivers/atm/idt77252.c b/drivers/atm/idt77252.c
index 1bdf104e90bb..b621f56a36be 100644
--- a/drivers/atm/idt77252.c
+++ b/drivers/atm/idt77252.c
@@ -2551,12 +2551,12 @@ done:
 		timeout = 5 * 1000;
 		while (atomic_read(&vc->scq->used) > 0) {
 			timeout = msleep_interruptible(timeout);
-			if (!timeout)
+			if (!timeout) {
+				pr_warn("%s: SCQ drain timeout: %u used\n",
+					card->name, atomic_read(&vc->scq->used));
 				break;
+			}
 		}
-		if (!timeout)
-			printk("%s: SCQ drain timeout: %u used\n",
-			       card->name, atomic_read(&vc->scq->used));
 
 		writel(TCMDQ_HALT | vc->index, SAR_REG_TCMDQ);
 		clear_scd(card, vc->scq, vc->class);
diff --git a/drivers/base/dma-contiguous.c b/drivers/base/dma-contiguous.c
index 83969f8c5727..6467c919c509 100644
--- a/drivers/base/dma-contiguous.c
+++ b/drivers/base/dma-contiguous.c
@@ -176,14 +176,24 @@ static int __init cma_activate_area(struct cma *cma)
 		base_pfn = pfn;
 		for (j = pageblock_nr_pages; j; --j, pfn++) {
 			WARN_ON_ONCE(!pfn_valid(pfn));
+			/*
+			 * alloc_contig_range requires the pfn range
+			 * specified to be in the same zone. Make this
+			 * simple by forcing the entire CMA resv range
+			 * to be in the same zone.
+			 */
 			if (page_zone(pfn_to_page(pfn)) != zone)
-				return -EINVAL;
+				goto err;
 		}
 		init_cma_reserved_pageblock(pfn_to_page(base_pfn));
 	} while (--i);
 
 	mutex_init(&cma->lock);
 	return 0;
+
+err:
+	kfree(cma->bitmap);
+	return -EINVAL;
 }
 
 static struct cma cma_areas[MAX_CMA_AREAS];
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 343ffad59377..bf412961a934 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -214,9 +214,6 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime,
 		pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev),
 			error, (unsigned long long)nsecs >> 10);
 	}
-
-	trace_device_pm_report_time(dev, info, nsecs, pm_verb(state.event),
-				    error);
 }
 
 /**
@@ -387,7 +384,9 @@ static int dpm_run_callback(pm_callback_t cb, struct device *dev,
 	calltime = initcall_debug_start(dev);
 
 	pm_dev_dbg(dev, state, info);
+	trace_device_pm_callback_start(dev, info, state.event);
 	error = cb(dev);
+	trace_device_pm_callback_end(dev, error);
 	suspend_report_result(cb, error);
 
 	initcall_debug_report(dev, calltime, error, state, info);
@@ -545,6 +544,7 @@ static void dpm_resume_noirq(pm_message_t state)
 	struct device *dev;
 	ktime_t starttime = ktime_get();
 
+	trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, true);
 	mutex_lock(&dpm_list_mtx);
 	pm_transition = state;
 
@@ -587,6 +587,7 @@ static void dpm_resume_noirq(pm_message_t state)
 	dpm_show_time(starttime, state, "noirq");
 	resume_device_irqs();
 	cpuidle_resume();
+	trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, false);
 }
 
 /**
@@ -664,6 +665,7 @@ static void dpm_resume_early(pm_message_t state)
 	struct device *dev;
 	ktime_t starttime = ktime_get();
 
+	trace_suspend_resume(TPS("dpm_resume_early"), state.event, true);
 	mutex_lock(&dpm_list_mtx);
 	pm_transition = state;
 
@@ -703,6 +705,7 @@ static void dpm_resume_early(pm_message_t state)
 	mutex_unlock(&dpm_list_mtx);
 	async_synchronize_full();
 	dpm_show_time(starttime, state, "early");
+	trace_suspend_resume(TPS("dpm_resume_early"), state.event, false);
 }
 
 /**
@@ -834,6 +837,7 @@ void dpm_resume(pm_message_t state)
 	struct device *dev;
 	ktime_t starttime = ktime_get();
 
+	trace_suspend_resume(TPS("dpm_resume"), state.event, true);
 	might_sleep();
 
 	mutex_lock(&dpm_list_mtx);
@@ -875,6 +879,7 @@ void dpm_resume(pm_message_t state)
 	dpm_show_time(starttime, state, NULL);
 
 	cpufreq_resume();
+	trace_suspend_resume(TPS("dpm_resume"), state.event, false);
 }
 
 /**
@@ -913,7 +918,9 @@ static void device_complete(struct device *dev, pm_message_t state)
 
 	if (callback) {
 		pm_dev_dbg(dev, state, info);
+		trace_device_pm_callback_start(dev, info, state.event);
 		callback(dev);
+		trace_device_pm_callback_end(dev, 0);
 	}
 
 	device_unlock(dev);
@@ -932,6 +939,7 @@ void dpm_complete(pm_message_t state)
 {
 	struct list_head list;
 
+	trace_suspend_resume(TPS("dpm_complete"), state.event, true);
 	might_sleep();
 
 	INIT_LIST_HEAD(&list);
@@ -951,6 +959,7 @@ void dpm_complete(pm_message_t state)
 	}
 	list_splice(&list, &dpm_list);
 	mutex_unlock(&dpm_list_mtx);
+	trace_suspend_resume(TPS("dpm_complete"), state.event, false);
 }
 
 /**
@@ -1086,6 +1095,7 @@ static int dpm_suspend_noirq(pm_message_t state)
 	ktime_t starttime = ktime_get();
 	int error = 0;
 
+	trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, true);
 	cpuidle_pause();
 	suspend_device_irqs();
 	mutex_lock(&dpm_list_mtx);
@@ -1126,6 +1136,7 @@ static int dpm_suspend_noirq(pm_message_t state)
 	} else {
 		dpm_show_time(starttime, state, "noirq");
 	}
+	trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, false);
 	return error;
 }
 
@@ -1222,6 +1233,7 @@ static int dpm_suspend_late(pm_message_t state)
 	ktime_t starttime = ktime_get();
 	int error = 0;
 
+	trace_suspend_resume(TPS("dpm_suspend_late"), state.event, true);
 	mutex_lock(&dpm_list_mtx);
 	pm_transition = state;
 	async_error = 0;
@@ -1257,6 +1269,7 @@ static int dpm_suspend_late(pm_message_t state)
 	} else {
 		dpm_show_time(starttime, state, "late");
 	}
+	trace_suspend_resume(TPS("dpm_suspend_late"), state.event, false);
 	return error;
 }
 
@@ -1295,7 +1308,9 @@ static int legacy_suspend(struct device *dev, pm_message_t state,
 
 	calltime = initcall_debug_start(dev);
 
+	trace_device_pm_callback_start(dev, info, state.event);
 	error = cb(dev, state);
+	trace_device_pm_callback_end(dev, error);
 	suspend_report_result(cb, error);
 
 	initcall_debug_report(dev, calltime, error, state, info);
@@ -1461,6 +1476,7 @@ int dpm_suspend(pm_message_t state)
 	ktime_t starttime = ktime_get();
 	int error = 0;
 
+	trace_suspend_resume(TPS("dpm_suspend"), state.event, true);
 	might_sleep();
 
 	cpufreq_suspend();
@@ -1498,6 +1514,7 @@ int dpm_suspend(pm_message_t state)
 		dpm_save_failed_step(SUSPEND_SUSPEND);
 	} else
 		dpm_show_time(starttime, state, NULL);
+	trace_suspend_resume(TPS("dpm_suspend"), state.event, false);
 	return error;
 }
 
@@ -1549,8 +1566,11 @@ static int device_prepare(struct device *dev, pm_message_t state)
 		callback = dev->driver->pm->prepare;
 	}
 
-	if (callback)
+	if (callback) {
+		trace_device_pm_callback_start(dev, info, state.event);
 		ret = callback(dev);
+		trace_device_pm_callback_end(dev, ret);
+	}
 
 	device_unlock(dev);
 
@@ -1582,6 +1602,7 @@ int dpm_prepare(pm_message_t state)
 {
 	int error = 0;
 
+	trace_suspend_resume(TPS("dpm_prepare"), state.event, true);
 	might_sleep();
 
 	mutex_lock(&dpm_list_mtx);
@@ -1612,6 +1633,7 @@ int dpm_prepare(pm_message_t state)
 		put_device(dev);
 	}
 	mutex_unlock(&dpm_list_mtx);
+	trace_suspend_resume(TPS("dpm_prepare"), state.event, false);
 	return error;
 }
 
diff --git a/drivers/base/syscore.c b/drivers/base/syscore.c
index e8d11b6630ee..dbb8350ea8dc 100644
--- a/drivers/base/syscore.c
+++ b/drivers/base/syscore.c
@@ -10,6 +10,7 @@
 #include <linux/mutex.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
+#include <trace/events/power.h>
 
 static LIST_HEAD(syscore_ops_list);
 static DEFINE_MUTEX(syscore_ops_lock);
@@ -49,6 +50,7 @@ int syscore_suspend(void)
 	struct syscore_ops *ops;
 	int ret = 0;
 
+	trace_suspend_resume(TPS("syscore_suspend"), 0, true);
 	pr_debug("Checking wakeup interrupts\n");
 
 	/* Return error code if there are any wakeup interrupts pending. */
@@ -70,6 +72,7 @@ int syscore_suspend(void)
 				"Interrupts enabled after %pF\n", ops->suspend);
 		}
 
+	trace_suspend_resume(TPS("syscore_suspend"), 0, false);
 	return 0;
 
  err_out:
@@ -92,6 +95,7 @@ void syscore_resume(void)
 {
 	struct syscore_ops *ops;
 
+	trace_suspend_resume(TPS("syscore_resume"), 0, true);
 	WARN_ONCE(!irqs_disabled(),
 		"Interrupts enabled before system core resume.\n");
 
@@ -103,6 +107,7 @@ void syscore_resume(void)
 			WARN_ONCE(!irqs_disabled(),
 				"Interrupts enabled after %pF\n", ops->resume);
 		}
+	trace_suspend_resume(TPS("syscore_resume"), 0, false);
 }
 EXPORT_SYMBOL_GPL(syscore_resume);
 #endif /* CONFIG_PM_SLEEP */
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index b6c8aaf4931b..5b17ec88ea05 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -1337,8 +1337,11 @@ int drbd_submit_peer_request(struct drbd_device *device,
 		return 0;
 	}
 
+	/* Discards don't have any payload.
+	 * But the scsi layer still expects a bio_vec it can use internally,
+	 * see sd_setup_discard_cmnd() and blk_add_request_payload(). */
 	if (peer_req->flags & EE_IS_TRIM)
-		nr_pages = 0; /* discards don't have any payload. */
+		nr_pages = 1;
 
 	/* In most cases, we will only need one bio.  But in case the lower
 	 * level restrictions happen to be different at this offset on this
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 677db049f55a..56d46ffb08e1 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -3777,7 +3777,7 @@ static void floppy_rb0_cb(struct bio *bio, int err)
 	int drive = cbdata->drive;
 
 	if (err) {
-		pr_info("floppy: error %d while reading block 0", err);
+		pr_info("floppy: error %d while reading block 0\n", err);
 		set_bit(FD_OPEN_SHOULD_FAIL_BIT, &UDRS->flags);
 	}
 	complete(&cbdata->complete);
diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 74abd49fabdc..295f3afbbef5 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -39,6 +39,7 @@
 #include <../drivers/ata/ahci.h>
 #include <linux/export.h>
 #include <linux/debugfs.h>
+#include <linux/prefetch.h>
 #include "mtip32xx.h"
 
 #define HW_CMD_SLOT_SZ		(MTIP_MAX_COMMAND_SLOTS * 32)
@@ -2380,6 +2381,8 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	/* Map the scatter list for DMA access */
 	nents = dma_map_sg(&dd->pdev->dev, command->sg, nents, dma_dir);
 
+	prefetch(&port->flags);
+
 	command->scatter_ents = nents;
 
 	/*
@@ -2392,7 +2395,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	fis = command->command;
 	fis->type        = 0x27;
 	fis->opts        = 1 << 7;
-	if (rq_data_dir(rq) == READ)
+	if (dma_dir == DMA_FROM_DEVICE)
 		fis->command = ATA_CMD_FPDMA_READ;
 	else
 		fis->command = ATA_CMD_FPDMA_WRITE;
@@ -2412,7 +2415,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	fis->res3        = 0;
 	fill_command_sg(dd, command, nents);
 
-	if (command->unaligned)
+	if (unlikely(command->unaligned))
 		fis->device |= 1 << 7;
 
 	/* Populate the command header */
@@ -2433,7 +2436,7 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
 	 * To prevent this command from being issued
 	 * if an internal command is in progress or error handling is active.
 	 */
-	if (port->flags & MTIP_PF_PAUSE_IO) {
+	if (unlikely(port->flags & MTIP_PF_PAUSE_IO)) {
 		set_bit(rq->tag, port->cmds_to_issue);
 		set_bit(MTIP_PF_ISSUE_CMDS_BIT, &port->flags);
 		return;
@@ -3754,7 +3757,7 @@ static bool mtip_check_unal_depth(struct blk_mq_hw_ctx *hctx,
 	struct driver_data *dd = hctx->queue->queuedata;
 	struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
 
-	if (!dd->unal_qdepth || rq_data_dir(rq) == READ)
+	if (rq_data_dir(rq) == READ || !dd->unal_qdepth)
 		return false;
 
 	/*
@@ -3776,11 +3779,11 @@ static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq)
 {
 	int ret;
 
-	if (mtip_check_unal_depth(hctx, rq))
+	if (unlikely(mtip_check_unal_depth(hctx, rq)))
 		return BLK_MQ_RQ_QUEUE_BUSY;
 
 	ret = mtip_submit_request(hctx, rq);
-	if (!ret)
+	if (likely(!ret))
 		return BLK_MQ_RQ_QUEUE_OK;
 
 	rq->errors = ret;
diff --git a/drivers/block/mtip32xx/mtip32xx.h b/drivers/block/mtip32xx/mtip32xx.h
index 4b9b554234bc..ba1b31ee22ec 100644
--- a/drivers/block/mtip32xx/mtip32xx.h
+++ b/drivers/block/mtip32xx/mtip32xx.h
@@ -493,19 +493,19 @@ struct driver_data {
 
 	struct workqueue_struct *isr_workq;
 
-	struct mtip_work work[MTIP_MAX_SLOT_GROUPS];
-
 	atomic_t irq_workers_active;
 
+	struct mtip_work work[MTIP_MAX_SLOT_GROUPS];
+
 	int isr_binding;
 
 	struct block_device *bdev;
 
-	int unal_qdepth; /* qdepth of unaligned IO queue */
-
 	struct list_head online_list; /* linkage for online list */
 
 	struct list_head remove_list; /* linkage for removing list */
+
+	int unal_qdepth; /* qdepth of unaligned IO queue */
 };
 
 #endif
diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c
index 77087a29b127..a3b042c4d448 100644
--- a/drivers/block/null_blk.c
+++ b/drivers/block/null_blk.c
@@ -79,7 +79,7 @@ MODULE_PARM_DESC(home_node, "Home node for the device");
 
 static int queue_mode = NULL_Q_MQ;
 module_param(queue_mode, int, S_IRUGO);
-MODULE_PARM_DESC(use_mq, "Use blk-mq interface (0=bio,1=rq,2=multiqueue)");
+MODULE_PARM_DESC(queue_mode, "Block interface to use (0=bio,1=rq,2=multiqueue)");
 
 static int gb = 250;
 module_param(gb, int, S_IRUGO);
@@ -227,7 +227,10 @@ static void null_cmd_end_timer(struct nullb_cmd *cmd)
 
 static void null_softirq_done_fn(struct request *rq)
 {
-	end_cmd(blk_mq_rq_to_pdu(rq));
+	if (queue_mode == NULL_Q_MQ)
+		end_cmd(blk_mq_rq_to_pdu(rq));
+	else
+		end_cmd(rq->special);
 }
 
 static inline void null_handle_cmd(struct nullb_cmd *cmd)
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index a842c71dcc21..02351e217165 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -10,10 +10,6 @@
  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
  */
 
 #include <linux/nvme.h>
@@ -46,16 +42,26 @@
 #include <scsi/sg.h>
 #include <asm-generic/io-64-nonatomic-lo-hi.h>
 
-#define NVME_Q_DEPTH 1024
+#include <trace/events/block.h>
+
+#define NVME_Q_DEPTH		1024
 #define SQ_SIZE(depth)		(depth * sizeof(struct nvme_command))
 #define CQ_SIZE(depth)		(depth * sizeof(struct nvme_completion))
-#define ADMIN_TIMEOUT	(60 * HZ)
-#define IOD_TIMEOUT	(4 * NVME_IO_TIMEOUT)
+#define ADMIN_TIMEOUT		(admin_timeout * HZ)
+#define IOD_TIMEOUT		(retry_time * HZ)
+
+static unsigned char admin_timeout = 60;
+module_param(admin_timeout, byte, 0644);
+MODULE_PARM_DESC(admin_timeout, "timeout in seconds for admin commands");
 
-unsigned char io_timeout = 30;
-module_param(io_timeout, byte, 0644);
+unsigned char nvme_io_timeout = 30;
+module_param_named(io_timeout, nvme_io_timeout, byte, 0644);
 MODULE_PARM_DESC(io_timeout, "timeout in seconds for I/O");
 
+static unsigned char retry_time = 30;
+module_param(retry_time, byte, 0644);
+MODULE_PARM_DESC(retry_time, "time in seconds to retry failed I/O");
+
 static int nvme_major;
 module_param(nvme_major, int, 0);
 
@@ -67,6 +73,7 @@ static LIST_HEAD(dev_list);
 static struct task_struct *nvme_thread;
 static struct workqueue_struct *nvme_workq;
 static wait_queue_head_t nvme_kthread_wait;
+static struct notifier_block nvme_nb;
 
 static void nvme_reset_failed_dev(struct work_struct *ws);
 
@@ -199,16 +206,13 @@ static int alloc_cmdid_killable(struct nvme_queue *nvmeq, void *ctx,
 #define CMD_CTX_CANCELLED	(0x30C + CMD_CTX_BASE)
 #define CMD_CTX_COMPLETED	(0x310 + CMD_CTX_BASE)
 #define CMD_CTX_INVALID		(0x314 + CMD_CTX_BASE)
-#define CMD_CTX_FLUSH		(0x318 + CMD_CTX_BASE)
-#define CMD_CTX_ABORT		(0x31C + CMD_CTX_BASE)
+#define CMD_CTX_ABORT		(0x318 + CMD_CTX_BASE)
 
 static void special_completion(struct nvme_queue *nvmeq, void *ctx,
 						struct nvme_completion *cqe)
 {
 	if (ctx == CMD_CTX_CANCELLED)
 		return;
-	if (ctx == CMD_CTX_FLUSH)
-		return;
 	if (ctx == CMD_CTX_ABORT) {
 		++nvmeq->dev->abort_limit;
 		return;
@@ -247,8 +251,9 @@ static void *free_cmdid(struct nvme_queue *nvmeq, int cmdid,
 	void *ctx;
 	struct nvme_cmd_info *info = nvme_cmd_info(nvmeq);
 
-	if (cmdid >= nvmeq->q_depth) {
-		*fn = special_completion;
+	if (cmdid >= nvmeq->q_depth || !info[cmdid].fn) {
+		if (fn)
+			*fn = special_completion;
 		return CMD_CTX_INVALID;
 	}
 	if (fn)
@@ -281,9 +286,17 @@ static struct nvme_queue *raw_nvmeq(struct nvme_dev *dev, int qid)
 
 static struct nvme_queue *get_nvmeq(struct nvme_dev *dev) __acquires(RCU)
 {
+	struct nvme_queue *nvmeq;
 	unsigned queue_id = get_cpu_var(*dev->io_queue);
+
 	rcu_read_lock();
-	return rcu_dereference(dev->queues[queue_id]);
+	nvmeq = rcu_dereference(dev->queues[queue_id]);
+	if (nvmeq)
+		return nvmeq;
+
+	rcu_read_unlock();
+	put_cpu_var(*dev->io_queue);
+	return NULL;
 }
 
 static void put_nvmeq(struct nvme_queue *nvmeq) __releases(RCU)
@@ -295,8 +308,15 @@ static void put_nvmeq(struct nvme_queue *nvmeq) __releases(RCU)
 static struct nvme_queue *lock_nvmeq(struct nvme_dev *dev, int q_idx)
 							__acquires(RCU)
 {
+	struct nvme_queue *nvmeq;
+
 	rcu_read_lock();
-	return rcu_dereference(dev->queues[q_idx]);
+	nvmeq = rcu_dereference(dev->queues[q_idx]);
+	if (nvmeq)
+		return nvmeq;
+
+	rcu_read_unlock();
+	return NULL;
 }
 
 static void unlock_nvmeq(struct nvme_queue *nvmeq) __releases(RCU)
@@ -387,25 +407,30 @@ void nvme_free_iod(struct nvme_dev *dev, struct nvme_iod *iod)
 static void nvme_start_io_acct(struct bio *bio)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
-	const int rw = bio_data_dir(bio);
-	int cpu = part_stat_lock();
-	part_round_stats(cpu, &disk->part0);
-	part_stat_inc(cpu, &disk->part0, ios[rw]);
-	part_stat_add(cpu, &disk->part0, sectors[rw], bio_sectors(bio));
-	part_inc_in_flight(&disk->part0, rw);
-	part_stat_unlock();
+	if (blk_queue_io_stat(disk->queue)) {
+		const int rw = bio_data_dir(bio);
+		int cpu = part_stat_lock();
+		part_round_stats(cpu, &disk->part0);
+		part_stat_inc(cpu, &disk->part0, ios[rw]);
+		part_stat_add(cpu, &disk->part0, sectors[rw],
+							bio_sectors(bio));
+		part_inc_in_flight(&disk->part0, rw);
+		part_stat_unlock();
+	}
 }
 
 static void nvme_end_io_acct(struct bio *bio, unsigned long start_time)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
-	const int rw = bio_data_dir(bio);
-	unsigned long duration = jiffies - start_time;
-	int cpu = part_stat_lock();
-	part_stat_add(cpu, &disk->part0, ticks[rw], duration);
-	part_round_stats(cpu, &disk->part0);
-	part_dec_in_flight(&disk->part0, rw);
-	part_stat_unlock();
+	if (blk_queue_io_stat(disk->queue)) {
+		const int rw = bio_data_dir(bio);
+		unsigned long duration = jiffies - start_time;
+		int cpu = part_stat_lock();
+		part_stat_add(cpu, &disk->part0, ticks[rw], duration);
+		part_round_stats(cpu, &disk->part0);
+		part_dec_in_flight(&disk->part0, rw);
+		part_stat_unlock();
+	}
 }
 
 static void bio_completion(struct nvme_queue *nvmeq, void *ctx,
@@ -414,6 +439,7 @@ static void bio_completion(struct nvme_queue *nvmeq, void *ctx,
 	struct nvme_iod *iod = ctx;
 	struct bio *bio = iod->private;
 	u16 status = le16_to_cpup(&cqe->status) >> 1;
+	int error = 0;
 
 	if (unlikely(status)) {
 		if (!(status & NVME_SC_DNR ||
@@ -426,6 +452,7 @@ static void bio_completion(struct nvme_queue *nvmeq, void *ctx,
 			wake_up(&nvmeq->sq_full);
 			return;
 		}
+		error = -EIO;
 	}
 	if (iod->nents) {
 		dma_unmap_sg(nvmeq->q_dmadev, iod->sg, iod->nents,
@@ -433,10 +460,9 @@ static void bio_completion(struct nvme_queue *nvmeq, void *ctx,
 		nvme_end_io_acct(bio, iod->start_time);
 	}
 	nvme_free_iod(nvmeq->dev, iod);
-	if (status)
-		bio_endio(bio, -EIO);
-	else
-		bio_endio(bio, 0);
+
+	trace_block_bio_complete(bdev_get_queue(bio->bi_bdev), bio, error);
+	bio_endio(bio, error);
 }
 
 /* length is in bytes.  gfp flags indicates whether we may sleep. */
@@ -525,6 +551,8 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
 	if (!split)
 		return -ENOMEM;
 
+	trace_block_split(bdev_get_queue(bio->bi_bdev), bio,
+					split->bi_iter.bi_sector);
 	bio_chain(split, bio);
 
 	if (!waitqueue_active(&nvmeq->sq_full))
@@ -627,16 +655,6 @@ static int nvme_submit_flush(struct nvme_queue *nvmeq, struct nvme_ns *ns,
 	return 0;
 }
 
-int nvme_submit_flush_data(struct nvme_queue *nvmeq, struct nvme_ns *ns)
-{
-	int cmdid = alloc_cmdid(nvmeq, (void *)CMD_CTX_FLUSH,
-					special_completion, NVME_IO_TIMEOUT);
-	if (unlikely(cmdid < 0))
-		return cmdid;
-
-	return nvme_submit_flush(nvmeq, ns, cmdid);
-}
-
 static int nvme_submit_iod(struct nvme_queue *nvmeq, struct nvme_iod *iod)
 {
 	struct bio *bio = iod->private;
@@ -652,7 +670,7 @@ static int nvme_submit_iod(struct nvme_queue *nvmeq, struct nvme_iod *iod)
 
 	if (bio->bi_rw & REQ_DISCARD)
 		return nvme_submit_discard(nvmeq, ns, bio, iod, cmdid);
-	if ((bio->bi_rw & REQ_FLUSH) && !iod->nents)
+	if (bio->bi_rw & REQ_FLUSH)
 		return nvme_submit_flush(nvmeq, ns, cmdid);
 
 	control = 0;
@@ -686,6 +704,26 @@ static int nvme_submit_iod(struct nvme_queue *nvmeq, struct nvme_iod *iod)
 	return 0;
 }
 
+static int nvme_split_flush_data(struct nvme_queue *nvmeq, struct bio *bio)
+{
+	struct bio *split = bio_clone(bio, GFP_ATOMIC);
+	if (!split)
+		return -ENOMEM;
+
+	split->bi_iter.bi_size = 0;
+	split->bi_phys_segments = 0;
+	bio->bi_rw &= ~REQ_FLUSH;
+	bio_chain(split, bio);
+
+	if (!waitqueue_active(&nvmeq->sq_full))
+		add_wait_queue(&nvmeq->sq_full, &nvmeq->sq_cong_wait);
+	bio_list_add(&nvmeq->sq_cong, split);
+	bio_list_add(&nvmeq->sq_cong, bio);
+	wake_up_process(nvme_thread);
+
+	return 0;
+}
+
 /*
  * Called with local interrupts disabled and the q_lock held.  May not sleep.
  */
@@ -696,11 +734,8 @@ static int nvme_submit_bio_queue(struct nvme_queue *nvmeq, struct nvme_ns *ns,
 	int psegs = bio_phys_segments(ns->queue, bio);
 	int result;
 
-	if ((bio->bi_rw & REQ_FLUSH) && psegs) {
-		result = nvme_submit_flush_data(nvmeq, ns);
-		if (result)
-			return result;
-	}
+	if ((bio->bi_rw & REQ_FLUSH) && psegs)
+		return nvme_split_flush_data(nvmeq, bio);
 
 	iod = nvme_alloc_iod(psegs, bio->bi_iter.bi_size, GFP_ATOMIC);
 	if (!iod)
@@ -795,7 +830,6 @@ static void nvme_make_request(struct request_queue *q, struct bio *bio)
 	int result = -EBUSY;
 
 	if (!nvmeq) {
-		put_nvmeq(NULL);
 		bio_endio(bio, -EIO);
 		return;
 	}
@@ -870,10 +904,8 @@ static int nvme_submit_sync_cmd(struct nvme_dev *dev, int q_idx,
 	struct nvme_queue *nvmeq;
 
 	nvmeq = lock_nvmeq(dev, q_idx);
-	if (!nvmeq) {
-		unlock_nvmeq(nvmeq);
+	if (!nvmeq)
 		return -ENODEV;
-	}
 
 	cmdinfo.task = current;
 	cmdinfo.status = -EINTR;
@@ -898,9 +930,10 @@ static int nvme_submit_sync_cmd(struct nvme_dev *dev, int q_idx,
 
 	if (cmdinfo.status == -EINTR) {
 		nvmeq = lock_nvmeq(dev, q_idx);
-		if (nvmeq)
+		if (nvmeq) {
 			nvme_abort_command(nvmeq, cmdid);
-		unlock_nvmeq(nvmeq);
+			unlock_nvmeq(nvmeq);
+		}
 		return -EINTR;
 	}
 
@@ -1358,7 +1391,8 @@ static int nvme_wait_ready(struct nvme_dev *dev, u64 cap, bool enabled)
 			return -EINTR;
 		if (time_after(jiffies, timeout)) {
 			dev_err(&dev->pci_dev->dev,
-				"Device not ready; aborting initialisation\n");
+				"Device not ready; aborting %s\n", enabled ?
+						"initialisation" : "reset");
 			return -ENODEV;
 		}
 	}
@@ -1481,7 +1515,11 @@ struct nvme_iod *nvme_map_user_pages(struct nvme_dev *dev, int write,
 		goto put_pages;
 	}
 
+	err = -ENOMEM;
 	iod = nvme_alloc_iod(count, length, GFP_KERNEL);
+	if (!iod)
+		goto put_pages;
+
 	sg = iod->sg;
 	sg_init_table(sg, count);
 	for (i = 0; i < count; i++) {
@@ -1494,7 +1532,6 @@ struct nvme_iod *nvme_map_user_pages(struct nvme_dev *dev, int write,
 	sg_mark_end(&sg[i - 1]);
 	iod->nents = count;
 
-	err = -ENOMEM;
 	nents = dma_map_sg(&dev->pci_dev->dev, sg, count,
 				write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
 	if (!nents)
@@ -1894,6 +1931,8 @@ static struct nvme_ns *nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid,
 	blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift);
 	if (dev->max_hw_sectors)
 		blk_queue_max_hw_sectors(ns->queue, dev->max_hw_sectors);
+	if (dev->vwc & NVME_CTRL_VWC_PRESENT)
+		blk_queue_flush(ns->queue, REQ_FLUSH | REQ_FUA);
 
 	disk->major = nvme_major;
 	disk->first_minor = 0;
@@ -2062,8 +2101,13 @@ static int set_queue_count(struct nvme_dev *dev, int count)
 
 	status = nvme_set_features(dev, NVME_FEAT_NUM_QUEUES, q_count, 0,
 								&result);
-	if (status)
-		return status < 0 ? -EIO : -EBUSY;
+	if (status < 0)
+		return status;
+	if (status > 0) {
+		dev_err(&dev->pci_dev->dev, "Could not set queue count (%d)\n",
+									status);
+		return -EBUSY;
+	}
 	return min(result & 0xffff, result >> 16) + 1;
 }
 
@@ -2072,14 +2116,25 @@ static size_t db_bar_size(struct nvme_dev *dev, unsigned nr_io_queues)
 	return 4096 + ((nr_io_queues + 1) * 8 * dev->db_stride);
 }
 
+static void nvme_cpu_workfn(struct work_struct *work)
+{
+	struct nvme_dev *dev = container_of(work, struct nvme_dev, cpu_work);
+	if (dev->initialized)
+		nvme_assign_io_queues(dev);
+}
+
 static int nvme_cpu_notify(struct notifier_block *self,
 				unsigned long action, void *hcpu)
 {
-	struct nvme_dev *dev = container_of(self, struct nvme_dev, nb);
+	struct nvme_dev *dev;
+
 	switch (action) {
 	case CPU_ONLINE:
 	case CPU_DEAD:
-		nvme_assign_io_queues(dev);
+		spin_lock(&dev_list_lock);
+		list_for_each_entry(dev, &dev_list, node)
+			schedule_work(&dev->cpu_work);
+		spin_unlock(&dev_list_lock);
 		break;
 	}
 	return NOTIFY_OK;
@@ -2148,11 +2203,6 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
 	nvme_free_queues(dev, nr_io_queues + 1);
 	nvme_assign_io_queues(dev);
 
-	dev->nb.notifier_call = &nvme_cpu_notify;
-	result = register_hotcpu_notifier(&dev->nb);
-	if (result)
-		goto free_queues;
-
 	return 0;
 
  free_queues:
@@ -2184,6 +2234,7 @@ static int nvme_dev_add(struct nvme_dev *dev)
 
 	res = nvme_identify(dev, 0, 1, dma_addr);
 	if (res) {
+		dev_err(&pdev->dev, "Identify Controller failed (%d)\n", res);
 		res = -EIO;
 		goto out;
 	}
@@ -2192,6 +2243,7 @@ static int nvme_dev_add(struct nvme_dev *dev)
 	nn = le32_to_cpup(&ctrl->nn);
 	dev->oncs = le16_to_cpup(&ctrl->oncs);
 	dev->abort_limit = ctrl->acl + 1;
+	dev->vwc = ctrl->vwc;
 	memcpy(dev->serial, ctrl->sn, sizeof(ctrl->sn));
 	memcpy(dev->model, ctrl->mn, sizeof(ctrl->mn));
 	memcpy(dev->firmware_rev, ctrl->fr, sizeof(ctrl->fr));
@@ -2450,8 +2502,6 @@ static void nvme_dev_shutdown(struct nvme_dev *dev)
 	int i;
 
 	dev->initialized = 0;
-	unregister_hotcpu_notifier(&dev->nb);
-
 	nvme_dev_list_remove(dev);
 
 	if (!dev->bar || (dev->bar && readl(&dev->bar->csts) == -1)) {
@@ -2722,6 +2772,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	INIT_LIST_HEAD(&dev->namespaces);
 	dev->reset_workfn = nvme_reset_failed_dev;
 	INIT_WORK(&dev->reset_work, nvme_reset_workfn);
+	INIT_WORK(&dev->cpu_work, nvme_cpu_workfn);
 	dev->pci_dev = pdev;
 	pci_set_drvdata(pdev, dev);
 	result = nvme_set_instance(dev);
@@ -2801,6 +2852,7 @@ static void nvme_remove(struct pci_dev *pdev)
 
 	pci_set_drvdata(pdev, NULL);
 	flush_work(&dev->reset_work);
+	flush_work(&dev->cpu_work);
 	misc_deregister(&dev->miscdev);
 	nvme_dev_remove(dev);
 	nvme_dev_shutdown(dev);
@@ -2889,11 +2941,18 @@ static int __init nvme_init(void)
 	else if (result > 0)
 		nvme_major = result;
 
-	result = pci_register_driver(&nvme_driver);
+	nvme_nb.notifier_call = &nvme_cpu_notify;
+	result = register_hotcpu_notifier(&nvme_nb);
 	if (result)
 		goto unregister_blkdev;
+
+	result = pci_register_driver(&nvme_driver);
+	if (result)
+		goto unregister_hotcpu;
 	return 0;
 
+ unregister_hotcpu:
+	unregister_hotcpu_notifier(&nvme_nb);
  unregister_blkdev:
 	unregister_blkdev(nvme_major, "nvme");
  kill_workq:
@@ -2904,9 +2963,11 @@ static int __init nvme_init(void)
 static void __exit nvme_exit(void)
 {
 	pci_unregister_driver(&nvme_driver);
+	unregister_hotcpu_notifier(&nvme_nb);
 	unregister_blkdev(nvme_major, "nvme");
 	destroy_workqueue(nvme_workq);
 	BUG_ON(nvme_thread && !IS_ERR(nvme_thread));
+	_nvme_check_size();
 }
 
 MODULE_AUTHOR("Matthew Wilcox <willy@linux.intel.com>");
diff --git a/drivers/block/nvme-scsi.c b/drivers/block/nvme-scsi.c
index 2c3f5be06da1..a4cd6d691c63 100644
--- a/drivers/block/nvme-scsi.c
+++ b/drivers/block/nvme-scsi.c
@@ -1,6 +1,6 @@
 /*
  * NVM Express device driver
- * Copyright (c) 2011, Intel Corporation.
+ * Copyright (c) 2011-2014, Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -10,10 +10,6 @@
  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
  */
 
 /*
@@ -243,8 +239,6 @@ static int sg_version_num = 30534;	/* 2 digits for each component */
 #define READ_CAP_16_RESP_SIZE				32
 
 /* NVMe Namespace and Command Defines */
-#define NVME_GET_SMART_LOG_PAGE				0x02
-#define NVME_GET_FEAT_TEMP_THRESH			0x04
 #define BYTES_TO_DWORDS					4
 #define NVME_MAX_FIRMWARE_SLOT				7
 
@@ -686,6 +680,7 @@ static int nvme_trans_standard_inquiry_page(struct nvme_ns *ns,
 	u8 resp_data_format = 0x02;
 	u8 protect;
 	u8 cmdque = 0x01 << 1;
+	u8 fw_offset = sizeof(dev->firmware_rev);
 
 	mem = dma_alloc_coherent(&dev->pci_dev->dev, sizeof(struct nvme_id_ns),
 				&dma_addr, GFP_KERNEL);
@@ -721,7 +716,11 @@ static int nvme_trans_standard_inquiry_page(struct nvme_ns *ns,
 	inq_response[7] = cmdque;	/* wbus16=0 | sync=0 | vs=0 */
 	strncpy(&inq_response[8], "NVMe    ", 8);
 	strncpy(&inq_response[16], dev->model, 16);
-	strncpy(&inq_response[32], dev->firmware_rev, 4);
+
+	while (dev->firmware_rev[fw_offset - 1] == ' ' && fw_offset > 4)
+		fw_offset--;
+	fw_offset -= 4;
+	strncpy(&inq_response[32], dev->firmware_rev + fw_offset, 4);
 
 	xfer_len = min(alloc_len, STANDARD_INQUIRY_LENGTH);
 	res = nvme_trans_copy_to_user(hdr, inq_response, xfer_len);
@@ -1018,8 +1017,8 @@ static int nvme_trans_log_info_exceptions(struct nvme_ns *ns,
 	c.common.opcode = nvme_admin_get_log_page;
 	c.common.nsid = cpu_to_le32(0xFFFFFFFF);
 	c.common.prp1 = cpu_to_le64(dma_addr);
-	c.common.cdw10[0] = cpu_to_le32(((sizeof(struct nvme_smart_log) /
-			BYTES_TO_DWORDS) << 16) | NVME_GET_SMART_LOG_PAGE);
+	c.common.cdw10[0] = cpu_to_le32((((sizeof(struct nvme_smart_log) /
+			BYTES_TO_DWORDS) - 1) << 16) | NVME_LOG_SMART);
 	res = nvme_submit_admin_cmd(dev, &c, NULL);
 	if (res != NVME_SC_SUCCESS) {
 		temp_c = LOG_TEMP_UNKNOWN;
@@ -1086,8 +1085,8 @@ static int nvme_trans_log_temperature(struct nvme_ns *ns, struct sg_io_hdr *hdr,
 	c.common.opcode = nvme_admin_get_log_page;
 	c.common.nsid = cpu_to_le32(0xFFFFFFFF);
 	c.common.prp1 = cpu_to_le64(dma_addr);
-	c.common.cdw10[0] = cpu_to_le32(((sizeof(struct nvme_smart_log) /
-			BYTES_TO_DWORDS) << 16) | NVME_GET_SMART_LOG_PAGE);
+	c.common.cdw10[0] = cpu_to_le32((((sizeof(struct nvme_smart_log) /
+			BYTES_TO_DWORDS) - 1) << 16) | NVME_LOG_SMART);
 	res = nvme_submit_admin_cmd(dev, &c, NULL);
 	if (res != NVME_SC_SUCCESS) {
 		temp_c_cur = LOG_TEMP_UNKNOWN;
@@ -1477,7 +1476,7 @@ static int nvme_trans_power_state(struct nvme_ns *ns, struct sg_io_hdr *hdr,
 		goto out_dma;
 	}
 	id_ctrl = mem;
-	lowest_pow_st = id_ctrl->npss - 1;
+	lowest_pow_st = max(POWER_STATE_0, (int)(id_ctrl->npss - 1));
 
 	switch (pc) {
 	case NVME_POWER_STATE_START_VALID:
@@ -1494,20 +1493,19 @@ static int nvme_trans_power_state(struct nvme_ns *ns, struct sg_io_hdr *hdr,
 		break;
 	case NVME_POWER_STATE_IDLE:
 		/* Action unspecified if POWER CONDITION MODIFIER != [0,1,2] */
-		/* min of desired state and (lps-1) because lps is STOP */
 		if (pcmod == 0x0)
-			ps_desired = min(POWER_STATE_1, (lowest_pow_st - 1));
+			ps_desired = POWER_STATE_1;
 		else if (pcmod == 0x1)
-			ps_desired = min(POWER_STATE_2, (lowest_pow_st - 1));
+			ps_desired = POWER_STATE_2;
 		else if (pcmod == 0x2)
-			ps_desired = min(POWER_STATE_3, (lowest_pow_st - 1));
+			ps_desired = POWER_STATE_3;
 		break;
 	case NVME_POWER_STATE_STANDBY:
 		/* Action unspecified if POWER CONDITION MODIFIER != [0,1] */
 		if (pcmod == 0x0)
-			ps_desired = max(0, (lowest_pow_st - 2));
+			ps_desired = max(POWER_STATE_0, (lowest_pow_st - 2));
 		else if (pcmod == 0x1)
-			ps_desired = max(0, (lowest_pow_st - 1));
+			ps_desired = max(POWER_STATE_0, (lowest_pow_st - 1));
 		break;
 	case NVME_POWER_STATE_LU_CONTROL:
 	default:
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index ef166ad2dbad..758ac442c5b5 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -704,6 +704,7 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 
 	rq = blk_get_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
 			     WRITE : READ, __GFP_WAIT);
+	blk_rq_set_block_pc(rq);
 
 	if (cgc->buflen) {
 		ret = blk_rq_map_kern(q, rq, cgc->buffer, cgc->buflen,
@@ -716,7 +717,6 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	memcpy(rq->cmd, cgc->cmd, CDROM_PACKET_SIZE);
 
 	rq->timeout = 60*HZ;
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	if (cgc->quiet)
 		rq->cmd_flags |= REQ_QUIET;
 
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 4c95b503b09e..b2c98c1bc037 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -541,7 +541,6 @@ static int rbd_open(struct block_device *bdev, fmode_t mode)
 		return -ENOENT;
 
 	(void) get_device(&rbd_dev->dev);
-	set_device_ro(bdev, rbd_dev->mapping.read_only);
 
 	return 0;
 }
@@ -559,10 +558,76 @@ static void rbd_release(struct gendisk *disk, fmode_t mode)
 	put_device(&rbd_dev->dev);
 }
 
+static int rbd_ioctl_set_ro(struct rbd_device *rbd_dev, unsigned long arg)
+{
+	int ret = 0;
+	int val;
+	bool ro;
+	bool ro_changed = false;
+
+	/* get_user() may sleep, so call it before taking rbd_dev->lock */
+	if (get_user(val, (int __user *)(arg)))
+		return -EFAULT;
+
+	ro = val ? true : false;
+	/* Snapshot doesn't allow to write*/
+	if (rbd_dev->spec->snap_id != CEPH_NOSNAP && !ro)
+		return -EROFS;
+
+	spin_lock_irq(&rbd_dev->lock);
+	/* prevent others open this device */
+	if (rbd_dev->open_count > 1) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	if (rbd_dev->mapping.read_only != ro) {
+		rbd_dev->mapping.read_only = ro;
+		ro_changed = true;
+	}
+
+out:
+	spin_unlock_irq(&rbd_dev->lock);
+	/* set_disk_ro() may sleep, so call it after releasing rbd_dev->lock */
+	if (ret == 0 && ro_changed)
+		set_disk_ro(rbd_dev->disk, ro ? 1 : 0);
+
+	return ret;
+}
+
+static int rbd_ioctl(struct block_device *bdev, fmode_t mode,
+			unsigned int cmd, unsigned long arg)
+{
+	struct rbd_device *rbd_dev = bdev->bd_disk->private_data;
+	int ret = 0;
+
+	switch (cmd) {
+	case BLKROSET:
+		ret = rbd_ioctl_set_ro(rbd_dev, arg);
+		break;
+	default:
+		ret = -ENOTTY;
+	}
+
+	return ret;
+}
+
+#ifdef CONFIG_COMPAT
+static int rbd_compat_ioctl(struct block_device *bdev, fmode_t mode,
+				unsigned int cmd, unsigned long arg)
+{
+	return rbd_ioctl(bdev, mode, cmd, arg);
+}
+#endif /* CONFIG_COMPAT */
+
 static const struct block_device_operations rbd_bd_ops = {
 	.owner			= THIS_MODULE,
 	.open			= rbd_open,
 	.release		= rbd_release,
+	.ioctl			= rbd_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl		= rbd_compat_ioctl,
+#endif
 };
 
 /*
@@ -1366,6 +1431,14 @@ static bool obj_request_exists_test(struct rbd_obj_request *obj_request)
 	return test_bit(OBJ_REQ_EXISTS, &obj_request->flags) != 0;
 }
 
+static bool obj_request_overlaps_parent(struct rbd_obj_request *obj_request)
+{
+	struct rbd_device *rbd_dev = obj_request->img_request->rbd_dev;
+
+	return obj_request->img_offset <
+	    round_up(rbd_dev->parent_overlap, rbd_obj_bytes(&rbd_dev->header));
+}
+
 static void rbd_obj_request_get(struct rbd_obj_request *obj_request)
 {
 	dout("%s: obj %p (was %d)\n", __func__, obj_request,
@@ -1382,6 +1455,13 @@ static void rbd_obj_request_put(struct rbd_obj_request *obj_request)
 	kref_put(&obj_request->kref, rbd_obj_request_destroy);
 }
 
+static void rbd_img_request_get(struct rbd_img_request *img_request)
+{
+	dout("%s: img %p (was %d)\n", __func__, img_request,
+	     atomic_read(&img_request->kref.refcount));
+	kref_get(&img_request->kref);
+}
+
 static bool img_request_child_test(struct rbd_img_request *img_request);
 static void rbd_parent_request_destroy(struct kref *kref);
 static void rbd_img_request_destroy(struct kref *kref);
@@ -2142,6 +2222,7 @@ static void rbd_img_obj_callback(struct rbd_obj_request *obj_request)
 	img_request->next_completion = which;
 out:
 	spin_unlock_irq(&img_request->completion_lock);
+	rbd_img_request_put(img_request);
 
 	if (!more)
 		rbd_img_request_complete(img_request);
@@ -2242,6 +2323,7 @@ static int rbd_img_request_fill(struct rbd_img_request *img_request,
 			goto out_unwind;
 		obj_request->osd_req = osd_req;
 		obj_request->callback = rbd_img_obj_callback;
+		rbd_img_request_get(img_request);
 
 		if (write_request) {
 			osd_req_op_alloc_hint_init(osd_req, which,
@@ -2674,7 +2756,7 @@ static int rbd_img_obj_request_submit(struct rbd_obj_request *obj_request)
 	 */
 	if (!img_request_write_test(img_request) ||
 		!img_request_layered_test(img_request) ||
-		rbd_dev->parent_overlap <= obj_request->img_offset ||
+		!obj_request_overlaps_parent(obj_request) ||
 		((known = obj_request_known_test(obj_request)) &&
 			obj_request_exists_test(obj_request))) {
 
@@ -2872,56 +2954,55 @@ static void rbd_watch_cb(u64 ver, u64 notify_id, u8 opcode, void *data)
 }
 
 /*
- * Request sync osd watch/unwatch.  The value of "start" determines
- * whether a watch request is being initiated or torn down.
+ * Initiate a watch request, synchronously.
  */
-static int __rbd_dev_header_watch_sync(struct rbd_device *rbd_dev, bool start)
+static int rbd_dev_header_watch_sync(struct rbd_device *rbd_dev)
 {
 	struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
 	struct rbd_obj_request *obj_request;
 	int ret;
 
-	rbd_assert(start ^ !!rbd_dev->watch_event);
-	rbd_assert(start ^ !!rbd_dev->watch_request);
+	rbd_assert(!rbd_dev->watch_event);
+	rbd_assert(!rbd_dev->watch_request);
 
-	if (start) {
-		ret = ceph_osdc_create_event(osdc, rbd_watch_cb, rbd_dev,
-						&rbd_dev->watch_event);
-		if (ret < 0)
-			return ret;
-		rbd_assert(rbd_dev->watch_event != NULL);
-	}
+	ret = ceph_osdc_create_event(osdc, rbd_watch_cb, rbd_dev,
+				     &rbd_dev->watch_event);
+	if (ret < 0)
+		return ret;
+
+	rbd_assert(rbd_dev->watch_event);
 
-	ret = -ENOMEM;
 	obj_request = rbd_obj_request_create(rbd_dev->header_name, 0, 0,
-							OBJ_REQUEST_NODATA);
-	if (!obj_request)
+					     OBJ_REQUEST_NODATA);
+	if (!obj_request) {
+		ret = -ENOMEM;
 		goto out_cancel;
+	}
 
 	obj_request->osd_req = rbd_osd_req_create(rbd_dev, true, 1,
 						  obj_request);
-	if (!obj_request->osd_req)
-		goto out_cancel;
+	if (!obj_request->osd_req) {
+		ret = -ENOMEM;
+		goto out_put;
+	}
 
-	if (start)
-		ceph_osdc_set_request_linger(osdc, obj_request->osd_req);
-	else
-		ceph_osdc_unregister_linger_request(osdc,
-					rbd_dev->watch_request->osd_req);
+	ceph_osdc_set_request_linger(osdc, obj_request->osd_req);
 
 	osd_req_op_watch_init(obj_request->osd_req, 0, CEPH_OSD_OP_WATCH,
-				rbd_dev->watch_event->cookie, 0, start ? 1 : 0);
+			      rbd_dev->watch_event->cookie, 0, 1);
 	rbd_osd_req_format_write(obj_request);
 
 	ret = rbd_obj_request_submit(osdc, obj_request);
 	if (ret)
-		goto out_cancel;
+		goto out_linger;
+
 	ret = rbd_obj_request_wait(obj_request);
 	if (ret)
-		goto out_cancel;
+		goto out_linger;
+
 	ret = obj_request->result;
 	if (ret)
-		goto out_cancel;
+		goto out_linger;
 
 	/*
 	 * A watch request is set to linger, so the underlying osd
@@ -2931,36 +3012,84 @@ static int __rbd_dev_header_watch_sync(struct rbd_device *rbd_dev, bool start)
 	 * it.  We'll drop that reference (below) after we've
 	 * unregistered it.
 	 */
-	if (start) {
-		rbd_dev->watch_request = obj_request;
+	rbd_dev->watch_request = obj_request;
 
-		return 0;
+	return 0;
+
+out_linger:
+	ceph_osdc_unregister_linger_request(osdc, obj_request->osd_req);
+out_put:
+	rbd_obj_request_put(obj_request);
+out_cancel:
+	ceph_osdc_cancel_event(rbd_dev->watch_event);
+	rbd_dev->watch_event = NULL;
+
+	return ret;
+}
+
+/*
+ * Tear down a watch request, synchronously.
+ */
+static int __rbd_dev_header_unwatch_sync(struct rbd_device *rbd_dev)
+{
+	struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
+	struct rbd_obj_request *obj_request;
+	int ret;
+
+	rbd_assert(rbd_dev->watch_event);
+	rbd_assert(rbd_dev->watch_request);
+
+	obj_request = rbd_obj_request_create(rbd_dev->header_name, 0, 0,
+					     OBJ_REQUEST_NODATA);
+	if (!obj_request) {
+		ret = -ENOMEM;
+		goto out_cancel;
+	}
+
+	obj_request->osd_req = rbd_osd_req_create(rbd_dev, true, 1,
+						  obj_request);
+	if (!obj_request->osd_req) {
+		ret = -ENOMEM;
+		goto out_put;
 	}
 
+	osd_req_op_watch_init(obj_request->osd_req, 0, CEPH_OSD_OP_WATCH,
+			      rbd_dev->watch_event->cookie, 0, 0);
+	rbd_osd_req_format_write(obj_request);
+
+	ret = rbd_obj_request_submit(osdc, obj_request);
+	if (ret)
+		goto out_put;
+
+	ret = rbd_obj_request_wait(obj_request);
+	if (ret)
+		goto out_put;
+
+	ret = obj_request->result;
+	if (ret)
+		goto out_put;
+
 	/* We have successfully torn down the watch request */
 
+	ceph_osdc_unregister_linger_request(osdc,
+					    rbd_dev->watch_request->osd_req);
 	rbd_obj_request_put(rbd_dev->watch_request);
 	rbd_dev->watch_request = NULL;
+
+out_put:
+	rbd_obj_request_put(obj_request);
 out_cancel:
-	/* Cancel the event if we're tearing down, or on error */
 	ceph_osdc_cancel_event(rbd_dev->watch_event);
 	rbd_dev->watch_event = NULL;
-	if (obj_request)
-		rbd_obj_request_put(obj_request);
 
 	return ret;
 }
 
-static int rbd_dev_header_watch_sync(struct rbd_device *rbd_dev)
-{
-	return __rbd_dev_header_watch_sync(rbd_dev, true);
-}
-
 static void rbd_dev_header_unwatch_sync(struct rbd_device *rbd_dev)
 {
 	int ret;
 
-	ret = __rbd_dev_header_watch_sync(rbd_dev, false);
+	ret = __rbd_dev_header_unwatch_sync(rbd_dev);
 	if (ret) {
 		rbd_warn(rbd_dev, "unable to tear down watch request: %d\n",
 			 ret);
@@ -3058,7 +3187,6 @@ static void rbd_request_fn(struct request_queue *q)
 		__releases(q->queue_lock) __acquires(q->queue_lock)
 {
 	struct rbd_device *rbd_dev = q->queuedata;
-	bool read_only = rbd_dev->mapping.read_only;
 	struct request *rq;
 	int result;
 
@@ -3094,7 +3222,7 @@ static void rbd_request_fn(struct request_queue *q)
 
 		if (write_request) {
 			result = -EROFS;
-			if (read_only)
+			if (rbd_dev->mapping.read_only)
 				goto end_request;
 			rbd_assert(rbd_dev->spec->snap_id == CEPH_NOSNAP);
 		}
@@ -4683,6 +4811,38 @@ out_err:
 }
 
 /*
+ * Return pool id (>= 0) or a negative error code.
+ */
+static int rbd_add_get_pool_id(struct rbd_client *rbdc, const char *pool_name)
+{
+	u64 newest_epoch;
+	unsigned long timeout = rbdc->client->options->mount_timeout * HZ;
+	int tries = 0;
+	int ret;
+
+again:
+	ret = ceph_pg_poolid_by_name(rbdc->client->osdc.osdmap, pool_name);
+	if (ret == -ENOENT && tries++ < 1) {
+		ret = ceph_monc_do_get_version(&rbdc->client->monc, "osdmap",
+					       &newest_epoch);
+		if (ret < 0)
+			return ret;
+
+		if (rbdc->client->osdc.osdmap->epoch < newest_epoch) {
+			ceph_monc_request_next_osdmap(&rbdc->client->monc);
+			(void) ceph_monc_wait_osdmap(&rbdc->client->monc,
+						     newest_epoch, timeout);
+			goto again;
+		} else {
+			/* the osdmap we have is new enough */
+			return -ENOENT;
+		}
+	}
+
+	return ret;
+}
+
+/*
  * An rbd format 2 image has a unique identifier, distinct from the
  * name given to it by the user.  Internally, that identifier is
  * what's used to specify the names of objects related to the image.
@@ -4752,7 +4912,7 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
 
 		image_id = ceph_extract_encoded_string(&p, p + ret,
 						NULL, GFP_NOIO);
-		ret = IS_ERR(image_id) ? PTR_ERR(image_id) : 0;
+		ret = PTR_ERR_OR_ZERO(image_id);
 		if (!ret)
 			rbd_dev->image_format = 2;
 	} else {
@@ -4907,6 +5067,7 @@ static int rbd_dev_device_setup(struct rbd_device *rbd_dev)
 	if (ret)
 		goto err_out_disk;
 	set_capacity(rbd_dev->disk, rbd_dev->mapping.size / SECTOR_SIZE);
+	set_disk_ro(rbd_dev->disk, rbd_dev->mapping.read_only);
 
 	ret = rbd_bus_add_dev(rbd_dev);
 	if (ret)
@@ -5053,7 +5214,6 @@ static ssize_t do_rbd_add(struct bus_type *bus,
 	struct rbd_options *rbd_opts = NULL;
 	struct rbd_spec *spec = NULL;
 	struct rbd_client *rbdc;
-	struct ceph_osd_client *osdc;
 	bool read_only;
 	int rc = -ENOMEM;
 
@@ -5075,8 +5235,7 @@ static ssize_t do_rbd_add(struct bus_type *bus,
 	}
 
 	/* pick the pool */
-	osdc = &rbdc->client->osdc;
-	rc = ceph_pg_poolid_by_name(osdc->osdmap, spec->pool_name);
+	rc = rbd_add_get_pool_id(rbdc, spec->pool_name);
 	if (rc < 0)
 		goto err_out_client;
 	spec->pool_id = (u64)rc;
@@ -5387,6 +5546,7 @@ err_out_slab:
 
 static void __exit rbd_exit(void)
 {
+	ida_destroy(&rbd_dev_id_ida);
 	rbd_sysfs_cleanup();
 	if (single_major)
 		unregister_blkdev(rbd_major, RBD_DRV_NAME);
diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
index 48eccb350180..089e72cd37be 100644
--- a/drivers/block/zram/zram_drv.c
+++ b/drivers/block/zram/zram_drv.c
@@ -622,8 +622,10 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity)
 	memset(&zram->stats, 0, sizeof(zram->stats));
 
 	zram->disksize = 0;
-	if (reset_capacity)
+	if (reset_capacity) {
 		set_capacity(zram->disk, 0);
+		revalidate_disk(zram->disk);
+	}
 	up_write(&zram->init_lock);
 }
 
@@ -664,6 +666,7 @@ static ssize_t disksize_store(struct device *dev,
 	zram->comp = comp;
 	zram->disksize = disksize;
 	set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);
+	revalidate_disk(zram->disk);
 	up_write(&zram->init_lock);
 	return len;
 
diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c
index a83b57e57b63..f98380648cb3 100644
--- a/drivers/bluetooth/ath3k.c
+++ b/drivers/bluetooth/ath3k.c
@@ -193,9 +193,10 @@ static int ath3k_load_firmware(struct usb_device *udev,
 	sent += 20;
 	count -= 20;
 
+	pipe = usb_sndbulkpipe(udev, 0x02);
+
 	while (count) {
 		size = min_t(uint, count, BULK_SIZE);
-		pipe = usb_sndbulkpipe(udev, 0x02);
 		memcpy(send_buf, firmware->data + sent, size);
 
 		err = usb_bulk_msg(udev, pipe, send_buf, size,
diff --git a/drivers/bluetooth/btmrvl_drv.h b/drivers/bluetooth/btmrvl_drv.h
index 7399303d7d99..dc79f88f8717 100644
--- a/drivers/bluetooth/btmrvl_drv.h
+++ b/drivers/bluetooth/btmrvl_drv.h
@@ -59,6 +59,8 @@ struct btmrvl_device {
 };
 
 struct btmrvl_adapter {
+	void *hw_regs_buf;
+	u8 *hw_regs;
 	u32 int_count;
 	struct sk_buff_head tx_queue;
 	u8 psmode;
@@ -140,7 +142,7 @@ void btmrvl_interrupt(struct btmrvl_private *priv);
 bool btmrvl_check_evtpkt(struct btmrvl_private *priv, struct sk_buff *skb);
 int btmrvl_process_event(struct btmrvl_private *priv, struct sk_buff *skb);
 
-int btmrvl_send_module_cfg_cmd(struct btmrvl_private *priv, int subcmd);
+int btmrvl_send_module_cfg_cmd(struct btmrvl_private *priv, u8 subcmd);
 int btmrvl_send_hscfg_cmd(struct btmrvl_private *priv);
 int btmrvl_enable_ps(struct btmrvl_private *priv);
 int btmrvl_prepare_command(struct btmrvl_private *priv);
diff --git a/drivers/bluetooth/btmrvl_main.c b/drivers/bluetooth/btmrvl_main.c
index 2c4997ce2484..e9dbddb0b8f1 100644
--- a/drivers/bluetooth/btmrvl_main.c
+++ b/drivers/bluetooth/btmrvl_main.c
@@ -24,6 +24,7 @@
 #include <net/bluetooth/hci_core.h>
 
 #include "btmrvl_drv.h"
+#include "btmrvl_sdio.h"
 
 #define VERSION "1.0"
 
@@ -201,7 +202,7 @@ static int btmrvl_send_sync_cmd(struct btmrvl_private *priv, u16 opcode,
 	return 0;
 }
 
-int btmrvl_send_module_cfg_cmd(struct btmrvl_private *priv, int subcmd)
+int btmrvl_send_module_cfg_cmd(struct btmrvl_private *priv, u8 subcmd)
 {
 	int ret;
 
@@ -337,10 +338,25 @@ static int btmrvl_tx_pkt(struct btmrvl_private *priv, struct sk_buff *skb)
 
 static void btmrvl_init_adapter(struct btmrvl_private *priv)
 {
+	int buf_size;
+
 	skb_queue_head_init(&priv->adapter->tx_queue);
 
 	priv->adapter->ps_state = PS_AWAKE;
 
+	buf_size = ALIGN_SZ(SDIO_BLOCK_SIZE, BTSDIO_DMA_ALIGN);
+	priv->adapter->hw_regs_buf = kzalloc(buf_size, GFP_KERNEL);
+	if (!priv->adapter->hw_regs_buf) {
+		priv->adapter->hw_regs = NULL;
+		BT_ERR("Unable to allocate buffer for hw_regs.");
+	} else {
+		priv->adapter->hw_regs =
+			(u8 *)ALIGN_ADDR(priv->adapter->hw_regs_buf,
+					 BTSDIO_DMA_ALIGN);
+		BT_DBG("hw_regs_buf=%p hw_regs=%p",
+		       priv->adapter->hw_regs_buf, priv->adapter->hw_regs);
+	}
+
 	init_waitqueue_head(&priv->adapter->cmd_wait_q);
 }
 
@@ -348,6 +364,7 @@ static void btmrvl_free_adapter(struct btmrvl_private *priv)
 {
 	skb_queue_purge(&priv->adapter->tx_queue);
 
+	kfree(priv->adapter->hw_regs_buf);
 	kfree(priv->adapter);
 
 	priv->adapter = NULL;
diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c
index 1b52c9f5230d..9dedca516ff5 100644
--- a/drivers/bluetooth/btmrvl_sdio.c
+++ b/drivers/bluetooth/btmrvl_sdio.c
@@ -64,6 +64,7 @@ static const struct btmrvl_sdio_card_reg btmrvl_reg_8688 = {
 	.io_port_0 = 0x00,
 	.io_port_1 = 0x01,
 	.io_port_2 = 0x02,
+	.int_read_to_clear = false,
 };
 static const struct btmrvl_sdio_card_reg btmrvl_reg_87xx = {
 	.cfg = 0x00,
@@ -80,6 +81,7 @@ static const struct btmrvl_sdio_card_reg btmrvl_reg_87xx = {
 	.io_port_0 = 0x78,
 	.io_port_1 = 0x79,
 	.io_port_2 = 0x7a,
+	.int_read_to_clear = false,
 };
 
 static const struct btmrvl_sdio_card_reg btmrvl_reg_88xx = {
@@ -97,6 +99,9 @@ static const struct btmrvl_sdio_card_reg btmrvl_reg_88xx = {
 	.io_port_0 = 0xd8,
 	.io_port_1 = 0xd9,
 	.io_port_2 = 0xda,
+	.int_read_to_clear = true,
+	.host_int_rsr = 0x01,
+	.card_misc_cfg = 0xcc,
 };
 
 static const struct btmrvl_sdio_device btmrvl_sdio_sd8688 = {
@@ -667,46 +672,78 @@ static int btmrvl_sdio_process_int_status(struct btmrvl_private *priv)
 	return 0;
 }
 
-static void btmrvl_sdio_interrupt(struct sdio_func *func)
+static int btmrvl_sdio_read_to_clear(struct btmrvl_sdio_card *card, u8 *ireg)
 {
-	struct btmrvl_private *priv;
-	struct btmrvl_sdio_card *card;
-	ulong flags;
-	u8 ireg = 0;
+	struct btmrvl_adapter *adapter = card->priv->adapter;
 	int ret;
 
-	card = sdio_get_drvdata(func);
-	if (!card || !card->priv) {
-		BT_ERR("sbi_interrupt(%p) card or priv is "
-				"NULL, card=%p\n", func, card);
-		return;
+	ret = sdio_readsb(card->func, adapter->hw_regs, 0, SDIO_BLOCK_SIZE);
+	if (ret) {
+		BT_ERR("sdio_readsb: read int hw_regs failed: %d", ret);
+		return ret;
 	}
 
-	priv = card->priv;
+	*ireg = adapter->hw_regs[card->reg->host_intstatus];
+	BT_DBG("hw_regs[%#x]=%#x", card->reg->host_intstatus, *ireg);
+
+	return 0;
+}
 
-	ireg = sdio_readb(card->func, card->reg->host_intstatus, &ret);
+static int btmrvl_sdio_write_to_clear(struct btmrvl_sdio_card *card, u8 *ireg)
+{
+	int ret;
+
+	*ireg = sdio_readb(card->func, card->reg->host_intstatus, &ret);
 	if (ret) {
-		BT_ERR("sdio_readb: read int status register failed");
-		return;
+		BT_ERR("sdio_readb: read int status failed: %d", ret);
+		return ret;
 	}
 
-	if (ireg != 0) {
+	if (*ireg) {
 		/*
 		 * DN_LD_HOST_INT_STATUS and/or UP_LD_HOST_INT_STATUS
 		 * Clear the interrupt status register and re-enable the
 		 * interrupt.
 		 */
-		BT_DBG("ireg = 0x%x", ireg);
+		BT_DBG("int_status = 0x%x", *ireg);
 
-		sdio_writeb(card->func, ~(ireg) & (DN_LD_HOST_INT_STATUS |
-					UP_LD_HOST_INT_STATUS),
-				card->reg->host_intstatus, &ret);
+		sdio_writeb(card->func, ~(*ireg) & (DN_LD_HOST_INT_STATUS |
+						    UP_LD_HOST_INT_STATUS),
+			    card->reg->host_intstatus, &ret);
 		if (ret) {
-			BT_ERR("sdio_writeb: clear int status register failed");
-			return;
+			BT_ERR("sdio_writeb: clear int status failed: %d", ret);
+			return ret;
 		}
 	}
 
+	return 0;
+}
+
+static void btmrvl_sdio_interrupt(struct sdio_func *func)
+{
+	struct btmrvl_private *priv;
+	struct btmrvl_sdio_card *card;
+	ulong flags;
+	u8 ireg = 0;
+	int ret;
+
+	card = sdio_get_drvdata(func);
+	if (!card || !card->priv) {
+		BT_ERR("sbi_interrupt(%p) card or priv is "
+				"NULL, card=%p\n", func, card);
+		return;
+	}
+
+	priv = card->priv;
+
+	if (card->reg->int_read_to_clear)
+		ret = btmrvl_sdio_read_to_clear(card, &ireg);
+	else
+		ret = btmrvl_sdio_write_to_clear(card, &ireg);
+
+	if (ret)
+		return;
+
 	spin_lock_irqsave(&priv->driver_lock, flags);
 	sdio_ireg |= ireg;
 	spin_unlock_irqrestore(&priv->driver_lock, flags);
@@ -777,6 +814,30 @@ static int btmrvl_sdio_register_dev(struct btmrvl_sdio_card *card)
 
 	BT_DBG("SDIO FUNC%d IO port: 0x%x", func->num, card->ioport);
 
+	if (card->reg->int_read_to_clear) {
+		reg = sdio_readb(func, card->reg->host_int_rsr, &ret);
+		if (ret < 0) {
+			ret = -EIO;
+			goto release_irq;
+		}
+		sdio_writeb(func, reg | 0x3f, card->reg->host_int_rsr, &ret);
+		if (ret < 0) {
+			ret = -EIO;
+			goto release_irq;
+		}
+
+		reg = sdio_readb(func, card->reg->card_misc_cfg, &ret);
+		if (ret < 0) {
+			ret = -EIO;
+			goto release_irq;
+		}
+		sdio_writeb(func, reg | 0x10, card->reg->card_misc_cfg, &ret);
+		if (ret < 0) {
+			ret = -EIO;
+			goto release_irq;
+		}
+	}
+
 	sdio_set_drvdata(func, card);
 
 	sdio_release_host(func);
diff --git a/drivers/bluetooth/btmrvl_sdio.h b/drivers/bluetooth/btmrvl_sdio.h
index 43d35a609ca9..d4dd3b0fa53d 100644
--- a/drivers/bluetooth/btmrvl_sdio.h
+++ b/drivers/bluetooth/btmrvl_sdio.h
@@ -78,6 +78,9 @@ struct btmrvl_sdio_card_reg {
 	u8 io_port_0;
 	u8 io_port_1;
 	u8 io_port_2;
+	bool int_read_to_clear;
+	u8 host_int_rsr;
+	u8 card_misc_cfg;
 };
 
 struct btmrvl_sdio_card {
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index a7dfbf9a3afb..a1c80b0c7663 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -49,6 +49,7 @@ static struct usb_driver btusb_driver;
 #define BTUSB_WRONG_SCO_MTU	0x40
 #define BTUSB_ATH3012		0x80
 #define BTUSB_INTEL		0x100
+#define BTUSB_BCM_PATCHRAM	0x200
 
 static const struct usb_device_id btusb_table[] = {
 	/* Generic Bluetooth USB device */
@@ -111,7 +112,8 @@ static const struct usb_device_id btusb_table[] = {
 	{ USB_VENDOR_AND_INTERFACE_INFO(0x0489, 0xff, 0x01, 0x01) },
 
 	/* Broadcom devices with vendor specific id */
-	{ USB_VENDOR_AND_INTERFACE_INFO(0x0a5c, 0xff, 0x01, 0x01) },
+	{ USB_VENDOR_AND_INTERFACE_INFO(0x0a5c, 0xff, 0x01, 0x01),
+	  .driver_info = BTUSB_BCM_PATCHRAM },
 
 	/* Belkin F8065bf - Broadcom based */
 	{ USB_VENDOR_AND_INTERFACE_INFO(0x050d, 0xff, 0x01, 0x01) },
@@ -1381,6 +1383,154 @@ exit_mfg_deactivate:
 	return 0;
 }
 
+static int btusb_setup_bcm_patchram(struct hci_dev *hdev)
+{
+	struct btusb_data *data = hci_get_drvdata(hdev);
+	struct usb_device *udev = data->udev;
+	char fw_name[64];
+	const struct firmware *fw;
+	const u8 *fw_ptr;
+	size_t fw_size;
+	const struct hci_command_hdr *cmd;
+	const u8 *cmd_param;
+	u16 opcode;
+	struct sk_buff *skb;
+	struct hci_rp_read_local_version *ver;
+	long ret;
+
+	snprintf(fw_name, sizeof(fw_name), "brcm/%s-%04x-%04x.hcd",
+		 udev->product ? udev->product : "BCM",
+		 le16_to_cpu(udev->descriptor.idVendor),
+		 le16_to_cpu(udev->descriptor.idProduct));
+
+	ret = request_firmware(&fw, fw_name, &hdev->dev);
+	if (ret < 0) {
+		BT_INFO("%s: BCM: patch %s not found", hdev->name,
+			fw_name);
+		return 0;
+	}
+
+	/* Reset */
+	skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		ret = PTR_ERR(skb);
+		BT_ERR("%s: HCI_OP_RESET failed (%ld)", hdev->name, ret);
+		goto done;
+	}
+	kfree_skb(skb);
+
+	/* Read Local Version Info */
+	skb = __hci_cmd_sync(hdev, HCI_OP_READ_LOCAL_VERSION, 0, NULL,
+			     HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		ret = PTR_ERR(skb);
+		BT_ERR("%s: HCI_OP_READ_LOCAL_VERSION failed (%ld)",
+			hdev->name, ret);
+		goto done;
+	}
+
+	if (skb->len != sizeof(*ver)) {
+		BT_ERR("%s: HCI_OP_READ_LOCAL_VERSION event length mismatch",
+			hdev->name);
+		kfree_skb(skb);
+		ret = -EIO;
+		goto done;
+	}
+
+	ver = (struct hci_rp_read_local_version *) skb->data;
+	BT_INFO("%s: BCM: patching hci_ver=%02x hci_rev=%04x lmp_ver=%02x "
+		"lmp_subver=%04x", hdev->name, ver->hci_ver, ver->hci_rev,
+		ver->lmp_ver, ver->lmp_subver);
+	kfree_skb(skb);
+
+	/* Start Download */
+	skb = __hci_cmd_sync(hdev, 0xfc2e, 0, NULL, HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		ret = PTR_ERR(skb);
+		BT_ERR("%s: BCM: Download Minidrv command failed (%ld)",
+			hdev->name, ret);
+		goto reset_fw;
+	}
+	kfree_skb(skb);
+
+	/* 50 msec delay after Download Minidrv completes */
+	msleep(50);
+
+	fw_ptr = fw->data;
+	fw_size = fw->size;
+
+	while (fw_size >= sizeof(*cmd)) {
+		cmd = (struct hci_command_hdr *) fw_ptr;
+		fw_ptr += sizeof(*cmd);
+		fw_size -= sizeof(*cmd);
+
+		if (fw_size < cmd->plen) {
+			BT_ERR("%s: BCM: patch %s is corrupted",
+				hdev->name, fw_name);
+			ret = -EINVAL;
+			goto reset_fw;
+		}
+
+		cmd_param = fw_ptr;
+		fw_ptr += cmd->plen;
+		fw_size -= cmd->plen;
+
+		opcode = le16_to_cpu(cmd->opcode);
+
+		skb = __hci_cmd_sync(hdev, opcode, cmd->plen, cmd_param,
+				     HCI_INIT_TIMEOUT);
+		if (IS_ERR(skb)) {
+			ret = PTR_ERR(skb);
+			BT_ERR("%s: BCM: patch command %04x failed (%ld)",
+				hdev->name, opcode, ret);
+			goto reset_fw;
+		}
+		kfree_skb(skb);
+	}
+
+	/* 250 msec delay after Launch Ram completes */
+	msleep(250);
+
+reset_fw:
+	/* Reset */
+	skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		ret = PTR_ERR(skb);
+		BT_ERR("%s: HCI_OP_RESET failed (%ld)", hdev->name, ret);
+		goto done;
+	}
+	kfree_skb(skb);
+
+	/* Read Local Version Info */
+	skb = __hci_cmd_sync(hdev, HCI_OP_READ_LOCAL_VERSION, 0, NULL,
+			     HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		ret = PTR_ERR(skb);
+		BT_ERR("%s: HCI_OP_READ_LOCAL_VERSION failed (%ld)",
+			hdev->name, ret);
+		goto done;
+	}
+
+	if (skb->len != sizeof(*ver)) {
+		BT_ERR("%s: HCI_OP_READ_LOCAL_VERSION event length mismatch",
+			hdev->name);
+		kfree_skb(skb);
+		ret = -EIO;
+		goto done;
+	}
+
+	ver = (struct hci_rp_read_local_version *) skb->data;
+	BT_INFO("%s: BCM: firmware hci_ver=%02x hci_rev=%04x lmp_ver=%02x "
+		"lmp_subver=%04x", hdev->name, ver->hci_ver, ver->hci_rev,
+		ver->lmp_ver, ver->lmp_subver);
+	kfree_skb(skb);
+
+done:
+	release_firmware(fw);
+
+	return ret;
+}
+
 static int btusb_probe(struct usb_interface *intf,
 				const struct usb_device_id *id)
 {
@@ -1486,6 +1636,9 @@ static int btusb_probe(struct usb_interface *intf,
 	if (id->driver_info & BTUSB_BCM92035)
 		hdev->setup = btusb_setup_bcm92035;
 
+	if (id->driver_info & BTUSB_BCM_PATCHRAM)
+		hdev->setup = btusb_setup_bcm_patchram;
+
 	if (id->driver_info & BTUSB_INTEL)
 		hdev->setup = btusb_setup_intel;
 
diff --git a/drivers/bluetooth/hci_h4.c b/drivers/bluetooth/hci_h4.c
index 7048a583fe51..66db9a803373 100644
--- a/drivers/bluetooth/hci_h4.c
+++ b/drivers/bluetooth/hci_h4.c
@@ -55,13 +55,6 @@ struct h4_struct {
 	struct sk_buff_head txq;
 };
 
-/* H4 receiver States */
-#define H4_W4_PACKET_TYPE	0
-#define H4_W4_EVENT_HDR		1
-#define H4_W4_ACL_HDR		2
-#define H4_W4_SCO_HDR		3
-#define H4_W4_DATA		4
-
 /* Initialize protocol */
 static int h4_open(struct hci_uart *hu)
 {
diff --git a/drivers/bus/Kconfig b/drivers/bus/Kconfig
index a118ec1650fa..1f37d9870e7a 100644
--- a/drivers/bus/Kconfig
+++ b/drivers/bus/Kconfig
@@ -45,7 +45,7 @@ config OMAP_INTERCONNECT
 
 config ARM_CCI
 	bool "ARM CCI driver support"
-	depends on ARM
+	depends on ARM && OF && CPU_V7
 	help
 	  Driver supporting the CCI cache coherent interconnect for ARM
 	  platforms.
diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
index 2a44767891f5..898b84bba28a 100644
--- a/drivers/cdrom/cdrom.c
+++ b/drivers/cdrom/cdrom.c
@@ -2184,6 +2184,7 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 			ret = -ENOMEM;
 			break;
 		}
+		blk_rq_set_block_pc(rq);
 
 		ret = blk_rq_map_user(q, rq, NULL, ubuf, len, GFP_KERNEL);
 		if (ret) {
@@ -2203,7 +2204,6 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 		rq->cmd[9] = 0xf8;
 
 		rq->cmd_len = 12;
-		rq->cmd_type = REQ_TYPE_BLOCK_PC;
 		rq->timeout = 60 * HZ;
 		bio = rq->bio;
 
diff --git a/drivers/char/hw_random/virtio-rng.c b/drivers/char/hw_random/virtio-rng.c
index 2ce0e225e58c..f3e71501de54 100644
--- a/drivers/char/hw_random/virtio-rng.c
+++ b/drivers/char/hw_random/virtio-rng.c
@@ -25,88 +25,115 @@
 #include <linux/virtio_rng.h>
 #include <linux/module.h>
 
-static struct virtqueue *vq;
-static unsigned int data_avail;
-static DECLARE_COMPLETION(have_data);
-static bool busy;
+static DEFINE_IDA(rng_index_ida);
+
+struct virtrng_info {
+	struct virtio_device *vdev;
+	struct hwrng hwrng;
+	struct virtqueue *vq;
+	unsigned int data_avail;
+	struct completion have_data;
+	bool busy;
+	char name[25];
+	int index;
+};
 
 static void random_recv_done(struct virtqueue *vq)
 {
+	struct virtrng_info *vi = vq->vdev->priv;
+
 	/* We can get spurious callbacks, e.g. shared IRQs + virtio_pci. */
-	if (!virtqueue_get_buf(vq, &data_avail))
+	if (!virtqueue_get_buf(vi->vq, &vi->data_avail))
 		return;
 
-	complete(&have_data);
+	complete(&vi->have_data);
 }
 
 /* The host will fill any buffer we give it with sweet, sweet randomness. */
-static void register_buffer(u8 *buf, size_t size)
+static void register_buffer(struct virtrng_info *vi, u8 *buf, size_t size)
 {
 	struct scatterlist sg;
 
 	sg_init_one(&sg, buf, size);
 
 	/* There should always be room for one buffer. */
-	virtqueue_add_inbuf(vq, &sg, 1, buf, GFP_KERNEL);
+	virtqueue_add_inbuf(vi->vq, &sg, 1, buf, GFP_KERNEL);
 
-	virtqueue_kick(vq);
+	virtqueue_kick(vi->vq);
 }
 
 static int virtio_read(struct hwrng *rng, void *buf, size_t size, bool wait)
 {
 	int ret;
+	struct virtrng_info *vi = (struct virtrng_info *)rng->priv;
 
-	if (!busy) {
-		busy = true;
-		init_completion(&have_data);
-		register_buffer(buf, size);
+	if (!vi->busy) {
+		vi->busy = true;
+		init_completion(&vi->have_data);
+		register_buffer(vi, buf, size);
 	}
 
 	if (!wait)
 		return 0;
 
-	ret = wait_for_completion_killable(&have_data);
+	ret = wait_for_completion_killable(&vi->have_data);
 	if (ret < 0)
 		return ret;
 
-	busy = false;
+	vi->busy = false;
 
-	return data_avail;
+	return vi->data_avail;
 }
 
 static void virtio_cleanup(struct hwrng *rng)
 {
-	if (busy)
-		wait_for_completion(&have_data);
-}
+	struct virtrng_info *vi = (struct virtrng_info *)rng->priv;
 
-
-static struct hwrng virtio_hwrng = {
-	.name		= "virtio",
-	.cleanup	= virtio_cleanup,
-	.read		= virtio_read,
-};
+	if (vi->busy)
+		wait_for_completion(&vi->have_data);
+}
 
 static int probe_common(struct virtio_device *vdev)
 {
-	int err;
+	int err, index;
+	struct virtrng_info *vi = NULL;
+
+	vi = kzalloc(sizeof(struct virtrng_info), GFP_KERNEL);
+	if (!vi)
+		return -ENOMEM;
 
-	if (vq) {
-		/* We only support one device for now */
-		return -EBUSY;
+	vi->index = index = ida_simple_get(&rng_index_ida, 0, 0, GFP_KERNEL);
+	if (index < 0) {
+		kfree(vi);
+		return index;
 	}
+	sprintf(vi->name, "virtio_rng.%d", index);
+	init_completion(&vi->have_data);
+
+	vi->hwrng = (struct hwrng) {
+		.read = virtio_read,
+		.cleanup = virtio_cleanup,
+		.priv = (unsigned long)vi,
+		.name = vi->name,
+	};
+	vdev->priv = vi;
+
 	/* We expect a single virtqueue. */
-	vq = virtio_find_single_vq(vdev, random_recv_done, "input");
-	if (IS_ERR(vq)) {
-		err = PTR_ERR(vq);
-		vq = NULL;
+	vi->vq = virtio_find_single_vq(vdev, random_recv_done, "input");
+	if (IS_ERR(vi->vq)) {
+		err = PTR_ERR(vi->vq);
+		vi->vq = NULL;
+		kfree(vi);
+		ida_simple_remove(&rng_index_ida, index);
 		return err;
 	}
 
-	err = hwrng_register(&virtio_hwrng);
+	err = hwrng_register(&vi->hwrng);
 	if (err) {
 		vdev->config->del_vqs(vdev);
-		vq = NULL;
+		vi->vq = NULL;
+		kfree(vi);
+		ida_simple_remove(&rng_index_ida, index);
 		return err;
 	}
 
@@ -115,11 +142,13 @@ static int probe_common(struct virtio_device *vdev)
 
 static void remove_common(struct virtio_device *vdev)
 {
+	struct virtrng_info *vi = vdev->priv;
 	vdev->config->reset(vdev);
-	busy = false;
-	hwrng_unregister(&virtio_hwrng);
+	vi->busy = false;
+	hwrng_unregister(&vi->hwrng);
 	vdev->config->del_vqs(vdev);
-	vq = NULL;
+	ida_simple_remove(&rng_index_ida, vi->index);
+	kfree(vi);
 }
 
 static int virtrng_probe(struct virtio_device *vdev)
diff --git a/drivers/char/i8k.c b/drivers/char/i8k.c
index d915707d2ba1..93dcad0c1cbe 100644
--- a/drivers/char/i8k.c
+++ b/drivers/char/i8k.c
@@ -138,7 +138,9 @@ static int i8k_smm(struct smm_regs *regs)
 	if (!alloc_cpumask_var(&old_mask, GFP_KERNEL))
 		return -ENOMEM;
 	cpumask_copy(old_mask, &current->cpus_allowed);
-	set_cpus_allowed_ptr(current, cpumask_of(0));
+	rc = set_cpus_allowed_ptr(current, cpumask_of(0));
+	if (rc)
+		goto out;
 	if (smp_processor_id() != 0) {
 		rc = -EBUSY;
 		goto out;
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 4ad71ef2cd59..0a7ac0a7b252 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -980,7 +980,6 @@ static void push_to_pool(struct work_struct *work)
 static size_t account(struct entropy_store *r, size_t nbytes, int min,
 		      int reserved)
 {
-	int have_bytes;
 	int entropy_count, orig;
 	size_t ibytes;
 
@@ -989,17 +988,19 @@ static size_t account(struct entropy_store *r, size_t nbytes, int min,
 	/* Can we pull enough? */
 retry:
 	entropy_count = orig = ACCESS_ONCE(r->entropy_count);
-	have_bytes = entropy_count >> (ENTROPY_SHIFT + 3);
 	ibytes = nbytes;
 	/* If limited, never pull more than available */
-	if (r->limit)
-		ibytes = min_t(size_t, ibytes, have_bytes - reserved);
+	if (r->limit) {
+		int have_bytes = entropy_count >> (ENTROPY_SHIFT + 3);
+
+		if ((have_bytes -= reserved) < 0)
+			have_bytes = 0;
+		ibytes = min_t(size_t, ibytes, have_bytes);
+	}
 	if (ibytes < min)
 		ibytes = 0;
-	if (have_bytes >= ibytes + reserved)
-		entropy_count -= ibytes << (ENTROPY_SHIFT + 3);
-	else
-		entropy_count = reserved << (ENTROPY_SHIFT + 3);
+	if ((entropy_count -= ibytes << (ENTROPY_SHIFT + 3)) < 0)
+		entropy_count = 0;
 
 	if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig)
 		goto retry;
diff --git a/drivers/char/raw.c b/drivers/char/raw.c
index 6e8d65e9b1d3..0102dc788608 100644
--- a/drivers/char/raw.c
+++ b/drivers/char/raw.c
@@ -284,10 +284,10 @@ static long raw_ctl_compat_ioctl(struct file *file, unsigned int cmd,
 #endif
 
 static const struct file_operations raw_fops = {
-	.read		= do_sync_read,
-	.aio_read	= generic_file_aio_read,
-	.write		= do_sync_write,
-	.aio_write	= blkdev_aio_write,
+	.read		= new_sync_read,
+	.read_iter	= generic_file_read_iter,
+	.write		= new_sync_write,
+	.write_iter	= blkdev_write_iter,
 	.fsync		= blkdev_fsync,
 	.open		= raw_open,
 	.release	= raw_release,
diff --git a/drivers/clk/clk-s2mps11.c b/drivers/clk/clk-s2mps11.c
index 9b7b5859a420..3757e9e72d37 100644
--- a/drivers/clk/clk-s2mps11.c
+++ b/drivers/clk/clk-s2mps11.c
@@ -230,16 +230,13 @@ static int s2mps11_clk_probe(struct platform_device *pdev)
 			goto err_reg;
 		}
 
-		s2mps11_clk->lookup = devm_kzalloc(&pdev->dev,
-					sizeof(struct clk_lookup), GFP_KERNEL);
+		s2mps11_clk->lookup = clkdev_alloc(s2mps11_clk->clk,
+					s2mps11_name(s2mps11_clk), NULL);
 		if (!s2mps11_clk->lookup) {
 			ret = -ENOMEM;
 			goto err_lup;
 		}
 
-		s2mps11_clk->lookup->con_id = s2mps11_name(s2mps11_clk);
-		s2mps11_clk->lookup->clk = s2mps11_clk->clk;
-
 		clkdev_add(s2mps11_clk->lookup);
 	}
 
diff --git a/drivers/clk/qcom/mmcc-msm8960.c b/drivers/clk/qcom/mmcc-msm8960.c
index 12f3c0b64fcd..4c449b3170f6 100644
--- a/drivers/clk/qcom/mmcc-msm8960.c
+++ b/drivers/clk/qcom/mmcc-msm8960.c
@@ -1209,7 +1209,7 @@ static struct clk_branch rot_clk = {
 
 static u8 mmcc_pxo_hdmi_map[] = {
 	[P_PXO]		= 0,
-	[P_HDMI_PLL]	= 2,
+	[P_HDMI_PLL]	= 3,
 };
 
 static const char *mmcc_pxo_hdmi[] = {
diff --git a/drivers/clk/samsung/clk-exynos4.c b/drivers/clk/samsung/clk-exynos4.c
index 4f150c9dd38c..7f4a473a7ad7 100644
--- a/drivers/clk/samsung/clk-exynos4.c
+++ b/drivers/clk/samsung/clk-exynos4.c
@@ -925,21 +925,13 @@ static struct samsung_gate_clock exynos4x12_gate_clks[] __initdata = {
 	GATE(CLK_RTC, "rtc", "aclk100", E4X12_GATE_IP_PERIR, 15,
 			0, 0),
 	GATE(CLK_KEYIF, "keyif", "aclk100", E4X12_GATE_IP_PERIR, 16, 0, 0),
-	GATE(CLK_SCLK_PWM_ISP, "sclk_pwm_isp", "div_pwm_isp",
-			E4X12_SRC_MASK_ISP, 0, CLK_SET_RATE_PARENT, 0),
-	GATE(CLK_SCLK_SPI0_ISP, "sclk_spi0_isp", "div_spi0_isp_pre",
-			E4X12_SRC_MASK_ISP, 4, CLK_SET_RATE_PARENT, 0),
-	GATE(CLK_SCLK_SPI1_ISP, "sclk_spi1_isp", "div_spi1_isp_pre",
-			E4X12_SRC_MASK_ISP, 8, CLK_SET_RATE_PARENT, 0),
-	GATE(CLK_SCLK_UART_ISP, "sclk_uart_isp", "div_uart_isp",
-			E4X12_SRC_MASK_ISP, 12, CLK_SET_RATE_PARENT, 0),
-	GATE(CLK_PWM_ISP_SCLK, "pwm_isp_sclk", "sclk_pwm_isp",
+	GATE(CLK_PWM_ISP_SCLK, "pwm_isp_sclk", "div_pwm_isp",
 			E4X12_GATE_IP_ISP, 0, 0, 0),
-	GATE(CLK_SPI0_ISP_SCLK, "spi0_isp_sclk", "sclk_spi0_isp",
+	GATE(CLK_SPI0_ISP_SCLK, "spi0_isp_sclk", "div_spi0_isp_pre",
 			E4X12_GATE_IP_ISP, 1, 0, 0),
-	GATE(CLK_SPI1_ISP_SCLK, "spi1_isp_sclk", "sclk_spi1_isp",
+	GATE(CLK_SPI1_ISP_SCLK, "spi1_isp_sclk", "div_spi1_isp_pre",
 			E4X12_GATE_IP_ISP, 2, 0, 0),
-	GATE(CLK_UART_ISP_SCLK, "uart_isp_sclk", "sclk_uart_isp",
+	GATE(CLK_UART_ISP_SCLK, "uart_isp_sclk", "div_uart_isp",
 			E4X12_GATE_IP_ISP, 3, 0, 0),
 	GATE(CLK_WDT, "watchdog", "aclk100", E4X12_GATE_IP_PERIR, 14, 0, 0),
 	GATE(CLK_PCM0, "pcm0", "aclk100", E4X12_GATE_IP_MAUDIO, 2,
diff --git a/drivers/clk/samsung/clk-exynos5250.c b/drivers/clk/samsung/clk-exynos5250.c
index 1fad4c5e3f5d..184f64293b26 100644
--- a/drivers/clk/samsung/clk-exynos5250.c
+++ b/drivers/clk/samsung/clk-exynos5250.c
@@ -661,7 +661,7 @@ static struct samsung_gate_clock exynos5250_gate_clks[] __initdata = {
 	GATE(CLK_RTC, "rtc", "div_aclk66", GATE_IP_PERIS, 20, 0, 0),
 	GATE(CLK_TMU, "tmu", "div_aclk66", GATE_IP_PERIS, 21, 0, 0),
 	GATE(CLK_SMMU_TV, "smmu_tv", "mout_aclk200_disp1_sub",
-			GATE_IP_DISP1, 2, 0, 0),
+			GATE_IP_DISP1, 9, 0, 0),
 	GATE(CLK_SMMU_FIMD1, "smmu_fimd1", "mout_aclk200_disp1_sub",
 			GATE_IP_DISP1, 8, 0, 0),
 	GATE(CLK_SMMU_2D, "smmu_2d", "div_aclk200", GATE_IP_ACP, 7, 0, 0),
diff --git a/drivers/clk/samsung/clk-exynos5420.c b/drivers/clk/samsung/clk-exynos5420.c
index 9d7d7eed03fd..a4e6cc782e5c 100644
--- a/drivers/clk/samsung/clk-exynos5420.c
+++ b/drivers/clk/samsung/clk-exynos5420.c
@@ -631,7 +631,8 @@ static struct samsung_mux_clock exynos5x_mux_clks[] __initdata = {
 			SRC_TOP4, 16, 1),
 	MUX(0, "mout_user_aclk266", mout_user_aclk266_p, SRC_TOP4, 20, 1),
 	MUX(0, "mout_user_aclk166", mout_user_aclk166_p, SRC_TOP4, 24, 1),
-	MUX(0, "mout_user_aclk333", mout_user_aclk333_p, SRC_TOP4, 28, 1),
+	MUX(CLK_MOUT_USER_ACLK333, "mout_user_aclk333", mout_user_aclk333_p,
+			SRC_TOP4, 28, 1),
 
 	MUX(0, "mout_user_aclk400_disp1", mout_user_aclk400_disp1_p,
 			SRC_TOP5, 0, 1),
@@ -684,7 +685,8 @@ static struct samsung_mux_clock exynos5x_mux_clks[] __initdata = {
 			SRC_TOP11, 12, 1),
 	MUX(0, "mout_sw_aclk266", mout_sw_aclk266_p, SRC_TOP11, 20, 1),
 	MUX(0, "mout_sw_aclk166", mout_sw_aclk166_p, SRC_TOP11, 24, 1),
-	MUX(0, "mout_sw_aclk333", mout_sw_aclk333_p, SRC_TOP11, 28, 1),
+	MUX(CLK_MOUT_SW_ACLK333, "mout_sw_aclk333", mout_sw_aclk333_p,
+			SRC_TOP11, 28, 1),
 
 	MUX(0, "mout_sw_aclk400_disp1", mout_sw_aclk400_disp1_p,
 			SRC_TOP12, 4, 1),
@@ -890,8 +892,6 @@ static struct samsung_gate_clock exynos5x_gate_clks[] __initdata = {
 			GATE_BUS_TOP, 9, CLK_IGNORE_UNUSED, 0),
 	GATE(0, "aclk66_psgen", "mout_user_aclk66_psgen",
 			GATE_BUS_TOP, 10, CLK_IGNORE_UNUSED, 0),
-	GATE(CLK_ACLK66_PERIC, "aclk66_peric", "mout_user_aclk66_peric",
-			GATE_BUS_TOP, 11, CLK_IGNORE_UNUSED, 0),
 	GATE(0, "aclk266_isp", "mout_user_aclk266_isp",
 			GATE_BUS_TOP, 13, 0, 0),
 	GATE(0, "aclk166", "mout_user_aclk166",
@@ -994,34 +994,61 @@ static struct samsung_gate_clock exynos5x_gate_clks[] __initdata = {
 			SRC_MASK_FSYS, 24, CLK_SET_RATE_PARENT, 0),
 
 	/* PERIC Block */
-	GATE(CLK_UART0, "uart0", "aclk66_peric", GATE_IP_PERIC, 0, 0, 0),
-	GATE(CLK_UART1, "uart1", "aclk66_peric", GATE_IP_PERIC, 1, 0, 0),
-	GATE(CLK_UART2, "uart2", "aclk66_peric", GATE_IP_PERIC, 2, 0, 0),
-	GATE(CLK_UART3, "uart3", "aclk66_peric", GATE_IP_PERIC, 3, 0, 0),
-	GATE(CLK_I2C0, "i2c0", "aclk66_peric", GATE_IP_PERIC, 6, 0, 0),
-	GATE(CLK_I2C1, "i2c1", "aclk66_peric", GATE_IP_PERIC, 7, 0, 0),
-	GATE(CLK_I2C2, "i2c2", "aclk66_peric", GATE_IP_PERIC, 8, 0, 0),
-	GATE(CLK_I2C3, "i2c3", "aclk66_peric", GATE_IP_PERIC, 9, 0, 0),
-	GATE(CLK_USI0, "usi0", "aclk66_peric", GATE_IP_PERIC, 10, 0, 0),
-	GATE(CLK_USI1, "usi1", "aclk66_peric", GATE_IP_PERIC, 11, 0, 0),
-	GATE(CLK_USI2, "usi2", "aclk66_peric", GATE_IP_PERIC, 12, 0, 0),
-	GATE(CLK_USI3, "usi3", "aclk66_peric", GATE_IP_PERIC, 13, 0, 0),
-	GATE(CLK_I2C_HDMI, "i2c_hdmi", "aclk66_peric", GATE_IP_PERIC, 14, 0, 0),
-	GATE(CLK_TSADC, "tsadc", "aclk66_peric", GATE_IP_PERIC, 15, 0, 0),
-	GATE(CLK_SPI0, "spi0", "aclk66_peric", GATE_IP_PERIC, 16, 0, 0),
-	GATE(CLK_SPI1, "spi1", "aclk66_peric", GATE_IP_PERIC, 17, 0, 0),
-	GATE(CLK_SPI2, "spi2", "aclk66_peric", GATE_IP_PERIC, 18, 0, 0),
-	GATE(CLK_I2S1, "i2s1", "aclk66_peric", GATE_IP_PERIC, 20, 0, 0),
-	GATE(CLK_I2S2, "i2s2", "aclk66_peric", GATE_IP_PERIC, 21, 0, 0),
-	GATE(CLK_PCM1, "pcm1", "aclk66_peric", GATE_IP_PERIC, 22, 0, 0),
-	GATE(CLK_PCM2, "pcm2", "aclk66_peric", GATE_IP_PERIC, 23, 0, 0),
-	GATE(CLK_PWM, "pwm", "aclk66_peric", GATE_IP_PERIC, 24, 0, 0),
-	GATE(CLK_SPDIF, "spdif", "aclk66_peric", GATE_IP_PERIC, 26, 0, 0),
-	GATE(CLK_USI4, "usi4", "aclk66_peric", GATE_IP_PERIC, 28, 0, 0),
-	GATE(CLK_USI5, "usi5", "aclk66_peric", GATE_IP_PERIC, 30, 0, 0),
-	GATE(CLK_USI6, "usi6", "aclk66_peric", GATE_IP_PERIC, 31, 0, 0),
-
-	GATE(CLK_KEYIF, "keyif", "aclk66_peric", GATE_BUS_PERIC, 22, 0, 0),
+	GATE(CLK_UART0, "uart0", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 0, 0, 0),
+	GATE(CLK_UART1, "uart1", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 1, 0, 0),
+	GATE(CLK_UART2, "uart2", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 2, 0, 0),
+	GATE(CLK_UART3, "uart3", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 3, 0, 0),
+	GATE(CLK_I2C0, "i2c0", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 6, 0, 0),
+	GATE(CLK_I2C1, "i2c1", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 7, 0, 0),
+	GATE(CLK_I2C2, "i2c2", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 8, 0, 0),
+	GATE(CLK_I2C3, "i2c3", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 9, 0, 0),
+	GATE(CLK_USI0, "usi0", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 10, 0, 0),
+	GATE(CLK_USI1, "usi1", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 11, 0, 0),
+	GATE(CLK_USI2, "usi2", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 12, 0, 0),
+	GATE(CLK_USI3, "usi3", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 13, 0, 0),
+	GATE(CLK_I2C_HDMI, "i2c_hdmi", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 14, 0, 0),
+	GATE(CLK_TSADC, "tsadc", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 15, 0, 0),
+	GATE(CLK_SPI0, "spi0", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 16, 0, 0),
+	GATE(CLK_SPI1, "spi1", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 17, 0, 0),
+	GATE(CLK_SPI2, "spi2", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 18, 0, 0),
+	GATE(CLK_I2S1, "i2s1", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 20, 0, 0),
+	GATE(CLK_I2S2, "i2s2", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 21, 0, 0),
+	GATE(CLK_PCM1, "pcm1", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 22, 0, 0),
+	GATE(CLK_PCM2, "pcm2", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 23, 0, 0),
+	GATE(CLK_PWM, "pwm", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 24, 0, 0),
+	GATE(CLK_SPDIF, "spdif", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 26, 0, 0),
+	GATE(CLK_USI4, "usi4", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 28, 0, 0),
+	GATE(CLK_USI5, "usi5", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 30, 0, 0),
+	GATE(CLK_USI6, "usi6", "mout_user_aclk66_peric",
+			GATE_IP_PERIC, 31, 0, 0),
+
+	GATE(CLK_KEYIF, "keyif", "mout_user_aclk66_peric",
+			GATE_BUS_PERIC, 22, 0, 0),
 
 	/* PERIS Block */
 	GATE(CLK_CHIPID, "chipid", "aclk66_psgen",
diff --git a/drivers/clk/samsung/clk-s3c2410.c b/drivers/clk/samsung/clk-s3c2410.c
index ba0716801db2..140f4733c02e 100644
--- a/drivers/clk/samsung/clk-s3c2410.c
+++ b/drivers/clk/samsung/clk-s3c2410.c
@@ -152,6 +152,11 @@ struct samsung_clock_alias s3c2410_common_aliases[] __initdata = {
 	ALIAS(HCLK, NULL, "hclk"),
 	ALIAS(MPLL, NULL, "mpll"),
 	ALIAS(FCLK, NULL, "fclk"),
+	ALIAS(PCLK, NULL, "watchdog"),
+	ALIAS(PCLK_SDI, NULL, "sdi"),
+	ALIAS(HCLK_NAND, NULL, "nand"),
+	ALIAS(PCLK_I2S, NULL, "iis"),
+	ALIAS(PCLK_I2C, NULL, "i2c"),
 };
 
 /* S3C2410 specific clocks */
@@ -378,7 +383,7 @@ void __init s3c2410_common_clk_init(struct device_node *np, unsigned long xti_f,
 	if (!np)
 		s3c2410_common_clk_register_fixed_ext(ctx, xti_f);
 
-	if (current_soc == 2410) {
+	if (current_soc == S3C2410) {
 		if (_get_rate("xti") == 12 * MHZ) {
 			s3c2410_plls[mpll].rate_table = pll_s3c2410_12mhz_tbl;
 			s3c2410_plls[upll].rate_table = pll_s3c2410_12mhz_tbl;
@@ -432,7 +437,7 @@ void __init s3c2410_common_clk_init(struct device_node *np, unsigned long xti_f,
 		samsung_clk_register_fixed_factor(ctx, s3c2410_ffactor,
 				ARRAY_SIZE(s3c2410_ffactor));
 		samsung_clk_register_alias(ctx, s3c2410_aliases,
-			ARRAY_SIZE(s3c2410_common_aliases));
+			ARRAY_SIZE(s3c2410_aliases));
 		break;
 	case S3C2440:
 		samsung_clk_register_mux(ctx, s3c2440_muxes,
diff --git a/drivers/clk/samsung/clk-s3c64xx.c b/drivers/clk/samsung/clk-s3c64xx.c
index efa16ee592c8..8889ff1c10fc 100644
--- a/drivers/clk/samsung/clk-s3c64xx.c
+++ b/drivers/clk/samsung/clk-s3c64xx.c
@@ -418,8 +418,10 @@ static struct samsung_clock_alias s3c64xx_clock_aliases[] = {
 	ALIAS(SCLK_MMC2, "s3c-sdhci.2", "mmc_busclk.2"),
 	ALIAS(SCLK_MMC1, "s3c-sdhci.1", "mmc_busclk.2"),
 	ALIAS(SCLK_MMC0, "s3c-sdhci.0", "mmc_busclk.2"),
-	ALIAS(SCLK_SPI1, "s3c6410-spi.1", "spi-bus"),
-	ALIAS(SCLK_SPI0, "s3c6410-spi.0", "spi-bus"),
+	ALIAS(PCLK_SPI1, "s3c6410-spi.1", "spi_busclk0"),
+	ALIAS(SCLK_SPI1, "s3c6410-spi.1", "spi_busclk2"),
+	ALIAS(PCLK_SPI0, "s3c6410-spi.0", "spi_busclk0"),
+	ALIAS(SCLK_SPI0, "s3c6410-spi.0", "spi_busclk2"),
 	ALIAS(SCLK_AUDIO1, "samsung-pcm.1", "audio-bus"),
 	ALIAS(SCLK_AUDIO1, "samsung-i2s.1", "audio-bus"),
 	ALIAS(SCLK_AUDIO0, "samsung-pcm.0", "audio-bus"),
diff --git a/drivers/clk/spear/spear3xx_clock.c b/drivers/clk/spear/spear3xx_clock.c
index c2d204315546..bb5f387774e2 100644
--- a/drivers/clk/spear/spear3xx_clock.c
+++ b/drivers/clk/spear/spear3xx_clock.c
@@ -211,7 +211,7 @@ static inline void spear310_clk_init(void) { }
 /* array of all spear 320 clock lookups */
 #ifdef CONFIG_MACH_SPEAR320
 
-#define SPEAR320_CONTROL_REG		(soc_config_base + 0x0000)
+#define SPEAR320_CONTROL_REG		(soc_config_base + 0x0010)
 #define SPEAR320_EXT_CTRL_REG		(soc_config_base + 0x0018)
 
 	#define SPEAR320_UARTX_PCLK_MASK		0x1
@@ -245,7 +245,8 @@ static const char *smii0_parents[] = { "smii_125m_pad", "ras_pll2_clk",
 	"ras_syn0_gclk", };
 static const char *uartx_parents[] = { "ras_syn1_gclk", "ras_apb_clk", };
 
-static void __init spear320_clk_init(void __iomem *soc_config_base)
+static void __init spear320_clk_init(void __iomem *soc_config_base,
+				     struct clk *ras_apb_clk)
 {
 	struct clk *clk;
 
@@ -342,6 +343,8 @@ static void __init spear320_clk_init(void __iomem *soc_config_base)
 			SPEAR320_CONTROL_REG, UART1_PCLK_SHIFT, UART1_PCLK_MASK,
 			0, &_lock);
 	clk_register_clkdev(clk, NULL, "a3000000.serial");
+	/* Enforce ras_apb_clk */
+	clk_set_parent(clk, ras_apb_clk);
 
 	clk = clk_register_mux(NULL, "uart2_clk", uartx_parents,
 			ARRAY_SIZE(uartx_parents),
@@ -349,6 +352,8 @@ static void __init spear320_clk_init(void __iomem *soc_config_base)
 			SPEAR320_EXT_CTRL_REG, SPEAR320_UART2_PCLK_SHIFT,
 			SPEAR320_UARTX_PCLK_MASK, 0, &_lock);
 	clk_register_clkdev(clk, NULL, "a4000000.serial");
+	/* Enforce ras_apb_clk */
+	clk_set_parent(clk, ras_apb_clk);
 
 	clk = clk_register_mux(NULL, "uart3_clk", uartx_parents,
 			ARRAY_SIZE(uartx_parents),
@@ -379,12 +384,12 @@ static void __init spear320_clk_init(void __iomem *soc_config_base)
 	clk_register_clkdev(clk, NULL, "60100000.serial");
 }
 #else
-static inline void spear320_clk_init(void __iomem *soc_config_base) { }
+static inline void spear320_clk_init(void __iomem *sb, struct clk *rc) { }
 #endif
 
 void __init spear3xx_clk_init(void __iomem *misc_base, void __iomem *soc_config_base)
 {
-	struct clk *clk, *clk1;
+	struct clk *clk, *clk1, *ras_apb_clk;
 
 	clk = clk_register_fixed_rate(NULL, "osc_32k_clk", NULL, CLK_IS_ROOT,
 			32000);
@@ -613,6 +618,7 @@ void __init spear3xx_clk_init(void __iomem *misc_base, void __iomem *soc_config_
 	clk = clk_register_gate(NULL, "ras_apb_clk", "apb_clk", 0, RAS_CLK_ENB,
 			RAS_APB_CLK_ENB, 0, &_lock);
 	clk_register_clkdev(clk, "ras_apb_clk", NULL);
+	ras_apb_clk = clk;
 
 	clk = clk_register_gate(NULL, "ras_32k_clk", "osc_32k_clk", 0,
 			RAS_CLK_ENB, RAS_32K_CLK_ENB, 0, &_lock);
@@ -659,5 +665,5 @@ void __init spear3xx_clk_init(void __iomem *misc_base, void __iomem *soc_config_
 	else if (of_machine_is_compatible("st,spear310"))
 		spear310_clk_init();
 	else if (of_machine_is_compatible("st,spear320"))
-		spear320_clk_init(soc_config_base);
+		spear320_clk_init(soc_config_base, ras_apb_clk);
 }
diff --git a/drivers/clk/sunxi/Makefile b/drivers/clk/sunxi/Makefile
index b5bac917612c..762fd64dbd1f 100644
--- a/drivers/clk/sunxi/Makefile
+++ b/drivers/clk/sunxi/Makefile
@@ -3,3 +3,7 @@
 #
 
 obj-y += clk-sunxi.o clk-factors.o
+obj-y += clk-a10-hosc.o
+obj-y += clk-a20-gmac.o
+
+obj-$(CONFIG_MFD_SUN6I_PRCM) += clk-sun6i-ar100.o clk-sun6i-apb0.o clk-sun6i-apb0-gates.o
diff --git a/drivers/clk/sunxi/clk-a10-hosc.c b/drivers/clk/sunxi/clk-a10-hosc.c
new file mode 100644
index 000000000000..0481d5d673d6
--- /dev/null
+++ b/drivers/clk/sunxi/clk-a10-hosc.c
@@ -0,0 +1,73 @@
+/*
+ * Copyright 2013 Emilio López
+ *
+ * Emilio López <emilio@elopez.com.ar>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+
+#define SUNXI_OSC24M_GATE	0
+
+static DEFINE_SPINLOCK(hosc_lock);
+
+static void __init sun4i_osc_clk_setup(struct device_node *node)
+{
+	struct clk *clk;
+	struct clk_fixed_rate *fixed;
+	struct clk_gate *gate;
+	const char *clk_name = node->name;
+	u32 rate;
+
+	if (of_property_read_u32(node, "clock-frequency", &rate))
+		return;
+
+	/* allocate fixed-rate and gate clock structs */
+	fixed = kzalloc(sizeof(struct clk_fixed_rate), GFP_KERNEL);
+	if (!fixed)
+		return;
+	gate = kzalloc(sizeof(struct clk_gate), GFP_KERNEL);
+	if (!gate)
+		goto err_free_fixed;
+
+	of_property_read_string(node, "clock-output-names", &clk_name);
+
+	/* set up gate and fixed rate properties */
+	gate->reg = of_iomap(node, 0);
+	gate->bit_idx = SUNXI_OSC24M_GATE;
+	gate->lock = &hosc_lock;
+	fixed->fixed_rate = rate;
+
+	clk = clk_register_composite(NULL, clk_name,
+			NULL, 0,
+			NULL, NULL,
+			&fixed->hw, &clk_fixed_rate_ops,
+			&gate->hw, &clk_gate_ops,
+			CLK_IS_ROOT);
+
+	if (IS_ERR(clk))
+		goto err_free_gate;
+
+	of_clk_add_provider(node, of_clk_src_simple_get, clk);
+	clk_register_clkdev(clk, clk_name, NULL);
+
+	return;
+
+err_free_gate:
+	kfree(gate);
+err_free_fixed:
+	kfree(fixed);
+}
+CLK_OF_DECLARE(sun4i_osc, "allwinner,sun4i-a10-osc-clk", sun4i_osc_clk_setup);
diff --git a/drivers/clk/sunxi/clk-a20-gmac.c b/drivers/clk/sunxi/clk-a20-gmac.c
new file mode 100644
index 000000000000..633ddc4389ef
--- /dev/null
+++ b/drivers/clk/sunxi/clk-a20-gmac.c
@@ -0,0 +1,119 @@
+/*
+ * Copyright 2013 Emilio López
+ * Emilio López <emilio@elopez.com.ar>
+ *
+ * Copyright 2013 Chen-Yu Tsai
+ * Chen-Yu Tsai <wens@csie.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/slab.h>
+
+static DEFINE_SPINLOCK(gmac_lock);
+
+/**
+ * sun7i_a20_gmac_clk_setup - Setup function for A20/A31 GMAC clock module
+ *
+ * This clock looks something like this
+ *                               ________________________
+ *  MII TX clock from PHY >-----|___________    _________|----> to GMAC core
+ *  GMAC Int. RGMII TX clk >----|___________\__/__gate---|----> to PHY
+ *  Ext. 125MHz RGMII TX clk >--|__divider__/            |
+ *                              |________________________|
+ *
+ * The external 125 MHz reference is optional, i.e. GMAC can use its
+ * internal TX clock just fine. The A31 GMAC clock module does not have
+ * the divider controls for the external reference.
+ *
+ * To keep it simple, let the GMAC use either the MII TX clock for MII mode,
+ * and its internal TX clock for GMII and RGMII modes. The GMAC driver should
+ * select the appropriate source and gate/ungate the output to the PHY.
+ *
+ * Only the GMAC should use this clock. Altering the clock so that it doesn't
+ * match the GMAC's operation parameters will result in the GMAC not being
+ * able to send traffic out. The GMAC driver should set the clock rate and
+ * enable/disable this clock to configure the required state. The clock
+ * driver then responds by auto-reparenting the clock.
+ */
+
+#define SUN7I_A20_GMAC_GPIT	2
+#define SUN7I_A20_GMAC_MASK	0x3
+#define SUN7I_A20_GMAC_PARENTS	2
+
+static void __init sun7i_a20_gmac_clk_setup(struct device_node *node)
+{
+	struct clk *clk;
+	struct clk_mux *mux;
+	struct clk_gate *gate;
+	const char *clk_name = node->name;
+	const char *parents[SUN7I_A20_GMAC_PARENTS];
+	void *reg;
+
+	if (of_property_read_string(node, "clock-output-names", &clk_name))
+		return;
+
+	/* allocate mux and gate clock structs */
+	mux = kzalloc(sizeof(struct clk_mux), GFP_KERNEL);
+	if (!mux)
+		return;
+
+	gate = kzalloc(sizeof(struct clk_gate), GFP_KERNEL);
+	if (!gate)
+		goto free_mux;
+
+	/* gmac clock requires exactly 2 parents */
+	parents[0] = of_clk_get_parent_name(node, 0);
+	parents[1] = of_clk_get_parent_name(node, 1);
+	if (!parents[0] || !parents[1])
+		goto free_gate;
+
+	reg = of_iomap(node, 0);
+	if (!reg)
+		goto free_gate;
+
+	/* set up gate and fixed rate properties */
+	gate->reg = reg;
+	gate->bit_idx = SUN7I_A20_GMAC_GPIT;
+	gate->lock = &gmac_lock;
+	mux->reg = reg;
+	mux->mask = SUN7I_A20_GMAC_MASK;
+	mux->flags = CLK_MUX_INDEX_BIT;
+	mux->lock = &gmac_lock;
+
+	clk = clk_register_composite(NULL, clk_name,
+			parents, SUN7I_A20_GMAC_PARENTS,
+			&mux->hw, &clk_mux_ops,
+			NULL, NULL,
+			&gate->hw, &clk_gate_ops,
+			0);
+
+	if (IS_ERR(clk))
+		goto iounmap_reg;
+
+	of_clk_add_provider(node, of_clk_src_simple_get, clk);
+	clk_register_clkdev(clk, clk_name, NULL);
+
+	return;
+
+iounmap_reg:
+	iounmap(reg);
+free_gate:
+	kfree(gate);
+free_mux:
+	kfree(mux);
+}
+CLK_OF_DECLARE(sun7i_a20_gmac, "allwinner,sun7i-a20-gmac-clk",
+		sun7i_a20_gmac_clk_setup);
diff --git a/drivers/clk/sunxi/clk-sun6i-apb0-gates.c b/drivers/clk/sunxi/clk-sun6i-apb0-gates.c
new file mode 100644
index 000000000000..670f90d629d7
--- /dev/null
+++ b/drivers/clk/sunxi/clk-sun6i-apb0-gates.c
@@ -0,0 +1,99 @@
+/*
+ * Copyright (C) 2014 Free Electrons
+ *
+ * License Terms: GNU General Public License v2
+ * Author: Boris BREZILLON <boris.brezillon@free-electrons.com>
+ *
+ * Allwinner A31 APB0 clock gates driver
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+#define SUN6I_APB0_GATES_MAX_SIZE	32
+
+static int sun6i_a31_apb0_gates_clk_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct clk_onecell_data *clk_data;
+	const char *clk_parent;
+	const char *clk_name;
+	struct resource *r;
+	void __iomem *reg;
+	int gate_id;
+	int ngates;
+	int i;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	reg = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(reg))
+		return PTR_ERR(reg);
+
+	clk_parent = of_clk_get_parent_name(np, 0);
+	if (!clk_parent)
+		return -EINVAL;
+
+	ngates = of_property_count_strings(np, "clock-output-names");
+	if (ngates < 0)
+		return ngates;
+
+	if (!ngates || ngates > SUN6I_APB0_GATES_MAX_SIZE)
+		return -EINVAL;
+
+	clk_data = devm_kzalloc(&pdev->dev, sizeof(struct clk_onecell_data),
+				GFP_KERNEL);
+	if (!clk_data)
+		return -ENOMEM;
+
+	clk_data->clks = devm_kzalloc(&pdev->dev,
+				      SUN6I_APB0_GATES_MAX_SIZE *
+				      sizeof(struct clk *),
+				      GFP_KERNEL);
+	if (!clk_data->clks)
+		return -ENOMEM;
+
+	for (i = 0; i < ngates; i++) {
+		of_property_read_string_index(np, "clock-output-names",
+					      i, &clk_name);
+
+		gate_id = i;
+		of_property_read_u32_index(np, "clock-indices", i, &gate_id);
+
+		WARN_ON(gate_id >= SUN6I_APB0_GATES_MAX_SIZE);
+		if (gate_id >= SUN6I_APB0_GATES_MAX_SIZE)
+			continue;
+
+		clk_data->clks[gate_id] = clk_register_gate(&pdev->dev,
+							    clk_name,
+							    clk_parent, 0,
+							    reg, gate_id,
+							    0, NULL);
+		WARN_ON(IS_ERR(clk_data->clks[gate_id]));
+	}
+
+	clk_data->clk_num = ngates;
+
+	return of_clk_add_provider(np, of_clk_src_onecell_get, clk_data);
+}
+
+const struct of_device_id sun6i_a31_apb0_gates_clk_dt_ids[] = {
+	{ .compatible = "allwinner,sun6i-a31-apb0-gates-clk" },
+	{ /* sentinel */ }
+};
+
+static struct platform_driver sun6i_a31_apb0_gates_clk_driver = {
+	.driver = {
+		.name = "sun6i-a31-apb0-gates-clk",
+		.owner = THIS_MODULE,
+		.of_match_table = sun6i_a31_apb0_gates_clk_dt_ids,
+	},
+	.probe = sun6i_a31_apb0_gates_clk_probe,
+};
+module_platform_driver(sun6i_a31_apb0_gates_clk_driver);
+
+MODULE_AUTHOR("Boris BREZILLON <boris.brezillon@free-electrons.com>");
+MODULE_DESCRIPTION("Allwinner A31 APB0 gate clocks driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/clk/sunxi/clk-sun6i-apb0.c b/drivers/clk/sunxi/clk-sun6i-apb0.c
new file mode 100644
index 000000000000..11f17c34c2ae
--- /dev/null
+++ b/drivers/clk/sunxi/clk-sun6i-apb0.c
@@ -0,0 +1,77 @@
+/*
+ * Copyright (C) 2014 Free Electrons
+ *
+ * License Terms: GNU General Public License v2
+ * Author: Boris BREZILLON <boris.brezillon@free-electrons.com>
+ *
+ * Allwinner A31 APB0 clock driver
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+/*
+ * The APB0 clk has a configurable divisor.
+ *
+ * We must use a clk_div_table and not a regular power of 2
+ * divisor here, because the first 2 values divide the clock
+ * by 2.
+ */
+static const struct clk_div_table sun6i_a31_apb0_divs[] = {
+	{ .val = 0, .div = 2, },
+	{ .val = 1, .div = 2, },
+	{ .val = 2, .div = 4, },
+	{ .val = 3, .div = 8, },
+	{ /* sentinel */ },
+};
+
+static int sun6i_a31_apb0_clk_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	const char *clk_name = np->name;
+	const char *clk_parent;
+	struct resource *r;
+	void __iomem *reg;
+	struct clk *clk;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	reg = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(reg))
+		return PTR_ERR(reg);
+
+	clk_parent = of_clk_get_parent_name(np, 0);
+	if (!clk_parent)
+		return -EINVAL;
+
+	of_property_read_string(np, "clock-output-names", &clk_name);
+
+	clk = clk_register_divider_table(&pdev->dev, clk_name, clk_parent,
+					 0, reg, 0, 2, 0, sun6i_a31_apb0_divs,
+					 NULL);
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
+
+	return of_clk_add_provider(np, of_clk_src_simple_get, clk);
+}
+
+const struct of_device_id sun6i_a31_apb0_clk_dt_ids[] = {
+	{ .compatible = "allwinner,sun6i-a31-apb0-clk" },
+	{ /* sentinel */ }
+};
+
+static struct platform_driver sun6i_a31_apb0_clk_driver = {
+	.driver = {
+		.name = "sun6i-a31-apb0-clk",
+		.owner = THIS_MODULE,
+		.of_match_table = sun6i_a31_apb0_clk_dt_ids,
+	},
+	.probe = sun6i_a31_apb0_clk_probe,
+};
+module_platform_driver(sun6i_a31_apb0_clk_driver);
+
+MODULE_AUTHOR("Boris BREZILLON <boris.brezillon@free-electrons.com>");
+MODULE_DESCRIPTION("Allwinner A31 APB0 clock Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/clk/sunxi/clk-sun6i-ar100.c b/drivers/clk/sunxi/clk-sun6i-ar100.c
new file mode 100644
index 000000000000..f73cc051f0dd
--- /dev/null
+++ b/drivers/clk/sunxi/clk-sun6i-ar100.c
@@ -0,0 +1,233 @@
+/*
+ * Copyright (C) 2014 Free Electrons
+ *
+ * License Terms: GNU General Public License v2
+ * Author: Boris BREZILLON <boris.brezillon@free-electrons.com>
+ *
+ * Allwinner A31 AR100 clock driver
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+#define SUN6I_AR100_MAX_PARENTS		4
+#define SUN6I_AR100_SHIFT_MASK		0x3
+#define SUN6I_AR100_SHIFT_MAX		SUN6I_AR100_SHIFT_MASK
+#define SUN6I_AR100_SHIFT_SHIFT		4
+#define SUN6I_AR100_DIV_MASK		0x1f
+#define SUN6I_AR100_DIV_MAX		(SUN6I_AR100_DIV_MASK + 1)
+#define SUN6I_AR100_DIV_SHIFT		8
+#define SUN6I_AR100_MUX_MASK		0x3
+#define SUN6I_AR100_MUX_SHIFT		16
+
+struct ar100_clk {
+	struct clk_hw hw;
+	void __iomem *reg;
+};
+
+static inline struct ar100_clk *to_ar100_clk(struct clk_hw *hw)
+{
+	return container_of(hw, struct ar100_clk, hw);
+}
+
+static unsigned long ar100_recalc_rate(struct clk_hw *hw,
+				       unsigned long parent_rate)
+{
+	struct ar100_clk *clk = to_ar100_clk(hw);
+	u32 val = readl(clk->reg);
+	int shift = (val >> SUN6I_AR100_SHIFT_SHIFT) & SUN6I_AR100_SHIFT_MASK;
+	int div = (val >> SUN6I_AR100_DIV_SHIFT) & SUN6I_AR100_DIV_MASK;
+
+	return (parent_rate >> shift) / (div + 1);
+}
+
+static long ar100_determine_rate(struct clk_hw *hw, unsigned long rate,
+				 unsigned long *best_parent_rate,
+				 struct clk **best_parent_clk)
+{
+	int nparents = __clk_get_num_parents(hw->clk);
+	long best_rate = -EINVAL;
+	int i;
+
+	*best_parent_clk = NULL;
+
+	for (i = 0; i < nparents; i++) {
+		unsigned long parent_rate;
+		unsigned long tmp_rate;
+		struct clk *parent;
+		unsigned long div;
+		int shift;
+
+		parent = clk_get_parent_by_index(hw->clk, i);
+		parent_rate = __clk_get_rate(parent);
+		div = DIV_ROUND_UP(parent_rate, rate);
+
+		/*
+		 * The AR100 clk contains 2 divisors:
+		 * - one power of 2 divisor
+		 * - one regular divisor
+		 *
+		 * First check if we can safely shift (or divide by a power
+		 * of 2) without losing precision on the requested rate.
+		 */
+		shift = ffs(div) - 1;
+		if (shift > SUN6I_AR100_SHIFT_MAX)
+			shift = SUN6I_AR100_SHIFT_MAX;
+
+		div >>= shift;
+
+		/*
+		 * Then if the divisor is still bigger than what the HW
+		 * actually supports, use a bigger shift (or power of 2
+		 * divider) value and accept to lose some precision.
+		 */
+		while (div > SUN6I_AR100_DIV_MAX) {
+			shift++;
+			div >>= 1;
+			if (shift > SUN6I_AR100_SHIFT_MAX)
+				break;
+		}
+
+		/*
+		 * If the shift value (or power of 2 divider) is bigger
+		 * than what the HW actually support, skip this parent.
+		 */
+		if (shift > SUN6I_AR100_SHIFT_MAX)
+			continue;
+
+		tmp_rate = (parent_rate >> shift) / div;
+		if (!*best_parent_clk || tmp_rate > best_rate) {
+			*best_parent_clk = parent;
+			*best_parent_rate = parent_rate;
+			best_rate = tmp_rate;
+		}
+	}
+
+	return best_rate;
+}
+
+static int ar100_set_parent(struct clk_hw *hw, u8 index)
+{
+	struct ar100_clk *clk = to_ar100_clk(hw);
+	u32 val = readl(clk->reg);
+
+	if (index >= SUN6I_AR100_MAX_PARENTS)
+		return -EINVAL;
+
+	val &= ~(SUN6I_AR100_MUX_MASK << SUN6I_AR100_MUX_SHIFT);
+	val |= (index << SUN6I_AR100_MUX_SHIFT);
+	writel(val, clk->reg);
+
+	return 0;
+}
+
+static u8 ar100_get_parent(struct clk_hw *hw)
+{
+	struct ar100_clk *clk = to_ar100_clk(hw);
+	return (readl(clk->reg) >> SUN6I_AR100_MUX_SHIFT) &
+	       SUN6I_AR100_MUX_MASK;
+}
+
+static int ar100_set_rate(struct clk_hw *hw, unsigned long rate,
+			  unsigned long parent_rate)
+{
+	unsigned long div = parent_rate / rate;
+	struct ar100_clk *clk = to_ar100_clk(hw);
+	u32 val = readl(clk->reg);
+	int shift;
+
+	if (parent_rate % rate)
+		return -EINVAL;
+
+	shift = ffs(div) - 1;
+	if (shift > SUN6I_AR100_SHIFT_MAX)
+		shift = SUN6I_AR100_SHIFT_MAX;
+
+	div >>= shift;
+
+	if (div > SUN6I_AR100_DIV_MAX)
+		return -EINVAL;
+
+	val &= ~((SUN6I_AR100_SHIFT_MASK << SUN6I_AR100_SHIFT_SHIFT) |
+		 (SUN6I_AR100_DIV_MASK << SUN6I_AR100_DIV_SHIFT));
+	val |= (shift << SUN6I_AR100_SHIFT_SHIFT) |
+	       (div << SUN6I_AR100_DIV_SHIFT);
+	writel(val, clk->reg);
+
+	return 0;
+}
+
+struct clk_ops ar100_ops = {
+	.recalc_rate = ar100_recalc_rate,
+	.determine_rate = ar100_determine_rate,
+	.set_parent = ar100_set_parent,
+	.get_parent = ar100_get_parent,
+	.set_rate = ar100_set_rate,
+};
+
+static int sun6i_a31_ar100_clk_probe(struct platform_device *pdev)
+{
+	const char *parents[SUN6I_AR100_MAX_PARENTS];
+	struct device_node *np = pdev->dev.of_node;
+	const char *clk_name = np->name;
+	struct clk_init_data init;
+	struct ar100_clk *ar100;
+	struct resource *r;
+	struct clk *clk;
+	int nparents;
+	int i;
+
+	ar100 = devm_kzalloc(&pdev->dev, sizeof(*ar100), GFP_KERNEL);
+	if (!ar100)
+		return -ENOMEM;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	ar100->reg = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(ar100->reg))
+		return PTR_ERR(ar100->reg);
+
+	nparents = of_clk_get_parent_count(np);
+	if (nparents > SUN6I_AR100_MAX_PARENTS)
+		nparents = SUN6I_AR100_MAX_PARENTS;
+
+	for (i = 0; i < nparents; i++)
+		parents[i] = of_clk_get_parent_name(np, i);
+
+	of_property_read_string(np, "clock-output-names", &clk_name);
+
+	init.name = clk_name;
+	init.ops = &ar100_ops;
+	init.parent_names = parents;
+	init.num_parents = nparents;
+	init.flags = 0;
+
+	ar100->hw.init = &init;
+
+	clk = clk_register(&pdev->dev, &ar100->hw);
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
+
+	return of_clk_add_provider(np, of_clk_src_simple_get, clk);
+}
+
+const struct of_device_id sun6i_a31_ar100_clk_dt_ids[] = {
+	{ .compatible = "allwinner,sun6i-a31-ar100-clk" },
+	{ /* sentinel */ }
+};
+
+static struct platform_driver sun6i_a31_ar100_clk_driver = {
+	.driver = {
+		.name = "sun6i-a31-ar100-clk",
+		.owner = THIS_MODULE,
+		.of_match_table = sun6i_a31_ar100_clk_dt_ids,
+	},
+	.probe = sun6i_a31_ar100_clk_probe,
+};
+module_platform_driver(sun6i_a31_ar100_clk_driver);
+
+MODULE_AUTHOR("Boris BREZILLON <boris.brezillon@free-electrons.com>");
+MODULE_DESCRIPTION("Allwinner A31 AR100 clock Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/clk/sunxi/clk-sunxi.c b/drivers/clk/sunxi/clk-sunxi.c
index 426483422d3d..fb2ce8440f0e 100644
--- a/drivers/clk/sunxi/clk-sunxi.c
+++ b/drivers/clk/sunxi/clk-sunxi.c
@@ -28,63 +28,6 @@ static DEFINE_SPINLOCK(clk_lock);
 #define SUNXI_MAX_PARENTS	5
 
 /**
- * sun4i_osc_clk_setup() - Setup function for gatable oscillator
- */
-
-#define SUNXI_OSC24M_GATE	0
-
-static void __init sun4i_osc_clk_setup(struct device_node *node)
-{
-	struct clk *clk;
-	struct clk_fixed_rate *fixed;
-	struct clk_gate *gate;
-	const char *clk_name = node->name;
-	u32 rate;
-
-	if (of_property_read_u32(node, "clock-frequency", &rate))
-		return;
-
-	/* allocate fixed-rate and gate clock structs */
-	fixed = kzalloc(sizeof(struct clk_fixed_rate), GFP_KERNEL);
-	if (!fixed)
-		return;
-	gate = kzalloc(sizeof(struct clk_gate), GFP_KERNEL);
-	if (!gate)
-		goto err_free_fixed;
-
-	of_property_read_string(node, "clock-output-names", &clk_name);
-
-	/* set up gate and fixed rate properties */
-	gate->reg = of_iomap(node, 0);
-	gate->bit_idx = SUNXI_OSC24M_GATE;
-	gate->lock = &clk_lock;
-	fixed->fixed_rate = rate;
-
-	clk = clk_register_composite(NULL, clk_name,
-			NULL, 0,
-			NULL, NULL,
-			&fixed->hw, &clk_fixed_rate_ops,
-			&gate->hw, &clk_gate_ops,
-			CLK_IS_ROOT);
-
-	if (IS_ERR(clk))
-		goto err_free_gate;
-
-	of_clk_add_provider(node, of_clk_src_simple_get, clk);
-	clk_register_clkdev(clk, clk_name, NULL);
-
-	return;
-
-err_free_gate:
-	kfree(gate);
-err_free_fixed:
-	kfree(fixed);
-}
-CLK_OF_DECLARE(sun4i_osc, "allwinner,sun4i-a10-osc-clk", sun4i_osc_clk_setup);
-
-
-
-/**
  * sun4i_get_pll1_factors() - calculates n, k, m, p factors for PLL1
  * PLL1 rate is calculated as follows
  * rate = (parent_rate * n * (k + 1) >> p) / (m + 1);
@@ -408,104 +351,6 @@ static void sun7i_a20_get_out_factors(u32 *freq, u32 parent_rate,
 	*p = calcp;
 }
 
-
-
-/**
- * sun7i_a20_gmac_clk_setup - Setup function for A20/A31 GMAC clock module
- *
- * This clock looks something like this
- *                               ________________________
- *  MII TX clock from PHY >-----|___________    _________|----> to GMAC core
- *  GMAC Int. RGMII TX clk >----|___________\__/__gate---|----> to PHY
- *  Ext. 125MHz RGMII TX clk >--|__divider__/            |
- *                              |________________________|
- *
- * The external 125 MHz reference is optional, i.e. GMAC can use its
- * internal TX clock just fine. The A31 GMAC clock module does not have
- * the divider controls for the external reference.
- *
- * To keep it simple, let the GMAC use either the MII TX clock for MII mode,
- * and its internal TX clock for GMII and RGMII modes. The GMAC driver should
- * select the appropriate source and gate/ungate the output to the PHY.
- *
- * Only the GMAC should use this clock. Altering the clock so that it doesn't
- * match the GMAC's operation parameters will result in the GMAC not being
- * able to send traffic out. The GMAC driver should set the clock rate and
- * enable/disable this clock to configure the required state. The clock
- * driver then responds by auto-reparenting the clock.
- */
-
-#define SUN7I_A20_GMAC_GPIT	2
-#define SUN7I_A20_GMAC_MASK	0x3
-#define SUN7I_A20_GMAC_PARENTS	2
-
-static void __init sun7i_a20_gmac_clk_setup(struct device_node *node)
-{
-	struct clk *clk;
-	struct clk_mux *mux;
-	struct clk_gate *gate;
-	const char *clk_name = node->name;
-	const char *parents[SUN7I_A20_GMAC_PARENTS];
-	void *reg;
-
-	if (of_property_read_string(node, "clock-output-names", &clk_name))
-		return;
-
-	/* allocate mux and gate clock structs */
-	mux = kzalloc(sizeof(struct clk_mux), GFP_KERNEL);
-	if (!mux)
-		return;
-
-	gate = kzalloc(sizeof(struct clk_gate), GFP_KERNEL);
-	if (!gate)
-		goto free_mux;
-
-	/* gmac clock requires exactly 2 parents */
-	parents[0] = of_clk_get_parent_name(node, 0);
-	parents[1] = of_clk_get_parent_name(node, 1);
-	if (!parents[0] || !parents[1])
-		goto free_gate;
-
-	reg = of_iomap(node, 0);
-	if (!reg)
-		goto free_gate;
-
-	/* set up gate and fixed rate properties */
-	gate->reg = reg;
-	gate->bit_idx = SUN7I_A20_GMAC_GPIT;
-	gate->lock = &clk_lock;
-	mux->reg = reg;
-	mux->mask = SUN7I_A20_GMAC_MASK;
-	mux->flags = CLK_MUX_INDEX_BIT;
-	mux->lock = &clk_lock;
-
-	clk = clk_register_composite(NULL, clk_name,
-			parents, SUN7I_A20_GMAC_PARENTS,
-			&mux->hw, &clk_mux_ops,
-			NULL, NULL,
-			&gate->hw, &clk_gate_ops,
-			0);
-
-	if (IS_ERR(clk))
-		goto iounmap_reg;
-
-	of_clk_add_provider(node, of_clk_src_simple_get, clk);
-	clk_register_clkdev(clk, clk_name, NULL);
-
-	return;
-
-iounmap_reg:
-	iounmap(reg);
-free_gate:
-	kfree(gate);
-free_mux:
-	kfree(mux);
-}
-CLK_OF_DECLARE(sun7i_a20_gmac, "allwinner,sun7i-a20-gmac-clk",
-		sun7i_a20_gmac_clk_setup);
-
-
-
 /**
  * clk_sunxi_mmc_phase_control() - configures MMC clock phase control
  */
@@ -1009,6 +854,11 @@ static const struct gates_data sun5i_a13_usb_gates_data __initconst = {
 	.reset_mask = 0x03,
 };
 
+static const struct gates_data sun6i_a31_usb_gates_data __initconst = {
+	.mask = { BIT(18) | BIT(17) | BIT(16) | BIT(10) | BIT(9) | BIT(8) },
+	.reset_mask = BIT(2) | BIT(1) | BIT(0),
+};
+
 static void __init sunxi_gates_clk_setup(struct device_node *node,
 					 struct gates_data *data)
 {
@@ -1304,6 +1154,7 @@ static const struct of_device_id clk_gates_match[] __initconst = {
 	{.compatible = "allwinner,sun6i-a31-apb2-gates-clk", .data = &sun6i_a31_apb2_gates_data,},
 	{.compatible = "allwinner,sun4i-a10-usb-clk", .data = &sun4i_a10_usb_gates_data,},
 	{.compatible = "allwinner,sun5i-a13-usb-clk", .data = &sun5i_a13_usb_gates_data,},
+	{.compatible = "allwinner,sun6i-a31-usb-clk", .data = &sun6i_a31_usb_gates_data,},
 	{}
 };
 
@@ -1321,33 +1172,10 @@ static void __init of_sunxi_table_clock_setup(const struct of_device_id *clk_mat
 	}
 }
 
-/**
- * System clock protection
- *
- * By enabling these critical clocks, we prevent their accidental gating
- * by the framework
- */
-static void __init sunxi_clock_protect(void)
+static void __init sunxi_init_clocks(const char *clocks[], int nclocks)
 {
-	struct clk *clk;
-
-	/* memory bus clock - sun5i+ */
-	clk = clk_get(NULL, "mbus");
-	if (!IS_ERR(clk)) {
-		clk_prepare_enable(clk);
-		clk_put(clk);
-	}
-
-	/* DDR clock - sun4i+ */
-	clk = clk_get(NULL, "pll5_ddr");
-	if (!IS_ERR(clk)) {
-		clk_prepare_enable(clk);
-		clk_put(clk);
-	}
-}
+	unsigned int i;
 
-static void __init sunxi_init_clocks(struct device_node *np)
-{
 	/* Register factor clocks */
 	of_sunxi_table_clock_setup(clk_factors_match, sunxi_factors_clk_setup);
 
@@ -1363,11 +1191,48 @@ static void __init sunxi_init_clocks(struct device_node *np)
 	/* Register gate clocks */
 	of_sunxi_table_clock_setup(clk_gates_match, sunxi_gates_clk_setup);
 
-	/* Enable core system clocks */
-	sunxi_clock_protect();
+	/* Protect the clocks that needs to stay on */
+	for (i = 0; i < nclocks; i++) {
+		struct clk *clk = clk_get(NULL, clocks[i]);
+
+		if (!IS_ERR(clk))
+			clk_prepare_enable(clk);
+	}
+}
+
+static const char *sun4i_a10_critical_clocks[] __initdata = {
+	"pll5_ddr",
+};
+
+static void __init sun4i_a10_init_clocks(struct device_node *node)
+{
+	sunxi_init_clocks(sun4i_a10_critical_clocks,
+			  ARRAY_SIZE(sun4i_a10_critical_clocks));
+}
+CLK_OF_DECLARE(sun4i_a10_clk_init, "allwinner,sun4i-a10", sun4i_a10_init_clocks);
+
+static const char *sun5i_critical_clocks[] __initdata = {
+	"mbus",
+	"pll5_ddr",
+};
+
+static void __init sun5i_init_clocks(struct device_node *node)
+{
+	sunxi_init_clocks(sun5i_critical_clocks,
+			  ARRAY_SIZE(sun5i_critical_clocks));
+}
+CLK_OF_DECLARE(sun5i_a10s_clk_init, "allwinner,sun5i-a10s", sun5i_init_clocks);
+CLK_OF_DECLARE(sun5i_a13_clk_init, "allwinner,sun5i-a13", sun5i_init_clocks);
+CLK_OF_DECLARE(sun7i_a20_clk_init, "allwinner,sun7i-a20", sun5i_init_clocks);
+
+static const char *sun6i_critical_clocks[] __initdata = {
+	"cpu",
+	"ahb1_sdram",
+};
+
+static void __init sun6i_init_clocks(struct device_node *node)
+{
+	sunxi_init_clocks(sun6i_critical_clocks,
+			  ARRAY_SIZE(sun6i_critical_clocks));
 }
-CLK_OF_DECLARE(sun4i_a10_clk_init, "allwinner,sun4i-a10", sunxi_init_clocks);
-CLK_OF_DECLARE(sun5i_a10s_clk_init, "allwinner,sun5i-a10s", sunxi_init_clocks);
-CLK_OF_DECLARE(sun5i_a13_clk_init, "allwinner,sun5i-a13", sunxi_init_clocks);
-CLK_OF_DECLARE(sun6i_a31_clk_init, "allwinner,sun6i-a31", sunxi_init_clocks);
-CLK_OF_DECLARE(sun7i_a20_clk_init, "allwinner,sun7i-a20", sunxi_init_clocks);
+CLK_OF_DECLARE(sun6i_a31_clk_init, "allwinner,sun6i-a31", sun6i_init_clocks);
diff --git a/drivers/clk/ti/Makefile b/drivers/clk/ti/Makefile
index 4319d4031aa3..ed4d0aaf8916 100644
--- a/drivers/clk/ti/Makefile
+++ b/drivers/clk/ti/Makefile
@@ -3,9 +3,11 @@ obj-y					+= clk.o autoidle.o clockdomain.o
 clk-common				= dpll.o composite.o divider.o gate.o \
 					  fixed-factor.o mux.o apll.o
 obj-$(CONFIG_SOC_AM33XX)		+= $(clk-common) clk-33xx.o
+obj-$(CONFIG_ARCH_OMAP2)		+= $(clk-common) interface.o clk-2xxx.o
 obj-$(CONFIG_ARCH_OMAP3)		+= $(clk-common) interface.o clk-3xxx.o
 obj-$(CONFIG_ARCH_OMAP4)		+= $(clk-common) clk-44xx.o
 obj-$(CONFIG_SOC_OMAP5)			+= $(clk-common) clk-54xx.o
-obj-$(CONFIG_SOC_DRA7XX)		+= $(clk-common) clk-7xx.o
+obj-$(CONFIG_SOC_DRA7XX)		+= $(clk-common) clk-7xx.o \
+					   clk-dra7-atl.o
 obj-$(CONFIG_SOC_AM43XX)		+= $(clk-common) clk-43xx.o
 endif
diff --git a/drivers/clk/ti/apll.c b/drivers/clk/ti/apll.c
index b986f61f5a77..72d97279eae1 100644
--- a/drivers/clk/ti/apll.c
+++ b/drivers/clk/ti/apll.c
@@ -77,13 +77,11 @@ static int dra7_apll_enable(struct clk_hw *hw)
 	if (i == MAX_APLL_WAIT_TRIES) {
 		pr_warn("clock: %s failed transition to '%s'\n",
 			clk_name, (state) ? "locked" : "bypassed");
-	} else {
+		r = -EBUSY;
+	} else
 		pr_debug("clock: %s transition to '%s' in %d loops\n",
 			 clk_name, (state) ? "locked" : "bypassed", i);
 
-		r = 0;
-	}
-
 	return r;
 }
 
@@ -221,3 +219,184 @@ cleanup:
 	kfree(init);
 }
 CLK_OF_DECLARE(dra7_apll_clock, "ti,dra7-apll-clock", of_dra7_apll_setup);
+
+#define OMAP2_EN_APLL_LOCKED	0x3
+#define OMAP2_EN_APLL_STOPPED	0x0
+
+static int omap2_apll_is_enabled(struct clk_hw *hw)
+{
+	struct clk_hw_omap *clk = to_clk_hw_omap(hw);
+	struct dpll_data *ad = clk->dpll_data;
+	u32 v;
+
+	v = ti_clk_ll_ops->clk_readl(ad->control_reg);
+	v &= ad->enable_mask;
+
+	v >>= __ffs(ad->enable_mask);
+
+	return v == OMAP2_EN_APLL_LOCKED ? 1 : 0;
+}
+
+static unsigned long omap2_apll_recalc(struct clk_hw *hw,
+				       unsigned long parent_rate)
+{
+	struct clk_hw_omap *clk = to_clk_hw_omap(hw);
+
+	if (omap2_apll_is_enabled(hw))
+		return clk->fixed_rate;
+
+	return 0;
+}
+
+static int omap2_apll_enable(struct clk_hw *hw)
+{
+	struct clk_hw_omap *clk = to_clk_hw_omap(hw);
+	struct dpll_data *ad = clk->dpll_data;
+	u32 v;
+	int i = 0;
+
+	v = ti_clk_ll_ops->clk_readl(ad->control_reg);
+	v &= ~ad->enable_mask;
+	v |= OMAP2_EN_APLL_LOCKED << __ffs(ad->enable_mask);
+	ti_clk_ll_ops->clk_writel(v, ad->control_reg);
+
+	while (1) {
+		v = ti_clk_ll_ops->clk_readl(ad->idlest_reg);
+		if (v & ad->idlest_mask)
+			break;
+		if (i > MAX_APLL_WAIT_TRIES)
+			break;
+		i++;
+		udelay(1);
+	}
+
+	if (i == MAX_APLL_WAIT_TRIES) {
+		pr_warn("%s failed to transition to locked\n",
+			__clk_get_name(clk->hw.clk));
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static void omap2_apll_disable(struct clk_hw *hw)
+{
+	struct clk_hw_omap *clk = to_clk_hw_omap(hw);
+	struct dpll_data *ad = clk->dpll_data;
+	u32 v;
+
+	v = ti_clk_ll_ops->clk_readl(ad->control_reg);
+	v &= ~ad->enable_mask;
+	v |= OMAP2_EN_APLL_STOPPED << __ffs(ad->enable_mask);
+	ti_clk_ll_ops->clk_writel(v, ad->control_reg);
+}
+
+static struct clk_ops omap2_apll_ops = {
+	.enable		= &omap2_apll_enable,
+	.disable	= &omap2_apll_disable,
+	.is_enabled	= &omap2_apll_is_enabled,
+	.recalc_rate	= &omap2_apll_recalc,
+};
+
+static void omap2_apll_set_autoidle(struct clk_hw_omap *clk, u32 val)
+{
+	struct dpll_data *ad = clk->dpll_data;
+	u32 v;
+
+	v = ti_clk_ll_ops->clk_readl(ad->autoidle_reg);
+	v &= ~ad->autoidle_mask;
+	v |= val << __ffs(ad->autoidle_mask);
+	ti_clk_ll_ops->clk_writel(v, ad->control_reg);
+}
+
+#define OMAP2_APLL_AUTOIDLE_LOW_POWER_STOP	0x3
+#define OMAP2_APLL_AUTOIDLE_DISABLE		0x0
+
+static void omap2_apll_allow_idle(struct clk_hw_omap *clk)
+{
+	omap2_apll_set_autoidle(clk, OMAP2_APLL_AUTOIDLE_LOW_POWER_STOP);
+}
+
+static void omap2_apll_deny_idle(struct clk_hw_omap *clk)
+{
+	omap2_apll_set_autoidle(clk, OMAP2_APLL_AUTOIDLE_DISABLE);
+}
+
+static struct clk_hw_omap_ops omap2_apll_hwops = {
+	.allow_idle	= &omap2_apll_allow_idle,
+	.deny_idle	= &omap2_apll_deny_idle,
+};
+
+static void __init of_omap2_apll_setup(struct device_node *node)
+{
+	struct dpll_data *ad = NULL;
+	struct clk_hw_omap *clk_hw = NULL;
+	struct clk_init_data *init = NULL;
+	struct clk *clk;
+	const char *parent_name;
+	u32 val;
+
+	ad = kzalloc(sizeof(*ad), GFP_KERNEL);
+	clk_hw = kzalloc(sizeof(*clk_hw), GFP_KERNEL);
+	init = kzalloc(sizeof(*init), GFP_KERNEL);
+
+	if (!ad || !clk_hw || !init)
+		goto cleanup;
+
+	clk_hw->dpll_data = ad;
+	clk_hw->hw.init = init;
+	init->ops = &omap2_apll_ops;
+	init->name = node->name;
+	clk_hw->ops = &omap2_apll_hwops;
+
+	init->num_parents = of_clk_get_parent_count(node);
+	if (init->num_parents != 1) {
+		pr_err("%s must have one parent\n", node->name);
+		goto cleanup;
+	}
+
+	parent_name = of_clk_get_parent_name(node, 0);
+	init->parent_names = &parent_name;
+
+	if (of_property_read_u32(node, "ti,clock-frequency", &val)) {
+		pr_err("%s missing clock-frequency\n", node->name);
+		goto cleanup;
+	}
+	clk_hw->fixed_rate = val;
+
+	if (of_property_read_u32(node, "ti,bit-shift", &val)) {
+		pr_err("%s missing bit-shift\n", node->name);
+		goto cleanup;
+	}
+
+	clk_hw->enable_bit = val;
+	ad->enable_mask = 0x3 << val;
+	ad->autoidle_mask = 0x3 << val;
+
+	if (of_property_read_u32(node, "ti,idlest-shift", &val)) {
+		pr_err("%s missing idlest-shift\n", node->name);
+		goto cleanup;
+	}
+
+	ad->idlest_mask = 1 << val;
+
+	ad->control_reg = ti_clk_get_reg_addr(node, 0);
+	ad->autoidle_reg = ti_clk_get_reg_addr(node, 1);
+	ad->idlest_reg = ti_clk_get_reg_addr(node, 2);
+
+	if (!ad->control_reg || !ad->autoidle_reg || !ad->idlest_reg)
+		goto cleanup;
+
+	clk = clk_register(NULL, &clk_hw->hw);
+	if (!IS_ERR(clk)) {
+		of_clk_add_provider(node, of_clk_src_simple_get, clk);
+		kfree(init);
+		return;
+	}
+cleanup:
+	kfree(ad);
+	kfree(clk_hw);
+	kfree(init);
+}
+CLK_OF_DECLARE(omap2_apll_clock, "ti,omap2-apll-clock",
+	       of_omap2_apll_setup);
diff --git a/drivers/clk/ti/clk-2xxx.c b/drivers/clk/ti/clk-2xxx.c
new file mode 100644
index 000000000000..c808ab3d2bb2
--- /dev/null
+++ b/drivers/clk/ti/clk-2xxx.c
@@ -0,0 +1,256 @@
+/*
+ * OMAP2 Clock init
+ *
+ * Copyright (C) 2013 Texas Instruments, Inc
+ *     Tero Kristo (t-kristo@ti.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/clk-provider.h>
+#include <linux/clk/ti.h>
+
+static struct ti_dt_clk omap2xxx_clks[] = {
+	DT_CLK(NULL, "func_32k_ck", "func_32k_ck"),
+	DT_CLK(NULL, "secure_32k_ck", "secure_32k_ck"),
+	DT_CLK(NULL, "virt_12m_ck", "virt_12m_ck"),
+	DT_CLK(NULL, "virt_13m_ck", "virt_13m_ck"),
+	DT_CLK(NULL, "virt_19200000_ck", "virt_19200000_ck"),
+	DT_CLK(NULL, "virt_26m_ck", "virt_26m_ck"),
+	DT_CLK(NULL, "aplls_clkin_ck", "aplls_clkin_ck"),
+	DT_CLK(NULL, "aplls_clkin_x2_ck", "aplls_clkin_x2_ck"),
+	DT_CLK(NULL, "osc_ck", "osc_ck"),
+	DT_CLK(NULL, "sys_ck", "sys_ck"),
+	DT_CLK(NULL, "alt_ck", "alt_ck"),
+	DT_CLK(NULL, "mcbsp_clks", "mcbsp_clks"),
+	DT_CLK(NULL, "dpll_ck", "dpll_ck"),
+	DT_CLK(NULL, "apll96_ck", "apll96_ck"),
+	DT_CLK(NULL, "apll54_ck", "apll54_ck"),
+	DT_CLK(NULL, "func_54m_ck", "func_54m_ck"),
+	DT_CLK(NULL, "core_ck", "core_ck"),
+	DT_CLK(NULL, "func_96m_ck", "func_96m_ck"),
+	DT_CLK(NULL, "func_48m_ck", "func_48m_ck"),
+	DT_CLK(NULL, "func_12m_ck", "func_12m_ck"),
+	DT_CLK(NULL, "sys_clkout_src", "sys_clkout_src"),
+	DT_CLK(NULL, "sys_clkout", "sys_clkout"),
+	DT_CLK(NULL, "emul_ck", "emul_ck"),
+	DT_CLK(NULL, "mpu_ck", "mpu_ck"),
+	DT_CLK(NULL, "dsp_fck", "dsp_fck"),
+	DT_CLK(NULL, "gfx_3d_fck", "gfx_3d_fck"),
+	DT_CLK(NULL, "gfx_2d_fck", "gfx_2d_fck"),
+	DT_CLK(NULL, "gfx_ick", "gfx_ick"),
+	DT_CLK("omapdss_dss", "ick", "dss_ick"),
+	DT_CLK(NULL, "dss_ick", "dss_ick"),
+	DT_CLK(NULL, "dss1_fck", "dss1_fck"),
+	DT_CLK(NULL, "dss2_fck", "dss2_fck"),
+	DT_CLK(NULL, "dss_54m_fck", "dss_54m_fck"),
+	DT_CLK(NULL, "core_l3_ck", "core_l3_ck"),
+	DT_CLK(NULL, "ssi_fck", "ssi_ssr_sst_fck"),
+	DT_CLK(NULL, "usb_l4_ick", "usb_l4_ick"),
+	DT_CLK(NULL, "l4_ck", "l4_ck"),
+	DT_CLK(NULL, "ssi_l4_ick", "ssi_l4_ick"),
+	DT_CLK(NULL, "gpt1_ick", "gpt1_ick"),
+	DT_CLK(NULL, "gpt1_fck", "gpt1_fck"),
+	DT_CLK(NULL, "gpt2_ick", "gpt2_ick"),
+	DT_CLK(NULL, "gpt2_fck", "gpt2_fck"),
+	DT_CLK(NULL, "gpt3_ick", "gpt3_ick"),
+	DT_CLK(NULL, "gpt3_fck", "gpt3_fck"),
+	DT_CLK(NULL, "gpt4_ick", "gpt4_ick"),
+	DT_CLK(NULL, "gpt4_fck", "gpt4_fck"),
+	DT_CLK(NULL, "gpt5_ick", "gpt5_ick"),
+	DT_CLK(NULL, "gpt5_fck", "gpt5_fck"),
+	DT_CLK(NULL, "gpt6_ick", "gpt6_ick"),
+	DT_CLK(NULL, "gpt6_fck", "gpt6_fck"),
+	DT_CLK(NULL, "gpt7_ick", "gpt7_ick"),
+	DT_CLK(NULL, "gpt7_fck", "gpt7_fck"),
+	DT_CLK(NULL, "gpt8_ick", "gpt8_ick"),
+	DT_CLK(NULL, "gpt8_fck", "gpt8_fck"),
+	DT_CLK(NULL, "gpt9_ick", "gpt9_ick"),
+	DT_CLK(NULL, "gpt9_fck", "gpt9_fck"),
+	DT_CLK(NULL, "gpt10_ick", "gpt10_ick"),
+	DT_CLK(NULL, "gpt10_fck", "gpt10_fck"),
+	DT_CLK(NULL, "gpt11_ick", "gpt11_ick"),
+	DT_CLK(NULL, "gpt11_fck", "gpt11_fck"),
+	DT_CLK(NULL, "gpt12_ick", "gpt12_ick"),
+	DT_CLK(NULL, "gpt12_fck", "gpt12_fck"),
+	DT_CLK("omap-mcbsp.1", "ick", "mcbsp1_ick"),
+	DT_CLK(NULL, "mcbsp1_ick", "mcbsp1_ick"),
+	DT_CLK(NULL, "mcbsp1_fck", "mcbsp1_fck"),
+	DT_CLK("omap-mcbsp.2", "ick", "mcbsp2_ick"),
+	DT_CLK(NULL, "mcbsp2_ick", "mcbsp2_ick"),
+	DT_CLK(NULL, "mcbsp2_fck", "mcbsp2_fck"),
+	DT_CLK("omap2_mcspi.1", "ick", "mcspi1_ick"),
+	DT_CLK(NULL, "mcspi1_ick", "mcspi1_ick"),
+	DT_CLK(NULL, "mcspi1_fck", "mcspi1_fck"),
+	DT_CLK("omap2_mcspi.2", "ick", "mcspi2_ick"),
+	DT_CLK(NULL, "mcspi2_ick", "mcspi2_ick"),
+	DT_CLK(NULL, "mcspi2_fck", "mcspi2_fck"),
+	DT_CLK(NULL, "uart1_ick", "uart1_ick"),
+	DT_CLK(NULL, "uart1_fck", "uart1_fck"),
+	DT_CLK(NULL, "uart2_ick", "uart2_ick"),
+	DT_CLK(NULL, "uart2_fck", "uart2_fck"),
+	DT_CLK(NULL, "uart3_ick", "uart3_ick"),
+	DT_CLK(NULL, "uart3_fck", "uart3_fck"),
+	DT_CLK(NULL, "gpios_ick", "gpios_ick"),
+	DT_CLK(NULL, "gpios_fck", "gpios_fck"),
+	DT_CLK("omap_wdt", "ick", "mpu_wdt_ick"),
+	DT_CLK(NULL, "mpu_wdt_ick", "mpu_wdt_ick"),
+	DT_CLK(NULL, "mpu_wdt_fck", "mpu_wdt_fck"),
+	DT_CLK(NULL, "sync_32k_ick", "sync_32k_ick"),
+	DT_CLK(NULL, "wdt1_ick", "wdt1_ick"),
+	DT_CLK(NULL, "omapctrl_ick", "omapctrl_ick"),
+	DT_CLK("omap24xxcam", "fck", "cam_fck"),
+	DT_CLK(NULL, "cam_fck", "cam_fck"),
+	DT_CLK("omap24xxcam", "ick", "cam_ick"),
+	DT_CLK(NULL, "cam_ick", "cam_ick"),
+	DT_CLK(NULL, "mailboxes_ick", "mailboxes_ick"),
+	DT_CLK(NULL, "wdt4_ick", "wdt4_ick"),
+	DT_CLK(NULL, "wdt4_fck", "wdt4_fck"),
+	DT_CLK(NULL, "mspro_ick", "mspro_ick"),
+	DT_CLK(NULL, "mspro_fck", "mspro_fck"),
+	DT_CLK(NULL, "fac_ick", "fac_ick"),
+	DT_CLK(NULL, "fac_fck", "fac_fck"),
+	DT_CLK("omap_hdq.0", "ick", "hdq_ick"),
+	DT_CLK(NULL, "hdq_ick", "hdq_ick"),
+	DT_CLK("omap_hdq.0", "fck", "hdq_fck"),
+	DT_CLK(NULL, "hdq_fck", "hdq_fck"),
+	DT_CLK("omap_i2c.1", "ick", "i2c1_ick"),
+	DT_CLK(NULL, "i2c1_ick", "i2c1_ick"),
+	DT_CLK("omap_i2c.2", "ick", "i2c2_ick"),
+	DT_CLK(NULL, "i2c2_ick", "i2c2_ick"),
+	DT_CLK(NULL, "gpmc_fck", "gpmc_fck"),
+	DT_CLK(NULL, "sdma_fck", "sdma_fck"),
+	DT_CLK(NULL, "sdma_ick", "sdma_ick"),
+	DT_CLK(NULL, "sdrc_ick", "sdrc_ick"),
+	DT_CLK(NULL, "des_ick", "des_ick"),
+	DT_CLK("omap-sham", "ick", "sha_ick"),
+	DT_CLK(NULL, "sha_ick", "sha_ick"),
+	DT_CLK("omap_rng", "ick", "rng_ick"),
+	DT_CLK(NULL, "rng_ick", "rng_ick"),
+	DT_CLK("omap-aes", "ick", "aes_ick"),
+	DT_CLK(NULL, "aes_ick", "aes_ick"),
+	DT_CLK(NULL, "pka_ick", "pka_ick"),
+	DT_CLK(NULL, "usb_fck", "usb_fck"),
+	DT_CLK(NULL, "timer_32k_ck", "func_32k_ck"),
+	DT_CLK(NULL, "timer_sys_ck", "sys_ck"),
+	DT_CLK(NULL, "timer_ext_ck", "alt_ck"),
+	{ .node_name = NULL },
+};
+
+static struct ti_dt_clk omap2420_clks[] = {
+	DT_CLK(NULL, "sys_clkout2_src", "sys_clkout2_src"),
+	DT_CLK(NULL, "sys_clkout2", "sys_clkout2"),
+	DT_CLK(NULL, "dsp_ick", "dsp_ick"),
+	DT_CLK(NULL, "iva1_ifck", "iva1_ifck"),
+	DT_CLK(NULL, "iva1_mpu_int_ifck", "iva1_mpu_int_ifck"),
+	DT_CLK(NULL, "wdt3_ick", "wdt3_ick"),
+	DT_CLK(NULL, "wdt3_fck", "wdt3_fck"),
+	DT_CLK("mmci-omap.0", "ick", "mmc_ick"),
+	DT_CLK(NULL, "mmc_ick", "mmc_ick"),
+	DT_CLK("mmci-omap.0", "fck", "mmc_fck"),
+	DT_CLK(NULL, "mmc_fck", "mmc_fck"),
+	DT_CLK(NULL, "eac_ick", "eac_ick"),
+	DT_CLK(NULL, "eac_fck", "eac_fck"),
+	DT_CLK(NULL, "i2c1_fck", "i2c1_fck"),
+	DT_CLK(NULL, "i2c2_fck", "i2c2_fck"),
+	DT_CLK(NULL, "vlynq_ick", "vlynq_ick"),
+	DT_CLK(NULL, "vlynq_fck", "vlynq_fck"),
+	DT_CLK("musb-hdrc", "fck", "osc_ck"),
+	{ .node_name = NULL },
+};
+
+static struct ti_dt_clk omap2430_clks[] = {
+	DT_CLK("twl", "fck", "osc_ck"),
+	DT_CLK(NULL, "iva2_1_ick", "iva2_1_ick"),
+	DT_CLK(NULL, "mdm_ick", "mdm_ick"),
+	DT_CLK(NULL, "mdm_osc_ck", "mdm_osc_ck"),
+	DT_CLK("omap-mcbsp.3", "ick", "mcbsp3_ick"),
+	DT_CLK(NULL, "mcbsp3_ick", "mcbsp3_ick"),
+	DT_CLK(NULL, "mcbsp3_fck", "mcbsp3_fck"),
+	DT_CLK("omap-mcbsp.4", "ick", "mcbsp4_ick"),
+	DT_CLK(NULL, "mcbsp4_ick", "mcbsp4_ick"),
+	DT_CLK(NULL, "mcbsp4_fck", "mcbsp4_fck"),
+	DT_CLK("omap-mcbsp.5", "ick", "mcbsp5_ick"),
+	DT_CLK(NULL, "mcbsp5_ick", "mcbsp5_ick"),
+	DT_CLK(NULL, "mcbsp5_fck", "mcbsp5_fck"),
+	DT_CLK("omap2_mcspi.3", "ick", "mcspi3_ick"),
+	DT_CLK(NULL, "mcspi3_ick", "mcspi3_ick"),
+	DT_CLK(NULL, "mcspi3_fck", "mcspi3_fck"),
+	DT_CLK(NULL, "icr_ick", "icr_ick"),
+	DT_CLK(NULL, "i2chs1_fck", "i2chs1_fck"),
+	DT_CLK(NULL, "i2chs2_fck", "i2chs2_fck"),
+	DT_CLK("musb-omap2430", "ick", "usbhs_ick"),
+	DT_CLK(NULL, "usbhs_ick", "usbhs_ick"),
+	DT_CLK("omap_hsmmc.0", "ick", "mmchs1_ick"),
+	DT_CLK(NULL, "mmchs1_ick", "mmchs1_ick"),
+	DT_CLK(NULL, "mmchs1_fck", "mmchs1_fck"),
+	DT_CLK("omap_hsmmc.1", "ick", "mmchs2_ick"),
+	DT_CLK(NULL, "mmchs2_ick", "mmchs2_ick"),
+	DT_CLK(NULL, "mmchs2_fck", "mmchs2_fck"),
+	DT_CLK(NULL, "gpio5_ick", "gpio5_ick"),
+	DT_CLK(NULL, "gpio5_fck", "gpio5_fck"),
+	DT_CLK(NULL, "mdm_intc_ick", "mdm_intc_ick"),
+	DT_CLK("omap_hsmmc.0", "mmchsdb_fck", "mmchsdb1_fck"),
+	DT_CLK(NULL, "mmchsdb1_fck", "mmchsdb1_fck"),
+	DT_CLK("omap_hsmmc.1", "mmchsdb_fck", "mmchsdb2_fck"),
+	DT_CLK(NULL, "mmchsdb2_fck", "mmchsdb2_fck"),
+	{ .node_name = NULL },
+};
+
+static const char *enable_init_clks[] = {
+	"apll96_ck",
+	"apll54_ck",
+	"sync_32k_ick",
+	"omapctrl_ick",
+	"gpmc_fck",
+	"sdrc_ick",
+};
+
+enum {
+	OMAP2_SOC_OMAP2420,
+	OMAP2_SOC_OMAP2430,
+};
+
+static int __init omap2xxx_dt_clk_init(int soc_type)
+{
+	ti_dt_clocks_register(omap2xxx_clks);
+
+	if (soc_type == OMAP2_SOC_OMAP2420)
+		ti_dt_clocks_register(omap2420_clks);
+	else
+		ti_dt_clocks_register(omap2430_clks);
+
+	omap2xxx_clkt_vps_init();
+
+	omap2_clk_disable_autoidle_all();
+
+	omap2_clk_enable_init_clocks(enable_init_clks,
+				     ARRAY_SIZE(enable_init_clks));
+
+	pr_info("Clocking rate (Crystal/DPLL/MPU): %ld.%01ld/%ld/%ld MHz\n",
+		(clk_get_rate(clk_get_sys(NULL, "sys_ck")) / 1000000),
+		(clk_get_rate(clk_get_sys(NULL, "sys_ck")) / 100000) % 10,
+		(clk_get_rate(clk_get_sys(NULL, "dpll_ck")) / 1000000),
+		(clk_get_rate(clk_get_sys(NULL, "mpu_ck")) / 1000000));
+
+	return 0;
+}
+
+int __init omap2420_dt_clk_init(void)
+{
+	return omap2xxx_dt_clk_init(OMAP2_SOC_OMAP2420);
+}
+
+int __init omap2430_dt_clk_init(void)
+{
+	return omap2xxx_dt_clk_init(OMAP2_SOC_OMAP2430);
+}
diff --git a/drivers/clk/ti/clk-43xx.c b/drivers/clk/ti/clk-43xx.c
index 527a43da3d33..3795fce8a830 100644
--- a/drivers/clk/ti/clk-43xx.c
+++ b/drivers/clk/ti/clk-43xx.c
@@ -116,9 +116,25 @@ static struct ti_dt_clk am43xx_clks[] = {
 
 int __init am43xx_dt_clk_init(void)
 {
+	struct clk *clk1, *clk2;
+
 	ti_dt_clocks_register(am43xx_clks);
 
 	omap2_clk_disable_autoidle_all();
 
+	/*
+	 * cpsw_cpts_rft_clk  has got the choice of 3 clocksources
+	 * dpll_core_m4_ck, dpll_core_m5_ck and dpll_disp_m2_ck.
+	 * By default dpll_core_m4_ck is selected, witn this as clock
+	 * source the CPTS doesnot work properly. It gives clockcheck errors
+	 * while running PTP.
+	 * clockcheck: clock jumped backward or running slower than expected!
+	 * By selecting dpll_core_m5_ck as the clocksource fixes this issue.
+	 * In AM335x dpll_core_m5_ck is the default clocksource.
+	 */
+	clk1 = clk_get_sys(NULL, "cpsw_cpts_rft_clk");
+	clk2 = clk_get_sys(NULL, "dpll_core_m5_ck");
+	clk_set_parent(clk1, clk2);
+
 	return 0;
 }
diff --git a/drivers/clk/ti/clk-54xx.c b/drivers/clk/ti/clk-54xx.c
index 08f3d1b915b3..5e183993e3ec 100644
--- a/drivers/clk/ti/clk-54xx.c
+++ b/drivers/clk/ti/clk-54xx.c
@@ -240,6 +240,12 @@ int __init omap5xxx_dt_clk_init(void)
 	if (rc)
 		pr_err("%s: failed to configure ABE DPLL!\n", __func__);
 
+	abe_dpll = clk_get_sys(NULL, "dpll_abe_m2x2_ck");
+	if (!rc)
+		rc = clk_set_rate(abe_dpll, OMAP5_DPLL_ABE_DEFFREQ * 2);
+	if (rc)
+		pr_err("%s: failed to configure ABE m2x2 DPLL!\n", __func__);
+
 	usb_dpll = clk_get_sys(NULL, "dpll_usb_ck");
 	rc = clk_set_rate(usb_dpll, OMAP5_DPLL_USB_DEFFREQ);
 	if (rc)
diff --git a/drivers/clk/ti/clk-7xx.c b/drivers/clk/ti/clk-7xx.c
index f7e40734c819..e1581335937d 100644
--- a/drivers/clk/ti/clk-7xx.c
+++ b/drivers/clk/ti/clk-7xx.c
@@ -24,7 +24,7 @@ static struct ti_dt_clk dra7xx_clks[] = {
 	DT_CLK(NULL, "atl_clkin0_ck", "atl_clkin0_ck"),
 	DT_CLK(NULL, "atl_clkin1_ck", "atl_clkin1_ck"),
 	DT_CLK(NULL, "atl_clkin2_ck", "atl_clkin2_ck"),
-	DT_CLK(NULL, "atlclkin3_ck", "atlclkin3_ck"),
+	DT_CLK(NULL, "atl_clkin3_ck", "atl_clkin3_ck"),
 	DT_CLK(NULL, "hdmi_clkin_ck", "hdmi_clkin_ck"),
 	DT_CLK(NULL, "mlb_clkin_ck", "mlb_clkin_ck"),
 	DT_CLK(NULL, "mlbp_clkin_ck", "mlbp_clkin_ck"),
diff --git a/drivers/clk/ti/clk-dra7-atl.c b/drivers/clk/ti/clk-dra7-atl.c
new file mode 100644
index 000000000000..4a65b410e4d5
--- /dev/null
+++ b/drivers/clk/ti/clk-dra7-atl.c
@@ -0,0 +1,312 @@
+/*
+ * DRA7 ATL (Audio Tracking Logic) clock driver
+ *
+ * Copyright (C) 2013 Texas Instruments, Inc.
+ *
+ * Peter Ujfalusi <peter.ujfalusi@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/module.h>
+#include <linux/clk-provider.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+
+#define DRA7_ATL_INSTANCES	4
+
+#define DRA7_ATL_PPMR_REG(id)		(0x200 + (id * 0x80))
+#define DRA7_ATL_BBSR_REG(id)		(0x204 + (id * 0x80))
+#define DRA7_ATL_ATLCR_REG(id)		(0x208 + (id * 0x80))
+#define DRA7_ATL_SWEN_REG(id)		(0x210 + (id * 0x80))
+#define DRA7_ATL_BWSMUX_REG(id)		(0x214 + (id * 0x80))
+#define DRA7_ATL_AWSMUX_REG(id)		(0x218 + (id * 0x80))
+#define DRA7_ATL_PCLKMUX_REG(id)	(0x21c + (id * 0x80))
+
+#define DRA7_ATL_SWEN			BIT(0)
+#define DRA7_ATL_DIVIDER_MASK		(0x1f)
+#define DRA7_ATL_PCLKMUX		BIT(0)
+struct dra7_atl_clock_info;
+
+struct dra7_atl_desc {
+	struct clk *clk;
+	struct clk_hw hw;
+	struct dra7_atl_clock_info *cinfo;
+	int id;
+
+	bool probed;		/* the driver for the IP has been loaded */
+	bool valid;		/* configured */
+	bool enabled;
+	u32 bws;		/* Baseband Word Select Mux */
+	u32 aws;		/* Audio Word Select Mux */
+	u32 divider;		/* Cached divider value */
+};
+
+struct dra7_atl_clock_info {
+	struct device *dev;
+	void __iomem *iobase;
+
+	struct dra7_atl_desc *cdesc;
+};
+
+#define to_atl_desc(_hw)	container_of(_hw, struct dra7_atl_desc, hw)
+
+static inline void atl_write(struct dra7_atl_clock_info *cinfo, u32 reg,
+			     u32 val)
+{
+	__raw_writel(val, cinfo->iobase + reg);
+}
+
+static inline int atl_read(struct dra7_atl_clock_info *cinfo, u32 reg)
+{
+	return __raw_readl(cinfo->iobase + reg);
+}
+
+static int atl_clk_enable(struct clk_hw *hw)
+{
+	struct dra7_atl_desc *cdesc = to_atl_desc(hw);
+
+	if (!cdesc->probed)
+		goto out;
+
+	if (unlikely(!cdesc->valid))
+		dev_warn(cdesc->cinfo->dev, "atl%d has not been configured\n",
+			 cdesc->id);
+	pm_runtime_get_sync(cdesc->cinfo->dev);
+
+	atl_write(cdesc->cinfo, DRA7_ATL_ATLCR_REG(cdesc->id),
+		  cdesc->divider - 1);
+	atl_write(cdesc->cinfo, DRA7_ATL_SWEN_REG(cdesc->id), DRA7_ATL_SWEN);
+
+out:
+	cdesc->enabled = true;
+
+	return 0;
+}
+
+static void atl_clk_disable(struct clk_hw *hw)
+{
+	struct dra7_atl_desc *cdesc = to_atl_desc(hw);
+
+	if (!cdesc->probed)
+		goto out;
+
+	atl_write(cdesc->cinfo, DRA7_ATL_SWEN_REG(cdesc->id), 0);
+	pm_runtime_put_sync(cdesc->cinfo->dev);
+
+out:
+	cdesc->enabled = false;
+}
+
+static int atl_clk_is_enabled(struct clk_hw *hw)
+{
+	struct dra7_atl_desc *cdesc = to_atl_desc(hw);
+
+	return cdesc->enabled;
+}
+
+static unsigned long atl_clk_recalc_rate(struct clk_hw *hw,
+					 unsigned long parent_rate)
+{
+	struct dra7_atl_desc *cdesc = to_atl_desc(hw);
+
+	return parent_rate / cdesc->divider;
+}
+
+static long atl_clk_round_rate(struct clk_hw *hw, unsigned long rate,
+			       unsigned long *parent_rate)
+{
+	unsigned divider;
+
+	divider = (*parent_rate + rate / 2) / rate;
+	if (divider > DRA7_ATL_DIVIDER_MASK + 1)
+		divider = DRA7_ATL_DIVIDER_MASK + 1;
+
+	return *parent_rate / divider;
+}
+
+static int atl_clk_set_rate(struct clk_hw *hw, unsigned long rate,
+			    unsigned long parent_rate)
+{
+	struct dra7_atl_desc *cdesc = to_atl_desc(hw);
+	u32 divider;
+
+	divider = ((parent_rate + rate / 2) / rate) - 1;
+	if (divider > DRA7_ATL_DIVIDER_MASK)
+		divider = DRA7_ATL_DIVIDER_MASK;
+
+	cdesc->divider = divider + 1;
+
+	return 0;
+}
+
+const struct clk_ops atl_clk_ops = {
+	.enable		= atl_clk_enable,
+	.disable	= atl_clk_disable,
+	.is_enabled	= atl_clk_is_enabled,
+	.recalc_rate	= atl_clk_recalc_rate,
+	.round_rate	= atl_clk_round_rate,
+	.set_rate	= atl_clk_set_rate,
+};
+
+static void __init of_dra7_atl_clock_setup(struct device_node *node)
+{
+	struct dra7_atl_desc *clk_hw = NULL;
+	struct clk_init_data init = { 0 };
+	const char **parent_names = NULL;
+	struct clk *clk;
+
+	clk_hw = kzalloc(sizeof(*clk_hw), GFP_KERNEL);
+	if (!clk_hw) {
+		pr_err("%s: could not allocate dra7_atl_desc\n", __func__);
+		return;
+	}
+
+	clk_hw->hw.init = &init;
+	clk_hw->divider = 1;
+	init.name = node->name;
+	init.ops = &atl_clk_ops;
+	init.flags = CLK_IGNORE_UNUSED;
+	init.num_parents = of_clk_get_parent_count(node);
+
+	if (init.num_parents != 1) {
+		pr_err("%s: atl clock %s must have 1 parent\n", __func__,
+		       node->name);
+		goto cleanup;
+	}
+
+	parent_names = kzalloc(sizeof(char *), GFP_KERNEL);
+
+	if (!parent_names)
+		goto cleanup;
+
+	parent_names[0] = of_clk_get_parent_name(node, 0);
+
+	init.parent_names = parent_names;
+
+	clk = clk_register(NULL, &clk_hw->hw);
+
+	if (!IS_ERR(clk)) {
+		of_clk_add_provider(node, of_clk_src_simple_get, clk);
+		return;
+	}
+cleanup:
+	kfree(parent_names);
+	kfree(clk_hw);
+}
+CLK_OF_DECLARE(dra7_atl_clock, "ti,dra7-atl-clock", of_dra7_atl_clock_setup);
+
+static int of_dra7_atl_clk_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct dra7_atl_clock_info *cinfo;
+	int i;
+	int ret = 0;
+
+	if (!node)
+		return -ENODEV;
+
+	cinfo = devm_kzalloc(&pdev->dev, sizeof(*cinfo), GFP_KERNEL);
+	if (!cinfo)
+		return -ENOMEM;
+
+	cinfo->iobase = of_iomap(node, 0);
+	cinfo->dev = &pdev->dev;
+	pm_runtime_enable(cinfo->dev);
+
+	pm_runtime_get_sync(cinfo->dev);
+	atl_write(cinfo, DRA7_ATL_PCLKMUX_REG(0), DRA7_ATL_PCLKMUX);
+
+	for (i = 0; i < DRA7_ATL_INSTANCES; i++) {
+		struct device_node *cfg_node;
+		char prop[5];
+		struct dra7_atl_desc *cdesc;
+		struct of_phandle_args clkspec;
+		struct clk *clk;
+		int rc;
+
+		rc = of_parse_phandle_with_args(node, "ti,provided-clocks",
+						NULL, i, &clkspec);
+
+		if (rc) {
+			pr_err("%s: failed to lookup atl clock %d\n", __func__,
+			       i);
+			return -EINVAL;
+		}
+
+		clk = of_clk_get_from_provider(&clkspec);
+
+		cdesc = to_atl_desc(__clk_get_hw(clk));
+		cdesc->cinfo = cinfo;
+		cdesc->id = i;
+
+		/* Get configuration for the ATL instances */
+		snprintf(prop, sizeof(prop), "atl%u", i);
+		cfg_node = of_find_node_by_name(node, prop);
+		if (cfg_node) {
+			ret = of_property_read_u32(cfg_node, "bws",
+						   &cdesc->bws);
+			ret |= of_property_read_u32(cfg_node, "aws",
+						    &cdesc->aws);
+			if (!ret) {
+				cdesc->valid = true;
+				atl_write(cinfo, DRA7_ATL_BWSMUX_REG(i),
+					  cdesc->bws);
+				atl_write(cinfo, DRA7_ATL_AWSMUX_REG(i),
+					  cdesc->aws);
+			}
+		}
+
+		cdesc->probed = true;
+		/*
+		 * Enable the clock if it has been asked prior to loading the
+		 * hw driver
+		 */
+		if (cdesc->enabled)
+			atl_clk_enable(__clk_get_hw(clk));
+	}
+	pm_runtime_put_sync(cinfo->dev);
+
+	return ret;
+}
+
+static int of_dra7_atl_clk_remove(struct platform_device *pdev)
+{
+	pm_runtime_disable(&pdev->dev);
+
+	return 0;
+}
+
+static struct of_device_id of_dra7_atl_clk_match_tbl[] = {
+	{ .compatible = "ti,dra7-atl", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, of_dra7_atl_clk_match_tbl);
+
+static struct platform_driver dra7_atl_clk_driver = {
+	.driver = {
+		.name = "dra7-atl",
+		.owner = THIS_MODULE,
+		.of_match_table = of_dra7_atl_clk_match_tbl,
+	},
+	.probe = of_dra7_atl_clk_probe,
+	.remove = of_dra7_atl_clk_remove,
+};
+
+module_platform_driver(dra7_atl_clk_driver);
+
+MODULE_DESCRIPTION("Clock driver for DRA7 Audio Tracking Logic");
+MODULE_ALIAS("platform:dra7-atl-clock");
+MODULE_AUTHOR("Peter Ujfalusi <peter.ujfalusi@ti.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/clk/ti/dpll.c b/drivers/clk/ti/dpll.c
index 7e498a44f97d..79791e1bf282 100644
--- a/drivers/clk/ti/dpll.c
+++ b/drivers/clk/ti/dpll.c
@@ -25,8 +25,6 @@
 #undef pr_fmt
 #define pr_fmt(fmt) "%s: " fmt, __func__
 
-#define DPLL_HAS_AUTOIDLE	0x1
-
 #if defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_SOC_OMAP5) || \
 	defined(CONFIG_SOC_DRA7XX)
 static const struct clk_ops dpll_m4xen_ck_ops = {
@@ -37,21 +35,18 @@ static const struct clk_ops dpll_m4xen_ck_ops = {
 	.set_rate	= &omap3_noncore_dpll_set_rate,
 	.get_parent	= &omap2_init_dpll_parent,
 };
+#else
+static const struct clk_ops dpll_m4xen_ck_ops = {};
 #endif
 
+#if defined(CONFIG_ARCH_OMAP3) || defined(CONFIG_ARCH_OMAP4) || \
+	defined(CONFIG_SOC_OMAP5) || defined(CONFIG_SOC_DRA7XX) || \
+	defined(CONFIG_SOC_AM33XX) || defined(CONFIG_SOC_AM43XX)
 static const struct clk_ops dpll_core_ck_ops = {
 	.recalc_rate	= &omap3_dpll_recalc,
 	.get_parent	= &omap2_init_dpll_parent,
 };
 
-#ifdef CONFIG_ARCH_OMAP3
-static const struct clk_ops omap3_dpll_core_ck_ops = {
-	.get_parent	= &omap2_init_dpll_parent,
-	.recalc_rate	= &omap3_dpll_recalc,
-	.round_rate	= &omap2_dpll_round_rate,
-};
-#endif
-
 static const struct clk_ops dpll_ck_ops = {
 	.enable		= &omap3_noncore_dpll_enable,
 	.disable	= &omap3_noncore_dpll_disable,
@@ -67,6 +62,33 @@ static const struct clk_ops dpll_no_gate_ck_ops = {
 	.round_rate	= &omap2_dpll_round_rate,
 	.set_rate	= &omap3_noncore_dpll_set_rate,
 };
+#else
+static const struct clk_ops dpll_core_ck_ops = {};
+static const struct clk_ops dpll_ck_ops = {};
+static const struct clk_ops dpll_no_gate_ck_ops = {};
+const struct clk_hw_omap_ops clkhwops_omap3_dpll = {};
+#endif
+
+#ifdef CONFIG_ARCH_OMAP2
+static const struct clk_ops omap2_dpll_core_ck_ops = {
+	.get_parent	= &omap2_init_dpll_parent,
+	.recalc_rate	= &omap2_dpllcore_recalc,
+	.round_rate	= &omap2_dpll_round_rate,
+	.set_rate	= &omap2_reprogram_dpllcore,
+};
+#else
+static const struct clk_ops omap2_dpll_core_ck_ops = {};
+#endif
+
+#ifdef CONFIG_ARCH_OMAP3
+static const struct clk_ops omap3_dpll_core_ck_ops = {
+	.get_parent	= &omap2_init_dpll_parent,
+	.recalc_rate	= &omap3_dpll_recalc,
+	.round_rate	= &omap2_dpll_round_rate,
+};
+#else
+static const struct clk_ops omap3_dpll_core_ck_ops = {};
+#endif
 
 #ifdef CONFIG_ARCH_OMAP3
 static const struct clk_ops omap3_dpll_ck_ops = {
@@ -139,7 +161,8 @@ cleanup:
 }
 
 #if defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_SOC_OMAP5) || \
-	defined(CONFIG_SOC_DRA7XX) || defined(CONFIG_SOC_AM33XX)
+	defined(CONFIG_SOC_DRA7XX) || defined(CONFIG_SOC_AM33XX) || \
+	defined(CONFIG_SOC_AM43XX)
 /**
  * ti_clk_register_dpll_x2 - Registers a DPLLx2 clock
  * @node: device node for this clock
@@ -193,14 +216,12 @@ static void ti_clk_register_dpll_x2(struct device_node *node,
  * @node: device node containing the DPLL info
  * @ops: ops for the DPLL
  * @ddt: DPLL data template to use
- * @init_flags: flags for controlling init types
  *
  * Initializes a DPLL clock from device tree data.
  */
 static void __init of_ti_dpll_setup(struct device_node *node,
 				    const struct clk_ops *ops,
-				    const struct dpll_data *ddt,
-				    u8 init_flags)
+				    const struct dpll_data *ddt)
 {
 	struct clk_hw_omap *clk_hw = NULL;
 	struct clk_init_data *init = NULL;
@@ -241,13 +262,30 @@ static void __init of_ti_dpll_setup(struct device_node *node,
 	init->parent_names = parent_names;
 
 	dd->control_reg = ti_clk_get_reg_addr(node, 0);
-	dd->idlest_reg = ti_clk_get_reg_addr(node, 1);
-	dd->mult_div1_reg = ti_clk_get_reg_addr(node, 2);
 
-	if (!dd->control_reg || !dd->idlest_reg || !dd->mult_div1_reg)
+	/*
+	 * Special case for OMAP2 DPLL, register order is different due to
+	 * missing idlest_reg, also clkhwops is different. Detected from
+	 * missing idlest_mask.
+	 */
+	if (!dd->idlest_mask) {
+		dd->mult_div1_reg = ti_clk_get_reg_addr(node, 1);
+#ifdef CONFIG_ARCH_OMAP2
+		clk_hw->ops = &clkhwops_omap2xxx_dpll;
+		omap2xxx_clkt_dpllcore_init(&clk_hw->hw);
+#endif
+	} else {
+		dd->idlest_reg = ti_clk_get_reg_addr(node, 1);
+		if (!dd->idlest_reg)
+			goto cleanup;
+
+		dd->mult_div1_reg = ti_clk_get_reg_addr(node, 2);
+	}
+
+	if (!dd->control_reg || !dd->mult_div1_reg)
 		goto cleanup;
 
-	if (init_flags & DPLL_HAS_AUTOIDLE) {
+	if (dd->autoidle_mask) {
 		dd->autoidle_reg = ti_clk_get_reg_addr(node, 3);
 		if (!dd->autoidle_reg)
 			goto cleanup;
@@ -285,7 +323,7 @@ CLK_OF_DECLARE(ti_omap4_dpll_x2_clock, "ti,omap4-dpll-x2-clock",
 	       of_ti_omap4_dpll_x2_setup);
 #endif
 
-#ifdef CONFIG_SOC_AM33XX
+#if defined(CONFIG_SOC_AM33XX) || defined(CONFIG_SOC_AM43XX)
 static void __init of_ti_am3_dpll_x2_setup(struct device_node *node)
 {
 	ti_clk_register_dpll_x2(node, &dpll_x2_ck_ops, NULL);
@@ -310,7 +348,7 @@ static void __init of_ti_omap3_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &omap3_dpll_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &omap3_dpll_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap3_dpll_clock, "ti,omap3-dpll-clock",
 	       of_ti_omap3_dpll_setup);
@@ -329,7 +367,7 @@ static void __init of_ti_omap3_core_dpll_setup(struct device_node *node)
 		.freqsel_mask = 0xf0,
 	};
 
-	of_ti_dpll_setup(node, &omap3_dpll_core_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &omap3_dpll_core_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap3_core_dpll_clock, "ti,omap3-dpll-core-clock",
 	       of_ti_omap3_core_dpll_setup);
@@ -349,7 +387,7 @@ static void __init of_ti_omap3_per_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_STOP) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &omap3_dpll_per_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &omap3_dpll_per_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap3_per_dpll_clock, "ti,omap3-dpll-per-clock",
 	       of_ti_omap3_per_dpll_setup);
@@ -371,7 +409,7 @@ static void __init of_ti_omap3_per_jtype_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_STOP) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &omap3_dpll_per_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &omap3_dpll_per_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap3_per_jtype_dpll_clock, "ti,omap3-dpll-per-j-type-clock",
 	       of_ti_omap3_per_jtype_dpll_setup);
@@ -391,11 +429,32 @@ static void __init of_ti_omap4_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &dpll_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap4_dpll_clock, "ti,omap4-dpll-clock",
 	       of_ti_omap4_dpll_setup);
 
+static void __init of_ti_omap5_mpu_dpll_setup(struct device_node *node)
+{
+	const struct dpll_data dd = {
+		.idlest_mask = 0x1,
+		.enable_mask = 0x7,
+		.autoidle_mask = 0x7,
+		.mult_mask = 0x7ff << 8,
+		.div1_mask = 0x7f,
+		.max_multiplier = 2047,
+		.max_divider = 128,
+		.dcc_mask = BIT(22),
+		.dcc_rate = 1400000000, /* DCC beyond 1.4GHz */
+		.min_divider = 1,
+		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
+	};
+
+	of_ti_dpll_setup(node, &dpll_ck_ops, &dd);
+}
+CLK_OF_DECLARE(of_ti_omap5_mpu_dpll_clock, "ti,omap5-mpu-dpll-clock",
+	       of_ti_omap5_mpu_dpll_setup);
+
 static void __init of_ti_omap4_core_dpll_setup(struct device_node *node)
 {
 	const struct dpll_data dd = {
@@ -410,7 +469,7 @@ static void __init of_ti_omap4_core_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_core_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &dpll_core_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap4_core_dpll_clock, "ti,omap4-dpll-core-clock",
 	       of_ti_omap4_core_dpll_setup);
@@ -433,7 +492,7 @@ static void __init of_ti_omap4_m4xen_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_m4xen_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &dpll_m4xen_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap4_m4xen_dpll_clock, "ti,omap4-dpll-m4xen-clock",
 	       of_ti_omap4_m4xen_dpll_setup);
@@ -454,7 +513,7 @@ static void __init of_ti_omap4_jtype_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_m4xen_ck_ops, &dd, DPLL_HAS_AUTOIDLE);
+	of_ti_dpll_setup(node, &dpll_m4xen_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_omap4_jtype_dpll_clock, "ti,omap4-dpll-j-type-clock",
 	       of_ti_omap4_jtype_dpll_setup);
@@ -465,7 +524,6 @@ static void __init of_ti_am3_no_gate_dpll_setup(struct device_node *node)
 	const struct dpll_data dd = {
 		.idlest_mask = 0x1,
 		.enable_mask = 0x7,
-		.autoidle_mask = 0x7,
 		.mult_mask = 0x7ff << 8,
 		.div1_mask = 0x7f,
 		.max_multiplier = 2047,
@@ -474,7 +532,7 @@ static void __init of_ti_am3_no_gate_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_no_gate_ck_ops, &dd, 0);
+	of_ti_dpll_setup(node, &dpll_no_gate_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_am3_no_gate_dpll_clock, "ti,am3-dpll-no-gate-clock",
 	       of_ti_am3_no_gate_dpll_setup);
@@ -484,7 +542,6 @@ static void __init of_ti_am3_jtype_dpll_setup(struct device_node *node)
 	const struct dpll_data dd = {
 		.idlest_mask = 0x1,
 		.enable_mask = 0x7,
-		.autoidle_mask = 0x7,
 		.mult_mask = 0x7ff << 8,
 		.div1_mask = 0x7f,
 		.max_multiplier = 4095,
@@ -494,7 +551,7 @@ static void __init of_ti_am3_jtype_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_ck_ops, &dd, 0);
+	of_ti_dpll_setup(node, &dpll_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_am3_jtype_dpll_clock, "ti,am3-dpll-j-type-clock",
 	       of_ti_am3_jtype_dpll_setup);
@@ -504,7 +561,6 @@ static void __init of_ti_am3_no_gate_jtype_dpll_setup(struct device_node *node)
 	const struct dpll_data dd = {
 		.idlest_mask = 0x1,
 		.enable_mask = 0x7,
-		.autoidle_mask = 0x7,
 		.mult_mask = 0x7ff << 8,
 		.div1_mask = 0x7f,
 		.max_multiplier = 2047,
@@ -514,7 +570,7 @@ static void __init of_ti_am3_no_gate_jtype_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_no_gate_ck_ops, &dd, 0);
+	of_ti_dpll_setup(node, &dpll_no_gate_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_am3_no_gate_jtype_dpll_clock,
 	       "ti,am3-dpll-no-gate-j-type-clock",
@@ -525,7 +581,6 @@ static void __init of_ti_am3_dpll_setup(struct device_node *node)
 	const struct dpll_data dd = {
 		.idlest_mask = 0x1,
 		.enable_mask = 0x7,
-		.autoidle_mask = 0x7,
 		.mult_mask = 0x7ff << 8,
 		.div1_mask = 0x7f,
 		.max_multiplier = 2047,
@@ -534,7 +589,7 @@ static void __init of_ti_am3_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_ck_ops, &dd, 0);
+	of_ti_dpll_setup(node, &dpll_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_am3_dpll_clock, "ti,am3-dpll-clock", of_ti_am3_dpll_setup);
 
@@ -543,7 +598,6 @@ static void __init of_ti_am3_core_dpll_setup(struct device_node *node)
 	const struct dpll_data dd = {
 		.idlest_mask = 0x1,
 		.enable_mask = 0x7,
-		.autoidle_mask = 0x7,
 		.mult_mask = 0x7ff << 8,
 		.div1_mask = 0x7f,
 		.max_multiplier = 2047,
@@ -552,7 +606,22 @@ static void __init of_ti_am3_core_dpll_setup(struct device_node *node)
 		.modes = (1 << DPLL_LOW_POWER_BYPASS) | (1 << DPLL_LOCKED),
 	};
 
-	of_ti_dpll_setup(node, &dpll_core_ck_ops, &dd, 0);
+	of_ti_dpll_setup(node, &dpll_core_ck_ops, &dd);
 }
 CLK_OF_DECLARE(ti_am3_core_dpll_clock, "ti,am3-dpll-core-clock",
 	       of_ti_am3_core_dpll_setup);
+
+static void __init of_ti_omap2_core_dpll_setup(struct device_node *node)
+{
+	const struct dpll_data dd = {
+		.enable_mask = 0x3,
+		.mult_mask = 0x3ff << 12,
+		.div1_mask = 0xf << 8,
+		.max_divider = 16,
+		.min_divider = 1,
+	};
+
+	of_ti_dpll_setup(node, &omap2_dpll_core_ck_ops, &dd);
+}
+CLK_OF_DECLARE(ti_omap2_core_dpll_clock, "ti,omap2-dpll-core-clock",
+	       of_ti_omap2_core_dpll_setup);
diff --git a/drivers/clk/ti/gate.c b/drivers/clk/ti/gate.c
index 58734817d502..b326d2797feb 100644
--- a/drivers/clk/ti/gate.c
+++ b/drivers/clk/ti/gate.c
@@ -185,7 +185,7 @@ of_ti_composite_no_wait_gate_clk_setup(struct device_node *node)
 CLK_OF_DECLARE(ti_composite_no_wait_gate_clk, "ti,composite-no-wait-gate-clock",
 	       of_ti_composite_no_wait_gate_clk_setup);
 
-#ifdef CONFIG_ARCH_OMAP3
+#if defined(CONFIG_ARCH_OMAP2) || defined(CONFIG_ARCH_OMAP3)
 static void __init of_ti_composite_interface_clk_setup(struct device_node *node)
 {
 	_of_ti_composite_gate_clk_setup(node, &clkhwops_iclk_wait);
diff --git a/drivers/clk/ti/interface.c b/drivers/clk/ti/interface.c
index 320a2b168bb2..9c3e8c4aaa40 100644
--- a/drivers/clk/ti/interface.c
+++ b/drivers/clk/ti/interface.c
@@ -94,6 +94,7 @@ static void __init of_ti_no_wait_interface_clk_setup(struct device_node *node)
 CLK_OF_DECLARE(ti_no_wait_interface_clk, "ti,omap3-no-wait-interface-clock",
 	       of_ti_no_wait_interface_clk_setup);
 
+#ifdef CONFIG_ARCH_OMAP3
 static void __init of_ti_hsotgusb_interface_clk_setup(struct device_node *node)
 {
 	_of_ti_interface_clk_setup(node,
@@ -123,3 +124,13 @@ static void __init of_ti_am35xx_interface_clk_setup(struct device_node *node)
 }
 CLK_OF_DECLARE(ti_am35xx_interface_clk, "ti,am35xx-interface-clock",
 	       of_ti_am35xx_interface_clk_setup);
+#endif
+
+#ifdef CONFIG_SOC_OMAP2430
+static void __init of_ti_omap2430_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node, &clkhwops_omap2430_i2chs_wait);
+}
+CLK_OF_DECLARE(ti_omap2430_interface_clk, "ti,omap2430-interface-clock",
+	       of_ti_omap2430_interface_clk_setup);
+#endif
diff --git a/drivers/clk/ti/mux.c b/drivers/clk/ti/mux.c
index 0197a478720c..e9d650e51287 100644
--- a/drivers/clk/ti/mux.c
+++ b/drivers/clk/ti/mux.c
@@ -160,7 +160,7 @@ static void of_mux_clk_setup(struct device_node *node)
 	u8 clk_mux_flags = 0;
 	u32 mask = 0;
 	u32 shift = 0;
-	u32 flags = 0;
+	u32 flags = CLK_SET_RATE_NO_REPARENT;
 
 	num_parents = of_clk_get_parent_count(node);
 	if (num_parents < 2) {
diff --git a/drivers/clocksource/exynos_mct.c b/drivers/clocksource/exynos_mct.c
index 8d6420013a04..ab51bf20a3ed 100644
--- a/drivers/clocksource/exynos_mct.c
+++ b/drivers/clocksource/exynos_mct.c
@@ -153,19 +153,16 @@ static void exynos4_mct_write(unsigned int value, unsigned long offset)
 }
 
 /* Clocksource handling */
-static void exynos4_mct_frc_start(u32 hi, u32 lo)
+static void exynos4_mct_frc_start(void)
 {
 	u32 reg;
 
-	exynos4_mct_write(lo, EXYNOS4_MCT_G_CNT_L);
-	exynos4_mct_write(hi, EXYNOS4_MCT_G_CNT_U);
-
 	reg = __raw_readl(reg_base + EXYNOS4_MCT_G_TCON);
 	reg |= MCT_G_TCON_START;
 	exynos4_mct_write(reg, EXYNOS4_MCT_G_TCON);
 }
 
-static cycle_t exynos4_frc_read(struct clocksource *cs)
+static cycle_t notrace _exynos4_frc_read(void)
 {
 	unsigned int lo, hi;
 	u32 hi2 = __raw_readl(reg_base + EXYNOS4_MCT_G_CNT_U);
@@ -179,9 +176,14 @@ static cycle_t exynos4_frc_read(struct clocksource *cs)
 	return ((cycle_t)hi << 32) | lo;
 }
 
+static cycle_t exynos4_frc_read(struct clocksource *cs)
+{
+	return _exynos4_frc_read();
+}
+
 static void exynos4_frc_resume(struct clocksource *cs)
 {
-	exynos4_mct_frc_start(0, 0);
+	exynos4_mct_frc_start();
 }
 
 struct clocksource mct_frc = {
@@ -195,12 +197,23 @@ struct clocksource mct_frc = {
 
 static u64 notrace exynos4_read_sched_clock(void)
 {
-	return exynos4_frc_read(&mct_frc);
+	return _exynos4_frc_read();
+}
+
+static struct delay_timer exynos4_delay_timer;
+
+static cycles_t exynos4_read_current_timer(void)
+{
+	return _exynos4_frc_read();
 }
 
 static void __init exynos4_clocksource_init(void)
 {
-	exynos4_mct_frc_start(0, 0);
+	exynos4_mct_frc_start();
+
+	exynos4_delay_timer.read_current_timer = &exynos4_read_current_timer;
+	exynos4_delay_timer.freq = clk_rate;
+	register_current_timer_delay(&exynos4_delay_timer);
 
 	if (clocksource_register_hz(&mct_frc, clk_rate))
 		panic("%s: can't register clocksource\n", mct_frc.name);
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index 1fbe11f2a146..ffe350f86bca 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -185,7 +185,9 @@ config CPU_FREQ_GOV_CONSERVATIVE
 
 config GENERIC_CPUFREQ_CPU0
 	tristate "Generic CPU0 cpufreq driver"
-	depends on HAVE_CLK && REGULATOR && OF && THERMAL && CPU_THERMAL
+	depends on HAVE_CLK && OF
+	# if CPU_THERMAL is on and THERMAL=m, CPU0 cannot be =y:
+	depends on !CPU_THERMAL || THERMAL
 	select PM_OPP
 	help
 	  This adds a generic cpufreq driver for CPU0 frequency management.
diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
index 36d20d0fce27..ebac67115009 100644
--- a/drivers/cpufreq/Kconfig.arm
+++ b/drivers/cpufreq/Kconfig.arm
@@ -5,8 +5,7 @@
 # big LITTLE core layer and glue drivers
 config ARM_BIG_LITTLE_CPUFREQ
 	tristate "Generic ARM big LITTLE CPUfreq driver"
-	depends on (BIG_LITTLE && ARM_CPU_TOPOLOGY) || (ARM64 && SMP)
-	depends on HAVE_CLK
+	depends on ARM && BIG_LITTLE && ARM_CPU_TOPOLOGY && HAVE_CLK
 	select PM_OPP
 	help
 	  This enables the Generic CPUfreq driver for ARM big.LITTLE platforms.
diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile
index 738c8b7b17dc..db6d9a2fea4d 100644
--- a/drivers/cpufreq/Makefile
+++ b/drivers/cpufreq/Makefile
@@ -49,7 +49,7 @@ obj-$(CONFIG_ARM_BIG_LITTLE_CPUFREQ)	+= arm_big_little.o
 # LITTLE drivers, so that it is probed last.
 obj-$(CONFIG_ARM_DT_BL_CPUFREQ)		+= arm_big_little_dt.o
 
-obj-$(CONFIG_ARCH_DAVINCI_DA850)	+= davinci-cpufreq.o
+obj-$(CONFIG_ARCH_DAVINCI)		+= davinci-cpufreq.o
 obj-$(CONFIG_UX500_SOC_DB8500)		+= dbx500-cpufreq.o
 obj-$(CONFIG_ARM_EXYNOS_CPUFREQ)	+= exynos-cpufreq.o
 obj-$(CONFIG_ARM_EXYNOS4210_CPUFREQ)	+= exynos4210-cpufreq.o
diff --git a/drivers/cpufreq/cpufreq-cpu0.c b/drivers/cpufreq/cpufreq-cpu0.c
index 09b9129c7bd3..ee1ae303a07c 100644
--- a/drivers/cpufreq/cpufreq-cpu0.c
+++ b/drivers/cpufreq/cpufreq-cpu0.c
@@ -104,7 +104,7 @@ static int cpu0_cpufreq_init(struct cpufreq_policy *policy)
 }
 
 static struct cpufreq_driver cpu0_cpufreq_driver = {
-	.flags = CPUFREQ_STICKY,
+	.flags = CPUFREQ_STICKY | CPUFREQ_NEED_INITIAL_FREQ_CHECK,
 	.verify = cpufreq_generic_frequency_table_verify,
 	.target_index = cpu0_set_target,
 	.get = cpufreq_generic_get,
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index ae11dd51f81d..62259d27f03e 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1816,20 +1816,55 @@ EXPORT_SYMBOL(cpufreq_unregister_notifier);
  *                              GOVERNORS                            *
  *********************************************************************/
 
+/* Must set freqs->new to intermediate frequency */
+static int __target_intermediate(struct cpufreq_policy *policy,
+				 struct cpufreq_freqs *freqs, int index)
+{
+	int ret;
+
+	freqs->new = cpufreq_driver->get_intermediate(policy, index);
+
+	/* We don't need to switch to intermediate freq */
+	if (!freqs->new)
+		return 0;
+
+	pr_debug("%s: cpu: %d, switching to intermediate freq: oldfreq: %u, intermediate freq: %u\n",
+		 __func__, policy->cpu, freqs->old, freqs->new);
+
+	cpufreq_freq_transition_begin(policy, freqs);
+	ret = cpufreq_driver->target_intermediate(policy, index);
+	cpufreq_freq_transition_end(policy, freqs, ret);
+
+	if (ret)
+		pr_err("%s: Failed to change to intermediate frequency: %d\n",
+		       __func__, ret);
+
+	return ret;
+}
+
 static int __target_index(struct cpufreq_policy *policy,
 			  struct cpufreq_frequency_table *freq_table, int index)
 {
-	struct cpufreq_freqs freqs;
+	struct cpufreq_freqs freqs = {.old = policy->cur, .flags = 0};
+	unsigned int intermediate_freq = 0;
 	int retval = -EINVAL;
 	bool notify;
 
 	notify = !(cpufreq_driver->flags & CPUFREQ_ASYNC_NOTIFICATION);
-
 	if (notify) {
-		freqs.old = policy->cur;
-		freqs.new = freq_table[index].frequency;
-		freqs.flags = 0;
+		/* Handle switching to intermediate frequency */
+		if (cpufreq_driver->get_intermediate) {
+			retval = __target_intermediate(policy, &freqs, index);
+			if (retval)
+				return retval;
+
+			intermediate_freq = freqs.new;
+			/* Set old freq to intermediate */
+			if (intermediate_freq)
+				freqs.old = freqs.new;
+		}
 
+		freqs.new = freq_table[index].frequency;
 		pr_debug("%s: cpu: %d, oldfreq: %u, new freq: %u\n",
 			 __func__, policy->cpu, freqs.old, freqs.new);
 
@@ -1841,9 +1876,23 @@ static int __target_index(struct cpufreq_policy *policy,
 		pr_err("%s: Failed to change cpu frequency: %d\n", __func__,
 		       retval);
 
-	if (notify)
+	if (notify) {
 		cpufreq_freq_transition_end(policy, &freqs, retval);
 
+		/*
+		 * Failed after setting to intermediate freq? Driver should have
+		 * reverted back to initial frequency and so should we. Check
+		 * here for intermediate_freq instead of get_intermediate, in
+		 * case we have't switched to intermediate freq at all.
+		 */
+		if (unlikely(retval && intermediate_freq)) {
+			freqs.old = intermediate_freq;
+			freqs.new = policy->restore_freq;
+			cpufreq_freq_transition_begin(policy, &freqs);
+			cpufreq_freq_transition_end(policy, &freqs, 0);
+		}
+	}
+
 	return retval;
 }
 
@@ -1875,6 +1924,9 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
 	if (target_freq == policy->cur)
 		return 0;
 
+	/* Save last value to restore later on errors */
+	policy->restore_freq = policy->cur;
+
 	if (cpufreq_driver->target)
 		retval = cpufreq_driver->target(policy, target_freq, relation);
 	else if (cpufreq_driver->target_index) {
@@ -2190,10 +2242,8 @@ int cpufreq_update_policy(unsigned int cpu)
 	struct cpufreq_policy new_policy;
 	int ret;
 
-	if (!policy) {
-		ret = -ENODEV;
-		goto no_policy;
-	}
+	if (!policy)
+		return -ENODEV;
 
 	down_write(&policy->rwsem);
 
@@ -2212,7 +2262,7 @@ int cpufreq_update_policy(unsigned int cpu)
 		new_policy.cur = cpufreq_driver->get(cpu);
 		if (WARN_ON(!new_policy.cur)) {
 			ret = -EIO;
-			goto no_policy;
+			goto unlock;
 		}
 
 		if (!policy->cur) {
@@ -2227,10 +2277,10 @@ int cpufreq_update_policy(unsigned int cpu)
 
 	ret = cpufreq_set_policy(policy, &new_policy);
 
+unlock:
 	up_write(&policy->rwsem);
 
 	cpufreq_cpu_put(policy);
-no_policy:
 	return ret;
 }
 EXPORT_SYMBOL(cpufreq_update_policy);
@@ -2361,7 +2411,8 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
 	    !(driver_data->setpolicy || driver_data->target_index ||
 		    driver_data->target) ||
 	     (driver_data->setpolicy && (driver_data->target_index ||
-		    driver_data->target)))
+		    driver_data->target)) ||
+	     (!!driver_data->get_intermediate != !!driver_data->target_intermediate))
 		return -EINVAL;
 
 	pr_debug("trying to register driver %s\n", driver_data->name);
diff --git a/drivers/cpufreq/cpufreq_governor.c b/drivers/cpufreq/cpufreq_governor.c
index e1c6433b16e0..1b44496b2d2b 100644
--- a/drivers/cpufreq/cpufreq_governor.c
+++ b/drivers/cpufreq/cpufreq_governor.c
@@ -36,14 +36,29 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 	struct od_dbs_tuners *od_tuners = dbs_data->tuners;
 	struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
 	struct cpufreq_policy *policy;
+	unsigned int sampling_rate;
 	unsigned int max_load = 0;
 	unsigned int ignore_nice;
 	unsigned int j;
 
-	if (dbs_data->cdata->governor == GOV_ONDEMAND)
+	if (dbs_data->cdata->governor == GOV_ONDEMAND) {
+		struct od_cpu_dbs_info_s *od_dbs_info =
+				dbs_data->cdata->get_cpu_dbs_info_s(cpu);
+
+		/*
+		 * Sometimes, the ondemand governor uses an additional
+		 * multiplier to give long delays. So apply this multiplier to
+		 * the 'sampling_rate', so as to keep the wake-up-from-idle
+		 * detection logic a bit conservative.
+		 */
+		sampling_rate = od_tuners->sampling_rate;
+		sampling_rate *= od_dbs_info->rate_mult;
+
 		ignore_nice = od_tuners->ignore_nice_load;
-	else
+	} else {
+		sampling_rate = cs_tuners->sampling_rate;
 		ignore_nice = cs_tuners->ignore_nice_load;
+	}
 
 	policy = cdbs->cur_policy;
 
@@ -96,7 +111,46 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 		if (unlikely(!wall_time || wall_time < idle_time))
 			continue;
 
-		load = 100 * (wall_time - idle_time) / wall_time;
+		/*
+		 * If the CPU had gone completely idle, and a task just woke up
+		 * on this CPU now, it would be unfair to calculate 'load' the
+		 * usual way for this elapsed time-window, because it will show
+		 * near-zero load, irrespective of how CPU intensive that task
+		 * actually is. This is undesirable for latency-sensitive bursty
+		 * workloads.
+		 *
+		 * To avoid this, we reuse the 'load' from the previous
+		 * time-window and give this task a chance to start with a
+		 * reasonably high CPU frequency. (However, we shouldn't over-do
+		 * this copy, lest we get stuck at a high load (high frequency)
+		 * for too long, even when the current system load has actually
+		 * dropped down. So we perform the copy only once, upon the
+		 * first wake-up from idle.)
+		 *
+		 * Detecting this situation is easy: the governor's deferrable
+		 * timer would not have fired during CPU-idle periods. Hence
+		 * an unusually large 'wall_time' (as compared to the sampling
+		 * rate) indicates this scenario.
+		 *
+		 * prev_load can be zero in two cases and we must recalculate it
+		 * for both cases:
+		 * - during long idle intervals
+		 * - explicitly set to zero
+		 */
+		if (unlikely(wall_time > (2 * sampling_rate) &&
+			     j_cdbs->prev_load)) {
+			load = j_cdbs->prev_load;
+
+			/*
+			 * Perform a destructive copy, to ensure that we copy
+			 * the previous load only once, upon the first wake-up
+			 * from idle.
+			 */
+			j_cdbs->prev_load = 0;
+		} else {
+			load = 100 * (wall_time - idle_time) / wall_time;
+			j_cdbs->prev_load = load;
+		}
 
 		if (load > max_load)
 			max_load = load;
@@ -318,11 +372,18 @@ int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 		for_each_cpu(j, policy->cpus) {
 			struct cpu_dbs_common_info *j_cdbs =
 				dbs_data->cdata->get_cpu_cdbs(j);
+			unsigned int prev_load;
 
 			j_cdbs->cpu = j;
 			j_cdbs->cur_policy = policy;
 			j_cdbs->prev_cpu_idle = get_cpu_idle_time(j,
 					       &j_cdbs->prev_cpu_wall, io_busy);
+
+			prev_load = (unsigned int)
+				(j_cdbs->prev_cpu_wall - j_cdbs->prev_cpu_idle);
+			j_cdbs->prev_load = 100 * prev_load /
+					(unsigned int) j_cdbs->prev_cpu_wall;
+
 			if (ignore_nice)
 				j_cdbs->prev_cpu_nice =
 					kcpustat_cpu(j).cpustat[CPUTIME_NICE];
diff --git a/drivers/cpufreq/cpufreq_governor.h b/drivers/cpufreq/cpufreq_governor.h
index bfb9ae14142c..cc401d147e72 100644
--- a/drivers/cpufreq/cpufreq_governor.h
+++ b/drivers/cpufreq/cpufreq_governor.h
@@ -134,6 +134,13 @@ struct cpu_dbs_common_info {
 	u64 prev_cpu_idle;
 	u64 prev_cpu_wall;
 	u64 prev_cpu_nice;
+	/*
+	 * Used to keep track of load in the previous interval. However, when
+	 * explicitly set to zero, it is used as a flag to ensure that we copy
+	 * the previous load to the current interval only once, upon the first
+	 * wake-up from idle.
+	 */
+	unsigned int prev_load;
 	struct cpufreq_policy *cur_policy;
 	struct delayed_work work;
 	/*
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index aebd4572eb6d..86631cb6f7de 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -128,6 +128,7 @@ static struct pstate_funcs pstate_funcs;
 
 struct perf_limits {
 	int no_turbo;
+	int turbo_disabled;
 	int max_perf_pct;
 	int min_perf_pct;
 	int32_t max_perf;
@@ -196,10 +197,7 @@ static signed int pid_calc(struct _pid *pid, int32_t busy)
 	pid->last_err = fp_error;
 
 	result = pterm + mul_fp(pid->integral, pid->i_gain) + dterm;
-	if (result >= 0)
-		result = result + (1 << (FRAC_BITS-1));
-	else
-		result = result - (1 << (FRAC_BITS-1));
+	result = result + (1 << (FRAC_BITS-1));
 	return (signed int)fp_toint(result);
 }
 
@@ -290,7 +288,10 @@ static ssize_t store_no_turbo(struct kobject *a, struct attribute *b,
 	if (ret != 1)
 		return -EINVAL;
 	limits.no_turbo = clamp_t(int, input, 0 , 1);
-
+	if (limits.turbo_disabled) {
+		pr_warn("Turbo disabled by BIOS or unavailable on processor\n");
+		limits.no_turbo = limits.turbo_disabled;
+	}
 	return count;
 }
 
@@ -360,21 +361,21 @@ static int byt_get_min_pstate(void)
 {
 	u64 value;
 	rdmsrl(BYT_RATIOS, value);
-	return (value >> 8) & 0x3F;
+	return (value >> 8) & 0x7F;
 }
 
 static int byt_get_max_pstate(void)
 {
 	u64 value;
 	rdmsrl(BYT_RATIOS, value);
-	return (value >> 16) & 0x3F;
+	return (value >> 16) & 0x7F;
 }
 
 static int byt_get_turbo_pstate(void)
 {
 	u64 value;
 	rdmsrl(BYT_TURBO_RATIOS, value);
-	return value & 0x3F;
+	return value & 0x7F;
 }
 
 static void byt_set_pstate(struct cpudata *cpudata, int pstate)
@@ -384,7 +385,7 @@ static void byt_set_pstate(struct cpudata *cpudata, int pstate)
 	u32 vid;
 
 	val = pstate << 8;
-	if (limits.no_turbo)
+	if (limits.no_turbo && !limits.turbo_disabled)
 		val |= (u64)1 << 32;
 
 	vid_fp = cpudata->vid.min + mul_fp(
@@ -408,8 +409,8 @@ static void byt_get_vid(struct cpudata *cpudata)
 
 
 	rdmsrl(BYT_VIDS, value);
-	cpudata->vid.min = int_tofp((value >> 8) & 0x3f);
-	cpudata->vid.max = int_tofp((value >> 16) & 0x3f);
+	cpudata->vid.min = int_tofp((value >> 8) & 0x7f);
+	cpudata->vid.max = int_tofp((value >> 16) & 0x7f);
 	cpudata->vid.ratio = div_fp(
 		cpudata->vid.max - cpudata->vid.min,
 		int_tofp(cpudata->pstate.max_pstate -
@@ -451,7 +452,7 @@ static void core_set_pstate(struct cpudata *cpudata, int pstate)
 	u64 val;
 
 	val = pstate << 8;
-	if (limits.no_turbo)
+	if (limits.no_turbo && !limits.turbo_disabled)
 		val |= (u64)1 << 32;
 
 	wrmsrl_on_cpu(cpudata->cpu, MSR_IA32_PERF_CTL, val);
@@ -691,23 +692,16 @@ MODULE_DEVICE_TABLE(x86cpu, intel_pstate_cpu_ids);
 
 static int intel_pstate_init_cpu(unsigned int cpunum)
 {
-
-	const struct x86_cpu_id *id;
 	struct cpudata *cpu;
 
-	id = x86_match_cpu(intel_pstate_cpu_ids);
-	if (!id)
-		return -ENODEV;
-
 	all_cpu_data[cpunum] = kzalloc(sizeof(struct cpudata), GFP_KERNEL);
 	if (!all_cpu_data[cpunum])
 		return -ENOMEM;
 
 	cpu = all_cpu_data[cpunum];
 
-	intel_pstate_get_cpu_pstates(cpu);
-
 	cpu->cpu = cpunum;
+	intel_pstate_get_cpu_pstates(cpu);
 
 	init_timer_deferrable(&cpu->timer);
 	cpu->timer.function = intel_pstate_timer_func;
@@ -750,7 +744,7 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 		limits.min_perf = int_tofp(1);
 		limits.max_perf_pct = 100;
 		limits.max_perf = int_tofp(1);
-		limits.no_turbo = 0;
+		limits.no_turbo = limits.turbo_disabled;
 		return 0;
 	}
 	limits.min_perf_pct = (policy->min * 100) / policy->cpuinfo.max_freq;
@@ -793,6 +787,7 @@ static int intel_pstate_cpu_init(struct cpufreq_policy *policy)
 {
 	struct cpudata *cpu;
 	int rc;
+	u64 misc_en;
 
 	rc = intel_pstate_init_cpu(policy->cpu);
 	if (rc)
@@ -800,8 +795,13 @@ static int intel_pstate_cpu_init(struct cpufreq_policy *policy)
 
 	cpu = all_cpu_data[policy->cpu];
 
-	if (!limits.no_turbo &&
-		limits.min_perf_pct == 100 && limits.max_perf_pct == 100)
+	rdmsrl(MSR_IA32_MISC_ENABLE, misc_en);
+	if (misc_en & MSR_IA32_MISC_ENABLE_TURBO_DISABLE ||
+		cpu->pstate.max_pstate == cpu->pstate.turbo_pstate) {
+		limits.turbo_disabled = 1;
+		limits.no_turbo = 1;
+	}
+	if (limits.min_perf_pct == 100 && limits.max_perf_pct == 100)
 		policy->policy = CPUFREQ_POLICY_PERFORMANCE;
 	else
 		policy->policy = CPUFREQ_POLICY_POWERSAVE;
diff --git a/drivers/cpufreq/ppc-corenet-cpufreq.c b/drivers/cpufreq/ppc-corenet-cpufreq.c
index 0af618abebaf..3607070797af 100644
--- a/drivers/cpufreq/ppc-corenet-cpufreq.c
+++ b/drivers/cpufreq/ppc-corenet-cpufreq.c
@@ -138,7 +138,7 @@ static int corenet_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	struct cpufreq_frequency_table *table;
 	struct cpu_data *data;
 	unsigned int cpu = policy->cpu;
-	u64 transition_latency_hz;
+	u64 u64temp;
 
 	np = of_get_cpu_node(cpu, NULL);
 	if (!np)
@@ -206,9 +206,10 @@ static int corenet_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	for_each_cpu(i, per_cpu(cpu_mask, cpu))
 		per_cpu(cpu_data, i) = data;
 
-	transition_latency_hz = 12ULL * NSEC_PER_SEC;
-	policy->cpuinfo.transition_latency =
-		do_div(transition_latency_hz, fsl_get_sys_freq());
+	/* Minimum transition latency is 12 platform clocks */
+	u64temp = 12ULL * NSEC_PER_SEC;
+	do_div(u64temp, fsl_get_sys_freq());
+	policy->cpuinfo.transition_latency = u64temp + 1;
 
 	of_node_put(np);
 
diff --git a/drivers/cpufreq/tegra-cpufreq.c b/drivers/cpufreq/tegra-cpufreq.c
index 6e774c6ac20b..8084c7f7e206 100644
--- a/drivers/cpufreq/tegra-cpufreq.c
+++ b/drivers/cpufreq/tegra-cpufreq.c
@@ -45,46 +45,54 @@ static struct clk *cpu_clk;
 static struct clk *pll_x_clk;
 static struct clk *pll_p_clk;
 static struct clk *emc_clk;
+static bool pll_x_prepared;
 
-static int tegra_cpu_clk_set_rate(unsigned long rate)
+static unsigned int tegra_get_intermediate(struct cpufreq_policy *policy,
+					   unsigned int index)
+{
+	unsigned int ifreq = clk_get_rate(pll_p_clk) / 1000;
+
+	/*
+	 * Don't switch to intermediate freq if:
+	 * - we are already at it, i.e. policy->cur == ifreq
+	 * - index corresponds to ifreq
+	 */
+	if ((freq_table[index].frequency == ifreq) || (policy->cur == ifreq))
+		return 0;
+
+	return ifreq;
+}
+
+static int tegra_target_intermediate(struct cpufreq_policy *policy,
+				     unsigned int index)
 {
 	int ret;
 
 	/*
 	 * Take an extra reference to the main pll so it doesn't turn
-	 * off when we move the cpu off of it
+	 * off when we move the cpu off of it as enabling it again while we
+	 * switch to it from tegra_target() would take additional time.
+	 *
+	 * When target-freq is equal to intermediate freq we don't need to
+	 * switch to an intermediate freq and so this routine isn't called.
+	 * Also, we wouldn't be using pll_x anymore and must not take extra
+	 * reference to it, as it can be disabled now to save some power.
 	 */
 	clk_prepare_enable(pll_x_clk);
 
 	ret = clk_set_parent(cpu_clk, pll_p_clk);
-	if (ret) {
-		pr_err("Failed to switch cpu to clock pll_p\n");
-		goto out;
-	}
-
-	if (rate == clk_get_rate(pll_p_clk))
-		goto out;
-
-	ret = clk_set_rate(pll_x_clk, rate);
-	if (ret) {
-		pr_err("Failed to change pll_x to %lu\n", rate);
-		goto out;
-	}
-
-	ret = clk_set_parent(cpu_clk, pll_x_clk);
-	if (ret) {
-		pr_err("Failed to switch cpu to clock pll_x\n");
-		goto out;
-	}
+	if (ret)
+		clk_disable_unprepare(pll_x_clk);
+	else
+		pll_x_prepared = true;
 
-out:
-	clk_disable_unprepare(pll_x_clk);
 	return ret;
 }
 
 static int tegra_target(struct cpufreq_policy *policy, unsigned int index)
 {
 	unsigned long rate = freq_table[index].frequency;
+	unsigned int ifreq = clk_get_rate(pll_p_clk) / 1000;
 	int ret = 0;
 
 	/*
@@ -98,10 +106,30 @@ static int tegra_target(struct cpufreq_policy *policy, unsigned int index)
 	else
 		clk_set_rate(emc_clk, 100000000);  /* emc 50Mhz */
 
-	ret = tegra_cpu_clk_set_rate(rate * 1000);
+	/*
+	 * target freq == pll_p, don't need to take extra reference to pll_x_clk
+	 * as it isn't used anymore.
+	 */
+	if (rate == ifreq)
+		return clk_set_parent(cpu_clk, pll_p_clk);
+
+	ret = clk_set_rate(pll_x_clk, rate * 1000);
+	/* Restore to earlier frequency on error, i.e. pll_x */
 	if (ret)
-		pr_err("cpu-tegra: Failed to set cpu frequency to %lu kHz\n",
-			rate);
+		pr_err("Failed to change pll_x to %lu\n", rate);
+
+	ret = clk_set_parent(cpu_clk, pll_x_clk);
+	/* This shouldn't fail while changing or restoring */
+	WARN_ON(ret);
+
+	/*
+	 * Drop count to pll_x clock only if we switched to intermediate freq
+	 * earlier while transitioning to a target frequency.
+	 */
+	if (pll_x_prepared) {
+		clk_disable_unprepare(pll_x_clk);
+		pll_x_prepared = false;
+	}
 
 	return ret;
 }
@@ -137,16 +165,18 @@ static int tegra_cpu_exit(struct cpufreq_policy *policy)
 }
 
 static struct cpufreq_driver tegra_cpufreq_driver = {
-	.flags		= CPUFREQ_NEED_INITIAL_FREQ_CHECK,
-	.verify		= cpufreq_generic_frequency_table_verify,
-	.target_index	= tegra_target,
-	.get		= cpufreq_generic_get,
-	.init		= tegra_cpu_init,
-	.exit		= tegra_cpu_exit,
-	.name		= "tegra",
-	.attr		= cpufreq_generic_attr,
+	.flags			= CPUFREQ_NEED_INITIAL_FREQ_CHECK,
+	.verify			= cpufreq_generic_frequency_table_verify,
+	.get_intermediate	= tegra_get_intermediate,
+	.target_intermediate	= tegra_target_intermediate,
+	.target_index		= tegra_target,
+	.get			= cpufreq_generic_get,
+	.init			= tegra_cpu_init,
+	.exit			= tegra_cpu_exit,
+	.name			= "tegra",
+	.attr			= cpufreq_generic_attr,
 #ifdef CONFIG_PM
-	.suspend	= cpufreq_generic_suspend,
+	.suspend		= cpufreq_generic_suspend,
 #endif
 };
 
diff --git a/drivers/cpuidle/Kconfig b/drivers/cpuidle/Kconfig
index f04e25f6c98d..1b96fb91d32c 100644
--- a/drivers/cpuidle/Kconfig
+++ b/drivers/cpuidle/Kconfig
@@ -35,6 +35,11 @@ depends on ARM
 source "drivers/cpuidle/Kconfig.arm"
 endmenu
 
+menu "MIPS CPU Idle Drivers"
+depends on MIPS
+source "drivers/cpuidle/Kconfig.mips"
+endmenu
+
 menu "POWERPC CPU Idle Drivers"
 depends on PPC
 source "drivers/cpuidle/Kconfig.powerpc"
diff --git a/drivers/cpuidle/Kconfig.mips b/drivers/cpuidle/Kconfig.mips
new file mode 100644
index 000000000000..0e70ee28a5ca
--- /dev/null
+++ b/drivers/cpuidle/Kconfig.mips
@@ -0,0 +1,17 @@
+#
+# MIPS CPU Idle Drivers
+#
+config MIPS_CPS_CPUIDLE
+	bool "CPU Idle driver for MIPS CPS platforms"
+	depends on CPU_IDLE
+	depends on SYS_SUPPORTS_MIPS_CPS
+	select ARCH_NEEDS_CPU_IDLE_COUPLED if MIPS_MT
+	select GENERIC_CLOCKEVENTS_BROADCAST if SMP
+	select MIPS_CPS_PM
+	default y
+	help
+	  Select this option to enable processor idle state management
+	  through cpuidle for systems built around the MIPS Coherent
+	  Processing System (CPS) architecture. In order to make use of
+	  the deepest idle states you will need to ensure that you are
+	  also using the CONFIG_MIPS_CPS SMP implementation.
diff --git a/drivers/cpuidle/Makefile b/drivers/cpuidle/Makefile
index 9b5b2b560d70..d8bb1ff72561 100644
--- a/drivers/cpuidle/Makefile
+++ b/drivers/cpuidle/Makefile
@@ -18,6 +18,10 @@ obj-$(CONFIG_ARM_AT91_CPUIDLE)          += cpuidle-at91.o
 obj-$(CONFIG_ARM_EXYNOS_CPUIDLE)        += cpuidle-exynos.o
 
 ###############################################################################
+# MIPS drivers
+obj-$(CONFIG_MIPS_CPS_CPUIDLE)		+= cpuidle-cps.o
+
+###############################################################################
 # POWERPC drivers
 obj-$(CONFIG_PSERIES_CPUIDLE)		+= cpuidle-pseries.o
 obj-$(CONFIG_POWERNV_CPUIDLE)		+= cpuidle-powernv.o
diff --git a/drivers/cpuidle/cpuidle-armada-370-xp.c b/drivers/cpuidle/cpuidle-armada-370-xp.c
index 28587d0f3947..a5fba0287bfb 100644
--- a/drivers/cpuidle/cpuidle-armada-370-xp.c
+++ b/drivers/cpuidle/cpuidle-armada-370-xp.c
@@ -55,7 +55,7 @@ static struct cpuidle_driver armada_370_xp_idle_driver = {
 		.power_usage		= 50,
 		.target_residency	= 100,
 		.flags			= CPUIDLE_FLAG_TIME_VALID,
-		.name			= "MV CPU IDLE",
+		.name			= "Idle",
 		.desc			= "CPU power down",
 	},
 	.states[2]		= {
@@ -65,7 +65,7 @@ static struct cpuidle_driver armada_370_xp_idle_driver = {
 		.target_residency	= 1000,
 		.flags			= CPUIDLE_FLAG_TIME_VALID |
 						ARMADA_370_XP_FLAG_DEEP_IDLE,
-		.name			= "MV CPU DEEP IDLE",
+		.name			= "Deep idle",
 		.desc			= "CPU and L2 Fabric power down",
 	},
 	.state_count = ARMADA_370_XP_MAX_STATES,
diff --git a/drivers/cpuidle/cpuidle-cps.c b/drivers/cpuidle/cpuidle-cps.c
new file mode 100644
index 000000000000..fc7b62720deb
--- /dev/null
+++ b/drivers/cpuidle/cpuidle-cps.c
@@ -0,0 +1,186 @@
+/*
+ * Copyright (C) 2014 Imagination Technologies
+ * Author: Paul Burton <paul.burton@imgtec.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/cpu_pm.h>
+#include <linux/cpuidle.h>
+#include <linux/init.h>
+
+#include <asm/idle.h>
+#include <asm/pm-cps.h>
+
+/* Enumeration of the various idle states this driver may enter */
+enum cps_idle_state {
+	STATE_WAIT = 0,		/* MIPS wait instruction, coherent */
+	STATE_NC_WAIT,		/* MIPS wait instruction, non-coherent */
+	STATE_CLOCK_GATED,	/* Core clock gated */
+	STATE_POWER_GATED,	/* Core power gated */
+	STATE_COUNT
+};
+
+static int cps_nc_enter(struct cpuidle_device *dev,
+			struct cpuidle_driver *drv, int index)
+{
+	enum cps_pm_state pm_state;
+	int err;
+
+	/*
+	 * At least one core must remain powered up & clocked in order for the
+	 * system to have any hope of functioning.
+	 *
+	 * TODO: don't treat core 0 specially, just prevent the final core
+	 * TODO: remap interrupt affinity temporarily
+	 */
+	if (!cpu_data[dev->cpu].core && (index > STATE_NC_WAIT))
+		index = STATE_NC_WAIT;
+
+	/* Select the appropriate cps_pm_state */
+	switch (index) {
+	case STATE_NC_WAIT:
+		pm_state = CPS_PM_NC_WAIT;
+		break;
+	case STATE_CLOCK_GATED:
+		pm_state = CPS_PM_CLOCK_GATED;
+		break;
+	case STATE_POWER_GATED:
+		pm_state = CPS_PM_POWER_GATED;
+		break;
+	default:
+		BUG();
+		return -EINVAL;
+	}
+
+	/* Notify listeners the CPU is about to power down */
+	if ((pm_state == CPS_PM_POWER_GATED) && cpu_pm_enter())
+		return -EINTR;
+
+	/* Enter that state */
+	err = cps_pm_enter_state(pm_state);
+
+	/* Notify listeners the CPU is back up */
+	if (pm_state == CPS_PM_POWER_GATED)
+		cpu_pm_exit();
+
+	return err ?: index;
+}
+
+static struct cpuidle_driver cps_driver = {
+	.name			= "cpc_cpuidle",
+	.owner			= THIS_MODULE,
+	.states = {
+		[STATE_WAIT] = MIPS_CPUIDLE_WAIT_STATE,
+		[STATE_NC_WAIT] = {
+			.enter	= cps_nc_enter,
+			.exit_latency		= 200,
+			.target_residency	= 450,
+			.flags	= CPUIDLE_FLAG_TIME_VALID,
+			.name	= "nc-wait",
+			.desc	= "non-coherent MIPS wait",
+		},
+		[STATE_CLOCK_GATED] = {
+			.enter	= cps_nc_enter,
+			.exit_latency		= 300,
+			.target_residency	= 700,
+			.flags	= CPUIDLE_FLAG_TIME_VALID |
+				  CPUIDLE_FLAG_TIMER_STOP,
+			.name	= "clock-gated",
+			.desc	= "core clock gated",
+		},
+		[STATE_POWER_GATED] = {
+			.enter	= cps_nc_enter,
+			.exit_latency		= 600,
+			.target_residency	= 1000,
+			.flags	= CPUIDLE_FLAG_TIME_VALID |
+				  CPUIDLE_FLAG_TIMER_STOP,
+			.name	= "power-gated",
+			.desc	= "core power gated",
+		},
+	},
+	.state_count		= STATE_COUNT,
+	.safe_state_index	= 0,
+};
+
+static void __init cps_cpuidle_unregister(void)
+{
+	int cpu;
+	struct cpuidle_device *device;
+
+	for_each_possible_cpu(cpu) {
+		device = &per_cpu(cpuidle_dev, cpu);
+		cpuidle_unregister_device(device);
+	}
+
+	cpuidle_unregister_driver(&cps_driver);
+}
+
+static int __init cps_cpuidle_init(void)
+{
+	int err, cpu, core, i;
+	struct cpuidle_device *device;
+
+	/* Detect supported states */
+	if (!cps_pm_support_state(CPS_PM_POWER_GATED))
+		cps_driver.state_count = STATE_CLOCK_GATED + 1;
+	if (!cps_pm_support_state(CPS_PM_CLOCK_GATED))
+		cps_driver.state_count = STATE_NC_WAIT + 1;
+	if (!cps_pm_support_state(CPS_PM_NC_WAIT))
+		cps_driver.state_count = STATE_WAIT + 1;
+
+	/* Inform the user if some states are unavailable */
+	if (cps_driver.state_count < STATE_COUNT) {
+		pr_info("cpuidle-cps: limited to ");
+		switch (cps_driver.state_count - 1) {
+		case STATE_WAIT:
+			pr_cont("coherent wait\n");
+			break;
+		case STATE_NC_WAIT:
+			pr_cont("non-coherent wait\n");
+			break;
+		case STATE_CLOCK_GATED:
+			pr_cont("clock gating\n");
+			break;
+		}
+	}
+
+	/*
+	 * Set the coupled flag on the appropriate states if this system
+	 * requires it.
+	 */
+	if (coupled_coherence)
+		for (i = STATE_NC_WAIT; i < cps_driver.state_count; i++)
+			cps_driver.states[i].flags |= CPUIDLE_FLAG_COUPLED;
+
+	err = cpuidle_register_driver(&cps_driver);
+	if (err) {
+		pr_err("Failed to register CPS cpuidle driver\n");
+		return err;
+	}
+
+	for_each_possible_cpu(cpu) {
+		core = cpu_data[cpu].core;
+		device = &per_cpu(cpuidle_dev, cpu);
+		device->cpu = cpu;
+#ifdef CONFIG_MIPS_MT
+		cpumask_copy(&device->coupled_cpus, &cpu_sibling_map[cpu]);
+#endif
+
+		err = cpuidle_register_device(device);
+		if (err) {
+			pr_err("Failed to register CPU%d cpuidle device\n",
+			       cpu);
+			goto err_out;
+		}
+	}
+
+	return 0;
+err_out:
+	cps_cpuidle_unregister();
+	return err;
+}
+device_initcall(cps_cpuidle_init);
diff --git a/drivers/cpuidle/cpuidle-powernv.c b/drivers/cpuidle/cpuidle-powernv.c
index 719f6fb5b1c3..74f5788d50b1 100644
--- a/drivers/cpuidle/cpuidle-powernv.c
+++ b/drivers/cpuidle/cpuidle-powernv.c
@@ -73,12 +73,10 @@ static int fastsleep_loop(struct cpuidle_device *dev,
 		return index;
 
 	new_lpcr = old_lpcr;
-	new_lpcr &= ~(LPCR_MER | LPCR_PECE); /* lpcr[mer] must be 0 */
-
-	/* exit powersave upon external interrupt, but not decrementer
-	 * interrupt.
+	/* Do not exit powersave upon decrementer as we've setup the timer
+	 * offload.
 	 */
-	new_lpcr |= LPCR_PECE0;
+	new_lpcr &= ~LPCR_PECE1;
 
 	mtspr(SPRN_LPCR, new_lpcr);
 	power7_sleep();
diff --git a/drivers/cpuidle/driver.c b/drivers/cpuidle/driver.c
index 136d6a283e0a..9634f20e3926 100644
--- a/drivers/cpuidle/driver.c
+++ b/drivers/cpuidle/driver.c
@@ -187,8 +187,11 @@ static int poll_idle(struct cpuidle_device *dev,
 
 	t1 = ktime_get();
 	local_irq_enable();
-	while (!need_resched())
-		cpu_relax();
+	if (!current_set_polling_and_test()) {
+		while (!need_resched())
+			cpu_relax();
+	}
+	current_clr_polling();
 
 	t2 = ktime_get();
 	diff = ktime_to_us(ktime_sub(t2, t1));
diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig
index f066fa23cc05..02f177aeb16c 100644
--- a/drivers/crypto/Kconfig
+++ b/drivers/crypto/Kconfig
@@ -313,7 +313,7 @@ config CRYPTO_DEV_S5P
 
 config CRYPTO_DEV_NX
 	bool "Support for IBM Power7+ in-Nest cryptographic acceleration"
-	depends on PPC64 && IBMVIO
+	depends on PPC64 && IBMVIO && !CPU_LITTLE_ENDIAN
 	default n
 	help
 	  Support for Power7+ in-Nest cryptographic acceleration.
diff --git a/drivers/crypto/caam/jr.c b/drivers/crypto/caam/jr.c
index 1d80bd3636c5..b512a4ba7569 100644
--- a/drivers/crypto/caam/jr.c
+++ b/drivers/crypto/caam/jr.c
@@ -453,8 +453,8 @@ static int caam_jr_probe(struct platform_device *pdev)
 	int error;
 
 	jrdev = &pdev->dev;
-	jrpriv = kmalloc(sizeof(struct caam_drv_private_jr),
-			 GFP_KERNEL);
+	jrpriv = devm_kmalloc(jrdev, sizeof(struct caam_drv_private_jr),
+			      GFP_KERNEL);
 	if (!jrpriv)
 		return -ENOMEM;
 
@@ -487,10 +487,8 @@ static int caam_jr_probe(struct platform_device *pdev)
 
 	/* Now do the platform independent part */
 	error = caam_jr_init(jrdev); /* now turn on hardware */
-	if (error) {
-		kfree(jrpriv);
+	if (error)
 		return error;
-	}
 
 	jrpriv->dev = jrdev;
 	spin_lock(&driver_data.jr_alloc_lock);
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 5c5863842de9..1eca7b9760e6 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -234,7 +234,7 @@ config PL330_DMA
 
 config PCH_DMA
 	tristate "Intel EG20T PCH / LAPIS Semicon IOH(ML7213/ML7223/ML7831) DMA"
-	depends on PCI && X86
+	depends on PCI && (X86_32 || COMPILE_TEST)
 	select DMA_ENGINE
 	help
 	  Enable support for Intel EG20T PCH DMA engine.
@@ -269,7 +269,7 @@ config MXS_DMA
 	select DMA_ENGINE
 	help
 	  Support the MXS DMA engine. This engine including APBH-DMA
-	  and APBX-DMA is integrated into Freescale i.MX23/28 chips.
+	  and APBX-DMA is integrated into Freescale i.MX23/28/MX6Q/MX6DL chips.
 
 config EP93XX_DMA
 	bool "Cirrus Logic EP93xx DMA support"
@@ -361,6 +361,20 @@ config FSL_EDMA
 	  multiplexing capability for DMA request sources(slot).
 	  This module can be found on Freescale Vybrid and LS-1 SoCs.
 
+config XILINX_VDMA
+	tristate "Xilinx AXI VDMA Engine"
+	depends on (ARCH_ZYNQ || MICROBLAZE)
+	select DMA_ENGINE
+	help
+	  Enable support for Xilinx AXI VDMA Soft IP.
+
+	  This engine provides high-bandwidth direct memory access
+	  between memory and AXI4-Stream video type target
+	  peripherals including peripherals which support AXI4-
+	  Stream Video Protocol.  It has two stream interfaces/
+	  channels, Memory Mapped to Stream (MM2S) and Stream to
+	  Memory Mapped (S2MM) for the data transfers.
+
 config DMA_ENGINE
 	bool
 
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 5150c82c9caf..c779e1eb2db2 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -46,3 +46,4 @@ obj-$(CONFIG_K3_DMA) += k3dma.o
 obj-$(CONFIG_MOXART_DMA) += moxart-dma.o
 obj-$(CONFIG_FSL_EDMA) += fsl-edma.o
 obj-$(CONFIG_QCOM_BAM_DMA) += qcom_bam_dma.o
+obj-y += xilinx/
diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c
index d028f36ae655..8f8b0b608875 100644
--- a/drivers/dma/cppi41.c
+++ b/drivers/dma/cppi41.c
@@ -86,6 +86,9 @@
 
 #define USBSS_IRQ_PD_COMP	(1 <<  2)
 
+/* Packet Descriptor */
+#define PD2_ZERO_LENGTH		(1 << 19)
+
 struct cppi41_channel {
 	struct dma_chan chan;
 	struct dma_async_tx_descriptor txd;
@@ -307,7 +310,7 @@ static irqreturn_t cppi41_irq(int irq, void *data)
 			__iormb();
 
 		while (val) {
-			u32 desc;
+			u32 desc, len;
 
 			q_num = __fls(val);
 			val &= ~(1 << q_num);
@@ -319,9 +322,13 @@ static irqreturn_t cppi41_irq(int irq, void *data)
 						q_num, desc);
 				continue;
 			}
-			c->residue = pd_trans_len(c->desc->pd6) -
-				pd_trans_len(c->desc->pd0);
 
+			if (c->desc->pd2 & PD2_ZERO_LENGTH)
+				len = 0;
+			else
+				len = pd_trans_len(c->desc->pd0);
+
+			c->residue = pd_trans_len(c->desc->pd6) - len;
 			dma_cookie_complete(&c->txd);
 			c->txd.callback(c->txd.callback_param);
 		}
diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c
index 7a740769c2fa..a27ded53ab4f 100644
--- a/drivers/dma/dw/core.c
+++ b/drivers/dma/dw/core.c
@@ -1493,6 +1493,13 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 	dw->regs = chip->regs;
 	chip->dw = dw;
 
+	dw->clk = devm_clk_get(chip->dev, "hclk");
+	if (IS_ERR(dw->clk))
+		return PTR_ERR(dw->clk);
+	err = clk_prepare_enable(dw->clk);
+	if (err)
+		return err;
+
 	dw_params = dma_read_byaddr(chip->regs, DW_PARAMS);
 	autocfg = dw_params >> DW_PARAMS_EN & 0x1;
 
@@ -1500,15 +1507,19 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 
 	if (!pdata && autocfg) {
 		pdata = devm_kzalloc(chip->dev, sizeof(*pdata), GFP_KERNEL);
-		if (!pdata)
-			return -ENOMEM;
+		if (!pdata) {
+			err = -ENOMEM;
+			goto err_pdata;
+		}
 
 		/* Fill platform data with the default values */
 		pdata->is_private = true;
 		pdata->chan_allocation_order = CHAN_ALLOCATION_ASCENDING;
 		pdata->chan_priority = CHAN_PRIORITY_ASCENDING;
-	} else if (!pdata || pdata->nr_channels > DW_DMA_MAX_NR_CHANNELS)
-		return -EINVAL;
+	} else if (!pdata || pdata->nr_channels > DW_DMA_MAX_NR_CHANNELS) {
+		err = -EINVAL;
+		goto err_pdata;
+	}
 
 	if (autocfg)
 		nr_channels = (dw_params >> DW_PARAMS_NR_CHAN & 0x7) + 1;
@@ -1517,13 +1528,10 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 
 	dw->chan = devm_kcalloc(chip->dev, nr_channels, sizeof(*dw->chan),
 				GFP_KERNEL);
-	if (!dw->chan)
-		return -ENOMEM;
-
-	dw->clk = devm_clk_get(chip->dev, "hclk");
-	if (IS_ERR(dw->clk))
-		return PTR_ERR(dw->clk);
-	clk_prepare_enable(dw->clk);
+	if (!dw->chan) {
+		err = -ENOMEM;
+		goto err_pdata;
+	}
 
 	/* Get hardware configuration parameters */
 	if (autocfg) {
@@ -1553,7 +1561,8 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 					 sizeof(struct dw_desc), 4, 0);
 	if (!dw->desc_pool) {
 		dev_err(chip->dev, "No memory for descriptors dma pool\n");
-		return -ENOMEM;
+		err = -ENOMEM;
+		goto err_pdata;
 	}
 
 	tasklet_init(&dw->tasklet, dw_dma_tasklet, (unsigned long)dw);
@@ -1561,7 +1570,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 	err = request_irq(chip->irq, dw_dma_interrupt, IRQF_SHARED,
 			  "dw_dmac", dw);
 	if (err)
-		return err;
+		goto err_pdata;
 
 	INIT_LIST_HEAD(&dw->dma.channels);
 	for (i = 0; i < nr_channels; i++) {
@@ -1650,12 +1659,20 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 
 	dma_writel(dw, CFG, DW_CFG_DMA_EN);
 
+	err = dma_async_device_register(&dw->dma);
+	if (err)
+		goto err_dma_register;
+
 	dev_info(chip->dev, "DesignWare DMA Controller, %d channels\n",
 		 nr_channels);
 
-	dma_async_device_register(&dw->dma);
-
 	return 0;
+
+err_dma_register:
+	free_irq(chip->irq, dw);
+err_pdata:
+	clk_disable_unprepare(dw->clk);
+	return err;
 }
 EXPORT_SYMBOL_GPL(dw_dma_probe);
 
@@ -1676,6 +1693,8 @@ int dw_dma_remove(struct dw_dma_chip *chip)
 		channel_clear_bit(dw, CH_EN, dwc->mask);
 	}
 
+	clk_disable_unprepare(dw->clk);
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dw_dma_remove);
diff --git a/drivers/dma/dw/pci.c b/drivers/dma/dw/pci.c
index fec59f1a77bb..39e30c3c7a9d 100644
--- a/drivers/dma/dw/pci.c
+++ b/drivers/dma/dw/pci.c
@@ -93,19 +93,13 @@ static int dw_pci_resume_early(struct device *dev)
 	return dw_dma_resume(chip);
 };
 
-#else /* !CONFIG_PM_SLEEP */
-
-#define dw_pci_suspend_late	NULL
-#define dw_pci_resume_early	NULL
-
-#endif /* !CONFIG_PM_SLEEP */
+#endif /* CONFIG_PM_SLEEP */
 
 static const struct dev_pm_ops dw_pci_dev_pm_ops = {
-	.suspend_late = dw_pci_suspend_late,
-	.resume_early = dw_pci_resume_early,
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(dw_pci_suspend_late, dw_pci_resume_early)
 };
 
-static DEFINE_PCI_DEVICE_TABLE(dw_pci_id_table) = {
+static const struct pci_device_id dw_pci_id_table[] = {
 	/* Medfield */
 	{ PCI_VDEVICE(INTEL, 0x0827), (kernel_ulong_t)&dw_pci_pdata },
 	{ PCI_VDEVICE(INTEL, 0x0830), (kernel_ulong_t)&dw_pci_pdata },
diff --git a/drivers/dma/dw/platform.c b/drivers/dma/dw/platform.c
index 453822cc4f9d..c5b339af6be5 100644
--- a/drivers/dma/dw/platform.c
+++ b/drivers/dma/dw/platform.c
@@ -256,7 +256,7 @@ MODULE_DEVICE_TABLE(acpi, dw_dma_acpi_id_table);
 
 #ifdef CONFIG_PM_SLEEP
 
-static int dw_suspend_noirq(struct device *dev)
+static int dw_suspend_late(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct dw_dma_chip *chip = platform_get_drvdata(pdev);
@@ -264,7 +264,7 @@ static int dw_suspend_noirq(struct device *dev)
 	return dw_dma_suspend(chip);
 }
 
-static int dw_resume_noirq(struct device *dev)
+static int dw_resume_early(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct dw_dma_chip *chip = platform_get_drvdata(pdev);
@@ -272,20 +272,10 @@ static int dw_resume_noirq(struct device *dev)
 	return dw_dma_resume(chip);
 }
 
-#else /* !CONFIG_PM_SLEEP */
-
-#define dw_suspend_noirq	NULL
-#define dw_resume_noirq		NULL
-
-#endif /* !CONFIG_PM_SLEEP */
+#endif /* CONFIG_PM_SLEEP */
 
 static const struct dev_pm_ops dw_dev_pm_ops = {
-	.suspend_noirq = dw_suspend_noirq,
-	.resume_noirq = dw_resume_noirq,
-	.freeze_noirq = dw_suspend_noirq,
-	.thaw_noirq = dw_resume_noirq,
-	.restore_noirq = dw_resume_noirq,
-	.poweroff_noirq = dw_suspend_noirq,
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(dw_suspend_late, dw_resume_early)
 };
 
 static struct platform_driver dw_driver = {
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index f157c6f76b32..e0fec68aed25 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -61,6 +61,16 @@ static u32 get_sr(struct fsldma_chan *chan)
 	return DMA_IN(chan, &chan->regs->sr, 32);
 }
 
+static void set_mr(struct fsldma_chan *chan, u32 val)
+{
+	DMA_OUT(chan, &chan->regs->mr, val, 32);
+}
+
+static u32 get_mr(struct fsldma_chan *chan)
+{
+	return DMA_IN(chan, &chan->regs->mr, 32);
+}
+
 static void set_cdar(struct fsldma_chan *chan, dma_addr_t addr)
 {
 	DMA_OUT(chan, &chan->regs->cdar, addr | FSL_DMA_SNEN, 64);
@@ -71,6 +81,11 @@ static dma_addr_t get_cdar(struct fsldma_chan *chan)
 	return DMA_IN(chan, &chan->regs->cdar, 64) & ~FSL_DMA_SNEN;
 }
 
+static void set_bcr(struct fsldma_chan *chan, u32 val)
+{
+	DMA_OUT(chan, &chan->regs->bcr, val, 32);
+}
+
 static u32 get_bcr(struct fsldma_chan *chan)
 {
 	return DMA_IN(chan, &chan->regs->bcr, 32);
@@ -135,7 +150,7 @@ static void set_ld_eol(struct fsldma_chan *chan, struct fsl_desc_sw *desc)
 static void dma_init(struct fsldma_chan *chan)
 {
 	/* Reset the channel */
-	DMA_OUT(chan, &chan->regs->mr, 0, 32);
+	set_mr(chan, 0);
 
 	switch (chan->feature & FSL_DMA_IP_MASK) {
 	case FSL_DMA_IP_85XX:
@@ -144,16 +159,15 @@ static void dma_init(struct fsldma_chan *chan)
 		 * EOLNIE - End of links interrupt enable
 		 * BWC - Bandwidth sharing among channels
 		 */
-		DMA_OUT(chan, &chan->regs->mr, FSL_DMA_MR_BWC
-				| FSL_DMA_MR_EIE | FSL_DMA_MR_EOLNIE, 32);
+		set_mr(chan, FSL_DMA_MR_BWC | FSL_DMA_MR_EIE
+			| FSL_DMA_MR_EOLNIE);
 		break;
 	case FSL_DMA_IP_83XX:
 		/* Set the channel to below modes:
 		 * EOTIE - End-of-transfer interrupt enable
 		 * PRC_RM - PCI read multiple
 		 */
-		DMA_OUT(chan, &chan->regs->mr, FSL_DMA_MR_EOTIE
-				| FSL_DMA_MR_PRC_RM, 32);
+		set_mr(chan, FSL_DMA_MR_EOTIE | FSL_DMA_MR_PRC_RM);
 		break;
 	}
 }
@@ -175,10 +189,10 @@ static void dma_start(struct fsldma_chan *chan)
 {
 	u32 mode;
 
-	mode = DMA_IN(chan, &chan->regs->mr, 32);
+	mode = get_mr(chan);
 
 	if (chan->feature & FSL_DMA_CHAN_PAUSE_EXT) {
-		DMA_OUT(chan, &chan->regs->bcr, 0, 32);
+		set_bcr(chan, 0);
 		mode |= FSL_DMA_MR_EMP_EN;
 	} else {
 		mode &= ~FSL_DMA_MR_EMP_EN;
@@ -191,7 +205,7 @@ static void dma_start(struct fsldma_chan *chan)
 		mode |= FSL_DMA_MR_CS;
 	}
 
-	DMA_OUT(chan, &chan->regs->mr, mode, 32);
+	set_mr(chan, mode);
 }
 
 static void dma_halt(struct fsldma_chan *chan)
@@ -200,7 +214,7 @@ static void dma_halt(struct fsldma_chan *chan)
 	int i;
 
 	/* read the mode register */
-	mode = DMA_IN(chan, &chan->regs->mr, 32);
+	mode = get_mr(chan);
 
 	/*
 	 * The 85xx controller supports channel abort, which will stop
@@ -209,14 +223,14 @@ static void dma_halt(struct fsldma_chan *chan)
 	 */
 	if ((chan->feature & FSL_DMA_IP_MASK) == FSL_DMA_IP_85XX) {
 		mode |= FSL_DMA_MR_CA;
-		DMA_OUT(chan, &chan->regs->mr, mode, 32);
+		set_mr(chan, mode);
 
 		mode &= ~FSL_DMA_MR_CA;
 	}
 
 	/* stop the DMA controller */
 	mode &= ~(FSL_DMA_MR_CS | FSL_DMA_MR_EMS_EN);
-	DMA_OUT(chan, &chan->regs->mr, mode, 32);
+	set_mr(chan, mode);
 
 	/* wait for the DMA controller to become idle */
 	for (i = 0; i < 100; i++) {
@@ -245,7 +259,7 @@ static void fsl_chan_set_src_loop_size(struct fsldma_chan *chan, int size)
 {
 	u32 mode;
 
-	mode = DMA_IN(chan, &chan->regs->mr, 32);
+	mode = get_mr(chan);
 
 	switch (size) {
 	case 0:
@@ -259,7 +273,7 @@ static void fsl_chan_set_src_loop_size(struct fsldma_chan *chan, int size)
 		break;
 	}
 
-	DMA_OUT(chan, &chan->regs->mr, mode, 32);
+	set_mr(chan, mode);
 }
 
 /**
@@ -277,7 +291,7 @@ static void fsl_chan_set_dst_loop_size(struct fsldma_chan *chan, int size)
 {
 	u32 mode;
 
-	mode = DMA_IN(chan, &chan->regs->mr, 32);
+	mode = get_mr(chan);
 
 	switch (size) {
 	case 0:
@@ -291,7 +305,7 @@ static void fsl_chan_set_dst_loop_size(struct fsldma_chan *chan, int size)
 		break;
 	}
 
-	DMA_OUT(chan, &chan->regs->mr, mode, 32);
+	set_mr(chan, mode);
 }
 
 /**
@@ -312,10 +326,10 @@ static void fsl_chan_set_request_count(struct fsldma_chan *chan, int size)
 
 	BUG_ON(size > 1024);
 
-	mode = DMA_IN(chan, &chan->regs->mr, 32);
+	mode = get_mr(chan);
 	mode |= (__ilog2(size) << 24) & 0x0f000000;
 
-	DMA_OUT(chan, &chan->regs->mr, mode, 32);
+	set_mr(chan, mode);
 }
 
 /**
@@ -404,6 +418,19 @@ static dma_cookie_t fsl_dma_tx_submit(struct dma_async_tx_descriptor *tx)
 }
 
 /**
+ * fsl_dma_free_descriptor - Free descriptor from channel's DMA pool.
+ * @chan : Freescale DMA channel
+ * @desc: descriptor to be freed
+ */
+static void fsl_dma_free_descriptor(struct fsldma_chan *chan,
+		struct fsl_desc_sw *desc)
+{
+	list_del(&desc->node);
+	chan_dbg(chan, "LD %p free\n", desc);
+	dma_pool_free(chan->desc_pool, desc, desc->async_tx.phys);
+}
+
+/**
  * fsl_dma_alloc_descriptor - Allocate descriptor from channel's DMA pool.
  * @chan : Freescale DMA channel
  *
@@ -426,14 +453,107 @@ static struct fsl_desc_sw *fsl_dma_alloc_descriptor(struct fsldma_chan *chan)
 	desc->async_tx.tx_submit = fsl_dma_tx_submit;
 	desc->async_tx.phys = pdesc;
 
-#ifdef FSL_DMA_LD_DEBUG
 	chan_dbg(chan, "LD %p allocated\n", desc);
-#endif
 
 	return desc;
 }
 
 /**
+ * fsl_chan_xfer_ld_queue - transfer any pending transactions
+ * @chan : Freescale DMA channel
+ *
+ * HARDWARE STATE: idle
+ * LOCKING: must hold chan->desc_lock
+ */
+static void fsl_chan_xfer_ld_queue(struct fsldma_chan *chan)
+{
+	struct fsl_desc_sw *desc;
+
+	/*
+	 * If the list of pending descriptors is empty, then we
+	 * don't need to do any work at all
+	 */
+	if (list_empty(&chan->ld_pending)) {
+		chan_dbg(chan, "no pending LDs\n");
+		return;
+	}
+
+	/*
+	 * The DMA controller is not idle, which means that the interrupt
+	 * handler will start any queued transactions when it runs after
+	 * this transaction finishes
+	 */
+	if (!chan->idle) {
+		chan_dbg(chan, "DMA controller still busy\n");
+		return;
+	}
+
+	/*
+	 * If there are some link descriptors which have not been
+	 * transferred, we need to start the controller
+	 */
+
+	/*
+	 * Move all elements from the queue of pending transactions
+	 * onto the list of running transactions
+	 */
+	chan_dbg(chan, "idle, starting controller\n");
+	desc = list_first_entry(&chan->ld_pending, struct fsl_desc_sw, node);
+	list_splice_tail_init(&chan->ld_pending, &chan->ld_running);
+
+	/*
+	 * The 85xx DMA controller doesn't clear the channel start bit
+	 * automatically at the end of a transfer. Therefore we must clear
+	 * it in software before starting the transfer.
+	 */
+	if ((chan->feature & FSL_DMA_IP_MASK) == FSL_DMA_IP_85XX) {
+		u32 mode;
+
+		mode = get_mr(chan);
+		mode &= ~FSL_DMA_MR_CS;
+		set_mr(chan, mode);
+	}
+
+	/*
+	 * Program the descriptor's address into the DMA controller,
+	 * then start the DMA transaction
+	 */
+	set_cdar(chan, desc->async_tx.phys);
+	get_cdar(chan);
+
+	dma_start(chan);
+	chan->idle = false;
+}
+
+/**
+ * fsldma_cleanup_descriptor - cleanup and free a single link descriptor
+ * @chan: Freescale DMA channel
+ * @desc: descriptor to cleanup and free
+ *
+ * This function is used on a descriptor which has been executed by the DMA
+ * controller. It will run any callbacks, submit any dependencies, and then
+ * free the descriptor.
+ */
+static void fsldma_cleanup_descriptor(struct fsldma_chan *chan,
+				      struct fsl_desc_sw *desc)
+{
+	struct dma_async_tx_descriptor *txd = &desc->async_tx;
+
+	/* Run the link descriptor callback function */
+	if (txd->callback) {
+		chan_dbg(chan, "LD %p callback\n", desc);
+		txd->callback(txd->callback_param);
+	}
+
+	/* Run any dependencies */
+	dma_run_dependencies(txd);
+
+	dma_descriptor_unmap(txd);
+	chan_dbg(chan, "LD %p free\n", desc);
+	dma_pool_free(chan->desc_pool, desc, txd->phys);
+}
+
+/**
  * fsl_dma_alloc_chan_resources - Allocate resources for DMA channel.
  * @chan : Freescale DMA channel
  *
@@ -477,13 +597,8 @@ static void fsldma_free_desc_list(struct fsldma_chan *chan,
 {
 	struct fsl_desc_sw *desc, *_desc;
 
-	list_for_each_entry_safe(desc, _desc, list, node) {
-		list_del(&desc->node);
-#ifdef FSL_DMA_LD_DEBUG
-		chan_dbg(chan, "LD %p free\n", desc);
-#endif
-		dma_pool_free(chan->desc_pool, desc, desc->async_tx.phys);
-	}
+	list_for_each_entry_safe(desc, _desc, list, node)
+		fsl_dma_free_descriptor(chan, desc);
 }
 
 static void fsldma_free_desc_list_reverse(struct fsldma_chan *chan,
@@ -491,13 +606,8 @@ static void fsldma_free_desc_list_reverse(struct fsldma_chan *chan,
 {
 	struct fsl_desc_sw *desc, *_desc;
 
-	list_for_each_entry_safe_reverse(desc, _desc, list, node) {
-		list_del(&desc->node);
-#ifdef FSL_DMA_LD_DEBUG
-		chan_dbg(chan, "LD %p free\n", desc);
-#endif
-		dma_pool_free(chan->desc_pool, desc, desc->async_tx.phys);
-	}
+	list_for_each_entry_safe_reverse(desc, _desc, list, node)
+		fsl_dma_free_descriptor(chan, desc);
 }
 
 /**
@@ -520,35 +630,6 @@ static void fsl_dma_free_chan_resources(struct dma_chan *dchan)
 }
 
 static struct dma_async_tx_descriptor *
-fsl_dma_prep_interrupt(struct dma_chan *dchan, unsigned long flags)
-{
-	struct fsldma_chan *chan;
-	struct fsl_desc_sw *new;
-
-	if (!dchan)
-		return NULL;
-
-	chan = to_fsl_chan(dchan);
-
-	new = fsl_dma_alloc_descriptor(chan);
-	if (!new) {
-		chan_err(chan, "%s\n", msg_ld_oom);
-		return NULL;
-	}
-
-	new->async_tx.cookie = -EBUSY;
-	new->async_tx.flags = flags;
-
-	/* Insert the link descriptor to the LD ring */
-	list_add_tail(&new->node, &new->tx_list);
-
-	/* Set End-of-link to the last link descriptor of new list */
-	set_ld_eol(chan, new);
-
-	return &new->async_tx;
-}
-
-static struct dma_async_tx_descriptor *
 fsl_dma_prep_memcpy(struct dma_chan *dchan,
 	dma_addr_t dma_dst, dma_addr_t dma_src,
 	size_t len, unsigned long flags)
@@ -817,105 +898,6 @@ static int fsl_dma_device_control(struct dma_chan *dchan,
 }
 
 /**
- * fsldma_cleanup_descriptor - cleanup and free a single link descriptor
- * @chan: Freescale DMA channel
- * @desc: descriptor to cleanup and free
- *
- * This function is used on a descriptor which has been executed by the DMA
- * controller. It will run any callbacks, submit any dependencies, and then
- * free the descriptor.
- */
-static void fsldma_cleanup_descriptor(struct fsldma_chan *chan,
-				      struct fsl_desc_sw *desc)
-{
-	struct dma_async_tx_descriptor *txd = &desc->async_tx;
-
-	/* Run the link descriptor callback function */
-	if (txd->callback) {
-#ifdef FSL_DMA_LD_DEBUG
-		chan_dbg(chan, "LD %p callback\n", desc);
-#endif
-		txd->callback(txd->callback_param);
-	}
-
-	/* Run any dependencies */
-	dma_run_dependencies(txd);
-
-	dma_descriptor_unmap(txd);
-#ifdef FSL_DMA_LD_DEBUG
-	chan_dbg(chan, "LD %p free\n", desc);
-#endif
-	dma_pool_free(chan->desc_pool, desc, txd->phys);
-}
-
-/**
- * fsl_chan_xfer_ld_queue - transfer any pending transactions
- * @chan : Freescale DMA channel
- *
- * HARDWARE STATE: idle
- * LOCKING: must hold chan->desc_lock
- */
-static void fsl_chan_xfer_ld_queue(struct fsldma_chan *chan)
-{
-	struct fsl_desc_sw *desc;
-
-	/*
-	 * If the list of pending descriptors is empty, then we
-	 * don't need to do any work at all
-	 */
-	if (list_empty(&chan->ld_pending)) {
-		chan_dbg(chan, "no pending LDs\n");
-		return;
-	}
-
-	/*
-	 * The DMA controller is not idle, which means that the interrupt
-	 * handler will start any queued transactions when it runs after
-	 * this transaction finishes
-	 */
-	if (!chan->idle) {
-		chan_dbg(chan, "DMA controller still busy\n");
-		return;
-	}
-
-	/*
-	 * If there are some link descriptors which have not been
-	 * transferred, we need to start the controller
-	 */
-
-	/*
-	 * Move all elements from the queue of pending transactions
-	 * onto the list of running transactions
-	 */
-	chan_dbg(chan, "idle, starting controller\n");
-	desc = list_first_entry(&chan->ld_pending, struct fsl_desc_sw, node);
-	list_splice_tail_init(&chan->ld_pending, &chan->ld_running);
-
-	/*
-	 * The 85xx DMA controller doesn't clear the channel start bit
-	 * automatically at the end of a transfer. Therefore we must clear
-	 * it in software before starting the transfer.
-	 */
-	if ((chan->feature & FSL_DMA_IP_MASK) == FSL_DMA_IP_85XX) {
-		u32 mode;
-
-		mode = DMA_IN(chan, &chan->regs->mr, 32);
-		mode &= ~FSL_DMA_MR_CS;
-		DMA_OUT(chan, &chan->regs->mr, mode, 32);
-	}
-
-	/*
-	 * Program the descriptor's address into the DMA controller,
-	 * then start the DMA transaction
-	 */
-	set_cdar(chan, desc->async_tx.phys);
-	get_cdar(chan);
-
-	dma_start(chan);
-	chan->idle = false;
-}
-
-/**
  * fsl_dma_memcpy_issue_pending - Issue the DMA start command
  * @chan : Freescale DMA channel
  */
@@ -1304,12 +1286,10 @@ static int fsldma_of_probe(struct platform_device *op)
 	fdev->irq = irq_of_parse_and_map(op->dev.of_node, 0);
 
 	dma_cap_set(DMA_MEMCPY, fdev->common.cap_mask);
-	dma_cap_set(DMA_INTERRUPT, fdev->common.cap_mask);
 	dma_cap_set(DMA_SG, fdev->common.cap_mask);
 	dma_cap_set(DMA_SLAVE, fdev->common.cap_mask);
 	fdev->common.device_alloc_chan_resources = fsl_dma_alloc_chan_resources;
 	fdev->common.device_free_chan_resources = fsl_dma_free_chan_resources;
-	fdev->common.device_prep_dma_interrupt = fsl_dma_prep_interrupt;
 	fdev->common.device_prep_dma_memcpy = fsl_dma_prep_memcpy;
 	fdev->common.device_prep_dma_sg = fsl_dma_prep_sg;
 	fdev->common.device_tx_status = fsl_tx_status;
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index 19041cefabb1..14867e3ac8ff 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -255,6 +255,7 @@ struct sdma_channel {
 	enum dma_slave_buswidth		word_size;
 	unsigned int			buf_tail;
 	unsigned int			num_bd;
+	unsigned int			period_len;
 	struct sdma_buffer_descriptor	*bd;
 	dma_addr_t			bd_phys;
 	unsigned int			pc_from_device, pc_to_device;
@@ -593,6 +594,12 @@ static void sdma_event_disable(struct sdma_channel *sdmac, unsigned int event)
 
 static void sdma_handle_channel_loop(struct sdma_channel *sdmac)
 {
+	if (sdmac->desc.callback)
+		sdmac->desc.callback(sdmac->desc.callback_param);
+}
+
+static void sdma_update_channel_loop(struct sdma_channel *sdmac)
+{
 	struct sdma_buffer_descriptor *bd;
 
 	/*
@@ -607,15 +614,10 @@ static void sdma_handle_channel_loop(struct sdma_channel *sdmac)
 
 		if (bd->mode.status & BD_RROR)
 			sdmac->status = DMA_ERROR;
-		else
-			sdmac->status = DMA_IN_PROGRESS;
 
 		bd->mode.status |= BD_DONE;
 		sdmac->buf_tail++;
 		sdmac->buf_tail %= sdmac->num_bd;
-
-		if (sdmac->desc.callback)
-			sdmac->desc.callback(sdmac->desc.callback_param);
 	}
 }
 
@@ -671,6 +673,9 @@ static irqreturn_t sdma_int_handler(int irq, void *dev_id)
 		int channel = fls(stat) - 1;
 		struct sdma_channel *sdmac = &sdma->channel[channel];
 
+		if (sdmac->flags & IMX_DMA_SG_LOOP)
+			sdma_update_channel_loop(sdmac);
+
 		tasklet_schedule(&sdmac->tasklet);
 
 		__clear_bit(channel, &stat);
@@ -1131,6 +1136,7 @@ static struct dma_async_tx_descriptor *sdma_prep_dma_cyclic(
 	sdmac->status = DMA_IN_PROGRESS;
 
 	sdmac->buf_tail = 0;
+	sdmac->period_len = period_len;
 
 	sdmac->flags |= IMX_DMA_SG_LOOP;
 	sdmac->direction = direction;
@@ -1227,9 +1233,15 @@ static enum dma_status sdma_tx_status(struct dma_chan *chan,
 				      struct dma_tx_state *txstate)
 {
 	struct sdma_channel *sdmac = to_sdma_chan(chan);
+	u32 residue;
+
+	if (sdmac->flags & IMX_DMA_SG_LOOP)
+		residue = (sdmac->num_bd - sdmac->buf_tail) * sdmac->period_len;
+	else
+		residue = sdmac->chn_count - sdmac->chn_real_count;
 
 	dma_set_tx_state(txstate, chan->completed_cookie, chan->cookie,
-			sdmac->chn_count - sdmac->chn_real_count);
+			 residue);
 
 	return sdmac->status;
 }
diff --git a/drivers/dma/mmp_pdma.c b/drivers/dma/mmp_pdma.c
index bf02e7beb51a..a7b186d536b3 100644
--- a/drivers/dma/mmp_pdma.c
+++ b/drivers/dma/mmp_pdma.c
@@ -29,8 +29,8 @@
 #define DALGN		0x00a0
 #define DINT		0x00f0
 #define DDADR		0x0200
-#define DSADR		0x0204
-#define DTADR		0x0208
+#define DSADR(n)	(0x0204 + ((n) << 4))
+#define DTADR(n)	(0x0208 + ((n) << 4))
 #define DCMD		0x020c
 
 #define DCSR_RUN	BIT(31)	/* Run Bit (read / write) */
@@ -277,7 +277,7 @@ static void mmp_pdma_free_phy(struct mmp_pdma_chan *pchan)
 		return;
 
 	/* clear the channel mapping in DRCMR */
-	reg = DRCMR(pchan->phy->vchan->drcmr);
+	reg = DRCMR(pchan->drcmr);
 	writel(0, pchan->phy->base + reg);
 
 	spin_lock_irqsave(&pdev->phy_lock, flags);
@@ -748,11 +748,92 @@ static int mmp_pdma_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
 	return 0;
 }
 
+static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan,
+				     dma_cookie_t cookie)
+{
+	struct mmp_pdma_desc_sw *sw;
+	u32 curr, residue = 0;
+	bool passed = false;
+	bool cyclic = chan->cyclic_first != NULL;
+
+	/*
+	 * If the channel does not have a phy pointer anymore, it has already
+	 * been completed. Therefore, its residue is 0.
+	 */
+	if (!chan->phy)
+		return 0;
+
+	if (chan->dir == DMA_DEV_TO_MEM)
+		curr = readl(chan->phy->base + DTADR(chan->phy->idx));
+	else
+		curr = readl(chan->phy->base + DSADR(chan->phy->idx));
+
+	list_for_each_entry(sw, &chan->chain_running, node) {
+		u32 start, end, len;
+
+		if (chan->dir == DMA_DEV_TO_MEM)
+			start = sw->desc.dtadr;
+		else
+			start = sw->desc.dsadr;
+
+		len = sw->desc.dcmd & DCMD_LENGTH;
+		end = start + len;
+
+		/*
+		 * 'passed' will be latched once we found the descriptor which
+		 * lies inside the boundaries of the curr pointer. All
+		 * descriptors that occur in the list _after_ we found that
+		 * partially handled descriptor are still to be processed and
+		 * are hence added to the residual bytes counter.
+		 */
+
+		if (passed) {
+			residue += len;
+		} else if (curr >= start && curr <= end) {
+			residue += end - curr;
+			passed = true;
+		}
+
+		/*
+		 * Descriptors that have the ENDIRQEN bit set mark the end of a
+		 * transaction chain, and the cookie assigned with it has been
+		 * returned previously from mmp_pdma_tx_submit().
+		 *
+		 * In case we have multiple transactions in the running chain,
+		 * and the cookie does not match the one the user asked us
+		 * about, reset the state variables and start over.
+		 *
+		 * This logic does not apply to cyclic transactions, where all
+		 * descriptors have the ENDIRQEN bit set, and for which we
+		 * can't have multiple transactions on one channel anyway.
+		 */
+		if (cyclic || !(sw->desc.dcmd & DCMD_ENDIRQEN))
+			continue;
+
+		if (sw->async_tx.cookie == cookie) {
+			return residue;
+		} else {
+			residue = 0;
+			passed = false;
+		}
+	}
+
+	/* We should only get here in case of cyclic transactions */
+	return residue;
+}
+
 static enum dma_status mmp_pdma_tx_status(struct dma_chan *dchan,
 					  dma_cookie_t cookie,
 					  struct dma_tx_state *txstate)
 {
-	return dma_cookie_status(dchan, cookie, txstate);
+	struct mmp_pdma_chan *chan = to_mmp_pdma_chan(dchan);
+	enum dma_status ret;
+
+	ret = dma_cookie_status(dchan, cookie, txstate);
+	if (likely(ret != DMA_ERROR))
+		dma_set_residue(txstate, mmp_pdma_residue(chan, cookie));
+
+	return ret;
 }
 
 /**
@@ -858,8 +939,7 @@ static int mmp_pdma_chan_init(struct mmp_pdma_device *pdev, int idx, int irq)
 	struct mmp_pdma_chan *chan;
 	int ret;
 
-	chan = devm_kzalloc(pdev->dev, sizeof(struct mmp_pdma_chan),
-			    GFP_KERNEL);
+	chan = devm_kzalloc(pdev->dev, sizeof(*chan), GFP_KERNEL);
 	if (chan == NULL)
 		return -ENOMEM;
 
@@ -946,8 +1026,7 @@ static int mmp_pdma_probe(struct platform_device *op)
 			irq_num++;
 	}
 
-	pdev->phy = devm_kcalloc(pdev->dev,
-				 dma_channels, sizeof(struct mmp_pdma_chan),
+	pdev->phy = devm_kcalloc(pdev->dev, dma_channels, sizeof(*pdev->phy),
 				 GFP_KERNEL);
 	if (pdev->phy == NULL)
 		return -ENOMEM;
diff --git a/drivers/dma/mpc512x_dma.c b/drivers/dma/mpc512x_dma.c
index 448750da4402..2ad43738ac8b 100644
--- a/drivers/dma/mpc512x_dma.c
+++ b/drivers/dma/mpc512x_dma.c
@@ -2,6 +2,7 @@
  * Copyright (C) Freescale Semicondutor, Inc. 2007, 2008.
  * Copyright (C) Semihalf 2009
  * Copyright (C) Ilya Yanok, Emcraft Systems 2010
+ * Copyright (C) Alexander Popov, Promcontroller 2014
  *
  * Written by Piotr Ziecik <kosmo@semihalf.com>. Hardware description
  * (defines, structures and comments) was taken from MPC5121 DMA driver
@@ -29,8 +30,18 @@
  */
 
 /*
- * This is initial version of MPC5121 DMA driver. Only memory to memory
- * transfers are supported (tested using dmatest module).
+ * MPC512x and MPC8308 DMA driver. It supports
+ * memory to memory data transfers (tested using dmatest module) and
+ * data transfers between memory and peripheral I/O memory
+ * by means of slave scatter/gather with these limitations:
+ *  - chunked transfers (described by s/g lists with more than one item)
+ *     are refused as long as proper support for scatter/gather is missing;
+ *  - transfers on MPC8308 always start from software as this SoC appears
+ *     not to have external request lines for peripheral flow control;
+ *  - only peripheral devices with 4-byte FIFO access register are supported;
+ *  - minimal memory <-> I/O memory transfer chunk is 4 bytes and consequently
+ *     source and destination addresses must be 4-byte aligned
+ *     and transfer size must be aligned on (4 * maxburst) boundary;
  */
 
 #include <linux/module.h>
@@ -52,9 +63,17 @@
 #define MPC_DMA_DESCRIPTORS	64
 
 /* Macro definitions */
-#define MPC_DMA_CHANNELS	64
 #define MPC_DMA_TCD_OFFSET	0x1000
 
+/*
+ * Maximum channel counts for individual hardware variants
+ * and the maximum channel count over all supported controllers,
+ * used for data structure size
+ */
+#define MPC8308_DMACHAN_MAX	16
+#define MPC512x_DMACHAN_MAX	64
+#define MPC_DMA_CHANNELS	64
+
 /* Arbitration mode of group and channel */
 #define MPC_DMA_DMACR_EDCG	(1 << 31)
 #define MPC_DMA_DMACR_ERGA	(1 << 3)
@@ -181,6 +200,7 @@ struct mpc_dma_desc {
 	dma_addr_t			tcd_paddr;
 	int				error;
 	struct list_head		node;
+	int				will_access_peripheral;
 };
 
 struct mpc_dma_chan {
@@ -193,6 +213,12 @@ struct mpc_dma_chan {
 	struct mpc_dma_tcd		*tcd;
 	dma_addr_t			tcd_paddr;
 
+	/* Settings for access to peripheral FIFO */
+	dma_addr_t			src_per_paddr;
+	u32				src_tcd_nunits;
+	dma_addr_t			dst_per_paddr;
+	u32				dst_tcd_nunits;
+
 	/* Lock for this structure */
 	spinlock_t			lock;
 };
@@ -243,8 +269,23 @@ static void mpc_dma_execute(struct mpc_dma_chan *mchan)
 	struct mpc_dma_desc *mdesc;
 	int cid = mchan->chan.chan_id;
 
-	/* Move all queued descriptors to active list */
-	list_splice_tail_init(&mchan->queued, &mchan->active);
+	while (!list_empty(&mchan->queued)) {
+		mdesc = list_first_entry(&mchan->queued,
+						struct mpc_dma_desc, node);
+		/*
+		 * Grab either several mem-to-mem transfer descriptors
+		 * or one peripheral transfer descriptor,
+		 * don't mix mem-to-mem and peripheral transfer descriptors
+		 * within the same 'active' list.
+		 */
+		if (mdesc->will_access_peripheral) {
+			if (list_empty(&mchan->active))
+				list_move_tail(&mdesc->node, &mchan->active);
+			break;
+		} else {
+			list_move_tail(&mdesc->node, &mchan->active);
+		}
+	}
 
 	/* Chain descriptors into one transaction */
 	list_for_each_entry(mdesc, &mchan->active, node) {
@@ -270,7 +311,17 @@ static void mpc_dma_execute(struct mpc_dma_chan *mchan)
 
 	if (first != prev)
 		mdma->tcd[cid].e_sg = 1;
-	out_8(&mdma->regs->dmassrt, cid);
+
+	if (mdma->is_mpc8308) {
+		/* MPC8308, no request lines, software initiated start */
+		out_8(&mdma->regs->dmassrt, cid);
+	} else if (first->will_access_peripheral) {
+		/* Peripherals involved, start by external request signal */
+		out_8(&mdma->regs->dmaserq, cid);
+	} else {
+		/* Memory to memory transfer, software initiated start */
+		out_8(&mdma->regs->dmassrt, cid);
+	}
 }
 
 /* Handle interrupt on one half of DMA controller (32 channels) */
@@ -588,6 +639,7 @@ mpc_dma_prep_memcpy(struct dma_chan *chan, dma_addr_t dst, dma_addr_t src,
 	}
 
 	mdesc->error = 0;
+	mdesc->will_access_peripheral = 0;
 	tcd = mdesc->tcd;
 
 	/* Prepare Transfer Control Descriptor for this transaction */
@@ -635,6 +687,193 @@ mpc_dma_prep_memcpy(struct dma_chan *chan, dma_addr_t dst, dma_addr_t src,
 	return &mdesc->desc;
 }
 
+static struct dma_async_tx_descriptor *
+mpc_dma_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
+		unsigned int sg_len, enum dma_transfer_direction direction,
+		unsigned long flags, void *context)
+{
+	struct mpc_dma *mdma = dma_chan_to_mpc_dma(chan);
+	struct mpc_dma_chan *mchan = dma_chan_to_mpc_dma_chan(chan);
+	struct mpc_dma_desc *mdesc = NULL;
+	dma_addr_t per_paddr;
+	u32 tcd_nunits;
+	struct mpc_dma_tcd *tcd;
+	unsigned long iflags;
+	struct scatterlist *sg;
+	size_t len;
+	int iter, i;
+
+	/* Currently there is no proper support for scatter/gather */
+	if (sg_len != 1)
+		return NULL;
+
+	if (!is_slave_direction(direction))
+		return NULL;
+
+	for_each_sg(sgl, sg, sg_len, i) {
+		spin_lock_irqsave(&mchan->lock, iflags);
+
+		mdesc = list_first_entry(&mchan->free,
+						struct mpc_dma_desc, node);
+		if (!mdesc) {
+			spin_unlock_irqrestore(&mchan->lock, iflags);
+			/* Try to free completed descriptors */
+			mpc_dma_process_completed(mdma);
+			return NULL;
+		}
+
+		list_del(&mdesc->node);
+
+		if (direction == DMA_DEV_TO_MEM) {
+			per_paddr = mchan->src_per_paddr;
+			tcd_nunits = mchan->src_tcd_nunits;
+		} else {
+			per_paddr = mchan->dst_per_paddr;
+			tcd_nunits = mchan->dst_tcd_nunits;
+		}
+
+		spin_unlock_irqrestore(&mchan->lock, iflags);
+
+		if (per_paddr == 0 || tcd_nunits == 0)
+			goto err_prep;
+
+		mdesc->error = 0;
+		mdesc->will_access_peripheral = 1;
+
+		/* Prepare Transfer Control Descriptor for this transaction */
+		tcd = mdesc->tcd;
+
+		memset(tcd, 0, sizeof(struct mpc_dma_tcd));
+
+		if (!IS_ALIGNED(sg_dma_address(sg), 4))
+			goto err_prep;
+
+		if (direction == DMA_DEV_TO_MEM) {
+			tcd->saddr = per_paddr;
+			tcd->daddr = sg_dma_address(sg);
+			tcd->soff = 0;
+			tcd->doff = 4;
+		} else {
+			tcd->saddr = sg_dma_address(sg);
+			tcd->daddr = per_paddr;
+			tcd->soff = 4;
+			tcd->doff = 0;
+		}
+
+		tcd->ssize = MPC_DMA_TSIZE_4;
+		tcd->dsize = MPC_DMA_TSIZE_4;
+
+		len = sg_dma_len(sg);
+		tcd->nbytes = tcd_nunits * 4;
+		if (!IS_ALIGNED(len, tcd->nbytes))
+			goto err_prep;
+
+		iter = len / tcd->nbytes;
+		if (iter >= 1 << 15) {
+			/* len is too big */
+			goto err_prep;
+		}
+		/* citer_linkch contains the high bits of iter */
+		tcd->biter = iter & 0x1ff;
+		tcd->biter_linkch = iter >> 9;
+		tcd->citer = tcd->biter;
+		tcd->citer_linkch = tcd->biter_linkch;
+
+		tcd->e_sg = 0;
+		tcd->d_req = 1;
+
+		/* Place descriptor in prepared list */
+		spin_lock_irqsave(&mchan->lock, iflags);
+		list_add_tail(&mdesc->node, &mchan->prepared);
+		spin_unlock_irqrestore(&mchan->lock, iflags);
+	}
+
+	return &mdesc->desc;
+
+err_prep:
+	/* Put the descriptor back */
+	spin_lock_irqsave(&mchan->lock, iflags);
+	list_add_tail(&mdesc->node, &mchan->free);
+	spin_unlock_irqrestore(&mchan->lock, iflags);
+
+	return NULL;
+}
+
+static int mpc_dma_device_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
+							unsigned long arg)
+{
+	struct mpc_dma_chan *mchan;
+	struct mpc_dma *mdma;
+	struct dma_slave_config *cfg;
+	unsigned long flags;
+
+	mchan = dma_chan_to_mpc_dma_chan(chan);
+	switch (cmd) {
+	case DMA_TERMINATE_ALL:
+		/* Disable channel requests */
+		mdma = dma_chan_to_mpc_dma(chan);
+
+		spin_lock_irqsave(&mchan->lock, flags);
+
+		out_8(&mdma->regs->dmacerq, chan->chan_id);
+		list_splice_tail_init(&mchan->prepared, &mchan->free);
+		list_splice_tail_init(&mchan->queued, &mchan->free);
+		list_splice_tail_init(&mchan->active, &mchan->free);
+
+		spin_unlock_irqrestore(&mchan->lock, flags);
+
+		return 0;
+
+	case DMA_SLAVE_CONFIG:
+		/*
+		 * Software constraints:
+		 *  - only transfers between a peripheral device and
+		 *     memory are supported;
+		 *  - only peripheral devices with 4-byte FIFO access register
+		 *     are supported;
+		 *  - minimal transfer chunk is 4 bytes and consequently
+		 *     source and destination addresses must be 4-byte aligned
+		 *     and transfer size must be aligned on (4 * maxburst)
+		 *     boundary;
+		 *  - during the transfer RAM address is being incremented by
+		 *     the size of minimal transfer chunk;
+		 *  - peripheral port's address is constant during the transfer.
+		 */
+
+		cfg = (void *)arg;
+
+		if (cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
+		    cfg->dst_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
+		    !IS_ALIGNED(cfg->src_addr, 4) ||
+		    !IS_ALIGNED(cfg->dst_addr, 4)) {
+			return -EINVAL;
+		}
+
+		spin_lock_irqsave(&mchan->lock, flags);
+
+		mchan->src_per_paddr = cfg->src_addr;
+		mchan->src_tcd_nunits = cfg->src_maxburst;
+		mchan->dst_per_paddr = cfg->dst_addr;
+		mchan->dst_tcd_nunits = cfg->dst_maxburst;
+
+		/* Apply defaults */
+		if (mchan->src_tcd_nunits == 0)
+			mchan->src_tcd_nunits = 1;
+		if (mchan->dst_tcd_nunits == 0)
+			mchan->dst_tcd_nunits = 1;
+
+		spin_unlock_irqrestore(&mchan->lock, flags);
+
+		return 0;
+
+	default:
+		/* Unknown command */
+		break;
+	}
+
+	return -ENXIO;
+}
+
 static int mpc_dma_probe(struct platform_device *op)
 {
 	struct device_node *dn = op->dev.of_node;
@@ -649,13 +888,15 @@ static int mpc_dma_probe(struct platform_device *op)
 	mdma = devm_kzalloc(dev, sizeof(struct mpc_dma), GFP_KERNEL);
 	if (!mdma) {
 		dev_err(dev, "Memory exhausted!\n");
-		return -ENOMEM;
+		retval = -ENOMEM;
+		goto err;
 	}
 
 	mdma->irq = irq_of_parse_and_map(dn, 0);
 	if (mdma->irq == NO_IRQ) {
 		dev_err(dev, "Error mapping IRQ!\n");
-		return -EINVAL;
+		retval = -EINVAL;
+		goto err;
 	}
 
 	if (of_device_is_compatible(dn, "fsl,mpc8308-dma")) {
@@ -663,14 +904,15 @@ static int mpc_dma_probe(struct platform_device *op)
 		mdma->irq2 = irq_of_parse_and_map(dn, 1);
 		if (mdma->irq2 == NO_IRQ) {
 			dev_err(dev, "Error mapping IRQ!\n");
-			return -EINVAL;
+			retval = -EINVAL;
+			goto err_dispose1;
 		}
 	}
 
 	retval = of_address_to_resource(dn, 0, &res);
 	if (retval) {
 		dev_err(dev, "Error parsing memory region!\n");
-		return retval;
+		goto err_dispose2;
 	}
 
 	regs_start = res.start;
@@ -678,31 +920,34 @@ static int mpc_dma_probe(struct platform_device *op)
 
 	if (!devm_request_mem_region(dev, regs_start, regs_size, DRV_NAME)) {
 		dev_err(dev, "Error requesting memory region!\n");
-		return -EBUSY;
+		retval = -EBUSY;
+		goto err_dispose2;
 	}
 
 	mdma->regs = devm_ioremap(dev, regs_start, regs_size);
 	if (!mdma->regs) {
 		dev_err(dev, "Error mapping memory region!\n");
-		return -ENOMEM;
+		retval = -ENOMEM;
+		goto err_dispose2;
 	}
 
 	mdma->tcd = (struct mpc_dma_tcd *)((u8 *)(mdma->regs)
 							+ MPC_DMA_TCD_OFFSET);
 
-	retval = devm_request_irq(dev, mdma->irq, &mpc_dma_irq, 0, DRV_NAME,
-									mdma);
+	retval = request_irq(mdma->irq, &mpc_dma_irq, 0, DRV_NAME, mdma);
 	if (retval) {
 		dev_err(dev, "Error requesting IRQ!\n");
-		return -EINVAL;
+		retval = -EINVAL;
+		goto err_dispose2;
 	}
 
 	if (mdma->is_mpc8308) {
-		retval = devm_request_irq(dev, mdma->irq2, &mpc_dma_irq, 0,
-				DRV_NAME, mdma);
+		retval = request_irq(mdma->irq2, &mpc_dma_irq, 0,
+							DRV_NAME, mdma);
 		if (retval) {
 			dev_err(dev, "Error requesting IRQ2!\n");
-			return -EINVAL;
+			retval = -EINVAL;
+			goto err_free1;
 		}
 	}
 
@@ -710,18 +955,21 @@ static int mpc_dma_probe(struct platform_device *op)
 
 	dma = &mdma->dma;
 	dma->dev = dev;
-	if (!mdma->is_mpc8308)
-		dma->chancnt = MPC_DMA_CHANNELS;
+	if (mdma->is_mpc8308)
+		dma->chancnt = MPC8308_DMACHAN_MAX;
 	else
-		dma->chancnt = 16; /* MPC8308 DMA has only 16 channels */
+		dma->chancnt = MPC512x_DMACHAN_MAX;
 	dma->device_alloc_chan_resources = mpc_dma_alloc_chan_resources;
 	dma->device_free_chan_resources = mpc_dma_free_chan_resources;
 	dma->device_issue_pending = mpc_dma_issue_pending;
 	dma->device_tx_status = mpc_dma_tx_status;
 	dma->device_prep_dma_memcpy = mpc_dma_prep_memcpy;
+	dma->device_prep_slave_sg = mpc_dma_prep_slave_sg;
+	dma->device_control = mpc_dma_device_control;
 
 	INIT_LIST_HEAD(&dma->channels);
 	dma_cap_set(DMA_MEMCPY, dma->cap_mask);
+	dma_cap_set(DMA_SLAVE, dma->cap_mask);
 
 	for (i = 0; i < dma->chancnt; i++) {
 		mchan = &mdma->channels[i];
@@ -747,7 +995,19 @@ static int mpc_dma_probe(struct platform_device *op)
 	 * - Round-robin group arbitration,
 	 * - Round-robin channel arbitration.
 	 */
-	if (!mdma->is_mpc8308) {
+	if (mdma->is_mpc8308) {
+		/* MPC8308 has 16 channels and lacks some registers */
+		out_be32(&mdma->regs->dmacr, MPC_DMA_DMACR_ERCA);
+
+		/* enable snooping */
+		out_be32(&mdma->regs->dmagpor, MPC_DMA_DMAGPOR_SNOOP_ENABLE);
+		/* Disable error interrupts */
+		out_be32(&mdma->regs->dmaeeil, 0);
+
+		/* Clear interrupts status */
+		out_be32(&mdma->regs->dmaintl, 0xFFFF);
+		out_be32(&mdma->regs->dmaerrl, 0xFFFF);
+	} else {
 		out_be32(&mdma->regs->dmacr, MPC_DMA_DMACR_EDCG |
 					MPC_DMA_DMACR_ERGA | MPC_DMA_DMACR_ERCA);
 
@@ -768,29 +1028,28 @@ static int mpc_dma_probe(struct platform_device *op)
 		/* Route interrupts to IPIC */
 		out_be32(&mdma->regs->dmaihsa, 0);
 		out_be32(&mdma->regs->dmailsa, 0);
-	} else {
-		/* MPC8308 has 16 channels and lacks some registers */
-		out_be32(&mdma->regs->dmacr, MPC_DMA_DMACR_ERCA);
-
-		/* enable snooping */
-		out_be32(&mdma->regs->dmagpor, MPC_DMA_DMAGPOR_SNOOP_ENABLE);
-		/* Disable error interrupts */
-		out_be32(&mdma->regs->dmaeeil, 0);
-
-		/* Clear interrupts status */
-		out_be32(&mdma->regs->dmaintl, 0xFFFF);
-		out_be32(&mdma->regs->dmaerrl, 0xFFFF);
 	}
 
 	/* Register DMA engine */
 	dev_set_drvdata(dev, mdma);
 	retval = dma_async_device_register(dma);
-	if (retval) {
-		devm_free_irq(dev, mdma->irq, mdma);
-		irq_dispose_mapping(mdma->irq);
-	}
+	if (retval)
+		goto err_free2;
 
 	return retval;
+
+err_free2:
+	if (mdma->is_mpc8308)
+		free_irq(mdma->irq2, mdma);
+err_free1:
+	free_irq(mdma->irq, mdma);
+err_dispose2:
+	if (mdma->is_mpc8308)
+		irq_dispose_mapping(mdma->irq2);
+err_dispose1:
+	irq_dispose_mapping(mdma->irq);
+err:
+	return retval;
 }
 
 static int mpc_dma_remove(struct platform_device *op)
@@ -799,7 +1058,11 @@ static int mpc_dma_remove(struct platform_device *op)
 	struct mpc_dma *mdma = dev_get_drvdata(dev);
 
 	dma_async_device_unregister(&mdma->dma);
-	devm_free_irq(dev, mdma->irq, mdma);
+	if (mdma->is_mpc8308) {
+		free_irq(mdma->irq2, mdma);
+		irq_dispose_mapping(mdma->irq2);
+	}
+	free_irq(mdma->irq, mdma);
 	irq_dispose_mapping(mdma->irq);
 
 	return 0;
@@ -807,6 +1070,7 @@ static int mpc_dma_remove(struct platform_device *op)
 
 static struct of_device_id mpc_dma_match[] = {
 	{ .compatible = "fsl,mpc5121-dma", },
+	{ .compatible = "fsl,mpc8308-dma", },
 	{},
 };
 
diff --git a/drivers/dma/pch_dma.c b/drivers/dma/pch_dma.c
index 05fa548bd659..9f9ca9fe5ce6 100644
--- a/drivers/dma/pch_dma.c
+++ b/drivers/dma/pch_dma.c
@@ -21,6 +21,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/init.h>
 #include <linux/pci.h>
+#include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <linux/pch_dma.h>
@@ -996,7 +997,7 @@ static void pch_dma_remove(struct pci_dev *pdev)
 #define PCI_DEVICE_ID_ML7831_DMA1_8CH	0x8810
 #define PCI_DEVICE_ID_ML7831_DMA2_4CH	0x8815
 
-DEFINE_PCI_DEVICE_TABLE(pch_dma_id_table) = {
+const struct pci_device_id pch_dma_id_table[] = {
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_EG20T_PCH_DMA_8CH), 8 },
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_EG20T_PCH_DMA_4CH), 4 },
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7213_DMA1_8CH), 8}, /* UART Video */
diff --git a/drivers/dma/s3c24xx-dma.c b/drivers/dma/s3c24xx-dma.c
index b209a0f17344..012520c9fd79 100644
--- a/drivers/dma/s3c24xx-dma.c
+++ b/drivers/dma/s3c24xx-dma.c
@@ -164,6 +164,7 @@ struct s3c24xx_sg {
  * @disrcc: value for source control register
  * @didstc: value for destination control register
  * @dcon: base value for dcon register
+ * @cyclic: indicate cyclic transfer
  */
 struct s3c24xx_txd {
 	struct virt_dma_desc vd;
@@ -173,6 +174,7 @@ struct s3c24xx_txd {
 	u32 disrcc;
 	u32 didstc;
 	u32 dcon;
+	bool cyclic;
 };
 
 struct s3c24xx_dma_chan;
@@ -669,8 +671,10 @@ static irqreturn_t s3c24xx_dma_irq(int irq, void *data)
 		/* when more sg's are in this txd, start the next one */
 		if (!list_is_last(txd->at, &txd->dsg_list)) {
 			txd->at = txd->at->next;
+			if (txd->cyclic)
+				vchan_cyclic_callback(&txd->vd);
 			s3c24xx_dma_start_next_sg(s3cchan, txd);
-		} else {
+		} else if (!txd->cyclic) {
 			s3cchan->at = NULL;
 			vchan_cookie_complete(&txd->vd);
 
@@ -682,6 +686,12 @@ static irqreturn_t s3c24xx_dma_irq(int irq, void *data)
 				s3c24xx_dma_start_next_txd(s3cchan);
 			else
 				s3c24xx_dma_phy_free(s3cchan);
+		} else {
+			vchan_cyclic_callback(&txd->vd);
+
+			/* Cyclic: reset at beginning */
+			txd->at = txd->dsg_list.next;
+			s3c24xx_dma_start_next_sg(s3cchan, txd);
 		}
 	}
 	spin_unlock(&s3cchan->vc.lock);
@@ -877,6 +887,104 @@ static struct dma_async_tx_descriptor *s3c24xx_dma_prep_memcpy(
 	return vchan_tx_prep(&s3cchan->vc, &txd->vd, flags);
 }
 
+static struct dma_async_tx_descriptor *s3c24xx_dma_prep_dma_cyclic(
+	struct dma_chan *chan, dma_addr_t addr, size_t size, size_t period,
+	enum dma_transfer_direction direction, unsigned long flags,
+	void *context)
+{
+	struct s3c24xx_dma_chan *s3cchan = to_s3c24xx_dma_chan(chan);
+	struct s3c24xx_dma_engine *s3cdma = s3cchan->host;
+	const struct s3c24xx_dma_platdata *pdata = s3cdma->pdata;
+	struct s3c24xx_dma_channel *cdata = &pdata->channels[s3cchan->id];
+	struct s3c24xx_txd *txd;
+	struct s3c24xx_sg *dsg;
+	unsigned sg_len;
+	dma_addr_t slave_addr;
+	u32 hwcfg = 0;
+	int i;
+
+	dev_dbg(&s3cdma->pdev->dev,
+		"prepare cyclic transaction of %zu bytes with period %zu from %s\n",
+		size, period, s3cchan->name);
+
+	if (!is_slave_direction(direction)) {
+		dev_err(&s3cdma->pdev->dev,
+			"direction %d unsupported\n", direction);
+		return NULL;
+	}
+
+	txd = s3c24xx_dma_get_txd();
+	if (!txd)
+		return NULL;
+
+	txd->cyclic = 1;
+
+	if (cdata->handshake)
+		txd->dcon |= S3C24XX_DCON_HANDSHAKE;
+
+	switch (cdata->bus) {
+	case S3C24XX_DMA_APB:
+		txd->dcon |= S3C24XX_DCON_SYNC_PCLK;
+		hwcfg |= S3C24XX_DISRCC_LOC_APB;
+		break;
+	case S3C24XX_DMA_AHB:
+		txd->dcon |= S3C24XX_DCON_SYNC_HCLK;
+		hwcfg |= S3C24XX_DISRCC_LOC_AHB;
+		break;
+	}
+
+	/*
+	 * Always assume our peripheral desintation is a fixed
+	 * address in memory.
+	 */
+	hwcfg |= S3C24XX_DISRCC_INC_FIXED;
+
+	/*
+	 * Individual dma operations are requested by the slave,
+	 * so serve only single atomic operations (S3C24XX_DCON_SERV_SINGLE).
+	 */
+	txd->dcon |= S3C24XX_DCON_SERV_SINGLE;
+
+	if (direction == DMA_MEM_TO_DEV) {
+		txd->disrcc = S3C24XX_DISRCC_LOC_AHB |
+			      S3C24XX_DISRCC_INC_INCREMENT;
+		txd->didstc = hwcfg;
+		slave_addr = s3cchan->cfg.dst_addr;
+		txd->width = s3cchan->cfg.dst_addr_width;
+	} else {
+		txd->disrcc = hwcfg;
+		txd->didstc = S3C24XX_DIDSTC_LOC_AHB |
+			      S3C24XX_DIDSTC_INC_INCREMENT;
+		slave_addr = s3cchan->cfg.src_addr;
+		txd->width = s3cchan->cfg.src_addr_width;
+	}
+
+	sg_len = size / period;
+
+	for (i = 0; i < sg_len; i++) {
+		dsg = kzalloc(sizeof(*dsg), GFP_NOWAIT);
+		if (!dsg) {
+			s3c24xx_dma_free_txd(txd);
+			return NULL;
+		}
+		list_add_tail(&dsg->node, &txd->dsg_list);
+
+		dsg->len = period;
+		/* Check last period length */
+		if (i == sg_len - 1)
+			dsg->len = size - period * i;
+		if (direction == DMA_MEM_TO_DEV) {
+			dsg->src_addr = addr + period * i;
+			dsg->dst_addr = slave_addr;
+		} else { /* DMA_DEV_TO_MEM */
+			dsg->src_addr = slave_addr;
+			dsg->dst_addr = addr + period * i;
+		}
+	}
+
+	return vchan_tx_prep(&s3cchan->vc, &txd->vd, flags);
+}
+
 static struct dma_async_tx_descriptor *s3c24xx_dma_prep_slave_sg(
 		struct dma_chan *chan, struct scatterlist *sgl,
 		unsigned int sg_len, enum dma_transfer_direction direction,
@@ -961,7 +1069,6 @@ static struct dma_async_tx_descriptor *s3c24xx_dma_prep_slave_sg(
 			dsg->src_addr = slave_addr;
 			dsg->dst_addr = sg_dma_address(sg);
 		}
-		break;
 	}
 
 	return vchan_tx_prep(&s3cchan->vc, &txd->vd, flags);
@@ -1198,6 +1305,7 @@ static int s3c24xx_dma_probe(struct platform_device *pdev)
 
 	/* Initialize slave engine for SoC internal dedicated peripherals */
 	dma_cap_set(DMA_SLAVE, s3cdma->slave.cap_mask);
+	dma_cap_set(DMA_CYCLIC, s3cdma->slave.cap_mask);
 	dma_cap_set(DMA_PRIVATE, s3cdma->slave.cap_mask);
 	s3cdma->slave.dev = &pdev->dev;
 	s3cdma->slave.device_alloc_chan_resources =
@@ -1207,6 +1315,7 @@ static int s3c24xx_dma_probe(struct platform_device *pdev)
 	s3cdma->slave.device_tx_status = s3c24xx_dma_tx_status;
 	s3cdma->slave.device_issue_pending = s3c24xx_dma_issue_pending;
 	s3cdma->slave.device_prep_slave_sg = s3c24xx_dma_prep_slave_sg;
+	s3cdma->slave.device_prep_dma_cyclic = s3c24xx_dma_prep_dma_cyclic;
 	s3cdma->slave.device_control = s3c24xx_dma_control;
 
 	/* Register as many memcpy channels as there are physical channels */
diff --git a/drivers/dma/sh/Kconfig b/drivers/dma/sh/Kconfig
index b4c813831006..0f719816c91b 100644
--- a/drivers/dma/sh/Kconfig
+++ b/drivers/dma/sh/Kconfig
@@ -4,7 +4,7 @@
 
 config SH_DMAE_BASE
 	bool "Renesas SuperH DMA Engine support"
-	depends on (SUPERH && SH_DMA) || (ARM && ARCH_SHMOBILE)
+	depends on (SUPERH && SH_DMA) || ARCH_SHMOBILE || COMPILE_TEST
 	depends on !SH_DMA_API
 	default y
 	select DMA_ENGINE
diff --git a/drivers/dma/sh/rcar-hpbdma.c b/drivers/dma/sh/rcar-hpbdma.c
index 3083d901a414..b212d9471ab5 100644
--- a/drivers/dma/sh/rcar-hpbdma.c
+++ b/drivers/dma/sh/rcar-hpbdma.c
@@ -18,6 +18,7 @@
 
 #include <linux/dmaengine.h>
 #include <linux/delay.h>
+#include <linux/err.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
diff --git a/drivers/dma/sh/shdma-base.c b/drivers/dma/sh/shdma-base.c
index 52396771acbe..b35007e21e6b 100644
--- a/drivers/dma/sh/shdma-base.c
+++ b/drivers/dma/sh/shdma-base.c
@@ -73,8 +73,7 @@ static void shdma_chan_xfer_ld_queue(struct shdma_chan *schan)
 static dma_cookie_t shdma_tx_submit(struct dma_async_tx_descriptor *tx)
 {
 	struct shdma_desc *chunk, *c, *desc =
-		container_of(tx, struct shdma_desc, async_tx),
-		*last = desc;
+		container_of(tx, struct shdma_desc, async_tx);
 	struct shdma_chan *schan = to_shdma_chan(tx->chan);
 	dma_async_tx_callback callback = tx->callback;
 	dma_cookie_t cookie;
@@ -98,19 +97,20 @@ static dma_cookie_t shdma_tx_submit(struct dma_async_tx_descriptor *tx)
 				      &chunk->node == &schan->ld_free))
 			break;
 		chunk->mark = DESC_SUBMITTED;
-		/* Callback goes to the last chunk */
-		chunk->async_tx.callback = NULL;
+		if (chunk->chunks == 1) {
+			chunk->async_tx.callback = callback;
+			chunk->async_tx.callback_param = tx->callback_param;
+		} else {
+			/* Callback goes to the last chunk */
+			chunk->async_tx.callback = NULL;
+		}
 		chunk->cookie = cookie;
 		list_move_tail(&chunk->node, &schan->ld_queue);
-		last = chunk;
 
 		dev_dbg(schan->dev, "submit #%d@%p on %d\n",
-			tx->cookie, &last->async_tx, schan->id);
+			tx->cookie, &chunk->async_tx, schan->id);
 	}
 
-	last->async_tx.callback = callback;
-	last->async_tx.callback_param = tx->callback_param;
-
 	if (power_up) {
 		int ret;
 		schan->pm_state = SHDMA_PM_BUSY;
@@ -304,6 +304,7 @@ static dma_async_tx_callback __ld_cleanup(struct shdma_chan *schan, bool all)
 	dma_async_tx_callback callback = NULL;
 	void *param = NULL;
 	unsigned long flags;
+	LIST_HEAD(cyclic_list);
 
 	spin_lock_irqsave(&schan->chan_lock, flags);
 	list_for_each_entry_safe(desc, _desc, &schan->ld_queue, node) {
@@ -369,10 +370,16 @@ static dma_async_tx_callback __ld_cleanup(struct shdma_chan *schan, bool all)
 		if (((desc->mark == DESC_COMPLETED ||
 		      desc->mark == DESC_WAITING) &&
 		     async_tx_test_ack(&desc->async_tx)) || all) {
-			/* Remove from ld_queue list */
-			desc->mark = DESC_IDLE;
 
-			list_move(&desc->node, &schan->ld_free);
+			if (all || !desc->cyclic) {
+				/* Remove from ld_queue list */
+				desc->mark = DESC_IDLE;
+				list_move(&desc->node, &schan->ld_free);
+			} else {
+				/* reuse as cyclic */
+				desc->mark = DESC_SUBMITTED;
+				list_move_tail(&desc->node, &cyclic_list);
+			}
 
 			if (list_empty(&schan->ld_queue)) {
 				dev_dbg(schan->dev, "Bring down channel %d\n", schan->id);
@@ -389,6 +396,8 @@ static dma_async_tx_callback __ld_cleanup(struct shdma_chan *schan, bool all)
 		 */
 		schan->dma_chan.completed_cookie = schan->dma_chan.cookie;
 
+	list_splice_tail(&cyclic_list, &schan->ld_queue);
+
 	spin_unlock_irqrestore(&schan->chan_lock, flags);
 
 	if (callback)
@@ -521,7 +530,7 @@ static struct shdma_desc *shdma_add_desc(struct shdma_chan *schan,
  */
 static struct dma_async_tx_descriptor *shdma_prep_sg(struct shdma_chan *schan,
 	struct scatterlist *sgl, unsigned int sg_len, dma_addr_t *addr,
-	enum dma_transfer_direction direction, unsigned long flags)
+	enum dma_transfer_direction direction, unsigned long flags, bool cyclic)
 {
 	struct scatterlist *sg;
 	struct shdma_desc *first = NULL, *new = NULL /* compiler... */;
@@ -569,7 +578,11 @@ static struct dma_async_tx_descriptor *shdma_prep_sg(struct shdma_chan *schan,
 			if (!new)
 				goto err_get_desc;
 
-			new->chunks = chunks--;
+			new->cyclic = cyclic;
+			if (cyclic)
+				new->chunks = 1;
+			else
+				new->chunks = chunks--;
 			list_add_tail(&new->node, &tx_list);
 		} while (len);
 	}
@@ -612,7 +625,8 @@ static struct dma_async_tx_descriptor *shdma_prep_memcpy(
 	sg_dma_address(&sg) = dma_src;
 	sg_dma_len(&sg) = len;
 
-	return shdma_prep_sg(schan, &sg, 1, &dma_dest, DMA_MEM_TO_MEM, flags);
+	return shdma_prep_sg(schan, &sg, 1, &dma_dest, DMA_MEM_TO_MEM,
+			     flags, false);
 }
 
 static struct dma_async_tx_descriptor *shdma_prep_slave_sg(
@@ -640,7 +654,58 @@ static struct dma_async_tx_descriptor *shdma_prep_slave_sg(
 	slave_addr = ops->slave_addr(schan);
 
 	return shdma_prep_sg(schan, sgl, sg_len, &slave_addr,
-			      direction, flags);
+			     direction, flags, false);
+}
+
+#define SHDMA_MAX_SG_LEN 32
+
+static struct dma_async_tx_descriptor *shdma_prep_dma_cyclic(
+	struct dma_chan *chan, dma_addr_t buf_addr, size_t buf_len,
+	size_t period_len, enum dma_transfer_direction direction,
+	unsigned long flags, void *context)
+{
+	struct shdma_chan *schan = to_shdma_chan(chan);
+	struct shdma_dev *sdev = to_shdma_dev(schan->dma_chan.device);
+	const struct shdma_ops *ops = sdev->ops;
+	unsigned int sg_len = buf_len / period_len;
+	int slave_id = schan->slave_id;
+	dma_addr_t slave_addr;
+	struct scatterlist sgl[SHDMA_MAX_SG_LEN];
+	int i;
+
+	if (!chan)
+		return NULL;
+
+	BUG_ON(!schan->desc_num);
+
+	if (sg_len > SHDMA_MAX_SG_LEN) {
+		dev_err(schan->dev, "sg length %d exceds limit %d",
+				sg_len, SHDMA_MAX_SG_LEN);
+		return NULL;
+	}
+
+	/* Someone calling slave DMA on a generic channel? */
+	if (slave_id < 0 || (buf_len < period_len)) {
+		dev_warn(schan->dev,
+			"%s: bad parameter: buf_len=%zu, period_len=%zu, id=%d\n",
+			__func__, buf_len, period_len, slave_id);
+		return NULL;
+	}
+
+	slave_addr = ops->slave_addr(schan);
+
+	sg_init_table(sgl, sg_len);
+	for (i = 0; i < sg_len; i++) {
+		dma_addr_t src = buf_addr + (period_len * i);
+
+		sg_set_page(&sgl[i], pfn_to_page(PFN_DOWN(src)), period_len,
+			    offset_in_page(src));
+		sg_dma_address(&sgl[i]) = src;
+		sg_dma_len(&sgl[i]) = period_len;
+	}
+
+	return shdma_prep_sg(schan, sgl, sg_len, &slave_addr,
+			     direction, flags, true);
 }
 
 static int shdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
@@ -915,6 +980,7 @@ int shdma_init(struct device *dev, struct shdma_dev *sdev,
 
 	/* Compulsory for DMA_SLAVE fields */
 	dma_dev->device_prep_slave_sg = shdma_prep_slave_sg;
+	dma_dev->device_prep_dma_cyclic = shdma_prep_dma_cyclic;
 	dma_dev->device_control = shdma_control;
 
 	dma_dev->dev = dev;
diff --git a/drivers/dma/sh/shdmac.c b/drivers/dma/sh/shdmac.c
index dda7e7563f5d..146d5df926db 100644
--- a/drivers/dma/sh/shdmac.c
+++ b/drivers/dma/sh/shdmac.c
@@ -18,21 +18,22 @@
  *
  */
 
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/err.h>
 #include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kdebug.h>
 #include <linux/module.h>
+#include <linux/notifier.h>
 #include <linux/of.h>
 #include <linux/of_device.h>
-#include <linux/slab.h>
-#include <linux/interrupt.h>
-#include <linux/dmaengine.h>
-#include <linux/delay.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/rculist.h>
 #include <linux/sh_dma.h>
-#include <linux/notifier.h>
-#include <linux/kdebug.h>
+#include <linux/slab.h>
 #include <linux/spinlock.h>
-#include <linux/rculist.h>
 
 #include "../dmaengine.h"
 #include "shdma.h"
diff --git a/drivers/dma/sh/sudmac.c b/drivers/dma/sh/sudmac.c
index 4e7df43b50d6..3ce103909896 100644
--- a/drivers/dma/sh/sudmac.c
+++ b/drivers/dma/sh/sudmac.c
@@ -14,12 +14,13 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/dmaengine.h>
+#include <linux/err.h>
 #include <linux/init.h>
-#include <linux/module.h>
-#include <linux/slab.h>
 #include <linux/interrupt.h>
-#include <linux/dmaengine.h>
+#include <linux/module.h>
 #include <linux/platform_device.h>
+#include <linux/slab.h>
 #include <linux/sudmac.h>
 
 struct sudmac_chan {
diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c
index bf18c786ed40..c7984459ede7 100644
--- a/drivers/dma/ste_dma40.c
+++ b/drivers/dma/ste_dma40.c
@@ -556,7 +556,6 @@ struct d40_gen_dmac {
  * later
  * @reg_val_backup_chan: Backup data for standard channel parameter registers.
  * @gcc_pwr_off_mask: Mask to maintain the channels that can be turned off.
- * @initialized: true if the dma has been initialized
  * @gen_dmac: the struct for generic registers values to represent u8500/8540
  * DMA controller
  */
@@ -594,7 +593,6 @@ struct d40_base {
 	u32				  reg_val_backup_v4[BACKUP_REGS_SZ_MAX];
 	u32				 *reg_val_backup_chan;
 	u16				  gcc_pwr_off_mask;
-	bool				  initialized;
 	struct d40_gen_dmac		  gen_dmac;
 };
 
@@ -1056,62 +1054,6 @@ static int d40_sg_2_dmalen(struct scatterlist *sgl, int sg_len,
 	return len;
 }
 
-
-#ifdef CONFIG_PM
-static void dma40_backup(void __iomem *baseaddr, u32 *backup,
-			 u32 *regaddr, int num, bool save)
-{
-	int i;
-
-	for (i = 0; i < num; i++) {
-		void __iomem *addr = baseaddr + regaddr[i];
-
-		if (save)
-			backup[i] = readl_relaxed(addr);
-		else
-			writel_relaxed(backup[i], addr);
-	}
-}
-
-static void d40_save_restore_registers(struct d40_base *base, bool save)
-{
-	int i;
-
-	/* Save/Restore channel specific registers */
-	for (i = 0; i < base->num_phy_chans; i++) {
-		void __iomem *addr;
-		int idx;
-
-		if (base->phy_res[i].reserved)
-			continue;
-
-		addr = base->virtbase + D40_DREG_PCBASE + i * D40_DREG_PCDELTA;
-		idx = i * ARRAY_SIZE(d40_backup_regs_chan);
-
-		dma40_backup(addr, &base->reg_val_backup_chan[idx],
-			     d40_backup_regs_chan,
-			     ARRAY_SIZE(d40_backup_regs_chan),
-			     save);
-	}
-
-	/* Save/Restore global registers */
-	dma40_backup(base->virtbase, base->reg_val_backup,
-		     d40_backup_regs, ARRAY_SIZE(d40_backup_regs),
-		     save);
-
-	/* Save/Restore registers only existing on dma40 v3 and later */
-	if (base->gen_dmac.backup)
-		dma40_backup(base->virtbase, base->reg_val_backup_v4,
-			     base->gen_dmac.backup,
-			base->gen_dmac.backup_size,
-			save);
-}
-#else
-static void d40_save_restore_registers(struct d40_base *base, bool save)
-{
-}
-#endif
-
 static int __d40_execute_command_phy(struct d40_chan *d40c,
 				     enum d40_command command)
 {
@@ -1495,8 +1437,8 @@ static int d40_pause(struct d40_chan *d40c)
 	if (!d40c->busy)
 		return 0;
 
-	pm_runtime_get_sync(d40c->base->dev);
 	spin_lock_irqsave(&d40c->lock, flags);
+	pm_runtime_get_sync(d40c->base->dev);
 
 	res = d40_channel_execute_command(d40c, D40_DMA_SUSPEND_REQ);
 
@@ -2998,18 +2940,88 @@ failure1:
 }
 
 /* Suspend resume functionality */
-#ifdef CONFIG_PM
-static int dma40_pm_suspend(struct device *dev)
+#ifdef CONFIG_PM_SLEEP
+static int dma40_suspend(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct d40_base *base = platform_get_drvdata(pdev);
-	int ret = 0;
+	int ret;
+
+	ret = pm_runtime_force_suspend(dev);
+	if (ret)
+		return ret;
 
 	if (base->lcpa_regulator)
 		ret = regulator_disable(base->lcpa_regulator);
 	return ret;
 }
 
+static int dma40_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct d40_base *base = platform_get_drvdata(pdev);
+	int ret = 0;
+
+	if (base->lcpa_regulator) {
+		ret = regulator_enable(base->lcpa_regulator);
+		if (ret)
+			return ret;
+	}
+
+	return pm_runtime_force_resume(dev);
+}
+#endif
+
+#ifdef CONFIG_PM
+static void dma40_backup(void __iomem *baseaddr, u32 *backup,
+			 u32 *regaddr, int num, bool save)
+{
+	int i;
+
+	for (i = 0; i < num; i++) {
+		void __iomem *addr = baseaddr + regaddr[i];
+
+		if (save)
+			backup[i] = readl_relaxed(addr);
+		else
+			writel_relaxed(backup[i], addr);
+	}
+}
+
+static void d40_save_restore_registers(struct d40_base *base, bool save)
+{
+	int i;
+
+	/* Save/Restore channel specific registers */
+	for (i = 0; i < base->num_phy_chans; i++) {
+		void __iomem *addr;
+		int idx;
+
+		if (base->phy_res[i].reserved)
+			continue;
+
+		addr = base->virtbase + D40_DREG_PCBASE + i * D40_DREG_PCDELTA;
+		idx = i * ARRAY_SIZE(d40_backup_regs_chan);
+
+		dma40_backup(addr, &base->reg_val_backup_chan[idx],
+			     d40_backup_regs_chan,
+			     ARRAY_SIZE(d40_backup_regs_chan),
+			     save);
+	}
+
+	/* Save/Restore global registers */
+	dma40_backup(base->virtbase, base->reg_val_backup,
+		     d40_backup_regs, ARRAY_SIZE(d40_backup_regs),
+		     save);
+
+	/* Save/Restore registers only existing on dma40 v3 and later */
+	if (base->gen_dmac.backup)
+		dma40_backup(base->virtbase, base->reg_val_backup_v4,
+			     base->gen_dmac.backup,
+			base->gen_dmac.backup_size,
+			save);
+}
+
 static int dma40_runtime_suspend(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -3030,36 +3042,20 @@ static int dma40_runtime_resume(struct device *dev)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct d40_base *base = platform_get_drvdata(pdev);
 
-	if (base->initialized)
-		d40_save_restore_registers(base, false);
+	d40_save_restore_registers(base, false);
 
 	writel_relaxed(D40_DREG_GCC_ENABLE_ALL,
 		       base->virtbase + D40_DREG_GCC);
 	return 0;
 }
-
-static int dma40_resume(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct d40_base *base = platform_get_drvdata(pdev);
-	int ret = 0;
-
-	if (base->lcpa_regulator)
-		ret = regulator_enable(base->lcpa_regulator);
-
-	return ret;
-}
+#endif
 
 static const struct dev_pm_ops dma40_pm_ops = {
-	.suspend		= dma40_pm_suspend,
-	.runtime_suspend	= dma40_runtime_suspend,
-	.runtime_resume		= dma40_runtime_resume,
-	.resume			= dma40_resume,
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(dma40_suspend, dma40_resume)
+	SET_PM_RUNTIME_PM_OPS(dma40_runtime_suspend,
+				dma40_runtime_resume,
+				NULL)
 };
-#define DMA40_PM_OPS	(&dma40_pm_ops)
-#else
-#define DMA40_PM_OPS	NULL
-#endif
 
 /* Initialization functions. */
 
@@ -3645,12 +3641,6 @@ static int __init d40_probe(struct platform_device *pdev)
 		goto failure;
 	}
 
-	pm_runtime_irq_safe(base->dev);
-	pm_runtime_set_autosuspend_delay(base->dev, DMA40_AUTOSUSPEND_DELAY);
-	pm_runtime_use_autosuspend(base->dev);
-	pm_runtime_enable(base->dev);
-	pm_runtime_resume(base->dev);
-
 	if (base->plat_data->use_esram_lcla) {
 
 		base->lcpa_regulator = regulator_get(base->dev, "lcla_esram");
@@ -3671,7 +3661,15 @@ static int __init d40_probe(struct platform_device *pdev)
 		}
 	}
 
-	base->initialized = true;
+	writel_relaxed(D40_DREG_GCC_ENABLE_ALL, base->virtbase + D40_DREG_GCC);
+
+	pm_runtime_irq_safe(base->dev);
+	pm_runtime_set_autosuspend_delay(base->dev, DMA40_AUTOSUSPEND_DELAY);
+	pm_runtime_use_autosuspend(base->dev);
+	pm_runtime_mark_last_busy(base->dev);
+	pm_runtime_set_active(base->dev);
+	pm_runtime_enable(base->dev);
+
 	ret = d40_dmaengine_init(base, num_reserved_chans);
 	if (ret)
 		goto failure;
@@ -3754,7 +3752,7 @@ static struct platform_driver d40_driver = {
 	.driver = {
 		.owner = THIS_MODULE,
 		.name  = D40_NAME,
-		.pm = DMA40_PM_OPS,
+		.pm = &dma40_pm_ops,
 		.of_match_table = d40_match,
 	},
 };
diff --git a/drivers/dma/xilinx/Makefile b/drivers/dma/xilinx/Makefile
new file mode 100644
index 000000000000..3c4e9f2fea28
--- /dev/null
+++ b/drivers/dma/xilinx/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_XILINX_VDMA) += xilinx_vdma.o
diff --git a/drivers/dma/xilinx/xilinx_vdma.c b/drivers/dma/xilinx/xilinx_vdma.c
new file mode 100644
index 000000000000..42a13e8d4607
--- /dev/null
+++ b/drivers/dma/xilinx/xilinx_vdma.c
@@ -0,0 +1,1379 @@
+/*
+ * DMA driver for Xilinx Video DMA Engine
+ *
+ * Copyright (C) 2010-2014 Xilinx, Inc. All rights reserved.
+ *
+ * Based on the Freescale DMA driver.
+ *
+ * Description:
+ * The AXI Video Direct Memory Access (AXI VDMA) core is a soft Xilinx IP
+ * core that provides high-bandwidth direct memory access between memory
+ * and AXI4-Stream type video target peripherals. The core provides efficient
+ * two dimensional DMA operations with independent asynchronous read (S2MM)
+ * and write (MM2S) channel operation. It can be configured to have either
+ * one channel or two channels. If configured as two channels, one is to
+ * transmit to the video device (MM2S) and another is to receive from the
+ * video device (S2MM). Initialization, status, interrupt and management
+ * registers are accessed through an AXI4-Lite slave interface.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/amba/xilinx_dma.h>
+#include <linux/bitops.h>
+#include <linux/dmapool.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of_address.h>
+#include <linux/of_dma.h>
+#include <linux/of_platform.h>
+#include <linux/of_irq.h>
+#include <linux/slab.h>
+
+#include "../dmaengine.h"
+
+/* Register/Descriptor Offsets */
+#define XILINX_VDMA_MM2S_CTRL_OFFSET		0x0000
+#define XILINX_VDMA_S2MM_CTRL_OFFSET		0x0030
+#define XILINX_VDMA_MM2S_DESC_OFFSET		0x0050
+#define XILINX_VDMA_S2MM_DESC_OFFSET		0x00a0
+
+/* Control Registers */
+#define XILINX_VDMA_REG_DMACR			0x0000
+#define XILINX_VDMA_DMACR_DELAY_MAX		0xff
+#define XILINX_VDMA_DMACR_DELAY_SHIFT		24
+#define XILINX_VDMA_DMACR_FRAME_COUNT_MAX	0xff
+#define XILINX_VDMA_DMACR_FRAME_COUNT_SHIFT	16
+#define XILINX_VDMA_DMACR_ERR_IRQ		BIT(14)
+#define XILINX_VDMA_DMACR_DLY_CNT_IRQ		BIT(13)
+#define XILINX_VDMA_DMACR_FRM_CNT_IRQ		BIT(12)
+#define XILINX_VDMA_DMACR_MASTER_SHIFT		8
+#define XILINX_VDMA_DMACR_FSYNCSRC_SHIFT	5
+#define XILINX_VDMA_DMACR_FRAMECNT_EN		BIT(4)
+#define XILINX_VDMA_DMACR_GENLOCK_EN		BIT(3)
+#define XILINX_VDMA_DMACR_RESET			BIT(2)
+#define XILINX_VDMA_DMACR_CIRC_EN		BIT(1)
+#define XILINX_VDMA_DMACR_RUNSTOP		BIT(0)
+#define XILINX_VDMA_DMACR_FSYNCSRC_MASK		GENMASK(6, 5)
+
+#define XILINX_VDMA_REG_DMASR			0x0004
+#define XILINX_VDMA_DMASR_EOL_LATE_ERR		BIT(15)
+#define XILINX_VDMA_DMASR_ERR_IRQ		BIT(14)
+#define XILINX_VDMA_DMASR_DLY_CNT_IRQ		BIT(13)
+#define XILINX_VDMA_DMASR_FRM_CNT_IRQ		BIT(12)
+#define XILINX_VDMA_DMASR_SOF_LATE_ERR		BIT(11)
+#define XILINX_VDMA_DMASR_SG_DEC_ERR		BIT(10)
+#define XILINX_VDMA_DMASR_SG_SLV_ERR		BIT(9)
+#define XILINX_VDMA_DMASR_EOF_EARLY_ERR		BIT(8)
+#define XILINX_VDMA_DMASR_SOF_EARLY_ERR		BIT(7)
+#define XILINX_VDMA_DMASR_DMA_DEC_ERR		BIT(6)
+#define XILINX_VDMA_DMASR_DMA_SLAVE_ERR		BIT(5)
+#define XILINX_VDMA_DMASR_DMA_INT_ERR		BIT(4)
+#define XILINX_VDMA_DMASR_IDLE			BIT(1)
+#define XILINX_VDMA_DMASR_HALTED		BIT(0)
+#define XILINX_VDMA_DMASR_DELAY_MASK		GENMASK(31, 24)
+#define XILINX_VDMA_DMASR_FRAME_COUNT_MASK	GENMASK(23, 16)
+
+#define XILINX_VDMA_REG_CURDESC			0x0008
+#define XILINX_VDMA_REG_TAILDESC		0x0010
+#define XILINX_VDMA_REG_REG_INDEX		0x0014
+#define XILINX_VDMA_REG_FRMSTORE		0x0018
+#define XILINX_VDMA_REG_THRESHOLD		0x001c
+#define XILINX_VDMA_REG_FRMPTR_STS		0x0024
+#define XILINX_VDMA_REG_PARK_PTR		0x0028
+#define XILINX_VDMA_PARK_PTR_WR_REF_SHIFT	8
+#define XILINX_VDMA_PARK_PTR_RD_REF_SHIFT	0
+#define XILINX_VDMA_REG_VDMA_VERSION		0x002c
+
+/* Register Direct Mode Registers */
+#define XILINX_VDMA_REG_VSIZE			0x0000
+#define XILINX_VDMA_REG_HSIZE			0x0004
+
+#define XILINX_VDMA_REG_FRMDLY_STRIDE		0x0008
+#define XILINX_VDMA_FRMDLY_STRIDE_FRMDLY_SHIFT	24
+#define XILINX_VDMA_FRMDLY_STRIDE_STRIDE_SHIFT	0
+
+#define XILINX_VDMA_REG_START_ADDRESS(n)	(0x000c + 4 * (n))
+
+/* HW specific definitions */
+#define XILINX_VDMA_MAX_CHANS_PER_DEVICE	0x2
+
+#define XILINX_VDMA_DMAXR_ALL_IRQ_MASK	\
+		(XILINX_VDMA_DMASR_FRM_CNT_IRQ | \
+		 XILINX_VDMA_DMASR_DLY_CNT_IRQ | \
+		 XILINX_VDMA_DMASR_ERR_IRQ)
+
+#define XILINX_VDMA_DMASR_ALL_ERR_MASK	\
+		(XILINX_VDMA_DMASR_EOL_LATE_ERR | \
+		 XILINX_VDMA_DMASR_SOF_LATE_ERR | \
+		 XILINX_VDMA_DMASR_SG_DEC_ERR | \
+		 XILINX_VDMA_DMASR_SG_SLV_ERR | \
+		 XILINX_VDMA_DMASR_EOF_EARLY_ERR | \
+		 XILINX_VDMA_DMASR_SOF_EARLY_ERR | \
+		 XILINX_VDMA_DMASR_DMA_DEC_ERR | \
+		 XILINX_VDMA_DMASR_DMA_SLAVE_ERR | \
+		 XILINX_VDMA_DMASR_DMA_INT_ERR)
+
+/*
+ * Recoverable errors are DMA Internal error, SOF Early, EOF Early
+ * and SOF Late. They are only recoverable when C_FLUSH_ON_FSYNC
+ * is enabled in the h/w system.
+ */
+#define XILINX_VDMA_DMASR_ERR_RECOVER_MASK	\
+		(XILINX_VDMA_DMASR_SOF_LATE_ERR | \
+		 XILINX_VDMA_DMASR_EOF_EARLY_ERR | \
+		 XILINX_VDMA_DMASR_SOF_EARLY_ERR | \
+		 XILINX_VDMA_DMASR_DMA_INT_ERR)
+
+/* Axi VDMA Flush on Fsync bits */
+#define XILINX_VDMA_FLUSH_S2MM		3
+#define XILINX_VDMA_FLUSH_MM2S		2
+#define XILINX_VDMA_FLUSH_BOTH		1
+
+/* Delay loop counter to prevent hardware failure */
+#define XILINX_VDMA_LOOP_COUNT		1000000
+
+/**
+ * struct xilinx_vdma_desc_hw - Hardware Descriptor
+ * @next_desc: Next Descriptor Pointer @0x00
+ * @pad1: Reserved @0x04
+ * @buf_addr: Buffer address @0x08
+ * @pad2: Reserved @0x0C
+ * @vsize: Vertical Size @0x10
+ * @hsize: Horizontal Size @0x14
+ * @stride: Number of bytes between the first
+ *	    pixels of each horizontal line @0x18
+ */
+struct xilinx_vdma_desc_hw {
+	u32 next_desc;
+	u32 pad1;
+	u32 buf_addr;
+	u32 pad2;
+	u32 vsize;
+	u32 hsize;
+	u32 stride;
+} __aligned(64);
+
+/**
+ * struct xilinx_vdma_tx_segment - Descriptor segment
+ * @hw: Hardware descriptor
+ * @node: Node in the descriptor segments list
+ * @phys: Physical address of segment
+ */
+struct xilinx_vdma_tx_segment {
+	struct xilinx_vdma_desc_hw hw;
+	struct list_head node;
+	dma_addr_t phys;
+} __aligned(64);
+
+/**
+ * struct xilinx_vdma_tx_descriptor - Per Transaction structure
+ * @async_tx: Async transaction descriptor
+ * @segments: TX segments list
+ * @node: Node in the channel descriptors list
+ */
+struct xilinx_vdma_tx_descriptor {
+	struct dma_async_tx_descriptor async_tx;
+	struct list_head segments;
+	struct list_head node;
+};
+
+/**
+ * struct xilinx_vdma_chan - Driver specific VDMA channel structure
+ * @xdev: Driver specific device structure
+ * @ctrl_offset: Control registers offset
+ * @desc_offset: TX descriptor registers offset
+ * @lock: Descriptor operation lock
+ * @pending_list: Descriptors waiting
+ * @active_desc: Active descriptor
+ * @allocated_desc: Allocated descriptor
+ * @done_list: Complete descriptors
+ * @common: DMA common channel
+ * @desc_pool: Descriptors pool
+ * @dev: The dma device
+ * @irq: Channel IRQ
+ * @id: Channel ID
+ * @direction: Transfer direction
+ * @num_frms: Number of frames
+ * @has_sg: Support scatter transfers
+ * @genlock: Support genlock mode
+ * @err: Channel has errors
+ * @tasklet: Cleanup work after irq
+ * @config: Device configuration info
+ * @flush_on_fsync: Flush on Frame sync
+ */
+struct xilinx_vdma_chan {
+	struct xilinx_vdma_device *xdev;
+	u32 ctrl_offset;
+	u32 desc_offset;
+	spinlock_t lock;
+	struct list_head pending_list;
+	struct xilinx_vdma_tx_descriptor *active_desc;
+	struct xilinx_vdma_tx_descriptor *allocated_desc;
+	struct list_head done_list;
+	struct dma_chan common;
+	struct dma_pool *desc_pool;
+	struct device *dev;
+	int irq;
+	int id;
+	enum dma_transfer_direction direction;
+	int num_frms;
+	bool has_sg;
+	bool genlock;
+	bool err;
+	struct tasklet_struct tasklet;
+	struct xilinx_vdma_config config;
+	bool flush_on_fsync;
+};
+
+/**
+ * struct xilinx_vdma_device - VDMA device structure
+ * @regs: I/O mapped base address
+ * @dev: Device Structure
+ * @common: DMA device structure
+ * @chan: Driver specific VDMA channel
+ * @has_sg: Specifies whether Scatter-Gather is present or not
+ * @flush_on_fsync: Flush on frame sync
+ */
+struct xilinx_vdma_device {
+	void __iomem *regs;
+	struct device *dev;
+	struct dma_device common;
+	struct xilinx_vdma_chan *chan[XILINX_VDMA_MAX_CHANS_PER_DEVICE];
+	bool has_sg;
+	u32 flush_on_fsync;
+};
+
+/* Macros */
+#define to_xilinx_chan(chan) \
+	container_of(chan, struct xilinx_vdma_chan, common)
+#define to_vdma_tx_descriptor(tx) \
+	container_of(tx, struct xilinx_vdma_tx_descriptor, async_tx)
+
+/* IO accessors */
+static inline u32 vdma_read(struct xilinx_vdma_chan *chan, u32 reg)
+{
+	return ioread32(chan->xdev->regs + reg);
+}
+
+static inline void vdma_write(struct xilinx_vdma_chan *chan, u32 reg, u32 value)
+{
+	iowrite32(value, chan->xdev->regs + reg);
+}
+
+static inline void vdma_desc_write(struct xilinx_vdma_chan *chan, u32 reg,
+				   u32 value)
+{
+	vdma_write(chan, chan->desc_offset + reg, value);
+}
+
+static inline u32 vdma_ctrl_read(struct xilinx_vdma_chan *chan, u32 reg)
+{
+	return vdma_read(chan, chan->ctrl_offset + reg);
+}
+
+static inline void vdma_ctrl_write(struct xilinx_vdma_chan *chan, u32 reg,
+				   u32 value)
+{
+	vdma_write(chan, chan->ctrl_offset + reg, value);
+}
+
+static inline void vdma_ctrl_clr(struct xilinx_vdma_chan *chan, u32 reg,
+				 u32 clr)
+{
+	vdma_ctrl_write(chan, reg, vdma_ctrl_read(chan, reg) & ~clr);
+}
+
+static inline void vdma_ctrl_set(struct xilinx_vdma_chan *chan, u32 reg,
+				 u32 set)
+{
+	vdma_ctrl_write(chan, reg, vdma_ctrl_read(chan, reg) | set);
+}
+
+/* -----------------------------------------------------------------------------
+ * Descriptors and segments alloc and free
+ */
+
+/**
+ * xilinx_vdma_alloc_tx_segment - Allocate transaction segment
+ * @chan: Driver specific VDMA channel
+ *
+ * Return: The allocated segment on success and NULL on failure.
+ */
+static struct xilinx_vdma_tx_segment *
+xilinx_vdma_alloc_tx_segment(struct xilinx_vdma_chan *chan)
+{
+	struct xilinx_vdma_tx_segment *segment;
+	dma_addr_t phys;
+
+	segment = dma_pool_alloc(chan->desc_pool, GFP_ATOMIC, &phys);
+	if (!segment)
+		return NULL;
+
+	memset(segment, 0, sizeof(*segment));
+	segment->phys = phys;
+
+	return segment;
+}
+
+/**
+ * xilinx_vdma_free_tx_segment - Free transaction segment
+ * @chan: Driver specific VDMA channel
+ * @segment: VDMA transaction segment
+ */
+static void xilinx_vdma_free_tx_segment(struct xilinx_vdma_chan *chan,
+					struct xilinx_vdma_tx_segment *segment)
+{
+	dma_pool_free(chan->desc_pool, segment, segment->phys);
+}
+
+/**
+ * xilinx_vdma_tx_descriptor - Allocate transaction descriptor
+ * @chan: Driver specific VDMA channel
+ *
+ * Return: The allocated descriptor on success and NULL on failure.
+ */
+static struct xilinx_vdma_tx_descriptor *
+xilinx_vdma_alloc_tx_descriptor(struct xilinx_vdma_chan *chan)
+{
+	struct xilinx_vdma_tx_descriptor *desc;
+	unsigned long flags;
+
+	if (chan->allocated_desc)
+		return chan->allocated_desc;
+
+	desc = kzalloc(sizeof(*desc), GFP_KERNEL);
+	if (!desc)
+		return NULL;
+
+	spin_lock_irqsave(&chan->lock, flags);
+	chan->allocated_desc = desc;
+	spin_unlock_irqrestore(&chan->lock, flags);
+
+	INIT_LIST_HEAD(&desc->segments);
+
+	return desc;
+}
+
+/**
+ * xilinx_vdma_free_tx_descriptor - Free transaction descriptor
+ * @chan: Driver specific VDMA channel
+ * @desc: VDMA transaction descriptor
+ */
+static void
+xilinx_vdma_free_tx_descriptor(struct xilinx_vdma_chan *chan,
+			       struct xilinx_vdma_tx_descriptor *desc)
+{
+	struct xilinx_vdma_tx_segment *segment, *next;
+
+	if (!desc)
+		return;
+
+	list_for_each_entry_safe(segment, next, &desc->segments, node) {
+		list_del(&segment->node);
+		xilinx_vdma_free_tx_segment(chan, segment);
+	}
+
+	kfree(desc);
+}
+
+/* Required functions */
+
+/**
+ * xilinx_vdma_free_desc_list - Free descriptors list
+ * @chan: Driver specific VDMA channel
+ * @list: List to parse and delete the descriptor
+ */
+static void xilinx_vdma_free_desc_list(struct xilinx_vdma_chan *chan,
+					struct list_head *list)
+{
+	struct xilinx_vdma_tx_descriptor *desc, *next;
+
+	list_for_each_entry_safe(desc, next, list, node) {
+		list_del(&desc->node);
+		xilinx_vdma_free_tx_descriptor(chan, desc);
+	}
+}
+
+/**
+ * xilinx_vdma_free_descriptors - Free channel descriptors
+ * @chan: Driver specific VDMA channel
+ */
+static void xilinx_vdma_free_descriptors(struct xilinx_vdma_chan *chan)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&chan->lock, flags);
+
+	xilinx_vdma_free_desc_list(chan, &chan->pending_list);
+	xilinx_vdma_free_desc_list(chan, &chan->done_list);
+
+	xilinx_vdma_free_tx_descriptor(chan, chan->active_desc);
+	chan->active_desc = NULL;
+
+	spin_unlock_irqrestore(&chan->lock, flags);
+}
+
+/**
+ * xilinx_vdma_free_chan_resources - Free channel resources
+ * @dchan: DMA channel
+ */
+static void xilinx_vdma_free_chan_resources(struct dma_chan *dchan)
+{
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+
+	dev_dbg(chan->dev, "Free all channel resources.\n");
+
+	xilinx_vdma_free_descriptors(chan);
+	dma_pool_destroy(chan->desc_pool);
+	chan->desc_pool = NULL;
+}
+
+/**
+ * xilinx_vdma_chan_desc_cleanup - Clean channel descriptors
+ * @chan: Driver specific VDMA channel
+ */
+static void xilinx_vdma_chan_desc_cleanup(struct xilinx_vdma_chan *chan)
+{
+	struct xilinx_vdma_tx_descriptor *desc, *next;
+	unsigned long flags;
+
+	spin_lock_irqsave(&chan->lock, flags);
+
+	list_for_each_entry_safe(desc, next, &chan->done_list, node) {
+		dma_async_tx_callback callback;
+		void *callback_param;
+
+		/* Remove from the list of running transactions */
+		list_del(&desc->node);
+
+		/* Run the link descriptor callback function */
+		callback = desc->async_tx.callback;
+		callback_param = desc->async_tx.callback_param;
+		if (callback) {
+			spin_unlock_irqrestore(&chan->lock, flags);
+			callback(callback_param);
+			spin_lock_irqsave(&chan->lock, flags);
+		}
+
+		/* Run any dependencies, then free the descriptor */
+		dma_run_dependencies(&desc->async_tx);
+		xilinx_vdma_free_tx_descriptor(chan, desc);
+	}
+
+	spin_unlock_irqrestore(&chan->lock, flags);
+}
+
+/**
+ * xilinx_vdma_do_tasklet - Schedule completion tasklet
+ * @data: Pointer to the Xilinx VDMA channel structure
+ */
+static void xilinx_vdma_do_tasklet(unsigned long data)
+{
+	struct xilinx_vdma_chan *chan = (struct xilinx_vdma_chan *)data;
+
+	xilinx_vdma_chan_desc_cleanup(chan);
+}
+
+/**
+ * xilinx_vdma_alloc_chan_resources - Allocate channel resources
+ * @dchan: DMA channel
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int xilinx_vdma_alloc_chan_resources(struct dma_chan *dchan)
+{
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+
+	/* Has this channel already been allocated? */
+	if (chan->desc_pool)
+		return 0;
+
+	/*
+	 * We need the descriptor to be aligned to 64bytes
+	 * for meeting Xilinx VDMA specification requirement.
+	 */
+	chan->desc_pool = dma_pool_create("xilinx_vdma_desc_pool",
+				chan->dev,
+				sizeof(struct xilinx_vdma_tx_segment),
+				__alignof__(struct xilinx_vdma_tx_segment), 0);
+	if (!chan->desc_pool) {
+		dev_err(chan->dev,
+			"unable to allocate channel %d descriptor pool\n",
+			chan->id);
+		return -ENOMEM;
+	}
+
+	dma_cookie_init(dchan);
+	return 0;
+}
+
+/**
+ * xilinx_vdma_tx_status - Get VDMA transaction status
+ * @dchan: DMA channel
+ * @cookie: Transaction identifier
+ * @txstate: Transaction state
+ *
+ * Return: DMA transaction status
+ */
+static enum dma_status xilinx_vdma_tx_status(struct dma_chan *dchan,
+					dma_cookie_t cookie,
+					struct dma_tx_state *txstate)
+{
+	return dma_cookie_status(dchan, cookie, txstate);
+}
+
+/**
+ * xilinx_vdma_is_running - Check if VDMA channel is running
+ * @chan: Driver specific VDMA channel
+ *
+ * Return: '1' if running, '0' if not.
+ */
+static bool xilinx_vdma_is_running(struct xilinx_vdma_chan *chan)
+{
+	return !(vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR) &
+		 XILINX_VDMA_DMASR_HALTED) &&
+		(vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR) &
+		 XILINX_VDMA_DMACR_RUNSTOP);
+}
+
+/**
+ * xilinx_vdma_is_idle - Check if VDMA channel is idle
+ * @chan: Driver specific VDMA channel
+ *
+ * Return: '1' if idle, '0' if not.
+ */
+static bool xilinx_vdma_is_idle(struct xilinx_vdma_chan *chan)
+{
+	return vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR) &
+		XILINX_VDMA_DMASR_IDLE;
+}
+
+/**
+ * xilinx_vdma_halt - Halt VDMA channel
+ * @chan: Driver specific VDMA channel
+ */
+static void xilinx_vdma_halt(struct xilinx_vdma_chan *chan)
+{
+	int loop = XILINX_VDMA_LOOP_COUNT;
+
+	vdma_ctrl_clr(chan, XILINX_VDMA_REG_DMACR, XILINX_VDMA_DMACR_RUNSTOP);
+
+	/* Wait for the hardware to halt */
+	do {
+		if (vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR) &
+		    XILINX_VDMA_DMASR_HALTED)
+			break;
+	} while (loop--);
+
+	if (!loop) {
+		dev_err(chan->dev, "Cannot stop channel %p: %x\n",
+			chan, vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR));
+		chan->err = true;
+	}
+
+	return;
+}
+
+/**
+ * xilinx_vdma_start - Start VDMA channel
+ * @chan: Driver specific VDMA channel
+ */
+static void xilinx_vdma_start(struct xilinx_vdma_chan *chan)
+{
+	int loop = XILINX_VDMA_LOOP_COUNT;
+
+	vdma_ctrl_set(chan, XILINX_VDMA_REG_DMACR, XILINX_VDMA_DMACR_RUNSTOP);
+
+	/* Wait for the hardware to start */
+	do {
+		if (!(vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR) &
+		      XILINX_VDMA_DMASR_HALTED))
+			break;
+	} while (loop--);
+
+	if (!loop) {
+		dev_err(chan->dev, "Cannot start channel %p: %x\n",
+			chan, vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR));
+
+		chan->err = true;
+	}
+
+	return;
+}
+
+/**
+ * xilinx_vdma_start_transfer - Starts VDMA transfer
+ * @chan: Driver specific channel struct pointer
+ */
+static void xilinx_vdma_start_transfer(struct xilinx_vdma_chan *chan)
+{
+	struct xilinx_vdma_config *config = &chan->config;
+	struct xilinx_vdma_tx_descriptor *desc;
+	unsigned long flags;
+	u32 reg;
+	struct xilinx_vdma_tx_segment *head, *tail = NULL;
+
+	if (chan->err)
+		return;
+
+	spin_lock_irqsave(&chan->lock, flags);
+
+	/* There's already an active descriptor, bail out. */
+	if (chan->active_desc)
+		goto out_unlock;
+
+	if (list_empty(&chan->pending_list))
+		goto out_unlock;
+
+	desc = list_first_entry(&chan->pending_list,
+				struct xilinx_vdma_tx_descriptor, node);
+
+	/* If it is SG mode and hardware is busy, cannot submit */
+	if (chan->has_sg && xilinx_vdma_is_running(chan) &&
+	    !xilinx_vdma_is_idle(chan)) {
+		dev_dbg(chan->dev, "DMA controller still busy\n");
+		goto out_unlock;
+	}
+
+	/*
+	 * If hardware is idle, then all descriptors on the running lists are
+	 * done, start new transfers
+	 */
+	if (chan->has_sg) {
+		head = list_first_entry(&desc->segments,
+					struct xilinx_vdma_tx_segment, node);
+		tail = list_entry(desc->segments.prev,
+				  struct xilinx_vdma_tx_segment, node);
+
+		vdma_ctrl_write(chan, XILINX_VDMA_REG_CURDESC, head->phys);
+	}
+
+	/* Configure the hardware using info in the config structure */
+	reg = vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR);
+
+	if (config->frm_cnt_en)
+		reg |= XILINX_VDMA_DMACR_FRAMECNT_EN;
+	else
+		reg &= ~XILINX_VDMA_DMACR_FRAMECNT_EN;
+
+	/*
+	 * With SG, start with circular mode, so that BDs can be fetched.
+	 * In direct register mode, if not parking, enable circular mode
+	 */
+	if (chan->has_sg || !config->park)
+		reg |= XILINX_VDMA_DMACR_CIRC_EN;
+
+	if (config->park)
+		reg &= ~XILINX_VDMA_DMACR_CIRC_EN;
+
+	vdma_ctrl_write(chan, XILINX_VDMA_REG_DMACR, reg);
+
+	if (config->park && (config->park_frm >= 0) &&
+			(config->park_frm < chan->num_frms)) {
+		if (chan->direction == DMA_MEM_TO_DEV)
+			vdma_write(chan, XILINX_VDMA_REG_PARK_PTR,
+				config->park_frm <<
+					XILINX_VDMA_PARK_PTR_RD_REF_SHIFT);
+		else
+			vdma_write(chan, XILINX_VDMA_REG_PARK_PTR,
+				config->park_frm <<
+					XILINX_VDMA_PARK_PTR_WR_REF_SHIFT);
+	}
+
+	/* Start the hardware */
+	xilinx_vdma_start(chan);
+
+	if (chan->err)
+		goto out_unlock;
+
+	/* Start the transfer */
+	if (chan->has_sg) {
+		vdma_ctrl_write(chan, XILINX_VDMA_REG_TAILDESC, tail->phys);
+	} else {
+		struct xilinx_vdma_tx_segment *segment, *last = NULL;
+		int i = 0;
+
+		list_for_each_entry(segment, &desc->segments, node) {
+			vdma_desc_write(chan,
+					XILINX_VDMA_REG_START_ADDRESS(i++),
+					segment->hw.buf_addr);
+			last = segment;
+		}
+
+		if (!last)
+			goto out_unlock;
+
+		/* HW expects these parameters to be same for one transaction */
+		vdma_desc_write(chan, XILINX_VDMA_REG_HSIZE, last->hw.hsize);
+		vdma_desc_write(chan, XILINX_VDMA_REG_FRMDLY_STRIDE,
+				last->hw.stride);
+		vdma_desc_write(chan, XILINX_VDMA_REG_VSIZE, last->hw.vsize);
+	}
+
+	list_del(&desc->node);
+	chan->active_desc = desc;
+
+out_unlock:
+	spin_unlock_irqrestore(&chan->lock, flags);
+}
+
+/**
+ * xilinx_vdma_issue_pending - Issue pending transactions
+ * @dchan: DMA channel
+ */
+static void xilinx_vdma_issue_pending(struct dma_chan *dchan)
+{
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+
+	xilinx_vdma_start_transfer(chan);
+}
+
+/**
+ * xilinx_vdma_complete_descriptor - Mark the active descriptor as complete
+ * @chan : xilinx DMA channel
+ *
+ * CONTEXT: hardirq
+ */
+static void xilinx_vdma_complete_descriptor(struct xilinx_vdma_chan *chan)
+{
+	struct xilinx_vdma_tx_descriptor *desc;
+	unsigned long flags;
+
+	spin_lock_irqsave(&chan->lock, flags);
+
+	desc = chan->active_desc;
+	if (!desc) {
+		dev_dbg(chan->dev, "no running descriptors\n");
+		goto out_unlock;
+	}
+
+	dma_cookie_complete(&desc->async_tx);
+	list_add_tail(&desc->node, &chan->done_list);
+
+	chan->active_desc = NULL;
+
+out_unlock:
+	spin_unlock_irqrestore(&chan->lock, flags);
+}
+
+/**
+ * xilinx_vdma_reset - Reset VDMA channel
+ * @chan: Driver specific VDMA channel
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int xilinx_vdma_reset(struct xilinx_vdma_chan *chan)
+{
+	int loop = XILINX_VDMA_LOOP_COUNT;
+	u32 tmp;
+
+	vdma_ctrl_set(chan, XILINX_VDMA_REG_DMACR, XILINX_VDMA_DMACR_RESET);
+
+	tmp = vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR) &
+		XILINX_VDMA_DMACR_RESET;
+
+	/* Wait for the hardware to finish reset */
+	do {
+		tmp = vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR) &
+			XILINX_VDMA_DMACR_RESET;
+	} while (loop-- && tmp);
+
+	if (!loop) {
+		dev_err(chan->dev, "reset timeout, cr %x, sr %x\n",
+			vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR),
+			vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR));
+		return -ETIMEDOUT;
+	}
+
+	chan->err = false;
+
+	return 0;
+}
+
+/**
+ * xilinx_vdma_chan_reset - Reset VDMA channel and enable interrupts
+ * @chan: Driver specific VDMA channel
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int xilinx_vdma_chan_reset(struct xilinx_vdma_chan *chan)
+{
+	int err;
+
+	/* Reset VDMA */
+	err = xilinx_vdma_reset(chan);
+	if (err)
+		return err;
+
+	/* Enable interrupts */
+	vdma_ctrl_set(chan, XILINX_VDMA_REG_DMACR,
+		      XILINX_VDMA_DMAXR_ALL_IRQ_MASK);
+
+	return 0;
+}
+
+/**
+ * xilinx_vdma_irq_handler - VDMA Interrupt handler
+ * @irq: IRQ number
+ * @data: Pointer to the Xilinx VDMA channel structure
+ *
+ * Return: IRQ_HANDLED/IRQ_NONE
+ */
+static irqreturn_t xilinx_vdma_irq_handler(int irq, void *data)
+{
+	struct xilinx_vdma_chan *chan = data;
+	u32 status;
+
+	/* Read the status and ack the interrupts. */
+	status = vdma_ctrl_read(chan, XILINX_VDMA_REG_DMASR);
+	if (!(status & XILINX_VDMA_DMAXR_ALL_IRQ_MASK))
+		return IRQ_NONE;
+
+	vdma_ctrl_write(chan, XILINX_VDMA_REG_DMASR,
+			status & XILINX_VDMA_DMAXR_ALL_IRQ_MASK);
+
+	if (status & XILINX_VDMA_DMASR_ERR_IRQ) {
+		/*
+		 * An error occurred. If C_FLUSH_ON_FSYNC is enabled and the
+		 * error is recoverable, ignore it. Otherwise flag the error.
+		 *
+		 * Only recoverable errors can be cleared in the DMASR register,
+		 * make sure not to write to other error bits to 1.
+		 */
+		u32 errors = status & XILINX_VDMA_DMASR_ALL_ERR_MASK;
+		vdma_ctrl_write(chan, XILINX_VDMA_REG_DMASR,
+				errors & XILINX_VDMA_DMASR_ERR_RECOVER_MASK);
+
+		if (!chan->flush_on_fsync ||
+		    (errors & ~XILINX_VDMA_DMASR_ERR_RECOVER_MASK)) {
+			dev_err(chan->dev,
+				"Channel %p has errors %x, cdr %x tdr %x\n",
+				chan, errors,
+				vdma_ctrl_read(chan, XILINX_VDMA_REG_CURDESC),
+				vdma_ctrl_read(chan, XILINX_VDMA_REG_TAILDESC));
+			chan->err = true;
+		}
+	}
+
+	if (status & XILINX_VDMA_DMASR_DLY_CNT_IRQ) {
+		/*
+		 * Device takes too long to do the transfer when user requires
+		 * responsiveness.
+		 */
+		dev_dbg(chan->dev, "Inter-packet latency too long\n");
+	}
+
+	if (status & XILINX_VDMA_DMASR_FRM_CNT_IRQ) {
+		xilinx_vdma_complete_descriptor(chan);
+		xilinx_vdma_start_transfer(chan);
+	}
+
+	tasklet_schedule(&chan->tasklet);
+	return IRQ_HANDLED;
+}
+
+/**
+ * xilinx_vdma_tx_submit - Submit DMA transaction
+ * @tx: Async transaction descriptor
+ *
+ * Return: cookie value on success and failure value on error
+ */
+static dma_cookie_t xilinx_vdma_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	struct xilinx_vdma_tx_descriptor *desc = to_vdma_tx_descriptor(tx);
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(tx->chan);
+	dma_cookie_t cookie;
+	unsigned long flags;
+	int err;
+
+	if (chan->err) {
+		/*
+		 * If reset fails, need to hard reset the system.
+		 * Channel is no longer functional
+		 */
+		err = xilinx_vdma_chan_reset(chan);
+		if (err < 0)
+			return err;
+	}
+
+	spin_lock_irqsave(&chan->lock, flags);
+
+	cookie = dma_cookie_assign(tx);
+
+	/* Append the transaction to the pending transactions queue. */
+	list_add_tail(&desc->node, &chan->pending_list);
+
+	/* Free the allocated desc */
+	chan->allocated_desc = NULL;
+
+	spin_unlock_irqrestore(&chan->lock, flags);
+
+	return cookie;
+}
+
+/**
+ * xilinx_vdma_dma_prep_interleaved - prepare a descriptor for a
+ *	DMA_SLAVE transaction
+ * @dchan: DMA channel
+ * @xt: Interleaved template pointer
+ * @flags: transfer ack flags
+ *
+ * Return: Async transaction descriptor on success and NULL on failure
+ */
+static struct dma_async_tx_descriptor *
+xilinx_vdma_dma_prep_interleaved(struct dma_chan *dchan,
+				 struct dma_interleaved_template *xt,
+				 unsigned long flags)
+{
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+	struct xilinx_vdma_tx_descriptor *desc;
+	struct xilinx_vdma_tx_segment *segment, *prev = NULL;
+	struct xilinx_vdma_desc_hw *hw;
+
+	if (!is_slave_direction(xt->dir))
+		return NULL;
+
+	if (!xt->numf || !xt->sgl[0].size)
+		return NULL;
+
+	/* Allocate a transaction descriptor. */
+	desc = xilinx_vdma_alloc_tx_descriptor(chan);
+	if (!desc)
+		return NULL;
+
+	dma_async_tx_descriptor_init(&desc->async_tx, &chan->common);
+	desc->async_tx.tx_submit = xilinx_vdma_tx_submit;
+	async_tx_ack(&desc->async_tx);
+
+	/* Allocate the link descriptor from DMA pool */
+	segment = xilinx_vdma_alloc_tx_segment(chan);
+	if (!segment)
+		goto error;
+
+	/* Fill in the hardware descriptor */
+	hw = &segment->hw;
+	hw->vsize = xt->numf;
+	hw->hsize = xt->sgl[0].size;
+	hw->stride = xt->sgl[0].icg <<
+			XILINX_VDMA_FRMDLY_STRIDE_STRIDE_SHIFT;
+	hw->stride |= chan->config.frm_dly <<
+			XILINX_VDMA_FRMDLY_STRIDE_FRMDLY_SHIFT;
+
+	if (xt->dir != DMA_MEM_TO_DEV)
+		hw->buf_addr = xt->dst_start;
+	else
+		hw->buf_addr = xt->src_start;
+
+	/* Link the previous next descriptor to current */
+	prev = list_last_entry(&desc->segments,
+				struct xilinx_vdma_tx_segment, node);
+	prev->hw.next_desc = segment->phys;
+
+	/* Insert the segment into the descriptor segments list. */
+	list_add_tail(&segment->node, &desc->segments);
+
+	prev = segment;
+
+	/* Link the last hardware descriptor with the first. */
+	segment = list_first_entry(&desc->segments,
+				   struct xilinx_vdma_tx_segment, node);
+	prev->hw.next_desc = segment->phys;
+
+	return &desc->async_tx;
+
+error:
+	xilinx_vdma_free_tx_descriptor(chan, desc);
+	return NULL;
+}
+
+/**
+ * xilinx_vdma_terminate_all - Halt the channel and free descriptors
+ * @chan: Driver specific VDMA Channel pointer
+ */
+static void xilinx_vdma_terminate_all(struct xilinx_vdma_chan *chan)
+{
+	/* Halt the DMA engine */
+	xilinx_vdma_halt(chan);
+
+	/* Remove and free all of the descriptors in the lists */
+	xilinx_vdma_free_descriptors(chan);
+}
+
+/**
+ * xilinx_vdma_channel_set_config - Configure VDMA channel
+ * Run-time configuration for Axi VDMA, supports:
+ * . halt the channel
+ * . configure interrupt coalescing and inter-packet delay threshold
+ * . start/stop parking
+ * . enable genlock
+ *
+ * @dchan: DMA channel
+ * @cfg: VDMA device configuration pointer
+ *
+ * Return: '0' on success and failure value on error
+ */
+int xilinx_vdma_channel_set_config(struct dma_chan *dchan,
+					struct xilinx_vdma_config *cfg)
+{
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+	u32 dmacr;
+
+	if (cfg->reset)
+		return xilinx_vdma_chan_reset(chan);
+
+	dmacr = vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR);
+
+	chan->config.frm_dly = cfg->frm_dly;
+	chan->config.park = cfg->park;
+
+	/* genlock settings */
+	chan->config.gen_lock = cfg->gen_lock;
+	chan->config.master = cfg->master;
+
+	if (cfg->gen_lock && chan->genlock) {
+		dmacr |= XILINX_VDMA_DMACR_GENLOCK_EN;
+		dmacr |= cfg->master << XILINX_VDMA_DMACR_MASTER_SHIFT;
+	}
+
+	chan->config.frm_cnt_en = cfg->frm_cnt_en;
+	if (cfg->park)
+		chan->config.park_frm = cfg->park_frm;
+	else
+		chan->config.park_frm = -1;
+
+	chan->config.coalesc = cfg->coalesc;
+	chan->config.delay = cfg->delay;
+
+	if (cfg->coalesc <= XILINX_VDMA_DMACR_FRAME_COUNT_MAX) {
+		dmacr |= cfg->coalesc << XILINX_VDMA_DMACR_FRAME_COUNT_SHIFT;
+		chan->config.coalesc = cfg->coalesc;
+	}
+
+	if (cfg->delay <= XILINX_VDMA_DMACR_DELAY_MAX) {
+		dmacr |= cfg->delay << XILINX_VDMA_DMACR_DELAY_SHIFT;
+		chan->config.delay = cfg->delay;
+	}
+
+	/* FSync Source selection */
+	dmacr &= ~XILINX_VDMA_DMACR_FSYNCSRC_MASK;
+	dmacr |= cfg->ext_fsync << XILINX_VDMA_DMACR_FSYNCSRC_SHIFT;
+
+	vdma_ctrl_write(chan, XILINX_VDMA_REG_DMACR, dmacr);
+
+	return 0;
+}
+EXPORT_SYMBOL(xilinx_vdma_channel_set_config);
+
+/**
+ * xilinx_vdma_device_control - Configure DMA channel of the device
+ * @dchan: DMA Channel pointer
+ * @cmd: DMA control command
+ * @arg: Channel configuration
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int xilinx_vdma_device_control(struct dma_chan *dchan,
+				      enum dma_ctrl_cmd cmd, unsigned long arg)
+{
+	struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+
+	if (cmd != DMA_TERMINATE_ALL)
+		return -ENXIO;
+
+	xilinx_vdma_terminate_all(chan);
+
+	return 0;
+}
+
+/* -----------------------------------------------------------------------------
+ * Probe and remove
+ */
+
+/**
+ * xilinx_vdma_chan_remove - Per Channel remove function
+ * @chan: Driver specific VDMA channel
+ */
+static void xilinx_vdma_chan_remove(struct xilinx_vdma_chan *chan)
+{
+	/* Disable all interrupts */
+	vdma_ctrl_clr(chan, XILINX_VDMA_REG_DMACR,
+		      XILINX_VDMA_DMAXR_ALL_IRQ_MASK);
+
+	if (chan->irq > 0)
+		free_irq(chan->irq, chan);
+
+	tasklet_kill(&chan->tasklet);
+
+	list_del(&chan->common.device_node);
+}
+
+/**
+ * xilinx_vdma_chan_probe - Per Channel Probing
+ * It get channel features from the device tree entry and
+ * initialize special channel handling routines
+ *
+ * @xdev: Driver specific device structure
+ * @node: Device node
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int xilinx_vdma_chan_probe(struct xilinx_vdma_device *xdev,
+				  struct device_node *node)
+{
+	struct xilinx_vdma_chan *chan;
+	bool has_dre = false;
+	u32 value, width;
+	int err;
+
+	/* Allocate and initialize the channel structure */
+	chan = devm_kzalloc(xdev->dev, sizeof(*chan), GFP_KERNEL);
+	if (!chan)
+		return -ENOMEM;
+
+	chan->dev = xdev->dev;
+	chan->xdev = xdev;
+	chan->has_sg = xdev->has_sg;
+
+	spin_lock_init(&chan->lock);
+	INIT_LIST_HEAD(&chan->pending_list);
+	INIT_LIST_HEAD(&chan->done_list);
+
+	/* Retrieve the channel properties from the device tree */
+	has_dre = of_property_read_bool(node, "xlnx,include-dre");
+
+	chan->genlock = of_property_read_bool(node, "xlnx,genlock-mode");
+
+	err = of_property_read_u32(node, "xlnx,datawidth", &value);
+	if (err) {
+		dev_err(xdev->dev, "missing xlnx,datawidth property\n");
+		return err;
+	}
+	width = value >> 3; /* Convert bits to bytes */
+
+	/* If data width is greater than 8 bytes, DRE is not in hw */
+	if (width > 8)
+		has_dre = false;
+
+	if (!has_dre)
+		xdev->common.copy_align = fls(width - 1);
+
+	if (of_device_is_compatible(node, "xlnx,axi-vdma-mm2s-channel")) {
+		chan->direction = DMA_MEM_TO_DEV;
+		chan->id = 0;
+
+		chan->ctrl_offset = XILINX_VDMA_MM2S_CTRL_OFFSET;
+		chan->desc_offset = XILINX_VDMA_MM2S_DESC_OFFSET;
+
+		if (xdev->flush_on_fsync == XILINX_VDMA_FLUSH_BOTH ||
+		    xdev->flush_on_fsync == XILINX_VDMA_FLUSH_MM2S)
+			chan->flush_on_fsync = true;
+	} else if (of_device_is_compatible(node,
+					    "xlnx,axi-vdma-s2mm-channel")) {
+		chan->direction = DMA_DEV_TO_MEM;
+		chan->id = 1;
+
+		chan->ctrl_offset = XILINX_VDMA_S2MM_CTRL_OFFSET;
+		chan->desc_offset = XILINX_VDMA_S2MM_DESC_OFFSET;
+
+		if (xdev->flush_on_fsync == XILINX_VDMA_FLUSH_BOTH ||
+		    xdev->flush_on_fsync == XILINX_VDMA_FLUSH_S2MM)
+			chan->flush_on_fsync = true;
+	} else {
+		dev_err(xdev->dev, "Invalid channel compatible node\n");
+		return -EINVAL;
+	}
+
+	/* Request the interrupt */
+	chan->irq = irq_of_parse_and_map(node, 0);
+	err = request_irq(chan->irq, xilinx_vdma_irq_handler, IRQF_SHARED,
+			  "xilinx-vdma-controller", chan);
+	if (err) {
+		dev_err(xdev->dev, "unable to request IRQ %d\n", chan->irq);
+		return err;
+	}
+
+	/* Initialize the tasklet */
+	tasklet_init(&chan->tasklet, xilinx_vdma_do_tasklet,
+			(unsigned long)chan);
+
+	/*
+	 * Initialize the DMA channel and add it to the DMA engine channels
+	 * list.
+	 */
+	chan->common.device = &xdev->common;
+
+	list_add_tail(&chan->common.device_node, &xdev->common.channels);
+	xdev->chan[chan->id] = chan;
+
+	/* Reset the channel */
+	err = xilinx_vdma_chan_reset(chan);
+	if (err < 0) {
+		dev_err(xdev->dev, "Reset channel failed\n");
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * of_dma_xilinx_xlate - Translation function
+ * @dma_spec: Pointer to DMA specifier as found in the device tree
+ * @ofdma: Pointer to DMA controller data
+ *
+ * Return: DMA channel pointer on success and NULL on error
+ */
+static struct dma_chan *of_dma_xilinx_xlate(struct of_phandle_args *dma_spec,
+						struct of_dma *ofdma)
+{
+	struct xilinx_vdma_device *xdev = ofdma->of_dma_data;
+	int chan_id = dma_spec->args[0];
+
+	if (chan_id >= XILINX_VDMA_MAX_CHANS_PER_DEVICE)
+		return NULL;
+
+	return dma_get_slave_channel(&xdev->chan[chan_id]->common);
+}
+
+/**
+ * xilinx_vdma_probe - Driver probe function
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int xilinx_vdma_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct xilinx_vdma_device *xdev;
+	struct device_node *child;
+	struct resource *io;
+	u32 num_frames;
+	int i, err;
+
+	/* Allocate and initialize the DMA engine structure */
+	xdev = devm_kzalloc(&pdev->dev, sizeof(*xdev), GFP_KERNEL);
+	if (!xdev)
+		return -ENOMEM;
+
+	xdev->dev = &pdev->dev;
+
+	/* Request and map I/O memory */
+	io = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	xdev->regs = devm_ioremap_resource(&pdev->dev, io);
+	if (IS_ERR(xdev->regs))
+		return PTR_ERR(xdev->regs);
+
+	/* Retrieve the DMA engine properties from the device tree */
+	xdev->has_sg = of_property_read_bool(node, "xlnx,include-sg");
+
+	err = of_property_read_u32(node, "xlnx,num-fstores", &num_frames);
+	if (err < 0) {
+		dev_err(xdev->dev, "missing xlnx,num-fstores property\n");
+		return err;
+	}
+
+	err = of_property_read_u32(node, "xlnx,flush-fsync",
+					&xdev->flush_on_fsync);
+	if (err < 0)
+		dev_warn(xdev->dev, "missing xlnx,flush-fsync property\n");
+
+	/* Initialize the DMA engine */
+	xdev->common.dev = &pdev->dev;
+
+	INIT_LIST_HEAD(&xdev->common.channels);
+	dma_cap_set(DMA_SLAVE, xdev->common.cap_mask);
+	dma_cap_set(DMA_PRIVATE, xdev->common.cap_mask);
+
+	xdev->common.device_alloc_chan_resources =
+				xilinx_vdma_alloc_chan_resources;
+	xdev->common.device_free_chan_resources =
+				xilinx_vdma_free_chan_resources;
+	xdev->common.device_prep_interleaved_dma =
+				xilinx_vdma_dma_prep_interleaved;
+	xdev->common.device_control = xilinx_vdma_device_control;
+	xdev->common.device_tx_status = xilinx_vdma_tx_status;
+	xdev->common.device_issue_pending = xilinx_vdma_issue_pending;
+
+	platform_set_drvdata(pdev, xdev);
+
+	/* Initialize the channels */
+	for_each_child_of_node(node, child) {
+		err = xilinx_vdma_chan_probe(xdev, child);
+		if (err < 0)
+			goto error;
+	}
+
+	for (i = 0; i < XILINX_VDMA_MAX_CHANS_PER_DEVICE; i++)
+		if (xdev->chan[i])
+			xdev->chan[i]->num_frms = num_frames;
+
+	/* Register the DMA engine with the core */
+	dma_async_device_register(&xdev->common);
+
+	err = of_dma_controller_register(node, of_dma_xilinx_xlate,
+					 xdev);
+	if (err < 0) {
+		dev_err(&pdev->dev, "Unable to register DMA to DT\n");
+		dma_async_device_unregister(&xdev->common);
+		goto error;
+	}
+
+	dev_info(&pdev->dev, "Xilinx AXI VDMA Engine Driver Probed!!\n");
+
+	return 0;
+
+error:
+	for (i = 0; i < XILINX_VDMA_MAX_CHANS_PER_DEVICE; i++)
+		if (xdev->chan[i])
+			xilinx_vdma_chan_remove(xdev->chan[i]);
+
+	return err;
+}
+
+/**
+ * xilinx_vdma_remove - Driver remove function
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: Always '0'
+ */
+static int xilinx_vdma_remove(struct platform_device *pdev)
+{
+	struct xilinx_vdma_device *xdev = platform_get_drvdata(pdev);
+	int i;
+
+	of_dma_controller_free(pdev->dev.of_node);
+
+	dma_async_device_unregister(&xdev->common);
+
+	for (i = 0; i < XILINX_VDMA_MAX_CHANS_PER_DEVICE; i++)
+		if (xdev->chan[i])
+			xilinx_vdma_chan_remove(xdev->chan[i]);
+
+	return 0;
+}
+
+static const struct of_device_id xilinx_vdma_of_ids[] = {
+	{ .compatible = "xlnx,axi-vdma-1.00.a",},
+	{}
+};
+
+static struct platform_driver xilinx_vdma_driver = {
+	.driver = {
+		.name = "xilinx-vdma",
+		.owner = THIS_MODULE,
+		.of_match_table = xilinx_vdma_of_ids,
+	},
+	.probe = xilinx_vdma_probe,
+	.remove = xilinx_vdma_remove,
+};
+
+module_platform_driver(xilinx_vdma_driver);
+
+MODULE_AUTHOR("Xilinx, Inc.");
+MODULE_DESCRIPTION("Xilinx VDMA driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/firmware/efi/efi-pstore.c b/drivers/firmware/efi/efi-pstore.c
index 4b9dc836dcf9..e992abc5ef26 100644
--- a/drivers/firmware/efi/efi-pstore.c
+++ b/drivers/firmware/efi/efi-pstore.c
@@ -40,7 +40,7 @@ struct pstore_read_data {
 static inline u64 generic_id(unsigned long timestamp,
 			     unsigned int part, int count)
 {
-	return (timestamp * 100 + part) * 1000 + count;
+	return ((u64) timestamp * 100 + part) * 1000 + count;
 }
 
 static int efi_pstore_read_func(struct efivar_entry *entry, void *data)
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index cd36deb619fa..eff1a2f22f09 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -353,10 +353,10 @@ static int __init fdt_find_uefi_params(unsigned long node, const char *uname,
 				       int depth, void *data)
 {
 	struct param_info *info = data;
-	void *prop, *dest;
-	unsigned long len;
+	const void *prop;
+	void *dest;
 	u64 val;
-	int i;
+	int i, len;
 
 	if (depth != 1 ||
 	    (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
diff --git a/drivers/firmware/efi/fdt.c b/drivers/firmware/efi/fdt.c
index 5c6a8e8a9580..82d774161cc9 100644
--- a/drivers/firmware/efi/fdt.c
+++ b/drivers/firmware/efi/fdt.c
@@ -63,7 +63,7 @@ static efi_status_t update_fdt(efi_system_table_t *sys_table, void *orig_fdt,
 	 */
 	prev = 0;
 	for (;;) {
-		const char *type, *name;
+		const char *type;
 		int len;
 
 		node = fdt_next_node(fdt, prev, NULL);
diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index d9c9cb4665db..2ebc9071e354 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -2614,7 +2614,7 @@ static struct gpio_desc *of_find_gpio(struct device *dev, const char *con_id,
 
 		desc = of_get_named_gpiod_flags(dev->of_node, prop_name, idx,
 						&of_flags);
-		if (!IS_ERR(desc))
+		if (!IS_ERR(desc) || (PTR_ERR(desc) == -EPROBE_DEFER))
 			break;
 	}
 
diff --git a/drivers/gpu/Makefile b/drivers/gpu/Makefile
index d8a22c2a579d..70da9eb52a42 100644
--- a/drivers/gpu/Makefile
+++ b/drivers/gpu/Makefile
@@ -1,2 +1,3 @@
 obj-y			+= drm/ vga/
 obj-$(CONFIG_TEGRA_HOST1X)	+= host1x/
+obj-$(CONFIG_IMX_IPUV3_CORE)	+= ipu-v3/
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index d1cc2f613a78..f5120046ff80 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -83,6 +83,8 @@ config DRM_KMS_CMA_HELPER
 
 source "drivers/gpu/drm/i2c/Kconfig"
 
+source "drivers/gpu/drm/bridge/Kconfig"
+
 config DRM_TDFX
 	tristate "3dfx Banshee/Voodoo3+"
 	depends on DRM && PCI
@@ -199,5 +201,3 @@ source "drivers/gpu/drm/msm/Kconfig"
 source "drivers/gpu/drm/tegra/Kconfig"
 
 source "drivers/gpu/drm/panel/Kconfig"
-
-source "drivers/gpu/drm/bridge/Kconfig"
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 48e38ba22783..dd2ba4269740 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -14,7 +14,7 @@ drm-y       :=	drm_auth.o drm_buffer.o drm_bufs.o drm_cache.o \
 		drm_info.o drm_debugfs.o drm_encoder_slave.o \
 		drm_trace_points.o drm_global.o drm_prime.o \
 		drm_rect.o drm_vma_manager.o drm_flip_work.o \
-		drm_plane_helper.o
+		drm_modeset_lock.o
 
 drm-$(CONFIG_COMPAT) += drm_ioc32.o
 drm-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_gem_cma_helper.o
@@ -23,7 +23,8 @@ drm-$(CONFIG_DRM_PANEL) += drm_panel.o
 
 drm-usb-y   := drm_usb.o
 
-drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o
+drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \
+		drm_plane_helper.o
 drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o
 drm_kms_helper-$(CONFIG_DRM_KMS_FB_HELPER) += drm_fb_helper.o
 drm_kms_helper-$(CONFIG_DRM_KMS_CMA_HELPER) += drm_fb_cma_helper.o
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 32982da82694..8ab3cd1a8cdb 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -173,7 +173,7 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 	if (ret)
 		goto err_kms;
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	if (ret)
 		goto err_kms;
 
@@ -402,7 +402,7 @@ static struct platform_driver armada_drm_platform_driver = {
 
 static int __init armada_drm_init(void)
 {
-	armada_drm_driver.num_ioctls = DRM_ARRAY_SIZE(armada_ioctls);
+	armada_drm_driver.num_ioctls = ARRAY_SIZE(armada_ioctls);
 	return platform_driver_register(&armada_drm_platform_driver);
 }
 module_init(armada_drm_init);
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 948cb14c561e..fd166f532ab9 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -181,10 +181,8 @@ void armada_fbdev_lastclose(struct drm_device *dev)
 {
 	struct armada_private *priv = dev->dev_private;
 
-	drm_modeset_lock_all(dev);
 	if (priv->fbdev)
-		drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-	drm_modeset_unlock_all(dev);
+		drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 }
 
 void armada_fbdev_fini(struct drm_device *dev)
diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c
index 887816f43476..bb9b642d8485 100644
--- a/drivers/gpu/drm/armada/armada_gem.c
+++ b/drivers/gpu/drm/armada/armada_gem.c
@@ -433,7 +433,6 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 
 	if (dobj->obj.filp) {
 		struct address_space *mapping;
-		gfp_t gfp;
 		int count;
 
 		count = dobj->obj.size / PAGE_SIZE;
@@ -441,12 +440,11 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 			goto free_sgt;
 
 		mapping = file_inode(dobj->obj.filp)->i_mapping;
-		gfp = mapping_gfp_mask(mapping);
 
 		for_each_sg(sgt->sgl, sg, count, i) {
 			struct page *page;
 
-			page = shmem_read_mapping_page_gfp(mapping, i, gfp);
+			page = shmem_read_mapping_page(mapping, i);
 			if (IS_ERR(page)) {
 				num = i;
 				goto release;
diff --git a/drivers/gpu/drm/ast/Makefile b/drivers/gpu/drm/ast/Makefile
index 8df4f284ee24..171aa0622b66 100644
--- a/drivers/gpu/drm/ast/Makefile
+++ b/drivers/gpu/drm/ast/Makefile
@@ -4,6 +4,6 @@
 
 ccflags-y := -Iinclude/drm
 
-ast-y := ast_drv.o ast_main.o ast_mode.o ast_fb.o ast_ttm.o ast_post.o
+ast-y := ast_drv.o ast_main.o ast_mode.o ast_fb.o ast_ttm.o ast_post.o ast_dp501.o
 
-obj-$(CONFIG_DRM_AST) := ast.o
\ No newline at end of file
+obj-$(CONFIG_DRM_AST) := ast.o
diff --git a/drivers/gpu/drm/ast/ast_dp501.c b/drivers/gpu/drm/ast/ast_dp501.c
new file mode 100644
index 000000000000..5da4b62285fa
--- /dev/null
+++ b/drivers/gpu/drm/ast/ast_dp501.c
@@ -0,0 +1,410 @@
+
+#include <linux/firmware.h>
+#include <drm/drmP.h>
+#include "ast_drv.h"
+MODULE_FIRMWARE("ast_dp501_fw.bin");
+
+int ast_load_dp501_microcode(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	static char *fw_name = "ast_dp501_fw.bin";
+	int err;
+	err = request_firmware(&ast->dp501_fw, fw_name, dev->dev);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void send_ack(struct ast_private *ast)
+{
+	u8 sendack;
+	sendack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0xff);
+	sendack |= 0x80;
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0x00, sendack);
+}
+
+static void send_nack(struct ast_private *ast)
+{
+	u8 sendack;
+	sendack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0xff);
+	sendack &= ~0x80;
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0x00, sendack);
+}
+
+static bool wait_ack(struct ast_private *ast)
+{
+	u8 waitack;
+	u32 retry = 0;
+	do {
+		waitack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd2, 0xff);
+		waitack &= 0x80;
+		udelay(100);
+	} while ((!waitack) && (retry++ < 1000));
+
+	if (retry < 1000)
+		return true;
+	else
+		return false;
+}
+
+static bool wait_nack(struct ast_private *ast)
+{
+	u8 waitack;
+	u32 retry = 0;
+	do {
+		waitack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd2, 0xff);
+		waitack &= 0x80;
+		udelay(100);
+	} while ((waitack) && (retry++ < 1000));
+
+	if (retry < 1000)
+		return true;
+	else
+		return false;
+}
+
+static void set_cmd_trigger(struct ast_private *ast)
+{
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, ~0x40, 0x40);
+}
+
+static void clear_cmd_trigger(struct ast_private *ast)
+{
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, ~0x40, 0x00);
+}
+
+#if 0
+static bool wait_fw_ready(struct ast_private *ast)
+{
+	u8 waitready;
+	u32 retry = 0;
+	do {
+		waitready = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd2, 0xff);
+		waitready &= 0x40;
+		udelay(100);
+	} while ((!waitready) && (retry++ < 1000));
+
+	if (retry < 1000)
+		return true;
+	else
+		return false;
+}
+#endif
+
+static bool ast_write_cmd(struct drm_device *dev, u8 data)
+{
+	struct ast_private *ast = dev->dev_private;
+	int retry = 0;
+	if (wait_nack(ast)) {
+		send_nack(ast);
+		ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9a, 0x00, data);
+		send_ack(ast);
+		set_cmd_trigger(ast);
+		do {
+			if (wait_ack(ast)) {
+				clear_cmd_trigger(ast);
+				send_nack(ast);
+				return true;
+			}
+		} while (retry++ < 100);
+	}
+	clear_cmd_trigger(ast);
+	send_nack(ast);
+	return false;
+}
+
+static bool ast_write_data(struct drm_device *dev, u8 data)
+{
+	struct ast_private *ast = dev->dev_private;
+
+	if (wait_nack(ast)) {
+		send_nack(ast);
+		ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9a, 0x00, data);
+		send_ack(ast);
+		if (wait_ack(ast)) {
+			send_nack(ast);
+			return true;
+		}
+	}
+	send_nack(ast);
+	return false;
+}
+
+#if 0
+static bool ast_read_data(struct drm_device *dev, u8 *data)
+{
+	struct ast_private *ast = dev->dev_private;
+	u8 tmp;
+
+	*data = 0;
+
+	if (wait_ack(ast) == false)
+		return false;
+	tmp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd3, 0xff);
+	*data = tmp;
+	if (wait_nack(ast) == false) {
+		send_nack(ast);
+		return false;
+	}
+	send_nack(ast);
+	return true;
+}
+
+static void clear_cmd(struct ast_private *ast)
+{
+	send_nack(ast);
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9a, 0x00, 0x00);
+}
+#endif
+
+void ast_set_dp501_video_output(struct drm_device *dev, u8 mode)
+{
+	ast_write_cmd(dev, 0x40);
+	ast_write_data(dev, mode);
+
+	msleep(10);
+}
+
+static u32 get_fw_base(struct ast_private *ast)
+{
+	return ast_mindwm(ast, 0x1e6e2104) & 0x7fffffff;
+}
+
+bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 i, data;
+	u32 boot_address;
+
+	data = ast_mindwm(ast, 0x1e6e2100) & 0x01;
+	if (data) {
+		boot_address = get_fw_base(ast);
+		for (i = 0; i < size; i += 4)
+			*(u32 *)(addr + i) = ast_mindwm(ast, boot_address + i);
+		return true;
+	}
+	return false;
+}
+
+bool ast_launch_m68k(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 i, data, len = 0;
+	u32 boot_address;
+	u8 *fw_addr = NULL;
+	u8 jreg;
+
+	data = ast_mindwm(ast, 0x1e6e2100) & 0x01;
+	if (!data) {
+
+		if (ast->dp501_fw_addr) {
+			fw_addr = ast->dp501_fw_addr;
+			len = 32*1024;
+		} else if (ast->dp501_fw) {
+			fw_addr = (u8 *)ast->dp501_fw->data;
+			len = ast->dp501_fw->size;
+		}
+		/* Get BootAddress */
+		ast_moutdwm(ast, 0x1e6e2000, 0x1688a8a8);
+		data = ast_mindwm(ast, 0x1e6e0004);
+		switch (data & 0x03) {
+		case 0:
+			boot_address = 0x44000000;
+			break;
+		default:
+		case 1:
+			boot_address = 0x48000000;
+			break;
+		case 2:
+			boot_address = 0x50000000;
+			break;
+		case 3:
+			boot_address = 0x60000000;
+			break;
+		}
+		boot_address -= 0x200000; /* -2MB */
+
+		/* copy image to buffer */
+		for (i = 0; i < len; i += 4) {
+			data = *(u32 *)(fw_addr + i);
+			ast_moutdwm(ast, boot_address + i, data);
+		}
+
+		/* Init SCU */
+		ast_moutdwm(ast, 0x1e6e2000, 0x1688a8a8);
+
+		/* Launch FW */
+		ast_moutdwm(ast, 0x1e6e2104, 0x80000000 + boot_address);
+		ast_moutdwm(ast, 0x1e6e2100, 1);
+
+		/* Update Scratch */
+		data = ast_mindwm(ast, 0x1e6e2040) & 0xfffff1ff;		/* D[11:9] = 100b: UEFI handling */
+		data |= 0x800;
+		ast_moutdwm(ast, 0x1e6e2040, data);
+
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x99, 0xfc); /* D[1:0]: Reserved Video Buffer */
+		jreg |= 0x02;
+		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x99, jreg);
+	}
+	return true;
+}
+
+u8 ast_get_dp501_max_clk(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 boot_address, offset, data;
+	u8 linkcap[4], linkrate, linklanes, maxclk = 0xff;
+
+	boot_address = get_fw_base(ast);
+
+	/* validate FW version */
+	offset = 0xf000;
+	data = ast_mindwm(ast, boot_address + offset);
+	if ((data & 0xf0) != 0x10) /* version: 1x */
+		return maxclk;
+
+	/* Read Link Capability */
+	offset  = 0xf014;
+	*(u32 *)linkcap = ast_mindwm(ast, boot_address + offset);
+	if (linkcap[2] == 0) {
+		linkrate = linkcap[0];
+		linklanes = linkcap[1];
+		data = (linkrate == 0x0a) ? (90 * linklanes) : (54 * linklanes);
+		if (data > 0xff)
+			data = 0xff;
+		maxclk = (u8)data;
+	}
+	return maxclk;
+}
+
+bool ast_dp501_read_edid(struct drm_device *dev, u8 *ediddata)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 i, boot_address, offset, data;
+
+	boot_address = get_fw_base(ast);
+
+	/* validate FW version */
+	offset = 0xf000;
+	data = ast_mindwm(ast, boot_address + offset);
+	if ((data & 0xf0) != 0x10)
+		return false;
+
+	/* validate PnP Monitor */
+	offset = 0xf010;
+	data = ast_mindwm(ast, boot_address + offset);
+	if (!(data & 0x01))
+		return false;
+
+	/* Read EDID */
+	offset = 0xf020;
+	for (i = 0; i < 128; i += 4) {
+		data = ast_mindwm(ast, boot_address + offset + i);
+		*(u32 *)(ediddata + i) = data;
+	}
+
+	return true;
+}
+
+static bool ast_init_dvo(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u8 jreg;
+	u32 data;
+	ast_write32(ast, 0xf004, 0x1e6e0000);
+	ast_write32(ast, 0xf000, 0x1);
+	ast_write32(ast, 0x12000, 0x1688a8a8);
+
+	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd0, 0xff);
+	if (!(jreg & 0x80)) {
+		/* Init SCU DVO Settings */
+		data = ast_read32(ast, 0x12008);
+		/* delay phase */
+		data &= 0xfffff8ff;
+		data |= 0x00000500;
+		ast_write32(ast, 0x12008, data);
+
+		if (ast->chip == AST2300) {
+			data = ast_read32(ast, 0x12084);
+			/* multi-pins for DVO single-edge */
+			data |= 0xfffe0000;
+			ast_write32(ast, 0x12084, data);
+
+			data = ast_read32(ast, 0x12088);
+			/* multi-pins for DVO single-edge */
+			data |= 0x000fffff;
+			ast_write32(ast, 0x12088, data);
+
+			data = ast_read32(ast, 0x12090);
+			/* multi-pins for DVO single-edge */
+			data &= 0xffffffcf;
+			data |= 0x00000020;
+			ast_write32(ast, 0x12090, data);
+		} else { /* AST2400 */
+			data = ast_read32(ast, 0x12088);
+			/* multi-pins for DVO single-edge */
+			data |= 0x30000000;
+			ast_write32(ast, 0x12088, data);
+
+			data = ast_read32(ast, 0x1208c);
+			/* multi-pins for DVO single-edge */
+			data |= 0x000000cf;
+			ast_write32(ast, 0x1208c, data);
+
+			data = ast_read32(ast, 0x120a4);
+			/* multi-pins for DVO single-edge */
+			data |= 0xffff0000;
+			ast_write32(ast, 0x120a4, data);
+
+			data = ast_read32(ast, 0x120a8);
+			/* multi-pins for DVO single-edge */
+			data |= 0x0000000f;
+			ast_write32(ast, 0x120a8, data);
+
+			data = ast_read32(ast, 0x12094);
+			/* multi-pins for DVO single-edge */
+			data |= 0x00000002;
+			ast_write32(ast, 0x12094, data);
+		}
+	}
+
+	/* Force to DVO */
+	data = ast_read32(ast, 0x1202c);
+	data &= 0xfffbffff;
+	ast_write32(ast, 0x1202c, data);
+
+	/* Init VGA DVO Settings */
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa3, 0xcf, 0x80);
+	return true;
+}
+
+void ast_init_3rdtx(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u8 jreg;
+	u32 data;
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd1, 0xff);
+		switch (jreg & 0x0e) {
+		case 0x04:
+			ast_init_dvo(dev);
+			break;
+		case 0x08:
+			ast_launch_m68k(dev);
+			break;
+		case 0x0c:
+			ast_init_dvo(dev);
+			break;
+		default:
+			if (ast->tx_chip_type == AST_TX_SIL164)
+				ast_init_dvo(dev);
+			else {
+				ast_write32(ast, 0x12000, 0x1688a8a8);
+				data = ast_read32(ast, 0x1202c);
+				data &= 0xfffcffff;
+				ast_write32(ast, 0, data);
+			}
+		}
+	}
+}
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index 5137f15dba19..44074fbcf7ff 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -94,9 +94,7 @@ static int ast_drm_thaw(struct drm_device *dev)
 	ast_post_gpu(dev);
 
 	drm_mode_config_reset(dev);
-	drm_modeset_lock_all(dev);
 	drm_helper_resume_force_mode(dev);
-	drm_modeset_unlock_all(dev);
 
 	console_lock();
 	ast_fbdev_set_suspend(dev, 0);
@@ -198,7 +196,6 @@ static const struct file_operations ast_fops = {
 
 static struct drm_driver driver = {
 	.driver_features = DRIVER_MODESET | DRIVER_GEM,
-	.dev_priv_size = 0,
 
 	.load = ast_driver_load,
 	.unload = ast_driver_unload,
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 9833a1b1acc1..5d6a87573c33 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -61,9 +61,17 @@ enum ast_chip {
 	AST2200,
 	AST2150,
 	AST2300,
+	AST2400,
 	AST1180,
 };
 
+enum ast_tx_chip {
+	AST_TX_NONE,
+	AST_TX_SIL164,
+	AST_TX_ITE66121,
+	AST_TX_DP501,
+};
+
 #define AST_DRAM_512Mx16 0
 #define AST_DRAM_1Gx16   1
 #define AST_DRAM_512Mx32 2
@@ -102,6 +110,12 @@ struct ast_private {
 	 * we have. */
 	struct ttm_bo_kmap_obj cache_kmap;
 	int next_cursor;
+	bool support_wide_screen;
+
+	enum ast_tx_chip tx_chip_type;
+	u8 dp501_maxclk;
+	u8 *dp501_fw_addr;
+	const struct firmware *dp501_fw;	/* dp501 fw */
 };
 
 int ast_driver_load(struct drm_device *dev, unsigned long flags);
@@ -368,4 +382,14 @@ int ast_mmap(struct file *filp, struct vm_area_struct *vma);
 
 /* ast post */
 void ast_post_gpu(struct drm_device *dev);
+u32 ast_mindwm(struct ast_private *ast, u32 r);
+void ast_moutdwm(struct ast_private *ast, u32 r, u32 v);
+/* ast dp501 */
+int ast_load_dp501_microcode(struct drm_device *dev);
+void ast_set_dp501_video_output(struct drm_device *dev, u8 mode);
+bool ast_launch_m68k(struct drm_device *dev);
+bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size);
+bool ast_dp501_read_edid(struct drm_device *dev, u8 *ediddata);
+u8 ast_get_dp501_max_clk(struct drm_device *dev);
+void ast_init_3rdtx(struct drm_device *dev);
 #endif
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 50535fd5a88d..a2cc6be97983 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -66,12 +66,16 @@ uint8_t ast_get_index_reg_mask(struct ast_private *ast,
 static int ast_detect_chip(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
+	uint32_t data, jreg;
 
 	if (dev->pdev->device == PCI_CHIP_AST1180) {
 		ast->chip = AST1100;
 		DRM_INFO("AST 1180 detected\n");
 	} else {
-		if (dev->pdev->revision >= 0x20) {
+		if (dev->pdev->revision >= 0x30) {
+			ast->chip = AST2400;
+			DRM_INFO("AST 2400 detected\n");
+		} else if (dev->pdev->revision >= 0x20) {
 			ast->chip = AST2300;
 			DRM_INFO("AST 2300 detected\n");
 		} else if (dev->pdev->revision >= 0x10) {
@@ -104,6 +108,59 @@ static int ast_detect_chip(struct drm_device *dev)
 			DRM_INFO("AST 2000 detected\n");
 		}
 	}
+
+	switch (ast->chip) {
+	case AST1180:
+		ast->support_wide_screen = true;
+		break;
+	case AST2000:
+		ast->support_wide_screen = false;
+		break;
+	default:
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd0, 0xff);
+		if (!(jreg & 0x80))
+			ast->support_wide_screen = true;
+		else if (jreg & 0x01)
+			ast->support_wide_screen = true;
+		else {
+			ast->support_wide_screen = false;
+			ast_write32(ast, 0xf004, 0x1e6e0000);
+			ast_write32(ast, 0xf000, 0x1);
+			data = ast_read32(ast, 0x1207c);
+			data &= 0x300;
+			if (ast->chip == AST2300 && data == 0x0) /* ast1300 */
+				ast->support_wide_screen = true;
+			if (ast->chip == AST2400 && data == 0x100) /* ast1400 */
+				ast->support_wide_screen = true;
+		}
+		break;
+	}
+
+	ast->tx_chip_type = AST_TX_NONE;
+	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa3, 0xff);
+	if (jreg & 0x80)
+		ast->tx_chip_type = AST_TX_SIL164;
+	if ((ast->chip == AST2300) || (ast->chip == AST2400)) {
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd1, 0xff);
+		switch (jreg) {
+		case 0x04:
+			ast->tx_chip_type = AST_TX_SIL164;
+			break;
+		case 0x08:
+			ast->dp501_fw_addr = kzalloc(32*1024, GFP_KERNEL);
+			if (ast->dp501_fw_addr) {
+				/* backup firmware */
+				if (ast_backup_fw(dev, ast->dp501_fw_addr, 32*1024)) {
+					kfree(ast->dp501_fw_addr);
+					ast->dp501_fw_addr = NULL;
+				}
+			}
+			/* fallthrough */
+		case 0x0c:
+			ast->tx_chip_type = AST_TX_DP501;
+		}
+	}
+
 	return 0;
 }
 
@@ -129,7 +186,7 @@ static int ast_get_dram_info(struct drm_device *dev)
 	else
 		ast->dram_bus_width = 32;
 
-	if (ast->chip == AST2300) {
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
 		switch (data & 0x03) {
 		case 0:
 			ast->dram_type = AST_DRAM_512Mx16;
@@ -257,17 +314,32 @@ static u32 ast_get_vram_info(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
 	u8 jreg;
-
+	u32 vram_size;
 	ast_open_key(ast);
 
+	vram_size = AST_VIDMEM_DEFAULT_SIZE;
 	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xaa, 0xff);
 	switch (jreg & 3) {
-	case 0: return AST_VIDMEM_SIZE_8M;
-	case 1: return AST_VIDMEM_SIZE_16M;
-	case 2: return AST_VIDMEM_SIZE_32M;
-	case 3: return AST_VIDMEM_SIZE_64M;
+	case 0: vram_size = AST_VIDMEM_SIZE_8M; break;
+	case 1: vram_size = AST_VIDMEM_SIZE_16M; break;
+	case 2: vram_size = AST_VIDMEM_SIZE_32M; break;
+	case 3: vram_size = AST_VIDMEM_SIZE_64M; break;
+	}
+
+	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x99, 0xff);
+	switch (jreg & 0x03) {
+	case 1:
+		vram_size -= 0x100000;
+		break;
+	case 2:
+		vram_size -= 0x200000;
+		break;
+	case 3:
+		vram_size -= 0x400000;
+		break;
 	}
-	return AST_VIDMEM_DEFAULT_SIZE;
+
+	return vram_size;
 }
 
 int ast_driver_load(struct drm_device *dev, unsigned long flags)
@@ -316,6 +388,7 @@ int ast_driver_load(struct drm_device *dev, unsigned long flags)
 	if (ast->chip == AST2100 ||
 	    ast->chip == AST2200 ||
 	    ast->chip == AST2300 ||
+	    ast->chip == AST2400 ||
 	    ast->chip == AST1180) {
 		dev->mode_config.max_width = 1920;
 		dev->mode_config.max_height = 2048;
@@ -343,6 +416,7 @@ int ast_driver_unload(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
 
+	kfree(ast->dp501_fw_addr);
 	ast_mode_fini(dev);
 	ast_fbdev_fini(dev);
 	drm_mode_config_cleanup(dev);
@@ -411,16 +485,13 @@ static void ast_bo_unref(struct ast_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
+
 void ast_gem_free_object(struct drm_gem_object *obj)
 {
 	struct ast_bo *ast_bo = gem_to_ast_bo(obj);
 
-	if (!ast_bo)
-		return;
 	ast_bo_unref(&ast_bo);
 }
 
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index a4afdc8bb578..114aee941d46 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -115,11 +115,17 @@ static bool ast_get_vbios_mode_info(struct drm_crtc *crtc, struct drm_display_mo
 		else
 			vbios_mode->enh_table = &res_1280x1024[refresh_rate_index];
 		break;
+	case 1360:
+		vbios_mode->enh_table = &res_1360x768[refresh_rate_index];
+		break;
 	case 1440:
 		vbios_mode->enh_table = &res_1440x900[refresh_rate_index];
 		break;
 	case 1600:
-		vbios_mode->enh_table = &res_1600x1200[refresh_rate_index];
+		if (crtc->mode.crtc_vdisplay == 900)
+			vbios_mode->enh_table = &res_1600x900[refresh_rate_index];
+		else
+			vbios_mode->enh_table = &res_1600x1200[refresh_rate_index];
 		break;
 	case 1680:
 		vbios_mode->enh_table = &res_1680x1050[refresh_rate_index];
@@ -175,14 +181,17 @@ static bool ast_get_vbios_mode_info(struct drm_crtc *crtc, struct drm_display_mo
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x8d, refresh_rate_index & 0xff);
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x8e, mode_id & 0xff);
 
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x91, 0xa8);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x92, crtc->primary->fb->bits_per_pixel);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x93, adjusted_mode->clock / 1000);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x94, adjusted_mode->crtc_hdisplay);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x95, adjusted_mode->crtc_hdisplay >> 8);
+		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x91, 0x00);
+		if (vbios_mode->enh_table->flags & NewModeInfo) {
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x91, 0xa8);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x92, crtc->primary->fb->bits_per_pixel);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x93, adjusted_mode->clock / 1000);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x94, adjusted_mode->crtc_hdisplay);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x95, adjusted_mode->crtc_hdisplay >> 8);
 
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x96, adjusted_mode->crtc_vdisplay);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x97, adjusted_mode->crtc_vdisplay >> 8);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x96, adjusted_mode->crtc_vdisplay);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x97, adjusted_mode->crtc_vdisplay >> 8);
+		}
 	}
 
 	return true;
@@ -389,7 +398,7 @@ static void ast_set_ext_reg(struct drm_crtc *crtc, struct drm_display_mode *mode
 	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa8, 0xfd, jregA8);
 
 	/* Set Threshold */
-	if (ast->chip == AST2300) {
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa7, 0x78);
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa6, 0x60);
 	} else if (ast->chip == AST2100 ||
@@ -451,9 +460,13 @@ static void ast_crtc_dpms(struct drm_crtc *crtc, int mode)
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 		ast_set_index_reg_mask(ast, AST_IO_SEQ_PORT, 0x1, 0xdf, 0);
+		if (ast->tx_chip_type == AST_TX_DP501)
+			ast_set_dp501_video_output(crtc->dev, 1);
 		ast_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_OFF:
+		if (ast->tx_chip_type == AST_TX_DP501)
+			ast_set_dp501_video_output(crtc->dev, 0);
 		ast_set_index_reg_mask(ast, AST_IO_SEQ_PORT, 0x1, 0xdf, 0x20);
 		break;
 	}
@@ -729,10 +742,24 @@ static int ast_encoder_init(struct drm_device *dev)
 static int ast_get_modes(struct drm_connector *connector)
 {
 	struct ast_connector *ast_connector = to_ast_connector(connector);
+	struct ast_private *ast = connector->dev->dev_private;
 	struct edid *edid;
 	int ret;
-
-	edid = drm_get_edid(connector, &ast_connector->i2c->adapter);
+	bool flags = false;
+	if (ast->tx_chip_type == AST_TX_DP501) {
+		ast->dp501_maxclk = 0xff;
+		edid = kmalloc(128, GFP_KERNEL);
+		if (!edid)
+			return -ENOMEM;
+
+		flags = ast_dp501_read_edid(connector->dev, (u8 *)edid);
+		if (flags)
+			ast->dp501_maxclk = ast_get_dp501_max_clk(connector->dev);
+		else
+			kfree(edid);
+	}
+	if (!flags)
+		edid = drm_get_edid(connector, &ast_connector->i2c->adapter);
 	if (edid) {
 		drm_mode_connector_update_edid_property(&ast_connector->base, edid);
 		ret = drm_add_edid_modes(connector, edid);
@@ -746,7 +773,56 @@ static int ast_get_modes(struct drm_connector *connector)
 static int ast_mode_valid(struct drm_connector *connector,
 			  struct drm_display_mode *mode)
 {
-	return MODE_OK;
+	struct ast_private *ast = connector->dev->dev_private;
+	int flags = MODE_NOMODE;
+	uint32_t jtemp;
+
+	if (ast->support_wide_screen) {
+		if ((mode->hdisplay == 1680) && (mode->vdisplay == 1050))
+			return MODE_OK;
+		if ((mode->hdisplay == 1280) && (mode->vdisplay == 800))
+			return MODE_OK;
+		if ((mode->hdisplay == 1440) && (mode->vdisplay == 900))
+			return MODE_OK;
+		if ((mode->hdisplay == 1360) && (mode->vdisplay == 768))
+			return MODE_OK;
+		if ((mode->hdisplay == 1600) && (mode->vdisplay == 900))
+			return MODE_OK;
+
+		if ((ast->chip == AST2100) || (ast->chip == AST2200) || (ast->chip == AST2300) || (ast->chip == AST2400) || (ast->chip == AST1180)) {
+			if ((mode->hdisplay == 1920) && (mode->vdisplay == 1080))
+				return MODE_OK;
+
+			if ((mode->hdisplay == 1920) && (mode->vdisplay == 1200)) {
+				jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd1, 0xff);
+				if (jtemp & 0x01)
+					return MODE_NOMODE;
+				else
+					return MODE_OK;
+			}
+		}
+	}
+	switch (mode->hdisplay) {
+	case 640:
+		if (mode->vdisplay == 480) flags = MODE_OK;
+		break;
+	case 800:
+		if (mode->vdisplay == 600) flags = MODE_OK;
+		break;
+	case 1024:
+		if (mode->vdisplay == 768) flags = MODE_OK;
+		break;
+	case 1280:
+		if (mode->vdisplay == 1024) flags = MODE_OK;
+		break;
+	case 1600:
+		if (mode->vdisplay == 1200) flags = MODE_OK;
+		break;
+	default:
+		return flags;
+	}
+
+	return flags;
 }
 
 static void ast_connector_destroy(struct drm_connector *connector)
diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c
index 635f6ffc27c2..38d437f3a267 100644
--- a/drivers/gpu/drm/ast/ast_post.c
+++ b/drivers/gpu/drm/ast/ast_post.c
@@ -78,7 +78,7 @@ ast_set_def_ext_reg(struct drm_device *dev)
 	for (i = 0x81; i <= 0x8f; i++)
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, i, 0x00);
 
-	if (ast->chip == AST2300) {
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
 		if (dev->pdev->revision >= 0x20)
 			ext_reg_info = extreginfo_ast2300;
 		else
@@ -102,23 +102,32 @@ ast_set_def_ext_reg(struct drm_device *dev)
 
 	/* Enable RAMDAC for A1 */
 	reg = 0x04;
-	if (ast->chip == AST2300)
+	if (ast->chip == AST2300 || ast->chip == AST2400)
 		reg |= 0x20;
 	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb6, 0xff, reg);
 }
 
-static inline u32 mindwm(struct ast_private *ast, u32 r)
+u32 ast_mindwm(struct ast_private *ast, u32 r)
 {
+	uint32_t data;
+
 	ast_write32(ast, 0xf004, r & 0xffff0000);
 	ast_write32(ast, 0xf000, 0x1);
 
+	do {
+		data = ast_read32(ast, 0xf004) & 0xffff0000;
+	} while (data != (r & 0xffff0000));
 	return ast_read32(ast, 0x10000 + (r & 0x0000ffff));
 }
 
-static inline void moutdwm(struct ast_private *ast, u32 r, u32 v)
+void ast_moutdwm(struct ast_private *ast, u32 r, u32 v)
 {
+	uint32_t data;
 	ast_write32(ast, 0xf004, r & 0xffff0000);
 	ast_write32(ast, 0xf000, 0x1);
+	do {
+		data = ast_read32(ast, 0xf004) & 0xffff0000;
+	} while (data != (r & 0xffff0000));
 	ast_write32(ast, 0x10000 + (r & 0x0000ffff), v);
 }
 
@@ -154,28 +163,28 @@ static u32 mmctestburst2_ast2150(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000001 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000001 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x40;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x40;
 		if (++timeout > TIMEOUT_AST2150) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0xffffffff;
 		}
 	} while (!data);
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000003 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000003 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x40;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x40;
 		if (++timeout > TIMEOUT_AST2150) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0xffffffff;
 		}
 	} while (!data);
-	data = (mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
+	data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 	return data;
 }
 
@@ -184,18 +193,18 @@ static u32 mmctestsingle2_ast2150(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x40;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x40;
 		if (++timeout > TIMEOUT_AST2150) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0xffffffff;
 		}
 	} while (!data);
-	data = (mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
+	data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 	return data;
 }
 #endif
@@ -215,7 +224,7 @@ static int cbrscan_ast2150(struct ast_private *ast, int busw)
 	u32 patcnt, loop;
 
 	for (patcnt = 0; patcnt < CBR_PATNUM_AST2150; patcnt++) {
-		moutdwm(ast, 0x1e6e007c, pattern_AST2150[patcnt]);
+		ast_moutdwm(ast, 0x1e6e007c, pattern_AST2150[patcnt]);
 		for (loop = 0; loop < CBR_PASSNUM_AST2150; loop++) {
 			if (cbrtest_ast2150(ast))
 				break;
@@ -237,7 +246,7 @@ cbr_start:
 	passcnt = 0;
 
 	for (dlli = 0; dlli < 100; dlli++) {
-		moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
+		ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
 		data = cbrscan_ast2150(ast, busw);
 		if (data != 0) {
 			if (data & 0x1) {
@@ -254,7 +263,7 @@ cbr_start:
 		goto cbr_start;
 
 	dlli = dll_min[0] + (((dll_max[0] - dll_min[0]) * 7) >> 4);
-	moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
+	ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
 }
 
 
@@ -365,10 +374,12 @@ void ast_post_gpu(struct drm_device *dev)
 	ast_open_key(ast);
 	ast_set_def_ext_reg(dev);
 
-	if (ast->chip == AST2300)
+	if (ast->chip == AST2300 || ast->chip == AST2400)
 		ast_init_dram_2300(dev);
 	else
 		ast_init_dram_reg(dev);
+
+	ast_init_3rdtx(dev);
 }
 
 /* AST 2300 DRAM settings */
@@ -403,6 +414,7 @@ struct ast2300_dram_param {
 /*
  * DQSI DLL CBR Setting
  */
+#define CBR_SIZE0            ((1  << 10) - 1)
 #define CBR_SIZE1            ((4  << 10) - 1)
 #define CBR_SIZE2            ((64 << 10) - 1)
 #define CBR_PASSNUM          5
@@ -423,88 +435,84 @@ static const u32 pattern[8] = {
 	0x7C61D253
 };
 
-#if 0 /* unused in DDX, included for completeness */
 static int mmc_test_burst(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x000000c1 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x000000c1 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x3000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x3000;
 		if (data & 0x2000) {
 			return 0;
 		}
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0;
 		}
 	} while (!data);
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 	return 1;
 }
-#endif
 
 static int mmc_test_burst2(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000041 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000041 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x1000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x1000;
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x0);
+			ast_moutdwm(ast, 0x1e6e0070, 0x0);
 			return -1;
 		}
 	} while (!data);
-	data = mindwm(ast, 0x1e6e0078);
+	data = ast_mindwm(ast, 0x1e6e0078);
 	data = (data | (data >> 16)) & 0xffff;
-	moutdwm(ast, 0x1e6e0070, 0x0);
+	ast_moutdwm(ast, 0x1e6e0070, 0x0);
 	return data;
 }
 
-#if 0 /* Unused in DDX here for completeness */
 static int mmc_test_single(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x000000c5 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x000000c5 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x3000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x3000;
 		if (data & 0x2000)
 			return 0;
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x0);
+			ast_moutdwm(ast, 0x1e6e0070, 0x0);
 			return 0;
 		}
 	} while (!data);
-	moutdwm(ast, 0x1e6e0070, 0x0);
+	ast_moutdwm(ast, 0x1e6e0070, 0x0);
 	return 1;
 }
-#endif
 
 static int mmc_test_single2(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x1000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x1000;
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x0);
+			ast_moutdwm(ast, 0x1e6e0070, 0x0);
 			return -1;
 		}
 	} while (!data);
-	data = mindwm(ast, 0x1e6e0078);
+	data = ast_mindwm(ast, 0x1e6e0078);
 	data = (data | (data >> 16)) & 0xffff;
-	moutdwm(ast, 0x1e6e0070, 0x0);
+	ast_moutdwm(ast, 0x1e6e0070, 0x0);
 	return data;
 }
 
@@ -533,7 +541,7 @@ static int cbr_scan(struct ast_private *ast)
 
 	data2 = 3;
 	for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) {
-		moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
+		ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
 		for (loop = 0; loop < CBR_PASSNUM2; loop++) {
 			if ((data = cbr_test(ast)) != 0) {
 				data2 &= data;
@@ -568,7 +576,7 @@ static u32 cbr_scan2(struct ast_private *ast)
 
 	data2 = 0xffff;
 	for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) {
-		moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
+		ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
 		for (loop = 0; loop < CBR_PASSNUM2; loop++) {
 			if ((data = cbr_test2(ast)) != 0) {
 				data2 &= data;
@@ -583,106 +591,35 @@ static u32 cbr_scan2(struct ast_private *ast)
 	return data2;
 }
 
-#if 0 /* unused in DDX - added for completeness */
-static void finetuneDQI(struct ast_private *ast, struct ast2300_dram_param *param)
+static u32 cbr_test3(struct ast_private *ast)
 {
-	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt;
-
-	gold_sadj[0] = (mindwm(ast, 0x1E6E0024) >> 16) & 0xffff;
-	gold_sadj[1] = gold_sadj[0] >> 8;
-	gold_sadj[0] = gold_sadj[0] & 0xff;
-	gold_sadj[0] = (gold_sadj[0] + gold_sadj[1]) >> 1;
-	gold_sadj[1] = gold_sadj[0];
-
-	for (cnt = 0; cnt < 16; cnt++) {
-		dllmin[cnt] = 0xff;
-		dllmax[cnt] = 0x0;
-	}
-	passcnt = 0;
-	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
+	if (!mmc_test_burst(ast, 0))
+		return 0;
+	if (!mmc_test_single(ast, 0))
+		return 0;
+	return 1;
+}
 
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE1);
-		data = cbr_scan2(ast);
-		if (data != 0) {
-			mask = 0x00010001;
-			for (cnt = 0; cnt < 16; cnt++) {
-				if (data & mask) {
-					if (dllmin[cnt] > dlli) {
-						dllmin[cnt] = dlli;
-					}
-					if (dllmax[cnt] < dlli) {
-						dllmax[cnt] = dlli;
-					}
-				}
-				mask <<= 1;
-			}
-			passcnt++;
-		} else if (passcnt >= CBR_THRESHOLD) {
-			break;
-		}
-	}
-	data = 0;
-	for (cnt = 0; cnt < 8; cnt++) {
-		data >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[0] >= dlli) {
-				dlli = (gold_sadj[0] - dlli) >> 1;
-				if (dlli > 3) {
-					dlli = 3;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[0]) >> 1;
-				if (dlli > 4) {
-					dlli = 4;
-				}
-				dlli = (8 - dlli) & 0x7;
-			}
-			data |= dlli << 21;
-		}
-	}
-	moutdwm(ast, 0x1E6E0080, data);
+static u32 cbr_scan3(struct ast_private *ast)
+{
+	u32 patcnt, loop;
 
-	data = 0;
-	for (cnt = 8; cnt < 16; cnt++) {
-		data >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[1] >= dlli) {
-				dlli = (gold_sadj[1] - dlli) >> 1;
-				if (dlli > 3) {
-					dlli = 3;
-				} else {
-					dlli = (dlli - 1) & 0x7;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[1]) >> 1;
-				dlli += 1;
-				if (dlli > 4) {
-					dlli = 4;
-				}
-				dlli = (8 - dlli) & 0x7;
-			}
-			data |= dlli << 21;
+	for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) {
+		ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
+		for (loop = 0; loop < 2; loop++) {
+			if (cbr_test3(ast))
+				break;
 		}
+		if (loop == 2)
+			return 0;
 	}
-	moutdwm(ast, 0x1E6E0084, data);
-
-} /* finetuneDQI */
-#endif
+	return 1;
+}
 
-static void finetuneDQI_L(struct ast_private *ast, struct ast2300_dram_param *param)
+static bool finetuneDQI_L(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt;
-
+	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt, retry = 0;
+	bool status = false;
 FINETUNE_START:
 	for (cnt = 0; cnt < 16; cnt++) {
 		dllmin[cnt] = 0xff;
@@ -690,16 +627,8 @@ FINETUNE_START:
 	}
 	passcnt = 0;
 	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
-
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE1);
+		ast_moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
+		ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE1);
 		data = cbr_scan2(ast);
 		if (data != 0) {
 			mask = 0x00010001;
@@ -727,9 +656,13 @@ FINETUNE_START:
 			passcnt++;
 		}
 	}
+	if (retry++ > 10)
+		goto FINETUNE_DONE;
 	if (passcnt != 16) {
 		goto FINETUNE_START;
 	}
+	status = true;
+FINETUNE_DONE:
 	gold_sadj[0] = gold_sadj[0] >> 4;
 	gold_sadj[1] = gold_sadj[0];
 
@@ -753,7 +686,7 @@ FINETUNE_START:
 			data |= dlli << 21;
 		}
 	}
-	moutdwm(ast, 0x1E6E0080, data);
+	ast_moutdwm(ast, 0x1E6E0080, data);
 
 	data = 0;
 	for (cnt = 8; cnt < 16; cnt++) {
@@ -778,162 +711,116 @@ FINETUNE_START:
 			data |= dlli << 21;
 		}
 	}
-	moutdwm(ast, 0x1E6E0084, data);
-
+	ast_moutdwm(ast, 0x1E6E0084, data);
+	return status;
 } /* finetuneDQI_L */
 
-static void finetuneDQI_L2(struct ast_private *ast, struct ast2300_dram_param *param)
+static void finetuneDQSI(struct ast_private *ast)
 {
-	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt, data2;
+	u32 dlli, dqsip, dqidly;
+	u32 reg_mcr18, reg_mcr0c, passcnt[2], diff;
+	u32 g_dqidly, g_dqsip, g_margin, g_side;
+	u16 pass[32][2][2];
+	char tag[2][76];
+
+	/* Disable DQI CBR */
+	reg_mcr0c  = ast_mindwm(ast, 0x1E6E000C);
+	reg_mcr18  = ast_mindwm(ast, 0x1E6E0018);
+	reg_mcr18 &= 0x0000ffff;
+	ast_moutdwm(ast, 0x1E6E0018, reg_mcr18);
 
-	for (cnt = 0; cnt < 16; cnt++) {
-		dllmin[cnt] = 0xff;
-		dllmax[cnt] = 0x0;
-	}
-	passcnt = 0;
 	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
-
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE2);
-		data = cbr_scan2(ast);
-		if (data != 0) {
-			mask = 0x00010001;
-			for (cnt = 0; cnt < 16; cnt++) {
-				if (data & mask) {
-					if (dllmin[cnt] > dlli) {
-						dllmin[cnt] = dlli;
-					}
-					if (dllmax[cnt] < dlli) {
-						dllmax[cnt] = dlli;
-					}
-				}
-				mask <<= 1;
-			}
-			passcnt++;
-		} else if (passcnt >= CBR_THRESHOLD2) {
-			break;
-		}
+		tag[0][dlli] = 0x0;
+		tag[1][dlli] = 0x0;
 	}
-	gold_sadj[0] = 0x0;
-	gold_sadj[1] = 0xFF;
-	for (cnt = 0; cnt < 8; cnt++) {
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			if (gold_sadj[0] < dllmin[cnt]) {
-				gold_sadj[0] = dllmin[cnt];
-			}
-			if (gold_sadj[1] > dllmax[cnt]) {
-				gold_sadj[1] = dllmax[cnt];
-			}
-		}
+	for (dqidly = 0; dqidly < 32; dqidly++) {
+		pass[dqidly][0][0] = 0xff;
+		pass[dqidly][0][1] = 0x0;
+		pass[dqidly][1][0] = 0xff;
+		pass[dqidly][1][1] = 0x0;
 	}
-	gold_sadj[0] = (gold_sadj[1] + gold_sadj[0]) >> 1;
-	gold_sadj[1] = mindwm(ast, 0x1E6E0080);
-
-	data = 0;
-	for (cnt = 0; cnt < 8; cnt++) {
-		data >>= 3;
-		data2 = gold_sadj[1] & 0x7;
-		gold_sadj[1] >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[0] >= dlli) {
-				dlli = (gold_sadj[0] - dlli) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 3) {
-					data2 = (data2 + dlli) & 0x7;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[0]) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 4) {
-					data2 = (data2 - dlli) & 0x7;
+	for (dqidly = 0; dqidly < 32; dqidly++) {
+		passcnt[0] = passcnt[1] = 0;
+		for (dqsip = 0; dqsip < 2; dqsip++) {
+			ast_moutdwm(ast, 0x1E6E000C, 0);
+			ast_moutdwm(ast, 0x1E6E0018, reg_mcr18 | (dqidly << 16) | (dqsip << 23));
+			ast_moutdwm(ast, 0x1E6E000C, reg_mcr0c);
+			for (dlli = 0; dlli < 76; dlli++) {
+				ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24));
+				ast_moutdwm(ast, 0x1E6E0070, 0);
+				ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE0);
+				if (cbr_scan3(ast)) {
+					if (dlli == 0)
+						break;
+					passcnt[dqsip]++;
+					tag[dqsip][dlli] = 'P';
+					if (dlli < pass[dqidly][dqsip][0])
+						pass[dqidly][dqsip][0] = (u16) dlli;
+					if (dlli > pass[dqidly][dqsip][1])
+						pass[dqidly][dqsip][1] = (u16) dlli;
+				} else if (passcnt[dqsip] >= 5)
+					break;
+				else {
+					pass[dqidly][dqsip][0] = 0xff;
+					pass[dqidly][dqsip][1] = 0x0;
 				}
 			}
 		}
-		data |= data2 << 21;
-	}
-	moutdwm(ast, 0x1E6E0080, data);
-
-	gold_sadj[0] = 0x0;
-	gold_sadj[1] = 0xFF;
-	for (cnt = 8; cnt < 16; cnt++) {
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			if (gold_sadj[0] < dllmin[cnt]) {
-				gold_sadj[0] = dllmin[cnt];
-			}
-			if (gold_sadj[1] > dllmax[cnt]) {
-				gold_sadj[1] = dllmax[cnt];
-			}
-		}
+		if (passcnt[0] == 0 && passcnt[1] == 0)
+			dqidly++;
 	}
-	gold_sadj[0] = (gold_sadj[1] + gold_sadj[0]) >> 1;
-	gold_sadj[1] = mindwm(ast, 0x1E6E0084);
-
-	data = 0;
-	for (cnt = 8; cnt < 16; cnt++) {
-		data >>= 3;
-		data2 = gold_sadj[1] & 0x7;
-		gold_sadj[1] >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[0] >= dlli) {
-				dlli = (gold_sadj[0] - dlli) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 3) {
-					data2 = (data2 + dlli) & 0x7;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[0]) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 4) {
-					data2 = (data2 - dlli) & 0x7;
-				}
+	/* Search margin */
+	g_dqidly = g_dqsip = g_margin = g_side = 0;
+
+	for (dqidly = 0; dqidly < 32; dqidly++) {
+		for (dqsip = 0; dqsip < 2; dqsip++) {
+			if (pass[dqidly][dqsip][0] > pass[dqidly][dqsip][1])
+				continue;
+			diff = pass[dqidly][dqsip][1] - pass[dqidly][dqsip][0];
+			if ((diff+2) < g_margin)
+				continue;
+			passcnt[0] = passcnt[1] = 0;
+			for (dlli = pass[dqidly][dqsip][0]; dlli > 0  && tag[dqsip][dlli] != 0; dlli--, passcnt[0]++);
+			for (dlli = pass[dqidly][dqsip][1]; dlli < 76 && tag[dqsip][dlli] != 0; dlli++, passcnt[1]++);
+			if (passcnt[0] > passcnt[1])
+				passcnt[0] = passcnt[1];
+			passcnt[1] = 0;
+			if (passcnt[0] > g_side)
+				passcnt[1] = passcnt[0] - g_side;
+			if (diff > (g_margin+1) && (passcnt[1] > 0 || passcnt[0] > 8)) {
+				g_margin = diff;
+				g_dqidly = dqidly;
+				g_dqsip  = dqsip;
+				g_side   = passcnt[0];
+			} else if (passcnt[1] > 1 && g_side < 8) {
+				if (diff > g_margin)
+					g_margin = diff;
+				g_dqidly = dqidly;
+				g_dqsip  = dqsip;
+				g_side   = passcnt[0];
 			}
 		}
-		data |= data2 << 21;
 	}
-	moutdwm(ast, 0x1E6E0084, data);
-
-} /* finetuneDQI_L2 */
+	reg_mcr18 = reg_mcr18 | (g_dqidly << 16) | (g_dqsip << 23);
+	ast_moutdwm(ast, 0x1E6E0018, reg_mcr18);
 
-static void cbr_dll2(struct ast_private *ast, struct ast2300_dram_param *param)
+}
+static bool cbr_dll2(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 dllmin[2], dllmax[2], dlli, data, data2, passcnt;
+	u32 dllmin[2], dllmax[2], dlli, data, passcnt, retry = 0;
+	bool status = false;
 
-
-	finetuneDQI_L(ast, param);
-	finetuneDQI_L2(ast, param);
+	finetuneDQSI(ast);
+	if (finetuneDQI_L(ast, param) == false)
+		return status;
 
 CBR_START2:
 	dllmin[0] = dllmin[1] = 0xff;
 	dllmax[0] = dllmax[1] = 0x0;
 	passcnt = 0;
 	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
-
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE2);
+		ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24));
+		ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE2);
 		data = cbr_scan(ast);
 		if (data != 0) {
 			if (data & 0x1) {
@@ -957,44 +844,31 @@ CBR_START2:
 			break;
 		}
 	}
+	if (retry++ > 10)
+		goto CBR_DONE2;
 	if (dllmax[0] == 0 || (dllmax[0]-dllmin[0]) < CBR_THRESHOLD) {
 		goto CBR_START2;
 	}
 	if (dllmax[1] == 0 || (dllmax[1]-dllmin[1]) < CBR_THRESHOLD) {
 		goto CBR_START2;
 	}
+	status = true;
+CBR_DONE2:
 	dlli  = (dllmin[1] + dllmax[1]) >> 1;
 	dlli <<= 8;
 	dlli += (dllmin[0] + dllmax[0]) >> 1;
-	moutdwm(ast, 0x1E6E0068, (mindwm(ast, 0x1E6E0068) & 0xFFFF) | (dlli << 16));
-
-	data  = (mindwm(ast, 0x1E6E0080) >> 24) & 0x1F;
-	data2 = (mindwm(ast, 0x1E6E0018) & 0xff80ffff) | (data << 16);
-	moutdwm(ast, 0x1E6E0018, data2);
-	moutdwm(ast, 0x1E6E0024, 0x8001 | (data << 1) | (param->dll2_finetune_step << 8));
-
-	/* Wait DQSI latch phase calibration */
-	moutdwm(ast, 0x1E6E0074, 0x00000010);
-	moutdwm(ast, 0x1E6E0070, 0x00000003);
-	do {
-		data = mindwm(ast, 0x1E6E0070);
-	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0070, 0x00000003);
-	do {
-		data = mindwm(ast, 0x1E6E0070);
-	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0068, ast_mindwm(ast, 0x1E720058) | (dlli << 16));
+	return status;
 } /* CBRDLL2 */
 
 static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *param)
 {
 	u32 trap, trap_AC2, trap_MRS;
 
-	moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
+	ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
 
 	/* Ger trap info */
-	trap = (mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
+	trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
 	trap_AC2  = 0x00020000 + (trap << 16);
 	trap_AC2 |= 0x00300000 + ((trap & 0x2) << 19);
 	trap_MRS  = 0x00000010 + (trap << 4);
@@ -1008,22 +882,35 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 	switch (param->dram_freq) {
 	case 336:
-		moutdwm(ast, 0x1E6E2020, 0x0190);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0190);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x22202725;
 		param->reg_AC2       = 0xAA007613 | trap_AC2;
 		param->reg_DQSIC     = 0x000000BA;
 		param->reg_MRS       = 0x04001400 | trap_MRS;
 		param->reg_EMRS      = 0x00000000;
-		param->reg_IOZ       = 0x00000034;
+		param->reg_IOZ       = 0x00000023;
 		param->reg_DQIDLY    = 0x00000074;
 		param->reg_FREQ      = 0x00004DC0;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 3;
+		switch (param->dram_chipid) {
+		default:
+		case AST_DRAM_512Mx16:
+		case AST_DRAM_1Gx16:
+			param->reg_AC2   = 0xAA007613 | trap_AC2;
+			break;
+		case AST_DRAM_2Gx16:
+			param->reg_AC2   = 0xAA00761C | trap_AC2;
+			break;
+		case AST_DRAM_4Gx16:
+			param->reg_AC2   = 0xAA007636 | trap_AC2;
+			break;
+		}
 		break;
 	default:
 	case 396:
-		moutdwm(ast, 0x1E6E2020, 0x03F1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x03F1);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x33302825;
 		param->reg_AC2       = 0xCC009617 | trap_AC2;
@@ -1033,7 +920,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->reg_IOZ       = 0x00000034;
 		param->reg_DRV       = 0x000000FA;
 		param->reg_DQIDLY    = 0x00000089;
-		param->reg_FREQ      = 0x000050C0;
+		param->reg_FREQ      = 0x00005040;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 4;
 
@@ -1053,14 +940,14 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		break;
 
 	case 408:
-		moutdwm(ast, 0x1E6E2020, 0x01F0);
+		ast_moutdwm(ast, 0x1E6E2020, 0x01F0);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x33302825;
 		param->reg_AC2       = 0xCC009617 | trap_AC2;
 		param->reg_DQSIC     = 0x000000E2;
 		param->reg_MRS       = 0x04001600 | trap_MRS;
 		param->reg_EMRS      = 0x00000000;
-		param->reg_IOZ       = 0x00000034;
+		param->reg_IOZ       = 0x00000023;
 		param->reg_DRV       = 0x000000FA;
 		param->reg_DQIDLY    = 0x00000089;
 		param->reg_FREQ      = 0x000050C0;
@@ -1083,7 +970,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 		break;
 	case 456:
-		moutdwm(ast, 0x1E6E2020, 0x0230);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0230);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x33302926;
 		param->reg_AC2       = 0xCD44961A;
@@ -1097,7 +984,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 4;
 		break;
 	case 504:
-		moutdwm(ast, 0x1E6E2020, 0x0270);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0270);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x33302926;
 		param->reg_AC2       = 0xDE44A61D;
@@ -1111,7 +998,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 4;
 		break;
 	case 528:
-		moutdwm(ast, 0x1E6E2020, 0x0290);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0290);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x33302926;
@@ -1127,7 +1014,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 576:
-		moutdwm(ast, 0x1E6E2020, 0x0140);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0140);
 		param->reg_MADJ      = 0x00136868;
 		param->reg_SADJ      = 0x00004534;
 		param->wodt          = 1;
@@ -1145,7 +1032,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 600:
-		moutdwm(ast, 0x1E6E2020, 0x02E1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x02E1);
 		param->reg_MADJ      = 0x00136868;
 		param->reg_SADJ      = 0x00004534;
 		param->wodt          = 1;
@@ -1163,7 +1050,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 624:
-		moutdwm(ast, 0x1E6E2020, 0x0160);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0160);
 		param->reg_MADJ      = 0x00136868;
 		param->reg_SADJ      = 0x00004534;
 		param->wodt          = 1;
@@ -1196,7 +1083,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 	case AST_DRAM_4Gx16:
 		param->dram_config = 0x133;
 		break;
-	}; /* switch size */
+	} /* switch size */
 
 	switch (param->vram_size) {
 	default:
@@ -1218,106 +1105,98 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 static void ddr3_init(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 data, data2;
+	u32 data, data2, retry = 0;
 
-	moutdwm(ast, 0x1E6E0000, 0xFC600309);
-	moutdwm(ast, 0x1E6E0018, 0x00000100);
-	moutdwm(ast, 0x1E6E0024, 0x00000000);
-	moutdwm(ast, 0x1E6E0034, 0x00000000);
+ddr3_init_start:
+	ast_moutdwm(ast, 0x1E6E0000, 0xFC600309);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00000100);
+	ast_moutdwm(ast, 0x1E6E0024, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0034, 0x00000000);
 	udelay(10);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
-	moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
+	ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
 	udelay(10);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
 	udelay(10);
 
-	moutdwm(ast, 0x1E6E0004, param->dram_config);
-	moutdwm(ast, 0x1E6E0008, 0x90040f);
-	moutdwm(ast, 0x1E6E0010, param->reg_AC1);
-	moutdwm(ast, 0x1E6E0014, param->reg_AC2);
-	moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
-	moutdwm(ast, 0x1E6E0080, 0x00000000);
-	moutdwm(ast, 0x1E6E0084, 0x00000000);
-	moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
-	moutdwm(ast, 0x1E6E0018, 0x4040A170);
-	moutdwm(ast, 0x1E6E0018, 0x20402370);
-	moutdwm(ast, 0x1E6E0038, 0x00000000);
-	moutdwm(ast, 0x1E6E0040, 0xFF444444);
-	moutdwm(ast, 0x1E6E0044, 0x22222222);
-	moutdwm(ast, 0x1E6E0048, 0x22222222);
-	moutdwm(ast, 0x1E6E004C, 0x00000002);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
-	moutdwm(ast, 0x1E6E0054, 0);
-	moutdwm(ast, 0x1E6E0060, param->reg_DRV);
-	moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0074, 0x00000000);
-	moutdwm(ast, 0x1E6E0078, 0x00000000);
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0004, param->dram_config);
+	ast_moutdwm(ast, 0x1E6E0008, 0x90040f);
+	ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1);
+	ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2);
+	ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
+	ast_moutdwm(ast, 0x1E6E0080, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0084, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
+	ast_moutdwm(ast, 0x1E6E0018, 0x4000A170);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00002370);
+	ast_moutdwm(ast, 0x1E6E0038, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0040, 0xFF444444);
+	ast_moutdwm(ast, 0x1E6E0044, 0x22222222);
+	ast_moutdwm(ast, 0x1E6E0048, 0x22222222);
+	ast_moutdwm(ast, 0x1E6E004C, 0x00000002);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0054, 0);
+	ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV);
+	ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0074, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0078, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
 	/* Wait MCLK2X lock to MCLK */
 	do {
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 	} while (!(data & 0x08000000));
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00005C04);
-	udelay(10);
-	moutdwm(ast, 0x1E6E000C, 0x00000000);
-	moutdwm(ast, 0x1E6E0034, 0x00000000);
-	data = mindwm(ast, 0x1E6E001C);
+	data = ast_mindwm(ast, 0x1E6E001C);
 	data = (data >> 8) & 0xff;
 	while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) {
-		data2 = (mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
+		data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
 		if ((data2 & 0xff) > param->madj_max) {
 			break;
 		}
-		moutdwm(ast, 0x1E6E0064, data2);
+		ast_moutdwm(ast, 0x1E6E0064, data2);
 		if (data2 & 0x00100000) {
 			data2 = ((data2 & 0xff) >> 3) + 3;
 		} else {
 			data2 = ((data2 & 0xff) >> 2) + 5;
 		}
-		data = mindwm(ast, 0x1E6E0068) & 0xffff00ff;
+		data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff;
 		data2 += data & 0xff;
 		data = data | (data2 << 8);
-		moutdwm(ast, 0x1E6E0068, data);
+		ast_moutdwm(ast, 0x1E6E0068, data);
 		udelay(10);
-		moutdwm(ast, 0x1E6E0064, mindwm(ast, 0x1E6E0064) | 0xC0000);
+		ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000);
 		udelay(10);
-		data = mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
-		moutdwm(ast, 0x1E6E0018, data);
+		data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		data = data | 0x200;
-		moutdwm(ast, 0x1E6E0018, data);
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		do {
-			data = mindwm(ast, 0x1E6E001C);
+			data = ast_mindwm(ast, 0x1E6E001C);
 		} while (!(data & 0x08000000));
 
-		moutdwm(ast, 0x1E6E0034, 0x00000001);
-		moutdwm(ast, 0x1E6E000C, 0x00005C04);
-		udelay(10);
-		moutdwm(ast, 0x1E6E000C, 0x00000000);
-		moutdwm(ast, 0x1E6E0034, 0x00000000);
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 		data = (data >> 8) & 0xff;
 	}
-	data = mindwm(ast, 0x1E6E0018) | 0xC00;
-	moutdwm(ast, 0x1E6E0018, data);
+	ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0068) & 0xffff);
+	data = ast_mindwm(ast, 0x1E6E0018) | 0xC00;
+	ast_moutdwm(ast, 0x1E6E0018, data);
 
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00000040);
+	ast_moutdwm(ast, 0x1E6E0034, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E000C, 0x00000040);
 	udelay(50);
 	/* Mode Register Setting */
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000005);
-	moutdwm(ast, 0x1E6E0028, 0x00000007);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS);
-	moutdwm(ast, 0x1E6E000C, 0x00005C08);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-
-	moutdwm(ast, 0x1E6E000C, 0x7FFF5C01);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000005);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000007);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS);
+	ast_moutdwm(ast, 0x1E6E000C, 0x00005C08);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+
+	ast_moutdwm(ast, 0x1E6E000C, 0x00005C01);
 	data = 0;
 	if (param->wodt) {
 		data = 0x300;
@@ -1325,30 +1204,23 @@ static void ddr3_init(struct ast_private *ast, struct ast2300_dram_param *param)
 	if (param->rodt) {
 		data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3);
 	}
-	moutdwm(ast, 0x1E6E0034, data | 0x3);
+	ast_moutdwm(ast, 0x1E6E0034, data | 0x3);
 
-	/* Wait DQI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0080);
-	} while (!(data & 0x40000000));
-	/* Wait DQSI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0020);
-	} while (!(data & 0x00000800));
 	/* Calibrate the DQSI delay */
-	cbr_dll2(ast, param);
+	if ((cbr_dll2(ast, param) == false) && (retry++ < 10))
+		goto ddr3_init_start;
 
-	moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
+	ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
 	/* ECC Memory Initialization */
 #ifdef ECC
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
-	moutdwm(ast, 0x1E6E0070, 0x221);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x221);
 	do {
-		data = mindwm(ast, 0x1E6E0070);
+		data = ast_mindwm(ast, 0x1E6E0070);
 	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
 #endif
 
 
@@ -1358,10 +1230,10 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 {
 	u32 trap, trap_AC2, trap_MRS;
 
-	moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
+	ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
 
 	/* Ger trap info */
-	trap = (mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
+	trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
 	trap_AC2  = (trap << 20) | (trap << 16);
 	trap_AC2 += 0x00110000;
 	trap_MRS  = 0x00000040 | (trap << 4);
@@ -1375,7 +1247,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 	switch (param->dram_freq) {
 	case 264:
-		moutdwm(ast, 0x1E6E2020, 0x0130);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0130);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x11101513;
 		param->reg_AC2       = 0x78117011;
@@ -1390,7 +1262,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 336:
-		moutdwm(ast, 0x1E6E2020, 0x0190);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0190);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x22202613;
 		param->reg_AC2       = 0xAA009016 | trap_AC2;
@@ -1403,10 +1275,25 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->reg_FREQ      = 0x00004DC0;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 3;
+		switch (param->dram_chipid) {
+		default:
+		case AST_DRAM_512Mx16:
+			param->reg_AC2   = 0xAA009012 | trap_AC2;
+			break;
+		case AST_DRAM_1Gx16:
+			param->reg_AC2   = 0xAA009016 | trap_AC2;
+			break;
+		case AST_DRAM_2Gx16:
+			param->reg_AC2   = 0xAA009023 | trap_AC2;
+			break;
+		case AST_DRAM_4Gx16:
+			param->reg_AC2   = 0xAA00903B | trap_AC2;
+			break;
+		}
 		break;
 	default:
 	case 396:
-		moutdwm(ast, 0x1E6E2020, 0x03F1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x03F1);
 		param->wodt          = 1;
 		param->rodt          = 0;
 		param->reg_AC1       = 0x33302714;
@@ -1417,7 +1304,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->reg_DRV       = 0x000000FA;
 		param->reg_IOZ       = 0x00000034;
 		param->reg_DQIDLY    = 0x00000089;
-		param->reg_FREQ      = 0x000050C0;
+		param->reg_FREQ      = 0x00005040;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 4;
 
@@ -1440,7 +1327,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		break;
 
 	case 408:
-		moutdwm(ast, 0x1E6E2020, 0x01F0);
+		ast_moutdwm(ast, 0x1E6E2020, 0x01F0);
 		param->wodt          = 1;
 		param->rodt          = 0;
 		param->reg_AC1       = 0x33302714;
@@ -1473,7 +1360,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 		break;
 	case 456:
-		moutdwm(ast, 0x1E6E2020, 0x0230);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0230);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x33302815;
 		param->reg_AC2       = 0xCD44B01E;
@@ -1488,7 +1375,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 504:
-		moutdwm(ast, 0x1E6E2020, 0x0261);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0261);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x33302815;
@@ -1504,7 +1391,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 528:
-		moutdwm(ast, 0x1E6E2020, 0x0120);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0120);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x33302815;
@@ -1520,7 +1407,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 552:
-		moutdwm(ast, 0x1E6E2020, 0x02A1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x02A1);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x43402915;
@@ -1536,7 +1423,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 576:
-		moutdwm(ast, 0x1E6E2020, 0x0140);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0140);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x43402915;
@@ -1567,7 +1454,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 	case AST_DRAM_4Gx16:
 		param->dram_config = 0x123;
 		break;
-	}; /* switch size */
+	} /* switch size */
 
 	switch (param->vram_size) {
 	default:
@@ -1588,110 +1475,102 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 static void ddr2_init(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 data, data2;
-
-	moutdwm(ast, 0x1E6E0000, 0xFC600309);
-	moutdwm(ast, 0x1E6E0018, 0x00000100);
-	moutdwm(ast, 0x1E6E0024, 0x00000000);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
-	moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
+	u32 data, data2, retry = 0;
+
+ddr2_init_start:
+	ast_moutdwm(ast, 0x1E6E0000, 0xFC600309);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00000100);
+	ast_moutdwm(ast, 0x1E6E0024, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
+	ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
 	udelay(10);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
 	udelay(10);
 
-	moutdwm(ast, 0x1E6E0004, param->dram_config);
-	moutdwm(ast, 0x1E6E0008, 0x90040f);
-	moutdwm(ast, 0x1E6E0010, param->reg_AC1);
-	moutdwm(ast, 0x1E6E0014, param->reg_AC2);
-	moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
-	moutdwm(ast, 0x1E6E0080, 0x00000000);
-	moutdwm(ast, 0x1E6E0084, 0x00000000);
-	moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
-	moutdwm(ast, 0x1E6E0018, 0x4040A130);
-	moutdwm(ast, 0x1E6E0018, 0x20402330);
-	moutdwm(ast, 0x1E6E0038, 0x00000000);
-	moutdwm(ast, 0x1E6E0040, 0xFF808000);
-	moutdwm(ast, 0x1E6E0044, 0x88848466);
-	moutdwm(ast, 0x1E6E0048, 0x44440008);
-	moutdwm(ast, 0x1E6E004C, 0x00000000);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
-	moutdwm(ast, 0x1E6E0054, 0);
-	moutdwm(ast, 0x1E6E0060, param->reg_DRV);
-	moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0074, 0x00000000);
-	moutdwm(ast, 0x1E6E0078, 0x00000000);
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0004, param->dram_config);
+	ast_moutdwm(ast, 0x1E6E0008, 0x90040f);
+	ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1);
+	ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2);
+	ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
+	ast_moutdwm(ast, 0x1E6E0080, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0084, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
+	ast_moutdwm(ast, 0x1E6E0018, 0x4000A130);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00002330);
+	ast_moutdwm(ast, 0x1E6E0038, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0040, 0xFF808000);
+	ast_moutdwm(ast, 0x1E6E0044, 0x88848466);
+	ast_moutdwm(ast, 0x1E6E0048, 0x44440008);
+	ast_moutdwm(ast, 0x1E6E004C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0054, 0);
+	ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV);
+	ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0074, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0078, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
 
 	/* Wait MCLK2X lock to MCLK */
 	do {
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 	} while (!(data & 0x08000000));
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00005C04);
-	udelay(10);
-	moutdwm(ast, 0x1E6E000C, 0x00000000);
-	moutdwm(ast, 0x1E6E0034, 0x00000000);
-	data = mindwm(ast, 0x1E6E001C);
+	data = ast_mindwm(ast, 0x1E6E001C);
 	data = (data >> 8) & 0xff;
 	while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) {
-		data2 = (mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
+		data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
 		if ((data2 & 0xff) > param->madj_max) {
 			break;
 		}
-		moutdwm(ast, 0x1E6E0064, data2);
+		ast_moutdwm(ast, 0x1E6E0064, data2);
 		if (data2 & 0x00100000) {
 			data2 = ((data2 & 0xff) >> 3) + 3;
 		} else {
 			data2 = ((data2 & 0xff) >> 2) + 5;
 		}
-		data = mindwm(ast, 0x1E6E0068) & 0xffff00ff;
+		data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff;
 		data2 += data & 0xff;
 		data = data | (data2 << 8);
-		moutdwm(ast, 0x1E6E0068, data);
+		ast_moutdwm(ast, 0x1E6E0068, data);
 		udelay(10);
-		moutdwm(ast, 0x1E6E0064, mindwm(ast, 0x1E6E0064) | 0xC0000);
+		ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000);
 		udelay(10);
-		data = mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
-		moutdwm(ast, 0x1E6E0018, data);
+		data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		data = data | 0x200;
-		moutdwm(ast, 0x1E6E0018, data);
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		do {
-			data = mindwm(ast, 0x1E6E001C);
+			data = ast_mindwm(ast, 0x1E6E001C);
 		} while (!(data & 0x08000000));
 
-		moutdwm(ast, 0x1E6E0034, 0x00000001);
-		moutdwm(ast, 0x1E6E000C, 0x00005C04);
-		udelay(10);
-		moutdwm(ast, 0x1E6E000C, 0x00000000);
-		moutdwm(ast, 0x1E6E0034, 0x00000000);
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 		data = (data >> 8) & 0xff;
 	}
-	data = mindwm(ast, 0x1E6E0018) | 0xC00;
-	moutdwm(ast, 0x1E6E0018, data);
+	ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0008) & 0xffff);
+	data = ast_mindwm(ast, 0x1E6E0018) | 0xC00;
+	ast_moutdwm(ast, 0x1E6E0018, data);
 
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0034, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E000C, 0x00000000);
 	udelay(50);
 	/* Mode Register Setting */
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000005);
-	moutdwm(ast, 0x1E6E0028, 0x00000007);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-
-	moutdwm(ast, 0x1E6E000C, 0x00005C08);
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS | 0x380);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-
-	moutdwm(ast, 0x1E6E000C, 0x7FFF5C01);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000005);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000007);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+
+	ast_moutdwm(ast, 0x1E6E000C, 0x00005C08);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS | 0x380);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+
+	ast_moutdwm(ast, 0x1E6E000C, 0x7FFF5C01);
 	data = 0;
 	if (param->wodt) {
 		data = 0x500;
@@ -1699,30 +1578,23 @@ static void ddr2_init(struct ast_private *ast, struct ast2300_dram_param *param)
 	if (param->rodt) {
 		data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3);
 	}
-	moutdwm(ast, 0x1E6E0034, data | 0x3);
-	moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
+	ast_moutdwm(ast, 0x1E6E0034, data | 0x3);
+	ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
 
-	/* Wait DQI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0080);
-	} while (!(data & 0x40000000));
-	/* Wait DQSI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0020);
-	} while (!(data & 0x00000800));
 	/* Calibrate the DQSI delay */
-	cbr_dll2(ast, param);
+	if ((cbr_dll2(ast, param) == false) && (retry++ < 10))
+		goto ddr2_init_start;
 
 	/* ECC Memory Initialization */
 #ifdef ECC
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
-	moutdwm(ast, 0x1E6E0070, 0x221);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x221);
 	do {
-		data = mindwm(ast, 0x1E6E0070);
+		data = ast_mindwm(ast, 0x1E6E0070);
 	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
 #endif
 
 }
@@ -1768,8 +1640,8 @@ static void ast_init_dram_2300(struct drm_device *dev)
 			ddr2_init(ast, &param);
 		}
 
-		temp = mindwm(ast, 0x1e6e2040);
-		moutdwm(ast, 0x1e6e2040, temp | 0x40);
+		temp = ast_mindwm(ast, 0x1e6e2040);
+		ast_moutdwm(ast, 0x1e6e2040, temp | 0x40);
 	}
 
 	/* wait ready */
diff --git a/drivers/gpu/drm/ast/ast_tables.h b/drivers/gpu/drm/ast/ast_tables.h
index 95fa6aba26bc..4c761dcea972 100644
--- a/drivers/gpu/drm/ast/ast_tables.h
+++ b/drivers/gpu/drm/ast/ast_tables.h
@@ -42,7 +42,7 @@
 #define HBorder                 0x00000020
 #define VBorder                 0x00000010
 #define WideScreenMode		0x00000100
-
+#define NewModeInfo		0x00000200
 
 /* DCLK Index */
 #define VCLK25_175     		0x00
@@ -67,6 +67,11 @@
 #define VCLK106_5   		0x12
 #define VCLK146_25  		0x13
 #define VCLK148_5   		0x14
+#define VCLK71      		0x15
+#define VCLK88_75   		0x16
+#define VCLK119     		0x17
+#define VCLK85_5     		0x18
+#define VCLK97_75     		0x19
 
 static struct ast_vbios_dclk_info dclk_table[] = {
 	{0x2C, 0xE7, 0x03},					/* 00: VCLK25_175	*/
@@ -90,6 +95,10 @@ static struct ast_vbios_dclk_info dclk_table[] = {
 	{0x28, 0x49, 0x80},					/* 12: VCLK106.5        */
 	{0x37, 0x49, 0x80},					/* 13: VCLK146.25       */
 	{0x1f, 0x45, 0x80},					/* 14: VCLK148.5        */
+	{0x47, 0x6c, 0x80},					/* 15: VCLK71       */
+	{0x25, 0x65, 0x80},					/* 16: VCLK88.75    */
+	{0x77, 0x58, 0x80},					/* 17: VCLK119      */
+	{0x32, 0x67, 0x80},				    /* 18: VCLK85_5     */
 };
 
 static struct ast_vbios_stdtable vbios_stdtable[] = {
@@ -225,41 +234,63 @@ static struct ast_vbios_enhtable res_1600x1200[] = {
 	 (SyncPP | Charx8Dot), 0xFF, 1, 0x33 },
 };
 
-static struct ast_vbios_enhtable res_1920x1200[] = {
-	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
-	 (SyncNP | Charx8Dot), 60, 1, 0x34 },
-	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
-	 (SyncNP | Charx8Dot), 0xFF, 1, 0x34 },
+/* 16:9 */
+static struct ast_vbios_enhtable res_1360x768[] = {
+	{1792, 1360, 64,112, 795,  768, 3, 6, VCLK85_5,	         /* 60Hz */
+	 (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x39 },
+	{1792, 1360, 64,112, 795,  768, 3, 6, VCLK85_5,	         /* end */
+	 (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x39 },
+};
+
+static struct ast_vbios_enhtable res_1600x900[] = {
+	{1760, 1600, 48, 32, 926,  900, 3, 5, VCLK97_75,	/* 60Hz CVT RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x3A },
+	{1760, 1600, 48, 32, 926,  900, 3, 5, VCLK97_75,	/* end */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x3A }
 };
 
+static struct ast_vbios_enhtable res_1920x1080[] = {
+	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x38 },
+	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x38 },
+};
+
+
 /* 16:10 */
 static struct ast_vbios_enhtable res_1280x800[] = {
+	{1440, 1280, 48, 32,  823,  800, 3, 6, VCLK71,	/* 60Hz RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 35 },
 	{1680, 1280, 72,128,  831,  800, 3, 6, VCLK83_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x35 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x35 },
 	{1680, 1280, 72,128,  831,  800, 3, 6, VCLK83_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x35 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x35 },
 
 };
 
 static struct ast_vbios_enhtable res_1440x900[] = {
+	{1600, 1440, 48, 32,  926,  900, 3, 6, VCLK88_75,	/* 60Hz RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x36 },
 	{1904, 1440, 80,152,  934,  900, 3, 6, VCLK106_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x36 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x36 },
 	{1904, 1440, 80,152,  934,  900, 3, 6, VCLK106_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x36 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x36 },
 };
 
 static struct ast_vbios_enhtable res_1680x1050[] = {
+	{1840, 1680, 48, 32, 1080, 1050, 3, 6, VCLK119,	/* 60Hz RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x37 },
 	{2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x37 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x37 },
 	{2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x37 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x37 },
 };
 
-/* HDTV */
-static struct ast_vbios_enhtable res_1920x1080[] = {
-	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
-	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x38 },
-	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
-	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x38 },
+static struct ast_vbios_enhtable res_1920x1200[] = {
+	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x34 },
+	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x34 },
 };
+
 #endif
diff --git a/drivers/gpu/drm/bochs/bochs_mm.c b/drivers/gpu/drm/bochs/bochs_mm.c
index f488be55d650..b9a695d92792 100644
--- a/drivers/gpu/drm/bochs/bochs_mm.c
+++ b/drivers/gpu/drm/bochs/bochs_mm.c
@@ -434,17 +434,13 @@ static void bochs_bo_unref(struct bochs_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
 
 void bochs_gem_free_object(struct drm_gem_object *obj)
 {
 	struct bochs_bo *bochs_bo = gem_to_bochs_bo(obj);
 
-	if (!bochs_bo)
-		return;
 	bochs_bo_unref(&bochs_bo);
 }
 
diff --git a/drivers/gpu/drm/bridge/ptn3460.c b/drivers/gpu/drm/bridge/ptn3460.c
index b171901a3553..98fd17ae4916 100644
--- a/drivers/gpu/drm/bridge/ptn3460.c
+++ b/drivers/gpu/drm/bridge/ptn3460.c
@@ -225,12 +225,6 @@ out:
 	return num_modes;
 }
 
-static int ptn3460_mode_valid(struct drm_connector *connector,
-		struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 struct drm_encoder *ptn3460_best_encoder(struct drm_connector *connector)
 {
 	struct ptn3460_bridge *ptn_bridge;
@@ -242,7 +236,6 @@ struct drm_encoder *ptn3460_best_encoder(struct drm_connector *connector)
 
 struct drm_connector_helper_funcs ptn3460_connector_helper_funcs = {
 	.get_modes = ptn3460_get_modes,
-	.mode_valid = ptn3460_mode_valid,
 	.best_encoder = ptn3460_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c
index 4b0170cf53fd..99c1983f99d2 100644
--- a/drivers/gpu/drm/cirrus/cirrus_main.c
+++ b/drivers/gpu/drm/cirrus/cirrus_main.c
@@ -264,17 +264,13 @@ static void cirrus_bo_unref(struct cirrus_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
 
 void cirrus_gem_free_object(struct drm_gem_object *obj)
 {
 	struct cirrus_bo *cirrus_bo = gem_to_cirrus_bo(obj);
 
-	if (!cirrus_bo)
-		return;
 	cirrus_bo_unref(&cirrus_bo);
 }
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c
index f59433b7610c..49332c5fe35b 100644
--- a/drivers/gpu/drm/cirrus/cirrus_mode.c
+++ b/drivers/gpu/drm/cirrus/cirrus_mode.c
@@ -505,13 +505,6 @@ static int cirrus_vga_get_modes(struct drm_connector *connector)
 	return count;
 }
 
-static int cirrus_vga_mode_valid(struct drm_connector *connector,
-				 struct drm_display_mode *mode)
-{
-	/* Any mode we've added is valid */
-	return MODE_OK;
-}
-
 static struct drm_encoder *cirrus_connector_best_encoder(struct drm_connector
 						  *connector)
 {
@@ -546,7 +539,6 @@ static void cirrus_connector_destroy(struct drm_connector *connector)
 
 struct drm_connector_helper_funcs cirrus_vga_connector_helper_funcs = {
 	.get_modes = cirrus_vga_get_modes,
-	.mode_valid = cirrus_vga_mode_valid,
 	.best_encoder = cirrus_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index edec31fe3fed..68175b54504b 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -363,7 +363,7 @@ static int drm_addmap_core(struct drm_device * dev, resource_size_t offset,
 		list->master = dev->primary->master;
 	*maplist = list;
 	return 0;
-	}
+}
 
 int drm_addmap(struct drm_device * dev, resource_size_t offset,
 	       unsigned int size, enum drm_map_type type,
@@ -656,13 +656,13 @@ int drm_addbufs_agp(struct drm_device * dev, struct drm_buf_desc * request)
 		DRM_DEBUG("zone invalid\n");
 		return -EINVAL;
 	}
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (dev->buf_use) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	atomic_inc(&dev->buf_alloc);
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	mutex_lock(&dev->struct_mutex);
 	entry = &dma->bufs[order];
@@ -805,13 +805,13 @@ int drm_addbufs_pci(struct drm_device * dev, struct drm_buf_desc * request)
 	page_order = order - PAGE_SHIFT > 0 ? order - PAGE_SHIFT : 0;
 	total = PAGE_SIZE << page_order;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (dev->buf_use) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	atomic_inc(&dev->buf_alloc);
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	mutex_lock(&dev->struct_mutex);
 	entry = &dma->bufs[order];
@@ -1015,13 +1015,13 @@ static int drm_addbufs_sg(struct drm_device * dev, struct drm_buf_desc * request
 	if (order < DRM_MIN_ORDER || order > DRM_MAX_ORDER)
 		return -EINVAL;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (dev->buf_use) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	atomic_inc(&dev->buf_alloc);
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	mutex_lock(&dev->struct_mutex);
 	entry = &dma->bufs[order];
@@ -1175,7 +1175,7 @@ int drm_addbufs(struct drm_device *dev, void *data,
  * \param arg pointer to a drm_buf_info structure.
  * \return zero on success or a negative number on failure.
  *
- * Increments drm_device::buf_use while holding the drm_device::count_lock
+ * Increments drm_device::buf_use while holding the drm_device::buf_lock
  * lock, preventing of allocating more buffers after this call. Information
  * about each requested buffer is then copied into user space.
  */
@@ -1196,13 +1196,13 @@ int drm_infobufs(struct drm_device *dev, void *data,
 	if (!dma)
 		return -EINVAL;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (atomic_read(&dev->buf_alloc)) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	++dev->buf_use;		/* Can't allocate more after this call */
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	for (i = 0, count = 0; i < DRM_MAX_ORDER + 1; i++) {
 		if (dma->bufs[i].buf_count)
@@ -1381,13 +1381,13 @@ int drm_mapbufs(struct drm_device *dev, void *data,
 	if (!dma)
 		return -EINVAL;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (atomic_read(&dev->buf_alloc)) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	dev->buf_use++;		/* Can't allocate more after this call */
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	if (request->count >= dma->buf_count) {
 		if ((dev->agp && (dma->flags & _DRM_DMA_USE_AGP))
diff --git a/drivers/gpu/drm/drm_cache.c b/drivers/gpu/drm/drm_cache.c
index 534cb89b160d..a6b690626a6b 100644
--- a/drivers/gpu/drm/drm_cache.c
+++ b/drivers/gpu/drm/drm_cache.c
@@ -131,14 +131,14 @@ drm_clflush_sg(struct sg_table *st)
 EXPORT_SYMBOL(drm_clflush_sg);
 
 void
-drm_clflush_virt_range(char *addr, unsigned long length)
+drm_clflush_virt_range(void *addr, unsigned long length)
 {
 #if defined(CONFIG_X86)
 	if (cpu_has_clflush) {
-		char *end = addr + length;
+		void *end = addr + length;
 		mb();
 		for (; addr < end; addr += boot_cpu_data.x86_clflush_size)
-			clflush(addr);
+			clflushopt(addr);
 		clflushopt(end - 1);
 		mb();
 		return;
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index d8b7099abece..fe94cc10cd35 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -37,6 +37,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_modeset_lock.h>
 
 #include "drm_crtc_internal.h"
 
@@ -50,12 +51,42 @@
  */
 void drm_modeset_lock_all(struct drm_device *dev)
 {
-	struct drm_crtc *crtc;
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_modeset_acquire_ctx *ctx;
+	int ret;
 
-	mutex_lock(&dev->mode_config.mutex);
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (WARN_ON(!ctx))
+		return;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		mutex_lock_nest_lock(&crtc->mutex, &dev->mode_config.mutex);
+	mutex_lock(&config->mutex);
+
+	drm_modeset_acquire_init(ctx, 0);
+
+retry:
+	ret = drm_modeset_lock(&config->connection_mutex, ctx);
+	if (ret)
+		goto fail;
+	ret = drm_modeset_lock_all_crtcs(dev, ctx);
+	if (ret)
+		goto fail;
+
+	WARN_ON(config->acquire_ctx);
+
+	/* now we hold the locks, so now that it is safe, stash the
+	 * ctx for drm_modeset_unlock_all():
+	 */
+	config->acquire_ctx = ctx;
+
+	drm_warn_on_modeset_not_all_locked(dev);
+
+	return;
+
+fail:
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(ctx);
+		goto retry;
+	}
 }
 EXPORT_SYMBOL(drm_modeset_lock_all);
 
@@ -67,10 +98,17 @@ EXPORT_SYMBOL(drm_modeset_lock_all);
  */
 void drm_modeset_unlock_all(struct drm_device *dev)
 {
-	struct drm_crtc *crtc;
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_modeset_acquire_ctx *ctx = config->acquire_ctx;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		mutex_unlock(&crtc->mutex);
+	if (WARN_ON(!ctx))
+		return;
+
+	config->acquire_ctx = NULL;
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+
+	kfree(ctx);
 
 	mutex_unlock(&dev->mode_config.mutex);
 }
@@ -91,8 +129,9 @@ void drm_warn_on_modeset_not_all_locked(struct drm_device *dev)
 		return;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		WARN_ON(!mutex_is_locked(&crtc->mutex));
+		WARN_ON(!drm_modeset_is_locked(&crtc->mutex));
 
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
 }
 EXPORT_SYMBOL(drm_warn_on_modeset_not_all_locked);
@@ -227,6 +266,7 @@ static const struct drm_prop_enum_list drm_encoder_enum_list[] =
 	{ DRM_MODE_ENCODER_TVDAC, "TV" },
 	{ DRM_MODE_ENCODER_VIRTUAL, "Virtual" },
 	{ DRM_MODE_ENCODER_DSI, "DSI" },
+	{ DRM_MODE_ENCODER_DPMST, "DP MST" },
 };
 
 static const struct drm_prop_enum_list drm_subpixel_enum_list[] =
@@ -256,46 +296,6 @@ void drm_connector_ida_destroy(void)
 }
 
 /**
- * drm_get_encoder_name - return a string for encoder
- * @encoder: encoder to compute name of
- *
- * Note that the buffer used by this function is globally shared and owned by
- * the function itself.
- *
- * FIXME: This isn't really multithreading safe.
- */
-const char *drm_get_encoder_name(const struct drm_encoder *encoder)
-{
-	static char buf[32];
-
-	snprintf(buf, 32, "%s-%d",
-		 drm_encoder_enum_list[encoder->encoder_type].name,
-		 encoder->base.id);
-	return buf;
-}
-EXPORT_SYMBOL(drm_get_encoder_name);
-
-/**
- * drm_get_connector_name - return a string for connector
- * @connector: connector to compute name of
- *
- * Note that the buffer used by this function is globally shared and owned by
- * the function itself.
- *
- * FIXME: This isn't really multithreading safe.
- */
-const char *drm_get_connector_name(const struct drm_connector *connector)
-{
-	static char buf[32];
-
-	snprintf(buf, 32, "%s-%d",
-		 drm_connector_enum_list[connector->connector_type].name,
-		 connector->connector_type_id);
-	return buf;
-}
-EXPORT_SYMBOL(drm_get_connector_name);
-
-/**
  * drm_get_connector_status_name - return a string for connector status
  * @status: connector status to compute name of
  *
@@ -409,6 +409,21 @@ void drm_mode_object_put(struct drm_device *dev,
 	mutex_unlock(&dev->mode_config.idr_mutex);
 }
 
+static struct drm_mode_object *_object_find(struct drm_device *dev,
+		uint32_t id, uint32_t type)
+{
+	struct drm_mode_object *obj = NULL;
+
+	mutex_lock(&dev->mode_config.idr_mutex);
+	obj = idr_find(&dev->mode_config.crtc_idr, id);
+	if (!obj || (type != DRM_MODE_OBJECT_ANY && obj->type != type) ||
+	    (obj->id != id))
+		obj = NULL;
+	mutex_unlock(&dev->mode_config.idr_mutex);
+
+	return obj;
+}
+
 /**
  * drm_mode_object_find - look up a drm object with static lifetime
  * @dev: drm device
@@ -416,7 +431,9 @@ void drm_mode_object_put(struct drm_device *dev,
  * @type: type of the mode object
  *
  * Note that framebuffers cannot be looked up with this functions - since those
- * are reference counted, they need special treatment.
+ * are reference counted, they need special treatment.  Even with
+ * DRM_MODE_OBJECT_ANY (although that will simply return NULL
+ * rather than WARN_ON()).
  */
 struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
 		uint32_t id, uint32_t type)
@@ -426,13 +443,10 @@ struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
 	/* Framebuffers are reference counted and need their own lookup
 	 * function.*/
 	WARN_ON(type == DRM_MODE_OBJECT_FB);
-
-	mutex_lock(&dev->mode_config.idr_mutex);
-	obj = idr_find(&dev->mode_config.crtc_idr, id);
-	if (!obj || (obj->type != type) || (obj->id != id))
+	obj = _object_find(dev, id, type);
+	/* don't leak out unref'd fb's */
+	if (obj && (obj->type == DRM_MODE_OBJECT_FB))
 		obj = NULL;
-	mutex_unlock(&dev->mode_config.idr_mutex);
-
 	return obj;
 }
 EXPORT_SYMBOL(drm_mode_object_find);
@@ -538,7 +552,7 @@ EXPORT_SYMBOL(drm_framebuffer_lookup);
  */
 void drm_framebuffer_unreference(struct drm_framebuffer *fb)
 {
-	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	DRM_DEBUG("%p: FB ID: %d (%d)\n", fb, fb->base.id, atomic_read(&fb->refcount.refcount));
 	kref_put(&fb->refcount, drm_framebuffer_free);
 }
 EXPORT_SYMBOL(drm_framebuffer_unreference);
@@ -551,7 +565,7 @@ EXPORT_SYMBOL(drm_framebuffer_unreference);
  */
 void drm_framebuffer_reference(struct drm_framebuffer *fb)
 {
-	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	DRM_DEBUG("%p: FB ID: %d (%d)\n", fb, fb->base.id, atomic_read(&fb->refcount.refcount));
 	kref_get(&fb->refcount);
 }
 EXPORT_SYMBOL(drm_framebuffer_reference);
@@ -563,7 +577,7 @@ static void drm_framebuffer_free_bug(struct kref *kref)
 
 static void __drm_framebuffer_unreference(struct drm_framebuffer *fb)
 {
-	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	DRM_DEBUG("%p: FB ID: %d (%d)\n", fb, fb->base.id, atomic_read(&fb->refcount.refcount));
 	kref_put(&fb->refcount, drm_framebuffer_free_bug);
 }
 
@@ -691,6 +705,8 @@ void drm_framebuffer_remove(struct drm_framebuffer *fb)
 }
 EXPORT_SYMBOL(drm_framebuffer_remove);
 
+DEFINE_WW_CLASS(crtc_ww_class);
+
 /**
  * drm_crtc_init_with_planes - Initialise a new CRTC object with
  *    specified primary and cursor planes.
@@ -710,6 +726,7 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 			      void *cursor,
 			      const struct drm_crtc_funcs *funcs)
 {
+	struct drm_mode_config *config = &dev->mode_config;
 	int ret;
 
 	crtc->dev = dev;
@@ -717,8 +734,9 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 	crtc->invert_dimensions = false;
 
 	drm_modeset_lock_all(dev);
-	mutex_init(&crtc->mutex);
-	mutex_lock_nest_lock(&crtc->mutex, &dev->mode_config.mutex);
+	drm_modeset_lock_init(&crtc->mutex);
+	/* dropped by _unlock_all(): */
+	drm_modeset_lock(&crtc->mutex, config->acquire_ctx);
 
 	ret = drm_mode_object_get(dev, &crtc->base, DRM_MODE_OBJECT_CRTC);
 	if (ret)
@@ -726,8 +744,8 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 
 	crtc->base.properties = &crtc->properties;
 
-	list_add_tail(&crtc->head, &dev->mode_config.crtc_list);
-	dev->mode_config.num_crtc++;
+	list_add_tail(&crtc->head, &config->crtc_list);
+	config->num_crtc++;
 
 	crtc->primary = primary;
 	if (primary)
@@ -755,6 +773,8 @@ void drm_crtc_cleanup(struct drm_crtc *crtc)
 	kfree(crtc->gamma_store);
 	crtc->gamma_store = NULL;
 
+	drm_modeset_lock_fini(&crtc->mutex);
+
 	drm_mode_object_put(dev, &crtc->base);
 	list_del(&crtc->head);
 	dev->mode_config.num_crtc--;
@@ -824,7 +844,7 @@ int drm_connector_init(struct drm_device *dev,
 
 	ret = drm_mode_object_get(dev, &connector->base, DRM_MODE_OBJECT_CONNECTOR);
 	if (ret)
-		goto out;
+		goto out_unlock;
 
 	connector->base.properties = &connector->properties;
 	connector->dev = dev;
@@ -834,9 +854,17 @@ int drm_connector_init(struct drm_device *dev,
 		ida_simple_get(connector_ida, 1, 0, GFP_KERNEL);
 	if (connector->connector_type_id < 0) {
 		ret = connector->connector_type_id;
-		drm_mode_object_put(dev, &connector->base);
-		goto out;
+		goto out_put;
+	}
+	connector->name =
+		kasprintf(GFP_KERNEL, "%s-%d",
+			  drm_connector_enum_list[connector_type].name,
+			  connector->connector_type_id);
+	if (!connector->name) {
+		ret = -ENOMEM;
+		goto out_put;
 	}
+
 	INIT_LIST_HEAD(&connector->probed_modes);
 	INIT_LIST_HEAD(&connector->modes);
 	connector->edid_blob_ptr = NULL;
@@ -853,7 +881,11 @@ int drm_connector_init(struct drm_device *dev,
 	drm_object_attach_property(&connector->base,
 				      dev->mode_config.dpms_property, 0);
 
- out:
+out_put:
+	if (ret)
+		drm_mode_object_put(dev, &connector->base);
+
+out_unlock:
 	drm_modeset_unlock_all(dev);
 
 	return ret;
@@ -881,6 +913,8 @@ void drm_connector_cleanup(struct drm_connector *connector)
 		   connector->connector_type_id);
 
 	drm_mode_object_put(dev, &connector->base);
+	kfree(connector->name);
+	connector->name = NULL;
 	list_del(&connector->head);
 	dev->mode_config.num_connector--;
 }
@@ -982,16 +1016,27 @@ int drm_encoder_init(struct drm_device *dev,
 
 	ret = drm_mode_object_get(dev, &encoder->base, DRM_MODE_OBJECT_ENCODER);
 	if (ret)
-		goto out;
+		goto out_unlock;
 
 	encoder->dev = dev;
 	encoder->encoder_type = encoder_type;
 	encoder->funcs = funcs;
+	encoder->name = kasprintf(GFP_KERNEL, "%s-%d",
+				  drm_encoder_enum_list[encoder_type].name,
+				  encoder->base.id);
+	if (!encoder->name) {
+		ret = -ENOMEM;
+		goto out_put;
+	}
 
 	list_add_tail(&encoder->head, &dev->mode_config.encoder_list);
 	dev->mode_config.num_encoder++;
 
- out:
+out_put:
+	if (ret)
+		drm_mode_object_put(dev, &encoder->base);
+
+out_unlock:
 	drm_modeset_unlock_all(dev);
 
 	return ret;
@@ -1009,6 +1054,8 @@ void drm_encoder_cleanup(struct drm_encoder *encoder)
 	struct drm_device *dev = encoder->dev;
 	drm_modeset_lock_all(dev);
 	drm_mode_object_put(dev, &encoder->base);
+	kfree(encoder->name);
+	encoder->name = NULL;
 	list_del(&encoder->head);
 	dev->mode_config.num_encoder--;
 	drm_modeset_unlock_all(dev);
@@ -1145,16 +1192,19 @@ EXPORT_SYMBOL(drm_plane_cleanup);
  */
 void drm_plane_force_disable(struct drm_plane *plane)
 {
+	struct drm_framebuffer *old_fb = plane->fb;
 	int ret;
 
-	if (!plane->fb)
+	if (!old_fb)
 		return;
 
 	ret = plane->funcs->disable_plane(plane);
-	if (ret)
+	if (ret) {
 		DRM_ERROR("failed to disable plane with busy fb\n");
+		return;
+	}
 	/* disconnect the plane from the fb and crtc: */
-	__drm_framebuffer_unreference(plane->fb);
+	__drm_framebuffer_unreference(old_fb);
 	plane->fb = NULL;
 	plane->crtc = NULL;
 }
@@ -1378,6 +1428,12 @@ static int drm_mode_group_init(struct drm_device *dev, struct drm_mode_group *gr
 	return 0;
 }
 
+void drm_mode_group_destroy(struct drm_mode_group *group)
+{
+	kfree(group->id_list);
+	group->id_list = NULL;
+}
+
 /*
  * NOTE: Driver's shouldn't ever call drm_mode_group_init_legacy_group - it is
  * the drm core's responsibility to set up mode control groups.
@@ -1614,7 +1670,7 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 					    &dev->mode_config.encoder_list,
 					    head) {
 				DRM_DEBUG_KMS("[ENCODER:%d:%s]\n", encoder->base.id,
-						drm_get_encoder_name(encoder));
+						encoder->name);
 				if (put_user(encoder->base.id, encoder_id +
 					     copied)) {
 					ret = -EFAULT;
@@ -1646,7 +1702,7 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 					    head) {
 				DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 					connector->base.id,
-					drm_get_connector_name(connector));
+					connector->name);
 				if (put_user(connector->base.id,
 					     connector_id + copied)) {
 					ret = -EFAULT;
@@ -1695,7 +1751,6 @@ int drm_mode_getcrtc(struct drm_device *dev,
 {
 	struct drm_mode_crtc *crtc_resp = data;
 	struct drm_crtc *crtc;
-	struct drm_mode_object *obj;
 	int ret = 0;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
@@ -1703,13 +1758,11 @@ int drm_mode_getcrtc(struct drm_device *dev,
 
 	drm_modeset_lock_all(dev);
 
-	obj = drm_mode_object_find(dev, crtc_resp->crtc_id,
-				   DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_resp->crtc_id);
+	if (!crtc) {
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 
 	crtc_resp->x = crtc->x;
 	crtc_resp->y = crtc->y;
@@ -1763,7 +1816,6 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 			  struct drm_file *file_priv)
 {
 	struct drm_mode_get_connector *out_resp = data;
-	struct drm_mode_object *obj;
 	struct drm_connector *connector;
 	struct drm_display_mode *mode;
 	int mode_count = 0;
@@ -1787,13 +1839,11 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 
 	mutex_lock(&dev->mode_config.mutex);
 
-	obj = drm_mode_object_find(dev, out_resp->connector_id,
-				   DRM_MODE_OBJECT_CONNECTOR);
-	if (!obj) {
+	connector = drm_connector_find(dev, out_resp->connector_id);
+	if (!connector) {
 		ret = -ENOENT;
 		goto out;
 	}
-	connector = obj_to_connector(obj);
 
 	props_count = connector->properties.count;
 
@@ -1821,10 +1871,12 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 	out_resp->mm_height = connector->display_info.height_mm;
 	out_resp->subpixel = connector->display_info.subpixel_order;
 	out_resp->connection = connector->status;
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	if (connector->encoder)
 		out_resp->encoder_id = connector->encoder->base.id;
 	else
 		out_resp->encoder_id = 0;
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 
 	/*
 	 * This ioctl is called twice, once to determine how much space is
@@ -1908,7 +1960,6 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 			struct drm_file *file_priv)
 {
 	struct drm_mode_get_encoder *enc_resp = data;
-	struct drm_mode_object *obj;
 	struct drm_encoder *encoder;
 	int ret = 0;
 
@@ -1916,13 +1967,11 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, enc_resp->encoder_id,
-				   DRM_MODE_OBJECT_ENCODER);
-	if (!obj) {
+	encoder = drm_encoder_find(dev, enc_resp->encoder_id);
+	if (!encoder) {
 		ret = -ENOENT;
 		goto out;
 	}
-	encoder = obj_to_encoder(obj);
 
 	if (encoder->crtc)
 		enc_resp->crtc_id = encoder->crtc->base.id;
@@ -2020,7 +2069,6 @@ int drm_mode_getplane(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv)
 {
 	struct drm_mode_get_plane *plane_resp = data;
-	struct drm_mode_object *obj;
 	struct drm_plane *plane;
 	uint32_t __user *format_ptr;
 	int ret = 0;
@@ -2029,13 +2077,11 @@ int drm_mode_getplane(struct drm_device *dev, void *data,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, plane_resp->plane_id,
-				   DRM_MODE_OBJECT_PLANE);
-	if (!obj) {
+	plane = drm_plane_find(dev, plane_resp->plane_id);
+	if (!plane) {
 		ret = -ENOENT;
 		goto out;
 	}
-	plane = obj_to_plane(obj);
 
 	if (plane->crtc)
 		plane_resp->crtc_id = plane->crtc->base.id;
@@ -2088,7 +2134,6 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv)
 {
 	struct drm_mode_set_plane *plane_req = data;
-	struct drm_mode_object *obj;
 	struct drm_plane *plane;
 	struct drm_crtc *crtc;
 	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
@@ -2103,35 +2148,42 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	 * First, find the plane, crtc, and fb objects.  If not available,
 	 * we don't bother to call the driver.
 	 */
-	obj = drm_mode_object_find(dev, plane_req->plane_id,
-				   DRM_MODE_OBJECT_PLANE);
-	if (!obj) {
+	plane = drm_plane_find(dev, plane_req->plane_id);
+	if (!plane) {
 		DRM_DEBUG_KMS("Unknown plane ID %d\n",
 			      plane_req->plane_id);
 		return -ENOENT;
 	}
-	plane = obj_to_plane(obj);
 
 	/* No fb means shut it down */
 	if (!plane_req->fb_id) {
 		drm_modeset_lock_all(dev);
 		old_fb = plane->fb;
-		plane->funcs->disable_plane(plane);
-		plane->crtc = NULL;
-		plane->fb = NULL;
+		ret = plane->funcs->disable_plane(plane);
+		if (!ret) {
+			plane->crtc = NULL;
+			plane->fb = NULL;
+		} else {
+			old_fb = NULL;
+		}
 		drm_modeset_unlock_all(dev);
 		goto out;
 	}
 
-	obj = drm_mode_object_find(dev, plane_req->crtc_id,
-				   DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, plane_req->crtc_id);
+	if (!crtc) {
 		DRM_DEBUG_KMS("Unknown crtc ID %d\n",
 			      plane_req->crtc_id);
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
+
+	/* Check whether this plane is usable on this CRTC */
+	if (!(plane->possible_crtcs & drm_crtc_mask(crtc))) {
+		DRM_DEBUG_KMS("Invalid crtc for plane\n");
+		ret = -EINVAL;
+		goto out;
+	}
 
 	fb = drm_framebuffer_lookup(dev, plane_req->fb_id);
 	if (!fb) {
@@ -2187,16 +2239,18 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	}
 
 	drm_modeset_lock_all(dev);
+	old_fb = plane->fb;
 	ret = plane->funcs->update_plane(plane, crtc, fb,
 					 plane_req->crtc_x, plane_req->crtc_y,
 					 plane_req->crtc_w, plane_req->crtc_h,
 					 plane_req->src_x, plane_req->src_y,
 					 plane_req->src_w, plane_req->src_h);
 	if (!ret) {
-		old_fb = plane->fb;
 		plane->crtc = crtc;
 		plane->fb = fb;
 		fb = NULL;
+	} else {
+		old_fb = NULL;
 	}
 	drm_modeset_unlock_all(dev);
 
@@ -2239,9 +2293,7 @@ int drm_mode_set_config_internal(struct drm_mode_set *set)
 	ret = crtc->funcs->set_config(set);
 	if (ret == 0) {
 		crtc->primary->crtc = crtc;
-
-		/* crtc->fb must be updated by ->set_config, enforces this. */
-		WARN_ON(fb != crtc->primary->fb);
+		crtc->primary->fb = fb;
 	}
 
 	list_for_each_entry(tmp, &crtc->dev->mode_config.crtc_list, head) {
@@ -2318,7 +2370,6 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 {
 	struct drm_mode_config *config = &dev->mode_config;
 	struct drm_mode_crtc *crtc_req = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	struct drm_connector **connector_set = NULL, *connector;
 	struct drm_framebuffer *fb = NULL;
@@ -2336,14 +2387,12 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 		return -ERANGE;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, crtc_req->crtc_id,
-				   DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_req->crtc_id);
+	if (!crtc) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", crtc_req->crtc_id);
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 	DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id);
 
 	if (crtc_req->mode_valid) {
@@ -2426,18 +2475,16 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 				goto out;
 			}
 
-			obj = drm_mode_object_find(dev, out_id,
-						   DRM_MODE_OBJECT_CONNECTOR);
-			if (!obj) {
+			connector = drm_connector_find(dev, out_id);
+			if (!connector) {
 				DRM_DEBUG_KMS("Connector id %d unknown\n",
 						out_id);
 				ret = -ENOENT;
 				goto out;
 			}
-			connector = obj_to_connector(obj);
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 					connector->base.id,
-					drm_get_connector_name(connector));
+					connector->name);
 
 			connector_set[i] = connector;
 		}
@@ -2466,7 +2513,6 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 				  struct drm_mode_cursor2 *req,
 				  struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	int ret = 0;
 
@@ -2476,14 +2522,13 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 	if (!req->flags || (~DRM_MODE_CURSOR_FLAGS & req->flags))
 		return -EINVAL;
 
-	obj = drm_mode_object_find(dev, req->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, req->crtc_id);
+	if (!crtc) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", req->crtc_id);
 		return -ENOENT;
 	}
-	crtc = obj_to_crtc(obj);
 
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	if (req->flags & DRM_MODE_CURSOR_BO) {
 		if (!crtc->funcs->cursor_set && !crtc->funcs->cursor_set2) {
 			ret = -ENXIO;
@@ -2507,7 +2552,7 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 		}
 	}
 out:
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 
 	return ret;
 
@@ -3097,6 +3142,8 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags,
 	if (!property)
 		return NULL;
 
+	property->dev = dev;
+
 	if (num_values) {
 		property->values = kzalloc(sizeof(uint64_t)*num_values, GFP_KERNEL);
 		if (!property->values)
@@ -3117,6 +3164,9 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags,
 	}
 
 	list_add_tail(&property->head, &dev->mode_config.property_list);
+
+	WARN_ON(!drm_property_type_valid(property));
+
 	return property;
 fail:
 	kfree(property->values);
@@ -3217,6 +3267,22 @@ struct drm_property *drm_property_create_bitmask(struct drm_device *dev,
 }
 EXPORT_SYMBOL(drm_property_create_bitmask);
 
+static struct drm_property *property_create_range(struct drm_device *dev,
+					 int flags, const char *name,
+					 uint64_t min, uint64_t max)
+{
+	struct drm_property *property;
+
+	property = drm_property_create(dev, flags, name, 2);
+	if (!property)
+		return NULL;
+
+	property->values[0] = min;
+	property->values[1] = max;
+
+	return property;
+}
+
 /**
  * drm_property_create - create a new ranged property type
  * @dev: drm device
@@ -3239,20 +3305,36 @@ struct drm_property *drm_property_create_range(struct drm_device *dev, int flags
 					 const char *name,
 					 uint64_t min, uint64_t max)
 {
+	return property_create_range(dev, DRM_MODE_PROP_RANGE | flags,
+			name, min, max);
+}
+EXPORT_SYMBOL(drm_property_create_range);
+
+struct drm_property *drm_property_create_signed_range(struct drm_device *dev,
+					 int flags, const char *name,
+					 int64_t min, int64_t max)
+{
+	return property_create_range(dev, DRM_MODE_PROP_SIGNED_RANGE | flags,
+			name, I642U64(min), I642U64(max));
+}
+EXPORT_SYMBOL(drm_property_create_signed_range);
+
+struct drm_property *drm_property_create_object(struct drm_device *dev,
+					 int flags, const char *name, uint32_t type)
+{
 	struct drm_property *property;
 
-	flags |= DRM_MODE_PROP_RANGE;
+	flags |= DRM_MODE_PROP_OBJECT;
 
-	property = drm_property_create(dev, flags, name, 2);
+	property = drm_property_create(dev, flags, name, 1);
 	if (!property)
 		return NULL;
 
-	property->values[0] = min;
-	property->values[1] = max;
+	property->values[0] = type;
 
 	return property;
 }
-EXPORT_SYMBOL(drm_property_create_range);
+EXPORT_SYMBOL(drm_property_create_object);
 
 /**
  * drm_property_add_enum - add a possible value to an enumeration property
@@ -3274,14 +3356,16 @@ int drm_property_add_enum(struct drm_property *property, int index,
 {
 	struct drm_property_enum *prop_enum;
 
-	if (!(property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)))
+	if (!(drm_property_type_is(property, DRM_MODE_PROP_ENUM) ||
+			drm_property_type_is(property, DRM_MODE_PROP_BITMASK)))
 		return -EINVAL;
 
 	/*
 	 * Bitmask enum properties have the additional constraint of values
 	 * from 0 to 63
 	 */
-	if ((property->flags & DRM_MODE_PROP_BITMASK) && (value > 63))
+	if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK) &&
+			(value > 63))
 		return -EINVAL;
 
 	if (!list_empty(&property->enum_blob_list)) {
@@ -3438,7 +3522,6 @@ EXPORT_SYMBOL(drm_object_property_get_value);
 int drm_mode_getproperty_ioctl(struct drm_device *dev,
 			       void *data, struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_mode_get_property *out_resp = data;
 	struct drm_property *property;
 	int enum_count = 0;
@@ -3457,17 +3540,17 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, out_resp->prop_id, DRM_MODE_OBJECT_PROPERTY);
-	if (!obj) {
+	property = drm_property_find(dev, out_resp->prop_id);
+	if (!property) {
 		ret = -ENOENT;
 		goto done;
 	}
-	property = obj_to_property(obj);
 
-	if (property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)) {
+	if (drm_property_type_is(property, DRM_MODE_PROP_ENUM) ||
+			drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) {
 		list_for_each_entry(prop_enum, &property->enum_blob_list, head)
 			enum_count++;
-	} else if (property->flags & DRM_MODE_PROP_BLOB) {
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) {
 		list_for_each_entry(prop_blob, &property->enum_blob_list, head)
 			blob_count++;
 	}
@@ -3489,7 +3572,8 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 	}
 	out_resp->count_values = value_count;
 
-	if (property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)) {
+	if (drm_property_type_is(property, DRM_MODE_PROP_ENUM) ||
+			drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) {
 		if ((out_resp->count_enum_blobs >= enum_count) && enum_count) {
 			copied = 0;
 			enum_ptr = (struct drm_mode_property_enum __user *)(unsigned long)out_resp->enum_blob_ptr;
@@ -3511,7 +3595,7 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 		out_resp->count_enum_blobs = enum_count;
 	}
 
-	if (property->flags & DRM_MODE_PROP_BLOB) {
+	if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) {
 		if ((out_resp->count_enum_blobs >= blob_count) && blob_count) {
 			copied = 0;
 			blob_id_ptr = (uint32_t __user *)(unsigned long)out_resp->enum_blob_ptr;
@@ -3590,7 +3674,6 @@ static void drm_property_destroy_blob(struct drm_device *dev,
 int drm_mode_getblob_ioctl(struct drm_device *dev,
 			   void *data, struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_mode_get_blob *out_resp = data;
 	struct drm_property_blob *blob;
 	int ret = 0;
@@ -3600,12 +3683,11 @@ int drm_mode_getblob_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, out_resp->blob_id, DRM_MODE_OBJECT_BLOB);
-	if (!obj) {
+	blob = drm_property_blob_find(dev, out_resp->blob_id);
+	if (!blob) {
 		ret = -ENOENT;
 		goto done;
 	}
-	blob = obj_to_blob(obj);
 
 	if (out_resp->length == blob->length) {
 		blob_ptr = (void __user *)(unsigned long)out_resp->data;
@@ -3667,19 +3749,40 @@ static bool drm_property_change_is_valid(struct drm_property *property,
 {
 	if (property->flags & DRM_MODE_PROP_IMMUTABLE)
 		return false;
-	if (property->flags & DRM_MODE_PROP_RANGE) {
+
+	if (drm_property_type_is(property, DRM_MODE_PROP_RANGE)) {
 		if (value < property->values[0] || value > property->values[1])
 			return false;
 		return true;
-	} else if (property->flags & DRM_MODE_PROP_BITMASK) {
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_SIGNED_RANGE)) {
+		int64_t svalue = U642I64(value);
+		if (svalue < U642I64(property->values[0]) ||
+				svalue > U642I64(property->values[1]))
+			return false;
+		return true;
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) {
 		int i;
 		uint64_t valid_mask = 0;
 		for (i = 0; i < property->num_values; i++)
 			valid_mask |= (1ULL << property->values[i]);
 		return !(value & ~valid_mask);
-	} else if (property->flags & DRM_MODE_PROP_BLOB) {
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) {
 		/* Only the driver knows */
 		return true;
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_OBJECT)) {
+		struct drm_mode_object *obj;
+		/* a zero value for an object property translates to null: */
+		if (value == 0)
+			return true;
+		/*
+		 * NOTE: use _object_find() directly to bypass restriction on
+		 * looking up refcnt'd objects (ie. fb's).  For a refcnt'd
+		 * object this could race against object finalization, so it
+		 * simply tells us that the object *was* valid.  Which is good
+		 * enough.
+		 */
+		obj = _object_find(property->dev, value, property->values[0]);
+		return obj != NULL;
 	} else {
 		int i;
 		for (i = 0; i < property->num_values; i++)
@@ -3987,7 +4090,6 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 			     void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_crtc_lut *crtc_lut = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	void *r_base, *g_base, *b_base;
 	int size;
@@ -3997,12 +4099,11 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_lut->crtc_id);
+	if (!crtc) {
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 
 	if (crtc->funcs->gamma_set == NULL) {
 		ret = -ENOSYS;
@@ -4061,7 +4162,6 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 			     void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_crtc_lut *crtc_lut = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	void *r_base, *g_base, *b_base;
 	int size;
@@ -4071,12 +4171,11 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_lut->crtc_id);
+	if (!crtc) {
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 
 	/* memcpy into gamma store */
 	if (crtc_lut->gamma_size != crtc->gamma_size) {
@@ -4129,7 +4228,6 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 			     void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_crtc_page_flip *page_flip = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
 	struct drm_pending_vblank_event *e = NULL;
@@ -4143,12 +4241,11 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 	if ((page_flip->flags & DRM_MODE_PAGE_FLIP_ASYNC) && !dev->mode_config.async_page_flip)
 		return -EINVAL;
 
-	obj = drm_mode_object_find(dev, page_flip->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj)
+	crtc = drm_crtc_find(dev, page_flip->crtc_id);
+	if (!crtc)
 		return -ENOENT;
-	crtc = obj_to_crtc(obj);
 
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	if (crtc->primary->fb == NULL) {
 		/* The framebuffer is currently unbound, presumably
 		 * due to a hotplug event, that userspace has not
@@ -4232,7 +4329,7 @@ out:
 		drm_framebuffer_unreference(fb);
 	if (old_fb)
 		drm_framebuffer_unreference(old_fb);
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 
 	return ret;
 }
@@ -4597,6 +4694,7 @@ EXPORT_SYMBOL(drm_format_vert_chroma_subsampling);
 void drm_mode_config_init(struct drm_device *dev)
 {
 	mutex_init(&dev->mode_config.mutex);
+	drm_modeset_lock_init(&dev->mode_config.connection_mutex);
 	mutex_init(&dev->mode_config.idr_mutex);
 	mutex_init(&dev->mode_config.fb_lock);
 	INIT_LIST_HEAD(&dev->mode_config.fb_list);
@@ -4696,5 +4794,6 @@ void drm_mode_config_cleanup(struct drm_device *dev)
 	}
 
 	idr_destroy(&dev->mode_config.crtc_idr);
+	drm_modeset_lock_fini(&dev->mode_config.connection_mutex);
 }
 EXPORT_SYMBOL(drm_mode_config_cleanup);
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 872ba11c4533..78b37f3febd3 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -93,8 +93,10 @@ bool drm_helper_encoder_in_use(struct drm_encoder *encoder)
 	 * We can expect this mutex to be locked if we are not panicking.
 	 * Locking is currently fubar in the panic handler.
 	 */
-	if (!oops_in_progress)
+	if (!oops_in_progress) {
 		WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+		WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+	}
 
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->encoder == encoder)
@@ -153,20 +155,14 @@ drm_encoder_disable(struct drm_encoder *encoder)
 static void __drm_helper_disable_unused_functions(struct drm_device *dev)
 {
 	struct drm_encoder *encoder;
-	struct drm_connector *connector;
 	struct drm_crtc *crtc;
 
 	drm_warn_on_modeset_not_all_locked(dev);
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		if (!connector->encoder)
-			continue;
-	}
-
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
 		if (!drm_helper_encoder_in_use(encoder)) {
 			drm_encoder_disable(encoder);
-			/* disconnector encoder from any connector */
+			/* disconnect encoder from any connector */
 			encoder->crtc = NULL;
 		}
 	}
@@ -349,7 +345,7 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 			continue;
 
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] set [MODE:%d:%s]\n",
-			encoder->base.id, drm_get_encoder_name(encoder),
+			encoder->base.id, encoder->name,
 			mode->base.id, mode->name);
 		encoder_funcs = encoder->helper_private;
 		encoder_funcs->mode_set(encoder, mode, adjusted_mode);
@@ -400,8 +396,7 @@ done:
 }
 EXPORT_SYMBOL(drm_crtc_helper_set_mode);
 
-
-static int
+static void
 drm_crtc_helper_disable(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -430,7 +425,6 @@ drm_crtc_helper_disable(struct drm_crtc *crtc)
 	}
 
 	__drm_helper_disable_unused_functions(dev);
-	return 0;
 }
 
 /**
@@ -481,7 +475,8 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 				(int)set->num_connectors, set->x, set->y);
 	} else {
 		DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id);
-		return drm_crtc_helper_disable(set->crtc);
+		drm_crtc_helper_disable(set->crtc);
+		return 0;
 	}
 
 	dev = set->crtc->dev;
@@ -620,11 +615,11 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 		}
 		if (new_crtc) {
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n",
-				connector->base.id, drm_get_connector_name(connector),
+				connector->base.id, connector->name,
 				new_crtc->base.id);
 		} else {
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n",
-				connector->base.id, drm_get_connector_name(connector));
+				connector->base.id, connector->name);
 		}
 	}
 
@@ -650,7 +645,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 			DRM_DEBUG_KMS("Setting connector DPMS state to on\n");
 			for (i = 0; i < set->num_connectors; i++) {
 				DRM_DEBUG_KMS("\t[CONNECTOR:%d:%s] set DPMS on\n", set->connectors[i]->base.id,
-					      drm_get_connector_name(set->connectors[i]));
+					      set->connectors[i]->name);
 				set->connectors[i]->funcs->dpms(set->connectors[i], DRM_MODE_DPMS_ON);
 			}
 		}
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 4b6e6f3ba0a1..08e33b8b13a4 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -206,13 +206,17 @@ i2c_dp_aux_prepare_bus(struct i2c_adapter *adapter)
  * i2c_dp_aux_add_bus() - register an i2c adapter using the aux ch helper
  * @adapter: i2c adapter to register
  *
- * This registers an i2c adapater that uses dp aux channel as it's underlaying
+ * This registers an i2c adapter that uses dp aux channel as it's underlaying
  * transport. The driver needs to fill out the &i2c_algo_dp_aux_data structure
  * and store it in the algo_data member of the @adapter argument. This will be
  * used by the i2c over dp aux algorithm to drive the hardware.
  *
  * RETURNS:
  * 0 on success, -ERRNO on failure.
+ *
+ * IMPORTANT:
+ * This interface is deprecated, please switch to the new dp aux helpers and
+ * drm_dp_aux_register().
  */
 int
 i2c_dp_aux_add_bus(struct i2c_adapter *adapter)
@@ -378,7 +382,10 @@ static int drm_dp_dpcd_access(struct drm_dp_aux *aux, u8 request,
 	 * transactions.
 	 */
 	for (retry = 0; retry < 7; retry++) {
+
+		mutex_lock(&aux->hw_mutex);
 		err = aux->transfer(aux, &msg);
+		mutex_unlock(&aux->hw_mutex);
 		if (err < 0) {
 			if (err == -EBUSY)
 				continue;
@@ -592,7 +599,9 @@ static int drm_dp_i2c_do_msg(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	 * before giving up the AUX transaction.
 	 */
 	for (retry = 0; retry < 7; retry++) {
+		mutex_lock(&aux->hw_mutex);
 		err = aux->transfer(aux, msg);
+		mutex_unlock(&aux->hw_mutex);
 		if (err < 0) {
 			if (err == -EBUSY)
 				continue;
@@ -725,13 +734,15 @@ static const struct i2c_algorithm drm_dp_i2c_algo = {
 };
 
 /**
- * drm_dp_aux_register_i2c_bus() - register an I2C adapter for I2C-over-AUX
+ * drm_dp_aux_register() - initialise and register aux channel
  * @aux: DisplayPort AUX channel
  *
  * Returns 0 on success or a negative error code on failure.
  */
-int drm_dp_aux_register_i2c_bus(struct drm_dp_aux *aux)
+int drm_dp_aux_register(struct drm_dp_aux *aux)
 {
+	mutex_init(&aux->hw_mutex);
+
 	aux->ddc.algo = &drm_dp_i2c_algo;
 	aux->ddc.algo_data = aux;
 	aux->ddc.retries = 3;
@@ -746,14 +757,14 @@ int drm_dp_aux_register_i2c_bus(struct drm_dp_aux *aux)
 
 	return i2c_add_adapter(&aux->ddc);
 }
-EXPORT_SYMBOL(drm_dp_aux_register_i2c_bus);
+EXPORT_SYMBOL(drm_dp_aux_register);
 
 /**
- * drm_dp_aux_unregister_i2c_bus() - unregister an I2C-over-AUX adapter
+ * drm_dp_aux_unregister() - unregister an AUX adapter
  * @aux: DisplayPort AUX channel
  */
-void drm_dp_aux_unregister_i2c_bus(struct drm_dp_aux *aux)
+void drm_dp_aux_unregister(struct drm_dp_aux *aux)
 {
 	i2c_del_adapter(&aux->ddc);
 }
-EXPORT_SYMBOL(drm_dp_aux_unregister_i2c_bus);
+EXPORT_SYMBOL(drm_dp_aux_unregister);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 03711d00aaae..8218078b6133 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -419,8 +419,9 @@ long drm_ioctl(struct file *filp,
 			retcode = -EFAULT;
 			goto err_i1;
 		}
-	} else
+	} else if (cmd & IOC_OUT) {
 		memset(kdata, 0, usize);
+	}
 
 	if (ioctl->flags & DRM_UNLOCKED)
 		retcode = func(dev, kdata, file_priv);
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index d4e3f9d9370f..dfa9769b26b5 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -70,6 +70,8 @@
 #define EDID_QUIRK_FORCE_REDUCED_BLANKING	(1 << 7)
 /* Force 8bpc */
 #define EDID_QUIRK_FORCE_8BPC			(1 << 8)
+/* Force 12bpc */
+#define EDID_QUIRK_FORCE_12BPC			(1 << 9)
 
 struct detailed_mode_closure {
 	struct drm_connector *connector;
@@ -125,6 +127,9 @@ static struct edid_quirk {
 	{ "SAM", 596, EDID_QUIRK_PREFER_LARGE_60 },
 	{ "SAM", 638, EDID_QUIRK_PREFER_LARGE_60 },
 
+	/* Sony PVM-2541A does up to 12 bpc, but only reports max 8 bpc */
+	{ "SNY", 0x2541, EDID_QUIRK_FORCE_12BPC },
+
 	/* ViewSonic VA2026w */
 	{ "VSC", 5020, EDID_QUIRK_FORCE_REDUCED_BLANKING },
 
@@ -984,9 +989,13 @@ static const u8 edid_header[] = {
 	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00
 };
 
- /*
- * Sanity check the header of the base EDID block.  Return 8 if the header
- * is perfect, down to 0 if it's totally wrong.
+/**
+ * drm_edid_header_is_valid - sanity check the header of the base EDID block
+ * @raw_edid: pointer to raw base EDID block
+ *
+ * Sanity check the header of the base EDID block.
+ *
+ * Return: 8 if the header is perfect, down to 0 if it's totally wrong.
  */
 int drm_edid_header_is_valid(const u8 *raw_edid)
 {
@@ -1005,9 +1014,16 @@ module_param_named(edid_fixup, edid_fixup, int, 0400);
 MODULE_PARM_DESC(edid_fixup,
 		 "Minimum number of valid EDID header bytes (0-8, default 6)");
 
-/*
- * Sanity check the EDID block (base or extension).  Return 0 if the block
- * doesn't check out, or 1 if it's valid.
+/**
+ * drm_edid_block_valid - Sanity check the EDID block (base or extension)
+ * @raw_edid: pointer to raw EDID block
+ * @block: type of block to validate (0 for base, extension otherwise)
+ * @print_bad_edid: if true, dump bad EDID blocks to the console
+ *
+ * Validate a base or extension EDID block and optionally dump bad blocks to
+ * the console.
+ *
+ * Return: True if the block is valid, false otherwise.
  */
 bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid)
 {
@@ -1077,6 +1093,8 @@ EXPORT_SYMBOL(drm_edid_block_valid);
  * @edid: EDID data
  *
  * Sanity-check an entire EDID record (including extensions)
+ *
+ * Return: True if the EDID data is valid, false otherwise.
  */
 bool drm_edid_is_valid(struct edid *edid)
 {
@@ -1096,18 +1114,15 @@ EXPORT_SYMBOL(drm_edid_is_valid);
 
 #define DDC_SEGMENT_ADDR 0x30
 /**
- * Get EDID information via I2C.
- *
- * @adapter : i2c device adaptor
+ * drm_do_probe_ddc_edid() - get EDID information via I2C
+ * @adapter: I2C device adaptor
  * @buf: EDID data buffer to be filled
  * @block: 128 byte EDID block to start fetching from
  * @len: EDID data buffer length to fetch
  *
- * Returns:
- *
- * 0 on success or -1 on failure.
+ * Try to fetch EDID information by calling I2C driver functions.
  *
- * Try to fetch EDID information by calling i2c driver function.
+ * Return: 0 on success or -1 on failure.
  */
 static int
 drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
@@ -1118,7 +1133,8 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
 	unsigned char xfers = segment ? 3 : 2;
 	int ret, retries = 5;
 
-	/* The core i2c driver will automatically retry the transfer if the
+	/*
+	 * The core I2C driver will automatically retry the transfer if the
 	 * adapter reports EAGAIN. However, we find that bit-banging transfers
 	 * are susceptible to errors under a heavily loaded machine and
 	 * generate spurious NAKs and timeouts. Retrying the transfer
@@ -1144,10 +1160,10 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
 			}
 		};
 
-	/*
-	 * Avoid sending the segment addr to not upset non-compliant ddc
-	 * monitors.
-	 */
+		/*
+		 * Avoid sending the segment addr to not upset non-compliant
+		 * DDC monitors.
+		 */
 		ret = i2c_transfer(adapter, &msgs[3 - xfers], xfers);
 
 		if (ret == -ENXIO) {
@@ -1216,7 +1232,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 		if (i == 4 && print_bad_edid) {
 			dev_warn(connector->dev->dev,
 			 "%s: Ignoring invalid EDID block %d.\n",
-			 drm_get_connector_name(connector), j);
+			 connector->name, j);
 
 			connector->bad_edid_counter++;
 		}
@@ -1236,7 +1252,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 carp:
 	if (print_bad_edid) {
 		dev_warn(connector->dev->dev, "%s: EDID block %d invalid.\n",
-			 drm_get_connector_name(connector), j);
+			 connector->name, j);
 	}
 	connector->bad_edid_counter++;
 
@@ -1246,12 +1262,10 @@ out:
 }
 
 /**
- * Probe DDC presence.
- * @adapter: i2c adapter to probe
+ * drm_probe_ddc() - probe DDC presence
+ * @adapter: I2C adapter to probe
  *
- * Returns:
- *
- * 1 on success
+ * Return: True on success, false on failure.
  */
 bool
 drm_probe_ddc(struct i2c_adapter *adapter)
@@ -1265,12 +1279,12 @@ EXPORT_SYMBOL(drm_probe_ddc);
 /**
  * drm_get_edid - get EDID data, if available
  * @connector: connector we're probing
- * @adapter: i2c adapter to use for DDC
+ * @adapter: I2C adapter to use for DDC
  *
- * Poke the given i2c channel to grab EDID data if possible.  If found,
+ * Poke the given I2C channel to grab EDID data if possible.  If found,
  * attach it to the connector.
  *
- * Return edid data or NULL if we couldn't find any.
+ * Return: Pointer to valid EDID or NULL if we couldn't find any.
  */
 struct edid *drm_get_edid(struct drm_connector *connector,
 			  struct i2c_adapter *adapter)
@@ -1288,7 +1302,7 @@ EXPORT_SYMBOL(drm_get_edid);
  * drm_edid_duplicate - duplicate an EDID and the extensions
  * @edid: EDID to duplicate
  *
- * Return duplicate edid or NULL on allocation failure.
+ * Return: Pointer to duplicated EDID or NULL on allocation failure.
  */
 struct edid *drm_edid_duplicate(const struct edid *edid)
 {
@@ -1411,7 +1425,8 @@ mode_is_rb(const struct drm_display_mode *mode)
  * @rb: Mode reduced-blanking-ness
  *
  * Walk the DMT mode list looking for a match for the given parameters.
- * Return a newly allocated copy of the mode, or NULL if not found.
+ *
+ * Return: A newly allocated copy of the mode, or NULL if not found.
  */
 struct drm_display_mode *drm_mode_find_dmt(struct drm_device *dev,
 					   int hsize, int vsize, int fresh,
@@ -1595,14 +1610,13 @@ bad_std_timing(u8 a, u8 b)
  * @connector: connector of for the EDID block
  * @edid: EDID block to scan
  * @t: standard timing params
- * @revision: standard timing level
  *
  * Take the standard timing params (in this case width, aspect, and refresh)
  * and convert them into a real mode using CVT/GTF/DMT.
  */
 static struct drm_display_mode *
 drm_mode_std(struct drm_connector *connector, struct edid *edid,
-	     struct std_timing *t, int revision)
+	     struct std_timing *t)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_display_mode *m, *mode = NULL;
@@ -1623,7 +1637,7 @@ drm_mode_std(struct drm_connector *connector, struct edid *edid,
 	vrefresh_rate = vfreq + 60;
 	/* the vdisplay is calculated based on the aspect ratio */
 	if (aspect_ratio == 0) {
-		if (revision < 3)
+		if (edid->revision < 3)
 			vsize = hsize;
 		else
 			vsize = (hsize * 10) / 16;
@@ -2140,7 +2154,7 @@ do_established_modes(struct detailed_timing *timing, void *c)
 
 /**
  * add_established_modes - get est. modes from EDID and add them
- * @connector: connector of for the EDID block
+ * @connector: connector to add mode(s) to
  * @edid: EDID block to scan
  *
  * Each EDID block contains a bitmap of the supported "established modes" list
@@ -2191,8 +2205,7 @@ do_standard_modes(struct detailed_timing *timing, void *c)
 			struct drm_display_mode *newmode;
 
 			std = &data->data.timings[i];
-			newmode = drm_mode_std(connector, edid, std,
-					       edid->revision);
+			newmode = drm_mode_std(connector, edid, std);
 			if (newmode) {
 				drm_mode_probed_add(connector, newmode);
 				closure->modes++;
@@ -2203,7 +2216,7 @@ do_standard_modes(struct detailed_timing *timing, void *c)
 
 /**
  * add_standard_modes - get std. modes from EDID and add them
- * @connector: connector of for the EDID block
+ * @connector: connector to add mode(s) to
  * @edid: EDID block to scan
  *
  * Standard modes can be calculated using the appropriate standard (DMT,
@@ -2221,8 +2234,7 @@ add_standard_modes(struct drm_connector *connector, struct edid *edid)
 		struct drm_display_mode *newmode;
 
 		newmode = drm_mode_std(connector, edid,
-				       &edid->standard_timings[i],
-				       edid->revision);
+				       &edid->standard_timings[i]);
 		if (newmode) {
 			drm_mode_probed_add(connector, newmode);
 			modes++;
@@ -2425,7 +2437,7 @@ cea_mode_alternate_clock(const struct drm_display_mode *cea_mode)
  * drm_match_cea_mode - look for a CEA mode matching given mode
  * @to_match: display mode
  *
- * Returns the CEA Video ID (VIC) of the mode or 0 if it isn't a CEA-861
+ * Return: The CEA Video ID (VIC) of the mode or 0 if it isn't a CEA-861
  * mode.
  */
 u8 drm_match_cea_mode(const struct drm_display_mode *to_match)
@@ -2452,6 +2464,22 @@ u8 drm_match_cea_mode(const struct drm_display_mode *to_match)
 }
 EXPORT_SYMBOL(drm_match_cea_mode);
 
+/**
+ * drm_get_cea_aspect_ratio - get the picture aspect ratio corresponding to
+ * the input VIC from the CEA mode list
+ * @video_code: ID given to each of the CEA modes
+ *
+ * Returns picture aspect ratio
+ */
+enum hdmi_picture_aspect drm_get_cea_aspect_ratio(const u8 video_code)
+{
+	/* return picture aspect ratio for video_code - 1 to access the
+	 * right array element
+	*/
+	return edid_cea_modes[video_code-1].picture_aspect_ratio;
+}
+EXPORT_SYMBOL(drm_get_cea_aspect_ratio);
+
 /*
  * Calculate the alternate clock for HDMI modes (those from the HDMI vendor
  * specific block).
@@ -3023,11 +3051,9 @@ monitor_name(struct detailed_timing *t, void *data)
  * @connector: connector corresponding to the HDMI/DP sink
  * @edid: EDID to parse
  *
- * Fill the ELD (EDID-Like Data) buffer for passing to the audio driver.
- * Some ELD fields are left to the graphics driver caller:
- * - Conn_Type
- * - HDCP
- * - Port_ID
+ * Fill the ELD (EDID-Like Data) buffer for passing to the audio driver. The
+ * Conn_Type, HDCP and Port_ID ELD fields are left for the graphics driver to
+ * fill in.
  */
 void drm_edid_to_eld(struct drm_connector *connector, struct edid *edid)
 {
@@ -3111,9 +3137,10 @@ EXPORT_SYMBOL(drm_edid_to_eld);
  * @sads: pointer that will be set to the extracted SADs
  *
  * Looks for CEA EDID block and extracts SADs (Short Audio Descriptors) from it.
- * Note: returned pointer needs to be kfreed
  *
- * Return number of found SADs or negative number on error.
+ * Note: The returned pointer needs to be freed using kfree().
+ *
+ * Return: The number of found SADs or negative number on error.
  */
 int drm_edid_to_sad(struct edid *edid, struct cea_sad **sads)
 {
@@ -3170,9 +3197,11 @@ EXPORT_SYMBOL(drm_edid_to_sad);
  * @sadb: pointer to the speaker block
  *
  * Looks for CEA EDID block and extracts the Speaker Allocation Data Block from it.
- * Note: returned pointer needs to be kfreed
  *
- * Return number of found Speaker Allocation Blocks or negative number on error.
+ * Note: The returned pointer needs to be freed using kfree().
+ *
+ * Return: The number of found Speaker Allocation Blocks or negative number on
+ * error.
  */
 int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb)
 {
@@ -3204,10 +3233,9 @@ int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb)
 
 			/* Speaker Allocation Data Block */
 			if (dbl == 3) {
-				*sadb = kmalloc(dbl, GFP_KERNEL);
+				*sadb = kmemdup(&db[1], dbl, GFP_KERNEL);
 				if (!*sadb)
 					return -ENOMEM;
-				memcpy(*sadb, &db[1], dbl);
 				count = dbl;
 				break;
 			}
@@ -3219,9 +3247,12 @@ int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb)
 EXPORT_SYMBOL(drm_edid_to_speaker_allocation);
 
 /**
- * drm_av_sync_delay - HDMI/DP sink audio-video sync delay in millisecond
+ * drm_av_sync_delay - compute the HDMI/DP sink audio-video sync delay
  * @connector: connector associated with the HDMI/DP sink
  * @mode: the display mode
+ *
+ * Return: The HDMI/DP sink's audio-video sync delay in milliseconds or 0 if
+ * the sink doesn't support audio or video.
  */
 int drm_av_sync_delay(struct drm_connector *connector,
 		      struct drm_display_mode *mode)
@@ -3263,6 +3294,9 @@ EXPORT_SYMBOL(drm_av_sync_delay);
  *
  * It's possible for one encoder to be associated with multiple HDMI/DP sinks.
  * The policy is now hard coded to simply use the first HDMI/DP sink's ELD.
+ *
+ * Return: The connector associated with the first HDMI/DP sink that has ELD
+ * attached to it.
  */
 struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
 				     struct drm_display_mode *mode)
@@ -3270,6 +3304,8 @@ struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
 	struct drm_connector *connector;
 	struct drm_device *dev = encoder->dev;
 
+	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->encoder == encoder && connector->eld[0])
 			return connector;
@@ -3279,11 +3315,12 @@ struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
 EXPORT_SYMBOL(drm_select_eld);
 
 /**
- * drm_detect_hdmi_monitor - detect whether monitor is hdmi.
+ * drm_detect_hdmi_monitor - detect whether monitor is HDMI
  * @edid: monitor EDID information
  *
  * Parse the CEA extension according to CEA-861-B.
- * Return true if HDMI, false if not or unknown.
+ *
+ * Return: True if the monitor is HDMI, false if not or unknown.
  */
 bool drm_detect_hdmi_monitor(struct edid *edid)
 {
@@ -3321,6 +3358,7 @@ EXPORT_SYMBOL(drm_detect_hdmi_monitor);
  * audio format, assume at least 'basic audio' support, even if 'basic
  * audio' is not defined in EDID.
  *
+ * Return: True if the monitor supports audio, false otherwise.
  */
 bool drm_detect_monitor_audio(struct edid *edid)
 {
@@ -3364,6 +3402,8 @@ EXPORT_SYMBOL(drm_detect_monitor_audio);
  * Check whether the monitor reports the RGB quantization range selection
  * as supported. The AVI infoframe can then be used to inform the monitor
  * which quantization range (full or limited) is used.
+ *
+ * Return: True if the RGB quantization range is selectable, false otherwise.
  */
 bool drm_rgb_quant_range_selectable(struct edid *edid)
 {
@@ -3390,16 +3430,119 @@ bool drm_rgb_quant_range_selectable(struct edid *edid)
 EXPORT_SYMBOL(drm_rgb_quant_range_selectable);
 
 /**
+ * drm_assign_hdmi_deep_color_info - detect whether monitor supports
+ * hdmi deep color modes and update drm_display_info if so.
+ *
+ * @edid: monitor EDID information
+ * @info: Updated with maximum supported deep color bpc and color format
+ *        if deep color supported.
+ *
+ * Parse the CEA extension according to CEA-861-B.
+ * Return true if HDMI deep color supported, false if not or unknown.
+ */
+static bool drm_assign_hdmi_deep_color_info(struct edid *edid,
+                                            struct drm_display_info *info,
+                                            struct drm_connector *connector)
+{
+	u8 *edid_ext, *hdmi;
+	int i;
+	int start_offset, end_offset;
+	unsigned int dc_bpc = 0;
+
+	edid_ext = drm_find_cea_extension(edid);
+	if (!edid_ext)
+		return false;
+
+	if (cea_db_offsets(edid_ext, &start_offset, &end_offset))
+		return false;
+
+	/*
+	 * Because HDMI identifier is in Vendor Specific Block,
+	 * search it from all data blocks of CEA extension.
+	 */
+	for_each_cea_db(edid_ext, i, start_offset, end_offset) {
+		if (cea_db_is_hdmi_vsdb(&edid_ext[i])) {
+			/* HDMI supports at least 8 bpc */
+			info->bpc = 8;
+
+			hdmi = &edid_ext[i];
+			if (cea_db_payload_len(hdmi) < 6)
+				return false;
+
+			if (hdmi[6] & DRM_EDID_HDMI_DC_30) {
+				dc_bpc = 10;
+				info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_30;
+				DRM_DEBUG("%s: HDMI sink does deep color 30.\n",
+						  connector->name);
+			}
+
+			if (hdmi[6] & DRM_EDID_HDMI_DC_36) {
+				dc_bpc = 12;
+				info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_36;
+				DRM_DEBUG("%s: HDMI sink does deep color 36.\n",
+						  connector->name);
+			}
+
+			if (hdmi[6] & DRM_EDID_HDMI_DC_48) {
+				dc_bpc = 16;
+				info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_48;
+				DRM_DEBUG("%s: HDMI sink does deep color 48.\n",
+						  connector->name);
+			}
+
+			if (dc_bpc > 0) {
+				DRM_DEBUG("%s: Assigning HDMI sink color depth as %d bpc.\n",
+						  connector->name, dc_bpc);
+				info->bpc = dc_bpc;
+
+				/*
+				 * Deep color support mandates RGB444 support for all video
+				 * modes and forbids YCRCB422 support for all video modes per
+				 * HDMI 1.3 spec.
+				 */
+				info->color_formats = DRM_COLOR_FORMAT_RGB444;
+
+				/* YCRCB444 is optional according to spec. */
+				if (hdmi[6] & DRM_EDID_HDMI_DC_Y444) {
+					info->color_formats |= DRM_COLOR_FORMAT_YCRCB444;
+					DRM_DEBUG("%s: HDMI sink does YCRCB444 in deep color.\n",
+							  connector->name);
+				}
+
+				/*
+				 * Spec says that if any deep color mode is supported at all,
+				 * then deep color 36 bit must be supported.
+				 */
+				if (!(hdmi[6] & DRM_EDID_HDMI_DC_36)) {
+					DRM_DEBUG("%s: HDMI sink should do DC_36, but does not!\n",
+							  connector->name);
+				}
+
+				return true;
+			}
+			else {
+				DRM_DEBUG("%s: No deep color support on this HDMI sink.\n",
+						  connector->name);
+			}
+		}
+	}
+
+	return false;
+}
+
+/**
  * drm_add_display_info - pull display info out if present
  * @edid: EDID data
  * @info: display info (attached to connector)
+ * @connector: connector whose edid is used to build display info
  *
  * Grab any available display info and stuff it into the drm_display_info
  * structure that's part of the connector.  Useful for tracking bpp and
  * color spaces.
  */
 static void drm_add_display_info(struct edid *edid,
-				 struct drm_display_info *info)
+                                 struct drm_display_info *info,
+                                 struct drm_connector *connector)
 {
 	u8 *edid_ext;
 
@@ -3429,6 +3572,9 @@ static void drm_add_display_info(struct edid *edid,
 			info->color_formats |= DRM_COLOR_FORMAT_YCRCB422;
 	}
 
+	/* HDMI deep color modes supported? Assign to info, if so */
+	drm_assign_hdmi_deep_color_info(edid, info, connector);
+
 	/* Only defined for 1.4 with digital displays */
 	if (edid->revision < 4)
 		return;
@@ -3458,6 +3604,9 @@ static void drm_add_display_info(struct edid *edid,
 		break;
 	}
 
+	DRM_DEBUG("%s: Assigning EDID-1.4 digital sink color depth as %d bpc.\n",
+			  connector->name, info->bpc);
+
 	info->color_formats |= DRM_COLOR_FORMAT_RGB444;
 	if (edid->features & DRM_EDID_FEATURE_RGB_YCRCB444)
 		info->color_formats |= DRM_COLOR_FORMAT_YCRCB444;
@@ -3468,11 +3617,11 @@ static void drm_add_display_info(struct edid *edid,
 /**
  * drm_add_edid_modes - add modes from EDID data, if available
  * @connector: connector we're probing
- * @edid: edid data
+ * @edid: EDID data
  *
  * Add the specified modes to the connector's mode list.
  *
- * Return number of modes added or 0 if we couldn't find any.
+ * Return: The number of modes added or 0 if we couldn't find any.
  */
 int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 {
@@ -3484,7 +3633,7 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 	}
 	if (!drm_edid_is_valid(edid)) {
 		dev_warn(connector->dev->dev, "%s: EDID invalid.\n",
-			 drm_get_connector_name(connector));
+			 connector->name);
 		return 0;
 	}
 
@@ -3516,11 +3665,14 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 	if (quirks & (EDID_QUIRK_PREFER_LARGE_60 | EDID_QUIRK_PREFER_LARGE_75))
 		edid_fixup_preferred(connector, quirks);
 
-	drm_add_display_info(edid, &connector->display_info);
+	drm_add_display_info(edid, &connector->display_info, connector);
 
 	if (quirks & EDID_QUIRK_FORCE_8BPC)
 		connector->display_info.bpc = 8;
 
+	if (quirks & EDID_QUIRK_FORCE_12BPC)
+		connector->display_info.bpc = 12;
+
 	return num_modes;
 }
 EXPORT_SYMBOL(drm_add_edid_modes);
@@ -3534,7 +3686,7 @@ EXPORT_SYMBOL(drm_add_edid_modes);
  * Add the specified modes to the connector's mode list. Only when the
  * hdisplay/vdisplay is not beyond the given limit, it will be added.
  *
- * Return number of modes added or 0 if we couldn't find any.
+ * Return: The number of modes added or 0 if we couldn't find any.
  */
 int drm_add_modes_noedid(struct drm_connector *connector,
 			int hdisplay, int vdisplay)
@@ -3573,13 +3725,22 @@ int drm_add_modes_noedid(struct drm_connector *connector,
 }
 EXPORT_SYMBOL(drm_add_modes_noedid);
 
+/**
+ * drm_set_preferred_mode - Sets the preferred mode of a connector
+ * @connector: connector whose mode list should be processed
+ * @hpref: horizontal resolution of preferred mode
+ * @vpref: vertical resolution of preferred mode
+ *
+ * Marks a mode as preferred if it matches the resolution specified by @hpref
+ * and @vpref.
+ */
 void drm_set_preferred_mode(struct drm_connector *connector,
 			   int hpref, int vpref)
 {
 	struct drm_display_mode *mode;
 
 	list_for_each_entry(mode, &connector->probed_modes, head) {
-		if (mode->hdisplay  == hpref &&
+		if (mode->hdisplay == hpref &&
 		    mode->vdisplay == vpref)
 			mode->type |= DRM_MODE_TYPE_PREFERRED;
 	}
@@ -3592,7 +3753,7 @@ EXPORT_SYMBOL(drm_set_preferred_mode);
  * @frame: HDMI AVI infoframe
  * @mode: DRM display mode
  *
- * Returns 0 on success or a negative error code on failure.
+ * Return: 0 on success or a negative error code on failure.
  */
 int
 drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
@@ -3613,6 +3774,12 @@ drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
 	frame->video_code = drm_match_cea_mode(mode);
 
 	frame->picture_aspect = HDMI_PICTURE_ASPECT_NONE;
+
+	/* Populate picture aspect ratio from CEA mode list */
+	if (frame->video_code > 0)
+		frame->picture_aspect = drm_get_cea_aspect_ratio(
+						frame->video_code);
+
 	frame->active_aspect = HDMI_ACTIVE_ASPECT_PICTURE;
 	frame->scan_mode = HDMI_SCAN_MODE_UNDERSCAN;
 
@@ -3657,7 +3824,7 @@ s3d_structure_from_display_mode(const struct drm_display_mode *mode)
  * 4k or stereoscopic 3D mode. So when giving any other mode as input this
  * function will return -EINVAL, error that can be safely ignored.
  *
- * Returns 0 on success or a negative error code on failure.
+ * Return: 0 on success or a negative error code on failure.
  */
 int
 drm_hdmi_vendor_infoframe_from_display_mode(struct hdmi_vendor_infoframe *frame,
diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c
index 1b4c7a5442c5..0a235fe61c9b 100644
--- a/drivers/gpu/drm/drm_edid_load.c
+++ b/drivers/gpu/drm/drm_edid_load.c
@@ -31,8 +31,9 @@ module_param_string(edid_firmware, edid_firmware, sizeof(edid_firmware), 0644);
 MODULE_PARM_DESC(edid_firmware, "Do not probe monitor, use specified EDID blob "
 	"from built-in data or /lib/firmware instead. ");
 
-#define GENERIC_EDIDS 5
+#define GENERIC_EDIDS 6
 static const char *generic_edid_name[GENERIC_EDIDS] = {
+	"edid/800x600.bin",
 	"edid/1024x768.bin",
 	"edid/1280x1024.bin",
 	"edid/1600x1200.bin",
@@ -44,6 +45,24 @@ static const u8 generic_edid[GENERIC_EDIDS][128] = {
 	{
 	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00,
 	0x31, 0xd8, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x05, 0x16, 0x01, 0x03, 0x6d, 0x1b, 0x14, 0x78,
+	0xea, 0x5e, 0xc0, 0xa4, 0x59, 0x4a, 0x98, 0x25,
+	0x20, 0x50, 0x54, 0x01, 0x00, 0x00, 0x45, 0x40,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0xa0, 0x0f,
+	0x20, 0x00, 0x31, 0x58, 0x1c, 0x20, 0x28, 0x80,
+	0x14, 0x00, 0x15, 0xd0, 0x10, 0x00, 0x00, 0x1e,
+	0x00, 0x00, 0x00, 0xff, 0x00, 0x4c, 0x69, 0x6e,
+	0x75, 0x78, 0x20, 0x23, 0x30, 0x0a, 0x20, 0x20,
+	0x20, 0x20, 0x00, 0x00, 0x00, 0xfd, 0x00, 0x3b,
+	0x3d, 0x24, 0x26, 0x05, 0x00, 0x0a, 0x20, 0x20,
+	0x20, 0x20, 0x20, 0x20, 0x00, 0x00, 0x00, 0xfc,
+	0x00, 0x4c, 0x69, 0x6e, 0x75, 0x78, 0x20, 0x53,
+	0x56, 0x47, 0x41, 0x0a, 0x20, 0x20, 0x00, 0xc2,
+	},
+	{
+	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00,
+	0x31, 0xd8, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
 	0x05, 0x16, 0x01, 0x03, 0x6d, 0x23, 0x1a, 0x78,
 	0xea, 0x5e, 0xc0, 0xa4, 0x59, 0x4a, 0x98, 0x25,
 	0x20, 0x50, 0x54, 0x00, 0x08, 0x00, 0x61, 0x40,
@@ -242,7 +261,7 @@ out:
 
 int drm_load_edid_firmware(struct drm_connector *connector)
 {
-	const char *connector_name = drm_get_connector_name(connector);
+	const char *connector_name = connector->name;
 	char *edidname = edid_firmware, *last, *colon;
 	int ret;
 	struct edid *edid;
diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
index 61b5a47ad239..f27c883be391 100644
--- a/drivers/gpu/drm/drm_fb_cma_helper.c
+++ b/drivers/gpu/drm/drm_fb_cma_helper.c
@@ -429,13 +429,8 @@ EXPORT_SYMBOL_GPL(drm_fbdev_cma_fini);
  */
 void drm_fbdev_cma_restore_mode(struct drm_fbdev_cma *fbdev_cma)
 {
-	if (fbdev_cma) {
-		struct drm_device *dev = fbdev_cma->fb_helper.dev;
-
-		drm_modeset_lock_all(dev);
-		drm_fb_helper_restore_fbdev_mode(&fbdev_cma->fb_helper);
-		drm_modeset_unlock_all(dev);
-	}
+	if (fbdev_cma)
+		drm_fb_helper_restore_fbdev_mode_unlocked(&fbdev_cma->fb_helper);
 }
 EXPORT_SYMBOL_GPL(drm_fbdev_cma_restore_mode);
 
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 04d3fd3658f3..d5d8cea1a679 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -45,13 +45,13 @@ static LIST_HEAD(kernel_fb_helper_list);
  * DOC: fbdev helpers
  *
  * The fb helper functions are useful to provide an fbdev on top of a drm kernel
- * mode setting driver. They can be used mostly independantely from the crtc
+ * mode setting driver. They can be used mostly independently from the crtc
  * helper functions used by many drivers to implement the kernel mode setting
  * interfaces.
  *
  * Initialization is done as a three-step process with drm_fb_helper_init(),
  * drm_fb_helper_single_add_all_connectors() and drm_fb_helper_initial_config().
- * Drivers with fancier requirements than the default beheviour can override the
+ * Drivers with fancier requirements than the default behaviour can override the
  * second step with their own code.  Teardown is done with drm_fb_helper_fini().
  *
  * At runtime drivers should restore the fbdev console by calling
@@ -59,7 +59,7 @@ static LIST_HEAD(kernel_fb_helper_list);
  * should also notify the fb helper code from updates to the output
  * configuration by calling drm_fb_helper_hotplug_event(). For easier
  * integration with the output polling code in drm_crtc_helper.c the modeset
- * code proves a ->output_poll_changed callback.
+ * code provides a ->output_poll_changed callback.
  *
  * All other functions exported by the fb helper library can be used to
  * implement the fbdev driver interface by the driver.
@@ -120,7 +120,7 @@ static int drm_fb_helper_parse_command_line(struct drm_fb_helper *fb_helper)
 		mode = &fb_helper_conn->cmdline_mode;
 
 		/* do something on return - turn off connector maybe */
-		if (fb_get_options(drm_get_connector_name(connector), &option))
+		if (fb_get_options(connector->name, &option))
 			continue;
 
 		if (drm_mode_parse_command_line_for_connector(option,
@@ -142,12 +142,12 @@ static int drm_fb_helper_parse_command_line(struct drm_fb_helper *fb_helper)
 				}
 
 				DRM_INFO("forcing %s connector %s\n",
-					 drm_get_connector_name(connector), s);
+					 connector->name, s);
 				connector->force = mode->force;
 			}
 
 			DRM_DEBUG_KMS("cmdline mode for connector %s %dx%d@%dHz%s%s%s\n",
-				      drm_get_connector_name(connector),
+				      connector->name,
 				      mode->xres, mode->yres,
 				      mode->refresh_specified ? mode->refresh : 60,
 				      mode->rb ? " reduced blanking" : "",
@@ -273,15 +273,7 @@ int drm_fb_helper_debug_leave(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_debug_leave);
 
-/**
- * drm_fb_helper_restore_fbdev_mode - restore fbdev configuration
- * @fb_helper: fbcon to restore
- *
- * This should be called from driver's drm ->lastclose callback
- * when implementing an fbcon on top of kms using this helper. This ensures that
- * the user isn't greeted with a black screen when e.g. X dies.
- */
-bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
+static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
 	struct drm_plane *plane;
@@ -311,7 +303,40 @@ bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 	}
 	return error;
 }
-EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode);
+/**
+ * drm_fb_helper_restore_fbdev_mode - restore fbdev configuration
+ * @fb_helper: fbcon to restore
+ *
+ * This should be called from driver's drm ->lastclose callback
+ * when implementing an fbcon on top of kms using this helper. This ensures that
+ * the user isn't greeted with a black screen when e.g. X dies.
+ *
+ * Use this variant if you need to bypass locking (panic), or already
+ * hold all modeset locks.  Otherwise use drm_fb_helper_restore_fbdev_mode_unlocked()
+ */
+static bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
+{
+	return restore_fbdev_mode(fb_helper);
+}
+
+/**
+ * drm_fb_helper_restore_fbdev_mode_unlocked - restore fbdev configuration
+ * @fb_helper: fbcon to restore
+ *
+ * This should be called from driver's drm ->lastclose callback
+ * when implementing an fbcon on top of kms using this helper. This ensures that
+ * the user isn't greeted with a black screen when e.g. X dies.
+ */
+bool drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper)
+{
+	struct drm_device *dev = fb_helper->dev;
+	bool ret;
+	drm_modeset_lock_all(dev);
+	ret = restore_fbdev_mode(fb_helper);
+	drm_modeset_unlock_all(dev);
+	return ret;
+}
+EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode_unlocked);
 
 /*
  * restore fbcon display for all kms driver's using this helper, used for sysrq
@@ -326,12 +351,25 @@ static bool drm_fb_helper_force_kernel_mode(void)
 		return false;
 
 	list_for_each_entry(helper, &kernel_fb_helper_list, kernel_fb_list) {
-		if (helper->dev->switch_power_state == DRM_SWITCH_POWER_OFF)
+		struct drm_device *dev = helper->dev;
+
+		if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
+			continue;
+
+		/* NOTE: we use lockless flag below to avoid grabbing other
+		 * modeset locks.  So just trylock the underlying mutex
+		 * directly:
+		 */
+		if (!mutex_trylock(&dev->mode_config.mutex)) {
+			error = true;
 			continue;
+		}
 
 		ret = drm_fb_helper_restore_fbdev_mode(helper);
 		if (ret)
 			error = true;
+
+		mutex_unlock(&dev->mode_config.mutex);
 	}
 	return error;
 }
@@ -811,7 +849,6 @@ EXPORT_SYMBOL(drm_fb_helper_check_var);
 int drm_fb_helper_set_par(struct fb_info *info)
 {
 	struct drm_fb_helper *fb_helper = info->par;
-	struct drm_device *dev = fb_helper->dev;
 	struct fb_var_screeninfo *var = &info->var;
 
 	if (var->pixclock != 0) {
@@ -819,9 +856,7 @@ int drm_fb_helper_set_par(struct fb_info *info)
 		return -EINVAL;
 	}
 
-	drm_modeset_lock_all(dev);
-	drm_fb_helper_restore_fbdev_mode(fb_helper);
-	drm_modeset_unlock_all(dev);
+	drm_fb_helper_restore_fbdev_mode_unlocked(fb_helper);
 
 	if (fb_helper->delayed_hotplug) {
 		fb_helper->delayed_hotplug = false;
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index e1eba0b7cd45..021fe5d11df5 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -43,8 +43,7 @@
 DEFINE_MUTEX(drm_global_mutex);
 EXPORT_SYMBOL(drm_global_mutex);
 
-static int drm_open_helper(struct inode *inode, struct file *filp,
-			   struct drm_minor *minor);
+static int drm_open_helper(struct file *filp, struct drm_minor *minor);
 
 static int drm_setup(struct drm_device * dev)
 {
@@ -95,7 +94,7 @@ int drm_open(struct inode *inode, struct file *filp)
 	/* share address_space across all char-devs of a single device */
 	filp->f_mapping = dev->anon_inode->i_mapping;
 
-	retcode = drm_open_helper(inode, filp, minor);
+	retcode = drm_open_helper(filp, minor);
 	if (retcode)
 		goto err_undo;
 	if (need_setup) {
@@ -171,7 +170,6 @@ static int drm_cpu_valid(void)
 /**
  * Called whenever a process opens /dev/drm.
  *
- * \param inode device inode.
  * \param filp file pointer.
  * \param minor acquired minor-object.
  * \return zero on success or a negative number on failure.
@@ -179,8 +177,7 @@ static int drm_cpu_valid(void)
  * Creates and initializes a drm_file structure for the file private data in \p
  * filp and add it into the double linked list in \p dev.
  */
-static int drm_open_helper(struct inode *inode, struct file *filp,
-			   struct drm_minor *minor)
+static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 {
 	struct drm_device *dev = minor->dev;
 	struct drm_file *priv;
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 9909bef59800..f7d71190aad5 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -474,21 +474,10 @@ struct page **drm_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask)
 			goto fail;
 		pages[i] = p;
 
-		/* There is a hypothetical issue w/ drivers that require
-		 * buffer memory in the low 4GB.. if the pages are un-
-		 * pinned, and swapped out, they can end up swapped back
-		 * in above 4GB.  If pages are already in memory, then
-		 * shmem_read_mapping_page_gfp will ignore the gfpmask,
-		 * even if the already in-memory page disobeys the mask.
-		 *
-		 * It is only a theoretical issue today, because none of
-		 * the devices with this limitation can be populated with
-		 * enough memory to trigger the issue.  But this BUG_ON()
-		 * is here as a reminder in case the problem with
-		 * shmem_read_mapping_page_gfp() isn't solved by the time
-		 * it does become a real issue.
-		 *
-		 * See this thread: http://lkml.org/lkml/2011/7/11/238
+		/* Make sure shmem keeps __GFP_DMA32 allocated pages in the
+		 * correct region during swapin. Note that this requires
+		 * __GFP_DMA32 to be set in mapping_gfp_mask(inode->i_mapping)
+		 * so shmem can relocate pages during swapin if required.
 		 */
 		BUG_ON((gfpmask & __GFP_DMA32) &&
 				(page_to_pfn(p) >= 0x00100000UL));
diff --git a/drivers/gpu/drm/drm_info.c b/drivers/gpu/drm/drm_info.c
index 7473035dd28b..86feedd5e6f6 100644
--- a/drivers/gpu/drm/drm_info.c
+++ b/drivers/gpu/drm/drm_info.c
@@ -47,18 +47,16 @@ int drm_name_info(struct seq_file *m, void *data)
 	struct drm_minor *minor = node->minor;
 	struct drm_device *dev = minor->dev;
 	struct drm_master *master = minor->master;
-	const char *bus_name;
 	if (!master)
 		return 0;
 
-	bus_name = dev->driver->bus->get_name(dev);
 	if (master->unique) {
 		seq_printf(m, "%s %s %s\n",
-			   bus_name,
+			   dev->driver->name,
 			   dev_name(dev->dev), master->unique);
 	} else {
 		seq_printf(m, "%s %s\n",
-			   bus_name, dev_name(dev->dev));
+			   dev->driver->name, dev_name(dev->dev));
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 93a42040bedb..69c61f392e66 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -72,9 +72,6 @@ static void
 drm_unset_busid(struct drm_device *dev,
 		struct drm_master *master)
 {
-	kfree(dev->devname);
-	dev->devname = NULL;
-
 	kfree(master->unique);
 	master->unique = NULL;
 	master->unique_len = 0;
@@ -93,7 +90,8 @@ drm_unset_busid(struct drm_device *dev,
  * Copies the bus id from userspace into drm_device::unique, and verifies that
  * it matches the device this DRM is attached to (EINVAL otherwise).  Deprecated
  * in interface version 1.1 and will return EBUSY when setversion has requested
- * version 1.1 or greater.
+ * version 1.1 or greater. Also note that KMS is all version 1.1 and later and
+ * UMS was only ever supported on pci devices.
  */
 int drm_setunique(struct drm_device *dev, void *data,
 		  struct drm_file *file_priv)
@@ -108,10 +106,13 @@ int drm_setunique(struct drm_device *dev, void *data,
 	if (!u->unique_len || u->unique_len > 1024)
 		return -EINVAL;
 
-	if (!dev->driver->bus->set_unique)
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return 0;
+
+	if (WARN_ON(!dev->pdev))
 		return -EINVAL;
 
-	ret = dev->driver->bus->set_unique(dev, master, u);
+	ret = drm_pci_set_unique(dev, master, u);
 	if (ret)
 		goto err;
 
@@ -130,13 +131,25 @@ static int drm_set_busid(struct drm_device *dev, struct drm_file *file_priv)
 	if (master->unique != NULL)
 		drm_unset_busid(dev, master);
 
-	ret = dev->driver->bus->set_busid(dev, master);
-	if (ret)
-		goto err;
+	if (dev->driver->bus && dev->driver->bus->set_busid) {
+		ret = dev->driver->bus->set_busid(dev, master);
+		if (ret) {
+			drm_unset_busid(dev, master);
+			return ret;
+		}
+	} else {
+		if (WARN(dev->unique == NULL,
+			 "No drm_bus.set_busid() implementation provided by "
+			 "%ps. Use drm_dev_set_unique() to set the unique "
+			 "name explicitly.", dev->driver))
+			return -EINVAL;
+
+		master->unique = kstrdup(dev->unique, GFP_KERNEL);
+		if (master->unique)
+			master->unique_len = strlen(dev->unique);
+	}
+
 	return 0;
-err:
-	drm_unset_busid(dev, master);
-	return ret;
 }
 
 /**
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index ec5c3f4cdd01..0de123afdb34 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -1,6 +1,5 @@
-/**
- * \file drm_irq.c
- * IRQ support
+/*
+ * drm_irq.c IRQ and vblank support
  *
  * \author Rickard E. (Rik) Faith <faith@valinux.com>
  * \author Gareth Hughes <gareth@valinux.com>
@@ -56,33 +55,6 @@
  */
 #define DRM_REDUNDANT_VBLIRQ_THRESH_NS 1000000
 
-/**
- * Get interrupt from bus id.
- *
- * \param inode device inode.
- * \param file_priv DRM file private.
- * \param cmd command.
- * \param arg user argument, pointing to a drm_irq_busid structure.
- * \return zero on success or a negative number on failure.
- *
- * Finds the PCI device with the specified bus id and gets its IRQ number.
- * This IOCTL is deprecated, and will now return EINVAL for any busid not equal
- * to that of the device that this DRM instance attached to.
- */
-int drm_irq_by_busid(struct drm_device *dev, void *data,
-		     struct drm_file *file_priv)
-{
-	struct drm_irq_busid *p = data;
-
-	if (!dev->driver->bus->irq_by_busid)
-		return -EINVAL;
-
-	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-		return -EINVAL;
-
-	return dev->driver->bus->irq_by_busid(dev, p);
-}
-
 /*
  * Clear vblank timestamp buffer for a crtc.
  */
@@ -167,33 +139,40 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 
 static void vblank_disable_fn(unsigned long arg)
 {
-	struct drm_device *dev = (struct drm_device *)arg;
+	struct drm_vblank_crtc *vblank = (void *)arg;
+	struct drm_device *dev = vblank->dev;
 	unsigned long irqflags;
-	int i;
+	int crtc = vblank->crtc;
 
 	if (!dev->vblank_disable_allowed)
 		return;
 
-	for (i = 0; i < dev->num_crtcs; i++) {
-		spin_lock_irqsave(&dev->vbl_lock, irqflags);
-		if (atomic_read(&dev->vblank[i].refcount) == 0 &&
-		    dev->vblank[i].enabled) {
-			DRM_DEBUG("disabling vblank on crtc %d\n", i);
-			vblank_disable_and_save(dev, i);
-		}
-		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
+	spin_lock_irqsave(&dev->vbl_lock, irqflags);
+	if (atomic_read(&vblank->refcount) == 0 && vblank->enabled) {
+		DRM_DEBUG("disabling vblank on crtc %d\n", crtc);
+		vblank_disable_and_save(dev, crtc);
 	}
+	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 }
 
+/**
+ * drm_vblank_cleanup - cleanup vblank support
+ * @dev: DRM device
+ *
+ * This function cleans up any resources allocated in drm_vblank_init.
+ */
 void drm_vblank_cleanup(struct drm_device *dev)
 {
+	int crtc;
+
 	/* Bail if the driver didn't call drm_vblank_init() */
 	if (dev->num_crtcs == 0)
 		return;
 
-	del_timer_sync(&dev->vblank_disable_timer);
-
-	vblank_disable_fn((unsigned long)dev);
+	for (crtc = 0; crtc < dev->num_crtcs; crtc++) {
+		del_timer_sync(&dev->vblank[crtc].disable_timer);
+		vblank_disable_fn((unsigned long)&dev->vblank[crtc]);
+	}
 
 	kfree(dev->vblank);
 
@@ -201,12 +180,20 @@ void drm_vblank_cleanup(struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_vblank_cleanup);
 
+/**
+ * drm_vblank_init - initialize vblank support
+ * @dev: drm_device
+ * @num_crtcs: number of crtcs supported by @dev
+ *
+ * This function initializes vblank support for @num_crtcs display pipelines.
+ *
+ * Returns:
+ * Zero on success or a negative error code on failure.
+ */
 int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 {
 	int i, ret = -ENOMEM;
 
-	setup_timer(&dev->vblank_disable_timer, vblank_disable_fn,
-		    (unsigned long)dev);
 	spin_lock_init(&dev->vbl_lock);
 	spin_lock_init(&dev->vblank_time_lock);
 
@@ -216,8 +203,13 @@ int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 	if (!dev->vblank)
 		goto err;
 
-	for (i = 0; i < num_crtcs; i++)
+	for (i = 0; i < num_crtcs; i++) {
+		dev->vblank[i].dev = dev;
+		dev->vblank[i].crtc = i;
 		init_waitqueue_head(&dev->vblank[i].queue);
+		setup_timer(&dev->vblank[i].disable_timer, vblank_disable_fn,
+			    (unsigned long)&dev->vblank[i]);
+	}
 
 	DRM_INFO("Supports vblank timestamp caching Rev 2 (21.10.2013).\n");
 
@@ -261,42 +253,42 @@ static void drm_irq_vgaarb_nokms(void *cookie, bool state)
 }
 
 /**
- * Install IRQ handler.
- *
- * \param dev DRM device.
+ * drm_irq_install - install IRQ handler
+ * @dev: DRM device
+ * @irq: IRQ number to install the handler for
  *
  * Initializes the IRQ related data. Installs the handler, calling the driver
- * \c irq_preinstall() and \c irq_postinstall() functions
- * before and after the installation.
+ * irq_preinstall() and irq_postinstall() functions before and after the
+ * installation.
+ *
+ * This is the simplified helper interface provided for drivers with no special
+ * needs. Drivers which need to install interrupt handlers for multiple
+ * interrupts must instead set drm_device->irq_enabled to signal the DRM core
+ * that vblank interrupts are available.
+ *
+ * Returns:
+ * Zero on success or a negative error code on failure.
  */
-int drm_irq_install(struct drm_device *dev)
+int drm_irq_install(struct drm_device *dev, int irq)
 {
 	int ret;
 	unsigned long sh_flags = 0;
-	char *irqname;
 
 	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
 		return -EINVAL;
 
-	if (drm_dev_to_irq(dev) == 0)
+	if (irq == 0)
 		return -EINVAL;
 
-	mutex_lock(&dev->struct_mutex);
-
 	/* Driver must have been initialized */
-	if (!dev->dev_private) {
-		mutex_unlock(&dev->struct_mutex);
+	if (!dev->dev_private)
 		return -EINVAL;
-	}
 
-	if (dev->irq_enabled) {
-		mutex_unlock(&dev->struct_mutex);
+	if (dev->irq_enabled)
 		return -EBUSY;
-	}
 	dev->irq_enabled = true;
-	mutex_unlock(&dev->struct_mutex);
 
-	DRM_DEBUG("irq=%d\n", drm_dev_to_irq(dev));
+	DRM_DEBUG("irq=%d\n", irq);
 
 	/* Before installing handler */
 	if (dev->driver->irq_preinstall)
@@ -306,18 +298,11 @@ int drm_irq_install(struct drm_device *dev)
 	if (drm_core_check_feature(dev, DRIVER_IRQ_SHARED))
 		sh_flags = IRQF_SHARED;
 
-	if (dev->devname)
-		irqname = dev->devname;
-	else
-		irqname = dev->driver->name;
-
-	ret = request_irq(drm_dev_to_irq(dev), dev->driver->irq_handler,
-			  sh_flags, irqname, dev);
+	ret = request_irq(irq, dev->driver->irq_handler,
+			  sh_flags, dev->driver->name, dev);
 
 	if (ret < 0) {
-		mutex_lock(&dev->struct_mutex);
 		dev->irq_enabled = false;
-		mutex_unlock(&dev->struct_mutex);
 		return ret;
 	}
 
@@ -329,12 +314,12 @@ int drm_irq_install(struct drm_device *dev)
 		ret = dev->driver->irq_postinstall(dev);
 
 	if (ret < 0) {
-		mutex_lock(&dev->struct_mutex);
 		dev->irq_enabled = false;
-		mutex_unlock(&dev->struct_mutex);
 		if (!drm_core_check_feature(dev, DRIVER_MODESET))
 			vga_client_register(dev->pdev, NULL, NULL, NULL);
-		free_irq(drm_dev_to_irq(dev), dev);
+		free_irq(irq, dev);
+	} else {
+		dev->irq = irq;
 	}
 
 	return ret;
@@ -342,11 +327,20 @@ int drm_irq_install(struct drm_device *dev)
 EXPORT_SYMBOL(drm_irq_install);
 
 /**
- * Uninstall the IRQ handler.
+ * drm_irq_uninstall - uninstall the IRQ handler
+ * @dev: DRM device
+ *
+ * Calls the driver's irq_uninstall() function and unregisters the IRQ handler.
+ * This should only be called by drivers which used drm_irq_install() to set up
+ * their interrupt handler. Other drivers must only reset
+ * drm_device->irq_enabled to false.
  *
- * \param dev DRM device.
+ * Note that for kernel modesetting drivers it is a bug if this function fails.
+ * The sanity checks are only to catch buggy user modesetting drivers which call
+ * the same function through an ioctl.
  *
- * Calls the driver's \c irq_uninstall() function, and stops the irq.
+ * Returns:
+ * Zero on success or a negative error code on failure.
  */
 int drm_irq_uninstall(struct drm_device *dev)
 {
@@ -357,10 +351,8 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
 		return -EINVAL;
 
-	mutex_lock(&dev->struct_mutex);
 	irq_enabled = dev->irq_enabled;
 	dev->irq_enabled = false;
-	mutex_unlock(&dev->struct_mutex);
 
 	/*
 	 * Wake up any waiters so they don't hang.
@@ -379,7 +371,7 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (!irq_enabled)
 		return -EINVAL;
 
-	DRM_DEBUG("irq=%d\n", drm_dev_to_irq(dev));
+	DRM_DEBUG("irq=%d\n", dev->irq);
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		vga_client_register(dev->pdev, NULL, NULL, NULL);
@@ -387,13 +379,13 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (dev->driver->irq_uninstall)
 		dev->driver->irq_uninstall(dev);
 
-	free_irq(drm_dev_to_irq(dev), dev);
+	free_irq(dev->irq, dev);
 
 	return 0;
 }
 EXPORT_SYMBOL(drm_irq_uninstall);
 
-/**
+/*
  * IRQ control ioctl.
  *
  * \param inode device inode.
@@ -408,43 +400,52 @@ int drm_control(struct drm_device *dev, void *data,
 		struct drm_file *file_priv)
 {
 	struct drm_control *ctl = data;
+	int ret = 0, irq;
 
 	/* if we haven't irq we fallback for compatibility reasons -
 	 * this used to be a separate function in drm_dma.h
 	 */
 
+	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
+		return 0;
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return 0;
+	/* UMS was only ever support on pci devices. */
+	if (WARN_ON(!dev->pdev))
+		return -EINVAL;
 
 	switch (ctl->func) {
 	case DRM_INST_HANDLER:
-		if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-			return 0;
-		if (drm_core_check_feature(dev, DRIVER_MODESET))
-			return 0;
+		irq = dev->pdev->irq;
+
 		if (dev->if_version < DRM_IF_VERSION(1, 2) &&
-		    ctl->irq != drm_dev_to_irq(dev))
+		    ctl->irq != irq)
 			return -EINVAL;
-		return drm_irq_install(dev);
+		mutex_lock(&dev->struct_mutex);
+		ret = drm_irq_install(dev, irq);
+		mutex_unlock(&dev->struct_mutex);
+
+		return ret;
 	case DRM_UNINST_HANDLER:
-		if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-			return 0;
-		if (drm_core_check_feature(dev, DRIVER_MODESET))
-			return 0;
-		return drm_irq_uninstall(dev);
+		mutex_lock(&dev->struct_mutex);
+		ret = drm_irq_uninstall(dev);
+		mutex_unlock(&dev->struct_mutex);
+
+		return ret;
 	default:
 		return -EINVAL;
 	}
 }
 
 /**
- * drm_calc_timestamping_constants - Calculate vblank timestamp constants
- *
- * @crtc drm_crtc whose timestamp constants should be updated.
- * @mode display mode containing the scanout timings
+ * drm_calc_timestamping_constants - calculate vblank timestamp constants
+ * @crtc: drm_crtc whose timestamp constants should be updated.
+ * @mode: display mode containing the scanout timings
  *
  * Calculate and store various constants which are later
  * needed by vblank and swap-completion timestamping, e.g,
  * by drm_calc_vbltimestamp_from_scanoutpos(). They are
- * derived from crtc's true scanout timing, so they take
+ * derived from CRTC's true scanout timing, so they take
  * things like panel scaling or other adjustments into account.
  */
 void drm_calc_timestamping_constants(struct drm_crtc *crtc,
@@ -489,11 +490,22 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 EXPORT_SYMBOL(drm_calc_timestamping_constants);
 
 /**
- * drm_calc_vbltimestamp_from_scanoutpos - helper routine for kms
- * drivers. Implements calculation of exact vblank timestamps from
- * given drm_display_mode timings and current video scanout position
- * of a crtc. This can be called from within get_vblank_timestamp()
- * implementation of a kms driver to implement the actual timestamping.
+ * drm_calc_vbltimestamp_from_scanoutpos - precise vblank timestamp helper
+ * @dev: DRM device
+ * @crtc: Which CRTC's vblank timestamp to retrieve
+ * @max_error: Desired maximum allowable error in timestamps (nanosecs)
+ *             On return contains true maximum error of timestamp
+ * @vblank_time: Pointer to struct timeval which should receive the timestamp
+ * @flags: Flags to pass to driver:
+ *         0 = Default,
+ *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl IRQ handler
+ * @refcrtc: CRTC which defines scanout timing
+ * @mode: mode which defines the scanout timings
+ *
+ * Implements calculation of exact vblank timestamps from given drm_display_mode
+ * timings and current video scanout position of a CRTC. This can be called from
+ * within get_vblank_timestamp() implementation of a kms driver to implement the
+ * actual timestamping.
  *
  * Should return timestamps conforming to the OML_sync_control OpenML
  * extension specification. The timestamp corresponds to the end of
@@ -508,21 +520,11 @@ EXPORT_SYMBOL(drm_calc_timestamping_constants);
  * returns as no operation if a doublescan or interlaced video mode is
  * active. Higher level code is expected to handle this.
  *
- * @dev: DRM device.
- * @crtc: Which crtc's vblank timestamp to retrieve.
- * @max_error: Desired maximum allowable error in timestamps (nanosecs).
- *             On return contains true maximum error of timestamp.
- * @vblank_time: Pointer to struct timeval which should receive the timestamp.
- * @flags: Flags to pass to driver:
- *         0 = Default.
- *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl irq handler.
- * @refcrtc: drm_crtc* of crtc which defines scanout timing.
- * @mode: mode which defines the scanout timings
- *
- * Returns negative value on error, failure or if not supported in current
+ * Returns:
+ * Negative value on error, failure or if not supported in current
  * video mode:
  *
- * -EINVAL   - Invalid crtc.
+ * -EINVAL   - Invalid CRTC.
  * -EAGAIN   - Temporary unavailable, e.g., called before initial modeset.
  * -ENOTSUPP - Function not supported in current display mode.
  * -EIO      - Failed, e.g., due to failed scanout position query.
@@ -671,23 +673,23 @@ static struct timeval get_drm_timestamp(void)
 
 /**
  * drm_get_last_vbltimestamp - retrieve raw timestamp for the most recent
- * vblank interval.
- *
+ * 			       vblank interval
  * @dev: DRM device
- * @crtc: which crtc's vblank timestamp to retrieve
+ * @crtc: which CRTC's vblank timestamp to retrieve
  * @tvblank: Pointer to target struct timeval which should receive the timestamp
  * @flags: Flags to pass to driver:
- *         0 = Default.
- *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl irq handler.
+ *         0 = Default,
+ *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl IRQ handler
  *
  * Fetches the system timestamp corresponding to the time of the most recent
- * vblank interval on specified crtc. May call into kms-driver to
+ * vblank interval on specified CRTC. May call into kms-driver to
  * compute the timestamp with a high-precision GPU specific method.
  *
  * Returns zero if timestamp originates from uncorrected do_gettimeofday()
  * call, i.e., it isn't very precisely locked to the true vblank.
  *
- * Returns non-zero if timestamp is considered to be very precise.
+ * Returns:
+ * Non-zero if timestamp is considered to be very precise, zero otherwise.
  */
 u32 drm_get_last_vbltimestamp(struct drm_device *dev, int crtc,
 			      struct timeval *tvblank, unsigned flags)
@@ -722,6 +724,9 @@ EXPORT_SYMBOL(drm_get_last_vbltimestamp);
  * Fetches the "cooked" vblank count value that represents the number of
  * vblank events since the system was booted, including lost events due to
  * modesetting activity.
+ *
+ * Returns:
+ * The software vblank counter.
  */
 u32 drm_vblank_count(struct drm_device *dev, int crtc)
 {
@@ -740,8 +745,7 @@ EXPORT_SYMBOL(drm_vblank_count);
  * Fetches the "cooked" vblank count value that represents the number of
  * vblank events since the system was booted, including lost events due to
  * modesetting activity. Returns corresponding system timestamp of the time
- * of the vblank interval that corresponds to the current value vblank counter
- * value.
+ * of the vblank interval that corresponds to the current vblank counter value.
  */
 u32 drm_vblank_count_and_time(struct drm_device *dev, int crtc,
 			      struct timeval *vblanktime)
@@ -870,6 +874,42 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 }
 
 /**
+ * drm_vblank_enable - enable the vblank interrupt on a CRTC
+ * @dev: DRM device
+ * @crtc: CRTC in question
+ */
+static int drm_vblank_enable(struct drm_device *dev, int crtc)
+{
+	int ret = 0;
+
+	assert_spin_locked(&dev->vbl_lock);
+
+	spin_lock(&dev->vblank_time_lock);
+
+	if (!dev->vblank[crtc].enabled) {
+		/*
+		 * Enable vblank irqs under vblank_time_lock protection.
+		 * All vblank count & timestamp updates are held off
+		 * until we are done reinitializing master counter and
+		 * timestamps. Filtercode in drm_handle_vblank() will
+		 * prevent double-accounting of same vblank interval.
+		 */
+		ret = dev->driver->enable_vblank(dev, crtc);
+		DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n", crtc, ret);
+		if (ret)
+			atomic_dec(&dev->vblank[crtc].refcount);
+		else {
+			dev->vblank[crtc].enabled = true;
+			drm_update_vblank_count(dev, crtc);
+		}
+	}
+
+	spin_unlock(&dev->vblank_time_lock);
+
+	return ret;
+}
+
+/**
  * drm_vblank_get - get a reference count on vblank events
  * @dev: DRM device
  * @crtc: which CRTC to own
@@ -877,36 +917,20 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
  * Acquire a reference count on vblank events to avoid having them disabled
  * while in use.
  *
- * RETURNS
+ * This is the legacy version of drm_crtc_vblank_get().
+ *
+ * Returns:
  * Zero on success, nonzero on failure.
  */
 int drm_vblank_get(struct drm_device *dev, int crtc)
 {
-	unsigned long irqflags, irqflags2;
+	unsigned long irqflags;
 	int ret = 0;
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/* Going from 0->1 means we have to enable interrupts again */
 	if (atomic_add_return(1, &dev->vblank[crtc].refcount) == 1) {
-		spin_lock_irqsave(&dev->vblank_time_lock, irqflags2);
-		if (!dev->vblank[crtc].enabled) {
-			/* Enable vblank irqs under vblank_time_lock protection.
-			 * All vblank count & timestamp updates are held off
-			 * until we are done reinitializing master counter and
-			 * timestamps. Filtercode in drm_handle_vblank() will
-			 * prevent double-accounting of same vblank interval.
-			 */
-			ret = dev->driver->enable_vblank(dev, crtc);
-			DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n",
-				  crtc, ret);
-			if (ret)
-				atomic_dec(&dev->vblank[crtc].refcount);
-			else {
-				dev->vblank[crtc].enabled = true;
-				drm_update_vblank_count(dev, crtc);
-			}
-		}
-		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2);
+		ret = drm_vblank_enable(dev, crtc);
 	} else {
 		if (!dev->vblank[crtc].enabled) {
 			atomic_dec(&dev->vblank[crtc].refcount);
@@ -920,12 +944,32 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 EXPORT_SYMBOL(drm_vblank_get);
 
 /**
+ * drm_crtc_vblank_get - get a reference count on vblank events
+ * @crtc: which CRTC to own
+ *
+ * Acquire a reference count on vblank events to avoid having them disabled
+ * while in use.
+ *
+ * This is the native kms version of drm_vblank_off().
+ *
+ * Returns:
+ * Zero on success, nonzero on failure.
+ */
+int drm_crtc_vblank_get(struct drm_crtc *crtc)
+{
+	return drm_vblank_get(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_get);
+
+/**
  * drm_vblank_put - give up ownership of vblank events
  * @dev: DRM device
  * @crtc: which counter to give up
  *
  * Release ownership of a given vblank counter, turning off interrupts
  * if possible. Disable interrupts after drm_vblank_offdelay milliseconds.
+ *
+ * This is the legacy version of drm_crtc_vblank_put().
  */
 void drm_vblank_put(struct drm_device *dev, int crtc)
 {
@@ -934,17 +978,39 @@ void drm_vblank_put(struct drm_device *dev, int crtc)
 	/* Last user schedules interrupt disable */
 	if (atomic_dec_and_test(&dev->vblank[crtc].refcount) &&
 	    (drm_vblank_offdelay > 0))
-		mod_timer(&dev->vblank_disable_timer,
+		mod_timer(&dev->vblank[crtc].disable_timer,
 			  jiffies + ((drm_vblank_offdelay * HZ)/1000));
 }
 EXPORT_SYMBOL(drm_vblank_put);
 
 /**
+ * drm_crtc_vblank_put - give up ownership of vblank events
+ * @crtc: which counter to give up
+ *
+ * Release ownership of a given vblank counter, turning off interrupts
+ * if possible. Disable interrupts after drm_vblank_offdelay milliseconds.
+ *
+ * This is the native kms version of drm_vblank_put().
+ */
+void drm_crtc_vblank_put(struct drm_crtc *crtc)
+{
+	drm_vblank_put(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_put);
+
+/**
  * drm_vblank_off - disable vblank events on a CRTC
  * @dev: DRM device
  * @crtc: CRTC in question
  *
- * Caller must hold event lock.
+ * Drivers can use this function to shut down the vblank interrupt handling when
+ * disabling a crtc. This function ensures that the latest vblank frame count is
+ * stored so that drm_vblank_on() can restore it again.
+ *
+ * Drivers must use this function when the hardware vblank counter can get
+ * reset, e.g. when suspending.
+ *
+ * This is the legacy version of drm_crtc_vblank_off().
  */
 void drm_vblank_off(struct drm_device *dev, int crtc)
 {
@@ -978,12 +1044,87 @@ void drm_vblank_off(struct drm_device *dev, int crtc)
 EXPORT_SYMBOL(drm_vblank_off);
 
 /**
+ * drm_crtc_vblank_off - disable vblank events on a CRTC
+ * @crtc: CRTC in question
+ *
+ * Drivers can use this function to shut down the vblank interrupt handling when
+ * disabling a crtc. This function ensures that the latest vblank frame count is
+ * stored so that drm_vblank_on can restore it again.
+ *
+ * Drivers must use this function when the hardware vblank counter can get
+ * reset, e.g. when suspending.
+ *
+ * This is the native kms version of drm_vblank_off().
+ */
+void drm_crtc_vblank_off(struct drm_crtc *crtc)
+{
+	drm_vblank_off(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_off);
+
+/**
+ * drm_vblank_on - enable vblank events on a CRTC
+ * @dev: DRM device
+ * @crtc: CRTC in question
+ *
+ * This functions restores the vblank interrupt state captured with
+ * drm_vblank_off() again. Note that calls to drm_vblank_on() and
+ * drm_vblank_off() can be unbalanced and so can also be unconditionaly called
+ * in driver load code to reflect the current hardware state of the crtc.
+ *
+ * This is the legacy version of drm_crtc_vblank_on().
+ */
+void drm_vblank_on(struct drm_device *dev, int crtc)
+{
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev->vbl_lock, irqflags);
+	/* re-enable interrupts if there's are users left */
+	if (atomic_read(&dev->vblank[crtc].refcount) != 0)
+		WARN_ON(drm_vblank_enable(dev, crtc));
+	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
+}
+EXPORT_SYMBOL(drm_vblank_on);
+
+/**
+ * drm_crtc_vblank_on - enable vblank events on a CRTC
+ * @crtc: CRTC in question
+ *
+ * This functions restores the vblank interrupt state captured with
+ * drm_vblank_off() again. Note that calls to drm_vblank_on() and
+ * drm_vblank_off() can be unbalanced and so can also be unconditionaly called
+ * in driver load code to reflect the current hardware state of the crtc.
+ *
+ * This is the native kms version of drm_vblank_on().
+ */
+void drm_crtc_vblank_on(struct drm_crtc *crtc)
+{
+	drm_vblank_on(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_on);
+
+/**
  * drm_vblank_pre_modeset - account for vblanks across mode sets
  * @dev: DRM device
  * @crtc: CRTC in question
  *
  * Account for vblank events across mode setting events, which will likely
  * reset the hardware frame counter.
+ *
+ * This is done by grabbing a temporary vblank reference to ensure that the
+ * vblank interrupt keeps running across the modeset sequence. With this the
+ * software-side vblank frame counting will ensure that there are no jumps or
+ * discontinuities.
+ *
+ * Unfortunately this approach is racy and also doesn't work when the vblank
+ * interrupt stops running, e.g. across system suspend resume. It is therefore
+ * highly recommended that drivers use the newer drm_vblank_off() and
+ * drm_vblank_on() instead. drm_vblank_pre_modeset() only works correctly when
+ * using "cooked" software vblank frame counters and not relying on any hardware
+ * counters.
+ *
+ * Drivers must call drm_vblank_post_modeset() when re-enabling the same crtc
+ * again.
  */
 void drm_vblank_pre_modeset(struct drm_device *dev, int crtc)
 {
@@ -1005,6 +1146,14 @@ void drm_vblank_pre_modeset(struct drm_device *dev, int crtc)
 }
 EXPORT_SYMBOL(drm_vblank_pre_modeset);
 
+/**
+ * drm_vblank_post_modeset - undo drm_vblank_pre_modeset changes
+ * @dev: DRM device
+ * @crtc: CRTC in question
+ *
+ * This function again drops the temporary vblank reference acquired in
+ * drm_vblank_pre_modeset.
+ */
 void drm_vblank_post_modeset(struct drm_device *dev, int crtc)
 {
 	unsigned long irqflags;
@@ -1026,7 +1175,7 @@ void drm_vblank_post_modeset(struct drm_device *dev, int crtc)
 }
 EXPORT_SYMBOL(drm_vblank_post_modeset);
 
-/**
+/*
  * drm_modeset_ctl - handle vblank event counter changes across mode switch
  * @DRM_IOCTL_ARGS: standard ioctl arguments
  *
@@ -1139,7 +1288,7 @@ err_put:
 	return ret;
 }
 
-/**
+/*
  * Wait for VBLANK.
  *
  * \param inode device inode.
@@ -1150,7 +1299,7 @@ err_put:
  *
  * This function enables the vblank interrupt on the pipe requested, then
  * sleeps waiting for the requested sequence number to occur, and drops
- * the vblank interrupt refcount afterwards. (vblank irq disable follows that
+ * the vblank interrupt refcount afterwards. (vblank IRQ disable follows that
  * after a timeout with no further vblank waits scheduled).
  */
 int drm_wait_vblank(struct drm_device *dev, void *data,
@@ -1160,9 +1309,8 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 	int ret;
 	unsigned int flags, seq, crtc, high_crtc;
 
-	if (drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-		if ((!drm_dev_to_irq(dev)) || (!dev->irq_enabled))
-			return -EINVAL;
+	if (!dev->irq_enabled)
+		return -EINVAL;
 
 	if (vblwait->request.type & _DRM_VBLANK_SIGNAL)
 		return -EINVAL;
@@ -1222,6 +1370,7 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 	DRM_WAIT_ON(ret, dev->vblank[crtc].queue, 3 * HZ,
 		    (((drm_vblank_count(dev, crtc) -
 		       vblwait->request.sequence) <= (1 << 23)) ||
+		     !dev->vblank[crtc].enabled ||
 		     !dev->irq_enabled));
 
 	if (ret != -EINTR) {
diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index 09821f46d768..e633df2f68d8 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -282,6 +282,14 @@ static int mipi_dsi_drv_remove(struct device *dev)
 	return drv->remove(dsi);
 }
 
+static void mipi_dsi_drv_shutdown(struct device *dev)
+{
+	struct mipi_dsi_driver *drv = to_mipi_dsi_driver(dev->driver);
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
+
+	drv->shutdown(dsi);
+}
+
 /**
  * mipi_dsi_driver_register - register a driver for DSI devices
  * @drv: DSI driver structure
@@ -293,6 +301,8 @@ int mipi_dsi_driver_register(struct mipi_dsi_driver *drv)
 		drv->driver.probe = mipi_dsi_drv_probe;
 	if (drv->remove)
 		drv->driver.remove = mipi_dsi_drv_remove;
+	if (drv->shutdown)
+		drv->driver.shutdown = mipi_dsi_drv_shutdown;
 
 	return driver_register(&drv->driver);
 }
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index 8b410576fce4..bedf1894e17e 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -1013,6 +1013,7 @@ EXPORT_SYMBOL(drm_mode_sort);
 /**
  * drm_mode_connector_list_update - update the mode list for the connector
  * @connector: the connector to update
+ * @merge_type_bits: whether to merge or overright type bits.
  *
  * This moves the modes from the @connector probed_modes list
  * to the actual mode list. It compares the probed mode against the current
@@ -1021,7 +1022,8 @@ EXPORT_SYMBOL(drm_mode_sort);
  * This is just a helper functions doesn't validate any modes itself and also
  * doesn't prune any invalid modes. Callers need to do that themselves.
  */
-void drm_mode_connector_list_update(struct drm_connector *connector)
+void drm_mode_connector_list_update(struct drm_connector *connector,
+				    bool merge_type_bits)
 {
 	struct drm_display_mode *mode;
 	struct drm_display_mode *pmode, *pt;
@@ -1039,7 +1041,10 @@ void drm_mode_connector_list_update(struct drm_connector *connector)
 				/* if equal delete the probed mode */
 				mode->status = pmode->status;
 				/* Merge type bits together */
-				mode->type |= pmode->type;
+				if (merge_type_bits)
+					mode->type |= pmode->type;
+				else
+					mode->type = pmode->type;
 				list_del(&pmode->head);
 				drm_mode_destroy(connector->dev, pmode);
 				break;
diff --git a/drivers/gpu/drm/drm_modeset_lock.c b/drivers/gpu/drm/drm_modeset_lock.c
new file mode 100644
index 000000000000..0dc57d5ecd10
--- /dev/null
+++ b/drivers/gpu/drm/drm_modeset_lock.c
@@ -0,0 +1,248 @@
+/*
+ * Copyright (C) 2014 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_modeset_lock.h>
+
+/**
+ * DOC: kms locking
+ *
+ * As KMS moves toward more fine grained locking, and atomic ioctl where
+ * userspace can indirectly control locking order, it becomes necessary
+ * to use ww_mutex and acquire-contexts to avoid deadlocks.  But because
+ * the locking is more distributed around the driver code, we want a bit
+ * of extra utility/tracking out of our acquire-ctx.  This is provided
+ * by drm_modeset_lock / drm_modeset_acquire_ctx.
+ *
+ * For basic principles of ww_mutex, see: Documentation/ww-mutex-design.txt
+ *
+ * The basic usage pattern is to:
+ *
+ *     drm_modeset_acquire_init(&ctx)
+ *   retry:
+ *     foreach (lock in random_ordered_set_of_locks) {
+ *       ret = drm_modeset_lock(lock, &ctx)
+ *       if (ret == -EDEADLK) {
+ *          drm_modeset_backoff(&ctx);
+ *          goto retry;
+ *       }
+ *     }
+ *
+ *     ... do stuff ...
+ *
+ *     drm_modeset_drop_locks(&ctx);
+ *     drm_modeset_acquire_fini(&ctx);
+ */
+
+
+/**
+ * drm_modeset_acquire_init - initialize acquire context
+ * @ctx: the acquire context
+ * @flags: for future
+ */
+void drm_modeset_acquire_init(struct drm_modeset_acquire_ctx *ctx,
+		uint32_t flags)
+{
+	memset(ctx, 0, sizeof(*ctx));
+	ww_acquire_init(&ctx->ww_ctx, &crtc_ww_class);
+	INIT_LIST_HEAD(&ctx->locked);
+}
+EXPORT_SYMBOL(drm_modeset_acquire_init);
+
+/**
+ * drm_modeset_acquire_fini - cleanup acquire context
+ * @ctx: the acquire context
+ */
+void drm_modeset_acquire_fini(struct drm_modeset_acquire_ctx *ctx)
+{
+	ww_acquire_fini(&ctx->ww_ctx);
+}
+EXPORT_SYMBOL(drm_modeset_acquire_fini);
+
+/**
+ * drm_modeset_drop_locks - drop all locks
+ * @ctx: the acquire context
+ *
+ * Drop all locks currently held against this acquire context.
+ */
+void drm_modeset_drop_locks(struct drm_modeset_acquire_ctx *ctx)
+{
+	WARN_ON(ctx->contended);
+	while (!list_empty(&ctx->locked)) {
+		struct drm_modeset_lock *lock;
+
+		lock = list_first_entry(&ctx->locked,
+				struct drm_modeset_lock, head);
+
+		drm_modeset_unlock(lock);
+	}
+}
+EXPORT_SYMBOL(drm_modeset_drop_locks);
+
+static inline int modeset_lock(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx,
+		bool interruptible, bool slow)
+{
+	int ret;
+
+	WARN_ON(ctx->contended);
+
+	if (interruptible && slow) {
+		ret = ww_mutex_lock_slow_interruptible(&lock->mutex, &ctx->ww_ctx);
+	} else if (interruptible) {
+		ret = ww_mutex_lock_interruptible(&lock->mutex, &ctx->ww_ctx);
+	} else if (slow) {
+		ww_mutex_lock_slow(&lock->mutex, &ctx->ww_ctx);
+		ret = 0;
+	} else {
+		ret = ww_mutex_lock(&lock->mutex, &ctx->ww_ctx);
+	}
+	if (!ret) {
+		WARN_ON(!list_empty(&lock->head));
+		list_add(&lock->head, &ctx->locked);
+	} else if (ret == -EALREADY) {
+		/* we already hold the lock.. this is fine.  For atomic
+		 * we will need to be able to drm_modeset_lock() things
+		 * without having to keep track of what is already locked
+		 * or not.
+		 */
+		ret = 0;
+	} else if (ret == -EDEADLK) {
+		ctx->contended = lock;
+	}
+
+	return ret;
+}
+
+static int modeset_backoff(struct drm_modeset_acquire_ctx *ctx,
+		bool interruptible)
+{
+	struct drm_modeset_lock *contended = ctx->contended;
+
+	ctx->contended = NULL;
+
+	if (WARN_ON(!contended))
+		return 0;
+
+	drm_modeset_drop_locks(ctx);
+
+	return modeset_lock(contended, ctx, interruptible, true);
+}
+
+/**
+ * drm_modeset_backoff - deadlock avoidance backoff
+ * @ctx: the acquire context
+ *
+ * If deadlock is detected (ie. drm_modeset_lock() returns -EDEADLK),
+ * you must call this function to drop all currently held locks and
+ * block until the contended lock becomes available.
+ */
+void drm_modeset_backoff(struct drm_modeset_acquire_ctx *ctx)
+{
+	modeset_backoff(ctx, false);
+}
+EXPORT_SYMBOL(drm_modeset_backoff);
+
+/**
+ * drm_modeset_backoff_interruptible - deadlock avoidance backoff
+ * @ctx: the acquire context
+ *
+ * Interruptible version of drm_modeset_backoff()
+ */
+int drm_modeset_backoff_interruptible(struct drm_modeset_acquire_ctx *ctx)
+{
+	return modeset_backoff(ctx, true);
+}
+EXPORT_SYMBOL(drm_modeset_backoff_interruptible);
+
+/**
+ * drm_modeset_lock - take modeset lock
+ * @lock: lock to take
+ * @ctx: acquire ctx
+ *
+ * If ctx is not NULL, then its ww acquire context is used and the
+ * lock will be tracked by the context and can be released by calling
+ * drm_modeset_drop_locks().  If -EDEADLK is returned, this means a
+ * deadlock scenario has been detected and it is an error to attempt
+ * to take any more locks without first calling drm_modeset_backoff().
+ */
+int drm_modeset_lock(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx)
+{
+	if (ctx)
+		return modeset_lock(lock, ctx, false, false);
+
+	ww_mutex_lock(&lock->mutex, NULL);
+	return 0;
+}
+EXPORT_SYMBOL(drm_modeset_lock);
+
+/**
+ * drm_modeset_lock_interruptible - take modeset lock
+ * @lock: lock to take
+ * @ctx: acquire ctx
+ *
+ * Interruptible version of drm_modeset_lock()
+ */
+int drm_modeset_lock_interruptible(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx)
+{
+	if (ctx)
+		return modeset_lock(lock, ctx, true, false);
+
+	return ww_mutex_lock_interruptible(&lock->mutex, NULL);
+}
+EXPORT_SYMBOL(drm_modeset_lock_interruptible);
+
+/**
+ * drm_modeset_unlock - drop modeset lock
+ * @lock: lock to release
+ */
+void drm_modeset_unlock(struct drm_modeset_lock *lock)
+{
+	list_del_init(&lock->head);
+	ww_mutex_unlock(&lock->mutex);
+}
+EXPORT_SYMBOL(drm_modeset_unlock);
+
+/* Temporary.. until we have sufficiently fine grained locking, there
+ * are a couple scenarios where it is convenient to grab all crtc locks.
+ * It is planned to remove this:
+ */
+int drm_modeset_lock_all_crtcs(struct drm_device *dev,
+		struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_crtc *crtc;
+	int ret = 0;
+
+	list_for_each_entry(crtc, &config->crtc_list, head) {
+		ret = drm_modeset_lock(&crtc->mutex, ctx);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_modeset_lock_all_crtcs);
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 9c696a5ad74d..020cfd934854 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -1,17 +1,3 @@
-/* drm_pci.h -- PCI DMA memory management wrappers for DRM -*- linux-c -*- */
-/**
- * \file drm_pci.c
- * \brief Functions and ioctls to manage PCI memory
- *
- * \warning These interfaces aren't stable yet.
- *
- * \todo Implement the remaining ioctl's for the PCI pools.
- * \todo The wrappers here are so thin that they would be better off inlined..
- *
- * \author José Fonseca <jrfonseca@tungstengraphics.com>
- * \author Leif Delgass <ldelgass@retinalburn.net>
- */
-
 /*
  * Copyright 2003 José Fonseca.
  * Copyright 2003 Leif Delgass.
@@ -42,12 +28,14 @@
 #include <linux/export.h>
 #include <drm/drmP.h>
 
-/**********************************************************************/
-/** \name PCI memory */
-/*@{*/
-
 /**
- * \brief Allocate a PCI consistent memory block, for DMA.
+ * drm_pci_alloc - Allocate a PCI consistent memory block, for DMA.
+ * @dev: DRM device
+ * @size: size of block to allocate
+ * @align: alignment of block
+ *
+ * Return: A handle to the allocated memory block on success or NULL on
+ * failure.
  */
 drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t align)
 {
@@ -88,8 +76,8 @@ drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t ali
 
 EXPORT_SYMBOL(drm_pci_alloc);
 
-/**
- * \brief Free a PCI consistent memory block without freeing its descriptor.
+/*
+ * Free a PCI consistent memory block without freeing its descriptor.
  *
  * This function is for internal use in the Linux-specific DRM core code.
  */
@@ -111,7 +99,9 @@ void __drm_pci_free(struct drm_device * dev, drm_dma_handle_t * dmah)
 }
 
 /**
- * \brief Free a PCI consistent memory block
+ * drm_pci_free - Free a PCI consistent memory block
+ * @dev: DRM device
+ * @dmah: handle to memory block
  */
 void drm_pci_free(struct drm_device * dev, drm_dma_handle_t * dmah)
 {
@@ -137,21 +127,9 @@ static int drm_get_pci_domain(struct drm_device *dev)
 	return pci_domain_nr(dev->pdev->bus);
 }
 
-static int drm_pci_get_irq(struct drm_device *dev)
-{
-	return dev->pdev->irq;
-}
-
-static const char *drm_pci_get_name(struct drm_device *dev)
-{
-	struct pci_driver *pdriver = dev->driver->kdriver.pci;
-	return pdriver->name;
-}
-
 static int drm_pci_set_busid(struct drm_device *dev, struct drm_master *master)
 {
 	int len, ret;
-	struct pci_driver *pdriver = dev->driver->kdriver.pci;
 	master->unique_len = 40;
 	master->unique_size = master->unique_len;
 	master->unique = kmalloc(master->unique_size, GFP_KERNEL);
@@ -173,29 +151,16 @@ static int drm_pci_set_busid(struct drm_device *dev, struct drm_master *master)
 	} else
 		master->unique_len = len;
 
-	dev->devname =
-		kmalloc(strlen(pdriver->name) +
-			master->unique_len + 2, GFP_KERNEL);
-
-	if (dev->devname == NULL) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	sprintf(dev->devname, "%s@%s", pdriver->name,
-		master->unique);
-
 	return 0;
 err:
 	return ret;
 }
 
-static int drm_pci_set_unique(struct drm_device *dev,
-			      struct drm_master *master,
-			      struct drm_unique *u)
+int drm_pci_set_unique(struct drm_device *dev,
+		       struct drm_master *master,
+		       struct drm_unique *u)
 {
 	int domain, bus, slot, func, ret;
-	const char *bus_name;
 
 	master->unique_len = u->unique_len;
 	master->unique_size = u->unique_len + 1;
@@ -212,17 +177,6 @@ static int drm_pci_set_unique(struct drm_device *dev,
 
 	master->unique[master->unique_len] = '\0';
 
-	bus_name = dev->driver->bus->get_name(dev);
-	dev->devname = kmalloc(strlen(bus_name) +
-			       strlen(master->unique) + 2, GFP_KERNEL);
-	if (!dev->devname) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	sprintf(dev->devname, "%s@%s", bus_name,
-		master->unique);
-
 	/* Return error if the busid submitted doesn't match the device's actual
 	 * busid.
 	 */
@@ -247,7 +201,6 @@ err:
 	return ret;
 }
 
-
 static int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
 {
 	if ((p->busnum >> 8) != drm_get_pci_domain(dev) ||
@@ -262,6 +215,36 @@ static int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
 	return 0;
 }
 
+/**
+ * drm_irq_by_busid - Get interrupt from bus ID
+ * @dev: DRM device
+ * @data: IOCTL parameter pointing to a drm_irq_busid structure
+ * @file_priv: DRM file private.
+ *
+ * Finds the PCI device with the specified bus id and gets its IRQ number.
+ * This IOCTL is deprecated, and will now return EINVAL for any busid not equal
+ * to that of the device that this DRM instance attached to.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int drm_irq_by_busid(struct drm_device *dev, void *data,
+		     struct drm_file *file_priv)
+{
+	struct drm_irq_busid *p = data;
+
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return -EINVAL;
+
+	/* UMS was only ever support on PCI devices. */
+	if (WARN_ON(!dev->pdev))
+		return -EINVAL;
+
+	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
+		return -EINVAL;
+
+	return drm_pci_irq_by_busid(dev, p);
+}
+
 static void drm_pci_agp_init(struct drm_device *dev)
 {
 	if (drm_core_check_feature(dev, DRIVER_USE_AGP)) {
@@ -287,24 +270,20 @@ void drm_pci_agp_destroy(struct drm_device *dev)
 }
 
 static struct drm_bus drm_pci_bus = {
-	.bus_type = DRIVER_BUS_PCI,
-	.get_irq = drm_pci_get_irq,
-	.get_name = drm_pci_get_name,
 	.set_busid = drm_pci_set_busid,
-	.set_unique = drm_pci_set_unique,
-	.irq_by_busid = drm_pci_irq_by_busid,
 };
 
 /**
- * Register.
- *
- * \param pdev - PCI device structure
- * \param ent entry from the PCI ID table with device type flags
- * \return zero on success or a negative number on failure.
+ * drm_get_pci_dev - Register a PCI device with the DRM subsystem
+ * @pdev: PCI device
+ * @ent: entry from the PCI ID table that matches @pdev
+ * @driver: DRM device driver
  *
  * Attempt to gets inter module "drm" information. If we are first
  * then register the character device and inter module information.
  * Try and register, if we fail to register, backout previous work.
+ *
+ * Return: 0 on success or a negative error code on failure.
  */
 int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent,
 		    struct drm_driver *driver)
@@ -357,15 +336,14 @@ err_free:
 EXPORT_SYMBOL(drm_get_pci_dev);
 
 /**
- * PCI device initialization. Called direct from modules at load time.
- *
- * \return zero on success or a negative number on failure.
+ * drm_pci_init - Register matching PCI devices with the DRM subsystem
+ * @driver: DRM device driver
+ * @pdriver: PCI device driver
  *
- * Initializes a drm_device structures,registering the
- * stubs and initializing the AGP device.
+ * Initializes a drm_device structures, registering the stubs and initializing
+ * the AGP device.
  *
- * Expands the \c DRIVER_PREINIT and \c DRIVER_POST_INIT macros before and
- * after the initialization for driver customization.
+ * Return: 0 on success or a negative error code on failure.
  */
 int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 {
@@ -375,7 +353,6 @@ int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 
 	DRM_DEBUG("\n");
 
-	driver->kdriver.pci = pdriver;
 	driver->bus = &drm_pci_bus;
 
 	if (driver->driver_features & DRIVER_MODESET)
@@ -453,11 +430,31 @@ int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 }
 
 void drm_pci_agp_destroy(struct drm_device *dev) {}
+
+int drm_irq_by_busid(struct drm_device *dev, void *data,
+		     struct drm_file *file_priv)
+{
+	return -EINVAL;
+}
+
+int drm_pci_set_unique(struct drm_device *dev,
+		       struct drm_master *master,
+		       struct drm_unique *u)
+{
+	return -EINVAL;
+}
 #endif
 
 EXPORT_SYMBOL(drm_pci_init);
 
-/*@}*/
+/**
+ * drm_pci_exit - Unregister matching PCI devices from the DRM subsystem
+ * @driver: DRM device driver
+ * @pdriver: PCI device driver
+ *
+ * Unregisters one or more devices matched by a PCI driver from the DRM
+ * subsystem.
+ */
 void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
 {
 	struct drm_device *dev, *tmp;
diff --git a/drivers/gpu/drm/drm_plane_helper.c b/drivers/gpu/drm/drm_plane_helper.c
index d2b1c03b3d71..6d133149cc74 100644
--- a/drivers/gpu/drm/drm_plane_helper.c
+++ b/drivers/gpu/drm/drm_plane_helper.c
@@ -25,7 +25,9 @@
 
 #include <linux/list.h>
 #include <drm/drmP.h>
+#include <drm/drm_plane_helper.h>
 #include <drm/drm_rect.h>
+#include <drm/drm_plane_helper.h>
 
 #define SUBPIXEL_MASK 0xffff
 
@@ -36,9 +38,9 @@
  * creating the primary plane.  However drivers that still call
  * drm_plane_init() will use this minimal format list as the default.
  */
-const static uint32_t safe_modeset_formats[] = {
-       DRM_FORMAT_XRGB8888,
-       DRM_FORMAT_ARGB8888,
+static const uint32_t safe_modeset_formats[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ARGB8888,
 };
 
 /*
@@ -54,6 +56,13 @@ static int get_connectors_for_crtc(struct drm_crtc *crtc,
 	struct drm_connector *connector;
 	int count = 0;
 
+	/*
+	 * Note: Once we change the plane hooks to more fine-grained locking we
+	 * need to grab the connection_mutex here to be able to make these
+	 * checks.
+	 */
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->encoder && connector->encoder->crtc == crtc) {
 			if (connector_list != NULL && count < num_connectors)
@@ -66,6 +75,79 @@ static int get_connectors_for_crtc(struct drm_crtc *crtc,
 }
 
 /**
+ * drm_plane_helper_check_update() - Check plane update for validity
+ * @plane: plane object to update
+ * @crtc: owning CRTC of owning plane
+ * @fb: framebuffer to flip onto plane
+ * @src: source coordinates in 16.16 fixed point
+ * @dest: integer destination coordinates
+ * @clip: integer clipping coordinates
+ * @min_scale: minimum @src:@dest scaling factor in 16.16 fixed point
+ * @max_scale: maximum @src:@dest scaling factor in 16.16 fixed point
+ * @can_position: is it legal to position the plane such that it
+ *                doesn't cover the entire crtc?  This will generally
+ *                only be false for primary planes.
+ * @can_update_disabled: can the plane be updated while the crtc
+ *                       is disabled?
+ * @visible: output parameter indicating whether plane is still visible after
+ *           clipping
+ *
+ * Checks that a desired plane update is valid.  Drivers that provide
+ * their own plane handling rather than helper-provided implementations may
+ * still wish to call this function to avoid duplication of error checking
+ * code.
+ *
+ * RETURNS:
+ * Zero if update appears valid, error code on failure
+ */
+int drm_plane_helper_check_update(struct drm_plane *plane,
+				    struct drm_crtc *crtc,
+				    struct drm_framebuffer *fb,
+				    struct drm_rect *src,
+				    struct drm_rect *dest,
+				    const struct drm_rect *clip,
+				    int min_scale,
+				    int max_scale,
+				    bool can_position,
+				    bool can_update_disabled,
+				    bool *visible)
+{
+	int hscale, vscale;
+
+	if (!crtc->enabled && !can_update_disabled) {
+		DRM_DEBUG_KMS("Cannot update plane of a disabled CRTC.\n");
+		return -EINVAL;
+	}
+
+	/* Check scaling */
+	hscale = drm_rect_calc_hscale(src, dest, min_scale, max_scale);
+	vscale = drm_rect_calc_vscale(src, dest, min_scale, max_scale);
+	if (hscale < 0 || vscale < 0) {
+		DRM_DEBUG_KMS("Invalid scaling of plane\n");
+		return -ERANGE;
+	}
+
+	*visible = drm_rect_clip_scaled(src, dest, clip, hscale, vscale);
+	if (!*visible)
+		/*
+		 * Plane isn't visible; some drivers can handle this
+		 * so we just return success here.  Drivers that can't
+		 * (including those that use the primary plane helper's
+		 * update function) will return an error from their
+		 * update_plane handler.
+		 */
+		return 0;
+
+	if (!can_position && !drm_rect_equals(dest, clip)) {
+		DRM_DEBUG_KMS("Plane must cover entire CRTC\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_plane_helper_check_update);
+
+/**
  * drm_primary_helper_update() - Helper for primary plane update
  * @plane: plane object to update
  * @crtc: owning CRTC of owning plane
@@ -113,56 +195,42 @@ int drm_primary_helper_update(struct drm_plane *plane, struct drm_crtc *crtc,
 		.x = src_x >> 16,
 		.y = src_y >> 16,
 	};
+	struct drm_rect src = {
+		.x1 = src_x,
+		.y1 = src_y,
+		.x2 = src_x + src_w,
+		.y2 = src_y + src_h,
+	};
 	struct drm_rect dest = {
 		.x1 = crtc_x,
 		.y1 = crtc_y,
 		.x2 = crtc_x + crtc_w,
 		.y2 = crtc_y + crtc_h,
 	};
-	struct drm_rect clip = {
+	const struct drm_rect clip = {
 		.x2 = crtc->mode.hdisplay,
 		.y2 = crtc->mode.vdisplay,
 	};
 	struct drm_connector **connector_list;
-	struct drm_framebuffer *tmpfb;
 	int num_connectors, ret;
+	bool visible;
 
-	if (!crtc->enabled) {
-		DRM_DEBUG_KMS("Cannot update primary plane of a disabled CRTC.\n");
-		return -EINVAL;
-	}
-
-	/* Disallow subpixel positioning */
-	if ((src_x | src_y | src_w | src_h) & SUBPIXEL_MASK) {
-		DRM_DEBUG_KMS("Primary plane does not support subpixel positioning\n");
-		return -EINVAL;
-	}
-
-	/* Primary planes are locked to their owning CRTC */
-	if (plane->possible_crtcs != drm_crtc_mask(crtc)) {
-		DRM_DEBUG_KMS("Cannot change primary plane CRTC\n");
-		return -EINVAL;
-	}
-
-	/* Disallow scaling */
-	if (crtc_w != src_w || crtc_h != src_h) {
-		DRM_DEBUG_KMS("Can't scale primary plane\n");
-		return -EINVAL;
-	}
-
-	/* Make sure primary plane covers entire CRTC */
-	drm_rect_intersect(&dest, &clip);
-	if (dest.x1 != 0 || dest.y1 != 0 ||
-	    dest.x2 != crtc->mode.hdisplay || dest.y2 != crtc->mode.vdisplay) {
-		DRM_DEBUG_KMS("Primary plane must cover entire CRTC\n");
-		return -EINVAL;
-	}
-
-	/* Framebuffer must be big enough to cover entire plane */
-	ret = drm_crtc_check_viewport(crtc, crtc_x, crtc_y, &crtc->mode, fb);
+	ret = drm_plane_helper_check_update(plane, crtc, fb,
+					    &src, &dest, &clip,
+					    DRM_PLANE_HELPER_NO_SCALING,
+					    DRM_PLANE_HELPER_NO_SCALING,
+					    false, false, &visible);
 	if (ret)
 		return ret;
 
+	if (!visible)
+		/*
+		 * Primary plane isn't visible.  Note that unless a driver
+		 * provides their own disable function, this will just
+		 * wind up returning -EINVAL to userspace.
+		 */
+		return plane->funcs->disable_plane(plane);
+
 	/* Find current connectors for CRTC */
 	num_connectors = get_connectors_for_crtc(crtc, NULL, 0);
 	BUG_ON(num_connectors == 0);
@@ -176,21 +244,14 @@ int drm_primary_helper_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	set.num_connectors = num_connectors;
 
 	/*
-	 * set_config() adjusts crtc->primary->fb; however the DRM setplane
-	 * code that called us expects to handle the framebuffer update and
-	 * reference counting; save and restore the current fb before
-	 * calling it.
-	 *
-	 * N.B., we call set_config() directly here rather than using
+	 * We call set_config() directly here rather than using
 	 * drm_mode_set_config_internal.  We're reprogramming the same
 	 * connectors that were already in use, so we shouldn't need the extra
 	 * cross-CRTC fb refcounting to accomodate stealing connectors.
 	 * drm_mode_setplane() already handles the basic refcounting for the
 	 * framebuffers involved in this operation.
 	 */
-	tmpfb = plane->fb;
 	ret = crtc->funcs->set_config(&set);
-	plane->fb = tmpfb;
 
 	kfree(connector_list);
 	return ret;
@@ -232,7 +293,6 @@ EXPORT_SYMBOL(drm_primary_helper_disable);
  */
 void drm_primary_helper_destroy(struct drm_plane *plane)
 {
-	plane->funcs->disable_plane(plane);
 	drm_plane_cleanup(plane);
 	kfree(plane);
 }
diff --git a/drivers/gpu/drm/drm_platform.c b/drivers/gpu/drm/drm_platform.c
index 319ff5385601..d5b76f148c12 100644
--- a/drivers/gpu/drm/drm_platform.c
+++ b/drivers/gpu/drm/drm_platform.c
@@ -68,16 +68,6 @@ err_free:
 	return ret;
 }
 
-static int drm_platform_get_irq(struct drm_device *dev)
-{
-	return platform_get_irq(dev->platformdev, 0);
-}
-
-static const char *drm_platform_get_name(struct drm_device *dev)
-{
-	return dev->platformdev->name;
-}
-
 static int drm_platform_set_busid(struct drm_device *dev, struct drm_master *master)
 {
 	int len, ret, id;
@@ -106,46 +96,30 @@ static int drm_platform_set_busid(struct drm_device *dev, struct drm_master *mas
 		goto err;
 	}
 
-	dev->devname =
-		kmalloc(strlen(dev->platformdev->name) +
-			master->unique_len + 2, GFP_KERNEL);
-
-	if (dev->devname == NULL) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	sprintf(dev->devname, "%s@%s", dev->platformdev->name,
-		master->unique);
 	return 0;
 err:
 	return ret;
 }
 
 static struct drm_bus drm_platform_bus = {
-	.bus_type = DRIVER_BUS_PLATFORM,
-	.get_irq = drm_platform_get_irq,
-	.get_name = drm_platform_get_name,
 	.set_busid = drm_platform_set_busid,
 };
 
 /**
- * Platform device initialization. Called direct from modules.
+ * drm_platform_init - Register a platform device with the DRM subsystem
+ * @driver: DRM device driver
+ * @platform_device: platform device to register
  *
- * \return zero on success or a negative number on failure.
- *
- * Initializes a drm_device structures,registering the
- * stubs
+ * Registers the specified DRM device driver and platform device with the DRM
+ * subsystem, initializing a drm_device structure and calling the driver's
+ * .load() function.
  *
- * Expands the \c DRIVER_PREINIT and \c DRIVER_POST_INIT macros before and
- * after the initialization for driver customization.
+ * Return: 0 on success or a negative error code on failure.
  */
-
 int drm_platform_init(struct drm_driver *driver, struct platform_device *platform_device)
 {
 	DRM_DEBUG("\n");
 
-	driver->kdriver.platform_device = platform_device;
 	driver->bus = &drm_platform_bus;
 	return drm_get_platform_dev(platform_device, driver);
 }
diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c
index e70f54d4a581..d22676b89cbb 100644
--- a/drivers/gpu/drm/drm_probe_helper.c
+++ b/drivers/gpu/drm/drm_probe_helper.c
@@ -82,26 +82,8 @@ static void drm_mode_validate_flag(struct drm_connector *connector,
 	return;
 }
 
-/**
- * drm_helper_probe_single_connector_modes - get complete set of display modes
- * @connector: connector to probe
- * @maxX: max width for modes
- * @maxY: max height for modes
- *
- * Based on the helper callbacks implemented by @connector try to detect all
- * valid modes.  Modes will first be added to the connector's probed_modes list,
- * then culled (based on validity and the @maxX, @maxY parameters) and put into
- * the normal modes list.
- *
- * Intended to be use as a generic implementation of the ->fill_modes()
- * @connector vfunc for drivers that use the crtc helpers for output mode
- * filtering and detection.
- *
- * Returns:
- * The number of modes found on @connector.
- */
-int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
-					    uint32_t maxX, uint32_t maxY)
+static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connector *connector,
+							      uint32_t maxX, uint32_t maxY, bool merge_type_bits)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_display_mode *mode;
@@ -114,7 +96,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id,
-			drm_get_connector_name(connector));
+			connector->name);
 	/* set all modes to the unverified state */
 	list_for_each_entry(mode, &connector->modes, head)
 		mode->status = MODE_UNVERIFIED;
@@ -138,7 +120,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 
 	if (connector->status == connector_status_disconnected) {
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] disconnected\n",
-			connector->base.id, drm_get_connector_name(connector));
+			connector->base.id, connector->name);
 		drm_mode_connector_update_edid_property(connector, NULL);
 		verbose_prune = false;
 		goto prune;
@@ -155,7 +137,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 	if (count == 0)
 		goto prune;
 
-	drm_mode_connector_list_update(connector);
+	drm_mode_connector_list_update(connector, merge_type_bits);
 
 	if (maxX && maxY)
 		drm_mode_validate_size(dev, &connector->modes, maxX, maxY);
@@ -169,7 +151,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 	drm_mode_validate_flag(connector, mode_flags);
 
 	list_for_each_entry(mode, &connector->modes, head) {
-		if (mode->status == MODE_OK)
+		if (mode->status == MODE_OK && connector_funcs->mode_valid)
 			mode->status = connector_funcs->mode_valid(connector,
 								   mode);
 	}
@@ -186,7 +168,7 @@ prune:
 	drm_mode_sort(&connector->modes);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] probed modes :\n", connector->base.id,
-			drm_get_connector_name(connector));
+			connector->name);
 	list_for_each_entry(mode, &connector->modes, head) {
 		drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V);
 		drm_mode_debug_printmodeline(mode);
@@ -194,9 +176,49 @@ prune:
 
 	return count;
 }
+
+/**
+ * drm_helper_probe_single_connector_modes - get complete set of display modes
+ * @connector: connector to probe
+ * @maxX: max width for modes
+ * @maxY: max height for modes
+ *
+ * Based on the helper callbacks implemented by @connector try to detect all
+ * valid modes.  Modes will first be added to the connector's probed_modes list,
+ * then culled (based on validity and the @maxX, @maxY parameters) and put into
+ * the normal modes list.
+ *
+ * Intended to be use as a generic implementation of the ->fill_modes()
+ * @connector vfunc for drivers that use the crtc helpers for output mode
+ * filtering and detection.
+ *
+ * Returns:
+ * The number of modes found on @connector.
+ */
+int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
+					    uint32_t maxX, uint32_t maxY)
+{
+	return drm_helper_probe_single_connector_modes_merge_bits(connector, maxX, maxY, true);
+}
 EXPORT_SYMBOL(drm_helper_probe_single_connector_modes);
 
 /**
+ * drm_helper_probe_single_connector_modes_nomerge - get complete set of display modes
+ * @connector: connector to probe
+ * @maxX: max width for modes
+ * @maxY: max height for modes
+ *
+ * This operates like drm_hehlper_probe_single_connector_modes except it
+ * replaces the mode bits instead of merging them for preferred modes.
+ */
+int drm_helper_probe_single_connector_modes_nomerge(struct drm_connector *connector,
+					    uint32_t maxX, uint32_t maxY)
+{
+	return drm_helper_probe_single_connector_modes_merge_bits(connector, maxX, maxY, false);
+}
+EXPORT_SYMBOL(drm_helper_probe_single_connector_modes_nomerge);
+
+/**
  * drm_kms_helper_hotplug_event - fire off KMS hotplug events
  * @dev: drm_device whose connector state changed
  *
@@ -264,7 +286,7 @@ static void output_poll_execute(struct work_struct *work)
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] "
 				      "status updated from %s to %s\n",
 				      connector->base.id,
-				      drm_get_connector_name(connector),
+				      connector->name,
 				      old, new);
 
 			changed = true;
@@ -409,7 +431,7 @@ bool drm_helper_hpd_irq_event(struct drm_device *dev)
 		connector->status = connector->funcs->detect(connector, false);
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n",
 			      connector->base.id,
-			      drm_get_connector_name(connector),
+			      connector->name,
 			      drm_get_connector_status_name(old_status),
 			      drm_get_connector_status_name(connector->status));
 		if (old_status != connector->status)
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index 4c24c3ac1efa..14d16464000a 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -1,16 +1,11 @@
-/**
- * \file drm_stub.h
- * Stub support
- *
- * \author Rickard E. (Rik) Faith <faith@valinux.com>
- */
-
 /*
  * Created: Fri Jan 19 10:48:35 2001 by faith@acm.org
  *
  * Copyright 2001 VA Linux Systems, Inc., Sunnyvale, California.
  * All Rights Reserved.
  *
+ * Author Rickard E. (Rik) Faith <faith@valinux.com>
+ *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
  * to deal in the Software without restriction, including without limitation
@@ -128,7 +123,10 @@ struct drm_master *drm_master_create(struct drm_minor *minor)
 	kref_init(&master->refcount);
 	spin_lock_init(&master->lock.spinlock);
 	init_waitqueue_head(&master->lock.lock_queue);
-	drm_ht_create(&master->magiclist, DRM_MAGIC_HASH_ORDER);
+	if (drm_ht_create(&master->magiclist, DRM_MAGIC_HASH_ORDER)) {
+		kfree(master);
+		return NULL;
+	}
 	INIT_LIST_HEAD(&master->magicfree);
 	master->minor = minor;
 
@@ -166,9 +164,6 @@ static void drm_master_destroy(struct kref *kref)
 		master->unique_len = 0;
 	}
 
-	kfree(dev->devname);
-	dev->devname = NULL;
-
 	list_for_each_entry_safe(pt, next, &master->magicfree, head) {
 		list_del(&pt->head);
 		drm_ht_remove_item(&master->magiclist, &pt->hash_item);
@@ -294,6 +289,7 @@ static void drm_minor_free(struct drm_device *dev, unsigned int type)
 
 	slot = drm_minor_get_slot(dev, type);
 	if (*slot) {
+		drm_mode_group_destroy(&(*slot)->mode_group);
 		kfree(*slot);
 		*slot = NULL;
 	}
@@ -424,11 +420,15 @@ void drm_minor_release(struct drm_minor *minor)
 }
 
 /**
- * Called via drm_exit() at module unload time or when pci device is
- * unplugged.
+ * drm_put_dev - Unregister and release a DRM device
+ * @dev: DRM device
  *
- * Cleans up all DRM device, calling drm_lastclose().
+ * Called at module unload time or when a PCI device is unplugged.
  *
+ * Use of this function is discouraged. It will eventually go away completely.
+ * Please use drm_dev_unregister() and drm_dev_unref() explicitly instead.
+ *
+ * Cleans up all DRM device, calling drm_lastclose().
  */
 void drm_put_dev(struct drm_device *dev)
 {
@@ -535,7 +535,7 @@ static void drm_fs_inode_free(struct inode *inode)
 }
 
 /**
- * drm_dev_alloc - Allocate new drm device
+ * drm_dev_alloc - Allocate new DRM device
  * @driver: DRM driver to allocate device for
  * @parent: Parent device object
  *
@@ -569,7 +569,7 @@ struct drm_device *drm_dev_alloc(struct drm_driver *driver,
 	INIT_LIST_HEAD(&dev->maplist);
 	INIT_LIST_HEAD(&dev->vblank_event_list);
 
-	spin_lock_init(&dev->count_lock);
+	spin_lock_init(&dev->buf_lock);
 	spin_lock_init(&dev->event_lock);
 	mutex_init(&dev->struct_mutex);
 	mutex_init(&dev->ctxlist_mutex);
@@ -648,9 +648,8 @@ static void drm_dev_release(struct kref *ref)
 	drm_minor_free(dev, DRM_MINOR_RENDER);
 	drm_minor_free(dev, DRM_MINOR_CONTROL);
 
-	kfree(dev->devname);
-
 	mutex_destroy(&dev->master_mutex);
+	kfree(dev->unique);
 	kfree(dev);
 }
 
@@ -690,6 +689,7 @@ EXPORT_SYMBOL(drm_dev_unref);
 /**
  * drm_dev_register - Register DRM device
  * @dev: Device to register
+ * @flags: Flags passed to the driver's .load() function
  *
  * Register the DRM device @dev with the system, advertise device to user-space
  * and start normal device operation. @dev must be allocated via drm_dev_alloc()
@@ -778,3 +778,28 @@ void drm_dev_unregister(struct drm_device *dev)
 	drm_minor_unregister(dev, DRM_MINOR_CONTROL);
 }
 EXPORT_SYMBOL(drm_dev_unregister);
+
+/**
+ * drm_dev_set_unique - Set the unique name of a DRM device
+ * @dev: device of which to set the unique name
+ * @fmt: format string for unique name
+ *
+ * Sets the unique name of a DRM device using the specified format string and
+ * a variable list of arguments. Drivers can use this at driver probe time if
+ * the unique name of the devices they drive is static.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int drm_dev_set_unique(struct drm_device *dev, const char *fmt, ...)
+{
+	va_list ap;
+
+	kfree(dev->unique);
+
+	va_start(ap, fmt);
+	dev->unique = kvasprintf(GFP_KERNEL, fmt, ap);
+	va_end(ap);
+
+	return dev->unique ? 0 : -ENOMEM;
+}
+EXPORT_SYMBOL(drm_dev_set_unique);
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index c22c3097c3e8..369b26278e76 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -380,9 +380,9 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 
 	connector->kdev = device_create(drm_class, dev->primary->kdev,
 					0, connector, "card%d-%s",
-					dev->primary->index, drm_get_connector_name(connector));
+					dev->primary->index, connector->name);
 	DRM_DEBUG("adding \"%s\" to sysfs\n",
-		  drm_get_connector_name(connector));
+		  connector->name);
 
 	if (IS_ERR(connector->kdev)) {
 		DRM_ERROR("failed to register connector device: %ld\n", PTR_ERR(connector->kdev));
@@ -460,7 +460,7 @@ void drm_sysfs_connector_remove(struct drm_connector *connector)
 	if (!connector->kdev)
 		return;
 	DRM_DEBUG("removing \"%s\" from sysfs\n",
-		  drm_get_connector_name(connector));
+		  connector->name);
 
 	for (i = 0; i < ARRAY_SIZE(connector_attrs); i++)
 		device_remove_file(connector->kdev, &connector_attrs[i]);
diff --git a/drivers/gpu/drm/drm_usb.c b/drivers/gpu/drm/drm_usb.c
index c3406aad2944..f2fe94aab901 100644
--- a/drivers/gpu/drm/drm_usb.c
+++ b/drivers/gpu/drm/drm_usb.c
@@ -36,16 +36,6 @@ err_free:
 }
 EXPORT_SYMBOL(drm_get_usb_dev);
 
-static int drm_usb_get_irq(struct drm_device *dev)
-{
-	return 0;
-}
-
-static const char *drm_usb_get_name(struct drm_device *dev)
-{
-	return "USB";
-}
-
 static int drm_usb_set_busid(struct drm_device *dev,
 			       struct drm_master *master)
 {
@@ -53,18 +43,24 @@ static int drm_usb_set_busid(struct drm_device *dev,
 }
 
 static struct drm_bus drm_usb_bus = {
-	.bus_type = DRIVER_BUS_USB,
-	.get_irq = drm_usb_get_irq,
-	.get_name = drm_usb_get_name,
 	.set_busid = drm_usb_set_busid,
 };
-    
+
+/**
+ * drm_usb_init - Register matching USB devices with the DRM subsystem
+ * @driver: DRM device driver
+ * @udriver: USB device driver
+ *
+ * Registers one or more devices matched by a USB driver with the DRM
+ * subsystem.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
 int drm_usb_init(struct drm_driver *driver, struct usb_driver *udriver)
 {
 	int res;
 	DRM_DEBUG("\n");
 
-	driver->kdriver.usb = udriver;
 	driver->bus = &drm_usb_bus;
 
 	res = usb_register(udriver);
@@ -72,6 +68,14 @@ int drm_usb_init(struct drm_driver *driver, struct usb_driver *udriver)
 }
 EXPORT_SYMBOL(drm_usb_init);
 
+/**
+ * drm_usb_exit - Unregister matching USB devices from the DRM subsystem
+ * @driver: DRM device driver
+ * @udriver: USB device driver
+ *
+ * Unregisters one or more devices matched by a USB driver from the DRM
+ * subsystem.
+ */
 void drm_usb_exit(struct drm_driver *driver,
 		  struct usb_driver *udriver)
 {
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 5bf5bca94f56..178d2a9672a8 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -26,14 +26,14 @@ config DRM_EXYNOS_DMABUF
 
 config DRM_EXYNOS_FIMD
 	bool "Exynos DRM FIMD"
-	depends on DRM_EXYNOS && !FB_S3C && !ARCH_MULTIPLATFORM
+	depends on DRM_EXYNOS && !FB_S3C
 	select FB_MODE_HELPERS
 	help
 	  Choose this option if you want to use Exynos FIMD for DRM.
 
 config DRM_EXYNOS_DPI
 	bool "EXYNOS DRM parallel output support"
-	depends on DRM_EXYNOS
+	depends on DRM_EXYNOS_FIMD
 	select DRM_PANEL
 	default n
 	help
@@ -41,7 +41,7 @@ config DRM_EXYNOS_DPI
 
 config DRM_EXYNOS_DSI
 	bool "EXYNOS DRM MIPI-DSI driver support"
-	depends on DRM_EXYNOS
+	depends on DRM_EXYNOS_FIMD
 	select DRM_MIPI_DSI
 	select DRM_PANEL
 	default n
@@ -50,7 +50,7 @@ config DRM_EXYNOS_DSI
 
 config DRM_EXYNOS_DP
 	bool "EXYNOS DRM DP driver support"
-	depends on DRM_EXYNOS && ARCH_EXYNOS
+	depends on DRM_EXYNOS_FIMD && ARCH_EXYNOS && (DRM_PTN3460=n || DRM_PTN3460=y || DRM_PTN3460=DRM_EXYNOS)
 	default DRM_EXYNOS
 	help
 	  This enables support for DP device.
diff --git a/drivers/gpu/drm/exynos/exynos_ddc.c b/drivers/gpu/drm/exynos/exynos_ddc.c
deleted file mode 100644
index 6a8c84e7c839..000000000000
--- a/drivers/gpu/drm/exynos/exynos_ddc.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/*
- * Copyright (C) 2011 Samsung Electronics Co.Ltd
- * Authors:
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *	Inki Dae <inki.dae@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <drm/drmP.h>
-
-#include <linux/kernel.h>
-#include <linux/i2c.h>
-#include <linux/of.h>
-
-#include "exynos_drm_drv.h"
-#include "exynos_hdmi.h"
-
-static int s5p_ddc_probe(struct i2c_client *client,
-			const struct i2c_device_id *dev_id)
-{
-	hdmi_attach_ddc_client(client);
-
-	dev_info(&client->adapter->dev,
-		"attached %s into i2c adapter successfully\n",
-		client->name);
-
-	return 0;
-}
-
-static int s5p_ddc_remove(struct i2c_client *client)
-{
-	dev_info(&client->adapter->dev,
-		"detached %s from i2c adapter successfully\n",
-		client->name);
-
-	return 0;
-}
-
-static struct of_device_id hdmiddc_match_types[] = {
-	{
-		.compatible = "samsung,exynos5-hdmiddc",
-	}, {
-		.compatible = "samsung,exynos4210-hdmiddc",
-	}, {
-		/* end node */
-	}
-};
-
-struct i2c_driver ddc_driver = {
-	.driver = {
-		.name = "exynos-hdmiddc",
-		.owner = THIS_MODULE,
-		.of_match_table = hdmiddc_match_types,
-	},
-	.probe		= s5p_ddc_probe,
-	.remove		= s5p_ddc_remove,
-	.command		= NULL,
-};
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c
index aed533bbfd31..a8ffc8c1477b 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.c
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.c
@@ -18,6 +18,9 @@
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/gpio.h>
+#include <linux/component.h>
 #include <linux/phy/phy.h>
 #include <video/of_display_timing.h>
 #include <video/of_videomode.h>
@@ -141,15 +144,15 @@ static int exynos_dp_read_edid(struct exynos_dp_device *dp)
 			return -EIO;
 		}
 
-		exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_TEST_REQUEST,
+		exynos_dp_read_byte_from_dpcd(dp, DP_TEST_REQUEST,
 					&test_vector);
-		if (test_vector & DPCD_TEST_EDID_READ) {
+		if (test_vector & DP_TEST_LINK_EDID_READ) {
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_EDID_CHECKSUM,
+				DP_TEST_EDID_CHECKSUM,
 				edid[EDID_BLOCK_LENGTH + EDID_CHECKSUM]);
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_RESPONSE,
-				DPCD_TEST_EDID_CHECKSUM_WRITE);
+				DP_TEST_RESPONSE,
+				DP_TEST_EDID_CHECKSUM_WRITE);
 		}
 	} else {
 		dev_info(dp->dev, "EDID data does not include any extensions.\n");
@@ -171,15 +174,15 @@ static int exynos_dp_read_edid(struct exynos_dp_device *dp)
 		}
 
 		exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_TEST_REQUEST,
+			DP_TEST_REQUEST,
 			&test_vector);
-		if (test_vector & DPCD_TEST_EDID_READ) {
+		if (test_vector & DP_TEST_LINK_EDID_READ) {
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_EDID_CHECKSUM,
+				DP_TEST_EDID_CHECKSUM,
 				edid[EDID_CHECKSUM]);
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_RESPONSE,
-				DPCD_TEST_EDID_CHECKSUM_WRITE);
+				DP_TEST_RESPONSE,
+				DP_TEST_EDID_CHECKSUM_WRITE);
 		}
 	}
 
@@ -193,8 +196,8 @@ static int exynos_dp_handle_edid(struct exynos_dp_device *dp)
 	int i;
 	int retval;
 
-	/* Read DPCD DPCD_ADDR_DPCD_REV~RECEIVE_PORT1_CAP_1 */
-	retval = exynos_dp_read_bytes_from_dpcd(dp, DPCD_ADDR_DPCD_REV,
+	/* Read DPCD DP_DPCD_REV~RECEIVE_PORT1_CAP_1 */
+	retval = exynos_dp_read_bytes_from_dpcd(dp, DP_DPCD_REV,
 				12, buf);
 	if (retval)
 		return retval;
@@ -214,14 +217,14 @@ static void exynos_dp_enable_rx_to_enhanced_mode(struct exynos_dp_device *dp,
 {
 	u8 data;
 
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_LANE_COUNT_SET, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_LANE_COUNT_SET, &data);
 
 	if (enable)
-		exynos_dp_write_byte_to_dpcd(dp, DPCD_ADDR_LANE_COUNT_SET,
-			DPCD_ENHANCED_FRAME_EN |
+		exynos_dp_write_byte_to_dpcd(dp, DP_LANE_COUNT_SET,
+			DP_LANE_COUNT_ENHANCED_FRAME_EN |
 			DPCD_LANE_COUNT_SET(data));
 	else
-		exynos_dp_write_byte_to_dpcd(dp, DPCD_ADDR_LANE_COUNT_SET,
+		exynos_dp_write_byte_to_dpcd(dp, DP_LANE_COUNT_SET,
 			DPCD_LANE_COUNT_SET(data));
 }
 
@@ -230,7 +233,7 @@ static int exynos_dp_is_enhanced_mode_available(struct exynos_dp_device *dp)
 	u8 data;
 	int retval;
 
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_MAX_LANE_COUNT, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_MAX_LANE_COUNT, &data);
 	retval = DPCD_ENHANCED_FRAME_CAP(data);
 
 	return retval;
@@ -250,8 +253,8 @@ static void exynos_dp_training_pattern_dis(struct exynos_dp_device *dp)
 	exynos_dp_set_training_pattern(dp, DP_NONE);
 
 	exynos_dp_write_byte_to_dpcd(dp,
-		DPCD_ADDR_TRAINING_PATTERN_SET,
-		DPCD_TRAINING_PATTERN_DISABLED);
+		DP_TRAINING_PATTERN_SET,
+		DP_TRAINING_PATTERN_DISABLE);
 }
 
 static void exynos_dp_set_lane_lane_pre_emphasis(struct exynos_dp_device *dp,
@@ -295,7 +298,7 @@ static int exynos_dp_link_start(struct exynos_dp_device *dp)
 	/* Setup RX configuration */
 	buf[0] = dp->link_train.link_rate;
 	buf[1] = dp->link_train.lane_count;
-	retval = exynos_dp_write_bytes_to_dpcd(dp, DPCD_ADDR_LINK_BW_SET,
+	retval = exynos_dp_write_bytes_to_dpcd(dp, DP_LINK_BW_SET,
 				2, buf);
 	if (retval)
 		return retval;
@@ -322,16 +325,16 @@ static int exynos_dp_link_start(struct exynos_dp_device *dp)
 
 	/* Set RX training pattern */
 	retval = exynos_dp_write_byte_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
-			DPCD_SCRAMBLING_DISABLED | DPCD_TRAINING_PATTERN_1);
+			DP_TRAINING_PATTERN_SET,
+			DP_LINK_SCRAMBLING_DISABLE | DP_TRAINING_PATTERN_1);
 	if (retval)
 		return retval;
 
 	for (lane = 0; lane < lane_count; lane++)
-		buf[lane] = DPCD_PRE_EMPHASIS_PATTERN2_LEVEL0 |
-			    DPCD_VOLTAGE_SWING_PATTERN1_LEVEL0;
+		buf[lane] = DP_TRAIN_PRE_EMPHASIS_0 |
+			    DP_TRAIN_VOLTAGE_SWING_400;
 
-	retval = exynos_dp_write_bytes_to_dpcd(dp, DPCD_ADDR_TRAINING_LANE0_SET,
+	retval = exynos_dp_write_bytes_to_dpcd(dp, DP_TRAINING_LANE0_SET,
 			lane_count, buf);
 
 	return retval;
@@ -352,7 +355,7 @@ static int exynos_dp_clock_recovery_ok(u8 link_status[2], int lane_count)
 
 	for (lane = 0; lane < lane_count; lane++) {
 		lane_status = exynos_dp_get_lane_status(link_status, lane);
-		if ((lane_status & DPCD_LANE_CR_DONE) == 0)
+		if ((lane_status & DP_LANE_CR_DONE) == 0)
 			return -EINVAL;
 	}
 	return 0;
@@ -364,13 +367,13 @@ static int exynos_dp_channel_eq_ok(u8 link_status[2], u8 link_align,
 	int lane;
 	u8 lane_status;
 
-	if ((link_align & DPCD_INTERLANE_ALIGN_DONE) == 0)
+	if ((link_align & DP_INTERLANE_ALIGN_DONE) == 0)
 		return -EINVAL;
 
 	for (lane = 0; lane < lane_count; lane++) {
 		lane_status = exynos_dp_get_lane_status(link_status, lane);
-		lane_status &= DPCD_CHANNEL_EQ_BITS;
-		if (lane_status != DPCD_CHANNEL_EQ_BITS)
+		lane_status &= DP_CHANNEL_EQ_BITS;
+		if (lane_status != DP_CHANNEL_EQ_BITS)
 			return -EINVAL;
 	}
 
@@ -468,9 +471,9 @@ static void exynos_dp_get_adjust_training_lane(struct exynos_dp_device *dp,
 				DPCD_PRE_EMPHASIS_SET(pre_emphasis);
 
 		if (voltage_swing == VOLTAGE_LEVEL_3)
-			training_lane |= DPCD_MAX_SWING_REACHED;
+			training_lane |= DP_TRAIN_MAX_SWING_REACHED;
 		if (pre_emphasis == PRE_EMPHASIS_LEVEL_3)
-			training_lane |= DPCD_MAX_PRE_EMPHASIS_REACHED;
+			training_lane |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
 
 		dp->link_train.training_lane[lane] = training_lane;
 	}
@@ -487,12 +490,12 @@ static int exynos_dp_process_clock_recovery(struct exynos_dp_device *dp)
 	lane_count = dp->link_train.lane_count;
 
 	retval =  exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_LANE0_1_STATUS, 2, link_status);
+			DP_LANE0_1_STATUS, 2, link_status);
 	if (retval)
 		return retval;
 
 	retval =  exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
+			DP_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
 	if (retval)
 		return retval;
 
@@ -501,9 +504,9 @@ static int exynos_dp_process_clock_recovery(struct exynos_dp_device *dp)
 		exynos_dp_set_training_pattern(dp, TRAINING_PTN2);
 
 		retval = exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TRAINING_PATTERN_SET,
-				DPCD_SCRAMBLING_DISABLED |
-				DPCD_TRAINING_PATTERN_2);
+				DP_TRAINING_PATTERN_SET,
+				DP_LINK_SCRAMBLING_DISABLE |
+				DP_TRAINING_PATTERN_2);
 		if (retval)
 			return retval;
 
@@ -543,7 +546,7 @@ static int exynos_dp_process_clock_recovery(struct exynos_dp_device *dp)
 			dp->link_train.training_lane[lane], lane);
 
 	retval = exynos_dp_write_bytes_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_LANE0_SET, lane_count,
+			DP_TRAINING_LANE0_SET, lane_count,
 			dp->link_train.training_lane);
 	if (retval)
 		return retval;
@@ -562,7 +565,7 @@ static int exynos_dp_process_equalizer_training(struct exynos_dp_device *dp)
 	lane_count = dp->link_train.lane_count;
 
 	retval = exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_LANE0_1_STATUS, 2, link_status);
+			DP_LANE0_1_STATUS, 2, link_status);
 	if (retval)
 		return retval;
 
@@ -572,12 +575,12 @@ static int exynos_dp_process_equalizer_training(struct exynos_dp_device *dp)
 	}
 
 	retval = exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
+			DP_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
 	if (retval)
 		return retval;
 
 	retval = exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_LANE_ALIGN_STATUS_UPDATED, &link_align);
+			DP_LANE_ALIGN_STATUS_UPDATED, &link_align);
 	if (retval)
 		return retval;
 
@@ -619,7 +622,7 @@ static int exynos_dp_process_equalizer_training(struct exynos_dp_device *dp)
 		exynos_dp_set_lane_link_training(dp,
 			dp->link_train.training_lane[lane], lane);
 
-	retval = exynos_dp_write_bytes_to_dpcd(dp, DPCD_ADDR_TRAINING_LANE0_SET,
+	retval = exynos_dp_write_bytes_to_dpcd(dp, DP_TRAINING_LANE0_SET,
 			lane_count, dp->link_train.training_lane);
 
 	return retval;
@@ -634,7 +637,7 @@ static void exynos_dp_get_max_rx_bandwidth(struct exynos_dp_device *dp,
 	 * For DP rev.1.1, Maximum link rate of Main Link lanes
 	 * 0x06 = 1.62 Gbps, 0x0a = 2.7 Gbps
 	 */
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_MAX_LINK_RATE, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_MAX_LINK_RATE, &data);
 	*bandwidth = data;
 }
 
@@ -647,7 +650,7 @@ static void exynos_dp_get_max_rx_lane_count(struct exynos_dp_device *dp,
 	 * For DP rev.1.1, Maximum number of Main Link lanes
 	 * 0x01 = 1 lane, 0x02 = 2 lanes, 0x04 = 4 lanes
 	 */
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_MAX_LANE_COUNT, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_MAX_LANE_COUNT, &data);
 	*lane_count = DPCD_MAX_LANE_COUNT(data);
 }
 
@@ -819,20 +822,20 @@ static void exynos_dp_enable_scramble(struct exynos_dp_device *dp, bool enable)
 		exynos_dp_enable_scrambling(dp);
 
 		exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
+			DP_TRAINING_PATTERN_SET,
 			&data);
 		exynos_dp_write_byte_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
-			(u8)(data & ~DPCD_SCRAMBLING_DISABLED));
+			DP_TRAINING_PATTERN_SET,
+			(u8)(data & ~DP_LINK_SCRAMBLING_DISABLE));
 	} else {
 		exynos_dp_disable_scrambling(dp);
 
 		exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
+			DP_TRAINING_PATTERN_SET,
 			&data);
 		exynos_dp_write_byte_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
-			(u8)(data | DPCD_SCRAMBLING_DISABLED));
+			DP_TRAINING_PATTERN_SET,
+			(u8)(data | DP_LINK_SCRAMBLING_DISABLE));
 	}
 }
 
@@ -949,12 +952,6 @@ static int exynos_dp_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int exynos_dp_mode_valid(struct drm_connector *connector,
-			struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *exynos_dp_best_encoder(
 			struct drm_connector *connector)
 {
@@ -965,20 +962,9 @@ static struct drm_encoder *exynos_dp_best_encoder(
 
 static struct drm_connector_helper_funcs exynos_dp_connector_helper_funcs = {
 	.get_modes = exynos_dp_get_modes,
-	.mode_valid = exynos_dp_mode_valid,
 	.best_encoder = exynos_dp_best_encoder,
 };
 
-static int exynos_dp_initialize(struct exynos_drm_display *display,
-				struct drm_device *drm_dev)
-{
-	struct exynos_dp_device *dp = display->ctx;
-
-	dp->drm_dev = drm_dev;
-
-	return 0;
-}
-
 static bool find_bridge(const char *compat, struct bridge_init *bridge)
 {
 	bridge->client = NULL;
@@ -1101,12 +1087,11 @@ static void exynos_dp_dpms(struct exynos_drm_display *display, int mode)
 		break;
 	default:
 		break;
-	};
+	}
 	dp->dpms_mode = mode;
 }
 
 static struct exynos_drm_display_ops exynos_dp_display_ops = {
-	.initialize = exynos_dp_initialize,
 	.create_connector = exynos_dp_create_connector,
 	.dpms = exynos_dp_dpms,
 };
@@ -1123,10 +1108,8 @@ static struct video_info *exynos_dp_dt_parse_pdata(struct device *dev)
 
 	dp_video_config = devm_kzalloc(dev,
 				sizeof(*dp_video_config), GFP_KERNEL);
-	if (!dp_video_config) {
-		dev_err(dev, "memory allocation for video config failed\n");
+	if (!dp_video_config)
 		return ERR_PTR(-ENOMEM);
-	}
 
 	dp_video_config->h_sync_polarity =
 		of_property_read_bool(dp_node, "hsync-active-high");
@@ -1185,10 +1168,7 @@ static int exynos_dp_dt_parse_phydata(struct exynos_dp_device *dp)
 	dp_phy_node = of_find_node_by_name(dp_phy_node, "dptx-phy");
 	if (!dp_phy_node) {
 		dp->phy = devm_phy_get(dp->dev, "dp");
-		if (IS_ERR(dp->phy))
-			return PTR_ERR(dp->phy);
-		else
-			return 0;
+		return PTR_ERR_OR_ZERO(dp->phy);
 	}
 
 	if (of_property_read_u32(dp_phy_node, "reg", &phy_base)) {
@@ -1230,19 +1210,20 @@ static int exynos_dp_dt_parse_panel(struct exynos_dp_device *dp)
 	return 0;
 }
 
-static int exynos_dp_probe(struct platform_device *pdev)
+static int exynos_dp_bind(struct device *dev, struct device *master, void *data)
 {
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm_dev = data;
 	struct resource *res;
 	struct exynos_dp_device *dp;
+	unsigned int irq_flags;
 
 	int ret = 0;
 
 	dp = devm_kzalloc(&pdev->dev, sizeof(struct exynos_dp_device),
 				GFP_KERNEL);
-	if (!dp) {
-		dev_err(&pdev->dev, "no memory for device data\n");
+	if (!dp)
 		return -ENOMEM;
-	}
 
 	dp->dev = &pdev->dev;
 	dp->dpms_mode = DRM_MODE_DPMS_OFF;
@@ -1273,7 +1254,30 @@ static int exynos_dp_probe(struct platform_device *pdev)
 	if (IS_ERR(dp->reg_base))
 		return PTR_ERR(dp->reg_base);
 
-	dp->irq = platform_get_irq(pdev, 0);
+	dp->hpd_gpio = of_get_named_gpio(dev->of_node, "samsung,hpd-gpio", 0);
+
+	if (gpio_is_valid(dp->hpd_gpio)) {
+		/*
+		 * Set up the hotplug GPIO from the device tree as an interrupt.
+		 * Simply specifying a different interrupt in the device tree
+		 * doesn't work since we handle hotplug rather differently when
+		 * using a GPIO.  We also need the actual GPIO specifier so
+		 * that we can get the current state of the GPIO.
+		 */
+		ret = devm_gpio_request_one(&pdev->dev, dp->hpd_gpio, GPIOF_IN,
+					    "hpd_gpio");
+		if (ret) {
+			dev_err(&pdev->dev, "failed to get hpd gpio\n");
+			return ret;
+		}
+		dp->irq = gpio_to_irq(dp->hpd_gpio);
+		irq_flags = IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING;
+	} else {
+		dp->hpd_gpio = -ENODEV;
+		dp->irq = platform_get_irq(pdev, 0);
+		irq_flags = 0;
+	}
+
 	if (dp->irq == -ENXIO) {
 		dev_err(&pdev->dev, "failed to get irq\n");
 		return -ENODEV;
@@ -1285,28 +1289,61 @@ static int exynos_dp_probe(struct platform_device *pdev)
 
 	exynos_dp_init_dp(dp);
 
-	ret = devm_request_irq(&pdev->dev, dp->irq, exynos_dp_irq_handler, 0,
-				"exynos-dp", dp);
+	ret = devm_request_irq(&pdev->dev, dp->irq, exynos_dp_irq_handler,
+			irq_flags, "exynos-dp", dp);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to request irq\n");
 		return ret;
 	}
 	disable_irq(dp->irq);
 
+	dp->drm_dev = drm_dev;
 	exynos_dp_display.ctx = dp;
 
 	platform_set_drvdata(pdev, &exynos_dp_display);
-	exynos_drm_display_register(&exynos_dp_display);
 
-	return 0;
+	return exynos_drm_create_enc_conn(drm_dev, &exynos_dp_display);
 }
 
-static int exynos_dp_remove(struct platform_device *pdev)
+static void exynos_dp_unbind(struct device *dev, struct device *master,
+				void *data)
 {
-	struct exynos_drm_display *display = platform_get_drvdata(pdev);
+	struct exynos_drm_display *display = dev_get_drvdata(dev);
+	struct exynos_dp_device *dp = display->ctx;
+	struct drm_encoder *encoder = dp->encoder;
 
 	exynos_dp_dpms(display, DRM_MODE_DPMS_OFF);
-	exynos_drm_display_unregister(&exynos_dp_display);
+
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&dp->connector);
+}
+
+static const struct component_ops exynos_dp_ops = {
+	.bind	= exynos_dp_bind,
+	.unbind	= exynos_dp_unbind,
+};
+
+static int exynos_dp_probe(struct platform_device *pdev)
+{
+	int ret;
+
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR,
+					exynos_dp_display.type);
+	if (ret)
+		return ret;
+
+	ret = component_add(&pdev->dev, &exynos_dp_ops);
+	if (ret)
+		exynos_drm_component_del(&pdev->dev,
+						EXYNOS_DEVICE_TYPE_CONNECTOR);
+
+	return ret;
+}
+
+static int exynos_dp_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &exynos_dp_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.h b/drivers/gpu/drm/exynos/exynos_dp_core.h
index d6a900d4ee40..02cc4f9ab903 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.h
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.h
@@ -14,6 +14,7 @@
 #define _EXYNOS_DP_CORE_H
 
 #include <drm/drm_crtc.h>
+#include <drm/drm_dp_helper.h>
 #include <drm/exynos_drm.h>
 
 #define DP_TIMEOUT_LOOP_COUNT 100
@@ -159,6 +160,7 @@ struct exynos_dp_device {
 	struct work_struct	hotplug_work;
 	struct phy		*phy;
 	int			dpms_mode;
+	int			hpd_gpio;
 
 	struct exynos_drm_panel_info panel;
 };
@@ -261,69 +263,17 @@ void exynos_dp_disable_scrambling(struct exynos_dp_device *dp);
 #define EDID_EXTENSION_FLAG			0x7e
 #define EDID_CHECKSUM				0x7f
 
-/* Definition for DPCD Register */
-#define DPCD_ADDR_DPCD_REV			0x0000
-#define DPCD_ADDR_MAX_LINK_RATE			0x0001
-#define DPCD_ADDR_MAX_LANE_COUNT		0x0002
-#define DPCD_ADDR_LINK_BW_SET			0x0100
-#define DPCD_ADDR_LANE_COUNT_SET		0x0101
-#define DPCD_ADDR_TRAINING_PATTERN_SET		0x0102
-#define DPCD_ADDR_TRAINING_LANE0_SET		0x0103
-#define DPCD_ADDR_LANE0_1_STATUS		0x0202
-#define DPCD_ADDR_LANE_ALIGN_STATUS_UPDATED	0x0204
-#define DPCD_ADDR_ADJUST_REQUEST_LANE0_1	0x0206
-#define DPCD_ADDR_ADJUST_REQUEST_LANE2_3	0x0207
-#define DPCD_ADDR_TEST_REQUEST			0x0218
-#define DPCD_ADDR_TEST_RESPONSE			0x0260
-#define DPCD_ADDR_TEST_EDID_CHECKSUM		0x0261
-#define DPCD_ADDR_SINK_POWER_STATE		0x0600
-
-/* DPCD_ADDR_MAX_LANE_COUNT */
+/* DP_MAX_LANE_COUNT */
 #define DPCD_ENHANCED_FRAME_CAP(x)		(((x) >> 7) & 0x1)
 #define DPCD_MAX_LANE_COUNT(x)			((x) & 0x1f)
 
-/* DPCD_ADDR_LANE_COUNT_SET */
-#define DPCD_ENHANCED_FRAME_EN			(0x1 << 7)
+/* DP_LANE_COUNT_SET */
 #define DPCD_LANE_COUNT_SET(x)			((x) & 0x1f)
 
-/* DPCD_ADDR_TRAINING_PATTERN_SET */
-#define DPCD_SCRAMBLING_DISABLED		(0x1 << 5)
-#define DPCD_SCRAMBLING_ENABLED			(0x0 << 5)
-#define DPCD_TRAINING_PATTERN_2			(0x2 << 0)
-#define DPCD_TRAINING_PATTERN_1			(0x1 << 0)
-#define DPCD_TRAINING_PATTERN_DISABLED		(0x0 << 0)
-
-/* DPCD_ADDR_TRAINING_LANE0_SET */
-#define DPCD_MAX_PRE_EMPHASIS_REACHED		(0x1 << 5)
+/* DP_TRAINING_LANE0_SET */
 #define DPCD_PRE_EMPHASIS_SET(x)		(((x) & 0x3) << 3)
 #define DPCD_PRE_EMPHASIS_GET(x)		(((x) >> 3) & 0x3)
-#define DPCD_PRE_EMPHASIS_PATTERN2_LEVEL0	(0x0 << 3)
-#define DPCD_MAX_SWING_REACHED			(0x1 << 2)
 #define DPCD_VOLTAGE_SWING_SET(x)		(((x) & 0x3) << 0)
 #define DPCD_VOLTAGE_SWING_GET(x)		(((x) >> 0) & 0x3)
-#define DPCD_VOLTAGE_SWING_PATTERN1_LEVEL0	(0x0 << 0)
-
-/* DPCD_ADDR_LANE0_1_STATUS */
-#define DPCD_LANE_SYMBOL_LOCKED			(0x1 << 2)
-#define DPCD_LANE_CHANNEL_EQ_DONE		(0x1 << 1)
-#define DPCD_LANE_CR_DONE			(0x1 << 0)
-#define DPCD_CHANNEL_EQ_BITS			(DPCD_LANE_CR_DONE|	\
-						 DPCD_LANE_CHANNEL_EQ_DONE|\
-						 DPCD_LANE_SYMBOL_LOCKED)
-
-/* DPCD_ADDR_LANE_ALIGN__STATUS_UPDATED */
-#define DPCD_LINK_STATUS_UPDATED		(0x1 << 7)
-#define DPCD_DOWNSTREAM_PORT_STATUS_CHANGED	(0x1 << 6)
-#define DPCD_INTERLANE_ALIGN_DONE		(0x1 << 0)
-
-/* DPCD_ADDR_TEST_REQUEST */
-#define DPCD_TEST_EDID_READ			(0x1 << 2)
-
-/* DPCD_ADDR_TEST_RESPONSE */
-#define DPCD_TEST_EDID_CHECKSUM_WRITE		(0x1 << 2)
-
-/* DPCD_ADDR_SINK_POWER_STATE */
-#define DPCD_SET_POWER_STATE_D0			(0x1 << 0)
-#define DPCD_SET_POWER_STATE_D4			(0x2 << 0)
 
 #endif /* _EXYNOS_DP_CORE_H */
diff --git a/drivers/gpu/drm/exynos/exynos_dp_reg.c b/drivers/gpu/drm/exynos/exynos_dp_reg.c
index b70da5052ff0..c1f87a2a9284 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_reg.c
+++ b/drivers/gpu/drm/exynos/exynos_dp_reg.c
@@ -13,6 +13,7 @@
 #include <linux/device.h>
 #include <linux/io.h>
 #include <linux/delay.h>
+#include <linux/gpio.h>
 
 #include "exynos_dp_core.h"
 #include "exynos_dp_reg.h"
@@ -326,6 +327,9 @@ void exynos_dp_clear_hotplug_interrupts(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
+	if (gpio_is_valid(dp->hpd_gpio))
+		return;
+
 	reg = HOTPLUG_CHG | HPD_LOST | PLUG;
 	writel(reg, dp->reg_base + EXYNOS_DP_COMMON_INT_STA_4);
 
@@ -337,6 +341,9 @@ void exynos_dp_init_hpd(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
+	if (gpio_is_valid(dp->hpd_gpio))
+		return;
+
 	exynos_dp_clear_hotplug_interrupts(dp);
 
 	reg = readl(dp->reg_base + EXYNOS_DP_SYS_CTL_3);
@@ -348,19 +355,27 @@ enum dp_irq_type exynos_dp_get_irq_type(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
-	/* Parse hotplug interrupt status register */
-	reg = readl(dp->reg_base + EXYNOS_DP_COMMON_INT_STA_4);
+	if (gpio_is_valid(dp->hpd_gpio)) {
+		reg = gpio_get_value(dp->hpd_gpio);
+		if (reg)
+			return DP_IRQ_TYPE_HP_CABLE_IN;
+		else
+			return DP_IRQ_TYPE_HP_CABLE_OUT;
+	} else {
+		/* Parse hotplug interrupt status register */
+		reg = readl(dp->reg_base + EXYNOS_DP_COMMON_INT_STA_4);
 
-	if (reg & PLUG)
-		return DP_IRQ_TYPE_HP_CABLE_IN;
+		if (reg & PLUG)
+			return DP_IRQ_TYPE_HP_CABLE_IN;
 
-	if (reg & HPD_LOST)
-		return DP_IRQ_TYPE_HP_CABLE_OUT;
+		if (reg & HPD_LOST)
+			return DP_IRQ_TYPE_HP_CABLE_OUT;
 
-	if (reg & HOTPLUG_CHG)
-		return DP_IRQ_TYPE_HP_CHANGE;
+		if (reg & HOTPLUG_CHG)
+			return DP_IRQ_TYPE_HP_CHANGE;
 
-	return DP_IRQ_TYPE_UNKNOWN;
+		return DP_IRQ_TYPE_UNKNOWN;
+	}
 }
 
 void exynos_dp_reset_aux(struct exynos_dp_device *dp)
@@ -386,7 +401,7 @@ void exynos_dp_init_aux(struct exynos_dp_device *dp)
 	/* Disable AUX transaction H/W retry */
 	reg = AUX_BIT_PERIOD_EXPECTED_DELAY(3) | AUX_HW_RETRY_COUNT_SEL(0)|
 		AUX_HW_RETRY_INTERVAL_600_MICROSECONDS;
-	writel(reg, dp->reg_base + EXYNOS_DP_AUX_HW_RETRY_CTL) ;
+	writel(reg, dp->reg_base + EXYNOS_DP_AUX_HW_RETRY_CTL);
 
 	/* Receive AUX Channel DEFER commands equal to DEFFER_COUNT*64 */
 	reg = DEFER_CTRL_EN | DEFER_COUNT(1);
@@ -402,9 +417,14 @@ int exynos_dp_get_plug_in_status(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
-	reg = readl(dp->reg_base + EXYNOS_DP_SYS_CTL_3);
-	if (reg & HPD_STATUS)
-		return 0;
+	if (gpio_is_valid(dp->hpd_gpio)) {
+		if (gpio_get_value(dp->hpd_gpio))
+			return 0;
+	} else {
+		reg = readl(dp->reg_base + EXYNOS_DP_SYS_CTL_3);
+		if (reg & HPD_STATUS)
+			return 0;
+	}
 
 	return -EINVAL;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_core.c b/drivers/gpu/drm/exynos/exynos_drm_core.c
index 0e9e06ce36b8..4c9f972eaa07 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_core.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_core.c
@@ -19,21 +19,19 @@
 #include "exynos_drm_fbdev.h"
 
 static LIST_HEAD(exynos_drm_subdrv_list);
-static LIST_HEAD(exynos_drm_manager_list);
-static LIST_HEAD(exynos_drm_display_list);
 
-static int exynos_drm_create_enc_conn(struct drm_device *dev,
+int exynos_drm_create_enc_conn(struct drm_device *dev,
 					struct exynos_drm_display *display)
 {
 	struct drm_encoder *encoder;
-	struct exynos_drm_manager *manager;
 	int ret;
 	unsigned long possible_crtcs = 0;
 
-	/* Find possible crtcs for this display */
-	list_for_each_entry(manager, &exynos_drm_manager_list, list)
-		if (manager->type == display->type)
-			possible_crtcs |= 1 << manager->pipe;
+	ret = exynos_drm_crtc_get_pipe_from_type(dev, display->type);
+	if (ret < 0)
+		return ret;
+
+	possible_crtcs |= 1 << ret;
 
 	/* create and initialize a encoder for this sub driver. */
 	encoder = exynos_drm_encoder_create(dev, display, possible_crtcs);
@@ -57,127 +55,29 @@ err_destroy_encoder:
 	return ret;
 }
 
-static int exynos_drm_subdrv_probe(struct drm_device *dev,
-					struct exynos_drm_subdrv *subdrv)
-{
-	if (subdrv->probe) {
-		int ret;
-
-		subdrv->drm_dev = dev;
-
-		/*
-		 * this probe callback would be called by sub driver
-		 * after setting of all resources to this sub driver,
-		 * such as clock, irq and register map are done or by load()
-		 * of exynos drm driver.
-		 *
-		 * P.S. note that this driver is considered for modularization.
-		 */
-		ret = subdrv->probe(dev, subdrv->dev);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-static void exynos_drm_subdrv_remove(struct drm_device *dev,
-				      struct exynos_drm_subdrv *subdrv)
-{
-	if (subdrv->remove)
-		subdrv->remove(dev, subdrv->dev);
-}
-
-int exynos_drm_initialize_managers(struct drm_device *dev)
+int exynos_drm_subdrv_register(struct exynos_drm_subdrv *subdrv)
 {
-	struct exynos_drm_manager *manager, *n;
-	int ret, pipe = 0;
-
-	list_for_each_entry(manager, &exynos_drm_manager_list, list) {
-		if (manager->ops->initialize) {
-			ret = manager->ops->initialize(manager, dev, pipe);
-			if (ret) {
-				DRM_ERROR("Mgr init [%d] failed with %d\n",
-						manager->type, ret);
-				goto err;
-			}
-		}
+	if (!subdrv)
+		return -EINVAL;
 
-		manager->drm_dev = dev;
-		manager->pipe = pipe++;
+	list_add_tail(&subdrv->list, &exynos_drm_subdrv_list);
 
-		ret = exynos_drm_crtc_create(manager);
-		if (ret) {
-			DRM_ERROR("CRTC create [%d] failed with %d\n",
-					manager->type, ret);
-			goto err;
-		}
-	}
 	return 0;
-
-err:
-	list_for_each_entry_safe(manager, n, &exynos_drm_manager_list, list) {
-		if (pipe-- > 0)
-			exynos_drm_manager_unregister(manager);
-		else
-			list_del(&manager->list);
-	}
-	return ret;
-}
-
-void exynos_drm_remove_managers(struct drm_device *dev)
-{
-	struct exynos_drm_manager *manager, *n;
-
-	list_for_each_entry_safe(manager, n, &exynos_drm_manager_list, list)
-		exynos_drm_manager_unregister(manager);
 }
+EXPORT_SYMBOL_GPL(exynos_drm_subdrv_register);
 
-int exynos_drm_initialize_displays(struct drm_device *dev)
+int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv)
 {
-	struct exynos_drm_display *display, *n;
-	int ret, initialized = 0;
-
-	list_for_each_entry(display, &exynos_drm_display_list, list) {
-		if (display->ops->initialize) {
-			ret = display->ops->initialize(display, dev);
-			if (ret) {
-				DRM_ERROR("Display init [%d] failed with %d\n",
-						display->type, ret);
-				goto err;
-			}
-		}
+	if (!subdrv)
+		return -EINVAL;
 
-		initialized++;
+	list_del(&subdrv->list);
 
-		ret = exynos_drm_create_enc_conn(dev, display);
-		if (ret) {
-			DRM_ERROR("Encoder create [%d] failed with %d\n",
-					display->type, ret);
-			goto err;
-		}
-	}
 	return 0;
-
-err:
-	list_for_each_entry_safe(display, n, &exynos_drm_display_list, list) {
-		if (initialized-- > 0)
-			exynos_drm_display_unregister(display);
-		else
-			list_del(&display->list);
-	}
-	return ret;
-}
-
-void exynos_drm_remove_displays(struct drm_device *dev)
-{
-	struct exynos_drm_display *display, *n;
-
-	list_for_each_entry_safe(display, n, &exynos_drm_display_list, list)
-		exynos_drm_display_unregister(display);
 }
+EXPORT_SYMBOL_GPL(exynos_drm_subdrv_unregister);
 
-int exynos_drm_device_register(struct drm_device *dev)
+int exynos_drm_device_subdrv_probe(struct drm_device *dev)
 {
 	struct exynos_drm_subdrv *subdrv, *n;
 	int err;
@@ -186,19 +86,28 @@ int exynos_drm_device_register(struct drm_device *dev)
 		return -EINVAL;
 
 	list_for_each_entry_safe(subdrv, n, &exynos_drm_subdrv_list, list) {
-		err = exynos_drm_subdrv_probe(dev, subdrv);
-		if (err) {
-			DRM_DEBUG("exynos drm subdrv probe failed.\n");
-			list_del(&subdrv->list);
-			continue;
+		if (subdrv->probe) {
+			subdrv->drm_dev = dev;
+
+			/*
+			 * this probe callback would be called by sub driver
+			 * after setting of all resources to this sub driver,
+			 * such as clock, irq and register map are done.
+			 */
+			err = subdrv->probe(dev, subdrv->dev);
+			if (err) {
+				DRM_DEBUG("exynos drm subdrv probe failed.\n");
+				list_del(&subdrv->list);
+				continue;
+			}
 		}
 	}
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(exynos_drm_device_register);
+EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_probe);
 
-int exynos_drm_device_unregister(struct drm_device *dev)
+int exynos_drm_device_subdrv_remove(struct drm_device *dev)
 {
 	struct exynos_drm_subdrv *subdrv;
 
@@ -208,66 +117,13 @@ int exynos_drm_device_unregister(struct drm_device *dev)
 	}
 
 	list_for_each_entry(subdrv, &exynos_drm_subdrv_list, list) {
-		exynos_drm_subdrv_remove(dev, subdrv);
+		if (subdrv->remove)
+			subdrv->remove(dev, subdrv->dev);
 	}
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(exynos_drm_device_unregister);
-
-int exynos_drm_manager_register(struct exynos_drm_manager *manager)
-{
-	BUG_ON(!manager->ops);
-	list_add_tail(&manager->list, &exynos_drm_manager_list);
-	return 0;
-}
-
-int exynos_drm_manager_unregister(struct exynos_drm_manager *manager)
-{
-	if (manager->ops->remove)
-		manager->ops->remove(manager);
-
-	list_del(&manager->list);
-	return 0;
-}
-
-int exynos_drm_display_register(struct exynos_drm_display *display)
-{
-	BUG_ON(!display->ops);
-	list_add_tail(&display->list, &exynos_drm_display_list);
-	return 0;
-}
-
-int exynos_drm_display_unregister(struct exynos_drm_display *display)
-{
-	if (display->ops->remove)
-		display->ops->remove(display);
-
-	list_del(&display->list);
-	return 0;
-}
-
-int exynos_drm_subdrv_register(struct exynos_drm_subdrv *subdrv)
-{
-	if (!subdrv)
-		return -EINVAL;
-
-	list_add_tail(&subdrv->list, &exynos_drm_subdrv_list);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(exynos_drm_subdrv_register);
-
-int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv)
-{
-	if (!subdrv)
-		return -EINVAL;
-
-	list_del(&subdrv->list);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(exynos_drm_subdrv_unregister);
+EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_remove);
 
 int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file)
 {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 1ef5ab9c9d51..95c9435d0266 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -368,6 +368,7 @@ int exynos_drm_crtc_create(struct exynos_drm_manager *manager)
 		return -ENOMEM;
 	}
 
+	manager->crtc = &exynos_crtc->drm_crtc;
 	crtc = &exynos_crtc->drm_crtc;
 
 	private->crtc[manager->pipe] = crtc;
@@ -491,3 +492,19 @@ void exynos_drm_crtc_complete_scanout(struct drm_framebuffer *fb)
 			manager->ops->wait_for_vblank(manager);
 	}
 }
+
+int exynos_drm_crtc_get_pipe_from_type(struct drm_device *drm_dev,
+					unsigned int out_type)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &drm_dev->mode_config.crtc_list, head) {
+		struct exynos_drm_crtc *exynos_crtc;
+
+		exynos_crtc = to_exynos_crtc(crtc);
+		if (exynos_crtc->manager->type == out_type)
+			return exynos_crtc->manager->pipe;
+	}
+
+	return -EPERM;
+}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
index c27b66cc5d24..9f74b10a8a01 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
@@ -32,4 +32,8 @@ void exynos_drm_crtc_plane_commit(struct drm_crtc *crtc, int zpos);
 void exynos_drm_crtc_plane_enable(struct drm_crtc *crtc, int zpos);
 void exynos_drm_crtc_plane_disable(struct drm_crtc *crtc, int zpos);
 
+/* This function gets pipe value to crtc device matched with out_type. */
+int exynos_drm_crtc_get_pipe_from_type(struct drm_device *drm_dev,
+					unsigned int out_type);
+
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index 2b09c7c0bfcc..9e530f205ad2 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -40,20 +40,10 @@ exynos_dpi_detect(struct drm_connector *connector, bool force)
 {
 	struct exynos_dpi *ctx = connector_to_dpi(connector);
 
-	/* panels supported only by boot-loader are always connected */
-	if (!ctx->panel_node)
-		return connector_status_connected;
-
-	if (!ctx->panel) {
-		ctx->panel = of_drm_find_panel(ctx->panel_node);
-		if (ctx->panel)
-			drm_panel_attach(ctx->panel, &ctx->connector);
-	}
-
-	if (ctx->panel)
-		return connector_status_connected;
+	if (ctx->panel && !ctx->panel->connector)
+		drm_panel_attach(ctx->panel, &ctx->connector);
 
-	return connector_status_disconnected;
+	return connector_status_connected;
 }
 
 static void exynos_dpi_connector_destroy(struct drm_connector *connector)
@@ -94,12 +84,6 @@ static int exynos_dpi_get_modes(struct drm_connector *connector)
 	return 0;
 }
 
-static int exynos_dpi_mode_valid(struct drm_connector *connector,
-				 struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *
 exynos_dpi_best_encoder(struct drm_connector *connector)
 {
@@ -110,7 +94,6 @@ exynos_dpi_best_encoder(struct drm_connector *connector)
 
 static struct drm_connector_helper_funcs exynos_dpi_connector_helper_funcs = {
 	.get_modes = exynos_dpi_get_modes,
-	.mode_valid = exynos_dpi_mode_valid,
 	.best_encoder = exynos_dpi_best_encoder,
 };
 
@@ -123,10 +106,7 @@ static int exynos_dpi_create_connector(struct exynos_drm_display *display,
 
 	ctx->encoder = encoder;
 
-	if (ctx->panel_node)
-		connector->polled = DRM_CONNECTOR_POLL_CONNECT;
-	else
-		connector->polled = DRM_CONNECTOR_POLL_HPD;
+	connector->polled = DRM_CONNECTOR_POLL_HPD;
 
 	ret = drm_connector_init(encoder->dev, connector,
 				 &exynos_dpi_connector_funcs,
@@ -172,7 +152,7 @@ static void exynos_dpi_dpms(struct exynos_drm_display *display, int mode)
 		break;
 	default:
 		break;
-	};
+	}
 	ctx->dpms_mode = mode;
 }
 
@@ -294,8 +274,10 @@ static int exynos_dpi_parse_dt(struct exynos_dpi *ctx)
 			return -ENOMEM;
 
 		ret = of_get_videomode(dn, vm, 0);
-		if (ret < 0)
+		if (ret < 0) {
+			devm_kfree(dev, vm);
 			return ret;
+		}
 
 		ctx->vm = vm;
 
@@ -308,32 +290,58 @@ static int exynos_dpi_parse_dt(struct exynos_dpi *ctx)
 	return 0;
 }
 
-int exynos_dpi_probe(struct device *dev)
+struct exynos_drm_display *exynos_dpi_probe(struct device *dev)
 {
 	struct exynos_dpi *ctx;
 	int ret;
 
+	ret = exynos_drm_component_add(dev,
+					EXYNOS_DEVICE_TYPE_CONNECTOR,
+					exynos_dpi_display.type);
+	if (ret)
+		return ERR_PTR(ret);
+
 	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
-		return -ENOMEM;
+		goto err_del_component;
 
 	ctx->dev = dev;
 	exynos_dpi_display.ctx = ctx;
 	ctx->dpms_mode = DRM_MODE_DPMS_OFF;
 
 	ret = exynos_dpi_parse_dt(ctx);
-	if (ret < 0)
-		return ret;
+	if (ret < 0) {
+		devm_kfree(dev, ctx);
+		goto err_del_component;
+	}
 
-	exynos_drm_display_register(&exynos_dpi_display);
+	if (ctx->panel_node) {
+		ctx->panel = of_drm_find_panel(ctx->panel_node);
+		if (!ctx->panel) {
+			exynos_drm_component_del(dev,
+						EXYNOS_DEVICE_TYPE_CONNECTOR);
+			return ERR_PTR(-EPROBE_DEFER);
+		}
+	}
 
-	return 0;
+	return &exynos_dpi_display;
+
+err_del_component:
+	exynos_drm_component_del(dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
+
+	return NULL;
 }
 
 int exynos_dpi_remove(struct device *dev)
 {
+	struct drm_encoder *encoder = exynos_dpi_display.encoder;
+	struct exynos_dpi *ctx = exynos_dpi_display.ctx;
+
 	exynos_dpi_dpms(&exynos_dpi_display, DRM_MODE_DPMS_OFF);
-	exynos_drm_display_unregister(&exynos_dpi_display);
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&ctx->connector);
+
+	exynos_drm_component_del(dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 2d27ba23a6a8..ab7d182063c3 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -16,6 +16,7 @@
 #include <drm/drm_crtc_helper.h>
 
 #include <linux/anon_inodes.h>
+#include <linux/component.h>
 
 #include <drm/exynos_drm.h>
 
@@ -40,9 +41,19 @@
 
 #define VBLANK_OFF_DELAY	50000
 
-/* platform device pointer for eynos drm device. */
 static struct platform_device *exynos_drm_pdev;
 
+static DEFINE_MUTEX(drm_component_lock);
+static LIST_HEAD(drm_component_list);
+
+struct component_dev {
+	struct list_head list;
+	struct device *crtc_dev;
+	struct device *conn_dev;
+	enum exynos_drm_output_type out_type;
+	unsigned int dev_type_flag;
+};
+
 static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 {
 	struct exynos_drm_private *private;
@@ -73,38 +84,21 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 
 	exynos_drm_mode_config_init(dev);
 
-	ret = exynos_drm_initialize_managers(dev);
-	if (ret)
-		goto err_mode_config_cleanup;
-
 	for (nr = 0; nr < MAX_PLANE; nr++) {
 		struct drm_plane *plane;
 		unsigned long possible_crtcs = (1 << MAX_CRTC) - 1;
 
 		plane = exynos_plane_init(dev, possible_crtcs, false);
 		if (!plane)
-			goto err_manager_cleanup;
+			goto err_mode_config_cleanup;
 	}
 
-	ret = exynos_drm_initialize_displays(dev);
-	if (ret)
-		goto err_manager_cleanup;
-
 	/* init kms poll for handling hpd */
 	drm_kms_helper_poll_init(dev);
 
 	ret = drm_vblank_init(dev, MAX_CRTC);
 	if (ret)
-		goto err_display_cleanup;
-
-	/*
-	 * probe sub drivers such as display controller and hdmi driver,
-	 * that were registered at probe() of platform driver
-	 * to the sub driver and create encoder and connector for them.
-	 */
-	ret = exynos_drm_device_register(dev);
-	if (ret)
-		goto err_vblank;
+		goto err_mode_config_cleanup;
 
 	/* setup possible_clones. */
 	exynos_drm_encoder_setup(dev);
@@ -113,17 +107,25 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 
 	platform_set_drvdata(dev->platformdev, dev);
 
+	/* Try to bind all sub drivers. */
+	ret = component_bind_all(dev->dev, dev);
+	if (ret)
+		goto err_cleanup_vblank;
+
+	/* Probe non kms sub drivers and virtual display driver. */
+	ret = exynos_drm_device_subdrv_probe(dev);
+	if (ret)
+		goto err_unbind_all;
+
 	/* force connectors detection */
 	drm_helper_hpd_irq_event(dev);
 
 	return 0;
 
-err_vblank:
+err_unbind_all:
+	component_unbind_all(dev->dev, dev);
+err_cleanup_vblank:
 	drm_vblank_cleanup(dev);
-err_display_cleanup:
-	exynos_drm_remove_displays(dev);
-err_manager_cleanup:
-	exynos_drm_remove_managers(dev);
 err_mode_config_cleanup:
 	drm_mode_config_cleanup(dev);
 	drm_release_iommu_mapping(dev);
@@ -135,17 +137,17 @@ err_free_private:
 
 static int exynos_drm_unload(struct drm_device *dev)
 {
+	exynos_drm_device_subdrv_remove(dev);
+
 	exynos_drm_fbdev_fini(dev);
-	exynos_drm_device_unregister(dev);
 	drm_vblank_cleanup(dev);
 	drm_kms_helper_poll_fini(dev);
-	exynos_drm_remove_displays(dev);
-	exynos_drm_remove_managers(dev);
 	drm_mode_config_cleanup(dev);
 
 	drm_release_iommu_mapping(dev);
 	kfree(dev->dev_private);
 
+	component_unbind_all(dev->dev, dev);
 	dev->dev_private = NULL;
 
 	return 0;
@@ -183,9 +185,9 @@ static int exynos_drm_resume(struct drm_device *dev)
 		if (connector->funcs->dpms)
 			connector->funcs->dpms(connector, connector->dpms);
 	}
+	drm_modeset_unlock_all(dev);
 
 	drm_helper_resume_force_mode(dev);
-	drm_modeset_unlock_all(dev);
 
 	return 0;
 }
@@ -323,8 +325,7 @@ static const struct file_operations exynos_drm_driver_fops = {
 };
 
 static struct drm_driver exynos_drm_driver = {
-	.driver_features	= DRIVER_HAVE_IRQ | DRIVER_MODESET |
-					DRIVER_GEM | DRIVER_PRIME,
+	.driver_features	= DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME,
 	.load			= exynos_drm_load,
 	.unload			= exynos_drm_unload,
 	.suspend		= exynos_drm_suspend,
@@ -355,27 +356,6 @@ static struct drm_driver exynos_drm_driver = {
 	.minor	= DRIVER_MINOR,
 };
 
-static int exynos_drm_platform_probe(struct platform_device *pdev)
-{
-	int ret;
-
-	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
-	if (ret)
-		return ret;
-
-	pm_runtime_enable(&pdev->dev);
-	pm_runtime_get_sync(&pdev->dev);
-
-	return drm_platform_init(&exynos_drm_driver, pdev);
-}
-
-static int exynos_drm_platform_remove(struct platform_device *pdev)
-{
-	drm_put_dev(platform_get_drvdata(pdev));
-
-	return 0;
-}
-
 #ifdef CONFIG_PM_SLEEP
 static int exynos_drm_sys_suspend(struct device *dev)
 {
@@ -400,196 +380,319 @@ static int exynos_drm_sys_resume(struct device *dev)
 }
 #endif
 
-#ifdef CONFIG_PM_RUNTIME
-static int exynos_drm_runtime_suspend(struct device *dev)
+static const struct dev_pm_ops exynos_drm_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(exynos_drm_sys_suspend, exynos_drm_sys_resume)
+};
+
+int exynos_drm_component_add(struct device *dev,
+				enum exynos_drm_device_type dev_type,
+				enum exynos_drm_output_type out_type)
 {
-	struct drm_device *drm_dev = dev_get_drvdata(dev);
-	pm_message_t message;
+	struct component_dev *cdev;
 
-	if (pm_runtime_suspended(dev))
-		return 0;
+	if (dev_type != EXYNOS_DEVICE_TYPE_CRTC &&
+			dev_type != EXYNOS_DEVICE_TYPE_CONNECTOR) {
+		DRM_ERROR("invalid device type.\n");
+		return -EINVAL;
+	}
 
-	message.event = PM_EVENT_SUSPEND;
-	return exynos_drm_suspend(drm_dev, message);
+	mutex_lock(&drm_component_lock);
+
+	/*
+	 * Make sure to check if there is a component which has two device
+	 * objects, for connector and for encoder/connector.
+	 * It should make sure that crtc and encoder/connector drivers are
+	 * ready before exynos drm core binds them.
+	 */
+	list_for_each_entry(cdev, &drm_component_list, list) {
+		if (cdev->out_type == out_type) {
+			/*
+			 * If crtc and encoder/connector device objects are
+			 * added already just return.
+			 */
+			if (cdev->dev_type_flag == (EXYNOS_DEVICE_TYPE_CRTC |
+						EXYNOS_DEVICE_TYPE_CONNECTOR)) {
+				mutex_unlock(&drm_component_lock);
+				return 0;
+			}
+
+			if (dev_type == EXYNOS_DEVICE_TYPE_CRTC) {
+				cdev->crtc_dev = dev;
+				cdev->dev_type_flag |= dev_type;
+			}
+
+			if (dev_type == EXYNOS_DEVICE_TYPE_CONNECTOR) {
+				cdev->conn_dev = dev;
+				cdev->dev_type_flag |= dev_type;
+			}
+
+			mutex_unlock(&drm_component_lock);
+			return 0;
+		}
+	}
+
+	mutex_unlock(&drm_component_lock);
+
+	cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
+	if (!cdev)
+		return -ENOMEM;
+
+	if (dev_type == EXYNOS_DEVICE_TYPE_CRTC)
+		cdev->crtc_dev = dev;
+	if (dev_type == EXYNOS_DEVICE_TYPE_CONNECTOR)
+		cdev->conn_dev = dev;
+
+	cdev->out_type = out_type;
+	cdev->dev_type_flag = dev_type;
+
+	mutex_lock(&drm_component_lock);
+	list_add_tail(&cdev->list, &drm_component_list);
+	mutex_unlock(&drm_component_lock);
+
+	return 0;
 }
 
-static int exynos_drm_runtime_resume(struct device *dev)
+void exynos_drm_component_del(struct device *dev,
+				enum exynos_drm_device_type dev_type)
 {
-	struct drm_device *drm_dev = dev_get_drvdata(dev);
+	struct component_dev *cdev, *next;
 
-	if (!pm_runtime_suspended(dev))
-		return 0;
+	mutex_lock(&drm_component_lock);
 
-	return exynos_drm_resume(drm_dev);
+	list_for_each_entry_safe(cdev, next, &drm_component_list, list) {
+		if (dev_type == EXYNOS_DEVICE_TYPE_CRTC) {
+			if (cdev->crtc_dev == dev) {
+				cdev->crtc_dev = NULL;
+				cdev->dev_type_flag &= ~dev_type;
+			}
+		}
+
+		if (dev_type == EXYNOS_DEVICE_TYPE_CONNECTOR) {
+			if (cdev->conn_dev == dev) {
+				cdev->conn_dev = NULL;
+				cdev->dev_type_flag &= ~dev_type;
+			}
+		}
+
+		/*
+		 * Release cdev object only in case that both of crtc and
+		 * encoder/connector device objects are NULL.
+		 */
+		if (!cdev->crtc_dev && !cdev->conn_dev) {
+			list_del(&cdev->list);
+			kfree(cdev);
+		}
+
+		break;
+	}
+
+	mutex_unlock(&drm_component_lock);
 }
-#endif
 
-static const struct dev_pm_ops exynos_drm_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(exynos_drm_sys_suspend, exynos_drm_sys_resume)
-	SET_RUNTIME_PM_OPS(exynos_drm_runtime_suspend,
-			exynos_drm_runtime_resume, NULL)
-};
+static int compare_of(struct device *dev, void *data)
+{
+	return dev == (struct device *)data;
+}
 
-static struct platform_driver exynos_drm_platform_driver = {
-	.probe		= exynos_drm_platform_probe,
-	.remove		= exynos_drm_platform_remove,
-	.driver		= {
-		.owner	= THIS_MODULE,
-		.name	= "exynos-drm",
-		.pm	= &exynos_drm_pm_ops,
-	},
+static int exynos_drm_add_components(struct device *dev, struct master *m)
+{
+	struct component_dev *cdev;
+	unsigned int attach_cnt = 0;
+
+	mutex_lock(&drm_component_lock);
+
+	list_for_each_entry(cdev, &drm_component_list, list) {
+		int ret;
+
+		/*
+		 * Add components to master only in case that crtc and
+		 * encoder/connector device objects exist.
+		 */
+		if (!cdev->crtc_dev || !cdev->conn_dev)
+			continue;
+
+		attach_cnt++;
+
+		mutex_unlock(&drm_component_lock);
+
+		/*
+		 * fimd and dpi modules have same device object so add
+		 * only crtc device object in this case.
+		 *
+		 * TODO. if dpi module follows driver-model driver then
+		 * below codes can be removed.
+		 */
+		if (cdev->crtc_dev == cdev->conn_dev) {
+			ret = component_master_add_child(m, compare_of,
+					cdev->crtc_dev);
+			if (ret < 0)
+				return ret;
+
+			goto out_lock;
+		}
+
+		/*
+		 * Do not chage below call order.
+		 * crtc device first should be added to master because
+		 * connector/encoder need pipe number of crtc when they
+		 * are created.
+		 */
+		ret = component_master_add_child(m, compare_of, cdev->crtc_dev);
+		ret |= component_master_add_child(m, compare_of,
+							cdev->conn_dev);
+		if (ret < 0)
+			return ret;
+
+out_lock:
+		mutex_lock(&drm_component_lock);
+	}
+
+	mutex_unlock(&drm_component_lock);
+
+	return attach_cnt ? 0 : -ENODEV;
+}
+
+static int exynos_drm_bind(struct device *dev)
+{
+	return drm_platform_init(&exynos_drm_driver, to_platform_device(dev));
+}
+
+static void exynos_drm_unbind(struct device *dev)
+{
+	drm_put_dev(dev_get_drvdata(dev));
+}
+
+static const struct component_master_ops exynos_drm_ops = {
+	.add_components = exynos_drm_add_components,
+	.bind		= exynos_drm_bind,
+	.unbind		= exynos_drm_unbind,
 };
 
-static int __init exynos_drm_init(void)
+static int exynos_drm_platform_probe(struct platform_device *pdev)
 {
 	int ret;
 
+	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
+	exynos_drm_driver.num_ioctls = ARRAY_SIZE(exynos_ioctls);
+
+#ifdef CONFIG_DRM_EXYNOS_FIMD
+	ret = platform_driver_register(&fimd_driver);
+	if (ret < 0)
+		return ret;
+#endif
+
 #ifdef CONFIG_DRM_EXYNOS_DP
 	ret = platform_driver_register(&dp_driver);
 	if (ret < 0)
-		goto out_dp;
+		goto err_unregister_fimd_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_DSI
 	ret = platform_driver_register(&dsi_driver);
 	if (ret < 0)
-		goto out_dsi;
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_FIMD
-	ret = platform_driver_register(&fimd_driver);
-	if (ret < 0)
-		goto out_fimd;
+		goto err_unregister_dp_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_HDMI
-	ret = platform_driver_register(&hdmi_driver);
-	if (ret < 0)
-		goto out_hdmi;
 	ret = platform_driver_register(&mixer_driver);
 	if (ret < 0)
-		goto out_mixer;
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	ret = platform_driver_register(&vidi_driver);
+		goto err_unregister_dsi_drv;
+	ret = platform_driver_register(&hdmi_driver);
 	if (ret < 0)
-		goto out_vidi;
+		goto err_unregister_mixer_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_G2D
 	ret = platform_driver_register(&g2d_driver);
 	if (ret < 0)
-		goto out_g2d;
+		goto err_unregister_hdmi_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_FIMC
 	ret = platform_driver_register(&fimc_driver);
 	if (ret < 0)
-		goto out_fimc;
+		goto err_unregister_g2d_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_ROTATOR
 	ret = platform_driver_register(&rotator_driver);
 	if (ret < 0)
-		goto out_rotator;
+		goto err_unregister_fimc_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_GSC
 	ret = platform_driver_register(&gsc_driver);
 	if (ret < 0)
-		goto out_gsc;
+		goto err_unregister_rotator_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_IPP
 	ret = platform_driver_register(&ipp_driver);
 	if (ret < 0)
-		goto out_ipp;
+		goto err_unregister_gsc_drv;
 
 	ret = exynos_platform_device_ipp_register();
 	if (ret < 0)
-		goto out_ipp_dev;
+		goto err_unregister_ipp_drv;
 #endif
 
-	ret = platform_driver_register(&exynos_drm_platform_driver);
+	ret = component_master_add(&pdev->dev, &exynos_drm_ops);
 	if (ret < 0)
-		goto out_drm;
-
-	exynos_drm_pdev = platform_device_register_simple("exynos-drm", -1,
-				NULL, 0);
-	if (IS_ERR(exynos_drm_pdev)) {
-		ret = PTR_ERR(exynos_drm_pdev);
-		goto out;
-	}
+		DRM_DEBUG_KMS("re-tried by last sub driver probed later.\n");
 
 	return 0;
 
-out:
-	platform_driver_unregister(&exynos_drm_platform_driver);
-
-out_drm:
 #ifdef CONFIG_DRM_EXYNOS_IPP
-	exynos_platform_device_ipp_unregister();
-out_ipp_dev:
+err_unregister_ipp_drv:
 	platform_driver_unregister(&ipp_driver);
-out_ipp:
+err_unregister_gsc_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_GSC
 	platform_driver_unregister(&gsc_driver);
-out_gsc:
+err_unregister_rotator_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_ROTATOR
 	platform_driver_unregister(&rotator_driver);
-out_rotator:
+err_unregister_fimc_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_FIMC
 	platform_driver_unregister(&fimc_driver);
-out_fimc:
+err_unregister_g2d_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_G2D
 	platform_driver_unregister(&g2d_driver);
-out_g2d:
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	platform_driver_unregister(&vidi_driver);
-out_vidi:
+err_unregister_hdmi_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_HDMI
-	platform_driver_unregister(&mixer_driver);
-out_mixer:
 	platform_driver_unregister(&hdmi_driver);
-out_hdmi:
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_FIMD
-	platform_driver_unregister(&fimd_driver);
-out_fimd:
+err_unregister_mixer_drv:
+	platform_driver_unregister(&mixer_driver);
+err_unregister_dsi_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_DSI
 	platform_driver_unregister(&dsi_driver);
-out_dsi:
+err_unregister_dp_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_DP
 	platform_driver_unregister(&dp_driver);
-out_dp:
+err_unregister_fimd_drv:
+#endif
+
+#ifdef CONFIG_DRM_EXYNOS_FIMD
+	platform_driver_unregister(&fimd_driver);
 #endif
 	return ret;
 }
 
-static void __exit exynos_drm_exit(void)
+static int exynos_drm_platform_remove(struct platform_device *pdev)
 {
-	platform_device_unregister(exynos_drm_pdev);
-
-	platform_driver_unregister(&exynos_drm_platform_driver);
-
 #ifdef CONFIG_DRM_EXYNOS_IPP
 	exynos_platform_device_ipp_unregister();
 	platform_driver_unregister(&ipp_driver);
@@ -616,10 +719,6 @@ static void __exit exynos_drm_exit(void)
 	platform_driver_unregister(&hdmi_driver);
 #endif
 
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	platform_driver_unregister(&vidi_driver);
-#endif
-
 #ifdef CONFIG_DRM_EXYNOS_FIMD
 	platform_driver_unregister(&fimd_driver);
 #endif
@@ -631,6 +730,59 @@ static void __exit exynos_drm_exit(void)
 #ifdef CONFIG_DRM_EXYNOS_DP
 	platform_driver_unregister(&dp_driver);
 #endif
+	component_master_del(&pdev->dev, &exynos_drm_ops);
+	return 0;
+}
+
+static struct platform_driver exynos_drm_platform_driver = {
+	.probe	= exynos_drm_platform_probe,
+	.remove	= exynos_drm_platform_remove,
+	.driver	= {
+		.owner	= THIS_MODULE,
+		.name	= "exynos-drm",
+		.pm	= &exynos_drm_pm_ops,
+	},
+};
+
+static int exynos_drm_init(void)
+{
+	int ret;
+
+	exynos_drm_pdev = platform_device_register_simple("exynos-drm", -1,
+								NULL, 0);
+	if (IS_ERR(exynos_drm_pdev))
+		return PTR_ERR(exynos_drm_pdev);
+
+#ifdef CONFIG_DRM_EXYNOS_VIDI
+	ret = exynos_drm_probe_vidi();
+	if (ret < 0)
+		goto err_unregister_pd;
+#endif
+
+	ret = platform_driver_register(&exynos_drm_platform_driver);
+	if (ret)
+		goto err_remove_vidi;
+
+	return 0;
+
+err_remove_vidi:
+#ifdef CONFIG_DRM_EXYNOS_VIDI
+	exynos_drm_remove_vidi();
+
+err_unregister_pd:
+#endif
+	platform_device_unregister(exynos_drm_pdev);
+
+	return ret;
+}
+
+static void exynos_drm_exit(void)
+{
+	platform_driver_unregister(&exynos_drm_platform_driver);
+#ifdef CONFIG_DRM_EXYNOS_VIDI
+	exynos_drm_remove_vidi();
+#endif
+	platform_device_unregister(exynos_drm_pdev);
 }
 
 module_init(exynos_drm_init);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index ce3e6a30deaa..06cde4506278 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -42,6 +42,13 @@ struct drm_connector;
 
 extern unsigned int drm_vblank_offdelay;
 
+/* This enumerates device type. */
+enum exynos_drm_device_type {
+	EXYNOS_DEVICE_TYPE_NONE,
+	EXYNOS_DEVICE_TYPE_CRTC,
+	EXYNOS_DEVICE_TYPE_CONNECTOR,
+};
+
 /* this enumerates display type. */
 enum exynos_drm_output_type {
 	EXYNOS_DISPLAY_TYPE_NONE,
@@ -122,7 +129,6 @@ struct exynos_drm_overlay {
  * Exynos DRM Display Structure.
  *	- this structure is common to analog tv, digital tv and lcd panel.
  *
- * @initialize: initializes the display with drm_dev
  * @remove: cleans up the display for removal
  * @mode_fixup: fix mode data comparing to hw specific display mode.
  * @mode_set: convert drm_display_mode to hw specific display mode and
@@ -133,8 +139,6 @@ struct exynos_drm_overlay {
  */
 struct exynos_drm_display;
 struct exynos_drm_display_ops {
-	int (*initialize)(struct exynos_drm_display *display,
-				struct drm_device *drm_dev);
 	int (*create_connector)(struct exynos_drm_display *display,
 				struct drm_encoder *encoder);
 	void (*remove)(struct exynos_drm_display *display);
@@ -172,8 +176,6 @@ struct exynos_drm_display {
 /*
  * Exynos drm manager ops
  *
- * @initialize: initializes the manager with drm_dev
- * @remove: cleans up the manager for removal
  * @dpms: control device power.
  * @mode_fixup: fix mode data before applying it
  * @mode_set: set the given mode to the manager
@@ -189,9 +191,6 @@ struct exynos_drm_display {
  */
 struct exynos_drm_manager;
 struct exynos_drm_manager_ops {
-	int (*initialize)(struct exynos_drm_manager *mgr,
-				struct drm_device *drm_dev, int pipe);
-	void (*remove)(struct exynos_drm_manager *mgr);
 	void (*dpms)(struct exynos_drm_manager *mgr, int mode);
 	bool (*mode_fixup)(struct exynos_drm_manager *mgr,
 				const struct drm_display_mode *mode,
@@ -215,6 +214,7 @@ struct exynos_drm_manager_ops {
  * @list: the list entry for this manager
  * @type: one of EXYNOS_DISPLAY_TYPE_LCD and HDMI.
  * @drm_dev: pointer to the drm device
+ * @crtc: crtc object.
  * @pipe: the pipe number for this crtc/manager
  * @ops: pointer to callbacks for exynos drm specific functionality
  * @ctx: A pointer to the manager's implementation specific context
@@ -223,6 +223,7 @@ struct exynos_drm_manager {
 	struct list_head list;
 	enum exynos_drm_output_type type;
 	struct drm_device *drm_dev;
+	struct drm_crtc *crtc;
 	int pipe;
 	struct exynos_drm_manager_ops *ops;
 	void *ctx;
@@ -254,6 +255,7 @@ struct drm_exynos_file_private {
  *	otherwise default one.
  * @da_space_size: size of device address space.
  *	if 0 then default value is used for it.
+ * @pipe: the pipe number for this crtc/manager.
  */
 struct exynos_drm_private {
 	struct drm_fb_helper *fb_helper;
@@ -271,6 +273,8 @@ struct exynos_drm_private {
 
 	unsigned long da_start;
 	unsigned long da_space_size;
+
+	unsigned int pipe;
 };
 
 /*
@@ -281,11 +285,11 @@ struct exynos_drm_private {
  * @drm_dev: pointer to drm_device and this pointer would be set
  *	when sub driver calls exynos_drm_subdrv_register().
  * @manager: subdrv has its own manager to control a hardware appropriately
- *	and we can access a hardware drawing on this manager.
+ *     and we can access a hardware drawing on this manager.
  * @probe: this callback would be called by exynos drm driver after
- *	subdrv is registered to it.
+ *     subdrv is registered to it.
  * @remove: this callback is used to release resources created
- *	by probe callback.
+ *     by probe callback.
  * @open: this would be called with drm device file open.
  * @close: this would be called with drm device file close.
  */
@@ -302,39 +306,14 @@ struct exynos_drm_subdrv {
 			struct drm_file *file);
 };
 
-/*
- * this function calls a probe callback registered to sub driver list and
- * create its own encoder and connector and then set drm_device object
- * to global one.
- */
-int exynos_drm_device_register(struct drm_device *dev);
-/*
- * this function calls a remove callback registered to sub driver list and
- * destroy its own encoder and connetor.
- */
-int exynos_drm_device_unregister(struct drm_device *dev);
-
-int exynos_drm_initialize_managers(struct drm_device *dev);
-void exynos_drm_remove_managers(struct drm_device *dev);
-int exynos_drm_initialize_displays(struct drm_device *dev);
-void exynos_drm_remove_displays(struct drm_device *dev);
-
-int exynos_drm_manager_register(struct exynos_drm_manager *manager);
-int exynos_drm_manager_unregister(struct exynos_drm_manager *manager);
-int exynos_drm_display_register(struct exynos_drm_display *display);
-int exynos_drm_display_unregister(struct exynos_drm_display *display);
-
-/*
- * this function would be called by sub drivers such as display controller
- * or hdmi driver to register this sub driver object to exynos drm driver
- * and when a sub driver is registered to exynos drm driver a probe callback
- * of the sub driver is called and creates its own encoder and connector.
- */
+ /* This function would be called by non kms drivers such as g2d and ipp. */
 int exynos_drm_subdrv_register(struct exynos_drm_subdrv *drm_subdrv);
 
 /* this function removes subdrv list from exynos drm driver */
 int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *drm_subdrv);
 
+int exynos_drm_device_subdrv_probe(struct drm_device *dev);
+int exynos_drm_device_subdrv_remove(struct drm_device *dev);
 int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file);
 void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file);
 
@@ -360,18 +339,40 @@ int exynos_platform_device_ipp_register(void);
 void exynos_platform_device_ipp_unregister(void);
 
 #ifdef CONFIG_DRM_EXYNOS_DPI
-int exynos_dpi_probe(struct device *dev);
+struct exynos_drm_display * exynos_dpi_probe(struct device *dev);
 int exynos_dpi_remove(struct device *dev);
 #else
-static inline int exynos_dpi_probe(struct device *dev) { return 0; }
+static inline struct exynos_drm_display *
+exynos_dpi_probe(struct device *dev) { return NULL; }
 static inline int exynos_dpi_remove(struct device *dev) { return 0; }
 #endif
 
+/*
+ * this function registers exynos drm vidi platform device/driver.
+ */
+int exynos_drm_probe_vidi(void);
+
+/*
+ * this function unregister exynos drm vidi platform device/driver.
+ */
+void exynos_drm_remove_vidi(void);
+
+/* This function creates a encoder and a connector, and initializes them. */
+int exynos_drm_create_enc_conn(struct drm_device *dev,
+				struct exynos_drm_display *display);
+
+int exynos_drm_component_add(struct device *dev,
+				enum exynos_drm_device_type dev_type,
+				enum exynos_drm_output_type out_type);
+
+void exynos_drm_component_del(struct device *dev,
+				enum exynos_drm_device_type dev_type);
+
+extern struct platform_driver fimd_driver;
 extern struct platform_driver dp_driver;
 extern struct platform_driver dsi_driver;
-extern struct platform_driver fimd_driver;
-extern struct platform_driver hdmi_driver;
 extern struct platform_driver mixer_driver;
+extern struct platform_driver hdmi_driver;
 extern struct platform_driver exynos_drm_common_hdmi_driver;
 extern struct platform_driver vidi_driver;
 extern struct platform_driver g2d_driver;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 4ac438187568..6302aa64f6c1 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -19,6 +19,7 @@
 #include <linux/irq.h>
 #include <linux/phy/phy.h>
 #include <linux/regulator/consumer.h>
+#include <linux/component.h>
 
 #include <video/mipi_display.h>
 #include <video/videomode.h>
@@ -1378,16 +1379,60 @@ end:
 	return ret;
 }
 
+static int exynos_dsi_bind(struct device *dev, struct device *master,
+				void *data)
+{
+	struct drm_device *drm_dev = data;
+	struct exynos_dsi *dsi;
+	int ret;
+
+	ret = exynos_drm_create_enc_conn(drm_dev, &exynos_dsi_display);
+	if (ret) {
+		DRM_ERROR("Encoder create [%d] failed with %d\n",
+				exynos_dsi_display.type, ret);
+		return ret;
+	}
+
+	dsi = exynos_dsi_display.ctx;
+
+	return mipi_dsi_host_register(&dsi->dsi_host);
+}
+
+static void exynos_dsi_unbind(struct device *dev, struct device *master,
+				void *data)
+{
+	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
+	struct drm_encoder *encoder = dsi->encoder;
+
+	exynos_dsi_dpms(&exynos_dsi_display, DRM_MODE_DPMS_OFF);
+
+	mipi_dsi_host_unregister(&dsi->dsi_host);
+
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&dsi->connector);
+}
+
+static const struct component_ops exynos_dsi_component_ops = {
+	.bind	= exynos_dsi_bind,
+	.unbind	= exynos_dsi_unbind,
+};
+
 static int exynos_dsi_probe(struct platform_device *pdev)
 {
 	struct resource *res;
 	struct exynos_dsi *dsi;
 	int ret;
 
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR,
+					exynos_dsi_display.type);
+	if (ret)
+		return ret;
+
 	dsi = devm_kzalloc(&pdev->dev, sizeof(*dsi), GFP_KERNEL);
 	if (!dsi) {
 		dev_err(&pdev->dev, "failed to allocate dsi object.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_del_component;
 	}
 
 	init_completion(&dsi->completed);
@@ -1401,7 +1446,7 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 
 	ret = exynos_dsi_parse_dt(dsi);
 	if (ret)
-		return ret;
+		goto err_del_component;
 
 	dsi->supplies[0].supply = "vddcore";
 	dsi->supplies[1].supply = "vddio";
@@ -1415,32 +1460,37 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 	dsi->pll_clk = devm_clk_get(&pdev->dev, "pll_clk");
 	if (IS_ERR(dsi->pll_clk)) {
 		dev_info(&pdev->dev, "failed to get dsi pll input clock\n");
-		return -EPROBE_DEFER;
+		ret = PTR_ERR(dsi->pll_clk);
+		goto err_del_component;
 	}
 
 	dsi->bus_clk = devm_clk_get(&pdev->dev, "bus_clk");
 	if (IS_ERR(dsi->bus_clk)) {
 		dev_info(&pdev->dev, "failed to get dsi bus clock\n");
-		return -EPROBE_DEFER;
+		ret = PTR_ERR(dsi->bus_clk);
+		goto err_del_component;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	dsi->reg_base = devm_ioremap_resource(&pdev->dev, res);
 	if (IS_ERR(dsi->reg_base)) {
 		dev_err(&pdev->dev, "failed to remap io region\n");
-		return PTR_ERR(dsi->reg_base);
+		ret = PTR_ERR(dsi->reg_base);
+		goto err_del_component;
 	}
 
 	dsi->phy = devm_phy_get(&pdev->dev, "dsim");
 	if (IS_ERR(dsi->phy)) {
 		dev_info(&pdev->dev, "failed to get dsim phy\n");
-		return -EPROBE_DEFER;
+		ret = PTR_ERR(dsi->phy);
+		goto err_del_component;
 	}
 
 	dsi->irq = platform_get_irq(pdev, 0);
 	if (dsi->irq < 0) {
 		dev_err(&pdev->dev, "failed to request dsi irq resource\n");
-		return dsi->irq;
+		ret = dsi->irq;
+		goto err_del_component;
 	}
 
 	irq_set_status_flags(dsi->irq, IRQ_NOAUTOEN);
@@ -1449,58 +1499,31 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 					dev_name(&pdev->dev), dsi);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to request dsi irq\n");
-		return ret;
+		goto err_del_component;
 	}
 
 	exynos_dsi_display.ctx = dsi;
 
 	platform_set_drvdata(pdev, &exynos_dsi_display);
-	exynos_drm_display_register(&exynos_dsi_display);
-
-	return mipi_dsi_host_register(&dsi->dsi_host);
-}
-
-static int exynos_dsi_remove(struct platform_device *pdev)
-{
-	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
-
-	exynos_dsi_dpms(&exynos_dsi_display, DRM_MODE_DPMS_OFF);
-
-	exynos_drm_display_unregister(&exynos_dsi_display);
-	mipi_dsi_host_unregister(&dsi->dsi_host);
-
-	return 0;
-}
 
-#if CONFIG_PM_SLEEP
-static int exynos_dsi_resume(struct device *dev)
-{
-	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
+	ret = component_add(&pdev->dev, &exynos_dsi_component_ops);
+	if (ret)
+		goto err_del_component;
 
-	if (dsi->state & DSIM_STATE_ENABLED) {
-		dsi->state &= ~DSIM_STATE_ENABLED;
-		exynos_dsi_enable(dsi);
-	}
+	return ret;
 
-	return 0;
+err_del_component:
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
+	return ret;
 }
 
-static int exynos_dsi_suspend(struct device *dev)
+static int exynos_dsi_remove(struct platform_device *pdev)
 {
-	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
-
-	if (dsi->state & DSIM_STATE_ENABLED) {
-		exynos_dsi_disable(dsi);
-		dsi->state |= DSIM_STATE_ENABLED;
-	}
+	component_del(&pdev->dev, &exynos_dsi_component_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 
 	return 0;
 }
-#endif
-
-static const struct dev_pm_ops exynos_dsi_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(exynos_dsi_suspend, exynos_dsi_resume)
-};
 
 static struct of_device_id exynos_dsi_of_match[] = {
 	{ .compatible = "samsung,exynos4210-mipi-dsi" },
@@ -1513,7 +1536,6 @@ struct platform_driver dsi_driver = {
 	.driver = {
 		   .name = "exynos-dsi",
 		   .owner = THIS_MODULE,
-		   .pm = &exynos_dsi_pm_ops,
 		   .of_match_table = exynos_dsi_of_match,
 	},
 };
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index addbf7536da4..d771b467cf0c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -121,16 +121,8 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 	offset = fbi->var.xoffset * (fb->bits_per_pixel >> 3);
 	offset += fbi->var.yoffset * fb->pitches[0];
 
-	dev->mode_config.fb_base = (resource_size_t)buffer->dma_addr;
 	fbi->screen_base = buffer->kvaddr + offset;
-	if (is_drm_iommu_supported(dev))
-		fbi->fix.smem_start = (unsigned long)
-			(page_to_phys(sg_page(buffer->sgt->sgl)) + offset);
-	else
-		fbi->fix.smem_start = (unsigned long)buffer->dma_addr;
-
 	fbi->screen_size = size;
-	fbi->fix.smem_len = size;
 
 	return 0;
 }
@@ -237,7 +229,7 @@ static struct drm_fb_helper_funcs exynos_drm_fb_helper_funcs = {
 	.fb_probe =	exynos_drm_fbdev_create,
 };
 
-bool exynos_drm_fbdev_is_anything_connected(struct drm_device *dev)
+static bool exynos_drm_fbdev_is_anything_connected(struct drm_device *dev)
 {
 	struct drm_connector *connector;
 	bool ret = false;
@@ -375,7 +367,5 @@ void exynos_drm_fbdev_restore_mode(struct drm_device *dev)
 	if (!private || !private->fb_helper)
 		return;
 
-	drm_modeset_lock_all(dev);
-	drm_fb_helper_restore_fbdev_mode(private->fb_helper);
-	drm_modeset_unlock_all(dev);
+	drm_fb_helper_restore_fbdev_mode_unlocked(private->fb_helper);
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index 30d76b2ff9c2..831dde9034c6 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -18,6 +18,7 @@
 #include <linux/clk.h>
 #include <linux/pm_runtime.h>
 #include <linux/of.h>
+#include <linux/spinlock.h>
 
 #include <drm/drmP.h>
 #include <drm/exynos_drm.h>
@@ -57,7 +58,6 @@
 #define FIMC_SHFACTOR	10
 #define FIMC_BUF_STOP	1
 #define FIMC_BUF_START	2
-#define FIMC_REG_SZ		32
 #define FIMC_WIDTH_ITU_709	1280
 #define FIMC_REFRESH_MAX	60
 #define FIMC_REFRESH_MIN	12
@@ -69,9 +69,6 @@
 #define get_fimc_context(dev)	platform_get_drvdata(to_platform_device(dev))
 #define get_ctx_from_ippdrv(ippdrv)	container_of(ippdrv,\
 					struct fimc_context, ippdrv);
-#define fimc_read(offset)		readl(ctx->regs + (offset))
-#define fimc_write(cfg, offset)	writel(cfg, ctx->regs + (offset))
-
 enum fimc_wb {
 	FIMC_WB_NONE,
 	FIMC_WB_A,
@@ -161,7 +158,7 @@ struct fimc_context {
 	struct exynos_drm_ippdrv	ippdrv;
 	struct resource	*regs_res;
 	void __iomem	*regs;
-	struct mutex	lock;
+	spinlock_t	lock;
 	struct clk	*clocks[FIMC_CLKS_MAX];
 	u32		clk_frequency;
 	struct regmap	*sysreg;
@@ -172,39 +169,53 @@ struct fimc_context {
 	bool	suspended;
 };
 
+static u32 fimc_read(struct fimc_context *ctx, u32 reg)
+{
+	return readl(ctx->regs + reg);
+}
+
+static void fimc_write(struct fimc_context *ctx, u32 val, u32 reg)
+{
+	writel(val, ctx->regs + reg);
+}
+
+static void fimc_set_bits(struct fimc_context *ctx, u32 reg, u32 bits)
+{
+	void __iomem *r = ctx->regs + reg;
+
+	writel(readl(r) | bits, r);
+}
+
+static void fimc_clear_bits(struct fimc_context *ctx, u32 reg, u32 bits)
+{
+	void __iomem *r = ctx->regs + reg;
+
+	writel(readl(r) & ~bits, r);
+}
+
 static void fimc_sw_reset(struct fimc_context *ctx)
 {
 	u32 cfg;
 
 	/* stop dma operation */
-	cfg = fimc_read(EXYNOS_CISTATUS);
-	if (EXYNOS_CISTATUS_GET_ENVID_STATUS(cfg)) {
-		cfg = fimc_read(EXYNOS_MSCTRL);
-		cfg &= ~EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg, EXYNOS_MSCTRL);
-	}
+	cfg = fimc_read(ctx, EXYNOS_CISTATUS);
+	if (EXYNOS_CISTATUS_GET_ENVID_STATUS(cfg))
+		fimc_clear_bits(ctx, EXYNOS_MSCTRL, EXYNOS_MSCTRL_ENVID);
 
-	cfg = fimc_read(EXYNOS_CISRCFMT);
-	cfg |= EXYNOS_CISRCFMT_ITU601_8BIT;
-	fimc_write(cfg, EXYNOS_CISRCFMT);
+	fimc_set_bits(ctx, EXYNOS_CISRCFMT, EXYNOS_CISRCFMT_ITU601_8BIT);
 
 	/* disable image capture */
-	cfg = fimc_read(EXYNOS_CIIMGCPT);
-	cfg &= ~(EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
-	fimc_write(cfg, EXYNOS_CIIMGCPT);
+	fimc_clear_bits(ctx, EXYNOS_CIIMGCPT,
+		EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
 
 	/* s/w reset */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg |= (EXYNOS_CIGCTRL_SWRST);
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_set_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_SWRST);
 
 	/* s/w reset complete */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg &= ~EXYNOS_CIGCTRL_SWRST;
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_SWRST);
 
 	/* reset sequence */
-	fimc_write(0x0, EXYNOS_CIFCNTSEQ);
+	fimc_write(ctx, 0x0, EXYNOS_CIFCNTSEQ);
 }
 
 static int fimc_set_camblk_fimd0_wb(struct fimc_context *ctx)
@@ -220,7 +231,7 @@ static void fimc_set_type_ctrl(struct fimc_context *ctx, enum fimc_wb wb)
 
 	DRM_DEBUG_KMS("wb[%d]\n", wb);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	cfg &= ~(EXYNOS_CIGCTRL_TESTPATTERN_MASK |
 		EXYNOS_CIGCTRL_SELCAM_ITU_MASK |
 		EXYNOS_CIGCTRL_SELCAM_MIPI_MASK |
@@ -246,7 +257,7 @@ static void fimc_set_type_ctrl(struct fimc_context *ctx, enum fimc_wb wb)
 		break;
 	}
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
 static void fimc_set_polarity(struct fimc_context *ctx,
@@ -259,7 +270,7 @@ static void fimc_set_polarity(struct fimc_context *ctx,
 	DRM_DEBUG_KMS("inv_href[%d]inv_hsync[%d]\n",
 		pol->inv_href, pol->inv_hsync);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	cfg &= ~(EXYNOS_CIGCTRL_INVPOLPCLK | EXYNOS_CIGCTRL_INVPOLVSYNC |
 		 EXYNOS_CIGCTRL_INVPOLHREF | EXYNOS_CIGCTRL_INVPOLHSYNC);
 
@@ -272,7 +283,7 @@ static void fimc_set_polarity(struct fimc_context *ctx,
 	if (pol->inv_hsync)
 		cfg |= EXYNOS_CIGCTRL_INVPOLHSYNC;
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
 static void fimc_handle_jpeg(struct fimc_context *ctx, bool enable)
@@ -281,70 +292,54 @@ static void fimc_handle_jpeg(struct fimc_context *ctx, bool enable)
 
 	DRM_DEBUG_KMS("enable[%d]\n", enable);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	if (enable)
 		cfg |= EXYNOS_CIGCTRL_CAM_JPEG;
 	else
 		cfg &= ~EXYNOS_CIGCTRL_CAM_JPEG;
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
-static void fimc_handle_irq(struct fimc_context *ctx, bool enable,
-		bool overflow, bool level)
+static void fimc_mask_irq(struct fimc_context *ctx, bool enable)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("enable[%d]overflow[%d]level[%d]\n",
-			enable, overflow, level);
+	DRM_DEBUG_KMS("enable[%d]\n", enable);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	if (enable) {
-		cfg &= ~(EXYNOS_CIGCTRL_IRQ_OVFEN | EXYNOS_CIGCTRL_IRQ_LEVEL);
-		cfg |= EXYNOS_CIGCTRL_IRQ_ENABLE;
-		if (overflow)
-			cfg |= EXYNOS_CIGCTRL_IRQ_OVFEN;
-		if (level)
-			cfg |= EXYNOS_CIGCTRL_IRQ_LEVEL;
+		cfg &= ~EXYNOS_CIGCTRL_IRQ_OVFEN;
+		cfg |= EXYNOS_CIGCTRL_IRQ_ENABLE | EXYNOS_CIGCTRL_IRQ_LEVEL;
 	} else
-		cfg &= ~(EXYNOS_CIGCTRL_IRQ_OVFEN | EXYNOS_CIGCTRL_IRQ_ENABLE);
-
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+		cfg &= ~EXYNOS_CIGCTRL_IRQ_ENABLE;
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
 static void fimc_clear_irq(struct fimc_context *ctx)
 {
-	u32 cfg;
-
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg |= EXYNOS_CIGCTRL_IRQ_CLR;
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_set_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_IRQ_CLR);
 }
 
 static bool fimc_check_ovf(struct fimc_context *ctx)
 {
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
-	u32 cfg, status, flag;
+	u32 status, flag;
 
-	status = fimc_read(EXYNOS_CISTATUS);
+	status = fimc_read(ctx, EXYNOS_CISTATUS);
 	flag = EXYNOS_CISTATUS_OVFIY | EXYNOS_CISTATUS_OVFICB |
 		EXYNOS_CISTATUS_OVFICR;
 
 	DRM_DEBUG_KMS("flag[0x%x]\n", flag);
 
 	if (status & flag) {
-		cfg = fimc_read(EXYNOS_CIWDOFST);
-		cfg |= (EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
+		fimc_set_bits(ctx, EXYNOS_CIWDOFST,
+			EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
 			EXYNOS_CIWDOFST_CLROVFICR);
-
-		fimc_write(cfg, EXYNOS_CIWDOFST);
-
-		cfg = fimc_read(EXYNOS_CIWDOFST);
-		cfg &= ~(EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
+		fimc_clear_bits(ctx, EXYNOS_CIWDOFST,
+			EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
 			EXYNOS_CIWDOFST_CLROVFICR);
 
-		fimc_write(cfg, EXYNOS_CIWDOFST);
-
 		dev_err(ippdrv->dev, "occurred overflow at %d, status 0x%x.\n",
 			ctx->id, status);
 		return true;
@@ -357,7 +352,7 @@ static bool fimc_check_frame_end(struct fimc_context *ctx)
 {
 	u32 cfg;
 
-	cfg = fimc_read(EXYNOS_CISTATUS);
+	cfg = fimc_read(ctx, EXYNOS_CISTATUS);
 
 	DRM_DEBUG_KMS("cfg[0x%x]\n", cfg);
 
@@ -365,7 +360,7 @@ static bool fimc_check_frame_end(struct fimc_context *ctx)
 		return false;
 
 	cfg &= ~(EXYNOS_CISTATUS_FRAMEEND);
-	fimc_write(cfg, EXYNOS_CISTATUS);
+	fimc_write(ctx, cfg, EXYNOS_CISTATUS);
 
 	return true;
 }
@@ -375,7 +370,7 @@ static int fimc_get_buf_id(struct fimc_context *ctx)
 	u32 cfg;
 	int frame_cnt, buf_id;
 
-	cfg = fimc_read(EXYNOS_CISTATUS2);
+	cfg = fimc_read(ctx, EXYNOS_CISTATUS2);
 	frame_cnt = EXYNOS_CISTATUS2_GET_FRAMECOUNT_BEFORE(cfg);
 
 	if (frame_cnt == 0)
@@ -402,13 +397,13 @@ static void fimc_handle_lastend(struct fimc_context *ctx, bool enable)
 
 	DRM_DEBUG_KMS("enable[%d]\n", enable);
 
-	cfg = fimc_read(EXYNOS_CIOCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIOCTRL);
 	if (enable)
 		cfg |= EXYNOS_CIOCTRL_LASTENDEN;
 	else
 		cfg &= ~EXYNOS_CIOCTRL_LASTENDEN;
 
-	fimc_write(cfg, EXYNOS_CIOCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIOCTRL);
 }
 
 
@@ -420,18 +415,18 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
 	/* RGB */
-	cfg = fimc_read(EXYNOS_CISCCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~EXYNOS_CISCCTRL_INRGB_FMT_RGB_MASK;
 
 	switch (fmt) {
 	case DRM_FORMAT_RGB565:
 		cfg |= EXYNOS_CISCCTRL_INRGB_FMT_RGB565;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	case DRM_FORMAT_RGB888:
 	case DRM_FORMAT_XRGB8888:
 		cfg |= EXYNOS_CISCCTRL_INRGB_FMT_RGB888;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	default:
 		/* bypass */
@@ -439,7 +434,7 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	}
 
 	/* YUV */
-	cfg = fimc_read(EXYNOS_MSCTRL);
+	cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg &= ~(EXYNOS_MSCTRL_ORDER2P_SHIFT_MASK |
 		EXYNOS_MSCTRL_C_INT_IN_2PLANE |
 		EXYNOS_MSCTRL_ORDER422_YCBYCR);
@@ -479,7 +474,7 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_MSCTRL);
+	fimc_write(ctx, cfg, EXYNOS_MSCTRL);
 
 	return 0;
 }
@@ -492,7 +487,7 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
-	cfg = fimc_read(EXYNOS_MSCTRL);
+	cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg &= ~EXYNOS_MSCTRL_INFORMAT_RGB;
 
 	switch (fmt) {
@@ -527,9 +522,9 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_MSCTRL);
+	fimc_write(ctx, cfg, EXYNOS_MSCTRL);
 
-	cfg = fimc_read(EXYNOS_CIDMAPARAM);
+	cfg = fimc_read(ctx, EXYNOS_CIDMAPARAM);
 	cfg &= ~EXYNOS_CIDMAPARAM_R_MODE_MASK;
 
 	if (fmt == DRM_FORMAT_NV12MT)
@@ -537,7 +532,7 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 	else
 		cfg |= EXYNOS_CIDMAPARAM_R_MODE_LINEAR;
 
-	fimc_write(cfg, EXYNOS_CIDMAPARAM);
+	fimc_write(ctx, cfg, EXYNOS_CIDMAPARAM);
 
 	return fimc_src_set_fmt_order(ctx, fmt);
 }
@@ -552,11 +547,11 @@ static int fimc_src_set_transf(struct device *dev,
 
 	DRM_DEBUG_KMS("degree[%d]flip[0x%x]\n", degree, flip);
 
-	cfg1 = fimc_read(EXYNOS_MSCTRL);
+	cfg1 = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg1 &= ~(EXYNOS_MSCTRL_FLIP_X_MIRROR |
 		EXYNOS_MSCTRL_FLIP_Y_MIRROR);
 
-	cfg2 = fimc_read(EXYNOS_CITRGFMT);
+	cfg2 = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg2 &= ~EXYNOS_CITRGFMT_INROT90_CLOCKWISE;
 
 	switch (degree) {
@@ -595,8 +590,8 @@ static int fimc_src_set_transf(struct device *dev,
 		return -EINVAL;
 	}
 
-	fimc_write(cfg1, EXYNOS_MSCTRL);
-	fimc_write(cfg2, EXYNOS_CITRGFMT);
+	fimc_write(ctx, cfg1, EXYNOS_MSCTRL);
+	fimc_write(ctx, cfg2, EXYNOS_CITRGFMT);
 	*swap = (cfg2 & EXYNOS_CITRGFMT_INROT90_CLOCKWISE) ? 1 : 0;
 
 	return 0;
@@ -621,17 +616,17 @@ static int fimc_set_window(struct fimc_context *ctx,
 	 * set window offset 1, 2 size
 	 * check figure 43-21 in user manual
 	 */
-	cfg = fimc_read(EXYNOS_CIWDOFST);
+	cfg = fimc_read(ctx, EXYNOS_CIWDOFST);
 	cfg &= ~(EXYNOS_CIWDOFST_WINHOROFST_MASK |
 		EXYNOS_CIWDOFST_WINVEROFST_MASK);
 	cfg |= (EXYNOS_CIWDOFST_WINHOROFST(h1) |
 		EXYNOS_CIWDOFST_WINVEROFST(v1));
 	cfg |= EXYNOS_CIWDOFST_WINOFSEN;
-	fimc_write(cfg, EXYNOS_CIWDOFST);
+	fimc_write(ctx, cfg, EXYNOS_CIWDOFST);
 
 	cfg = (EXYNOS_CIWDOFST2_WINHOROFST2(h2) |
 		EXYNOS_CIWDOFST2_WINVEROFST2(v2));
-	fimc_write(cfg, EXYNOS_CIWDOFST2);
+	fimc_write(ctx, cfg, EXYNOS_CIWDOFST2);
 
 	return 0;
 }
@@ -651,7 +646,7 @@ static int fimc_src_set_size(struct device *dev, int swap,
 	cfg = (EXYNOS_ORGISIZE_HORIZONTAL(img_sz.hsize) |
 		EXYNOS_ORGISIZE_VERTICAL(img_sz.vsize));
 
-	fimc_write(cfg, EXYNOS_ORGISIZE);
+	fimc_write(ctx, cfg, EXYNOS_ORGISIZE);
 
 	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]\n", pos->x, pos->y, pos->w, pos->h);
 
@@ -663,12 +658,12 @@ static int fimc_src_set_size(struct device *dev, int swap,
 	}
 
 	/* set input DMA image size */
-	cfg = fimc_read(EXYNOS_CIREAL_ISIZE);
+	cfg = fimc_read(ctx, EXYNOS_CIREAL_ISIZE);
 	cfg &= ~(EXYNOS_CIREAL_ISIZE_HEIGHT_MASK |
 		EXYNOS_CIREAL_ISIZE_WIDTH_MASK);
 	cfg |= (EXYNOS_CIREAL_ISIZE_WIDTH(img_pos.w) |
 		EXYNOS_CIREAL_ISIZE_HEIGHT(img_pos.h));
-	fimc_write(cfg, EXYNOS_CIREAL_ISIZE);
+	fimc_write(ctx, cfg, EXYNOS_CIREAL_ISIZE);
 
 	/*
 	 * set input FIFO image size
@@ -677,18 +672,18 @@ static int fimc_src_set_size(struct device *dev, int swap,
 	cfg = (EXYNOS_CISRCFMT_ITU601_8BIT |
 		EXYNOS_CISRCFMT_SOURCEHSIZE(img_sz.hsize) |
 		EXYNOS_CISRCFMT_SOURCEVSIZE(img_sz.vsize));
-	fimc_write(cfg, EXYNOS_CISRCFMT);
+	fimc_write(ctx, cfg, EXYNOS_CISRCFMT);
 
 	/* offset Y(RGB), Cb, Cr */
 	cfg = (EXYNOS_CIIYOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIIYOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIIYOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIIYOFF);
 	cfg = (EXYNOS_CIICBOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIICBOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIICBOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIICBOFF);
 	cfg = (EXYNOS_CIICROFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIICROFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIICROFF);
+	fimc_write(ctx, cfg, EXYNOS_CIICROFF);
 
 	return fimc_set_window(ctx, &img_pos, &img_sz);
 }
@@ -722,25 +717,25 @@ static int fimc_src_set_addr(struct device *dev,
 	switch (buf_type) {
 	case IPP_BUF_ENQUEUE:
 		config = &property->config[EXYNOS_DRM_OPS_SRC];
-		fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_Y],
+		fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_Y],
 			EXYNOS_CIIYSA(buf_id));
 
 		if (config->fmt == DRM_FORMAT_YVU420) {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIICBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIICRSA(buf_id));
 		} else {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIICBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIICRSA(buf_id));
 		}
 		break;
 	case IPP_BUF_DEQUEUE:
-		fimc_write(0x0, EXYNOS_CIIYSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIICBSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIICRSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIIYSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIICBSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIICRSA(buf_id));
 		break;
 	default:
 		/* bypass */
@@ -765,22 +760,22 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
 	/* RGB */
-	cfg = fimc_read(EXYNOS_CISCCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~EXYNOS_CISCCTRL_OUTRGB_FMT_RGB_MASK;
 
 	switch (fmt) {
 	case DRM_FORMAT_RGB565:
 		cfg |= EXYNOS_CISCCTRL_OUTRGB_FMT_RGB565;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	case DRM_FORMAT_RGB888:
 		cfg |= EXYNOS_CISCCTRL_OUTRGB_FMT_RGB888;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	case DRM_FORMAT_XRGB8888:
 		cfg |= (EXYNOS_CISCCTRL_OUTRGB_FMT_RGB888 |
 			EXYNOS_CISCCTRL_EXTRGB_EXTENSION);
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		break;
 	default:
 		/* bypass */
@@ -788,7 +783,7 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	}
 
 	/* YUV */
-	cfg = fimc_read(EXYNOS_CIOCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIOCTRL);
 	cfg &= ~(EXYNOS_CIOCTRL_ORDER2P_MASK |
 		EXYNOS_CIOCTRL_ORDER422_MASK |
 		EXYNOS_CIOCTRL_YCBCR_PLANE_MASK);
@@ -830,7 +825,7 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_CIOCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIOCTRL);
 
 	return 0;
 }
@@ -843,16 +838,16 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
-	cfg = fimc_read(EXYNOS_CIEXTEN);
+	cfg = fimc_read(ctx, EXYNOS_CIEXTEN);
 
 	if (fmt == DRM_FORMAT_AYUV) {
 		cfg |= EXYNOS_CIEXTEN_YUV444_OUT;
-		fimc_write(cfg, EXYNOS_CIEXTEN);
+		fimc_write(ctx, cfg, EXYNOS_CIEXTEN);
 	} else {
 		cfg &= ~EXYNOS_CIEXTEN_YUV444_OUT;
-		fimc_write(cfg, EXYNOS_CIEXTEN);
+		fimc_write(ctx, cfg, EXYNOS_CIEXTEN);
 
-		cfg = fimc_read(EXYNOS_CITRGFMT);
+		cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 		cfg &= ~EXYNOS_CITRGFMT_OUTFORMAT_MASK;
 
 		switch (fmt) {
@@ -885,10 +880,10 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 			return -EINVAL;
 		}
 
-		fimc_write(cfg, EXYNOS_CITRGFMT);
+		fimc_write(ctx, cfg, EXYNOS_CITRGFMT);
 	}
 
-	cfg = fimc_read(EXYNOS_CIDMAPARAM);
+	cfg = fimc_read(ctx, EXYNOS_CIDMAPARAM);
 	cfg &= ~EXYNOS_CIDMAPARAM_W_MODE_MASK;
 
 	if (fmt == DRM_FORMAT_NV12MT)
@@ -896,7 +891,7 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 	else
 		cfg |= EXYNOS_CIDMAPARAM_W_MODE_LINEAR;
 
-	fimc_write(cfg, EXYNOS_CIDMAPARAM);
+	fimc_write(ctx, cfg, EXYNOS_CIDMAPARAM);
 
 	return fimc_dst_set_fmt_order(ctx, fmt);
 }
@@ -911,7 +906,7 @@ static int fimc_dst_set_transf(struct device *dev,
 
 	DRM_DEBUG_KMS("degree[%d]flip[0x%x]\n", degree, flip);
 
-	cfg = fimc_read(EXYNOS_CITRGFMT);
+	cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg &= ~EXYNOS_CITRGFMT_FLIP_MASK;
 	cfg &= ~EXYNOS_CITRGFMT_OUTROT90_CLOCKWISE;
 
@@ -951,53 +946,23 @@ static int fimc_dst_set_transf(struct device *dev,
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_CITRGFMT);
+	fimc_write(ctx, cfg, EXYNOS_CITRGFMT);
 	*swap = (cfg & EXYNOS_CITRGFMT_OUTROT90_CLOCKWISE) ? 1 : 0;
 
 	return 0;
 }
 
-static int fimc_get_ratio_shift(u32 src, u32 dst, u32 *ratio, u32 *shift)
-{
-	DRM_DEBUG_KMS("src[%d]dst[%d]\n", src, dst);
-
-	if (src >= dst * 64) {
-		DRM_ERROR("failed to make ratio and shift.\n");
-		return -EINVAL;
-	} else if (src >= dst * 32) {
-		*ratio = 32;
-		*shift = 5;
-	} else if (src >= dst * 16) {
-		*ratio = 16;
-		*shift = 4;
-	} else if (src >= dst * 8) {
-		*ratio = 8;
-		*shift = 3;
-	} else if (src >= dst * 4) {
-		*ratio = 4;
-		*shift = 2;
-	} else if (src >= dst * 2) {
-		*ratio = 2;
-		*shift = 1;
-	} else {
-		*ratio = 1;
-		*shift = 0;
-	}
-
-	return 0;
-}
-
 static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 		struct drm_exynos_pos *src, struct drm_exynos_pos *dst)
 {
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
 	u32 cfg, cfg_ext, shfactor;
 	u32 pre_dst_width, pre_dst_height;
-	u32 pre_hratio, hfactor, pre_vratio, vfactor;
+	u32 hfactor, vfactor;
 	int ret = 0;
 	u32 src_w, src_h, dst_w, dst_h;
 
-	cfg_ext = fimc_read(EXYNOS_CITRGFMT);
+	cfg_ext = fimc_read(ctx, EXYNOS_CITRGFMT);
 	if (cfg_ext & EXYNOS_CITRGFMT_INROT90_CLOCKWISE) {
 		src_w = src->h;
 		src_h = src->w;
@@ -1014,24 +979,24 @@ static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 		dst_h = dst->h;
 	}
 
-	ret = fimc_get_ratio_shift(src_w, dst_w, &pre_hratio, &hfactor);
-	if (ret) {
+	/* fimc_ippdrv_check_property assures that dividers are not null */
+	hfactor = fls(src_w / dst_w / 2);
+	if (hfactor > FIMC_SHFACTOR / 2) {
 		dev_err(ippdrv->dev, "failed to get ratio horizontal.\n");
-		return ret;
+		return -EINVAL;
 	}
 
-	ret = fimc_get_ratio_shift(src_h, dst_h, &pre_vratio, &vfactor);
-	if (ret) {
+	vfactor = fls(src_h / dst_h / 2);
+	if (vfactor > FIMC_SHFACTOR / 2) {
 		dev_err(ippdrv->dev, "failed to get ratio vertical.\n");
-		return ret;
+		return -EINVAL;
 	}
 
-	pre_dst_width = src_w / pre_hratio;
-	pre_dst_height = src_h / pre_vratio;
+	pre_dst_width = src_w >> hfactor;
+	pre_dst_height = src_h >> vfactor;
 	DRM_DEBUG_KMS("pre_dst_width[%d]pre_dst_height[%d]\n",
 		pre_dst_width, pre_dst_height);
-	DRM_DEBUG_KMS("pre_hratio[%d]hfactor[%d]pre_vratio[%d]vfactor[%d]\n",
-		pre_hratio, hfactor, pre_vratio, vfactor);
+	DRM_DEBUG_KMS("hfactor[%d]vfactor[%d]\n", hfactor, vfactor);
 
 	sc->hratio = (src_w << 14) / (dst_w << hfactor);
 	sc->vratio = (src_h << 14) / (dst_h << vfactor);
@@ -1044,13 +1009,13 @@ static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 	DRM_DEBUG_KMS("shfactor[%d]\n", shfactor);
 
 	cfg = (EXYNOS_CISCPRERATIO_SHFACTOR(shfactor) |
-		EXYNOS_CISCPRERATIO_PREHORRATIO(pre_hratio) |
-		EXYNOS_CISCPRERATIO_PREVERRATIO(pre_vratio));
-	fimc_write(cfg, EXYNOS_CISCPRERATIO);
+		EXYNOS_CISCPRERATIO_PREHORRATIO(1 << hfactor) |
+		EXYNOS_CISCPRERATIO_PREVERRATIO(1 << vfactor));
+	fimc_write(ctx, cfg, EXYNOS_CISCPRERATIO);
 
 	cfg = (EXYNOS_CISCPREDST_PREDSTWIDTH(pre_dst_width) |
 		EXYNOS_CISCPREDST_PREDSTHEIGHT(pre_dst_height));
-	fimc_write(cfg, EXYNOS_CISCPREDST);
+	fimc_write(ctx, cfg, EXYNOS_CISCPREDST);
 
 	return ret;
 }
@@ -1064,7 +1029,7 @@ static void fimc_set_scaler(struct fimc_context *ctx, struct fimc_scaler *sc)
 	DRM_DEBUG_KMS("hratio[%d]vratio[%d]\n",
 		sc->hratio, sc->vratio);
 
-	cfg = fimc_read(EXYNOS_CISCCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~(EXYNOS_CISCCTRL_SCALERBYPASS |
 		EXYNOS_CISCCTRL_SCALEUP_H | EXYNOS_CISCCTRL_SCALEUP_V |
 		EXYNOS_CISCCTRL_MAIN_V_RATIO_MASK |
@@ -1084,14 +1049,14 @@ static void fimc_set_scaler(struct fimc_context *ctx, struct fimc_scaler *sc)
 
 	cfg |= (EXYNOS_CISCCTRL_MAINHORRATIO((sc->hratio >> 6)) |
 		EXYNOS_CISCCTRL_MAINVERRATIO((sc->vratio >> 6)));
-	fimc_write(cfg, EXYNOS_CISCCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 
-	cfg_ext = fimc_read(EXYNOS_CIEXTEN);
+	cfg_ext = fimc_read(ctx, EXYNOS_CIEXTEN);
 	cfg_ext &= ~EXYNOS_CIEXTEN_MAINHORRATIO_EXT_MASK;
 	cfg_ext &= ~EXYNOS_CIEXTEN_MAINVERRATIO_EXT_MASK;
 	cfg_ext |= (EXYNOS_CIEXTEN_MAINHORRATIO_EXT(sc->hratio) |
 		EXYNOS_CIEXTEN_MAINVERRATIO_EXT(sc->vratio));
-	fimc_write(cfg_ext, EXYNOS_CIEXTEN);
+	fimc_write(ctx, cfg_ext, EXYNOS_CIEXTEN);
 }
 
 static int fimc_dst_set_size(struct device *dev, int swap,
@@ -1109,12 +1074,12 @@ static int fimc_dst_set_size(struct device *dev, int swap,
 	cfg = (EXYNOS_ORGOSIZE_HORIZONTAL(img_sz.hsize) |
 		EXYNOS_ORGOSIZE_VERTICAL(img_sz.vsize));
 
-	fimc_write(cfg, EXYNOS_ORGOSIZE);
+	fimc_write(ctx, cfg, EXYNOS_ORGOSIZE);
 
 	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]\n", pos->x, pos->y, pos->w, pos->h);
 
 	/* CSC ITU */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	cfg &= ~EXYNOS_CIGCTRL_CSC_MASK;
 
 	if (sz->hsize >= FIMC_WIDTH_ITU_709)
@@ -1122,7 +1087,7 @@ static int fimc_dst_set_size(struct device *dev, int swap,
 	else
 		cfg |= EXYNOS_CIGCTRL_CSC_ITU601;
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 
 	if (swap) {
 		img_pos.w = pos->h;
@@ -1132,41 +1097,38 @@ static int fimc_dst_set_size(struct device *dev, int swap,
 	}
 
 	/* target image size */
-	cfg = fimc_read(EXYNOS_CITRGFMT);
+	cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg &= ~(EXYNOS_CITRGFMT_TARGETH_MASK |
 		EXYNOS_CITRGFMT_TARGETV_MASK);
 	cfg |= (EXYNOS_CITRGFMT_TARGETHSIZE(img_pos.w) |
 		EXYNOS_CITRGFMT_TARGETVSIZE(img_pos.h));
-	fimc_write(cfg, EXYNOS_CITRGFMT);
+	fimc_write(ctx, cfg, EXYNOS_CITRGFMT);
 
 	/* target area */
 	cfg = EXYNOS_CITAREA_TARGET_AREA(img_pos.w * img_pos.h);
-	fimc_write(cfg, EXYNOS_CITAREA);
+	fimc_write(ctx, cfg, EXYNOS_CITAREA);
 
 	/* offset Y(RGB), Cb, Cr */
 	cfg = (EXYNOS_CIOYOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIOYOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIOYOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIOYOFF);
 	cfg = (EXYNOS_CIOCBOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIOCBOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIOCBOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIOCBOFF);
 	cfg = (EXYNOS_CIOCROFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIOCROFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIOCROFF);
+	fimc_write(ctx, cfg, EXYNOS_CIOCROFF);
 
 	return 0;
 }
 
-static int fimc_dst_get_buf_seq(struct fimc_context *ctx)
+static int fimc_dst_get_buf_count(struct fimc_context *ctx)
 {
-	u32 cfg, i, buf_num = 0;
-	u32 mask = 0x00000001;
+	u32 cfg, buf_num;
 
-	cfg = fimc_read(EXYNOS_CIFCNTSEQ);
+	cfg = fimc_read(ctx, EXYNOS_CIFCNTSEQ);
 
-	for (i = 0; i < FIMC_REG_SZ; i++)
-		if (cfg & (mask << i))
-			buf_num++;
+	buf_num = hweight32(cfg);
 
 	DRM_DEBUG_KMS("buf_num[%d]\n", buf_num);
 
@@ -1181,13 +1143,14 @@ static int fimc_dst_set_buf_seq(struct fimc_context *ctx, u32 buf_id,
 	u32 cfg;
 	u32 mask = 0x00000001 << buf_id;
 	int ret = 0;
+	unsigned long flags;
 
 	DRM_DEBUG_KMS("buf_id[%d]buf_type[%d]\n", buf_id, buf_type);
 
-	mutex_lock(&ctx->lock);
+	spin_lock_irqsave(&ctx->lock, flags);
 
 	/* mask register set */
-	cfg = fimc_read(EXYNOS_CIFCNTSEQ);
+	cfg = fimc_read(ctx, EXYNOS_CIFCNTSEQ);
 
 	switch (buf_type) {
 	case IPP_BUF_ENQUEUE:
@@ -1205,20 +1168,20 @@ static int fimc_dst_set_buf_seq(struct fimc_context *ctx, u32 buf_id,
 	/* sequence id */
 	cfg &= ~mask;
 	cfg |= (enable << buf_id);
-	fimc_write(cfg, EXYNOS_CIFCNTSEQ);
+	fimc_write(ctx, cfg, EXYNOS_CIFCNTSEQ);
 
 	/* interrupt enable */
 	if (buf_type == IPP_BUF_ENQUEUE &&
-	    fimc_dst_get_buf_seq(ctx) >= FIMC_BUF_START)
-		fimc_handle_irq(ctx, true, false, true);
+	    fimc_dst_get_buf_count(ctx) >= FIMC_BUF_START)
+		fimc_mask_irq(ctx, true);
 
 	/* interrupt disable */
 	if (buf_type == IPP_BUF_DEQUEUE &&
-	    fimc_dst_get_buf_seq(ctx) <= FIMC_BUF_STOP)
-		fimc_handle_irq(ctx, false, false, true);
+	    fimc_dst_get_buf_count(ctx) <= FIMC_BUF_STOP)
+		fimc_mask_irq(ctx, false);
 
 err_unlock:
-	mutex_unlock(&ctx->lock);
+	spin_unlock_irqrestore(&ctx->lock, flags);
 	return ret;
 }
 
@@ -1252,25 +1215,25 @@ static int fimc_dst_set_addr(struct device *dev,
 	case IPP_BUF_ENQUEUE:
 		config = &property->config[EXYNOS_DRM_OPS_DST];
 
-		fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_Y],
+		fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_Y],
 			EXYNOS_CIOYSA(buf_id));
 
 		if (config->fmt == DRM_FORMAT_YVU420) {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIOCBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIOCRSA(buf_id));
 		} else {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIOCBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIOCRSA(buf_id));
 		}
 		break;
 	case IPP_BUF_DEQUEUE:
-		fimc_write(0x0, EXYNOS_CIOYSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIOCBSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIOCRSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIOYSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIOCBSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIOCRSA(buf_id));
 		break;
 	default:
 		/* bypass */
@@ -1342,11 +1305,7 @@ static irqreturn_t fimc_irq_handler(int irq, void *dev_id)
 
 static int fimc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 {
-	struct drm_exynos_ipp_prop_list *prop_list;
-
-	prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL);
-	if (!prop_list)
-		return -ENOMEM;
+	struct drm_exynos_ipp_prop_list *prop_list = &ippdrv->prop_list;
 
 	prop_list->version = 1;
 	prop_list->writeback = 1;
@@ -1371,8 +1330,6 @@ static int fimc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 	prop_list->scale_min.hsize = FIMC_SCALE_MIN;
 	prop_list->scale_min.vsize = FIMC_SCALE_MIN;
 
-	ippdrv->prop_list = prop_list;
-
 	return 0;
 }
 
@@ -1395,7 +1352,7 @@ static int fimc_ippdrv_check_property(struct device *dev,
 {
 	struct fimc_context *ctx = get_fimc_context(dev);
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
-	struct drm_exynos_ipp_prop_list *pp = ippdrv->prop_list;
+	struct drm_exynos_ipp_prop_list *pp = &ippdrv->prop_list;
 	struct drm_exynos_ipp_config *config;
 	struct drm_exynos_pos *pos;
 	struct drm_exynos_sz *sz;
@@ -1508,15 +1465,15 @@ static void fimc_clear_addr(struct fimc_context *ctx)
 	int i;
 
 	for (i = 0; i < FIMC_MAX_SRC; i++) {
-		fimc_write(0, EXYNOS_CIIYSA(i));
-		fimc_write(0, EXYNOS_CIICBSA(i));
-		fimc_write(0, EXYNOS_CIICRSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIIYSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIICBSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIICRSA(i));
 	}
 
 	for (i = 0; i < FIMC_MAX_DST; i++) {
-		fimc_write(0, EXYNOS_CIOYSA(i));
-		fimc_write(0, EXYNOS_CIOCBSA(i));
-		fimc_write(0, EXYNOS_CIOCRSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIOYSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIOCBSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIOCRSA(i));
 	}
 }
 
@@ -1556,7 +1513,7 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 
 	property = &c_node->property;
 
-	fimc_handle_irq(ctx, true, false, true);
+	fimc_mask_irq(ctx, true);
 
 	for_each_ipp_ops(i) {
 		config = &property->config[i];
@@ -1582,10 +1539,10 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 		fimc_handle_lastend(ctx, false);
 
 		/* setup dma */
-		cfg0 = fimc_read(EXYNOS_MSCTRL);
+		cfg0 = fimc_read(ctx, EXYNOS_MSCTRL);
 		cfg0 &= ~EXYNOS_MSCTRL_INPUT_MASK;
 		cfg0 |= EXYNOS_MSCTRL_INPUT_MEMORY;
-		fimc_write(cfg0, EXYNOS_MSCTRL);
+		fimc_write(ctx, cfg0, EXYNOS_MSCTRL);
 		break;
 	case IPP_CMD_WB:
 		fimc_set_type_ctrl(ctx, FIMC_WB_A);
@@ -1610,41 +1567,33 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 	}
 
 	/* Reset status */
-	fimc_write(0x0, EXYNOS_CISTATUS);
+	fimc_write(ctx, 0x0, EXYNOS_CISTATUS);
 
-	cfg0 = fimc_read(EXYNOS_CIIMGCPT);
+	cfg0 = fimc_read(ctx, EXYNOS_CIIMGCPT);
 	cfg0 &= ~EXYNOS_CIIMGCPT_IMGCPTEN_SC;
 	cfg0 |= EXYNOS_CIIMGCPT_IMGCPTEN_SC;
 
 	/* Scaler */
-	cfg1 = fimc_read(EXYNOS_CISCCTRL);
+	cfg1 = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg1 &= ~EXYNOS_CISCCTRL_SCAN_MASK;
 	cfg1 |= (EXYNOS_CISCCTRL_PROGRESSIVE |
 		EXYNOS_CISCCTRL_SCALERSTART);
 
-	fimc_write(cfg1, EXYNOS_CISCCTRL);
+	fimc_write(ctx, cfg1, EXYNOS_CISCCTRL);
 
 	/* Enable image capture*/
 	cfg0 |= EXYNOS_CIIMGCPT_IMGCPTEN;
-	fimc_write(cfg0, EXYNOS_CIIMGCPT);
+	fimc_write(ctx, cfg0, EXYNOS_CIIMGCPT);
 
 	/* Disable frame end irq */
-	cfg0 = fimc_read(EXYNOS_CIGCTRL);
-	cfg0 &= ~EXYNOS_CIGCTRL_IRQ_END_DISABLE;
-	fimc_write(cfg0, EXYNOS_CIGCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_IRQ_END_DISABLE);
 
-	cfg0 = fimc_read(EXYNOS_CIOCTRL);
-	cfg0 &= ~EXYNOS_CIOCTRL_WEAVE_MASK;
-	fimc_write(cfg0, EXYNOS_CIOCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CIOCTRL, EXYNOS_CIOCTRL_WEAVE_MASK);
 
 	if (cmd == IPP_CMD_M2M) {
-		cfg0 = fimc_read(EXYNOS_MSCTRL);
-		cfg0 |= EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg0, EXYNOS_MSCTRL);
+		fimc_set_bits(ctx, EXYNOS_MSCTRL, EXYNOS_MSCTRL_ENVID);
 
-		cfg0 = fimc_read(EXYNOS_MSCTRL);
-		cfg0 |= EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg0, EXYNOS_MSCTRL);
+		fimc_set_bits(ctx, EXYNOS_MSCTRL, EXYNOS_MSCTRL_ENVID);
 	}
 
 	return 0;
@@ -1661,10 +1610,10 @@ static void fimc_ippdrv_stop(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 	switch (cmd) {
 	case IPP_CMD_M2M:
 		/* Source clear */
-		cfg = fimc_read(EXYNOS_MSCTRL);
+		cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 		cfg &= ~EXYNOS_MSCTRL_INPUT_MASK;
 		cfg &= ~EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg, EXYNOS_MSCTRL);
+		fimc_write(ctx, cfg, EXYNOS_MSCTRL);
 		break;
 	case IPP_CMD_WB:
 		exynos_drm_ippnb_send_event(IPP_SET_WRITEBACK, (void *)&set_wb);
@@ -1675,25 +1624,20 @@ static void fimc_ippdrv_stop(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 		break;
 	}
 
-	fimc_handle_irq(ctx, false, false, true);
+	fimc_mask_irq(ctx, false);
 
 	/* reset sequence */
-	fimc_write(0x0, EXYNOS_CIFCNTSEQ);
+	fimc_write(ctx, 0x0, EXYNOS_CIFCNTSEQ);
 
 	/* Scaler disable */
-	cfg = fimc_read(EXYNOS_CISCCTRL);
-	cfg &= ~EXYNOS_CISCCTRL_SCALERSTART;
-	fimc_write(cfg, EXYNOS_CISCCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CISCCTRL, EXYNOS_CISCCTRL_SCALERSTART);
 
 	/* Disable image capture */
-	cfg = fimc_read(EXYNOS_CIIMGCPT);
-	cfg &= ~(EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
-	fimc_write(cfg, EXYNOS_CIIMGCPT);
+	fimc_clear_bits(ctx, EXYNOS_CIIMGCPT,
+		EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
 
 	/* Enable frame end irq */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg |= EXYNOS_CIGCTRL_IRQ_END_DISABLE;
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_set_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_IRQ_END_DISABLE);
 }
 
 static void fimc_put_clocks(struct fimc_context *ctx)
@@ -1848,7 +1792,7 @@ static int fimc_probe(struct platform_device *pdev)
 
 	DRM_DEBUG_KMS("id[%d]ippdrv[0x%x]\n", ctx->id, (int)ippdrv);
 
-	mutex_init(&ctx->lock);
+	spin_lock_init(&ctx->lock);
 	platform_set_drvdata(pdev, ctx);
 
 	pm_runtime_set_active(dev);
@@ -1879,7 +1823,6 @@ static int fimc_remove(struct platform_device *pdev)
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
 
 	exynos_drm_ippdrv_unregister(ippdrv);
-	mutex_destroy(&ctx->lock);
 
 	fimc_put_clocks(ctx);
 	pm_runtime_set_suspended(dev);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 40fd6ccfcd6f..33161ad38201 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -19,6 +19,7 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/component.h>
 
 #include <video/of_display_timing.h>
 #include <video/of_videomode.h>
@@ -38,6 +39,7 @@
  */
 
 #define FIMD_DEFAULT_FRAMERATE 60
+#define MIN_FB_WIDTH_FOR_16WORD_BURST 128
 
 /* position control register for hardware window 0, 2 ~ 4.*/
 #define VIDOSD_A(win)		(VIDOSD_BASE + 0x00 + (win) * 16)
@@ -122,6 +124,7 @@ struct fimd_context {
 
 	struct exynos_drm_panel_info panel;
 	struct fimd_driver_data *driver_data;
+	struct exynos_drm_display *display;
 };
 
 static const struct of_device_id fimd_driver_dt_match[] = {
@@ -143,13 +146,57 @@ static inline struct fimd_driver_data *drm_fimd_get_driver_data(
 	return (struct fimd_driver_data *)of_id->data;
 }
 
+static void fimd_wait_for_vblank(struct exynos_drm_manager *mgr)
+{
+	struct fimd_context *ctx = mgr->ctx;
+
+	if (ctx->suspended)
+		return;
+
+	atomic_set(&ctx->wait_vsync_event, 1);
+
+	/*
+	 * wait for FIMD to signal VSYNC interrupt or return after
+	 * timeout which is set to 50ms (refresh rate of 20).
+	 */
+	if (!wait_event_timeout(ctx->wait_vsync_queue,
+				!atomic_read(&ctx->wait_vsync_event),
+				HZ/20))
+		DRM_DEBUG_KMS("vblank wait timed out.\n");
+}
+
+
+static void fimd_clear_channel(struct exynos_drm_manager *mgr)
+{
+	struct fimd_context *ctx = mgr->ctx;
+	int win, ch_enabled = 0;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	/* Check if any channel is enabled. */
+	for (win = 0; win < WINDOWS_NR; win++) {
+		u32 val = readl(ctx->regs + SHADOWCON);
+		if (val & SHADOWCON_CHx_ENABLE(win)) {
+			val &= ~SHADOWCON_CHx_ENABLE(win);
+			writel(val, ctx->regs + SHADOWCON);
+			ch_enabled = 1;
+		}
+	}
+
+	/* Wait for vsync, as disable channel takes effect at next vsync */
+	if (ch_enabled)
+		fimd_wait_for_vblank(mgr);
+}
+
 static int fimd_mgr_initialize(struct exynos_drm_manager *mgr,
-			struct drm_device *drm_dev, int pipe)
+			struct drm_device *drm_dev)
 {
 	struct fimd_context *ctx = mgr->ctx;
+	struct exynos_drm_private *priv;
+	priv = drm_dev->dev_private;
 
-	ctx->drm_dev = drm_dev;
-	ctx->pipe = pipe;
+	mgr->drm_dev = ctx->drm_dev = drm_dev;
+	mgr->pipe = ctx->pipe = priv->pipe++;
 
 	/*
 	 * enable drm irq mode.
@@ -169,8 +216,14 @@ static int fimd_mgr_initialize(struct exynos_drm_manager *mgr,
 	drm_dev->vblank_disable_allowed = true;
 
 	/* attach this sub driver to iommu mapping if supported. */
-	if (is_drm_iommu_supported(ctx->drm_dev))
+	if (is_drm_iommu_supported(ctx->drm_dev)) {
+		/*
+		 * If any channel is already active, iommu will throw
+		 * a PAGE FAULT when enabled. So clear any channel if enabled.
+		 */
+		fimd_clear_channel(mgr);
 		drm_iommu_attach_device(ctx->drm_dev, ctx->dev);
+	}
 
 	return 0;
 }
@@ -324,25 +377,6 @@ static void fimd_disable_vblank(struct exynos_drm_manager *mgr)
 	}
 }
 
-static void fimd_wait_for_vblank(struct exynos_drm_manager *mgr)
-{
-	struct fimd_context *ctx = mgr->ctx;
-
-	if (ctx->suspended)
-		return;
-
-	atomic_set(&ctx->wait_vsync_event, 1);
-
-	/*
-	 * wait for FIMD to signal VSYNC interrupt or return after
-	 * timeout which is set to 50ms (refresh rate of 20).
-	 */
-	if (!wait_event_timeout(ctx->wait_vsync_queue,
-				!atomic_read(&ctx->wait_vsync_event),
-				HZ/20))
-		DRM_DEBUG_KMS("vblank wait timed out.\n");
-}
-
 static void fimd_win_mode_set(struct exynos_drm_manager *mgr,
 			struct exynos_drm_overlay *overlay)
 {
@@ -446,6 +480,19 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win)
 
 	DRM_DEBUG_KMS("bpp = %d\n", win_data->bpp);
 
+	/*
+	 * In case of exynos, setting dma-burst to 16Word causes permanent
+	 * tearing for very small buffers, e.g. cursor buffer. Burst Mode
+	 * switching which is based on overlay size is not recommended as
+	 * overlay size varies alot towards the end of the screen and rapid
+	 * movement causes unstable DMA which results into iommu crash/tear.
+	 */
+
+	if (win_data->fb_width < MIN_FB_WIDTH_FOR_16WORD_BURST) {
+		val &= ~WINCONx_BURSTLEN_MASK;
+		val |= WINCONx_BURSTLEN_4WORD;
+	}
+
 	writel(val, ctx->regs + WINCON(win));
 }
 
@@ -656,19 +703,6 @@ static void fimd_win_disable(struct exynos_drm_manager *mgr, int zpos)
 	win_data->enabled = false;
 }
 
-static void fimd_clear_win(struct fimd_context *ctx, int win)
-{
-	writel(0, ctx->regs + WINCON(win));
-	writel(0, ctx->regs + VIDOSD_A(win));
-	writel(0, ctx->regs + VIDOSD_B(win));
-	writel(0, ctx->regs + VIDOSD_C(win));
-
-	if (win == 1 || win == 2)
-		writel(0, ctx->regs + VIDOSD_D(win));
-
-	fimd_shadow_protect_win(ctx, win, false);
-}
-
 static void fimd_window_suspend(struct exynos_drm_manager *mgr)
 {
 	struct fimd_context *ctx = mgr->ctx;
@@ -707,6 +741,8 @@ static void fimd_apply(struct exynos_drm_manager *mgr)
 		win_data = &ctx->win_data[i];
 		if (win_data->enabled)
 			fimd_win_commit(mgr, i);
+		else
+			fimd_win_disable(mgr, i);
 	}
 
 	fimd_commit(mgr);
@@ -803,8 +839,6 @@ static void fimd_dpms(struct exynos_drm_manager *mgr, int mode)
 }
 
 static struct exynos_drm_manager_ops fimd_manager_ops = {
-	.initialize = fimd_mgr_initialize,
-	.remove = fimd_mgr_remove,
 	.dpms = fimd_dpms,
 	.mode_fixup = fimd_mode_fixup,
 	.mode_set = fimd_mode_set,
@@ -849,20 +883,64 @@ out:
 	return IRQ_HANDLED;
 }
 
+static int fimd_bind(struct device *dev, struct device *master, void *data)
+{
+	struct fimd_context *ctx = fimd_manager.ctx;
+	struct drm_device *drm_dev = data;
+
+	fimd_mgr_initialize(&fimd_manager, drm_dev);
+	exynos_drm_crtc_create(&fimd_manager);
+	if (ctx->display)
+		exynos_drm_create_enc_conn(drm_dev, ctx->display);
+
+	return 0;
+
+}
+
+static void fimd_unbind(struct device *dev, struct device *master,
+			void *data)
+{
+	struct exynos_drm_manager *mgr = dev_get_drvdata(dev);
+	struct fimd_context *ctx = fimd_manager.ctx;
+	struct drm_crtc *crtc = mgr->crtc;
+
+	fimd_dpms(mgr, DRM_MODE_DPMS_OFF);
+
+	if (ctx->display)
+		exynos_dpi_remove(dev);
+
+	fimd_mgr_remove(mgr);
+
+	crtc->funcs->destroy(crtc);
+}
+
+static const struct component_ops fimd_component_ops = {
+	.bind	= fimd_bind,
+	.unbind = fimd_unbind,
+};
+
 static int fimd_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct fimd_context *ctx;
 	struct resource *res;
-	int win;
 	int ret = -EINVAL;
 
-	if (!dev->of_node)
-		return -ENODEV;
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC,
+					fimd_manager.type);
+	if (ret)
+		return ret;
+
+	if (!dev->of_node) {
+		ret = -ENODEV;
+		goto err_del_component;
+	}
 
 	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	if (!ctx) {
+		ret = -ENOMEM;
+		goto err_del_component;
+	}
 
 	ctx->dev = dev;
 	ctx->suspended = true;
@@ -875,32 +953,37 @@ static int fimd_probe(struct platform_device *pdev)
 	ctx->bus_clk = devm_clk_get(dev, "fimd");
 	if (IS_ERR(ctx->bus_clk)) {
 		dev_err(dev, "failed to get bus clock\n");
-		return PTR_ERR(ctx->bus_clk);
+		ret = PTR_ERR(ctx->bus_clk);
+		goto err_del_component;
 	}
 
 	ctx->lcd_clk = devm_clk_get(dev, "sclk_fimd");
 	if (IS_ERR(ctx->lcd_clk)) {
 		dev_err(dev, "failed to get lcd clock\n");
-		return PTR_ERR(ctx->lcd_clk);
+		ret = PTR_ERR(ctx->lcd_clk);
+		goto err_del_component;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 
 	ctx->regs = devm_ioremap_resource(dev, res);
-	if (IS_ERR(ctx->regs))
-		return PTR_ERR(ctx->regs);
+	if (IS_ERR(ctx->regs)) {
+		ret = PTR_ERR(ctx->regs);
+		goto err_del_component;
+	}
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_IRQ, "vsync");
 	if (!res) {
 		dev_err(dev, "irq request failed.\n");
-		return -ENXIO;
+		ret = -ENXIO;
+		goto err_del_component;
 	}
 
 	ret = devm_request_irq(dev, res->start, fimd_irq_handler,
 							0, "drm_fimd", ctx);
 	if (ret) {
 		dev_err(dev, "irq request failed.\n");
-		return ret;
+		goto err_del_component;
 	}
 
 	ctx->driver_data = drm_fimd_get_driver_data(pdev);
@@ -910,30 +993,34 @@ static int fimd_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, &fimd_manager);
 
 	fimd_manager.ctx = ctx;
-	exynos_drm_manager_register(&fimd_manager);
 
-	exynos_dpi_probe(ctx->dev);
+	ctx->display = exynos_dpi_probe(dev);
+	if (IS_ERR(ctx->display))
+		return PTR_ERR(ctx->display);
 
-	pm_runtime_enable(dev);
+	pm_runtime_enable(&pdev->dev);
 
-	for (win = 0; win < WINDOWS_NR; win++)
-		fimd_clear_win(ctx, win);
+	ret = component_add(&pdev->dev, &fimd_component_ops);
+	if (ret)
+		goto err_disable_pm_runtime;
 
-	return 0;
+	return ret;
+
+err_disable_pm_runtime:
+	pm_runtime_disable(&pdev->dev);
+
+err_del_component:
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
+	return ret;
 }
 
 static int fimd_remove(struct platform_device *pdev)
 {
-	struct exynos_drm_manager *mgr = platform_get_drvdata(pdev);
-
-	exynos_dpi_remove(&pdev->dev);
-
-	exynos_drm_manager_unregister(&fimd_manager);
-
-	fimd_dpms(mgr, DRM_MODE_DPMS_OFF);
-
 	pm_runtime_disable(&pdev->dev);
 
+	component_del(&pdev->dev, &fimd_component_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index 42d2904d88c7..163a054922cb 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -612,22 +612,20 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 	args->pitch = args->width * ((args->bpp + 7) / 8);
 	args->size = args->pitch * args->height;
 
-	exynos_gem_obj = exynos_drm_gem_create(dev, EXYNOS_BO_CONTIG |
-						EXYNOS_BO_WC, args->size);
-	/*
-	 * If physically contiguous memory allocation fails and if IOMMU is
-	 * supported then try to get buffer from non physically contiguous
-	 * memory area.
-	 */
-	if (IS_ERR(exynos_gem_obj) && is_drm_iommu_supported(dev)) {
-		dev_warn(dev->dev, "contiguous FB allocation failed, falling back to non-contiguous\n");
+	if (is_drm_iommu_supported(dev)) {
+		exynos_gem_obj = exynos_drm_gem_create(dev,
+			EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC,
+			args->size);
+	} else {
 		exynos_gem_obj = exynos_drm_gem_create(dev,
-					EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC,
-					args->size);
+			EXYNOS_BO_CONTIG | EXYNOS_BO_WC,
+			args->size);
 	}
 
-	if (IS_ERR(exynos_gem_obj))
+	if (IS_ERR(exynos_gem_obj)) {
+		dev_warn(dev->dev, "FB allocation failed.\n");
 		return PTR_ERR(exynos_gem_obj);
+	}
 
 	ret = exynos_drm_gem_handle_create(&exynos_gem_obj->base, file_priv,
 			&args->handle);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index fa75059a6104..9e3ff1672965 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -1335,11 +1335,7 @@ static irqreturn_t gsc_irq_handler(int irq, void *dev_id)
 
 static int gsc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 {
-	struct drm_exynos_ipp_prop_list *prop_list;
-
-	prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL);
-	if (!prop_list)
-		return -ENOMEM;
+	struct drm_exynos_ipp_prop_list *prop_list = &ippdrv->prop_list;
 
 	prop_list->version = 1;
 	prop_list->writeback = 1;
@@ -1363,8 +1359,6 @@ static int gsc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 	prop_list->scale_min.hsize = GSC_SCALE_MIN;
 	prop_list->scale_min.vsize = GSC_SCALE_MIN;
 
-	ippdrv->prop_list = prop_list;
-
 	return 0;
 }
 
@@ -1387,7 +1381,7 @@ static int gsc_ippdrv_check_property(struct device *dev,
 {
 	struct gsc_context *ctx = get_gsc_context(dev);
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
-	struct drm_exynos_ipp_prop_list *pp = ippdrv->prop_list;
+	struct drm_exynos_ipp_prop_list *pp = &ippdrv->prop_list;
 	struct drm_exynos_ipp_config *config;
 	struct drm_exynos_pos *pos;
 	struct drm_exynos_sz *sz;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
index 3d78144387ac..a1888e128f1d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
@@ -167,6 +167,13 @@ static int ipp_create_id(struct idr *id_idr, struct mutex *lock, void *obj,
 	return 0;
 }
 
+static void ipp_remove_id(struct idr *id_idr, struct mutex *lock, u32 id)
+{
+	mutex_lock(lock);
+	idr_remove(id_idr, id);
+	mutex_unlock(lock);
+}
+
 static void *ipp_find_obj(struct idr *id_idr, struct mutex *lock, u32 id)
 {
 	void *obj;
@@ -276,11 +283,6 @@ static struct exynos_drm_ippdrv *ipp_find_drv_by_handle(u32 prop_id)
 
 	DRM_DEBUG_KMS("prop_id[%d]\n", prop_id);
 
-	if (list_empty(&exynos_drm_ippdrv_list)) {
-		DRM_DEBUG_KMS("ippdrv_list is empty.\n");
-		return ERR_PTR(-ENODEV);
-	}
-
 	/*
 	 * This case is search ipp driver by prop_id handle.
 	 * sometimes, ipp subsystem find driver by prop_id.
@@ -289,11 +291,14 @@ static struct exynos_drm_ippdrv *ipp_find_drv_by_handle(u32 prop_id)
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
 		DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]\n", count++, (int)ippdrv);
 
-		if (!list_empty(&ippdrv->cmd_list)) {
-			list_for_each_entry(c_node, &ippdrv->cmd_list, list)
-				if (c_node->property.prop_id == prop_id)
-					return ippdrv;
+		mutex_lock(&ippdrv->cmd_lock);
+		list_for_each_entry(c_node, &ippdrv->cmd_list, list) {
+			if (c_node->property.prop_id == prop_id) {
+				mutex_unlock(&ippdrv->cmd_lock);
+				return ippdrv;
+			}
 		}
+		mutex_unlock(&ippdrv->cmd_lock);
 	}
 
 	return ERR_PTR(-ENODEV);
@@ -325,6 +330,7 @@ int exynos_drm_ipp_get_property(struct drm_device *drm_dev, void *data,
 	if (!prop_list->ipp_id) {
 		list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list)
 			count++;
+
 		/*
 		 * Supports ippdrv list count for user application.
 		 * First step user application getting ippdrv count.
@@ -346,7 +352,7 @@ int exynos_drm_ipp_get_property(struct drm_device *drm_dev, void *data,
 			return PTR_ERR(ippdrv);
 		}
 
-		prop_list = ippdrv->prop_list;
+		*prop_list = ippdrv->prop_list;
 	}
 
 	return 0;
@@ -386,9 +392,11 @@ static int ipp_find_and_set_property(struct drm_exynos_ipp_property *property)
 	 * when we find this command no using prop_id.
 	 * return property information set in this command node.
 	 */
+	mutex_lock(&ippdrv->cmd_lock);
 	list_for_each_entry(c_node, &ippdrv->cmd_list, list) {
 		if ((c_node->property.prop_id == prop_id) &&
 		    (c_node->state == IPP_STATE_STOP)) {
+			mutex_unlock(&ippdrv->cmd_lock);
 			DRM_DEBUG_KMS("found cmd[%d]ippdrv[0x%x]\n",
 				property->cmd, (int)ippdrv);
 
@@ -396,6 +404,7 @@ static int ipp_find_and_set_property(struct drm_exynos_ipp_property *property)
 			return 0;
 		}
 	}
+	mutex_unlock(&ippdrv->cmd_lock);
 
 	DRM_ERROR("failed to search property.\n");
 
@@ -499,7 +508,7 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 	c_node->start_work = ipp_create_cmd_work();
 	if (IS_ERR(c_node->start_work)) {
 		DRM_ERROR("failed to create start work.\n");
-		goto err_clear;
+		goto err_remove_id;
 	}
 
 	c_node->stop_work = ipp_create_cmd_work();
@@ -514,7 +523,7 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 		goto err_free_stop;
 	}
 
-	mutex_init(&c_node->cmd_lock);
+	mutex_init(&c_node->lock);
 	mutex_init(&c_node->mem_lock);
 	mutex_init(&c_node->event_lock);
 
@@ -526,7 +535,9 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 
 	INIT_LIST_HEAD(&c_node->event_list);
 	list_splice_init(&priv->event_list, &c_node->event_list);
+	mutex_lock(&ippdrv->cmd_lock);
 	list_add_tail(&c_node->list, &ippdrv->cmd_list);
+	mutex_unlock(&ippdrv->cmd_lock);
 
 	/* make dedicated state without m2m */
 	if (!ipp_is_m2m_cmd(property->cmd))
@@ -538,18 +549,24 @@ err_free_stop:
 	kfree(c_node->stop_work);
 err_free_start:
 	kfree(c_node->start_work);
+err_remove_id:
+	ipp_remove_id(&ctx->prop_idr, &ctx->prop_lock, property->prop_id);
 err_clear:
 	kfree(c_node);
 	return ret;
 }
 
-static void ipp_clean_cmd_node(struct drm_exynos_ipp_cmd_node *c_node)
+static void ipp_clean_cmd_node(struct ipp_context *ctx,
+				struct drm_exynos_ipp_cmd_node *c_node)
 {
 	/* delete list */
 	list_del(&c_node->list);
 
+	ipp_remove_id(&ctx->prop_idr, &ctx->prop_lock,
+			c_node->property.prop_id);
+
 	/* destroy mutex */
-	mutex_destroy(&c_node->cmd_lock);
+	mutex_destroy(&c_node->lock);
 	mutex_destroy(&c_node->mem_lock);
 	mutex_destroy(&c_node->event_lock);
 
@@ -567,17 +584,10 @@ static int ipp_check_mem_list(struct drm_exynos_ipp_cmd_node *c_node)
 	struct list_head *head;
 	int ret, i, count[EXYNOS_DRM_OPS_MAX] = { 0, };
 
-	mutex_lock(&c_node->mem_lock);
-
 	for_each_ipp_ops(i) {
 		/* source/destination memory list */
 		head = &c_node->mem_list[i];
 
-		if (list_empty(head)) {
-			DRM_DEBUG_KMS("%s memory empty.\n", i ? "dst" : "src");
-			continue;
-		}
-
 		/* find memory node entry */
 		list_for_each_entry(m_node, head, list) {
 			DRM_DEBUG_KMS("%s,count[%d]m_node[0x%x]\n",
@@ -602,8 +612,6 @@ static int ipp_check_mem_list(struct drm_exynos_ipp_cmd_node *c_node)
 		ret = max(count[EXYNOS_DRM_OPS_SRC],
 			count[EXYNOS_DRM_OPS_DST]);
 
-	mutex_unlock(&c_node->mem_lock);
-
 	return ret;
 }
 
@@ -646,16 +654,13 @@ static int ipp_set_mem_node(struct exynos_drm_ippdrv *ippdrv,
 		return -EFAULT;
 	}
 
-	mutex_lock(&c_node->mem_lock);
-
 	DRM_DEBUG_KMS("ops_id[%d]\n", m_node->ops_id);
 
 	/* get operations callback */
 	ops = ippdrv->ops[m_node->ops_id];
 	if (!ops) {
 		DRM_ERROR("not support ops.\n");
-		ret = -EFAULT;
-		goto err_unlock;
+		return -EFAULT;
 	}
 
 	/* set address and enable irq */
@@ -664,12 +669,10 @@ static int ipp_set_mem_node(struct exynos_drm_ippdrv *ippdrv,
 			m_node->buf_id, IPP_BUF_ENQUEUE);
 		if (ret) {
 			DRM_ERROR("failed to set addr.\n");
-			goto err_unlock;
+			return ret;
 		}
 	}
 
-err_unlock:
-	mutex_unlock(&c_node->mem_lock);
 	return ret;
 }
 
@@ -684,11 +687,9 @@ static struct drm_exynos_ipp_mem_node
 	void *addr;
 	int i;
 
-	mutex_lock(&c_node->mem_lock);
-
 	m_node = kzalloc(sizeof(*m_node), GFP_KERNEL);
 	if (!m_node)
-		goto err_unlock;
+		return ERR_PTR(-ENOMEM);
 
 	/* clear base address for error handling */
 	memset(&buf_info, 0x0, sizeof(buf_info));
@@ -722,15 +723,14 @@ static struct drm_exynos_ipp_mem_node
 
 	m_node->filp = file;
 	m_node->buf_info = buf_info;
+	mutex_lock(&c_node->mem_lock);
 	list_add_tail(&m_node->list, &c_node->mem_list[qbuf->ops_id]);
-
 	mutex_unlock(&c_node->mem_lock);
+
 	return m_node;
 
 err_clear:
 	kfree(m_node);
-err_unlock:
-	mutex_unlock(&c_node->mem_lock);
 	return ERR_PTR(-EFAULT);
 }
 
@@ -747,13 +747,6 @@ static int ipp_put_mem_node(struct drm_device *drm_dev,
 		return -EFAULT;
 	}
 
-	if (list_empty(&m_node->list)) {
-		DRM_ERROR("empty memory node.\n");
-		return -ENOMEM;
-	}
-
-	mutex_lock(&c_node->mem_lock);
-
 	DRM_DEBUG_KMS("ops_id[%d]\n", m_node->ops_id);
 
 	/* put gem buffer */
@@ -768,8 +761,6 @@ static int ipp_put_mem_node(struct drm_device *drm_dev,
 	list_del(&m_node->list);
 	kfree(m_node);
 
-	mutex_unlock(&c_node->mem_lock);
-
 	return 0;
 }
 
@@ -805,7 +796,9 @@ static int ipp_get_event(struct drm_device *drm_dev,
 	e->base.event = &e->event.base;
 	e->base.file_priv = file;
 	e->base.destroy = ipp_free_event;
+	mutex_lock(&c_node->event_lock);
 	list_add_tail(&e->base.link, &c_node->event_list);
+	mutex_unlock(&c_node->event_lock);
 
 	return 0;
 }
@@ -816,11 +809,7 @@ static void ipp_put_event(struct drm_exynos_ipp_cmd_node *c_node,
 	struct drm_exynos_ipp_send_event *e, *te;
 	int count = 0;
 
-	if (list_empty(&c_node->event_list)) {
-		DRM_DEBUG_KMS("event_list is empty.\n");
-		return;
-	}
-
+	mutex_lock(&c_node->event_lock);
 	list_for_each_entry_safe(e, te, &c_node->event_list, base.link) {
 		DRM_DEBUG_KMS("count[%d]e[0x%x]\n", count++, (int)e);
 
@@ -841,9 +830,13 @@ static void ipp_put_event(struct drm_exynos_ipp_cmd_node *c_node,
 			/* delete list */
 			list_del(&e->base.link);
 			kfree(e);
-			return;
+			goto out_unlock;
 		}
 	}
+
+out_unlock:
+	mutex_unlock(&c_node->event_lock);
+	return;
 }
 
 static void ipp_handle_cmd_work(struct device *dev,
@@ -887,7 +880,9 @@ static int ipp_queue_buf_with_run(struct device *dev,
 		return 0;
 	}
 
+	mutex_lock(&c_node->mem_lock);
 	if (!ipp_check_mem_list(c_node)) {
+		mutex_unlock(&c_node->mem_lock);
 		DRM_DEBUG_KMS("empty memory.\n");
 		return 0;
 	}
@@ -904,10 +899,12 @@ static int ipp_queue_buf_with_run(struct device *dev,
 	} else {
 		ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 		if (ret) {
+			mutex_unlock(&c_node->mem_lock);
 			DRM_ERROR("failed to set m node.\n");
 			return ret;
 		}
 	}
+	mutex_unlock(&c_node->mem_lock);
 
 	return 0;
 }
@@ -918,15 +915,15 @@ static void ipp_clean_queue_buf(struct drm_device *drm_dev,
 {
 	struct drm_exynos_ipp_mem_node *m_node, *tm_node;
 
-	if (!list_empty(&c_node->mem_list[qbuf->ops_id])) {
-		/* delete list */
-		list_for_each_entry_safe(m_node, tm_node,
-			&c_node->mem_list[qbuf->ops_id], list) {
-			if (m_node->buf_id == qbuf->buf_id &&
-			    m_node->ops_id == qbuf->ops_id)
-				ipp_put_mem_node(drm_dev, c_node, m_node);
-		}
+	/* delete list */
+	mutex_lock(&c_node->mem_lock);
+	list_for_each_entry_safe(m_node, tm_node,
+		&c_node->mem_list[qbuf->ops_id], list) {
+		if (m_node->buf_id == qbuf->buf_id &&
+		    m_node->ops_id == qbuf->ops_id)
+			ipp_put_mem_node(drm_dev, c_node, m_node);
 	}
+	mutex_unlock(&c_node->mem_lock);
 }
 
 int exynos_drm_ipp_queue_buf(struct drm_device *drm_dev, void *data,
@@ -998,7 +995,7 @@ int exynos_drm_ipp_queue_buf(struct drm_device *drm_dev, void *data,
 		}
 		break;
 	case IPP_BUF_DEQUEUE:
-		mutex_lock(&c_node->cmd_lock);
+		mutex_lock(&c_node->lock);
 
 		/* put event for destination buffer */
 		if (qbuf->ops_id == EXYNOS_DRM_OPS_DST)
@@ -1006,7 +1003,7 @@ int exynos_drm_ipp_queue_buf(struct drm_device *drm_dev, void *data,
 
 		ipp_clean_queue_buf(drm_dev, c_node, qbuf);
 
-		mutex_unlock(&c_node->cmd_lock);
+		mutex_unlock(&c_node->lock);
 		break;
 	default:
 		DRM_ERROR("invalid buffer control.\n");
@@ -1109,12 +1106,12 @@ int exynos_drm_ipp_cmd_ctrl(struct drm_device *drm_dev, void *data,
 	case IPP_CTRL_PLAY:
 		if (pm_runtime_suspended(ippdrv->dev))
 			pm_runtime_get_sync(ippdrv->dev);
+
 		c_node->state = IPP_STATE_START;
 
 		cmd_work = c_node->start_work;
 		cmd_work->ctrl = cmd_ctrl->ctrl;
 		ipp_handle_cmd_work(dev, ippdrv, cmd_work, c_node);
-		c_node->state = IPP_STATE_START;
 		break;
 	case IPP_CTRL_STOP:
 		cmd_work = c_node->stop_work;
@@ -1129,10 +1126,12 @@ int exynos_drm_ipp_cmd_ctrl(struct drm_device *drm_dev, void *data,
 
 		c_node->state = IPP_STATE_STOP;
 		ippdrv->dedicated = false;
-		ipp_clean_cmd_node(c_node);
+		mutex_lock(&ippdrv->cmd_lock);
+		ipp_clean_cmd_node(ctx, c_node);
 
 		if (list_empty(&ippdrv->cmd_list))
 			pm_runtime_put_sync(ippdrv->dev);
+		mutex_unlock(&ippdrv->cmd_lock);
 		break;
 	case IPP_CTRL_PAUSE:
 		cmd_work = c_node->stop_work;
@@ -1260,9 +1259,11 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 	/* store command info in ippdrv */
 	ippdrv->c_node = c_node;
 
+	mutex_lock(&c_node->mem_lock);
 	if (!ipp_check_mem_list(c_node)) {
 		DRM_DEBUG_KMS("empty memory.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_unlock;
 	}
 
 	/* set current property in ippdrv */
@@ -1270,7 +1271,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 	if (ret) {
 		DRM_ERROR("failed to set property.\n");
 		ippdrv->c_node = NULL;
-		return ret;
+		goto err_unlock;
 	}
 
 	/* check command */
@@ -1285,7 +1286,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			if (!m_node) {
 				DRM_ERROR("failed to get node.\n");
 				ret = -EFAULT;
-				return ret;
+				goto err_unlock;
 			}
 
 			DRM_DEBUG_KMS("m_node[0x%x]\n", (int)m_node);
@@ -1293,7 +1294,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
 				DRM_ERROR("failed to set m node.\n");
-				return ret;
+				goto err_unlock;
 			}
 		}
 		break;
@@ -1305,7 +1306,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
 				DRM_ERROR("failed to set m node.\n");
-				return ret;
+				goto err_unlock;
 			}
 		}
 		break;
@@ -1317,14 +1318,16 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
 				DRM_ERROR("failed to set m node.\n");
-				return ret;
+				goto err_unlock;
 			}
 		}
 		break;
 	default:
 		DRM_ERROR("invalid operations.\n");
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_unlock;
 	}
+	mutex_unlock(&c_node->mem_lock);
 
 	DRM_DEBUG_KMS("cmd[%d]\n", property->cmd);
 
@@ -1333,11 +1336,17 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 		ret = ippdrv->start(ippdrv->dev, property->cmd);
 		if (ret) {
 			DRM_ERROR("failed to start ops.\n");
+			ippdrv->c_node = NULL;
 			return ret;
 		}
 	}
 
 	return 0;
+
+err_unlock:
+	mutex_unlock(&c_node->mem_lock);
+	ippdrv->c_node = NULL;
+	return ret;
 }
 
 static int ipp_stop_property(struct drm_device *drm_dev,
@@ -1354,6 +1363,8 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 	/* put event */
 	ipp_put_event(c_node, NULL);
 
+	mutex_lock(&c_node->mem_lock);
+
 	/* check command */
 	switch (property->cmd) {
 	case IPP_CMD_M2M:
@@ -1361,11 +1372,6 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 			/* source/destination memory list */
 			head = &c_node->mem_list[i];
 
-			if (list_empty(head)) {
-				DRM_DEBUG_KMS("mem_list is empty.\n");
-				break;
-			}
-
 			list_for_each_entry_safe(m_node, tm_node,
 				head, list) {
 				ret = ipp_put_mem_node(drm_dev, c_node,
@@ -1381,11 +1387,6 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 		/* destination memory list */
 		head = &c_node->mem_list[EXYNOS_DRM_OPS_DST];
 
-		if (list_empty(head)) {
-			DRM_DEBUG_KMS("mem_list is empty.\n");
-			break;
-		}
-
 		list_for_each_entry_safe(m_node, tm_node, head, list) {
 			ret = ipp_put_mem_node(drm_dev, c_node, m_node);
 			if (ret) {
@@ -1398,11 +1399,6 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 		/* source memory list */
 		head = &c_node->mem_list[EXYNOS_DRM_OPS_SRC];
 
-		if (list_empty(head)) {
-			DRM_DEBUG_KMS("mem_list is empty.\n");
-			break;
-		}
-
 		list_for_each_entry_safe(m_node, tm_node, head, list) {
 			ret = ipp_put_mem_node(drm_dev, c_node, m_node);
 			if (ret) {
@@ -1418,6 +1414,8 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 	}
 
 err_clear:
+	mutex_unlock(&c_node->mem_lock);
+
 	/* stop operations */
 	if (ippdrv->stop)
 		ippdrv->stop(ippdrv->dev, property->cmd);
@@ -1446,7 +1444,7 @@ void ipp_sched_cmd(struct work_struct *work)
 		return;
 	}
 
-	mutex_lock(&c_node->cmd_lock);
+	mutex_lock(&c_node->lock);
 
 	property = &c_node->property;
 
@@ -1494,7 +1492,7 @@ void ipp_sched_cmd(struct work_struct *work)
 	DRM_DEBUG_KMS("ctrl[%d] done.\n", cmd_work->ctrl);
 
 err_unlock:
-	mutex_unlock(&c_node->cmd_lock);
+	mutex_unlock(&c_node->lock);
 }
 
 static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
@@ -1524,14 +1522,18 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 		return -EINVAL;
 	}
 
+	mutex_lock(&c_node->event_lock);
 	if (list_empty(&c_node->event_list)) {
 		DRM_DEBUG_KMS("event list is empty.\n");
-		return 0;
+		ret = 0;
+		goto err_event_unlock;
 	}
 
+	mutex_lock(&c_node->mem_lock);
 	if (!ipp_check_mem_list(c_node)) {
 		DRM_DEBUG_KMS("empty memory.\n");
-		return 0;
+		ret = 0;
+		goto err_mem_unlock;
 	}
 
 	/* check command */
@@ -1545,7 +1547,8 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 				struct drm_exynos_ipp_mem_node, list);
 			if (!m_node) {
 				DRM_ERROR("empty memory node.\n");
-				return -ENOMEM;
+				ret = -ENOMEM;
+				goto err_mem_unlock;
 			}
 
 			tbuf_id[i] = m_node->buf_id;
@@ -1567,7 +1570,8 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 		m_node = ipp_find_mem_node(c_node, &qbuf);
 		if (!m_node) {
 			DRM_ERROR("empty memory node.\n");
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto err_mem_unlock;
 		}
 
 		tbuf_id[EXYNOS_DRM_OPS_DST] = m_node->buf_id;
@@ -1584,7 +1588,8 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 			struct drm_exynos_ipp_mem_node, list);
 		if (!m_node) {
 			DRM_ERROR("empty memory node.\n");
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto err_mem_unlock;
 		}
 
 		tbuf_id[EXYNOS_DRM_OPS_SRC] = m_node->buf_id;
@@ -1595,8 +1600,10 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 		break;
 	default:
 		DRM_ERROR("invalid operations.\n");
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_mem_unlock;
 	}
+	mutex_unlock(&c_node->mem_lock);
 
 	if (tbuf_id[EXYNOS_DRM_OPS_DST] != buf_id[EXYNOS_DRM_OPS_DST])
 		DRM_ERROR("failed to match buf_id[%d %d]prop_id[%d]\n",
@@ -1611,11 +1618,6 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 	e = list_first_entry(&c_node->event_list,
 		struct drm_exynos_ipp_send_event, base.link);
 
-	if (!e) {
-		DRM_ERROR("empty event.\n");
-		return -EINVAL;
-	}
-
 	do_gettimeofday(&now);
 	DRM_DEBUG_KMS("tv_sec[%ld]tv_usec[%ld]\n", now.tv_sec, now.tv_usec);
 	e->event.tv_sec = now.tv_sec;
@@ -1630,11 +1632,18 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 	list_move_tail(&e->base.link, &e->base.file_priv->event_list);
 	wake_up_interruptible(&e->base.file_priv->event_wait);
 	spin_unlock_irqrestore(&drm_dev->event_lock, flags);
+	mutex_unlock(&c_node->event_lock);
 
 	DRM_DEBUG_KMS("done cmd[%d]prop_id[%d]buf_id[%d]\n",
 		property->cmd, property->prop_id, tbuf_id[EXYNOS_DRM_OPS_DST]);
 
 	return 0;
+
+err_mem_unlock:
+	mutex_unlock(&c_node->mem_lock);
+err_event_unlock:
+	mutex_unlock(&c_node->event_lock);
+	return ret;
 }
 
 void ipp_sched_event(struct work_struct *work)
@@ -1676,8 +1685,6 @@ void ipp_sched_event(struct work_struct *work)
 		goto err_completion;
 	}
 
-	mutex_lock(&c_node->event_lock);
-
 	ret = ipp_send_event(ippdrv, c_node, event_work->buf_id);
 	if (ret) {
 		DRM_ERROR("failed to send event.\n");
@@ -1687,8 +1694,6 @@ void ipp_sched_event(struct work_struct *work)
 err_completion:
 	if (ipp_is_m2m_cmd(c_node->property.cmd))
 		complete(&c_node->start_complete);
-
-	mutex_unlock(&c_node->event_lock);
 }
 
 static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
@@ -1699,23 +1704,21 @@ static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 
 	/* get ipp driver entry */
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
+		u32 ipp_id;
+
 		ippdrv->drm_dev = drm_dev;
 
 		ret = ipp_create_id(&ctx->ipp_idr, &ctx->ipp_lock, ippdrv,
-			&ippdrv->ipp_id);
-		if (ret) {
+				    &ipp_id);
+		if (ret || ipp_id == 0) {
 			DRM_ERROR("failed to create id.\n");
-			goto err_idr;
+			goto err;
 		}
 
 		DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]ipp_id[%d]\n",
-			count++, (int)ippdrv, ippdrv->ipp_id);
+			count++, (int)ippdrv, ipp_id);
 
-		if (ippdrv->ipp_id == 0) {
-			DRM_ERROR("failed to get ipp_id[%d]\n",
-				ippdrv->ipp_id);
-			goto err_idr;
-		}
+		ippdrv->prop_list.ipp_id = ipp_id;
 
 		/* store parent device for node */
 		ippdrv->parent_dev = dev;
@@ -1724,39 +1727,46 @@ static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 		ippdrv->event_workq = ctx->event_workq;
 		ippdrv->sched_event = ipp_sched_event;
 		INIT_LIST_HEAD(&ippdrv->cmd_list);
+		mutex_init(&ippdrv->cmd_lock);
 
 		if (is_drm_iommu_supported(drm_dev)) {
 			ret = drm_iommu_attach_device(drm_dev, ippdrv->dev);
 			if (ret) {
 				DRM_ERROR("failed to activate iommu\n");
-				goto err_iommu;
+				goto err;
 			}
 		}
 	}
 
 	return 0;
 
-err_iommu:
+err:
 	/* get ipp driver entry */
-	list_for_each_entry_reverse(ippdrv, &exynos_drm_ippdrv_list, drv_list)
+	list_for_each_entry_continue_reverse(ippdrv, &exynos_drm_ippdrv_list,
+						drv_list) {
 		if (is_drm_iommu_supported(drm_dev))
 			drm_iommu_detach_device(drm_dev, ippdrv->dev);
 
-err_idr:
-	idr_destroy(&ctx->ipp_idr);
-	idr_destroy(&ctx->prop_idr);
+		ipp_remove_id(&ctx->ipp_idr, &ctx->ipp_lock,
+				ippdrv->prop_list.ipp_id);
+	}
+
 	return ret;
 }
 
 static void ipp_subdrv_remove(struct drm_device *drm_dev, struct device *dev)
 {
 	struct exynos_drm_ippdrv *ippdrv;
+	struct ipp_context *ctx = get_ipp_context(dev);
 
 	/* get ipp driver entry */
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
 		if (is_drm_iommu_supported(drm_dev))
 			drm_iommu_detach_device(drm_dev, ippdrv->dev);
 
+		ipp_remove_id(&ctx->ipp_idr, &ctx->ipp_lock,
+				ippdrv->prop_list.ipp_id);
+
 		ippdrv->drm_dev = NULL;
 		exynos_drm_ippdrv_unregister(ippdrv);
 	}
@@ -1787,20 +1797,14 @@ static void ipp_subdrv_close(struct drm_device *drm_dev, struct device *dev,
 	struct drm_exynos_file_private *file_priv = file->driver_priv;
 	struct exynos_drm_ipp_private *priv = file_priv->ipp_priv;
 	struct exynos_drm_ippdrv *ippdrv = NULL;
+	struct ipp_context *ctx = get_ipp_context(dev);
 	struct drm_exynos_ipp_cmd_node *c_node, *tc_node;
 	int count = 0;
 
 	DRM_DEBUG_KMS("for priv[0x%x]\n", (int)priv);
 
-	if (list_empty(&exynos_drm_ippdrv_list)) {
-		DRM_DEBUG_KMS("ippdrv_list is empty.\n");
-		goto err_clear;
-	}
-
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
-		if (list_empty(&ippdrv->cmd_list))
-			continue;
-
+		mutex_lock(&ippdrv->cmd_lock);
 		list_for_each_entry_safe(c_node, tc_node,
 			&ippdrv->cmd_list, list) {
 			DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]\n",
@@ -1820,14 +1824,14 @@ static void ipp_subdrv_close(struct drm_device *drm_dev, struct device *dev,
 				}
 
 				ippdrv->dedicated = false;
-				ipp_clean_cmd_node(c_node);
+				ipp_clean_cmd_node(ctx, c_node);
 				if (list_empty(&ippdrv->cmd_list))
 					pm_runtime_put_sync(ippdrv->dev);
 			}
 		}
+		mutex_unlock(&ippdrv->cmd_lock);
 	}
 
-err_clear:
 	kfree(priv);
 	return;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.h b/drivers/gpu/drm/exynos/exynos_drm_ipp.h
index ab1634befc05..7aaeaae757c2 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.h
@@ -52,7 +52,7 @@ struct drm_exynos_ipp_cmd_work {
  * @list: list head to command queue information.
  * @event_list: list head of event.
  * @mem_list: list head to source,destination memory queue information.
- * @cmd_lock: lock for synchronization of access to ioctl.
+ * @lock: lock for synchronization of access to ioctl.
  * @mem_lock: lock for synchronization of access to memory nodes.
  * @event_lock: lock for synchronization of access to scheduled event.
  * @start_complete: completion of start of command.
@@ -68,7 +68,7 @@ struct drm_exynos_ipp_cmd_node {
 	struct list_head	list;
 	struct list_head	event_list;
 	struct list_head	mem_list[EXYNOS_DRM_OPS_MAX];
-	struct mutex	cmd_lock;
+	struct mutex	lock;
 	struct mutex	mem_lock;
 	struct mutex	event_lock;
 	struct completion	start_complete;
@@ -83,7 +83,7 @@ struct drm_exynos_ipp_cmd_node {
 /*
  * A structure of buffer information.
  *
- * @gem_objs: Y, Cb, Cr each gem object.
+ * @handles: Y, Cb, Cr each gem object handle.
  * @base: Y, Cb, Cr each planar address.
  */
 struct drm_exynos_ipp_buf_info {
@@ -142,12 +142,12 @@ struct exynos_drm_ipp_ops {
  * @parent_dev: parent device information.
  * @dev: platform device.
  * @drm_dev: drm device.
- * @ipp_id: id of ipp driver.
  * @dedicated: dedicated ipp device.
  * @ops: source, destination operations.
  * @event_workq: event work queue.
  * @c_node: current command information.
  * @cmd_list: list head for command information.
+ * @cmd_lock: lock for synchronization of access to cmd_list.
  * @prop_list: property informations of current ipp driver.
  * @check_property: check property about format, size, buffer.
  * @reset: reset ipp block.
@@ -160,13 +160,13 @@ struct exynos_drm_ippdrv {
 	struct device	*parent_dev;
 	struct device	*dev;
 	struct drm_device	*drm_dev;
-	u32	ipp_id;
 	bool	dedicated;
 	struct exynos_drm_ipp_ops	*ops[EXYNOS_DRM_OPS_MAX];
 	struct workqueue_struct	*event_workq;
 	struct drm_exynos_ipp_cmd_node *c_node;
 	struct list_head	cmd_list;
-	struct drm_exynos_ipp_prop_list *prop_list;
+	struct mutex	cmd_lock;
+	struct drm_exynos_ipp_prop_list prop_list;
 
 	int (*check_property)(struct device *dev,
 		struct drm_exynos_ipp_property *property);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
index 7b901688defa..f01fbb6dc1f0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
@@ -158,8 +158,9 @@ static irqreturn_t rotator_irq_handler(int irq, void *arg)
 			rot->cur_buf_id[EXYNOS_DRM_OPS_DST];
 		queue_work(ippdrv->event_workq,
 			(struct work_struct *)event_work);
-	} else
+	} else {
 		DRM_ERROR("the SFR is set illegally\n");
+	}
 
 	return IRQ_HANDLED;
 }
@@ -469,11 +470,7 @@ static struct exynos_drm_ipp_ops rot_dst_ops = {
 
 static int rotator_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 {
-	struct drm_exynos_ipp_prop_list *prop_list;
-
-	prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL);
-	if (!prop_list)
-		return -ENOMEM;
+	struct drm_exynos_ipp_prop_list *prop_list = &ippdrv->prop_list;
 
 	prop_list->version = 1;
 	prop_list->flip = (1 << EXYNOS_DRM_FLIP_VERTICAL) |
@@ -486,8 +483,6 @@ static int rotator_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 	prop_list->crop = 0;
 	prop_list->scale = 0;
 
-	ippdrv->prop_list = prop_list;
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 852f2dadaebd..2fb8705d6461 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -51,6 +51,7 @@ struct vidi_context {
 	struct drm_crtc			*crtc;
 	struct drm_encoder		*encoder;
 	struct drm_connector		connector;
+	struct exynos_drm_subdrv	subdrv;
 	struct vidi_win_data		win_data[WINDOWS_NR];
 	struct edid			*raw_edid;
 	unsigned int			clkdiv;
@@ -294,14 +295,13 @@ static void vidi_dpms(struct exynos_drm_manager *mgr, int mode)
 }
 
 static int vidi_mgr_initialize(struct exynos_drm_manager *mgr,
-			struct drm_device *drm_dev, int pipe)
+			struct drm_device *drm_dev)
 {
 	struct vidi_context *ctx = mgr->ctx;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
 
-	DRM_ERROR("vidi initialize ct=%p dev=%p pipe=%d\n", ctx, drm_dev, pipe);
-
-	ctx->drm_dev = drm_dev;
-	ctx->pipe = pipe;
+	mgr->drm_dev = ctx->drm_dev = drm_dev;
+	mgr->pipe = ctx->pipe = priv->pipe++;
 
 	/*
 	 * enable drm irq mode.
@@ -324,7 +324,6 @@ static int vidi_mgr_initialize(struct exynos_drm_manager *mgr,
 }
 
 static struct exynos_drm_manager_ops vidi_manager_ops = {
-	.initialize = vidi_mgr_initialize,
 	.dpms = vidi_dpms,
 	.commit = vidi_commit,
 	.enable_vblank = vidi_enable_vblank,
@@ -533,12 +532,6 @@ static int vidi_get_modes(struct drm_connector *connector)
 	return drm_add_edid_modes(connector, edid);
 }
 
-static int vidi_mode_valid(struct drm_connector *connector,
-			struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *vidi_best_encoder(struct drm_connector *connector)
 {
 	struct vidi_context *ctx = ctx_from_connector(connector);
@@ -548,7 +541,6 @@ static struct drm_encoder *vidi_best_encoder(struct drm_connector *connector)
 
 static struct drm_connector_helper_funcs vidi_connector_helper_funcs = {
 	.get_modes = vidi_get_modes,
-	.mode_valid = vidi_mode_valid,
 	.best_encoder = vidi_best_encoder,
 };
 
@@ -586,13 +578,38 @@ static struct exynos_drm_display vidi_display = {
 	.ops = &vidi_display_ops,
 };
 
+static int vidi_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
+{
+	struct exynos_drm_manager *mgr = get_vidi_mgr(dev);
+	struct vidi_context *ctx = mgr->ctx;
+	struct drm_crtc *crtc = ctx->crtc;
+	int ret;
+
+	vidi_mgr_initialize(mgr, drm_dev);
+
+	ret = exynos_drm_crtc_create(&vidi_manager);
+	if (ret) {
+		DRM_ERROR("failed to create crtc.\n");
+		return ret;
+	}
+
+	ret = exynos_drm_create_enc_conn(drm_dev, &vidi_display);
+	if (ret) {
+		crtc->funcs->destroy(crtc);
+		DRM_ERROR("failed to create encoder and connector.\n");
+		return ret;
+	}
+
+	return 0;
+}
+
 static int vidi_probe(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
+	struct exynos_drm_subdrv *subdrv;
 	struct vidi_context *ctx;
 	int ret;
 
-	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
+	ctx = devm_kzalloc(&pdev->dev, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return -ENOMEM;
 
@@ -607,28 +624,43 @@ static int vidi_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, &vidi_manager);
 
-	ret = device_create_file(dev, &dev_attr_connection);
-	if (ret < 0)
-		DRM_INFO("failed to create connection sysfs.\n");
+	subdrv = &ctx->subdrv;
+	subdrv->dev = &pdev->dev;
+	subdrv->probe = vidi_subdrv_probe;
+
+	ret = exynos_drm_subdrv_register(subdrv);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to register drm vidi device\n");
+		return ret;
+	}
 
-	exynos_drm_manager_register(&vidi_manager);
-	exynos_drm_display_register(&vidi_display);
+	ret = device_create_file(&pdev->dev, &dev_attr_connection);
+	if (ret < 0) {
+		exynos_drm_subdrv_unregister(subdrv);
+		DRM_INFO("failed to create connection sysfs.\n");
+	}
 
 	return 0;
 }
 
 static int vidi_remove(struct platform_device *pdev)
 {
-	struct vidi_context *ctx = platform_get_drvdata(pdev);
-
-	exynos_drm_display_unregister(&vidi_display);
-	exynos_drm_manager_unregister(&vidi_manager);
+	struct exynos_drm_manager *mgr = platform_get_drvdata(pdev);
+	struct vidi_context *ctx = mgr->ctx;
+	struct drm_encoder *encoder = ctx->encoder;
+	struct drm_crtc *crtc = mgr->crtc;
 
 	if (ctx->raw_edid != (struct edid *)fake_edid_info) {
 		kfree(ctx->raw_edid);
 		ctx->raw_edid = NULL;
+
+		return -EINVAL;
 	}
 
+	crtc->funcs->destroy(crtc);
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&ctx->connector);
+
 	return 0;
 }
 
@@ -640,3 +672,31 @@ struct platform_driver vidi_driver = {
 		.owner	= THIS_MODULE,
 	},
 };
+
+int exynos_drm_probe_vidi(void)
+{
+	struct platform_device *pdev;
+	int ret;
+
+	pdev = platform_device_register_simple("exynos-drm-vidi", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	ret = platform_driver_register(&vidi_driver);
+	if (ret) {
+		platform_device_unregister(pdev);
+		return ret;
+	}
+
+	return ret;
+}
+
+void exynos_drm_remove_vidi(void)
+{
+	struct vidi_context *ctx = vidi_manager.ctx;
+	struct exynos_drm_subdrv *subdrv = &ctx->subdrv;
+	struct platform_device *pdev = to_platform_device(subdrv->dev);
+
+	platform_driver_unregister(&vidi_driver);
+	platform_device_unregister(pdev);
+}
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 9a6d652a3ef2..aa259b0a873a 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -33,13 +33,17 @@
 #include <linux/regulator/consumer.h>
 #include <linux/io.h>
 #include <linux/of.h>
-#include <linux/i2c.h>
+#include <linux/of_address.h>
 #include <linux/of_gpio.h>
 #include <linux/hdmi.h>
+#include <linux/component.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
 
 #include <drm/exynos_drm.h>
 
 #include "exynos_drm_drv.h"
+#include "exynos_drm_crtc.h"
 #include "exynos_mixer.h"
 
 #include <linux/gpio.h>
@@ -48,6 +52,8 @@
 #define get_hdmi_display(dev)	platform_get_drvdata(to_platform_device(dev))
 #define ctx_from_connector(c)	container_of(c, struct hdmi_context, connector)
 
+#define HOTPLUG_DEBOUNCE_MS		1100
+
 /* AVI header and aspect ratio */
 #define HDMI_AVI_VERSION		0x02
 #define HDMI_AVI_LENGTH		0x0D
@@ -66,6 +72,8 @@ enum hdmi_type {
 
 struct hdmi_driver_data {
 	unsigned int type;
+	const struct hdmiphy_config *phy_confs;
+	unsigned int phy_conf_count;
 	unsigned int is_apb_phy:1;
 };
 
@@ -74,7 +82,6 @@ struct hdmi_resources {
 	struct clk			*sclk_hdmi;
 	struct clk			*sclk_pixel;
 	struct clk			*sclk_hdmiphy;
-	struct clk			*hdmiphy;
 	struct clk			*mout_hdmi;
 	struct regulator_bulk_data	*regul_bulk;
 	int				regul_count;
@@ -185,17 +192,23 @@ struct hdmi_context {
 
 	void __iomem			*regs;
 	int				irq;
+	struct delayed_work		hotplug_work;
 
 	struct i2c_adapter		*ddc_adpt;
 	struct i2c_client		*hdmiphy_port;
 
 	/* current hdmiphy conf regs */
+	struct drm_display_mode		current_mode;
 	struct hdmi_conf_regs		mode_conf;
 
 	struct hdmi_resources		res;
 
 	int				hpd_gpio;
+	void __iomem			*regs_hdmiphy;
+	const struct hdmiphy_config		*phy_confs;
+	unsigned int			phy_conf_count;
 
+	struct regmap			*pmureg;
 	enum hdmi_type			type;
 };
 
@@ -204,14 +217,6 @@ struct hdmiphy_config {
 	u8 conf[32];
 };
 
-struct hdmi_driver_data exynos4212_hdmi_driver_data = {
-	.type	= HDMI_TYPE14,
-};
-
-struct hdmi_driver_data exynos5_hdmi_driver_data = {
-	.type	= HDMI_TYPE14,
-};
-
 /* list of phy config settings */
 static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 	{
@@ -319,18 +324,18 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 	{
 		.pixel_clock = 71000000,
 		.conf = {
-			0x01, 0x91, 0x1e, 0x15, 0x40, 0x3c, 0xce, 0x08,
-			0x04, 0x20, 0xb2, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x3b, 0x35, 0x40, 0x0c, 0x04, 0x08,
+			0x85, 0xa0, 0x63, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0xad, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 	{
 		.pixel_clock = 73250000,
 		.conf = {
-			0x01, 0xd1, 0x1f, 0x15, 0x40, 0x18, 0xe9, 0x08,
-			0x02, 0xa0, 0xb7, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x3d, 0x35, 0x40, 0x18, 0x02, 0x08,
+			0x83, 0xa0, 0x6e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0xa8, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
@@ -362,15 +367,6 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 		},
 	},
 	{
-		.pixel_clock = 88750000,
-		.conf = {
-			0x01, 0x91, 0x25, 0x17, 0x40, 0x30, 0xfe, 0x08,
-			0x06, 0x20, 0xde, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-			0x54, 0x8a, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
-		},
-	},
-	{
 		.pixel_clock = 106500000,
 		.conf = {
 			0x01, 0xd1, 0x2c, 0x12, 0x40, 0x0c, 0x09, 0x08,
@@ -391,18 +387,18 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 	{
 		.pixel_clock = 115500000,
 		.conf = {
-			0x01, 0xd1, 0x30, 0x1a, 0x40, 0x40, 0x10, 0x04,
-			0x04, 0xa0, 0x21, 0xd9, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x30, 0x12, 0x40, 0x40, 0x10, 0x08,
+			0x80, 0x80, 0x21, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0xaa, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 	{
 		.pixel_clock = 119000000,
 		.conf = {
-			0x01, 0x91, 0x32, 0x14, 0x40, 0x60, 0xd8, 0x08,
-			0x06, 0x20, 0x2a, 0xd9, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x32, 0x1a, 0x40, 0x30, 0xd8, 0x08,
+			0x04, 0xa0, 0x2a, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0x9d, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
@@ -426,6 +422,183 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 	},
 };
 
+static const struct hdmiphy_config hdmiphy_5420_configs[] = {
+	{
+		.pixel_clock = 25200000,
+		.conf = {
+			0x01, 0x52, 0x3F, 0x55, 0x40, 0x01, 0x00, 0xC8,
+			0x82, 0xC8, 0xBD, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x06, 0x80, 0x01, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xF4, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 27000000,
+		.conf = {
+			0x01, 0xD1, 0x22, 0x51, 0x40, 0x08, 0xFC, 0xE0,
+			0x98, 0xE8, 0xCB, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x06, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xE4, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 27027000,
+		.conf = {
+			0x01, 0xD1, 0x2D, 0x72, 0x40, 0x64, 0x12, 0xC8,
+			0x43, 0xE8, 0x0E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x26, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xE3, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 36000000,
+		.conf = {
+			0x01, 0x51, 0x2D, 0x55, 0x40, 0x40, 0x00, 0xC8,
+			0x02, 0xC8, 0x0E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xAB, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 40000000,
+		.conf = {
+			0x01, 0xD1, 0x21, 0x31, 0x40, 0x3C, 0x28, 0xC8,
+			0x87, 0xE8, 0xC8, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x9A, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 65000000,
+		.conf = {
+			0x01, 0xD1, 0x36, 0x34, 0x40, 0x0C, 0x04, 0xC8,
+			0x82, 0xE8, 0x45, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xBD, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 71000000,
+		.conf = {
+			0x01, 0xD1, 0x3B, 0x35, 0x40, 0x0C, 0x04, 0xC8,
+			0x85, 0xE8, 0x63, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x57, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 73250000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x10, 0x40, 0x78, 0x8D, 0xC8,
+			0x81, 0xE8, 0xB7, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x56, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xA8, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 74176000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x10, 0x40, 0x5B, 0xEF, 0xC8,
+			0x81, 0xE8, 0xB9, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x56, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xA6, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 74250000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x10, 0x40, 0x40, 0xF8, 0x08,
+			0x81, 0xE8, 0xBA, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x26, 0x80, 0x09, 0x84, 0x05, 0x22, 0x24, 0x66,
+			0x54, 0xA5, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 83500000,
+		.conf = {
+			0x01, 0xD1, 0x23, 0x11, 0x40, 0x0C, 0xFB, 0xC8,
+			0x85, 0xE8, 0xD1, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x4A, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 88750000,
+		.conf = {
+			0x01, 0xD1, 0x25, 0x11, 0x40, 0x18, 0xFF, 0xC8,
+			0x83, 0xE8, 0xDE, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x45, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 106500000,
+		.conf = {
+			0x01, 0xD1, 0x2C, 0x12, 0x40, 0x0C, 0x09, 0xC8,
+			0x84, 0xE8, 0x0A, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x73, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 108000000,
+		.conf = {
+			0x01, 0x51, 0x2D, 0x15, 0x40, 0x01, 0x00, 0xC8,
+			0x82, 0xC8, 0x0E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xC7, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 115500000,
+		.conf = {
+			0x01, 0xD1, 0x30, 0x14, 0x40, 0x0C, 0x03, 0xC8,
+			0x88, 0xE8, 0x21, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x6A, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 146250000,
+		.conf = {
+			0x01, 0xD1, 0x3D, 0x15, 0x40, 0x18, 0xFD, 0xC8,
+			0x83, 0xE8, 0x6E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x54, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 148500000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x00, 0x40, 0x40, 0xF8, 0x08,
+			0x81, 0xE8, 0xBA, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x26, 0x80, 0x09, 0x84, 0x05, 0x22, 0x24, 0x66,
+			0x54, 0x4B, 0x25, 0x03, 0x00, 0x80, 0x01, 0x80,
+		},
+	},
+};
+
+static struct hdmi_driver_data exynos5420_hdmi_driver_data = {
+	.type		= HDMI_TYPE14,
+	.phy_confs	= hdmiphy_5420_configs,
+	.phy_conf_count	= ARRAY_SIZE(hdmiphy_5420_configs),
+	.is_apb_phy	= 1,
+};
+
+static struct hdmi_driver_data exynos4212_hdmi_driver_data = {
+	.type		= HDMI_TYPE14,
+	.phy_confs	= hdmiphy_v14_configs,
+	.phy_conf_count	= ARRAY_SIZE(hdmiphy_v14_configs),
+	.is_apb_phy	= 0,
+};
+
+static struct hdmi_driver_data exynos5_hdmi_driver_data = {
+	.type		= HDMI_TYPE14,
+	.phy_confs	= hdmiphy_v13_configs,
+	.phy_conf_count	= ARRAY_SIZE(hdmiphy_v13_configs),
+	.is_apb_phy	= 0,
+};
+
 static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id)
 {
 	return readl(hdata->regs + reg_id);
@@ -445,6 +618,48 @@ static inline void hdmi_reg_writemask(struct hdmi_context *hdata,
 	writel(value, hdata->regs + reg_id);
 }
 
+static int hdmiphy_reg_writeb(struct hdmi_context *hdata,
+			u32 reg_offset, u8 value)
+{
+	if (hdata->hdmiphy_port) {
+		u8 buffer[2];
+		int ret;
+
+		buffer[0] = reg_offset;
+		buffer[1] = value;
+
+		ret = i2c_master_send(hdata->hdmiphy_port, buffer, 2);
+		if (ret == 2)
+			return 0;
+		return ret;
+	} else {
+		writeb(value, hdata->regs_hdmiphy + (reg_offset<<2));
+		return 0;
+	}
+}
+
+static int hdmiphy_reg_write_buf(struct hdmi_context *hdata,
+			u32 reg_offset, const u8 *buf, u32 len)
+{
+	if ((reg_offset + len) > 32)
+		return -EINVAL;
+
+	if (hdata->hdmiphy_port) {
+		int ret;
+
+		ret = i2c_master_send(hdata->hdmiphy_port, buf, len);
+		if (ret == len)
+			return 0;
+		return ret;
+	} else {
+		int i;
+		for (i = 0; i < len; i++)
+			writeb(buf[i], hdata->regs_hdmiphy +
+				((reg_offset + i)<<2));
+		return 0;
+	}
+}
+
 static void hdmi_v13_regs_dump(struct hdmi_context *hdata, char *prefix)
 {
 #define DUMPREG(reg_id) \
@@ -809,6 +1024,8 @@ static enum drm_connector_status hdmi_detect(struct drm_connector *connector,
 {
 	struct hdmi_context *hdata = ctx_from_connector(connector);
 
+	hdata->hpd = gpio_get_value(hdata->hpd_gpio);
+
 	return hdata->hpd ? connector_status_connected :
 			connector_status_disconnected;
 }
@@ -848,20 +1065,10 @@ static int hdmi_get_modes(struct drm_connector *connector)
 
 static int hdmi_find_phy_conf(struct hdmi_context *hdata, u32 pixel_clock)
 {
-	const struct hdmiphy_config *confs;
-	int count, i;
-
-	if (hdata->type == HDMI_TYPE13) {
-		confs = hdmiphy_v13_configs;
-		count = ARRAY_SIZE(hdmiphy_v13_configs);
-	} else if (hdata->type == HDMI_TYPE14) {
-		confs = hdmiphy_v14_configs;
-		count = ARRAY_SIZE(hdmiphy_v14_configs);
-	} else
-		return -EINVAL;
+	int i;
 
-	for (i = 0; i < count; i++)
-		if (confs[i].pixel_clock == pixel_clock)
+	for (i = 0; i < hdata->phy_conf_count; i++)
+		if (hdata->phy_confs[i].pixel_clock == pixel_clock)
 			return i;
 
 	DRM_DEBUG_KMS("Could not find phy config for %d\n", pixel_clock);
@@ -928,16 +1135,6 @@ static int hdmi_create_connector(struct exynos_drm_display *display,
 	return 0;
 }
 
-static int hdmi_initialize(struct exynos_drm_display *display,
-			struct drm_device *drm_dev)
-{
-	struct hdmi_context *hdata = display->ctx;
-
-	hdata->drm_dev = drm_dev;
-
-	return 0;
-}
-
 static void hdmi_mode_fixup(struct exynos_drm_display *display,
 				struct drm_connector *connector,
 				const struct drm_display_mode *mode,
@@ -1136,20 +1333,15 @@ static void hdmi_audio_control(struct hdmi_context *hdata, bool onoff)
 			HDMI_ASP_EN : HDMI_ASP_DIS, HDMI_ASP_MASK);
 }
 
-static void hdmi_conf_reset(struct hdmi_context *hdata)
+static void hdmi_start(struct hdmi_context *hdata, bool start)
 {
-	u32 reg;
+	u32 val = start ? HDMI_TG_EN : 0;
 
-	if (hdata->type == HDMI_TYPE13)
-		reg = HDMI_V13_CORE_RSTOUT;
-	else
-		reg = HDMI_CORE_RSTOUT;
+	if (hdata->current_mode.flags & DRM_MODE_FLAG_INTERLACE)
+		val |= HDMI_FIELD_EN;
 
-	/* resetting HDMI core */
-	hdmi_reg_writemask(hdata, reg,  0, HDMI_CORE_SW_RSTOUT);
-	usleep_range(10000, 12000);
-	hdmi_reg_writemask(hdata, reg, ~0, HDMI_CORE_SW_RSTOUT);
-	usleep_range(10000, 12000);
+	hdmi_reg_writemask(hdata, HDMI_CON_0, val, HDMI_EN);
+	hdmi_reg_writemask(hdata, HDMI_TG_CMD, val, HDMI_TG_EN | HDMI_FIELD_EN);
 }
 
 static void hdmi_conf_init(struct hdmi_context *hdata)
@@ -1163,6 +1355,8 @@ static void hdmi_conf_init(struct hdmi_context *hdata)
 	/* choose HDMI mode */
 	hdmi_reg_writemask(hdata, HDMI_MODE_SEL,
 		HDMI_MODE_HDMI_EN, HDMI_MODE_MASK);
+	/* Apply Video preable and Guard band in HDMI mode only */
+	hdmi_reg_writeb(hdata, HDMI_CON_2, 0);
 	/* disable bluescreen */
 	hdmi_reg_writemask(hdata, HDMI_CON_0, 0, HDMI_BLUE_SCR_EN);
 
@@ -1286,12 +1480,7 @@ static void hdmi_v13_mode_apply(struct hdmi_context *hdata)
 	clk_prepare_enable(hdata->res.sclk_hdmi);
 
 	/* enable HDMI and timing generator */
-	hdmi_reg_writemask(hdata, HDMI_CON_0, ~0, HDMI_EN);
-	if (core->int_pro_mode[0])
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN |
-				HDMI_FIELD_EN);
-	else
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN);
+	hdmi_start(hdata, true);
 }
 
 static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
@@ -1453,12 +1642,7 @@ static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
 	clk_prepare_enable(hdata->res.sclk_hdmi);
 
 	/* enable HDMI and timing generator */
-	hdmi_reg_writemask(hdata, HDMI_CON_0, ~0, HDMI_EN);
-	if (core->int_pro_mode[0])
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN |
-				HDMI_FIELD_EN);
-	else
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN);
+	hdmi_start(hdata, true);
 }
 
 static void hdmi_mode_apply(struct hdmi_context *hdata)
@@ -1499,32 +1683,51 @@ static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 
 static void hdmiphy_poweron(struct hdmi_context *hdata)
 {
-	if (hdata->type == HDMI_TYPE14)
-		hdmi_reg_writemask(hdata, HDMI_PHY_CON_0, 0,
-			HDMI_PHY_POWER_OFF_EN);
+	if (hdata->type != HDMI_TYPE14)
+		return;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_ENABLE_MODE_SET);
+	/* Phy Power On */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_POWER,
+				HDMI_PHY_POWER_ON);
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_DISABLE_MODE_SET);
+	/* PHY SW Reset */
+	hdmiphy_conf_reset(hdata);
 }
 
 static void hdmiphy_poweroff(struct hdmi_context *hdata)
 {
-	if (hdata->type == HDMI_TYPE14)
-		hdmi_reg_writemask(hdata, HDMI_PHY_CON_0, ~0,
-			HDMI_PHY_POWER_OFF_EN);
+	if (hdata->type != HDMI_TYPE14)
+		return;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* PHY SW Reset */
+	hdmiphy_conf_reset(hdata);
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_ENABLE_MODE_SET);
+
+	/* PHY Power Off */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_POWER,
+				HDMI_PHY_POWER_OFF);
+
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_DISABLE_MODE_SET);
 }
 
 static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 {
-	const u8 *hdmiphy_data;
-	u8 buffer[32];
-	u8 operation[2];
-	u8 read_buffer[32] = {0, };
 	int ret;
 	int i;
 
-	if (!hdata->hdmiphy_port) {
-		DRM_ERROR("hdmiphy is not attached\n");
-		return;
-	}
-
 	/* pixel clock */
 	i = hdmi_find_phy_conf(hdata, hdata->mode_conf.pixel_clock);
 	if (i < 0) {
@@ -1532,39 +1735,21 @@ static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 		return;
 	}
 
-	if (hdata->type == HDMI_TYPE13)
-		hdmiphy_data = hdmiphy_v13_configs[i].conf;
-	else
-		hdmiphy_data = hdmiphy_v14_configs[i].conf;
-
-	memcpy(buffer, hdmiphy_data, 32);
-	ret = i2c_master_send(hdata->hdmiphy_port, buffer, 32);
-	if (ret != 32) {
-		DRM_ERROR("failed to configure HDMIPHY via I2C\n");
+	ret = hdmiphy_reg_write_buf(hdata, 0, hdata->phy_confs[i].conf, 32);
+	if (ret) {
+		DRM_ERROR("failed to configure hdmiphy\n");
 		return;
 	}
 
 	usleep_range(10000, 12000);
 
-	/* operation mode */
-	operation[0] = 0x1f;
-	operation[1] = 0x80;
-
-	ret = i2c_master_send(hdata->hdmiphy_port, operation, 2);
-	if (ret != 2) {
+	ret = hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_DISABLE_MODE_SET);
+	if (ret) {
 		DRM_ERROR("failed to enable hdmiphy\n");
 		return;
 	}
 
-	ret = i2c_master_recv(hdata->hdmiphy_port, read_buffer, 32);
-	if (ret < 0) {
-		DRM_ERROR("failed to read hdmiphy config\n");
-		return;
-	}
-
-	for (i = 0; i < ret; i++)
-		DRM_DEBUG_KMS("hdmiphy[0x%02x] write[0x%02x] - "
-			"recv [0x%02x]\n", i, buffer[i], read_buffer[i]);
 }
 
 static void hdmi_conf_apply(struct hdmi_context *hdata)
@@ -1573,7 +1758,7 @@ static void hdmi_conf_apply(struct hdmi_context *hdata)
 	hdmiphy_conf_apply(hdata);
 
 	mutex_lock(&hdata->hdmi_mutex);
-	hdmi_conf_reset(hdata);
+	hdmi_start(hdata, false);
 	hdmi_conf_init(hdata);
 	mutex_unlock(&hdata->hdmi_mutex);
 
@@ -1814,6 +1999,9 @@ static void hdmi_mode_set(struct exynos_drm_display *display,
 		m->vrefresh, (m->flags & DRM_MODE_FLAG_INTERLACE) ?
 		"INTERLACED" : "PROGERESSIVE");
 
+	/* preserve mode information for later use. */
+	drm_mode_copy(&hdata->current_mode, mode);
+
 	if (hdata->type == HDMI_TYPE13)
 		hdmi_v13_mode_set(hdata, mode);
 	else
@@ -1854,7 +2042,10 @@ static void hdmi_poweron(struct exynos_drm_display *display)
 	if (regulator_bulk_enable(res->regul_count, res->regul_bulk))
 		DRM_DEBUG_KMS("failed to enable regulator bulk\n");
 
-	clk_prepare_enable(res->hdmiphy);
+	/* set pmu hdmiphy control bit to enable hdmiphy */
+	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
+			PMU_HDMI_PHY_ENABLE_BIT, 1);
+
 	clk_prepare_enable(res->hdmi);
 	clk_prepare_enable(res->sclk_hdmi);
 
@@ -1872,16 +2063,20 @@ static void hdmi_poweroff(struct exynos_drm_display *display)
 		goto out;
 	mutex_unlock(&hdata->hdmi_mutex);
 
-	/*
-	 * The TV power domain needs any condition of hdmiphy to turn off and
-	 * its reset state seems to meet the condition.
-	 */
-	hdmiphy_conf_reset(hdata);
+	/* HDMI System Disable */
+	hdmi_reg_writemask(hdata, HDMI_CON_0, 0, HDMI_EN);
+
 	hdmiphy_poweroff(hdata);
 
+	cancel_delayed_work(&hdata->hotplug_work);
+
 	clk_disable_unprepare(res->sclk_hdmi);
 	clk_disable_unprepare(res->hdmi);
-	clk_disable_unprepare(res->hdmiphy);
+
+	/* reset pmu hdmiphy control bit to disable hdmiphy */
+	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
+			PMU_HDMI_PHY_ENABLE_BIT, 0);
+
 	regulator_bulk_disable(res->regul_count, res->regul_bulk);
 
 	pm_runtime_put_sync(hdata->dev);
@@ -1895,6 +2090,11 @@ out:
 
 static void hdmi_dpms(struct exynos_drm_display *display, int mode)
 {
+	struct hdmi_context *hdata = display->ctx;
+	struct drm_encoder *encoder = hdata->encoder;
+	struct drm_crtc *crtc = encoder->crtc;
+	struct drm_crtc_helper_funcs *funcs = NULL;
+
 	DRM_DEBUG_KMS("mode %d\n", mode);
 
 	switch (mode) {
@@ -1904,6 +2104,20 @@ static void hdmi_dpms(struct exynos_drm_display *display, int mode)
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
+		/*
+		 * The SFRs of VP and Mixer are updated by Vertical Sync of
+		 * Timing generator which is a part of HDMI so the sequence
+		 * to disable TV Subsystem should be as following,
+		 *	VP -> Mixer -> HDMI
+		 *
+		 * Below codes will try to disable Mixer and VP(if used)
+		 * prior to disabling HDMI.
+		 */
+		if (crtc)
+			funcs = crtc->helper_private;
+		if (funcs && funcs->dpms)
+			(*funcs->dpms)(crtc, mode);
+
 		hdmi_poweroff(display);
 		break;
 	default:
@@ -1913,7 +2127,6 @@ static void hdmi_dpms(struct exynos_drm_display *display, int mode)
 }
 
 static struct exynos_drm_display_ops hdmi_display_ops = {
-	.initialize	= hdmi_initialize,
 	.create_connector = hdmi_create_connector,
 	.mode_fixup	= hdmi_mode_fixup,
 	.mode_set	= hdmi_mode_set,
@@ -1926,9 +2139,11 @@ static struct exynos_drm_display hdmi_display = {
 	.ops = &hdmi_display_ops,
 };
 
-static irqreturn_t hdmi_irq_thread(int irq, void *arg)
+static void hdmi_hotplug_work_func(struct work_struct *work)
 {
-	struct hdmi_context *hdata = arg;
+	struct hdmi_context *hdata;
+
+	hdata = container_of(work, struct hdmi_context, hotplug_work.work);
 
 	mutex_lock(&hdata->hdmi_mutex);
 	hdata->hpd = gpio_get_value(hdata->hpd_gpio);
@@ -1936,6 +2151,14 @@ static irqreturn_t hdmi_irq_thread(int irq, void *arg)
 
 	if (hdata->drm_dev)
 		drm_helper_hpd_irq_event(hdata->drm_dev);
+}
+
+static irqreturn_t hdmi_irq_thread(int irq, void *arg)
+{
+	struct hdmi_context *hdata = arg;
+
+	mod_delayed_work(system_wq, &hdata->hotplug_work,
+			msecs_to_jiffies(HOTPLUG_DEBOUNCE_MS));
 
 	return IRQ_HANDLED;
 }
@@ -1954,37 +2177,35 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 	DRM_DEBUG_KMS("HDMI resource init\n");
 
-	memset(res, 0, sizeof(*res));
-
 	/* get clocks, power */
 	res->hdmi = devm_clk_get(dev, "hdmi");
 	if (IS_ERR(res->hdmi)) {
 		DRM_ERROR("failed to get clock 'hdmi'\n");
+		ret = PTR_ERR(res->hdmi);
 		goto fail;
 	}
 	res->sclk_hdmi = devm_clk_get(dev, "sclk_hdmi");
 	if (IS_ERR(res->sclk_hdmi)) {
 		DRM_ERROR("failed to get clock 'sclk_hdmi'\n");
+		ret = PTR_ERR(res->sclk_hdmi);
 		goto fail;
 	}
 	res->sclk_pixel = devm_clk_get(dev, "sclk_pixel");
 	if (IS_ERR(res->sclk_pixel)) {
 		DRM_ERROR("failed to get clock 'sclk_pixel'\n");
+		ret = PTR_ERR(res->sclk_pixel);
 		goto fail;
 	}
 	res->sclk_hdmiphy = devm_clk_get(dev, "sclk_hdmiphy");
 	if (IS_ERR(res->sclk_hdmiphy)) {
 		DRM_ERROR("failed to get clock 'sclk_hdmiphy'\n");
-		goto fail;
-	}
-	res->hdmiphy = devm_clk_get(dev, "hdmiphy");
-	if (IS_ERR(res->hdmiphy)) {
-		DRM_ERROR("failed to get clock 'hdmiphy'\n");
+		ret = PTR_ERR(res->sclk_hdmiphy);
 		goto fail;
 	}
 	res->mout_hdmi = devm_clk_get(dev, "mout_hdmi");
 	if (IS_ERR(res->mout_hdmi)) {
 		DRM_ERROR("failed to get clock 'mout_hdmi'\n");
+		ret = PTR_ERR(res->mout_hdmi);
 		goto fail;
 	}
 
@@ -1992,8 +2213,10 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 	res->regul_bulk = devm_kzalloc(dev, ARRAY_SIZE(supply) *
 		sizeof(res->regul_bulk[0]), GFP_KERNEL);
-	if (!res->regul_bulk)
+	if (!res->regul_bulk) {
+		ret = -ENOMEM;
 		goto fail;
+	}
 	for (i = 0; i < ARRAY_SIZE(supply); ++i) {
 		res->regul_bulk[i].supply = supply[i];
 		res->regul_bulk[i].consumer = NULL;
@@ -2001,14 +2224,14 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(supply), res->regul_bulk);
 	if (ret) {
 		DRM_ERROR("failed to get regulators\n");
-		goto fail;
+		return ret;
 	}
 	res->regul_count = ARRAY_SIZE(supply);
 
-	return 0;
+	return ret;
 fail:
 	DRM_ERROR("HDMI resource init - failed\n");
-	return -ENODEV;
+	return ret;
 }
 
 static struct s5p_hdmi_platform_data *drm_hdmi_dt_parse_pdata
@@ -2043,42 +2266,105 @@ static struct of_device_id hdmi_match_types[] = {
 		.compatible = "samsung,exynos4212-hdmi",
 		.data = &exynos4212_hdmi_driver_data,
 	}, {
+		.compatible = "samsung,exynos5420-hdmi",
+		.data = &exynos5420_hdmi_driver_data,
+	}, {
 		/* end node */
 	}
 };
 
+static int hdmi_bind(struct device *dev, struct device *master, void *data)
+{
+	struct drm_device *drm_dev = data;
+	struct hdmi_context *hdata;
+
+	hdata = hdmi_display.ctx;
+	hdata->drm_dev = drm_dev;
+
+	return exynos_drm_create_enc_conn(drm_dev, &hdmi_display);
+}
+
+static void hdmi_unbind(struct device *dev, struct device *master, void *data)
+{
+	struct exynos_drm_display *display = get_hdmi_display(dev);
+	struct drm_encoder *encoder = display->encoder;
+	struct hdmi_context *hdata = display->ctx;
+
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&hdata->connector);
+}
+
+static const struct component_ops hdmi_component_ops = {
+	.bind	= hdmi_bind,
+	.unbind = hdmi_unbind,
+};
+
+static struct device_node *hdmi_legacy_ddc_dt_binding(struct device *dev)
+{
+	const char *compatible_str = "samsung,exynos4210-hdmiddc";
+	struct device_node *np;
+
+	np = of_find_compatible_node(NULL, NULL, compatible_str);
+	if (np)
+		return of_get_next_parent(np);
+
+	return NULL;
+}
+
+static struct device_node *hdmi_legacy_phy_dt_binding(struct device *dev)
+{
+	const char *compatible_str = "samsung,exynos4212-hdmiphy";
+
+	return of_find_compatible_node(NULL, NULL, compatible_str);
+}
+
 static int hdmi_probe(struct platform_device *pdev)
 {
+	struct device_node *ddc_node, *phy_node;
+	struct s5p_hdmi_platform_data *pdata;
+	struct hdmi_driver_data *drv_data;
+	const struct of_device_id *match;
 	struct device *dev = &pdev->dev;
 	struct hdmi_context *hdata;
-	struct s5p_hdmi_platform_data *pdata;
 	struct resource *res;
-	const struct of_device_id *match;
-	struct device_node *ddc_node, *phy_node;
-	struct hdmi_driver_data *drv_data;
 	int ret;
 
-	 if (!dev->of_node)
-		return -ENODEV;
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR,
+					hdmi_display.type);
+	if (ret)
+		return ret;
+
+	if (!dev->of_node) {
+		ret = -ENODEV;
+		goto err_del_component;
+	}
 
 	pdata = drm_hdmi_dt_parse_pdata(dev);
-	if (!pdata)
-		return -EINVAL;
+	if (!pdata) {
+		ret = -EINVAL;
+		goto err_del_component;
+	}
 
 	hdata = devm_kzalloc(dev, sizeof(struct hdmi_context), GFP_KERNEL);
-	if (!hdata)
-		return -ENOMEM;
+	if (!hdata) {
+		ret = -ENOMEM;
+		goto err_del_component;
+	}
 
 	mutex_init(&hdata->hdmi_mutex);
 
 	platform_set_drvdata(pdev, &hdmi_display);
 
 	match = of_match_node(hdmi_match_types, dev->of_node);
-	if (!match)
-		return -ENODEV;
+	if (!match) {
+		ret = -ENODEV;
+		goto err_del_component;
+	}
 
 	drv_data = (struct hdmi_driver_data *)match->data;
 	hdata->type = drv_data->type;
+	hdata->phy_confs = drv_data->phy_confs;
+	hdata->phy_conf_count = drv_data->phy_conf_count;
 
 	hdata->hpd_gpio = pdata->hpd_gpio;
 	hdata->dev = dev;
@@ -2086,35 +2372,44 @@ static int hdmi_probe(struct platform_device *pdev)
 	ret = hdmi_resources_init(hdata);
 	if (ret) {
 		DRM_ERROR("hdmi_resources_init failed\n");
-		return -EINVAL;
+		return ret;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	hdata->regs = devm_ioremap_resource(dev, res);
-	if (IS_ERR(hdata->regs))
-		return PTR_ERR(hdata->regs);
+	if (IS_ERR(hdata->regs)) {
+		ret = PTR_ERR(hdata->regs);
+		goto err_del_component;
+	}
 
 	ret = devm_gpio_request(dev, hdata->hpd_gpio, "HPD");
 	if (ret) {
 		DRM_ERROR("failed to request HPD gpio\n");
-		return ret;
+		goto err_del_component;
 	}
 
+	ddc_node = hdmi_legacy_ddc_dt_binding(dev);
+	if (ddc_node)
+		goto out_get_ddc_adpt;
+
 	/* DDC i2c driver */
 	ddc_node = of_parse_phandle(dev->of_node, "ddc", 0);
 	if (!ddc_node) {
 		DRM_ERROR("Failed to find ddc node in device tree\n");
-		return -ENODEV;
+		ret = -ENODEV;
+		goto err_del_component;
 	}
+
+out_get_ddc_adpt:
 	hdata->ddc_adpt = of_find_i2c_adapter_by_node(ddc_node);
 	if (!hdata->ddc_adpt) {
 		DRM_ERROR("Failed to get ddc i2c adapter by node\n");
-		return -ENODEV;
+		return -EPROBE_DEFER;
 	}
 
-	/* Not support APB PHY yet. */
-	if (drv_data->is_apb_phy)
-		return -EPERM;
+	phy_node = hdmi_legacy_phy_dt_binding(dev);
+	if (phy_node)
+		goto out_get_phy_port;
 
 	/* hdmiphy i2c driver */
 	phy_node = of_parse_phandle(dev->of_node, "phy", 0);
@@ -2123,11 +2418,22 @@ static int hdmi_probe(struct platform_device *pdev)
 		ret = -ENODEV;
 		goto err_ddc;
 	}
-	hdata->hdmiphy_port = of_find_i2c_device_by_node(phy_node);
-	if (!hdata->hdmiphy_port) {
-		DRM_ERROR("Failed to get hdmi phy i2c client from node\n");
-		ret = -ENODEV;
-		goto err_ddc;
+
+out_get_phy_port:
+	if (drv_data->is_apb_phy) {
+		hdata->regs_hdmiphy = of_iomap(phy_node, 0);
+		if (!hdata->regs_hdmiphy) {
+			DRM_ERROR("failed to ioremap hdmi phy\n");
+			ret = -ENOMEM;
+			goto err_ddc;
+		}
+	} else {
+		hdata->hdmiphy_port = of_find_i2c_device_by_node(phy_node);
+		if (!hdata->hdmiphy_port) {
+			DRM_ERROR("Failed to get hdmi phy i2c client\n");
+			ret = -EPROBE_DEFER;
+			goto err_ddc;
+		}
 	}
 
 	hdata->irq = gpio_to_irq(hdata->hpd_gpio);
@@ -2139,6 +2445,8 @@ static int hdmi_probe(struct platform_device *pdev)
 
 	hdata->hpd = gpio_get_value(hdata->hpd_gpio);
 
+	INIT_DELAYED_WORK(&hdata->hotplug_work, hdmi_hotplug_work_func);
+
 	ret = devm_request_threaded_irq(dev, hdata->irq, NULL,
 			hdmi_irq_thread, IRQF_TRIGGER_RISING |
 			IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
@@ -2148,30 +2456,51 @@ static int hdmi_probe(struct platform_device *pdev)
 		goto err_hdmiphy;
 	}
 
-	pm_runtime_enable(dev);
+	hdata->pmureg = syscon_regmap_lookup_by_phandle(dev->of_node,
+			"samsung,syscon-phandle");
+	if (IS_ERR(hdata->pmureg)) {
+		DRM_ERROR("syscon regmap lookup failed.\n");
+		ret = -EPROBE_DEFER;
+		goto err_hdmiphy;
+	}
 
+	pm_runtime_enable(dev);
 	hdmi_display.ctx = hdata;
-	exynos_drm_display_register(&hdmi_display);
 
-	return 0;
+	ret = component_add(&pdev->dev, &hdmi_component_ops);
+	if (ret)
+		goto err_disable_pm_runtime;
+
+	return ret;
+
+err_disable_pm_runtime:
+	pm_runtime_disable(dev);
 
 err_hdmiphy:
-	put_device(&hdata->hdmiphy_port->dev);
+	if (hdata->hdmiphy_port)
+		put_device(&hdata->hdmiphy_port->dev);
 err_ddc:
 	put_device(&hdata->ddc_adpt->dev);
+
+err_del_component:
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
+
 	return ret;
 }
 
 static int hdmi_remove(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
-	struct exynos_drm_display *display = get_hdmi_display(dev);
-	struct hdmi_context *hdata = display->ctx;
+	struct hdmi_context *hdata = hdmi_display.ctx;
+
+	cancel_delayed_work_sync(&hdata->hotplug_work);
 
 	put_device(&hdata->hdmiphy_port->dev);
 	put_device(&hdata->ddc_adpt->dev);
+
 	pm_runtime_disable(&pdev->dev);
+	component_del(&pdev->dev, &hdmi_component_ops);
 
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.h b/drivers/gpu/drm/exynos/exynos_hdmi.h
deleted file mode 100644
index 0ddf3957de15..000000000000
--- a/drivers/gpu/drm/exynos/exynos_hdmi.h
+++ /dev/null
@@ -1,23 +0,0 @@
-/*
- *
- * Copyright (c) 2011 Samsung Electronics Co., Ltd.
- * Authors:
- *	Inki Dae <inki.dae@samsung.com>
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-
-#ifndef _EXYNOS_HDMI_H_
-#define _EXYNOS_HDMI_H_
-
-void hdmi_attach_ddc_client(struct i2c_client *ddc);
-void hdmi_attach_hdmiphy_client(struct i2c_client *hdmiphy);
-
-extern struct i2c_driver hdmiphy_driver;
-extern struct i2c_driver ddc_driver;
-
-#endif
diff --git a/drivers/gpu/drm/exynos/exynos_hdmiphy.c b/drivers/gpu/drm/exynos/exynos_hdmiphy.c
deleted file mode 100644
index 59abb1494ceb..000000000000
--- a/drivers/gpu/drm/exynos/exynos_hdmiphy.c
+++ /dev/null
@@ -1,65 +0,0 @@
-/*
- * Copyright (C) 2011 Samsung Electronics Co.Ltd
- * Authors:
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *	Inki Dae <inki.dae@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <drm/drmP.h>
-
-#include <linux/kernel.h>
-#include <linux/i2c.h>
-#include <linux/of.h>
-
-#include "exynos_drm_drv.h"
-#include "exynos_hdmi.h"
-
-
-static int hdmiphy_probe(struct i2c_client *client,
-	const struct i2c_device_id *id)
-{
-	hdmi_attach_hdmiphy_client(client);
-
-	dev_info(&client->adapter->dev, "attached s5p_hdmiphy "
-		"into i2c adapter successfully\n");
-
-	return 0;
-}
-
-static int hdmiphy_remove(struct i2c_client *client)
-{
-	dev_info(&client->adapter->dev, "detached s5p_hdmiphy "
-		"from i2c adapter successfully\n");
-
-	return 0;
-}
-
-static struct of_device_id hdmiphy_match_types[] = {
-	{
-		.compatible = "samsung,exynos5-hdmiphy",
-	}, {
-		.compatible = "samsung,exynos4210-hdmiphy",
-	}, {
-		.compatible = "samsung,exynos4212-hdmiphy",
-	}, {
-		/* end node */
-	}
-};
-
-struct i2c_driver hdmiphy_driver = {
-	.driver = {
-		.name	= "exynos-hdmiphy",
-		.owner	= THIS_MODULE,
-		.of_match_table = hdmiphy_match_types,
-	},
-	.probe		= hdmiphy_probe,
-	.remove		= hdmiphy_remove,
-	.command		= NULL,
-};
-EXPORT_SYMBOL(hdmiphy_driver);
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index ce288818d2c0..7529946d0a74 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -31,6 +31,7 @@
 #include <linux/clk.h>
 #include <linux/regulator/consumer.h>
 #include <linux/of.h>
+#include <linux/component.h>
 
 #include <drm/exynos_drm.h>
 
@@ -376,6 +377,20 @@ static void mixer_run(struct mixer_context *ctx)
 	mixer_regs_dump(ctx);
 }
 
+static void mixer_stop(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	int timeout = 20;
+
+	mixer_reg_writemask(res, MXR_STATUS, 0, MXR_STATUS_REG_RUN);
+
+	while (!(mixer_reg_read(res, MXR_STATUS) & MXR_STATUS_REG_IDLE) &&
+			--timeout)
+		usleep_range(10000, 12000);
+
+	mixer_regs_dump(ctx);
+}
+
 static void vp_video_buffer(struct mixer_context *ctx, int win)
 {
 	struct mixer_resources *res = &ctx->mixer_res;
@@ -496,13 +511,8 @@ static void vp_video_buffer(struct mixer_context *ctx, int win)
 static void mixer_layer_update(struct mixer_context *ctx)
 {
 	struct mixer_resources *res = &ctx->mixer_res;
-	u32 val;
 
-	val = mixer_reg_read(res, MXR_CFG);
-
-	/* allow one update per vsync only */
-	if (!(val & MXR_CFG_LAYER_UPDATE_COUNT_MASK))
-		mixer_reg_writemask(res, MXR_CFG, ~0, MXR_CFG_LAYER_UPDATE);
+	mixer_reg_writemask(res, MXR_CFG, ~0, MXR_CFG_LAYER_UPDATE);
 }
 
 static void mixer_graph_buffer(struct mixer_context *ctx, int win)
@@ -830,13 +840,15 @@ static int vp_resources_init(struct mixer_context *mixer_ctx)
 }
 
 static int mixer_initialize(struct exynos_drm_manager *mgr,
-			struct drm_device *drm_dev, int pipe)
+			struct drm_device *drm_dev)
 {
 	int ret;
 	struct mixer_context *mixer_ctx = mgr->ctx;
+	struct exynos_drm_private *priv;
+	priv = drm_dev->dev_private;
 
-	mixer_ctx->drm_dev = drm_dev;
-	mixer_ctx->pipe = pipe;
+	mgr->drm_dev = mixer_ctx->drm_dev = drm_dev;
+	mgr->pipe = mixer_ctx->pipe = priv->pipe++;
 
 	/* acquire resources: regs, irqs, clocks */
 	ret = mixer_resources_init(mixer_ctx);
@@ -1007,6 +1019,8 @@ static void mixer_wait_for_vblank(struct exynos_drm_manager *mgr)
 	}
 	mutex_unlock(&mixer_ctx->mixer_mutex);
 
+	drm_vblank_get(mgr->crtc->dev, mixer_ctx->pipe);
+
 	atomic_set(&mixer_ctx->wait_vsync_event, 1);
 
 	/*
@@ -1017,6 +1031,8 @@ static void mixer_wait_for_vblank(struct exynos_drm_manager *mgr)
 				!atomic_read(&mixer_ctx->wait_vsync_event),
 				HZ/20))
 		DRM_DEBUG_KMS("vblank wait timed out.\n");
+
+	drm_vblank_put(mgr->crtc->dev, mixer_ctx->pipe);
 }
 
 static void mixer_window_suspend(struct exynos_drm_manager *mgr)
@@ -1058,7 +1074,7 @@ static void mixer_poweron(struct exynos_drm_manager *mgr)
 		mutex_unlock(&ctx->mixer_mutex);
 		return;
 	}
-	ctx->powered = true;
+
 	mutex_unlock(&ctx->mixer_mutex);
 
 	pm_runtime_get_sync(ctx->dev);
@@ -1069,6 +1085,12 @@ static void mixer_poweron(struct exynos_drm_manager *mgr)
 		clk_prepare_enable(res->sclk_mixer);
 	}
 
+	mutex_lock(&ctx->mixer_mutex);
+	ctx->powered = true;
+	mutex_unlock(&ctx->mixer_mutex);
+
+	mixer_reg_writemask(res, MXR_STATUS, ~0, MXR_STATUS_SOFT_RESET);
+
 	mixer_reg_write(res, MXR_INT_EN, ctx->int_en);
 	mixer_win_reset(ctx);
 
@@ -1081,14 +1103,21 @@ static void mixer_poweroff(struct exynos_drm_manager *mgr)
 	struct mixer_resources *res = &ctx->mixer_res;
 
 	mutex_lock(&ctx->mixer_mutex);
-	if (!ctx->powered)
-		goto out;
+	if (!ctx->powered) {
+		mutex_unlock(&ctx->mixer_mutex);
+		return;
+	}
 	mutex_unlock(&ctx->mixer_mutex);
 
+	mixer_stop(ctx);
 	mixer_window_suspend(mgr);
 
 	ctx->int_en = mixer_reg_read(res, MXR_INT_EN);
 
+	mutex_lock(&ctx->mixer_mutex);
+	ctx->powered = false;
+	mutex_unlock(&ctx->mixer_mutex);
+
 	clk_disable_unprepare(res->mixer);
 	if (ctx->vp_enabled) {
 		clk_disable_unprepare(res->vp);
@@ -1096,12 +1125,6 @@ static void mixer_poweroff(struct exynos_drm_manager *mgr)
 	}
 
 	pm_runtime_put_sync(ctx->dev);
-
-	mutex_lock(&ctx->mixer_mutex);
-	ctx->powered = false;
-
-out:
-	mutex_unlock(&ctx->mixer_mutex);
 }
 
 static void mixer_dpms(struct exynos_drm_manager *mgr, int mode)
@@ -1142,8 +1165,6 @@ int mixer_check_mode(struct drm_display_mode *mode)
 }
 
 static struct exynos_drm_manager_ops mixer_manager_ops = {
-	.initialize		= mixer_initialize,
-	.remove			= mixer_mgr_remove,
 	.dpms			= mixer_dpms,
 	.enable_vblank		= mixer_enable_vblank,
 	.disable_vblank		= mixer_disable_vblank,
@@ -1200,11 +1221,13 @@ static struct of_device_id mixer_match_types[] = {
 	}
 };
 
-static int mixer_probe(struct platform_device *pdev)
+static int mixer_bind(struct device *dev, struct device *manager, void *data)
 {
-	struct device *dev = &pdev->dev;
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm_dev = data;
 	struct mixer_context *ctx;
 	struct mixer_drv_data *drv;
+	int ret;
 
 	dev_info(dev, "probe start\n");
 
@@ -1233,19 +1256,61 @@ static int mixer_probe(struct platform_device *pdev)
 	atomic_set(&ctx->wait_vsync_event, 0);
 
 	mixer_manager.ctx = ctx;
+	ret = mixer_initialize(&mixer_manager, drm_dev);
+	if (ret)
+		return ret;
+
 	platform_set_drvdata(pdev, &mixer_manager);
-	exynos_drm_manager_register(&mixer_manager);
+	ret = exynos_drm_crtc_create(&mixer_manager);
+	if (ret) {
+		mixer_mgr_remove(&mixer_manager);
+		return ret;
+	}
 
 	pm_runtime_enable(dev);
 
 	return 0;
 }
 
-static int mixer_remove(struct platform_device *pdev)
+static void mixer_unbind(struct device *dev, struct device *master, void *data)
 {
-	dev_info(&pdev->dev, "remove successful\n");
+	struct exynos_drm_manager *mgr = dev_get_drvdata(dev);
+	struct drm_crtc *crtc = mgr->crtc;
+
+	dev_info(dev, "remove successful\n");
 
-	pm_runtime_disable(&pdev->dev);
+	mixer_mgr_remove(mgr);
+
+	pm_runtime_disable(dev);
+
+	crtc->funcs->destroy(crtc);
+}
+
+static const struct component_ops mixer_component_ops = {
+	.bind	= mixer_bind,
+	.unbind	= mixer_unbind,
+};
+
+static int mixer_probe(struct platform_device *pdev)
+{
+	int ret;
+
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC,
+					mixer_manager.type);
+	if (ret)
+		return ret;
+
+	ret = component_add(&pdev->dev, &mixer_component_ops);
+	if (ret)
+		exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
+
+	return ret;
+}
+
+static int mixer_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &mixer_component_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/regs-hdmi.h b/drivers/gpu/drm/exynos/regs-hdmi.h
index ef1b3eb3ba6e..3f35ac6d8a47 100644
--- a/drivers/gpu/drm/exynos/regs-hdmi.h
+++ b/drivers/gpu/drm/exynos/regs-hdmi.h
@@ -578,4 +578,20 @@
 #define HDMI_TG_VACT_ST4_H		HDMI_TG_BASE(0x0074)
 #define HDMI_TG_3D			HDMI_TG_BASE(0x00F0)
 
+/* HDMI PHY Registers Offsets*/
+#define HDMIPHY_POWER		(0x74 >> 2)
+#define HDMIPHY_MODE_SET_DONE		(0x7c >> 2)
+
+/* HDMI PHY Values */
+#define HDMI_PHY_POWER_ON              0x80
+#define HDMI_PHY_POWER_OFF             0xff
+
+/* HDMI PHY Values */
+#define HDMI_PHY_DISABLE_MODE_SET	0x80
+#define HDMI_PHY_ENABLE_MODE_SET	0x00
+
+/* PMU Registers for PHY */
+#define PMU_HDMI_PHY_CONTROL		0x700
+#define PMU_HDMI_PHY_ENABLE_BIT		BIT(0)
+
 #endif /* SAMSUNG_REGS_HDMI_H */
diff --git a/drivers/gpu/drm/exynos/regs-mixer.h b/drivers/gpu/drm/exynos/regs-mixer.h
index 4537026bc385..5f32e1a29411 100644
--- a/drivers/gpu/drm/exynos/regs-mixer.h
+++ b/drivers/gpu/drm/exynos/regs-mixer.h
@@ -78,6 +78,7 @@
 #define MXR_STATUS_BIG_ENDIAN		(1 << 3)
 #define MXR_STATUS_ENDIAN_MASK		(1 << 3)
 #define MXR_STATUS_SYNC_ENABLE		(1 << 2)
+#define MXR_STATUS_REG_IDLE		(1 << 1)
 #define MXR_STATUS_REG_RUN		(1 << 0)
 
 /* bits for MXR_CFG */
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
index 489ffd2c66e5..87885d8c06e8 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
@@ -148,7 +148,7 @@ static int handle_dsi_error(struct mdfld_dsi_pkg_sender *sender, u32 mask)
 		break;
 	case BIT(14):
 		/*wait for all fifo empty*/
-		/*wait_for_all_fifos_empty(sender)*/;
+		/*wait_for_all_fifos_empty(sender)*/
 		break;
 	case BIT(15):
 		dev_dbg(sender->dev->dev, "No Action required\n");
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index b686e56646eb..6e8fe9ec02b5 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -112,11 +112,9 @@ static void psb_driver_lastclose(struct drm_device *dev)
 	struct drm_psb_private *dev_priv = dev->dev_private;
 	struct psb_fbdev *fbdev = dev_priv->fbdev;
 
-	drm_modeset_lock_all(dev);
-	ret = drm_fb_helper_restore_fbdev_mode(&fbdev->psb_fb_helper);
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(&fbdev->psb_fb_helper);
 	if (ret)
 		DRM_DEBUG("failed to restore crtc mode\n");
-	drm_modeset_unlock_all(dev);
 
 	return;
 }
@@ -354,7 +352,7 @@ static int psb_driver_load(struct drm_device *dev, unsigned long flags)
 	PSB_WVDC32(0xFFFFFFFF, PSB_INT_MASK_R);
 	spin_unlock_irqrestore(&dev_priv->irqmask_lock, irqflags);
 
-	drm_irq_install(dev);
+	drm_irq_install(dev, dev->pdev->irq);
 
 	dev->vblank_disable_allowed = true;
 	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
@@ -479,7 +477,7 @@ static struct drm_driver driver = {
 	.lastclose = psb_driver_lastclose,
 	.preclose = psb_driver_preclose,
 
-	.num_ioctls = DRM_ARRAY_SIZE(psb_ioctls),
+	.num_ioctls = ARRAY_SIZE(psb_ioctls),
 	.device_is_agp = psb_driver_device_is_agp,
 	.irq_preinstall = psb_irq_preinstall,
 	.irq_postinstall = psb_irq_postinstall,
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 48af5cac1902..ac357b02bd35 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -568,11 +568,11 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 
 static uint8_t tda998x_cksum(uint8_t *buf, size_t bytes)
 {
-	uint8_t sum = 0;
+	int sum = 0;
 
 	while (bytes--)
-		sum += *buf++;
-	return (255 - sum) + 1;
+		sum -= *buf++;
+	return sum;
 }
 
 #define HB(x) (x)
@@ -810,6 +810,12 @@ static int
 tda998x_encoder_mode_valid(struct drm_encoder *encoder,
 			  struct drm_display_mode *mode)
 {
+	if (mode->clock > 150000)
+		return MODE_CLOCK_HIGH;
+	if (mode->htotal >= BIT(13))
+		return MODE_BAD_HVALUE;
+	if (mode->vtotal >= BIT(11))
+		return MODE_BAD_VVALUE;
 	return MODE_OK;
 }
 
@@ -1048,8 +1054,8 @@ read_edid_block(struct drm_encoder *encoder, uint8_t *buf, int blk)
 			return i;
 		}
 	} else {
-		for (i = 10; i > 0; i--) {
-			msleep(10);
+		for (i = 100; i > 0; i--) {
+			msleep(1);
 			ret = reg_read(priv, REG_INT_FLAGS_2);
 			if (ret < 0)
 				return ret;
@@ -1183,7 +1189,6 @@ static void
 tda998x_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct tda998x_priv *priv = to_tda998x_priv(encoder);
-	drm_i2c_encoder_destroy(encoder);
 
 	/* disable all IRQs and free the IRQ handler */
 	cec_write(priv, REG_CEC_RXSHPDINTENA, 0);
@@ -1193,6 +1198,7 @@ tda998x_encoder_destroy(struct drm_encoder *encoder)
 
 	if (priv->cec)
 		i2c_unregister_device(priv->cec);
+	drm_i2c_encoder_destroy(encoder);
 	kfree(priv);
 }
 
diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c
index aeace37415aa..e88bac1d781f 100644
--- a/drivers/gpu/drm/i810/i810_dma.c
+++ b/drivers/gpu/drm/i810/i810_dma.c
@@ -1251,7 +1251,7 @@ const struct drm_ioctl_desc i810_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I810_FLIP, i810_flip_bufs, DRM_AUTH|DRM_UNLOCKED),
 };
 
-int i810_max_ioctl = DRM_ARRAY_SIZE(i810_ioctls);
+int i810_max_ioctl = ARRAY_SIZE(i810_ioctls);
 
 /**
  * Determine if the device really is AGP or not.
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index bea2d67196fb..437e1824d0bf 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -5,6 +5,7 @@ config DRM_I915
 	depends on (AGP || AGP=n)
 	select INTEL_GTT
 	select AGP_INTEL if AGP
+	select INTERVAL_TREE
 	# we need shmfs for the swappable backing store, and in particular
 	# the shmem_readpage() which depends upon tmpfs
 	select SHMEM
@@ -71,7 +72,7 @@ config DRM_I915_PRELIMINARY_HW_SUPPORT
 
 config DRM_I915_UMS
 	bool "Enable userspace modesetting on Intel hardware (DEPRECATED)"
-	depends on DRM_I915
+	depends on DRM_I915 && BROKEN
 	default n
 	help
 	  Choose this option if you still need userspace modesetting.
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index b1445b73465b..cad1683d8bb5 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -18,6 +18,7 @@ i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
 # GEM code
 i915-y += i915_cmd_parser.o \
 	  i915_gem_context.o \
+	  i915_gem_render_state.o \
 	  i915_gem_debug.o \
 	  i915_gem_dmabuf.o \
 	  i915_gem_evict.o \
@@ -26,12 +27,18 @@ i915-y += i915_cmd_parser.o \
 	  i915_gem.o \
 	  i915_gem_stolen.o \
 	  i915_gem_tiling.o \
+	  i915_gem_userptr.o \
 	  i915_gpu_error.o \
 	  i915_irq.o \
 	  i915_trace_points.o \
 	  intel_ringbuffer.o \
 	  intel_uncore.o
 
+# autogenerated null render state
+i915-y += intel_renderstate_gen6.o \
+	  intel_renderstate_gen7.o \
+	  intel_renderstate_gen8.o
+
 # modesetting core code
 i915-y += intel_bios.o \
 	  intel_display.o \
@@ -55,6 +62,7 @@ i915-y += dvo_ch7017.o \
 	  intel_dsi_cmd.o \
 	  intel_dsi.o \
 	  intel_dsi_pll.o \
+	  intel_dsi_panel_vbt.o \
 	  intel_dvo.o \
 	  intel_hdmi.o \
 	  intel_i2c.o \
diff --git a/drivers/gpu/drm/i915/dvo_ch7xxx.c b/drivers/gpu/drm/i915/dvo_ch7xxx.c
index a0f5bdd69491..80449f475960 100644
--- a/drivers/gpu/drm/i915/dvo_ch7xxx.c
+++ b/drivers/gpu/drm/i915/dvo_ch7xxx.c
@@ -160,7 +160,7 @@ static bool ch7xxx_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!ch7xxx->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from %s:%02x.\n",
diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c
index 0f1865d7d4d8..0f2587ff347c 100644
--- a/drivers/gpu/drm/i915/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/dvo_ivch.c
@@ -195,7 +195,7 @@ static bool ivch_read(struct intel_dvo_device *dvo, int addr, uint16_t *data)
 	if (i2c_transfer(adapter, msgs, 3) == 3) {
 		*data = (in_buf[1] << 8) | in_buf[0];
 		return true;
-	};
+	}
 
 	if (!priv->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from "
diff --git a/drivers/gpu/drm/i915/dvo_ns2501.c b/drivers/gpu/drm/i915/dvo_ns2501.c
index 8155ded79079..74f2af7c2d3e 100644
--- a/drivers/gpu/drm/i915/dvo_ns2501.c
+++ b/drivers/gpu/drm/i915/dvo_ns2501.c
@@ -121,7 +121,7 @@ static bool ns2501_readb(struct intel_dvo_device *dvo, int addr, uint8_t * ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!ns->quiet) {
 		DRM_DEBUG_KMS
@@ -233,9 +233,8 @@ static enum drm_mode_status ns2501_mode_valid(struct intel_dvo_device *dvo,
 					      struct drm_display_mode *mode)
 {
 	DRM_DEBUG_KMS
-	    ("%s: is mode valid (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d)\n",
-	     __FUNCTION__, mode->hdisplay, mode->htotal, mode->vdisplay,
-	     mode->vtotal);
+	    ("is mode valid (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d)\n",
+	     mode->hdisplay, mode->htotal, mode->vdisplay, mode->vtotal);
 
 	/*
 	 * Currently, these are all the modes I have data from.
@@ -261,9 +260,8 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 	struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv);
 
 	DRM_DEBUG_KMS
-	    ("%s: set mode (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d).\n",
-	     __FUNCTION__, mode->hdisplay, mode->htotal, mode->vdisplay,
-	     mode->vtotal);
+	    ("set mode (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d).\n",
+	     mode->hdisplay, mode->htotal, mode->vdisplay, mode->vtotal);
 
 	/*
 	 * Where do I find the native resolution for which scaling is not required???
@@ -277,8 +275,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 		if (mode->hdisplay == 800 && mode->vdisplay == 600) {
 			/* mode 277 */
 			ns->reg_8_shadow &= ~NS2501_8_BPAS;
-			DRM_DEBUG_KMS("%s: switching to 800x600\n",
-				      __FUNCTION__);
+			DRM_DEBUG_KMS("switching to 800x600\n");
 
 			/*
 			 * No, I do not know where this data comes from.
@@ -341,8 +338,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 
 		} else if (mode->hdisplay == 640 && mode->vdisplay == 480) {
 			/* mode 274 */
-			DRM_DEBUG_KMS("%s: switching to 640x480\n",
-				      __FUNCTION__);
+			DRM_DEBUG_KMS("switching to 640x480\n");
 			/*
 			 * No, I do not know where this data comes from.
 			 * It is just what the video bios left in the DVO, so
@@ -406,8 +402,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 
 		} else if (mode->hdisplay == 1024 && mode->vdisplay == 768) {
 			/* mode 280 */
-			DRM_DEBUG_KMS("%s: switching to 1024x768\n",
-				      __FUNCTION__);
+			DRM_DEBUG_KMS("switching to 1024x768\n");
 			/*
 			 * This might or might not work, actually. I'm silently
 			 * assuming here that the native panel resolution is
@@ -458,8 +453,7 @@ static void ns2501_dpms(struct intel_dvo_device *dvo, bool enable)
 	struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv);
 	unsigned char ch;
 
-	DRM_DEBUG_KMS("%s: Trying set the dpms of the DVO to %i\n",
-		      __FUNCTION__, enable);
+	DRM_DEBUG_KMS("Trying set the dpms of the DVO to %i\n", enable);
 
 	ch = ns->reg_8_shadow;
 
diff --git a/drivers/gpu/drm/i915/dvo_sil164.c b/drivers/gpu/drm/i915/dvo_sil164.c
index 7b3e9e936200..fa0114967076 100644
--- a/drivers/gpu/drm/i915/dvo_sil164.c
+++ b/drivers/gpu/drm/i915/dvo_sil164.c
@@ -93,7 +93,7 @@ static bool sil164_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!sil->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from %s:%02x.\n",
diff --git a/drivers/gpu/drm/i915/dvo_tfp410.c b/drivers/gpu/drm/i915/dvo_tfp410.c
index 12ea4b164692..7853719a0e81 100644
--- a/drivers/gpu/drm/i915/dvo_tfp410.c
+++ b/drivers/gpu/drm/i915/dvo_tfp410.c
@@ -118,7 +118,7 @@ static bool tfp410_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!tfp->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from %s:%02x.\n",
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 4cf6d020d513..9d7954366bd2 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -28,7 +28,7 @@
 #include "i915_drv.h"
 
 /**
- * DOC: i915 batch buffer command parser
+ * DOC: batch buffer command parser
  *
  * Motivation:
  * Certain OpenGL features (e.g. transform feedback, performance monitoring)
@@ -86,6 +86,367 @@
  * general bitmasking mechanism.
  */
 
+#define STD_MI_OPCODE_MASK  0xFF800000
+#define STD_3D_OPCODE_MASK  0xFFFF0000
+#define STD_2D_OPCODE_MASK  0xFFC00000
+#define STD_MFX_OPCODE_MASK 0xFFFF0000
+
+#define CMD(op, opm, f, lm, fl, ...)				\
+	{							\
+		.flags = (fl) | ((f) ? CMD_DESC_FIXED : 0),	\
+		.cmd = { (op), (opm) }, 			\
+		.length = { (lm) },				\
+		__VA_ARGS__					\
+	}
+
+/* Convenience macros to compress the tables */
+#define SMI STD_MI_OPCODE_MASK
+#define S3D STD_3D_OPCODE_MASK
+#define S2D STD_2D_OPCODE_MASK
+#define SMFX STD_MFX_OPCODE_MASK
+#define F true
+#define S CMD_DESC_SKIP
+#define R CMD_DESC_REJECT
+#define W CMD_DESC_REGISTER
+#define B CMD_DESC_BITMASK
+#define M CMD_DESC_MASTER
+
+/*            Command                          Mask   Fixed Len   Action
+	      ---------------------------------------------------------- */
+static const struct drm_i915_cmd_descriptor common_cmds[] = {
+	CMD(  MI_NOOP,                          SMI,    F,  1,      S  ),
+	CMD(  MI_USER_INTERRUPT,                SMI,    F,  1,      R  ),
+	CMD(  MI_WAIT_FOR_EVENT,                SMI,    F,  1,      M  ),
+	CMD(  MI_ARB_CHECK,                     SMI,    F,  1,      S  ),
+	CMD(  MI_REPORT_HEAD,                   SMI,    F,  1,      S  ),
+	CMD(  MI_SUSPEND_FLUSH,                 SMI,    F,  1,      S  ),
+	CMD(  MI_SEMAPHORE_MBOX,                SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_STORE_DWORD_INDEX,             SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_LOAD_REGISTER_IMM(1),          SMI,   !F,  0xFF,   W,
+	      .reg = { .offset = 1, .mask = 0x007FFFFC }               ),
+	CMD(  MI_STORE_REGISTER_MEM(1),         SMI,   !F,  0xFF,   W | B,
+	      .reg = { .offset = 1, .mask = 0x007FFFFC },
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_LOAD_REGISTER_MEM,             SMI,   !F,  0xFF,   W | B,
+	      .reg = { .offset = 1, .mask = 0x007FFFFC },
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_BATCH_BUFFER_START,            SMI,   !F,  0xFF,   S  ),
+};
+
+static const struct drm_i915_cmd_descriptor render_cmds[] = {
+	CMD(  MI_FLUSH,                         SMI,    F,  1,      S  ),
+	CMD(  MI_ARB_ON_OFF,                    SMI,    F,  1,      R  ),
+	CMD(  MI_PREDICATE,                     SMI,    F,  1,      S  ),
+	CMD(  MI_TOPOLOGY_FILTER,               SMI,    F,  1,      S  ),
+	CMD(  MI_DISPLAY_FLIP,                  SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_SET_CONTEXT,                   SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_URB_CLEAR,                     SMI,   !F,  0xFF,   S  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_CLFLUSH,                       SMI,   !F,  0x3FF,  B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_REPORT_PERF_COUNT,             SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 1,
+			.mask = MI_REPORT_PERF_COUNT_GGTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_CONDITIONAL_BATCH_BUFFER_END,  SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  GFX_OP_3DSTATE_VF_STATISTICS,     S3D,    F,  1,      S  ),
+	CMD(  PIPELINE_SELECT,                  S3D,    F,  1,      S  ),
+	CMD(  MEDIA_VFE_STATE,			S3D,   !F,  0xFFFF, B,
+	      .bits = {{
+			.offset = 2,
+			.mask = MEDIA_VFE_STATE_MMIO_ACCESS_MASK,
+			.expected = 0,
+	      }},						       ),
+	CMD(  GPGPU_OBJECT,                     S3D,   !F,  0xFF,   S  ),
+	CMD(  GPGPU_WALKER,                     S3D,   !F,  0xFF,   S  ),
+	CMD(  GFX_OP_3DSTATE_SO_DECL_LIST,      S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_PIPE_CONTROL(5),           S3D,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 1,
+			.mask = (PIPE_CONTROL_MMIO_WRITE | PIPE_CONTROL_NOTIFY),
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+		        .mask = (PIPE_CONTROL_GLOBAL_GTT_IVB |
+				 PIPE_CONTROL_STORE_DATA_INDEX),
+			.expected = 0,
+			.condition_offset = 1,
+			.condition_mask = PIPE_CONTROL_POST_SYNC_OP_MASK,
+	      }},						       ),
+};
+
+static const struct drm_i915_cmd_descriptor hsw_render_cmds[] = {
+	CMD(  MI_SET_PREDICATE,                 SMI,    F,  1,      S  ),
+	CMD(  MI_RS_CONTROL,                    SMI,    F,  1,      S  ),
+	CMD(  MI_URB_ATOMIC_ALLOC,              SMI,    F,  1,      S  ),
+	CMD(  MI_RS_CONTEXT,                    SMI,    F,  1,      S  ),
+	CMD(  MI_LOAD_SCAN_LINES_INCL,          SMI,   !F,  0x3F,   M  ),
+	CMD(  MI_LOAD_SCAN_LINES_EXCL,          SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_LOAD_REGISTER_REG,             SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_RS_STORE_DATA_IMM,             SMI,   !F,  0xFF,   S  ),
+	CMD(  MI_LOAD_URB_MEM,                  SMI,   !F,  0xFF,   S  ),
+	CMD(  MI_STORE_URB_MEM,                 SMI,   !F,  0xFF,   S  ),
+	CMD(  GFX_OP_3DSTATE_DX9_CONSTANTF_VS,  S3D,   !F,  0x7FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_DX9_CONSTANTF_PS,  S3D,   !F,  0x7FF,  S  ),
+
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_VS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_GS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_HS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_DS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_PS,  S3D,   !F,  0x1FF,  S  ),
+};
+
+static const struct drm_i915_cmd_descriptor video_cmds[] = {
+	CMD(  MI_ARB_ON_OFF,                    SMI,    F,  1,      R  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_FLUSH_DW,                      SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_FLUSH_DW_NOTIFY,
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+			.mask = MI_FLUSH_DW_USE_GTT,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      },
+	      {
+			.offset = 0,
+			.mask = MI_FLUSH_DW_STORE_INDEX,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      }},						       ),
+	CMD(  MI_CONDITIONAL_BATCH_BUFFER_END,  SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	/*
+	 * MFX_WAIT doesn't fit the way we handle length for most commands.
+	 * It has a length field but it uses a non-standard length bias.
+	 * It is always 1 dword though, so just treat it as fixed length.
+	 */
+	CMD(  MFX_WAIT,                         SMFX,   F,  1,      S  ),
+};
+
+static const struct drm_i915_cmd_descriptor vecs_cmds[] = {
+	CMD(  MI_ARB_ON_OFF,                    SMI,    F,  1,      R  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_FLUSH_DW,                      SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_FLUSH_DW_NOTIFY,
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+			.mask = MI_FLUSH_DW_USE_GTT,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      },
+	      {
+			.offset = 0,
+			.mask = MI_FLUSH_DW_STORE_INDEX,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      }},						       ),
+	CMD(  MI_CONDITIONAL_BATCH_BUFFER_END,  SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+};
+
+static const struct drm_i915_cmd_descriptor blt_cmds[] = {
+	CMD(  MI_DISPLAY_FLIP,                  SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0x3FF,  B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_FLUSH_DW,                      SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_FLUSH_DW_NOTIFY,
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+			.mask = MI_FLUSH_DW_USE_GTT,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      },
+	      {
+			.offset = 0,
+			.mask = MI_FLUSH_DW_STORE_INDEX,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      }},						       ),
+	CMD(  COLOR_BLT,                        S2D,   !F,  0x3F,   S  ),
+	CMD(  SRC_COPY_BLT,                     S2D,   !F,  0x3F,   S  ),
+};
+
+static const struct drm_i915_cmd_descriptor hsw_blt_cmds[] = {
+	CMD(  MI_LOAD_SCAN_LINES_INCL,          SMI,   !F,  0x3F,   M  ),
+	CMD(  MI_LOAD_SCAN_LINES_EXCL,          SMI,   !F,  0x3F,   R  ),
+};
+
+#undef CMD
+#undef SMI
+#undef S3D
+#undef S2D
+#undef SMFX
+#undef F
+#undef S
+#undef R
+#undef W
+#undef B
+#undef M
+
+static const struct drm_i915_cmd_table gen7_render_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ render_cmds, ARRAY_SIZE(render_cmds) },
+};
+
+static const struct drm_i915_cmd_table hsw_render_ring_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ render_cmds, ARRAY_SIZE(render_cmds) },
+	{ hsw_render_cmds, ARRAY_SIZE(hsw_render_cmds) },
+};
+
+static const struct drm_i915_cmd_table gen7_video_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ video_cmds, ARRAY_SIZE(video_cmds) },
+};
+
+static const struct drm_i915_cmd_table hsw_vebox_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ vecs_cmds, ARRAY_SIZE(vecs_cmds) },
+};
+
+static const struct drm_i915_cmd_table gen7_blt_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ blt_cmds, ARRAY_SIZE(blt_cmds) },
+};
+
+static const struct drm_i915_cmd_table hsw_blt_ring_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ blt_cmds, ARRAY_SIZE(blt_cmds) },
+	{ hsw_blt_cmds, ARRAY_SIZE(hsw_blt_cmds) },
+};
+
+/*
+ * Register whitelists, sorted by increasing register offset.
+ *
+ * Some registers that userspace accesses are 64 bits. The register
+ * access commands only allow 32-bit accesses. Hence, we have to include
+ * entries for both halves of the 64-bit registers.
+ */
+
+/* Convenience macro for adding 64-bit registers */
+#define REG64(addr) (addr), (addr + sizeof(u32))
+
+static const u32 gen7_render_regs[] = {
+	REG64(HS_INVOCATION_COUNT),
+	REG64(DS_INVOCATION_COUNT),
+	REG64(IA_VERTICES_COUNT),
+	REG64(IA_PRIMITIVES_COUNT),
+	REG64(VS_INVOCATION_COUNT),
+	REG64(GS_INVOCATION_COUNT),
+	REG64(GS_PRIMITIVES_COUNT),
+	REG64(CL_INVOCATION_COUNT),
+	REG64(CL_PRIMITIVES_COUNT),
+	REG64(PS_INVOCATION_COUNT),
+	REG64(PS_DEPTH_COUNT),
+	OACONTROL, /* Only allowed for LRI and SRM. See below. */
+	GEN7_3DPRIM_END_OFFSET,
+	GEN7_3DPRIM_START_VERTEX,
+	GEN7_3DPRIM_VERTEX_COUNT,
+	GEN7_3DPRIM_INSTANCE_COUNT,
+	GEN7_3DPRIM_START_INSTANCE,
+	GEN7_3DPRIM_BASE_VERTEX,
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(0)),
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(1)),
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(2)),
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(3)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(0)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(1)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(2)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(3)),
+	GEN7_SO_WRITE_OFFSET(0),
+	GEN7_SO_WRITE_OFFSET(1),
+	GEN7_SO_WRITE_OFFSET(2),
+	GEN7_SO_WRITE_OFFSET(3),
+};
+
+static const u32 gen7_blt_regs[] = {
+	BCS_SWCTRL,
+};
+
+static const u32 ivb_master_regs[] = {
+	FORCEWAKE_MT,
+	DERRMR,
+	GEN7_PIPE_DE_LOAD_SL(PIPE_A),
+	GEN7_PIPE_DE_LOAD_SL(PIPE_B),
+	GEN7_PIPE_DE_LOAD_SL(PIPE_C),
+};
+
+static const u32 hsw_master_regs[] = {
+	FORCEWAKE_MT,
+	DERRMR,
+};
+
+#undef REG64
+
 static u32 gen7_render_get_cmd_length_mask(u32 cmd_header)
 {
 	u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT;
@@ -137,15 +498,18 @@ static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header)
 	return 0;
 }
 
-static void validate_cmds_sorted(struct intel_ring_buffer *ring)
+static bool validate_cmds_sorted(struct intel_engine_cs *ring,
+				 const struct drm_i915_cmd_table *cmd_tables,
+				 int cmd_table_count)
 {
 	int i;
+	bool ret = true;
 
-	if (!ring->cmd_tables || ring->cmd_table_count == 0)
-		return;
+	if (!cmd_tables || cmd_table_count == 0)
+		return true;
 
-	for (i = 0; i < ring->cmd_table_count; i++) {
-		const struct drm_i915_cmd_table *table = &ring->cmd_tables[i];
+	for (i = 0; i < cmd_table_count; i++) {
+		const struct drm_i915_cmd_table *table = &cmd_tables[i];
 		u32 previous = 0;
 		int j;
 
@@ -154,35 +518,107 @@ static void validate_cmds_sorted(struct intel_ring_buffer *ring)
 				&table->table[i];
 			u32 curr = desc->cmd.value & desc->cmd.mask;
 
-			if (curr < previous)
+			if (curr < previous) {
 				DRM_ERROR("CMD: table not sorted ring=%d table=%d entry=%d cmd=0x%08X prev=0x%08X\n",
 					  ring->id, i, j, curr, previous);
+				ret = false;
+			}
 
 			previous = curr;
 		}
 	}
+
+	return ret;
 }
 
-static void check_sorted(int ring_id, const u32 *reg_table, int reg_count)
+static bool check_sorted(int ring_id, const u32 *reg_table, int reg_count)
 {
 	int i;
 	u32 previous = 0;
+	bool ret = true;
 
 	for (i = 0; i < reg_count; i++) {
 		u32 curr = reg_table[i];
 
-		if (curr < previous)
+		if (curr < previous) {
 			DRM_ERROR("CMD: table not sorted ring=%d entry=%d reg=0x%08X prev=0x%08X\n",
 				  ring_id, i, curr, previous);
+			ret = false;
+		}
 
 		previous = curr;
 	}
+
+	return ret;
+}
+
+static bool validate_regs_sorted(struct intel_engine_cs *ring)
+{
+	return check_sorted(ring->id, ring->reg_table, ring->reg_count) &&
+		check_sorted(ring->id, ring->master_reg_table,
+			     ring->master_reg_count);
+}
+
+struct cmd_node {
+	const struct drm_i915_cmd_descriptor *desc;
+	struct hlist_node node;
+};
+
+/*
+ * Different command ranges have different numbers of bits for the opcode. For
+ * example, MI commands use bits 31:23 while 3D commands use bits 31:16. The
+ * problem is that, for example, MI commands use bits 22:16 for other fields
+ * such as GGTT vs PPGTT bits. If we include those bits in the mask then when
+ * we mask a command from a batch it could hash to the wrong bucket due to
+ * non-opcode bits being set. But if we don't include those bits, some 3D
+ * commands may hash to the same bucket due to not including opcode bits that
+ * make the command unique. For now, we will risk hashing to the same bucket.
+ *
+ * If we attempt to generate a perfect hash, we should be able to look at bits
+ * 31:29 of a command from a batch buffer and use the full mask for that
+ * client. The existing INSTR_CLIENT_MASK/SHIFT defines can be used for this.
+ */
+#define CMD_HASH_MASK STD_MI_OPCODE_MASK
+
+static int init_hash_table(struct intel_engine_cs *ring,
+			   const struct drm_i915_cmd_table *cmd_tables,
+			   int cmd_table_count)
+{
+	int i, j;
+
+	hash_init(ring->cmd_hash);
+
+	for (i = 0; i < cmd_table_count; i++) {
+		const struct drm_i915_cmd_table *table = &cmd_tables[i];
+
+		for (j = 0; j < table->count; j++) {
+			const struct drm_i915_cmd_descriptor *desc =
+				&table->table[j];
+			struct cmd_node *desc_node =
+				kmalloc(sizeof(*desc_node), GFP_KERNEL);
+
+			if (!desc_node)
+				return -ENOMEM;
+
+			desc_node->desc = desc;
+			hash_add(ring->cmd_hash, &desc_node->node,
+				 desc->cmd.value & CMD_HASH_MASK);
+		}
+	}
+
+	return 0;
 }
 
-static void validate_regs_sorted(struct intel_ring_buffer *ring)
+static void fini_hash_table(struct intel_engine_cs *ring)
 {
-	check_sorted(ring->id, ring->reg_table, ring->reg_count);
-	check_sorted(ring->id, ring->master_reg_table, ring->master_reg_count);
+	struct hlist_node *tmp;
+	struct cmd_node *desc_node;
+	int i;
+
+	hash_for_each_safe(ring->cmd_hash, i, tmp, desc_node, node) {
+		hash_del(&desc_node->node);
+		kfree(desc_node);
+	}
 }
 
 /**
@@ -190,25 +626,74 @@ static void validate_regs_sorted(struct intel_ring_buffer *ring)
  * @ring: the ringbuffer to initialize
  *
  * Optionally initializes fields related to batch buffer command parsing in the
- * struct intel_ring_buffer based on whether the platform requires software
+ * struct intel_engine_cs based on whether the platform requires software
  * command parsing.
+ *
+ * Return: non-zero if initialization fails
  */
-void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring)
+int i915_cmd_parser_init_ring(struct intel_engine_cs *ring)
 {
+	const struct drm_i915_cmd_table *cmd_tables;
+	int cmd_table_count;
+	int ret;
+
 	if (!IS_GEN7(ring->dev))
-		return;
+		return 0;
 
 	switch (ring->id) {
 	case RCS:
+		if (IS_HASWELL(ring->dev)) {
+			cmd_tables = hsw_render_ring_cmds;
+			cmd_table_count =
+				ARRAY_SIZE(hsw_render_ring_cmds);
+		} else {
+			cmd_tables = gen7_render_cmds;
+			cmd_table_count = ARRAY_SIZE(gen7_render_cmds);
+		}
+
+		ring->reg_table = gen7_render_regs;
+		ring->reg_count = ARRAY_SIZE(gen7_render_regs);
+
+		if (IS_HASWELL(ring->dev)) {
+			ring->master_reg_table = hsw_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(hsw_master_regs);
+		} else {
+			ring->master_reg_table = ivb_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(ivb_master_regs);
+		}
+
 		ring->get_cmd_length_mask = gen7_render_get_cmd_length_mask;
 		break;
 	case VCS:
+		cmd_tables = gen7_video_cmds;
+		cmd_table_count = ARRAY_SIZE(gen7_video_cmds);
 		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
 	case BCS:
+		if (IS_HASWELL(ring->dev)) {
+			cmd_tables = hsw_blt_ring_cmds;
+			cmd_table_count = ARRAY_SIZE(hsw_blt_ring_cmds);
+		} else {
+			cmd_tables = gen7_blt_cmds;
+			cmd_table_count = ARRAY_SIZE(gen7_blt_cmds);
+		}
+
+		ring->reg_table = gen7_blt_regs;
+		ring->reg_count = ARRAY_SIZE(gen7_blt_regs);
+
+		if (IS_HASWELL(ring->dev)) {
+			ring->master_reg_table = hsw_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(hsw_master_regs);
+		} else {
+			ring->master_reg_table = ivb_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(ivb_master_regs);
+		}
+
 		ring->get_cmd_length_mask = gen7_blt_get_cmd_length_mask;
 		break;
 	case VECS:
+		cmd_tables = hsw_vebox_cmds;
+		cmd_table_count = ARRAY_SIZE(hsw_vebox_cmds);
 		/* VECS can use the same length_mask function as VCS */
 		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
@@ -218,18 +703,45 @@ void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring)
 		BUG();
 	}
 
-	validate_cmds_sorted(ring);
-	validate_regs_sorted(ring);
+	BUG_ON(!validate_cmds_sorted(ring, cmd_tables, cmd_table_count));
+	BUG_ON(!validate_regs_sorted(ring));
+
+	ret = init_hash_table(ring, cmd_tables, cmd_table_count);
+	if (ret) {
+		DRM_ERROR("CMD: cmd_parser_init failed!\n");
+		fini_hash_table(ring);
+		return ret;
+	}
+
+	ring->needs_cmd_parser = true;
+
+	return 0;
+}
+
+/**
+ * i915_cmd_parser_fini_ring() - clean up cmd parser related fields
+ * @ring: the ringbuffer to clean up
+ *
+ * Releases any resources related to command parsing that may have been
+ * initialized for the specified ring.
+ */
+void i915_cmd_parser_fini_ring(struct intel_engine_cs *ring)
+{
+	if (!ring->needs_cmd_parser)
+		return;
+
+	fini_hash_table(ring);
 }
 
 static const struct drm_i915_cmd_descriptor*
-find_cmd_in_table(const struct drm_i915_cmd_table *table,
+find_cmd_in_table(struct intel_engine_cs *ring,
 		  u32 cmd_header)
 {
-	int i;
+	struct cmd_node *desc_node;
 
-	for (i = 0; i < table->count; i++) {
-		const struct drm_i915_cmd_descriptor *desc = &table->table[i];
+	hash_for_each_possible(ring->cmd_hash, desc_node, node,
+			       cmd_header & CMD_HASH_MASK) {
+		const struct drm_i915_cmd_descriptor *desc = desc_node->desc;
 		u32 masked_cmd = desc->cmd.mask & cmd_header;
 		u32 masked_value = desc->cmd.value & desc->cmd.mask;
 
@@ -249,20 +761,16 @@ find_cmd_in_table(const struct drm_i915_cmd_table *table,
  * ring's default length encoding and returns default_desc.
  */
 static const struct drm_i915_cmd_descriptor*
-find_cmd(struct intel_ring_buffer *ring,
+find_cmd(struct intel_engine_cs *ring,
 	 u32 cmd_header,
 	 struct drm_i915_cmd_descriptor *default_desc)
 {
+	const struct drm_i915_cmd_descriptor *desc;
 	u32 mask;
-	int i;
 
-	for (i = 0; i < ring->cmd_table_count; i++) {
-		const struct drm_i915_cmd_descriptor *desc;
-
-		desc = find_cmd_in_table(&ring->cmd_tables[i], cmd_header);
-		if (desc)
-			return desc;
-	}
+	desc = find_cmd_in_table(ring, cmd_header);
+	if (desc)
+		return desc;
 
 	mask = ring->get_cmd_length_mask(cmd_header);
 	if (!mask)
@@ -329,15 +837,112 @@ finish:
  *
  * Return: true if the ring requires software command parsing
  */
-bool i915_needs_cmd_parser(struct intel_ring_buffer *ring)
+bool i915_needs_cmd_parser(struct intel_engine_cs *ring)
 {
-	/* No command tables indicates a platform without parsing */
-	if (!ring->cmd_tables)
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+
+	if (!ring->needs_cmd_parser)
+		return false;
+
+	/*
+	 * XXX: VLV is Gen7 and therefore has cmd_tables, but has PPGTT
+	 * disabled. That will cause all of the parser's PPGTT checks to
+	 * fail. For now, disable parsing when PPGTT is off.
+	 */
+	if (!dev_priv->mm.aliasing_ppgtt)
 		return false;
 
 	return (i915.enable_cmd_parser == 1);
 }
 
+static bool check_cmd(const struct intel_engine_cs *ring,
+		      const struct drm_i915_cmd_descriptor *desc,
+		      const u32 *cmd,
+		      const bool is_master,
+		      bool *oacontrol_set)
+{
+	if (desc->flags & CMD_DESC_REJECT) {
+		DRM_DEBUG_DRIVER("CMD: Rejected command: 0x%08X\n", *cmd);
+		return false;
+	}
+
+	if ((desc->flags & CMD_DESC_MASTER) && !is_master) {
+		DRM_DEBUG_DRIVER("CMD: Rejected master-only command: 0x%08X\n",
+				 *cmd);
+		return false;
+	}
+
+	if (desc->flags & CMD_DESC_REGISTER) {
+		u32 reg_addr = cmd[desc->reg.offset] & desc->reg.mask;
+
+		/*
+		 * OACONTROL requires some special handling for writes. We
+		 * want to make sure that any batch which enables OA also
+		 * disables it before the end of the batch. The goal is to
+		 * prevent one process from snooping on the perf data from
+		 * another process. To do that, we need to check the value
+		 * that will be written to the register. Hence, limit
+		 * OACONTROL writes to only MI_LOAD_REGISTER_IMM commands.
+		 */
+		if (reg_addr == OACONTROL) {
+			if (desc->cmd.value == MI_LOAD_REGISTER_MEM)
+				return false;
+
+			if (desc->cmd.value == MI_LOAD_REGISTER_IMM(1))
+				*oacontrol_set = (cmd[2] != 0);
+		}
+
+		if (!valid_reg(ring->reg_table,
+			       ring->reg_count, reg_addr)) {
+			if (!is_master ||
+			    !valid_reg(ring->master_reg_table,
+				       ring->master_reg_count,
+				       reg_addr)) {
+				DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
+						 reg_addr,
+						 *cmd,
+						 ring->id);
+				return false;
+			}
+		}
+	}
+
+	if (desc->flags & CMD_DESC_BITMASK) {
+		int i;
+
+		for (i = 0; i < MAX_CMD_DESC_BITMASKS; i++) {
+			u32 dword;
+
+			if (desc->bits[i].mask == 0)
+				break;
+
+			if (desc->bits[i].condition_mask != 0) {
+				u32 offset =
+					desc->bits[i].condition_offset;
+				u32 condition = cmd[offset] &
+					desc->bits[i].condition_mask;
+
+				if (condition == 0)
+					continue;
+			}
+
+			dword = cmd[desc->bits[i].offset] &
+				desc->bits[i].mask;
+
+			if (dword != desc->bits[i].expected) {
+				DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (ring=%d)\n",
+						 *cmd,
+						 desc->bits[i].mask,
+						 desc->bits[i].expected,
+						 dword, ring->id);
+				return false;
+			}
+		}
+	}
+
+	return true;
+}
+
 #define LENGTH_BIAS 2
 
 /**
@@ -352,7 +957,7 @@ bool i915_needs_cmd_parser(struct intel_ring_buffer *ring)
  *
  * Return: non-zero if the parser finds violations or otherwise fails
  */
-int i915_parse_cmds(struct intel_ring_buffer *ring,
+int i915_parse_cmds(struct intel_engine_cs *ring,
 		    struct drm_i915_gem_object *batch_obj,
 		    u32 batch_start_offset,
 		    bool is_master)
@@ -361,6 +966,7 @@ int i915_parse_cmds(struct intel_ring_buffer *ring,
 	u32 *cmd, *batch_base, *batch_end;
 	struct drm_i915_cmd_descriptor default_desc = { 0 };
 	int needs_clflush = 0;
+	bool oacontrol_set = false; /* OACONTROL tracking. See check_cmd() */
 
 	ret = i915_gem_obj_prepare_shmem_read(batch_obj, &needs_clflush);
 	if (ret) {
@@ -402,76 +1008,27 @@ int i915_parse_cmds(struct intel_ring_buffer *ring,
 			length = ((*cmd & desc->length.mask) + LENGTH_BIAS);
 
 		if ((batch_end - cmd) < length) {
-			DRM_DEBUG_DRIVER("CMD: Command length exceeds batch length: 0x%08X length=%d batchlen=%td\n",
+			DRM_DEBUG_DRIVER("CMD: Command length exceeds batch length: 0x%08X length=%u batchlen=%td\n",
 					 *cmd,
 					 length,
-					 (unsigned long)(batch_end - cmd));
+					 batch_end - cmd);
 			ret = -EINVAL;
 			break;
 		}
 
-		if (desc->flags & CMD_DESC_REJECT) {
-			DRM_DEBUG_DRIVER("CMD: Rejected command: 0x%08X\n", *cmd);
+		if (!check_cmd(ring, desc, cmd, is_master, &oacontrol_set)) {
 			ret = -EINVAL;
 			break;
 		}
 
-		if ((desc->flags & CMD_DESC_MASTER) && !is_master) {
-			DRM_DEBUG_DRIVER("CMD: Rejected master-only command: 0x%08X\n",
-					 *cmd);
-			ret = -EINVAL;
-			break;
-		}
-
-		if (desc->flags & CMD_DESC_REGISTER) {
-			u32 reg_addr = cmd[desc->reg.offset] & desc->reg.mask;
-
-			if (!valid_reg(ring->reg_table,
-				       ring->reg_count, reg_addr)) {
-				if (!is_master ||
-				    !valid_reg(ring->master_reg_table,
-					       ring->master_reg_count,
-					       reg_addr)) {
-					DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
-							 reg_addr,
-							 *cmd,
-							 ring->id);
-					ret = -EINVAL;
-					break;
-				}
-			}
-		}
-
-		if (desc->flags & CMD_DESC_BITMASK) {
-			int i;
-
-			for (i = 0; i < MAX_CMD_DESC_BITMASKS; i++) {
-				u32 dword;
-
-				if (desc->bits[i].mask == 0)
-					break;
-
-				dword = cmd[desc->bits[i].offset] &
-					desc->bits[i].mask;
-
-				if (dword != desc->bits[i].expected) {
-					DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (ring=%d)\n",
-							 *cmd,
-							 desc->bits[i].mask,
-							 desc->bits[i].expected,
-							 dword, ring->id);
-					ret = -EINVAL;
-					break;
-				}
-			}
-
-			if (ret)
-				break;
-		}
-
 		cmd += length;
 	}
 
+	if (oacontrol_set) {
+		DRM_DEBUG_DRIVER("CMD: batch set OACONTROL but did not clear it\n");
+		ret = -EINVAL;
+	}
+
 	if (cmd >= batch_end) {
 		DRM_DEBUG_DRIVER("CMD: Got to the end of the buffer w/o a BBE cmd!\n");
 		ret = -EINVAL;
@@ -483,3 +1040,22 @@ int i915_parse_cmds(struct intel_ring_buffer *ring,
 
 	return ret;
 }
+
+/**
+ * i915_cmd_parser_get_version() - get the cmd parser version number
+ *
+ * The cmd parser maintains a simple increasing integer version number suitable
+ * for passing to userspace clients to determine what operations are permitted.
+ *
+ * Return: the current version number of the cmd parser
+ */
+int i915_cmd_parser_get_version(void)
+{
+	/*
+	 * Command parser version history
+	 *
+	 * 1. Initial version. Checks batches and reports violations, but leaves
+	 *    hardware parsing enabled (so does not allow new use cases).
+	 */
+	return 1;
+}
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 195fe5bc0aac..b8c689202c40 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -79,7 +79,7 @@ drm_add_fake_info_node(struct drm_minor *minor,
 
 static int i915_capabilities(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	const struct intel_device_info *info = INTEL_INFO(dev);
 
@@ -172,7 +172,7 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		seq_printf(m, " (%s)", obj->ring->name);
 }
 
-static void describe_ctx(struct seq_file *m, struct i915_hw_context *ctx)
+static void describe_ctx(struct seq_file *m, struct intel_context *ctx)
 {
 	seq_putc(m, ctx->is_initialized ? 'I' : 'i');
 	seq_putc(m, ctx->remap_slice ? 'R' : 'r');
@@ -181,7 +181,7 @@ static void describe_ctx(struct seq_file *m, struct i915_hw_context *ctx)
 
 static int i915_gem_object_list_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	uintptr_t list = (uintptr_t) node->info_ent->data;
 	struct list_head *head;
 	struct drm_device *dev = node->minor->dev;
@@ -239,7 +239,7 @@ static int obj_rank_by_stolen(void *priv,
 
 static int i915_gem_stolen_list_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
@@ -371,7 +371,7 @@ static int per_file_stats(int id, void *ptr, void *data)
 
 static int i915_gem_object_info(struct seq_file *m, void* data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 count, mappable_count, purgeable_count;
@@ -446,7 +446,9 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 
 		memset(&stats, 0, sizeof(stats));
 		stats.file_priv = file->driver_priv;
+		spin_lock(&file->table_lock);
 		idr_for_each(&file->object_idr, per_file_stats, &stats);
+		spin_unlock(&file->table_lock);
 		/*
 		 * Although we have a valid reference on file->pid, that does
 		 * not guarantee that the task_struct who called get_pid() is
@@ -474,7 +476,7 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 
 static int i915_gem_gtt_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	uintptr_t list = (uintptr_t) node->info_ent->data;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -509,12 +511,12 @@ static int i915_gem_gtt_info(struct seq_file *m, void *data)
 
 static int i915_gem_pageflip_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	unsigned long flags;
 	struct intel_crtc *crtc;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		const char pipe = pipe_name(crtc->pipe);
 		const char plane = plane_name(crtc->plane);
 		struct intel_unpin_work *work;
@@ -559,10 +561,10 @@ static int i915_gem_pageflip_info(struct seq_file *m, void *data)
 
 static int i915_gem_request_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	struct drm_i915_gem_request *gem_request;
 	int ret, count, i;
 
@@ -594,7 +596,7 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 }
 
 static void i915_ring_seqno_info(struct seq_file *m,
-				 struct intel_ring_buffer *ring)
+				 struct intel_engine_cs *ring)
 {
 	if (ring->get_seqno) {
 		seq_printf(m, "Current sequence (%s): %u\n",
@@ -604,10 +606,10 @@ static void i915_ring_seqno_info(struct seq_file *m,
 
 static int i915_gem_seqno_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
@@ -627,10 +629,10 @@ static int i915_gem_seqno_info(struct seq_file *m, void *data)
 
 static int i915_interrupt_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i, pipe;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
@@ -638,7 +640,47 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 		return ret;
 	intel_runtime_pm_get(dev_priv);
 
-	if (INTEL_INFO(dev)->gen >= 8) {
+	if (IS_CHERRYVIEW(dev)) {
+		int i;
+		seq_printf(m, "Master Interrupt Control:\t%08x\n",
+			   I915_READ(GEN8_MASTER_IRQ));
+
+		seq_printf(m, "Display IER:\t%08x\n",
+			   I915_READ(VLV_IER));
+		seq_printf(m, "Display IIR:\t%08x\n",
+			   I915_READ(VLV_IIR));
+		seq_printf(m, "Display IIR_RW:\t%08x\n",
+			   I915_READ(VLV_IIR_RW));
+		seq_printf(m, "Display IMR:\t%08x\n",
+			   I915_READ(VLV_IMR));
+		for_each_pipe(pipe)
+			seq_printf(m, "Pipe %c stat:\t%08x\n",
+				   pipe_name(pipe),
+				   I915_READ(PIPESTAT(pipe)));
+
+		seq_printf(m, "Port hotplug:\t%08x\n",
+			   I915_READ(PORT_HOTPLUG_EN));
+		seq_printf(m, "DPFLIPSTAT:\t%08x\n",
+			   I915_READ(VLV_DPFLIPSTAT));
+		seq_printf(m, "DPINVGTT:\t%08x\n",
+			   I915_READ(DPINVGTT));
+
+		for (i = 0; i < 4; i++) {
+			seq_printf(m, "GT Interrupt IMR %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IMR(i)));
+			seq_printf(m, "GT Interrupt IIR %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IIR(i)));
+			seq_printf(m, "GT Interrupt IER %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IER(i)));
+		}
+
+		seq_printf(m, "PCU interrupt mask:\t%08x\n",
+			   I915_READ(GEN8_PCU_IMR));
+		seq_printf(m, "PCU interrupt identity:\t%08x\n",
+			   I915_READ(GEN8_PCU_IIR));
+		seq_printf(m, "PCU interrupt enable:\t%08x\n",
+			   I915_READ(GEN8_PCU_IER));
+	} else if (INTEL_INFO(dev)->gen >= 8) {
 		seq_printf(m, "Master Interrupt Control:\t%08x\n",
 			   I915_READ(GEN8_MASTER_IRQ));
 
@@ -768,7 +810,7 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 
 static int i915_gem_fence_regs_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i, ret;
@@ -797,10 +839,10 @@ static int i915_gem_fence_regs_info(struct seq_file *m, void *data)
 
 static int i915_hws_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	const u32 *hws;
 	int i;
 
@@ -945,7 +987,7 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_next_seqno_fops,
 
 static int i915_rstdby_delays(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u16 crstanddelay;
@@ -966,9 +1008,9 @@ static int i915_rstdby_delays(struct seq_file *m, void *unused)
 	return 0;
 }
 
-static int i915_cur_delayinfo(struct seq_file *m, void *unused)
+static int i915_frequency_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret = 0;
@@ -991,6 +1033,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 		u32 rp_state_limits = I915_READ(GEN6_RP_STATE_LIMITS);
 		u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+		u32 rpmodectl, rpinclimit, rpdeclimit;
 		u32 rpstat, cagf, reqf;
 		u32 rpupei, rpcurup, rpprevup;
 		u32 rpdownei, rpcurdown, rpprevdown;
@@ -1011,6 +1054,10 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 			reqf >>= 25;
 		reqf *= GT_FREQUENCY_MULTIPLIER;
 
+		rpmodectl = I915_READ(GEN6_RP_CONTROL);
+		rpinclimit = I915_READ(GEN6_RP_UP_THRESHOLD);
+		rpdeclimit = I915_READ(GEN6_RP_DOWN_THRESHOLD);
+
 		rpstat = I915_READ(GEN6_RPSTAT1);
 		rpupei = I915_READ(GEN6_RP_CUR_UP_EI);
 		rpcurup = I915_READ(GEN6_RP_CUR_UP);
@@ -1027,14 +1074,23 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		gen6_gt_force_wake_put(dev_priv, FORCEWAKE_ALL);
 		mutex_unlock(&dev->struct_mutex);
 
+		seq_printf(m, "PM IER=0x%08x IMR=0x%08x ISR=0x%08x IIR=0x%08x, MASK=0x%08x\n",
+			   I915_READ(GEN6_PMIER),
+			   I915_READ(GEN6_PMIMR),
+			   I915_READ(GEN6_PMISR),
+			   I915_READ(GEN6_PMIIR),
+			   I915_READ(GEN6_PMINTRMSK));
 		seq_printf(m, "GT_PERF_STATUS: 0x%08x\n", gt_perf_status);
-		seq_printf(m, "RPSTAT1: 0x%08x\n", rpstat);
 		seq_printf(m, "Render p-state ratio: %d\n",
 			   (gt_perf_status & 0xff00) >> 8);
 		seq_printf(m, "Render p-state VID: %d\n",
 			   gt_perf_status & 0xff);
 		seq_printf(m, "Render p-state limit: %d\n",
 			   rp_state_limits & 0xff);
+		seq_printf(m, "RPSTAT1: 0x%08x\n", rpstat);
+		seq_printf(m, "RPMODECTL: 0x%08x\n", rpmodectl);
+		seq_printf(m, "RPINCLIMIT: 0x%08x\n", rpinclimit);
+		seq_printf(m, "RPDECLIMIT: 0x%08x\n", rpdeclimit);
 		seq_printf(m, "RPNSWREQ: %dMHz\n", reqf);
 		seq_printf(m, "CAGF: %dMHz\n", cagf);
 		seq_printf(m, "RP CUR UP EI: %dus\n", rpupei &
@@ -1094,7 +1150,7 @@ out:
 
 static int i915_delayfreq_table(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 delayfreq;
@@ -1125,7 +1181,7 @@ static inline int MAP_TO_MV(int map)
 
 static int i915_inttoext_table(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 inttoext;
@@ -1149,7 +1205,7 @@ static int i915_inttoext_table(struct seq_file *m, void *unused)
 
 static int ironlake_drpc_info(struct seq_file *m)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 rgvmodectl, rstdbyctl;
@@ -1219,15 +1275,19 @@ static int ironlake_drpc_info(struct seq_file *m)
 static int vlv_drpc_info(struct seq_file *m)
 {
 
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 rpmodectl1, rcctl1;
 	unsigned fw_rendercount = 0, fw_mediacount = 0;
 
+	intel_runtime_pm_get(dev_priv);
+
 	rpmodectl1 = I915_READ(GEN6_RP_CONTROL);
 	rcctl1 = I915_READ(GEN6_RC_CONTROL);
 
+	intel_runtime_pm_put(dev_priv);
+
 	seq_printf(m, "Video Turbo Mode: %s\n",
 		   yesno(rpmodectl1 & GEN6_RP_MEDIA_TURBO));
 	seq_printf(m, "Turbo enabled: %s\n",
@@ -1247,6 +1307,11 @@ static int vlv_drpc_info(struct seq_file *m)
 			(I915_READ(VLV_GTLC_PW_STATUS) &
 				VLV_GTLC_PW_MEDIA_STATUS_MASK) ? "Up" : "Down");
 
+	seq_printf(m, "Render RC6 residency since boot: %u\n",
+		   I915_READ(VLV_GT_RENDER_RC6));
+	seq_printf(m, "Media RC6 residency since boot: %u\n",
+		   I915_READ(VLV_GT_MEDIA_RC6));
+
 	spin_lock_irq(&dev_priv->uncore.lock);
 	fw_rendercount = dev_priv->uncore.fw_rendercount;
 	fw_mediacount = dev_priv->uncore.fw_mediacount;
@@ -1263,7 +1328,7 @@ static int vlv_drpc_info(struct seq_file *m)
 static int gen6_drpc_info(struct seq_file *m)
 {
 
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 rpmodectl1, gt_core_status, rcctl1, rc6vids = 0;
@@ -1362,7 +1427,7 @@ static int gen6_drpc_info(struct seq_file *m)
 
 static int i915_drpc_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 
 	if (IS_VALLEYVIEW(dev))
@@ -1375,7 +1440,7 @@ static int i915_drpc_info(struct seq_file *m, void *unused)
 
 static int i915_fbc_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1437,7 +1502,7 @@ static int i915_fbc_status(struct seq_file *m, void *unused)
 
 static int i915_ips_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1460,7 +1525,7 @@ static int i915_ips_status(struct seq_file *m, void *unused)
 
 static int i915_sr_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool sr_enabled = false;
@@ -1486,7 +1551,7 @@ static int i915_sr_status(struct seq_file *m, void *unused)
 
 static int i915_emon_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long temp, chipset, gfx;
@@ -1514,7 +1579,7 @@ static int i915_emon_status(struct seq_file *m, void *unused)
 
 static int i915_ring_freq_table(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret = 0;
@@ -1557,7 +1622,7 @@ out:
 
 static int i915_gfxec(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -1577,7 +1642,7 @@ static int i915_gfxec(struct seq_file *m, void *unused)
 
 static int i915_opregion(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_opregion *opregion = &dev_priv->opregion;
@@ -1605,7 +1670,7 @@ out:
 
 static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct intel_fbdev *ifbdev = NULL;
 	struct intel_framebuffer *fb;
@@ -1651,11 +1716,11 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 
 static int i915_context_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
-	struct i915_hw_context *ctx;
+	struct intel_engine_cs *ring;
+	struct intel_context *ctx;
 	int ret, i;
 
 	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
@@ -1675,6 +1740,9 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	}
 
 	list_for_each_entry(ctx, &dev_priv->context_list, link) {
+		if (ctx->obj == NULL)
+			continue;
+
 		seq_puts(m, "HW context ");
 		describe_ctx(m, ctx);
 		for_each_ring(ring, dev_priv, i)
@@ -1692,7 +1760,7 @@ static int i915_context_status(struct seq_file *m, void *unused)
 
 static int i915_gen6_forcewake_count_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned forcewake_count = 0, fw_rendercount = 0, fw_mediacount = 0;
@@ -1740,7 +1808,7 @@ static const char *swizzle_string(unsigned swizzle)
 
 static int i915_swizzle_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -1788,10 +1856,14 @@ static int i915_swizzle_info(struct seq_file *m, void *data)
 
 static int per_file_ctx(int id, void *ptr, void *data)
 {
-	struct i915_hw_context *ctx = ptr;
+	struct intel_context *ctx = ptr;
 	struct seq_file *m = data;
 	struct i915_hw_ppgtt *ppgtt = ctx_to_ppgtt(ctx);
 
+	if (i915_gem_context_is_default(ctx))
+		seq_puts(m, "  default context:\n");
+	else
+		seq_printf(m, "  context %d:\n", ctx->id);
 	ppgtt->debug_dump(ppgtt, m);
 
 	return 0;
@@ -1800,7 +1872,7 @@ static int per_file_ctx(int id, void *ptr, void *data)
 static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
 	int unused, i;
 
@@ -1816,8 +1888,7 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 			u64 pdp = I915_READ(ring->mmio_base + offset + 4);
 			pdp <<= 32;
 			pdp |= I915_READ(ring->mmio_base + offset);
-			for (i = 0; i < 4; i++)
-				seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
+			seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
 		}
 	}
 }
@@ -1825,7 +1896,7 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	struct drm_file *file;
 	int i;
 
@@ -1852,12 +1923,9 @@ static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct drm_i915_file_private *file_priv = file->driver_priv;
-		struct i915_hw_ppgtt *pvt_ppgtt;
 
-		pvt_ppgtt = ctx_to_ppgtt(file_priv->private_default_ctx);
 		seq_printf(m, "proc: %s\n",
 			   get_pid_task(file->pid, PIDTYPE_PID)->comm);
-		seq_puts(m, "  default context:\n");
 		idr_for_each(&file_priv->context_idr, per_file_ctx, m);
 	}
 	seq_printf(m, "ECOCHK: 0x%08x\n", I915_READ(GAM_ECOCHK));
@@ -1865,7 +1933,7 @@ static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 
 static int i915_ppgtt_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1885,56 +1953,9 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 	return 0;
 }
 
-static int i915_dpio_info(struct seq_file *m, void *data)
-{
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret;
-
-
-	if (!IS_VALLEYVIEW(dev)) {
-		seq_puts(m, "unsupported\n");
-		return 0;
-	}
-
-	ret = mutex_lock_interruptible(&dev_priv->dpio_lock);
-	if (ret)
-		return ret;
-
-	seq_printf(m, "DPIO_CTL: 0x%08x\n", I915_READ(DPIO_CTL));
-
-	seq_printf(m, "DPIO PLL DW3 CH0 : 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW3(0)));
-	seq_printf(m, "DPIO PLL DW3 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW3(1)));
-
-	seq_printf(m, "DPIO PLL DW5 CH0: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW5(0)));
-	seq_printf(m, "DPIO PLL DW5 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW5(1)));
-
-	seq_printf(m, "DPIO PLL DW7 CH0: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW7(0)));
-	seq_printf(m, "DPIO PLL DW7 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW7(1)));
-
-	seq_printf(m, "DPIO PLL DW10 CH0: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW10(0)));
-	seq_printf(m, "DPIO PLL DW10 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW10(1)));
-
-	seq_printf(m, "DPIO_FASTCLK_DISABLE: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_CMN_DW0));
-
-	mutex_unlock(&dev_priv->dpio_lock);
-
-	return 0;
-}
-
 static int i915_llc(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -2040,11 +2061,11 @@ static int i915_energy_uJ(struct seq_file *m, void *data)
 
 static int i915_pc8_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (!IS_HASWELL(dev)) {
+	if (!IS_HASWELL(dev) && !IS_BROADWELL(dev)) {
 		seq_puts(m, "not supported\n");
 		return 0;
 	}
@@ -2115,7 +2136,7 @@ static const char *power_domain_str(enum intel_display_power_domain domain)
 
 static int i915_power_domain_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
@@ -2170,7 +2191,7 @@ static void intel_encoder_info(struct seq_file *m,
 			       struct intel_crtc *intel_crtc,
 			       struct intel_encoder *intel_encoder)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct intel_connector *intel_connector;
@@ -2178,12 +2199,12 @@ static void intel_encoder_info(struct seq_file *m,
 
 	encoder = &intel_encoder->base;
 	seq_printf(m, "\tencoder %d: type: %s, connectors:\n",
-		   encoder->base.id, drm_get_encoder_name(encoder));
+		   encoder->base.id, encoder->name);
 	for_each_connector_on_encoder(dev, encoder, intel_connector) {
 		struct drm_connector *connector = &intel_connector->base;
 		seq_printf(m, "\t\tconnector %d: type: %s, status: %s",
 			   connector->base.id,
-			   drm_get_connector_name(connector),
+			   connector->name,
 			   drm_get_connector_status_name(connector->status));
 		if (connector->status == connector_status_connected) {
 			struct drm_display_mode *mode = &crtc->mode;
@@ -2197,7 +2218,7 @@ static void intel_encoder_info(struct seq_file *m,
 
 static void intel_crtc_info(struct seq_file *m, struct intel_crtc *intel_crtc)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct intel_encoder *intel_encoder;
@@ -2254,7 +2275,7 @@ static void intel_connector_info(struct seq_file *m,
 	struct drm_display_mode *mode;
 
 	seq_printf(m, "connector %d: type %s, status: %s\n",
-		   connector->base.id, drm_get_connector_name(connector),
+		   connector->base.id, connector->name,
 		   drm_get_connector_status_name(connector->status));
 	if (connector->status == connector_status_connected) {
 		seq_printf(m, "\tname: %s\n", connector->display_info.name);
@@ -2286,10 +2307,8 @@ static bool cursor_active(struct drm_device *dev, int pipe)
 
 	if (IS_845G(dev) || IS_I865G(dev))
 		state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
-	else if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev))
-		state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 	else
-		state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
+		state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 
 	return state;
 }
@@ -2299,10 +2318,7 @@ static bool cursor_position(struct drm_device *dev, int pipe, int *x, int *y)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 pos;
 
-	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev))
-		pos = I915_READ(CURPOS_IVB(pipe));
-	else
-		pos = I915_READ(CURPOS(pipe));
+	pos = I915_READ(CURPOS(pipe));
 
 	*x = (pos >> CURSOR_X_SHIFT) & CURSOR_POS_MASK;
 	if (pos & (CURSOR_POS_SIGN << CURSOR_X_SHIFT))
@@ -2317,7 +2333,7 @@ static bool cursor_position(struct drm_device *dev, int pipe, int *x, int *y)
 
 static int i915_display_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc;
@@ -2327,7 +2343,7 @@ static int i915_display_info(struct seq_file *m, void *unused)
 	drm_modeset_lock_all(dev);
 	seq_printf(m, "CRTC info\n");
 	seq_printf(m, "---------\n");
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		bool active;
 		int x, y;
 
@@ -2339,10 +2355,14 @@ static int i915_display_info(struct seq_file *m, void *unused)
 
 			active = cursor_position(dev, crtc->pipe, &x, &y);
 			seq_printf(m, "\tcursor visible? %s, position (%d, %d), addr 0x%08x, active? %s\n",
-				   yesno(crtc->cursor_visible),
+				   yesno(crtc->cursor_base),
 				   x, y, crtc->cursor_addr,
 				   yesno(active));
 		}
+
+		seq_printf(m, "\tunderrun reporting: cpu=%s pch=%s \n",
+			   yesno(!crtc->cpu_fifo_underrun_disabled),
+			   yesno(!crtc->pch_fifo_underrun_disabled));
 	}
 
 	seq_printf(m, "\n");
@@ -2595,7 +2615,7 @@ static int i9xx_pipe_crc_auto_source(struct drm_device *dev, enum pipe pipe,
 
 	*source = INTEL_PIPE_CRC_SOURCE_PIPE;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
 			    base.head) {
 		if (!encoder->base.crtc)
@@ -2631,7 +2651,7 @@ static int i9xx_pipe_crc_auto_source(struct drm_device *dev, enum pipe pipe,
 			break;
 		}
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -3106,7 +3126,7 @@ static const struct file_operations i915_display_crc_ctl_fops = {
 static void wm_latency_show(struct seq_file *m, const uint16_t wm[5])
 {
 	struct drm_device *dev = m->private;
-	int num_levels = IS_HASWELL(dev) || IS_BROADWELL(dev) ? 5 : 4;
+	int num_levels = ilk_wm_max_level(dev) + 1;
 	int level;
 
 	drm_modeset_lock_all(dev);
@@ -3189,7 +3209,7 @@ static ssize_t wm_latency_write(struct file *file, const char __user *ubuf,
 	struct seq_file *m = file->private_data;
 	struct drm_device *dev = m->private;
 	uint16_t new[5] = { 0 };
-	int num_levels = IS_HASWELL(dev) || IS_BROADWELL(dev) ? 5 : 4;
+	int num_levels = ilk_wm_max_level(dev) + 1;
 	int level;
 	int ret;
 	char tmp[32];
@@ -3286,9 +3306,15 @@ static int
 i915_wedged_set(void *data, u64 val)
 {
 	struct drm_device *dev = data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	intel_runtime_pm_get(dev_priv);
 
 	i915_handle_error(dev, val,
 			  "Manually setting wedged to %llu", val);
+
+	intel_runtime_pm_put(dev_priv);
+
 	return 0;
 }
 
@@ -3774,7 +3800,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gem_hws_bsd", i915_hws_info, 0, (void *)VCS},
 	{"i915_gem_hws_vebox", i915_hws_info, 0, (void *)VECS},
 	{"i915_rstdby_delays", i915_rstdby_delays, 0},
-	{"i915_cur_delayinfo", i915_cur_delayinfo, 0},
+	{"i915_frequency_info", i915_frequency_info, 0},
 	{"i915_delayfreq_table", i915_delayfreq_table, 0},
 	{"i915_inttoext_table", i915_inttoext_table, 0},
 	{"i915_drpc_info", i915_drpc_info, 0},
@@ -3790,7 +3816,6 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gen6_forcewake_count", i915_gen6_forcewake_count_info, 0},
 	{"i915_swizzle_info", i915_swizzle_info, 0},
 	{"i915_ppgtt_info", i915_ppgtt_info, 0},
-	{"i915_dpio", i915_dpio_info, 0},
 	{"i915_llc", i915_llc, 0},
 	{"i915_edp_psr_status", i915_edp_psr_status, 0},
 	{"i915_sink_crc_eDP1", i915_sink_crc, 0},
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index eedb023af27d..d44344140627 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -36,6 +36,8 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include <linux/pci.h>
+#include <linux/console.h>
+#include <linux/vt.h>
 #include <linux/vgaarb.h>
 #include <linux/acpi.h>
 #include <linux/pnp.h>
@@ -44,6 +46,7 @@
 #include <acpi/video.h>
 #include <linux/pm.h>
 #include <linux/pm_runtime.h>
+#include <linux/oom.h>
 
 #define LP_RING(d) (&((struct drm_i915_private *)(d))->ring[RCS])
 
@@ -63,7 +66,7 @@
  * has access to the ring.
  */
 #define RING_LOCK_TEST_WITH_RETURN(dev, file) do {			\
-	if (LP_RING(dev->dev_private)->obj == NULL)			\
+	if (LP_RING(dev->dev_private)->buffer->obj == NULL)			\
 		LOCK_TEST_WITH_RETURN(dev, file);			\
 } while (0)
 
@@ -119,7 +122,7 @@ static void i915_write_hws_pga(struct drm_device *dev)
 static void i915_free_hws(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
 
 	if (dev_priv->status_page_dmah) {
 		drm_pci_free(dev, dev_priv->status_page_dmah);
@@ -139,7 +142,8 @@ void i915_kernel_lost_context(struct drm_device * dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_master_private *master_priv;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 
 	/*
 	 * We should never lose context on the ring with modesetting
@@ -148,17 +152,17 @@ void i915_kernel_lost_context(struct drm_device * dev)
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return;
 
-	ring->head = I915_READ_HEAD(ring) & HEAD_ADDR;
-	ring->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
-	ring->space = ring->head - (ring->tail + I915_RING_FREE_SPACE);
-	if (ring->space < 0)
-		ring->space += ring->size;
+	ringbuf->head = I915_READ_HEAD(ring) & HEAD_ADDR;
+	ringbuf->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
+	ringbuf->space = ringbuf->head - (ringbuf->tail + I915_RING_FREE_SPACE);
+	if (ringbuf->space < 0)
+		ringbuf->space += ringbuf->size;
 
 	if (!dev->primary->master)
 		return;
 
 	master_priv = dev->primary->master->driver_priv;
-	if (ring->head == ring->tail && master_priv->sarea_priv)
+	if (ringbuf->head == ringbuf->tail && master_priv->sarea_priv)
 		master_priv->sarea_priv->perf_boxes |= I915_BOX_RING_EMPTY;
 }
 
@@ -201,7 +205,7 @@ static int i915_initialize(struct drm_device * dev, drm_i915_init_t * init)
 	}
 
 	if (init->ring_size != 0) {
-		if (LP_RING(dev_priv)->obj != NULL) {
+		if (LP_RING(dev_priv)->buffer->obj != NULL) {
 			i915_dma_cleanup(dev);
 			DRM_ERROR("Client tried to initialize ringbuffer in "
 				  "GEM mode\n");
@@ -234,11 +238,11 @@ static int i915_initialize(struct drm_device * dev, drm_i915_init_t * init)
 static int i915_dma_resume(struct drm_device * dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
 
 	DRM_DEBUG_DRIVER("%s\n", __func__);
 
-	if (ring->virtual_start == NULL) {
+	if (ring->buffer->virtual_start == NULL) {
 		DRM_ERROR("can not ioremap virtual address for"
 			  " ring buffer\n");
 		return -ENOMEM;
@@ -360,7 +364,7 @@ static int i915_emit_cmds(struct drm_device * dev, int *buffer, int dwords)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i, ret;
 
-	if ((dwords+1) * sizeof(int) >= LP_RING(dev_priv)->size - 8)
+	if ((dwords+1) * sizeof(int) >= LP_RING(dev_priv)->buffer->size - 8)
 		return -EINVAL;
 
 	for (i = 0; i < dwords;) {
@@ -782,7 +786,7 @@ static int i915_wait_irq(struct drm_device * dev, int irq_nr)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_master_private *master_priv = dev->primary->master->driver_priv;
 	int ret = 0;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
 
 	DRM_DEBUG_DRIVER("irq_nr=%d breadcrumb=%d\n", irq_nr,
 		  READ_BREADCRUMB(dev_priv));
@@ -823,7 +827,7 @@ static int i915_irq_emit(struct drm_device *dev, void *data,
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return -ENODEV;
 
-	if (!dev_priv || !LP_RING(dev_priv)->virtual_start) {
+	if (!dev_priv || !LP_RING(dev_priv)->buffer->virtual_start) {
 		DRM_ERROR("called with no initialization\n");
 		return -EINVAL;
 	}
@@ -1017,6 +1021,9 @@ static int i915_getparam(struct drm_device *dev, void *data,
 	case I915_PARAM_HAS_EXEC_HANDLE_LUT:
 		value = 1;
 		break;
+	case I915_PARAM_CMD_PARSER_VERSION:
+		value = i915_cmd_parser_get_version();
+		break;
 	default:
 		DRM_DEBUG("Unknown parameter %d\n", param->param);
 		return -EINVAL;
@@ -1070,7 +1077,7 @@ static int i915_set_status_page(struct drm_device *dev, void *data,
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	drm_i915_hws_addr_t *hws = data;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return -ENODEV;
@@ -1277,12 +1284,13 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_
 static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	bool can_switch;
 
-	spin_lock(&dev->count_lock);
-	can_switch = (dev->open_count == 0);
-	spin_unlock(&dev->count_lock);
-	return can_switch;
+	/*
+	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
+	 * locking inversion with the driver load path. And the access here is
+	 * completely racy anyway. So don't bother with locking for now.
+	 */
+	return dev->open_count == 0;
 }
 
 static const struct vga_switcheroo_client_ops i915_switcheroo_ops = {
@@ -1326,7 +1334,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	intel_power_domains_init_hw(dev_priv);
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, dev->pdev->irq);
 	if (ret)
 		goto cleanup_gem_stolen;
 
@@ -1336,7 +1344,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	ret = i915_gem_init(dev);
 	if (ret)
-		goto cleanup_power;
+		goto cleanup_irq;
 
 	INIT_WORK(&dev_priv->console_resume_work, intel_console_resume);
 
@@ -1345,10 +1353,8 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	/* Always safe in the mode setting case. */
 	/* FIXME: do pre/post-mode set stuff in core KMS code */
 	dev->vblank_disable_allowed = true;
-	if (INTEL_INFO(dev)->num_pipes == 0) {
-		intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
+	if (INTEL_INFO(dev)->num_pipes == 0)
 		return 0;
-	}
 
 	ret = intel_fbdev_init(dev);
 	if (ret)
@@ -1382,9 +1388,7 @@ cleanup_gem:
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 	WARN_ON(dev_priv->mm.aliasing_ppgtt);
-	drm_mm_takedown(&dev_priv->gtt.base.mm);
-cleanup_power:
-	intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
+cleanup_irq:
 	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
@@ -1447,6 +1451,39 @@ static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
 }
 #endif
 
+#if !defined(CONFIG_VGA_CONSOLE)
+static int i915_kick_out_vgacon(struct drm_i915_private *dev_priv)
+{
+	return 0;
+}
+#elif !defined(CONFIG_DUMMY_CONSOLE)
+static int i915_kick_out_vgacon(struct drm_i915_private *dev_priv)
+{
+	return -ENODEV;
+}
+#else
+static int i915_kick_out_vgacon(struct drm_i915_private *dev_priv)
+{
+	int ret = 0;
+
+	DRM_INFO("Replacing VGA console driver\n");
+
+	console_lock();
+	if (con_is_bound(&vga_con))
+		ret = do_take_over_console(&dummy_con, 0, MAX_NR_CONSOLES - 1, 1);
+	if (ret == 0) {
+		ret = do_unregister_con_driver(&vga_con);
+
+		/* Ignore "already unregistered". */
+		if (ret == -ENODEV)
+			ret = 0;
+	}
+	console_unlock();
+
+	return ret;
+}
+#endif
+
 static void i915_dump_device_info(struct drm_i915_private *dev_priv)
 {
 	const struct intel_device_info *info = &dev_priv->info;
@@ -1620,8 +1657,15 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (ret)
 		goto out_regs;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
+	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
+		ret = i915_kick_out_vgacon(dev_priv);
+		if (ret) {
+			DRM_ERROR("failed to remove conflicting VGA console\n");
+			goto out_gtt;
+		}
+
 		i915_kick_out_firmware_fb(dev_priv);
+	}
 
 	pci_set_master(dev->pdev);
 
@@ -1739,8 +1783,8 @@ out_power_well:
 	intel_power_domains_remove(dev_priv);
 	drm_vblank_cleanup(dev);
 out_gem_unload:
-	if (dev_priv->mm.inactive_shrinker.scan_objects)
-		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
+	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
+	unregister_shrinker(&dev_priv->mm.shrinker);
 
 	if (dev->pdev->msi_enabled)
 		pci_disable_msi(dev->pdev);
@@ -1753,8 +1797,6 @@ out_mtrrfree:
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
 	io_mapping_free(dev_priv->gtt.mappable);
 out_gtt:
-	list_del(&dev_priv->gtt.base.global_link);
-	drm_mm_takedown(&dev_priv->gtt.base.mm);
 	dev_priv->gtt.base.cleanup(&dev_priv->gtt.base);
 out_regs:
 	intel_uncore_fini(dev);
@@ -1791,8 +1833,8 @@ int i915_driver_unload(struct drm_device *dev)
 
 	i915_teardown_sysfs(dev);
 
-	if (dev_priv->mm.inactive_shrinker.scan_objects)
-		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
+	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
+	unregister_shrinker(&dev_priv->mm.shrinker);
 
 	io_mapping_free(dev_priv->gtt.mappable);
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
@@ -1843,7 +1885,6 @@ int i915_driver_unload(struct drm_device *dev)
 			i915_free_hws(dev);
 	}
 
-	list_del(&dev_priv->gtt.base.global_link);
 	WARN_ON(!list_empty(&dev_priv->vm_list));
 
 	drm_vblank_cleanup(dev);
@@ -1864,7 +1905,7 @@ int i915_driver_unload(struct drm_device *dev)
 		kmem_cache_destroy(dev_priv->slab);
 
 	pci_dev_put(dev_priv->bridge_dev);
-	kfree(dev->dev_private);
+	kfree(dev_priv);
 
 	return 0;
 }
@@ -1925,6 +1966,8 @@ void i915_driver_postclose(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
+	if (file_priv && file_priv->bsd_ring)
+		file_priv->bsd_ring = NULL;
 	kfree(file_priv);
 }
 
@@ -1978,9 +2021,10 @@ const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_GET_RESET_STATS, i915_get_reset_stats_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_USERPTR, i915_gem_userptr_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
 
-int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls);
+int i915_max_ioctl = ARRAY_SIZE(i915_ioctls);
 
 /*
  * This is really ugly: Because old userspace abused the linux agp interface to
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 82f4d1f47d3b..651e65e051c0 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -36,6 +36,7 @@
 
 #include <linux/console.h>
 #include <linux/module.h>
+#include <linux/pm_runtime.h>
 #include <drm/drm_crtc_helper.h>
 
 static struct drm_driver driver;
@@ -49,12 +50,30 @@ static struct drm_driver driver;
 	.dpll_md_offsets = { DPLL_A_MD_OFFSET, DPLL_B_MD_OFFSET }, \
 	.palette_offsets = { PALETTE_A_OFFSET, PALETTE_B_OFFSET }
 
+#define GEN_CHV_PIPEOFFSETS \
+	.pipe_offsets = { PIPE_A_OFFSET, PIPE_B_OFFSET, \
+			  CHV_PIPE_C_OFFSET }, \
+	.trans_offsets = { TRANSCODER_A_OFFSET, TRANSCODER_B_OFFSET, \
+			   CHV_TRANSCODER_C_OFFSET, }, \
+	.dpll_offsets = { DPLL_A_OFFSET, DPLL_B_OFFSET, \
+			  CHV_DPLL_C_OFFSET }, \
+	.dpll_md_offsets = { DPLL_A_MD_OFFSET, DPLL_B_MD_OFFSET, \
+			     CHV_DPLL_C_MD_OFFSET }, \
+	.palette_offsets = { PALETTE_A_OFFSET, PALETTE_B_OFFSET, \
+			     CHV_PALETTE_C_OFFSET }
+
+#define CURSOR_OFFSETS \
+	.cursor_offsets = { CURSOR_A_OFFSET, CURSOR_B_OFFSET, CHV_CURSOR_C_OFFSET }
+
+#define IVB_CURSOR_OFFSETS \
+	.cursor_offsets = { CURSOR_A_OFFSET, IVB_CURSOR_B_OFFSET, IVB_CURSOR_C_OFFSET }
 
 static const struct intel_device_info intel_i830_info = {
 	.gen = 2, .is_mobile = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_845g_info = {
@@ -62,6 +81,7 @@ static const struct intel_device_info intel_845g_info = {
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i85x_info = {
@@ -71,6 +91,7 @@ static const struct intel_device_info intel_i85x_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i865g_info = {
@@ -78,6 +99,7 @@ static const struct intel_device_info intel_i865g_info = {
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i915g_info = {
@@ -85,6 +107,7 @@ static const struct intel_device_info intel_i915g_info = {
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 static const struct intel_device_info intel_i915gm_info = {
 	.gen = 3, .is_mobile = 1, .num_pipes = 2,
@@ -94,12 +117,14 @@ static const struct intel_device_info intel_i915gm_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 static const struct intel_device_info intel_i945g_info = {
 	.gen = 3, .has_hotplug = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 static const struct intel_device_info intel_i945gm_info = {
 	.gen = 3, .is_i945gm = 1, .is_mobile = 1, .num_pipes = 2,
@@ -109,6 +134,7 @@ static const struct intel_device_info intel_i945gm_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i965g_info = {
@@ -117,6 +143,7 @@ static const struct intel_device_info intel_i965g_info = {
 	.has_overlay = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i965gm_info = {
@@ -126,6 +153,7 @@ static const struct intel_device_info intel_i965gm_info = {
 	.supports_tv = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_g33_info = {
@@ -134,6 +162,7 @@ static const struct intel_device_info intel_g33_info = {
 	.has_overlay = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_g45_info = {
@@ -141,6 +170,7 @@ static const struct intel_device_info intel_g45_info = {
 	.has_pipe_cxsr = 1, .has_hotplug = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_gm45_info = {
@@ -150,6 +180,7 @@ static const struct intel_device_info intel_gm45_info = {
 	.supports_tv = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_pineview_info = {
@@ -157,6 +188,7 @@ static const struct intel_device_info intel_pineview_info = {
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.has_overlay = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ironlake_d_info = {
@@ -164,6 +196,7 @@ static const struct intel_device_info intel_ironlake_d_info = {
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ironlake_m_info = {
@@ -172,6 +205,7 @@ static const struct intel_device_info intel_ironlake_m_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_sandybridge_d_info = {
@@ -181,6 +215,7 @@ static const struct intel_device_info intel_sandybridge_d_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.has_llc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_sandybridge_m_info = {
@@ -190,6 +225,7 @@ static const struct intel_device_info intel_sandybridge_m_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.has_llc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 #define GEN7_FEATURES  \
@@ -203,6 +239,7 @@ static const struct intel_device_info intel_ivybridge_d_info = {
 	GEN7_FEATURES,
 	.is_ivybridge = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ivybridge_m_info = {
@@ -210,6 +247,7 @@ static const struct intel_device_info intel_ivybridge_m_info = {
 	.is_ivybridge = 1,
 	.is_mobile = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ivybridge_q_info = {
@@ -217,6 +255,7 @@ static const struct intel_device_info intel_ivybridge_q_info = {
 	.is_ivybridge = 1,
 	.num_pipes = 0, /* legal, last one wins */
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_valleyview_m_info = {
@@ -228,6 +267,7 @@ static const struct intel_device_info intel_valleyview_m_info = {
 	.has_fbc = 0, /* legal, last one wins */
 	.has_llc = 0, /* legal, last one wins */
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_valleyview_d_info = {
@@ -238,6 +278,7 @@ static const struct intel_device_info intel_valleyview_d_info = {
 	.has_fbc = 0, /* legal, last one wins */
 	.has_llc = 0, /* legal, last one wins */
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_haswell_d_info = {
@@ -247,6 +288,7 @@ static const struct intel_device_info intel_haswell_d_info = {
 	.has_fpga_dbg = 1,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_haswell_m_info = {
@@ -257,6 +299,7 @@ static const struct intel_device_info intel_haswell_m_info = {
 	.has_fpga_dbg = 1,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_broadwell_d_info = {
@@ -267,6 +310,7 @@ static const struct intel_device_info intel_broadwell_d_info = {
 	.has_ddi = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_broadwell_m_info = {
@@ -277,6 +321,40 @@ static const struct intel_device_info intel_broadwell_m_info = {
 	.has_ddi = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
+static const struct intel_device_info intel_broadwell_gt3d_info = {
+	.gen = 8, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.has_llc = 1,
+	.has_ddi = 1,
+	.has_fbc = 1,
+	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
+static const struct intel_device_info intel_broadwell_gt3m_info = {
+	.gen = 8, .is_mobile = 1, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.has_llc = 1,
+	.has_ddi = 1,
+	.has_fbc = 1,
+	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
+static const struct intel_device_info intel_cherryview_info = {
+	.is_preliminary = 1,
+	.gen = 8, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+	.is_valleyview = 1,
+	.display_mmio_offset = VLV_DISPLAY_BASE,
+	GEN_CHV_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 /*
@@ -311,8 +389,11 @@ static const struct intel_device_info intel_broadwell_m_info = {
 	INTEL_HSW_M_IDS(&intel_haswell_m_info), \
 	INTEL_VLV_M_IDS(&intel_valleyview_m_info),	\
 	INTEL_VLV_D_IDS(&intel_valleyview_d_info),	\
-	INTEL_BDW_M_IDS(&intel_broadwell_m_info),	\
-	INTEL_BDW_D_IDS(&intel_broadwell_d_info)
+	INTEL_BDW_GT12M_IDS(&intel_broadwell_m_info),	\
+	INTEL_BDW_GT12D_IDS(&intel_broadwell_d_info),	\
+	INTEL_BDW_GT3M_IDS(&intel_broadwell_gt3m_info),	\
+	INTEL_BDW_GT3D_IDS(&intel_broadwell_gt3d_info), \
+	INTEL_CHV_IDS(&intel_cherryview_info)
 
 static const struct pci_device_id pciidlist[] = {		/* aka */
 	INTEL_PCI_IDS,
@@ -445,18 +526,20 @@ static int i915_drm_freeze(struct drm_device *dev)
 			return error;
 		}
 
-		cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work);
-
 		drm_irq_uninstall(dev);
 		dev_priv->enable_hotplug_processing = false;
+
+		intel_disable_gt_powersave(dev);
+
 		/*
 		 * Disable CRTCs directly since we want to preserve sw state
 		 * for _thaw.
 		 */
-		mutex_lock(&dev->mode_config.mutex);
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		drm_modeset_lock_all(dev);
+		for_each_crtc(dev, crtc) {
 			dev_priv->display.crtc_disable(crtc);
-		mutex_unlock(&dev->mode_config.mutex);
+		}
+		drm_modeset_unlock_all(dev);
 
 		intel_modeset_suspend_hw(dev);
 	}
@@ -519,24 +602,6 @@ void intel_console_resume(struct work_struct *work)
 	console_unlock();
 }
 
-static void intel_resume_hotplug(struct drm_device *dev)
-{
-	struct drm_mode_config *mode_config = &dev->mode_config;
-	struct intel_encoder *encoder;
-
-	mutex_lock(&mode_config->mutex);
-	DRM_DEBUG_KMS("running encoder hotplug functions\n");
-
-	list_for_each_entry(encoder, &mode_config->encoder_list, base.head)
-		if (encoder->hot_plug)
-			encoder->hot_plug(encoder);
-
-	mutex_unlock(&mode_config->mutex);
-
-	/* Just fire off a uevent and let userspace tell us what to do */
-	drm_helper_hpd_irq_event(dev);
-}
-
 static int i915_drm_thaw_early(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -551,7 +616,6 @@ static int i915_drm_thaw_early(struct drm_device *dev)
 static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int error = 0;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET) &&
 	    restore_gtt_mappings) {
@@ -569,12 +633,14 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 		drm_mode_config_reset(dev);
 
 		mutex_lock(&dev->struct_mutex);
-
-		error = i915_gem_init_hw(dev);
+		if (i915_gem_init_hw(dev)) {
+			DRM_ERROR("failed to re-initialize GPU, declaring wedged!\n");
+			atomic_set_mask(I915_WEDGED, &dev_priv->gpu_error.reset_counter);
+		}
 		mutex_unlock(&dev->struct_mutex);
 
 		/* We need working interrupts for modeset enabling ... */
-		drm_irq_install(dev);
+		drm_irq_install(dev, dev->pdev->irq);
 
 		intel_modeset_init_hw(dev);
 
@@ -591,7 +657,7 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 		intel_hpd_init(dev);
 		dev_priv->enable_hotplug_processing = true;
 		/* Config may have changed between suspend and resume */
-		intel_resume_hotplug(dev);
+		drm_helper_hpd_irq_event(dev);
 	}
 
 	intel_opregion_init(dev);
@@ -613,7 +679,7 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 	mutex_unlock(&dev_priv->modeset_restore_lock);
 
 	intel_runtime_pm_put(dev_priv);
-	return error;
+	return 0;
 }
 
 static int i915_drm_thaw(struct drm_device *dev)
@@ -746,18 +812,20 @@ int i915_reset(struct drm_device *dev)
 			return ret;
 		}
 
-		drm_irq_uninstall(dev);
-		drm_irq_install(dev);
+		/*
+		 * FIXME: This races pretty badly against concurrent holders of
+		 * ring interrupts. This is possible since we've started to drop
+		 * dev->struct_mutex in select places when waiting for the gpu.
+		 */
 
-		/* rps/rc6 re-init is necessary to restore state lost after the
-		 * reset and the re-install of drm irq. Skip for ironlake per
+		/*
+		 * rps/rc6 re-init is necessary to restore state lost after the
+		 * reset and the re-install of gt irqs. Skip for ironlake per
 		 * previous concerns that it doesn't respond well to some forms
-		 * of re-init after reset. */
-		if (INTEL_INFO(dev)->gen > 5) {
-			mutex_lock(&dev->struct_mutex);
-			intel_enable_gt_powersave(dev);
-			mutex_unlock(&dev->struct_mutex);
-		}
+		 * of re-init after reset.
+		 */
+		if (INTEL_INFO(dev)->gen > 5)
+			intel_reset_gt_powersave(dev);
 
 		intel_hpd_init(dev);
 	} else {
@@ -891,21 +959,453 @@ static int i915_pm_poweroff(struct device *dev)
 	return i915_drm_freeze(drm_dev);
 }
 
-static int i915_runtime_suspend(struct device *device)
+static int hsw_runtime_suspend(struct drm_i915_private *dev_priv)
+{
+	hsw_enable_pc8(dev_priv);
+
+	return 0;
+}
+
+static int snb_runtime_resume(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	intel_init_pch_refclk(dev);
+
+	return 0;
+}
+
+static int hsw_runtime_resume(struct drm_i915_private *dev_priv)
+{
+	hsw_disable_pc8(dev_priv);
+
+	return 0;
+}
+
+/*
+ * Save all Gunit registers that may be lost after a D3 and a subsequent
+ * S0i[R123] transition. The list of registers needing a save/restore is
+ * defined in the VLV2_S0IXRegs document. This documents marks all Gunit
+ * registers in the following way:
+ * - Driver: saved/restored by the driver
+ * - Punit : saved/restored by the Punit firmware
+ * - No, w/o marking: no need to save/restore, since the register is R/O or
+ *                    used internally by the HW in a way that doesn't depend
+ *                    keeping the content across a suspend/resume.
+ * - Debug : used for debugging
+ *
+ * We save/restore all registers marked with 'Driver', with the following
+ * exceptions:
+ * - Registers out of use, including also registers marked with 'Debug'.
+ *   These have no effect on the driver's operation, so we don't save/restore
+ *   them to reduce the overhead.
+ * - Registers that are fully setup by an initialization function called from
+ *   the resume path. For example many clock gating and RPS/RC6 registers.
+ * - Registers that provide the right functionality with their reset defaults.
+ *
+ * TODO: Except for registers that based on the above 3 criteria can be safely
+ * ignored, we save/restore all others, practically treating the HW context as
+ * a black-box for the driver. Further investigation is needed to reduce the
+ * saved/restored registers even further, by following the same 3 criteria.
+ */
+static void vlv_save_gunit_s0ix_state(struct drm_i915_private *dev_priv)
+{
+	struct vlv_s0ix_state *s = &dev_priv->vlv_s0ix_state;
+	int i;
+
+	/* GAM 0x4000-0x4770 */
+	s->wr_watermark		= I915_READ(GEN7_WR_WATERMARK);
+	s->gfx_prio_ctrl	= I915_READ(GEN7_GFX_PRIO_CTRL);
+	s->arb_mode		= I915_READ(ARB_MODE);
+	s->gfx_pend_tlb0	= I915_READ(GEN7_GFX_PEND_TLB0);
+	s->gfx_pend_tlb1	= I915_READ(GEN7_GFX_PEND_TLB1);
+
+	for (i = 0; i < ARRAY_SIZE(s->lra_limits); i++)
+		s->lra_limits[i] = I915_READ(GEN7_LRA_LIMITS_BASE + i * 4);
+
+	s->media_max_req_count	= I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
+	s->gfx_max_req_count	= I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
+
+	s->render_hwsp		= I915_READ(RENDER_HWS_PGA_GEN7);
+	s->ecochk		= I915_READ(GAM_ECOCHK);
+	s->bsd_hwsp		= I915_READ(BSD_HWS_PGA_GEN7);
+	s->blt_hwsp		= I915_READ(BLT_HWS_PGA_GEN7);
+
+	s->tlb_rd_addr		= I915_READ(GEN7_TLB_RD_ADDR);
+
+	/* MBC 0x9024-0x91D0, 0x8500 */
+	s->g3dctl		= I915_READ(VLV_G3DCTL);
+	s->gsckgctl		= I915_READ(VLV_GSCKGCTL);
+	s->mbctl		= I915_READ(GEN6_MBCTL);
+
+	/* GCP 0x9400-0x9424, 0x8100-0x810C */
+	s->ucgctl1		= I915_READ(GEN6_UCGCTL1);
+	s->ucgctl3		= I915_READ(GEN6_UCGCTL3);
+	s->rcgctl1		= I915_READ(GEN6_RCGCTL1);
+	s->rcgctl2		= I915_READ(GEN6_RCGCTL2);
+	s->rstctl		= I915_READ(GEN6_RSTCTL);
+	s->misccpctl		= I915_READ(GEN7_MISCCPCTL);
+
+	/* GPM 0xA000-0xAA84, 0x8000-0x80FC */
+	s->gfxpause		= I915_READ(GEN6_GFXPAUSE);
+	s->rpdeuhwtc		= I915_READ(GEN6_RPDEUHWTC);
+	s->rpdeuc		= I915_READ(GEN6_RPDEUC);
+	s->ecobus		= I915_READ(ECOBUS);
+	s->pwrdwnupctl		= I915_READ(VLV_PWRDWNUPCTL);
+	s->rp_down_timeout	= I915_READ(GEN6_RP_DOWN_TIMEOUT);
+	s->rp_deucsw		= I915_READ(GEN6_RPDEUCSW);
+	s->rcubmabdtmr		= I915_READ(GEN6_RCUBMABDTMR);
+	s->rcedata		= I915_READ(VLV_RCEDATA);
+	s->spare2gh		= I915_READ(VLV_SPAREG2H);
+
+	/* Display CZ domain, 0x4400C-0x4402C, 0x4F000-0x4F11F */
+	s->gt_imr		= I915_READ(GTIMR);
+	s->gt_ier		= I915_READ(GTIER);
+	s->pm_imr		= I915_READ(GEN6_PMIMR);
+	s->pm_ier		= I915_READ(GEN6_PMIER);
+
+	for (i = 0; i < ARRAY_SIZE(s->gt_scratch); i++)
+		s->gt_scratch[i] = I915_READ(GEN7_GT_SCRATCH_BASE + i * 4);
+
+	/* GT SA CZ domain, 0x100000-0x138124 */
+	s->tilectl		= I915_READ(TILECTL);
+	s->gt_fifoctl		= I915_READ(GTFIFOCTL);
+	s->gtlc_wake_ctrl	= I915_READ(VLV_GTLC_WAKE_CTRL);
+	s->gtlc_survive		= I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	s->pmwgicz		= I915_READ(VLV_PMWGICZ);
+
+	/* Gunit-Display CZ domain, 0x182028-0x1821CF */
+	s->gu_ctl0		= I915_READ(VLV_GU_CTL0);
+	s->gu_ctl1		= I915_READ(VLV_GU_CTL1);
+	s->clock_gate_dis2	= I915_READ(VLV_GUNIT_CLOCK_GATE2);
+
+	/*
+	 * Not saving any of:
+	 * DFT,		0x9800-0x9EC0
+	 * SARB,	0xB000-0xB1FC
+	 * GAC,		0x5208-0x524C, 0x14000-0x14C000
+	 * PCI CFG
+	 */
+}
+
+static void vlv_restore_gunit_s0ix_state(struct drm_i915_private *dev_priv)
+{
+	struct vlv_s0ix_state *s = &dev_priv->vlv_s0ix_state;
+	u32 val;
+	int i;
+
+	/* GAM 0x4000-0x4770 */
+	I915_WRITE(GEN7_WR_WATERMARK,	s->wr_watermark);
+	I915_WRITE(GEN7_GFX_PRIO_CTRL,	s->gfx_prio_ctrl);
+	I915_WRITE(ARB_MODE,		s->arb_mode | (0xffff << 16));
+	I915_WRITE(GEN7_GFX_PEND_TLB0,	s->gfx_pend_tlb0);
+	I915_WRITE(GEN7_GFX_PEND_TLB1,	s->gfx_pend_tlb1);
+
+	for (i = 0; i < ARRAY_SIZE(s->lra_limits); i++)
+		I915_WRITE(GEN7_LRA_LIMITS_BASE + i * 4, s->lra_limits[i]);
+
+	I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->media_max_req_count);
+	I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->gfx_max_req_count);
+
+	I915_WRITE(RENDER_HWS_PGA_GEN7,	s->render_hwsp);
+	I915_WRITE(GAM_ECOCHK,		s->ecochk);
+	I915_WRITE(BSD_HWS_PGA_GEN7,	s->bsd_hwsp);
+	I915_WRITE(BLT_HWS_PGA_GEN7,	s->blt_hwsp);
+
+	I915_WRITE(GEN7_TLB_RD_ADDR,	s->tlb_rd_addr);
+
+	/* MBC 0x9024-0x91D0, 0x8500 */
+	I915_WRITE(VLV_G3DCTL,		s->g3dctl);
+	I915_WRITE(VLV_GSCKGCTL,	s->gsckgctl);
+	I915_WRITE(GEN6_MBCTL,		s->mbctl);
+
+	/* GCP 0x9400-0x9424, 0x8100-0x810C */
+	I915_WRITE(GEN6_UCGCTL1,	s->ucgctl1);
+	I915_WRITE(GEN6_UCGCTL3,	s->ucgctl3);
+	I915_WRITE(GEN6_RCGCTL1,	s->rcgctl1);
+	I915_WRITE(GEN6_RCGCTL2,	s->rcgctl2);
+	I915_WRITE(GEN6_RSTCTL,		s->rstctl);
+	I915_WRITE(GEN7_MISCCPCTL,	s->misccpctl);
+
+	/* GPM 0xA000-0xAA84, 0x8000-0x80FC */
+	I915_WRITE(GEN6_GFXPAUSE,	s->gfxpause);
+	I915_WRITE(GEN6_RPDEUHWTC,	s->rpdeuhwtc);
+	I915_WRITE(GEN6_RPDEUC,		s->rpdeuc);
+	I915_WRITE(ECOBUS,		s->ecobus);
+	I915_WRITE(VLV_PWRDWNUPCTL,	s->pwrdwnupctl);
+	I915_WRITE(GEN6_RP_DOWN_TIMEOUT,s->rp_down_timeout);
+	I915_WRITE(GEN6_RPDEUCSW,	s->rp_deucsw);
+	I915_WRITE(GEN6_RCUBMABDTMR,	s->rcubmabdtmr);
+	I915_WRITE(VLV_RCEDATA,		s->rcedata);
+	I915_WRITE(VLV_SPAREG2H,	s->spare2gh);
+
+	/* Display CZ domain, 0x4400C-0x4402C, 0x4F000-0x4F11F */
+	I915_WRITE(GTIMR,		s->gt_imr);
+	I915_WRITE(GTIER,		s->gt_ier);
+	I915_WRITE(GEN6_PMIMR,		s->pm_imr);
+	I915_WRITE(GEN6_PMIER,		s->pm_ier);
+
+	for (i = 0; i < ARRAY_SIZE(s->gt_scratch); i++)
+		I915_WRITE(GEN7_GT_SCRATCH_BASE + i * 4, s->gt_scratch[i]);
+
+	/* GT SA CZ domain, 0x100000-0x138124 */
+	I915_WRITE(TILECTL,			s->tilectl);
+	I915_WRITE(GTFIFOCTL,			s->gt_fifoctl);
+	/*
+	 * Preserve the GT allow wake and GFX force clock bit, they are not
+	 * be restored, as they are used to control the s0ix suspend/resume
+	 * sequence by the caller.
+	 */
+	val = I915_READ(VLV_GTLC_WAKE_CTRL);
+	val &= VLV_GTLC_ALLOWWAKEREQ;
+	val |= s->gtlc_wake_ctrl & ~VLV_GTLC_ALLOWWAKEREQ;
+	I915_WRITE(VLV_GTLC_WAKE_CTRL, val);
+
+	val = I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	val &= VLV_GFX_CLK_FORCE_ON_BIT;
+	val |= s->gtlc_survive & ~VLV_GFX_CLK_FORCE_ON_BIT;
+	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG, val);
+
+	I915_WRITE(VLV_PMWGICZ,			s->pmwgicz);
+
+	/* Gunit-Display CZ domain, 0x182028-0x1821CF */
+	I915_WRITE(VLV_GU_CTL0,			s->gu_ctl0);
+	I915_WRITE(VLV_GU_CTL1,			s->gu_ctl1);
+	I915_WRITE(VLV_GUNIT_CLOCK_GATE2,	s->clock_gate_dis2);
+}
+
+int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool force_on)
+{
+	u32 val;
+	int err;
+
+	val = I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	WARN_ON(!!(val & VLV_GFX_CLK_FORCE_ON_BIT) == force_on);
+
+#define COND (I915_READ(VLV_GTLC_SURVIVABILITY_REG) & VLV_GFX_CLK_STATUS_BIT)
+	/* Wait for a previous force-off to settle */
+	if (force_on) {
+		err = wait_for(!COND, 20);
+		if (err) {
+			DRM_ERROR("timeout waiting for GFX clock force-off (%08x)\n",
+				  I915_READ(VLV_GTLC_SURVIVABILITY_REG));
+			return err;
+		}
+	}
+
+	val = I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	val &= ~VLV_GFX_CLK_FORCE_ON_BIT;
+	if (force_on)
+		val |= VLV_GFX_CLK_FORCE_ON_BIT;
+	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG, val);
+
+	if (!force_on)
+		return 0;
+
+	err = wait_for(COND, 20);
+	if (err)
+		DRM_ERROR("timeout waiting for GFX clock force-on (%08x)\n",
+			  I915_READ(VLV_GTLC_SURVIVABILITY_REG));
+
+	return err;
+#undef COND
+}
+
+static int vlv_allow_gt_wake(struct drm_i915_private *dev_priv, bool allow)
+{
+	u32 val;
+	int err = 0;
+
+	val = I915_READ(VLV_GTLC_WAKE_CTRL);
+	val &= ~VLV_GTLC_ALLOWWAKEREQ;
+	if (allow)
+		val |= VLV_GTLC_ALLOWWAKEREQ;
+	I915_WRITE(VLV_GTLC_WAKE_CTRL, val);
+	POSTING_READ(VLV_GTLC_WAKE_CTRL);
+
+#define COND (!!(I915_READ(VLV_GTLC_PW_STATUS) & VLV_GTLC_ALLOWWAKEACK) == \
+	      allow)
+	err = wait_for(COND, 1);
+	if (err)
+		DRM_ERROR("timeout disabling GT waking\n");
+	return err;
+#undef COND
+}
+
+static int vlv_wait_for_gt_wells(struct drm_i915_private *dev_priv,
+				 bool wait_for_on)
+{
+	u32 mask;
+	u32 val;
+	int err;
+
+	mask = VLV_GTLC_PW_MEDIA_STATUS_MASK | VLV_GTLC_PW_RENDER_STATUS_MASK;
+	val = wait_for_on ? mask : 0;
+#define COND ((I915_READ(VLV_GTLC_PW_STATUS) & mask) == val)
+	if (COND)
+		return 0;
+
+	DRM_DEBUG_KMS("waiting for GT wells to go %s (%08x)\n",
+			wait_for_on ? "on" : "off",
+			I915_READ(VLV_GTLC_PW_STATUS));
+
+	/*
+	 * RC6 transitioning can be delayed up to 2 msec (see
+	 * valleyview_enable_rps), use 3 msec for safety.
+	 */
+	err = wait_for(COND, 3);
+	if (err)
+		DRM_ERROR("timeout waiting for GT wells to go %s\n",
+			  wait_for_on ? "on" : "off");
+
+	return err;
+#undef COND
+}
+
+static void vlv_check_no_gt_access(struct drm_i915_private *dev_priv)
+{
+	if (!(I915_READ(VLV_GTLC_PW_STATUS) & VLV_GTLC_ALLOWWAKEERR))
+		return;
+
+	DRM_ERROR("GT register access while GT waking disabled\n");
+	I915_WRITE(VLV_GTLC_PW_STATUS, VLV_GTLC_ALLOWWAKEERR);
+}
+
+static int vlv_runtime_suspend(struct drm_i915_private *dev_priv)
+{
+	u32 mask;
+	int err;
+
+	/*
+	 * Bspec defines the following GT well on flags as debug only, so
+	 * don't treat them as hard failures.
+	 */
+	(void)vlv_wait_for_gt_wells(dev_priv, false);
+
+	mask = VLV_GTLC_RENDER_CTX_EXISTS | VLV_GTLC_MEDIA_CTX_EXISTS;
+	WARN_ON((I915_READ(VLV_GTLC_WAKE_CTRL) & mask) != mask);
+
+	vlv_check_no_gt_access(dev_priv);
+
+	err = vlv_force_gfx_clock(dev_priv, true);
+	if (err)
+		goto err1;
+
+	err = vlv_allow_gt_wake(dev_priv, false);
+	if (err)
+		goto err2;
+	vlv_save_gunit_s0ix_state(dev_priv);
+
+	err = vlv_force_gfx_clock(dev_priv, false);
+	if (err)
+		goto err2;
+
+	return 0;
+
+err2:
+	/* For safety always re-enable waking and disable gfx clock forcing */
+	vlv_allow_gt_wake(dev_priv, true);
+err1:
+	vlv_force_gfx_clock(dev_priv, false);
+
+	return err;
+}
+
+static int vlv_runtime_resume(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	int err;
+	int ret;
+
+	/*
+	 * If any of the steps fail just try to continue, that's the best we
+	 * can do at this point. Return the first error code (which will also
+	 * leave RPM permanently disabled).
+	 */
+	ret = vlv_force_gfx_clock(dev_priv, true);
+
+	vlv_restore_gunit_s0ix_state(dev_priv);
+
+	err = vlv_allow_gt_wake(dev_priv, true);
+	if (!ret)
+		ret = err;
+
+	err = vlv_force_gfx_clock(dev_priv, false);
+	if (!ret)
+		ret = err;
+
+	vlv_check_no_gt_access(dev_priv);
+
+	intel_init_clock_gating(dev);
+	i915_gem_restore_fences(dev);
+
+	return ret;
+}
+
+static int intel_runtime_suspend(struct device *device)
 {
 	struct pci_dev *pdev = to_pci_dev(device);
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	if (WARN_ON_ONCE(!(dev_priv->rps.enabled && intel_enable_rc6(dev))))
+		return -ENODEV;
 
 	WARN_ON(!HAS_RUNTIME_PM(dev));
 	assert_force_wake_inactive(dev_priv);
 
 	DRM_DEBUG_KMS("Suspending device\n");
 
-	if (HAS_PC8(dev))
-		hsw_enable_pc8(dev_priv);
+	/*
+	 * We could deadlock here in case another thread holding struct_mutex
+	 * calls RPM suspend concurrently, since the RPM suspend will wait
+	 * first for this RPM suspend to finish. In this case the concurrent
+	 * RPM resume will be followed by its RPM suspend counterpart. Still
+	 * for consistency return -EAGAIN, which will reschedule this suspend.
+	 */
+	if (!mutex_trylock(&dev->struct_mutex)) {
+		DRM_DEBUG_KMS("device lock contention, deffering suspend\n");
+		/*
+		 * Bump the expiration timestamp, otherwise the suspend won't
+		 * be rescheduled.
+		 */
+		pm_runtime_mark_last_busy(device);
 
+		return -EAGAIN;
+	}
+	/*
+	 * We are safe here against re-faults, since the fault handler takes
+	 * an RPM reference.
+	 */
 	i915_gem_release_all_mmaps(dev_priv);
+	mutex_unlock(&dev->struct_mutex);
+
+	/*
+	 * rps.work can't be rearmed here, since we get here only after making
+	 * sure the GPU is idle and the RPS freq is set to the minimum. See
+	 * intel_mark_idle().
+	 */
+	cancel_work_sync(&dev_priv->rps.work);
+	intel_runtime_pm_disable_interrupts(dev);
+
+	if (IS_GEN6(dev)) {
+		ret = 0;
+	} else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
+		ret = hsw_runtime_suspend(dev_priv);
+	} else if (IS_VALLEYVIEW(dev)) {
+		ret = vlv_runtime_suspend(dev_priv);
+	} else {
+		ret = -ENODEV;
+		WARN_ON(1);
+	}
+
+	if (ret) {
+		DRM_ERROR("Runtime suspend failed, disabling it (%d)\n", ret);
+		intel_runtime_pm_restore_interrupts(dev);
+
+		return ret;
+	}
 
 	del_timer_sync(&dev_priv->gpu_error.hangcheck_timer);
 	dev_priv->pm.suspended = true;
@@ -923,11 +1423,12 @@ static int i915_runtime_suspend(struct device *device)
 	return 0;
 }
 
-static int i915_runtime_resume(struct device *device)
+static int intel_runtime_resume(struct device *device)
 {
 	struct pci_dev *pdev = to_pci_dev(device);
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
 
 	WARN_ON(!HAS_RUNTIME_PM(dev));
 
@@ -936,11 +1437,33 @@ static int i915_runtime_resume(struct device *device)
 	intel_opregion_notify_adapter(dev, PCI_D0);
 	dev_priv->pm.suspended = false;
 
-	if (HAS_PC8(dev))
-		hsw_disable_pc8(dev_priv);
+	if (IS_GEN6(dev)) {
+		ret = snb_runtime_resume(dev_priv);
+	} else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
+		ret = hsw_runtime_resume(dev_priv);
+	} else if (IS_VALLEYVIEW(dev)) {
+		ret = vlv_runtime_resume(dev_priv);
+	} else {
+		WARN_ON(1);
+		ret = -ENODEV;
+	}
 
-	DRM_DEBUG_KMS("Device resumed\n");
-	return 0;
+	/*
+	 * No point of rolling back things in case of an error, as the best
+	 * we can do is to hope that things will still work (and disable RPM).
+	 */
+	i915_gem_init_swizzling(dev);
+	gen6_update_ring_freq(dev);
+
+	intel_runtime_pm_restore_interrupts(dev);
+	intel_reset_gt_powersave(dev);
+
+	if (ret)
+		DRM_ERROR("Runtime resume failed, disabling it (%d)\n", ret);
+	else
+		DRM_DEBUG_KMS("Device resumed\n");
+
+	return ret;
 }
 
 static const struct dev_pm_ops i915_pm_ops = {
@@ -954,8 +1477,8 @@ static const struct dev_pm_ops i915_pm_ops = {
 	.poweroff = i915_pm_poweroff,
 	.restore_early = i915_pm_resume_early,
 	.restore = i915_pm_resume,
-	.runtime_suspend = i915_runtime_suspend,
-	.runtime_resume = i915_runtime_resume,
+	.runtime_suspend = intel_runtime_suspend,
+	.runtime_resume = intel_runtime_resume,
 };
 
 static const struct vm_operations_struct i915_gem_vm_ops = {
@@ -1062,6 +1585,7 @@ static int __init i915_init(void)
 		driver.get_vblank_timestamp = NULL;
 #ifndef CONFIG_DRM_I915_UMS
 		/* Silently fail loading to not upset userspace. */
+		DRM_DEBUG_DRIVER("KMS and UMS disabled.\n");
 		return 0;
 #endif
 	}
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 388c028e223c..374f964323ad 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -35,11 +35,13 @@
 #include "i915_reg.h"
 #include "intel_bios.h"
 #include "intel_ringbuffer.h"
+#include "i915_gem_gtt.h"
 #include <linux/io-mapping.h>
 #include <linux/i2c.h>
 #include <linux/i2c-algo-bit.h>
 #include <drm/intel-gtt.h>
 #include <linux/backlight.h>
+#include <linux/hashtable.h>
 #include <linux/intel-iommu.h>
 #include <linux/kref.h>
 #include <linux/pm_qos.h>
@@ -91,7 +93,7 @@ enum port {
 };
 #define port_name(p) ((p) + 'A')
 
-#define I915_NUM_PHYS_VLV 1
+#define I915_NUM_PHYS_VLV 2
 
 enum dpio_channel {
 	DPIO_CH0,
@@ -162,6 +164,12 @@ enum hpd_pin {
 #define for_each_pipe(p) for ((p) = 0; (p) < INTEL_INFO(dev)->num_pipes; (p)++)
 #define for_each_sprite(p, s) for ((s) = 0; (s) < INTEL_INFO(dev)->num_sprites[(p)]; (s)++)
 
+#define for_each_crtc(dev, crtc) \
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+
+#define for_each_intel_crtc(dev, intel_crtc) \
+	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head)
+
 #define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \
 	list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \
 		if ((intel_encoder)->base.crtc == (__crtc))
@@ -171,6 +179,7 @@ enum hpd_pin {
 		if ((intel_connector)->base.encoder == (__encoder))
 
 struct drm_i915_private;
+struct i915_mmu_object;
 
 enum intel_dpll_id {
 	DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */
@@ -312,7 +321,6 @@ struct drm_i915_error_state {
 	u32 gab_ctl;
 	u32 gfx_mode;
 	u32 extra_instdone[I915_NUM_INSTDONE_REG];
-	u32 pipestat[I915_MAX_PIPES];
 	u64 fence[I915_MAX_NUM_FENCES];
 	struct intel_overlay_error_state *overlay;
 	struct intel_display_error_state *display;
@@ -346,7 +354,7 @@ struct drm_i915_error_state {
 		u64 bbaddr;
 		u64 acthd;
 		u32 fault_reg;
-		u32 faddr;
+		u64 faddr;
 		u32 rc_psmi; /* sleep state */
 		u32 semaphore_mboxes[I915_NUM_RINGS - 1];
 
@@ -385,6 +393,7 @@ struct drm_i915_error_state {
 		u32 tiling:2;
 		u32 dirty:1;
 		u32 purgeable:1;
+		u32 userptr:1;
 		s32 ring:4;
 		u32 cache_level:3;
 	} **active_bo, **pinned_bo;
@@ -449,10 +458,11 @@ struct drm_i915_display_funcs {
 	int (*queue_flip)(struct drm_device *dev, struct drm_crtc *crtc,
 			  struct drm_framebuffer *fb,
 			  struct drm_i915_gem_object *obj,
+			  struct intel_engine_cs *ring,
 			  uint32_t flags);
-	int (*update_primary_plane)(struct drm_crtc *crtc,
-				    struct drm_framebuffer *fb,
-				    int x, int y);
+	void (*update_primary_plane)(struct drm_crtc *crtc,
+				     struct drm_framebuffer *fb,
+				     int x, int y);
 	void (*hpd_irq_setup)(struct drm_device *dev);
 	/* clock updates for mode set */
 	/* cursor updates */
@@ -545,6 +555,7 @@ struct intel_device_info {
 	int dpll_offsets[I915_MAX_PIPES];
 	int dpll_md_offsets[I915_MAX_PIPES];
 	int palette_offsets[I915_MAX_PIPES];
+	int cursor_offsets[I915_MAX_PIPES];
 };
 
 #undef DEFINE_FLAG
@@ -560,168 +571,6 @@ enum i915_cache_level {
 	I915_CACHE_WT, /* hsw:gt3e WriteThrough for scanouts */
 };
 
-typedef uint32_t gen6_gtt_pte_t;
-
-/**
- * A VMA represents a GEM BO that is bound into an address space. Therefore, a
- * VMA's presence cannot be guaranteed before binding, or after unbinding the
- * object into/from the address space.
- *
- * To make things as simple as possible (ie. no refcounting), a VMA's lifetime
- * will always be <= an objects lifetime. So object refcounting should cover us.
- */
-struct i915_vma {
-	struct drm_mm_node node;
-	struct drm_i915_gem_object *obj;
-	struct i915_address_space *vm;
-
-	/** This object's place on the active/inactive lists */
-	struct list_head mm_list;
-
-	struct list_head vma_link; /* Link in the object's VMA list */
-
-	/** This vma's place in the batchbuffer or on the eviction list */
-	struct list_head exec_list;
-
-	/**
-	 * Used for performing relocations during execbuffer insertion.
-	 */
-	struct hlist_node exec_node;
-	unsigned long exec_handle;
-	struct drm_i915_gem_exec_object2 *exec_entry;
-
-	/**
-	 * How many users have pinned this object in GTT space. The following
-	 * users can each hold at most one reference: pwrite/pread, pin_ioctl
-	 * (via user_pin_count), execbuffer (objects are not allowed multiple
-	 * times for the same batchbuffer), and the framebuffer code. When
-	 * switching/pageflipping, the framebuffer code has at most two buffers
-	 * pinned per crtc.
-	 *
-	 * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3
-	 * bits with absolutely no headroom. So use 4 bits. */
-	unsigned int pin_count:4;
-#define DRM_I915_GEM_OBJECT_MAX_PIN_COUNT 0xf
-
-	/** Unmap an object from an address space. This usually consists of
-	 * setting the valid PTE entries to a reserved scratch page. */
-	void (*unbind_vma)(struct i915_vma *vma);
-	/* Map an object into an address space with the given cache flags. */
-#define GLOBAL_BIND (1<<0)
-	void (*bind_vma)(struct i915_vma *vma,
-			 enum i915_cache_level cache_level,
-			 u32 flags);
-};
-
-struct i915_address_space {
-	struct drm_mm mm;
-	struct drm_device *dev;
-	struct list_head global_link;
-	unsigned long start;		/* Start offset always 0 for dri2 */
-	size_t total;		/* size addr space maps (ex. 2GB for ggtt) */
-
-	struct {
-		dma_addr_t addr;
-		struct page *page;
-	} scratch;
-
-	/**
-	 * List of objects currently involved in rendering.
-	 *
-	 * Includes buffers having the contents of their GPU caches
-	 * flushed, not necessarily primitives.  last_rendering_seqno
-	 * represents when the rendering involved will be completed.
-	 *
-	 * A reference is held on the buffer while on this list.
-	 */
-	struct list_head active_list;
-
-	/**
-	 * LRU list of objects which are not in the ringbuffer and
-	 * are ready to unbind, but are still in the GTT.
-	 *
-	 * last_rendering_seqno is 0 while an object is in this list.
-	 *
-	 * A reference is not held on the buffer while on this list,
-	 * as merely being GTT-bound shouldn't prevent its being
-	 * freed, and we'll pull it off the list in the free path.
-	 */
-	struct list_head inactive_list;
-
-	/* FIXME: Need a more generic return type */
-	gen6_gtt_pte_t (*pte_encode)(dma_addr_t addr,
-				     enum i915_cache_level level,
-				     bool valid); /* Create a valid PTE */
-	void (*clear_range)(struct i915_address_space *vm,
-			    uint64_t start,
-			    uint64_t length,
-			    bool use_scratch);
-	void (*insert_entries)(struct i915_address_space *vm,
-			       struct sg_table *st,
-			       uint64_t start,
-			       enum i915_cache_level cache_level);
-	void (*cleanup)(struct i915_address_space *vm);
-};
-
-/* The Graphics Translation Table is the way in which GEN hardware translates a
- * Graphics Virtual Address into a Physical Address. In addition to the normal
- * collateral associated with any va->pa translations GEN hardware also has a
- * portion of the GTT which can be mapped by the CPU and remain both coherent
- * and correct (in cases like swizzling). That region is referred to as GMADR in
- * the spec.
- */
-struct i915_gtt {
-	struct i915_address_space base;
-	size_t stolen_size;		/* Total size of stolen memory */
-
-	unsigned long mappable_end;	/* End offset that we can CPU map */
-	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
-	phys_addr_t mappable_base;	/* PA of our GMADR */
-
-	/** "Graphics Stolen Memory" holds the global PTEs */
-	void __iomem *gsm;
-
-	bool do_idle_maps;
-
-	int mtrr;
-
-	/* global gtt ops */
-	int (*gtt_probe)(struct drm_device *dev, size_t *gtt_total,
-			  size_t *stolen, phys_addr_t *mappable_base,
-			  unsigned long *mappable_end);
-};
-#define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
-
-#define GEN8_LEGACY_PDPS 4
-struct i915_hw_ppgtt {
-	struct i915_address_space base;
-	struct kref ref;
-	struct drm_mm_node node;
-	unsigned num_pd_entries;
-	unsigned num_pd_pages; /* gen8+ */
-	union {
-		struct page **pt_pages;
-		struct page **gen8_pt_pages[GEN8_LEGACY_PDPS];
-	};
-	struct page *pd_pages;
-	union {
-		uint32_t pd_offset;
-		dma_addr_t pd_dma_addr[GEN8_LEGACY_PDPS];
-	};
-	union {
-		dma_addr_t *pt_dma_addr;
-		dma_addr_t *gen8_pt_dma_addr[4];
-	};
-
-	struct i915_hw_context *ctx;
-
-	int (*enable)(struct i915_hw_ppgtt *ppgtt);
-	int (*switch_mm)(struct i915_hw_ppgtt *ppgtt,
-			 struct intel_ring_buffer *ring,
-			 bool synchronous);
-	void (*debug_dump)(struct i915_hw_ppgtt *ppgtt, struct seq_file *m);
-};
-
 struct i915_ctx_hang_stats {
 	/* This context had batch pending when hang was declared */
 	unsigned batch_pending;
@@ -738,13 +587,13 @@ struct i915_ctx_hang_stats {
 
 /* This must match up with the value previously used for execbuf2.rsvd1. */
 #define DEFAULT_CONTEXT_ID 0
-struct i915_hw_context {
+struct intel_context {
 	struct kref ref;
 	int id;
 	bool is_initialized;
 	uint8_t remap_slice;
 	struct drm_i915_file_private *file_priv;
-	struct intel_ring_buffer *last_ring;
+	struct intel_engine_cs *last_ring;
 	struct drm_i915_gem_object *obj;
 	struct i915_ctx_hang_stats hang_stats;
 	struct i915_address_space *vm;
@@ -782,6 +631,10 @@ struct i915_fbc {
 	} no_fbc_reason;
 };
 
+struct i915_drrs {
+	struct intel_connector *connector;
+};
+
 struct i915_psr {
 	bool sink_support;
 	bool source_ok;
@@ -803,6 +656,7 @@ enum intel_sbi_destination {
 #define QUIRK_PIPEA_FORCE (1<<0)
 #define QUIRK_LVDS_SSC_DISABLE (1<<1)
 #define QUIRK_INVERT_BRIGHTNESS (1<<2)
+#define QUIRK_BACKLIGHT_PRESENT (1<<3)
 
 struct intel_fbdev;
 struct intel_fbc_work;
@@ -965,6 +819,67 @@ struct i915_suspend_saved_registers {
 	u32 savePCH_PORT_HOTPLUG;
 };
 
+struct vlv_s0ix_state {
+	/* GAM */
+	u32 wr_watermark;
+	u32 gfx_prio_ctrl;
+	u32 arb_mode;
+	u32 gfx_pend_tlb0;
+	u32 gfx_pend_tlb1;
+	u32 lra_limits[GEN7_LRA_LIMITS_REG_NUM];
+	u32 media_max_req_count;
+	u32 gfx_max_req_count;
+	u32 render_hwsp;
+	u32 ecochk;
+	u32 bsd_hwsp;
+	u32 blt_hwsp;
+	u32 tlb_rd_addr;
+
+	/* MBC */
+	u32 g3dctl;
+	u32 gsckgctl;
+	u32 mbctl;
+
+	/* GCP */
+	u32 ucgctl1;
+	u32 ucgctl3;
+	u32 rcgctl1;
+	u32 rcgctl2;
+	u32 rstctl;
+	u32 misccpctl;
+
+	/* GPM */
+	u32 gfxpause;
+	u32 rpdeuhwtc;
+	u32 rpdeuc;
+	u32 ecobus;
+	u32 pwrdwnupctl;
+	u32 rp_down_timeout;
+	u32 rp_deucsw;
+	u32 rcubmabdtmr;
+	u32 rcedata;
+	u32 spare2gh;
+
+	/* Display 1 CZ domain */
+	u32 gt_imr;
+	u32 gt_ier;
+	u32 pm_imr;
+	u32 pm_ier;
+	u32 gt_scratch[GEN7_GT_SCRATCH_REG_NUM];
+
+	/* GT SA CZ domain */
+	u32 tilectl;
+	u32 gt_fifoctl;
+	u32 gtlc_wake_ctrl;
+	u32 gtlc_survive;
+	u32 pmwgicz;
+
+	/* Display 2 CZ domain */
+	u32 gu_ctl0;
+	u32 gu_ctl1;
+	u32 clock_gate_dis2;
+};
+
 struct intel_gen6_power_mgmt {
 	/* work and pm_iir are protected by dev_priv->irq_lock */
 	struct work_struct work;
@@ -1063,6 +978,8 @@ struct i915_power_well {
 	bool always_on;
 	/* power well enable/disable usage count */
 	int count;
+	/* cached hw enabled state */
+	bool hw_enabled;
 	unsigned long domains;
 	unsigned long data;
 	const struct i915_power_well_ops *ops;
@@ -1074,6 +991,7 @@ struct i915_power_domains {
 	 * time are on. They are kept on until after the first modeset.
 	 */
 	bool init_power_on;
+	bool initializing;
 	int power_well_count;
 
 	struct mutex lock;
@@ -1132,7 +1050,8 @@ struct i915_gem_mm {
 	/** PPGTT used for aliasing the PPGTT with the GTT */
 	struct i915_hw_ppgtt *aliasing_ppgtt;
 
-	struct shrinker inactive_shrinker;
+	struct notifier_block oom_notifier;
+	struct shrinker shrinker;
 	bool shrinker_no_lock_stealing;
 
 	/** LRU list of objects with fence regs on them. */
@@ -1170,6 +1089,9 @@ struct i915_gem_mm {
 	 */
 	bool busy;
 
+	/* the indicator for dispatch video commands on two BSD rings */
+	int bsd_ring_dispatch_index;
+
 	/** Bit 6 swizzling required for X tiling */
 	uint32_t bit_6_swizzle_x;
 	/** Bit 6 swizzling required for Y tiling */
@@ -1245,8 +1167,12 @@ struct i915_gpu_error {
 	 */
 	wait_queue_head_t reset_queue;
 
-	/* For gpu hang simulation. */
-	unsigned int stop_rings;
+	/* Userspace knobs for gpu hang simulation;
+	 * combines both a ring mask, and extra flags
+	 */
+	u32 stop_rings;
+#define I915_STOP_RING_ALLOW_BAN       (1 << 31)
+#define I915_STOP_RING_ALLOW_WARN      (1 << 30)
 
 	/* For missed irq/seqno simulation. */
 	unsigned int test_irq_rings;
@@ -1266,6 +1192,12 @@ struct ddi_vbt_port_info {
 	uint8_t supports_dp:1;
 };
 
+enum drrs_support_type {
+	DRRS_NOT_SUPPORTED = 0,
+	STATIC_DRRS_SUPPORT = 1,
+	SEAMLESS_DRRS_SUPPORT = 2
+};
+
 struct intel_vbt_data {
 	struct drm_display_mode *lfp_lvds_vbt_mode; /* if any */
 	struct drm_display_mode *sdvo_lvds_vbt_mode; /* if any */
@@ -1278,9 +1210,12 @@ struct intel_vbt_data {
 	unsigned int lvds_use_ssc:1;
 	unsigned int display_clock_mode:1;
 	unsigned int fdi_rx_polarity_inverted:1;
+	unsigned int has_mipi:1;
 	int lvds_ssc_freq;
 	unsigned int bios_lvds_val; /* initial [PCH_]LVDS reg val in VBIOS */
 
+	enum drrs_support_type drrs_type;
+
 	/* eDP */
 	int edp_rate;
 	int edp_lanes;
@@ -1299,7 +1234,14 @@ struct intel_vbt_data {
 
 	/* MIPI DSI */
 	struct {
+		u16 port;
 		u16 panel_id;
+		struct mipi_config *config;
+		struct mipi_pps_data *pps;
+		u8 seq_version;
+		u32 size;
+		u8 *data;
+		u8 *sequence[MIPI_SEQ_MAX];
 	} dsi;
 
 	int crt_ddc_pin;
@@ -1351,23 +1293,13 @@ struct ilk_wm_values {
  * goes back to false exactly before we reenable the IRQs. We use this variable
  * to check if someone is trying to enable/disable IRQs while they're supposed
  * to be disabled. This shouldn't happen and we'll print some error messages in
- * case it happens, but if it actually happens we'll also update the variables
- * inside struct regsave so when we restore the IRQs they will contain the
- * latest expected values.
+ * case it happens.
  *
  * For more, read the Documentation/power/runtime_pm.txt.
  */
 struct i915_runtime_pm {
 	bool suspended;
 	bool irqs_disabled;
-
-	struct {
-		uint32_t deimr;
-		uint32_t sdeimr;
-		uint32_t gtimr;
-		uint32_t gtier;
-		uint32_t gen6_pmimr;
-	} regsave;
 };
 
 enum intel_pipe_crc_source {
@@ -1400,7 +1332,7 @@ struct intel_pipe_crc {
 	wait_queue_head_t wq;
 };
 
-typedef struct drm_i915_private {
+struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *slab;
 
@@ -1424,10 +1356,13 @@ typedef struct drm_i915_private {
 	 */
 	uint32_t gpio_mmio_base;
 
+	/* MMIO base address for MIPI regs */
+	uint32_t mipi_mmio_base;
+
 	wait_queue_head_t gmbus_wait_queue;
 
 	struct pci_dev *bridge_dev;
-	struct intel_ring_buffer ring[I915_NUM_RINGS];
+	struct intel_engine_cs ring[I915_NUM_RINGS];
 	uint32_t last_seqno, next_seqno;
 
 	drm_dma_handle_t *status_page_dmah;
@@ -1469,6 +1404,7 @@ typedef struct drm_i915_private {
 	struct timer_list hotplug_reenable_timer;
 
 	struct i915_fbc fbc;
+	struct i915_drrs drrs;
 	struct intel_opregion opregion;
 	struct intel_vbt_data vbt;
 
@@ -1486,6 +1422,7 @@ typedef struct drm_i915_private {
 	int num_fence_regs; /* 8 on pre-965, 16 otherwise */
 
 	unsigned int fsb_freq, mem_freq, is_ddr3;
+	unsigned int vlv_cdclk_freq;
 
 	/**
 	 * wq - Driver workqueue for GEM.
@@ -1509,9 +1446,12 @@ typedef struct drm_i915_private {
 	struct mutex modeset_restore_lock;
 
 	struct list_head vm_list; /* Global list of all address spaces */
-	struct i915_gtt gtt; /* VMA representing the global address space */
+	struct i915_gtt gtt; /* VM representing the global address space */
 
 	struct i915_gem_mm mm;
+#if defined(CONFIG_MMU_NOTIFIER)
+	DECLARE_HASHTABLE(mmu_notifiers, 7);
+#endif
 
 	/* Kernel Modesetting */
 
@@ -1580,6 +1520,7 @@ typedef struct drm_i915_private {
 
 	u32 suspend_count;
 	struct i915_suspend_saved_registers regfile;
+	struct vlv_s0ix_state vlv_s0ix_state;
 
 	struct {
 		/*
@@ -1605,7 +1546,12 @@ typedef struct drm_i915_private {
 	struct i915_dri1_state dri1;
 	/* Old ums support infrastructure, same warning applies. */
 	struct i915_ums_state ums;
-} drm_i915_private_t;
+
+	/*
+	 * NOTE: This is the dri1/ums dungeon, don't add stuff here. Your patch
+	 * will be rejected. Instead look for a better place.
+	 */
+};
 
 static inline struct drm_i915_private *to_i915(const struct drm_device *dev)
 {
@@ -1642,6 +1588,8 @@ struct drm_i915_gem_object_ops {
 	 */
 	int (*get_pages)(struct drm_i915_gem_object *);
 	void (*put_pages)(struct drm_i915_gem_object *);
+	int (*dmabuf_export)(struct drm_i915_gem_object *);
+	void (*release)(struct drm_i915_gem_object *);
 };
 
 struct drm_i915_gem_object {
@@ -1732,7 +1680,7 @@ struct drm_i915_gem_object {
 	void *dma_buf_vmapping;
 	int vmapping_count;
 
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 
 	/** Breadcrumb of last rendering to the buffer. */
 	uint32_t last_read_seqno;
@@ -1755,8 +1703,20 @@ struct drm_i915_gem_object {
 
 	/** for phy allocated objects */
 	drm_dma_handle_t *phys_handle;
-};
 
+	union {
+		struct i915_gem_userptr {
+			uintptr_t ptr;
+			unsigned read_only :1;
+			unsigned workers :4;
+#define I915_GEM_USERPTR_MAX_WORKERS 15
+
+			struct mm_struct *mm;
+			struct i915_mmu_object *mn;
+			struct work_struct *work;
+		} userptr;
+	};
+};
 #define to_intel_bo(x) container_of(x, struct drm_i915_gem_object, base)
 
 /**
@@ -1771,7 +1731,7 @@ struct drm_i915_gem_object {
  */
 struct drm_i915_gem_request {
 	/** On Which ring this request was generated */
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 
 	/** GEM sequence number associated with this request. */
 	uint32_t seqno;
@@ -1783,7 +1743,7 @@ struct drm_i915_gem_request {
 	u32 tail;
 
 	/** Context related to this request */
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 
 	/** Batch buffer related to this request if any */
 	struct drm_i915_gem_object *batch_obj;
@@ -1810,8 +1770,8 @@ struct drm_i915_file_private {
 	} mm;
 	struct idr context_idr;
 
-	struct i915_hw_context *private_default_ctx;
 	atomic_t rps_wait_boost;
+	struct  intel_engine_cs *bsd_ring;
 };
 
 /*
@@ -1879,11 +1839,17 @@ struct drm_i915_cmd_descriptor {
 	 * the expected value, the parser rejects it. Only valid if flags has
 	 * the CMD_DESC_BITMASK bit set. Only entries where mask is non-zero
 	 * are valid.
+	 *
+	 * If the check specifies a non-zero condition_mask then the parser
+	 * only performs the check when the bits specified by condition_mask
+	 * are non-zero.
 	 */
 	struct {
 		u32 offset;
 		u32 mask;
 		u32 expected;
+		u32 condition_offset;
+		u32 condition_mask;
 	} bits[MAX_CMD_DESC_BITMASKS];
 };
 
@@ -1925,8 +1891,9 @@ struct drm_i915_cmd_table {
 				 (dev)->pdev->device == 0x0106 || \
 				 (dev)->pdev->device == 0x010A)
 #define IS_VALLEYVIEW(dev)	(INTEL_INFO(dev)->is_valleyview)
+#define IS_CHERRYVIEW(dev)	(INTEL_INFO(dev)->is_valleyview && IS_GEN8(dev))
 #define IS_HASWELL(dev)	(INTEL_INFO(dev)->is_haswell)
-#define IS_BROADWELL(dev)	(INTEL_INFO(dev)->gen == 8)
+#define IS_BROADWELL(dev)	(!INTEL_INFO(dev)->is_valleyview && IS_GEN8(dev))
 #define IS_MOBILE(dev)		(INTEL_INFO(dev)->is_mobile)
 #define IS_HSW_EARLY_SDV(dev)	(IS_HASWELL(dev) && \
 				 ((dev)->pdev->device & 0xFF00) == 0x0C00)
@@ -1962,17 +1929,21 @@ struct drm_i915_cmd_table {
 #define BSD_RING		(1<<VCS)
 #define BLT_RING		(1<<BCS)
 #define VEBOX_RING		(1<<VECS)
-#define HAS_BSD(dev)            (INTEL_INFO(dev)->ring_mask & BSD_RING)
-#define HAS_BLT(dev)            (INTEL_INFO(dev)->ring_mask & BLT_RING)
-#define HAS_VEBOX(dev)            (INTEL_INFO(dev)->ring_mask & VEBOX_RING)
-#define HAS_LLC(dev)            (INTEL_INFO(dev)->has_llc)
-#define HAS_WT(dev)            (IS_HASWELL(dev) && to_i915(dev)->ellc_size)
+#define BSD2_RING		(1<<VCS2)
+#define HAS_BSD(dev)		(INTEL_INFO(dev)->ring_mask & BSD_RING)
+#define HAS_BSD2(dev)		(INTEL_INFO(dev)->ring_mask & BSD2_RING)
+#define HAS_BLT(dev)		(INTEL_INFO(dev)->ring_mask & BLT_RING)
+#define HAS_VEBOX(dev)		(INTEL_INFO(dev)->ring_mask & VEBOX_RING)
+#define HAS_LLC(dev)		(INTEL_INFO(dev)->has_llc)
+#define HAS_WT(dev)		((IS_HASWELL(dev) || IS_BROADWELL(dev)) && \
+				 to_i915(dev)->ellc_size)
 #define I915_NEED_GFX_HWS(dev)	(INTEL_INFO(dev)->need_gfx_hws)
 
 #define HAS_HW_CONTEXTS(dev)	(INTEL_INFO(dev)->gen >= 6)
-#define HAS_ALIASING_PPGTT(dev)	(INTEL_INFO(dev)->gen >= 6 && !IS_VALLEYVIEW(dev))
-#define HAS_PPGTT(dev)		(INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev) \
-				 && !IS_BROADWELL(dev))
+#define HAS_ALIASING_PPGTT(dev)	(INTEL_INFO(dev)->gen >= 6 && \
+				 (!IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)))
+#define HAS_PPGTT(dev)		(INTEL_INFO(dev)->gen >= 7 \
+				 && !IS_GEN8(dev))
 #define USES_PPGTT(dev)		intel_enable_ppgtt(dev, false)
 #define USES_FULL_PPGTT(dev)	intel_enable_ppgtt(dev, true)
 
@@ -2010,8 +1981,8 @@ struct drm_i915_cmd_table {
 #define HAS_DDI(dev)		(INTEL_INFO(dev)->has_ddi)
 #define HAS_FPGA_DBG_UNCLAIMED(dev)	(INTEL_INFO(dev)->has_fpga_dbg)
 #define HAS_PSR(dev)		(IS_HASWELL(dev) || IS_BROADWELL(dev))
-#define HAS_PC8(dev)		(IS_HASWELL(dev)) /* XXX HSW:ULX */
-#define HAS_RUNTIME_PM(dev)	(IS_HASWELL(dev))
+#define HAS_RUNTIME_PM(dev)	(IS_GEN6(dev) || IS_HASWELL(dev) || \
+				 IS_BROADWELL(dev) || IS_VALLEYVIEW(dev))
 
 #define INTEL_PCH_DEVICE_ID_MASK		0xff00
 #define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
@@ -2068,6 +2039,7 @@ struct i915_params {
 	bool prefault_disable;
 	bool reset;
 	bool disable_display;
+	bool disable_vtd_wa;
 };
 extern struct i915_params i915 __read_mostly;
 
@@ -2096,6 +2068,7 @@ extern unsigned long i915_chipset_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_mch_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_gfx_val(struct drm_i915_private *dev_priv);
 extern void i915_update_gfx_val(struct drm_i915_private *dev_priv);
+int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool on);
 
 extern void intel_console_resume(struct work_struct *work);
 
@@ -2170,6 +2143,9 @@ int i915_gem_set_tiling(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 int i915_gem_get_tiling(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
+int i915_gem_init_userptr(struct drm_device *dev);
+int i915_gem_userptr_ioctl(struct drm_device *dev, void *data,
+			   struct drm_file *file);
 int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
@@ -2227,9 +2203,9 @@ static inline void i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
 
 int __must_check i915_mutex_lock_interruptible(struct drm_device *dev);
 int i915_gem_object_sync(struct drm_i915_gem_object *obj,
-			 struct intel_ring_buffer *to);
+			 struct intel_engine_cs *to);
 void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct intel_ring_buffer *ring);
+			     struct intel_engine_cs *ring);
 int i915_gem_dumb_create(struct drm_file *file_priv,
 			 struct drm_device *dev,
 			 struct drm_mode_create_dumb *args);
@@ -2249,31 +2225,14 @@ int __must_check i915_gem_set_seqno(struct drm_device *dev, u32 seqno);
 int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_object_put_fence(struct drm_i915_gem_object *obj);
 
-static inline bool
-i915_gem_object_pin_fence(struct drm_i915_gem_object *obj)
-{
-	if (obj->fence_reg != I915_FENCE_REG_NONE) {
-		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-		dev_priv->fence_regs[obj->fence_reg].pin_count++;
-		return true;
-	} else
-		return false;
-}
-
-static inline void
-i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
-{
-	if (obj->fence_reg != I915_FENCE_REG_NONE) {
-		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-		WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count <= 0);
-		dev_priv->fence_regs[obj->fence_reg].pin_count--;
-	}
-}
+bool i915_gem_object_pin_fence(struct drm_i915_gem_object *obj);
+void i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj);
 
 struct drm_i915_gem_request *
-i915_gem_find_active_request(struct intel_ring_buffer *ring);
+i915_gem_find_active_request(struct intel_engine_cs *ring);
 
 bool i915_gem_retire_requests(struct drm_device *dev);
+void i915_gem_retire_requests_ring(struct intel_engine_cs *ring);
 int __must_check i915_gem_check_wedge(struct i915_gpu_error *error,
 				      bool interruptible);
 static inline bool i915_reset_in_progress(struct i915_gpu_error *error)
@@ -2292,23 +2251,35 @@ static inline u32 i915_reset_count(struct i915_gpu_error *error)
 	return ((atomic_read(&error->reset_counter) & ~I915_WEDGED) + 1) / 2;
 }
 
+static inline bool i915_stop_ring_allow_ban(struct drm_i915_private *dev_priv)
+{
+	return dev_priv->gpu_error.stop_rings == 0 ||
+		dev_priv->gpu_error.stop_rings & I915_STOP_RING_ALLOW_BAN;
+}
+
+static inline bool i915_stop_ring_allow_warn(struct drm_i915_private *dev_priv)
+{
+	return dev_priv->gpu_error.stop_rings == 0 ||
+		dev_priv->gpu_error.stop_rings & I915_STOP_RING_ALLOW_WARN;
+}
+
 void i915_gem_reset(struct drm_device *dev);
 bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, bool force);
 int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_init(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
-int i915_gem_l3_remap(struct intel_ring_buffer *ring, int slice);
+int i915_gem_l3_remap(struct intel_engine_cs *ring, int slice);
 void i915_gem_init_swizzling(struct drm_device *dev);
 void i915_gem_cleanup_ringbuffer(struct drm_device *dev);
 int __must_check i915_gpu_idle(struct drm_device *dev);
 int __must_check i915_gem_suspend(struct drm_device *dev);
-int __i915_add_request(struct intel_ring_buffer *ring,
+int __i915_add_request(struct intel_engine_cs *ring,
 		       struct drm_file *file,
 		       struct drm_i915_gem_object *batch_obj,
 		       u32 *seqno);
 #define i915_add_request(ring, seqno) \
 	__i915_add_request(ring, NULL, NULL, seqno)
-int __must_check i915_wait_seqno(struct intel_ring_buffer *ring,
+int __must_check i915_wait_seqno(struct intel_engine_cs *ring,
 				 uint32_t seqno);
 int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
 int __must_check
@@ -2319,7 +2290,7 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write);
 int __must_check
 i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 				     u32 alignment,
-				     struct intel_ring_buffer *pipelined);
+				     struct intel_engine_cs *pipelined);
 void i915_gem_object_unpin_from_display_plane(struct drm_i915_gem_object *obj);
 int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj,
 				int align);
@@ -2416,22 +2387,22 @@ void i915_gem_context_reset(struct drm_device *dev);
 int i915_gem_context_open(struct drm_device *dev, struct drm_file *file);
 int i915_gem_context_enable(struct drm_i915_private *dev_priv);
 void i915_gem_context_close(struct drm_device *dev, struct drm_file *file);
-int i915_switch_context(struct intel_ring_buffer *ring,
-			struct i915_hw_context *to);
-struct i915_hw_context *
+int i915_switch_context(struct intel_engine_cs *ring,
+			struct intel_context *to);
+struct intel_context *
 i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id);
 void i915_gem_context_free(struct kref *ctx_ref);
-static inline void i915_gem_context_reference(struct i915_hw_context *ctx)
+static inline void i915_gem_context_reference(struct intel_context *ctx)
 {
 	kref_get(&ctx->ref);
 }
 
-static inline void i915_gem_context_unreference(struct i915_hw_context *ctx)
+static inline void i915_gem_context_unreference(struct intel_context *ctx)
 {
 	kref_put(&ctx->ref, i915_gem_context_free);
 }
 
-static inline bool i915_gem_context_is_default(const struct i915_hw_context *c)
+static inline bool i915_gem_context_is_default(const struct intel_context *c)
 {
 	return c->id == DEFAULT_CONTEXT_ID;
 }
@@ -2441,6 +2412,8 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file);
 
+/* i915_gem_render_state.c */
+int i915_gem_render_state_init(struct intel_engine_cs *ring);
 /* i915_gem_evict.c */
 int __must_check i915_gem_evict_something(struct drm_device *dev,
 					  struct i915_address_space *vm,
@@ -2453,23 +2426,12 @@ int __must_check i915_gem_evict_something(struct drm_device *dev,
 int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle);
 int i915_gem_evict_everything(struct drm_device *dev);
 
-/* i915_gem_gtt.c */
-void i915_check_and_clear_faults(struct drm_device *dev);
-void i915_gem_suspend_gtt_mappings(struct drm_device *dev);
-void i915_gem_restore_gtt_mappings(struct drm_device *dev);
-int __must_check i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj);
-void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj);
-void i915_gem_init_global_gtt(struct drm_device *dev);
-void i915_gem_setup_global_gtt(struct drm_device *dev, unsigned long start,
-			       unsigned long mappable_end, unsigned long end);
-int i915_gem_gtt_init(struct drm_device *dev);
+/* belongs in i915_gem_gtt.h */
 static inline void i915_gem_chipset_flush(struct drm_device *dev)
 {
 	if (INTEL_INFO(dev)->gen < 6)
 		intel_gtt_chipset_flush();
 }
-int i915_gem_init_ppgtt(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt);
-bool intel_enable_ppgtt(struct drm_device *dev, bool full);
 
 /* i915_gem_stolen.c */
 int i915_gem_init_stolen(struct drm_device *dev);
@@ -2537,9 +2499,11 @@ void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone);
 const char *i915_cache_level_str(int type);
 
 /* i915_cmd_parser.c */
-void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring);
-bool i915_needs_cmd_parser(struct intel_ring_buffer *ring);
-int i915_parse_cmds(struct intel_ring_buffer *ring,
+int i915_cmd_parser_get_version(void);
+int i915_cmd_parser_init_ring(struct intel_engine_cs *ring);
+void i915_cmd_parser_fini_ring(struct intel_engine_cs *ring);
+bool i915_needs_cmd_parser(struct intel_engine_cs *ring);
+int i915_parse_cmds(struct intel_engine_cs *ring,
 		    struct drm_i915_gem_object *batch_obj,
 		    u32 batch_start_offset,
 		    bool is_master);
@@ -2688,20 +2652,6 @@ void vlv_flisdsi_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
 int vlv_gpu_freq(struct drm_i915_private *dev_priv, int val);
 int vlv_freq_opcode(struct drm_i915_private *dev_priv, int val);
 
-void vlv_force_wake_get(struct drm_i915_private *dev_priv, int fw_engine);
-void vlv_force_wake_put(struct drm_i915_private *dev_priv, int fw_engine);
-
-#define FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg) \
-	(((reg) >= 0x2000 && (reg) < 0x4000) ||\
-	((reg) >= 0x5000 && (reg) < 0x8000) ||\
-	((reg) >= 0xB000 && (reg) < 0x12000) ||\
-	((reg) >= 0x2E000 && (reg) < 0x30000))
-
-#define FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(reg)\
-	(((reg) >= 0x12000 && (reg) < 0x14000) ||\
-	((reg) >= 0x22000 && (reg) < 0x24000) ||\
-	((reg) >= 0x30000 && (reg) < 0x40000))
-
 #define FORCEWAKE_RENDER	(1 << 0)
 #define FORCEWAKE_MEDIA		(1 << 1)
 #define FORCEWAKE_ALL		(FORCEWAKE_RENDER | FORCEWAKE_MEDIA)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 3326770c9ed2..f36126383d26 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -31,6 +31,7 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include <linux/oom.h>
 #include <linux/shmem_fs.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
@@ -43,6 +44,8 @@ static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *o
 static __must_check int
 i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 			       bool readonly);
+static void
+i915_gem_object_retire(struct drm_i915_gem_object *obj);
 
 static void i915_gem_write_fence(struct drm_device *dev, int reg,
 				 struct drm_i915_gem_object *obj);
@@ -50,14 +53,15 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj,
 					 struct drm_i915_fence_reg *fence,
 					 bool enable);
 
-static unsigned long i915_gem_inactive_count(struct shrinker *shrinker,
+static unsigned long i915_gem_shrinker_count(struct shrinker *shrinker,
 					     struct shrink_control *sc);
-static unsigned long i915_gem_inactive_scan(struct shrinker *shrinker,
+static unsigned long i915_gem_shrinker_scan(struct shrinker *shrinker,
 					    struct shrink_control *sc);
+static int i915_gem_shrinker_oom(struct notifier_block *nb,
+				 unsigned long event,
+				 void *ptr);
 static unsigned long i915_gem_purge(struct drm_i915_private *dev_priv, long target);
 static unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
-static void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
-static void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
 
 static bool cpu_cache_is_coherent(struct drm_device *dev,
 				  enum i915_cache_level level)
@@ -470,6 +474,8 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 		ret = i915_gem_object_wait_rendering(obj, true);
 		if (ret)
 			return ret;
+
+		i915_gem_object_retire(obj);
 	}
 
 	ret = i915_gem_object_get_pages(obj);
@@ -885,6 +891,8 @@ i915_gem_shmem_pwrite(struct drm_device *dev,
 		ret = i915_gem_object_wait_rendering(obj, false);
 		if (ret)
 			return ret;
+
+		i915_gem_object_retire(obj);
 	}
 	/* Same trick applies to invalidate partially written cachelines read
 	 * before writing. */
@@ -1088,7 +1096,7 @@ i915_gem_check_wedge(struct i915_gpu_error *error,
  * equal.
  */
 static int
-i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno)
+i915_gem_check_olr(struct intel_engine_cs *ring, u32 seqno)
 {
 	int ret;
 
@@ -1107,7 +1115,7 @@ static void fake_irq(unsigned long data)
 }
 
 static bool missed_irq(struct drm_i915_private *dev_priv,
-		       struct intel_ring_buffer *ring)
+		       struct intel_engine_cs *ring)
 {
 	return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings);
 }
@@ -1138,7 +1146,7 @@ static bool can_wait_boost(struct drm_i915_file_private *file_priv)
  * Returns 0 if the seqno was found within the alloted time. Else returns the
  * errno with remaining time filled in timeout argument.
  */
-static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
+static int __wait_seqno(struct intel_engine_cs *ring, u32 seqno,
 			unsigned reset_counter,
 			bool interruptible,
 			struct timespec *timeout,
@@ -1245,7 +1253,7 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
  * request and object lists appropriately for that event.
  */
 int
-i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
+i915_wait_seqno(struct intel_engine_cs *ring, uint32_t seqno)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1270,9 +1278,10 @@ i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 
 static int
 i915_gem_object_wait_rendering__tail(struct drm_i915_gem_object *obj,
-				     struct intel_ring_buffer *ring)
+				     struct intel_engine_cs *ring)
 {
-	i915_gem_retire_requests_ring(ring);
+	if (!obj->active)
+		return 0;
 
 	/* Manually manage the write flush as we may have not yet
 	 * retired the buffer.
@@ -1282,7 +1291,6 @@ i915_gem_object_wait_rendering__tail(struct drm_i915_gem_object *obj,
 	 * we know we have passed the last write.
 	 */
 	obj->last_write_seqno = 0;
-	obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
 
 	return 0;
 }
@@ -1295,7 +1303,7 @@ static __must_check int
 i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 			       bool readonly)
 {
-	struct intel_ring_buffer *ring = obj->ring;
+	struct intel_engine_cs *ring = obj->ring;
 	u32 seqno;
 	int ret;
 
@@ -1320,7 +1328,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = obj->ring;
+	struct intel_engine_cs *ring = obj->ring;
 	unsigned reset_counter;
 	u32 seqno;
 	int ret;
@@ -1536,7 +1544,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	/* Access to snoopable pages through the GTT is incoherent. */
 	if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(dev)) {
-		ret = -EINVAL;
+		ret = -EFAULT;
 		goto unlock;
 	}
 
@@ -1803,12 +1811,16 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data,
 	return i915_gem_mmap_gtt(file, dev, args->handle, &args->offset);
 }
 
+static inline int
+i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj)
+{
+	return obj->madv == I915_MADV_DONTNEED;
+}
+
 /* Immediately discard the backing storage */
 static void
 i915_gem_object_truncate(struct drm_i915_gem_object *obj)
 {
-	struct inode *inode;
-
 	i915_gem_object_free_mmap_offset(obj);
 
 	if (obj->base.filp == NULL)
@@ -1819,16 +1831,28 @@ i915_gem_object_truncate(struct drm_i915_gem_object *obj)
 	 * To do this we must instruct the shmfs to drop all of its
 	 * backing pages, *now*.
 	 */
-	inode = file_inode(obj->base.filp);
-	shmem_truncate_range(inode, 0, (loff_t)-1);
-
+	shmem_truncate_range(file_inode(obj->base.filp), 0, (loff_t)-1);
 	obj->madv = __I915_MADV_PURGED;
 }
 
-static inline int
-i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj)
+/* Try to discard unwanted pages */
+static void
+i915_gem_object_invalidate(struct drm_i915_gem_object *obj)
 {
-	return obj->madv == I915_MADV_DONTNEED;
+	struct address_space *mapping;
+
+	switch (obj->madv) {
+	case I915_MADV_DONTNEED:
+		i915_gem_object_truncate(obj);
+	case __I915_MADV_PURGED:
+		return;
+	}
+
+	if (obj->base.filp == NULL)
+		return;
+
+	mapping = file_inode(obj->base.filp)->i_mapping,
+	invalidate_mapping_pages(mapping, 0, (loff_t)-1);
 }
 
 static void
@@ -1893,8 +1917,7 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)
 	ops->put_pages(obj);
 	obj->pages = NULL;
 
-	if (i915_gem_object_is_purgeable(obj))
-		i915_gem_object_truncate(obj);
+	i915_gem_object_invalidate(obj);
 
 	return 0;
 }
@@ -1903,58 +1926,58 @@ static unsigned long
 __i915_gem_shrink(struct drm_i915_private *dev_priv, long target,
 		  bool purgeable_only)
 {
-	struct list_head still_bound_list;
-	struct drm_i915_gem_object *obj, *next;
+	struct list_head still_in_list;
+	struct drm_i915_gem_object *obj;
 	unsigned long count = 0;
 
-	list_for_each_entry_safe(obj, next,
-				 &dev_priv->mm.unbound_list,
-				 global_list) {
-		if ((i915_gem_object_is_purgeable(obj) || !purgeable_only) &&
-		    i915_gem_object_put_pages(obj) == 0) {
-			count += obj->base.size >> PAGE_SHIFT;
-			if (count >= target)
-				return count;
-		}
-	}
-
 	/*
-	 * As we may completely rewrite the bound list whilst unbinding
+	 * As we may completely rewrite the (un)bound list whilst unbinding
 	 * (due to retiring requests) we have to strictly process only
 	 * one element of the list at the time, and recheck the list
 	 * on every iteration.
+	 *
+	 * In particular, we must hold a reference whilst removing the
+	 * object as we may end up waiting for and/or retiring the objects.
+	 * This might release the final reference (held by the active list)
+	 * and result in the object being freed from under us. This is
+	 * similar to the precautions the eviction code must take whilst
+	 * removing objects.
+	 *
+	 * Also note that although these lists do not hold a reference to
+	 * the object we can safely grab one here: The final object
+	 * unreferencing and the bound_list are both protected by the
+	 * dev->struct_mutex and so we won't ever be able to observe an
+	 * object on the bound_list with a reference count equals 0.
 	 */
-	INIT_LIST_HEAD(&still_bound_list);
+	INIT_LIST_HEAD(&still_in_list);
+	while (count < target && !list_empty(&dev_priv->mm.unbound_list)) {
+		obj = list_first_entry(&dev_priv->mm.unbound_list,
+				       typeof(*obj), global_list);
+		list_move_tail(&obj->global_list, &still_in_list);
+
+		if (!i915_gem_object_is_purgeable(obj) && purgeable_only)
+			continue;
+
+		drm_gem_object_reference(&obj->base);
+
+		if (i915_gem_object_put_pages(obj) == 0)
+			count += obj->base.size >> PAGE_SHIFT;
+
+		drm_gem_object_unreference(&obj->base);
+	}
+	list_splice(&still_in_list, &dev_priv->mm.unbound_list);
+
+	INIT_LIST_HEAD(&still_in_list);
 	while (count < target && !list_empty(&dev_priv->mm.bound_list)) {
 		struct i915_vma *vma, *v;
 
 		obj = list_first_entry(&dev_priv->mm.bound_list,
 				       typeof(*obj), global_list);
-		list_move_tail(&obj->global_list, &still_bound_list);
+		list_move_tail(&obj->global_list, &still_in_list);
 
 		if (!i915_gem_object_is_purgeable(obj) && purgeable_only)
 			continue;
 
-		/*
-		 * Hold a reference whilst we unbind this object, as we may
-		 * end up waiting for and retiring requests. This might
-		 * release the final reference (held by the active list)
-		 * and result in the object being freed from under us.
-		 * in this object being freed.
-		 *
-		 * Note 1: Shrinking the bound list is special since only active
-		 * (and hence bound objects) can contain such limbo objects, so
-		 * we don't need special tricks for shrinking the unbound list.
-		 * The only other place where we have to be careful with active
-		 * objects suddenly disappearing due to retiring requests is the
-		 * eviction code.
-		 *
-		 * Note 2: Even though the bound list doesn't hold a reference
-		 * to the object we can safely grab one here: The final object
-		 * unreferencing and the bound_list are both protected by the
-		 * dev->struct_mutex and so we won't ever be able to observe an
-		 * object on the bound_list with a reference count equals 0.
-		 */
 		drm_gem_object_reference(&obj->base);
 
 		list_for_each_entry_safe(vma, v, &obj->vma_list, vma_link)
@@ -1966,7 +1989,7 @@ __i915_gem_shrink(struct drm_i915_private *dev_priv, long target,
 
 		drm_gem_object_unreference(&obj->base);
 	}
-	list_splice(&still_bound_list, &dev_priv->mm.bound_list);
+	list_splice(&still_in_list, &dev_priv->mm.bound_list);
 
 	return count;
 }
@@ -1980,17 +2003,8 @@ i915_gem_purge(struct drm_i915_private *dev_priv, long target)
 static unsigned long
 i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_gem_object *obj, *next;
-	long freed = 0;
-
 	i915_gem_evict_everything(dev_priv->dev);
-
-	list_for_each_entry_safe(obj, next, &dev_priv->mm.unbound_list,
-				 global_list) {
-		if (i915_gem_object_put_pages(obj) == 0)
-			freed += obj->base.size >> PAGE_SHIFT;
-	}
-	return freed;
+	return __i915_gem_shrink(dev_priv, LONG_MAX, false);
 }
 
 static int
@@ -2094,7 +2108,19 @@ err_pages:
 		page_cache_release(sg_page_iter_page(&sg_iter));
 	sg_free_table(st);
 	kfree(st);
-	return PTR_ERR(page);
+
+	/* shmemfs first checks if there is enough memory to allocate the page
+	 * and reports ENOSPC should there be insufficient, along with the usual
+	 * ENOMEM for a genuine allocation failure.
+	 *
+	 * We use ENOSPC in our driver to mean that we have run out of aperture
+	 * space and so want to translate the error from shmemfs back to our
+	 * usual understanding of ENOMEM.
+	 */
+	if (PTR_ERR(page) == -ENOSPC)
+		return -ENOMEM;
+	else
+		return PTR_ERR(page);
 }
 
 /* Ensure that the associated pages are gathered from the backing storage
@@ -2131,7 +2157,7 @@ i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 
 static void
 i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
-			       struct intel_ring_buffer *ring)
+			       struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2169,7 +2195,7 @@ i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
 }
 
 void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct intel_ring_buffer *ring)
+			     struct intel_engine_cs *ring)
 {
 	list_move_tail(&vma->mm_list, &vma->vm->active_list);
 	return i915_gem_object_move_to_active(vma->obj, ring);
@@ -2207,11 +2233,24 @@ i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
 	WARN_ON(i915_verify_lists(dev));
 }
 
+static void
+i915_gem_object_retire(struct drm_i915_gem_object *obj)
+{
+	struct intel_engine_cs *ring = obj->ring;
+
+	if (ring == NULL)
+		return;
+
+	if (i915_seqno_passed(ring->get_seqno(ring, true),
+			      obj->last_read_seqno))
+		i915_gem_object_move_to_inactive(obj);
+}
+
 static int
 i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i, j;
 
 	/* Carefully retire all requests without writing to the rings */
@@ -2226,8 +2265,8 @@ i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
 	for_each_ring(ring, dev_priv, i) {
 		intel_ring_init_seqno(ring, seqno);
 
-		for (j = 0; j < ARRAY_SIZE(ring->sync_seqno); j++)
-			ring->sync_seqno[j] = 0;
+		for (j = 0; j < ARRAY_SIZE(ring->semaphore.sync_seqno); j++)
+			ring->semaphore.sync_seqno[j] = 0;
 	}
 
 	return 0;
@@ -2277,7 +2316,7 @@ i915_gem_get_seqno(struct drm_device *dev, u32 *seqno)
 	return 0;
 }
 
-int __i915_add_request(struct intel_ring_buffer *ring,
+int __i915_add_request(struct intel_engine_cs *ring,
 		       struct drm_file *file,
 		       struct drm_i915_gem_object *obj,
 		       u32 *out_seqno)
@@ -2382,7 +2421,7 @@ i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
 }
 
 static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
-				   const struct i915_hw_context *ctx)
+				   const struct intel_context *ctx)
 {
 	unsigned long elapsed;
 
@@ -2395,8 +2434,9 @@ static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
 		if (!i915_gem_context_is_default(ctx)) {
 			DRM_DEBUG("context hanging too fast, banning!\n");
 			return true;
-		} else if (dev_priv->gpu_error.stop_rings == 0) {
-			DRM_ERROR("gpu hanging too fast, banning!\n");
+		} else if (i915_stop_ring_allow_ban(dev_priv)) {
+			if (i915_stop_ring_allow_warn(dev_priv))
+				DRM_ERROR("gpu hanging too fast, banning!\n");
 			return true;
 		}
 	}
@@ -2405,7 +2445,7 @@ static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
 }
 
 static void i915_set_reset_status(struct drm_i915_private *dev_priv,
-				  struct i915_hw_context *ctx,
+				  struct intel_context *ctx,
 				  const bool guilty)
 {
 	struct i915_ctx_hang_stats *hs;
@@ -2436,7 +2476,7 @@ static void i915_gem_free_request(struct drm_i915_gem_request *request)
 }
 
 struct drm_i915_gem_request *
-i915_gem_find_active_request(struct intel_ring_buffer *ring)
+i915_gem_find_active_request(struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_request *request;
 	u32 completed_seqno;
@@ -2454,7 +2494,7 @@ i915_gem_find_active_request(struct intel_ring_buffer *ring)
 }
 
 static void i915_gem_reset_ring_status(struct drm_i915_private *dev_priv,
-				       struct intel_ring_buffer *ring)
+				       struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_request *request;
 	bool ring_hung;
@@ -2473,7 +2513,7 @@ static void i915_gem_reset_ring_status(struct drm_i915_private *dev_priv,
 }
 
 static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
-					struct intel_ring_buffer *ring)
+					struct intel_engine_cs *ring)
 {
 	while (!list_empty(&ring->active_list)) {
 		struct drm_i915_gem_object *obj;
@@ -2501,6 +2541,11 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 
 		i915_gem_free_request(request);
 	}
+
+	/* These may not have been flush before the reset, do so now */
+	kfree(ring->preallocated_lazy_request);
+	ring->preallocated_lazy_request = NULL;
+	ring->outstanding_lazy_seqno = 0;
 }
 
 void i915_gem_restore_fences(struct drm_device *dev)
@@ -2527,7 +2572,7 @@ void i915_gem_restore_fences(struct drm_device *dev)
 void i915_gem_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	/*
@@ -2541,8 +2586,6 @@ void i915_gem_reset(struct drm_device *dev)
 	for_each_ring(ring, dev_priv, i)
 		i915_gem_reset_ring_cleanup(dev_priv, ring);
 
-	i915_gem_cleanup_ringbuffer(dev);
-
 	i915_gem_context_reset(dev);
 
 	i915_gem_restore_fences(dev);
@@ -2551,8 +2594,8 @@ void i915_gem_reset(struct drm_device *dev)
 /**
  * This function clears the request list as sequence numbers are passed.
  */
-static void
-i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
+void
+i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 {
 	uint32_t seqno;
 
@@ -2597,7 +2640,7 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
 		 * of tail of the request to update the last known position
 		 * of the GPU head.
 		 */
-		ring->last_retired_head = request->tail;
+		ring->buffer->last_retired_head = request->tail;
 
 		i915_gem_free_request(request);
 	}
@@ -2615,7 +2658,7 @@ bool
 i915_gem_retire_requests(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	bool idle = true;
 	int i;
 
@@ -2709,7 +2752,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_wait *args = data;
 	struct drm_i915_gem_object *obj;
-	struct intel_ring_buffer *ring = NULL;
+	struct intel_engine_cs *ring = NULL;
 	struct timespec timeout_stack, *timeout = NULL;
 	unsigned reset_counter;
 	u32 seqno = 0;
@@ -2780,9 +2823,9 @@ out:
  */
 int
 i915_gem_object_sync(struct drm_i915_gem_object *obj,
-		     struct intel_ring_buffer *to)
+		     struct intel_engine_cs *to)
 {
-	struct intel_ring_buffer *from = obj->ring;
+	struct intel_engine_cs *from = obj->ring;
 	u32 seqno;
 	int ret, idx;
 
@@ -2795,7 +2838,7 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 	idx = intel_ring_sync_index(from, to);
 
 	seqno = obj->last_read_seqno;
-	if (seqno <= from->sync_seqno[idx])
+	if (seqno <= from->semaphore.sync_seqno[idx])
 		return 0;
 
 	ret = i915_gem_check_olr(obj->ring, seqno);
@@ -2803,13 +2846,13 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 		return ret;
 
 	trace_i915_gem_ring_sync_to(from, to, seqno);
-	ret = to->sync_to(to, from, seqno);
+	ret = to->semaphore.sync_to(to, from, seqno);
 	if (!ret)
 		/* We use last_read_seqno because sync_to()
 		 * might have just caused seqno wrap under
 		 * the radar.
 		 */
-		from->sync_seqno[idx] = obj->last_read_seqno;
+		from->semaphore.sync_seqno[idx] = obj->last_read_seqno;
 
 	return ret;
 }
@@ -2865,12 +2908,14 @@ int i915_vma_unbind(struct i915_vma *vma)
 	 * cause memory corruption through use-after-free.
 	 */
 
-	i915_gem_object_finish_gtt(obj);
+	if (i915_is_ggtt(vma->vm)) {
+		i915_gem_object_finish_gtt(obj);
 
-	/* release the fence reg _after_ flushing */
-	ret = i915_gem_object_put_fence(obj);
-	if (ret)
-		return ret;
+		/* release the fence reg _after_ flushing */
+		ret = i915_gem_object_put_fence(obj);
+		if (ret)
+			return ret;
+	}
 
 	trace_i915_vma_unbind(vma);
 
@@ -2903,7 +2948,7 @@ int i915_vma_unbind(struct i915_vma *vma)
 int i915_gpu_idle(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i;
 
 	/* Flush everything onto the inactive list. */
@@ -3144,6 +3189,9 @@ i915_gem_object_put_fence(struct drm_i915_gem_object *obj)
 
 	fence = &dev_priv->fence_regs[obj->fence_reg];
 
+	if (WARN_ON(fence->pin_count))
+		return -EBUSY;
+
 	i915_gem_object_fence_lost(obj);
 	i915_gem_object_update_fence(obj, fence, false);
 
@@ -3548,6 +3596,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	if (ret)
 		return ret;
 
+	i915_gem_object_retire(obj);
 	i915_gem_object_flush_cpu_write_domain(obj, false);
 
 	/* Serialise direct access to this object with the barriers for
@@ -3646,6 +3695,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		 * in obj->write_domain and have been skipping the clflushes.
 		 * Just set it to the CPU cache for now.
 		 */
+		i915_gem_object_retire(obj);
 		WARN_ON(obj->base.write_domain & ~I915_GEM_DOMAIN_CPU);
 
 		old_read_domains = obj->base.read_domains;
@@ -3743,6 +3793,15 @@ unlock:
 
 static bool is_pin_display(struct drm_i915_gem_object *obj)
 {
+	struct i915_vma *vma;
+
+	if (list_empty(&obj->vma_list))
+		return false;
+
+	vma = i915_gem_obj_to_ggtt(obj);
+	if (!vma)
+		return false;
+
 	/* There are 3 sources that pin objects:
 	 *   1. The display engine (scanouts, sprites, cursors);
 	 *   2. Reservations for execbuffer;
@@ -3754,7 +3813,7 @@ static bool is_pin_display(struct drm_i915_gem_object *obj)
 	 * subtracting the potential reference by the user, any pin_count
 	 * remains, it must be due to another use by the display engine.
 	 */
-	return i915_gem_obj_to_ggtt(obj)->pin_count - !!obj->user_pin_count;
+	return vma->pin_count - !!obj->user_pin_count;
 }
 
 /*
@@ -3765,9 +3824,10 @@ static bool is_pin_display(struct drm_i915_gem_object *obj)
 int
 i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 				     u32 alignment,
-				     struct intel_ring_buffer *pipelined)
+				     struct intel_engine_cs *pipelined)
 {
 	u32 old_read_domains, old_write_domain;
+	bool was_pin_display;
 	int ret;
 
 	if (pipelined != obj->ring) {
@@ -3779,6 +3839,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	/* Mark the pin_display early so that we account for the
 	 * display coherency whilst setting up the cache domains.
 	 */
+	was_pin_display = obj->pin_display;
 	obj->pin_display = true;
 
 	/* The display engine is not coherent with the LLC cache on gen6.  As
@@ -3821,7 +3882,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	return 0;
 
 err_unpin_display:
-	obj->pin_display = is_pin_display(obj);
+	WARN_ON(was_pin_display != is_pin_display(obj));
+	obj->pin_display = was_pin_display;
 	return ret;
 }
 
@@ -3868,6 +3930,7 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 	if (ret)
 		return ret;
 
+	i915_gem_object_retire(obj);
 	i915_gem_object_flush_gtt_write_domain(obj);
 
 	old_write_domain = obj->base.write_domain;
@@ -3917,7 +3980,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 	unsigned long recent_enough = jiffies - msecs_to_jiffies(20);
 	struct drm_i915_gem_request *request;
-	struct intel_ring_buffer *ring = NULL;
+	struct intel_engine_cs *ring = NULL;
 	unsigned reset_counter;
 	u32 seqno = 0;
 	int ret;
@@ -3976,9 +4039,13 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj,
 		    uint32_t alignment,
 		    uint64_t flags)
 {
+	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	struct i915_vma *vma;
 	int ret;
 
+	if (WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base))
+		return -ENODEV;
+
 	if (WARN_ON(flags & (PIN_GLOBAL | PIN_MAPPABLE) && !i915_is_ggtt(vm)))
 		return -EINVAL;
 
@@ -4032,6 +4099,32 @@ i915_gem_object_ggtt_unpin(struct drm_i915_gem_object *obj)
 		obj->pin_mappable = false;
 }
 
+bool
+i915_gem_object_pin_fence(struct drm_i915_gem_object *obj)
+{
+	if (obj->fence_reg != I915_FENCE_REG_NONE) {
+		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+		struct i915_vma *ggtt_vma = i915_gem_obj_to_ggtt(obj);
+
+		WARN_ON(!ggtt_vma ||
+			dev_priv->fence_regs[obj->fence_reg].pin_count >
+			ggtt_vma->pin_count);
+		dev_priv->fence_regs[obj->fence_reg].pin_count++;
+		return true;
+	} else
+		return false;
+}
+
+void
+i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
+{
+	if (obj->fence_reg != I915_FENCE_REG_NONE) {
+		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+		WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count <= 0);
+		dev_priv->fence_regs[obj->fence_reg].pin_count--;
+	}
+}
+
 int
 i915_gem_pin_ioctl(struct drm_device *dev, void *data,
 		   struct drm_file *file)
@@ -4292,6 +4385,30 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 	return obj;
 }
 
+static bool discard_backing_storage(struct drm_i915_gem_object *obj)
+{
+	/* If we are the last user of the backing storage (be it shmemfs
+	 * pages or stolen etc), we know that the pages are going to be
+	 * immediately released. In this case, we can then skip copying
+	 * back the contents from the GPU.
+	 */
+
+	if (obj->madv != I915_MADV_WILLNEED)
+		return false;
+
+	if (obj->base.filp == NULL)
+		return true;
+
+	/* At first glance, this looks racy, but then again so would be
+	 * userspace racing mmap against close. However, the first external
+	 * reference to the filp can only be obtained through the
+	 * i915_gem_mmap_ioctl() which safeguards us against the user
+	 * acquiring such a reference whilst we are in the middle of
+	 * freeing the object.
+	 */
+	return atomic_long_read(&obj->base.filp->f_count) == 1;
+}
+
 void i915_gem_free_object(struct drm_gem_object *gem_obj)
 {
 	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
@@ -4329,6 +4446,8 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 
 	if (WARN_ON(obj->pages_pin_count))
 		obj->pages_pin_count = 0;
+	if (discard_backing_storage(obj))
+		obj->madv = I915_MADV_DONTNEED;
 	i915_gem_object_put_pages(obj);
 	i915_gem_object_free_mmap_offset(obj);
 	i915_gem_object_release_stolen(obj);
@@ -4338,6 +4457,9 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	if (obj->base.import_attach)
 		drm_prime_gem_destroy(&obj->base, NULL);
 
+	if (obj->ops->release)
+		obj->ops->release(obj);
+
 	drm_gem_object_release(&obj->base);
 	i915_gem_info_remove_obj(dev_priv, obj->base.size);
 
@@ -4371,6 +4493,17 @@ void i915_gem_vma_destroy(struct i915_vma *vma)
 	kfree(vma);
 }
 
+static void
+i915_gem_stop_ringbuffers(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring;
+	int i;
+
+	for_each_ring(ring, dev_priv, i)
+		intel_stop_ring_buffer(ring);
+}
+
 int
 i915_gem_suspend(struct drm_device *dev)
 {
@@ -4392,7 +4525,7 @@ i915_gem_suspend(struct drm_device *dev)
 		i915_gem_evict_everything(dev);
 
 	i915_kernel_lost_context(dev);
-	i915_gem_cleanup_ringbuffer(dev);
+	i915_gem_stop_ringbuffers(dev);
 
 	/* Hack!  Don't let anybody do execbuf while we don't control the chip.
 	 * We need to replace this with a semaphore, or something.
@@ -4413,7 +4546,7 @@ err:
 	return ret;
 }
 
-int i915_gem_l3_remap(struct intel_ring_buffer *ring, int slice)
+int i915_gem_l3_remap(struct intel_engine_cs *ring, int slice)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4512,13 +4645,20 @@ static int i915_gem_init_rings(struct drm_device *dev)
 			goto cleanup_blt_ring;
 	}
 
+	if (HAS_BSD2(dev)) {
+		ret = intel_init_bsd2_ring_buffer(dev);
+		if (ret)
+			goto cleanup_vebox_ring;
+	}
 
 	ret = i915_gem_set_seqno(dev, ((u32)~0 - 0x1000));
 	if (ret)
-		goto cleanup_vebox_ring;
+		goto cleanup_bsd2_ring;
 
 	return 0;
 
+cleanup_bsd2_ring:
+	intel_cleanup_ring_buffer(&dev_priv->ring[VCS2]);
 cleanup_vebox_ring:
 	intel_cleanup_ring_buffer(&dev_priv->ring[VECS]);
 cleanup_blt_ring:
@@ -4576,15 +4716,11 @@ i915_gem_init_hw(struct drm_device *dev)
 	 * the do_switch), but before enabling PPGTT. So don't move this.
 	 */
 	ret = i915_gem_context_enable(dev_priv);
-	if (ret) {
+	if (ret && ret != -EIO) {
 		DRM_ERROR("Context enable failed %d\n", ret);
-		goto err_out;
+		i915_gem_cleanup_ringbuffer(dev);
 	}
 
-	return 0;
-
-err_out:
-	i915_gem_cleanup_ringbuffer(dev);
 	return ret;
 }
 
@@ -4597,11 +4733,13 @@ int i915_gem_init(struct drm_device *dev)
 
 	if (IS_VALLEYVIEW(dev)) {
 		/* VLVA0 (potential hack), BIOS isn't actually waking us */
-		I915_WRITE(VLV_GTLC_WAKE_CTRL, 1);
-		if (wait_for((I915_READ(VLV_GTLC_PW_STATUS) & 1) == 1, 10))
+		I915_WRITE(VLV_GTLC_WAKE_CTRL, VLV_GTLC_ALLOWWAKEREQ);
+		if (wait_for((I915_READ(VLV_GTLC_PW_STATUS) &
+			      VLV_GTLC_ALLOWWAKEACK), 10))
 			DRM_DEBUG_DRIVER("allow wake ack timed out\n");
 	}
 
+	i915_gem_init_userptr(dev);
 	i915_gem_init_global_gtt(dev);
 
 	ret = i915_gem_context_init(dev);
@@ -4611,25 +4749,28 @@ int i915_gem_init(struct drm_device *dev)
 	}
 
 	ret = i915_gem_init_hw(dev);
-	mutex_unlock(&dev->struct_mutex);
-	if (ret) {
-		WARN_ON(dev_priv->mm.aliasing_ppgtt);
-		i915_gem_context_fini(dev);
-		drm_mm_takedown(&dev_priv->gtt.base.mm);
-		return ret;
+	if (ret == -EIO) {
+		/* Allow ring initialisation to fail by marking the GPU as
+		 * wedged. But we only want to do this where the GPU is angry,
+		 * for all other failure, such as an allocation failure, bail.
+		 */
+		DRM_ERROR("Failed to initialize GPU, declaring it wedged\n");
+		atomic_set_mask(I915_WEDGED, &dev_priv->gpu_error.reset_counter);
+		ret = 0;
 	}
+	mutex_unlock(&dev->struct_mutex);
 
 	/* Allow hardware batchbuffers unless told otherwise, but not for KMS. */
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		dev_priv->dri1.allow_batchbuffer = 1;
-	return 0;
+	return ret;
 }
 
 void
 i915_gem_cleanup_ringbuffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	for_each_ring(ring, dev_priv, i)
@@ -4661,16 +4802,15 @@ i915_gem_entervt_ioctl(struct drm_device *dev, void *data,
 	}
 
 	BUG_ON(!list_empty(&dev_priv->gtt.base.active_list));
-	mutex_unlock(&dev->struct_mutex);
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, dev->pdev->irq);
 	if (ret)
 		goto cleanup_ringbuffer;
+	mutex_unlock(&dev->struct_mutex);
 
 	return 0;
 
 cleanup_ringbuffer:
-	mutex_lock(&dev->struct_mutex);
 	i915_gem_cleanup_ringbuffer(dev);
 	dev_priv->ums.mm_suspended = 1;
 	mutex_unlock(&dev->struct_mutex);
@@ -4685,7 +4825,9 @@ i915_gem_leavevt_ioctl(struct drm_device *dev, void *data,
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return 0;
 
+	mutex_lock(&dev->struct_mutex);
 	drm_irq_uninstall(dev);
+	mutex_unlock(&dev->struct_mutex);
 
 	return i915_gem_suspend(dev);
 }
@@ -4704,7 +4846,7 @@ i915_gem_lastclose(struct drm_device *dev)
 }
 
 static void
-init_ring_lists(struct intel_ring_buffer *ring)
+init_ring_lists(struct intel_engine_cs *ring)
 {
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
@@ -4752,7 +4894,7 @@ i915_gem_load(struct drm_device *dev)
 	init_waitqueue_head(&dev_priv->gpu_error.reset_queue);
 
 	/* On GEN3 we really need to make sure the ARB C3 LP bit is set */
-	if (IS_GEN3(dev)) {
+	if (!drm_core_check_feature(dev, DRIVER_MODESET) && IS_GEN3(dev)) {
 		I915_WRITE(MI_ARB_STATE,
 			   _MASKED_BIT_ENABLE(MI_ARB_C3_LP_WRITE_ENABLE));
 	}
@@ -4779,10 +4921,13 @@ i915_gem_load(struct drm_device *dev)
 
 	dev_priv->mm.interruptible = true;
 
-	dev_priv->mm.inactive_shrinker.scan_objects = i915_gem_inactive_scan;
-	dev_priv->mm.inactive_shrinker.count_objects = i915_gem_inactive_count;
-	dev_priv->mm.inactive_shrinker.seeks = DEFAULT_SEEKS;
-	register_shrinker(&dev_priv->mm.inactive_shrinker);
+	dev_priv->mm.shrinker.scan_objects = i915_gem_shrinker_scan;
+	dev_priv->mm.shrinker.count_objects = i915_gem_shrinker_count;
+	dev_priv->mm.shrinker.seeks = DEFAULT_SEEKS;
+	register_shrinker(&dev_priv->mm.shrinker);
+
+	dev_priv->mm.oom_notifier.notifier_call = i915_gem_shrinker_oom;
+	register_oom_notifier(&dev_priv->mm.oom_notifier);
 }
 
 void i915_gem_release(struct drm_device *dev, struct drm_file *file)
@@ -4857,27 +5002,46 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
 #endif
 }
 
+static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock)
+{
+	if (!mutex_trylock(&dev->struct_mutex)) {
+		if (!mutex_is_locked_by(&dev->struct_mutex, current))
+			return false;
+
+		if (to_i915(dev)->mm.shrinker_no_lock_stealing)
+			return false;
+
+		*unlock = false;
+	} else
+		*unlock = true;
+
+	return true;
+}
+
+static int num_vma_bound(struct drm_i915_gem_object *obj)
+{
+	struct i915_vma *vma;
+	int count = 0;
+
+	list_for_each_entry(vma, &obj->vma_list, vma_link)
+		if (drm_mm_node_allocated(&vma->node))
+			count++;
+
+	return count;
+}
+
 static unsigned long
-i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc)
+i915_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 {
 	struct drm_i915_private *dev_priv =
-		container_of(shrinker,
-			     struct drm_i915_private,
-			     mm.inactive_shrinker);
+		container_of(shrinker, struct drm_i915_private, mm.shrinker);
 	struct drm_device *dev = dev_priv->dev;
 	struct drm_i915_gem_object *obj;
-	bool unlock = true;
 	unsigned long count;
+	bool unlock;
 
-	if (!mutex_trylock(&dev->struct_mutex)) {
-		if (!mutex_is_locked_by(&dev->struct_mutex, current))
-			return 0;
-
-		if (dev_priv->mm.shrinker_no_lock_stealing)
-			return 0;
-
-		unlock = false;
-	}
+	if (!i915_gem_shrinker_lock(dev, &unlock))
+		return 0;
 
 	count = 0;
 	list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list)
@@ -4885,10 +5049,8 @@ i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc)
 			count += obj->base.size >> PAGE_SHIFT;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		if (obj->active)
-			continue;
-
-		if (!i915_gem_obj_is_pinned(obj) && obj->pages_pin_count == 0)
+		if (!i915_gem_obj_is_pinned(obj) &&
+		    obj->pages_pin_count == num_vma_bound(obj))
 			count += obj->base.size >> PAGE_SHIFT;
 	}
 
@@ -4961,44 +5123,99 @@ unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o,
 }
 
 static unsigned long
-i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc)
+i915_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc)
 {
 	struct drm_i915_private *dev_priv =
-		container_of(shrinker,
-			     struct drm_i915_private,
-			     mm.inactive_shrinker);
+		container_of(shrinker, struct drm_i915_private, mm.shrinker);
 	struct drm_device *dev = dev_priv->dev;
 	unsigned long freed;
-	bool unlock = true;
+	bool unlock;
 
-	if (!mutex_trylock(&dev->struct_mutex)) {
-		if (!mutex_is_locked_by(&dev->struct_mutex, current))
-			return SHRINK_STOP;
-
-		if (dev_priv->mm.shrinker_no_lock_stealing)
-			return SHRINK_STOP;
-
-		unlock = false;
-	}
+	if (!i915_gem_shrinker_lock(dev, &unlock))
+		return SHRINK_STOP;
 
 	freed = i915_gem_purge(dev_priv, sc->nr_to_scan);
 	if (freed < sc->nr_to_scan)
 		freed += __i915_gem_shrink(dev_priv,
 					   sc->nr_to_scan - freed,
 					   false);
-	if (freed < sc->nr_to_scan)
-		freed += i915_gem_shrink_all(dev_priv);
-
 	if (unlock)
 		mutex_unlock(&dev->struct_mutex);
 
 	return freed;
 }
 
+static int
+i915_gem_shrinker_oom(struct notifier_block *nb, unsigned long event, void *ptr)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(nb, struct drm_i915_private, mm.oom_notifier);
+	struct drm_device *dev = dev_priv->dev;
+	struct drm_i915_gem_object *obj;
+	unsigned long timeout = msecs_to_jiffies(5000) + 1;
+	unsigned long pinned, bound, unbound, freed;
+	bool was_interruptible;
+	bool unlock;
+
+	while (!i915_gem_shrinker_lock(dev, &unlock) && --timeout)
+		schedule_timeout_killable(1);
+	if (timeout == 0) {
+		pr_err("Unable to purge GPU memory due lock contention.\n");
+		return NOTIFY_DONE;
+	}
+
+	was_interruptible = dev_priv->mm.interruptible;
+	dev_priv->mm.interruptible = false;
+
+	freed = i915_gem_shrink_all(dev_priv);
+
+	dev_priv->mm.interruptible = was_interruptible;
+
+	/* Because we may be allocating inside our own driver, we cannot
+	 * assert that there are no objects with pinned pages that are not
+	 * being pointed to by hardware.
+	 */
+	unbound = bound = pinned = 0;
+	list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list) {
+		if (!obj->base.filp) /* not backed by a freeable object */
+			continue;
+
+		if (obj->pages_pin_count)
+			pinned += obj->base.size;
+		else
+			unbound += obj->base.size;
+	}
+	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
+		if (!obj->base.filp)
+			continue;
+
+		if (obj->pages_pin_count)
+			pinned += obj->base.size;
+		else
+			bound += obj->base.size;
+	}
+
+	if (unlock)
+		mutex_unlock(&dev->struct_mutex);
+
+	pr_info("Purging GPU memory, %lu bytes freed, %lu bytes still pinned.\n",
+		freed, pinned);
+	if (unbound || bound)
+		pr_err("%lu and %lu bytes still available in the "
+		       "bound and unbound GPU page lists.\n",
+		       bound, unbound);
+
+	*(unsigned long *)ptr += freed;
+	return NOTIFY_DONE;
+}
+
 struct i915_vma *i915_gem_obj_to_ggtt(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
 
+	/* This WARN has probably outlived its usefulness (callers already
+	 * WARN if they don't find the GGTT vma they expect). When removing,
+	 * remember to remove the pre-check in is_pin_display() as well */
 	if (WARN_ON(list_empty(&obj->vma_list)))
 		return NULL;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index d72db15afa02..a5ddf3bce9c3 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -178,7 +178,7 @@ static int get_context_size(struct drm_device *dev)
 
 void i915_gem_context_free(struct kref *ctx_ref)
 {
-	struct i915_hw_context *ctx = container_of(ctx_ref,
+	struct intel_context *ctx = container_of(ctx_ref,
 						   typeof(*ctx), ref);
 	struct i915_hw_ppgtt *ppgtt = NULL;
 
@@ -199,7 +199,7 @@ void i915_gem_context_free(struct kref *ctx_ref)
 }
 
 static struct i915_hw_ppgtt *
-create_vm_for_ctx(struct drm_device *dev, struct i915_hw_context *ctx)
+create_vm_for_ctx(struct drm_device *dev, struct intel_context *ctx)
 {
 	struct i915_hw_ppgtt *ppgtt;
 	int ret;
@@ -218,12 +218,12 @@ create_vm_for_ctx(struct drm_device *dev, struct i915_hw_context *ctx)
 	return ppgtt;
 }
 
-static struct i915_hw_context *
+static struct intel_context *
 __create_hw_context(struct drm_device *dev,
 		  struct drm_i915_file_private *file_priv)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
@@ -240,7 +240,15 @@ __create_hw_context(struct drm_device *dev,
 			goto err_out;
 		}
 
-		if (INTEL_INFO(dev)->gen >= 7) {
+		/*
+		 * Try to make the context utilize L3 as well as LLC.
+		 *
+		 * On VLV we don't have L3 controls in the PTEs so we
+		 * shouldn't touch the cache level, especially as that
+		 * would make the object snooped which might have a
+		 * negative performance impact.
+		 */
+		if (INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev)) {
 			ret = i915_gem_object_set_cache_level(ctx->obj,
 							      I915_CACHE_L3_LLC);
 			/* Failure shouldn't ever happen this early */
@@ -277,14 +285,14 @@ err_out:
  * context state of the GPU for applications that don't utilize HW contexts, as
  * well as an idle case.
  */
-static struct i915_hw_context *
+static struct intel_context *
 i915_gem_create_context(struct drm_device *dev,
 			struct drm_i915_file_private *file_priv,
 			bool create_vm)
 {
 	const bool is_global_default_ctx = file_priv == NULL;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret = 0;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
@@ -356,8 +364,8 @@ void i915_gem_context_reset(struct drm_device *dev)
 	/* Prevent the hardware from restoring the last context (which hung) on
 	 * the next switch */
 	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_ring_buffer *ring = &dev_priv->ring[i];
-		struct i915_hw_context *dctx = ring->default_context;
+		struct intel_engine_cs *ring = &dev_priv->ring[i];
+		struct intel_context *dctx = ring->default_context;
 
 		/* Do a fake switch to the default context */
 		if (ring->last_context == dctx)
@@ -383,7 +391,7 @@ void i915_gem_context_reset(struct drm_device *dev)
 int i915_gem_context_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int i;
 
 	/* Init should only be called once per module load. Eventually the
@@ -418,7 +426,7 @@ int i915_gem_context_init(struct drm_device *dev)
 void i915_gem_context_fini(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *dctx = dev_priv->ring[RCS].default_context;
+	struct intel_context *dctx = dev_priv->ring[RCS].default_context;
 	int i;
 
 	if (dctx->obj) {
@@ -441,10 +449,12 @@ void i915_gem_context_fini(struct drm_device *dev)
 			i915_gem_context_unreference(dctx);
 			dev_priv->ring[RCS].last_context = NULL;
 		}
+
+		i915_gem_object_ggtt_unpin(dctx->obj);
 	}
 
 	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_ring_buffer *ring = &dev_priv->ring[i];
+		struct intel_engine_cs *ring = &dev_priv->ring[i];
 
 		if (ring->last_context)
 			i915_gem_context_unreference(ring->last_context);
@@ -453,13 +463,12 @@ void i915_gem_context_fini(struct drm_device *dev)
 		ring->last_context = NULL;
 	}
 
-	i915_gem_object_ggtt_unpin(dctx->obj);
 	i915_gem_context_unreference(dctx);
 }
 
 int i915_gem_context_enable(struct drm_i915_private *dev_priv)
 {
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i;
 
 	/* This is the only place the aliasing PPGTT gets enabled, which means
@@ -486,11 +495,7 @@ int i915_gem_context_enable(struct drm_i915_private *dev_priv)
 
 static int context_idr_cleanup(int id, void *p, void *data)
 {
-	struct i915_hw_context *ctx = p;
-
-	/* Ignore the default context because close will handle it */
-	if (i915_gem_context_is_default(ctx))
-		return 0;
+	struct intel_context *ctx = p;
 
 	i915_gem_context_unreference(ctx);
 	return 0;
@@ -499,17 +504,17 @@ static int context_idr_cleanup(int id, void *p, void *data)
 int i915_gem_context_open(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct intel_context *ctx;
 
 	idr_init(&file_priv->context_idr);
 
 	mutex_lock(&dev->struct_mutex);
-	file_priv->private_default_ctx =
-		i915_gem_create_context(dev, file_priv, USES_FULL_PPGTT(dev));
+	ctx = i915_gem_create_context(dev, file_priv, USES_FULL_PPGTT(dev));
 	mutex_unlock(&dev->struct_mutex);
 
-	if (IS_ERR(file_priv->private_default_ctx)) {
+	if (IS_ERR(ctx)) {
 		idr_destroy(&file_priv->context_idr);
-		return PTR_ERR(file_priv->private_default_ctx);
+		return PTR_ERR(ctx);
 	}
 
 	return 0;
@@ -521,16 +526,14 @@ void i915_gem_context_close(struct drm_device *dev, struct drm_file *file)
 
 	idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL);
 	idr_destroy(&file_priv->context_idr);
-
-	i915_gem_context_unreference(file_priv->private_default_ctx);
 }
 
-struct i915_hw_context *
+struct intel_context *
 i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id)
 {
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 
-	ctx = (struct i915_hw_context *)idr_find(&file_priv->context_idr, id);
+	ctx = (struct intel_context *)idr_find(&file_priv->context_idr, id);
 	if (!ctx)
 		return ERR_PTR(-ENOENT);
 
@@ -538,8 +541,8 @@ i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id)
 }
 
 static inline int
-mi_set_context(struct intel_ring_buffer *ring,
-	       struct i915_hw_context *new_context,
+mi_set_context(struct intel_engine_cs *ring,
+	       struct intel_context *new_context,
 	       u32 hw_flags)
 {
 	int ret;
@@ -549,7 +552,7 @@ mi_set_context(struct intel_ring_buffer *ring,
 	 * explicitly, so we rely on the value at ring init, stored in
 	 * itlb_before_ctx_switch.
 	 */
-	if (IS_GEN6(ring->dev) && ring->itlb_before_ctx_switch) {
+	if (IS_GEN6(ring->dev)) {
 		ret = ring->flush(ring, I915_GEM_GPU_DOMAINS, 0);
 		if (ret)
 			return ret;
@@ -559,8 +562,8 @@ mi_set_context(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw */
-	if (IS_GEN7(ring->dev))
+	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */
+	if (INTEL_INFO(ring->dev)->gen >= 7)
 		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE);
 	else
 		intel_ring_emit(ring, MI_NOOP);
@@ -578,7 +581,7 @@ mi_set_context(struct intel_ring_buffer *ring,
 	 */
 	intel_ring_emit(ring, MI_NOOP);
 
-	if (IS_GEN7(ring->dev))
+	if (INTEL_INFO(ring->dev)->gen >= 7)
 		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE);
 	else
 		intel_ring_emit(ring, MI_NOOP);
@@ -588,13 +591,14 @@ mi_set_context(struct intel_ring_buffer *ring,
 	return ret;
 }
 
-static int do_switch(struct intel_ring_buffer *ring,
-		     struct i915_hw_context *to)
+static int do_switch(struct intel_engine_cs *ring,
+		     struct intel_context *to)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	struct i915_hw_context *from = ring->last_context;
+	struct intel_context *from = ring->last_context;
 	struct i915_hw_ppgtt *ppgtt = ctx_to_ppgtt(to);
 	u32 hw_flags = 0;
+	bool uninitialized = false;
 	int ret, i;
 
 	if (from != NULL && ring == &dev_priv->ring[RCS]) {
@@ -693,6 +697,7 @@ static int do_switch(struct intel_ring_buffer *ring,
 		i915_gem_context_unreference(from);
 	}
 
+	uninitialized = !to->is_initialized && from == NULL;
 	to->is_initialized = true;
 
 done:
@@ -700,6 +705,12 @@ done:
 	ring->last_context = to;
 	to->last_ring = ring;
 
+	if (uninitialized) {
+		ret = i915_gem_render_state_init(ring);
+		if (ret)
+			DRM_ERROR("init render state: %d\n", ret);
+	}
+
 	return 0;
 
 unpin_out:
@@ -718,8 +729,8 @@ unpin_out:
  * it will have a refoucnt > 1. This allows us to destroy the context abstract
  * object while letting the normal object tracking destroy the backing BO.
  */
-int i915_switch_context(struct intel_ring_buffer *ring,
-			struct i915_hw_context *to)
+int i915_switch_context(struct intel_engine_cs *ring,
+			struct intel_context *to)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
@@ -748,7 +759,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_gem_context_create *args = data;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	if (!hw_context_enabled(dev))
@@ -774,7 +785,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_gem_context_destroy *args = data;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	if (args->ctx_id == DEFAULT_CONTEXT_ID)
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index 9bb533e0d762..580aa42443ed 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -161,12 +161,8 @@ static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr)
 {
 	struct drm_i915_gem_object *obj = dma_buf_to_obj(dma_buf);
 	struct drm_device *dev = obj->base.dev;
-	int ret;
-
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return;
 
+	mutex_lock(&dev->struct_mutex);
 	if (--obj->vmapping_count == 0) {
 		vunmap(obj->dma_buf_vmapping);
 		obj->dma_buf_vmapping = NULL;
@@ -233,6 +229,14 @@ static const struct dma_buf_ops i915_dmabuf_ops =  {
 struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
 				      struct drm_gem_object *gem_obj, int flags)
 {
+	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
+
+	if (obj->ops->dmabuf_export) {
+		int ret = obj->ops->dmabuf_export(obj);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+
 	return dma_buf_export(gem_obj, &i915_dmabuf_ops, gem_obj->size, flags);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 20fef6c50267..3a30133f93e8 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -265,10 +265,12 @@ static inline int use_cpu_reloc(struct drm_i915_gem_object *obj)
 
 static int
 relocate_entry_cpu(struct drm_i915_gem_object *obj,
-		   struct drm_i915_gem_relocation_entry *reloc)
+		   struct drm_i915_gem_relocation_entry *reloc,
+		   uint64_t target_offset)
 {
 	struct drm_device *dev = obj->base.dev;
 	uint32_t page_offset = offset_in_page(reloc->offset);
+	uint64_t delta = reloc->delta + target_offset;
 	char *vaddr;
 	int ret;
 
@@ -278,7 +280,7 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj,
 
 	vaddr = kmap_atomic(i915_gem_object_get_page(obj,
 				reloc->offset >> PAGE_SHIFT));
-	*(uint32_t *)(vaddr + page_offset) = reloc->delta;
+	*(uint32_t *)(vaddr + page_offset) = lower_32_bits(delta);
 
 	if (INTEL_INFO(dev)->gen >= 8) {
 		page_offset = offset_in_page(page_offset + sizeof(uint32_t));
@@ -289,7 +291,7 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj,
 			    (reloc->offset + sizeof(uint32_t)) >> PAGE_SHIFT));
 		}
 
-		*(uint32_t *)(vaddr + page_offset) = 0;
+		*(uint32_t *)(vaddr + page_offset) = upper_32_bits(delta);
 	}
 
 	kunmap_atomic(vaddr);
@@ -299,10 +301,12 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj,
 
 static int
 relocate_entry_gtt(struct drm_i915_gem_object *obj,
-		   struct drm_i915_gem_relocation_entry *reloc)
+		   struct drm_i915_gem_relocation_entry *reloc,
+		   uint64_t target_offset)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint64_t delta = reloc->delta + target_offset;
 	uint32_t __iomem *reloc_entry;
 	void __iomem *reloc_page;
 	int ret;
@@ -321,7 +325,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 			reloc->offset & PAGE_MASK);
 	reloc_entry = (uint32_t __iomem *)
 		(reloc_page + offset_in_page(reloc->offset));
-	iowrite32(reloc->delta, reloc_entry);
+	iowrite32(lower_32_bits(delta), reloc_entry);
 
 	if (INTEL_INFO(dev)->gen >= 8) {
 		reloc_entry += 1;
@@ -334,7 +338,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 			reloc_entry = reloc_page;
 		}
 
-		iowrite32(0, reloc_entry);
+		iowrite32(upper_32_bits(delta), reloc_entry);
 	}
 
 	io_mapping_unmap_atomic(reloc_page);
@@ -351,7 +355,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	struct drm_gem_object *target_obj;
 	struct drm_i915_gem_object *target_i915_obj;
 	struct i915_vma *target_vma;
-	uint32_t target_offset;
+	uint64_t target_offset;
 	int ret;
 
 	/* we've already hold a reference to all valid objects */
@@ -429,11 +433,10 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	if (obj->active && in_atomic())
 		return -EFAULT;
 
-	reloc->delta += target_offset;
 	if (use_cpu_reloc(obj))
-		ret = relocate_entry_cpu(obj, reloc);
+		ret = relocate_entry_cpu(obj, reloc, target_offset);
 	else
-		ret = relocate_entry_gtt(obj, reloc);
+		ret = relocate_entry_gtt(obj, reloc, target_offset);
 
 	if (ret)
 		return ret;
@@ -541,7 +544,7 @@ need_reloc_mappable(struct i915_vma *vma)
 
 static int
 i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
-				struct intel_ring_buffer *ring,
+				struct intel_engine_cs *ring,
 				bool *need_reloc)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
@@ -628,7 +631,7 @@ eb_vma_misplaced(struct i915_vma *vma, bool has_fenced_gpu_access)
 }
 
 static int
-i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
+i915_gem_execbuffer_reserve(struct intel_engine_cs *ring,
 			    struct list_head *vmas,
 			    bool *need_relocs)
 {
@@ -642,6 +645,8 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 	if (list_empty(vmas))
 		return 0;
 
+	i915_gem_retire_requests_ring(ring);
+
 	vm = list_first_entry(vmas, struct i915_vma, exec_list)->vm;
 
 	INIT_LIST_HEAD(&ordered_vmas);
@@ -727,7 +732,7 @@ static int
 i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 				  struct drm_i915_gem_execbuffer2 *args,
 				  struct drm_file *file,
-				  struct intel_ring_buffer *ring,
+				  struct intel_engine_cs *ring,
 				  struct eb_vmas *eb,
 				  struct drm_i915_gem_exec_object2 *exec)
 {
@@ -843,7 +848,7 @@ err:
 }
 
 static int
-i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
+i915_gem_execbuffer_move_to_gpu(struct intel_engine_cs *ring,
 				struct list_head *vmas)
 {
 	struct i915_vma *vma;
@@ -926,11 +931,11 @@ validate_exec_list(struct drm_i915_gem_exec_object2 *exec,
 	return 0;
 }
 
-static struct i915_hw_context *
+static struct intel_context *
 i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
-			  struct intel_ring_buffer *ring, const u32 ctx_id)
+			  struct intel_engine_cs *ring, const u32 ctx_id)
 {
-	struct i915_hw_context *ctx = NULL;
+	struct intel_context *ctx = NULL;
 	struct i915_ctx_hang_stats *hs;
 
 	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_ID)
@@ -951,7 +956,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 
 static void
 i915_gem_execbuffer_move_to_active(struct list_head *vmas,
-				   struct intel_ring_buffer *ring)
+				   struct intel_engine_cs *ring)
 {
 	struct i915_vma *vma;
 
@@ -974,6 +979,9 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 			if (i915_gem_obj_ggtt_bound(obj) &&
 			    i915_gem_obj_to_ggtt(obj)->pin_count)
 				intel_mark_fb_busy(obj, ring);
+
+			/* update for the implicit flush after a batch */
+			obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
 		}
 
 		trace_i915_gem_object_change_domain(obj, old_read, old_write);
@@ -983,7 +991,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 static void
 i915_gem_execbuffer_retire_commands(struct drm_device *dev,
 				    struct drm_file *file,
-				    struct intel_ring_buffer *ring,
+				    struct intel_engine_cs *ring,
 				    struct drm_i915_gem_object *obj)
 {
 	/* Unconditionally force add_request to emit a full flush. */
@@ -995,13 +1003,15 @@ i915_gem_execbuffer_retire_commands(struct drm_device *dev,
 
 static int
 i915_reset_gen7_sol_offsets(struct drm_device *dev,
-			    struct intel_ring_buffer *ring)
+			    struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret, i;
 
-	if (!IS_GEN7(dev) || ring != &dev_priv->ring[RCS])
-		return 0;
+	if (!IS_GEN7(dev) || ring != &dev_priv->ring[RCS]) {
+		DRM_DEBUG("sol reset is gen7/rcs only\n");
+		return -EINVAL;
+	}
 
 	ret = intel_ring_begin(ring, 4 * 3);
 	if (ret)
@@ -1018,6 +1028,37 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 	return 0;
 }
 
+/**
+ * Find one BSD ring to dispatch the corresponding BSD command.
+ * The Ring ID is returned.
+ */
+static int gen8_dispatch_bsd_ring(struct drm_device *dev,
+				  struct drm_file *file)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+
+	/* Check whether the file_priv is using one ring */
+	if (file_priv->bsd_ring)
+		return file_priv->bsd_ring->id;
+	else {
+		/* If no, use the ping-pong mechanism to select one ring */
+		int ring_id;
+
+		mutex_lock(&dev->struct_mutex);
+		if (dev_priv->mm.bsd_ring_dispatch_index == 0) {
+			ring_id = VCS;
+			dev_priv->mm.bsd_ring_dispatch_index = 1;
+		} else {
+			ring_id = VCS2;
+			dev_priv->mm.bsd_ring_dispatch_index = 0;
+		}
+		file_priv->bsd_ring = &dev_priv->ring[ring_id];
+		mutex_unlock(&dev->struct_mutex);
+		return ring_id;
+	}
+}
+
 static struct drm_i915_gem_object *
 eb_get_batch(struct eb_vmas *eb)
 {
@@ -1047,11 +1088,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	struct eb_vmas *eb;
 	struct drm_i915_gem_object *batch_obj;
 	struct drm_clip_rect *cliprects = NULL;
-	struct intel_ring_buffer *ring;
-	struct i915_hw_context *ctx;
+	struct intel_engine_cs *ring;
+	struct intel_context *ctx;
 	struct i915_address_space *vm;
 	const u32 ctx_id = i915_execbuffer2_get_context_id(*args);
-	u32 exec_start = args->batch_start_offset, exec_len;
+	u64 exec_start = args->batch_start_offset, exec_len;
 	u32 mask, flags;
 	int ret, mode, i;
 	bool need_relocs;
@@ -1073,7 +1114,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (args->flags & I915_EXEC_IS_PINNED)
 		flags |= I915_DISPATCH_PINNED;
 
-	if ((args->flags & I915_EXEC_RING_MASK) > I915_NUM_RINGS) {
+	if ((args->flags & I915_EXEC_RING_MASK) > LAST_USER_RING) {
 		DRM_DEBUG("execbuf with unknown ring: %d\n",
 			  (int)(args->flags & I915_EXEC_RING_MASK));
 		return -EINVAL;
@@ -1081,7 +1122,14 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	if ((args->flags & I915_EXEC_RING_MASK) == I915_EXEC_DEFAULT)
 		ring = &dev_priv->ring[RCS];
-	else
+	else if ((args->flags & I915_EXEC_RING_MASK) == I915_EXEC_BSD) {
+		if (HAS_BSD2(dev)) {
+			int ring_id;
+			ring_id = gen8_dispatch_bsd_ring(dev, file);
+			ring = &dev_priv->ring[ring_id];
+		} else
+			ring = &dev_priv->ring[VCS];
+	} else
 		ring = &dev_priv->ring[(args->flags & I915_EXEC_RING_MASK) - 1];
 
 	if (!intel_ring_initialized(ring)) {
@@ -1096,14 +1144,22 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	case I915_EXEC_CONSTANTS_REL_GENERAL:
 	case I915_EXEC_CONSTANTS_ABSOLUTE:
 	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (ring == &dev_priv->ring[RCS] &&
-		    mode != dev_priv->relative_constants_mode) {
-			if (INTEL_INFO(dev)->gen < 4)
+		if (mode != 0 && ring != &dev_priv->ring[RCS]) {
+			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
+			return -EINVAL;
+		}
+
+		if (mode != dev_priv->relative_constants_mode) {
+			if (INTEL_INFO(dev)->gen < 4) {
+				DRM_DEBUG("no rel constants on pre-gen4\n");
 				return -EINVAL;
+			}
 
 			if (INTEL_INFO(dev)->gen > 5 &&
-			    mode == I915_EXEC_CONSTANTS_REL_SURFACE)
+			    mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
+				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
 				return -EINVAL;
+			}
 
 			/* The HW changed the meaning on this bit on gen6 */
 			if (INTEL_INFO(dev)->gen >= 6)
@@ -1151,6 +1207,16 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 			ret = -EFAULT;
 			goto pre_mutex_err;
 		}
+	} else {
+		if (args->DR4 == 0xffffffff) {
+			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
+			args->DR4 = 0;
+		}
+
+		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
+			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
+			return -EINVAL;
+		}
 	}
 
 	intel_runtime_pm_get(dev_priv);
@@ -1170,7 +1236,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		mutex_unlock(&dev->struct_mutex);
 		ret = PTR_ERR(ctx);
 		goto pre_mutex_err;
-	} 
+	}
 
 	i915_gem_context_reference(ctx);
 
@@ -1180,6 +1246,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	eb = eb_create(args);
 	if (eb == NULL) {
+		i915_gem_context_unreference(ctx);
 		mutex_unlock(&dev->struct_mutex);
 		ret = -ENOMEM;
 		goto pre_mutex_err;
@@ -1430,6 +1497,11 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
+	if (args->rsvd2 != 0) {
+		DRM_DEBUG("dirty rvsd2 field\n");
+		return -EINVAL;
+	}
+
 	exec2_list = kmalloc(sizeof(*exec2_list)*args->buffer_count,
 			     GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
 	if (exec2_list == NULL)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 5deb22864c52..8b3cde703364 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -30,7 +30,8 @@
 #include "i915_trace.h"
 #include "intel_drv.h"
 
-static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv);
+static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv);
+static void chv_setup_private_ppat(struct drm_i915_private *dev_priv);
 
 bool intel_enable_ppgtt(struct drm_device *dev, bool full)
 {
@@ -65,59 +66,6 @@ static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
 	return HAS_ALIASING_PPGTT(dev) ? 1 : 0;
 }
 
-#define GEN6_PPGTT_PD_ENTRIES 512
-#define I915_PPGTT_PT_ENTRIES (PAGE_SIZE / sizeof(gen6_gtt_pte_t))
-typedef uint64_t gen8_gtt_pte_t;
-typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
-
-/* PPGTT stuff */
-#define GEN6_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0xff0))
-#define HSW_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0x7f0))
-
-#define GEN6_PDE_VALID			(1 << 0)
-/* gen6+ has bit 11-4 for physical addr bit 39-32 */
-#define GEN6_PDE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
-
-#define GEN6_PTE_VALID			(1 << 0)
-#define GEN6_PTE_UNCACHED		(1 << 1)
-#define HSW_PTE_UNCACHED		(0)
-#define GEN6_PTE_CACHE_LLC		(2 << 1)
-#define GEN7_PTE_CACHE_L3_LLC		(3 << 1)
-#define GEN6_PTE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
-#define HSW_PTE_ADDR_ENCODE(addr)	HSW_GTT_ADDR_ENCODE(addr)
-
-/* Cacheability Control is a 4-bit value. The low three bits are stored in *
- * bits 3:1 of the PTE, while the fourth bit is stored in bit 11 of the PTE.
- */
-#define HSW_CACHEABILITY_CONTROL(bits)	((((bits) & 0x7) << 1) | \
-					 (((bits) & 0x8) << (11 - 3)))
-#define HSW_WB_LLC_AGE3			HSW_CACHEABILITY_CONTROL(0x2)
-#define HSW_WB_LLC_AGE0			HSW_CACHEABILITY_CONTROL(0x3)
-#define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
-#define HSW_WB_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x8)
-#define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
-#define HSW_WT_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x7)
-
-#define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
-#define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
-
-/* GEN8 legacy style addressis defined as a 3 level page table:
- * 31:30 | 29:21 | 20:12 |  11:0
- * PDPE  |  PDE  |  PTE  | offset
- * The difference as compared to normal x86 3 level page table is the PDPEs are
- * programmed via register.
- */
-#define GEN8_PDPE_SHIFT			30
-#define GEN8_PDPE_MASK			0x3
-#define GEN8_PDE_SHIFT			21
-#define GEN8_PDE_MASK			0x1ff
-#define GEN8_PTE_SHIFT			12
-#define GEN8_PTE_MASK			0x1ff
-
-#define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
-#define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
-#define PPAT_CACHED_INDEX		_PAGE_PAT /* WB LLCeLLC */
-#define PPAT_DISPLAY_ELLC_INDEX		_PAGE_PCD /* WT eLLC */
 
 static void ppgtt_bind_vma(struct i915_vma *vma,
 			   enum i915_cache_level cache_level,
@@ -131,10 +79,19 @@ static inline gen8_gtt_pte_t gen8_pte_encode(dma_addr_t addr,
 {
 	gen8_gtt_pte_t pte = valid ? _PAGE_PRESENT | _PAGE_RW : 0;
 	pte |= addr;
-	if (level != I915_CACHE_NONE)
-		pte |= PPAT_CACHED_INDEX;
-	else
+
+	switch (level) {
+	case I915_CACHE_NONE:
 		pte |= PPAT_UNCACHED_INDEX;
+		break;
+	case I915_CACHE_WT:
+		pte |= PPAT_DISPLAY_ELLC_INDEX;
+		break;
+	default:
+		pte |= PPAT_CACHED_INDEX;
+		break;
+	}
+
 	return pte;
 }
 
@@ -197,9 +154,6 @@ static gen6_gtt_pte_t ivb_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
-#define BYT_PTE_WRITEABLE		(1 << 1)
-#define BYT_PTE_SNOOPED_BY_CPU_CACHES	(1 << 2)
-
 static gen6_gtt_pte_t byt_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
 				     bool valid)
@@ -253,7 +207,7 @@ static gen6_gtt_pte_t iris_pte_encode(dma_addr_t addr,
 }
 
 /* Broadwell Page Directory Pointer Descriptors */
-static int gen8_write_pdp(struct intel_ring_buffer *ring, unsigned entry,
+static int gen8_write_pdp(struct intel_engine_cs *ring, unsigned entry,
 			   uint64_t val, bool synchronous)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -283,7 +237,7 @@ static int gen8_write_pdp(struct intel_ring_buffer *ring, unsigned entry,
 }
 
 static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_ring_buffer *ring,
+			  struct intel_engine_cs *ring,
 			  bool synchronous)
 {
 	int i, ret;
@@ -332,6 +286,8 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 			num_entries--;
 		}
 
+		if (!HAS_LLC(ppgtt->base.dev))
+			drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
 		kunmap_atomic(pt_vaddr);
 
 		pte = 0;
@@ -368,6 +324,8 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 			gen8_pte_encode(sg_page_iter_dma_address(&sg_iter),
 					cache_level, true);
 		if (++pte == GEN8_PTES_PER_PAGE) {
+			if (!HAS_LLC(ppgtt->base.dev))
+				drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
 			kunmap_atomic(pt_vaddr);
 			pt_vaddr = NULL;
 			if (++pde == GEN8_PDES_PER_PAGE) {
@@ -377,8 +335,11 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 			pte = 0;
 		}
 	}
-	if (pt_vaddr)
+	if (pt_vaddr) {
+		if (!HAS_LLC(ppgtt->base.dev))
+			drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
 		kunmap_atomic(pt_vaddr);
+	}
 }
 
 static void gen8_free_page_tables(struct page **pt_pages)
@@ -641,6 +602,8 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt, uint64_t size)
 			pd_vaddr[j] = gen8_pde_encode(ppgtt->base.dev, addr,
 						      I915_CACHE_LLC);
 		}
+		if (!HAS_LLC(ppgtt->base.dev))
+			drm_clflush_virt_range(pd_vaddr, PAGE_SIZE);
 		kunmap_atomic(pd_vaddr);
 	}
 
@@ -753,7 +716,7 @@ static uint32_t get_pd_offset(struct i915_hw_ppgtt *ppgtt)
 }
 
 static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			 struct intel_ring_buffer *ring,
+			 struct intel_engine_cs *ring,
 			 bool synchronous)
 {
 	struct drm_device *dev = ppgtt->base.dev;
@@ -797,7 +760,7 @@ static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 }
 
 static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_ring_buffer *ring,
+			  struct intel_engine_cs *ring,
 			  bool synchronous)
 {
 	struct drm_device *dev = ppgtt->base.dev;
@@ -848,7 +811,7 @@ static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 }
 
 static int gen6_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_ring_buffer *ring,
+			  struct intel_engine_cs *ring,
 			  bool synchronous)
 {
 	struct drm_device *dev = ppgtt->base.dev;
@@ -869,7 +832,7 @@ static int gen8_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int j, ret;
 
 	for_each_ring(ring, dev_priv, j) {
@@ -899,7 +862,7 @@ static int gen7_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	uint32_t ecochk, ecobits;
 	int i;
 
@@ -938,7 +901,7 @@ static int gen6_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	uint32_t ecochk, gab_ctl, ecobits;
 	int i;
 
@@ -1067,8 +1030,6 @@ static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
 
 static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)
 {
-#define GEN6_PD_ALIGN (PAGE_SIZE * 16)
-#define GEN6_PD_SIZE (GEN6_PPGTT_PD_ENTRIES * PAGE_SIZE)
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool retried = false;
@@ -1084,8 +1045,7 @@ alloc:
 						  &ppgtt->node, GEN6_PD_SIZE,
 						  GEN6_PD_ALIGN, 0,
 						  0, dev_priv->gtt.base.total,
-						  DRM_MM_SEARCH_DEFAULT,
-						  DRM_MM_CREATE_DEFAULT);
+						  DRM_MM_TOPDOWN);
 	if (ret == -ENOSPC && !retried) {
 		ret = i915_gem_evict_something(dev, &dev_priv->gtt.base,
 					       GEN6_PD_SIZE, GEN6_PD_ALIGN,
@@ -1311,7 +1271,7 @@ static void undo_idling(struct drm_i915_private *dev_priv, bool interruptible)
 void i915_check_and_clear_faults(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	if (INTEL_INFO(dev)->gen < 6)
@@ -1386,7 +1346,11 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 
 
 	if (INTEL_INFO(dev)->gen >= 8) {
-		gen8_setup_private_ppat(dev_priv);
+		if (IS_CHERRYVIEW(dev))
+			chv_setup_private_ppat(dev_priv);
+		else
+			bdw_setup_private_ppat(dev_priv);
+
 		return;
 	}
 
@@ -1438,7 +1402,7 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 		(gen8_gtt_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
 	int i = 0;
 	struct sg_page_iter sg_iter;
-	dma_addr_t addr;
+	dma_addr_t addr = 0;
 
 	for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
 		addr = sg_dma_address(sg_iter.sg) +
@@ -1811,9 +1775,27 @@ static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
 	bdw_gmch_ctl &= BDW_GMCH_GGMS_MASK;
 	if (bdw_gmch_ctl)
 		bdw_gmch_ctl = 1 << bdw_gmch_ctl;
+
+#ifdef CONFIG_X86_32
+	/* Limit 32b platforms to a 2GB GGTT: 4 << 20 / pte size * PAGE_SIZE */
+	if (bdw_gmch_ctl > 4)
+		bdw_gmch_ctl = 4;
+#endif
+
 	return bdw_gmch_ctl << 20;
 }
 
+static inline unsigned int chv_get_total_gtt_size(u16 gmch_ctrl)
+{
+	gmch_ctrl >>= SNB_GMCH_GGMS_SHIFT;
+	gmch_ctrl &= SNB_GMCH_GGMS_MASK;
+
+	if (gmch_ctrl)
+		return 1 << (20 + gmch_ctrl);
+
+	return 0;
+}
+
 static inline size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
 {
 	snb_gmch_ctl >>= SNB_GMCH_GMS_SHIFT;
@@ -1828,6 +1810,24 @@ static inline size_t gen8_get_stolen_size(u16 bdw_gmch_ctl)
 	return bdw_gmch_ctl << 25; /* 32 MB units */
 }
 
+static size_t chv_get_stolen_size(u16 gmch_ctrl)
+{
+	gmch_ctrl >>= SNB_GMCH_GMS_SHIFT;
+	gmch_ctrl &= SNB_GMCH_GMS_MASK;
+
+	/*
+	 * 0x0  to 0x10: 32MB increments starting at 0MB
+	 * 0x11 to 0x16: 4MB increments starting at 8MB
+	 * 0x17 to 0x1d: 4MB increments start at 36MB
+	 */
+	if (gmch_ctrl < 0x11)
+		return gmch_ctrl << 25;
+	else if (gmch_ctrl < 0x17)
+		return (gmch_ctrl - 0x11 + 2) << 22;
+	else
+		return (gmch_ctrl - 0x17 + 9) << 22;
+}
+
 static int ggtt_probe_common(struct drm_device *dev,
 			     size_t gtt_size)
 {
@@ -1858,19 +1858,8 @@ static int ggtt_probe_common(struct drm_device *dev,
 /* The GGTT and PPGTT need a private PPAT setup in order to handle cacheability
  * bits. When using advanced contexts each context stores its own PAT, but
  * writing this data shouldn't be harmful even in those cases. */
-static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv)
-{
-#define GEN8_PPAT_UC		(0<<0)
-#define GEN8_PPAT_WC		(1<<0)
-#define GEN8_PPAT_WT		(2<<0)
-#define GEN8_PPAT_WB		(3<<0)
-#define GEN8_PPAT_ELLC_OVERRIDE	(0<<2)
-/* FIXME(BDW): Bspec is completely confused about cache control bits. */
-#define GEN8_PPAT_LLC		(1<<2)
-#define GEN8_PPAT_LLCELLC	(2<<2)
-#define GEN8_PPAT_LLCeLLC	(3<<2)
-#define GEN8_PPAT_AGE(x)	(x<<4)
-#define GEN8_PPAT(i, x) ((uint64_t) (x) << ((i) * 8))
+static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv)
+{
 	uint64_t pat;
 
 	pat = GEN8_PPAT(0, GEN8_PPAT_WB | GEN8_PPAT_LLC)     | /* for normal objects, no eLLC */
@@ -1888,6 +1877,33 @@ static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
 }
 
+static void chv_setup_private_ppat(struct drm_i915_private *dev_priv)
+{
+	uint64_t pat;
+
+	/*
+	 * Map WB on BDW to snooped on CHV.
+	 *
+	 * Only the snoop bit has meaning for CHV, the rest is
+	 * ignored.
+	 *
+	 * Note that the harware enforces snooping for all page
+	 * table accesses. The snoop bit is actually ignored for
+	 * PDEs.
+	 */
+	pat = GEN8_PPAT(0, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(1, 0) |
+	      GEN8_PPAT(2, 0) |
+	      GEN8_PPAT(3, 0) |
+	      GEN8_PPAT(4, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(5, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(6, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(7, CHV_PPAT_SNOOP);
+
+	I915_WRITE(GEN8_PRIVATE_PAT, pat);
+	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
+}
+
 static int gen8_gmch_probe(struct drm_device *dev,
 			   size_t *gtt_total,
 			   size_t *stolen,
@@ -1908,12 +1924,20 @@ static int gen8_gmch_probe(struct drm_device *dev,
 
 	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
 
-	*stolen = gen8_get_stolen_size(snb_gmch_ctl);
+	if (IS_CHERRYVIEW(dev)) {
+		*stolen = chv_get_stolen_size(snb_gmch_ctl);
+		gtt_size = chv_get_total_gtt_size(snb_gmch_ctl);
+	} else {
+		*stolen = gen8_get_stolen_size(snb_gmch_ctl);
+		gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
+	}
 
-	gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
 	*gtt_total = (gtt_size / sizeof(gen8_gtt_pte_t)) << PAGE_SHIFT;
 
-	gen8_setup_private_ppat(dev_priv);
+	if (IS_CHERRYVIEW(dev))
+		chv_setup_private_ppat(dev_priv);
+	else
+		bdw_setup_private_ppat(dev_priv);
 
 	ret = ggtt_probe_common(dev, gtt_size);
 
@@ -1968,7 +1992,10 @@ static void gen6_gmch_remove(struct i915_address_space *vm)
 
 	struct i915_gtt *gtt = container_of(vm, struct i915_gtt, base);
 
-	drm_mm_takedown(&vm->mm);
+	if (drm_mm_initialized(&vm->mm)) {
+		drm_mm_takedown(&vm->mm);
+		list_del(&vm->global_link);
+	}
 	iounmap(gtt->gsm);
 	teardown_scratch_page(vm->dev);
 }
@@ -2001,6 +2028,10 @@ static int i915_gmch_probe(struct drm_device *dev,
 
 static void i915_gmch_remove(struct i915_address_space *vm)
 {
+	if (drm_mm_initialized(&vm->mm)) {
+		drm_mm_takedown(&vm->mm);
+		list_del(&vm->global_link);
+	}
 	intel_gmch_remove();
 }
 
@@ -2043,6 +2074,10 @@ int i915_gem_gtt_init(struct drm_device *dev)
 		 gtt->base.total >> 20);
 	DRM_DEBUG_DRIVER("GMADR size = %ldM\n", gtt->mappable_end >> 20);
 	DRM_DEBUG_DRIVER("GTT stolen size = %zdM\n", gtt->stolen_size >> 20);
+#ifdef CONFIG_INTEL_IOMMU
+	if (intel_iommu_gfx_mapped)
+		DRM_INFO("VT-d active for gfx access\n");
+#endif
 	/*
 	 * i915.enable_ppgtt is read-only, so do an early pass to validate the
 	 * user's requested state against the hardware/driver capabilities.  We
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
new file mode 100644
index 000000000000..1b96a06be3cb
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -0,0 +1,284 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Please try to maintain the following order within this file unless it makes
+ * sense to do otherwise. From top to bottom:
+ * 1. typedefs
+ * 2. #defines, and macros
+ * 3. structure definitions
+ * 4. function prototypes
+ *
+ * Within each section, please try to order by generation in ascending order,
+ * from top to bottom (ie. gen6 on the top, gen8 on the bottom).
+ */
+
+#ifndef __I915_GEM_GTT_H__
+#define __I915_GEM_GTT_H__
+
+typedef uint32_t gen6_gtt_pte_t;
+typedef uint64_t gen8_gtt_pte_t;
+typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
+
+#define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
+
+#define I915_PPGTT_PT_ENTRIES		(PAGE_SIZE / sizeof(gen6_gtt_pte_t))
+/* gen6-hsw has bit 11-4 for physical addr bit 39-32 */
+#define GEN6_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0xff0))
+#define GEN6_PTE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
+#define GEN6_PDE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
+#define GEN6_PTE_CACHE_LLC		(2 << 1)
+#define GEN6_PTE_UNCACHED		(1 << 1)
+#define GEN6_PTE_VALID			(1 << 0)
+
+#define GEN6_PPGTT_PD_ENTRIES		512
+#define GEN6_PD_SIZE			(GEN6_PPGTT_PD_ENTRIES * PAGE_SIZE)
+#define GEN6_PD_ALIGN			(PAGE_SIZE * 16)
+#define GEN6_PDE_VALID			(1 << 0)
+
+#define GEN7_PTE_CACHE_L3_LLC		(3 << 1)
+
+#define BYT_PTE_SNOOPED_BY_CPU_CACHES	(1 << 2)
+#define BYT_PTE_WRITEABLE		(1 << 1)
+
+/* Cacheability Control is a 4-bit value. The low three bits are stored in bits
+ * 3:1 of the PTE, while the fourth bit is stored in bit 11 of the PTE.
+ */
+#define HSW_CACHEABILITY_CONTROL(bits)	((((bits) & 0x7) << 1) | \
+					 (((bits) & 0x8) << (11 - 3)))
+#define HSW_WB_LLC_AGE3			HSW_CACHEABILITY_CONTROL(0x2)
+#define HSW_WB_LLC_AGE0			HSW_CACHEABILITY_CONTROL(0x3)
+#define HSW_WB_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x8)
+#define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
+#define HSW_WT_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x7)
+#define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
+#define HSW_PTE_UNCACHED		(0)
+#define HSW_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0x7f0))
+#define HSW_PTE_ADDR_ENCODE(addr)	HSW_GTT_ADDR_ENCODE(addr)
+
+/* GEN8 legacy style address is defined as a 3 level page table:
+ * 31:30 | 29:21 | 20:12 |  11:0
+ * PDPE  |  PDE  |  PTE  | offset
+ * The difference as compared to normal x86 3 level page table is the PDPEs are
+ * programmed via register.
+ */
+#define GEN8_PDPE_SHIFT			30
+#define GEN8_PDPE_MASK			0x3
+#define GEN8_PDE_SHIFT			21
+#define GEN8_PDE_MASK			0x1ff
+#define GEN8_PTE_SHIFT			12
+#define GEN8_PTE_MASK			0x1ff
+#define GEN8_LEGACY_PDPS		4
+#define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
+#define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
+
+#define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
+#define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
+#define PPAT_CACHED_INDEX		_PAGE_PAT /* WB LLCeLLC */
+#define PPAT_DISPLAY_ELLC_INDEX		_PAGE_PCD /* WT eLLC */
+
+#define CHV_PPAT_SNOOP			(1<<6)
+#define GEN8_PPAT_AGE(x)		(x<<4)
+#define GEN8_PPAT_LLCeLLC		(3<<2)
+#define GEN8_PPAT_LLCELLC		(2<<2)
+#define GEN8_PPAT_LLC			(1<<2)
+#define GEN8_PPAT_WB			(3<<0)
+#define GEN8_PPAT_WT			(2<<0)
+#define GEN8_PPAT_WC			(1<<0)
+#define GEN8_PPAT_UC			(0<<0)
+#define GEN8_PPAT_ELLC_OVERRIDE		(0<<2)
+#define GEN8_PPAT(i, x)			((uint64_t) (x) << ((i) * 8))
+
+enum i915_cache_level;
+/**
+ * A VMA represents a GEM BO that is bound into an address space. Therefore, a
+ * VMA's presence cannot be guaranteed before binding, or after unbinding the
+ * object into/from the address space.
+ *
+ * To make things as simple as possible (ie. no refcounting), a VMA's lifetime
+ * will always be <= an objects lifetime. So object refcounting should cover us.
+ */
+struct i915_vma {
+	struct drm_mm_node node;
+	struct drm_i915_gem_object *obj;
+	struct i915_address_space *vm;
+
+	/** This object's place on the active/inactive lists */
+	struct list_head mm_list;
+
+	struct list_head vma_link; /* Link in the object's VMA list */
+
+	/** This vma's place in the batchbuffer or on the eviction list */
+	struct list_head exec_list;
+
+	/**
+	 * Used for performing relocations during execbuffer insertion.
+	 */
+	struct hlist_node exec_node;
+	unsigned long exec_handle;
+	struct drm_i915_gem_exec_object2 *exec_entry;
+
+	/**
+	 * How many users have pinned this object in GTT space. The following
+	 * users can each hold at most one reference: pwrite/pread, pin_ioctl
+	 * (via user_pin_count), execbuffer (objects are not allowed multiple
+	 * times for the same batchbuffer), and the framebuffer code. When
+	 * switching/pageflipping, the framebuffer code has at most two buffers
+	 * pinned per crtc.
+	 *
+	 * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3
+	 * bits with absolutely no headroom. So use 4 bits. */
+	unsigned int pin_count:4;
+#define DRM_I915_GEM_OBJECT_MAX_PIN_COUNT 0xf
+
+	/** Unmap an object from an address space. This usually consists of
+	 * setting the valid PTE entries to a reserved scratch page. */
+	void (*unbind_vma)(struct i915_vma *vma);
+	/* Map an object into an address space with the given cache flags. */
+#define GLOBAL_BIND (1<<0)
+	void (*bind_vma)(struct i915_vma *vma,
+			 enum i915_cache_level cache_level,
+			 u32 flags);
+};
+
+struct i915_address_space {
+	struct drm_mm mm;
+	struct drm_device *dev;
+	struct list_head global_link;
+	unsigned long start;		/* Start offset always 0 for dri2 */
+	size_t total;		/* size addr space maps (ex. 2GB for ggtt) */
+
+	struct {
+		dma_addr_t addr;
+		struct page *page;
+	} scratch;
+
+	/**
+	 * List of objects currently involved in rendering.
+	 *
+	 * Includes buffers having the contents of their GPU caches
+	 * flushed, not necessarily primitives.  last_rendering_seqno
+	 * represents when the rendering involved will be completed.
+	 *
+	 * A reference is held on the buffer while on this list.
+	 */
+	struct list_head active_list;
+
+	/**
+	 * LRU list of objects which are not in the ringbuffer and
+	 * are ready to unbind, but are still in the GTT.
+	 *
+	 * last_rendering_seqno is 0 while an object is in this list.
+	 *
+	 * A reference is not held on the buffer while on this list,
+	 * as merely being GTT-bound shouldn't prevent its being
+	 * freed, and we'll pull it off the list in the free path.
+	 */
+	struct list_head inactive_list;
+
+	/* FIXME: Need a more generic return type */
+	gen6_gtt_pte_t (*pte_encode)(dma_addr_t addr,
+				     enum i915_cache_level level,
+				     bool valid); /* Create a valid PTE */
+	void (*clear_range)(struct i915_address_space *vm,
+			    uint64_t start,
+			    uint64_t length,
+			    bool use_scratch);
+	void (*insert_entries)(struct i915_address_space *vm,
+			       struct sg_table *st,
+			       uint64_t start,
+			       enum i915_cache_level cache_level);
+	void (*cleanup)(struct i915_address_space *vm);
+};
+
+/* The Graphics Translation Table is the way in which GEN hardware translates a
+ * Graphics Virtual Address into a Physical Address. In addition to the normal
+ * collateral associated with any va->pa translations GEN hardware also has a
+ * portion of the GTT which can be mapped by the CPU and remain both coherent
+ * and correct (in cases like swizzling). That region is referred to as GMADR in
+ * the spec.
+ */
+struct i915_gtt {
+	struct i915_address_space base;
+	size_t stolen_size;		/* Total size of stolen memory */
+
+	unsigned long mappable_end;	/* End offset that we can CPU map */
+	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
+	phys_addr_t mappable_base;	/* PA of our GMADR */
+
+	/** "Graphics Stolen Memory" holds the global PTEs */
+	void __iomem *gsm;
+
+	bool do_idle_maps;
+
+	int mtrr;
+
+	/* global gtt ops */
+	int (*gtt_probe)(struct drm_device *dev, size_t *gtt_total,
+			  size_t *stolen, phys_addr_t *mappable_base,
+			  unsigned long *mappable_end);
+};
+
+struct i915_hw_ppgtt {
+	struct i915_address_space base;
+	struct kref ref;
+	struct drm_mm_node node;
+	unsigned num_pd_entries;
+	unsigned num_pd_pages; /* gen8+ */
+	union {
+		struct page **pt_pages;
+		struct page **gen8_pt_pages[GEN8_LEGACY_PDPS];
+	};
+	struct page *pd_pages;
+	union {
+		uint32_t pd_offset;
+		dma_addr_t pd_dma_addr[GEN8_LEGACY_PDPS];
+	};
+	union {
+		dma_addr_t *pt_dma_addr;
+		dma_addr_t *gen8_pt_dma_addr[4];
+	};
+
+	struct intel_context *ctx;
+
+	int (*enable)(struct i915_hw_ppgtt *ppgtt);
+	int (*switch_mm)(struct i915_hw_ppgtt *ppgtt,
+			 struct intel_engine_cs *ring,
+			 bool synchronous);
+	void (*debug_dump)(struct i915_hw_ppgtt *ppgtt, struct seq_file *m);
+};
+
+int i915_gem_gtt_init(struct drm_device *dev);
+void i915_gem_init_global_gtt(struct drm_device *dev);
+void i915_gem_setup_global_gtt(struct drm_device *dev, unsigned long start,
+			       unsigned long mappable_end, unsigned long end);
+
+bool intel_enable_ppgtt(struct drm_device *dev, bool full);
+int i915_gem_init_ppgtt(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt);
+
+void i915_check_and_clear_faults(struct drm_device *dev);
+void i915_gem_suspend_gtt_mappings(struct drm_device *dev);
+void i915_gem_restore_gtt_mappings(struct drm_device *dev);
+
+int __must_check i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj);
+void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj);
+
+#endif
diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.c b/drivers/gpu/drm/i915/i915_gem_render_state.c
new file mode 100644
index 000000000000..3521f998a178
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_render_state.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Mika Kuoppala <mika.kuoppala@intel.com>
+ *
+ */
+
+#include "i915_drv.h"
+#include "intel_renderstate.h"
+
+struct i915_render_state {
+	struct drm_i915_gem_object *obj;
+	unsigned long ggtt_offset;
+	void *batch;
+	u32 size;
+	u32 len;
+};
+
+static struct i915_render_state *render_state_alloc(struct drm_device *dev)
+{
+	struct i915_render_state *so;
+	struct page *page;
+	int ret;
+
+	so = kzalloc(sizeof(*so), GFP_KERNEL);
+	if (!so)
+		return ERR_PTR(-ENOMEM);
+
+	so->obj = i915_gem_alloc_object(dev, 4096);
+	if (so->obj == NULL) {
+		ret = -ENOMEM;
+		goto free;
+	}
+	so->size = 4096;
+
+	ret = i915_gem_obj_ggtt_pin(so->obj, 4096, 0);
+	if (ret)
+		goto free_gem;
+
+	BUG_ON(so->obj->pages->nents != 1);
+	page = sg_page(so->obj->pages->sgl);
+
+	so->batch = kmap(page);
+	if (!so->batch) {
+		ret = -ENOMEM;
+		goto unpin;
+	}
+
+	so->ggtt_offset = i915_gem_obj_ggtt_offset(so->obj);
+
+	return so;
+unpin:
+	i915_gem_object_ggtt_unpin(so->obj);
+free_gem:
+	drm_gem_object_unreference(&so->obj->base);
+free:
+	kfree(so);
+	return ERR_PTR(ret);
+}
+
+static void render_state_free(struct i915_render_state *so)
+{
+	kunmap(so->batch);
+	i915_gem_object_ggtt_unpin(so->obj);
+	drm_gem_object_unreference(&so->obj->base);
+	kfree(so);
+}
+
+static const struct intel_renderstate_rodata *
+render_state_get_rodata(struct drm_device *dev, const int gen)
+{
+	switch (gen) {
+	case 6:
+		return &gen6_null_state;
+	case 7:
+		return &gen7_null_state;
+	case 8:
+		return &gen8_null_state;
+	}
+
+	return NULL;
+}
+
+static int render_state_setup(const int gen,
+			      const struct intel_renderstate_rodata *rodata,
+			      struct i915_render_state *so)
+{
+	const u64 goffset = i915_gem_obj_ggtt_offset(so->obj);
+	u32 reloc_index = 0;
+	u32 * const d = so->batch;
+	unsigned int i = 0;
+	int ret;
+
+	if (!rodata || rodata->batch_items * 4 > so->size)
+		return -EINVAL;
+
+	ret = i915_gem_object_set_to_cpu_domain(so->obj, true);
+	if (ret)
+		return ret;
+
+	while (i < rodata->batch_items) {
+		u32 s = rodata->batch[i];
+
+		if (reloc_index < rodata->reloc_items &&
+		    i * 4  == rodata->reloc[reloc_index]) {
+
+			s += goffset & 0xffffffff;
+
+			/* We keep batch offsets max 32bit */
+			if (gen >= 8) {
+				if (i + 1 >= rodata->batch_items ||
+				    rodata->batch[i + 1] != 0)
+					return -EINVAL;
+
+				d[i] = s;
+				i++;
+				s = (goffset & 0xffffffff00000000ull) >> 32;
+			}
+
+			reloc_index++;
+		}
+
+		d[i] = s;
+		i++;
+	}
+
+	ret = i915_gem_object_set_to_gtt_domain(so->obj, false);
+	if (ret)
+		return ret;
+
+	if (rodata->reloc_items != reloc_index) {
+		DRM_ERROR("not all relocs resolved, %d out of %d\n",
+			  reloc_index, rodata->reloc_items);
+		return -EINVAL;
+	}
+
+	so->len = rodata->batch_items * 4;
+
+	return 0;
+}
+
+int i915_gem_render_state_init(struct intel_engine_cs *ring)
+{
+	const int gen = INTEL_INFO(ring->dev)->gen;
+	struct i915_render_state *so;
+	const struct intel_renderstate_rodata *rodata;
+	int ret;
+
+	if (WARN_ON(ring->id != RCS))
+		return -ENOENT;
+
+	rodata = render_state_get_rodata(ring->dev, gen);
+	if (rodata == NULL)
+		return 0;
+
+	so = render_state_alloc(ring->dev);
+	if (IS_ERR(so))
+		return PTR_ERR(so);
+
+	ret = render_state_setup(gen, rodata, so);
+	if (ret)
+		goto out;
+
+	ret = ring->dispatch_execbuffer(ring,
+					i915_gem_obj_ggtt_offset(so->obj),
+					so->len,
+					I915_DISPATCH_SECURE);
+	if (ret)
+		goto out;
+
+	i915_vma_move_to_active(i915_gem_obj_to_ggtt(so->obj), ring);
+
+	ret = __i915_add_request(ring, NULL, so->obj, NULL);
+	/* __i915_add_request moves object to inactive if it fails */
+out:
+	render_state_free(so);
+	return ret;
+}
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 62ef55ba061c..7465ab0fd396 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -74,6 +74,50 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 	if (base == 0)
 		return 0;
 
+	/* make sure we don't clobber the GTT if it's within stolen memory */
+	if (INTEL_INFO(dev)->gen <= 4 && !IS_G33(dev) && !IS_G4X(dev)) {
+		struct {
+			u32 start, end;
+		} stolen[2] = {
+			{ .start = base, .end = base + dev_priv->gtt.stolen_size, },
+			{ .start = base, .end = base + dev_priv->gtt.stolen_size, },
+		};
+		u64 gtt_start, gtt_end;
+
+		gtt_start = I915_READ(PGTBL_CTL);
+		if (IS_GEN4(dev))
+			gtt_start = (gtt_start & PGTBL_ADDRESS_LO_MASK) |
+				(gtt_start & PGTBL_ADDRESS_HI_MASK) << 28;
+		else
+			gtt_start &= PGTBL_ADDRESS_LO_MASK;
+		gtt_end = gtt_start + gtt_total_entries(dev_priv->gtt) * 4;
+
+		if (gtt_start >= stolen[0].start && gtt_start < stolen[0].end)
+			stolen[0].end = gtt_start;
+		if (gtt_end > stolen[1].start && gtt_end <= stolen[1].end)
+			stolen[1].start = gtt_end;
+
+		/* pick the larger of the two chunks */
+		if (stolen[0].end - stolen[0].start >
+		    stolen[1].end - stolen[1].start) {
+			base = stolen[0].start;
+			dev_priv->gtt.stolen_size = stolen[0].end - stolen[0].start;
+		} else {
+			base = stolen[1].start;
+			dev_priv->gtt.stolen_size = stolen[1].end - stolen[1].start;
+		}
+
+		if (stolen[0].start != stolen[1].start ||
+		    stolen[0].end != stolen[1].end) {
+			DRM_DEBUG_KMS("GTT within stolen memory at 0x%llx-0x%llx\n",
+				      (unsigned long long) gtt_start,
+				      (unsigned long long) gtt_end - 1);
+			DRM_DEBUG_KMS("Stolen memory adjusted to 0x%x-0x%x\n",
+				      base, base + (u32) dev_priv->gtt.stolen_size - 1);
+		}
+	}
+
+
 	/* Verify that nothing else uses this physical address. Stolen
 	 * memory should be reserved by the BIOS and hidden from the
 	 * kernel. So if the region is already marked as busy, something
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
new file mode 100644
index 000000000000..21ea92886a56
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -0,0 +1,711 @@
+/*
+ * Copyright © 2012-2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#include "drmP.h"
+#include "i915_drm.h"
+#include "i915_drv.h"
+#include "i915_trace.h"
+#include "intel_drv.h"
+#include <linux/mmu_context.h>
+#include <linux/mmu_notifier.h>
+#include <linux/mempolicy.h>
+#include <linux/swap.h>
+
+#if defined(CONFIG_MMU_NOTIFIER)
+#include <linux/interval_tree.h>
+
+struct i915_mmu_notifier {
+	spinlock_t lock;
+	struct hlist_node node;
+	struct mmu_notifier mn;
+	struct rb_root objects;
+	struct drm_device *dev;
+	struct mm_struct *mm;
+	struct work_struct work;
+	unsigned long count;
+	unsigned long serial;
+};
+
+struct i915_mmu_object {
+	struct i915_mmu_notifier *mmu;
+	struct interval_tree_node it;
+	struct drm_i915_gem_object *obj;
+};
+
+static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
+						       struct mm_struct *mm,
+						       unsigned long start,
+						       unsigned long end)
+{
+	struct i915_mmu_notifier *mn = container_of(_mn, struct i915_mmu_notifier, mn);
+	struct interval_tree_node *it = NULL;
+	unsigned long serial = 0;
+
+	end--; /* interval ranges are inclusive, but invalidate range is exclusive */
+	while (start < end) {
+		struct drm_i915_gem_object *obj;
+
+		obj = NULL;
+		spin_lock(&mn->lock);
+		if (serial == mn->serial)
+			it = interval_tree_iter_next(it, start, end);
+		else
+			it = interval_tree_iter_first(&mn->objects, start, end);
+		if (it != NULL) {
+			obj = container_of(it, struct i915_mmu_object, it)->obj;
+			drm_gem_object_reference(&obj->base);
+			serial = mn->serial;
+		}
+		spin_unlock(&mn->lock);
+		if (obj == NULL)
+			return;
+
+		mutex_lock(&mn->dev->struct_mutex);
+		/* Cancel any active worker and force us to re-evaluate gup */
+		obj->userptr.work = NULL;
+
+		if (obj->pages != NULL) {
+			struct drm_i915_private *dev_priv = to_i915(mn->dev);
+			struct i915_vma *vma, *tmp;
+			bool was_interruptible;
+
+			was_interruptible = dev_priv->mm.interruptible;
+			dev_priv->mm.interruptible = false;
+
+			list_for_each_entry_safe(vma, tmp, &obj->vma_list, vma_link) {
+				int ret = i915_vma_unbind(vma);
+				WARN_ON(ret && ret != -EIO);
+			}
+			WARN_ON(i915_gem_object_put_pages(obj));
+
+			dev_priv->mm.interruptible = was_interruptible;
+		}
+
+		start = obj->userptr.ptr + obj->base.size;
+
+		drm_gem_object_unreference(&obj->base);
+		mutex_unlock(&mn->dev->struct_mutex);
+	}
+}
+
+static const struct mmu_notifier_ops i915_gem_userptr_notifier = {
+	.invalidate_range_start = i915_gem_userptr_mn_invalidate_range_start,
+};
+
+static struct i915_mmu_notifier *
+__i915_mmu_notifier_lookup(struct drm_device *dev, struct mm_struct *mm)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_mmu_notifier *mmu;
+
+	/* Protected by dev->struct_mutex */
+	hash_for_each_possible(dev_priv->mmu_notifiers, mmu, node, (unsigned long)mm)
+		if (mmu->mm == mm)
+			return mmu;
+
+	return NULL;
+}
+
+static struct i915_mmu_notifier *
+i915_mmu_notifier_get(struct drm_device *dev, struct mm_struct *mm)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_mmu_notifier *mmu;
+	int ret;
+
+	lockdep_assert_held(&dev->struct_mutex);
+
+	mmu = __i915_mmu_notifier_lookup(dev, mm);
+	if (mmu)
+		return mmu;
+
+	mmu = kmalloc(sizeof(*mmu), GFP_KERNEL);
+	if (mmu == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	spin_lock_init(&mmu->lock);
+	mmu->dev = dev;
+	mmu->mn.ops = &i915_gem_userptr_notifier;
+	mmu->mm = mm;
+	mmu->objects = RB_ROOT;
+	mmu->count = 0;
+	mmu->serial = 0;
+
+	/* Protected by mmap_sem (write-lock) */
+	ret = __mmu_notifier_register(&mmu->mn, mm);
+	if (ret) {
+		kfree(mmu);
+		return ERR_PTR(ret);
+	}
+
+	/* Protected by dev->struct_mutex */
+	hash_add(dev_priv->mmu_notifiers, &mmu->node, (unsigned long)mm);
+	return mmu;
+}
+
+static void
+__i915_mmu_notifier_destroy_worker(struct work_struct *work)
+{
+	struct i915_mmu_notifier *mmu = container_of(work, typeof(*mmu), work);
+	mmu_notifier_unregister(&mmu->mn, mmu->mm);
+	kfree(mmu);
+}
+
+static void
+__i915_mmu_notifier_destroy(struct i915_mmu_notifier *mmu)
+{
+	lockdep_assert_held(&mmu->dev->struct_mutex);
+
+	/* Protected by dev->struct_mutex */
+	hash_del(&mmu->node);
+
+	/* Our lock ordering is: mmap_sem, mmu_notifier_scru, struct_mutex.
+	 * We enter the function holding struct_mutex, therefore we need
+	 * to drop our mutex prior to calling mmu_notifier_unregister in
+	 * order to prevent lock inversion (and system-wide deadlock)
+	 * between the mmap_sem and struct-mutex. Hence we defer the
+	 * unregistration to a workqueue where we hold no locks.
+	 */
+	INIT_WORK(&mmu->work, __i915_mmu_notifier_destroy_worker);
+	schedule_work(&mmu->work);
+}
+
+static void __i915_mmu_notifier_update_serial(struct i915_mmu_notifier *mmu)
+{
+	if (++mmu->serial == 0)
+		mmu->serial = 1;
+}
+
+static void
+i915_mmu_notifier_del(struct i915_mmu_notifier *mmu,
+		      struct i915_mmu_object *mn)
+{
+	lockdep_assert_held(&mmu->dev->struct_mutex);
+
+	spin_lock(&mmu->lock);
+	interval_tree_remove(&mn->it, &mmu->objects);
+	__i915_mmu_notifier_update_serial(mmu);
+	spin_unlock(&mmu->lock);
+
+	/* Protected against _add() by dev->struct_mutex */
+	if (--mmu->count == 0)
+		__i915_mmu_notifier_destroy(mmu);
+}
+
+static int
+i915_mmu_notifier_add(struct i915_mmu_notifier *mmu,
+		      struct i915_mmu_object *mn)
+{
+	struct interval_tree_node *it;
+	int ret;
+
+	ret = i915_mutex_lock_interruptible(mmu->dev);
+	if (ret)
+		return ret;
+
+	/* Make sure we drop the final active reference (and thereby
+	 * remove the objects from the interval tree) before we do
+	 * the check for overlapping objects.
+	 */
+	i915_gem_retire_requests(mmu->dev);
+
+	/* Disallow overlapping userptr objects */
+	spin_lock(&mmu->lock);
+	it = interval_tree_iter_first(&mmu->objects,
+				      mn->it.start, mn->it.last);
+	if (it) {
+		struct drm_i915_gem_object *obj;
+
+		/* We only need to check the first object in the range as it
+		 * either has cancelled gup work queued and we need to
+		 * return back to the user to give time for the gup-workers
+		 * to flush their object references upon which the object will
+		 * be removed from the interval-tree, or the the range is
+		 * still in use by another client and the overlap is invalid.
+		 */
+
+		obj = container_of(it, struct i915_mmu_object, it)->obj;
+		ret = obj->userptr.workers ? -EAGAIN : -EINVAL;
+	} else {
+		interval_tree_insert(&mn->it, &mmu->objects);
+		__i915_mmu_notifier_update_serial(mmu);
+		ret = 0;
+	}
+	spin_unlock(&mmu->lock);
+	mutex_unlock(&mmu->dev->struct_mutex);
+
+	return ret;
+}
+
+static void
+i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
+{
+	struct i915_mmu_object *mn;
+
+	mn = obj->userptr.mn;
+	if (mn == NULL)
+		return;
+
+	i915_mmu_notifier_del(mn->mmu, mn);
+	obj->userptr.mn = NULL;
+}
+
+static int
+i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
+				    unsigned flags)
+{
+	struct i915_mmu_notifier *mmu;
+	struct i915_mmu_object *mn;
+	int ret;
+
+	if (flags & I915_USERPTR_UNSYNCHRONIZED)
+		return capable(CAP_SYS_ADMIN) ? 0 : -EPERM;
+
+	down_write(&obj->userptr.mm->mmap_sem);
+	ret = i915_mutex_lock_interruptible(obj->base.dev);
+	if (ret == 0) {
+		mmu = i915_mmu_notifier_get(obj->base.dev, obj->userptr.mm);
+		if (!IS_ERR(mmu))
+			mmu->count++; /* preemptive add to act as a refcount */
+		else
+			ret = PTR_ERR(mmu);
+		mutex_unlock(&obj->base.dev->struct_mutex);
+	}
+	up_write(&obj->userptr.mm->mmap_sem);
+	if (ret)
+		return ret;
+
+	mn = kzalloc(sizeof(*mn), GFP_KERNEL);
+	if (mn == NULL) {
+		ret = -ENOMEM;
+		goto destroy_mmu;
+	}
+
+	mn->mmu = mmu;
+	mn->it.start = obj->userptr.ptr;
+	mn->it.last = mn->it.start + obj->base.size - 1;
+	mn->obj = obj;
+
+	ret = i915_mmu_notifier_add(mmu, mn);
+	if (ret)
+		goto free_mn;
+
+	obj->userptr.mn = mn;
+	return 0;
+
+free_mn:
+	kfree(mn);
+destroy_mmu:
+	mutex_lock(&obj->base.dev->struct_mutex);
+	if (--mmu->count == 0)
+		__i915_mmu_notifier_destroy(mmu);
+	mutex_unlock(&obj->base.dev->struct_mutex);
+	return ret;
+}
+
+#else
+
+static void
+i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
+{
+}
+
+static int
+i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
+				    unsigned flags)
+{
+	if ((flags & I915_USERPTR_UNSYNCHRONIZED) == 0)
+		return -ENODEV;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	return 0;
+}
+#endif
+
+struct get_pages_work {
+	struct work_struct work;
+	struct drm_i915_gem_object *obj;
+	struct task_struct *task;
+};
+
+
+#if IS_ENABLED(CONFIG_SWIOTLB)
+#define swiotlb_active() swiotlb_nr_tbl()
+#else
+#define swiotlb_active() 0
+#endif
+
+static int
+st_set_pages(struct sg_table **st, struct page **pvec, int num_pages)
+{
+	struct scatterlist *sg;
+	int ret, n;
+
+	*st = kmalloc(sizeof(**st), GFP_KERNEL);
+	if (*st == NULL)
+		return -ENOMEM;
+
+	if (swiotlb_active()) {
+		ret = sg_alloc_table(*st, num_pages, GFP_KERNEL);
+		if (ret)
+			goto err;
+
+		for_each_sg((*st)->sgl, sg, num_pages, n)
+			sg_set_page(sg, pvec[n], PAGE_SIZE, 0);
+	} else {
+		ret = sg_alloc_table_from_pages(*st, pvec, num_pages,
+						0, num_pages << PAGE_SHIFT,
+						GFP_KERNEL);
+		if (ret)
+			goto err;
+	}
+
+	return 0;
+
+err:
+	kfree(*st);
+	*st = NULL;
+	return ret;
+}
+
+static void
+__i915_gem_userptr_get_pages_worker(struct work_struct *_work)
+{
+	struct get_pages_work *work = container_of(_work, typeof(*work), work);
+	struct drm_i915_gem_object *obj = work->obj;
+	struct drm_device *dev = obj->base.dev;
+	const int num_pages = obj->base.size >> PAGE_SHIFT;
+	struct page **pvec;
+	int pinned, ret;
+
+	ret = -ENOMEM;
+	pinned = 0;
+
+	pvec = kmalloc(num_pages*sizeof(struct page *),
+		       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
+	if (pvec == NULL)
+		pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
+	if (pvec != NULL) {
+		struct mm_struct *mm = obj->userptr.mm;
+
+		down_read(&mm->mmap_sem);
+		while (pinned < num_pages) {
+			ret = get_user_pages(work->task, mm,
+					     obj->userptr.ptr + pinned * PAGE_SIZE,
+					     num_pages - pinned,
+					     !obj->userptr.read_only, 0,
+					     pvec + pinned, NULL);
+			if (ret < 0)
+				break;
+
+			pinned += ret;
+		}
+		up_read(&mm->mmap_sem);
+	}
+
+	mutex_lock(&dev->struct_mutex);
+	if (obj->userptr.work != &work->work) {
+		ret = 0;
+	} else if (pinned == num_pages) {
+		ret = st_set_pages(&obj->pages, pvec, num_pages);
+		if (ret == 0) {
+			list_add_tail(&obj->global_list, &to_i915(dev)->mm.unbound_list);
+			pinned = 0;
+		}
+	}
+
+	obj->userptr.work = ERR_PTR(ret);
+	obj->userptr.workers--;
+	drm_gem_object_unreference(&obj->base);
+	mutex_unlock(&dev->struct_mutex);
+
+	release_pages(pvec, pinned, 0);
+	drm_free_large(pvec);
+
+	put_task_struct(work->task);
+	kfree(work);
+}
+
+static int
+i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
+{
+	const int num_pages = obj->base.size >> PAGE_SHIFT;
+	struct page **pvec;
+	int pinned, ret;
+
+	/* If userspace should engineer that these pages are replaced in
+	 * the vma between us binding this page into the GTT and completion
+	 * of rendering... Their loss. If they change the mapping of their
+	 * pages they need to create a new bo to point to the new vma.
+	 *
+	 * However, that still leaves open the possibility of the vma
+	 * being copied upon fork. Which falls under the same userspace
+	 * synchronisation issue as a regular bo, except that this time
+	 * the process may not be expecting that a particular piece of
+	 * memory is tied to the GPU.
+	 *
+	 * Fortunately, we can hook into the mmu_notifier in order to
+	 * discard the page references prior to anything nasty happening
+	 * to the vma (discard or cloning) which should prevent the more
+	 * egregious cases from causing harm.
+	 */
+
+	pvec = NULL;
+	pinned = 0;
+	if (obj->userptr.mm == current->mm) {
+		pvec = kmalloc(num_pages*sizeof(struct page *),
+			       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
+		if (pvec == NULL) {
+			pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
+			if (pvec == NULL)
+				return -ENOMEM;
+		}
+
+		pinned = __get_user_pages_fast(obj->userptr.ptr, num_pages,
+					       !obj->userptr.read_only, pvec);
+	}
+	if (pinned < num_pages) {
+		if (pinned < 0) {
+			ret = pinned;
+			pinned = 0;
+		} else {
+			/* Spawn a worker so that we can acquire the
+			 * user pages without holding our mutex. Access
+			 * to the user pages requires mmap_sem, and we have
+			 * a strict lock ordering of mmap_sem, struct_mutex -
+			 * we already hold struct_mutex here and so cannot
+			 * call gup without encountering a lock inversion.
+			 *
+			 * Userspace will keep on repeating the operation
+			 * (thanks to EAGAIN) until either we hit the fast
+			 * path or the worker completes. If the worker is
+			 * cancelled or superseded, the task is still run
+			 * but the results ignored. (This leads to
+			 * complications that we may have a stray object
+			 * refcount that we need to be wary of when
+			 * checking for existing objects during creation.)
+			 * If the worker encounters an error, it reports
+			 * that error back to this function through
+			 * obj->userptr.work = ERR_PTR.
+			 */
+			ret = -EAGAIN;
+			if (obj->userptr.work == NULL &&
+			    obj->userptr.workers < I915_GEM_USERPTR_MAX_WORKERS) {
+				struct get_pages_work *work;
+
+				work = kmalloc(sizeof(*work), GFP_KERNEL);
+				if (work != NULL) {
+					obj->userptr.work = &work->work;
+					obj->userptr.workers++;
+
+					work->obj = obj;
+					drm_gem_object_reference(&obj->base);
+
+					work->task = current;
+					get_task_struct(work->task);
+
+					INIT_WORK(&work->work, __i915_gem_userptr_get_pages_worker);
+					schedule_work(&work->work);
+				} else
+					ret = -ENOMEM;
+			} else {
+				if (IS_ERR(obj->userptr.work)) {
+					ret = PTR_ERR(obj->userptr.work);
+					obj->userptr.work = NULL;
+				}
+			}
+		}
+	} else {
+		ret = st_set_pages(&obj->pages, pvec, num_pages);
+		if (ret == 0) {
+			obj->userptr.work = NULL;
+			pinned = 0;
+		}
+	}
+
+	release_pages(pvec, pinned, 0);
+	drm_free_large(pvec);
+	return ret;
+}
+
+static void
+i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj)
+{
+	struct scatterlist *sg;
+	int i;
+
+	BUG_ON(obj->userptr.work != NULL);
+
+	if (obj->madv != I915_MADV_WILLNEED)
+		obj->dirty = 0;
+
+	for_each_sg(obj->pages->sgl, sg, obj->pages->nents, i) {
+		struct page *page = sg_page(sg);
+
+		if (obj->dirty)
+			set_page_dirty(page);
+
+		mark_page_accessed(page);
+		page_cache_release(page);
+	}
+	obj->dirty = 0;
+
+	sg_free_table(obj->pages);
+	kfree(obj->pages);
+}
+
+static void
+i915_gem_userptr_release(struct drm_i915_gem_object *obj)
+{
+	i915_gem_userptr_release__mmu_notifier(obj);
+
+	if (obj->userptr.mm) {
+		mmput(obj->userptr.mm);
+		obj->userptr.mm = NULL;
+	}
+}
+
+static int
+i915_gem_userptr_dmabuf_export(struct drm_i915_gem_object *obj)
+{
+	if (obj->userptr.mn)
+		return 0;
+
+	return i915_gem_userptr_init__mmu_notifier(obj, 0);
+}
+
+static const struct drm_i915_gem_object_ops i915_gem_userptr_ops = {
+	.dmabuf_export = i915_gem_userptr_dmabuf_export,
+	.get_pages = i915_gem_userptr_get_pages,
+	.put_pages = i915_gem_userptr_put_pages,
+	.release = i915_gem_userptr_release,
+};
+
+/**
+ * Creates a new mm object that wraps some normal memory from the process
+ * context - user memory.
+ *
+ * We impose several restrictions upon the memory being mapped
+ * into the GPU.
+ * 1. It must be page aligned (both start/end addresses, i.e ptr and size).
+ * 2. It cannot overlap any other userptr object in the same address space.
+ * 3. It must be normal system memory, not a pointer into another map of IO
+ *    space (e.g. it must not be a GTT mmapping of another object).
+ * 4. We only allow a bo as large as we could in theory map into the GTT,
+ *    that is we limit the size to the total size of the GTT.
+ * 5. The bo is marked as being snoopable. The backing pages are left
+ *    accessible directly by the CPU, but reads and writes by the GPU may
+ *    incur the cost of a snoop (unless you have an LLC architecture).
+ *
+ * Synchronisation between multiple users and the GPU is left to userspace
+ * through the normal set-domain-ioctl. The kernel will enforce that the
+ * GPU relinquishes the VMA before it is returned back to the system
+ * i.e. upon free(), munmap() or process termination. However, the userspace
+ * malloc() library may not immediately relinquish the VMA after free() and
+ * instead reuse it whilst the GPU is still reading and writing to the VMA.
+ * Caveat emptor.
+ *
+ * Also note, that the object created here is not currently a "first class"
+ * object, in that several ioctls are banned. These are the CPU access
+ * ioctls: mmap(), pwrite and pread. In practice, you are expected to use
+ * direct access via your pointer rather than use those ioctls.
+ *
+ * If you think this is a good interface to use to pass GPU memory between
+ * drivers, please use dma-buf instead. In fact, wherever possible use
+ * dma-buf instead.
+ */
+int
+i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_userptr *args = data;
+	struct drm_i915_gem_object *obj;
+	int ret;
+	u32 handle;
+
+	if (args->flags & ~(I915_USERPTR_READ_ONLY |
+			    I915_USERPTR_UNSYNCHRONIZED))
+		return -EINVAL;
+
+	if (offset_in_page(args->user_ptr | args->user_size))
+		return -EINVAL;
+
+	if (args->user_size > dev_priv->gtt.base.total)
+		return -E2BIG;
+
+	if (!access_ok(args->flags & I915_USERPTR_READ_ONLY ? VERIFY_READ : VERIFY_WRITE,
+		       (char __user *)(unsigned long)args->user_ptr, args->user_size))
+		return -EFAULT;
+
+	if (args->flags & I915_USERPTR_READ_ONLY) {
+		/* On almost all of the current hw, we cannot tell the GPU that a
+		 * page is readonly, so this is just a placeholder in the uAPI.
+		 */
+		return -ENODEV;
+	}
+
+	/* Allocate the new object */
+	obj = i915_gem_object_alloc(dev);
+	if (obj == NULL)
+		return -ENOMEM;
+
+	drm_gem_private_object_init(dev, &obj->base, args->user_size);
+	i915_gem_object_init(obj, &i915_gem_userptr_ops);
+	obj->cache_level = I915_CACHE_LLC;
+	obj->base.write_domain = I915_GEM_DOMAIN_CPU;
+	obj->base.read_domains = I915_GEM_DOMAIN_CPU;
+
+	obj->userptr.ptr = args->user_ptr;
+	obj->userptr.read_only = !!(args->flags & I915_USERPTR_READ_ONLY);
+
+	/* And keep a pointer to the current->mm for resolving the user pages
+	 * at binding. This means that we need to hook into the mmu_notifier
+	 * in order to detect if the mmu is destroyed.
+	 */
+	ret = -ENOMEM;
+	if ((obj->userptr.mm = get_task_mm(current)))
+		ret = i915_gem_userptr_init__mmu_notifier(obj, args->flags);
+	if (ret == 0)
+		ret = drm_gem_handle_create(file, &obj->base, &handle);
+
+	/* drop reference from allocate - handle holds it now */
+	drm_gem_object_unreference_unlocked(&obj->base);
+	if (ret)
+		return ret;
+
+	args->handle = handle;
+	return 0;
+}
+
+int
+i915_gem_init_userptr(struct drm_device *dev)
+{
+#if defined(CONFIG_MMU_NOTIFIER)
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	hash_init(dev_priv->mmu_notifiers);
+#endif
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 12f1d43b2d68..66cf41765bf9 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -42,6 +42,7 @@ static const char *ring_str(int ring)
 	case VCS: return "bsd";
 	case BCS: return "blt";
 	case VECS: return "vebox";
+	case VCS2: return "bsd2";
 	default: return "";
 	}
 }
@@ -204,6 +205,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 		err_puts(m, tiling_flag(err->tiling));
 		err_puts(m, dirty_flag(err->dirty));
 		err_puts(m, purgeable_flag(err->purgeable));
+		err_puts(m, err->userptr ? " userptr" : "");
 		err_puts(m, err->ring != -1 ? " " : "");
 		err_puts(m, ring_str(err->ring));
 		err_puts(m, i915_cache_level_str(err->cache_level));
@@ -257,7 +259,8 @@ static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 		err_printf(m, "  INSTPS: 0x%08x\n", ring->instps);
 	}
 	err_printf(m, "  INSTPM: 0x%08x\n", ring->instpm);
-	err_printf(m, "  FADDR: 0x%08x\n", ring->faddr);
+	err_printf(m, "  FADDR: 0x%08x %08x\n", upper_32_bits(ring->faddr),
+		   lower_32_bits(ring->faddr));
 	if (INTEL_INFO(dev)->gen >= 6) {
 		err_printf(m, "  RC PSMI: 0x%08x\n", ring->rc_psmi);
 		err_printf(m, "  FAULT_REG: 0x%08x\n", ring->fault_reg);
@@ -452,16 +455,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			err_printf(m, "%s --- HW Context = 0x%08x\n",
 				   dev_priv->ring[i].name,
 				   obj->gtt_offset);
-			offset = 0;
-			for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
-				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
-					   offset,
-					   obj->pages[0][elt],
-					   obj->pages[0][elt+1],
-					   obj->pages[0][elt+2],
-					   obj->pages[0][elt+3]);
-					offset += 16;
-			}
+			print_error_obj(m, obj);
 		}
 	}
 
@@ -648,6 +642,7 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 	err->tiling = obj->tiling_mode;
 	err->dirty = obj->dirty;
 	err->purgeable = obj->madv != I915_MADV_WILLNEED;
+	err->userptr = obj->userptr.mm != NULL;
 	err->ring = obj->ring ? obj->ring->id : -1;
 	err->cache_level = obj->cache_level;
 }
@@ -752,7 +747,7 @@ static void i915_gem_record_fences(struct drm_device *dev,
 }
 
 static void i915_record_ring_state(struct drm_device *dev,
-				   struct intel_ring_buffer *ring,
+				   struct intel_engine_cs *ring,
 				   struct drm_i915_error_ring *ering)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -764,14 +759,14 @@ static void i915_record_ring_state(struct drm_device *dev,
 			= I915_READ(RING_SYNC_0(ring->mmio_base));
 		ering->semaphore_mboxes[1]
 			= I915_READ(RING_SYNC_1(ring->mmio_base));
-		ering->semaphore_seqno[0] = ring->sync_seqno[0];
-		ering->semaphore_seqno[1] = ring->sync_seqno[1];
+		ering->semaphore_seqno[0] = ring->semaphore.sync_seqno[0];
+		ering->semaphore_seqno[1] = ring->semaphore.sync_seqno[1];
 	}
 
 	if (HAS_VEBOX(dev)) {
 		ering->semaphore_mboxes[2] =
 			I915_READ(RING_SYNC_2(ring->mmio_base));
-		ering->semaphore_seqno[2] = ring->sync_seqno[2];
+		ering->semaphore_seqno[2] = ring->semaphore.sync_seqno[2];
 	}
 
 	if (INTEL_INFO(dev)->gen >= 4) {
@@ -781,8 +776,10 @@ static void i915_record_ring_state(struct drm_device *dev,
 		ering->instdone = I915_READ(RING_INSTDONE(ring->mmio_base));
 		ering->instps = I915_READ(RING_INSTPS(ring->mmio_base));
 		ering->bbaddr = I915_READ(RING_BBADDR(ring->mmio_base));
-		if (INTEL_INFO(dev)->gen >= 8)
+		if (INTEL_INFO(dev)->gen >= 8) {
+			ering->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(ring->mmio_base)) << 32;
 			ering->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(ring->mmio_base)) << 32;
+		}
 		ering->bbstate = I915_READ(RING_BBSTATE(ring->mmio_base));
 	} else {
 		ering->faddr = I915_READ(DMA_FADD_I8XX);
@@ -828,8 +825,8 @@ static void i915_record_ring_state(struct drm_device *dev,
 		ering->hws = I915_READ(mmio);
 	}
 
-	ering->cpu_ring_head = ring->head;
-	ering->cpu_ring_tail = ring->tail;
+	ering->cpu_ring_head = ring->buffer->head;
+	ering->cpu_ring_tail = ring->buffer->tail;
 
 	ering->hangcheck_score = ring->hangcheck.score;
 	ering->hangcheck_action = ring->hangcheck.action;
@@ -862,7 +859,7 @@ static void i915_record_ring_state(struct drm_device *dev,
 }
 
 
-static void i915_gem_record_active_context(struct intel_ring_buffer *ring,
+static void i915_gem_record_active_context(struct intel_engine_cs *ring,
 					   struct drm_i915_error_state *error,
 					   struct drm_i915_error_ring *ering)
 {
@@ -875,10 +872,7 @@ static void i915_gem_record_active_context(struct intel_ring_buffer *ring,
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		if ((error->ccid & PAGE_MASK) == i915_gem_obj_ggtt_offset(obj)) {
-			ering->ctx = i915_error_object_create_sized(dev_priv,
-								    obj,
-								    &dev_priv->gtt.base,
-								    1);
+			ering->ctx = i915_error_ggtt_object_create(dev_priv, obj);
 			break;
 		}
 	}
@@ -892,7 +886,9 @@ static void i915_gem_record_rings(struct drm_device *dev,
 	int i, count;
 
 	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_ring_buffer *ring = &dev_priv->ring[i];
+		struct intel_engine_cs *ring = &dev_priv->ring[i];
+
+		error->ring[i].pid = -1;
 
 		if (ring->dev == NULL)
 			continue;
@@ -901,7 +897,6 @@ static void i915_gem_record_rings(struct drm_device *dev,
 
 		i915_record_ring_state(dev, ring, &error->ring[i]);
 
-		error->ring[i].pid = -1;
 		request = i915_gem_find_active_request(ring);
 		if (request) {
 			/* We need to copy these to an anonymous buffer
@@ -936,7 +931,7 @@ static void i915_gem_record_rings(struct drm_device *dev,
 		}
 
 		error->ring[i].ringbuffer =
-			i915_error_ggtt_object_create(dev_priv, ring->obj);
+			i915_error_ggtt_object_create(dev_priv, ring->buffer->obj);
 
 		if (ring->status_page.obj)
 			error->ring[i].hws_page =
@@ -1037,7 +1032,6 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 				   struct drm_i915_error_state *error)
 {
 	struct drm_device *dev = dev_priv->dev;
-	int pipe;
 
 	/* General organization
 	 * 1. Registers specific to a single generation
@@ -1062,9 +1056,6 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 		error->gfx_mode = I915_READ(GFX_MODE);
 	}
 
-	if (IS_GEN2(dev))
-		error->ier = I915_READ16(IER);
-
 	/* 2: Registers which belong to multiple generations */
 	if (INTEL_INFO(dev)->gen >= 7)
 		error->forcewake = I915_READ(FORCEWAKE_MT);
@@ -1088,9 +1079,10 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 	if (HAS_PCH_SPLIT(dev))
 		error->ier = I915_READ(DEIER) | I915_READ(GTIER);
 	else {
-		error->ier = I915_READ(IER);
-		for_each_pipe(pipe)
-			error->pipestat[pipe] = I915_READ(PIPESTAT(pipe));
+		if (IS_GEN2(dev))
+			error->ier = I915_READ16(IER);
+		else
+			error->ier = I915_READ(IER);
 	}
 
 	/* 4: Everything else */
diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c
index 3c59584161c2..2e0613e26251 100644
--- a/drivers/gpu/drm/i915/i915_ioc32.c
+++ b/drivers/gpu/drm/i915/i915_ioc32.c
@@ -208,7 +208,7 @@ long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(i915_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(i915_compat_ioctls))
 		fn = i915_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 0b99de95593b..267f069765ad 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -80,17 +80,64 @@ static const u32 hpd_status_i915[] = { /* i915 and valleyview are the same */
 	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS
 };
 
+/* IIR can theoretically queue up two events. Be paranoid. */
+#define GEN8_IRQ_RESET_NDX(type, which) do { \
+	I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
+	POSTING_READ(GEN8_##type##_IMR(which)); \
+	I915_WRITE(GEN8_##type##_IER(which), 0); \
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+	POSTING_READ(GEN8_##type##_IIR(which)); \
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+	POSTING_READ(GEN8_##type##_IIR(which)); \
+} while (0)
+
+#define GEN5_IRQ_RESET(type) do { \
+	I915_WRITE(type##IMR, 0xffffffff); \
+	POSTING_READ(type##IMR); \
+	I915_WRITE(type##IER, 0); \
+	I915_WRITE(type##IIR, 0xffffffff); \
+	POSTING_READ(type##IIR); \
+	I915_WRITE(type##IIR, 0xffffffff); \
+	POSTING_READ(type##IIR); \
+} while (0)
+
+/*
+ * We should clear IMR at preinstall/uninstall, and just check at postinstall.
+ */
+#define GEN5_ASSERT_IIR_IS_ZERO(reg) do { \
+	u32 val = I915_READ(reg); \
+	if (val) { \
+		WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n", \
+		     (reg), val); \
+		I915_WRITE((reg), 0xffffffff); \
+		POSTING_READ(reg); \
+		I915_WRITE((reg), 0xffffffff); \
+		POSTING_READ(reg); \
+	} \
+} while (0)
+
+#define GEN8_IRQ_INIT_NDX(type, which, imr_val, ier_val) do { \
+	GEN5_ASSERT_IIR_IS_ZERO(GEN8_##type##_IIR(which)); \
+	I915_WRITE(GEN8_##type##_IMR(which), (imr_val)); \
+	I915_WRITE(GEN8_##type##_IER(which), (ier_val)); \
+	POSTING_READ(GEN8_##type##_IER(which)); \
+} while (0)
+
+#define GEN5_IRQ_INIT(type, imr_val, ier_val) do { \
+	GEN5_ASSERT_IIR_IS_ZERO(type##IIR); \
+	I915_WRITE(type##IMR, (imr_val)); \
+	I915_WRITE(type##IER, (ier_val)); \
+	POSTING_READ(type##IER); \
+} while (0)
+
 /* For display hotplug interrupt */
 static void
 ironlake_enable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
 {
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.deimr &= ~mask;
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	if ((dev_priv->irq_mask & mask) != 0) {
 		dev_priv->irq_mask &= ~mask;
@@ -104,11 +151,8 @@ ironlake_disable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
 {
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.deimr |= mask;
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	if ((dev_priv->irq_mask & mask) != mask) {
 		dev_priv->irq_mask |= mask;
@@ -129,13 +173,8 @@ static void ilk_update_gt_irq(struct drm_i915_private *dev_priv,
 {
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.gtimr &= ~interrupt_mask;
-		dev_priv->pm.regsave.gtimr |= (~enabled_irq_mask &
-						interrupt_mask);
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	dev_priv->gt_irq_mask &= ~interrupt_mask;
 	dev_priv->gt_irq_mask |= (~enabled_irq_mask & interrupt_mask);
@@ -167,13 +206,8 @@ static void snb_update_pm_irq(struct drm_i915_private *dev_priv,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.gen6_pmimr &= ~interrupt_mask;
-		dev_priv->pm.regsave.gen6_pmimr |= (~enabled_irq_mask &
-						     interrupt_mask);
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	new_val = dev_priv->pm_irq_mask;
 	new_val &= ~interrupt_mask;
@@ -214,6 +248,46 @@ static bool ivb_can_enable_err_int(struct drm_device *dev)
 	return true;
 }
 
+/**
+  * bdw_update_pm_irq - update GT interrupt 2
+  * @dev_priv: driver private
+  * @interrupt_mask: mask of interrupt bits to update
+  * @enabled_irq_mask: mask of interrupt bits to enable
+  *
+  * Copied from the snb function, updated with relevant register offsets
+  */
+static void bdw_update_pm_irq(struct drm_i915_private *dev_priv,
+			      uint32_t interrupt_mask,
+			      uint32_t enabled_irq_mask)
+{
+	uint32_t new_val;
+
+	assert_spin_locked(&dev_priv->irq_lock);
+
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
+		return;
+
+	new_val = dev_priv->pm_irq_mask;
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != dev_priv->pm_irq_mask) {
+		dev_priv->pm_irq_mask = new_val;
+		I915_WRITE(GEN8_GT_IMR(2), dev_priv->pm_irq_mask);
+		POSTING_READ(GEN8_GT_IMR(2));
+	}
+}
+
+void bdw_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask)
+{
+	bdw_update_pm_irq(dev_priv, mask, mask);
+}
+
+void bdw_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask)
+{
+	bdw_update_pm_irq(dev_priv, mask, 0);
+}
+
 static bool cpt_can_enable_serr_int(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -232,16 +306,51 @@ static bool cpt_can_enable_serr_int(struct drm_device *dev)
 	return true;
 }
 
-static void i9xx_clear_fifo_underrun(struct drm_device *dev, enum pipe pipe)
+void i9xx_check_fifo_underruns(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *crtc;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+
+	for_each_intel_crtc(dev, crtc) {
+		u32 reg = PIPESTAT(crtc->pipe);
+		u32 pipestat;
+
+		if (crtc->cpu_fifo_underrun_disabled)
+			continue;
+
+		pipestat = I915_READ(reg) & 0xffff0000;
+		if ((pipestat & PIPE_FIFO_UNDERRUN_STATUS) == 0)
+			continue;
+
+		I915_WRITE(reg, pipestat | PIPE_FIFO_UNDERRUN_STATUS);
+		POSTING_READ(reg);
+
+		DRM_ERROR("pipe %c underrun\n", pipe_name(crtc->pipe));
+	}
+
+	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
+
+static void i9xx_set_fifo_underrun_reporting(struct drm_device *dev,
+					     enum pipe pipe,
+					     bool enable, bool old)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 reg = PIPESTAT(pipe);
-	u32 pipestat = I915_READ(reg) & 0x7fff0000;
+	u32 pipestat = I915_READ(reg) & 0xffff0000;
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	I915_WRITE(reg, pipestat | PIPE_FIFO_UNDERRUN_STATUS);
-	POSTING_READ(reg);
+	if (enable) {
+		I915_WRITE(reg, pipestat | PIPE_FIFO_UNDERRUN_STATUS);
+		POSTING_READ(reg);
+	} else {
+		if (old && pipestat & PIPE_FIFO_UNDERRUN_STATUS)
+			DRM_ERROR("pipe %c underrun\n", pipe_name(pipe));
+	}
 }
 
 static void ironlake_set_fifo_underrun_reporting(struct drm_device *dev,
@@ -258,7 +367,8 @@ static void ironlake_set_fifo_underrun_reporting(struct drm_device *dev,
 }
 
 static void ivybridge_set_fifo_underrun_reporting(struct drm_device *dev,
-						  enum pipe pipe, bool enable)
+						  enum pipe pipe,
+						  bool enable, bool old)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	if (enable) {
@@ -269,15 +379,12 @@ static void ivybridge_set_fifo_underrun_reporting(struct drm_device *dev,
 
 		ironlake_enable_display_irq(dev_priv, DE_ERR_INT_IVB);
 	} else {
-		bool was_enabled = !(I915_READ(DEIMR) & DE_ERR_INT_IVB);
-
-		/* Change the state _after_ we've read out the current one. */
 		ironlake_disable_display_irq(dev_priv, DE_ERR_INT_IVB);
 
-		if (!was_enabled &&
-		    (I915_READ(GEN7_ERR_INT) & ERR_INT_FIFO_UNDERRUN(pipe))) {
-			DRM_DEBUG_KMS("uncleared fifo underrun on pipe %c\n",
-				      pipe_name(pipe));
+		if (old &&
+		    I915_READ(GEN7_ERR_INT) & ERR_INT_FIFO_UNDERRUN(pipe)) {
+			DRM_ERROR("uncleared fifo underrun on pipe %c\n",
+				  pipe_name(pipe));
 		}
 	}
 }
@@ -313,14 +420,8 @@ static void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled &&
-	    (interrupt_mask & SDE_HOTPLUG_MASK_CPT)) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.sdeimr &= ~interrupt_mask;
-		dev_priv->pm.regsave.sdeimr |= (~enabled_irq_mask &
-						 interrupt_mask);
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	I915_WRITE(SDEIMR, sdeimr);
 	POSTING_READ(SDEIMR);
@@ -346,7 +447,7 @@ static void ibx_set_fifo_underrun_reporting(struct drm_device *dev,
 
 static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
 					    enum transcoder pch_transcoder,
-					    bool enable)
+					    bool enable, bool old)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -359,16 +460,12 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
 
 		ibx_enable_display_interrupt(dev_priv, SDE_ERROR_CPT);
 	} else {
-		uint32_t tmp = I915_READ(SERR_INT);
-		bool was_enabled = !(I915_READ(SDEIMR) & SDE_ERROR_CPT);
-
-		/* Change the state _after_ we've read out the current one. */
 		ibx_disable_display_interrupt(dev_priv, SDE_ERROR_CPT);
 
-		if (!was_enabled &&
-		    (tmp & SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder))) {
-			DRM_DEBUG_KMS("uncleared pch fifo underrun on pch transcoder %c\n",
-				      transcoder_name(pch_transcoder));
+		if (old && I915_READ(SERR_INT) &
+		    SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder)) {
+			DRM_ERROR("uncleared pch fifo underrun on pch transcoder %c\n",
+				  transcoder_name(pch_transcoder));
 		}
 	}
 }
@@ -387,34 +484,29 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
  *
  * Returns the previous state of underrun reporting.
  */
-bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
-					     enum pipe pipe, bool enable)
+static bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
+						    enum pipe pipe, bool enable)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	bool ret;
+	bool old;
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	ret = !intel_crtc->cpu_fifo_underrun_disabled;
-
-	if (enable == ret)
-		goto done;
-
+	old = !intel_crtc->cpu_fifo_underrun_disabled;
 	intel_crtc->cpu_fifo_underrun_disabled = !enable;
 
-	if (enable && (INTEL_INFO(dev)->gen < 5 || IS_VALLEYVIEW(dev)))
-		i9xx_clear_fifo_underrun(dev, pipe);
+	if (INTEL_INFO(dev)->gen < 5 || IS_VALLEYVIEW(dev))
+		i9xx_set_fifo_underrun_reporting(dev, pipe, enable, old);
 	else if (IS_GEN5(dev) || IS_GEN6(dev))
 		ironlake_set_fifo_underrun_reporting(dev, pipe, enable);
 	else if (IS_GEN7(dev))
-		ivybridge_set_fifo_underrun_reporting(dev, pipe, enable);
+		ivybridge_set_fifo_underrun_reporting(dev, pipe, enable, old);
 	else if (IS_GEN8(dev))
 		broadwell_set_fifo_underrun_reporting(dev, pipe, enable);
 
-done:
-	return ret;
+	return old;
 }
 
 bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
@@ -463,7 +555,7 @@ bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pch_transcoder];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	unsigned long flags;
-	bool ret;
+	bool old;
 
 	/*
 	 * NOTE: Pre-LPT has a fixed cpu pipe -> pch transcoder mapping, but LPT
@@ -476,21 +568,16 @@ bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 
-	ret = !intel_crtc->pch_fifo_underrun_disabled;
-
-	if (enable == ret)
-		goto done;
-
+	old = !intel_crtc->pch_fifo_underrun_disabled;
 	intel_crtc->pch_fifo_underrun_disabled = !enable;
 
 	if (HAS_PCH_IBX(dev))
 		ibx_set_fifo_underrun_reporting(dev, pch_transcoder, enable);
 	else
-		cpt_set_fifo_underrun_reporting(dev, pch_transcoder, enable);
+		cpt_set_fifo_underrun_reporting(dev, pch_transcoder, enable, old);
 
-done:
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
-	return ret;
+	return old;
 }
 
 
@@ -503,8 +590,10 @@ __i915_enable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (WARN_ON_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
-	                 status_mask & ~PIPESTAT_INT_STATUS_MASK))
+	if (WARN_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
+		      status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: enable_mask=0x%x, status_mask=0x%x\n",
+		      pipe_name(pipe), enable_mask, status_mask))
 		return;
 
 	if ((pipestat & enable_mask) == enable_mask)
@@ -527,8 +616,10 @@ __i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (WARN_ON_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
-	                 status_mask & ~PIPESTAT_INT_STATUS_MASK))
+	if (WARN_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
+		      status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: enable_mask=0x%x, status_mask=0x%x\n",
+		      pipe_name(pipe), enable_mask, status_mask))
 		return;
 
 	if ((pipestat & enable_mask) == 0)
@@ -546,11 +637,17 @@ static u32 vlv_get_pipestat_enable_mask(struct drm_device *dev, u32 status_mask)
 	u32 enable_mask = status_mask << 16;
 
 	/*
-	 * On pipe A we don't support the PSR interrupt yet, on pipe B the
-	 * same bit MBZ.
+	 * On pipe A we don't support the PSR interrupt yet,
+	 * on pipe B and C the same bit MBZ.
 	 */
 	if (WARN_ON_ONCE(status_mask & PIPE_A_PSR_STATUS_VLV))
 		return 0;
+	/*
+	 * On pipe B and C we don't support the PSR interrupt yet, on pipe
+	 * A the same bit is for perf counters which we don't use either.
+	 */
+	if (WARN_ON_ONCE(status_mask & PIPE_B_PSR_STATUS_VLV))
+		return 0;
 
 	enable_mask &= ~(PIPE_FIFO_UNDERRUN_STATUS |
 			 SPRITE0_FLIP_DONE_INT_EN_VLV |
@@ -637,6 +734,56 @@ i915_pipe_enabled(struct drm_device *dev, int pipe)
 	}
 }
 
+/*
+ * This timing diagram depicts the video signal in and
+ * around the vertical blanking period.
+ *
+ * Assumptions about the fictitious mode used in this example:
+ *  vblank_start >= 3
+ *  vsync_start = vblank_start + 1
+ *  vsync_end = vblank_start + 2
+ *  vtotal = vblank_start + 3
+ *
+ *           start of vblank:
+ *           latch double buffered registers
+ *           increment frame counter (ctg+)
+ *           generate start of vblank interrupt (gen4+)
+ *           |
+ *           |          frame start:
+ *           |          generate frame start interrupt (aka. vblank interrupt) (gmch)
+ *           |          may be shifted forward 1-3 extra lines via PIPECONF
+ *           |          |
+ *           |          |  start of vsync:
+ *           |          |  generate vsync interrupt
+ *           |          |  |
+ * ___xxxx___    ___xxxx___    ___xxxx___    ___xxxx___    ___xxxx___    ___xxxx
+ *       .   \hs/   .      \hs/          \hs/          \hs/   .      \hs/
+ * ----va---> <-----------------vb--------------------> <--------va-------------
+ *       |          |       <----vs----->                     |
+ * -vbs-----> <---vbs+1---> <---vbs+2---> <-----0-----> <-----1-----> <-----2--- (scanline counter gen2)
+ * -vbs-2---> <---vbs-1---> <---vbs-----> <---vbs+1---> <---vbs+2---> <-----0--- (scanline counter gen3+)
+ * -vbs-2---> <---vbs-2---> <---vbs-1---> <---vbs-----> <---vbs+1---> <---vbs+2- (scanline counter hsw+ hdmi)
+ *       |          |                                         |
+ *       last visible pixel                                   first visible pixel
+ *                  |                                         increment frame counter (gen3/4)
+ *                  pixel counter = vblank_start * htotal     pixel counter = 0 (gen3/4)
+ *
+ * x  = horizontal active
+ * _  = horizontal blanking
+ * hs = horizontal sync
+ * va = vertical active
+ * vb = vertical blanking
+ * vs = vertical sync
+ * vbs = vblank_start (number)
+ *
+ * Summary:
+ * - most events happen at the start of horizontal sync
+ * - frame start happens at the start of horizontal blank, 1-4 lines
+ *   (depending on PIPECONF settings) after the start of vblank
+ * - gen3/4 pixel and frame counter are synchronized with the start
+ *   of horizontal active on the first line of vertical active
+ */
+
 static u32 i8xx_get_vblank_counter(struct drm_device *dev, int pipe)
 {
 	/* Gen2 doesn't have a hardware frame counter */
@@ -651,7 +798,7 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long high_frame;
 	unsigned long low_frame;
-	u32 high1, high2, low, pixel, vbl_start;
+	u32 high1, high2, low, pixel, vbl_start, hsync_start, htotal;
 
 	if (!i915_pipe_enabled(dev, pipe)) {
 		DRM_DEBUG_DRIVER("trying to get vblank count for disabled "
@@ -665,17 +812,28 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 		const struct drm_display_mode *mode =
 			&intel_crtc->config.adjusted_mode;
 
-		vbl_start = mode->crtc_vblank_start * mode->crtc_htotal;
+		htotal = mode->crtc_htotal;
+		hsync_start = mode->crtc_hsync_start;
+		vbl_start = mode->crtc_vblank_start;
+		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+			vbl_start = DIV_ROUND_UP(vbl_start, 2);
 	} else {
 		enum transcoder cpu_transcoder = (enum transcoder) pipe;
-		u32 htotal;
 
 		htotal = ((I915_READ(HTOTAL(cpu_transcoder)) >> 16) & 0x1fff) + 1;
+		hsync_start = (I915_READ(HSYNC(cpu_transcoder))  & 0x1fff) + 1;
 		vbl_start = (I915_READ(VBLANK(cpu_transcoder)) & 0x1fff) + 1;
-
-		vbl_start *= htotal;
+		if ((I915_READ(PIPECONF(cpu_transcoder)) &
+		     PIPECONF_INTERLACE_MASK) != PIPECONF_PROGRESSIVE)
+			vbl_start = DIV_ROUND_UP(vbl_start, 2);
 	}
 
+	/* Convert to pixel count */
+	vbl_start *= htotal;
+
+	/* Start of vblank event occurs at start of hsync */
+	vbl_start -= htotal - hsync_start;
+
 	high_frame = PIPEFRAME(pipe);
 	low_frame = PIPEFRAMEPIXEL(pipe);
 
@@ -719,24 +877,28 @@ static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
 /* raw reads, only for fast reads of display block, no need for forcewake etc. */
 #define __raw_i915_read32(dev_priv__, reg__) readl((dev_priv__)->regs + (reg__))
 
-static bool ilk_pipe_in_vblank_locked(struct drm_device *dev, enum pipe pipe)
+static int __intel_get_crtc_scanline(struct intel_crtc *crtc)
 {
+	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t status;
-	int reg;
+	const struct drm_display_mode *mode = &crtc->config.adjusted_mode;
+	enum pipe pipe = crtc->pipe;
+	int position, vtotal;
 
-	if (INTEL_INFO(dev)->gen >= 8) {
-		status = GEN8_PIPE_VBLANK;
-		reg = GEN8_DE_PIPE_ISR(pipe);
-	} else if (INTEL_INFO(dev)->gen >= 7) {
-		status = DE_PIPE_VBLANK_IVB(pipe);
-		reg = DEISR;
-	} else {
-		status = DE_PIPE_VBLANK(pipe);
-		reg = DEISR;
-	}
+	vtotal = mode->crtc_vtotal;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		vtotal /= 2;
 
-	return __raw_i915_read32(dev_priv, reg) & status;
+	if (IS_GEN2(dev))
+		position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN2;
+	else
+		position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3;
+
+	/*
+	 * See update_scanline_offset() for the details on the
+	 * scanline_offset adjustment.
+	 */
+	return (position + crtc->scanline_offset) % vtotal;
 }
 
 static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
@@ -748,7 +910,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	const struct drm_display_mode *mode = &intel_crtc->config.adjusted_mode;
 	int position;
-	int vbl_start, vbl_end, htotal, vtotal;
+	int vbl_start, vbl_end, hsync_start, htotal, vtotal;
 	bool in_vbl = true;
 	int ret = 0;
 	unsigned long irqflags;
@@ -760,6 +922,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	}
 
 	htotal = mode->crtc_htotal;
+	hsync_start = mode->crtc_hsync_start;
 	vtotal = mode->crtc_vtotal;
 	vbl_start = mode->crtc_vblank_start;
 	vbl_end = mode->crtc_vblank_end;
@@ -778,7 +941,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	 * following code must not block on uncore.lock.
 	 */
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-	
+
 	/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
 
 	/* Get optional system timestamp before query. */
@@ -789,68 +952,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 		/* No obvious pixelcount register. Only query vertical
 		 * scanout position from Display scan line register.
 		 */
-		if (IS_GEN2(dev))
-			position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN2;
-		else
-			position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3;
-
-		if (HAS_DDI(dev)) {
-			/*
-			 * On HSW HDMI outputs there seems to be a 2 line
-			 * difference, whereas eDP has the normal 1 line
-			 * difference that earlier platforms have. External
-			 * DP is unknown. For now just check for the 2 line
-			 * difference case on all output types on HSW+.
-			 *
-			 * This might misinterpret the scanline counter being
-			 * one line too far along on eDP, but that's less
-			 * dangerous than the alternative since that would lead
-			 * the vblank timestamp code astray when it sees a
-			 * scanline count before vblank_start during a vblank
-			 * interrupt.
-			 */
-			in_vbl = ilk_pipe_in_vblank_locked(dev, pipe);
-			if ((in_vbl && (position == vbl_start - 2 ||
-					position == vbl_start - 1)) ||
-			    (!in_vbl && (position == vbl_end - 2 ||
-					 position == vbl_end - 1)))
-				position = (position + 2) % vtotal;
-		} else if (HAS_PCH_SPLIT(dev)) {
-			/*
-			 * The scanline counter increments at the leading edge
-			 * of hsync, ie. it completely misses the active portion
-			 * of the line. Fix up the counter at both edges of vblank
-			 * to get a more accurate picture whether we're in vblank
-			 * or not.
-			 */
-			in_vbl = ilk_pipe_in_vblank_locked(dev, pipe);
-			if ((in_vbl && position == vbl_start - 1) ||
-			    (!in_vbl && position == vbl_end - 1))
-				position = (position + 1) % vtotal;
-		} else {
-			/*
-			 * ISR vblank status bits don't work the way we'd want
-			 * them to work on non-PCH platforms (for
-			 * ilk_pipe_in_vblank_locked()), and there doesn't
-			 * appear any other way to determine if we're currently
-			 * in vblank.
-			 *
-			 * Instead let's assume that we're already in vblank if
-			 * we got called from the vblank interrupt and the
-			 * scanline counter value indicates that we're on the
-			 * line just prior to vblank start. This should result
-			 * in the correct answer, unless the vblank interrupt
-			 * delivery really got delayed for almost exactly one
-			 * full frame/field.
-			 */
-			if (flags & DRM_CALLED_FROM_VBLIRQ &&
-			    position == vbl_start - 1) {
-				position = (position + 1) % vtotal;
-
-				/* Signal this correction as "applied". */
-				ret |= 0x8;
-			}
-		}
+		position = __intel_get_crtc_scanline(intel_crtc);
 	} else {
 		/* Have access to pixelcount since start of frame.
 		 * We can split this into vertical and horizontal
@@ -862,6 +964,29 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 		vbl_start *= htotal;
 		vbl_end *= htotal;
 		vtotal *= htotal;
+
+		/*
+		 * In interlaced modes, the pixel counter counts all pixels,
+		 * so one field will have htotal more pixels. In order to avoid
+		 * the reported position from jumping backwards when the pixel
+		 * counter is beyond the length of the shorter field, just
+		 * clamp the position the length of the shorter field. This
+		 * matches how the scanline counter based position works since
+		 * the scanline counter doesn't count the two half lines.
+		 */
+		if (position >= vtotal)
+			position = vtotal - 1;
+
+		/*
+		 * Start of vblank interrupt is triggered at start of hsync,
+		 * just prior to the first active line of vblank. However we
+		 * consider lines to start at the leading edge of horizontal
+		 * active. So, should we get here before we've crossed into
+		 * the horizontal active of the first line in vblank, we would
+		 * not set the DRM_SCANOUTPOS_INVBL flag. In order to fix that,
+		 * always add htotal-hsync_start to the current pixel position.
+		 */
+		position = (position + htotal - hsync_start) % vtotal;
 	}
 
 	/* Get optional system timestamp after query. */
@@ -900,6 +1025,19 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	return ret;
 }
 
+int intel_get_crtc_scanline(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	unsigned long irqflags;
+	int position;
+
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+	position = __intel_get_crtc_scanline(crtc);
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+
+	return position;
+}
+
 static int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
 			      int *max_error,
 			      struct timeval *vblank_time,
@@ -945,7 +1083,7 @@ static bool intel_hpd_irq_event(struct drm_device *dev,
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n",
 		      connector->base.id,
-		      drm_get_connector_name(connector),
+		      connector->name,
 		      drm_get_connector_status_name(old_status),
 		      drm_get_connector_status_name(connector->status));
 
@@ -990,7 +1128,7 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		    connector->polled == DRM_CONNECTOR_POLL_HPD) {
 			DRM_INFO("HPD interrupt storm detected on connector %s: "
 				 "switching from hotplug detection to polling\n",
-				drm_get_connector_name(connector));
+				connector->name);
 			dev_priv->hpd_stats[intel_encoder->hpd_pin].hpd_mark = HPD_DISABLED;
 			connector->polled = DRM_CONNECTOR_POLL_CONNECT
 				| DRM_CONNECTOR_POLL_DISCONNECT;
@@ -998,7 +1136,7 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		}
 		if (hpd_event_bits & (1 << intel_encoder->hpd_pin)) {
 			DRM_DEBUG_KMS("Connector %s (pin %i) received hotplug event.\n",
-				      drm_get_connector_name(connector), intel_encoder->hpd_pin);
+				      connector->name, intel_encoder->hpd_pin);
 		}
 	}
 	 /* if there were no outputs to poll, poll was disabled,
@@ -1073,9 +1211,9 @@ static void ironlake_rps_change_irq_handler(struct drm_device *dev)
 }
 
 static void notify_ring(struct drm_device *dev,
-			struct intel_ring_buffer *ring)
+			struct intel_engine_cs *ring)
 {
-	if (ring->obj == NULL)
+	if (!intel_ring_initialized(ring))
 		return;
 
 	trace_i915_gem_request_complete(ring);
@@ -1094,8 +1232,12 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	spin_lock_irq(&dev_priv->irq_lock);
 	pm_iir = dev_priv->rps.pm_iir;
 	dev_priv->rps.pm_iir = 0;
-	/* Make sure not to corrupt PMIMR state used by ringbuffer code */
-	snb_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	if (IS_BROADWELL(dev_priv->dev))
+		bdw_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	else {
+		/* Make sure not to corrupt PMIMR state used by ringbuffer */
+		snb_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	}
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	/* Make sure we didn't queue anything we're not going to process. */
@@ -1292,6 +1434,19 @@ static void snb_gt_irq_handler(struct drm_device *dev,
 		ivybridge_parity_error_irq_handler(dev, gt_iir);
 }
 
+static void gen8_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
+{
+	if ((pm_iir & dev_priv->pm_rps_events) == 0)
+		return;
+
+	spin_lock(&dev_priv->irq_lock);
+	dev_priv->rps.pm_iir |= pm_iir & dev_priv->pm_rps_events;
+	bdw_disable_pm_irq(dev_priv, pm_iir & dev_priv->pm_rps_events);
+	spin_unlock(&dev_priv->irq_lock);
+
+	queue_work(dev_priv->wq, &dev_priv->rps.work);
+}
+
 static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
 				       struct drm_i915_private *dev_priv,
 				       u32 master_ctl)
@@ -1315,18 +1470,32 @@ static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
 			DRM_ERROR("The master control interrupt lied (GT0)!\n");
 	}
 
-	if (master_ctl & GEN8_GT_VCS1_IRQ) {
+	if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) {
 		tmp = I915_READ(GEN8_GT_IIR(1));
 		if (tmp) {
 			ret = IRQ_HANDLED;
 			vcs = tmp >> GEN8_VCS1_IRQ_SHIFT;
 			if (vcs & GT_RENDER_USER_INTERRUPT)
 				notify_ring(dev, &dev_priv->ring[VCS]);
+			vcs = tmp >> GEN8_VCS2_IRQ_SHIFT;
+			if (vcs & GT_RENDER_USER_INTERRUPT)
+				notify_ring(dev, &dev_priv->ring[VCS2]);
 			I915_WRITE(GEN8_GT_IIR(1), tmp);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT1)!\n");
 	}
 
+	if (master_ctl & GEN8_GT_PM_IRQ) {
+		tmp = I915_READ(GEN8_GT_IIR(2));
+		if (tmp & dev_priv->pm_rps_events) {
+			ret = IRQ_HANDLED;
+			gen8_rps_irq_handler(dev_priv, tmp);
+			I915_WRITE(GEN8_GT_IIR(2),
+				   tmp & dev_priv->pm_rps_events);
+		} else
+			DRM_ERROR("The master control interrupt lied (PM)!\n");
+	}
+
 	if (master_ctl & GEN8_GT_VECS_IRQ) {
 		tmp = I915_READ(GEN8_GT_IIR(3));
 		if (tmp) {
@@ -1549,6 +1718,19 @@ static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 	}
 }
 
+static bool intel_pipe_handle_vblank(struct drm_device *dev, enum pipe pipe)
+{
+	struct intel_crtc *crtc;
+
+	if (!drm_handle_vblank(dev, pipe))
+		return false;
+
+	crtc = to_intel_crtc(intel_get_crtc_for_pipe(dev, pipe));
+	wake_up(&crtc->vbl_wait);
+
+	return true;
+}
+
 static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1578,6 +1760,9 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 		case PIPE_B:
 			iir_bit = I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
 			break;
+		case PIPE_C:
+			iir_bit = I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+			break;
 		}
 		if (iir & iir_bit)
 			mask |= dev_priv->pipestat_irq_mask[pipe];
@@ -1600,7 +1785,7 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 
 	for_each_pipe(pipe) {
 		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
 		if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV) {
 			intel_prepare_page_flip(dev, pipe);
@@ -1619,9 +1804,36 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 		gmbus_irq_handler(dev);
 }
 
+static void i9xx_hpd_irq_handler(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
+
+	if (IS_G4X(dev)) {
+		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
+
+		intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_g4x);
+	} else {
+		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
+
+		intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
+	}
+
+	if ((IS_G4X(dev) || IS_VALLEYVIEW(dev)) &&
+	    hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
+		dp_aux_irq_handler(dev);
+
+	I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
+	/*
+	 * Make sure hotplug status is cleared before we clear IIR, or else we
+	 * may miss hotplug events.
+	 */
+	POSTING_READ(PORT_HOTPLUG_STAT);
+}
+
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 iir, gt_iir, pm_iir;
 	irqreturn_t ret = IRQ_NONE;
@@ -1641,19 +1853,8 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 		valleyview_pipestat_irq_handler(dev, iir);
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if (iir & I915_DISPLAY_PORT_INTERRUPT) {
-			u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
-
-			intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
-
-			if (hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
-				dp_aux_irq_handler(dev);
-
-			I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-			I915_READ(PORT_HOTPLUG_STAT);
-		}
-
+		if (iir & I915_DISPLAY_PORT_INTERRUPT)
+			i9xx_hpd_irq_handler(dev);
 
 		if (pm_iir)
 			gen6_rps_irq_handler(dev_priv, pm_iir);
@@ -1667,6 +1868,40 @@ out:
 	return ret;
 }
 
+static irqreturn_t cherryview_irq_handler(int irq, void *arg)
+{
+	struct drm_device *dev = arg;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 master_ctl, iir;
+	irqreturn_t ret = IRQ_NONE;
+
+	for (;;) {
+		master_ctl = I915_READ(GEN8_MASTER_IRQ) & ~GEN8_MASTER_IRQ_CONTROL;
+		iir = I915_READ(VLV_IIR);
+
+		if (master_ctl == 0 && iir == 0)
+			break;
+
+		I915_WRITE(GEN8_MASTER_IRQ, 0);
+
+		gen8_gt_irq_handler(dev, dev_priv, master_ctl);
+
+		valleyview_pipestat_irq_handler(dev, iir);
+
+		/* Consume port.  Then clear IIR or we'll miss events */
+		i9xx_hpd_irq_handler(dev);
+
+		I915_WRITE(VLV_IIR, iir);
+
+		I915_WRITE(GEN8_MASTER_IRQ, DE_MASTER_IRQ_CONTROL);
+		POSTING_READ(GEN8_MASTER_IRQ);
+
+		ret = IRQ_HANDLED;
+	}
+
+	return ret;
+}
+
 static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1827,7 +2062,7 @@ static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 
 	for_each_pipe(pipe) {
 		if (de_iir & DE_PIPE_VBLANK(pipe))
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
 		if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe))
 			if (intel_set_cpu_fifo_underrun_reporting(dev, pipe, false))
@@ -1877,7 +2112,7 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 
 	for_each_pipe(pipe) {
 		if (de_iir & (DE_PIPE_VBLANK_IVB(pipe)))
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
 		/* plane/pipes map 1:1 on ilk+ */
 		if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe)) {
@@ -1899,7 +2134,7 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 
 static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 de_iir, gt_iir, de_ier, sde_ier = 0;
 	irqreturn_t ret = IRQ_NONE;
@@ -2020,9 +2255,9 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 
 		pipe_iir = I915_READ(GEN8_DE_PIPE_IIR(pipe));
 		if (pipe_iir & GEN8_PIPE_VBLANK)
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
-		if (pipe_iir & GEN8_PIPE_FLIP_DONE) {
+		if (pipe_iir & GEN8_PIPE_PRIMARY_FLIP_DONE) {
 			intel_prepare_page_flip(dev, pipe);
 			intel_finish_page_flip_plane(dev, pipe);
 		}
@@ -2075,7 +2310,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 			       bool reset_completed)
 {
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	/*
@@ -2137,6 +2372,14 @@ static void i915_error_work_func(struct work_struct *work)
 				   reset_event);
 
 		/*
+		 * In most cases it's guaranteed that we get here with an RPM
+		 * reference held, for example because there is a pending GPU
+		 * request that won't finish until the reset is done. This
+		 * isn't the case at least when we get here by doing a
+		 * simulated reset via debugs, so get an RPM reference.
+		 */
+		intel_runtime_pm_get(dev_priv);
+		/*
 		 * All state reset _must_ be completed before we update the
 		 * reset counter, for otherwise waiters might miss the reset
 		 * pending state and not properly drop locks, resulting in
@@ -2146,6 +2389,8 @@ static void i915_error_work_func(struct work_struct *work)
 
 		intel_display_handle_reset(dev);
 
+		intel_runtime_pm_put(dev_priv);
+
 		if (ret == 0) {
 			/*
 			 * After all the gem state is reset, increment the reset
@@ -2383,10 +2628,6 @@ static int i915_enable_vblank(struct drm_device *dev, int pipe)
 	else
 		i915_enable_pipestat(dev_priv, pipe,
 				     PIPE_VBLANK_INTERRUPT_STATUS);
-
-	/* maintain vblank delivery even in deep C-states */
-	if (INTEL_INFO(dev)->gen == 3)
-		I915_WRITE(INSTPM, _MASKED_BIT_DISABLE(INSTPM_AGPBUSY_DIS));
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
 	return 0;
@@ -2450,9 +2691,6 @@ static void i915_disable_vblank(struct drm_device *dev, int pipe)
 	unsigned long irqflags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	if (INTEL_INFO(dev)->gen == 3)
-		I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_AGPBUSY_DIS));
-
 	i915_disable_pipestat(dev_priv, pipe,
 			      PIPE_VBLANK_INTERRUPT_STATUS |
 			      PIPE_START_VBLANK_INTERRUPT_STATUS);
@@ -2498,29 +2736,77 @@ static void gen8_disable_vblank(struct drm_device *dev, int pipe)
 }
 
 static u32
-ring_last_seqno(struct intel_ring_buffer *ring)
+ring_last_seqno(struct intel_engine_cs *ring)
 {
 	return list_entry(ring->request_list.prev,
 			  struct drm_i915_gem_request, list)->seqno;
 }
 
 static bool
-ring_idle(struct intel_ring_buffer *ring, u32 seqno)
+ring_idle(struct intel_engine_cs *ring, u32 seqno)
 {
 	return (list_empty(&ring->request_list) ||
 		i915_seqno_passed(seqno, ring_last_seqno(ring)));
 }
 
-static struct intel_ring_buffer *
-semaphore_waits_for(struct intel_ring_buffer *ring, u32 *seqno)
+static bool
+ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr)
+{
+	if (INTEL_INFO(dev)->gen >= 8) {
+		/*
+		 * FIXME: gen8 semaphore support - currently we don't emit
+		 * semaphores on bdw anyway, but this needs to be addressed when
+		 * we merge that code.
+		 */
+		return false;
+	} else {
+		ipehr &= ~MI_SEMAPHORE_SYNC_MASK;
+		return ipehr == (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE |
+				 MI_SEMAPHORE_REGISTER);
+	}
+}
+
+static struct intel_engine_cs *
+semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct intel_engine_cs *signaller;
+	int i;
+
+	if (INTEL_INFO(dev_priv->dev)->gen >= 8) {
+		/*
+		 * FIXME: gen8 semaphore support - currently we don't emit
+		 * semaphores on bdw anyway, but this needs to be addressed when
+		 * we merge that code.
+		 */
+		return NULL;
+	} else {
+		u32 sync_bits = ipehr & MI_SEMAPHORE_SYNC_MASK;
+
+		for_each_ring(signaller, dev_priv, i) {
+			if(ring == signaller)
+				continue;
+
+			if (sync_bits == signaller->semaphore.mbox.wait[ring->id])
+				return signaller;
+		}
+	}
+
+	DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x\n",
+		  ring->id, ipehr);
+
+	return NULL;
+}
+
+static struct intel_engine_cs *
+semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 cmd, ipehr, head;
 	int i;
 
 	ipehr = I915_READ(RING_IPEHR(ring->mmio_base));
-	if ((ipehr & ~(0x3 << 16)) !=
-	    (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE | MI_SEMAPHORE_REGISTER))
+	if (!ipehr_is_semaphore_wait(ring->dev, ipehr))
 		return NULL;
 
 	/*
@@ -2538,10 +2824,10 @@ semaphore_waits_for(struct intel_ring_buffer *ring, u32 *seqno)
 		 * our ring is smaller than what the hardware (and hence
 		 * HEAD_ADDR) allows. Also handles wrap-around.
 		 */
-		head &= ring->size - 1;
+		head &= ring->buffer->size - 1;
 
 		/* This here seems to blow up */
-		cmd = ioread32(ring->virtual_start + head);
+		cmd = ioread32(ring->buffer->virtual_start + head);
 		if (cmd == ipehr)
 			break;
 
@@ -2551,20 +2837,24 @@ semaphore_waits_for(struct intel_ring_buffer *ring, u32 *seqno)
 	if (!i)
 		return NULL;
 
-	*seqno = ioread32(ring->virtual_start + head + 4) + 1;
-	return &dev_priv->ring[(ring->id + (((ipehr >> 17) & 1) + 1)) % 3];
+	*seqno = ioread32(ring->buffer->virtual_start + head + 4) + 1;
+	return semaphore_wait_to_signaller_ring(ring, ipehr);
 }
 
-static int semaphore_passed(struct intel_ring_buffer *ring)
+static int semaphore_passed(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	struct intel_ring_buffer *signaller;
+	struct intel_engine_cs *signaller;
 	u32 seqno, ctl;
 
-	ring->hangcheck.deadlock = true;
+	ring->hangcheck.deadlock++;
 
 	signaller = semaphore_waits_for(ring, &seqno);
-	if (signaller == NULL || signaller->hangcheck.deadlock)
+	if (signaller == NULL)
+		return -1;
+
+	/* Prevent pathological recursion due to driver bugs */
+	if (signaller->hangcheck.deadlock >= I915_NUM_RINGS)
 		return -1;
 
 	/* cursory check for an unkickable deadlock */
@@ -2572,20 +2862,26 @@ static int semaphore_passed(struct intel_ring_buffer *ring)
 	if (ctl & RING_WAIT_SEMAPHORE && semaphore_passed(signaller) < 0)
 		return -1;
 
-	return i915_seqno_passed(signaller->get_seqno(signaller, false), seqno);
+	if (i915_seqno_passed(signaller->get_seqno(signaller, false), seqno))
+		return 1;
+
+	if (signaller->hangcheck.deadlock)
+		return -1;
+
+	return 0;
 }
 
 static void semaphore_clear_deadlocks(struct drm_i915_private *dev_priv)
 {
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	for_each_ring(ring, dev_priv, i)
-		ring->hangcheck.deadlock = false;
+		ring->hangcheck.deadlock = 0;
 }
 
 static enum intel_ring_hangcheck_action
-ring_stuck(struct intel_ring_buffer *ring, u64 acthd)
+ring_stuck(struct intel_engine_cs *ring, u64 acthd)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2641,7 +2937,7 @@ static void i915_hangcheck_elapsed(unsigned long data)
 {
 	struct drm_device *dev = (struct drm_device *)data;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 	int busy_count = 0, rings_hung = 0;
 	bool stuck[I915_NUM_RINGS] = { 0 };
@@ -2759,57 +3055,63 @@ void i915_queue_hangcheck(struct drm_device *dev)
 		  round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES));
 }
 
-static void ibx_irq_preinstall(struct drm_device *dev)
+static void ibx_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (HAS_PCH_NOP(dev))
 		return;
 
-	/* south display irq */
-	I915_WRITE(SDEIMR, 0xffffffff);
-	/*
-	 * SDEIER is also touched by the interrupt handler to work around missed
-	 * PCH interrupts. Hence we can't update it after the interrupt handler
-	 * is enabled - instead we unconditionally enable all PCH interrupt
-	 * sources here, but then only unmask them as needed with SDEIMR.
-	 */
+	GEN5_IRQ_RESET(SDE);
+
+	if (HAS_PCH_CPT(dev) || HAS_PCH_LPT(dev))
+		I915_WRITE(SERR_INT, 0xffffffff);
+}
+
+/*
+ * SDEIER is also touched by the interrupt handler to work around missed PCH
+ * interrupts. Hence we can't update it after the interrupt handler is enabled -
+ * instead we unconditionally enable all PCH interrupt sources here, but then
+ * only unmask them as needed with SDEIMR.
+ *
+ * This function needs to be called before interrupts are enabled.
+ */
+static void ibx_irq_pre_postinstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (HAS_PCH_NOP(dev))
+		return;
+
+	WARN_ON(I915_READ(SDEIER) != 0);
 	I915_WRITE(SDEIER, 0xffffffff);
 	POSTING_READ(SDEIER);
 }
 
-static void gen5_gt_irq_preinstall(struct drm_device *dev)
+static void gen5_gt_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	/* and GT */
-	I915_WRITE(GTIMR, 0xffffffff);
-	I915_WRITE(GTIER, 0x0);
-	POSTING_READ(GTIER);
-
-	if (INTEL_INFO(dev)->gen >= 6) {
-		/* and PM */
-		I915_WRITE(GEN6_PMIMR, 0xffffffff);
-		I915_WRITE(GEN6_PMIER, 0x0);
-		POSTING_READ(GEN6_PMIER);
-	}
+	GEN5_IRQ_RESET(GT);
+	if (INTEL_INFO(dev)->gen >= 6)
+		GEN5_IRQ_RESET(GEN6_PM);
 }
 
 /* drm_dma.h hooks
 */
-static void ironlake_irq_preinstall(struct drm_device *dev)
+static void ironlake_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	I915_WRITE(HWSTAM, 0xeffe);
+	I915_WRITE(HWSTAM, 0xffffffff);
 
-	I915_WRITE(DEIMR, 0xffffffff);
-	I915_WRITE(DEIER, 0x0);
-	POSTING_READ(DEIER);
+	GEN5_IRQ_RESET(DE);
+	if (IS_GEN7(dev))
+		I915_WRITE(GEN7_ERR_INT, 0xffffffff);
 
-	gen5_gt_irq_preinstall(dev);
+	gen5_gt_irq_reset(dev);
 
-	ibx_irq_preinstall(dev);
+	ibx_irq_reset(dev);
 }
 
 static void valleyview_irq_preinstall(struct drm_device *dev)
@@ -2827,7 +3129,7 @@ static void valleyview_irq_preinstall(struct drm_device *dev)
 	I915_WRITE(GTIIR, I915_READ(GTIIR));
 	I915_WRITE(GTIIR, I915_READ(GTIIR));
 
-	gen5_gt_irq_preinstall(dev);
+	gen5_gt_irq_reset(dev);
 
 	I915_WRITE(DPINVGTT, 0xff);
 
@@ -2841,7 +3143,15 @@ static void valleyview_irq_preinstall(struct drm_device *dev)
 	POSTING_READ(VLV_IER);
 }
 
-static void gen8_irq_preinstall(struct drm_device *dev)
+static void gen8_gt_irq_reset(struct drm_i915_private *dev_priv)
+{
+	GEN8_IRQ_RESET_NDX(GT, 0);
+	GEN8_IRQ_RESET_NDX(GT, 1);
+	GEN8_IRQ_RESET_NDX(GT, 2);
+	GEN8_IRQ_RESET_NDX(GT, 3);
+}
+
+static void gen8_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
@@ -2849,43 +3159,44 @@ static void gen8_irq_preinstall(struct drm_device *dev)
 	I915_WRITE(GEN8_MASTER_IRQ, 0);
 	POSTING_READ(GEN8_MASTER_IRQ);
 
-	/* IIR can theoretically queue up two events. Be paranoid */
-#define GEN8_IRQ_INIT_NDX(type, which) do { \
-		I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IMR(which)); \
-		I915_WRITE(GEN8_##type##_IER(which), 0); \
-		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IIR(which)); \
-		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-	} while (0)
-
-#define GEN8_IRQ_INIT(type) do { \
-		I915_WRITE(GEN8_##type##_IMR, 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IMR); \
-		I915_WRITE(GEN8_##type##_IER, 0); \
-		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IIR); \
-		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
-	} while (0)
-
-	GEN8_IRQ_INIT_NDX(GT, 0);
-	GEN8_IRQ_INIT_NDX(GT, 1);
-	GEN8_IRQ_INIT_NDX(GT, 2);
-	GEN8_IRQ_INIT_NDX(GT, 3);
+	gen8_gt_irq_reset(dev_priv);
 
-	for_each_pipe(pipe) {
-		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe);
-	}
+	for_each_pipe(pipe)
+		GEN8_IRQ_RESET_NDX(DE_PIPE, pipe);
+
+	GEN5_IRQ_RESET(GEN8_DE_PORT_);
+	GEN5_IRQ_RESET(GEN8_DE_MISC_);
+	GEN5_IRQ_RESET(GEN8_PCU_);
+
+	ibx_irq_reset(dev);
+}
+
+static void cherryview_irq_preinstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe;
+
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	POSTING_READ(GEN8_MASTER_IRQ);
 
-	GEN8_IRQ_INIT(DE_PORT);
-	GEN8_IRQ_INIT(DE_MISC);
-	GEN8_IRQ_INIT(PCU);
-#undef GEN8_IRQ_INIT
-#undef GEN8_IRQ_INIT_NDX
+	gen8_gt_irq_reset(dev_priv);
+
+	GEN5_IRQ_RESET(GEN8_PCU_);
 
 	POSTING_READ(GEN8_PCU_IIR);
 
-	ibx_irq_preinstall(dev);
+	I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
+
+	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
+
+	for_each_pipe(pipe)
+		I915_WRITE(PIPESTAT(pipe), 0xffff);
+
+	I915_WRITE(VLV_IMR, 0xffffffff);
+	I915_WRITE(VLV_IER, 0x0);
+	I915_WRITE(VLV_IIR, 0xffffffff);
+	POSTING_READ(VLV_IIR);
 }
 
 static void ibx_hpd_irq_setup(struct drm_device *dev)
@@ -2931,15 +3242,12 @@ static void ibx_irq_postinstall(struct drm_device *dev)
 	if (HAS_PCH_NOP(dev))
 		return;
 
-	if (HAS_PCH_IBX(dev)) {
+	if (HAS_PCH_IBX(dev))
 		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
-	} else {
+	else
 		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
 
-		I915_WRITE(SERR_INT, I915_READ(SERR_INT));
-	}
-
-	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
+	GEN5_ASSERT_IIR_IS_ZERO(SDEIIR);
 	I915_WRITE(SDEIMR, ~mask);
 }
 
@@ -2965,10 +3273,7 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev)
 		gt_irqs |= GT_BLT_USER_INTERRUPT | GT_BSD_USER_INTERRUPT;
 	}
 
-	I915_WRITE(GTIIR, I915_READ(GTIIR));
-	I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
-	I915_WRITE(GTIER, gt_irqs);
-	POSTING_READ(GTIER);
+	GEN5_IRQ_INIT(GT, dev_priv->gt_irq_mask, gt_irqs);
 
 	if (INTEL_INFO(dev)->gen >= 6) {
 		pm_irqs |= dev_priv->pm_rps_events;
@@ -2977,10 +3282,7 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev)
 			pm_irqs |= PM_VEBOX_USER_INTERRUPT;
 
 		dev_priv->pm_irq_mask = 0xffffffff;
-		I915_WRITE(GEN6_PMIIR, I915_READ(GEN6_PMIIR));
-		I915_WRITE(GEN6_PMIMR, dev_priv->pm_irq_mask);
-		I915_WRITE(GEN6_PMIER, pm_irqs);
-		POSTING_READ(GEN6_PMIER);
+		GEN5_IRQ_INIT(GEN6_PM, dev_priv->pm_irq_mask, pm_irqs);
 	}
 }
 
@@ -2997,8 +3299,6 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 				DE_PLANEA_FLIP_DONE_IVB | DE_AUX_CHANNEL_A_IVB);
 		extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB |
 			      DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB);
-
-		I915_WRITE(GEN7_ERR_INT, I915_READ(GEN7_ERR_INT));
 	} else {
 		display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT |
 				DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE |
@@ -3011,11 +3311,11 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 
 	dev_priv->irq_mask = ~display_mask;
 
-	/* should always can generate irq */
-	I915_WRITE(DEIIR, I915_READ(DEIIR));
-	I915_WRITE(DEIMR, dev_priv->irq_mask);
-	I915_WRITE(DEIER, display_mask | extra_mask);
-	POSTING_READ(DEIER);
+	I915_WRITE(HWSTAM, 0xeffe);
+
+	ibx_irq_pre_postinstall(dev);
+
+	GEN5_IRQ_INIT(DE, dev_priv->irq_mask, display_mask | extra_mask);
 
 	gen5_gt_irq_postinstall(dev);
 
@@ -3175,21 +3475,16 @@ static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv)
 		GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT
 		};
 
-	for (i = 0; i < ARRAY_SIZE(gt_interrupts); i++) {
-		u32 tmp = I915_READ(GEN8_GT_IIR(i));
-		if (tmp)
-			DRM_ERROR("Interrupt (%d) should have been masked in pre-install 0x%08x\n",
-				  i, tmp);
-		I915_WRITE(GEN8_GT_IMR(i), ~gt_interrupts[i]);
-		I915_WRITE(GEN8_GT_IER(i), gt_interrupts[i]);
-	}
-	POSTING_READ(GEN8_GT_IER(0));
+	for (i = 0; i < ARRAY_SIZE(gt_interrupts); i++)
+		GEN8_IRQ_INIT_NDX(GT, i, ~gt_interrupts[i], gt_interrupts[i]);
+
+	dev_priv->pm_irq_mask = 0xffffffff;
 }
 
 static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
-	uint32_t de_pipe_masked = GEN8_PIPE_FLIP_DONE |
+	uint32_t de_pipe_masked = GEN8_PIPE_PRIMARY_FLIP_DONE |
 		GEN8_PIPE_CDCLK_CRC_DONE |
 		GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
 	uint32_t de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK |
@@ -3199,25 +3494,19 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	dev_priv->de_irq_mask[PIPE_B] = ~de_pipe_masked;
 	dev_priv->de_irq_mask[PIPE_C] = ~de_pipe_masked;
 
-	for_each_pipe(pipe) {
-		u32 tmp = I915_READ(GEN8_DE_PIPE_IIR(pipe));
-		if (tmp)
-			DRM_ERROR("Interrupt (%d) should have been masked in pre-install 0x%08x\n",
-				  pipe, tmp);
-		I915_WRITE(GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
-		I915_WRITE(GEN8_DE_PIPE_IER(pipe), de_pipe_enables);
-	}
-	POSTING_READ(GEN8_DE_PIPE_ISR(0));
+	for_each_pipe(pipe)
+		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe, dev_priv->de_irq_mask[pipe],
+				  de_pipe_enables);
 
-	I915_WRITE(GEN8_DE_PORT_IMR, ~GEN8_AUX_CHANNEL_A);
-	I915_WRITE(GEN8_DE_PORT_IER, GEN8_AUX_CHANNEL_A);
-	POSTING_READ(GEN8_DE_PORT_IER);
+	GEN5_IRQ_INIT(GEN8_DE_PORT_, ~GEN8_AUX_CHANNEL_A, GEN8_AUX_CHANNEL_A);
 }
 
 static int gen8_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	ibx_irq_pre_postinstall(dev);
+
 	gen8_gt_irq_postinstall(dev_priv);
 	gen8_de_irq_postinstall(dev_priv);
 
@@ -3229,44 +3518,55 @@ static int gen8_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
-static void gen8_irq_uninstall(struct drm_device *dev)
+static int cherryview_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 enable_mask = I915_DISPLAY_PORT_INTERRUPT |
+		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
+		I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+	u32 pipestat_enable = PLANE_FLIP_DONE_INT_STATUS_VLV |
+		PIPE_CRC_DONE_INTERRUPT_STATUS;
+	unsigned long irqflags;
 	int pipe;
 
-	if (!dev_priv)
-		return;
+	/*
+	 * Leave vblank interrupts masked initially.  enable/disable will
+	 * toggle them based on usage.
+	 */
+	dev_priv->irq_mask = ~enable_mask;
 
-	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	for_each_pipe(pipe)
+		I915_WRITE(PIPESTAT(pipe), 0xffff);
 
-#define GEN8_IRQ_FINI_NDX(type, which) do { \
-		I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
-		I915_WRITE(GEN8_##type##_IER(which), 0); \
-		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-	} while (0)
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
+	for_each_pipe(pipe)
+		i915_enable_pipestat(dev_priv, pipe, pipestat_enable);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
-#define GEN8_IRQ_FINI(type) do { \
-		I915_WRITE(GEN8_##type##_IMR, 0xffffffff); \
-		I915_WRITE(GEN8_##type##_IER, 0); \
-		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
-	} while (0)
+	I915_WRITE(VLV_IIR, 0xffffffff);
+	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
+	I915_WRITE(VLV_IER, enable_mask);
 
-	GEN8_IRQ_FINI_NDX(GT, 0);
-	GEN8_IRQ_FINI_NDX(GT, 1);
-	GEN8_IRQ_FINI_NDX(GT, 2);
-	GEN8_IRQ_FINI_NDX(GT, 3);
+	gen8_gt_irq_postinstall(dev_priv);
 
-	for_each_pipe(pipe) {
-		GEN8_IRQ_FINI_NDX(DE_PIPE, pipe);
-	}
+	I915_WRITE(GEN8_MASTER_IRQ, MASTER_INTERRUPT_ENABLE);
+	POSTING_READ(GEN8_MASTER_IRQ);
 
-	GEN8_IRQ_FINI(DE_PORT);
-	GEN8_IRQ_FINI(DE_MISC);
-	GEN8_IRQ_FINI(PCU);
-#undef GEN8_IRQ_FINI
-#undef GEN8_IRQ_FINI_NDX
+	return 0;
+}
 
-	POSTING_READ(GEN8_PCU_IIR);
+static void gen8_irq_uninstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!dev_priv)
+		return;
+
+	intel_hpd_irq_uninstall(dev_priv);
+
+	gen8_irq_reset(dev);
 }
 
 static void valleyview_irq_uninstall(struct drm_device *dev)
@@ -3278,6 +3578,8 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 	if (!dev_priv)
 		return;
 
+	I915_WRITE(VLV_MASTER_IER, 0);
+
 	intel_hpd_irq_uninstall(dev_priv);
 
 	for_each_pipe(pipe)
@@ -3300,35 +3602,67 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 	POSTING_READ(VLV_IER);
 }
 
-static void ironlake_irq_uninstall(struct drm_device *dev)
+static void cherryview_irq_uninstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe;
 
 	if (!dev_priv)
 		return;
 
-	intel_hpd_irq_uninstall(dev_priv);
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	POSTING_READ(GEN8_MASTER_IRQ);
 
-	I915_WRITE(HWSTAM, 0xffffffff);
+#define GEN8_IRQ_FINI_NDX(type, which)				\
+do {								\
+	I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff);	\
+	I915_WRITE(GEN8_##type##_IER(which), 0);		\
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff);	\
+	POSTING_READ(GEN8_##type##_IIR(which));			\
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff);	\
+} while (0)
+
+#define GEN8_IRQ_FINI(type)				\
+do {							\
+	I915_WRITE(GEN8_##type##_IMR, 0xffffffff);	\
+	I915_WRITE(GEN8_##type##_IER, 0);		\
+	I915_WRITE(GEN8_##type##_IIR, 0xffffffff);	\
+	POSTING_READ(GEN8_##type##_IIR);		\
+	I915_WRITE(GEN8_##type##_IIR, 0xffffffff);	\
+} while (0)
 
-	I915_WRITE(DEIMR, 0xffffffff);
-	I915_WRITE(DEIER, 0x0);
-	I915_WRITE(DEIIR, I915_READ(DEIIR));
-	if (IS_GEN7(dev))
-		I915_WRITE(GEN7_ERR_INT, I915_READ(GEN7_ERR_INT));
+	GEN8_IRQ_FINI_NDX(GT, 0);
+	GEN8_IRQ_FINI_NDX(GT, 1);
+	GEN8_IRQ_FINI_NDX(GT, 2);
+	GEN8_IRQ_FINI_NDX(GT, 3);
 
-	I915_WRITE(GTIMR, 0xffffffff);
-	I915_WRITE(GTIER, 0x0);
-	I915_WRITE(GTIIR, I915_READ(GTIIR));
+	GEN8_IRQ_FINI(PCU);
 
-	if (HAS_PCH_NOP(dev))
+#undef GEN8_IRQ_FINI
+#undef GEN8_IRQ_FINI_NDX
+
+	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
+
+	for_each_pipe(pipe)
+		I915_WRITE(PIPESTAT(pipe), 0xffff);
+
+	I915_WRITE(VLV_IMR, 0xffffffff);
+	I915_WRITE(VLV_IER, 0x0);
+	I915_WRITE(VLV_IIR, 0xffffffff);
+	POSTING_READ(VLV_IIR);
+}
+
+static void ironlake_irq_uninstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!dev_priv)
 		return;
 
-	I915_WRITE(SDEIMR, 0xffffffff);
-	I915_WRITE(SDEIER, 0x0);
-	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
-	if (HAS_PCH_CPT(dev) || HAS_PCH_LPT(dev))
-		I915_WRITE(SERR_INT, I915_READ(SERR_INT));
+	intel_hpd_irq_uninstall(dev_priv);
+
+	ironlake_irq_reset(dev);
 }
 
 static void i8xx_irq_preinstall(struct drm_device * dev)
@@ -3386,7 +3720,7 @@ static bool i8xx_handle_vblank(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u16 flip_pending = DISPLAY_PLANE_FLIP_PENDING(plane);
 
-	if (!drm_handle_vblank(dev, pipe))
+	if (!intel_pipe_handle_vblank(dev, pipe))
 		return false;
 
 	if ((iir & flip_pending) == 0)
@@ -3410,7 +3744,7 @@ static bool i8xx_handle_vblank(struct drm_device *dev,
 
 static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u16 iir, new_iir;
 	u32 pipe_stats[2];
@@ -3571,7 +3905,7 @@ static bool i915_handle_vblank(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 flip_pending = DISPLAY_PLANE_FLIP_PENDING(plane);
 
-	if (!drm_handle_vblank(dev, pipe))
+	if (!intel_pipe_handle_vblank(dev, pipe))
 		return false;
 
 	if ((iir & flip_pending) == 0)
@@ -3595,7 +3929,7 @@ static bool i915_handle_vblank(struct drm_device *dev,
 
 static irqreturn_t i915_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 iir, new_iir, pipe_stats[I915_MAX_PIPES];
 	unsigned long irqflags;
@@ -3636,16 +3970,9 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 			break;
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if ((I915_HAS_HOTPLUG(dev)) &&
-		    (iir & I915_DISPLAY_PORT_INTERRUPT)) {
-			u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
-
-			intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
-
-			I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-			POSTING_READ(PORT_HOTPLUG_STAT);
-		}
+		if (I915_HAS_HOTPLUG(dev) &&
+		    iir & I915_DISPLAY_PORT_INTERRUPT)
+			i9xx_hpd_irq_handler(dev);
 
 		I915_WRITE(IIR, iir & ~flip_mask);
 		new_iir = I915_READ(IIR); /* Flush posted writes */
@@ -3832,7 +4159,7 @@ static void i915_hpd_irq_setup(struct drm_device *dev)
 
 static irqreturn_t i965_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 iir, new_iir;
 	u32 pipe_stats[I915_MAX_PIPES];
@@ -3879,22 +4206,8 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		ret = IRQ_HANDLED;
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if (iir & I915_DISPLAY_PORT_INTERRUPT) {
-			u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-			u32 hotplug_trigger = hotplug_status & (IS_G4X(dev) ?
-								  HOTPLUG_INT_STATUS_G4X :
-								  HOTPLUG_INT_STATUS_I915);
-
-			intel_hpd_irq_handler(dev, hotplug_trigger,
-					      IS_G4X(dev) ? hpd_status_g4x : hpd_status_i915);
-
-			if (IS_G4X(dev) &&
-			    (hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X))
-				dp_aux_irq_handler(dev);
-
-			I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-			I915_READ(PORT_HOTPLUG_STAT);
-		}
+		if (iir & I915_DISPLAY_PORT_INTERRUPT)
+			i9xx_hpd_irq_handler(dev);
 
 		I915_WRITE(IIR, iir & ~flip_mask);
 		new_iir = I915_READ(IIR); /* Flush posted writes */
@@ -3997,7 +4310,7 @@ static void intel_hpd_irq_reenable(unsigned long data)
 			if (intel_connector->encoder->hpd_pin == i) {
 				if (connector->polled != intel_connector->polled)
 					DRM_DEBUG_DRIVER("Reenabling HPD on connector %s\n",
-							 drm_get_connector_name(connector));
+							 connector->name);
 				connector->polled = intel_connector->polled;
 				if (!connector->polled)
 					connector->polled = DRM_CONNECTOR_POLL_HPD;
@@ -4045,7 +4358,15 @@ void intel_irq_init(struct drm_device *dev)
 		dev->driver->get_scanout_position = i915_get_crtc_scanoutpos;
 	}
 
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		dev->driver->irq_handler = cherryview_irq_handler;
+		dev->driver->irq_preinstall = cherryview_irq_preinstall;
+		dev->driver->irq_postinstall = cherryview_irq_postinstall;
+		dev->driver->irq_uninstall = cherryview_irq_uninstall;
+		dev->driver->enable_vblank = valleyview_enable_vblank;
+		dev->driver->disable_vblank = valleyview_disable_vblank;
+		dev_priv->display.hpd_irq_setup = i915_hpd_irq_setup;
+	} else if (IS_VALLEYVIEW(dev)) {
 		dev->driver->irq_handler = valleyview_irq_handler;
 		dev->driver->irq_preinstall = valleyview_irq_preinstall;
 		dev->driver->irq_postinstall = valleyview_irq_postinstall;
@@ -4055,7 +4376,7 @@ void intel_irq_init(struct drm_device *dev)
 		dev_priv->display.hpd_irq_setup = i915_hpd_irq_setup;
 	} else if (IS_GEN8(dev)) {
 		dev->driver->irq_handler = gen8_irq_handler;
-		dev->driver->irq_preinstall = gen8_irq_preinstall;
+		dev->driver->irq_preinstall = gen8_irq_reset;
 		dev->driver->irq_postinstall = gen8_irq_postinstall;
 		dev->driver->irq_uninstall = gen8_irq_uninstall;
 		dev->driver->enable_vblank = gen8_enable_vblank;
@@ -4063,7 +4384,7 @@ void intel_irq_init(struct drm_device *dev)
 		dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
-		dev->driver->irq_preinstall = ironlake_irq_preinstall;
+		dev->driver->irq_preinstall = ironlake_irq_reset;
 		dev->driver->irq_postinstall = ironlake_irq_postinstall;
 		dev->driver->irq_uninstall = ironlake_irq_uninstall;
 		dev->driver->enable_vblank = ironlake_enable_vblank;
@@ -4121,57 +4442,20 @@ void intel_hpd_init(struct drm_device *dev)
 }
 
 /* Disable interrupts so we can allow runtime PM. */
-void hsw_runtime_pm_disable_interrupts(struct drm_device *dev)
+void intel_runtime_pm_disable_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-
-	dev_priv->pm.regsave.deimr = I915_READ(DEIMR);
-	dev_priv->pm.regsave.sdeimr = I915_READ(SDEIMR);
-	dev_priv->pm.regsave.gtimr = I915_READ(GTIMR);
-	dev_priv->pm.regsave.gtier = I915_READ(GTIER);
-	dev_priv->pm.regsave.gen6_pmimr = I915_READ(GEN6_PMIMR);
-
-	ironlake_disable_display_irq(dev_priv, 0xffffffff);
-	ibx_disable_display_interrupt(dev_priv, 0xffffffff);
-	ilk_disable_gt_irq(dev_priv, 0xffffffff);
-	snb_disable_pm_irq(dev_priv, 0xffffffff);
 
+	dev->driver->irq_uninstall(dev);
 	dev_priv->pm.irqs_disabled = true;
-
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
 /* Restore interrupts so we can recover from runtime PM. */
-void hsw_runtime_pm_restore_interrupts(struct drm_device *dev)
+void intel_runtime_pm_restore_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long irqflags;
-	uint32_t val;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-
-	val = I915_READ(DEIMR);
-	WARN(val != 0xffffffff, "DEIMR is 0x%08x\n", val);
-
-	val = I915_READ(SDEIMR);
-	WARN(val != 0xffffffff, "SDEIMR is 0x%08x\n", val);
-
-	val = I915_READ(GTIMR);
-	WARN(val != 0xffffffff, "GTIMR is 0x%08x\n", val);
-
-	val = I915_READ(GEN6_PMIMR);
-	WARN(val != 0xffffffff, "GEN6_PMIMR is 0x%08x\n", val);
 
 	dev_priv->pm.irqs_disabled = false;
-
-	ironlake_enable_display_irq(dev_priv, ~dev_priv->pm.regsave.deimr);
-	ibx_enable_display_interrupt(dev_priv, ~dev_priv->pm.regsave.sdeimr);
-	ilk_enable_gt_irq(dev_priv, ~dev_priv->pm.regsave.gtimr);
-	snb_enable_pm_irq(dev_priv, ~dev_priv->pm.regsave.gen6_pmimr);
-	I915_WRITE(GTIER, dev_priv->pm.regsave.gtier);
-
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	dev->driver->irq_preinstall(dev);
+	dev->driver->irq_postinstall(dev);
 }
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index d1d7980f0e01..d05a2afa17dc 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -46,7 +46,8 @@ struct i915_params i915 __read_mostly = {
 	.reset = true,
 	.invert_brightness = 0,
 	.disable_display = 0,
-	.enable_cmd_parser = 0,
+	.enable_cmd_parser = 1,
+	.disable_vtd_wa = 0,
 };
 
 module_param_named(modeset, i915.modeset, int, 0400);
@@ -149,6 +150,9 @@ MODULE_PARM_DESC(invert_brightness,
 module_param_named(disable_display, i915.disable_display, bool, 0600);
 MODULE_PARM_DESC(disable_display, "Disable display (default: false)");
 
+module_param_named(disable_vtd_wa, i915.disable_vtd_wa, bool, 0600);
+MODULE_PARM_DESC(disable_vtd_wa, "Disable all VT-d workarounds (default: false)");
+
 module_param_named(enable_cmd_parser, i915.enable_cmd_parser, int, 0600);
 MODULE_PARM_DESC(enable_cmd_parser,
-		 "Enable command parsing (1=enabled, 0=disabled [default])");
+		 "Enable command parsing (1=enabled [default], 0=disabled)");
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c77af69c2d8f..a5bab61bfc00 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -29,6 +29,8 @@
 #define _TRANSCODER(tran, a, b) ((a) + (tran)*((b)-(a)))
 
 #define _PORT(port, a, b) ((a) + (port)*((b)-(a)))
+#define _PIPE3(pipe, a, b, c) (pipe < 2 ? _PIPE(pipe, a, b) : c)
+#define _PORT3(port, a, b, c) (port < 2 ? _PORT(port, a, b) : c)
 
 #define _MASKED_BIT_ENABLE(a) (((a) << 16) | (a))
 #define _MASKED_BIT_DISABLE(a) ((a) << 16)
@@ -77,13 +79,19 @@
 
 /* Graphics reset regs */
 #define I965_GDRST 0xc0 /* PCI config register */
-#define ILK_GDSR 0x2ca4 /* MCHBAR offset */
 #define  GRDOM_FULL	(0<<2)
 #define  GRDOM_RENDER	(1<<2)
 #define  GRDOM_MEDIA	(3<<2)
 #define  GRDOM_MASK	(3<<2)
 #define  GRDOM_RESET_ENABLE (1<<0)
 
+#define ILK_GDSR 0x2ca4 /* MCHBAR offset */
+#define  ILK_GRDOM_FULL		(0<<1)
+#define  ILK_GRDOM_RENDER	(1<<1)
+#define  ILK_GRDOM_MEDIA	(3<<1)
+#define  ILK_GRDOM_MASK		(3<<1)
+#define  ILK_GRDOM_RESET_ENABLE (1<<0)
+
 #define GEN6_MBCUNIT_SNPCR	0x900c /* for LLC config */
 #define   GEN6_MBC_SNPCR_SHIFT	21
 #define   GEN6_MBC_SNPCR_MASK	(3<<21)
@@ -92,6 +100,9 @@
 #define   GEN6_MBC_SNPCR_LOW	(2<<21)
 #define   GEN6_MBC_SNPCR_MIN	(3<<21) /* only 1/16th of the cache is shared */
 
+#define VLV_G3DCTL		0x9024
+#define VLV_GSCKGCTL		0x9028
+
 #define GEN6_MBCTL		0x0907c
 #define   GEN6_MBCTL_ENABLE_BOOT_FETCH	(1 << 4)
 #define   GEN6_MBCTL_CTX_FETCH_NEEDED	(1 << 3)
@@ -190,6 +201,8 @@
  * Memory interface instructions used by the kernel
  */
 #define MI_INSTR(opcode, flags) (((opcode) << 23) | (flags))
+/* Many MI commands use bit 22 of the header dword for GGTT vs PPGTT */
+#define  MI_GLOBAL_GTT    (1<<22)
 
 #define MI_NOOP			MI_INSTR(0, 0)
 #define MI_USER_INTERRUPT	MI_INSTR(0x02, 0)
@@ -244,7 +257,8 @@
 #define   MI_SEMAPHORE_SYNC_BVE	    (0<<16) /* VECS wait for BCS  (VEBSYNC) */
 #define   MI_SEMAPHORE_SYNC_VVE	    (1<<16) /* VECS wait for VCS  (VEVSYNC) */
 #define   MI_SEMAPHORE_SYNC_RVE	    (2<<16) /* VECS wait for RCS  (VERSYNC) */
-#define   MI_SEMAPHORE_SYNC_INVALID  (3<<16)
+#define   MI_SEMAPHORE_SYNC_INVALID (3<<16)
+#define   MI_SEMAPHORE_SYNC_MASK    (3<<16)
 #define MI_SET_CONTEXT		MI_INSTR(0x18, 0)
 #define   MI_MM_SPACE_GTT		(1<<8)
 #define   MI_MM_SPACE_PHYSICAL		(0<<8)
@@ -262,13 +276,16 @@
  * - One can actually load arbitrary many arbitrary registers: Simply issue x
  *   address/value pairs. Don't overdue it, though, x <= 2^4 must hold!
  */
-#define MI_LOAD_REGISTER_IMM(x)	MI_INSTR(0x22, 2*x-1)
-#define MI_STORE_REGISTER_MEM(x) MI_INSTR(0x24, 2*x-1)
+#define MI_LOAD_REGISTER_IMM(x)	MI_INSTR(0x22, 2*(x)-1)
+#define MI_STORE_REGISTER_MEM(x) MI_INSTR(0x24, 2*(x)-1)
+#define MI_STORE_REGISTER_MEM_GEN8(x) MI_INSTR(0x24, 3*(x)-1)
 #define   MI_SRM_LRM_GLOBAL_GTT		(1<<22)
 #define MI_FLUSH_DW		MI_INSTR(0x26, 1) /* for GEN6 */
 #define   MI_FLUSH_DW_STORE_INDEX	(1<<21)
 #define   MI_INVALIDATE_TLB		(1<<18)
 #define   MI_FLUSH_DW_OP_STOREDW	(1<<14)
+#define   MI_FLUSH_DW_OP_MASK		(3<<14)
+#define   MI_FLUSH_DW_NOTIFY		(1<<8)
 #define   MI_INVALIDATE_BSD		(1<<7)
 #define   MI_FLUSH_DW_USE_GTT		(1<<2)
 #define   MI_FLUSH_DW_USE_PPGTT		(0<<2)
@@ -330,9 +347,12 @@
 #define   DISPLAY_PLANE_B           (1<<20)
 #define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|(len-2))
 #define   PIPE_CONTROL_GLOBAL_GTT_IVB			(1<<24) /* gen7+ */
+#define   PIPE_CONTROL_MMIO_WRITE			(1<<23)
+#define   PIPE_CONTROL_STORE_DATA_INDEX			(1<<21)
 #define   PIPE_CONTROL_CS_STALL				(1<<20)
 #define   PIPE_CONTROL_TLB_INVALIDATE			(1<<18)
 #define   PIPE_CONTROL_QW_WRITE				(1<<14)
+#define   PIPE_CONTROL_POST_SYNC_OP_MASK                (3<<14)
 #define   PIPE_CONTROL_DEPTH_STALL			(1<<13)
 #define   PIPE_CONTROL_WRITE_FLUSH			(1<<12)
 #define   PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH	(1<<12) /* gen6+ */
@@ -347,6 +367,94 @@
 #define   PIPE_CONTROL_DEPTH_CACHE_FLUSH		(1<<0)
 #define   PIPE_CONTROL_GLOBAL_GTT (1<<2) /* in addr dword */
 
+/*
+ * Commands used only by the command parser
+ */
+#define MI_SET_PREDICATE        MI_INSTR(0x01, 0)
+#define MI_ARB_CHECK            MI_INSTR(0x05, 0)
+#define MI_RS_CONTROL           MI_INSTR(0x06, 0)
+#define MI_URB_ATOMIC_ALLOC     MI_INSTR(0x09, 0)
+#define MI_PREDICATE            MI_INSTR(0x0C, 0)
+#define MI_RS_CONTEXT           MI_INSTR(0x0F, 0)
+#define MI_TOPOLOGY_FILTER      MI_INSTR(0x0D, 0)
+#define MI_LOAD_SCAN_LINES_EXCL MI_INSTR(0x13, 0)
+#define MI_URB_CLEAR            MI_INSTR(0x19, 0)
+#define MI_UPDATE_GTT           MI_INSTR(0x23, 0)
+#define MI_CLFLUSH              MI_INSTR(0x27, 0)
+#define MI_REPORT_PERF_COUNT    MI_INSTR(0x28, 0)
+#define   MI_REPORT_PERF_COUNT_GGTT (1<<0)
+#define MI_LOAD_REGISTER_MEM    MI_INSTR(0x29, 0)
+#define MI_LOAD_REGISTER_REG    MI_INSTR(0x2A, 0)
+#define MI_RS_STORE_DATA_IMM    MI_INSTR(0x2B, 0)
+#define MI_LOAD_URB_MEM         MI_INSTR(0x2C, 0)
+#define MI_STORE_URB_MEM        MI_INSTR(0x2D, 0)
+#define MI_CONDITIONAL_BATCH_BUFFER_END MI_INSTR(0x36, 0)
+
+#define PIPELINE_SELECT                ((0x3<<29)|(0x1<<27)|(0x1<<24)|(0x4<<16))
+#define GFX_OP_3DSTATE_VF_STATISTICS   ((0x3<<29)|(0x1<<27)|(0x0<<24)|(0xB<<16))
+#define MEDIA_VFE_STATE                ((0x3<<29)|(0x2<<27)|(0x0<<24)|(0x0<<16))
+#define  MEDIA_VFE_STATE_MMIO_ACCESS_MASK (0x18)
+#define GPGPU_OBJECT                   ((0x3<<29)|(0x2<<27)|(0x1<<24)|(0x4<<16))
+#define GPGPU_WALKER                   ((0x3<<29)|(0x2<<27)|(0x1<<24)|(0x5<<16))
+#define GFX_OP_3DSTATE_DX9_CONSTANTF_VS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x39<<16))
+#define GFX_OP_3DSTATE_DX9_CONSTANTF_PS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x3A<<16))
+#define GFX_OP_3DSTATE_SO_DECL_LIST \
+	((0x3<<29)|(0x3<<27)|(0x1<<24)|(0x17<<16))
+
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_VS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x43<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_GS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x44<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_HS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x45<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_DS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x46<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_PS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x47<<16))
+
+#define MFX_WAIT  ((0x3<<29)|(0x1<<27)|(0x0<<16))
+
+#define COLOR_BLT     ((0x2<<29)|(0x40<<22))
+#define SRC_COPY_BLT  ((0x2<<29)|(0x43<<22))
+
+/*
+ * Registers used only by the command parser
+ */
+#define BCS_SWCTRL 0x22200
+
+#define HS_INVOCATION_COUNT 0x2300
+#define DS_INVOCATION_COUNT 0x2308
+#define IA_VERTICES_COUNT   0x2310
+#define IA_PRIMITIVES_COUNT 0x2318
+#define VS_INVOCATION_COUNT 0x2320
+#define GS_INVOCATION_COUNT 0x2328
+#define GS_PRIMITIVES_COUNT 0x2330
+#define CL_INVOCATION_COUNT 0x2338
+#define CL_PRIMITIVES_COUNT 0x2340
+#define PS_INVOCATION_COUNT 0x2348
+#define PS_DEPTH_COUNT      0x2350
+
+/* There are the 4 64-bit counter registers, one for each stream output */
+#define GEN7_SO_NUM_PRIMS_WRITTEN(n) (0x5200 + (n) * 8)
+
+#define GEN7_SO_PRIM_STORAGE_NEEDED(n)  (0x5240 + (n) * 8)
+
+#define GEN7_3DPRIM_END_OFFSET          0x2420
+#define GEN7_3DPRIM_START_VERTEX        0x2430
+#define GEN7_3DPRIM_VERTEX_COUNT        0x2434
+#define GEN7_3DPRIM_INSTANCE_COUNT      0x2438
+#define GEN7_3DPRIM_START_INSTANCE      0x243C
+#define GEN7_3DPRIM_BASE_VERTEX         0x2440
+
+#define OACONTROL 0x2360
+
+#define _GEN7_PIPEA_DE_LOAD_SL	0x70068
+#define _GEN7_PIPEB_DE_LOAD_SL	0x71068
+#define GEN7_PIPE_DE_LOAD_SL(pipe) _PIPE(pipe, \
+					 _GEN7_PIPEA_DE_LOAD_SL, \
+					 _GEN7_PIPEB_DE_LOAD_SL)
 
 /*
  * Reset registers
@@ -370,6 +478,7 @@
 #define   IOSF_PORT_PUNIT			0x4
 #define   IOSF_PORT_NC				0x11
 #define   IOSF_PORT_DPIO			0x12
+#define   IOSF_PORT_DPIO_2			0x1a
 #define   IOSF_PORT_GPIO_NC			0x13
 #define   IOSF_PORT_CCK				0x14
 #define   IOSF_PORT_CCU				0xA9
@@ -381,9 +490,6 @@
 /* See configdb bunit SB addr map */
 #define BUNIT_REG_BISOC				0x11
 
-#define PUNIT_OPCODE_REG_READ			6
-#define PUNIT_OPCODE_REG_WRITE			7
-
 #define PUNIT_REG_DSPFREQ			0x36
 #define   DSPFREQSTAT_SHIFT			30
 #define   DSPFREQSTAT_MASK			(0x3 << DSPFREQSTAT_SHIFT)
@@ -469,16 +575,91 @@ enum punit_power_well {
 #define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
 #define CCK_DISPLAY_CLOCK_CONTROL		0x6b
 
-/*
- * DPIO - a special bus for various display related registers to hide behind
+/**
+ * DOC: DPIO
+ *
+ * VLV and CHV have slightly peculiar display PHYs for driving DP/HDMI
+ * ports. DPIO is the name given to such a display PHY. These PHYs
+ * don't follow the standard programming model using direct MMIO
+ * registers, and instead their registers must be accessed trough IOSF
+ * sideband. VLV has one such PHY for driving ports B and C, and CHV
+ * adds another PHY for driving port D. Each PHY responds to specific
+ * IOSF-SB port.
+ *
+ * Each display PHY is made up of one or two channels. Each channel
+ * houses a common lane part which contains the PLL and other common
+ * logic. CH0 common lane also contains the IOSF-SB logic for the
+ * Common Register Interface (CRI) ie. the DPIO registers. CRI clock
+ * must be running when any DPIO registers are accessed.
+ *
+ * In addition to having their own registers, the PHYs are also
+ * controlled through some dedicated signals from the display
+ * controller. These include PLL reference clock enable, PLL enable,
+ * and CRI clock selection, for example.
  *
- * DPIO is VLV only.
+ * Eeach channel also has two splines (also called data lanes), and
+ * each spline is made up of one Physical Access Coding Sub-Layer
+ * (PCS) block and two TX lanes. So each channel has two PCS blocks
+ * and four TX lanes. The TX lanes are used as DP lanes or TMDS
+ * data/clock pairs depending on the output type.
+ *
+ * Additionally the PHY also contains an AUX lane with AUX blocks
+ * for each channel. This is used for DP AUX communication, but
+ * this fact isn't really relevant for the driver since AUX is
+ * controlled from the display controller side. No DPIO registers
+ * need to be accessed during AUX communication,
+ *
+ * Generally the common lane corresponds to the pipe and
+ * the spline (PCS/TX) correponds to the port.
+ *
+ * For dual channel PHY (VLV/CHV):
+ *
+ *  pipe A == CMN/PLL/REF CH0
+ *
+ *  pipe B == CMN/PLL/REF CH1
+ *
+ *  port B == PCS/TX CH0
+ *
+ *  port C == PCS/TX CH1
+ *
+ * This is especially important when we cross the streams
+ * ie. drive port B with pipe B, or port C with pipe A.
+ *
+ * For single channel PHY (CHV):
+ *
+ *  pipe C == CMN/PLL/REF CH0
+ *
+ *  port D == PCS/TX CH0
+ *
+ * Note: digital port B is DDI0, digital port C is DDI1,
+ * digital port D is DDI2
+ */
+/*
+ * Dual channel PHY (VLV/CHV)
+ * ---------------------------------
+ * |      CH0      |      CH1      |
+ * |  CMN/PLL/REF  |  CMN/PLL/REF  |
+ * |---------------|---------------| Display PHY
+ * | PCS01 | PCS23 | PCS01 | PCS23 |
+ * |-------|-------|-------|-------|
+ * |TX0|TX1|TX2|TX3|TX0|TX1|TX2|TX3|
+ * ---------------------------------
+ * |     DDI0      |     DDI1      | DP/HDMI ports
+ * ---------------------------------
  *
- * Note: digital port B is DDI0, digital pot C is DDI1
+ * Single channel PHY (CHV)
+ * -----------------
+ * |      CH0      |
+ * |  CMN/PLL/REF  |
+ * |---------------| Display PHY
+ * | PCS01 | PCS23 |
+ * |-------|-------|
+ * |TX0|TX1|TX2|TX3|
+ * -----------------
+ * |     DDI2      | DP/HDMI port
+ * -----------------
  */
 #define DPIO_DEVFN			0
-#define DPIO_OPCODE_REG_WRITE		1
-#define DPIO_OPCODE_REG_READ		0
 
 #define DPIO_CTL			(VLV_DISPLAY_BASE + 0x2110)
 #define  DPIO_MODSEL1			(1<<3) /* if ref clk b == 27 */
@@ -555,14 +736,29 @@ enum punit_power_well {
 #define   DPIO_PCS_TX_LANE1_RESET	(1<<7)
 #define VLV_PCS_DW0(ch) _PORT(ch, _VLV_PCS_DW0_CH0, _VLV_PCS_DW0_CH1)
 
+#define _VLV_PCS01_DW0_CH0		0x200
+#define _VLV_PCS23_DW0_CH0		0x400
+#define _VLV_PCS01_DW0_CH1		0x2600
+#define _VLV_PCS23_DW0_CH1		0x2800
+#define VLV_PCS01_DW0(ch) _PORT(ch, _VLV_PCS01_DW0_CH0, _VLV_PCS01_DW0_CH1)
+#define VLV_PCS23_DW0(ch) _PORT(ch, _VLV_PCS23_DW0_CH0, _VLV_PCS23_DW0_CH1)
+
 #define _VLV_PCS_DW1_CH0		0x8204
 #define _VLV_PCS_DW1_CH1		0x8404
+#define   CHV_PCS_REQ_SOFTRESET_EN	(1<<23)
 #define   DPIO_PCS_CLK_CRI_RXEB_EIOS_EN	(1<<22)
 #define   DPIO_PCS_CLK_CRI_RXDIGFILTSG_EN (1<<21)
 #define   DPIO_PCS_CLK_DATAWIDTH_SHIFT	(6)
 #define   DPIO_PCS_CLK_SOFT_RESET	(1<<5)
 #define VLV_PCS_DW1(ch) _PORT(ch, _VLV_PCS_DW1_CH0, _VLV_PCS_DW1_CH1)
 
+#define _VLV_PCS01_DW1_CH0		0x204
+#define _VLV_PCS23_DW1_CH0		0x404
+#define _VLV_PCS01_DW1_CH1		0x2604
+#define _VLV_PCS23_DW1_CH1		0x2804
+#define VLV_PCS01_DW1(ch) _PORT(ch, _VLV_PCS01_DW1_CH0, _VLV_PCS01_DW1_CH1)
+#define VLV_PCS23_DW1(ch) _PORT(ch, _VLV_PCS23_DW1_CH0, _VLV_PCS23_DW1_CH1)
+
 #define _VLV_PCS_DW8_CH0		0x8220
 #define _VLV_PCS_DW8_CH1		0x8420
 #define VLV_PCS_DW8(ch) _PORT(ch, _VLV_PCS_DW8_CH0, _VLV_PCS_DW8_CH1)
@@ -578,6 +774,19 @@ enum punit_power_well {
 #define _VLV_PCS_DW9_CH1		0x8424
 #define	VLV_PCS_DW9(ch) _PORT(ch, _VLV_PCS_DW9_CH0, _VLV_PCS_DW9_CH1)
 
+#define _CHV_PCS_DW10_CH0		0x8228
+#define _CHV_PCS_DW10_CH1		0x8428
+#define   DPIO_PCS_SWING_CALC_TX0_TX2	(1<<30)
+#define   DPIO_PCS_SWING_CALC_TX1_TX3	(1<<31)
+#define CHV_PCS_DW10(ch) _PORT(ch, _CHV_PCS_DW10_CH0, _CHV_PCS_DW10_CH1)
+
+#define _VLV_PCS01_DW10_CH0		0x0228
+#define _VLV_PCS23_DW10_CH0		0x0428
+#define _VLV_PCS01_DW10_CH1		0x2628
+#define _VLV_PCS23_DW10_CH1		0x2828
+#define VLV_PCS01_DW10(port) _PORT(port, _VLV_PCS01_DW10_CH0, _VLV_PCS01_DW10_CH1)
+#define VLV_PCS23_DW10(port) _PORT(port, _VLV_PCS23_DW10_CH0, _VLV_PCS23_DW10_CH1)
+
 #define _VLV_PCS_DW11_CH0		0x822c
 #define _VLV_PCS_DW11_CH1		0x842c
 #define VLV_PCS_DW11(ch) _PORT(ch, _VLV_PCS_DW11_CH0, _VLV_PCS_DW11_CH1)
@@ -596,14 +805,21 @@ enum punit_power_well {
 
 #define _VLV_TX_DW2_CH0			0x8288
 #define _VLV_TX_DW2_CH1			0x8488
+#define   DPIO_SWING_MARGIN_SHIFT	16
+#define   DPIO_SWING_MARGIN_MASK	(0xff << DPIO_SWING_MARGIN_SHIFT)
+#define   DPIO_UNIQ_TRANS_SCALE_SHIFT	8
 #define VLV_TX_DW2(ch) _PORT(ch, _VLV_TX_DW2_CH0, _VLV_TX_DW2_CH1)
 
 #define _VLV_TX_DW3_CH0			0x828c
 #define _VLV_TX_DW3_CH1			0x848c
+/* The following bit for CHV phy */
+#define   DPIO_TX_UNIQ_TRANS_SCALE_EN	(1<<27)
 #define VLV_TX_DW3(ch) _PORT(ch, _VLV_TX_DW3_CH0, _VLV_TX_DW3_CH1)
 
 #define _VLV_TX_DW4_CH0			0x8290
 #define _VLV_TX_DW4_CH1			0x8490
+#define   DPIO_SWING_DEEMPH9P5_SHIFT	24
+#define   DPIO_SWING_DEEMPH9P5_MASK	(0xff << DPIO_SWING_DEEMPH9P5_SHIFT)
 #define VLV_TX_DW4(ch) _PORT(ch, _VLV_TX_DW4_CH0, _VLV_TX_DW4_CH1)
 
 #define _VLV_TX3_DW4_CH0		0x690
@@ -623,6 +839,73 @@ enum punit_power_well {
 #define _VLV_TX_DW14_CH1		0x84b8
 #define VLV_TX_DW14(ch) _PORT(ch, _VLV_TX_DW14_CH0, _VLV_TX_DW14_CH1)
 
+/* CHV dpPhy registers */
+#define _CHV_PLL_DW0_CH0		0x8000
+#define _CHV_PLL_DW0_CH1		0x8180
+#define CHV_PLL_DW0(ch) _PIPE(ch, _CHV_PLL_DW0_CH0, _CHV_PLL_DW0_CH1)
+
+#define _CHV_PLL_DW1_CH0		0x8004
+#define _CHV_PLL_DW1_CH1		0x8184
+#define   DPIO_CHV_N_DIV_SHIFT		8
+#define   DPIO_CHV_M1_DIV_BY_2		(0 << 0)
+#define CHV_PLL_DW1(ch) _PIPE(ch, _CHV_PLL_DW1_CH0, _CHV_PLL_DW1_CH1)
+
+#define _CHV_PLL_DW2_CH0		0x8008
+#define _CHV_PLL_DW2_CH1		0x8188
+#define CHV_PLL_DW2(ch) _PIPE(ch, _CHV_PLL_DW2_CH0, _CHV_PLL_DW2_CH1)
+
+#define _CHV_PLL_DW3_CH0		0x800c
+#define _CHV_PLL_DW3_CH1		0x818c
+#define  DPIO_CHV_FRAC_DIV_EN		(1 << 16)
+#define  DPIO_CHV_FIRST_MOD		(0 << 8)
+#define  DPIO_CHV_SECOND_MOD		(1 << 8)
+#define  DPIO_CHV_FEEDFWD_GAIN_SHIFT	0
+#define CHV_PLL_DW3(ch) _PIPE(ch, _CHV_PLL_DW3_CH0, _CHV_PLL_DW3_CH1)
+
+#define _CHV_PLL_DW6_CH0		0x8018
+#define _CHV_PLL_DW6_CH1		0x8198
+#define   DPIO_CHV_GAIN_CTRL_SHIFT	16
+#define	  DPIO_CHV_INT_COEFF_SHIFT	8
+#define   DPIO_CHV_PROP_COEFF_SHIFT	0
+#define CHV_PLL_DW6(ch) _PIPE(ch, _CHV_PLL_DW6_CH0, _CHV_PLL_DW6_CH1)
+
+#define _CHV_CMN_DW13_CH0		0x8134
+#define _CHV_CMN_DW0_CH1		0x8080
+#define   DPIO_CHV_S1_DIV_SHIFT		21
+#define   DPIO_CHV_P1_DIV_SHIFT		13 /* 3 bits */
+#define   DPIO_CHV_P2_DIV_SHIFT		8  /* 5 bits */
+#define   DPIO_CHV_K_DIV_SHIFT		4
+#define   DPIO_PLL_FREQLOCK		(1 << 1)
+#define   DPIO_PLL_LOCK			(1 << 0)
+#define CHV_CMN_DW13(ch) _PIPE(ch, _CHV_CMN_DW13_CH0, _CHV_CMN_DW0_CH1)
+
+#define _CHV_CMN_DW14_CH0		0x8138
+#define _CHV_CMN_DW1_CH1		0x8084
+#define   DPIO_AFC_RECAL		(1 << 14)
+#define   DPIO_DCLKP_EN			(1 << 13)
+#define CHV_CMN_DW14(ch) _PIPE(ch, _CHV_CMN_DW14_CH0, _CHV_CMN_DW1_CH1)
+
+#define CHV_CMN_DW30			0x8178
+#define   DPIO_LRC_BYPASS		(1 << 3)
+
+#define _TXLANE(ch, lane, offset) ((ch ? 0x2400 : 0) + \
+					(lane) * 0x200 + (offset))
+
+#define CHV_TX_DW0(ch, lane) _TXLANE(ch, lane, 0x80)
+#define CHV_TX_DW1(ch, lane) _TXLANE(ch, lane, 0x84)
+#define CHV_TX_DW2(ch, lane) _TXLANE(ch, lane, 0x88)
+#define CHV_TX_DW3(ch, lane) _TXLANE(ch, lane, 0x8c)
+#define CHV_TX_DW4(ch, lane) _TXLANE(ch, lane, 0x90)
+#define CHV_TX_DW5(ch, lane) _TXLANE(ch, lane, 0x94)
+#define CHV_TX_DW6(ch, lane) _TXLANE(ch, lane, 0x98)
+#define CHV_TX_DW7(ch, lane) _TXLANE(ch, lane, 0x9c)
+#define CHV_TX_DW8(ch, lane) _TXLANE(ch, lane, 0xa0)
+#define CHV_TX_DW9(ch, lane) _TXLANE(ch, lane, 0xa4)
+#define CHV_TX_DW10(ch, lane) _TXLANE(ch, lane, 0xa8)
+#define CHV_TX_DW11(ch, lane) _TXLANE(ch, lane, 0xac)
+#define   DPIO_FRC_LATENCY_SHFIT	8
+#define CHV_TX_DW14(ch, lane) _TXLANE(ch, lane, 0xb8)
+#define   DPIO_UPAR_SHIFT		30
 /*
  * Fence registers
  */
@@ -659,10 +942,14 @@ enum punit_power_well {
 /*
  * Instruction and interrupt control regs
  */
+#define PGTBL_CTL	0x02020
+#define   PGTBL_ADDRESS_LO_MASK	0xfffff000 /* bits [31:12] */
+#define   PGTBL_ADDRESS_HI_MASK	0x000000f0 /* bits [35:32] (gen4) */
 #define PGTBL_ER	0x02024
 #define RENDER_RING_BASE	0x02000
 #define BSD_RING_BASE		0x04000
 #define GEN6_BSD_RING_BASE	0x12000
+#define GEN8_BSD2_RING_BASE	0x1c000
 #define VEBOX_RING_BASE		0x1a000
 #define BLT_RING_BASE		0x22000
 #define RING_TAIL(base)		((base)+0x30)
@@ -688,9 +975,20 @@ enum punit_power_well {
 #define RING_MAX_IDLE(base)	((base)+0x54)
 #define RING_HWS_PGA(base)	((base)+0x80)
 #define RING_HWS_PGA_GEN6(base)	((base)+0x2080)
-#define ARB_MODE		0x04030
+
+#define GEN7_WR_WATERMARK	0x4028
+#define GEN7_GFX_PRIO_CTRL	0x402C
+#define ARB_MODE		0x4030
 #define   ARB_MODE_SWIZZLE_SNB	(1<<4)
 #define   ARB_MODE_SWIZZLE_IVB	(1<<5)
+#define GEN7_GFX_PEND_TLB0	0x4034
+#define GEN7_GFX_PEND_TLB1	0x4038
+/* L3, CVS, ZTLB, RCC, CASC LRA min, max values */
+#define GEN7_LRA_LIMITS_BASE	0x403C
+#define GEN7_LRA_LIMITS_REG_NUM	13
+#define GEN7_MEDIA_MAX_REQ_COUNT	0x4070
+#define GEN7_GFX_MAX_REQ_COUNT		0x4074
+
 #define GAMTARBMODE		0x04a08
 #define   ARB_MODE_BWGTLB_DISABLE (1<<9)
 #define   ARB_MODE_SWIZZLE_BDW	(1<<1)
@@ -725,6 +1023,9 @@ enum punit_power_well {
 #define   RING_WAIT_I8XX	(1<<0) /* gen2, PRBx_HEAD */
 #define   RING_WAIT		(1<<11) /* gen3+, PRBx_CTL */
 #define   RING_WAIT_SEMAPHORE	(1<<10) /* gen6+ */
+
+#define GEN7_TLB_RD_ADDR	0x4700
+
 #if 0
 #define PRB0_TAIL	0x02030
 #define PRB0_HEAD	0x02034
@@ -748,6 +1049,7 @@ enum punit_power_well {
 #define RING_INSTDONE(base)	((base)+0x6c)
 #define RING_INSTPS(base)	((base)+0x70)
 #define RING_DMA_FADD(base)	((base)+0x78)
+#define RING_DMA_FADD_UDW(base)	((base)+0x60) /* gen8+ */
 #define RING_INSTPM(base)	((base)+0xc0)
 #define RING_MI_MODE(base)	((base)+0x9c)
 #define INSTPS		0x02070 /* 965+ only */
@@ -842,21 +1144,26 @@ enum punit_power_well {
 #define GFX_MODE_GEN7	0x0229c
 #define RING_MODE_GEN7(ring)	((ring)->mmio_base+0x29c)
 #define   GFX_RUN_LIST_ENABLE		(1<<15)
-#define   GFX_TLB_INVALIDATE_ALWAYS	(1<<13)
+#define   GFX_TLB_INVALIDATE_EXPLICIT	(1<<13)
 #define   GFX_SURFACE_FAULT_ENABLE	(1<<12)
 #define   GFX_REPLAY_MODE		(1<<11)
 #define   GFX_PSMI_GRANULARITY		(1<<10)
 #define   GFX_PPGTT_ENABLE		(1<<9)
 
 #define VLV_DISPLAY_BASE 0x180000
+#define VLV_MIPI_BASE VLV_DISPLAY_BASE
 
+#define VLV_GU_CTL0	(VLV_DISPLAY_BASE + 0x2030)
+#define VLV_GU_CTL1	(VLV_DISPLAY_BASE + 0x2034)
 #define SCPD0		0x0209c /* 915+ only */
 #define IER		0x020a0
 #define IIR		0x020a4
 #define IMR		0x020a8
 #define ISR		0x020ac
 #define VLV_GUNIT_CLOCK_GATE	(VLV_DISPLAY_BASE + 0x2060)
+#define   GINT_DIS		(1<<22)
 #define   GCFG_DIS		(1<<8)
+#define VLV_GUNIT_CLOCK_GATE2	(VLV_DISPLAY_BASE + 0x2064)
 #define VLV_IIR_RW	(VLV_DISPLAY_BASE + 0x2084)
 #define VLV_IER		(VLV_DISPLAY_BASE + 0x20a0)
 #define VLV_IIR		(VLV_DISPLAY_BASE + 0x20a4)
@@ -875,7 +1182,7 @@ enum punit_power_well {
 #define   I915_ERROR_INSTRUCTION			(1<<0)
 #define INSTPM	        0x020c0
 #define   INSTPM_SELF_EN (1<<12) /* 915GM only */
-#define   INSTPM_AGPBUSY_DIS (1<<11) /* gen3: when disabled, pending interrupts
+#define   INSTPM_AGPBUSY_INT_EN (1<<11) /* gen3: when disabled, pending interrupts
 					will not assert AGPBUSY# and will only
 					be delivered when out of C3. */
 #define   INSTPM_FORCE_ORDERING				(1<<7) /* GEN6+ */
@@ -956,6 +1263,10 @@ enum punit_power_well {
 #define   MI_ARB_DISPLAY_PRIORITY_A_B		(0 << 0)	/* display A > display B */
 #define   MI_ARB_DISPLAY_PRIORITY_B_A		(1 << 0)	/* display B > display A */
 
+#define MI_STATE	0x020e4 /* gen2 only */
+#define   MI_AGPBUSY_INT_EN			(1 << 1) /* 85x only */
+#define   MI_AGPBUSY_830_MODE			(1 << 0) /* 85x only */
+
 #define CACHE_MODE_0	0x02120 /* 915+ only */
 #define   CM0_PIPELINED_RENDER_FLUSH_DISABLE (1<<8)
 #define   CM0_IZ_OPT_DISABLE      (1<<6)
@@ -973,6 +1284,7 @@ enum punit_power_well {
 #define   ECO_FLIP_DONE		(1<<0)
 
 #define CACHE_MODE_0_GEN7	0x7000 /* IVB+ */
+#define RC_OP_FLUSH_ENABLE (1<<0)
 #define   HIZ_RAW_STALL_OPT_DISABLE (1<<2)
 #define CACHE_MODE_1		0x7004 /* IVB+ */
 #define   PIXEL_SUBSPAN_COLLECT_OPT_DISABLE	(1<<6)
@@ -984,6 +1296,7 @@ enum punit_power_well {
 
 #define GEN6_RC_SLEEP_PSMI_CONTROL	0x2050
 #define   GEN8_RC_SEMA_IDLE_MSG_DISABLE	(1 << 12)
+#define   GEN8_FF_DOP_CLOCK_GATE_DISABLE	(1<<10)
 
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	0x12050
 #define   GEN6_BSD_SLEEP_MSG_DISABLE	(1 << 0)
@@ -1024,24 +1337,43 @@ enum punit_power_well {
 
 /* These are all the "old" interrupts */
 #define ILK_BSD_USER_INTERRUPT				(1<<5)
+
+#define I915_PM_INTERRUPT				(1<<31)
+#define I915_ISP_INTERRUPT				(1<<22)
+#define I915_LPE_PIPE_B_INTERRUPT			(1<<21)
+#define I915_LPE_PIPE_A_INTERRUPT			(1<<20)
+#define I915_MIPIB_INTERRUPT				(1<<19)
+#define I915_MIPIA_INTERRUPT				(1<<18)
 #define I915_PIPE_CONTROL_NOTIFY_INTERRUPT		(1<<18)
 #define I915_DISPLAY_PORT_INTERRUPT			(1<<17)
+#define I915_DISPLAY_PIPE_C_HBLANK_INTERRUPT		(1<<16)
+#define I915_MASTER_ERROR_INTERRUPT			(1<<15)
 #define I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT	(1<<15)
+#define I915_DISPLAY_PIPE_B_HBLANK_INTERRUPT		(1<<14)
 #define I915_GMCH_THERMAL_SENSOR_EVENT_INTERRUPT	(1<<14) /* p-state */
+#define I915_DISPLAY_PIPE_A_HBLANK_INTERRUPT		(1<<13)
 #define I915_HWB_OOM_INTERRUPT				(1<<13)
+#define I915_LPE_PIPE_C_INTERRUPT			(1<<12)
 #define I915_SYNC_STATUS_INTERRUPT			(1<<12)
+#define I915_MISC_INTERRUPT				(1<<11)
 #define I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT	(1<<11)
+#define I915_DISPLAY_PIPE_C_VBLANK_INTERRUPT		(1<<10)
 #define I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT	(1<<10)
+#define I915_DISPLAY_PIPE_C_EVENT_INTERRUPT		(1<<9)
 #define I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT	(1<<9)
+#define I915_DISPLAY_PIPE_C_DPBM_INTERRUPT		(1<<8)
 #define I915_DISPLAY_PLANE_C_FLIP_PENDING_INTERRUPT	(1<<8)
 #define I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT		(1<<7)
 #define I915_DISPLAY_PIPE_A_EVENT_INTERRUPT		(1<<6)
 #define I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT		(1<<5)
 #define I915_DISPLAY_PIPE_B_EVENT_INTERRUPT		(1<<4)
+#define I915_DISPLAY_PIPE_A_DPBM_INTERRUPT		(1<<3)
+#define I915_DISPLAY_PIPE_B_DPBM_INTERRUPT		(1<<2)
 #define I915_DEBUG_INTERRUPT				(1<<2)
+#define I915_WINVALID_INTERRUPT				(1<<1)
 #define I915_USER_INTERRUPT				(1<<1)
 #define I915_ASLE_INTERRUPT				(1<<0)
-#define I915_BSD_USER_INTERRUPT				(1 << 25)
+#define I915_BSD_USER_INTERRUPT				(1<<25)
 
 #define GEN6_BSD_RNCID			0x12198
 
@@ -1198,6 +1530,7 @@ enum punit_power_well {
 #define   GMBUS_PORT_SSC	1
 #define   GMBUS_PORT_VGADDC	2
 #define   GMBUS_PORT_PANEL	3
+#define   GMBUS_PORT_DPD_CHV	3 /* HDMID_CHV */
 #define   GMBUS_PORT_DPC	4 /* HDMIC */
 #define   GMBUS_PORT_DPB	5 /* SDVO, HDMIB */
 #define   GMBUS_PORT_DPD	6 /* HDMID */
@@ -1239,6 +1572,7 @@ enum punit_power_well {
  */
 #define DPLL_A_OFFSET 0x6014
 #define DPLL_B_OFFSET 0x6018
+#define CHV_DPLL_C_OFFSET 0x6030
 #define DPLL(pipe) (dev_priv->info.dpll_offsets[pipe] + \
 		    dev_priv->info.display_mmio_offset)
 
@@ -1273,10 +1607,23 @@ enum punit_power_well {
 #define   DPLL_LOCK_VLV			(1<<15)
 #define   DPLL_INTEGRATED_CRI_CLK_VLV	(1<<14)
 #define   DPLL_INTEGRATED_CLOCK_VLV	(1<<13)
+#define   DPLL_SSC_REF_CLOCK_CHV	(1<<13)
 #define   DPLL_PORTC_READY_MASK		(0xf << 4)
 #define   DPLL_PORTB_READY_MASK		(0xf)
 
 #define   DPLL_FPA01_P1_POST_DIV_MASK_I830	0x001f0000
+
+/* Additional CHV pll/phy registers */
+#define DPIO_PHY_STATUS			(VLV_DISPLAY_BASE + 0x6240)
+#define   DPLL_PORTD_READY_MASK		(0xf)
+#define DISPLAY_PHY_CONTROL (VLV_DISPLAY_BASE + 0x60100)
+#define   PHY_COM_LANE_RESET_DEASSERT(phy, val) \
+				((phy == DPIO_PHY0) ? (val | 1) : (val | 2))
+#define   PHY_COM_LANE_RESET_ASSERT(phy, val) \
+				((phy == DPIO_PHY0) ? (val & ~1) : (val & ~2))
+#define DISPLAY_PHY_STATUS (VLV_DISPLAY_BASE + 0x60104)
+#define   PHY_POWERGOOD(phy)	((phy == DPIO_PHY0) ? (1<<31) : (1<<30))
+
 /*
  * The i830 generation, in LVDS mode, defines P1 as the bit number set within
  * this field (only one bit may be set).
@@ -1317,6 +1664,7 @@ enum punit_power_well {
 
 #define DPLL_A_MD_OFFSET 0x601c /* 965+ only */
 #define DPLL_B_MD_OFFSET 0x6020 /* 965+ only */
+#define CHV_DPLL_C_MD_OFFSET 0x603c
 #define DPLL_MD(pipe) (dev_priv->info.dpll_md_offsets[pipe] + \
 		       dev_priv->info.display_mmio_offset)
 
@@ -1416,7 +1764,7 @@ enum punit_power_well {
 # define DPIOUNIT_CLOCK_GATE_DISABLE		(1 << 6) /* 915-945 */
 # define OVFUNIT_CLOCK_GATE_DISABLE		(1 << 5)
 # define OVBUNIT_CLOCK_GATE_DISABLE		(1 << 4)
-/**
+/*
  * This bit must be set on the 830 to prevent hangs when turning off the
  * overlay scaler.
  */
@@ -1436,12 +1784,12 @@ enum punit_power_well {
 # define COLOR_CALCULATOR_CLOCK_GATE_DISABLE	(1 << 7)
 # define MOTION_COMP_CLOCK_GATE_DISABLE		(1 << 6)
 # define MAG_CLOCK_GATE_DISABLE			(1 << 5)
-/** This bit must be unset on 855,865 */
+/* This bit must be unset on 855,865 */
 # define MECI_CLOCK_GATE_DISABLE		(1 << 4)
 # define DCMP_CLOCK_GATE_DISABLE		(1 << 3)
 # define MEC_CLOCK_GATE_DISABLE			(1 << 2)
 # define MECO_CLOCK_GATE_DISABLE		(1 << 1)
-/** This bit must be set on 855,865. */
+/* This bit must be set on 855,865. */
 # define SV_CLOCK_GATE_DISABLE			(1 << 0)
 # define I915_MPEG_CLOCK_GATE_DISABLE		(1 << 16)
 # define I915_VLD_IP_PR_CLOCK_GATE_DISABLE	(1 << 15)
@@ -1462,14 +1810,14 @@ enum punit_power_well {
 # define I915_BY_CLOCK_GATE_DISABLE		(1 << 0)
 
 # define I965_RCZ_CLOCK_GATE_DISABLE		(1 << 30)
-/** This bit must always be set on 965G/965GM */
+/* This bit must always be set on 965G/965GM */
 # define I965_RCC_CLOCK_GATE_DISABLE		(1 << 29)
 # define I965_RCPB_CLOCK_GATE_DISABLE		(1 << 28)
 # define I965_DAP_CLOCK_GATE_DISABLE		(1 << 27)
 # define I965_ROC_CLOCK_GATE_DISABLE		(1 << 26)
 # define I965_GW_CLOCK_GATE_DISABLE		(1 << 25)
 # define I965_TD_CLOCK_GATE_DISABLE		(1 << 24)
-/** This bit must always be set on 965G */
+/* This bit must always be set on 965G */
 # define I965_ISC_CLOCK_GATE_DISABLE		(1 << 23)
 # define I965_IC_CLOCK_GATE_DISABLE		(1 << 22)
 # define I965_EU_CLOCK_GATE_DISABLE		(1 << 21)
@@ -1494,6 +1842,10 @@ enum punit_power_well {
 #define VF_UNIT_CLOCK_GATE_DISABLE		(1 << 9)
 #define GS_UNIT_CLOCK_GATE_DISABLE		(1 << 7)
 #define CL_UNIT_CLOCK_GATE_DISABLE		(1 << 6)
+
+#define VDECCLK_GATE_D		0x620C		/* g4x only */
+#define  VCP_UNIT_CLOCK_GATE_DISABLE		(1 << 4)
+
 #define RAMCLK_GATE_D		0x6210		/* CRL only */
 #define DEUC			0x6214          /* CRL only */
 
@@ -1513,6 +1865,7 @@ enum punit_power_well {
  */
 #define PALETTE_A_OFFSET 0xa000
 #define PALETTE_B_OFFSET 0xa800
+#define CHV_PALETTE_C_OFFSET 0xc000
 #define PALETTE(pipe) (dev_priv->info.palette_offsets[pipe] + \
 		       dev_priv->info.display_mmio_offset)
 
@@ -1535,7 +1888,7 @@ enum punit_power_well {
 /* Memory controller frequency in MCHBAR for Haswell (possible SNB+) */
 #define DCLK (MCHBAR_MIRROR_BASE_SNB + 0x5e04)
 
-/** 915-945 and GM965 MCH register controlling DRAM channel access */
+/* 915-945 and GM965 MCH register controlling DRAM channel access */
 #define DCC			0x10200
 #define DCC_ADDRESSING_MODE_SINGLE_CHANNEL		(0 << 0)
 #define DCC_ADDRESSING_MODE_DUAL_CHANNEL_ASYMMETRIC	(1 << 0)
@@ -1544,15 +1897,15 @@ enum punit_power_well {
 #define DCC_CHANNEL_XOR_DISABLE				(1 << 10)
 #define DCC_CHANNEL_XOR_BIT_17				(1 << 9)
 
-/** Pineview MCH register contains DDR3 setting */
+/* Pineview MCH register contains DDR3 setting */
 #define CSHRDDR3CTL            0x101a8
 #define CSHRDDR3CTL_DDR3       (1 << 2)
 
-/** 965 MCH register controlling DRAM channel configuration */
+/* 965 MCH register controlling DRAM channel configuration */
 #define C0DRB3			0x10206
 #define C1DRB3			0x10606
 
-/** snb MCH registers for reading the DRAM channel configuration */
+/* snb MCH registers for reading the DRAM channel configuration */
 #define MAD_DIMM_C0			(MCHBAR_MIRROR_BASE_SNB + 0x5004)
 #define MAD_DIMM_C1			(MCHBAR_MIRROR_BASE_SNB + 0x5008)
 #define MAD_DIMM_C2			(MCHBAR_MIRROR_BASE_SNB + 0x500C)
@@ -1574,7 +1927,7 @@ enum punit_power_well {
 #define   MAD_DIMM_A_SIZE_SHIFT		0
 #define   MAD_DIMM_A_SIZE_MASK		(0xff << MAD_DIMM_A_SIZE_SHIFT)
 
-/** snb MCH registers for priority tuning */
+/* snb MCH registers for priority tuning */
 #define MCH_SSKPD			(MCHBAR_MIRROR_BASE_SNB + 0x5d10)
 #define   MCH_SSKPD_WM0_MASK		0x3f
 #define   MCH_SSKPD_WM0_VAL		0xc
@@ -2002,6 +2355,7 @@ enum punit_power_well {
 #define TRANSCODER_A_OFFSET 0x60000
 #define TRANSCODER_B_OFFSET 0x61000
 #define TRANSCODER_C_OFFSET 0x62000
+#define CHV_TRANSCODER_C_OFFSET 0x63000
 #define TRANSCODER_EDP_OFFSET 0x6f000
 
 #define _TRANSCODER2(pipe, reg) (dev_priv->info.trans_offsets[(pipe)] - \
@@ -2226,6 +2580,7 @@ enum punit_power_well {
 #define GEN3_SDVOC	0x61160
 #define GEN4_HDMIB	GEN3_SDVOB
 #define GEN4_HDMIC	GEN3_SDVOC
+#define CHV_HDMID	0x6116C
 #define PCH_SDVOB	0xe1140
 #define PCH_HDMIB	PCH_SDVOB
 #define PCH_HDMIC	0xe1150
@@ -2246,7 +2601,7 @@ enum punit_power_well {
 #define   SDVO_PIPE_B_SELECT			(1 << 30)
 #define   SDVO_STALL_SELECT			(1 << 29)
 #define   SDVO_INTERRUPT_ENABLE			(1 << 26)
-/**
+/*
  * 915G/GM SDVO pixel multiplier.
  * Programmed value is multiplier - 1, up to 5x.
  * \sa DPLL_MD_UDI_MULTIPLIER_MASK
@@ -2286,6 +2641,10 @@ enum punit_power_well {
 #define   SDVO_PIPE_SEL_CPT(pipe)		((pipe) << 29)
 #define   SDVO_PIPE_SEL_MASK_CPT		(3 << 29)
 
+/* CHV SDVO/HDMI bits: */
+#define   SDVO_PIPE_SEL_CHV(pipe)		((pipe) << 24)
+#define   SDVO_PIPE_SEL_MASK_CHV		(3 << 24)
+
 
 /* DVO port control */
 #define DVOA			0x61120
@@ -2556,65 +2915,65 @@ enum punit_power_well {
 
 /* TV port control */
 #define TV_CTL			0x68000
-/** Enables the TV encoder */
+/* Enables the TV encoder */
 # define TV_ENC_ENABLE			(1 << 31)
-/** Sources the TV encoder input from pipe B instead of A. */
+/* Sources the TV encoder input from pipe B instead of A. */
 # define TV_ENC_PIPEB_SELECT		(1 << 30)
-/** Outputs composite video (DAC A only) */
+/* Outputs composite video (DAC A only) */
 # define TV_ENC_OUTPUT_COMPOSITE	(0 << 28)
-/** Outputs SVideo video (DAC B/C) */
+/* Outputs SVideo video (DAC B/C) */
 # define TV_ENC_OUTPUT_SVIDEO		(1 << 28)
-/** Outputs Component video (DAC A/B/C) */
+/* Outputs Component video (DAC A/B/C) */
 # define TV_ENC_OUTPUT_COMPONENT	(2 << 28)
-/** Outputs Composite and SVideo (DAC A/B/C) */
+/* Outputs Composite and SVideo (DAC A/B/C) */
 # define TV_ENC_OUTPUT_SVIDEO_COMPOSITE	(3 << 28)
 # define TV_TRILEVEL_SYNC		(1 << 21)
-/** Enables slow sync generation (945GM only) */
+/* Enables slow sync generation (945GM only) */
 # define TV_SLOW_SYNC			(1 << 20)
-/** Selects 4x oversampling for 480i and 576p */
+/* Selects 4x oversampling for 480i and 576p */
 # define TV_OVERSAMPLE_4X		(0 << 18)
-/** Selects 2x oversampling for 720p and 1080i */
+/* Selects 2x oversampling for 720p and 1080i */
 # define TV_OVERSAMPLE_2X		(1 << 18)
-/** Selects no oversampling for 1080p */
+/* Selects no oversampling for 1080p */
 # define TV_OVERSAMPLE_NONE		(2 << 18)
-/** Selects 8x oversampling */
+/* Selects 8x oversampling */
 # define TV_OVERSAMPLE_8X		(3 << 18)
-/** Selects progressive mode rather than interlaced */
+/* Selects progressive mode rather than interlaced */
 # define TV_PROGRESSIVE			(1 << 17)
-/** Sets the colorburst to PAL mode.  Required for non-M PAL modes. */
+/* Sets the colorburst to PAL mode.  Required for non-M PAL modes. */
 # define TV_PAL_BURST			(1 << 16)
-/** Field for setting delay of Y compared to C */
+/* Field for setting delay of Y compared to C */
 # define TV_YC_SKEW_MASK		(7 << 12)
-/** Enables a fix for 480p/576p standard definition modes on the 915GM only */
+/* Enables a fix for 480p/576p standard definition modes on the 915GM only */
 # define TV_ENC_SDP_FIX			(1 << 11)
-/**
+/*
  * Enables a fix for the 915GM only.
  *
  * Not sure what it does.
  */
 # define TV_ENC_C0_FIX			(1 << 10)
-/** Bits that must be preserved by software */
+/* Bits that must be preserved by software */
 # define TV_CTL_SAVE			((1 << 11) | (3 << 9) | (7 << 6) | 0xf)
 # define TV_FUSE_STATE_MASK		(3 << 4)
-/** Read-only state that reports all features enabled */
+/* Read-only state that reports all features enabled */
 # define TV_FUSE_STATE_ENABLED		(0 << 4)
-/** Read-only state that reports that Macrovision is disabled in hardware*/
+/* Read-only state that reports that Macrovision is disabled in hardware*/
 # define TV_FUSE_STATE_NO_MACROVISION	(1 << 4)
-/** Read-only state that reports that TV-out is disabled in hardware. */
+/* Read-only state that reports that TV-out is disabled in hardware. */
 # define TV_FUSE_STATE_DISABLED		(2 << 4)
-/** Normal operation */
+/* Normal operation */
 # define TV_TEST_MODE_NORMAL		(0 << 0)
-/** Encoder test pattern 1 - combo pattern */
+/* Encoder test pattern 1 - combo pattern */
 # define TV_TEST_MODE_PATTERN_1		(1 << 0)
-/** Encoder test pattern 2 - full screen vertical 75% color bars */
+/* Encoder test pattern 2 - full screen vertical 75% color bars */
 # define TV_TEST_MODE_PATTERN_2		(2 << 0)
-/** Encoder test pattern 3 - full screen horizontal 75% color bars */
+/* Encoder test pattern 3 - full screen horizontal 75% color bars */
 # define TV_TEST_MODE_PATTERN_3		(3 << 0)
-/** Encoder test pattern 4 - random noise */
+/* Encoder test pattern 4 - random noise */
 # define TV_TEST_MODE_PATTERN_4		(4 << 0)
-/** Encoder test pattern 5 - linear color ramps */
+/* Encoder test pattern 5 - linear color ramps */
 # define TV_TEST_MODE_PATTERN_5		(5 << 0)
-/**
+/*
  * This test mode forces the DACs to 50% of full output.
  *
  * This is used for load detection in combination with TVDAC_SENSE_MASK
@@ -2624,35 +2983,35 @@ enum punit_power_well {
 
 #define TV_DAC			0x68004
 # define TV_DAC_SAVE		0x00ffff00
-/**
+/*
  * Reports that DAC state change logic has reported change (RO).
  *
  * This gets cleared when TV_DAC_STATE_EN is cleared
 */
 # define TVDAC_STATE_CHG		(1 << 31)
 # define TVDAC_SENSE_MASK		(7 << 28)
-/** Reports that DAC A voltage is above the detect threshold */
+/* Reports that DAC A voltage is above the detect threshold */
 # define TVDAC_A_SENSE			(1 << 30)
-/** Reports that DAC B voltage is above the detect threshold */
+/* Reports that DAC B voltage is above the detect threshold */
 # define TVDAC_B_SENSE			(1 << 29)
-/** Reports that DAC C voltage is above the detect threshold */
+/* Reports that DAC C voltage is above the detect threshold */
 # define TVDAC_C_SENSE			(1 << 28)
-/**
+/*
  * Enables DAC state detection logic, for load-based TV detection.
  *
  * The PLL of the chosen pipe (in TV_CTL) must be running, and the encoder set
  * to off, for load detection to work.
  */
 # define TVDAC_STATE_CHG_EN		(1 << 27)
-/** Sets the DAC A sense value to high */
+/* Sets the DAC A sense value to high */
 # define TVDAC_A_SENSE_CTL		(1 << 26)
-/** Sets the DAC B sense value to high */
+/* Sets the DAC B sense value to high */
 # define TVDAC_B_SENSE_CTL		(1 << 25)
-/** Sets the DAC C sense value to high */
+/* Sets the DAC C sense value to high */
 # define TVDAC_C_SENSE_CTL		(1 << 24)
-/** Overrides the ENC_ENABLE and DAC voltage levels */
+/* Overrides the ENC_ENABLE and DAC voltage levels */
 # define DAC_CTL_OVERRIDE		(1 << 7)
-/** Sets the slew rate.  Must be preserved in software */
+/* Sets the slew rate.  Must be preserved in software */
 # define ENC_TVDAC_SLEW_FAST		(1 << 6)
 # define DAC_A_1_3_V			(0 << 4)
 # define DAC_A_1_1_V			(1 << 4)
@@ -2667,7 +3026,7 @@ enum punit_power_well {
 # define DAC_C_0_7_V			(2 << 0)
 # define DAC_C_MASK			(3 << 0)
 
-/**
+/*
  * CSC coefficients are stored in a floating point format with 9 bits of
  * mantissa and 2 or 3 bits of exponent.  The exponent is represented as 2**-n,
  * where 2-bit exponents are unsigned n, and 3-bit exponents are signed n with
@@ -2682,7 +3041,7 @@ enum punit_power_well {
 #define TV_CSC_Y2		0x68014
 # define TV_BY_MASK			0x07ff0000
 # define TV_BY_SHIFT			16
-/**
+/*
  * Y attenuation for component video.
  *
  * Stored in 1.9 fixed point.
@@ -2699,7 +3058,7 @@ enum punit_power_well {
 #define TV_CSC_U2		0x6801c
 # define TV_BU_MASK			0x07ff0000
 # define TV_BU_SHIFT			16
-/**
+/*
  * U attenuation for component video.
  *
  * Stored in 1.9 fixed point.
@@ -2716,7 +3075,7 @@ enum punit_power_well {
 #define TV_CSC_V2		0x68024
 # define TV_BV_MASK			0x07ff0000
 # define TV_BV_SHIFT			16
-/**
+/*
  * V attenuation for component video.
  *
  * Stored in 1.9 fixed point.
@@ -2725,74 +3084,74 @@ enum punit_power_well {
 # define TV_AV_SHIFT			0
 
 #define TV_CLR_KNOBS		0x68028
-/** 2s-complement brightness adjustment */
+/* 2s-complement brightness adjustment */
 # define TV_BRIGHTNESS_MASK		0xff000000
 # define TV_BRIGHTNESS_SHIFT		24
-/** Contrast adjustment, as a 2.6 unsigned floating point number */
+/* Contrast adjustment, as a 2.6 unsigned floating point number */
 # define TV_CONTRAST_MASK		0x00ff0000
 # define TV_CONTRAST_SHIFT		16
-/** Saturation adjustment, as a 2.6 unsigned floating point number */
+/* Saturation adjustment, as a 2.6 unsigned floating point number */
 # define TV_SATURATION_MASK		0x0000ff00
 # define TV_SATURATION_SHIFT		8
-/** Hue adjustment, as an integer phase angle in degrees */
+/* Hue adjustment, as an integer phase angle in degrees */
 # define TV_HUE_MASK			0x000000ff
 # define TV_HUE_SHIFT			0
 
 #define TV_CLR_LEVEL		0x6802c
-/** Controls the DAC level for black */
+/* Controls the DAC level for black */
 # define TV_BLACK_LEVEL_MASK		0x01ff0000
 # define TV_BLACK_LEVEL_SHIFT		16
-/** Controls the DAC level for blanking */
+/* Controls the DAC level for blanking */
 # define TV_BLANK_LEVEL_MASK		0x000001ff
 # define TV_BLANK_LEVEL_SHIFT		0
 
 #define TV_H_CTL_1		0x68030
-/** Number of pixels in the hsync. */
+/* Number of pixels in the hsync. */
 # define TV_HSYNC_END_MASK		0x1fff0000
 # define TV_HSYNC_END_SHIFT		16
-/** Total number of pixels minus one in the line (display and blanking). */
+/* Total number of pixels minus one in the line (display and blanking). */
 # define TV_HTOTAL_MASK			0x00001fff
 # define TV_HTOTAL_SHIFT		0
 
 #define TV_H_CTL_2		0x68034
-/** Enables the colorburst (needed for non-component color) */
+/* Enables the colorburst (needed for non-component color) */
 # define TV_BURST_ENA			(1 << 31)
-/** Offset of the colorburst from the start of hsync, in pixels minus one. */
+/* Offset of the colorburst from the start of hsync, in pixels minus one. */
 # define TV_HBURST_START_SHIFT		16
 # define TV_HBURST_START_MASK		0x1fff0000
-/** Length of the colorburst */
+/* Length of the colorburst */
 # define TV_HBURST_LEN_SHIFT		0
 # define TV_HBURST_LEN_MASK		0x0001fff
 
 #define TV_H_CTL_3		0x68038
-/** End of hblank, measured in pixels minus one from start of hsync */
+/* End of hblank, measured in pixels minus one from start of hsync */
 # define TV_HBLANK_END_SHIFT		16
 # define TV_HBLANK_END_MASK		0x1fff0000
-/** Start of hblank, measured in pixels minus one from start of hsync */
+/* Start of hblank, measured in pixels minus one from start of hsync */
 # define TV_HBLANK_START_SHIFT		0
 # define TV_HBLANK_START_MASK		0x0001fff
 
 #define TV_V_CTL_1		0x6803c
-/** XXX */
+/* XXX */
 # define TV_NBR_END_SHIFT		16
 # define TV_NBR_END_MASK		0x07ff0000
-/** XXX */
+/* XXX */
 # define TV_VI_END_F1_SHIFT		8
 # define TV_VI_END_F1_MASK		0x00003f00
-/** XXX */
+/* XXX */
 # define TV_VI_END_F2_SHIFT		0
 # define TV_VI_END_F2_MASK		0x0000003f
 
 #define TV_V_CTL_2		0x68040
-/** Length of vsync, in half lines */
+/* Length of vsync, in half lines */
 # define TV_VSYNC_LEN_MASK		0x07ff0000
 # define TV_VSYNC_LEN_SHIFT		16
-/** Offset of the start of vsync in field 1, measured in one less than the
+/* Offset of the start of vsync in field 1, measured in one less than the
  * number of half lines.
  */
 # define TV_VSYNC_START_F1_MASK		0x00007f00
 # define TV_VSYNC_START_F1_SHIFT	8
-/**
+/*
  * Offset of the start of vsync in field 2, measured in one less than the
  * number of half lines.
  */
@@ -2800,17 +3159,17 @@ enum punit_power_well {
 # define TV_VSYNC_START_F2_SHIFT	0
 
 #define TV_V_CTL_3		0x68044
-/** Enables generation of the equalization signal */
+/* Enables generation of the equalization signal */
 # define TV_EQUAL_ENA			(1 << 31)
-/** Length of vsync, in half lines */
+/* Length of vsync, in half lines */
 # define TV_VEQ_LEN_MASK		0x007f0000
 # define TV_VEQ_LEN_SHIFT		16
-/** Offset of the start of equalization in field 1, measured in one less than
+/* Offset of the start of equalization in field 1, measured in one less than
  * the number of half lines.
  */
 # define TV_VEQ_START_F1_MASK		0x0007f00
 # define TV_VEQ_START_F1_SHIFT		8
-/**
+/*
  * Offset of the start of equalization in field 2, measured in one less than
  * the number of half lines.
  */
@@ -2818,13 +3177,13 @@ enum punit_power_well {
 # define TV_VEQ_START_F2_SHIFT		0
 
 #define TV_V_CTL_4		0x68048
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F1_MASK	0x003f0000
 # define TV_VBURST_START_F1_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2832,13 +3191,13 @@ enum punit_power_well {
 # define TV_VBURST_END_F1_SHIFT		0
 
 #define TV_V_CTL_5		0x6804c
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F2_MASK	0x003f0000
 # define TV_VBURST_START_F2_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2846,13 +3205,13 @@ enum punit_power_well {
 # define TV_VBURST_END_F2_SHIFT		0
 
 #define TV_V_CTL_6		0x68050
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F3_MASK	0x003f0000
 # define TV_VBURST_START_F3_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2860,13 +3219,13 @@ enum punit_power_well {
 # define TV_VBURST_END_F3_SHIFT		0
 
 #define TV_V_CTL_7		0x68054
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F4_MASK	0x003f0000
 # define TV_VBURST_START_F4_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2874,56 +3233,56 @@ enum punit_power_well {
 # define TV_VBURST_END_F4_SHIFT		0
 
 #define TV_SC_CTL_1		0x68060
-/** Turns on the first subcarrier phase generation DDA */
+/* Turns on the first subcarrier phase generation DDA */
 # define TV_SC_DDA1_EN			(1 << 31)
-/** Turns on the first subcarrier phase generation DDA */
+/* Turns on the first subcarrier phase generation DDA */
 # define TV_SC_DDA2_EN			(1 << 30)
-/** Turns on the first subcarrier phase generation DDA */
+/* Turns on the first subcarrier phase generation DDA */
 # define TV_SC_DDA3_EN			(1 << 29)
-/** Sets the subcarrier DDA to reset frequency every other field */
+/* Sets the subcarrier DDA to reset frequency every other field */
 # define TV_SC_RESET_EVERY_2		(0 << 24)
-/** Sets the subcarrier DDA to reset frequency every fourth field */
+/* Sets the subcarrier DDA to reset frequency every fourth field */
 # define TV_SC_RESET_EVERY_4		(1 << 24)
-/** Sets the subcarrier DDA to reset frequency every eighth field */
+/* Sets the subcarrier DDA to reset frequency every eighth field */
 # define TV_SC_RESET_EVERY_8		(2 << 24)
-/** Sets the subcarrier DDA to never reset the frequency */
+/* Sets the subcarrier DDA to never reset the frequency */
 # define TV_SC_RESET_NEVER		(3 << 24)
-/** Sets the peak amplitude of the colorburst.*/
+/* Sets the peak amplitude of the colorburst.*/
 # define TV_BURST_LEVEL_MASK		0x00ff0000
 # define TV_BURST_LEVEL_SHIFT		16
-/** Sets the increment of the first subcarrier phase generation DDA */
+/* Sets the increment of the first subcarrier phase generation DDA */
 # define TV_SCDDA1_INC_MASK		0x00000fff
 # define TV_SCDDA1_INC_SHIFT		0
 
 #define TV_SC_CTL_2		0x68064
-/** Sets the rollover for the second subcarrier phase generation DDA */
+/* Sets the rollover for the second subcarrier phase generation DDA */
 # define TV_SCDDA2_SIZE_MASK		0x7fff0000
 # define TV_SCDDA2_SIZE_SHIFT		16
-/** Sets the increent of the second subcarrier phase generation DDA */
+/* Sets the increent of the second subcarrier phase generation DDA */
 # define TV_SCDDA2_INC_MASK		0x00007fff
 # define TV_SCDDA2_INC_SHIFT		0
 
 #define TV_SC_CTL_3		0x68068
-/** Sets the rollover for the third subcarrier phase generation DDA */
+/* Sets the rollover for the third subcarrier phase generation DDA */
 # define TV_SCDDA3_SIZE_MASK		0x7fff0000
 # define TV_SCDDA3_SIZE_SHIFT		16
-/** Sets the increent of the third subcarrier phase generation DDA */
+/* Sets the increent of the third subcarrier phase generation DDA */
 # define TV_SCDDA3_INC_MASK		0x00007fff
 # define TV_SCDDA3_INC_SHIFT		0
 
 #define TV_WIN_POS		0x68070
-/** X coordinate of the display from the start of horizontal active */
+/* X coordinate of the display from the start of horizontal active */
 # define TV_XPOS_MASK			0x1fff0000
 # define TV_XPOS_SHIFT			16
-/** Y coordinate of the display from the start of vertical active (NBR) */
+/* Y coordinate of the display from the start of vertical active (NBR) */
 # define TV_YPOS_MASK			0x00000fff
 # define TV_YPOS_SHIFT			0
 
 #define TV_WIN_SIZE		0x68074
-/** Horizontal size of the display window, measured in pixels*/
+/* Horizontal size of the display window, measured in pixels*/
 # define TV_XSIZE_MASK			0x1fff0000
 # define TV_XSIZE_SHIFT			16
-/**
+/*
  * Vertical size of the display window, measured in pixels.
  *
  * Must be even for interlaced modes.
@@ -2932,28 +3291,28 @@ enum punit_power_well {
 # define TV_YSIZE_SHIFT			0
 
 #define TV_FILTER_CTL_1		0x68080
-/**
+/*
  * Enables automatic scaling calculation.
  *
  * If set, the rest of the registers are ignored, and the calculated values can
  * be read back from the register.
  */
 # define TV_AUTO_SCALE			(1 << 31)
-/**
+/*
  * Disables the vertical filter.
  *
  * This is required on modes more than 1024 pixels wide */
 # define TV_V_FILTER_BYPASS		(1 << 29)
-/** Enables adaptive vertical filtering */
+/* Enables adaptive vertical filtering */
 # define TV_VADAPT			(1 << 28)
 # define TV_VADAPT_MODE_MASK		(3 << 26)
-/** Selects the least adaptive vertical filtering mode */
+/* Selects the least adaptive vertical filtering mode */
 # define TV_VADAPT_MODE_LEAST		(0 << 26)
-/** Selects the moderately adaptive vertical filtering mode */
+/* Selects the moderately adaptive vertical filtering mode */
 # define TV_VADAPT_MODE_MODERATE	(1 << 26)
-/** Selects the most adaptive vertical filtering mode */
+/* Selects the most adaptive vertical filtering mode */
 # define TV_VADAPT_MODE_MOST		(3 << 26)
-/**
+/*
  * Sets the horizontal scaling factor.
  *
  * This should be the fractional part of the horizontal scaling factor divided
@@ -2965,14 +3324,14 @@ enum punit_power_well {
 # define TV_HSCALE_FRAC_SHIFT		0
 
 #define TV_FILTER_CTL_2		0x68084
-/**
+/*
  * Sets the integer part of the 3.15 fixed-point vertical scaling factor.
  *
  * TV_VSCALE should be (src height - 1) / ((interlace * dest height) - 1)
  */
 # define TV_VSCALE_INT_MASK		0x00038000
 # define TV_VSCALE_INT_SHIFT		15
-/**
+/*
  * Sets the fractional part of the 3.15 fixed-point vertical scaling factor.
  *
  * \sa TV_VSCALE_INT_MASK
@@ -2981,7 +3340,7 @@ enum punit_power_well {
 # define TV_VSCALE_FRAC_SHIFT		0
 
 #define TV_FILTER_CTL_3		0x68088
-/**
+/*
  * Sets the integer part of the 3.15 fixed-point vertical scaling factor.
  *
  * TV_VSCALE should be (src height - 1) / (1/4 * (dest height - 1))
@@ -2990,7 +3349,7 @@ enum punit_power_well {
  */
 # define TV_VSCALE_IP_INT_MASK		0x00038000
 # define TV_VSCALE_IP_INT_SHIFT		15
-/**
+/*
  * Sets the fractional part of the 3.15 fixed-point vertical scaling factor.
  *
  * For progressive modes, TV_VSCALE_IP_INT should be set to zeroes.
@@ -3002,26 +3361,26 @@ enum punit_power_well {
 
 #define TV_CC_CONTROL		0x68090
 # define TV_CC_ENABLE			(1 << 31)
-/**
+/*
  * Specifies which field to send the CC data in.
  *
  * CC data is usually sent in field 0.
  */
 # define TV_CC_FID_MASK			(1 << 27)
 # define TV_CC_FID_SHIFT		27
-/** Sets the horizontal position of the CC data.  Usually 135. */
+/* Sets the horizontal position of the CC data.  Usually 135. */
 # define TV_CC_HOFF_MASK		0x03ff0000
 # define TV_CC_HOFF_SHIFT		16
-/** Sets the vertical position of the CC data.  Usually 21 */
+/* Sets the vertical position of the CC data.  Usually 21 */
 # define TV_CC_LINE_MASK		0x0000003f
 # define TV_CC_LINE_SHIFT		0
 
 #define TV_CC_DATA		0x68094
 # define TV_CC_RDY			(1 << 31)
-/** Second word of CC data to be transmitted. */
+/* Second word of CC data to be transmitted. */
 # define TV_CC_DATA_2_MASK		0x007f0000
 # define TV_CC_DATA_2_SHIFT		16
-/** First word of CC data to be transmitted. */
+/* First word of CC data to be transmitted. */
 # define TV_CC_DATA_1_MASK		0x0000007f
 # define TV_CC_DATA_1_SHIFT		0
 
@@ -3043,6 +3402,8 @@ enum punit_power_well {
 #define   DP_PORT_EN			(1 << 31)
 #define   DP_PIPEB_SELECT		(1 << 30)
 #define   DP_PIPE_MASK			(1 << 30)
+#define   DP_PIPE_SELECT_CHV(pipe)	((pipe) << 16)
+#define   DP_PIPE_MASK_CHV		(3 << 16)
 
 /* Link training mode - select a suitable mode for each stage */
 #define   DP_LINK_TRAIN_PAT_1		(0 << 28)
@@ -3090,32 +3451,32 @@ enum punit_power_well {
 #define   DP_PLL_FREQ_160MHZ		(1 << 16)
 #define   DP_PLL_FREQ_MASK		(3 << 16)
 
-/** locked once port is enabled */
+/* locked once port is enabled */
 #define   DP_PORT_REVERSAL		(1 << 15)
 
 /* eDP */
 #define   DP_PLL_ENABLE			(1 << 14)
 
-/** sends the clock on lane 15 of the PEG for debug */
+/* sends the clock on lane 15 of the PEG for debug */
 #define   DP_CLOCK_OUTPUT_ENABLE	(1 << 13)
 
 #define   DP_SCRAMBLING_DISABLE		(1 << 12)
 #define   DP_SCRAMBLING_DISABLE_IRONLAKE	(1 << 7)
 
-/** limit RGB values to avoid confusing TVs */
+/* limit RGB values to avoid confusing TVs */
 #define   DP_COLOR_RANGE_16_235		(1 << 8)
 
-/** Turn on the audio link */
+/* Turn on the audio link */
 #define   DP_AUDIO_OUTPUT_ENABLE	(1 << 6)
 
-/** vs and hs sync polarity */
+/* vs and hs sync polarity */
 #define   DP_SYNC_VS_HIGH		(1 << 4)
 #define   DP_SYNC_HS_HIGH		(1 << 3)
 
-/** A fantasy */
+/* A fantasy */
 #define   DP_DETECTED			(1 << 2)
 
-/** The aux channel provides a way to talk to the
+/* The aux channel provides a way to talk to the
  * signal sink for DDC etc. Max packet size supported
  * is 20 bytes in each direction, hence the 5 fixed
  * data registers
@@ -3258,6 +3619,7 @@ enum punit_power_well {
 #define   PIPECONF_INTERLACED_DBL_ILK		(4 << 21) /* ilk/snb only */
 #define   PIPECONF_PFIT_PF_INTERLACED_DBL_ILK	(5 << 21) /* ilk/snb only */
 #define   PIPECONF_INTERLACE_MODE_MASK		(7 << 21)
+#define   PIPECONF_EDP_RR_MODE_SWITCH		(1 << 20)
 #define   PIPECONF_CXSR_DOWNCLOCK	(1<<16)
 #define   PIPECONF_COLOR_RANGE_SELECT	(1 << 13)
 #define   PIPECONF_BPC_MASK	(0x7 << 5)
@@ -3276,6 +3638,7 @@ enum punit_power_well {
 #define   SPRITE1_FLIP_DONE_INT_EN_VLV		(1UL<<30)
 #define   PIPE_CRC_ERROR_ENABLE			(1UL<<29)
 #define   PIPE_CRC_DONE_ENABLE			(1UL<<28)
+#define   PERF_COUNTER2_INTERRUPT_EN		(1UL<<27)
 #define   PIPE_GMBUS_EVENT_ENABLE		(1UL<<27)
 #define   PLANE_FLIP_DONE_INT_EN_VLV		(1UL<<26)
 #define   PIPE_HOTPLUG_INTERRUPT_ENABLE		(1UL<<26)
@@ -3287,8 +3650,10 @@ enum punit_power_well {
 #define   PIPE_ODD_FIELD_INTERRUPT_ENABLE	(1UL<<21)
 #define   PIPE_EVEN_FIELD_INTERRUPT_ENABLE	(1UL<<20)
 #define   PIPE_B_PSR_INTERRUPT_ENABLE_VLV	(1UL<<19)
+#define   PERF_COUNTER_INTERRUPT_EN		(1UL<<19)
 #define   PIPE_HOTPLUG_TV_INTERRUPT_ENABLE	(1UL<<18) /* pre-965 */
 #define   PIPE_START_VBLANK_INTERRUPT_ENABLE	(1UL<<18) /* 965 or later */
+#define   PIPE_FRAMESTART_INTERRUPT_ENABLE	(1UL<<17)
 #define   PIPE_VBLANK_INTERRUPT_ENABLE		(1UL<<17)
 #define   PIPEA_HBLANK_INT_EN_VLV		(1UL<<16)
 #define   PIPE_OVERLAY_UPDATED_ENABLE		(1UL<<16)
@@ -3296,6 +3661,7 @@ enum punit_power_well {
 #define   SPRITE0_FLIP_DONE_INT_STATUS_VLV	(1UL<<14)
 #define   PIPE_CRC_ERROR_INTERRUPT_STATUS	(1UL<<13)
 #define   PIPE_CRC_DONE_INTERRUPT_STATUS	(1UL<<12)
+#define   PERF_COUNTER2_INTERRUPT_STATUS	(1UL<<11)
 #define   PIPE_GMBUS_INTERRUPT_STATUS		(1UL<<11)
 #define   PLANE_FLIP_DONE_INT_STATUS_VLV	(1UL<<10)
 #define   PIPE_HOTPLUG_INTERRUPT_STATUS		(1UL<<10)
@@ -3304,20 +3670,25 @@ enum punit_power_well {
 #define   PIPE_DPST_EVENT_STATUS		(1UL<<7)
 #define   PIPE_LEGACY_BLC_EVENT_STATUS		(1UL<<6)
 #define   PIPE_A_PSR_STATUS_VLV			(1UL<<6)
+#define   PIPE_LEGACY_BLC_EVENT_STATUS		(1UL<<6)
 #define   PIPE_ODD_FIELD_INTERRUPT_STATUS	(1UL<<5)
 #define   PIPE_EVEN_FIELD_INTERRUPT_STATUS	(1UL<<4)
 #define   PIPE_B_PSR_STATUS_VLV			(1UL<<3)
+#define   PERF_COUNTER_INTERRUPT_STATUS		(1UL<<3)
 #define   PIPE_HOTPLUG_TV_INTERRUPT_STATUS	(1UL<<2) /* pre-965 */
 #define   PIPE_START_VBLANK_INTERRUPT_STATUS	(1UL<<2) /* 965 or later */
+#define   PIPE_FRAMESTART_INTERRUPT_STATUS	(1UL<<1)
 #define   PIPE_VBLANK_INTERRUPT_STATUS		(1UL<<1)
+#define   PIPE_HBLANK_INT_STATUS		(1UL<<0)
 #define   PIPE_OVERLAY_UPDATED_STATUS		(1UL<<0)
 
 #define PIPESTAT_INT_ENABLE_MASK		0x7fff0000
 #define PIPESTAT_INT_STATUS_MASK		0x0000ffff
 
-#define PIPE_A_OFFSET	0x70000
-#define PIPE_B_OFFSET	0x71000
-#define PIPE_C_OFFSET	0x72000
+#define PIPE_A_OFFSET		0x70000
+#define PIPE_B_OFFSET		0x71000
+#define PIPE_C_OFFSET		0x72000
+#define CHV_PIPE_C_OFFSET	0x74000
 /*
  * There's actually no pipe EDP. Some pipe registers have
  * simply shifted from the pipe to the transcoder, while
@@ -3355,14 +3726,25 @@ enum punit_power_well {
 #define   SPRITED_FLIP_DONE_INT_EN		(1<<26)
 #define   SPRITEC_FLIP_DONE_INT_EN		(1<<25)
 #define   PLANEB_FLIP_DONE_INT_EN		(1<<24)
+#define   PIPE_PSR_INT_EN			(1<<22)
 #define   PIPEA_LINE_COMPARE_INT_EN		(1<<21)
 #define   PIPEA_HLINE_INT_EN			(1<<20)
 #define   PIPEA_VBLANK_INT_EN			(1<<19)
 #define   SPRITEB_FLIP_DONE_INT_EN		(1<<18)
 #define   SPRITEA_FLIP_DONE_INT_EN		(1<<17)
 #define   PLANEA_FLIPDONE_INT_EN		(1<<16)
-
-#define DPINVGTT				(VLV_DISPLAY_BASE + 0x7002c) /* VLV only */
+#define   PIPEC_LINE_COMPARE_INT_EN		(1<<13)
+#define   PIPEC_HLINE_INT_EN			(1<<12)
+#define   PIPEC_VBLANK_INT_EN			(1<<11)
+#define   SPRITEF_FLIPDONE_INT_EN		(1<<10)
+#define   SPRITEE_FLIPDONE_INT_EN		(1<<9)
+#define   PLANEC_FLIPDONE_INT_EN		(1<<8)
+
+#define DPINVGTT				(VLV_DISPLAY_BASE + 0x7002c) /* VLV/CHV only */
+#define   SPRITEF_INVALID_GTT_INT_EN		(1<<27)
+#define   SPRITEE_INVALID_GTT_INT_EN		(1<<26)
+#define   PLANEC_INVALID_GTT_INT_EN		(1<<25)
+#define   CURSORC_INVALID_GTT_INT_EN		(1<<24)
 #define   CURSORB_INVALID_GTT_INT_EN		(1<<23)
 #define   CURSORA_INVALID_GTT_INT_EN		(1<<22)
 #define   SPRITED_INVALID_GTT_INT_EN		(1<<21)
@@ -3372,6 +3754,11 @@ enum punit_power_well {
 #define   SPRITEA_INVALID_GTT_INT_EN		(1<<17)
 #define   PLANEA_INVALID_GTT_INT_EN		(1<<16)
 #define   DPINVGTT_EN_MASK			0xff0000
+#define   DPINVGTT_EN_MASK_CHV			0xfff0000
+#define   SPRITEF_INVALID_GTT_STATUS		(1<<11)
+#define   SPRITEE_INVALID_GTT_STATUS		(1<<10)
+#define   PLANEC_INVALID_GTT_STATUS		(1<<9)
+#define   CURSORC_INVALID_GTT_STATUS		(1<<8)
 #define   CURSORB_INVALID_GTT_STATUS		(1<<7)
 #define   CURSORA_INVALID_GTT_STATUS		(1<<6)
 #define   SPRITED_INVALID_GTT_STATUS		(1<<5)
@@ -3381,6 +3768,7 @@ enum punit_power_well {
 #define   SPRITEA_INVALID_GTT_STATUS		(1<<1)
 #define   PLANEA_INVALID_GTT_STATUS		(1<<0)
 #define   DPINVGTT_STATUS_MASK			0xff
+#define   DPINVGTT_STATUS_MASK_CHV		0xfff
 
 #define DSPARB			0x70030
 #define   DSPARB_CSTART_MASK	(0x7f << 7)
@@ -3420,14 +3808,43 @@ enum punit_power_well {
 #define DDL_CURSORA_PRECISION_32	(1<<31)
 #define DDL_CURSORA_PRECISION_16	(0<<31)
 #define DDL_CURSORA_SHIFT		24
+#define DDL_SPRITEB_PRECISION_32	(1<<23)
+#define DDL_SPRITEB_PRECISION_16	(0<<23)
+#define DDL_SPRITEB_SHIFT		16
+#define DDL_SPRITEA_PRECISION_32	(1<<15)
+#define DDL_SPRITEA_PRECISION_16	(0<<15)
+#define DDL_SPRITEA_SHIFT		8
 #define DDL_PLANEA_PRECISION_32		(1<<7)
 #define DDL_PLANEA_PRECISION_16		(0<<7)
+#define DDL_PLANEA_SHIFT		0
+
 #define VLV_DDL2			(VLV_DISPLAY_BASE + 0x70054)
 #define DDL_CURSORB_PRECISION_32	(1<<31)
 #define DDL_CURSORB_PRECISION_16	(0<<31)
 #define DDL_CURSORB_SHIFT		24
+#define DDL_SPRITED_PRECISION_32	(1<<23)
+#define DDL_SPRITED_PRECISION_16	(0<<23)
+#define DDL_SPRITED_SHIFT		16
+#define DDL_SPRITEC_PRECISION_32	(1<<15)
+#define DDL_SPRITEC_PRECISION_16	(0<<15)
+#define DDL_SPRITEC_SHIFT		8
 #define DDL_PLANEB_PRECISION_32		(1<<7)
 #define DDL_PLANEB_PRECISION_16		(0<<7)
+#define DDL_PLANEB_SHIFT		0
+
+#define VLV_DDL3			(VLV_DISPLAY_BASE + 0x70058)
+#define DDL_CURSORC_PRECISION_32	(1<<31)
+#define DDL_CURSORC_PRECISION_16	(0<<31)
+#define DDL_CURSORC_SHIFT		24
+#define DDL_SPRITEF_PRECISION_32	(1<<23)
+#define DDL_SPRITEF_PRECISION_16	(0<<23)
+#define DDL_SPRITEF_SHIFT		16
+#define DDL_SPRITEE_PRECISION_32	(1<<15)
+#define DDL_SPRITEE_PRECISION_16	(0<<15)
+#define DDL_SPRITEE_SHIFT		8
+#define DDL_PLANEC_PRECISION_32		(1<<7)
+#define DDL_PLANEC_PRECISION_16		(0<<7)
+#define DDL_PLANEC_SHIFT		0
 
 /* FIFO watermark sizes etc */
 #define G4X_FIFO_LINE_SIZE	64
@@ -3535,12 +3952,13 @@ enum punit_power_well {
 #define   PIPE_PIXEL_MASK         0x00ffffff
 #define   PIPE_PIXEL_SHIFT        0
 /* GM45+ just has to be different */
-#define _PIPEA_FRMCOUNT_GM45	(dev_priv->info.display_mmio_offset + 0x70040)
-#define _PIPEA_FLIPCOUNT_GM45	(dev_priv->info.display_mmio_offset + 0x70044)
-#define PIPE_FRMCOUNT_GM45(pipe) _PIPE(pipe, _PIPEA_FRMCOUNT_GM45, _PIPEB_FRMCOUNT_GM45)
+#define _PIPEA_FRMCOUNT_GM45	0x70040
+#define _PIPEA_FLIPCOUNT_GM45	0x70044
+#define PIPE_FRMCOUNT_GM45(pipe) _PIPE2(pipe, _PIPEA_FRMCOUNT_GM45)
+#define PIPE_FLIPCOUNT_GM45(pipe) _PIPE2(pipe, _PIPEA_FLIPCOUNT_GM45)
 
 /* Cursor A & B regs */
-#define _CURACNTR		(dev_priv->info.display_mmio_offset + 0x70080)
+#define _CURACNTR		0x70080
 /* Old style CUR*CNTR flags (desktop 8xx) */
 #define   CURSOR_ENABLE		0x80000000
 #define   CURSOR_GAMMA_ENABLE	0x40000000
@@ -3567,28 +3985,34 @@ enum punit_power_well {
 #define   MCURSOR_PIPE_B	(1 << 28)
 #define   MCURSOR_GAMMA_ENABLE  (1 << 26)
 #define   CURSOR_TRICKLE_FEED_DISABLE	(1 << 14)
-#define _CURABASE		(dev_priv->info.display_mmio_offset + 0x70084)
-#define _CURAPOS		(dev_priv->info.display_mmio_offset + 0x70088)
+#define _CURABASE		0x70084
+#define _CURAPOS		0x70088
 #define   CURSOR_POS_MASK       0x007FF
 #define   CURSOR_POS_SIGN       0x8000
 #define   CURSOR_X_SHIFT        0
 #define   CURSOR_Y_SHIFT        16
 #define CURSIZE			0x700a0
-#define _CURBCNTR		(dev_priv->info.display_mmio_offset + 0x700c0)
-#define _CURBBASE		(dev_priv->info.display_mmio_offset + 0x700c4)
-#define _CURBPOS		(dev_priv->info.display_mmio_offset + 0x700c8)
+#define _CURBCNTR		0x700c0
+#define _CURBBASE		0x700c4
+#define _CURBPOS		0x700c8
 
 #define _CURBCNTR_IVB		0x71080
 #define _CURBBASE_IVB		0x71084
 #define _CURBPOS_IVB		0x71088
 
-#define CURCNTR(pipe) _PIPE(pipe, _CURACNTR, _CURBCNTR)
-#define CURBASE(pipe) _PIPE(pipe, _CURABASE, _CURBBASE)
-#define CURPOS(pipe) _PIPE(pipe, _CURAPOS, _CURBPOS)
+#define _CURSOR2(pipe, reg) (dev_priv->info.cursor_offsets[(pipe)] - \
+	dev_priv->info.cursor_offsets[PIPE_A] + (reg) + \
+	dev_priv->info.display_mmio_offset)
+
+#define CURCNTR(pipe) _CURSOR2(pipe, _CURACNTR)
+#define CURBASE(pipe) _CURSOR2(pipe, _CURABASE)
+#define CURPOS(pipe) _CURSOR2(pipe, _CURAPOS)
 
-#define CURCNTR_IVB(pipe) _PIPE(pipe, _CURACNTR, _CURBCNTR_IVB)
-#define CURBASE_IVB(pipe) _PIPE(pipe, _CURABASE, _CURBBASE_IVB)
-#define CURPOS_IVB(pipe) _PIPE(pipe, _CURAPOS, _CURBPOS_IVB)
+#define CURSOR_A_OFFSET 0x70080
+#define CURSOR_B_OFFSET 0x700c0
+#define CHV_CURSOR_C_OFFSET 0x700e0
+#define IVB_CURSOR_B_OFFSET 0x71080
+#define IVB_CURSOR_C_OFFSET 0x72080
 
 /* Display A control */
 #define _DSPACNTR				0x70180
@@ -4093,6 +4517,7 @@ enum punit_power_well {
 #define  GEN8_DE_PIPE_A_IRQ		(1<<16)
 #define  GEN8_DE_PIPE_IRQ(pipe)		(1<<(16+pipe))
 #define  GEN8_GT_VECS_IRQ		(1<<6)
+#define  GEN8_GT_PM_IRQ			(1<<4)
 #define  GEN8_GT_VCS2_IRQ		(1<<3)
 #define  GEN8_GT_VCS1_IRQ		(1<<2)
 #define  GEN8_GT_BCS_IRQ		(1<<1)
@@ -4120,7 +4545,7 @@ enum punit_power_well {
 #define  GEN8_PIPE_SPRITE_FAULT		(1 << 9)
 #define  GEN8_PIPE_PRIMARY_FAULT	(1 << 8)
 #define  GEN8_PIPE_SPRITE_FLIP_DONE	(1 << 5)
-#define  GEN8_PIPE_FLIP_DONE		(1 << 4)
+#define  GEN8_PIPE_PRIMARY_FLIP_DONE	(1 << 4)
 #define  GEN8_PIPE_SCAN_LINE_EVENT	(1 << 2)
 #define  GEN8_PIPE_VSYNC		(1 << 1)
 #define  GEN8_PIPE_VBLANK		(1 << 0)
@@ -4832,6 +5257,8 @@ enum punit_power_well {
 #define  PORT_TRANS_SEL_CPT(pipe)	((pipe) << 29)
 #define  PORT_TO_PIPE(val)	(((val) & (1<<30)) >> 30)
 #define  PORT_TO_PIPE_CPT(val)	(((val) & PORT_TRANS_SEL_MASK) >> 29)
+#define  SDVO_PORT_TO_PIPE_CHV(val)	(((val) & (3<<24)) >> 24)
+#define  DP_PORT_TO_PIPE_CHV(val)	(((val) & (3<<16)) >> 16)
 
 #define TRANS_DP_CTL_A		0xe0300
 #define TRANS_DP_CTL_B		0xe1300
@@ -4888,6 +5315,8 @@ enum punit_power_well {
 
 #define  EDP_LINK_TRAIN_VOL_EMP_MASK_IVB	(0x3f<<22)
 
+#define  VLV_PMWGICZ				0x1300a4
+
 #define  FORCEWAKE				0xA18C
 #define  FORCEWAKE_VLV				0x1300b0
 #define  FORCEWAKE_ACK_VLV			0x1300b4
@@ -4896,15 +5325,22 @@ enum punit_power_well {
 #define  FORCEWAKE_ACK_HSW			0x130044
 #define  FORCEWAKE_ACK				0x130090
 #define  VLV_GTLC_WAKE_CTRL			0x130090
+#define   VLV_GTLC_RENDER_CTX_EXISTS		(1 << 25)
+#define   VLV_GTLC_MEDIA_CTX_EXISTS		(1 << 24)
+#define   VLV_GTLC_ALLOWWAKEREQ			(1 << 0)
+
 #define  VLV_GTLC_PW_STATUS			0x130094
-#define VLV_GTLC_PW_RENDER_STATUS_MASK		0x80
-#define VLV_GTLC_PW_MEDIA_STATUS_MASK		0x20
+#define   VLV_GTLC_ALLOWWAKEACK			(1 << 0)
+#define   VLV_GTLC_ALLOWWAKEERR			(1 << 1)
+#define   VLV_GTLC_PW_MEDIA_STATUS_MASK		(1 << 5)
+#define   VLV_GTLC_PW_RENDER_STATUS_MASK	(1 << 7)
 #define  FORCEWAKE_MT				0xa188 /* multi-threaded */
 #define   FORCEWAKE_KERNEL			0x1
 #define   FORCEWAKE_USER			0x2
 #define  FORCEWAKE_MT_ACK			0x130040
 #define  ECOBUS					0xa180
 #define    FORCEWAKE_MT_ENABLE			(1<<5)
+#define  VLV_SPAREG2H				0xA194
 
 #define  GTFIFODBG				0x120000
 #define    GT_FIFO_SBDROPERR			(1<<6)
@@ -4924,6 +5360,7 @@ enum punit_power_well {
 #define  HSW_EDRAM_PRESENT			0x120010
 
 #define GEN6_UCGCTL1				0x9400
+# define GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE		(1 << 16)
 # define GEN6_BLBUNIT_CLOCK_GATE_DISABLE		(1 << 5)
 # define GEN6_CSUNIT_CLOCK_GATE_DISABLE			(1 << 7)
 
@@ -4934,12 +5371,19 @@ enum punit_power_well {
 # define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE		(1 << 12)
 # define GEN6_RCCUNIT_CLOCK_GATE_DISABLE		(1 << 11)
 
+#define GEN6_UCGCTL3				0x9408
+
 #define GEN7_UCGCTL4				0x940c
 #define  GEN7_L3BANK2X_CLOCK_GATE_DISABLE	(1<<25)
 
+#define GEN6_RCGCTL1				0x9410
+#define GEN6_RCGCTL2				0x9414
+#define GEN6_RSTCTL				0x9420
+
 #define GEN8_UCGCTL6				0x9430
 #define   GEN8_SDEUNIT_CLOCK_GATE_DISABLE	(1<<14)
 
+#define GEN6_GFXPAUSE				0xA000
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
 #define   GEN6_FREQUENCY(x)			((x)<<25)
@@ -4992,6 +5436,9 @@ enum punit_power_well {
 #define GEN6_RP_UP_EI				0xA068
 #define GEN6_RP_DOWN_EI				0xA06C
 #define GEN6_RP_IDLE_HYSTERSIS			0xA070
+#define GEN6_RPDEUHWTC				0xA080
+#define GEN6_RPDEUC				0xA084
+#define GEN6_RPDEUCSW				0xA088
 #define GEN6_RC_STATE				0xA094
 #define GEN6_RC1_WAKE_RATE_LIMIT		0xA098
 #define GEN6_RC6_WAKE_RATE_LIMIT		0xA09C
@@ -4999,11 +5446,15 @@ enum punit_power_well {
 #define GEN6_RC_EVALUATION_INTERVAL		0xA0A8
 #define GEN6_RC_IDLE_HYSTERSIS			0xA0AC
 #define GEN6_RC_SLEEP				0xA0B0
+#define GEN6_RCUBMABDTMR			0xA0B0
 #define GEN6_RC1e_THRESHOLD			0xA0B4
 #define GEN6_RC6_THRESHOLD			0xA0B8
 #define GEN6_RC6p_THRESHOLD			0xA0BC
+#define VLV_RCEDATA				0xA0BC
 #define GEN6_RC6pp_THRESHOLD			0xA0C0
 #define GEN6_PMINTRMSK				0xA168
+#define GEN8_PMINTR_REDIRECT_TO_NON_DISP	(1<<31)
+#define VLV_PWRDWNUPCTL				0xA294
 
 #define GEN6_PMISR				0x44020
 #define GEN6_PMIMR				0x44024 /* rps_lock */
@@ -5020,6 +5471,9 @@ enum punit_power_well {
 						 GEN6_PM_RP_DOWN_THRESHOLD | \
 						 GEN6_PM_RP_DOWN_TIMEOUT)
 
+#define GEN7_GT_SCRATCH_BASE			0x4F100
+#define GEN7_GT_SCRATCH_REG_NUM			8
+
 #define VLV_GTLC_SURVIVABILITY_REG              0x130098
 #define VLV_GFX_CLK_STATUS_BIT			(1<<3)
 #define VLV_GFX_CLK_FORCE_ON_BIT		(1<<2)
@@ -5030,6 +5484,9 @@ enum punit_power_well {
 #define   VLV_MEDIA_RC6_COUNT_EN		(1<<1)
 #define   VLV_RENDER_RC6_COUNT_EN		(1<<0)
 #define GEN6_GT_GFX_RC6				0x138108
+#define VLV_GT_RENDER_RC6			0x138108
+#define VLV_GT_MEDIA_RC6			0x13810C
+
 #define GEN6_GT_GFX_RC6p			0x13810C
 #define GEN6_GT_GFX_RC6pp			0x138110
 
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 56785e8fb2eb..043123c77a1f 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -328,8 +328,6 @@ int i915_save_state(struct drm_device *dev)
 		}
 	}
 
-	intel_disable_gt_powersave(dev);
-
 	/* Cache mode state */
 	if (INTEL_INFO(dev)->gen < 7)
 		dev_priv->regfile.saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0);
diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index 9c57029f6f4b..86ce39aad0ff 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -186,7 +186,7 @@ i915_l3_write(struct file *filp, struct kobject *kobj,
 	struct drm_minor *dminor = dev_to_drm_minor(dev);
 	struct drm_device *drm_dev = dminor->dev;
 	struct drm_i915_private *dev_priv = drm_dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	u32 *temp = NULL; /* Just here to make handling failures easy */
 	int slice = (int)(uintptr_t)attr->private;
 	int ret;
@@ -263,6 +263,8 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 
 	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
+	intel_runtime_pm_get(dev_priv);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 	if (IS_VALLEYVIEW(dev_priv->dev)) {
 		u32 freq;
@@ -273,6 +275,8 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 	}
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
+	intel_runtime_pm_put(dev_priv);
+
 	return snprintf(buf, PAGE_SIZE, "%d\n", ret);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index 23c26f1f8b37..f5aa0067755a 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -7,6 +7,7 @@
 
 #include <drm/drmP.h>
 #include "i915_drv.h"
+#include "intel_drv.h"
 #include "intel_ringbuffer.h"
 
 #undef TRACE_SYSTEM
@@ -14,6 +15,80 @@
 #define TRACE_SYSTEM_STRING __stringify(TRACE_SYSTEM)
 #define TRACE_INCLUDE_FILE i915_trace
 
+/* pipe updates */
+
+TRACE_EVENT(i915_pipe_update_start,
+	    TP_PROTO(struct intel_crtc *crtc, u32 min, u32 max),
+	    TP_ARGS(crtc, min, max),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     __field(u32, min)
+			     __field(u32, max)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
+										       crtc->pipe);
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   __entry->min = min;
+			   __entry->max = max;
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
+		      pipe_name(__entry->pipe), __entry->frame,
+		       __entry->scanline, __entry->min, __entry->max)
+);
+
+TRACE_EVENT(i915_pipe_update_vblank_evaded,
+	    TP_PROTO(struct intel_crtc *crtc, u32 min, u32 max, u32 frame),
+	    TP_ARGS(crtc, min, max, frame),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     __field(u32, min)
+			     __field(u32, max)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = frame;
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   __entry->min = min;
+			   __entry->max = max;
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
+		      pipe_name(__entry->pipe), __entry->frame,
+		       __entry->scanline, __entry->min, __entry->max)
+);
+
+TRACE_EVENT(i915_pipe_update_end,
+	    TP_PROTO(struct intel_crtc *crtc, u32 frame),
+	    TP_ARGS(crtc, frame),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = frame;
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u",
+		      pipe_name(__entry->pipe), __entry->frame,
+		      __entry->scanline)
+);
+
 /* object tracking */
 
 TRACE_EVENT(i915_gem_object_create,
@@ -251,8 +326,8 @@ TRACE_EVENT(i915_gem_evict_vm,
 );
 
 TRACE_EVENT(i915_gem_ring_sync_to,
-	    TP_PROTO(struct intel_ring_buffer *from,
-		     struct intel_ring_buffer *to,
+	    TP_PROTO(struct intel_engine_cs *from,
+		     struct intel_engine_cs *to,
 		     u32 seqno),
 	    TP_ARGS(from, to, seqno),
 
@@ -277,7 +352,7 @@ TRACE_EVENT(i915_gem_ring_sync_to,
 );
 
 TRACE_EVENT(i915_gem_ring_dispatch,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno, u32 flags),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno, u32 flags),
 	    TP_ARGS(ring, seqno, flags),
 
 	    TP_STRUCT__entry(
@@ -300,7 +375,7 @@ TRACE_EVENT(i915_gem_ring_dispatch,
 );
 
 TRACE_EVENT(i915_gem_ring_flush,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 invalidate, u32 flush),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 invalidate, u32 flush),
 	    TP_ARGS(ring, invalidate, flush),
 
 	    TP_STRUCT__entry(
@@ -323,7 +398,7 @@ TRACE_EVENT(i915_gem_ring_flush,
 );
 
 DECLARE_EVENT_CLASS(i915_gem_request,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno),
 
 	    TP_STRUCT__entry(
@@ -343,12 +418,12 @@ DECLARE_EVENT_CLASS(i915_gem_request,
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno)
 );
 
 TRACE_EVENT(i915_gem_request_complete,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring),
 
 	    TP_STRUCT__entry(
@@ -368,12 +443,12 @@ TRACE_EVENT(i915_gem_request_complete,
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_retire,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno)
 );
 
 TRACE_EVENT(i915_gem_request_wait_begin,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno),
 
 	    TP_STRUCT__entry(
@@ -402,12 +477,12 @@ TRACE_EVENT(i915_gem_request_wait_begin,
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_end,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno)
 );
 
 DECLARE_EVENT_CLASS(i915_ring,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring),
 
 	    TP_STRUCT__entry(
@@ -424,12 +499,12 @@ DECLARE_EVENT_CLASS(i915_ring,
 );
 
 DEFINE_EVENT(i915_ring, i915_ring_wait_begin,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring)
 );
 
 DEFINE_EVENT(i915_ring, i915_ring_wait_end,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring)
 );
 
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index aff4a113cda3..827498e081df 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -49,13 +49,19 @@ find_section(struct bdb_header *bdb, int section_id)
 	total = bdb->bdb_size;
 
 	/* walk the sections looking for section_id */
-	while (index < total) {
+	while (index + 3 < total) {
 		current_id = *(base + index);
 		index++;
+
 		current_size = *((u16 *)(base + index));
 		index += 2;
+
+		if (index + current_size > total)
+			return NULL;
+
 		if (current_id == section_id)
 			return base + index;
+
 		index += current_size;
 	}
 
@@ -206,7 +212,7 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 	const struct lvds_dvo_timing *panel_dvo_timing;
 	const struct lvds_fp_timing *fp_timing;
 	struct drm_display_mode *panel_fixed_mode;
-	int i, downclock;
+	int i, downclock, drrs_mode;
 
 	lvds_options = find_section(bdb, BDB_LVDS_OPTIONS);
 	if (!lvds_options)
@@ -218,6 +224,28 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 
 	panel_type = lvds_options->panel_type;
 
+	drrs_mode = (lvds_options->dps_panel_type_bits
+				>> (panel_type * 2)) & MODE_MASK;
+	/*
+	 * VBT has static DRRS = 0 and seamless DRRS = 2.
+	 * The below piece of code is required to adjust vbt.drrs_type
+	 * to match the enum drrs_support_type.
+	 */
+	switch (drrs_mode) {
+	case 0:
+		dev_priv->vbt.drrs_type = STATIC_DRRS_SUPPORT;
+		DRM_DEBUG_KMS("DRRS supported mode is static\n");
+		break;
+	case 2:
+		dev_priv->vbt.drrs_type = SEAMLESS_DRRS_SUPPORT;
+		DRM_DEBUG_KMS("DRRS supported mode is seamless\n");
+		break;
+	default:
+		dev_priv->vbt.drrs_type = DRRS_NOT_SUPPORTED;
+		DRM_DEBUG_KMS("DRRS not supported (VBT input)\n");
+		break;
+	}
+
 	lvds_lfp_data = find_section(bdb, BDB_LVDS_LFP_DATA);
 	if (!lvds_lfp_data)
 		return;
@@ -287,9 +315,6 @@ parse_lfp_backlight(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 	const struct bdb_lfp_backlight_data *backlight_data;
 	const struct bdb_lfp_backlight_data_entry *entry;
 
-	/* Err to enabling backlight if no backlight block. */
-	dev_priv->vbt.backlight.present = true;
-
 	backlight_data = find_section(bdb, BDB_LVDS_BACKLIGHT);
 	if (!backlight_data)
 		return;
@@ -526,6 +551,16 @@ parse_driver_features(struct drm_i915_private *dev_priv,
 
 	if (driver->dual_frequency)
 		dev_priv->render_reclock_avail = true;
+
+	DRM_DEBUG_KMS("DRRS State Enabled:%d\n", driver->drrs_enabled);
+	/*
+	 * If DRRS is not supported, drrs_type has to be set to 0.
+	 * This is because, VBT is configured in such a way that
+	 * static DRRS is 0 and DRRS not supported is represented by
+	 * driver->drrs_enabled=false
+	 */
+	if (!driver->drrs_enabled)
+		dev_priv->vbt.drrs_type = DRRS_NOT_SUPPORTED;
 }
 
 static void
@@ -628,19 +663,221 @@ parse_edp(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 	}
 }
 
+static u8 *goto_next_sequence(u8 *data, int *size)
+{
+	u16 len;
+	int tmp = *size;
+
+	if (--tmp < 0)
+		return NULL;
+
+	/* goto first element */
+	data++;
+	while (1) {
+		switch (*data) {
+		case MIPI_SEQ_ELEM_SEND_PKT:
+			/*
+			 * skip by this element payload size
+			 * skip elem id, command flag and data type
+			 */
+			tmp -= 5;
+			if (tmp < 0)
+				return NULL;
+
+			data += 3;
+			len = *((u16 *)data);
+
+			tmp -= len;
+			if (tmp < 0)
+				return NULL;
+
+			/* skip by len */
+			data = data + 2 + len;
+			break;
+		case MIPI_SEQ_ELEM_DELAY:
+			/* skip by elem id, and delay is 4 bytes */
+			tmp -= 5;
+			if (tmp < 0)
+				return NULL;
+
+			data += 5;
+			break;
+		case MIPI_SEQ_ELEM_GPIO:
+			tmp -= 3;
+			if (tmp < 0)
+				return NULL;
+
+			data += 3;
+			break;
+		default:
+			DRM_ERROR("Unknown element\n");
+			return NULL;
+		}
+
+		/* end of sequence ? */
+		if (*data == 0)
+			break;
+	}
+
+	/* goto next sequence or end of block byte */
+	if (--tmp < 0)
+		return NULL;
+
+	data++;
+
+	/* update amount of data left for the sequence block to be parsed */
+	*size = tmp;
+	return data;
+}
+
 static void
 parse_mipi(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 {
-	struct bdb_mipi *mipi;
+	struct bdb_mipi_config *start;
+	struct bdb_mipi_sequence *sequence;
+	struct mipi_config *config;
+	struct mipi_pps_data *pps;
+	u8 *data, *seq_data;
+	int i, panel_id, seq_size;
+	u16 block_size;
+
+	/* parse MIPI blocks only if LFP type is MIPI */
+	if (!dev_priv->vbt.has_mipi)
+		return;
+
+	/* Initialize this to undefined indicating no generic MIPI support */
+	dev_priv->vbt.dsi.panel_id = MIPI_DSI_UNDEFINED_PANEL_ID;
 
-	mipi = find_section(bdb, BDB_MIPI_CONFIG);
-	if (!mipi) {
-		DRM_DEBUG_KMS("No MIPI BDB found");
+	/* Block #40 is already parsed and panel_fixed_mode is
+	 * stored in dev_priv->lfp_lvds_vbt_mode
+	 * resuse this when needed
+	 */
+
+	/* Parse #52 for panel index used from panel_type already
+	 * parsed
+	 */
+	start = find_section(bdb, BDB_MIPI_CONFIG);
+	if (!start) {
+		DRM_DEBUG_KMS("No MIPI config BDB found");
 		return;
 	}
 
-	/* XXX: add more info */
+	DRM_DEBUG_DRIVER("Found MIPI Config block, panel index = %d\n",
+								panel_type);
+
+	/*
+	 * get hold of the correct configuration block and pps data as per
+	 * the panel_type as index
+	 */
+	config = &start->config[panel_type];
+	pps = &start->pps[panel_type];
+
+	/* store as of now full data. Trim when we realise all is not needed */
+	dev_priv->vbt.dsi.config = kmemdup(config, sizeof(struct mipi_config), GFP_KERNEL);
+	if (!dev_priv->vbt.dsi.config)
+		return;
+
+	dev_priv->vbt.dsi.pps = kmemdup(pps, sizeof(struct mipi_pps_data), GFP_KERNEL);
+	if (!dev_priv->vbt.dsi.pps) {
+		kfree(dev_priv->vbt.dsi.config);
+		return;
+	}
+
+	/* We have mandatory mipi config blocks. Initialize as generic panel */
 	dev_priv->vbt.dsi.panel_id = MIPI_DSI_GENERIC_PANEL_ID;
+
+	/* Check if we have sequence block as well */
+	sequence = find_section(bdb, BDB_MIPI_SEQUENCE);
+	if (!sequence) {
+		DRM_DEBUG_KMS("No MIPI Sequence found, parsing complete\n");
+		return;
+	}
+
+	DRM_DEBUG_DRIVER("Found MIPI sequence block\n");
+
+	block_size = get_blocksize(sequence);
+
+	/*
+	 * parse the sequence block for individual sequences
+	 */
+	dev_priv->vbt.dsi.seq_version = sequence->version;
+
+	seq_data = &sequence->data[0];
+
+	/*
+	 * sequence block is variable length and hence we need to parse and
+	 * get the sequence data for specific panel id
+	 */
+	for (i = 0; i < MAX_MIPI_CONFIGURATIONS; i++) {
+		panel_id = *seq_data;
+		seq_size = *((u16 *) (seq_data + 1));
+		if (panel_id == panel_type)
+			break;
+
+		/* skip the sequence including seq header of 3 bytes */
+		seq_data = seq_data + 3 + seq_size;
+		if ((seq_data - &sequence->data[0]) > block_size) {
+			DRM_ERROR("Sequence start is beyond sequence block size, corrupted sequence block\n");
+			return;
+		}
+	}
+
+	if (i == MAX_MIPI_CONFIGURATIONS) {
+		DRM_ERROR("Sequence block detected but no valid configuration\n");
+		return;
+	}
+
+	/* check if found sequence is completely within the sequence block
+	 * just being paranoid */
+	if (seq_size > block_size) {
+		DRM_ERROR("Corrupted sequence/size, bailing out\n");
+		return;
+	}
+
+	/* skip the panel id(1 byte) and seq size(2 bytes) */
+	dev_priv->vbt.dsi.data = kmemdup(seq_data + 3, seq_size, GFP_KERNEL);
+	if (!dev_priv->vbt.dsi.data)
+		return;
+
+	/*
+	 * loop into the sequence data and split into multiple sequneces
+	 * There are only 5 types of sequences as of now
+	 */
+	data = dev_priv->vbt.dsi.data;
+	dev_priv->vbt.dsi.size = seq_size;
+
+	/* two consecutive 0x00 indicate end of all sequences */
+	while (1) {
+		int seq_id = *data;
+		if (MIPI_SEQ_MAX > seq_id && seq_id > MIPI_SEQ_UNDEFINED) {
+			dev_priv->vbt.dsi.sequence[seq_id] = data;
+			DRM_DEBUG_DRIVER("Found mipi sequence - %d\n", seq_id);
+		} else {
+			DRM_ERROR("undefined sequence\n");
+			goto err;
+		}
+
+		/* partial parsing to skip elements */
+		data = goto_next_sequence(data, &seq_size);
+
+		if (data == NULL) {
+			DRM_ERROR("Sequence elements going beyond block itself. Sequence block parsing failed\n");
+			goto err;
+		}
+
+		if (*data == 0)
+			break; /* end of sequence reached */
+	}
+
+	DRM_DEBUG_DRIVER("MIPI related vbt parsing complete\n");
+	return;
+err:
+	kfree(dev_priv->vbt.dsi.data);
+	dev_priv->vbt.dsi.data = NULL;
+
+	/* error during parsing so set all pointers to null
+	 * because of partial parsing */
+	memset(dev_priv->vbt.dsi.sequence, 0, MIPI_SEQ_MAX);
 }
 
 static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
@@ -823,6 +1060,15 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 			/* skip the device block if device type is invalid */
 			continue;
 		}
+
+		if (p_child->common.dvo_port >= DVO_PORT_MIPIA
+		    && p_child->common.dvo_port <= DVO_PORT_MIPID
+		    &&p_child->common.device_type & DEVICE_TYPE_MIPI_OUTPUT) {
+			DRM_DEBUG_KMS("Found MIPI as LFP\n");
+			dev_priv->vbt.has_mipi = 1;
+			dev_priv->vbt.dsi.port = p_child->common.dvo_port;
+		}
+
 		child_dev_ptr = dev_priv->vbt.child_dev + count;
 		count++;
 		memcpy((void *)child_dev_ptr, (void *)p_child,
@@ -839,6 +1085,9 @@ init_vbt_defaults(struct drm_i915_private *dev_priv)
 
 	dev_priv->vbt.crt_ddc_pin = GMBUS_PORT_VGADDC;
 
+	/* Default to having backlight */
+	dev_priv->vbt.backlight.present = true;
+
 	/* LFP panel data */
 	dev_priv->vbt.lvds_dither = 1;
 	dev_priv->vbt.lvds_vbt = 0;
@@ -893,6 +1142,46 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = {
 	{ }
 };
 
+static struct bdb_header *validate_vbt(char *base, size_t size,
+				       struct vbt_header *vbt,
+				       const char *source)
+{
+	size_t offset;
+	struct bdb_header *bdb;
+
+	if (vbt == NULL) {
+		DRM_DEBUG_DRIVER("VBT signature missing\n");
+		return NULL;
+	}
+
+	offset = (char *)vbt - base;
+	if (offset + sizeof(struct vbt_header) > size) {
+		DRM_DEBUG_DRIVER("VBT header incomplete\n");
+		return NULL;
+	}
+
+	if (memcmp(vbt->signature, "$VBT", 4)) {
+		DRM_DEBUG_DRIVER("VBT invalid signature\n");
+		return NULL;
+	}
+
+	offset += vbt->bdb_offset;
+	if (offset + sizeof(struct bdb_header) > size) {
+		DRM_DEBUG_DRIVER("BDB header incomplete\n");
+		return NULL;
+	}
+
+	bdb = (struct bdb_header *)(base + offset);
+	if (offset + bdb->bdb_size > size) {
+		DRM_DEBUG_DRIVER("BDB incomplete\n");
+		return NULL;
+	}
+
+	DRM_DEBUG_KMS("Using VBT from %s: %20s\n",
+		      source, vbt->signature);
+	return bdb;
+}
+
 /**
  * intel_parse_bios - find VBT and initialize settings from the BIOS
  * @dev: DRM device
@@ -916,20 +1205,13 @@ intel_parse_bios(struct drm_device *dev)
 	init_vbt_defaults(dev_priv);
 
 	/* XXX Should this validation be moved to intel_opregion.c? */
-	if (!dmi_check_system(intel_no_opregion_vbt) && dev_priv->opregion.vbt) {
-		struct vbt_header *vbt = dev_priv->opregion.vbt;
-		if (memcmp(vbt->signature, "$VBT", 4) == 0) {
-			DRM_DEBUG_KMS("Using VBT from OpRegion: %20s\n",
-					 vbt->signature);
-			bdb = (struct bdb_header *)((char *)vbt + vbt->bdb_offset);
-		} else
-			dev_priv->opregion.vbt = NULL;
-	}
+	if (!dmi_check_system(intel_no_opregion_vbt) && dev_priv->opregion.vbt)
+		bdb = validate_vbt((char *)dev_priv->opregion.header, OPREGION_SIZE,
+				   (struct vbt_header *)dev_priv->opregion.vbt,
+				   "OpRegion");
 
 	if (bdb == NULL) {
-		struct vbt_header *vbt = NULL;
-		size_t size;
-		int i;
+		size_t i, size;
 
 		bios = pci_map_rom(pdev, &size);
 		if (!bios)
@@ -937,19 +1219,18 @@ intel_parse_bios(struct drm_device *dev)
 
 		/* Scour memory looking for the VBT signature */
 		for (i = 0; i + 4 < size; i++) {
-			if (!memcmp(bios + i, "$VBT", 4)) {
-				vbt = (struct vbt_header *)(bios + i);
+			if (memcmp(bios + i, "$VBT", 4) == 0) {
+				bdb = validate_vbt(bios, size,
+						   (struct vbt_header *)(bios + i),
+						   "PCI ROM");
 				break;
 			}
 		}
 
-		if (!vbt) {
-			DRM_DEBUG_DRIVER("VBT signature missing\n");
+		if (!bdb) {
 			pci_unmap_rom(pdev, bios);
 			return -1;
 		}
-
-		bdb = (struct bdb_header *)(bios + i + vbt->bdb_offset);
 	}
 
 	/* Grab useful general definitions */
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index f27f7b282465..b98667796337 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -282,6 +282,9 @@ struct bdb_general_definitions {
 	union child_device_config devices[0];
 } __packed;
 
+/* Mask for DRRS / Panel Channel / SSC / BLT control bits extraction */
+#define MODE_MASK		0x3
+
 struct bdb_lvds_options {
 	u8 panel_type;
 	u8 rsvd1;
@@ -294,6 +297,18 @@ struct bdb_lvds_options {
 	u8 lvds_edid:1;
 	u8 rsvd2:1;
 	u8 rsvd4;
+	/* LVDS Panel channel bits stored here */
+	u32 lvds_panel_channel_bits;
+	/* LVDS SSC (Spread Spectrum Clock) bits stored here. */
+	u16 ssc_bits;
+	u16 ssc_freq;
+	u16 ssc_ddt;
+	/* Panel color depth defined here */
+	u16 panel_color_depth;
+	/* LVDS panel type bits stored here */
+	u32 dps_panel_type_bits;
+	/* LVDS backlight control type bits stored here */
+	u32 blt_control_type_bits;
 } __packed;
 
 /* LFP pointer table contains entries to the struct below */
@@ -482,6 +497,20 @@ struct bdb_driver_features {
 
 	u8 hdmi_termination;
 	u8 custom_vbt_version;
+	/* Driver features data block */
+	u16 rmpm_enabled:1;
+	u16 s2ddt_enabled:1;
+	u16 dpst_enabled:1;
+	u16 bltclt_enabled:1;
+	u16 adb_enabled:1;
+	u16 drrs_enabled:1;
+	u16 grs_enabled:1;
+	u16 gpmt_enabled:1;
+	u16 tbt_enabled:1;
+	u16 psr_enabled:1;
+	u16 ips_enabled:1;
+	u16 reserved3:4;
+	u16 pc_feature_valid:1;
 } __packed;
 
 #define EDP_18BPP	0
@@ -714,6 +743,10 @@ int intel_parse_bios(struct drm_device *dev);
 #define DVO_PORT_DPC	8
 #define DVO_PORT_DPD	9
 #define DVO_PORT_DPA	10
+#define DVO_PORT_MIPIA	21
+#define DVO_PORT_MIPIB	22
+#define DVO_PORT_MIPIC	23
+#define DVO_PORT_MIPID	24
 
 /* Block 52 contains MIPI Panel info
  * 6 such enteries will there. Index into correct
@@ -870,4 +903,35 @@ struct bdb_mipi_sequence {
 	u8 data[0];
 };
 
+/* MIPI Sequnece Block definitions */
+enum mipi_seq {
+	MIPI_SEQ_UNDEFINED = 0,
+	MIPI_SEQ_ASSERT_RESET,
+	MIPI_SEQ_INIT_OTP,
+	MIPI_SEQ_DISPLAY_ON,
+	MIPI_SEQ_DISPLAY_OFF,
+	MIPI_SEQ_DEASSERT_RESET,
+	MIPI_SEQ_MAX
+};
+
+enum mipi_seq_element {
+	MIPI_SEQ_ELEM_UNDEFINED = 0,
+	MIPI_SEQ_ELEM_SEND_PKT,
+	MIPI_SEQ_ELEM_DELAY,
+	MIPI_SEQ_ELEM_GPIO,
+	MIPI_SEQ_ELEM_STATUS,
+	MIPI_SEQ_ELEM_MAX
+};
+
+enum mipi_gpio_pin_index {
+	MIPI_GPIO_UNDEFINED = 0,
+	MIPI_GPIO_PANEL_ENABLE,
+	MIPI_GPIO_BL_ENABLE,
+	MIPI_GPIO_PWM_ENABLE,
+	MIPI_GPIO_RESET_N,
+	MIPI_GPIO_PWR_DOWN_R,
+	MIPI_GPIO_STDBY_RST_N,
+	MIPI_GPIO_MAX
+};
+
 #endif /* _I830_BIOS_H_ */
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index aa5a3dc43342..5a045d3bd77e 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -144,28 +144,49 @@ static void intel_crt_set_dpms(struct intel_encoder *encoder, int mode)
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
-	u32 temp;
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
+	u32 adpa;
+
+	if (INTEL_INFO(dev)->gen >= 5)
+		adpa = ADPA_HOTPLUG_BITS;
+	else
+		adpa = 0;
 
-	temp = I915_READ(crt->adpa_reg);
-	temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE);
-	temp &= ~ADPA_DAC_ENABLE;
+	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
+		adpa |= ADPA_HSYNC_ACTIVE_HIGH;
+	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
+		adpa |= ADPA_VSYNC_ACTIVE_HIGH;
+
+	/* For CPT allow 3 pipe config, for others just use A or B */
+	if (HAS_PCH_LPT(dev))
+		; /* Those bits don't exist here */
+	else if (HAS_PCH_CPT(dev))
+		adpa |= PORT_TRANS_SEL_CPT(crtc->pipe);
+	else if (crtc->pipe == 0)
+		adpa |= ADPA_PIPE_A_SELECT;
+	else
+		adpa |= ADPA_PIPE_B_SELECT;
+
+	if (!HAS_PCH_SPLIT(dev))
+		I915_WRITE(BCLRPAT(crtc->pipe), 0);
 
 	switch (mode) {
 	case DRM_MODE_DPMS_ON:
-		temp |= ADPA_DAC_ENABLE;
+		adpa |= ADPA_DAC_ENABLE;
 		break;
 	case DRM_MODE_DPMS_STANDBY:
-		temp |= ADPA_DAC_ENABLE | ADPA_HSYNC_CNTL_DISABLE;
+		adpa |= ADPA_DAC_ENABLE | ADPA_HSYNC_CNTL_DISABLE;
 		break;
 	case DRM_MODE_DPMS_SUSPEND:
-		temp |= ADPA_DAC_ENABLE | ADPA_VSYNC_CNTL_DISABLE;
+		adpa |= ADPA_DAC_ENABLE | ADPA_VSYNC_CNTL_DISABLE;
 		break;
 	case DRM_MODE_DPMS_OFF:
-		temp |= ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE;
+		adpa |= ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE;
 		break;
 	}
 
-	I915_WRITE(crt->adpa_reg, temp);
+	I915_WRITE(crt->adpa_reg, adpa);
 }
 
 static void intel_disable_crt(struct intel_encoder *encoder)
@@ -274,42 +295,6 @@ static bool intel_crt_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_crt_mode_set(struct intel_encoder *encoder)
-{
-
-	struct drm_device *dev = encoder->base.dev;
-	struct intel_crt *crt = intel_encoder_to_crt(encoder);
-	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
-	u32 adpa;
-
-	if (INTEL_INFO(dev)->gen >= 5)
-		adpa = ADPA_HOTPLUG_BITS;
-	else
-		adpa = 0;
-
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
-		adpa |= ADPA_HSYNC_ACTIVE_HIGH;
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
-		adpa |= ADPA_VSYNC_ACTIVE_HIGH;
-
-	/* For CPT allow 3 pipe config, for others just use A or B */
-	if (HAS_PCH_LPT(dev))
-		; /* Those bits don't exist here */
-	else if (HAS_PCH_CPT(dev))
-		adpa |= PORT_TRANS_SEL_CPT(crtc->pipe);
-	else if (crtc->pipe == 0)
-		adpa |= ADPA_PIPE_A_SELECT;
-	else
-		adpa |= ADPA_PIPE_B_SELECT;
-
-	if (!HAS_PCH_SPLIT(dev))
-		I915_WRITE(BCLRPAT(crtc->pipe), 0);
-
-	I915_WRITE(crt->adpa_reg, adpa);
-}
-
 static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
@@ -645,11 +630,12 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 	enum intel_display_power_domain power_domain;
 	enum drm_connector_status status;
 	struct intel_load_detect_pipe tmp;
+	struct drm_modeset_acquire_ctx ctx;
 
 	intel_runtime_pm_get(dev_priv);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] force=%d\n",
-		      connector->base.id, drm_get_connector_name(connector),
+		      connector->base.id, connector->name,
 		      force);
 
 	power_domain = intel_display_port_power_domain(intel_encoder);
@@ -688,12 +674,12 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 	}
 
 	/* for pre-945g platforms use load detect */
-	if (intel_get_load_detect_pipe(connector, NULL, &tmp)) {
+	if (intel_get_load_detect_pipe(connector, NULL, &tmp, &ctx)) {
 		if (intel_crt_detect_ddc(connector))
 			status = connector_status_connected;
 		else
 			status = intel_crt_load_detect(crt);
-		intel_release_load_detect_pipe(connector, &tmp);
+		intel_release_load_detect_pipe(connector, &tmp, &ctx);
 	} else
 		status = connector_status_unknown;
 
@@ -867,7 +853,6 @@ void intel_crt_init(struct drm_device *dev)
 		crt->adpa_reg = ADPA;
 
 	crt->base.compute_config = intel_crt_compute_config;
-	crt->base.mode_set = intel_crt_mode_set;
 	crt->base.disable = intel_disable_crt;
 	crt->base.enable = intel_enable_crt;
 	if (I915_HAS_HOTPLUG(dev))
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 0ad4e9600063..b17b9c7c769f 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -364,55 +364,6 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 	DRM_ERROR("FDI link training failed!\n");
 }
 
-static void intel_ddi_mode_set(struct intel_encoder *encoder)
-{
-	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	int port = intel_ddi_get_encoder_port(encoder);
-	int pipe = crtc->pipe;
-	int type = encoder->type;
-	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
-
-	DRM_DEBUG_KMS("Preparing DDI mode on port %c, pipe %c\n",
-		      port_name(port), pipe_name(pipe));
-
-	crtc->eld_vld = false;
-	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
-		struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
-		struct intel_digital_port *intel_dig_port =
-			enc_to_dig_port(&encoder->base);
-
-		intel_dp->DP = intel_dig_port->saved_port_bits |
-			       DDI_BUF_CTL_ENABLE | DDI_BUF_EMP_400MV_0DB_HSW;
-		intel_dp->DP |= DDI_PORT_WIDTH(intel_dp->lane_count);
-
-		if (intel_dp->has_audio) {
-			DRM_DEBUG_DRIVER("DP audio on pipe %c on DDI\n",
-					 pipe_name(crtc->pipe));
-
-			/* write eld */
-			DRM_DEBUG_DRIVER("DP audio: write eld information\n");
-			intel_write_eld(&encoder->base, adjusted_mode);
-		}
-	} else if (type == INTEL_OUTPUT_HDMI) {
-		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
-
-		if (intel_hdmi->has_audio) {
-			/* Proper support for digital audio needs a new logic
-			 * and a new set of registers, so we leave it for future
-			 * patch bombing.
-			 */
-			DRM_DEBUG_DRIVER("HDMI audio on pipe %c on DDI\n",
-					 pipe_name(crtc->pipe));
-
-			/* write eld */
-			DRM_DEBUG_DRIVER("HDMI audio: write eld information\n");
-			intel_write_eld(&encoder->base, adjusted_mode);
-		}
-
-		intel_hdmi->set_infoframes(&encoder->base, adjusted_mode);
-	}
-}
-
 static struct intel_encoder *
 intel_ddi_get_crtc_encoder(struct drm_crtc *crtc)
 {
@@ -1062,9 +1013,7 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 	}
 
 	if (type == INTEL_OUTPUT_HDMI) {
-		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
-
-		if (intel_hdmi->has_hdmi_sink)
+		if (intel_crtc->config.has_hdmi_sink)
 			temp |= TRANS_DDI_MODE_SELECT_HDMI;
 		else
 			temp |= TRANS_DDI_MODE_SELECT_DVI;
@@ -1293,28 +1242,48 @@ void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc)
 static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_crtc *crtc = encoder->crtc;
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc *crtc = to_intel_crtc(encoder->crtc);
 	enum port port = intel_ddi_get_encoder_port(intel_encoder);
 	int type = intel_encoder->type;
 
+	if (crtc->config.has_audio) {
+		DRM_DEBUG_DRIVER("Audio on pipe %c on DDI\n",
+				 pipe_name(crtc->pipe));
+
+		/* write eld */
+		DRM_DEBUG_DRIVER("DDI audio: write eld information\n");
+		intel_write_eld(encoder, &crtc->config.adjusted_mode);
+	}
+
 	if (type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 		intel_edp_panel_on(intel_dp);
 	}
 
-	WARN_ON(intel_crtc->ddi_pll_sel == PORT_CLK_SEL_NONE);
-	I915_WRITE(PORT_CLK_SEL(port), intel_crtc->ddi_pll_sel);
+	WARN_ON(crtc->ddi_pll_sel == PORT_CLK_SEL_NONE);
+	I915_WRITE(PORT_CLK_SEL(port), crtc->ddi_pll_sel);
 
 	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+		struct intel_digital_port *intel_dig_port =
+			enc_to_dig_port(encoder);
+
+		intel_dp->DP = intel_dig_port->saved_port_bits |
+			       DDI_BUF_CTL_ENABLE | DDI_BUF_EMP_400MV_0DB_HSW;
+		intel_dp->DP |= DDI_PORT_WIDTH(intel_dp->lane_count);
 
 		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
 		intel_dp_start_link_train(intel_dp);
 		intel_dp_complete_link_train(intel_dp);
 		if (port != PORT_A)
 			intel_dp_stop_link_train(intel_dp);
+	} else if (type == INTEL_OUTPUT_HDMI) {
+		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+
+		intel_hdmi->set_infoframes(encoder,
+					   crtc->config.has_hdmi_sink,
+					   &crtc->config.adjusted_mode);
 	}
 }
 
@@ -1385,7 +1354,8 @@ static void intel_enable_ddi(struct intel_encoder *intel_encoder)
 		intel_edp_psr_enable(intel_dp);
 	}
 
-	if (intel_crtc->eld_vld && type != INTEL_OUTPUT_EDP) {
+	if (intel_crtc->config.has_audio) {
+		intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
 		tmp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
 		tmp |= ((AUDIO_OUTPUT_ENABLE_A | AUDIO_ELD_VALID_A) << (pipe * 4));
 		I915_WRITE(HSW_AUD_PIN_ELD_CP_VLD, tmp);
@@ -1403,11 +1373,14 @@ static void intel_disable_ddi(struct intel_encoder *intel_encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
 
-	if (intel_crtc->eld_vld && type != INTEL_OUTPUT_EDP) {
+	/* We can't touch HSW_AUD_PIN_ELD_CP_VLD uncionditionally because this
+	 * register is part of the power well on Haswell. */
+	if (intel_crtc->config.has_audio) {
 		tmp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
 		tmp &= ~((AUDIO_OUTPUT_ENABLE_A | AUDIO_ELD_VALID_A) <<
 			 (pipe * 4));
 		I915_WRITE(HSW_AUD_PIN_ELD_CP_VLD, tmp);
+		intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO);
 	}
 
 	if (type == INTEL_OUTPUT_EDP) {
@@ -1580,6 +1553,7 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 
 	switch (temp & TRANS_DDI_MODE_SELECT_MASK) {
 	case TRANS_DDI_MODE_SELECT_HDMI:
+		pipe_config->has_hdmi_sink = true;
 	case TRANS_DDI_MODE_SELECT_DVI:
 	case TRANS_DDI_MODE_SELECT_FDI:
 		break;
@@ -1592,6 +1566,12 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 		break;
 	}
 
+	if (intel_display_power_enabled(dev_priv, POWER_DOMAIN_AUDIO)) {
+		temp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
+		if (temp & (AUDIO_OUTPUT_ENABLE_A << (intel_crtc->pipe * 4)))
+			pipe_config->has_audio = true;
+	}
+
 	if (encoder->type == INTEL_OUTPUT_EDP && dev_priv->vbt.edp_bpp &&
 	    pipe_config->pipe_bpp > dev_priv->vbt.edp_bpp) {
 		/*
@@ -1708,7 +1688,6 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_ddi_compute_config;
-	intel_encoder->mode_set = intel_ddi_mode_set;
 	intel_encoder->enable = intel_enable_ddi;
 	intel_encoder->pre_enable = intel_ddi_pre_enable;
 	intel_encoder->disable = intel_disable_ddi;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 5b60e25baa32..e27e7804c0b9 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -41,6 +41,9 @@
 #include <drm/drm_crtc_helper.h>
 #include <linux/dma_remapping.h>
 
+#define DIV_ROUND_CLOSEST_ULL(ll, d)	\
+	({ unsigned long long _tmp = (ll)+(d)/2; do_div(_tmp, d); _tmp; })
+
 static void intel_increase_pllclock(struct drm_crtc *crtc);
 static void intel_crtc_update_cursor(struct drm_crtc *crtc, bool on);
 
@@ -55,6 +58,15 @@ static int intel_framebuffer_init(struct drm_device *dev,
 				  struct intel_framebuffer *ifb,
 				  struct drm_mode_fb_cmd2 *mode_cmd,
 				  struct drm_i915_gem_object *obj);
+static void intel_dp_set_m_n(struct intel_crtc *crtc);
+static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc);
+static void intel_set_pipe_timings(struct intel_crtc *intel_crtc);
+static void intel_cpu_transcoder_set_m_n(struct intel_crtc *crtc,
+					 struct intel_link_m_n *m_n);
+static void ironlake_set_pipeconf(struct drm_crtc *crtc);
+static void haswell_set_pipeconf(struct drm_crtc *crtc);
+static void intel_set_pipe_csc(struct drm_crtc *crtc);
+static void vlv_prepare_pll(struct intel_crtc *crtc);
 
 typedef struct {
 	int	min, max;
@@ -328,6 +340,22 @@ static const intel_limit_t intel_limits_vlv = {
 	.p2 = { .p2_slow = 2, .p2_fast = 20 }, /* slow=min, fast=max */
 };
 
+static const intel_limit_t intel_limits_chv = {
+	/*
+	 * These are the data rate limits (measured in fast clocks)
+	 * since those are the strictest limits we have.  The fast
+	 * clock and actual rate limits are more relaxed, so checking
+	 * them would make no difference.
+	 */
+	.dot = { .min = 25000 * 5, .max = 540000 * 5},
+	.vco = { .min = 4860000, .max = 6700000 },
+	.n = { .min = 1, .max = 1 },
+	.m1 = { .min = 2, .max = 2 },
+	.m2 = { .min = 24 << 22, .max = 175 << 22 },
+	.p1 = { .min = 2, .max = 4 },
+	.p2 = {	.p2_slow = 1, .p2_fast = 14 },
+};
+
 static void vlv_clock(int refclk, intel_clock_t *clock)
 {
 	clock->m = clock->m1 * clock->m2;
@@ -412,6 +440,8 @@ static const intel_limit_t *intel_limit(struct drm_crtc *crtc, int refclk)
 			limit = &intel_limits_pineview_lvds;
 		else
 			limit = &intel_limits_pineview_sdvo;
+	} else if (IS_CHERRYVIEW(dev)) {
+		limit = &intel_limits_chv;
 	} else if (IS_VALLEYVIEW(dev)) {
 		limit = &intel_limits_vlv;
 	} else if (!IS_GEN2(dev)) {
@@ -456,6 +486,17 @@ static void i9xx_clock(int refclk, intel_clock_t *clock)
 	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
 }
 
+static void chv_clock(int refclk, intel_clock_t *clock)
+{
+	clock->m = clock->m1 * clock->m2;
+	clock->p = clock->p1 * clock->p2;
+	if (WARN_ON(clock->n == 0 || clock->p == 0))
+		return;
+	clock->vco = DIV_ROUND_CLOSEST_ULL((uint64_t)refclk * clock->m,
+			clock->n << 22);
+	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
+}
+
 #define INTELPllInvalid(s)   do { /* DRM_DEBUG(s); */ return false; } while (0)
 /**
  * Returns whether the given set of divisors are valid for a given refclk with
@@ -731,6 +772,58 @@ vlv_find_best_dpll(const intel_limit_t *limit, struct drm_crtc *crtc,
 	return found;
 }
 
+static bool
+chv_find_best_dpll(const intel_limit_t *limit, struct drm_crtc *crtc,
+		   int target, int refclk, intel_clock_t *match_clock,
+		   intel_clock_t *best_clock)
+{
+	struct drm_device *dev = crtc->dev;
+	intel_clock_t clock;
+	uint64_t m2;
+	int found = false;
+
+	memset(best_clock, 0, sizeof(*best_clock));
+
+	/*
+	 * Based on hardware doc, the n always set to 1, and m1 always
+	 * set to 2.  If requires to support 200Mhz refclk, we need to
+	 * revisit this because n may not 1 anymore.
+	 */
+	clock.n = 1, clock.m1 = 2;
+	target *= 5;	/* fast clock */
+
+	for (clock.p1 = limit->p1.max; clock.p1 >= limit->p1.min; clock.p1--) {
+		for (clock.p2 = limit->p2.p2_fast;
+				clock.p2 >= limit->p2.p2_slow;
+				clock.p2 -= clock.p2 > 10 ? 2 : 1) {
+
+			clock.p = clock.p1 * clock.p2;
+
+			m2 = DIV_ROUND_CLOSEST_ULL(((uint64_t)target * clock.p *
+					clock.n) << 22, refclk * clock.m1);
+
+			if (m2 > INT_MAX/clock.m1)
+				continue;
+
+			clock.m2 = m2;
+
+			chv_clock(refclk, &clock);
+
+			if (!intel_PLL_is_valid(dev, limit, &clock))
+				continue;
+
+			/* based on hardware requirement, prefer bigger p
+			 */
+			if (clock.p > best_clock->p) {
+				*best_clock = clock;
+				found = true;
+			}
+		}
+	}
+
+	return found;
+}
+
 bool intel_crtc_active(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -765,7 +858,7 @@ static void g4x_wait_for_vblank(struct drm_device *dev, int pipe)
 	frame = I915_READ(frame_reg);
 
 	if (wait_for(I915_READ_NOTRACE(frame_reg) != frame, 50))
-		DRM_DEBUG_KMS("vblank wait timed out\n");
+		WARN(1, "vblank wait timed out\n");
 }
 
 /**
@@ -878,7 +971,7 @@ bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 	u32 bit;
 
 	if (HAS_PCH_IBX(dev_priv->dev)) {
-		switch(port->port) {
+		switch (port->port) {
 		case PORT_B:
 			bit = SDE_PORTB_HOTPLUG;
 			break;
@@ -892,7 +985,7 @@ bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 			return true;
 		}
 	} else {
-		switch(port->port) {
+		switch (port->port) {
 		case PORT_B:
 			bit = SDE_PORTB_HOTPLUG_CPT;
 			break;
@@ -1097,10 +1190,8 @@ static void assert_cursor(struct drm_i915_private *dev_priv,
 
 	if (IS_845G(dev) || IS_I865G(dev))
 		cur_state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
-	else if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev))
-		cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 	else
-		cur_state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
+		cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 
 	WARN(cur_state != state,
 	     "cursor on pipe %c assertion failure (expected %s, current %s)\n",
@@ -1253,6 +1344,9 @@ static bool dp_pipe_enabled(struct drm_i915_private *dev_priv,
 		u32	trans_dp_ctl = I915_READ(trans_dp_ctl_reg);
 		if ((trans_dp_ctl & TRANS_DP_PORT_SEL_MASK) != port_sel)
 			return false;
+	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+		if ((val & DP_PIPE_MASK_CHV) != DP_PIPE_SELECT_CHV(pipe))
+			return false;
 	} else {
 		if ((val & DP_PIPE_MASK) != (pipe << 30))
 			return false;
@@ -1269,6 +1363,9 @@ static bool hdmi_pipe_enabled(struct drm_i915_private *dev_priv,
 	if (HAS_PCH_CPT(dev_priv->dev)) {
 		if ((val & SDVO_PIPE_SEL_MASK_CPT) != SDVO_PIPE_SEL_CPT(pipe))
 			return false;
+	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+		if ((val & SDVO_PIPE_SEL_MASK_CHV) != SDVO_PIPE_SEL_CHV(pipe))
+			return false;
 	} else {
 		if ((val & SDVO_PIPE_SEL_MASK) != SDVO_PIPE_SEL(pipe))
 			return false;
@@ -1367,7 +1464,17 @@ static void intel_init_dpio(struct drm_device *dev)
 	if (!IS_VALLEYVIEW(dev))
 		return;
 
-	DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
+	/*
+	 * IOSF_PORT_DPIO is used for VLV x2 PHY (DP/HDMI B and C),
+	 * CHV x1 PHY (DP/HDMI D)
+	 * IOSF_PORT_DPIO_2 is used for CHV x2 PHY (DP/HDMI B and C)
+	 */
+	if (IS_CHERRYVIEW(dev)) {
+		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO_2;
+		DPIO_PHY_IOSF_PORT(DPIO_PHY1) = IOSF_PORT_DPIO;
+	} else {
+		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
+	}
 }
 
 static void intel_reset_dpio(struct drm_device *dev)
@@ -1377,25 +1484,48 @@ static void intel_reset_dpio(struct drm_device *dev)
 	if (!IS_VALLEYVIEW(dev))
 		return;
 
-	/*
-	 * Enable the CRI clock source so we can get at the display and the
-	 * reference clock for VGA hotplug / manual detection.
-	 */
-	I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
-		   DPLL_REFA_CLK_ENABLE_VLV |
-		   DPLL_INTEGRATED_CRI_CLK_VLV);
+	if (IS_CHERRYVIEW(dev)) {
+		enum dpio_phy phy;
+		u32 val;
 
-	/*
-	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
-	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
-	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
-	 *   b.	The other bits such as sfr settings / modesel may all be set
-	 *      to 0.
-	 *
-	 * This should only be done on init and resume from S3 with both
-	 * PLLs disabled, or we risk losing DPIO and PLL synchronization.
-	 */
-	I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
+		for (phy = DPIO_PHY0; phy < I915_NUM_PHYS_VLV; phy++) {
+			/* Poll for phypwrgood signal */
+			if (wait_for(I915_READ(DISPLAY_PHY_STATUS) &
+						PHY_POWERGOOD(phy), 1))
+				DRM_ERROR("Display PHY %d is not power up\n", phy);
+
+			/*
+			 * Deassert common lane reset for PHY.
+			 *
+			 * This should only be done on init and resume from S3
+			 * with both PLLs disabled, or we risk losing DPIO and
+			 * PLL synchronization.
+			 */
+			val = I915_READ(DISPLAY_PHY_CONTROL);
+			I915_WRITE(DISPLAY_PHY_CONTROL,
+				PHY_COM_LANE_RESET_DEASSERT(phy, val));
+		}
+
+	} else {
+		/*
+		 * If DPIO has already been reset, e.g. by BIOS, just skip all
+		 * this.
+		 */
+		if (I915_READ(DPIO_CTL) & DPIO_CMNRST)
+			return;
+
+		/*
+		 * From VLV2A0_DP_eDP_HDMI_DPIO_driver_vbios_notes_11.docx:
+		 * Need to assert and de-assert PHY SB reset by gating the
+		 * common lane power, then un-gating it.
+		 * Simply ungating isn't enough to reset the PHY enough to get
+		 * ports and lanes running.
+		 */
+		__vlv_set_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC,
+				     false);
+		__vlv_set_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC,
+				     true);
+	}
 }
 
 static void vlv_enable_pll(struct intel_crtc *crtc)
@@ -1436,6 +1566,44 @@ static void vlv_enable_pll(struct intel_crtc *crtc)
 	udelay(150); /* wait for warmup */
 }
 
+static void chv_enable_pll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = crtc->pipe;
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	u32 tmp;
+
+	assert_pipe_disabled(dev_priv, crtc->pipe);
+
+	BUG_ON(!IS_CHERRYVIEW(dev_priv->dev));
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Enable back the 10bit clock to display controller */
+	tmp = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port));
+	tmp |= DPIO_DCLKP_EN;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port), tmp);
+
+	/*
+	 * Need to wait > 100ns between dclkp clock enable bit and PLL enable.
+	 */
+	udelay(1);
+
+	/* Enable PLL */
+	I915_WRITE(DPLL(pipe), crtc->config.dpll_hw_state.dpll);
+
+	/* Check PLL is locked */
+	if (wait_for(((I915_READ(DPLL(pipe)) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1))
+		DRM_ERROR("PLL %d failed to lock\n", pipe);
+
+	/* not sure when this should be written */
+	I915_WRITE(DPLL_MD(pipe), crtc->config.dpll_hw_state.dpll_md);
+	POSTING_READ(DPLL_MD(pipe));
+
+	mutex_unlock(&dev_priv->dpio_lock);
+}
+
 static void i9xx_enable_pll(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -1519,45 +1687,92 @@ static void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 		val = DPLL_INTEGRATED_CRI_CLK_VLV | DPLL_REFA_CLK_ENABLE_VLV;
 	I915_WRITE(DPLL(pipe), val);
 	POSTING_READ(DPLL(pipe));
+
+}
+
+static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
+{
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	u32 val;
+
+	/* Make sure the pipe isn't still relying on us */
+	assert_pipe_disabled(dev_priv, pipe);
+
+	/* Set PLL en = 0 */
+	val = DPLL_SSC_REF_CLOCK_CHV;
+	if (pipe != PIPE_A)
+		val |= DPLL_INTEGRATED_CRI_CLK_VLV;
+	I915_WRITE(DPLL(pipe), val);
+	POSTING_READ(DPLL(pipe));
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Disable 10bit clock to display controller */
+	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port));
+	val &= ~DPIO_DCLKP_EN;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port), val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 		struct intel_digital_port *dport)
 {
 	u32 port_mask;
+	int dpll_reg;
 
 	switch (dport->port) {
 	case PORT_B:
 		port_mask = DPLL_PORTB_READY_MASK;
+		dpll_reg = DPLL(0);
 		break;
 	case PORT_C:
 		port_mask = DPLL_PORTC_READY_MASK;
+		dpll_reg = DPLL(0);
+		break;
+	case PORT_D:
+		port_mask = DPLL_PORTD_READY_MASK;
+		dpll_reg = DPIO_PHY_STATUS;
 		break;
 	default:
 		BUG();
 	}
 
-	if (wait_for((I915_READ(DPLL(0)) & port_mask) == 0, 1000))
+	if (wait_for((I915_READ(dpll_reg) & port_mask) == 0, 1000))
 		WARN(1, "timed out waiting for port %c ready: 0x%08x\n",
-		     port_name(dport->port), I915_READ(DPLL(0)));
+		     port_name(dport->port), I915_READ(dpll_reg));
+}
+
+static void intel_prepare_shared_dpll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
+
+	WARN_ON(!pll->refcount);
+	if (pll->active == 0) {
+		DRM_DEBUG_DRIVER("setting up %s\n", pll->name);
+		WARN_ON(pll->on);
+		assert_shared_dpll_disabled(dev_priv, pll);
+
+		pll->mode_set(dev_priv, pll);
+	}
 }
 
 /**
- * ironlake_enable_shared_dpll - enable PCH PLL
+ * intel_enable_shared_dpll - enable PCH PLL
  * @dev_priv: i915 private structure
  * @pipe: pipe PLL to enable
  *
  * The PCH PLL needs to be enabled before the PCH transcoder, since it
  * drives the transcoder clock.
  */
-static void ironlake_enable_shared_dpll(struct intel_crtc *crtc)
+static void intel_enable_shared_dpll(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
 
-	/* PCH PLLs only available on ILK, SNB and IVB */
-	BUG_ON(INTEL_INFO(dev)->gen < 5);
 	if (WARN_ON(pll == NULL))
 		return;
 
@@ -1804,16 +2019,6 @@ static void intel_enable_pipe(struct intel_crtc *crtc)
 
 	I915_WRITE(reg, val | PIPECONF_ENABLE);
 	POSTING_READ(reg);
-
-	/*
-	 * There's no guarantee the pipe will really start running now. It
-	 * depends on the Gen, the output type and the relative order between
-	 * pipe and plane enabling. Avoid waiting on HSW+ since it's not
-	 * necessary.
-	 * TODO: audit the previous gens.
-	 */
-	if (INTEL_INFO(dev)->gen <= 7 && !IS_HASWELL(dev))
-		intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
 /**
@@ -1882,6 +2087,7 @@ void intel_flush_primary_plane(struct drm_i915_private *dev_priv,
 static void intel_enable_primary_hw_plane(struct drm_i915_private *dev_priv,
 					  enum plane plane, enum pipe pipe)
 {
+	struct drm_device *dev = dev_priv->dev;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
 	int reg;
@@ -1890,18 +2096,25 @@ static void intel_enable_primary_hw_plane(struct drm_i915_private *dev_priv,
 	/* If the pipe isn't enabled, we can't pump pixels and may hang */
 	assert_pipe_enabled(dev_priv, pipe);
 
-	WARN(intel_crtc->primary_enabled, "Primary plane already enabled\n");
+	if (intel_crtc->primary_enabled)
+		return;
 
 	intel_crtc->primary_enabled = true;
 
 	reg = DSPCNTR(plane);
 	val = I915_READ(reg);
-	if (val & DISPLAY_PLANE_ENABLE)
-		return;
+	WARN_ON(val & DISPLAY_PLANE_ENABLE);
 
 	I915_WRITE(reg, val | DISPLAY_PLANE_ENABLE);
 	intel_flush_primary_plane(dev_priv, plane);
-	intel_wait_for_vblank(dev_priv->dev, pipe);
+
+	/*
+	 * BDW signals flip done immediately if the plane
+	 * is disabled, even if the plane enable is already
+	 * armed to occur at the next vblank :(
+	 */
+	if (IS_BROADWELL(dev))
+		intel_wait_for_vblank(dev, intel_crtc->pipe);
 }
 
 /**
@@ -1920,18 +2133,17 @@ static void intel_disable_primary_hw_plane(struct drm_i915_private *dev_priv,
 	int reg;
 	u32 val;
 
-	WARN(!intel_crtc->primary_enabled, "Primary plane already disabled\n");
+	if (!intel_crtc->primary_enabled)
+		return;
 
 	intel_crtc->primary_enabled = false;
 
 	reg = DSPCNTR(plane);
 	val = I915_READ(reg);
-	if ((val & DISPLAY_PLANE_ENABLE) == 0)
-		return;
+	WARN_ON((val & DISPLAY_PLANE_ENABLE) == 0);
 
 	I915_WRITE(reg, val & ~DISPLAY_PLANE_ENABLE);
 	intel_flush_primary_plane(dev_priv, plane);
-	intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
 static bool need_vtd_wa(struct drm_device *dev)
@@ -1954,7 +2166,7 @@ static int intel_align_height(struct drm_device *dev, int height, bool tiled)
 int
 intel_pin_and_fence_fb_obj(struct drm_device *dev,
 			   struct drm_i915_gem_object *obj,
-			   struct intel_ring_buffer *pipelined)
+			   struct intel_engine_cs *pipelined)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 alignment;
@@ -2134,7 +2346,7 @@ static void intel_find_plane_obj(struct intel_crtc *intel_crtc,
 	 * Failed to alloc the obj, check to see if we should share
 	 * an fb with another CRTC instead
 	 */
-	list_for_each_entry(c, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, c) {
 		i = to_intel_crtc(c);
 
 		if (c == &intel_crtc->base)
@@ -2152,9 +2364,9 @@ static void intel_find_plane_obj(struct intel_crtc *intel_crtc,
 	}
 }
 
-static int i9xx_update_primary_plane(struct drm_crtc *crtc,
-				     struct drm_framebuffer *fb,
-				     int x, int y)
+static void i9xx_update_primary_plane(struct drm_crtc *crtc,
+				      struct drm_framebuffer *fb,
+				      int x, int y)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2166,15 +2378,6 @@ static int i9xx_update_primary_plane(struct drm_crtc *crtc,
 	u32 dspcntr;
 	u32 reg;
 
-	switch (plane) {
-	case 0:
-	case 1:
-		break;
-	default:
-		DRM_ERROR("Can't update plane %c in SAREA\n", plane_name(plane));
-		return -EINVAL;
-	}
-
 	intel_fb = to_intel_framebuffer(fb);
 	obj = intel_fb->obj;
 
@@ -2249,13 +2452,11 @@ static int i9xx_update_primary_plane(struct drm_crtc *crtc,
 	} else
 		I915_WRITE(DSPADDR(plane), i915_gem_obj_ggtt_offset(obj) + linear_offset);
 	POSTING_READ(reg);
-
-	return 0;
 }
 
-static int ironlake_update_primary_plane(struct drm_crtc *crtc,
-					 struct drm_framebuffer *fb,
-					 int x, int y)
+static void ironlake_update_primary_plane(struct drm_crtc *crtc,
+					  struct drm_framebuffer *fb,
+					  int x, int y)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2267,16 +2468,6 @@ static int ironlake_update_primary_plane(struct drm_crtc *crtc,
 	u32 dspcntr;
 	u32 reg;
 
-	switch (plane) {
-	case 0:
-	case 1:
-	case 2:
-		break;
-	default:
-		DRM_ERROR("Can't update plane %c in SAREA\n", plane_name(plane));
-		return -EINVAL;
-	}
-
 	intel_fb = to_intel_framebuffer(fb);
 	obj = intel_fb->obj;
 
@@ -2343,8 +2534,6 @@ static int ironlake_update_primary_plane(struct drm_crtc *crtc,
 		I915_WRITE(DSPLINOFF(plane), linear_offset);
 	}
 	POSTING_READ(reg);
-
-	return 0;
 }
 
 /* Assume fb object is pinned & idle & fenced and just update base pointers */
@@ -2359,7 +2548,9 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		dev_priv->display.disable_fbc(dev);
 	intel_increase_pllclock(crtc);
 
-	return dev_priv->display.update_primary_plane(crtc, fb, x, y);
+	dev_priv->display.update_primary_plane(crtc, fb, x, y);
+
+	return 0;
 }
 
 void intel_display_handle_reset(struct drm_device *dev)
@@ -2381,7 +2572,7 @@ void intel_display_handle_reset(struct drm_device *dev)
 	 * pending_flip_queue really got woken up.
 	 */
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 		enum plane plane = intel_crtc->plane;
 
@@ -2389,10 +2580,10 @@ void intel_display_handle_reset(struct drm_device *dev)
 		intel_finish_page_flip_plane(dev, plane);
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-		mutex_lock(&crtc->mutex);
+		drm_modeset_lock(&crtc->mutex, NULL);
 		/*
 		 * FIXME: Once we have proper support for primary planes (and
 		 * disabling them without disabling the entire crtc) allow again
@@ -2403,7 +2594,7 @@ void intel_display_handle_reset(struct drm_device *dev)
 							       crtc->primary->fb,
 							       crtc->x,
 							       crtc->y);
-		mutex_unlock(&crtc->mutex);
+		drm_modeset_unlock(&crtc->mutex);
 	}
 }
 
@@ -2518,14 +2709,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		intel_crtc->config.pipe_src_h = adjusted_mode->crtc_vdisplay;
 	}
 
-	ret = dev_priv->display.update_primary_plane(crtc, fb, x, y);
-	if (ret) {
-		mutex_lock(&dev->struct_mutex);
-		intel_unpin_fb_obj(to_intel_framebuffer(fb)->obj);
-		mutex_unlock(&dev->struct_mutex);
-		DRM_ERROR("failed to update base address\n");
-		return ret;
-	}
+	dev_priv->display.update_primary_plane(crtc, fb, x, y);
 
 	old_fb = crtc->primary->fb;
 	crtc->primary->fb = fb;
@@ -2628,12 +2812,10 @@ static void ironlake_fdi_link_train(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	u32 reg, temp, tries;
 
-	/* FDI needs bits from pipe & plane first */
+	/* FDI needs bits from pipe first */
 	assert_pipe_enabled(dev_priv, pipe);
-	assert_plane_enabled(dev_priv, plane);
 
 	/* Train 1: umask FDI RX Interrupt symbol_lock and bit_lock bit
 	   for train result */
@@ -3064,9 +3246,8 @@ static void ironlake_fdi_disable(struct drm_crtc *crtc)
 	udelay(100);
 
 	/* Ironlake workaround, disable clock pointer after downing FDI */
-	if (HAS_PCH_IBX(dev)) {
+	if (HAS_PCH_IBX(dev))
 		I915_WRITE(FDI_RX_CHICKEN(pipe), FDI_RX_PHASE_SYNC_POINTER_OVR);
-	}
 
 	/* still set train pattern 1 */
 	reg = FDI_TX_CTL(pipe);
@@ -3104,7 +3285,7 @@ bool intel_has_pending_fb_unpin(struct drm_device *dev)
 	 * cannot claim and pin a new fb without at least acquring the
 	 * struct_mutex and so serialising with us.
 	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (atomic_read(&crtc->unpin_work_count) == 0)
 			continue;
 
@@ -3117,7 +3298,7 @@ bool intel_has_pending_fb_unpin(struct drm_device *dev)
 	return false;
 }
 
-static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
+void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3127,8 +3308,9 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
 
 	WARN_ON(waitqueue_active(&dev_priv->pending_flip_queue));
 
-	wait_event(dev_priv->pending_flip_queue,
-		   !intel_crtc_has_pending_flip(crtc));
+	WARN_ON(wait_event_timeout(dev_priv->pending_flip_queue,
+				   !intel_crtc_has_pending_flip(crtc),
+				   60*HZ) == 0);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_finish_fb(crtc->primary->fb);
@@ -3341,7 +3523,7 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 	 * Note that enable_shared_dpll tries to do the right thing, but
 	 * get_shared_dpll unconditionally resets the pll - we need that to have
 	 * the right LVDS enable sequence. */
-	ironlake_enable_shared_dpll(intel_crtc);
+	intel_enable_shared_dpll(intel_crtc);
 
 	/* set transcoder timing, panel must allow it */
 	assert_panel_unlocked(dev_priv, pipe);
@@ -3445,6 +3627,8 @@ static struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc)
 		DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
 			      crtc->base.base.id, pll->name);
 
+		WARN_ON(pll->refcount);
+
 		goto found;
 	}
 
@@ -3478,20 +3662,13 @@ static struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc)
 	return NULL;
 
 found:
+	if (pll->refcount == 0)
+		pll->hw_state = crtc->config.dpll_hw_state;
+
 	crtc->config.shared_dpll = i;
 	DRM_DEBUG_DRIVER("using %s for pipe %c\n", pll->name,
 			 pipe_name(crtc->pipe));
 
-	if (pll->active == 0) {
-		memcpy(&pll->hw_state, &crtc->config.dpll_hw_state,
-		       sizeof(pll->hw_state));
-
-		DRM_DEBUG_DRIVER("setting up %s\n", pll->name);
-		WARN_ON(pll->on);
-		assert_shared_dpll_disabled(dev_priv, pll);
-
-		pll->mode_set(dev_priv, pll);
-	}
 	pll->refcount++;
 
 	return pll;
@@ -3562,17 +3739,17 @@ static void intel_disable_planes(struct drm_crtc *crtc)
 
 void hsw_enable_ips(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (!crtc->config.ips_enabled)
 		return;
 
-	/* We can only enable IPS after we enable a plane and wait for a vblank.
-	 * We guarantee that the plane is enabled by calling intel_enable_ips
-	 * only after intel_enable_plane. And intel_enable_plane already waits
-	 * for a vblank, so all we need to do here is to enable the IPS bit. */
+	/* We can only enable IPS after we enable a plane and wait for a vblank */
+	intel_wait_for_vblank(dev, crtc->pipe);
+
 	assert_plane_enabled(dev_priv, crtc->plane);
-	if (IS_BROADWELL(crtc->base.dev)) {
+	if (IS_BROADWELL(dev)) {
 		mutex_lock(&dev_priv->rps.hw_lock);
 		WARN_ON(sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0xc0000000));
 		mutex_unlock(&dev_priv->rps.hw_lock);
@@ -3602,10 +3779,13 @@ void hsw_disable_ips(struct intel_crtc *crtc)
 		return;
 
 	assert_plane_enabled(dev_priv, crtc->plane);
-	if (IS_BROADWELL(crtc->base.dev)) {
+	if (IS_BROADWELL(dev)) {
 		mutex_lock(&dev_priv->rps.hw_lock);
 		WARN_ON(sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0));
 		mutex_unlock(&dev_priv->rps.hw_lock);
+		/* wait for pcode to finish disabling IPS, which may take up to 42ms */
+		if (wait_for((I915_READ(IPS_CTL) & IPS_ENABLE) == 0, 42))
+			DRM_ERROR("Timed out waiting for IPS disable\n");
 	} else {
 		I915_WRITE(IPS_CTL, 0);
 		POSTING_READ(IPS_CTL);
@@ -3662,6 +3842,94 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 		hsw_enable_ips(intel_crtc);
 }
 
+static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
+{
+	if (!enable && intel_crtc->overlay) {
+		struct drm_device *dev = intel_crtc->base.dev;
+		struct drm_i915_private *dev_priv = dev->dev_private;
+
+		mutex_lock(&dev->struct_mutex);
+		dev_priv->mm.interruptible = false;
+		(void) intel_overlay_switch_off(intel_crtc->overlay);
+		dev_priv->mm.interruptible = true;
+		mutex_unlock(&dev->struct_mutex);
+	}
+
+	/* Let userspace switch the overlay on again. In most cases userspace
+	 * has to recompute where to put it anyway.
+	 */
+}
+
+/**
+ * i9xx_fixup_plane - ugly workaround for G45 to fire up the hardware
+ * cursor plane briefly if not already running after enabling the display
+ * plane.
+ * This workaround avoids occasional blank screens when self refresh is
+ * enabled.
+ */
+static void
+g4x_fixup_plane(struct drm_i915_private *dev_priv, enum pipe pipe)
+{
+	u32 cntl = I915_READ(CURCNTR(pipe));
+
+	if ((cntl & CURSOR_MODE) == 0) {
+		u32 fw_bcl_self = I915_READ(FW_BLC_SELF);
+
+		I915_WRITE(FW_BLC_SELF, fw_bcl_self & ~FW_BLC_SELF_EN);
+		I915_WRITE(CURCNTR(pipe), CURSOR_MODE_64_ARGB_AX);
+		intel_wait_for_vblank(dev_priv->dev, pipe);
+		I915_WRITE(CURCNTR(pipe), cntl);
+		I915_WRITE(CURBASE(pipe), I915_READ(CURBASE(pipe)));
+		I915_WRITE(FW_BLC_SELF, fw_bcl_self);
+	}
+}
+
+static void intel_crtc_enable_planes(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	int plane = intel_crtc->plane;
+
+	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
+	intel_enable_planes(crtc);
+	/* The fixup needs to happen before cursor is enabled */
+	if (IS_G4X(dev))
+		g4x_fixup_plane(dev_priv, pipe);
+	intel_crtc_update_cursor(crtc, true);
+	intel_crtc_dpms_overlay(intel_crtc, true);
+
+	hsw_enable_ips(intel_crtc);
+
+	mutex_lock(&dev->struct_mutex);
+	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
+	mutex_unlock(&dev->struct_mutex);
+}
+
+static void intel_crtc_disable_planes(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	int plane = intel_crtc->plane;
+
+	intel_crtc_wait_for_pending_flips(crtc);
+	drm_crtc_vblank_off(crtc);
+
+	if (dev_priv->fbc.plane == plane)
+		intel_disable_fbc(dev);
+
+	hsw_disable_ips(intel_crtc);
+
+	intel_crtc_dpms_overlay(intel_crtc, false);
+	intel_crtc_update_cursor(crtc, false);
+	intel_disable_planes(crtc);
+	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
+}
+
 static void ironlake_crtc_enable(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -3669,13 +3937,35 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
+	enum plane plane = intel_crtc->plane;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	if (intel_crtc->config.has_pch_encoder)
+		intel_prepare_shared_dpll(intel_crtc);
+
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	if (intel_crtc->config.has_pch_encoder) {
+		intel_cpu_transcoder_set_m_n(intel_crtc,
+					     &intel_crtc->config.fdi_m_n);
+	}
+
+	ironlake_set_pipeconf(crtc);
+
+	/* Set up the display plane register */
+	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
 	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
@@ -3705,32 +3995,19 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
 
 	if (intel_crtc->config.has_pch_encoder)
 		ironlake_pch_enable(crtc);
 
-	mutex_lock(&dev->struct_mutex);
-	intel_update_fbc(dev);
-	mutex_unlock(&dev->struct_mutex);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
 
 	if (HAS_PCH_CPT(dev))
 		cpt_verify_modeset(dev, intel_crtc->pipe);
 
-	/*
-	 * There seems to be a race in PCH platform hw (at least on some
-	 * outputs) where an enabled pipe still completes any pageflip right
-	 * away (as if the pipe is off) instead of waiting for vblank. As soon
-	 * as the first vblank happend, everything works as expected. Hence just
-	 * wait for one vblank before returning to avoid strange things
-	 * happening.
-	 */
-	intel_wait_for_vblank(dev, intel_crtc->pipe);
+	intel_crtc_enable_planes(crtc);
+
+	drm_crtc_vblank_on(crtc);
 }
 
 /* IPS only exists on ULT machines and is tied to pipe A. */
@@ -3739,47 +4016,6 @@ static bool hsw_crtc_supports_ips(struct intel_crtc *crtc)
 	return HAS_IPS(crtc->base.dev) && crtc->pipe == PIPE_A;
 }
 
-static void haswell_crtc_enable_planes(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
-
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
-
-	hsw_enable_ips(intel_crtc);
-
-	mutex_lock(&dev->struct_mutex);
-	intel_update_fbc(dev);
-	mutex_unlock(&dev->struct_mutex);
-}
-
-static void haswell_crtc_disable_planes(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
-
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
-
-	/* FBC must be disabled before disabling the plane on HSW. */
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
-
-	hsw_disable_ips(intel_crtc);
-
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
-}
-
 /*
  * This implements the workaround described in the "notes" section of the mode
  * set sequence documentation. When going from no pipes or single pipe to
@@ -3793,7 +4029,7 @@ static void haswell_mode_set_planes_workaround(struct intel_crtc *crtc)
 
 	/* We want to get the other_active_crtc only if there's only 1 other
 	 * active crtc. */
-	list_for_each_entry(crtc_it, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc_it) {
 		if (!crtc_it->active || crtc_it == crtc)
 			continue;
 
@@ -3816,12 +4052,34 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
+	enum plane plane = intel_crtc->plane;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	if (intel_crtc->config.has_pch_encoder) {
+		intel_cpu_transcoder_set_m_n(intel_crtc,
+					     &intel_crtc->config.fdi_m_n);
+	}
+
+	haswell_set_pipeconf(crtc);
+
+	intel_set_pipe_csc(crtc);
+
+	/* Set up the display plane register */
+	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE | DISPPLANE_PIPE_CSC_ENABLE);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
 	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
@@ -3862,7 +4120,9 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	/* If we change the relative order between pipe/planes enabling, we need
 	 * to change the workaround. */
 	haswell_mode_set_planes_workaround(intel_crtc);
-	haswell_crtc_enable_planes(crtc);
+	intel_crtc_enable_planes(crtc);
+
+	drm_crtc_vblank_on(crtc);
 }
 
 static void ironlake_pfit_disable(struct intel_crtc *crtc)
@@ -3887,26 +4147,16 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	u32 reg, temp;
 
-
 	if (!intel_crtc->active)
 		return;
 
+	intel_crtc_disable_planes(crtc);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
 
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
-
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
-
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
-
 	if (intel_crtc->config.has_pch_encoder)
 		intel_set_pch_fifo_underrun_reporting(dev, pipe, false);
 
@@ -3950,6 +4200,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
 	mutex_unlock(&dev->struct_mutex);
 }
 
@@ -3965,7 +4216,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	if (!intel_crtc->active)
 		return;
 
-	haswell_crtc_disable_planes(crtc);
+	intel_crtc_disable_planes(crtc);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		intel_opregion_notify_encoder(encoder, false);
@@ -3997,6 +4248,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
 	mutex_unlock(&dev->struct_mutex);
 }
 
@@ -4011,48 +4263,6 @@ static void haswell_crtc_off(struct drm_crtc *crtc)
 	intel_ddi_put_crtc_pll(crtc);
 }
 
-static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
-{
-	if (!enable && intel_crtc->overlay) {
-		struct drm_device *dev = intel_crtc->base.dev;
-		struct drm_i915_private *dev_priv = dev->dev_private;
-
-		mutex_lock(&dev->struct_mutex);
-		dev_priv->mm.interruptible = false;
-		(void) intel_overlay_switch_off(intel_crtc->overlay);
-		dev_priv->mm.interruptible = true;
-		mutex_unlock(&dev->struct_mutex);
-	}
-
-	/* Let userspace switch the overlay on again. In most cases userspace
-	 * has to recompute where to put it anyway.
-	 */
-}
-
-/**
- * i9xx_fixup_plane - ugly workaround for G45 to fire up the hardware
- * cursor plane briefly if not already running after enabling the display
- * plane.
- * This workaround avoids occasional blank screens when self refresh is
- * enabled.
- */
-static void
-g4x_fixup_plane(struct drm_i915_private *dev_priv, enum pipe pipe)
-{
-	u32 cntl = I915_READ(CURCNTR(pipe));
-
-	if ((cntl & CURSOR_MODE) == 0) {
-		u32 fw_bcl_self = I915_READ(FW_BLC_SELF);
-
-		I915_WRITE(FW_BLC_SELF, fw_bcl_self & ~FW_BLC_SELF_EN);
-		I915_WRITE(CURCNTR(pipe), CURSOR_MODE_64_ARGB_AX);
-		intel_wait_for_vblank(dev_priv->dev, pipe);
-		I915_WRITE(CURCNTR(pipe), cntl);
-		I915_WRITE(CURBASE(pipe), I915_READ(CURBASE(pipe)));
-		I915_WRITE(FW_BLC_SELF, fw_bcl_self);
-	}
-}
-
 static void i9xx_pfit_enable(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -4164,7 +4374,7 @@ static void modeset_update_crtc_power_domains(struct drm_device *dev)
 	 * First get all needed power domains, then put all unneeded, to avoid
 	 * any unnecessary toggling of the power wells.
 	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		enum intel_display_power_domain domain;
 
 		if (!crtc->base.enabled)
@@ -4176,7 +4386,7 @@ static void modeset_update_crtc_power_domains(struct drm_device *dev)
 			intel_display_power_get(dev_priv, domain);
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		enum intel_display_power_domain domain;
 
 		for_each_power_domain(domain, crtc->enabled_power_domains)
@@ -4207,6 +4417,9 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, cmd;
 
+	WARN_ON(valleyview_cur_cdclk(dev_priv) != dev_priv->vlv_cdclk_freq);
+	dev_priv->vlv_cdclk_freq = cdclk;
+
 	if (cdclk >= 320) /* jump to highest voltage for 400MHz too */
 		cmd = 2;
 	else if (cdclk == 266)
@@ -4261,7 +4474,7 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	intel_i2c_reset(dev);
 }
 
-static int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
+int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
 {
 	int cur_cdclk, vco;
 	int divider;
@@ -4282,10 +4495,6 @@ static int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
 static int valleyview_calc_cdclk(struct drm_i915_private *dev_priv,
 				 int max_pixclk)
 {
-	int cur_cdclk;
-
-	cur_cdclk = valleyview_cur_cdclk(dev_priv);
-
 	/*
 	 * Really only a few cases to deal with, as only 4 CDclks are supported:
 	 *   200MHz
@@ -4311,8 +4520,7 @@ static int intel_mode_max_pixclk(struct drm_i915_private *dev_priv)
 	struct intel_crtc *intel_crtc;
 	int max_pixclk = 0;
 
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		if (intel_crtc->new_enabled)
 			max_pixclk = max(max_pixclk,
 					 intel_crtc->new_config->adjusted_mode.crtc_clock);
@@ -4327,14 +4535,13 @@ static void valleyview_modeset_global_pipes(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc;
 	int max_pixclk = intel_mode_max_pixclk(dev_priv);
-	int cur_cdclk = valleyview_cur_cdclk(dev_priv);
 
-	if (valleyview_calc_cdclk(dev_priv, max_pixclk) == cur_cdclk)
+	if (valleyview_calc_cdclk(dev_priv, max_pixclk) ==
+	    dev_priv->vlv_cdclk_freq)
 		return;
 
 	/* disable/enable all currently active pipes while we change cdclk */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head)
+	for_each_intel_crtc(dev, intel_crtc)
 		if (intel_crtc->base.enabled)
 			*prepare_pipes |= (1 << intel_crtc->pipe);
 }
@@ -4343,10 +4550,9 @@ static void valleyview_modeset_global_resources(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int max_pixclk = intel_mode_max_pixclk(dev_priv);
-	int cur_cdclk = valleyview_cur_cdclk(dev_priv);
 	int req_cdclk = valleyview_calc_cdclk(dev_priv, max_pixclk);
 
-	if (req_cdclk != cur_cdclk)
+	if (req_cdclk != dev_priv->vlv_cdclk_freq)
 		valleyview_set_cdclk(dev, req_cdclk);
 	modeset_update_crtc_power_domains(dev);
 }
@@ -4360,22 +4566,56 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
 	bool is_dsi;
+	u32 dspcntr;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	is_dsi = intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI);
+
+	if (!is_dsi && !IS_CHERRYVIEW(dev))
+		vlv_prepare_pll(intel_crtc);
+
+	/* Set up the display plane register */
+	dspcntr = DISPPLANE_GAMMA_ENABLE;
+
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	/* pipesrc and dspsize control the size that is scaled from,
+	 * which should always be the user's requested size.
+	 */
+	I915_WRITE(DSPSIZE(plane),
+		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
+		   (intel_crtc->config.pipe_src_w - 1));
+	I915_WRITE(DSPPOS(plane), 0);
+
+	i9xx_set_pipeconf(intel_crtc);
+
+	I915_WRITE(DSPCNTR(plane), dspcntr);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
+	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_pll_enable)
 			encoder->pre_pll_enable(encoder);
 
-	is_dsi = intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI);
-
-	if (!is_dsi)
-		vlv_enable_pll(intel_crtc);
+	if (!is_dsi) {
+		if (IS_CHERRYVIEW(dev))
+			chv_enable_pll(intel_crtc);
+		else
+			vlv_enable_pll(intel_crtc);
+	}
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
@@ -4387,15 +4627,25 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
-	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
-
-	intel_update_fbc(dev);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
+
+	intel_crtc_enable_planes(crtc);
+
+	drm_crtc_vblank_on(crtc);
+
+	/* Underruns don't raise interrupts, so check manually. */
+	i9xx_check_fifo_underruns(dev);
+}
+
+static void i9xx_set_pll_dividers(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(FP0(crtc->pipe), crtc->config.dpll_hw_state.fp0);
+	I915_WRITE(FP1(crtc->pipe), crtc->config.dpll_hw_state.fp1);
 }
 
 static void i9xx_crtc_enable(struct drm_crtc *crtc)
@@ -4406,14 +4656,49 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
+	u32 dspcntr;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	i9xx_set_pll_dividers(intel_crtc);
+
+	/* Set up the display plane register */
+	dspcntr = DISPPLANE_GAMMA_ENABLE;
+
+	if (pipe == 0)
+		dspcntr &= ~DISPPLANE_SEL_PIPE_MASK;
+	else
+		dspcntr |= DISPPLANE_SEL_PIPE_B;
+
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	/* pipesrc and dspsize control the size that is scaled from,
+	 * which should always be the user's requested size.
+	 */
+	I915_WRITE(DSPSIZE(plane),
+		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
+		   (intel_crtc->config.pipe_src_w - 1));
+	I915_WRITE(DSPPOS(plane), 0);
+
+	i9xx_set_pipeconf(intel_crtc);
+
+	I915_WRITE(DSPCNTR(plane), dspcntr);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
+	if (!IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
 			encoder->pre_enable(encoder);
@@ -4426,21 +4711,26 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
-	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	/* The fixup needs to happen before cursor is enabled */
-	if (IS_G4X(dev))
-		g4x_fixup_plane(dev_priv, pipe);
-	intel_crtc_update_cursor(crtc, true);
-
-	/* Give the overlay scaler a chance to enable if it's on this pipe */
-	intel_crtc_dpms_overlay(intel_crtc, true);
-
-	intel_update_fbc(dev);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
+
+	intel_crtc_enable_planes(crtc);
+
+	/*
+	 * Gen2 reports pipe underruns whenever all planes are disabled.
+	 * So don't enable underrun reporting before at least some planes
+	 * are enabled.
+	 * FIXME: Need to fix the logic to work when we turn off all planes
+	 * but leave the pipe running.
+	 */
+	if (IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
+
+	drm_crtc_vblank_on(crtc);
+
+	/* Underruns don't raise interrupts, so check manually. */
+	i9xx_check_fifo_underruns(dev);
 }
 
 static void i9xx_pfit_disable(struct intel_crtc *crtc)
@@ -4465,27 +4755,31 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 
 	if (!intel_crtc->active)
 		return;
 
-	for_each_encoder_on_crtc(dev, crtc, encoder)
-		encoder->disable(encoder);
+	/*
+	 * Gen2 reports pipe underruns whenever all planes are disabled.
+	 * So diasble underrun reporting before all the planes get disabled.
+	 * FIXME: Need to fix the logic to work when we turn off all planes
+	 * but leave the pipe running.
+	 */
+	if (IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 
-	/* Give the overlay scaler a chance to disable if it's on this pipe */
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
+	intel_crtc_disable_planes(crtc);
 
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		encoder->disable(encoder);
 
-	intel_crtc_dpms_overlay(intel_crtc, false);
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
+	/*
+	 * On gen2 planes are double buffered but the pipe isn't, so we must
+	 * wait for planes to fully turn off before disabling the pipe.
+	 */
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
-	intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 	intel_disable_pipe(dev_priv, pipe);
 
 	i9xx_pfit_disable(intel_crtc);
@@ -4494,15 +4788,25 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 		if (encoder->post_disable)
 			encoder->post_disable(encoder);
 
-	if (IS_VALLEYVIEW(dev) && !intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI))
-		vlv_disable_pll(dev_priv, pipe);
-	else if (!IS_VALLEYVIEW(dev))
-		i9xx_disable_pll(dev_priv, pipe);
+	if (!intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI)) {
+		if (IS_CHERRYVIEW(dev))
+			chv_disable_pll(dev_priv, pipe);
+		else if (IS_VALLEYVIEW(dev))
+			vlv_disable_pll(dev_priv, pipe);
+		else
+			i9xx_disable_pll(dev_priv, pipe);
+	}
+
+	if (!IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 
 	intel_crtc->active = false;
 	intel_update_watermarks(crtc);
 
+	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
+	mutex_unlock(&dev->struct_mutex);
 }
 
 static void i9xx_crtc_off(struct drm_crtc *crtc)
@@ -4565,13 +4869,11 @@ static void intel_crtc_disable(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_connector *connector;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 	/* crtc should still be enabled when we disable it. */
 	WARN_ON(!crtc->enabled);
 
 	dev_priv->display.crtc_disable(crtc);
-	intel_crtc->eld_vld = false;
 	intel_crtc_update_sarea(crtc, false);
 	dev_priv->display.off(crtc);
 
@@ -4635,7 +4937,7 @@ static void intel_connector_check_state(struct intel_connector *connector)
 
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 			      connector->base.base.id,
-			      drm_get_connector_name(&connector->base));
+			      connector->base.name);
 
 		WARN(connector->base.dpms == DRM_MODE_DPMS_OFF,
 		     "wrong connector dpms state\n");
@@ -5039,8 +5341,6 @@ static void i9xx_update_pll_dividers(struct intel_crtc *crtc,
 				     intel_clock_t *reduced_clock)
 {
 	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = crtc->pipe;
 	u32 fp, fp2 = 0;
 
 	if (IS_PINEVIEW(dev)) {
@@ -5053,17 +5353,14 @@ static void i9xx_update_pll_dividers(struct intel_crtc *crtc,
 			fp2 = i9xx_dpll_compute_fp(reduced_clock);
 	}
 
-	I915_WRITE(FP0(pipe), fp);
 	crtc->config.dpll_hw_state.fp0 = fp;
 
 	crtc->lowfreq_avail = false;
 	if (intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_LVDS) &&
 	    reduced_clock && i915.powersave) {
-		I915_WRITE(FP1(pipe), fp2);
 		crtc->config.dpll_hw_state.fp1 = fp2;
 		crtc->lowfreq_avail = true;
 	} else {
-		I915_WRITE(FP1(pipe), fp);
 		crtc->config.dpll_hw_state.fp1 = fp;
 	}
 }
@@ -5141,12 +5438,34 @@ static void intel_dp_set_m_n(struct intel_crtc *crtc)
 
 static void vlv_update_pll(struct intel_crtc *crtc)
 {
+	u32 dpll, dpll_md;
+
+	/*
+	 * Enable DPIO clock input. We should never disable the reference
+	 * clock for pipe B, since VGA hotplug / manual detection depends
+	 * on it.
+	 */
+	dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV |
+		DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV;
+	/* We should never disable this, set it here for state tracking */
+	if (crtc->pipe == PIPE_B)
+		dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
+	dpll |= DPLL_VCO_ENABLE;
+	crtc->config.dpll_hw_state.dpll = dpll;
+
+	dpll_md = (crtc->config.pixel_multiplier - 1)
+		<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
+	crtc->config.dpll_hw_state.dpll_md = dpll_md;
+}
+
+static void vlv_prepare_pll(struct intel_crtc *crtc)
+{
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = crtc->pipe;
-	u32 dpll, mdiv;
+	u32 mdiv;
 	u32 bestn, bestm1, bestm2, bestp1, bestp2;
-	u32 coreclk, reg_val, dpll_md;
+	u32 coreclk, reg_val;
 
 	mutex_lock(&dev_priv->dpio_lock);
 
@@ -5159,7 +5478,7 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	/* See eDP HDMI DPIO driver vbios notes doc */
 
 	/* PLL B needs special handling */
-	if (pipe)
+	if (pipe == PIPE_B)
 		vlv_pllb_recal_opamp(dev_priv, pipe);
 
 	/* Set up Tx target for periodic Rcomp update */
@@ -5203,7 +5522,7 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	if (intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_EDP) ||
 	    intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_DISPLAYPORT)) {
 		/* Use SSC source */
-		if (!pipe)
+		if (pipe == PIPE_A)
 			vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW5(pipe),
 					 0x0df40000);
 		else
@@ -5211,7 +5530,7 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 					 0x0df70000);
 	} else { /* HDMI or VGA */
 		/* Use bend source */
-		if (!pipe)
+		if (pipe == PIPE_A)
 			vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW5(pipe),
 					 0x0df70000);
 		else
@@ -5227,26 +5546,84 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW7(pipe), coreclk);
 
 	vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW11(pipe), 0x87871000);
+	mutex_unlock(&dev_priv->dpio_lock);
+}
+
+static void chv_update_pll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = crtc->pipe;
+	int dpll_reg = DPLL(crtc->pipe);
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	u32 loopfilter, intcoeff;
+	u32 bestn, bestm1, bestm2, bestp1, bestp2, bestm2_frac;
+	int refclk;
+
+	crtc->config.dpll_hw_state.dpll = DPLL_SSC_REF_CLOCK_CHV |
+		DPLL_REFA_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS |
+		DPLL_VCO_ENABLE;
+	if (pipe != PIPE_A)
+		crtc->config.dpll_hw_state.dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
+
+	crtc->config.dpll_hw_state.dpll_md =
+		(crtc->config.pixel_multiplier - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT;
+
+	bestn = crtc->config.dpll.n;
+	bestm2_frac = crtc->config.dpll.m2 & 0x3fffff;
+	bestm1 = crtc->config.dpll.m1;
+	bestm2 = crtc->config.dpll.m2 >> 22;
+	bestp1 = crtc->config.dpll.p1;
+	bestp2 = crtc->config.dpll.p2;
 
 	/*
-	 * Enable DPIO clock input. We should never disable the reference
-	 * clock for pipe B, since VGA hotplug / manual detection depends
-	 * on it.
+	 * Enable Refclk and SSC
 	 */
-	dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV |
-		DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV;
-	/* We should never disable this, set it here for state tracking */
-	if (pipe == PIPE_B)
-		dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
-	dpll |= DPLL_VCO_ENABLE;
-	crtc->config.dpll_hw_state.dpll = dpll;
+	I915_WRITE(dpll_reg,
+		   crtc->config.dpll_hw_state.dpll & ~DPLL_VCO_ENABLE);
 
-	dpll_md = (crtc->config.pixel_multiplier - 1)
-		<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
-	crtc->config.dpll_hw_state.dpll_md = dpll_md;
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* p1 and p2 divider */
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW13(port),
+			5 << DPIO_CHV_S1_DIV_SHIFT |
+			bestp1 << DPIO_CHV_P1_DIV_SHIFT |
+			bestp2 << DPIO_CHV_P2_DIV_SHIFT |
+			1 << DPIO_CHV_K_DIV_SHIFT);
+
+	/* Feedback post-divider - m2 */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW0(port), bestm2);
+
+	/* Feedback refclk divider - n and m1 */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW1(port),
+			DPIO_CHV_M1_DIV_BY_2 |
+			1 << DPIO_CHV_N_DIV_SHIFT);
+
+	/* M2 fraction division */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW2(port), bestm2_frac);
+
+	/* M2 fraction division enable */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW3(port),
+		       DPIO_CHV_FRAC_DIV_EN |
+		       (2 << DPIO_CHV_FEEDFWD_GAIN_SHIFT));
+
+	/* Loop filter */
+	refclk = i9xx_get_refclk(&crtc->base, 0);
+	loopfilter = 5 << DPIO_CHV_PROP_COEFF_SHIFT |
+		2 << DPIO_CHV_GAIN_CTRL_SHIFT;
+	if (refclk == 100000)
+		intcoeff = 11;
+	else if (refclk == 38400)
+		intcoeff = 10;
+	else
+		intcoeff = 9;
+	loopfilter |= intcoeff << DPIO_CHV_INT_COEFF_SHIFT;
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW6(port), loopfilter);
 
-	if (crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(crtc);
+	/* AFC Recal */
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port),
+			vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port)) |
+			DPIO_AFC_RECAL);
 
 	mutex_unlock(&dev_priv->dpio_lock);
 }
@@ -5325,9 +5702,6 @@ static void i9xx_update_pll(struct intel_crtc *crtc,
 			<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
 		crtc->config.dpll_hw_state.dpll_md = dpll_md;
 	}
-
-	if (crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(crtc);
 }
 
 static void i8xx_update_pll(struct intel_crtc *crtc,
@@ -5567,16 +5941,12 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	int refclk, num_connectors = 0;
 	intel_clock_t clock, reduced_clock;
-	u32 dspcntr;
 	bool ok, has_reduced_clock = false;
 	bool is_lvds = false, is_dsi = false;
 	struct intel_encoder *encoder;
 	const intel_limit_t *limit;
-	int ret;
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		switch (encoder->type) {
@@ -5592,7 +5962,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	}
 
 	if (is_dsi)
-		goto skip_dpll;
+		return 0;
 
 	if (!intel_crtc->config.clock_set) {
 		refclk = i9xx_get_refclk(crtc, num_connectors);
@@ -5637,43 +6007,17 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 		i8xx_update_pll(intel_crtc,
 				has_reduced_clock ? &reduced_clock : NULL,
 				num_connectors);
+	} else if (IS_CHERRYVIEW(dev)) {
+		chv_update_pll(intel_crtc);
 	} else if (IS_VALLEYVIEW(dev)) {
 		vlv_update_pll(intel_crtc);
 	} else {
 		i9xx_update_pll(intel_crtc,
 				has_reduced_clock ? &reduced_clock : NULL,
-                                num_connectors);
-	}
-
-skip_dpll:
-	/* Set up the display plane register */
-	dspcntr = DISPPLANE_GAMMA_ENABLE;
-
-	if (!IS_VALLEYVIEW(dev)) {
-		if (pipe == 0)
-			dspcntr &= ~DISPPLANE_SEL_PIPE_MASK;
-		else
-			dspcntr |= DISPPLANE_SEL_PIPE_B;
+				num_connectors);
 	}
 
-	intel_set_pipe_timings(intel_crtc);
-
-	/* pipesrc and dspsize control the size that is scaled from,
-	 * which should always be the user's requested size.
-	 */
-	I915_WRITE(DSPSIZE(plane),
-		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
-		   (intel_crtc->config.pipe_src_w - 1));
-	I915_WRITE(DSPPOS(plane), 0);
-
-	i9xx_set_pipeconf(intel_crtc);
-
-	I915_WRITE(DSPCNTR(plane), dspcntr);
-	POSTING_READ(DSPCNTR(plane));
-
-	ret = intel_pipe_set_base(crtc, x, y, fb);
-
-	return ret;
+	return 0;
 }
 
 static void i9xx_get_pfit_config(struct intel_crtc *crtc,
@@ -5793,6 +6137,36 @@ static void i9xx_get_plane_config(struct intel_crtc *crtc,
 
 }
 
+static void chv_crtc_clock_get(struct intel_crtc *crtc,
+			       struct intel_crtc_config *pipe_config)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = pipe_config->cpu_transcoder;
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	intel_clock_t clock;
+	u32 cmn_dw13, pll_dw0, pll_dw1, pll_dw2;
+	int refclk = 100000;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	cmn_dw13 = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW13(port));
+	pll_dw0 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW0(port));
+	pll_dw1 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW1(port));
+	pll_dw2 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW2(port));
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	clock.m1 = (pll_dw1 & 0x7) == DPIO_CHV_M1_DIV_BY_2 ? 2 : 0;
+	clock.m2 = ((pll_dw0 & 0xff) << 22) | (pll_dw2 & 0x3fffff);
+	clock.n = (pll_dw1 >> DPIO_CHV_N_DIV_SHIFT) & 0xf;
+	clock.p1 = (cmn_dw13 >> DPIO_CHV_P1_DIV_SHIFT) & 0x7;
+	clock.p2 = (cmn_dw13 >> DPIO_CHV_P2_DIV_SHIFT) & 0x1f;
+
+	chv_clock(refclk, &clock);
+
+	/* clock.dot is the fast clock */
+	pipe_config->port_clock = clock.dot / 5;
+}
+
 static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 				 struct intel_crtc_config *pipe_config)
 {
@@ -5827,6 +6201,9 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 		}
 	}
 
+	if (IS_VALLEYVIEW(dev) && (tmp & PIPECONF_COLOR_RANGE_SELECT))
+		pipe_config->limited_color_range = true;
+
 	if (INTEL_INFO(dev)->gen < 4)
 		pipe_config->double_wide = tmp & PIPECONF_DOUBLE_WIDE;
 
@@ -5862,7 +6239,9 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 						     DPLL_PORTB_READY_MASK);
 	}
 
-	if (IS_VALLEYVIEW(dev))
+	if (IS_CHERRYVIEW(dev))
+		chv_crtc_clock_get(crtc, pipe_config);
+	else if (IS_VALLEYVIEW(dev))
 		vlv_crtc_clock_get(crtc, pipe_config);
 	else
 		i9xx_crtc_clock_get(crtc, pipe_config);
@@ -5983,8 +6362,7 @@ static void ironlake_init_pch_refclk(struct drm_device *dev)
 			if (intel_panel_use_ssc(dev_priv) && can_ssc) {
 				DRM_DEBUG_KMS("Using SSC on eDP\n");
 				val |= DREF_CPU_SOURCE_OUTPUT_DOWNSPREAD;
-			}
-			else
+			} else
 				val |= DREF_CPU_SOURCE_OUTPUT_NONSPREAD;
 		} else
 			val |= DREF_CPU_SOURCE_OUTPUT_DISABLE;
@@ -6563,10 +6941,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 				  struct drm_framebuffer *fb)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	int num_connectors = 0;
 	intel_clock_t clock, reduced_clock;
 	u32 dpll = 0, fp = 0, fp2 = 0;
@@ -6574,7 +6949,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	bool is_lvds = false;
 	struct intel_encoder *encoder;
 	struct intel_shared_dpll *pll;
-	int ret;
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		switch (encoder->type) {
@@ -6624,36 +6998,18 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 		pll = intel_get_shared_dpll(intel_crtc);
 		if (pll == NULL) {
 			DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
-					 pipe_name(pipe));
+					 pipe_name(intel_crtc->pipe));
 			return -EINVAL;
 		}
 	} else
 		intel_put_shared_dpll(intel_crtc);
 
-	if (intel_crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(intel_crtc);
-
 	if (is_lvds && has_reduced_clock && i915.powersave)
 		intel_crtc->lowfreq_avail = true;
 	else
 		intel_crtc->lowfreq_avail = false;
 
-	intel_set_pipe_timings(intel_crtc);
-
-	if (intel_crtc->config.has_pch_encoder) {
-		intel_cpu_transcoder_set_m_n(intel_crtc,
-					     &intel_crtc->config.fdi_m_n);
-	}
-
-	ironlake_set_pipeconf(crtc);
-
-	/* Set up the display plane register */
-	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE);
-	POSTING_READ(DSPCNTR(plane));
-
-	ret = intel_pipe_set_base(crtc, x, y, fb);
-
-	return ret;
+	return 0;
 }
 
 static void intel_pch_transcoder_get_m_n(struct intel_crtc *crtc,
@@ -6831,6 +7187,9 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 		break;
 	}
 
+	if (tmp & PIPECONF_COLOR_RANGE_SELECT)
+		pipe_config->limited_color_range = true;
+
 	if (I915_READ(PCH_TRANSCONF(crtc->pipe)) & TRANS_ENABLE) {
 		struct intel_shared_dpll *pll;
 
@@ -6880,10 +7239,8 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
 	struct intel_crtc *crtc;
-	unsigned long irqflags;
-	uint32_t val;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head)
+	for_each_intel_crtc(dev, crtc)
 		WARN(crtc->active, "CRTC for pipe %c enabled\n",
 		     pipe_name(crtc->pipe));
 
@@ -6902,14 +7259,29 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	     "Utility pin enabled\n");
 	WARN(I915_READ(PCH_GTC_CTL) & PCH_GTC_ENABLE, "PCH GTC enabled\n");
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	val = I915_READ(DEIMR);
-	WARN((val | DE_PCH_EVENT_IVB) != 0xffffffff,
-	     "Unexpected DEIMR bits enabled: 0x%x\n", val);
-	val = I915_READ(SDEIMR);
-	WARN((val | SDE_HOTPLUG_MASK_CPT) != 0xffffffff,
-	     "Unexpected SDEIMR bits enabled: 0x%x\n", val);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	/*
+	 * In theory we can still leave IRQs enabled, as long as only the HPD
+	 * interrupts remain enabled. We used to check for that, but since it's
+	 * gen-specific and since we only disable LCPLL after we fully disable
+	 * the interrupts, the check below should be enough.
+	 */
+	WARN(!dev_priv->pm.irqs_disabled, "IRQs enabled\n");
+}
+
+static void hsw_write_dcomp(struct drm_i915_private *dev_priv, uint32_t val)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	if (IS_HASWELL(dev)) {
+		mutex_lock(&dev_priv->rps.hw_lock);
+		if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP,
+					    val))
+			DRM_ERROR("Failed to disable D_COMP\n");
+		mutex_unlock(&dev_priv->rps.hw_lock);
+	} else {
+		I915_WRITE(D_COMP, val);
+	}
+	POSTING_READ(D_COMP);
 }
 
 /*
@@ -6949,11 +7321,7 @@ static void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
 
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_DISABLE;
-	mutex_lock(&dev_priv->rps.hw_lock);
-	if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val))
-		DRM_ERROR("Failed to disable D_COMP\n");
-	mutex_unlock(&dev_priv->rps.hw_lock);
-	POSTING_READ(D_COMP);
+	hsw_write_dcomp(dev_priv, val);
 	ndelay(100);
 
 	if (wait_for((I915_READ(D_COMP) & D_COMP_RCOMP_IN_PROGRESS) == 0, 1))
@@ -7008,11 +7376,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_FORCE;
 	val &= ~D_COMP_COMP_DISABLE;
-	mutex_lock(&dev_priv->rps.hw_lock);
-	if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val))
-		DRM_ERROR("Failed to enable D_COMP\n");
-	mutex_unlock(&dev_priv->rps.hw_lock);
-	POSTING_READ(D_COMP);
+	hsw_write_dcomp(dev_priv, val);
 
 	val = I915_READ(LCPLL_CTL);
 	val &= ~LCPLL_PLL_DISABLE;
@@ -7066,8 +7430,6 @@ void hsw_enable_pc8(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	uint32_t val;
 
-	WARN_ON(!HAS_PC8(dev));
-
 	DRM_DEBUG_KMS("Enabling package C8+\n");
 
 	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
@@ -7077,7 +7439,6 @@ void hsw_enable_pc8(struct drm_i915_private *dev_priv)
 	}
 
 	lpt_disable_clkout_dp(dev);
-	hsw_runtime_pm_disable_interrupts(dev);
 	hsw_disable_lcpll(dev_priv, true, true);
 }
 
@@ -7086,12 +7447,9 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	uint32_t val;
 
-	WARN_ON(!HAS_PC8(dev));
-
 	DRM_DEBUG_KMS("Disabling package C8+\n");
 
 	hsw_restore_lcpll(dev_priv);
-	hsw_runtime_pm_restore_interrupts(dev);
 	lpt_init_pch_refclk(dev);
 
 	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
@@ -7101,10 +7459,11 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv)
 	}
 
 	intel_prepare_ddi(dev);
-	i915_gem_init_swizzling(dev);
-	mutex_lock(&dev_priv->rps.hw_lock);
-	gen6_update_ring_freq(dev);
-	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+static void snb_modeset_global_resources(struct drm_device *dev)
+{
+	modeset_update_crtc_power_domains(dev);
 }
 
 static void haswell_modeset_global_resources(struct drm_device *dev)
@@ -7116,39 +7475,15 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 				 int x, int y,
 				 struct drm_framebuffer *fb)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int plane = intel_crtc->plane;
-	int ret;
 
 	if (!intel_ddi_pll_select(intel_crtc))
 		return -EINVAL;
 	intel_ddi_pll_enable(intel_crtc);
 
-	if (intel_crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(intel_crtc);
-
 	intel_crtc->lowfreq_avail = false;
 
-	intel_set_pipe_timings(intel_crtc);
-
-	if (intel_crtc->config.has_pch_encoder) {
-		intel_cpu_transcoder_set_m_n(intel_crtc,
-					     &intel_crtc->config.fdi_m_n);
-	}
-
-	haswell_set_pipeconf(crtc);
-
-	intel_set_pipe_csc(crtc);
-
-	/* Set up the display plane register */
-	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE | DISPPLANE_PIPE_CSC_ENABLE);
-	POSTING_READ(DSPCNTR(plane));
-
-	ret = intel_pipe_set_base(crtc, x, y, fb);
-
-	return ret;
+	return 0;
 }
 
 static bool haswell_get_pipe_config(struct intel_crtc *crtc,
@@ -7228,38 +7563,6 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	return true;
 }
 
-static int intel_crtc_mode_set(struct drm_crtc *crtc,
-			       int x, int y,
-			       struct drm_framebuffer *fb)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_encoder *encoder;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_display_mode *mode = &intel_crtc->config.requested_mode;
-	int pipe = intel_crtc->pipe;
-	int ret;
-
-	drm_vblank_pre_modeset(dev, pipe);
-
-	ret = dev_priv->display.crtc_mode_set(crtc, x, y, fb);
-
-	drm_vblank_post_modeset(dev, pipe);
-
-	if (ret != 0)
-		return ret;
-
-	for_each_encoder_on_crtc(dev, crtc, encoder) {
-		DRM_DEBUG_KMS("[ENCODER:%d:%s] set [MODE:%d:%s]\n",
-			encoder->base.base.id,
-			drm_get_encoder_name(&encoder->base),
-			mode->base.id, mode->name);
-		encoder->mode_set(encoder);
-	}
-
-	return 0;
-}
-
 static struct {
 	int clock;
 	u32 config;
@@ -7374,8 +7677,6 @@ static void haswell_write_eld(struct drm_connector *connector,
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
-	struct drm_device *dev = crtc->dev;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t eldv;
 	uint32_t i;
 	int len;
@@ -7387,17 +7688,14 @@ static void haswell_write_eld(struct drm_connector *connector,
 	int aud_config = HSW_AUD_CFG(pipe);
 	int aud_cntrl_st2 = HSW_AUD_PIN_ELD_CP_VLD;
 
-
-	DRM_DEBUG_DRIVER("HDMI: Haswell Audio initialize....\n");
-
 	/* Audio output enable */
 	DRM_DEBUG_DRIVER("HDMI audio: enable codec\n");
 	tmp = I915_READ(aud_cntrl_st2);
 	tmp |= (AUDIO_OUTPUT_ENABLE_A << (pipe * 4));
 	I915_WRITE(aud_cntrl_st2, tmp);
+	POSTING_READ(aud_cntrl_st2);
 
-	/* Wait for 1 vertical blank */
-	intel_wait_for_vblank(dev, pipe);
+	assert_pipe_disabled(dev_priv, to_intel_crtc(crtc)->pipe);
 
 	/* Set ELD valid state */
 	tmp = I915_READ(aud_cntrl_st2);
@@ -7417,7 +7715,6 @@ static void haswell_write_eld(struct drm_connector *connector,
 	DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(pipe));
 
 	eldv = AUDIO_ELD_VALID_A << (pipe * 4);
-	intel_crtc->eld_vld = true;
 
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
 		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
@@ -7564,9 +7861,9 @@ void intel_write_eld(struct drm_encoder *encoder,
 
 	DRM_DEBUG_DRIVER("ELD on [CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
 			 connector->base.id,
-			 drm_get_connector_name(connector),
+			 connector->name,
 			 connector->encoder->base.id,
-			 drm_get_encoder_name(connector->encoder));
+			 connector->encoder->name);
 
 	connector->eld[6] = drm_av_sync_delay(connector, mode) / 2;
 
@@ -7579,29 +7876,33 @@ static void i845_update_cursor(struct drm_crtc *crtc, u32 base)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	bool visible = base != 0;
-	u32 cntl;
-
-	if (intel_crtc->cursor_visible == visible)
-		return;
+	uint32_t cntl;
 
-	cntl = I915_READ(_CURACNTR);
-	if (visible) {
+	if (base != intel_crtc->cursor_base) {
 		/* On these chipsets we can only modify the base whilst
 		 * the cursor is disabled.
 		 */
+		if (intel_crtc->cursor_cntl) {
+			I915_WRITE(_CURACNTR, 0);
+			POSTING_READ(_CURACNTR);
+			intel_crtc->cursor_cntl = 0;
+		}
+
 		I915_WRITE(_CURABASE, base);
+		POSTING_READ(_CURABASE);
+	}
 
-		cntl &= ~(CURSOR_FORMAT_MASK);
-		/* XXX width must be 64, stride 256 => 0x00 << 28 */
-		cntl |= CURSOR_ENABLE |
+	/* XXX width must be 64, stride 256 => 0x00 << 28 */
+	cntl = 0;
+	if (base)
+		cntl = (CURSOR_ENABLE |
 			CURSOR_GAMMA_ENABLE |
-			CURSOR_FORMAT_ARGB;
-	} else
-		cntl &= ~(CURSOR_ENABLE | CURSOR_GAMMA_ENABLE);
-	I915_WRITE(_CURACNTR, cntl);
-
-	intel_crtc->cursor_visible = visible;
+			CURSOR_FORMAT_ARGB);
+	if (intel_crtc->cursor_cntl != cntl) {
+		I915_WRITE(_CURACNTR, cntl);
+		POSTING_READ(_CURACNTR);
+		intel_crtc->cursor_cntl = cntl;
+	}
 }
 
 static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
@@ -7610,16 +7911,12 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	bool visible = base != 0;
+	uint32_t cntl;
 
-	if (intel_crtc->cursor_visible != visible) {
-		int16_t width = intel_crtc->cursor_width;
-		uint32_t cntl = I915_READ(CURCNTR(pipe));
-		if (base) {
-			cntl &= ~(CURSOR_MODE | MCURSOR_PIPE_SELECT);
-			cntl |= MCURSOR_GAMMA_ENABLE;
-
-			switch (width) {
+	cntl = 0;
+	if (base) {
+		cntl = MCURSOR_GAMMA_ENABLE;
+		switch (intel_crtc->cursor_width) {
 			case 64:
 				cntl |= CURSOR_MODE_64_ARGB_AX;
 				break;
@@ -7632,18 +7929,16 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
 			default:
 				WARN_ON(1);
 				return;
-			}
-			cntl |= pipe << 28; /* Connect to correct pipe */
-		} else {
-			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
-			cntl |= CURSOR_MODE_DISABLE;
 		}
+		cntl |= pipe << 28; /* Connect to correct pipe */
+	}
+	if (intel_crtc->cursor_cntl != cntl) {
 		I915_WRITE(CURCNTR(pipe), cntl);
-
-		intel_crtc->cursor_visible = visible;
+		POSTING_READ(CURCNTR(pipe));
+		intel_crtc->cursor_cntl = cntl;
 	}
+
 	/* and commit changes on next vblank */
-	POSTING_READ(CURCNTR(pipe));
 	I915_WRITE(CURBASE(pipe), base);
 	POSTING_READ(CURBASE(pipe));
 }
@@ -7654,15 +7949,12 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	bool visible = base != 0;
-
-	if (intel_crtc->cursor_visible != visible) {
-		int16_t width = intel_crtc->cursor_width;
-		uint32_t cntl = I915_READ(CURCNTR_IVB(pipe));
-		if (base) {
-			cntl &= ~CURSOR_MODE;
-			cntl |= MCURSOR_GAMMA_ENABLE;
-			switch (width) {
+	uint32_t cntl;
+
+	cntl = 0;
+	if (base) {
+		cntl = MCURSOR_GAMMA_ENABLE;
+		switch (intel_crtc->cursor_width) {
 			case 64:
 				cntl |= CURSOR_MODE_64_ARGB_AX;
 				break;
@@ -7675,23 +7967,20 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 			default:
 				WARN_ON(1);
 				return;
-			}
-		} else {
-			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
-			cntl |= CURSOR_MODE_DISABLE;
 		}
-		if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
-			cntl |= CURSOR_PIPE_CSC_ENABLE;
-			cntl &= ~CURSOR_TRICKLE_FEED_DISABLE;
-		}
-		I915_WRITE(CURCNTR_IVB(pipe), cntl);
+	}
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
+		cntl |= CURSOR_PIPE_CSC_ENABLE;
 
-		intel_crtc->cursor_visible = visible;
+	if (intel_crtc->cursor_cntl != cntl) {
+		I915_WRITE(CURCNTR(pipe), cntl);
+		POSTING_READ(CURCNTR(pipe));
+		intel_crtc->cursor_cntl = cntl;
 	}
+
 	/* and commit changes on next vblank */
-	POSTING_READ(CURCNTR_IVB(pipe));
-	I915_WRITE(CURBASE_IVB(pipe), base);
-	POSTING_READ(CURBASE_IVB(pipe));
+	I915_WRITE(CURBASE(pipe), base);
+	POSTING_READ(CURBASE(pipe));
 }
 
 /* If no-part of the cursor is visible on the framebuffer, then the GPU may hang... */
@@ -7705,7 +7994,6 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	int x = intel_crtc->cursor_x;
 	int y = intel_crtc->cursor_y;
 	u32 base = 0, pos = 0;
-	bool visible;
 
 	if (on)
 		base = intel_crtc->cursor_addr;
@@ -7734,20 +8022,18 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	}
 	pos |= y << CURSOR_Y_SHIFT;
 
-	visible = base != 0;
-	if (!visible && !intel_crtc->cursor_visible)
+	if (base == 0 && intel_crtc->cursor_base == 0)
 		return;
 
-	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev)) {
-		I915_WRITE(CURPOS_IVB(pipe), pos);
+	I915_WRITE(CURPOS(pipe), pos);
+
+	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev))
 		ivb_update_cursor(crtc, base);
-	} else {
-		I915_WRITE(CURPOS(pipe), pos);
-		if (IS_845G(dev) || IS_I865G(dev))
-			i845_update_cursor(crtc, base);
-		else
-			i9xx_update_cursor(crtc, base);
-	}
+	else if (IS_845G(dev) || IS_I865G(dev))
+		i845_update_cursor(crtc, base);
+	else
+		i9xx_update_cursor(crtc, base);
+	intel_crtc->cursor_base = base;
 }
 
 static int intel_crtc_cursor_set(struct drm_crtc *crtc,
@@ -8015,7 +8301,8 @@ mode_fits_in_fbdev(struct drm_device *dev,
 
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct drm_display_mode *mode,
-				struct intel_load_detect_pipe *old)
+				struct intel_load_detect_pipe *old,
+				struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_crtc *intel_crtc;
 	struct intel_encoder *intel_encoder =
@@ -8025,11 +8312,19 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	struct drm_crtc *crtc = NULL;
 	struct drm_device *dev = encoder->dev;
 	struct drm_framebuffer *fb;
-	int i = -1;
+	struct drm_mode_config *config = &dev->mode_config;
+	int ret, i = -1;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector),
-		      encoder->base.id, drm_get_encoder_name(encoder));
+		      connector->base.id, connector->name,
+		      encoder->base.id, encoder->name);
+
+	drm_modeset_acquire_init(ctx, 0);
+
+retry:
+	ret = drm_modeset_lock(&config->connection_mutex, ctx);
+	if (ret)
+		goto fail_unlock;
 
 	/*
 	 * Algorithm gets a little messy:
@@ -8045,7 +8340,9 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	if (encoder->crtc) {
 		crtc = encoder->crtc;
 
-		mutex_lock(&crtc->mutex);
+		ret = drm_modeset_lock(&crtc->mutex, ctx);
+		if (ret)
+			goto fail_unlock;
 
 		old->dpms_mode = connector->dpms;
 		old->load_detect_temp = false;
@@ -8058,7 +8355,7 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	}
 
 	/* Find an unused one (if possible) */
-	list_for_each_entry(possible_crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, possible_crtc) {
 		i++;
 		if (!(encoder->possible_crtcs & (1 << i)))
 			continue;
@@ -8073,10 +8370,12 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	 */
 	if (!crtc) {
 		DRM_DEBUG_KMS("no pipe available for load-detect\n");
-		return false;
+		goto fail_unlock;
 	}
 
-	mutex_lock(&crtc->mutex);
+	ret = drm_modeset_lock(&crtc->mutex, ctx);
+	if (ret)
+		goto fail_unlock;
 	intel_encoder->new_crtc = to_intel_crtc(crtc);
 	to_intel_connector(connector)->new_encoder = intel_encoder;
 
@@ -8126,12 +8425,21 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 		intel_crtc->new_config = &intel_crtc->config;
 	else
 		intel_crtc->new_config = NULL;
-	mutex_unlock(&crtc->mutex);
+fail_unlock:
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(ctx);
+		goto retry;
+	}
+
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+
 	return false;
 }
 
 void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old)
+				    struct intel_load_detect_pipe *old,
+				    struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_encoder *intel_encoder =
 		intel_attached_encoder(connector);
@@ -8140,8 +8448,8 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector),
-		      encoder->base.id, drm_get_encoder_name(encoder));
+		      connector->base.id, connector->name,
+		      encoder->base.id, encoder->name);
 
 	if (old->load_detect_temp) {
 		to_intel_connector(connector)->new_encoder = NULL;
@@ -8155,7 +8463,7 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 			drm_framebuffer_unreference(old->release_fb);
 		}
 
-		mutex_unlock(&crtc->mutex);
+		goto unlock;
 		return;
 	}
 
@@ -8163,7 +8471,9 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	if (old->dpms_mode != DRM_MODE_DPMS_ON)
 		connector->funcs->dpms(connector, old->dpms_mode);
 
-	mutex_unlock(&crtc->mutex);
+unlock:
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
 }
 
 static int i9xx_pll_refclk(struct drm_device *dev,
@@ -8449,7 +8759,7 @@ void intel_mark_idle(struct drm_device *dev)
 	if (!i915.powersave)
 		goto out;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		if (!crtc->primary->fb)
 			continue;
 
@@ -8464,7 +8774,7 @@ out:
 }
 
 void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			struct intel_ring_buffer *ring)
+			struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_crtc *crtc;
@@ -8472,7 +8782,7 @@ void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
 	if (!i915.powersave)
 		return;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		if (!crtc->primary->fb)
 			continue;
 
@@ -8560,7 +8870,7 @@ static void do_intel_finish_page_flip(struct drm_device *dev,
 	if (work->event)
 		drm_send_vblank_event(dev, intel_crtc->pipe, work->event);
 
-	drm_vblank_put(dev, intel_crtc->pipe);
+	drm_crtc_vblank_put(crtc);
 
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
@@ -8587,6 +8897,48 @@ void intel_finish_page_flip_plane(struct drm_device *dev, int plane)
 	do_intel_finish_page_flip(dev, crtc);
 }
 
+/* Is 'a' after or equal to 'b'? */
+static bool g4x_flip_count_after_eq(u32 a, u32 b)
+{
+	return !((a - b) & 0x80000000);
+}
+
+static bool page_flip_finished(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/*
+	 * The relevant registers doen't exist on pre-ctg.
+	 * As the flip done interrupt doesn't trigger for mmio
+	 * flips on gmch platforms, a flip count check isn't
+	 * really needed there. But since ctg has the registers,
+	 * include it in the check anyway.
+	 */
+	if (INTEL_INFO(dev)->gen < 5 && !IS_G4X(dev))
+		return true;
+
+	/*
+	 * A DSPSURFLIVE check isn't enough in case the mmio and CS flips
+	 * used the same base address. In that case the mmio flip might
+	 * have completed, but the CS hasn't even executed the flip yet.
+	 *
+	 * A flip count check isn't enough as the CS might have updated
+	 * the base address just after start of vblank, but before we
+	 * managed to process the interrupt. This means we'd complete the
+	 * CS flip too soon.
+	 *
+	 * Combining both checks should get us a good enough result. It may
+	 * still happen that the CS flip has been executed, but has not
+	 * yet actually completed. But in case the base address is the same
+	 * anyway, we don't really care.
+	 */
+	return (I915_READ(DSPSURFLIVE(crtc->plane)) & ~0xfff) ==
+		crtc->unpin_work->gtt_offset &&
+		g4x_flip_count_after_eq(I915_READ(PIPE_FLIPCOUNT_GM45(crtc->pipe)),
+				    crtc->unpin_work->flip_count);
+}
+
 void intel_prepare_page_flip(struct drm_device *dev, int plane)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -8599,12 +8951,12 @@ void intel_prepare_page_flip(struct drm_device *dev, int plane)
 	 * is also accompanied by a spurious intel_prepare_page_flip().
 	 */
 	spin_lock_irqsave(&dev->event_lock, flags);
-	if (intel_crtc->unpin_work)
+	if (intel_crtc->unpin_work && page_flip_finished(intel_crtc))
 		atomic_inc_not_zero(&intel_crtc->unpin_work->pending);
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
 
-inline static void intel_mark_page_flip_active(struct intel_crtc *intel_crtc)
+static inline void intel_mark_page_flip_active(struct intel_crtc *intel_crtc)
 {
 	/* Ensure that the work item is consistent when activating it ... */
 	smp_wmb();
@@ -8617,21 +8969,16 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 6);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	/* Can't queue multiple flips, so wait for the previous
 	 * one to finish before executing the next.
@@ -8645,38 +8992,28 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, 0); /* aux display base address, unused */
 
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen3_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 6);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	if (intel_crtc->plane)
 		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
@@ -8687,38 +9024,29 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, MI_NOOP);
 
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen4_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pf, pipesrc;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 4);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	/* i965+ uses the linear or tiled offsets from the
 	 * Display Registers (which do not change across a page-flip)
@@ -8727,8 +9055,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring,
-			(i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset) |
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset |
 			obj->tiling_mode);
 
 	/* XXX Enabling the panel-fitter across page-flip is so far
@@ -8742,37 +9069,28 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen6_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	uint32_t pf, pipesrc;
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 4);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0] | obj->tiling_mode);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 
 	/* Contrary to the suggestions in the documentation,
 	 * "Enable Panel Fitter" does not seem to be required when page
@@ -8787,34 +9105,20 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen7_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_ring_buffer *ring;
 	uint32_t plane_bit = 0;
 	int len, ret;
 
-	ring = obj->ring;
-	if (IS_VALLEYVIEW(dev) || ring == NULL || ring->id != RCS)
-		ring = &dev_priv->ring[BCS];
-
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
-	switch(intel_crtc->plane) {
+	switch (intel_crtc->plane) {
 	case PLANE_A:
 		plane_bit = MI_DISPLAY_FLIP_IVB_PLANE_A;
 		break;
@@ -8826,13 +9130,20 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 		break;
 	default:
 		WARN_ONCE(1, "unknown plane in flip command\n");
-		ret = -ENODEV;
-		goto err_unpin;
+		return -ENODEV;
 	}
 
 	len = 4;
-	if (ring->id == RCS)
+	if (ring->id == RCS) {
 		len += 6;
+		/*
+		 * On Gen 8, SRM is now taking an extra dword to accommodate
+		 * 48bits addresses, and we need a NOOP for the batch size to
+		 * stay even.
+		 */
+		if (IS_GEN8(dev))
+			len += 2;
+	}
 
 	/*
 	 * BSpec MI_DISPLAY_FLIP for IVB:
@@ -8846,11 +9157,11 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	 */
 	ret = intel_ring_cacheline_align(ring);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	ret = intel_ring_begin(ring, len);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	/* Unmask the flip-done completion message. Note that the bspec says that
 	 * we should do this for both the BCS and RCS, and that we must not unmask
@@ -8867,31 +9178,35 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 		intel_ring_emit(ring, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
 					DERRMR_PIPEB_PRI_FLIP_DONE |
 					DERRMR_PIPEC_PRI_FLIP_DONE));
-		intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1) |
-				MI_SRM_LRM_GLOBAL_GTT);
+		if (IS_GEN8(dev))
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM_GEN8(1) |
+					      MI_SRM_LRM_GLOBAL_GTT);
+		else
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1) |
+					      MI_SRM_LRM_GLOBAL_GTT);
 		intel_ring_emit(ring, DERRMR);
 		intel_ring_emit(ring, ring->scratch.gtt_offset + 256);
+		if (IS_GEN8(dev)) {
+			intel_ring_emit(ring, 0);
+			intel_ring_emit(ring, MI_NOOP);
+		}
 	}
 
 	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
 	intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode));
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, (MI_NOOP));
 
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_default_queue_flip(struct drm_device *dev,
 				    struct drm_crtc *crtc,
 				    struct drm_framebuffer *fb,
 				    struct drm_i915_gem_object *obj,
+				    struct intel_engine_cs *ring,
 				    uint32_t flags)
 {
 	return -ENODEV;
@@ -8908,6 +9223,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct drm_i915_gem_object *obj = to_intel_framebuffer(fb)->obj;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_unpin_work *work;
+	struct intel_engine_cs *ring;
 	unsigned long flags;
 	int ret;
 
@@ -8936,7 +9252,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	work->old_fb_obj = to_intel_framebuffer(old_fb)->obj;
 	INIT_WORK(&work->work, intel_unpin_work_fn);
 
-	ret = drm_vblank_get(dev, intel_crtc->pipe);
+	ret = drm_crtc_vblank_get(crtc);
 	if (ret)
 		goto free_work;
 
@@ -8945,7 +9261,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (intel_crtc->unpin_work) {
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 		kfree(work);
-		drm_vblank_put(dev, intel_crtc->pipe);
+		drm_crtc_vblank_put(crtc);
 
 		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
 		return -EBUSY;
@@ -8973,10 +9289,30 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	atomic_inc(&intel_crtc->unpin_work_count);
 	intel_crtc->reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
-	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, page_flip_flags);
+	if (INTEL_INFO(dev)->gen >= 5 || IS_G4X(dev))
+		work->flip_count = I915_READ(PIPE_FLIPCOUNT_GM45(intel_crtc->pipe)) + 1;
+
+	if (IS_VALLEYVIEW(dev)) {
+		ring = &dev_priv->ring[BCS];
+	} else if (INTEL_INFO(dev)->gen >= 7) {
+		ring = obj->ring;
+		if (ring == NULL || ring->id != RCS)
+			ring = &dev_priv->ring[BCS];
+	} else {
+		ring = &dev_priv->ring[RCS];
+	}
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
 	if (ret)
 		goto cleanup_pending;
 
+	work->gtt_offset =
+		i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset;
+
+	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, ring, page_flip_flags);
+	if (ret)
+		goto cleanup_unpin;
+
 	intel_disable_fbc(dev);
 	intel_mark_fb_busy(obj, NULL);
 	mutex_unlock(&dev->struct_mutex);
@@ -8985,6 +9321,8 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	return 0;
 
+cleanup_unpin:
+	intel_unpin_fb_obj(obj);
 cleanup_pending:
 	atomic_dec(&intel_crtc->unpin_work_count);
 	crtc->primary->fb = old_fb;
@@ -8997,7 +9335,7 @@ cleanup:
 	intel_crtc->unpin_work = NULL;
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
-	drm_vblank_put(dev, intel_crtc->pipe);
+	drm_crtc_vblank_put(crtc);
 free_work:
 	kfree(work);
 
@@ -9040,8 +9378,7 @@ static void intel_modeset_update_staged_output_state(struct drm_device *dev)
 			to_intel_crtc(encoder->base.crtc);
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = crtc->base.enabled;
 
 		if (crtc->new_enabled)
@@ -9072,21 +9409,20 @@ static void intel_modeset_commit_output_state(struct drm_device *dev)
 		encoder->base.crtc = &encoder->new_crtc->base;
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->base.enabled = crtc->new_enabled;
 	}
 }
 
 static void
-connected_sink_compute_bpp(struct intel_connector * connector,
+connected_sink_compute_bpp(struct intel_connector *connector,
 			   struct intel_crtc_config *pipe_config)
 {
 	int bpp = pipe_config->pipe_bpp;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] checking for sink bpp constrains\n",
 		connector->base.base.id,
-		drm_get_connector_name(&connector->base));
+		connector->base.name);
 
 	/* Don't use an invalid EDID bpc value */
 	if (connector->base.display_info.bpc &&
@@ -9427,8 +9763,7 @@ intel_modeset_affected_pipes(struct drm_crtc *crtc, unsigned *modeset_pipes,
 	}
 
 	/* Check for pipes that will be enabled/disabled ... */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		if (intel_crtc->base.enabled == intel_crtc->new_enabled)
 			continue;
 
@@ -9501,8 +9836,7 @@ intel_modeset_update_state(struct drm_device *dev, unsigned prepare_pipes)
 	intel_modeset_commit_output_state(dev);
 
 	/* Double check state. */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		WARN_ON(intel_crtc->base.enabled != intel_crtc_in_use(&intel_crtc->base));
 		WARN_ON(intel_crtc->new_config &&
 			intel_crtc->new_config != &intel_crtc->config);
@@ -9631,6 +9965,12 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_I(adjusted_mode.crtc_vsync_end);
 
 	PIPE_CONF_CHECK_I(pixel_multiplier);
+	PIPE_CONF_CHECK_I(has_hdmi_sink);
+	if ((INTEL_INFO(dev)->gen < 8 && !IS_HASWELL(dev)) ||
+	    IS_VALLEYVIEW(dev))
+		PIPE_CONF_CHECK_I(limited_color_range);
+
+	PIPE_CONF_CHECK_I(has_audio);
 
 	PIPE_CONF_CHECK_FLAGS(adjusted_mode.flags,
 			      DRM_MODE_FLAG_INTERLACE);
@@ -9728,7 +10068,7 @@ check_encoder_state(struct drm_device *dev)
 
 		DRM_DEBUG_KMS("[ENCODER:%d:%s]\n",
 			      encoder->base.base.id,
-			      drm_get_encoder_name(&encoder->base));
+			      encoder->base.name);
 
 		WARN(&encoder->new_crtc->base != encoder->base.crtc,
 		     "encoder's stage crtc doesn't match current crtc\n");
@@ -9780,8 +10120,7 @@ check_crtc_state(struct drm_device *dev)
 	struct intel_encoder *encoder;
 	struct intel_crtc_config pipe_config;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		bool enabled = false;
 		bool active = false;
 
@@ -9870,8 +10209,7 @@ check_shared_dpll_state(struct drm_device *dev)
 		     "pll on state mismatch (expected %i, found %i)\n",
 		     pll->on, active);
 
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-				    base.head) {
+		for_each_intel_crtc(dev, crtc) {
 			if (crtc->base.enabled && intel_crtc_to_shared_dpll(crtc) == pll)
 				enabled_crtcs++;
 			if (crtc->active && intel_crtc_to_shared_dpll(crtc) == pll)
@@ -9911,6 +10249,44 @@ void ironlake_check_encoder_dotclock(const struct intel_crtc_config *pipe_config
 	     pipe_config->adjusted_mode.crtc_clock, dotclock);
 }
 
+static void update_scanline_offset(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+
+	/*
+	 * The scanline counter increments at the leading edge of hsync.
+	 *
+	 * On most platforms it starts counting from vtotal-1 on the
+	 * first active line. That means the scanline counter value is
+	 * always one less than what we would expect. Ie. just after
+	 * start of vblank, which also occurs at start of hsync (on the
+	 * last active line), the scanline counter will read vblank_start-1.
+	 *
+	 * On gen2 the scanline counter starts counting from 1 instead
+	 * of vtotal-1, so we have to subtract one (or rather add vtotal-1
+	 * to keep the value positive), instead of adding one.
+	 *
+	 * On HSW+ the behaviour of the scanline counter depends on the output
+	 * type. For DP ports it behaves like most other platforms, but on HDMI
+	 * there's an extra 1 line difference. So we need to add two instead of
+	 * one to the value.
+	 */
+	if (IS_GEN2(dev)) {
+		const struct drm_display_mode *mode = &crtc->config.adjusted_mode;
+		int vtotal;
+
+		vtotal = mode->crtc_vtotal;
+		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+			vtotal /= 2;
+
+		crtc->scanline_offset = vtotal - 1;
+	} else if (HAS_DDI(dev) &&
+		   intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_HDMI)) {
+		crtc->scanline_offset = 2;
+	} else
+		crtc->scanline_offset = 1;
+}
+
 static int __intel_set_mode(struct drm_crtc *crtc,
 			    struct drm_display_mode *mode,
 			    int x, int y, struct drm_framebuffer *fb)
@@ -10002,15 +10378,38 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 	 * on the DPLL.
 	 */
 	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
-		ret = intel_crtc_mode_set(&intel_crtc->base,
-					  x, y, fb);
+		struct drm_framebuffer *old_fb;
+
+		mutex_lock(&dev->struct_mutex);
+		ret = intel_pin_and_fence_fb_obj(dev,
+						 to_intel_framebuffer(fb)->obj,
+						 NULL);
+		if (ret != 0) {
+			DRM_ERROR("pin & fence failed\n");
+			mutex_unlock(&dev->struct_mutex);
+			goto done;
+		}
+		old_fb = crtc->primary->fb;
+		if (old_fb)
+			intel_unpin_fb_obj(to_intel_framebuffer(old_fb)->obj);
+		mutex_unlock(&dev->struct_mutex);
+
+		crtc->primary->fb = fb;
+		crtc->x = x;
+		crtc->y = y;
+
+		ret = dev_priv->display.crtc_mode_set(&intel_crtc->base,
+						      x, y, fb);
 		if (ret)
 			goto done;
 	}
 
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
-	for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc)
+	for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc) {
+		update_scanline_offset(intel_crtc);
+
 		dev_priv->display.crtc_enable(&intel_crtc->base);
+	}
 
 	/* FIXME: add subpixel order */
 done:
@@ -10086,7 +10485,7 @@ static int intel_set_config_save_state(struct drm_device *dev,
 	 * restored, not the drivers personal bookkeeping.
 	 */
 	count = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		config->save_crtc_enabled[count++] = crtc->enabled;
 	}
 
@@ -10112,7 +10511,7 @@ static void intel_set_config_restore_state(struct drm_device *dev,
 	int count;
 
 	count = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = config->save_crtc_enabled[count++];
 
 		if (crtc->new_enabled)
@@ -10236,7 +10635,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n",
 				connector->base.base.id,
-				drm_get_connector_name(&connector->base));
+				connector->base.name);
 		}
 
 
@@ -10271,7 +10670,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n",
 			connector->base.base.id,
-			drm_get_connector_name(&connector->base),
+			connector->base.name,
 			new_crtc->base.id);
 	}
 
@@ -10302,8 +10701,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 	}
 	/* Now we've also updated encoder->new_crtc for all encoders. */
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = false;
 
 		list_for_each_entry(encoder,
@@ -10516,7 +10914,7 @@ static void ibx_pch_dpll_disable(struct drm_i915_private *dev_priv,
 	struct intel_crtc *crtc;
 
 	/* Make sure no transcoder isn't still depending on us. */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (intel_crtc_to_shared_dpll(crtc) == pll)
 			assert_pch_transcoder_disabled(dev_priv, crtc->pipe);
 	}
@@ -10573,16 +10971,6 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 
 	drm_crtc_init(dev, &intel_crtc->base, &intel_crtc_funcs);
 
-	if (IS_GEN2(dev)) {
-		intel_crtc->max_cursor_width = GEN2_CURSOR_WIDTH;
-		intel_crtc->max_cursor_height = GEN2_CURSOR_HEIGHT;
-	} else {
-		intel_crtc->max_cursor_width = CURSOR_WIDTH;
-		intel_crtc->max_cursor_height = CURSOR_HEIGHT;
-	}
-	dev->mode_config.cursor_width = intel_crtc->max_cursor_width;
-	dev->mode_config.cursor_height = intel_crtc->max_cursor_height;
-
 	drm_mode_crtc_set_gamma_size(&intel_crtc->base, 256);
 	for (i = 0; i < 256; i++) {
 		intel_crtc->lut_r[i] = i;
@@ -10601,19 +10989,27 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 		intel_crtc->plane = !pipe;
 	}
 
+	intel_crtc->cursor_base = ~0;
+	intel_crtc->cursor_cntl = ~0;
+
+	init_waitqueue_head(&intel_crtc->vbl_wait);
+
 	BUG_ON(pipe >= ARRAY_SIZE(dev_priv->plane_to_crtc_mapping) ||
 	       dev_priv->plane_to_crtc_mapping[intel_crtc->plane] != NULL);
 	dev_priv->plane_to_crtc_mapping[intel_crtc->plane] = &intel_crtc->base;
 	dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base;
 
 	drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs);
+
+	WARN_ON(drm_crtc_index(&intel_crtc->base) != intel_crtc->pipe);
 }
 
 enum pipe intel_get_pipe_from_connector(struct intel_connector *connector)
 {
 	struct drm_encoder *encoder = connector->base.encoder;
+	struct drm_device *dev = connector->base.dev;
 
-	WARN_ON(!mutex_is_locked(&connector->base.dev->mode_config.mutex));
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	if (!encoder)
 		return INVALID_PIPE;
@@ -10701,6 +11097,22 @@ const char *intel_output_name(int output)
 	return names[output];
 }
 
+static bool intel_crt_present(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (IS_ULT(dev))
+		return false;
+
+	if (IS_CHERRYVIEW(dev))
+		return false;
+
+	if (IS_VALLEYVIEW(dev) && !dev_priv->vbt.int_crt_support)
+		return false;
+
+	return true;
+}
+
 static void intel_setup_outputs(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -10709,7 +11121,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 
 	intel_lvds_init(dev);
 
-	if (!IS_ULT(dev))
+	if (intel_crt_present(dev))
 		intel_crt_init(dev);
 
 	if (HAS_DDI(dev)) {
@@ -10773,6 +11185,15 @@ static void intel_setup_outputs(struct drm_device *dev)
 				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
 		}
 
+		if (IS_CHERRYVIEW(dev)) {
+			if (I915_READ(VLV_DISPLAY_BASE + CHV_HDMID) & SDVO_DETECTED) {
+				intel_hdmi_init(dev, VLV_DISPLAY_BASE + CHV_HDMID,
+						PORT_D);
+				if (I915_READ(VLV_DISPLAY_BASE + DP_D) & DP_DETECTED)
+					intel_dp_init(dev, VLV_DISPLAY_BASE + DP_D, PORT_D);
+			}
+		}
+
 		intel_dsi_init(dev);
 	} else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) {
 		bool found = false;
@@ -11002,6 +11423,8 @@ static void intel_init_display(struct drm_device *dev)
 
 	if (HAS_PCH_SPLIT(dev) || IS_G4X(dev))
 		dev_priv->display.find_dpll = g4x_find_best_dpll;
+	else if (IS_CHERRYVIEW(dev))
+		dev_priv->display.find_dpll = chv_find_best_dpll;
 	else if (IS_VALLEYVIEW(dev))
 		dev_priv->display.find_dpll = vlv_find_best_dpll;
 	else if (IS_PINEVIEW(dev))
@@ -11083,6 +11506,8 @@ static void intel_init_display(struct drm_device *dev)
 		} else if (IS_GEN6(dev)) {
 			dev_priv->display.fdi_link_train = gen6_fdi_link_train;
 			dev_priv->display.write_eld = ironlake_write_eld;
+			dev_priv->display.modeset_global_resources =
+				snb_modeset_global_resources;
 		} else if (IS_IVYBRIDGE(dev)) {
 			/* FIXME: detect B0+ stepping and use auto training */
 			dev_priv->display.fdi_link_train = ivb_manual_fdi_link_train;
@@ -11166,6 +11591,14 @@ static void quirk_invert_brightness(struct drm_device *dev)
 	DRM_INFO("applying inverted panel brightness quirk\n");
 }
 
+/* Some VBT's incorrectly indicate no backlight is present */
+static void quirk_backlight_present(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	dev_priv->quirks |= QUIRK_BACKLIGHT_PRESENT;
+	DRM_INFO("applying backlight present quirk\n");
+}
+
 struct intel_quirk {
 	int device;
 	int subsystem_vendor;
@@ -11211,9 +11644,6 @@ static struct intel_quirk intel_quirks[] = {
 	/* ThinkPad T60 needs pipe A force quirk (bug #16494) */
 	{ 0x2782, 0x17aa, 0x201a, quirk_pipea_force },
 
-	/* 830 needs to leave pipe A & dpll A up */
-	{ 0x3577, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
-
 	/* Lenovo U160 cannot use SSC on LVDS */
 	{ 0x0046, 0x17aa, 0x3920, quirk_ssc_force_disable },
 
@@ -11237,6 +11667,12 @@ static struct intel_quirk intel_quirks[] = {
 
 	/* Acer Aspire 5336 */
 	{ 0x2a42, 0x1025, 0x048a, quirk_invert_brightness },
+
+	/* Acer C720 and C720P Chromebooks (Celeron 2955U) have backlights */
+	{ 0x0a06, 0x1025, 0x0a11, quirk_backlight_present },
+
+	/* Toshiba CB35 Chromebook (Celeron 2955U) */
+	{ 0x0a06, 0x1179, 0x0a88, quirk_backlight_present },
 };
 
 static void intel_init_quirks(struct drm_device *dev)
@@ -11287,9 +11723,7 @@ void intel_modeset_init_hw(struct drm_device *dev)
 
 	intel_reset_dpio(dev);
 
-	mutex_lock(&dev->struct_mutex);
 	intel_enable_gt_powersave(dev);
-	mutex_unlock(&dev->struct_mutex);
 }
 
 void intel_modeset_suspend_hw(struct drm_device *dev)
@@ -11333,6 +11767,15 @@ void intel_modeset_init(struct drm_device *dev)
 		dev->mode_config.max_width = 8192;
 		dev->mode_config.max_height = 8192;
 	}
+
+	if (IS_GEN2(dev)) {
+		dev->mode_config.cursor_width = GEN2_CURSOR_WIDTH;
+		dev->mode_config.cursor_height = GEN2_CURSOR_HEIGHT;
+	} else {
+		dev->mode_config.cursor_width = MAX_CURSOR_WIDTH;
+		dev->mode_config.cursor_height = MAX_CURSOR_HEIGHT;
+	}
+
 	dev->mode_config.fb_base = dev_priv->gtt.mappable_base;
 
 	DRM_DEBUG_KMS("%d display pipe%s available.\n",
@@ -11362,12 +11805,11 @@ void intel_modeset_init(struct drm_device *dev)
 	/* Just in case the BIOS is doing something questionable. */
 	intel_disable_fbc(dev);
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	intel_modeset_setup_hw_state(dev, false);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (!crtc->active)
 			continue;
 
@@ -11395,6 +11837,7 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 	struct intel_connector *connector;
 	struct drm_connector *crt = NULL;
 	struct intel_load_detect_pipe load_detect_temp;
+	struct drm_modeset_acquire_ctx ctx;
 
 	/* We can't just switch on the pipe A, we need to set things up with a
 	 * proper mode and output configuration. As a gross hack, enable pipe A
@@ -11411,8 +11854,8 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 	if (!crt)
 		return;
 
-	if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp))
-		intel_release_load_detect_pipe(crt, &load_detect_temp);
+	if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp, &ctx))
+		intel_release_load_detect_pipe(crt, &load_detect_temp, &ctx);
 
 
 }
@@ -11447,6 +11890,12 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 	reg = PIPECONF(crtc->config.cpu_transcoder);
 	I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK);
 
+	/* restore vblank interrupts to correct state */
+	if (crtc->active)
+		drm_vblank_on(dev, crtc->pipe);
+	else
+		drm_vblank_off(dev, crtc->pipe);
+
 	/* We need to sanitize the plane -> pipe mapping first because this will
 	 * disable the crtc (and hence change the state) if it is wrong. Note
 	 * that gen4+ has a fixed plane -> pipe mapping.  */
@@ -11525,16 +11974,25 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 			encoder->base.crtc = NULL;
 		}
 	}
-	if (crtc->active) {
+
+	if (crtc->active || IS_VALLEYVIEW(dev) || INTEL_INFO(dev)->gen < 5) {
 		/*
 		 * We start out with underrun reporting disabled to avoid races.
 		 * For correct bookkeeping mark this on active crtcs.
 		 *
+		 * Also on gmch platforms we dont have any hardware bits to
+		 * disable the underrun reporting. Which means we need to start
+		 * out with underrun reporting disabled also on inactive pipes,
+		 * since otherwise we'll complain about the garbage we read when
+		 * e.g. coming up after runtime pm.
+		 *
 		 * No protection against concurrent access is required - at
 		 * worst a fifo underrun happens which also sets this to false.
 		 */
 		crtc->cpu_fifo_underrun_disabled = true;
 		crtc->pch_fifo_underrun_disabled = true;
+
+		update_scanline_offset(crtc);
 	}
 }
 
@@ -11552,7 +12010,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	if (encoder->connectors_active && !has_active_crtc) {
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] has active connectors but no active pipe!\n",
 			      encoder->base.base.id,
-			      drm_get_encoder_name(&encoder->base));
+			      encoder->base.name);
 
 		/* Connector is active, but has no active pipe. This is
 		 * fallout from our resume register restoring. Disable
@@ -11560,7 +12018,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 		if (encoder->base.crtc) {
 			DRM_DEBUG_KMS("[ENCODER:%d:%s] manually disabled\n",
 				      encoder->base.base.id,
-				      drm_get_encoder_name(&encoder->base));
+				      encoder->base.name);
 			encoder->disable(encoder);
 		}
 		encoder->base.crtc = NULL;
@@ -11611,6 +12069,16 @@ void i915_redisable_vga(struct drm_device *dev)
 	i915_redisable_vga_power_on(dev);
 }
 
+static bool primary_get_hw_state(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+
+	if (!crtc->active)
+		return false;
+
+	return I915_READ(DSPCNTR(crtc->plane)) & DISPLAY_PLANE_ENABLE;
+}
+
 static void intel_modeset_readout_hw_state(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -11620,8 +12088,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 	struct intel_connector *connector;
 	int i;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		memset(&crtc->config, 0, sizeof(crtc->config));
 
 		crtc->config.quirks |= PIPE_CONFIG_QUIRK_INHERITED_MODE;
@@ -11630,7 +12097,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 								 &crtc->config);
 
 		crtc->base.enabled = crtc->active;
-		crtc->primary_enabled = crtc->active;
+		crtc->primary_enabled = primary_get_hw_state(crtc);
 
 		DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n",
 			      crtc->base.base.id,
@@ -11646,8 +12113,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 
 		pll->on = pll->get_hw_state(dev_priv, pll, &pll->hw_state);
 		pll->active = 0;
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-				    base.head) {
+		for_each_intel_crtc(dev, crtc) {
 			if (crtc->active && intel_crtc_to_shared_dpll(crtc) == pll)
 				pll->active++;
 		}
@@ -11672,7 +12138,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		encoder->connectors_active = false;
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] hw state readout: %s, pipe %c\n",
 			      encoder->base.base.id,
-			      drm_get_encoder_name(&encoder->base),
+			      encoder->base.name,
 			      encoder->base.crtc ? "enabled" : "disabled",
 			      pipe_name(pipe));
 	}
@@ -11689,7 +12155,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		}
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] hw state readout: %s\n",
 			      connector->base.base.id,
-			      drm_get_connector_name(&connector->base),
+			      connector->base.name,
 			      connector->base.encoder ? "enabled" : "disabled");
 	}
 }
@@ -11712,8 +12178,7 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 	 * Note that this could go away if we move to using crtc_config
 	 * checking everywhere.
 	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (crtc->active && i915.fastboot) {
 			intel_mode_from_pipe_config(&crtc->base.mode, &crtc->config);
 			DRM_DEBUG_KMS("[CRTC:%d] found active mode: ",
@@ -11789,7 +12254,7 @@ void intel_modeset_gem_init(struct drm_device *dev)
 	 * for this.
 	 */
 	mutex_lock(&dev->struct_mutex);
-	list_for_each_entry(c, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, c) {
 		if (!c->primary->fb)
 			continue;
 
@@ -11835,7 +12300,7 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
 	intel_unregister_dsm_handler();
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		/* Skip inactive CRTCs */
 		if (!crtc->primary->fb)
 			continue;
@@ -11933,6 +12398,7 @@ struct intel_display_error_state {
 	struct intel_pipe_error_state {
 		bool power_domain_on;
 		u32 source;
+		u32 stat;
 	} pipe[I915_MAX_PIPES];
 
 	struct intel_plane_error_state {
@@ -11985,20 +12451,14 @@ intel_display_capture_error_state(struct drm_device *dev)
 
 	for_each_pipe(i) {
 		error->pipe[i].power_domain_on =
-			intel_display_power_enabled_sw(dev_priv,
-						       POWER_DOMAIN_PIPE(i));
+			intel_display_power_enabled_unlocked(dev_priv,
+							   POWER_DOMAIN_PIPE(i));
 		if (!error->pipe[i].power_domain_on)
 			continue;
 
-		if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev)) {
-			error->cursor[i].control = I915_READ(CURCNTR(i));
-			error->cursor[i].position = I915_READ(CURPOS(i));
-			error->cursor[i].base = I915_READ(CURBASE(i));
-		} else {
-			error->cursor[i].control = I915_READ(CURCNTR_IVB(i));
-			error->cursor[i].position = I915_READ(CURPOS_IVB(i));
-			error->cursor[i].base = I915_READ(CURBASE_IVB(i));
-		}
+		error->cursor[i].control = I915_READ(CURCNTR(i));
+		error->cursor[i].position = I915_READ(CURPOS(i));
+		error->cursor[i].base = I915_READ(CURBASE(i));
 
 		error->plane[i].control = I915_READ(DSPCNTR(i));
 		error->plane[i].stride = I915_READ(DSPSTRIDE(i));
@@ -12014,6 +12474,9 @@ intel_display_capture_error_state(struct drm_device *dev)
 		}
 
 		error->pipe[i].source = I915_READ(PIPESRC(i));
+
+		if (!HAS_PCH_SPLIT(dev))
+			error->pipe[i].stat = I915_READ(PIPESTAT(i));
 	}
 
 	error->num_transcoders = INTEL_INFO(dev)->num_pipes;
@@ -12024,7 +12487,7 @@ intel_display_capture_error_state(struct drm_device *dev)
 		enum transcoder cpu_transcoder = transcoders[i];
 
 		error->transcoder[i].power_domain_on =
-			intel_display_power_enabled_sw(dev_priv,
+			intel_display_power_enabled_unlocked(dev_priv,
 				POWER_DOMAIN_TRANSCODER(cpu_transcoder));
 		if (!error->transcoder[i].power_domain_on)
 			continue;
@@ -12064,6 +12527,7 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 		err_printf(m, "  Power: %s\n",
 			   error->pipe[i].power_domain_on ? "on" : "off");
 		err_printf(m, "  SRC: %08x\n", error->pipe[i].source);
+		err_printf(m, "  STAT: %08x\n", error->pipe[i].stat);
 
 		err_printf(m, "Plane [%d]:\n", i);
 		err_printf(m, "  CNTR: %08x\n", error->plane[i].control);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 2a00cb828d20..075170d1844f 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -28,6 +28,8 @@
 #include <linux/i2c.h>
 #include <linux/slab.h>
 #include <linux/export.h>
+#include <linux/notifier.h>
+#include <linux/reboot.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_crtc_helper.h>
@@ -64,6 +66,24 @@ static const struct dp_link_dpll vlv_dpll[] = {
 		{ .p1 = 2, .p2 = 2, .n = 1, .m1 = 2, .m2 = 27 } }
 };
 
+/*
+ * CHV supports eDP 1.4 that have  more link rates.
+ * Below only provides the fixed rate but exclude variable rate.
+ */
+static const struct dp_link_dpll chv_dpll[] = {
+	/*
+	 * CHV requires to program fractional division for m2.
+	 * m2 is stored in fixed point format using formula below
+	 * (m2_int << 22) | m2_fraction
+	 */
+	{ DP_LINK_BW_1_62,	/* m2_int = 32, m2_fraction = 1677722 */
+		{ .p1 = 4, .p2 = 2, .n = 1, .m1 = 2, .m2 = 0x819999a } },
+	{ DP_LINK_BW_2_7,	/* m2_int = 27, m2_fraction = 0 */
+		{ .p1 = 4, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } },
+	{ DP_LINK_BW_5_4,	/* m2_int = 27, m2_fraction = 0 */
+		{ .p1 = 2, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } }
+};
+
 /**
  * is_edp - is the given port attached to an eDP panel (either CPU or PCH)
  * @intel_dp: DP struct
@@ -318,6 +338,37 @@ static u32 _pp_stat_reg(struct intel_dp *intel_dp)
 		return VLV_PIPE_PP_STATUS(vlv_power_sequencer_pipe(intel_dp));
 }
 
+/* Reboot notifier handler to shutdown panel power to guarantee T12 timing
+   This function only applicable when panel PM state is not to be tracked */
+static int edp_notify_handler(struct notifier_block *this, unsigned long code,
+			      void *unused)
+{
+	struct intel_dp *intel_dp = container_of(this, typeof(* intel_dp),
+						 edp_notifier);
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 pp_div;
+	u32 pp_ctrl_reg, pp_div_reg;
+	enum pipe pipe = vlv_power_sequencer_pipe(intel_dp);
+
+	if (!is_edp(intel_dp) || code != SYS_RESTART)
+		return 0;
+
+	if (IS_VALLEYVIEW(dev)) {
+		pp_ctrl_reg = VLV_PIPE_PP_CONTROL(pipe);
+		pp_div_reg  = VLV_PIPE_PP_DIVISOR(pipe);
+		pp_div = I915_READ(pp_div_reg);
+		pp_div &= PP_REFERENCE_DIVIDER_MASK;
+
+		/* 0x1F write to PP_DIV_REG sets max cycle delay */
+		I915_WRITE(pp_div_reg, pp_div | 0x1F);
+		I915_WRITE(pp_ctrl_reg, PANEL_UNLOCK_REGS | PANEL_POWER_OFF);
+		msleep(intel_dp->panel_power_cycle_delay);
+	}
+
+	return 0;
+}
+
 static bool edp_have_panel_power(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
@@ -330,8 +381,12 @@ static bool edp_have_panel_vdd(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
+	enum intel_display_power_domain power_domain;
 
-	return !dev_priv->pm.suspended &&
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	return intel_display_power_enabled(dev_priv, power_domain) &&
 	       (I915_READ(_pp_ctrl_reg(intel_dp)) & EDP_FORCE_VDD) != 0;
 }
 
@@ -697,9 +752,9 @@ intel_dp_aux_init(struct intel_dp *intel_dp, struct intel_connector *connector)
 	DRM_DEBUG_KMS("registering %s bus for %s\n", name,
 		      connector->base.kdev->kobj.name);
 
-	ret = drm_dp_aux_register_i2c_bus(&intel_dp->aux);
+	ret = drm_dp_aux_register(&intel_dp->aux);
 	if (ret < 0) {
-		DRM_ERROR("drm_dp_aux_register_i2c_bus() for %s failed (%d)\n",
+		DRM_ERROR("drm_dp_aux_register() for %s failed (%d)\n",
 			  name, ret);
 		return;
 	}
@@ -709,7 +764,7 @@ intel_dp_aux_init(struct intel_dp *intel_dp, struct intel_connector *connector)
 				intel_dp->aux.ddc.dev.kobj.name);
 	if (ret < 0) {
 		DRM_ERROR("sysfs_create_link() for %s failed (%d)\n", name, ret);
-		drm_dp_aux_unregister_i2c_bus(&intel_dp->aux);
+		drm_dp_aux_unregister(&intel_dp->aux);
 	}
 }
 
@@ -739,6 +794,9 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 	} else if (HAS_PCH_SPLIT(dev)) {
 		divisor = pch_dpll;
 		count = ARRAY_SIZE(pch_dpll);
+	} else if (IS_CHERRYVIEW(dev)) {
+		divisor = chv_dpll;
+		count = ARRAY_SIZE(chv_dpll);
 	} else if (IS_VALLEYVIEW(dev)) {
 		divisor = vlv_dpll;
 		count = ARRAY_SIZE(vlv_dpll);
@@ -755,6 +813,20 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 	}
 }
 
+static void
+intel_dp_set_m2_n2(struct intel_crtc *crtc, struct intel_link_m_n *m_n)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum transcoder transcoder = crtc->config.cpu_transcoder;
+
+	I915_WRITE(PIPE_DATA_M2(transcoder),
+		TU_SIZE(m_n->tu) | m_n->gmch_m);
+	I915_WRITE(PIPE_DATA_N2(transcoder), m_n->gmch_n);
+	I915_WRITE(PIPE_LINK_M2(transcoder), m_n->link_m);
+	I915_WRITE(PIPE_LINK_N2(transcoder), m_n->link_n);
+}
+
 bool
 intel_dp_compute_config(struct intel_encoder *encoder,
 			struct intel_crtc_config *pipe_config)
@@ -780,6 +852,7 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 		pipe_config->has_pch_encoder = true;
 
 	pipe_config->has_dp_encoder = true;
+	pipe_config->has_audio = intel_dp->has_audio;
 
 	if (is_edp(intel_dp) && intel_connector->panel.fixed_mode) {
 		intel_fixed_panel_mode(intel_connector->panel.fixed_mode,
@@ -880,6 +953,14 @@ found:
 			       pipe_config->port_clock,
 			       &pipe_config->dp_m_n);
 
+	if (intel_connector->panel.downclock_mode != NULL &&
+		intel_dp->drrs_state.type == SEAMLESS_DRRS_SUPPORT) {
+			intel_link_compute_m_n(bpp, lane_count,
+				intel_connector->panel.downclock_mode->clock,
+				pipe_config->port_clock,
+				&pipe_config->dp_m2_n2);
+	}
+
 	intel_dp_set_clock(encoder, pipe_config, intel_dp->link_bw);
 
 	return true;
@@ -915,7 +996,7 @@ static void ironlake_set_pll_cpu_edp(struct intel_dp *intel_dp)
 	udelay(500);
 }
 
-static void intel_dp_mode_set(struct intel_encoder *encoder)
+static void intel_dp_prepare(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -950,7 +1031,7 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 	intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
 	intel_dp->DP |= DP_PORT_WIDTH(intel_dp->lane_count);
 
-	if (intel_dp->has_audio) {
+	if (crtc->config.has_audio) {
 		DRM_DEBUG_DRIVER("Enabling DP audio on pipe %c\n",
 				 pipe_name(crtc->pipe));
 		intel_dp->DP |= DP_AUDIO_OUTPUT_ENABLE;
@@ -983,14 +1064,15 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
-		if (crtc->pipe == 1)
-			intel_dp->DP |= DP_PIPEB_SELECT;
+		if (!IS_CHERRYVIEW(dev)) {
+			if (crtc->pipe == 1)
+				intel_dp->DP |= DP_PIPEB_SELECT;
+		} else {
+			intel_dp->DP |= DP_PIPE_SELECT_CHV(crtc->pipe);
+		}
 	} else {
 		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 	}
-
-	if (port == PORT_A && !IS_VALLEYVIEW(dev))
-		ironlake_set_pll_cpu_edp(intel_dp);
 }
 
 #define IDLE_ON_MASK		(PP_ON | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
@@ -1082,7 +1164,10 @@ static  u32 ironlake_get_pp_control(struct intel_dp *intel_dp)
 static bool _edp_panel_vdd_on(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	u32 pp;
 	u32 pp_stat_reg, pp_ctrl_reg;
 	bool need_to_disable = !intel_dp->want_panel_vdd;
@@ -1095,7 +1180,8 @@ static bool _edp_panel_vdd_on(struct intel_dp *intel_dp)
 	if (edp_have_panel_vdd(intel_dp))
 		return need_to_disable;
 
-	intel_runtime_pm_get(dev_priv);
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
 
 	DRM_DEBUG_KMS("Turning eDP VDD on\n");
 
@@ -1139,9 +1225,14 @@ static void edp_panel_vdd_off_sync(struct intel_dp *intel_dp)
 	u32 pp;
 	u32 pp_stat_reg, pp_ctrl_reg;
 
-	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	if (!intel_dp->want_panel_vdd && edp_have_panel_vdd(intel_dp)) {
+		struct intel_digital_port *intel_dig_port =
+						dp_to_dig_port(intel_dp);
+		struct intel_encoder *intel_encoder = &intel_dig_port->base;
+		enum intel_display_power_domain power_domain;
+
 		DRM_DEBUG_KMS("Turning eDP VDD off\n");
 
 		pp = ironlake_get_pp_control(intel_dp);
@@ -1160,7 +1251,8 @@ static void edp_panel_vdd_off_sync(struct intel_dp *intel_dp)
 		if ((pp & POWER_TARGET_ON) == 0)
 			intel_dp->last_power_cycle = jiffies;
 
-		intel_runtime_pm_put(dev_priv);
+		power_domain = intel_display_port_power_domain(intel_encoder);
+		intel_display_power_put(dev_priv, power_domain);
 	}
 }
 
@@ -1170,9 +1262,9 @@ static void edp_panel_vdd_work(struct work_struct *__work)
 						 struct intel_dp, panel_vdd_work);
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	edp_panel_vdd_off_sync(intel_dp);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 }
 
 static void edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
@@ -1244,8 +1336,11 @@ void intel_edp_panel_on(struct intel_dp *intel_dp)
 
 void intel_edp_panel_off(struct intel_dp *intel_dp)
 {
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	u32 pp;
 	u32 pp_ctrl_reg;
 
@@ -1275,7 +1370,8 @@ void intel_edp_panel_off(struct intel_dp *intel_dp)
 	wait_panel_off(intel_dp);
 
 	/* We got a reference when we enabled the VDD. */
-	intel_runtime_pm_put(dev_priv);
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_put(dev_priv, power_domain);
 }
 
 void intel_edp_backlight_on(struct intel_dp *intel_dp)
@@ -1432,6 +1528,8 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 
 	if (port == PORT_A && IS_GEN7(dev) && !IS_VALLEYVIEW(dev)) {
 		*pipe = PORT_TO_PIPE_CPT(tmp);
+	} else if (IS_CHERRYVIEW(dev)) {
+		*pipe = DP_PORT_TO_PIPE_CHV(tmp);
 	} else if (!HAS_PCH_CPT(dev) || port == PORT_A) {
 		*pipe = PORT_TO_PIPE(tmp);
 	} else {
@@ -1479,8 +1577,11 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	int dotclock;
 
+	tmp = I915_READ(intel_dp->output_reg);
+	if (tmp & DP_AUDIO_OUTPUT_ENABLE)
+		pipe_config->has_audio = true;
+
 	if ((port == PORT_A) || !HAS_PCH_CPT(dev)) {
-		tmp = I915_READ(intel_dp->output_reg);
 		if (tmp & DP_SYNC_HS_HIGH)
 			flags |= DRM_MODE_FLAG_PHSYNC;
 		else
@@ -1816,17 +1917,59 @@ static void intel_disable_dp(struct intel_encoder *encoder)
 		intel_dp_link_down(intel_dp);
 }
 
-static void intel_post_disable_dp(struct intel_encoder *encoder)
+static void g4x_post_disable_dp(struct intel_encoder *encoder)
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	enum port port = dp_to_dig_port(intel_dp)->port;
+
+	if (port != PORT_A)
+		return;
+
+	intel_dp_link_down(intel_dp);
+	ironlake_edp_pll_off(intel_dp);
+}
+
+static void vlv_post_disable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+
+	intel_dp_link_down(intel_dp);
+}
+
+static void chv_post_disable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 val;
 
-	if (port == PORT_A || IS_VALLEYVIEW(dev)) {
-		intel_dp_link_down(intel_dp);
-		if (!IS_VALLEYVIEW(dev))
-			ironlake_edp_pll_off(intel_dp);
-	}
+	intel_dp_link_down(intel_dp);
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Propagate soft reset to data lane reset */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 static void intel_enable_dp(struct intel_encoder *encoder)
@@ -1868,8 +2011,13 @@ static void g4x_pre_enable_dp(struct intel_encoder *encoder)
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
 
-	if (dport->port == PORT_A)
+	intel_dp_prepare(encoder);
+
+	/* Only ilk+ has port A */
+	if (dport->port == PORT_A) {
+		ironlake_set_pll_cpu_edp(intel_dp);
 		ironlake_edp_pll_on(intel_dp);
+	}
 }
 
 static void vlv_pre_enable_dp(struct intel_encoder *encoder)
@@ -1921,6 +2069,8 @@ static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 
+	intel_dp_prepare(encoder);
+
 	/* Program Tx lane resets to default */
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW0(port),
@@ -1939,6 +2089,69 @@ static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
+static void chv_pre_enable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct edp_power_seq power_seq;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
+	int data, i;
+	u32 val;
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Deassert soft data lane reset*/
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	/* Program Tx lane latency optimal setting*/
+	for (i = 0; i < 4; i++) {
+		/* Set the latency optimal bit */
+		data = (i == 1) ? 0x0 : 0x6;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW11(ch, i),
+				data << DPIO_FRC_LATENCY_SHFIT);
+
+		/* Set the upar bit */
+		data = (i == 1) ? 0x0 : 0x1;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW14(ch, i),
+				data << DPIO_UPAR_SHIFT);
+	}
+
+	/* Data lane stagger programming */
+	/* FIXME: Fix up value only after power analysis */
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	if (is_edp(intel_dp)) {
+		/* init power sequencer on this pipe and port */
+		intel_dp_init_panel_power_sequencer(dev, intel_dp, &power_seq);
+		intel_dp_init_panel_power_sequencer_registers(dev, intel_dp,
+							      &power_seq);
+	}
+
+	intel_enable_dp(encoder);
+
+	vlv_wait_port_ready(dev_priv, dport);
+}
+
 /*
  * Native read with retry for link status and receiver capability reads for
  * cases where the sink may still be asleep.
@@ -2163,6 +2376,166 @@ static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
 	return 0;
 }
 
+static uint32_t intel_chv_signal_levels(struct intel_dp *intel_dp)
+{
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
+	struct intel_crtc *intel_crtc = to_intel_crtc(dport->base.base.crtc);
+	u32 deemph_reg_value, margin_reg_value, val;
+	uint8_t train_set = intel_dp->train_set[0];
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	enum pipe pipe = intel_crtc->pipe;
+	int i;
+
+	switch (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) {
+	case DP_TRAIN_PRE_EMPHASIS_0:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 128;
+			margin_reg_value = 52;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			deemph_reg_value = 128;
+			margin_reg_value = 77;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			deemph_reg_value = 128;
+			margin_reg_value = 102;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_1200:
+			deemph_reg_value = 128;
+			margin_reg_value = 154;
+			/* FIXME extra to set for 1200 */
+			break;
+		default:
+			return 0;
+		}
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_3_5:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 85;
+			margin_reg_value = 78;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			deemph_reg_value = 85;
+			margin_reg_value = 116;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			deemph_reg_value = 85;
+			margin_reg_value = 154;
+			break;
+		default:
+			return 0;
+		}
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_6:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 64;
+			margin_reg_value = 104;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			deemph_reg_value = 64;
+			margin_reg_value = 154;
+			break;
+		default:
+			return 0;
+		}
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_9_5:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 43;
+			margin_reg_value = 154;
+			break;
+		default:
+			return 0;
+		}
+		break;
+	default:
+		return 0;
+	}
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Clear calc init */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* Program swing deemph */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW4(ch, i));
+		val &= ~DPIO_SWING_DEEMPH9P5_MASK;
+		val |= deemph_reg_value << DPIO_SWING_DEEMPH9P5_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW4(ch, i), val);
+	}
+
+	/* Program swing margin */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+		val &= ~DPIO_SWING_MARGIN_MASK;
+		val |= margin_reg_value << DPIO_SWING_MARGIN_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
+	}
+
+	/* Disable unique transition scale */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+		val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
+	}
+
+	if (((train_set & DP_TRAIN_PRE_EMPHASIS_MASK)
+			== DP_TRAIN_PRE_EMPHASIS_0) &&
+		((train_set & DP_TRAIN_VOLTAGE_SWING_MASK)
+			== DP_TRAIN_VOLTAGE_SWING_1200)) {
+
+		/*
+		 * The document said it needs to set bit 27 for ch0 and bit 26
+		 * for ch1. Might be a typo in the doc.
+		 * For now, for this unique transition scale selection, set bit
+		 * 27 for ch0 and ch1.
+		 */
+		for (i = 0; i < 4; i++) {
+			val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+			val |= DPIO_TX_UNIQ_TRANS_SCALE_EN;
+			vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
+		}
+
+		for (i = 0; i < 4; i++) {
+			val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+			val &= ~(0xff << DPIO_UNIQ_TRANS_SCALE_SHIFT);
+			val |= (0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT);
+			vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
+		}
+	}
+
+	/* Start swing calculation */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* LRC Bypass */
+	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
+	val |= DPIO_LRC_BYPASS;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	return 0;
+}
+
 static void
 intel_get_adjust_train(struct intel_dp *intel_dp,
 		       const uint8_t link_status[DP_LINK_STATUS_SIZE])
@@ -2377,6 +2750,9 @@ intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
 	} else if (IS_HASWELL(dev)) {
 		signal_levels = intel_hsw_signal_levels(train_set);
 		mask = DDI_BUF_EMP_MASK;
+	} else if (IS_CHERRYVIEW(dev)) {
+		signal_levels = intel_chv_signal_levels(intel_dp);
+		mask = 0;
 	} else if (IS_VALLEYVIEW(dev)) {
 		signal_levels = intel_vlv_signal_levels(intel_dp);
 		mask = 0;
@@ -2743,22 +3119,7 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 		to_intel_crtc(intel_dig_port->base.base.crtc);
 	uint32_t DP = intel_dp->DP;
 
-	/*
-	 * DDI code has a strict mode set sequence and we should try to respect
-	 * it, otherwise we might hang the machine in many different ways. So we
-	 * really should be disabling the port only on a complete crtc_disable
-	 * sequence. This function is just called under two conditions on DDI
-	 * code:
-	 * - Link train failed while doing crtc_enable, and on this case we
-	 *   really should respect the mode set sequence and wait for a
-	 *   crtc_disable.
-	 * - Someone turned the monitor off and intel_dp_check_link_status
-	 *   called us. We don't need to disable the whole port on this case, so
-	 *   when someone turns the monitor on again,
-	 *   intel_ddi_prepare_link_retrain will take care of redoing the link
-	 *   train.
-	 */
-	if (HAS_DDI(dev))
+	if (WARN_ON(HAS_DDI(dev)))
 		return;
 
 	if (WARN_ON((I915_READ(intel_dp->output_reg) & DP_PORT_EN) == 0))
@@ -2775,9 +3136,6 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 	}
 	POSTING_READ(intel_dp->output_reg);
 
-	/* We don't really know why we're doing this */
-	intel_wait_for_vblank(dev, intel_crtc->pipe);
-
 	if (HAS_PCH_IBX(dev) &&
 	    I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
 		struct drm_crtc *crtc = intel_dig_port->base.base.crtc;
@@ -2948,6 +3306,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 	u8 sink_irq_vector;
 	u8 link_status[DP_LINK_STATUS_SIZE];
 
+	/* FIXME: This access isn't protected by any locks. */
 	if (!intel_encoder->connectors_active)
 		return;
 
@@ -2980,7 +3339,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 
 	if (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) {
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
-			      drm_get_encoder_name(&intel_encoder->base));
+			      intel_encoder->base.name);
 		intel_dp_start_link_train(intel_dp);
 		intel_dp_complete_link_train(intel_dp);
 		intel_dp_stop_link_train(intel_dp);
@@ -3166,7 +3525,7 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	intel_display_power_get(dev_priv, power_domain);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	intel_dp->has_audio = false;
 
@@ -3374,13 +3733,17 @@ void intel_dp_encoder_destroy(struct drm_encoder *encoder)
 	struct intel_dp *intel_dp = &intel_dig_port->dp;
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 
-	drm_dp_aux_unregister_i2c_bus(&intel_dp->aux);
+	drm_dp_aux_unregister(&intel_dp->aux);
 	drm_encoder_cleanup(encoder);
 	if (is_edp(intel_dp)) {
 		cancel_delayed_work_sync(&intel_dp->panel_vdd_work);
-		mutex_lock(&dev->mode_config.mutex);
+		drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 		edp_panel_vdd_off_sync(intel_dp);
-		mutex_unlock(&dev->mode_config.mutex);
+		drm_modeset_unlock(&dev->mode_config.connection_mutex);
+		if (intel_dp->edp_notifier.notifier_call) {
+			unregister_reboot_notifier(&intel_dp->edp_notifier);
+			intel_dp->edp_notifier.notifier_call = NULL;
+		}
 	}
 	kfree(intel_dig_port);
 }
@@ -3651,6 +4014,130 @@ intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
 		      I915_READ(pp_div_reg));
 }
 
+void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_encoder *encoder;
+	struct intel_dp *intel_dp = NULL;
+	struct intel_crtc_config *config = NULL;
+	struct intel_crtc *intel_crtc = NULL;
+	struct intel_connector *intel_connector = dev_priv->drrs.connector;
+	u32 reg, val;
+	enum edp_drrs_refresh_rate_type index = DRRS_HIGH_RR;
+
+	if (refresh_rate <= 0) {
+		DRM_DEBUG_KMS("Refresh rate should be positive non-zero.\n");
+		return;
+	}
+
+	if (intel_connector == NULL) {
+		DRM_DEBUG_KMS("DRRS supported for eDP only.\n");
+		return;
+	}
+
+	if (INTEL_INFO(dev)->gen < 8 && intel_edp_is_psr_enabled(dev)) {
+		DRM_DEBUG_KMS("DRRS is disabled as PSR is enabled\n");
+		return;
+	}
+
+	encoder = intel_attached_encoder(&intel_connector->base);
+	intel_dp = enc_to_intel_dp(&encoder->base);
+	intel_crtc = encoder->new_crtc;
+
+	if (!intel_crtc) {
+		DRM_DEBUG_KMS("DRRS: intel_crtc not initialized\n");
+		return;
+	}
+
+	config = &intel_crtc->config;
+
+	if (intel_dp->drrs_state.type < SEAMLESS_DRRS_SUPPORT) {
+		DRM_DEBUG_KMS("Only Seamless DRRS supported.\n");
+		return;
+	}
+
+	if (intel_connector->panel.downclock_mode->vrefresh == refresh_rate)
+		index = DRRS_LOW_RR;
+
+	if (index == intel_dp->drrs_state.refresh_rate_type) {
+		DRM_DEBUG_KMS(
+			"DRRS requested for previously set RR...ignoring\n");
+		return;
+	}
+
+	if (!intel_crtc->active) {
+		DRM_DEBUG_KMS("eDP encoder disabled. CRTC not Active\n");
+		return;
+	}
+
+	if (INTEL_INFO(dev)->gen > 6 && INTEL_INFO(dev)->gen < 8) {
+		reg = PIPECONF(intel_crtc->config.cpu_transcoder);
+		val = I915_READ(reg);
+		if (index > DRRS_HIGH_RR) {
+			val |= PIPECONF_EDP_RR_MODE_SWITCH;
+			intel_dp_set_m2_n2(intel_crtc, &config->dp_m2_n2);
+		} else {
+			val &= ~PIPECONF_EDP_RR_MODE_SWITCH;
+		}
+		I915_WRITE(reg, val);
+	}
+
+	/*
+	 * mutex taken to ensure that there is no race between differnt
+	 * drrs calls trying to update refresh rate. This scenario may occur
+	 * in future when idleness detection based DRRS in kernel and
+	 * possible calls from user space to set differnt RR are made.
+	 */
+
+	mutex_lock(&intel_dp->drrs_state.mutex);
+
+	intel_dp->drrs_state.refresh_rate_type = index;
+
+	mutex_unlock(&intel_dp->drrs_state.mutex);
+
+	DRM_DEBUG_KMS("eDP Refresh Rate set to : %dHz\n", refresh_rate);
+}
+
+static struct drm_display_mode *
+intel_dp_drrs_init(struct intel_digital_port *intel_dig_port,
+			struct intel_connector *intel_connector,
+			struct drm_display_mode *fixed_mode)
+{
+	struct drm_connector *connector = &intel_connector->base;
+	struct intel_dp *intel_dp = &intel_dig_port->dp;
+	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_display_mode *downclock_mode = NULL;
+
+	if (INTEL_INFO(dev)->gen <= 6) {
+		DRM_DEBUG_KMS("DRRS supported for Gen7 and above\n");
+		return NULL;
+	}
+
+	if (dev_priv->vbt.drrs_type != SEAMLESS_DRRS_SUPPORT) {
+		DRM_INFO("VBT doesn't support DRRS\n");
+		return NULL;
+	}
+
+	downclock_mode = intel_find_panel_downclock
+					(dev, fixed_mode, connector);
+
+	if (!downclock_mode) {
+		DRM_INFO("DRRS not supported\n");
+		return NULL;
+	}
+
+	dev_priv->drrs.connector = intel_connector;
+
+	mutex_init(&intel_dp->drrs_state.mutex);
+
+	intel_dp->drrs_state.type = dev_priv->vbt.drrs_type;
+
+	intel_dp->drrs_state.refresh_rate_type = DRRS_HIGH_RR;
+	DRM_INFO("seamless DRRS supported for eDP panel.\n");
+	return downclock_mode;
+}
+
 static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 				     struct intel_connector *intel_connector,
 				     struct edp_power_seq *power_seq)
@@ -3661,10 +4148,13 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_display_mode *fixed_mode = NULL;
+	struct drm_display_mode *downclock_mode = NULL;
 	bool has_dpcd;
 	struct drm_display_mode *scan;
 	struct edid *edid;
 
+	intel_dp->drrs_state.type = DRRS_NOT_SUPPORTED;
+
 	if (!is_edp(intel_dp))
 		return true;
 
@@ -3715,6 +4205,9 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	list_for_each_entry(scan, &connector->probed_modes, head) {
 		if ((scan->type & DRM_MODE_TYPE_PREFERRED)) {
 			fixed_mode = drm_mode_duplicate(dev, scan);
+			downclock_mode = intel_dp_drrs_init(
+						intel_dig_port,
+						intel_connector, fixed_mode);
 			break;
 		}
 	}
@@ -3728,7 +4221,12 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	}
 	mutex_unlock(&dev->mode_config.mutex);
 
-	intel_panel_init(&intel_connector->panel, fixed_mode, NULL);
+	if (IS_VALLEYVIEW(dev)) {
+		intel_dp->edp_notifier.notifier_call = edp_notify_handler;
+		register_reboot_notifier(&intel_dp->edp_notifier);
+	}
+
+	intel_panel_init(&intel_connector->panel, fixed_mode, downclock_mode);
 	intel_panel_setup_backlight(connector);
 
 	return true;
@@ -3826,12 +4324,12 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_dp->psr_setup_done = false;
 
 	if (!intel_edp_init_connector(intel_dp, intel_connector, &power_seq)) {
-		drm_dp_aux_unregister_i2c_bus(&intel_dp->aux);
+		drm_dp_aux_unregister(&intel_dp->aux);
 		if (is_edp(intel_dp)) {
 			cancel_delayed_work_sync(&intel_dp->panel_vdd_work);
-			mutex_lock(&dev->mode_config.mutex);
+			drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 			edp_panel_vdd_off_sync(intel_dp);
-			mutex_unlock(&dev->mode_config.mutex);
+			drm_modeset_unlock(&dev->mode_config.connection_mutex);
 		}
 		drm_sysfs_connector_remove(connector);
 		drm_connector_cleanup(connector);
@@ -3877,25 +4375,36 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_dp_compute_config;
-	intel_encoder->mode_set = intel_dp_mode_set;
 	intel_encoder->disable = intel_disable_dp;
-	intel_encoder->post_disable = intel_post_disable_dp;
 	intel_encoder->get_hw_state = intel_dp_get_hw_state;
 	intel_encoder->get_config = intel_dp_get_config;
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		intel_encoder->pre_enable = chv_pre_enable_dp;
+		intel_encoder->enable = vlv_enable_dp;
+		intel_encoder->post_disable = chv_post_disable_dp;
+	} else if (IS_VALLEYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = vlv_dp_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_pre_enable_dp;
 		intel_encoder->enable = vlv_enable_dp;
+		intel_encoder->post_disable = vlv_post_disable_dp;
 	} else {
 		intel_encoder->pre_enable = g4x_pre_enable_dp;
 		intel_encoder->enable = g4x_enable_dp;
+		intel_encoder->post_disable = g4x_post_disable_dp;
 	}
 
 	intel_dig_port->port = port;
 	intel_dig_port->dp.output_reg = output_reg;
 
 	intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
-	intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	if (IS_CHERRYVIEW(dev)) {
+		if (port == PORT_D)
+			intel_encoder->crtc_mask = 1 << 2;
+		else
+			intel_encoder->crtc_mask = (1 << 0) | (1 << 1);
+	} else {
+		intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	}
 	intel_encoder->cloneable = 0;
 	intel_encoder->hot_plug = intel_dp_hot_plug;
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 328b1a70264b..f67340ed2c12 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -81,8 +81,8 @@
 /* Maximum cursor sizes */
 #define GEN2_CURSOR_WIDTH 64
 #define GEN2_CURSOR_HEIGHT 64
-#define CURSOR_WIDTH 256
-#define CURSOR_HEIGHT 256
+#define MAX_CURSOR_WIDTH 256
+#define MAX_CURSOR_HEIGHT 256
 
 #define INTEL_I2C_BUS_DVO 1
 #define INTEL_I2C_BUS_SDVO 2
@@ -106,8 +106,8 @@
 #define INTEL_DVO_CHIP_TMDS 2
 #define INTEL_DVO_CHIP_TVOUT 4
 
-#define INTEL_DSI_COMMAND_MODE	0
-#define INTEL_DSI_VIDEO_MODE	1
+#define INTEL_DSI_VIDEO_MODE	0
+#define INTEL_DSI_COMMAND_MODE	1
 
 struct intel_framebuffer {
 	struct drm_framebuffer base;
@@ -273,6 +273,13 @@ struct intel_crtc_config {
 	 * accordingly. */
 	bool has_dp_encoder;
 
+	/* Whether we should send NULL infoframes. Required for audio. */
+	bool has_hdmi_sink;
+
+	/* Audio enabled on this pipe. Only valid if either has_hdmi_sink or
+	 * has_dp_encoder is set. */
+	bool has_audio;
+
 	/*
 	 * Enable dithering, used when the selected pipe bpp doesn't match the
 	 * plane bpp.
@@ -306,6 +313,9 @@ struct intel_crtc_config {
 	int pipe_bpp;
 	struct intel_link_m_n dp_m_n;
 
+	/* m2_n2 for eDP downclock */
+	struct intel_link_m_n dp_m2_n2;
+
 	/*
 	 * Frequence the dpll for the port should run at. Differs from the
 	 * adjusted dotclock e.g. for DP or 12bpc hdmi mode. This is also
@@ -343,6 +353,9 @@ struct intel_pipe_wm {
 	struct intel_wm_level wm[5];
 	uint32_t linetime;
 	bool fbc_wm_enabled;
+	bool pipe_enabled;
+	bool sprites_enabled;
+	bool sprites_scaled;
 };
 
 struct intel_crtc {
@@ -357,7 +370,6 @@ struct intel_crtc {
 	 */
 	bool active;
 	unsigned long enabled_power_domains;
-	bool eld_vld;
 	bool primary_enabled; /* is the primary plane (partially) visible? */
 	bool lowfreq_avail;
 	struct intel_overlay *overlay;
@@ -374,8 +386,8 @@ struct intel_crtc {
 	uint32_t cursor_addr;
 	int16_t cursor_x, cursor_y;
 	int16_t cursor_width, cursor_height;
-	int16_t max_cursor_width, max_cursor_height;
-	bool cursor_visible;
+	uint32_t cursor_cntl;
+	uint32_t cursor_base;
 
 	struct intel_plane_config plane_config;
 	struct intel_crtc_config config;
@@ -396,6 +408,10 @@ struct intel_crtc {
 		/* watermarks currently being used  */
 		struct intel_pipe_wm active;
 	} wm;
+
+	wait_queue_head_t vbl_wait;
+
+	int scanline_offset;
 };
 
 struct intel_plane_wm_parameters {
@@ -479,11 +495,23 @@ struct intel_hdmi {
 				enum hdmi_infoframe_type type,
 				const void *frame, ssize_t len);
 	void (*set_infoframes)(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode);
 };
 
 #define DP_MAX_DOWNSTREAM_PORTS		0x10
 
+/**
+ * HIGH_RR is the highest eDP panel refresh rate read from EDID
+ * LOW_RR is the lowest eDP panel refresh rate found from EDID
+ * parsing for same resolution.
+ */
+enum edp_drrs_refresh_rate_type {
+	DRRS_HIGH_RR,
+	DRRS_LOW_RR,
+	DRRS_MAX_RR, /* RR count */
+};
+
 struct intel_dp {
 	uint32_t output_reg;
 	uint32_t aux_ch_ctl_reg;
@@ -510,6 +538,8 @@ struct intel_dp {
 	unsigned long last_power_on;
 	unsigned long last_backlight_off;
 	bool psr_setup_done;
+	struct notifier_block edp_notifier;
+
 	bool use_tps3;
 	struct intel_connector *attached_connector;
 
@@ -522,6 +552,12 @@ struct intel_dp {
 				     bool has_aux_irq,
 				     int send_bytes,
 				     uint32_t aux_clock_divider);
+	struct {
+		enum drrs_support_type type;
+		enum edp_drrs_refresh_rate_type refresh_rate_type;
+		struct mutex mutex;
+	} drrs_state;
+
 };
 
 struct intel_digital_port {
@@ -537,6 +573,7 @@ vlv_dport_to_channel(struct intel_digital_port *dport)
 {
 	switch (dport->port) {
 	case PORT_B:
+	case PORT_D:
 		return DPIO_CH0;
 	case PORT_C:
 		return DPIO_CH1;
@@ -545,6 +582,20 @@ vlv_dport_to_channel(struct intel_digital_port *dport)
 	}
 }
 
+static inline int
+vlv_pipe_to_channel(enum pipe pipe)
+{
+	switch (pipe) {
+	case PIPE_A:
+	case PIPE_C:
+		return DPIO_CH0;
+	case PIPE_B:
+		return DPIO_CH1;
+	default:
+		BUG();
+	}
+}
+
 static inline struct drm_crtc *
 intel_get_crtc_for_pipe(struct drm_device *dev, int pipe)
 {
@@ -569,6 +620,8 @@ struct intel_unpin_work {
 #define INTEL_FLIP_INACTIVE	0
 #define INTEL_FLIP_PENDING	1
 #define INTEL_FLIP_COMPLETE	2
+	u32 flip_count;
+	u32 gtt_offset;
 	bool enable_stall_check;
 };
 
@@ -620,8 +673,6 @@ hdmi_to_dig_port(struct intel_hdmi *intel_hdmi)
 /* i915_irq.c */
 bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
 					   enum pipe pipe, bool enable);
-bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
-					     enum pipe pipe, bool enable);
 bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 					   enum transcoder pch_transcoder,
 					   bool enable);
@@ -629,8 +680,12 @@ void ilk_enable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
 void ilk_disable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
 void snb_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
 void snb_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
-void hsw_runtime_pm_disable_interrupts(struct drm_device *dev);
-void hsw_runtime_pm_restore_interrupts(struct drm_device *dev);
+void bdw_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void bdw_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void intel_runtime_pm_disable_interrupts(struct drm_device *dev);
+void intel_runtime_pm_restore_interrupts(struct drm_device *dev);
+int intel_get_crtc_scanline(struct intel_crtc *crtc);
+void i9xx_check_fifo_underruns(struct drm_device *dev);
 
 
 /* intel_crt.c */
@@ -666,9 +721,10 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 const char *intel_output_name(int output);
 bool intel_has_pending_fb_unpin(struct drm_device *dev);
 int intel_pch_rawclk(struct drm_device *dev);
+int valleyview_cur_cdclk(struct drm_i915_private *dev_priv);
 void intel_mark_busy(struct drm_device *dev);
 void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			struct intel_ring_buffer *ring);
+			struct intel_engine_cs *ring);
 void intel_mark_idle(struct drm_device *dev);
 void intel_crtc_restore_mode(struct drm_crtc *crtc);
 void intel_crtc_update_dpms(struct drm_crtc *crtc);
@@ -695,12 +751,14 @@ void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 			 struct intel_digital_port *dport);
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct drm_display_mode *mode,
-				struct intel_load_detect_pipe *old);
+				struct intel_load_detect_pipe *old,
+				struct drm_modeset_acquire_ctx *ctx);
 void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old);
+				    struct intel_load_detect_pipe *old,
+				    struct drm_modeset_acquire_ctx *ctx);
 int intel_pin_and_fence_fb_obj(struct drm_device *dev,
 			       struct drm_i915_gem_object *obj,
-			       struct intel_ring_buffer *pipelined);
+			       struct intel_engine_cs *pipelined);
 void intel_unpin_fb_obj(struct drm_i915_gem_object *obj);
 struct drm_framebuffer *
 __intel_framebuffer_create(struct drm_device *dev,
@@ -751,6 +809,8 @@ int valleyview_get_vco(struct drm_i915_private *dev_priv);
 void intel_mode_from_pipe_config(struct drm_display_mode *mode,
 				 struct intel_crtc_config *pipe_config);
 int intel_format_to_fourcc(int format);
+void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc);
+
 
 /* intel_dp.c */
 void intel_dp_init(struct drm_device *dev, int output_reg, enum port port);
@@ -774,7 +834,7 @@ void intel_edp_panel_off(struct intel_dp *intel_dp);
 void intel_edp_psr_enable(struct intel_dp *intel_dp);
 void intel_edp_psr_disable(struct intel_dp *intel_dp);
 void intel_edp_psr_update(struct drm_device *dev);
-
+void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate);
 
 /* intel_dsi.c */
 bool intel_dsi_init(struct drm_device *dev);
@@ -876,6 +936,7 @@ extern struct drm_display_mode *intel_find_panel_downclock(
 /* intel_pm.c */
 void intel_init_clock_gating(struct drm_device *dev);
 void intel_suspend_hw(struct drm_device *dev);
+int ilk_wm_max_level(const struct drm_device *dev);
 void intel_update_watermarks(struct drm_crtc *crtc);
 void intel_update_sprite_watermarks(struct drm_plane *plane,
 				    struct drm_crtc *crtc,
@@ -891,8 +952,8 @@ int intel_power_domains_init(struct drm_i915_private *);
 void intel_power_domains_remove(struct drm_i915_private *);
 bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
 				 enum intel_display_power_domain domain);
-bool intel_display_power_enabled_sw(struct drm_i915_private *dev_priv,
-				    enum intel_display_power_domain domain);
+bool intel_display_power_enabled_unlocked(struct drm_i915_private *dev_priv,
+					  enum intel_display_power_domain domain);
 void intel_display_power_get(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain);
 void intel_display_power_put(struct drm_i915_private *dev_priv,
@@ -902,6 +963,7 @@ void intel_init_gt_powersave(struct drm_device *dev);
 void intel_cleanup_gt_powersave(struct drm_device *dev);
 void intel_enable_gt_powersave(struct drm_device *dev);
 void intel_disable_gt_powersave(struct drm_device *dev);
+void intel_reset_gt_powersave(struct drm_device *dev);
 void ironlake_teardown_rc6(struct drm_device *dev);
 void gen6_update_ring_freq(struct drm_device *dev);
 void gen6_rps_idle(struct drm_i915_private *dev_priv);
@@ -909,11 +971,13 @@ void gen6_rps_boost(struct drm_i915_private *dev_priv);
 void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv);
 void intel_aux_display_runtime_put(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_get(struct drm_i915_private *dev_priv);
+void intel_runtime_pm_get_noresume(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_put(struct drm_i915_private *dev_priv);
 void intel_init_runtime_pm(struct drm_i915_private *dev_priv);
 void intel_fini_runtime_pm(struct drm_i915_private *dev_priv);
 void ilk_wm_get_hw_state(struct drm_device *dev);
-
+void __vlv_set_power_well(struct drm_i915_private *dev_priv,
+			  enum punit_power_well power_well_id, bool enable);
 
 /* intel_sdvo.c */
 bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob);
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 33656647f8bc..3fd082933c87 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -35,6 +35,11 @@
 
 /* the sub-encoders aka panel drivers */
 static const struct intel_dsi_device intel_dsi_devices[] = {
+	{
+		.panel_id = MIPI_DSI_GENERIC_PANEL_ID,
+		.name = "vbt-generic-dsi-vid-mode-display",
+		.dev_ops = &vbt_generic_dsi_display_ops,
+	},
 };
 
 static void band_gap_reset(struct drm_i915_private *dev_priv)
@@ -59,12 +64,12 @@ static struct intel_dsi *intel_attached_dsi(struct drm_connector *connector)
 
 static inline bool is_vid_mode(struct intel_dsi *intel_dsi)
 {
-	return intel_dsi->dev.type == INTEL_DSI_VIDEO_MODE;
+	return intel_dsi->operation_mode == INTEL_DSI_VIDEO_MODE;
 }
 
 static inline bool is_cmd_mode(struct intel_dsi *intel_dsi)
 {
-	return intel_dsi->dev.type == INTEL_DSI_COMMAND_MODE;
+	return intel_dsi->operation_mode == INTEL_DSI_COMMAND_MODE;
 }
 
 static void intel_dsi_hot_plug(struct intel_encoder *encoder)
@@ -94,13 +99,6 @@ static bool intel_dsi_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_dsi_pre_pll_enable(struct intel_encoder *encoder)
-{
-	DRM_DEBUG_KMS("\n");
-
-	vlv_enable_dsi_pll(encoder);
-}
-
 static void intel_dsi_device_ready(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
@@ -110,32 +108,27 @@ static void intel_dsi_device_ready(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
+	mutex_lock(&dev_priv->dpio_lock);
+	/* program rcomp for compliance, reduce from 50 ohms to 45 ohms
+	 * needed everytime after power gate */
+	vlv_flisdsi_write(dev_priv, 0x04, 0x0004);
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	/* bandgap reset is needed after everytime we do power gate */
+	band_gap_reset(dev_priv);
+
+	I915_WRITE(MIPI_DEVICE_READY(pipe), ULPS_STATE_ENTER);
+	usleep_range(2500, 3000);
+
 	val = I915_READ(MIPI_PORT_CTRL(pipe));
 	I915_WRITE(MIPI_PORT_CTRL(pipe), val | LP_OUTPUT_HOLD);
 	usleep_range(1000, 1500);
-	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY | ULPS_STATE_EXIT);
-	usleep_range(2000, 2500);
-	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY);
-	usleep_range(2000, 2500);
-	I915_WRITE(MIPI_DEVICE_READY(pipe), 0x00);
-	usleep_range(2000, 2500);
-	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY);
-	usleep_range(2000, 2500);
-}
-static void intel_dsi_pre_enable(struct intel_encoder *encoder)
-{
-	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
-
-	DRM_DEBUG_KMS("\n");
-
-	if (intel_dsi->dev.dev_ops->panel_reset)
-		intel_dsi->dev.dev_ops->panel_reset(&intel_dsi->dev);
 
-	/* put device in ready state */
-	intel_dsi_device_ready(encoder);
+	I915_WRITE(MIPI_DEVICE_READY(pipe), ULPS_STATE_EXIT);
+	usleep_range(2500, 3000);
 
-	if (intel_dsi->dev.dev_ops->send_otp_cmds)
-		intel_dsi->dev.dev_ops->send_otp_cmds(&intel_dsi->dev);
+	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY);
+	usleep_range(2500, 3000);
 }
 
 static void intel_dsi_enable(struct intel_encoder *encoder)
@@ -153,18 +146,78 @@ static void intel_dsi_enable(struct intel_encoder *encoder)
 		I915_WRITE(MIPI_MAX_RETURN_PKT_SIZE(pipe), 8 * 4);
 	else {
 		msleep(20); /* XXX */
-		dpi_send_cmd(intel_dsi, TURN_ON);
+		dpi_send_cmd(intel_dsi, TURN_ON, DPI_LP_MODE_EN);
 		msleep(100);
 
+		if (intel_dsi->dev.dev_ops->enable)
+			intel_dsi->dev.dev_ops->enable(&intel_dsi->dev);
+
 		/* assert ip_tg_enable signal */
 		temp = I915_READ(MIPI_PORT_CTRL(pipe)) & ~LANE_CONFIGURATION_MASK;
 		temp = temp | intel_dsi->port_bits;
 		I915_WRITE(MIPI_PORT_CTRL(pipe), temp | DPI_ENABLE);
 		POSTING_READ(MIPI_PORT_CTRL(pipe));
 	}
+}
+
+static void intel_dsi_pre_enable(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 tmp;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* Disable DPOunit clock gating, can stall pipe
+	 * and we need DPLL REFA always enabled */
+	tmp = I915_READ(DPLL(pipe));
+	tmp |= DPLL_REFA_CLK_ENABLE_VLV;
+	I915_WRITE(DPLL(pipe), tmp);
+
+	tmp = I915_READ(DSPCLK_GATE_D);
+	tmp |= DPOUNIT_CLOCK_GATE_DISABLE;
+	I915_WRITE(DSPCLK_GATE_D, tmp);
+
+	/* put device in ready state */
+	intel_dsi_device_ready(encoder);
+
+	msleep(intel_dsi->panel_on_delay);
+
+	if (intel_dsi->dev.dev_ops->panel_reset)
+		intel_dsi->dev.dev_ops->panel_reset(&intel_dsi->dev);
+
+	if (intel_dsi->dev.dev_ops->send_otp_cmds)
+		intel_dsi->dev.dev_ops->send_otp_cmds(&intel_dsi->dev);
+
+	/* Enable port in pre-enable phase itself because as per hw team
+	 * recommendation, port should be enabled befor plane & pipe */
+	intel_dsi_enable(encoder);
+}
+
+static void intel_dsi_enable_nop(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+
+	/* for DSI port enable has to be done before pipe
+	 * and plane enable, so port enable is done in
+	 * pre_enable phase itself unlike other encoders
+	 */
+}
+
+static void intel_dsi_pre_disable(struct intel_encoder *encoder)
+{
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+
+	DRM_DEBUG_KMS("\n");
 
-	if (intel_dsi->dev.dev_ops->enable)
-		intel_dsi->dev.dev_ops->enable(&intel_dsi->dev);
+	if (is_vid_mode(intel_dsi)) {
+		/* Send Shutdown command to the panel in LP mode */
+		dpi_send_cmd(intel_dsi, SHUTDOWN, DPI_LP_MODE_EN);
+		msleep(10);
+	}
 }
 
 static void intel_dsi_disable(struct intel_encoder *encoder)
@@ -179,9 +232,6 @@ static void intel_dsi_disable(struct intel_encoder *encoder)
 	DRM_DEBUG_KMS("\n");
 
 	if (is_vid_mode(intel_dsi)) {
-		dpi_send_cmd(intel_dsi, SHUTDOWN);
-		msleep(10);
-
 		/* de-assert ip_tg_enable signal */
 		temp = I915_READ(MIPI_PORT_CTRL(pipe));
 		I915_WRITE(MIPI_PORT_CTRL(pipe), temp & ~DPI_ENABLE);
@@ -190,6 +240,23 @@ static void intel_dsi_disable(struct intel_encoder *encoder)
 		msleep(2);
 	}
 
+	/* Panel commands can be sent when clock is in LP11 */
+	I915_WRITE(MIPI_DEVICE_READY(pipe), 0x0);
+
+	temp = I915_READ(MIPI_CTRL(pipe));
+	temp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
+	I915_WRITE(MIPI_CTRL(pipe), temp |
+			intel_dsi->escape_clk_div <<
+			ESCAPE_CLOCK_DIVIDER_SHIFT);
+
+	I915_WRITE(MIPI_EOT_DISABLE(pipe), CLOCKSTOP);
+
+	temp = I915_READ(MIPI_DSI_FUNC_PRG(pipe));
+	temp &= ~VID_MODE_FORMAT_MASK;
+	I915_WRITE(MIPI_DSI_FUNC_PRG(pipe), temp);
+
+	I915_WRITE(MIPI_DEVICE_READY(pipe), 0x1);
+
 	/* if disable packets are sent before sending shutdown packet then in
 	 * some next enable sequence send turn on packet error is observed */
 	if (intel_dsi->dev.dev_ops->disable)
@@ -205,38 +272,50 @@ static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
-	I915_WRITE(MIPI_DEVICE_READY(pipe), ULPS_STATE_ENTER);
+	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY | ULPS_STATE_ENTER);
 	usleep_range(2000, 2500);
 
-	I915_WRITE(MIPI_DEVICE_READY(pipe), ULPS_STATE_EXIT);
+	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY | ULPS_STATE_EXIT);
 	usleep_range(2000, 2500);
 
-	I915_WRITE(MIPI_DEVICE_READY(pipe), ULPS_STATE_ENTER);
+	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY | ULPS_STATE_ENTER);
 	usleep_range(2000, 2500);
 
-	val = I915_READ(MIPI_PORT_CTRL(pipe));
-	I915_WRITE(MIPI_PORT_CTRL(pipe), val & ~LP_OUTPUT_HOLD);
-	usleep_range(1000, 1500);
-
 	if (wait_for(((I915_READ(MIPI_PORT_CTRL(pipe)) & AFE_LATCHOUT)
 					== 0x00000), 30))
 		DRM_ERROR("DSI LP not going Low\n");
 
+	val = I915_READ(MIPI_PORT_CTRL(pipe));
+	I915_WRITE(MIPI_PORT_CTRL(pipe), val & ~LP_OUTPUT_HOLD);
+	usleep_range(1000, 1500);
+
 	I915_WRITE(MIPI_DEVICE_READY(pipe), 0x00);
 	usleep_range(2000, 2500);
 
 	vlv_disable_dsi_pll(encoder);
 }
+
 static void intel_dsi_post_disable(struct intel_encoder *encoder)
 {
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	u32 val;
 
 	DRM_DEBUG_KMS("\n");
 
+	intel_dsi_disable(encoder);
+
 	intel_dsi_clear_device_ready(encoder);
 
+	val = I915_READ(DSPCLK_GATE_D);
+	val &= ~DPOUNIT_CLOCK_GATE_DISABLE;
+	I915_WRITE(DSPCLK_GATE_D, val);
+
 	if (intel_dsi->dev.dev_ops->disable_panel_power)
 		intel_dsi->dev.dev_ops->disable_panel_power(&intel_dsi->dev);
+
+	msleep(intel_dsi->panel_off_delay);
+	msleep(intel_dsi->panel_pwr_cycle_delay);
 }
 
 static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
@@ -364,7 +443,7 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 	I915_WRITE(MIPI_VBP_COUNT(pipe), vbp);
 }
 
-static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
+static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	struct drm_device *dev = encoder->dev;
@@ -379,9 +458,6 @@ static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
 
 	DRM_DEBUG_KMS("pipe %c\n", pipe_name(pipe));
 
-	/* XXX: Location of the call */
-	band_gap_reset(dev_priv);
-
 	/* escape clock divider, 20MHz, shared for A and C. device ready must be
 	 * off when doing this! txclkesc? */
 	tmp = I915_READ(MIPI_CTRL(0));
@@ -452,10 +528,20 @@ static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
 	/* dphy stuff */
 
 	/* in terms of low power clock */
-	I915_WRITE(MIPI_INIT_COUNT(pipe), txclkesc(ESCAPE_CLOCK_DIVIDER_1, 100));
+	I915_WRITE(MIPI_INIT_COUNT(pipe), txclkesc(intel_dsi->escape_clk_div, 100));
+
+	val = 0;
+	if (intel_dsi->eotp_pkt == 0)
+		val |= EOT_DISABLE;
+
+	if (intel_dsi->clock_stop)
+		val |= CLOCKSTOP;
 
 	/* recovery disables */
-	I915_WRITE(MIPI_EOT_DISABLE(pipe), intel_dsi->eot_disable);
+	I915_WRITE(MIPI_EOT_DISABLE(pipe), val);
+
+	/* in terms of low power clock */
+	I915_WRITE(MIPI_INIT_COUNT(pipe), intel_dsi->init_count);
 
 	/* in terms of txbyteclkhs. actual high to low switch +
 	 * MIPI_STOP_STATE_STALL * MIPI_LP_BYTECLK.
@@ -484,9 +570,23 @@ static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
 		   intel_dsi->clk_hs_to_lp_count << HS_LP_PWR_SW_CNT_SHIFT);
 
 	if (is_vid_mode(intel_dsi))
+		/* Some panels might have resolution which is not a multiple of
+		 * 64 like 1366 x 768. Enable RANDOM resolution support for such
+		 * panels by default */
 		I915_WRITE(MIPI_VIDEO_MODE_FORMAT(pipe),
 				intel_dsi->video_frmt_cfg_bits |
-				intel_dsi->video_mode_format);
+				intel_dsi->video_mode_format |
+				IP_TG_CONFIG |
+				RANDOM_DPI_DISPLAY_RESOLUTION);
+}
+
+static void intel_dsi_pre_pll_enable(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+
+	intel_dsi_prepare(encoder);
+
+	vlv_enable_dsi_pll(encoder);
 }
 
 static enum drm_connector_status
@@ -566,11 +666,16 @@ bool intel_dsi_init(struct drm_device *dev)
 	struct intel_connector *intel_connector;
 	struct drm_connector *connector;
 	struct drm_display_mode *fixed_mode = NULL;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	const struct intel_dsi_device *dsi;
 	unsigned int i;
 
 	DRM_DEBUG_KMS("\n");
 
+	/* There is no detection method for MIPI so rely on VBT */
+	if (!dev_priv->vbt.has_mipi)
+		return false;
+
 	intel_dsi = kzalloc(sizeof(*intel_dsi), GFP_KERNEL);
 	if (!intel_dsi)
 		return false;
@@ -585,6 +690,13 @@ bool intel_dsi_init(struct drm_device *dev)
 	encoder = &intel_encoder->base;
 	intel_dsi->attached_connector = intel_connector;
 
+	if (IS_VALLEYVIEW(dev)) {
+		dev_priv->mipi_mmio_base = VLV_MIPI_BASE;
+	} else {
+		DRM_ERROR("Unsupported Mipi device to reg base");
+		return false;
+	}
+
 	connector = &intel_connector->base;
 
 	drm_encoder_init(dev, encoder, &intel_dsi_funcs, DRM_MODE_ENCODER_DSI);
@@ -594,9 +706,8 @@ bool intel_dsi_init(struct drm_device *dev)
 	intel_encoder->compute_config = intel_dsi_compute_config;
 	intel_encoder->pre_pll_enable = intel_dsi_pre_pll_enable;
 	intel_encoder->pre_enable = intel_dsi_pre_enable;
-	intel_encoder->enable = intel_dsi_enable;
-	intel_encoder->mode_set = intel_dsi_mode_set;
-	intel_encoder->disable = intel_dsi_disable;
+	intel_encoder->enable = intel_dsi_enable_nop;
+	intel_encoder->disable = intel_dsi_pre_disable;
 	intel_encoder->post_disable = intel_dsi_post_disable;
 	intel_encoder->get_hw_state = intel_dsi_get_hw_state;
 	intel_encoder->get_config = intel_dsi_get_config;
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index b4a27cec882f..31db33d3e5cc 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -31,7 +31,6 @@
 struct intel_dsi_device {
 	unsigned int panel_id;
 	const char *name;
-	int type;
 	const struct intel_dsi_dev_ops *dev_ops;
 	void *dev_priv;
 };
@@ -85,6 +84,9 @@ struct intel_dsi {
 	/* virtual channel */
 	int channel;
 
+	/* Video mode or command mode */
+	u16 operation_mode;
+
 	/* number of DSI lanes */
 	unsigned int lane_count;
 
@@ -95,8 +97,10 @@ struct intel_dsi {
 	u32 video_mode_format;
 
 	/* eot for MIPI_EOT_DISABLE register */
-	u32 eot_disable;
+	u8 eotp_pkt;
+	u8 clock_stop;
 
+	u8 escape_clk_div;
 	u32 port_bits;
 	u32 bw_timer;
 	u32 dphy_reg;
@@ -110,6 +114,15 @@ struct intel_dsi {
 	u16 hs_to_lp_count;
 	u16 clk_lp_to_hs_count;
 	u16 clk_hs_to_lp_count;
+
+	u16 init_count;
+
+	/* all delays in ms */
+	u16 backlight_off_delay;
+	u16 backlight_on_delay;
+	u16 panel_on_delay;
+	u16 panel_off_delay;
+	u16 panel_pwr_cycle_delay;
 };
 
 static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
@@ -120,4 +133,6 @@ static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 extern void vlv_enable_dsi_pll(struct intel_encoder *encoder);
 extern void vlv_disable_dsi_pll(struct intel_encoder *encoder);
 
+extern struct intel_dsi_dev_ops vbt_generic_dsi_display_ops;
+
 #endif /* _INTEL_DSI_H */
diff --git a/drivers/gpu/drm/i915/intel_dsi_cmd.c b/drivers/gpu/drm/i915/intel_dsi_cmd.c
index 7c40f981d2c7..933c86305237 100644
--- a/drivers/gpu/drm/i915/intel_dsi_cmd.c
+++ b/drivers/gpu/drm/i915/intel_dsi_cmd.c
@@ -389,7 +389,7 @@ int dsi_vc_generic_read(struct intel_dsi *intel_dsi, int channel,
  *
  * XXX: commands with data in MIPI_DPI_DATA?
  */
-int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd)
+int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs)
 {
 	struct drm_encoder *encoder = &intel_dsi->base.base;
 	struct drm_device *dev = encoder->dev;
@@ -399,17 +399,11 @@ int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd)
 	u32 mask;
 
 	/* XXX: pipe, hs */
-	if (intel_dsi->hs)
+	if (hs)
 		cmd &= ~DPI_LP_MODE;
 	else
 		cmd |= DPI_LP_MODE;
 
-	/* DPI virtual channel?! */
-
-	mask = DPI_FIFO_EMPTY;
-	if (wait_for((I915_READ(MIPI_GEN_FIFO_STAT(pipe)) & mask) == mask, 50))
-		DRM_ERROR("Timeout waiting for DPI FIFO empty.\n");
-
 	/* clear bit */
 	I915_WRITE(MIPI_INTR_STAT(pipe), SPL_PKT_SENT_INTERRUPT);
 
diff --git a/drivers/gpu/drm/i915/intel_dsi_cmd.h b/drivers/gpu/drm/i915/intel_dsi_cmd.h
index 54c8a234a2e0..9a18cbfa5460 100644
--- a/drivers/gpu/drm/i915/intel_dsi_cmd.h
+++ b/drivers/gpu/drm/i915/intel_dsi_cmd.h
@@ -33,6 +33,9 @@
 #include "intel_drv.h"
 #include "intel_dsi.h"
 
+#define DPI_LP_MODE_EN	false
+#define DPI_HS_MODE_EN	true
+
 void dsi_hs_mode_enable(struct intel_dsi *intel_dsi, bool enable);
 
 int dsi_vc_dcs_write(struct intel_dsi *intel_dsi, int channel,
@@ -47,7 +50,7 @@ int dsi_vc_dcs_read(struct intel_dsi *intel_dsi, int channel, u8 dcs_cmd,
 int dsi_vc_generic_read(struct intel_dsi *intel_dsi, int channel,
 			u8 *reqdata, int reqlen, u8 *buf, int buflen);
 
-int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd);
+int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs);
 
 /* XXX: questionable write helpers */
 static inline int dsi_vc_dcs_write_0(struct intel_dsi *intel_dsi,
diff --git a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
new file mode 100644
index 000000000000..21a0d348cedc
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
@@ -0,0 +1,589 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Author: Shobhit Kumar <shobhit.kumar@intel.com>
+ *
+ */
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_edid.h>
+#include <drm/i915_drm.h>
+#include <linux/slab.h>
+#include <video/mipi_display.h>
+#include <asm/intel-mid.h>
+#include <video/mipi_display.h>
+#include "i915_drv.h"
+#include "intel_drv.h"
+#include "intel_dsi.h"
+#include "intel_dsi_cmd.h"
+
+#define MIPI_TRANSFER_MODE_SHIFT	0
+#define MIPI_VIRTUAL_CHANNEL_SHIFT	1
+#define MIPI_PORT_SHIFT			3
+
+#define PREPARE_CNT_MAX		0x3F
+#define EXIT_ZERO_CNT_MAX	0x3F
+#define CLK_ZERO_CNT_MAX	0xFF
+#define TRAIL_CNT_MAX		0x1F
+
+#define NS_KHZ_RATIO 1000000
+
+#define GPI0_NC_0_HV_DDI0_HPD           0x4130
+#define GPIO_NC_0_HV_DDI0_PAD           0x4138
+#define GPIO_NC_1_HV_DDI0_DDC_SDA       0x4120
+#define GPIO_NC_1_HV_DDI0_DDC_SDA_PAD   0x4128
+#define GPIO_NC_2_HV_DDI0_DDC_SCL       0x4110
+#define GPIO_NC_2_HV_DDI0_DDC_SCL_PAD   0x4118
+#define GPIO_NC_3_PANEL0_VDDEN          0x4140
+#define GPIO_NC_3_PANEL0_VDDEN_PAD      0x4148
+#define GPIO_NC_4_PANEL0_BLKEN          0x4150
+#define GPIO_NC_4_PANEL0_BLKEN_PAD      0x4158
+#define GPIO_NC_5_PANEL0_BLKCTL         0x4160
+#define GPIO_NC_5_PANEL0_BLKCTL_PAD     0x4168
+#define GPIO_NC_6_PCONF0                0x4180
+#define GPIO_NC_6_PAD                   0x4188
+#define GPIO_NC_7_PCONF0                0x4190
+#define GPIO_NC_7_PAD                   0x4198
+#define GPIO_NC_8_PCONF0                0x4170
+#define GPIO_NC_8_PAD                   0x4178
+#define GPIO_NC_9_PCONF0                0x4100
+#define GPIO_NC_9_PAD                   0x4108
+#define GPIO_NC_10_PCONF0               0x40E0
+#define GPIO_NC_10_PAD                  0x40E8
+#define GPIO_NC_11_PCONF0               0x40F0
+#define GPIO_NC_11_PAD                  0x40F8
+
+struct gpio_table {
+	u16 function_reg;
+	u16 pad_reg;
+	u8 init;
+};
+
+static struct gpio_table gtable[] = {
+	{ GPI0_NC_0_HV_DDI0_HPD, GPIO_NC_0_HV_DDI0_PAD, 0 },
+	{ GPIO_NC_1_HV_DDI0_DDC_SDA, GPIO_NC_1_HV_DDI0_DDC_SDA_PAD, 0 },
+	{ GPIO_NC_2_HV_DDI0_DDC_SCL, GPIO_NC_2_HV_DDI0_DDC_SCL_PAD, 0 },
+	{ GPIO_NC_3_PANEL0_VDDEN, GPIO_NC_3_PANEL0_VDDEN_PAD, 0 },
+	{ GPIO_NC_4_PANEL0_BLKEN, GPIO_NC_4_PANEL0_BLKEN_PAD, 0 },
+	{ GPIO_NC_5_PANEL0_BLKCTL, GPIO_NC_5_PANEL0_BLKCTL_PAD, 0 },
+	{ GPIO_NC_6_PCONF0, GPIO_NC_6_PAD, 0 },
+	{ GPIO_NC_7_PCONF0, GPIO_NC_7_PAD, 0 },
+	{ GPIO_NC_8_PCONF0, GPIO_NC_8_PAD, 0 },
+	{ GPIO_NC_9_PCONF0, GPIO_NC_9_PAD, 0 },
+	{ GPIO_NC_10_PCONF0, GPIO_NC_10_PAD, 0},
+	{ GPIO_NC_11_PCONF0, GPIO_NC_11_PAD, 0}
+};
+
+static u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi, u8 *data)
+{
+	u8 type, byte, mode, vc, port;
+	u16 len;
+
+	byte = *data++;
+	mode = (byte >> MIPI_TRANSFER_MODE_SHIFT) & 0x1;
+	vc = (byte >> MIPI_VIRTUAL_CHANNEL_SHIFT) & 0x3;
+	port = (byte >> MIPI_PORT_SHIFT) & 0x3;
+
+	/* LP or HS mode */
+	intel_dsi->hs = mode;
+
+	/* get packet type and increment the pointer */
+	type = *data++;
+
+	len = *((u16 *) data);
+	data += 2;
+
+	switch (type) {
+	case MIPI_DSI_GENERIC_SHORT_WRITE_0_PARAM:
+		dsi_vc_generic_write_0(intel_dsi, vc);
+		break;
+	case MIPI_DSI_GENERIC_SHORT_WRITE_1_PARAM:
+		dsi_vc_generic_write_1(intel_dsi, vc, *data);
+		break;
+	case MIPI_DSI_GENERIC_SHORT_WRITE_2_PARAM:
+		dsi_vc_generic_write_2(intel_dsi, vc, *data, *(data + 1));
+		break;
+	case MIPI_DSI_GENERIC_READ_REQUEST_0_PARAM:
+	case MIPI_DSI_GENERIC_READ_REQUEST_1_PARAM:
+	case MIPI_DSI_GENERIC_READ_REQUEST_2_PARAM:
+		DRM_DEBUG_DRIVER("Generic Read not yet implemented or used\n");
+		break;
+	case MIPI_DSI_GENERIC_LONG_WRITE:
+		dsi_vc_generic_write(intel_dsi, vc, data, len);
+		break;
+	case MIPI_DSI_DCS_SHORT_WRITE:
+		dsi_vc_dcs_write_0(intel_dsi, vc, *data);
+		break;
+	case MIPI_DSI_DCS_SHORT_WRITE_PARAM:
+		dsi_vc_dcs_write_1(intel_dsi, vc, *data, *(data + 1));
+		break;
+	case MIPI_DSI_DCS_READ:
+		DRM_DEBUG_DRIVER("DCS Read not yet implemented or used\n");
+		break;
+	case MIPI_DSI_DCS_LONG_WRITE:
+		dsi_vc_dcs_write(intel_dsi, vc, data, len);
+		break;
+	};
+
+	data += len;
+
+	return data;
+}
+
+static u8 *mipi_exec_delay(struct intel_dsi *intel_dsi, u8 *data)
+{
+	u32 delay = *((u32 *) data);
+
+	usleep_range(delay, delay + 10);
+	data += 4;
+
+	return data;
+}
+
+static u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, u8 *data)
+{
+	u8 gpio, action;
+	u16 function, pad;
+	u32 val;
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	gpio = *data++;
+
+	/* pull up/down */
+	action = *data++;
+
+	function = gtable[gpio].function_reg;
+	pad = gtable[gpio].pad_reg;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	if (!gtable[gpio].init) {
+		/* program the function */
+		/* FIXME: remove constant below */
+		vlv_gpio_nc_write(dev_priv, function, 0x2000CC00);
+		gtable[gpio].init = 1;
+	}
+
+	val = 0x4 | action;
+
+	/* pull up/down */
+	vlv_gpio_nc_write(dev_priv, pad, val);
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	return data;
+}
+
+typedef u8 * (*fn_mipi_elem_exec)(struct intel_dsi *intel_dsi, u8 *data);
+static const fn_mipi_elem_exec exec_elem[] = {
+	NULL, /* reserved */
+	mipi_exec_send_packet,
+	mipi_exec_delay,
+	mipi_exec_gpio,
+	NULL, /* status read; later */
+};
+
+/*
+ * MIPI Sequence from VBT #53 parsing logic
+ * We have already separated each seqence during bios parsing
+ * Following is generic execution function for any sequence
+ */
+
+static const char * const seq_name[] = {
+	"UNDEFINED",
+	"MIPI_SEQ_ASSERT_RESET",
+	"MIPI_SEQ_INIT_OTP",
+	"MIPI_SEQ_DISPLAY_ON",
+	"MIPI_SEQ_DISPLAY_OFF",
+	"MIPI_SEQ_DEASSERT_RESET"
+};
+
+static void generic_exec_sequence(struct intel_dsi *intel_dsi, char *sequence)
+{
+	u8 *data = sequence;
+	fn_mipi_elem_exec mipi_elem_exec;
+	int index;
+
+	if (!sequence)
+		return;
+
+	DRM_DEBUG_DRIVER("Starting MIPI sequence - %s\n", seq_name[*data]);
+
+	/* go to the first element of the sequence */
+	data++;
+
+	/* parse each byte till we reach end of sequence byte - 0x00 */
+	while (1) {
+		index = *data;
+		mipi_elem_exec = exec_elem[index];
+		if (!mipi_elem_exec) {
+			DRM_ERROR("Unsupported MIPI element, skipping sequence execution\n");
+			return;
+		}
+
+		/* goto element payload */
+		data++;
+
+		/* execute the element specific rotines */
+		data = mipi_elem_exec(intel_dsi, data);
+
+		/*
+		 * After processing the element, data should point to
+		 * next element or end of sequence
+		 * check if have we reached end of sequence
+		 */
+		if (*data == 0x00)
+			break;
+	}
+}
+
+static bool generic_init(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct mipi_config *mipi_config = dev_priv->vbt.dsi.config;
+	struct mipi_pps_data *pps = dev_priv->vbt.dsi.pps;
+	struct drm_display_mode *mode = dev_priv->vbt.lfp_lvds_vbt_mode;
+	u32 bits_per_pixel = 24;
+	u32 tlpx_ns, extra_byte_count, bitrate, tlpx_ui;
+	u32 ui_num, ui_den;
+	u32 prepare_cnt, exit_zero_cnt, clk_zero_cnt, trail_cnt;
+	u32 ths_prepare_ns, tclk_trail_ns;
+	u32 tclk_prepare_clkzero, ths_prepare_hszero;
+	u32 lp_to_hs_switch, hs_to_lp_switch;
+
+	DRM_DEBUG_KMS("\n");
+
+	intel_dsi->eotp_pkt = mipi_config->eot_pkt_disabled ? 0 : 1;
+	intel_dsi->clock_stop = mipi_config->enable_clk_stop ? 1 : 0;
+	intel_dsi->lane_count = mipi_config->lane_cnt + 1;
+	intel_dsi->pixel_format = mipi_config->videomode_color_format << 7;
+
+	if (intel_dsi->pixel_format == VID_MODE_FORMAT_RGB666)
+		bits_per_pixel = 18;
+	else if (intel_dsi->pixel_format == VID_MODE_FORMAT_RGB565)
+		bits_per_pixel = 16;
+
+	bitrate = (mode->clock * bits_per_pixel) / intel_dsi->lane_count;
+
+	intel_dsi->operation_mode = mipi_config->is_cmd_mode;
+	intel_dsi->video_mode_format = mipi_config->video_transfer_mode;
+	intel_dsi->escape_clk_div = mipi_config->byte_clk_sel;
+	intel_dsi->lp_rx_timeout = mipi_config->lp_rx_timeout;
+	intel_dsi->turn_arnd_val = mipi_config->turn_around_timeout;
+	intel_dsi->rst_timer_val = mipi_config->device_reset_timer;
+	intel_dsi->init_count = mipi_config->master_init_timer;
+	intel_dsi->bw_timer = mipi_config->dbi_bw_timer;
+	intel_dsi->video_frmt_cfg_bits = mipi_config->bta_enabled ? DISABLE_VIDEO_BTA : 0;
+
+	switch (intel_dsi->escape_clk_div) {
+	case 0:
+		tlpx_ns = 50;
+		break;
+	case 1:
+		tlpx_ns = 100;
+		break;
+
+	case 2:
+		tlpx_ns = 200;
+		break;
+	default:
+		tlpx_ns = 50;
+		break;
+	}
+
+	switch (intel_dsi->lane_count) {
+	case 1:
+	case 2:
+		extra_byte_count = 2;
+		break;
+	case 3:
+		extra_byte_count = 4;
+		break;
+	case 4:
+	default:
+		extra_byte_count = 3;
+		break;
+	}
+
+	/*
+	 * ui(s) = 1/f [f in hz]
+	 * ui(ns) = 10^9 / (f*10^6) [f in Mhz] -> 10^3/f(Mhz)
+	 */
+
+	/* in Kbps */
+	ui_num = NS_KHZ_RATIO;
+	ui_den = bitrate;
+
+	tclk_prepare_clkzero = mipi_config->tclk_prepare_clkzero;
+	ths_prepare_hszero = mipi_config->ths_prepare_hszero;
+
+	/*
+	 * B060
+	 * LP byte clock = TLPX/ (8UI)
+	 */
+	intel_dsi->lp_byte_clk = DIV_ROUND_UP(tlpx_ns * ui_den, 8 * ui_num);
+
+	/* count values in UI = (ns value) * (bitrate / (2 * 10^6))
+	 *
+	 * Since txddrclkhs_i is 2xUI, all the count values programmed in
+	 * DPHY param register are divided by 2
+	 *
+	 * prepare count
+	 */
+	ths_prepare_ns = max(mipi_config->ths_prepare, mipi_config->tclk_prepare);
+	prepare_cnt = DIV_ROUND_UP(ths_prepare_ns * ui_den, ui_num * 2);
+
+	/* exit zero count */
+	exit_zero_cnt = DIV_ROUND_UP(
+				(ths_prepare_hszero - ths_prepare_ns) * ui_den,
+				ui_num * 2
+				);
+
+	/*
+	 * Exit zero  is unified val ths_zero and ths_exit
+	 * minimum value for ths_exit = 110ns
+	 * min (exit_zero_cnt * 2) = 110/UI
+	 * exit_zero_cnt = 55/UI
+	 */
+	 if (exit_zero_cnt < (55 * ui_den / ui_num))
+		if ((55 * ui_den) % ui_num)
+			exit_zero_cnt += 1;
+
+	/* clk zero count */
+	clk_zero_cnt = DIV_ROUND_UP(
+			(tclk_prepare_clkzero -	ths_prepare_ns)
+			* ui_den, 2 * ui_num);
+
+	/* trail count */
+	tclk_trail_ns = max(mipi_config->tclk_trail, mipi_config->ths_trail);
+	trail_cnt = DIV_ROUND_UP(tclk_trail_ns * ui_den, 2 * ui_num);
+
+	if (prepare_cnt > PREPARE_CNT_MAX ||
+		exit_zero_cnt > EXIT_ZERO_CNT_MAX ||
+		clk_zero_cnt > CLK_ZERO_CNT_MAX ||
+		trail_cnt > TRAIL_CNT_MAX)
+		DRM_DEBUG_DRIVER("Values crossing maximum limits, restricting to max values\n");
+
+	if (prepare_cnt > PREPARE_CNT_MAX)
+		prepare_cnt = PREPARE_CNT_MAX;
+
+	if (exit_zero_cnt > EXIT_ZERO_CNT_MAX)
+		exit_zero_cnt = EXIT_ZERO_CNT_MAX;
+
+	if (clk_zero_cnt > CLK_ZERO_CNT_MAX)
+		clk_zero_cnt = CLK_ZERO_CNT_MAX;
+
+	if (trail_cnt > TRAIL_CNT_MAX)
+		trail_cnt = TRAIL_CNT_MAX;
+
+	/* B080 */
+	intel_dsi->dphy_reg = exit_zero_cnt << 24 | trail_cnt << 16 |
+						clk_zero_cnt << 8 | prepare_cnt;
+
+	/*
+	 * LP to HS switch count = 4TLPX + PREP_COUNT * 2 + EXIT_ZERO_COUNT * 2
+	 *					+ 10UI + Extra Byte Count
+	 *
+	 * HS to LP switch count = THS-TRAIL + 2TLPX + Extra Byte Count
+	 * Extra Byte Count is calculated according to number of lanes.
+	 * High Low Switch Count is the Max of LP to HS and
+	 * HS to LP switch count
+	 *
+	 */
+	tlpx_ui = DIV_ROUND_UP(tlpx_ns * ui_den, ui_num);
+
+	/* B044 */
+	/* FIXME:
+	 * The comment above does not match with the code */
+	lp_to_hs_switch = DIV_ROUND_UP(4 * tlpx_ui + prepare_cnt * 2 +
+						exit_zero_cnt * 2 + 10, 8);
+
+	hs_to_lp_switch = DIV_ROUND_UP(mipi_config->ths_trail + 2 * tlpx_ui, 8);
+
+	intel_dsi->hs_to_lp_count = max(lp_to_hs_switch, hs_to_lp_switch);
+	intel_dsi->hs_to_lp_count += extra_byte_count;
+
+	/* B088 */
+	/* LP -> HS for clock lanes
+	 * LP clk sync + LP11 + LP01 + tclk_prepare + tclk_zero +
+	 *						extra byte count
+	 * 2TPLX + 1TLPX + 1 TPLX(in ns) + prepare_cnt * 2 + clk_zero_cnt *
+	 *					2(in UI) + extra byte count
+	 * In byteclks = (4TLPX + prepare_cnt * 2 + clk_zero_cnt *2 (in UI)) /
+	 *					8 + extra byte count
+	 */
+	intel_dsi->clk_lp_to_hs_count =
+		DIV_ROUND_UP(
+			4 * tlpx_ui + prepare_cnt * 2 +
+			clk_zero_cnt * 2,
+			8);
+
+	intel_dsi->clk_lp_to_hs_count += extra_byte_count;
+
+	/* HS->LP for Clock Lanes
+	 * Low Power clock synchronisations + 1Tx byteclk + tclk_trail +
+	 *						Extra byte count
+	 * 2TLPX + 8UI + (trail_count*2)(in UI) + Extra byte count
+	 * In byteclks = (2*TLpx(in UI) + trail_count*2 +8)(in UI)/8 +
+	 *						Extra byte count
+	 */
+	intel_dsi->clk_hs_to_lp_count =
+		DIV_ROUND_UP(2 * tlpx_ui + trail_cnt * 2 + 8,
+			8);
+	intel_dsi->clk_hs_to_lp_count += extra_byte_count;
+
+	DRM_DEBUG_KMS("Eot %s\n", intel_dsi->eotp_pkt ? "enabled" : "disabled");
+	DRM_DEBUG_KMS("Clockstop %s\n", intel_dsi->clock_stop ?
+						"disabled" : "enabled");
+	DRM_DEBUG_KMS("Mode %s\n", intel_dsi->operation_mode ? "command" : "video");
+	DRM_DEBUG_KMS("Pixel Format %d\n", intel_dsi->pixel_format);
+	DRM_DEBUG_KMS("TLPX %d\n", intel_dsi->escape_clk_div);
+	DRM_DEBUG_KMS("LP RX Timeout 0x%x\n", intel_dsi->lp_rx_timeout);
+	DRM_DEBUG_KMS("Turnaround Timeout 0x%x\n", intel_dsi->turn_arnd_val);
+	DRM_DEBUG_KMS("Init Count 0x%x\n", intel_dsi->init_count);
+	DRM_DEBUG_KMS("HS to LP Count 0x%x\n", intel_dsi->hs_to_lp_count);
+	DRM_DEBUG_KMS("LP Byte Clock %d\n", intel_dsi->lp_byte_clk);
+	DRM_DEBUG_KMS("DBI BW Timer 0x%x\n", intel_dsi->bw_timer);
+	DRM_DEBUG_KMS("LP to HS Clock Count 0x%x\n", intel_dsi->clk_lp_to_hs_count);
+	DRM_DEBUG_KMS("HS to LP Clock Count 0x%x\n", intel_dsi->clk_hs_to_lp_count);
+	DRM_DEBUG_KMS("BTA %s\n",
+			intel_dsi->video_frmt_cfg_bits & DISABLE_VIDEO_BTA ?
+			"disabled" : "enabled");
+
+	/* delays in VBT are in unit of 100us, so need to convert
+	 * here in ms
+	 * Delay (100us) * 100 /1000 = Delay / 10 (ms) */
+	intel_dsi->backlight_off_delay = pps->bl_disable_delay / 10;
+	intel_dsi->backlight_on_delay = pps->bl_enable_delay / 10;
+	intel_dsi->panel_on_delay = pps->panel_on_delay / 10;
+	intel_dsi->panel_off_delay = pps->panel_off_delay / 10;
+	intel_dsi->panel_pwr_cycle_delay = pps->panel_power_cycle_delay / 10;
+
+	return true;
+}
+
+static int generic_mode_valid(struct intel_dsi_device *dsi,
+		   struct drm_display_mode *mode)
+{
+	return MODE_OK;
+}
+
+static bool generic_mode_fixup(struct intel_dsi_device *dsi,
+		    const struct drm_display_mode *mode,
+		    struct drm_display_mode *adjusted_mode) {
+	return true;
+}
+
+static void generic_panel_reset(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_ASSERT_RESET];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_disable_panel_power(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_DEASSERT_RESET];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_send_otp_cmds(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_INIT_OTP];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_enable(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_DISPLAY_ON];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_disable(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_DISPLAY_OFF];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static enum drm_connector_status generic_detect(struct intel_dsi_device *dsi)
+{
+	return connector_status_connected;
+}
+
+static bool generic_get_hw_state(struct intel_dsi_device *dev)
+{
+	return true;
+}
+
+static struct drm_display_mode *generic_get_modes(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->vbt.lfp_lvds_vbt_mode->type |= DRM_MODE_TYPE_PREFERRED;
+	return dev_priv->vbt.lfp_lvds_vbt_mode;
+}
+
+static void generic_destroy(struct intel_dsi_device *dsi) { }
+
+/* Callbacks. We might not need them all. */
+struct intel_dsi_dev_ops vbt_generic_dsi_display_ops = {
+	.init = generic_init,
+	.mode_valid = generic_mode_valid,
+	.mode_fixup = generic_mode_fixup,
+	.panel_reset = generic_panel_reset,
+	.disable_panel_power = generic_disable_panel_power,
+	.send_otp_cmds = generic_send_otp_cmds,
+	.enable = generic_enable,
+	.disable = generic_disable,
+	.detect = generic_detect,
+	.get_hw_state = generic_get_hw_state,
+	.get_modes = generic_get_modes,
+	.destroy = generic_destroy,
+};
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 7fe3feedfe03..a3631c0a5c28 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -285,7 +285,7 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_dvo_mode_set(struct intel_encoder *encoder)
+static void intel_dvo_pre_enable(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -343,7 +343,7 @@ intel_dvo_detect(struct drm_connector *connector, bool force)
 {
 	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 	return intel_dvo->dev.dev_ops->detect(&intel_dvo->dev);
 }
 
@@ -475,7 +475,7 @@ void intel_dvo_init(struct drm_device *dev)
 	intel_encoder->get_hw_state = intel_dvo_get_hw_state;
 	intel_encoder->get_config = intel_dvo_get_config;
 	intel_encoder->compute_config = intel_dvo_compute_config;
-	intel_encoder->mode_set = intel_dvo_mode_set;
+	intel_encoder->pre_enable = intel_dvo_pre_enable;
 	intel_connector->get_hw_state = intel_dvo_connector_get_hw_state;
 	intel_connector->unregister = intel_connector_unregister;
 
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index f73ba5e6b7a8..088fe9378a4c 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -343,15 +343,15 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			num_connectors_detected++;
 
 		if (!enabled[i]) {
-			DRM_DEBUG_KMS("connector %d not enabled, skipping\n",
-				      connector->base.id);
+			DRM_DEBUG_KMS("connector %s not enabled, skipping\n",
+				      connector->name);
 			continue;
 		}
 
 		encoder = connector->encoder;
 		if (!encoder || WARN_ON(!encoder->crtc)) {
-			DRM_DEBUG_KMS("connector %d has no encoder or crtc, skipping\n",
-				      connector->base.id);
+			DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n",
+				      connector->name);
 			enabled[i] = false;
 			continue;
 		}
@@ -373,16 +373,16 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			}
 		}
 
-		DRM_DEBUG_KMS("looking for cmdline mode on connector %d\n",
-			      fb_conn->connector->base.id);
+		DRM_DEBUG_KMS("looking for cmdline mode on connector %s\n",
+			      connector->name);
 
 		/* go for command line mode first */
 		modes[i] = drm_pick_cmdline_mode(fb_conn, width, height);
 
 		/* try for preferred next */
 		if (!modes[i]) {
-			DRM_DEBUG_KMS("looking for preferred mode on connector %d\n",
-				      fb_conn->connector->base.id);
+			DRM_DEBUG_KMS("looking for preferred mode on connector %s\n",
+				      connector->name);
 			modes[i] = drm_has_preferred_mode(fb_conn, width,
 							  height);
 		}
@@ -390,7 +390,7 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 		/* No preferred mode marked by the EDID? Are there any modes? */
 		if (!modes[i] && !list_empty(&connector->modes)) {
 			DRM_DEBUG_KMS("using first mode listed on connector %s\n",
-				      drm_get_connector_name(connector));
+				      connector->name);
 			modes[i] = list_first_entry(&connector->modes,
 						    struct drm_display_mode,
 						    head);
@@ -409,16 +409,20 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			 * since the fb helper layer wants a pointer to
 			 * something we own.
 			 */
+			DRM_DEBUG_KMS("looking for current mode on connector %s\n",
+				      connector->name);
 			intel_mode_from_pipe_config(&encoder->crtc->hwmode,
 						    &to_intel_crtc(encoder->crtc)->config);
 			modes[i] = &encoder->crtc->hwmode;
 		}
 		crtcs[i] = new_crtc;
 
-		DRM_DEBUG_KMS("connector %s on crtc %d: %s\n",
-			      drm_get_connector_name(connector),
+		DRM_DEBUG_KMS("connector %s on pipe %d [CRTC:%d]: %dx%d%s\n",
+			      connector->name,
+			      pipe_name(to_intel_crtc(encoder->crtc)->pipe),
 			      encoder->crtc->base.id,
-			      modes[i]->name);
+			      modes[i]->hdisplay, modes[i]->vdisplay,
+			      modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" :"");
 
 		fallback = false;
 	}
@@ -497,7 +501,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		return false;
 
 	/* Find the largest fb */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		intel_crtc = to_intel_crtc(crtc);
 
 		if (!intel_crtc->active || !crtc->primary->fb) {
@@ -521,7 +525,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	}
 
 	/* Now make sure all the pipes will fit into it */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		unsigned int cur_size;
 
 		intel_crtc = to_intel_crtc(crtc);
@@ -586,7 +590,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	drm_framebuffer_reference(&ifbdev->fb->base);
 
 	/* Final pass to check if any active pipes don't have fbs */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		intel_crtc = to_intel_crtc(crtc);
 
 		if (!intel_crtc->active)
@@ -692,11 +696,7 @@ void intel_fbdev_restore_mode(struct drm_device *dev)
 	if (!dev_priv->fbdev)
 		return;
 
-	drm_modeset_lock_all(dev);
-
-	ret = drm_fb_helper_restore_fbdev_mode(&dev_priv->fbdev->helper);
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(&dev_priv->fbdev->helper);
 	if (ret)
 		DRM_DEBUG("failed to restore crtc mode\n");
-
-	drm_modeset_unlock_all(dev);
 }
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 157267aa3561..eee2bbec2958 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -418,6 +418,7 @@ intel_hdmi_set_hdmi_infoframe(struct drm_encoder *encoder,
 }
 
 static void g4x_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -440,7 +441,7 @@ static void g4x_set_infoframes(struct drm_encoder *encoder,
 	 * either. */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~VIDEO_DIP_ENABLE;
@@ -471,6 +472,7 @@ static void g4x_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void ibx_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -486,7 +488,7 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 	/* See the big comment in g4x_set_infoframes() */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~VIDEO_DIP_ENABLE;
@@ -518,6 +520,7 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void cpt_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -531,7 +534,7 @@ static void cpt_set_infoframes(struct drm_encoder *encoder,
 	/* See the big comment in g4x_set_infoframes() */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~(VIDEO_DIP_ENABLE | VIDEO_DIP_ENABLE_AVI);
@@ -554,20 +557,23 @@ static void cpt_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void vlv_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	u32 reg = VLV_TVIDEO_DIP_CTL(intel_crtc->pipe);
 	u32 val = I915_READ(reg);
+	u32 port = VIDEO_DIP_PORT(intel_dig_port->port);
 
 	assert_hdmi_port_disabled(intel_hdmi);
 
 	/* See the big comment in g4x_set_infoframes() */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~VIDEO_DIP_ENABLE;
@@ -576,9 +582,19 @@ static void vlv_set_infoframes(struct drm_encoder *encoder,
 		return;
 	}
 
+	if (port != (val & VIDEO_DIP_PORT_MASK)) {
+		if (val & VIDEO_DIP_ENABLE) {
+			val &= ~VIDEO_DIP_ENABLE;
+			I915_WRITE(reg, val);
+			POSTING_READ(reg);
+		}
+		val &= ~VIDEO_DIP_PORT_MASK;
+		val |= port;
+	}
+
 	val |= VIDEO_DIP_ENABLE;
-	val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
-		 VIDEO_DIP_ENABLE_GCP);
+	val &= ~(VIDEO_DIP_ENABLE_AVI | VIDEO_DIP_ENABLE_VENDOR |
+		 VIDEO_DIP_ENABLE_GAMUT | VIDEO_DIP_ENABLE_GCP);
 
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
@@ -589,6 +605,7 @@ static void vlv_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void hsw_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -599,7 +616,7 @@ static void hsw_set_infoframes(struct drm_encoder *encoder,
 
 	assert_hdmi_port_disabled(intel_hdmi);
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		I915_WRITE(reg, 0);
 		POSTING_READ(reg);
 		return;
@@ -616,7 +633,7 @@ static void hsw_set_infoframes(struct drm_encoder *encoder,
 	intel_hdmi_set_hdmi_infoframe(encoder, adjusted_mode);
 }
 
-static void intel_hdmi_mode_set(struct intel_encoder *encoder)
+static void intel_hdmi_prepare(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -638,27 +655,26 @@ static void intel_hdmi_mode_set(struct intel_encoder *encoder)
 	else
 		hdmi_val |= SDVO_COLOR_FORMAT_8bpc;
 
-	/* Required on CPT */
-	if (intel_hdmi->has_hdmi_sink && HAS_PCH_CPT(dev))
+	if (crtc->config.has_hdmi_sink)
 		hdmi_val |= HDMI_MODE_SELECT_HDMI;
 
-	if (intel_hdmi->has_audio) {
+	if (crtc->config.has_audio) {
+		WARN_ON(!crtc->config.has_hdmi_sink);
 		DRM_DEBUG_DRIVER("Enabling HDMI audio on pipe %c\n",
 				 pipe_name(crtc->pipe));
 		hdmi_val |= SDVO_AUDIO_ENABLE;
-		hdmi_val |= HDMI_MODE_SELECT_HDMI;
 		intel_write_eld(&encoder->base, adjusted_mode);
 	}
 
 	if (HAS_PCH_CPT(dev))
 		hdmi_val |= SDVO_PIPE_SEL_CPT(crtc->pipe);
+	else if (IS_CHERRYVIEW(dev))
+		hdmi_val |= SDVO_PIPE_SEL_CHV(crtc->pipe);
 	else
 		hdmi_val |= SDVO_PIPE_SEL(crtc->pipe);
 
 	I915_WRITE(intel_hdmi->hdmi_reg, hdmi_val);
 	POSTING_READ(intel_hdmi->hdmi_reg);
-
-	intel_hdmi->set_infoframes(&encoder->base, adjusted_mode);
 }
 
 static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
@@ -681,6 +697,8 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
 
 	if (HAS_PCH_CPT(dev))
 		*pipe = PORT_TO_PIPE_CPT(tmp);
+	else if (IS_CHERRYVIEW(dev))
+		*pipe = SDVO_PORT_TO_PIPE_CHV(tmp);
 	else
 		*pipe = PORT_TO_PIPE(tmp);
 
@@ -707,6 +725,12 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	else
 		flags |= DRM_MODE_FLAG_NVSYNC;
 
+	if (tmp & HDMI_MODE_SELECT_HDMI)
+		pipe_config->has_hdmi_sink = true;
+
+	if (tmp & HDMI_MODE_SELECT_HDMI)
+		pipe_config->has_audio = true;
+
 	pipe_config->adjusted_mode.flags |= flags;
 
 	if ((tmp & SDVO_COLOR_FORMAT_MASK) == HDMI_COLOR_FORMAT_12bpc)
@@ -729,7 +753,7 @@ static void intel_enable_hdmi(struct intel_encoder *encoder)
 	u32 temp;
 	u32 enable_bits = SDVO_ENABLE;
 
-	if (intel_hdmi->has_audio)
+	if (intel_crtc->config.has_audio)
 		enable_bits |= SDVO_AUDIO_ENABLE;
 
 	temp = I915_READ(intel_hdmi->hdmi_reg);
@@ -883,9 +907,11 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 	int portclock_limit = hdmi_portclock_limit(intel_hdmi, false);
 	int desired_bpp;
 
+	pipe_config->has_hdmi_sink = intel_hdmi->has_hdmi_sink;
+
 	if (intel_hdmi->color_range_auto) {
 		/* See CEA-861-E - 5.1 Default Encoding Parameters */
-		if (intel_hdmi->has_hdmi_sink &&
+		if (pipe_config->has_hdmi_sink &&
 		    drm_match_cea_mode(adjusted_mode) > 1)
 			intel_hdmi->color_range = HDMI_COLOR_RANGE_16_235;
 		else
@@ -898,13 +924,16 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 	if (HAS_PCH_SPLIT(dev) && !HAS_DDI(dev))
 		pipe_config->has_pch_encoder = true;
 
+	if (pipe_config->has_hdmi_sink && intel_hdmi->has_audio)
+		pipe_config->has_audio = true;
+
 	/*
 	 * HDMI is either 12 or 8, so if the display lets 10bpc sneak
 	 * through, clamp it down. Note that g4x/vlv don't support 12bpc hdmi
 	 * outputs. We also need to check that the higher clock still fits
 	 * within limits.
 	 */
-	if (pipe_config->pipe_bpp > 8*3 && intel_hdmi->has_hdmi_sink &&
+	if (pipe_config->pipe_bpp > 8*3 && pipe_config->has_hdmi_sink &&
 	    clock_12bpc <= portclock_limit &&
 	    hdmi_12bpc_possible(encoder->new_crtc)) {
 		DRM_DEBUG_KMS("picking bpc to 12 for HDMI output\n");
@@ -944,7 +973,7 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status status = connector_status_disconnected;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	power_domain = intel_display_port_power_domain(intel_encoder);
 	intel_display_power_get(dev_priv, power_domain);
@@ -1104,20 +1133,34 @@ done:
 	return 0;
 }
 
+static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
+{
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *adjusted_mode =
+		&intel_crtc->config.adjusted_mode;
+
+	intel_hdmi_prepare(encoder);
+
+	intel_hdmi->set_infoframes(&encoder->base,
+				   intel_crtc->config.has_hdmi_sink,
+				   adjusted_mode);
+}
+
 static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
+	struct intel_hdmi *intel_hdmi = &dport->hdmi;
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *adjusted_mode =
+		&intel_crtc->config.adjusted_mode;
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 	u32 val;
 
-	if (!IS_VALLEYVIEW(dev))
-		return;
-
 	/* Enable clock channels for this port */
 	mutex_lock(&dev_priv->dpio_lock);
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW8(port));
@@ -1144,6 +1187,10 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW23(port), 0x00400888);
 	mutex_unlock(&dev_priv->dpio_lock);
 
+	intel_hdmi->set_infoframes(&encoder->base,
+				   intel_crtc->config.has_hdmi_sink,
+				   adjusted_mode);
+
 	intel_enable_hdmi(encoder);
 
 	vlv_wait_port_ready(dev_priv, dport);
@@ -1159,8 +1206,7 @@ static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 
-	if (!IS_VALLEYVIEW(dev))
-		return;
+	intel_hdmi_prepare(encoder);
 
 	/* Program Tx lane resets to default */
 	mutex_lock(&dev_priv->dpio_lock);
@@ -1199,6 +1245,152 @@ static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
+static void chv_hdmi_post_disable(struct intel_encoder *encoder)
+{
+	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 val;
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Propagate soft reset to data lane reset */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+}
+
+static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
+{
+	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
+	int data, i;
+	u32 val;
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Deassert soft data lane reset*/
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	/* Program Tx latency optimal setting */
+	for (i = 0; i < 4; i++) {
+		/* Set the latency optimal bit */
+		data = (i == 1) ? 0x0 : 0x6;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW11(ch, i),
+				data << DPIO_FRC_LATENCY_SHFIT);
+
+		/* Set the upar bit */
+		data = (i == 1) ? 0x0 : 0x1;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW14(ch, i),
+				data << DPIO_UPAR_SHIFT);
+	}
+
+	/* Data lane stagger programming */
+	/* FIXME: Fix up value only after power analysis */
+
+	/* Clear calc init */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* FIXME: Program the support xxx V-dB */
+	/* Use 800mV-0dB */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW4(ch, i));
+		val &= ~DPIO_SWING_DEEMPH9P5_MASK;
+		val |= 128 << DPIO_SWING_DEEMPH9P5_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW4(ch, i), val);
+	}
+
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+		val &= ~DPIO_SWING_MARGIN_MASK;
+		val |= 102 << DPIO_SWING_MARGIN_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
+	}
+
+	/* Disable unique transition scale */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+		val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
+	}
+
+	/* Additional steps for 1200mV-0dB */
+#if 0
+	val = vlv_dpio_read(dev_priv, pipe, VLV_TX_DW3(ch));
+	if (ch)
+		val |= DPIO_TX_UNIQ_TRANS_SCALE_CH1;
+	else
+		val |= DPIO_TX_UNIQ_TRANS_SCALE_CH0;
+	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW3(ch), val);
+
+	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW2(ch),
+			vlv_dpio_read(dev_priv, pipe, VLV_TX_DW2(ch)) |
+				(0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT));
+#endif
+	/* Start swing calculation */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* LRC Bypass */
+	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
+	val |= DPIO_LRC_BYPASS;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	intel_enable_hdmi(encoder);
+
+	vlv_wait_port_ready(dev_priv, dport);
+}
+
 static void intel_hdmi_destroy(struct drm_connector *connector)
 {
 	drm_connector_cleanup(connector);
@@ -1259,7 +1451,10 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 		intel_encoder->hpd_pin = HPD_PORT_C;
 		break;
 	case PORT_D:
-		intel_hdmi->ddc_bus = GMBUS_PORT_DPD;
+		if (IS_CHERRYVIEW(dev))
+			intel_hdmi->ddc_bus = GMBUS_PORT_DPD_CHV;
+		else
+			intel_hdmi->ddc_bus = GMBUS_PORT_DPD;
 		intel_encoder->hpd_pin = HPD_PORT_D;
 		break;
 	case PORT_A:
@@ -1329,21 +1524,32 @@ void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_hdmi_compute_config;
-	intel_encoder->mode_set = intel_hdmi_mode_set;
 	intel_encoder->disable = intel_disable_hdmi;
 	intel_encoder->get_hw_state = intel_hdmi_get_hw_state;
 	intel_encoder->get_config = intel_hdmi_get_config;
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		intel_encoder->pre_enable = chv_hdmi_pre_enable;
+		intel_encoder->enable = vlv_enable_hdmi;
+		intel_encoder->post_disable = chv_hdmi_post_disable;
+	} else if (IS_VALLEYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = vlv_hdmi_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_hdmi_pre_enable;
 		intel_encoder->enable = vlv_enable_hdmi;
 		intel_encoder->post_disable = vlv_hdmi_post_disable;
 	} else {
+		intel_encoder->pre_enable = intel_hdmi_pre_enable;
 		intel_encoder->enable = intel_enable_hdmi;
 	}
 
 	intel_encoder->type = INTEL_OUTPUT_HDMI;
-	intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	if (IS_CHERRYVIEW(dev)) {
+		if (port == PORT_D)
+			intel_encoder->crtc_mask = 1 << 2;
+		else
+			intel_encoder->crtc_mask = (1 << 0) | (1 << 1);
+	} else {
+		intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	}
 	intel_encoder->cloneable = 1 << INTEL_OUTPUT_ANALOG;
 	/*
 	 * BSpec is unclear about HDMI+HDMI cloning on g4x, but it seems
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index f1ecf916474a..23126023aeba 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -111,13 +111,6 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 
 	pipe_config->adjusted_mode.flags |= flags;
 
-	/* gen2/3 store dither state in pfit control, needs to match */
-	if (INTEL_INFO(dev)->gen < 4) {
-		tmp = I915_READ(PFIT_CONTROL);
-
-		pipe_config->gmch_pfit.control |= tmp & PANEL_8TO6_DITHER_ENABLE;
-	}
-
 	dotclock = pipe_config->port_clock;
 
 	if (HAS_PCH_SPLIT(dev_priv->dev))
@@ -126,10 +119,6 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 	pipe_config->adjusted_mode.crtc_clock = dotclock;
 }
 
-/* The LVDS pin pair needs to be on before the DPLLs are enabled.
- * This is an exception to the general rule that mode_set doesn't turn
- * things on.
- */
 static void intel_pre_enable_lvds(struct intel_encoder *encoder)
 {
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
@@ -331,15 +320,6 @@ static bool intel_lvds_compute_config(struct intel_encoder *intel_encoder,
 	return true;
 }
 
-static void intel_lvds_mode_set(struct intel_encoder *encoder)
-{
-	/*
-	 * We don't do anything here, the LVDS port is fully set up in the pre
-	 * enable hook - the ordering constraints for enabling the lvds port vs.
-	 * enabling the display pll are too strict.
-	 */
-}
-
 /**
  * Detect the LVDS connection.
  *
@@ -354,7 +334,7 @@ intel_lvds_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status status;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	status = intel_panel_detect(dev);
 	if (status != connector_status_unknown)
@@ -953,7 +933,6 @@ void intel_lvds_init(struct drm_device *dev)
 	intel_encoder->enable = intel_enable_lvds;
 	intel_encoder->pre_enable = intel_pre_enable_lvds;
 	intel_encoder->compute_config = intel_lvds_compute_config;
-	intel_encoder->mode_set = intel_lvds_mode_set;
 	intel_encoder->disable = intel_disable_lvds;
 	intel_encoder->get_hw_state = intel_lvds_get_hw_state;
 	intel_encoder->get_config = intel_lvds_get_config;
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index acde2945eb8a..4f6b53998d79 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -403,6 +403,15 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 
 	DRM_DEBUG_DRIVER("bclp = 0x%08x\n", bclp);
 
+	/*
+	 * If the acpi_video interface is not supposed to be used, don't
+	 * bother processing backlight level change requests from firmware.
+	 */
+	if (!acpi_video_verify_backlight_support()) {
+		DRM_DEBUG_KMS("opregion backlight request ignored\n");
+		return 0;
+	}
+
 	if (!(bclp & ASLE_BCLP_VALID))
 		return ASLC_BACKLIGHT_FAILED;
 
@@ -410,7 +419,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 	if (bclp > 255)
 		return ASLC_BACKLIGHT_FAILED;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 
 	/*
 	 * Update backlight on all connectors that support backlight (usually
@@ -421,7 +430,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 		intel_panel_set_backlight(intel_connector, bclp, 255);
 	iowrite32(DIV_ROUND_UP(bclp * 100, 255) | ASLE_CBLV_VALID, &asle->cblv);
 
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 129db0c7d835..daa118978eec 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -213,7 +213,7 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	BUG_ON(overlay->last_flip_req);
@@ -236,7 +236,7 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	BUG_ON(overlay->active);
@@ -263,7 +263,7 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	u32 flip_addr = overlay->flip_addr;
 	u32 tmp;
 	int ret;
@@ -320,7 +320,7 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	u32 flip_addr = overlay->flip_addr;
 	int ret;
 
@@ -363,7 +363,7 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	if (overlay->last_flip_req == 0)
@@ -389,7 +389,7 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	/* Only wait if there is actually an old frame to release to
@@ -688,7 +688,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
 	u32 swidth, swidthsw, sheight, ostride;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	BUG_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 	BUG_ON(!overlay);
 
 	ret = intel_overlay_release_old_vid(overlay);
@@ -793,7 +793,7 @@ int intel_overlay_switch_off(struct intel_overlay *overlay)
 	int ret;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	BUG_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	ret = intel_overlay_recover_from_interrupt(overlay);
 	if (ret != 0)
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index cb8cfb7e0974..628cd8938274 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -42,6 +42,59 @@ intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
 	drm_mode_set_crtcinfo(adjusted_mode, 0);
 }
 
+/**
+ * intel_find_panel_downclock - find the reduced downclock for LVDS in EDID
+ * @dev: drm device
+ * @fixed_mode : panel native mode
+ * @connector: LVDS/eDP connector
+ *
+ * Return downclock_avail
+ * Find the reduced downclock for LVDS/eDP in EDID.
+ */
+struct drm_display_mode *
+intel_find_panel_downclock(struct drm_device *dev,
+			struct drm_display_mode *fixed_mode,
+			struct drm_connector *connector)
+{
+	struct drm_display_mode *scan, *tmp_mode;
+	int temp_downclock;
+
+	temp_downclock = fixed_mode->clock;
+	tmp_mode = NULL;
+
+	list_for_each_entry(scan, &connector->probed_modes, head) {
+		/*
+		 * If one mode has the same resolution with the fixed_panel
+		 * mode while they have the different refresh rate, it means
+		 * that the reduced downclock is found. In such
+		 * case we can set the different FPx0/1 to dynamically select
+		 * between low and high frequency.
+		 */
+		if (scan->hdisplay == fixed_mode->hdisplay &&
+		    scan->hsync_start == fixed_mode->hsync_start &&
+		    scan->hsync_end == fixed_mode->hsync_end &&
+		    scan->htotal == fixed_mode->htotal &&
+		    scan->vdisplay == fixed_mode->vdisplay &&
+		    scan->vsync_start == fixed_mode->vsync_start &&
+		    scan->vsync_end == fixed_mode->vsync_end &&
+		    scan->vtotal == fixed_mode->vtotal) {
+			if (scan->clock < temp_downclock) {
+				/*
+				 * The downclock is already found. But we
+				 * expect to find the lower downclock.
+				 */
+				temp_downclock = scan->clock;
+				tmp_mode = scan;
+			}
+		}
+	}
+
+	if (temp_downclock < fixed_mode->clock)
+		return drm_mode_duplicate(dev, tmp_mode);
+	else
+		return NULL;
+}
+
 /* adjusted_mode has been preset to be the panel's fixed mode */
 void
 intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
@@ -308,21 +361,43 @@ void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 		pfit_control |= ((intel_crtc->pipe << PFIT_PIPE_SHIFT) |
 				 PFIT_FILTER_FUZZY);
 
+	/* Make sure pre-965 set dither correctly for 18bpp panels. */
+	if (INTEL_INFO(dev)->gen < 4 && pipe_config->pipe_bpp == 18)
+		pfit_control |= PANEL_8TO6_DITHER_ENABLE;
+
 out:
 	if ((pfit_control & PFIT_ENABLE) == 0) {
 		pfit_control = 0;
 		pfit_pgm_ratios = 0;
 	}
 
-	/* Make sure pre-965 set dither correctly for 18bpp panels. */
-	if (INTEL_INFO(dev)->gen < 4 && pipe_config->pipe_bpp == 18)
-		pfit_control |= PANEL_8TO6_DITHER_ENABLE;
-
 	pipe_config->gmch_pfit.control = pfit_control;
 	pipe_config->gmch_pfit.pgm_ratios = pfit_pgm_ratios;
 	pipe_config->gmch_pfit.lvds_border_bits = border;
 }
 
+enum drm_connector_status
+intel_panel_detect(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Assume that the BIOS does not lie through the OpRegion... */
+	if (!i915.panel_ignore_lid && dev_priv->opregion.lid_state) {
+		return ioread32(dev_priv->opregion.lid_state) & 0x1 ?
+			connector_status_connected :
+			connector_status_disconnected;
+	}
+
+	switch (i915.panel_ignore_lid) {
+	case -2:
+		return connector_status_connected;
+	case -1:
+		return connector_status_disconnected;
+	default:
+		return connector_status_unknown;
+	}
+}
+
 static u32 intel_panel_compute_brightness(struct intel_connector *connector,
 					  u32 val)
 {
@@ -723,9 +798,6 @@ static void i965_enable_backlight(struct intel_connector *connector)
 	ctl = freq << 16;
 	I915_WRITE(BLC_PWM_CTL, ctl);
 
-	/* XXX: combine this into above write? */
-	intel_panel_actually_set_backlight(connector, panel->backlight.level);
-
 	ctl2 = BLM_PIPE(pipe);
 	if (panel->backlight.combination_mode)
 		ctl2 |= BLM_COMBINATION_MODE;
@@ -734,6 +806,8 @@ static void i965_enable_backlight(struct intel_connector *connector)
 	I915_WRITE(BLC_PWM_CTL2, ctl2);
 	POSTING_READ(BLC_PWM_CTL2);
 	I915_WRITE(BLC_PWM_CTL2, ctl2 | BLM_PWM_ENABLE);
+
+	intel_panel_actually_set_backlight(connector, panel->backlight.level);
 }
 
 static void vlv_enable_backlight(struct intel_connector *connector)
@@ -795,40 +869,18 @@ void intel_panel_enable_backlight(struct intel_connector *connector)
 	spin_unlock_irqrestore(&dev_priv->backlight_lock, flags);
 }
 
-enum drm_connector_status
-intel_panel_detect(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	/* Assume that the BIOS does not lie through the OpRegion... */
-	if (!i915.panel_ignore_lid && dev_priv->opregion.lid_state) {
-		return ioread32(dev_priv->opregion.lid_state) & 0x1 ?
-			connector_status_connected :
-			connector_status_disconnected;
-	}
-
-	switch (i915.panel_ignore_lid) {
-	case -2:
-		return connector_status_connected;
-	case -1:
-		return connector_status_disconnected;
-	default:
-		return connector_status_unknown;
-	}
-}
-
 #if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE)
 static int intel_backlight_device_update_status(struct backlight_device *bd)
 {
 	struct intel_connector *connector = bl_get_data(bd);
 	struct drm_device *dev = connector->base.dev;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	DRM_DEBUG_KMS("updating intel_backlight, brightness=%d/%d\n",
 		      bd->props.brightness, bd->props.max_brightness);
 	intel_panel_set_backlight(connector, bd->props.brightness,
 				  bd->props.max_brightness);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 	return 0;
 }
 
@@ -840,9 +892,9 @@ static int intel_backlight_device_get_brightness(struct backlight_device *bd)
 	int ret;
 
 	intel_runtime_pm_get(dev_priv);
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	ret = intel_panel_get_backlight(connector);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 	intel_runtime_pm_put(dev_priv);
 
 	return ret;
@@ -1066,8 +1118,12 @@ int intel_panel_setup_backlight(struct drm_connector *connector)
 	int ret;
 
 	if (!dev_priv->vbt.backlight.present) {
-		DRM_DEBUG_KMS("native backlight control not available per VBT\n");
-		return 0;
+		if (dev_priv->quirks & QUIRK_BACKLIGHT_PRESENT) {
+			DRM_DEBUG_KMS("no backlight present per VBT, but present per quirk\n");
+		} else {
+			DRM_DEBUG_KMS("no backlight present per VBT\n");
+			return 0;
+		}
 	}
 
 	/* set level and max in panel struct */
@@ -1077,7 +1133,7 @@ int intel_panel_setup_backlight(struct drm_connector *connector)
 
 	if (ret) {
 		DRM_DEBUG_KMS("failed to setup backlight for connector %s\n",
-			      drm_get_connector_name(connector));
+			      connector->name);
 		return ret;
 	}
 
@@ -1103,59 +1159,6 @@ void intel_panel_destroy_backlight(struct drm_connector *connector)
 	intel_backlight_device_unregister(intel_connector);
 }
 
-/**
- * intel_find_panel_downclock - find the reduced downclock for LVDS in EDID
- * @dev: drm device
- * @fixed_mode : panel native mode
- * @connector: LVDS/eDP connector
- *
- * Return downclock_avail
- * Find the reduced downclock for LVDS/eDP in EDID.
- */
-struct drm_display_mode *
-intel_find_panel_downclock(struct drm_device *dev,
-			struct drm_display_mode *fixed_mode,
-			struct drm_connector *connector)
-{
-	struct drm_display_mode *scan, *tmp_mode;
-	int temp_downclock;
-
-	temp_downclock = fixed_mode->clock;
-	tmp_mode = NULL;
-
-	list_for_each_entry(scan, &connector->probed_modes, head) {
-		/*
-		 * If one mode has the same resolution with the fixed_panel
-		 * mode while they have the different refresh rate, it means
-		 * that the reduced downclock is found. In such
-		 * case we can set the different FPx0/1 to dynamically select
-		 * between low and high frequency.
-		 */
-		if (scan->hdisplay == fixed_mode->hdisplay &&
-		    scan->hsync_start == fixed_mode->hsync_start &&
-		    scan->hsync_end == fixed_mode->hsync_end &&
-		    scan->htotal == fixed_mode->htotal &&
-		    scan->vdisplay == fixed_mode->vdisplay &&
-		    scan->vsync_start == fixed_mode->vsync_start &&
-		    scan->vsync_end == fixed_mode->vsync_end &&
-		    scan->vtotal == fixed_mode->vtotal) {
-			if (scan->clock < temp_downclock) {
-				/*
-				 * The downclock is already found. But we
-				 * expect to find the lower downclock.
-				 */
-				temp_downclock = scan->clock;
-				tmp_mode = scan;
-			}
-		}
-	}
-
-	if (temp_downclock < fixed_mode->clock)
-		return drm_mode_duplicate(dev, tmp_mode);
-	else
-		return NULL;
-}
-
 /* Set up chip specific backlight functions */
 void intel_panel_init_backlight_funcs(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d93dcf683e8c..ee72807069e4 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -487,7 +487,7 @@ void intel_update_fbc(struct drm_device *dev)
 	 *   - new fb is too large to fit in compressed buffer
 	 *   - going to an unsupported config (interlace, pixel multiply, etc.)
 	 */
-	list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, tmp_crtc) {
 		if (intel_crtc_active(tmp_crtc) &&
 		    to_intel_crtc(tmp_crtc)->primary_enabled) {
 			if (crtc) {
@@ -511,8 +511,7 @@ void intel_update_fbc(struct drm_device *dev)
 	obj = intel_fb->obj;
 	adjusted_mode = &intel_crtc->config.adjusted_mode;
 
-	if (i915.enable_fbc < 0 &&
-	    INTEL_INFO(dev)->gen <= 7 && !IS_HASWELL(dev)) {
+	if (i915.enable_fbc < 0) {
 		if (set_no_fbc_reason(dev_priv, FBC_CHIP_DEFAULT))
 			DRM_DEBUG_KMS("disabled per chip default\n");
 		goto out_disable;
@@ -1010,7 +1009,7 @@ static struct drm_crtc *single_enabled_crtc(struct drm_device *dev)
 {
 	struct drm_crtc *crtc, *enabled = NULL;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		if (intel_crtc_active(crtc)) {
 			if (enabled)
 				return NULL;
@@ -1831,6 +1830,40 @@ static unsigned int ilk_display_fifo_size(const struct drm_device *dev)
 		return 512;
 }
 
+static unsigned int ilk_plane_wm_reg_max(const struct drm_device *dev,
+					 int level, bool is_sprite)
+{
+	if (INTEL_INFO(dev)->gen >= 8)
+		/* BDW primary/sprite plane watermarks */
+		return level == 0 ? 255 : 2047;
+	else if (INTEL_INFO(dev)->gen >= 7)
+		/* IVB/HSW primary/sprite plane watermarks */
+		return level == 0 ? 127 : 1023;
+	else if (!is_sprite)
+		/* ILK/SNB primary plane watermarks */
+		return level == 0 ? 127 : 511;
+	else
+		/* ILK/SNB sprite plane watermarks */
+		return level == 0 ? 63 : 255;
+}
+
+static unsigned int ilk_cursor_wm_reg_max(const struct drm_device *dev,
+					  int level)
+{
+	if (INTEL_INFO(dev)->gen >= 7)
+		return level == 0 ? 63 : 255;
+	else
+		return level == 0 ? 31 : 63;
+}
+
+static unsigned int ilk_fbc_wm_reg_max(const struct drm_device *dev)
+{
+	if (INTEL_INFO(dev)->gen >= 8)
+		return 31;
+	else
+		return 15;
+}
+
 /* Calculate the maximum primary/sprite plane watermark */
 static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 				     int level,
@@ -1839,7 +1872,6 @@ static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 				     bool is_sprite)
 {
 	unsigned int fifo_size = ilk_display_fifo_size(dev);
-	unsigned int max;
 
 	/* if sprites aren't enabled, sprites get nothing */
 	if (is_sprite && !config->sprites_enabled)
@@ -1870,19 +1902,7 @@ static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 	}
 
 	/* clamp to max that the registers can hold */
-	if (INTEL_INFO(dev)->gen >= 8)
-		max = level == 0 ? 255 : 2047;
-	else if (INTEL_INFO(dev)->gen >= 7)
-		/* IVB/HSW primary/sprite plane watermarks */
-		max = level == 0 ? 127 : 1023;
-	else if (!is_sprite)
-		/* ILK/SNB primary plane watermarks */
-		max = level == 0 ? 127 : 511;
-	else
-		/* ILK/SNB sprite plane watermarks */
-		max = level == 0 ? 63 : 255;
-
-	return min(fifo_size, max);
+	return min(fifo_size, ilk_plane_wm_reg_max(dev, level, is_sprite));
 }
 
 /* Calculate the maximum cursor plane watermark */
@@ -1895,20 +1915,7 @@ static unsigned int ilk_cursor_wm_max(const struct drm_device *dev,
 		return 64;
 
 	/* otherwise just report max that registers can hold */
-	if (INTEL_INFO(dev)->gen >= 7)
-		return level == 0 ? 63 : 255;
-	else
-		return level == 0 ? 31 : 63;
-}
-
-/* Calculate the maximum FBC watermark */
-static unsigned int ilk_fbc_wm_max(const struct drm_device *dev)
-{
-	/* max that registers can hold */
-	if (INTEL_INFO(dev)->gen >= 8)
-		return 31;
-	else
-		return 15;
+	return ilk_cursor_wm_reg_max(dev, level);
 }
 
 static void ilk_compute_wm_maximums(const struct drm_device *dev,
@@ -1920,7 +1927,17 @@ static void ilk_compute_wm_maximums(const struct drm_device *dev,
 	max->pri = ilk_plane_wm_max(dev, level, config, ddb_partitioning, false);
 	max->spr = ilk_plane_wm_max(dev, level, config, ddb_partitioning, true);
 	max->cur = ilk_cursor_wm_max(dev, level, config);
-	max->fbc = ilk_fbc_wm_max(dev);
+	max->fbc = ilk_fbc_wm_reg_max(dev);
+}
+
+static void ilk_compute_wm_reg_maximums(struct drm_device *dev,
+					int level,
+					struct ilk_wm_maximums *max)
+{
+	max->pri = ilk_plane_wm_reg_max(dev, level, false);
+	max->spr = ilk_plane_wm_reg_max(dev, level, true);
+	max->cur = ilk_cursor_wm_reg_max(dev, level);
+	max->fbc = ilk_fbc_wm_reg_max(dev);
 }
 
 static bool ilk_validate_wm_level(int level,
@@ -2059,7 +2076,7 @@ static void intel_fixup_cur_wm_latency(struct drm_device *dev, uint16_t wm[5])
 		wm[3] *= 2;
 }
 
-static int ilk_wm_max_level(const struct drm_device *dev)
+int ilk_wm_max_level(const struct drm_device *dev)
 {
 	/* how many WM levels are we expecting */
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
@@ -2155,38 +2172,52 @@ static void ilk_setup_wm_latency(struct drm_device *dev)
 }
 
 static void ilk_compute_wm_parameters(struct drm_crtc *crtc,
-				      struct ilk_pipe_wm_parameters *p,
-				      struct intel_wm_config *config)
+				      struct ilk_pipe_wm_parameters *p)
 {
 	struct drm_device *dev = crtc->dev;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
 	struct drm_plane *plane;
 
-	p->active = intel_crtc_active(crtc);
-	if (p->active) {
-		p->pipe_htotal = intel_crtc->config.adjusted_mode.crtc_htotal;
-		p->pixel_rate = ilk_pipe_pixel_rate(dev, crtc);
-		p->pri.bytes_per_pixel = crtc->primary->fb->bits_per_pixel / 8;
-		p->cur.bytes_per_pixel = 4;
-		p->pri.horiz_pixels = intel_crtc->config.pipe_src_w;
-		p->cur.horiz_pixels = intel_crtc->cursor_width;
-		/* TODO: for now, assume primary and cursor planes are always enabled. */
-		p->pri.enabled = true;
-		p->cur.enabled = true;
-	}
+	if (!intel_crtc_active(crtc))
+		return;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		config->num_pipes_active += intel_crtc_active(crtc);
+	p->active = true;
+	p->pipe_htotal = intel_crtc->config.adjusted_mode.crtc_htotal;
+	p->pixel_rate = ilk_pipe_pixel_rate(dev, crtc);
+	p->pri.bytes_per_pixel = crtc->primary->fb->bits_per_pixel / 8;
+	p->cur.bytes_per_pixel = 4;
+	p->pri.horiz_pixels = intel_crtc->config.pipe_src_w;
+	p->cur.horiz_pixels = intel_crtc->cursor_width;
+	/* TODO: for now, assume primary and cursor planes are always enabled. */
+	p->pri.enabled = true;
+	p->cur.enabled = true;
 
 	drm_for_each_legacy_plane(plane, &dev->mode_config.plane_list) {
 		struct intel_plane *intel_plane = to_intel_plane(plane);
 
-		if (intel_plane->pipe == pipe)
+		if (intel_plane->pipe == pipe) {
 			p->spr = intel_plane->wm;
+			break;
+		}
+	}
+}
 
-		config->sprites_enabled |= intel_plane->wm.enabled;
-		config->sprites_scaled |= intel_plane->wm.scaled;
+static void ilk_compute_wm_config(struct drm_device *dev,
+				  struct intel_wm_config *config)
+{
+	struct intel_crtc *intel_crtc;
+
+	/* Compute the currently _active_ config */
+	for_each_intel_crtc(dev, intel_crtc) {
+		const struct intel_pipe_wm *wm = &intel_crtc->wm.active;
+
+		if (!wm->pipe_enabled)
+			continue;
+
+		config->sprites_enabled |= wm->sprites_enabled;
+		config->sprites_scaled |= wm->sprites_scaled;
+		config->num_pipes_active++;
 	}
 }
 
@@ -2206,8 +2237,9 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	};
 	struct ilk_wm_maximums max;
 
-	/* LP0 watermarks always use 1/2 DDB partitioning */
-	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
+	pipe_wm->pipe_enabled = params->active;
+	pipe_wm->sprites_enabled = params->spr.enabled;
+	pipe_wm->sprites_scaled = params->spr.scaled;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
 	if (INTEL_INFO(dev)->gen <= 6 && params->spr.enabled)
@@ -2217,15 +2249,37 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	if (params->spr.scaled)
 		max_level = 0;
 
-	for (level = 0; level <= max_level; level++)
-		ilk_compute_wm_level(dev_priv, level, params,
-				     &pipe_wm->wm[level]);
+	ilk_compute_wm_level(dev_priv, 0, params, &pipe_wm->wm[0]);
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
 
+	/* LP0 watermarks always use 1/2 DDB partitioning */
+	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
+
 	/* At least LP0 must be valid */
-	return ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]);
+	if (!ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]))
+		return false;
+
+	ilk_compute_wm_reg_maximums(dev, 1, &max);
+
+	for (level = 1; level <= max_level; level++) {
+		struct intel_wm_level wm = {};
+
+		ilk_compute_wm_level(dev_priv, level, params, &wm);
+
+		/*
+		 * Disable any watermark level that exceeds the
+		 * register maximums since such watermarks are
+		 * always invalid.
+		 */
+		if (!ilk_validate_wm_level(level, &max, &wm))
+			break;
+
+		pipe_wm->wm[level] = wm;
+	}
+
+	return true;
 }
 
 /*
@@ -2237,20 +2291,28 @@ static void ilk_merge_wm_level(struct drm_device *dev,
 {
 	const struct intel_crtc *intel_crtc;
 
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) {
-		const struct intel_wm_level *wm =
-			&intel_crtc->wm.active.wm[level];
+	ret_wm->enable = true;
+
+	for_each_intel_crtc(dev, intel_crtc) {
+		const struct intel_pipe_wm *active = &intel_crtc->wm.active;
+		const struct intel_wm_level *wm = &active->wm[level];
+
+		if (!active->pipe_enabled)
+			continue;
 
+		/*
+		 * The watermark values may have been used in the past,
+		 * so we must maintain them in the registers for some
+		 * time even if the level is now disabled.
+		 */
 		if (!wm->enable)
-			return;
+			ret_wm->enable = false;
 
 		ret_wm->pri_val = max(ret_wm->pri_val, wm->pri_val);
 		ret_wm->spr_val = max(ret_wm->spr_val, wm->spr_val);
 		ret_wm->cur_val = max(ret_wm->cur_val, wm->cur_val);
 		ret_wm->fbc_val = max(ret_wm->fbc_val, wm->fbc_val);
 	}
-
-	ret_wm->enable = true;
 }
 
 /*
@@ -2262,6 +2324,7 @@ static void ilk_wm_merge(struct drm_device *dev,
 			 struct intel_pipe_wm *merged)
 {
 	int level, max_level = ilk_wm_max_level(dev);
+	int last_enabled_level = max_level;
 
 	/* ILK/SNB/IVB: LP1+ watermarks only w/ single pipe */
 	if ((INTEL_INFO(dev)->gen <= 6 || IS_IVYBRIDGE(dev)) &&
@@ -2277,15 +2340,19 @@ static void ilk_wm_merge(struct drm_device *dev,
 
 		ilk_merge_wm_level(dev, level, wm);
 
-		if (!ilk_validate_wm_level(level, max, wm))
-			break;
+		if (level > last_enabled_level)
+			wm->enable = false;
+		else if (!ilk_validate_wm_level(level, max, wm))
+			/* make sure all following levels get disabled */
+			last_enabled_level = level - 1;
 
 		/*
 		 * The spec says it is preferred to disable
 		 * FBC WMs instead of disabling a WM level.
 		 */
 		if (wm->fbc_val > max->fbc) {
-			merged->fbc_wm_enabled = false;
+			if (wm->enable)
+				merged->fbc_wm_enabled = false;
 			wm->fbc_val = 0;
 		}
 	}
@@ -2340,14 +2407,19 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 		level = ilk_wm_lp_to_level(wm_lp, merged);
 
 		r = &merged->wm[level];
-		if (!r->enable)
-			break;
 
-		results->wm_lp[wm_lp - 1] = WM3_LP_EN |
+		/*
+		 * Maintain the watermark values even if the level is
+		 * disabled. Doing otherwise could cause underruns.
+		 */
+		results->wm_lp[wm_lp - 1] =
 			(ilk_wm_lp_latency(dev, level) << WM1_LP_LATENCY_SHIFT) |
 			(r->pri_val << WM1_LP_SR_SHIFT) |
 			r->cur_val;
 
+		if (r->enable)
+			results->wm_lp[wm_lp - 1] |= WM1_LP_SR_EN;
+
 		if (INTEL_INFO(dev)->gen >= 8)
 			results->wm_lp[wm_lp - 1] |=
 				r->fbc_val << WM1_LP_FBC_SHIFT_BDW;
@@ -2355,6 +2427,10 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 			results->wm_lp[wm_lp - 1] |=
 				r->fbc_val << WM1_LP_FBC_SHIFT;
 
+		/*
+		 * Always set WM1S_LP_EN when spr_val != 0, even if the
+		 * level is disabled. Doing otherwise could cause underruns.
+		 */
 		if (INTEL_INFO(dev)->gen <= 6 && r->spr_val) {
 			WARN_ON(wm_lp != 1);
 			results->wm_lp_spr[wm_lp - 1] = WM1S_LP_EN | r->spr_val;
@@ -2363,7 +2439,7 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 	}
 
 	/* LP0 register values */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		enum pipe pipe = intel_crtc->pipe;
 		const struct intel_wm_level *r =
 			&intel_crtc->wm.active.wm[0];
@@ -2598,7 +2674,7 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct intel_wm_config config = {};
 
-	ilk_compute_wm_parameters(crtc, &params, &config);
+	ilk_compute_wm_parameters(crtc, &params);
 
 	intel_compute_pipe_wm(crtc, &params, &pipe_wm);
 
@@ -2607,6 +2683,8 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 
 	intel_crtc->wm.active = pipe_wm;
 
+	ilk_compute_wm_config(dev, &config);
+
 	ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_1_2, &max);
 	ilk_wm_merge(dev, &config, &max, &lp_wm_1_2);
 
@@ -2673,7 +2751,9 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		hw->wm_linetime[pipe] = I915_READ(PIPE_WM_LINETIME(pipe));
 
-	if (intel_crtc_active(crtc)) {
+	active->pipe_enabled = intel_crtc_active(crtc);
+
+	if (active->pipe_enabled) {
 		u32 tmp = hw->wm_pipe[pipe];
 
 		/*
@@ -2706,7 +2786,7 @@ void ilk_wm_get_hw_state(struct drm_device *dev)
 	struct ilk_wm_values *hw = &dev_priv->wm.hw;
 	struct drm_crtc *crtc;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+	for_each_crtc(dev, crtc)
 		ilk_pipe_wm_get_hw_state(crtc);
 
 	hw->wm_lp[0] = I915_READ(WM1_LP_ILK);
@@ -2714,8 +2794,10 @@ void ilk_wm_get_hw_state(struct drm_device *dev)
 	hw->wm_lp[2] = I915_READ(WM3_LP_ILK);
 
 	hw->wm_lp_spr[0] = I915_READ(WM1S_LP_ILK);
-	hw->wm_lp_spr[1] = I915_READ(WM2S_LP_IVB);
-	hw->wm_lp_spr[2] = I915_READ(WM3S_LP_IVB);
+	if (INTEL_INFO(dev)->gen >= 7) {
+		hw->wm_lp_spr[1] = I915_READ(WM2S_LP_IVB);
+		hw->wm_lp_spr[2] = I915_READ(WM3S_LP_IVB);
+	}
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		hw->partitioning = (I915_READ(WM_MISC) & WM_MISC_DATA_PARTITION_5_6) ?
@@ -3071,6 +3153,9 @@ static u32 gen6_rps_pm_mask(struct drm_i915_private *dev_priv, u8 val)
 	if (INTEL_INFO(dev_priv->dev)->gen <= 7 && !IS_HASWELL(dev_priv->dev))
 		mask |= GEN6_PM_RP_UP_EI_EXPIRED;
 
+	if (IS_GEN8(dev_priv->dev))
+		mask |= GEN8_PMINTR_REDIRECT_TO_NON_DISP;
+
 	return ~mask;
 }
 
@@ -3091,7 +3176,7 @@ void gen6_set_rps(struct drm_device *dev, u8 val)
 	if (val != dev_priv->rps.cur_freq) {
 		gen6_set_rps_thresholds(dev_priv, val);
 
-		if (IS_HASWELL(dev))
+		if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 			I915_WRITE(GEN6_RPNSWREQ,
 				   HSW_FREQUENCY(val));
 		else
@@ -3124,6 +3209,14 @@ void gen6_set_rps(struct drm_device *dev, u8 val)
 */
 static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 {
+	struct drm_device *dev = dev_priv->dev;
+
+	/* Latest VLV doesn't need to force the gfx clock */
+	if (dev->pdev->revision >= 0xd) {
+		valleyview_set_rps(dev_priv->dev, dev_priv->rps.min_freq_softlimit);
+		return;
+	}
+
 	/*
 	 * When we are idle.  Drop to min voltage state.
 	 */
@@ -3134,16 +3227,7 @@ static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 	/* Mask turbo interrupt so that they will not come in between */
 	I915_WRITE(GEN6_PMINTRMSK, 0xffffffff);
 
-	/* Bring up the Gfx clock */
-	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG,
-		I915_READ(VLV_GTLC_SURVIVABILITY_REG) |
-				VLV_GFX_CLK_FORCE_ON_BIT);
-
-	if (wait_for(((VLV_GFX_CLK_STATUS_BIT &
-		I915_READ(VLV_GTLC_SURVIVABILITY_REG)) != 0), 5)) {
-			DRM_ERROR("GFX_CLK_ON request timed out\n");
-		return;
-	}
+	vlv_force_gfx_clock(dev_priv, true);
 
 	dev_priv->rps.cur_freq = dev_priv->rps.min_freq_softlimit;
 
@@ -3154,10 +3238,7 @@ static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 				& GENFREQSTATUS) == 0, 5))
 		DRM_ERROR("timed out waiting for Punit\n");
 
-	/* Release the Gfx clock */
-	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG,
-		I915_READ(VLV_GTLC_SURVIVABILITY_REG) &
-				~VLV_GFX_CLK_FORCE_ON_BIT);
+	vlv_force_gfx_clock(dev_priv, false);
 
 	I915_WRITE(GEN6_PMINTRMSK,
 		   gen6_rps_pm_mask(dev_priv, dev_priv->rps.cur_freq));
@@ -3215,6 +3296,26 @@ void valleyview_set_rps(struct drm_device *dev, u8 val)
 	trace_intel_gpu_freq_change(vlv_gpu_freq(dev_priv, val));
 }
 
+static void gen8_disable_rps_interrupts(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(GEN6_PMINTRMSK, ~GEN8_PMINTR_REDIRECT_TO_NON_DISP);
+	I915_WRITE(GEN8_GT_IER(2), I915_READ(GEN8_GT_IER(2)) &
+				   ~dev_priv->pm_rps_events);
+	/* Complete PM interrupt masking here doesn't race with the rps work
+	 * item again unmasking PM interrupts because that is using a different
+	 * register (GEN8_GT_IMR(2)) to mask PM interrupts. The only risk is in
+	 * leaving stale bits in GEN8_GT_IIR(2) and GEN8_GT_IMR(2) which
+	 * gen8_enable_rps will clean up. */
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	dev_priv->rps.pm_iir = 0;
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	I915_WRITE(GEN8_GT_IIR(2), dev_priv->pm_rps_events);
+}
+
 static void gen6_disable_rps_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3241,7 +3342,10 @@ static void gen6_disable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 	I915_WRITE(GEN6_RPNSWREQ, 1 << 31);
 
-	gen6_disable_rps_interrupts(dev);
+	if (IS_BROADWELL(dev))
+		gen8_disable_rps_interrupts(dev);
+	else
+		gen6_disable_rps_interrupts(dev);
 }
 
 static void valleyview_disable_rps(struct drm_device *dev)
@@ -3255,21 +3359,44 @@ static void valleyview_disable_rps(struct drm_device *dev)
 
 static void intel_print_rc6_info(struct drm_device *dev, u32 mode)
 {
+	if (IS_VALLEYVIEW(dev)) {
+		if (mode & (GEN7_RC_CTL_TO_MODE | GEN6_RC_CTL_EI_MODE(1)))
+			mode = GEN6_RC_CTL_RC6_ENABLE;
+		else
+			mode = 0;
+	}
 	DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n",
 		 (mode & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off",
 		 (mode & GEN6_RC_CTL_RC6p_ENABLE) ? "on" : "off",
 		 (mode & GEN6_RC_CTL_RC6pp_ENABLE) ? "on" : "off");
 }
 
-int intel_enable_rc6(const struct drm_device *dev)
+static int sanitize_rc6_option(const struct drm_device *dev, int enable_rc6)
 {
 	/* No RC6 before Ironlake */
 	if (INTEL_INFO(dev)->gen < 5)
 		return 0;
 
+	/* RC6 is only on Ironlake mobile not on desktop */
+	if (INTEL_INFO(dev)->gen == 5 && !IS_IRONLAKE_M(dev))
+		return 0;
+
 	/* Respect the kernel parameter if it is set */
-	if (i915.enable_rc6 >= 0)
-		return i915.enable_rc6;
+	if (enable_rc6 >= 0) {
+		int mask;
+
+		if (INTEL_INFO(dev)->gen == 6 || IS_IVYBRIDGE(dev))
+			mask = INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE |
+			       INTEL_RC6pp_ENABLE;
+		else
+			mask = INTEL_RC6_ENABLE;
+
+		if ((enable_rc6 & mask) != enable_rc6)
+			DRM_INFO("Adjusting RC6 mask to %d (requested %d, valid %d)\n",
+				 enable_rc6 & mask, enable_rc6, mask);
+
+		return enable_rc6 & mask;
+	}
 
 	/* Disable RC6 on Ironlake */
 	if (INTEL_INFO(dev)->gen == 5)
@@ -3281,6 +3408,22 @@ int intel_enable_rc6(const struct drm_device *dev)
 	return INTEL_RC6_ENABLE;
 }
 
+int intel_enable_rc6(const struct drm_device *dev)
+{
+	return i915.enable_rc6;
+}
+
+static void gen8_enable_rps_interrupts(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	WARN_ON(dev_priv->rps.pm_iir);
+	bdw_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	I915_WRITE(GEN8_GT_IIR(2), dev_priv->pm_rps_events);
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
 static void gen6_enable_rps_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3292,10 +3435,31 @@ static void gen6_enable_rps_interrupts(struct drm_device *dev)
 	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
+static void parse_rp_state_cap(struct drm_i915_private *dev_priv, u32 rp_state_cap)
+{
+	/* All of these values are in units of 50MHz */
+	dev_priv->rps.cur_freq		= 0;
+	/* static values from HW: RP0 < RPe < RP1 < RPn (min_freq) */
+	dev_priv->rps.rp1_freq		= (rp_state_cap >>  8) & 0xff;
+	dev_priv->rps.rp0_freq		= (rp_state_cap >>  0) & 0xff;
+	dev_priv->rps.min_freq		= (rp_state_cap >> 16) & 0xff;
+	/* XXX: only BYT has a special efficient freq */
+	dev_priv->rps.efficient_freq	= dev_priv->rps.rp1_freq;
+	/* hw_max = RP0 until we check for overclocking */
+	dev_priv->rps.max_freq		= dev_priv->rps.rp0_freq;
+
+	/* Preserve min/max settings in case of re-init */
+	if (dev_priv->rps.max_freq_softlimit == 0)
+		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
+
+	if (dev_priv->rps.min_freq_softlimit == 0)
+		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+}
+
 static void gen8_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	uint32_t rc6_mask = 0, rp_state_cap;
 	int unused;
 
@@ -3310,6 +3474,7 @@ static void gen8_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 
 	rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+	parse_rp_state_cap(dev_priv, rp_state_cap);
 
 	/* 2b: Program RC6 thresholds.*/
 	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
@@ -3329,8 +3494,10 @@ static void gen8_enable_rps(struct drm_device *dev)
 				    rc6_mask);
 
 	/* 4 Program defaults and thresholds for RPS*/
-	I915_WRITE(GEN6_RPNSWREQ, HSW_FREQUENCY(10)); /* Request 500 MHz */
-	I915_WRITE(GEN6_RC_VIDEO_FREQ, HSW_FREQUENCY(12)); /* Request 600 MHz */
+	I915_WRITE(GEN6_RPNSWREQ,
+		   HSW_FREQUENCY(dev_priv->rps.rp1_freq));
+	I915_WRITE(GEN6_RC_VIDEO_FREQ,
+		   HSW_FREQUENCY(dev_priv->rps.rp1_freq));
 	/* NB: Docs say 1s, and 1000000 - which aren't equivalent */
 	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 100000000 / 128); /* 1 second timeout */
 
@@ -3359,7 +3526,7 @@ static void gen8_enable_rps(struct drm_device *dev)
 
 	gen6_set_rps(dev, (I915_READ(GEN6_GT_PERF_STATUS) & 0xff00) >> 8);
 
-	gen6_enable_rps_interrupts(dev);
+	gen8_enable_rps_interrupts(dev);
 
 	gen6_gt_force_wake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -3367,7 +3534,7 @@ static void gen8_enable_rps(struct drm_device *dev)
 static void gen6_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	u32 rp_state_cap;
 	u32 gt_perf_status;
 	u32 rc6vids, pcu_mbox = 0, rc6_mask = 0;
@@ -3396,23 +3563,7 @@ static void gen6_enable_rps(struct drm_device *dev)
 	rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
 	gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 
-	/* All of these values are in units of 50MHz */
-	dev_priv->rps.cur_freq		= 0;
-	/* static values from HW: RP0 < RPe < RP1 < RPn (min_freq) */
-	dev_priv->rps.rp1_freq		= (rp_state_cap >>  8) & 0xff;
-	dev_priv->rps.rp0_freq		= (rp_state_cap >>  0) & 0xff;
-	dev_priv->rps.min_freq		= (rp_state_cap >> 16) & 0xff;
-	/* XXX: only BYT has a special efficient freq */
-	dev_priv->rps.efficient_freq	= dev_priv->rps.rp1_freq;
-	/* hw_max = RP0 until we check for overclocking */
-	dev_priv->rps.max_freq		= dev_priv->rps.rp0_freq;
-
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
-
-	if (dev_priv->rps.min_freq_softlimit == 0)
-		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+	parse_rp_state_cap(dev_priv, rp_state_cap);
 
 	/* disable the counters and set deterministic thresholds */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -3494,7 +3645,7 @@ static void gen6_enable_rps(struct drm_device *dev)
 	gen6_gt_force_wake_put(dev_priv, FORCEWAKE_ALL);
 }
 
-void gen6_update_ring_freq(struct drm_device *dev)
+static void __gen6_update_ring_freq(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int min_freq = 15;
@@ -3564,6 +3715,18 @@ void gen6_update_ring_freq(struct drm_device *dev)
 	}
 }
 
+void gen6_update_ring_freq(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (INTEL_INFO(dev)->gen < 6 || IS_VALLEYVIEW(dev))
+		return;
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+	__gen6_update_ring_freq(dev);
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
 int valleyview_rps_max_freq(struct drm_i915_private *dev_priv)
 {
 	u32 val, rp0;
@@ -3658,10 +3821,49 @@ static void valleyview_cleanup_pctx(struct drm_device *dev)
 	dev_priv->vlv_pctx = NULL;
 }
 
+static void valleyview_init_gt_powersave(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	valleyview_setup_pctx(dev);
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+
+	dev_priv->rps.max_freq = valleyview_rps_max_freq(dev_priv);
+	dev_priv->rps.rp0_freq = dev_priv->rps.max_freq;
+	DRM_DEBUG_DRIVER("max GPU freq: %d MHz (%u)\n",
+			 vlv_gpu_freq(dev_priv, dev_priv->rps.max_freq),
+			 dev_priv->rps.max_freq);
+
+	dev_priv->rps.efficient_freq = valleyview_rps_rpe_freq(dev_priv);
+	DRM_DEBUG_DRIVER("RPe GPU freq: %d MHz (%u)\n",
+			 vlv_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
+			 dev_priv->rps.efficient_freq);
+
+	dev_priv->rps.min_freq = valleyview_rps_min_freq(dev_priv);
+	DRM_DEBUG_DRIVER("min GPU freq: %d MHz (%u)\n",
+			 vlv_gpu_freq(dev_priv, dev_priv->rps.min_freq),
+			 dev_priv->rps.min_freq);
+
+	/* Preserve min/max settings in case of re-init */
+	if (dev_priv->rps.max_freq_softlimit == 0)
+		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
+
+	if (dev_priv->rps.min_freq_softlimit == 0)
+		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+static void valleyview_cleanup_gt_powersave(struct drm_device *dev)
+{
+	valleyview_cleanup_pctx(dev);
+}
+
 static void valleyview_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	u32 gtfifodbg, val, rc6_mode = 0;
 	int i;
 
@@ -3724,29 +3926,6 @@ static void valleyview_enable_rps(struct drm_device *dev)
 			 vlv_gpu_freq(dev_priv, dev_priv->rps.cur_freq),
 			 dev_priv->rps.cur_freq);
 
-	dev_priv->rps.max_freq = valleyview_rps_max_freq(dev_priv);
-	dev_priv->rps.rp0_freq  = dev_priv->rps.max_freq;
-	DRM_DEBUG_DRIVER("max GPU freq: %d MHz (%u)\n",
-			 vlv_gpu_freq(dev_priv, dev_priv->rps.max_freq),
-			 dev_priv->rps.max_freq);
-
-	dev_priv->rps.efficient_freq = valleyview_rps_rpe_freq(dev_priv);
-	DRM_DEBUG_DRIVER("RPe GPU freq: %d MHz (%u)\n",
-			 vlv_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
-			 dev_priv->rps.efficient_freq);
-
-	dev_priv->rps.min_freq = valleyview_rps_min_freq(dev_priv);
-	DRM_DEBUG_DRIVER("min GPU freq: %d MHz (%u)\n",
-			 vlv_gpu_freq(dev_priv, dev_priv->rps.min_freq),
-			 dev_priv->rps.min_freq);
-
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
-
-	if (dev_priv->rps.min_freq_softlimit == 0)
-		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
-
 	DRM_DEBUG_DRIVER("setting GPU freq to %d MHz (%u)\n",
 			 vlv_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
 			 dev_priv->rps.efficient_freq);
@@ -3815,7 +3994,7 @@ static int ironlake_setup_rc6(struct drm_device *dev)
 static void ironlake_enable_rc6(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	bool was_interruptible;
 	int ret;
 
@@ -3873,7 +4052,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)
 	I915_WRITE(PWRCTXA, i915_gem_obj_ggtt_offset(dev_priv->ips.pwrctx) | PWRCTX_EN);
 	I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
 
-	intel_print_rc6_info(dev, INTEL_RC6_ENABLE);
+	intel_print_rc6_info(dev, GEN6_RC_CTL_RC6_ENABLE);
 }
 
 static unsigned long intel_pxfreq(u32 vidfreq)
@@ -4327,7 +4506,7 @@ EXPORT_SYMBOL_GPL(i915_gpu_lower);
 bool i915_gpu_busy(void)
 {
 	struct drm_i915_private *dev_priv;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	bool ret = false;
 	int i;
 
@@ -4487,14 +4666,16 @@ static void intel_init_emon(struct drm_device *dev)
 
 void intel_init_gt_powersave(struct drm_device *dev)
 {
+	i915.enable_rc6 = sanitize_rc6_option(dev, i915.enable_rc6);
+
 	if (IS_VALLEYVIEW(dev))
-		valleyview_setup_pctx(dev);
+		valleyview_init_gt_powersave(dev);
 }
 
 void intel_cleanup_gt_powersave(struct drm_device *dev)
 {
 	if (IS_VALLEYVIEW(dev))
-		valleyview_cleanup_pctx(dev);
+		valleyview_cleanup_gt_powersave(dev);
 }
 
 void intel_disable_gt_powersave(struct drm_device *dev)
@@ -4507,8 +4688,10 @@ void intel_disable_gt_powersave(struct drm_device *dev)
 	if (IS_IRONLAKE_M(dev)) {
 		ironlake_disable_drps(dev);
 		ironlake_disable_rc6(dev);
-	} else if (INTEL_INFO(dev)->gen >= 6) {
-		cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work);
+	} else if (IS_GEN6(dev) || IS_GEN7(dev) || IS_BROADWELL(dev)) {
+		if (cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work))
+			intel_runtime_pm_put(dev_priv);
+
 		cancel_work_sync(&dev_priv->rps.work);
 		mutex_lock(&dev_priv->rps.hw_lock);
 		if (IS_VALLEYVIEW(dev))
@@ -4533,13 +4716,15 @@ static void intel_gen6_powersave_work(struct work_struct *work)
 		valleyview_enable_rps(dev);
 	} else if (IS_BROADWELL(dev)) {
 		gen8_enable_rps(dev);
-		gen6_update_ring_freq(dev);
+		__gen6_update_ring_freq(dev);
 	} else {
 		gen6_enable_rps(dev);
-		gen6_update_ring_freq(dev);
+		__gen6_update_ring_freq(dev);
 	}
 	dev_priv->rps.enabled = true;
 	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	intel_runtime_pm_put(dev_priv);
 }
 
 void intel_enable_gt_powersave(struct drm_device *dev)
@@ -4547,20 +4732,38 @@ void intel_enable_gt_powersave(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (IS_IRONLAKE_M(dev)) {
+		mutex_lock(&dev->struct_mutex);
 		ironlake_enable_drps(dev);
 		ironlake_enable_rc6(dev);
 		intel_init_emon(dev);
-	} else if (IS_GEN6(dev) || IS_GEN7(dev)) {
+		mutex_unlock(&dev->struct_mutex);
+	} else if (IS_GEN6(dev) || IS_GEN7(dev) || IS_BROADWELL(dev)) {
 		/*
 		 * PCU communication is slow and this doesn't need to be
 		 * done at any specific time, so do this out of our fast path
 		 * to make resume and init faster.
+		 *
+		 * We depend on the HW RC6 power context save/restore
+		 * mechanism when entering D3 through runtime PM suspend. So
+		 * disable RPM until RPS/RC6 is properly setup. We can only
+		 * get here via the driver load/system resume/runtime resume
+		 * paths, so the _noresume version is enough (and in case of
+		 * runtime resume it's necessary).
 		 */
-		schedule_delayed_work(&dev_priv->rps.delayed_resume_work,
-				      round_jiffies_up_relative(HZ));
+		if (schedule_delayed_work(&dev_priv->rps.delayed_resume_work,
+					   round_jiffies_up_relative(HZ)))
+			intel_runtime_pm_get_noresume(dev_priv);
 	}
 }
 
+void intel_reset_gt_powersave(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->rps.enabled = false;
+	intel_enable_gt_powersave(dev);
+}
+
 static void ibx_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4666,6 +4869,9 @@ static void ironlake_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(CACHE_MODE_0,
 		   _MASKED_BIT_ENABLE(CM0_PIPELINED_RENDER_FLUSH_DISABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:ilk */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	g4x_disable_trickle_feed(dev);
 
 	ibx_init_clock_gating(dev);
@@ -4741,6 +4947,9 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 		I915_WRITE(GEN6_GT_MODE,
 			   _MASKED_BIT_ENABLE(GEN6_TD_FOUR_ROW_DISPATCH_DISABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:snb */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/*
 	 * BSpec recoomends 8x4 when MSAA is used,
 	 * however in practice 16x4 seems fastest.
@@ -4909,6 +5118,10 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN7_HALF_SLICE_CHICKEN1,
 		   _MASKED_BIT_ENABLE(GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE));
 
+	/* WaDisableDopClockGating:bdw May not be needed for production */
+	I915_WRITE(GEN7_ROW_CHICKEN2,
+		   _MASKED_BIT_ENABLE(DOP_CLOCK_GATING_DISABLE));
+
 	/* WaSwitchSolVfFArbitrationPriority:bdw */
 	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | HSW_ECOCHK_ARB_PRIO_SOL);
 
@@ -4980,6 +5193,9 @@ static void haswell_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN7_FF_THREAD_MODE,
 		   I915_READ(GEN7_FF_THREAD_MODE) & ~GEN7_FF_VS_REF_CNT_FFME);
 
+	/* WaDisable_RenderCache_OperationalFlush:hsw */
+	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/* enable HiZ Raw Stall Optimization */
 	I915_WRITE(CACHE_MODE_0_GEN7,
 		   _MASKED_BIT_DISABLE(HIZ_RAW_STALL_OPT_DISABLE));
@@ -5032,6 +5248,9 @@ static void ivybridge_init_clock_gating(struct drm_device *dev)
 		I915_WRITE(GEN7_HALF_SLICE_CHICKEN1,
 			   _MASKED_BIT_ENABLE(GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:ivb */
+	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/* Apply the WaDisableRHWOOptimizationForRenderHang:ivb workaround. */
 	I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1,
 		   GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC);
@@ -5126,6 +5345,10 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	}
 	DRM_DEBUG_DRIVER("DDR speed: %d MHz", dev_priv->mem_freq);
 
+	dev_priv->vlv_cdclk_freq = valleyview_cur_cdclk(dev_priv);
+	DRM_DEBUG_DRIVER("Current CD clock rate: %d MHz",
+			 dev_priv->vlv_cdclk_freq);
+
 	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
 
 	/* WaDisableEarlyCull:vlv */
@@ -5143,6 +5366,9 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 		   _MASKED_BIT_ENABLE(GEN7_MAX_PS_THREAD_DEP |
 				      GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:vlv */
+	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/* WaForceL3Serialization:vlv */
 	I915_WRITE(GEN7_L3SQCREG4, I915_READ(GEN7_L3SQCREG4) &
 		   ~L3SQ_URB_READ_CAM_MATCH_DISABLE);
@@ -5165,8 +5391,11 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN6_UCGCTL2,
 		   GEN6_RCZUNIT_CLOCK_GATE_DISABLE);
 
-	/* WaDisableL3Bank2xClockGate:vlv */
-	I915_WRITE(GEN7_UCGCTL4, GEN7_L3BANK2X_CLOCK_GATE_DISABLE);
+	/* WaDisableL3Bank2xClockGate:vlv
+	 * Disabling L3 clock gating- MMIO 940c[25] = 1
+	 * Set bit 25, to disable L3_BANK_2x_CLK_GATING */
+	I915_WRITE(GEN7_UCGCTL4,
+		   I915_READ(GEN7_UCGCTL4) | GEN7_L3BANK2X_CLOCK_GATE_DISABLE);
 
 	I915_WRITE(MI_ARB_VLV, MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE);
 
@@ -5191,6 +5420,59 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(VLV_GUNIT_CLOCK_GATE, GCFG_DIS);
 }
 
+static void cherryview_init_clock_gating(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
+
+	I915_WRITE(MI_ARB_VLV, MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE);
+
+	/* WaDisablePartialInstShootdown:chv */
+	I915_WRITE(GEN8_ROW_CHICKEN,
+		   _MASKED_BIT_ENABLE(PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE));
+
+	/* WaDisableThreadStallDopClockGating:chv */
+	I915_WRITE(GEN8_ROW_CHICKEN,
+		   _MASKED_BIT_ENABLE(STALL_DOP_GATING_DISABLE));
+
+	/* WaVSRefCountFullforceMissDisable:chv */
+	/* WaDSRefCountFullforceMissDisable:chv */
+	I915_WRITE(GEN7_FF_THREAD_MODE,
+		   I915_READ(GEN7_FF_THREAD_MODE) &
+		   ~(GEN8_FF_DS_REF_CNT_FFME | GEN7_FF_VS_REF_CNT_FFME));
+
+	/* WaDisableSemaphoreAndSyncFlipWait:chv */
+	I915_WRITE(GEN6_RC_SLEEP_PSMI_CONTROL,
+		   _MASKED_BIT_ENABLE(GEN8_RC_SEMA_IDLE_MSG_DISABLE));
+
+	/* WaDisableCSUnitClockGating:chv */
+	I915_WRITE(GEN6_UCGCTL1, I915_READ(GEN6_UCGCTL1) |
+		   GEN6_CSUNIT_CLOCK_GATE_DISABLE);
+
+	/* WaDisableSDEUnitClockGating:chv */
+	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
+		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+
+	/* WaDisableSamplerPowerBypass:chv (pre-production hw) */
+	I915_WRITE(HALF_SLICE_CHICKEN3,
+		   _MASKED_BIT_ENABLE(GEN8_SAMPLER_POWER_BYPASS_DIS));
+
+	/* WaDisableGunitClockGating:chv (pre-production hw) */
+	I915_WRITE(VLV_GUNIT_CLOCK_GATE, I915_READ(VLV_GUNIT_CLOCK_GATE) |
+		   GINT_DIS);
+
+	/* WaDisableFfDopClockGating:chv (pre-production hw) */
+	I915_WRITE(GEN6_RC_SLEEP_PSMI_CONTROL,
+		   _MASKED_BIT_ENABLE(GEN8_FF_DOP_CLOCK_GATE_DISABLE));
+
+	/* WaDisableDopClockGating:chv (pre-production hw) */
+	I915_WRITE(GEN7_ROW_CHICKEN2,
+		   _MASKED_BIT_ENABLE(DOP_CLOCK_GATING_DISABLE));
+	I915_WRITE(GEN6_UCGCTL1, I915_READ(GEN6_UCGCTL1) |
+		   GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE);
+}
+
 static void g4x_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -5212,6 +5494,9 @@ static void g4x_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(CACHE_MODE_0,
 		   _MASKED_BIT_ENABLE(CM0_PIPELINED_RENDER_FLUSH_DISABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:g4x */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	g4x_disable_trickle_feed(dev);
 }
 
@@ -5226,6 +5511,9 @@ static void crestline_init_clock_gating(struct drm_device *dev)
 	I915_WRITE16(DEUC, 0);
 	I915_WRITE(MI_ARB_STATE,
 		   _MASKED_BIT_ENABLE(MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE));
+
+	/* WaDisable_RenderCache_OperationalFlush:gen4 */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
 }
 
 static void broadwater_init_clock_gating(struct drm_device *dev)
@@ -5240,6 +5528,9 @@ static void broadwater_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(RENCLK_GATE_D2, 0);
 	I915_WRITE(MI_ARB_STATE,
 		   _MASKED_BIT_ENABLE(MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE));
+
+	/* WaDisable_RenderCache_OperationalFlush:gen4 */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
 }
 
 static void gen3_init_clock_gating(struct drm_device *dev)
@@ -5256,6 +5547,12 @@ static void gen3_init_clock_gating(struct drm_device *dev)
 
 	/* IIR "flip pending" means done if this bit is set */
 	I915_WRITE(ECOSKPD, _MASKED_BIT_DISABLE(ECO_FLIP_DONE));
+
+	/* interrupts should cause a wake up from C3 */
+	I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_AGPBUSY_INT_EN));
+
+	/* On GEN3 we really need to make sure the ARB C3 LP bit is set */
+	I915_WRITE(MI_ARB_STATE, _MASKED_BIT_ENABLE(MI_ARB_C3_LP_WRITE_ENABLE));
 }
 
 static void i85x_init_clock_gating(struct drm_device *dev)
@@ -5263,6 +5560,10 @@ static void i85x_init_clock_gating(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	I915_WRITE(RENCLK_GATE_D1, SV_CLOCK_GATE_DISABLE);
+
+	/* interrupts should cause a wake up from C3 */
+	I915_WRITE(MI_STATE, _MASKED_BIT_ENABLE(MI_AGPBUSY_INT_EN) |
+		   _MASKED_BIT_DISABLE(MI_AGPBUSY_830_MODE));
 }
 
 static void i830_init_clock_gating(struct drm_device *dev)
@@ -5310,18 +5611,8 @@ static bool hsw_power_well_enabled(struct drm_i915_private *dev_priv,
 		     (HSW_PWR_WELL_ENABLE_REQUEST | HSW_PWR_WELL_STATE_ENABLED);
 }
 
-bool intel_display_power_enabled_sw(struct drm_i915_private *dev_priv,
-				    enum intel_display_power_domain domain)
-{
-	struct i915_power_domains *power_domains;
-
-	power_domains = &dev_priv->power_domains;
-
-	return power_domains->domain_use_count[domain];
-}
-
-bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
-				 enum intel_display_power_domain domain)
+bool intel_display_power_enabled_unlocked(struct drm_i915_private *dev_priv,
+					  enum intel_display_power_domain domain)
 {
 	struct i915_power_domains *power_domains;
 	struct i915_power_well *power_well;
@@ -5335,21 +5626,34 @@ bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
 
 	is_enabled = true;
 
-	mutex_lock(&power_domains->lock);
 	for_each_power_well_rev(i, power_well, BIT(domain), power_domains) {
 		if (power_well->always_on)
 			continue;
 
-		if (!power_well->ops->is_enabled(dev_priv, power_well)) {
+		if (!power_well->hw_enabled) {
 			is_enabled = false;
 			break;
 		}
 	}
-	mutex_unlock(&power_domains->lock);
 
 	return is_enabled;
 }
 
+bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
+				 enum intel_display_power_domain domain)
+{
+	struct i915_power_domains *power_domains;
+	bool ret;
+
+	power_domains = &dev_priv->power_domains;
+
+	mutex_lock(&power_domains->lock);
+	ret = intel_display_power_enabled_unlocked(dev_priv, domain);
+	mutex_unlock(&power_domains->lock);
+
+	return ret;
+}
+
 /*
  * Starting with Haswell, we have a "Power Down Well" that can be turned off
  * when not needed anymore. We have 4 registers that can request the power well
@@ -5392,33 +5696,6 @@ static void hsw_power_well_post_enable(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void reset_vblank_counter(struct drm_device *dev, enum pipe pipe)
-{
-	assert_spin_locked(&dev->vbl_lock);
-
-	dev->vblank[pipe].last = 0;
-}
-
-static void hsw_power_well_post_disable(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-	enum pipe pipe;
-	unsigned long irqflags;
-
-	/*
-	 * After this, the registers on the pipes that are part of the power
-	 * well will become zero, so we have to adjust our counters according to
-	 * that.
-	 *
-	 * FIXME: Should we do this in general in drm_vblank_post_modeset?
-	 */
-	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	for_each_pipe(pipe)
-		if (pipe != PIPE_A)
-			reset_vblank_counter(dev, pipe);
-	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
-}
-
 static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 			       struct i915_power_well *power_well, bool enable)
 {
@@ -5447,8 +5724,6 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 			I915_WRITE(HSW_PWR_WELL_DRIVER, 0);
 			POSTING_READ(HSW_PWR_WELL_DRIVER);
 			DRM_DEBUG_KMS("Requesting to disable the power well\n");
-
-			hsw_power_well_post_disable(dev_priv);
 		}
 	}
 }
@@ -5489,13 +5764,34 @@ static bool i9xx_always_on_power_well_enabled(struct drm_i915_private *dev_priv,
 	return true;
 }
 
-static void vlv_set_power_well(struct drm_i915_private *dev_priv,
-			       struct i915_power_well *power_well, bool enable)
+void __vlv_set_power_well(struct drm_i915_private *dev_priv,
+			  enum punit_power_well power_well_id, bool enable)
 {
-	enum punit_power_well power_well_id = power_well->data;
+	struct drm_device *dev = dev_priv->dev;
 	u32 mask;
 	u32 state;
 	u32 ctrl;
+	enum pipe pipe;
+
+	if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC) {
+		if (enable) {
+			/*
+			 * Enable the CRI clock source so we can get at the
+			 * display and the reference clock for VGA
+			 * hotplug / manual detection.
+			 */
+			I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
+				   DPLL_REFA_CLK_ENABLE_VLV |
+				   DPLL_INTEGRATED_CRI_CLK_VLV);
+			udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
+		} else {
+			for_each_pipe(pipe)
+				assert_pll_disabled(dev_priv, pipe);
+			/* Assert common reset */
+			I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) &
+				   ~DPIO_CMNRST);
+		}
+	}
 
 	mask = PUNIT_PWRGT_MASK(power_well_id);
 	state = enable ? PUNIT_PWRGT_PWR_ON(power_well_id) :
@@ -5523,6 +5819,28 @@ static void vlv_set_power_well(struct drm_i915_private *dev_priv,
 
 out:
 	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	/*
+	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
+	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
+	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
+	 *   b.	The other bits such as sfr settings / modesel may all
+	 *	be set to 0.
+	 *
+	 * This should only be done on init and resume from S3 with
+	 * both PLLs disabled, or we risk losing DPIO and PLL
+	 * synchronization.
+	 */
+	if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC && enable)
+		I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
+}
+
+static void vlv_set_power_well(struct drm_i915_private *dev_priv,
+			       struct i915_power_well *power_well, bool enable)
+{
+	enum punit_power_well power_well_id = power_well->data;
+
+	__vlv_set_power_well(dev_priv, power_well_id, enable);
 }
 
 static void vlv_power_well_sync_hw(struct drm_i915_private *dev_priv,
@@ -5591,11 +5909,13 @@ static void vlv_display_power_well_enable(struct drm_i915_private *dev_priv,
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	/*
-	 * During driver initialization we need to defer enabling hotplug
-	 * processing until fbdev is set up.
+	 * During driver initialization/resume we can avoid restoring the
+	 * part of the HW/SW state that will be inited anyway explicitly.
 	 */
-	if (dev_priv->enable_hotplug_processing)
-		intel_hpd_init(dev_priv->dev);
+	if (dev_priv->power_domains.initializing)
+		return;
+
+	intel_hpd_init(dev_priv->dev);
 
 	i915_redisable_vga_power_on(dev_priv->dev);
 }
@@ -5603,23 +5923,12 @@ static void vlv_display_power_well_enable(struct drm_i915_private *dev_priv,
 static void vlv_display_power_well_disable(struct drm_i915_private *dev_priv,
 					   struct i915_power_well *power_well)
 {
-	struct drm_device *dev = dev_priv->dev;
-	enum pipe pipe;
-
 	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DISP2D);
 
 	spin_lock_irq(&dev_priv->irq_lock);
-	for_each_pipe(pipe)
-		__intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
-
 	valleyview_disable_display_irqs(dev_priv);
 	spin_unlock_irq(&dev_priv->irq_lock);
 
-	spin_lock_irq(&dev->vbl_lock);
-	for_each_pipe(pipe)
-		reset_vblank_counter(dev, pipe);
-	spin_unlock_irq(&dev->vbl_lock);
-
 	vlv_set_power_well(dev_priv, power_well, false);
 }
 
@@ -5663,6 +5972,7 @@ void intel_display_power_get(struct drm_i915_private *dev_priv,
 		if (!power_well->count++) {
 			DRM_DEBUG_KMS("enabling %s\n", power_well->name);
 			power_well->ops->enable(dev_priv, power_well);
+			power_well->hw_enabled = true;
 		}
 
 		check_power_well_state(dev_priv, power_well);
@@ -5692,6 +6002,7 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 
 		if (!--power_well->count && i915.disable_power_well) {
 			DRM_DEBUG_KMS("disabling %s\n", power_well->name);
+			power_well->hw_enabled = false;
 			power_well->ops->disable(dev_priv, power_well);
 		}
 
@@ -5706,33 +6017,56 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 static struct i915_power_domains *hsw_pwr;
 
 /* Display audio driver power well request */
-void i915_request_power_well(void)
+int i915_request_power_well(void)
 {
 	struct drm_i915_private *dev_priv;
 
-	if (WARN_ON(!hsw_pwr))
-		return;
+	if (!hsw_pwr)
+		return -ENODEV;
 
 	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
 				power_domains);
 	intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(i915_request_power_well);
 
 /* Display audio driver power well release */
-void i915_release_power_well(void)
+int i915_release_power_well(void)
 {
 	struct drm_i915_private *dev_priv;
 
-	if (WARN_ON(!hsw_pwr))
-		return;
+	if (!hsw_pwr)
+		return -ENODEV;
 
 	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
 				power_domains);
 	intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(i915_release_power_well);
 
+/*
+ * Private interface for the audio driver to get CDCLK in kHz.
+ *
+ * Caller must request power well using i915_request_power_well() prior to
+ * making the call.
+ */
+int i915_get_cdclk_freq(void)
+{
+	struct drm_i915_private *dev_priv;
+
+	if (!hsw_pwr)
+		return -ENODEV;
+
+	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
+				power_domains);
+
+	return intel_ddi_get_cdclk_freq(dev_priv);
+}
+EXPORT_SYMBOL_GPL(i915_get_cdclk_freq);
+
+
 #define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)
 
 #define HSW_ALWAYS_ON_POWER_DOMAINS (			\
@@ -5867,12 +6201,6 @@ static struct i915_power_well vlv_power_wells[] = {
 		.ops = &vlv_display_power_well_ops,
 	},
 	{
-		.name = "dpio-common",
-		.domains = VLV_DPIO_CMN_BC_POWER_DOMAINS,
-		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
-		.ops = &vlv_dpio_power_well_ops,
-	},
-	{
 		.name = "dpio-tx-b-01",
 		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
 			   VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS |
@@ -5908,6 +6236,12 @@ static struct i915_power_well vlv_power_wells[] = {
 		.ops = &vlv_dpio_power_well_ops,
 		.data = PUNIT_POWER_WELL_DPIO_TX_C_LANES_23,
 	},
+	{
+		.name = "dpio-common",
+		.domains = VLV_DPIO_CMN_BC_POWER_DOMAINS,
+		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
+		.ops = &vlv_dpio_power_well_ops,
+	},
 };
 
 #define set_power_wells(power_domains, __power_wells) ({		\
@@ -5952,16 +6286,23 @@ static void intel_power_domains_resume(struct drm_i915_private *dev_priv)
 	int i;
 
 	mutex_lock(&power_domains->lock);
-	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains)
+	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains) {
 		power_well->ops->sync_hw(dev_priv, power_well);
+		power_well->hw_enabled = power_well->ops->is_enabled(dev_priv,
+								     power_well);
+	}
 	mutex_unlock(&power_domains->lock);
 }
 
 void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
 {
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+
+	power_domains->initializing = true;
 	/* For now, we need the power well to be always enabled. */
 	intel_display_set_init_power(dev_priv, true);
 	intel_power_domains_resume(dev_priv);
+	power_domains->initializing = false;
 }
 
 void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv)
@@ -5986,6 +6327,18 @@ void intel_runtime_pm_get(struct drm_i915_private *dev_priv)
 	WARN(dev_priv->pm.suspended, "Device still suspended.\n");
 }
 
+void intel_runtime_pm_get_noresume(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct device *device = &dev->pdev->dev;
+
+	if (!HAS_RUNTIME_PM(dev))
+		return;
+
+	WARN(dev_priv->pm.suspended, "Getting nosync-ref while suspended.\n");
+	pm_runtime_get_noresume(device);
+}
+
 void intel_runtime_pm_put(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
@@ -6008,6 +6361,15 @@ void intel_init_runtime_pm(struct drm_i915_private *dev_priv)
 
 	pm_runtime_set_active(device);
 
+	/*
+	 * RPM depends on RC6 to save restore the GT HW context, so make RC6 a
+	 * requirement.
+	 */
+	if (!intel_enable_rc6(dev)) {
+		DRM_INFO("RC6 disabled, disabling runtime PM support\n");
+		return;
+	}
+
 	pm_runtime_set_autosuspend_delay(device, 10000); /* 10s */
 	pm_runtime_mark_last_busy(device);
 	pm_runtime_use_autosuspend(device);
@@ -6023,6 +6385,9 @@ void intel_fini_runtime_pm(struct drm_i915_private *dev_priv)
 	if (!HAS_RUNTIME_PM(dev))
 		return;
 
+	if (!intel_enable_rc6(dev))
+		return;
+
 	/* Make sure we're not suspended first. */
 	pm_runtime_get_sync(device);
 	pm_runtime_disable(device);
@@ -6087,6 +6452,10 @@ void intel_init_pm(struct drm_device *dev)
 			dev_priv->display.init_clock_gating = haswell_init_clock_gating;
 		else if (INTEL_INFO(dev)->gen == 8)
 			dev_priv->display.init_clock_gating = gen8_init_clock_gating;
+	} else if (IS_CHERRYVIEW(dev)) {
+		dev_priv->display.update_wm = valleyview_update_wm;
+		dev_priv->display.init_clock_gating =
+			cherryview_init_clock_gating;
 	} else if (IS_VALLEYVIEW(dev)) {
 		dev_priv->display.update_wm = valleyview_update_wm;
 		dev_priv->display.init_clock_gating =
diff --git a/drivers/gpu/drm/i915/intel_renderstate.h b/drivers/gpu/drm/i915/intel_renderstate.h
new file mode 100644
index 000000000000..a5e783a9928a
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate.h
@@ -0,0 +1,48 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _INTEL_RENDERSTATE_H
+#define _INTEL_RENDERSTATE_H
+
+#include <linux/types.h>
+
+struct intel_renderstate_rodata {
+	const u32 *reloc;
+	const u32 reloc_items;
+	const u32 *batch;
+	const u32 batch_items;
+};
+
+extern const struct intel_renderstate_rodata gen6_null_state;
+extern const struct intel_renderstate_rodata gen7_null_state;
+extern const struct intel_renderstate_rodata gen8_null_state;
+
+#define RO_RENDERSTATE(_g)						\
+	const struct intel_renderstate_rodata gen ## _g ## _null_state = { \
+		.reloc = gen ## _g ## _null_state_relocs,		\
+		.reloc_items = sizeof(gen ## _g ## _null_state_relocs)/4, \
+		.batch = gen ## _g ## _null_state_batch,		\
+		.batch_items = sizeof(gen ## _g ## _null_state_batch)/4, \
+	}
+
+#endif /* INTEL_RENDERSTATE_H */
diff --git a/drivers/gpu/drm/i915/intel_renderstate_gen6.c b/drivers/gpu/drm/i915/intel_renderstate_gen6.c
new file mode 100644
index 000000000000..740538ad0977
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate_gen6.c
@@ -0,0 +1,289 @@
+#include "intel_renderstate.h"
+
+static const u32 gen6_null_state_relocs[] = {
+	0x00000020,
+	0x00000024,
+	0x0000002c,
+	0x000001e0,
+	0x000001e4,
+};
+
+static const u32 gen6_null_state_batch[] = {
+	0x69040000,
+	0x790d0001,
+	0x00000000,
+	0x00000000,
+	0x78180000,
+	0x00000001,
+	0x61010008,
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000001,
+	0x61020000,
+	0x00000000,
+	0x78050001,
+	0x00000018,
+	0x00000000,
+	0x780d1002,
+	0x00000000,
+	0x00000000,
+	0x00000420,
+	0x78150003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78100004,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78160003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78110005,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78120002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78170003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x79050005,
+	0xe0040000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x79100000,
+	0x00000000,
+	0x79000002,
+	0xffffffff,
+	0x00000000,
+	0x00000000,
+	0x780e0002,
+	0x00000441,
+	0x00000401,
+	0x00000401,
+	0x78021002,
+	0x00000000,
+	0x00000000,
+	0x00000400,
+	0x78130012,
+	0x00400810,
+	0x00000000,
+	0x20000000,
+	0x04000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78140007,
+	0x00000280,
+	0x08080000,
+	0x00000000,
+	0x00060000,
+	0x4e080002,
+	0x00100400,
+	0x00000000,
+	0x00000000,
+	0x78090005,
+	0x02000000,
+	0x22220000,
+	0x02f60000,
+	0x11330000,
+	0x02850004,
+	0x11220000,
+	0x78011002,
+	0x00000000,
+	0x00000000,
+	0x00000200,
+	0x78080003,
+	0x00002000,
+	0x00000448,	 /* reloc */
+	0x00000448,	 /* reloc */
+	0x00000000,
+	0x05000000,	 /* cmds end */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000220,	 /* state start */
+	0x00000240,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0060005a,
+	0x204077be,
+	0x000000c0,
+	0x008d0040,
+	0x0060005a,
+	0x206077be,
+	0x000000c0,
+	0x008d0080,
+	0x0060005a,
+	0x208077be,
+	0x000000d0,
+	0x008d0040,
+	0x0060005a,
+	0x20a077be,
+	0x000000d0,
+	0x008d0080,
+	0x00000201,
+	0x20080061,
+	0x00000000,
+	0x00000000,
+	0x00600001,
+	0x20200022,
+	0x008d0000,
+	0x00000000,
+	0x02800031,
+	0x21c01cc9,
+	0x00000020,
+	0x0a8a0001,
+	0x00600001,
+	0x204003be,
+	0x008d01c0,
+	0x00000000,
+	0x00600001,
+	0x206003be,
+	0x008d01e0,
+	0x00000000,
+	0x00600001,
+	0x208003be,
+	0x008d0200,
+	0x00000000,
+	0x00600001,
+	0x20a003be,
+	0x008d0220,
+	0x00000000,
+	0x00600001,
+	0x20c003be,
+	0x008d0240,
+	0x00000000,
+	0x00600001,
+	0x20e003be,
+	0x008d0260,
+	0x00000000,
+	0x00600001,
+	0x210003be,
+	0x008d0280,
+	0x00000000,
+	0x00600001,
+	0x212003be,
+	0x008d02a0,
+	0x00000000,
+	0x05800031,
+	0x24001cc8,
+	0x00000040,
+	0x90019000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x30000000,
+	0x00000124,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xf99a130c,
+	0x799a130c,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x80000031,
+	0x00000003,
+	0x00000000,	 /* state end */
+};
+
+RO_RENDERSTATE(6);
diff --git a/drivers/gpu/drm/i915/intel_renderstate_gen7.c b/drivers/gpu/drm/i915/intel_renderstate_gen7.c
new file mode 100644
index 000000000000..6fa7ff2a1298
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate_gen7.c
@@ -0,0 +1,253 @@
+#include "intel_renderstate.h"
+
+static const u32 gen7_null_state_relocs[] = {
+	0x0000000c,
+	0x00000010,
+	0x00000018,
+	0x000001ec,
+};
+
+static const u32 gen7_null_state_batch[] = {
+	0x69040000,
+	0x61010008,
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000001,
+	0x790d0002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78180000,
+	0x00000001,
+	0x79160000,
+	0x00000008,
+	0x78300000,
+	0x02010040,
+	0x78310000,
+	0x04000000,
+	0x78320000,
+	0x04000000,
+	0x78330000,
+	0x02000000,
+	0x78100004,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781b0005,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781c0002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781d0004,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78110005,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78120002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78210000,
+	0x00000000,
+	0x78130005,
+	0x00000000,
+	0x20000000,
+	0x04000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78140001,
+	0x20000800,
+	0x00000000,
+	0x781e0001,
+	0x00000000,
+	0x00000000,
+	0x78050005,
+	0xe0040000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78040001,
+	0x00000000,
+	0x00000000,
+	0x78240000,
+	0x00000240,
+	0x78230000,
+	0x00000260,
+	0x782f0000,
+	0x00000280,
+	0x781f000c,
+	0x00400810,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78200006,
+	0x000002c0,
+	0x08080000,
+	0x00000000,
+	0x28000402,
+	0x00060000,
+	0x00000000,
+	0x00000000,
+	0x78090005,
+	0x02000000,
+	0x22220000,
+	0x02f60000,
+	0x11230000,
+	0x02f60004,
+	0x11230000,
+	0x78080003,
+	0x00006008,
+	0x00000340,	 /* reloc */
+	0xffffffff,
+	0x00000000,
+	0x782a0000,
+	0x00000360,
+	0x79000002,
+	0xffffffff,
+	0x00000000,
+	0x00000000,
+	0x7b000005,
+	0x0000000f,
+	0x00000003,
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000000,
+	0x05000000,	 /* cmds end */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000031,	 /* state start */
+	0x00000003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xf99a130c,
+	0x799a130c,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000492,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0080005a,
+	0x2e2077bd,
+	0x000000c0,
+	0x008d0040,
+	0x0080005a,
+	0x2e6077bd,
+	0x000000d0,
+	0x008d0040,
+	0x02800031,
+	0x21801fa9,
+	0x008d0e20,
+	0x08840001,
+	0x00800001,
+	0x2e2003bd,
+	0x008d0180,
+	0x00000000,
+	0x00800001,
+	0x2e6003bd,
+	0x008d01c0,
+	0x00000000,
+	0x00800001,
+	0x2ea003bd,
+	0x008d0200,
+	0x00000000,
+	0x00800001,
+	0x2ee003bd,
+	0x008d0240,
+	0x00000000,
+	0x05800031,
+	0x20001fa8,
+	0x008d0e20,
+	0x90031000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000380,
+	0x000003a0,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,	 /* state end */
+};
+
+RO_RENDERSTATE(7);
diff --git a/drivers/gpu/drm/i915/intel_renderstate_gen8.c b/drivers/gpu/drm/i915/intel_renderstate_gen8.c
new file mode 100644
index 000000000000..5c875615d42a
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate_gen8.c
@@ -0,0 +1,479 @@
+#include "intel_renderstate.h"
+
+static const u32 gen8_null_state_relocs[] = {
+	0x00000048,
+	0x00000050,
+	0x00000060,
+	0x000003ec,
+};
+
+static const u32 gen8_null_state_batch[] = {
+	0x69040000,
+	0x61020001,
+	0x00000000,
+	0x00000000,
+	0x79120000,
+	0x00000000,
+	0x79130000,
+	0x00000000,
+	0x79140000,
+	0x00000000,
+	0x79150000,
+	0x00000000,
+	0x79160000,
+	0x00000000,
+	0x6101000e,
+	0x00000001,
+	0x00000000,
+	0x00000001,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0xfffff001,
+	0x00001001,
+	0xfffff001,
+	0x00001001,
+	0x78230000,
+	0x000006e0,
+	0x78210000,
+	0x00000700,
+	0x78300000,
+	0x08010040,
+	0x78330000,
+	0x08000000,
+	0x78310000,
+	0x08000000,
+	0x78320000,
+	0x08000000,
+	0x78240000,
+	0x00000641,
+	0x780e0000,
+	0x00000601,
+	0x780d0000,
+	0x00000000,
+	0x78180000,
+	0x00000001,
+	0x78520003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78190009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781b0007,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78270000,
+	0x00000000,
+	0x782c0000,
+	0x00000000,
+	0x781c0002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78160009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78110008,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78290000,
+	0x00000000,
+	0x782e0000,
+	0x00000000,
+	0x781a0009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781d0007,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78280000,
+	0x00000000,
+	0x782d0000,
+	0x00000000,
+	0x78260000,
+	0x00000000,
+	0x782b0000,
+	0x00000000,
+	0x78150009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78100007,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781e0003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78120002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781f0002,
+	0x30400820,
+	0x00000000,
+	0x00000000,
+	0x78510009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78500003,
+	0x00210000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78130002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x782a0000,
+	0x00000480,
+	0x782f0000,
+	0x00000540,
+	0x78140000,
+	0x00000800,
+	0x78170009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x7820000a,
+	0x00000580,
+	0x00000000,
+	0x08080000,
+	0x00000000,
+	0x00000000,
+	0x1f000002,
+	0x00060000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x784d0000,
+	0x40000000,
+	0x784f0000,
+	0x80000100,
+	0x780f0000,
+	0x00000740,
+	0x78050006,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78070003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78060003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78040001,
+	0x00000000,
+	0x00000001,
+	0x79000002,
+	0xffffffff,
+	0x00000000,
+	0x00000000,
+	0x78080003,
+	0x00006000,
+	0x000005e0,	 /* reloc */
+	0x00000000,
+	0x00000000,
+	0x78090005,
+	0x02000000,
+	0x22220000,
+	0x02f60000,
+	0x11230000,
+	0x02850004,
+	0x11230000,
+	0x784b0000,
+	0x0000000f,
+	0x78490001,
+	0x00000000,
+	0x00000000,
+	0x7b000005,
+	0x00000000,
+	0x00000003,
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000000,
+	0x05000000,	 /* cmds end */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x000004c0,	 /* state start */
+	0x00000500,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000092,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0060005a,
+	0x21403ae8,
+	0x3a0000c0,
+	0x008d0040,
+	0x0060005a,
+	0x21603ae8,
+	0x3a0000c0,
+	0x008d0080,
+	0x0060005a,
+	0x21803ae8,
+	0x3a0000d0,
+	0x008d0040,
+	0x0060005a,
+	0x21a03ae8,
+	0x3a0000d0,
+	0x008d0080,
+	0x02800031,
+	0x2e0022e8,
+	0x0e000140,
+	0x08840001,
+	0x05800031,
+	0x200022e0,
+	0x0e000e00,
+	0x90031000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xf99a130c,
+	0x799a130c,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x3f800000,
+	0x00000000,
+	0x3f800000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,	 /* state end */
+};
+
+RO_RENDERSTATE(8);
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 79fb4cc2137c..279488addf3f 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -33,26 +33,44 @@
 #include "i915_trace.h"
 #include "intel_drv.h"
 
-static inline int ring_space(struct intel_ring_buffer *ring)
+/* Early gen2 devices have a cacheline of just 32 bytes, using 64 is overkill,
+ * but keeps the logic simple. Indeed, the whole purpose of this macro is just
+ * to give some inclination as to some of the magic values used in the various
+ * workarounds!
+ */
+#define CACHELINE_BYTES 64
+
+static inline int __ring_space(int head, int tail, int size)
 {
-	int space = (ring->head & HEAD_ADDR) - (ring->tail + I915_RING_FREE_SPACE);
+	int space = head - (tail + I915_RING_FREE_SPACE);
 	if (space < 0)
-		space += ring->size;
+		space += size;
 	return space;
 }
 
-void __intel_ring_advance(struct intel_ring_buffer *ring)
+static inline int ring_space(struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	return __ring_space(ringbuf->head & HEAD_ADDR, ringbuf->tail, ringbuf->size);
+}
+
+static bool intel_ring_stopped(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	return dev_priv->gpu_error.stop_rings & intel_ring_flag(ring);
+}
 
-	ring->tail &= ring->size - 1;
-	if (dev_priv->gpu_error.stop_rings & intel_ring_flag(ring))
+void __intel_ring_advance(struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	ringbuf->tail &= ringbuf->size - 1;
+	if (intel_ring_stopped(ring))
 		return;
-	ring->write_tail(ring, ring->tail);
+	ring->write_tail(ring, ringbuf->tail);
 }
 
 static int
-gen2_render_ring_flush(struct intel_ring_buffer *ring,
+gen2_render_ring_flush(struct intel_engine_cs *ring,
 		       u32	invalidate_domains,
 		       u32	flush_domains)
 {
@@ -78,7 +96,7 @@ gen2_render_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen4_render_ring_flush(struct intel_ring_buffer *ring,
+gen4_render_ring_flush(struct intel_engine_cs *ring,
 		       u32	invalidate_domains,
 		       u32	flush_domains)
 {
@@ -173,9 +191,9 @@ gen4_render_ring_flush(struct intel_ring_buffer *ring,
  * really our business.  That leaves only stall at scoreboard.
  */
 static int
-intel_emit_post_sync_nonzero_flush(struct intel_ring_buffer *ring)
+intel_emit_post_sync_nonzero_flush(struct intel_engine_cs *ring)
 {
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 
@@ -208,11 +226,11 @@ intel_emit_post_sync_nonzero_flush(struct intel_ring_buffer *ring)
 }
 
 static int
-gen6_render_ring_flush(struct intel_ring_buffer *ring,
+gen6_render_ring_flush(struct intel_engine_cs *ring,
                          u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* Force SNB workarounds for PIPE_CONTROL flushes */
@@ -260,7 +278,7 @@ gen6_render_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen7_render_ring_cs_stall_wa(struct intel_ring_buffer *ring)
+gen7_render_ring_cs_stall_wa(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -278,7 +296,7 @@ gen7_render_ring_cs_stall_wa(struct intel_ring_buffer *ring)
 	return 0;
 }
 
-static int gen7_ring_fbc_flush(struct intel_ring_buffer *ring, u32 value)
+static int gen7_ring_fbc_flush(struct intel_engine_cs *ring, u32 value)
 {
 	int ret;
 
@@ -302,11 +320,11 @@ static int gen7_ring_fbc_flush(struct intel_ring_buffer *ring, u32 value)
 }
 
 static int
-gen7_render_ring_flush(struct intel_ring_buffer *ring,
+gen7_render_ring_flush(struct intel_engine_cs *ring,
 		       u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/*
@@ -363,11 +381,11 @@ gen7_render_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen8_render_ring_flush(struct intel_ring_buffer *ring,
+gen8_render_ring_flush(struct intel_engine_cs *ring,
 		       u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	flags |= PIPE_CONTROL_CS_STALL;
@@ -403,14 +421,14 @@ gen8_render_ring_flush(struct intel_ring_buffer *ring,
 
 }
 
-static void ring_write_tail(struct intel_ring_buffer *ring,
+static void ring_write_tail(struct intel_engine_cs *ring,
 			    u32 value)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	I915_WRITE_TAIL(ring, value);
 }
 
-u64 intel_ring_get_active_head(struct intel_ring_buffer *ring)
+u64 intel_ring_get_active_head(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u64 acthd;
@@ -426,7 +444,7 @@ u64 intel_ring_get_active_head(struct intel_ring_buffer *ring)
 	return acthd;
 }
 
-static void ring_setup_phys_status_page(struct intel_ring_buffer *ring)
+static void ring_setup_phys_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 addr;
@@ -437,7 +455,7 @@ static void ring_setup_phys_status_page(struct intel_ring_buffer *ring)
 	I915_WRITE(HWS_PGA, addr);
 }
 
-static bool stop_ring(struct intel_ring_buffer *ring)
+static bool stop_ring(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = to_i915(ring->dev);
 
@@ -461,11 +479,12 @@ static bool stop_ring(struct intel_ring_buffer *ring)
 	return (I915_READ_HEAD(ring) & HEAD_ADDR) == 0;
 }
 
-static int init_ring_common(struct intel_ring_buffer *ring)
+static int init_ring_common(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_object *obj = ring->obj;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct drm_i915_gem_object *obj = ringbuf->obj;
 	int ret = 0;
 
 	gen6_gt_force_wake_get(dev_priv, FORCEWAKE_ALL);
@@ -504,7 +523,7 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 	 * register values. */
 	I915_WRITE_START(ring, i915_gem_obj_ggtt_offset(obj));
 	I915_WRITE_CTL(ring,
-			((ring->size - PAGE_SIZE) & RING_NR_PAGES)
+			((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES)
 			| RING_VALID);
 
 	/* If the head is still not zero, the ring is dead */
@@ -512,12 +531,11 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 		     I915_READ_START(ring) == i915_gem_obj_ggtt_offset(obj) &&
 		     (I915_READ_HEAD(ring) & HEAD_ADDR) == 0, 50)) {
 		DRM_ERROR("%s initialization failed "
-				"ctl %08x head %08x tail %08x start %08x\n",
-				ring->name,
-				I915_READ_CTL(ring),
-				I915_READ_HEAD(ring),
-				I915_READ_TAIL(ring),
-				I915_READ_START(ring));
+			  "ctl %08x (valid? %d) head %08x tail %08x start %08x [expected %08lx]\n",
+			  ring->name,
+			  I915_READ_CTL(ring), I915_READ_CTL(ring) & RING_VALID,
+			  I915_READ_HEAD(ring), I915_READ_TAIL(ring),
+			  I915_READ_START(ring), (unsigned long)i915_gem_obj_ggtt_offset(obj));
 		ret = -EIO;
 		goto out;
 	}
@@ -525,10 +543,10 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 	if (!drm_core_check_feature(ring->dev, DRIVER_MODESET))
 		i915_kernel_lost_context(ring->dev);
 	else {
-		ring->head = I915_READ_HEAD(ring);
-		ring->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
-		ring->space = ring_space(ring);
-		ring->last_retired_head = -1;
+		ringbuf->head = I915_READ_HEAD(ring);
+		ringbuf->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
+		ringbuf->space = ring_space(ring);
+		ringbuf->last_retired_head = -1;
 	}
 
 	memset(&ring->hangcheck, 0, sizeof(ring->hangcheck));
@@ -540,7 +558,7 @@ out:
 }
 
 static int
-init_pipe_control(struct intel_ring_buffer *ring)
+init_pipe_control(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -581,7 +599,7 @@ err:
 	return ret;
 }
 
-static int init_render_ring(struct intel_ring_buffer *ring)
+static int init_render_ring(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -595,19 +613,21 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 	 * to use MI_WAIT_FOR_EVENT within the CS. It should already be
 	 * programmed to '1' on all products.
 	 *
-	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv,bdw
+	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv,bdw,chv
 	 */
 	if (INTEL_INFO(dev)->gen >= 6)
 		I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(ASYNC_FLIP_PERF_DISABLE));
 
 	/* Required for the hardware to program scanline values for waiting */
+	/* WaEnableFlushTlbInvalidationMode:snb */
 	if (INTEL_INFO(dev)->gen == 6)
 		I915_WRITE(GFX_MODE,
-			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_ALWAYS));
+			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_EXPLICIT));
 
+	/* WaBCSVCSTlbInvalidationMode:ivb,vlv,hsw */
 	if (IS_GEN7(dev))
 		I915_WRITE(GFX_MODE_GEN7,
-			   _MASKED_BIT_DISABLE(GFX_TLB_INVALIDATE_ALWAYS) |
+			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_EXPLICIT) |
 			   _MASKED_BIT_ENABLE(GFX_REPLAY_MODE));
 
 	if (INTEL_INFO(dev)->gen >= 5) {
@@ -624,13 +644,6 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 		 */
 		I915_WRITE(CACHE_MODE_0,
 			   _MASKED_BIT_DISABLE(CM0_STC_EVICT_DISABLE_LRA_SNB));
-
-		/* This is not explicitly set for GEN6, so read the register.
-		 * see intel_ring_mi_set_context() for why we care.
-		 * TODO: consider explicitly setting the bit for GEN5
-		 */
-		ring->itlb_before_ctx_switch =
-			!!(I915_READ(GFX_MODE) & GFX_TLB_INVALIDATE_ALWAYS);
 	}
 
 	if (INTEL_INFO(dev)->gen >= 6)
@@ -642,7 +655,7 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 	return ret;
 }
 
-static void render_ring_cleanup(struct intel_ring_buffer *ring)
+static void render_ring_cleanup(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 
@@ -658,20 +671,46 @@ static void render_ring_cleanup(struct intel_ring_buffer *ring)
 	ring->scratch.obj = NULL;
 }
 
-static void
-update_mboxes(struct intel_ring_buffer *ring,
-	      u32 mmio_offset)
+static int gen6_signal(struct intel_engine_cs *signaller,
+		       unsigned int num_dwords)
 {
-/* NB: In order to be able to do semaphore MBOX updates for varying number
- * of rings, it's easiest if we round up each individual update to a
- * multiple of 2 (since ring updates must always be a multiple of 2)
- * even though the actual update only requires 3 dwords.
- */
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *useless;
+	int i, ret;
+
+	/* NB: In order to be able to do semaphore MBOX updates for varying
+	 * number of rings, it's easiest if we round up each individual update
+	 * to a multiple of 2 (since ring updates must always be a multiple of
+	 * 2) even though the actual update only requires 3 dwords.
+	 */
 #define MBOX_UPDATE_DWORDS 4
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-	intel_ring_emit(ring, mmio_offset);
-	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
-	intel_ring_emit(ring, MI_NOOP);
+	if (i915_semaphore_is_enabled(dev))
+		num_dwords += ((I915_NUM_RINGS-1) * MBOX_UPDATE_DWORDS);
+	else
+		return intel_ring_begin(signaller, num_dwords);
+
+	ret = intel_ring_begin(signaller, num_dwords);
+	if (ret)
+		return ret;
+#undef MBOX_UPDATE_DWORDS
+
+	for_each_ring(useless, dev_priv, i) {
+		u32 mbox_reg = signaller->semaphore.mbox.signal[i];
+		if (mbox_reg != GEN6_NOSYNC) {
+			intel_ring_emit(signaller, MI_LOAD_REGISTER_IMM(1));
+			intel_ring_emit(signaller, mbox_reg);
+			intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+			intel_ring_emit(signaller, MI_NOOP);
+		} else {
+			intel_ring_emit(signaller, MI_NOOP);
+			intel_ring_emit(signaller, MI_NOOP);
+			intel_ring_emit(signaller, MI_NOOP);
+			intel_ring_emit(signaller, MI_NOOP);
+		}
+	}
+
+	return 0;
 }
 
 /**
@@ -684,29 +723,14 @@ update_mboxes(struct intel_ring_buffer *ring,
  * This acts like a signal in the canonical semaphore.
  */
 static int
-gen6_add_request(struct intel_ring_buffer *ring)
+gen6_add_request(struct intel_engine_cs *ring)
 {
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *useless;
-	int i, ret, num_dwords = 4;
-
-	if (i915_semaphore_is_enabled(dev))
-		num_dwords += ((I915_NUM_RINGS-1) * MBOX_UPDATE_DWORDS);
-#undef MBOX_UPDATE_DWORDS
+	int ret;
 
-	ret = intel_ring_begin(ring, num_dwords);
+	ret = ring->semaphore.signal(ring, 4);
 	if (ret)
 		return ret;
 
-	if (i915_semaphore_is_enabled(dev)) {
-		for_each_ring(useless, dev_priv, i) {
-			u32 mbox_reg = ring->signal_mbox[i];
-			if (mbox_reg != GEN6_NOSYNC)
-				update_mboxes(ring, mbox_reg);
-		}
-	}
-
 	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
 	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
 	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
@@ -731,14 +755,15 @@ static inline bool i915_gem_has_seqno_wrapped(struct drm_device *dev,
  * @seqno - seqno which the waiter will block on
  */
 static int
-gen6_ring_sync(struct intel_ring_buffer *waiter,
-	       struct intel_ring_buffer *signaller,
+gen6_ring_sync(struct intel_engine_cs *waiter,
+	       struct intel_engine_cs *signaller,
 	       u32 seqno)
 {
-	int ret;
 	u32 dw1 = MI_SEMAPHORE_MBOX |
 		  MI_SEMAPHORE_COMPARE |
 		  MI_SEMAPHORE_REGISTER;
+	u32 wait_mbox = signaller->semaphore.mbox.wait[waiter->id];
+	int ret;
 
 	/* Throughout all of the GEM code, seqno passed implies our current
 	 * seqno is >= the last seqno executed. However for hardware the
@@ -746,8 +771,7 @@ gen6_ring_sync(struct intel_ring_buffer *waiter,
 	 */
 	seqno -= 1;
 
-	WARN_ON(signaller->semaphore_register[waiter->id] ==
-		MI_SEMAPHORE_SYNC_INVALID);
+	WARN_ON(wait_mbox == MI_SEMAPHORE_SYNC_INVALID);
 
 	ret = intel_ring_begin(waiter, 4);
 	if (ret)
@@ -755,9 +779,7 @@ gen6_ring_sync(struct intel_ring_buffer *waiter,
 
 	/* If seqno wrap happened, omit the wait with no-ops */
 	if (likely(!i915_gem_has_seqno_wrapped(waiter->dev, seqno))) {
-		intel_ring_emit(waiter,
-				dw1 |
-				signaller->semaphore_register[waiter->id]);
+		intel_ring_emit(waiter, dw1 | wait_mbox);
 		intel_ring_emit(waiter, seqno);
 		intel_ring_emit(waiter, 0);
 		intel_ring_emit(waiter, MI_NOOP);
@@ -782,9 +804,9 @@ do {									\
 } while (0)
 
 static int
-pc_render_add_request(struct intel_ring_buffer *ring)
+pc_render_add_request(struct intel_engine_cs *ring)
 {
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* For Ironlake, MI_USER_INTERRUPT was deprecated and apparently
@@ -806,15 +828,15 @@ pc_render_add_request(struct intel_ring_buffer *ring)
 	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, 0);
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128; /* write to separate cachelines */
+	scratch_addr += 2 * CACHELINE_BYTES; /* write to separate cachelines */
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
 
 	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
@@ -830,7 +852,7 @@ pc_render_add_request(struct intel_ring_buffer *ring)
 }
 
 static u32
-gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+gen6_ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 {
 	/* Workaround to force correct ordering between irq and seqno writes on
 	 * ivb (and maybe also on snb) by reading from a CS register (like
@@ -844,31 +866,31 @@ gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
 }
 
 static u32
-ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 {
 	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
 }
 
 static void
-ring_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+ring_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
 }
 
 static u32
-pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+pc_render_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 {
 	return ring->scratch.cpu_page[0];
 }
 
 static void
-pc_render_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+pc_render_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	ring->scratch.cpu_page[0] = seqno;
 }
 
 static bool
-gen5_ring_get_irq(struct intel_ring_buffer *ring)
+gen5_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -886,7 +908,7 @@ gen5_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-gen5_ring_put_irq(struct intel_ring_buffer *ring)
+gen5_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -899,7 +921,7 @@ gen5_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-i9xx_ring_get_irq(struct intel_ring_buffer *ring)
+i9xx_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -920,7 +942,7 @@ i9xx_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-i9xx_ring_put_irq(struct intel_ring_buffer *ring)
+i9xx_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -936,7 +958,7 @@ i9xx_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-i8xx_ring_get_irq(struct intel_ring_buffer *ring)
+i8xx_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -957,7 +979,7 @@ i8xx_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-i8xx_ring_put_irq(struct intel_ring_buffer *ring)
+i8xx_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -972,7 +994,7 @@ i8xx_ring_put_irq(struct intel_ring_buffer *ring)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
 
-void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
+void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -989,6 +1011,11 @@ void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
 		case BCS:
 			mmio = BLT_HWS_PGA_GEN7;
 			break;
+		/*
+		 * VCS2 actually doesn't exist on Gen7. Only shut up
+		 * gcc switch check warning
+		 */
+		case VCS2:
 		case VCS:
 			mmio = BSD_HWS_PGA_GEN7;
 			break;
@@ -1030,7 +1057,7 @@ void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
 }
 
 static int
-bsd_ring_flush(struct intel_ring_buffer *ring,
+bsd_ring_flush(struct intel_engine_cs *ring,
 	       u32     invalidate_domains,
 	       u32     flush_domains)
 {
@@ -1047,7 +1074,7 @@ bsd_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-i9xx_add_request(struct intel_ring_buffer *ring)
+i9xx_add_request(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -1065,7 +1092,7 @@ i9xx_add_request(struct intel_ring_buffer *ring)
 }
 
 static bool
-gen6_ring_get_irq(struct intel_ring_buffer *ring)
+gen6_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1090,7 +1117,7 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-gen6_ring_put_irq(struct intel_ring_buffer *ring)
+gen6_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1108,7 +1135,7 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-hsw_vebox_get_irq(struct intel_ring_buffer *ring)
+hsw_vebox_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1128,7 +1155,7 @@ hsw_vebox_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-hsw_vebox_put_irq(struct intel_ring_buffer *ring)
+hsw_vebox_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1146,7 +1173,7 @@ hsw_vebox_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-gen8_ring_get_irq(struct intel_ring_buffer *ring)
+gen8_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1172,7 +1199,7 @@ gen8_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-gen8_ring_put_irq(struct intel_ring_buffer *ring)
+gen8_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1192,8 +1219,8 @@ gen8_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static int
-i965_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			 u32 offset, u32 length,
+i965_dispatch_execbuffer(struct intel_engine_cs *ring,
+			 u64 offset, u32 length,
 			 unsigned flags)
 {
 	int ret;
@@ -1215,8 +1242,8 @@ i965_dispatch_execbuffer(struct intel_ring_buffer *ring,
 /* Just userspace ABI convention to limit the wa batch bo to a resonable size */
 #define I830_BATCH_LIMIT (256*1024)
 static int
-i830_dispatch_execbuffer(struct intel_ring_buffer *ring,
-				u32 offset, u32 len,
+i830_dispatch_execbuffer(struct intel_engine_cs *ring,
+				u64 offset, u32 len,
 				unsigned flags)
 {
 	int ret;
@@ -1266,8 +1293,8 @@ i830_dispatch_execbuffer(struct intel_ring_buffer *ring,
 }
 
 static int
-i915_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			 u32 offset, u32 len,
+i915_dispatch_execbuffer(struct intel_engine_cs *ring,
+			 u64 offset, u32 len,
 			 unsigned flags)
 {
 	int ret;
@@ -1283,7 +1310,7 @@ i915_dispatch_execbuffer(struct intel_ring_buffer *ring,
 	return 0;
 }
 
-static void cleanup_status_page(struct intel_ring_buffer *ring)
+static void cleanup_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_object *obj;
 
@@ -1297,50 +1324,44 @@ static void cleanup_status_page(struct intel_ring_buffer *ring)
 	ring->status_page.obj = NULL;
 }
 
-static int init_status_page(struct intel_ring_buffer *ring)
+static int init_status_page(struct intel_engine_cs *ring)
 {
-	struct drm_device *dev = ring->dev;
 	struct drm_i915_gem_object *obj;
-	int ret;
 
-	obj = i915_gem_alloc_object(dev, 4096);
-	if (obj == NULL) {
-		DRM_ERROR("Failed to allocate status page\n");
-		ret = -ENOMEM;
-		goto err;
-	}
+	if ((obj = ring->status_page.obj) == NULL) {
+		int ret;
 
-	ret = i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
-	if (ret)
-		goto err_unref;
+		obj = i915_gem_alloc_object(ring->dev, 4096);
+		if (obj == NULL) {
+			DRM_ERROR("Failed to allocate status page\n");
+			return -ENOMEM;
+		}
 
-	ret = i915_gem_obj_ggtt_pin(obj, 4096, 0);
-	if (ret)
-		goto err_unref;
+		ret = i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+		if (ret)
+			goto err_unref;
+
+		ret = i915_gem_obj_ggtt_pin(obj, 4096, 0);
+		if (ret) {
+err_unref:
+			drm_gem_object_unreference(&obj->base);
+			return ret;
+		}
+
+		ring->status_page.obj = obj;
+	}
 
 	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(obj);
 	ring->status_page.page_addr = kmap(sg_page(obj->pages->sgl));
-	if (ring->status_page.page_addr == NULL) {
-		ret = -ENOMEM;
-		goto err_unpin;
-	}
-	ring->status_page.obj = obj;
 	memset(ring->status_page.page_addr, 0, PAGE_SIZE);
 
 	DRM_DEBUG_DRIVER("%s hws offset: 0x%08x\n",
 			ring->name, ring->status_page.gfx_addr);
 
 	return 0;
-
-err_unpin:
-	i915_gem_object_ggtt_unpin(obj);
-err_unref:
-	drm_gem_object_unreference(&obj->base);
-err:
-	return ret;
 }
 
-static int init_phys_status_page(struct intel_ring_buffer *ring)
+static int init_phys_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
@@ -1357,44 +1378,24 @@ static int init_phys_status_page(struct intel_ring_buffer *ring)
 	return 0;
 }
 
-static int intel_init_ring_buffer(struct drm_device *dev,
-				  struct intel_ring_buffer *ring)
+static int allocate_ring_buffer(struct intel_engine_cs *ring)
 {
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	struct drm_i915_gem_object *obj;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	ring->dev = dev;
-	INIT_LIST_HEAD(&ring->active_list);
-	INIT_LIST_HEAD(&ring->request_list);
-	ring->size = 32 * PAGE_SIZE;
-	memset(ring->sync_seqno, 0, sizeof(ring->sync_seqno));
-
-	init_waitqueue_head(&ring->irq_queue);
-
-	if (I915_NEED_GFX_HWS(dev)) {
-		ret = init_status_page(ring);
-		if (ret)
-			return ret;
-	} else {
-		BUG_ON(ring->id != RCS);
-		ret = init_phys_status_page(ring);
-		if (ret)
-			return ret;
-	}
+	if (intel_ring_initialized(ring))
+		return 0;
 
 	obj = NULL;
 	if (!HAS_LLC(dev))
-		obj = i915_gem_object_create_stolen(dev, ring->size);
+		obj = i915_gem_object_create_stolen(dev, ringbuf->size);
 	if (obj == NULL)
-		obj = i915_gem_alloc_object(dev, ring->size);
-	if (obj == NULL) {
-		DRM_ERROR("Failed to allocate ringbuffer\n");
-		ret = -ENOMEM;
-		goto err_hws;
-	}
-
-	ring->obj = obj;
+		obj = i915_gem_alloc_object(dev, ringbuf->size);
+	if (obj == NULL)
+		return -ENOMEM;
 
 	ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, PIN_MAPPABLE);
 	if (ret)
@@ -1404,65 +1405,102 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	if (ret)
 		goto err_unpin;
 
-	ring->virtual_start =
+	ringbuf->virtual_start =
 		ioremap_wc(dev_priv->gtt.mappable_base + i915_gem_obj_ggtt_offset(obj),
-			   ring->size);
-	if (ring->virtual_start == NULL) {
-		DRM_ERROR("Failed to map ringbuffer.\n");
+				ringbuf->size);
+	if (ringbuf->virtual_start == NULL) {
 		ret = -EINVAL;
 		goto err_unpin;
 	}
 
-	ret = ring->init(ring);
-	if (ret)
-		goto err_unmap;
+	ringbuf->obj = obj;
+	return 0;
+
+err_unpin:
+	i915_gem_object_ggtt_unpin(obj);
+err_unref:
+	drm_gem_object_unreference(&obj->base);
+	return ret;
+}
+
+static int intel_init_ring_buffer(struct drm_device *dev,
+				  struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	int ret;
+
+	if (ringbuf == NULL) {
+		ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
+		if (!ringbuf)
+			return -ENOMEM;
+		ring->buffer = ringbuf;
+	}
+
+	ring->dev = dev;
+	INIT_LIST_HEAD(&ring->active_list);
+	INIT_LIST_HEAD(&ring->request_list);
+	ringbuf->size = 32 * PAGE_SIZE;
+	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
+
+	init_waitqueue_head(&ring->irq_queue);
+
+	if (I915_NEED_GFX_HWS(dev)) {
+		ret = init_status_page(ring);
+		if (ret)
+			goto error;
+	} else {
+		BUG_ON(ring->id != RCS);
+		ret = init_phys_status_page(ring);
+		if (ret)
+			goto error;
+	}
+
+	ret = allocate_ring_buffer(ring);
+	if (ret) {
+		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n", ring->name, ret);
+		goto error;
+	}
 
 	/* Workaround an erratum on the i830 which causes a hang if
 	 * the TAIL pointer points to within the last 2 cachelines
 	 * of the buffer.
 	 */
-	ring->effective_size = ring->size;
-	if (IS_I830(ring->dev) || IS_845G(ring->dev))
-		ring->effective_size -= 128;
+	ringbuf->effective_size = ringbuf->size;
+	if (IS_I830(dev) || IS_845G(dev))
+		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
 
-	i915_cmd_parser_init_ring(ring);
+	ret = i915_cmd_parser_init_ring(ring);
+	if (ret)
+		goto error;
+
+	ret = ring->init(ring);
+	if (ret)
+		goto error;
 
 	return 0;
 
-err_unmap:
-	iounmap(ring->virtual_start);
-err_unpin:
-	i915_gem_object_ggtt_unpin(obj);
-err_unref:
-	drm_gem_object_unreference(&obj->base);
-	ring->obj = NULL;
-err_hws:
-	cleanup_status_page(ring);
+error:
+	kfree(ringbuf);
+	ring->buffer = NULL;
 	return ret;
 }
 
-void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
+void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 {
-	struct drm_i915_private *dev_priv;
-	int ret;
+	struct drm_i915_private *dev_priv = to_i915(ring->dev);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 
-	if (ring->obj == NULL)
+	if (!intel_ring_initialized(ring))
 		return;
 
-	/* Disable the ring buffer. The ring must be idle at this point */
-	dev_priv = ring->dev->dev_private;
-	ret = intel_ring_idle(ring);
-	if (ret && !i915_reset_in_progress(&dev_priv->gpu_error))
-		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  ring->name, ret);
-
-	I915_WRITE_CTL(ring, 0);
+	intel_stop_ring_buffer(ring);
+	WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
 
-	iounmap(ring->virtual_start);
+	iounmap(ringbuf->virtual_start);
 
-	i915_gem_object_ggtt_unpin(ring->obj);
-	drm_gem_object_unreference(&ring->obj->base);
-	ring->obj = NULL;
+	i915_gem_object_ggtt_unpin(ringbuf->obj);
+	drm_gem_object_unreference(&ringbuf->obj->base);
+	ringbuf->obj = NULL;
 	ring->preallocated_lazy_request = NULL;
 	ring->outstanding_lazy_seqno = 0;
 
@@ -1470,44 +1508,34 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
 		ring->cleanup(ring);
 
 	cleanup_status_page(ring);
+
+	i915_cmd_parser_fini_ring(ring);
+
+	kfree(ringbuf);
+	ring->buffer = NULL;
 }
 
-static int intel_ring_wait_request(struct intel_ring_buffer *ring, int n)
+static int intel_ring_wait_request(struct intel_engine_cs *ring, int n)
 {
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	struct drm_i915_gem_request *request;
-	u32 seqno = 0, tail;
+	u32 seqno = 0;
 	int ret;
 
-	if (ring->last_retired_head != -1) {
-		ring->head = ring->last_retired_head;
-		ring->last_retired_head = -1;
+	if (ringbuf->last_retired_head != -1) {
+		ringbuf->head = ringbuf->last_retired_head;
+		ringbuf->last_retired_head = -1;
 
-		ring->space = ring_space(ring);
-		if (ring->space >= n)
+		ringbuf->space = ring_space(ring);
+		if (ringbuf->space >= n)
 			return 0;
 	}
 
 	list_for_each_entry(request, &ring->request_list, list) {
-		int space;
-
-		if (request->tail == -1)
-			continue;
-
-		space = request->tail - (ring->tail + I915_RING_FREE_SPACE);
-		if (space < 0)
-			space += ring->size;
-		if (space >= n) {
+		if (__ring_space(request->tail, ringbuf->tail, ringbuf->size) >= n) {
 			seqno = request->seqno;
-			tail = request->tail;
 			break;
 		}
-
-		/* Consume this request in case we need more space than
-		 * is available and so need to prevent a race between
-		 * updating last_retired_head and direct reads of
-		 * I915_RING_HEAD. It also provides a nice sanity check.
-		 */
-		request->tail = -1;
 	}
 
 	if (seqno == 0)
@@ -1517,18 +1545,19 @@ static int intel_ring_wait_request(struct intel_ring_buffer *ring, int n)
 	if (ret)
 		return ret;
 
-	ring->head = tail;
-	ring->space = ring_space(ring);
-	if (WARN_ON(ring->space < n))
-		return -ENOSPC;
+	i915_gem_retire_requests_ring(ring);
+	ringbuf->head = ringbuf->last_retired_head;
+	ringbuf->last_retired_head = -1;
 
+	ringbuf->space = ring_space(ring);
 	return 0;
 }
 
-static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
+static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	unsigned long end;
 	int ret;
 
@@ -1539,7 +1568,6 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 	/* force the tail write in case we have been skipping them */
 	__intel_ring_advance(ring);
 
-	trace_i915_ring_wait_begin(ring);
 	/* With GEM the hangcheck timer should kick us out of the loop,
 	 * leaving it early runs the risk of corrupting GEM state (due
 	 * to running on almost untested codepaths). But on resume
@@ -1547,12 +1575,13 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 	 * case by choosing an insanely large timeout. */
 	end = jiffies + 60 * HZ;
 
+	trace_i915_ring_wait_begin(ring);
 	do {
-		ring->head = I915_READ_HEAD(ring);
-		ring->space = ring_space(ring);
-		if (ring->space >= n) {
-			trace_i915_ring_wait_end(ring);
-			return 0;
+		ringbuf->head = I915_READ_HEAD(ring);
+		ringbuf->space = ring_space(ring);
+		if (ringbuf->space >= n) {
+			ret = 0;
+			break;
 		}
 
 		if (!drm_core_check_feature(dev, DRIVER_MODESET) &&
@@ -1564,38 +1593,49 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 
 		msleep(1);
 
+		if (dev_priv->mm.interruptible && signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+
 		ret = i915_gem_check_wedge(&dev_priv->gpu_error,
 					   dev_priv->mm.interruptible);
 		if (ret)
-			return ret;
-	} while (!time_after(jiffies, end));
+			break;
+
+		if (time_after(jiffies, end)) {
+			ret = -EBUSY;
+			break;
+		}
+	} while (1);
 	trace_i915_ring_wait_end(ring);
-	return -EBUSY;
+	return ret;
 }
 
-static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring)
+static int intel_wrap_ring_buffer(struct intel_engine_cs *ring)
 {
 	uint32_t __iomem *virt;
-	int rem = ring->size - ring->tail;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	int rem = ringbuf->size - ringbuf->tail;
 
-	if (ring->space < rem) {
+	if (ringbuf->space < rem) {
 		int ret = ring_wait_for_space(ring, rem);
 		if (ret)
 			return ret;
 	}
 
-	virt = ring->virtual_start + ring->tail;
+	virt = ringbuf->virtual_start + ringbuf->tail;
 	rem /= 4;
 	while (rem--)
 		iowrite32(MI_NOOP, virt++);
 
-	ring->tail = 0;
-	ring->space = ring_space(ring);
+	ringbuf->tail = 0;
+	ringbuf->space = ring_space(ring);
 
 	return 0;
 }
 
-int intel_ring_idle(struct intel_ring_buffer *ring)
+int intel_ring_idle(struct intel_engine_cs *ring)
 {
 	u32 seqno;
 	int ret;
@@ -1619,7 +1659,7 @@ int intel_ring_idle(struct intel_ring_buffer *ring)
 }
 
 static int
-intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
+intel_ring_alloc_seqno(struct intel_engine_cs *ring)
 {
 	if (ring->outstanding_lazy_seqno)
 		return 0;
@@ -1637,18 +1677,19 @@ intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
 	return i915_gem_get_seqno(ring->dev, &ring->outstanding_lazy_seqno);
 }
 
-static int __intel_ring_prepare(struct intel_ring_buffer *ring,
+static int __intel_ring_prepare(struct intel_engine_cs *ring,
 				int bytes)
 {
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	int ret;
 
-	if (unlikely(ring->tail + bytes > ring->effective_size)) {
+	if (unlikely(ringbuf->tail + bytes > ringbuf->effective_size)) {
 		ret = intel_wrap_ring_buffer(ring);
 		if (unlikely(ret))
 			return ret;
 	}
 
-	if (unlikely(ring->space < bytes)) {
+	if (unlikely(ringbuf->space < bytes)) {
 		ret = ring_wait_for_space(ring, bytes);
 		if (unlikely(ret))
 			return ret;
@@ -1657,7 +1698,7 @@ static int __intel_ring_prepare(struct intel_ring_buffer *ring,
 	return 0;
 }
 
-int intel_ring_begin(struct intel_ring_buffer *ring,
+int intel_ring_begin(struct intel_engine_cs *ring,
 		     int num_dwords)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -1677,19 +1718,20 @@ int intel_ring_begin(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	ring->space -= num_dwords * sizeof(uint32_t);
+	ring->buffer->space -= num_dwords * sizeof(uint32_t);
 	return 0;
 }
 
 /* Align the ring tail to a cacheline boundary */
-int intel_ring_cacheline_align(struct intel_ring_buffer *ring)
+int intel_ring_cacheline_align(struct intel_engine_cs *ring)
 {
-	int num_dwords = (64 - (ring->tail & 63)) / sizeof(uint32_t);
+	int num_dwords = (ring->buffer->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
 	int ret;
 
 	if (num_dwords == 0)
 		return 0;
 
+	num_dwords = CACHELINE_BYTES / sizeof(uint32_t) - num_dwords;
 	ret = intel_ring_begin(ring, num_dwords);
 	if (ret)
 		return ret;
@@ -1702,7 +1744,7 @@ int intel_ring_cacheline_align(struct intel_ring_buffer *ring)
 	return 0;
 }
 
-void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
+void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
@@ -1719,7 +1761,7 @@ void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
 	ring->hangcheck.seqno = seqno;
 }
 
-static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
+static void gen6_bsd_ring_write_tail(struct intel_engine_cs *ring,
 				     u32 value)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -1752,7 +1794,7 @@ static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
 		   _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
 }
 
-static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
+static int gen6_bsd_ring_flush(struct intel_engine_cs *ring,
 			       u32 invalidate, u32 flush)
 {
 	uint32_t cmd;
@@ -1788,8 +1830,8 @@ static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			      u32 offset, u32 len,
+gen8_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
+			      u64 offset, u32 len,
 			      unsigned flags)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -1803,8 +1845,8 @@ gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 
 	/* FIXME(BDW): Address space and security selectors. */
 	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8));
-	intel_ring_emit(ring, offset);
-	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, lower_32_bits(offset));
+	intel_ring_emit(ring, upper_32_bits(offset));
 	intel_ring_emit(ring, MI_NOOP);
 	intel_ring_advance(ring);
 
@@ -1812,8 +1854,8 @@ gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 }
 
 static int
-hsw_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			      u32 offset, u32 len,
+hsw_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
+			      u64 offset, u32 len,
 			      unsigned flags)
 {
 	int ret;
@@ -1833,8 +1875,8 @@ hsw_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 }
 
 static int
-gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			      u32 offset, u32 len,
+gen6_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
+			      u64 offset, u32 len,
 			      unsigned flags)
 {
 	int ret;
@@ -1855,7 +1897,7 @@ gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 
 /* Blitter support (SandyBridge+) */
 
-static int gen6_ring_flush(struct intel_ring_buffer *ring,
+static int gen6_ring_flush(struct intel_engine_cs *ring,
 			   u32 invalidate, u32 flush)
 {
 	struct drm_device *dev = ring->dev;
@@ -1898,7 +1940,7 @@ static int gen6_ring_flush(struct intel_ring_buffer *ring,
 int intel_init_render_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 
 	ring->name = "render ring";
 	ring->id = RCS;
@@ -1920,15 +1962,24 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
 		ring->get_seqno = gen6_ring_get_seqno;
 		ring->set_seqno = ring_set_seqno;
-		ring->sync_to = gen6_ring_sync;
-		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_RV;
-		ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_RB;
-		ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_RVE;
-		ring->signal_mbox[RCS] = GEN6_NOSYNC;
-		ring->signal_mbox[VCS] = GEN6_VRSYNC;
-		ring->signal_mbox[BCS] = GEN6_BRSYNC;
-		ring->signal_mbox[VECS] = GEN6_VERSYNC;
+		ring->semaphore.sync_to = gen6_ring_sync;
+		ring->semaphore.signal = gen6_signal;
+		/*
+		 * The current semaphore is only applied on pre-gen8 platform.
+		 * And there is no VCS2 ring on the pre-gen8 platform. So the
+		 * semaphore between RCS and VCS2 is initialized as INVALID.
+		 * Gen8 will initialize the sema between VCS2 and RCS later.
+		 */
+		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
+		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
+		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
+		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
+		ring->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
+		ring->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
+		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	} else if (IS_GEN5(dev)) {
 		ring->add_request = pc_render_add_request;
 		ring->flush = gen4_render_ring_flush;
@@ -1999,16 +2050,25 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	int ret;
 
+	if (ringbuf == NULL) {
+		ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
+		if (!ringbuf)
+			return -ENOMEM;
+		ring->buffer = ringbuf;
+	}
+
 	ring->name = "render ring";
 	ring->id = RCS;
 	ring->mmio_base = RENDER_RING_BASE;
 
 	if (INTEL_INFO(dev)->gen >= 6) {
 		/* non-kms not supported on gen6+ */
-		return -ENODEV;
+		ret = -ENODEV;
+		goto err_ringbuf;
 	}
 
 	/* Note: gem is not supported on gen5/ilk without kms (the corresponding
@@ -2043,31 +2103,39 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size)
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
 
-	ring->size = size;
-	ring->effective_size = ring->size;
+	ringbuf->size = size;
+	ringbuf->effective_size = ringbuf->size;
 	if (IS_I830(ring->dev) || IS_845G(ring->dev))
-		ring->effective_size -= 128;
+		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
 
-	ring->virtual_start = ioremap_wc(start, size);
-	if (ring->virtual_start == NULL) {
+	ringbuf->virtual_start = ioremap_wc(start, size);
+	if (ringbuf->virtual_start == NULL) {
 		DRM_ERROR("can not ioremap virtual address for"
 			  " ring buffer\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_ringbuf;
 	}
 
 	if (!I915_NEED_GFX_HWS(dev)) {
 		ret = init_phys_status_page(ring);
 		if (ret)
-			return ret;
+			goto err_vstart;
 	}
 
 	return 0;
+
+err_vstart:
+	iounmap(ringbuf->virtual_start);
+err_ringbuf:
+	kfree(ringbuf);
+	ring->buffer = NULL;
+	return ret;
 }
 
 int intel_init_bsd_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[VCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[VCS];
 
 	ring->name = "bsd ring";
 	ring->id = VCS;
@@ -2096,15 +2164,24 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 			ring->dispatch_execbuffer =
 				gen6_ring_dispatch_execbuffer;
 		}
-		ring->sync_to = gen6_ring_sync;
-		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VR;
-		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_VB;
-		ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_VVE;
-		ring->signal_mbox[RCS] = GEN6_RVSYNC;
-		ring->signal_mbox[VCS] = GEN6_NOSYNC;
-		ring->signal_mbox[BCS] = GEN6_BVSYNC;
-		ring->signal_mbox[VECS] = GEN6_VEVSYNC;
+		ring->semaphore.sync_to = gen6_ring_sync;
+		ring->semaphore.signal = gen6_signal;
+		/*
+		 * The current semaphore is only applied on pre-gen8 platform.
+		 * And there is no VCS2 ring on the pre-gen8 platform. So the
+		 * semaphore between VCS and VCS2 is initialized as INVALID.
+		 * Gen8 will initialize the sema between VCS2 and VCS later.
+		 */
+		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
+		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
+		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
+		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
+		ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
+		ring->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
+		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	} else {
 		ring->mmio_base = BSD_RING_BASE;
 		ring->flush = bsd_ring_flush;
@@ -2127,10 +2204,63 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 	return intel_init_ring_buffer(dev, ring);
 }
 
+/**
+ * Initialize the second BSD ring for Broadwell GT3.
+ * It is noted that this only exists on Broadwell GT3.
+ */
+int intel_init_bsd2_ring_buffer(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring = &dev_priv->ring[VCS2];
+
+	if ((INTEL_INFO(dev)->gen != 8)) {
+		DRM_ERROR("No dual-BSD ring on non-BDW machine\n");
+		return -EINVAL;
+	}
+
+	ring->name = "bds2_ring";
+	ring->id = VCS2;
+
+	ring->write_tail = ring_write_tail;
+	ring->mmio_base = GEN8_BSD2_RING_BASE;
+	ring->flush = gen6_bsd_ring_flush;
+	ring->add_request = gen6_add_request;
+	ring->get_seqno = gen6_ring_get_seqno;
+	ring->set_seqno = ring_set_seqno;
+	ring->irq_enable_mask =
+			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT;
+	ring->irq_get = gen8_ring_get_irq;
+	ring->irq_put = gen8_ring_put_irq;
+	ring->dispatch_execbuffer =
+			gen8_ring_dispatch_execbuffer;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	/*
+	 * The current semaphore is only applied on the pre-gen8. And there
+	 * is no bsd2 ring on the pre-gen8. So now the semaphore_register
+	 * between VCS2 and other ring is initialized as invalid.
+	 * Gen8 will initialize the sema between VCS2 and other ring later.
+	 */
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+
+	ring->init = init_ring_common;
+
+	return intel_init_ring_buffer(dev, ring);
+}
+
 int intel_init_blt_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[BCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[BCS];
 
 	ring->name = "blitter ring";
 	ring->id = BCS;
@@ -2153,15 +2283,24 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 		ring->irq_put = gen6_ring_put_irq;
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 	}
-	ring->sync_to = gen6_ring_sync;
-	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_BR;
-	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_BV;
-	ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_BVE;
-	ring->signal_mbox[RCS] = GEN6_RBSYNC;
-	ring->signal_mbox[VCS] = GEN6_VBSYNC;
-	ring->signal_mbox[BCS] = GEN6_NOSYNC;
-	ring->signal_mbox[VECS] = GEN6_VEBSYNC;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	/*
+	 * The current semaphore is only applied on pre-gen8 platform. And
+	 * there is no VCS2 ring on the pre-gen8 platform. So the semaphore
+	 * between BCS and VCS2 is initialized as INVALID.
+	 * Gen8 will initialize the sema between BCS and VCS2 later.
+	 */
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);
@@ -2170,7 +2309,7 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 int intel_init_vebox_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[VECS];
+	struct intel_engine_cs *ring = &dev_priv->ring[VECS];
 
 	ring->name = "video enhancement ring";
 	ring->id = VECS;
@@ -2194,22 +2333,25 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev)
 		ring->irq_put = hsw_vebox_put_irq;
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 	}
-	ring->sync_to = gen6_ring_sync;
-	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VER;
-	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_VEV;
-	ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_VEB;
-	ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->signal_mbox[RCS] = GEN6_RVESYNC;
-	ring->signal_mbox[VCS] = GEN6_VVESYNC;
-	ring->signal_mbox[BCS] = GEN6_BVESYNC;
-	ring->signal_mbox[VECS] = GEN6_NOSYNC;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);
 }
 
 int
-intel_ring_flush_all_caches(struct intel_ring_buffer *ring)
+intel_ring_flush_all_caches(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -2227,7 +2369,7 @@ intel_ring_flush_all_caches(struct intel_ring_buffer *ring)
 }
 
 int
-intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring)
+intel_ring_invalidate_all_caches(struct intel_engine_cs *ring)
 {
 	uint32_t flush_domains;
 	int ret;
@@ -2245,3 +2387,19 @@ intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring)
 	ring->gpu_caches_dirty = false;
 	return 0;
 }
+
+void
+intel_stop_ring_buffer(struct intel_engine_cs *ring)
+{
+	int ret;
+
+	if (!intel_ring_initialized(ring))
+		return;
+
+	ret = intel_ring_idle(ring);
+	if (ret && !i915_reset_in_progress(&to_i915(ring->dev)->gpu_error))
+		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
+			  ring->name, ret);
+
+	stop_ring(ring);
+}
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 2b91c4b4d34b..e72017bdcd7f 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -1,6 +1,10 @@
 #ifndef _INTEL_RINGBUFFER_H_
 #define _INTEL_RINGBUFFER_H_
 
+#include <linux/hashtable.h>
+
+#define I915_CMD_HASH_ORDER 9
+
 /*
  * Gen2 BSpec "1. Programming Environment" / 1.4.4.6 "Ring Buffer Use"
  * Gen3 BSpec "vol1c Memory Interface Functions" / 2.3.4.5 "Ring Buffer Use"
@@ -51,79 +55,96 @@ struct intel_ring_hangcheck {
 	u32 seqno;
 	int score;
 	enum intel_ring_hangcheck_action action;
-	bool deadlock;
+	int deadlock;
 };
 
-struct  intel_ring_buffer {
+struct intel_ringbuffer {
+	struct drm_i915_gem_object *obj;
+	void __iomem *virtual_start;
+
+	u32 head;
+	u32 tail;
+	int space;
+	int size;
+	int effective_size;
+
+	/** We track the position of the requests in the ring buffer, and
+	 * when each is retired we increment last_retired_head as the GPU
+	 * must have finished processing the request and so we know we
+	 * can advance the ringbuffer up to that position.
+	 *
+	 * last_retired_head is set to -1 after the value is consumed so
+	 * we can detect new retirements.
+	 */
+	u32 last_retired_head;
+};
+
+struct  intel_engine_cs {
 	const char	*name;
 	enum intel_ring_id {
 		RCS = 0x0,
 		VCS,
 		BCS,
 		VECS,
+		VCS2
 	} id;
-#define I915_NUM_RINGS 4
+#define I915_NUM_RINGS 5
+#define LAST_USER_RING (VECS + 1)
 	u32		mmio_base;
-	void		__iomem *virtual_start;
 	struct		drm_device *dev;
-	struct		drm_i915_gem_object *obj;
+	struct intel_ringbuffer *buffer;
 
-	u32		head;
-	u32		tail;
-	int		space;
-	int		size;
-	int		effective_size;
 	struct intel_hw_status_page status_page;
 
-	/** We track the position of the requests in the ring buffer, and
-	 * when each is retired we increment last_retired_head as the GPU
-	 * must have finished processing the request and so we know we
-	 * can advance the ringbuffer up to that position.
-	 *
-	 * last_retired_head is set to -1 after the value is consumed so
-	 * we can detect new retirements.
-	 */
-	u32		last_retired_head;
-
 	unsigned irq_refcount; /* protected by dev_priv->irq_lock */
 	u32		irq_enable_mask;	/* bitmask to enable ring interrupt */
 	u32		trace_irq_seqno;
-	u32		sync_seqno[I915_NUM_RINGS-1];
-	bool __must_check (*irq_get)(struct intel_ring_buffer *ring);
-	void		(*irq_put)(struct intel_ring_buffer *ring);
+	bool __must_check (*irq_get)(struct intel_engine_cs *ring);
+	void		(*irq_put)(struct intel_engine_cs *ring);
 
-	int		(*init)(struct intel_ring_buffer *ring);
+	int		(*init)(struct intel_engine_cs *ring);
 
-	void		(*write_tail)(struct intel_ring_buffer *ring,
+	void		(*write_tail)(struct intel_engine_cs *ring,
 				      u32 value);
-	int __must_check (*flush)(struct intel_ring_buffer *ring,
+	int __must_check (*flush)(struct intel_engine_cs *ring,
 				  u32	invalidate_domains,
 				  u32	flush_domains);
-	int		(*add_request)(struct intel_ring_buffer *ring);
+	int		(*add_request)(struct intel_engine_cs *ring);
 	/* Some chipsets are not quite as coherent as advertised and need
 	 * an expensive kick to force a true read of the up-to-date seqno.
 	 * However, the up-to-date seqno is not always required and the last
 	 * seen value is good enough. Note that the seqno will always be
 	 * monotonic, even if not coherent.
 	 */
-	u32		(*get_seqno)(struct intel_ring_buffer *ring,
+	u32		(*get_seqno)(struct intel_engine_cs *ring,
 				     bool lazy_coherency);
-	void		(*set_seqno)(struct intel_ring_buffer *ring,
+	void		(*set_seqno)(struct intel_engine_cs *ring,
 				     u32 seqno);
-	int		(*dispatch_execbuffer)(struct intel_ring_buffer *ring,
-					       u32 offset, u32 length,
+	int		(*dispatch_execbuffer)(struct intel_engine_cs *ring,
+					       u64 offset, u32 length,
 					       unsigned flags);
 #define I915_DISPATCH_SECURE 0x1
 #define I915_DISPATCH_PINNED 0x2
-	void		(*cleanup)(struct intel_ring_buffer *ring);
-	int		(*sync_to)(struct intel_ring_buffer *ring,
-				   struct intel_ring_buffer *to,
-				   u32 seqno);
+	void		(*cleanup)(struct intel_engine_cs *ring);
 
-	/* our mbox written by others */
-	u32		semaphore_register[I915_NUM_RINGS];
-	/* mboxes this ring signals to */
-	u32		signal_mbox[I915_NUM_RINGS];
+	struct {
+		u32	sync_seqno[I915_NUM_RINGS-1];
+
+		struct {
+			/* our mbox written by others */
+			u32		wait[I915_NUM_RINGS];
+			/* mboxes this ring signals to */
+			u32		signal[I915_NUM_RINGS];
+		} mbox;
+
+		/* AKA wait() */
+		int	(*sync_to)(struct intel_engine_cs *ring,
+				   struct intel_engine_cs *to,
+				   u32 seqno);
+		int	(*signal)(struct intel_engine_cs *signaller,
+				  /* num_dwords needed by caller */
+				  unsigned int num_dwords);
+	} semaphore;
 
 	/**
 	 * List of objects currently involved in rendering from the
@@ -153,12 +174,8 @@ struct  intel_ring_buffer {
 
 	wait_queue_head_t irq_queue;
 
-	/**
-	 * Do an explicit TLB flush before MI_SET_CONTEXT
-	 */
-	bool itlb_before_ctx_switch;
-	struct i915_hw_context *default_context;
-	struct i915_hw_context *last_context;
+	struct intel_context *default_context;
+	struct intel_context *last_context;
 
 	struct intel_ring_hangcheck hangcheck;
 
@@ -168,12 +185,13 @@ struct  intel_ring_buffer {
 		volatile u32 *cpu_page;
 	} scratch;
 
+	bool needs_cmd_parser;
+
 	/*
-	 * Tables of commands the command parser needs to know about
+	 * Table of commands the command parser needs to know about
 	 * for this ring.
 	 */
-	const struct drm_i915_cmd_table *cmd_tables;
-	int cmd_table_count;
+	DECLARE_HASHTABLE(cmd_hash, I915_CMD_HASH_ORDER);
 
 	/*
 	 * Table of registers allowed in commands that read/write registers.
@@ -202,20 +220,20 @@ struct  intel_ring_buffer {
 };
 
 static inline bool
-intel_ring_initialized(struct intel_ring_buffer *ring)
+intel_ring_initialized(struct intel_engine_cs *ring)
 {
-	return ring->obj != NULL;
+	return ring->buffer && ring->buffer->obj;
 }
 
 static inline unsigned
-intel_ring_flag(struct intel_ring_buffer *ring)
+intel_ring_flag(struct intel_engine_cs *ring)
 {
 	return 1 << ring->id;
 }
 
 static inline u32
-intel_ring_sync_index(struct intel_ring_buffer *ring,
-		      struct intel_ring_buffer *other)
+intel_ring_sync_index(struct intel_engine_cs *ring,
+		      struct intel_engine_cs *other)
 {
 	int idx;
 
@@ -233,7 +251,7 @@ intel_ring_sync_index(struct intel_ring_buffer *ring,
 }
 
 static inline u32
-intel_read_status_page(struct intel_ring_buffer *ring,
+intel_read_status_page(struct intel_engine_cs *ring,
 		       int reg)
 {
 	/* Ensure that the compiler doesn't optimize away the load. */
@@ -242,7 +260,7 @@ intel_read_status_page(struct intel_ring_buffer *ring,
 }
 
 static inline void
-intel_write_status_page(struct intel_ring_buffer *ring,
+intel_write_status_page(struct intel_engine_cs *ring,
 			int reg, u32 value)
 {
 	ring->status_page.page_addr[reg] = value;
@@ -267,47 +285,51 @@ intel_write_status_page(struct intel_ring_buffer *ring,
 #define I915_GEM_HWS_SCRATCH_INDEX	0x30
 #define I915_GEM_HWS_SCRATCH_ADDR (I915_GEM_HWS_SCRATCH_INDEX << MI_STORE_DWORD_INDEX_SHIFT)
 
-void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring);
+void intel_stop_ring_buffer(struct intel_engine_cs *ring);
+void intel_cleanup_ring_buffer(struct intel_engine_cs *ring);
 
-int __must_check intel_ring_begin(struct intel_ring_buffer *ring, int n);
-int __must_check intel_ring_cacheline_align(struct intel_ring_buffer *ring);
-static inline void intel_ring_emit(struct intel_ring_buffer *ring,
+int __must_check intel_ring_begin(struct intel_engine_cs *ring, int n);
+int __must_check intel_ring_cacheline_align(struct intel_engine_cs *ring);
+static inline void intel_ring_emit(struct intel_engine_cs *ring,
 				   u32 data)
 {
-	iowrite32(data, ring->virtual_start + ring->tail);
-	ring->tail += 4;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
+	ringbuf->tail += 4;
 }
-static inline void intel_ring_advance(struct intel_ring_buffer *ring)
+static inline void intel_ring_advance(struct intel_engine_cs *ring)
 {
-	ring->tail &= ring->size - 1;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	ringbuf->tail &= ringbuf->size - 1;
 }
-void __intel_ring_advance(struct intel_ring_buffer *ring);
+void __intel_ring_advance(struct intel_engine_cs *ring);
 
-int __must_check intel_ring_idle(struct intel_ring_buffer *ring);
-void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno);
-int intel_ring_flush_all_caches(struct intel_ring_buffer *ring);
-int intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring);
+int __must_check intel_ring_idle(struct intel_engine_cs *ring);
+void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno);
+int intel_ring_flush_all_caches(struct intel_engine_cs *ring);
+int intel_ring_invalidate_all_caches(struct intel_engine_cs *ring);
 
 int intel_init_render_ring_buffer(struct drm_device *dev);
 int intel_init_bsd_ring_buffer(struct drm_device *dev);
+int intel_init_bsd2_ring_buffer(struct drm_device *dev);
 int intel_init_blt_ring_buffer(struct drm_device *dev);
 int intel_init_vebox_ring_buffer(struct drm_device *dev);
 
-u64 intel_ring_get_active_head(struct intel_ring_buffer *ring);
-void intel_ring_setup_status_page(struct intel_ring_buffer *ring);
+u64 intel_ring_get_active_head(struct intel_engine_cs *ring);
+void intel_ring_setup_status_page(struct intel_engine_cs *ring);
 
-static inline u32 intel_ring_get_tail(struct intel_ring_buffer *ring)
+static inline u32 intel_ring_get_tail(struct intel_engine_cs *ring)
 {
-	return ring->tail;
+	return ring->buffer->tail;
 }
 
-static inline u32 intel_ring_get_seqno(struct intel_ring_buffer *ring)
+static inline u32 intel_ring_get_seqno(struct intel_engine_cs *ring)
 {
 	BUG_ON(ring->outstanding_lazy_seqno == 0);
 	return ring->outstanding_lazy_seqno;
 }
 
-static inline void i915_trace_irq_get(struct intel_ring_buffer *ring, u32 seqno)
+static inline void i915_trace_irq_get(struct intel_engine_cs *ring, u32 seqno)
 {
 	if (ring->trace_irq_seqno == 0 && ring->irq_get(ring))
 		ring->trace_irq_seqno = seqno;
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 46be00d66df3..20375cc7f82d 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1153,20 +1153,21 @@ static bool intel_sdvo_compute_config(struct intel_encoder *encoder,
 	pipe_config->pixel_multiplier =
 		intel_sdvo_get_pixel_multiplier(adjusted_mode);
 
+	pipe_config->has_hdmi_sink = intel_sdvo->has_hdmi_monitor;
+
 	if (intel_sdvo->color_range_auto) {
 		/* See CEA-861-E - 5.1 Default Encoding Parameters */
 		/* FIXME: This bit is only valid when using TMDS encoding and 8
 		 * bit per color mode. */
-		if (intel_sdvo->has_hdmi_monitor &&
+		if (pipe_config->has_hdmi_sink &&
 		    drm_match_cea_mode(adjusted_mode) > 1)
-			intel_sdvo->color_range = HDMI_COLOR_RANGE_16_235;
-		else
-			intel_sdvo->color_range = 0;
+			pipe_config->limited_color_range = true;
+	} else {
+		if (pipe_config->has_hdmi_sink &&
+		    intel_sdvo->color_range == HDMI_COLOR_RANGE_16_235)
+			pipe_config->limited_color_range = true;
 	}
 
-	if (intel_sdvo->color_range)
-		pipe_config->limited_color_range = true;
-
 	/* Clock computation needs to happen after pixel multiplier. */
 	if (intel_sdvo->is_tv)
 		i9xx_adjust_sdvo_tv_clock(pipe_config);
@@ -1174,7 +1175,7 @@ static bool intel_sdvo_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
+static void intel_sdvo_pre_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1223,7 +1224,7 @@ static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
 	if (!intel_sdvo_set_target_input(intel_sdvo))
 		return;
 
-	if (intel_sdvo->has_hdmi_monitor) {
+	if (crtc->config.has_hdmi_sink) {
 		intel_sdvo_set_encode(intel_sdvo, SDVO_ENCODE_HDMI);
 		intel_sdvo_set_colorimetry(intel_sdvo,
 					   SDVO_COLORIMETRY_RGB256);
@@ -1258,8 +1259,8 @@ static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
 		/* The real mode polarity is set by the SDVO commands, using
 		 * struct intel_sdvo_dtd. */
 		sdvox = SDVO_VSYNC_ACTIVE_HIGH | SDVO_HSYNC_ACTIVE_HIGH;
-		if (!HAS_PCH_SPLIT(dev) && intel_sdvo->is_hdmi)
-			sdvox |= intel_sdvo->color_range;
+		if (!HAS_PCH_SPLIT(dev) && crtc->config.limited_color_range)
+			sdvox |= HDMI_COLOR_RANGE_16_235;
 		if (INTEL_INFO(dev)->gen < 5)
 			sdvox |= SDVO_BORDER_ENABLE;
 	} else {
@@ -1349,6 +1350,8 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 	u8 val;
 	bool ret;
 
+	sdvox = I915_READ(intel_sdvo->sdvo_reg);
+
 	ret = intel_sdvo_get_input_timing(intel_sdvo, &dtd);
 	if (!ret) {
 		/* Some sdvo encoders are not spec compliant and don't
@@ -1377,13 +1380,14 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 	 * other platfroms.
 	 */
 	if (IS_I915G(dev) || IS_I915GM(dev)) {
-		sdvox = I915_READ(intel_sdvo->sdvo_reg);
 		pipe_config->pixel_multiplier =
 			((sdvox & SDVO_PORT_MULTIPLY_MASK)
 			 >> SDVO_PORT_MULTIPLY_SHIFT) + 1;
 	}
 
-	dotclock = pipe_config->port_clock / pipe_config->pixel_multiplier;
+	dotclock = pipe_config->port_clock;
+	if (pipe_config->pixel_multiplier)
+		dotclock /= pipe_config->pixel_multiplier;
 
 	if (HAS_PCH_SPLIT(dev))
 		ironlake_check_encoder_dotclock(pipe_config, dotclock);
@@ -1406,6 +1410,15 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 		}
 	}
 
+	if (sdvox & HDMI_COLOR_RANGE_16_235)
+		pipe_config->limited_color_range = true;
+
+	if (intel_sdvo_get_value(intel_sdvo, SDVO_CMD_GET_ENCODE,
+				 &val, 1)) {
+		if (val == SDVO_ENCODE_HDMI)
+			pipe_config->has_hdmi_sink = true;
+	}
+
 	WARN(encoder_pixel_multiplier != pipe_config->pixel_multiplier,
 	     "SDVO pixel multiplier mismatch, port: %i, encoder: %i\n",
 	     pipe_config->pixel_multiplier, encoder_pixel_multiplier);
@@ -1732,7 +1745,7 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status ret;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	if (!intel_sdvo_get_value(intel_sdvo,
 				  SDVO_CMD_GET_ATTACHED_DISPLAYS,
@@ -1794,7 +1807,7 @@ static void intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 	struct edid *edid;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	/* set the bus switch and get the modes */
 	edid = intel_sdvo_get_edid(connector);
@@ -1892,7 +1905,7 @@ static void intel_sdvo_get_tv_modes(struct drm_connector *connector)
 	int i;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	/* Read the list of supported input resolutions for the selected TV
 	 * format.
@@ -1929,7 +1942,7 @@ static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
 	struct drm_display_mode *newmode;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	/*
 	 * Fetch modes from VBT. For SDVO prefer the VBT mode since some
@@ -2999,7 +3012,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
 
 	intel_encoder->compute_config = intel_sdvo_compute_config;
 	intel_encoder->disable = intel_disable_sdvo;
-	intel_encoder->mode_set = intel_sdvo_mode_set;
+	intel_encoder->pre_enable = intel_sdvo_pre_enable;
 	intel_encoder->enable = intel_enable_sdvo;
 	intel_encoder->get_hw_state = intel_sdvo_get_hw_state;
 	intel_encoder->get_config = intel_sdvo_get_config;
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 0954f132726e..01d841ea3140 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -29,12 +29,21 @@
  * IOSF sideband, see VLV2_SidebandMsg_HAS.docx and
  * VLV_VLV2_PUNIT_HAS_0.8.docx
  */
+
+/* Standard MMIO read, non-posted */
+#define SB_MRD_NP	0x00
+/* Standard MMIO write, non-posted */
+#define SB_MWR_NP	0x01
+/* Private register read, double-word addressing, non-posted */
+#define SB_CRRDDA_NP	0x06
+/* Private register write, double-word addressing, non-posted */
+#define SB_CRWRDA_NP	0x07
+
 static int vlv_sideband_rw(struct drm_i915_private *dev_priv, u32 devfn,
 			   u32 port, u32 opcode, u32 addr, u32 *val)
 {
 	u32 cmd, be = 0xf, bar = 0;
-	bool is_read = (opcode == PUNIT_OPCODE_REG_READ ||
-			opcode == DPIO_OPCODE_REG_READ);
+	bool is_read = (opcode == SB_MRD_NP || opcode == SB_CRRDDA_NP);
 
 	cmd = (devfn << IOSF_DEVFN_SHIFT) | (opcode << IOSF_OPCODE_SHIFT) |
 		(port << IOSF_PORT_SHIFT) | (be << IOSF_BYTE_ENABLES_SHIFT) |
@@ -74,7 +83,7 @@ u32 vlv_punit_read(struct drm_i915_private *dev_priv, u8 addr)
 
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_PUNIT,
-			PUNIT_OPCODE_REG_READ, addr, &val);
+			SB_CRRDDA_NP, addr, &val);
 	mutex_unlock(&dev_priv->dpio_lock);
 
 	return val;
@@ -86,7 +95,7 @@ void vlv_punit_write(struct drm_i915_private *dev_priv, u8 addr, u32 val)
 
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_PUNIT,
-			PUNIT_OPCODE_REG_WRITE, addr, &val);
+			SB_CRWRDA_NP, addr, &val);
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
@@ -95,7 +104,7 @@ u32 vlv_bunit_read(struct drm_i915_private *dev_priv, u32 reg)
 	u32 val = 0;
 
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_BUNIT,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 
 	return val;
 }
@@ -103,7 +112,7 @@ u32 vlv_bunit_read(struct drm_i915_private *dev_priv, u32 reg)
 void vlv_bunit_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_BUNIT,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
@@ -114,7 +123,7 @@ u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
 
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_NC,
-			PUNIT_OPCODE_REG_READ, addr, &val);
+			SB_CRRDDA_NP, addr, &val);
 	mutex_unlock(&dev_priv->dpio_lock);
 
 	return val;
@@ -124,56 +133,56 @@ u32 vlv_gpio_nc_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPIO_NC,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_gpio_nc_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPIO_NC,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_cck_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCK,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_cck_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCK,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_ccu_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCU,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_ccu_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCU,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_gps_core_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPS_CORE,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_gps_core_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPS_CORE,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg)
@@ -181,14 +190,22 @@ u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg)
 	u32 val = 0;
 
 	vlv_sideband_rw(dev_priv, DPIO_DEVFN, DPIO_PHY_IOSF_PORT(DPIO_PHY(pipe)),
-			DPIO_OPCODE_REG_READ, reg, &val);
+			SB_MRD_NP, reg, &val);
+
+	/*
+	 * FIXME: There might be some registers where all 1's is a valid value,
+	 * so ideally we should check the register offset instead...
+	 */
+	WARN(val == 0xffffffff, "DPIO read pipe %c reg 0x%x == 0x%x\n",
+	     pipe_name(pipe), reg, val);
+
 	return val;
 }
 
 void vlv_dpio_write(struct drm_i915_private *dev_priv, enum pipe pipe, int reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, DPIO_DEVFN, DPIO_PHY_IOSF_PORT(DPIO_PHY(pipe)),
-			DPIO_OPCODE_REG_WRITE, reg, &val);
+			SB_MWR_NP, reg, &val);
 }
 
 /* SBI access */
@@ -253,13 +270,13 @@ void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 u32 vlv_flisdsi_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
-	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI,
-					DPIO_OPCODE_REG_READ, reg, &val);
+	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI, SB_CRRDDA_NP,
+			reg, &val);
 	return val;
 }
 
 void vlv_flisdsi_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
-	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI,
-					DPIO_OPCODE_REG_WRITE, reg, &val);
+	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI, SB_CRWRDA_NP,
+			reg, &val);
 }
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 336ae6c602f2..9a17b4e92ef4 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -37,6 +37,106 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
+static int usecs_to_scanlines(const struct drm_display_mode *mode, int usecs)
+{
+	/* paranoia */
+	if (!mode->crtc_htotal)
+		return 1;
+
+	return DIV_ROUND_UP(usecs * mode->crtc_clock, 1000 * mode->crtc_htotal);
+}
+
+static bool intel_pipe_update_start(struct intel_crtc *crtc, uint32_t *start_vbl_count)
+{
+	struct drm_device *dev = crtc->base.dev;
+	const struct drm_display_mode *mode = &crtc->config.adjusted_mode;
+	enum pipe pipe = crtc->pipe;
+	long timeout = msecs_to_jiffies_timeout(1);
+	int scanline, min, max, vblank_start;
+	DEFINE_WAIT(wait);
+
+	WARN_ON(!drm_modeset_is_locked(&crtc->base.mutex));
+
+	vblank_start = mode->crtc_vblank_start;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		vblank_start = DIV_ROUND_UP(vblank_start, 2);
+
+	/* FIXME needs to be calibrated sensibly */
+	min = vblank_start - usecs_to_scanlines(mode, 100);
+	max = vblank_start - 1;
+
+	if (min <= 0 || max <= 0)
+		return false;
+
+	if (WARN_ON(drm_vblank_get(dev, pipe)))
+		return false;
+
+	local_irq_disable();
+
+	trace_i915_pipe_update_start(crtc, min, max);
+
+	for (;;) {
+		/*
+		 * prepare_to_wait() has a memory barrier, which guarantees
+		 * other CPUs can see the task state update by the time we
+		 * read the scanline.
+		 */
+		prepare_to_wait(&crtc->vbl_wait, &wait, TASK_UNINTERRUPTIBLE);
+
+		scanline = intel_get_crtc_scanline(crtc);
+		if (scanline < min || scanline > max)
+			break;
+
+		if (timeout <= 0) {
+			DRM_ERROR("Potential atomic update failure on pipe %c\n",
+				  pipe_name(crtc->pipe));
+			break;
+		}
+
+		local_irq_enable();
+
+		timeout = schedule_timeout(timeout);
+
+		local_irq_disable();
+	}
+
+	finish_wait(&crtc->vbl_wait, &wait);
+
+	drm_vblank_put(dev, pipe);
+
+	*start_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+
+	trace_i915_pipe_update_vblank_evaded(crtc, min, max, *start_vbl_count);
+
+	return true;
+}
+
+static void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count)
+{
+	struct drm_device *dev = crtc->base.dev;
+	enum pipe pipe = crtc->pipe;
+	u32 end_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+
+	trace_i915_pipe_update_end(crtc, end_vbl_count);
+
+	local_irq_enable();
+
+	if (start_vbl_count != end_vbl_count)
+		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u)\n",
+			  pipe_name(pipe), start_vbl_count, end_vbl_count);
+}
+
+static void intel_update_primary_plane(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	int reg = DSPCNTR(crtc->plane);
+
+	if (crtc->primary_enabled)
+		I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
+	else
+		I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
+}
+
 static void
 vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 		 struct drm_framebuffer *fb,
@@ -48,11 +148,14 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(dplane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
 	u32 sprctl;
 	unsigned long sprsurf_offset, linear_offset;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 start_vbl_count;
+	bool atomic_update;
 
 	sprctl = I915_READ(SPCNTR(pipe, plane));
 
@@ -131,6 +234,10 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 							fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
+
 	I915_WRITE(SPSTRIDE(pipe, plane), fb->pitches[0]);
 	I915_WRITE(SPPOS(pipe, plane), (crtc_y << 16) | crtc_x);
 
@@ -143,7 +250,11 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 	I915_WRITE(SPCNTR(pipe, plane), sprctl);
 	I915_WRITE(SPSURF(pipe, plane), i915_gem_obj_ggtt_offset(obj) +
 		   sprsurf_offset);
-	POSTING_READ(SPSURF(pipe, plane));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 }
 
 static void
@@ -152,14 +263,25 @@ vlv_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc)
 	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(dplane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
+	u32 start_vbl_count;
+	bool atomic_update;
+
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
 
 	I915_WRITE(SPCNTR(pipe, plane), I915_READ(SPCNTR(pipe, plane)) &
 		   ~SP_ENABLE);
 	/* Activate double buffered register update */
 	I915_WRITE(SPSURF(pipe, plane), 0);
-	POSTING_READ(SPSURF(pipe, plane));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 
 	intel_update_sprite_watermarks(dplane, crtc, 0, 0, false, false);
 }
@@ -226,10 +348,13 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	u32 sprctl, sprscale = 0;
 	unsigned long sprsurf_offset, linear_offset;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 start_vbl_count;
+	bool atomic_update;
 
 	sprctl = I915_READ(SPRCTL(pipe));
 
@@ -299,6 +424,10 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 					       pixel_size, fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
+
 	I915_WRITE(SPRSTRIDE(pipe), fb->pitches[0]);
 	I915_WRITE(SPRPOS(pipe), (crtc_y << 16) | crtc_x);
 
@@ -317,7 +446,11 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	I915_WRITE(SPRCTL(pipe), sprctl);
 	I915_WRITE(SPRSURF(pipe),
 		   i915_gem_obj_ggtt_offset(obj) + sprsurf_offset);
-	POSTING_READ(SPRSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 }
 
 static void
@@ -326,7 +459,14 @@ ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
+	u32 start_vbl_count;
+	bool atomic_update;
+
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
 
 	I915_WRITE(SPRCTL(pipe), I915_READ(SPRCTL(pipe)) & ~SPRITE_ENABLE);
 	/* Can't leave the scaler enabled... */
@@ -334,7 +474,11 @@ ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 		I915_WRITE(SPRSCALE(pipe), 0);
 	/* Activate double buffered register update */
 	I915_WRITE(SPRSURF(pipe), 0);
-	POSTING_READ(SPRSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 
 	/*
 	 * Avoid underruns when disabling the sprite.
@@ -410,10 +554,13 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	unsigned long dvssurf_offset, linear_offset;
 	u32 dvscntr, dvsscale;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 start_vbl_count;
+	bool atomic_update;
 
 	dvscntr = I915_READ(DVSCNTR(pipe));
 
@@ -478,6 +625,10 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 					       pixel_size, fb->pitches[0]);
 	linear_offset -= dvssurf_offset;
 
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
+
 	I915_WRITE(DVSSTRIDE(pipe), fb->pitches[0]);
 	I915_WRITE(DVSPOS(pipe), (crtc_y << 16) | crtc_x);
 
@@ -491,7 +642,11 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	I915_WRITE(DVSCNTR(pipe), dvscntr);
 	I915_WRITE(DVSSURF(pipe),
 		   i915_gem_obj_ggtt_offset(obj) + dvssurf_offset);
-	POSTING_READ(DVSSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 }
 
 static void
@@ -500,14 +655,25 @@ ilk_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
+	u32 start_vbl_count;
+	bool atomic_update;
+
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
 
 	I915_WRITE(DVSCNTR(pipe), I915_READ(DVSCNTR(pipe)) & ~DVS_ENABLE);
 	/* Disable the scaler */
 	I915_WRITE(DVSSCALE(pipe), 0);
 	/* Flush double buffered register updates */
 	I915_WRITE(DVSSURF(pipe), 0);
-	POSTING_READ(DVSSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 
 	/*
 	 * Avoid underruns when disabling the sprite.
@@ -519,20 +685,18 @@ ilk_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 }
 
 static void
-intel_enable_primary(struct drm_crtc *crtc)
+intel_post_enable_primary(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int reg = DSPCNTR(intel_crtc->plane);
 
-	if (intel_crtc->primary_enabled)
-		return;
-
-	intel_crtc->primary_enabled = true;
-
-	I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+	/*
+	 * BDW signals flip done immediately if the plane
+	 * is disabled, even if the plane enable is already
+	 * armed to occur at the next vblank :(
+	 */
+	if (IS_BROADWELL(dev))
+		intel_wait_for_vblank(dev, intel_crtc->pipe);
 
 	/*
 	 * FIXME IPS should be fine as long as one plane is
@@ -540,10 +704,7 @@ intel_enable_primary(struct drm_crtc *crtc)
 	 * when going from primary only to sprite only and vice
 	 * versa.
 	 */
-	if (intel_crtc->config.ips_enabled) {
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-		hsw_enable_ips(intel_crtc);
-	}
+	hsw_enable_ips(intel_crtc);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
@@ -551,17 +712,11 @@ intel_enable_primary(struct drm_crtc *crtc)
 }
 
 static void
-intel_disable_primary(struct drm_crtc *crtc)
+intel_pre_disable_primary(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int reg = DSPCNTR(intel_crtc->plane);
-
-	if (!intel_crtc->primary_enabled)
-		return;
-
-	intel_crtc->primary_enabled = false;
 
 	mutex_lock(&dev->struct_mutex);
 	if (dev_priv->fbc.plane == intel_crtc->plane)
@@ -575,9 +730,6 @@ intel_disable_primary(struct drm_crtc *crtc)
 	 * versa.
 	 */
 	hsw_disable_ips(intel_crtc);
-
-	I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
 }
 
 static int
@@ -671,7 +823,7 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_i915_gem_object *obj = intel_fb->obj;
 	struct drm_i915_gem_object *old_obj = intel_plane->obj;
 	int ret;
-	bool disable_primary = false;
+	bool primary_enabled;
 	bool visible;
 	int hscale, vscale;
 	int max_scale, min_scale;
@@ -842,8 +994,8 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	 * If the sprite is completely covering the primary plane,
 	 * we can disable the primary and save power.
 	 */
-	disable_primary = drm_rect_equals(&dst, &clip) && !colorkey_enabled(intel_plane);
-	WARN_ON(disable_primary && !visible && intel_crtc->active);
+	primary_enabled = !drm_rect_equals(&dst, &clip) || colorkey_enabled(intel_plane);
+	WARN_ON(!primary_enabled && !visible && intel_crtc->active);
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -870,12 +1022,15 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	intel_plane->obj = obj;
 
 	if (intel_crtc->active) {
-		/*
-		 * Be sure to re-enable the primary before the sprite is no longer
-		 * covering it fully.
-		 */
-		if (!disable_primary)
-			intel_enable_primary(crtc);
+		bool primary_was_enabled = intel_crtc->primary_enabled;
+
+		intel_crtc->primary_enabled = primary_enabled;
+
+		if (primary_was_enabled != primary_enabled)
+			intel_crtc_wait_for_pending_flips(crtc);
+
+		if (primary_was_enabled && !primary_enabled)
+			intel_pre_disable_primary(crtc);
 
 		if (visible)
 			intel_plane->update_plane(plane, crtc, fb, obj,
@@ -884,8 +1039,8 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		else
 			intel_plane->disable_plane(plane, crtc);
 
-		if (disable_primary)
-			intel_disable_primary(crtc);
+		if (!primary_was_enabled && primary_enabled)
+			intel_post_enable_primary(crtc);
 	}
 
 	/* Unpin old obj after new one is active to avoid ugliness */
@@ -923,8 +1078,14 @@ intel_disable_plane(struct drm_plane *plane)
 	intel_crtc = to_intel_crtc(plane->crtc);
 
 	if (intel_crtc->active) {
-		intel_enable_primary(plane->crtc);
+		bool primary_was_enabled = intel_crtc->primary_enabled;
+
+		intel_crtc->primary_enabled = true;
+
 		intel_plane->disable_plane(plane, plane->crtc);
+
+		if (!primary_was_enabled && intel_crtc->primary_enabled)
+			intel_post_enable_primary(plane->crtc);
 	}
 
 	if (intel_plane->obj) {
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index bafe92e317d5..67c6c9a2eb1c 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -934,7 +934,86 @@ intel_tv_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_tv_mode_set(struct intel_encoder *encoder)
+static void
+set_tv_mode_timings(struct drm_i915_private *dev_priv,
+		    const struct tv_mode *tv_mode,
+		    bool burst_ena)
+{
+	u32 hctl1, hctl2, hctl3;
+	u32 vctl1, vctl2, vctl3, vctl4, vctl5, vctl6, vctl7;
+
+	hctl1 = (tv_mode->hsync_end << TV_HSYNC_END_SHIFT) |
+		(tv_mode->htotal << TV_HTOTAL_SHIFT);
+
+	hctl2 = (tv_mode->hburst_start << 16) |
+		(tv_mode->hburst_len << TV_HBURST_LEN_SHIFT);
+
+	if (burst_ena)
+		hctl2 |= TV_BURST_ENA;
+
+	hctl3 = (tv_mode->hblank_start << TV_HBLANK_START_SHIFT) |
+		(tv_mode->hblank_end << TV_HBLANK_END_SHIFT);
+
+	vctl1 = (tv_mode->nbr_end << TV_NBR_END_SHIFT) |
+		(tv_mode->vi_end_f1 << TV_VI_END_F1_SHIFT) |
+		(tv_mode->vi_end_f2 << TV_VI_END_F2_SHIFT);
+
+	vctl2 = (tv_mode->vsync_len << TV_VSYNC_LEN_SHIFT) |
+		(tv_mode->vsync_start_f1 << TV_VSYNC_START_F1_SHIFT) |
+		(tv_mode->vsync_start_f2 << TV_VSYNC_START_F2_SHIFT);
+
+	vctl3 = (tv_mode->veq_len << TV_VEQ_LEN_SHIFT) |
+		(tv_mode->veq_start_f1 << TV_VEQ_START_F1_SHIFT) |
+		(tv_mode->veq_start_f2 << TV_VEQ_START_F2_SHIFT);
+
+	if (tv_mode->veq_ena)
+		vctl3 |= TV_EQUAL_ENA;
+
+	vctl4 = (tv_mode->vburst_start_f1 << TV_VBURST_START_F1_SHIFT) |
+		(tv_mode->vburst_end_f1 << TV_VBURST_END_F1_SHIFT);
+
+	vctl5 = (tv_mode->vburst_start_f2 << TV_VBURST_START_F2_SHIFT) |
+		(tv_mode->vburst_end_f2 << TV_VBURST_END_F2_SHIFT);
+
+	vctl6 = (tv_mode->vburst_start_f3 << TV_VBURST_START_F3_SHIFT) |
+		(tv_mode->vburst_end_f3 << TV_VBURST_END_F3_SHIFT);
+
+	vctl7 = (tv_mode->vburst_start_f4 << TV_VBURST_START_F4_SHIFT) |
+		(tv_mode->vburst_end_f4 << TV_VBURST_END_F4_SHIFT);
+
+	I915_WRITE(TV_H_CTL_1, hctl1);
+	I915_WRITE(TV_H_CTL_2, hctl2);
+	I915_WRITE(TV_H_CTL_3, hctl3);
+	I915_WRITE(TV_V_CTL_1, vctl1);
+	I915_WRITE(TV_V_CTL_2, vctl2);
+	I915_WRITE(TV_V_CTL_3, vctl3);
+	I915_WRITE(TV_V_CTL_4, vctl4);
+	I915_WRITE(TV_V_CTL_5, vctl5);
+	I915_WRITE(TV_V_CTL_6, vctl6);
+	I915_WRITE(TV_V_CTL_7, vctl7);
+}
+
+static void set_color_conversion(struct drm_i915_private *dev_priv,
+				 const struct color_conversion *color_conversion)
+{
+	if (!color_conversion)
+		return;
+
+	I915_WRITE(TV_CSC_Y, (color_conversion->ry << 16) |
+		   color_conversion->gy);
+	I915_WRITE(TV_CSC_Y2, (color_conversion->by << 16) |
+		   color_conversion->ay);
+	I915_WRITE(TV_CSC_U, (color_conversion->ru << 16) |
+		   color_conversion->gu);
+	I915_WRITE(TV_CSC_U2, (color_conversion->bu << 16) |
+		   color_conversion->au);
+	I915_WRITE(TV_CSC_V, (color_conversion->rv << 16) |
+		   color_conversion->gv);
+	I915_WRITE(TV_CSC_V2, (color_conversion->bv << 16) |
+		   color_conversion->av);
+}
+
+static void intel_tv_pre_enable(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -942,14 +1021,13 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 	struct intel_tv *intel_tv = enc_to_tv(encoder);
 	const struct tv_mode *tv_mode = intel_tv_mode_find(intel_tv);
 	u32 tv_ctl;
-	u32 hctl1, hctl2, hctl3;
-	u32 vctl1, vctl2, vctl3, vctl4, vctl5, vctl6, vctl7;
 	u32 scctl1, scctl2, scctl3;
 	int i, j;
 	const struct video_levels *video_levels;
 	const struct color_conversion *color_conversion;
 	bool burst_ena;
-	int pipe = intel_crtc->pipe;
+	int xpos = 0x0, ypos = 0x0;
+	unsigned int xsize, ysize;
 
 	if (!tv_mode)
 		return;	/* can't happen (mode_prepare prevents this) */
@@ -982,44 +1060,6 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		burst_ena = tv_mode->burst_ena;
 		break;
 	}
-	hctl1 = (tv_mode->hsync_end << TV_HSYNC_END_SHIFT) |
-		(tv_mode->htotal << TV_HTOTAL_SHIFT);
-
-	hctl2 = (tv_mode->hburst_start << 16) |
-		(tv_mode->hburst_len << TV_HBURST_LEN_SHIFT);
-
-	if (burst_ena)
-		hctl2 |= TV_BURST_ENA;
-
-	hctl3 = (tv_mode->hblank_start << TV_HBLANK_START_SHIFT) |
-		(tv_mode->hblank_end << TV_HBLANK_END_SHIFT);
-
-	vctl1 = (tv_mode->nbr_end << TV_NBR_END_SHIFT) |
-		(tv_mode->vi_end_f1 << TV_VI_END_F1_SHIFT) |
-		(tv_mode->vi_end_f2 << TV_VI_END_F2_SHIFT);
-
-	vctl2 = (tv_mode->vsync_len << TV_VSYNC_LEN_SHIFT) |
-		(tv_mode->vsync_start_f1 << TV_VSYNC_START_F1_SHIFT) |
-		(tv_mode->vsync_start_f2 << TV_VSYNC_START_F2_SHIFT);
-
-	vctl3 = (tv_mode->veq_len << TV_VEQ_LEN_SHIFT) |
-		(tv_mode->veq_start_f1 << TV_VEQ_START_F1_SHIFT) |
-		(tv_mode->veq_start_f2 << TV_VEQ_START_F2_SHIFT);
-
-	if (tv_mode->veq_ena)
-		vctl3 |= TV_EQUAL_ENA;
-
-	vctl4 = (tv_mode->vburst_start_f1 << TV_VBURST_START_F1_SHIFT) |
-		(tv_mode->vburst_end_f1 << TV_VBURST_END_F1_SHIFT);
-
-	vctl5 = (tv_mode->vburst_start_f2 << TV_VBURST_START_F2_SHIFT) |
-		(tv_mode->vburst_end_f2 << TV_VBURST_END_F2_SHIFT);
-
-	vctl6 = (tv_mode->vburst_start_f3 << TV_VBURST_START_F3_SHIFT) |
-		(tv_mode->vburst_end_f3 << TV_VBURST_END_F3_SHIFT);
-
-	vctl7 = (tv_mode->vburst_start_f4 << TV_VBURST_START_F4_SHIFT) |
-		(tv_mode->vburst_end_f4 << TV_VBURST_END_F4_SHIFT);
 
 	if (intel_crtc->pipe == 1)
 		tv_ctl |= TV_ENC_PIPEB_SELECT;
@@ -1051,37 +1091,16 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		tv_mode->dda3_inc << TV_SCDDA3_INC_SHIFT;
 
 	/* Enable two fixes for the chips that need them. */
-	if (dev->pdev->device < 0x2772)
+	if (IS_I915GM(dev))
 		tv_ctl |= TV_ENC_C0_FIX | TV_ENC_SDP_FIX;
 
-	I915_WRITE(TV_H_CTL_1, hctl1);
-	I915_WRITE(TV_H_CTL_2, hctl2);
-	I915_WRITE(TV_H_CTL_3, hctl3);
-	I915_WRITE(TV_V_CTL_1, vctl1);
-	I915_WRITE(TV_V_CTL_2, vctl2);
-	I915_WRITE(TV_V_CTL_3, vctl3);
-	I915_WRITE(TV_V_CTL_4, vctl4);
-	I915_WRITE(TV_V_CTL_5, vctl5);
-	I915_WRITE(TV_V_CTL_6, vctl6);
-	I915_WRITE(TV_V_CTL_7, vctl7);
+	set_tv_mode_timings(dev_priv, tv_mode, burst_ena);
+
 	I915_WRITE(TV_SC_CTL_1, scctl1);
 	I915_WRITE(TV_SC_CTL_2, scctl2);
 	I915_WRITE(TV_SC_CTL_3, scctl3);
 
-	if (color_conversion) {
-		I915_WRITE(TV_CSC_Y, (color_conversion->ry << 16) |
-			   color_conversion->gy);
-		I915_WRITE(TV_CSC_Y2, (color_conversion->by << 16) |
-			   color_conversion->ay);
-		I915_WRITE(TV_CSC_U, (color_conversion->ru << 16) |
-			   color_conversion->gu);
-		I915_WRITE(TV_CSC_U2, (color_conversion->bu << 16) |
-			   color_conversion->au);
-		I915_WRITE(TV_CSC_V, (color_conversion->rv << 16) |
-			   color_conversion->gv);
-		I915_WRITE(TV_CSC_V2, (color_conversion->bv << 16) |
-			   color_conversion->av);
-	}
+	set_color_conversion(dev_priv, color_conversion);
 
 	if (INTEL_INFO(dev)->gen >= 4)
 		I915_WRITE(TV_CLR_KNOBS, 0x00404000);
@@ -1092,46 +1111,25 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		I915_WRITE(TV_CLR_LEVEL,
 			   ((video_levels->black << TV_BLACK_LEVEL_SHIFT) |
 			    (video_levels->blank << TV_BLANK_LEVEL_SHIFT)));
-	{
-		int pipeconf_reg = PIPECONF(pipe);
-		int dspcntr_reg = DSPCNTR(intel_crtc->plane);
-		int pipeconf = I915_READ(pipeconf_reg);
-		int dspcntr = I915_READ(dspcntr_reg);
-		int xpos = 0x0, ypos = 0x0;
-		unsigned int xsize, ysize;
-		/* Pipe must be off here */
-		I915_WRITE(dspcntr_reg, dspcntr & ~DISPLAY_PLANE_ENABLE);
-		intel_flush_primary_plane(dev_priv, intel_crtc->plane);
-
-		/* Wait for vblank for the disable to take effect */
-		if (IS_GEN2(dev))
-			intel_wait_for_vblank(dev, intel_crtc->pipe);
-
-		I915_WRITE(pipeconf_reg, pipeconf & ~PIPECONF_ENABLE);
-		/* Wait for vblank for the disable to take effect. */
-		intel_wait_for_pipe_off(dev, intel_crtc->pipe);
-
-		/* Filter ctl must be set before TV_WIN_SIZE */
-		I915_WRITE(TV_FILTER_CTL_1, TV_AUTO_SCALE);
-		xsize = tv_mode->hblank_start - tv_mode->hblank_end;
-		if (tv_mode->progressive)
-			ysize = tv_mode->nbr_end + 1;
-		else
-			ysize = 2*tv_mode->nbr_end + 1;
-
-		xpos += intel_tv->margin[TV_MARGIN_LEFT];
-		ypos += intel_tv->margin[TV_MARGIN_TOP];
-		xsize -= (intel_tv->margin[TV_MARGIN_LEFT] +
-			  intel_tv->margin[TV_MARGIN_RIGHT]);
-		ysize -= (intel_tv->margin[TV_MARGIN_TOP] +
-			  intel_tv->margin[TV_MARGIN_BOTTOM]);
-		I915_WRITE(TV_WIN_POS, (xpos<<16)|ypos);
-		I915_WRITE(TV_WIN_SIZE, (xsize<<16)|ysize);
-
-		I915_WRITE(pipeconf_reg, pipeconf);
-		I915_WRITE(dspcntr_reg, dspcntr);
-		intel_flush_primary_plane(dev_priv, intel_crtc->plane);
-	}
+
+	assert_pipe_disabled(dev_priv, intel_crtc->pipe);
+
+	/* Filter ctl must be set before TV_WIN_SIZE */
+	I915_WRITE(TV_FILTER_CTL_1, TV_AUTO_SCALE);
+	xsize = tv_mode->hblank_start - tv_mode->hblank_end;
+	if (tv_mode->progressive)
+		ysize = tv_mode->nbr_end + 1;
+	else
+		ysize = 2*tv_mode->nbr_end + 1;
+
+	xpos += intel_tv->margin[TV_MARGIN_LEFT];
+	ypos += intel_tv->margin[TV_MARGIN_TOP];
+	xsize -= (intel_tv->margin[TV_MARGIN_LEFT] +
+		  intel_tv->margin[TV_MARGIN_RIGHT]);
+	ysize -= (intel_tv->margin[TV_MARGIN_TOP] +
+		  intel_tv->margin[TV_MARGIN_BOTTOM]);
+	I915_WRITE(TV_WIN_POS, (xpos<<16)|ypos);
+	I915_WRITE(TV_WIN_SIZE, (xsize<<16)|ysize);
 
 	j = 0;
 	for (i = 0; i < 60; i++)
@@ -1316,17 +1314,18 @@ intel_tv_detect(struct drm_connector *connector, bool force)
 	int type;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] force=%d\n",
-		      connector->base.id, drm_get_connector_name(connector),
+		      connector->base.id, connector->name,
 		      force);
 
 	mode = reported_modes[0];
 
 	if (force) {
 		struct intel_load_detect_pipe tmp;
+		struct drm_modeset_acquire_ctx ctx;
 
-		if (intel_get_load_detect_pipe(connector, &mode, &tmp)) {
+		if (intel_get_load_detect_pipe(connector, &mode, &tmp, &ctx)) {
 			type = intel_tv_detect_type(intel_tv, connector);
-			intel_release_load_detect_pipe(connector, &tmp);
+			intel_release_load_detect_pipe(connector, &tmp, &ctx);
 		} else
 			return connector_status_unknown;
 	} else
@@ -1634,7 +1633,7 @@ intel_tv_init(struct drm_device *dev)
 
 	intel_encoder->compute_config = intel_tv_compute_config;
 	intel_encoder->get_config = intel_tv_get_config;
-	intel_encoder->mode_set = intel_tv_mode_set;
+	intel_encoder->pre_enable = intel_tv_pre_enable;
 	intel_encoder->enable = intel_enable_tv;
 	intel_encoder->disable = intel_disable_tv;
 	intel_encoder->get_hw_state = intel_tv_get_hw_state;
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index d0c75779d3f6..4f6fef7ac069 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -255,8 +255,7 @@ static void __vlv_force_wake_put(struct drm_i915_private *dev_priv,
 
 }
 
-void vlv_force_wake_get(struct drm_i915_private *dev_priv,
-						int fw_engine)
+static void vlv_force_wake_get(struct drm_i915_private *dev_priv, int fw_engine)
 {
 	unsigned long irqflags;
 
@@ -275,8 +274,7 @@ void vlv_force_wake_get(struct drm_i915_private *dev_priv,
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
-void vlv_force_wake_put(struct drm_i915_private *dev_priv,
-						int fw_engine)
+static void vlv_force_wake_put(struct drm_i915_private *dev_priv, int fw_engine)
 {
 	unsigned long irqflags;
 
@@ -322,7 +320,8 @@ static void intel_uncore_forcewake_reset(struct drm_device *dev, bool restore)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
 
-	del_timer_sync(&dev_priv->uncore.force_wake_timer);
+	if (del_timer_sync(&dev_priv->uncore.force_wake_timer))
+		gen6_force_wake_timer((unsigned long)dev_priv);
 
 	/* Hold uncore.lock across reset to prevent any register access
 	 * with forcewake not set correctly
@@ -374,7 +373,7 @@ void intel_uncore_early_sanitize(struct drm_device *dev)
 	if (HAS_FPGA_DBG_UNCLAIMED(dev))
 		__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
 
-	if (IS_HASWELL(dev) &&
+	if ((IS_HASWELL(dev) || IS_BROADWELL(dev)) &&
 	    (__raw_i915_read32(dev_priv, HSW_EDRAM_PRESENT) == 1)) {
 		/* The docs do not explain exactly how the calculation can be
 		 * made. It is somewhat guessable, but for now, it's always
@@ -395,26 +394,8 @@ void intel_uncore_early_sanitize(struct drm_device *dev)
 
 void intel_uncore_sanitize(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 reg_val;
-
 	/* BIOS often leaves RC6 enabled, but disable it for hw init */
 	intel_disable_gt_powersave(dev);
-
-	/* Turn off power gate, require especially for the BIOS less system */
-	if (IS_VALLEYVIEW(dev)) {
-
-		mutex_lock(&dev_priv->rps.hw_lock);
-		reg_val = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS);
-
-		if (reg_val & (PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_RENDER) |
-			       PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_MEDIA) |
-			       PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_DISP2D)))
-			vlv_punit_write(dev_priv, PUNIT_REG_PWRGT_CTRL, 0x0);
-
-		mutex_unlock(&dev_priv->rps.hw_lock);
-
-	}
 }
 
 /*
@@ -488,6 +469,17 @@ void assert_force_wake_inactive(struct drm_i915_private *dev_priv)
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
 	 ((reg) < 0x40000 && (reg) != FORCEWAKE)
 
+#define FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg) \
+	(((reg) >= 0x2000 && (reg) < 0x4000) ||\
+	((reg) >= 0x5000 && (reg) < 0x8000) ||\
+	((reg) >= 0xB000 && (reg) < 0x12000) ||\
+	((reg) >= 0x2E000 && (reg) < 0x30000))
+
+#define FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(reg)\
+	(((reg) >= 0x12000 && (reg) < 0x14000) ||\
+	((reg) >= 0x22000 && (reg) < 0x24000) ||\
+	((reg) >= 0x30000 && (reg) < 0x40000))
+
 static void
 ilk_dummy_write(struct drm_i915_private *dev_priv)
 {
@@ -854,12 +846,15 @@ void intel_uncore_fini(struct drm_device *dev)
 	intel_uncore_forcewake_reset(dev, false);
 }
 
+#define GEN_RANGE(l, h) GENMASK(h, l)
+
 static const struct register_whitelist {
 	uint64_t offset;
 	uint32_t size;
-	uint32_t gen_bitmask; /* support gens, 0x10 for 4, 0x30 for 4 and 5, etc. */
+	/* supported gens, 0x10 for 4, 0x30 for 4 and 5, etc. */
+	uint32_t gen_bitmask;
 } whitelist[] = {
-	{ RING_TIMESTAMP(RENDER_RING_BASE), 8, 0x1F0 },
+	{ RING_TIMESTAMP(RENDER_RING_BASE), 8, GEN_RANGE(4, 8) },
 };
 
 int i915_reg_read_ioctl(struct drm_device *dev,
@@ -911,7 +906,7 @@ int i915_get_reset_stats_ioctl(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_reset_stats *args = data;
 	struct i915_ctx_hang_stats *hs;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	if (args->flags || args->pad)
@@ -955,6 +950,9 @@ static int i965_do_reset(struct drm_device *dev)
 {
 	int ret;
 
+	/* FIXME: i965g/gm need a display save/restore for gpu reset. */
+	return -ENODEV;
+
 	/*
 	 * Set the domains we want to reset (GRDOM/bits 2 and 3) as
 	 * well as the reset bit (GR/bit 0).  Setting the GR bit
@@ -966,7 +964,6 @@ static int i965_do_reset(struct drm_device *dev)
 	if (ret)
 		return ret;
 
-	/* We can't reset render&media without also resetting display ... */
 	pci_write_config_byte(dev->pdev, I965_GDRST,
 			      GRDOM_MEDIA | GRDOM_RESET_ENABLE);
 
@@ -979,26 +976,58 @@ static int i965_do_reset(struct drm_device *dev)
 	return 0;
 }
 
+static int g4x_do_reset(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	pci_write_config_byte(dev->pdev, I965_GDRST,
+			      GRDOM_RENDER | GRDOM_RESET_ENABLE);
+	ret =  wait_for(i965_reset_complete(dev), 500);
+	if (ret)
+		return ret;
+
+	/* WaVcpClkGateDisableForMediaReset:ctg,elk */
+	I915_WRITE(VDECCLK_GATE_D, I915_READ(VDECCLK_GATE_D) | VCP_UNIT_CLOCK_GATE_DISABLE);
+	POSTING_READ(VDECCLK_GATE_D);
+
+	pci_write_config_byte(dev->pdev, I965_GDRST,
+			      GRDOM_MEDIA | GRDOM_RESET_ENABLE);
+	ret =  wait_for(i965_reset_complete(dev), 500);
+	if (ret)
+		return ret;
+
+	/* WaVcpClkGateDisableForMediaReset:ctg,elk */
+	I915_WRITE(VDECCLK_GATE_D, I915_READ(VDECCLK_GATE_D) & ~VCP_UNIT_CLOCK_GATE_DISABLE);
+	POSTING_READ(VDECCLK_GATE_D);
+
+	pci_write_config_byte(dev->pdev, I965_GDRST, 0);
+
+	return 0;
+}
+
 static int ironlake_do_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 gdrst;
 	int ret;
 
-	gdrst = I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR);
-	gdrst &= ~GRDOM_MASK;
 	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR,
-		   gdrst | GRDOM_RENDER | GRDOM_RESET_ENABLE);
-	ret = wait_for(I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) & 0x1, 500);
+		   ILK_GRDOM_RENDER | ILK_GRDOM_RESET_ENABLE);
+	ret = wait_for((I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) &
+			ILK_GRDOM_RESET_ENABLE) == 0, 500);
 	if (ret)
 		return ret;
 
-	/* We can't reset render&media without also resetting display ... */
-	gdrst = I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR);
-	gdrst &= ~GRDOM_MASK;
 	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR,
-		   gdrst | GRDOM_MEDIA | GRDOM_RESET_ENABLE);
-	return wait_for(I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) & 0x1, 500);
+		   ILK_GRDOM_MEDIA | ILK_GRDOM_RESET_ENABLE);
+	ret = wait_for((I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) &
+			ILK_GRDOM_RESET_ENABLE) == 0, 500);
+	if (ret)
+		return ret;
+
+	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR, 0);
+
+	return 0;
 }
 
 static int gen6_do_reset(struct drm_device *dev)
@@ -1029,7 +1058,11 @@ int intel_gpu_reset(struct drm_device *dev)
 	case 7:
 	case 6: return gen6_do_reset(dev);
 	case 5: return ironlake_do_reset(dev);
-	case 4: return i965_do_reset(dev);
+	case 4:
+		if (IS_G4X(dev))
+			return g4x_do_reset(dev);
+		else
+			return i965_do_reset(dev);
 	default: return -ENODEV;
 	}
 }
diff --git a/drivers/gpu/drm/mga/mga_ioc32.c b/drivers/gpu/drm/mga/mga_ioc32.c
index 86b4bb804852..729bfd56b55f 100644
--- a/drivers/gpu/drm/mga/mga_ioc32.c
+++ b/drivers/gpu/drm/mga/mga_ioc32.c
@@ -214,7 +214,7 @@ long mga_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(mga_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(mga_compat_ioctls))
 		fn = mga_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/mga/mga_state.c b/drivers/gpu/drm/mga/mga_state.c
index 314685b7f41f..792f924496fc 100644
--- a/drivers/gpu/drm/mga/mga_state.c
+++ b/drivers/gpu/drm/mga/mga_state.c
@@ -1020,7 +1020,7 @@ static int mga_getparam(struct drm_device *dev, void *data, struct drm_file *fil
 
 	switch (param->param) {
 	case MGA_PARAM_IRQ_NR:
-		value = drm_dev_to_irq(dev);
+		value = dev->pdev->irq;
 		break;
 	case MGA_PARAM_CARD_TYPE:
 		value = dev_priv->chipset;
@@ -1099,4 +1099,4 @@ const struct drm_ioctl_desc mga_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(MGA_DMA_BOOTSTRAP, mga_dma_bootstrap, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 };
 
-int mga_max_ioctl = DRM_ARRAY_SIZE(mga_ioctls);
+int mga_max_ioctl = ARRAY_SIZE(mga_ioctls);
diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c
index 26868e5c55b0..f6b283b8375e 100644
--- a/drivers/gpu/drm/mgag200/mgag200_main.c
+++ b/drivers/gpu/drm/mgag200/mgag200_main.c
@@ -322,17 +322,13 @@ static void mgag200_bo_unref(struct mgag200_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
 
 void mgag200_gem_free_object(struct drm_gem_object *obj)
 {
 	struct mgag200_bo *mgag200_bo = gem_to_mga_bo(obj);
 
-	if (!mgag200_bo)
-		return;
 	mgag200_bo_unref(&mgag200_bo);
 }
 
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index b6984971ce0c..f12388967856 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -3,7 +3,7 @@ config DRM_MSM
 	tristate "MSM DRM"
 	depends on DRM
 	depends on MSM_IOMMU
-	depends on ARCH_MSM8960 || (ARM && COMPILE_TEST)
+	depends on ARCH_QCOM || (ARM && COMPILE_TEST)
 	select DRM_KMS_HELPER
 	select SHMEM
 	select TMPFS
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 5e1e6b0cd8ac..93ca49c8df44 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -34,6 +34,8 @@ msm-y := \
 	msm_gem_submit.o \
 	msm_gpu.o \
 	msm_iommu.o \
+	msm_perf.o \
+	msm_rd.o \
 	msm_ringbuffer.o
 
 msm-$(CONFIG_DRM_MSM_FBDEV) += msm_fbdev.o
diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
index f20fbde5dc49..942e09d898a8 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
@@ -207,11 +207,11 @@ static int a3xx_hw_init(struct msm_gpu *gpu)
 	/* Turn on performance counters: */
 	gpu_write(gpu, REG_A3XX_RBBM_PERFCTR_CTL, 0x01);
 
-	/* Set SP perfcounter 7 to count SP_FS_FULL_ALU_INSTRUCTIONS
-	 * we will use this to augment our hang detection:
-	 */
-	gpu_write(gpu, REG_A3XX_SP_PERFCOUNTER7_SELECT,
-			SP_FS_FULL_ALU_INSTRUCTIONS);
+	/* Enable the perfcntrs that we use.. */
+	for (i = 0; i < gpu->num_perfcntrs; i++) {
+		const struct msm_gpu_perfcntr *perfcntr = &gpu->perfcntrs[i];
+		gpu_write(gpu, perfcntr->select_reg, perfcntr->select_val);
+	}
 
 	gpu_write(gpu, REG_A3XX_RBBM_INT_0_MASK, A3XX_INT0_MASK);
 
@@ -465,6 +465,13 @@ static const struct adreno_gpu_funcs funcs = {
 	},
 };
 
+static const struct msm_gpu_perfcntr perfcntrs[] = {
+	{ REG_A3XX_SP_PERFCOUNTER6_SELECT, REG_A3XX_RBBM_PERFCTR_SP_6_LO,
+			SP_ALU_ACTIVE_CYCLES, "ALUACTIVE" },
+	{ REG_A3XX_SP_PERFCOUNTER7_SELECT, REG_A3XX_RBBM_PERFCTR_SP_7_LO,
+			SP_FS_FULL_ALU_INSTRUCTIONS, "ALUFULL" },
+};
+
 struct msm_gpu *a3xx_gpu_init(struct drm_device *dev)
 {
 	struct a3xx_gpu *a3xx_gpu = NULL;
@@ -504,6 +511,9 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev)
 	DBG("fast_rate=%u, slow_rate=%u, bus_freq=%u",
 			gpu->fast_rate, gpu->slow_rate, gpu->bus_freq);
 
+	gpu->perfcntrs = perfcntrs;
+	gpu->num_perfcntrs = ARRAY_SIZE(perfcntrs);
+
 	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, config->rev);
 	if (ret)
 		goto fail;
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index ae750f6928c1..7f7aadef8a82 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -277,6 +277,7 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 	static const char *hpd_reg_names[] = {"hpd-gdsc", "hpd-5v"};
 	static const char *pwr_reg_names[] = {"core-vdda", "core-vcc"};
 	static const char *hpd_clk_names[] = {"iface_clk", "core_clk", "mdp_core_clk"};
+	static unsigned long hpd_clk_freq[] = {0, 19200000, 0};
 	static const char *pwr_clk_names[] = {"extp_clk", "alt_iface_clk"};
 
 	config.phy_init      = hdmi_phy_8x74_init;
@@ -286,6 +287,7 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 	config.pwr_reg_names = pwr_reg_names;
 	config.pwr_reg_cnt   = ARRAY_SIZE(pwr_reg_names);
 	config.hpd_clk_names = hpd_clk_names;
+	config.hpd_freq      = hpd_clk_freq;
 	config.hpd_clk_cnt   = ARRAY_SIZE(hpd_clk_names);
 	config.pwr_clk_names = pwr_clk_names;
 	config.pwr_clk_cnt   = ARRAY_SIZE(pwr_clk_names);
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.h b/drivers/gpu/drm/msm/hdmi/hdmi.h
index 9fafee6a3e43..9d7723c6528a 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.h
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.h
@@ -87,6 +87,7 @@ struct hdmi_platform_config {
 
 	/* clks that need to be on for hpd: */
 	const char **hpd_clk_names;
+	const long unsigned *hpd_freq;
 	int hpd_clk_cnt;
 
 	/* clks that need to be on for screen pwr (ie pixel clk): */
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
index 7dedfdd12075..28f7e3ec6c28 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
@@ -127,6 +127,14 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector)
 	}
 
 	for (i = 0; i < config->hpd_clk_cnt; i++) {
+		if (config->hpd_freq && config->hpd_freq[i]) {
+			ret = clk_set_rate(hdmi->hpd_clks[i],
+					config->hpd_freq[i]);
+			if (ret)
+				dev_warn(dev->dev, "failed to set clk %s (%d)\n",
+						config->hpd_clk_names[i], ret);
+		}
+
 		ret = clk_prepare_enable(hdmi->hpd_clks[i]);
 		if (ret) {
 			dev_err(dev->dev, "failed to enable hpd clk: %s (%d)\n",
@@ -247,36 +255,49 @@ void hdmi_connector_irq(struct drm_connector *connector)
 	}
 }
 
+static enum drm_connector_status detect_reg(struct hdmi *hdmi)
+{
+	uint32_t hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
+	return (hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED) ?
+			connector_status_connected : connector_status_disconnected;
+}
+
+static enum drm_connector_status detect_gpio(struct hdmi *hdmi)
+{
+	const struct hdmi_platform_config *config = hdmi->config;
+	return gpio_get_value(config->hpd_gpio) ?
+			connector_status_connected :
+			connector_status_disconnected;
+}
+
 static enum drm_connector_status hdmi_connector_detect(
 		struct drm_connector *connector, bool force)
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
 	struct hdmi *hdmi = hdmi_connector->hdmi;
-	const struct hdmi_platform_config *config = hdmi->config;
-	uint32_t hpd_int_status;
+	enum drm_connector_status stat_gpio, stat_reg;
 	int retry = 20;
 
-	hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
+	do {
+		stat_gpio = detect_gpio(hdmi);
+		stat_reg  = detect_reg(hdmi);
 
-	/* sense seems to in some cases be momentarily de-asserted, don't
-	 * let that trick us into thinking the monitor is gone:
-	 */
-	while (retry-- && !(hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED)) {
-		/* hdmi debounce logic seems to get stuck sometimes,
-		 * read directly the gpio to get a second opinion:
-		 */
-		if (gpio_get_value(config->hpd_gpio)) {
-			DBG("gpio tells us we are connected!");
-			hpd_int_status |= HDMI_HPD_INT_STATUS_CABLE_DETECTED;
+		if (stat_gpio == stat_reg)
 			break;
-		}
+
 		mdelay(10);
-		hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
-		DBG("status=%08x", hpd_int_status);
+	} while (--retry);
+
+	/* the status we get from reading gpio seems to be more reliable,
+	 * so trust that one the most if we didn't manage to get hdmi and
+	 * gpio status to agree:
+	 */
+	if (stat_gpio != stat_reg) {
+		DBG("HDMI_HPD_INT_STATUS tells us: %d", stat_reg);
+		DBG("hpd gpio tells us: %d", stat_gpio);
 	}
 
-	return (hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED) ?
-			connector_status_connected : connector_status_disconnected;
+	return stat_gpio;
 }
 
 static void hdmi_connector_destroy(struct drm_connector *connector)
@@ -389,7 +410,8 @@ struct drm_connector *hdmi_connector_init(struct hdmi *hdmi)
 			DRM_MODE_CONNECTOR_HDMIA);
 	drm_connector_helper_add(connector, &hdmi_connector_helper_funcs);
 
-	connector->polled = DRM_CONNECTOR_POLL_HPD;
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT |
+			DRM_CONNECTOR_POLL_DISCONNECT;
 
 	connector->interlace_allowed = 1;
 	connector->doublescan_allowed = 0;
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
index ef9957dbac94..74cebb51e8c2 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
@@ -217,8 +217,6 @@ static void mdp4_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
 
-	mdp4_crtc->plane->funcs->destroy(mdp4_crtc->plane);
-
 	drm_crtc_cleanup(crtc);
 	drm_flip_work_cleanup(&mdp4_crtc->unref_fb_work);
 	drm_flip_work_cleanup(&mdp4_crtc->unref_cursor_work);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
index 6ea10bdb6e8f..ebe2e60f3ab1 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
@@ -195,8 +195,6 @@ static void mdp5_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc);
 
-	mdp5_crtc->plane->funcs->destroy(mdp5_crtc->plane);
-
 	drm_crtc_cleanup(crtc);
 	drm_flip_work_cleanup(&mdp5_crtc->unref_fb_work);
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index ee8446c1b5f6..71510ee26e96 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -20,6 +20,10 @@
 #include "msm_mmu.h"
 #include "mdp5_kms.h"
 
+static const char *iommu_ports[] = {
+		"mdp_0",
+};
+
 static struct mdp5_platform_config *mdp5_get_config(struct platform_device *dev);
 
 static int mdp5_hw_init(struct msm_kms *kms)
@@ -104,6 +108,12 @@ static void mdp5_preclose(struct msm_kms *kms, struct drm_file *file)
 static void mdp5_destroy(struct msm_kms *kms)
 {
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	struct msm_mmu *mmu = mdp5_kms->mmu;
+
+	if (mmu) {
+		mmu->funcs->detach(mmu, iommu_ports, ARRAY_SIZE(iommu_ports));
+		mmu->funcs->destroy(mmu);
+	}
 	kfree(mdp5_kms);
 }
 
@@ -216,10 +226,6 @@ fail:
 	return ret;
 }
 
-static const char *iommu_ports[] = {
-		"mdp_0",
-};
-
 static int get_clk(struct platform_device *pdev, struct clk **clkp,
 		const char *name)
 {
@@ -280,12 +286,22 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 		goto fail;
 	}
 
-	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk") ||
-			get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk") ||
-			get_clk(pdev, &mdp5_kms->src_clk, "core_clk_src") ||
-			get_clk(pdev, &mdp5_kms->core_clk, "core_clk") ||
-			get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk") ||
-			get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk");
+	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->src_clk, "core_clk_src");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->core_clk, "core_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk");
 	if (ret)
 		goto fail;
 
@@ -307,17 +323,23 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 		mmu = msm_iommu_new(dev, config->iommu);
 		if (IS_ERR(mmu)) {
 			ret = PTR_ERR(mmu);
+			dev_err(dev->dev, "failed to init iommu: %d\n", ret);
 			goto fail;
 		}
+
 		ret = mmu->funcs->attach(mmu, iommu_ports,
 				ARRAY_SIZE(iommu_ports));
-		if (ret)
+		if (ret) {
+			dev_err(dev->dev, "failed to attach iommu: %d\n", ret);
+			mmu->funcs->destroy(mmu);
 			goto fail;
+		}
 	} else {
 		dev_info(dev->dev, "no iommu, fallback to phys "
 				"contig buffers for scanout\n");
 		mmu = NULL;
 	}
+	mdp5_kms->mmu = mmu;
 
 	mdp5_kms->id = msm_register_mmu(dev, mmu);
 	if (mdp5_kms->id < 0) {
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
index c8b1a2522c25..6e981b692d1d 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
@@ -33,6 +33,7 @@ struct mdp5_kms {
 
 	/* mapper-id used to request GEM buffer mapped for scanout: */
 	int id;
+	struct msm_mmu *mmu;
 
 	/* for tracking smp allocation amongst pipes: */
 	mdp5_smp_state_t smp_state;
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index 47f7bbb9c15a..f3daec4412ad 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -85,8 +85,11 @@ static int mdp5_plane_disable(struct drm_plane *plane)
 static void mdp5_plane_destroy(struct drm_plane *plane)
 {
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
+	struct msm_drm_private *priv = plane->dev->dev_private;
+
+	if (priv->kms)
+		mdp5_plane_disable(plane);
 
-	mdp5_plane_disable(plane);
 	drm_plane_cleanup(plane);
 
 	kfree(mdp5_plane);
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index f9de156b9e65..9a5d87db5c23 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -159,7 +159,7 @@ static int msm_unload(struct drm_device *dev)
 static int get_mdp_ver(struct platform_device *pdev)
 {
 #ifdef CONFIG_OF
-	const static struct of_device_id match_types[] = { {
+	static const struct of_device_id match_types[] = { {
 		.compatible = "qcom,mdss_mdp",
 		.data	= (void	*)5,
 	}, {
@@ -220,7 +220,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 		 * is bogus, but non-null if allocation succeeded:
 		 */
 		p = dma_alloc_attrs(dev->dev, size,
-				&priv->vram.paddr, 0, &attrs);
+				&priv->vram.paddr, GFP_KERNEL, &attrs);
 		if (!p) {
 			dev_err(dev->dev, "failed to allocate VRAM\n");
 			priv->vram.paddr = 0;
@@ -288,7 +288,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 	}
 
 	pm_runtime_get_sync(dev->dev);
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	pm_runtime_put_sync(dev->dev);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to install IRQ handler\n");
@@ -299,6 +299,10 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 	priv->fbdev = msm_fbdev_init(dev);
 #endif
 
+	ret = msm_debugfs_late_init(dev);
+	if (ret)
+		goto fail;
+
 	drm_kms_helper_poll_init(dev);
 
 	return 0;
@@ -382,11 +386,8 @@ static void msm_preclose(struct drm_device *dev, struct drm_file *file)
 static void msm_lastclose(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
-	if (priv->fbdev) {
-		drm_modeset_lock_all(dev);
-		drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-		drm_modeset_unlock_all(dev);
-	}
+	if (priv->fbdev)
+		drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 }
 
 static irqreturn_t msm_irq(int irq, void *arg)
@@ -531,6 +532,41 @@ static struct drm_info_list msm_debugfs_list[] = {
 		{ "fb", show_locked, 0, msm_fb_show },
 };
 
+static int late_init_minor(struct drm_minor *minor)
+{
+	int ret;
+
+	if (!minor)
+		return 0;
+
+	ret = msm_rd_debugfs_init(minor);
+	if (ret) {
+		dev_err(minor->dev->dev, "could not install rd debugfs\n");
+		return ret;
+	}
+
+	ret = msm_perf_debugfs_init(minor);
+	if (ret) {
+		dev_err(minor->dev->dev, "could not install perf debugfs\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+int msm_debugfs_late_init(struct drm_device *dev)
+{
+	int ret;
+	ret = late_init_minor(dev->primary);
+	if (ret)
+		return ret;
+	ret = late_init_minor(dev->render);
+	if (ret)
+		return ret;
+	ret = late_init_minor(dev->control);
+	return ret;
+}
+
 static int msm_debugfs_init(struct drm_minor *minor)
 {
 	struct drm_device *dev = minor->dev;
@@ -545,13 +581,17 @@ static int msm_debugfs_init(struct drm_minor *minor)
 		return ret;
 	}
 
-	return ret;
+	return 0;
 }
 
 static void msm_debugfs_cleanup(struct drm_minor *minor)
 {
 	drm_debugfs_remove_files(msm_debugfs_list,
 			ARRAY_SIZE(msm_debugfs_list), minor);
+	if (!minor->dev->dev_private)
+		return;
+	msm_rd_debugfs_cleanup(minor);
+	msm_perf_debugfs_cleanup(minor);
 }
 #endif
 
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 9d10ee0b5aac..8a2c5fd0893e 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -33,7 +33,7 @@
 #include <asm/sizes.h>
 
 
-#if defined(CONFIG_COMPILE_TEST) && !defined(CONFIG_ARCH_MSM)
+#if defined(CONFIG_COMPILE_TEST) && !defined(CONFIG_ARCH_QCOM)
 /* stubs we need for compile-test: */
 static inline struct device *msm_iommu_get_ctx(const char *ctx_name)
 {
@@ -55,6 +55,9 @@ static inline struct device *msm_iommu_get_ctx(const char *ctx_name)
 struct msm_kms;
 struct msm_gpu;
 struct msm_mmu;
+struct msm_rd_state;
+struct msm_perf_state;
+struct msm_gem_submit;
 
 #define NUM_DOMAINS 2    /* one for KMS, then one per gpu core (?) */
 
@@ -82,6 +85,9 @@ struct msm_drm_private {
 	uint32_t next_fence, completed_fence;
 	wait_queue_head_t fence_event;
 
+	struct msm_rd_state *rd;
+	struct msm_perf_state *perf;
+
 	/* list of GEM objects: */
 	struct list_head inactive_list;
 
@@ -204,6 +210,15 @@ void __exit hdmi_unregister(void);
 void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m);
 void msm_gem_describe_objects(struct list_head *list, struct seq_file *m);
 void msm_framebuffer_describe(struct drm_framebuffer *fb, struct seq_file *m);
+int msm_debugfs_late_init(struct drm_device *dev);
+int msm_rd_debugfs_init(struct drm_minor *minor);
+void msm_rd_debugfs_cleanup(struct drm_minor *minor);
+void msm_rd_dump_submit(struct msm_gem_submit *submit);
+int msm_perf_debugfs_init(struct drm_minor *minor);
+void msm_perf_debugfs_cleanup(struct drm_minor *minor);
+#else
+static inline int msm_debugfs_late_init(struct drm_device *dev) { return 0; }
+static inline void msm_rd_dump_submit(struct msm_gem_submit *submit) {}
 #endif
 
 void __iomem *msm_ioremap(struct platform_device *pdev, const char *name,
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index a752ab83b810..5107fc4826bc 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -59,7 +59,7 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	struct drm_framebuffer *fb = NULL;
 	struct fb_info *fbi = NULL;
 	struct drm_mode_fb_cmd2 mode_cmd = {0};
-	dma_addr_t paddr;
+	uint32_t paddr;
 	int ret, size;
 
 	sizes->surface_bpp = 32;
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index bb8026daebc9..690d7e7b6d1e 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -278,6 +278,7 @@ int msm_gem_get_iova_locked(struct drm_gem_object *obj, int id,
 		uint32_t *iova)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	struct drm_device *dev = obj->dev;
 	int ret = 0;
 
 	if (!msm_obj->domain[id].iova) {
@@ -285,6 +286,11 @@ int msm_gem_get_iova_locked(struct drm_gem_object *obj, int id,
 		struct msm_mmu *mmu = priv->mmus[id];
 		struct page **pages = get_pages(obj);
 
+		if (!mmu) {
+			dev_err(dev->dev, "null MMU pointer\n");
+			return -EINVAL;
+		}
+
 		if (IS_ERR(pages))
 			return PTR_ERR(pages);
 
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 3246bb46c4f2..bfb052688f8e 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -90,6 +90,7 @@ struct msm_gem_submit {
 		uint32_t type;
 		uint32_t size;  /* in dwords */
 		uint32_t iova;
+		uint32_t idx;   /* cmdstream buffer idx in bos[] */
 	} cmd[MAX_CMDS];
 	struct {
 		uint32_t flags;
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 1f1f4cffdaed..cd0554f68316 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -402,6 +402,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		submit->cmd[i].type = submit_cmd.type;
 		submit->cmd[i].size = submit_cmd.size / 4;
 		submit->cmd[i].iova = iova + submit_cmd.submit_offset;
+		submit->cmd[i].idx  = submit_cmd.submit_idx;
 
 		if (submit->valid)
 			continue;
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 3e667ca1f2b9..c6322197db8c 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -320,6 +320,101 @@ static void hangcheck_handler(unsigned long data)
 }
 
 /*
+ * Performance Counters:
+ */
+
+/* called under perf_lock */
+static int update_hw_cntrs(struct msm_gpu *gpu, uint32_t ncntrs, uint32_t *cntrs)
+{
+	uint32_t current_cntrs[ARRAY_SIZE(gpu->last_cntrs)];
+	int i, n = min(ncntrs, gpu->num_perfcntrs);
+
+	/* read current values: */
+	for (i = 0; i < gpu->num_perfcntrs; i++)
+		current_cntrs[i] = gpu_read(gpu, gpu->perfcntrs[i].sample_reg);
+
+	/* update cntrs: */
+	for (i = 0; i < n; i++)
+		cntrs[i] = current_cntrs[i] - gpu->last_cntrs[i];
+
+	/* save current values: */
+	for (i = 0; i < gpu->num_perfcntrs; i++)
+		gpu->last_cntrs[i] = current_cntrs[i];
+
+	return n;
+}
+
+static void update_sw_cntrs(struct msm_gpu *gpu)
+{
+	ktime_t time;
+	uint32_t elapsed;
+	unsigned long flags;
+
+	spin_lock_irqsave(&gpu->perf_lock, flags);
+	if (!gpu->perfcntr_active)
+		goto out;
+
+	time = ktime_get();
+	elapsed = ktime_to_us(ktime_sub(time, gpu->last_sample.time));
+
+	gpu->totaltime += elapsed;
+	if (gpu->last_sample.active)
+		gpu->activetime += elapsed;
+
+	gpu->last_sample.active = msm_gpu_active(gpu);
+	gpu->last_sample.time = time;
+
+out:
+	spin_unlock_irqrestore(&gpu->perf_lock, flags);
+}
+
+void msm_gpu_perfcntr_start(struct msm_gpu *gpu)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&gpu->perf_lock, flags);
+	/* we could dynamically enable/disable perfcntr registers too.. */
+	gpu->last_sample.active = msm_gpu_active(gpu);
+	gpu->last_sample.time = ktime_get();
+	gpu->activetime = gpu->totaltime = 0;
+	gpu->perfcntr_active = true;
+	update_hw_cntrs(gpu, 0, NULL);
+	spin_unlock_irqrestore(&gpu->perf_lock, flags);
+}
+
+void msm_gpu_perfcntr_stop(struct msm_gpu *gpu)
+{
+	gpu->perfcntr_active = false;
+}
+
+/* returns -errno or # of cntrs sampled */
+int msm_gpu_perfcntr_sample(struct msm_gpu *gpu, uint32_t *activetime,
+		uint32_t *totaltime, uint32_t ncntrs, uint32_t *cntrs)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&gpu->perf_lock, flags);
+
+	if (!gpu->perfcntr_active) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	*activetime = gpu->activetime;
+	*totaltime = gpu->totaltime;
+
+	gpu->activetime = gpu->totaltime = 0;
+
+	ret = update_hw_cntrs(gpu, ncntrs, cntrs);
+
+out:
+	spin_unlock_irqrestore(&gpu->perf_lock, flags);
+
+	return ret;
+}
+
+/*
  * Cmdstream submission/retirement:
  */
 
@@ -361,6 +456,7 @@ void msm_gpu_retire(struct msm_gpu *gpu)
 {
 	struct msm_drm_private *priv = gpu->dev->dev_private;
 	queue_work(priv->wq, &gpu->retire_work);
+	update_sw_cntrs(gpu);
 }
 
 /* add bo's to gpu's ring, and kick gpu: */
@@ -377,6 +473,12 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 
 	inactive_cancel(gpu);
 
+	msm_rd_dump_submit(submit);
+
+	gpu->submitted_fence = submit->fence;
+
+	update_sw_cntrs(gpu);
+
 	ret = gpu->funcs->submit(gpu, submit, ctx);
 	priv->lastctx = ctx;
 
@@ -429,6 +531,9 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	struct iommu_domain *iommu;
 	int i, ret;
 
+	if (WARN_ON(gpu->num_perfcntrs > ARRAY_SIZE(gpu->last_cntrs)))
+		gpu->num_perfcntrs = ARRAY_SIZE(gpu->last_cntrs);
+
 	gpu->dev = drm;
 	gpu->funcs = funcs;
 	gpu->name = name;
@@ -444,6 +549,8 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	setup_timer(&gpu->hangcheck_timer, hangcheck_handler,
 			(unsigned long)gpu);
 
+	spin_lock_init(&gpu->perf_lock);
+
 	BUG_ON(ARRAY_SIZE(clk_names) != ARRAY_SIZE(gpu->grp_clks));
 
 	/* Map registers: */
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index fad27008922f..9b579b792840 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -25,6 +25,7 @@
 #include "msm_ringbuffer.h"
 
 struct msm_gem_submit;
+struct msm_gpu_perfcntr;
 
 /* So far, with hardware that I've seen to date, we can have:
  *  + zero, one, or two z180 2d cores
@@ -64,6 +65,18 @@ struct msm_gpu {
 	struct drm_device *dev;
 	const struct msm_gpu_funcs *funcs;
 
+	/* performance counters (hw & sw): */
+	spinlock_t perf_lock;
+	bool perfcntr_active;
+	struct {
+		bool active;
+		ktime_t time;
+	} last_sample;
+	uint32_t totaltime, activetime;    /* sw counters */
+	uint32_t last_cntrs[5];            /* hw counters */
+	const struct msm_gpu_perfcntr *perfcntrs;
+	uint32_t num_perfcntrs;
+
 	struct msm_ringbuffer *rb;
 	uint32_t rb_iova;
 
@@ -113,6 +126,19 @@ static inline bool msm_gpu_active(struct msm_gpu *gpu)
 	return gpu->submitted_fence > gpu->funcs->last_fence(gpu);
 }
 
+/* Perf-Counters:
+ * The select_reg and select_val are just there for the benefit of the child
+ * class that actually enables the perf counter..  but msm_gpu base class
+ * will handle sampling/displaying the counters.
+ */
+
+struct msm_gpu_perfcntr {
+	uint32_t select_reg;
+	uint32_t sample_reg;
+	uint32_t select_val;
+	const char *name;
+};
+
 static inline void gpu_write(struct msm_gpu *gpu, u32 reg, u32 data)
 {
 	msm_writel(data, gpu->mmio + (reg << 2));
@@ -126,6 +152,11 @@ static inline u32 gpu_read(struct msm_gpu *gpu, u32 reg)
 int msm_gpu_pm_suspend(struct msm_gpu *gpu);
 int msm_gpu_pm_resume(struct msm_gpu *gpu);
 
+void msm_gpu_perfcntr_start(struct msm_gpu *gpu);
+void msm_gpu_perfcntr_stop(struct msm_gpu *gpu);
+int msm_gpu_perfcntr_sample(struct msm_gpu *gpu, uint32_t *activetime,
+		uint32_t *totaltime, uint32_t ncntrs, uint32_t *cntrs);
+
 void msm_gpu_retire(struct msm_gpu *gpu);
 int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		struct msm_file_private *ctx);
diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 92b745986231..4b2ad9181edf 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -28,7 +28,7 @@ static int msm_fault_handler(struct iommu_domain *iommu, struct device *dev,
 		unsigned long iova, int flags, void *arg)
 {
 	DBG("*** fault: iova=%08lx, flags=%d", iova, flags);
-	return 0;
+	return -ENOSYS;
 }
 
 static int msm_iommu_attach(struct msm_mmu *mmu, const char **names, int cnt)
@@ -40,8 +40,10 @@ static int msm_iommu_attach(struct msm_mmu *mmu, const char **names, int cnt)
 	for (i = 0; i < cnt; i++) {
 		struct device *msm_iommu_get_ctx(const char *ctx_name);
 		struct device *ctx = msm_iommu_get_ctx(names[i]);
-		if (IS_ERR_OR_NULL(ctx))
+		if (IS_ERR_OR_NULL(ctx)) {
+			dev_warn(dev->dev, "couldn't get %s context", names[i]);
 			continue;
+		}
 		ret = iommu_attach_device(iommu->domain, ctx);
 		if (ret) {
 			dev_warn(dev->dev, "could not attach iommu to %s", names[i]);
@@ -52,6 +54,20 @@ static int msm_iommu_attach(struct msm_mmu *mmu, const char **names, int cnt)
 	return 0;
 }
 
+static void msm_iommu_detach(struct msm_mmu *mmu, const char **names, int cnt)
+{
+	struct msm_iommu *iommu = to_msm_iommu(mmu);
+	int i;
+
+	for (i = 0; i < cnt; i++) {
+		struct device *msm_iommu_get_ctx(const char *ctx_name);
+		struct device *ctx = msm_iommu_get_ctx(names[i]);
+		if (IS_ERR_OR_NULL(ctx))
+			continue;
+		iommu_detach_device(iommu->domain, ctx);
+	}
+}
+
 static int msm_iommu_map(struct msm_mmu *mmu, uint32_t iova,
 		struct sg_table *sgt, unsigned len, int prot)
 {
@@ -110,7 +126,7 @@ static int msm_iommu_unmap(struct msm_mmu *mmu, uint32_t iova,
 
 		VERB("unmap[%d]: %08x(%x)", i, iova, bytes);
 
-		BUG_ON(!IS_ALIGNED(bytes, PAGE_SIZE));
+		BUG_ON(!PAGE_ALIGNED(bytes));
 
 		da += bytes;
 	}
@@ -127,6 +143,7 @@ static void msm_iommu_destroy(struct msm_mmu *mmu)
 
 static const struct msm_mmu_funcs funcs = {
 		.attach = msm_iommu_attach,
+		.detach = msm_iommu_detach,
 		.map = msm_iommu_map,
 		.unmap = msm_iommu_unmap,
 		.destroy = msm_iommu_destroy,
diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
index 030324482b4a..21da6d154f71 100644
--- a/drivers/gpu/drm/msm/msm_mmu.h
+++ b/drivers/gpu/drm/msm/msm_mmu.h
@@ -22,6 +22,7 @@
 
 struct msm_mmu_funcs {
 	int (*attach)(struct msm_mmu *mmu, const char **names, int cnt);
+	void (*detach)(struct msm_mmu *mmu, const char **names, int cnt);
 	int (*map)(struct msm_mmu *mmu, uint32_t iova, struct sg_table *sgt,
 			unsigned len, int prot);
 	int (*unmap)(struct msm_mmu *mmu, uint32_t iova, struct sg_table *sgt,
diff --git a/drivers/gpu/drm/msm/msm_perf.c b/drivers/gpu/drm/msm/msm_perf.c
new file mode 100644
index 000000000000..830857c47c86
--- /dev/null
+++ b/drivers/gpu/drm/msm/msm_perf.c
@@ -0,0 +1,275 @@
+/*
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* For profiling, userspace can:
+ *
+ *   tail -f /sys/kernel/debug/dri/<minor>/gpu
+ *
+ * This will enable performance counters/profiling to track the busy time
+ * and any gpu specific performance counters that are supported.
+ */
+
+#ifdef CONFIG_DEBUG_FS
+
+#include <linux/debugfs.h>
+
+#include "msm_drv.h"
+#include "msm_gpu.h"
+
+struct msm_perf_state {
+	struct drm_device *dev;
+
+	bool open;
+	int cnt;
+	struct mutex read_lock;
+
+	char buf[256];
+	int buftot, bufpos;
+
+	unsigned long next_jiffies;
+
+	struct dentry *ent;
+	struct drm_info_node *node;
+};
+
+#define SAMPLE_TIME (HZ/4)
+
+/* wait for next sample time: */
+static int wait_sample(struct msm_perf_state *perf)
+{
+	unsigned long start_jiffies = jiffies;
+
+	if (time_after(perf->next_jiffies, start_jiffies)) {
+		unsigned long remaining_jiffies =
+			perf->next_jiffies - start_jiffies;
+		int ret = schedule_timeout_interruptible(remaining_jiffies);
+		if (ret > 0) {
+			/* interrupted */
+			return -ERESTARTSYS;
+		}
+	}
+	perf->next_jiffies += SAMPLE_TIME;
+	return 0;
+}
+
+static int refill_buf(struct msm_perf_state *perf)
+{
+	struct msm_drm_private *priv = perf->dev->dev_private;
+	struct msm_gpu *gpu = priv->gpu;
+	char *ptr = perf->buf;
+	int rem = sizeof(perf->buf);
+	int i, n;
+
+	if ((perf->cnt++ % 32) == 0) {
+		/* Header line: */
+		n = snprintf(ptr, rem, "%%BUSY");
+		ptr += n;
+		rem -= n;
+
+		for (i = 0; i < gpu->num_perfcntrs; i++) {
+			const struct msm_gpu_perfcntr *perfcntr = &gpu->perfcntrs[i];
+			n = snprintf(ptr, rem, "\t%s", perfcntr->name);
+			ptr += n;
+			rem -= n;
+		}
+	} else {
+		/* Sample line: */
+		uint32_t activetime = 0, totaltime = 0;
+		uint32_t cntrs[5];
+		uint32_t val;
+		int ret;
+
+		/* sleep until next sample time: */
+		ret = wait_sample(perf);
+		if (ret)
+			return ret;
+
+		ret = msm_gpu_perfcntr_sample(gpu, &activetime, &totaltime,
+				ARRAY_SIZE(cntrs), cntrs);
+		if (ret < 0)
+			return ret;
+
+		val = totaltime ? 1000 * activetime / totaltime : 0;
+		n = snprintf(ptr, rem, "%3d.%d%%", val / 10, val % 10);
+		ptr += n;
+		rem -= n;
+
+		for (i = 0; i < ret; i++) {
+			/* cycle counters (I think).. convert to MHz.. */
+			val = cntrs[i] / 10000;
+			n = snprintf(ptr, rem, "\t%5d.%02d",
+					val / 100, val % 100);
+			ptr += n;
+			rem -= n;
+		}
+	}
+
+	n = snprintf(ptr, rem, "\n");
+	ptr += n;
+	rem -= n;
+
+	perf->bufpos = 0;
+	perf->buftot = ptr - perf->buf;
+
+	return 0;
+}
+
+static ssize_t perf_read(struct file *file, char __user *buf,
+		size_t sz, loff_t *ppos)
+{
+	struct msm_perf_state *perf = file->private_data;
+	int n = 0, ret;
+
+	mutex_lock(&perf->read_lock);
+
+	if (perf->bufpos >= perf->buftot) {
+		ret = refill_buf(perf);
+		if (ret)
+			goto out;
+	}
+
+	n = min((int)sz, perf->buftot - perf->bufpos);
+	ret = copy_to_user(buf, &perf->buf[perf->bufpos], n);
+	if (ret)
+		goto out;
+
+	perf->bufpos += n;
+	*ppos += n;
+
+out:
+	mutex_unlock(&perf->read_lock);
+	if (ret)
+		return ret;
+	return n;
+}
+
+static int perf_open(struct inode *inode, struct file *file)
+{
+	struct msm_perf_state *perf = inode->i_private;
+	struct drm_device *dev = perf->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_gpu *gpu = priv->gpu;
+	int ret = 0;
+
+	mutex_lock(&dev->struct_mutex);
+
+	if (perf->open || !gpu) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	file->private_data = perf;
+	perf->open = true;
+	perf->cnt = 0;
+	perf->buftot = 0;
+	perf->bufpos = 0;
+	msm_gpu_perfcntr_start(gpu);
+	perf->next_jiffies = jiffies + SAMPLE_TIME;
+
+out:
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
+}
+
+static int perf_release(struct inode *inode, struct file *file)
+{
+	struct msm_perf_state *perf = inode->i_private;
+	struct msm_drm_private *priv = perf->dev->dev_private;
+	msm_gpu_perfcntr_stop(priv->gpu);
+	perf->open = false;
+	return 0;
+}
+
+
+static const struct file_operations perf_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.open = perf_open,
+	.read = perf_read,
+	.llseek = no_llseek,
+	.release = perf_release,
+};
+
+int msm_perf_debugfs_init(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_perf_state *perf;
+
+	/* only create on first minor: */
+	if (priv->perf)
+		return 0;
+
+	perf = kzalloc(sizeof(*perf), GFP_KERNEL);
+	if (!perf)
+		return -ENOMEM;
+
+	perf->dev = minor->dev;
+
+	mutex_init(&perf->read_lock);
+	priv->perf = perf;
+
+	perf->node = kzalloc(sizeof(*perf->node), GFP_KERNEL);
+	if (!perf->node)
+		goto fail;
+
+	perf->ent = debugfs_create_file("perf", S_IFREG | S_IRUGO,
+			minor->debugfs_root, perf, &perf_debugfs_fops);
+	if (!perf->ent) {
+		DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/perf\n",
+				minor->debugfs_root->d_name.name);
+		goto fail;
+	}
+
+	perf->node->minor = minor;
+	perf->node->dent  = perf->ent;
+	perf->node->info_ent = NULL;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&perf->node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+
+fail:
+	msm_perf_debugfs_cleanup(minor);
+	return -1;
+}
+
+void msm_perf_debugfs_cleanup(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_perf_state *perf = priv->perf;
+
+	if (!perf)
+		return;
+
+	priv->perf = NULL;
+
+	debugfs_remove(perf->ent);
+
+	if (perf->node) {
+		mutex_lock(&minor->debugfs_lock);
+		list_del(&perf->node->list);
+		mutex_unlock(&minor->debugfs_lock);
+		kfree(perf->node);
+	}
+
+	mutex_destroy(&perf->read_lock);
+
+	kfree(perf);
+}
+
+#endif
diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c
new file mode 100644
index 000000000000..9a78c48817c6
--- /dev/null
+++ b/drivers/gpu/drm/msm/msm_rd.c
@@ -0,0 +1,337 @@
+/*
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* For debugging crashes, userspace can:
+ *
+ *   tail -f /sys/kernel/debug/dri/<minor>/rd > logfile.rd
+ *
+ * To log the cmdstream in a format that is understood by freedreno/cffdump
+ * utility.  By comparing the last successfully completed fence #, to the
+ * cmdstream for the next fence, you can narrow down which process and submit
+ * caused the gpu crash/lockup.
+ *
+ * This bypasses drm_debugfs_create_files() mainly because we need to use
+ * our own fops for a bit more control.  In particular, we don't want to
+ * do anything if userspace doesn't have the debugfs file open.
+ */
+
+#ifdef CONFIG_DEBUG_FS
+
+#include <linux/kfifo.h>
+#include <linux/debugfs.h>
+#include <linux/circ_buf.h>
+#include <linux/wait.h>
+
+#include "msm_drv.h"
+#include "msm_gpu.h"
+#include "msm_gem.h"
+
+enum rd_sect_type {
+	RD_NONE,
+	RD_TEST,       /* ascii text */
+	RD_CMD,        /* ascii text */
+	RD_GPUADDR,    /* u32 gpuaddr, u32 size */
+	RD_CONTEXT,    /* raw dump */
+	RD_CMDSTREAM,  /* raw dump */
+	RD_CMDSTREAM_ADDR, /* gpu addr of cmdstream */
+	RD_PARAM,      /* u32 param_type, u32 param_val, u32 bitlen */
+	RD_FLUSH,      /* empty, clear previous params */
+	RD_PROGRAM,    /* shader program, raw dump */
+	RD_VERT_SHADER,
+	RD_FRAG_SHADER,
+	RD_BUFFER_CONTENTS,
+	RD_GPU_ID,
+};
+
+#define BUF_SZ 512  /* should be power of 2 */
+
+/* space used: */
+#define circ_count(circ) \
+	(CIRC_CNT((circ)->head, (circ)->tail, BUF_SZ))
+#define circ_count_to_end(circ) \
+	(CIRC_CNT_TO_END((circ)->head, (circ)->tail, BUF_SZ))
+/* space available: */
+#define circ_space(circ) \
+	(CIRC_SPACE((circ)->head, (circ)->tail, BUF_SZ))
+#define circ_space_to_end(circ) \
+	(CIRC_SPACE_TO_END((circ)->head, (circ)->tail, BUF_SZ))
+
+struct msm_rd_state {
+	struct drm_device *dev;
+
+	bool open;
+
+	struct dentry *ent;
+	struct drm_info_node *node;
+
+	/* current submit to read out: */
+	struct msm_gem_submit *submit;
+
+	/* fifo access is synchronized on the producer side by
+	 * struct_mutex held by submit code (otherwise we could
+	 * end up w/ cmds logged in different order than they
+	 * were executed).  And read_lock synchronizes the reads
+	 */
+	struct mutex read_lock;
+
+	wait_queue_head_t fifo_event;
+	struct circ_buf fifo;
+
+	char buf[BUF_SZ];
+};
+
+static void rd_write(struct msm_rd_state *rd, const void *buf, int sz)
+{
+	struct circ_buf *fifo = &rd->fifo;
+	const char *ptr = buf;
+
+	while (sz > 0) {
+		char *fptr = &fifo->buf[fifo->head];
+		int n;
+
+		wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0);
+
+		n = min(sz, circ_space_to_end(&rd->fifo));
+		memcpy(fptr, ptr, n);
+
+		fifo->head = (fifo->head + n) & (BUF_SZ - 1);
+		sz  -= n;
+		ptr += n;
+
+		wake_up_all(&rd->fifo_event);
+	}
+}
+
+static void rd_write_section(struct msm_rd_state *rd,
+		enum rd_sect_type type, const void *buf, int sz)
+{
+	rd_write(rd, &type, 4);
+	rd_write(rd, &sz, 4);
+	rd_write(rd, buf, sz);
+}
+
+static ssize_t rd_read(struct file *file, char __user *buf,
+		size_t sz, loff_t *ppos)
+{
+	struct msm_rd_state *rd = file->private_data;
+	struct circ_buf *fifo = &rd->fifo;
+	const char *fptr = &fifo->buf[fifo->tail];
+	int n = 0, ret = 0;
+
+	mutex_lock(&rd->read_lock);
+
+	ret = wait_event_interruptible(rd->fifo_event,
+			circ_count(&rd->fifo) > 0);
+	if (ret)
+		goto out;
+
+	n = min_t(int, sz, circ_count_to_end(&rd->fifo));
+	ret = copy_to_user(buf, fptr, n);
+	if (ret)
+		goto out;
+
+	fifo->tail = (fifo->tail + n) & (BUF_SZ - 1);
+	*ppos += n;
+
+	wake_up_all(&rd->fifo_event);
+
+out:
+	mutex_unlock(&rd->read_lock);
+	if (ret)
+		return ret;
+	return n;
+}
+
+static int rd_open(struct inode *inode, struct file *file)
+{
+	struct msm_rd_state *rd = inode->i_private;
+	struct drm_device *dev = rd->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_gpu *gpu = priv->gpu;
+	uint64_t val;
+	uint32_t gpu_id;
+	int ret = 0;
+
+	mutex_lock(&dev->struct_mutex);
+
+	if (rd->open || !gpu) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	file->private_data = rd;
+	rd->open = true;
+
+	/* the parsing tools need to know gpu-id to know which
+	 * register database to load.
+	 */
+	gpu->funcs->get_param(gpu, MSM_PARAM_GPU_ID, &val);
+	gpu_id = val;
+
+	rd_write_section(rd, RD_GPU_ID, &gpu_id, sizeof(gpu_id));
+
+out:
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
+}
+
+static int rd_release(struct inode *inode, struct file *file)
+{
+	struct msm_rd_state *rd = inode->i_private;
+	rd->open = false;
+	return 0;
+}
+
+
+static const struct file_operations rd_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.open = rd_open,
+	.read = rd_read,
+	.llseek = no_llseek,
+	.release = rd_release,
+};
+
+int msm_rd_debugfs_init(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_rd_state *rd;
+
+	/* only create on first minor: */
+	if (priv->rd)
+		return 0;
+
+	rd = kzalloc(sizeof(*rd), GFP_KERNEL);
+	if (!rd)
+		return -ENOMEM;
+
+	rd->dev = minor->dev;
+	rd->fifo.buf = rd->buf;
+
+	mutex_init(&rd->read_lock);
+	priv->rd = rd;
+
+	init_waitqueue_head(&rd->fifo_event);
+
+	rd->node = kzalloc(sizeof(*rd->node), GFP_KERNEL);
+	if (!rd->node)
+		goto fail;
+
+	rd->ent = debugfs_create_file("rd", S_IFREG | S_IRUGO,
+			minor->debugfs_root, rd, &rd_debugfs_fops);
+	if (!rd->ent) {
+		DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/rd\n",
+				minor->debugfs_root->d_name.name);
+		goto fail;
+	}
+
+	rd->node->minor = minor;
+	rd->node->dent  = rd->ent;
+	rd->node->info_ent = NULL;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&rd->node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+
+fail:
+	msm_rd_debugfs_cleanup(minor);
+	return -1;
+}
+
+void msm_rd_debugfs_cleanup(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_rd_state *rd = priv->rd;
+
+	if (!rd)
+		return;
+
+	priv->rd = NULL;
+
+	debugfs_remove(rd->ent);
+
+	if (rd->node) {
+		mutex_lock(&minor->debugfs_lock);
+		list_del(&rd->node->list);
+		mutex_unlock(&minor->debugfs_lock);
+		kfree(rd->node);
+	}
+
+	mutex_destroy(&rd->read_lock);
+
+	kfree(rd);
+}
+
+/* called under struct_mutex */
+void msm_rd_dump_submit(struct msm_gem_submit *submit)
+{
+	struct drm_device *dev = submit->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_rd_state *rd = priv->rd;
+	char msg[128];
+	int i, n;
+
+	if (!rd->open)
+		return;
+
+	/* writing into fifo is serialized by caller, and
+	 * rd->read_lock is used to serialize the reads
+	 */
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	n = snprintf(msg, sizeof(msg), "%.*s/%d: fence=%u",
+			TASK_COMM_LEN, current->comm, task_pid_nr(current),
+			submit->fence);
+
+	rd_write_section(rd, RD_CMD, msg, ALIGN(n, 4));
+
+	/* could be nice to have an option (module-param?) to snapshot
+	 * all the bo's associated with the submit.  Handy to see vtx
+	 * buffers, etc.  For now just the cmdstream bo's is enough.
+	 */
+
+	for (i = 0; i < submit->nr_cmds; i++) {
+		uint32_t idx  = submit->cmd[i].idx;
+		uint32_t iova = submit->cmd[i].iova;
+		uint32_t szd  = submit->cmd[i].size; /* in dwords */
+		struct msm_gem_object *obj = submit->bos[idx].obj;
+		const char *buf = msm_gem_vaddr_locked(&obj->base);
+
+		buf += iova - submit->bos[idx].iova;
+
+		rd_write_section(rd, RD_GPUADDR,
+				(uint32_t[2]){ iova, szd * 4 }, 8);
+		rd_write_section(rd, RD_BUFFER_CONTENTS,
+				buf, szd * 4);
+
+		switch (submit->cmd[i].type) {
+		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
+			/* ignore IB-targets, we've logged the buffer, the
+			 * parser tool will follow the IB based on the logged
+			 * buffer/gpuaddr, so nothing more to do.
+			 */
+			break;
+		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
+		case MSM_SUBMIT_CMD_BUF:
+			rd_write_section(rd, RD_CMDSTREAM_ADDR,
+					(uint32_t[2]){ iova, szd }, 8);
+			break;
+		}
+	}
+}
+#endif
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index b7d216264775..8b307e143632 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -102,6 +102,7 @@ nouveau-y += core/subdev/fb/nvaa.o
 nouveau-y += core/subdev/fb/nvaf.o
 nouveau-y += core/subdev/fb/nvc0.o
 nouveau-y += core/subdev/fb/nve0.o
+nouveau-y += core/subdev/fb/gk20a.o
 nouveau-y += core/subdev/fb/gm107.o
 nouveau-y += core/subdev/fb/ramnv04.o
 nouveau-y += core/subdev/fb/ramnv10.o
@@ -117,25 +118,33 @@ nouveau-y += core/subdev/fb/ramnva3.o
 nouveau-y += core/subdev/fb/ramnvaa.o
 nouveau-y += core/subdev/fb/ramnvc0.o
 nouveau-y += core/subdev/fb/ramnve0.o
+nouveau-y += core/subdev/fb/ramgk20a.o
 nouveau-y += core/subdev/fb/ramgm107.o
 nouveau-y += core/subdev/fb/sddr3.o
 nouveau-y += core/subdev/fb/gddr5.o
 nouveau-y += core/subdev/gpio/base.o
 nouveau-y += core/subdev/gpio/nv10.o
 nouveau-y += core/subdev/gpio/nv50.o
+nouveau-y += core/subdev/gpio/nv92.o
 nouveau-y += core/subdev/gpio/nvd0.o
 nouveau-y += core/subdev/gpio/nve0.o
 nouveau-y += core/subdev/i2c/base.o
 nouveau-y += core/subdev/i2c/anx9805.o
 nouveau-y += core/subdev/i2c/aux.o
 nouveau-y += core/subdev/i2c/bit.o
+nouveau-y += core/subdev/i2c/pad.o
+nouveau-y += core/subdev/i2c/padnv04.o
+nouveau-y += core/subdev/i2c/padnv94.o
 nouveau-y += core/subdev/i2c/nv04.o
 nouveau-y += core/subdev/i2c/nv4e.o
 nouveau-y += core/subdev/i2c/nv50.o
 nouveau-y += core/subdev/i2c/nv94.o
 nouveau-y += core/subdev/i2c/nvd0.o
+nouveau-y += core/subdev/i2c/gf117.o
+nouveau-y += core/subdev/i2c/nve0.o
 nouveau-y += core/subdev/ibus/nvc0.o
 nouveau-y += core/subdev/ibus/nve0.o
+nouveau-y += core/subdev/ibus/gk20a.o
 nouveau-y += core/subdev/instmem/base.o
 nouveau-y += core/subdev/instmem/nv04.o
 nouveau-y += core/subdev/instmem/nv40.o
@@ -214,6 +223,9 @@ nouveau-y += core/engine/device/nvc0.o
 nouveau-y += core/engine/device/nve0.o
 nouveau-y += core/engine/device/gm100.o
 nouveau-y += core/engine/disp/base.o
+nouveau-y += core/engine/disp/conn.o
+nouveau-y += core/engine/disp/outp.o
+nouveau-y += core/engine/disp/outpdp.o
 nouveau-y += core/engine/disp/nv04.o
 nouveau-y += core/engine/disp/nv50.o
 nouveau-y += core/engine/disp/nv84.o
@@ -245,6 +257,7 @@ nouveau-y += core/engine/fifo/nv50.o
 nouveau-y += core/engine/fifo/nv84.o
 nouveau-y += core/engine/fifo/nvc0.o
 nouveau-y += core/engine/fifo/nve0.o
+nouveau-y += core/engine/fifo/gk20a.o
 nouveau-y += core/engine/fifo/nv108.o
 nouveau-y += core/engine/graph/ctxnv40.o
 nouveau-y += core/engine/graph/ctxnv50.o
@@ -255,6 +268,7 @@ nouveau-y += core/engine/graph/ctxnvc8.o
 nouveau-y += core/engine/graph/ctxnvd7.o
 nouveau-y += core/engine/graph/ctxnvd9.o
 nouveau-y += core/engine/graph/ctxnve4.o
+nouveau-y += core/engine/graph/ctxgk20a.o
 nouveau-y += core/engine/graph/ctxnvf0.o
 nouveau-y += core/engine/graph/ctxnv108.o
 nouveau-y += core/engine/graph/ctxgm107.o
@@ -275,6 +289,7 @@ nouveau-y += core/engine/graph/nvc8.o
 nouveau-y += core/engine/graph/nvd7.o
 nouveau-y += core/engine/graph/nvd9.o
 nouveau-y += core/engine/graph/nve4.o
+nouveau-y += core/engine/graph/gk20a.o
 nouveau-y += core/engine/graph/nvf0.o
 nouveau-y += core/engine/graph/nv108.o
 nouveau-y += core/engine/graph/gm107.o
diff --git a/drivers/gpu/drm/nouveau/core/core/event.c b/drivers/gpu/drm/nouveau/core/core/event.c
index 3f3c76581a9e..ae81d3b5d8b7 100644
--- a/drivers/gpu/drm/nouveau/core/core/event.c
+++ b/drivers/gpu/drm/nouveau/core/core/event.c
@@ -28,14 +28,20 @@ nouveau_event_put(struct nouveau_eventh *handler)
 {
 	struct nouveau_event *event = handler->event;
 	unsigned long flags;
-	if (__test_and_clear_bit(NVKM_EVENT_ENABLE, &handler->flags)) {
-		spin_lock_irqsave(&event->refs_lock, flags);
-		if (!--event->index[handler->index].refs) {
+	u32 m, t;
+
+	if (!__test_and_clear_bit(NVKM_EVENT_ENABLE, &handler->flags))
+		return;
+
+	spin_lock_irqsave(&event->refs_lock, flags);
+	for (m = handler->types; t = __ffs(m), m; m &= ~(1 << t)) {
+		if (!--event->refs[handler->index * event->types_nr + t]) {
 			if (event->disable)
-				event->disable(event, handler->index);
+				event->disable(event, 1 << t, handler->index);
 		}
-		spin_unlock_irqrestore(&event->refs_lock, flags);
+
 	}
+	spin_unlock_irqrestore(&event->refs_lock, flags);
 }
 
 void
@@ -43,14 +49,20 @@ nouveau_event_get(struct nouveau_eventh *handler)
 {
 	struct nouveau_event *event = handler->event;
 	unsigned long flags;
-	if (!__test_and_set_bit(NVKM_EVENT_ENABLE, &handler->flags)) {
-		spin_lock_irqsave(&event->refs_lock, flags);
-		if (!event->index[handler->index].refs++) {
+	u32 m, t;
+
+	if (__test_and_set_bit(NVKM_EVENT_ENABLE, &handler->flags))
+		return;
+
+	spin_lock_irqsave(&event->refs_lock, flags);
+	for (m = handler->types; t = __ffs(m), m; m &= ~(1 << t)) {
+		if (!event->refs[handler->index * event->types_nr + t]++) {
 			if (event->enable)
-				event->enable(event, handler->index);
+				event->enable(event, 1 << t, handler->index);
 		}
-		spin_unlock_irqrestore(&event->refs_lock, flags);
+
 	}
+	spin_unlock_irqrestore(&event->refs_lock, flags);
 }
 
 static void
@@ -65,38 +77,47 @@ nouveau_event_fini(struct nouveau_eventh *handler)
 }
 
 static int
-nouveau_event_init(struct nouveau_event *event, int index,
-		   int (*func)(void *, int), void *priv,
+nouveau_event_init(struct nouveau_event *event, u32 types, int index,
+		   int (*func)(void *, u32, int), void *priv,
 		   struct nouveau_eventh *handler)
 {
 	unsigned long flags;
 
+	if (types & ~((1 << event->types_nr) - 1))
+		return -EINVAL;
 	if (index >= event->index_nr)
 		return -EINVAL;
 
 	handler->event = event;
 	handler->flags = 0;
+	handler->types = types;
 	handler->index = index;
 	handler->func = func;
 	handler->priv = priv;
 
 	spin_lock_irqsave(&event->list_lock, flags);
-	list_add_tail(&handler->head, &event->index[index].list);
+	list_add_tail(&handler->head, &event->list[index]);
 	spin_unlock_irqrestore(&event->list_lock, flags);
 	return 0;
 }
 
 int
-nouveau_event_new(struct nouveau_event *event, int index,
-		  int (*func)(void *, int), void *priv,
+nouveau_event_new(struct nouveau_event *event, u32 types, int index,
+		  int (*func)(void *, u32, int), void *priv,
 		  struct nouveau_eventh **phandler)
 {
 	struct nouveau_eventh *handler;
 	int ret = -ENOMEM;
 
+	if (event->check) {
+		ret = event->check(event, types, index);
+		if (ret)
+			return ret;
+	}
+
 	handler = *phandler = kmalloc(sizeof(*handler), GFP_KERNEL);
 	if (handler) {
-		ret = nouveau_event_init(event, index, func, priv, handler);
+		ret = nouveau_event_init(event, types, index, func, priv, handler);
 		if (ret)
 			kfree(handler);
 	}
@@ -116,7 +137,7 @@ nouveau_event_ref(struct nouveau_eventh *handler, struct nouveau_eventh **ref)
 }
 
 void
-nouveau_event_trigger(struct nouveau_event *event, int index)
+nouveau_event_trigger(struct nouveau_event *event, u32 types, int index)
 {
 	struct nouveau_eventh *handler;
 	unsigned long flags;
@@ -125,10 +146,15 @@ nouveau_event_trigger(struct nouveau_event *event, int index)
 		return;
 
 	spin_lock_irqsave(&event->list_lock, flags);
-	list_for_each_entry(handler, &event->index[index].list, head) {
-		if (test_bit(NVKM_EVENT_ENABLE, &handler->flags) &&
-		    handler->func(handler->priv, index) == NVKM_EVENT_DROP)
-			nouveau_event_put(handler);
+	list_for_each_entry(handler, &event->list[index], head) {
+		if (!test_bit(NVKM_EVENT_ENABLE, &handler->flags))
+			continue;
+		if (!(handler->types & types))
+			continue;
+		if (handler->func(handler->priv, handler->types & types, index)
+				!= NVKM_EVENT_DROP)
+			continue;
+		nouveau_event_put(handler);
 	}
 	spin_unlock_irqrestore(&event->list_lock, flags);
 }
@@ -144,20 +170,27 @@ nouveau_event_destroy(struct nouveau_event **pevent)
 }
 
 int
-nouveau_event_create(int index_nr, struct nouveau_event **pevent)
+nouveau_event_create(int types_nr, int index_nr, struct nouveau_event **pevent)
 {
 	struct nouveau_event *event;
 	int i;
 
-	event = *pevent = kzalloc(sizeof(*event) + index_nr *
-				  sizeof(event->index[0]), GFP_KERNEL);
+	event = *pevent = kzalloc(sizeof(*event) + (index_nr * types_nr) *
+				  sizeof(event->refs[0]), GFP_KERNEL);
 	if (!event)
 		return -ENOMEM;
 
+	event->list = kmalloc(sizeof(*event->list) * index_nr, GFP_KERNEL);
+	if (!event->list) {
+		kfree(event);
+		return -ENOMEM;
+	}
+
 	spin_lock_init(&event->list_lock);
 	spin_lock_init(&event->refs_lock);
 	for (i = 0; i < index_nr; i++)
-		INIT_LIST_HEAD(&event->index[i].list);
+		INIT_LIST_HEAD(&event->list[i]);
+	event->types_nr = types_nr;
 	event->index_nr = index_nr;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/core/object.c b/drivers/gpu/drm/nouveau/core/core/object.c
index 7f48e288215f..124538555904 100644
--- a/drivers/gpu/drm/nouveau/core/core/object.c
+++ b/drivers/gpu/drm/nouveau/core/core/object.c
@@ -156,7 +156,7 @@ nouveau_object_ctor(struct nouveau_object *parent,
 	}
 
 	if (ret == 0) {
-		nv_debug(object, "created\n");
+		nv_trace(object, "created\n");
 		atomic_set(&object->refcount, 1);
 	}
 
@@ -166,7 +166,7 @@ nouveau_object_ctor(struct nouveau_object *parent,
 static void
 nouveau_object_dtor(struct nouveau_object *object)
 {
-	nv_debug(object, "destroying\n");
+	nv_trace(object, "destroying\n");
 	nv_ofuncs(object)->dtor(object);
 }
 
@@ -337,7 +337,7 @@ nouveau_object_inc(struct nouveau_object *object)
 		goto fail_self;
 	}
 
-	nv_debug(object, "initialised\n");
+	nv_trace(object, "initialised\n");
 	return 0;
 
 fail_self:
@@ -375,7 +375,7 @@ nouveau_object_decf(struct nouveau_object *object)
 	if (object->parent)
 		nouveau_object_dec(object->parent, false);
 
-	nv_debug(object, "stopped\n");
+	nv_trace(object, "stopped\n");
 	return 0;
 }
 
@@ -411,7 +411,7 @@ nouveau_object_decs(struct nouveau_object *object)
 		}
 	}
 
-	nv_debug(object, "suspended\n");
+	nv_trace(object, "suspended\n");
 	return 0;
 
 fail_parent:
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/gm100.c b/drivers/gpu/drm/nouveau/core/engine/device/gm100.c
index d258c21c4a22..a520029e25d9 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/gm100.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/gm100.c
@@ -60,8 +60,8 @@ gm100_identify(struct nouveau_device *device)
 	case 0x117:
 		device->cname = "GM107";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 #if 0
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv04.c b/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
index 0a51ff4e9e00..40b29d0214cb 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
@@ -47,7 +47,7 @@ nv04_identify(struct nouveau_device *device)
 	case 0x04:
 		device->cname = "NV04";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv04_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -65,7 +65,7 @@ nv04_identify(struct nouveau_device *device)
 	case 0x05:
 		device->cname = "NV05";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv05_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv10.c b/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
index e008de8b51b0..5f7c25ff523d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
@@ -48,8 +48,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x10:
 		device->cname = "NV10";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -65,8 +65,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x15:
 		device->cname = "NV15";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -84,8 +84,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x16:
 		device->cname = "NV16";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -103,8 +103,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x1a:
 		device->cname = "nForce";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -122,8 +122,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x11:
 		device->cname = "NV11";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -141,8 +141,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x17:
 		device->cname = "NV17";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -160,8 +160,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x1f:
 		device->cname = "nForce2";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -179,8 +179,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x18:
 		device->cname = "NV18";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv20.c b/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
index 7b629a3aed05..75fed11bba0a 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
@@ -49,8 +49,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x20:
 		device->cname = "NV20";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -68,8 +68,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x25:
 		device->cname = "NV25";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -87,8 +87,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x28:
 		device->cname = "NV28";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -106,8 +106,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x2a:
 		device->cname = "NV2A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv30.c b/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
index 7dfddd5a1908..36919d7db7cc 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
@@ -49,8 +49,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x30:
 		device->cname = "NV30";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -68,8 +68,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x35:
 		device->cname = "NV35";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -87,8 +87,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x31:
 		device->cname = "NV31";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -107,8 +107,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x36:
 		device->cname = "NV36";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -127,8 +127,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x34:
 		device->cname = "NV34";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
index 7c1ce6cf4f1f..1130a62be2c7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
@@ -53,8 +53,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x40:
 		device->cname = "NV40";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -76,8 +76,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x41:
 		device->cname = "NV41";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -99,8 +99,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x42:
 		device->cname = "NV42";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -122,8 +122,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x43:
 		device->cname = "NV43";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -145,8 +145,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x45:
 		device->cname = "NV45";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -168,8 +168,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x47:
 		device->cname = "G70";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -191,8 +191,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x49:
 		device->cname = "G71";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -214,8 +214,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4b:
 		device->cname = "G73";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -237,8 +237,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x44:
 		device->cname = "NV44";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -260,8 +260,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x46:
 		device->cname = "G72";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -283,8 +283,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4a:
 		device->cname = "NV44A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -306,8 +306,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4c:
 		device->cname = "C61";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -329,8 +329,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4e:
 		device->cname = "C51";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv4e_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv4e_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -352,8 +352,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x63:
 		device->cname = "C73";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -375,8 +375,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x67:
 		device->cname = "C67";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -398,8 +398,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x68:
 		device->cname = "C68";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
index 66499fa0f758..ef0b0bde1a91 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
@@ -60,8 +60,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x50:
 		device->cname = "G80";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv50_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -85,8 +85,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x84:
 		device->cname = "G84";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv50_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -113,8 +113,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x86:
 		device->cname = "G86";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv50_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -141,8 +141,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x92:
 		device->cname = "G92";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -169,8 +169,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x94:
 		device->cname = "G94";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -197,8 +197,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x96:
 		device->cname = "G96";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -225,8 +225,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x98:
 		device->cname = "G98";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -253,8 +253,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa0:
 		device->cname = "G200";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -281,8 +281,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xaa:
 		device->cname = "MCP77/MCP78";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nvaa_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -309,8 +309,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xac:
 		device->cname = "MCP79/MCP7A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nvaa_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -337,8 +337,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa3:
 		device->cname = "GT215";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -367,8 +367,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa5:
 		device->cname = "GT216";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -396,8 +396,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa8:
 		device->cname = "GT218";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -425,8 +425,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xaf:
 		device->cname = "MCP89";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
index 2075b3027052..8d55ed633b19 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
@@ -60,8 +60,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc0:
 		device->cname = "GF100";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -92,8 +92,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc4:
 		device->cname = "GF104";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -124,8 +124,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc3:
 		device->cname = "GF106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -155,8 +155,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xce:
 		device->cname = "GF114";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -187,8 +187,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xcf:
 		device->cname = "GF116";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -219,8 +219,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc1:
 		device->cname = "GF108";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -250,8 +250,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc8:
 		device->cname = "GF110";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -282,8 +282,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xd9:
 		device->cname = "GF119";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nvd0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -313,8 +313,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xd7:
 		device->cname = "GF117";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nvd0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  gf117_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
index 9784cbf8a9d2..2d1e97d4264f 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
@@ -60,8 +60,8 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe4:
 		device->cname = "GK104";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -93,8 +93,8 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe7:
 		device->cname = "GK107";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -126,8 +126,8 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe6:
 		device->cname = "GK106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -156,11 +156,61 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
 		break;
+	case 0xea:
+		device->cname = "GK20A";
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_TIMER  ] = &gk20a_timer_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  gk20a_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_IBUS   ] = &gk20a_ibus_oclass;
+		device->oclass[NVDEV_SUBDEV_INSTMEM] = nv50_instmem_oclass;
+		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  gk20a_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_GR     ] =  gk20a_graph_oclass;
+		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
+		break;
 	case 0xf0:
 		device->cname = "GK110";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
+		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nvc0_devinit_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_LTCG   ] =  gf100_ltcg_oclass;
+		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
+		device->oclass[NVDEV_SUBDEV_INSTMEM] =  nv50_instmem_oclass;
+		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
+		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_GR     ] =  nvf0_graph_oclass;
+		device->oclass[NVDEV_ENGINE_DISP   ] =  nvf0_disp_oclass;
+		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
+		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
+		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
+		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
+		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
+		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvf0_perfmon_oclass;
+		break;
+	case 0xf1:
+		device->cname = "GK110B";
+		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -184,18 +234,16 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
-#if 0
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
-#endif
 		device->oclass[NVDEV_ENGINE_PERFMON] = &nvf0_perfmon_oclass;
 		break;
 	case 0x108:
 		device->cname = "GK208";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -219,11 +267,9 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
-#if 0
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
-#endif
 		break;
 	default:
 		nv_fatal(device, "unknown Kepler chipset\n");
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/base.c b/drivers/gpu/drm/nouveau/core/engine/disp/base.c
index 7a5cae42834f..9c38c5e40500 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/base.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/base.c
@@ -22,13 +22,89 @@
  * Authors: Ben Skeggs
  */
 
-#include <engine/disp.h>
+#include "priv.h"
+#include "outp.h"
+#include "conn.h"
+
+static int
+nouveau_disp_hpd_check(struct nouveau_event *event, u32 types, int index)
+{
+	struct nouveau_disp *disp = event->priv;
+	struct nvkm_output *outp;
+	list_for_each_entry(outp, &disp->outp, head) {
+		if (outp->conn->index == index) {
+			if (outp->conn->hpd.event)
+				return 0;
+			break;
+		}
+	}
+	return -ENOSYS;
+}
+
+int
+_nouveau_disp_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_disp *disp = (void *)object;
+	struct nvkm_output *outp;
+	int ret;
+
+	list_for_each_entry(outp, &disp->outp, head) {
+		ret = nv_ofuncs(outp)->fini(nv_object(outp), suspend);
+		if (ret && suspend)
+			goto fail_outp;
+	}
+
+	return nouveau_engine_fini(&disp->base, suspend);
+
+fail_outp:
+	list_for_each_entry_continue_reverse(outp, &disp->outp, head) {
+		nv_ofuncs(outp)->init(nv_object(outp));
+	}
+
+	return ret;
+}
+
+int
+_nouveau_disp_init(struct nouveau_object *object)
+{
+	struct nouveau_disp *disp = (void *)object;
+	struct nvkm_output *outp;
+	int ret;
+
+	ret = nouveau_engine_init(&disp->base);
+	if (ret)
+		return ret;
+
+	list_for_each_entry(outp, &disp->outp, head) {
+		ret = nv_ofuncs(outp)->init(nv_object(outp));
+		if (ret)
+			goto fail_outp;
+	}
+
+	return ret;
+
+fail_outp:
+	list_for_each_entry_continue_reverse(outp, &disp->outp, head) {
+		nv_ofuncs(outp)->fini(nv_object(outp), false);
+	}
+
+	return ret;
+}
 
 void
 _nouveau_disp_dtor(struct nouveau_object *object)
 {
 	struct nouveau_disp *disp = (void *)object;
+	struct nvkm_output *outp, *outt;
+
 	nouveau_event_destroy(&disp->vblank);
+
+	if (disp->outp.next) {
+		list_for_each_entry_safe(outp, outt, &disp->outp, head) {
+			nouveau_object_ref(NULL, (struct nouveau_object **)&outp);
+		}
+	}
+
 	nouveau_engine_destroy(&disp->base);
 }
 
@@ -39,8 +115,15 @@ nouveau_disp_create_(struct nouveau_object *parent,
 		     const char *intname, const char *extname,
 		     int length, void **pobject)
 {
+	struct nouveau_disp_impl *impl = (void *)oclass;
+	struct nouveau_bios *bios = nouveau_bios(parent);
 	struct nouveau_disp *disp;
-	int ret;
+	struct nouveau_oclass **sclass;
+	struct nouveau_object *object;
+	struct dcb_output dcbE;
+	u8  hpd = 0, ver, hdr;
+	u32 data;
+	int ret, i;
 
 	ret = nouveau_engine_create_(parent, engine, oclass, true,
 				     intname, extname, length, pobject);
@@ -48,5 +131,42 @@ nouveau_disp_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	return nouveau_event_create(heads, &disp->vblank);
+	INIT_LIST_HEAD(&disp->outp);
+
+	/* create output objects for each display path in the vbios */
+	i = -1;
+	while ((data = dcb_outp_parse(bios, ++i, &ver, &hdr, &dcbE))) {
+		if (dcbE.type == DCB_OUTPUT_UNUSED)
+			continue;
+		if (dcbE.type == DCB_OUTPUT_EOL)
+			break;
+		data = dcbE.location << 4 | dcbE.type;
+
+		oclass = nvkm_output_oclass;
+		sclass = impl->outp;
+		while (sclass && sclass[0]) {
+			if (sclass[0]->handle == data) {
+				oclass = sclass[0];
+				break;
+			}
+			sclass++;
+		}
+
+		nouveau_object_ctor(*pobject, *pobject, oclass,
+				    &dcbE, i, &object);
+		hpd = max(hpd, (u8)(dcbE.connector + 1));
+	}
+
+	ret = nouveau_event_create(3, hpd, &disp->hpd);
+	if (ret)
+		return ret;
+
+	disp->hpd->priv = disp;
+	disp->hpd->check = nouveau_disp_hpd_check;
+
+	ret = nouveau_event_create(1, heads, &disp->vblank);
+	if (ret)
+		return ret;
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/conn.c b/drivers/gpu/drm/nouveau/core/engine/disp/conn.c
new file mode 100644
index 000000000000..4ffbc70ecf5a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/conn.c
@@ -0,0 +1,172 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/gpio.h>
+
+#include "conn.h"
+#include "outp.h"
+
+static void
+nvkm_connector_hpd_work(struct work_struct *w)
+{
+	struct nvkm_connector *conn = container_of(w, typeof(*conn), hpd.work);
+	struct nouveau_disp *disp = nouveau_disp(conn);
+	struct nouveau_gpio *gpio = nouveau_gpio(conn);
+	u32 send = NVKM_HPD_UNPLUG;
+	if (gpio->get(gpio, 0, DCB_GPIO_UNUSED, conn->hpd.event->index))
+		send = NVKM_HPD_PLUG;
+	nouveau_event_trigger(disp->hpd, send, conn->index);
+	nouveau_event_get(conn->hpd.event);
+}
+
+static int
+nvkm_connector_hpd(void *data, u32 type, int index)
+{
+	struct nvkm_connector *conn = data;
+	DBG("HPD: %d\n", type);
+	schedule_work(&conn->hpd.work);
+	return NVKM_EVENT_DROP;
+}
+
+int
+_nvkm_connector_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_connector *conn = (void *)object;
+	if (conn->hpd.event)
+		nouveau_event_put(conn->hpd.event);
+	return nouveau_object_fini(&conn->base, suspend);
+}
+
+int
+_nvkm_connector_init(struct nouveau_object *object)
+{
+	struct nvkm_connector *conn = (void *)object;
+	int ret = nouveau_object_init(&conn->base);
+	if (ret == 0) {
+		if (conn->hpd.event)
+			nouveau_event_get(conn->hpd.event);
+	}
+	return ret;
+}
+
+void
+_nvkm_connector_dtor(struct nouveau_object *object)
+{
+	struct nvkm_connector *conn = (void *)object;
+	nouveau_event_ref(NULL, &conn->hpd.event);
+	nouveau_object_destroy(&conn->base);
+}
+
+int
+nvkm_connector_create_(struct nouveau_object *parent,
+		       struct nouveau_object *engine,
+		       struct nouveau_oclass *oclass,
+		       struct nvbios_connE *info, int index,
+		       int length, void **pobject)
+{
+	static const u8 hpd[] = { 0x07, 0x08, 0x51, 0x52, 0x5e, 0x5f, 0x60 };
+	struct nouveau_gpio *gpio = nouveau_gpio(parent);
+	struct nouveau_disp *disp = (void *)engine;
+	struct nvkm_connector *conn;
+	struct nvkm_output *outp;
+	struct dcb_gpio_func func;
+	int ret;
+
+	list_for_each_entry(outp, &disp->outp, head) {
+		if (outp->conn && outp->conn->index == index) {
+			atomic_inc(&nv_object(outp->conn)->refcount);
+			*pobject = outp->conn;
+			return 1;
+		}
+	}
+
+	ret = nouveau_object_create_(parent, engine, oclass, 0, length, pobject);
+	conn = *pobject;
+	if (ret)
+		return ret;
+
+	conn->info = *info;
+	conn->index = index;
+
+	DBG("type %02x loc %d hpd %02x dp %x di %x sr %x lcdid %x\n",
+	    info->type, info->location, info->hpd, info->dp,
+	    info->di, info->sr, info->lcdid);
+
+	if ((info->hpd = ffs(info->hpd))) {
+		if (--info->hpd >= ARRAY_SIZE(hpd)) {
+			ERR("hpd %02x unknown\n", info->hpd);
+			goto done;
+		}
+		info->hpd = hpd[info->hpd];
+
+		ret = gpio->find(gpio, 0, info->hpd, DCB_GPIO_UNUSED, &func);
+		if (ret) {
+			ERR("func %02x lookup failed, %d\n", info->hpd, ret);
+			goto done;
+		}
+
+		ret = nouveau_event_new(gpio->events, NVKM_GPIO_TOGGLED,
+					func.line, nvkm_connector_hpd,
+					conn, &conn->hpd.event);
+		if (ret) {
+			ERR("func %02x failed, %d\n", info->hpd, ret);
+		} else {
+			DBG("func %02x (HPD)\n", info->hpd);
+		}
+	}
+
+done:
+	INIT_WORK(&conn->hpd.work, nvkm_connector_hpd_work);
+	return 0;
+}
+
+int
+_nvkm_connector_ctor(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, void *info, u32 index,
+		     struct nouveau_object **pobject)
+{
+	struct nvkm_connector *conn;
+	int ret;
+
+	ret = nvkm_connector_create(parent, engine, oclass, info, index, &conn);
+	*pobject = nv_object(conn);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass *
+nvkm_connector_oclass = &(struct nvkm_connector_impl) {
+	.base = {
+		.handle = 0,
+		.ofuncs = &(struct nouveau_ofuncs) {
+			.ctor = _nvkm_connector_ctor,
+			.dtor = _nvkm_connector_dtor,
+			.init = _nvkm_connector_init,
+			.fini = _nvkm_connector_fini,
+		},
+	},
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/conn.h b/drivers/gpu/drm/nouveau/core/engine/disp/conn.h
new file mode 100644
index 000000000000..035ebeacbb1c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/conn.h
@@ -0,0 +1,59 @@
+#ifndef __NVKM_DISP_CONN_H__
+#define __NVKM_DISP_CONN_H__
+
+#include "priv.h"
+
+struct nvkm_connector {
+	struct nouveau_object base;
+	struct list_head head;
+
+	struct nvbios_connE info;
+	int index;
+
+	struct {
+		struct nouveau_eventh *event;
+		struct work_struct work;
+	} hpd;
+};
+
+#define nvkm_connector_create(p,e,c,b,i,d)                                     \
+	nvkm_connector_create_((p), (e), (c), (b), (i), sizeof(**d), (void **)d)
+#define nvkm_connector_destroy(d) ({                                           \
+	struct nvkm_connector *disp = (d);                                     \
+	_nvkm_connector_dtor(nv_object(disp));                                 \
+})
+#define nvkm_connector_init(d) ({                                              \
+	struct nvkm_connector *disp = (d);                                     \
+	_nvkm_connector_init(nv_object(disp));                                 \
+})
+#define nvkm_connector_fini(d,s) ({                                            \
+	struct nvkm_connector *disp = (d);                                     \
+	_nvkm_connector_fini(nv_object(disp), (s));                            \
+})
+
+int nvkm_connector_create_(struct nouveau_object *, struct nouveau_object *,
+			   struct nouveau_oclass *, struct nvbios_connE *,
+			   int, int, void **);
+
+int  _nvkm_connector_ctor(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, void *, u32,
+			  struct nouveau_object **);
+void _nvkm_connector_dtor(struct nouveau_object *);
+int  _nvkm_connector_init(struct nouveau_object *);
+int  _nvkm_connector_fini(struct nouveau_object *, bool);
+
+struct nvkm_connector_impl {
+	struct nouveau_oclass base;
+};
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nvkm_connector *_conn = (void *)conn;                           \
+	nv_##l(nv_object(conn)->engine, "%02x:%02x%02x: "f, _conn->index,      \
+	       _conn->info.location, _conn->info.type, ##a);                   \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
index 3ca2d25b7f5e..5a5b59b21130 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
@@ -30,42 +30,38 @@
 
 #include <engine/disp.h>
 
-#include "dport.h"
+#include <core/class.h>
 
-#define DBG(fmt, args...) nv_debug(dp->disp, "DP:%04x:%04x: " fmt,             \
-				   dp->outp->hasht, dp->outp->hashm, ##args)
-#define ERR(fmt, args...) nv_error(dp->disp, "DP:%04x:%04x: " fmt,             \
-				   dp->outp->hasht, dp->outp->hashm, ##args)
+#include "dport.h"
+#include "outpdp.h"
 
 /******************************************************************************
  * link training
  *****************************************************************************/
 struct dp_state {
-	const struct nouveau_dp_func *func;
-	struct nouveau_disp *disp;
-	struct dcb_output *outp;
-	struct nvbios_dpout info;
-	u8 version;
-	struct nouveau_i2c_port *aux;
-	int head;
-	u8  dpcd[4];
+	struct nvkm_output_dp *outp;
 	int link_nr;
 	u32 link_bw;
 	u8  stat[6];
 	u8  conf[4];
+	bool pc2;
+	u8  pc2stat;
+	u8  pc2conf[2];
 };
 
 static int
 dp_set_link_config(struct dp_state *dp)
 {
-	struct nouveau_disp *disp = dp->disp;
+	struct nvkm_output_dp_impl *impl = (void *)nv_oclass(dp->outp);
+	struct nvkm_output_dp *outp = dp->outp;
+	struct nouveau_disp *disp = nouveau_disp(outp);
 	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nvbios_init init = {
-		.subdev = nv_subdev(dp->disp),
+		.subdev = nv_subdev(disp),
 		.bios = bios,
 		.offset = 0x0000,
-		.outp = dp->outp,
-		.crtc = dp->head,
+		.outp = &outp->base.info,
+		.crtc = -1,
 		.execute = 1,
 	};
 	u32 lnkcmp;
@@ -75,8 +71,8 @@ dp_set_link_config(struct dp_state *dp)
 	DBG("%d lanes at %d KB/s\n", dp->link_nr, dp->link_bw);
 
 	/* set desired link configuration on the source */
-	if ((lnkcmp = dp->info.lnkcmp)) {
-		if (dp->version < 0x30) {
+	if ((lnkcmp = dp->outp->info.lnkcmp)) {
+		if (outp->version < 0x30) {
 			while ((dp->link_bw / 10) < nv_ro16(bios, lnkcmp))
 				lnkcmp += 4;
 			init.offset = nv_ro16(bios, lnkcmp + 2);
@@ -89,73 +85,112 @@ dp_set_link_config(struct dp_state *dp)
 		nvbios_exec(&init);
 	}
 
-	ret = dp->func->lnk_ctl(dp->disp, dp->outp, dp->head,
-				dp->link_nr, dp->link_bw / 27000,
-				dp->dpcd[DPCD_RC02] &
-					 DPCD_RC02_ENHANCED_FRAME_CAP);
+	ret = impl->lnk_ctl(outp, dp->link_nr, dp->link_bw / 27000,
+			    outp->dpcd[DPCD_RC02] &
+				       DPCD_RC02_ENHANCED_FRAME_CAP);
 	if (ret) {
-		ERR("lnk_ctl failed with %d\n", ret);
+		if (ret < 0)
+			ERR("lnk_ctl failed with %d\n", ret);
 		return ret;
 	}
 
+	impl->lnk_pwr(outp, dp->link_nr);
+
 	/* set desired link configuration on the sink */
 	sink[0] = dp->link_bw / 27000;
 	sink[1] = dp->link_nr;
-	if (dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
+	if (outp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
 		sink[1] |= DPCD_LC01_ENHANCED_FRAME_EN;
 
-	return nv_wraux(dp->aux, DPCD_LC00, sink, 2);
+	return nv_wraux(outp->base.edid, DPCD_LC00_LINK_BW_SET, sink, 2);
 }
 
 static void
 dp_set_training_pattern(struct dp_state *dp, u8 pattern)
 {
+	struct nvkm_output_dp_impl *impl = (void *)nv_oclass(dp->outp);
+	struct nvkm_output_dp *outp = dp->outp;
 	u8 sink_tp;
 
 	DBG("training pattern %d\n", pattern);
-	dp->func->pattern(dp->disp, dp->outp, dp->head, pattern);
+	impl->pattern(outp, pattern);
 
-	nv_rdaux(dp->aux, DPCD_LC02, &sink_tp, 1);
+	nv_rdaux(outp->base.edid, DPCD_LC02, &sink_tp, 1);
 	sink_tp &= ~DPCD_LC02_TRAINING_PATTERN_SET;
 	sink_tp |= pattern;
-	nv_wraux(dp->aux, DPCD_LC02, &sink_tp, 1);
+	nv_wraux(outp->base.edid, DPCD_LC02, &sink_tp, 1);
 }
 
 static int
-dp_link_train_commit(struct dp_state *dp)
+dp_link_train_commit(struct dp_state *dp, bool pc)
 {
-	int i;
+	struct nvkm_output_dp_impl *impl = (void *)nv_oclass(dp->outp);
+	struct nvkm_output_dp *outp = dp->outp;
+	int ret, i;
 
 	for (i = 0; i < dp->link_nr; i++) {
 		u8 lane = (dp->stat[4 + (i >> 1)] >> ((i & 1) * 4)) & 0xf;
+		u8 lpc2 = (dp->pc2stat >> (i * 2)) & 0x3;
 		u8 lpre = (lane & 0x0c) >> 2;
 		u8 lvsw = (lane & 0x03) >> 0;
+		u8 hivs = 3 - lpre;
+		u8 hipe = 3;
+		u8 hipc = 3;
+
+		if (lpc2 >= hipc)
+			lpc2 = hipc | DPCD_LC0F_LANE0_MAX_POST_CURSOR2_REACHED;
+		if (lpre >= hipe) {
+			lpre = hipe | DPCD_LC03_MAX_SWING_REACHED; /* yes. */
+			lvsw = hivs = 3 - (lpre & 3);
+		} else
+		if (lvsw >= hivs) {
+			lvsw = hivs | DPCD_LC03_MAX_SWING_REACHED;
+		}
 
 		dp->conf[i] = (lpre << 3) | lvsw;
-		if (lvsw == 3)
-			dp->conf[i] |= DPCD_LC03_MAX_SWING_REACHED;
-		if (lpre == 3)
-			dp->conf[i] |= DPCD_LC03_MAX_PRE_EMPHASIS_REACHED;
+		dp->pc2conf[i >> 1] |= lpc2 << ((i & 1) * 4);
+
+		DBG("config lane %d %02x %02x\n", i, dp->conf[i], lpc2);
+		impl->drv_ctl(outp, i, lvsw & 3, lpre & 3, lpc2 & 3);
+	}
+
+	ret = nv_wraux(outp->base.edid, DPCD_LC03(0), dp->conf, 4);
+	if (ret)
+		return ret;
 
-		DBG("config lane %d %02x\n", i, dp->conf[i]);
-		dp->func->drv_ctl(dp->disp, dp->outp, dp->head, i, lvsw, lpre);
+	if (pc) {
+		ret = nv_wraux(outp->base.edid, DPCD_LC0F, dp->pc2conf, 2);
+		if (ret)
+			return ret;
 	}
 
-	return nv_wraux(dp->aux, DPCD_LC03(0), dp->conf, 4);
+	return 0;
 }
 
 static int
-dp_link_train_update(struct dp_state *dp, u32 delay)
+dp_link_train_update(struct dp_state *dp, bool pc, u32 delay)
 {
+	struct nvkm_output_dp *outp = dp->outp;
 	int ret;
 
-	udelay(delay);
+	if (outp->dpcd[DPCD_RC0E_AUX_RD_INTERVAL])
+		mdelay(outp->dpcd[DPCD_RC0E_AUX_RD_INTERVAL] * 4);
+	else
+		udelay(delay);
 
-	ret = nv_rdaux(dp->aux, DPCD_LS02, dp->stat, 6);
+	ret = nv_rdaux(outp->base.edid, DPCD_LS02, dp->stat, 6);
 	if (ret)
 		return ret;
 
-	DBG("status %6ph\n", dp->stat);
+	if (pc) {
+		ret = nv_rdaux(outp->base.edid, DPCD_LS0C, &dp->pc2stat, 1);
+		if (ret)
+			dp->pc2stat = 0x00;
+		DBG("status %6ph pc2 %02x\n", dp->stat, dp->pc2stat);
+	} else {
+		DBG("status %6ph\n", dp->stat);
+	}
+
 	return 0;
 }
 
@@ -169,8 +204,8 @@ dp_link_train_cr(struct dp_state *dp)
 	dp_set_training_pattern(dp, 1);
 
 	do {
-		if (dp_link_train_commit(dp) ||
-		    dp_link_train_update(dp, 100))
+		if (dp_link_train_commit(dp, false) ||
+		    dp_link_train_update(dp, false, 100))
 			break;
 
 		cr_done = true;
@@ -196,13 +231,19 @@ dp_link_train_cr(struct dp_state *dp)
 static int
 dp_link_train_eq(struct dp_state *dp)
 {
+	struct nvkm_output_dp *outp = dp->outp;
 	bool eq_done = false, cr_done = true;
 	int tries = 0, i;
 
-	dp_set_training_pattern(dp, 2);
+	if (outp->dpcd[2] & DPCD_RC02_TPS3_SUPPORTED)
+		dp_set_training_pattern(dp, 3);
+	else
+		dp_set_training_pattern(dp, 2);
 
 	do {
-		if (dp_link_train_update(dp, 400))
+		if ((tries &&
+		    dp_link_train_commit(dp, dp->pc2)) ||
+		    dp_link_train_update(dp, dp->pc2, 400))
 			break;
 
 		eq_done = !!(dp->stat[2] & DPCD_LS04_INTERLANE_ALIGN_DONE);
@@ -214,9 +255,6 @@ dp_link_train_eq(struct dp_state *dp)
 			    !(lane & DPCD_LS02_LANE0_SYMBOL_LOCKED))
 				eq_done = false;
 		}
-
-		if (dp_link_train_commit(dp))
-			break;
 	} while (!eq_done && cr_done && ++tries <= 5);
 
 	return eq_done ? 0 : -1;
@@ -225,121 +263,109 @@ dp_link_train_eq(struct dp_state *dp)
 static void
 dp_link_train_init(struct dp_state *dp, bool spread)
 {
+	struct nvkm_output_dp *outp = dp->outp;
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nvbios_init init = {
-		.subdev = nv_subdev(dp->disp),
-		.bios = nouveau_bios(dp->disp),
-		.outp = dp->outp,
-		.crtc = dp->head,
+		.subdev = nv_subdev(disp),
+		.bios = bios,
+		.outp = &outp->base.info,
+		.crtc = -1,
 		.execute = 1,
 	};
 
 	/* set desired spread */
 	if (spread)
-		init.offset = dp->info.script[2];
+		init.offset = outp->info.script[2];
 	else
-		init.offset = dp->info.script[3];
+		init.offset = outp->info.script[3];
 	nvbios_exec(&init);
 
 	/* pre-train script */
-	init.offset = dp->info.script[0];
+	init.offset = outp->info.script[0];
 	nvbios_exec(&init);
 }
 
 static void
 dp_link_train_fini(struct dp_state *dp)
 {
+	struct nvkm_output_dp *outp = dp->outp;
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nvbios_init init = {
-		.subdev = nv_subdev(dp->disp),
-		.bios = nouveau_bios(dp->disp),
-		.outp = dp->outp,
-		.crtc = dp->head,
+		.subdev = nv_subdev(disp),
+		.bios = bios,
+		.outp = &outp->base.info,
+		.crtc = -1,
 		.execute = 1,
 	};
 
 	/* post-train script */
-	init.offset = dp->info.script[1],
+	init.offset = outp->info.script[1],
 	nvbios_exec(&init);
 }
 
-int
-nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
-		 struct dcb_output *outp, int head, u32 datarate)
+static const struct dp_rates {
+	u32 rate;
+	u8  bw;
+	u8  nr;
+} nouveau_dp_rates[] = {
+	{ 2160000, 0x14, 4 },
+	{ 1080000, 0x0a, 4 },
+	{ 1080000, 0x14, 2 },
+	{  648000, 0x06, 4 },
+	{  540000, 0x0a, 2 },
+	{  540000, 0x14, 1 },
+	{  324000, 0x06, 2 },
+	{  270000, 0x0a, 1 },
+	{  162000, 0x06, 1 },
+	{}
+};
+
+void
+nouveau_dp_train(struct work_struct *w)
 {
-	struct nouveau_bios *bios = nouveau_bios(disp);
-	struct nouveau_i2c *i2c = nouveau_i2c(disp);
+	struct nvkm_output_dp *outp = container_of(w, typeof(*outp), lt.work);
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	const struct dp_rates *cfg = nouveau_dp_rates;
 	struct dp_state _dp = {
-		.disp = disp,
-		.func = func,
 		.outp = outp,
-		.head = head,
 	}, *dp = &_dp;
-	const u32 bw_list[] = { 540000, 270000, 162000, 0 };
-	const u32 *link_bw = bw_list;
-	u8  hdr, cnt, len;
-	u32 data;
+	u32 datarate = 0;
 	int ret;
 
-	/* find the bios displayport data relevant to this output */
-	data = nvbios_dpout_match(bios, outp->hasht, outp->hashm, &dp->version,
-				 &hdr, &cnt, &len, &dp->info);
-	if (!data) {
-		ERR("bios data not found\n");
-		return -EINVAL;
-	}
-
-	/* acquire the aux channel and fetch some info about the display */
-	if (outp->location)
-		dp->aux = i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(outp->extdev));
-	else
-		dp->aux = i2c->find(i2c, NV_I2C_TYPE_DCBI2C(outp->i2c_index));
-	if (!dp->aux) {
-		ERR("no aux channel?!\n");
-		return -ENODEV;
-	}
-
-	ret = nv_rdaux(dp->aux, 0x00000, dp->dpcd, sizeof(dp->dpcd));
-	if (ret) {
-		/* it's possible the display has been unplugged before we
-		 * get here.  we still need to execute the full set of
-		 * vbios scripts, and program the OR at a high enough
-		 * frequency to satisfy the target mode.  failure to do
-		 * so results at best in an UPDATE hanging, and at worst
-		 * with PDISP running away to join the circus.
-		 */
-		dp->dpcd[1] = link_bw[0] / 27000;
-		dp->dpcd[2] = 4;
-		dp->dpcd[3] = 0x00;
-		ERR("failed to read DPCD\n");
-	}
-
 	/* bring capabilities within encoder limits */
-	if ((dp->dpcd[2] & 0x1f) > dp->outp->dpconf.link_nr) {
-		dp->dpcd[2] &= ~0x1f;
-		dp->dpcd[2] |= dp->outp->dpconf.link_nr;
+	if (nv_mclass(disp) < NVD0_DISP_CLASS)
+		outp->dpcd[2] &= ~DPCD_RC02_TPS3_SUPPORTED;
+	if ((outp->dpcd[2] & 0x1f) > outp->base.info.dpconf.link_nr) {
+		outp->dpcd[2] &= ~DPCD_RC02_MAX_LANE_COUNT;
+		outp->dpcd[2] |= outp->base.info.dpconf.link_nr;
+	}
+	if (outp->dpcd[1] > outp->base.info.dpconf.link_bw)
+		outp->dpcd[1] = outp->base.info.dpconf.link_bw;
+	dp->pc2 = outp->dpcd[2] & DPCD_RC02_TPS3_SUPPORTED;
+
+	/* restrict link config to the lowest required rate, if requested */
+	if (datarate) {
+		datarate = (datarate / 8) * 10; /* 8B/10B coding overhead */
+		while (cfg[1].rate >= datarate)
+			cfg++;
 	}
-	if (dp->dpcd[1] > dp->outp->dpconf.link_bw)
-		dp->dpcd[1] = dp->outp->dpconf.link_bw;
+	cfg--;
 
-	/* adjust required bandwidth for 8B/10B coding overhead */
-	datarate = (datarate / 8) * 10;
+	/* disable link interrupt handling during link training */
+	nouveau_event_put(outp->irq);
 
 	/* enable down-spreading and execute pre-train script from vbios */
-	dp_link_train_init(dp, dp->dpcd[3] & 0x01);
+	dp_link_train_init(dp, outp->dpcd[3] & 0x01);
 
-	/* start off at highest link rate supported by encoder and display */
-	while (*link_bw > (dp->dpcd[1] * 27000))
-		link_bw++;
-
-	while ((ret = -EIO) && link_bw[0]) {
-		/* find minimum required lane count at this link rate */
-		dp->link_nr = dp->dpcd[2] & DPCD_RC02_MAX_LANE_COUNT;
-		while ((dp->link_nr >> 1) * link_bw[0] > datarate)
-			dp->link_nr >>= 1;
-
-		/* drop link rate to minimum with this lane count */
-		while ((link_bw[1] * dp->link_nr) > datarate)
-			link_bw++;
-		dp->link_bw = link_bw[0];
+	while (ret = -EIO, (++cfg)->rate) {
+		/* select next configuration supported by encoder and sink */
+		while (cfg->nr > (outp->dpcd[2] & DPCD_RC02_MAX_LANE_COUNT) ||
+		       cfg->bw > (outp->dpcd[DPCD_RC01_MAX_LINK_RATE]))
+			cfg++;
+		dp->link_bw = cfg->bw * 27000;
+		dp->link_nr = cfg->nr;
 
 		/* program selected link configuration */
 		ret = dp_set_link_config(dp);
@@ -356,17 +382,18 @@ nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
 			 */
 			break;
 		}
-
-		/* retry at lower rate */
-		link_bw++;
 	}
 
-	/* finish link training */
+	/* finish link training and execute post-train script from vbios */
 	dp_set_training_pattern(dp, 0);
 	if (ret < 0)
 		ERR("link training failed\n");
 
-	/* execute post-train script from vbios */
 	dp_link_train_fini(dp);
-	return (ret < 0) ? false : true;
+
+	/* signal completion and enable link interrupt handling */
+	DBG("training complete\n");
+	atomic_set(&outp->lt.done, 1);
+	wake_up(&outp->lt.wait);
+	nouveau_event_get(outp->irq);
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.h b/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
index 0e1bbd18ff6c..5628d2d5ec71 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
@@ -2,19 +2,18 @@
 #define __NVKM_DISP_DPORT_H__
 
 /* DPCD Receiver Capabilities */
-#define DPCD_RC00                                                       0x00000
-#define DPCD_RC00_DPCD_REV                                                 0xff
-#define DPCD_RC01                                                       0x00001
-#define DPCD_RC01_MAX_LINK_RATE                                            0xff
+#define DPCD_RC00_DPCD_REV                                              0x00000
+#define DPCD_RC01_MAX_LINK_RATE                                         0x00001
 #define DPCD_RC02                                                       0x00002
 #define DPCD_RC02_ENHANCED_FRAME_CAP                                       0x80
+#define DPCD_RC02_TPS3_SUPPORTED                                           0x40
 #define DPCD_RC02_MAX_LANE_COUNT                                           0x1f
 #define DPCD_RC03                                                       0x00003
 #define DPCD_RC03_MAX_DOWNSPREAD                                           0x01
+#define DPCD_RC0E_AUX_RD_INTERVAL                                       0x0000e
 
 /* DPCD Link Configuration */
-#define DPCD_LC00                                                       0x00100
-#define DPCD_LC00_LINK_BW_SET                                              0xff
+#define DPCD_LC00_LINK_BW_SET                                           0x00100
 #define DPCD_LC01                                                       0x00101
 #define DPCD_LC01_ENHANCED_FRAME_EN                                        0x80
 #define DPCD_LC01_LANE_COUNT_SET                                           0x1f
@@ -25,6 +24,16 @@
 #define DPCD_LC03_PRE_EMPHASIS_SET                                         0x18
 #define DPCD_LC03_MAX_SWING_REACHED                                        0x04
 #define DPCD_LC03_VOLTAGE_SWING_SET                                        0x03
+#define DPCD_LC0F                                                       0x0010f
+#define DPCD_LC0F_LANE1_MAX_POST_CURSOR2_REACHED                           0x40
+#define DPCD_LC0F_LANE1_POST_CURSOR2_SET                                   0x30
+#define DPCD_LC0F_LANE0_MAX_POST_CURSOR2_REACHED                           0x04
+#define DPCD_LC0F_LANE0_POST_CURSOR2_SET                                   0x03
+#define DPCD_LC10                                                       0x00110
+#define DPCD_LC10_LANE3_MAX_POST_CURSOR2_REACHED                           0x40
+#define DPCD_LC10_LANE3_POST_CURSOR2_SET                                   0x30
+#define DPCD_LC10_LANE2_MAX_POST_CURSOR2_REACHED                           0x04
+#define DPCD_LC10_LANE2_POST_CURSOR2_SET                                   0x03
 
 /* DPCD Link/Sink Status */
 #define DPCD_LS02                                                       0x00202
@@ -55,24 +64,12 @@
 #define DPCD_LS07_LANE3_VOLTAGE_SWING                                      0x30
 #define DPCD_LS07_LANE2_PRE_EMPHASIS                                       0x0c
 #define DPCD_LS07_LANE2_VOLTAGE_SWING                                      0x03
+#define DPCD_LS0C                                                       0x0020c
+#define DPCD_LS0C_LANE3_POST_CURSOR2                                       0xc0
+#define DPCD_LS0C_LANE2_POST_CURSOR2                                       0x30
+#define DPCD_LS0C_LANE1_POST_CURSOR2                                       0x0c
+#define DPCD_LS0C_LANE0_POST_CURSOR2                                       0x03
 
-struct nouveau_disp;
-struct dcb_output;
-
-struct nouveau_dp_func {
-	int (*pattern)(struct nouveau_disp *, struct dcb_output *,
-		       int head, int pattern);
-	int (*lnk_ctl)(struct nouveau_disp *, struct dcb_output *, int head,
-		       int link_nr, int link_bw, bool enh_frame);
-	int (*drv_ctl)(struct nouveau_disp *, struct dcb_output *, int head,
-		       int lane, int swing, int preem);
-};
-
-extern const struct nouveau_dp_func nv94_sor_dp_func;
-extern const struct nouveau_dp_func nvd0_sor_dp_func;
-extern const struct nouveau_dp_func nv50_pior_dp_func;
-
-int nouveau_dp_train(struct nouveau_disp *, const struct nouveau_dp_func *,
-		     struct dcb_output *, int, u32);
+void nouveau_dp_train(struct work_struct *);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c b/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c
index cf6f59677b74..9fc7447fec90 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c
@@ -81,7 +81,6 @@ gm107_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
@@ -94,6 +93,7 @@ gm107_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nve0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nve0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
index 6c89af792889..a32666ed0c47 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
@@ -51,6 +51,14 @@ nv04_disp_scanoutpos(struct nouveau_object *object, u32 mthd,
 	args->htotal  = nv_rd32(priv, 0x680824 + (head * 0x2000)) & 0xffff;
 	args->hblanke = args->htotal - 1;
 
+	/*
+	 * If output is vga instead of digital then vtotal/htotal is invalid
+	 * so we have to give up and trigger the timestamping fallback in the
+	 * drm core.
+	 */
+	if (!args->vtotal || !args->htotal)
+		return -ENOTSUPP;
+
 	args->time[0] = ktime_to_ns(ktime_get());
 	line = nv_rd32(priv, 0x600868 + (head * 0x2000));
 	args->time[1] = ktime_to_ns(ktime_get());
@@ -78,13 +86,13 @@ nv04_disp_sclass[] = {
  ******************************************************************************/
 
 static void
-nv04_disp_vblank_enable(struct nouveau_event *event, int head)
+nv04_disp_vblank_enable(struct nouveau_event *event, int type, int head)
 {
 	nv_wr32(event->priv, 0x600140 + (head * 0x2000) , 0x00000001);
 }
 
 static void
-nv04_disp_vblank_disable(struct nouveau_event *event, int head)
+nv04_disp_vblank_disable(struct nouveau_event *event, int type, int head)
 {
 	nv_wr32(event->priv, 0x600140 + (head * 0x2000) , 0x00000000);
 }
@@ -98,12 +106,12 @@ nv04_disp_intr(struct nouveau_subdev *subdev)
 	u32 pvideo;
 
 	if (crtc0 & 0x00000001) {
-		nouveau_event_trigger(priv->base.vblank, 0);
+		nouveau_event_trigger(priv->base.vblank, 1, 0);
 		nv_wr32(priv, 0x600100, 0x00000001);
 	}
 
 	if (crtc1 & 0x00000001) {
-		nouveau_event_trigger(priv->base.vblank, 1);
+		nouveau_event_trigger(priv->base.vblank, 1, 1);
 		nv_wr32(priv, 0x602100, 0x00000001);
 	}
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
index 9a0cab9c3adb..2283c442a10d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
@@ -829,13 +829,13 @@ nv50_disp_base_scanoutpos(struct nouveau_object *object, u32 mthd,
 }
 
 static void
-nv50_disp_base_vblank_enable(struct nouveau_event *event, int head)
+nv50_disp_base_vblank_enable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x61002c, (4 << head), (4 << head));
 }
 
 static void
-nv50_disp_base_vblank_disable(struct nouveau_event *event, int head)
+nv50_disp_base_vblank_disable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x61002c, (4 << head), 0);
 }
@@ -1114,19 +1114,20 @@ nv50_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 	nv_wr32(priv, 0x610080 + (chid * 0x08), 0x90000000);
 }
 
-static u16
-exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
-	    struct dcb_output *dcb, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+static struct nvkm_output *
+exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl,
+	    u32 *data, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 	    struct nvbios_outp *info)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
-	u16 mask, type, data;
+	struct nvkm_output *outp;
+	u16 mask, type;
 
-	if (outp < 4) {
+	if (or < 4) {
 		type = DCB_OUTPUT_ANALOG;
 		mask = 0;
 	} else
-	if (outp < 8) {
+	if (or < 8) {
 		switch (ctrl & 0x00000f00) {
 		case 0x00000000: type = DCB_OUTPUT_LVDS; mask = 1; break;
 		case 0x00000100: type = DCB_OUTPUT_TMDS; mask = 1; break;
@@ -1136,45 +1137,48 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 		case 0x00000900: type = DCB_OUTPUT_DP; mask = 2; break;
 		default:
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
-			return 0x0000;
+			return NULL;
 		}
-		outp -= 4;
+		or  -= 4;
 	} else {
-		outp = outp - 8;
+		or   = or - 8;
 		type = 0x0010;
 		mask = 0;
 		switch (ctrl & 0x00000f00) {
-		case 0x00000000: type |= priv->pior.type[outp]; break;
+		case 0x00000000: type |= priv->pior.type[or]; break;
 		default:
 			nv_error(priv, "unknown PIOR mc 0x%08x\n", ctrl);
-			return 0x0000;
+			return NULL;
 		}
 	}
 
 	mask  = 0x00c0 & (mask << 6);
-	mask |= 0x0001 << outp;
+	mask |= 0x0001 << or;
 	mask |= 0x0100 << head;
 
-	data = dcb_outp_match(bios, type, mask, ver, hdr, dcb);
-	if (!data)
-		return 0x0000;
-
-	/* off-chip encoders require matching the exact encoder type */
-	if (dcb->location != 0)
-		type |= dcb->extdev << 8;
+	list_for_each_entry(outp, &priv->base.outp, head) {
+		if ((outp->info.hasht & 0xff) == type &&
+		    (outp->info.hashm & mask) == mask) {
+			*data = nvbios_outp_match(bios, outp->info.hasht,
+							outp->info.hashm,
+						  ver, hdr, cnt, len, info);
+			if (!*data)
+				return NULL;
+			return outp;
+		}
+	}
 
-	return nvbios_outp_match(bios, type, mask, ver, hdr, cnt, len, info);
+	return NULL;
 }
 
-static bool
+static struct nvkm_output *
 exec_script(struct nv50_disp_priv *priv, int head, int id)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info;
-	struct dcb_output dcb;
 	u8  ver, hdr, cnt, len;
-	u16 data;
-	u32 ctrl = 0x00000000;
+	u32 data, ctrl = 0;
 	u32 reg;
 	int i;
 
@@ -1204,36 +1208,35 @@ exec_script(struct nv50_disp_priv *priv, int head, int id)
 	}
 
 	if (!(ctrl & (1 << head)))
-		return false;
+		return NULL;
 	i--;
 
-	data = exec_lookup(priv, head, i, ctrl, &dcb, &ver, &hdr, &cnt, &len, &info);
-	if (data) {
+	outp = exec_lookup(priv, head, i, ctrl, &data, &ver, &hdr, &cnt, &len, &info);
+	if (outp) {
 		struct nvbios_init init = {
 			.subdev = nv_subdev(priv),
 			.bios = bios,
 			.offset = info.script[id],
-			.outp = &dcb,
+			.outp = &outp->info,
 			.crtc = head,
 			.execute = 1,
 		};
 
-		return nvbios_exec(&init) == 0;
+		nvbios_exec(&init);
 	}
 
-	return false;
+	return outp;
 }
 
-static u32
-exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
-	    struct dcb_output *outp)
+static struct nvkm_output *
+exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk, u32 *conf)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info1;
 	struct nvbios_ocfg info2;
 	u8  ver, hdr, cnt, len;
-	u32 ctrl = 0x00000000;
-	u32 data, conf = ~0;
+	u32 data, ctrl = 0;
 	u32 reg;
 	int i;
 
@@ -1263,37 +1266,37 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 	}
 
 	if (!(ctrl & (1 << head)))
-		return conf;
+		return NULL;
 	i--;
 
-	data = exec_lookup(priv, head, i, ctrl, outp, &ver, &hdr, &cnt, &len, &info1);
-	if (!data)
-		return conf;
+	outp = exec_lookup(priv, head, i, ctrl, &data, &ver, &hdr, &cnt, &len, &info1);
+	if (!outp)
+		return NULL;
 
-	if (outp->location == 0) {
-		switch (outp->type) {
+	if (outp->info.location == 0) {
+		switch (outp->info.type) {
 		case DCB_OUTPUT_TMDS:
-			conf = (ctrl & 0x00000f00) >> 8;
+			*conf = (ctrl & 0x00000f00) >> 8;
 			if (pclk >= 165000)
-				conf |= 0x0100;
+				*conf |= 0x0100;
 			break;
 		case DCB_OUTPUT_LVDS:
-			conf = priv->sor.lvdsconf;
+			*conf = priv->sor.lvdsconf;
 			break;
 		case DCB_OUTPUT_DP:
-			conf = (ctrl & 0x00000f00) >> 8;
+			*conf = (ctrl & 0x00000f00) >> 8;
 			break;
 		case DCB_OUTPUT_ANALOG:
 		default:
-			conf = 0x00ff;
+			*conf = 0x00ff;
 			break;
 		}
 	} else {
-		conf = (ctrl & 0x00000f00) >> 8;
+		*conf = (ctrl & 0x00000f00) >> 8;
 		pclk = pclk / 2;
 	}
 
-	data = nvbios_ocfg_match(bios, data, conf, &ver, &hdr, &cnt, &len, &info2);
+	data = nvbios_ocfg_match(bios, data, *conf, &ver, &hdr, &cnt, &len, &info2);
 	if (data && id < 0xff) {
 		data = nvbios_oclk_match(bios, info2.clkcmp[id], pclk);
 		if (data) {
@@ -1301,7 +1304,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 				.subdev = nv_subdev(priv),
 				.bios = bios,
 				.offset = data,
-				.outp = outp,
+				.outp = &outp->info,
 				.crtc = head,
 				.execute = 1,
 			};
@@ -1310,7 +1313,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 		}
 	}
 
-	return conf;
+	return outp;
 }
 
 static void
@@ -1322,7 +1325,35 @@ nv50_disp_intr_unk10_0(struct nv50_disp_priv *priv, int head)
 static void
 nv50_disp_intr_unk20_0(struct nv50_disp_priv *priv, int head)
 {
-	exec_script(priv, head, 2);
+	struct nvkm_output *outp = exec_script(priv, head, 2);
+
+	/* the binary driver does this outside of the supervisor handling
+	 * (after the third supervisor from a detach).  we (currently?)
+	 * allow both detach/attach to happen in the same set of
+	 * supervisor interrupts, so it would make sense to execute this
+	 * (full power down?) script after all the detach phases of the
+	 * supervisor handling.  like with training if needed from the
+	 * second supervisor, nvidia doesn't do this, so who knows if it's
+	 * entirely safe, but it does appear to work..
+	 *
+	 * without this script being run, on some configurations i've
+	 * seen, switching from DP to TMDS on a DP connector may result
+	 * in a blank screen (SOR_PWR off/on can restore it)
+	 */
+	if (outp && outp->info.type == DCB_OUTPUT_DP) {
+		struct nvkm_output_dp *outpdp = (void *)outp;
+		struct nvbios_init init = {
+			.subdev = nv_subdev(priv),
+			.bios = nouveau_bios(priv),
+			.outp = &outp->info,
+			.crtc = head,
+			.offset = outpdp->info.script[4],
+			.execute = 1,
+		};
+
+		nvbios_exec(&init);
+		atomic_set(&outpdp->lt.done, 0);
+	}
 }
 
 static void
@@ -1444,56 +1475,83 @@ nv50_disp_intr_unk20_2_dp(struct nv50_disp_priv *priv,
 static void
 nv50_disp_intr_unk20_2(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
+	struct nvkm_output *outp;
 	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
 	u32 hval, hreg = 0x614200 + (head * 0x800);
 	u32 oval, oreg;
-	u32 mask;
-	u32 conf = exec_clkcmp(priv, head, 0xff, pclk, &outp);
-	if (conf != ~0) {
-		if (outp.location == 0 && outp.type == DCB_OUTPUT_DP) {
-			u32 soff = (ffs(outp.or) - 1) * 0x08;
-			u32 ctrl = nv_rd32(priv, 0x610794 + soff);
-			u32 datarate;
-
-			switch ((ctrl & 0x000f0000) >> 16) {
-			case 6: datarate = pclk * 30 / 8; break;
-			case 5: datarate = pclk * 24 / 8; break;
-			case 2:
-			default:
-				datarate = pclk * 18 / 8;
-				break;
-			}
+	u32 mask, conf;
 
-			nouveau_dp_train(&priv->base, priv->sor.dp,
-					 &outp, head, datarate);
-		}
+	outp = exec_clkcmp(priv, head, 0xff, pclk, &conf);
+	if (!outp)
+		return;
+
+	/* we allow both encoder attach and detach operations to occur
+	 * within a single supervisor (ie. modeset) sequence.  the
+	 * encoder detach scripts quite often switch off power to the
+	 * lanes, which requires the link to be re-trained.
+	 *
+	 * this is not generally an issue as the sink "must" (heh)
+	 * signal an irq when it's lost sync so the driver can
+	 * re-train.
+	 *
+	 * however, on some boards, if one does not configure at least
+	 * the gpu side of the link *before* attaching, then various
+	 * things can go horribly wrong (PDISP disappearing from mmio,
+	 * third supervisor never happens, etc).
+	 *
+	 * the solution is simply to retrain here, if necessary.  last
+	 * i checked, the binary driver userspace does not appear to
+	 * trigger this situation (it forces an UPDATE between steps).
+	 */
+	if (outp->info.type == DCB_OUTPUT_DP) {
+		u32 soff = (ffs(outp->info.or) - 1) * 0x08;
+		u32 ctrl, datarate;
 
-		exec_clkcmp(priv, head, 0, pclk, &outp);
-
-		if (!outp.location && outp.type == DCB_OUTPUT_ANALOG) {
-			oreg = 0x614280 + (ffs(outp.or) - 1) * 0x800;
-			oval = 0x00000000;
-			hval = 0x00000000;
-			mask = 0xffffffff;
-		} else
-		if (!outp.location) {
-			if (outp.type == DCB_OUTPUT_DP)
-				nv50_disp_intr_unk20_2_dp(priv, &outp, pclk);
-			oreg = 0x614300 + (ffs(outp.or) - 1) * 0x800;
-			oval = (conf & 0x0100) ? 0x00000101 : 0x00000000;
-			hval = 0x00000000;
-			mask = 0x00000707;
+		if (outp->info.location == 0) {
+			ctrl = nv_rd32(priv, 0x610794 + soff);
+			soff = 1;
 		} else {
-			oreg = 0x614380 + (ffs(outp.or) - 1) * 0x800;
-			oval = 0x00000001;
-			hval = 0x00000001;
-			mask = 0x00000707;
+			ctrl = nv_rd32(priv, 0x610b80 + soff);
+			soff = 2;
+		}
+
+		switch ((ctrl & 0x000f0000) >> 16) {
+		case 6: datarate = pclk * 30; break;
+		case 5: datarate = pclk * 24; break;
+		case 2:
+		default:
+			datarate = pclk * 18;
+			break;
 		}
 
-		nv_mask(priv, hreg, 0x0000000f, hval);
-		nv_mask(priv, oreg, mask, oval);
+		if (nvkm_output_dp_train(outp, datarate / soff, true))
+			ERR("link not trained before attach\n");
 	}
+
+	exec_clkcmp(priv, head, 0, pclk, &conf);
+
+	if (!outp->info.location && outp->info.type == DCB_OUTPUT_ANALOG) {
+		oreg = 0x614280 + (ffs(outp->info.or) - 1) * 0x800;
+		oval = 0x00000000;
+		hval = 0x00000000;
+		mask = 0xffffffff;
+	} else
+	if (!outp->info.location) {
+		if (outp->info.type == DCB_OUTPUT_DP)
+			nv50_disp_intr_unk20_2_dp(priv, &outp->info, pclk);
+		oreg = 0x614300 + (ffs(outp->info.or) - 1) * 0x800;
+		oval = (conf & 0x0100) ? 0x00000101 : 0x00000000;
+		hval = 0x00000000;
+		mask = 0x00000707;
+	} else {
+		oreg = 0x614380 + (ffs(outp->info.or) - 1) * 0x800;
+		oval = 0x00000001;
+		hval = 0x00000001;
+		mask = 0x00000707;
+	}
+
+	nv_mask(priv, hreg, 0x0000000f, hval);
+	nv_mask(priv, oreg, mask, oval);
 }
 
 /* If programming a TMDS output on a SOR that can also be configured for
@@ -1521,30 +1579,16 @@ nv50_disp_intr_unk40_0_tmds(struct nv50_disp_priv *priv, struct dcb_output *outp
 static void
 nv50_disp_intr_unk40_0(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
+	struct nvkm_output *outp;
 	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
-	if (exec_clkcmp(priv, head, 1, pclk, &outp) != ~0) {
-		if (outp.location == 0 && outp.type == DCB_OUTPUT_TMDS)
-			nv50_disp_intr_unk40_0_tmds(priv, &outp);
-		else
-		if (outp.location == 1 && outp.type == DCB_OUTPUT_DP) {
-			u32 soff = (ffs(outp.or) - 1) * 0x08;
-			u32 ctrl = nv_rd32(priv, 0x610b84 + soff);
-			u32 datarate;
-
-			switch ((ctrl & 0x000f0000) >> 16) {
-			case 6: datarate = pclk * 30 / 8; break;
-			case 5: datarate = pclk * 24 / 8; break;
-			case 2:
-			default:
-				datarate = pclk * 18 / 8;
-				break;
-			}
+	u32 conf;
 
-			nouveau_dp_train(&priv->base, priv->pior.dp,
-					 &outp, head, datarate);
-		}
-	}
+	outp = exec_clkcmp(priv, head, 1, pclk, &conf);
+	if (!outp)
+		return;
+
+	if (outp->info.location == 0 && outp->info.type == DCB_OUTPUT_TMDS)
+		nv50_disp_intr_unk40_0_tmds(priv, &outp->info);
 }
 
 void
@@ -1610,13 +1654,13 @@ nv50_disp_intr(struct nouveau_subdev *subdev)
 	}
 
 	if (intr1 & 0x00000004) {
-		nouveau_event_trigger(priv->base.vblank, 0);
+		nouveau_event_trigger(priv->base.vblank, 1, 0);
 		nv_wr32(priv, 0x610024, 0x00000004);
 		intr1 &= ~0x00000004;
 	}
 
 	if (intr1 & 0x00000008) {
-		nouveau_event_trigger(priv->base.vblank, 1);
+		nouveau_event_trigger(priv->base.vblank, 1, 1);
 		nv_wr32(priv, 0x610024, 0x00000008);
 		intr1 &= ~0x00000008;
 	}
@@ -1656,11 +1700,16 @@ nv50_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
 struct nouveau_oclass *
+nv50_disp_outp_sclass[] = {
+	&nv50_pior_dp_impl.base.base,
+	NULL
+};
+
+struct nouveau_oclass *
 nv50_disp_oclass = &(struct nv50_disp_impl) {
 	.base.base.handle = NV_ENGINE(DISP, 0x50),
 	.base.base.ofuncs = &(struct nouveau_ofuncs) {
@@ -1669,6 +1718,7 @@ nv50_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv50_disp_outp_sclass,
 	.mthd.core = &nv50_disp_mast_mthd_chan,
 	.mthd.base = &nv50_disp_sync_mthd_chan,
 	.mthd.ovly = &nv50_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
index 48d59db47f0d..1a886472b6f5 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
@@ -11,6 +11,8 @@
 
 #include "dport.h"
 #include "priv.h"
+#include "outp.h"
+#include "outpdp.h"
 
 struct nv50_disp_impl {
 	struct nouveau_disp_impl base;
@@ -43,13 +45,11 @@ struct nv50_disp_priv {
 		int (*hda_eld)(struct nv50_disp_priv *, int sor, u8 *, u32);
 		int (*hdmi)(struct nv50_disp_priv *, int head, int sor, u32);
 		u32 lvdsconf;
-		const struct nouveau_dp_func *dp;
 	} sor;
 	struct {
 		int nr;
 		int (*power)(struct nv50_disp_priv *, int ext, u32 data);
 		u8 type[3];
-		const struct nouveau_dp_func *dp;
 	} pior;
 };
 
@@ -199,4 +199,14 @@ void nvd0_disp_intr(struct nouveau_subdev *);
 extern const struct nv50_disp_mthd_chan nve0_disp_mast_mthd_chan;
 extern const struct nv50_disp_mthd_chan nve0_disp_ovly_mthd_chan;
 
+extern struct nvkm_output_dp_impl nv50_pior_dp_impl;
+extern struct nouveau_oclass *nv50_disp_outp_sclass[];
+
+extern struct nvkm_output_dp_impl nv94_sor_dp_impl;
+int nv94_sor_dp_lnk_pwr(struct nvkm_output_dp *, int);
+extern struct nouveau_oclass *nv94_disp_outp_sclass[];
+
+extern struct nvkm_output_dp_impl nvd0_sor_dp_impl;
+extern struct nouveau_oclass *nvd0_disp_outp_sclass[];
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
index 98c5b19bc2b0..1cc62e434683 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
@@ -264,7 +264,6 @@ nv84_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
@@ -277,6 +276,7 @@ nv84_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv50_disp_outp_sclass,
 	.mthd.core = &nv84_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nv84_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
index 6844061c7e04..4f718a9f5aef 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
@@ -77,6 +77,7 @@ nv94_disp_base_omthds[] = {
 	{ SOR_MTHD(NV50_DISP_SOR_PWR)         , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
+	{ SOR_MTHD(NV94_DISP_SOR_DP_PWR)      , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
 	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
@@ -122,13 +123,18 @@ nv94_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
-	priv->sor.dp = &nv94_sor_dp_func;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
 struct nouveau_oclass *
+nv94_disp_outp_sclass[] = {
+	&nv50_pior_dp_impl.base.base,
+	&nv94_sor_dp_impl.base.base,
+	NULL
+};
+
+struct nouveau_oclass *
 nv94_disp_oclass = &(struct nv50_disp_impl) {
 	.base.base.handle = NV_ENGINE(DISP, 0x88),
 	.base.base.ofuncs = &(struct nouveau_ofuncs) {
@@ -137,6 +143,7 @@ nv94_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv94_disp_outp_sclass,
 	.mthd.core = &nv94_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nv84_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
index 88c96241c02a..6237a9a36f70 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
@@ -126,7 +126,6 @@ nva0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
@@ -139,6 +138,7 @@ nva0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv50_disp_outp_sclass,
 	.mthd.core = &nv84_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nva0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c b/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
index 46cb2ce0e82a..019124d4782b 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
@@ -50,6 +50,7 @@ nva3_disp_base_omthds[] = {
 	{ SOR_MTHD(NVA3_DISP_SOR_HDA_ELD)     , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
+	{ SOR_MTHD(NV94_DISP_SOR_DP_PWR)      , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
 	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
@@ -96,9 +97,7 @@ nva3_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nva3_hda_eld;
 	priv->sor.hdmi = nva3_hdmi_ctrl;
-	priv->sor.dp = &nv94_sor_dp_func;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
@@ -111,6 +110,7 @@ nva3_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv94_disp_outp_sclass,
 	.mthd.core = &nv94_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nv84_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
index 876de9ac3793..fa30d8196f35 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
@@ -748,13 +748,13 @@ nvd0_disp_base_scanoutpos(struct nouveau_object *object, u32 mthd,
 }
 
 static void
-nvd0_disp_base_vblank_enable(struct nouveau_event *event, int head)
+nvd0_disp_base_vblank_enable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x6100c0 + (head * 0x800), 0x00000001, 0x00000001);
 }
 
 static void
-nvd0_disp_base_vblank_disable(struct nouveau_event *event, int head)
+nvd0_disp_base_vblank_disable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x6100c0 + (head * 0x800), 0x00000001, 0x00000000);
 }
@@ -887,6 +887,7 @@ nvd0_disp_base_omthds[] = {
 	{ SOR_MTHD(NVA3_DISP_SOR_HDA_ELD)     , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
+	{ SOR_MTHD(NV94_DISP_SOR_DP_PWR)      , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
 	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
@@ -915,19 +916,20 @@ nvd0_disp_sclass[] = {
  * Display engine implementation
  ******************************************************************************/
 
-static u16
-exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
-	    struct dcb_output *dcb, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+static struct nvkm_output *
+exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl,
+	    u32 *data, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 	    struct nvbios_outp *info)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
-	u16 mask, type, data;
+	struct nvkm_output *outp;
+	u16 mask, type;
 
-	if (outp < 4) {
+	if (or < 4) {
 		type = DCB_OUTPUT_ANALOG;
 		mask = 0;
 	} else {
-		outp -= 4;
+		or -= 4;
 		switch (ctrl & 0x00000f00) {
 		case 0x00000000: type = DCB_OUTPUT_LVDS; mask = 1; break;
 		case 0x00000100: type = DCB_OUTPUT_TMDS; mask = 1; break;
@@ -939,101 +941,106 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
 			return 0x0000;
 		}
-		dcb->sorconf.link = mask;
 	}
 
 	mask  = 0x00c0 & (mask << 6);
-	mask |= 0x0001 << outp;
+	mask |= 0x0001 << or;
 	mask |= 0x0100 << head;
 
-	data = dcb_outp_match(bios, type, mask, ver, hdr, dcb);
-	if (!data)
-		return 0x0000;
+	list_for_each_entry(outp, &priv->base.outp, head) {
+		if ((outp->info.hasht & 0xff) == type &&
+		    (outp->info.hashm & mask) == mask) {
+			*data = nvbios_outp_match(bios, outp->info.hasht,
+							outp->info.hashm,
+						  ver, hdr, cnt, len, info);
+			if (!*data)
+				return NULL;
+			return outp;
+		}
+	}
 
-	return nvbios_outp_match(bios, type, mask, ver, hdr, cnt, len, info);
+	return NULL;
 }
 
-static bool
+static struct nvkm_output *
 exec_script(struct nv50_disp_priv *priv, int head, int id)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info;
-	struct dcb_output dcb;
 	u8  ver, hdr, cnt, len;
-	u32 ctrl = 0x00000000;
-	u16 data;
-	int outp;
+	u32 data, ctrl = 0;
+	int or;
 
-	for (outp = 0; !(ctrl & (1 << head)) && outp < 8; outp++) {
-		ctrl = nv_rd32(priv, 0x640180 + (outp * 0x20));
+	for (or = 0; !(ctrl & (1 << head)) && or < 8; or++) {
+		ctrl = nv_rd32(priv, 0x640180 + (or * 0x20));
 		if (ctrl & (1 << head))
 			break;
 	}
 
-	if (outp == 8)
-		return false;
+	if (or == 8)
+		return NULL;
 
-	data = exec_lookup(priv, head, outp, ctrl, &dcb, &ver, &hdr, &cnt, &len, &info);
-	if (data) {
+	outp = exec_lookup(priv, head, or, ctrl, &data, &ver, &hdr, &cnt, &len, &info);
+	if (outp) {
 		struct nvbios_init init = {
 			.subdev = nv_subdev(priv),
 			.bios = bios,
 			.offset = info.script[id],
-			.outp = &dcb,
+			.outp = &outp->info,
 			.crtc = head,
 			.execute = 1,
 		};
 
-		return nvbios_exec(&init) == 0;
+		nvbios_exec(&init);
 	}
 
-	return false;
+	return outp;
 }
 
-static u32
-exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
-	    u32 pclk, struct dcb_output *dcb)
+static struct nvkm_output *
+exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk, u32 *conf)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info1;
 	struct nvbios_ocfg info2;
 	u8  ver, hdr, cnt, len;
-	u32 ctrl = 0x00000000;
-	u32 data, conf = ~0;
-	int outp;
+	u32 data, ctrl = 0;
+	int or;
 
-	for (outp = 0; !(ctrl & (1 << head)) && outp < 8; outp++) {
-		ctrl = nv_rd32(priv, 0x660180 + (outp * 0x20));
+	for (or = 0; !(ctrl & (1 << head)) && or < 8; or++) {
+		ctrl = nv_rd32(priv, 0x660180 + (or * 0x20));
 		if (ctrl & (1 << head))
 			break;
 	}
 
-	if (outp == 8)
-		return conf;
+	if (or == 8)
+		return NULL;
 
-	data = exec_lookup(priv, head, outp, ctrl, dcb, &ver, &hdr, &cnt, &len, &info1);
-	if (data == 0x0000)
-		return conf;
+	outp = exec_lookup(priv, head, or, ctrl, &data, &ver, &hdr, &cnt, &len, &info1);
+	if (!outp)
+		return NULL;
 
-	switch (dcb->type) {
+	switch (outp->info.type) {
 	case DCB_OUTPUT_TMDS:
-		conf = (ctrl & 0x00000f00) >> 8;
+		*conf = (ctrl & 0x00000f00) >> 8;
 		if (pclk >= 165000)
-			conf |= 0x0100;
+			*conf |= 0x0100;
 		break;
 	case DCB_OUTPUT_LVDS:
-		conf = priv->sor.lvdsconf;
+		*conf = priv->sor.lvdsconf;
 		break;
 	case DCB_OUTPUT_DP:
-		conf = (ctrl & 0x00000f00) >> 8;
+		*conf = (ctrl & 0x00000f00) >> 8;
 		break;
 	case DCB_OUTPUT_ANALOG:
 	default:
-		conf = 0x00ff;
+		*conf = 0x00ff;
 		break;
 	}
 
-	data = nvbios_ocfg_match(bios, data, conf, &ver, &hdr, &cnt, &len, &info2);
+	data = nvbios_ocfg_match(bios, data, *conf, &ver, &hdr, &cnt, &len, &info2);
 	if (data && id < 0xff) {
 		data = nvbios_oclk_match(bios, info2.clkcmp[id], pclk);
 		if (data) {
@@ -1041,7 +1048,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
 				.subdev = nv_subdev(priv),
 				.bios = bios,
 				.offset = data,
-				.outp = dcb,
+				.outp = &outp->info,
 				.crtc = head,
 				.execute = 1,
 			};
@@ -1050,7 +1057,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
 		}
 	}
 
-	return conf;
+	return outp;
 }
 
 static void
@@ -1062,7 +1069,23 @@ nvd0_disp_intr_unk1_0(struct nv50_disp_priv *priv, int head)
 static void
 nvd0_disp_intr_unk2_0(struct nv50_disp_priv *priv, int head)
 {
-	exec_script(priv, head, 2);
+	struct nvkm_output *outp = exec_script(priv, head, 2);
+
+	/* see note in nv50_disp_intr_unk20_0() */
+	if (outp && outp->info.type == DCB_OUTPUT_DP) {
+		struct nvkm_output_dp *outpdp = (void *)outp;
+		struct nvbios_init init = {
+			.subdev = nv_subdev(priv),
+			.bios = nouveau_bios(priv),
+			.outp = &outp->info,
+			.crtc = head,
+			.offset = outpdp->info.script[4],
+			.execute = 1,
+		};
+
+		nvbios_exec(&init);
+		atomic_set(&outpdp->lt.done, 0);
+	}
 }
 
 static void
@@ -1124,49 +1147,52 @@ nvd0_disp_intr_unk2_2_tu(struct nv50_disp_priv *priv, int head,
 static void
 nvd0_disp_intr_unk2_2(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
+	struct nvkm_output *outp;
 	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
-	u32 conf = exec_clkcmp(priv, head, 0xff, pclk, &outp);
-	if (conf != ~0) {
-		u32 addr, data;
-
-		if (outp.type == DCB_OUTPUT_DP) {
-			u32 sync = nv_rd32(priv, 0x660404 + (head * 0x300));
-			switch ((sync & 0x000003c0) >> 6) {
-			case 6: pclk = pclk * 30 / 8; break;
-			case 5: pclk = pclk * 24 / 8; break;
-			case 2:
-			default:
-				pclk = pclk * 18 / 8;
-				break;
-			}
-
-			nouveau_dp_train(&priv->base, priv->sor.dp,
-					 &outp, head, pclk);
+	u32 conf, addr, data;
+
+	outp = exec_clkcmp(priv, head, 0xff, pclk, &conf);
+	if (!outp)
+		return;
+
+	/* see note in nv50_disp_intr_unk20_2() */
+	if (outp->info.type == DCB_OUTPUT_DP) {
+		u32 sync = nv_rd32(priv, 0x660404 + (head * 0x300));
+		switch ((sync & 0x000003c0) >> 6) {
+		case 6: pclk = pclk * 30; break;
+		case 5: pclk = pclk * 24; break;
+		case 2:
+		default:
+			pclk = pclk * 18;
+			break;
 		}
 
-		exec_clkcmp(priv, head, 0, pclk, &outp);
+		if (nvkm_output_dp_train(outp, pclk, true))
+			ERR("link not trained before attach\n");
+	}
 
-		if (outp.type == DCB_OUTPUT_ANALOG) {
-			addr = 0x612280 + (ffs(outp.or) - 1) * 0x800;
-			data = 0x00000000;
-		} else {
-			if (outp.type == DCB_OUTPUT_DP)
-				nvd0_disp_intr_unk2_2_tu(priv, head, &outp);
-			addr = 0x612300 + (ffs(outp.or) - 1) * 0x800;
-			data = (conf & 0x0100) ? 0x00000101 : 0x00000000;
-		}
+	exec_clkcmp(priv, head, 0, pclk, &conf);
 
-		nv_mask(priv, addr, 0x00000707, data);
+	if (outp->info.type == DCB_OUTPUT_ANALOG) {
+		addr = 0x612280 + (ffs(outp->info.or) - 1) * 0x800;
+		data = 0x00000000;
+	} else {
+		if (outp->info.type == DCB_OUTPUT_DP)
+			nvd0_disp_intr_unk2_2_tu(priv, head, &outp->info);
+		addr = 0x612300 + (ffs(outp->info.or) - 1) * 0x800;
+		data = (conf & 0x0100) ? 0x00000101 : 0x00000000;
 	}
+
+	nv_mask(priv, addr, 0x00000707, data);
 }
 
 static void
 nvd0_disp_intr_unk4_0(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
 	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
-	exec_clkcmp(priv, head, 1, pclk, &outp);
+	u32 conf;
+
+	exec_clkcmp(priv, head, 1, pclk, &conf);
 }
 
 void
@@ -1240,7 +1266,7 @@ nvd0_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 		 chid, (mthd & 0x0000ffc), data, mthd, unkn);
 
 	if (chid == 0) {
-		switch (mthd) {
+		switch (mthd & 0xffc) {
 		case 0x0080:
 			nv50_disp_mthd_chan(priv, NV_DBG_ERROR, chid - 0,
 					    impl->mthd.core);
@@ -1250,7 +1276,7 @@ nvd0_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 		}
 	} else
 	if (chid <= 4) {
-		switch (mthd) {
+		switch (mthd & 0xffc) {
 		case 0x0080:
 			nv50_disp_mthd_chan(priv, NV_DBG_ERROR, chid - 1,
 					    impl->mthd.base);
@@ -1260,7 +1286,7 @@ nvd0_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 		}
 	} else
 	if (chid <= 8) {
-		switch (mthd) {
+		switch (mthd & 0xffc) {
 		case 0x0080:
 			nv50_disp_mthd_chan(priv, NV_DBG_ERROR, chid - 5,
 					    impl->mthd.ovly);
@@ -1317,7 +1343,7 @@ nvd0_disp_intr(struct nouveau_subdev *subdev)
 		if (mask & intr) {
 			u32 stat = nv_rd32(priv, 0x6100bc + (i * 0x800));
 			if (stat & 0x00000001)
-				nouveau_event_trigger(priv->base.vblank, i);
+				nouveau_event_trigger(priv->base.vblank, 1, i);
 			nv_mask(priv, 0x6100bc + (i * 0x800), 0, 0);
 			nv_rd32(priv, 0x6100c0 + (i * 0x800));
 		}
@@ -1352,11 +1378,16 @@ nvd0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
 struct nouveau_oclass *
+nvd0_disp_outp_sclass[] = {
+	&nvd0_sor_dp_impl.base.base,
+	NULL
+};
+
+struct nouveau_oclass *
 nvd0_disp_oclass = &(struct nv50_disp_impl) {
 	.base.base.handle = NV_ENGINE(DISP, 0x90),
 	.base.base.ofuncs = &(struct nouveau_ofuncs) {
@@ -1365,6 +1396,7 @@ nvd0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nvd0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nvd0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
index 44e0b8f34c1a..11328e3f5df1 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
@@ -246,7 +246,6 @@ nve0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
@@ -259,6 +258,7 @@ nve0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nve0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nve0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c
index 482585d375fa..104388081d73 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c
@@ -81,7 +81,6 @@ nvf0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
@@ -94,6 +93,7 @@ nvf0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nve0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nve0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outp.c b/drivers/gpu/drm/nouveau/core/engine/disp/outp.c
new file mode 100644
index 000000000000..ad9ba7ccec7f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outp.c
@@ -0,0 +1,137 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/i2c.h>
+#include <subdev/bios.h>
+#include <subdev/bios/conn.h>
+
+#include "outp.h"
+
+int
+_nvkm_output_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_output *outp = (void *)object;
+	nv_ofuncs(outp->conn)->fini(nv_object(outp->conn), suspend);
+	return nouveau_object_fini(&outp->base, suspend);
+}
+
+int
+_nvkm_output_init(struct nouveau_object *object)
+{
+	struct nvkm_output *outp = (void *)object;
+	int ret = nouveau_object_init(&outp->base);
+	if (ret == 0)
+		nv_ofuncs(outp->conn)->init(nv_object(outp->conn));
+	return 0;
+}
+
+void
+_nvkm_output_dtor(struct nouveau_object *object)
+{
+	struct nvkm_output *outp = (void *)object;
+	list_del(&outp->head);
+	nouveau_object_ref(NULL, (void *)&outp->conn);
+	nouveau_object_destroy(&outp->base);
+}
+
+int
+nvkm_output_create_(struct nouveau_object *parent,
+		    struct nouveau_object *engine,
+		    struct nouveau_oclass *oclass,
+		    struct dcb_output *dcbE, int index,
+		    int length, void **pobject)
+{
+	struct nouveau_bios *bios = nouveau_bios(engine);
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nouveau_disp *disp = (void *)engine;
+	struct nvbios_connE connE;
+	struct nvkm_output *outp;
+	u8  ver, hdr;
+	u32 data;
+	int ret;
+
+	ret = nouveau_object_create_(parent, engine, oclass, 0, length, pobject);
+	outp = *pobject;
+	if (ret)
+		return ret;
+
+	outp->info = *dcbE;
+	outp->index = index;
+
+	DBG("type %02x loc %d or %d link %d con %x edid %x bus %d head %x\n",
+	    dcbE->type, dcbE->location, dcbE->or, dcbE->type >= 2 ?
+	    dcbE->sorconf.link : 0, dcbE->connector, dcbE->i2c_index,
+	    dcbE->bus, dcbE->heads);
+
+	outp->port = i2c->find(i2c, outp->info.i2c_index);
+	outp->edid = outp->port;
+
+	data = nvbios_connEp(bios, outp->info.connector, &ver, &hdr, &connE);
+	if (!data) {
+		DBG("vbios connector data not found\n");
+		memset(&connE, 0x00, sizeof(connE));
+		connE.type = DCB_CONNECTOR_NONE;
+	}
+
+	ret = nouveau_object_ctor(parent, engine, nvkm_connector_oclass,
+				 &connE, outp->info.connector,
+				 (struct nouveau_object **)&outp->conn);
+	if (ret < 0) {
+		ERR("error %d creating connector, disabling\n", ret);
+		return ret;
+	}
+
+	list_add_tail(&outp->head, &disp->outp);
+	return 0;
+}
+
+int
+_nvkm_output_ctor(struct nouveau_object *parent,
+		  struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *dcbE, u32 index,
+		  struct nouveau_object **pobject)
+{
+	struct nvkm_output *outp;
+	int ret;
+
+	ret = nvkm_output_create(parent, engine, oclass, dcbE, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass *
+nvkm_output_oclass = &(struct nvkm_output_impl) {
+	.base = {
+		.handle = 0,
+		.ofuncs = &(struct nouveau_ofuncs) {
+			.ctor = _nvkm_output_ctor,
+			.dtor = _nvkm_output_dtor,
+			.init = _nvkm_output_init,
+			.fini = _nvkm_output_fini,
+		},
+	},
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outp.h b/drivers/gpu/drm/nouveau/core/engine/disp/outp.h
new file mode 100644
index 000000000000..bc76fbf85710
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outp.h
@@ -0,0 +1,59 @@
+#ifndef __NVKM_DISP_OUTP_H__
+#define __NVKM_DISP_OUTP_H__
+
+#include "priv.h"
+
+struct nvkm_output {
+	struct nouveau_object base;
+	struct list_head head;
+
+	struct dcb_output info;
+	int index;
+
+	struct nouveau_i2c_port *port;
+	struct nouveau_i2c_port *edid;
+
+	struct nvkm_connector *conn;
+};
+
+#define nvkm_output_create(p,e,c,b,i,d)                                        \
+	nvkm_output_create_((p), (e), (c), (b), (i), sizeof(**d), (void **)d)
+#define nvkm_output_destroy(d) ({                                              \
+	struct nvkm_output *_outp = (d);                                       \
+	_nvkm_output_dtor(nv_object(_outp));                                   \
+})
+#define nvkm_output_init(d) ({                                                 \
+	struct nvkm_output *_outp = (d);                                       \
+	_nvkm_output_init(nv_object(_outp));                                   \
+})
+#define nvkm_output_fini(d,s) ({                                               \
+	struct nvkm_output *_outp = (d);                                       \
+	_nvkm_output_fini(nv_object(_outp), (s));                              \
+})
+
+int nvkm_output_create_(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, struct dcb_output *,
+			int, int, void **);
+
+int  _nvkm_output_ctor(struct nouveau_object *, struct nouveau_object *,
+		       struct nouveau_oclass *, void *, u32,
+		       struct nouveau_object **);
+void _nvkm_output_dtor(struct nouveau_object *);
+int  _nvkm_output_init(struct nouveau_object *);
+int  _nvkm_output_fini(struct nouveau_object *, bool);
+
+struct nvkm_output_impl {
+	struct nouveau_oclass base;
+};
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nvkm_output *_outp = (void *)outp;                              \
+	nv_##l(nv_object(outp)->engine, "%02x:%04x:%04x: "f, _outp->index,     \
+	       _outp->info.hasht, _outp->info.hashm, ##a);                     \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.c b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.c
new file mode 100644
index 000000000000..eb2d7789555d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.c
@@ -0,0 +1,278 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/i2c.h>
+
+#include "outpdp.h"
+#include "conn.h"
+#include "dport.h"
+
+int
+nvkm_output_dp_train(struct nvkm_output *base, u32 datarate, bool wait)
+{
+	struct nvkm_output_dp *outp = (void *)base;
+	bool retrain = true;
+	u8 link[2], stat[3];
+	u32 linkrate;
+	int ret, i;
+
+	/* check that the link is trained at a high enough rate */
+	ret = nv_rdaux(outp->base.edid, DPCD_LC00_LINK_BW_SET, link, 2);
+	if (ret) {
+		DBG("failed to read link config, assuming no sink\n");
+		goto done;
+	}
+
+	linkrate = link[0] * 27000 * (link[1] & DPCD_LC01_LANE_COUNT_SET);
+	linkrate = (linkrate * 8) / 10; /* 8B/10B coding overhead */
+	datarate = (datarate + 9) / 10; /* -> decakilobits */
+	if (linkrate < datarate) {
+		DBG("link not trained at sufficient rate\n");
+		goto done;
+	}
+
+	/* check that link is still trained */
+	ret = nv_rdaux(outp->base.edid, DPCD_LS02, stat, 3);
+	if (ret) {
+		DBG("failed to read link status, assuming no sink\n");
+		goto done;
+	}
+
+	if (stat[2] & DPCD_LS04_INTERLANE_ALIGN_DONE) {
+		for (i = 0; i < (link[1] & DPCD_LC01_LANE_COUNT_SET); i++) {
+			u8 lane = (stat[i >> 1] >> ((i & 1) * 4)) & 0x0f;
+			if (!(lane & DPCD_LS02_LANE0_CR_DONE) ||
+			    !(lane & DPCD_LS02_LANE0_CHANNEL_EQ_DONE) ||
+			    !(lane & DPCD_LS02_LANE0_SYMBOL_LOCKED)) {
+				DBG("lane %d not equalised\n", lane);
+				goto done;
+			}
+		}
+		retrain = false;
+	} else {
+		DBG("no inter-lane alignment\n");
+	}
+
+done:
+	if (retrain || !atomic_read(&outp->lt.done)) {
+		/* no sink, but still need to configure source */
+		if (outp->dpcd[DPCD_RC00_DPCD_REV] == 0x00) {
+			outp->dpcd[DPCD_RC01_MAX_LINK_RATE] =
+				outp->base.info.dpconf.link_bw;
+			outp->dpcd[DPCD_RC02] =
+				outp->base.info.dpconf.link_nr;
+		}
+		atomic_set(&outp->lt.done, 0);
+		schedule_work(&outp->lt.work);
+	} else {
+		nouveau_event_get(outp->irq);
+	}
+
+	if (wait) {
+		if (!wait_event_timeout(outp->lt.wait,
+					atomic_read(&outp->lt.done),
+					msecs_to_jiffies(2000)))
+			ret = -ETIMEDOUT;
+	}
+
+	return ret;
+}
+
+static void
+nvkm_output_dp_enable(struct nvkm_output_dp *outp, bool present)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
+	if (present) {
+		if (!outp->present) {
+			nouveau_i2c(port)->acquire_pad(port, 0);
+			DBG("aux power -> always\n");
+			outp->present = true;
+		}
+		nvkm_output_dp_train(&outp->base, 0, true);
+	} else {
+		if (outp->present) {
+			nouveau_i2c(port)->release_pad(port);
+			DBG("aux power -> demand\n");
+			outp->present = false;
+		}
+		atomic_set(&outp->lt.done, 0);
+	}
+}
+
+static void
+nvkm_output_dp_detect(struct nvkm_output_dp *outp)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
+	int ret = nouveau_i2c(port)->acquire_pad(port, 0);
+	if (ret == 0) {
+		ret = nv_rdaux(outp->base.edid, DPCD_RC00_DPCD_REV,
+			       outp->dpcd, sizeof(outp->dpcd));
+		nvkm_output_dp_enable(outp, ret == 0);
+		nouveau_i2c(port)->release_pad(port);
+	}
+}
+
+static void
+nvkm_output_dp_service_work(struct work_struct *work)
+{
+	struct nvkm_output_dp *outp = container_of(work, typeof(*outp), work);
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	int type = atomic_xchg(&outp->pending, 0);
+	u32 send = 0;
+
+	if (type & (NVKM_I2C_PLUG | NVKM_I2C_UNPLUG)) {
+		nvkm_output_dp_detect(outp);
+		if (type & NVKM_I2C_UNPLUG)
+			send |= NVKM_HPD_UNPLUG;
+		if (type & NVKM_I2C_PLUG)
+			send |= NVKM_HPD_PLUG;
+		nouveau_event_get(outp->base.conn->hpd.event);
+	}
+
+	if (type & NVKM_I2C_IRQ) {
+		nvkm_output_dp_train(&outp->base, 0, true);
+		send |= NVKM_HPD_IRQ;
+	}
+
+	nouveau_event_trigger(disp->hpd, send, outp->base.info.connector);
+}
+
+static int
+nvkm_output_dp_service(void *data, u32 type, int index)
+{
+	struct nvkm_output_dp *outp = data;
+	DBG("HPD: %d\n", type);
+	atomic_or(type, &outp->pending);
+	schedule_work(&outp->work);
+	return NVKM_EVENT_DROP;
+}
+
+int
+_nvkm_output_dp_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_output_dp *outp = (void *)object;
+	nouveau_event_put(outp->irq);
+	nvkm_output_dp_enable(outp, false);
+	return nvkm_output_fini(&outp->base, suspend);
+}
+
+int
+_nvkm_output_dp_init(struct nouveau_object *object)
+{
+	struct nvkm_output_dp *outp = (void *)object;
+	nvkm_output_dp_detect(outp);
+	return nvkm_output_init(&outp->base);
+}
+
+void
+_nvkm_output_dp_dtor(struct nouveau_object *object)
+{
+	struct nvkm_output_dp *outp = (void *)object;
+	nouveau_event_ref(NULL, &outp->irq);
+	nvkm_output_destroy(&outp->base);
+}
+
+int
+nvkm_output_dp_create_(struct nouveau_object *parent,
+		       struct nouveau_object *engine,
+		       struct nouveau_oclass *oclass,
+		       struct dcb_output *info, int index,
+		       int length, void **pobject)
+{
+	struct nouveau_bios *bios = nouveau_bios(parent);
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nvkm_output_dp *outp;
+	u8  hdr, cnt, len;
+	u32 data;
+	int ret;
+
+	ret = nvkm_output_create_(parent, engine, oclass, info, index,
+				  length, pobject);
+	outp = *pobject;
+	if (ret)
+		return ret;
+
+	nouveau_event_ref(NULL, &outp->base.conn->hpd.event);
+
+	/* access to the aux channel is not optional... */
+	if (!outp->base.edid) {
+		ERR("aux channel not found\n");
+		return -ENODEV;
+	}
+
+	/* nor is the bios data for this output... */
+	data = nvbios_dpout_match(bios, outp->base.info.hasht,
+				  outp->base.info.hashm, &outp->version,
+				  &hdr, &cnt, &len, &outp->info);
+	if (!data) {
+		ERR("no bios dp data\n");
+		return -ENODEV;
+	}
+
+	DBG("bios dp %02x %02x %02x %02x\n", outp->version, hdr, cnt, len);
+
+	/* link training */
+	INIT_WORK(&outp->lt.work, nouveau_dp_train);
+	init_waitqueue_head(&outp->lt.wait);
+	atomic_set(&outp->lt.done, 0);
+
+	/* link maintenance */
+	ret = nouveau_event_new(i2c->ntfy, NVKM_I2C_IRQ, outp->base.edid->index,
+				nvkm_output_dp_service, outp, &outp->irq);
+	if (ret) {
+		ERR("error monitoring aux irq event: %d\n", ret);
+		return ret;
+	}
+
+	INIT_WORK(&outp->work, nvkm_output_dp_service_work);
+
+	/* hotplug detect, replaces gpio-based mechanism with aux events */
+	ret = nouveau_event_new(i2c->ntfy, NVKM_I2C_PLUG | NVKM_I2C_UNPLUG,
+				outp->base.edid->index,
+				nvkm_output_dp_service, outp,
+			       &outp->base.conn->hpd.event);
+	if (ret) {
+		ERR("error monitoring aux hpd events: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+int
+_nvkm_output_dp_ctor(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, void *info, u32 index,
+		     struct nouveau_object **pobject)
+{
+	struct nvkm_output_dp *outp;
+	int ret;
+
+	ret = nvkm_output_dp_create(parent, engine, oclass, info, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.h b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.h
new file mode 100644
index 000000000000..ff33ba12cb67
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.h
@@ -0,0 +1,65 @@
+#ifndef __NVKM_DISP_OUTP_DP_H__
+#define __NVKM_DISP_OUTP_DP_H__
+
+#include <subdev/bios.h>
+#include <subdev/bios/dp.h>
+
+#include "outp.h"
+
+struct nvkm_output_dp {
+	struct nvkm_output base;
+
+	struct nvbios_dpout info;
+	u8 version;
+
+	struct nouveau_eventh *irq;
+	struct nouveau_eventh *hpd;
+	struct work_struct work;
+	atomic_t pending;
+	bool present;
+	u8 dpcd[16];
+
+	struct {
+		struct work_struct work;
+		wait_queue_head_t wait;
+		atomic_t done;
+	} lt;
+};
+
+#define nvkm_output_dp_create(p,e,c,b,i,d)                                     \
+	nvkm_output_dp_create_((p), (e), (c), (b), (i), sizeof(**d), (void **)d)
+#define nvkm_output_dp_destroy(d) ({                                           \
+	struct nvkm_output_dp *_outp = (d);                                    \
+	_nvkm_output_dp_dtor(nv_object(_outp));                                \
+})
+#define nvkm_output_dp_init(d) ({                                              \
+	struct nvkm_output_dp *_outp = (d);                                    \
+	_nvkm_output_dp_init(nv_object(_outp));                                \
+})
+#define nvkm_output_dp_fini(d,s) ({                                            \
+	struct nvkm_output_dp *_outp = (d);                                    \
+	_nvkm_output_dp_fini(nv_object(_outp), (s));                           \
+})
+
+int nvkm_output_dp_create_(struct nouveau_object *, struct nouveau_object *,
+			   struct nouveau_oclass *, struct dcb_output *,
+			   int, int, void **);
+
+int  _nvkm_output_dp_ctor(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, void *, u32,
+			  struct nouveau_object **);
+void _nvkm_output_dp_dtor(struct nouveau_object *);
+int  _nvkm_output_dp_init(struct nouveau_object *);
+int  _nvkm_output_dp_fini(struct nouveau_object *, bool);
+
+struct nvkm_output_dp_impl {
+	struct nvkm_output_impl base;
+	int (*pattern)(struct nvkm_output_dp *, int);
+	int (*lnk_pwr)(struct nvkm_output_dp *, int nr);
+	int (*lnk_ctl)(struct nvkm_output_dp *, int nr, int bw, bool ef);
+	int (*drv_ctl)(struct nvkm_output_dp *, int ln, int vs, int pe, int pc);
+};
+
+int nvkm_output_dp_train(struct nvkm_output *, u32 rate, bool wait);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
index 2c8ce351b52d..fe0f256f11bf 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
@@ -33,68 +33,107 @@
 #include "nv50.h"
 
 /******************************************************************************
- * DisplayPort
+ * TMDS
  *****************************************************************************/
-static struct nouveau_i2c_port *
-nv50_pior_dp_find(struct nouveau_disp *disp, struct dcb_output *outp)
+
+static int
+nv50_pior_tmds_ctor(struct nouveau_object *parent,
+		    struct nouveau_object *engine,
+		    struct nouveau_oclass *oclass, void *info, u32 index,
+		    struct nouveau_object **pobject)
 {
-	struct nouveau_i2c *i2c = nouveau_i2c(disp);
-	return i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(outp->extdev));
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nvkm_output *outp;
+	int ret;
+
+	ret = nvkm_output_create(parent, engine, oclass, info, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	outp->edid = i2c->find_type(i2c, NV_I2C_TYPE_EXTDDC(outp->info.extdev));
+	return 0;
 }
 
+struct nvkm_output_impl
+nv50_pior_tmds_impl = {
+	.base.handle = DCB_OUTPUT_TMDS | 0x0100,
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_pior_tmds_ctor,
+		.dtor = _nvkm_output_dtor,
+		.init = _nvkm_output_init,
+		.fini = _nvkm_output_fini,
+	},
+};
+
+/******************************************************************************
+ * DisplayPort
+ *****************************************************************************/
+
 static int
-nv50_pior_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
-		     int head, int pattern)
+nv50_pior_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
-	struct nouveau_i2c_port *port;
-	int ret = -EINVAL;
-
-	port = nv50_pior_dp_find(disp, outp);
-	if (port) {
-		if (port->func->pattern)
-			ret = port->func->pattern(port, pattern);
-		else
-			ret = 0;
-	}
-
-	return ret;
+	struct nouveau_i2c_port *port = outp->base.edid;
+	if (port && port->func->pattern)
+		return port->func->pattern(port, pattern);
+	return port ? 0 : -ENODEV;
 }
 
 static int
-nv50_pior_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		     int head, int lane_nr, int link_bw, bool enh)
+nv50_pior_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
 {
-	struct nouveau_i2c_port *port;
-	int ret = -EINVAL;
+	return 0;
+}
 
-	port = nv50_pior_dp_find(disp, outp);
+static int
+nv50_pior_dp_lnk_ctl(struct nvkm_output_dp *outp, int nr, int bw, bool ef)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
 	if (port && port->func->lnk_ctl)
-		ret = port->func->lnk_ctl(port, lane_nr, link_bw, enh);
+		return port->func->lnk_ctl(port, nr, bw, ef);
+	return port ? 0 : -ENODEV;
+}
 
-	return ret;
+static int
+nv50_pior_dp_drv_ctl(struct nvkm_output_dp *outp, int ln, int vs, int pe, int pc)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
+	if (port && port->func->drv_ctl)
+		return port->func->drv_ctl(port, ln, vs, pe);
+	return port ? 0 : -ENODEV;
 }
 
 static int
-nv50_pior_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		     int head, int lane, int vsw, int pre)
+nv50_pior_dp_ctor(struct nouveau_object *parent,
+		  struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *info, u32 index,
+		  struct nouveau_object **pobject)
 {
-	struct nouveau_i2c_port *port;
-	int ret = -EINVAL;
-
-	port = nv50_pior_dp_find(disp, outp);
-	if (port) {
-		if (port->func->drv_ctl)
-			ret = port->func->drv_ctl(port, lane, vsw, pre);
-		else
-			ret = 0;
-	}
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nvkm_output_dp *outp;
+	int ret;
 
-	return ret;
+	ret = nvkm_output_dp_create(parent, engine, oclass, info, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	outp->base.edid = i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(
+					 outp->base.info.extdev));
+	return 0;
 }
 
-const struct nouveau_dp_func
-nv50_pior_dp_func = {
+struct nvkm_output_dp_impl
+nv50_pior_dp_impl = {
+	.base.base.handle = DCB_OUTPUT_DP | 0x0010,
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_pior_dp_ctor,
+		.dtor = _nvkm_output_dp_dtor,
+		.init = _nvkm_output_dp_init,
+		.fini = _nvkm_output_dp_fini,
+	},
 	.pattern = nv50_pior_dp_pattern,
+	.lnk_pwr = nv50_pior_dp_lnk_pwr,
 	.lnk_ctl = nv50_pior_dp_lnk_ctl,
 	.drv_ctl = nv50_pior_dp_drv_ctl,
 };
@@ -102,6 +141,7 @@ nv50_pior_dp_func = {
 /******************************************************************************
  * General PIOR handling
  *****************************************************************************/
+
 int
 nv50_pior_power(struct nv50_disp_priv *priv, int or, u32 data)
 {
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/priv.h b/drivers/gpu/drm/nouveau/core/engine/disp/priv.h
index cc3c7a4ca747..26e9a42569c7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/priv.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/priv.h
@@ -1,10 +1,42 @@
 #ifndef __NVKM_DISP_PRIV_H__
 #define __NVKM_DISP_PRIV_H__
 
+#include <subdev/bios.h>
+#include <subdev/bios/dcb.h>
+#include <subdev/bios/conn.h>
+
 #include <engine/disp.h>
 
 struct nouveau_disp_impl {
 	struct nouveau_oclass base;
+	struct nouveau_oclass **outp;
+	struct nouveau_oclass **conn;
 };
 
+#define nouveau_disp_create(p,e,c,h,i,x,d)                                     \
+	nouveau_disp_create_((p), (e), (c), (h), (i), (x),                     \
+			     sizeof(**d), (void **)d)
+#define nouveau_disp_destroy(d) ({                                             \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_dtor(nv_object(disp));                                   \
+})
+#define nouveau_disp_init(d) ({                                                \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_init(nv_object(disp));                                   \
+})
+#define nouveau_disp_fini(d,s) ({                                              \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_fini(nv_object(disp), (s));                              \
+})
+
+int  nouveau_disp_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int heads,
+			  const char *, const char *, int, void **);
+void _nouveau_disp_dtor(struct nouveau_object *);
+int  _nouveau_disp_init(struct nouveau_object *);
+int  _nouveau_disp_fini(struct nouveau_object *, bool);
+
+extern struct nouveau_oclass *nvkm_output_oclass;
+extern struct nouveau_oclass *nvkm_connector_oclass;
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
index 526b75242899..7a1ebdfa9e1b 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
@@ -47,8 +47,12 @@ int
 nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 {
 	struct nv50_disp_priv *priv = (void *)object->engine;
+	const u8  type = (mthd & NV50_DISP_SOR_MTHD_TYPE) >> 12;
 	const u8  head = (mthd & NV50_DISP_SOR_MTHD_HEAD) >> 3;
+	const u8  link = (mthd & NV50_DISP_SOR_MTHD_LINK) >> 2;
 	const u8    or = (mthd & NV50_DISP_SOR_MTHD_OR);
+	const u16 mask = (0x0100 << head) | (0x0040 << link) | (0x0001 << or);
+	struct nvkm_output *outp = NULL, *temp;
 	u32 data;
 	int ret = -EINVAL;
 
@@ -56,6 +60,13 @@ nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 		return -EINVAL;
 	data = *(u32 *)args;
 
+	list_for_each_entry(temp, &priv->base.outp, head) {
+		if ((temp->info.hasht & 0xff) == type &&
+		    (temp->info.hashm & mask) == mask) {
+			outp = temp;
+			break;
+		}
+	}
 
 	switch (mthd & ~0x3f) {
 	case NV50_DISP_SOR_PWR:
@@ -71,6 +82,24 @@ nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 		priv->sor.lvdsconf = data & NV50_DISP_SOR_LVDS_SCRIPT_ID;
 		ret = 0;
 		break;
+	case NV94_DISP_SOR_DP_PWR:
+		if (outp) {
+			struct nvkm_output_dp *outpdp = (void *)outp;
+			switch (data) {
+			case NV94_DISP_SOR_DP_PWR_STATE_OFF:
+				nouveau_event_put(outpdp->irq);
+				((struct nvkm_output_dp_impl *)nv_oclass(outp))
+					->lnk_pwr(outpdp, 0);
+				atomic_set(&outpdp->lt.done, 0);
+				break;
+			case NV94_DISP_SOR_DP_PWR_STATE_ON:
+				nvkm_output_dp_train(&outpdp->base, 0, true);
+				break;
+			default:
+				return -EINVAL;
+			}
+		}
+		break;
 	default:
 		BUG_ON(1);
 	}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
index eea3ef59693d..05487cda84a8 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
@@ -29,19 +29,21 @@
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/dp.h>
 #include <subdev/bios/init.h>
+#include <subdev/timer.h>
 
 #include "nv50.h"
+#include "outpdp.h"
 
 static inline u32
-nv94_sor_soff(struct dcb_output *outp)
+nv94_sor_soff(struct nvkm_output_dp *outp)
 {
-	return (ffs(outp->or) - 1) * 0x800;
+	return (ffs(outp->base.info.or) - 1) * 0x800;
 }
 
 static inline u32
-nv94_sor_loff(struct dcb_output *outp)
+nv94_sor_loff(struct nvkm_output_dp *outp)
 {
-	return nv94_sor_soff(outp) + !(outp->sorconf.link & 1) * 0x80;
+	return nv94_sor_soff(outp) + !(outp->base.info.sorconf.link & 1) * 0x80;
 }
 
 static inline u32
@@ -55,77 +57,96 @@ nv94_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
 }
 
 static int
-nv94_sor_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int pattern)
+nv94_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 loff = nv94_sor_loff(outp);
 	nv_mask(priv, 0x61c10c + loff, 0x0f000000, pattern << 24);
 	return 0;
 }
 
+int
+nv94_sor_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
+{
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
+	const u32 soff = nv94_sor_soff(outp);
+	const u32 loff = nv94_sor_loff(outp);
+	u32 mask = 0, i;
+
+	for (i = 0; i < nr; i++)
+		mask |= 1 << (nv94_sor_dp_lane_map(priv, i) >> 3);
+
+	nv_mask(priv, 0x61c130 + loff, 0x0000000f, mask);
+	nv_mask(priv, 0x61c034 + soff, 0x80000000, 0x80000000);
+	nv_wait(priv, 0x61c034 + soff, 0x80000000, 0x00000000);
+	return 0;
+}
+
 static int
-nv94_sor_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int link_nr, int link_bw, bool enh_frame)
+nv94_sor_dp_lnk_ctl(struct nvkm_output_dp *outp, int nr, int bw, bool ef)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 soff = nv94_sor_soff(outp);
 	const u32 loff = nv94_sor_loff(outp);
 	u32 dpctrl = 0x00000000;
 	u32 clksor = 0x00000000;
-	u32 lane = 0;
-	int i;
 
-	dpctrl |= ((1 << link_nr) - 1) << 16;
-	if (enh_frame)
+	dpctrl |= ((1 << nr) - 1) << 16;
+	if (ef)
 		dpctrl |= 0x00004000;
-	if (link_bw > 0x06)
+	if (bw > 0x06)
 		clksor |= 0x00040000;
 
-	for (i = 0; i < link_nr; i++)
-		lane |= 1 << (nv94_sor_dp_lane_map(priv, i) >> 3);
-
 	nv_mask(priv, 0x614300 + soff, 0x000c0000, clksor);
 	nv_mask(priv, 0x61c10c + loff, 0x001f4000, dpctrl);
-	nv_mask(priv, 0x61c130 + loff, 0x0000000f, lane);
 	return 0;
 }
 
 static int
-nv94_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int lane, int swing, int preem)
+nv94_sor_dp_drv_ctl(struct nvkm_output_dp *outp, int ln, int vs, int pe, int pc)
 {
-	struct nouveau_bios *bios = nouveau_bios(disp);
-	struct nv50_disp_priv *priv = (void *)disp;
-	const u32 shift = nv94_sor_dp_lane_map(priv, lane);
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	const u32 shift = nv94_sor_dp_lane_map(priv, ln);
 	const u32 loff = nv94_sor_loff(outp);
 	u32 addr, data[3];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
 
-	addr = nvbios_dpout_match(bios, outp->hasht, outp->hashm,
+	addr = nvbios_dpout_match(bios, outp->base.info.hasht,
+					outp->base.info.hashm,
 				 &ver, &hdr, &cnt, &len, &info);
 	if (!addr)
 		return -ENODEV;
 
-	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem,
+	addr = nvbios_dpcfg_match(bios, addr, 0, vs, pe,
 				 &ver, &hdr, &cnt, &len, &ocfg);
 	if (!addr)
 		return -EINVAL;
 
 	data[0] = nv_rd32(priv, 0x61c118 + loff) & ~(0x000000ff << shift);
 	data[1] = nv_rd32(priv, 0x61c120 + loff) & ~(0x000000ff << shift);
-	data[2] = nv_rd32(priv, 0x61c130 + loff) & ~(0x0000ff00);
-	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.drv << shift));
-	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pre << shift));
-	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.unk << 8));
+	data[2] = nv_rd32(priv, 0x61c130 + loff);
+	if ((data[2] & 0x0000ff00) < (ocfg.tx_pu << 8) || ln == 0)
+		data[2] = (data[2] & ~0x0000ff00) | (ocfg.tx_pu << 8);
+	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.dc << shift));
+	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pe << shift));
+	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.tx_pu << 8));
 	return 0;
 }
 
-const struct nouveau_dp_func
-nv94_sor_dp_func = {
+struct nvkm_output_dp_impl
+nv94_sor_dp_impl = {
+	.base.base.handle = DCB_OUTPUT_DP,
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nvkm_output_dp_ctor,
+		.dtor = _nvkm_output_dp_dtor,
+		.init = _nvkm_output_dp_init,
+		.fini = _nvkm_output_dp_fini,
+	},
 	.pattern = nv94_sor_dp_pattern,
+	.lnk_pwr = nv94_sor_dp_lnk_pwr,
 	.lnk_ctl = nv94_sor_dp_lnk_ctl,
 	.drv_ctl = nv94_sor_dp_drv_ctl,
 };
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
index d2df572f16a3..97f0e9cd3d40 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
@@ -29,19 +29,20 @@
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/dp.h>
 #include <subdev/bios/init.h>
+#include <subdev/timer.h>
 
 #include "nv50.h"
 
 static inline u32
-nvd0_sor_soff(struct dcb_output *outp)
+nvd0_sor_soff(struct nvkm_output_dp *outp)
 {
-	return (ffs(outp->or) - 1) * 0x800;
+	return (ffs(outp->base.info.or) - 1) * 0x800;
 }
 
 static inline u32
-nvd0_sor_loff(struct dcb_output *outp)
+nvd0_sor_loff(struct nvkm_output_dp *outp)
 {
-	return nvd0_sor_soff(outp) + !(outp->sorconf.link & 1) * 0x80;
+	return nvd0_sor_soff(outp) + !(outp->base.info.sorconf.link & 1) * 0x80;
 }
 
 static inline u32
@@ -52,77 +53,80 @@ nvd0_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
 }
 
 static int
-nvd0_sor_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int pattern)
+nvd0_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 loff = nvd0_sor_loff(outp);
 	nv_mask(priv, 0x61c110 + loff, 0x0f0f0f0f, 0x01010101 * pattern);
 	return 0;
 }
 
 static int
-nvd0_sor_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int link_nr, int link_bw, bool enh_frame)
+nvd0_sor_dp_lnk_ctl(struct nvkm_output_dp *outp, int nr, int bw, bool ef)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 soff = nvd0_sor_soff(outp);
 	const u32 loff = nvd0_sor_loff(outp);
 	u32 dpctrl = 0x00000000;
 	u32 clksor = 0x00000000;
-	u32 lane = 0;
-	int i;
 
-	clksor |= link_bw << 18;
-	dpctrl |= ((1 << link_nr) - 1) << 16;
-	if (enh_frame)
+	clksor |= bw << 18;
+	dpctrl |= ((1 << nr) - 1) << 16;
+	if (ef)
 		dpctrl |= 0x00004000;
 
-	for (i = 0; i < link_nr; i++)
-		lane |= 1 << (nvd0_sor_dp_lane_map(priv, i) >> 3);
-
 	nv_mask(priv, 0x612300 + soff, 0x007c0000, clksor);
 	nv_mask(priv, 0x61c10c + loff, 0x001f4000, dpctrl);
-	nv_mask(priv, 0x61c130 + loff, 0x0000000f, lane);
 	return 0;
 }
 
 static int
-nvd0_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int lane, int swing, int preem)
+nvd0_sor_dp_drv_ctl(struct nvkm_output_dp *outp, int ln, int vs, int pe, int pc)
 {
-	struct nouveau_bios *bios = nouveau_bios(disp);
-	struct nv50_disp_priv *priv = (void *)disp;
-	const u32 shift = nvd0_sor_dp_lane_map(priv, lane);
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	const u32 shift = nvd0_sor_dp_lane_map(priv, ln);
 	const u32 loff = nvd0_sor_loff(outp);
-	u32 addr, data[3];
+	u32 addr, data[4];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
 
-	addr = nvbios_dpout_match(bios, outp->hasht, outp->hashm,
+	addr = nvbios_dpout_match(bios, outp->base.info.hasht,
+					outp->base.info.hashm,
 				 &ver, &hdr, &cnt, &len, &info);
 	if (!addr)
 		return -ENODEV;
 
-	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem,
+	addr = nvbios_dpcfg_match(bios, addr, pc, vs, pe,
 				 &ver, &hdr, &cnt, &len, &ocfg);
 	if (!addr)
 		return -EINVAL;
 
 	data[0] = nv_rd32(priv, 0x61c118 + loff) & ~(0x000000ff << shift);
 	data[1] = nv_rd32(priv, 0x61c120 + loff) & ~(0x000000ff << shift);
-	data[2] = nv_rd32(priv, 0x61c130 + loff) & ~(0x0000ff00);
-	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.drv << shift));
-	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pre << shift));
-	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.unk << 8));
-	nv_mask(priv, 0x61c13c + loff, 0x00000000, 0x00000000);
+	data[2] = nv_rd32(priv, 0x61c130 + loff);
+	if ((data[2] & 0x0000ff00) < (ocfg.tx_pu << 8) || ln == 0)
+		data[2] = (data[2] & ~0x0000ff00) | (ocfg.tx_pu << 8);
+	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.dc << shift));
+	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pe << shift));
+	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.tx_pu << 8));
+	data[3] = nv_rd32(priv, 0x61c13c + loff) & ~(0x000000ff << shift);
+	nv_wr32(priv, 0x61c13c + loff, data[3] | (ocfg.pc << shift));
 	return 0;
 }
 
-const struct nouveau_dp_func
-nvd0_sor_dp_func = {
+struct nvkm_output_dp_impl
+nvd0_sor_dp_impl = {
+	.base.base.handle = DCB_OUTPUT_DP,
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nvkm_output_dp_ctor,
+		.dtor = _nvkm_output_dp_dtor,
+		.init = _nvkm_output_dp_init,
+		.fini = _nvkm_output_dp_fini,
+	},
 	.pattern = nvd0_sor_dp_pattern,
+	.lnk_pwr = nv94_sor_dp_lnk_pwr,
 	.lnk_ctl = nvd0_sor_dp_lnk_ctl,
 	.drv_ctl = nvd0_sor_dp_drv_ctl,
 };
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/base.c b/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
index 6f9041ced9a2..56ed3d73bf8e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
@@ -91,7 +91,7 @@ nouveau_fifo_channel_create_(struct nouveau_object *parent,
 	if (!chan->user)
 		return -EFAULT;
 
-	nouveau_event_trigger(priv->cevent, 0);
+	nouveau_event_trigger(priv->cevent, 1, 0);
 
 	chan->size = size;
 	return 0;
@@ -194,11 +194,11 @@ nouveau_fifo_create_(struct nouveau_object *parent,
 	if (!priv->channel)
 		return -ENOMEM;
 
-	ret = nouveau_event_create(1, &priv->cevent);
+	ret = nouveau_event_create(1, 1, &priv->cevent);
 	if (ret)
 		return ret;
 
-	ret = nouveau_event_create(1, &priv->uevent);
+	ret = nouveau_event_create(1, 1, &priv->uevent);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/gk20a.c b/drivers/gpu/drm/nouveau/core/engine/fifo/gk20a.c
new file mode 100644
index 000000000000..327456eae963
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/gk20a.c
@@ -0,0 +1,35 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "nve0.h"
+
+struct nouveau_oclass *
+gk20a_fifo_oclass = &(struct nve0_fifo_impl) {
+	.base.handle = NV_ENGINE(FIFO, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nve0_fifo_ctor,
+		.dtor = nve0_fifo_dtor,
+		.init = nve0_fifo_init,
+		.fini = nve0_fifo_fini,
+	},
+	.channels = 128,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
index 54f26cc801c7..c61b16a63884 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
@@ -539,7 +539,7 @@ nv04_fifo_intr(struct nouveau_subdev *subdev)
 			}
 
 			if (status & 0x40000000) {
-				nouveau_event_trigger(priv->base.uevent, 0);
+				nouveau_event_trigger(priv->base.uevent, 1, 0);
 				nv_wr32(priv, 0x002100, 0x40000000);
 				status &= ~0x40000000;
 			}
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
index fe0f41e65d9b..6e5ac16e5460 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
@@ -389,14 +389,14 @@ nv84_fifo_cclass = {
  ******************************************************************************/
 
 static void
-nv84_fifo_uevent_enable(struct nouveau_event *event, int index)
+nv84_fifo_uevent_enable(struct nouveau_event *event, int type, int index)
 {
 	struct nv84_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x40000000, 0x40000000);
 }
 
 static void
-nv84_fifo_uevent_disable(struct nouveau_event *event, int index)
+nv84_fifo_uevent_disable(struct nouveau_event *event, int type, int index)
 {
 	struct nv84_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x40000000, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
index fa1e719872b7..ae4a4dc5642a 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
@@ -730,7 +730,7 @@ nvc0_fifo_intr_engine_unit(struct nvc0_fifo_priv *priv, int engn)
 	for (unkn = 0; unkn < 8; unkn++) {
 		u32 ints = (intr >> (unkn * 0x04)) & inte;
 		if (ints & 0x1) {
-			nouveau_event_trigger(priv->base.uevent, 0);
+			nouveau_event_trigger(priv->base.uevent, 1, 0);
 			ints &= ~1;
 		}
 		if (ints) {
@@ -827,14 +827,14 @@ nvc0_fifo_intr(struct nouveau_subdev *subdev)
 }
 
 static void
-nvc0_fifo_uevent_enable(struct nouveau_event *event, int index)
+nvc0_fifo_uevent_enable(struct nouveau_event *event, int type, int index)
 {
 	struct nvc0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x80000000);
 }
 
 static void
-nvc0_fifo_uevent_disable(struct nouveau_event *event, int index)
+nvc0_fifo_uevent_disable(struct nouveau_event *event, int type, int index)
 {
 	struct nvc0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
index a9a1a9c9f9f2..298063edb92d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
@@ -859,7 +859,7 @@ nve0_fifo_intr_runlist(struct nve0_fifo_priv *priv)
 static void
 nve0_fifo_intr_engine(struct nve0_fifo_priv *priv)
 {
-	nouveau_event_trigger(priv->base.uevent, 0);
+	nouveau_event_trigger(priv->base.uevent, 1, 0);
 }
 
 static void
@@ -952,14 +952,14 @@ nve0_fifo_intr(struct nouveau_subdev *subdev)
 }
 
 static void
-nve0_fifo_uevent_enable(struct nouveau_event *event, int index)
+nve0_fifo_uevent_enable(struct nouveau_event *event, int type, int index)
 {
 	struct nve0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x80000000);
 }
 
 static void
-nve0_fifo_uevent_disable(struct nouveau_event *event, int index)
+nve0_fifo_uevent_disable(struct nouveau_event *event, int type, int index)
 {
 	struct nve0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h
index 014344ebee66..e96b32bb1bbc 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h
@@ -8,6 +8,7 @@ int  nve0_fifo_ctor(struct nouveau_object *, struct nouveau_object *,
 		    struct nouveau_object **);
 void nve0_fifo_dtor(struct nouveau_object *);
 int  nve0_fifo_init(struct nouveau_object *);
+int  nve0_fifo_fini(struct nouveau_object *, bool);
 
 struct nve0_fifo_impl {
 	struct nouveau_oclass base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxgk20a.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxgk20a.c
new file mode 100644
index 000000000000..224ee0287ab7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxgk20a.c
@@ -0,0 +1,53 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "ctxnvc0.h"
+
+static const struct nvc0_graph_pack
+gk20a_grctx_pack_mthd[] = {
+	{ nve4_grctx_init_a097_0, 0xa297 },
+	{ nvc0_grctx_init_902d_0, 0x902d },
+	{}
+};
+
+struct nouveau_oclass *
+gk20a_grctx_oclass = &(struct nvc0_grctx_oclass) {
+	.base.handle = NV_ENGCTX(GR, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_graph_context_ctor,
+		.dtor = nvc0_graph_context_dtor,
+		.init = _nouveau_graph_context_init,
+		.fini = _nouveau_graph_context_fini,
+		.rd32 = _nouveau_graph_context_rd32,
+		.wr32 = _nouveau_graph_context_wr32,
+	},
+	.main  = nve4_grctx_generate_main,
+	.mods  = nve4_grctx_generate_mods,
+	.unkn  = nve4_grctx_generate_unkn,
+	.hub   = nve4_grctx_pack_hub,
+	.gpc   = nve4_grctx_pack_gpc,
+	.zcull = nvc0_grctx_pack_zcull,
+	.tpc   = nve4_grctx_pack_tpc,
+	.ppc   = nve4_grctx_pack_ppc,
+	.icmd  = nve4_grctx_pack_icmd,
+	.mthd  = gk20a_grctx_pack_mthd,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c
index 48351b4d6d6b..8de4a4291548 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c
@@ -545,10 +545,12 @@ nv108_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 	mmio_list(0x408010, 0x80000000,  0, 0);
 	mmio_list(0x419004, 0x00000000,  8, 1);
 	mmio_list(0x419008, 0x00000000,  0, 0);
+	mmio_list(0x4064cc, 0x80000000,  0, 0);
 	mmio_list(0x408004, 0x00000000,  8, 0);
 	mmio_list(0x408008, 0x80000030,  0, 0);
 	mmio_list(0x418808, 0x00000000,  8, 0);
 	mmio_list(0x41880c, 0x80000030,  0, 0);
+	mmio_list(0x4064c8, 0x00c20200,  0, 0);
 	mmio_list(0x418810, 0x80000000, 12, 2);
 	mmio_list(0x419848, 0x10000000, 12, 2);
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h
index 9c815d1f99ef..8da8b627b9d0 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h
@@ -69,7 +69,9 @@ extern struct nouveau_oclass *nvd7_grctx_oclass;
 extern struct nouveau_oclass *nvd9_grctx_oclass;
 
 extern struct nouveau_oclass *nve4_grctx_oclass;
+extern struct nouveau_oclass *gk20a_grctx_oclass;
 void nve4_grctx_generate_main(struct nvc0_graph_priv *, struct nvc0_grctx *);
+void nve4_grctx_generate_mods(struct nvc0_graph_priv *, struct nvc0_grctx *);
 void nve4_grctx_generate_unkn(struct nvc0_graph_priv *);
 void nve4_grctx_generate_r418bb8(struct nvc0_graph_priv *);
 
@@ -151,6 +153,13 @@ extern const struct nvc0_graph_init nve4_grctx_init_gpm_0[];
 
 extern const struct nvc0_graph_init nve4_grctx_init_pes_0[];
 
+extern const struct nvc0_graph_pack nve4_grctx_pack_hub[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_gpc[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_tpc[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_ppc[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_icmd[];
+extern const struct nvc0_graph_init nve4_grctx_init_a097_0[];
+
 extern const struct nvc0_graph_pack nvf0_grctx_pack_mthd[];
 
 extern const struct nvc0_graph_init nvf0_grctx_init_pri_0[];
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c
index 49a14b116a5f..c5b249238587 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c
@@ -272,13 +272,13 @@ nve4_grctx_init_icmd_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_icmd[] = {
 	{ nve4_grctx_init_icmd_0 },
 	{}
 };
 
-static const struct nvc0_graph_init
+const struct nvc0_graph_init
 nve4_grctx_init_a097_0[] = {
 	{ 0x000800,   8, 0x40, 0x00000000 },
 	{ 0x000804,   8, 0x40, 0x00000000 },
@@ -697,7 +697,7 @@ nve4_grctx_init_be_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_hub[] = {
 	{ nvc0_grctx_init_main_0 },
 	{ nve4_grctx_init_fe_0 },
@@ -737,7 +737,7 @@ nve4_grctx_init_gpm_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_gpc[] = {
 	{ nvc0_grctx_init_gpc_unk_0 },
 	{ nvd9_grctx_init_prop_0 },
@@ -802,7 +802,7 @@ nve4_grctx_init_sm_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_tpc[] = {
 	{ nvd7_grctx_init_pe_0 },
 	{ nve4_grctx_init_tex_0 },
@@ -826,7 +826,7 @@ nve4_grctx_init_cbm_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_ppc[] = {
 	{ nve4_grctx_init_pes_0 },
 	{ nve4_grctx_init_cbm_0 },
@@ -838,7 +838,7 @@ nve4_grctx_pack_ppc[] = {
  * PGRAPH context implementation
  ******************************************************************************/
 
-static void
+void
 nve4_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 {
 	u32 magic[GPC_MAX][2];
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c
index 0fab95e49f53..dec03f04114d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c
@@ -842,7 +842,7 @@ nvf0_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 		u16 magic3 = 0x0648;
 		magic[gpc][0]  = 0x10000000 | (magic0 << 16) | offset;
 		magic[gpc][1]  = 0x00000000 | (magic1 << 16);
-		offset += 0x0324 * (priv->tpc_nr[gpc] - 1);;
+		offset += 0x0324 * (priv->tpc_nr[gpc] - 1);
 		magic[gpc][2]  = 0x10000000 | (magic2 << 16) | offset;
 		magic[gpc][3]  = 0x00000000 | (magic3 << 16);
 		offset += 0x0324;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/gpc.fuc b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/gpc.fuc
index 2f7345f7fe07..7445f12b1d9e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/gpc.fuc
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/gpc.fuc
@@ -54,7 +54,7 @@ mmio_list_base:
 #ifdef INCLUDE_CODE
 // reports an exception to the host
 //
-// In: $r15 error code (see nvc0.fuc)
+// In: $r15 error code (see os.h)
 //
 error:
 	push $r14
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hub.fuc b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hub.fuc
index c8ddb8d71b91..b4ad18bf5a26 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hub.fuc
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hub.fuc
@@ -49,7 +49,7 @@ hub_mmio_list_next:
 #ifdef INCLUDE_CODE
 // reports an exception to the host
 //
-// In: $r15 error code (see nvc0.fuc)
+// In: $r15 error code (see os.h)
 //
 error:
 	nv_iowr(NV_PGRAPH_FECS_CC_SCRATCH_VAL(5), 0, $r15)
@@ -343,13 +343,25 @@ ih:
 	ih_no_ctxsw:
 	and $r11 $r10 NV_PGRAPH_FECS_INTR_FWMTHD
 	bra e #ih_no_fwmthd
-		// none we handle, ack, and fall-through to unhandled
+		// none we handle; report to host and ack
+		nv_rd32($r15, NV_PGRAPH_TRAPPED_DATA_LO)
+		nv_iowr(NV_PGRAPH_FECS_CC_SCRATCH_VAL(4), 0, $r15)
+		nv_rd32($r15, NV_PGRAPH_TRAPPED_ADDR)
+		nv_iowr(NV_PGRAPH_FECS_CC_SCRATCH_VAL(3), 0, $r15)
+		extr $r14 $r15 16:18
+		shl b32 $r14 $r14 2
+		imm32($r15, NV_PGRAPH_FE_OBJECT_TABLE(0))
+		add b32 $r14 $r15
+		call(nv_rd32)
+		nv_iowr(NV_PGRAPH_FECS_CC_SCRATCH_VAL(2), 0, $r15)
+		mov $r15 E_BAD_FWMTHD
+		call(error)
 		mov $r11 0x100
 		nv_wr32(0x400144, $r11)
 
 	// anything we didn't handle, bring it to the host's attention
 	ih_no_fwmthd:
-	mov $r11 0x104 // FIFO | CHSW
+	mov $r11 0x504 // FIFO | CHSW | FWMTHD
 	not b32 $r11
 	and $r11 $r10 $r11
 	bra e #ih_no_other
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubgm107.fuc5.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubgm107.fuc5.h
index 214dd16ec566..5f953c5c20b7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubgm107.fuc5.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubgm107.fuc5.h
@@ -478,10 +478,10 @@ uint32_t gm107_grhub_code[] = {
 	0x01040080,
 	0xbd0001f6,
 	0x01004104,
-	0x627e020f,
-	0x717e0006,
+	0xa87e020f,
+	0xb77e0006,
 	0x100f0006,
-	0x0006b37e,
+	0x0006f97e,
 	0x98000e98,
 	0x207e010f,
 	0x14950001,
@@ -523,8 +523,8 @@ uint32_t gm107_grhub_code[] = {
 	0x800040b7,
 	0xf40132b6,
 	0x000fb41b,
-	0x0006b37e,
-	0x627e000f,
+	0x0006f97e,
+	0xa87e000f,
 	0x00800006,
 	0x01f60201,
 	0xbd04bd00,
@@ -554,7 +554,7 @@ uint32_t gm107_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0231f401,
-	0x0008367e,
+	0x00087c7e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -563,7 +563,7 @@ uint32_t gm107_grhub_code[] = {
 	0x37008006,
 	0x0009f602,
 	0x31f404bd,
-	0x08367e01,
+	0x087c7e01,
 	0xf094bd00,
 	0x00800699,
 	0x09f60217,
@@ -572,7 +572,7 @@ uint32_t gm107_grhub_code[] = {
 	0x20f92f0e,
 	0x32f412b2,
 	0x0232f401,
-	0x0008367e,
+	0x00087c7e,
 	0x008020fc,
 	0x02f602c0,
 	0xf404bd00,
@@ -580,7 +580,7 @@ uint32_t gm107_grhub_code[] = {
 	0x23c8130e,
 	0x0d0bf41f,
 	0xf40131f4,
-	0x367e0232,
+	0x7c7e0232,
 /* 0x054e: chsw_done */
 	0x01020008,
 	0x02c30080,
@@ -593,7 +593,7 @@ uint32_t gm107_grhub_code[] = {
 	0xb0ff2a0e,
 	0x1bf401e4,
 	0x7ef2b20c,
-	0xf40007d6,
+	0xf400081c,
 /* 0x057a: main_not_ctx_chan */
 	0xe4b0400e,
 	0x2c1bf402,
@@ -602,7 +602,7 @@ uint32_t gm107_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0232f401,
-	0x0008367e,
+	0x00087c7e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -642,238 +642,238 @@ uint32_t gm107_grhub_code[] = {
 /* 0x061a: ih_no_ctxsw */
 	0xabe40000,
 	0x0bf40400,
-	0x01004b10,
-	0x448ebfb2,
-	0x8f7e4001,
-/* 0x062e: ih_no_fwmthd */
-	0x044b0000,
-	0xffb0bd01,
-	0x0bf4b4ab,
-	0x0700800c,
-	0x000bf603,
-/* 0x0642: ih_no_other */
-	0x004004bd,
-	0x000af601,
-	0xf0fc04bd,
-	0xd0fce0fc,
-	0xa0fcb0fc,
-	0x80fc90fc,
-	0xfc0088fe,
-	0x0032f480,
-/* 0x0662: ctx_4170s */
-	0xf5f001f8,
-	0x8effb210,
-	0x7e404170,
-	0xf800008f,
-/* 0x0671: ctx_4170w */
-	0x41708e00,
+	0x07088e56,
 	0x00657e40,
-	0xf0ffb200,
-	0x1bf410f4,
-/* 0x0683: ctx_redswitch */
-	0x4e00f8f3,
-	0xe5f00200,
-	0x20e5f040,
-	0x8010e5f0,
-	0xf6018500,
-	0x04bd000e,
-/* 0x069a: ctx_redswitch_delay */
-	0xf2b6080f,
-	0xfd1bf401,
-	0x0400e5f1,
-	0x0100e5f1,
-	0x01850080,
-	0xbd000ef6,
-/* 0x06b3: ctx_86c */
-	0x8000f804,
-	0xf6022300,
+	0x80ffb200,
+	0xf6020400,
 	0x04bd000f,
-	0x148effb2,
-	0x8f7e408a,
-	0xffb20000,
-	0x41a88c8e,
+	0x4007048e,
+	0x0000657e,
+	0x0080ffb2,
+	0x0ff60203,
+	0xc704bd00,
+	0xee9450fe,
+	0x07008f02,
+	0x00efbb40,
+	0x0000657e,
+	0x02020080,
+	0xbd000ff6,
+	0x7e030f04,
+	0x4b0002f8,
+	0xbfb20100,
+	0x4001448e,
 	0x00008f7e,
-/* 0x06d2: ctx_mem */
-	0x008000f8,
-	0x0ff60284,
-/* 0x06db: ctx_mem_wait */
-	0x8f04bd00,
-	0xcf028400,
-	0xfffd00ff,
-	0xf61bf405,
-/* 0x06ea: ctx_load */
-	0x94bd00f8,
-	0x800599f0,
-	0xf6023700,
-	0x04bd0009,
-	0xb87e0c0a,
-	0xf4bd0000,
-	0x02890080,
+/* 0x0674: ih_no_fwmthd */
+	0xbd05044b,
+	0xb4abffb0,
+	0x800c0bf4,
+	0xf6030700,
+	0x04bd000b,
+/* 0x0688: ih_no_other */
+	0xf6010040,
+	0x04bd000a,
+	0xe0fcf0fc,
+	0xb0fcd0fc,
+	0x90fca0fc,
+	0x88fe80fc,
+	0xf480fc00,
+	0x01f80032,
+/* 0x06a8: ctx_4170s */
+	0xb210f5f0,
+	0x41708eff,
+	0x008f7e40,
+/* 0x06b7: ctx_4170w */
+	0x8e00f800,
+	0x7e404170,
+	0xb2000065,
+	0x10f4f0ff,
+	0xf8f31bf4,
+/* 0x06c9: ctx_redswitch */
+	0x02004e00,
+	0xf040e5f0,
+	0xe5f020e5,
+	0x85008010,
+	0x000ef601,
+	0x080f04bd,
+/* 0x06e0: ctx_redswitch_delay */
+	0xf401f2b6,
+	0xe5f1fd1b,
+	0xe5f10400,
+	0x00800100,
+	0x0ef60185,
+	0xf804bd00,
+/* 0x06f9: ctx_86c */
+	0x23008000,
+	0x000ff602,
+	0xffb204bd,
+	0x408a148e,
+	0x00008f7e,
+	0x8c8effb2,
+	0x8f7e41a8,
+	0x00f80000,
+/* 0x0718: ctx_mem */
+	0x02840080,
 	0xbd000ff6,
-	0xc1008004,
-	0x0002f602,
-	0x008004bd,
-	0x02f60283,
-	0x0f04bd00,
-	0x06d27e07,
-	0xc0008000,
-	0x0002f602,
-	0x0bfe04bd,
-	0x1f2af000,
-	0xb60424b6,
-	0x94bd0220,
-	0x800899f0,
-	0xf6023700,
-	0x04bd0009,
-	0x02810080,
-	0xbd0002f6,
-	0x0000d204,
-	0x25f08000,
-	0x88008002,
-	0x0002f602,
-	0x100104bd,
-	0xf0020042,
-	0x12fa0223,
-	0xbd03f805,
-	0x0899f094,
-	0x02170080,
-	0xbd0009f6,
-	0x81019804,
-	0x981814b6,
-	0x25b68002,
-	0x0512fd08,
-	0xbd1601b5,
-	0x0999f094,
-	0x02370080,
-	0xbd0009f6,
-	0x81008004,
-	0x0001f602,
-	0x010204bd,
-	0x02880080,
+/* 0x0721: ctx_mem_wait */
+	0x84008f04,
+	0x00ffcf02,
+	0xf405fffd,
+	0x00f8f61b,
+/* 0x0730: ctx_load */
+	0x99f094bd,
+	0x37008005,
+	0x0009f602,
+	0x0c0a04bd,
+	0x0000b87e,
+	0x0080f4bd,
+	0x0ff60289,
+	0x8004bd00,
+	0xf602c100,
+	0x04bd0002,
+	0x02830080,
 	0xbd0002f6,
-	0x01004104,
-	0xfa0613f0,
-	0x03f80501,
+	0x7e070f04,
+	0x80000718,
+	0xf602c000,
+	0x04bd0002,
+	0xf0000bfe,
+	0x24b61f2a,
+	0x0220b604,
 	0x99f094bd,
-	0x17008009,
+	0x37008008,
 	0x0009f602,
-	0x94bd04bd,
-	0x800599f0,
+	0x008004bd,
+	0x02f60281,
+	0xd204bd00,
+	0x80000000,
+	0x800225f0,
+	0xf6028800,
+	0x04bd0002,
+	0x00421001,
+	0x0223f002,
+	0xf80512fa,
+	0xf094bd03,
+	0x00800899,
+	0x09f60217,
+	0x9804bd00,
+	0x14b68101,
+	0x80029818,
+	0xfd0825b6,
+	0x01b50512,
+	0xf094bd16,
+	0x00800999,
+	0x09f60237,
+	0x8004bd00,
+	0xf6028100,
+	0x04bd0001,
+	0x00800102,
+	0x02f60288,
+	0x4104bd00,
+	0x13f00100,
+	0x0501fa06,
+	0x94bd03f8,
+	0x800999f0,
 	0xf6021700,
 	0x04bd0009,
-/* 0x07d6: ctx_chan */
-	0xea7e00f8,
-	0x0c0a0006,
-	0x0000b87e,
-	0xd27e050f,
-	0x00f80006,
-/* 0x07e8: ctx_mmio_exec */
-	0x80410398,
+	0x99f094bd,
+	0x17008005,
+	0x0009f602,
+	0x00f804bd,
+/* 0x081c: ctx_chan */
+	0x0007307e,
+	0xb87e0c0a,
+	0x050f0000,
+	0x0007187e,
+/* 0x082e: ctx_mmio_exec */
+	0x039800f8,
+	0x81008041,
+	0x0003f602,
+	0x34bd04bd,
+/* 0x083c: ctx_mmio_loop */
+	0xf4ff34c4,
+	0x00450e1b,
+	0x0653f002,
+	0xf80535fa,
+/* 0x084d: ctx_mmio_pull */
+	0x804e9803,
+	0x7e814f98,
+	0xb600008f,
+	0x12b60830,
+	0xdf1bf401,
+/* 0x0860: ctx_mmio_done */
+	0x80160398,
 	0xf6028100,
 	0x04bd0003,
-/* 0x07f6: ctx_mmio_loop */
-	0x34c434bd,
-	0x0e1bf4ff,
-	0xf0020045,
-	0x35fa0653,
-/* 0x0807: ctx_mmio_pull */
-	0x9803f805,
-	0x4f98804e,
-	0x008f7e81,
-	0x0830b600,
-	0xf40112b6,
-/* 0x081a: ctx_mmio_done */
-	0x0398df1b,
-	0x81008016,
-	0x0003f602,
-	0x00b504bd,
-	0x01004140,
-	0xfa0613f0,
-	0x03f80601,
-/* 0x0836: ctx_xfer */
-	0x040e00f8,
-	0x03020080,
-	0xbd000ef6,
-/* 0x0841: ctx_xfer_idle */
-	0x00008e04,
-	0x00eecf03,
-	0x2000e4f1,
-	0xf4f51bf4,
-	0x02f40611,
-/* 0x0855: ctx_xfer_pre */
-	0x7e100f0c,
-	0xf40006b3,
-/* 0x085e: ctx_xfer_pre_load */
-	0x020f1b11,
-	0x0006627e,
-	0x0006717e,
-	0x0006837e,
-	0x627ef4bd,
-	0xea7e0006,
-/* 0x0876: ctx_xfer_exec */
-	0x01980006,
-	0x8024bd16,
-	0xf6010500,
-	0x04bd0002,
-	0x008e1fb2,
-	0x8f7e41a5,
-	0xfcf00000,
-	0x022cf001,
-	0xfd0124b6,
-	0xffb205f2,
-	0x41a5048e,
+	0x414000b5,
+	0x13f00100,
+	0x0601fa06,
+	0x00f803f8,
+/* 0x087c: ctx_xfer */
+	0x0080040e,
+	0x0ef60302,
+/* 0x0887: ctx_xfer_idle */
+	0x8e04bd00,
+	0xcf030000,
+	0xe4f100ee,
+	0x1bf42000,
+	0x0611f4f5,
+/* 0x089b: ctx_xfer_pre */
+	0x0f0c02f4,
+	0x06f97e10,
+	0x1b11f400,
+/* 0x08a4: ctx_xfer_pre_load */
+	0xa87e020f,
+	0xb77e0006,
+	0xc97e0006,
+	0xf4bd0006,
+	0x0006a87e,
+	0x0007307e,
+/* 0x08bc: ctx_xfer_exec */
+	0xbd160198,
+	0x05008024,
+	0x0002f601,
+	0x1fb204bd,
+	0x41a5008e,
 	0x00008f7e,
-	0x0002167e,
-	0xfc8024bd,
-	0x02f60247,
-	0xf004bd00,
-	0x20b6012c,
-	0x4afc8003,
-	0x0002f602,
-	0xacf004bd,
-	0x06a5f001,
-	0x0c98000b,
-	0x010d9800,
-	0x3d7e000e,
-	0x080a0001,
-	0x0000ec7e,
-	0x00020a7e,
-	0x0a1201f4,
-	0x00b87e0c,
-	0x7e050f00,
-	0xf40006d2,
-/* 0x08f2: ctx_xfer_post */
-	0x020f2d02,
-	0x0006627e,
-	0xb37ef4bd,
-	0x277e0006,
-	0x717e0002,
+	0xf001fcf0,
+	0x24b6022c,
+	0x05f2fd01,
+	0x048effb2,
+	0x8f7e41a5,
+	0x167e0000,
+	0x24bd0002,
+	0x0247fc80,
+	0xbd0002f6,
+	0x012cf004,
+	0x800320b6,
+	0xf6024afc,
+	0x04bd0002,
+	0xf001acf0,
+	0x000b06a5,
+	0x98000c98,
+	0x000e010d,
+	0x00013d7e,
+	0xec7e080a,
+	0x0a7e0000,
+	0x01f40002,
+	0x7e0c0a12,
+	0x0f0000b8,
+	0x07187e05,
+	0x2d02f400,
+/* 0x0938: ctx_xfer_post */
+	0xa87e020f,
 	0xf4bd0006,
-	0x0006627e,
-	0x981011f4,
-	0x11fd4001,
-	0x070bf405,
-	0x0007e87e,
-/* 0x091c: ctx_xfer_no_post_mmio */
-/* 0x091c: ctx_xfer_done */
-	0x000000f8,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
+	0x0006f97e,
+	0x0002277e,
+	0x0006b77e,
+	0xa87ef4bd,
+	0x11f40006,
+	0x40019810,
+	0xf40511fd,
+	0x2e7e070b,
+/* 0x0962: ctx_xfer_no_post_mmio */
+/* 0x0962: ctx_xfer_done */
+	0x00f80008,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnv108.fuc5.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnv108.fuc5.h
index 64dfd75192bf..e49b5a877ae4 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnv108.fuc5.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnv108.fuc5.h
@@ -478,10 +478,10 @@ uint32_t nv108_grhub_code[] = {
 	0x01040080,
 	0xbd0001f6,
 	0x01004104,
-	0x627e020f,
-	0x717e0006,
+	0xa87e020f,
+	0xb77e0006,
 	0x100f0006,
-	0x0006b37e,
+	0x0006f97e,
 	0x98000e98,
 	0x207e010f,
 	0x14950001,
@@ -523,8 +523,8 @@ uint32_t nv108_grhub_code[] = {
 	0x800040b7,
 	0xf40132b6,
 	0x000fb41b,
-	0x0006b37e,
-	0x627e000f,
+	0x0006f97e,
+	0xa87e000f,
 	0x00800006,
 	0x01f60201,
 	0xbd04bd00,
@@ -554,7 +554,7 @@ uint32_t nv108_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0231f401,
-	0x0008367e,
+	0x00087c7e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -563,7 +563,7 @@ uint32_t nv108_grhub_code[] = {
 	0x37008006,
 	0x0009f602,
 	0x31f404bd,
-	0x08367e01,
+	0x087c7e01,
 	0xf094bd00,
 	0x00800699,
 	0x09f60217,
@@ -572,7 +572,7 @@ uint32_t nv108_grhub_code[] = {
 	0x20f92f0e,
 	0x32f412b2,
 	0x0232f401,
-	0x0008367e,
+	0x00087c7e,
 	0x008020fc,
 	0x02f602c0,
 	0xf404bd00,
@@ -580,7 +580,7 @@ uint32_t nv108_grhub_code[] = {
 	0x23c8130e,
 	0x0d0bf41f,
 	0xf40131f4,
-	0x367e0232,
+	0x7c7e0232,
 /* 0x054e: chsw_done */
 	0x01020008,
 	0x02c30080,
@@ -593,7 +593,7 @@ uint32_t nv108_grhub_code[] = {
 	0xb0ff2a0e,
 	0x1bf401e4,
 	0x7ef2b20c,
-	0xf40007d6,
+	0xf400081c,
 /* 0x057a: main_not_ctx_chan */
 	0xe4b0400e,
 	0x2c1bf402,
@@ -602,7 +602,7 @@ uint32_t nv108_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0232f401,
-	0x0008367e,
+	0x00087c7e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -642,238 +642,238 @@ uint32_t nv108_grhub_code[] = {
 /* 0x061a: ih_no_ctxsw */
 	0xabe40000,
 	0x0bf40400,
-	0x01004b10,
-	0x448ebfb2,
-	0x8f7e4001,
-/* 0x062e: ih_no_fwmthd */
-	0x044b0000,
-	0xffb0bd01,
-	0x0bf4b4ab,
-	0x0700800c,
-	0x000bf603,
-/* 0x0642: ih_no_other */
-	0x004004bd,
-	0x000af601,
-	0xf0fc04bd,
-	0xd0fce0fc,
-	0xa0fcb0fc,
-	0x80fc90fc,
-	0xfc0088fe,
-	0x0032f480,
-/* 0x0662: ctx_4170s */
-	0xf5f001f8,
-	0x8effb210,
-	0x7e404170,
-	0xf800008f,
-/* 0x0671: ctx_4170w */
-	0x41708e00,
+	0x07088e56,
 	0x00657e40,
-	0xf0ffb200,
-	0x1bf410f4,
-/* 0x0683: ctx_redswitch */
-	0x4e00f8f3,
-	0xe5f00200,
-	0x20e5f040,
-	0x8010e5f0,
-	0xf6018500,
-	0x04bd000e,
-/* 0x069a: ctx_redswitch_delay */
-	0xf2b6080f,
-	0xfd1bf401,
-	0x0400e5f1,
-	0x0100e5f1,
-	0x01850080,
-	0xbd000ef6,
-/* 0x06b3: ctx_86c */
-	0x8000f804,
-	0xf6022300,
+	0x80ffb200,
+	0xf6020400,
 	0x04bd000f,
-	0x148effb2,
-	0x8f7e408a,
-	0xffb20000,
-	0x41a88c8e,
+	0x4007048e,
+	0x0000657e,
+	0x0080ffb2,
+	0x0ff60203,
+	0xc704bd00,
+	0xee9450fe,
+	0x07008f02,
+	0x00efbb40,
+	0x0000657e,
+	0x02020080,
+	0xbd000ff6,
+	0x7e030f04,
+	0x4b0002f8,
+	0xbfb20100,
+	0x4001448e,
 	0x00008f7e,
-/* 0x06d2: ctx_mem */
-	0x008000f8,
-	0x0ff60284,
-/* 0x06db: ctx_mem_wait */
-	0x8f04bd00,
-	0xcf028400,
-	0xfffd00ff,
-	0xf61bf405,
-/* 0x06ea: ctx_load */
-	0x94bd00f8,
-	0x800599f0,
-	0xf6023700,
-	0x04bd0009,
-	0xb87e0c0a,
-	0xf4bd0000,
-	0x02890080,
+/* 0x0674: ih_no_fwmthd */
+	0xbd05044b,
+	0xb4abffb0,
+	0x800c0bf4,
+	0xf6030700,
+	0x04bd000b,
+/* 0x0688: ih_no_other */
+	0xf6010040,
+	0x04bd000a,
+	0xe0fcf0fc,
+	0xb0fcd0fc,
+	0x90fca0fc,
+	0x88fe80fc,
+	0xf480fc00,
+	0x01f80032,
+/* 0x06a8: ctx_4170s */
+	0xb210f5f0,
+	0x41708eff,
+	0x008f7e40,
+/* 0x06b7: ctx_4170w */
+	0x8e00f800,
+	0x7e404170,
+	0xb2000065,
+	0x10f4f0ff,
+	0xf8f31bf4,
+/* 0x06c9: ctx_redswitch */
+	0x02004e00,
+	0xf040e5f0,
+	0xe5f020e5,
+	0x85008010,
+	0x000ef601,
+	0x080f04bd,
+/* 0x06e0: ctx_redswitch_delay */
+	0xf401f2b6,
+	0xe5f1fd1b,
+	0xe5f10400,
+	0x00800100,
+	0x0ef60185,
+	0xf804bd00,
+/* 0x06f9: ctx_86c */
+	0x23008000,
+	0x000ff602,
+	0xffb204bd,
+	0x408a148e,
+	0x00008f7e,
+	0x8c8effb2,
+	0x8f7e41a8,
+	0x00f80000,
+/* 0x0718: ctx_mem */
+	0x02840080,
 	0xbd000ff6,
-	0xc1008004,
-	0x0002f602,
-	0x008004bd,
-	0x02f60283,
-	0x0f04bd00,
-	0x06d27e07,
-	0xc0008000,
-	0x0002f602,
-	0x0bfe04bd,
-	0x1f2af000,
-	0xb60424b6,
-	0x94bd0220,
-	0x800899f0,
-	0xf6023700,
-	0x04bd0009,
-	0x02810080,
-	0xbd0002f6,
-	0x0000d204,
-	0x25f08000,
-	0x88008002,
-	0x0002f602,
-	0x100104bd,
-	0xf0020042,
-	0x12fa0223,
-	0xbd03f805,
-	0x0899f094,
-	0x02170080,
-	0xbd0009f6,
-	0x81019804,
-	0x981814b6,
-	0x25b68002,
-	0x0512fd08,
-	0xbd1601b5,
-	0x0999f094,
-	0x02370080,
-	0xbd0009f6,
-	0x81008004,
-	0x0001f602,
-	0x010204bd,
-	0x02880080,
+/* 0x0721: ctx_mem_wait */
+	0x84008f04,
+	0x00ffcf02,
+	0xf405fffd,
+	0x00f8f61b,
+/* 0x0730: ctx_load */
+	0x99f094bd,
+	0x37008005,
+	0x0009f602,
+	0x0c0a04bd,
+	0x0000b87e,
+	0x0080f4bd,
+	0x0ff60289,
+	0x8004bd00,
+	0xf602c100,
+	0x04bd0002,
+	0x02830080,
 	0xbd0002f6,
-	0x01004104,
-	0xfa0613f0,
-	0x03f80501,
+	0x7e070f04,
+	0x80000718,
+	0xf602c000,
+	0x04bd0002,
+	0xf0000bfe,
+	0x24b61f2a,
+	0x0220b604,
 	0x99f094bd,
-	0x17008009,
+	0x37008008,
 	0x0009f602,
-	0x94bd04bd,
-	0x800599f0,
+	0x008004bd,
+	0x02f60281,
+	0xd204bd00,
+	0x80000000,
+	0x800225f0,
+	0xf6028800,
+	0x04bd0002,
+	0x00421001,
+	0x0223f002,
+	0xf80512fa,
+	0xf094bd03,
+	0x00800899,
+	0x09f60217,
+	0x9804bd00,
+	0x14b68101,
+	0x80029818,
+	0xfd0825b6,
+	0x01b50512,
+	0xf094bd16,
+	0x00800999,
+	0x09f60237,
+	0x8004bd00,
+	0xf6028100,
+	0x04bd0001,
+	0x00800102,
+	0x02f60288,
+	0x4104bd00,
+	0x13f00100,
+	0x0501fa06,
+	0x94bd03f8,
+	0x800999f0,
 	0xf6021700,
 	0x04bd0009,
-/* 0x07d6: ctx_chan */
-	0xea7e00f8,
-	0x0c0a0006,
-	0x0000b87e,
-	0xd27e050f,
-	0x00f80006,
-/* 0x07e8: ctx_mmio_exec */
-	0x80410398,
+	0x99f094bd,
+	0x17008005,
+	0x0009f602,
+	0x00f804bd,
+/* 0x081c: ctx_chan */
+	0x0007307e,
+	0xb87e0c0a,
+	0x050f0000,
+	0x0007187e,
+/* 0x082e: ctx_mmio_exec */
+	0x039800f8,
+	0x81008041,
+	0x0003f602,
+	0x34bd04bd,
+/* 0x083c: ctx_mmio_loop */
+	0xf4ff34c4,
+	0x00450e1b,
+	0x0653f002,
+	0xf80535fa,
+/* 0x084d: ctx_mmio_pull */
+	0x804e9803,
+	0x7e814f98,
+	0xb600008f,
+	0x12b60830,
+	0xdf1bf401,
+/* 0x0860: ctx_mmio_done */
+	0x80160398,
 	0xf6028100,
 	0x04bd0003,
-/* 0x07f6: ctx_mmio_loop */
-	0x34c434bd,
-	0x0e1bf4ff,
-	0xf0020045,
-	0x35fa0653,
-/* 0x0807: ctx_mmio_pull */
-	0x9803f805,
-	0x4f98804e,
-	0x008f7e81,
-	0x0830b600,
-	0xf40112b6,
-/* 0x081a: ctx_mmio_done */
-	0x0398df1b,
-	0x81008016,
-	0x0003f602,
-	0x00b504bd,
-	0x01004140,
-	0xfa0613f0,
-	0x03f80601,
-/* 0x0836: ctx_xfer */
-	0x040e00f8,
-	0x03020080,
-	0xbd000ef6,
-/* 0x0841: ctx_xfer_idle */
-	0x00008e04,
-	0x00eecf03,
-	0x2000e4f1,
-	0xf4f51bf4,
-	0x02f40611,
-/* 0x0855: ctx_xfer_pre */
-	0x7e100f0c,
-	0xf40006b3,
-/* 0x085e: ctx_xfer_pre_load */
-	0x020f1b11,
-	0x0006627e,
-	0x0006717e,
-	0x0006837e,
-	0x627ef4bd,
-	0xea7e0006,
-/* 0x0876: ctx_xfer_exec */
-	0x01980006,
-	0x8024bd16,
-	0xf6010500,
-	0x04bd0002,
-	0x008e1fb2,
-	0x8f7e41a5,
-	0xfcf00000,
-	0x022cf001,
-	0xfd0124b6,
-	0xffb205f2,
-	0x41a5048e,
+	0x414000b5,
+	0x13f00100,
+	0x0601fa06,
+	0x00f803f8,
+/* 0x087c: ctx_xfer */
+	0x0080040e,
+	0x0ef60302,
+/* 0x0887: ctx_xfer_idle */
+	0x8e04bd00,
+	0xcf030000,
+	0xe4f100ee,
+	0x1bf42000,
+	0x0611f4f5,
+/* 0x089b: ctx_xfer_pre */
+	0x0f0c02f4,
+	0x06f97e10,
+	0x1b11f400,
+/* 0x08a4: ctx_xfer_pre_load */
+	0xa87e020f,
+	0xb77e0006,
+	0xc97e0006,
+	0xf4bd0006,
+	0x0006a87e,
+	0x0007307e,
+/* 0x08bc: ctx_xfer_exec */
+	0xbd160198,
+	0x05008024,
+	0x0002f601,
+	0x1fb204bd,
+	0x41a5008e,
 	0x00008f7e,
-	0x0002167e,
-	0xfc8024bd,
-	0x02f60247,
-	0xf004bd00,
-	0x20b6012c,
-	0x4afc8003,
-	0x0002f602,
-	0xacf004bd,
-	0x06a5f001,
-	0x0c98000b,
-	0x010d9800,
-	0x3d7e000e,
-	0x080a0001,
-	0x0000ec7e,
-	0x00020a7e,
-	0x0a1201f4,
-	0x00b87e0c,
-	0x7e050f00,
-	0xf40006d2,
-/* 0x08f2: ctx_xfer_post */
-	0x020f2d02,
-	0x0006627e,
-	0xb37ef4bd,
-	0x277e0006,
-	0x717e0002,
+	0xf001fcf0,
+	0x24b6022c,
+	0x05f2fd01,
+	0x048effb2,
+	0x8f7e41a5,
+	0x167e0000,
+	0x24bd0002,
+	0x0247fc80,
+	0xbd0002f6,
+	0x012cf004,
+	0x800320b6,
+	0xf6024afc,
+	0x04bd0002,
+	0xf001acf0,
+	0x000b06a5,
+	0x98000c98,
+	0x000e010d,
+	0x00013d7e,
+	0xec7e080a,
+	0x0a7e0000,
+	0x01f40002,
+	0x7e0c0a12,
+	0x0f0000b8,
+	0x07187e05,
+	0x2d02f400,
+/* 0x0938: ctx_xfer_post */
+	0xa87e020f,
 	0xf4bd0006,
-	0x0006627e,
-	0x981011f4,
-	0x11fd4001,
-	0x070bf405,
-	0x0007e87e,
-/* 0x091c: ctx_xfer_no_post_mmio */
-/* 0x091c: ctx_xfer_done */
-	0x000000f8,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
+	0x0006f97e,
+	0x0002277e,
+	0x0006b77e,
+	0xa87ef4bd,
+	0x11f40006,
+	0x40019810,
+	0xf40511fd,
+	0x2e7e070b,
+/* 0x0962: ctx_xfer_no_post_mmio */
+/* 0x0962: ctx_xfer_done */
+	0x00f80008,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvc0.fuc.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvc0.fuc.h
index f8f7b278a13f..92dfe6a4ac87 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvc0.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvc0.fuc.h
@@ -528,10 +528,10 @@ uint32_t nvc0_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0xb521f502,
-	0xc721f507,
-	0x10f7f007,
-	0x081421f5,
+	0x0d21f502,
+	0x1f21f508,
+	0x10f7f008,
+	0x086c21f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t nvc0_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x081421f5,
+	0x086c21f5,
 	0xf500f7f0,
-	0xf107b521,
+	0xf1080d21,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,8 +610,8 @@ uint32_t nvc0_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0xe821f502,
-	0xf094bd09,
+	0x4021f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -621,7 +621,7 @@ uint32_t nvc0_grhub_code[] = {
 	0x0203f00f,
 	0xbd0009d0,
 	0x0131f404,
-	0x09e821f5,
+	0x0a4021f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t nvc0_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc09e821,
+	0xfc0a4021,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t nvc0_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x09e821f5,
+	0x0a4021f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t nvc0_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0x7821f502,
+	0xd021f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,8 +664,8 @@ uint32_t nvc0_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0xe821f502,
-	0xf094bd09,
+	0x4021f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -710,18 +710,40 @@ uint32_t nvc0_grhub_code[] = {
 /* 0x072b: ih_no_ctxsw */
 	0xe40421f4,
 	0xf40400ab,
-	0xb7f1140b,
+	0xe7f16c0b,
+	0xe3f00708,
+	0x6821f440,
+	0xf102ffb9,
+	0xf0040007,
+	0x0fd00203,
+	0xf104bd00,
+	0xf00704e7,
+	0x21f440e3,
+	0x02ffb968,
+	0x030007f1,
+	0xd00203f0,
+	0x04bd000f,
+	0x9450fec7,
+	0xf7f102ee,
+	0xf3f00700,
+	0x00efbb40,
+	0xf16821f4,
+	0xf0020007,
+	0x0fd00203,
+	0xf004bd00,
+	0x21f503f7,
+	0xb7f1037e,
 	0xbfb90100,
 	0x44e7f102,
 	0x40e3f001,
-/* 0x0743: ih_no_fwmthd */
+/* 0x079b: ih_no_fwmthd */
 	0xf19d21f4,
-	0xbd0104b7,
+	0xbd0504b7,
 	0xb4abffb0,
 	0xf10f0bf4,
 	0xf0070007,
 	0x0bd00303,
-/* 0x075b: ih_no_other */
+/* 0x07b3: ih_no_other */
 	0xf104bd00,
 	0xf0010007,
 	0x0ad00003,
@@ -731,36 +753,36 @@ uint32_t nvc0_grhub_code[] = {
 	0xfc90fca0,
 	0x0088fe80,
 	0x32f480fc,
-/* 0x077f: ctx_4160s */
+/* 0x07d7: ctx_4160s */
 	0xf001f800,
 	0xffb901f7,
 	0x60e7f102,
 	0x40e3f041,
-/* 0x078f: ctx_4160s_wait */
+/* 0x07e7: ctx_4160s_wait */
 	0xf19d21f4,
 	0xf04160e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf404ffc8,
 	0x00f8f00b,
-/* 0x07a4: ctx_4160c */
+/* 0x07fc: ctx_4160c */
 	0xffb9f4bd,
 	0x60e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x07b5: ctx_4170s */
+/* 0x080d: ctx_4170s */
 	0x10f5f000,
 	0xf102ffb9,
 	0xf04170e7,
 	0x21f440e3,
-/* 0x07c7: ctx_4170w */
+/* 0x081f: ctx_4170w */
 	0xf100f89d,
 	0xf04170e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf410f4f0,
 	0x00f8f01b,
-/* 0x07dc: ctx_redswitch */
+/* 0x0834: ctx_redswitch */
 	0x0200e7f1,
 	0xf040e5f0,
 	0xe5f020e5,
@@ -768,7 +790,7 @@ uint32_t nvc0_grhub_code[] = {
 	0x0103f085,
 	0xbd000ed0,
 	0x08f7f004,
-/* 0x07f8: ctx_redswitch_delay */
+/* 0x0850: ctx_redswitch_delay */
 	0xf401f2b6,
 	0xe5f1fd1b,
 	0xe5f10400,
@@ -776,7 +798,7 @@ uint32_t nvc0_grhub_code[] = {
 	0x03f08500,
 	0x000ed001,
 	0x00f804bd,
-/* 0x0814: ctx_86c */
+/* 0x086c: ctx_86c */
 	0x1b0007f1,
 	0xd00203f0,
 	0x04bd000f,
@@ -787,16 +809,16 @@ uint32_t nvc0_grhub_code[] = {
 	0xa86ce7f1,
 	0xf441e3f0,
 	0x00f89d21,
-/* 0x083c: ctx_mem */
+/* 0x0894: ctx_mem */
 	0x840007f1,
 	0xd00203f0,
 	0x04bd000f,
-/* 0x0848: ctx_mem_wait */
+/* 0x08a0: ctx_mem_wait */
 	0x8400f7f1,
 	0xcf02f3f0,
 	0xfffd00ff,
 	0xf31bf405,
-/* 0x085a: ctx_load */
+/* 0x08b2: ctx_load */
 	0x94bd00f8,
 	0xf10599f0,
 	0xf00f0007,
@@ -814,7 +836,7 @@ uint32_t nvc0_grhub_code[] = {
 	0x02d00203,
 	0xf004bd00,
 	0x21f507f7,
-	0x07f1083c,
+	0x07f10894,
 	0x03f0c000,
 	0x0002d002,
 	0x0bfe04bd,
@@ -869,31 +891,31 @@ uint32_t nvc0_grhub_code[] = {
 	0x03f01700,
 	0x0009d002,
 	0x00f804bd,
-/* 0x0978: ctx_chan */
-	0x077f21f5,
-	0x085a21f5,
+/* 0x09d0: ctx_chan */
+	0x07d721f5,
+	0x08b221f5,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x3c21f505,
-	0xa421f508,
-/* 0x0993: ctx_mmio_exec */
+	0x9421f505,
+	0xfc21f508,
+/* 0x09eb: ctx_mmio_exec */
 	0x9800f807,
 	0x07f14103,
 	0x03f08100,
 	0x0003d002,
 	0x34bd04bd,
-/* 0x09a4: ctx_mmio_loop */
+/* 0x09fc: ctx_mmio_loop */
 	0xf4ff34c4,
 	0x57f10f1b,
 	0x53f00200,
 	0x0535fa06,
-/* 0x09b6: ctx_mmio_pull */
+/* 0x0a0e: ctx_mmio_pull */
 	0x4e9803f8,
 	0x814f9880,
 	0xb69d21f4,
 	0x12b60830,
 	0xdf1bf401,
-/* 0x09c8: ctx_mmio_done */
+/* 0x0a20: ctx_mmio_done */
 	0xf1160398,
 	0xf0810007,
 	0x03d00203,
@@ -902,30 +924,30 @@ uint32_t nvc0_grhub_code[] = {
 	0x13f00100,
 	0x0601fa06,
 	0x00f803f8,
-/* 0x09e8: ctx_xfer */
+/* 0x0a40: ctx_xfer */
 	0xf104e7f0,
 	0xf0020007,
 	0x0ed00303,
-/* 0x09f7: ctx_xfer_idle */
+/* 0x0a4f: ctx_xfer_idle */
 	0xf104bd00,
 	0xf00000e7,
 	0xeecf03e3,
 	0x00e4f100,
 	0xf21bf420,
 	0xf40611f4,
-/* 0x0a0e: ctx_xfer_pre */
+/* 0x0a66: ctx_xfer_pre */
 	0xf7f01102,
-	0x1421f510,
-	0x7f21f508,
+	0x6c21f510,
+	0xd721f508,
 	0x1c11f407,
-/* 0x0a1c: ctx_xfer_pre_load */
+/* 0x0a74: ctx_xfer_pre_load */
 	0xf502f7f0,
-	0xf507b521,
-	0xf507c721,
-	0xbd07dc21,
-	0xb521f5f4,
-	0x5a21f507,
-/* 0x0a35: ctx_xfer_exec */
+	0xf5080d21,
+	0xf5081f21,
+	0xbd083421,
+	0x0d21f5f4,
+	0xb221f508,
+/* 0x0a8d: ctx_xfer_exec */
 	0x16019808,
 	0x07f124bd,
 	0x03f00500,
@@ -960,23 +982,65 @@ uint32_t nvc0_grhub_code[] = {
 	0x1301f402,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x3c21f505,
+	0x9421f505,
 	0x3202f408,
-/* 0x0ac4: ctx_xfer_post */
+/* 0x0b1c: ctx_xfer_post */
 	0xf502f7f0,
-	0xbd07b521,
-	0x1421f5f4,
+	0xbd080d21,
+	0x6c21f5f4,
 	0x7f21f508,
-	0xc721f502,
-	0xf5f4bd07,
-	0xf407b521,
+	0x1f21f502,
+	0xf5f4bd08,
+	0xf4080d21,
 	0x01981011,
 	0x0511fd40,
 	0xf5070bf4,
-/* 0x0aef: ctx_xfer_no_post_mmio */
-	0xf5099321,
-/* 0x0af3: ctx_xfer_done */
-	0xf807a421,
+/* 0x0b47: ctx_xfer_no_post_mmio */
+	0xf509eb21,
+/* 0x0b4b: ctx_xfer_done */
+	0xf807fc21,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvd7.fuc.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvd7.fuc.h
index 624215a005b0..62b0c7601d8b 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvd7.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvd7.fuc.h
@@ -528,10 +528,10 @@ uint32_t nvd7_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0xb521f502,
-	0xc721f507,
-	0x10f7f007,
-	0x081421f5,
+	0x0d21f502,
+	0x1f21f508,
+	0x10f7f008,
+	0x086c21f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t nvd7_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x081421f5,
+	0x086c21f5,
 	0xf500f7f0,
-	0xf107b521,
+	0xf1080d21,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,8 +610,8 @@ uint32_t nvd7_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0xe821f502,
-	0xf094bd09,
+	0x4021f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -621,7 +621,7 @@ uint32_t nvd7_grhub_code[] = {
 	0x0203f00f,
 	0xbd0009d0,
 	0x0131f404,
-	0x09e821f5,
+	0x0a4021f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t nvd7_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc09e821,
+	0xfc0a4021,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t nvd7_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x09e821f5,
+	0x0a4021f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t nvd7_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0x7821f502,
+	0xd021f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,8 +664,8 @@ uint32_t nvd7_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0xe821f502,
-	0xf094bd09,
+	0x4021f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -710,18 +710,40 @@ uint32_t nvd7_grhub_code[] = {
 /* 0x072b: ih_no_ctxsw */
 	0xe40421f4,
 	0xf40400ab,
-	0xb7f1140b,
+	0xe7f16c0b,
+	0xe3f00708,
+	0x6821f440,
+	0xf102ffb9,
+	0xf0040007,
+	0x0fd00203,
+	0xf104bd00,
+	0xf00704e7,
+	0x21f440e3,
+	0x02ffb968,
+	0x030007f1,
+	0xd00203f0,
+	0x04bd000f,
+	0x9450fec7,
+	0xf7f102ee,
+	0xf3f00700,
+	0x00efbb40,
+	0xf16821f4,
+	0xf0020007,
+	0x0fd00203,
+	0xf004bd00,
+	0x21f503f7,
+	0xb7f1037e,
 	0xbfb90100,
 	0x44e7f102,
 	0x40e3f001,
-/* 0x0743: ih_no_fwmthd */
+/* 0x079b: ih_no_fwmthd */
 	0xf19d21f4,
-	0xbd0104b7,
+	0xbd0504b7,
 	0xb4abffb0,
 	0xf10f0bf4,
 	0xf0070007,
 	0x0bd00303,
-/* 0x075b: ih_no_other */
+/* 0x07b3: ih_no_other */
 	0xf104bd00,
 	0xf0010007,
 	0x0ad00003,
@@ -731,36 +753,36 @@ uint32_t nvd7_grhub_code[] = {
 	0xfc90fca0,
 	0x0088fe80,
 	0x32f480fc,
-/* 0x077f: ctx_4160s */
+/* 0x07d7: ctx_4160s */
 	0xf001f800,
 	0xffb901f7,
 	0x60e7f102,
 	0x40e3f041,
-/* 0x078f: ctx_4160s_wait */
+/* 0x07e7: ctx_4160s_wait */
 	0xf19d21f4,
 	0xf04160e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf404ffc8,
 	0x00f8f00b,
-/* 0x07a4: ctx_4160c */
+/* 0x07fc: ctx_4160c */
 	0xffb9f4bd,
 	0x60e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x07b5: ctx_4170s */
+/* 0x080d: ctx_4170s */
 	0x10f5f000,
 	0xf102ffb9,
 	0xf04170e7,
 	0x21f440e3,
-/* 0x07c7: ctx_4170w */
+/* 0x081f: ctx_4170w */
 	0xf100f89d,
 	0xf04170e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf410f4f0,
 	0x00f8f01b,
-/* 0x07dc: ctx_redswitch */
+/* 0x0834: ctx_redswitch */
 	0x0200e7f1,
 	0xf040e5f0,
 	0xe5f020e5,
@@ -768,7 +790,7 @@ uint32_t nvd7_grhub_code[] = {
 	0x0103f085,
 	0xbd000ed0,
 	0x08f7f004,
-/* 0x07f8: ctx_redswitch_delay */
+/* 0x0850: ctx_redswitch_delay */
 	0xf401f2b6,
 	0xe5f1fd1b,
 	0xe5f10400,
@@ -776,7 +798,7 @@ uint32_t nvd7_grhub_code[] = {
 	0x03f08500,
 	0x000ed001,
 	0x00f804bd,
-/* 0x0814: ctx_86c */
+/* 0x086c: ctx_86c */
 	0x1b0007f1,
 	0xd00203f0,
 	0x04bd000f,
@@ -787,16 +809,16 @@ uint32_t nvd7_grhub_code[] = {
 	0xa86ce7f1,
 	0xf441e3f0,
 	0x00f89d21,
-/* 0x083c: ctx_mem */
+/* 0x0894: ctx_mem */
 	0x840007f1,
 	0xd00203f0,
 	0x04bd000f,
-/* 0x0848: ctx_mem_wait */
+/* 0x08a0: ctx_mem_wait */
 	0x8400f7f1,
 	0xcf02f3f0,
 	0xfffd00ff,
 	0xf31bf405,
-/* 0x085a: ctx_load */
+/* 0x08b2: ctx_load */
 	0x94bd00f8,
 	0xf10599f0,
 	0xf00f0007,
@@ -814,7 +836,7 @@ uint32_t nvd7_grhub_code[] = {
 	0x02d00203,
 	0xf004bd00,
 	0x21f507f7,
-	0x07f1083c,
+	0x07f10894,
 	0x03f0c000,
 	0x0002d002,
 	0x0bfe04bd,
@@ -869,31 +891,31 @@ uint32_t nvd7_grhub_code[] = {
 	0x03f01700,
 	0x0009d002,
 	0x00f804bd,
-/* 0x0978: ctx_chan */
-	0x077f21f5,
-	0x085a21f5,
+/* 0x09d0: ctx_chan */
+	0x07d721f5,
+	0x08b221f5,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x3c21f505,
-	0xa421f508,
-/* 0x0993: ctx_mmio_exec */
+	0x9421f505,
+	0xfc21f508,
+/* 0x09eb: ctx_mmio_exec */
 	0x9800f807,
 	0x07f14103,
 	0x03f08100,
 	0x0003d002,
 	0x34bd04bd,
-/* 0x09a4: ctx_mmio_loop */
+/* 0x09fc: ctx_mmio_loop */
 	0xf4ff34c4,
 	0x57f10f1b,
 	0x53f00200,
 	0x0535fa06,
-/* 0x09b6: ctx_mmio_pull */
+/* 0x0a0e: ctx_mmio_pull */
 	0x4e9803f8,
 	0x814f9880,
 	0xb69d21f4,
 	0x12b60830,
 	0xdf1bf401,
-/* 0x09c8: ctx_mmio_done */
+/* 0x0a20: ctx_mmio_done */
 	0xf1160398,
 	0xf0810007,
 	0x03d00203,
@@ -902,30 +924,30 @@ uint32_t nvd7_grhub_code[] = {
 	0x13f00100,
 	0x0601fa06,
 	0x00f803f8,
-/* 0x09e8: ctx_xfer */
+/* 0x0a40: ctx_xfer */
 	0xf104e7f0,
 	0xf0020007,
 	0x0ed00303,
-/* 0x09f7: ctx_xfer_idle */
+/* 0x0a4f: ctx_xfer_idle */
 	0xf104bd00,
 	0xf00000e7,
 	0xeecf03e3,
 	0x00e4f100,
 	0xf21bf420,
 	0xf40611f4,
-/* 0x0a0e: ctx_xfer_pre */
+/* 0x0a66: ctx_xfer_pre */
 	0xf7f01102,
-	0x1421f510,
-	0x7f21f508,
+	0x6c21f510,
+	0xd721f508,
 	0x1c11f407,
-/* 0x0a1c: ctx_xfer_pre_load */
+/* 0x0a74: ctx_xfer_pre_load */
 	0xf502f7f0,
-	0xf507b521,
-	0xf507c721,
-	0xbd07dc21,
-	0xb521f5f4,
-	0x5a21f507,
-/* 0x0a35: ctx_xfer_exec */
+	0xf5080d21,
+	0xf5081f21,
+	0xbd083421,
+	0x0d21f5f4,
+	0xb221f508,
+/* 0x0a8d: ctx_xfer_exec */
 	0x16019808,
 	0x07f124bd,
 	0x03f00500,
@@ -960,23 +982,65 @@ uint32_t nvd7_grhub_code[] = {
 	0x1301f402,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x3c21f505,
+	0x9421f505,
 	0x3202f408,
-/* 0x0ac4: ctx_xfer_post */
+/* 0x0b1c: ctx_xfer_post */
 	0xf502f7f0,
-	0xbd07b521,
-	0x1421f5f4,
+	0xbd080d21,
+	0x6c21f5f4,
 	0x7f21f508,
-	0xc721f502,
-	0xf5f4bd07,
-	0xf407b521,
+	0x1f21f502,
+	0xf5f4bd08,
+	0xf4080d21,
 	0x01981011,
 	0x0511fd40,
 	0xf5070bf4,
-/* 0x0aef: ctx_xfer_no_post_mmio */
-	0xf5099321,
-/* 0x0af3: ctx_xfer_done */
-	0xf807a421,
+/* 0x0b47: ctx_xfer_no_post_mmio */
+	0xf509eb21,
+/* 0x0b4b: ctx_xfer_done */
+	0xf807fc21,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnve0.fuc.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnve0.fuc.h
index 6547b3dfc7ed..51c3797d8537 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnve0.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnve0.fuc.h
@@ -528,10 +528,10 @@ uint32_t nve0_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0x7f21f502,
-	0x9121f507,
+	0xd721f502,
+	0xe921f507,
 	0x10f7f007,
-	0x07de21f5,
+	0x083621f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t nve0_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x07de21f5,
+	0x083621f5,
 	0xf500f7f0,
-	0xf1077f21,
+	0xf107d721,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,8 +610,8 @@ uint32_t nve0_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0xaa21f502,
-	0xf094bd09,
+	0x0221f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -621,7 +621,7 @@ uint32_t nve0_grhub_code[] = {
 	0x0203f00f,
 	0xbd0009d0,
 	0x0131f404,
-	0x09aa21f5,
+	0x0a0221f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t nve0_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc09aa21,
+	0xfc0a0221,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t nve0_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x09aa21f5,
+	0x0a0221f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t nve0_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0x4221f502,
+	0x9a21f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,8 +664,8 @@ uint32_t nve0_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0xaa21f502,
-	0xf094bd09,
+	0x0221f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -710,18 +710,40 @@ uint32_t nve0_grhub_code[] = {
 /* 0x072b: ih_no_ctxsw */
 	0xe40421f4,
 	0xf40400ab,
-	0xb7f1140b,
+	0xe7f16c0b,
+	0xe3f00708,
+	0x6821f440,
+	0xf102ffb9,
+	0xf0040007,
+	0x0fd00203,
+	0xf104bd00,
+	0xf00704e7,
+	0x21f440e3,
+	0x02ffb968,
+	0x030007f1,
+	0xd00203f0,
+	0x04bd000f,
+	0x9450fec7,
+	0xf7f102ee,
+	0xf3f00700,
+	0x00efbb40,
+	0xf16821f4,
+	0xf0020007,
+	0x0fd00203,
+	0xf004bd00,
+	0x21f503f7,
+	0xb7f1037e,
 	0xbfb90100,
 	0x44e7f102,
 	0x40e3f001,
-/* 0x0743: ih_no_fwmthd */
+/* 0x079b: ih_no_fwmthd */
 	0xf19d21f4,
-	0xbd0104b7,
+	0xbd0504b7,
 	0xb4abffb0,
 	0xf10f0bf4,
 	0xf0070007,
 	0x0bd00303,
-/* 0x075b: ih_no_other */
+/* 0x07b3: ih_no_other */
 	0xf104bd00,
 	0xf0010007,
 	0x0ad00003,
@@ -731,19 +753,19 @@ uint32_t nve0_grhub_code[] = {
 	0xfc90fca0,
 	0x0088fe80,
 	0x32f480fc,
-/* 0x077f: ctx_4170s */
+/* 0x07d7: ctx_4170s */
 	0xf001f800,
 	0xffb910f5,
 	0x70e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x0791: ctx_4170w */
+/* 0x07e9: ctx_4170w */
 	0x70e7f100,
 	0x40e3f041,
 	0xb96821f4,
 	0xf4f002ff,
 	0xf01bf410,
-/* 0x07a6: ctx_redswitch */
+/* 0x07fe: ctx_redswitch */
 	0xe7f100f8,
 	0xe5f00200,
 	0x20e5f040,
@@ -751,7 +773,7 @@ uint32_t nve0_grhub_code[] = {
 	0xf0850007,
 	0x0ed00103,
 	0xf004bd00,
-/* 0x07c2: ctx_redswitch_delay */
+/* 0x081a: ctx_redswitch_delay */
 	0xf2b608f7,
 	0xfd1bf401,
 	0x0400e5f1,
@@ -759,7 +781,7 @@ uint32_t nve0_grhub_code[] = {
 	0x850007f1,
 	0xd00103f0,
 	0x04bd000e,
-/* 0x07de: ctx_86c */
+/* 0x0836: ctx_86c */
 	0x07f100f8,
 	0x03f01b00,
 	0x000fd002,
@@ -770,17 +792,17 @@ uint32_t nve0_grhub_code[] = {
 	0xe7f102ff,
 	0xe3f0a86c,
 	0x9d21f441,
-/* 0x0806: ctx_mem */
+/* 0x085e: ctx_mem */
 	0x07f100f8,
 	0x03f08400,
 	0x000fd002,
-/* 0x0812: ctx_mem_wait */
+/* 0x086a: ctx_mem_wait */
 	0xf7f104bd,
 	0xf3f08400,
 	0x00ffcf02,
 	0xf405fffd,
 	0x00f8f31b,
-/* 0x0824: ctx_load */
+/* 0x087c: ctx_load */
 	0x99f094bd,
 	0x0007f105,
 	0x0203f00f,
@@ -797,7 +819,7 @@ uint32_t nve0_grhub_code[] = {
 	0x0203f083,
 	0xbd0002d0,
 	0x07f7f004,
-	0x080621f5,
+	0x085e21f5,
 	0xc00007f1,
 	0xd00203f0,
 	0x04bd0002,
@@ -852,29 +874,29 @@ uint32_t nve0_grhub_code[] = {
 	0x170007f1,
 	0xd00203f0,
 	0x04bd0009,
-/* 0x0942: ctx_chan */
+/* 0x099a: ctx_chan */
 	0x21f500f8,
-	0xa7f00824,
+	0xa7f0087c,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf8080621,
-/* 0x0955: ctx_mmio_exec */
+	0xf8085e21,
+/* 0x09ad: ctx_mmio_exec */
 	0x41039800,
 	0x810007f1,
 	0xd00203f0,
 	0x04bd0003,
-/* 0x0966: ctx_mmio_loop */
+/* 0x09be: ctx_mmio_loop */
 	0x34c434bd,
 	0x0f1bf4ff,
 	0x020057f1,
 	0xfa0653f0,
 	0x03f80535,
-/* 0x0978: ctx_mmio_pull */
+/* 0x09d0: ctx_mmio_pull */
 	0x98804e98,
 	0x21f4814f,
 	0x0830b69d,
 	0xf40112b6,
-/* 0x098a: ctx_mmio_done */
+/* 0x09e2: ctx_mmio_done */
 	0x0398df1b,
 	0x0007f116,
 	0x0203f081,
@@ -883,30 +905,30 @@ uint32_t nve0_grhub_code[] = {
 	0x010017f1,
 	0xfa0613f0,
 	0x03f80601,
-/* 0x09aa: ctx_xfer */
+/* 0x0a02: ctx_xfer */
 	0xe7f000f8,
 	0x0007f104,
 	0x0303f002,
 	0xbd000ed0,
-/* 0x09b9: ctx_xfer_idle */
+/* 0x0a11: ctx_xfer_idle */
 	0x00e7f104,
 	0x03e3f000,
 	0xf100eecf,
 	0xf42000e4,
 	0x11f4f21b,
 	0x0d02f406,
-/* 0x09d0: ctx_xfer_pre */
+/* 0x0a28: ctx_xfer_pre */
 	0xf510f7f0,
-	0xf407de21,
-/* 0x09da: ctx_xfer_pre_load */
+	0xf4083621,
+/* 0x0a32: ctx_xfer_pre_load */
 	0xf7f01c11,
-	0x7f21f502,
-	0x9121f507,
-	0xa621f507,
+	0xd721f502,
+	0xe921f507,
+	0xfe21f507,
 	0xf5f4bd07,
-	0xf5077f21,
-/* 0x09f3: ctx_xfer_exec */
-	0x98082421,
+	0xf507d721,
+/* 0x0a4b: ctx_xfer_exec */
+	0x98087c21,
 	0x24bd1601,
 	0x050007f1,
 	0xd00103f0,
@@ -941,21 +963,21 @@ uint32_t nve0_grhub_code[] = {
 	0xa7f01301,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf4080621,
-/* 0x0a82: ctx_xfer_post */
+	0xf4085e21,
+/* 0x0ada: ctx_xfer_post */
 	0xf7f02e02,
-	0x7f21f502,
+	0xd721f502,
 	0xf5f4bd07,
-	0xf507de21,
+	0xf5083621,
 	0xf5027f21,
-	0xbd079121,
-	0x7f21f5f4,
+	0xbd07e921,
+	0xd721f5f4,
 	0x1011f407,
 	0xfd400198,
 	0x0bf40511,
-	0x5521f507,
-/* 0x0aad: ctx_xfer_no_post_mmio */
-/* 0x0aad: ctx_xfer_done */
+	0xad21f507,
+/* 0x0b05: ctx_xfer_no_post_mmio */
+/* 0x0b05: ctx_xfer_done */
 	0x0000f809,
 	0x00000000,
 	0x00000000,
@@ -977,4 +999,46 @@ uint32_t nve0_grhub_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvf0.fuc.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvf0.fuc.h
index a5aee5a4302f..a0af4b703a8e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvf0.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/hubnvf0.fuc.h
@@ -528,10 +528,10 @@ uint32_t nvf0_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0x7f21f502,
-	0x9121f507,
+	0xd721f502,
+	0xe921f507,
 	0x10f7f007,
-	0x07de21f5,
+	0x083621f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t nvf0_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x07de21f5,
+	0x083621f5,
 	0xf500f7f0,
-	0xf1077f21,
+	0xf107d721,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,8 +610,8 @@ uint32_t nvf0_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0xaa21f502,
-	0xf094bd09,
+	0x0221f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -621,7 +621,7 @@ uint32_t nvf0_grhub_code[] = {
 	0x0203f037,
 	0xbd0009d0,
 	0x0131f404,
-	0x09aa21f5,
+	0x0a0221f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t nvf0_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc09aa21,
+	0xfc0a0221,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t nvf0_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x09aa21f5,
+	0x0a0221f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t nvf0_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0x4221f502,
+	0x9a21f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,8 +664,8 @@ uint32_t nvf0_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0xaa21f502,
-	0xf094bd09,
+	0x0221f502,
+	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
 	0x0009d002,
@@ -710,18 +710,40 @@ uint32_t nvf0_grhub_code[] = {
 /* 0x072b: ih_no_ctxsw */
 	0xe40421f4,
 	0xf40400ab,
-	0xb7f1140b,
+	0xe7f16c0b,
+	0xe3f00708,
+	0x6821f440,
+	0xf102ffb9,
+	0xf0040007,
+	0x0fd00203,
+	0xf104bd00,
+	0xf00704e7,
+	0x21f440e3,
+	0x02ffb968,
+	0x030007f1,
+	0xd00203f0,
+	0x04bd000f,
+	0x9450fec7,
+	0xf7f102ee,
+	0xf3f00700,
+	0x00efbb40,
+	0xf16821f4,
+	0xf0020007,
+	0x0fd00203,
+	0xf004bd00,
+	0x21f503f7,
+	0xb7f1037e,
 	0xbfb90100,
 	0x44e7f102,
 	0x40e3f001,
-/* 0x0743: ih_no_fwmthd */
+/* 0x079b: ih_no_fwmthd */
 	0xf19d21f4,
-	0xbd0104b7,
+	0xbd0504b7,
 	0xb4abffb0,
 	0xf10f0bf4,
 	0xf0070007,
 	0x0bd00303,
-/* 0x075b: ih_no_other */
+/* 0x07b3: ih_no_other */
 	0xf104bd00,
 	0xf0010007,
 	0x0ad00003,
@@ -731,19 +753,19 @@ uint32_t nvf0_grhub_code[] = {
 	0xfc90fca0,
 	0x0088fe80,
 	0x32f480fc,
-/* 0x077f: ctx_4170s */
+/* 0x07d7: ctx_4170s */
 	0xf001f800,
 	0xffb910f5,
 	0x70e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x0791: ctx_4170w */
+/* 0x07e9: ctx_4170w */
 	0x70e7f100,
 	0x40e3f041,
 	0xb96821f4,
 	0xf4f002ff,
 	0xf01bf410,
-/* 0x07a6: ctx_redswitch */
+/* 0x07fe: ctx_redswitch */
 	0xe7f100f8,
 	0xe5f00200,
 	0x20e5f040,
@@ -751,7 +773,7 @@ uint32_t nvf0_grhub_code[] = {
 	0xf0850007,
 	0x0ed00103,
 	0xf004bd00,
-/* 0x07c2: ctx_redswitch_delay */
+/* 0x081a: ctx_redswitch_delay */
 	0xf2b608f7,
 	0xfd1bf401,
 	0x0400e5f1,
@@ -759,7 +781,7 @@ uint32_t nvf0_grhub_code[] = {
 	0x850007f1,
 	0xd00103f0,
 	0x04bd000e,
-/* 0x07de: ctx_86c */
+/* 0x0836: ctx_86c */
 	0x07f100f8,
 	0x03f02300,
 	0x000fd002,
@@ -770,17 +792,17 @@ uint32_t nvf0_grhub_code[] = {
 	0xe7f102ff,
 	0xe3f0a88c,
 	0x9d21f441,
-/* 0x0806: ctx_mem */
+/* 0x085e: ctx_mem */
 	0x07f100f8,
 	0x03f08400,
 	0x000fd002,
-/* 0x0812: ctx_mem_wait */
+/* 0x086a: ctx_mem_wait */
 	0xf7f104bd,
 	0xf3f08400,
 	0x00ffcf02,
 	0xf405fffd,
 	0x00f8f31b,
-/* 0x0824: ctx_load */
+/* 0x087c: ctx_load */
 	0x99f094bd,
 	0x0007f105,
 	0x0203f037,
@@ -797,7 +819,7 @@ uint32_t nvf0_grhub_code[] = {
 	0x0203f083,
 	0xbd0002d0,
 	0x07f7f004,
-	0x080621f5,
+	0x085e21f5,
 	0xc00007f1,
 	0xd00203f0,
 	0x04bd0002,
@@ -852,29 +874,29 @@ uint32_t nvf0_grhub_code[] = {
 	0x170007f1,
 	0xd00203f0,
 	0x04bd0009,
-/* 0x0942: ctx_chan */
+/* 0x099a: ctx_chan */
 	0x21f500f8,
-	0xa7f00824,
+	0xa7f0087c,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf8080621,
-/* 0x0955: ctx_mmio_exec */
+	0xf8085e21,
+/* 0x09ad: ctx_mmio_exec */
 	0x41039800,
 	0x810007f1,
 	0xd00203f0,
 	0x04bd0003,
-/* 0x0966: ctx_mmio_loop */
+/* 0x09be: ctx_mmio_loop */
 	0x34c434bd,
 	0x0f1bf4ff,
 	0x020057f1,
 	0xfa0653f0,
 	0x03f80535,
-/* 0x0978: ctx_mmio_pull */
+/* 0x09d0: ctx_mmio_pull */
 	0x98804e98,
 	0x21f4814f,
 	0x0830b69d,
 	0xf40112b6,
-/* 0x098a: ctx_mmio_done */
+/* 0x09e2: ctx_mmio_done */
 	0x0398df1b,
 	0x0007f116,
 	0x0203f081,
@@ -883,30 +905,30 @@ uint32_t nvf0_grhub_code[] = {
 	0x010017f1,
 	0xfa0613f0,
 	0x03f80601,
-/* 0x09aa: ctx_xfer */
+/* 0x0a02: ctx_xfer */
 	0xe7f000f8,
 	0x0007f104,
 	0x0303f002,
 	0xbd000ed0,
-/* 0x09b9: ctx_xfer_idle */
+/* 0x0a11: ctx_xfer_idle */
 	0x00e7f104,
 	0x03e3f000,
 	0xf100eecf,
 	0xf42000e4,
 	0x11f4f21b,
 	0x0d02f406,
-/* 0x09d0: ctx_xfer_pre */
+/* 0x0a28: ctx_xfer_pre */
 	0xf510f7f0,
-	0xf407de21,
-/* 0x09da: ctx_xfer_pre_load */
+	0xf4083621,
+/* 0x0a32: ctx_xfer_pre_load */
 	0xf7f01c11,
-	0x7f21f502,
-	0x9121f507,
-	0xa621f507,
+	0xd721f502,
+	0xe921f507,
+	0xfe21f507,
 	0xf5f4bd07,
-	0xf5077f21,
-/* 0x09f3: ctx_xfer_exec */
-	0x98082421,
+	0xf507d721,
+/* 0x0a4b: ctx_xfer_exec */
+	0x98087c21,
 	0x24bd1601,
 	0x050007f1,
 	0xd00103f0,
@@ -941,21 +963,21 @@ uint32_t nvf0_grhub_code[] = {
 	0xa7f01301,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf4080621,
-/* 0x0a82: ctx_xfer_post */
+	0xf4085e21,
+/* 0x0ada: ctx_xfer_post */
 	0xf7f02e02,
-	0x7f21f502,
+	0xd721f502,
 	0xf5f4bd07,
-	0xf507de21,
+	0xf5083621,
 	0xf5027f21,
-	0xbd079121,
-	0x7f21f5f4,
+	0xbd07e921,
+	0xd721f5f4,
 	0x1011f407,
 	0xfd400198,
 	0x0bf40511,
-	0x5521f507,
-/* 0x0aad: ctx_xfer_no_post_mmio */
-/* 0x0aad: ctx_xfer_done */
+	0xad21f507,
+/* 0x0b05: ctx_xfer_no_post_mmio */
+/* 0x0b05: ctx_xfer_done */
 	0x0000f809,
 	0x00000000,
 	0x00000000,
@@ -977,4 +999,46 @@ uint32_t nvf0_grhub_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/macros.fuc b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/macros.fuc
index a47d49db5232..2a0b0f844299 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/macros.fuc
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/macros.fuc
@@ -30,6 +30,12 @@
 #define GK110 0xf0
 #define GK208 0x108
 
+#define NV_PGRAPH_TRAPPED_ADDR                                         0x400704
+#define NV_PGRAPH_TRAPPED_DATA_LO                                      0x400708
+#define NV_PGRAPH_TRAPPED_DATA_HI                                      0x40070c
+
+#define NV_PGRAPH_FE_OBJECT_TABLE(n)                        ((n) * 4 + 0x400700)
+
 #define NV_PGRAPH_FECS_INTR_ACK                                        0x409004
 #define NV_PGRAPH_FECS_INTR                                            0x409008
 #define NV_PGRAPH_FECS_INTR_FWMTHD                                   0x00000400
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/os.h b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/os.h
index fd1d380de094..1718ae4e8224 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/fuc/os.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/fuc/os.h
@@ -3,5 +3,6 @@
 
 #define E_BAD_COMMAND  0x00000001
 #define E_CMD_OVERFLOW 0x00000002
+#define E_BAD_FWMTHD   0x00000003
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/gk20a.c b/drivers/gpu/drm/nouveau/core/engine/graph/gk20a.c
new file mode 100644
index 000000000000..83048a56430d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/gk20a.c
@@ -0,0 +1,47 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "nvc0.h"
+#include "ctxnvc0.h"
+
+static struct nouveau_oclass
+gk20a_graph_sclass[] = {
+	{ 0x902d, &nouveau_object_ofuncs },
+	{ 0xa040, &nouveau_object_ofuncs },
+	{ 0xa297, &nouveau_object_ofuncs },
+	{ 0xa0c0, &nouveau_object_ofuncs },
+	{}
+};
+
+struct nouveau_oclass *
+gk20a_graph_oclass = &(struct nvc0_graph_oclass) {
+	.base.handle = NV_ENGINE(GR, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_graph_ctor,
+		.dtor = nvc0_graph_dtor,
+		.init = nve4_graph_init,
+		.fini = nve4_graph_fini,
+	},
+	.cclass = &gk20a_grctx_oclass,
+	.sclass = gk20a_graph_sclass,
+	.mmio = nve4_graph_pack_mmio,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
index 2c7809e1a09b..20665c21d80e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
@@ -901,7 +901,7 @@ nv50_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		nv_engine(priv)->sclass = nvaf_graph_sclass;
 		break;
 
-	};
+	}
 
 	/* unfortunate hw bug workaround... */
 	if (nv_device(priv)->chipset != 0x50 &&
@@ -976,7 +976,6 @@ nv50_graph_init(struct nouveau_object *object)
 		break;
 	case 0xa0:
 	default:
-		nv_wr32(priv, 0x402cc0, 0x00000000);
 		if (nv_device(priv)->chipset == 0xa0 ||
 		    nv_device(priv)->chipset == 0xaa ||
 		    nv_device(priv)->chipset == 0xac) {
@@ -991,10 +990,10 @@ nv50_graph_init(struct nouveau_object *object)
 
 	/* zero out zcull regions */
 	for (i = 0; i < 8; i++) {
-		nv_wr32(priv, 0x402c20 + (i * 8), 0x00000000);
-		nv_wr32(priv, 0x402c24 + (i * 8), 0x00000000);
-		nv_wr32(priv, 0x402c28 + (i * 8), 0x00000000);
-		nv_wr32(priv, 0x402c2c + (i * 8), 0x00000000);
+		nv_wr32(priv, 0x402c20 + (i * 0x10), 0x00000000);
+		nv_wr32(priv, 0x402c24 + (i * 0x10), 0x00000000);
+		nv_wr32(priv, 0x402c28 + (i * 0x10), 0x00000000);
+		nv_wr32(priv, 0x402c2c + (i * 0x10), 0x00000000);
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
index f3c7329da0a0..aa0838916354 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
@@ -789,17 +789,40 @@ nvc0_graph_ctxctl_debug(struct nvc0_graph_priv *priv)
 static void
 nvc0_graph_ctxctl_isr(struct nvc0_graph_priv *priv)
 {
-	u32 ustat = nv_rd32(priv, 0x409c18);
+	u32 stat = nv_rd32(priv, 0x409c18);
 
-	if (ustat & 0x00000001)
-		nv_error(priv, "CTXCTL ucode error\n");
-	if (ustat & 0x00080000)
-		nv_error(priv, "CTXCTL watchdog timeout\n");
-	if (ustat & ~0x00080001)
-		nv_error(priv, "CTXCTL 0x%08x\n", ustat);
+	if (stat & 0x00000001) {
+		u32 code = nv_rd32(priv, 0x409814);
+		if (code == E_BAD_FWMTHD) {
+			u32 class = nv_rd32(priv, 0x409808);
+			u32  addr = nv_rd32(priv, 0x40980c);
+			u32  subc = (addr & 0x00070000) >> 16;
+			u32  mthd = (addr & 0x00003ffc);
+			u32  data = nv_rd32(priv, 0x409810);
+
+			nv_error(priv, "FECS MTHD subc %d class 0x%04x "
+				       "mthd 0x%04x data 0x%08x\n",
+				 subc, class, mthd, data);
+
+			nv_wr32(priv, 0x409c20, 0x00000001);
+			stat &= ~0x00000001;
+		} else {
+			nv_error(priv, "FECS ucode error %d\n", code);
+		}
+	}
 
-	nvc0_graph_ctxctl_debug(priv);
-	nv_wr32(priv, 0x409c20, ustat);
+	if (stat & 0x00080000) {
+		nv_error(priv, "FECS watchdog timeout\n");
+		nvc0_graph_ctxctl_debug(priv);
+		nv_wr32(priv, 0x409c20, 0x00080000);
+		stat &= ~0x00080000;
+	}
+
+	if (stat) {
+		nv_error(priv, "FECS 0x%08x\n", stat);
+		nvc0_graph_ctxctl_debug(priv);
+		nv_wr32(priv, 0x409c20, stat);
+	}
 }
 
 static void
@@ -894,6 +917,10 @@ nvc0_graph_init_fw(struct nvc0_graph_priv *priv, u32 fuc_base,
 			nv_wr32(priv, fuc_base + 0x0188, i >> 6);
 		nv_wr32(priv, fuc_base + 0x0184, code->data[i]);
 	}
+
+	/* code must be padded to 0x40 words */
+	for (; i & 0x3f; i++)
+		nv_wr32(priv, fuc_base + 0x0184, 0);
 }
 
 static void
@@ -1259,10 +1286,14 @@ nvc0_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvc0_graph_oclass *oclass = (void *)bclass;
 	struct nouveau_device *device = nv_device(parent);
 	struct nvc0_graph_priv *priv;
+	bool use_ext_fw, enable;
 	int ret, i;
 
-	ret = nouveau_graph_create(parent, engine, bclass,
-				   (oclass->fecs.ucode != NULL), &priv);
+	use_ext_fw = nouveau_boolopt(device->cfgopt, "NvGrUseFW",
+				     oclass->fecs.ucode == NULL);
+	enable = use_ext_fw || oclass->fecs.ucode != NULL;
+
+	ret = nouveau_graph_create(parent, engine, bclass, enable, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -1272,7 +1303,7 @@ nvc0_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 
 	priv->base.units = nvc0_graph_units;
 
-	if (nouveau_boolopt(device->cfgopt, "NvGrUseFW", false)) {
+	if (use_ext_fw) {
 		nv_info(priv, "using external firmware\n");
 		if (nvc0_graph_ctor_fw(priv, "fuc409c", &priv->fuc409c) ||
 		    nvc0_graph_ctor_fw(priv, "fuc409d", &priv->fuc409d) ||
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h
index 90d44616c876..ffc289198dd8 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h
@@ -38,6 +38,8 @@
 #include <engine/fifo.h>
 #include <engine/graph.h>
 
+#include "fuc/os.h"
+
 #define GPC_MAX 32
 #define TPC_MAX (GPC_MAX * 8)
 
@@ -116,6 +118,7 @@ int  nvc0_graph_ctor(struct nouveau_object *, struct nouveau_object *,
 		     struct nouveau_object **);
 void nvc0_graph_dtor(struct nouveau_object *);
 int  nvc0_graph_init(struct nouveau_object *);
+int  nve4_graph_fini(struct nouveau_object *, bool);
 int  nve4_graph_init(struct nouveau_object *);
 
 extern struct nouveau_oclass nvc0_graph_sclass[];
@@ -217,6 +220,7 @@ extern const struct nvc0_graph_init nve4_graph_init_main_0[];
 extern const struct nvc0_graph_init nve4_graph_init_tpccs_0[];
 extern const struct nvc0_graph_init nve4_graph_init_pe_0[];
 extern const struct nvc0_graph_init nve4_graph_init_be_0[];
+extern const struct nvc0_graph_pack nve4_graph_pack_mmio[];
 
 extern const struct nvc0_graph_init nvf0_graph_init_fe_0[];
 extern const struct nvc0_graph_init nvf0_graph_init_sked_0[];
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c b/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c
index f7c011217175..51e0c075ad34 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c
@@ -151,7 +151,7 @@ nve4_graph_init_be_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_graph_pack_mmio[] = {
 	{ nve4_graph_init_main_0 },
 	{ nvc0_graph_init_fe_0 },
@@ -189,7 +189,7 @@ nve4_graph_pack_mmio[] = {
  * PGRAPH engine/subdev functions
  ******************************************************************************/
 
-static int
+int
 nve4_graph_fini(struct nouveau_object *object, bool suspend)
 {
 	struct nvc0_graph_priv *priv = (void *)object;
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
index 5ce686ee729e..f3b4d9dbf23c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
@@ -124,7 +124,7 @@ nv50_software_sclass[] = {
  ******************************************************************************/
 
 static int
-nv50_software_vblsem_release(void *data, int head)
+nv50_software_vblsem_release(void *data, u32 type, int head)
 {
 	struct nv50_software_chan *chan = data;
 	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
@@ -183,7 +183,7 @@ nv50_software_context_ctor(struct nouveau_object *parent,
 		return -ENOMEM;
 
 	for (i = 0; i < chan->vblank.nr_event; i++) {
-		ret = nouveau_event_new(pdisp->vblank, i, pclass->vblank,
+		ret = nouveau_event_new(pdisp->vblank, 1, i, pclass->vblank,
 					chan, &chan->vblank.event[i]);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.h b/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
index 2de370c21279..bb49a7a20857 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
@@ -19,7 +19,7 @@ int  nv50_software_ctor(struct nouveau_object *, struct nouveau_object *,
 
 struct nv50_software_cclass {
 	struct nouveau_oclass base;
-	int (*vblank)(void *, int);
+	int (*vblank)(void *, u32, int);
 };
 
 struct nv50_software_chan {
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
index f9430c1bf3e5..135c20f38356 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
@@ -104,7 +104,7 @@ nvc0_software_sclass[] = {
  ******************************************************************************/
 
 static int
-nvc0_software_vblsem_release(void *data, int head)
+nvc0_software_vblsem_release(void *data, u32 type, int head)
 {
 	struct nv50_software_chan *chan = data;
 	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
diff --git a/drivers/gpu/drm/nouveau/core/include/core/class.h b/drivers/gpu/drm/nouveau/core/include/core/class.h
index 9c0cd73462d9..e0c812bc884f 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/class.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/class.h
@@ -295,6 +295,10 @@ struct nv04_display_scanoutpos {
 #define NV84_DISP_SOR_HDMI_PWR_REKEY                                 0x0000007f
 #define NV50_DISP_SOR_LVDS_SCRIPT                                    0x00013000
 #define NV50_DISP_SOR_LVDS_SCRIPT_ID                                 0x0000ffff
+#define NV94_DISP_SOR_DP_PWR                                         0x00016000
+#define NV94_DISP_SOR_DP_PWR_STATE                                   0x00000001
+#define NV94_DISP_SOR_DP_PWR_STATE_OFF                               0x00000000
+#define NV94_DISP_SOR_DP_PWR_STATE_ON                                0x00000001
 
 #define NV50_DISP_DAC_MTHD                                           0x00020000
 #define NV50_DISP_DAC_MTHD_TYPE                                      0x0000f000
diff --git a/drivers/gpu/drm/nouveau/core/include/core/event.h b/drivers/gpu/drm/nouveau/core/include/core/event.h
index 5d539ebff3ed..ba3f1a76a815 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/event.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/event.h
@@ -12,32 +12,33 @@ struct nouveau_eventh {
 	struct nouveau_event *event;
 	struct list_head head;
 	unsigned long flags;
+	u32 types;
 	int index;
-	int (*func)(void *, int);
+	int (*func)(void *, u32, int);
 	void *priv;
 };
 
 struct nouveau_event {
-	spinlock_t list_lock;
-	spinlock_t refs_lock;
-
 	void *priv;
-	void (*enable)(struct nouveau_event *, int index);
-	void (*disable)(struct nouveau_event *, int index);
+	int (*check)(struct nouveau_event *, u32 type, int index);
+	void (*enable)(struct nouveau_event *, int type, int index);
+	void (*disable)(struct nouveau_event *, int type, int index);
 
+	int types_nr;
 	int index_nr;
-	struct {
-		struct list_head list;
-		int refs;
-	} index[];
+
+	spinlock_t list_lock;
+	struct list_head *list;
+	spinlock_t refs_lock;
+	int refs[];
 };
 
-int  nouveau_event_create(int index_nr, struct nouveau_event **);
+int  nouveau_event_create(int types_nr, int index_nr, struct nouveau_event **);
 void nouveau_event_destroy(struct nouveau_event **);
-void nouveau_event_trigger(struct nouveau_event *, int index);
+void nouveau_event_trigger(struct nouveau_event *, u32 types, int index);
 
-int  nouveau_event_new(struct nouveau_event *, int index,
-		       int (*func)(void *, int), void *,
+int  nouveau_event_new(struct nouveau_event *, u32 types, int index,
+		       int (*func)(void *, u32, int), void *,
 		       struct nouveau_eventh **);
 void nouveau_event_ref(struct nouveau_eventh *, struct nouveau_eventh **);
 void nouveau_event_get(struct nouveau_eventh *);
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/disp.h b/drivers/gpu/drm/nouveau/core/include/engine/disp.h
index fd0c68804de3..fde842896806 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/disp.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/disp.h
@@ -6,8 +6,19 @@
 #include <core/device.h>
 #include <core/event.h>
 
+enum nvkm_hpd_event {
+	NVKM_HPD_PLUG = 1,
+	NVKM_HPD_UNPLUG = 2,
+	NVKM_HPD_IRQ = 4,
+	NVKM_HPD = (NVKM_HPD_PLUG | NVKM_HPD_UNPLUG | NVKM_HPD_IRQ)
+};
+
 struct nouveau_disp {
 	struct nouveau_engine base;
+
+	struct list_head outp;
+	struct nouveau_event *hpd;
+
 	struct nouveau_event *vblank;
 };
 
@@ -17,25 +28,6 @@ nouveau_disp(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_ENGINE_DISP];
 }
 
-#define nouveau_disp_create(p,e,c,h,i,x,d)                                     \
-	nouveau_disp_create_((p), (e), (c), (h), (i), (x),                     \
-			     sizeof(**d), (void **)d)
-#define nouveau_disp_destroy(d) ({                                             \
-	struct nouveau_disp *disp = (d);                                       \
-	_nouveau_disp_dtor(nv_object(disp));                                   \
-})
-#define nouveau_disp_init(d)                                                   \
-	nouveau_engine_init(&(d)->base)
-#define nouveau_disp_fini(d,s)                                                 \
-	nouveau_engine_fini(&(d)->base, (s))
-
-int  nouveau_disp_create_(struct nouveau_object *, struct nouveau_object *,
-			  struct nouveau_oclass *, int heads,
-			  const char *, const char *, int, void **);
-void _nouveau_disp_dtor(struct nouveau_object *);
-#define _nouveau_disp_init _nouveau_engine_init
-#define _nouveau_disp_fini _nouveau_engine_fini
-
 extern struct nouveau_oclass *nv04_disp_oclass;
 extern struct nouveau_oclass *nv50_disp_oclass;
 extern struct nouveau_oclass *nv84_disp_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
index 26b6b2bb1112..b639eb2c74ff 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
@@ -109,6 +109,7 @@ extern struct nouveau_oclass *nv50_fifo_oclass;
 extern struct nouveau_oclass *nv84_fifo_oclass;
 extern struct nouveau_oclass *nvc0_fifo_oclass;
 extern struct nouveau_oclass *nve0_fifo_oclass;
+extern struct nouveau_oclass *gk20a_fifo_oclass;
 extern struct nouveau_oclass *nv108_fifo_oclass;
 
 void nv04_fifo_intr(struct nouveau_subdev *);
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/graph.h b/drivers/gpu/drm/nouveau/core/include/engine/graph.h
index 871edfdf3d5b..8c1d4772da0c 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/graph.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/graph.h
@@ -68,6 +68,7 @@ extern struct nouveau_oclass *nvc8_graph_oclass;
 extern struct nouveau_oclass *nvd7_graph_oclass;
 extern struct nouveau_oclass *nvd9_graph_oclass;
 extern struct nouveau_oclass *nve4_graph_oclass;
+extern struct nouveau_oclass *gk20a_graph_oclass;
 extern struct nouveau_oclass *nvf0_graph_oclass;
 extern struct nouveau_oclass *nv108_graph_oclass;
 extern struct nouveau_oclass *gm107_graph_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h
index a32feb3f3fb6..f3930c27cb7a 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h
@@ -22,7 +22,25 @@ enum dcb_connector_type {
 	DCB_CONNECTOR_NONE = 0xff
 };
 
-u16 dcb_conntab(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
-u16 dcb_conn(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len);
+struct nvbios_connT {
+};
+
+u32 nvbios_connTe(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u32 nvbios_connTp(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+		  struct nvbios_connT *info);
+
+struct nvbios_connE {
+	u8 type;
+	u8 location;
+	u8 hpd;
+	u8 dp;
+	u8 di;
+	u8 sr;
+	u8 lcdid;
+};
+
+u32 nvbios_connEe(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *hdr);
+u32 nvbios_connEp(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *hdr,
+		  struct nvbios_connE *info);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h
index 6e54218b55fc..728206e21777 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h
@@ -17,9 +17,10 @@ u16 nvbios_dpout_match(struct nouveau_bios *, u16 type, u16 mask,
 		       struct nvbios_dpout *);
 
 struct nvbios_dpcfg {
-	u8 drv;
-	u8 pre;
-	u8 unk;
+	u8 pc;
+	u8 dc;
+	u8 pe;
+	u8 tx_pu;
 };
 
 u16
@@ -27,7 +28,7 @@ nvbios_dpcfg_parse(struct nouveau_bios *, u16 outp, u8 idx,
 		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		   struct nvbios_dpcfg *);
 u16
-nvbios_dpcfg_match(struct nouveau_bios *, u16 outp, u8 un, u8 vs, u8 pe,
+nvbios_dpcfg_match(struct nouveau_bios *, u16 outp, u8 pc, u8 vs, u8 pe,
 		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		   struct nvbios_dpcfg *);
 
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
index 8f4ced75444a..c01e29c9f89a 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
@@ -77,6 +77,8 @@ struct nouveau_clock {
 	int tstate; /* thermal adjustment (max-) */
 	int dstate; /* display adjustment (min+) */
 
+	bool allow_reclock;
+
 	int  (*read)(struct nouveau_clock *, enum nv_clk_src);
 	int  (*calc)(struct nouveau_clock *, struct nouveau_cstate *);
 	int  (*prog)(struct nouveau_clock *);
@@ -106,8 +108,8 @@ struct nouveau_clocks {
 	int mdiv;
 };
 
-#define nouveau_clock_create(p,e,o,i,d)                                        \
-	nouveau_clock_create_((p), (e), (o), (i), sizeof(**d), (void **)d)
+#define nouveau_clock_create(p,e,o,i,r,d)                                      \
+	nouveau_clock_create_((p), (e), (o), (i), (r), sizeof(**d), (void **)d)
 #define nouveau_clock_destroy(p) ({                                            \
 	struct nouveau_clock *clk = (p);                                       \
 	_nouveau_clock_dtor(nv_object(clk));                                   \
@@ -121,7 +123,7 @@ struct nouveau_clocks {
 
 int  nouveau_clock_create_(struct nouveau_object *, struct nouveau_object *,
 			   struct nouveau_oclass *,
-			   struct nouveau_clocks *, int, void **);
+			   struct nouveau_clocks *, bool, int, void **);
 void _nouveau_clock_dtor(struct nouveau_object *);
 int _nouveau_clock_init(struct nouveau_object *);
 #define _nouveau_clock_fini _nouveau_subdev_fini
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/fb.h b/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
index 58c7ccdebb01..871e73914b24 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
@@ -105,6 +105,7 @@ extern struct nouveau_oclass *nvaa_fb_oclass;
 extern struct nouveau_oclass *nvaf_fb_oclass;
 extern struct nouveau_oclass *nvc0_fb_oclass;
 extern struct nouveau_oclass *nve0_fb_oclass;
+extern struct nouveau_oclass *gk20a_fb_oclass;
 extern struct nouveau_oclass *gm107_fb_oclass;
 
 #include <subdev/bios/ramcfg.h>
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h b/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
index c85b9f1579ad..612d82ab683d 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
@@ -8,17 +8,18 @@
 #include <subdev/bios.h>
 #include <subdev/bios/gpio.h>
 
+enum nvkm_gpio_event {
+	NVKM_GPIO_HI = 1,
+	NVKM_GPIO_LO = 2,
+	NVKM_GPIO_TOGGLED = (NVKM_GPIO_HI | NVKM_GPIO_LO),
+};
+
 struct nouveau_gpio {
 	struct nouveau_subdev base;
 
 	struct nouveau_event *events;
 
-	/* hardware interfaces */
 	void (*reset)(struct nouveau_gpio *, u8 func);
-	int  (*drive)(struct nouveau_gpio *, int line, int dir, int out);
-	int  (*sense)(struct nouveau_gpio *, int line);
-
-	/* software interfaces */
 	int  (*find)(struct nouveau_gpio *, int idx, u8 tag, u8 line,
 		     struct dcb_gpio_func *);
 	int  (*set)(struct nouveau_gpio *, int idx, u8 tag, u8 line, int state);
@@ -31,23 +32,10 @@ nouveau_gpio(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_GPIO];
 }
 
-#define nouveau_gpio_create(p,e,o,l,d)                                         \
-	nouveau_gpio_create_((p), (e), (o), (l), sizeof(**d), (void **)d)
-#define nouveau_gpio_destroy(p) ({                                             \
-	struct nouveau_gpio *gpio = (p);                                       \
-	_nouveau_gpio_dtor(nv_object(gpio));                                   \
-})
-#define nouveau_gpio_fini(p,s)                                                 \
-	nouveau_subdev_fini(&(p)->base, (s))
-
-int  nouveau_gpio_create_(struct nouveau_object *, struct nouveau_object *,
-			  struct nouveau_oclass *, int, int, void **);
-void _nouveau_gpio_dtor(struct nouveau_object *);
-int  nouveau_gpio_init(struct nouveau_gpio *);
-
-extern struct nouveau_oclass nv10_gpio_oclass;
-extern struct nouveau_oclass nv50_gpio_oclass;
-extern struct nouveau_oclass nvd0_gpio_oclass;
-extern struct nouveau_oclass nve0_gpio_oclass;
+extern struct nouveau_oclass *nv10_gpio_oclass;
+extern struct nouveau_oclass *nv50_gpio_oclass;
+extern struct nouveau_oclass *nv92_gpio_oclass;
+extern struct nouveau_oclass *nvd0_gpio_oclass;
+extern struct nouveau_oclass *nve0_gpio_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
index 7f50a858b16f..825f7bb46b67 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
@@ -14,52 +14,41 @@
 #define NV_I2C_TYPE_EXTDDC(e) (0x0005 | (e) << 8)
 #define NV_I2C_TYPE_EXTAUX(e) (0x0006 | (e) << 8)
 
+enum nvkm_i2c_event {
+	NVKM_I2C_PLUG = 1,
+	NVKM_I2C_UNPLUG = 2,
+	NVKM_I2C_IRQ = 4,
+	NVKM_I2C_DONE = 8,
+	NVKM_I2C_ANY = (NVKM_I2C_PLUG |
+			NVKM_I2C_UNPLUG |
+			NVKM_I2C_IRQ |
+			NVKM_I2C_DONE),
+};
+
 struct nouveau_i2c_port {
 	struct nouveau_object base;
 	struct i2c_adapter adapter;
+	struct mutex mutex;
 
 	struct list_head head;
 	u8  index;
+	int aux;
 
 	const struct nouveau_i2c_func *func;
 };
 
 struct nouveau_i2c_func {
-	void (*acquire)(struct nouveau_i2c_port *);
-	void (*release)(struct nouveau_i2c_port *);
-
 	void (*drive_scl)(struct nouveau_i2c_port *, int);
 	void (*drive_sda)(struct nouveau_i2c_port *, int);
 	int  (*sense_scl)(struct nouveau_i2c_port *);
 	int  (*sense_sda)(struct nouveau_i2c_port *);
 
-	int  (*aux)(struct nouveau_i2c_port *, u8, u32, u8 *, u8);
+	int  (*aux)(struct nouveau_i2c_port *, bool, u8, u32, u8 *, u8);
 	int  (*pattern)(struct nouveau_i2c_port *, int pattern);
 	int  (*lnk_ctl)(struct nouveau_i2c_port *, int nr, int bw, bool enh);
 	int  (*drv_ctl)(struct nouveau_i2c_port *, int lane, int sw, int pe);
 };
 
-#define nouveau_i2c_port_create(p,e,o,i,a,f,d)                                 \
-	nouveau_i2c_port_create_((p), (e), (o), (i), (a), (f),                 \
-				 sizeof(**d), (void **)d)
-#define nouveau_i2c_port_destroy(p) ({                                         \
-	struct nouveau_i2c_port *port = (p);                                   \
-	_nouveau_i2c_port_dtor(nv_object(i2c));                                \
-})
-#define nouveau_i2c_port_init(p)                                               \
-	nouveau_object_init(&(p)->base)
-#define nouveau_i2c_port_fini(p,s)                                             \
-	nouveau_object_fini(&(p)->base, (s))
-
-int nouveau_i2c_port_create_(struct nouveau_object *, struct nouveau_object *,
-			     struct nouveau_oclass *, u8,
-			     const struct i2c_algorithm *,
-			     const struct nouveau_i2c_func *,
-			     int, void **);
-void _nouveau_i2c_port_dtor(struct nouveau_object *);
-#define _nouveau_i2c_port_init nouveau_object_init
-#define _nouveau_i2c_port_fini nouveau_object_fini
-
 struct nouveau_i2c_board_info {
 	struct i2c_board_info dev;
 	u8 udelay; /* set to 0 to use the standard delay */
@@ -67,13 +56,20 @@ struct nouveau_i2c_board_info {
 
 struct nouveau_i2c {
 	struct nouveau_subdev base;
+	struct nouveau_event *ntfy;
 
 	struct nouveau_i2c_port *(*find)(struct nouveau_i2c *, u8 index);
 	struct nouveau_i2c_port *(*find_type)(struct nouveau_i2c *, u16 type);
+	int  (*acquire_pad)(struct nouveau_i2c_port *, unsigned long timeout);
+	void (*release_pad)(struct nouveau_i2c_port *);
+	int  (*acquire)(struct nouveau_i2c_port *, unsigned long timeout);
+	void (*release)(struct nouveau_i2c_port *);
 	int (*identify)(struct nouveau_i2c *, int index,
 			const char *what, struct nouveau_i2c_board_info *,
 			bool (*match)(struct nouveau_i2c_port *,
 				      struct i2c_board_info *, void *), void *);
+
+	wait_queue_head_t wait;
 	struct list_head ports;
 };
 
@@ -83,37 +79,13 @@ nouveau_i2c(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_I2C];
 }
 
-#define nouveau_i2c_create(p,e,o,s,d)                                          \
-	nouveau_i2c_create_((p), (e), (o), (s), sizeof(**d), (void **)d)
-#define nouveau_i2c_destroy(p) ({                                              \
-	struct nouveau_i2c *i2c = (p);                                         \
-	_nouveau_i2c_dtor(nv_object(i2c));                                     \
-})
-#define nouveau_i2c_init(p) ({                                                 \
-	struct nouveau_i2c *i2c = (p);                                         \
-	_nouveau_i2c_init(nv_object(i2c));                                     \
-})
-#define nouveau_i2c_fini(p,s) ({                                               \
-	struct nouveau_i2c *i2c = (p);                                         \
-	_nouveau_i2c_fini(nv_object(i2c), (s));                                \
-})
-
-int nouveau_i2c_create_(struct nouveau_object *, struct nouveau_object *,
-			struct nouveau_oclass *, struct nouveau_oclass *,
-			int, void **);
-void _nouveau_i2c_dtor(struct nouveau_object *);
-int  _nouveau_i2c_init(struct nouveau_object *);
-int  _nouveau_i2c_fini(struct nouveau_object *, bool);
-
-extern struct nouveau_oclass nv04_i2c_oclass;
-extern struct nouveau_oclass nv4e_i2c_oclass;
-extern struct nouveau_oclass nv50_i2c_oclass;
-extern struct nouveau_oclass nv94_i2c_oclass;
-extern struct nouveau_oclass nvd0_i2c_oclass;
-extern struct nouveau_oclass nouveau_anx9805_sclass[];
-
-extern const struct i2c_algorithm nouveau_i2c_bit_algo;
-extern const struct i2c_algorithm nouveau_i2c_aux_algo;
+extern struct nouveau_oclass *nv04_i2c_oclass;
+extern struct nouveau_oclass *nv4e_i2c_oclass;
+extern struct nouveau_oclass *nv50_i2c_oclass;
+extern struct nouveau_oclass *nv94_i2c_oclass;
+extern struct nouveau_oclass *nvd0_i2c_oclass;
+extern struct nouveau_oclass *gf117_i2c_oclass;
+extern struct nouveau_oclass *nve0_i2c_oclass;
 
 static inline int
 nv_rdi2cr(struct nouveau_i2c_port *port, u8 addr, u8 reg)
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h b/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h
index 88814f159d89..31df634c0fdc 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h
@@ -30,5 +30,6 @@ nouveau_ibus(void *obj)
 
 extern struct nouveau_oclass nvc0_ibus_oclass;
 extern struct nouveau_oclass nve0_ibus_oclass;
+extern struct nouveau_oclass gk20a_ibus_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bar/base.c b/drivers/gpu/drm/nouveau/core/subdev/bar/base.c
index bdf594116f3f..73b1ed20c8d5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bar/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bar/base.c
@@ -118,8 +118,10 @@ nouveau_bar_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	bar->iomem = ioremap(nv_device_resource_start(device, 3),
-			     nv_device_resource_len(device, 3));
+	if (nv_device_resource_len(device, 3) != 0)
+		bar->iomem = ioremap(nv_device_resource_start(device, 3),
+				     nv_device_resource_len(device, 3));
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c
index 3f30db62e656..ca8139b9ab27 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c
@@ -30,14 +30,16 @@
 
 #include "priv.h"
 
+struct nvc0_bar_priv_vm {
+	struct nouveau_gpuobj *mem;
+	struct nouveau_gpuobj *pgd;
+	struct nouveau_vm *vm;
+};
+
 struct nvc0_bar_priv {
 	struct nouveau_bar base;
 	spinlock_t lock;
-	struct {
-		struct nouveau_gpuobj *mem;
-		struct nouveau_gpuobj *pgd;
-		struct nouveau_vm *vm;
-	} bar[2];
+	struct nvc0_bar_priv_vm bar[2];
 };
 
 static int
@@ -79,87 +81,87 @@ nvc0_bar_unmap(struct nouveau_bar *bar, struct nouveau_vma *vma)
 }
 
 static int
-nvc0_bar_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
+nvc0_bar_init_vm(struct nvc0_bar_priv *priv, struct nvc0_bar_priv_vm *bar_vm,
+		 int bar_nr)
 {
-	struct nouveau_device *device = nv_device(parent);
-	struct nvc0_bar_priv *priv;
-	struct nouveau_gpuobj *mem;
+	struct nouveau_device *device = nv_device(&priv->base);
 	struct nouveau_vm *vm;
+	resource_size_t bar_len;
 	int ret;
 
-	ret = nouveau_bar_create(parent, engine, oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	/* BAR3 */
 	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x1000, 0, 0,
-				&priv->bar[0].mem);
-	mem = priv->bar[0].mem;
+				&bar_vm->mem);
 	if (ret)
 		return ret;
 
 	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x8000, 0, 0,
-				&priv->bar[0].pgd);
+				&bar_vm->pgd);
 	if (ret)
 		return ret;
 
-	ret = nouveau_vm_new(device, 0, nv_device_resource_len(device, 3), 0, &vm);
+	bar_len = nv_device_resource_len(device, bar_nr);
+
+	ret = nouveau_vm_new(device, 0, bar_len, 0, &vm);
 	if (ret)
 		return ret;
 
 	atomic_inc(&vm->engref[NVDEV_SUBDEV_BAR]);
 
-	ret = nouveau_gpuobj_new(nv_object(priv), NULL,
-				 (nv_device_resource_len(device, 3) >> 12) * 8,
-				 0x1000, NVOBJ_FLAG_ZERO_ALLOC,
-				 &vm->pgt[0].obj[0]);
-	vm->pgt[0].refcount[0] = 1;
-	if (ret)
-		return ret;
+	/*
+	 * Bootstrap page table lookup.
+	 */
+	if (bar_nr == 3) {
+		ret = nouveau_gpuobj_new(nv_object(priv), NULL,
+					 (bar_len >> 12) * 8, 0x1000,
+					 NVOBJ_FLAG_ZERO_ALLOC,
+					&vm->pgt[0].obj[0]);
+		vm->pgt[0].refcount[0] = 1;
+		if (ret)
+			return ret;
+	}
 
-	ret = nouveau_vm_ref(vm, &priv->bar[0].vm, priv->bar[0].pgd);
+	ret = nouveau_vm_ref(vm, &bar_vm->vm, bar_vm->pgd);
 	nouveau_vm_ref(NULL, &vm, NULL);
 	if (ret)
 		return ret;
 
-	nv_wo32(mem, 0x0200, lower_32_bits(priv->bar[0].pgd->addr));
-	nv_wo32(mem, 0x0204, upper_32_bits(priv->bar[0].pgd->addr));
-	nv_wo32(mem, 0x0208, lower_32_bits(nv_device_resource_len(device, 3) - 1));
-	nv_wo32(mem, 0x020c, upper_32_bits(nv_device_resource_len(device, 3) - 1));
+	nv_wo32(bar_vm->mem, 0x0200, lower_32_bits(bar_vm->pgd->addr));
+	nv_wo32(bar_vm->mem, 0x0204, upper_32_bits(bar_vm->pgd->addr));
+	nv_wo32(bar_vm->mem, 0x0208, lower_32_bits(bar_len - 1));
+	nv_wo32(bar_vm->mem, 0x020c, upper_32_bits(bar_len - 1));
 
-	/* BAR1 */
-	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x1000, 0, 0,
-				&priv->bar[1].mem);
-	mem = priv->bar[1].mem;
-	if (ret)
-		return ret;
+	return 0;
+}
 
-	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x8000, 0, 0,
-				&priv->bar[1].pgd);
-	if (ret)
-		return ret;
+static int
+nvc0_bar_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nouveau_device *device = nv_device(parent);
+	struct nvc0_bar_priv *priv;
+	bool has_bar3 = nv_device_resource_len(device, 3) != 0;
+	int ret;
 
-	ret = nouveau_vm_new(device, 0, nv_device_resource_len(device, 1), 0, &vm);
+	ret = nouveau_bar_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	atomic_inc(&vm->engref[NVDEV_SUBDEV_BAR]);
+	/* BAR3 */
+	if (has_bar3) {
+		ret = nvc0_bar_init_vm(priv, &priv->bar[0], 3);
+		if (ret)
+			return ret;
+		priv->base.alloc = nouveau_bar_alloc;
+		priv->base.kmap = nvc0_bar_kmap;
+	}
 
-	ret = nouveau_vm_ref(vm, &priv->bar[1].vm, priv->bar[1].pgd);
-	nouveau_vm_ref(NULL, &vm, NULL);
+	/* BAR1 */
+	ret = nvc0_bar_init_vm(priv, &priv->bar[1], 1);
 	if (ret)
 		return ret;
 
-	nv_wo32(mem, 0x0200, lower_32_bits(priv->bar[1].pgd->addr));
-	nv_wo32(mem, 0x0204, upper_32_bits(priv->bar[1].pgd->addr));
-	nv_wo32(mem, 0x0208, lower_32_bits(nv_device_resource_len(device, 1) - 1));
-	nv_wo32(mem, 0x020c, upper_32_bits(nv_device_resource_len(device, 1) - 1));
-
-	priv->base.alloc = nouveau_bar_alloc;
-	priv->base.kmap = nvc0_bar_kmap;
 	priv->base.umap = nvc0_bar_umap;
 	priv->base.unmap = nvc0_bar_unmap;
 	priv->base.flush = nv84_bar_flush;
@@ -201,7 +203,9 @@ nvc0_bar_init(struct nouveau_object *object)
 	nv_mask(priv, 0x100c80, 0x00000001, 0x00000000);
 
 	nv_wr32(priv, 0x001704, 0x80000000 | priv->bar[1].mem->addr >> 12);
-	nv_wr32(priv, 0x001714, 0xc0000000 | priv->bar[0].mem->addr >> 12);
+	if (priv->bar[0].mem)
+		nv_wr32(priv, 0x001714,
+			0xc0000000 | priv->bar[0].mem->addr >> 12);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
index 222e8ebb669d..d45704a2c2df 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
@@ -183,10 +183,11 @@ nouveau_bios_shadow_prom(struct nouveau_bios *bios)
 		goto out;
 
 	bios->data = kmalloc(bios->size, GFP_KERNEL);
-	if (bios->data) {
-		for (i = 0; i < bios->size; i += 4)
-			((u32 *)bios->data)[i/4] = nv_rd32(bios, 0x300000 + i);
-	}
+	if (!bios->data)
+		goto out;
+
+	for (i = 0; i < bios->size; i += 4)
+		((u32 *)bios->data)[i/4] = nv_rd32(bios, 0x300000 + i);
 
 	/* check the PCI record header */
 	pcir = nv_ro16(bios, 0x0018);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c b/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c
index 5ac010efd959..2ede3bcd96a1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c
@@ -28,12 +28,12 @@
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/conn.h>
 
-u16
-dcb_conntab(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+u32
+nvbios_connTe(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 {
-	u16 dcb = dcb_table(bios, ver, hdr, cnt, len);
+	u32 dcb = dcb_table(bios, ver, hdr, cnt, len);
 	if (dcb && *ver >= 0x30 && *hdr >= 0x16) {
-		u16 data = nv_ro16(bios, dcb + 0x14);
+		u32 data = nv_ro16(bios, dcb + 0x14);
 		if (data) {
 			*ver = nv_ro08(bios, data + 0);
 			*hdr = nv_ro08(bios, data + 1);
@@ -42,15 +42,59 @@ dcb_conntab(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 			return data;
 		}
 	}
-	return 0x0000;
+	return 0x00000000;
 }
 
-u16
-dcb_conn(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len)
+u32
+nvbios_connTp(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+	      struct nvbios_connT *info)
+{
+	u32 data = nvbios_connTe(bios, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!data * *ver) {
+	case 0x30:
+	case 0x40:
+		return data;
+	default:
+		break;
+	}
+	return 0x00000000;
+}
+
+u32
+nvbios_connEe(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len)
 {
 	u8  hdr, cnt;
-	u16 data = dcb_conntab(bios, ver, &hdr, &cnt, len);
+	u32 data = nvbios_connTe(bios, ver, &hdr, &cnt, len);
 	if (data && idx < cnt)
 		return data + hdr + (idx * *len);
-	return 0x0000;
+	return 0x00000000;
+}
+
+u32
+nvbios_connEp(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len,
+	      struct nvbios_connE *info)
+{
+	u32 data = nvbios_connEe(bios, idx, ver, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!data * *ver) {
+	case 0x30:
+	case 0x40:
+		info->type     =  nv_ro08(bios, data + 0x00);
+		info->location =  nv_ro08(bios, data + 0x01) & 0x0f;
+		info->hpd      = (nv_ro08(bios, data + 0x01) & 0x30) >> 4;
+		info->dp       = (nv_ro08(bios, data + 0x01) & 0xc0) >> 6;
+		if (*len < 4)
+			return data;
+		info->hpd     |= (nv_ro08(bios, data + 0x02) & 0x03) << 2;
+		info->dp      |=  nv_ro08(bios, data + 0x02) & 0x0c;
+		info->di       = (nv_ro08(bios, data + 0x02) & 0xf0) >> 4;
+		info->hpd     |= (nv_ro08(bios, data + 0x03) & 0x07) << 4;
+		info->sr       = (nv_ro08(bios, data + 0x03) & 0x08) >> 3;
+		info->lcdid    = (nv_ro08(bios, data + 0x03) & 0x70) >> 4;
+		return data;
+	default:
+		break;
+	}
+	return 0x00000000;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c b/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
index 7628fe759220..f309dd657250 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
@@ -162,18 +162,20 @@ nvbios_dpcfg_parse(struct nouveau_bios *bios, u16 outp, u8 idx,
 		   struct nvbios_dpcfg *info)
 {
 	u16 data = nvbios_dpcfg_entry(bios, outp, idx, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
 	if (data) {
 		switch (*ver) {
 		case 0x21:
-			info->drv = nv_ro08(bios, data + 0x02);
-			info->pre = nv_ro08(bios, data + 0x03);
-			info->unk = nv_ro08(bios, data + 0x04);
+			info->dc    = nv_ro08(bios, data + 0x02);
+			info->pe    = nv_ro08(bios, data + 0x03);
+			info->tx_pu = nv_ro08(bios, data + 0x04);
 			break;
 		case 0x30:
 		case 0x40:
-			info->drv = nv_ro08(bios, data + 0x01);
-			info->pre = nv_ro08(bios, data + 0x02);
-			info->unk = nv_ro08(bios, data + 0x03);
+			info->pc    = nv_ro08(bios, data + 0x00);
+			info->dc    = nv_ro08(bios, data + 0x01);
+			info->pe    = nv_ro08(bios, data + 0x02);
+			info->tx_pu = nv_ro08(bios, data + 0x03);
 			break;
 		default:
 			data = 0x0000;
@@ -184,7 +186,7 @@ nvbios_dpcfg_parse(struct nouveau_bios *bios, u16 outp, u8 idx,
 }
 
 u16
-nvbios_dpcfg_match(struct nouveau_bios *bios, u16 outp, u8 un, u8 vs, u8 pe,
+nvbios_dpcfg_match(struct nouveau_bios *bios, u16 outp, u8 pc, u8 vs, u8 pe,
 		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		   struct nvbios_dpcfg *info)
 {
@@ -193,16 +195,15 @@ nvbios_dpcfg_match(struct nouveau_bios *bios, u16 outp, u8 un, u8 vs, u8 pe,
 
 	if (*ver >= 0x30) {
 		const u8 vsoff[] = { 0, 4, 7, 9 };
-		idx = (un * 10) + vsoff[vs] + pe;
+		idx = (pc * 10) + vsoff[vs] + pe;
 	} else {
-		while ((data = nvbios_dpcfg_entry(bios, outp, idx,
+		while ((data = nvbios_dpcfg_entry(bios, outp, ++idx,
 						  ver, hdr, cnt, len))) {
 			if (nv_ro08(bios, data + 0x00) == vs &&
 			    nv_ro08(bios, data + 0x01) == pe)
 				break;
-			idx++;
 		}
 	}
 
-	return nvbios_dpcfg_parse(bios, outp, pe, ver, hdr, cnt, len, info);
+	return nvbios_dpcfg_parse(bios, outp, idx, ver, hdr, cnt, len, info);
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
index acaeaf79e3f0..626380f9e4c0 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
@@ -98,15 +98,16 @@ static u8
 init_conn(struct nvbios_init *init)
 {
 	struct nouveau_bios *bios = init->bios;
-	u8  ver, len;
-	u16 conn;
+	struct nvbios_connE connE;
+	u8  ver, hdr;
+	u32 conn;
 
 	if (init_exec(init)) {
 		if (init->outp) {
 			conn = init->outp->connector;
-			conn = dcb_conn(bios, conn, &ver, &len);
+			conn = nvbios_connEp(bios, conn, &ver, &hdr, &connE);
 			if (conn)
-				return nv_ro08(bios, conn);
+				return connE.type;
 		}
 
 		error("script needs connector type\n");
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/base.c b/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
index dd62baead39c..22351f594d2a 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
@@ -346,8 +346,8 @@ nouveau_clock_ustate_update(struct nouveau_clock *clk, int req)
 	struct nouveau_pstate *pstate;
 	int i = 0;
 
-	/* YKW repellant */
-	return -ENOSYS;
+	if (!clk->allow_reclock)
+		return -ENOSYS;
 
 	if (req != -1 && req != -2) {
 		list_for_each_entry(pstate, &clk->states, head) {
@@ -456,6 +456,7 @@ nouveau_clock_create_(struct nouveau_object *parent,
 		      struct nouveau_object *engine,
 		      struct nouveau_oclass *oclass,
 		      struct nouveau_clocks *clocks,
+		      bool allow_reclock,
 		      int length, void **object)
 {
 	struct nouveau_device *device = nv_device(parent);
@@ -478,6 +479,8 @@ nouveau_clock_create_(struct nouveau_object *parent,
 		ret = nouveau_pstate_new(clk, idx++);
 	} while (ret == 0);
 
+	clk->allow_reclock = allow_reclock;
+
 	mode = nouveau_stropt(device->cfgopt, "NvClkMode", &arglen);
 	if (mode) {
 		if (!strncasecmpz(mode, "disabled", arglen)) {
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
index b74db6cfc4e2..eb2d4425a49e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
@@ -82,7 +82,8 @@ nv04_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv04_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nv04_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nv04_domain, false,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
index db7346f79080..8a9e16839791 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
@@ -213,7 +213,8 @@ nv40_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv40_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nv40_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nv40_domain, true,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
index 250a6d96016b..8c132772ba9e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
@@ -507,7 +507,7 @@ nv50_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	int ret;
 
 	ret = nouveau_clock_create(parent, engine, oclass, pclass->domains,
-				  &priv);
+				   false, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
index 4f5a1373f002..9fb58354a80b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
@@ -302,7 +302,8 @@ nva3_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nva3_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nva3_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nva3_domain, false,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
index 7a723b4f564d..6a65fc9e9663 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
@@ -421,7 +421,8 @@ nvaa_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvaa_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nvaa_domains, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nvaa_domains, true,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
index c3105720ed24..dbf8517f54da 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
@@ -437,7 +437,8 @@ nvc0_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvc0_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nvc0_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nvc0_domain, false,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
index d3c37c96f0e7..0e62a3240144 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
@@ -307,7 +307,6 @@ calc_clk(struct nve0_clock_priv *priv,
 		info->dsrc = src0;
 		if (div0) {
 			info->ddiv |= 0x80000000;
-			info->ddiv |= div0 << 8;
 			info->ddiv |= div0;
 		}
 		if (div1D) {
@@ -352,7 +351,7 @@ nve0_clock_prog_0(struct nve0_clock_priv *priv, int clk)
 {
 	struct nve0_clock_info *info = &priv->eng[clk];
 	if (!info->ssel) {
-		nv_mask(priv, 0x1371d0 + (clk * 0x04), 0x80003f3f, info->ddiv);
+		nv_mask(priv, 0x1371d0 + (clk * 0x04), 0x8000003f, info->ddiv);
 		nv_wr32(priv, 0x137160 + (clk * 0x04), info->dsrc);
 	}
 }
@@ -389,7 +388,10 @@ static void
 nve0_clock_prog_3(struct nve0_clock_priv *priv, int clk)
 {
 	struct nve0_clock_info *info = &priv->eng[clk];
-	nv_mask(priv, 0x137250 + (clk * 0x04), 0x00003f3f, info->mdiv);
+	if (info->ssel)
+		nv_mask(priv, 0x137250 + (clk * 0x04), 0x00003f00, info->mdiv);
+	else
+		nv_mask(priv, 0x137250 + (clk * 0x04), 0x0000003f, info->mdiv);
 }
 
 static void
@@ -473,7 +475,8 @@ nve0_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nve0_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nve0_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nve0_domain, true,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
new file mode 100644
index 000000000000..a16024a74771
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
@@ -0,0 +1,56 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "nvc0.h"
+
+struct gk20a_fb_priv {
+	struct nouveau_fb base;
+};
+
+static int
+gk20a_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	     struct nouveau_oclass *oclass, void *data, u32 size,
+	     struct nouveau_object **pobject)
+{
+	struct gk20a_fb_priv *priv;
+	int ret;
+
+	ret = nouveau_fb_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass *
+gk20a_fb_oclass = &(struct nouveau_fb_impl) {
+	.base.handle = NV_SUBDEV(FB, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = gk20a_fb_ctor,
+		.dtor = _nouveau_fb_dtor,
+		.init = _nouveau_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.memtype = nvc0_fb_memtype_valid,
+	.ram = &gk20a_ram_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h b/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
index da74c889aed4..82273f832e42 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
@@ -32,6 +32,7 @@ extern struct nouveau_oclass nva3_ram_oclass;
 extern struct nouveau_oclass nvaa_ram_oclass;
 extern struct nouveau_oclass nvc0_ram_oclass;
 extern struct nouveau_oclass nve0_ram_oclass;
+extern struct nouveau_oclass gk20a_ram_oclass;
 extern struct nouveau_oclass gm107_ram_oclass;
 
 int nouveau_sddr3_calc(struct nouveau_ram *ram);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h b/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h
index 0f57fcfe0bbf..2af9cfd2c60f 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h
@@ -26,7 +26,7 @@ ramfuc_reg2(u32 addr1, u32 addr2)
 	};
 }
 
-static inline struct ramfuc_reg
+static noinline struct ramfuc_reg
 ramfuc_reg(u32 addr)
 {
 	return ramfuc_reg2(addr, addr);
@@ -107,7 +107,7 @@ ramfuc_nsec(struct ramfuc *ram, u32 nsec)
 
 #define ram_init(s,p)       ramfuc_init(&(s)->base, (p))
 #define ram_exec(s,e)       ramfuc_exec(&(s)->base, (e))
-#define ram_have(s,r)       ((s)->r_##r.addr != 0x000000)
+#define ram_have(s,r)       ((s)->r_##r.addr[0] != 0x000000)
 #define ram_rd32(s,r)       ramfuc_rd32(&(s)->base, &(s)->r_##r)
 #define ram_wr32(s,r,d)     ramfuc_wr32(&(s)->base, &(s)->r_##r, (d))
 #define ram_nuke(s,r)       ramfuc_nuke(&(s)->base, &(s)->r_##r)
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramgk20a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramgk20a.c
new file mode 100644
index 000000000000..4d77d75e4673
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramgk20a.c
@@ -0,0 +1,152 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "priv.h"
+
+#include <subdev/fb.h>
+
+struct gk20a_mem {
+	struct nouveau_mem base;
+	void *cpuaddr;
+	dma_addr_t handle;
+};
+#define to_gk20a_mem(m) container_of(m, struct gk20a_mem, base)
+
+static void
+gk20a_ram_put(struct nouveau_fb *pfb, struct nouveau_mem **pmem)
+{
+	struct device *dev = nv_device_base(nv_device(pfb));
+	struct gk20a_mem *mem = to_gk20a_mem(*pmem);
+
+	*pmem = NULL;
+	if (unlikely(mem == NULL))
+		return;
+
+	if (likely(mem->cpuaddr))
+		dma_free_coherent(dev, mem->base.size << PAGE_SHIFT,
+				  mem->cpuaddr, mem->handle);
+
+	kfree(mem->base.pages);
+	kfree(mem);
+}
+
+static int
+gk20a_ram_get(struct nouveau_fb *pfb, u64 size, u32 align, u32 ncmin,
+	     u32 memtype, struct nouveau_mem **pmem)
+{
+	struct device *dev = nv_device_base(nv_device(pfb));
+	struct gk20a_mem *mem;
+	u32 type = memtype & 0xff;
+	u32 npages, order;
+	int i;
+
+	nv_debug(pfb, "%s: size: %llx align: %x, ncmin: %x\n", __func__, size,
+		 align, ncmin);
+
+	npages = size >> PAGE_SHIFT;
+	if (npages == 0)
+		npages = 1;
+
+	if (align == 0)
+		align = PAGE_SIZE;
+	align >>= PAGE_SHIFT;
+
+	/* round alignment to the next power of 2, if needed */
+	order = fls(align);
+	if ((align & (align - 1)) == 0)
+		order--;
+	align = BIT(order);
+
+	/* ensure returned address is correctly aligned */
+	npages = max(align, npages);
+
+	mem = kzalloc(sizeof(*mem), GFP_KERNEL);
+	if (!mem)
+		return -ENOMEM;
+
+	mem->base.size = npages;
+	mem->base.memtype = type;
+
+	mem->base.pages = kzalloc(sizeof(dma_addr_t) * npages, GFP_KERNEL);
+	if (!mem->base.pages) {
+		kfree(mem);
+		return -ENOMEM;
+	}
+
+	*pmem = &mem->base;
+
+	mem->cpuaddr = dma_alloc_coherent(dev, npages << PAGE_SHIFT,
+					  &mem->handle, GFP_KERNEL);
+	if (!mem->cpuaddr) {
+		nv_error(pfb, "%s: cannot allocate memory!\n", __func__);
+		gk20a_ram_put(pfb, pmem);
+		return -ENOMEM;
+	}
+
+	align <<= PAGE_SHIFT;
+
+	/* alignment check */
+	if (unlikely(mem->handle & (align - 1)))
+		nv_warn(pfb, "memory not aligned as requested: %pad (0x%x)\n",
+			&mem->handle, align);
+
+	nv_debug(pfb, "alloc size: 0x%x, align: 0x%x, paddr: %pad, vaddr: %p\n",
+		 npages << PAGE_SHIFT, align, &mem->handle, mem->cpuaddr);
+
+	for (i = 0; i < npages; i++)
+		mem->base.pages[i] = mem->handle + (PAGE_SIZE * i);
+
+	mem->base.offset = (u64)mem->base.pages[0];
+
+	return 0;
+}
+
+static int
+gk20a_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 datasize,
+	      struct nouveau_object **pobject)
+{
+	struct nouveau_ram *ram;
+	int ret;
+
+	ret = nouveau_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+	ram->type = NV_MEM_TYPE_STOLEN;
+	ram->size = get_num_physpages() << PAGE_SHIFT;
+
+	ram->get = gk20a_ram_get;
+	ram->put = gk20a_ram_put;
+
+	return 0;
+}
+
+struct nouveau_oclass
+gk20a_ram_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = gk20a_ram_ctor,
+		.dtor = _nouveau_ram_dtor,
+		.init = _nouveau_ram_init,
+		.fini = _nouveau_ram_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
index ef91b6e893af..e5d12c24cc43 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
@@ -211,7 +211,7 @@ nv50_ram_prog(struct nouveau_fb *pfb)
 	struct nv50_ram *ram = (void *)pfb->ram;
 	struct nv50_ramseq *hwsq = &ram->hwsq;
 
-	ram_exec(hwsq, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(hwsq, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
index 6eb97f16fbda..8076fb195dd5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
@@ -309,7 +309,7 @@ nva3_ram_prog(struct nouveau_fb *pfb)
 	struct nouveau_device *device = nv_device(pfb);
 	struct nva3_ram *ram = (void *)pfb->ram;
 	struct nva3_ramfuc *fuc = &ram->fuc;
-	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
index 8edc92224c84..5a6a5027f749 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
@@ -408,7 +408,7 @@ nvc0_ram_prog(struct nouveau_fb *pfb)
 	struct nouveau_device *device = nv_device(pfb);
 	struct nvc0_ram *ram = (void *)pfb->ram;
 	struct nvc0_ramfuc *fuc = &ram->fuc;
-	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
index 16752192cf87..c5b46e302319 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
@@ -200,6 +200,7 @@ r1373f4_init(struct nve0_ramfuc *fuc)
 	/* (re)program mempll, if required */
 	if (ram->mode == 2) {
 		ram_mask(fuc, 0x1373f4, 0x00010000, 0x00000000);
+		ram_mask(fuc, 0x132000, 0x80000000, 0x80000000);
 		ram_mask(fuc, 0x132000, 0x00000001, 0x00000000);
 		ram_mask(fuc, 0x132004, 0x103fffff, mcoef);
 		ram_mask(fuc, 0x132000, 0x00000001, 0x00000001);
@@ -262,8 +263,8 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 	struct nve0_ram *ram = (void *)pfb->ram;
 	struct nve0_ramfuc *fuc = &ram->fuc;
 	struct nouveau_ram_data *next = ram->base.next;
-	int vc = !(next->bios.ramcfg_11_02_08);
-	int mv = !(next->bios.ramcfg_11_02_04);
+	int vc = !next->bios.ramcfg_11_02_08;
+	int mv = !next->bios.ramcfg_11_02_04;
 	u32 mask, data;
 
 	ram_mask(fuc, 0x10f808, 0x40000000, 0x40000000);
@@ -370,8 +371,8 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 		}
 	}
 
-	if ( (next->bios.ramcfg_11_02_40) ||
-	     (next->bios.ramcfg_11_07_10)) {
+	if (next->bios.ramcfg_11_02_40 ||
+	    next->bios.ramcfg_11_07_10) {
 		ram_mask(fuc, 0x132040, 0x00010000, 0x00010000);
 		ram_nsec(fuc, 20000);
 	}
@@ -417,7 +418,7 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 		ram_mask(fuc, 0x10f694, 0xff00ff00, data);
 	}
 
-	if (ram->mode == 2 && (next->bios.ramcfg_11_08_10))
+	if (ram->mode == 2 && next->bios.ramcfg_11_08_10)
 		data = 0x00000080;
 	else
 		data = 0x00000000;
@@ -425,13 +426,13 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 
 	mask = 0x00070000;
 	data = 0x00000000;
-	if (!(next->bios.ramcfg_11_02_80))
+	if (!next->bios.ramcfg_11_02_80)
 		data |= 0x03000000;
-	if (!(next->bios.ramcfg_11_02_40))
+	if (!next->bios.ramcfg_11_02_40)
 		data |= 0x00002000;
-	if (!(next->bios.ramcfg_11_07_10))
+	if (!next->bios.ramcfg_11_07_10)
 		data |= 0x00004000;
-	if (!(next->bios.ramcfg_11_07_08))
+	if (!next->bios.ramcfg_11_07_08)
 		data |= 0x00000003;
 	else
 		data |= 0x74000000;
@@ -486,7 +487,7 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 
 	data = mask = 0x00000000;
 	if (NOTE00(ramcfg_02_03 != 0)) {
-		data |= (next->bios.ramcfg_11_02_03) << 8;
+		data |= next->bios.ramcfg_11_02_03 << 8;
 		mask |= 0x00000300;
 	}
 	if (NOTE00(ramcfg_01_10)) {
@@ -498,7 +499,7 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 
 	data = mask = 0x00000000;
 	if (NOTE00(timing_30_07 != 0)) {
-		data |= (next->bios.timing_20_30_07) << 28;
+		data |= next->bios.timing_20_30_07 << 28;
 		mask |= 0x70000000;
 	}
 	if (NOTE00(ramcfg_01_01)) {
@@ -510,7 +511,7 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 
 	data = mask = 0x00000000;
 	if (NOTE00(timing_30_07 != 0)) {
-		data |= (next->bios.timing_20_30_07) << 28;
+		data |= next->bios.timing_20_30_07 << 28;
 		mask |= 0x70000000;
 	}
 	if (NOTE00(ramcfg_01_02)) {
@@ -522,16 +523,16 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 
 	mask = 0x33f00000;
 	data = 0x00000000;
-	if (!(next->bios.ramcfg_11_01_04))
+	if (!next->bios.ramcfg_11_01_04)
 		data |= 0x20200000;
-	if (!(next->bios.ramcfg_11_07_80))
+	if (!next->bios.ramcfg_11_07_80)
 		data |= 0x12800000;
 	/*XXX: see note above about there probably being some condition
 	 *     for the 10f824 stuff that uses ramcfg 3...
 	 */
-	if ( (next->bios.ramcfg_11_03_f0)) {
+	if (next->bios.ramcfg_11_03_f0) {
 		if (next->bios.rammap_11_08_0c) {
-			if (!(next->bios.ramcfg_11_07_80))
+			if (!next->bios.ramcfg_11_07_80)
 				mask |= 0x00000020;
 			else
 				data |= 0x00000020;
@@ -563,7 +564,7 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 		ram_wait(fuc, 0x100710, 0x80000000, 0x80000000, 200000);
 	}
 
-	data = (next->bios.timing_20_30_07) << 8;
+	data = next->bios.timing_20_30_07 << 8;
 	if (next->bios.ramcfg_11_01_01)
 		data |= 0x80000000;
 	ram_mask(fuc, 0x100778, 0x00000700, data);
@@ -588,7 +589,7 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
 	ram_wr32(fuc, 0x10f210, 0x80000000); /* REFRESH_AUTO = 1 */
 
-	if ((next->bios.ramcfg_11_08_10) && (ram->mode == 2) /*XXX*/) {
+	if (next->bios.ramcfg_11_08_10 && (ram->mode == 2) /*XXX*/) {
 		u32 temp = ram_mask(fuc, 0x10f294, 0xff000000, 0x24000000);
 		nve0_ram_train(fuc, 0xbc0e0000, 0xa4010000); /*XXX*/
 		ram_nsec(fuc, 1000);
@@ -621,8 +622,8 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 	data  = ram_rd32(fuc, 0x10f978);
 	data &= ~0x00046144;
 	data |=  0x0000000b;
-	if (!(next->bios.ramcfg_11_07_08)) {
-		if (!(next->bios.ramcfg_11_07_04))
+	if (!next->bios.ramcfg_11_07_08) {
+		if (!next->bios.ramcfg_11_07_04)
 			data |= 0x0000200c;
 		else
 			data |= 0x00000000;
@@ -636,11 +637,11 @@ nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
 		ram_wr32(fuc, 0x10f830, data);
 	}
 
-	if (!(next->bios.ramcfg_11_07_08)) {
+	if (!next->bios.ramcfg_11_07_08) {
 		data = 0x88020000;
-		if ( (next->bios.ramcfg_11_07_04))
+		if ( next->bios.ramcfg_11_07_04)
 			data |= 0x10000000;
-		if (!(next->bios.rammap_11_08_10))
+		if (!next->bios.rammap_11_08_10)
 			data |= 0x00080000;
 	} else {
 		data = 0xa40e0000;
@@ -689,8 +690,8 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 	const u32 runk0 = ram->fN1 << 16;
 	const u32 runk1 = ram->fN1;
 	struct nouveau_ram_data *next = ram->base.next;
-	int vc = !(next->bios.ramcfg_11_02_08);
-	int mv = !(next->bios.ramcfg_11_02_04);
+	int vc = !next->bios.ramcfg_11_02_08;
+	int mv = !next->bios.ramcfg_11_02_04;
 	u32 mask, data;
 
 	ram_mask(fuc, 0x10f808, 0x40000000, 0x40000000);
@@ -705,7 +706,7 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 	}
 
 	ram_mask(fuc, 0x10f200, 0x00000800, 0x00000000);
-	if ((next->bios.ramcfg_11_03_f0))
+	if (next->bios.ramcfg_11_03_f0)
 		ram_mask(fuc, 0x10f808, 0x04000000, 0x04000000);
 
 	ram_wr32(fuc, 0x10f314, 0x00000001); /* PRECHARGE */
@@ -761,7 +762,7 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 
 	ram_mask(fuc, 0x1373f4, 0x00000000, 0x00010010);
 	data  = ram_rd32(fuc, 0x1373ec) & ~0x00030000;
-	data |= (next->bios.ramcfg_11_03_30) << 12;
+	data |= next->bios.ramcfg_11_03_30 << 16;
 	ram_wr32(fuc, 0x1373ec, data);
 	ram_mask(fuc, 0x1373f4, 0x00000003, 0x00000000);
 	ram_mask(fuc, 0x1373f4, 0x00000010, 0x00000000);
@@ -793,8 +794,8 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 		}
 	}
 
-	if ( (next->bios.ramcfg_11_02_40) ||
-	     (next->bios.ramcfg_11_07_10)) {
+	if (next->bios.ramcfg_11_02_40 ||
+	    next->bios.ramcfg_11_07_10) {
 		ram_mask(fuc, 0x132040, 0x00010000, 0x00010000);
 		ram_nsec(fuc, 20000);
 	}
@@ -810,13 +811,13 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 
 	mask = 0x00010000;
 	data = 0x00000000;
-	if (!(next->bios.ramcfg_11_02_80))
+	if (!next->bios.ramcfg_11_02_80)
 		data |= 0x03000000;
-	if (!(next->bios.ramcfg_11_02_40))
+	if (!next->bios.ramcfg_11_02_40)
 		data |= 0x00002000;
-	if (!(next->bios.ramcfg_11_07_10))
+	if (!next->bios.ramcfg_11_07_10)
 		data |= 0x00004000;
-	if (!(next->bios.ramcfg_11_07_08))
+	if (!next->bios.ramcfg_11_07_08)
 		data |= 0x00000003;
 	else
 		data |= 0x14000000;
@@ -844,16 +845,16 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 
 	mask = 0x33f00000;
 	data = 0x00000000;
-	if (!(next->bios.ramcfg_11_01_04))
+	if (!next->bios.ramcfg_11_01_04)
 		data |= 0x20200000;
-	if (!(next->bios.ramcfg_11_07_80))
+	if (!next->bios.ramcfg_11_07_80)
 		data |= 0x12800000;
 	/*XXX: see note above about there probably being some condition
 	 *     for the 10f824 stuff that uses ramcfg 3...
 	 */
-	if ( (next->bios.ramcfg_11_03_f0)) {
+	if (next->bios.ramcfg_11_03_f0) {
 		if (next->bios.rammap_11_08_0c) {
-			if (!(next->bios.ramcfg_11_07_80))
+			if (!next->bios.ramcfg_11_07_80)
 				mask |= 0x00000020;
 			else
 				data |= 0x00000020;
@@ -876,7 +877,7 @@ nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
 		data = next->bios.timing_20_2c_1fc0;
 	ram_mask(fuc, 0x10f24c, 0x7f000000, data << 24);
 
-	ram_mask(fuc, 0x10f224, 0x001f0000, next->bios.timing_20_30_f8);
+	ram_mask(fuc, 0x10f224, 0x001f0000, next->bios.timing_20_30_f8 << 16);
 
 	ram_wr32(fuc, 0x10f090, 0x4000007f);
 	ram_nsec(fuc, 1000);
@@ -1111,7 +1112,7 @@ nve0_ram_prog(struct nouveau_fb *pfb)
 	struct nouveau_device *device = nv_device(pfb);
 	struct nve0_ram *ram = (void *)pfb->ram;
 	struct nve0_ramfuc *fuc = &ram->fuc;
-	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return (ram->base.next == &ram->base.xition);
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
index f572c2804c32..45e0202f3151 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
@@ -22,21 +22,24 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/gpio.h>
 #include <subdev/bios.h>
 #include <subdev/bios/gpio.h>
 
+#include "priv.h"
+
 static int
 nouveau_gpio_drive(struct nouveau_gpio *gpio,
 		   int idx, int line, int dir, int out)
 {
-	return gpio->drive ? gpio->drive(gpio, line, dir, out) : -ENODEV;
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	return impl->drive ? impl->drive(gpio, line, dir, out) : -ENODEV;
 }
 
 static int
 nouveau_gpio_sense(struct nouveau_gpio *gpio, int idx, int line)
 {
-	return gpio->sense ? gpio->sense(gpio, line) : -ENODEV;
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	return impl->sense ? impl->sense(gpio, line) : -ENODEV;
 }
 
 static int
@@ -102,6 +105,80 @@ nouveau_gpio_get(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line)
 	return ret;
 }
 
+static void
+nouveau_gpio_intr_disable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(event->priv);
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	impl->intr_mask(gpio, type, 1 << index, 0);
+}
+
+static void
+nouveau_gpio_intr_enable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(event->priv);
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	impl->intr_mask(gpio, type, 1 << index, 1 << index);
+}
+
+static void
+nouveau_gpio_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(subdev);
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	u32 hi, lo, e, i;
+
+	impl->intr_stat(gpio, &hi, &lo);
+
+	for (i = 0; e = 0, (hi | lo) && i < impl->lines; i++) {
+		if (hi & (1 << i))
+			e |= NVKM_GPIO_HI;
+		if (lo & (1 << i))
+			e |= NVKM_GPIO_LO;
+		nouveau_event_trigger(gpio->events, e, i);
+	}
+}
+
+int
+_nouveau_gpio_fini(struct nouveau_object *object, bool suspend)
+{
+	const struct nouveau_gpio_impl *impl = (void *)object->oclass;
+	struct nouveau_gpio *gpio = nouveau_gpio(object);
+	u32 mask = (1 << impl->lines) - 1;
+
+	impl->intr_mask(gpio, NVKM_GPIO_TOGGLED, mask, 0);
+	impl->intr_stat(gpio, &mask, &mask);
+
+	return nouveau_subdev_fini(&gpio->base, suspend);
+}
+
+static struct dmi_system_id gpio_reset_ids[] = {
+	{
+		.ident = "Apple Macbook 10,1",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro10,1"),
+		}
+	},
+	{ }
+};
+
+int
+_nouveau_gpio_init(struct nouveau_object *object)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(object);
+	int ret;
+
+	ret = nouveau_subdev_init(&gpio->base);
+	if (ret)
+		return ret;
+
+	if (gpio->reset && dmi_check_system(gpio_reset_ids))
+		gpio->reset(gpio, DCB_GPIO_UNUSED);
+
+	return ret;
+}
+
 void
 _nouveau_gpio_dtor(struct nouveau_object *object)
 {
@@ -113,9 +190,10 @@ _nouveau_gpio_dtor(struct nouveau_object *object)
 int
 nouveau_gpio_create_(struct nouveau_object *parent,
 		     struct nouveau_object *engine,
-		     struct nouveau_oclass *oclass, int lines,
+		     struct nouveau_oclass *oclass,
 		     int length, void **pobject)
 {
+	const struct nouveau_gpio_impl *impl = (void *)oclass;
 	struct nouveau_gpio *gpio;
 	int ret;
 
@@ -125,34 +203,34 @@ nouveau_gpio_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	ret = nouveau_event_create(lines, &gpio->events);
-	if (ret)
-		return ret;
-
 	gpio->find = nouveau_gpio_find;
 	gpio->set  = nouveau_gpio_set;
 	gpio->get  = nouveau_gpio_get;
+	gpio->reset = impl->reset;
+
+	ret = nouveau_event_create(2, impl->lines, &gpio->events);
+	if (ret)
+		return ret;
+
+	gpio->events->priv = gpio;
+	gpio->events->enable = nouveau_gpio_intr_enable;
+	gpio->events->disable = nouveau_gpio_intr_disable;
+	nv_subdev(gpio)->intr = nouveau_gpio_intr;
 	return 0;
 }
 
-static struct dmi_system_id gpio_reset_ids[] = {
-	{
-		.ident = "Apple Macbook 10,1",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro10,1"),
-		}
-	},
-	{ }
-};
-
 int
-nouveau_gpio_init(struct nouveau_gpio *gpio)
+_nouveau_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 size,
+		   struct nouveau_object **pobject)
 {
-	int ret = nouveau_subdev_init(&gpio->base);
-	if (ret == 0 && gpio->reset) {
-		if (dmi_check_system(gpio_reset_ids))
-			gpio->reset(gpio, DCB_GPIO_UNUSED);
-	}
-	return ret;
+	struct nouveau_gpio *gpio;
+	int ret;
+
+	ret = nouveau_gpio_create(parent, engine, oclass, &gpio);
+	*pobject = nv_object(gpio);
+	if (ret)
+		return ret;
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
index 76d5d5465ddd..27ad23eaf185 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
@@ -26,10 +26,6 @@
 
 #include "priv.h"
 
-struct nv10_gpio_priv {
-	struct nouveau_gpio base;
-};
-
 static int
 nv10_gpio_sense(struct nouveau_gpio *gpio, int line)
 {
@@ -83,95 +79,38 @@ nv10_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 }
 
 static void
-nv10_gpio_intr(struct nouveau_subdev *subdev)
-{
-	struct nv10_gpio_priv *priv = (void *)subdev;
-	u32 intr = nv_rd32(priv, 0x001104);
-	u32 hi = (intr & 0x0000ffff) >> 0;
-	u32 lo = (intr & 0xffff0000) >> 16;
-	int i;
-
-	for (i = 0; (hi | lo) && i < 32; i++) {
-		if ((hi | lo) & (1 << i))
-			nouveau_event_trigger(priv->base.events, i);
-	}
-
-	nv_wr32(priv, 0x001104, intr);
-}
-
-static void
-nv10_gpio_intr_enable(struct nouveau_event *event, int line)
-{
-	nv_wr32(event->priv, 0x001104, 0x00010001 << line);
-	nv_mask(event->priv, 0x001144, 0x00010001 << line, 0x00010001 << line);
-}
-
-static void
-nv10_gpio_intr_disable(struct nouveau_event *event, int line)
-{
-	nv_wr32(event->priv, 0x001104, 0x00010001 << line);
-	nv_mask(event->priv, 0x001144, 0x00010001 << line, 0x00000000);
-}
-
-static int
-nv10_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
+nv10_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
 {
-	struct nv10_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass, 16, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.drive = nv10_gpio_drive;
-	priv->base.sense = nv10_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nv10_gpio_intr_enable;
-	priv->base.events->disable = nv10_gpio_intr_disable;
-	nv_subdev(priv)->intr = nv10_gpio_intr;
-	return 0;
+	u32 intr = nv_rd32(gpio, 0x001104);
+	u32 stat = nv_rd32(gpio, 0x001144) & intr;
+	*lo = (stat & 0xffff0000) >> 16;
+	*hi = (stat & 0x0000ffff);
+	nv_wr32(gpio, 0x001104, intr);
 }
 
 static void
-nv10_gpio_dtor(struct nouveau_object *object)
-{
-	struct nv10_gpio_priv *priv = (void *)object;
-	nouveau_gpio_destroy(&priv->base);
-}
-
-static int
-nv10_gpio_init(struct nouveau_object *object)
-{
-	struct nv10_gpio_priv *priv = (void *)object;
-	int ret;
-
-	ret = nouveau_gpio_init(&priv->base);
-	if (ret)
-		return ret;
-
-	nv_wr32(priv, 0x001144, 0x00000000);
-	nv_wr32(priv, 0x001104, 0xffffffff);
-	return 0;
-}
-
-static int
-nv10_gpio_fini(struct nouveau_object *object, bool suspend)
+nv10_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
 {
-	struct nv10_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0x001144, 0x00000000);
-	return nouveau_gpio_fini(&priv->base, suspend);
+	u32 inte = nv_rd32(gpio, 0x001144);
+	if (type & NVKM_GPIO_LO)
+		inte = (inte & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte = (inte & ~mask) | data;
+	nv_wr32(gpio, 0x001144, inte);
 }
 
-struct nouveau_oclass
-nv10_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0x10),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv10_gpio_ctor,
-		.dtor = nv10_gpio_dtor,
-		.init = nv10_gpio_init,
-		.fini = nv10_gpio_fini,
+struct nouveau_oclass *
+nv10_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0x10),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 16,
+	.intr_stat = nv10_gpio_intr_stat,
+	.intr_mask = nv10_gpio_intr_mask,
+	.drive = nv10_gpio_drive,
+	.sense = nv10_gpio_sense,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
index 2ef774731629..1864fa98e6b1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
@@ -24,15 +24,10 @@
 
 #include "priv.h"
 
-struct nv50_gpio_priv {
-	struct nouveau_gpio base;
-};
-
-static void
+void
 nv50_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 {
 	struct nouveau_bios *bios = nouveau_bios(gpio);
-	struct nv50_gpio_priv *priv = (void *)gpio;
 	u8 ver, len;
 	u16 entry;
 	int ent = -1;
@@ -55,7 +50,7 @@ nv50_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 
 		gpio->set(gpio, 0, func, line, defs);
 
-		nv_mask(priv, reg, 0x00010001 << lsh, val << lsh);
+		nv_mask(gpio, reg, 0x00010001 << lsh, val << lsh);
 	}
 }
 
@@ -72,7 +67,7 @@ nv50_gpio_location(int line, u32 *reg, u32 *shift)
 	return 0;
 }
 
-static int
+int
 nv50_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 {
 	u32 reg, shift;
@@ -84,7 +79,7 @@ nv50_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 	return 0;
 }
 
-static int
+int
 nv50_gpio_sense(struct nouveau_gpio *gpio, int line)
 {
 	u32 reg, shift;
@@ -95,119 +90,40 @@ nv50_gpio_sense(struct nouveau_gpio *gpio, int line)
 	return !!(nv_rd32(gpio, reg) & (4 << shift));
 }
 
-void
-nv50_gpio_intr(struct nouveau_subdev *subdev)
-{
-	struct nv50_gpio_priv *priv = (void *)subdev;
-	u32 intr0, intr1 = 0;
-	u32 hi, lo;
-	int i;
-
-	intr0 = nv_rd32(priv, 0xe054) & nv_rd32(priv, 0xe050);
-	if (nv_device(priv)->chipset > 0x92)
-		intr1 = nv_rd32(priv, 0xe074) & nv_rd32(priv, 0xe070);
-
-	hi = (intr0 & 0x0000ffff) | (intr1 << 16);
-	lo = (intr0 >> 16) | (intr1 & 0xffff0000);
-
-	for (i = 0; (hi | lo) && i < 32; i++) {
-		if ((hi | lo) & (1 << i))
-			nouveau_event_trigger(priv->base.events, i);
-	}
-
-	nv_wr32(priv, 0xe054, intr0);
-	if (nv_device(priv)->chipset > 0x92)
-		nv_wr32(priv, 0xe074, intr1);
-}
-
-void
-nv50_gpio_intr_enable(struct nouveau_event *event, int line)
-{
-	const u32 addr = line < 16 ? 0xe050 : 0xe070;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x04, mask);
-	nv_mask(event->priv, addr + 0x00, mask, mask);
-}
-
-void
-nv50_gpio_intr_disable(struct nouveau_event *event, int line)
-{
-	const u32 addr = line < 16 ? 0xe050 : 0xe070;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x04, mask);
-	nv_mask(event->priv, addr + 0x00, mask, 0x00000000);
-}
-
-static int
-nv50_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
-{
-	struct nv50_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass,
-				  nv_device(parent)->chipset > 0x92 ? 32 : 16,
-				  &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.reset = nv50_gpio_reset;
-	priv->base.drive = nv50_gpio_drive;
-	priv->base.sense = nv50_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nv50_gpio_intr_enable;
-	priv->base.events->disable = nv50_gpio_intr_disable;
-	nv_subdev(priv)->intr = nv50_gpio_intr;
-	return 0;
-}
-
-void
-nv50_gpio_dtor(struct nouveau_object *object)
-{
-	struct nv50_gpio_priv *priv = (void *)object;
-	nouveau_gpio_destroy(&priv->base);
-}
-
-int
-nv50_gpio_init(struct nouveau_object *object)
+static void
+nv50_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
 {
-	struct nv50_gpio_priv *priv = (void *)object;
-	int ret;
-
-	ret = nouveau_gpio_init(&priv->base);
-	if (ret)
-		return ret;
-
-	/* disable, and ack any pending gpio interrupts */
-	nv_wr32(priv, 0xe050, 0x00000000);
-	nv_wr32(priv, 0xe054, 0xffffffff);
-	if (nv_device(priv)->chipset > 0x92) {
-		nv_wr32(priv, 0xe070, 0x00000000);
-		nv_wr32(priv, 0xe074, 0xffffffff);
-	}
-
-	return 0;
+	u32 intr = nv_rd32(gpio, 0x00e054);
+	u32 stat = nv_rd32(gpio, 0x00e050) & intr;
+	*lo = (stat & 0xffff0000) >> 16;
+	*hi = (stat & 0x0000ffff);
+	nv_wr32(gpio, 0x00e054, intr);
 }
 
-int
-nv50_gpio_fini(struct nouveau_object *object, bool suspend)
+static void
+nv50_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
 {
-	struct nv50_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0xe050, 0x00000000);
-	if (nv_device(priv)->chipset > 0x92)
-		nv_wr32(priv, 0xe070, 0x00000000);
-	return nouveau_gpio_fini(&priv->base, suspend);
+	u32 inte = nv_rd32(gpio, 0x00e050);
+	if (type & NVKM_GPIO_LO)
+		inte = (inte & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte = (inte & ~mask) | data;
+	nv_wr32(gpio, 0x00e050, inte);
 }
 
-struct nouveau_oclass
-nv50_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_gpio_ctor,
-		.dtor = nv50_gpio_dtor,
-		.init = nv50_gpio_init,
-		.fini = nv50_gpio_fini,
+struct nouveau_oclass *
+nv50_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 16,
+	.intr_stat = nv50_gpio_intr_stat,
+	.intr_mask = nv50_gpio_intr_mask,
+	.drive = nv50_gpio_drive,
+	.sense = nv50_gpio_sense,
+	.reset = nv50_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv92.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv92.c
new file mode 100644
index 000000000000..252083d376f5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv92.c
@@ -0,0 +1,74 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "priv.h"
+
+void
+nv92_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
+{
+	u32 intr0 = nv_rd32(gpio, 0x00e054);
+	u32 intr1 = nv_rd32(gpio, 0x00e074);
+	u32 stat0 = nv_rd32(gpio, 0x00e050) & intr0;
+	u32 stat1 = nv_rd32(gpio, 0x00e070) & intr1;
+	*lo = (stat1 & 0xffff0000) | (stat0 >> 16);
+	*hi = (stat1 << 16) | (stat0 & 0x0000ffff);
+	nv_wr32(gpio, 0x00e054, intr0);
+	nv_wr32(gpio, 0x00e074, intr1);
+}
+
+void
+nv92_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
+{
+	u32 inte0 = nv_rd32(gpio, 0x00e050);
+	u32 inte1 = nv_rd32(gpio, 0x00e070);
+	if (type & NVKM_GPIO_LO)
+		inte0 = (inte0 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte0 = (inte0 & ~(mask & 0xffff)) | (data & 0xffff);
+	mask >>= 16;
+	data >>= 16;
+	if (type & NVKM_GPIO_LO)
+		inte1 = (inte1 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte1 = (inte1 & ~mask) | data;
+	nv_wr32(gpio, 0x00e050, inte0);
+	nv_wr32(gpio, 0x00e070, inte1);
+}
+
+struct nouveau_oclass *
+nv92_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0x92),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
+	},
+	.lines = 32,
+	.intr_stat = nv92_gpio_intr_stat,
+	.intr_mask = nv92_gpio_intr_mask,
+	.drive = nv50_gpio_drive,
+	.sense = nv50_gpio_sense,
+	.reset = nv50_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
index 010431e3acec..a4682b0956ad 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
@@ -24,15 +24,10 @@
 
 #include "priv.h"
 
-struct nvd0_gpio_priv {
-	struct nouveau_gpio base;
-};
-
 void
 nvd0_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 {
 	struct nouveau_bios *bios = nouveau_bios(gpio);
-	struct nvd0_gpio_priv *priv = (void *)gpio;
 	u8 ver, len;
 	u16 entry;
 	int ent = -1;
@@ -51,9 +46,9 @@ nvd0_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 
 		gpio->set(gpio, 0, func, line, defs);
 
-		nv_mask(priv, 0x00d610 + (line * 4), 0xff, unk0);
+		nv_mask(gpio, 0x00d610 + (line * 4), 0xff, unk0);
 		if (unk1--)
-			nv_mask(priv, 0x00d740 + (unk1 * 4), 0xff, line);
+			nv_mask(gpio, 0x00d740 + (unk1 * 4), 0xff, line);
 	}
 }
 
@@ -72,36 +67,19 @@ nvd0_gpio_sense(struct nouveau_gpio *gpio, int line)
 	return !!(nv_rd32(gpio, 0x00d610 + (line * 4)) & 0x00004000);
 }
 
-static int
-nvd0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
-{
-	struct nvd0_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass, 32, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.reset = nvd0_gpio_reset;
-	priv->base.drive = nvd0_gpio_drive;
-	priv->base.sense = nvd0_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nv50_gpio_intr_enable;
-	priv->base.events->disable = nv50_gpio_intr_disable;
-	nv_subdev(priv)->intr = nv50_gpio_intr;
-	return 0;
-}
-
-struct nouveau_oclass
-nvd0_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0xd0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvd0_gpio_ctor,
-		.dtor = nv50_gpio_dtor,
-		.init = nv50_gpio_init,
-		.fini = nv50_gpio_fini,
+struct nouveau_oclass *
+nvd0_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0xd0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 32,
+	.intr_stat = nv92_gpio_intr_stat,
+	.intr_mask = nv92_gpio_intr_mask,
+	.drive = nvd0_gpio_drive,
+	.sense = nvd0_gpio_sense,
+	.reset = nvd0_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
index 16b8c5bf5efa..e1145b48c76c 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
@@ -24,108 +24,51 @@
 
 #include "priv.h"
 
-struct nve0_gpio_priv {
-	struct nouveau_gpio base;
-};
-
-void
-nve0_gpio_intr(struct nouveau_subdev *subdev)
+static void
+nve0_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
 {
-	struct nve0_gpio_priv *priv = (void *)subdev;
-	u32 intr0 = nv_rd32(priv, 0xdc00) & nv_rd32(priv, 0xdc08);
-	u32 intr1 = nv_rd32(priv, 0xdc80) & nv_rd32(priv, 0xdc88);
-	u32 hi = (intr0 & 0x0000ffff) | (intr1 << 16);
-	u32 lo = (intr0 >> 16) | (intr1 & 0xffff0000);
-	int i;
-
-	for (i = 0; (hi | lo) && i < 32; i++) {
-		if ((hi | lo) & (1 << i))
-			nouveau_event_trigger(priv->base.events, i);
-	}
-
-	nv_wr32(priv, 0xdc00, intr0);
-	nv_wr32(priv, 0xdc88, intr1);
+	u32 intr0 = nv_rd32(gpio, 0x00dc00);
+	u32 intr1 = nv_rd32(gpio, 0x00dc80);
+	u32 stat0 = nv_rd32(gpio, 0x00dc08) & intr0;
+	u32 stat1 = nv_rd32(gpio, 0x00dc88) & intr1;
+	*lo = (stat1 & 0xffff0000) | (stat0 >> 16);
+	*hi = (stat1 << 16) | (stat0 & 0x0000ffff);
+	nv_wr32(gpio, 0x00dc00, intr0);
+	nv_wr32(gpio, 0x00dc80, intr1);
 }
 
 void
-nve0_gpio_intr_enable(struct nouveau_event *event, int line)
+nve0_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
 {
-	const u32 addr = line < 16 ? 0xdc00 : 0xdc80;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x08, mask);
-	nv_mask(event->priv, addr + 0x00, mask, mask);
-}
-
-void
-nve0_gpio_intr_disable(struct nouveau_event *event, int line)
-{
-	const u32 addr = line < 16 ? 0xdc00 : 0xdc80;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x08, mask);
-	nv_mask(event->priv, addr + 0x00, mask, 0x00000000);
-}
-
-int
-nve0_gpio_fini(struct nouveau_object *object, bool suspend)
-{
-	struct nve0_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0xdc08, 0x00000000);
-	nv_wr32(priv, 0xdc88, 0x00000000);
-	return nouveau_gpio_fini(&priv->base, suspend);
-}
-
-int
-nve0_gpio_init(struct nouveau_object *object)
-{
-	struct nve0_gpio_priv *priv = (void *)object;
-	int ret;
-
-	ret = nouveau_gpio_init(&priv->base);
-	if (ret)
-		return ret;
-
-	nv_wr32(priv, 0xdc00, 0xffffffff);
-	nv_wr32(priv, 0xdc80, 0xffffffff);
-	return 0;
-}
-
-void
-nve0_gpio_dtor(struct nouveau_object *object)
-{
-	struct nve0_gpio_priv *priv = (void *)object;
-	nouveau_gpio_destroy(&priv->base);
-}
-
-static int
-nve0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
-{
-	struct nve0_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass, 32, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.reset = nvd0_gpio_reset;
-	priv->base.drive = nvd0_gpio_drive;
-	priv->base.sense = nvd0_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nve0_gpio_intr_enable;
-	priv->base.events->disable = nve0_gpio_intr_disable;
-	nv_subdev(priv)->intr = nve0_gpio_intr;
-	return 0;
+	u32 inte0 = nv_rd32(gpio, 0x00dc08);
+	u32 inte1 = nv_rd32(gpio, 0x00dc88);
+	if (type & NVKM_GPIO_LO)
+		inte0 = (inte0 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte0 = (inte0 & ~(mask & 0xffff)) | (data & 0xffff);
+	mask >>= 16;
+	data >>= 16;
+	if (type & NVKM_GPIO_LO)
+		inte1 = (inte1 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte1 = (inte1 & ~mask) | data;
+	nv_wr32(gpio, 0x00dc08, inte0);
+	nv_wr32(gpio, 0x00dc88, inte1);
 }
 
-struct nouveau_oclass
-nve0_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0xe0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nve0_gpio_ctor,
-		.dtor = nv50_gpio_dtor,
-		.init = nve0_gpio_init,
-		.fini = nve0_gpio_fini,
+struct nouveau_oclass *
+nve0_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0xe0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 32,
+	.intr_stat = nve0_gpio_intr_stat,
+	.intr_mask = nve0_gpio_intr_mask,
+	.drive = nvd0_gpio_drive,
+	.sense = nvd0_gpio_sense,
+	.reset = nvd0_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h b/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
index 2ee1c895c782..e1724dfc86ae 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
@@ -3,15 +3,65 @@
 
 #include <subdev/gpio.h>
 
-void nv50_gpio_dtor(struct nouveau_object *);
-int  nv50_gpio_init(struct nouveau_object *);
-int  nv50_gpio_fini(struct nouveau_object *, bool);
-void nv50_gpio_intr(struct nouveau_subdev *);
-void nv50_gpio_intr_enable(struct nouveau_event *, int line);
-void nv50_gpio_intr_disable(struct nouveau_event *, int line);
+#define nouveau_gpio_create(p,e,o,d)                                           \
+	nouveau_gpio_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_gpio_destroy(p) ({                                             \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_dtor(nv_object(gpio));                                   \
+})
+#define nouveau_gpio_init(p) ({                                                \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_init(nv_object(gpio));                                   \
+})
+#define nouveau_gpio_fini(p,s) ({                                              \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_fini(nv_object(gpio), (s));                              \
+})
+
+int  nouveau_gpio_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int, void **);
+int  _nouveau_gpio_ctor(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, void *, u32,
+			struct nouveau_object **);
+void _nouveau_gpio_dtor(struct nouveau_object *);
+int  _nouveau_gpio_init(struct nouveau_object *);
+int  _nouveau_gpio_fini(struct nouveau_object *, bool);
+
+struct nouveau_gpio_impl {
+	struct nouveau_oclass base;
+	int lines;
+
+	/* read and ack pending interrupts, returning only data
+	 * for lines that have not been masked off, while still
+	 * performing the ack for anything that was pending.
+	 */
+	void (*intr_stat)(struct nouveau_gpio *, u32 *, u32 *);
+
+	/* mask on/off interrupts for hi/lo transitions on a
+	 * given set of gpio lines
+	 */
+	void (*intr_mask)(struct nouveau_gpio *, u32, u32, u32);
+
+	/* configure gpio direction and output value */
+	int  (*drive)(struct nouveau_gpio *, int line, int dir, int out);
+
+	/* sense current state of given gpio line */
+	int  (*sense)(struct nouveau_gpio *, int line);
+
+	/*XXX*/
+	void (*reset)(struct nouveau_gpio *, u8);
+};
+
+void nv50_gpio_reset(struct nouveau_gpio *, u8);
+int  nv50_gpio_drive(struct nouveau_gpio *, int, int, int);
+int  nv50_gpio_sense(struct nouveau_gpio *, int);
+
+void nv92_gpio_intr_stat(struct nouveau_gpio *, u32 *, u32 *);
+void nv92_gpio_intr_mask(struct nouveau_gpio *, u32, u32, u32);
 
 void nvd0_gpio_reset(struct nouveau_gpio *, u8);
 int  nvd0_gpio_drive(struct nouveau_gpio *, int, int, int);
 int  nvd0_gpio_sense(struct nouveau_gpio *, int);
 
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
index 4b195ac4da66..2c2731a6cf91 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs <bskeggs@redhat.com>
  */
 
-#include <subdev/i2c.h>
+#include "port.h"
 
 struct anx9805_i2c_port {
 	struct nouveau_i2c_port base;
@@ -37,6 +37,8 @@ anx9805_train(struct nouveau_i2c_port *port, int link_nr, int link_bw, bool enh)
 	struct nouveau_i2c_port *mast = (void *)nv_object(chan)->parent;
 	u8 tmp, i;
 
+	DBG("ANX9805 train %d 0x%02x %d\n", link_nr, link_bw, enh);
+
 	nv_wri2cr(mast, chan->addr, 0xa0, link_bw);
 	nv_wri2cr(mast, chan->addr, 0xa1, link_nr | (enh ? 0x80 : 0x00));
 	nv_wri2cr(mast, chan->addr, 0xa2, 0x01);
@@ -60,21 +62,29 @@ anx9805_train(struct nouveau_i2c_port *port, int link_nr, int link_bw, bool enh)
 }
 
 static int
-anx9805_aux(struct nouveau_i2c_port *port, u8 type, u32 addr, u8 *data, u8 size)
+anx9805_aux(struct nouveau_i2c_port *port, bool retry,
+	    u8 type, u32 addr, u8 *data, u8 size)
 {
 	struct anx9805_i2c_port *chan = (void *)port;
 	struct nouveau_i2c_port *mast = (void *)nv_object(chan)->parent;
 	int i, ret = -ETIMEDOUT;
+	u8 buf[16] = {};
 	u8 tmp;
 
+	DBG("%02x %05x %d\n", type, addr, size);
+
 	tmp = nv_rdi2cr(mast, chan->ctrl, 0x07) & ~0x04;
 	nv_wri2cr(mast, chan->ctrl, 0x07, tmp | 0x04);
 	nv_wri2cr(mast, chan->ctrl, 0x07, tmp);
 	nv_wri2cr(mast, chan->ctrl, 0xf7, 0x01);
 
 	nv_wri2cr(mast, chan->addr, 0xe4, 0x80);
-	for (i = 0; !(type & 1) && i < size; i++)
-		nv_wri2cr(mast, chan->addr, 0xf0 + i, data[i]);
+	if (!(type & 1)) {
+		memcpy(buf, data, size);
+		DBG("%16ph", buf);
+		for (i = 0; i < size; i++)
+			nv_wri2cr(mast, chan->addr, 0xf0 + i, buf[i]);
+	}
 	nv_wri2cr(mast, chan->addr, 0xe5, ((size - 1) << 4) | type);
 	nv_wri2cr(mast, chan->addr, 0xe6, (addr & 0x000ff) >>  0);
 	nv_wri2cr(mast, chan->addr, 0xe7, (addr & 0x0ff00) >>  8);
@@ -93,8 +103,13 @@ anx9805_aux(struct nouveau_i2c_port *port, u8 type, u32 addr, u8 *data, u8 size)
 		goto done;
 	}
 
-	for (i = 0; (type & 1) && i < size; i++)
-		data[i] = nv_rdi2cr(mast, chan->addr, 0xf0 + i);
+	if (type & 1) {
+		for (i = 0; i < size; i++)
+			buf[i] = nv_rdi2cr(mast, chan->addr, 0xf0 + i);
+		DBG("%16ph", buf);
+		memcpy(data, buf, size);
+	}
+
 	ret = 0;
 done:
 	nv_wri2cr(mast, chan->ctrl, 0xf7, 0x01);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
index 5de074ad170b..02eb42be2e9e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
@@ -22,15 +22,19 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/i2c.h>
+#include "priv.h"
 
 int
 nv_rdaux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 {
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
 	if (port->func->aux) {
-		if (port->func->acquire)
-			port->func->acquire(port);
-		return port->func->aux(port, 9, addr, data, size);
+		int ret = i2c->acquire(port, 0);
+		if (ret == 0) {
+			ret = port->func->aux(port, true, 9, addr, data, size);
+			i2c->release(port);
+		}
+		return ret;
 	}
 	return -ENODEV;
 }
@@ -38,10 +42,14 @@ nv_rdaux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 int
 nv_wraux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 {
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
 	if (port->func->aux) {
-		if (port->func->acquire)
-			port->func->acquire(port);
-		return port->func->aux(port, 8, addr, data, size);
+		int ret = i2c->acquire(port, 0);
+		if (ret == 0) {
+			ret = port->func->aux(port, true, 8, addr, data, size);
+			i2c->release(port);
+		}
+		return ret;
 	}
 	return -ENODEV;
 }
@@ -50,13 +58,16 @@ static int
 aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 {
 	struct nouveau_i2c_port *port = adap->algo_data;
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
 	struct i2c_msg *msg = msgs;
 	int ret, mcnt = num;
 
 	if (!port->func->aux)
 		return -ENODEV;
-	if ( port->func->acquire)
-		port->func->acquire(port);
+
+	ret = i2c->acquire(port, 0);
+	if (ret)
+		return ret;
 
 	while (mcnt--) {
 		u8 remaining = msg->len;
@@ -74,9 +85,11 @@ aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 			if (mcnt || remaining > 16)
 				cmd |= 4; /* MOT */
 
-			ret = port->func->aux(port, cmd, msg->addr, ptr, cnt);
-			if (ret < 0)
+			ret = port->func->aux(port, true, cmd, msg->addr, ptr, cnt);
+			if (ret < 0) {
+				i2c->release(port);
 				return ret;
+			}
 
 			ptr += cnt;
 			remaining -= cnt;
@@ -85,6 +98,7 @@ aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 		msg++;
 	}
 
+	i2c->release(port);
 	return num;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
index 378e05b88e6f..09ba2cc851cf 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
@@ -23,13 +23,16 @@
  */
 
 #include <core/option.h>
+#include <core/event.h>
 
 #include <subdev/bios.h>
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/i2c.h>
-#include <subdev/i2c.h>
 #include <subdev/vga.h>
 
+#include "priv.h"
+#include "pad.h"
+
 /******************************************************************************
  * interface to linux i2c bit-banging algorithm
  *****************************************************************************/
@@ -45,9 +48,15 @@ nouveau_i2c_pre_xfer(struct i2c_adapter *adap)
 {
 	struct i2c_algo_bit_data *bit = adap->algo_data;
 	struct nouveau_i2c_port *port = bit->data;
-	if (port->func->acquire)
-		port->func->acquire(port);
-	return 0;
+	return nouveau_i2c(port)->acquire(port, bit->timeout);
+}
+
+static void
+nouveau_i2c_post_xfer(struct i2c_adapter *adap)
+{
+	struct i2c_algo_bit_data *bit = adap->algo_data;
+	struct nouveau_i2c_port *port = bit->data;
+	return nouveau_i2c(port)->release(port);
 }
 
 static void
@@ -82,6 +91,15 @@ nouveau_i2c_getsda(void *data)
  * base i2c "port" class implementation
  *****************************************************************************/
 
+int
+_nouveau_i2c_port_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_i2c_port *port = (void *)object;
+	struct nvkm_i2c_pad *pad = nvkm_i2c_pad(port);
+	nv_ofuncs(pad)->fini(nv_object(pad), suspend);
+	return nouveau_object_fini(&port->base, suspend);
+}
+
 void
 _nouveau_i2c_port_dtor(struct nouveau_object *object)
 {
@@ -98,7 +116,7 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 			 const struct nouveau_i2c_func *func,
 			 int size, void **pobject)
 {
-	struct nouveau_device *device = nv_device(parent);
+	struct nouveau_device *device = nv_device(engine);
 	struct nouveau_i2c *i2c = (void *)engine;
 	struct nouveau_i2c_port *port;
 	int ret;
@@ -113,8 +131,9 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 	port->adapter.owner = THIS_MODULE;
 	port->adapter.dev.parent = nv_device_base(device);
 	port->index = index;
+	port->aux = -1;
 	port->func = func;
-	i2c_set_adapdata(&port->adapter, i2c);
+	mutex_init(&port->mutex);
 
 	if ( algo == &nouveau_i2c_bit_algo &&
 	    !nouveau_boolopt(device->cfgopt, "NvI2C", CSTMSEL)) {
@@ -128,6 +147,7 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 		bit->timeout = usecs_to_jiffies(2200);
 		bit->data = port;
 		bit->pre_xfer = nouveau_i2c_pre_xfer;
+		bit->post_xfer = nouveau_i2c_post_xfer;
 		bit->setsda = nouveau_i2c_setsda;
 		bit->setscl = nouveau_i2c_setscl;
 		bit->getsda = nouveau_i2c_getsda;
@@ -141,7 +161,6 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 		ret = i2c_add_adapter(&port->adapter);
 	}
 
-	/* drop port's i2c subdev refcount, i2c handles this itself */
 	if (ret == 0)
 		list_add_tail(&port->head, &i2c->ports);
 	return ret;
@@ -193,6 +212,75 @@ nouveau_i2c_find_type(struct nouveau_i2c *i2c, u16 type)
 	return NULL;
 }
 
+static void
+nouveau_i2c_release_pad(struct nouveau_i2c_port *port)
+{
+	struct nvkm_i2c_pad *pad = nvkm_i2c_pad(port);
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
+
+	if (atomic_dec_and_test(&nv_object(pad)->usecount)) {
+		nv_ofuncs(pad)->fini(nv_object(pad), false);
+		wake_up_all(&i2c->wait);
+	}
+}
+
+static int
+nouveau_i2c_try_acquire_pad(struct nouveau_i2c_port *port)
+{
+	struct nvkm_i2c_pad *pad = nvkm_i2c_pad(port);
+
+	if (atomic_add_return(1, &nv_object(pad)->usecount) != 1) {
+		struct nouveau_object *owner = (void *)pad->port;
+		do {
+			if (owner == (void *)port)
+				return 0;
+			owner = owner->parent;
+		} while(owner);
+		nouveau_i2c_release_pad(port);
+		return -EBUSY;
+	}
+
+	pad->next = port;
+	nv_ofuncs(pad)->init(nv_object(pad));
+	return 0;
+}
+
+static int
+nouveau_i2c_acquire_pad(struct nouveau_i2c_port *port, unsigned long timeout)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
+
+	if (timeout) {
+		if (wait_event_timeout(i2c->wait,
+				       nouveau_i2c_try_acquire_pad(port) == 0,
+				       timeout) == 0)
+			return -EBUSY;
+	} else {
+		wait_event(i2c->wait, nouveau_i2c_try_acquire_pad(port) == 0);
+	}
+
+	return 0;
+}
+
+static void
+nouveau_i2c_release(struct nouveau_i2c_port *port)
+__releases(pad->mutex)
+{
+	nouveau_i2c(port)->release_pad(port);
+	mutex_unlock(&port->mutex);
+}
+
+static int
+nouveau_i2c_acquire(struct nouveau_i2c_port *port, unsigned long timeout)
+__acquires(pad->mutex)
+{
+	int ret;
+	mutex_lock(&port->mutex);
+	if ((ret = nouveau_i2c(port)->acquire_pad(port, timeout)))
+		mutex_unlock(&port->mutex);
+	return ret;
+}
+
 static int
 nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
 		     struct nouveau_i2c_board_info *info,
@@ -237,11 +325,59 @@ nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
 	return -ENODEV;
 }
 
+static void
+nouveau_i2c_intr_disable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(event->priv);
+	struct nouveau_i2c_port *port = i2c->find(i2c, index);
+	const struct nouveau_i2c_impl *impl = (void *)nv_object(i2c)->oclass;
+	if (port && port->aux >= 0)
+		impl->aux_mask(i2c, type, 1 << port->aux, 0);
+}
+
+static void
+nouveau_i2c_intr_enable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(event->priv);
+	struct nouveau_i2c_port *port = i2c->find(i2c, index);
+	const struct nouveau_i2c_impl *impl = (void *)nv_object(i2c)->oclass;
+	if (port && port->aux >= 0)
+		impl->aux_mask(i2c, type, 1 << port->aux, 1 << port->aux);
+}
+
+static void
+nouveau_i2c_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_i2c_impl *impl = (void *)nv_oclass(subdev);
+	struct nouveau_i2c *i2c = nouveau_i2c(subdev);
+	struct nouveau_i2c_port *port;
+	u32 hi, lo, rq, tx, e;
+
+	if (impl->aux_stat) {
+		impl->aux_stat(i2c, &hi, &lo, &rq, &tx);
+		if (hi || lo || rq || tx) {
+			list_for_each_entry(port, &i2c->ports, head) {
+				if (e = 0, port->aux < 0)
+					continue;
+
+				if (hi & (1 << port->aux)) e |= NVKM_I2C_PLUG;
+				if (lo & (1 << port->aux)) e |= NVKM_I2C_UNPLUG;
+				if (rq & (1 << port->aux)) e |= NVKM_I2C_IRQ;
+				if (tx & (1 << port->aux)) e |= NVKM_I2C_DONE;
+
+				nouveau_event_trigger(i2c->ntfy, e, port->index);
+			}
+		}
+	}
+}
+
 int
 _nouveau_i2c_fini(struct nouveau_object *object, bool suspend)
 {
+	struct nouveau_i2c_impl *impl = (void *)nv_oclass(object);
 	struct nouveau_i2c *i2c = (void *)object;
 	struct nouveau_i2c_port *port;
+	u32 mask;
 	int ret;
 
 	list_for_each_entry(port, &i2c->ports, head) {
@@ -250,6 +386,11 @@ _nouveau_i2c_fini(struct nouveau_object *object, bool suspend)
 			goto fail;
 	}
 
+	if ((mask = (1 << impl->aux) - 1), impl->aux_stat) {
+		impl->aux_mask(i2c, NVKM_I2C_ANY, mask, 0);
+		impl->aux_stat(i2c, &mask, &mask, &mask, &mask);
+	}
+
 	return nouveau_subdev_fini(&i2c->base, suspend);
 fail:
 	list_for_each_entry_continue_reverse(port, &i2c->ports, head) {
@@ -290,6 +431,8 @@ _nouveau_i2c_dtor(struct nouveau_object *object)
 	struct nouveau_i2c *i2c = (void *)object;
 	struct nouveau_i2c_port *port, *temp;
 
+	nouveau_event_destroy(&i2c->ntfy);
+
 	list_for_each_entry_safe(port, temp, &i2c->ports, head) {
 		nouveau_object_ref(NULL, (struct nouveau_object **)&port);
 	}
@@ -306,14 +449,14 @@ int
 nouveau_i2c_create_(struct nouveau_object *parent,
 		    struct nouveau_object *engine,
 		    struct nouveau_oclass *oclass,
-		    struct nouveau_oclass *sclass,
 		    int length, void **pobject)
 {
+	const struct nouveau_i2c_impl *impl = (void *)oclass;
 	struct nouveau_bios *bios = nouveau_bios(parent);
 	struct nouveau_i2c *i2c;
 	struct nouveau_object *object;
 	struct dcb_i2c_entry info;
-	int ret, i, j, index = -1;
+	int ret, i, j, index = -1, pad;
 	struct dcb_output outp;
 	u8  ver, hdr;
 	u32 data;
@@ -324,24 +467,48 @@ nouveau_i2c_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
+	nv_subdev(i2c)->intr = nouveau_i2c_intr;
 	i2c->find = nouveau_i2c_find;
 	i2c->find_type = nouveau_i2c_find_type;
+	i2c->acquire_pad = nouveau_i2c_acquire_pad;
+	i2c->release_pad = nouveau_i2c_release_pad;
+	i2c->acquire = nouveau_i2c_acquire;
+	i2c->release = nouveau_i2c_release;
 	i2c->identify = nouveau_i2c_identify;
+	init_waitqueue_head(&i2c->wait);
 	INIT_LIST_HEAD(&i2c->ports);
 
 	while (!dcb_i2c_parse(bios, ++index, &info)) {
 		if (info.type == DCB_I2C_UNUSED)
 			continue;
 
-		oclass = sclass;
+		if (info.share != DCB_I2C_UNUSED) {
+			if (info.type == DCB_I2C_NVIO_AUX)
+				pad = info.drive;
+			else
+				pad = info.share;
+			oclass = impl->pad_s;
+		} else {
+			pad = 0x100 + info.drive;
+			oclass = impl->pad_x;
+		}
+
+		ret = nouveau_object_ctor(NULL, *pobject, oclass,
+					  NULL, pad, &parent);
+		if (ret < 0)
+			continue;
+
+		oclass = impl->sclass;
 		do {
 			ret = -EINVAL;
 			if (oclass->handle == info.type) {
-				ret = nouveau_object_ctor(*pobject, *pobject,
+				ret = nouveau_object_ctor(parent, *pobject,
 							  oclass, &info,
 							  index, &object);
 			}
 		} while (ret && (++oclass)->handle);
+
+		nouveau_object_ref(NULL, &parent);
 	}
 
 	/* in addition to the busses specified in the i2c table, there
@@ -380,5 +547,28 @@ nouveau_i2c_create_(struct nouveau_object *parent,
 		}
 	}
 
+	ret = nouveau_event_create(4, index, &i2c->ntfy);
+	if (ret)
+		return ret;
+
+	i2c->ntfy->priv = i2c;
+	i2c->ntfy->enable = nouveau_i2c_intr_enable;
+	i2c->ntfy->disable = nouveau_i2c_intr_disable;
+	return 0;
+}
+
+int
+_nouveau_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **pobject)
+{
+	struct nouveau_i2c *i2c;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, &i2c);
+	*pobject = nv_object(i2c);
+	if (ret)
+		return ret;
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
index a6e72d3b06b5..813ffc96e864 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include "subdev/i2c.h"
+#include "priv.h"
 
 #ifdef CONFIG_NOUVEAU_I2C_INTERNAL
 #define T_TIMEOUT  2200000
@@ -187,8 +187,9 @@ i2c_bit_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 	struct i2c_msg *msg = msgs;
 	int ret = 0, mcnt = num;
 
-	if (port->func->acquire)
-		port->func->acquire(port);
+	ret = nouveau_i2c(port)->acquire(port, nsecs_to_jiffies(T_TIMEOUT));
+	if (ret)
+		return ret;
 
 	while (!ret && mcnt--) {
 		u8 remaining = msg->len;
@@ -210,6 +211,7 @@ i2c_bit_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 	}
 
 	i2c_stop(port);
+	nouveau_i2c(port)->release(port);
 	return (ret < 0) ? ret : num;
 }
 #else
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/gf117.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/gf117.c
new file mode 100644
index 000000000000..fa891c39866b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/gf117.c
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+struct nouveau_oclass *
+gf117_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0xd7),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+	.sclass = nvd0_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
index 860d5d2365da..b1725bdea967 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
@@ -22,9 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/i2c.h>
 #include <subdev/vga.h>
 
+#include "priv.h"
+
 struct nv04_i2c_priv {
 	struct nouveau_i2c base;
 };
@@ -115,29 +116,15 @@ nv04_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv04_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv04_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv04_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv04_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x04),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv04_i2c_ctor,
+struct nouveau_oclass *
+nv04_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x04),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv04_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
index 0c2655a03bb4..f16c87ce5ba1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
@@ -22,9 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/i2c.h>
 #include <subdev/vga.h>
 
+#include "priv.h"
+
 struct nv4e_i2c_priv {
 	struct nouveau_i2c base;
 };
@@ -107,29 +108,15 @@ nv4e_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv4e_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv4e_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv4e_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv4e_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x4e),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv4e_i2c_ctor,
+struct nouveau_oclass *
+nv4e_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x4e),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv4e_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
index a8d67a287704..7b8756d4df08 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
@@ -121,29 +121,15 @@ nv50_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv50_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv50_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv50_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_i2c_ctor,
+struct nouveau_oclass *
+nv50_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv50_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
index 4e5ba48ebf5a..5d2a77421c74 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
@@ -1,7 +1,7 @@
 #ifndef __NV50_I2C_H__
 #define __NV50_I2C_H__
 
-#include <subdev/i2c.h>
+#include "priv.h"
 
 struct nv50_i2c_priv {
 	struct nouveau_i2c base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
index df6d3e4b68be..f59c3a255462 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
@@ -24,6 +24,36 @@
 
 #include "nv50.h"
 
+void
+nv94_aux_stat(struct nouveau_i2c *i2c, u32 *hi, u32 *lo, u32 *rq, u32 *tx)
+{
+	u32 intr = nv_rd32(i2c, 0x00e06c);
+	u32 stat = nv_rd32(i2c, 0x00e068) & intr, i;
+	for (i = 0, *hi = *lo = *rq = *tx = 0; i < 8; i++) {
+		if ((stat & (1 << (i * 4)))) *hi |= 1 << i;
+		if ((stat & (2 << (i * 4)))) *lo |= 1 << i;
+		if ((stat & (4 << (i * 4)))) *rq |= 1 << i;
+		if ((stat & (8 << (i * 4)))) *tx |= 1 << i;
+	}
+	nv_wr32(i2c, 0x00e06c, intr);
+}
+
+void
+nv94_aux_mask(struct nouveau_i2c *i2c, u32 type, u32 mask, u32 data)
+{
+	u32 temp = nv_rd32(i2c, 0x00e068), i;
+	for (i = 0; i < 8; i++) {
+		if (mask & (1 << i)) {
+			if (!(data & (1 << i))) {
+				temp &= ~(type << (i * 4));
+				continue;
+			}
+			temp |= type << (i * 4);
+		}
+	}
+	nv_wr32(i2c, 0x00e068, temp);
+}
+
 #define AUX_DBG(fmt, args...) nv_debug(aux, "AUXCH(%d): " fmt, ch, ##args)
 #define AUX_ERR(fmt, args...) nv_error(aux, "AUXCH(%d): " fmt, ch, ##args)
 
@@ -69,7 +99,8 @@ auxch_init(struct nouveau_i2c *aux, int ch)
 }
 
 int
-nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
+nv94_aux(struct nouveau_i2c_port *base, bool retry,
+	 u8 type, u32 addr, u8 *data, u8 size)
 {
 	struct nouveau_i2c *aux = nouveau_i2c(base);
 	struct nv50_i2c_port *port = (void *)base;
@@ -105,9 +136,8 @@ nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
 	ctrl |= size - 1;
 	nv_wr32(aux, 0x00e4e0 + (ch * 0x50), addr);
 
-	/* retry transaction a number of times on failure... */
-	ret = -EREMOTEIO;
-	for (retries = 0; retries < 32; retries++) {
+	/* (maybe) retry transaction a number of times on failure... */
+	for (retries = 0; !ret && retries < 32; retries++) {
 		/* reset, and delay a while if this is a retry */
 		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x80000000 | ctrl);
 		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x00000000 | ctrl);
@@ -123,16 +153,21 @@ nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
 			udelay(1);
 			if (!timeout--) {
 				AUX_ERR("tx req timeout 0x%08x\n", ctrl);
+				ret = -EIO;
 				goto out;
 			}
 		} while (ctrl & 0x00010000);
+		ret = 1;
 
 		/* read status, and check if transaction completed ok */
 		stat = nv_mask(aux, 0x00e4e8 + (ch * 0x50), 0, 0);
-		if (!(stat & 0x000f0f00)) {
-			ret = 0;
-			break;
-		}
+		if ((stat & 0x000f0000) == 0x00080000 ||
+		    (stat & 0x000f0000) == 0x00020000)
+			ret = retry ? 0 : 1;
+		if ((stat & 0x00000100))
+			ret = -ETIMEDOUT;
+		if ((stat & 0x00000e00))
+			ret = -EIO;
 
 		AUX_DBG("%02d 0x%08x 0x%08x\n", retries, ctrl, stat);
 	}
@@ -147,29 +182,11 @@ nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
 
 out:
 	auxch_fini(aux, ch);
-	return ret;
-}
-
-void
-nv94_i2c_acquire(struct nouveau_i2c_port *base)
-{
-	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
-	struct nv50_i2c_port *port = (void *)base;
-	if (port->ctrl) {
-		nv_mask(priv, port->ctrl + 0x0c, 0x00000001, 0x00000000);
-		nv_mask(priv, port->ctrl + 0x00, 0x0000f003, port->data);
-	}
-}
-
-void
-nv94_i2c_release(struct nouveau_i2c_port *base)
-{
+	return ret < 0 ? ret : (stat & 0x000f0000) >> 16;
 }
 
 static const struct nouveau_i2c_func
 nv94_i2c_func = {
-	.acquire   = nv94_i2c_acquire,
-	.release   = nv94_i2c_release,
 	.drive_scl = nv50_i2c_drive_scl,
 	.drive_sda = nv50_i2c_drive_sda,
 	.sense_scl = nv50_i2c_sense_scl,
@@ -206,8 +223,6 @@ nv94_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 
 static const struct nouveau_i2c_func
 nv94_aux_func = {
-	.acquire   = nv94_i2c_acquire,
-	.release   = nv94_i2c_release,
 	.aux       = nv94_aux,
 };
 
@@ -227,6 +242,7 @@ nv94_aux_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
+	port->base.aux = info->drive;
 	port->addr = info->drive;
 	if (info->share != DCB_I2C_UNUSED) {
 		port->ctrl = 0x00e500 + (info->drive * 0x50);
@@ -257,29 +273,19 @@ nv94_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv94_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv94_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv94_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x94),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv94_i2c_ctor,
+struct nouveau_oclass *
+nv94_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x94),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv94_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv94_i2c_pad_oclass,
+	.aux = 4,
+	.aux_stat = nv94_aux_stat,
+	.aux_mask = nv94_aux_mask,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
index 29967d30f97c..364ddb1c5f03 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
@@ -42,8 +42,6 @@ nvd0_i2c_sense_sda(struct nouveau_i2c_port *base)
 
 static const struct nouveau_i2c_func
 nvd0_i2c_func = {
-	.acquire   = nv94_i2c_acquire,
-	.release   = nv94_i2c_release,
 	.drive_scl = nv50_i2c_drive_scl,
 	.drive_sda = nv50_i2c_drive_sda,
 	.sense_scl = nvd0_i2c_sense_scl,
@@ -75,7 +73,7 @@ nvd0_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-static struct nouveau_oclass
+struct nouveau_oclass
 nvd0_i2c_sclass[] = {
 	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT),
 	  .ofuncs = &(struct nouveau_ofuncs) {
@@ -96,29 +94,19 @@ nvd0_i2c_sclass[] = {
 	{}
 };
 
-static int
-nvd0_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nvd0_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nvd0_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0xd0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvd0_i2c_ctor,
+struct nouveau_oclass *
+nvd0_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0xd0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nvd0_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv94_i2c_pad_oclass,
+	.aux = 4,
+	.aux_stat = nv94_aux_stat,
+	.aux_mask = nv94_aux_mask,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nve0.c
new file mode 100644
index 000000000000..cae77e1ad8dc
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nve0.c
@@ -0,0 +1,72 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+static void
+nve0_aux_stat(struct nouveau_i2c *i2c, u32 *hi, u32 *lo, u32 *rq, u32 *tx)
+{
+	u32 intr = nv_rd32(i2c, 0x00dc60);
+	u32 stat = nv_rd32(i2c, 0x00dc68) & intr, i;
+	for (i = 0, *hi = *lo = *rq = *tx = 0; i < 8; i++) {
+		if ((stat & (1 << (i * 4)))) *hi |= 1 << i;
+		if ((stat & (2 << (i * 4)))) *lo |= 1 << i;
+		if ((stat & (4 << (i * 4)))) *rq |= 1 << i;
+		if ((stat & (8 << (i * 4)))) *tx |= 1 << i;
+	}
+	nv_wr32(i2c, 0x00dc60, intr);
+}
+
+static void
+nve0_aux_mask(struct nouveau_i2c *i2c, u32 type, u32 mask, u32 data)
+{
+	u32 temp = nv_rd32(i2c, 0x00dc68), i;
+	for (i = 0; i < 8; i++) {
+		if (mask & (1 << i)) {
+			if (!(data & (1 << i))) {
+				temp &= ~(type << (i * 4));
+				continue;
+			}
+			temp |= type << (i * 4);
+		}
+	}
+	nv_wr32(i2c, 0x00dc68, temp);
+}
+
+struct nouveau_oclass *
+nve0_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0xe0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+	.sclass = nvd0_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv94_i2c_pad_oclass,
+	.aux = 4,
+	.aux_stat = nve0_aux_stat,
+	.aux_mask = nve0_aux_mask,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.c
new file mode 100644
index 000000000000..e9e412477c12
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.c
@@ -0,0 +1,84 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "pad.h"
+
+int
+_nvkm_i2c_pad_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_i2c_pad *pad = (void *)object;
+	DBG("-> NULL\n");
+	pad->port = NULL;
+	return nouveau_object_fini(&pad->base, suspend);
+}
+
+int
+_nvkm_i2c_pad_init(struct nouveau_object *object)
+{
+	struct nvkm_i2c_pad *pad = (void *)object;
+	DBG("-> PORT:%02x\n", pad->next->index);
+	pad->port = pad->next;
+	return nouveau_object_init(&pad->base);
+}
+
+int
+nvkm_i2c_pad_create_(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, int index,
+		     int size, void **pobject)
+{
+	struct nouveau_i2c *i2c = (void *)engine;
+	struct nouveau_i2c_port *port;
+	struct nvkm_i2c_pad *pad;
+	int ret;
+
+	list_for_each_entry(port, &i2c->ports, head) {
+		pad = nvkm_i2c_pad(port);
+		if (pad->index == index) {
+			atomic_inc(&nv_object(pad)->refcount);
+			*pobject = pad;
+			return 1;
+		}
+	}
+
+	ret = nouveau_object_create_(parent, engine, oclass, 0, size, pobject);
+	pad = *pobject;
+	if (ret)
+		return ret;
+
+	pad->index = index;
+	return 0;
+}
+
+int
+_nvkm_i2c_pad_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct nvkm_i2c_pad *pad;
+	int ret;
+	ret = nvkm_i2c_pad_create(parent, engine, oclass, index, &pad);
+	*pobject = nv_object(pad);
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.h
new file mode 100644
index 000000000000..452ac10c3004
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.h
@@ -0,0 +1,58 @@
+#ifndef __NVKM_I2C_PAD_H__
+#define __NVKM_I2C_PAD_H__
+
+#include "priv.h"
+
+struct nvkm_i2c_pad {
+	struct nouveau_object base;
+	int index;
+	struct nouveau_i2c_port *port;
+	struct nouveau_i2c_port *next;
+};
+
+static inline struct nvkm_i2c_pad *
+nvkm_i2c_pad(struct nouveau_i2c_port *port)
+{
+	struct nouveau_object *pad = nv_object(port);
+	while (pad->parent)
+		pad = pad->parent;
+	return (void *)pad;
+}
+
+#define nvkm_i2c_pad_create(p,e,o,i,d)                                         \
+	nvkm_i2c_pad_create_((p), (e), (o), (i), sizeof(**d), (void **)d)
+#define nvkm_i2c_pad_destroy(p) ({                                             \
+	struct nvkm_i2c_pad *_p = (p);                                         \
+	_nvkm_i2c_pad_dtor(nv_object(_p));                                     \
+})
+#define nvkm_i2c_pad_init(p) ({                                                \
+	struct nvkm_i2c_pad *_p = (p);                                         \
+	_nvkm_i2c_pad_init(nv_object(_p));                                     \
+})
+#define nvkm_i2c_pad_fini(p,s) ({                                              \
+	struct nvkm_i2c_pad *_p = (p);                                         \
+	_nvkm_i2c_pad_fini(nv_object(_p), (s));                                \
+})
+
+int nvkm_i2c_pad_create_(struct nouveau_object *, struct nouveau_object *,
+			 struct nouveau_oclass *, int index, int, void **);
+
+int _nvkm_i2c_pad_ctor(struct nouveau_object *, struct nouveau_object *,
+		       struct nouveau_oclass *, void *, u32,
+		       struct nouveau_object **);
+#define _nvkm_i2c_pad_dtor nouveau_object_destroy
+int _nvkm_i2c_pad_init(struct nouveau_object *);
+int _nvkm_i2c_pad_fini(struct nouveau_object *, bool);
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nvkm_i2c_pad *_pad = (void *)pad;                               \
+	nv_##l(nv_object(_pad)->engine, "PAD:%c:%02x: "f,                      \
+	       _pad->index >= 0x100 ? 'X' : 'S',                               \
+	       _pad->index >= 0x100 ? _pad->index - 0x100 : _pad->index, ##a); \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv04.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv04.c
new file mode 100644
index 000000000000..2c4b61296dd1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv04.c
@@ -0,0 +1,35 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "pad.h"
+
+struct nouveau_oclass
+nv04_i2c_pad_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nvkm_i2c_pad_ctor,
+		.dtor = _nvkm_i2c_pad_dtor,
+		.init = _nvkm_i2c_pad_init,
+		.fini = _nvkm_i2c_pad_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv94.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv94.c
new file mode 100644
index 000000000000..0dc6753014f0
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv94.c
@@ -0,0 +1,86 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "pad.h"
+
+struct nv94_i2c_pad {
+	struct nvkm_i2c_pad base;
+	int addr;
+};
+
+static int
+nv94_i2c_pad_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_i2c *i2c = (void *)object->engine;
+	struct nv94_i2c_pad *pad = (void *)object;
+	nv_mask(i2c, 0x00e50c + pad->addr, 0x00000001, 0x00000001);
+	return nvkm_i2c_pad_fini(&pad->base, suspend);
+}
+
+static int
+nv94_i2c_pad_init(struct nouveau_object *object)
+{
+	struct nouveau_i2c *i2c = (void *)object->engine;
+	struct nv94_i2c_pad *pad = (void *)object;
+
+	switch (nv_oclass(pad->base.next)->handle) {
+	case NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_AUX):
+		nv_mask(i2c, 0x00e500 + pad->addr, 0x0000c003, 0x00000002);
+		break;
+	case NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT):
+	default:
+		nv_mask(i2c, 0x00e500 + pad->addr, 0x0000c003, 0x0000c001);
+		break;
+	}
+
+	nv_mask(i2c, 0x00e50c + pad->addr, 0x00000001, 0x00000000);
+	return nvkm_i2c_pad_init(&pad->base);
+}
+
+static int
+nv94_i2c_pad_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 index,
+		  struct nouveau_object **pobject)
+{
+	struct nv94_i2c_pad *pad;
+	int ret;
+
+	ret = nvkm_i2c_pad_create(parent, engine, oclass, index, &pad);
+	*pobject = nv_object(pad);
+	if (ret)
+		return ret;
+
+	pad->addr = index * 0x50;;
+	return 0;
+}
+
+struct nouveau_oclass
+nv94_i2c_pad_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv94_i2c_pad_ctor,
+		.dtor = _nvkm_i2c_pad_dtor,
+		.init = nv94_i2c_pad_init,
+		.fini = nv94_i2c_pad_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/port.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/port.h
new file mode 100644
index 000000000000..a8ff6e077af5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/port.h
@@ -0,0 +1,15 @@
+#ifndef __NVKM_I2C_PORT_H__
+#define __NVKM_I2C_PORT_H__
+
+#include "priv.h"
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nouveau_i2c_port *_port = (void *)port;                         \
+	nv_##l(nv_object(_port)->engine, "PORT:%02x: "f, _port->index, ##a);   \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/priv.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/priv.h
new file mode 100644
index 000000000000..780090b6425a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/priv.h
@@ -0,0 +1,85 @@
+#ifndef __NVKM_I2C_H__
+#define __NVKM_I2C_H__
+
+#include <subdev/i2c.h>
+
+extern struct nouveau_oclass nv04_i2c_pad_oclass;
+extern struct nouveau_oclass nv94_i2c_pad_oclass;
+
+#define nouveau_i2c_port_create(p,e,o,i,a,f,d)                                 \
+	nouveau_i2c_port_create_((p), (e), (o), (i), (a), (f),                 \
+				 sizeof(**d), (void **)d)
+#define nouveau_i2c_port_destroy(p) ({                                         \
+	struct nouveau_i2c_port *port = (p);                                   \
+	_nouveau_i2c_port_dtor(nv_object(i2c));                                \
+})
+#define nouveau_i2c_port_init(p)                                               \
+	nouveau_object_init(&(p)->base)
+#define nouveau_i2c_port_fini(p,s)                                             \
+	nouveau_object_fini(&(p)->base, (s))
+
+int nouveau_i2c_port_create_(struct nouveau_object *, struct nouveau_object *,
+			     struct nouveau_oclass *, u8,
+			     const struct i2c_algorithm *,
+			     const struct nouveau_i2c_func *,
+			     int, void **);
+void _nouveau_i2c_port_dtor(struct nouveau_object *);
+#define _nouveau_i2c_port_init nouveau_object_init
+int  _nouveau_i2c_port_fini(struct nouveau_object *, bool);
+
+#define nouveau_i2c_create(p,e,o,d)                                            \
+	nouveau_i2c_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_i2c_destroy(p) ({                                              \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_dtor(nv_object(i2c));                                     \
+})
+#define nouveau_i2c_init(p) ({                                                 \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_init(nv_object(i2c));                                     \
+})
+#define nouveau_i2c_fini(p,s) ({                                               \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_fini(nv_object(i2c), (s));                                \
+})
+
+int nouveau_i2c_create_(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, int, void **);
+int  _nouveau_i2c_ctor(struct nouveau_object *, struct nouveau_object *,
+		       struct nouveau_oclass *, void *, u32,
+		       struct nouveau_object **);
+void _nouveau_i2c_dtor(struct nouveau_object *);
+int  _nouveau_i2c_init(struct nouveau_object *);
+int  _nouveau_i2c_fini(struct nouveau_object *, bool);
+
+extern struct nouveau_oclass nouveau_anx9805_sclass[];
+extern struct nouveau_oclass nvd0_i2c_sclass[];
+
+extern const struct i2c_algorithm nouveau_i2c_bit_algo;
+extern const struct i2c_algorithm nouveau_i2c_aux_algo;
+
+struct nouveau_i2c_impl {
+	struct nouveau_oclass base;
+
+	/* supported i2c port classes */
+	struct nouveau_oclass *sclass;
+	struct nouveau_oclass *pad_x;
+	struct nouveau_oclass *pad_s;
+
+	/* number of native dp aux channels present */
+	int aux;
+
+	/* read and ack pending interrupts, returning only data
+	 * for ports that have not been masked off, while still
+	 * performing the ack for anything that was pending.
+	 */
+	void (*aux_stat)(struct nouveau_i2c *, u32 *, u32 *, u32 *, u32 *);
+
+	/* mask on/off interrupt types for a given set of auxch
+	 */
+	void (*aux_mask)(struct nouveau_i2c *, u32, u32, u32);
+};
+
+void nv94_aux_stat(struct nouveau_i2c *, u32 *, u32 *, u32 *, u32 *);
+void nv94_aux_mask(struct nouveau_i2c *, u32, u32, u32);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/ibus/gk20a.c b/drivers/gpu/drm/nouveau/core/subdev/ibus/gk20a.c
new file mode 100644
index 000000000000..245f0ebaa6af
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/ibus/gk20a.c
@@ -0,0 +1,103 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include <subdev/ibus.h>
+#include <subdev/timer.h>
+
+struct gk20a_ibus_priv {
+	struct nouveau_ibus base;
+};
+
+static void
+gk20a_ibus_init_priv_ring(struct gk20a_ibus_priv *priv)
+{
+	nv_mask(priv, 0x137250, 0x3f, 0);
+
+	nv_mask(priv, 0x000200, 0x20, 0);
+	usleep_range(20, 30);
+	nv_mask(priv, 0x000200, 0x20, 0x20);
+
+	nv_wr32(priv, 0x12004c, 0x4);
+	nv_wr32(priv, 0x122204, 0x2);
+	nv_rd32(priv, 0x122204);
+}
+
+static void
+gk20a_ibus_intr(struct nouveau_subdev *subdev)
+{
+	struct gk20a_ibus_priv *priv = (void *)subdev;
+	u32 status0 = nv_rd32(priv, 0x120058);
+
+	if (status0 & 0x7) {
+		nv_debug(priv, "resetting priv ring\n");
+		gk20a_ibus_init_priv_ring(priv);
+	}
+
+	/* Acknowledge interrupt */
+	nv_mask(priv, 0x12004c, 0x2, 0x2);
+
+	if (!nv_wait(subdev, 0x12004c, 0x3f, 0x00))
+		nv_warn(priv, "timeout waiting for ringmaster ack\n");
+}
+
+static int
+gk20a_ibus_init(struct nouveau_object *object)
+{
+	struct gk20a_ibus_priv *priv = (void *)object;
+	int ret;
+
+	ret = _nouveau_ibus_init(object);
+	if (ret)
+		return ret;
+
+	gk20a_ibus_init_priv_ring(priv);
+
+	return 0;
+}
+
+static int
+gk20a_ibus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
+{
+	struct gk20a_ibus_priv *priv;
+	int ret;
+
+	ret = nouveau_ibus_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->intr = gk20a_ibus_intr;
+	return 0;
+}
+
+struct nouveau_oclass
+gk20a_ibus_oclass = {
+	.handle = NV_SUBDEV(IBUS, 0xea),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = gk20a_ibus_ctor,
+		.dtor = _nouveau_ibus_dtor,
+		.init = gk20a_ibus_init,
+		.fini = _nouveau_ibus_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/ibus/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/ibus/nve0.c
index 7120124dceac..ebef970a0645 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/ibus/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/ibus/nve0.c
@@ -95,6 +95,23 @@ nve0_ibus_intr(struct nouveau_subdev *subdev)
 }
 
 static int
+nve0_ibus_init(struct nouveau_object *object)
+{
+	struct nve0_ibus_priv *priv = (void *)object;
+	int ret = nouveau_ibus_init(&priv->base);
+	if (ret == 0) {
+		nv_mask(priv, 0x122318, 0x0003ffff, 0x00001000);
+		nv_mask(priv, 0x12231c, 0x0003ffff, 0x00000200);
+		nv_mask(priv, 0x122310, 0x0003ffff, 0x00000800);
+		nv_mask(priv, 0x122348, 0x0003ffff, 0x00000100);
+		nv_mask(priv, 0x1223b0, 0x0003ffff, 0x00000fff);
+		nv_mask(priv, 0x122348, 0x0003ffff, 0x00000200);
+		nv_mask(priv, 0x122358, 0x0003ffff, 0x00002880);
+	}
+	return ret;
+}
+
+static int
 nve0_ibus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
 	       struct nouveau_object **pobject)
@@ -117,7 +134,7 @@ nve0_ibus_oclass = {
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nve0_ibus_ctor,
 		.dtor = _nouveau_ibus_dtor,
-		.init = _nouveau_ibus_init,
+		.init = nve0_ibus_init,
 		.fini = _nouveau_ibus_fini,
 	},
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
index e8822a934c48..9ca93e2718f7 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
@@ -26,6 +26,7 @@
 
 const struct nouveau_mc_intr
 nv50_mc_intr[] = {
+	{ 0x04000000, NVDEV_ENGINE_DISP },  /* DISP before FIFO, so pageflip-timestamping works! */
 	{ 0x00000001, NVDEV_ENGINE_MPEG },
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
 	{ 0x00001000, NVDEV_ENGINE_GR },
@@ -33,8 +34,8 @@ nv50_mc_intr[] = {
 	{ 0x00008000, NVDEV_ENGINE_BSP },	/* NV84- */
 	{ 0x00020000, NVDEV_ENGINE_VP },	/* NV84- */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
-	{ 0x00200000, NVDEV_SUBDEV_GPIO },
-	{ 0x04000000, NVDEV_ENGINE_DISP },
+	{ 0x00200000, NVDEV_SUBDEV_GPIO },	/* PMGR->GPIO */
+	{ 0x00200000, NVDEV_SUBDEV_I2C }, 	/* PMGR->I2C/AUX */
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{ 0x0002d101, NVDEV_SUBDEV_FB },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
index f8a6f18e2d34..3c76d9038f38 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
@@ -26,6 +26,7 @@
 
 static const struct nouveau_mc_intr
 nv98_mc_intr[] = {
+	{ 0x04000000, NVDEV_ENGINE_DISP },  /* DISP first, so pageflip timestamps work */
 	{ 0x00000001, NVDEV_ENGINE_PPP },
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
 	{ 0x00001000, NVDEV_ENGINE_GR },
@@ -35,9 +36,9 @@ nv98_mc_intr[] = {
 	{ 0x00040000, NVDEV_SUBDEV_PWR },	/* NVA3:NVC0 */
 	{ 0x00080000, NVDEV_SUBDEV_THERM },	/* NVA3:NVC0 */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
-	{ 0x00200000, NVDEV_SUBDEV_GPIO },
+	{ 0x00200000, NVDEV_SUBDEV_GPIO },	/* PMGR->GPIO */
+	{ 0x00200000, NVDEV_SUBDEV_I2C }, 	/* PMGR->I2C/AUX */
 	{ 0x00400000, NVDEV_ENGINE_COPY0 },	/* NVA3-     */
-	{ 0x04000000, NVDEV_ENGINE_DISP },
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{ 0x0042d101, NVDEV_SUBDEV_FB },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
index 34472d317097..f9c6a678b47d 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
@@ -26,6 +26,7 @@
 
 const struct nouveau_mc_intr
 nvc0_mc_intr[] = {
+	{ 0x04000000, NVDEV_ENGINE_DISP },  /* DISP first, so pageflip timestamps work. */
 	{ 0x00000001, NVDEV_ENGINE_PPP },
 	{ 0x00000020, NVDEV_ENGINE_COPY0 },
 	{ 0x00000040, NVDEV_ENGINE_COPY1 },
@@ -37,10 +38,10 @@ nvc0_mc_intr[] = {
 	{ 0x00040000, NVDEV_SUBDEV_THERM },
 	{ 0x00020000, NVDEV_ENGINE_VP },
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
-	{ 0x00200000, NVDEV_SUBDEV_GPIO },
+	{ 0x00200000, NVDEV_SUBDEV_GPIO },	/* PMGR->GPIO */
+	{ 0x00200000, NVDEV_SUBDEV_I2C },	/* PMGR->I2C/AUX */
 	{ 0x01000000, NVDEV_SUBDEV_PWR },
 	{ 0x02000000, NVDEV_SUBDEV_LTCG },
-	{ 0x04000000, NVDEV_ENGINE_DISP },
 	{ 0x08000000, NVDEV_SUBDEV_FB },
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x40000000, NVDEV_SUBDEV_IBUS },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c
index 64f8b4702bf7..fcaabe8456e3 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c
@@ -150,7 +150,7 @@ mxm_dcb_sanitise_entry(struct nouveau_bios *bios, void *data, int idx, u16 pdcb)
 	 * common example is DP->eDP.
 	 */
 	conn  = bios->data;
-	conn += dcb_conn(bios, (ctx.outp[0] & 0x0000f000) >> 12, &ver, &len);
+	conn += nvbios_connEe(bios, (ctx.outp[0] & 0x0000f000) >> 12, &ver, &len);
 	type  = conn[0];
 	switch (ctx.desc.conn_type) {
 	case 0x01: /* LVDS */
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc
index 2284ecb1c9b8..c2bb616a8da5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc
@@ -83,7 +83,7 @@ host_send:
 		// increment GET
 		add b32 $r1 0x1
 		and $r14 $r1 #fifo_qmaskf
-		nv_iowr(NV_PPWR_FIFO_GET(0), $r1)
+		nv_iowr(NV_PPWR_FIFO_GET(0), $r14)
 		bra #host_send
 	host_send_done:
 	ret
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h
index 4bd43a99fdcc..39a5dc150a05 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h
@@ -1018,7 +1018,7 @@ uint32_t nv108_pwr_code[] = {
 	0xb600023f,
 	0x1ec40110,
 	0x04b0400f,
-	0xbd0001f6,
+	0xbd000ef6,
 	0xc70ef404,
 /* 0x0328: host_send_done */
 /* 0x032a: host_recv */
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h
index 5a73fa620978..254205cd5166 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h
@@ -1124,7 +1124,7 @@ uint32_t nva3_pwr_code[] = {
 	0x0f1ec401,
 	0x04b007f1,
 	0xd00604b6,
-	0x04bd0001,
+	0x04bd000e,
 /* 0x03cb: host_send_done */
 	0xf8ba0ef4,
 /* 0x03cd: host_recv */
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h
index 4dba00d2dd1a..7ac87405d01b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h
@@ -1124,7 +1124,7 @@ uint32_t nvc0_pwr_code[] = {
 	0x0f1ec401,
 	0x04b007f1,
 	0xd00604b6,
-	0x04bd0001,
+	0x04bd000e,
 /* 0x03cb: host_send_done */
 	0xf8ba0ef4,
 /* 0x03cd: host_recv */
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h
index 5e24c6bc041d..cd9ff1a73284 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h
@@ -1033,7 +1033,7 @@ uint32_t nvd0_pwr_code[] = {
 	0xb6026b21,
 	0x1ec40110,
 	0xb007f10f,
-	0x0001d004,
+	0x000ed004,
 	0x0ef404bd,
 /* 0x0365: host_send_done */
 /* 0x0367: host_recv */
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
index 7610fc5f8fa2..ca9ad9fd47be 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
@@ -60,9 +60,9 @@ static struct nouveau_i2c_board_info
 nv_board_infos[] = {
 	{ { I2C_BOARD_INFO("w83l785ts", 0x2d) }, 0 },
 	{ { I2C_BOARD_INFO("w83781d", 0x2d) }, 0  },
-	{ { I2C_BOARD_INFO("adt7473", 0x2e) }, 20  },
-	{ { I2C_BOARD_INFO("adt7473", 0x2d) }, 20  },
-	{ { I2C_BOARD_INFO("adt7473", 0x2c) }, 20  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2e) }, 40  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2d) }, 40  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2c) }, 40  },
 	{ { I2C_BOARD_INFO("f75375", 0x2e) }, 0  },
 	{ { I2C_BOARD_INFO("lm99", 0x4c) }, 0  },
 	{ { I2C_BOARD_INFO("lm90", 0x4c) }, 0  },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
index 3b2c4580098b..0478b2e3fb1d 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
@@ -36,7 +36,7 @@ nva3_therm_fan_sense(struct nouveau_therm *therm)
 	u32 tach = nv_rd32(therm, 0x00e728) & 0x0000ffff;
 	u32 ctrl = nv_rd32(therm, 0x00e720);
 	if (ctrl & 0x00000001)
-		return tach * 60;
+		return tach * 60 / 2;
 	return -ENODEV;
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dac.c b/drivers/gpu/drm/nouveau/dispnv04/dac.c
index 434b920f6bd4..a96dda48718e 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dac.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dac.c
@@ -414,7 +414,7 @@ static void nv04_dac_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 drm_get_connector_name(&nouveau_encoder_connector_get(nv_encoder)->base),
+		 nouveau_encoder_connector_get(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
index a2d669b4acf2..e57babb206d3 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
@@ -477,7 +477,7 @@ static void nv04_dfp_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 drm_get_connector_name(&nouveau_encoder_connector_get(nv_encoder)->base),
+		 nouveau_encoder_connector_get(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c
index 2f1ed61f7c8c..4342fdaee707 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c
@@ -115,7 +115,7 @@ nv04_display_create(struct drm_device *dev)
 				 &dev->mode_config.connector_list, head) {
 		if (!connector->encoder_ids[0]) {
 			NV_WARN(drm, "%s has no encoders, removing\n",
-				drm_get_connector_name(connector));
+				connector->name);
 			connector->funcs->destroy(connector);
 		}
 	}
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
index 244822df8ffc..8667620b703a 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
@@ -171,7 +171,8 @@ static void nv04_tv_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 drm_get_connector_name(&nouveau_encoder_connector_get(nv_encoder)->base), nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
+		 nouveau_encoder_connector_get(nv_encoder)->base.name,
+		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
 static void nv04_tv_destroy(struct drm_encoder *encoder)
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
index acef48f4a4ea..195bd8e86c6a 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
@@ -612,8 +612,7 @@ static void nv17_tv_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_INFO(drm, "Output %s is running on CRTC %d using output %c\n",
-		drm_get_connector_name(
-			&nouveau_encoder_connector_get(nv_encoder)->base),
+		nouveau_encoder_connector_get(nv_encoder)->base.name,
 		nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index d07ce028af51..1fa222e8f007 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -44,6 +44,7 @@
 
 #include <subdev/i2c.h>
 #include <subdev/gpio.h>
+#include <engine/disp.h>
 
 MODULE_PARM_DESC(tv_disable, "Disable TV-out detection");
 static int nouveau_tv_disable = 0;
@@ -75,7 +76,8 @@ find_encoder(struct drm_connector *connector, int type)
 			continue;
 		nv_encoder = nouveau_encoder(obj_to_encoder(obj));
 
-		if (type == DCB_OUTPUT_ANY || nv_encoder->dcb->type == type)
+		if (type == DCB_OUTPUT_ANY ||
+		    (nv_encoder->dcb && nv_encoder->dcb->type == type))
 			return nv_encoder;
 	}
 
@@ -100,22 +102,24 @@ static void
 nouveau_connector_destroy(struct drm_connector *connector)
 {
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
-	nouveau_event_ref(NULL, &nv_connector->hpd_func);
+	nouveau_event_ref(NULL, &nv_connector->hpd);
 	kfree(nv_connector->edid);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
+	if (nv_connector->aux.transfer)
+		drm_dp_aux_unregister(&nv_connector->aux);
 	kfree(connector);
 }
 
-static struct nouveau_i2c_port *
-nouveau_connector_ddc_detect(struct drm_connector *connector,
-			     struct nouveau_encoder **pnv_encoder)
+static struct nouveau_encoder *
+nouveau_connector_ddc_detect(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	struct nouveau_i2c_port *port = NULL;
+	struct nouveau_encoder *nv_encoder;
+	struct drm_mode_object *obj;
 	int i, panel = -ENODEV;
 
 	/* eDP panels need powering on by us (if the VBIOS doesn't default it
@@ -130,13 +134,9 @@ nouveau_connector_ddc_detect(struct drm_connector *connector,
 		}
 	}
 
-	for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++) {
-		struct nouveau_encoder *nv_encoder;
-		struct drm_mode_object *obj;
-		int id;
-
-		id = connector->encoder_ids[i];
-		if (!id)
+	for (i = 0; nv_encoder = NULL, i < DRM_CONNECTOR_MAX_ENCODER; i++) {
+		int id = connector->encoder_ids[i];
+		if (id == 0)
 			break;
 
 		obj = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_ENCODER);
@@ -144,22 +144,24 @@ nouveau_connector_ddc_detect(struct drm_connector *connector,
 			continue;
 		nv_encoder = nouveau_encoder(obj_to_encoder(obj));
 
-		port = nv_encoder->i2c;
-		if (port && nv_probe_i2c(port, 0x50)) {
-			*pnv_encoder = nv_encoder;
-			break;
+		if (nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+			int ret = nouveau_dp_detect(nv_encoder);
+			if (ret == 0)
+				break;
+		} else
+		if (nv_encoder->i2c) {
+			if (nv_probe_i2c(nv_encoder->i2c, 0x50))
+				break;
 		}
-
-		port = NULL;
 	}
 
 	/* eDP panel not detected, restore panel power GPIO to previous
 	 * state to avoid confusing the SOR for other output types.
 	 */
-	if (!port && panel == 0)
+	if (!nv_encoder && panel == 0)
 		gpio->set(gpio, 0, DCB_GPIO_PANEL_POWER, 0xff, panel);
 
-	return port;
+	return nv_encoder;
 }
 
 static struct nouveau_encoder *
@@ -258,25 +260,17 @@ nouveau_connector_detect(struct drm_connector *connector, bool force)
 	if (ret < 0 && ret != -EACCES)
 		return conn_status;
 
-	i2c = nouveau_connector_ddc_detect(connector, &nv_encoder);
-	if (i2c) {
+	nv_encoder = nouveau_connector_ddc_detect(connector);
+	if (nv_encoder && (i2c = nv_encoder->i2c) != NULL) {
 		nv_connector->edid = drm_get_edid(connector, &i2c->adapter);
 		drm_mode_connector_update_edid_property(connector,
 							nv_connector->edid);
 		if (!nv_connector->edid) {
 			NV_ERROR(drm, "DDC responded, but no EDID for %s\n",
-				 drm_get_connector_name(connector));
+				 connector->name);
 			goto detect_analog;
 		}
 
-		if (nv_encoder->dcb->type == DCB_OUTPUT_DP &&
-		    !nouveau_dp_detect(to_drm_encoder(nv_encoder))) {
-			NV_ERROR(drm, "Detected %s, but failed init\n",
-				 drm_get_connector_name(connector));
-			conn_status = connector_status_disconnected;
-			goto out;
-		}
-
 		/* Override encoder type for DVI-I based on whether EDID
 		 * says the display is digital or analog, both use the
 		 * same i2c channel so the value returned from ddc_detect
@@ -437,7 +431,7 @@ nouveau_connector_force(struct drm_connector *connector)
 	nv_encoder = find_encoder(connector, type);
 	if (!nv_encoder) {
 		NV_ERROR(drm, "can't find encoder to force %s on!\n",
-			 drm_get_connector_name(connector));
+			 connector->name);
 		connector->status = connector_status_disconnected;
 		return;
 	}
@@ -912,33 +906,103 @@ nouveau_connector_funcs_lvds = {
 };
 
 static void
+nouveau_connector_dp_dpms(struct drm_connector *connector, int mode)
+{
+	struct nouveau_encoder *nv_encoder = NULL;
+
+	if (connector->encoder)
+		nv_encoder = nouveau_encoder(connector->encoder);
+	if (nv_encoder && nv_encoder->dcb &&
+	    nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+		if (mode == DRM_MODE_DPMS_ON) {
+			u8 data = DP_SET_POWER_D0;
+			nv_wraux(nv_encoder->i2c, DP_SET_POWER, &data, 1);
+			usleep_range(1000, 2000);
+		} else {
+			u8 data = DP_SET_POWER_D3;
+			nv_wraux(nv_encoder->i2c, DP_SET_POWER, &data, 1);
+		}
+	}
+
+	drm_helper_connector_dpms(connector, mode);
+}
+
+static const struct drm_connector_funcs
+nouveau_connector_funcs_dp = {
+	.dpms = nouveau_connector_dp_dpms,
+	.save = NULL,
+	.restore = NULL,
+	.detect = nouveau_connector_detect,
+	.destroy = nouveau_connector_destroy,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.set_property = nouveau_connector_set_property,
+	.force = nouveau_connector_force
+};
+
+static void
 nouveau_connector_hotplug_work(struct work_struct *work)
 {
 	struct nouveau_connector *nv_connector =
-		container_of(work, struct nouveau_connector, hpd_work);
+		container_of(work, typeof(*nv_connector), work);
 	struct drm_connector *connector = &nv_connector->base;
-	struct drm_device *dev = connector->dev;
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	bool plugged = gpio->get(gpio, 0, nv_connector->hpd.func, 0xff);
+	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	const char *name = connector->name;
 
-	NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un",
-		 drm_get_connector_name(connector));
+	if (nv_connector->status & NVKM_HPD_IRQ) {
+	} else {
+		bool plugged = (nv_connector->status != NVKM_HPD_UNPLUG);
 
-	if (plugged)
-		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
-	else
-		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+		NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un", name);
 
-	drm_helper_hpd_irq_event(dev);
+		if (plugged)
+			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
+		else
+			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+		drm_helper_hpd_irq_event(connector->dev);
+	}
+
+	nouveau_event_get(nv_connector->hpd);
 }
 
 static int
-nouveau_connector_hotplug(void *data, int index)
+nouveau_connector_hotplug(void *data, u32 type, int index)
 {
 	struct nouveau_connector *nv_connector = data;
-	schedule_work(&nv_connector->hpd_work);
-	return NVKM_EVENT_KEEP;
+	nv_connector->status = type;
+	schedule_work(&nv_connector->work);
+	return NVKM_EVENT_DROP;
+}
+
+static ssize_t
+nouveau_connector_aux_xfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
+{
+	struct nouveau_connector *nv_connector =
+		container_of(aux, typeof(*nv_connector), aux);
+	struct nouveau_encoder *nv_encoder;
+	struct nouveau_i2c_port *port;
+	int ret;
+
+	nv_encoder = find_encoder(&nv_connector->base, DCB_OUTPUT_DP);
+	if (!nv_encoder || !(port = nv_encoder->i2c))
+		return -ENODEV;
+	if (WARN_ON(msg->size > 16))
+		return -E2BIG;
+	if (msg->size == 0)
+		return msg->size;
+
+	ret = nouveau_i2c(port)->acquire(port, 0);
+	if (ret)
+		return ret;
+
+	ret = port->func->aux(port, false, msg->request, msg->address,
+			      msg->buffer, msg->size);
+	nouveau_i2c(port)->release(port);
+	if (ret >= 0) {
+		msg->reply = ret;
+		return msg->size;
+	}
+
+	return ret;
 }
 
 static int
@@ -974,9 +1038,9 @@ nouveau_connector_create(struct drm_device *dev, int index)
 {
 	const struct drm_connector_funcs *funcs = &nouveau_connector_funcs;
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
 	struct nouveau_display *disp = nouveau_display(dev);
 	struct nouveau_connector *nv_connector = NULL;
+	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
 	struct drm_connector *connector;
 	int type, ret = 0;
 	bool dummy;
@@ -992,33 +1056,15 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		return ERR_PTR(-ENOMEM);
 
 	connector = &nv_connector->base;
-	INIT_WORK(&nv_connector->hpd_work, nouveau_connector_hotplug_work);
 	nv_connector->index = index;
 
 	/* attempt to parse vbios connector type and hotplug gpio */
 	nv_connector->dcb = olddcb_conn(dev, index);
 	if (nv_connector->dcb) {
-		static const u8 hpd[16] = {
-			0xff, 0x07, 0x08, 0xff, 0xff, 0x51, 0x52, 0xff,
-			0xff, 0xff, 0xff, 0xff, 0xff, 0x5e, 0x5f, 0x60,
-		};
-
 		u32 entry = ROM16(nv_connector->dcb[0]);
 		if (olddcb_conntab(dev)[3] >= 4)
 			entry |= (u32)ROM16(nv_connector->dcb[2]) << 16;
 
-		ret = gpio->find(gpio, 0, hpd[ffs((entry & 0x07033000) >> 12)],
-				 DCB_GPIO_UNUSED, &nv_connector->hpd);
-		if (ret)
-			nv_connector->hpd.func = DCB_GPIO_UNUSED;
-
-		if (nv_connector->hpd.func != DCB_GPIO_UNUSED) {
-			nouveau_event_new(gpio->events, nv_connector->hpd.line,
-					  nouveau_connector_hotplug,
-					  nv_connector,
-					 &nv_connector->hpd_func);
-		}
-
 		nv_connector->type = nv_connector->dcb[0];
 		if (drm_conntype_from_dcb(nv_connector->type) ==
 					  DRM_MODE_CONNECTOR_Unknown) {
@@ -1040,7 +1086,6 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 	} else {
 		nv_connector->type = DCB_CONNECTOR_NONE;
-		nv_connector->hpd.func = DCB_GPIO_UNUSED;
 	}
 
 	/* no vbios data, or an unknown dcb connector type - attempt to
@@ -1080,8 +1125,8 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 	}
 
-	type = drm_conntype_from_dcb(nv_connector->type);
-	if (type == DRM_MODE_CONNECTOR_LVDS) {
+	switch ((type = drm_conntype_from_dcb(nv_connector->type))) {
+	case DRM_MODE_CONNECTOR_LVDS:
 		ret = nouveau_bios_parse_lvds_table(dev, 0, &dummy, &dummy);
 		if (ret) {
 			NV_ERROR(drm, "Error parsing LVDS table, disabling\n");
@@ -1090,8 +1135,23 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 
 		funcs = &nouveau_connector_funcs_lvds;
-	} else {
+		break;
+	case DRM_MODE_CONNECTOR_DisplayPort:
+	case DRM_MODE_CONNECTOR_eDP:
+		nv_connector->aux.dev = dev->dev;
+		nv_connector->aux.transfer = nouveau_connector_aux_xfer;
+		ret = drm_dp_aux_register(&nv_connector->aux);
+		if (ret) {
+			NV_ERROR(drm, "failed to register aux channel\n");
+			kfree(nv_connector);
+			return ERR_PTR(ret);
+		}
+
+		funcs = &nouveau_connector_funcs_dp;
+		break;
+	default:
 		funcs = &nouveau_connector_funcs;
+		break;
 	}
 
 	/* defaults, will get overridden in detect() */
@@ -1166,10 +1226,16 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		break;
 	}
 
-	connector->polled = DRM_CONNECTOR_POLL_CONNECT;
-	if (nv_connector->hpd.func != DCB_GPIO_UNUSED)
+	ret = nouveau_event_new(pdisp->hpd, NVKM_HPD, index,
+				nouveau_connector_hotplug,
+				nv_connector, &nv_connector->hpd);
+	if (ret)
+		connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+	else
 		connector->polled = DRM_CONNECTOR_POLL_HPD;
 
+	INIT_WORK(&nv_connector->work, nouveau_connector_hotplug_work);
+
 	drm_sysfs_connector_add(connector);
 	return connector;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index 264a778f473b..8861b6c579ad 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -28,12 +28,12 @@
 #define __NOUVEAU_CONNECTOR_H__
 
 #include <drm/drm_edid.h>
+#include <drm/drm_dp_helper.h>
 #include "nouveau_crtc.h"
 
 #include <core/event.h>
 
 #include <subdev/bios.h>
-#include <subdev/bios/gpio.h>
 
 struct nouveau_i2c_port;
 
@@ -67,9 +67,11 @@ struct nouveau_connector {
 	u8 index;
 	u8 *dcb;
 
-	struct dcb_gpio_func hpd;
-	struct work_struct hpd_work;
-	struct nouveau_eventh *hpd_func;
+	struct nouveau_eventh *hpd;
+	u32 status;
+	struct work_struct work;
+
+	struct drm_dp_aux aux;
 
 	int dithering_mode;
 	int dithering_depth;
diff --git a/drivers/gpu/drm/nouveau/nouveau_crtc.h b/drivers/gpu/drm/nouveau/nouveau_crtc.h
index d1e5890784d7..a0534489d23f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_crtc.h
+++ b/drivers/gpu/drm/nouveau/nouveau_crtc.h
@@ -74,7 +74,7 @@ struct nouveau_crtc {
 
 static inline struct nouveau_crtc *nouveau_crtc(struct drm_crtc *crtc)
 {
-	return container_of(crtc, struct nouveau_crtc, base);
+	return crtc ? container_of(crtc, struct nouveau_crtc, base) : NULL;
 }
 
 static inline struct drm_crtc *to_drm_crtc(struct nouveau_crtc *crtc)
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index da764a4ed958..47ad74255bf1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -42,7 +42,7 @@
 #include <core/class.h>
 
 static int
-nouveau_display_vblank_handler(void *data, int head)
+nouveau_display_vblank_handler(void *data, u32 type, int head)
 {
 	struct nouveau_drm *drm = data;
 	drm_handle_vblank(drm->dev, head);
@@ -178,7 +178,7 @@ nouveau_display_vblank_init(struct drm_device *dev)
 		return -ENOMEM;
 
 	for (i = 0; i < dev->mode_config.num_crtc; i++) {
-		ret = nouveau_event_new(pdisp->vblank, i,
+		ret = nouveau_event_new(pdisp->vblank, 1, i,
 					nouveau_display_vblank_handler,
 					drm, &disp->vblank[i]);
 		if (ret) {
@@ -393,7 +393,7 @@ nouveau_display_init(struct drm_device *dev)
 	/* enable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (conn->hpd_func) nouveau_event_get(conn->hpd_func);
+		if (conn->hpd) nouveau_event_get(conn->hpd);
 	}
 
 	return ret;
@@ -408,7 +408,7 @@ nouveau_display_fini(struct drm_device *dev)
 	/* disable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (conn->hpd_func) nouveau_event_put(conn->hpd_func);
+		if (conn->hpd) nouveau_event_put(conn->hpd);
 	}
 
 	drm_kms_helper_poll_disable(dev);
@@ -736,6 +736,9 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		  fb->bits_per_pixel, fb->pitches[0], crtc->x, crtc->y,
 		  new_bo->bo.offset };
 
+	/* Keep vblanks on during flip, for the target crtc of this flip */
+	drm_vblank_get(dev, nouveau_crtc(crtc)->index);
+
 	/* Emit a page flip */
 	if (nv_device(drm->device)->card_type >= NV_50) {
 		ret = nv50_display_flip_next(crtc, fb, chan, swap_interval);
@@ -779,6 +782,7 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	return 0;
 
 fail_unreserve:
+	drm_vblank_put(dev, nouveau_crtc(crtc)->index);
 	ttm_bo_unreserve(&old_bo->bo);
 fail_unpin:
 	mutex_unlock(&chan->cli->mutex);
@@ -798,6 +802,7 @@ nouveau_finish_page_flip(struct nouveau_channel *chan,
 	struct drm_device *dev = drm->dev;
 	struct nouveau_page_flip_state *s;
 	unsigned long flags;
+	int crtcid = -1;
 
 	spin_lock_irqsave(&dev->event_lock, flags);
 
@@ -808,8 +813,16 @@ nouveau_finish_page_flip(struct nouveau_channel *chan,
 	}
 
 	s = list_first_entry(&fctx->flip, struct nouveau_page_flip_state, head);
-	if (s->event)
-		drm_send_vblank_event(dev, s->crtc, s->event);
+	if (s->event) {
+		/* Vblank timestamps/counts are only correct on >= NV-50 */
+		if (nv_device(drm->device)->card_type >= NV_50)
+			crtcid = s->crtc;
+
+		drm_send_vblank_event(dev, crtcid, s->event);
+	}
+
+	/* Give up ownership of vblank for page-flipped crtc */
+	drm_vblank_put(dev, s->crtc);
 
 	list_del(&s->head);
 	if (ps)
diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c
index 36fd22500569..5675ffc175ae 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dp.c
@@ -55,11 +55,10 @@ nouveau_dp_probe_oui(struct drm_device *dev, struct nouveau_i2c_port *auxch,
 
 }
 
-bool
-nouveau_dp_detect(struct drm_encoder *encoder)
+int
+nouveau_dp_detect(struct nouveau_encoder *nv_encoder)
 {
-	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct drm_device *dev = encoder->dev;
+	struct drm_device *dev = nv_encoder->base.base.dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_i2c_port *auxch;
 	u8 *dpcd = nv_encoder->dp.dpcd;
@@ -67,11 +66,11 @@ nouveau_dp_detect(struct drm_encoder *encoder)
 
 	auxch = nv_encoder->i2c;
 	if (!auxch)
-		return false;
+		return -ENODEV;
 
 	ret = nv_rdaux(auxch, DP_DPCD_REV, dpcd, 8);
 	if (ret)
-		return false;
+		return ret;
 
 	nv_encoder->dp.link_bw = 27000 * dpcd[1];
 	nv_encoder->dp.link_nr = dpcd[2] & DP_MAX_LANE_COUNT_MASK;
@@ -91,6 +90,5 @@ nouveau_dp_detect(struct drm_encoder *encoder)
 		     nv_encoder->dp.link_nr, nv_encoder->dp.link_bw);
 
 	nouveau_dp_probe_oui(dev, auxch, dpcd);
-
-	return true;
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index ddd83756b9a2..5425ffe3931d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -652,12 +652,12 @@ int nouveau_pmops_resume(struct device *dev)
 	ret = nouveau_do_resume(drm_dev);
 	if (ret)
 		return ret;
-	if (drm_dev->mode_config.num_crtc)
-		nouveau_fbcon_set_suspend(drm_dev, 0);
 
-	nouveau_fbcon_zfill_all(drm_dev);
-	if (drm_dev->mode_config.num_crtc)
+	if (drm_dev->mode_config.num_crtc) {
 		nouveau_display_resume(drm_dev);
+		nouveau_fbcon_set_suspend(drm_dev, 0);
+	}
+
 	return 0;
 }
 
@@ -683,11 +683,12 @@ static int nouveau_pmops_thaw(struct device *dev)
 	ret = nouveau_do_resume(drm_dev);
 	if (ret)
 		return ret;
-	if (drm_dev->mode_config.num_crtc)
-		nouveau_fbcon_set_suspend(drm_dev, 0);
-	nouveau_fbcon_zfill_all(drm_dev);
-	if (drm_dev->mode_config.num_crtc)
+
+	if (drm_dev->mode_config.num_crtc) {
 		nouveau_display_resume(drm_dev);
+		nouveau_fbcon_set_suspend(drm_dev, 0);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h
index 24660c0f713d..5f0e37fc2849 100644
--- a/drivers/gpu/drm/nouveau/nouveau_encoder.h
+++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h
@@ -46,6 +46,7 @@ struct nouveau_encoder {
 	/* different to drm_encoder.crtc, this reflects what's
 	 * actually programmed on the hw, not the proposed crtc */
 	struct drm_crtc *crtc;
+	u32 ctrl;
 
 	struct drm_display_mode mode;
 	int last_dpms;
@@ -84,9 +85,7 @@ get_slave_funcs(struct drm_encoder *enc)
 }
 
 /* nouveau_dp.c */
-bool nouveau_dp_detect(struct drm_encoder *);
-void nouveau_dp_dpms(struct drm_encoder *, int mode, u32 datarate,
-		     struct nouveau_object *);
+int nouveau_dp_detect(struct nouveau_encoder *);
 
 struct nouveau_connector *
 nouveau_encoder_connector_get(struct nouveau_encoder *encoder);
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index 64a42cfd3717..191665ee7f52 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -531,17 +531,10 @@ nouveau_fbcon_set_suspend(struct drm_device *dev, int state)
 		if (state == 1)
 			nouveau_fbcon_save_disable_accel(dev);
 		fb_set_suspend(drm->fbcon->helper.fbdev, state);
-		if (state == 0)
+		if (state == 0) {
 			nouveau_fbcon_restore_accel(dev);
+			nouveau_fbcon_zfill(dev, drm->fbcon);
+		}
 		console_unlock();
 	}
 }
-
-void
-nouveau_fbcon_zfill_all(struct drm_device *dev)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	if (drm->fbcon) {
-		nouveau_fbcon_zfill(dev, drm->fbcon);
-	}
-}
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.h b/drivers/gpu/drm/nouveau/nouveau_fbcon.h
index fdfc0c94fbcc..fcff797d2084 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.h
@@ -61,7 +61,6 @@ void nouveau_fbcon_gpu_lockup(struct fb_info *info);
 int nouveau_fbcon_init(struct drm_device *dev);
 void nouveau_fbcon_fini(struct drm_device *dev);
 void nouveau_fbcon_set_suspend(struct drm_device *dev, int state);
-void nouveau_fbcon_zfill_all(struct drm_device *dev);
 void nouveau_fbcon_save_disable_accel(struct drm_device *dev);
 void nouveau_fbcon_restore_accel(struct drm_device *dev);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 90074d620e31..ab5ea3b0d666 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -166,7 +166,7 @@ nouveau_fence_done(struct nouveau_fence *fence)
 }
 
 static int
-nouveau_fence_wait_uevent_handler(void *data, int index)
+nouveau_fence_wait_uevent_handler(void *data, u32 type, int index)
 {
 	struct nouveau_fence_priv *priv = data;
 	wake_up_all(&priv->waiting);
@@ -183,7 +183,7 @@ nouveau_fence_wait_uevent(struct nouveau_fence *fence, bool intr)
 	struct nouveau_eventh *handler;
 	int ret = 0;
 
-	ret = nouveau_event_new(pfifo->uevent, 0,
+	ret = nouveau_event_new(pfifo->uevent, 1, 0,
 				nouveau_fence_wait_uevent_handler,
 				priv, &handler);
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nouveau_ioc32.c b/drivers/gpu/drm/nouveau/nouveau_ioc32.c
index c1a7e5a73a26..462679a8fec5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ioc32.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ioc32.c
@@ -57,7 +57,7 @@ long nouveau_compat_ioctl(struct file *filp, unsigned int cmd,
 		return drm_compat_ioctl(filp, cmd, arg);
 
 #if 0
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(mga_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(mga_compat_ioctls))
 		fn = nouveau_compat_ioctls[nr - DRM_COMMAND_BASE];
 #endif
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/nouveau/nouveau_vga.c b/drivers/gpu/drm/nouveau/nouveau_vga.c
index fb84da3cb50d..4f4c3fec6916 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vga.c
+++ b/drivers/gpu/drm/nouveau/nouveau_vga.c
@@ -64,12 +64,13 @@ static bool
 nouveau_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	bool can_switch;
 
-	spin_lock(&dev->count_lock);
-	can_switch = (dev->open_count == 0);
-	spin_unlock(&dev->count_lock);
-	return can_switch;
+	/*
+	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
+	 * locking inversion with the driver load path. And the access here is
+	 * completely racy anyway. So don't bother with locking for now.
+	 */
+	return dev->open_count == 0;
 }
 
 static const struct vga_switcheroo_client_ops
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 58af547b0b93..4c534b7b04da 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -1,4 +1,4 @@
-	/*
+/*
  * Copyright 2011 Red Hat Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
@@ -26,6 +26,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_dp_helper.h>
 
 #include "nouveau_drm.h"
 #include "nouveau_dma.h"
@@ -957,7 +958,7 @@ nv50_crtc_prepare(struct drm_crtc *crtc)
 
 	nv50_display_flip_stop(crtc);
 
-	push = evo_wait(mast, 2);
+	push = evo_wait(mast, 6);
 	if (push) {
 		if (nv50_vers(mast) < NV84_DISP_MAST_CLASS) {
 			evo_mthd(push, 0x0874 + (nv_crtc->index * 0x400), 1);
@@ -1207,6 +1208,7 @@ static void
 nv50_crtc_disable(struct drm_crtc *crtc)
 {
 	struct nv50_head *head = nv50_head(crtc);
+	evo_sync(crtc->dev);
 	if (head->image)
 		nouveau_bo_unpin(head->image);
 	nouveau_bo_ref(NULL, &head->image);
@@ -1700,10 +1702,9 @@ nv50_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode)
 }
 
 static void
-nv50_hdmi_disconnect(struct drm_encoder *encoder)
+nv50_hdmi_disconnect(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc)
 {
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
 	struct nv50_disp *disp = nv50_disp(encoder->dev);
 	const u32 moff = (nv_crtc->index << 3) | nv_encoder->or;
 
@@ -1722,7 +1723,7 @@ nv50_sor_dpms(struct drm_encoder *encoder, int mode)
 	struct drm_device *dev = encoder->dev;
 	struct nv50_disp *disp = nv50_disp(dev);
 	struct drm_encoder *partner;
-	int or = nv_encoder->or;
+	u32 mthd;
 
 	nv_encoder->last_dpms = mode;
 
@@ -1740,7 +1741,18 @@ nv50_sor_dpms(struct drm_encoder *encoder, int mode)
 		}
 	}
 
-	nv_call(disp->core, NV50_DISP_SOR_PWR + or, (mode == DRM_MODE_DPMS_ON));
+	mthd  = (ffs(nv_encoder->dcb->heads) - 1) << 3;
+	mthd |= (ffs(nv_encoder->dcb->sorconf.link) - 1) << 2;
+	mthd |= nv_encoder->or;
+
+	if (nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+		nv_call(disp->core, NV50_DISP_SOR_PWR | mthd, 1);
+		mthd |= NV94_DISP_SOR_DP_PWR;
+	} else {
+		mthd |= NV50_DISP_SOR_PWR;
+	}
+
+	nv_call(disp->core, mthd, (mode == DRM_MODE_DPMS_ON));
 }
 
 static bool
@@ -1764,33 +1776,36 @@ nv50_sor_mode_fixup(struct drm_encoder *encoder,
 }
 
 static void
-nv50_sor_disconnect(struct drm_encoder *encoder)
+nv50_sor_ctrl(struct nouveau_encoder *nv_encoder, u32 mask, u32 data)
 {
-	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nv50_mast *mast = nv50_mast(encoder->dev);
-	const int or = nv_encoder->or;
-	u32 *push;
-
-	if (nv_encoder->crtc) {
-		nv50_crtc_prepare(nv_encoder->crtc);
-
-		push = evo_wait(mast, 4);
-		if (push) {
-			if (nv50_vers(mast) < NVD0_DISP_MAST_CLASS) {
-				evo_mthd(push, 0x0600 + (or * 0x40), 1);
-				evo_data(push, 0x00000000);
-			} else {
-				evo_mthd(push, 0x0200 + (or * 0x20), 1);
-				evo_data(push, 0x00000000);
-			}
-			evo_kick(push, mast);
+	struct nv50_mast *mast = nv50_mast(nv_encoder->base.base.dev);
+	u32 temp = (nv_encoder->ctrl & ~mask) | (data & mask), *push;
+	if (temp != nv_encoder->ctrl && (push = evo_wait(mast, 2))) {
+		if (nv50_vers(mast) < NVD0_DISP_MAST_CLASS) {
+			evo_mthd(push, 0x0600 + (nv_encoder->or * 0x40), 1);
+			evo_data(push, (nv_encoder->ctrl = temp));
+		} else {
+			evo_mthd(push, 0x0200 + (nv_encoder->or * 0x20), 1);
+			evo_data(push, (nv_encoder->ctrl = temp));
 		}
-
-		nv50_hdmi_disconnect(encoder);
+		evo_kick(push, mast);
 	}
+}
+
+static void
+nv50_sor_disconnect(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
 
 	nv_encoder->last_dpms = DRM_MODE_DPMS_OFF;
 	nv_encoder->crtc = NULL;
+
+	if (nv_crtc) {
+		nv50_crtc_prepare(&nv_crtc->base);
+		nv50_sor_ctrl(nv_encoder, 1 << nv_crtc->index, 0);
+		nv50_hdmi_disconnect(&nv_encoder->base.base, nv_crtc);
+	}
 }
 
 static void
@@ -1810,12 +1825,14 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
 	struct nouveau_connector *nv_connector;
 	struct nvbios *bios = &drm->vbios;
-	u32 *push, lvds = 0;
+	u32 lvds = 0, mask, ctrl;
 	u8 owner = 1 << nv_crtc->index;
 	u8 proto = 0xf;
 	u8 depth = 0x0;
 
 	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	nv_encoder->crtc = encoder->crtc;
+
 	switch (nv_encoder->dcb->type) {
 	case DCB_OUTPUT_TMDS:
 		if (nv_encoder->dcb->sorconf.link & 1) {
@@ -1827,7 +1844,7 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 			proto = 0x2;
 		}
 
-		nv50_hdmi_mode_set(encoder, mode);
+		nv50_hdmi_mode_set(&nv_encoder->base.base, mode);
 		break;
 	case DCB_OUTPUT_LVDS:
 		proto = 0x0;
@@ -1883,19 +1900,11 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 		break;
 	}
 
-	nv50_sor_dpms(encoder, DRM_MODE_DPMS_ON);
+	nv50_sor_dpms(&nv_encoder->base.base, DRM_MODE_DPMS_ON);
 
-	push = evo_wait(nv50_mast(dev), 8);
-	if (push) {
-		if (nv50_vers(mast) < NVD0_DISP_CLASS) {
-			u32 ctrl = (depth << 16) | (proto << 8) | owner;
-			if (mode->flags & DRM_MODE_FLAG_NHSYNC)
-				ctrl |= 0x00001000;
-			if (mode->flags & DRM_MODE_FLAG_NVSYNC)
-				ctrl |= 0x00002000;
-			evo_mthd(push, 0x0600 + (nv_encoder->or * 0x040), 1);
-			evo_data(push, ctrl);
-		} else {
+	if (nv50_vers(mast) >= NVD0_DISP_CLASS) {
+		u32 *push = evo_wait(mast, 3);
+		if (push) {
 			u32 magic = 0x31ec6000 | (nv_crtc->index << 25);
 			u32 syncs = 0x00000001;
 
@@ -1910,14 +1919,21 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 			evo_mthd(push, 0x0404 + (nv_crtc->index * 0x300), 2);
 			evo_data(push, syncs | (depth << 6));
 			evo_data(push, magic);
-			evo_mthd(push, 0x0200 + (nv_encoder->or * 0x020), 1);
-			evo_data(push, owner | (proto << 8));
+			evo_kick(push, mast);
 		}
 
-		evo_kick(push, mast);
+		ctrl = proto << 8;
+		mask = 0x00000f00;
+	} else {
+		ctrl = (depth << 16) | (proto << 8);
+		if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+			ctrl |= 0x00001000;
+		if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+			ctrl |= 0x00002000;
+		mask = 0x000f3f00;
 	}
 
-	nv_encoder->crtc = encoder->crtc;
+	nv50_sor_ctrl(nv_encoder, mask | owner, ctrl | owner);
 }
 
 static void
@@ -2295,7 +2311,7 @@ nv50_display_create(struct drm_device *dev)
 			continue;
 
 		NV_WARN(drm, "%s has no encoders, removing\n",
-			drm_get_connector_name(connector));
+			connector->name);
 		connector->funcs->destroy(connector);
 	}
 
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index e3c47a8005ff..2d28dc337cfb 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -319,13 +319,13 @@ static void page_flip_worker(struct work_struct *work)
 	struct drm_display_mode *mode = &crtc->mode;
 	struct drm_gem_object *bo;
 
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	omap_plane_mode_set(omap_crtc->plane, crtc, crtc->primary->fb,
 			0, 0, mode->hdisplay, mode->vdisplay,
 			crtc->x << 16, crtc->y << 16,
 			mode->hdisplay << 16, mode->vdisplay << 16,
 			vblank_cb, crtc);
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 
 	bo = omap_framebuffer_bo(crtc->primary->fb, 0);
 	drm_gem_object_unreference_unlocked(bo);
@@ -465,7 +465,7 @@ static void apply_worker(struct work_struct *work)
 	 * the callbacks and list modification all serialized
 	 * with respect to modesetting ioctls from userspace.
 	 */
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	dispc_runtime_get();
 
 	/*
@@ -510,7 +510,7 @@ static void apply_worker(struct work_struct *work)
 
 out:
 	dispc_runtime_put();
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 }
 
 int omap_crtc_apply(struct drm_crtc *crtc,
@@ -518,7 +518,7 @@ int omap_crtc_apply(struct drm_crtc *crtc,
 {
 	struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
 
-	WARN_ON(!mutex_is_locked(&crtc->mutex));
+	WARN_ON(!drm_modeset_is_locked(&crtc->mutex));
 
 	/* no need to queue it again if it is already queued: */
 	if (apply->queued)
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index c8270e4b26f3..002b9721e85a 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -588,9 +588,7 @@ static void dev_lastclose(struct drm_device *dev)
 		}
 	}
 
-	drm_modeset_lock_all(dev);
-	ret = drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-	drm_modeset_unlock_all(dev);
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 	if (ret)
 		DBG("failed to restore crtc mode");
 }
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index 8b019602ffe6..2a5cacdc344b 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -346,6 +346,7 @@ void omap_framebuffer_flush(struct drm_framebuffer *fb,
 
 	VERB("flush: %d,%d %dx%d, fb=%p", x, y, w, h, fb);
 
+	/* FIXME: This is racy - no protection against modeset config changes. */
 	while ((connector = omap_framebuffer_get_next_connector(fb, connector))) {
 		/* only consider connectors that are part of a chain */
 		if (connector->encoder && connector->encoder->crtc) {
diff --git a/drivers/gpu/drm/panel/panel-ld9040.c b/drivers/gpu/drm/panel/panel-ld9040.c
index 1f1f8371a199..db1601fdbe29 100644
--- a/drivers/gpu/drm/panel/panel-ld9040.c
+++ b/drivers/gpu/drm/panel/panel-ld9040.c
@@ -27,6 +27,7 @@
 #define MCS_ELVSS_ON		0xb1
 #define MCS_USER_SETTING	0xf0
 #define MCS_DISPCTL		0xf2
+#define MCS_POWER_CTRL		0xf4
 #define MCS_GTCON		0xf7
 #define MCS_PANEL_CONDITION	0xf8
 #define MCS_GAMMA_SET1		0xf9
@@ -182,6 +183,8 @@ static void ld9040_init(struct ld9040 *ctx)
 	ld9040_dcs_write_seq_static(ctx, MCS_DISPCTL,
 		0x02, 0x08, 0x08, 0x10, 0x10);
 	ld9040_dcs_write_seq_static(ctx, MCS_MANPWR, 0x04);
+	ld9040_dcs_write_seq_static(ctx, MCS_POWER_CTRL,
+		0x0a, 0x87, 0x25, 0x6a, 0x44, 0x02, 0x88);
 	ld9040_dcs_write_seq_static(ctx, MCS_ELVSS_ON, 0x0d, 0x00, 0x16);
 	ld9040_dcs_write_seq_static(ctx, MCS_GTCON, 0x09, 0x00, 0x00);
 	ld9040_brightness_set(ctx);
diff --git a/drivers/gpu/drm/panel/panel-s6e8aa0.c b/drivers/gpu/drm/panel/panel-s6e8aa0.c
index 35941d2412b8..06e57a26db7a 100644
--- a/drivers/gpu/drm/panel/panel-s6e8aa0.c
+++ b/drivers/gpu/drm/panel/panel-s6e8aa0.c
@@ -847,6 +847,7 @@ static void s6e8aa0_read_mtp_id(struct s6e8aa0 *ctx)
 	if (i >= ARRAY_SIZE(s6e8aa0_variants)) {
 		dev_err(ctx->dev, "unsupported display version %d\n", id[1]);
 		ctx->error = -EINVAL;
+		return;
 	}
 
 	ctx->variant = &s6e8aa0_variants[i];
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 309f29e9234a..a25136132c31 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -262,6 +262,13 @@ static int panel_simple_remove(struct device *dev)
 	return 0;
 }
 
+static void panel_simple_shutdown(struct device *dev)
+{
+	struct panel_simple *panel = dev_get_drvdata(dev);
+
+	panel_simple_disable(&panel->base);
+}
+
 static const struct drm_display_mode auo_b101aw03_mode = {
 	.clock = 51450,
 	.hdisplay = 1024,
@@ -284,6 +291,28 @@ static const struct panel_desc auo_b101aw03 = {
 	},
 };
 
+static const struct drm_display_mode auo_b133xtn01_mode = {
+	.clock = 69500,
+	.hdisplay = 1366,
+	.hsync_start = 1366 + 48,
+	.hsync_end = 1366 + 48 + 32,
+	.htotal = 1366 + 48 + 32 + 20,
+	.vdisplay = 768,
+	.vsync_start = 768 + 3,
+	.vsync_end = 768 + 3 + 6,
+	.vtotal = 768 + 3 + 6 + 13,
+	.vrefresh = 60,
+};
+
+static const struct panel_desc auo_b133xtn01 = {
+	.modes = &auo_b133xtn01_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 293,
+		.height = 165,
+	},
+};
+
 static const struct drm_display_mode chunghwa_claa101wa01a_mode = {
 	.clock = 72070,
 	.hdisplay = 1366,
@@ -328,6 +357,52 @@ static const struct panel_desc chunghwa_claa101wb01 = {
 	},
 };
 
+static const struct drm_display_mode edt_et057090dhu_mode = {
+	.clock = 25175,
+	.hdisplay = 640,
+	.hsync_start = 640 + 16,
+	.hsync_end = 640 + 16 + 30,
+	.htotal = 640 + 16 + 30 + 114,
+	.vdisplay = 480,
+	.vsync_start = 480 + 10,
+	.vsync_end = 480 + 10 + 3,
+	.vtotal = 480 + 10 + 3 + 32,
+	.vrefresh = 60,
+	.flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC,
+};
+
+static const struct panel_desc edt_et057090dhu = {
+	.modes = &edt_et057090dhu_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 115,
+		.height = 86,
+	},
+};
+
+static const struct drm_display_mode edt_etm0700g0dh6_mode = {
+	.clock = 33260,
+	.hdisplay = 800,
+	.hsync_start = 800 + 40,
+	.hsync_end = 800 + 40 + 128,
+	.htotal = 800 + 40 + 128 + 88,
+	.vdisplay = 480,
+	.vsync_start = 480 + 10,
+	.vsync_end = 480 + 10 + 2,
+	.vtotal = 480 + 10 + 2 + 33,
+	.vrefresh = 60,
+	.flags = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
+};
+
+static const struct panel_desc edt_etm0700g0dh6 = {
+	.modes = &edt_etm0700g0dh6_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+};
+
 static const struct drm_display_mode lg_lp129qe_mode = {
 	.clock = 285250,
 	.hdisplay = 2560,
@@ -377,12 +452,24 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "auo,b101aw03",
 		.data = &auo_b101aw03,
 	}, {
+		.compatible = "auo,b133xtn01",
+		.data = &auo_b133xtn01,
+	}, {
 		.compatible = "chunghwa,claa101wa01a",
 		.data = &chunghwa_claa101wa01a
 	}, {
 		.compatible = "chunghwa,claa101wb01",
 		.data = &chunghwa_claa101wb01
 	}, {
+		.compatible = "edt,et057090dhu",
+		.data = &edt_et057090dhu,
+	}, {
+		.compatible = "edt,et070080dh6",
+		.data = &edt_etm0700g0dh6,
+	}, {
+		.compatible = "edt,etm0700g0dh6",
+		.data = &edt_etm0700g0dh6,
+	}, {
 		.compatible = "lg,lp129qe",
 		.data = &lg_lp129qe,
 	}, {
@@ -412,6 +499,11 @@ static int panel_simple_platform_remove(struct platform_device *pdev)
 	return panel_simple_remove(&pdev->dev);
 }
 
+static void panel_simple_platform_shutdown(struct platform_device *pdev)
+{
+	panel_simple_shutdown(&pdev->dev);
+}
+
 static struct platform_driver panel_simple_platform_driver = {
 	.driver = {
 		.name = "panel-simple",
@@ -420,6 +512,7 @@ static struct platform_driver panel_simple_platform_driver = {
 	},
 	.probe = panel_simple_platform_probe,
 	.remove = panel_simple_platform_remove,
+	.shutdown = panel_simple_platform_shutdown,
 };
 
 struct panel_desc_dsi {
@@ -561,6 +654,11 @@ static int panel_simple_dsi_remove(struct mipi_dsi_device *dsi)
 	return panel_simple_remove(&dsi->dev);
 }
 
+static void panel_simple_dsi_shutdown(struct mipi_dsi_device *dsi)
+{
+	panel_simple_shutdown(&dsi->dev);
+}
+
 static struct mipi_dsi_driver panel_simple_dsi_driver = {
 	.driver = {
 		.name = "panel-simple-dsi",
@@ -569,6 +667,7 @@ static struct mipi_dsi_driver panel_simple_dsi_driver = {
 	},
 	.probe = panel_simple_dsi_probe,
 	.remove = panel_simple_dsi_remove,
+	.shutdown = panel_simple_dsi_shutdown,
 };
 
 static int __init panel_simple_init(void)
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 41bdd174657e..5d7ea2461852 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -574,6 +574,10 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc,
 			   bo->surf.height, bo->surf.stride, bo->surf.format);
 		qxl_io_create_primary(qdev, base_offset, bo);
 		bo->is_primary = true;
+	}
+
+	if (bo->is_primary) {
+		DRM_DEBUG_KMS("setting surface_id to 0 for primary surface %d on crtc %d\n", bo->surface_id, qcrtc->index);
 		surf_id = 0;
 	} else {
 		surf_id = bo->surface_id;
@@ -841,7 +845,7 @@ static const struct drm_connector_funcs qxl_connector_funcs = {
 	.save = qxl_conn_save,
 	.restore = qxl_conn_restore,
 	.detect = qxl_conn_detect,
-	.fill_modes = drm_helper_probe_single_connector_modes,
+	.fill_modes = drm_helper_probe_single_connector_modes_nomerge,
 	.set_property = qxl_conn_set_property,
 	.destroy = qxl_conn_destroy,
 };
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index fee8748bdca5..6e936634d65c 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -214,7 +214,6 @@ static struct pci_driver qxl_pci_driver = {
 static struct drm_driver qxl_driver = {
 	.driver_features = DRIVER_GEM | DRIVER_MODESET |
 			   DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED,
-	.dev_priv_size = 0,
 	.load = qxl_driver_load,
 	.unload = qxl_driver_unload,
 
diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c
index 0bb86e6d41b4..b110883f8253 100644
--- a/drivers/gpu/drm/qxl/qxl_ioctl.c
+++ b/drivers/gpu/drm/qxl/qxl_ioctl.c
@@ -451,4 +451,4 @@ const struct drm_ioctl_desc qxl_ioctls[] = {
 			  DRM_AUTH|DRM_UNLOCKED),
 };
 
-int qxl_max_ioctls = DRM_ARRAY_SIZE(qxl_ioctls);
+int qxl_max_ioctls = ARRAY_SIZE(qxl_ioctls);
diff --git a/drivers/gpu/drm/qxl/qxl_irq.c b/drivers/gpu/drm/qxl/qxl_irq.c
index 28f84b4fce32..34d6a85e9023 100644
--- a/drivers/gpu/drm/qxl/qxl_irq.c
+++ b/drivers/gpu/drm/qxl/qxl_irq.c
@@ -87,7 +87,7 @@ int qxl_irq_init(struct qxl_device *qdev)
 	atomic_set(&qdev->irq_received_cursor, 0);
 	atomic_set(&qdev->irq_received_io_cmd, 0);
 	qdev->irq_received_error = 0;
-	ret = drm_irq_install(qdev->ddev);
+	ret = drm_irq_install(qdev->ddev, qdev->ddev->pdev->irq);
 	qdev->ram_header->int_mask = QXL_INTERRUPT_MASK;
 	if (unlikely(ret != 0)) {
 		DRM_ERROR("Failed installing irq: %d\n", ret);
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index d52c27527b9a..71a1baeac14e 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -109,13 +109,11 @@ static const struct vm_operations_struct *ttm_vm_ops;
 static int qxl_ttm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *bo;
-	struct qxl_device *qdev;
 	int r;
 
 	bo = (struct ttm_buffer_object *)vma->vm_private_data;
 	if (bo == NULL)
 		return VM_FAULT_NOPAGE;
-	qdev = qxl_get_qdev(bo->bdev);
 	r = ttm_vm_ops->fault(vma, vmf);
 	return r;
 }
@@ -162,10 +160,6 @@ static int qxl_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
 static int qxl_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 			     struct ttm_mem_type_manager *man)
 {
-	struct qxl_device *qdev;
-
-	qdev = qxl_get_qdev(bdev);
-
 	switch (type) {
 	case TTM_PL_SYSTEM:
 		/* System memory */
diff --git a/drivers/gpu/drm/r128/r128_ioc32.c b/drivers/gpu/drm/r128/r128_ioc32.c
index b0d0fd3e4376..663f38c63ba6 100644
--- a/drivers/gpu/drm/r128/r128_ioc32.c
+++ b/drivers/gpu/drm/r128/r128_ioc32.c
@@ -203,7 +203,7 @@ long r128_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(r128_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(r128_compat_ioctls))
 		fn = r128_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/r128/r128_state.c b/drivers/gpu/drm/r128/r128_state.c
index e806dacd452f..575e986f82a7 100644
--- a/drivers/gpu/drm/r128/r128_state.c
+++ b/drivers/gpu/drm/r128/r128_state.c
@@ -1594,7 +1594,7 @@ static int r128_getparam(struct drm_device *dev, void *data, struct drm_file *fi
 
 	switch (param->param) {
 	case R128_PARAM_IRQ_NR:
-		value = drm_dev_to_irq(dev);
+		value = dev->pdev->irq;
 		break;
 	default:
 		return -EINVAL;
@@ -1641,4 +1641,4 @@ const struct drm_ioctl_desc r128_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(R128_GETPARAM, r128_getparam, DRM_AUTH),
 };
 
-int r128_max_ioctl = DRM_ARRAY_SIZE(r128_ioctls);
+int r128_max_ioctl = ARRAY_SIZE(r128_ioctls);
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index 09433534dc47..dbcbfe80aac0 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -72,7 +72,7 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
 	radeon_cs.o radeon_bios.o radeon_benchmark.o r100.o r300.o r420.o \
 	rs400.o rs600.o rs690.o rv515.o r520.o r600.o rv770.o radeon_test.o \
 	r200.o radeon_legacy_tv.o r600_cs.o r600_blit_shaders.o \
-	radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \
+	radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o dce3_1_afmt.o \
 	evergreen.o evergreen_cs.o evergreen_blit_shaders.o \
 	evergreen_hdmi.o radeon_trace_points.o ni.o cayman_blit_shaders.o \
 	atombios_encoders.o radeon_semaphore.o radeon_sa.o atombios_i2c.o si.o \
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index e911898348f8..a03c73411a56 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -557,6 +557,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 	u32 adjusted_clock = mode->clock;
 	int encoder_mode = atombios_get_encoder_mode(encoder);
 	u32 dp_clock = mode->clock;
+	u32 clock = mode->clock;
 	int bpc = radeon_crtc->bpc;
 	bool is_duallink = radeon_dig_monitor_is_duallink(encoder, mode->clock);
 
@@ -632,6 +633,24 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 			radeon_crtc->pll_flags |= RADEON_PLL_USE_REF_DIV;
 	}
 
+	/* adjust pll for deep color modes */
+	if (encoder_mode == ATOM_ENCODER_MODE_HDMI) {
+		switch (bpc) {
+		case 8:
+		default:
+			break;
+		case 10:
+			clock = (clock * 5) / 4;
+			break;
+		case 12:
+			clock = (clock * 3) / 2;
+			break;
+		case 16:
+			clock = clock * 2;
+			break;
+		}
+	}
+
 	/* DCE3+ has an AdjustDisplayPll that will adjust the pixel clock
 	 * accordingly based on the encoder/transmitter to work around
 	 * special hw requirements.
@@ -653,7 +672,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 			switch (crev) {
 			case 1:
 			case 2:
-				args.v1.usPixelClock = cpu_to_le16(mode->clock / 10);
+				args.v1.usPixelClock = cpu_to_le16(clock / 10);
 				args.v1.ucTransmitterID = radeon_encoder->encoder_id;
 				args.v1.ucEncodeMode = encoder_mode;
 				if (radeon_crtc->ss_enabled && radeon_crtc->ss.percentage)
@@ -665,7 +684,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 				adjusted_clock = le16_to_cpu(args.v1.usPixelClock) * 10;
 				break;
 			case 3:
-				args.v3.sInput.usPixelClock = cpu_to_le16(mode->clock / 10);
+				args.v3.sInput.usPixelClock = cpu_to_le16(clock / 10);
 				args.v3.sInput.ucTransmitterID = radeon_encoder->encoder_id;
 				args.v3.sInput.ucEncodeMode = encoder_mode;
 				args.v3.sInput.ucDispPllConfig = 0;
@@ -679,10 +698,6 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 					args.v3.sInput.usPixelClock = cpu_to_le16(dp_clock / 10);
 				} else if (radeon_encoder->devices & (ATOM_DEVICE_DFP_SUPPORT)) {
 					struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-					if (encoder_mode == ATOM_ENCODER_MODE_HDMI)
-						/* deep color support */
-						args.v3.sInput.usPixelClock =
-							cpu_to_le16((mode->clock * bpc / 8) / 10);
 					if (dig->coherent_mode)
 						args.v3.sInput.ucDispPllConfig |=
 							DISPPLL_CONFIG_COHERENT_MODE;
@@ -862,14 +877,21 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc,
 			args.v5.ucMiscInfo = 0; /* HDMI depth, etc. */
 			if (ss_enabled && (ss->type & ATOM_EXTERNAL_SS_MASK))
 				args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_REF_DIV_SRC;
-			switch (bpc) {
-			case 8:
-			default:
-				args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_24BPP;
-				break;
-			case 10:
-				args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_30BPP;
-				break;
+			if (encoder_mode == ATOM_ENCODER_MODE_HDMI) {
+				switch (bpc) {
+				case 8:
+				default:
+					args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_24BPP;
+					break;
+				case 10:
+					/* yes this is correct, the atom define is wrong */
+					args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_32BPP;
+					break;
+				case 12:
+					/* yes this is correct, the atom define is wrong */
+					args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_30BPP;
+					break;
+				}
 			}
 			args.v5.ucTransmitterID = encoder_id;
 			args.v5.ucEncoderMode = encoder_mode;
@@ -884,20 +906,22 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc,
 			args.v6.ucMiscInfo = 0; /* HDMI depth, etc. */
 			if (ss_enabled && (ss->type & ATOM_EXTERNAL_SS_MASK))
 				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_REF_DIV_SRC;
-			switch (bpc) {
-			case 8:
-			default:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_24BPP;
-				break;
-			case 10:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_30BPP;
-				break;
-			case 12:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_36BPP;
-				break;
-			case 16:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_48BPP;
-				break;
+			if (encoder_mode == ATOM_ENCODER_MODE_HDMI) {
+				switch (bpc) {
+				case 8:
+				default:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_24BPP;
+					break;
+				case 10:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_30BPP_V6;
+					break;
+				case 12:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_36BPP_V6;
+					break;
+				case 16:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_48BPP;
+					break;
+				}
 			}
 			args.v6.ucTransmitterID = encoder_id;
 			args.v6.ucEncoderMode = encoder_mode;
@@ -938,6 +962,9 @@ static bool atombios_crtc_prepare_pll(struct drm_crtc *crtc, struct drm_display_
 		struct radeon_connector_atom_dig *dig_connector =
 			radeon_connector->con_priv;
 		int dp_clock;
+
+		/* Assign mode clock for hdmi deep color max clock limit check */
+		radeon_connector->pixelclock_for_modeset = mode->clock;
 		radeon_crtc->bpc = radeon_get_monitor_bpc(connector);
 
 		switch (encoder_mode) {
@@ -1019,10 +1046,17 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
 	struct radeon_encoder *radeon_encoder =
 		to_radeon_encoder(radeon_crtc->encoder);
 	u32 pll_clock = mode->clock;
+	u32 clock = mode->clock;
 	u32 ref_div = 0, fb_div = 0, frac_fb_div = 0, post_div = 0;
 	struct radeon_pll *pll;
 	int encoder_mode = atombios_get_encoder_mode(radeon_crtc->encoder);
 
+	/* pass the actual clock to atombios_crtc_program_pll for DCE5,6 for HDMI */
+	if (ASIC_IS_DCE5(rdev) &&
+	    (encoder_mode == ATOM_ENCODER_MODE_HDMI) &&
+	    (radeon_crtc->bpc > 8))
+		clock = radeon_crtc->adjusted_clock;
+
 	switch (radeon_crtc->pll_id) {
 	case ATOM_PPLL1:
 		pll = &rdev->clock.p1pll;
@@ -1057,7 +1091,7 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
 				 radeon_crtc->crtc_id, &radeon_crtc->ss);
 
 	atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id,
-				  encoder_mode, radeon_encoder->encoder_id, mode->clock,
+				  encoder_mode, radeon_encoder->encoder_id, clock,
 				  ref_div, fb_div, frac_fb_div, post_div,
 				  radeon_crtc->bpc, radeon_crtc->ss_enabled, &radeon_crtc->ss);
 
@@ -1102,6 +1136,7 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 	u32 fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_NONE);
 	u32 tmp, viewport_w, viewport_h;
 	int r;
+	bool bypass_lut = false;
 
 	/* no fb bound */
 	if (!atomic && !crtc->primary->fb) {
@@ -1140,33 +1175,73 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 	radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL);
 	radeon_bo_unreserve(rbo);
 
-	switch (target_fb->bits_per_pixel) {
-	case 8:
+	switch (target_fb->pixel_format) {
+	case DRM_FORMAT_C8:
 		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_8BPP) |
 			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_INDEXED));
 		break;
-	case 15:
+	case DRM_FORMAT_XRGB4444:
+	case DRM_FORMAT_ARGB4444:
+		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_16BPP) |
+			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB4444));
+#ifdef __BIG_ENDIAN
+		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN16);
+#endif
+		break;
+	case DRM_FORMAT_XRGB1555:
+	case DRM_FORMAT_ARGB1555:
 		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_16BPP) |
 			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB1555));
+#ifdef __BIG_ENDIAN
+		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN16);
+#endif
+		break;
+	case DRM_FORMAT_BGRX5551:
+	case DRM_FORMAT_BGRA5551:
+		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_16BPP) |
+			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_BGRA5551));
+#ifdef __BIG_ENDIAN
+		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN16);
+#endif
 		break;
-	case 16:
+	case DRM_FORMAT_RGB565:
 		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_16BPP) |
 			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB565));
 #ifdef __BIG_ENDIAN
 		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN16);
 #endif
 		break;
-	case 24:
-	case 32:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_ARGB8888:
 		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_32BPP) |
 			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB8888));
 #ifdef __BIG_ENDIAN
 		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN32);
 #endif
 		break;
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_ARGB2101010:
+		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_32BPP) |
+			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB2101010));
+#ifdef __BIG_ENDIAN
+		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN32);
+#endif
+		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
+		bypass_lut = true;
+		break;
+	case DRM_FORMAT_BGRX1010102:
+	case DRM_FORMAT_BGRA1010102:
+		fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_32BPP) |
+			     EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_BGRA1010102));
+#ifdef __BIG_ENDIAN
+		fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN32);
+#endif
+		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
+		bypass_lut = true;
+		break;
 	default:
-		DRM_ERROR("Unsupported screen depth %d\n",
-			  target_fb->bits_per_pixel);
+		DRM_ERROR("Unsupported screen format %s\n",
+			  drm_get_format_name(target_fb->pixel_format));
 		return -EINVAL;
 	}
 
@@ -1295,6 +1370,18 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 	WREG32(EVERGREEN_GRPH_CONTROL + radeon_crtc->crtc_offset, fb_format);
 	WREG32(EVERGREEN_GRPH_SWAP_CONTROL + radeon_crtc->crtc_offset, fb_swap);
 
+	/*
+	 * The LUT only has 256 slots for indexing by a 8 bpc fb. Bypass the LUT
+	 * for > 8 bpc scanout to avoid truncation of fb indices to 8 msb's, to
+	 * retain the full precision throughout the pipeline.
+	 */
+	WREG32_P(EVERGREEN_GRPH_LUT_10BIT_BYPASS_CONTROL + radeon_crtc->crtc_offset,
+		 (bypass_lut ? EVERGREEN_LUT_10BIT_BYPASS_EN : 0),
+		 ~EVERGREEN_LUT_10BIT_BYPASS_EN);
+
+	if (bypass_lut)
+		DRM_DEBUG_KMS("Bypassing hardware LUT due to 10 bit fb scanout.\n");
+
 	WREG32(EVERGREEN_GRPH_SURFACE_OFFSET_X + radeon_crtc->crtc_offset, 0);
 	WREG32(EVERGREEN_GRPH_SURFACE_OFFSET_Y + radeon_crtc->crtc_offset, 0);
 	WREG32(EVERGREEN_GRPH_X_START + radeon_crtc->crtc_offset, 0);
@@ -1362,6 +1449,7 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 	u32 fb_swap = R600_D1GRPH_SWAP_ENDIAN_NONE;
 	u32 tmp, viewport_w, viewport_h;
 	int r;
+	bool bypass_lut = false;
 
 	/* no fb bound */
 	if (!atomic && !crtc->primary->fb) {
@@ -1399,18 +1487,30 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 	radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL);
 	radeon_bo_unreserve(rbo);
 
-	switch (target_fb->bits_per_pixel) {
-	case 8:
+	switch (target_fb->pixel_format) {
+	case DRM_FORMAT_C8:
 		fb_format =
 		    AVIVO_D1GRPH_CONTROL_DEPTH_8BPP |
 		    AVIVO_D1GRPH_CONTROL_8BPP_INDEXED;
 		break;
-	case 15:
+	case DRM_FORMAT_XRGB4444:
+	case DRM_FORMAT_ARGB4444:
+		fb_format =
+		    AVIVO_D1GRPH_CONTROL_DEPTH_16BPP |
+		    AVIVO_D1GRPH_CONTROL_16BPP_ARGB4444;
+#ifdef __BIG_ENDIAN
+		fb_swap = R600_D1GRPH_SWAP_ENDIAN_16BIT;
+#endif
+		break;
+	case DRM_FORMAT_XRGB1555:
 		fb_format =
 		    AVIVO_D1GRPH_CONTROL_DEPTH_16BPP |
 		    AVIVO_D1GRPH_CONTROL_16BPP_ARGB1555;
+#ifdef __BIG_ENDIAN
+		fb_swap = R600_D1GRPH_SWAP_ENDIAN_16BIT;
+#endif
 		break;
-	case 16:
+	case DRM_FORMAT_RGB565:
 		fb_format =
 		    AVIVO_D1GRPH_CONTROL_DEPTH_16BPP |
 		    AVIVO_D1GRPH_CONTROL_16BPP_RGB565;
@@ -1418,8 +1518,8 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 		fb_swap = R600_D1GRPH_SWAP_ENDIAN_16BIT;
 #endif
 		break;
-	case 24:
-	case 32:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_ARGB8888:
 		fb_format =
 		    AVIVO_D1GRPH_CONTROL_DEPTH_32BPP |
 		    AVIVO_D1GRPH_CONTROL_32BPP_ARGB8888;
@@ -1427,9 +1527,20 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 		fb_swap = R600_D1GRPH_SWAP_ENDIAN_32BIT;
 #endif
 		break;
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_ARGB2101010:
+		fb_format =
+		    AVIVO_D1GRPH_CONTROL_DEPTH_32BPP |
+		    AVIVO_D1GRPH_CONTROL_32BPP_ARGB2101010;
+#ifdef __BIG_ENDIAN
+		fb_swap = R600_D1GRPH_SWAP_ENDIAN_32BIT;
+#endif
+		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
+		bypass_lut = true;
+		break;
 	default:
-		DRM_ERROR("Unsupported screen depth %d\n",
-			  target_fb->bits_per_pixel);
+		DRM_ERROR("Unsupported screen format %s\n",
+			  drm_get_format_name(target_fb->pixel_format));
 		return -EINVAL;
 	}
 
@@ -1468,6 +1579,13 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 	if (rdev->family >= CHIP_R600)
 		WREG32(R600_D1GRPH_SWAP_CONTROL + radeon_crtc->crtc_offset, fb_swap);
 
+	/* LUT only has 256 slots for 8 bpc fb. Bypass for > 8 bpc scanout for precision */
+	WREG32_P(AVIVO_D1GRPH_LUT_SEL + radeon_crtc->crtc_offset,
+		 (bypass_lut ? AVIVO_LUT_10BIT_BYPASS_EN : 0), ~AVIVO_LUT_10BIT_BYPASS_EN);
+
+	if (bypass_lut)
+		DRM_DEBUG_KMS("Bypassing hardware LUT due to 10 bit fb scanout.\n");
+
 	WREG32(AVIVO_D1GRPH_SURFACE_OFFSET_X + radeon_crtc->crtc_offset, 0);
 	WREG32(AVIVO_D1GRPH_SURFACE_OFFSET_Y + radeon_crtc->crtc_offset, 0);
 	WREG32(AVIVO_D1GRPH_X_START + radeon_crtc->crtc_offset, 0);
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 54e4f52549af..b1e11f8434e2 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -95,9 +95,12 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessAuxChannelTransaction);
 	unsigned char *base;
 	int recv_bytes;
+	int r = 0;
 
 	memset(&args, 0, sizeof(args));
 
+	mutex_lock(&chan->mutex);
+
 	base = (unsigned char *)(rdev->mode_info.atom_context->scratch + 1);
 
 	radeon_atom_copy_swap(base, send, send_bytes, true);
@@ -117,19 +120,22 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 	/* timeout */
 	if (args.v1.ucReplyStatus == 1) {
 		DRM_DEBUG_KMS("dp_aux_ch timeout\n");
-		return -ETIMEDOUT;
+		r = -ETIMEDOUT;
+		goto done;
 	}
 
 	/* flags not zero */
 	if (args.v1.ucReplyStatus == 2) {
 		DRM_DEBUG_KMS("dp_aux_ch flags not zero\n");
-		return -EBUSY;
+		r = -EIO;
+		goto done;
 	}
 
 	/* error */
 	if (args.v1.ucReplyStatus == 3) {
 		DRM_DEBUG_KMS("dp_aux_ch error\n");
-		return -EIO;
+		r = -EIO;
+		goto done;
 	}
 
 	recv_bytes = args.v1.ucDataOutLen;
@@ -139,7 +145,11 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 	if (recv && recv_size)
 		radeon_atom_copy_swap(recv, base + 16, recv_bytes, false);
 
-	return recv_bytes;
+	r = recv_bytes;
+done:
+	mutex_unlock(&chan->mutex);
+
+	return r;
 }
 
 #define BARE_ADDRESS_SIZE 3
@@ -212,11 +222,12 @@ void radeon_dp_aux_init(struct radeon_connector *radeon_connector)
 	radeon_connector->ddc_bus->rec.hpd = radeon_connector->hpd.hpd;
 	radeon_connector->ddc_bus->aux.dev = radeon_connector->base.kdev;
 	radeon_connector->ddc_bus->aux.transfer = radeon_dp_aux_transfer;
-	ret = drm_dp_aux_register_i2c_bus(&radeon_connector->ddc_bus->aux);
+
+	ret = drm_dp_aux_register(&radeon_connector->ddc_bus->aux);
 	if (!ret)
 		radeon_connector->ddc_bus->has_aux = true;
 
-	WARN(ret, "drm_dp_aux_register_i2c_bus() failed with error %d\n", ret);
+	WARN(ret, "drm_dp_aux_register() failed with error %d\n", ret);
 }
 
 /***** general DP utility functions *****/
@@ -281,6 +292,19 @@ static int dp_get_max_dp_pix_clock(int link_rate,
 
 /***** radeon specific DP functions *****/
 
+static int radeon_dp_get_max_link_rate(struct drm_connector *connector,
+				       u8 dpcd[DP_DPCD_SIZE])
+{
+	int max_link_rate;
+
+	if (radeon_connector_is_dp12_capable(connector))
+		max_link_rate = min(drm_dp_max_link_rate(dpcd), 540000);
+	else
+		max_link_rate = min(drm_dp_max_link_rate(dpcd), 270000);
+
+	return max_link_rate;
+}
+
 /* First get the min lane# when low rate is used according to pixel clock
  * (prefer low rate), second check max lane# supported by DP panel,
  * if the max lane# < low rate lane# then use max lane# instead.
@@ -290,7 +314,7 @@ static int radeon_dp_get_dp_lane_number(struct drm_connector *connector,
 					int pix_clock)
 {
 	int bpp = convert_bpc_to_bpp(radeon_get_monitor_bpc(connector));
-	int max_link_rate = drm_dp_max_link_rate(dpcd);
+	int max_link_rate = radeon_dp_get_max_link_rate(connector, dpcd);
 	int max_lane_num = drm_dp_max_lane_count(dpcd);
 	int lane_num;
 	int max_dp_pix_clock;
@@ -328,7 +352,7 @@ static int radeon_dp_get_dp_link_clock(struct drm_connector *connector,
 			return 540000;
 	}
 
-	return drm_dp_max_link_rate(dpcd);
+	return radeon_dp_get_max_link_rate(connector, dpcd);
 }
 
 static u8 radeon_dp_encoder_service(struct radeon_device *rdev,
@@ -379,16 +403,18 @@ bool radeon_dp_getdpcd(struct radeon_connector *radeon_connector)
 {
 	struct radeon_connector_atom_dig *dig_connector = radeon_connector->con_priv;
 	u8 msg[DP_DPCD_SIZE];
-	int ret, i;
+	int ret;
+
+	char dpcd_hex_dump[DP_DPCD_SIZE * 3];
 
 	ret = drm_dp_dpcd_read(&radeon_connector->ddc_bus->aux, DP_DPCD_REV, msg,
 			       DP_DPCD_SIZE);
 	if (ret > 0) {
 		memcpy(dig_connector->dpcd, msg, DP_DPCD_SIZE);
-		DRM_DEBUG_KMS("DPCD: ");
-		for (i = 0; i < DP_DPCD_SIZE; i++)
-			DRM_DEBUG_KMS("%02x ", msg[i]);
-		DRM_DEBUG_KMS("\n");
+
+		hex_dump_to_buffer(dig_connector->dpcd, sizeof(dig_connector->dpcd),
+				   32, 1, dpcd_hex_dump, sizeof(dpcd_hex_dump), false);
+		DRM_DEBUG_KMS("DPCD: %s\n", dpcd_hex_dump);
 
 		radeon_dp_probe_oui(radeon_connector);
 
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index e6eb5097597f..2b2908440644 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -1884,8 +1884,11 @@ atombios_set_encoder_crtc_source(struct drm_encoder *encoder)
 					args.v2.ucEncodeMode = ATOM_ENCODER_MODE_CRT;
 				else
 					args.v2.ucEncodeMode = atombios_get_encoder_mode(encoder);
-			} else
+			} else if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
+				args.v2.ucEncodeMode = ATOM_ENCODER_MODE_LVDS;
+			} else {
 				args.v2.ucEncodeMode = atombios_get_encoder_mode(encoder);
+			}
 			switch (radeon_encoder->encoder_id) {
 			case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
 			case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1:
diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c
index b5162c3b6111..9c570fb15b8c 100644
--- a/drivers/gpu/drm/radeon/atombios_i2c.c
+++ b/drivers/gpu/drm/radeon/atombios_i2c.c
@@ -43,15 +43,19 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessI2cChannelTransaction);
 	unsigned char *base;
 	u16 out = cpu_to_le16(0);
+	int r = 0;
 
 	memset(&args, 0, sizeof(args));
 
+	mutex_lock(&chan->mutex);
+
 	base = (unsigned char *)rdev->mode_info.atom_context->scratch;
 
 	if (flags & HW_I2C_WRITE) {
 		if (num > ATOM_MAX_HW_I2C_WRITE) {
 			DRM_ERROR("hw i2c: tried to write too many bytes (%d vs 3)\n", num);
-			return -EINVAL;
+			r = -EINVAL;
+			goto done;
 		}
 		if (buf == NULL)
 			args.ucRegIndex = 0;
@@ -65,7 +69,8 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	} else {
 		if (num > ATOM_MAX_HW_I2C_READ) {
 			DRM_ERROR("hw i2c: tried to read too many bytes (%d vs 255)\n", num);
-			return -EINVAL;
+			r = -EINVAL;
+			goto done;
 		}
 		args.ucRegIndex = 0;
 		args.lpI2CDataOut = 0;
@@ -82,13 +87,17 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	/* error */
 	if (args.ucStatus != HW_ASSISTED_I2C_STATUS_SUCCESS) {
 		DRM_DEBUG_KMS("hw_i2c error\n");
-		return -EIO;
+		r = -EIO;
+		goto done;
 	}
 
 	if (!(flags & HW_I2C_WRITE))
 		radeon_atom_copy_swap(buf, base, num, false);
 
-	return 0;
+done:
+	mutex_unlock(&chan->mutex);
+
+	return r;
 }
 
 int radeon_atom_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 10dae4106c08..584090ac3eb9 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -1179,7 +1179,7 @@ static int ci_stop_dpm(struct radeon_device *rdev)
 	tmp &= ~GLOBAL_PWRMGT_EN;
 	WREG32_SMC(GENERAL_PWRMGT, tmp);
 
-	tmp = RREG32(SCLK_PWRMGT_CNTL);
+	tmp = RREG32_SMC(SCLK_PWRMGT_CNTL);
 	tmp &= ~DYNAMIC_PM_EN;
 	WREG32_SMC(SCLK_PWRMGT_CNTL, tmp);
 
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index d2fd98968085..0b2471107137 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -80,6 +80,7 @@ extern int sumo_rlc_init(struct radeon_device *rdev);
 extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 extern void si_rlc_reset(struct radeon_device *rdev);
 extern void si_init_uvd_internal_cg(struct radeon_device *rdev);
+static u32 cik_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh);
 extern int cik_sdma_resume(struct radeon_device *rdev);
 extern void cik_sdma_enable(struct radeon_device *rdev, bool enable);
 extern void cik_sdma_fini(struct radeon_device *rdev);
@@ -3257,7 +3258,7 @@ static void cik_gpu_init(struct radeon_device *rdev)
 	u32 mc_shared_chmap, mc_arb_ramcfg;
 	u32 hdp_host_path_cntl;
 	u32 tmp;
-	int i, j;
+	int i, j, k;
 
 	switch (rdev->family) {
 	case CHIP_BONAIRE:
@@ -3446,6 +3447,15 @@ static void cik_gpu_init(struct radeon_device *rdev)
 		     rdev->config.cik.max_sh_per_se,
 		     rdev->config.cik.max_backends_per_se);
 
+	for (i = 0; i < rdev->config.cik.max_shader_engines; i++) {
+		for (j = 0; j < rdev->config.cik.max_sh_per_se; j++) {
+			for (k = 0; k < rdev->config.cik.max_cu_per_sh; k++) {
+				rdev->config.cik.active_cus +=
+					hweight32(cik_get_cu_active_bitmap(rdev, i, j));
+			}
+		}
+	}
+
 	/* set HW defaults for 3D engine */
 	WREG32(CP_MEQ_THRESHOLDS, MEQ1_START(0x30) | MEQ2_START(0x60));
 
@@ -3698,7 +3708,7 @@ bool cik_semaphore_ring_emit(struct radeon_device *rdev,
 	unsigned sel = emit_wait ? PACKET3_SEM_SEL_WAIT : PACKET3_SEM_SEL_SIGNAL;
 
 	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xffff) | sel);
 
 	return true;
@@ -3818,7 +3828,7 @@ void cik_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 			radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 			radeon_ring_write(ring, WRITE_DATA_DST_SEL(1));
 			radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
-			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr));
 			radeon_ring_write(ring, next_rptr);
 		}
 
@@ -5396,6 +5406,7 @@ static int cik_pcie_gart_enable(struct radeon_device *rdev)
 	WREG32(MC_VM_MX_L1_TLB_CNTL,
 	       (0xA << 7) |
 	       ENABLE_L1_TLB |
+	       ENABLE_L1_FRAGMENT_PROCESSING |
 	       SYSTEM_ACCESS_MODE_NOT_IN_SYS |
 	       ENABLE_ADVANCED_DRIVER_MODEL |
 	       SYSTEM_APERTURE_UNMAPPED_ACCESS_PASS_THRU);
@@ -5408,7 +5419,8 @@ static int cik_pcie_gart_enable(struct radeon_device *rdev)
 	       CONTEXT1_IDENTITY_ACCESS_MODE(1));
 	WREG32(VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS | INVALIDATE_L2_CACHE);
 	WREG32(VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY |
-	       L2_CACHE_BIGK_FRAGMENT_SIZE(6));
+	       BANK_SELECT(4) |
+	       L2_CACHE_BIGK_FRAGMENT_SIZE(4));
 	/* setup context0 */
 	WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR, rdev->mc.gtt_start >> 12);
 	WREG32(VM_CONTEXT0_PAGE_TABLE_END_ADDR, rdev->mc.gtt_end >> 12);
@@ -5444,6 +5456,7 @@ static int cik_pcie_gart_enable(struct radeon_device *rdev)
 	       (u32)(rdev->dummy_page.addr >> 12));
 	WREG32(VM_CONTEXT1_CNTL2, 4);
 	WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) |
+				PAGE_TABLE_BLOCK_SIZE(radeon_vm_block_size - 9) |
 				RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT |
 				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
@@ -7450,7 +7463,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.cik.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -7476,7 +7489,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.cik.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
@@ -7502,7 +7515,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_flip(rdev, 2);
+						radeon_crtc_handle_vblank(rdev, 2);
 					rdev->irq.stat_regs.cik.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
@@ -7528,7 +7541,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_flip(rdev, 3);
+						radeon_crtc_handle_vblank(rdev, 3);
 					rdev->irq.stat_regs.cik.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
@@ -7554,7 +7567,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_flip(rdev, 4);
+						radeon_crtc_handle_vblank(rdev, 4);
 					rdev->irq.stat_regs.cik.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
@@ -7580,7 +7593,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_flip(rdev, 5);
+						radeon_crtc_handle_vblank(rdev, 5);
 					rdev->irq.stat_regs.cik.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
@@ -7663,14 +7676,16 @@ restart_ih:
 			addr = RREG32(VM_CONTEXT1_PROTECTION_FAULT_ADDR);
 			status = RREG32(VM_CONTEXT1_PROTECTION_FAULT_STATUS);
 			mc_client = RREG32(VM_CONTEXT1_PROTECTION_FAULT_MCCLIENT);
+			/* reset addr and status */
+			WREG32_P(VM_CONTEXT1_CNTL2, 1, ~1);
+			if (addr == 0x0 && status == 0x0)
+				break;
 			dev_err(rdev->dev, "GPU fault detected: %d 0x%08x\n", src_id, src_data);
 			dev_err(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
 				addr);
 			dev_err(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 				status);
 			cik_vm_decode_fault(rdev, status, addr, mc_client);
-			/* reset addr and status */
-			WREG32_P(VM_CONTEXT1_CNTL2, 1, ~1);
 			break;
 		case 167: /* VCE */
 			DRM_DEBUG("IH: VCE int: 0x%08x\n", src_data);
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index 72e464c79a88..8e9d0f1d858e 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -141,7 +141,7 @@ void cik_sdma_ring_ib_execute(struct radeon_device *rdev,
 		next_rptr += 4;
 		radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0));
 		radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
-		radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+		radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr));
 		radeon_ring_write(ring, 1); /* number of DWs to follow */
 		radeon_ring_write(ring, next_rptr);
 	}
@@ -151,7 +151,7 @@ void cik_sdma_ring_ib_execute(struct radeon_device *rdev,
 		radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_NOP, 0, 0));
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_INDIRECT_BUFFER, 0, extra_bits));
 	radeon_ring_write(ring, ib->gpu_addr & 0xffffffe0); /* base must be 32 byte aligned */
-	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xffffffff);
+	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr));
 	radeon_ring_write(ring, ib->length_dw);
 
 }
@@ -203,8 +203,8 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
 
 	/* write the fence */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_FENCE, 0, 0));
-	radeon_ring_write(ring, addr & 0xffffffff);
-	radeon_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
+	radeon_ring_write(ring, upper_32_bits(addr));
 	radeon_ring_write(ring, fence->seq);
 	/* generate an interrupt */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_TRAP, 0, 0));
@@ -233,7 +233,7 @@ bool cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
 
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SEMAPHORE, 0, extra_bits));
 	radeon_ring_write(ring, addr & 0xfffffff8);
-	radeon_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	radeon_ring_write(ring, upper_32_bits(addr));
 
 	return true;
 }
@@ -551,10 +551,10 @@ int cik_copy_dma(struct radeon_device *rdev,
 		radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_COPY, SDMA_COPY_SUB_OPCODE_LINEAR, 0));
 		radeon_ring_write(ring, cur_size_in_bytes);
 		radeon_ring_write(ring, 0); /* src/dst endian swap */
-		radeon_ring_write(ring, src_offset & 0xffffffff);
-		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xffffffff);
-		radeon_ring_write(ring, dst_offset & 0xffffffff);
-		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(src_offset));
+		radeon_ring_write(ring, upper_32_bits(src_offset));
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
+		radeon_ring_write(ring, upper_32_bits(dst_offset));
 		src_offset += cur_size_in_bytes;
 		dst_offset += cur_size_in_bytes;
 	}
@@ -605,7 +605,7 @@ int cik_sdma_ring_test(struct radeon_device *rdev,
 	}
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0));
 	radeon_ring_write(ring, rdev->vram_scratch.gpu_addr & 0xfffffffc);
-	radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xffffffff);
+	radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr));
 	radeon_ring_write(ring, 1); /* number of DWs to follow */
 	radeon_ring_write(ring, 0xDEADBEEF);
 	radeon_ring_unlock_commit(rdev, ring);
@@ -660,7 +660,7 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 
 	ib.ptr[0] = SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
 	ib.ptr[1] = rdev->vram_scratch.gpu_addr & 0xfffffffc;
-	ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xffffffff;
+	ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr);
 	ib.ptr[3] = 1;
 	ib.ptr[4] = 0xDEADBEEF;
 	ib.length_dw = 5;
@@ -742,7 +742,26 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 
 	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
 
-	if (flags & R600_PTE_SYSTEM) {
+	if (flags == R600_PTE_GART) {
+		uint64_t src = rdev->gart.table_addr + (addr >> 12) * 8;
+		while (count) {
+			unsigned bytes = count * 8;
+			if (bytes > 0x1FFFF8)
+				bytes = 0x1FFFF8;
+
+			ib->ptr[ib->length_dw++] = SDMA_PACKET(SDMA_OPCODE_COPY, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
+			ib->ptr[ib->length_dw++] = bytes;
+			ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */
+			ib->ptr[ib->length_dw++] = lower_32_bits(src);
+			ib->ptr[ib->length_dw++] = upper_32_bits(src);
+			ib->ptr[ib->length_dw++] = lower_32_bits(pe);
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
+
+			pe += bytes;
+			src += bytes;
+			count -= bytes / 8;
+		}
+	} else if (flags & R600_PTE_SYSTEM) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index dd7926394a8f..0c6e1b55d968 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -482,6 +482,7 @@
 #define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
 #define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
 #define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
+#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
 #define VM_CONTEXT1_CNTL				0x1414
 #define VM_CONTEXT0_CNTL2				0x1430
 #define VM_CONTEXT1_CNTL2				0x1434
@@ -1751,12 +1752,12 @@
 #define		EOP_TC_WB_ACTION_EN                     (1 << 15) /* L2 */
 #define		EOP_TCL1_ACTION_EN                      (1 << 16)
 #define		EOP_TC_ACTION_EN                        (1 << 17) /* L2 */
+#define		EOP_TCL2_VOLATILE                       (1 << 24)
 #define		EOP_CACHE_POLICY(x)                     ((x) << 25)
                 /* 0 - LRU
 		 * 1 - Stream
 		 * 2 - Bypass
 		 */
-#define		EOP_TCL2_VOLATILE                       (1 << 27)
 #define		DATA_SEL(x)                             ((x) << 29)
                 /* 0 - discard
 		 * 1 - send low 32bit data
diff --git a/drivers/gpu/drm/radeon/clearstate_cayman.h b/drivers/gpu/drm/radeon/clearstate_cayman.h
index aa908c55a513..e48a14037b76 100644
--- a/drivers/gpu/drm/radeon/clearstate_cayman.h
+++ b/drivers/gpu/drm/radeon/clearstate_cayman.h
@@ -1050,7 +1050,7 @@ static const struct cs_extent_def SECT_CONTEXT_defs[] =
     {SECT_CONTEXT_def_5, 0x0000a29e, 5 },
     {SECT_CONTEXT_def_6, 0x0000a2a5, 56 },
     {SECT_CONTEXT_def_7, 0x0000a2de, 290 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const u32 SECT_CLEAR_def_1[] =
 {
@@ -1061,7 +1061,7 @@ static const u32 SECT_CLEAR_def_1[] =
 static const struct cs_extent_def SECT_CLEAR_defs[] =
 {
     {SECT_CLEAR_def_1, 0x0000ffc0, 3 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const u32 SECT_CTRLCONST_def_1[] =
 {
@@ -1071,11 +1071,11 @@ static const u32 SECT_CTRLCONST_def_1[] =
 static const struct cs_extent_def SECT_CTRLCONST_defs[] =
 {
     {SECT_CTRLCONST_def_1, 0x0000f3fc, 2 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const struct cs_section_def cayman_cs_data[] = {
     { SECT_CONTEXT_defs, SECT_CONTEXT },
     { SECT_CLEAR_defs, SECT_CLEAR },
     { SECT_CTRLCONST_defs, SECT_CTRLCONST },
-    { 0, SECT_NONE }
+    { NULL, SECT_NONE }
 };
diff --git a/drivers/gpu/drm/radeon/clearstate_ci.h b/drivers/gpu/drm/radeon/clearstate_ci.h
index c3982f9475fb..f55d06664e31 100644
--- a/drivers/gpu/drm/radeon/clearstate_ci.h
+++ b/drivers/gpu/drm/radeon/clearstate_ci.h
@@ -936,9 +936,9 @@ static const struct cs_extent_def ci_SECT_CONTEXT_defs[] =
     {ci_SECT_CONTEXT_def_5, 0x0000a2a0, 2 },
     {ci_SECT_CONTEXT_def_6, 0x0000a2a3, 1 },
     {ci_SECT_CONTEXT_def_7, 0x0000a2a5, 233 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const struct cs_section_def ci_cs_data[] = {
     { ci_SECT_CONTEXT_defs, SECT_CONTEXT },
-    { 0, SECT_NONE }
+    { NULL, SECT_NONE }
 };
diff --git a/drivers/gpu/drm/radeon/clearstate_si.h b/drivers/gpu/drm/radeon/clearstate_si.h
index b994cb2a35a0..66e39cdb5cb0 100644
--- a/drivers/gpu/drm/radeon/clearstate_si.h
+++ b/drivers/gpu/drm/radeon/clearstate_si.h
@@ -933,9 +933,9 @@ static const struct cs_extent_def si_SECT_CONTEXT_defs[] =
     {si_SECT_CONTEXT_def_5, 0x0000a2a1, 1 },
     {si_SECT_CONTEXT_def_6, 0x0000a2a3, 1 },
     {si_SECT_CONTEXT_def_7, 0x0000a2a5, 233 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const struct cs_section_def si_cs_data[] = {
     { si_SECT_CONTEXT_defs, SECT_CONTEXT },
-    { 0, SECT_NONE }
+    { NULL, SECT_NONE }
 };
diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c
index 5a9a5f4d7888..47d31e915758 100644
--- a/drivers/gpu/drm/radeon/cypress_dpm.c
+++ b/drivers/gpu/drm/radeon/cypress_dpm.c
@@ -1551,7 +1551,7 @@ int cypress_populate_smc_voltage_tables(struct radeon_device *rdev,
 
 		table->voltageMaskTable.highMask[RV770_SMC_VOLTAGEMASK_VDDCI] = 0;
 		table->voltageMaskTable.lowMask[RV770_SMC_VOLTAGEMASK_VDDCI] =
-			cpu_to_be32(eg_pi->vddc_voltage_table.mask_low);
+			cpu_to_be32(eg_pi->vddci_voltage_table.mask_low);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/radeon/dce3_1_afmt.c b/drivers/gpu/drm/radeon/dce3_1_afmt.c
new file mode 100644
index 000000000000..51800e340a57
--- /dev/null
+++ b/drivers/gpu/drm/radeon/dce3_1_afmt.c
@@ -0,0 +1,244 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ * Copyright 2014 Rafał Miłecki
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <linux/hdmi.h>
+#include <drm/drmP.h>
+#include "radeon.h"
+#include "radeon_asic.h"
+#include "r600d.h"
+
+static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
+{
+	struct radeon_device *rdev = encoder->dev->dev_private;
+	struct drm_connector *connector;
+	struct radeon_connector *radeon_connector = NULL;
+	u32 tmp;
+	u8 *sadb;
+	int sad_count;
+
+	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder) {
+			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
+	}
+
+	if (!radeon_connector) {
+		DRM_ERROR("Couldn't find encoder's connector\n");
+		return;
+	}
+
+	sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb);
+	if (sad_count < 0) {
+		DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
+		return;
+	}
+
+	/* program the speaker allocation */
+	tmp = RREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER);
+	tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK);
+	/* set HDMI mode */
+	tmp |= HDMI_CONNECTION;
+	if (sad_count)
+		tmp |= SPEAKER_ALLOCATION(sadb[0]);
+	else
+		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
+	WREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER, tmp);
+
+	kfree(sadb);
+}
+
+static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder)
+{
+	struct radeon_device *rdev = encoder->dev->dev_private;
+	struct drm_connector *connector;
+	struct radeon_connector *radeon_connector = NULL;
+	struct cea_sad *sads;
+	int i, sad_count;
+
+	static const u16 eld_reg_to_type[][2] = {
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR0, HDMI_AUDIO_CODING_TYPE_PCM },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR1, HDMI_AUDIO_CODING_TYPE_AC3 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR2, HDMI_AUDIO_CODING_TYPE_MPEG1 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR3, HDMI_AUDIO_CODING_TYPE_MP3 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR4, HDMI_AUDIO_CODING_TYPE_MPEG2 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR5, HDMI_AUDIO_CODING_TYPE_AAC_LC },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR6, HDMI_AUDIO_CODING_TYPE_DTS },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR7, HDMI_AUDIO_CODING_TYPE_ATRAC },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR9, HDMI_AUDIO_CODING_TYPE_EAC3 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR10, HDMI_AUDIO_CODING_TYPE_DTS_HD },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR11, HDMI_AUDIO_CODING_TYPE_MLP },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
+	};
+
+	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder) {
+			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
+	}
+
+	if (!radeon_connector) {
+		DRM_ERROR("Couldn't find encoder's connector\n");
+		return;
+	}
+
+	sad_count = drm_edid_to_sad(radeon_connector->edid, &sads);
+	if (sad_count < 0) {
+		DRM_ERROR("Couldn't read SADs: %d\n", sad_count);
+		return;
+	}
+	BUG_ON(!sads);
+
+	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
+		u32 value = 0;
+		u8 stereo_freqs = 0;
+		int max_channels = -1;
+		int j;
+
+		for (j = 0; j < sad_count; j++) {
+			struct cea_sad *sad = &sads[j];
+
+			if (sad->format == eld_reg_to_type[i][1]) {
+				if (sad->channels > max_channels) {
+					value = MAX_CHANNELS(sad->channels) |
+						DESCRIPTOR_BYTE_2(sad->byte2) |
+						SUPPORTED_FREQUENCIES(sad->freq);
+					max_channels = sad->channels;
+				}
+
+				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
+					stereo_freqs |= sad->freq;
+				else
+					break;
+			}
+		}
+
+		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
+
+		WREG32(eld_reg_to_type[i][0], value);
+	}
+
+	kfree(sads);
+}
+
+/*
+ * update the info frames with the data from the current display mode
+ */
+void dce3_1_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
+	struct hdmi_avi_infoframe frame;
+	uint32_t offset;
+	ssize_t err;
+
+	if (!dig || !dig->afmt)
+		return;
+
+	/* Silent, r600_hdmi_enable will raise WARN for us */
+	if (!dig->afmt->enabled)
+		return;
+	offset = dig->afmt->offset;
+
+	/* disable audio prior to setting up hw */
+	dig->afmt->pin = r600_audio_get_pin(rdev);
+	r600_audio_enable(rdev, dig->afmt->pin, false);
+
+	r600_audio_set_dto(encoder, mode->clock);
+
+	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
+	       HDMI0_NULL_SEND); /* send null packets when required */
+
+	WREG32(HDMI0_AUDIO_CRC_CONTROL + offset, 0x1000);
+
+	if (ASIC_IS_DCE32(rdev)) {
+		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
+		       HDMI0_AUDIO_PACKETS_PER_LINE(3)); /* should be suffient for all audio modes and small enough for all hblanks */
+		WREG32(AFMT_AUDIO_PACKET_CONTROL + offset,
+		       AFMT_AUDIO_SAMPLE_SEND | /* send audio packets */
+		       AFMT_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
+	} else {
+		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		       HDMI0_AUDIO_SAMPLE_SEND | /* send audio packets */
+		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
+		       HDMI0_AUDIO_PACKETS_PER_LINE(3) | /* should be suffient for all audio modes and small enough for all hblanks */
+		       HDMI0_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
+	}
+
+	if (ASIC_IS_DCE32(rdev)) {
+		dce3_2_afmt_write_speaker_allocation(encoder);
+		dce3_2_afmt_write_sad_regs(encoder);
+	}
+
+	WREG32(HDMI0_ACR_PACKET_CONTROL + offset,
+	       HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */
+	       HDMI0_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
+
+	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
+	       HDMI0_NULL_SEND | /* send null packets when required */
+	       HDMI0_GC_SEND | /* send general control packets */
+	       HDMI0_GC_CONT); /* send general control packets every frame */
+
+	/* TODO: HDMI0_AUDIO_INFO_UPDATE */
+	WREG32(HDMI0_INFOFRAME_CONTROL0 + offset,
+	       HDMI0_AVI_INFO_SEND | /* enable AVI info frames */
+	       HDMI0_AVI_INFO_CONT | /* send AVI info frames every frame/field */
+	       HDMI0_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */
+	       HDMI0_AUDIO_INFO_CONT); /* send audio info frames every frame/field */
+
+	WREG32(HDMI0_INFOFRAME_CONTROL1 + offset,
+	       HDMI0_AVI_INFO_LINE(2) | /* anything other than 0 */
+	       HDMI0_AUDIO_INFO_LINE(2)); /* anything other than 0 */
+
+	WREG32(HDMI0_GC + offset, 0); /* unset HDMI0_GC_AVMUTE */
+
+	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
+	if (err < 0) {
+		DRM_ERROR("failed to setup AVI infoframe: %zd\n", err);
+		return;
+	}
+
+	err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		DRM_ERROR("failed to pack AVI infoframe: %zd\n", err);
+		return;
+	}
+
+	r600_hdmi_update_avi_infoframe(encoder, buffer, sizeof(buffer));
+	r600_hdmi_update_ACR(encoder, mode->clock);
+
+	/* it's unknown what these bits do excatly, but it's indeed quite useful for debugging */
+	WREG32(HDMI0_RAMP_CONTROL0 + offset, 0x00FFFFFF);
+	WREG32(HDMI0_RAMP_CONTROL1 + offset, 0x007FFFFF);
+	WREG32(HDMI0_RAMP_CONTROL2 + offset, 0x00000001);
+	WREG32(HDMI0_RAMP_CONTROL3 + offset, 0x00000001);
+
+	r600_hdmi_audio_workaround(encoder);
+
+	/* enable audio after to setting up hw */
+	r600_audio_enable(rdev, dig->afmt->pin, true);
+}
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 0f7a51a3694f..f7ece0ff431b 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -189,7 +189,7 @@ static const u32 evergreen_golden_registers[] =
 	0x8c1c, 0xffffffff, 0x00001010,
 	0x28350, 0xffffffff, 0x00000000,
 	0xa008, 0xffffffff, 0x00010000,
-	0x5cc, 0xffffffff, 0x00000001,
+	0x5c4, 0xffffffff, 0x00000001,
 	0x9508, 0xffffffff, 0x00000002,
 	0x913c, 0x0000000f, 0x0000000a
 };
@@ -476,7 +476,7 @@ static const u32 cedar_golden_registers[] =
 	0x8c1c, 0xffffffff, 0x00001010,
 	0x28350, 0xffffffff, 0x00000000,
 	0xa008, 0xffffffff, 0x00010000,
-	0x5cc, 0xffffffff, 0x00000001,
+	0x5c4, 0xffffffff, 0x00000001,
 	0x9508, 0xffffffff, 0x00000002
 };
 
@@ -635,7 +635,7 @@ static const u32 juniper_mgcg_init[] =
 static const u32 supersumo_golden_registers[] =
 {
 	0x5eb4, 0xffffffff, 0x00000002,
-	0x5cc, 0xffffffff, 0x00000001,
+	0x5c4, 0xffffffff, 0x00000001,
 	0x7030, 0xffffffff, 0x00000011,
 	0x7c30, 0xffffffff, 0x00000011,
 	0x6104, 0x01000300, 0x00000000,
@@ -719,7 +719,7 @@ static const u32 sumo_golden_registers[] =
 static const u32 wrestler_golden_registers[] =
 {
 	0x5eb4, 0xffffffff, 0x00000002,
-	0x5cc, 0xffffffff, 0x00000001,
+	0x5c4, 0xffffffff, 0x00000001,
 	0x7030, 0xffffffff, 0x00000011,
 	0x7c30, 0xffffffff, 0x00000011,
 	0x6104, 0x01000300, 0x00000000,
@@ -1301,36 +1301,6 @@ void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc)
 }
 
 /**
- * radeon_irq_kms_pflip_irq_get - pre-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to prepare for pageflip on
- *
- * Pre-pageflip callback (evergreen+).
- * Enables the pageflip irq (vblank irq).
- */
-void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* enable the pflip int */
-	radeon_irq_kms_pflip_irq_get(rdev, crtc);
-}
-
-/**
- * evergreen_post_page_flip - pos-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to cleanup pageflip on
- *
- * Post-pageflip callback (evergreen+).
- * Disables the pageflip irq (vblank irq).
- */
-void evergreen_post_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* disable the pflip int */
-	radeon_irq_kms_pflip_irq_put(rdev, crtc);
-}
-
-/**
  * evergreen_page_flip - pageflip callback.
  *
  * @rdev: radeon_device pointer
@@ -1343,7 +1313,7 @@ void evergreen_post_page_flip(struct radeon_device *rdev, int crtc)
  * double buffered update to take place.
  * Returns the current update pending status.
  */
-u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
@@ -1375,9 +1345,23 @@ u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~EVERGREEN_GRPH_UPDATE_LOCK;
 	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+}
+
+/**
+ * evergreen_page_flip_pending - check if page flip is still pending
+ *
+ * @rdev: radeon_device pointer
+ * @crtc_id: crtc to check
+ *
+ * Returns the current update pending status.
+ */
+bool evergreen_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 
 	/* Return current update_pending status: */
-	return RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING;
+	return !!(RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) &
+		EVERGREEN_GRPH_SURFACE_UPDATE_PENDING);
 }
 
 /* get temperature in millidegrees */
@@ -3353,6 +3337,18 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 			disabled_rb_mask &= ~(1 << i);
 	}
 
+	for (i = 0; i < rdev->config.evergreen.num_ses; i++) {
+		u32 simd_disable_bitmap;
+
+		WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		simd_disable_bitmap = (RREG32(CC_GC_SHADER_PIPE_CONFIG) & 0xffff0000) >> 16;
+		simd_disable_bitmap |= 0xffffffff << rdev->config.evergreen.max_simds;
+		tmp <<= 16;
+		tmp |= simd_disable_bitmap;
+	}
+	rdev->config.evergreen.active_simds = hweight32(~tmp);
+
 	WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 	WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 
@@ -4810,7 +4806,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -4836,7 +4832,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
@@ -4862,7 +4858,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_flip(rdev, 2);
+						radeon_crtc_handle_vblank(rdev, 2);
 					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
@@ -4888,7 +4884,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_flip(rdev, 3);
+						radeon_crtc_handle_vblank(rdev, 3);
 					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
@@ -4914,7 +4910,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_flip(rdev, 4);
+						radeon_crtc_handle_vblank(rdev, 4);
 					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
@@ -4940,7 +4936,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_flip(rdev, 5);
+						radeon_crtc_handle_vblank(rdev, 5);
 					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
@@ -5070,14 +5066,16 @@ restart_ih:
 		case 147:
 			addr = RREG32(VM_CONTEXT1_PROTECTION_FAULT_ADDR);
 			status = RREG32(VM_CONTEXT1_PROTECTION_FAULT_STATUS);
+			/* reset addr and status */
+			WREG32_P(VM_CONTEXT1_CNTL2, 1, ~1);
+			if (addr == 0x0 && status == 0x0)
+				break;
 			dev_err(rdev->dev, "GPU fault detected: %d 0x%08x\n", src_id, src_data);
 			dev_err(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
 				addr);
 			dev_err(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 				status);
 			cayman_vm_decode_fault(rdev, status, addr);
-			/* reset addr and status */
-			WREG32_P(VM_CONTEXT1_CNTL2, 1, ~1);
 			break;
 		case 176: /* CP_INT in ring buffer */
 		case 177: /* CP_INT in IB1 */
diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 05b0c95813fd..1ec0e6e83f9f 100644
--- a/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -293,10 +293,13 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
 	struct hdmi_avi_infoframe frame;
 	uint32_t offset;
 	ssize_t err;
+	uint32_t val;
+	int bpc = 8;
 
 	if (!dig || !dig->afmt)
 		return;
@@ -306,6 +309,12 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 		return;
 	offset = dig->afmt->offset;
 
+	/* hdmi deep color mode general control packets setup, if bpc > 8 */
+	if (encoder->crtc) {
+		struct radeon_crtc *radeon_crtc = to_radeon_crtc(encoder->crtc);
+		bpc = radeon_crtc->bpc;
+	}
+
 	/* disable audio prior to setting up hw */
 	if (ASIC_IS_DCE6(rdev)) {
 		dig->afmt->pin = dce6_audio_get_pin(rdev);
@@ -322,6 +331,35 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 
 	WREG32(AFMT_AUDIO_CRC_CONTROL + offset, 0x1000);
 
+	val = RREG32(HDMI_CONTROL + offset);
+	val &= ~HDMI_DEEP_COLOR_ENABLE;
+	val &= ~HDMI_DEEP_COLOR_DEPTH_MASK;
+
+	switch (bpc) {
+		case 0:
+		case 6:
+		case 8:
+		case 16:
+		default:
+			DRM_DEBUG("%s: Disabling hdmi deep color for %d bpc.\n",
+					 connector->name, bpc);
+			break;
+		case 10:
+			val |= HDMI_DEEP_COLOR_ENABLE;
+			val |= HDMI_DEEP_COLOR_DEPTH(HDMI_30BIT_DEEP_COLOR);
+			DRM_DEBUG("%s: Enabling hdmi deep color 30 for 10 bpc.\n",
+					 connector->name);
+			break;
+		case 12:
+			val |= HDMI_DEEP_COLOR_ENABLE;
+			val |= HDMI_DEEP_COLOR_DEPTH(HDMI_36BIT_DEEP_COLOR);
+			DRM_DEBUG("%s: Enabling hdmi deep color 36 for 12 bpc.\n",
+					 connector->name);
+			break;
+	}
+
+	WREG32(HDMI_CONTROL + offset, val);
+
 	WREG32(HDMI_VBI_PACKET_CONTROL + offset,
 	       HDMI_NULL_SEND | /* send null packets when required */
 	       HDMI_GC_SEND | /* send general control packets */
@@ -348,9 +386,13 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 
 	/* fglrx clears sth in AFMT_AUDIO_PACKET_CONTROL2 here */
 
-	WREG32(HDMI_ACR_PACKET_CONTROL + offset,
-	       HDMI_ACR_SOURCE | /* select SW CTS value */
-	       HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
+	if (bpc > 8)
+		WREG32(HDMI_ACR_PACKET_CONTROL + offset,
+		       HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
+	else
+		WREG32(HDMI_ACR_PACKET_CONTROL + offset,
+		       HDMI_ACR_SOURCE | /* select SW CTS value */
+		       HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
 
 	evergreen_hdmi_update_ACR(encoder, mode->clock);
 
diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h
index a0f63ff5a5e9..333d143fca2c 100644
--- a/drivers/gpu/drm/radeon/evergreen_reg.h
+++ b/drivers/gpu/drm/radeon/evergreen_reg.h
@@ -116,6 +116,8 @@
 #       define EVERGREEN_GRPH_ARRAY_LINEAR_ALIGNED      1
 #       define EVERGREEN_GRPH_ARRAY_1D_TILED_THIN1      2
 #       define EVERGREEN_GRPH_ARRAY_2D_TILED_THIN1      4
+#define EVERGREEN_GRPH_LUT_10BIT_BYPASS_CONTROL         0x6808
+#       define EVERGREEN_LUT_10BIT_BYPASS_EN            (1 << 8)
 #define EVERGREEN_GRPH_SWAP_CONTROL                     0x680c
 #       define EVERGREEN_GRPH_ENDIAN_SWAP(x)            (((x) & 0x3) << 0)
 #       define EVERGREEN_GRPH_ENDIAN_NONE               0
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index f9c7963b3ee6..b066d6711b8d 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -517,10 +517,11 @@
 #       define HDMI_ERROR_ACK                (1 << 8)
 #       define HDMI_ERROR_MASK               (1 << 9)
 #       define HDMI_DEEP_COLOR_ENABLE        (1 << 24)
-#       define HDMI_DEEP_COLOR_DEPTH         (((x) & 3) << 28)
+#       define HDMI_DEEP_COLOR_DEPTH(x)      (((x) & 3) << 28)
 #       define HDMI_24BIT_DEEP_COLOR         0
 #       define HDMI_30BIT_DEEP_COLOR         1
 #       define HDMI_36BIT_DEEP_COLOR         2
+#       define HDMI_DEEP_COLOR_DEPTH_MASK    (3 << 28)
 #define HDMI_STATUS                          0x7034
 #       define HDMI_ACTIVE_AVMUTE            (1 << 0)
 #       define HDMI_AUDIO_PACKET_ERROR       (1 << 16)
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index 3f6e817d97ee..9ef8c38f2d66 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -2726,7 +2726,7 @@ int kv_dpm_init(struct radeon_device *rdev)
 	pi->caps_sclk_ds = true;
 	pi->enable_auto_thermal_throttling = true;
 	pi->disable_nb_ps3_in_battery = false;
-	pi->bapm_enable = false;
+	pi->bapm_enable = true;
 	pi->voltage_drop_t = 0;
 	pi->caps_sclk_throttle_low_notification = false;
 	pi->caps_fps = false; /* true? */
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index d246e043421a..5a33ca681867 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1057,6 +1057,18 @@ static void cayman_gpu_init(struct radeon_device *rdev)
 			disabled_rb_mask &= ~(1 << i);
 	}
 
+	for (i = 0; i < rdev->config.cayman.max_shader_engines; i++) {
+		u32 simd_disable_bitmap;
+
+		WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		simd_disable_bitmap = (RREG32(CC_GC_SHADER_PIPE_CONFIG) & 0xffff0000) >> 16;
+		simd_disable_bitmap |= 0xffffffff << rdev->config.cayman.max_simds_per_se;
+		tmp <<= 16;
+		tmp |= simd_disable_bitmap;
+	}
+	rdev->config.cayman.active_simds = hweight32(~tmp);
+
 	WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 	WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 
@@ -1228,12 +1240,14 @@ static int cayman_pcie_gart_enable(struct radeon_device *rdev)
 	       SYSTEM_APERTURE_UNMAPPED_ACCESS_PASS_THRU);
 	/* Setup L2 cache */
 	WREG32(VM_L2_CNTL, ENABLE_L2_CACHE |
+	       ENABLE_L2_FRAGMENT_PROCESSING |
 	       ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE |
 	       ENABLE_L2_PDE0_CACHE_LRU_UPDATE_BY_WRITE |
 	       EFFECTIVE_L2_QUEUE_SIZE(7) |
 	       CONTEXT1_IDENTITY_ACCESS_MODE(1));
 	WREG32(VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS | INVALIDATE_L2_CACHE);
 	WREG32(VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY |
+	       BANK_SELECT(6) |
 	       L2_CACHE_BIGK_FRAGMENT_SIZE(6));
 	/* setup context0 */
 	WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR, rdev->mc.gtt_start >> 12);
@@ -1266,6 +1280,7 @@ static int cayman_pcie_gart_enable(struct radeon_device *rdev)
 	       (u32)(rdev->dummy_page.addr >> 12));
 	WREG32(VM_CONTEXT1_CNTL2, 4);
 	WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) |
+				PAGE_TABLE_BLOCK_SIZE(radeon_vm_block_size - 9) |
 				RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT |
 				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
@@ -1343,7 +1358,7 @@ void cayman_fence_ring_emit(struct radeon_device *rdev,
 	/* EVENT_WRITE_EOP - flush caches, send int */
 	radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
 	radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
 	radeon_ring_write(ring, fence->seq);
 	radeon_ring_write(ring, 0);
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index 004c931606c4..01fc4888e6fe 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -1315,7 +1315,7 @@ static void ni_populate_smc_voltage_tables(struct radeon_device *rdev,
 
 		table->voltageMaskTable.highMask[NISLANDS_SMC_VOLTAGEMASK_VDDCI] = 0;
 		table->voltageMaskTable.lowMask[NISLANDS_SMC_VOLTAGEMASK_VDDCI] =
-			cpu_to_be32(eg_pi->vddc_voltage_table.mask_low);
+			cpu_to_be32(eg_pi->vddci_voltage_table.mask_low);
 	}
 }
 
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index d996033c243e..2e12e4d69253 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -128,6 +128,7 @@
 #define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
 #define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
 #define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
+#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
 #define VM_CONTEXT1_CNTL				0x1414
 #define VM_CONTEXT0_CNTL2				0x1430
 #define VM_CONTEXT1_CNTL2				0x1434
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index b6c32640df20..1544efcf1c3a 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -142,36 +142,6 @@ void r100_wait_for_vblank(struct radeon_device *rdev, int crtc)
 }
 
 /**
- * r100_pre_page_flip - pre-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to prepare for pageflip on
- *
- * Pre-pageflip callback (r1xx-r4xx).
- * Enables the pageflip irq (vblank irq).
- */
-void r100_pre_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* enable the pflip int */
-	radeon_irq_kms_pflip_irq_get(rdev, crtc);
-}
-
-/**
- * r100_post_page_flip - pos-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to cleanup pageflip on
- *
- * Post-pageflip callback (r1xx-r4xx).
- * Disables the pageflip irq (vblank irq).
- */
-void r100_post_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* disable the pflip int */
-	radeon_irq_kms_pflip_irq_put(rdev, crtc);
-}
-
-/**
  * r100_page_flip - pageflip callback.
  *
  * @rdev: radeon_device pointer
@@ -182,9 +152,8 @@ void r100_post_page_flip(struct radeon_device *rdev, int crtc)
  * During vblank we take the crtc lock and wait for the update_pending
  * bit to go high, when it does, we release the lock, and allow the
  * double buffered update to take place.
- * Returns the current update pending status.
  */
-u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = ((u32)crtc_base) | RADEON_CRTC_OFFSET__OFFSET_LOCK;
@@ -206,8 +175,24 @@ u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	tmp &= ~RADEON_CRTC_OFFSET__OFFSET_LOCK;
 	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
 
+}
+
+/**
+ * r100_page_flip_pending - check if page flip is still pending
+ *
+ * @rdev: radeon_device pointer
+ * @crtc_id: crtc to check
+ *
+ * Check if the last pagefilp is still pending (r1xx-r4xx).
+ * Returns the current update pending status.
+ */
+bool r100_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+
 	/* Return current update_pending status: */
-	return RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET;
+	return !!(RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) &
+		RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET);
 }
 
 /**
@@ -697,15 +682,11 @@ void r100_pci_gart_disable(struct radeon_device *rdev)
 	WREG32(RADEON_AIC_HI_ADDR, 0);
 }
 
-int r100_pci_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void r100_pci_gart_set_page(struct radeon_device *rdev, unsigned i,
+			    uint64_t addr)
 {
 	u32 *gtt = rdev->gart.ptr;
-
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
 	gtt[i] = cpu_to_le32(lower_32_bits(addr));
-	return 0;
 }
 
 void r100_pci_gart_fini(struct radeon_device *rdev)
@@ -794,7 +775,7 @@ int r100_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[0]))
-				radeon_crtc_handle_flip(rdev, 0);
+				radeon_crtc_handle_vblank(rdev, 0);
 		}
 		if (status & RADEON_CRTC2_VBLANK_STAT) {
 			if (rdev->irq.crtc_vblank_int[1]) {
@@ -803,7 +784,7 @@ int r100_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[1]))
-				radeon_crtc_handle_flip(rdev, 1);
+				radeon_crtc_handle_vblank(rdev, 1);
 		}
 		if (status & RADEON_FP_DETECT_STAT) {
 			queue_hotplug = true;
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 206caf9700b7..3c21d77a483d 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -72,13 +72,11 @@ void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev)
 #define R300_PTE_WRITEABLE (1 << 2)
 #define R300_PTE_READABLE  (1 << 3)
 
-int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
+			      uint64_t addr)
 {
 	void __iomem *ptr = rdev->gart.ptr;
 
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
 	addr = (lower_32_bits(addr) >> 8) |
 	       ((upper_32_bits(addr) & 0xff) << 24) |
 	       R300_PTE_WRITEABLE | R300_PTE_READABLE;
@@ -86,7 +84,6 @@ int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
 	 * on powerpc without HW swappers, it'll get swapped on way
 	 * into VRAM - so no need for cpu_to_le32 on VRAM tables */
 	writel(addr, ((void __iomem *)ptr) + (i * 4));
-	return 0;
 }
 
 int rv370_pcie_gart_init(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/r500_reg.h b/drivers/gpu/drm/radeon/r500_reg.h
index 1dd0d32993d5..136b7bc7cd20 100644
--- a/drivers/gpu/drm/radeon/r500_reg.h
+++ b/drivers/gpu/drm/radeon/r500_reg.h
@@ -402,6 +402,7 @@
  * block and vice versa.  This applies to GRPH, CUR, etc.
  */
 #define AVIVO_D1GRPH_LUT_SEL                                    0x6108
+#       define AVIVO_LUT_10BIT_BYPASS_EN                        (1 << 8)
 #define AVIVO_D1GRPH_PRIMARY_SURFACE_ADDRESS                    0x6110
 #define R700_D1GRPH_PRIMARY_SURFACE_ADDRESS_HIGH                0x6914
 #define R700_D2GRPH_PRIMARY_SURFACE_ADDRESS_HIGH                0x6114
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index bbc189fd3ddc..c66952d4b00c 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1958,6 +1958,9 @@ static void r600_gpu_init(struct radeon_device *rdev)
 	if (tmp < rdev->config.r600.max_simds) {
 		rdev->config.r600.max_simds = tmp;
 	}
+	tmp = rdev->config.r600.max_simds -
+		r600_count_pipe_bits((cc_gc_shader_pipe_config >> 16) & R6XX_MAX_SIMDS_MASK);
+	rdev->config.r600.active_simds = tmp;
 
 	disabled_rb_mask = (RREG32(CC_RB_BACKEND_DISABLE) >> 16) & R6XX_MAX_BACKENDS_MASK;
 	tmp = (tiling_config & PIPE_TILING__MASK) >> PIPE_TILING__SHIFT;
@@ -2724,7 +2727,7 @@ void r600_fence_ring_emit(struct radeon_device *rdev,
 		/* EVENT_WRITE_EOP - flush caches, send int */
 		radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
 		radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
-		radeon_ring_write(ring, addr & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(addr));
 		radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
 		radeon_ring_write(ring, fence->seq);
 		radeon_ring_write(ring, 0);
@@ -2763,7 +2766,7 @@ bool r600_semaphore_ring_emit(struct radeon_device *rdev,
 		sel |= PACKET3_SEM_WAIT_ON_SIGNAL;
 
 	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | sel);
 
 	return true;
@@ -2824,9 +2827,9 @@ int r600_copy_cpdma(struct radeon_device *rdev,
 		if (size_in_bytes == 0)
 			tmp |= PACKET3_CP_DMA_CP_SYNC;
 		radeon_ring_write(ring, PACKET3(PACKET3_CP_DMA, 4));
-		radeon_ring_write(ring, src_offset & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(src_offset));
 		radeon_ring_write(ring, tmp);
-		radeon_ring_write(ring, dst_offset & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
 		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
 		radeon_ring_write(ring, cur_size_in_bytes);
 		src_offset += cur_size_in_bytes;
@@ -3876,7 +3879,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -3902,7 +3905,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index 85a2bb28aed2..26ef8ced6f89 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -133,7 +133,7 @@ struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock)
 /*
  * update the N and CTS parameters for a given pixel clock rate
  */
-static void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
+void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -142,21 +142,33 @@ static void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	uint32_t offset = dig->afmt->offset;
 
-	WREG32(HDMI0_ACR_32_0 + offset, HDMI0_ACR_CTS_32(acr.cts_32khz));
-	WREG32(HDMI0_ACR_32_1 + offset, acr.n_32khz);
-
-	WREG32(HDMI0_ACR_44_0 + offset, HDMI0_ACR_CTS_44(acr.cts_44_1khz));
-	WREG32(HDMI0_ACR_44_1 + offset, acr.n_44_1khz);
-
-	WREG32(HDMI0_ACR_48_0 + offset, HDMI0_ACR_CTS_48(acr.cts_48khz));
-	WREG32(HDMI0_ACR_48_1 + offset, acr.n_48khz);
+	WREG32_P(HDMI0_ACR_32_0 + offset,
+		 HDMI0_ACR_CTS_32(acr.cts_32khz),
+		 ~HDMI0_ACR_CTS_32_MASK);
+	WREG32_P(HDMI0_ACR_32_1 + offset,
+		 HDMI0_ACR_N_32(acr.n_32khz),
+		 ~HDMI0_ACR_N_32_MASK);
+
+	WREG32_P(HDMI0_ACR_44_0 + offset,
+		 HDMI0_ACR_CTS_44(acr.cts_44_1khz),
+		 ~HDMI0_ACR_CTS_44_MASK);
+	WREG32_P(HDMI0_ACR_44_1 + offset,
+		 HDMI0_ACR_N_44(acr.n_44_1khz),
+		 ~HDMI0_ACR_N_44_MASK);
+
+	WREG32_P(HDMI0_ACR_48_0 + offset,
+		 HDMI0_ACR_CTS_48(acr.cts_48khz),
+		 ~HDMI0_ACR_CTS_48_MASK);
+	WREG32_P(HDMI0_ACR_48_1 + offset,
+		 HDMI0_ACR_N_48(acr.n_48khz),
+		 ~HDMI0_ACR_N_48_MASK);
 }
 
 /*
  * build a HDMI Video Info Frame
  */
-static void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder,
-					   void *buffer, size_t size)
+void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder, void *buffer,
+				    size_t size)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -231,7 +243,7 @@ int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder)
 /*
  * write the audio workaround status to the hardware
  */
-static void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
+void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -250,7 +262,7 @@ static void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
 		 value, ~HDMI0_AUDIO_TEST_EN);
 }
 
-static void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
+void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -320,121 +332,6 @@ static void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 	}
 }
 
-static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
-{
-	struct radeon_device *rdev = encoder->dev->dev_private;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
-	u32 tmp;
-	u8 *sadb;
-	int sad_count;
-
-	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb);
-	if (sad_count < 0) {
-		DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
-		return;
-	}
-
-	/* program the speaker allocation */
-	tmp = RREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER);
-	tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK);
-	/* set HDMI mode */
-	tmp |= HDMI_CONNECTION;
-	if (sad_count)
-		tmp |= SPEAKER_ALLOCATION(sadb[0]);
-	else
-		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
-	WREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER, tmp);
-
-	kfree(sadb);
-}
-
-static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder)
-{
-	struct radeon_device *rdev = encoder->dev->dev_private;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
-	struct cea_sad *sads;
-	int i, sad_count;
-
-	static const u16 eld_reg_to_type[][2] = {
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR0, HDMI_AUDIO_CODING_TYPE_PCM },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR1, HDMI_AUDIO_CODING_TYPE_AC3 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR2, HDMI_AUDIO_CODING_TYPE_MPEG1 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR3, HDMI_AUDIO_CODING_TYPE_MP3 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR4, HDMI_AUDIO_CODING_TYPE_MPEG2 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR5, HDMI_AUDIO_CODING_TYPE_AAC_LC },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR6, HDMI_AUDIO_CODING_TYPE_DTS },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR7, HDMI_AUDIO_CODING_TYPE_ATRAC },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR9, HDMI_AUDIO_CODING_TYPE_EAC3 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR10, HDMI_AUDIO_CODING_TYPE_DTS_HD },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR11, HDMI_AUDIO_CODING_TYPE_MLP },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
-	};
-
-	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	sad_count = drm_edid_to_sad(radeon_connector->edid, &sads);
-	if (sad_count < 0) {
-		DRM_ERROR("Couldn't read SADs: %d\n", sad_count);
-		return;
-	}
-	BUG_ON(!sads);
-
-	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
-		u32 value = 0;
-		u8 stereo_freqs = 0;
-		int max_channels = -1;
-		int j;
-
-		for (j = 0; j < sad_count; j++) {
-			struct cea_sad *sad = &sads[j];
-
-			if (sad->format == eld_reg_to_type[i][1]) {
-				if (sad->channels > max_channels) {
-					value = MAX_CHANNELS(sad->channels) |
-						DESCRIPTOR_BYTE_2(sad->byte2) |
-						SUPPORTED_FREQUENCIES(sad->freq);
-					max_channels = sad->channels;
-				}
-
-				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
-					stereo_freqs |= sad->freq;
-				else
-					break;
-			}
-		}
-
-		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
-
-		WREG32(eld_reg_to_type[i][0], value);
-	}
-
-	kfree(sads);
-}
-
 /*
  * update the info frames with the data from the current display mode
  */
@@ -447,6 +344,7 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
 	struct hdmi_avi_infoframe frame;
 	uint32_t offset;
+	uint32_t acr_ctl;
 	ssize_t err;
 
 	if (!dig || !dig->afmt)
@@ -463,52 +361,44 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 
 	r600_audio_set_dto(encoder, mode->clock);
 
-	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
-	       HDMI0_NULL_SEND); /* send null packets when required */
-
-	WREG32(HDMI0_AUDIO_CRC_CONTROL + offset, 0x1000);
-
-	if (ASIC_IS_DCE32(rdev)) {
-		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
-		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
-		       HDMI0_AUDIO_PACKETS_PER_LINE(3)); /* should be suffient for all audio modes and small enough for all hblanks */
-		WREG32(AFMT_AUDIO_PACKET_CONTROL + offset,
-		       AFMT_AUDIO_SAMPLE_SEND | /* send audio packets */
-		       AFMT_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
-	} else {
-		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
-		       HDMI0_AUDIO_SAMPLE_SEND | /* send audio packets */
-		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
-		       HDMI0_AUDIO_PACKETS_PER_LINE(3) | /* should be suffient for all audio modes and small enough for all hblanks */
-		       HDMI0_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
-	}
-
-	if (ASIC_IS_DCE32(rdev)) {
-		dce3_2_afmt_write_speaker_allocation(encoder);
-		dce3_2_afmt_write_sad_regs(encoder);
-	}
-
-	WREG32(HDMI0_ACR_PACKET_CONTROL + offset,
-	       HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */
-	       HDMI0_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
-
-	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
-	       HDMI0_NULL_SEND | /* send null packets when required */
-	       HDMI0_GC_SEND | /* send general control packets */
-	       HDMI0_GC_CONT); /* send general control packets every frame */
-
-	/* TODO: HDMI0_AUDIO_INFO_UPDATE */
-	WREG32(HDMI0_INFOFRAME_CONTROL0 + offset,
-	       HDMI0_AVI_INFO_SEND | /* enable AVI info frames */
-	       HDMI0_AVI_INFO_CONT | /* send AVI info frames every frame/field */
-	       HDMI0_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */
-	       HDMI0_AUDIO_INFO_CONT); /* send audio info frames every frame/field */
-
-	WREG32(HDMI0_INFOFRAME_CONTROL1 + offset,
-	       HDMI0_AVI_INFO_LINE(2) | /* anything other than 0 */
-	       HDMI0_AUDIO_INFO_LINE(2)); /* anything other than 0 */
-
-	WREG32(HDMI0_GC + offset, 0); /* unset HDMI0_GC_AVMUTE */
+	WREG32_P(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		 HDMI0_AUDIO_SAMPLE_SEND | /* send audio packets */
+		 HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
+		 HDMI0_AUDIO_PACKETS_PER_LINE(3) | /* should be suffient for all audio modes and small enough for all hblanks */
+		 HDMI0_60958_CS_UPDATE, /* allow 60958 channel status fields to be updated */
+		 ~(HDMI0_AUDIO_SAMPLE_SEND |
+		   HDMI0_AUDIO_DELAY_EN_MASK |
+		   HDMI0_AUDIO_PACKETS_PER_LINE_MASK |
+		   HDMI0_60958_CS_UPDATE));
+
+	/* DCE 3.0 uses register that's normally for CRC_CONTROL */
+	acr_ctl = ASIC_IS_DCE3(rdev) ? DCE3_HDMI0_ACR_PACKET_CONTROL :
+				       HDMI0_ACR_PACKET_CONTROL;
+	WREG32_P(acr_ctl + offset,
+		 HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */
+		 HDMI0_ACR_AUTO_SEND, /* allow hw to sent ACR packets when required */
+		 ~(HDMI0_ACR_SOURCE |
+		   HDMI0_ACR_AUTO_SEND));
+
+	WREG32_OR(HDMI0_VBI_PACKET_CONTROL + offset,
+		  HDMI0_NULL_SEND | /* send null packets when required */
+		  HDMI0_GC_SEND | /* send general control packets */
+		  HDMI0_GC_CONT); /* send general control packets every frame */
+
+	WREG32_OR(HDMI0_INFOFRAME_CONTROL0 + offset,
+		  HDMI0_AVI_INFO_SEND | /* enable AVI info frames */
+		  HDMI0_AVI_INFO_CONT | /* send AVI info frames every frame/field */
+		  HDMI0_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */
+		  HDMI0_AUDIO_INFO_UPDATE); /* required for audio info values to be updated */
+
+	WREG32_P(HDMI0_INFOFRAME_CONTROL1 + offset,
+		 HDMI0_AVI_INFO_LINE(2) | /* anything other than 0 */
+		 HDMI0_AUDIO_INFO_LINE(2), /* anything other than 0 */
+		 ~(HDMI0_AVI_INFO_LINE_MASK |
+		   HDMI0_AUDIO_INFO_LINE_MASK));
+
+	WREG32_AND(HDMI0_GC + offset,
+		   ~HDMI0_GC_AVMUTE); /* unset HDMI0_GC_AVMUTE */
 
 	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
 	if (err < 0) {
@@ -523,22 +413,45 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 	}
 
 	r600_hdmi_update_avi_infoframe(encoder, buffer, sizeof(buffer));
+
+	/* fglrx duplicates INFOFRAME_CONTROL0 & INFOFRAME_CONTROL1 ops here */
+
+	WREG32_AND(HDMI0_GENERIC_PACKET_CONTROL + offset,
+		   ~(HDMI0_GENERIC0_SEND |
+		     HDMI0_GENERIC0_CONT |
+		     HDMI0_GENERIC0_UPDATE |
+		     HDMI0_GENERIC1_SEND |
+		     HDMI0_GENERIC1_CONT |
+		     HDMI0_GENERIC0_LINE_MASK |
+		     HDMI0_GENERIC1_LINE_MASK));
+
 	r600_hdmi_update_ACR(encoder, mode->clock);
 
+	WREG32_P(HDMI0_60958_0 + offset,
+		 HDMI0_60958_CS_CHANNEL_NUMBER_L(1),
+		 ~(HDMI0_60958_CS_CHANNEL_NUMBER_L_MASK |
+		   HDMI0_60958_CS_CLOCK_ACCURACY_MASK));
+
+	WREG32_P(HDMI0_60958_1 + offset,
+		 HDMI0_60958_CS_CHANNEL_NUMBER_R(2),
+		 ~HDMI0_60958_CS_CHANNEL_NUMBER_R_MASK);
+
 	/* it's unknown what these bits do excatly, but it's indeed quite useful for debugging */
 	WREG32(HDMI0_RAMP_CONTROL0 + offset, 0x00FFFFFF);
 	WREG32(HDMI0_RAMP_CONTROL1 + offset, 0x007FFFFF);
 	WREG32(HDMI0_RAMP_CONTROL2 + offset, 0x00000001);
 	WREG32(HDMI0_RAMP_CONTROL3 + offset, 0x00000001);
 
-	r600_hdmi_audio_workaround(encoder);
-
 	/* enable audio after to setting up hw */
 	r600_audio_enable(rdev, dig->afmt->pin, true);
 }
 
-/*
- * update settings with current parameters from audio engine
+/**
+ * r600_hdmi_update_audio_settings - Update audio infoframe
+ *
+ * @encoder: drm encoder
+ *
+ * Gets info about current audio stream and updates audio infoframe.
  */
 void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 {
@@ -550,7 +463,7 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 	uint8_t buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AUDIO_INFOFRAME_SIZE];
 	struct hdmi_audio_infoframe frame;
 	uint32_t offset;
-	uint32_t iec;
+	uint32_t value;
 	ssize_t err;
 
 	if (!dig->afmt || !dig->afmt->enabled)
@@ -563,60 +476,6 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 	DRM_DEBUG("0x%02X IEC60958 status bits and 0x%02X category code\n",
 		  (int)audio.status_bits, (int)audio.category_code);
 
-	iec = 0;
-	if (audio.status_bits & AUDIO_STATUS_PROFESSIONAL)
-		iec |= 1 << 0;
-	if (audio.status_bits & AUDIO_STATUS_NONAUDIO)
-		iec |= 1 << 1;
-	if (audio.status_bits & AUDIO_STATUS_COPYRIGHT)
-		iec |= 1 << 2;
-	if (audio.status_bits & AUDIO_STATUS_EMPHASIS)
-		iec |= 1 << 3;
-
-	iec |= HDMI0_60958_CS_CATEGORY_CODE(audio.category_code);
-
-	switch (audio.rate) {
-	case 32000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x3);
-		break;
-	case 44100:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x0);
-		break;
-	case 48000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x2);
-		break;
-	case 88200:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x8);
-		break;
-	case 96000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0xa);
-		break;
-	case 176400:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0xc);
-		break;
-	case 192000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0xe);
-		break;
-	}
-
-	WREG32(HDMI0_60958_0 + offset, iec);
-
-	iec = 0;
-	switch (audio.bits_per_sample) {
-	case 16:
-		iec |= HDMI0_60958_CS_WORD_LENGTH(0x2);
-		break;
-	case 20:
-		iec |= HDMI0_60958_CS_WORD_LENGTH(0x3);
-		break;
-	case 24:
-		iec |= HDMI0_60958_CS_WORD_LENGTH(0xb);
-		break;
-	}
-	if (audio.status_bits & AUDIO_STATUS_V)
-		iec |= 0x5 << 16;
-	WREG32_P(HDMI0_60958_1 + offset, iec, ~0x5000f);
-
 	err = hdmi_audio_infoframe_init(&frame);
 	if (err < 0) {
 		DRM_ERROR("failed to setup audio infoframe\n");
@@ -631,8 +490,22 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 		return;
 	}
 
+	value = RREG32(HDMI0_AUDIO_PACKET_CONTROL + offset);
+	if (value & HDMI0_AUDIO_TEST_EN)
+		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		       value & ~HDMI0_AUDIO_TEST_EN);
+
+	WREG32_OR(HDMI0_CONTROL + offset,
+		  HDMI0_ERROR_ACK);
+
+	WREG32_AND(HDMI0_INFOFRAME_CONTROL0 + offset,
+		   ~HDMI0_AUDIO_INFO_SOURCE);
+
 	r600_hdmi_update_audio_infoframe(encoder, buffer, sizeof(buffer));
-	r600_hdmi_audio_workaround(encoder);
+
+	WREG32_OR(HDMI0_INFOFRAME_CONTROL0 + offset,
+		  HDMI0_AUDIO_INFO_CONT |
+		  HDMI0_AUDIO_INFO_UPDATE);
 }
 
 /*
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 37455f65107f..f94e7a9afe75 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -1029,15 +1029,18 @@
 #define HDMI0_AUDIO_PACKET_CONTROL   0x7408
 #       define HDMI0_AUDIO_SAMPLE_SEND  (1 << 0)
 #       define HDMI0_AUDIO_DELAY_EN(x)  (((x) & 3) << 4)
+#       define HDMI0_AUDIO_DELAY_EN_MASK	(3 << 4)
 #       define HDMI0_AUDIO_SEND_MAX_PACKETS  (1 << 8)
 #       define HDMI0_AUDIO_TEST_EN         (1 << 12)
 #       define HDMI0_AUDIO_PACKETS_PER_LINE(x)  (((x) & 0x1f) << 16)
+#       define HDMI0_AUDIO_PACKETS_PER_LINE_MASK	(0x1f << 16)
 #       define HDMI0_AUDIO_CHANNEL_SWAP    (1 << 24)
 #       define HDMI0_60958_CS_UPDATE       (1 << 26)
 #       define HDMI0_AZ_FORMAT_WTRIG_MASK  (1 << 28)
 #       define HDMI0_AZ_FORMAT_WTRIG_ACK   (1 << 29)
 #define HDMI0_AUDIO_CRC_CONTROL      0x740c
 #       define HDMI0_AUDIO_CRC_EN    (1 << 0)
+#define DCE3_HDMI0_ACR_PACKET_CONTROL	0x740c
 #define HDMI0_VBI_PACKET_CONTROL     0x7410
 #       define HDMI0_NULL_SEND       (1 << 0)
 #       define HDMI0_GC_SEND         (1 << 4)
@@ -1054,7 +1057,9 @@
 #       define HDMI0_MPEG_INFO_UPDATE  (1 << 10)
 #define HDMI0_INFOFRAME_CONTROL1     0x7418
 #       define HDMI0_AVI_INFO_LINE(x)  (((x) & 0x3f) << 0)
+#       define HDMI0_AVI_INFO_LINE_MASK		(0x3f << 0)
 #       define HDMI0_AUDIO_INFO_LINE(x)  (((x) & 0x3f) << 8)
+#       define HDMI0_AUDIO_INFO_LINE_MASK	(0x3f << 8)
 #       define HDMI0_MPEG_INFO_LINE(x)  (((x) & 0x3f) << 16)
 #define HDMI0_GENERIC_PACKET_CONTROL 0x741c
 #       define HDMI0_GENERIC0_SEND   (1 << 0)
@@ -1063,7 +1068,9 @@
 #       define HDMI0_GENERIC1_SEND   (1 << 4)
 #       define HDMI0_GENERIC1_CONT   (1 << 5)
 #       define HDMI0_GENERIC0_LINE(x)  (((x) & 0x3f) << 16)
+#       define HDMI0_GENERIC0_LINE_MASK		(0x3f << 16)
 #       define HDMI0_GENERIC1_LINE(x)  (((x) & 0x3f) << 24)
+#       define HDMI0_GENERIC1_LINE_MASK		(0x3f << 24)
 #define HDMI0_GC                     0x7428
 #       define HDMI0_GC_AVMUTE       (1 << 0)
 #define HDMI0_AVI_INFO0              0x7454
@@ -1119,16 +1126,22 @@
 #define HDMI0_GENERIC1_6             0x74a8
 #define HDMI0_ACR_32_0               0x74ac
 #       define HDMI0_ACR_CTS_32(x)   (((x) & 0xfffff) << 12)
+#       define HDMI0_ACR_CTS_32_MASK		(0xfffff << 12)
 #define HDMI0_ACR_32_1               0x74b0
 #       define HDMI0_ACR_N_32(x)   (((x) & 0xfffff) << 0)
+#       define HDMI0_ACR_N_32_MASK		(0xfffff << 0)
 #define HDMI0_ACR_44_0               0x74b4
 #       define HDMI0_ACR_CTS_44(x)   (((x) & 0xfffff) << 12)
+#       define HDMI0_ACR_CTS_44_MASK		(0xfffff << 12)
 #define HDMI0_ACR_44_1               0x74b8
 #       define HDMI0_ACR_N_44(x)   (((x) & 0xfffff) << 0)
+#       define HDMI0_ACR_N_44_MASK		(0xfffff << 0)
 #define HDMI0_ACR_48_0               0x74bc
 #       define HDMI0_ACR_CTS_48(x)   (((x) & 0xfffff) << 12)
+#       define HDMI0_ACR_CTS_48_MASK		(0xfffff << 12)
 #define HDMI0_ACR_48_1               0x74c0
 #       define HDMI0_ACR_N_48(x)   (((x) & 0xfffff) << 0)
+#       define HDMI0_ACR_N_48_MASK		(0xfffff << 0)
 #define HDMI0_ACR_STATUS_0           0x74c4
 #define HDMI0_ACR_STATUS_1           0x74c8
 #define HDMI0_AUDIO_INFO0            0x74cc
@@ -1148,14 +1161,17 @@
 #       define HDMI0_60958_CS_CATEGORY_CODE(x)      (((x) & 0xff) << 8)
 #       define HDMI0_60958_CS_SOURCE_NUMBER(x)      (((x) & 0xf) << 16)
 #       define HDMI0_60958_CS_CHANNEL_NUMBER_L(x)   (((x) & 0xf) << 20)
+#       define HDMI0_60958_CS_CHANNEL_NUMBER_L_MASK	(0xf << 20)
 #       define HDMI0_60958_CS_SAMPLING_FREQUENCY(x) (((x) & 0xf) << 24)
 #       define HDMI0_60958_CS_CLOCK_ACCURACY(x)     (((x) & 3) << 28)
+#       define HDMI0_60958_CS_CLOCK_ACCURACY_MASK	(3 << 28)
 #define HDMI0_60958_1                0x74d8
 #       define HDMI0_60958_CS_WORD_LENGTH(x)        (((x) & 0xf) << 0)
 #       define HDMI0_60958_CS_ORIGINAL_SAMPLING_FREQUENCY(x)   (((x) & 0xf) << 4)
 #       define HDMI0_60958_CS_VALID_L(x)   (((x) & 1) << 16)
 #       define HDMI0_60958_CS_VALID_R(x)   (((x) & 1) << 18)
 #       define HDMI0_60958_CS_CHANNEL_NUMBER_R(x)   (((x) & 0xf) << 20)
+#       define HDMI0_60958_CS_CHANNEL_NUMBER_R_MASK	(0xf << 20)
 #define HDMI0_ACR_PACKET_CONTROL     0x74dc
 #       define HDMI0_ACR_SEND        (1 << 0)
 #       define HDMI0_ACR_CONT        (1 << 1)
@@ -1166,6 +1182,7 @@
 #       define HDMI0_ACR_48          3
 #       define HDMI0_ACR_SOURCE      (1 << 8) /* 0 - hw; 1 - cts value */
 #       define HDMI0_ACR_AUTO_SEND   (1 << 12)
+#define DCE3_HDMI0_AUDIO_CRC_CONTROL	0x74dc
 #define HDMI0_RAMP_CONTROL0          0x74e0
 #       define HDMI0_RAMP_MAX_COUNT(x)   (((x) & 0xffffff) << 0)
 #define HDMI0_RAMP_CONTROL1          0x74e4
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 8149e7cf4303..29d9cc04c04e 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -100,6 +100,9 @@ extern int radeon_dpm;
 extern int radeon_aspm;
 extern int radeon_runtime_pm;
 extern int radeon_hard_reset;
+extern int radeon_vm_size;
+extern int radeon_vm_block_size;
+extern int radeon_deep_color;
 
 /*
  * Copy from radeon_drv.h so we don't have to include both and have conflicting
@@ -676,14 +679,16 @@ void radeon_doorbell_free(struct radeon_device *rdev, u32 doorbell);
  * IRQS.
  */
 
-struct radeon_unpin_work {
-	struct work_struct work;
-	struct radeon_device *rdev;
-	int crtc_id;
-	struct radeon_fence *fence;
+struct radeon_flip_work {
+	struct work_struct		flip_work;
+	struct work_struct		unpin_work;
+	struct radeon_device		*rdev;
+	int				crtc_id;
+	struct drm_framebuffer		*fb;
 	struct drm_pending_vblank_event *event;
-	struct radeon_bo *old_rbo;
-	u64 new_crtc_base;
+	struct radeon_bo		*old_rbo;
+	struct radeon_bo		*new_rbo;
+	struct radeon_fence		*fence;
 };
 
 struct r500_irq_stat_regs {
@@ -745,10 +750,6 @@ union radeon_irq_stat_regs {
 	struct cik_irq_stat_regs cik;
 };
 
-#define RADEON_MAX_HPD_PINS 7
-#define RADEON_MAX_CRTCS 6
-#define RADEON_MAX_AFMT_BLOCKS 7
-
 struct radeon_irq {
 	bool				installed;
 	spinlock_t			lock;
@@ -835,13 +836,8 @@ struct radeon_mec {
 /* maximum number of VMIDs */
 #define RADEON_NUM_VM	16
 
-/* defines number of bits in page table versus page directory,
- * a page is 4KB so we have 12 bits offset, 9 bits in the page
- * table and the remaining 19 bits are in the page directory */
-#define RADEON_VM_BLOCK_SIZE   9
-
 /* number of entries in page table */
-#define RADEON_VM_PTE_COUNT (1 << RADEON_VM_BLOCK_SIZE)
+#define RADEON_VM_PTE_COUNT (1 << radeon_vm_block_size)
 
 /* PTBs (Page Table Blocks) need to be aligned to 32K */
 #define RADEON_VM_PTB_ALIGN_SIZE   32768
@@ -854,6 +850,15 @@ struct radeon_mec {
 #define R600_PTE_READABLE	(1 << 5)
 #define R600_PTE_WRITEABLE	(1 << 6)
 
+/* PTE (Page Table Entry) fragment field for different page sizes */
+#define R600_PTE_FRAG_4KB	(0 << 7)
+#define R600_PTE_FRAG_64KB	(4 << 7)
+#define R600_PTE_FRAG_256KB	(6 << 7)
+
+/* flags used for GART page table entries on R600+ */
+#define R600_PTE_GART	( R600_PTE_VALID | R600_PTE_SYSTEM | R600_PTE_SNOOPED \
+			| R600_PTE_READABLE | R600_PTE_WRITEABLE)
+
 struct radeon_vm_pt {
 	struct radeon_bo		*bo;
 	uint64_t			addr;
@@ -986,8 +991,8 @@ struct radeon_cs_reloc {
 	struct radeon_bo		*robj;
 	struct ttm_validate_buffer	tv;
 	uint64_t			gpu_offset;
-	unsigned			domain;
-	unsigned			alt_domain;
+	unsigned			prefered_domains;
+	unsigned			allowed_domains;
 	uint32_t			tiling_flags;
 	uint32_t			handle;
 };
@@ -1771,7 +1776,8 @@ struct radeon_asic {
 	/* gart */
 	struct {
 		void (*tlb_flush)(struct radeon_device *rdev);
-		int (*set_page)(struct radeon_device *rdev, int i, uint64_t addr);
+		void (*set_page)(struct radeon_device *rdev, unsigned i,
+				 uint64_t addr);
 	} gart;
 	struct {
 		int (*init)(struct radeon_device *rdev);
@@ -1883,9 +1889,8 @@ struct radeon_asic {
 	} dpm;
 	/* pageflipping */
 	struct {
-		void (*pre_page_flip)(struct radeon_device *rdev, int crtc);
-		u32 (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base);
-		void (*post_page_flip)(struct radeon_device *rdev, int crtc);
+		void (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base);
+		bool (*page_flip_pending)(struct radeon_device *rdev, int crtc);
 	} pflip;
 };
 
@@ -1924,6 +1929,7 @@ struct r600_asic {
 	unsigned		tiling_group_size;
 	unsigned		tile_config;
 	unsigned		backend_map;
+	unsigned		active_simds;
 };
 
 struct rv770_asic {
@@ -1949,6 +1955,7 @@ struct rv770_asic {
 	unsigned		tiling_group_size;
 	unsigned		tile_config;
 	unsigned		backend_map;
+	unsigned		active_simds;
 };
 
 struct evergreen_asic {
@@ -1975,6 +1982,7 @@ struct evergreen_asic {
 	unsigned tiling_group_size;
 	unsigned tile_config;
 	unsigned backend_map;
+	unsigned active_simds;
 };
 
 struct cayman_asic {
@@ -2013,6 +2021,7 @@ struct cayman_asic {
 	unsigned multi_gpu_tile_size;
 
 	unsigned tile_config;
+	unsigned active_simds;
 };
 
 struct si_asic {
@@ -2043,6 +2052,7 @@ struct si_asic {
 
 	unsigned tile_config;
 	uint32_t tile_mode_array[32];
+	uint32_t active_cus;
 };
 
 struct cik_asic {
@@ -2074,6 +2084,7 @@ struct cik_asic {
 	unsigned tile_config;
 	uint32_t tile_mode_array[32];
 	uint32_t macrotile_mode_array[16];
+	uint32_t active_cus;
 };
 
 union radeon_asic_config {
@@ -2745,9 +2756,8 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
 #define radeon_pm_finish(rdev) (rdev)->asic->pm.finish((rdev))
 #define radeon_pm_init_profile(rdev) (rdev)->asic->pm.init_profile((rdev))
 #define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm.get_dynpm_state((rdev))
-#define radeon_pre_page_flip(rdev, crtc) (rdev)->asic->pflip.pre_page_flip((rdev), (crtc))
 #define radeon_page_flip(rdev, crtc, base) (rdev)->asic->pflip.page_flip((rdev), (crtc), (base))
-#define radeon_post_page_flip(rdev, crtc) (rdev)->asic->pflip.post_page_flip((rdev), (crtc))
+#define radeon_page_flip_pending(rdev, crtc) (rdev)->asic->pflip.page_flip_pending((rdev), (crtc))
 #define radeon_wait_for_vblank(rdev, crtc) (rdev)->asic->display.wait_for_vblank((rdev), (crtc))
 #define radeon_mc_wait_for_idle(rdev) (rdev)->asic->mc_wait_for_idle((rdev))
 #define radeon_get_xclk(rdev) (rdev)->asic->get_xclk((rdev))
diff --git a/drivers/gpu/drm/radeon/radeon_agp.c b/drivers/gpu/drm/radeon/radeon_agp.c
index 42433344cb1b..a9297b2c3524 100644
--- a/drivers/gpu/drm/radeon/radeon_agp.c
+++ b/drivers/gpu/drm/radeon/radeon_agp.c
@@ -117,9 +117,6 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = {
 	/* ATI Host Bridge / RV280 [M9+] Needs AGPMode 1 (phoronix forum) */
 	{ PCI_VENDOR_ID_ATI, 0xcbb2, PCI_VENDOR_ID_ATI, 0x5c61,
 		PCI_VENDOR_ID_SONY, 0x8175, 1},
-	/* HP Host Bridge / R300 [FireGL X1] Needs AGPMode 2 (fdo #7770) */
-	{ PCI_VENDOR_ID_HP, 0x122e, PCI_VENDOR_ID_ATI, 0x4e47,
-		PCI_VENDOR_ID_ATI, 0x0152, 2},
 	{ 0, 0, 0, 0, 0, 0, 0 },
 };
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e5f0177bea1e..34b9aa9e3c06 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -248,9 +248,8 @@ static struct radeon_asic r100_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -315,9 +314,8 @@ static struct radeon_asic r200_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -396,9 +394,8 @@ static struct radeon_asic r300_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -463,9 +460,8 @@ static struct radeon_asic r300_asic_pcie = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -530,9 +526,8 @@ static struct radeon_asic r420_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -597,9 +592,8 @@ static struct radeon_asic rs400_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -666,9 +660,8 @@ static struct radeon_asic rs600_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -735,9 +728,8 @@ static struct radeon_asic rs690_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -802,9 +794,8 @@ static struct radeon_asic rv515_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -869,9 +860,8 @@ static struct radeon_asic r520_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -968,9 +958,8 @@ static struct radeon_asic r600_asic = {
 		.get_temperature = &rv6xx_get_temp,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -1059,9 +1048,8 @@ static struct radeon_asic rv6xx_asic = {
 		.force_performance_level = &rv6xx_dpm_force_performance_level,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -1150,9 +1138,8 @@ static struct radeon_asic rs780_asic = {
 		.force_performance_level = &rs780_dpm_force_performance_level,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -1201,7 +1188,7 @@ static struct radeon_asic rv770_asic = {
 		.set_backlight_level = &atombios_set_backlight_level,
 		.get_backlight_level = &atombios_get_backlight_level,
 		.hdmi_enable = &r600_hdmi_enable,
-		.hdmi_setmode = &r600_hdmi_setmode,
+		.hdmi_setmode = &dce3_1_hdmi_setmode,
 	},
 	.copy = {
 		.blit = &r600_copy_cpdma,
@@ -1256,9 +1243,8 @@ static struct radeon_asic rv770_asic = {
 		.vblank_too_short = &rv770_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rv770_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rv770_page_flip_pending,
 	},
 };
 
@@ -1375,9 +1361,8 @@ static struct radeon_asic evergreen_asic = {
 		.vblank_too_short = &cypress_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1467,9 +1452,8 @@ static struct radeon_asic sumo_asic = {
 		.force_performance_level = &sumo_dpm_force_performance_level,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1560,9 +1544,8 @@ static struct radeon_asic btc_asic = {
 		.vblank_too_short = &btc_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1704,9 +1687,8 @@ static struct radeon_asic cayman_asic = {
 		.vblank_too_short = &ni_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1805,9 +1787,8 @@ static struct radeon_asic trinity_asic = {
 		.enable_bapm = &trinity_dpm_enable_bapm,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1936,9 +1917,8 @@ static struct radeon_asic si_asic = {
 		.vblank_too_short = &ni_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -2049,8 +2029,8 @@ static struct radeon_asic ci_asic = {
 		.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
 		.dma = &cik_copy_dma,
 		.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.copy = &cik_copy_cpdma,
-		.copy_ring_index = RADEON_RING_TYPE_GFX_INDEX,
+		.copy = &cik_copy_dma,
+		.copy_ring_index = R600_RING_TYPE_DMA_INDEX,
 	},
 	.surface = {
 		.set_reg = r600_set_surface_reg,
@@ -2099,9 +2079,8 @@ static struct radeon_asic ci_asic = {
 		.powergate_uvd = &ci_dpm_powergate_uvd,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -2204,9 +2183,8 @@ static struct radeon_asic kv_asic = {
 		.enable_bapm = &kv_dpm_enable_bapm,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 3d55a3a39e82..01e7c0ad8f01 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -67,7 +67,8 @@ bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int r100_asic_reset(struct radeon_device *rdev);
 u32 r100_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void r100_pci_gart_tlb_flush(struct radeon_device *rdev);
-int r100_pci_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+void r100_pci_gart_set_page(struct radeon_device *rdev, unsigned i,
+			    uint64_t addr);
 void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring);
 int r100_irq_set(struct radeon_device *rdev);
 int r100_irq_process(struct radeon_device *rdev);
@@ -135,9 +136,9 @@ extern void r100_pm_prepare(struct radeon_device *rdev);
 extern void r100_pm_finish(struct radeon_device *rdev);
 extern void r100_pm_init_profile(struct radeon_device *rdev);
 extern void r100_pm_get_dynpm_state(struct radeon_device *rdev);
-extern void r100_pre_page_flip(struct radeon_device *rdev, int crtc);
-extern u32 r100_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
-extern void r100_post_page_flip(struct radeon_device *rdev, int crtc);
+extern void r100_page_flip(struct radeon_device *rdev, int crtc,
+			   u64 crtc_base);
+extern bool r100_page_flip_pending(struct radeon_device *rdev, int crtc);
 extern void r100_wait_for_vblank(struct radeon_device *rdev, int crtc);
 extern int r100_mc_wait_for_idle(struct radeon_device *rdev);
 
@@ -171,7 +172,8 @@ extern void r300_fence_ring_emit(struct radeon_device *rdev,
 				struct radeon_fence *fence);
 extern int r300_cs_parse(struct radeon_cs_parser *p);
 extern void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev);
-extern int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+extern void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
+				     uint64_t addr);
 extern void rv370_set_pcie_lanes(struct radeon_device *rdev, int lanes);
 extern int rv370_get_pcie_lanes(struct radeon_device *rdev);
 extern void r300_set_reg_safe(struct radeon_device *rdev);
@@ -206,7 +208,8 @@ extern void rs400_fini(struct radeon_device *rdev);
 extern int rs400_suspend(struct radeon_device *rdev);
 extern int rs400_resume(struct radeon_device *rdev);
 void rs400_gart_tlb_flush(struct radeon_device *rdev);
-int rs400_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+void rs400_gart_set_page(struct radeon_device *rdev, unsigned i,
+			 uint64_t addr);
 uint32_t rs400_mc_rreg(struct radeon_device *rdev, uint32_t reg);
 void rs400_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 int rs400_gart_init(struct radeon_device *rdev);
@@ -229,7 +232,8 @@ int rs600_irq_process(struct radeon_device *rdev);
 void rs600_irq_disable(struct radeon_device *rdev);
 u32 rs600_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void rs600_gart_tlb_flush(struct radeon_device *rdev);
-int rs600_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+void rs600_gart_set_page(struct radeon_device *rdev, unsigned i,
+			 uint64_t addr);
 uint32_t rs600_mc_rreg(struct radeon_device *rdev, uint32_t reg);
 void rs600_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 void rs600_bandwidth_update(struct radeon_device *rdev);
@@ -241,9 +245,9 @@ void rs600_hpd_set_polarity(struct radeon_device *rdev,
 extern void rs600_pm_misc(struct radeon_device *rdev);
 extern void rs600_pm_prepare(struct radeon_device *rdev);
 extern void rs600_pm_finish(struct radeon_device *rdev);
-extern void rs600_pre_page_flip(struct radeon_device *rdev, int crtc);
-extern u32 rs600_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
-extern void rs600_post_page_flip(struct radeon_device *rdev, int crtc);
+extern void rs600_page_flip(struct radeon_device *rdev, int crtc,
+			    u64 crtc_base);
+extern bool rs600_page_flip_pending(struct radeon_device *rdev, int crtc);
 void rs600_set_safe_registers(struct radeon_device *rdev);
 extern void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc);
 extern int rs600_mc_wait_for_idle(struct radeon_device *rdev);
@@ -387,6 +391,11 @@ void r600_rlc_stop(struct radeon_device *rdev);
 int r600_audio_init(struct radeon_device *rdev);
 struct r600_audio_pin r600_audio_status(struct radeon_device *rdev);
 void r600_audio_fini(struct radeon_device *rdev);
+void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock);
+void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder, void *buffer,
+				    size_t size);
+void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock);
+void r600_hdmi_audio_workaround(struct drm_encoder *encoder);
 int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder);
 void r600_hdmi_update_audio_settings(struct drm_encoder *encoder);
 void r600_hdmi_enable(struct drm_encoder *encoder, bool enable);
@@ -447,7 +456,8 @@ void rv770_fini(struct radeon_device *rdev);
 int rv770_suspend(struct radeon_device *rdev);
 int rv770_resume(struct radeon_device *rdev);
 void rv770_pm_misc(struct radeon_device *rdev);
-u32 rv770_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+void rv770_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+bool rv770_page_flip_pending(struct radeon_device *rdev, int crtc);
 void r700_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 void r700_cp_stop(struct radeon_device *rdev);
 void r700_cp_fini(struct radeon_device *rdev);
@@ -458,6 +468,8 @@ int rv770_copy_dma(struct radeon_device *rdev,
 u32 rv770_get_xclk(struct radeon_device *rdev);
 int rv770_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
 int rv770_get_temp(struct radeon_device *rdev);
+/* hdmi */
+void dce3_1_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode);
 /* rv7xx pm */
 int rv770_dpm_init(struct radeon_device *rdev);
 int rv770_dpm_enable(struct radeon_device *rdev);
@@ -513,9 +525,9 @@ extern void sumo_pm_init_profile(struct radeon_device *rdev);
 extern void btc_pm_init_profile(struct radeon_device *rdev);
 int sumo_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
 int evergreen_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
-extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc);
-extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
-extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc);
+extern void evergreen_page_flip(struct radeon_device *rdev, int crtc,
+				u64 crtc_base);
+extern bool evergreen_page_flip_pending(struct radeon_device *rdev, int crtc);
 extern void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc);
 void evergreen_disable_interrupt_state(struct radeon_device *rdev);
 int evergreen_mc_wait_for_idle(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 30844814c25a..173f378428a9 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -1227,11 +1227,19 @@ bool radeon_atom_get_clock_info(struct drm_device *dev)
 			rdev->clock.default_dispclk =
 				le32_to_cpu(firmware_info->info_21.ulDefaultDispEngineClkFreq);
 			if (rdev->clock.default_dispclk == 0) {
-				if (ASIC_IS_DCE5(rdev))
+				if (ASIC_IS_DCE6(rdev))
+					rdev->clock.default_dispclk = 60000; /* 600 Mhz */
+				else if (ASIC_IS_DCE5(rdev))
 					rdev->clock.default_dispclk = 54000; /* 540 Mhz */
 				else
 					rdev->clock.default_dispclk = 60000; /* 600 Mhz */
 			}
+			/* set a reasonable default for DP */
+			if (ASIC_IS_DCE6(rdev) && (rdev->clock.default_dispclk < 53900)) {
+				DRM_INFO("Changing default dispclk from %dMhz to 600Mhz\n",
+					 rdev->clock.default_dispclk / 100);
+				rdev->clock.default_dispclk = 60000;
+			}
 			rdev->clock.dp_extclk =
 				le16_to_cpu(firmware_info->info_21.usUniphyDPModeExtClkFreq);
 			rdev->clock.current_dispclk = rdev->clock.default_dispclk;
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index 9ab30976287d..6a03624fadaa 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -626,7 +626,7 @@ static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
 	    vhdr->DeviceID != rdev->pdev->device) {
 		DRM_INFO("ACPI VFCT table is not for this card\n");
 		goto out_unmap;
-	};
+	}
 
 	if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + vhdr->ImageLength > tbl_size) {
 		DRM_ERROR("ACPI VFCT image truncated\n");
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index ea50e0ae7bf7..44831197e82e 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -48,6 +48,7 @@ void radeon_connector_hotplug(struct drm_connector *connector)
 	radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
 
 	/* if the connector is already off, don't turn it back on */
+	/* FIXME: This access isn't protected by any locks. */
 	if (connector->dpms != DRM_MODE_DPMS_ON)
 		return;
 
@@ -100,6 +101,7 @@ int radeon_get_monitor_bpc(struct drm_connector *connector)
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	struct radeon_connector_atom_dig *dig_connector;
 	int bpc = 8;
+	int mode_clock, max_tmds_clock;
 
 	switch (connector->connector_type) {
 	case DRM_MODE_CONNECTOR_DVII:
@@ -145,6 +147,64 @@ int radeon_get_monitor_bpc(struct drm_connector *connector)
 		}
 		break;
 	}
+
+	if (drm_detect_hdmi_monitor(radeon_connector->edid)) {
+		/* hdmi deep color only implemented on DCE4+ */
+		if ((bpc > 8) && !ASIC_IS_DCE4(rdev)) {
+			DRM_DEBUG("%s: HDMI deep color %d bpc unsupported. Using 8 bpc.\n",
+					  connector->name, bpc);
+			bpc = 8;
+		}
+
+		/*
+		 * Pre DCE-8 hw can't handle > 12 bpc, and more than 12 bpc doesn't make
+		 * much sense without support for > 12 bpc framebuffers. RGB 4:4:4 at
+		 * 12 bpc is always supported on hdmi deep color sinks, as this is
+		 * required by the HDMI-1.3 spec. Clamp to a safe 12 bpc maximum.
+		 */
+		if (bpc > 12) {
+			DRM_DEBUG("%s: HDMI deep color %d bpc unsupported. Using 12 bpc.\n",
+					  connector->name, bpc);
+			bpc = 12;
+		}
+
+		/* Any defined maximum tmds clock limit we must not exceed? */
+		if (connector->max_tmds_clock > 0) {
+			/* mode_clock is clock in kHz for mode to be modeset on this connector */
+			mode_clock = radeon_connector->pixelclock_for_modeset;
+
+			/* Maximum allowable input clock in kHz */
+			max_tmds_clock = connector->max_tmds_clock * 1000;
+
+			DRM_DEBUG("%s: hdmi mode dotclock %d kHz, max tmds input clock %d kHz.\n",
+					  connector->name, mode_clock, max_tmds_clock);
+
+			/* Check if bpc is within clock limit. Try to degrade gracefully otherwise */
+			if ((bpc == 12) && (mode_clock * 3/2 > max_tmds_clock)) {
+				if ((connector->display_info.edid_hdmi_dc_modes & DRM_EDID_HDMI_DC_30) &&
+					(mode_clock * 5/4 <= max_tmds_clock))
+					bpc = 10;
+				else
+					bpc = 8;
+
+				DRM_DEBUG("%s: HDMI deep color 12 bpc exceeds max tmds clock. Using %d bpc.\n",
+						  connector->name, bpc);
+			}
+
+			if ((bpc == 10) && (mode_clock * 5/4 > max_tmds_clock)) {
+				bpc = 8;
+				DRM_DEBUG("%s: HDMI deep color 10 bpc exceeds max tmds clock. Using %d bpc.\n",
+						  connector->name, bpc);
+			}
+		}
+	}
+
+	if ((radeon_deep_color == 0) && (bpc > 8))
+		bpc = 8;
+
+	DRM_DEBUG("%s: Display bpc=%d, returned bpc=%d\n",
+			  connector->name, connector->display_info.bpc, bpc);
+
 	return bpc;
 }
 
@@ -260,13 +320,17 @@ radeon_connector_analog_encoder_conflict_solve(struct drm_connector *connector,
 					continue;
 
 				if (priority == true) {
-					DRM_DEBUG_KMS("1: conflicting encoders switching off %s\n", drm_get_connector_name(conflict));
-					DRM_DEBUG_KMS("in favor of %s\n", drm_get_connector_name(connector));
+					DRM_DEBUG_KMS("1: conflicting encoders switching off %s\n",
+						      conflict->name);
+					DRM_DEBUG_KMS("in favor of %s\n",
+						      connector->name);
 					conflict->status = connector_status_disconnected;
 					radeon_connector_update_scratch_regs(conflict, connector_status_disconnected);
 				} else {
-					DRM_DEBUG_KMS("2: conflicting encoders switching off %s\n", drm_get_connector_name(connector));
-					DRM_DEBUG_KMS("in favor of %s\n", drm_get_connector_name(conflict));
+					DRM_DEBUG_KMS("2: conflicting encoders switching off %s\n",
+						      connector->name);
+					DRM_DEBUG_KMS("in favor of %s\n",
+						      conflict->name);
 					current_status = connector_status_disconnected;
 				}
 				break;
@@ -787,7 +851,7 @@ radeon_vga_detect(struct drm_connector *connector, bool force)
 
 		if (!radeon_connector->edid) {
 			DRM_ERROR("%s: probed a monitor but no|invalid EDID\n",
-					drm_get_connector_name(connector));
+					connector->name);
 			ret = connector_status_connected;
 		} else {
 			radeon_connector->use_digital = !!(radeon_connector->edid->input & DRM_EDID_INPUT_DIGITAL);
@@ -1010,12 +1074,13 @@ radeon_dvi_detect(struct drm_connector *connector, bool force)
 
 		if (!radeon_connector->edid) {
 			DRM_ERROR("%s: probed a monitor but no|invalid EDID\n",
-					drm_get_connector_name(connector));
+					connector->name);
 			/* rs690 seems to have a problem with connectors not existing and always
 			 * return a block of 0's. If we see this just stop polling on this output */
 			if ((rdev->family == CHIP_RS690 || rdev->family == CHIP_RS740) && radeon_connector->base.null_edid_counter) {
 				ret = connector_status_disconnected;
-				DRM_ERROR("%s: detected RS690 floating bus bug, stopping ddc detect\n", drm_get_connector_name(connector));
+				DRM_ERROR("%s: detected RS690 floating bus bug, stopping ddc detect\n",
+					  connector->name);
 				radeon_connector->ddc_bus = NULL;
 			} else {
 				ret = connector_status_connected;
@@ -1226,17 +1291,15 @@ static int radeon_dvi_mode_valid(struct drm_connector *connector,
 		    (radeon_connector->connector_object_id == CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D) ||
 		    (radeon_connector->connector_object_id == CONNECTOR_OBJECT_ID_HDMI_TYPE_B))
 			return MODE_OK;
-		else if (radeon_connector->connector_object_id == CONNECTOR_OBJECT_ID_HDMI_TYPE_A) {
-			if (ASIC_IS_DCE6(rdev)) {
-				/* HDMI 1.3+ supports max clock of 340 Mhz */
-				if (mode->clock > 340000)
-					return MODE_CLOCK_HIGH;
-				else
-					return MODE_OK;
-			} else
+		else if (ASIC_IS_DCE6(rdev) && drm_detect_hdmi_monitor(radeon_connector->edid)) {
+			/* HDMI 1.3+ supports max clock of 340 Mhz */
+			if (mode->clock > 340000)
 				return MODE_CLOCK_HIGH;
-		} else
+			else
+				return MODE_OK;
+		} else {
 			return MODE_CLOCK_HIGH;
+		}
 	}
 
 	/* check against the max pixel clock */
@@ -1387,7 +1450,7 @@ bool radeon_connector_is_dp12_capable(struct drm_connector *connector)
 	struct radeon_device *rdev = dev->dev_private;
 
 	if (ASIC_IS_DCE5(rdev) &&
-	    (rdev->clock.dp_extclk >= 53900) &&
+	    (rdev->clock.default_dispclk >= 53900) &&
 	    radeon_connector_encoder_is_hbr2(connector)) {
 		return true;
 	}
@@ -1487,6 +1550,8 @@ out:
 static int radeon_dp_mode_valid(struct drm_connector *connector,
 				  struct drm_display_mode *mode)
 {
+	struct drm_device *dev = connector->dev;
+	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	struct radeon_connector_atom_dig *radeon_dig_connector = radeon_connector->con_priv;
 
@@ -1517,14 +1582,23 @@ static int radeon_dp_mode_valid(struct drm_connector *connector,
 					return MODE_PANEL;
 			}
 		}
-		return MODE_OK;
 	} else {
 		if ((radeon_dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) ||
-		    (radeon_dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP))
+		    (radeon_dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP)) {
 			return radeon_dp_mode_valid_helper(connector, mode);
-		else
-			return MODE_OK;
+		} else {
+			if (ASIC_IS_DCE6(rdev) && drm_detect_hdmi_monitor(radeon_connector->edid)) {
+				/* HDMI 1.3+ supports max clock of 340 Mhz */
+				if (mode->clock > 340000)
+					return MODE_CLOCK_HIGH;
+			} else {
+				if (mode->clock > 165000)
+					return MODE_CLOCK_HIGH;
+			}
+		}
 	}
+
+	return MODE_OK;
 }
 
 static const struct drm_connector_helper_funcs radeon_dp_connector_helper_funcs = {
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 41ecf8a60611..71a143461478 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -140,10 +140,10 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 		if (p->ring == R600_RING_TYPE_UVD_INDEX &&
 		    (i == 0 || drm_pci_device_is_agp(p->rdev->ddev))) {
 			/* TODO: is this still needed for NI+ ? */
-			p->relocs[i].domain =
+			p->relocs[i].prefered_domains =
 				RADEON_GEM_DOMAIN_VRAM;
 
-			p->relocs[i].alt_domain =
+			p->relocs[i].allowed_domains =
 				RADEON_GEM_DOMAIN_VRAM;
 
 			/* prioritize this over any other relocation */
@@ -158,10 +158,10 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 				return -EINVAL;
 			}
 
-			p->relocs[i].domain = domain;
+			p->relocs[i].prefered_domains = domain;
 			if (domain == RADEON_GEM_DOMAIN_VRAM)
 				domain |= RADEON_GEM_DOMAIN_GTT;
-			p->relocs[i].alt_domain = domain;
+			p->relocs[i].allowed_domains = domain;
 		}
 
 		p->relocs[i].tv.bo = &p->relocs[i].robj->tbo;
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 2cd144c378d6..03686fab842d 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1052,6 +1052,43 @@ static void radeon_check_arguments(struct radeon_device *rdev)
 		radeon_agpmode = 0;
 		break;
 	}
+
+	if (!radeon_check_pot_argument(radeon_vm_size)) {
+		dev_warn(rdev->dev, "VM size (%d) must be a power of 2\n",
+			 radeon_vm_size);
+		radeon_vm_size = 4096;
+	}
+
+	if (radeon_vm_size < 4) {
+		dev_warn(rdev->dev, "VM size (%d) to small, min is 4MB\n",
+			 radeon_vm_size);
+		radeon_vm_size = 4096;
+	}
+
+       /*
+        * Max GPUVM size for Cayman, SI and CI are 40 bits.
+        */
+	if (radeon_vm_size > 1024*1024) {
+		dev_warn(rdev->dev, "VM size (%d) to large, max is 1TB\n",
+			 radeon_vm_size);
+		radeon_vm_size = 4096;
+	}
+
+	/* defines number of bits in page table versus page directory,
+	 * a page is 4KB so we have 12 bits offset, minimum 9 bits in the
+	 * page table and the remaining bits are in the page directory */
+	if (radeon_vm_block_size < 9) {
+		dev_warn(rdev->dev, "VM page table size (%d) to small\n",
+			 radeon_vm_block_size);
+		radeon_vm_block_size = 9;
+	}
+
+	if (radeon_vm_block_size > 24 ||
+	    radeon_vm_size < (1ull << radeon_vm_block_size)) {
+		dev_warn(rdev->dev, "VM page table size (%d) to large\n",
+			 radeon_vm_block_size);
+		radeon_vm_block_size = 9;
+	}
 }
 
 /**
@@ -1126,12 +1163,13 @@ static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
 static bool radeon_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	bool can_switch;
 
-	spin_lock(&dev->count_lock);
-	can_switch = (dev->open_count == 0);
-	spin_unlock(&dev->count_lock);
-	return can_switch;
+	/*
+	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
+	 * locking inversion with the driver load path. And the access here is
+	 * completely racy anyway. So don't bother with locking for now.
+	 */
+	return dev->open_count == 0;
 }
 
 static const struct vga_switcheroo_client_ops radeon_switcheroo_ops = {
@@ -1196,17 +1234,16 @@ int radeon_device_init(struct radeon_device *rdev,
 	if (r)
 		return r;
 
+	radeon_check_arguments(rdev);
 	/* Adjust VM size here.
-	 * Currently set to 4GB ((1 << 20) 4k pages).
-	 * Max GPUVM size for cayman and SI is 40 bits.
+	 * Max GPUVM size for cayman+ is 40 bits.
 	 */
-	rdev->vm_manager.max_pfn = 1 << 20;
+	rdev->vm_manager.max_pfn = radeon_vm_size << 8;
 
 	/* Set asic functions */
 	r = radeon_asic_init(rdev);
 	if (r)
 		return r;
-	radeon_check_arguments(rdev);
 
 	/* all of the newer IGP chips have an internal gart
 	 * However some rs4xx report as AGP, so remove that here.
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 356b733caafe..13896edcf0b6 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -66,7 +66,8 @@ static void avivo_crtc_load_lut(struct drm_crtc *crtc)
 			     (radeon_crtc->lut_b[i] << 0));
 	}
 
-	WREG32(AVIVO_D1GRPH_LUT_SEL + radeon_crtc->crtc_offset, radeon_crtc->crtc_id);
+	/* Only change bit 0 of LUT_SEL, other bits are set elsewhere */
+	WREG32_P(AVIVO_D1GRPH_LUT_SEL + radeon_crtc->crtc_offset, radeon_crtc->crtc_id, ~1);
 }
 
 static void dce4_crtc_load_lut(struct drm_crtc *crtc)
@@ -249,16 +250,21 @@ static void radeon_crtc_destroy(struct drm_crtc *crtc)
 	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
 
 	drm_crtc_cleanup(crtc);
+	destroy_workqueue(radeon_crtc->flip_queue);
 	kfree(radeon_crtc);
 }
 
-/*
- * Handle unpin events outside the interrupt handler proper.
+/**
+ * radeon_unpin_work_func - unpin old buffer object
+ *
+ * @__work - kernel work item
+ *
+ * Unpin the old frame buffer object outside of the interrupt handler
  */
 static void radeon_unpin_work_func(struct work_struct *__work)
 {
-	struct radeon_unpin_work *work =
-		container_of(__work, struct radeon_unpin_work, work);
+	struct radeon_flip_work *work =
+		container_of(__work, struct radeon_flip_work, unpin_work);
 	int r;
 
 	/* unpin of the old buffer */
@@ -276,10 +282,9 @@ static void radeon_unpin_work_func(struct work_struct *__work)
 	kfree(work);
 }
 
-void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
+void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
-	struct radeon_unpin_work *work;
 	unsigned long flags;
 	u32 update_pending;
 	int vpos, hpos;
@@ -289,24 +294,16 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 		return;
 
 	spin_lock_irqsave(&rdev->ddev->event_lock, flags);
-	work = radeon_crtc->unpin_work;
-	if (work == NULL ||
-	    (work->fence && !radeon_fence_signaled(work->fence))) {
+	if (radeon_crtc->flip_status != RADEON_FLIP_SUBMITTED) {
+		DRM_DEBUG_DRIVER("radeon_crtc->flip_status = %d != "
+				 "RADEON_FLIP_SUBMITTED(%d)\n",
+				 radeon_crtc->flip_status,
+				 RADEON_FLIP_SUBMITTED);
 		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
 		return;
 	}
-	/* New pageflip, or just completion of a previous one? */
-	if (!radeon_crtc->deferred_flip_completion) {
-		/* do the flip (mmio) */
-		update_pending = radeon_page_flip(rdev, crtc_id, work->new_crtc_base);
-	} else {
-		/* This is just a completion of a flip queued in crtc
-		 * at last invocation. Make sure we go directly to
-		 * completion routine.
-		 */
-		update_pending = 0;
-		radeon_crtc->deferred_flip_completion = 0;
-	}
+
+	update_pending = radeon_page_flip_pending(rdev, crtc_id);
 
 	/* Has the pageflip already completed in crtc, or is it certain
 	 * to complete in this vblank?
@@ -324,19 +321,43 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 		 */
 		update_pending = 0;
 	}
-	if (update_pending) {
-		/* crtc didn't flip in this target vblank interval,
-		 * but flip is pending in crtc. It will complete it
-		 * in next vblank interval, so complete the flip at
-		 * next vblank irq.
-		 */
-		radeon_crtc->deferred_flip_completion = 1;
+	spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
+	if (!update_pending)
+		radeon_crtc_handle_flip(rdev, crtc_id);
+}
+
+/**
+ * radeon_crtc_handle_flip - page flip completed
+ *
+ * @rdev: radeon device pointer
+ * @crtc_id: crtc number this event is for
+ *
+ * Called when we are sure that a page flip for this crtc is completed.
+ */
+void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	struct radeon_flip_work *work;
+	unsigned long flags;
+
+	/* this can happen at init */
+	if (radeon_crtc == NULL)
+		return;
+
+	spin_lock_irqsave(&rdev->ddev->event_lock, flags);
+	work = radeon_crtc->flip_work;
+	if (radeon_crtc->flip_status != RADEON_FLIP_SUBMITTED) {
+		DRM_DEBUG_DRIVER("radeon_crtc->flip_status = %d != "
+				 "RADEON_FLIP_SUBMITTED(%d)\n",
+				 radeon_crtc->flip_status,
+				 RADEON_FLIP_SUBMITTED);
 		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
 		return;
 	}
 
-	/* Pageflip (will be) certainly completed in this vblank. Clean up. */
-	radeon_crtc->unpin_work = NULL;
+	/* Pageflip completed. Clean up. */
+	radeon_crtc->flip_status = RADEON_FLIP_NONE;
+	radeon_crtc->flip_work = NULL;
 
 	/* wakeup userspace */
 	if (work->event)
@@ -346,84 +367,70 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 
 	drm_vblank_put(rdev->ddev, radeon_crtc->crtc_id);
 	radeon_fence_unref(&work->fence);
-	radeon_post_page_flip(work->rdev, work->crtc_id);
-	schedule_work(&work->work);
+	radeon_irq_kms_pflip_irq_put(rdev, work->crtc_id);
+	queue_work(radeon_crtc->flip_queue, &work->unpin_work);
 }
 
-static int radeon_crtc_page_flip(struct drm_crtc *crtc,
-				 struct drm_framebuffer *fb,
-				 struct drm_pending_vblank_event *event,
-				 uint32_t page_flip_flags)
+/**
+ * radeon_flip_work_func - page flip framebuffer
+ *
+ * @work - kernel work item
+ *
+ * Wait for the buffer object to become idle and do the actual page flip
+ */
+static void radeon_flip_work_func(struct work_struct *__work)
 {
-	struct drm_device *dev = crtc->dev;
-	struct radeon_device *rdev = dev->dev_private;
-	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
-	struct radeon_framebuffer *old_radeon_fb;
-	struct radeon_framebuffer *new_radeon_fb;
-	struct drm_gem_object *obj;
-	struct radeon_bo *rbo;
-	struct radeon_unpin_work *work;
-	unsigned long flags;
-	u32 tiling_flags, pitch_pixels;
-	u64 base;
-	int r;
+	struct radeon_flip_work *work =
+		container_of(__work, struct radeon_flip_work, flip_work);
+	struct radeon_device *rdev = work->rdev;
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[work->crtc_id];
 
-	work = kzalloc(sizeof *work, GFP_KERNEL);
-	if (work == NULL)
-		return -ENOMEM;
+	struct drm_crtc *crtc = &radeon_crtc->base;
+	struct drm_framebuffer *fb = work->fb;
 
-	work->event = event;
-	work->rdev = rdev;
-	work->crtc_id = radeon_crtc->crtc_id;
-	old_radeon_fb = to_radeon_framebuffer(crtc->primary->fb);
-	new_radeon_fb = to_radeon_framebuffer(fb);
-	/* schedule unpin of the old buffer */
-	obj = old_radeon_fb->obj;
-	/* take a reference to the old object */
-	drm_gem_object_reference(obj);
-	rbo = gem_to_radeon_bo(obj);
-	work->old_rbo = rbo;
-	obj = new_radeon_fb->obj;
-	rbo = gem_to_radeon_bo(obj);
+	uint32_t tiling_flags, pitch_pixels;
+	uint64_t base;
 
-	spin_lock(&rbo->tbo.bdev->fence_lock);
-	if (rbo->tbo.sync_obj)
-		work->fence = radeon_fence_ref(rbo->tbo.sync_obj);
-	spin_unlock(&rbo->tbo.bdev->fence_lock);
+	unsigned long flags;
+	int r;
 
-	INIT_WORK(&work->work, radeon_unpin_work_func);
+        down_read(&rdev->exclusive_lock);
+	while (work->fence) {
+		r = radeon_fence_wait(work->fence, false);
+		if (r == -EDEADLK) {
+			up_read(&rdev->exclusive_lock);
+			r = radeon_gpu_reset(rdev);
+			down_read(&rdev->exclusive_lock);
+		}
 
-	/* We borrow the event spin lock for protecting unpin_work */
-	spin_lock_irqsave(&dev->event_lock, flags);
-	if (radeon_crtc->unpin_work) {
-		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
-		r = -EBUSY;
-		goto unlock_free;
+		if (r) {
+			DRM_ERROR("failed to wait on page flip fence (%d)!\n",
+				  r);
+			goto cleanup;
+		} else
+			radeon_fence_unref(&work->fence);
 	}
-	radeon_crtc->unpin_work = work;
-	radeon_crtc->deferred_flip_completion = 0;
-	spin_unlock_irqrestore(&dev->event_lock, flags);
 
 	/* pin the new buffer */
 	DRM_DEBUG_DRIVER("flip-ioctl() cur_fbo = %p, cur_bbo = %p\n",
-			 work->old_rbo, rbo);
+			 work->old_rbo, work->new_rbo);
 
-	r = radeon_bo_reserve(rbo, false);
+	r = radeon_bo_reserve(work->new_rbo, false);
 	if (unlikely(r != 0)) {
 		DRM_ERROR("failed to reserve new rbo buffer before flip\n");
-		goto pflip_cleanup;
+		goto cleanup;
 	}
 	/* Only 27 bit offset for legacy CRTC */
-	r = radeon_bo_pin_restricted(rbo, RADEON_GEM_DOMAIN_VRAM,
+	r = radeon_bo_pin_restricted(work->new_rbo, RADEON_GEM_DOMAIN_VRAM,
 				     ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, &base);
 	if (unlikely(r != 0)) {
-		radeon_bo_unreserve(rbo);
+		radeon_bo_unreserve(work->new_rbo);
 		r = -EINVAL;
 		DRM_ERROR("failed to pin new rbo buffer before flip\n");
-		goto pflip_cleanup;
+		goto cleanup;
 	}
-	radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL);
-	radeon_bo_unreserve(rbo);
+	radeon_bo_get_tiling_flags(work->new_rbo, &tiling_flags, NULL);
+	radeon_bo_unreserve(work->new_rbo);
 
 	if (!ASIC_IS_AVIVO(rdev)) {
 		/* crtc offset is from display base addr not FB location */
@@ -461,44 +468,109 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 		base &= ~7;
 	}
 
-	spin_lock_irqsave(&dev->event_lock, flags);
-	work->new_crtc_base = base;
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-
-	/* update crtc fb */
-	crtc->primary->fb = fb;
-
-	r = drm_vblank_get(dev, radeon_crtc->crtc_id);
+	r = drm_vblank_get(crtc->dev, radeon_crtc->crtc_id);
 	if (r) {
 		DRM_ERROR("failed to get vblank before flip\n");
-		goto pflip_cleanup1;
+		goto pflip_cleanup;
 	}
 
+	/* We borrow the event spin lock for protecting flip_work */
+	spin_lock_irqsave(&crtc->dev->event_lock, flags);
+
 	/* set the proper interrupt */
-	radeon_pre_page_flip(rdev, radeon_crtc->crtc_id);
+	radeon_irq_kms_pflip_irq_get(rdev, radeon_crtc->crtc_id);
 
-	return 0;
+	/* do the flip (mmio) */
+	radeon_page_flip(rdev, radeon_crtc->crtc_id, base);
+
+	radeon_crtc->flip_status = RADEON_FLIP_SUBMITTED;
+	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+	up_read(&rdev->exclusive_lock);
+
+	return;
 
-pflip_cleanup1:
-	if (unlikely(radeon_bo_reserve(rbo, false) != 0)) {
+pflip_cleanup:
+	if (unlikely(radeon_bo_reserve(work->new_rbo, false) != 0)) {
 		DRM_ERROR("failed to reserve new rbo in error path\n");
-		goto pflip_cleanup;
+		goto cleanup;
 	}
-	if (unlikely(radeon_bo_unpin(rbo) != 0)) {
+	if (unlikely(radeon_bo_unpin(work->new_rbo) != 0)) {
 		DRM_ERROR("failed to unpin new rbo in error path\n");
 	}
-	radeon_bo_unreserve(rbo);
+	radeon_bo_unreserve(work->new_rbo);
 
-pflip_cleanup:
-	spin_lock_irqsave(&dev->event_lock, flags);
-	radeon_crtc->unpin_work = NULL;
-unlock_free:
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-	drm_gem_object_unreference_unlocked(old_radeon_fb->obj);
+cleanup:
+	drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
 	radeon_fence_unref(&work->fence);
 	kfree(work);
+	up_read(&rdev->exclusive_lock);
+}
+
+static int radeon_crtc_page_flip(struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_pending_vblank_event *event,
+				 uint32_t page_flip_flags)
+{
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+	struct radeon_framebuffer *old_radeon_fb;
+	struct radeon_framebuffer *new_radeon_fb;
+	struct drm_gem_object *obj;
+	struct radeon_flip_work *work;
+	unsigned long flags;
 
-	return r;
+	work = kzalloc(sizeof *work, GFP_KERNEL);
+	if (work == NULL)
+		return -ENOMEM;
+
+	INIT_WORK(&work->flip_work, radeon_flip_work_func);
+	INIT_WORK(&work->unpin_work, radeon_unpin_work_func);
+
+	work->rdev = rdev;
+	work->crtc_id = radeon_crtc->crtc_id;
+	work->fb = fb;
+	work->event = event;
+
+	/* schedule unpin of the old buffer */
+	old_radeon_fb = to_radeon_framebuffer(crtc->primary->fb);
+	obj = old_radeon_fb->obj;
+
+	/* take a reference to the old object */
+	drm_gem_object_reference(obj);
+	work->old_rbo = gem_to_radeon_bo(obj);
+
+	new_radeon_fb = to_radeon_framebuffer(fb);
+	obj = new_radeon_fb->obj;
+	work->new_rbo = gem_to_radeon_bo(obj);
+
+	spin_lock(&work->new_rbo->tbo.bdev->fence_lock);
+	if (work->new_rbo->tbo.sync_obj)
+		work->fence = radeon_fence_ref(work->new_rbo->tbo.sync_obj);
+	spin_unlock(&work->new_rbo->tbo.bdev->fence_lock);
+
+	/* We borrow the event spin lock for protecting flip_work */
+	spin_lock_irqsave(&crtc->dev->event_lock, flags);
+
+	if (radeon_crtc->flip_status != RADEON_FLIP_NONE) {
+		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+		drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
+		radeon_fence_unref(&work->fence);
+		kfree(work);
+		return -EBUSY;
+	}
+	radeon_crtc->flip_status = RADEON_FLIP_PENDING;
+	radeon_crtc->flip_work = work;
+
+	/* update crtc fb */
+	crtc->primary->fb = fb;
+
+	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+
+	queue_work(radeon_crtc->flip_queue, &work->flip_work);
+
+	return 0;
 }
 
 static int
@@ -568,6 +640,7 @@ static void radeon_crtc_init(struct drm_device *dev, int index)
 
 	drm_mode_crtc_set_gamma_size(&radeon_crtc->base, 256);
 	radeon_crtc->crtc_id = index;
+	radeon_crtc->flip_queue = create_singlethread_workqueue("radeon-crtc");
 	rdev->mode_info.crtcs[index] = radeon_crtc;
 
 	if (rdev->family >= CHIP_BONAIRE) {
@@ -661,7 +734,7 @@ static void radeon_print_display_setup(struct drm_device *dev)
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		radeon_connector = to_radeon_connector(connector);
 		DRM_INFO("Connector %d:\n", i);
-		DRM_INFO("  %s\n", drm_get_connector_name(connector));
+		DRM_INFO("  %s\n", connector->name);
 		if (radeon_connector->hpd.hpd != RADEON_HPD_NONE)
 			DRM_INFO("  %s\n", hpd_names[radeon_connector->hpd.hpd]);
 		if (radeon_connector->ddc_bus) {
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index c00a2f585185..cb1421369e3a 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -81,9 +81,10 @@
  *   2.37.0 - allow GS ring setup on r6xx/r7xx
  *   2.38.0 - RADEON_GEM_OP (GET_INITIAL_DOMAIN, SET_INITIAL_DOMAIN),
  *            CIK: 1D and linear tiling modes contain valid PIPE_CONFIG
+ *   2.39.0 - Add INFO query for number of active CUs
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	38
+#define KMS_DRIVER_MINOR	39
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
@@ -172,6 +173,9 @@ int radeon_dpm = -1;
 int radeon_aspm = -1;
 int radeon_runtime_pm = -1;
 int radeon_hard_reset = 0;
+int radeon_vm_size = 4096;
+int radeon_vm_block_size = 9;
+int radeon_deep_color = 0;
 
 MODULE_PARM_DESC(no_wb, "Disable AGP writeback for scratch registers");
 module_param_named(no_wb, radeon_no_wb, int, 0444);
@@ -239,6 +243,15 @@ module_param_named(runpm, radeon_runtime_pm, int, 0444);
 MODULE_PARM_DESC(hard_reset, "PCI config reset (1 = force enable, 0 = disable (default))");
 module_param_named(hard_reset, radeon_hard_reset, int, 0444);
 
+MODULE_PARM_DESC(vm_size, "VM address space size in megabytes (default 4GB)");
+module_param_named(vm_size, radeon_vm_size, int, 0444);
+
+MODULE_PARM_DESC(vm_block_size, "VM page table size in bits (default 9)");
+module_param_named(vm_block_size, radeon_vm_block_size, int, 0444);
+
+MODULE_PARM_DESC(deep_color, "Deep Color support (1 = enable, 0 = disable (default))");
+module_param_named(deep_color, radeon_deep_color, int, 0444);
+
 static struct pci_device_id pciidlist[] = {
 	radeon_PCI_IDS
 };
@@ -519,7 +532,6 @@ static struct drm_driver kms_driver = {
 	    DRIVER_USE_AGP |
 	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
 	    DRIVER_PRIME | DRIVER_RENDER,
-	.dev_priv_size = 0,
 	.load = radeon_driver_load_kms,
 	.open = radeon_driver_open_kms,
 	.preclose = radeon_driver_preclose_kms,
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index a77b1c13ea43..913787085dfa 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -819,15 +819,35 @@ static int radeon_debugfs_fence_info(struct seq_file *m, void *data)
 	return 0;
 }
 
+/**
+ * radeon_debugfs_gpu_reset - manually trigger a gpu reset
+ *
+ * Manually trigger a gpu reset at the next fence wait.
+ */
+static int radeon_debugfs_gpu_reset(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct radeon_device *rdev = dev->dev_private;
+
+	down_read(&rdev->exclusive_lock);
+	seq_printf(m, "%d\n", rdev->needs_reset);
+	rdev->needs_reset = true;
+	up_read(&rdev->exclusive_lock);
+
+	return 0;
+}
+
 static struct drm_info_list radeon_debugfs_fence_list[] = {
 	{"radeon_fence_info", &radeon_debugfs_fence_info, 0, NULL},
+	{"radeon_gpu_reset", &radeon_debugfs_gpu_reset, 0, NULL}
 };
 #endif
 
 int radeon_debugfs_fence_init(struct radeon_device *rdev)
 {
 #if defined(CONFIG_DEBUG_FS)
-	return radeon_debugfs_add_files(rdev, radeon_debugfs_fence_list, 1);
+	return radeon_debugfs_add_files(rdev, radeon_debugfs_fence_list, 2);
 #else
 	return 0;
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c
index 7b944142a9fd..add622008407 100644
--- a/drivers/gpu/drm/radeon/radeon_i2c.c
+++ b/drivers/gpu/drm/radeon/radeon_i2c.c
@@ -94,6 +94,8 @@ static int pre_xfer(struct i2c_adapter *i2c_adap)
 	struct radeon_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t temp;
 
+	mutex_lock(&i2c->mutex);
+
 	/* RV410 appears to have a bug where the hw i2c in reset
 	 * holds the i2c port in a bad state - switch hw i2c away before
 	 * doing DDC - do this for all r200s/r300s/r400s for safety sake
@@ -170,6 +172,8 @@ static void post_xfer(struct i2c_adapter *i2c_adap)
 	temp = RREG32(rec->mask_data_reg) & ~rec->mask_data_mask;
 	WREG32(rec->mask_data_reg, temp);
 	temp = RREG32(rec->mask_data_reg);
+
+	mutex_unlock(&i2c->mutex);
 }
 
 static int get_clock(void *i2c_priv)
@@ -813,6 +817,8 @@ static int radeon_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
 	struct radeon_i2c_bus_rec *rec = &i2c->rec;
 	int ret = 0;
 
+	mutex_lock(&i2c->mutex);
+
 	switch (rdev->family) {
 	case CHIP_R100:
 	case CHIP_RV100:
@@ -879,6 +885,8 @@ static int radeon_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
 		break;
 	}
 
+	mutex_unlock(&i2c->mutex);
+
 	return ret;
 }
 
@@ -919,6 +927,7 @@ struct radeon_i2c_chan *radeon_i2c_create(struct drm_device *dev,
 	i2c->adapter.dev.parent = &dev->pdev->dev;
 	i2c->dev = dev;
 	i2c_set_adapdata(&i2c->adapter, i2c);
+	mutex_init(&i2c->mutex);
 	if (rec->mm_i2c ||
 	    (rec->hw_capable &&
 	     radeon_hw_i2c &&
@@ -979,7 +988,7 @@ void radeon_i2c_destroy(struct radeon_i2c_chan *i2c)
 		return;
 	i2c_del_adapter(&i2c->adapter);
 	if (i2c->has_aux)
-		drm_dp_aux_unregister_i2c_bus(&i2c->aux);
+		drm_dp_aux_unregister(&i2c->aux);
 	kfree(i2c);
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_ioc32.c b/drivers/gpu/drm/radeon/radeon_ioc32.c
index bdb0f93e73bc..0b98ea134579 100644
--- a/drivers/gpu/drm/radeon/radeon_ioc32.c
+++ b/drivers/gpu/drm/radeon/radeon_ioc32.c
@@ -399,7 +399,7 @@ long radeon_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(radeon_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(radeon_compat_ioctls))
 		fn = radeon_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 089c9ffb0aa9..16807afab362 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -287,7 +287,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
 	INIT_WORK(&rdev->reset_work, radeon_irq_reset_work_func);
 
 	rdev->irq.installed = true;
-	r = drm_irq_install(rdev->ddev);
+	r = drm_irq_install(rdev->ddev, rdev->ddev->pdev->irq);
 	if (r) {
 		rdev->irq.installed = false;
 		flush_work(&rdev->hotplug_work);
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index eaaedba04675..35d931881b4b 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -513,6 +513,22 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		value_size = sizeof(uint64_t);
 		value64 = atomic64_read(&rdev->gtt_usage);
 		break;
+	case RADEON_INFO_ACTIVE_CU_COUNT:
+		if (rdev->family >= CHIP_BONAIRE)
+			*value = rdev->config.cik.active_cus;
+		else if (rdev->family >= CHIP_TAHITI)
+			*value = rdev->config.si.active_cus;
+		else if (rdev->family >= CHIP_CAYMAN)
+			*value = rdev->config.cayman.active_simds;
+		else if (rdev->family >= CHIP_CEDAR)
+			*value = rdev->config.evergreen.active_simds;
+		else if (rdev->family >= CHIP_RV770)
+			*value = rdev->config.rv770.active_simds;
+		else if (rdev->family >= CHIP_R600)
+			*value = rdev->config.r600.active_simds;
+		else
+			*value = 1;
+		break;
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->request);
 		return -EINVAL;
@@ -859,4 +875,4 @@ const struct drm_ioctl_desc radeon_ioctls_kms[] = {
 	DRM_IOCTL_DEF_DRV(RADEON_GEM_VA, radeon_gem_va_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(RADEON_GEM_OP, radeon_gem_op_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
-int radeon_max_kms_ioctl = DRM_ARRAY_SIZE(radeon_ioctls_kms);
+int radeon_max_kms_ioctl = ARRAY_SIZE(radeon_ioctls_kms);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 6ddf31a2d34e..0592ddb0904b 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -46,6 +46,10 @@ struct radeon_device;
 #define to_radeon_encoder(x) container_of(x, struct radeon_encoder, base)
 #define to_radeon_framebuffer(x) container_of(x, struct radeon_framebuffer, base)
 
+#define RADEON_MAX_HPD_PINS 7
+#define RADEON_MAX_CRTCS 6
+#define RADEON_MAX_AFMT_BLOCKS 7
+
 enum radeon_rmx_type {
 	RMX_OFF,
 	RMX_FULL,
@@ -191,6 +195,7 @@ struct radeon_i2c_chan {
 	struct radeon_i2c_bus_rec rec;
 	struct drm_dp_aux aux;
 	bool has_aux;
+	struct mutex mutex;
 };
 
 /* mostly for macs, but really any system without connector tables */
@@ -232,8 +237,8 @@ struct radeon_mode_info {
 	struct card_info *atom_card_info;
 	enum radeon_connector_table connector_table;
 	bool mode_config_initialized;
-	struct radeon_crtc *crtcs[6];
-	struct radeon_afmt *afmt[7];
+	struct radeon_crtc *crtcs[RADEON_MAX_CRTCS];
+	struct radeon_afmt *afmt[RADEON_MAX_AFMT_BLOCKS];
 	/* DVI-I properties */
 	struct drm_property *coherent_mode_property;
 	/* DAC enable load detect */
@@ -301,6 +306,12 @@ struct radeon_atom_ss {
 	uint16_t amount;
 };
 
+enum radeon_flip_status {
+	RADEON_FLIP_NONE,
+	RADEON_FLIP_PENDING,
+	RADEON_FLIP_SUBMITTED
+};
+
 struct radeon_crtc {
 	struct drm_crtc base;
 	int crtc_id;
@@ -324,8 +335,9 @@ struct radeon_crtc {
 	struct drm_display_mode native_mode;
 	int pll_id;
 	/* page flipping */
-	struct radeon_unpin_work *unpin_work;
-	int deferred_flip_completion;
+	struct workqueue_struct *flip_queue;
+	struct radeon_flip_work *flip_work;
+	enum radeon_flip_status flip_status;
 	/* pll sharing */
 	struct radeon_atom_ss ss;
 	bool ss_enabled;
@@ -505,6 +517,7 @@ struct radeon_connector {
 	struct radeon_i2c_chan *router_bus;
 	enum radeon_connector_audio audio;
 	enum radeon_connector_dither dither;
+	int pixelclock_for_modeset;
 };
 
 struct radeon_framebuffer {
@@ -906,6 +919,7 @@ bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj)
 
 void radeon_fb_output_poll_changed(struct radeon_device *rdev);
 
+void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id);
 void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id);
 
 int radeon_align_pitch(struct radeon_device *rdev, int width, int bpp, bool tiled);
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 4faa4d6f9bb4..6c717b257d6d 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -446,7 +446,7 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 	list_for_each_entry(lobj, head, tv.head) {
 		bo = lobj->robj;
 		if (!bo->pin_count) {
-			u32 domain = lobj->domain;
+			u32 domain = lobj->prefered_domains;
 			u32 current_domain =
 				radeon_mem_type_to_domain(bo->tbo.mem.mem_type);
 
@@ -458,7 +458,7 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 			 * into account. We don't want to disallow buffer moves
 			 * completely.
 			 */
-			if ((lobj->alt_domain & current_domain) != 0 &&
+			if ((lobj->allowed_domains & current_domain) != 0 &&
 			    (domain & current_domain) == 0 && /* will be moved */
 			    bytes_moved > bytes_moved_threshold) {
 				/* don't move it */
@@ -476,8 +476,9 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 				       initial_bytes_moved;
 
 			if (unlikely(r)) {
-				if (r != -ERESTARTSYS && domain != lobj->alt_domain) {
-					domain = lobj->alt_domain;
+				if (r != -ERESTARTSYS &&
+				    domain != lobj->allowed_domains) {
+					domain = lobj->allowed_domains;
 					goto retry;
 				}
 				ttm_eu_backoff_reservation(ticket, head);
@@ -730,7 +731,7 @@ int radeon_bo_wait(struct radeon_bo *bo, u32 *mem_type, bool no_wait)
 {
 	int r;
 
-	r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, 0);
+	r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, NULL);
 	if (unlikely(r != 0))
 		return r;
 	spin_lock(&bo->tbo.bdev->fence_lock);
diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
index 9e7b25a0629d..5a873f31a171 100644
--- a/drivers/gpu/drm/radeon/radeon_object.h
+++ b/drivers/gpu/drm/radeon/radeon_object.h
@@ -65,7 +65,7 @@ static inline int radeon_bo_reserve(struct radeon_bo *bo, bool no_intr)
 {
 	int r;
 
-	r = ttm_bo_reserve(&bo->tbo, !no_intr, false, false, 0);
+	r = ttm_bo_reserve(&bo->tbo, !no_intr, false, false, NULL);
 	if (unlikely(r != 0)) {
 		if (r != -ERESTARTSYS)
 			dev_err(bo->rdev->dev, "%p reserve failed\n", bo);
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 2bdae61c0ac0..e447e390d09a 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -73,8 +73,10 @@ void radeon_pm_acpi_event_handler(struct radeon_device *rdev)
 			rdev->pm.dpm.ac_power = true;
 		else
 			rdev->pm.dpm.ac_power = false;
-		if (rdev->asic->dpm.enable_bapm)
-			radeon_dpm_enable_bapm(rdev, rdev->pm.dpm.ac_power);
+		if (rdev->family == CHIP_ARUBA) {
+			if (rdev->asic->dpm.enable_bapm)
+				radeon_dpm_enable_bapm(rdev, rdev->pm.dpm.ac_power);
+		}
 		mutex_unlock(&rdev->pm.mutex);
         } else if (rdev->pm.pm_method == PM_METHOD_PROFILE) {
 		if (rdev->pm.profile == PM_PROFILE_AUTO) {
@@ -984,6 +986,8 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
 		if (enable) {
 			mutex_lock(&rdev->pm.mutex);
 			rdev->pm.dpm.uvd_active = true;
+			/* disable this for now */
+#if 0
 			if ((rdev->pm.dpm.sd == 1) && (rdev->pm.dpm.hd == 0))
 				dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_SD;
 			else if ((rdev->pm.dpm.sd == 2) && (rdev->pm.dpm.hd == 0))
@@ -993,6 +997,7 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
 			else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 2))
 				dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD2;
 			else
+#endif
 				dpm_state = POWER_STATE_TYPE_INTERNAL_UVD;
 			rdev->pm.dpm.state = dpm_state;
 			mutex_unlock(&rdev->pm.mutex);
diff --git a/drivers/gpu/drm/radeon/radeon_state.c b/drivers/gpu/drm/radeon/radeon_state.c
index 956ab7f14e16..23bb64fd775f 100644
--- a/drivers/gpu/drm/radeon/radeon_state.c
+++ b/drivers/gpu/drm/radeon/radeon_state.c
@@ -3054,7 +3054,7 @@ static int radeon_cp_getparam(struct drm_device *dev, void *data, struct drm_fil
 		if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_R600)
 			value = 0;
 		else
-			value = drm_dev_to_irq(dev);
+			value = dev->pdev->irq;
 		break;
 	case RADEON_PARAM_GART_BASE:
 		value = dev_priv->gart_vm_start;
@@ -3258,4 +3258,4 @@ struct drm_ioctl_desc radeon_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(RADEON_CS, r600_cs_legacy_ioctl, DRM_AUTH)
 };
 
-int radeon_max_ioctl = DRM_ARRAY_SIZE(radeon_ioctls);
+int radeon_max_ioctl = ARRAY_SIZE(radeon_ioctls);
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 1b65ae2433cd..a4ad270e8261 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -812,7 +812,8 @@ void radeon_uvd_note_usage(struct radeon_device *rdev)
 		    (rdev->pm.dpm.hd != hd)) {
 			rdev->pm.dpm.sd = sd;
 			rdev->pm.dpm.hd = hd;
-			streams_changed = true;
+			/* disable this for now */
+			/*streams_changed = true;*/
 		}
 	}
 
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index 3971d968af6c..aa21c31a846c 100644
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -66,6 +66,7 @@ int radeon_vce_init(struct radeon_device *rdev)
 	case CHIP_BONAIRE:
 	case CHIP_KAVERI:
 	case CHIP_KABINI:
+	case CHIP_HAWAII:
 	case CHIP_MULLINS:
 		fw_name = FIRMWARE_BONAIRE;
 		break;
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index c11b71d249e3..eecff6bbd341 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -59,7 +59,7 @@
  */
 static unsigned radeon_vm_num_pdes(struct radeon_device *rdev)
 {
-	return rdev->vm_manager.max_pfn >> RADEON_VM_BLOCK_SIZE;
+	return rdev->vm_manager.max_pfn >> radeon_vm_block_size;
 }
 
 /**
@@ -140,8 +140,8 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
 	/* add the vm page table to the list */
 	list[0].gobj = NULL;
 	list[0].robj = vm->page_directory;
-	list[0].domain = RADEON_GEM_DOMAIN_VRAM;
-	list[0].alt_domain = RADEON_GEM_DOMAIN_VRAM;
+	list[0].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
+	list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
 	list[0].tv.bo = &vm->page_directory->tbo;
 	list[0].tiling_flags = 0;
 	list[0].handle = 0;
@@ -153,8 +153,8 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
 
 		list[idx].gobj = NULL;
 		list[idx].robj = vm->page_tables[i].bo;
-		list[idx].domain = RADEON_GEM_DOMAIN_VRAM;
-		list[idx].alt_domain = RADEON_GEM_DOMAIN_VRAM;
+		list[idx].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
+		list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
 		list[idx].tv.bo = &list[idx].robj->tbo;
 		list[idx].tiling_flags = 0;
 		list[idx].handle = 0;
@@ -474,8 +474,10 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 	bo_va->valid = false;
 	list_move(&bo_va->vm_list, head);
 
-	soffset = (soffset / RADEON_GPU_PAGE_SIZE) >> RADEON_VM_BLOCK_SIZE;
-	eoffset = (eoffset / RADEON_GPU_PAGE_SIZE) >> RADEON_VM_BLOCK_SIZE;
+	soffset = (soffset / RADEON_GPU_PAGE_SIZE) >> radeon_vm_block_size;
+	eoffset = (eoffset / RADEON_GPU_PAGE_SIZE) >> radeon_vm_block_size;
+
+	BUG_ON(eoffset >= radeon_vm_num_pdes(rdev));
 
 	if (eoffset > vm->max_pde_used)
 		vm->max_pde_used = eoffset;
@@ -493,7 +495,7 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 		mutex_unlock(&vm->mutex);
 
 		r = radeon_bo_create(rdev, RADEON_VM_PTE_COUNT * 8,
-				     RADEON_GPU_PAGE_SIZE, false, 
+				     RADEON_GPU_PAGE_SIZE, true,
 				     RADEON_GEM_DOMAIN_VRAM, NULL, &pt);
 		if (r)
 			return r;
@@ -583,10 +585,9 @@ static uint32_t radeon_vm_page_flags(uint32_t flags)
 int radeon_vm_update_page_directory(struct radeon_device *rdev,
 				    struct radeon_vm *vm)
 {
-	static const uint32_t incr = RADEON_VM_PTE_COUNT * 8;
-
 	struct radeon_bo *pd = vm->page_directory;
 	uint64_t pd_addr = radeon_bo_gpu_offset(pd);
+	uint32_t incr = RADEON_VM_PTE_COUNT * 8;
 	uint64_t last_pde = ~0, last_pt = ~0;
 	unsigned count = 0, pt_idx, ndw;
 	struct radeon_ib ib;
@@ -660,6 +661,84 @@ int radeon_vm_update_page_directory(struct radeon_device *rdev,
 }
 
 /**
+ * radeon_vm_frag_ptes - add fragment information to PTEs
+ *
+ * @rdev: radeon_device pointer
+ * @ib: IB for the update
+ * @pe_start: first PTE to handle
+ * @pe_end: last PTE to handle
+ * @addr: addr those PTEs should point to
+ * @flags: hw mapping flags
+ *
+ * Global and local mutex must be locked!
+ */
+static void radeon_vm_frag_ptes(struct radeon_device *rdev,
+				struct radeon_ib *ib,
+				uint64_t pe_start, uint64_t pe_end,
+				uint64_t addr, uint32_t flags)
+{
+	/**
+	 * The MC L1 TLB supports variable sized pages, based on a fragment
+	 * field in the PTE. When this field is set to a non-zero value, page
+	 * granularity is increased from 4KB to (1 << (12 + frag)). The PTE
+	 * flags are considered valid for all PTEs within the fragment range
+	 * and corresponding mappings are assumed to be physically contiguous.
+	 *
+	 * The L1 TLB can store a single PTE for the whole fragment,
+	 * significantly increasing the space available for translation
+	 * caching. This leads to large improvements in throughput when the
+	 * TLB is under pressure.
+	 *
+	 * The L2 TLB distributes small and large fragments into two
+	 * asymmetric partitions. The large fragment cache is significantly
+	 * larger. Thus, we try to use large fragments wherever possible.
+	 * Userspace can support this by aligning virtual base address and
+	 * allocation size to the fragment size.
+	 */
+
+	/* NI is optimized for 256KB fragments, SI and newer for 64KB */
+	uint64_t frag_flags = rdev->family == CHIP_CAYMAN ?
+			R600_PTE_FRAG_256KB : R600_PTE_FRAG_64KB;
+	uint64_t frag_align = rdev->family == CHIP_CAYMAN ? 0x200 : 0x80;
+
+	uint64_t frag_start = ALIGN(pe_start, frag_align);
+	uint64_t frag_end = pe_end & ~(frag_align - 1);
+
+	unsigned count;
+
+	/* system pages are non continuously */
+	if ((flags & R600_PTE_SYSTEM) || !(flags & R600_PTE_VALID) ||
+	    (frag_start >= frag_end)) {
+
+		count = (pe_end - pe_start) / 8;
+		radeon_asic_vm_set_page(rdev, ib, pe_start, addr, count,
+					RADEON_GPU_PAGE_SIZE, flags);
+		return;
+	}
+
+	/* handle the 4K area at the beginning */
+	if (pe_start != frag_start) {
+		count = (frag_start - pe_start) / 8;
+		radeon_asic_vm_set_page(rdev, ib, pe_start, addr, count,
+					RADEON_GPU_PAGE_SIZE, flags);
+		addr += RADEON_GPU_PAGE_SIZE * count;
+	}
+
+	/* handle the area in the middle */
+	count = (frag_end - frag_start) / 8;
+	radeon_asic_vm_set_page(rdev, ib, frag_start, addr, count,
+				RADEON_GPU_PAGE_SIZE, flags | frag_flags);
+
+	/* handle the 4K area at the end */
+	if (frag_end != pe_end) {
+		addr += RADEON_GPU_PAGE_SIZE * count;
+		count = (pe_end - frag_end) / 8;
+		radeon_asic_vm_set_page(rdev, ib, frag_end, addr, count,
+					RADEON_GPU_PAGE_SIZE, flags);
+	}
+}
+
+/**
  * radeon_vm_update_ptes - make sure that page tables are valid
  *
  * @rdev: radeon_device pointer
@@ -679,8 +758,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 				  uint64_t start, uint64_t end,
 				  uint64_t dst, uint32_t flags)
 {
-	static const uint64_t mask = RADEON_VM_PTE_COUNT - 1;
-
+	uint64_t mask = RADEON_VM_PTE_COUNT - 1;
 	uint64_t last_pte = ~0, last_dst = ~0;
 	unsigned count = 0;
 	uint64_t addr;
@@ -690,7 +768,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 
 	/* walk over the address space and update the page tables */
 	for (addr = start; addr < end; ) {
-		uint64_t pt_idx = addr >> RADEON_VM_BLOCK_SIZE;
+		uint64_t pt_idx = addr >> radeon_vm_block_size;
 		struct radeon_bo *pt = vm->page_tables[pt_idx].bo;
 		unsigned nptes;
 		uint64_t pte;
@@ -708,10 +786,9 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 		if ((last_pte + 8 * count) != pte) {
 
 			if (count) {
-				radeon_asic_vm_set_page(rdev, ib, last_pte,
-							last_dst, count,
-							RADEON_GPU_PAGE_SIZE,
-							flags);
+				radeon_vm_frag_ptes(rdev, ib, last_pte,
+						    last_pte + 8 * count,
+						    last_dst, flags);
 			}
 
 			count = nptes;
@@ -726,9 +803,9 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 	}
 
 	if (count) {
-		radeon_asic_vm_set_page(rdev, ib, last_pte,
-					last_dst, count,
-					RADEON_GPU_PAGE_SIZE, flags);
+		radeon_vm_frag_ptes(rdev, ib, last_pte,
+				    last_pte + 8 * count,
+				    last_dst, flags);
 	}
 }
 
@@ -796,13 +873,13 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
 	/* padding, etc. */
 	ndw = 64;
 
-	if (RADEON_VM_BLOCK_SIZE > 11)
+	if (radeon_vm_block_size > 11)
 		/* reserve space for one header for every 2k dwords */
 		ndw += (nptes >> 11) * 4;
 	else
 		/* reserve space for one header for
 		    every (1 << BLOCK_SIZE) entries */
-		ndw += (nptes >> RADEON_VM_BLOCK_SIZE) * 4;
+		ndw += (nptes >> radeon_vm_block_size) * 4;
 
 	/* reserve space for pte addresses */
 	ndw += nptes * 2;
@@ -892,6 +969,8 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev,
  */
 int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
 {
+	const unsigned align = min(RADEON_VM_PTB_ALIGN_SIZE,
+		RADEON_VM_PTE_COUNT * 8);
 	unsigned pd_size, pd_entries, pts_size;
 	int r;
 
@@ -913,7 +992,7 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
 		return -ENOMEM;
 	}
 
-	r = radeon_bo_create(rdev, pd_size, RADEON_VM_PTB_ALIGN_SIZE, false,
+	r = radeon_bo_create(rdev, pd_size, align, true,
 			     RADEON_GEM_DOMAIN_VRAM, NULL,
 			     &vm->page_directory);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index 130d5cc50d43..a0f96decece3 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -212,21 +212,16 @@ void rs400_gart_fini(struct radeon_device *rdev)
 #define RS400_PTE_WRITEABLE (1 << 2)
 #define RS400_PTE_READABLE  (1 << 3)
 
-int rs400_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void rs400_gart_set_page(struct radeon_device *rdev, unsigned i, uint64_t addr)
 {
 	uint32_t entry;
 	u32 *gtt = rdev->gart.ptr;
 
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
-
 	entry = (lower_32_bits(addr) & PAGE_MASK) |
 		((upper_32_bits(addr) & 0xff) << 4) |
 		RS400_PTE_WRITEABLE | RS400_PTE_READABLE;
 	entry = cpu_to_le32(entry);
 	gtt[i] = entry;
-	return 0;
 }
 
 int rs400_mc_wait_for_idle(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 72d3616de08e..d1a35cb1c91d 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -109,19 +109,7 @@ void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc)
 	}
 }
 
-void rs600_pre_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* enable the pflip int */
-	radeon_irq_kms_pflip_irq_get(rdev, crtc);
-}
-
-void rs600_post_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* disable the pflip int */
-	radeon_irq_kms_pflip_irq_put(rdev, crtc);
-}
-
-u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
@@ -148,9 +136,15 @@ u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~AVIVO_D1GRPH_UPDATE_LOCK;
 	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+}
+
+bool rs600_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 
 	/* Return current update_pending status: */
-	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
+	return !!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) &
+		AVIVO_D1GRPH_SURFACE_UPDATE_PENDING);
 }
 
 void avivo_program_fmt(struct drm_encoder *encoder)
@@ -632,24 +626,16 @@ static void rs600_gart_fini(struct radeon_device *rdev)
 	radeon_gart_table_vram_free(rdev);
 }
 
-#define R600_PTE_VALID     (1 << 0)
-#define R600_PTE_SYSTEM    (1 << 1)
-#define R600_PTE_SNOOPED   (1 << 2)
-#define R600_PTE_READABLE  (1 << 5)
-#define R600_PTE_WRITEABLE (1 << 6)
-
-int rs600_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void rs600_gart_set_page(struct radeon_device *rdev, unsigned i, uint64_t addr)
 {
 	void __iomem *ptr = (void *)rdev->gart.ptr;
 
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
 	addr = addr & 0xFFFFFFFFFFFFF000ULL;
-	addr |= R600_PTE_VALID | R600_PTE_SYSTEM | R600_PTE_SNOOPED;
-	addr |= R600_PTE_READABLE | R600_PTE_WRITEABLE;
+	if (addr == rdev->dummy_page.addr)
+		addr |= R600_PTE_SYSTEM | R600_PTE_SNOOPED;
+	else
+		addr |= R600_PTE_GART;
 	writeq(addr, ptr + (i * 8));
-	return 0;
 }
 
 int rs600_irq_set(struct radeon_device *rdev)
@@ -787,7 +773,7 @@ int rs600_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[0]))
-				radeon_crtc_handle_flip(rdev, 0);
+				radeon_crtc_handle_vblank(rdev, 0);
 		}
 		if (G_007EDC_LB_D2_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			if (rdev->irq.crtc_vblank_int[1]) {
@@ -796,7 +782,7 @@ int rs600_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[1]))
-				radeon_crtc_handle_flip(rdev, 1);
+				radeon_crtc_handle_vblank(rdev, 1);
 		}
 		if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			queue_hotplug = true;
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index fef310773aad..da8703d8d455 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -801,7 +801,7 @@ u32 rv770_get_xclk(struct radeon_device *rdev)
 	return reference_clock;
 }
 
-u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
@@ -835,9 +835,15 @@ u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~AVIVO_D1GRPH_UPDATE_LOCK;
 	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+}
+
+bool rv770_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 
 	/* Return current update_pending status: */
-	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
+	return !!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) &
+		AVIVO_D1GRPH_SURFACE_UPDATE_PENDING);
 }
 
 /* get temperature in millidegrees */
@@ -1321,6 +1327,9 @@ static void rv770_gpu_init(struct radeon_device *rdev)
 	if (tmp < rdev->config.rv770.max_simds) {
 		rdev->config.rv770.max_simds = tmp;
 	}
+	tmp = rdev->config.rv770.max_simds -
+		r600_count_pipe_bits((cc_gc_shader_pipe_config >> 16) & R7XX_MAX_SIMDS_MASK);
+	rdev->config.rv770.active_simds = tmp;
 
 	switch (rdev->config.rv770.max_tile_pipes) {
 	case 1:
diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c
index da041a43d82e..3c76e1dcdf04 100644
--- a/drivers/gpu/drm/radeon/rv770_dpm.c
+++ b/drivers/gpu/drm/radeon/rv770_dpm.c
@@ -2329,12 +2329,6 @@ void rv770_get_engine_memory_ss(struct radeon_device *rdev)
 	pi->mclk_ss = radeon_atombios_get_asic_ss_info(rdev, &ss,
 						       ASIC_INTERNAL_MEMORY_SS, 0);
 
-	/* disable ss, causes hangs on some cayman boards */
-	if (rdev->family == CHIP_CAYMAN) {
-		pi->sclk_ss = false;
-		pi->mclk_ss = false;
-	}
-
 	if (pi->sclk_ss || pi->mclk_ss)
 		pi->dynamic_ss = true;
 	else
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 22a63c98ba14..eba0225259a4 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -71,6 +71,7 @@ MODULE_FIRMWARE("radeon/HAINAN_mc2.bin");
 MODULE_FIRMWARE("radeon/HAINAN_rlc.bin");
 MODULE_FIRMWARE("radeon/HAINAN_smc.bin");
 
+static u32 si_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh);
 static void si_pcie_gen3_enable(struct radeon_device *rdev);
 static void si_program_aspm(struct radeon_device *rdev);
 extern void sumo_rlc_fini(struct radeon_device *rdev);
@@ -2900,7 +2901,7 @@ static void si_gpu_init(struct radeon_device *rdev)
 	u32 sx_debug_1;
 	u32 hdp_host_path_cntl;
 	u32 tmp;
-	int i, j;
+	int i, j, k;
 
 	switch (rdev->family) {
 	case CHIP_TAHITI:
@@ -3098,6 +3099,14 @@ static void si_gpu_init(struct radeon_device *rdev)
 		     rdev->config.si.max_sh_per_se,
 		     rdev->config.si.max_cu_per_sh);
 
+	for (i = 0; i < rdev->config.si.max_shader_engines; i++) {
+		for (j = 0; j < rdev->config.si.max_sh_per_se; j++) {
+			for (k = 0; k < rdev->config.si.max_cu_per_sh; k++) {
+				rdev->config.si.active_cus +=
+					hweight32(si_get_cu_active_bitmap(rdev, i, j));
+			}
+		}
+	}
 
 	/* set HW defaults for 3D engine */
 	WREG32(CP_QUEUE_THRESHOLDS, (ROQ_IB1_START(0x16) |
@@ -3186,7 +3195,7 @@ void si_fence_ring_emit(struct radeon_device *rdev,
 	/* EVENT_WRITE_EOP - flush caches, send int */
 	radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
 	radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | EVENT_INDEX(5));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
 	radeon_ring_write(ring, fence->seq);
 	radeon_ring_write(ring, 0);
@@ -3219,7 +3228,7 @@ void si_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 			radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 			radeon_ring_write(ring, (1 << 8));
 			radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
-			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr));
 			radeon_ring_write(ring, next_rptr);
 		}
 
@@ -4044,18 +4053,21 @@ static int si_pcie_gart_enable(struct radeon_device *rdev)
 	WREG32(MC_VM_MX_L1_TLB_CNTL,
 	       (0xA << 7) |
 	       ENABLE_L1_TLB |
+	       ENABLE_L1_FRAGMENT_PROCESSING |
 	       SYSTEM_ACCESS_MODE_NOT_IN_SYS |
 	       ENABLE_ADVANCED_DRIVER_MODEL |
 	       SYSTEM_APERTURE_UNMAPPED_ACCESS_PASS_THRU);
 	/* Setup L2 cache */
 	WREG32(VM_L2_CNTL, ENABLE_L2_CACHE |
+	       ENABLE_L2_FRAGMENT_PROCESSING |
 	       ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE |
 	       ENABLE_L2_PDE0_CACHE_LRU_UPDATE_BY_WRITE |
 	       EFFECTIVE_L2_QUEUE_SIZE(7) |
 	       CONTEXT1_IDENTITY_ACCESS_MODE(1));
 	WREG32(VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS | INVALIDATE_L2_CACHE);
 	WREG32(VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY |
-	       L2_CACHE_BIGK_FRAGMENT_SIZE(0));
+	       BANK_SELECT(4) |
+	       L2_CACHE_BIGK_FRAGMENT_SIZE(4));
 	/* setup context0 */
 	WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR, rdev->mc.gtt_start >> 12);
 	WREG32(VM_CONTEXT0_PAGE_TABLE_END_ADDR, rdev->mc.gtt_end >> 12);
@@ -4092,6 +4104,7 @@ static int si_pcie_gart_enable(struct radeon_device *rdev)
 	       (u32)(rdev->dummy_page.addr >> 12));
 	WREG32(VM_CONTEXT1_CNTL2, 4);
 	WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) |
+				PAGE_TABLE_BLOCK_SIZE(radeon_vm_block_size - 9) |
 				RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT |
 				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
@@ -6151,7 +6164,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -6177,7 +6190,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
@@ -6203,7 +6216,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_flip(rdev, 2);
+						radeon_crtc_handle_vblank(rdev, 2);
 					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
@@ -6229,7 +6242,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_flip(rdev, 3);
+						radeon_crtc_handle_vblank(rdev, 3);
 					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
@@ -6255,7 +6268,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_flip(rdev, 4);
+						radeon_crtc_handle_vblank(rdev, 4);
 					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
@@ -6281,7 +6294,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_flip(rdev, 5);
+						radeon_crtc_handle_vblank(rdev, 5);
 					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
@@ -6363,14 +6376,16 @@ restart_ih:
 		case 147:
 			addr = RREG32(VM_CONTEXT1_PROTECTION_FAULT_ADDR);
 			status = RREG32(VM_CONTEXT1_PROTECTION_FAULT_STATUS);
+			/* reset addr and status */
+			WREG32_P(VM_CONTEXT1_CNTL2, 1, ~1);
+			if (addr == 0x0 && status == 0x0)
+				break;
 			dev_err(rdev->dev, "GPU fault detected: %d 0x%08x\n", src_id, src_data);
 			dev_err(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
 				addr);
 			dev_err(rdev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 				status);
 			si_vm_decode_fault(rdev, status, addr);
-			/* reset addr and status */
-			WREG32_P(VM_CONTEXT1_CNTL2, 1, ~1);
 			break;
 		case 176: /* RINGID0 CP_INT */
 			radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index de0ca070122f..e24c94b6d14d 100644
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -79,7 +79,25 @@ void si_dma_vm_set_page(struct radeon_device *rdev,
 
 	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
 
-	if (flags & R600_PTE_SYSTEM) {
+	if (flags == R600_PTE_GART) {
+		uint64_t src = rdev->gart.table_addr + (addr >> 12) * 8;
+		while (count) {
+			unsigned bytes = count * 8;
+			if (bytes > 0xFFFF8)
+				bytes = 0xFFFF8;
+
+			ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_COPY,
+							      1, 0, 0, bytes);
+			ib->ptr[ib->length_dw++] = lower_32_bits(pe);
+			ib->ptr[ib->length_dw++] = lower_32_bits(src);
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
+			ib->ptr[ib->length_dw++] = upper_32_bits(src) & 0xff;
+
+			pe += bytes;
+			src += bytes;
+			count -= bytes / 8;
+		}
+	} else if (flags & R600_PTE_SYSTEM) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
@@ -202,8 +220,8 @@ int si_copy_dma(struct radeon_device *rdev,
 			cur_size_in_bytes = 0xFFFFF;
 		size_in_bytes -= cur_size_in_bytes;
 		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_COPY, 1, 0, 0, cur_size_in_bytes));
-		radeon_ring_write(ring, dst_offset & 0xffffffff);
-		radeon_ring_write(ring, src_offset & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
+		radeon_ring_write(ring, lower_32_bits(src_offset));
 		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
 		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xff);
 		src_offset += cur_size_in_bytes;
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 9a3567bedaae..58918868f894 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -1948,6 +1948,10 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 			si_pi->cac_weights = cac_weights_cape_verde_pro;
 			si_pi->dte_data = dte_data_cape_verde;
 			break;
+		case 0x682C:
+			si_pi->cac_weights = cac_weights_cape_verde_pro;
+			si_pi->dte_data = dte_data_sun_xt;
+			break;
 		case 0x6825:
 		case 0x6827:
 			si_pi->cac_weights = cac_weights_heathrow;
@@ -1971,10 +1975,9 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 			si_pi->dte_data = dte_data_venus_xt;
 			break;
 		case 0x6823:
-			si_pi->cac_weights = cac_weights_chelsea_pro;
-			si_pi->dte_data = dte_data_venus_pro;
-			break;
 		case 0x682B:
+		case 0x6822:
+		case 0x682A:
 			si_pi->cac_weights = cac_weights_chelsea_pro;
 			si_pi->dte_data = dte_data_venus_pro;
 			break;
@@ -1988,6 +1991,7 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 		case 0x6601:
 		case 0x6621:
 		case 0x6603:
+		case 0x6605:
 			si_pi->cac_weights = cac_weights_mars_pro;
 			si_pi->lcac_config = lcac_mars_pro;
 			si_pi->cac_override = cac_override_oland;
@@ -1998,6 +2002,7 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 		case 0x6600:
 		case 0x6606:
 		case 0x6620:
+		case 0x6604:
 			si_pi->cac_weights = cac_weights_mars_xt;
 			si_pi->lcac_config = lcac_mars_pro;
 			si_pi->cac_override = cac_override_oland;
@@ -2006,6 +2011,8 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 			update_dte_from_pl2 = true;
 			break;
 		case 0x6611:
+		case 0x6613:
+		case 0x6608:
 			si_pi->cac_weights = cac_weights_oland_pro;
 			si_pi->lcac_config = lcac_mars_pro;
 			si_pi->cac_override = cac_override_oland;
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index 7321283602ce..fd414d34d885 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -362,6 +362,7 @@
 #define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
 #define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
 #define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
+#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
 #define VM_CONTEXT1_CNTL				0x1414
 #define VM_CONTEXT0_CNTL2				0x1430
 #define VM_CONTEXT1_CNTL2				0x1434
diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c
index 2a2822c03329..20da6ff183df 100644
--- a/drivers/gpu/drm/radeon/trinity_dpm.c
+++ b/drivers/gpu/drm/radeon/trinity_dpm.c
@@ -1874,7 +1874,15 @@ int trinity_dpm_init(struct radeon_device *rdev)
 	for (i = 0; i < SUMO_MAX_HARDWARE_POWERLEVELS; i++)
 		pi->at[i] = TRINITY_AT_DFLT;
 
-	pi->enable_bapm = false;
+	/* There are stability issues reported on latops with
+	 * bapm installed when switching between AC and battery
+	 * power.  At the same time, some desktop boards hang
+	 * if it's not enabled and dpm is enabled.
+	 */
+	if (rdev->flags & RADEON_IS_MOBILITY)
+		pi->enable_bapm = false;
+	else
+		pi->enable_bapm = true;
 	pi->enable_nbps_policy = true;
 	pi->enable_sclk_ds = true;
 	pi->enable_gfx_power_gating = true;
diff --git a/drivers/gpu/drm/radeon/uvd_v2_2.c b/drivers/gpu/drm/radeon/uvd_v2_2.c
index d1771004cb52..8bfdadd56598 100644
--- a/drivers/gpu/drm/radeon/uvd_v2_2.c
+++ b/drivers/gpu/drm/radeon/uvd_v2_2.c
@@ -45,7 +45,7 @@ void uvd_v2_2_fence_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, PACKET0(UVD_CONTEXT_ID, 0));
 	radeon_ring_write(ring, fence->seq);
 	radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_DATA0, 0));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_DATA1, 0));
 	radeon_ring_write(ring, upper_32_bits(addr) & 0xff);
 	radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_CMD, 0));
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index d8e835ac2c5e..2e3d7b5b0ad7 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -1,6 +1,7 @@
 config DRM_RCAR_DU
 	tristate "DRM Support for R-Car Display Unit"
 	depends on DRM && ARM
+	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
@@ -12,6 +13,7 @@ config DRM_RCAR_DU
 config DRM_RCAR_LVDS
 	bool "R-Car DU LVDS Encoder Support"
 	depends on DRM_RCAR_DU
+	depends on ARCH_R8A7790 || ARCH_R8A7791 || COMPILE_TEST
 	help
 	  Enable support the R-Car Display Unit embedded LVDS encoders
 	  (currently only on R8A7790).
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c b/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
index 4f3ba93cd91d..289048d1c7b2 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
@@ -57,15 +57,8 @@ static int rcar_du_lvds_connector_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int rcar_du_lvds_connector_mode_valid(struct drm_connector *connector,
-					    struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static const struct drm_connector_helper_funcs connector_helper_funcs = {
 	.get_modes = rcar_du_lvds_connector_get_modes,
-	.mode_valid = rcar_du_lvds_connector_mode_valid,
 	.best_encoder = rcar_du_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c b/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c
index 41d563adfeaa..ccfe64c7188f 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c
@@ -25,15 +25,8 @@ static int rcar_du_vga_connector_get_modes(struct drm_connector *connector)
 	return 0;
 }
 
-static int rcar_du_vga_connector_mode_valid(struct drm_connector *connector,
-					    struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static const struct drm_connector_helper_funcs connector_helper_funcs = {
 	.get_modes = rcar_du_vga_connector_get_modes,
-	.mode_valid = rcar_du_vga_connector_mode_valid,
 	.best_encoder = rcar_du_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/savage/savage_bci.c b/drivers/gpu/drm/savage/savage_bci.c
index d2b2df9e26f3..c97cdc9ab239 100644
--- a/drivers/gpu/drm/savage/savage_bci.c
+++ b/drivers/gpu/drm/savage/savage_bci.c
@@ -1079,4 +1079,4 @@ const struct drm_ioctl_desc savage_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(SAVAGE_BCI_EVENT_WAIT, savage_bci_event_wait, DRM_AUTH),
 };
 
-int savage_max_ioctl = DRM_ARRAY_SIZE(savage_ioctls);
+int savage_max_ioctl = ARRAY_SIZE(savage_ioctls);
diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/shmobile/Kconfig
index 2ee44ca9d67f..a50fe0eeaa0d 100644
--- a/drivers/gpu/drm/shmobile/Kconfig
+++ b/drivers/gpu/drm/shmobile/Kconfig
@@ -1,6 +1,7 @@
 config DRM_SHMOBILE
 	tristate "DRM Support for SH Mobile"
-	depends on DRM && (ARM || SUPERH)
+	depends on DRM && ARM
+	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select BACKLIGHT_CLASS_DEVICE
 	select DRM_KMS_HELPER
 	select DRM_KMS_FB_HELPER
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index e9e5e6d368cc..faf176b2daf9 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -674,12 +674,6 @@ static int shmob_drm_connector_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int shmob_drm_connector_mode_valid(struct drm_connector *connector,
-					  struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *
 shmob_drm_connector_best_encoder(struct drm_connector *connector)
 {
@@ -690,7 +684,6 @@ shmob_drm_connector_best_encoder(struct drm_connector *connector)
 
 static const struct drm_connector_helper_funcs connector_helper_funcs = {
 	.get_modes = shmob_drm_connector_get_modes,
-	.mode_valid = shmob_drm_connector_mode_valid,
 	.best_encoder = shmob_drm_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index c839c9c89efb..82c84c7fd4f6 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -185,7 +185,7 @@ static int shmob_drm_load(struct drm_device *dev, unsigned long flags)
 		goto done;
 	}
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	if (ret < 0) {
 		dev_err(&pdev->dev, "failed to install IRQ handler\n");
 		goto done;
diff --git a/drivers/gpu/drm/sis/sis_mm.c b/drivers/gpu/drm/sis/sis_mm.c
index 0573be0d2933..77f288e4a0a6 100644
--- a/drivers/gpu/drm/sis/sis_mm.c
+++ b/drivers/gpu/drm/sis/sis_mm.c
@@ -359,4 +359,4 @@ const struct drm_ioctl_desc sis_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(SIS_FB_INIT, sis_fb_init, DRM_AUTH | DRM_MASTER | DRM_ROOT_ONLY),
 };
 
-int sis_max_ioctl = DRM_ARRAY_SIZE(sis_ioctls);
+int sis_max_ioctl = ARRAY_SIZE(sis_ioctls);
diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile
index d43f21bb4596..2c66a8db9da4 100644
--- a/drivers/gpu/drm/tegra/Makefile
+++ b/drivers/gpu/drm/tegra/Makefile
@@ -1,7 +1,6 @@
 ccflags-$(CONFIG_DRM_TEGRA_DEBUG) += -DDEBUG
 
 tegra-drm-y := \
-	bus.o \
 	drm.o \
 	gem.o \
 	fb.o \
diff --git a/drivers/gpu/drm/tegra/bus.c b/drivers/gpu/drm/tegra/bus.c
deleted file mode 100644
index 71cef5c13dc8..000000000000
--- a/drivers/gpu/drm/tegra/bus.c
+++ /dev/null
@@ -1,75 +0,0 @@
-/*
- * Copyright (C) 2013 NVIDIA Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include "drm.h"
-
-static int drm_host1x_set_busid(struct drm_device *dev,
-				struct drm_master *master)
-{
-	const char *device = dev_name(dev->dev);
-	const char *driver = dev->driver->name;
-	const char *bus = dev->dev->bus->name;
-	int length;
-
-	master->unique_len = strlen(bus) + 1 + strlen(device);
-	master->unique_size = master->unique_len;
-
-	master->unique = kmalloc(master->unique_len + 1, GFP_KERNEL);
-	if (!master->unique)
-		return -ENOMEM;
-
-	snprintf(master->unique, master->unique_len + 1, "%s:%s", bus, device);
-
-	length = strlen(driver) + 1 + master->unique_len;
-
-	dev->devname = kmalloc(length + 1, GFP_KERNEL);
-	if (!dev->devname)
-		return -ENOMEM;
-
-	snprintf(dev->devname, length + 1, "%s@%s", driver, master->unique);
-
-	return 0;
-}
-
-static struct drm_bus drm_host1x_bus = {
-	.bus_type = DRIVER_BUS_HOST1X,
-	.set_busid = drm_host1x_set_busid,
-};
-
-int drm_host1x_init(struct drm_driver *driver, struct host1x_device *device)
-{
-	struct drm_device *drm;
-	int ret;
-
-	driver->bus = &drm_host1x_bus;
-
-	drm = drm_dev_alloc(driver, &device->dev);
-	if (!drm)
-		return -ENOMEM;
-
-	ret = drm_dev_register(drm, 0);
-	if (ret)
-		goto err_free;
-
-	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n", driver->name,
-		 driver->major, driver->minor, driver->patchlevel,
-		 driver->date, drm->primary->index);
-
-	return 0;
-
-err_free:
-	drm_dev_unref(drm);
-	return ret;
-}
-
-void drm_host1x_exit(struct drm_driver *driver, struct host1x_device *device)
-{
-	struct tegra_drm *tegra = dev_get_drvdata(&device->dev);
-
-	drm_put_dev(tegra->drm);
-}
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index edb871d7d395..ef40381f3909 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -17,6 +17,7 @@
 
 struct tegra_dc_soc_info {
 	bool supports_interlacing;
+	bool supports_cursor;
 };
 
 struct tegra_plane {
@@ -29,6 +30,254 @@ static inline struct tegra_plane *to_tegra_plane(struct drm_plane *plane)
 	return container_of(plane, struct tegra_plane, base);
 }
 
+static unsigned int tegra_dc_format(uint32_t format, uint32_t *swap)
+{
+	/* assume no swapping of fetched data */
+	if (swap)
+		*swap = BYTE_SWAP_NOSWAP;
+
+	switch (format) {
+	case DRM_FORMAT_XBGR8888:
+		return WIN_COLOR_DEPTH_R8G8B8A8;
+
+	case DRM_FORMAT_XRGB8888:
+		return WIN_COLOR_DEPTH_B8G8R8A8;
+
+	case DRM_FORMAT_RGB565:
+		return WIN_COLOR_DEPTH_B5G6R5;
+
+	case DRM_FORMAT_UYVY:
+		return WIN_COLOR_DEPTH_YCbCr422;
+
+	case DRM_FORMAT_YUYV:
+		if (swap)
+			*swap = BYTE_SWAP_SWAP2;
+
+		return WIN_COLOR_DEPTH_YCbCr422;
+
+	case DRM_FORMAT_YUV420:
+		return WIN_COLOR_DEPTH_YCbCr420P;
+
+	case DRM_FORMAT_YUV422:
+		return WIN_COLOR_DEPTH_YCbCr422P;
+
+	default:
+		break;
+	}
+
+	WARN(1, "unsupported pixel format %u, using default\n", format);
+	return WIN_COLOR_DEPTH_B8G8R8A8;
+}
+
+static bool tegra_dc_format_is_yuv(unsigned int format, bool *planar)
+{
+	switch (format) {
+	case WIN_COLOR_DEPTH_YCbCr422:
+	case WIN_COLOR_DEPTH_YUV422:
+		if (planar)
+			*planar = false;
+
+		return true;
+
+	case WIN_COLOR_DEPTH_YCbCr420P:
+	case WIN_COLOR_DEPTH_YUV420P:
+	case WIN_COLOR_DEPTH_YCbCr422P:
+	case WIN_COLOR_DEPTH_YUV422P:
+	case WIN_COLOR_DEPTH_YCbCr422R:
+	case WIN_COLOR_DEPTH_YUV422R:
+	case WIN_COLOR_DEPTH_YCbCr422RA:
+	case WIN_COLOR_DEPTH_YUV422RA:
+		if (planar)
+			*planar = true;
+
+		return true;
+	}
+
+	return false;
+}
+
+static inline u32 compute_dda_inc(unsigned int in, unsigned int out, bool v,
+				  unsigned int bpp)
+{
+	fixed20_12 outf = dfixed_init(out);
+	fixed20_12 inf = dfixed_init(in);
+	u32 dda_inc;
+	int max;
+
+	if (v)
+		max = 15;
+	else {
+		switch (bpp) {
+		case 2:
+			max = 8;
+			break;
+
+		default:
+			WARN_ON_ONCE(1);
+			/* fallthrough */
+		case 4:
+			max = 4;
+			break;
+		}
+	}
+
+	outf.full = max_t(u32, outf.full - dfixed_const(1), dfixed_const(1));
+	inf.full -= dfixed_const(1);
+
+	dda_inc = dfixed_div(inf, outf);
+	dda_inc = min_t(u32, dda_inc, dfixed_const(max));
+
+	return dda_inc;
+}
+
+static inline u32 compute_initial_dda(unsigned int in)
+{
+	fixed20_12 inf = dfixed_init(in);
+	return dfixed_frac(inf);
+}
+
+static int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
+				 const struct tegra_dc_window *window)
+{
+	unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp;
+	unsigned long value;
+	bool yuv, planar;
+
+	/*
+	 * For YUV planar modes, the number of bytes per pixel takes into
+	 * account only the luma component and therefore is 1.
+	 */
+	yuv = tegra_dc_format_is_yuv(window->format, &planar);
+	if (!yuv)
+		bpp = window->bits_per_pixel / 8;
+	else
+		bpp = planar ? 1 : 2;
+
+	value = WINDOW_A_SELECT << index;
+	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
+
+	tegra_dc_writel(dc, window->format, DC_WIN_COLOR_DEPTH);
+	tegra_dc_writel(dc, window->swap, DC_WIN_BYTE_SWAP);
+
+	value = V_POSITION(window->dst.y) | H_POSITION(window->dst.x);
+	tegra_dc_writel(dc, value, DC_WIN_POSITION);
+
+	value = V_SIZE(window->dst.h) | H_SIZE(window->dst.w);
+	tegra_dc_writel(dc, value, DC_WIN_SIZE);
+
+	h_offset = window->src.x * bpp;
+	v_offset = window->src.y;
+	h_size = window->src.w * bpp;
+	v_size = window->src.h;
+
+	value = V_PRESCALED_SIZE(v_size) | H_PRESCALED_SIZE(h_size);
+	tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE);
+
+	/*
+	 * For DDA computations the number of bytes per pixel for YUV planar
+	 * modes needs to take into account all Y, U and V components.
+	 */
+	if (yuv && planar)
+		bpp = 2;
+
+	h_dda = compute_dda_inc(window->src.w, window->dst.w, false, bpp);
+	v_dda = compute_dda_inc(window->src.h, window->dst.h, true, bpp);
+
+	value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda);
+	tegra_dc_writel(dc, value, DC_WIN_DDA_INC);
+
+	h_dda = compute_initial_dda(window->src.x);
+	v_dda = compute_initial_dda(window->src.y);
+
+	tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA);
+	tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA);
+
+	tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE);
+	tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE);
+
+	tegra_dc_writel(dc, window->base[0], DC_WINBUF_START_ADDR);
+
+	if (yuv && planar) {
+		tegra_dc_writel(dc, window->base[1], DC_WINBUF_START_ADDR_U);
+		tegra_dc_writel(dc, window->base[2], DC_WINBUF_START_ADDR_V);
+		value = window->stride[1] << 16 | window->stride[0];
+		tegra_dc_writel(dc, value, DC_WIN_LINE_STRIDE);
+	} else {
+		tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE);
+	}
+
+	if (window->bottom_up)
+		v_offset += window->src.h - 1;
+
+	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
+	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
+
+	if (window->tiled) {
+		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
+			DC_WIN_BUFFER_ADDR_MODE_TILE;
+	} else {
+		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
+			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
+
+	value = WIN_ENABLE;
+
+	if (yuv) {
+		/* setup default colorspace conversion coefficients */
+		tegra_dc_writel(dc, 0x00f0, DC_WIN_CSC_YOF);
+		tegra_dc_writel(dc, 0x012a, DC_WIN_CSC_KYRGB);
+		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KUR);
+		tegra_dc_writel(dc, 0x0198, DC_WIN_CSC_KVR);
+		tegra_dc_writel(dc, 0x039b, DC_WIN_CSC_KUG);
+		tegra_dc_writel(dc, 0x032f, DC_WIN_CSC_KVG);
+		tegra_dc_writel(dc, 0x0204, DC_WIN_CSC_KUB);
+		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KVB);
+
+		value |= CSC_ENABLE;
+	} else if (window->bits_per_pixel < 24) {
+		value |= COLOR_EXPAND;
+	}
+
+	if (window->bottom_up)
+		value |= V_DIRECTION;
+
+	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+
+	/*
+	 * Disable blending and assume Window A is the bottom-most window,
+	 * Window C is the top-most window and Window B is in the middle.
+	 */
+	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_NOKEY);
+	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_1WIN);
+
+	switch (index) {
+	case 0:
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
+		break;
+
+	case 1:
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
+		break;
+
+	case 2:
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_3WIN_XY);
+		break;
+	}
+
+	tegra_dc_writel(dc, WIN_A_UPDATE << index, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, WIN_A_ACT_REQ << index, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
 static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 			      struct drm_framebuffer *fb, int crtc_x,
 			      int crtc_y, unsigned int crtc_w,
@@ -49,7 +298,7 @@ static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	window.dst.y = crtc_y;
 	window.dst.w = crtc_w;
 	window.dst.h = crtc_h;
-	window.format = tegra_dc_format(fb->pixel_format);
+	window.format = tegra_dc_format(fb->pixel_format, &window.swap);
 	window.bits_per_pixel = fb->bits_per_pixel;
 	window.bottom_up = tegra_fb_is_bottom_up(fb);
 	window.tiled = tegra_fb_is_tiled(fb);
@@ -117,6 +366,7 @@ static const uint32_t plane_formats[] = {
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_UYVY,
+	DRM_FORMAT_YUYV,
 	DRM_FORMAT_YUV420,
 	DRM_FORMAT_YUV422,
 };
@@ -150,9 +400,9 @@ static int tegra_dc_add_planes(struct drm_device *drm, struct tegra_dc *dc)
 static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 			     struct drm_framebuffer *fb)
 {
-	unsigned int format = tegra_dc_format(fb->pixel_format);
 	struct tegra_bo *bo = tegra_fb_get_plane(fb, 0);
 	unsigned int h_offset = 0, v_offset = 0;
+	unsigned int format, swap;
 	unsigned long value;
 
 	tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER);
@@ -162,7 +412,10 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 
 	tegra_dc_writel(dc, bo->paddr + value, DC_WINBUF_START_ADDR);
 	tegra_dc_writel(dc, fb->pitches[0], DC_WIN_LINE_STRIDE);
+
+	format = tegra_dc_format(fb->pixel_format, &swap);
 	tegra_dc_writel(dc, format, DC_WIN_COLOR_DEPTH);
+	tegra_dc_writel(dc, swap, DC_WIN_BYTE_SWAP);
 
 	if (tegra_fb_is_tiled(fb)) {
 		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
@@ -177,13 +430,13 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 	/* make sure bottom-up buffers are properly displayed */
 	if (tegra_fb_is_bottom_up(fb)) {
 		value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
-		value |= INVERT_V;
+		value |= V_DIRECTION;
 		tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
 
 		v_offset += fb->height - 1;
 	} else {
 		value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
-		value &= ~INVERT_V;
+		value &= ~V_DIRECTION;
 		tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
 	}
 
@@ -225,6 +478,109 @@ void tegra_dc_disable_vblank(struct tegra_dc *dc)
 	spin_unlock_irqrestore(&dc->lock, flags);
 }
 
+static int tegra_dc_cursor_set2(struct drm_crtc *crtc, struct drm_file *file,
+				uint32_t handle, uint32_t width,
+				uint32_t height, int32_t hot_x, int32_t hot_y)
+{
+	unsigned long value = CURSOR_CLIP_DISPLAY;
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	struct drm_gem_object *gem;
+	struct tegra_bo *bo = NULL;
+
+	if (!dc->soc->supports_cursor)
+		return -ENXIO;
+
+	if (width != height)
+		return -EINVAL;
+
+	switch (width) {
+	case 32:
+		value |= CURSOR_SIZE_32x32;
+		break;
+
+	case 64:
+		value |= CURSOR_SIZE_64x64;
+		break;
+
+	case 128:
+		value |= CURSOR_SIZE_128x128;
+
+	case 256:
+		value |= CURSOR_SIZE_256x256;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (handle) {
+		gem = drm_gem_object_lookup(crtc->dev, file, handle);
+		if (!gem)
+			return -ENOENT;
+
+		bo = to_tegra_bo(gem);
+	}
+
+	if (bo) {
+		unsigned long addr = (bo->paddr & 0xfffffc00) >> 10;
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		unsigned long high = (bo->paddr & 0xfffffffc) >> 32;
+#endif
+
+		tegra_dc_writel(dc, value | addr, DC_DISP_CURSOR_START_ADDR);
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		tegra_dc_writel(dc, high, DC_DISP_CURSOR_START_ADDR_HI);
+#endif
+
+		value = tegra_dc_readl(dc, DC_DISP_DISP_WIN_OPTIONS);
+		value |= CURSOR_ENABLE;
+		tegra_dc_writel(dc, value, DC_DISP_DISP_WIN_OPTIONS);
+
+		value = tegra_dc_readl(dc, DC_DISP_BLEND_CURSOR_CONTROL);
+		value &= ~CURSOR_DST_BLEND_MASK;
+		value &= ~CURSOR_SRC_BLEND_MASK;
+		value |= CURSOR_MODE_NORMAL;
+		value |= CURSOR_DST_BLEND_NEG_K1_TIMES_SRC;
+		value |= CURSOR_SRC_BLEND_K1_TIMES_SRC;
+		value |= CURSOR_ALPHA;
+		tegra_dc_writel(dc, value, DC_DISP_BLEND_CURSOR_CONTROL);
+	} else {
+		value = tegra_dc_readl(dc, DC_DISP_DISP_WIN_OPTIONS);
+		value &= ~CURSOR_ENABLE;
+		tegra_dc_writel(dc, value, DC_DISP_DISP_WIN_OPTIONS);
+	}
+
+	tegra_dc_writel(dc, CURSOR_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, CURSOR_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	tegra_dc_writel(dc, GENERAL_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
+static int tegra_dc_cursor_move(struct drm_crtc *crtc, int x, int y)
+{
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	unsigned long value;
+
+	if (!dc->soc->supports_cursor)
+		return -ENXIO;
+
+	value = ((y & 0x3fff) << 16) | (x & 0x3fff);
+	tegra_dc_writel(dc, value, DC_DISP_CURSOR_POSITION);
+
+	tegra_dc_writel(dc, CURSOR_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, CURSOR_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	/* XXX: only required on generations earlier than Tegra124? */
+	tegra_dc_writel(dc, GENERAL_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
 static void tegra_dc_finish_page_flip(struct tegra_dc *dc)
 {
 	struct drm_device *drm = dc->base.dev;
@@ -301,6 +657,8 @@ static void tegra_dc_destroy(struct drm_crtc *crtc)
 }
 
 static const struct drm_crtc_funcs tegra_crtc_funcs = {
+	.cursor_set2 = tegra_dc_cursor_set2,
+	.cursor_move = tegra_dc_cursor_move,
 	.page_flip = tegra_dc_page_flip,
 	.set_config = drm_crtc_helper_set_config,
 	.destroy = tegra_dc_destroy,
@@ -334,52 +692,11 @@ static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc,
 	return true;
 }
 
-static inline u32 compute_dda_inc(unsigned int in, unsigned int out, bool v,
-				  unsigned int bpp)
-{
-	fixed20_12 outf = dfixed_init(out);
-	fixed20_12 inf = dfixed_init(in);
-	u32 dda_inc;
-	int max;
-
-	if (v)
-		max = 15;
-	else {
-		switch (bpp) {
-		case 2:
-			max = 8;
-			break;
-
-		default:
-			WARN_ON_ONCE(1);
-			/* fallthrough */
-		case 4:
-			max = 4;
-			break;
-		}
-	}
-
-	outf.full = max_t(u32, outf.full - dfixed_const(1), dfixed_const(1));
-	inf.full -= dfixed_const(1);
-
-	dda_inc = dfixed_div(inf, outf);
-	dda_inc = min_t(u32, dda_inc, dfixed_const(max));
-
-	return dda_inc;
-}
-
-static inline u32 compute_initial_dda(unsigned int in)
-{
-	fixed20_12 inf = dfixed_init(in);
-	return dfixed_frac(inf);
-}
-
 static int tegra_dc_set_timings(struct tegra_dc *dc,
 				struct drm_display_mode *mode)
 {
-	/* TODO: For HDMI compliance, h & v ref_to_sync should be set to 1 */
-	unsigned int h_ref_to_sync = 0;
-	unsigned int v_ref_to_sync = 0;
+	unsigned int h_ref_to_sync = 1;
+	unsigned int v_ref_to_sync = 1;
 	unsigned long value;
 
 	tegra_dc_writel(dc, 0x0, DC_DISP_DISP_TIMING_OPTIONS);
@@ -406,13 +723,14 @@ static int tegra_dc_set_timings(struct tegra_dc *dc,
 }
 
 static int tegra_crtc_setup_clk(struct drm_crtc *crtc,
-				struct drm_display_mode *mode,
-				unsigned long *div)
+				struct drm_display_mode *mode)
 {
-	unsigned long pclk = mode->clock * 1000, rate;
+	unsigned long pclk = mode->clock * 1000;
 	struct tegra_dc *dc = to_tegra_dc(crtc);
 	struct tegra_output *output = NULL;
 	struct drm_encoder *encoder;
+	unsigned int div;
+	u32 value;
 	long err;
 
 	list_for_each_entry(encoder, &crtc->dev->mode_config.encoder_list, head)
@@ -425,221 +743,23 @@ static int tegra_crtc_setup_clk(struct drm_crtc *crtc,
 		return -ENODEV;
 
 	/*
-	 * This assumes that the display controller will divide its parent
-	 * clock by 2 to generate the pixel clock.
+	 * This assumes that the parent clock is pll_d_out0 or pll_d2_out
+	 * respectively, each of which divides the base pll_d by 2.
 	 */
-	err = tegra_output_setup_clock(output, dc->clk, pclk * 2);
+	err = tegra_output_setup_clock(output, dc->clk, pclk, &div);
 	if (err < 0) {
 		dev_err(dc->dev, "failed to setup clock: %ld\n", err);
 		return err;
 	}
 
-	rate = clk_get_rate(dc->clk);
-	*div = (rate * 2 / pclk) - 2;
-
-	DRM_DEBUG_KMS("rate: %lu, div: %lu\n", rate, *div);
-
-	return 0;
-}
-
-static bool tegra_dc_format_is_yuv(unsigned int format, bool *planar)
-{
-	switch (format) {
-	case WIN_COLOR_DEPTH_YCbCr422:
-	case WIN_COLOR_DEPTH_YUV422:
-		if (planar)
-			*planar = false;
-
-		return true;
-
-	case WIN_COLOR_DEPTH_YCbCr420P:
-	case WIN_COLOR_DEPTH_YUV420P:
-	case WIN_COLOR_DEPTH_YCbCr422P:
-	case WIN_COLOR_DEPTH_YUV422P:
-	case WIN_COLOR_DEPTH_YCbCr422R:
-	case WIN_COLOR_DEPTH_YUV422R:
-	case WIN_COLOR_DEPTH_YCbCr422RA:
-	case WIN_COLOR_DEPTH_YUV422RA:
-		if (planar)
-			*planar = true;
-
-		return true;
-	}
-
-	return false;
-}
-
-int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
-			  const struct tegra_dc_window *window)
-{
-	unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp;
-	unsigned long value;
-	bool yuv, planar;
-
-	/*
-	 * For YUV planar modes, the number of bytes per pixel takes into
-	 * account only the luma component and therefore is 1.
-	 */
-	yuv = tegra_dc_format_is_yuv(window->format, &planar);
-	if (!yuv)
-		bpp = window->bits_per_pixel / 8;
-	else
-		bpp = planar ? 1 : 2;
-
-	value = WINDOW_A_SELECT << index;
-	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
-
-	tegra_dc_writel(dc, window->format, DC_WIN_COLOR_DEPTH);
-	tegra_dc_writel(dc, 0, DC_WIN_BYTE_SWAP);
-
-	value = V_POSITION(window->dst.y) | H_POSITION(window->dst.x);
-	tegra_dc_writel(dc, value, DC_WIN_POSITION);
-
-	value = V_SIZE(window->dst.h) | H_SIZE(window->dst.w);
-	tegra_dc_writel(dc, value, DC_WIN_SIZE);
-
-	h_offset = window->src.x * bpp;
-	v_offset = window->src.y;
-	h_size = window->src.w * bpp;
-	v_size = window->src.h;
-
-	value = V_PRESCALED_SIZE(v_size) | H_PRESCALED_SIZE(h_size);
-	tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE);
-
-	/*
-	 * For DDA computations the number of bytes per pixel for YUV planar
-	 * modes needs to take into account all Y, U and V components.
-	 */
-	if (yuv && planar)
-		bpp = 2;
-
-	h_dda = compute_dda_inc(window->src.w, window->dst.w, false, bpp);
-	v_dda = compute_dda_inc(window->src.h, window->dst.h, true, bpp);
-
-	value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda);
-	tegra_dc_writel(dc, value, DC_WIN_DDA_INC);
-
-	h_dda = compute_initial_dda(window->src.x);
-	v_dda = compute_initial_dda(window->src.y);
-
-	tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA);
-	tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA);
-
-	tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE);
-	tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE);
-
-	tegra_dc_writel(dc, window->base[0], DC_WINBUF_START_ADDR);
-
-	if (yuv && planar) {
-		tegra_dc_writel(dc, window->base[1], DC_WINBUF_START_ADDR_U);
-		tegra_dc_writel(dc, window->base[2], DC_WINBUF_START_ADDR_V);
-		value = window->stride[1] << 16 | window->stride[0];
-		tegra_dc_writel(dc, value, DC_WIN_LINE_STRIDE);
-	} else {
-		tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE);
-	}
-
-	if (window->bottom_up)
-		v_offset += window->src.h - 1;
+	DRM_DEBUG_KMS("rate: %lu, div: %u\n", clk_get_rate(dc->clk), div);
 
-	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
-	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
-
-	if (window->tiled) {
-		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
-			DC_WIN_BUFFER_ADDR_MODE_TILE;
-	} else {
-		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
-			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
-	}
-
-	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
-
-	value = WIN_ENABLE;
-
-	if (yuv) {
-		/* setup default colorspace conversion coefficients */
-		tegra_dc_writel(dc, 0x00f0, DC_WIN_CSC_YOF);
-		tegra_dc_writel(dc, 0x012a, DC_WIN_CSC_KYRGB);
-		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KUR);
-		tegra_dc_writel(dc, 0x0198, DC_WIN_CSC_KVR);
-		tegra_dc_writel(dc, 0x039b, DC_WIN_CSC_KUG);
-		tegra_dc_writel(dc, 0x032f, DC_WIN_CSC_KVG);
-		tegra_dc_writel(dc, 0x0204, DC_WIN_CSC_KUB);
-		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KVB);
-
-		value |= CSC_ENABLE;
-	} else if (window->bits_per_pixel < 24) {
-		value |= COLOR_EXPAND;
-	}
-
-	if (window->bottom_up)
-		value |= INVERT_V;
-
-	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
-
-	/*
-	 * Disable blending and assume Window A is the bottom-most window,
-	 * Window C is the top-most window and Window B is in the middle.
-	 */
-	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_NOKEY);
-	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_1WIN);
-
-	switch (index) {
-	case 0:
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_X);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
-		break;
-
-	case 1:
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
-		break;
-
-	case 2:
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_Y);
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_3WIN_XY);
-		break;
-	}
-
-	tegra_dc_writel(dc, WIN_A_UPDATE << index, DC_CMD_STATE_CONTROL);
-	tegra_dc_writel(dc, WIN_A_ACT_REQ << index, DC_CMD_STATE_CONTROL);
+	value = SHIFT_CLK_DIVIDER(div) | PIXEL_CLK_DIVIDER_PCD1;
+	tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL);
 
 	return 0;
 }
 
-unsigned int tegra_dc_format(uint32_t format)
-{
-	switch (format) {
-	case DRM_FORMAT_XBGR8888:
-		return WIN_COLOR_DEPTH_R8G8B8A8;
-
-	case DRM_FORMAT_XRGB8888:
-		return WIN_COLOR_DEPTH_B8G8R8A8;
-
-	case DRM_FORMAT_RGB565:
-		return WIN_COLOR_DEPTH_B5G6R5;
-
-	case DRM_FORMAT_UYVY:
-		return WIN_COLOR_DEPTH_YCbCr422;
-
-	case DRM_FORMAT_YUV420:
-		return WIN_COLOR_DEPTH_YCbCr420P;
-
-	case DRM_FORMAT_YUV422:
-		return WIN_COLOR_DEPTH_YCbCr422P;
-
-	default:
-		break;
-	}
-
-	WARN(1, "unsupported pixel format %u, using default\n", format);
-	return WIN_COLOR_DEPTH_B8G8R8A8;
-}
-
 static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 			       struct drm_display_mode *mode,
 			       struct drm_display_mode *adjusted,
@@ -648,12 +768,12 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 	struct tegra_bo *bo = tegra_fb_get_plane(crtc->primary->fb, 0);
 	struct tegra_dc *dc = to_tegra_dc(crtc);
 	struct tegra_dc_window window;
-	unsigned long div, value;
+	u32 value;
 	int err;
 
 	drm_vblank_pre_modeset(crtc->dev, dc->pipe);
 
-	err = tegra_crtc_setup_clk(crtc, mode, &div);
+	err = tegra_crtc_setup_clk(crtc, mode);
 	if (err) {
 		dev_err(dc->dev, "failed to setup clock for CRTC: %d\n", err);
 		return err;
@@ -669,9 +789,6 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 		tegra_dc_writel(dc, value, DC_DISP_INTERLACE_CONTROL);
 	}
 
-	value = SHIFT_CLK_DIVIDER(div) | PIXEL_CLK_DIVIDER_PCD1;
-	tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL);
-
 	/* setup window parameters */
 	memset(&window, 0, sizeof(window));
 	window.src.x = 0;
@@ -682,7 +799,8 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 	window.dst.y = 0;
 	window.dst.w = mode->hdisplay;
 	window.dst.h = mode->vdisplay;
-	window.format = tegra_dc_format(crtc->primary->fb->pixel_format);
+	window.format = tegra_dc_format(crtc->primary->fb->pixel_format,
+					&window.swap);
 	window.bits_per_pixel = crtc->primary->fb->bits_per_pixel;
 	window.stride[0] = crtc->primary->fb->pitches[0];
 	window.base[0] = bo->paddr;
@@ -728,10 +846,6 @@ static void tegra_crtc_prepare(struct drm_crtc *crtc)
 		WIN_A_OF_INT | WIN_B_OF_INT | WIN_C_OF_INT;
 	tegra_dc_writel(dc, value, DC_CMD_INT_POLARITY);
 
-	value = PW0_ENABLE | PW1_ENABLE | PW2_ENABLE | PW3_ENABLE |
-		PW4_ENABLE | PM0_ENABLE | PM1_ENABLE;
-	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_POWER_CONTROL);
-
 	/* initialize timer */
 	value = CURSOR_THRESHOLD(0) | WINDOW_A_THRESHOLD(0x20) |
 		WINDOW_B_THRESHOLD(0x20) | WINDOW_C_THRESHOLD(0x20);
@@ -991,6 +1105,8 @@ static int tegra_dc_show_regs(struct seq_file *s, void *data)
 	DUMP_REG(DC_DISP_SD_BL_CONTROL);
 	DUMP_REG(DC_DISP_SD_HW_K_VALUES);
 	DUMP_REG(DC_DISP_SD_MAN_K_VALUES);
+	DUMP_REG(DC_DISP_CURSOR_START_ADDR_HI);
+	DUMP_REG(DC_DISP_BLEND_CURSOR_CONTROL);
 	DUMP_REG(DC_WIN_WIN_OPTIONS);
 	DUMP_REG(DC_WIN_BYTE_SWAP);
 	DUMP_REG(DC_WIN_BUFFER_CONTROL);
@@ -1096,26 +1212,26 @@ static int tegra_dc_debugfs_exit(struct tegra_dc *dc)
 
 static int tegra_dc_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_dc *dc = host1x_client_to_dc(client);
 	int err;
 
-	drm_crtc_init(tegra->drm, &dc->base, &tegra_crtc_funcs);
+	drm_crtc_init(drm, &dc->base, &tegra_crtc_funcs);
 	drm_mode_crtc_set_gamma_size(&dc->base, 256);
 	drm_crtc_helper_add(&dc->base, &tegra_crtc_helper_funcs);
 
-	err = tegra_dc_rgb_init(tegra->drm, dc);
+	err = tegra_dc_rgb_init(drm, dc);
 	if (err < 0 && err != -ENODEV) {
 		dev_err(dc->dev, "failed to initialize RGB output: %d\n", err);
 		return err;
 	}
 
-	err = tegra_dc_add_planes(tegra->drm, dc);
+	err = tegra_dc_add_planes(drm, dc);
 	if (err < 0)
 		return err;
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_dc_debugfs_init(dc, tegra->drm->primary);
+		err = tegra_dc_debugfs_init(dc, drm->primary);
 		if (err < 0)
 			dev_err(dc->dev, "debugfs setup failed: %d\n", err);
 	}
@@ -1160,14 +1276,17 @@ static const struct host1x_client_ops dc_client_ops = {
 
 static const struct tegra_dc_soc_info tegra20_dc_soc_info = {
 	.supports_interlacing = false,
+	.supports_cursor = false,
 };
 
 static const struct tegra_dc_soc_info tegra30_dc_soc_info = {
 	.supports_interlacing = false,
+	.supports_cursor = false,
 };
 
 static const struct tegra_dc_soc_info tegra124_dc_soc_info = {
 	.supports_interlacing = true,
+	.supports_cursor = true,
 };
 
 static const struct of_device_id tegra_dc_of_match[] = {
diff --git a/drivers/gpu/drm/tegra/dc.h b/drivers/gpu/drm/tegra/dc.h
index c94101494826..78c5feff95d2 100644
--- a/drivers/gpu/drm/tegra/dc.h
+++ b/drivers/gpu/drm/tegra/dc.h
@@ -67,10 +67,12 @@
 #define WIN_A_ACT_REQ   (1 <<  1)
 #define WIN_B_ACT_REQ   (1 <<  2)
 #define WIN_C_ACT_REQ   (1 <<  3)
+#define CURSOR_ACT_REQ  (1 <<  7)
 #define GENERAL_UPDATE  (1 <<  8)
 #define WIN_A_UPDATE    (1 <<  9)
 #define WIN_B_UPDATE    (1 << 10)
 #define WIN_C_UPDATE    (1 << 11)
+#define CURSOR_UPDATE   (1 << 15)
 #define NC_HOST_TRIG    (1 << 24)
 
 #define DC_CMD_DISPLAY_WINDOW_HEADER		0x042
@@ -116,9 +118,10 @@
 #define DC_DISP_DISP_SIGNAL_OPTIONS1		0x401
 
 #define DC_DISP_DISP_WIN_OPTIONS		0x402
-#define HDMI_ENABLE (1 << 30)
-#define DSI_ENABLE  (1 << 29)
-#define SOR_ENABLE  (1 << 25)
+#define HDMI_ENABLE	(1 << 30)
+#define DSI_ENABLE	(1 << 29)
+#define SOR_ENABLE	(1 << 25)
+#define CURSOR_ENABLE	(1 << 16)
 
 #define DC_DISP_DISP_MEM_HIGH_PRIORITY		0x403
 #define CURSOR_THRESHOLD(x)   (((x) & 0x03) << 24)
@@ -266,6 +269,14 @@
 #define DC_DISP_CURSOR_BACKGROUND		0x43d
 
 #define DC_DISP_CURSOR_START_ADDR		0x43e
+#define CURSOR_CLIP_DISPLAY	(0 << 28)
+#define CURSOR_CLIP_WIN_A	(1 << 28)
+#define CURSOR_CLIP_WIN_B	(2 << 28)
+#define CURSOR_CLIP_WIN_C	(3 << 28)
+#define CURSOR_SIZE_32x32	(0 << 24)
+#define CURSOR_SIZE_64x64	(1 << 24)
+#define CURSOR_SIZE_128x128	(2 << 24)
+#define CURSOR_SIZE_256x256	(3 << 24)
 #define DC_DISP_CURSOR_START_ADDR_NS		0x43f
 
 #define DC_DISP_CURSOR_POSITION			0x440
@@ -302,6 +313,19 @@
 #define  INTERLACE_START  (1 << 1)
 #define  INTERLACE_ENABLE (1 << 0)
 
+#define DC_DISP_CURSOR_START_ADDR_HI		0x4ec
+#define DC_DISP_BLEND_CURSOR_CONTROL		0x4f1
+#define CURSOR_MODE_LEGACY			(0 << 24)
+#define CURSOR_MODE_NORMAL			(1 << 24)
+#define CURSOR_DST_BLEND_ZERO			(0 << 16)
+#define CURSOR_DST_BLEND_K1			(1 << 16)
+#define CURSOR_DST_BLEND_NEG_K1_TIMES_SRC	(2 << 16)
+#define CURSOR_DST_BLEND_MASK			(3 << 16)
+#define CURSOR_SRC_BLEND_K1			(0 << 8)
+#define CURSOR_SRC_BLEND_K1_TIMES_SRC		(1 << 8)
+#define CURSOR_SRC_BLEND_MASK			(3 << 8)
+#define CURSOR_ALPHA				0xff
+
 #define DC_WIN_CSC_YOF				0x611
 #define DC_WIN_CSC_KYRGB			0x612
 #define DC_WIN_CSC_KUR				0x613
@@ -312,7 +336,8 @@
 #define DC_WIN_CSC_KVB				0x618
 
 #define DC_WIN_WIN_OPTIONS			0x700
-#define INVERT_V     (1 <<  2)
+#define H_DIRECTION  (1 <<  0)
+#define V_DIRECTION  (1 <<  2)
 #define COLOR_EXPAND (1 <<  6)
 #define CSC_ENABLE   (1 << 18)
 #define WIN_ENABLE   (1 << 30)
diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c
index 005c19bd92df..3f132e356e9c 100644
--- a/drivers/gpu/drm/tegra/dpaux.c
+++ b/drivers/gpu/drm/tegra/dpaux.c
@@ -15,6 +15,7 @@
 #include <linux/platform_device.h>
 #include <linux/reset.h>
 #include <linux/regulator/consumer.h>
+#include <linux/workqueue.h>
 
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_panel.h>
@@ -41,6 +42,7 @@ struct tegra_dpaux {
 	struct regulator *vdd;
 
 	struct completion complete;
+	struct work_struct work;
 	struct list_head list;
 };
 
@@ -49,6 +51,11 @@ static inline struct tegra_dpaux *to_dpaux(struct drm_dp_aux *aux)
 	return container_of(aux, struct tegra_dpaux, aux);
 }
 
+static inline struct tegra_dpaux *work_to_dpaux(struct work_struct *work)
+{
+	return container_of(work, struct tegra_dpaux, work);
+}
+
 static inline unsigned long tegra_dpaux_readl(struct tegra_dpaux *dpaux,
 					      unsigned long offset)
 {
@@ -231,6 +238,14 @@ static ssize_t tegra_dpaux_transfer(struct drm_dp_aux *aux,
 	return ret;
 }
 
+static void tegra_dpaux_hotplug(struct work_struct *work)
+{
+	struct tegra_dpaux *dpaux = work_to_dpaux(work);
+
+	if (dpaux->output)
+		drm_helper_hpd_irq_event(dpaux->output->connector.dev);
+}
+
 static irqreturn_t tegra_dpaux_irq(int irq, void *data)
 {
 	struct tegra_dpaux *dpaux = data;
@@ -241,16 +256,8 @@ static irqreturn_t tegra_dpaux_irq(int irq, void *data)
 	value = tegra_dpaux_readl(dpaux, DPAUX_INTR_AUX);
 	tegra_dpaux_writel(dpaux, value, DPAUX_INTR_AUX);
 
-	if (value & DPAUX_INTR_PLUG_EVENT) {
-		if (dpaux->output) {
-			drm_helper_hpd_irq_event(dpaux->output->connector.dev);
-		}
-	}
-
-	if (value & DPAUX_INTR_UNPLUG_EVENT) {
-		if (dpaux->output)
-			drm_helper_hpd_irq_event(dpaux->output->connector.dev);
-	}
+	if (value & (DPAUX_INTR_PLUG_EVENT | DPAUX_INTR_UNPLUG_EVENT))
+		schedule_work(&dpaux->work);
 
 	if (value & DPAUX_INTR_IRQ_EVENT) {
 		/* TODO: handle this */
@@ -273,6 +280,7 @@ static int tegra_dpaux_probe(struct platform_device *pdev)
 	if (!dpaux)
 		return -ENOMEM;
 
+	INIT_WORK(&dpaux->work, tegra_dpaux_hotplug);
 	init_completion(&dpaux->complete);
 	INIT_LIST_HEAD(&dpaux->list);
 	dpaux->dev = &pdev->dev;
@@ -332,7 +340,7 @@ static int tegra_dpaux_probe(struct platform_device *pdev)
 	dpaux->aux.transfer = tegra_dpaux_transfer;
 	dpaux->aux.dev = &pdev->dev;
 
-	err = drm_dp_aux_register_i2c_bus(&dpaux->aux);
+	err = drm_dp_aux_register(&dpaux->aux);
 	if (err < 0)
 		return err;
 
@@ -355,12 +363,14 @@ static int tegra_dpaux_remove(struct platform_device *pdev)
 {
 	struct tegra_dpaux *dpaux = platform_get_drvdata(pdev);
 
-	drm_dp_aux_unregister_i2c_bus(&dpaux->aux);
+	drm_dp_aux_unregister(&dpaux->aux);
 
 	mutex_lock(&dpaux_lock);
 	list_del(&dpaux->list);
 	mutex_unlock(&dpaux_lock);
 
+	cancel_work_sync(&dpaux->work);
+
 	clk_disable_unprepare(dpaux->clk_parent);
 	reset_control_assert(dpaux->rst);
 	clk_disable_unprepare(dpaux->clk);
@@ -404,6 +414,7 @@ int tegra_dpaux_attach(struct tegra_dpaux *dpaux, struct tegra_output *output)
 	unsigned long timeout;
 	int err;
 
+	output->connector.polled = DRM_CONNECTOR_POLL_HPD;
 	dpaux->output = output;
 
 	err = regulator_enable(dpaux->vdd);
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 6f5b6e2f552e..3396f9f6a9f7 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -33,7 +33,6 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 	if (!tegra)
 		return -ENOMEM;
 
-	dev_set_drvdata(drm->dev, tegra);
 	mutex_init(&tegra->clients_lock);
 	INIT_LIST_HEAD(&tegra->clients);
 	drm->dev_private = tegra;
@@ -640,14 +639,40 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra,
 	return 0;
 }
 
-static int host1x_drm_probe(struct host1x_device *device)
+static int host1x_drm_probe(struct host1x_device *dev)
 {
-	return drm_host1x_init(&tegra_drm_driver, device);
+	struct drm_driver *driver = &tegra_drm_driver;
+	struct drm_device *drm;
+	int err;
+
+	drm = drm_dev_alloc(driver, &dev->dev);
+	if (!drm)
+		return -ENOMEM;
+
+	drm_dev_set_unique(drm, dev_name(&dev->dev));
+	dev_set_drvdata(&dev->dev, drm);
+
+	err = drm_dev_register(drm, 0);
+	if (err < 0)
+		goto unref;
+
+	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n", driver->name,
+		 driver->major, driver->minor, driver->patchlevel,
+		 driver->date, drm->primary->index);
+
+	return 0;
+
+unref:
+	drm_dev_unref(drm);
+	return err;
 }
 
-static int host1x_drm_remove(struct host1x_device *device)
+static int host1x_drm_remove(struct host1x_device *dev)
 {
-	drm_host1x_exit(&tegra_drm_driver, device);
+	struct drm_device *drm = dev_get_drvdata(&dev->dev);
+
+	drm_dev_unregister(drm);
+	drm_dev_unref(drm);
 
 	return 0;
 }
@@ -666,6 +691,7 @@ static const struct of_device_id host1x_drm_subdevs[] = {
 	{ .compatible = "nvidia,tegra114-gr3d", },
 	{ .compatible = "nvidia,tegra124-dc", },
 	{ .compatible = "nvidia,tegra124-sor", },
+	{ .compatible = "nvidia,tegra124-hdmi", },
 	{ /* sentinel */ }
 };
 
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index 126332c3ecbb..6b8fe9d86ed4 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -80,13 +80,13 @@ host1x_to_drm_client(struct host1x_client *client)
 	return container_of(client, struct tegra_drm_client, base);
 }
 
-extern int tegra_drm_register_client(struct tegra_drm *tegra,
-				     struct tegra_drm_client *client);
-extern int tegra_drm_unregister_client(struct tegra_drm *tegra,
-				       struct tegra_drm_client *client);
+int tegra_drm_register_client(struct tegra_drm *tegra,
+			      struct tegra_drm_client *client);
+int tegra_drm_unregister_client(struct tegra_drm *tegra,
+				struct tegra_drm_client *client);
 
-extern int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm);
-extern int tegra_drm_exit(struct tegra_drm *tegra);
+int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm);
+int tegra_drm_exit(struct tegra_drm *tegra);
 
 struct tegra_dc_soc_info;
 struct tegra_output;
@@ -156,6 +156,7 @@ struct tegra_dc_window {
 	} dst;
 	unsigned int bits_per_pixel;
 	unsigned int format;
+	unsigned int swap;
 	unsigned int stride[2];
 	unsigned long base[3];
 	bool bottom_up;
@@ -163,19 +164,15 @@ struct tegra_dc_window {
 };
 
 /* from dc.c */
-extern unsigned int tegra_dc_format(uint32_t format);
-extern int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
-				 const struct tegra_dc_window *window);
-extern void tegra_dc_enable_vblank(struct tegra_dc *dc);
-extern void tegra_dc_disable_vblank(struct tegra_dc *dc);
-extern void tegra_dc_cancel_page_flip(struct drm_crtc *crtc,
-				      struct drm_file *file);
+void tegra_dc_enable_vblank(struct tegra_dc *dc);
+void tegra_dc_disable_vblank(struct tegra_dc *dc);
+void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file);
 
 struct tegra_output_ops {
 	int (*enable)(struct tegra_output *output);
 	int (*disable)(struct tegra_output *output);
 	int (*setup_clock)(struct tegra_output *output, struct clk *clk,
-			   unsigned long pclk);
+			   unsigned long pclk, unsigned int *div);
 	int (*check_mode)(struct tegra_output *output,
 			  struct drm_display_mode *mode,
 			  enum drm_mode_status *status);
@@ -233,10 +230,11 @@ static inline int tegra_output_disable(struct tegra_output *output)
 }
 
 static inline int tegra_output_setup_clock(struct tegra_output *output,
-					   struct clk *clk, unsigned long pclk)
+					   struct clk *clk, unsigned long pclk,
+					   unsigned int *div)
 {
 	if (output && output->ops && output->ops->setup_clock)
-		return output->ops->setup_clock(output, clk, pclk);
+		return output->ops->setup_clock(output, clk, pclk, div);
 
 	return output ? -ENOSYS : -EINVAL;
 }
@@ -251,27 +249,21 @@ static inline int tegra_output_check_mode(struct tegra_output *output,
 	return output ? -ENOSYS : -EINVAL;
 }
 
-/* from bus.c */
-int drm_host1x_init(struct drm_driver *driver, struct host1x_device *device);
-void drm_host1x_exit(struct drm_driver *driver, struct host1x_device *device);
-
 /* from rgb.c */
-extern int tegra_dc_rgb_probe(struct tegra_dc *dc);
-extern int tegra_dc_rgb_remove(struct tegra_dc *dc);
-extern int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc);
-extern int tegra_dc_rgb_exit(struct tegra_dc *dc);
+int tegra_dc_rgb_probe(struct tegra_dc *dc);
+int tegra_dc_rgb_remove(struct tegra_dc *dc);
+int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc);
+int tegra_dc_rgb_exit(struct tegra_dc *dc);
 
 /* from output.c */
-extern int tegra_output_probe(struct tegra_output *output);
-extern int tegra_output_remove(struct tegra_output *output);
-extern int tegra_output_init(struct drm_device *drm, struct tegra_output *output);
-extern int tegra_output_exit(struct tegra_output *output);
+int tegra_output_probe(struct tegra_output *output);
+int tegra_output_remove(struct tegra_output *output);
+int tegra_output_init(struct drm_device *drm, struct tegra_output *output);
+int tegra_output_exit(struct tegra_output *output);
 
 /* from dpaux.c */
-
 struct tegra_dpaux;
 struct drm_dp_link;
-struct drm_dp_aux;
 
 struct tegra_dpaux *tegra_dpaux_find_by_of_node(struct device_node *np);
 enum drm_connector_status tegra_dpaux_detect(struct tegra_dpaux *dpaux);
@@ -288,10 +280,10 @@ struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer,
 				    unsigned int index);
 bool tegra_fb_is_bottom_up(struct drm_framebuffer *framebuffer);
 bool tegra_fb_is_tiled(struct drm_framebuffer *framebuffer);
-extern int tegra_drm_fb_init(struct drm_device *drm);
-extern void tegra_drm_fb_exit(struct drm_device *drm);
+int tegra_drm_fb_init(struct drm_device *drm);
+void tegra_drm_fb_exit(struct drm_device *drm);
 #ifdef CONFIG_DRM_TEGRA_FBDEV
-extern void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev);
+void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev);
 #endif
 
 extern struct platform_driver tegra_dc_driver;
diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c
index 0e599f0417c0..bd56f2affa78 100644
--- a/drivers/gpu/drm/tegra/dsi.c
+++ b/drivers/gpu/drm/tegra/dsi.c
@@ -14,6 +14,8 @@
 #include <linux/platform_device.h>
 #include <linux/reset.h>
 
+#include <linux/regulator/consumer.h>
+
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
 
@@ -43,11 +45,15 @@ struct tegra_dsi {
 	struct drm_minor *minor;
 	struct dentry *debugfs;
 
+	unsigned long flags;
 	enum mipi_dsi_pixel_format format;
 	unsigned int lanes;
 
 	struct tegra_mipi_device *mipi;
 	struct mipi_dsi_host host;
+
+	struct regulator *vdd;
+	bool enabled;
 };
 
 static inline struct tegra_dsi *
@@ -244,8 +250,10 @@ static int tegra_dsi_debugfs_exit(struct tegra_dsi *dsi)
 #define PKT_LP		(1 << 30)
 #define NUM_PKT_SEQ	12
 
-/* non-burst mode with sync-end */
-static const u32 pkt_seq_vnb_syne[NUM_PKT_SEQ] = {
+/*
+ * non-burst mode with sync pulses
+ */
+static const u32 pkt_seq_video_non_burst_sync_pulses[NUM_PKT_SEQ] = {
 	[ 0] = PKT_ID0(MIPI_DSI_V_SYNC_START) | PKT_LEN0(0) |
 	       PKT_ID1(MIPI_DSI_BLANKING_PACKET) | PKT_LEN1(1) |
 	       PKT_ID2(MIPI_DSI_H_SYNC_END) | PKT_LEN2(0) |
@@ -280,6 +288,36 @@ static const u32 pkt_seq_vnb_syne[NUM_PKT_SEQ] = {
 	       PKT_ID2(MIPI_DSI_BLANKING_PACKET) | PKT_LEN2(4),
 };
 
+/*
+ * non-burst mode with sync events
+ */
+static const u32 pkt_seq_video_non_burst_sync_events[NUM_PKT_SEQ] = {
+	[ 0] = PKT_ID0(MIPI_DSI_V_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 1] = 0,
+	[ 2] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 3] = 0,
+	[ 4] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 5] = 0,
+	[ 6] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_BLANKING_PACKET) | PKT_LEN1(2) |
+	       PKT_ID2(MIPI_DSI_PACKED_PIXEL_STREAM_24) | PKT_LEN2(3),
+	[ 7] = PKT_ID0(MIPI_DSI_BLANKING_PACKET) | PKT_LEN0(4),
+	[ 8] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 9] = 0,
+	[10] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_BLANKING_PACKET) | PKT_LEN1(2) |
+	       PKT_ID2(MIPI_DSI_PACKED_PIXEL_STREAM_24) | PKT_LEN2(3),
+	[11] = PKT_ID0(MIPI_DSI_BLANKING_PACKET) | PKT_LEN0(4),
+};
+
 static int tegra_dsi_set_phy_timing(struct tegra_dsi *dsi)
 {
 	struct mipi_dphy_timing timing;
@@ -361,28 +399,70 @@ static int tegra_dsi_get_muldiv(enum mipi_dsi_pixel_format format,
 	return 0;
 }
 
+static int tegra_dsi_get_format(enum mipi_dsi_pixel_format format,
+				enum tegra_dsi_format *fmt)
+{
+	switch (format) {
+	case MIPI_DSI_FMT_RGB888:
+		*fmt = TEGRA_DSI_FORMAT_24P;
+		break;
+
+	case MIPI_DSI_FMT_RGB666:
+		*fmt = TEGRA_DSI_FORMAT_18NP;
+		break;
+
+	case MIPI_DSI_FMT_RGB666_PACKED:
+		*fmt = TEGRA_DSI_FORMAT_18P;
+		break;
+
+	case MIPI_DSI_FMT_RGB565:
+		*fmt = TEGRA_DSI_FORMAT_16P;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int tegra_output_dsi_enable(struct tegra_output *output)
 {
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct drm_display_mode *mode = &dc->base.mode;
 	unsigned int hact, hsw, hbp, hfp, i, mul, div;
 	struct tegra_dsi *dsi = to_dsi(output);
-	/* FIXME: don't hardcode this */
-	const u32 *pkt_seq = pkt_seq_vnb_syne;
+	enum tegra_dsi_format format;
 	unsigned long value;
+	const u32 *pkt_seq;
 	int err;
 
+	if (dsi->enabled)
+		return 0;
+
+	if (dsi->flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE) {
+		DRM_DEBUG_KMS("Non-burst video mode with sync pulses\n");
+		pkt_seq = pkt_seq_video_non_burst_sync_pulses;
+	} else {
+		DRM_DEBUG_KMS("Non-burst video mode with sync events\n");
+		pkt_seq = pkt_seq_video_non_burst_sync_events;
+	}
+
 	err = tegra_dsi_get_muldiv(dsi->format, &mul, &div);
 	if (err < 0)
 		return err;
 
+	err = tegra_dsi_get_format(dsi->format, &format);
+	if (err < 0)
+		return err;
+
 	err = clk_enable(dsi->clk);
 	if (err < 0)
 		return err;
 
 	reset_control_deassert(dsi->rst);
 
-	value = DSI_CONTROL_CHANNEL(0) | DSI_CONTROL_FORMAT(dsi->format) |
+	value = DSI_CONTROL_CHANNEL(0) | DSI_CONTROL_FORMAT(format) |
 		DSI_CONTROL_LANES(dsi->lanes - 1) |
 		DSI_CONTROL_SOURCE(dc->pipe);
 	tegra_dsi_writel(dsi, value, DSI_CONTROL);
@@ -454,6 +534,8 @@ static int tegra_output_dsi_enable(struct tegra_output *output)
 	value |= DSI_POWER_CONTROL_ENABLE;
 	tegra_dsi_writel(dsi, value, DSI_POWER_CONTROL);
 
+	dsi->enabled = true;
+
 	return 0;
 }
 
@@ -463,9 +545,12 @@ static int tegra_output_dsi_disable(struct tegra_output *output)
 	struct tegra_dsi *dsi = to_dsi(output);
 	unsigned long value;
 
+	if (!dsi->enabled)
+		return 0;
+
 	/* disable DSI controller */
 	value = tegra_dsi_readl(dsi, DSI_POWER_CONTROL);
-	value &= DSI_POWER_CONTROL_ENABLE;
+	value &= ~DSI_POWER_CONTROL_ENABLE;
 	tegra_dsi_writel(dsi, value, DSI_POWER_CONTROL);
 
 	/*
@@ -492,30 +577,44 @@ static int tegra_output_dsi_disable(struct tegra_output *output)
 
 	clk_disable(dsi->clk);
 
+	dsi->enabled = false;
+
 	return 0;
 }
 
 static int tegra_output_dsi_setup_clock(struct tegra_output *output,
-					struct clk *clk, unsigned long pclk)
+					struct clk *clk, unsigned long pclk,
+					unsigned int *divp)
 {
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct drm_display_mode *mode = &dc->base.mode;
 	unsigned int timeout, mul, div, vrefresh;
 	struct tegra_dsi *dsi = to_dsi(output);
 	unsigned long bclk, plld, value;
-	struct clk *base;
 	int err;
 
 	err = tegra_dsi_get_muldiv(dsi->format, &mul, &div);
 	if (err < 0)
 		return err;
 
+	DRM_DEBUG_KMS("mul: %u, div: %u, lanes: %u\n", mul, div, dsi->lanes);
 	vrefresh = drm_mode_vrefresh(mode);
+	DRM_DEBUG_KMS("vrefresh: %u\n", vrefresh);
 
-	pclk = mode->htotal * mode->vtotal * vrefresh;
+	/* compute byte clock */
 	bclk = (pclk * mul) / (div * dsi->lanes);
-	plld = DIV_ROUND_UP(bclk * 8, 1000000);
-	pclk = (plld * 1000000) / 2;
+
+	/*
+	 * Compute bit clock and round up to the next MHz.
+	 */
+	plld = DIV_ROUND_UP(bclk * 8, 1000000) * 1000000;
+
+	/*
+	 * We divide the frequency by two here, but we make up for that by
+	 * setting the shift clock divider (further below) to half of the
+	 * correct value.
+	 */
+	plld /= 2;
 
 	err = clk_set_parent(clk, dsi->clk_parent);
 	if (err < 0) {
@@ -523,20 +622,26 @@ static int tegra_output_dsi_setup_clock(struct tegra_output *output,
 		return err;
 	}
 
-	base = clk_get_parent(dsi->clk_parent);
-
-	/*
-	 * This assumes that the parent clock is pll_d_out0 or pll_d2_out
-	 * respectively, each of which divides the base pll_d by 2.
-	 */
-	err = clk_set_rate(base, pclk * 2);
+	err = clk_set_rate(dsi->clk_parent, plld);
 	if (err < 0) {
 		dev_err(dsi->dev, "failed to set base clock rate to %lu Hz\n",
-			pclk * 2);
+			plld);
 		return err;
 	}
 
 	/*
+	 * Derive pixel clock from bit clock using the shift clock divider.
+	 * Note that this is only half of what we would expect, but we need
+	 * that to make up for the fact that we divided the bit clock by a
+	 * factor of two above.
+	 *
+	 * It's not clear exactly why this is necessary, but the display is
+	 * not working properly otherwise. Perhaps the PLLs cannot generate
+	 * frequencies sufficiently high.
+	 */
+	*divp = ((8 * mul) / (div * dsi->lanes)) - 2;
+
+	/*
 	 * XXX: Move the below somewhere else so that we don't need to have
 	 * access to the vrefresh in this function?
 	 */
@@ -610,61 +715,32 @@ static int tegra_dsi_pad_calibrate(struct tegra_dsi *dsi)
 
 static int tegra_dsi_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_dsi *dsi = host1x_client_to_dsi(client);
-	unsigned long value, i;
 	int err;
 
 	dsi->output.type = TEGRA_OUTPUT_DSI;
 	dsi->output.dev = client->dev;
 	dsi->output.ops = &dsi_ops;
 
-	err = tegra_output_init(tegra->drm, &dsi->output);
+	err = tegra_output_init(drm, &dsi->output);
 	if (err < 0) {
 		dev_err(client->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_dsi_debugfs_init(dsi, tegra->drm->primary);
+		err = tegra_dsi_debugfs_init(dsi, drm->primary);
 		if (err < 0)
 			dev_err(dsi->dev, "debugfs setup failed: %d\n", err);
 	}
 
-	/*
-	 * enable high-speed mode, checksum generation, ECC generation and
-	 * disable raw mode
-	 */
-	value = tegra_dsi_readl(dsi, DSI_HOST_CONTROL);
-	value |= DSI_HOST_CONTROL_ECC | DSI_HOST_CONTROL_CS |
-		 DSI_HOST_CONTROL_HS;
-	value &= ~DSI_HOST_CONTROL_RAW;
-	tegra_dsi_writel(dsi, value, DSI_HOST_CONTROL);
-
-	tegra_dsi_writel(dsi, 0, DSI_SOL_DELAY);
-	tegra_dsi_writel(dsi, 0, DSI_MAX_THRESHOLD);
-
-	tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_CONTROL);
-
-	for (i = 0; i < 8; i++) {
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_0 + i);
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_8 + i);
-	}
-
-	for (i = 0; i < 12; i++)
-		tegra_dsi_writel(dsi, 0, DSI_PKT_SEQ_0_LO + i);
-
-	tegra_dsi_writel(dsi, 0, DSI_DCS_CMDS);
-
 	err = tegra_dsi_pad_calibrate(dsi);
 	if (err < 0) {
 		dev_err(dsi->dev, "MIPI calibration failed: %d\n", err);
 		return err;
 	}
 
-	tegra_dsi_writel(dsi, DSI_POWER_CONTROL_ENABLE, DSI_POWER_CONTROL);
-	usleep_range(300, 1000);
-
 	return 0;
 }
 
@@ -715,66 +791,13 @@ static int tegra_dsi_setup_clocks(struct tegra_dsi *dsi)
 	return 0;
 }
 
-static void tegra_dsi_initialize(struct tegra_dsi *dsi)
-{
-	unsigned int i;
-
-	tegra_dsi_writel(dsi, 0, DSI_POWER_CONTROL);
-
-	tegra_dsi_writel(dsi, 0, DSI_INT_ENABLE);
-	tegra_dsi_writel(dsi, 0, DSI_INT_STATUS);
-	tegra_dsi_writel(dsi, 0, DSI_INT_MASK);
-
-	tegra_dsi_writel(dsi, 0, DSI_HOST_CONTROL);
-	tegra_dsi_writel(dsi, 0, DSI_CONTROL);
-
-	tegra_dsi_writel(dsi, 0, DSI_SOL_DELAY);
-	tegra_dsi_writel(dsi, 0, DSI_MAX_THRESHOLD);
-
-	tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_CONTROL);
-
-	for (i = 0; i < 8; i++) {
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_0 + i);
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_8 + i);
-	}
-
-	for (i = 0; i < 12; i++)
-		tegra_dsi_writel(dsi, 0, DSI_PKT_SEQ_0_LO + i);
-
-	tegra_dsi_writel(dsi, 0, DSI_DCS_CMDS);
-
-	for (i = 0; i < 4; i++)
-		tegra_dsi_writel(dsi, 0, DSI_PKT_LEN_0_1 + i);
-
-	tegra_dsi_writel(dsi, 0x00000000, DSI_PHY_TIMING_0);
-	tegra_dsi_writel(dsi, 0x00000000, DSI_PHY_TIMING_1);
-	tegra_dsi_writel(dsi, 0x000000ff, DSI_PHY_TIMING_2);
-	tegra_dsi_writel(dsi, 0x00000000, DSI_BTA_TIMING);
-
-	tegra_dsi_writel(dsi, 0, DSI_TIMEOUT_0);
-	tegra_dsi_writel(dsi, 0, DSI_TIMEOUT_1);
-	tegra_dsi_writel(dsi, 0, DSI_TO_TALLY);
-
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_0);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_CD);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CD_STATUS);
-	tegra_dsi_writel(dsi, 0, DSI_VIDEO_MODE_CONTROL);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_1);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_2);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_3);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_4);
-
-	tegra_dsi_writel(dsi, 0, DSI_GANGED_MODE_CONTROL);
-	tegra_dsi_writel(dsi, 0, DSI_GANGED_MODE_START);
-	tegra_dsi_writel(dsi, 0, DSI_GANGED_MODE_SIZE);
-}
-
 static int tegra_dsi_host_attach(struct mipi_dsi_host *host,
 				 struct mipi_dsi_device *device)
 {
 	struct tegra_dsi *dsi = host_to_tegra(host);
 	struct tegra_output *output = &dsi->output;
 
+	dsi->flags = device->mode_flags;
 	dsi->format = device->format;
 	dsi->lanes = device->lanes;
 
@@ -829,6 +852,7 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 	 * attaches to the DSI host, the parameters will be taken from
 	 * the attached device.
 	 */
+	dsi->flags = MIPI_DSI_MODE_VIDEO;
 	dsi->format = MIPI_DSI_FMT_RGB888;
 	dsi->lanes = 4;
 
@@ -872,6 +896,18 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 		return err;
 	}
 
+	dsi->vdd = devm_regulator_get(&pdev->dev, "avdd-dsi-csi");
+	if (IS_ERR(dsi->vdd)) {
+		dev_err(&pdev->dev, "cannot get VDD supply\n");
+		return PTR_ERR(dsi->vdd);
+	}
+
+	err = regulator_enable(dsi->vdd);
+	if (err < 0) {
+		dev_err(&pdev->dev, "cannot enable VDD supply\n");
+		return err;
+	}
+
 	err = tegra_dsi_setup_clocks(dsi);
 	if (err < 0) {
 		dev_err(&pdev->dev, "cannot setup clocks\n");
@@ -883,8 +919,6 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 	if (IS_ERR(dsi->regs))
 		return PTR_ERR(dsi->regs);
 
-	tegra_dsi_initialize(dsi);
-
 	dsi->mipi = tegra_mipi_request(&pdev->dev);
 	if (IS_ERR(dsi->mipi))
 		return PTR_ERR(dsi->mipi);
@@ -929,9 +963,11 @@ static int tegra_dsi_remove(struct platform_device *pdev)
 	mipi_dsi_host_unregister(&dsi->host);
 	tegra_mipi_free(dsi->mipi);
 
+	regulator_disable(dsi->vdd);
 	clk_disable_unprepare(dsi->clk_parent);
 	clk_disable_unprepare(dsi->clk_lp);
 	clk_disable_unprepare(dsi->clk);
+	reset_control_assert(dsi->rst);
 
 	err = tegra_output_remove(&dsi->output);
 	if (err < 0) {
diff --git a/drivers/gpu/drm/tegra/dsi.h b/drivers/gpu/drm/tegra/dsi.h
index 1db5cc24ea91..5ce610d08d77 100644
--- a/drivers/gpu/drm/tegra/dsi.h
+++ b/drivers/gpu/drm/tegra/dsi.h
@@ -117,4 +117,14 @@
 #define DSI_INIT_SEQ_DATA_14		0x5e
 #define DSI_INIT_SEQ_DATA_15		0x5f
 
+/*
+ * pixel format as used in the DSI_CONTROL_FORMAT field
+ */
+enum tegra_dsi_format {
+	TEGRA_DSI_FORMAT_16P,
+	TEGRA_DSI_FORMAT_18NP,
+	TEGRA_DSI_FORMAT_18P,
+	TEGRA_DSI_FORMAT_24P,
+};
+
 #endif
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index f7fca09d4921..9798a7080322 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -346,11 +346,8 @@ static void tegra_fbdev_free(struct tegra_fbdev *fbdev)
 
 void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev)
 {
-	if (fbdev) {
-		drm_modeset_lock_all(fbdev->base.dev);
-		drm_fb_helper_restore_fbdev_mode(&fbdev->base);
-		drm_modeset_unlock_all(fbdev->base.dev);
-	}
+	if (fbdev)
+		drm_fb_helper_restore_fbdev_mode_unlocked(&fbdev->base);
 }
 
 static void tegra_fb_output_poll_changed(struct drm_device *drm)
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index bcf9895cef9f..aa85b7b26f10 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -169,7 +169,8 @@ err:
 	return ERR_PTR(ret);
 }
 
-struct tegra_bo *tegra_bo_import(struct drm_device *drm, struct dma_buf *buf)
+static struct tegra_bo *tegra_bo_import(struct drm_device *drm,
+					struct dma_buf *buf)
 {
 	struct dma_buf_attachment *attach;
 	struct tegra_bo *bo;
diff --git a/drivers/gpu/drm/tegra/gr2d.c b/drivers/gpu/drm/tegra/gr2d.c
index 2c7ca748edf5..7c53941f2a9e 100644
--- a/drivers/gpu/drm/tegra/gr2d.c
+++ b/drivers/gpu/drm/tegra/gr2d.c
@@ -28,7 +28,7 @@ static inline struct gr2d *to_gr2d(struct tegra_drm_client *client)
 static int gr2d_init(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	unsigned long flags = HOST1X_SYNCPT_HAS_BASE;
 	struct gr2d *gr2d = to_gr2d(drm);
 
@@ -42,17 +42,17 @@ static int gr2d_init(struct host1x_client *client)
 		return -ENOMEM;
 	}
 
-	return tegra_drm_register_client(tegra, drm);
+	return tegra_drm_register_client(dev->dev_private, drm);
 }
 
 static int gr2d_exit(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	struct gr2d *gr2d = to_gr2d(drm);
 	int err;
 
-	err = tegra_drm_unregister_client(tegra, drm);
+	err = tegra_drm_unregister_client(dev->dev_private, drm);
 	if (err < 0)
 		return err;
 
diff --git a/drivers/gpu/drm/tegra/gr3d.c b/drivers/gpu/drm/tegra/gr3d.c
index 0cbb24b1ae04..30f5ba9bd6d0 100644
--- a/drivers/gpu/drm/tegra/gr3d.c
+++ b/drivers/gpu/drm/tegra/gr3d.c
@@ -37,7 +37,7 @@ static inline struct gr3d *to_gr3d(struct tegra_drm_client *client)
 static int gr3d_init(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	unsigned long flags = HOST1X_SYNCPT_HAS_BASE;
 	struct gr3d *gr3d = to_gr3d(drm);
 
@@ -51,17 +51,17 @@ static int gr3d_init(struct host1x_client *client)
 		return -ENOMEM;
 	}
 
-	return tegra_drm_register_client(tegra, drm);
+	return tegra_drm_register_client(dev->dev_private, drm);
 }
 
 static int gr3d_exit(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	struct gr3d *gr3d = to_gr3d(drm);
 	int err;
 
-	err = tegra_drm_unregister_client(tegra, drm);
+	err = tegra_drm_unregister_client(dev->dev_private, drm);
 	if (err < 0)
 		return err;
 
diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c
index 6928015d11a4..ba067bb767e3 100644
--- a/drivers/gpu/drm/tegra/hdmi.c
+++ b/drivers/gpu/drm/tegra/hdmi.c
@@ -42,8 +42,9 @@ struct tegra_hdmi {
 	struct device *dev;
 	bool enabled;
 
-	struct regulator *vdd;
+	struct regulator *hdmi;
 	struct regulator *pll;
+	struct regulator *vdd;
 
 	void __iomem *regs;
 	unsigned int irq;
@@ -317,6 +318,85 @@ static const struct tmds_config tegra114_tmds_config[] = {
 	},
 };
 
+static const struct tmds_config tegra124_tmds_config[] = {
+	{ /* 480p/576p / 25.2MHz/27MHz modes */
+		.pclk = 27000000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(0) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_LOADADJ(3) | SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_0_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_10_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 720p / 74.25MHz modes */
+		.pclk = 74250000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(1) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_PE_EN | SOR_PLL_LOADADJ(3) |
+			SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_15_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_10_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 1080p / 148.5MHz modes */
+		.pclk = 148500000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(3) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_PE_EN | SOR_PLL_LOADADJ(3) |
+			SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_10_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_12_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 225/297MHz modes */
+		.pclk = UINT_MAX,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(0xf) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_LOADADJ(3) | SOR_PLL_TMDS_TERMADJ(7)
+			| SOR_PLL_TMDS_TERM_ENABLE,
+		.pe_current = PE_CURRENT0(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_0_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_19_200_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_800_mA),
+	},
+};
+
 static const struct tegra_hdmi_audio_config *
 tegra_hdmi_get_audio_config(unsigned int audio_freq, unsigned int pclk)
 {
@@ -716,13 +796,9 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 		return err;
 	}
 
-	/*
-	 * This assumes that the display controller will divide its parent
-	 * clock by 2 to generate the pixel clock.
-	 */
-	err = tegra_output_setup_clock(output, hdmi->clk, pclk * 2);
+	err = regulator_enable(hdmi->vdd);
 	if (err < 0) {
-		dev_err(hdmi->dev, "failed to setup clock: %d\n", err);
+		dev_err(hdmi->dev, "failed to enable VDD regulator: %d\n", err);
 		return err;
 	}
 
@@ -730,7 +806,7 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	if (err < 0)
 		return err;
 
-	err = clk_enable(hdmi->clk);
+	err = clk_prepare_enable(hdmi->clk);
 	if (err < 0) {
 		dev_err(hdmi->dev, "failed to enable clock: %d\n", err);
 		return err;
@@ -740,6 +816,17 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	usleep_range(1000, 2000);
 	reset_control_deassert(hdmi->rst);
 
+	/* power up sequence */
+	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_PLL0);
+	value &= ~SOR_PLL_PDBG;
+	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_PLL0);
+
+	usleep_range(10, 20);
+
+	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_PLL0);
+	value &= ~SOR_PLL_PWR;
+	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_PLL0);
+
 	tegra_dc_writel(dc, VSYNC_H_POSITION(1),
 			DC_DISP_DISP_TIMING_OPTIONS);
 	tegra_dc_writel(dc, DITHER_CONTROL_DISABLE | BASE_COLOR_SIZE888,
@@ -838,9 +925,13 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_SEQ_INST(0));
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_SEQ_INST(8));
 
-	value = 0x1c800;
+	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_CSTM);
 	value &= ~SOR_CSTM_ROTCLK(~0);
 	value |= SOR_CSTM_ROTCLK(2);
+	value |= SOR_CSTM_PLLDIV;
+	value &= ~SOR_CSTM_LVDS_ENABLE;
+	value &= ~SOR_CSTM_MODE_MASK;
+	value |= SOR_CSTM_MODE_TMDS;
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_CSTM);
 
 	/* start SOR */
@@ -930,10 +1021,18 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 	 * sure it's only executed when the output is attached to one.
 	 */
 	if (dc) {
+		/*
+		 * XXX: We can't do this here because it causes HDMI to go
+		 * into an erroneous state with the result that HDMI won't
+		 * properly work once disabled. See also a similar symptom
+		 * for the SOR output.
+		 */
+		/*
 		value = tegra_dc_readl(dc, DC_CMD_DISPLAY_POWER_CONTROL);
 		value &= ~(PW0_ENABLE | PW1_ENABLE | PW2_ENABLE | PW3_ENABLE |
 			   PW4_ENABLE | PM0_ENABLE | PM1_ENABLE);
 		tegra_dc_writel(dc, value, DC_CMD_DISPLAY_POWER_CONTROL);
+		*/
 
 		value = tegra_dc_readl(dc, DC_CMD_DISPLAY_COMMAND);
 		value &= ~DISP_CTRL_MODE_MASK;
@@ -947,8 +1046,9 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 		tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL);
 	}
 
+	clk_disable_unprepare(hdmi->clk);
 	reset_control_assert(hdmi->rst);
-	clk_disable(hdmi->clk);
+	regulator_disable(hdmi->vdd);
 	regulator_disable(hdmi->pll);
 
 	hdmi->enabled = false;
@@ -957,10 +1057,10 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 }
 
 static int tegra_output_hdmi_setup_clock(struct tegra_output *output,
-					 struct clk *clk, unsigned long pclk)
+					 struct clk *clk, unsigned long pclk,
+					 unsigned int *div)
 {
 	struct tegra_hdmi *hdmi = to_hdmi(output);
-	struct clk *base;
 	int err;
 
 	err = clk_set_parent(clk, hdmi->clk_parent);
@@ -969,17 +1069,12 @@ static int tegra_output_hdmi_setup_clock(struct tegra_output *output,
 		return err;
 	}
 
-	base = clk_get_parent(hdmi->clk_parent);
-
-	/*
-	 * This assumes that the parent clock is pll_d_out0 or pll_d2_out
-	 * respectively, each of which divides the base pll_d by 2.
-	 */
-	err = clk_set_rate(base, pclk * 2);
+	err = clk_set_rate(hdmi->clk_parent, pclk);
 	if (err < 0)
-		dev_err(output->dev,
-			"failed to set base clock rate to %lu Hz\n",
-			pclk * 2);
+		dev_err(output->dev, "failed to set clock rate to %lu Hz\n",
+			pclk);
+
+	*div = 0;
 
 	return 0;
 }
@@ -1017,7 +1112,7 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data)
 	struct tegra_hdmi *hdmi = node->info_ent->data;
 	int err;
 
-	err = clk_enable(hdmi->clk);
+	err = clk_prepare_enable(hdmi->clk);
 	if (err)
 		return err;
 
@@ -1186,7 +1281,7 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data)
 
 #undef DUMP_REG
 
-	clk_disable(hdmi->clk);
+	clk_disable_unprepare(hdmi->clk);
 
 	return 0;
 }
@@ -1252,33 +1347,33 @@ static int tegra_hdmi_debugfs_exit(struct tegra_hdmi *hdmi)
 
 static int tegra_hdmi_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client);
 	int err;
 
-	err = regulator_enable(hdmi->vdd);
-	if (err < 0) {
-		dev_err(client->dev, "failed to enable VDD regulator: %d\n",
-			err);
-		return err;
-	}
-
 	hdmi->output.type = TEGRA_OUTPUT_HDMI;
 	hdmi->output.dev = client->dev;
 	hdmi->output.ops = &hdmi_ops;
 
-	err = tegra_output_init(tegra->drm, &hdmi->output);
+	err = tegra_output_init(drm, &hdmi->output);
 	if (err < 0) {
 		dev_err(client->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_hdmi_debugfs_init(hdmi, tegra->drm->primary);
+		err = tegra_hdmi_debugfs_init(hdmi, drm->primary);
 		if (err < 0)
 			dev_err(client->dev, "debugfs setup failed: %d\n", err);
 	}
 
+	err = regulator_enable(hdmi->hdmi);
+	if (err < 0) {
+		dev_err(client->dev, "failed to enable HDMI regulator: %d\n",
+			err);
+		return err;
+	}
+
 	return 0;
 }
 
@@ -1287,6 +1382,8 @@ static int tegra_hdmi_exit(struct host1x_client *client)
 	struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client);
 	int err;
 
+	regulator_disable(hdmi->hdmi);
+
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
 		err = tegra_hdmi_debugfs_exit(hdmi);
 		if (err < 0)
@@ -1306,8 +1403,6 @@ static int tegra_hdmi_exit(struct host1x_client *client)
 		return err;
 	}
 
-	regulator_disable(hdmi->vdd);
-
 	return 0;
 }
 
@@ -1340,7 +1435,16 @@ static const struct tegra_hdmi_config tegra114_hdmi_config = {
 	.has_sor_io_peak_current = true,
 };
 
+static const struct tegra_hdmi_config tegra124_hdmi_config = {
+	.tmds = tegra124_tmds_config,
+	.num_tmds = ARRAY_SIZE(tegra124_tmds_config),
+	.fuse_override_offset = HDMI_NV_PDISP_SOR_PAD_CTLS0,
+	.fuse_override_value = 1 << 31,
+	.has_sor_io_peak_current = true,
+};
+
 static const struct of_device_id tegra_hdmi_of_match[] = {
+	{ .compatible = "nvidia,tegra124-hdmi", .data = &tegra124_hdmi_config },
 	{ .compatible = "nvidia,tegra114-hdmi", .data = &tegra114_hdmi_config },
 	{ .compatible = "nvidia,tegra30-hdmi", .data = &tegra30_hdmi_config },
 	{ .compatible = "nvidia,tegra20-hdmi", .data = &tegra20_hdmi_config },
@@ -1381,28 +1485,20 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 		return PTR_ERR(hdmi->rst);
 	}
 
-	err = clk_prepare(hdmi->clk);
-	if (err < 0)
-		return err;
-
 	hdmi->clk_parent = devm_clk_get(&pdev->dev, "parent");
 	if (IS_ERR(hdmi->clk_parent))
 		return PTR_ERR(hdmi->clk_parent);
 
-	err = clk_prepare(hdmi->clk_parent);
-	if (err < 0)
-		return err;
-
 	err = clk_set_parent(hdmi->clk, hdmi->clk_parent);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to setup clocks: %d\n", err);
 		return err;
 	}
 
-	hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd");
-	if (IS_ERR(hdmi->vdd)) {
-		dev_err(&pdev->dev, "failed to get VDD regulator\n");
-		return PTR_ERR(hdmi->vdd);
+	hdmi->hdmi = devm_regulator_get(&pdev->dev, "hdmi");
+	if (IS_ERR(hdmi->hdmi)) {
+		dev_err(&pdev->dev, "failed to get HDMI regulator\n");
+		return PTR_ERR(hdmi->hdmi);
 	}
 
 	hdmi->pll = devm_regulator_get(&pdev->dev, "pll");
@@ -1411,6 +1507,12 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 		return PTR_ERR(hdmi->pll);
 	}
 
+	hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd");
+	if (IS_ERR(hdmi->vdd)) {
+		dev_err(&pdev->dev, "failed to get VDD regulator\n");
+		return PTR_ERR(hdmi->vdd);
+	}
+
 	hdmi->output.dev = &pdev->dev;
 
 	err = tegra_output_probe(&hdmi->output);
@@ -1462,8 +1564,8 @@ static int tegra_hdmi_remove(struct platform_device *pdev)
 		return err;
 	}
 
-	clk_unprepare(hdmi->clk_parent);
-	clk_unprepare(hdmi->clk);
+	clk_disable_unprepare(hdmi->clk_parent);
+	clk_disable_unprepare(hdmi->clk);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/tegra/hdmi.h b/drivers/gpu/drm/tegra/hdmi.h
index 0aebc485f7fa..919a19df4e1b 100644
--- a/drivers/gpu/drm/tegra/hdmi.h
+++ b/drivers/gpu/drm/tegra/hdmi.h
@@ -190,6 +190,11 @@
 
 #define HDMI_NV_PDISP_SOR_CSTM					0x5a
 #define SOR_CSTM_ROTCLK(x) (((x) & 0xf) << 24)
+#define SOR_CSTM_PLLDIV (1 << 21)
+#define SOR_CSTM_LVDS_ENABLE (1 << 16)
+#define SOR_CSTM_MODE_LVDS (0 << 12)
+#define SOR_CSTM_MODE_TMDS (1 << 12)
+#define SOR_CSTM_MODE_MASK (3 << 12)
 
 #define HDMI_NV_PDISP_SOR_LVDS					0x5b
 #define HDMI_NV_PDISP_SOR_CRCA					0x5c
diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index 0266fb40479e..d6af9be48f42 100644
--- a/drivers/gpu/drm/tegra/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -159,11 +159,38 @@ static int tegra_output_rgb_disable(struct tegra_output *output)
 }
 
 static int tegra_output_rgb_setup_clock(struct tegra_output *output,
-					struct clk *clk, unsigned long pclk)
+					struct clk *clk, unsigned long pclk,
+					unsigned int *div)
 {
 	struct tegra_rgb *rgb = to_rgb(output);
+	int err;
+
+	err = clk_set_parent(clk, rgb->clk_parent);
+	if (err < 0) {
+		dev_err(output->dev, "failed to set parent: %d\n", err);
+		return err;
+	}
 
-	return clk_set_parent(clk, rgb->clk_parent);
+	/*
+	 * We may not want to change the frequency of the parent clock, since
+	 * it may be a parent for other peripherals. This is due to the fact
+	 * that on Tegra20 there's only a single clock dedicated to display
+	 * (pll_d_out0), whereas later generations have a second one that can
+	 * be used to independently drive a second output (pll_d2_out0).
+	 *
+	 * As a way to support multiple outputs on Tegra20 as well, pll_p is
+	 * typically used as the parent clock for the display controllers.
+	 * But this comes at a cost: pll_p is the parent of several other
+	 * peripherals, so its frequency shouldn't change out of the blue.
+	 *
+	 * The best we can do at this point is to use the shift clock divider
+	 * and hope that the desired frequency can be matched (or at least
+	 * matched sufficiently close that the panel will still work).
+	 */
+
+	*div = ((clk_get_rate(clk) * 2) / pclk) - 2;
+
+	return 0;
 }
 
 static int tegra_output_rgb_check_mode(struct tegra_output *output,
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index 49ef5729f435..27c979b50111 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/clk.h>
+#include <linux/debugfs.h>
 #include <linux/io.h>
 #include <linux/platform_device.h>
 #include <linux/reset.h>
@@ -33,7 +34,23 @@ struct tegra_sor {
 
 	struct tegra_dpaux *dpaux;
 
+	struct mutex lock;
 	bool enabled;
+
+	struct dentry *debugfs;
+};
+
+struct tegra_sor_config {
+	u32 bits_per_pixel;
+
+	u32 active_polarity;
+	u32 active_count;
+	u32 tu_size;
+	u32 active_frac;
+	u32 watermark;
+
+	u32 hblank_symbols;
+	u32 vblank_symbols;
 };
 
 static inline struct tegra_sor *
@@ -289,34 +306,232 @@ static int tegra_sor_power_up(struct tegra_sor *sor, unsigned long timeout)
 	return -ETIMEDOUT;
 }
 
+struct tegra_sor_params {
+	/* number of link clocks per line */
+	unsigned int num_clocks;
+	/* ratio between input and output */
+	u64 ratio;
+	/* precision factor */
+	u64 precision;
+
+	unsigned int active_polarity;
+	unsigned int active_count;
+	unsigned int active_frac;
+	unsigned int tu_size;
+	unsigned int error;
+};
+
+static int tegra_sor_compute_params(struct tegra_sor *sor,
+				    struct tegra_sor_params *params,
+				    unsigned int tu_size)
+{
+	u64 active_sym, active_count, frac, approx;
+	u32 active_polarity, active_frac = 0;
+	const u64 f = params->precision;
+	s64 error;
+
+	active_sym = params->ratio * tu_size;
+	active_count = div_u64(active_sym, f) * f;
+	frac = active_sym - active_count;
+
+	/* fraction < 0.5 */
+	if (frac >= (f / 2)) {
+		active_polarity = 1;
+		frac = f - frac;
+	} else {
+		active_polarity = 0;
+	}
+
+	if (frac != 0) {
+		frac = div_u64(f * f,  frac); /* 1/fraction */
+		if (frac <= (15 * f)) {
+			active_frac = div_u64(frac, f);
+
+			/* round up */
+			if (active_polarity)
+				active_frac++;
+		} else {
+			active_frac = active_polarity ? 1 : 15;
+		}
+	}
+
+	if (active_frac == 1)
+		active_polarity = 0;
+
+	if (active_polarity == 1) {
+		if (active_frac) {
+			approx = active_count + (active_frac * (f - 1)) * f;
+			approx = div_u64(approx, active_frac * f);
+		} else {
+			approx = active_count + f;
+		}
+	} else {
+		if (active_frac)
+			approx = active_count + div_u64(f, active_frac);
+		else
+			approx = active_count;
+	}
+
+	error = div_s64(active_sym - approx, tu_size);
+	error *= params->num_clocks;
+
+	if (error <= 0 && abs64(error) < params->error) {
+		params->active_count = div_u64(active_count, f);
+		params->active_polarity = active_polarity;
+		params->active_frac = active_frac;
+		params->error = abs64(error);
+		params->tu_size = tu_size;
+
+		if (error == 0)
+			return true;
+	}
+
+	return false;
+}
+
+static int tegra_sor_calc_config(struct tegra_sor *sor,
+				 struct drm_display_mode *mode,
+				 struct tegra_sor_config *config,
+				 struct drm_dp_link *link)
+{
+	const u64 f = 100000, link_rate = link->rate * 1000;
+	const u64 pclk = mode->clock * 1000;
+	u64 input, output, watermark, num;
+	struct tegra_sor_params params;
+	u32 num_syms_per_line;
+	unsigned int i;
+
+	if (!link_rate || !link->num_lanes || !pclk || !config->bits_per_pixel)
+		return -EINVAL;
+
+	output = link_rate * 8 * link->num_lanes;
+	input = pclk * config->bits_per_pixel;
+
+	if (input >= output)
+		return -ERANGE;
+
+	memset(&params, 0, sizeof(params));
+	params.ratio = div64_u64(input * f, output);
+	params.num_clocks = div_u64(link_rate * mode->hdisplay, pclk);
+	params.precision = f;
+	params.error = 64 * f;
+	params.tu_size = 64;
+
+	for (i = params.tu_size; i >= 32; i--)
+		if (tegra_sor_compute_params(sor, &params, i))
+			break;
+
+	if (params.active_frac == 0) {
+		config->active_polarity = 0;
+		config->active_count = params.active_count;
+
+		if (!params.active_polarity)
+			config->active_count--;
+
+		config->tu_size = params.tu_size;
+		config->active_frac = 1;
+	} else {
+		config->active_polarity = params.active_polarity;
+		config->active_count = params.active_count;
+		config->active_frac = params.active_frac;
+		config->tu_size = params.tu_size;
+	}
+
+	dev_dbg(sor->dev,
+		"polarity: %d active count: %d tu size: %d active frac: %d\n",
+		config->active_polarity, config->active_count,
+		config->tu_size, config->active_frac);
+
+	watermark = params.ratio * config->tu_size * (f - params.ratio);
+	watermark = div_u64(watermark, f);
+
+	watermark = div_u64(watermark + params.error, f);
+	config->watermark = watermark + (config->bits_per_pixel / 8) + 2;
+	num_syms_per_line = (mode->hdisplay * config->bits_per_pixel) *
+			    (link->num_lanes * 8);
+
+	if (config->watermark > 30) {
+		config->watermark = 30;
+		dev_err(sor->dev,
+			"unable to compute TU size, forcing watermark to %u\n",
+			config->watermark);
+	} else if (config->watermark > num_syms_per_line) {
+		config->watermark = num_syms_per_line;
+		dev_err(sor->dev, "watermark too high, forcing to %u\n",
+			config->watermark);
+	}
+
+	/* compute the number of symbols per horizontal blanking interval */
+	num = ((mode->htotal - mode->hdisplay) - 7) * link_rate;
+	config->hblank_symbols = div_u64(num, pclk);
+
+	if (link->capabilities & DP_LINK_CAP_ENHANCED_FRAMING)
+		config->hblank_symbols -= 3;
+
+	config->hblank_symbols -= 12 / link->num_lanes;
+
+	/* compute the number of symbols per vertical blanking interval */
+	num = (mode->hdisplay - 25) * link_rate;
+	config->vblank_symbols = div_u64(num, pclk);
+	config->vblank_symbols -= 36 / link->num_lanes + 4;
+
+	dev_dbg(sor->dev, "blank symbols: H:%u V:%u\n", config->hblank_symbols,
+		config->vblank_symbols);
+
+	return 0;
+}
+
 static int tegra_output_sor_enable(struct tegra_output *output)
 {
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct drm_display_mode *mode = &dc->base.mode;
 	unsigned int vbe, vse, hbe, hse, vbs, hbs, i;
 	struct tegra_sor *sor = to_sor(output);
+	struct tegra_sor_config config;
+	struct drm_dp_link link;
+	struct drm_dp_aux *aux;
 	unsigned long value;
-	int err;
+	int err = 0;
+
+	mutex_lock(&sor->lock);
 
 	if (sor->enabled)
-		return 0;
+		goto unlock;
 
 	err = clk_prepare_enable(sor->clk);
 	if (err < 0)
-		return err;
+		goto unlock;
 
 	reset_control_deassert(sor->rst);
 
+	/* FIXME: properly convert to struct drm_dp_aux */
+	aux = (struct drm_dp_aux *)sor->dpaux;
+
 	if (sor->dpaux) {
 		err = tegra_dpaux_enable(sor->dpaux);
 		if (err < 0)
 			dev_err(sor->dev, "failed to enable DP: %d\n", err);
+
+		err = drm_dp_link_probe(aux, &link);
+		if (err < 0) {
+			dev_err(sor->dev, "failed to probe eDP link: %d\n",
+				err);
+			return err;
+		}
 	}
 
 	err = clk_set_parent(sor->clk, sor->clk_safe);
 	if (err < 0)
 		dev_err(sor->dev, "failed to set safe parent clock: %d\n", err);
 
+	memset(&config, 0, sizeof(config));
+	config.bits_per_pixel = 24; /* XXX: don't hardcode? */
+
+	err = tegra_sor_calc_config(sor, mode, &config, &link);
+	if (err < 0)
+		dev_err(sor->dev, "failed to compute link configuration: %d\n",
+			err);
+
 	value = tegra_sor_readl(sor, SOR_CLK_CNTRL);
 	value &= ~SOR_CLK_CNTRL_DP_CLK_SEL_MASK;
 	value |= SOR_CLK_CNTRL_DP_CLK_SEL_SINGLE_DPCLK;
@@ -385,7 +600,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_io_rail_power_on(TEGRA_IO_RAIL_LVDS);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power on I/O rail: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	usleep_range(5, 100);
@@ -419,15 +634,29 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	if (err < 0)
 		dev_err(sor->dev, "failed to set DP parent clock: %d\n", err);
 
-	/* power dplanes (XXX parameterize based on link?) */
+	/* power DP lanes */
 	value = tegra_sor_readl(sor, SOR_DP_PADCTL_0);
-	value |= SOR_DP_PADCTL_PD_TXD_3 | SOR_DP_PADCTL_PD_TXD_0 |
-		 SOR_DP_PADCTL_PD_TXD_1 | SOR_DP_PADCTL_PD_TXD_2;
+
+	if (link.num_lanes <= 2)
+		value &= ~(SOR_DP_PADCTL_PD_TXD_3 | SOR_DP_PADCTL_PD_TXD_2);
+	else
+		value |= SOR_DP_PADCTL_PD_TXD_3 | SOR_DP_PADCTL_PD_TXD_2;
+
+	if (link.num_lanes <= 1)
+		value &= ~SOR_DP_PADCTL_PD_TXD_1;
+	else
+		value |= SOR_DP_PADCTL_PD_TXD_1;
+
+	if (link.num_lanes == 0)
+		value &= ~SOR_DP_PADCTL_PD_TXD_0;
+	else
+		value |= SOR_DP_PADCTL_PD_TXD_0;
+
 	tegra_sor_writel(sor, value, SOR_DP_PADCTL_0);
 
 	value = tegra_sor_readl(sor, SOR_DP_LINKCTL_0);
 	value &= ~SOR_DP_LINKCTL_LANE_COUNT_MASK;
-	value |= SOR_DP_LINKCTL_LANE_COUNT(4);
+	value |= SOR_DP_LINKCTL_LANE_COUNT(link.num_lanes);
 	tegra_sor_writel(sor, value, SOR_DP_LINKCTL_0);
 
 	/* start lane sequencer */
@@ -443,10 +672,10 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 		usleep_range(250, 1000);
 	}
 
-	/* set link bandwidth (2.7 GHz, XXX: parameterize based on link?) */
+	/* set link bandwidth */
 	value = tegra_sor_readl(sor, SOR_CLK_CNTRL);
 	value &= ~SOR_CLK_CNTRL_DP_LINK_SPEED_MASK;
-	value |= SOR_CLK_CNTRL_DP_LINK_SPEED_G2_70;
+	value |= drm_dp_link_rate_to_bw_code(link.rate) << 2;
 	tegra_sor_writel(sor, value, SOR_CLK_CNTRL);
 
 	/* set linkctl */
@@ -454,7 +683,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	value |= SOR_DP_LINKCTL_ENABLE;
 
 	value &= ~SOR_DP_LINKCTL_TU_SIZE_MASK;
-	value |= SOR_DP_LINKCTL_TU_SIZE(59); /* XXX: don't hardcode? */
+	value |= SOR_DP_LINKCTL_TU_SIZE(config.tu_size);
 
 	value |= SOR_DP_LINKCTL_ENHANCED_FRAME;
 	tegra_sor_writel(sor, value, SOR_DP_LINKCTL_0);
@@ -470,28 +699,31 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 
 	value = tegra_sor_readl(sor, SOR_DP_CONFIG_0);
 	value &= ~SOR_DP_CONFIG_WATERMARK_MASK;
-	value |= SOR_DP_CONFIG_WATERMARK(14); /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_WATERMARK(config.watermark);
 
 	value &= ~SOR_DP_CONFIG_ACTIVE_SYM_COUNT_MASK;
-	value |= SOR_DP_CONFIG_ACTIVE_SYM_COUNT(47); /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_ACTIVE_SYM_COUNT(config.active_count);
 
 	value &= ~SOR_DP_CONFIG_ACTIVE_SYM_FRAC_MASK;
-	value |= SOR_DP_CONFIG_ACTIVE_SYM_FRAC(9); /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_ACTIVE_SYM_FRAC(config.active_frac);
 
-	value &= ~SOR_DP_CONFIG_ACTIVE_SYM_POLARITY; /* XXX: don't hardcode? */
+	if (config.active_polarity)
+		value |= SOR_DP_CONFIG_ACTIVE_SYM_POLARITY;
+	else
+		value &= ~SOR_DP_CONFIG_ACTIVE_SYM_POLARITY;
 
 	value |= SOR_DP_CONFIG_ACTIVE_SYM_ENABLE;
-	value |= SOR_DP_CONFIG_DISPARITY_NEGATIVE; /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_DISPARITY_NEGATIVE;
 	tegra_sor_writel(sor, value, SOR_DP_CONFIG_0);
 
 	value = tegra_sor_readl(sor, SOR_DP_AUDIO_HBLANK_SYMBOLS);
 	value &= ~SOR_DP_AUDIO_HBLANK_SYMBOLS_MASK;
-	value |= 137; /* XXX: don't hardcode? */
+	value |= config.hblank_symbols & 0xffff;
 	tegra_sor_writel(sor, value, SOR_DP_AUDIO_HBLANK_SYMBOLS);
 
 	value = tegra_sor_readl(sor, SOR_DP_AUDIO_VBLANK_SYMBOLS);
 	value &= ~SOR_DP_AUDIO_VBLANK_SYMBOLS_MASK;
-	value |= 2368; /* XXX: don't hardcode? */
+	value |= config.vblank_symbols & 0xffff;
 	tegra_sor_writel(sor, value, SOR_DP_AUDIO_VBLANK_SYMBOLS);
 
 	/* enable pad calibration logic */
@@ -500,30 +732,27 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	tegra_sor_writel(sor, value, SOR_DP_PADCTL_0);
 
 	if (sor->dpaux) {
-		/* FIXME: properly convert to struct drm_dp_aux */
-		struct drm_dp_aux *aux = (struct drm_dp_aux *)sor->dpaux;
-		struct drm_dp_link link;
 		u8 rate, lanes;
 
 		err = drm_dp_link_probe(aux, &link);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to probe eDP link: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		err = drm_dp_link_power_up(aux, &link);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to power up eDP link: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		err = drm_dp_link_configure(aux, &link);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to configure eDP link: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		rate = drm_dp_link_rate_to_bw_code(link.rate);
@@ -558,7 +787,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 		if (err < 0) {
 			dev_err(sor->dev, "DP fast link training failed: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		dev_dbg(sor->dev, "fast link training succeeded\n");
@@ -567,7 +796,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_sor_power_up(sor, 250);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power up SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	/* start display controller in continuous mode */
@@ -586,12 +815,26 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	 * configure panel (24bpp, vsync-, hsync-, DP-A protocol, complete
 	 * raster, associate with display controller)
 	 */
-	value = SOR_STATE_ASY_PIXELDEPTH_BPP_24_444 |
-		SOR_STATE_ASY_VSYNCPOL |
+	value = SOR_STATE_ASY_VSYNCPOL |
 		SOR_STATE_ASY_HSYNCPOL |
 		SOR_STATE_ASY_PROTOCOL_DP_A |
 		SOR_STATE_ASY_CRC_MODE_COMPLETE |
 		SOR_STATE_ASY_OWNER(dc->pipe + 1);
+
+	switch (config.bits_per_pixel) {
+	case 24:
+		value |= SOR_STATE_ASY_PIXELDEPTH_BPP_24_444;
+		break;
+
+	case 18:
+		value |= SOR_STATE_ASY_PIXELDEPTH_BPP_18_444;
+		break;
+
+	default:
+		BUG();
+		break;
+	}
+
 	tegra_sor_writel(sor, value, SOR_STATE_1);
 
 	/*
@@ -620,11 +863,8 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	value = ((vbs & 0x7fff) << 16) | (hbs & 0x7fff);
 	tegra_sor_writel(sor, value, SOR_HEAD_STATE_4(0));
 
-	/* XXX interlaced mode */
-	tegra_sor_writel(sor, 0x00000001, SOR_HEAD_STATE_5(0));
-
 	/* CSTM (LVDS, link A/B, upper) */
-	value = SOR_CSTM_LVDS | SOR_CSTM_LINK_ACT_B | SOR_CSTM_LINK_ACT_B |
+	value = SOR_CSTM_LVDS | SOR_CSTM_LINK_ACT_A | SOR_CSTM_LINK_ACT_B |
 		SOR_CSTM_UPPER;
 	tegra_sor_writel(sor, value, SOR_CSTM);
 
@@ -632,7 +872,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_sor_setup_pwm(sor, 250);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to setup PWM: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	value = tegra_dc_readl(dc, DC_DISP_DISP_WIN_OPTIONS);
@@ -644,18 +884,20 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_sor_attach(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to attach SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	err = tegra_sor_wakeup(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to enable DC: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	sor->enabled = true;
 
-	return 0;
+unlock:
+	mutex_unlock(&sor->lock);
+	return err;
 }
 
 static int tegra_sor_detach(struct tegra_sor *sor)
@@ -740,7 +982,7 @@ static int tegra_sor_power_down(struct tegra_sor *sor)
 	tegra_sor_writel(sor, value, SOR_DP_PADCTL_0);
 
 	/* stop lane sequencer */
-	value = SOR_LANE_SEQ_CTL_TRIGGER | SOR_LANE_SEQ_CTL_SEQUENCE_DOWN |
+	value = SOR_LANE_SEQ_CTL_TRIGGER | SOR_LANE_SEQ_CTL_SEQUENCE_UP |
 		SOR_LANE_SEQ_CTL_POWER_STATE_DOWN;
 	tegra_sor_writel(sor, value, SOR_LANE_SEQ_CTL);
 
@@ -783,15 +1025,17 @@ static int tegra_output_sor_disable(struct tegra_output *output)
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct tegra_sor *sor = to_sor(output);
 	unsigned long value;
-	int err;
+	int err = 0;
+
+	mutex_lock(&sor->lock);
 
 	if (!sor->enabled)
-		return 0;
+		goto unlock;
 
 	err = tegra_sor_detach(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to detach SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	tegra_sor_writel(sor, 0, SOR_STATE_1);
@@ -832,21 +1076,21 @@ static int tegra_output_sor_disable(struct tegra_output *output)
 	err = tegra_sor_power_down(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power down SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	if (sor->dpaux) {
 		err = tegra_dpaux_disable(sor->dpaux);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to disable DP: %d\n", err);
-			return err;
+			goto unlock;
 		}
 	}
 
 	err = tegra_io_rail_power_off(TEGRA_IO_RAIL_LVDS);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power off I/O rail: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	reset_control_assert(sor->rst);
@@ -854,18 +1098,18 @@ static int tegra_output_sor_disable(struct tegra_output *output)
 
 	sor->enabled = false;
 
-	return 0;
+unlock:
+	mutex_unlock(&sor->lock);
+	return err;
 }
 
 static int tegra_output_sor_setup_clock(struct tegra_output *output,
-					struct clk *clk, unsigned long pclk)
+					struct clk *clk, unsigned long pclk,
+					unsigned int *div)
 {
 	struct tegra_sor *sor = to_sor(output);
 	int err;
 
-	/* round to next MHz */
-	pclk = DIV_ROUND_UP(pclk / 2, 1000000) * 1000000;
-
 	err = clk_set_parent(clk, sor->clk_parent);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to set parent clock: %d\n", err);
@@ -874,11 +1118,12 @@ static int tegra_output_sor_setup_clock(struct tegra_output *output,
 
 	err = clk_set_rate(sor->clk_parent, pclk);
 	if (err < 0) {
-		dev_err(sor->dev, "failed to set base clock rate to %lu Hz\n",
-			pclk * 2);
+		dev_err(sor->dev, "failed to set clock rate to %lu Hz\n", pclk);
 		return err;
 	}
 
+	*div = 0;
+
 	return 0;
 }
 
@@ -914,9 +1159,124 @@ static const struct tegra_output_ops sor_ops = {
 	.detect = tegra_output_sor_detect,
 };
 
+static int tegra_sor_crc_open(struct inode *inode, struct file *file)
+{
+	file->private_data = inode->i_private;
+
+	return 0;
+}
+
+static int tegra_sor_crc_release(struct inode *inode, struct file *file)
+{
+	return 0;
+}
+
+static int tegra_sor_crc_wait(struct tegra_sor *sor, unsigned long timeout)
+{
+	u32 value;
+
+	timeout = jiffies + msecs_to_jiffies(timeout);
+
+	while (time_before(jiffies, timeout)) {
+		value = tegra_sor_readl(sor, SOR_CRC_A);
+		if (value & SOR_CRC_A_VALID)
+			return 0;
+
+		usleep_range(100, 200);
+	}
+
+	return -ETIMEDOUT;
+}
+
+static ssize_t tegra_sor_crc_read(struct file *file, char __user *buffer,
+				  size_t size, loff_t *ppos)
+{
+	struct tegra_sor *sor = file->private_data;
+	ssize_t num, err;
+	char buf[10];
+	u32 value;
+
+	mutex_lock(&sor->lock);
+
+	if (!sor->enabled) {
+		err = -EAGAIN;
+		goto unlock;
+	}
+
+	value = tegra_sor_readl(sor, SOR_STATE_1);
+	value &= ~SOR_STATE_ASY_CRC_MODE_MASK;
+	tegra_sor_writel(sor, value, SOR_STATE_1);
+
+	value = tegra_sor_readl(sor, SOR_CRC_CNTRL);
+	value |= SOR_CRC_CNTRL_ENABLE;
+	tegra_sor_writel(sor, value, SOR_CRC_CNTRL);
+
+	value = tegra_sor_readl(sor, SOR_TEST);
+	value &= ~SOR_TEST_CRC_POST_SERIALIZE;
+	tegra_sor_writel(sor, value, SOR_TEST);
+
+	err = tegra_sor_crc_wait(sor, 100);
+	if (err < 0)
+		goto unlock;
+
+	tegra_sor_writel(sor, SOR_CRC_A_RESET, SOR_CRC_A);
+	value = tegra_sor_readl(sor, SOR_CRC_B);
+
+	num = scnprintf(buf, sizeof(buf), "%08x\n", value);
+
+	err = simple_read_from_buffer(buffer, size, ppos, buf, num);
+
+unlock:
+	mutex_unlock(&sor->lock);
+	return err;
+}
+
+static const struct file_operations tegra_sor_crc_fops = {
+	.owner = THIS_MODULE,
+	.open = tegra_sor_crc_open,
+	.read = tegra_sor_crc_read,
+	.release = tegra_sor_crc_release,
+};
+
+static int tegra_sor_debugfs_init(struct tegra_sor *sor,
+				  struct drm_minor *minor)
+{
+	struct dentry *entry;
+	int err = 0;
+
+	sor->debugfs = debugfs_create_dir("sor", minor->debugfs_root);
+	if (!sor->debugfs)
+		return -ENOMEM;
+
+	entry = debugfs_create_file("crc", 0644, sor->debugfs, sor,
+				    &tegra_sor_crc_fops);
+	if (!entry) {
+		dev_err(sor->dev,
+			"cannot create /sys/kernel/debug/dri/%s/sor/crc\n",
+			minor->debugfs_root->d_name.name);
+		err = -ENOMEM;
+		goto remove;
+	}
+
+	return err;
+
+remove:
+	debugfs_remove(sor->debugfs);
+	sor->debugfs = NULL;
+	return err;
+}
+
+static int tegra_sor_debugfs_exit(struct tegra_sor *sor)
+{
+	debugfs_remove_recursive(sor->debugfs);
+	sor->debugfs = NULL;
+
+	return 0;
+}
+
 static int tegra_sor_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_sor *sor = host1x_client_to_sor(client);
 	int err;
 
@@ -928,12 +1288,18 @@ static int tegra_sor_init(struct host1x_client *client)
 	sor->output.dev = sor->dev;
 	sor->output.ops = &sor_ops;
 
-	err = tegra_output_init(tegra->drm, &sor->output);
+	err = tegra_output_init(drm, &sor->output);
 	if (err < 0) {
 		dev_err(sor->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
+	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
+		err = tegra_sor_debugfs_init(sor, drm->primary);
+		if (err < 0)
+			dev_err(sor->dev, "debugfs setup failed: %d\n", err);
+	}
+
 	if (sor->dpaux) {
 		err = tegra_dpaux_attach(sor->dpaux, &sor->output);
 		if (err < 0) {
@@ -964,6 +1330,12 @@ static int tegra_sor_exit(struct host1x_client *client)
 		}
 	}
 
+	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
+		err = tegra_sor_debugfs_exit(sor);
+		if (err < 0)
+			dev_err(sor->dev, "debugfs cleanup failed: %d\n", err);
+	}
+
 	err = tegra_output_exit(&sor->output);
 	if (err < 0) {
 		dev_err(sor->dev, "output cleanup failed: %d\n", err);
@@ -1045,6 +1417,8 @@ static int tegra_sor_probe(struct platform_device *pdev)
 	sor->client.ops = &sor_client_ops;
 	sor->client.dev = &pdev->dev;
 
+	mutex_init(&sor->lock);
+
 	err = host1x_client_register(&sor->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register host1x client: %d\n",
diff --git a/drivers/gpu/drm/tegra/sor.h b/drivers/gpu/drm/tegra/sor.h
index f4156d54cd05..a5f8853fedb5 100644
--- a/drivers/gpu/drm/tegra/sor.h
+++ b/drivers/gpu/drm/tegra/sor.h
@@ -47,6 +47,7 @@
 #define SOR_HEAD_STATE_4(x) (0x0d + (x))
 #define SOR_HEAD_STATE_5(x) (0x0f + (x))
 #define SOR_CRC_CNTRL 0x11
+#define  SOR_CRC_CNTRL_ENABLE			(1 << 0)
 #define SOR_DP_DEBUG_MVID 0x12
 
 #define SOR_CLK_CNTRL 0x13
@@ -69,6 +70,7 @@
 #define  SOR_PWR_NORMAL_STATE_PU		(1 << 0)
 
 #define SOR_TEST 0x16
+#define  SOR_TEST_CRC_POST_SERIALIZE		(1 << 23)
 #define  SOR_TEST_ATTACHED			(1 << 10)
 #define  SOR_TEST_HEAD_MODE_MASK		(3 << 8)
 #define  SOR_TEST_HEAD_MODE_AWAKE		(2 << 8)
@@ -115,6 +117,8 @@
 
 #define SOR_LVDS 0x1c
 #define SOR_CRC_A 0x1d
+#define  SOR_CRC_A_VALID			(1 << 0)
+#define  SOR_CRC_A_RESET			(1 << 0)
 #define SOR_CRC_B 0x1e
 #define SOR_BLANK 0x1f
 #define SOR_SEQ_CTL 0x20
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
index 171a8203892c..b20b69488dc9 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -268,7 +268,7 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 	}
 
 	pm_runtime_get_sync(dev->dev);
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	pm_runtime_put_sync(dev->dev);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to install IRQ handler\n");
diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index afdf383f630a..7094b92d1ec7 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -294,6 +294,7 @@ int udl_driver_load(struct drm_device *dev, unsigned long flags)
 	dev->dev_private = udl;
 
 	if (!udl_parse_vendor_descriptor(dev, dev->usbdev)) {
+		ret = -ENODEV;
 		DRM_ERROR("firmware not recognized. Assume incompatible device\n");
 		goto err;
 	}
diff --git a/drivers/gpu/drm/via/via_dma.c b/drivers/gpu/drm/via/via_dma.c
index a18479c6b6da..6fc0648dd37f 100644
--- a/drivers/gpu/drm/via/via_dma.c
+++ b/drivers/gpu/drm/via/via_dma.c
@@ -737,4 +737,4 @@ const struct drm_ioctl_desc via_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(VIA_BLIT_SYNC, via_dma_blit_sync, DRM_AUTH)
 };
 
-int via_max_ioctl = DRM_ARRAY_SIZE(via_ioctls);
+int via_max_ioctl = ARRAY_SIZE(via_ioctls);
diff --git a/drivers/gpu/drm/via/via_mm.c b/drivers/gpu/drm/via/via_mm.c
index 927889105483..d70b1e1544bf 100644
--- a/drivers/gpu/drm/via/via_mm.c
+++ b/drivers/gpu/drm/via/via_mm.c
@@ -79,7 +79,7 @@ int via_final_context(struct drm_device *dev, int context)
 
 	/* Linux specific until context tracking code gets ported to BSD */
 	/* Last context, perform cleanup */
-	if (list_is_singular(&dev->ctxlist) && dev->dev_private) {
+	if (list_is_singular(&dev->ctxlist)) {
 		DRM_DEBUG("Last Context\n");
 		drm_irq_uninstall(dev);
 		via_cleanup_futex(dev_priv);
diff --git a/drivers/gpu/drm/vmwgfx/Kconfig b/drivers/gpu/drm/vmwgfx/Kconfig
index b71bcd0bfbbf..67720f70fe29 100644
--- a/drivers/gpu/drm/vmwgfx/Kconfig
+++ b/drivers/gpu/drm/vmwgfx/Kconfig
@@ -1,11 +1,14 @@
 config DRM_VMWGFX
 	tristate "DRM driver for VMware Virtual GPU"
-	depends on DRM && PCI && FB
+	depends on DRM && PCI
 	select FB_DEFERRED_IO
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
 	select DRM_TTM
+	# Only needed for the transitional use of drm_crtc_init - can be removed
+	# again once vmwgfx sets up the primary plane itself.
+	select DRM_KMS_HELPER
 	help
 	  Choose this option if you would like to run 3D acceleration
 	  in a VMware virtual machine.
@@ -14,7 +17,7 @@ config DRM_VMWGFX
 	  The compiled module will be called "vmwgfx.ko".
 
 config DRM_VMWGFX_FBCON
-	depends on DRM_VMWGFX
+	depends on DRM_VMWGFX && FB
 	bool "Enable framebuffer console under vmwgfx by default"
 	help
 	   Choose this option if you are shipping a new vmwgfx
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 4a223bbea3b3..246a62bab378 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -806,7 +806,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	}
 
 	if (dev_priv->capabilities & SVGA_CAP_IRQMASK) {
-		ret = drm_irq_install(dev);
+		ret = drm_irq_install(dev, dev->pdev->irq);
 		if (ret != 0) {
 			DRM_ERROR("Failed installing irq: %d\n", ret);
 			goto out_no_irq;
@@ -1417,7 +1417,7 @@ static struct drm_driver driver = {
 	.enable_vblank = vmw_enable_vblank,
 	.disable_vblank = vmw_disable_vblank,
 	.ioctls = vmw_ioctls,
-	.num_ioctls = DRM_ARRAY_SIZE(vmw_ioctls),
+	.num_ioctls = ARRAY_SIZE(vmw_ioctls),
 	.master_create = vmw_master_create,
 	.master_destroy = vmw_master_destroy,
 	.master_set = vmw_master_set,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c
index a89ad938eacf..b031b48dbb3c 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c
@@ -179,7 +179,6 @@ static int vmw_fb_set_par(struct fb_info *info)
 		vmw_write(vmw_priv, SVGA_REG_DISPLAY_POSITION_Y, info->var.yoffset);
 		vmw_write(vmw_priv, SVGA_REG_DISPLAY_WIDTH, info->var.xres);
 		vmw_write(vmw_priv, SVGA_REG_DISPLAY_HEIGHT, info->var.yres);
-		vmw_write(vmw_priv, SVGA_REG_BYTES_PER_LINE, info->fix.line_length);
 		vmw_write(vmw_priv, SVGA_REG_DISPLAY_ID, SVGA_ID_INVALID);
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index a2dde5ad8138..8f3edc4710f2 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -187,7 +187,7 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	 * can do this since the caller in the drm core doesn't check anything
 	 * which is protected by any looks.
 	 */
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 	drm_modeset_lock_all(dev_priv->dev);
 
 	/* A lot of the code assumes this */
@@ -252,7 +252,7 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	ret = 0;
 out:
 	drm_modeset_unlock_all(dev_priv->dev);
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 
 	return ret;
 }
@@ -273,7 +273,7 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 	 * can do this since the caller in the drm core doesn't check anything
 	 * which is protected by any looks.
 	 */
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 	drm_modeset_lock_all(dev_priv->dev);
 
 	vmw_cursor_update_position(dev_priv, shown,
@@ -281,7 +281,7 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 				   du->cursor_y + du->hotspot_y);
 
 	drm_modeset_unlock_all(dev_priv->dev);
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 
 	return 0;
 }
@@ -2001,7 +2001,7 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector,
 	if (du->pref_mode)
 		list_move(&du->pref_mode->head, &connector->probed_modes);
 
-	drm_mode_connector_list_update(connector);
+	drm_mode_connector_list_update(connector, true);
 
 	return 1;
 }
diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c
index ccdd2e6da5e3..aaf54859adb0 100644
--- a/drivers/gpu/host1x/bus.c
+++ b/drivers/gpu/host1x/bus.c
@@ -216,8 +216,8 @@ int host1x_device_exit(struct host1x_device *device)
 }
 EXPORT_SYMBOL(host1x_device_exit);
 
-static int host1x_register_client(struct host1x *host1x,
-				  struct host1x_client *client)
+static int host1x_add_client(struct host1x *host1x,
+			     struct host1x_client *client)
 {
 	struct host1x_device *device;
 	struct host1x_subdev *subdev;
@@ -238,8 +238,8 @@ static int host1x_register_client(struct host1x *host1x,
 	return -ENODEV;
 }
 
-static int host1x_unregister_client(struct host1x *host1x,
-				    struct host1x_client *client)
+static int host1x_del_client(struct host1x *host1x,
+			     struct host1x_client *client)
 {
 	struct host1x_device *device, *dt;
 	struct host1x_subdev *subdev;
@@ -503,7 +503,7 @@ int host1x_client_register(struct host1x_client *client)
 	mutex_lock(&devices_lock);
 
 	list_for_each_entry(host1x, &devices, list) {
-		err = host1x_register_client(host1x, client);
+		err = host1x_add_client(host1x, client);
 		if (!err) {
 			mutex_unlock(&devices_lock);
 			return 0;
@@ -529,7 +529,7 @@ int host1x_client_unregister(struct host1x_client *client)
 	mutex_lock(&devices_lock);
 
 	list_for_each_entry(host1x, &devices, list) {
-		err = host1x_unregister_client(host1x, client);
+		err = host1x_del_client(host1x, client);
 		if (!err) {
 			mutex_unlock(&devices_lock);
 			return 0;
diff --git a/drivers/gpu/ipu-v3/Kconfig b/drivers/gpu/ipu-v3/Kconfig
new file mode 100644
index 000000000000..2f228a2f2a48
--- /dev/null
+++ b/drivers/gpu/ipu-v3/Kconfig
@@ -0,0 +1,7 @@
+config IMX_IPUV3_CORE
+	tristate "IPUv3 core support"
+	depends on SOC_IMX5 || SOC_IMX6Q || SOC_IMX6SL || ARCH_MULTIPLATFORM
+	depends on RESET_CONTROLLER
+	help
+	  Choose this if you have a i.MX5/6 system and want to use the Image
+	  Processing Unit. This option only enables IPU base support.
diff --git a/drivers/staging/imx-drm/ipu-v3/Makefile b/drivers/gpu/ipu-v3/Makefile
index 28ed72e98a96..1887972b4ac2 100644
--- a/drivers/staging/imx-drm/ipu-v3/Makefile
+++ b/drivers/gpu/ipu-v3/Makefile
@@ -1,3 +1,3 @@
-obj-$(CONFIG_DRM_IMX_IPUV3_CORE) += imx-ipu-v3.o
+obj-$(CONFIG_IMX_IPUV3_CORE) += imx-ipu-v3.o
 
-imx-ipu-v3-objs := ipu-common.o ipu-dc.o ipu-di.o ipu-dp.o ipu-dmfc.o
+imx-ipu-v3-objs := ipu-common.o ipu-dc.o ipu-di.o ipu-dp.o ipu-dmfc.o ipu-smfc.o
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
index a1f7b2001c8a..04e7b2eafbdd 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-common.c
+++ b/drivers/gpu/ipu-v3/ipu-common.c
@@ -31,7 +31,7 @@
 
 #include <drm/drm_fourcc.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 static inline u32 ipu_cm_read(struct ipu_soc *ipu, unsigned offset)
@@ -661,6 +661,39 @@ int ipu_module_disable(struct ipu_soc *ipu, u32 mask)
 }
 EXPORT_SYMBOL_GPL(ipu_module_disable);
 
+int ipu_csi_enable(struct ipu_soc *ipu, int csi)
+{
+	return ipu_module_enable(ipu, csi ? IPU_CONF_CSI1_EN : IPU_CONF_CSI0_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_csi_enable);
+
+int ipu_csi_disable(struct ipu_soc *ipu, int csi)
+{
+	return ipu_module_disable(ipu, csi ? IPU_CONF_CSI1_EN : IPU_CONF_CSI0_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_csi_disable);
+
+int ipu_smfc_enable(struct ipu_soc *ipu)
+{
+	return ipu_module_enable(ipu, IPU_CONF_SMFC_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_enable);
+
+int ipu_smfc_disable(struct ipu_soc *ipu)
+{
+	return ipu_module_disable(ipu, IPU_CONF_SMFC_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_disable);
+
+int ipu_idmac_get_current_buffer(struct ipuv3_channel *channel)
+{
+	struct ipu_soc *ipu = channel->ipu;
+	unsigned int chno = channel->num;
+
+	return (ipu_cm_read(ipu, IPU_CHA_CUR_BUF(chno)) & idma_mask(chno)) ? 1 : 0;
+}
+EXPORT_SYMBOL_GPL(ipu_idmac_get_current_buffer);
+
 void ipu_idmac_select_buffer(struct ipuv3_channel *channel, u32 buf_num)
 {
 	struct ipu_soc *ipu = channel->ipu;
@@ -896,8 +929,17 @@ static int ipu_submodules_init(struct ipu_soc *ipu,
 		goto err_dp;
 	}
 
+	ret = ipu_smfc_init(ipu, dev, ipu_base +
+			devtype->cm_ofs + IPU_CM_SMFC_REG_OFS);
+	if (ret) {
+		unit = "smfc";
+		goto err_smfc;
+	}
+
 	return 0;
 
+err_smfc:
+	ipu_dp_exit(ipu);
 err_dp:
 	ipu_dmfc_exit(ipu);
 err_dmfc:
@@ -977,6 +1019,7 @@ EXPORT_SYMBOL_GPL(ipu_idmac_channel_irq);
 
 static void ipu_submodules_exit(struct ipu_soc *ipu)
 {
+	ipu_smfc_exit(ipu);
 	ipu_dp_exit(ipu);
 	ipu_dmfc_exit(ipu);
 	ipu_dc_exit(ipu);
@@ -1001,6 +1044,7 @@ static void platform_device_unregister_children(struct platform_device *pdev)
 struct ipu_platform_reg {
 	struct ipu_client_platformdata pdata;
 	const char *name;
+	int reg_offset;
 };
 
 static const struct ipu_platform_reg client_reg[] = {
@@ -1022,13 +1066,29 @@ static const struct ipu_platform_reg client_reg[] = {
 			.dma[1] = -EINVAL,
 		},
 		.name = "imx-ipuv3-crtc",
+	}, {
+		.pdata = {
+			.csi = 0,
+			.dma[0] = IPUV3_CHANNEL_CSI0,
+			.dma[1] = -EINVAL,
+		},
+		.reg_offset = IPU_CM_CSI0_REG_OFS,
+		.name = "imx-ipuv3-camera",
+	}, {
+		.pdata = {
+			.csi = 1,
+			.dma[0] = IPUV3_CHANNEL_CSI1,
+			.dma[1] = -EINVAL,
+		},
+		.reg_offset = IPU_CM_CSI1_REG_OFS,
+		.name = "imx-ipuv3-camera",
 	},
 };
 
 static DEFINE_MUTEX(ipu_client_id_mutex);
 static int ipu_client_id;
 
-static int ipu_add_client_devices(struct ipu_soc *ipu)
+static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base)
 {
 	struct device *dev = ipu->dev;
 	unsigned i;
@@ -1042,9 +1102,19 @@ static int ipu_add_client_devices(struct ipu_soc *ipu)
 	for (i = 0; i < ARRAY_SIZE(client_reg); i++) {
 		const struct ipu_platform_reg *reg = &client_reg[i];
 		struct platform_device *pdev;
-
-		pdev = platform_device_register_data(dev, reg->name,
-			id++, &reg->pdata, sizeof(reg->pdata));
+		struct resource res;
+
+		if (reg->reg_offset) {
+			memset(&res, 0, sizeof(res));
+			res.flags = IORESOURCE_MEM;
+			res.start = ipu_base + ipu->devtype->cm_ofs + reg->reg_offset;
+			res.end = res.start + PAGE_SIZE - 1;
+			pdev = platform_device_register_resndata(dev, reg->name,
+				id++, &res, 1, &reg->pdata, sizeof(reg->pdata));
+		} else {
+			pdev = platform_device_register_data(dev, reg->name,
+				id++, &reg->pdata, sizeof(reg->pdata));
+		}
 
 		if (IS_ERR(pdev))
 			goto err_register;
@@ -1241,7 +1311,7 @@ static int ipu_probe(struct platform_device *pdev)
 	if (ret)
 		goto failed_submodules_init;
 
-	ret = ipu_add_client_devices(ipu);
+	ret = ipu_add_client_devices(ipu, ipu_base);
 	if (ret) {
 		dev_err(&pdev->dev, "adding client devices failed with %d\n",
 				ret);
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-dc.c b/drivers/gpu/ipu-v3/ipu-dc.c
index 784a4a13eac3..2326c752d89b 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-dc.c
+++ b/drivers/gpu/ipu-v3/ipu-dc.c
@@ -21,8 +21,7 @@
 #include <linux/interrupt.h>
 #include <linux/io.h>
 
-#include "../imx-drm.h"
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 #define DC_MAP_CONF_PTR(n)	(0x108 + ((n) & ~0x1) * 2)
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-di.c b/drivers/gpu/ipu-v3/ipu-di.c
index 849b3e120ef0..c490ba4384fc 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-di.c
+++ b/drivers/gpu/ipu-v3/ipu-di.c
@@ -20,7 +20,7 @@
 #include <linux/err.h>
 #include <linux/platform_device.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 struct ipu_di {
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-dmfc.c b/drivers/gpu/ipu-v3/ipu-dmfc.c
index 59f182b28fc1..042c3958e2a0 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-dmfc.c
+++ b/drivers/gpu/ipu-v3/ipu-dmfc.c
@@ -17,7 +17,7 @@
 #include <linux/errno.h>
 #include <linux/io.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 #define DMFC_RD_CHAN		0x0000
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-dp.c b/drivers/gpu/ipu-v3/ipu-dp.c
index d90f82a87d19..98686edbcdbb 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-dp.c
+++ b/drivers/gpu/ipu-v3/ipu-dp.c
@@ -19,7 +19,7 @@
 #include <linux/io.h>
 #include <linux/err.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 #define DP_SYNC 0
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-prv.h b/drivers/gpu/ipu-v3/ipu-prv.h
index bfc1b3366488..c93f50ec04f7 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-prv.h
+++ b/drivers/gpu/ipu-v3/ipu-prv.h
@@ -22,7 +22,7 @@ struct ipu_soc;
 #include <linux/clk.h>
 #include <linux/platform_device.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 
 #define IPUV3_CHANNEL_CSI0			 0
 #define IPUV3_CHANNEL_CSI1			 1
@@ -151,6 +151,8 @@ struct ipuv3_channel {
 struct ipu_dc_priv;
 struct ipu_dmfc_priv;
 struct ipu_di;
+struct ipu_smfc_priv;
+
 struct ipu_devtype;
 
 struct ipu_soc {
@@ -178,6 +180,7 @@ struct ipu_soc {
 	struct ipu_dp_priv	*dp_priv;
 	struct ipu_dmfc_priv	*dmfc_priv;
 	struct ipu_di		*di_priv[2];
+	struct ipu_smfc_priv	*smfc_priv;
 };
 
 void ipu_srm_dp_sync_update(struct ipu_soc *ipu);
@@ -206,4 +209,7 @@ void ipu_dc_exit(struct ipu_soc *ipu);
 int ipu_cpmem_init(struct ipu_soc *ipu, struct device *dev, unsigned long base);
 void ipu_cpmem_exit(struct ipu_soc *ipu);
 
+int ipu_smfc_init(struct ipu_soc *ipu, struct device *dev, unsigned long base);
+void ipu_smfc_exit(struct ipu_soc *ipu);
+
 #endif				/* __IPU_PRV_H__ */
diff --git a/drivers/gpu/ipu-v3/ipu-smfc.c b/drivers/gpu/ipu-v3/ipu-smfc.c
new file mode 100644
index 000000000000..e4f85ad286fc
--- /dev/null
+++ b/drivers/gpu/ipu-v3/ipu-smfc.c
@@ -0,0 +1,97 @@
+/*
+ * Copyright 2008-2010 Freescale Semiconductor, Inc. All Rights Reserved.
+ *
+ * The code contained herein is licensed under the GNU General Public
+ * License. You may obtain a copy of the GNU General Public License
+ * Version 2 or later at the following locations:
+ *
+ * http://www.opensource.org/licenses/gpl-license.html
+ * http://www.gnu.org/copyleft/gpl.html
+ */
+#define DEBUG
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/errno.h>
+#include <linux/spinlock.h>
+#include <linux/delay.h>
+#include <linux/clk.h>
+#include <video/imx-ipu-v3.h>
+
+#include "ipu-prv.h"
+
+struct ipu_smfc_priv {
+	void __iomem *base;
+	spinlock_t lock;
+};
+
+/*SMFC Registers */
+#define SMFC_MAP	0x0000
+#define SMFC_WMC	0x0004
+#define SMFC_BS		0x0008
+
+int ipu_smfc_set_burstsize(struct ipu_soc *ipu, int channel, int burstsize)
+{
+	struct ipu_smfc_priv *smfc = ipu->smfc_priv;
+	unsigned long flags;
+	u32 val, shift;
+
+	spin_lock_irqsave(&smfc->lock, flags);
+
+	shift = channel * 4;
+	val = readl(smfc->base + SMFC_BS);
+	val &= ~(0xf << shift);
+	val |= burstsize << shift;
+	writel(val, smfc->base + SMFC_BS);
+
+	spin_unlock_irqrestore(&smfc->lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_set_burstsize);
+
+int ipu_smfc_map_channel(struct ipu_soc *ipu, int channel, int csi_id, int mipi_id)
+{
+	struct ipu_smfc_priv *smfc = ipu->smfc_priv;
+	unsigned long flags;
+	u32 val, shift;
+
+	spin_lock_irqsave(&smfc->lock, flags);
+
+	shift = channel * 3;
+	val = readl(smfc->base + SMFC_MAP);
+	val &= ~(0x7 << shift);
+	val |= ((csi_id << 2) | mipi_id) << shift;
+	writel(val, smfc->base + SMFC_MAP);
+
+	spin_unlock_irqrestore(&smfc->lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_map_channel);
+
+int ipu_smfc_init(struct ipu_soc *ipu, struct device *dev,
+		  unsigned long base)
+{
+	struct ipu_smfc_priv *smfc;
+
+	smfc = devm_kzalloc(dev, sizeof(*smfc), GFP_KERNEL);
+	if (!smfc)
+		return -ENOMEM;
+
+	ipu->smfc_priv = smfc;
+	spin_lock_init(&smfc->lock);
+
+	smfc->base = devm_ioremap(dev, base, PAGE_SIZE);
+	if (!smfc->base)
+		return -ENOMEM;
+
+	pr_debug("%s: ioremap 0x%08lx -> %p\n", __func__, base, smfc->base);
+
+	return 0;
+}
+
+void ipu_smfc_exit(struct ipu_soc *ipu)
+{
+}
diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index ec0ae2d1686a..6866448083b2 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -623,7 +623,8 @@ static int vga_switcheroo_runtime_suspend(struct device *dev)
 	ret = dev->bus->pm->runtime_suspend(dev);
 	if (ret)
 		return ret;
-
+	if (vgasr_priv.handler->switchto)
+		vgasr_priv.handler->switchto(VGA_SWITCHEROO_IGD);
 	vga_switcheroo_power_switch(pdev, VGA_SWITCHEROO_OFF);
 	return 0;
 }
diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
index 800c8b60f7a2..5e79c6ad914f 100644
--- a/drivers/hid/Kconfig
+++ b/drivers/hid/Kconfig
@@ -810,7 +810,7 @@ config HID_ZYDACRON
 
 config HID_SENSOR_HUB
 	tristate "HID Sensors framework support"
-	depends on HID
+	depends on HID && HAS_IOMEM
 	select MFD_CORE
 	default n
 	---help---
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 6d00bb9366fa..48b66bbffc94 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -323,6 +323,7 @@
 
 #define USB_VENDOR_ID_ETURBOTOUCH	0x22b9
 #define USB_DEVICE_ID_ETURBOTOUCH	0x0006
+#define USB_DEVICE_ID_ETURBOTOUCH_2968	0x2968
 
 #define USB_VENDOR_ID_EZKEY		0x0518
 #define USB_DEVICE_ID_BTC_8193		0x0002
@@ -715,6 +716,8 @@
 
 #define USB_VENDOR_ID_PENMOUNT		0x14e1
 #define USB_DEVICE_ID_PENMOUNT_PCI	0x3500
+#define USB_DEVICE_ID_PENMOUNT_1610	0x1610
+#define USB_DEVICE_ID_PENMOUNT_1640	0x1640
 
 #define USB_VENDOR_ID_PETALYNX		0x18b1
 #define USB_DEVICE_ID_PETALYNX_MAXTER_REMOTE	0x0037
diff --git a/drivers/hid/hid-lg4ff.c b/drivers/hid/hid-lg4ff.c
index 24883b4d1a49..cc2bd2022198 100644
--- a/drivers/hid/hid-lg4ff.c
+++ b/drivers/hid/hid-lg4ff.c
@@ -52,7 +52,7 @@ static void hid_lg4ff_set_range_g25(struct hid_device *hid, u16 range);
 static ssize_t lg4ff_range_show(struct device *dev, struct device_attribute *attr, char *buf);
 static ssize_t lg4ff_range_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count);
 
-static DEVICE_ATTR(range, S_IRWXU | S_IRWXG | S_IRWXO, lg4ff_range_show, lg4ff_range_store);
+static DEVICE_ATTR(range, S_IRWXU | S_IRWXG | S_IROTH, lg4ff_range_show, lg4ff_range_store);
 
 struct lg4ff_device_entry {
 	__u32 product_id;
diff --git a/drivers/hid/hid-picolcd_fb.c b/drivers/hid/hid-picolcd_fb.c
index c930ab8554ea..7f965e231433 100644
--- a/drivers/hid/hid-picolcd_fb.c
+++ b/drivers/hid/hid-picolcd_fb.c
@@ -501,7 +501,7 @@ static ssize_t picolcd_fb_update_rate_store(struct device *dev,
 	return count;
 }
 
-static DEVICE_ATTR(fb_update_rate, 0666, picolcd_fb_update_rate_show,
+static DEVICE_ATTR(fb_update_rate, 0664, picolcd_fb_update_rate_show,
 		picolcd_fb_update_rate_store);
 
 /* initialize Framebuffer device */
diff --git a/drivers/hid/hid-rmi.c b/drivers/hid/hid-rmi.c
index 2451c7e5febd..578bbe65902b 100644
--- a/drivers/hid/hid-rmi.c
+++ b/drivers/hid/hid-rmi.c
@@ -428,6 +428,7 @@ static int rmi_raw_event(struct hid_device *hdev,
 	return 0;
 }
 
+#ifdef CONFIG_PM
 static int rmi_post_reset(struct hid_device *hdev)
 {
 	return rmi_set_mode(hdev, RMI_MODE_ATTN_REPORTS);
@@ -437,6 +438,7 @@ static int rmi_post_resume(struct hid_device *hdev)
 {
 	return rmi_set_mode(hdev, RMI_MODE_ATTN_REPORTS);
 }
+#endif /* CONFIG_PM */
 
 #define RMI4_MAX_PAGE 0xff
 #define RMI4_PAGE_SIZE 0x0100
diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c
index a8d5c8faf8cf..e244e449cbba 100644
--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c
@@ -159,17 +159,18 @@ int sensor_hub_register_callback(struct hid_sensor_hub_device *hsdev,
 {
 	struct hid_sensor_hub_callbacks_list *callback;
 	struct sensor_hub_data *pdata = hid_get_drvdata(hsdev->hdev);
+	unsigned long flags;
 
-	spin_lock(&pdata->dyn_callback_lock);
+	spin_lock_irqsave(&pdata->dyn_callback_lock, flags);
 	list_for_each_entry(callback, &pdata->dyn_callback_list, list)
 		if (callback->usage_id == usage_id &&
 						callback->hsdev == hsdev) {
-			spin_unlock(&pdata->dyn_callback_lock);
+			spin_unlock_irqrestore(&pdata->dyn_callback_lock, flags);
 			return -EINVAL;
 		}
 	callback = kzalloc(sizeof(*callback), GFP_ATOMIC);
 	if (!callback) {
-		spin_unlock(&pdata->dyn_callback_lock);
+		spin_unlock_irqrestore(&pdata->dyn_callback_lock, flags);
 		return -ENOMEM;
 	}
 	callback->hsdev = hsdev;
@@ -177,7 +178,7 @@ int sensor_hub_register_callback(struct hid_sensor_hub_device *hsdev,
 	callback->usage_id = usage_id;
 	callback->priv = NULL;
 	list_add_tail(&callback->list, &pdata->dyn_callback_list);
-	spin_unlock(&pdata->dyn_callback_lock);
+	spin_unlock_irqrestore(&pdata->dyn_callback_lock, flags);
 
 	return 0;
 }
@@ -188,8 +189,9 @@ int sensor_hub_remove_callback(struct hid_sensor_hub_device *hsdev,
 {
 	struct hid_sensor_hub_callbacks_list *callback;
 	struct sensor_hub_data *pdata = hid_get_drvdata(hsdev->hdev);
+	unsigned long flags;
 
-	spin_lock(&pdata->dyn_callback_lock);
+	spin_lock_irqsave(&pdata->dyn_callback_lock, flags);
 	list_for_each_entry(callback, &pdata->dyn_callback_list, list)
 		if (callback->usage_id == usage_id &&
 						callback->hsdev == hsdev) {
@@ -197,7 +199,7 @@ int sensor_hub_remove_callback(struct hid_sensor_hub_device *hsdev,
 			kfree(callback);
 			break;
 		}
-	spin_unlock(&pdata->dyn_callback_lock);
+	spin_unlock_irqrestore(&pdata->dyn_callback_lock, flags);
 
 	return 0;
 }
@@ -378,15 +380,16 @@ static int sensor_hub_suspend(struct hid_device *hdev, pm_message_t message)
 {
 	struct sensor_hub_data *pdata = hid_get_drvdata(hdev);
 	struct hid_sensor_hub_callbacks_list *callback;
+	unsigned long flags;
 
 	hid_dbg(hdev, " sensor_hub_suspend\n");
-	spin_lock(&pdata->dyn_callback_lock);
+	spin_lock_irqsave(&pdata->dyn_callback_lock, flags);
 	list_for_each_entry(callback, &pdata->dyn_callback_list, list) {
 		if (callback->usage_callback->suspend)
 			callback->usage_callback->suspend(
 					callback->hsdev, callback->priv);
 	}
-	spin_unlock(&pdata->dyn_callback_lock);
+	spin_unlock_irqrestore(&pdata->dyn_callback_lock, flags);
 
 	return 0;
 }
@@ -395,15 +398,16 @@ static int sensor_hub_resume(struct hid_device *hdev)
 {
 	struct sensor_hub_data *pdata = hid_get_drvdata(hdev);
 	struct hid_sensor_hub_callbacks_list *callback;
+	unsigned long flags;
 
 	hid_dbg(hdev, " sensor_hub_resume\n");
-	spin_lock(&pdata->dyn_callback_lock);
+	spin_lock_irqsave(&pdata->dyn_callback_lock, flags);
 	list_for_each_entry(callback, &pdata->dyn_callback_list, list) {
 		if (callback->usage_callback->resume)
 			callback->usage_callback->resume(
 					callback->hsdev, callback->priv);
 	}
-	spin_unlock(&pdata->dyn_callback_lock);
+	spin_unlock_irqrestore(&pdata->dyn_callback_lock, flags);
 
 	return 0;
 }
@@ -632,6 +636,7 @@ static int sensor_hub_probe(struct hid_device *hdev,
 			if (name == NULL) {
 				hid_err(hdev, "Failed MFD device name\n");
 					ret = -ENOMEM;
+					kfree(hsdev);
 					goto err_no_mem;
 			}
 			sd->hid_sensor_hub_client_devs[
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 59badc10a08c..31e6727cd009 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -49,6 +49,7 @@ static const struct hid_blacklist {
 
 	{ USB_VENDOR_ID_EMS, USB_DEVICE_ID_EMS_TRIO_LINKER_PLUS_II, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_ETURBOTOUCH, USB_DEVICE_ID_ETURBOTOUCH, HID_QUIRK_MULTI_INPUT },
+	{ USB_VENDOR_ID_ETURBOTOUCH, USB_DEVICE_ID_ETURBOTOUCH_2968, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_GREENASIA, USB_DEVICE_ID_GREENASIA_DUAL_USB_JOYPAD, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_PANTHERLORD, USB_DEVICE_ID_PANTHERLORD_TWIN_USB_JOYSTICK, HID_QUIRK_MULTI_INPUT | HID_QUIRK_SKIP_OUTPUT_REPORTS },
 	{ USB_VENDOR_ID_PLAYDOTCOM, USB_DEVICE_ID_PLAYDOTCOM_EMS_USBII, HID_QUIRK_MULTI_INPUT },
@@ -76,6 +77,8 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_MSI, USB_DEVICE_ID_MSI_GX680R_LED_PANEL, HID_QUIRK_NO_INIT_REPORTS },
 	{ USB_VENDOR_ID_NEXIO, USB_DEVICE_ID_NEXIO_MULTITOUCH_PTI0750, HID_QUIRK_NO_INIT_REPORTS },
 	{ USB_VENDOR_ID_NOVATEK, USB_DEVICE_ID_NOVATEK_MOUSE, HID_QUIRK_NO_INIT_REPORTS },
+	{ USB_VENDOR_ID_PENMOUNT, USB_DEVICE_ID_PENMOUNT_1610, HID_QUIRK_NOGET },
+	{ USB_VENDOR_ID_PENMOUNT, USB_DEVICE_ID_PENMOUNT_1640, HID_QUIRK_NOGET },
 	{ USB_VENDOR_ID_PIXART, USB_DEVICE_ID_PIXART_OPTICAL_TOUCH_SCREEN, HID_QUIRK_NO_INIT_REPORTS },
 	{ USB_VENDOR_ID_PIXART, USB_DEVICE_ID_PIXART_OPTICAL_TOUCH_SCREEN1, HID_QUIRK_NO_INIT_REPORTS },
 	{ USB_VENDOR_ID_PIXART, USB_DEVICE_ID_PIXART_OPTICAL_TOUCH_SCREEN2, HID_QUIRK_NO_INIT_REPORTS },
diff --git a/drivers/hsi/clients/Kconfig b/drivers/hsi/clients/Kconfig
index 71b9f9ab86e4..bc60dec3f586 100644
--- a/drivers/hsi/clients/Kconfig
+++ b/drivers/hsi/clients/Kconfig
@@ -15,7 +15,7 @@ config NOKIA_MODEM
 
 config SSI_PROTOCOL
 	tristate "SSI protocol"
-	depends on HSI && PHONET && (OMAP_SSI=y || OMAP_SSI=m)
+	depends on HSI && PHONET && OMAP_SSI
 	help
 	If you say Y here, you will enable the SSI protocol aka McSAAB.
 
diff --git a/drivers/hsi/controllers/omap_ssi_port.c b/drivers/hsi/controllers/omap_ssi_port.c
index b8693f0b27fe..29aea0b93360 100644
--- a/drivers/hsi/controllers/omap_ssi_port.c
+++ b/drivers/hsi/controllers/omap_ssi_port.c
@@ -1116,8 +1116,7 @@ static int __init ssi_port_probe(struct platform_device *pd)
 
 	dev_dbg(&pd->dev, "init ssi port...\n");
 
-	err = ref_module(THIS_MODULE, ssi->owner);
-	if (err) {
+	if (!try_module_get(ssi->owner)) {
 		dev_err(&pd->dev, "could not increment parent module refcount (err=%d)\n",
 			err);
 		return -ENODEV;
@@ -1254,6 +1253,7 @@ static int __exit ssi_port_remove(struct platform_device *pd)
 
 	omap_ssi->port[omap_port->port_id] = NULL;
 	platform_set_drvdata(pd, NULL);
+	module_put(ssi->owner);
 	pm_runtime_disable(&pd->dev);
 
 	return 0;
diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index 6c8b032cacba..ed9350d42764 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -404,7 +404,7 @@ static u32  next_vp;
  * performance critical channels (IDE, SCSI and Network) will be uniformly
  * distributed across all available CPUs.
  */
-static void init_vp_index(struct vmbus_channel *channel, uuid_le *type_guid)
+static void init_vp_index(struct vmbus_channel *channel, const uuid_le *type_guid)
 {
 	u32 cur_cpu;
 	int i;
diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c
index e84f4526eb36..ae22e3c1fc4c 100644
--- a/drivers/hv/connection.c
+++ b/drivers/hv/connection.c
@@ -339,9 +339,13 @@ static void process_chn_event(u32 relid)
 		 */
 
 		do {
-			hv_begin_read(&channel->inbound);
+			if (read_state)
+				hv_begin_read(&channel->inbound);
 			channel->onchannel_callback(arg);
-			bytes_to_read = hv_end_read(&channel->inbound);
+			if (read_state)
+				bytes_to_read = hv_end_read(&channel->inbound);
+			else
+				bytes_to_read = 0;
 		} while (read_state && (bytes_to_read != 0));
 	} else {
 		pr_err("no channel callback for relid - %u\n", relid);
diff --git a/drivers/hv/hv_kvp.c b/drivers/hv/hv_kvp.c
index ea852537307e..521c14625b3a 100644
--- a/drivers/hv/hv_kvp.c
+++ b/drivers/hv/hv_kvp.c
@@ -127,6 +127,17 @@ kvp_work_func(struct work_struct *dummy)
 	kvp_respond_to_host(NULL, HV_E_FAIL);
 }
 
+static void poll_channel(struct vmbus_channel *channel)
+{
+	if (channel->target_cpu != smp_processor_id())
+		smp_call_function_single(channel->target_cpu,
+					 hv_kvp_onchannelcallback,
+					 channel, true);
+	else
+		hv_kvp_onchannelcallback(channel);
+}
+
+
 static int kvp_handle_handshake(struct hv_kvp_msg *msg)
 {
 	int ret = 1;
@@ -155,7 +166,7 @@ static int kvp_handle_handshake(struct hv_kvp_msg *msg)
 		kvp_register(dm_reg_value);
 		kvp_transaction.active = false;
 		if (kvp_transaction.kvp_context)
-			hv_kvp_onchannelcallback(kvp_transaction.kvp_context);
+			poll_channel(kvp_transaction.kvp_context);
 	}
 	return ret;
 }
@@ -568,7 +579,7 @@ response_done:
 
 	vmbus_sendpacket(channel, recv_buffer, buf_len, req_id,
 				VM_PKT_DATA_INBAND, 0);
-
+	poll_channel(channel);
 }
 
 /*
@@ -603,7 +614,7 @@ void hv_kvp_onchannelcallback(void *context)
 		return;
 	}
 
-	vmbus_recvpacket(channel, recv_buffer, PAGE_SIZE * 2, &recvlen,
+	vmbus_recvpacket(channel, recv_buffer, PAGE_SIZE * 4, &recvlen,
 			 &requestid);
 
 	if (recvlen > 0) {
diff --git a/drivers/hv/hv_util.c b/drivers/hv/hv_util.c
index dd761806f0e8..3b9c9ef0deb8 100644
--- a/drivers/hv/hv_util.c
+++ b/drivers/hv/hv_util.c
@@ -319,7 +319,7 @@ static int util_probe(struct hv_device *dev,
 		(struct hv_util_service *)dev_id->driver_data;
 	int ret;
 
-	srv->recv_buffer = kmalloc(PAGE_SIZE * 2, GFP_KERNEL);
+	srv->recv_buffer = kmalloc(PAGE_SIZE * 4, GFP_KERNEL);
 	if (!srv->recv_buffer)
 		return -ENOMEM;
 	if (srv->util_init) {
diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h
index 18d1a8404cbc..22b750749a39 100644
--- a/drivers/hv/hyperv_vmbus.h
+++ b/drivers/hv/hyperv_vmbus.h
@@ -649,9 +649,9 @@ extern struct vmbus_connection vmbus_connection;
 
 /* General vmbus interface */
 
-struct hv_device *vmbus_device_create(uuid_le *type,
-					 uuid_le *instance,
-					 struct vmbus_channel *channel);
+struct hv_device *vmbus_device_create(const uuid_le *type,
+				      const uuid_le *instance,
+				      struct vmbus_channel *channel);
 
 int vmbus_device_register(struct hv_device *child_device_obj);
 void vmbus_device_unregister(struct hv_device *device_obj);
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 8e53a3c2607e..4d6b26979fbd 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -435,7 +435,7 @@ static int vmbus_uevent(struct device *device, struct kobj_uevent_env *env)
 	return ret;
 }
 
-static uuid_le null_guid;
+static const uuid_le null_guid;
 
 static inline bool is_null_guid(const __u8 *guid)
 {
@@ -450,7 +450,7 @@ static inline bool is_null_guid(const __u8 *guid)
  */
 static const struct hv_vmbus_device_id *hv_vmbus_get_id(
 					const struct hv_vmbus_device_id *id,
-					__u8 *guid)
+					const __u8 *guid)
 {
 	for (; !is_null_guid(id->guid); id++)
 		if (!memcmp(&id->guid, guid, sizeof(uuid_le)))
@@ -779,9 +779,9 @@ EXPORT_SYMBOL_GPL(vmbus_driver_unregister);
  * vmbus_device_create - Creates and registers a new child device
  * on the vmbus.
  */
-struct hv_device *vmbus_device_create(uuid_le *type,
-					    uuid_le *instance,
-					    struct vmbus_channel *channel)
+struct hv_device *vmbus_device_create(const uuid_le *type,
+				      const uuid_le *instance,
+				      struct vmbus_channel *channel)
 {
 	struct hv_device *child_device_obj;
 
diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 00343166feb1..02d3d85829f3 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -1052,7 +1052,7 @@ config SENSORS_PC87427
 	  will be called pc87427.
 
 config SENSORS_NTC_THERMISTOR
-	tristate "NTC thermistor support"
+	tristate "NTC thermistor support from Murata"
 	depends on !OF || IIO=n || IIO
 	help
 	  This driver supports NTC thermistors sensor reading and its
@@ -1060,7 +1060,8 @@ config SENSORS_NTC_THERMISTOR
 	  send notifications about the temperature.
 
 	  Currently, this driver supports
-	  NCP15WB473, NCP18WB473, NCP21WB473, NCP03WB473, and NCP15WL333.
+	  NCP15WB473, NCP18WB473, NCP21WB473, NCP03WB473, and NCP15WL333
+	  from Murata.
 
 	  This driver can also be built as a module.  If so, the module
 	  will be called ntc-thermistor.
@@ -1124,6 +1125,16 @@ config SENSORS_SHT21
 	  This driver can also be built as a module.  If so, the module
 	  will be called sht21.
 
+config SENSORS_SHTC1
+	tristate "Sensiron humidity and temperature sensors. SHTC1 and compat."
+	depends on I2C
+	help
+	  If you say yes here you get support for the Sensiron SHTC1 and SHTW1
+	  humidity and temperature sensors.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called shtc1.
+
 config SENSORS_S3C
 	tristate "Samsung built-in ADC"
 	depends on S3C_ADC
@@ -1166,6 +1177,7 @@ config SENSORS_DME1737
 config SENSORS_EMC1403
 	tristate "SMSC EMC1403/23 thermal sensor"
 	depends on I2C
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for the SMSC EMC1403/23
 	  temperature monitoring chip.
diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
index 11798ad7e801..3dc0f02f71d2 100644
--- a/drivers/hwmon/Makefile
+++ b/drivers/hwmon/Makefile
@@ -126,6 +126,7 @@ obj-$(CONFIG_SENSORS_SCH5627)	+= sch5627.o
 obj-$(CONFIG_SENSORS_SCH5636)	+= sch5636.o
 obj-$(CONFIG_SENSORS_SHT15)	+= sht15.o
 obj-$(CONFIG_SENSORS_SHT21)	+= sht21.o
+obj-$(CONFIG_SENSORS_SHTC1)	+= shtc1.o
 obj-$(CONFIG_SENSORS_SIS5595)	+= sis5595.o
 obj-$(CONFIG_SENSORS_SMM665)	+= smm665.o
 obj-$(CONFIG_SENSORS_SMSC47B397)+= smsc47b397.o
diff --git a/drivers/hwmon/adc128d818.c b/drivers/hwmon/adc128d818.c
index 5ffd81f19d01..0625e50d7a6e 100644
--- a/drivers/hwmon/adc128d818.c
+++ b/drivers/hwmon/adc128d818.c
@@ -239,50 +239,50 @@ static ssize_t adc128_show_alarm(struct device *dev,
 	return sprintf(buf, "%u\n", !!(alarms & mask));
 }
 
-static SENSOR_DEVICE_ATTR_2(in0_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 0, 0);
+static SENSOR_DEVICE_ATTR_2(in0_input, S_IRUGO,
+			    adc128_show_in, NULL, 0, 0);
 static SENSOR_DEVICE_ATTR_2(in0_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 0, 1);
 static SENSOR_DEVICE_ATTR_2(in0_max, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 0, 2);
 
-static SENSOR_DEVICE_ATTR_2(in1_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 1, 0);
+static SENSOR_DEVICE_ATTR_2(in1_input, S_IRUGO,
+			    adc128_show_in, NULL, 1, 0);
 static SENSOR_DEVICE_ATTR_2(in1_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 1, 1);
 static SENSOR_DEVICE_ATTR_2(in1_max, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 1, 2);
 
-static SENSOR_DEVICE_ATTR_2(in2_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 2, 0);
+static SENSOR_DEVICE_ATTR_2(in2_input, S_IRUGO,
+			    adc128_show_in, NULL, 2, 0);
 static SENSOR_DEVICE_ATTR_2(in2_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 2, 1);
 static SENSOR_DEVICE_ATTR_2(in2_max, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 2, 2);
 
-static SENSOR_DEVICE_ATTR_2(in3_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 3, 0);
+static SENSOR_DEVICE_ATTR_2(in3_input, S_IRUGO,
+			    adc128_show_in, NULL, 3, 0);
 static SENSOR_DEVICE_ATTR_2(in3_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 3, 1);
 static SENSOR_DEVICE_ATTR_2(in3_max, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 3, 2);
 
-static SENSOR_DEVICE_ATTR_2(in4_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 4, 0);
+static SENSOR_DEVICE_ATTR_2(in4_input, S_IRUGO,
+			    adc128_show_in, NULL, 4, 0);
 static SENSOR_DEVICE_ATTR_2(in4_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 4, 1);
 static SENSOR_DEVICE_ATTR_2(in4_max, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 4, 2);
 
-static SENSOR_DEVICE_ATTR_2(in5_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 5, 0);
+static SENSOR_DEVICE_ATTR_2(in5_input, S_IRUGO,
+			    adc128_show_in, NULL, 5, 0);
 static SENSOR_DEVICE_ATTR_2(in5_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 5, 1);
 static SENSOR_DEVICE_ATTR_2(in5_max, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 5, 2);
 
-static SENSOR_DEVICE_ATTR_2(in6_input, S_IWUSR | S_IRUGO,
-			    adc128_show_in, adc128_set_in, 6, 0);
+static SENSOR_DEVICE_ATTR_2(in6_input, S_IRUGO,
+			    adc128_show_in, NULL, 6, 0);
 static SENSOR_DEVICE_ATTR_2(in6_min, S_IWUSR | S_IRUGO,
 			    adc128_show_in, adc128_set_in, 6, 1);
 static SENSOR_DEVICE_ATTR_2(in6_max, S_IWUSR | S_IRUGO,
diff --git a/drivers/hwmon/adm1021.c b/drivers/hwmon/adm1021.c
index 3eb4281689b5..d74241bb278c 100644
--- a/drivers/hwmon/adm1021.c
+++ b/drivers/hwmon/adm1021.c
@@ -185,7 +185,7 @@ static ssize_t set_temp_max(struct device *dev,
 	struct adm1021_data *data = dev_get_drvdata(dev);
 	struct i2c_client *client = data->client;
 	long temp;
-	int err;
+	int reg_val, err;
 
 	err = kstrtol(buf, 10, &temp);
 	if (err)
@@ -193,10 +193,11 @@ static ssize_t set_temp_max(struct device *dev,
 	temp /= 1000;
 
 	mutex_lock(&data->update_lock);
-	data->temp_max[index] = clamp_val(temp, -128, 127);
+	reg_val = clamp_val(temp, -128, 127);
+	data->temp_max[index] = reg_val * 1000;
 	if (!read_only)
 		i2c_smbus_write_byte_data(client, ADM1021_REG_TOS_W(index),
-					  data->temp_max[index]);
+					  reg_val);
 	mutex_unlock(&data->update_lock);
 
 	return count;
@@ -210,7 +211,7 @@ static ssize_t set_temp_min(struct device *dev,
 	struct adm1021_data *data = dev_get_drvdata(dev);
 	struct i2c_client *client = data->client;
 	long temp;
-	int err;
+	int reg_val, err;
 
 	err = kstrtol(buf, 10, &temp);
 	if (err)
@@ -218,10 +219,11 @@ static ssize_t set_temp_min(struct device *dev,
 	temp /= 1000;
 
 	mutex_lock(&data->update_lock);
-	data->temp_min[index] = clamp_val(temp, -128, 127);
+	reg_val = clamp_val(temp, -128, 127);
+	data->temp_min[index] = reg_val * 1000;
 	if (!read_only)
 		i2c_smbus_write_byte_data(client, ADM1021_REG_THYST_W(index),
-					  data->temp_min[index]);
+					  reg_val);
 	mutex_unlock(&data->update_lock);
 
 	return count;
diff --git a/drivers/hwmon/adm1029.c b/drivers/hwmon/adm1029.c
index 78339e880bd6..2804571b269e 100644
--- a/drivers/hwmon/adm1029.c
+++ b/drivers/hwmon/adm1029.c
@@ -232,6 +232,9 @@ static ssize_t set_fan_div(struct device *dev,
 	/* Update the value */
 	reg = (reg & 0x3F) | (val << 6);
 
+	/* Update the cache */
+	data->fan_div[attr->index] = reg;
+
 	/* Write value */
 	i2c_smbus_write_byte_data(client,
 				  ADM1029_REG_FAN_DIV[attr->index], reg);
diff --git a/drivers/hwmon/adm1031.c b/drivers/hwmon/adm1031.c
index a8a540ca8c34..51c1a5a165ab 100644
--- a/drivers/hwmon/adm1031.c
+++ b/drivers/hwmon/adm1031.c
@@ -365,6 +365,7 @@ set_auto_temp_min(struct device *dev, struct device_attribute *attr,
 	if (ret)
 		return ret;
 
+	val = clamp_val(val, 0, 127000);
 	mutex_lock(&data->update_lock);
 	data->auto_temp[nr] = AUTO_TEMP_MIN_TO_REG(val, data->auto_temp[nr]);
 	adm1031_write_value(client, ADM1031_REG_AUTO_TEMP(nr),
@@ -394,6 +395,7 @@ set_auto_temp_max(struct device *dev, struct device_attribute *attr,
 	if (ret)
 		return ret;
 
+	val = clamp_val(val, 0, 127000);
 	mutex_lock(&data->update_lock);
 	data->temp_max[nr] = AUTO_TEMP_MAX_TO_REG(val, data->auto_temp[nr],
 						  data->pwm[nr]);
@@ -696,7 +698,7 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *attr,
 	if (ret)
 		return ret;
 
-	val = clamp_val(val, -55000, nr == 0 ? 127750 : 127875);
+	val = clamp_val(val, -55000, 127000);
 	mutex_lock(&data->update_lock);
 	data->temp_min[nr] = TEMP_TO_REG(val);
 	adm1031_write_value(client, ADM1031_REG_TEMP_MIN(nr),
@@ -717,7 +719,7 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *attr,
 	if (ret)
 		return ret;
 
-	val = clamp_val(val, -55000, nr == 0 ? 127750 : 127875);
+	val = clamp_val(val, -55000, 127000);
 	mutex_lock(&data->update_lock);
 	data->temp_max[nr] = TEMP_TO_REG(val);
 	adm1031_write_value(client, ADM1031_REG_TEMP_MAX(nr),
@@ -738,7 +740,7 @@ static ssize_t set_temp_crit(struct device *dev, struct device_attribute *attr,
 	if (ret)
 		return ret;
 
-	val = clamp_val(val, -55000, nr == 0 ? 127750 : 127875);
+	val = clamp_val(val, -55000, 127000);
 	mutex_lock(&data->update_lock);
 	data->temp_crit[nr] = TEMP_TO_REG(val);
 	adm1031_write_value(client, ADM1031_REG_TEMP_CRIT(nr),
diff --git a/drivers/hwmon/amc6821.c b/drivers/hwmon/amc6821.c
index eea817296513..9f2be3dd28f3 100644
--- a/drivers/hwmon/amc6821.c
+++ b/drivers/hwmon/amc6821.c
@@ -704,7 +704,7 @@ static SENSOR_DEVICE_ATTR(temp1_max_alarm, S_IRUGO,
 	get_temp_alarm, NULL, IDX_TEMP1_MAX);
 static SENSOR_DEVICE_ATTR(temp1_crit_alarm, S_IRUGO,
 	get_temp_alarm, NULL, IDX_TEMP1_CRIT);
-static SENSOR_DEVICE_ATTR(temp2_input, S_IRUGO | S_IWUSR,
+static SENSOR_DEVICE_ATTR(temp2_input, S_IRUGO,
 	get_temp, NULL, IDX_TEMP2_INPUT);
 static SENSOR_DEVICE_ATTR(temp2_min, S_IRUGO | S_IWUSR, get_temp,
 	set_temp, IDX_TEMP2_MIN);
diff --git a/drivers/hwmon/atxp1.c b/drivers/hwmon/atxp1.c
index 6edce42c61d5..2ae8a304b5ef 100644
--- a/drivers/hwmon/atxp1.c
+++ b/drivers/hwmon/atxp1.c
@@ -45,30 +45,6 @@ MODULE_AUTHOR("Sebastian Witt <se.witt@gmx.net>");
 
 static const unsigned short normal_i2c[] = { 0x37, 0x4e, I2C_CLIENT_END };
 
-static int atxp1_probe(struct i2c_client *client,
-		       const struct i2c_device_id *id);
-static int atxp1_remove(struct i2c_client *client);
-static struct atxp1_data *atxp1_update_device(struct device *dev);
-static int atxp1_detect(struct i2c_client *client, struct i2c_board_info *info);
-
-static const struct i2c_device_id atxp1_id[] = {
-	{ "atxp1", 0 },
-	{ }
-};
-MODULE_DEVICE_TABLE(i2c, atxp1_id);
-
-static struct i2c_driver atxp1_driver = {
-	.class		= I2C_CLASS_HWMON,
-	.driver = {
-		.name	= "atxp1",
-	},
-	.probe		= atxp1_probe,
-	.remove		= atxp1_remove,
-	.id_table	= atxp1_id,
-	.detect		= atxp1_detect,
-	.address_list	= normal_i2c,
-};
-
 struct atxp1_data {
 	struct device *hwmon_dev;
 	struct mutex update_lock;
@@ -386,4 +362,22 @@ static int atxp1_remove(struct i2c_client *client)
 	return 0;
 };
 
+static const struct i2c_device_id atxp1_id[] = {
+	{ "atxp1", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, atxp1_id);
+
+static struct i2c_driver atxp1_driver = {
+	.class		= I2C_CLASS_HWMON,
+	.driver = {
+		.name	= "atxp1",
+	},
+	.probe		= atxp1_probe,
+	.remove		= atxp1_remove,
+	.id_table	= atxp1_id,
+	.detect		= atxp1_detect,
+	.address_list	= normal_i2c,
+};
+
 module_i2c_driver(atxp1_driver);
diff --git a/drivers/hwmon/emc2103.c b/drivers/hwmon/emc2103.c
index fd892dd48e4c..78002de46cb6 100644
--- a/drivers/hwmon/emc2103.c
+++ b/drivers/hwmon/emc2103.c
@@ -250,9 +250,7 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *da,
 	if (result < 0)
 		return result;
 
-	val = DIV_ROUND_CLOSEST(val, 1000);
-	if ((val < -63) || (val > 127))
-		return -EINVAL;
+	val = clamp_val(DIV_ROUND_CLOSEST(val, 1000), -63, 127);
 
 	mutex_lock(&data->update_lock);
 	data->temp_min[nr] = val;
@@ -274,9 +272,7 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *da,
 	if (result < 0)
 		return result;
 
-	val = DIV_ROUND_CLOSEST(val, 1000);
-	if ((val < -63) || (val > 127))
-		return -EINVAL;
+	val = clamp_val(DIV_ROUND_CLOSEST(val, 1000), -63, 127);
 
 	mutex_lock(&data->update_lock);
 	data->temp_max[nr] = val;
@@ -390,15 +386,14 @@ static ssize_t set_fan_target(struct device *dev, struct device_attribute *da,
 {
 	struct emc2103_data *data = emc2103_update_device(dev);
 	struct i2c_client *client = to_i2c_client(dev);
-	long rpm_target;
+	unsigned long rpm_target;
 
-	int result = kstrtol(buf, 10, &rpm_target);
+	int result = kstrtoul(buf, 10, &rpm_target);
 	if (result < 0)
 		return result;
 
 	/* Datasheet states 16384 as maximum RPM target (table 3.2) */
-	if ((rpm_target < 0) || (rpm_target > 16384))
-		return -EINVAL;
+	rpm_target = clamp_val(rpm_target, 0, 16384);
 
 	mutex_lock(&data->update_lock);
 
diff --git a/drivers/hwmon/gpio-fan.c b/drivers/hwmon/gpio-fan.c
index ba35e4d530b5..2566c43dd1e9 100644
--- a/drivers/hwmon/gpio-fan.c
+++ b/drivers/hwmon/gpio-fan.c
@@ -538,7 +538,7 @@ static int gpio_fan_probe(struct platform_device *pdev)
 
 	/* Make this driver part of hwmon class. */
 	fan_data->hwmon_dev = hwmon_device_register_with_groups(&pdev->dev,
-						"gpio-fan", fan_data,
+						"gpio_fan", fan_data,
 						gpio_fan_groups);
 	if (IS_ERR(fan_data->hwmon_dev))
 		return PTR_ERR(fan_data->hwmon_dev);
diff --git a/drivers/hwmon/ina2xx.c b/drivers/hwmon/ina2xx.c
index 93d26e8af3e2..bfd3f3eeabcd 100644
--- a/drivers/hwmon/ina2xx.c
+++ b/drivers/hwmon/ina2xx.c
@@ -148,7 +148,8 @@ static int ina2xx_get_value(struct ina2xx_data *data, u8 reg)
 
 	switch (reg) {
 	case INA2XX_SHUNT_VOLTAGE:
-		val = DIV_ROUND_CLOSEST(data->regs[reg],
+		/* signed register */
+		val = DIV_ROUND_CLOSEST((s16)data->regs[reg],
 					data->config->shunt_div);
 		break;
 	case INA2XX_BUS_VOLTAGE:
@@ -160,8 +161,8 @@ static int ina2xx_get_value(struct ina2xx_data *data, u8 reg)
 		val = data->regs[reg] * data->config->power_lsb;
 		break;
 	case INA2XX_CURRENT:
-		/* LSB=1mA (selected). Is in mA */
-		val = data->regs[reg];
+		/* signed register, LSB=1mA (selected), in mA */
+		val = (s16)data->regs[reg];
 		break;
 	default:
 		/* programmer goofed */
diff --git a/drivers/hwmon/lm85.c b/drivers/hwmon/lm85.c
index bed4af358308..b0129a54e1a6 100644
--- a/drivers/hwmon/lm85.c
+++ b/drivers/hwmon/lm85.c
@@ -5,7 +5,7 @@
  * Copyright (c) 2002, 2003  Philip Pokorny <ppokorny@penguincomputing.com>
  * Copyright (c) 2003        Margit Schubert-While <margitsw@t-online.de>
  * Copyright (c) 2004        Justin Thiessen <jthiessen@penguincomputing.com>
- * Copyright (C) 2007--2009  Jean Delvare <jdelvare@suse.de>
+ * Copyright (C) 2007--2014  Jean Delvare <jdelvare@suse.de>
  *
  * Chip details at	      <http://www.national.com/ds/LM/LM85.pdf>
  *
@@ -39,7 +39,7 @@
 static const unsigned short normal_i2c[] = { 0x2c, 0x2d, 0x2e, I2C_CLIENT_END };
 
 enum chips {
-	any_chip, lm85b, lm85c,
+	lm85,
 	adm1027, adt7463, adt7468,
 	emc6d100, emc6d102, emc6d103, emc6d103s
 };
@@ -75,9 +75,6 @@ enum chips {
 #define LM85_COMPANY_NATIONAL		0x01
 #define LM85_COMPANY_ANALOG_DEV		0x41
 #define LM85_COMPANY_SMSC		0x5c
-#define LM85_VERSTEP_VMASK              0xf0
-#define LM85_VERSTEP_GENERIC		0x60
-#define LM85_VERSTEP_GENERIC2		0x70
 #define LM85_VERSTEP_LM85C		0x60
 #define LM85_VERSTEP_LM85B		0x62
 #define LM85_VERSTEP_LM96000_1		0x68
@@ -351,9 +348,9 @@ static const struct i2c_device_id lm85_id[] = {
 	{ "adm1027", adm1027 },
 	{ "adt7463", adt7463 },
 	{ "adt7468", adt7468 },
-	{ "lm85", any_chip },
-	{ "lm85b", lm85b },
-	{ "lm85c", lm85c },
+	{ "lm85", lm85 },
+	{ "lm85b", lm85 },
+	{ "lm85c", lm85 },
 	{ "emc6d100", emc6d100 },
 	{ "emc6d101", emc6d100 },
 	{ "emc6d102", emc6d102 },
@@ -1281,7 +1278,7 @@ static int lm85_detect(struct i2c_client *client, struct i2c_board_info *info)
 {
 	struct i2c_adapter *adapter = client->adapter;
 	int address = client->addr;
-	const char *type_name;
+	const char *type_name = NULL;
 	int company, verstep;
 
 	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) {
@@ -1297,16 +1294,6 @@ static int lm85_detect(struct i2c_client *client, struct i2c_board_info *info)
 		"Detecting device at 0x%02x with COMPANY: 0x%02x and VERSTEP: 0x%02x\n",
 		address, company, verstep);
 
-	/* All supported chips have the version in common */
-	if ((verstep & LM85_VERSTEP_VMASK) != LM85_VERSTEP_GENERIC &&
-	    (verstep & LM85_VERSTEP_VMASK) != LM85_VERSTEP_GENERIC2) {
-		dev_dbg(&adapter->dev,
-			"Autodetection failed: unsupported version\n");
-		return -ENODEV;
-	}
-	type_name = "lm85";
-
-	/* Now, refine the detection */
 	if (company == LM85_COMPANY_NATIONAL) {
 		switch (verstep) {
 		case LM85_VERSTEP_LM85C:
@@ -1323,6 +1310,7 @@ static int lm85_detect(struct i2c_client *client, struct i2c_board_info *info)
 					"Found Winbond WPCD377I, ignoring\n");
 				return -ENODEV;
 			}
+			type_name = "lm85";
 			break;
 		}
 	} else if (company == LM85_COMPANY_ANALOG_DEV) {
@@ -1357,12 +1345,11 @@ static int lm85_detect(struct i2c_client *client, struct i2c_board_info *info)
 			type_name = "emc6d103s";
 			break;
 		}
-	} else {
-		dev_dbg(&adapter->dev,
-			"Autodetection failed: unknown vendor\n");
-		return -ENODEV;
 	}
 
+	if (!type_name)
+		return -ENODEV;
+
 	strlcpy(info->type, type_name, I2C_NAME_SIZE);
 
 	return 0;
diff --git a/drivers/hwmon/ltc4151.c b/drivers/hwmon/ltc4151.c
index af81be1237c9..c86a18402496 100644
--- a/drivers/hwmon/ltc4151.c
+++ b/drivers/hwmon/ltc4151.c
@@ -47,7 +47,7 @@
 #define LTC4151_ADIN_L	0x05
 
 struct ltc4151_data {
-	struct device *hwmon_dev;
+	struct i2c_client *client;
 
 	struct mutex update_lock;
 	bool valid;
@@ -59,8 +59,8 @@ struct ltc4151_data {
 
 static struct ltc4151_data *ltc4151_update_device(struct device *dev)
 {
-	struct i2c_client *client = to_i2c_client(dev);
-	struct ltc4151_data *data = i2c_get_clientdata(client);
+	struct ltc4151_data *data = dev_get_drvdata(dev);
+	struct i2c_client *client = data->client;
 	struct ltc4151_data *ret = data;
 
 	mutex_lock(&data->update_lock);
@@ -159,7 +159,7 @@ static SENSOR_DEVICE_ATTR(curr1_input, S_IRUGO, ltc4151_show_value, NULL,
  * Finally, construct an array of pointers to members of the above objects,
  * as required for sysfs_create_group()
  */
-static struct attribute *ltc4151_attributes[] = {
+static struct attribute *ltc4151_attrs[] = {
 	&sensor_dev_attr_in1_input.dev_attr.attr,
 	&sensor_dev_attr_in2_input.dev_attr.attr,
 
@@ -167,54 +167,30 @@ static struct attribute *ltc4151_attributes[] = {
 
 	NULL,
 };
-
-static const struct attribute_group ltc4151_group = {
-	.attrs = ltc4151_attributes,
-};
+ATTRIBUTE_GROUPS(ltc4151);
 
 static int ltc4151_probe(struct i2c_client *client,
 			 const struct i2c_device_id *id)
 {
 	struct i2c_adapter *adapter = client->adapter;
+	struct device *dev = &client->dev;
 	struct ltc4151_data *data;
-	int ret;
+	struct device *hwmon_dev;
 
 	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA))
 		return -ENODEV;
 
-	data = devm_kzalloc(&client->dev, sizeof(*data), GFP_KERNEL);
+	data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 
-	i2c_set_clientdata(client, data);
+	data->client = client;
 	mutex_init(&data->update_lock);
 
-	/* Register sysfs hooks */
-	ret = sysfs_create_group(&client->dev.kobj, &ltc4151_group);
-	if (ret)
-		return ret;
-
-	data->hwmon_dev = hwmon_device_register(&client->dev);
-	if (IS_ERR(data->hwmon_dev)) {
-		ret = PTR_ERR(data->hwmon_dev);
-		goto out_hwmon_device_register;
-	}
-
-	return 0;
-
-out_hwmon_device_register:
-	sysfs_remove_group(&client->dev.kobj, &ltc4151_group);
-	return ret;
-}
-
-static int ltc4151_remove(struct i2c_client *client)
-{
-	struct ltc4151_data *data = i2c_get_clientdata(client);
-
-	hwmon_device_unregister(data->hwmon_dev);
-	sysfs_remove_group(&client->dev.kobj, &ltc4151_group);
-
-	return 0;
+	hwmon_dev = devm_hwmon_device_register_with_groups(dev, client->name,
+							   data,
+							   ltc4151_groups);
+	return PTR_ERR_OR_ZERO(hwmon_dev);
 }
 
 static const struct i2c_device_id ltc4151_id[] = {
@@ -229,7 +205,6 @@ static struct i2c_driver ltc4151_driver = {
 		.name	= "ltc4151",
 	},
 	.probe		= ltc4151_probe,
-	.remove		= ltc4151_remove,
 	.id_table	= ltc4151_id,
 };
 
diff --git a/drivers/hwmon/ntc_thermistor.c b/drivers/hwmon/ntc_thermistor.c
index e76feb86a1d4..ae66f42c4d6d 100644
--- a/drivers/hwmon/ntc_thermistor.c
+++ b/drivers/hwmon/ntc_thermistor.c
@@ -163,6 +163,18 @@ static int ntc_adc_iio_read(struct ntc_thermistor_platform_data *pdata)
 }
 
 static const struct of_device_id ntc_match[] = {
+	{ .compatible = "murata,ncp15wb473",
+		.data = &ntc_thermistor_id[0] },
+	{ .compatible = "murata,ncp18wb473",
+		.data = &ntc_thermistor_id[1] },
+	{ .compatible = "murata,ncp21wb473",
+		.data = &ntc_thermistor_id[2] },
+	{ .compatible = "murata,ncp03wb473",
+		.data = &ntc_thermistor_id[3] },
+	{ .compatible = "murata,ncp15wl333",
+		.data = &ntc_thermistor_id[4] },
+
+	/* Usage of vendor name "ntc" is deprecated */
 	{ .compatible = "ntc,ncp15wb473",
 		.data = &ntc_thermistor_id[0] },
 	{ .compatible = "ntc,ncp18wb473",
@@ -500,7 +512,7 @@ static int ntc_thermistor_probe(struct platform_device *pdev)
 	}
 
 	dev_info(&pdev->dev, "Thermistor type: %s successfully probed.\n",
-								pdev->name);
+								pdev_id->name);
 
 	return 0;
 err_after_sysfs:
@@ -534,7 +546,7 @@ static struct platform_driver ntc_thermistor_driver = {
 
 module_platform_driver(ntc_thermistor_driver);
 
-MODULE_DESCRIPTION("NTC Thermistor Driver");
+MODULE_DESCRIPTION("NTC Thermistor Driver from Murata");
 MODULE_AUTHOR("MyungJoo Ham <myungjoo.ham@samsung.com>");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:ntc-thermistor");
diff --git a/drivers/hwmon/shtc1.c b/drivers/hwmon/shtc1.c
new file mode 100644
index 000000000000..decd7df995ab
--- /dev/null
+++ b/drivers/hwmon/shtc1.c
@@ -0,0 +1,251 @@
+/* Sensirion SHTC1 humidity and temperature sensor driver
+ *
+ * Copyright (C) 2014 Sensirion AG, Switzerland
+ * Author: Johannes Winkelmann <johannes.winkelmann@sensirion.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/i2c.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#include <linux/err.h>
+#include <linux/delay.h>
+#include <linux/platform_data/shtc1.h>
+
+/* commands (high precision mode) */
+static const unsigned char shtc1_cmd_measure_blocking_hpm[]    = { 0x7C, 0xA2 };
+static const unsigned char shtc1_cmd_measure_nonblocking_hpm[] = { 0x78, 0x66 };
+
+/* commands (low precision mode) */
+static const unsigned char shtc1_cmd_measure_blocking_lpm[]    = { 0x64, 0x58 };
+static const unsigned char shtc1_cmd_measure_nonblocking_lpm[] = { 0x60, 0x9c };
+
+/* command for reading the ID register */
+static const unsigned char shtc1_cmd_read_id_reg[]	       = { 0xef, 0xc8 };
+
+/* constants for reading the ID register */
+#define SHTC1_ID	  0x07
+#define SHTC1_ID_REG_MASK 0x1f
+
+/* delays for non-blocking i2c commands, both in us */
+#define SHTC1_NONBLOCKING_WAIT_TIME_HPM  14400
+#define SHTC1_NONBLOCKING_WAIT_TIME_LPM   1000
+
+#define SHTC1_CMD_LENGTH      2
+#define SHTC1_RESPONSE_LENGTH 6
+
+struct shtc1_data {
+	struct i2c_client *client;
+	struct mutex update_lock;
+	bool valid;
+	unsigned long last_updated; /* in jiffies */
+
+	const unsigned char *command;
+	unsigned int nonblocking_wait_time; /* in us */
+
+	struct shtc1_platform_data setup;
+
+	int temperature; /* 1000 * temperature in dgr C */
+	int humidity; /* 1000 * relative humidity in %RH */
+};
+
+static int shtc1_update_values(struct i2c_client *client,
+			       struct shtc1_data *data,
+			       char *buf, int bufsize)
+{
+	int ret = i2c_master_send(client, data->command, SHTC1_CMD_LENGTH);
+	if (ret != SHTC1_CMD_LENGTH) {
+		dev_err(&client->dev, "failed to send command: %d\n", ret);
+		return ret < 0 ? ret : -EIO;
+	}
+
+	/*
+	 * In blocking mode (clock stretching mode) the I2C bus
+	 * is blocked for other traffic, thus the call to i2c_master_recv()
+	 * will wait until the data is ready. For non blocking mode, we
+	 * have to wait ourselves.
+	 */
+	if (!data->setup.blocking_io)
+		usleep_range(data->nonblocking_wait_time,
+			     data->nonblocking_wait_time + 1000);
+
+	ret = i2c_master_recv(client, buf, bufsize);
+	if (ret != bufsize) {
+		dev_err(&client->dev, "failed to read values: %d\n", ret);
+		return ret < 0 ? ret : -EIO;
+	}
+
+	return 0;
+}
+
+/* sysfs attributes */
+static struct shtc1_data *shtc1_update_client(struct device *dev)
+{
+	struct shtc1_data *data = dev_get_drvdata(dev);
+	struct i2c_client *client = data->client;
+	unsigned char buf[SHTC1_RESPONSE_LENGTH];
+	int val;
+	int ret = 0;
+
+	mutex_lock(&data->update_lock);
+
+	if (time_after(jiffies, data->last_updated + HZ / 10) || !data->valid) {
+		ret = shtc1_update_values(client, data, buf, sizeof(buf));
+		if (ret)
+			goto out;
+
+		/*
+		 * From datasheet:
+		 * T = -45 + 175 * ST / 2^16
+		 * RH = 100 * SRH / 2^16
+		 *
+		 * Adapted for integer fixed point (3 digit) arithmetic.
+		 */
+		val = be16_to_cpup((__be16 *)buf);
+		data->temperature = ((21875 * val) >> 13) - 45000;
+		val = be16_to_cpup((__be16 *)(buf + 3));
+		data->humidity = ((12500 * val) >> 13);
+
+		data->last_updated = jiffies;
+		data->valid = true;
+	}
+
+out:
+	mutex_unlock(&data->update_lock);
+
+	return ret == 0 ? data : ERR_PTR(ret);
+}
+
+static ssize_t temp1_input_show(struct device *dev,
+				struct device_attribute *attr,
+				char *buf)
+{
+	struct shtc1_data *data = shtc1_update_client(dev);
+	if (IS_ERR(data))
+		return PTR_ERR(data);
+
+	return sprintf(buf, "%d\n", data->temperature);
+}
+
+static ssize_t humidity1_input_show(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	struct shtc1_data *data = shtc1_update_client(dev);
+	if (IS_ERR(data))
+		return PTR_ERR(data);
+
+	return sprintf(buf, "%d\n", data->humidity);
+}
+
+static DEVICE_ATTR_RO(temp1_input);
+static DEVICE_ATTR_RO(humidity1_input);
+
+static struct attribute *shtc1_attrs[] = {
+	&dev_attr_temp1_input.attr,
+	&dev_attr_humidity1_input.attr,
+	NULL
+};
+
+ATTRIBUTE_GROUPS(shtc1);
+
+static void shtc1_select_command(struct shtc1_data *data)
+{
+	if (data->setup.high_precision) {
+		data->command = data->setup.blocking_io ?
+				shtc1_cmd_measure_blocking_hpm :
+				shtc1_cmd_measure_nonblocking_hpm;
+		data->nonblocking_wait_time = SHTC1_NONBLOCKING_WAIT_TIME_HPM;
+
+	} else {
+		data->command = data->setup.blocking_io ?
+				shtc1_cmd_measure_blocking_lpm :
+				shtc1_cmd_measure_nonblocking_lpm;
+		data->nonblocking_wait_time = SHTC1_NONBLOCKING_WAIT_TIME_LPM;
+	}
+}
+
+static int shtc1_probe(struct i2c_client *client,
+		       const struct i2c_device_id *id)
+{
+	int ret;
+	char id_reg[2];
+	struct shtc1_data *data;
+	struct device *hwmon_dev;
+	struct i2c_adapter *adap = client->adapter;
+	struct device *dev = &client->dev;
+
+	if (!i2c_check_functionality(adap, I2C_FUNC_I2C)) {
+		dev_err(dev, "plain i2c transactions not supported\n");
+		return -ENODEV;
+	}
+
+	ret = i2c_master_send(client, shtc1_cmd_read_id_reg, SHTC1_CMD_LENGTH);
+	if (ret != SHTC1_CMD_LENGTH) {
+		dev_err(dev, "could not send read_id_reg command: %d\n", ret);
+		return ret < 0 ? ret : -ENODEV;
+	}
+	ret = i2c_master_recv(client, id_reg, sizeof(id_reg));
+	if (ret != sizeof(id_reg)) {
+		dev_err(dev, "could not read ID register: %d\n", ret);
+		return -ENODEV;
+	}
+	if ((id_reg[1] & SHTC1_ID_REG_MASK) != SHTC1_ID) {
+		dev_err(dev, "ID register doesn't match\n");
+		return -ENODEV;
+	}
+
+	data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	data->setup.blocking_io = false;
+	data->setup.high_precision = true;
+	data->client = client;
+
+	if (client->dev.platform_data)
+		data->setup = *(struct shtc1_platform_data *)dev->platform_data;
+	shtc1_select_command(data);
+	mutex_init(&data->update_lock);
+
+	hwmon_dev = devm_hwmon_device_register_with_groups(dev,
+							   client->name,
+							   data,
+							   shtc1_groups);
+	if (IS_ERR(hwmon_dev))
+		dev_dbg(dev, "unable to register hwmon device\n");
+
+	return PTR_ERR_OR_ZERO(hwmon_dev);
+}
+
+/* device ID table */
+static const struct i2c_device_id shtc1_id[] = {
+	{ "shtc1", 0 },
+	{ "shtw1", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, shtc1_id);
+
+static struct i2c_driver shtc1_i2c_driver = {
+	.driver.name  = "shtc1",
+	.probe        = shtc1_probe,
+	.id_table     = shtc1_id,
+};
+
+module_i2c_driver(shtc1_i2c_driver);
+
+MODULE_AUTHOR("Johannes Winkelmann <johannes.winkelmann@sensirion.com>");
+MODULE_DESCRIPTION("Sensirion SHTC1 humidity and temperature sensor driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/hwmon/vexpress.c b/drivers/hwmon/vexpress.c
index 611f34c7333d..c53619086f33 100644
--- a/drivers/hwmon/vexpress.c
+++ b/drivers/hwmon/vexpress.c
@@ -27,17 +27,8 @@
 struct vexpress_hwmon_data {
 	struct device *hwmon_dev;
 	struct regmap *reg;
-	const char *name;
 };
 
-static ssize_t vexpress_hwmon_name_show(struct device *dev,
-		struct device_attribute *dev_attr, char *buffer)
-{
-	struct vexpress_hwmon_data *data = dev_get_drvdata(dev);
-
-	return sprintf(buffer, "%s\n", data->name);
-}
-
 static ssize_t vexpress_hwmon_label_show(struct device *dev,
 		struct device_attribute *dev_attr, char *buffer)
 {
@@ -95,16 +86,6 @@ static umode_t vexpress_hwmon_attr_is_visible(struct kobject *kobj,
 	return attr->mode;
 }
 
-static DEVICE_ATTR(name, S_IRUGO, vexpress_hwmon_name_show, NULL);
-
-#define VEXPRESS_HWMON_ATTRS(_name, _label_attr, _input_attr)	\
-struct attribute *vexpress_hwmon_attrs_##_name[] = {		\
-	&dev_attr_name.attr,					\
-	&dev_attr_##_label_attr.attr,				\
-	&sensor_dev_attr_##_input_attr.dev_attr.attr,		\
-	NULL							\
-}
-
 struct vexpress_hwmon_type {
 	const char *name;
 	const struct attribute_group **attr_groups;
@@ -114,7 +95,11 @@ struct vexpress_hwmon_type {
 static DEVICE_ATTR(in1_label, S_IRUGO, vexpress_hwmon_label_show, NULL);
 static SENSOR_DEVICE_ATTR(in1_input, S_IRUGO, vexpress_hwmon_u32_show,
 		NULL, 1000);
-static VEXPRESS_HWMON_ATTRS(volt, in1_label, in1_input);
+static struct attribute *vexpress_hwmon_attrs_volt[] = {
+	&dev_attr_in1_label.attr,
+	&sensor_dev_attr_in1_input.dev_attr.attr,
+	NULL
+};
 static struct attribute_group vexpress_hwmon_group_volt = {
 	.is_visible = vexpress_hwmon_attr_is_visible,
 	.attrs = vexpress_hwmon_attrs_volt,
@@ -131,7 +116,11 @@ static struct vexpress_hwmon_type vexpress_hwmon_volt = {
 static DEVICE_ATTR(curr1_label, S_IRUGO, vexpress_hwmon_label_show, NULL);
 static SENSOR_DEVICE_ATTR(curr1_input, S_IRUGO, vexpress_hwmon_u32_show,
 		NULL, 1000);
-static VEXPRESS_HWMON_ATTRS(amp, curr1_label, curr1_input);
+static struct attribute *vexpress_hwmon_attrs_amp[] = {
+	&dev_attr_curr1_label.attr,
+	&sensor_dev_attr_curr1_input.dev_attr.attr,
+	NULL
+};
 static struct attribute_group vexpress_hwmon_group_amp = {
 	.is_visible = vexpress_hwmon_attr_is_visible,
 	.attrs = vexpress_hwmon_attrs_amp,
@@ -147,7 +136,11 @@ static struct vexpress_hwmon_type vexpress_hwmon_amp = {
 static DEVICE_ATTR(temp1_label, S_IRUGO, vexpress_hwmon_label_show, NULL);
 static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, vexpress_hwmon_u32_show,
 		NULL, 1000);
-static VEXPRESS_HWMON_ATTRS(temp, temp1_label, temp1_input);
+static struct attribute *vexpress_hwmon_attrs_temp[] = {
+	&dev_attr_temp1_label.attr,
+	&sensor_dev_attr_temp1_input.dev_attr.attr,
+	NULL
+};
 static struct attribute_group vexpress_hwmon_group_temp = {
 	.is_visible = vexpress_hwmon_attr_is_visible,
 	.attrs = vexpress_hwmon_attrs_temp,
@@ -163,7 +156,11 @@ static struct vexpress_hwmon_type vexpress_hwmon_temp = {
 static DEVICE_ATTR(power1_label, S_IRUGO, vexpress_hwmon_label_show, NULL);
 static SENSOR_DEVICE_ATTR(power1_input, S_IRUGO, vexpress_hwmon_u32_show,
 		NULL, 1);
-static VEXPRESS_HWMON_ATTRS(power, power1_label, power1_input);
+static struct attribute *vexpress_hwmon_attrs_power[] = {
+	&dev_attr_power1_label.attr,
+	&sensor_dev_attr_power1_input.dev_attr.attr,
+	NULL
+};
 static struct attribute_group vexpress_hwmon_group_power = {
 	.is_visible = vexpress_hwmon_attr_is_visible,
 	.attrs = vexpress_hwmon_attrs_power,
@@ -179,7 +176,11 @@ static struct vexpress_hwmon_type vexpress_hwmon_power = {
 static DEVICE_ATTR(energy1_label, S_IRUGO, vexpress_hwmon_label_show, NULL);
 static SENSOR_DEVICE_ATTR(energy1_input, S_IRUGO, vexpress_hwmon_u64_show,
 		NULL, 1);
-static VEXPRESS_HWMON_ATTRS(energy, energy1_label, energy1_input);
+static struct attribute *vexpress_hwmon_attrs_energy[] = {
+	&dev_attr_energy1_label.attr,
+	&sensor_dev_attr_energy1_input.dev_attr.attr,
+	NULL
+};
 static struct attribute_group vexpress_hwmon_group_energy = {
 	.is_visible = vexpress_hwmon_attr_is_visible,
 	.attrs = vexpress_hwmon_attrs_energy,
@@ -218,7 +219,6 @@ MODULE_DEVICE_TABLE(of, vexpress_hwmon_of_match);
 
 static int vexpress_hwmon_probe(struct platform_device *pdev)
 {
-	int err;
 	const struct of_device_id *match;
 	struct vexpress_hwmon_data *data;
 	const struct vexpress_hwmon_type *type;
@@ -232,45 +232,19 @@ static int vexpress_hwmon_probe(struct platform_device *pdev)
 	if (!match)
 		return -ENODEV;
 	type = match->data;
-	data->name = type->name;
 
 	data->reg = devm_regmap_init_vexpress_config(&pdev->dev);
 	if (IS_ERR(data->reg))
 		return PTR_ERR(data->reg);
 
-	err = sysfs_create_groups(&pdev->dev.kobj, type->attr_groups);
-	if (err)
-		goto error;
-
-	data->hwmon_dev = hwmon_device_register(&pdev->dev);
-	if (IS_ERR(data->hwmon_dev)) {
-		err = PTR_ERR(data->hwmon_dev);
-		goto error;
-	}
-
-	return 0;
-
-error:
-	sysfs_remove_group(&pdev->dev.kobj, match->data);
-	return err;
-}
-
-static int vexpress_hwmon_remove(struct platform_device *pdev)
-{
-	struct vexpress_hwmon_data *data = platform_get_drvdata(pdev);
-	const struct of_device_id *match;
-
-	hwmon_device_unregister(data->hwmon_dev);
-
-	match = of_match_device(vexpress_hwmon_of_match, &pdev->dev);
-	sysfs_remove_group(&pdev->dev.kobj, match->data);
+	data->hwmon_dev = devm_hwmon_device_register_with_groups(&pdev->dev,
+			type->name, data, type->attr_groups);
 
-	return 0;
+	return PTR_ERR_OR_ZERO(data->hwmon_dev);
 }
 
 static struct platform_driver vexpress_hwmon_driver = {
 	.probe = vexpress_hwmon_probe,
-	.remove = vexpress_hwmon_remove,
 	.driver	= {
 		.name = DRVNAME,
 		.owner = THIS_MODULE,
diff --git a/drivers/hwmon/w83l786ng.c b/drivers/hwmon/w83l786ng.c
index 6ed76ceb9270..32487c19cbfc 100644
--- a/drivers/hwmon/w83l786ng.c
+++ b/drivers/hwmon/w83l786ng.c
@@ -249,7 +249,7 @@ static ssize_t show_##reg(struct device *dev, struct device_attribute *attr, \
 	int nr = to_sensor_dev_attr(attr)->index; \
 	struct w83l786ng_data *data = w83l786ng_update_device(dev); \
 	return sprintf(buf, "%d\n", \
-		FAN_FROM_REG(data->fan[nr], DIV_FROM_REG(data->fan_div[nr]))); \
+		FAN_FROM_REG(data->reg[nr], DIV_FROM_REG(data->fan_div[nr]))); \
 }
 
 show_fan_reg(fan);
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 620d1004a1e7..9f7d5859cf65 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -676,6 +676,16 @@ config I2C_RIIC
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-riic.
 
+config I2C_RK3X
+	tristate "Rockchip RK3xxx I2C adapter"
+	depends on OF
+	help
+	  Say Y here to include support for the I2C adapter in Rockchip RK3xxx
+	  SoCs.
+
+	  This driver can also be built as a module. If so, the module will
+	  be called i2c-rk3x.
+
 config HAVE_S3C2410_I2C
 	bool
 	help
@@ -764,6 +774,19 @@ config I2C_STU300
 	  This driver can also be built as a module. If so, the module
 	  will be called i2c-stu300.
 
+config I2C_SUN6I_P2WI
+	tristate "Allwinner sun6i internal P2WI controller"
+	depends on RESET_CONTROLLER
+	depends on MACH_SUN6I || COMPILE_TEST
+	help
+	  If you say yes to this option, support will be included for the
+	  P2WI (Push/Pull 2 Wire Interface) controller embedded in some sunxi
+	  SOCs.
+	  The P2WI looks like an SMBus controller (which supports only byte
+	  accesses), except that it only supports one slave device.
+	  This interface is used to connect to specific PMIC devices (like the
+	  AXP221).
+
 config I2C_TEGRA
 	tristate "NVIDIA Tegra internal I2C controller"
 	depends on ARCH_TEGRA
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index 298692cc6000..dd9a7f8e873f 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -66,6 +66,7 @@ obj-$(CONFIG_I2C_PXA)		+= i2c-pxa.o
 obj-$(CONFIG_I2C_PXA_PCI)	+= i2c-pxa-pci.o
 obj-$(CONFIG_I2C_QUP)		+= i2c-qup.o
 obj-$(CONFIG_I2C_RIIC)		+= i2c-riic.o
+obj-$(CONFIG_I2C_RK3X)		+= i2c-rk3x.o
 obj-$(CONFIG_I2C_S3C2410)	+= i2c-s3c2410.o
 obj-$(CONFIG_I2C_S6000)		+= i2c-s6000.o
 obj-$(CONFIG_I2C_SH7760)	+= i2c-sh7760.o
@@ -74,6 +75,7 @@ obj-$(CONFIG_I2C_SIMTEC)	+= i2c-simtec.o
 obj-$(CONFIG_I2C_SIRF)		+= i2c-sirf.o
 obj-$(CONFIG_I2C_ST)		+= i2c-st.o
 obj-$(CONFIG_I2C_STU300)	+= i2c-stu300.o
+obj-$(CONFIG_I2C_SUN6I_P2WI)	+= i2c-sun6i-p2wi.o
 obj-$(CONFIG_I2C_TEGRA)		+= i2c-tegra.o
 obj-$(CONFIG_I2C_VERSATILE)	+= i2c-versatile.o
 obj-$(CONFIG_I2C_WMT)		+= i2c-wmt.o
diff --git a/drivers/i2c/busses/i2c-rk3x.c b/drivers/i2c/busses/i2c-rk3x.c
new file mode 100644
index 000000000000..a9791509966a
--- /dev/null
+++ b/drivers/i2c/busses/i2c-rk3x.c
@@ -0,0 +1,763 @@
+/*
+ * Driver for I2C adapter in Rockchip RK3xxx SoC
+ *
+ * Max Schwarz <max.schwarz@online.de>
+ * based on the patches by Rockchip Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/i2c.h>
+#include <linux/interrupt.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/spinlock.h>
+#include <linux/clk.h>
+#include <linux/wait.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
+
+
+/* Register Map */
+#define REG_CON        0x00 /* control register */
+#define REG_CLKDIV     0x04 /* clock divisor register */
+#define REG_MRXADDR    0x08 /* slave address for REGISTER_TX */
+#define REG_MRXRADDR   0x0c /* slave register address for REGISTER_TX */
+#define REG_MTXCNT     0x10 /* number of bytes to be transmitted */
+#define REG_MRXCNT     0x14 /* number of bytes to be received */
+#define REG_IEN        0x18 /* interrupt enable */
+#define REG_IPD        0x1c /* interrupt pending */
+#define REG_FCNT       0x20 /* finished count */
+
+/* Data buffer offsets */
+#define TXBUFFER_BASE 0x100
+#define RXBUFFER_BASE 0x200
+
+/* REG_CON bits */
+#define REG_CON_EN        BIT(0)
+enum {
+	REG_CON_MOD_TX = 0,      /* transmit data */
+	REG_CON_MOD_REGISTER_TX, /* select register and restart */
+	REG_CON_MOD_RX,          /* receive data */
+	REG_CON_MOD_REGISTER_RX, /* broken: transmits read addr AND writes
+				  * register addr */
+};
+#define REG_CON_MOD(mod)  ((mod) << 1)
+#define REG_CON_MOD_MASK  (BIT(1) | BIT(2))
+#define REG_CON_START     BIT(3)
+#define REG_CON_STOP      BIT(4)
+#define REG_CON_LASTACK   BIT(5) /* 1: send NACK after last received byte */
+#define REG_CON_ACTACK    BIT(6) /* 1: stop if NACK is received */
+
+/* REG_MRXADDR bits */
+#define REG_MRXADDR_VALID(x) BIT(24 + (x)) /* [x*8+7:x*8] of MRX[R]ADDR valid */
+
+/* REG_IEN/REG_IPD bits */
+#define REG_INT_BTF       BIT(0) /* a byte was transmitted */
+#define REG_INT_BRF       BIT(1) /* a byte was received */
+#define REG_INT_MBTF      BIT(2) /* master data transmit finished */
+#define REG_INT_MBRF      BIT(3) /* master data receive finished */
+#define REG_INT_START     BIT(4) /* START condition generated */
+#define REG_INT_STOP      BIT(5) /* STOP condition generated */
+#define REG_INT_NAKRCV    BIT(6) /* NACK received */
+#define REG_INT_ALL       0x7f
+
+/* Constants */
+#define WAIT_TIMEOUT      200 /* ms */
+#define DEFAULT_SCL_RATE  (100 * 1000) /* Hz */
+
+enum rk3x_i2c_state {
+	STATE_IDLE,
+	STATE_START,
+	STATE_READ,
+	STATE_WRITE,
+	STATE_STOP
+};
+
+/**
+ * @grf_offset: offset inside the grf regmap for setting the i2c type
+ */
+struct rk3x_i2c_soc_data {
+	int grf_offset;
+};
+
+struct rk3x_i2c {
+	struct i2c_adapter adap;
+	struct device *dev;
+	struct rk3x_i2c_soc_data *soc_data;
+
+	/* Hardware resources */
+	void __iomem *regs;
+	struct clk *clk;
+
+	/* Settings */
+	unsigned int scl_frequency;
+
+	/* Synchronization & notification */
+	spinlock_t lock;
+	wait_queue_head_t wait;
+	bool busy;
+
+	/* Current message */
+	struct i2c_msg *msg;
+	u8 addr;
+	unsigned int mode;
+	bool is_last_msg;
+
+	/* I2C state machine */
+	enum rk3x_i2c_state state;
+	unsigned int processed; /* sent/received bytes */
+	int error;
+};
+
+static inline void i2c_writel(struct rk3x_i2c *i2c, u32 value,
+			      unsigned int offset)
+{
+	writel(value, i2c->regs + offset);
+}
+
+static inline u32 i2c_readl(struct rk3x_i2c *i2c, unsigned int offset)
+{
+	return readl(i2c->regs + offset);
+}
+
+/* Reset all interrupt pending bits */
+static inline void rk3x_i2c_clean_ipd(struct rk3x_i2c *i2c)
+{
+	i2c_writel(i2c, REG_INT_ALL, REG_IPD);
+}
+
+/**
+ * Generate a START condition, which triggers a REG_INT_START interrupt.
+ */
+static void rk3x_i2c_start(struct rk3x_i2c *i2c)
+{
+	u32 val;
+
+	rk3x_i2c_clean_ipd(i2c);
+	i2c_writel(i2c, REG_INT_START, REG_IEN);
+
+	/* enable adapter with correct mode, send START condition */
+	val = REG_CON_EN | REG_CON_MOD(i2c->mode) | REG_CON_START;
+
+	/* if we want to react to NACK, set ACTACK bit */
+	if (!(i2c->msg->flags & I2C_M_IGNORE_NAK))
+		val |= REG_CON_ACTACK;
+
+	i2c_writel(i2c, val, REG_CON);
+}
+
+/**
+ * Generate a STOP condition, which triggers a REG_INT_STOP interrupt.
+ *
+ * @error: Error code to return in rk3x_i2c_xfer
+ */
+static void rk3x_i2c_stop(struct rk3x_i2c *i2c, int error)
+{
+	unsigned int ctrl;
+
+	i2c->processed = 0;
+	i2c->msg = NULL;
+	i2c->error = error;
+
+	if (i2c->is_last_msg) {
+		/* Enable stop interrupt */
+		i2c_writel(i2c, REG_INT_STOP, REG_IEN);
+
+		i2c->state = STATE_STOP;
+
+		ctrl = i2c_readl(i2c, REG_CON);
+		ctrl |= REG_CON_STOP;
+		i2c_writel(i2c, ctrl, REG_CON);
+	} else {
+		/* Signal rk3x_i2c_xfer to start the next message. */
+		i2c->busy = false;
+		i2c->state = STATE_IDLE;
+
+		/*
+		 * The HW is actually not capable of REPEATED START. But we can
+		 * get the intended effect by resetting its internal state
+		 * and issuing an ordinary START.
+		 */
+		i2c_writel(i2c, 0, REG_CON);
+
+		/* signal that we are finished with the current msg */
+		wake_up(&i2c->wait);
+	}
+}
+
+/**
+ * Setup a read according to i2c->msg
+ */
+static void rk3x_i2c_prepare_read(struct rk3x_i2c *i2c)
+{
+	unsigned int len = i2c->msg->len - i2c->processed;
+	u32 con;
+
+	con = i2c_readl(i2c, REG_CON);
+
+	/*
+	 * The hw can read up to 32 bytes at a time. If we need more than one
+	 * chunk, send an ACK after the last byte of the current chunk.
+	 */
+	if (unlikely(len > 32)) {
+		len = 32;
+		con &= ~REG_CON_LASTACK;
+	} else {
+		con |= REG_CON_LASTACK;
+	}
+
+	/* make sure we are in plain RX mode if we read a second chunk */
+	if (i2c->processed != 0) {
+		con &= ~REG_CON_MOD_MASK;
+		con |= REG_CON_MOD(REG_CON_MOD_RX);
+	}
+
+	i2c_writel(i2c, con, REG_CON);
+	i2c_writel(i2c, len, REG_MRXCNT);
+}
+
+/**
+ * Fill the transmit buffer with data from i2c->msg
+ */
+static void rk3x_i2c_fill_transmit_buf(struct rk3x_i2c *i2c)
+{
+	unsigned int i, j;
+	u32 cnt = 0;
+	u32 val;
+	u8 byte;
+
+	for (i = 0; i < 8; ++i) {
+		val = 0;
+		for (j = 0; j < 4; ++j) {
+			if (i2c->processed == i2c->msg->len)
+				break;
+
+			if (i2c->processed == 0 && cnt == 0)
+				byte = (i2c->addr & 0x7f) << 1;
+			else
+				byte = i2c->msg->buf[i2c->processed++];
+
+			val |= byte << (j * 8);
+			cnt++;
+		}
+
+		i2c_writel(i2c, val, TXBUFFER_BASE + 4 * i);
+
+		if (i2c->processed == i2c->msg->len)
+			break;
+	}
+
+	i2c_writel(i2c, cnt, REG_MTXCNT);
+}
+
+
+/* IRQ handlers for individual states */
+
+static void rk3x_i2c_handle_start(struct rk3x_i2c *i2c, unsigned int ipd)
+{
+	if (!(ipd & REG_INT_START)) {
+		rk3x_i2c_stop(i2c, -EIO);
+		dev_warn(i2c->dev, "unexpected irq in START: 0x%x\n", ipd);
+		rk3x_i2c_clean_ipd(i2c);
+		return;
+	}
+
+	/* ack interrupt */
+	i2c_writel(i2c, REG_INT_START, REG_IPD);
+
+	/* disable start bit */
+	i2c_writel(i2c, i2c_readl(i2c, REG_CON) & ~REG_CON_START, REG_CON);
+
+	/* enable appropriate interrupts and transition */
+	if (i2c->mode == REG_CON_MOD_TX) {
+		i2c_writel(i2c, REG_INT_MBTF | REG_INT_NAKRCV, REG_IEN);
+		i2c->state = STATE_WRITE;
+		rk3x_i2c_fill_transmit_buf(i2c);
+	} else {
+		/* in any other case, we are going to be reading. */
+		i2c_writel(i2c, REG_INT_MBRF | REG_INT_NAKRCV, REG_IEN);
+		i2c->state = STATE_READ;
+		rk3x_i2c_prepare_read(i2c);
+	}
+}
+
+static void rk3x_i2c_handle_write(struct rk3x_i2c *i2c, unsigned int ipd)
+{
+	if (!(ipd & REG_INT_MBTF)) {
+		rk3x_i2c_stop(i2c, -EIO);
+		dev_err(i2c->dev, "unexpected irq in WRITE: 0x%x\n", ipd);
+		rk3x_i2c_clean_ipd(i2c);
+		return;
+	}
+
+	/* ack interrupt */
+	i2c_writel(i2c, REG_INT_MBTF, REG_IPD);
+
+	/* are we finished? */
+	if (i2c->processed == i2c->msg->len)
+		rk3x_i2c_stop(i2c, i2c->error);
+	else
+		rk3x_i2c_fill_transmit_buf(i2c);
+}
+
+static void rk3x_i2c_handle_read(struct rk3x_i2c *i2c, unsigned int ipd)
+{
+	unsigned int i;
+	unsigned int len = i2c->msg->len - i2c->processed;
+	u32 uninitialized_var(val);
+	u8 byte;
+
+	/* we only care for MBRF here. */
+	if (!(ipd & REG_INT_MBRF))
+		return;
+
+	/* ack interrupt */
+	i2c_writel(i2c, REG_INT_MBRF, REG_IPD);
+
+	/* read the data from receive buffer */
+	for (i = 0; i < len; ++i) {
+		if (i % 4 == 0)
+			val = i2c_readl(i2c, RXBUFFER_BASE + (i / 4) * 4);
+
+		byte = (val >> ((i % 4) * 8)) & 0xff;
+		i2c->msg->buf[i2c->processed++] = byte;
+	}
+
+	/* are we finished? */
+	if (i2c->processed == i2c->msg->len)
+		rk3x_i2c_stop(i2c, i2c->error);
+	else
+		rk3x_i2c_prepare_read(i2c);
+}
+
+static void rk3x_i2c_handle_stop(struct rk3x_i2c *i2c, unsigned int ipd)
+{
+	unsigned int con;
+
+	if (!(ipd & REG_INT_STOP)) {
+		rk3x_i2c_stop(i2c, -EIO);
+		dev_err(i2c->dev, "unexpected irq in STOP: 0x%x\n", ipd);
+		rk3x_i2c_clean_ipd(i2c);
+		return;
+	}
+
+	/* ack interrupt */
+	i2c_writel(i2c, REG_INT_STOP, REG_IPD);
+
+	/* disable STOP bit */
+	con = i2c_readl(i2c, REG_CON);
+	con &= ~REG_CON_STOP;
+	i2c_writel(i2c, con, REG_CON);
+
+	i2c->busy = false;
+	i2c->state = STATE_IDLE;
+
+	/* signal rk3x_i2c_xfer that we are finished */
+	wake_up(&i2c->wait);
+}
+
+static irqreturn_t rk3x_i2c_irq(int irqno, void *dev_id)
+{
+	struct rk3x_i2c *i2c = dev_id;
+	unsigned int ipd;
+
+	spin_lock(&i2c->lock);
+
+	ipd = i2c_readl(i2c, REG_IPD);
+	if (i2c->state == STATE_IDLE) {
+		dev_warn(i2c->dev, "irq in STATE_IDLE, ipd = 0x%x\n", ipd);
+		rk3x_i2c_clean_ipd(i2c);
+		goto out;
+	}
+
+	dev_dbg(i2c->dev, "IRQ: state %d, ipd: %x\n", i2c->state, ipd);
+
+	/* Clean interrupt bits we don't care about */
+	ipd &= ~(REG_INT_BRF | REG_INT_BTF);
+
+	if (ipd & REG_INT_NAKRCV) {
+		/*
+		 * We got a NACK in the last operation. Depending on whether
+		 * IGNORE_NAK is set, we have to stop the operation and report
+		 * an error.
+		 */
+		i2c_writel(i2c, REG_INT_NAKRCV, REG_IPD);
+
+		ipd &= ~REG_INT_NAKRCV;
+
+		if (!(i2c->msg->flags & I2C_M_IGNORE_NAK))
+			rk3x_i2c_stop(i2c, -ENXIO);
+	}
+
+	/* is there anything left to handle? */
+	if (unlikely(ipd == 0))
+		goto out;
+
+	switch (i2c->state) {
+	case STATE_START:
+		rk3x_i2c_handle_start(i2c, ipd);
+		break;
+	case STATE_WRITE:
+		rk3x_i2c_handle_write(i2c, ipd);
+		break;
+	case STATE_READ:
+		rk3x_i2c_handle_read(i2c, ipd);
+		break;
+	case STATE_STOP:
+		rk3x_i2c_handle_stop(i2c, ipd);
+		break;
+	case STATE_IDLE:
+		break;
+	}
+
+out:
+	spin_unlock(&i2c->lock);
+	return IRQ_HANDLED;
+}
+
+static void rk3x_i2c_set_scl_rate(struct rk3x_i2c *i2c, unsigned long scl_rate)
+{
+	unsigned long i2c_rate = clk_get_rate(i2c->clk);
+	unsigned int div;
+
+	/* SCL rate = (clk rate) / (8 * DIV) */
+	div = DIV_ROUND_UP(i2c_rate, scl_rate * 8);
+
+	/* The lower and upper half of the CLKDIV reg describe the length of
+	 * SCL low & high periods. */
+	div = DIV_ROUND_UP(div, 2);
+
+	i2c_writel(i2c, (div << 16) | (div & 0xffff), REG_CLKDIV);
+}
+
+/**
+ * Setup I2C registers for an I2C operation specified by msgs, num.
+ *
+ * Must be called with i2c->lock held.
+ *
+ * @msgs: I2C msgs to process
+ * @num: Number of msgs
+ *
+ * returns: Number of I2C msgs processed or negative in case of error
+ */
+static int rk3x_i2c_setup(struct rk3x_i2c *i2c, struct i2c_msg *msgs, int num)
+{
+	u32 addr = (msgs[0].addr & 0x7f) << 1;
+	int ret = 0;
+
+	/*
+	 * The I2C adapter can issue a small (len < 4) write packet before
+	 * reading. This speeds up SMBus-style register reads.
+	 * The MRXADDR/MRXRADDR hold the slave address and the slave register
+	 * address in this case.
+	 */
+
+	if (num >= 2 && msgs[0].len < 4 &&
+	    !(msgs[0].flags & I2C_M_RD) && (msgs[1].flags & I2C_M_RD)) {
+		u32 reg_addr = 0;
+		int i;
+
+		dev_dbg(i2c->dev, "Combined write/read from addr 0x%x\n",
+			addr >> 1);
+
+		/* Fill MRXRADDR with the register address(es) */
+		for (i = 0; i < msgs[0].len; ++i) {
+			reg_addr |= msgs[0].buf[i] << (i * 8);
+			reg_addr |= REG_MRXADDR_VALID(i);
+		}
+
+		/* msgs[0] is handled by hw. */
+		i2c->msg = &msgs[1];
+
+		i2c->mode = REG_CON_MOD_REGISTER_TX;
+
+		i2c_writel(i2c, addr | REG_MRXADDR_VALID(0), REG_MRXADDR);
+		i2c_writel(i2c, reg_addr, REG_MRXRADDR);
+
+		ret = 2;
+	} else {
+		/*
+		 * We'll have to do it the boring way and process the msgs
+		 * one-by-one.
+		 */
+
+		if (msgs[0].flags & I2C_M_RD) {
+			addr |= 1; /* set read bit */
+
+			/*
+			 * We have to transmit the slave addr first. Use
+			 * MOD_REGISTER_TX for that purpose.
+			 */
+			i2c->mode = REG_CON_MOD_REGISTER_TX;
+			i2c_writel(i2c, addr | REG_MRXADDR_VALID(0),
+				   REG_MRXADDR);
+			i2c_writel(i2c, 0, REG_MRXRADDR);
+		} else {
+			i2c->mode = REG_CON_MOD_TX;
+		}
+
+		i2c->msg = &msgs[0];
+
+		ret = 1;
+	}
+
+	i2c->addr = msgs[0].addr;
+	i2c->busy = true;
+	i2c->state = STATE_START;
+	i2c->processed = 0;
+	i2c->error = 0;
+
+	rk3x_i2c_clean_ipd(i2c);
+
+	return ret;
+}
+
+static int rk3x_i2c_xfer(struct i2c_adapter *adap,
+			 struct i2c_msg *msgs, int num)
+{
+	struct rk3x_i2c *i2c = (struct rk3x_i2c *)adap->algo_data;
+	unsigned long timeout, flags;
+	int ret = 0;
+	int i;
+
+	spin_lock_irqsave(&i2c->lock, flags);
+
+	clk_enable(i2c->clk);
+
+	/* The clock rate might have changed, so setup the divider again */
+	rk3x_i2c_set_scl_rate(i2c, i2c->scl_frequency);
+
+	i2c->is_last_msg = false;
+
+	/*
+	 * Process msgs. We can handle more than one message at once (see
+	 * rk3x_i2c_setup()).
+	 */
+	for (i = 0; i < num; i += ret) {
+		ret = rk3x_i2c_setup(i2c, msgs + i, num - i);
+
+		if (ret < 0) {
+			dev_err(i2c->dev, "rk3x_i2c_setup() failed\n");
+			break;
+		}
+
+		if (i + ret >= num)
+			i2c->is_last_msg = true;
+
+		spin_unlock_irqrestore(&i2c->lock, flags);
+
+		rk3x_i2c_start(i2c);
+
+		timeout = wait_event_timeout(i2c->wait, !i2c->busy,
+					     msecs_to_jiffies(WAIT_TIMEOUT));
+
+		spin_lock_irqsave(&i2c->lock, flags);
+
+		if (timeout == 0) {
+			dev_err(i2c->dev, "timeout, ipd: 0x%02x, state: %d\n",
+				i2c_readl(i2c, REG_IPD), i2c->state);
+
+			/* Force a STOP condition without interrupt */
+			i2c_writel(i2c, 0, REG_IEN);
+			i2c_writel(i2c, REG_CON_EN | REG_CON_STOP, REG_CON);
+
+			i2c->state = STATE_IDLE;
+
+			ret = -ETIMEDOUT;
+			break;
+		}
+
+		if (i2c->error) {
+			ret = i2c->error;
+			break;
+		}
+	}
+
+	clk_disable(i2c->clk);
+	spin_unlock_irqrestore(&i2c->lock, flags);
+
+	return ret;
+}
+
+static u32 rk3x_i2c_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL | I2C_FUNC_PROTOCOL_MANGLING;
+}
+
+static const struct i2c_algorithm rk3x_i2c_algorithm = {
+	.master_xfer		= rk3x_i2c_xfer,
+	.functionality		= rk3x_i2c_func,
+};
+
+static struct rk3x_i2c_soc_data soc_data[3] = {
+	{ .grf_offset = 0x154 }, /* rk3066 */
+	{ .grf_offset = 0x0a4 }, /* rk3188 */
+	{ .grf_offset = -1 },    /* no I2C switching needed */
+};
+
+static const struct of_device_id rk3x_i2c_match[] = {
+	{ .compatible = "rockchip,rk3066-i2c", .data = (void *)&soc_data[0] },
+	{ .compatible = "rockchip,rk3188-i2c", .data = (void *)&soc_data[1] },
+	{ .compatible = "rockchip,rk3288-i2c", .data = (void *)&soc_data[2] },
+	{},
+};
+
+static int rk3x_i2c_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	const struct of_device_id *match;
+	struct rk3x_i2c *i2c;
+	struct resource *mem;
+	int ret = 0;
+	int bus_nr;
+	u32 value;
+	int irq;
+
+	i2c = devm_kzalloc(&pdev->dev, sizeof(struct rk3x_i2c), GFP_KERNEL);
+	if (!i2c)
+		return -ENOMEM;
+
+	match = of_match_node(rk3x_i2c_match, np);
+	i2c->soc_data = (struct rk3x_i2c_soc_data *)match->data;
+
+	if (of_property_read_u32(pdev->dev.of_node, "clock-frequency",
+				 &i2c->scl_frequency)) {
+		dev_info(&pdev->dev, "using default SCL frequency: %d\n",
+			 DEFAULT_SCL_RATE);
+		i2c->scl_frequency = DEFAULT_SCL_RATE;
+	}
+
+	if (i2c->scl_frequency == 0 || i2c->scl_frequency > 400 * 1000) {
+		dev_warn(&pdev->dev, "invalid SCL frequency specified.\n");
+		dev_warn(&pdev->dev, "using default SCL frequency: %d\n",
+			 DEFAULT_SCL_RATE);
+		i2c->scl_frequency = DEFAULT_SCL_RATE;
+	}
+
+	strlcpy(i2c->adap.name, "rk3x-i2c", sizeof(i2c->adap.name));
+	i2c->adap.owner = THIS_MODULE;
+	i2c->adap.algo = &rk3x_i2c_algorithm;
+	i2c->adap.retries = 3;
+	i2c->adap.dev.of_node = np;
+	i2c->adap.algo_data = i2c;
+	i2c->adap.dev.parent = &pdev->dev;
+
+	i2c->dev = &pdev->dev;
+
+	spin_lock_init(&i2c->lock);
+	init_waitqueue_head(&i2c->wait);
+
+	i2c->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(i2c->clk)) {
+		dev_err(&pdev->dev, "cannot get clock\n");
+		return PTR_ERR(i2c->clk);
+	}
+
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	i2c->regs = devm_ioremap_resource(&pdev->dev, mem);
+	if (IS_ERR(i2c->regs))
+		return PTR_ERR(i2c->regs);
+
+	/* Try to set the I2C adapter number from dt */
+	bus_nr = of_alias_get_id(np, "i2c");
+
+	/*
+	 * Switch to new interface if the SoC also offers the old one.
+	 * The control bit is located in the GRF register space.
+	 */
+	if (i2c->soc_data->grf_offset >= 0) {
+		struct regmap *grf;
+
+		grf = syscon_regmap_lookup_by_phandle(np, "rockchip,grf");
+		if (IS_ERR(grf)) {
+			dev_err(&pdev->dev,
+				"rk3x-i2c needs 'rockchip,grf' property\n");
+			return PTR_ERR(grf);
+		}
+
+		if (bus_nr < 0) {
+			dev_err(&pdev->dev, "rk3x-i2c needs i2cX alias");
+			return -EINVAL;
+		}
+
+		/* 27+i: write mask, 11+i: value */
+		value = BIT(27 + bus_nr) | BIT(11 + bus_nr);
+
+		ret = regmap_write(grf, i2c->soc_data->grf_offset, value);
+		if (ret != 0) {
+			dev_err(i2c->dev, "Could not write to GRF: %d\n", ret);
+			return ret;
+		}
+	}
+
+	/* IRQ setup */
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(&pdev->dev, "cannot find rk3x IRQ\n");
+		return irq;
+	}
+
+	ret = devm_request_irq(&pdev->dev, irq, rk3x_i2c_irq,
+			       0, dev_name(&pdev->dev), i2c);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "cannot request IRQ\n");
+		return ret;
+	}
+
+	platform_set_drvdata(pdev, i2c);
+
+	ret = clk_prepare(i2c->clk);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "Could not prepare clock\n");
+		return ret;
+	}
+
+	ret = i2c_add_adapter(&i2c->adap);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "Could not register adapter\n");
+		goto err_clk;
+	}
+
+	dev_info(&pdev->dev, "Initialized RK3xxx I2C bus at %p\n", i2c->regs);
+
+	return 0;
+
+err_clk:
+	clk_unprepare(i2c->clk);
+	return ret;
+}
+
+static int rk3x_i2c_remove(struct platform_device *pdev)
+{
+	struct rk3x_i2c *i2c = platform_get_drvdata(pdev);
+
+	i2c_del_adapter(&i2c->adap);
+	clk_unprepare(i2c->clk);
+
+	return 0;
+}
+
+static struct platform_driver rk3x_i2c_driver = {
+	.probe   = rk3x_i2c_probe,
+	.remove  = rk3x_i2c_remove,
+	.driver  = {
+		.owner = THIS_MODULE,
+		.name  = "rk3x-i2c",
+		.of_match_table = rk3x_i2c_match,
+	},
+};
+
+module_platform_driver(rk3x_i2c_driver);
+
+MODULE_DESCRIPTION("Rockchip RK3xxx I2C Bus driver");
+MODULE_AUTHOR("Max Schwarz <max.schwarz@online.de>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/i2c/busses/i2c-sun6i-p2wi.c b/drivers/i2c/busses/i2c-sun6i-p2wi.c
new file mode 100644
index 000000000000..4d75d4759709
--- /dev/null
+++ b/drivers/i2c/busses/i2c-sun6i-p2wi.c
@@ -0,0 +1,344 @@
+/*
+ * P2WI (Push-Pull Two Wire Interface) bus driver.
+ *
+ * Author: Boris BREZILLON <boris.brezillon@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2.  This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ *
+ * The P2WI controller looks like an SMBus controller which only supports byte
+ * data transfers. But, it differs from standard SMBus protocol on several
+ * aspects:
+ * - it supports only one slave device, and thus drop the address field
+ * - it adds a parity bit every 8bits of data
+ * - only one read access is required to read a byte (instead of a write
+ *   followed by a read access in standard SMBus protocol)
+ * - there's no Ack bit after each byte transfer
+ *
+ * This means this bus cannot be used to interface with standard SMBus
+ * devices (the only known device to support this interface is the AXP221
+ * PMIC).
+ *
+ */
+#include <linux/clk.h>
+#include <linux/i2c.h>
+#include <linux/io.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/reset.h>
+
+
+/* P2WI registers */
+#define P2WI_CTRL		0x0
+#define P2WI_CCR		0x4
+#define P2WI_INTE		0x8
+#define P2WI_INTS		0xc
+#define P2WI_DADDR0		0x10
+#define P2WI_DADDR1		0x14
+#define P2WI_DLEN		0x18
+#define P2WI_DATA0		0x1c
+#define P2WI_DATA1		0x20
+#define P2WI_LCR		0x24
+#define P2WI_PMCR		0x28
+
+/* CTRL fields */
+#define P2WI_CTRL_START_TRANS		BIT(7)
+#define P2WI_CTRL_ABORT_TRANS		BIT(6)
+#define P2WI_CTRL_GLOBAL_INT_ENB	BIT(1)
+#define P2WI_CTRL_SOFT_RST		BIT(0)
+
+/* CLK CTRL fields */
+#define P2WI_CCR_SDA_OUT_DELAY(v)	(((v) & 0x7) << 8)
+#define P2WI_CCR_MAX_CLK_DIV		0xff
+#define P2WI_CCR_CLK_DIV(v)		((v) & P2WI_CCR_MAX_CLK_DIV)
+
+/* STATUS fields */
+#define P2WI_INTS_TRANS_ERR_ID(v)	(((v) >> 8) & 0xff)
+#define P2WI_INTS_LOAD_BSY		BIT(2)
+#define P2WI_INTS_TRANS_ERR		BIT(1)
+#define P2WI_INTS_TRANS_OVER		BIT(0)
+
+/* DATA LENGTH fields*/
+#define P2WI_DLEN_READ			BIT(4)
+#define P2WI_DLEN_DATA_LENGTH(v)	((v - 1) & 0x7)
+
+/* LINE CTRL fields*/
+#define P2WI_LCR_SCL_STATE		BIT(5)
+#define P2WI_LCR_SDA_STATE		BIT(4)
+#define P2WI_LCR_SCL_CTL		BIT(3)
+#define P2WI_LCR_SCL_CTL_EN		BIT(2)
+#define P2WI_LCR_SDA_CTL		BIT(1)
+#define P2WI_LCR_SDA_CTL_EN		BIT(0)
+
+/* PMU MODE CTRL fields */
+#define P2WI_PMCR_PMU_INIT_SEND		BIT(31)
+#define P2WI_PMCR_PMU_INIT_DATA(v)	(((v) & 0xff) << 16)
+#define P2WI_PMCR_PMU_MODE_REG(v)	(((v) & 0xff) << 8)
+#define P2WI_PMCR_PMU_DEV_ADDR(v)	((v) & 0xff)
+
+#define P2WI_MAX_FREQ			6000000
+
+struct p2wi {
+	struct i2c_adapter adapter;
+	struct completion complete;
+	unsigned int status;
+	void __iomem *regs;
+	struct clk *clk;
+	struct reset_control *rstc;
+	int slave_addr;
+};
+
+static irqreturn_t p2wi_interrupt(int irq, void *dev_id)
+{
+	struct p2wi *p2wi = dev_id;
+	unsigned long status;
+
+	status = readl(p2wi->regs + P2WI_INTS);
+	p2wi->status = status;
+
+	/* Clear interrupts */
+	status &= (P2WI_INTS_LOAD_BSY | P2WI_INTS_TRANS_ERR |
+		   P2WI_INTS_TRANS_OVER);
+	writel(status, p2wi->regs + P2WI_INTS);
+
+	complete(&p2wi->complete);
+
+	return IRQ_HANDLED;
+}
+
+static u32 p2wi_functionality(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_SMBUS_BYTE_DATA;
+}
+
+static int p2wi_smbus_xfer(struct i2c_adapter *adap, u16 addr,
+			   unsigned short flags, char read_write,
+			   u8 command, int size, union i2c_smbus_data *data)
+{
+	struct p2wi *p2wi = i2c_get_adapdata(adap);
+	unsigned long dlen = P2WI_DLEN_DATA_LENGTH(1);
+
+	if (p2wi->slave_addr >= 0 && addr != p2wi->slave_addr) {
+		dev_err(&adap->dev, "invalid P2WI address\n");
+		return -EINVAL;
+	}
+
+	if (!data)
+		return -EINVAL;
+
+	writel(command, p2wi->regs + P2WI_DADDR0);
+
+	if (read_write == I2C_SMBUS_READ)
+		dlen |= P2WI_DLEN_READ;
+	else
+		writel(data->byte, p2wi->regs + P2WI_DATA0);
+
+	writel(dlen, p2wi->regs + P2WI_DLEN);
+
+	if (readl(p2wi->regs + P2WI_CTRL) & P2WI_CTRL_START_TRANS) {
+		dev_err(&adap->dev, "P2WI bus busy\n");
+		return -EBUSY;
+	}
+
+	reinit_completion(&p2wi->complete);
+
+	writel(P2WI_INTS_LOAD_BSY | P2WI_INTS_TRANS_ERR | P2WI_INTS_TRANS_OVER,
+	       p2wi->regs + P2WI_INTE);
+
+	writel(P2WI_CTRL_START_TRANS | P2WI_CTRL_GLOBAL_INT_ENB,
+	       p2wi->regs + P2WI_CTRL);
+
+	wait_for_completion(&p2wi->complete);
+
+	if (p2wi->status & P2WI_INTS_LOAD_BSY) {
+		dev_err(&adap->dev, "P2WI bus busy\n");
+		return -EBUSY;
+	}
+
+	if (p2wi->status & P2WI_INTS_TRANS_ERR) {
+		dev_err(&adap->dev, "P2WI bus xfer error\n");
+		return -ENXIO;
+	}
+
+	if (read_write == I2C_SMBUS_READ)
+		data->byte = readl(p2wi->regs + P2WI_DATA0);
+
+	return 0;
+}
+
+static const struct i2c_algorithm p2wi_algo = {
+	.smbus_xfer = p2wi_smbus_xfer,
+	.functionality = p2wi_functionality,
+};
+
+static const struct of_device_id p2wi_of_match_table[] = {
+	{ .compatible = "allwinner,sun6i-a31-p2wi" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, p2wi_of_match_table);
+
+static int p2wi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *np = dev->of_node;
+	struct device_node *childnp;
+	unsigned long parent_clk_freq;
+	u32 clk_freq = 100000;
+	struct resource *r;
+	struct p2wi *p2wi;
+	u32 slave_addr;
+	int clk_div;
+	int irq;
+	int ret;
+
+	of_property_read_u32(np, "clock-frequency", &clk_freq);
+	if (clk_freq > P2WI_MAX_FREQ) {
+		dev_err(dev,
+			"required clock-frequency (%u Hz) is too high (max = 6MHz)",
+			clk_freq);
+		return -EINVAL;
+	}
+
+	if (of_get_child_count(np) > 1) {
+		dev_err(dev, "P2WI only supports one slave device\n");
+		return -EINVAL;
+	}
+
+	p2wi = devm_kzalloc(dev, sizeof(struct p2wi), GFP_KERNEL);
+	if (!p2wi)
+		return -ENOMEM;
+
+	p2wi->slave_addr = -1;
+
+	/*
+	 * Authorize a p2wi node without any children to be able to use an
+	 * i2c-dev from userpace.
+	 * In this case the slave_addr is set to -1 and won't be checked when
+	 * launching a P2WI transfer.
+	 */
+	childnp = of_get_next_available_child(np, NULL);
+	if (childnp) {
+		ret = of_property_read_u32(childnp, "reg", &slave_addr);
+		if (ret) {
+			dev_err(dev, "invalid slave address on node %s\n",
+				childnp->full_name);
+			return -EINVAL;
+		}
+
+		p2wi->slave_addr = slave_addr;
+	}
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	p2wi->regs = devm_ioremap_resource(dev, r);
+	if (IS_ERR(p2wi->regs))
+		return PTR_ERR(p2wi->regs);
+
+	strlcpy(p2wi->adapter.name, pdev->name, sizeof(p2wi->adapter.name));
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(dev, "failed to retrieve irq: %d\n", irq);
+		return irq;
+	}
+
+	p2wi->clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(p2wi->clk)) {
+		ret = PTR_ERR(p2wi->clk);
+		dev_err(dev, "failed to retrieve clk: %d\n", ret);
+		return ret;
+	}
+
+	ret = clk_prepare_enable(p2wi->clk);
+	if (ret) {
+		dev_err(dev, "failed to enable clk: %d\n", ret);
+		return ret;
+	}
+
+	parent_clk_freq = clk_get_rate(p2wi->clk);
+
+	p2wi->rstc = devm_reset_control_get(dev, NULL);
+	if (IS_ERR(p2wi->rstc)) {
+		ret = PTR_ERR(p2wi->rstc);
+		dev_err(dev, "failed to retrieve reset controller: %d\n", ret);
+		goto err_clk_disable;
+	}
+
+	ret = reset_control_deassert(p2wi->rstc);
+	if (ret) {
+		dev_err(dev, "failed to deassert reset line: %d\n", ret);
+		goto err_clk_disable;
+	}
+
+	init_completion(&p2wi->complete);
+	p2wi->adapter.dev.parent = dev;
+	p2wi->adapter.algo = &p2wi_algo;
+	p2wi->adapter.owner = THIS_MODULE;
+	p2wi->adapter.dev.of_node = pdev->dev.of_node;
+	platform_set_drvdata(pdev, p2wi);
+	i2c_set_adapdata(&p2wi->adapter, p2wi);
+
+	ret = devm_request_irq(dev, irq, p2wi_interrupt, 0, pdev->name, p2wi);
+	if (ret) {
+		dev_err(dev, "can't register interrupt handler irq%d: %d\n",
+			irq, ret);
+		goto err_reset_assert;
+	}
+
+	writel(P2WI_CTRL_SOFT_RST, p2wi->regs + P2WI_CTRL);
+
+	clk_div = parent_clk_freq / clk_freq;
+	if (!clk_div) {
+		dev_warn(dev,
+			 "clock-frequency is too high, setting it to %lu Hz\n",
+			 parent_clk_freq);
+		clk_div = 1;
+	} else if (clk_div > P2WI_CCR_MAX_CLK_DIV) {
+		dev_warn(dev,
+			 "clock-frequency is too low, setting it to %lu Hz\n",
+			 parent_clk_freq / P2WI_CCR_MAX_CLK_DIV);
+		clk_div = P2WI_CCR_MAX_CLK_DIV;
+	}
+
+	writel(P2WI_CCR_SDA_OUT_DELAY(1) | P2WI_CCR_CLK_DIV(clk_div),
+	       p2wi->regs + P2WI_CCR);
+
+	ret = i2c_add_adapter(&p2wi->adapter);
+	if (!ret)
+		return 0;
+
+err_reset_assert:
+	reset_control_assert(p2wi->rstc);
+
+err_clk_disable:
+	clk_disable_unprepare(p2wi->clk);
+
+	return ret;
+}
+
+static int p2wi_remove(struct platform_device *dev)
+{
+	struct p2wi *p2wi = platform_get_drvdata(dev);
+
+	reset_control_assert(p2wi->rstc);
+	clk_disable_unprepare(p2wi->clk);
+	i2c_del_adapter(&p2wi->adapter);
+
+	return 0;
+}
+
+static struct platform_driver p2wi_driver = {
+	.probe	= p2wi_probe,
+	.remove	= p2wi_remove,
+	.driver	= {
+		.owner = THIS_MODULE,
+		.name = "i2c-sunxi-p2wi",
+		.of_match_table = p2wi_of_match_table,
+	},
+};
+module_platform_driver(p2wi_driver);
+
+MODULE_AUTHOR("Boris BREZILLON <boris.brezillon@free-electrons.com>");
+MODULE_DESCRIPTION("Allwinner P2WI driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/i2c/muxes/Kconfig b/drivers/i2c/muxes/Kconfig
index f7f9865b8b89..f6d313e528de 100644
--- a/drivers/i2c/muxes/Kconfig
+++ b/drivers/i2c/muxes/Kconfig
@@ -40,6 +40,7 @@ config I2C_MUX_PCA9541
 
 config I2C_MUX_PCA954x
 	tristate "Philips PCA954x I2C Mux/switches"
+	depends on GPIOLIB
 	help
 	  If you say yes here you get support for the Philips PCA954x
 	  I2C mux/switch devices.
diff --git a/drivers/iio/accel/hid-sensor-accel-3d.c b/drivers/iio/accel/hid-sensor-accel-3d.c
index 69abf9163df7..54e464e4bb72 100644
--- a/drivers/iio/accel/hid-sensor-accel-3d.c
+++ b/drivers/iio/accel/hid-sensor-accel-3d.c
@@ -110,7 +110,6 @@ static int accel_3d_read_raw(struct iio_dev *indio_dev,
 	struct accel_3d_state *accel_state = iio_priv(indio_dev);
 	int report_id = -1;
 	u32 address;
-	int ret;
 	int ret_type;
 	s32 poll_value;
 
@@ -151,14 +150,12 @@ static int accel_3d_read_raw(struct iio_dev *indio_dev,
 		ret_type = IIO_VAL_INT;
 		break;
 	case IIO_CHAN_INFO_SAMP_FREQ:
-		ret = hid_sensor_read_samp_freq_value(
+		ret_type = hid_sensor_read_samp_freq_value(
 			&accel_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	case IIO_CHAN_INFO_HYSTERESIS:
-		ret = hid_sensor_read_raw_hyst_value(
+		ret_type = hid_sensor_read_raw_hyst_value(
 			&accel_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	default:
 		ret_type = -EINVAL;
diff --git a/drivers/iio/adc/ad799x.c b/drivers/iio/adc/ad799x.c
index 39b4cb48d738..6eba301ee03d 100644
--- a/drivers/iio/adc/ad799x.c
+++ b/drivers/iio/adc/ad799x.c
@@ -427,9 +427,12 @@ static int ad799x_write_event_value(struct iio_dev *indio_dev,
 	int ret;
 	struct ad799x_state *st = iio_priv(indio_dev);
 
+	if (val < 0 || val > RES_MASK(chan->scan_type.realbits))
+		return -EINVAL;
+
 	mutex_lock(&indio_dev->mlock);
 	ret = ad799x_i2c_write16(st, ad799x_threshold_reg(chan, dir, info),
-		val);
+		val << chan->scan_type.shift);
 	mutex_unlock(&indio_dev->mlock);
 
 	return ret;
@@ -452,7 +455,8 @@ static int ad799x_read_event_value(struct iio_dev *indio_dev,
 	mutex_unlock(&indio_dev->mlock);
 	if (ret < 0)
 		return ret;
-	*val = valin;
+	*val = (valin >> chan->scan_type.shift) &
+		RES_MASK(chan->scan_type.realbits);
 
 	return IIO_VAL_INT;
 }
diff --git a/drivers/iio/adc/at91_adc.c b/drivers/iio/adc/at91_adc.c
index 3b5bacd4d8da..2b6a9ce9927c 100644
--- a/drivers/iio/adc/at91_adc.c
+++ b/drivers/iio/adc/at91_adc.c
@@ -510,12 +510,11 @@ static int at91_adc_channel_init(struct iio_dev *idev)
 	return idev->num_channels;
 }
 
-static u8 at91_adc_get_trigger_value_by_name(struct iio_dev *idev,
+static int at91_adc_get_trigger_value_by_name(struct iio_dev *idev,
 					     struct at91_adc_trigger *triggers,
 					     const char *trigger_name)
 {
 	struct at91_adc_state *st = iio_priv(idev);
-	u8 value = 0;
 	int i;
 
 	for (i = 0; i < st->trigger_number; i++) {
@@ -528,15 +527,16 @@ static u8 at91_adc_get_trigger_value_by_name(struct iio_dev *idev,
 			return -ENOMEM;
 
 		if (strcmp(trigger_name, name) == 0) {
-			value = triggers[i].value;
 			kfree(name);
-			break;
+			if (triggers[i].value == 0)
+				return -EINVAL;
+			return triggers[i].value;
 		}
 
 		kfree(name);
 	}
 
-	return value;
+	return -EINVAL;
 }
 
 static int at91_adc_configure_trigger(struct iio_trigger *trig, bool state)
@@ -546,14 +546,14 @@ static int at91_adc_configure_trigger(struct iio_trigger *trig, bool state)
 	struct iio_buffer *buffer = idev->buffer;
 	struct at91_adc_reg_desc *reg = st->registers;
 	u32 status = at91_adc_readl(st, reg->trigger_register);
-	u8 value;
+	int value;
 	u8 bit;
 
 	value = at91_adc_get_trigger_value_by_name(idev,
 						   st->trigger_list,
 						   idev->trig->name);
-	if (value == 0)
-		return -EINVAL;
+	if (value < 0)
+		return value;
 
 	if (state) {
 		st->buffer = kmalloc(idev->scan_bytes, GFP_KERNEL);
diff --git a/drivers/iio/adc/men_z188_adc.c b/drivers/iio/adc/men_z188_adc.c
index 6989c16aec2b..b58d6302521f 100644
--- a/drivers/iio/adc/men_z188_adc.c
+++ b/drivers/iio/adc/men_z188_adc.c
@@ -121,8 +121,8 @@ static int men_z188_probe(struct mcb_device *dev,
 	indio_dev->num_channels = ARRAY_SIZE(z188_adc_iio_channels);
 
 	mem = mcb_request_mem(dev, "z188-adc");
-	if (!mem)
-		return -ENOMEM;
+	if (IS_ERR(mem))
+		return PTR_ERR(mem);
 
 	adc->base = ioremap(mem->start, resource_size(mem));
 	if (adc->base == NULL)
diff --git a/drivers/iio/adc/ti_am335x_adc.c b/drivers/iio/adc/ti_am335x_adc.c
index a4db3026bec6..d5dc4c6ce86c 100644
--- a/drivers/iio/adc/ti_am335x_adc.c
+++ b/drivers/iio/adc/ti_am335x_adc.c
@@ -374,7 +374,7 @@ static int tiadc_read_raw(struct iio_dev *indio_dev,
 			return -EAGAIN;
 		}
 	}
-	map_val = chan->channel + TOTAL_CHANNELS;
+	map_val = adc_dev->channel_step[chan->scan_index];
 
 	/*
 	 * We check the complete FIFO. We programmed just one entry but in case
diff --git a/drivers/iio/adc/twl4030-madc.c b/drivers/iio/adc/twl4030-madc.c
index 7de1c4c87942..eb86786e698e 100644
--- a/drivers/iio/adc/twl4030-madc.c
+++ b/drivers/iio/adc/twl4030-madc.c
@@ -645,6 +645,7 @@ int twl4030_get_madc_conversion(int channel_no)
 	req.channels = (1 << channel_no);
 	req.method = TWL4030_MADC_SW2;
 	req.active = 0;
+	req.raw = 0;
 	req.func_cb = NULL;
 	ret = twl4030_madc_conversion(&req);
 	if (ret < 0)
diff --git a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c
index 73282cee0c81..a3109a6f4d86 100644
--- a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c
+++ b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c
@@ -75,6 +75,9 @@ int hid_sensor_power_state(struct hid_sensor_common *st, bool state)
 					(s32)report_val);
 	}
 
+	sensor_hub_get_feature(st->hsdev, st->power_state.report_id,
+					st->power_state.index,
+					&state_val);
 	return 0;
 }
 EXPORT_SYMBOL(hid_sensor_power_state);
diff --git a/drivers/iio/gyro/hid-sensor-gyro-3d.c b/drivers/iio/gyro/hid-sensor-gyro-3d.c
index 40f4e4935d0d..fa034a3dad78 100644
--- a/drivers/iio/gyro/hid-sensor-gyro-3d.c
+++ b/drivers/iio/gyro/hid-sensor-gyro-3d.c
@@ -110,7 +110,6 @@ static int gyro_3d_read_raw(struct iio_dev *indio_dev,
 	struct gyro_3d_state *gyro_state = iio_priv(indio_dev);
 	int report_id = -1;
 	u32 address;
-	int ret;
 	int ret_type;
 	s32 poll_value;
 
@@ -151,14 +150,12 @@ static int gyro_3d_read_raw(struct iio_dev *indio_dev,
 		ret_type = IIO_VAL_INT;
 		break;
 	case IIO_CHAN_INFO_SAMP_FREQ:
-		ret = hid_sensor_read_samp_freq_value(
+		ret_type = hid_sensor_read_samp_freq_value(
 			&gyro_state->common_attributes, val, val2);
-			ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	case IIO_CHAN_INFO_HYSTERESIS:
-		ret = hid_sensor_read_raw_hyst_value(
+		ret_type = hid_sensor_read_raw_hyst_value(
 			&gyro_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	default:
 		ret_type = -EINVAL;
diff --git a/drivers/iio/inkern.c b/drivers/iio/inkern.c
index d833d55052ea..c7497009d60a 100644
--- a/drivers/iio/inkern.c
+++ b/drivers/iio/inkern.c
@@ -183,7 +183,7 @@ static struct iio_channel *of_iio_channel_get_by_name(struct device_node *np,
 		else if (name && index >= 0) {
 			pr_err("ERROR: could not get IIO channel %s:%s(%i)\n",
 				np->full_name, name ? name : "", index);
-			return chan;
+			return NULL;
 		}
 
 		/*
@@ -193,8 +193,9 @@ static struct iio_channel *of_iio_channel_get_by_name(struct device_node *np,
 		 */
 		np = np->parent;
 		if (np && !of_get_property(np, "io-channel-ranges", NULL))
-			break;
+			return NULL;
 	}
+
 	return chan;
 }
 
@@ -317,6 +318,7 @@ struct iio_channel *iio_channel_get(struct device *dev,
 		if (channel != NULL)
 			return channel;
 	}
+
 	return iio_channel_get_sys(name, channel_name);
 }
 EXPORT_SYMBOL_GPL(iio_channel_get);
diff --git a/drivers/iio/light/hid-sensor-als.c b/drivers/iio/light/hid-sensor-als.c
index f34c94380b41..96e71e103ea7 100644
--- a/drivers/iio/light/hid-sensor-als.c
+++ b/drivers/iio/light/hid-sensor-als.c
@@ -79,7 +79,6 @@ static int als_read_raw(struct iio_dev *indio_dev,
 	struct als_state *als_state = iio_priv(indio_dev);
 	int report_id = -1;
 	u32 address;
-	int ret;
 	int ret_type;
 	s32 poll_value;
 
@@ -129,14 +128,12 @@ static int als_read_raw(struct iio_dev *indio_dev,
 		ret_type = IIO_VAL_INT;
 		break;
 	case IIO_CHAN_INFO_SAMP_FREQ:
-		ret = hid_sensor_read_samp_freq_value(
+		ret_type = hid_sensor_read_samp_freq_value(
 				&als_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	case IIO_CHAN_INFO_HYSTERESIS:
-		ret = hid_sensor_read_raw_hyst_value(
+		ret_type = hid_sensor_read_raw_hyst_value(
 				&als_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	default:
 		ret_type = -EINVAL;
diff --git a/drivers/iio/light/hid-sensor-prox.c b/drivers/iio/light/hid-sensor-prox.c
index d203ef4d892f..412bae86d6ae 100644
--- a/drivers/iio/light/hid-sensor-prox.c
+++ b/drivers/iio/light/hid-sensor-prox.c
@@ -74,7 +74,6 @@ static int prox_read_raw(struct iio_dev *indio_dev,
 	struct prox_state *prox_state = iio_priv(indio_dev);
 	int report_id = -1;
 	u32 address;
-	int ret;
 	int ret_type;
 	s32 poll_value;
 
@@ -125,14 +124,12 @@ static int prox_read_raw(struct iio_dev *indio_dev,
 		ret_type = IIO_VAL_INT;
 		break;
 	case IIO_CHAN_INFO_SAMP_FREQ:
-		ret = hid_sensor_read_samp_freq_value(
+		ret_type = hid_sensor_read_samp_freq_value(
 				&prox_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	case IIO_CHAN_INFO_HYSTERESIS:
-		ret = hid_sensor_read_raw_hyst_value(
+		ret_type = hid_sensor_read_raw_hyst_value(
 				&prox_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	default:
 		ret_type = -EINVAL;
diff --git a/drivers/iio/light/tcs3472.c b/drivers/iio/light/tcs3472.c
index fe063a0a21cd..752569985d1d 100644
--- a/drivers/iio/light/tcs3472.c
+++ b/drivers/iio/light/tcs3472.c
@@ -52,6 +52,7 @@
 
 struct tcs3472_data {
 	struct i2c_client *client;
+	struct mutex lock;
 	u8 enable;
 	u8 control;
 	u8 atime;
@@ -116,10 +117,17 @@ static int tcs3472_read_raw(struct iio_dev *indio_dev,
 
 	switch (mask) {
 	case IIO_CHAN_INFO_RAW:
+		if (iio_buffer_enabled(indio_dev))
+			return -EBUSY;
+
+		mutex_lock(&data->lock);
 		ret = tcs3472_req_data(data);
-		if (ret < 0)
+		if (ret < 0) {
+			mutex_unlock(&data->lock);
 			return ret;
+		}
 		ret = i2c_smbus_read_word_data(data->client, chan->address);
+		mutex_unlock(&data->lock);
 		if (ret < 0)
 			return ret;
 		*val = ret;
@@ -255,6 +263,7 @@ static int tcs3472_probe(struct i2c_client *client,
 	data = iio_priv(indio_dev);
 	i2c_set_clientdata(client, indio_dev);
 	data->client = client;
+	mutex_init(&data->lock);
 
 	indio_dev->dev.parent = &client->dev;
 	indio_dev->info = &tcs3472_info;
diff --git a/drivers/iio/magnetometer/ak8975.c b/drivers/iio/magnetometer/ak8975.c
index 09ea5c481f4c..ea08313af0d2 100644
--- a/drivers/iio/magnetometer/ak8975.c
+++ b/drivers/iio/magnetometer/ak8975.c
@@ -373,8 +373,6 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 {
 	struct ak8975_data *data = iio_priv(indio_dev);
 	struct i2c_client *client = data->client;
-	u16 meas_reg;
-	s16 raw;
 	int ret;
 
 	mutex_lock(&data->lock);
@@ -422,16 +420,11 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 		dev_err(&client->dev, "Read axis data fails\n");
 		goto exit;
 	}
-	meas_reg = ret;
 
 	mutex_unlock(&data->lock);
 
-	/* Endian conversion of the measured values. */
-	raw = (s16) (le16_to_cpu(meas_reg));
-
 	/* Clamp to valid range. */
-	raw = clamp_t(s16, raw, -4096, 4095);
-	*val = raw;
+	*val = clamp_t(s16, ret, -4096, 4095);
 	return IIO_VAL_INT;
 
 exit:
diff --git a/drivers/iio/magnetometer/hid-sensor-magn-3d.c b/drivers/iio/magnetometer/hid-sensor-magn-3d.c
index 41cf29e2a371..b2b0937d5133 100644
--- a/drivers/iio/magnetometer/hid-sensor-magn-3d.c
+++ b/drivers/iio/magnetometer/hid-sensor-magn-3d.c
@@ -110,7 +110,6 @@ static int magn_3d_read_raw(struct iio_dev *indio_dev,
 	struct magn_3d_state *magn_state = iio_priv(indio_dev);
 	int report_id = -1;
 	u32 address;
-	int ret;
 	int ret_type;
 	s32 poll_value;
 
@@ -153,14 +152,12 @@ static int magn_3d_read_raw(struct iio_dev *indio_dev,
 		ret_type = IIO_VAL_INT;
 		break;
 	case IIO_CHAN_INFO_SAMP_FREQ:
-		ret = hid_sensor_read_samp_freq_value(
+		ret_type = hid_sensor_read_samp_freq_value(
 			&magn_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	case IIO_CHAN_INFO_HYSTERESIS:
-		ret = hid_sensor_read_raw_hyst_value(
+		ret_type = hid_sensor_read_raw_hyst_value(
 			&magn_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	default:
 		ret_type = -EINVAL;
diff --git a/drivers/iio/pressure/hid-sensor-press.c b/drivers/iio/pressure/hid-sensor-press.c
index 1cd190c73788..2c0d2a4fed8c 100644
--- a/drivers/iio/pressure/hid-sensor-press.c
+++ b/drivers/iio/pressure/hid-sensor-press.c
@@ -78,7 +78,6 @@ static int press_read_raw(struct iio_dev *indio_dev,
 	struct press_state *press_state = iio_priv(indio_dev);
 	int report_id = -1;
 	u32 address;
-	int ret;
 	int ret_type;
 	s32 poll_value;
 
@@ -128,14 +127,12 @@ static int press_read_raw(struct iio_dev *indio_dev,
 		ret_type = IIO_VAL_INT;
 		break;
 	case IIO_CHAN_INFO_SAMP_FREQ:
-		ret = hid_sensor_read_samp_freq_value(
+		ret_type = hid_sensor_read_samp_freq_value(
 				&press_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	case IIO_CHAN_INFO_HYSTERESIS:
-		ret = hid_sensor_read_raw_hyst_value(
+		ret_type = hid_sensor_read_raw_hyst_value(
 				&press_state->common_attributes, val, val2);
-		ret_type = IIO_VAL_INT_PLUS_MICRO;
 		break;
 	default:
 		ret_type = -EINVAL;
diff --git a/drivers/iio/pressure/mpl3115.c b/drivers/iio/pressure/mpl3115.c
index ba6d0c520e63..01b2e0b18878 100644
--- a/drivers/iio/pressure/mpl3115.c
+++ b/drivers/iio/pressure/mpl3115.c
@@ -98,7 +98,7 @@ static int mpl3115_read_raw(struct iio_dev *indio_dev,
 			mutex_unlock(&data->lock);
 			if (ret < 0)
 				return ret;
-			*val = sign_extend32(be32_to_cpu(tmp) >> 12, 23);
+			*val = be32_to_cpu(tmp) >> 12;
 			return IIO_VAL_INT;
 		case IIO_TEMP: /* in 0.0625 celsius / LSB */
 			mutex_lock(&data->lock);
@@ -112,7 +112,7 @@ static int mpl3115_read_raw(struct iio_dev *indio_dev,
 			mutex_unlock(&data->lock);
 			if (ret < 0)
 				return ret;
-			*val = sign_extend32(be32_to_cpu(tmp) >> 20, 15);
+			*val = sign_extend32(be32_to_cpu(tmp) >> 20, 11);
 			return IIO_VAL_INT;
 		default:
 			return -EINVAL;
@@ -185,7 +185,7 @@ static const struct iio_chan_spec mpl3115_channels[] = {
 			BIT(IIO_CHAN_INFO_SCALE),
 		.scan_index = 0,
 		.scan_type = {
-			.sign = 's',
+			.sign = 'u',
 			.realbits = 20,
 			.storagebits = 32,
 			.shift = 12,
diff --git a/drivers/infiniband/Makefile b/drivers/infiniband/Makefile
index bf508b5550c4..dc21836b5a8d 100644
--- a/drivers/infiniband/Makefile
+++ b/drivers/infiniband/Makefile
@@ -1,18 +1,3 @@
 obj-$(CONFIG_INFINIBAND)		+= core/
-obj-$(CONFIG_INFINIBAND_MTHCA)		+= hw/mthca/
-obj-$(CONFIG_INFINIBAND_IPATH)		+= hw/ipath/
-obj-$(CONFIG_INFINIBAND_QIB)		+= hw/qib/
-obj-$(CONFIG_INFINIBAND_EHCA)		+= hw/ehca/
-obj-$(CONFIG_INFINIBAND_AMSO1100)	+= hw/amso1100/
-obj-$(CONFIG_INFINIBAND_CXGB3)		+= hw/cxgb3/
-obj-$(CONFIG_INFINIBAND_CXGB4)		+= hw/cxgb4/
-obj-$(CONFIG_MLX4_INFINIBAND)		+= hw/mlx4/
-obj-$(CONFIG_MLX5_INFINIBAND)		+= hw/mlx5/
-obj-$(CONFIG_INFINIBAND_NES)		+= hw/nes/
-obj-$(CONFIG_INFINIBAND_OCRDMA)		+= hw/ocrdma/
-obj-$(CONFIG_INFINIBAND_USNIC)		+= hw/usnic/
-obj-$(CONFIG_INFINIBAND_IPOIB)		+= ulp/ipoib/
-obj-$(CONFIG_INFINIBAND_SRP)		+= ulp/srp/
-obj-$(CONFIG_INFINIBAND_SRPT)		+= ulp/srpt/
-obj-$(CONFIG_INFINIBAND_ISER)		+= ulp/iser/
-obj-$(CONFIG_INFINIBAND_ISERT)		+= ulp/isert/
+obj-$(CONFIG_INFINIBAND)		+= hw/
+obj-$(CONFIG_INFINIBAND)		+= ulp/
diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile
index 3ab3865544bb..ffd0af6734af 100644
--- a/drivers/infiniband/core/Makefile
+++ b/drivers/infiniband/core/Makefile
@@ -18,7 +18,7 @@ ib_sa-y :=			sa_query.o multicast.o
 
 ib_cm-y :=			cm.o
 
-iw_cm-y :=			iwcm.o
+iw_cm-y :=			iwcm.o iwpm_util.o iwpm_msg.o
 
 rdma_cm-y :=			cma.o
 
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 42c3058e6e9c..d570030d899c 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -3607,7 +3607,8 @@ static int cma_get_id_stats(struct sk_buff *skb, struct netlink_callback *cb)
 
 			id_stats = ibnl_put_msg(skb, &nlh, cb->nlh->nlmsg_seq,
 						sizeof *id_stats, RDMA_NL_RDMA_CM,
-						RDMA_NL_RDMA_CM_ID_STATS);
+						RDMA_NL_RDMA_CM_ID_STATS,
+						NLM_F_MULTI);
 			if (!id_stats)
 				goto out;
 
diff --git a/drivers/infiniband/core/iwpm_msg.c b/drivers/infiniband/core/iwpm_msg.c
new file mode 100644
index 000000000000..b85ddbc979e0
--- /dev/null
+++ b/drivers/infiniband/core/iwpm_msg.c
@@ -0,0 +1,685 @@
+/*
+ * Copyright (c) 2014 Intel Corporation. All rights reserved.
+ * Copyright (c) 2014 Chelsio, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "iwpm_util.h"
+
+static const char iwpm_ulib_name[] = "iWarpPortMapperUser";
+static int iwpm_ulib_version = 3;
+static int iwpm_user_pid = IWPM_PID_UNDEFINED;
+static atomic_t echo_nlmsg_seq;
+
+int iwpm_valid_pid(void)
+{
+	return iwpm_user_pid > 0;
+}
+EXPORT_SYMBOL(iwpm_valid_pid);
+
+/*
+ * iwpm_register_pid - Send a netlink query to user space
+ *                     for the iwarp port mapper pid
+ *
+ * nlmsg attributes:
+ *	[IWPM_NLA_REG_PID_SEQ]
+ *	[IWPM_NLA_REG_IF_NAME]
+ *	[IWPM_NLA_REG_IBDEV_NAME]
+ *	[IWPM_NLA_REG_ULIB_NAME]
+ */
+int iwpm_register_pid(struct iwpm_dev_data *pm_msg, u8 nl_client)
+{
+	struct sk_buff *skb = NULL;
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	struct nlmsghdr *nlh;
+	u32 msg_seq;
+	const char *err_str = "";
+	int ret = -EINVAL;
+
+	if (!iwpm_valid_client(nl_client)) {
+		err_str = "Invalid port mapper client";
+		goto pid_query_error;
+	}
+	if (iwpm_registered_client(nl_client))
+		return 0;
+	skb = iwpm_create_nlmsg(RDMA_NL_IWPM_REG_PID, &nlh, nl_client);
+	if (!skb) {
+		err_str = "Unable to create a nlmsg";
+		goto pid_query_error;
+	}
+	nlh->nlmsg_seq = iwpm_get_nlmsg_seq();
+	nlmsg_request = iwpm_get_nlmsg_request(nlh->nlmsg_seq, nl_client, GFP_KERNEL);
+	if (!nlmsg_request) {
+		err_str = "Unable to allocate netlink request";
+		goto pid_query_error;
+	}
+	msg_seq = atomic_read(&echo_nlmsg_seq);
+
+	/* fill in the pid request message */
+	err_str = "Unable to put attribute of the nlmsg";
+	ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, IWPM_NLA_REG_PID_SEQ);
+	if (ret)
+		goto pid_query_error;
+	ret = ibnl_put_attr(skb, nlh, IWPM_IFNAME_SIZE,
+				pm_msg->if_name, IWPM_NLA_REG_IF_NAME);
+	if (ret)
+		goto pid_query_error;
+	ret = ibnl_put_attr(skb, nlh, IWPM_DEVNAME_SIZE,
+				pm_msg->dev_name, IWPM_NLA_REG_IBDEV_NAME);
+	if (ret)
+		goto pid_query_error;
+	ret = ibnl_put_attr(skb, nlh, IWPM_ULIBNAME_SIZE,
+				(char *)iwpm_ulib_name, IWPM_NLA_REG_ULIB_NAME);
+	if (ret)
+		goto pid_query_error;
+
+	pr_debug("%s: Multicasting a nlmsg (dev = %s ifname = %s iwpm = %s)\n",
+		__func__, pm_msg->dev_name, pm_msg->if_name, iwpm_ulib_name);
+
+	ret = ibnl_multicast(skb, nlh, RDMA_NL_GROUP_IWPM, GFP_KERNEL);
+	if (ret) {
+		skb = NULL; /* skb is freed in the netlink send-op handling */
+		iwpm_set_registered(nl_client, 1);
+		iwpm_user_pid = IWPM_PID_UNAVAILABLE;
+		err_str = "Unable to send a nlmsg";
+		goto pid_query_error;
+	}
+	nlmsg_request->req_buffer = pm_msg;
+	ret = iwpm_wait_complete_req(nlmsg_request);
+	return ret;
+pid_query_error:
+	pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client);
+	if (skb)
+		dev_kfree_skb(skb);
+	if (nlmsg_request)
+		iwpm_free_nlmsg_request(&nlmsg_request->kref);
+	return ret;
+}
+EXPORT_SYMBOL(iwpm_register_pid);
+
+/*
+ * iwpm_add_mapping - Send a netlink add mapping message
+ *                    to the port mapper
+ * nlmsg attributes:
+ *	[IWPM_NLA_MANAGE_MAPPING_SEQ]
+ *	[IWPM_NLA_MANAGE_ADDR]
+ */
+int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
+{
+	struct sk_buff *skb = NULL;
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	struct nlmsghdr *nlh;
+	u32 msg_seq;
+	const char *err_str = "";
+	int ret = -EINVAL;
+
+	if (!iwpm_valid_client(nl_client)) {
+		err_str = "Invalid port mapper client";
+		goto add_mapping_error;
+	}
+	if (!iwpm_registered_client(nl_client)) {
+		err_str = "Unregistered port mapper client";
+		goto add_mapping_error;
+	}
+	if (!iwpm_valid_pid())
+		return 0;
+	skb = iwpm_create_nlmsg(RDMA_NL_IWPM_ADD_MAPPING, &nlh, nl_client);
+	if (!skb) {
+		err_str = "Unable to create a nlmsg";
+		goto add_mapping_error;
+	}
+	nlh->nlmsg_seq = iwpm_get_nlmsg_seq();
+	nlmsg_request = iwpm_get_nlmsg_request(nlh->nlmsg_seq, nl_client, GFP_KERNEL);
+	if (!nlmsg_request) {
+		err_str = "Unable to allocate netlink request";
+		goto add_mapping_error;
+	}
+	msg_seq = atomic_read(&echo_nlmsg_seq);
+	/* fill in the add mapping message */
+	err_str = "Unable to put attribute of the nlmsg";
+	ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq,
+				IWPM_NLA_MANAGE_MAPPING_SEQ);
+	if (ret)
+		goto add_mapping_error;
+	ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage),
+				&pm_msg->loc_addr, IWPM_NLA_MANAGE_ADDR);
+	if (ret)
+		goto add_mapping_error;
+	nlmsg_request->req_buffer = pm_msg;
+
+	ret = ibnl_unicast(skb, nlh, iwpm_user_pid);
+	if (ret) {
+		skb = NULL; /* skb is freed in the netlink send-op handling */
+		iwpm_user_pid = IWPM_PID_UNDEFINED;
+		err_str = "Unable to send a nlmsg";
+		goto add_mapping_error;
+	}
+	ret = iwpm_wait_complete_req(nlmsg_request);
+	return ret;
+add_mapping_error:
+	pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client);
+	if (skb)
+		dev_kfree_skb(skb);
+	if (nlmsg_request)
+		iwpm_free_nlmsg_request(&nlmsg_request->kref);
+	return ret;
+}
+EXPORT_SYMBOL(iwpm_add_mapping);
+
+/*
+ * iwpm_add_and_query_mapping - Send a netlink add and query
+ *                              mapping message to the port mapper
+ * nlmsg attributes:
+ *	[IWPM_NLA_QUERY_MAPPING_SEQ]
+ *	[IWPM_NLA_QUERY_LOCAL_ADDR]
+ *	[IWPM_NLA_QUERY_REMOTE_ADDR]
+ */
+int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
+{
+	struct sk_buff *skb = NULL;
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	struct nlmsghdr *nlh;
+	u32 msg_seq;
+	const char *err_str = "";
+	int ret = -EINVAL;
+
+	if (!iwpm_valid_client(nl_client)) {
+		err_str = "Invalid port mapper client";
+		goto query_mapping_error;
+	}
+	if (!iwpm_registered_client(nl_client)) {
+		err_str = "Unregistered port mapper client";
+		goto query_mapping_error;
+	}
+	if (!iwpm_valid_pid())
+		return 0;
+	ret = -ENOMEM;
+	skb = iwpm_create_nlmsg(RDMA_NL_IWPM_QUERY_MAPPING, &nlh, nl_client);
+	if (!skb) {
+		err_str = "Unable to create a nlmsg";
+		goto query_mapping_error;
+	}
+	nlh->nlmsg_seq = iwpm_get_nlmsg_seq();
+	nlmsg_request = iwpm_get_nlmsg_request(nlh->nlmsg_seq,
+				nl_client, GFP_KERNEL);
+	if (!nlmsg_request) {
+		err_str = "Unable to allocate netlink request";
+		goto query_mapping_error;
+	}
+	msg_seq = atomic_read(&echo_nlmsg_seq);
+
+	/* fill in the query message */
+	err_str = "Unable to put attribute of the nlmsg";
+	ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq,
+				IWPM_NLA_QUERY_MAPPING_SEQ);
+	if (ret)
+		goto query_mapping_error;
+	ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage),
+				&pm_msg->loc_addr, IWPM_NLA_QUERY_LOCAL_ADDR);
+	if (ret)
+		goto query_mapping_error;
+	ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage),
+				&pm_msg->rem_addr, IWPM_NLA_QUERY_REMOTE_ADDR);
+	if (ret)
+		goto query_mapping_error;
+	nlmsg_request->req_buffer = pm_msg;
+
+	ret = ibnl_unicast(skb, nlh, iwpm_user_pid);
+	if (ret) {
+		skb = NULL; /* skb is freed in the netlink send-op handling */
+		err_str = "Unable to send a nlmsg";
+		goto query_mapping_error;
+	}
+	ret = iwpm_wait_complete_req(nlmsg_request);
+	return ret;
+query_mapping_error:
+	pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client);
+	if (skb)
+		dev_kfree_skb(skb);
+	if (nlmsg_request)
+		iwpm_free_nlmsg_request(&nlmsg_request->kref);
+	return ret;
+}
+EXPORT_SYMBOL(iwpm_add_and_query_mapping);
+
+/*
+ * iwpm_remove_mapping - Send a netlink remove mapping message
+ *                       to the port mapper
+ * nlmsg attributes:
+ *	[IWPM_NLA_MANAGE_MAPPING_SEQ]
+ *	[IWPM_NLA_MANAGE_ADDR]
+ */
+int iwpm_remove_mapping(struct sockaddr_storage *local_addr, u8 nl_client)
+{
+	struct sk_buff *skb = NULL;
+	struct nlmsghdr *nlh;
+	u32 msg_seq;
+	const char *err_str = "";
+	int ret = -EINVAL;
+
+	if (!iwpm_valid_client(nl_client)) {
+		err_str = "Invalid port mapper client";
+		goto remove_mapping_error;
+	}
+	if (!iwpm_registered_client(nl_client)) {
+		err_str = "Unregistered port mapper client";
+		goto remove_mapping_error;
+	}
+	if (!iwpm_valid_pid())
+		return 0;
+	skb = iwpm_create_nlmsg(RDMA_NL_IWPM_REMOVE_MAPPING, &nlh, nl_client);
+	if (!skb) {
+		ret = -ENOMEM;
+		err_str = "Unable to create a nlmsg";
+		goto remove_mapping_error;
+	}
+	msg_seq = atomic_read(&echo_nlmsg_seq);
+	nlh->nlmsg_seq = iwpm_get_nlmsg_seq();
+	err_str = "Unable to put attribute of the nlmsg";
+	ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq,
+				IWPM_NLA_MANAGE_MAPPING_SEQ);
+	if (ret)
+		goto remove_mapping_error;
+	ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage),
+				local_addr, IWPM_NLA_MANAGE_ADDR);
+	if (ret)
+		goto remove_mapping_error;
+
+	ret = ibnl_unicast(skb, nlh, iwpm_user_pid);
+	if (ret) {
+		skb = NULL; /* skb is freed in the netlink send-op handling */
+		iwpm_user_pid = IWPM_PID_UNDEFINED;
+		err_str = "Unable to send a nlmsg";
+		goto remove_mapping_error;
+	}
+	iwpm_print_sockaddr(local_addr,
+			"remove_mapping: Local sockaddr:");
+	return 0;
+remove_mapping_error:
+	pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client);
+	if (skb)
+		dev_kfree_skb_any(skb);
+	return ret;
+}
+EXPORT_SYMBOL(iwpm_remove_mapping);
+
+/* netlink attribute policy for the received response to register pid request */
+static const struct nla_policy resp_reg_policy[IWPM_NLA_RREG_PID_MAX] = {
+	[IWPM_NLA_RREG_PID_SEQ]     = { .type = NLA_U32 },
+	[IWPM_NLA_RREG_IBDEV_NAME]  = { .type = NLA_STRING,
+					.len = IWPM_DEVNAME_SIZE - 1 },
+	[IWPM_NLA_RREG_ULIB_NAME]   = { .type = NLA_STRING,
+					.len = IWPM_ULIBNAME_SIZE - 1 },
+	[IWPM_NLA_RREG_ULIB_VER]    = { .type = NLA_U16 },
+	[IWPM_NLA_RREG_PID_ERR]     = { .type = NLA_U16 }
+};
+
+/*
+ * iwpm_register_pid_cb - Process a port mapper response to
+ *                        iwpm_register_pid()
+ */
+int iwpm_register_pid_cb(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	struct nlattr *nltb[IWPM_NLA_RREG_PID_MAX];
+	struct iwpm_dev_data *pm_msg;
+	char *dev_name, *iwpm_name;
+	u32 msg_seq;
+	u8 nl_client;
+	u16 iwpm_version;
+	const char *msg_type = "Register Pid response";
+
+	if (iwpm_parse_nlmsg(cb, IWPM_NLA_RREG_PID_MAX,
+				resp_reg_policy, nltb, msg_type))
+		return -EINVAL;
+
+	msg_seq = nla_get_u32(nltb[IWPM_NLA_RREG_PID_SEQ]);
+	nlmsg_request = iwpm_find_nlmsg_request(msg_seq);
+	if (!nlmsg_request) {
+		pr_info("%s: Could not find a matching request (seq = %u)\n",
+				 __func__, msg_seq);
+		return -EINVAL;
+	}
+	pm_msg = nlmsg_request->req_buffer;
+	nl_client = nlmsg_request->nl_client;
+	dev_name = (char *)nla_data(nltb[IWPM_NLA_RREG_IBDEV_NAME]);
+	iwpm_name = (char *)nla_data(nltb[IWPM_NLA_RREG_ULIB_NAME]);
+	iwpm_version = nla_get_u16(nltb[IWPM_NLA_RREG_ULIB_VER]);
+
+	/* check device name, ulib name and version */
+	if (strcmp(pm_msg->dev_name, dev_name) ||
+			strcmp(iwpm_ulib_name, iwpm_name) ||
+			iwpm_version != iwpm_ulib_version) {
+
+		pr_info("%s: Incorrect info (dev = %s name = %s version = %d)\n",
+				__func__, dev_name, iwpm_name, iwpm_version);
+		nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR;
+		goto register_pid_response_exit;
+	}
+	iwpm_user_pid = cb->nlh->nlmsg_pid;
+	atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
+	pr_debug("%s: iWarp Port Mapper (pid = %d) is available!\n",
+			__func__, iwpm_user_pid);
+	if (iwpm_valid_client(nl_client))
+		iwpm_set_registered(nl_client, 1);
+register_pid_response_exit:
+	nlmsg_request->request_done = 1;
+	/* always for found nlmsg_request */
+	kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request);
+	barrier();
+	wake_up(&nlmsg_request->waitq);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_register_pid_cb);
+
+/* netlink attribute policy for the received response to add mapping request */
+static const struct nla_policy resp_add_policy[IWPM_NLA_RMANAGE_MAPPING_MAX] = {
+	[IWPM_NLA_MANAGE_MAPPING_SEQ]     = { .type = NLA_U32 },
+	[IWPM_NLA_MANAGE_ADDR]            = { .len = sizeof(struct sockaddr_storage) },
+	[IWPM_NLA_MANAGE_MAPPED_LOC_ADDR] = { .len = sizeof(struct sockaddr_storage) },
+	[IWPM_NLA_RMANAGE_MAPPING_ERR]	  = { .type = NLA_U16 }
+};
+
+/*
+ * iwpm_add_mapping_cb - Process a port mapper response to
+ *                       iwpm_add_mapping()
+ */
+int iwpm_add_mapping_cb(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct iwpm_sa_data *pm_msg;
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	struct nlattr *nltb[IWPM_NLA_RMANAGE_MAPPING_MAX];
+	struct sockaddr_storage *local_sockaddr;
+	struct sockaddr_storage *mapped_sockaddr;
+	const char *msg_type;
+	u32 msg_seq;
+
+	msg_type = "Add Mapping response";
+	if (iwpm_parse_nlmsg(cb, IWPM_NLA_RMANAGE_MAPPING_MAX,
+				resp_add_policy, nltb, msg_type))
+		return -EINVAL;
+
+	atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
+
+	msg_seq = nla_get_u32(nltb[IWPM_NLA_MANAGE_MAPPING_SEQ]);
+	nlmsg_request = iwpm_find_nlmsg_request(msg_seq);
+	if (!nlmsg_request) {
+		pr_info("%s: Could not find a matching request (seq = %u)\n",
+				 __func__, msg_seq);
+		return -EINVAL;
+	}
+	pm_msg = nlmsg_request->req_buffer;
+	local_sockaddr = (struct sockaddr_storage *)
+			nla_data(nltb[IWPM_NLA_MANAGE_ADDR]);
+	mapped_sockaddr = (struct sockaddr_storage *)
+			nla_data(nltb[IWPM_NLA_MANAGE_MAPPED_LOC_ADDR]);
+
+	if (iwpm_compare_sockaddr(local_sockaddr, &pm_msg->loc_addr)) {
+		nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR;
+		goto add_mapping_response_exit;
+	}
+	if (mapped_sockaddr->ss_family != local_sockaddr->ss_family) {
+		pr_info("%s: Sockaddr family doesn't match the requested one\n",
+				__func__);
+		nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR;
+		goto add_mapping_response_exit;
+	}
+	memcpy(&pm_msg->mapped_loc_addr, mapped_sockaddr,
+			sizeof(*mapped_sockaddr));
+	iwpm_print_sockaddr(&pm_msg->loc_addr,
+			"add_mapping: Local sockaddr:");
+	iwpm_print_sockaddr(&pm_msg->mapped_loc_addr,
+			"add_mapping: Mapped local sockaddr:");
+
+add_mapping_response_exit:
+	nlmsg_request->request_done = 1;
+	/* always for found request */
+	kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request);
+	barrier();
+	wake_up(&nlmsg_request->waitq);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_add_mapping_cb);
+
+/* netlink attribute policy for the response to add and query mapping request */
+static const struct nla_policy resp_query_policy[IWPM_NLA_RQUERY_MAPPING_MAX] = {
+	[IWPM_NLA_QUERY_MAPPING_SEQ]      = { .type = NLA_U32 },
+	[IWPM_NLA_QUERY_LOCAL_ADDR]       = { .len = sizeof(struct sockaddr_storage) },
+	[IWPM_NLA_QUERY_REMOTE_ADDR]      = { .len = sizeof(struct sockaddr_storage) },
+	[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR] = { .len = sizeof(struct sockaddr_storage) },
+	[IWPM_NLA_RQUERY_MAPPED_REM_ADDR] = { .len = sizeof(struct sockaddr_storage) },
+	[IWPM_NLA_RQUERY_MAPPING_ERR]	  = { .type = NLA_U16 }
+};
+
+/*
+ * iwpm_add_and_query_mapping_cb - Process a port mapper response to
+ *                                 iwpm_add_and_query_mapping()
+ */
+int iwpm_add_and_query_mapping_cb(struct sk_buff *skb,
+				struct netlink_callback *cb)
+{
+	struct iwpm_sa_data *pm_msg;
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	struct nlattr *nltb[IWPM_NLA_RQUERY_MAPPING_MAX];
+	struct sockaddr_storage *local_sockaddr, *remote_sockaddr;
+	struct sockaddr_storage *mapped_loc_sockaddr, *mapped_rem_sockaddr;
+	const char *msg_type;
+	u32 msg_seq;
+	u16 err_code;
+
+	msg_type = "Query Mapping response";
+	if (iwpm_parse_nlmsg(cb, IWPM_NLA_RQUERY_MAPPING_MAX,
+				resp_query_policy, nltb, msg_type))
+		return -EINVAL;
+	atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
+
+	msg_seq = nla_get_u32(nltb[IWPM_NLA_QUERY_MAPPING_SEQ]);
+	nlmsg_request = iwpm_find_nlmsg_request(msg_seq);
+	if (!nlmsg_request) {
+		pr_info("%s: Could not find a matching request (seq = %u)\n",
+				 __func__, msg_seq);
+			return -EINVAL;
+	}
+	pm_msg = nlmsg_request->req_buffer;
+	local_sockaddr = (struct sockaddr_storage *)
+			nla_data(nltb[IWPM_NLA_QUERY_LOCAL_ADDR]);
+	remote_sockaddr = (struct sockaddr_storage *)
+			nla_data(nltb[IWPM_NLA_QUERY_REMOTE_ADDR]);
+	mapped_loc_sockaddr = (struct sockaddr_storage *)
+			nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR]);
+	mapped_rem_sockaddr = (struct sockaddr_storage *)
+			nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_REM_ADDR]);
+
+	err_code = nla_get_u16(nltb[IWPM_NLA_RQUERY_MAPPING_ERR]);
+	if (err_code == IWPM_REMOTE_QUERY_REJECT) {
+		pr_info("%s: Received a Reject (pid = %u, echo seq = %u)\n",
+			__func__, cb->nlh->nlmsg_pid, msg_seq);
+		nlmsg_request->err_code = IWPM_REMOTE_QUERY_REJECT;
+	}
+	if (iwpm_compare_sockaddr(local_sockaddr, &pm_msg->loc_addr) ||
+		iwpm_compare_sockaddr(remote_sockaddr, &pm_msg->rem_addr)) {
+		pr_info("%s: Incorrect local sockaddr\n", __func__);
+		nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR;
+		goto query_mapping_response_exit;
+	}
+	if (mapped_loc_sockaddr->ss_family != local_sockaddr->ss_family ||
+		mapped_rem_sockaddr->ss_family != remote_sockaddr->ss_family) {
+		pr_info("%s: Sockaddr family doesn't match the requested one\n",
+				__func__);
+		nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR;
+		goto query_mapping_response_exit;
+	}
+	memcpy(&pm_msg->mapped_loc_addr, mapped_loc_sockaddr,
+			sizeof(*mapped_loc_sockaddr));
+	memcpy(&pm_msg->mapped_rem_addr, mapped_rem_sockaddr,
+			sizeof(*mapped_rem_sockaddr));
+
+	iwpm_print_sockaddr(&pm_msg->loc_addr,
+			"query_mapping: Local sockaddr:");
+	iwpm_print_sockaddr(&pm_msg->mapped_loc_addr,
+			"query_mapping: Mapped local sockaddr:");
+	iwpm_print_sockaddr(&pm_msg->rem_addr,
+			"query_mapping: Remote sockaddr:");
+	iwpm_print_sockaddr(&pm_msg->mapped_rem_addr,
+			"query_mapping: Mapped remote sockaddr:");
+query_mapping_response_exit:
+	nlmsg_request->request_done = 1;
+	/* always for found request */
+	kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request);
+	barrier();
+	wake_up(&nlmsg_request->waitq);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_add_and_query_mapping_cb);
+
+/* netlink attribute policy for the received request for mapping info */
+static const struct nla_policy resp_mapinfo_policy[IWPM_NLA_MAPINFO_REQ_MAX] = {
+	[IWPM_NLA_MAPINFO_ULIB_NAME] = { .type = NLA_STRING,
+					.len = IWPM_ULIBNAME_SIZE - 1 },
+	[IWPM_NLA_MAPINFO_ULIB_VER]  = { .type = NLA_U16 }
+};
+
+/*
+ * iwpm_mapping_info_cb - Process a port mapper request for mapping info
+ */
+int iwpm_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct nlattr *nltb[IWPM_NLA_MAPINFO_REQ_MAX];
+	const char *msg_type = "Mapping Info response";
+	int iwpm_pid;
+	u8 nl_client;
+	char *iwpm_name;
+	u16 iwpm_version;
+	int ret = -EINVAL;
+
+	if (iwpm_parse_nlmsg(cb, IWPM_NLA_MAPINFO_REQ_MAX,
+				resp_mapinfo_policy, nltb, msg_type)) {
+		pr_info("%s: Unable to parse nlmsg\n", __func__);
+		return ret;
+	}
+	iwpm_name = (char *)nla_data(nltb[IWPM_NLA_MAPINFO_ULIB_NAME]);
+	iwpm_version = nla_get_u16(nltb[IWPM_NLA_MAPINFO_ULIB_VER]);
+	if (strcmp(iwpm_ulib_name, iwpm_name) ||
+			iwpm_version != iwpm_ulib_version) {
+		pr_info("%s: Invalid port mapper name = %s version = %d\n",
+				__func__, iwpm_name, iwpm_version);
+		return ret;
+	}
+	nl_client = RDMA_NL_GET_CLIENT(cb->nlh->nlmsg_type);
+	if (!iwpm_valid_client(nl_client)) {
+		pr_info("%s: Invalid port mapper client = %d\n",
+				__func__, nl_client);
+		return ret;
+	}
+	iwpm_set_registered(nl_client, 0);
+	atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
+	if (!iwpm_mapinfo_available())
+		return 0;
+	iwpm_pid = cb->nlh->nlmsg_pid;
+	pr_debug("%s: iWarp Port Mapper (pid = %d) is available!\n",
+		 __func__, iwpm_pid);
+	ret = iwpm_send_mapinfo(nl_client, iwpm_pid);
+	return ret;
+}
+EXPORT_SYMBOL(iwpm_mapping_info_cb);
+
+/* netlink attribute policy for the received mapping info ack */
+static const struct nla_policy ack_mapinfo_policy[IWPM_NLA_MAPINFO_NUM_MAX] = {
+	[IWPM_NLA_MAPINFO_SEQ]    =   { .type = NLA_U32 },
+	[IWPM_NLA_MAPINFO_SEND_NUM] = { .type = NLA_U32 },
+	[IWPM_NLA_MAPINFO_ACK_NUM] =  { .type = NLA_U32 }
+};
+
+/*
+ * iwpm_ack_mapping_info_cb - Process a port mapper ack for
+ *                            the provided mapping info records
+ */
+int iwpm_ack_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct nlattr *nltb[IWPM_NLA_MAPINFO_NUM_MAX];
+	u32 mapinfo_send, mapinfo_ack;
+	const char *msg_type = "Mapping Info Ack";
+
+	if (iwpm_parse_nlmsg(cb, IWPM_NLA_MAPINFO_NUM_MAX,
+				ack_mapinfo_policy, nltb, msg_type))
+		return -EINVAL;
+	mapinfo_send = nla_get_u32(nltb[IWPM_NLA_MAPINFO_SEND_NUM]);
+	mapinfo_ack = nla_get_u32(nltb[IWPM_NLA_MAPINFO_ACK_NUM]);
+	if (mapinfo_ack != mapinfo_send)
+		pr_info("%s: Invalid mapinfo number (sent = %u ack-ed = %u)\n",
+			__func__, mapinfo_send, mapinfo_ack);
+	atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_ack_mapping_info_cb);
+
+/* netlink attribute policy for the received port mapper error message */
+static const struct nla_policy map_error_policy[IWPM_NLA_ERR_MAX] = {
+	[IWPM_NLA_ERR_SEQ]        = { .type = NLA_U32 },
+	[IWPM_NLA_ERR_CODE]       = { .type = NLA_U16 },
+};
+
+/*
+ * iwpm_mapping_error_cb - Process a port mapper error message
+ */
+int iwpm_mapping_error_cb(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	int nl_client = RDMA_NL_GET_CLIENT(cb->nlh->nlmsg_type);
+	struct nlattr *nltb[IWPM_NLA_ERR_MAX];
+	u32 msg_seq;
+	u16 err_code;
+	const char *msg_type = "Mapping Error Msg";
+
+	if (iwpm_parse_nlmsg(cb, IWPM_NLA_ERR_MAX,
+				map_error_policy, nltb, msg_type))
+		return -EINVAL;
+
+	msg_seq = nla_get_u32(nltb[IWPM_NLA_ERR_SEQ]);
+	err_code = nla_get_u16(nltb[IWPM_NLA_ERR_CODE]);
+	pr_info("%s: Received msg seq = %u err code = %u client = %d\n",
+				__func__, msg_seq, err_code, nl_client);
+	/* look for nlmsg_request */
+	nlmsg_request = iwpm_find_nlmsg_request(msg_seq);
+	if (!nlmsg_request) {
+		/* not all errors have associated requests */
+		pr_debug("Could not find matching req (seq = %u)\n", msg_seq);
+		return 0;
+	}
+	atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
+	nlmsg_request->err_code = err_code;
+	nlmsg_request->request_done = 1;
+	/* always for found request */
+	kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request);
+	barrier();
+	wake_up(&nlmsg_request->waitq);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_mapping_error_cb);
diff --git a/drivers/infiniband/core/iwpm_util.c b/drivers/infiniband/core/iwpm_util.c
new file mode 100644
index 000000000000..69e9f84c1605
--- /dev/null
+++ b/drivers/infiniband/core/iwpm_util.c
@@ -0,0 +1,607 @@
+/*
+ * Copyright (c) 2014 Chelsio, Inc. All rights reserved.
+ * Copyright (c) 2014 Intel Corporation. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *	  copyright notice, this list of conditions and the following
+ *	  disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *	  copyright notice, this list of conditions and the following
+ *	  disclaimer in the documentation and/or other materials
+ *	  provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "iwpm_util.h"
+
+#define IWPM_HASH_BUCKET_SIZE	512
+#define IWPM_HASH_BUCKET_MASK	(IWPM_HASH_BUCKET_SIZE - 1)
+
+static LIST_HEAD(iwpm_nlmsg_req_list);
+static DEFINE_SPINLOCK(iwpm_nlmsg_req_lock);
+
+static struct hlist_head *iwpm_hash_bucket;
+static DEFINE_SPINLOCK(iwpm_mapinfo_lock);
+
+static DEFINE_MUTEX(iwpm_admin_lock);
+static struct iwpm_admin_data iwpm_admin;
+
+int iwpm_init(u8 nl_client)
+{
+	if (iwpm_valid_client(nl_client))
+		return -EINVAL;
+	mutex_lock(&iwpm_admin_lock);
+	if (atomic_read(&iwpm_admin.refcount) == 0) {
+		iwpm_hash_bucket = kzalloc(IWPM_HASH_BUCKET_SIZE *
+					sizeof(struct hlist_head), GFP_KERNEL);
+		if (!iwpm_hash_bucket) {
+			mutex_unlock(&iwpm_admin_lock);
+			pr_err("%s Unable to create mapinfo hash table\n", __func__);
+			return -ENOMEM;
+		}
+	}
+	atomic_inc(&iwpm_admin.refcount);
+	mutex_unlock(&iwpm_admin_lock);
+	iwpm_set_valid(nl_client, 1);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_init);
+
+static void free_hash_bucket(void);
+
+int iwpm_exit(u8 nl_client)
+{
+
+	if (!iwpm_valid_client(nl_client))
+		return -EINVAL;
+	mutex_lock(&iwpm_admin_lock);
+	if (atomic_read(&iwpm_admin.refcount) == 0) {
+		mutex_unlock(&iwpm_admin_lock);
+		pr_err("%s Incorrect usage - negative refcount\n", __func__);
+		return -EINVAL;
+	}
+	if (atomic_dec_and_test(&iwpm_admin.refcount)) {
+		free_hash_bucket();
+		pr_debug("%s: Mapinfo hash table is destroyed\n", __func__);
+	}
+	mutex_unlock(&iwpm_admin_lock);
+	iwpm_set_valid(nl_client, 0);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_exit);
+
+static struct hlist_head *get_hash_bucket_head(struct sockaddr_storage *,
+					       struct sockaddr_storage *);
+
+int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr,
+			struct sockaddr_storage *mapped_sockaddr,
+			u8 nl_client)
+{
+	struct hlist_head *hash_bucket_head;
+	struct iwpm_mapping_info *map_info;
+	unsigned long flags;
+
+	if (!iwpm_valid_client(nl_client))
+		return -EINVAL;
+	map_info = kzalloc(sizeof(struct iwpm_mapping_info), GFP_KERNEL);
+	if (!map_info) {
+		pr_err("%s: Unable to allocate a mapping info\n", __func__);
+		return -ENOMEM;
+	}
+	memcpy(&map_info->local_sockaddr, local_sockaddr,
+	       sizeof(struct sockaddr_storage));
+	memcpy(&map_info->mapped_sockaddr, mapped_sockaddr,
+	       sizeof(struct sockaddr_storage));
+	map_info->nl_client = nl_client;
+
+	spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
+	if (iwpm_hash_bucket) {
+		hash_bucket_head = get_hash_bucket_head(
+					&map_info->local_sockaddr,
+					&map_info->mapped_sockaddr);
+		hlist_add_head(&map_info->hlist_node, hash_bucket_head);
+	}
+	spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags);
+	return 0;
+}
+EXPORT_SYMBOL(iwpm_create_mapinfo);
+
+int iwpm_remove_mapinfo(struct sockaddr_storage *local_sockaddr,
+			struct sockaddr_storage *mapped_local_addr)
+{
+	struct hlist_node *tmp_hlist_node;
+	struct hlist_head *hash_bucket_head;
+	struct iwpm_mapping_info *map_info = NULL;
+	unsigned long flags;
+	int ret = -EINVAL;
+
+	spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
+	if (iwpm_hash_bucket) {
+		hash_bucket_head = get_hash_bucket_head(
+					local_sockaddr,
+					mapped_local_addr);
+		hlist_for_each_entry_safe(map_info, tmp_hlist_node,
+					hash_bucket_head, hlist_node) {
+
+			if (!iwpm_compare_sockaddr(&map_info->mapped_sockaddr,
+						mapped_local_addr)) {
+
+				hlist_del_init(&map_info->hlist_node);
+				kfree(map_info);
+				ret = 0;
+				break;
+			}
+		}
+	}
+	spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags);
+	return ret;
+}
+EXPORT_SYMBOL(iwpm_remove_mapinfo);
+
+static void free_hash_bucket(void)
+{
+	struct hlist_node *tmp_hlist_node;
+	struct iwpm_mapping_info *map_info;
+	unsigned long flags;
+	int i;
+
+	/* remove all the mapinfo data from the list */
+	spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
+	for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) {
+		hlist_for_each_entry_safe(map_info, tmp_hlist_node,
+			&iwpm_hash_bucket[i], hlist_node) {
+
+				hlist_del_init(&map_info->hlist_node);
+				kfree(map_info);
+			}
+	}
+	/* free the hash list */
+	kfree(iwpm_hash_bucket);
+	iwpm_hash_bucket = NULL;
+	spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags);
+}
+
+struct iwpm_nlmsg_request *iwpm_get_nlmsg_request(__u32 nlmsg_seq,
+					u8 nl_client, gfp_t gfp)
+{
+	struct iwpm_nlmsg_request *nlmsg_request = NULL;
+	unsigned long flags;
+
+	nlmsg_request = kzalloc(sizeof(struct iwpm_nlmsg_request), gfp);
+	if (!nlmsg_request) {
+		pr_err("%s Unable to allocate a nlmsg_request\n", __func__);
+		return NULL;
+	}
+	spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags);
+	list_add_tail(&nlmsg_request->inprocess_list, &iwpm_nlmsg_req_list);
+	spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags);
+
+	kref_init(&nlmsg_request->kref);
+	kref_get(&nlmsg_request->kref);
+	nlmsg_request->nlmsg_seq = nlmsg_seq;
+	nlmsg_request->nl_client = nl_client;
+	nlmsg_request->request_done = 0;
+	nlmsg_request->err_code = 0;
+	return nlmsg_request;
+}
+
+void iwpm_free_nlmsg_request(struct kref *kref)
+{
+	struct iwpm_nlmsg_request *nlmsg_request;
+	unsigned long flags;
+
+	nlmsg_request = container_of(kref, struct iwpm_nlmsg_request, kref);
+
+	spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags);
+	list_del_init(&nlmsg_request->inprocess_list);
+	spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags);
+
+	if (!nlmsg_request->request_done)
+		pr_debug("%s Freeing incomplete nlmsg request (seq = %u).\n",
+			__func__, nlmsg_request->nlmsg_seq);
+	kfree(nlmsg_request);
+}
+
+struct iwpm_nlmsg_request *iwpm_find_nlmsg_request(__u32 echo_seq)
+{
+	struct iwpm_nlmsg_request *nlmsg_request;
+	struct iwpm_nlmsg_request *found_request = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags);
+	list_for_each_entry(nlmsg_request, &iwpm_nlmsg_req_list,
+			    inprocess_list) {
+		if (nlmsg_request->nlmsg_seq == echo_seq) {
+			found_request = nlmsg_request;
+			kref_get(&nlmsg_request->kref);
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags);
+	return found_request;
+}
+
+int iwpm_wait_complete_req(struct iwpm_nlmsg_request *nlmsg_request)
+{
+	int ret;
+	init_waitqueue_head(&nlmsg_request->waitq);
+
+	ret = wait_event_timeout(nlmsg_request->waitq,
+			(nlmsg_request->request_done != 0), IWPM_NL_TIMEOUT);
+	if (!ret) {
+		ret = -EINVAL;
+		pr_info("%s: Timeout %d sec for netlink request (seq = %u)\n",
+			__func__, (IWPM_NL_TIMEOUT/HZ), nlmsg_request->nlmsg_seq);
+	} else {
+		ret = nlmsg_request->err_code;
+	}
+	kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request);
+	return ret;
+}
+
+int iwpm_get_nlmsg_seq(void)
+{
+	return atomic_inc_return(&iwpm_admin.nlmsg_seq);
+}
+
+int iwpm_valid_client(u8 nl_client)
+{
+	if (nl_client >= RDMA_NL_NUM_CLIENTS)
+		return 0;
+	return iwpm_admin.client_list[nl_client];
+}
+
+void iwpm_set_valid(u8 nl_client, int valid)
+{
+	if (nl_client >= RDMA_NL_NUM_CLIENTS)
+		return;
+	iwpm_admin.client_list[nl_client] = valid;
+}
+
+/* valid client */
+int iwpm_registered_client(u8 nl_client)
+{
+	return iwpm_admin.reg_list[nl_client];
+}
+
+/* valid client */
+void iwpm_set_registered(u8 nl_client, int reg)
+{
+	iwpm_admin.reg_list[nl_client] = reg;
+}
+
+int iwpm_compare_sockaddr(struct sockaddr_storage *a_sockaddr,
+				struct sockaddr_storage *b_sockaddr)
+{
+	if (a_sockaddr->ss_family != b_sockaddr->ss_family)
+		return 1;
+	if (a_sockaddr->ss_family == AF_INET) {
+		struct sockaddr_in *a4_sockaddr =
+			(struct sockaddr_in *)a_sockaddr;
+		struct sockaddr_in *b4_sockaddr =
+			(struct sockaddr_in *)b_sockaddr;
+		if (!memcmp(&a4_sockaddr->sin_addr,
+			&b4_sockaddr->sin_addr, sizeof(struct in_addr))
+			&& a4_sockaddr->sin_port == b4_sockaddr->sin_port)
+				return 0;
+
+	} else if (a_sockaddr->ss_family == AF_INET6) {
+		struct sockaddr_in6 *a6_sockaddr =
+			(struct sockaddr_in6 *)a_sockaddr;
+		struct sockaddr_in6 *b6_sockaddr =
+			(struct sockaddr_in6 *)b_sockaddr;
+		if (!memcmp(&a6_sockaddr->sin6_addr,
+			&b6_sockaddr->sin6_addr, sizeof(struct in6_addr))
+			&& a6_sockaddr->sin6_port == b6_sockaddr->sin6_port)
+				return 0;
+
+	} else {
+		pr_err("%s: Invalid sockaddr family\n", __func__);
+	}
+	return 1;
+}
+
+struct sk_buff *iwpm_create_nlmsg(u32 nl_op, struct nlmsghdr **nlh,
+						int nl_client)
+{
+	struct sk_buff *skb = NULL;
+
+	skb = dev_alloc_skb(NLMSG_GOODSIZE);
+	if (!skb) {
+		pr_err("%s Unable to allocate skb\n", __func__);
+		goto create_nlmsg_exit;
+	}
+	if (!(ibnl_put_msg(skb, nlh, 0, 0, nl_client, nl_op,
+			   NLM_F_REQUEST))) {
+		pr_warn("%s: Unable to put the nlmsg header\n", __func__);
+		dev_kfree_skb(skb);
+		skb = NULL;
+	}
+create_nlmsg_exit:
+	return skb;
+}
+
+int iwpm_parse_nlmsg(struct netlink_callback *cb, int policy_max,
+				   const struct nla_policy *nlmsg_policy,
+				   struct nlattr *nltb[], const char *msg_type)
+{
+	int nlh_len = 0;
+	int ret;
+	const char *err_str = "";
+
+	ret = nlmsg_validate(cb->nlh, nlh_len, policy_max-1, nlmsg_policy);
+	if (ret) {
+		err_str = "Invalid attribute";
+		goto parse_nlmsg_error;
+	}
+	ret = nlmsg_parse(cb->nlh, nlh_len, nltb, policy_max-1, nlmsg_policy);
+	if (ret) {
+		err_str = "Unable to parse the nlmsg";
+		goto parse_nlmsg_error;
+	}
+	ret = iwpm_validate_nlmsg_attr(nltb, policy_max);
+	if (ret) {
+		err_str = "Invalid NULL attribute";
+		goto parse_nlmsg_error;
+	}
+	return 0;
+parse_nlmsg_error:
+	pr_warn("%s: %s (msg type %s ret = %d)\n",
+			__func__, err_str, msg_type, ret);
+	return ret;
+}
+
+void iwpm_print_sockaddr(struct sockaddr_storage *sockaddr, char *msg)
+{
+	struct sockaddr_in6 *sockaddr_v6;
+	struct sockaddr_in *sockaddr_v4;
+
+	switch (sockaddr->ss_family) {
+	case AF_INET:
+		sockaddr_v4 = (struct sockaddr_in *)sockaddr;
+		pr_debug("%s IPV4 %pI4: %u(0x%04X)\n",
+			msg, &sockaddr_v4->sin_addr,
+			ntohs(sockaddr_v4->sin_port),
+			ntohs(sockaddr_v4->sin_port));
+		break;
+	case AF_INET6:
+		sockaddr_v6 = (struct sockaddr_in6 *)sockaddr;
+		pr_debug("%s IPV6 %pI6: %u(0x%04X)\n",
+			msg, &sockaddr_v6->sin6_addr,
+			ntohs(sockaddr_v6->sin6_port),
+			ntohs(sockaddr_v6->sin6_port));
+		break;
+	default:
+		break;
+	}
+}
+
+static u32 iwpm_ipv6_jhash(struct sockaddr_in6 *ipv6_sockaddr)
+{
+	u32 ipv6_hash = jhash(&ipv6_sockaddr->sin6_addr, sizeof(struct in6_addr), 0);
+	u32 hash = jhash_2words(ipv6_hash, (__force u32) ipv6_sockaddr->sin6_port, 0);
+	return hash;
+}
+
+static u32 iwpm_ipv4_jhash(struct sockaddr_in *ipv4_sockaddr)
+{
+	u32 ipv4_hash = jhash(&ipv4_sockaddr->sin_addr, sizeof(struct in_addr), 0);
+	u32 hash = jhash_2words(ipv4_hash, (__force u32) ipv4_sockaddr->sin_port, 0);
+	return hash;
+}
+
+static struct hlist_head *get_hash_bucket_head(struct sockaddr_storage
+					       *local_sockaddr,
+					       struct sockaddr_storage
+					       *mapped_sockaddr)
+{
+	u32 local_hash, mapped_hash, hash;
+
+	if (local_sockaddr->ss_family == AF_INET) {
+		local_hash = iwpm_ipv4_jhash((struct sockaddr_in *) local_sockaddr);
+		mapped_hash = iwpm_ipv4_jhash((struct sockaddr_in *) mapped_sockaddr);
+
+	} else if (local_sockaddr->ss_family == AF_INET6) {
+		local_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) local_sockaddr);
+		mapped_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) mapped_sockaddr);
+	} else {
+		pr_err("%s: Invalid sockaddr family\n", __func__);
+		return NULL;
+	}
+
+	if (local_hash == mapped_hash) /* if port mapper isn't available */
+		hash = local_hash;
+	else
+		hash = jhash_2words(local_hash, mapped_hash, 0);
+
+	return &iwpm_hash_bucket[hash & IWPM_HASH_BUCKET_MASK];
+}
+
+static int send_mapinfo_num(u32 mapping_num, u8 nl_client, int iwpm_pid)
+{
+	struct sk_buff *skb = NULL;
+	struct nlmsghdr *nlh;
+	u32 msg_seq;
+	const char *err_str = "";
+	int ret = -EINVAL;
+
+	skb = iwpm_create_nlmsg(RDMA_NL_IWPM_MAPINFO_NUM, &nlh, nl_client);
+	if (!skb) {
+		err_str = "Unable to create a nlmsg";
+		goto mapinfo_num_error;
+	}
+	nlh->nlmsg_seq = iwpm_get_nlmsg_seq();
+	msg_seq = 0;
+	err_str = "Unable to put attribute of mapinfo number nlmsg";
+	ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, IWPM_NLA_MAPINFO_SEQ);
+	if (ret)
+		goto mapinfo_num_error;
+	ret = ibnl_put_attr(skb, nlh, sizeof(u32),
+				&mapping_num, IWPM_NLA_MAPINFO_SEND_NUM);
+	if (ret)
+		goto mapinfo_num_error;
+	ret = ibnl_unicast(skb, nlh, iwpm_pid);
+	if (ret) {
+		skb = NULL;
+		err_str = "Unable to send a nlmsg";
+		goto mapinfo_num_error;
+	}
+	pr_debug("%s: Sent mapping number = %d\n", __func__, mapping_num);
+	return 0;
+mapinfo_num_error:
+	pr_info("%s: %s\n", __func__, err_str);
+	if (skb)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+static int send_nlmsg_done(struct sk_buff *skb, u8 nl_client, int iwpm_pid)
+{
+	struct nlmsghdr *nlh = NULL;
+	int ret = 0;
+
+	if (!skb)
+		return ret;
+	if (!(ibnl_put_msg(skb, &nlh, 0, 0, nl_client,
+			   RDMA_NL_IWPM_MAPINFO, NLM_F_MULTI))) {
+		pr_warn("%s Unable to put NLMSG_DONE\n", __func__);
+		return -ENOMEM;
+	}
+	nlh->nlmsg_type = NLMSG_DONE;
+	ret = ibnl_unicast(skb, (struct nlmsghdr *)skb->data, iwpm_pid);
+	if (ret)
+		pr_warn("%s Unable to send a nlmsg\n", __func__);
+	return ret;
+}
+
+int iwpm_send_mapinfo(u8 nl_client, int iwpm_pid)
+{
+	struct iwpm_mapping_info *map_info;
+	struct sk_buff *skb = NULL;
+	struct nlmsghdr *nlh;
+	int skb_num = 0, mapping_num = 0;
+	int i = 0, nlmsg_bytes = 0;
+	unsigned long flags;
+	const char *err_str = "";
+	int ret;
+
+	skb = dev_alloc_skb(NLMSG_GOODSIZE);
+	if (!skb) {
+		ret = -ENOMEM;
+		err_str = "Unable to allocate skb";
+		goto send_mapping_info_exit;
+	}
+	skb_num++;
+	spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
+	for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) {
+		hlist_for_each_entry(map_info, &iwpm_hash_bucket[i],
+				     hlist_node) {
+			if (map_info->nl_client != nl_client)
+				continue;
+			nlh = NULL;
+			if (!(ibnl_put_msg(skb, &nlh, 0, 0, nl_client,
+					RDMA_NL_IWPM_MAPINFO, NLM_F_MULTI))) {
+				ret = -ENOMEM;
+				err_str = "Unable to put the nlmsg header";
+				goto send_mapping_info_unlock;
+			}
+			err_str = "Unable to put attribute of the nlmsg";
+			ret = ibnl_put_attr(skb, nlh,
+					sizeof(struct sockaddr_storage),
+					&map_info->local_sockaddr,
+					IWPM_NLA_MAPINFO_LOCAL_ADDR);
+			if (ret)
+				goto send_mapping_info_unlock;
+
+			ret = ibnl_put_attr(skb, nlh,
+					sizeof(struct sockaddr_storage),
+					&map_info->mapped_sockaddr,
+					IWPM_NLA_MAPINFO_MAPPED_ADDR);
+			if (ret)
+				goto send_mapping_info_unlock;
+
+			iwpm_print_sockaddr(&map_info->local_sockaddr,
+				"send_mapping_info: Local sockaddr:");
+			iwpm_print_sockaddr(&map_info->mapped_sockaddr,
+				"send_mapping_info: Mapped local sockaddr:");
+			mapping_num++;
+			nlmsg_bytes += nlh->nlmsg_len;
+
+			/* check if all mappings can fit in one skb */
+			if (NLMSG_GOODSIZE - nlmsg_bytes < nlh->nlmsg_len * 2) {
+				/* and leave room for NLMSG_DONE */
+				nlmsg_bytes = 0;
+				skb_num++;
+				spin_unlock_irqrestore(&iwpm_mapinfo_lock,
+						       flags);
+				/* send the skb */
+				ret = send_nlmsg_done(skb, nl_client, iwpm_pid);
+				skb = NULL;
+				if (ret) {
+					err_str = "Unable to send map info";
+					goto send_mapping_info_exit;
+				}
+				if (skb_num == IWPM_MAPINFO_SKB_COUNT) {
+					ret = -ENOMEM;
+					err_str = "Insufficient skbs for map info";
+					goto send_mapping_info_exit;
+				}
+				skb = dev_alloc_skb(NLMSG_GOODSIZE);
+				if (!skb) {
+					ret = -ENOMEM;
+					err_str = "Unable to allocate skb";
+					goto send_mapping_info_exit;
+				}
+				spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
+			}
+		}
+	}
+send_mapping_info_unlock:
+	spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags);
+send_mapping_info_exit:
+	if (ret) {
+		pr_warn("%s: %s (ret = %d)\n", __func__, err_str, ret);
+		if (skb)
+			dev_kfree_skb(skb);
+		return ret;
+	}
+	send_nlmsg_done(skb, nl_client, iwpm_pid);
+	return send_mapinfo_num(mapping_num, nl_client, iwpm_pid);
+}
+
+int iwpm_mapinfo_available(void)
+{
+	unsigned long flags;
+	int full_bucket = 0, i = 0;
+
+	spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
+	if (iwpm_hash_bucket) {
+		for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) {
+			if (!hlist_empty(&iwpm_hash_bucket[i])) {
+				full_bucket = 1;
+				break;
+			}
+		}
+	}
+	spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags);
+	return full_bucket;
+}
diff --git a/drivers/infiniband/core/iwpm_util.h b/drivers/infiniband/core/iwpm_util.h
new file mode 100644
index 000000000000..9777c869a140
--- /dev/null
+++ b/drivers/infiniband/core/iwpm_util.h
@@ -0,0 +1,238 @@
+/*
+ * Copyright (c) 2014 Intel Corporation. All rights reserved.
+ * Copyright (c) 2014 Chelsio, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *	  copyright notice, this list of conditions and the following
+ *	  disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *	  copyright notice, this list of conditions and the following
+ *	  disclaimer in the documentation and/or other materials
+ *	  provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _IWPM_UTIL_H
+#define _IWPM_UTIL_H
+
+#include <linux/module.h>
+#include <linux/io.h>
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <linux/spinlock.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/delay.h>
+#include <linux/workqueue.h>
+#include <linux/mutex.h>
+#include <linux/jhash.h>
+#include <linux/kref.h>
+#include <net/netlink.h>
+#include <linux/errno.h>
+#include <rdma/iw_portmap.h>
+#include <rdma/rdma_netlink.h>
+
+
+#define IWPM_NL_RETRANS		3
+#define IWPM_NL_TIMEOUT		(10*HZ)
+#define IWPM_MAPINFO_SKB_COUNT	20
+
+#define IWPM_PID_UNDEFINED     -1
+#define IWPM_PID_UNAVAILABLE   -2
+
+struct iwpm_nlmsg_request {
+	struct list_head    inprocess_list;
+	__u32               nlmsg_seq;
+	void                *req_buffer;
+	u8	            nl_client;
+	u8                  request_done;
+	u16                 err_code;
+	wait_queue_head_t   waitq;
+	struct kref         kref;
+};
+
+struct iwpm_mapping_info {
+	struct hlist_node hlist_node;
+	struct sockaddr_storage local_sockaddr;
+	struct sockaddr_storage mapped_sockaddr;
+	u8     nl_client;
+};
+
+struct iwpm_admin_data {
+	atomic_t refcount;
+	atomic_t nlmsg_seq;
+	int      client_list[RDMA_NL_NUM_CLIENTS];
+	int      reg_list[RDMA_NL_NUM_CLIENTS];
+};
+
+/**
+ * iwpm_get_nlmsg_request - Allocate and initialize netlink message request
+ * @nlmsg_seq: Sequence number of the netlink message
+ * @nl_client: The index of the netlink client
+ * @gfp: Indicates how the memory for the request should be allocated
+ *
+ * Returns the newly allocated netlink request object if successful,
+ * otherwise returns NULL
+ */
+struct iwpm_nlmsg_request *iwpm_get_nlmsg_request(__u32 nlmsg_seq,
+						u8 nl_client, gfp_t gfp);
+
+/**
+ * iwpm_free_nlmsg_request - Deallocate netlink message request
+ * @kref: Holds reference of netlink message request
+ */
+void iwpm_free_nlmsg_request(struct kref *kref);
+
+/**
+ * iwpm_find_nlmsg_request - Find netlink message request in the request list
+ * @echo_seq: Sequence number of the netlink request to find
+ *
+ * Returns the found netlink message request,
+ * if not found, returns NULL
+ */
+struct iwpm_nlmsg_request *iwpm_find_nlmsg_request(__u32 echo_seq);
+
+/**
+ * iwpm_wait_complete_req - Block while servicing the netlink request
+ * @nlmsg_request: Netlink message request to service
+ *
+ * Wakes up, after the request is completed or expired
+ * Returns 0 if the request is complete without error
+ */
+int iwpm_wait_complete_req(struct iwpm_nlmsg_request *nlmsg_request);
+
+/**
+ * iwpm_get_nlmsg_seq - Get the sequence number for a netlink
+ *			message to send to the port mapper
+ *
+ * Returns the sequence number for the netlink message.
+ */
+int iwpm_get_nlmsg_seq(void);
+
+/**
+ * iwpm_valid_client - Check if the port mapper client is valid
+ * @nl_client: The index of the netlink client
+ *
+ * Valid clients need to call iwpm_init() before using
+ * the port mapper
+ */
+int iwpm_valid_client(u8 nl_client);
+
+/**
+ * iwpm_set_valid - Set the port mapper client to valid or not
+ * @nl_client: The index of the netlink client
+ * @valid: 1 if valid or 0 if invalid
+ */
+void iwpm_set_valid(u8 nl_client, int valid);
+
+/**
+ * iwpm_registered_client - Check if the port mapper client is registered
+ * @nl_client: The index of the netlink client
+ *
+ * Call iwpm_register_pid() to register a client
+ */
+int iwpm_registered_client(u8 nl_client);
+
+/**
+ * iwpm_set_registered - Set the port mapper client to registered or not
+ * @nl_client: The index of the netlink client
+ * @reg: 1 if registered or 0 if not
+ */
+void iwpm_set_registered(u8 nl_client, int reg);
+
+/**
+ * iwpm_send_mapinfo - Send local and mapped IPv4/IPv6 address info of
+ *                     a client to the user space port mapper
+ * @nl_client: The index of the netlink client
+ * @iwpm_pid: The pid of the user space port mapper
+ *
+ * If successful, returns the number of sent mapping info records
+ */
+int iwpm_send_mapinfo(u8 nl_client, int iwpm_pid);
+
+/**
+ * iwpm_mapinfo_available - Check if any mapping info records is available
+ *		            in the hash table
+ *
+ * Returns 1 if mapping information is available, otherwise returns 0
+ */
+int iwpm_mapinfo_available(void);
+
+/**
+ * iwpm_compare_sockaddr - Compare two sockaddr storage structs
+ *
+ * Returns 0 if they are holding the same ip/tcp address info,
+ * otherwise returns 1
+ */
+int iwpm_compare_sockaddr(struct sockaddr_storage *a_sockaddr,
+			struct sockaddr_storage *b_sockaddr);
+
+/**
+ * iwpm_validate_nlmsg_attr - Check for NULL netlink attributes
+ * @nltb: Holds address of each netlink message attributes
+ * @nla_count: Number of netlink message attributes
+ *
+ * Returns error if any of the nla_count attributes is NULL
+ */
+static inline int iwpm_validate_nlmsg_attr(struct nlattr *nltb[],
+					   int nla_count)
+{
+	int i;
+	for (i = 1; i < nla_count; i++) {
+		if (!nltb[i])
+			return -EINVAL;
+	}
+	return 0;
+}
+
+/**
+ * iwpm_create_nlmsg - Allocate skb and form a netlink message
+ * @nl_op: Netlink message opcode
+ * @nlh: Holds address of the netlink message header in skb
+ * @nl_client: The index of the netlink client
+ *
+ * Returns the newly allcated skb, or NULL if the tailroom of the skb
+ * is insufficient to store the message header and payload
+ */
+struct sk_buff *iwpm_create_nlmsg(u32 nl_op, struct nlmsghdr **nlh,
+					int nl_client);
+
+/**
+ * iwpm_parse_nlmsg - Validate and parse the received netlink message
+ * @cb: Netlink callback structure
+ * @policy_max: Maximum attribute type to be expected
+ * @nlmsg_policy: Validation policy
+ * @nltb: Array to store policy_max parsed elements
+ * @msg_type: Type of netlink message
+ *
+ * Returns 0 on success or a negative error code
+ */
+int iwpm_parse_nlmsg(struct netlink_callback *cb, int policy_max,
+				const struct nla_policy *nlmsg_policy,
+				struct nlattr *nltb[], const char *msg_type);
+
+/**
+ * iwpm_print_sockaddr - Print IPv4/IPv6 address and TCP port
+ * @sockaddr: Socket address to print
+ * @msg: Message to print
+ */
+void iwpm_print_sockaddr(struct sockaddr_storage *sockaddr, char *msg);
+#endif
diff --git a/drivers/infiniband/core/netlink.c b/drivers/infiniband/core/netlink.c
index a1e9cba84944..23dd5a5c7597 100644
--- a/drivers/infiniband/core/netlink.c
+++ b/drivers/infiniband/core/netlink.c
@@ -103,13 +103,13 @@ int ibnl_remove_client(int index)
 EXPORT_SYMBOL(ibnl_remove_client);
 
 void *ibnl_put_msg(struct sk_buff *skb, struct nlmsghdr **nlh, int seq,
-		   int len, int client, int op)
+		   int len, int client, int op, int flags)
 {
 	unsigned char *prev_tail;
 
 	prev_tail = skb_tail_pointer(skb);
 	*nlh = nlmsg_put(skb, 0, seq, RDMA_NL_GET_TYPE(client, op),
-			 len, NLM_F_MULTI);
+			 len, flags);
 	if (!*nlh)
 		goto out_nlmsg_trim;
 	(*nlh)->nlmsg_len = skb_tail_pointer(skb) - prev_tail;
@@ -172,6 +172,20 @@ static void ibnl_rcv(struct sk_buff *skb)
 	mutex_unlock(&ibnl_mutex);
 }
 
+int ibnl_unicast(struct sk_buff *skb, struct nlmsghdr *nlh,
+			__u32 pid)
+{
+	return nlmsg_unicast(nls, skb, pid);
+}
+EXPORT_SYMBOL(ibnl_unicast);
+
+int ibnl_multicast(struct sk_buff *skb, struct nlmsghdr *nlh,
+			unsigned int group, gfp_t flags)
+{
+	return nlmsg_multicast(nls, skb, 0, group, flags);
+}
+EXPORT_SYMBOL(ibnl_multicast);
+
 int __init ibnl_init(void)
 {
 	struct netlink_kernel_cfg cfg = {
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index f820958e4047..233eaf541f55 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -618,7 +618,7 @@ static void init_mad(struct ib_sa_mad *mad, struct ib_mad_agent *agent)
 
 static int send_mad(struct ib_sa_query *query, int timeout_ms, gfp_t gfp_mask)
 {
-	bool preload = gfp_mask & __GFP_WAIT;
+	bool preload = !!(gfp_mask & __GFP_WAIT);
 	unsigned long flags;
 	int ret, id;
 
diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index 7d3292c7b4b4..cbd0383f622e 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -429,15 +429,19 @@ static void ib_port_release(struct kobject *kobj)
 	struct attribute *a;
 	int i;
 
-	for (i = 0; (a = p->gid_group.attrs[i]); ++i)
-		kfree(a);
+	if (p->gid_group.attrs) {
+		for (i = 0; (a = p->gid_group.attrs[i]); ++i)
+			kfree(a);
 
-	kfree(p->gid_group.attrs);
+		kfree(p->gid_group.attrs);
+	}
 
-	for (i = 0; (a = p->pkey_group.attrs[i]); ++i)
-		kfree(a);
+	if (p->pkey_group.attrs) {
+		for (i = 0; (a = p->pkey_group.attrs[i]); ++i)
+			kfree(a);
 
-	kfree(p->pkey_group.attrs);
+		kfree(p->pkey_group.attrs);
+	}
 
 	kfree(p);
 }
@@ -534,10 +538,12 @@ static int add_port(struct ib_device *device, int port_num,
 	p->port_num   = port_num;
 
 	ret = kobject_init_and_add(&p->kobj, &port_type,
-				   kobject_get(device->ports_parent),
+				   device->ports_parent,
 				   "%d", port_num);
-	if (ret)
-		goto err_put;
+	if (ret) {
+		kfree(p);
+		return ret;
+	}
 
 	ret = sysfs_create_group(&p->kobj, &pma_group);
 	if (ret)
@@ -585,6 +591,7 @@ err_free_pkey:
 		kfree(p->pkey_group.attrs[i]);
 
 	kfree(p->pkey_group.attrs);
+	p->pkey_group.attrs = NULL;
 
 err_remove_gid:
 	sysfs_remove_group(&p->kobj, &p->gid_group);
@@ -594,13 +601,13 @@ err_free_gid:
 		kfree(p->gid_group.attrs[i]);
 
 	kfree(p->gid_group.attrs);
+	p->gid_group.attrs = NULL;
 
 err_remove_pma:
 	sysfs_remove_group(&p->kobj, &pma_group);
 
 err_put:
-	kobject_put(device->ports_parent);
-	kfree(p);
+	kobject_put(&p->kobj);
 	return ret;
 }
 
@@ -809,6 +816,22 @@ static struct attribute_group iw_stats_group = {
 	.attrs	= iw_proto_stats_attrs,
 };
 
+static void free_port_list_attributes(struct ib_device *device)
+{
+	struct kobject *p, *t;
+
+	list_for_each_entry_safe(p, t, &device->port_list, entry) {
+		struct ib_port *port = container_of(p, struct ib_port, kobj);
+		list_del(&p->entry);
+		sysfs_remove_group(p, &pma_group);
+		sysfs_remove_group(p, &port->pkey_group);
+		sysfs_remove_group(p, &port->gid_group);
+		kobject_put(p);
+	}
+
+	kobject_put(device->ports_parent);
+}
+
 int ib_device_register_sysfs(struct ib_device *device,
 			     int (*port_callback)(struct ib_device *,
 						  u8, struct kobject *))
@@ -835,7 +858,7 @@ int ib_device_register_sysfs(struct ib_device *device,
 	}
 
 	device->ports_parent = kobject_create_and_add("ports",
-					kobject_get(&class_dev->kobj));
+						      &class_dev->kobj);
 	if (!device->ports_parent) {
 		ret = -ENOMEM;
 		goto err_put;
@@ -862,21 +885,7 @@ int ib_device_register_sysfs(struct ib_device *device,
 	return 0;
 
 err_put:
-	{
-		struct kobject *p, *t;
-		struct ib_port *port;
-
-		list_for_each_entry_safe(p, t, &device->port_list, entry) {
-			list_del(&p->entry);
-			port = container_of(p, struct ib_port, kobj);
-			sysfs_remove_group(p, &pma_group);
-			sysfs_remove_group(p, &port->pkey_group);
-			sysfs_remove_group(p, &port->gid_group);
-			kobject_put(p);
-		}
-	}
-
-	kobject_put(&class_dev->kobj);
+	free_port_list_attributes(device);
 
 err_unregister:
 	device_unregister(class_dev);
@@ -887,22 +896,18 @@ err:
 
 void ib_device_unregister_sysfs(struct ib_device *device)
 {
-	struct kobject *p, *t;
-	struct ib_port *port;
-
 	/* Hold kobject until ib_dealloc_device() */
-	kobject_get(&device->dev.kobj);
+	struct kobject *kobj_dev = kobject_get(&device->dev.kobj);
+	int i;
 
-	list_for_each_entry_safe(p, t, &device->port_list, entry) {
-		list_del(&p->entry);
-		port = container_of(p, struct ib_port, kobj);
-		sysfs_remove_group(p, &pma_group);
-		sysfs_remove_group(p, &port->pkey_group);
-		sysfs_remove_group(p, &port->gid_group);
-		kobject_put(p);
-	}
+	if (device->node_type == RDMA_NODE_RNIC && device->get_protocol_stats)
+		sysfs_remove_group(kobj_dev, &iw_stats_group);
+
+	free_port_list_attributes(device);
+
+	for (i = 0; i < ARRAY_SIZE(ib_class_attributes); ++i)
+		device_remove_file(&device->dev, ib_class_attributes[i]);
 
-	kobject_put(device->ports_parent);
 	device_unregister(&device->dev);
 }
 
diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c
index f0d588f8859e..1acb99100556 100644
--- a/drivers/infiniband/core/user_mad.c
+++ b/drivers/infiniband/core/user_mad.c
@@ -98,7 +98,7 @@ struct ib_umad_port {
 
 struct ib_umad_device {
 	int                  start_port, end_port;
-	struct kref          ref;
+	struct kobject       kobj;
 	struct ib_umad_port  port[0];
 };
 
@@ -134,14 +134,18 @@ static DECLARE_BITMAP(dev_map, IB_UMAD_MAX_PORTS);
 static void ib_umad_add_one(struct ib_device *device);
 static void ib_umad_remove_one(struct ib_device *device);
 
-static void ib_umad_release_dev(struct kref *ref)
+static void ib_umad_release_dev(struct kobject *kobj)
 {
 	struct ib_umad_device *dev =
-		container_of(ref, struct ib_umad_device, ref);
+		container_of(kobj, struct ib_umad_device, kobj);
 
 	kfree(dev);
 }
 
+static struct kobj_type ib_umad_dev_ktype = {
+	.release = ib_umad_release_dev,
+};
+
 static int hdr_size(struct ib_umad_file *file)
 {
 	return file->use_pkey_index ? sizeof (struct ib_user_mad_hdr) :
@@ -780,27 +784,19 @@ static int ib_umad_open(struct inode *inode, struct file *filp)
 {
 	struct ib_umad_port *port;
 	struct ib_umad_file *file;
-	int ret;
+	int ret = -ENXIO;
 
 	port = container_of(inode->i_cdev, struct ib_umad_port, cdev);
-	if (port)
-		kref_get(&port->umad_dev->ref);
-	else
-		return -ENXIO;
 
 	mutex_lock(&port->file_mutex);
 
-	if (!port->ib_dev) {
-		ret = -ENXIO;
+	if (!port->ib_dev)
 		goto out;
-	}
 
+	ret = -ENOMEM;
 	file = kzalloc(sizeof *file, GFP_KERNEL);
-	if (!file) {
-		kref_put(&port->umad_dev->ref, ib_umad_release_dev);
-		ret = -ENOMEM;
+	if (!file)
 		goto out;
-	}
 
 	mutex_init(&file->mutex);
 	spin_lock_init(&file->send_lock);
@@ -814,6 +810,13 @@ static int ib_umad_open(struct inode *inode, struct file *filp)
 	list_add_tail(&file->port_list, &port->file_list);
 
 	ret = nonseekable_open(inode, filp);
+	if (ret) {
+		list_del(&file->port_list);
+		kfree(file);
+		goto out;
+	}
+
+	kobject_get(&port->umad_dev->kobj);
 
 out:
 	mutex_unlock(&port->file_mutex);
@@ -852,7 +855,7 @@ static int ib_umad_close(struct inode *inode, struct file *filp)
 	mutex_unlock(&file->port->file_mutex);
 
 	kfree(file);
-	kref_put(&dev->ref, ib_umad_release_dev);
+	kobject_put(&dev->kobj);
 
 	return 0;
 }
@@ -880,10 +883,6 @@ static int ib_umad_sm_open(struct inode *inode, struct file *filp)
 	int ret;
 
 	port = container_of(inode->i_cdev, struct ib_umad_port, sm_cdev);
-	if (port)
-		kref_get(&port->umad_dev->ref);
-	else
-		return -ENXIO;
 
 	if (filp->f_flags & O_NONBLOCK) {
 		if (down_trylock(&port->sm_sem)) {
@@ -898,17 +897,27 @@ static int ib_umad_sm_open(struct inode *inode, struct file *filp)
 	}
 
 	ret = ib_modify_port(port->ib_dev, port->port_num, 0, &props);
-	if (ret) {
-		up(&port->sm_sem);
-		goto fail;
-	}
+	if (ret)
+		goto err_up_sem;
 
 	filp->private_data = port;
 
-	return nonseekable_open(inode, filp);
+	ret = nonseekable_open(inode, filp);
+	if (ret)
+		goto err_clr_sm_cap;
+
+	kobject_get(&port->umad_dev->kobj);
+
+	return 0;
+
+err_clr_sm_cap:
+	swap(props.set_port_cap_mask, props.clr_port_cap_mask);
+	ib_modify_port(port->ib_dev, port->port_num, 0, &props);
+
+err_up_sem:
+	up(&port->sm_sem);
 
 fail:
-	kref_put(&port->umad_dev->ref, ib_umad_release_dev);
 	return ret;
 }
 
@@ -927,7 +936,7 @@ static int ib_umad_sm_close(struct inode *inode, struct file *filp)
 
 	up(&port->sm_sem);
 
-	kref_put(&port->umad_dev->ref, ib_umad_release_dev);
+	kobject_put(&port->umad_dev->kobj);
 
 	return ret;
 }
@@ -995,6 +1004,7 @@ static int find_overflow_devnum(void)
 }
 
 static int ib_umad_init_port(struct ib_device *device, int port_num,
+			     struct ib_umad_device *umad_dev,
 			     struct ib_umad_port *port)
 {
 	int devnum;
@@ -1027,6 +1037,7 @@ static int ib_umad_init_port(struct ib_device *device, int port_num,
 
 	cdev_init(&port->cdev, &umad_fops);
 	port->cdev.owner = THIS_MODULE;
+	port->cdev.kobj.parent = &umad_dev->kobj;
 	kobject_set_name(&port->cdev.kobj, "umad%d", port->dev_num);
 	if (cdev_add(&port->cdev, base, 1))
 		goto err_cdev;
@@ -1045,6 +1056,7 @@ static int ib_umad_init_port(struct ib_device *device, int port_num,
 	base += IB_UMAD_MAX_PORTS;
 	cdev_init(&port->sm_cdev, &umad_sm_fops);
 	port->sm_cdev.owner = THIS_MODULE;
+	port->sm_cdev.kobj.parent = &umad_dev->kobj;
 	kobject_set_name(&port->sm_cdev.kobj, "issm%d", port->dev_num);
 	if (cdev_add(&port->sm_cdev, base, 1))
 		goto err_sm_cdev;
@@ -1138,7 +1150,7 @@ static void ib_umad_add_one(struct ib_device *device)
 	if (!umad_dev)
 		return;
 
-	kref_init(&umad_dev->ref);
+	kobject_init(&umad_dev->kobj, &ib_umad_dev_ktype);
 
 	umad_dev->start_port = s;
 	umad_dev->end_port   = e;
@@ -1146,7 +1158,8 @@ static void ib_umad_add_one(struct ib_device *device)
 	for (i = s; i <= e; ++i) {
 		umad_dev->port[i - s].umad_dev = umad_dev;
 
-		if (ib_umad_init_port(device, i, &umad_dev->port[i - s]))
+		if (ib_umad_init_port(device, i, umad_dev,
+				      &umad_dev->port[i - s]))
 			goto err;
 	}
 
@@ -1158,7 +1171,7 @@ err:
 	while (--i >= s)
 		ib_umad_kill_port(&umad_dev->port[i - s]);
 
-	kref_put(&umad_dev->ref, ib_umad_release_dev);
+	kobject_put(&umad_dev->kobj);
 }
 
 static void ib_umad_remove_one(struct ib_device *device)
@@ -1172,7 +1185,7 @@ static void ib_umad_remove_one(struct ib_device *device)
 	for (i = 0; i <= umad_dev->end_port - umad_dev->start_port; ++i)
 		ib_umad_kill_port(&umad_dev->port[i]);
 
-	kref_put(&umad_dev->ref, ib_umad_release_dev);
+	kobject_put(&umad_dev->kobj);
 }
 
 static char *umad_devnode(struct device *dev, umode_t *mode)
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 92525f855d82..c2b89cc5dbca 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -48,7 +48,7 @@
 
 #include "core_priv.h"
 
-int ib_rate_to_mult(enum ib_rate rate)
+__attribute_const__ int ib_rate_to_mult(enum ib_rate rate)
 {
 	switch (rate) {
 	case IB_RATE_2_5_GBPS: return  1;
@@ -65,7 +65,7 @@ int ib_rate_to_mult(enum ib_rate rate)
 }
 EXPORT_SYMBOL(ib_rate_to_mult);
 
-enum ib_rate mult_to_ib_rate(int mult)
+__attribute_const__ enum ib_rate mult_to_ib_rate(int mult)
 {
 	switch (mult) {
 	case 1:  return IB_RATE_2_5_GBPS;
@@ -82,7 +82,7 @@ enum ib_rate mult_to_ib_rate(int mult)
 }
 EXPORT_SYMBOL(mult_to_ib_rate);
 
-int ib_rate_to_mbps(enum ib_rate rate)
+__attribute_const__ int ib_rate_to_mbps(enum ib_rate rate)
 {
 	switch (rate) {
 	case IB_RATE_2_5_GBPS: return 2500;
@@ -107,7 +107,7 @@ int ib_rate_to_mbps(enum ib_rate rate)
 }
 EXPORT_SYMBOL(ib_rate_to_mbps);
 
-enum rdma_transport_type
+__attribute_const__ enum rdma_transport_type
 rdma_node_get_transport(enum rdma_node_type node_type)
 {
 	switch (node_type) {
diff --git a/drivers/infiniband/hw/Makefile b/drivers/infiniband/hw/Makefile
new file mode 100644
index 000000000000..e900b03531a9
--- /dev/null
+++ b/drivers/infiniband/hw/Makefile
@@ -0,0 +1,12 @@
+obj-$(CONFIG_INFINIBAND_MTHCA)		+= mthca/
+obj-$(CONFIG_INFINIBAND_IPATH)		+= ipath/
+obj-$(CONFIG_INFINIBAND_QIB)		+= qib/
+obj-$(CONFIG_INFINIBAND_EHCA)		+= ehca/
+obj-$(CONFIG_INFINIBAND_AMSO1100)	+= amso1100/
+obj-$(CONFIG_INFINIBAND_CXGB3)		+= cxgb3/
+obj-$(CONFIG_INFINIBAND_CXGB4)		+= cxgb4/
+obj-$(CONFIG_MLX4_INFINIBAND)		+= mlx4/
+obj-$(CONFIG_MLX5_INFINIBAND)		+= mlx5/
+obj-$(CONFIG_INFINIBAND_NES)		+= nes/
+obj-$(CONFIG_INFINIBAND_OCRDMA)		+= ocrdma/
+obj-$(CONFIG_INFINIBAND_USNIC)		+= usnic/
diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c
index c3f5aca4ef00..de1c61b417d6 100644
--- a/drivers/infiniband/hw/cxgb3/cxio_hal.c
+++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c
@@ -735,14 +735,12 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry,
 			((perm & TPT_MW_BIND) ? F_TPT_MW_BIND_ENABLE : 0) |
 			V_TPT_ADDR_TYPE((zbva ? TPT_ZBTO : TPT_VATO)) |
 			V_TPT_PAGE_SIZE(page_size));
-		tpt.rsvd_pbl_addr = reset_tpt_entry ? 0 :
-				    cpu_to_be32(V_TPT_PBL_ADDR(PBL_OFF(rdev_p, pbl_addr)>>3));
+		tpt.rsvd_pbl_addr = cpu_to_be32(V_TPT_PBL_ADDR(PBL_OFF(rdev_p, pbl_addr)>>3));
 		tpt.len = cpu_to_be32(len);
 		tpt.va_hi = cpu_to_be32((u32) (to >> 32));
 		tpt.va_low_or_fbo = cpu_to_be32((u32) (to & 0xFFFFFFFFULL));
 		tpt.rsvd_bind_cnt_or_pstag = 0;
-		tpt.rsvd_pbl_size = reset_tpt_entry ? 0 :
-				  cpu_to_be32(V_TPT_PBL_SIZE(pbl_size >> 2));
+		tpt.rsvd_pbl_size = cpu_to_be32(V_TPT_PBL_SIZE(pbl_size >> 2));
 	}
 	err = cxio_hal_ctrl_qp_write_mem(rdev_p,
 				       stag_idx +
diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
index 095bb046e2c8..cb78b1e9bcd9 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
@@ -418,6 +418,7 @@ static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp)
 	skb->priority = CPL_PRIORITY_DATA;
 	set_arp_failure_handler(skb, abort_arp_failure);
 	req = (struct cpl_abort_req *) skb_put(skb, sizeof(*req));
+	memset(req, 0, sizeof(*req));
 	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_HOST_ABORT_CON_REQ));
 	req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid));
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ABORT_REQ, ep->hwtid));
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 1f863a96a480..5e153f6d4b48 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2009-2010 Chelsio, Inc. All rights reserved.
+ * Copyright (c) 2009-2014 Chelsio, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -47,6 +47,8 @@
 #include <net/ip6_route.h>
 #include <net/addrconf.h>
 
+#include <rdma/ib_addr.h>
+
 #include "iw_cxgb4.h"
 
 static char *states[] = {
@@ -232,12 +234,16 @@ static void release_tid(struct c4iw_rdev *rdev, u32 hwtid, struct sk_buff *skb)
 
 static void set_emss(struct c4iw_ep *ep, u16 opt)
 {
-	ep->emss = ep->com.dev->rdev.lldi.mtus[GET_TCPOPT_MSS(opt)] - 40;
+	ep->emss = ep->com.dev->rdev.lldi.mtus[GET_TCPOPT_MSS(opt)] -
+		   sizeof(struct iphdr) - sizeof(struct tcphdr);
 	ep->mss = ep->emss;
 	if (GET_TCPOPT_TSTAMP(opt))
 		ep->emss -= 12;
 	if (ep->emss < 128)
 		ep->emss = 128;
+	if (ep->emss & 7)
+		PDBG("Warning: misaligned mtu idx %u mss %u emss=%u\n",
+		     GET_TCPOPT_MSS(opt), ep->mss, ep->emss);
 	PDBG("%s mss_idx %u mss %u emss=%u\n", __func__, GET_TCPOPT_MSS(opt),
 	     ep->mss, ep->emss);
 }
@@ -294,6 +300,12 @@ void _c4iw_free_ep(struct kref *kref)
 		dst_release(ep->dst);
 		cxgb4_l2t_release(ep->l2t);
 	}
+	if (test_bit(RELEASE_MAPINFO, &ep->com.flags)) {
+		print_addr(&ep->com, __func__, "remove_mapinfo/mapping");
+		iwpm_remove_mapinfo(&ep->com.local_addr,
+				    &ep->com.mapped_local_addr);
+		iwpm_remove_mapping(&ep->com.local_addr, RDMA_NL_C4IW);
+	}
 	kfree(ep);
 }
 
@@ -341,10 +353,7 @@ static struct sk_buff *get_skb(struct sk_buff *skb, int len, gfp_t gfp)
 
 static struct net_device *get_real_dev(struct net_device *egress_dev)
 {
-	struct net_device *phys_dev = egress_dev;
-	if (egress_dev->priv_flags & IFF_802_1Q_VLAN)
-		phys_dev = vlan_dev_real_dev(egress_dev);
-	return phys_dev;
+	return rdma_vlan_dev_real_dev(egress_dev) ? : egress_dev;
 }
 
 static int our_interface(struct c4iw_dev *dev, struct net_device *egress_dev)
@@ -468,7 +477,7 @@ static void send_flowc(struct c4iw_ep *ep, struct sk_buff *skb)
 	flowc->mnemval[5].mnemonic = FW_FLOWC_MNEM_RCVNXT;
 	flowc->mnemval[5].val = cpu_to_be32(ep->rcv_seq);
 	flowc->mnemval[6].mnemonic = FW_FLOWC_MNEM_SNDBUF;
-	flowc->mnemval[6].val = cpu_to_be32(snd_win);
+	flowc->mnemval[6].val = cpu_to_be32(ep->snd_win);
 	flowc->mnemval[7].mnemonic = FW_FLOWC_MNEM_MSS;
 	flowc->mnemval[7].val = cpu_to_be32(ep->emss);
 	/* Pad WR to 16 byte boundary */
@@ -528,6 +537,49 @@ static int send_abort(struct c4iw_ep *ep, struct sk_buff *skb, gfp_t gfp)
 	return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
 }
 
+/*
+ * c4iw_form_pm_msg - Form a port mapper message with mapping info
+ */
+static void c4iw_form_pm_msg(struct c4iw_ep *ep,
+				struct iwpm_sa_data *pm_msg)
+{
+	memcpy(&pm_msg->loc_addr, &ep->com.local_addr,
+		sizeof(ep->com.local_addr));
+	memcpy(&pm_msg->rem_addr, &ep->com.remote_addr,
+		sizeof(ep->com.remote_addr));
+}
+
+/*
+ * c4iw_form_reg_msg - Form a port mapper message with dev info
+ */
+static void c4iw_form_reg_msg(struct c4iw_dev *dev,
+				struct iwpm_dev_data *pm_msg)
+{
+	memcpy(pm_msg->dev_name, dev->ibdev.name, IWPM_DEVNAME_SIZE);
+	memcpy(pm_msg->if_name, dev->rdev.lldi.ports[0]->name,
+				IWPM_IFNAME_SIZE);
+}
+
+static void c4iw_record_pm_msg(struct c4iw_ep *ep,
+			struct iwpm_sa_data *pm_msg)
+{
+	memcpy(&ep->com.mapped_local_addr, &pm_msg->mapped_loc_addr,
+		sizeof(ep->com.mapped_local_addr));
+	memcpy(&ep->com.mapped_remote_addr, &pm_msg->mapped_rem_addr,
+		sizeof(ep->com.mapped_remote_addr));
+}
+
+static void best_mtu(const unsigned short *mtus, unsigned short mtu,
+		     unsigned int *idx, int use_ts)
+{
+	unsigned short hdr_size = sizeof(struct iphdr) +
+				  sizeof(struct tcphdr) +
+				  (use_ts ? 12 : 0);
+	unsigned short data_size = mtu - hdr_size;
+
+	cxgb4_best_aligned_mtu(mtus, hdr_size, data_size, 8, idx);
+}
+
 static int send_connect(struct c4iw_ep *ep)
 {
 	struct cpl_act_open_req *req;
@@ -546,10 +598,15 @@ static int send_connect(struct c4iw_ep *ep)
 	int sizev6 = is_t4(ep->com.dev->rdev.lldi.adapter_type) ?
 				sizeof(struct cpl_act_open_req6) :
 				sizeof(struct cpl_t5_act_open_req6);
-	struct sockaddr_in *la = (struct sockaddr_in *)&ep->com.local_addr;
-	struct sockaddr_in *ra = (struct sockaddr_in *)&ep->com.remote_addr;
-	struct sockaddr_in6 *la6 = (struct sockaddr_in6 *)&ep->com.local_addr;
-	struct sockaddr_in6 *ra6 = (struct sockaddr_in6 *)&ep->com.remote_addr;
+	struct sockaddr_in *la = (struct sockaddr_in *)
+				 &ep->com.mapped_local_addr;
+	struct sockaddr_in *ra = (struct sockaddr_in *)
+				 &ep->com.mapped_remote_addr;
+	struct sockaddr_in6 *la6 = (struct sockaddr_in6 *)
+				   &ep->com.mapped_local_addr;
+	struct sockaddr_in6 *ra6 = (struct sockaddr_in6 *)
+				   &ep->com.mapped_remote_addr;
+	int win;
 
 	wrlen = (ep->com.remote_addr.ss_family == AF_INET) ?
 			roundup(sizev4, 16) :
@@ -565,8 +622,18 @@ static int send_connect(struct c4iw_ep *ep)
 	}
 	set_wr_txq(skb, CPL_PRIORITY_SETUP, ep->ctrlq_idx);
 
-	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
+	best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx,
+		 enable_tcp_timestamps);
 	wscale = compute_wscale(rcv_win);
+
+	/*
+	 * Specify the largest window that will fit in opt0. The
+	 * remainder will be specified in the rx_data_ack.
+	 */
+	win = ep->rcv_win >> 10;
+	if (win > RCV_BUFSIZ_MASK)
+		win = RCV_BUFSIZ_MASK;
+
 	opt0 = (nocong ? NO_CONG(1) : 0) |
 	       KEEP_ALIVE(1) |
 	       DELACK(1) |
@@ -577,7 +644,7 @@ static int send_connect(struct c4iw_ep *ep)
 	       SMAC_SEL(ep->smac_idx) |
 	       DSCP(ep->tos) |
 	       ULP_MODE(ULP_MODE_TCPDDP) |
-	       RCV_BUFSIZ(rcv_win>>10);
+	       RCV_BUFSIZ(win);
 	opt2 = RX_CHANNEL(0) |
 	       CCTRL_ECN(enable_ecn) |
 	       RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
@@ -633,6 +700,13 @@ static int send_connect(struct c4iw_ep *ep)
 			req6->opt2 = cpu_to_be32(opt2);
 		}
 	} else {
+		u32 isn = (prandom_u32() & ~7UL) - 1;
+
+		opt2 |= T5_OPT_2_VALID;
+		opt2 |= CONG_CNTRL_VALID; /* OPT_2_ISS for T5 */
+		if (peer2peer)
+			isn += 4;
+
 		if (ep->com.remote_addr.ss_family == AF_INET) {
 			t5_req = (struct cpl_t5_act_open_req *)
 				 skb_put(skb, wrlen);
@@ -649,6 +723,9 @@ static int send_connect(struct c4iw_ep *ep)
 						     cxgb4_select_ntuple(
 					     ep->com.dev->rdev.lldi.ports[0],
 					     ep->l2t)));
+			t5_req->rsvd = cpu_to_be32(isn);
+			PDBG("%s snd_isn %u\n", __func__,
+			     be32_to_cpu(t5_req->rsvd));
 			t5_req->opt2 = cpu_to_be32(opt2);
 		} else {
 			t5_req6 = (struct cpl_t5_act_open_req6 *)
@@ -672,6 +749,9 @@ static int send_connect(struct c4iw_ep *ep)
 							cxgb4_select_ntuple(
 						ep->com.dev->rdev.lldi.ports[0],
 						ep->l2t));
+			t5_req6->rsvd = cpu_to_be32(isn);
+			PDBG("%s snd_isn %u\n", __func__,
+			     be32_to_cpu(t5_req6->rsvd));
 			t5_req6->opt2 = cpu_to_be32(opt2);
 		}
 	}
@@ -1145,6 +1225,14 @@ static int update_rx_credits(struct c4iw_ep *ep, u32 credits)
 		return 0;
 	}
 
+	/*
+	 * If we couldn't specify the entire rcv window at connection setup
+	 * due to the limit in the number of bits in the RCV_BUFSIZ field,
+	 * then add the overage in to the credits returned.
+	 */
+	if (ep->rcv_win > RCV_BUFSIZ_MASK * 1024)
+		credits += ep->rcv_win - RCV_BUFSIZ_MASK * 1024;
+
 	req = (struct cpl_rx_data_ack *) skb_put(skb, wrlen);
 	memset(req, 0, wrlen);
 	INIT_TP_WR(req, ep->hwtid);
@@ -1618,6 +1706,7 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 	unsigned int mtu_idx;
 	int wscale;
 	struct sockaddr_in *sin;
+	int win;
 
 	skb = get_skb(NULL, sizeof(*req), GFP_KERNEL);
 	req = (struct fw_ofld_connection_wr *)__skb_put(skb, sizeof(*req));
@@ -1627,10 +1716,10 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 	req->le.filter = cpu_to_be32(cxgb4_select_ntuple(
 				     ep->com.dev->rdev.lldi.ports[0],
 				     ep->l2t));
-	sin = (struct sockaddr_in *)&ep->com.local_addr;
+	sin = (struct sockaddr_in *)&ep->com.mapped_local_addr;
 	req->le.lport = sin->sin_port;
 	req->le.u.ipv4.lip = sin->sin_addr.s_addr;
-	sin = (struct sockaddr_in *)&ep->com.remote_addr;
+	sin = (struct sockaddr_in *)&ep->com.mapped_remote_addr;
 	req->le.pport = sin->sin_port;
 	req->le.u.ipv4.pip = sin->sin_addr.s_addr;
 	req->tcb.t_state_to_astid =
@@ -1640,8 +1729,18 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 			htons(F_FW_OFLD_CONNECTION_WR_CPLRXDATAACK);
 	req->tcb.tx_max = (__force __be32) jiffies;
 	req->tcb.rcv_adv = htons(1);
-	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
+	best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx,
+		 enable_tcp_timestamps);
 	wscale = compute_wscale(rcv_win);
+
+	/*
+	 * Specify the largest window that will fit in opt0. The
+	 * remainder will be specified in the rx_data_ack.
+	 */
+	win = ep->rcv_win >> 10;
+	if (win > RCV_BUFSIZ_MASK)
+		win = RCV_BUFSIZ_MASK;
+
 	req->tcb.opt0 = (__force __be64) (TCAM_BYPASS(1) |
 		(nocong ? NO_CONG(1) : 0) |
 		KEEP_ALIVE(1) |
@@ -1653,7 +1752,7 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 		SMAC_SEL(ep->smac_idx) |
 		DSCP(ep->tos) |
 		ULP_MODE(ULP_MODE_TCPDDP) |
-		RCV_BUFSIZ(rcv_win >> 10));
+		RCV_BUFSIZ(win));
 	req->tcb.opt2 = (__force __be32) (PACE(1) |
 		TX_QUEUE(ep->com.dev->rdev.lldi.tx_modq[ep->tx_chan]) |
 		RX_CHANNEL(0) |
@@ -1690,6 +1789,13 @@ static int is_neg_adv(unsigned int status)
 	       status == CPL_ERR_KEEPALV_NEG_ADVICE;
 }
 
+static void set_tcp_window(struct c4iw_ep *ep, struct port_info *pi)
+{
+	ep->snd_win = snd_win;
+	ep->rcv_win = rcv_win;
+	PDBG("%s snd_win %d rcv_win %d\n", __func__, ep->snd_win, ep->rcv_win);
+}
+
 #define ACT_OPEN_RETRY_COUNT 2
 
 static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip,
@@ -1738,6 +1844,7 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip,
 		ep->ctrlq_idx = cxgb4_port_idx(pdev);
 		ep->rss_qid = cdev->rdev.lldi.rxq_ids[
 			cxgb4_port_idx(pdev) * step];
+		set_tcp_window(ep, (struct port_info *)netdev_priv(pdev));
 		dev_put(pdev);
 	} else {
 		pdev = get_real_dev(n->dev);
@@ -1746,16 +1853,17 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip,
 		if (!ep->l2t)
 			goto out;
 		ep->mtu = dst_mtu(dst);
-		ep->tx_chan = cxgb4_port_chan(n->dev);
-		ep->smac_idx = (cxgb4_port_viid(n->dev) & 0x7F) << 1;
+		ep->tx_chan = cxgb4_port_chan(pdev);
+		ep->smac_idx = (cxgb4_port_viid(pdev) & 0x7F) << 1;
 		step = cdev->rdev.lldi.ntxq /
 			cdev->rdev.lldi.nchan;
-		ep->txq_idx = cxgb4_port_idx(n->dev) * step;
-		ep->ctrlq_idx = cxgb4_port_idx(n->dev);
+		ep->txq_idx = cxgb4_port_idx(pdev) * step;
+		ep->ctrlq_idx = cxgb4_port_idx(pdev);
 		step = cdev->rdev.lldi.nrxq /
 			cdev->rdev.lldi.nchan;
 		ep->rss_qid = cdev->rdev.lldi.rxq_ids[
-			cxgb4_port_idx(n->dev) * step];
+			cxgb4_port_idx(pdev) * step];
+		set_tcp_window(ep, (struct port_info *)netdev_priv(pdev));
 
 		if (clear_mpa_v1) {
 			ep->retry_with_mpa_v1 = 0;
@@ -1870,10 +1978,10 @@ static int act_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
 	struct sockaddr_in6 *ra6;
 
 	ep = lookup_atid(t, atid);
-	la = (struct sockaddr_in *)&ep->com.local_addr;
-	ra = (struct sockaddr_in *)&ep->com.remote_addr;
-	la6 = (struct sockaddr_in6 *)&ep->com.local_addr;
-	ra6 = (struct sockaddr_in6 *)&ep->com.remote_addr;
+	la = (struct sockaddr_in *)&ep->com.mapped_local_addr;
+	ra = (struct sockaddr_in *)&ep->com.mapped_remote_addr;
+	la6 = (struct sockaddr_in6 *)&ep->com.mapped_local_addr;
+	ra6 = (struct sockaddr_in6 *)&ep->com.mapped_remote_addr;
 
 	PDBG("%s ep %p atid %u status %u errno %d\n", __func__, ep, atid,
 	     status, status2errno(status));
@@ -1986,13 +2094,36 @@ static void accept_cr(struct c4iw_ep *ep, struct sk_buff *skb,
 	u64 opt0;
 	u32 opt2;
 	int wscale;
+	struct cpl_t5_pass_accept_rpl *rpl5 = NULL;
+	int win;
 
 	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
 	BUG_ON(skb_cloned(skb));
-	skb_trim(skb, sizeof(*rpl));
+
 	skb_get(skb);
-	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
+	rpl = cplhdr(skb);
+	if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) {
+		skb_trim(skb, roundup(sizeof(*rpl5), 16));
+		rpl5 = (void *)rpl;
+		INIT_TP_WR(rpl5, ep->hwtid);
+	} else {
+		skb_trim(skb, sizeof(*rpl));
+		INIT_TP_WR(rpl, ep->hwtid);
+	}
+	OPCODE_TID(rpl) = cpu_to_be32(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL,
+						    ep->hwtid));
+
+	best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx,
+		 enable_tcp_timestamps && req->tcpopt.tstamp);
 	wscale = compute_wscale(rcv_win);
+
+	/*
+	 * Specify the largest window that will fit in opt0. The
+	 * remainder will be specified in the rx_data_ack.
+	 */
+	win = ep->rcv_win >> 10;
+	if (win > RCV_BUFSIZ_MASK)
+		win = RCV_BUFSIZ_MASK;
 	opt0 = (nocong ? NO_CONG(1) : 0) |
 	       KEEP_ALIVE(1) |
 	       DELACK(1) |
@@ -2003,7 +2134,7 @@ static void accept_cr(struct c4iw_ep *ep, struct sk_buff *skb,
 	       SMAC_SEL(ep->smac_idx) |
 	       DSCP(ep->tos >> 2) |
 	       ULP_MODE(ULP_MODE_TCPDDP) |
-	       RCV_BUFSIZ(rcv_win>>10);
+	       RCV_BUFSIZ(win);
 	opt2 = RX_CHANNEL(0) |
 	       RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
 
@@ -2023,14 +2154,18 @@ static void accept_cr(struct c4iw_ep *ep, struct sk_buff *skb,
 			opt2 |= CCTRL_ECN(1);
 	}
 	if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) {
+		u32 isn = (prandom_u32() & ~7UL) - 1;
 		opt2 |= T5_OPT_2_VALID;
 		opt2 |= V_CONG_CNTRL(CONG_ALG_TAHOE);
+		opt2 |= CONG_CNTRL_VALID; /* OPT_2_ISS for T5 */
+		rpl5 = (void *)rpl;
+		memset(&rpl5->iss, 0, roundup(sizeof(*rpl5)-sizeof(*rpl), 16));
+		if (peer2peer)
+			isn += 4;
+		rpl5->iss = cpu_to_be32(isn);
+		PDBG("%s iss %u\n", __func__, be32_to_cpu(rpl5->iss));
 	}
 
-	rpl = cplhdr(skb);
-	INIT_TP_WR(rpl, ep->hwtid);
-	OPCODE_TID(rpl) = cpu_to_be32(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL,
-				      ep->hwtid));
 	rpl->opt0 = cpu_to_be64(opt0);
 	rpl->opt2 = cpu_to_be32(opt2);
 	set_wr_txq(skb, CPL_PRIORITY_SETUP, ep->ctrlq_idx);
@@ -2095,6 +2230,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 	int err;
 	u16 peer_mss = ntohs(req->tcpopt.mss);
 	int iptype;
+	unsigned short hdrs;
 
 	parent_ep = lookup_stid(t, stid);
 	if (!parent_ep) {
@@ -2152,8 +2288,10 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 		goto reject;
 	}
 
-	if (peer_mss && child_ep->mtu > (peer_mss + 40))
-		child_ep->mtu = peer_mss + 40;
+	hdrs = sizeof(struct iphdr) + sizeof(struct tcphdr) +
+	       ((enable_tcp_timestamps && req->tcpopt.tstamp) ? 12 : 0);
+	if (peer_mss && child_ep->mtu > (peer_mss + hdrs))
+		child_ep->mtu = peer_mss + hdrs;
 
 	state_set(&child_ep->com, CONNECTING);
 	child_ep->com.dev = dev;
@@ -2730,13 +2868,15 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	struct c4iw_dev *dev = to_c4iw_dev(cm_id->device);
 	struct c4iw_ep *ep;
 	int err = 0;
-	struct sockaddr_in *laddr = (struct sockaddr_in *)&cm_id->local_addr;
-	struct sockaddr_in *raddr = (struct sockaddr_in *)&cm_id->remote_addr;
-	struct sockaddr_in6 *laddr6 = (struct sockaddr_in6 *)&cm_id->local_addr;
-	struct sockaddr_in6 *raddr6 = (struct sockaddr_in6 *)
-				      &cm_id->remote_addr;
+	struct sockaddr_in *laddr;
+	struct sockaddr_in *raddr;
+	struct sockaddr_in6 *laddr6;
+	struct sockaddr_in6 *raddr6;
+	struct iwpm_dev_data pm_reg_msg;
+	struct iwpm_sa_data pm_msg;
 	__u8 *ra;
 	int iptype;
+	int iwpm_err = 0;
 
 	if ((conn_param->ord > c4iw_max_read_depth) ||
 	    (conn_param->ird > c4iw_max_read_depth)) {
@@ -2767,7 +2907,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	if (!ep->com.qp) {
 		PDBG("%s qpn 0x%x not found!\n", __func__, conn_param->qpn);
 		err = -EINVAL;
-		goto fail2;
+		goto fail1;
 	}
 	ref_qp(ep);
 	PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn,
@@ -2780,10 +2920,50 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	if (ep->atid == -1) {
 		printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__);
 		err = -ENOMEM;
-		goto fail2;
+		goto fail1;
 	}
 	insert_handle(dev, &dev->atid_idr, ep, ep->atid);
 
+	memcpy(&ep->com.local_addr, &cm_id->local_addr,
+	       sizeof(ep->com.local_addr));
+	memcpy(&ep->com.remote_addr, &cm_id->remote_addr,
+	       sizeof(ep->com.remote_addr));
+
+	/* No port mapper available, go with the specified peer information */
+	memcpy(&ep->com.mapped_local_addr, &cm_id->local_addr,
+	       sizeof(ep->com.mapped_local_addr));
+	memcpy(&ep->com.mapped_remote_addr, &cm_id->remote_addr,
+	       sizeof(ep->com.mapped_remote_addr));
+
+	c4iw_form_reg_msg(dev, &pm_reg_msg);
+	iwpm_err = iwpm_register_pid(&pm_reg_msg, RDMA_NL_C4IW);
+	if (iwpm_err) {
+		PDBG("%s: Port Mapper reg pid fail (err = %d).\n",
+			__func__, iwpm_err);
+	}
+	if (iwpm_valid_pid() && !iwpm_err) {
+		c4iw_form_pm_msg(ep, &pm_msg);
+		iwpm_err = iwpm_add_and_query_mapping(&pm_msg, RDMA_NL_C4IW);
+		if (iwpm_err)
+			PDBG("%s: Port Mapper query fail (err = %d).\n",
+				__func__, iwpm_err);
+		else
+			c4iw_record_pm_msg(ep, &pm_msg);
+	}
+	if (iwpm_create_mapinfo(&ep->com.local_addr,
+				&ep->com.mapped_local_addr, RDMA_NL_C4IW)) {
+		iwpm_remove_mapping(&ep->com.local_addr, RDMA_NL_C4IW);
+		err = -ENOMEM;
+		goto fail1;
+	}
+	print_addr(&ep->com, __func__, "add_query/create_mapinfo");
+	set_bit(RELEASE_MAPINFO, &ep->com.flags);
+
+	laddr = (struct sockaddr_in *)&ep->com.mapped_local_addr;
+	raddr = (struct sockaddr_in *)&ep->com.mapped_remote_addr;
+	laddr6 = (struct sockaddr_in6 *)&ep->com.mapped_local_addr;
+	raddr6 = (struct sockaddr_in6 *) &ep->com.mapped_remote_addr;
+
 	if (cm_id->remote_addr.ss_family == AF_INET) {
 		iptype = 4;
 		ra = (__u8 *)&raddr->sin_addr;
@@ -2794,7 +2974,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 		if ((__force int)raddr->sin_addr.s_addr == INADDR_ANY) {
 			err = pick_local_ipaddrs(dev, cm_id);
 			if (err)
-				goto fail2;
+				goto fail1;
 		}
 
 		/* find a route */
@@ -2814,7 +2994,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 		if (ipv6_addr_type(&raddr6->sin6_addr) == IPV6_ADDR_ANY) {
 			err = pick_local_ip6addrs(dev, cm_id);
 			if (err)
-				goto fail2;
+				goto fail1;
 		}
 
 		/* find a route */
@@ -2830,13 +3010,13 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	if (!ep->dst) {
 		printk(KERN_ERR MOD "%s - cannot find route.\n", __func__);
 		err = -EHOSTUNREACH;
-		goto fail3;
+		goto fail2;
 	}
 
 	err = import_ep(ep, iptype, ra, ep->dst, ep->com.dev, true);
 	if (err) {
 		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
-		goto fail4;
+		goto fail3;
 	}
 
 	PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n",
@@ -2845,10 +3025,6 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 
 	state_set(&ep->com, CONNECTING);
 	ep->tos = 0;
-	memcpy(&ep->com.local_addr, &cm_id->local_addr,
-	       sizeof(ep->com.local_addr));
-	memcpy(&ep->com.remote_addr, &cm_id->remote_addr,
-	       sizeof(ep->com.remote_addr));
 
 	/* send connect request to rnic */
 	err = send_connect(ep);
@@ -2856,12 +3032,12 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 		goto out;
 
 	cxgb4_l2t_release(ep->l2t);
-fail4:
-	dst_release(ep->dst);
 fail3:
+	dst_release(ep->dst);
+fail2:
 	remove_handle(ep->com.dev, &ep->com.dev->atid_idr, ep->atid);
 	cxgb4_free_atid(ep->com.dev->rdev.lldi.tids, ep->atid);
-fail2:
+fail1:
 	cm_id->rem_ref(cm_id);
 	c4iw_put_ep(&ep->com);
 out:
@@ -2871,7 +3047,8 @@ out:
 static int create_server6(struct c4iw_dev *dev, struct c4iw_listen_ep *ep)
 {
 	int err;
-	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&ep->com.local_addr;
+	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)
+				    &ep->com.mapped_local_addr;
 
 	c4iw_init_wr_wait(&ep->com.wr_wait);
 	err = cxgb4_create_server6(ep->com.dev->rdev.lldi.ports[0],
@@ -2892,7 +3069,8 @@ static int create_server6(struct c4iw_dev *dev, struct c4iw_listen_ep *ep)
 static int create_server4(struct c4iw_dev *dev, struct c4iw_listen_ep *ep)
 {
 	int err;
-	struct sockaddr_in *sin = (struct sockaddr_in *)&ep->com.local_addr;
+	struct sockaddr_in *sin = (struct sockaddr_in *)
+				  &ep->com.mapped_local_addr;
 
 	if (dev->rdev.lldi.enable_fw_ofld_conn) {
 		do {
@@ -2927,6 +3105,9 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog)
 	int err = 0;
 	struct c4iw_dev *dev = to_c4iw_dev(cm_id->device);
 	struct c4iw_listen_ep *ep;
+	struct iwpm_dev_data pm_reg_msg;
+	struct iwpm_sa_data pm_msg;
+	int iwpm_err = 0;
 
 	might_sleep();
 
@@ -2961,6 +3142,37 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog)
 		goto fail2;
 	}
 	insert_handle(dev, &dev->stid_idr, ep, ep->stid);
+
+	/* No port mapper available, go with the specified info */
+	memcpy(&ep->com.mapped_local_addr, &cm_id->local_addr,
+	       sizeof(ep->com.mapped_local_addr));
+
+	c4iw_form_reg_msg(dev, &pm_reg_msg);
+	iwpm_err = iwpm_register_pid(&pm_reg_msg, RDMA_NL_C4IW);
+	if (iwpm_err) {
+		PDBG("%s: Port Mapper reg pid fail (err = %d).\n",
+			__func__, iwpm_err);
+	}
+	if (iwpm_valid_pid() && !iwpm_err) {
+		memcpy(&pm_msg.loc_addr, &ep->com.local_addr,
+				sizeof(ep->com.local_addr));
+		iwpm_err = iwpm_add_mapping(&pm_msg, RDMA_NL_C4IW);
+		if (iwpm_err)
+			PDBG("%s: Port Mapper query fail (err = %d).\n",
+				__func__, iwpm_err);
+		else
+			memcpy(&ep->com.mapped_local_addr,
+				&pm_msg.mapped_loc_addr,
+				sizeof(ep->com.mapped_local_addr));
+	}
+	if (iwpm_create_mapinfo(&ep->com.local_addr,
+				&ep->com.mapped_local_addr, RDMA_NL_C4IW)) {
+		err = -ENOMEM;
+		goto fail3;
+	}
+	print_addr(&ep->com, __func__, "add_mapping/create_mapinfo");
+
+	set_bit(RELEASE_MAPINFO, &ep->com.flags);
 	state_set(&ep->com, LISTEN);
 	if (ep->com.local_addr.ss_family == AF_INET)
 		err = create_server4(dev, ep);
@@ -2970,6 +3182,8 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog)
 		cm_id->provider_data = ep;
 		goto out;
 	}
+
+fail3:
 	cxgb4_free_stid(ep->com.dev->rdev.lldi.tids, ep->stid,
 			ep->com.local_addr.ss_family);
 fail2:
diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c
index cfaa56ada189..c04292c950f1 100644
--- a/drivers/infiniband/hw/cxgb4/cq.c
+++ b/drivers/infiniband/hw/cxgb4/cq.c
@@ -134,7 +134,8 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq,
 			V_FW_RI_RES_WR_IQANUS(0) |
 			V_FW_RI_RES_WR_IQANUD(1) |
 			F_FW_RI_RES_WR_IQANDST |
-			V_FW_RI_RES_WR_IQANDSTINDEX(*rdev->lldi.rxq_ids));
+			V_FW_RI_RES_WR_IQANDSTINDEX(
+				rdev->lldi.ciq_ids[cq->vector]));
 	res->u.cq.iqdroprss_to_iqesize = cpu_to_be16(
 			F_FW_RI_RES_WR_IQDROPRSS |
 			V_FW_RI_RES_WR_IQPCIECH(2) |
@@ -870,6 +871,9 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries,
 
 	rhp = to_c4iw_dev(ibdev);
 
+	if (vector >= rhp->rdev.lldi.nciq)
+		return ERR_PTR(-EINVAL);
+
 	chp = kzalloc(sizeof(*chp), GFP_KERNEL);
 	if (!chp)
 		return ERR_PTR(-ENOMEM);
@@ -915,6 +919,7 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries,
 	}
 	chp->cq.size = hwentries;
 	chp->cq.memsize = memsize;
+	chp->cq.vector = vector;
 
 	ret = create_cq(&rhp->rdev, &chp->cq,
 			ucontext ? &ucontext->uctx : &rhp->rdev.uctx);
@@ -940,7 +945,6 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries,
 		if (!mm2)
 			goto err4;
 
-		memset(&uresp, 0, sizeof(uresp));
 		uresp.qid_mask = rhp->rdev.cqmask;
 		uresp.cqid = chp->cq.cqid;
 		uresp.size = chp->cq.size;
@@ -951,7 +955,8 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries,
 		uresp.gts_key = ucontext->key;
 		ucontext->key += PAGE_SIZE;
 		spin_unlock(&ucontext->mmap_lock);
-		ret = ib_copy_to_udata(udata, &uresp, sizeof uresp);
+		ret = ib_copy_to_udata(udata, &uresp,
+				       sizeof(uresp) - sizeof(uresp.reserved));
 		if (ret)
 			goto err5;
 
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index f4fa50a609e2..dd93aadc996e 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -77,6 +77,16 @@ struct c4iw_debugfs_data {
 	int pos;
 };
 
+/* registered cxgb4 netlink callbacks */
+static struct ibnl_client_cbs c4iw_nl_cb_table[] = {
+	[RDMA_NL_IWPM_REG_PID] = {.dump = iwpm_register_pid_cb},
+	[RDMA_NL_IWPM_ADD_MAPPING] = {.dump = iwpm_add_mapping_cb},
+	[RDMA_NL_IWPM_QUERY_MAPPING] = {.dump = iwpm_add_and_query_mapping_cb},
+	[RDMA_NL_IWPM_HANDLE_ERR] = {.dump = iwpm_mapping_error_cb},
+	[RDMA_NL_IWPM_MAPINFO] = {.dump = iwpm_mapping_info_cb},
+	[RDMA_NL_IWPM_MAPINFO_NUM] = {.dump = iwpm_ack_mapping_info_cb}
+};
+
 static int count_idrs(int id, void *p, void *data)
 {
 	int *countp = data;
@@ -113,35 +123,49 @@ static int dump_qp(int id, void *p, void *data)
 				&qp->ep->com.local_addr;
 			struct sockaddr_in *rsin = (struct sockaddr_in *)
 				&qp->ep->com.remote_addr;
+			struct sockaddr_in *mapped_lsin = (struct sockaddr_in *)
+				&qp->ep->com.mapped_local_addr;
+			struct sockaddr_in *mapped_rsin = (struct sockaddr_in *)
+				&qp->ep->com.mapped_remote_addr;
 
 			cc = snprintf(qpd->buf + qpd->pos, space,
 				      "rc qp sq id %u rq id %u state %u "
 				      "onchip %u ep tid %u state %u "
-				      "%pI4:%u->%pI4:%u\n",
+				      "%pI4:%u/%u->%pI4:%u/%u\n",
 				      qp->wq.sq.qid, qp->wq.rq.qid,
 				      (int)qp->attr.state,
 				      qp->wq.sq.flags & T4_SQ_ONCHIP,
 				      qp->ep->hwtid, (int)qp->ep->com.state,
 				      &lsin->sin_addr, ntohs(lsin->sin_port),
-				      &rsin->sin_addr, ntohs(rsin->sin_port));
+				      ntohs(mapped_lsin->sin_port),
+				      &rsin->sin_addr, ntohs(rsin->sin_port),
+				      ntohs(mapped_rsin->sin_port));
 		} else {
 			struct sockaddr_in6 *lsin6 = (struct sockaddr_in6 *)
 				&qp->ep->com.local_addr;
 			struct sockaddr_in6 *rsin6 = (struct sockaddr_in6 *)
 				&qp->ep->com.remote_addr;
+			struct sockaddr_in6 *mapped_lsin6 =
+				(struct sockaddr_in6 *)
+				&qp->ep->com.mapped_local_addr;
+			struct sockaddr_in6 *mapped_rsin6 =
+				(struct sockaddr_in6 *)
+				&qp->ep->com.mapped_remote_addr;
 
 			cc = snprintf(qpd->buf + qpd->pos, space,
 				      "rc qp sq id %u rq id %u state %u "
 				      "onchip %u ep tid %u state %u "
-				      "%pI6:%u->%pI6:%u\n",
+				      "%pI6:%u/%u->%pI6:%u/%u\n",
 				      qp->wq.sq.qid, qp->wq.rq.qid,
 				      (int)qp->attr.state,
 				      qp->wq.sq.flags & T4_SQ_ONCHIP,
 				      qp->ep->hwtid, (int)qp->ep->com.state,
 				      &lsin6->sin6_addr,
 				      ntohs(lsin6->sin6_port),
+				      ntohs(mapped_lsin6->sin6_port),
 				      &rsin6->sin6_addr,
-				      ntohs(rsin6->sin6_port));
+				      ntohs(rsin6->sin6_port),
+				      ntohs(mapped_rsin6->sin6_port));
 		}
 	} else
 		cc = snprintf(qpd->buf + qpd->pos, space,
@@ -386,31 +410,43 @@ static int dump_ep(int id, void *p, void *data)
 			&ep->com.local_addr;
 		struct sockaddr_in *rsin = (struct sockaddr_in *)
 			&ep->com.remote_addr;
+		struct sockaddr_in *mapped_lsin = (struct sockaddr_in *)
+			&ep->com.mapped_local_addr;
+		struct sockaddr_in *mapped_rsin = (struct sockaddr_in *)
+			&ep->com.mapped_remote_addr;
 
 		cc = snprintf(epd->buf + epd->pos, space,
 			      "ep %p cm_id %p qp %p state %d flags 0x%lx "
 			      "history 0x%lx hwtid %d atid %d "
-			      "%pI4:%d <-> %pI4:%d\n",
+			      "%pI4:%d/%d <-> %pI4:%d/%d\n",
 			      ep, ep->com.cm_id, ep->com.qp,
 			      (int)ep->com.state, ep->com.flags,
 			      ep->com.history, ep->hwtid, ep->atid,
 			      &lsin->sin_addr, ntohs(lsin->sin_port),
-			      &rsin->sin_addr, ntohs(rsin->sin_port));
+			      ntohs(mapped_lsin->sin_port),
+			      &rsin->sin_addr, ntohs(rsin->sin_port),
+			      ntohs(mapped_rsin->sin_port));
 	} else {
 		struct sockaddr_in6 *lsin6 = (struct sockaddr_in6 *)
 			&ep->com.local_addr;
 		struct sockaddr_in6 *rsin6 = (struct sockaddr_in6 *)
 			&ep->com.remote_addr;
+		struct sockaddr_in6 *mapped_lsin6 = (struct sockaddr_in6 *)
+			&ep->com.mapped_local_addr;
+		struct sockaddr_in6 *mapped_rsin6 = (struct sockaddr_in6 *)
+			&ep->com.mapped_remote_addr;
 
 		cc = snprintf(epd->buf + epd->pos, space,
 			      "ep %p cm_id %p qp %p state %d flags 0x%lx "
 			      "history 0x%lx hwtid %d atid %d "
-			      "%pI6:%d <-> %pI6:%d\n",
+			      "%pI6:%d/%d <-> %pI6:%d/%d\n",
 			      ep, ep->com.cm_id, ep->com.qp,
 			      (int)ep->com.state, ep->com.flags,
 			      ep->com.history, ep->hwtid, ep->atid,
 			      &lsin6->sin6_addr, ntohs(lsin6->sin6_port),
-			      &rsin6->sin6_addr, ntohs(rsin6->sin6_port));
+			      ntohs(mapped_lsin6->sin6_port),
+			      &rsin6->sin6_addr, ntohs(rsin6->sin6_port),
+			      ntohs(mapped_rsin6->sin6_port));
 	}
 	if (cc < space)
 		epd->pos += cc;
@@ -431,23 +467,29 @@ static int dump_listen_ep(int id, void *p, void *data)
 	if (ep->com.local_addr.ss_family == AF_INET) {
 		struct sockaddr_in *lsin = (struct sockaddr_in *)
 			&ep->com.local_addr;
+		struct sockaddr_in *mapped_lsin = (struct sockaddr_in *)
+			&ep->com.mapped_local_addr;
 
 		cc = snprintf(epd->buf + epd->pos, space,
 			      "ep %p cm_id %p state %d flags 0x%lx stid %d "
-			      "backlog %d %pI4:%d\n",
+			      "backlog %d %pI4:%d/%d\n",
 			      ep, ep->com.cm_id, (int)ep->com.state,
 			      ep->com.flags, ep->stid, ep->backlog,
-			      &lsin->sin_addr, ntohs(lsin->sin_port));
+			      &lsin->sin_addr, ntohs(lsin->sin_port),
+			      ntohs(mapped_lsin->sin_port));
 	} else {
 		struct sockaddr_in6 *lsin6 = (struct sockaddr_in6 *)
 			&ep->com.local_addr;
+		struct sockaddr_in6 *mapped_lsin6 = (struct sockaddr_in6 *)
+			&ep->com.mapped_local_addr;
 
 		cc = snprintf(epd->buf + epd->pos, space,
 			      "ep %p cm_id %p state %d flags 0x%lx stid %d "
-			      "backlog %d %pI6:%d\n",
+			      "backlog %d %pI6:%d/%d\n",
 			      ep, ep->com.cm_id, (int)ep->com.state,
 			      ep->com.flags, ep->stid, ep->backlog,
-			      &lsin6->sin6_addr, ntohs(lsin6->sin6_port));
+			      &lsin6->sin6_addr, ntohs(lsin6->sin6_port),
+			      ntohs(mapped_lsin6->sin6_port));
 	}
 	if (cc < space)
 		epd->pos += cc;
@@ -687,6 +729,7 @@ static void c4iw_dealloc(struct uld_ctx *ctx)
 	if (ctx->dev->rdev.oc_mw_kva)
 		iounmap(ctx->dev->rdev.oc_mw_kva);
 	ib_dealloc_device(&ctx->dev->ibdev);
+	iwpm_exit(RDMA_NL_C4IW);
 	ctx->dev = NULL;
 }
 
@@ -736,6 +779,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 			pci_resource_len(devp->rdev.lldi.pdev, 2));
 		if (!devp->rdev.bar2_kva) {
 			pr_err(MOD "Unable to ioremap BAR2\n");
+			ib_dealloc_device(&devp->ibdev);
 			return ERR_PTR(-EINVAL);
 		}
 	} else if (ocqp_supported(infop)) {
@@ -747,6 +791,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 			devp->rdev.lldi.vr->ocq.size);
 		if (!devp->rdev.oc_mw_kva) {
 			pr_err(MOD "Unable to ioremap onchip mem\n");
+			ib_dealloc_device(&devp->ibdev);
 			return ERR_PTR(-EINVAL);
 		}
 	}
@@ -780,6 +825,14 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 					c4iw_debugfs_root);
 		setup_debugfs(devp);
 	}
+
+	ret = iwpm_init(RDMA_NL_C4IW);
+	if (ret) {
+		pr_err("port mapper initialization failed with %d\n", ret);
+		ib_dealloc_device(&devp->ibdev);
+		return ERR_PTR(ret);
+	}
+
 	return devp;
 }
 
@@ -1274,6 +1327,11 @@ static int __init c4iw_init_module(void)
 		printk(KERN_WARNING MOD
 		       "could not create debugfs entry, continuing\n");
 
+	if (ibnl_add_client(RDMA_NL_C4IW, RDMA_NL_IWPM_NUM_OPS,
+			    c4iw_nl_cb_table))
+		pr_err("%s[%u]: Failed to add netlink callback\n"
+		       , __func__, __LINE__);
+
 	cxgb4_register_uld(CXGB4_ULD_RDMA, &c4iw_uld_info);
 
 	return 0;
@@ -1291,6 +1349,7 @@ static void __exit c4iw_exit_module(void)
 	}
 	mutex_unlock(&dev_mutex);
 	cxgb4_unregister_uld(CXGB4_ULD_RDMA);
+	ibnl_remove_client(RDMA_NL_C4IW);
 	c4iw_cm_term();
 	debugfs_remove_recursive(c4iw_debugfs_root);
 }
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 7474b490760a..125bc5d1e175 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -52,6 +52,8 @@
 
 #include <rdma/ib_verbs.h>
 #include <rdma/iw_cm.h>
+#include <rdma/rdma_netlink.h>
+#include <rdma/iw_portmap.h>
 
 #include "cxgb4.h"
 #include "cxgb4_uld.h"
@@ -728,6 +730,7 @@ enum c4iw_ep_flags {
 	CLOSE_SENT		= 3,
 	TIMEOUT                 = 4,
 	QP_REFERENCED           = 5,
+	RELEASE_MAPINFO		= 6,
 };
 
 enum c4iw_ep_history {
@@ -764,6 +767,8 @@ struct c4iw_ep_common {
 	struct mutex mutex;
 	struct sockaddr_storage local_addr;
 	struct sockaddr_storage remote_addr;
+	struct sockaddr_storage mapped_local_addr;
+	struct sockaddr_storage mapped_remote_addr;
 	struct c4iw_wr_wait wr_wait;
 	unsigned long flags;
 	unsigned long history;
@@ -805,7 +810,48 @@ struct c4iw_ep {
 	u8 retry_with_mpa_v1;
 	u8 tried_with_mpa_v1;
 	unsigned int retry_count;
-};
+	int snd_win;
+	int rcv_win;
+};
+
+static inline void print_addr(struct c4iw_ep_common *epc, const char *func,
+			      const char *msg)
+{
+
+#define SINA(a) (&(((struct sockaddr_in *)(a))->sin_addr.s_addr))
+#define SINP(a) ntohs(((struct sockaddr_in *)(a))->sin_port)
+#define SIN6A(a) (&(((struct sockaddr_in6 *)(a))->sin6_addr))
+#define SIN6P(a) ntohs(((struct sockaddr_in6 *)(a))->sin6_port)
+
+	if (c4iw_debug) {
+		switch (epc->local_addr.ss_family) {
+		case AF_INET:
+			PDBG("%s %s %pI4:%u/%u <-> %pI4:%u/%u\n",
+			     func, msg, SINA(&epc->local_addr),
+			     SINP(&epc->local_addr),
+			     SINP(&epc->mapped_local_addr),
+			     SINA(&epc->remote_addr),
+			     SINP(&epc->remote_addr),
+			     SINP(&epc->mapped_remote_addr));
+			break;
+		case AF_INET6:
+			PDBG("%s %s %pI6:%u/%u <-> %pI6:%u/%u\n",
+			     func, msg, SIN6A(&epc->local_addr),
+			     SIN6P(&epc->local_addr),
+			     SIN6P(&epc->mapped_local_addr),
+			     SIN6A(&epc->remote_addr),
+			     SIN6P(&epc->remote_addr),
+			     SIN6P(&epc->mapped_remote_addr));
+			break;
+		default:
+			break;
+		}
+	}
+#undef SINA
+#undef SINP
+#undef SIN6A
+#undef SIN6P
+}
 
 static inline struct c4iw_ep *to_ep(struct iw_cm_id *cm_id)
 {
diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c
index a94a3e12c349..b1d305338de6 100644
--- a/drivers/infiniband/hw/cxgb4/provider.c
+++ b/drivers/infiniband/hw/cxgb4/provider.c
@@ -122,7 +122,7 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev,
 	INIT_LIST_HEAD(&context->mmaps);
 	spin_lock_init(&context->mmap_lock);
 
-	if (udata->outlen < sizeof(uresp)) {
+	if (udata->outlen < sizeof(uresp) - sizeof(uresp.reserved)) {
 		if (!warned++)
 			pr_err(MOD "Warning - downlevel libcxgb4 (non-fatal), device status page disabled.");
 		rhp->rdev.flags |= T4_STATUS_PAGE_DISABLED;
@@ -140,7 +140,8 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev,
 		context->key += PAGE_SIZE;
 		spin_unlock(&context->mmap_lock);
 
-		ret = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
+		ret = ib_copy_to_udata(udata, &uresp,
+				       sizeof(uresp) - sizeof(uresp.reserved));
 		if (ret)
 			goto err_mm;
 
@@ -499,7 +500,7 @@ int c4iw_register_device(struct c4iw_dev *dev)
 	dev->ibdev.node_type = RDMA_NODE_RNIC;
 	memcpy(dev->ibdev.node_desc, C4IW_NODE_DESC, sizeof(C4IW_NODE_DESC));
 	dev->ibdev.phys_port_cnt = dev->rdev.lldi.nports;
-	dev->ibdev.num_comp_vectors = 1;
+	dev->ibdev.num_comp_vectors =  dev->rdev.lldi.nciq;
 	dev->ibdev.dma_device = &(dev->rdev.lldi.pdev->dev);
 	dev->ibdev.query_device = c4iw_query_device;
 	dev->ibdev.query_port = c4iw_query_port;
diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h
index 2178f3198410..68b0a6bf4eb0 100644
--- a/drivers/infiniband/hw/cxgb4/t4.h
+++ b/drivers/infiniband/hw/cxgb4/t4.h
@@ -542,6 +542,7 @@ struct t4_cq {
 	size_t memsize;
 	__be64 bits_type_ts;
 	u32 cqid;
+	int vector;
 	u16 size; /* including status page */
 	u16 cidx;
 	u16 sw_pidx;
diff --git a/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h b/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h
index 6121ca08fe58..91289a051af9 100644
--- a/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h
+++ b/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h
@@ -848,6 +848,7 @@ enum {                     /* TCP congestion control algorithms */
 #define V_CONG_CNTRL(x) ((x) << S_CONG_CNTRL)
 #define G_CONG_CNTRL(x) (((x) >> S_CONG_CNTRL) & M_CONG_CNTRL)
 
+#define CONG_CNTRL_VALID   (1 << 18)
 #define T5_OPT_2_VALID       (1 << 31)
 
 #endif /* _T4FW_RI_API_H_ */
diff --git a/drivers/infiniband/hw/cxgb4/user.h b/drivers/infiniband/hw/cxgb4/user.h
index 11ccd276e5d9..cbd0ce170728 100644
--- a/drivers/infiniband/hw/cxgb4/user.h
+++ b/drivers/infiniband/hw/cxgb4/user.h
@@ -48,6 +48,7 @@ struct c4iw_create_cq_resp {
 	__u32 cqid;
 	__u32 size;
 	__u32 qid_mask;
+	__u32 reserved; /* explicit padding (optional for i386) */
 };
 
 
@@ -74,5 +75,6 @@ struct c4iw_create_qp_resp {
 struct c4iw_alloc_ucontext_resp {
 	__u64 status_page_key;
 	__u32 status_page_size;
+	__u32 reserved; /* explicit padding (optional for i386) */
 };
 #endif
diff --git a/drivers/infiniband/hw/ipath/ipath_diag.c b/drivers/infiniband/hw/ipath/ipath_diag.c
index e2f9a51f4a38..45802e97332e 100644
--- a/drivers/infiniband/hw/ipath/ipath_diag.c
+++ b/drivers/infiniband/hw/ipath/ipath_diag.c
@@ -346,6 +346,10 @@ static ssize_t ipath_diagpkt_write(struct file *fp,
 			ret = -EFAULT;
 			goto bail;
 		}
+		dp.len = odp.len;
+		dp.unit = odp.unit;
+		dp.data = odp.data;
+		dp.pbc_wd = 0;
 	} else {
 		ret = -EINVAL;
 		goto bail;
diff --git a/drivers/infiniband/hw/ipath/ipath_intr.c b/drivers/infiniband/hw/ipath/ipath_intr.c
index 26dfbc8ee0f1..01ba792791a0 100644
--- a/drivers/infiniband/hw/ipath/ipath_intr.c
+++ b/drivers/infiniband/hw/ipath/ipath_intr.c
@@ -70,7 +70,7 @@ void ipath_disarm_senderrbufs(struct ipath_devdata *dd)
 	if (sbuf[0] || sbuf[1] || (piobcnt > 128 && (sbuf[2] || sbuf[3]))) {
 		int i;
 		if (ipath_debug & (__IPATH_PKTDBG|__IPATH_DBG) &&
-			dd->ipath_lastcancel > jiffies) {
+			time_after(dd->ipath_lastcancel, jiffies)) {
 			__IPATH_DBG_WHICH(__IPATH_PKTDBG|__IPATH_DBG,
 					  "SendbufErrs %lx %lx", sbuf[0],
 					  sbuf[1]);
@@ -755,7 +755,7 @@ static int handle_errors(struct ipath_devdata *dd, ipath_err_t errs)
 
 	/* likely due to cancel; so suppress message unless verbose */
 	if ((errs & (INFINIPATH_E_SPKTLEN | INFINIPATH_E_SPIOARMLAUNCH)) &&
-		dd->ipath_lastcancel > jiffies) {
+		time_after(dd->ipath_lastcancel, jiffies)) {
 		/* armlaunch takes precedence; it often causes both. */
 		ipath_cdbg(VERBOSE,
 			"Suppressed %s error (%llx) after sendbuf cancel\n",
diff --git a/drivers/infiniband/hw/ipath/ipath_sdma.c b/drivers/infiniband/hw/ipath/ipath_sdma.c
index 98ac18ec977e..17a517766ad2 100644
--- a/drivers/infiniband/hw/ipath/ipath_sdma.c
+++ b/drivers/infiniband/hw/ipath/ipath_sdma.c
@@ -247,7 +247,7 @@ static void sdma_abort_task(unsigned long opaque)
 
 	/* ipath_sdma_abort() is done, waiting for interrupt */
 	if (status == IPATH_SDMA_ABORT_DISARMED) {
-		if (jiffies < dd->ipath_sdma_abort_intr_timeout)
+		if (time_before(jiffies, dd->ipath_sdma_abort_intr_timeout))
 			goto resched_noprint;
 		/* give up, intr got lost somewhere */
 		ipath_dbg("give up waiting for SDMADISABLED intr\n");
@@ -341,7 +341,7 @@ resched:
 	 * JAG - this is bad to just have default be a loop without
 	 * state change
 	 */
-	if (jiffies > dd->ipath_sdma_abort_jiffies) {
+	if (time_after(jiffies, dd->ipath_sdma_abort_jiffies)) {
 		ipath_dbg("looping with status 0x%08lx\n",
 			  dd->ipath_sdma_status);
 		dd->ipath_sdma_abort_jiffies = jiffies + 5 * HZ;
diff --git a/drivers/infiniband/hw/mlx4/ah.c b/drivers/infiniband/hw/mlx4/ah.c
index 170dca608042..2d8c3397774f 100644
--- a/drivers/infiniband/hw/mlx4/ah.c
+++ b/drivers/infiniband/hw/mlx4/ah.c
@@ -73,7 +73,7 @@ static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr
 {
 	struct mlx4_ib_dev *ibdev = to_mdev(pd->device);
 	struct mlx4_dev *dev = ibdev->dev;
-	int is_mcast;
+	int is_mcast = 0;
 	struct in6_addr in6;
 	u16 vlan_tag;
 
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 5f640814cc81..1066eec854a9 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -102,7 +102,7 @@ static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf *
 	int err;
 
 	err = mlx4_buf_alloc(dev->dev, nent * dev->dev->caps.cqe_size,
-			     PAGE_SIZE * 2, &buf->buf);
+			     PAGE_SIZE * 2, &buf->buf, GFP_KERNEL);
 
 	if (err)
 		goto out;
@@ -113,7 +113,7 @@ static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf *
 	if (err)
 		goto err_buf;
 
-	err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf);
+	err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf, GFP_KERNEL);
 	if (err)
 		goto err_mtt;
 
@@ -209,7 +209,7 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector
 
 		uar = &to_mucontext(context)->uar;
 	} else {
-		err = mlx4_db_alloc(dev->dev, &cq->db, 1);
+		err = mlx4_db_alloc(dev->dev, &cq->db, 1, GFP_KERNEL);
 		if (err)
 			goto err_cq;
 
diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index fd36ec672632..287ad0564acd 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -478,10 +478,6 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port,
 	if (!tun_ctx || tun_ctx->state != DEMUX_PV_STATE_ACTIVE)
 		return -EAGAIN;
 
-	/* QP0 forwarding only for Dom0 */
-	if (!dest_qpt && (mlx4_master_func_num(dev->dev) != slave))
-		return -EINVAL;
-
 	if (!dest_qpt)
 		tun_qp = &tun_ctx->qp[0];
 	else
@@ -667,6 +663,21 @@ static int mlx4_ib_demux_mad(struct ib_device *ibdev, u8 port,
 	}
 	/* Class-specific handling */
 	switch (mad->mad_hdr.mgmt_class) {
+	case IB_MGMT_CLASS_SUBN_LID_ROUTED:
+	case IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE:
+		/* 255 indicates the dom0 */
+		if (slave != 255 && slave != mlx4_master_func_num(dev->dev)) {
+			if (!mlx4_vf_smi_enabled(dev->dev, slave, port))
+				return -EPERM;
+			/* for a VF. drop unsolicited MADs */
+			if (!(mad->mad_hdr.method & IB_MGMT_METHOD_RESP)) {
+				mlx4_ib_warn(ibdev, "demux QP0. rejecting unsolicited mad for slave %d class 0x%x, method 0x%x\n",
+					     slave, mad->mad_hdr.mgmt_class,
+					     mad->mad_hdr.method);
+				return -EINVAL;
+			}
+		}
+		break;
 	case IB_MGMT_CLASS_SUBN_ADM:
 		if (mlx4_ib_demux_sa_handler(ibdev, port, slave,
 					     (struct ib_sa_mad *) mad))
@@ -1165,10 +1176,6 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port,
 	if (!sqp_ctx || sqp_ctx->state != DEMUX_PV_STATE_ACTIVE)
 		return -EAGAIN;
 
-	/* QP0 forwarding only for Dom0 */
-	if (dest_qpt == IB_QPT_SMI && (mlx4_master_func_num(dev->dev) != slave))
-		return -EINVAL;
-
 	if (dest_qpt == IB_QPT_SMI) {
 		src_qpnum = 0;
 		sqp = &sqp_ctx->qp[0];
@@ -1285,11 +1292,6 @@ static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc
 			     "belongs to another slave\n", wc->src_qp);
 		return;
 	}
-	if (slave != mlx4_master_func_num(dev->dev) && !(wc->src_qp & 0x2)) {
-		mlx4_ib_warn(ctx->ib_dev, "can't multiplex bad sqp:%d: "
-			     "non-master trying to send QP0 packets\n", wc->src_qp);
-		return;
-	}
 
 	/* Map transaction ID */
 	ib_dma_sync_single_for_cpu(ctx->ib_dev, tun_qp->ring[wr_ix].map,
@@ -1317,6 +1319,12 @@ static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc
 
 	/* Class-specific handling */
 	switch (tunnel->mad.mad_hdr.mgmt_class) {
+	case IB_MGMT_CLASS_SUBN_LID_ROUTED:
+	case IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE:
+		if (slave != mlx4_master_func_num(dev->dev) &&
+		    !mlx4_vf_smi_enabled(dev->dev, slave, ctx->port))
+			return;
+		break;
 	case IB_MGMT_CLASS_SUBN_ADM:
 		if (mlx4_ib_multiplex_sa_handler(ctx->ib_dev, ctx->port, slave,
 			      (struct ib_sa_mad *) &tunnel->mad))
@@ -1749,9 +1757,9 @@ static int create_pv_resources(struct ib_device *ibdev, int slave, int port,
 		return -EEXIST;
 
 	ctx->state = DEMUX_PV_STATE_STARTING;
-	/* have QP0 only on port owner, and only if link layer is IB */
-	if (ctx->slave == mlx4_master_func_num(to_mdev(ctx->ib_dev)->dev) &&
-	    rdma_port_get_link_layer(ibdev, ctx->port) == IB_LINK_LAYER_INFINIBAND)
+	/* have QP0 only if link layer is IB */
+	if (rdma_port_get_link_layer(ibdev, ctx->port) ==
+	    IB_LINK_LAYER_INFINIBAND)
 		ctx->has_smi = 1;
 
 	if (ctx->has_smi) {
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 199c7896f081..0f7027e7db13 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -545,12 +545,11 @@ static int mlx4_ib_modify_device(struct ib_device *ibdev, int mask,
 	return 0;
 }
 
-static int mlx4_SET_PORT(struct mlx4_ib_dev *dev, u8 port, int reset_qkey_viols,
-			 u32 cap_mask)
+static int mlx4_ib_SET_PORT(struct mlx4_ib_dev *dev, u8 port, int reset_qkey_viols,
+			    u32 cap_mask)
 {
 	struct mlx4_cmd_mailbox *mailbox;
 	int err;
-	u8 is_eth = dev->dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH;
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev->dev);
 	if (IS_ERR(mailbox))
@@ -564,8 +563,8 @@ static int mlx4_SET_PORT(struct mlx4_ib_dev *dev, u8 port, int reset_qkey_viols,
 		((__be32 *) mailbox->buf)[1] = cpu_to_be32(cap_mask);
 	}
 
-	err = mlx4_cmd(dev->dev, mailbox->dma, port, is_eth, MLX4_CMD_SET_PORT,
-		       MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE);
+	err = mlx4_cmd(dev->dev, mailbox->dma, port, 0, MLX4_CMD_SET_PORT,
+		       MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED);
 
 	mlx4_free_cmd_mailbox(dev->dev, mailbox);
 	return err;
@@ -574,11 +573,20 @@ static int mlx4_SET_PORT(struct mlx4_ib_dev *dev, u8 port, int reset_qkey_viols,
 static int mlx4_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 			       struct ib_port_modify *props)
 {
+	struct mlx4_ib_dev *mdev = to_mdev(ibdev);
+	u8 is_eth = mdev->dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH;
 	struct ib_port_attr attr;
 	u32 cap_mask;
 	int err;
 
-	mutex_lock(&to_mdev(ibdev)->cap_mask_mutex);
+	/* return OK if this is RoCE. CM calls ib_modify_port() regardless
+	 * of whether port link layer is ETH or IB. For ETH ports, qkey
+	 * violations and port capabilities are not meaningful.
+	 */
+	if (is_eth)
+		return 0;
+
+	mutex_lock(&mdev->cap_mask_mutex);
 
 	err = mlx4_ib_query_port(ibdev, port, &attr);
 	if (err)
@@ -587,9 +595,9 @@ static int mlx4_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 	cap_mask = (attr.port_cap_flags | props->set_port_cap_mask) &
 		~props->clr_port_cap_mask;
 
-	err = mlx4_SET_PORT(to_mdev(ibdev), port,
-			    !!(mask & IB_PORT_RESET_QKEY_CNTR),
-			    cap_mask);
+	err = mlx4_ib_SET_PORT(mdev, port,
+			       !!(mask & IB_PORT_RESET_QKEY_CNTR),
+			       cap_mask);
 
 out:
 	mutex_unlock(&to_mdev(ibdev)->cap_mask_mutex);
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index 66b0b7dbd9f4..369da3ca5d64 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -156,6 +156,7 @@ enum mlx4_ib_qp_flags {
 	MLX4_IB_QP_LSO = IB_QP_CREATE_IPOIB_UD_LSO,
 	MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK = IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK,
 	MLX4_IB_QP_NETIF = IB_QP_CREATE_NETIF_QP,
+	MLX4_IB_QP_CREATE_USE_GFP_NOIO = IB_QP_CREATE_USE_GFP_NOIO,
 	MLX4_IB_SRIOV_TUNNEL_QP = 1 << 30,
 	MLX4_IB_SRIOV_SQP = 1 << 31,
 };
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index dc57482ae7af..67780452f0cf 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -608,9 +608,20 @@ static int qp_has_rq(struct ib_qp_init_attr *attr)
 	return !attr->srq;
 }
 
+static int qp0_enabled_vf(struct mlx4_dev *dev, int qpn)
+{
+	int i;
+	for (i = 0; i < dev->caps.num_ports; i++) {
+		if (qpn == dev->caps.qp0_proxy[i])
+			return !!dev->caps.qp0_qkey[i];
+	}
+	return 0;
+}
+
 static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 			    struct ib_qp_init_attr *init_attr,
-			    struct ib_udata *udata, int sqpn, struct mlx4_ib_qp **caller_qp)
+			    struct ib_udata *udata, int sqpn, struct mlx4_ib_qp **caller_qp,
+			    gfp_t gfp)
 {
 	int qpn;
 	int err;
@@ -625,10 +636,13 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		     !(init_attr->create_flags & MLX4_IB_SRIOV_SQP))) {
 			if (init_attr->qp_type == IB_QPT_GSI)
 				qp_type = MLX4_IB_QPT_PROXY_GSI;
-			else if (mlx4_is_master(dev->dev))
-				qp_type = MLX4_IB_QPT_PROXY_SMI_OWNER;
-			else
-				qp_type = MLX4_IB_QPT_PROXY_SMI;
+			else {
+				if (mlx4_is_master(dev->dev) ||
+				    qp0_enabled_vf(dev->dev, sqpn))
+					qp_type = MLX4_IB_QPT_PROXY_SMI_OWNER;
+				else
+					qp_type = MLX4_IB_QPT_PROXY_SMI;
+			}
 		}
 		qpn = sqpn;
 		/* add extra sg entry for tunneling */
@@ -643,7 +657,9 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 			return -EINVAL;
 		if (tnl_init->proxy_qp_type == IB_QPT_GSI)
 			qp_type = MLX4_IB_QPT_TUN_GSI;
-		else if (tnl_init->slave == mlx4_master_func_num(dev->dev))
+		else if (tnl_init->slave == mlx4_master_func_num(dev->dev) ||
+			 mlx4_vf_smi_enabled(dev->dev, tnl_init->slave,
+					     tnl_init->port))
 			qp_type = MLX4_IB_QPT_TUN_SMI_OWNER;
 		else
 			qp_type = MLX4_IB_QPT_TUN_SMI;
@@ -658,14 +674,14 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		if (qp_type == MLX4_IB_QPT_SMI || qp_type == MLX4_IB_QPT_GSI ||
 		    (qp_type & (MLX4_IB_QPT_PROXY_SMI | MLX4_IB_QPT_PROXY_SMI_OWNER |
 				MLX4_IB_QPT_PROXY_GSI | MLX4_IB_QPT_TUN_SMI_OWNER))) {
-			sqp = kzalloc(sizeof (struct mlx4_ib_sqp), GFP_KERNEL);
+			sqp = kzalloc(sizeof (struct mlx4_ib_sqp), gfp);
 			if (!sqp)
 				return -ENOMEM;
 			qp = &sqp->qp;
 			qp->pri.vid = 0xFFFF;
 			qp->alt.vid = 0xFFFF;
 		} else {
-			qp = kzalloc(sizeof (struct mlx4_ib_qp), GFP_KERNEL);
+			qp = kzalloc(sizeof (struct mlx4_ib_qp), gfp);
 			if (!qp)
 				return -ENOMEM;
 			qp->pri.vid = 0xFFFF;
@@ -748,14 +764,14 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 			goto err;
 
 		if (qp_has_rq(init_attr)) {
-			err = mlx4_db_alloc(dev->dev, &qp->db, 0);
+			err = mlx4_db_alloc(dev->dev, &qp->db, 0, gfp);
 			if (err)
 				goto err;
 
 			*qp->db.db = 0;
 		}
 
-		if (mlx4_buf_alloc(dev->dev, qp->buf_size, PAGE_SIZE * 2, &qp->buf)) {
+		if (mlx4_buf_alloc(dev->dev, qp->buf_size, PAGE_SIZE * 2, &qp->buf, gfp)) {
 			err = -ENOMEM;
 			goto err_db;
 		}
@@ -765,13 +781,12 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		if (err)
 			goto err_buf;
 
-		err = mlx4_buf_write_mtt(dev->dev, &qp->mtt, &qp->buf);
+		err = mlx4_buf_write_mtt(dev->dev, &qp->mtt, &qp->buf, gfp);
 		if (err)
 			goto err_mtt;
 
-		qp->sq.wrid  = kmalloc(qp->sq.wqe_cnt * sizeof (u64), GFP_KERNEL);
-		qp->rq.wrid  = kmalloc(qp->rq.wqe_cnt * sizeof (u64), GFP_KERNEL);
-
+		qp->sq.wrid  = kmalloc(qp->sq.wqe_cnt * sizeof (u64), gfp);
+		qp->rq.wrid  = kmalloc(qp->rq.wqe_cnt * sizeof (u64), gfp);
 		if (!qp->sq.wrid || !qp->rq.wrid) {
 			err = -ENOMEM;
 			goto err_wrid;
@@ -801,7 +816,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 			goto err_proxy;
 	}
 
-	err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp);
+	err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp, gfp);
 	if (err)
 		goto err_qpn;
 
@@ -1040,7 +1055,10 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	struct mlx4_ib_qp *qp = NULL;
 	int err;
 	u16 xrcdn = 0;
+	gfp_t gfp;
 
+	gfp = (init_attr->create_flags & MLX4_IB_QP_CREATE_USE_GFP_NOIO) ?
+		GFP_NOIO : GFP_KERNEL;
 	/*
 	 * We only support LSO, vendor flag1, and multicast loopback blocking,
 	 * and only for kernel UD QPs.
@@ -1049,7 +1067,8 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 					MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK |
 					MLX4_IB_SRIOV_TUNNEL_QP |
 					MLX4_IB_SRIOV_SQP |
-					MLX4_IB_QP_NETIF))
+					MLX4_IB_QP_NETIF |
+					MLX4_IB_QP_CREATE_USE_GFP_NOIO))
 		return ERR_PTR(-EINVAL);
 
 	if (init_attr->create_flags & IB_QP_CREATE_NETIF_QP) {
@@ -1059,7 +1078,7 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 
 	if (init_attr->create_flags &&
 	    (udata ||
-	     ((init_attr->create_flags & ~MLX4_IB_SRIOV_SQP) &&
+	     ((init_attr->create_flags & ~(MLX4_IB_SRIOV_SQP | MLX4_IB_QP_CREATE_USE_GFP_NOIO)) &&
 	      init_attr->qp_type != IB_QPT_UD) ||
 	     ((init_attr->create_flags & MLX4_IB_SRIOV_SQP) &&
 	      init_attr->qp_type > IB_QPT_GSI)))
@@ -1079,7 +1098,7 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	case IB_QPT_RC:
 	case IB_QPT_UC:
 	case IB_QPT_RAW_PACKET:
-		qp = kzalloc(sizeof *qp, GFP_KERNEL);
+		qp = kzalloc(sizeof *qp, gfp);
 		if (!qp)
 			return ERR_PTR(-ENOMEM);
 		qp->pri.vid = 0xFFFF;
@@ -1088,7 +1107,7 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	case IB_QPT_UD:
 	{
 		err = create_qp_common(to_mdev(pd->device), pd, init_attr,
-				       udata, 0, &qp);
+				       udata, 0, &qp, gfp);
 		if (err)
 			return ERR_PTR(err);
 
@@ -1106,7 +1125,7 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 
 		err = create_qp_common(to_mdev(pd->device), pd, init_attr, udata,
 				       get_sqp_num(to_mdev(pd->device), init_attr),
-				       &qp);
+				       &qp, gfp);
 		if (err)
 			return ERR_PTR(err);
 
@@ -1938,6 +1957,19 @@ out:
 	return err;
 }
 
+static int vf_get_qp0_qkey(struct mlx4_dev *dev, int qpn, u32 *qkey)
+{
+	int i;
+	for (i = 0; i < dev->caps.num_ports; i++) {
+		if (qpn == dev->caps.qp0_proxy[i] ||
+		    qpn == dev->caps.qp0_tunnel[i]) {
+			*qkey = dev->caps.qp0_qkey[i];
+			return 0;
+		}
+	}
+	return -EINVAL;
+}
+
 static int build_sriov_qp0_header(struct mlx4_ib_sqp *sqp,
 				  struct ib_send_wr *wr,
 				  void *wqe, unsigned *mlx_seg_len)
@@ -1995,8 +2027,13 @@ static int build_sriov_qp0_header(struct mlx4_ib_sqp *sqp,
 			cpu_to_be32(mdev->dev->caps.qp0_tunnel[sqp->qp.port - 1]);
 
 	sqp->ud_header.bth.psn = cpu_to_be32((sqp->send_psn++) & ((1 << 24) - 1));
-	if (mlx4_get_parav_qkey(mdev->dev, sqp->qp.mqp.qpn, &qkey))
-		return -EINVAL;
+	if (mlx4_is_master(mdev->dev)) {
+		if (mlx4_get_parav_qkey(mdev->dev, sqp->qp.mqp.qpn, &qkey))
+			return -EINVAL;
+	} else {
+		if (vf_get_qp0_qkey(mdev->dev, sqp->qp.mqp.qpn, &qkey))
+			return -EINVAL;
+	}
 	sqp->ud_header.deth.qkey = cpu_to_be32(qkey);
 	sqp->ud_header.deth.source_qpn = cpu_to_be32(sqp->qp.mqp.qpn);
 
@@ -2378,7 +2415,8 @@ static void set_datagram_seg(struct mlx4_wqe_datagram_seg *dseg,
 
 static void set_tunnel_datagram_seg(struct mlx4_ib_dev *dev,
 				    struct mlx4_wqe_datagram_seg *dseg,
-				    struct ib_send_wr *wr, enum ib_qp_type qpt)
+				    struct ib_send_wr *wr,
+				    enum mlx4_ib_qp_type qpt)
 {
 	union mlx4_ext_av *av = &to_mah(wr->wr.ud.ah)->av;
 	struct mlx4_av sqp_av = {0};
@@ -2391,8 +2429,10 @@ static void set_tunnel_datagram_seg(struct mlx4_ib_dev *dev,
 			cpu_to_be32(0xf0000000);
 
 	memcpy(dseg->av, &sqp_av, sizeof (struct mlx4_av));
-	/* This function used only for sending on QP1 proxies */
-	dseg->dqpn = cpu_to_be32(dev->dev->caps.qp1_tunnel[port - 1]);
+	if (qpt == MLX4_IB_QPT_PROXY_GSI)
+		dseg->dqpn = cpu_to_be32(dev->dev->caps.qp1_tunnel[port - 1]);
+	else
+		dseg->dqpn = cpu_to_be32(dev->dev->caps.qp0_tunnel[port - 1]);
 	/* Use QKEY from the QP context, which is set by master */
 	dseg->qkey = cpu_to_be32(IB_QP_SET_QKEY);
 }
@@ -2687,11 +2727,6 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			break;
 
 		case MLX4_IB_QPT_PROXY_SMI_OWNER:
-			if (unlikely(!mlx4_is_master(to_mdev(ibqp->device)->dev))) {
-				err = -ENOSYS;
-				*bad_wr = wr;
-				goto out;
-			}
 			err = build_sriov_qp0_header(to_msqp(qp), wr, ctrl, &seglen);
 			if (unlikely(err)) {
 				*bad_wr = wr;
@@ -2708,16 +2743,13 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			size += seglen / 16;
 			break;
 		case MLX4_IB_QPT_PROXY_SMI:
-			/* don't allow QP0 sends on guests */
-			err = -ENOSYS;
-			*bad_wr = wr;
-			goto out;
 		case MLX4_IB_QPT_PROXY_GSI:
 			/* If we are tunneling special qps, this is a UD qp.
 			 * In this case we first add a UD segment targeting
 			 * the tunnel qp, and then add a header with address
 			 * information */
-			set_tunnel_datagram_seg(to_mdev(ibqp->device), wqe, wr, ibqp->qp_type);
+			set_tunnel_datagram_seg(to_mdev(ibqp->device), wqe, wr,
+						qp->mlx4_ib_qp_type);
 			wqe  += sizeof (struct mlx4_wqe_datagram_seg);
 			size += sizeof (struct mlx4_wqe_datagram_seg) / 16;
 			build_tunnel_header(wr, wqe, &seglen);
diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c
index 60c5fb025fc7..62d9285300af 100644
--- a/drivers/infiniband/hw/mlx4/srq.c
+++ b/drivers/infiniband/hw/mlx4/srq.c
@@ -134,13 +134,14 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd,
 		if (err)
 			goto err_mtt;
 	} else {
-		err = mlx4_db_alloc(dev->dev, &srq->db, 0);
+		err = mlx4_db_alloc(dev->dev, &srq->db, 0, GFP_KERNEL);
 		if (err)
 			goto err_srq;
 
 		*srq->db.db = 0;
 
-		if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, &srq->buf)) {
+		if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, &srq->buf,
+				   GFP_KERNEL)) {
 			err = -ENOMEM;
 			goto err_db;
 		}
@@ -165,7 +166,7 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd,
 		if (err)
 			goto err_buf;
 
-		err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf);
+		err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf, GFP_KERNEL);
 		if (err)
 			goto err_mtt;
 
diff --git a/drivers/infiniband/hw/mlx4/sysfs.c b/drivers/infiniband/hw/mlx4/sysfs.c
index 5a38e43eca65..cb4c66e723b5 100644
--- a/drivers/infiniband/hw/mlx4/sysfs.c
+++ b/drivers/infiniband/hw/mlx4/sysfs.c
@@ -389,8 +389,10 @@ struct mlx4_port {
 	struct mlx4_ib_dev    *dev;
 	struct attribute_group pkey_group;
 	struct attribute_group gid_group;
-	u8                     port_num;
+	struct device_attribute	enable_smi_admin;
+	struct device_attribute	smi_enabled;
 	int		       slave;
+	u8                     port_num;
 };
 
 
@@ -558,6 +560,101 @@ err:
 	return NULL;
 }
 
+static ssize_t sysfs_show_smi_enabled(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	struct mlx4_port *p =
+		container_of(attr, struct mlx4_port, smi_enabled);
+	ssize_t len = 0;
+
+	if (mlx4_vf_smi_enabled(p->dev->dev, p->slave, p->port_num))
+		len = sprintf(buf, "%d\n", 1);
+	else
+		len = sprintf(buf, "%d\n", 0);
+
+	return len;
+}
+
+static ssize_t sysfs_show_enable_smi_admin(struct device *dev,
+					   struct device_attribute *attr,
+					   char *buf)
+{
+	struct mlx4_port *p =
+		container_of(attr, struct mlx4_port, enable_smi_admin);
+	ssize_t len = 0;
+
+	if (mlx4_vf_get_enable_smi_admin(p->dev->dev, p->slave, p->port_num))
+		len = sprintf(buf, "%d\n", 1);
+	else
+		len = sprintf(buf, "%d\n", 0);
+
+	return len;
+}
+
+static ssize_t sysfs_store_enable_smi_admin(struct device *dev,
+					    struct device_attribute *attr,
+					    const char *buf, size_t count)
+{
+	struct mlx4_port *p =
+		container_of(attr, struct mlx4_port, enable_smi_admin);
+	int enable;
+
+	if (sscanf(buf, "%i", &enable) != 1 ||
+	    enable < 0 || enable > 1)
+		return -EINVAL;
+
+	if (mlx4_vf_set_enable_smi_admin(p->dev->dev, p->slave, p->port_num, enable))
+		return -EINVAL;
+	return count;
+}
+
+static int add_vf_smi_entries(struct mlx4_port *p)
+{
+	int is_eth = rdma_port_get_link_layer(&p->dev->ib_dev, p->port_num) ==
+			IB_LINK_LAYER_ETHERNET;
+	int ret;
+
+	/* do not display entries if eth transport, or if master */
+	if (is_eth || p->slave == mlx4_master_func_num(p->dev->dev))
+		return 0;
+
+	sysfs_attr_init(&p->smi_enabled.attr);
+	p->smi_enabled.show = sysfs_show_smi_enabled;
+	p->smi_enabled.store = NULL;
+	p->smi_enabled.attr.name = "smi_enabled";
+	p->smi_enabled.attr.mode = 0444;
+	ret = sysfs_create_file(&p->kobj, &p->smi_enabled.attr);
+	if (ret) {
+		pr_err("failed to create smi_enabled\n");
+		return ret;
+	}
+
+	sysfs_attr_init(&p->enable_smi_admin.attr);
+	p->enable_smi_admin.show = sysfs_show_enable_smi_admin;
+	p->enable_smi_admin.store = sysfs_store_enable_smi_admin;
+	p->enable_smi_admin.attr.name = "enable_smi_admin";
+	p->enable_smi_admin.attr.mode = 0644;
+	ret = sysfs_create_file(&p->kobj, &p->enable_smi_admin.attr);
+	if (ret) {
+		pr_err("failed to create enable_smi_admin\n");
+		sysfs_remove_file(&p->kobj, &p->smi_enabled.attr);
+		return ret;
+	}
+	return 0;
+}
+
+static void remove_vf_smi_entries(struct mlx4_port *p)
+{
+	int is_eth = rdma_port_get_link_layer(&p->dev->ib_dev, p->port_num) ==
+			IB_LINK_LAYER_ETHERNET;
+
+	if (is_eth || p->slave == mlx4_master_func_num(p->dev->dev))
+		return;
+
+	sysfs_remove_file(&p->kobj, &p->smi_enabled.attr);
+	sysfs_remove_file(&p->kobj, &p->enable_smi_admin.attr);
+}
+
 static int add_port(struct mlx4_ib_dev *dev, int port_num, int slave)
 {
 	struct mlx4_port *p;
@@ -602,6 +699,10 @@ static int add_port(struct mlx4_ib_dev *dev, int port_num, int slave)
 	if (ret)
 		goto err_free_gid;
 
+	ret = add_vf_smi_entries(p);
+	if (ret)
+		goto err_free_gid;
+
 	list_add_tail(&p->kobj.entry, &dev->pkeys.pkey_port_list[slave]);
 	return 0;
 
@@ -669,6 +770,7 @@ err_add:
 		mport = container_of(p, struct mlx4_port, kobj);
 		sysfs_remove_group(p, &mport->pkey_group);
 		sysfs_remove_group(p, &mport->gid_group);
+		remove_vf_smi_entries(mport);
 		kobject_put(p);
 	}
 	kobject_put(dev->dev_ports_parent[slave]);
@@ -713,6 +815,7 @@ static void unregister_pkey_tree(struct mlx4_ib_dev *device)
 			port = container_of(p, struct mlx4_port, kobj);
 			sysfs_remove_group(p, &port->pkey_group);
 			sysfs_remove_group(p, &port->gid_group);
+			remove_vf_smi_entries(port);
 			kobject_put(p);
 			kobject_put(device->dev_ports_parent[slave]);
 		}
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 62bb6b49dc1d..8ae4f896cb41 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -32,6 +32,7 @@
 
 #include <linux/kref.h>
 #include <rdma/ib_umem.h>
+#include <rdma/ib_user_verbs.h>
 #include "mlx5_ib.h"
 #include "user.h"
 
@@ -602,14 +603,24 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 			  int *cqe_size, int *index, int *inlen)
 {
 	struct mlx5_ib_create_cq ucmd;
+	size_t ucmdlen;
 	int page_shift;
 	int npages;
 	int ncont;
 	int err;
 
-	if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd)))
+	ucmdlen =
+		(udata->inlen - sizeof(struct ib_uverbs_cmd_hdr) <
+		 sizeof(ucmd)) ? (sizeof(ucmd) -
+				  sizeof(ucmd.reserved)) : sizeof(ucmd);
+
+	if (ib_copy_from_udata(&ucmd, udata, ucmdlen))
 		return -EFAULT;
 
+	if (ucmdlen == sizeof(ucmd) &&
+	    ucmd.reserved != 0)
+		return -EINVAL;
+
 	if (ucmd.cqe_size != 64 && ucmd.cqe_size != 128)
 		return -EINVAL;
 
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 50541586e0a6..f2ccf1a5a291 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -264,8 +264,6 @@ struct mlx5_ib_mr {
 	__be64			*pas;
 	dma_addr_t		dma;
 	int			npages;
-	struct completion	done;
-	enum ib_wc_status	status;
 	struct mlx5_ib_dev     *dev;
 	struct mlx5_create_mkey_mbox_out out;
 	struct mlx5_core_sig_ctx    *sig;
@@ -277,6 +275,17 @@ struct mlx5_ib_fast_reg_page_list {
 	dma_addr_t			map;
 };
 
+struct mlx5_ib_umr_context {
+	enum ib_wc_status	status;
+	struct completion	done;
+};
+
+static inline void mlx5_ib_init_umr_context(struct mlx5_ib_umr_context *context)
+{
+	context->status = -1;
+	init_completion(&context->done);
+}
+
 struct umr_common {
 	struct ib_pd	*pd;
 	struct ib_cq	*cq;
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 81392b26d078..afa873bd028e 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -73,6 +73,8 @@ static void reg_mr_callback(int status, void *context)
 	struct mlx5_cache_ent *ent = &cache->ent[c];
 	u8 key;
 	unsigned long flags;
+	struct mlx5_mr_table *table = &dev->mdev.priv.mr_table;
+	int err;
 
 	spin_lock_irqsave(&ent->lock, flags);
 	ent->pending--;
@@ -107,6 +109,13 @@ static void reg_mr_callback(int status, void *context)
 	ent->cur++;
 	ent->size++;
 	spin_unlock_irqrestore(&ent->lock, flags);
+
+	write_lock_irqsave(&table->lock, flags);
+	err = radix_tree_insert(&table->tree, mlx5_base_mkey(mr->mmr.key),
+				&mr->mmr);
+	if (err)
+		pr_err("Error inserting to mr tree. 0x%x\n", -err);
+	write_unlock_irqrestore(&table->lock, flags);
 }
 
 static int add_keys(struct mlx5_ib_dev *dev, int c, int num)
@@ -699,7 +708,7 @@ static void prep_umr_unreg_wqe(struct mlx5_ib_dev *dev,
 
 void mlx5_umr_cq_handler(struct ib_cq *cq, void *cq_context)
 {
-	struct mlx5_ib_mr *mr;
+	struct mlx5_ib_umr_context *context;
 	struct ib_wc wc;
 	int err;
 
@@ -712,9 +721,9 @@ void mlx5_umr_cq_handler(struct ib_cq *cq, void *cq_context)
 		if (err == 0)
 			break;
 
-		mr = (struct mlx5_ib_mr *)(unsigned long)wc.wr_id;
-		mr->status = wc.status;
-		complete(&mr->done);
+		context = (struct mlx5_ib_umr_context *) (unsigned long) wc.wr_id;
+		context->status = wc.status;
+		complete(&context->done);
 	}
 	ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
 }
@@ -726,11 +735,12 @@ static struct mlx5_ib_mr *reg_umr(struct ib_pd *pd, struct ib_umem *umem,
 	struct mlx5_ib_dev *dev = to_mdev(pd->device);
 	struct device *ddev = dev->ib_dev.dma_device;
 	struct umr_common *umrc = &dev->umrc;
+	struct mlx5_ib_umr_context umr_context;
 	struct ib_send_wr wr, *bad;
 	struct mlx5_ib_mr *mr;
 	struct ib_sge sg;
 	int size = sizeof(u64) * npages;
-	int err;
+	int err = 0;
 	int i;
 
 	for (i = 0; i < 1; i++) {
@@ -751,7 +761,7 @@ static struct mlx5_ib_mr *reg_umr(struct ib_pd *pd, struct ib_umem *umem,
 	mr->pas = kmalloc(size + MLX5_UMR_ALIGN - 1, GFP_KERNEL);
 	if (!mr->pas) {
 		err = -ENOMEM;
-		goto error;
+		goto free_mr;
 	}
 
 	mlx5_ib_populate_pas(dev, umem, page_shift,
@@ -760,44 +770,46 @@ static struct mlx5_ib_mr *reg_umr(struct ib_pd *pd, struct ib_umem *umem,
 	mr->dma = dma_map_single(ddev, mr_align(mr->pas, MLX5_UMR_ALIGN), size,
 				 DMA_TO_DEVICE);
 	if (dma_mapping_error(ddev, mr->dma)) {
-		kfree(mr->pas);
 		err = -ENOMEM;
-		goto error;
+		goto free_pas;
 	}
 
 	memset(&wr, 0, sizeof(wr));
-	wr.wr_id = (u64)(unsigned long)mr;
+	wr.wr_id = (u64)(unsigned long)&umr_context;
 	prep_umr_reg_wqe(pd, &wr, &sg, mr->dma, npages, mr->mmr.key, page_shift, virt_addr, len, access_flags);
 
-	/* We serialize polls so one process does not kidnap another's
-	 * completion. This is not a problem since wr is completed in
-	 * around 1 usec
-	 */
+	mlx5_ib_init_umr_context(&umr_context);
 	down(&umrc->sem);
-	init_completion(&mr->done);
 	err = ib_post_send(umrc->qp, &wr, &bad);
 	if (err) {
 		mlx5_ib_warn(dev, "post send failed, err %d\n", err);
-		up(&umrc->sem);
-		goto error;
+		goto unmap_dma;
+	} else {
+		wait_for_completion(&umr_context.done);
+		if (umr_context.status != IB_WC_SUCCESS) {
+			mlx5_ib_warn(dev, "reg umr failed\n");
+			err = -EFAULT;
+		}
 	}
-	wait_for_completion(&mr->done);
-	up(&umrc->sem);
 
+	mr->mmr.iova = virt_addr;
+	mr->mmr.size = len;
+	mr->mmr.pd = to_mpd(pd)->pdn;
+
+unmap_dma:
+	up(&umrc->sem);
 	dma_unmap_single(ddev, mr->dma, size, DMA_TO_DEVICE);
+
+free_pas:
 	kfree(mr->pas);
 
-	if (mr->status != IB_WC_SUCCESS) {
-		mlx5_ib_warn(dev, "reg umr failed\n");
-		err = -EFAULT;
-		goto error;
+free_mr:
+	if (err) {
+		free_cached_mr(dev, mr);
+		return ERR_PTR(err);
 	}
 
 	return mr;
-
-error:
-	free_cached_mr(dev, mr);
-	return ERR_PTR(err);
 }
 
 static struct mlx5_ib_mr *reg_create(struct ib_pd *pd, u64 virt_addr,
@@ -926,24 +938,26 @@ error:
 static int unreg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
 {
 	struct umr_common *umrc = &dev->umrc;
+	struct mlx5_ib_umr_context umr_context;
 	struct ib_send_wr wr, *bad;
 	int err;
 
 	memset(&wr, 0, sizeof(wr));
-	wr.wr_id = (u64)(unsigned long)mr;
+	wr.wr_id = (u64)(unsigned long)&umr_context;
 	prep_umr_unreg_wqe(dev, &wr, mr->mmr.key);
 
+	mlx5_ib_init_umr_context(&umr_context);
 	down(&umrc->sem);
-	init_completion(&mr->done);
 	err = ib_post_send(umrc->qp, &wr, &bad);
 	if (err) {
 		up(&umrc->sem);
 		mlx5_ib_dbg(dev, "err %d\n", err);
 		goto error;
+	} else {
+		wait_for_completion(&umr_context.done);
+		up(&umrc->sem);
 	}
-	wait_for_completion(&mr->done);
-	up(&umrc->sem);
-	if (mr->status != IB_WC_SUCCESS) {
+	if (umr_context.status != IB_WC_SUCCESS) {
 		mlx5_ib_warn(dev, "unreg umr failed\n");
 		err = -EFAULT;
 		goto error;
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index dc930ed21eca..d13ddf1c0033 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -574,6 +574,10 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	uar_index = uuarn_to_uar_index(&context->uuari, uuarn);
 	mlx5_ib_dbg(dev, "uuarn 0x%x, uar_index 0x%x\n", uuarn, uar_index);
 
+	qp->rq.offset = 0;
+	qp->sq.wqe_shift = ilog2(MLX5_SEND_WQE_BB);
+	qp->sq.offset = qp->rq.wqe_cnt << qp->rq.wqe_shift;
+
 	err = set_user_buf_size(dev, qp, &ucmd);
 	if (err)
 		goto err_uuar;
@@ -2078,6 +2082,7 @@ static int mlx5_set_bsf(struct ib_mr *sig_mr,
 	struct ib_sig_domain *wire = &sig_attrs->wire;
 	int ret, selector;
 
+	memset(bsf, 0, sizeof(*bsf));
 	switch (sig_attrs->mem.sig_type) {
 	case IB_SIG_TYPE_T10_DIF:
 		if (sig_attrs->wire.sig_type != IB_SIG_TYPE_T10_DIF)
@@ -2090,9 +2095,11 @@ static int mlx5_set_bsf(struct ib_mr *sig_mr,
 			/* Same block structure */
 			basic->bsf_size_sbs = 1 << 4;
 			if (mem->sig.dif.bg_type == wire->sig.dif.bg_type)
-				basic->wire.copy_byte_mask = 0xff;
-			else
-				basic->wire.copy_byte_mask = 0x3f;
+				basic->wire.copy_byte_mask |= 0xc0;
+			if (mem->sig.dif.app_tag == wire->sig.dif.app_tag)
+				basic->wire.copy_byte_mask |= 0x30;
+			if (mem->sig.dif.ref_tag == wire->sig.dif.ref_tag)
+				basic->wire.copy_byte_mask |= 0x0f;
 		} else
 			basic->wire.bs_selector = bs_selector(wire->sig.dif.pi_interval);
 
@@ -2131,9 +2138,13 @@ static int set_sig_data_segment(struct ib_send_wr *wr, struct mlx5_ib_qp *qp,
 	int ret;
 	int wqe_size;
 
-	if (!wr->wr.sig_handover.prot) {
+	if (!wr->wr.sig_handover.prot ||
+	    (data_key == wr->wr.sig_handover.prot->lkey &&
+	     data_va == wr->wr.sig_handover.prot->addr &&
+	     data_len == wr->wr.sig_handover.prot->length)) {
 		/**
 		 * Source domain doesn't contain signature information
+		 * or data and protection are interleaved in memory.
 		 * So need construct:
 		 *                  ------------------
 		 *                 |     data_klm     |
@@ -2187,23 +2198,13 @@ static int set_sig_data_segment(struct ib_send_wr *wr, struct mlx5_ib_qp *qp,
 		data_sentry->bcount = cpu_to_be16(block_size);
 		data_sentry->key = cpu_to_be32(data_key);
 		data_sentry->va = cpu_to_be64(data_va);
+		data_sentry->stride = cpu_to_be16(block_size);
+
 		prot_sentry->bcount = cpu_to_be16(prot_size);
 		prot_sentry->key = cpu_to_be32(prot_key);
+		prot_sentry->va = cpu_to_be64(prot_va);
+		prot_sentry->stride = cpu_to_be16(prot_size);
 
-		if (prot_key == data_key && prot_va == data_va) {
-			/**
-			 * The data and protection are interleaved
-			 * in a single memory region
-			 **/
-			prot_sentry->va = cpu_to_be64(data_va + block_size);
-			prot_sentry->stride = cpu_to_be16(block_size + prot_size);
-			data_sentry->stride = prot_sentry->stride;
-		} else {
-			/* The data and protection are two different buffers */
-			prot_sentry->va = cpu_to_be64(prot_va);
-			data_sentry->stride = cpu_to_be16(block_size);
-			prot_sentry->stride = cpu_to_be16(prot_size);
-		}
 		wqe_size = ALIGN(sizeof(*sblock_ctrl) + sizeof(*data_sentry) +
 				 sizeof(*prot_sentry), 64);
 	}
@@ -2275,7 +2276,10 @@ static int set_sig_umr_wr(struct ib_send_wr *wr, struct mlx5_ib_qp *qp,
 
 	/* length of the protected region, data + protection */
 	region_len = wr->sg_list->length;
-	if (wr->wr.sig_handover.prot)
+	if (wr->wr.sig_handover.prot &&
+	    (wr->wr.sig_handover.prot->lkey != wr->sg_list->lkey  ||
+	     wr->wr.sig_handover.prot->addr != wr->sg_list->addr  ||
+	     wr->wr.sig_handover.prot->length != wr->sg_list->length))
 		region_len += wr->wr.sig_handover.prot->length;
 
 	/**
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index 210b3eaf188a..384af6dec5eb 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -35,6 +35,7 @@
 #include <linux/mlx5/srq.h>
 #include <linux/slab.h>
 #include <rdma/ib_umem.h>
+#include <rdma/ib_user_verbs.h>
 
 #include "mlx5_ib.h"
 #include "user.h"
@@ -78,16 +79,27 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 {
 	struct mlx5_ib_dev *dev = to_mdev(pd->device);
 	struct mlx5_ib_create_srq ucmd;
+	size_t ucmdlen;
 	int err;
 	int npages;
 	int page_shift;
 	int ncont;
 	u32 offset;
 
-	if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd))) {
+	ucmdlen =
+		(udata->inlen - sizeof(struct ib_uverbs_cmd_hdr) <
+		 sizeof(ucmd)) ? (sizeof(ucmd) -
+				  sizeof(ucmd.reserved)) : sizeof(ucmd);
+
+	if (ib_copy_from_udata(&ucmd, udata, ucmdlen)) {
 		mlx5_ib_dbg(dev, "failed copy udata\n");
 		return -EFAULT;
 	}
+
+	if (ucmdlen == sizeof(ucmd) &&
+	    ucmd.reserved != 0)
+		return -EINVAL;
+
 	srq->wq_sig = !!(ucmd.flags & MLX5_SRQ_FLAG_SIGNATURE);
 
 	srq->umem = ib_umem_get(pd->uobject->context, ucmd.buf_addr, buf_size,
diff --git a/drivers/infiniband/hw/mlx5/user.h b/drivers/infiniband/hw/mlx5/user.h
index 0f4f8e42a17f..d0ba264ac1ed 100644
--- a/drivers/infiniband/hw/mlx5/user.h
+++ b/drivers/infiniband/hw/mlx5/user.h
@@ -91,6 +91,7 @@ struct mlx5_ib_create_cq {
 	__u64	buf_addr;
 	__u64	db_addr;
 	__u32	cqe_size;
+	__u32	reserved; /* explicit padding (optional on i386) */
 };
 
 struct mlx5_ib_create_cq_resp {
@@ -109,6 +110,7 @@ struct mlx5_ib_create_srq {
 	__u64	buf_addr;
 	__u64	db_addr;
 	__u32	flags;
+	__u32	reserved; /* explicit padding (optional on i386) */
 };
 
 struct mlx5_ib_create_srq_resp {
diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c
index 353c7b05a90a..3b2a6dc8ea99 100644
--- a/drivers/infiniband/hw/nes/nes.c
+++ b/drivers/infiniband/hw/nes/nes.c
@@ -68,7 +68,6 @@ MODULE_VERSION(DRV_VERSION);
 int max_mtu = 9000;
 int interrupt_mod_interval = 0;
 
-
 /* Interoperability */
 int mpa_version = 1;
 module_param(mpa_version, int, 0644);
@@ -112,6 +111,16 @@ static struct pci_device_id nes_pci_table[] = {
 
 MODULE_DEVICE_TABLE(pci, nes_pci_table);
 
+/* registered nes netlink callbacks */
+static struct ibnl_client_cbs nes_nl_cb_table[] = {
+	[RDMA_NL_IWPM_REG_PID] = {.dump = iwpm_register_pid_cb},
+	[RDMA_NL_IWPM_ADD_MAPPING] = {.dump = iwpm_add_mapping_cb},
+	[RDMA_NL_IWPM_QUERY_MAPPING] = {.dump = iwpm_add_and_query_mapping_cb},
+	[RDMA_NL_IWPM_HANDLE_ERR] = {.dump = iwpm_mapping_error_cb},
+	[RDMA_NL_IWPM_MAPINFO] = {.dump = iwpm_mapping_info_cb},
+	[RDMA_NL_IWPM_MAPINFO_NUM] = {.dump = iwpm_ack_mapping_info_cb}
+};
+
 static int nes_inetaddr_event(struct notifier_block *, unsigned long, void *);
 static int nes_net_event(struct notifier_block *, unsigned long, void *);
 static int nes_notifiers_registered;
@@ -672,6 +681,17 @@ static int nes_probe(struct pci_dev *pcidev, const struct pci_device_id *ent)
 	}
 	nes_notifiers_registered++;
 
+	if (ibnl_add_client(RDMA_NL_NES, RDMA_NL_IWPM_NUM_OPS, nes_nl_cb_table))
+		printk(KERN_ERR PFX "%s[%u]: Failed to add netlink callback\n",
+			__func__, __LINE__);
+
+	ret = iwpm_init(RDMA_NL_NES);
+	if (ret) {
+		printk(KERN_ERR PFX "%s: port mapper initialization failed\n",
+				pci_name(pcidev));
+		goto bail7;
+	}
+
 	INIT_DELAYED_WORK(&nesdev->work, nes_recheck_link_status);
 
 	/* Initialize network devices */
@@ -710,6 +730,7 @@ static int nes_probe(struct pci_dev *pcidev, const struct pci_device_id *ent)
 
 	nes_debug(NES_DBG_INIT, "netdev_count=%d, nesadapter->netdev_count=%d\n",
 			nesdev->netdev_count, nesdev->nesadapter->netdev_count);
+	ibnl_remove_client(RDMA_NL_NES);
 
 	nes_notifiers_registered--;
 	if (nes_notifiers_registered == 0) {
@@ -773,6 +794,8 @@ static void nes_remove(struct pci_dev *pcidev)
 				nesdev->nesadapter->netdev_count--;
 			}
 		}
+	ibnl_remove_client(RDMA_NL_NES);
+	iwpm_exit(RDMA_NL_NES);
 
 	nes_notifiers_registered--;
 	if (nes_notifiers_registered == 0) {
diff --git a/drivers/infiniband/hw/nes/nes.h b/drivers/infiniband/hw/nes/nes.h
index 33cc58941a3e..bd9d132f11c7 100644
--- a/drivers/infiniband/hw/nes/nes.h
+++ b/drivers/infiniband/hw/nes/nes.h
@@ -51,6 +51,8 @@
 #include <rdma/ib_pack.h>
 #include <rdma/rdma_cm.h>
 #include <rdma/iw_cm.h>
+#include <rdma/rdma_netlink.h>
+#include <rdma/iw_portmap.h>
 
 #define NES_SEND_FIRST_WRITE
 
@@ -130,6 +132,7 @@
 #define NES_DBG_IW_TX       0x00040000
 #define NES_DBG_SHUTDOWN    0x00080000
 #define NES_DBG_PAU         0x00100000
+#define NES_DBG_NLMSG       0x00200000
 #define NES_DBG_RSVD1       0x10000000
 #define NES_DBG_RSVD2       0x20000000
 #define NES_DBG_RSVD3       0x40000000
diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c
index dfa9df484505..6f09a72e78d7 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2006 - 2011 Intel Corporation.  All rights reserved.
+ * Copyright (c) 2006 - 2014 Intel Corporation.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -59,6 +59,7 @@
 #include <net/route.h>
 #include <net/ip_fib.h>
 #include <net/tcp.h>
+#include <linux/fcntl.h>
 
 #include "nes.h"
 
@@ -166,7 +167,6 @@ int nes_rem_ref_cm_node(struct nes_cm_node *cm_node)
 {
 	return rem_ref_cm_node(cm_node->cm_core, cm_node);
 }
-
 /**
  * create_event
  */
@@ -482,11 +482,11 @@ static void form_cm_frame(struct sk_buff *skb,
 	iph->ttl = 0x40;
 	iph->protocol = 0x06;   /* IPPROTO_TCP */
 
-	iph->saddr = htonl(cm_node->loc_addr);
-	iph->daddr = htonl(cm_node->rem_addr);
+	iph->saddr = htonl(cm_node->mapped_loc_addr);
+	iph->daddr = htonl(cm_node->mapped_rem_addr);
 
-	tcph->source = htons(cm_node->loc_port);
-	tcph->dest = htons(cm_node->rem_port);
+	tcph->source = htons(cm_node->mapped_loc_port);
+	tcph->dest = htons(cm_node->mapped_rem_port);
 	tcph->seq = htonl(cm_node->tcp_cntxt.loc_seq_num);
 
 	if (flags & SET_ACK) {
@@ -525,6 +525,100 @@ static void form_cm_frame(struct sk_buff *skb,
 	cm_packets_created++;
 }
 
+/*
+ * nes_create_sockaddr - Record ip addr and tcp port in a sockaddr struct
+ */
+static void nes_create_sockaddr(__be32 ip_addr, __be16 port,
+				struct sockaddr_storage *addr)
+{
+	struct sockaddr_in *nes_sockaddr = (struct sockaddr_in *)addr;
+	nes_sockaddr->sin_family = AF_INET;
+	memcpy(&nes_sockaddr->sin_addr.s_addr, &ip_addr, sizeof(__be32));
+	nes_sockaddr->sin_port = port;
+}
+
+/*
+ * nes_create_mapinfo - Create a mapinfo object in the port mapper data base
+ */
+static int nes_create_mapinfo(struct nes_cm_info *cm_info)
+{
+	struct sockaddr_storage local_sockaddr;
+	struct sockaddr_storage mapped_sockaddr;
+
+	nes_create_sockaddr(htonl(cm_info->loc_addr), htons(cm_info->loc_port),
+				&local_sockaddr);
+	nes_create_sockaddr(htonl(cm_info->mapped_loc_addr),
+			htons(cm_info->mapped_loc_port), &mapped_sockaddr);
+
+	return iwpm_create_mapinfo(&local_sockaddr,
+				&mapped_sockaddr, RDMA_NL_NES);
+}
+
+/*
+ * nes_remove_mapinfo - Remove a mapinfo object from the port mapper data base
+ *                      and send a remove mapping op message to
+ *                      the userspace port mapper
+ */
+static int nes_remove_mapinfo(u32 loc_addr, u16 loc_port,
+			u32 mapped_loc_addr, u16 mapped_loc_port)
+{
+	struct sockaddr_storage local_sockaddr;
+	struct sockaddr_storage mapped_sockaddr;
+
+	nes_create_sockaddr(htonl(loc_addr), htons(loc_port), &local_sockaddr);
+	nes_create_sockaddr(htonl(mapped_loc_addr), htons(mapped_loc_port),
+				&mapped_sockaddr);
+
+	iwpm_remove_mapinfo(&local_sockaddr, &mapped_sockaddr);
+	return iwpm_remove_mapping(&local_sockaddr, RDMA_NL_NES);
+}
+
+/*
+ * nes_form_pm_msg - Form a port mapper message with mapping info
+ */
+static void nes_form_pm_msg(struct nes_cm_info *cm_info,
+				struct iwpm_sa_data *pm_msg)
+{
+	nes_create_sockaddr(htonl(cm_info->loc_addr), htons(cm_info->loc_port),
+				&pm_msg->loc_addr);
+	nes_create_sockaddr(htonl(cm_info->rem_addr), htons(cm_info->rem_port),
+				&pm_msg->rem_addr);
+}
+
+/*
+ * nes_form_reg_msg - Form a port mapper message with dev info
+ */
+static void nes_form_reg_msg(struct nes_vnic *nesvnic,
+			struct iwpm_dev_data *pm_msg)
+{
+	memcpy(pm_msg->dev_name, nesvnic->nesibdev->ibdev.name,
+				IWPM_DEVNAME_SIZE);
+	memcpy(pm_msg->if_name, nesvnic->netdev->name, IWPM_IFNAME_SIZE);
+}
+
+/*
+ * nes_record_pm_msg - Save the received mapping info
+ */
+static void nes_record_pm_msg(struct nes_cm_info *cm_info,
+			struct iwpm_sa_data *pm_msg)
+{
+	struct sockaddr_in *mapped_loc_addr =
+			(struct sockaddr_in *)&pm_msg->mapped_loc_addr;
+	struct sockaddr_in *mapped_rem_addr =
+			(struct sockaddr_in *)&pm_msg->mapped_rem_addr;
+
+	if (mapped_loc_addr->sin_family == AF_INET) {
+		cm_info->mapped_loc_addr =
+			ntohl(mapped_loc_addr->sin_addr.s_addr);
+		cm_info->mapped_loc_port = ntohs(mapped_loc_addr->sin_port);
+	}
+	if (mapped_rem_addr->sin_family == AF_INET) {
+		cm_info->mapped_rem_addr =
+			ntohl(mapped_rem_addr->sin_addr.s_addr);
+		cm_info->mapped_rem_port = ntohs(mapped_rem_addr->sin_port);
+	}
+}
+
 /**
  * print_core - dump a cm core
  */
@@ -1147,8 +1241,11 @@ static struct nes_cm_node *find_node(struct nes_cm_core *cm_core,
 			  loc_addr, loc_port,
 			  cm_node->rem_addr, cm_node->rem_port,
 			  rem_addr, rem_port);
-		if ((cm_node->loc_addr == loc_addr) && (cm_node->loc_port == loc_port) &&
-		    (cm_node->rem_addr == rem_addr) && (cm_node->rem_port == rem_port)) {
+		if ((cm_node->mapped_loc_addr == loc_addr) &&
+			(cm_node->mapped_loc_port == loc_port) &&
+			(cm_node->mapped_rem_addr == rem_addr) &&
+			(cm_node->mapped_rem_port == rem_port)) {
+
 			add_ref_cm_node(cm_node);
 			spin_unlock_irqrestore(&cm_core->ht_lock, flags);
 			return cm_node;
@@ -1165,18 +1262,28 @@ static struct nes_cm_node *find_node(struct nes_cm_core *cm_core,
  * find_listener - find a cm node listening on this addr-port pair
  */
 static struct nes_cm_listener *find_listener(struct nes_cm_core *cm_core,
-					     nes_addr_t dst_addr, u16 dst_port, enum nes_cm_listener_state listener_state)
+					nes_addr_t dst_addr, u16 dst_port,
+					enum nes_cm_listener_state listener_state, int local)
 {
 	unsigned long flags;
 	struct nes_cm_listener *listen_node;
+	nes_addr_t listen_addr;
+	u16 listen_port;
 
 	/* walk list and find cm_node associated with this session ID */
 	spin_lock_irqsave(&cm_core->listen_list_lock, flags);
 	list_for_each_entry(listen_node, &cm_core->listen_list.list, list) {
+		if (local) {
+			listen_addr = listen_node->loc_addr;
+			listen_port = listen_node->loc_port;
+		} else {
+			listen_addr = listen_node->mapped_loc_addr;
+			listen_port = listen_node->mapped_loc_port;
+		}
 		/* compare node pair, return node handle if a match */
-		if (((listen_node->loc_addr == dst_addr) ||
-		     listen_node->loc_addr == 0x00000000) &&
-		    (listen_node->loc_port == dst_port) &&
+		if (((listen_addr == dst_addr) ||
+		     listen_addr == 0x00000000) &&
+		    (listen_port == dst_port) &&
 		    (listener_state & listen_node->listener_state)) {
 			atomic_inc(&listen_node->ref_count);
 			spin_unlock_irqrestore(&cm_core->listen_list_lock, flags);
@@ -1189,7 +1296,6 @@ static struct nes_cm_listener *find_listener(struct nes_cm_core *cm_core,
 	return NULL;
 }
 
-
 /**
  * add_hte_node - add a cm node to the hash table
  */
@@ -1310,9 +1416,20 @@ static int mini_cm_dec_refcnt_listen(struct nes_cm_core *cm_core,
 
 		spin_unlock_irqrestore(&cm_core->listen_list_lock, flags);
 
-		if (listener->nesvnic)
-			nes_manage_apbvt(listener->nesvnic, listener->loc_port,
-					 PCI_FUNC(listener->nesvnic->nesdev->pcidev->devfn), NES_MANAGE_APBVT_DEL);
+		if (listener->nesvnic) {
+			nes_manage_apbvt(listener->nesvnic,
+				listener->mapped_loc_port,
+				PCI_FUNC(listener->nesvnic->nesdev->pcidev->devfn),
+				NES_MANAGE_APBVT_DEL);
+
+			nes_remove_mapinfo(listener->loc_addr,
+					listener->loc_port,
+					listener->mapped_loc_addr,
+					listener->mapped_loc_port);
+			nes_debug(NES_DBG_NLMSG,
+					"Delete APBVT mapped_loc_port = %04X\n",
+					listener->mapped_loc_port);
+		}
 
 		nes_debug(NES_DBG_CM, "destroying listener (%p)\n", listener);
 
@@ -1454,6 +1571,11 @@ static struct nes_cm_node *make_cm_node(struct nes_cm_core *cm_core,
 	cm_node->loc_port = cm_info->loc_port;
 	cm_node->rem_port = cm_info->rem_port;
 
+	cm_node->mapped_loc_addr = cm_info->mapped_loc_addr;
+	cm_node->mapped_rem_addr = cm_info->mapped_rem_addr;
+	cm_node->mapped_loc_port = cm_info->mapped_loc_port;
+	cm_node->mapped_rem_port = cm_info->mapped_rem_port;
+
 	cm_node->mpa_frame_rev = mpa_version;
 	cm_node->send_rdma0_op = SEND_RDMA_READ_ZERO;
 	cm_node->mpav2_ird_ord = 0;
@@ -1500,8 +1622,10 @@ static struct nes_cm_node *make_cm_node(struct nes_cm_core *cm_core,
 	cm_node->loopbackpartner = NULL;
 
 	/* get the mac addr for the remote node */
-	oldarpindex = nes_arp_table(nesdev, cm_node->rem_addr, NULL, NES_ARP_RESOLVE);
-	arpindex = nes_addr_resolve_neigh(nesvnic, cm_info->rem_addr, oldarpindex);
+	oldarpindex = nes_arp_table(nesdev, cm_node->mapped_rem_addr,
+				NULL, NES_ARP_RESOLVE);
+	arpindex = nes_addr_resolve_neigh(nesvnic,
+				cm_node->mapped_rem_addr, oldarpindex);
 	if (arpindex < 0) {
 		kfree(cm_node);
 		return NULL;
@@ -1563,11 +1687,14 @@ static int rem_ref_cm_node(struct nes_cm_core *cm_core,
 		mini_cm_dec_refcnt_listen(cm_core, cm_node->listener, 0);
 	} else {
 		if (cm_node->apbvt_set && cm_node->nesvnic) {
-			nes_manage_apbvt(cm_node->nesvnic, cm_node->loc_port,
-					 PCI_FUNC(
-						 cm_node->nesvnic->nesdev->pcidev->devfn),
+			nes_manage_apbvt(cm_node->nesvnic, cm_node->mapped_loc_port,
+					 PCI_FUNC(cm_node->nesvnic->nesdev->pcidev->devfn),
 					 NES_MANAGE_APBVT_DEL);
 		}
+		nes_debug(NES_DBG_NLMSG, "Delete APBVT mapped_loc_port = %04X\n",
+					cm_node->mapped_loc_port);
+		nes_remove_mapinfo(cm_node->loc_addr, cm_node->loc_port,
+			cm_node->mapped_loc_addr, cm_node->mapped_loc_port);
 	}
 
 	atomic_dec(&cm_core->node_cnt);
@@ -2235,17 +2362,21 @@ static void process_packet(struct nes_cm_node *cm_node, struct sk_buff *skb,
  * mini_cm_listen - create a listen node with params
  */
 static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core,
-					      struct nes_vnic *nesvnic, struct nes_cm_info *cm_info)
+			struct nes_vnic *nesvnic, struct nes_cm_info *cm_info)
 {
 	struct nes_cm_listener *listener;
+	struct iwpm_dev_data pm_reg_msg;
+	struct iwpm_sa_data pm_msg;
 	unsigned long flags;
+	int iwpm_err = 0;
 
 	nes_debug(NES_DBG_CM, "Search for 0x%08x : 0x%04x\n",
 		  cm_info->loc_addr, cm_info->loc_port);
 
 	/* cannot have multiple matching listeners */
-	listener = find_listener(cm_core, htonl(cm_info->loc_addr),
-				 htons(cm_info->loc_port), NES_CM_LISTENER_EITHER_STATE);
+	listener = find_listener(cm_core, cm_info->loc_addr, cm_info->loc_port,
+				NES_CM_LISTENER_EITHER_STATE, 1);
+
 	if (listener && listener->listener_state == NES_CM_LISTENER_ACTIVE_STATE) {
 		/* find automatically incs ref count ??? */
 		atomic_dec(&listener->ref_count);
@@ -2254,6 +2385,22 @@ static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core,
 	}
 
 	if (!listener) {
+		nes_form_reg_msg(nesvnic, &pm_reg_msg);
+		iwpm_err = iwpm_register_pid(&pm_reg_msg, RDMA_NL_NES);
+		if (iwpm_err) {
+			nes_debug(NES_DBG_NLMSG,
+			"Port Mapper reg pid fail (err = %d).\n", iwpm_err);
+		}
+		if (iwpm_valid_pid() && !iwpm_err) {
+			nes_form_pm_msg(cm_info, &pm_msg);
+			iwpm_err = iwpm_add_mapping(&pm_msg, RDMA_NL_NES);
+			if (iwpm_err)
+				nes_debug(NES_DBG_NLMSG,
+				"Port Mapper query fail (err = %d).\n", iwpm_err);
+			else
+				nes_record_pm_msg(cm_info, &pm_msg);
+		}
+
 		/* create a CM listen node (1/2 node to compare incoming traffic to) */
 		listener = kzalloc(sizeof(*listener), GFP_ATOMIC);
 		if (!listener) {
@@ -2261,8 +2408,10 @@ static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core,
 			return NULL;
 		}
 
-		listener->loc_addr = htonl(cm_info->loc_addr);
-		listener->loc_port = htons(cm_info->loc_port);
+		listener->loc_addr = cm_info->loc_addr;
+		listener->loc_port = cm_info->loc_port;
+		listener->mapped_loc_addr = cm_info->mapped_loc_addr;
+		listener->mapped_loc_port = cm_info->mapped_loc_port;
 		listener->reused_node = 0;
 
 		atomic_set(&listener->ref_count, 1);
@@ -2324,14 +2473,18 @@ static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core,
 
 	if (cm_info->loc_addr == cm_info->rem_addr) {
 		loopbackremotelistener = find_listener(cm_core,
-						       ntohl(nesvnic->local_ipaddr), cm_node->rem_port,
-						       NES_CM_LISTENER_ACTIVE_STATE);
+			cm_node->mapped_loc_addr, cm_node->mapped_rem_port,
+			NES_CM_LISTENER_ACTIVE_STATE, 0);
 		if (loopbackremotelistener == NULL) {
 			create_event(cm_node, NES_CM_EVENT_ABORTED);
 		} else {
 			loopback_cm_info = *cm_info;
 			loopback_cm_info.loc_port = cm_info->rem_port;
 			loopback_cm_info.rem_port = cm_info->loc_port;
+			loopback_cm_info.mapped_loc_port =
+				cm_info->mapped_rem_port;
+			loopback_cm_info.mapped_rem_port =
+				cm_info->mapped_loc_port;
 			loopback_cm_info.cm_id = loopbackremotelistener->cm_id;
 			loopbackremotenode = make_cm_node(cm_core, nesvnic,
 							  &loopback_cm_info, loopbackremotelistener);
@@ -2560,6 +2713,12 @@ static int mini_cm_recv_pkt(struct nes_cm_core *cm_core,
 	nfo.rem_addr = ntohl(iph->saddr);
 	nfo.rem_port = ntohs(tcph->source);
 
+	/* If port mapper is available these should be mapped address info */
+	nfo.mapped_loc_addr = ntohl(iph->daddr);
+	nfo.mapped_loc_port = ntohs(tcph->dest);
+	nfo.mapped_rem_addr = ntohl(iph->saddr);
+	nfo.mapped_rem_port = ntohs(tcph->source);
+
 	tmp_daddr = cpu_to_be32(iph->daddr);
 	tmp_saddr = cpu_to_be32(iph->saddr);
 
@@ -2568,8 +2727,8 @@ static int mini_cm_recv_pkt(struct nes_cm_core *cm_core,
 
 	do {
 		cm_node = find_node(cm_core,
-				    nfo.rem_port, nfo.rem_addr,
-				    nfo.loc_port, nfo.loc_addr);
+				    nfo.mapped_rem_port, nfo.mapped_rem_addr,
+				    nfo.mapped_loc_port, nfo.mapped_loc_addr);
 
 		if (!cm_node) {
 			/* Only type of packet accepted are for */
@@ -2578,9 +2737,9 @@ static int mini_cm_recv_pkt(struct nes_cm_core *cm_core,
 				skb_handled = 0;
 				break;
 			}
-			listener = find_listener(cm_core, nfo.loc_addr,
-						 nfo.loc_port,
-						 NES_CM_LISTENER_ACTIVE_STATE);
+			listener = find_listener(cm_core, nfo.mapped_loc_addr,
+					nfo.mapped_loc_port,
+					NES_CM_LISTENER_ACTIVE_STATE, 0);
 			if (!listener) {
 				nfo.cm_id = NULL;
 				nfo.conn_type = 0;
@@ -3184,10 +3343,12 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 
 	nes_cm_init_tsa_conn(nesqp, cm_node);
 
-	nesqp->nesqp_context->tcpPorts[0] = cpu_to_le16(ntohs(laddr->sin_port));
-	nesqp->nesqp_context->tcpPorts[1] = cpu_to_le16(ntohs(raddr->sin_port));
+	nesqp->nesqp_context->tcpPorts[0] =
+				cpu_to_le16(cm_node->mapped_loc_port);
+	nesqp->nesqp_context->tcpPorts[1] =
+				cpu_to_le16(cm_node->mapped_rem_port);
 
-	nesqp->nesqp_context->ip0 = cpu_to_le32(ntohl(raddr->sin_addr.s_addr));
+	nesqp->nesqp_context->ip0 = cpu_to_le32(cm_node->mapped_rem_addr);
 
 	nesqp->nesqp_context->misc2 |= cpu_to_le32(
 		(u32)PCI_FUNC(nesdev->pcidev->devfn) <<
@@ -3211,9 +3372,9 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	memset(&nes_quad, 0, sizeof(nes_quad));
 	nes_quad.DstIpAdrIndex =
 		cpu_to_le32((u32)PCI_FUNC(nesdev->pcidev->devfn) << 24);
-	nes_quad.SrcIpadr = raddr->sin_addr.s_addr;
-	nes_quad.TcpPorts[0] = raddr->sin_port;
-	nes_quad.TcpPorts[1] = laddr->sin_port;
+	nes_quad.SrcIpadr = htonl(cm_node->mapped_rem_addr);
+	nes_quad.TcpPorts[0] = htons(cm_node->mapped_rem_port);
+	nes_quad.TcpPorts[1] = htons(cm_node->mapped_loc_port);
 
 	/* Produce hash key */
 	crc_value = get_crc_value(&nes_quad);
@@ -3315,6 +3476,9 @@ int nes_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	int apbvt_set = 0;
 	struct sockaddr_in *laddr = (struct sockaddr_in *)&cm_id->local_addr;
 	struct sockaddr_in *raddr = (struct sockaddr_in *)&cm_id->remote_addr;
+	struct iwpm_dev_data pm_reg_msg;
+	struct iwpm_sa_data pm_msg;
+	int iwpm_err = 0;
 
 	if (cm_id->remote_addr.ss_family != AF_INET)
 		return -ENOSYS;
@@ -3352,20 +3516,44 @@ int nes_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	nes_debug(NES_DBG_CM, "mpa private data len =%u\n",
 		  conn_param->private_data_len);
 
+	/* set up the connection params for the node */
+	cm_info.loc_addr = ntohl(laddr->sin_addr.s_addr);
+	cm_info.loc_port = ntohs(laddr->sin_port);
+	cm_info.rem_addr = ntohl(raddr->sin_addr.s_addr);
+	cm_info.rem_port = ntohs(raddr->sin_port);
+	cm_info.cm_id = cm_id;
+	cm_info.conn_type = NES_CM_IWARP_CONN_TYPE;
+
+	/* No port mapper available, go with the specified peer information */
+	cm_info.mapped_loc_addr = cm_info.loc_addr;
+	cm_info.mapped_loc_port = cm_info.loc_port;
+	cm_info.mapped_rem_addr = cm_info.rem_addr;
+	cm_info.mapped_rem_port = cm_info.rem_port;
+
+	nes_form_reg_msg(nesvnic, &pm_reg_msg);
+	iwpm_err = iwpm_register_pid(&pm_reg_msg, RDMA_NL_NES);
+	if (iwpm_err) {
+		nes_debug(NES_DBG_NLMSG,
+			"Port Mapper reg pid fail (err = %d).\n", iwpm_err);
+	}
+	if (iwpm_valid_pid() && !iwpm_err) {
+		nes_form_pm_msg(&cm_info, &pm_msg);
+		iwpm_err = iwpm_add_and_query_mapping(&pm_msg, RDMA_NL_NES);
+		if (iwpm_err)
+			nes_debug(NES_DBG_NLMSG,
+			"Port Mapper query fail (err = %d).\n", iwpm_err);
+		else
+			nes_record_pm_msg(&cm_info, &pm_msg);
+	}
+
 	if (laddr->sin_addr.s_addr != raddr->sin_addr.s_addr) {
-		nes_manage_apbvt(nesvnic, ntohs(laddr->sin_port),
-				 PCI_FUNC(nesdev->pcidev->devfn),
-				 NES_MANAGE_APBVT_ADD);
+		nes_manage_apbvt(nesvnic, cm_info.mapped_loc_port,
+			PCI_FUNC(nesdev->pcidev->devfn), NES_MANAGE_APBVT_ADD);
 		apbvt_set = 1;
 	}
 
-	/* set up the connection params for the node */
-	cm_info.loc_addr = htonl(laddr->sin_addr.s_addr);
-	cm_info.loc_port = htons(laddr->sin_port);
-	cm_info.rem_addr = htonl(raddr->sin_addr.s_addr);
-	cm_info.rem_port = htons(raddr->sin_port);
-	cm_info.cm_id = cm_id;
-	cm_info.conn_type = NES_CM_IWARP_CONN_TYPE;
+	if (nes_create_mapinfo(&cm_info))
+		return -ENOMEM;
 
 	cm_id->add_ref(cm_id);
 
@@ -3375,10 +3563,14 @@ int nes_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 					  &cm_info);
 	if (!cm_node) {
 		if (apbvt_set)
-			nes_manage_apbvt(nesvnic, ntohs(laddr->sin_port),
+			nes_manage_apbvt(nesvnic, cm_info.mapped_loc_port,
 					 PCI_FUNC(nesdev->pcidev->devfn),
 					 NES_MANAGE_APBVT_DEL);
 
+		nes_debug(NES_DBG_NLMSG, "Delete mapped_loc_port = %04X\n",
+				cm_info.mapped_loc_port);
+		nes_remove_mapinfo(cm_info.loc_addr, cm_info.loc_port,
+			cm_info.mapped_loc_addr, cm_info.mapped_loc_port);
 		cm_id->rem_ref(cm_id);
 		return -ENOMEM;
 	}
@@ -3424,13 +3616,16 @@ int nes_create_listen(struct iw_cm_id *cm_id, int backlog)
 			nesvnic->local_ipaddr, laddr->sin_addr.s_addr);
 
 	/* setup listen params in our api call struct */
-	cm_info.loc_addr = nesvnic->local_ipaddr;
-	cm_info.loc_port = laddr->sin_port;
+	cm_info.loc_addr = ntohl(nesvnic->local_ipaddr);
+	cm_info.loc_port = ntohs(laddr->sin_port);
 	cm_info.backlog = backlog;
 	cm_info.cm_id = cm_id;
 
 	cm_info.conn_type = NES_CM_IWARP_CONN_TYPE;
 
+	/* No port mapper available, go with the specified info */
+	cm_info.mapped_loc_addr = cm_info.loc_addr;
+	cm_info.mapped_loc_port = cm_info.loc_port;
 
 	cm_node = g_cm_core->api->listen(g_cm_core, nesvnic, &cm_info);
 	if (!cm_node) {
@@ -3442,7 +3637,10 @@ int nes_create_listen(struct iw_cm_id *cm_id, int backlog)
 	cm_id->provider_data = cm_node;
 
 	if (!cm_node->reused_node) {
-		err = nes_manage_apbvt(nesvnic, ntohs(laddr->sin_port),
+		if (nes_create_mapinfo(&cm_info))
+			return -ENOMEM;
+
+		err = nes_manage_apbvt(nesvnic, cm_node->mapped_loc_port,
 				       PCI_FUNC(nesvnic->nesdev->pcidev->devfn),
 				       NES_MANAGE_APBVT_ADD);
 		if (err) {
@@ -3567,9 +3765,11 @@ static void cm_event_connected(struct nes_cm_event *event)
 	nes_cm_init_tsa_conn(nesqp, cm_node);
 
 	/* set the QP tsa context */
-	nesqp->nesqp_context->tcpPorts[0] = cpu_to_le16(ntohs(laddr->sin_port));
-	nesqp->nesqp_context->tcpPorts[1] = cpu_to_le16(ntohs(raddr->sin_port));
-	nesqp->nesqp_context->ip0 = cpu_to_le32(ntohl(raddr->sin_addr.s_addr));
+	nesqp->nesqp_context->tcpPorts[0] =
+			cpu_to_le16(cm_node->mapped_loc_port);
+	nesqp->nesqp_context->tcpPorts[1] =
+			cpu_to_le16(cm_node->mapped_rem_port);
+	nesqp->nesqp_context->ip0 = cpu_to_le32(cm_node->mapped_rem_addr);
 
 	nesqp->nesqp_context->misc2 |= cpu_to_le32(
 			(u32)PCI_FUNC(nesdev->pcidev->devfn) <<
@@ -3599,9 +3799,9 @@ static void cm_event_connected(struct nes_cm_event *event)
 
 	nes_quad.DstIpAdrIndex =
 		cpu_to_le32((u32)PCI_FUNC(nesdev->pcidev->devfn) << 24);
-	nes_quad.SrcIpadr = raddr->sin_addr.s_addr;
-	nes_quad.TcpPorts[0] = raddr->sin_port;
-	nes_quad.TcpPorts[1] = laddr->sin_port;
+	nes_quad.SrcIpadr = htonl(cm_node->mapped_rem_addr);
+	nes_quad.TcpPorts[0] = htons(cm_node->mapped_rem_port);
+	nes_quad.TcpPorts[1] = htons(cm_node->mapped_loc_port);
 
 	/* Produce hash key */
 	crc_value = get_crc_value(&nes_quad);
@@ -3629,7 +3829,7 @@ static void cm_event_connected(struct nes_cm_event *event)
 	cm_event.ird = cm_node->ird_size;
 	cm_event.ord = cm_node->ord_size;
 
-	cm_event_laddr->sin_addr.s_addr = event->cm_info.rem_addr;
+	cm_event_laddr->sin_addr.s_addr = htonl(event->cm_info.rem_addr);
 	ret = cm_id->event_handler(cm_id, &cm_event);
 	nes_debug(NES_DBG_CM, "OFA CM event_handler returned, ret=%d\n", ret);
 
diff --git a/drivers/infiniband/hw/nes/nes_cm.h b/drivers/infiniband/hw/nes/nes_cm.h
index 522c99cd07c4..f522cf639789 100644
--- a/drivers/infiniband/hw/nes/nes_cm.h
+++ b/drivers/infiniband/hw/nes/nes_cm.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2006 - 2011 Intel Corporation.  All rights reserved.
+ * Copyright (c) 2006 - 2014 Intel Corporation.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -293,8 +293,8 @@ struct nes_cm_listener {
 	struct list_head           list;
 	struct nes_cm_core         *cm_core;
 	u8                         loc_mac[ETH_ALEN];
-	nes_addr_t                 loc_addr;
-	u16                        loc_port;
+	nes_addr_t                 loc_addr, mapped_loc_addr;
+	u16                        loc_port, mapped_loc_port;
 	struct iw_cm_id            *cm_id;
 	enum nes_cm_conn_type      conn_type;
 	atomic_t                   ref_count;
@@ -308,7 +308,9 @@ struct nes_cm_listener {
 /* per connection node and node state information */
 struct nes_cm_node {
 	nes_addr_t                loc_addr, rem_addr;
+	nes_addr_t                mapped_loc_addr, mapped_rem_addr;
 	u16                       loc_port, rem_port;
+	u16                       mapped_loc_port, mapped_rem_port;
 
 	u8                        loc_mac[ETH_ALEN];
 	u8                        rem_mac[ETH_ALEN];
@@ -364,6 +366,10 @@ struct nes_cm_info {
 	u16 rem_port;
 	nes_addr_t loc_addr;
 	nes_addr_t rem_addr;
+	u16 mapped_loc_port;
+	u16 mapped_rem_port;
+	nes_addr_t mapped_loc_addr;
+	nes_addr_t mapped_rem_addr;
 
 	enum nes_cm_conn_type  conn_type;
 	int backlog;
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c
index 6c54106f5e64..41a9aec9998d 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c
@@ -510,16 +510,9 @@ exit:
 	return status;
 }
 
-static int ocrdma_debugfs_open(struct inode *inode, struct file *file)
-{
-	if (inode->i_private)
-		file->private_data = inode->i_private;
-	return 0;
-}
-
 static const struct file_operations ocrdma_dbg_ops = {
 	.owner = THIS_MODULE,
-	.open = ocrdma_debugfs_open,
+	.open = simple_open,
 	.read = ocrdma_dbgfs_ops_read,
 };
 
diff --git a/drivers/infiniband/hw/qib/qib_init.c b/drivers/infiniband/hw/qib/qib_init.c
index 5b7aeb224a30..8d3c78ddc906 100644
--- a/drivers/infiniband/hw/qib/qib_init.c
+++ b/drivers/infiniband/hw/qib/qib_init.c
@@ -1272,7 +1272,7 @@ static int qib_notify_dca(struct notifier_block *nb, unsigned long event,
  * Do all the generic driver unit- and chip-independent memory
  * allocation and initialization.
  */
-static int __init qlogic_ib_init(void)
+static int __init qib_ib_init(void)
 {
 	int ret;
 
@@ -1316,12 +1316,12 @@ bail:
 	return ret;
 }
 
-module_init(qlogic_ib_init);
+module_init(qib_ib_init);
 
 /*
  * Do the non-unit driver cleanup, memory free, etc. at unload.
  */
-static void __exit qlogic_ib_cleanup(void)
+static void __exit qib_ib_cleanup(void)
 {
 	int ret;
 
@@ -1346,7 +1346,7 @@ static void __exit qlogic_ib_cleanup(void)
 	qib_dev_cleanup();
 }
 
-module_exit(qlogic_ib_cleanup);
+module_exit(qib_ib_cleanup);
 
 /* this can only be called after a successful initialization */
 static void cleanup_device_data(struct qib_devdata *dd)
diff --git a/drivers/infiniband/hw/qib/qib_mad.c b/drivers/infiniband/hw/qib/qib_mad.c
index edad991d60ed..22c720e5740d 100644
--- a/drivers/infiniband/hw/qib/qib_mad.c
+++ b/drivers/infiniband/hw/qib/qib_mad.c
@@ -1028,7 +1028,7 @@ static int set_pkeys(struct qib_devdata *dd, u8 port, u16 *pkeys)
 
 		event.event = IB_EVENT_PKEY_CHANGE;
 		event.device = &dd->verbs_dev.ibdev;
-		event.element.port_num = 1;
+		event.element.port_num = port;
 		ib_dispatch_event(&event);
 	}
 	return 0;
diff --git a/drivers/infiniband/hw/qib/qib_qp.c b/drivers/infiniband/hw/qib/qib_qp.c
index 0cad0c40d742..7fcc150d603c 100644
--- a/drivers/infiniband/hw/qib/qib_qp.c
+++ b/drivers/infiniband/hw/qib/qib_qp.c
@@ -985,7 +985,8 @@ struct ib_qp *qib_create_qp(struct ib_pd *ibpd,
 	struct ib_qp *ret;
 
 	if (init_attr->cap.max_send_sge > ib_qib_max_sges ||
-	    init_attr->cap.max_send_wr > ib_qib_max_qp_wrs) {
+	    init_attr->cap.max_send_wr > ib_qib_max_qp_wrs ||
+	    init_attr->create_flags) {
 		ret = ERR_PTR(-EINVAL);
 		goto bail;
 	}
diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
index d48d2c0a2e3c..53bd6a2d9cdb 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
@@ -466,6 +466,9 @@ struct ib_qp *usnic_ib_create_qp(struct ib_pd *pd,
 	ucontext = to_uucontext(pd->uobject->context);
 	us_ibdev = to_usdev(pd->device);
 
+	if (init_attr->create_flags)
+		return ERR_PTR(-EINVAL);
+
 	err = ib_copy_from_udata(&cmd, udata, sizeof(cmd));
 	if (err) {
 		usnic_err("%s: cannot copy udata for create_qp\n",
diff --git a/drivers/infiniband/hw/usnic/usnic_uiom_interval_tree.c b/drivers/infiniband/hw/usnic/usnic_uiom_interval_tree.c
index d135ad90d914..3a4288e0fbac 100644
--- a/drivers/infiniband/hw/usnic/usnic_uiom_interval_tree.c
+++ b/drivers/infiniband/hw/usnic/usnic_uiom_interval_tree.c
@@ -1,3 +1,21 @@
+/*
+ * Copyright (c) 2014, Cisco Systems, Inc. All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
 #include <linux/init.h>
 #include <linux/list.h>
 #include <linux/slab.h>
diff --git a/drivers/infiniband/ulp/Makefile b/drivers/infiniband/ulp/Makefile
new file mode 100644
index 000000000000..f3c7dcf03098
--- /dev/null
+++ b/drivers/infiniband/ulp/Makefile
@@ -0,0 +1,5 @@
+obj-$(CONFIG_INFINIBAND_IPOIB)		+= ipoib/
+obj-$(CONFIG_INFINIBAND_SRP)		+= srp/
+obj-$(CONFIG_INFINIBAND_SRPT)		+= srpt/
+obj-$(CONFIG_INFINIBAND_ISER)		+= iser/
+obj-$(CONFIG_INFINIBAND_ISERT)		+= isert/
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
index 1377f85911c2..933efcea0d03 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
@@ -1030,10 +1030,20 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_
 		.cap.max_send_sge	= 1,
 		.sq_sig_type		= IB_SIGNAL_ALL_WR,
 		.qp_type		= IB_QPT_RC,
-		.qp_context		= tx
+		.qp_context		= tx,
+		.create_flags		= IB_QP_CREATE_USE_GFP_NOIO
 	};
 
-	return ib_create_qp(priv->pd, &attr);
+	struct ib_qp *tx_qp;
+
+	tx_qp = ib_create_qp(priv->pd, &attr);
+	if (PTR_ERR(tx_qp) == -EINVAL) {
+		ipoib_warn(priv, "can't use GFP_NOIO for QPs on device %s, using GFP_KERNEL\n",
+			   priv->ca->name);
+		attr.create_flags &= ~IB_QP_CREATE_USE_GFP_NOIO;
+		tx_qp = ib_create_qp(priv->pd, &attr);
+	}
+	return tx_qp;
 }
 
 static int ipoib_cm_send_req(struct net_device *dev,
@@ -1104,12 +1114,14 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn,
 	struct ipoib_dev_priv *priv = netdev_priv(p->dev);
 	int ret;
 
-	p->tx_ring = vzalloc(ipoib_sendq_size * sizeof *p->tx_ring);
+	p->tx_ring = __vmalloc(ipoib_sendq_size * sizeof *p->tx_ring,
+			       GFP_NOIO, PAGE_KERNEL);
 	if (!p->tx_ring) {
 		ipoib_warn(priv, "failed to allocate tx ring\n");
 		ret = -ENOMEM;
 		goto err_tx;
 	}
+	memset(p->tx_ring, 0, ipoib_sendq_size * sizeof *p->tx_ring);
 
 	p->qp = ipoib_cm_create_tx_qp(p->dev, p);
 	if (IS_ERR(p->qp)) {
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
index c4b3940845e6..078cadd6c797 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
@@ -105,5 +105,5 @@ static const struct ethtool_ops ipoib_ethtool_ops = {
 
 void ipoib_set_ethtool_ops(struct net_device *dev)
 {
-	SET_ETHTOOL_OPS(dev, &ipoib_ethtool_ops);
+	dev->ethtool_ops = &ipoib_ethtool_ops;
 }
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index 25f195ef44b0..eb7973957a6e 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -99,6 +99,7 @@ MODULE_PARM_DESC(pi_enable, "Enable T10-PI offload support (default:disabled)");
 module_param_named(pi_guard, iser_pi_guard, int, 0644);
 MODULE_PARM_DESC(pi_guard, "T10-PI guard_type, 0:CRC|1:IP_CSUM (default:CRC)");
 
+static struct workqueue_struct *release_wq;
 struct iser_global ig;
 
 void
@@ -337,24 +338,6 @@ iscsi_iser_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 	return cls_conn;
 }
 
-static void
-iscsi_iser_conn_destroy(struct iscsi_cls_conn *cls_conn)
-{
-	struct iscsi_conn *conn = cls_conn->dd_data;
-	struct iser_conn *ib_conn = conn->dd_data;
-
-	iscsi_conn_teardown(cls_conn);
-	/*
-	 * Userspace will normally call the stop callback and
-	 * already have freed the ib_conn, but if it goofed up then
-	 * we free it here.
-	 */
-	if (ib_conn) {
-		ib_conn->iscsi_conn = NULL;
-		iser_conn_put(ib_conn, 1); /* deref iscsi/ib conn unbinding */
-	}
-}
-
 static int
 iscsi_iser_conn_bind(struct iscsi_cls_session *cls_session,
 		     struct iscsi_cls_conn *cls_conn, uint64_t transport_eph,
@@ -392,29 +375,39 @@ iscsi_iser_conn_bind(struct iscsi_cls_session *cls_session,
 	conn->dd_data = ib_conn;
 	ib_conn->iscsi_conn = conn;
 
-	iser_conn_get(ib_conn); /* ref iscsi/ib conn binding */
 	return 0;
 }
 
+static int
+iscsi_iser_conn_start(struct iscsi_cls_conn *cls_conn)
+{
+	struct iscsi_conn *iscsi_conn;
+	struct iser_conn *ib_conn;
+
+	iscsi_conn = cls_conn->dd_data;
+	ib_conn = iscsi_conn->dd_data;
+	reinit_completion(&ib_conn->stop_completion);
+
+	return iscsi_conn_start(cls_conn);
+}
+
 static void
 iscsi_iser_conn_stop(struct iscsi_cls_conn *cls_conn, int flag)
 {
 	struct iscsi_conn *conn = cls_conn->dd_data;
 	struct iser_conn *ib_conn = conn->dd_data;
 
+	iser_dbg("stopping iscsi_conn: %p, ib_conn: %p\n", conn, ib_conn);
+	iscsi_conn_stop(cls_conn, flag);
+
 	/*
 	 * Userspace may have goofed up and not bound the connection or
 	 * might have only partially setup the connection.
 	 */
 	if (ib_conn) {
-		iscsi_conn_stop(cls_conn, flag);
-		/*
-		 * There is no unbind event so the stop callback
-		 * must release the ref from the bind.
-		 */
-		iser_conn_put(ib_conn, 1); /* deref iscsi/ib conn unbinding */
+		conn->dd_data = NULL;
+		complete(&ib_conn->stop_completion);
 	}
-	conn->dd_data = NULL;
 }
 
 static void iscsi_iser_session_destroy(struct iscsi_cls_session *cls_session)
@@ -515,28 +508,28 @@ iscsi_iser_set_param(struct iscsi_cls_conn *cls_conn,
 	case ISCSI_PARAM_HDRDGST_EN:
 		sscanf(buf, "%d", &value);
 		if (value) {
-			iser_err("DataDigest wasn't negotiated to None");
+			iser_err("DataDigest wasn't negotiated to None\n");
 			return -EPROTO;
 		}
 		break;
 	case ISCSI_PARAM_DATADGST_EN:
 		sscanf(buf, "%d", &value);
 		if (value) {
-			iser_err("DataDigest wasn't negotiated to None");
+			iser_err("DataDigest wasn't negotiated to None\n");
 			return -EPROTO;
 		}
 		break;
 	case ISCSI_PARAM_IFMARKER_EN:
 		sscanf(buf, "%d", &value);
 		if (value) {
-			iser_err("IFMarker wasn't negotiated to No");
+			iser_err("IFMarker wasn't negotiated to No\n");
 			return -EPROTO;
 		}
 		break;
 	case ISCSI_PARAM_OFMARKER_EN:
 		sscanf(buf, "%d", &value);
 		if (value) {
-			iser_err("OFMarker wasn't negotiated to No");
+			iser_err("OFMarker wasn't negotiated to No\n");
 			return -EPROTO;
 		}
 		break;
@@ -652,19 +645,20 @@ iscsi_iser_ep_disconnect(struct iscsi_endpoint *ep)
 	struct iser_conn *ib_conn;
 
 	ib_conn = ep->dd_data;
-	if (ib_conn->iscsi_conn)
-		/*
-		 * Must suspend xmit path if the ep is bound to the
-		 * iscsi_conn, so we know we are not accessing the ib_conn
-		 * when we free it.
-		 *
-		 * This may not be bound if the ep poll failed.
-		 */
-		iscsi_suspend_tx(ib_conn->iscsi_conn);
-
-
-	iser_info("ib conn %p state %d\n", ib_conn, ib_conn->state);
+	iser_info("ep %p ib conn %p state %d\n", ep, ib_conn, ib_conn->state);
 	iser_conn_terminate(ib_conn);
+
+	/*
+	 * if iser_conn and iscsi_conn are bound, we must wait iscsi_conn_stop
+	 * call and ISER_CONN_DOWN state before freeing the iser resources.
+	 * otherwise we are safe to free resources immediately.
+	 */
+	if (ib_conn->iscsi_conn) {
+		INIT_WORK(&ib_conn->release_work, iser_release_work);
+		queue_work(release_wq, &ib_conn->release_work);
+	} else {
+		iser_conn_release(ib_conn);
+	}
 }
 
 static umode_t iser_attr_is_visible(int param_type, int param)
@@ -748,13 +742,13 @@ static struct iscsi_transport iscsi_iser_transport = {
 	/* connection management */
 	.create_conn            = iscsi_iser_conn_create,
 	.bind_conn              = iscsi_iser_conn_bind,
-	.destroy_conn           = iscsi_iser_conn_destroy,
+	.destroy_conn           = iscsi_conn_teardown,
 	.attr_is_visible	= iser_attr_is_visible,
 	.set_param              = iscsi_iser_set_param,
 	.get_conn_param		= iscsi_conn_get_param,
 	.get_ep_param		= iscsi_iser_get_ep_param,
 	.get_session_param	= iscsi_session_get_param,
-	.start_conn             = iscsi_conn_start,
+	.start_conn             = iscsi_iser_conn_start,
 	.stop_conn              = iscsi_iser_conn_stop,
 	/* iscsi host params */
 	.get_host_param		= iscsi_host_get_param,
@@ -801,6 +795,12 @@ static int __init iser_init(void)
 	mutex_init(&ig.connlist_mutex);
 	INIT_LIST_HEAD(&ig.connlist);
 
+	release_wq = alloc_workqueue("release workqueue", 0, 0);
+	if (!release_wq) {
+		iser_err("failed to allocate release workqueue\n");
+		return -ENOMEM;
+	}
+
 	iscsi_iser_scsi_transport = iscsi_register_transport(
 							&iscsi_iser_transport);
 	if (!iscsi_iser_scsi_transport) {
@@ -819,7 +819,24 @@ register_transport_failure:
 
 static void __exit iser_exit(void)
 {
+	struct iser_conn *ib_conn, *n;
+	int connlist_empty;
+
 	iser_dbg("Removing iSER datamover...\n");
+	destroy_workqueue(release_wq);
+
+	mutex_lock(&ig.connlist_mutex);
+	connlist_empty = list_empty(&ig.connlist);
+	mutex_unlock(&ig.connlist_mutex);
+
+	if (!connlist_empty) {
+		iser_err("Error cleanup stage completed but we still have iser "
+			 "connections, destroying them anyway.\n");
+		list_for_each_entry_safe(ib_conn, n, &ig.connlist, conn_list) {
+			iser_conn_release(ib_conn);
+		}
+	}
+
 	iscsi_unregister_transport(&iscsi_iser_transport);
 	kmem_cache_destroy(ig.desc_cache);
 }
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h
index 324129f80d40..97cd385bf7f7 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.h
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.h
@@ -69,7 +69,7 @@
 
 #define DRV_NAME	"iser"
 #define PFX		DRV_NAME ": "
-#define DRV_VER		"1.3"
+#define DRV_VER		"1.4"
 
 #define iser_dbg(fmt, arg...)				\
 	do {						\
@@ -333,6 +333,8 @@ struct iser_conn {
 	int                          post_recv_buf_count; /* posted rx count  */
 	atomic_t                     post_send_buf_count; /* posted tx count   */
 	char 			     name[ISER_OBJECT_NAME_SIZE];
+	struct work_struct	     release_work;
+	struct completion	     stop_completion;
 	struct list_head	     conn_list;       /* entry in ig conn list */
 
 	char  			     *login_buf;
@@ -417,12 +419,12 @@ void iscsi_iser_recv(struct iscsi_conn *conn,
 
 void iser_conn_init(struct iser_conn *ib_conn);
 
-void iser_conn_get(struct iser_conn *ib_conn);
-
-int iser_conn_put(struct iser_conn *ib_conn, int destroy_cma_id_allowed);
+void iser_conn_release(struct iser_conn *ib_conn);
 
 void iser_conn_terminate(struct iser_conn *ib_conn);
 
+void iser_release_work(struct work_struct *work);
+
 void iser_rcv_completion(struct iser_rx_desc *desc,
 			 unsigned long    dto_xfer_len,
 			struct iser_conn *ib_conn);
diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c
index 2e2d903db838..8d44a4060634 100644
--- a/drivers/infiniband/ulp/iser/iser_initiator.c
+++ b/drivers/infiniband/ulp/iser/iser_initiator.c
@@ -41,11 +41,11 @@
 #include "iscsi_iser.h"
 
 /* Register user buffer memory and initialize passive rdma
- *  dto descriptor. Total data size is stored in
- *  iser_task->data[ISER_DIR_IN].data_len
+ *  dto descriptor. Data size is stored in
+ *  task->data[ISER_DIR_IN].data_len, Protection size
+ *  os stored in task->prot[ISER_DIR_IN].data_len
  */
-static int iser_prepare_read_cmd(struct iscsi_task *task,
-				 unsigned int edtl)
+static int iser_prepare_read_cmd(struct iscsi_task *task)
 
 {
 	struct iscsi_iser_task *iser_task = task->dd_data;
@@ -73,14 +73,6 @@ static int iser_prepare_read_cmd(struct iscsi_task *task,
 			return err;
 	}
 
-	if (edtl > iser_task->data[ISER_DIR_IN].data_len) {
-		iser_err("Total data length: %ld, less than EDTL: "
-			 "%d, in READ cmd BHS itt: %d, conn: 0x%p\n",
-			 iser_task->data[ISER_DIR_IN].data_len, edtl,
-			 task->itt, iser_task->ib_conn);
-		return -EINVAL;
-	}
-
 	err = device->iser_reg_rdma_mem(iser_task, ISER_DIR_IN);
 	if (err) {
 		iser_err("Failed to set up Data-IN RDMA\n");
@@ -100,8 +92,9 @@ static int iser_prepare_read_cmd(struct iscsi_task *task,
 }
 
 /* Register user buffer memory and initialize passive rdma
- *  dto descriptor. Total data size is stored in
- *  task->data[ISER_DIR_OUT].data_len
+ *  dto descriptor. Data size is stored in
+ *  task->data[ISER_DIR_OUT].data_len, Protection size
+ *  is stored at task->prot[ISER_DIR_OUT].data_len
  */
 static int
 iser_prepare_write_cmd(struct iscsi_task *task,
@@ -135,14 +128,6 @@ iser_prepare_write_cmd(struct iscsi_task *task,
 			return err;
 	}
 
-	if (edtl > iser_task->data[ISER_DIR_OUT].data_len) {
-		iser_err("Total data length: %ld, less than EDTL: %d, "
-			 "in WRITE cmd BHS itt: %d, conn: 0x%p\n",
-			 iser_task->data[ISER_DIR_OUT].data_len,
-			 edtl, task->itt, task->conn);
-		return -EINVAL;
-	}
-
 	err = device->iser_reg_rdma_mem(iser_task, ISER_DIR_OUT);
 	if (err != 0) {
 		iser_err("Failed to register write cmd RDMA mem\n");
@@ -417,11 +402,12 @@ int iser_send_command(struct iscsi_conn *conn,
 	if (scsi_prot_sg_count(sc)) {
 		prot_buf->buf  = scsi_prot_sglist(sc);
 		prot_buf->size = scsi_prot_sg_count(sc);
-		prot_buf->data_len = sc->prot_sdb->length;
+		prot_buf->data_len = data_buf->data_len >>
+				     ilog2(sc->device->sector_size) * 8;
 	}
 
 	if (hdr->flags & ISCSI_FLAG_CMD_READ) {
-		err = iser_prepare_read_cmd(task, edtl);
+		err = iser_prepare_read_cmd(task);
 		if (err)
 			goto send_command_error;
 	}
diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c
index 32849f2becde..ea01075f9f9b 100644
--- a/drivers/infiniband/ulp/iser/iser_verbs.c
+++ b/drivers/infiniband/ulp/iser/iser_verbs.c
@@ -581,14 +581,30 @@ static int iser_conn_state_comp_exch(struct iser_conn *ib_conn,
 	return ret;
 }
 
+void iser_release_work(struct work_struct *work)
+{
+	struct iser_conn *ib_conn;
+
+	ib_conn = container_of(work, struct iser_conn, release_work);
+
+	/* wait for .conn_stop callback */
+	wait_for_completion(&ib_conn->stop_completion);
+
+	/* wait for the qp`s post send and post receive buffers to empty */
+	wait_event_interruptible(ib_conn->wait,
+				 ib_conn->state == ISER_CONN_DOWN);
+
+	iser_conn_release(ib_conn);
+}
+
 /**
  * Frees all conn objects and deallocs conn descriptor
  */
-static void iser_conn_release(struct iser_conn *ib_conn, int can_destroy_id)
+void iser_conn_release(struct iser_conn *ib_conn)
 {
 	struct iser_device  *device = ib_conn->device;
 
-	BUG_ON(ib_conn->state != ISER_CONN_DOWN);
+	BUG_ON(ib_conn->state == ISER_CONN_UP);
 
 	mutex_lock(&ig.connlist_mutex);
 	list_del(&ib_conn->conn_list);
@@ -600,27 +616,13 @@ static void iser_conn_release(struct iser_conn *ib_conn, int can_destroy_id)
 	if (device != NULL)
 		iser_device_try_release(device);
 	/* if cma handler context, the caller actually destroy the id */
-	if (ib_conn->cma_id != NULL && can_destroy_id) {
+	if (ib_conn->cma_id != NULL) {
 		rdma_destroy_id(ib_conn->cma_id);
 		ib_conn->cma_id = NULL;
 	}
 	iscsi_destroy_endpoint(ib_conn->ep);
 }
 
-void iser_conn_get(struct iser_conn *ib_conn)
-{
-	atomic_inc(&ib_conn->refcount);
-}
-
-int iser_conn_put(struct iser_conn *ib_conn, int can_destroy_id)
-{
-	if (atomic_dec_and_test(&ib_conn->refcount)) {
-		iser_conn_release(ib_conn, can_destroy_id);
-		return 1;
-	}
-	return 0;
-}
-
 /**
  * triggers start of the disconnect procedures and wait for them to be done
  */
@@ -638,24 +640,19 @@ void iser_conn_terminate(struct iser_conn *ib_conn)
 	if (err)
 		iser_err("Failed to disconnect, conn: 0x%p err %d\n",
 			 ib_conn,err);
-
-	wait_event_interruptible(ib_conn->wait,
-				 ib_conn->state == ISER_CONN_DOWN);
-
-	iser_conn_put(ib_conn, 1); /* deref ib conn deallocate */
 }
 
-static int iser_connect_error(struct rdma_cm_id *cma_id)
+static void iser_connect_error(struct rdma_cm_id *cma_id)
 {
 	struct iser_conn *ib_conn;
+
 	ib_conn = (struct iser_conn *)cma_id->context;
 
 	ib_conn->state = ISER_CONN_DOWN;
 	wake_up_interruptible(&ib_conn->wait);
-	return iser_conn_put(ib_conn, 0); /* deref ib conn's cma id */
 }
 
-static int iser_addr_handler(struct rdma_cm_id *cma_id)
+static void iser_addr_handler(struct rdma_cm_id *cma_id)
 {
 	struct iser_device *device;
 	struct iser_conn   *ib_conn;
@@ -664,7 +661,8 @@ static int iser_addr_handler(struct rdma_cm_id *cma_id)
 	device = iser_device_find_by_ib_device(cma_id);
 	if (!device) {
 		iser_err("device lookup/creation failed\n");
-		return iser_connect_error(cma_id);
+		iser_connect_error(cma_id);
+		return;
 	}
 
 	ib_conn = (struct iser_conn *)cma_id->context;
@@ -686,13 +684,12 @@ static int iser_addr_handler(struct rdma_cm_id *cma_id)
 	ret = rdma_resolve_route(cma_id, 1000);
 	if (ret) {
 		iser_err("resolve route failed: %d\n", ret);
-		return iser_connect_error(cma_id);
+		iser_connect_error(cma_id);
+		return;
 	}
-
-	return 0;
 }
 
-static int iser_route_handler(struct rdma_cm_id *cma_id)
+static void iser_route_handler(struct rdma_cm_id *cma_id)
 {
 	struct rdma_conn_param conn_param;
 	int    ret;
@@ -720,9 +717,9 @@ static int iser_route_handler(struct rdma_cm_id *cma_id)
 		goto failure;
 	}
 
-	return 0;
+	return;
 failure:
-	return iser_connect_error(cma_id);
+	iser_connect_error(cma_id);
 }
 
 static void iser_connected_handler(struct rdma_cm_id *cma_id)
@@ -735,14 +732,13 @@ static void iser_connected_handler(struct rdma_cm_id *cma_id)
 	iser_info("remote qpn:%x my qpn:%x\n", attr.dest_qp_num, cma_id->qp->qp_num);
 
 	ib_conn = (struct iser_conn *)cma_id->context;
-	ib_conn->state = ISER_CONN_UP;
-	wake_up_interruptible(&ib_conn->wait);
+	if (iser_conn_state_comp_exch(ib_conn, ISER_CONN_PENDING, ISER_CONN_UP))
+		wake_up_interruptible(&ib_conn->wait);
 }
 
-static int iser_disconnected_handler(struct rdma_cm_id *cma_id)
+static void iser_disconnected_handler(struct rdma_cm_id *cma_id)
 {
 	struct iser_conn *ib_conn;
-	int ret;
 
 	ib_conn = (struct iser_conn *)cma_id->context;
 
@@ -762,24 +758,19 @@ static int iser_disconnected_handler(struct rdma_cm_id *cma_id)
 		ib_conn->state = ISER_CONN_DOWN;
 		wake_up_interruptible(&ib_conn->wait);
 	}
-
-	ret = iser_conn_put(ib_conn, 0); /* deref ib conn's cma id */
-	return ret;
 }
 
 static int iser_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
 {
-	int ret = 0;
-
 	iser_info("event %d status %d conn %p id %p\n",
 		  event->event, event->status, cma_id->context, cma_id);
 
 	switch (event->event) {
 	case RDMA_CM_EVENT_ADDR_RESOLVED:
-		ret = iser_addr_handler(cma_id);
+		iser_addr_handler(cma_id);
 		break;
 	case RDMA_CM_EVENT_ROUTE_RESOLVED:
-		ret = iser_route_handler(cma_id);
+		iser_route_handler(cma_id);
 		break;
 	case RDMA_CM_EVENT_ESTABLISHED:
 		iser_connected_handler(cma_id);
@@ -789,18 +780,18 @@ static int iser_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *eve
 	case RDMA_CM_EVENT_CONNECT_ERROR:
 	case RDMA_CM_EVENT_UNREACHABLE:
 	case RDMA_CM_EVENT_REJECTED:
-		ret = iser_connect_error(cma_id);
+		iser_connect_error(cma_id);
 		break;
 	case RDMA_CM_EVENT_DISCONNECTED:
 	case RDMA_CM_EVENT_DEVICE_REMOVAL:
 	case RDMA_CM_EVENT_ADDR_CHANGE:
-		ret = iser_disconnected_handler(cma_id);
+		iser_disconnected_handler(cma_id);
 		break;
 	default:
 		iser_err("Unexpected RDMA CM event (%d)\n", event->event);
 		break;
 	}
-	return ret;
+	return 0;
 }
 
 void iser_conn_init(struct iser_conn *ib_conn)
@@ -809,7 +800,7 @@ void iser_conn_init(struct iser_conn *ib_conn)
 	init_waitqueue_head(&ib_conn->wait);
 	ib_conn->post_recv_buf_count = 0;
 	atomic_set(&ib_conn->post_send_buf_count, 0);
-	atomic_set(&ib_conn->refcount, 1); /* ref ib conn allocation */
+	init_completion(&ib_conn->stop_completion);
 	INIT_LIST_HEAD(&ib_conn->conn_list);
 	spin_lock_init(&ib_conn->lock);
 }
@@ -837,7 +828,6 @@ int iser_connect(struct iser_conn   *ib_conn,
 
 	ib_conn->state = ISER_CONN_PENDING;
 
-	iser_conn_get(ib_conn); /* ref ib conn's cma id */
 	ib_conn->cma_id = rdma_create_id(iser_cma_handler,
 					     (void *)ib_conn,
 					     RDMA_PS_TCP, IB_QPT_RC);
@@ -874,9 +864,8 @@ id_failure:
 	ib_conn->cma_id = NULL;
 addr_failure:
 	ib_conn->state = ISER_CONN_DOWN;
-	iser_conn_put(ib_conn, 1); /* deref ib conn's cma id */
 connect_failure:
-	iser_conn_put(ib_conn, 1); /* deref ib conn deallocate */
+	iser_conn_release(ib_conn);
 	return err;
 }
 
diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index b9d647468b99..d4c7928a0f36 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -663,8 +663,9 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
 
 	pi_support = np->tpg_np->tpg->tpg_attrib.t10_pi;
 	if (pi_support && !device->pi_capable) {
-		pr_err("Protection information requested but not supported\n");
-		ret = -EINVAL;
+		pr_err("Protection information requested but not supported, "
+		       "rejecting connect request\n");
+		ret = rdma_reject(cma_id, NULL, 0);
 		goto out_mr;
 	}
 
@@ -787,14 +788,12 @@ isert_disconnect_work(struct work_struct *work)
 		isert_put_conn(isert_conn);
 		return;
 	}
-	if (!isert_conn->logout_posted) {
-		pr_debug("Calling rdma_disconnect for !logout_posted from"
-			 " isert_disconnect_work\n");
+
+	if (isert_conn->disconnect) {
+		/* Send DREQ/DREP towards our initiator */
 		rdma_disconnect(isert_conn->conn_cm_id);
-		mutex_unlock(&isert_conn->conn_mutex);
-		iscsit_cause_connection_reinstatement(isert_conn->conn, 0);
-		goto wake_up;
 	}
+
 	mutex_unlock(&isert_conn->conn_mutex);
 
 wake_up:
@@ -803,10 +802,11 @@ wake_up:
 }
 
 static void
-isert_disconnected_handler(struct rdma_cm_id *cma_id)
+isert_disconnected_handler(struct rdma_cm_id *cma_id, bool disconnect)
 {
 	struct isert_conn *isert_conn = (struct isert_conn *)cma_id->context;
 
+	isert_conn->disconnect = disconnect;
 	INIT_WORK(&isert_conn->conn_logout_work, isert_disconnect_work);
 	schedule_work(&isert_conn->conn_logout_work);
 }
@@ -815,29 +815,28 @@ static int
 isert_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
 {
 	int ret = 0;
+	bool disconnect = false;
 
 	pr_debug("isert_cma_handler: event %d status %d conn %p id %p\n",
 		 event->event, event->status, cma_id->context, cma_id);
 
 	switch (event->event) {
 	case RDMA_CM_EVENT_CONNECT_REQUEST:
-		pr_debug("RDMA_CM_EVENT_CONNECT_REQUEST: >>>>>>>>>>>>>>>\n");
 		ret = isert_connect_request(cma_id, event);
 		break;
 	case RDMA_CM_EVENT_ESTABLISHED:
-		pr_debug("RDMA_CM_EVENT_ESTABLISHED >>>>>>>>>>>>>>\n");
 		isert_connected_handler(cma_id);
 		break;
-	case RDMA_CM_EVENT_DISCONNECTED:
-		pr_debug("RDMA_CM_EVENT_DISCONNECTED: >>>>>>>>>>>>>>\n");
-		isert_disconnected_handler(cma_id);
-		break;
-	case RDMA_CM_EVENT_DEVICE_REMOVAL:
-	case RDMA_CM_EVENT_ADDR_CHANGE:
+	case RDMA_CM_EVENT_ADDR_CHANGE:    /* FALLTHRU */
+	case RDMA_CM_EVENT_DISCONNECTED:   /* FALLTHRU */
+	case RDMA_CM_EVENT_DEVICE_REMOVAL: /* FALLTHRU */
+		disconnect = true;
+	case RDMA_CM_EVENT_TIMEWAIT_EXIT:  /* FALLTHRU */
+		isert_disconnected_handler(cma_id, disconnect);
 		break;
 	case RDMA_CM_EVENT_CONNECT_ERROR:
 	default:
-		pr_err("Unknown RDMA CMA event: %d\n", event->event);
+		pr_err("Unhandled RDMA CMA event: %d\n", event->event);
 		break;
 	}
 
@@ -1054,7 +1053,9 @@ isert_put_login_tx(struct iscsi_conn *conn, struct iscsi_login *login,
 	}
 	if (!login->login_failed) {
 		if (login->login_complete) {
-			if (isert_conn->conn_device->use_fastreg) {
+			if (!conn->sess->sess_ops->SessionType &&
+			    isert_conn->conn_device->use_fastreg) {
+				/* Normal Session and fastreg is used */
 				u8 pi_support = login->np->tpg_np->tpg->tpg_attrib.t10_pi;
 
 				ret = isert_conn_create_fastreg_pool(isert_conn,
@@ -1824,11 +1825,8 @@ isert_do_control_comp(struct work_struct *work)
 		break;
 	case ISTATE_SEND_LOGOUTRSP:
 		pr_debug("Calling iscsit_logout_post_handler >>>>>>>>>>>>>>\n");
-		/*
-		 * Call atomic_dec(&isert_conn->post_send_buf_count)
-		 * from isert_wait_conn()
-		 */
-		isert_conn->logout_posted = true;
+
+		atomic_dec(&isert_conn->post_send_buf_count);
 		iscsit_logout_post_handler(cmd, cmd->conn);
 		break;
 	case ISTATE_SEND_TEXTRSP:
@@ -2034,6 +2032,8 @@ isert_cq_rx_comp_err(struct isert_conn *isert_conn)
 	isert_conn->state = ISER_CONN_DOWN;
 	mutex_unlock(&isert_conn->conn_mutex);
 
+	iscsit_cause_connection_reinstatement(isert_conn->conn, 0);
+
 	complete(&isert_conn->conn_wait_comp_err);
 }
 
@@ -2320,7 +2320,7 @@ isert_put_text_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn)
 	int rc;
 
 	isert_create_send_desc(isert_conn, isert_cmd, &isert_cmd->tx_desc);
-	rc = iscsit_build_text_rsp(cmd, conn, hdr);
+	rc = iscsit_build_text_rsp(cmd, conn, hdr, ISCSI_INFINIBAND);
 	if (rc < 0)
 		return rc;
 
@@ -3156,9 +3156,14 @@ accept_wait:
 		return -ENODEV;
 
 	spin_lock_bh(&np->np_thread_lock);
-	if (np->np_thread_state == ISCSI_NP_THREAD_RESET) {
+	if (np->np_thread_state >= ISCSI_NP_THREAD_RESET) {
 		spin_unlock_bh(&np->np_thread_lock);
-		pr_debug("ISCSI_NP_THREAD_RESET for isert_accept_np\n");
+		pr_debug("np_thread_state %d for isert_accept_np\n",
+			 np->np_thread_state);
+		/**
+		 * No point in stalling here when np_thread
+		 * is in state RESET/SHUTDOWN/EXIT - bail
+		 **/
 		return -ENODEV;
 	}
 	spin_unlock_bh(&np->np_thread_lock);
@@ -3208,15 +3213,9 @@ static void isert_wait_conn(struct iscsi_conn *conn)
 	struct isert_conn *isert_conn = conn->context;
 
 	pr_debug("isert_wait_conn: Starting \n");
-	/*
-	 * Decrement post_send_buf_count for special case when called
-	 * from isert_do_control_comp() -> iscsit_logout_post_handler()
-	 */
-	mutex_lock(&isert_conn->conn_mutex);
-	if (isert_conn->logout_posted)
-		atomic_dec(&isert_conn->post_send_buf_count);
 
-	if (isert_conn->conn_cm_id && isert_conn->state != ISER_CONN_DOWN) {
+	mutex_lock(&isert_conn->conn_mutex);
+	if (isert_conn->conn_cm_id) {
 		pr_debug("Calling rdma_disconnect from isert_wait_conn\n");
 		rdma_disconnect(isert_conn->conn_cm_id);
 	}
@@ -3293,6 +3292,7 @@ destroy_rx_wq:
 
 static void __exit isert_exit(void)
 {
+	flush_scheduled_work();
 	destroy_workqueue(isert_comp_wq);
 	destroy_workqueue(isert_rx_wq);
 	iscsit_unregister_transport(&iser_target_transport);
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index da6612e68000..04f51f7bf614 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -116,7 +116,6 @@ struct isert_device;
 
 struct isert_conn {
 	enum iser_conn_state	state;
-	bool			logout_posted;
 	int			post_recv_buf_count;
 	atomic_t		post_send_buf_count;
 	u32			responder_resources;
@@ -151,6 +150,7 @@ struct isert_conn {
 #define ISERT_COMP_BATCH_COUNT	8
 	int			conn_comp_batch;
 	struct llist_head	conn_comp_llist;
+	bool                    disconnect;
 };
 
 #define ISERT_MAX_CQ 64
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index 66a908bf3fb9..e3c2c5b4297f 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -30,7 +30,7 @@
  * SOFTWARE.
  */
 
-#define pr_fmt(fmt) PFX fmt
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/module.h>
 #include <linux/init.h>
@@ -66,6 +66,8 @@ static unsigned int srp_sg_tablesize;
 static unsigned int cmd_sg_entries;
 static unsigned int indirect_sg_entries;
 static bool allow_ext_sg;
+static bool prefer_fr;
+static bool register_always;
 static int topspin_workarounds = 1;
 
 module_param(srp_sg_tablesize, uint, 0444);
@@ -87,6 +89,14 @@ module_param(topspin_workarounds, int, 0444);
 MODULE_PARM_DESC(topspin_workarounds,
 		 "Enable workarounds for Topspin/Cisco SRP target bugs if != 0");
 
+module_param(prefer_fr, bool, 0444);
+MODULE_PARM_DESC(prefer_fr,
+"Whether to use fast registration if both FMR and fast registration are supported");
+
+module_param(register_always, bool, 0444);
+MODULE_PARM_DESC(register_always,
+		 "Use memory registration even for contiguous memory regions");
+
 static struct kernel_param_ops srp_tmo_ops;
 
 static int srp_reconnect_delay = 10;
@@ -288,28 +298,174 @@ static int srp_new_cm_id(struct srp_target_port *target)
 	return 0;
 }
 
+static struct ib_fmr_pool *srp_alloc_fmr_pool(struct srp_target_port *target)
+{
+	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_fmr_pool_param fmr_param;
+
+	memset(&fmr_param, 0, sizeof(fmr_param));
+	fmr_param.pool_size	    = target->scsi_host->can_queue;
+	fmr_param.dirty_watermark   = fmr_param.pool_size / 4;
+	fmr_param.cache		    = 1;
+	fmr_param.max_pages_per_fmr = dev->max_pages_per_mr;
+	fmr_param.page_shift	    = ilog2(dev->mr_page_size);
+	fmr_param.access	    = (IB_ACCESS_LOCAL_WRITE |
+				       IB_ACCESS_REMOTE_WRITE |
+				       IB_ACCESS_REMOTE_READ);
+
+	return ib_create_fmr_pool(dev->pd, &fmr_param);
+}
+
+/**
+ * srp_destroy_fr_pool() - free the resources owned by a pool
+ * @pool: Fast registration pool to be destroyed.
+ */
+static void srp_destroy_fr_pool(struct srp_fr_pool *pool)
+{
+	int i;
+	struct srp_fr_desc *d;
+
+	if (!pool)
+		return;
+
+	for (i = 0, d = &pool->desc[0]; i < pool->size; i++, d++) {
+		if (d->frpl)
+			ib_free_fast_reg_page_list(d->frpl);
+		if (d->mr)
+			ib_dereg_mr(d->mr);
+	}
+	kfree(pool);
+}
+
+/**
+ * srp_create_fr_pool() - allocate and initialize a pool for fast registration
+ * @device:            IB device to allocate fast registration descriptors for.
+ * @pd:                Protection domain associated with the FR descriptors.
+ * @pool_size:         Number of descriptors to allocate.
+ * @max_page_list_len: Maximum fast registration work request page list length.
+ */
+static struct srp_fr_pool *srp_create_fr_pool(struct ib_device *device,
+					      struct ib_pd *pd, int pool_size,
+					      int max_page_list_len)
+{
+	struct srp_fr_pool *pool;
+	struct srp_fr_desc *d;
+	struct ib_mr *mr;
+	struct ib_fast_reg_page_list *frpl;
+	int i, ret = -EINVAL;
+
+	if (pool_size <= 0)
+		goto err;
+	ret = -ENOMEM;
+	pool = kzalloc(sizeof(struct srp_fr_pool) +
+		       pool_size * sizeof(struct srp_fr_desc), GFP_KERNEL);
+	if (!pool)
+		goto err;
+	pool->size = pool_size;
+	pool->max_page_list_len = max_page_list_len;
+	spin_lock_init(&pool->lock);
+	INIT_LIST_HEAD(&pool->free_list);
+
+	for (i = 0, d = &pool->desc[0]; i < pool->size; i++, d++) {
+		mr = ib_alloc_fast_reg_mr(pd, max_page_list_len);
+		if (IS_ERR(mr)) {
+			ret = PTR_ERR(mr);
+			goto destroy_pool;
+		}
+		d->mr = mr;
+		frpl = ib_alloc_fast_reg_page_list(device, max_page_list_len);
+		if (IS_ERR(frpl)) {
+			ret = PTR_ERR(frpl);
+			goto destroy_pool;
+		}
+		d->frpl = frpl;
+		list_add_tail(&d->entry, &pool->free_list);
+	}
+
+out:
+	return pool;
+
+destroy_pool:
+	srp_destroy_fr_pool(pool);
+
+err:
+	pool = ERR_PTR(ret);
+	goto out;
+}
+
+/**
+ * srp_fr_pool_get() - obtain a descriptor suitable for fast registration
+ * @pool: Pool to obtain descriptor from.
+ */
+static struct srp_fr_desc *srp_fr_pool_get(struct srp_fr_pool *pool)
+{
+	struct srp_fr_desc *d = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pool->lock, flags);
+	if (!list_empty(&pool->free_list)) {
+		d = list_first_entry(&pool->free_list, typeof(*d), entry);
+		list_del(&d->entry);
+	}
+	spin_unlock_irqrestore(&pool->lock, flags);
+
+	return d;
+}
+
+/**
+ * srp_fr_pool_put() - put an FR descriptor back in the free list
+ * @pool: Pool the descriptor was allocated from.
+ * @desc: Pointer to an array of fast registration descriptor pointers.
+ * @n:    Number of descriptors to put back.
+ *
+ * Note: The caller must already have queued an invalidation request for
+ * desc->mr->rkey before calling this function.
+ */
+static void srp_fr_pool_put(struct srp_fr_pool *pool, struct srp_fr_desc **desc,
+			    int n)
+{
+	unsigned long flags;
+	int i;
+
+	spin_lock_irqsave(&pool->lock, flags);
+	for (i = 0; i < n; i++)
+		list_add(&desc[i]->entry, &pool->free_list);
+	spin_unlock_irqrestore(&pool->lock, flags);
+}
+
+static struct srp_fr_pool *srp_alloc_fr_pool(struct srp_target_port *target)
+{
+	struct srp_device *dev = target->srp_host->srp_dev;
+
+	return srp_create_fr_pool(dev->dev, dev->pd,
+				  target->scsi_host->can_queue,
+				  dev->max_pages_per_mr);
+}
+
 static int srp_create_target_ib(struct srp_target_port *target)
 {
+	struct srp_device *dev = target->srp_host->srp_dev;
 	struct ib_qp_init_attr *init_attr;
 	struct ib_cq *recv_cq, *send_cq;
 	struct ib_qp *qp;
+	struct ib_fmr_pool *fmr_pool = NULL;
+	struct srp_fr_pool *fr_pool = NULL;
+	const int m = 1 + dev->use_fast_reg;
 	int ret;
 
 	init_attr = kzalloc(sizeof *init_attr, GFP_KERNEL);
 	if (!init_attr)
 		return -ENOMEM;
 
-	recv_cq = ib_create_cq(target->srp_host->srp_dev->dev,
-			       srp_recv_completion, NULL, target,
+	recv_cq = ib_create_cq(dev->dev, srp_recv_completion, NULL, target,
 			       target->queue_size, target->comp_vector);
 	if (IS_ERR(recv_cq)) {
 		ret = PTR_ERR(recv_cq);
 		goto err;
 	}
 
-	send_cq = ib_create_cq(target->srp_host->srp_dev->dev,
-			       srp_send_completion, NULL, target,
-			       target->queue_size, target->comp_vector);
+	send_cq = ib_create_cq(dev->dev, srp_send_completion, NULL, target,
+			       m * target->queue_size, target->comp_vector);
 	if (IS_ERR(send_cq)) {
 		ret = PTR_ERR(send_cq);
 		goto err_recv_cq;
@@ -318,16 +474,16 @@ static int srp_create_target_ib(struct srp_target_port *target)
 	ib_req_notify_cq(recv_cq, IB_CQ_NEXT_COMP);
 
 	init_attr->event_handler       = srp_qp_event;
-	init_attr->cap.max_send_wr     = target->queue_size;
+	init_attr->cap.max_send_wr     = m * target->queue_size;
 	init_attr->cap.max_recv_wr     = target->queue_size;
 	init_attr->cap.max_recv_sge    = 1;
 	init_attr->cap.max_send_sge    = 1;
-	init_attr->sq_sig_type         = IB_SIGNAL_ALL_WR;
+	init_attr->sq_sig_type         = IB_SIGNAL_REQ_WR;
 	init_attr->qp_type             = IB_QPT_RC;
 	init_attr->send_cq             = send_cq;
 	init_attr->recv_cq             = recv_cq;
 
-	qp = ib_create_qp(target->srp_host->srp_dev->pd, init_attr);
+	qp = ib_create_qp(dev->pd, init_attr);
 	if (IS_ERR(qp)) {
 		ret = PTR_ERR(qp);
 		goto err_send_cq;
@@ -337,6 +493,30 @@ static int srp_create_target_ib(struct srp_target_port *target)
 	if (ret)
 		goto err_qp;
 
+	if (dev->use_fast_reg && dev->has_fr) {
+		fr_pool = srp_alloc_fr_pool(target);
+		if (IS_ERR(fr_pool)) {
+			ret = PTR_ERR(fr_pool);
+			shost_printk(KERN_WARNING, target->scsi_host, PFX
+				     "FR pool allocation failed (%d)\n", ret);
+			goto err_qp;
+		}
+		if (target->fr_pool)
+			srp_destroy_fr_pool(target->fr_pool);
+		target->fr_pool = fr_pool;
+	} else if (!dev->use_fast_reg && dev->has_fmr) {
+		fmr_pool = srp_alloc_fmr_pool(target);
+		if (IS_ERR(fmr_pool)) {
+			ret = PTR_ERR(fmr_pool);
+			shost_printk(KERN_WARNING, target->scsi_host, PFX
+				     "FMR pool allocation failed (%d)\n", ret);
+			goto err_qp;
+		}
+		if (target->fmr_pool)
+			ib_destroy_fmr_pool(target->fmr_pool);
+		target->fmr_pool = fmr_pool;
+	}
+
 	if (target->qp)
 		ib_destroy_qp(target->qp);
 	if (target->recv_cq)
@@ -371,8 +551,16 @@ err:
  */
 static void srp_free_target_ib(struct srp_target_port *target)
 {
+	struct srp_device *dev = target->srp_host->srp_dev;
 	int i;
 
+	if (dev->use_fast_reg) {
+		if (target->fr_pool)
+			srp_destroy_fr_pool(target->fr_pool);
+	} else {
+		if (target->fmr_pool)
+			ib_destroy_fmr_pool(target->fmr_pool);
+	}
 	ib_destroy_qp(target->qp);
 	ib_destroy_cq(target->send_cq);
 	ib_destroy_cq(target->recv_cq);
@@ -577,7 +765,8 @@ static void srp_disconnect_target(struct srp_target_port *target)
 
 static void srp_free_req_data(struct srp_target_port *target)
 {
-	struct ib_device *ibdev = target->srp_host->srp_dev->dev;
+	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_device *ibdev = dev->dev;
 	struct srp_request *req;
 	int i;
 
@@ -586,7 +775,10 @@ static void srp_free_req_data(struct srp_target_port *target)
 
 	for (i = 0; i < target->req_ring_size; ++i) {
 		req = &target->req_ring[i];
-		kfree(req->fmr_list);
+		if (dev->use_fast_reg)
+			kfree(req->fr_list);
+		else
+			kfree(req->fmr_list);
 		kfree(req->map_page);
 		if (req->indirect_dma_addr) {
 			ib_dma_unmap_single(ibdev, req->indirect_dma_addr,
@@ -605,6 +797,7 @@ static int srp_alloc_req_data(struct srp_target_port *target)
 	struct srp_device *srp_dev = target->srp_host->srp_dev;
 	struct ib_device *ibdev = srp_dev->dev;
 	struct srp_request *req;
+	void *mr_list;
 	dma_addr_t dma_addr;
 	int i, ret = -ENOMEM;
 
@@ -617,12 +810,20 @@ static int srp_alloc_req_data(struct srp_target_port *target)
 
 	for (i = 0; i < target->req_ring_size; ++i) {
 		req = &target->req_ring[i];
-		req->fmr_list = kmalloc(target->cmd_sg_cnt * sizeof(void *),
-					GFP_KERNEL);
-		req->map_page = kmalloc(SRP_FMR_SIZE * sizeof(void *),
-					GFP_KERNEL);
+		mr_list = kmalloc(target->cmd_sg_cnt * sizeof(void *),
+				  GFP_KERNEL);
+		if (!mr_list)
+			goto out;
+		if (srp_dev->use_fast_reg)
+			req->fr_list = mr_list;
+		else
+			req->fmr_list = mr_list;
+		req->map_page = kmalloc(srp_dev->max_pages_per_mr *
+					sizeof(void *), GFP_KERNEL);
+		if (!req->map_page)
+			goto out;
 		req->indirect_desc = kmalloc(target->indirect_size, GFP_KERNEL);
-		if (!req->fmr_list || !req->map_page || !req->indirect_desc)
+		if (!req->indirect_desc)
 			goto out;
 
 		dma_addr = ib_dma_map_single(ibdev, req->indirect_desc,
@@ -759,21 +960,56 @@ static int srp_connect_target(struct srp_target_port *target)
 	}
 }
 
+static int srp_inv_rkey(struct srp_target_port *target, u32 rkey)
+{
+	struct ib_send_wr *bad_wr;
+	struct ib_send_wr wr = {
+		.opcode		    = IB_WR_LOCAL_INV,
+		.wr_id		    = LOCAL_INV_WR_ID_MASK,
+		.next		    = NULL,
+		.num_sge	    = 0,
+		.send_flags	    = 0,
+		.ex.invalidate_rkey = rkey,
+	};
+
+	return ib_post_send(target->qp, &wr, &bad_wr);
+}
+
 static void srp_unmap_data(struct scsi_cmnd *scmnd,
 			   struct srp_target_port *target,
 			   struct srp_request *req)
 {
-	struct ib_device *ibdev = target->srp_host->srp_dev->dev;
-	struct ib_pool_fmr **pfmr;
+	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_device *ibdev = dev->dev;
+	int i, res;
 
 	if (!scsi_sglist(scmnd) ||
 	    (scmnd->sc_data_direction != DMA_TO_DEVICE &&
 	     scmnd->sc_data_direction != DMA_FROM_DEVICE))
 		return;
 
-	pfmr = req->fmr_list;
-	while (req->nfmr--)
-		ib_fmr_pool_unmap(*pfmr++);
+	if (dev->use_fast_reg) {
+		struct srp_fr_desc **pfr;
+
+		for (i = req->nmdesc, pfr = req->fr_list; i > 0; i--, pfr++) {
+			res = srp_inv_rkey(target, (*pfr)->mr->rkey);
+			if (res < 0) {
+				shost_printk(KERN_ERR, target->scsi_host, PFX
+				  "Queueing INV WR for rkey %#x failed (%d)\n",
+				  (*pfr)->mr->rkey, res);
+				queue_work(system_long_wq,
+					   &target->tl_err_work);
+			}
+		}
+		if (req->nmdesc)
+			srp_fr_pool_put(target->fr_pool, req->fr_list,
+					req->nmdesc);
+	} else {
+		struct ib_pool_fmr **pfmr;
+
+		for (i = req->nmdesc, pfmr = req->fmr_list; i > 0; i--, pfmr++)
+			ib_fmr_pool_unmap(*pfmr);
+	}
 
 	ib_dma_unmap_sg(ibdev, scsi_sglist(scmnd), scsi_sg_count(scmnd),
 			scmnd->sc_data_direction);
@@ -813,6 +1049,10 @@ static struct scsi_cmnd *srp_claim_req(struct srp_target_port *target,
 
 /**
  * srp_free_req() - Unmap data and add request to the free request list.
+ * @target: SRP target port.
+ * @req:    Request to be freed.
+ * @scmnd:  SCSI command associated with @req.
+ * @req_lim_delta: Amount to be added to @target->req_lim.
  */
 static void srp_free_req(struct srp_target_port *target,
 			 struct srp_request *req, struct scsi_cmnd *scmnd,
@@ -882,21 +1122,19 @@ static int srp_rport_reconnect(struct srp_rport *rport)
 	 * callbacks will have finished before a new QP is allocated.
 	 */
 	ret = srp_new_cm_id(target);
-	/*
-	 * Whether or not creating a new CM ID succeeded, create a new
-	 * QP. This guarantees that all completion callback function
-	 * invocations have finished before request resetting starts.
-	 */
-	if (ret == 0)
-		ret = srp_create_target_ib(target);
-	else
-		srp_create_target_ib(target);
 
 	for (i = 0; i < target->req_ring_size; ++i) {
 		struct srp_request *req = &target->req_ring[i];
 		srp_finish_req(target, req, NULL, DID_RESET << 16);
 	}
 
+	/*
+	 * Whether or not creating a new CM ID succeeded, create a new
+	 * QP. This guarantees that all callback functions for the old QP have
+	 * finished before any send requests are posted on the new QP.
+	 */
+	ret += srp_create_target_ib(target);
+
 	INIT_LIST_HEAD(&target->free_tx);
 	for (i = 0; i < target->queue_size; ++i)
 		list_add(&target->tx_ring[i]->list, &target->free_tx);
@@ -928,33 +1166,87 @@ static void srp_map_desc(struct srp_map_state *state, dma_addr_t dma_addr,
 static int srp_map_finish_fmr(struct srp_map_state *state,
 			      struct srp_target_port *target)
 {
-	struct srp_device *dev = target->srp_host->srp_dev;
 	struct ib_pool_fmr *fmr;
 	u64 io_addr = 0;
 
-	if (!state->npages)
-		return 0;
-
-	if (state->npages == 1) {
-		srp_map_desc(state, state->base_dma_addr, state->fmr_len,
-			     target->rkey);
-		state->npages = state->fmr_len = 0;
-		return 0;
-	}
-
-	fmr = ib_fmr_pool_map_phys(dev->fmr_pool, state->pages,
+	fmr = ib_fmr_pool_map_phys(target->fmr_pool, state->pages,
 				   state->npages, io_addr);
 	if (IS_ERR(fmr))
 		return PTR_ERR(fmr);
 
 	*state->next_fmr++ = fmr;
-	state->nfmr++;
+	state->nmdesc++;
+
+	srp_map_desc(state, 0, state->dma_len, fmr->fmr->rkey);
 
-	srp_map_desc(state, 0, state->fmr_len, fmr->fmr->rkey);
-	state->npages = state->fmr_len = 0;
 	return 0;
 }
 
+static int srp_map_finish_fr(struct srp_map_state *state,
+			     struct srp_target_port *target)
+{
+	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_send_wr *bad_wr;
+	struct ib_send_wr wr;
+	struct srp_fr_desc *desc;
+	u32 rkey;
+
+	desc = srp_fr_pool_get(target->fr_pool);
+	if (!desc)
+		return -ENOMEM;
+
+	rkey = ib_inc_rkey(desc->mr->rkey);
+	ib_update_fast_reg_key(desc->mr, rkey);
+
+	memcpy(desc->frpl->page_list, state->pages,
+	       sizeof(state->pages[0]) * state->npages);
+
+	memset(&wr, 0, sizeof(wr));
+	wr.opcode = IB_WR_FAST_REG_MR;
+	wr.wr_id = FAST_REG_WR_ID_MASK;
+	wr.wr.fast_reg.iova_start = state->base_dma_addr;
+	wr.wr.fast_reg.page_list = desc->frpl;
+	wr.wr.fast_reg.page_list_len = state->npages;
+	wr.wr.fast_reg.page_shift = ilog2(dev->mr_page_size);
+	wr.wr.fast_reg.length = state->dma_len;
+	wr.wr.fast_reg.access_flags = (IB_ACCESS_LOCAL_WRITE |
+				       IB_ACCESS_REMOTE_READ |
+				       IB_ACCESS_REMOTE_WRITE);
+	wr.wr.fast_reg.rkey = desc->mr->lkey;
+
+	*state->next_fr++ = desc;
+	state->nmdesc++;
+
+	srp_map_desc(state, state->base_dma_addr, state->dma_len,
+		     desc->mr->rkey);
+
+	return ib_post_send(target->qp, &wr, &bad_wr);
+}
+
+static int srp_finish_mapping(struct srp_map_state *state,
+			      struct srp_target_port *target)
+{
+	int ret = 0;
+
+	if (state->npages == 0)
+		return 0;
+
+	if (state->npages == 1 && !register_always)
+		srp_map_desc(state, state->base_dma_addr, state->dma_len,
+			     target->rkey);
+	else
+		ret = target->srp_host->srp_dev->use_fast_reg ?
+			srp_map_finish_fr(state, target) :
+			srp_map_finish_fmr(state, target);
+
+	if (ret == 0) {
+		state->npages = 0;
+		state->dma_len = 0;
+	}
+
+	return ret;
+}
+
 static void srp_map_update_start(struct srp_map_state *state,
 				 struct scatterlist *sg, int sg_index,
 				 dma_addr_t dma_addr)
@@ -967,7 +1259,7 @@ static void srp_map_update_start(struct srp_map_state *state,
 static int srp_map_sg_entry(struct srp_map_state *state,
 			    struct srp_target_port *target,
 			    struct scatterlist *sg, int sg_index,
-			    int use_fmr)
+			    bool use_mr)
 {
 	struct srp_device *dev = target->srp_host->srp_dev;
 	struct ib_device *ibdev = dev->dev;
@@ -979,23 +1271,25 @@ static int srp_map_sg_entry(struct srp_map_state *state,
 	if (!dma_len)
 		return 0;
 
-	if (use_fmr == SRP_MAP_NO_FMR) {
-		/* Once we're in direct map mode for a request, we don't
-		 * go back to FMR mode, so no need to update anything
+	if (!use_mr) {
+		/*
+		 * Once we're in direct map mode for a request, we don't
+		 * go back to FMR or FR mode, so no need to update anything
 		 * other than the descriptor.
 		 */
 		srp_map_desc(state, dma_addr, dma_len, target->rkey);
 		return 0;
 	}
 
-	/* If we start at an offset into the FMR page, don't merge into
-	 * the current FMR. Finish it out, and use the kernel's MR for this
-	 * sg entry. This is to avoid potential bugs on some SRP targets
-	 * that were never quite defined, but went away when the initiator
-	 * avoided using FMR on such page fragments.
+	/*
+	 * Since not all RDMA HW drivers support non-zero page offsets for
+	 * FMR, if we start at an offset into a page, don't merge into the
+	 * current FMR mapping. Finish it out, and use the kernel's MR for
+	 * this sg entry.
 	 */
-	if (dma_addr & ~dev->fmr_page_mask || dma_len > dev->fmr_max_size) {
-		ret = srp_map_finish_fmr(state, target);
+	if ((!dev->use_fast_reg && dma_addr & ~dev->mr_page_mask) ||
+	    dma_len > dev->mr_max_size) {
+		ret = srp_finish_mapping(state, target);
 		if (ret)
 			return ret;
 
@@ -1004,52 +1298,106 @@ static int srp_map_sg_entry(struct srp_map_state *state,
 		return 0;
 	}
 
-	/* If this is the first sg to go into the FMR, save our position.
-	 * We need to know the first unmapped entry, its index, and the
-	 * first unmapped address within that entry to be able to restart
-	 * mapping after an error.
+	/*
+	 * If this is the first sg that will be mapped via FMR or via FR, save
+	 * our position. We need to know the first unmapped entry, its index,
+	 * and the first unmapped address within that entry to be able to
+	 * restart mapping after an error.
 	 */
 	if (!state->unmapped_sg)
 		srp_map_update_start(state, sg, sg_index, dma_addr);
 
 	while (dma_len) {
-		if (state->npages == SRP_FMR_SIZE) {
-			ret = srp_map_finish_fmr(state, target);
+		unsigned offset = dma_addr & ~dev->mr_page_mask;
+		if (state->npages == dev->max_pages_per_mr || offset != 0) {
+			ret = srp_finish_mapping(state, target);
 			if (ret)
 				return ret;
 
 			srp_map_update_start(state, sg, sg_index, dma_addr);
 		}
 
-		len = min_t(unsigned int, dma_len, dev->fmr_page_size);
+		len = min_t(unsigned int, dma_len, dev->mr_page_size - offset);
 
 		if (!state->npages)
 			state->base_dma_addr = dma_addr;
-		state->pages[state->npages++] = dma_addr;
-		state->fmr_len += len;
+		state->pages[state->npages++] = dma_addr & dev->mr_page_mask;
+		state->dma_len += len;
 		dma_addr += len;
 		dma_len -= len;
 	}
 
-	/* If the last entry of the FMR wasn't a full page, then we need to
+	/*
+	 * If the last entry of the MR wasn't a full page, then we need to
 	 * close it out and start a new one -- we can only merge at page
 	 * boundries.
 	 */
 	ret = 0;
-	if (len != dev->fmr_page_size) {
-		ret = srp_map_finish_fmr(state, target);
+	if (len != dev->mr_page_size) {
+		ret = srp_finish_mapping(state, target);
 		if (!ret)
 			srp_map_update_start(state, NULL, 0, 0);
 	}
 	return ret;
 }
 
+static int srp_map_sg(struct srp_map_state *state,
+		      struct srp_target_port *target, struct srp_request *req,
+		      struct scatterlist *scat, int count)
+{
+	struct srp_device *dev = target->srp_host->srp_dev;
+	struct ib_device *ibdev = dev->dev;
+	struct scatterlist *sg;
+	int i;
+	bool use_mr;
+
+	state->desc	= req->indirect_desc;
+	state->pages	= req->map_page;
+	if (dev->use_fast_reg) {
+		state->next_fr = req->fr_list;
+		use_mr = !!target->fr_pool;
+	} else {
+		state->next_fmr = req->fmr_list;
+		use_mr = !!target->fmr_pool;
+	}
+
+	for_each_sg(scat, sg, count, i) {
+		if (srp_map_sg_entry(state, target, sg, i, use_mr)) {
+			/*
+			 * Memory registration failed, so backtrack to the
+			 * first unmapped entry and continue on without using
+			 * memory registration.
+			 */
+			dma_addr_t dma_addr;
+			unsigned int dma_len;
+
+backtrack:
+			sg = state->unmapped_sg;
+			i = state->unmapped_index;
+
+			dma_addr = ib_sg_dma_address(ibdev, sg);
+			dma_len = ib_sg_dma_len(ibdev, sg);
+			dma_len -= (state->unmapped_addr - dma_addr);
+			dma_addr = state->unmapped_addr;
+			use_mr = false;
+			srp_map_desc(state, dma_addr, dma_len, target->rkey);
+		}
+	}
+
+	if (use_mr && srp_finish_mapping(state, target))
+		goto backtrack;
+
+	req->nmdesc = state->nmdesc;
+
+	return 0;
+}
+
 static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_target_port *target,
 			struct srp_request *req)
 {
-	struct scatterlist *scat, *sg;
+	struct scatterlist *scat;
 	struct srp_cmd *cmd = req->cmd->buf;
-	int i, len, nents, count, use_fmr;
+	int len, nents, count;
 	struct srp_device *dev;
 	struct ib_device *ibdev;
 	struct srp_map_state state;
@@ -1081,7 +1429,7 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_target_port *target,
 	fmt = SRP_DATA_DESC_DIRECT;
 	len = sizeof (struct srp_cmd) +	sizeof (struct srp_direct_buf);
 
-	if (count == 1) {
+	if (count == 1 && !register_always) {
 		/*
 		 * The midlayer only generated a single gather/scatter
 		 * entry, or DMA mapping coalesced everything to a
@@ -1094,13 +1442,13 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_target_port *target,
 		buf->key = cpu_to_be32(target->rkey);
 		buf->len = cpu_to_be32(ib_sg_dma_len(ibdev, scat));
 
-		req->nfmr = 0;
+		req->nmdesc = 0;
 		goto map_complete;
 	}
 
-	/* We have more than one scatter/gather entry, so build our indirect
-	 * descriptor table, trying to merge as many entries with FMR as we
-	 * can.
+	/*
+	 * We have more than one scatter/gather entry, so build our indirect
+	 * descriptor table, trying to merge as many entries as we can.
 	 */
 	indirect_hdr = (void *) cmd->add_data;
 
@@ -1108,35 +1456,7 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_target_port *target,
 				   target->indirect_size, DMA_TO_DEVICE);
 
 	memset(&state, 0, sizeof(state));
-	state.desc	= req->indirect_desc;
-	state.pages	= req->map_page;
-	state.next_fmr	= req->fmr_list;
-
-	use_fmr = dev->fmr_pool ? SRP_MAP_ALLOW_FMR : SRP_MAP_NO_FMR;
-
-	for_each_sg(scat, sg, count, i) {
-		if (srp_map_sg_entry(&state, target, sg, i, use_fmr)) {
-			/* FMR mapping failed, so backtrack to the first
-			 * unmapped entry and continue on without using FMR.
-			 */
-			dma_addr_t dma_addr;
-			unsigned int dma_len;
-
-backtrack:
-			sg = state.unmapped_sg;
-			i = state.unmapped_index;
-
-			dma_addr = ib_sg_dma_address(ibdev, sg);
-			dma_len = ib_sg_dma_len(ibdev, sg);
-			dma_len -= (state.unmapped_addr - dma_addr);
-			dma_addr = state.unmapped_addr;
-			use_fmr = SRP_MAP_NO_FMR;
-			srp_map_desc(&state, dma_addr, dma_len, target->rkey);
-		}
-	}
-
-	if (use_fmr == SRP_MAP_ALLOW_FMR && srp_map_finish_fmr(&state, target))
-		goto backtrack;
+	srp_map_sg(&state, target, req, scat, count);
 
 	/* We've mapped the request, now pull as much of the indirect
 	 * descriptor table as we can into the command buffer. If this
@@ -1144,9 +1464,9 @@ backtrack:
 	 * guaranteed to fit into the command, as the SCSI layer won't
 	 * give us more S/G entries than we allow.
 	 */
-	req->nfmr = state.nfmr;
 	if (state.ndesc == 1) {
-		/* FMR mapping was able to collapse this to one entry,
+		/*
+		 * Memory registration collapsed the sg-list into one entry,
 		 * so use a direct descriptor.
 		 */
 		struct srp_direct_buf *buf = (void *) cmd->add_data;
@@ -1455,6 +1775,7 @@ static void srp_handle_recv(struct srp_target_port *target, struct ib_wc *wc)
 
 /**
  * srp_tl_err_work() - handle a transport layer error
+ * @work: Work structure embedded in an SRP target port.
  *
  * Note: This function may get invoked before the rport has been created,
  * hence the target->rport test.
@@ -1468,14 +1789,24 @@ static void srp_tl_err_work(struct work_struct *work)
 		srp_start_tl_fail_timers(target->rport);
 }
 
-static void srp_handle_qp_err(enum ib_wc_status wc_status, bool send_err,
-			      struct srp_target_port *target)
+static void srp_handle_qp_err(u64 wr_id, enum ib_wc_status wc_status,
+			      bool send_err, struct srp_target_port *target)
 {
 	if (target->connected && !target->qp_in_error) {
-		shost_printk(KERN_ERR, target->scsi_host,
-			     PFX "failed %s status %d\n",
-			     send_err ? "send" : "receive",
-			     wc_status);
+		if (wr_id & LOCAL_INV_WR_ID_MASK) {
+			shost_printk(KERN_ERR, target->scsi_host, PFX
+				     "LOCAL_INV failed with status %d\n",
+				     wc_status);
+		} else if (wr_id & FAST_REG_WR_ID_MASK) {
+			shost_printk(KERN_ERR, target->scsi_host, PFX
+				     "FAST_REG_MR failed status %d\n",
+				     wc_status);
+		} else {
+			shost_printk(KERN_ERR, target->scsi_host,
+				     PFX "failed %s status %d for iu %p\n",
+				     send_err ? "send" : "receive",
+				     wc_status, (void *)(uintptr_t)wr_id);
+		}
 		queue_work(system_long_wq, &target->tl_err_work);
 	}
 	target->qp_in_error = true;
@@ -1491,7 +1822,7 @@ static void srp_recv_completion(struct ib_cq *cq, void *target_ptr)
 		if (likely(wc.status == IB_WC_SUCCESS)) {
 			srp_handle_recv(target, &wc);
 		} else {
-			srp_handle_qp_err(wc.status, false, target);
+			srp_handle_qp_err(wc.wr_id, wc.status, false, target);
 		}
 	}
 }
@@ -1507,7 +1838,7 @@ static void srp_send_completion(struct ib_cq *cq, void *target_ptr)
 			iu = (struct srp_iu *) (uintptr_t) wc.wr_id;
 			list_add(&iu->list, &target->free_tx);
 		} else {
-			srp_handle_qp_err(wc.status, true, target);
+			srp_handle_qp_err(wc.wr_id, wc.status, true, target);
 		}
 	}
 }
@@ -1521,7 +1852,7 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 	struct srp_cmd *cmd;
 	struct ib_device *dev;
 	unsigned long flags;
-	int len, result;
+	int len, ret;
 	const bool in_scsi_eh = !in_interrupt() && current == shost->ehandler;
 
 	/*
@@ -1533,12 +1864,9 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 	if (in_scsi_eh)
 		mutex_lock(&rport->mutex);
 
-	result = srp_chkready(target->rport);
-	if (unlikely(result)) {
-		scmnd->result = result;
-		scmnd->scsi_done(scmnd);
-		goto unlock_rport;
-	}
+	scmnd->result = srp_chkready(target->rport);
+	if (unlikely(scmnd->result))
+		goto err;
 
 	spin_lock_irqsave(&target->lock, flags);
 	iu = __srp_get_tx_iu(target, SRP_IU_CMD);
@@ -1553,7 +1881,6 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 	ib_dma_sync_single_for_cpu(dev, iu->dma, target->max_iu_len,
 				   DMA_TO_DEVICE);
 
-	scmnd->result        = 0;
 	scmnd->host_scribble = (void *) req;
 
 	cmd = iu->buf;
@@ -1570,7 +1897,15 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 	len = srp_map_data(scmnd, target, req);
 	if (len < 0) {
 		shost_printk(KERN_ERR, target->scsi_host,
-			     PFX "Failed to map data\n");
+			     PFX "Failed to map data (%d)\n", len);
+		/*
+		 * If we ran out of memory descriptors (-ENOMEM) because an
+		 * application is queuing many requests with more than
+		 * max_pages_per_mr sg-list elements, tell the SCSI mid-layer
+		 * to reduce queue depth temporarily.
+		 */
+		scmnd->result = len == -ENOMEM ?
+			DID_OK << 16 | QUEUE_FULL << 1 : DID_ERROR << 16;
 		goto err_iu;
 	}
 
@@ -1582,11 +1917,13 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 		goto err_unmap;
 	}
 
+	ret = 0;
+
 unlock_rport:
 	if (in_scsi_eh)
 		mutex_unlock(&rport->mutex);
 
-	return 0;
+	return ret;
 
 err_unmap:
 	srp_unmap_data(scmnd, target, req);
@@ -1594,16 +1931,27 @@ err_unmap:
 err_iu:
 	srp_put_tx_iu(target, iu, SRP_IU_CMD);
 
+	/*
+	 * Avoid that the loops that iterate over the request ring can
+	 * encounter a dangling SCSI command pointer.
+	 */
+	req->scmnd = NULL;
+
 	spin_lock_irqsave(&target->lock, flags);
 	list_add(&req->list, &target->free_reqs);
 
 err_unlock:
 	spin_unlock_irqrestore(&target->lock, flags);
 
-	if (in_scsi_eh)
-		mutex_unlock(&rport->mutex);
+err:
+	if (scmnd->result) {
+		scmnd->scsi_done(scmnd);
+		ret = 0;
+	} else {
+		ret = SCSI_MLQUEUE_HOST_BUSY;
+	}
 
-	return SCSI_MLQUEUE_HOST_BUSY;
+	goto unlock_rport;
 }
 
 /*
@@ -2310,6 +2658,8 @@ static struct class srp_class = {
 
 /**
  * srp_conn_unique() - check whether the connection to a target is unique
+ * @host:   SRP host.
+ * @target: SRP target port.
  */
 static bool srp_conn_unique(struct srp_host *host,
 			    struct srp_target_port *target)
@@ -2605,7 +2955,8 @@ static ssize_t srp_create_target(struct device *dev,
 		container_of(dev, struct srp_host, dev);
 	struct Scsi_Host *target_host;
 	struct srp_target_port *target;
-	struct ib_device *ibdev = host->srp_dev->dev;
+	struct srp_device *srp_dev = host->srp_dev;
+	struct ib_device *ibdev = srp_dev->dev;
 	int ret;
 
 	target_host = scsi_host_alloc(&srp_template,
@@ -2650,9 +3001,9 @@ static ssize_t srp_create_target(struct device *dev,
 		goto err;
 	}
 
-	if (!host->srp_dev->fmr_pool && !target->allow_ext_sg &&
-				target->cmd_sg_cnt < target->sg_tablesize) {
-		pr_warn("No FMR pool and no external indirect descriptors, limiting sg_tablesize to cmd_sg_cnt\n");
+	if (!srp_dev->has_fmr && !srp_dev->has_fr && !target->allow_ext_sg &&
+	    target->cmd_sg_cnt < target->sg_tablesize) {
+		pr_warn("No MR pool and no external indirect descriptors, limiting sg_tablesize to cmd_sg_cnt\n");
 		target->sg_tablesize = target->cmd_sg_cnt;
 	}
 
@@ -2790,9 +3141,9 @@ static void srp_add_one(struct ib_device *device)
 {
 	struct srp_device *srp_dev;
 	struct ib_device_attr *dev_attr;
-	struct ib_fmr_pool_param fmr_param;
 	struct srp_host *host;
-	int max_pages_per_fmr, fmr_page_shift, s, e, p;
+	int mr_page_shift, s, e, p;
+	u64 max_pages_per_mr;
 
 	dev_attr = kmalloc(sizeof *dev_attr, GFP_KERNEL);
 	if (!dev_attr)
@@ -2807,15 +3158,39 @@ static void srp_add_one(struct ib_device *device)
 	if (!srp_dev)
 		goto free_attr;
 
+	srp_dev->has_fmr = (device->alloc_fmr && device->dealloc_fmr &&
+			    device->map_phys_fmr && device->unmap_fmr);
+	srp_dev->has_fr = (dev_attr->device_cap_flags &
+			   IB_DEVICE_MEM_MGT_EXTENSIONS);
+	if (!srp_dev->has_fmr && !srp_dev->has_fr)
+		dev_warn(&device->dev, "neither FMR nor FR is supported\n");
+
+	srp_dev->use_fast_reg = (srp_dev->has_fr &&
+				 (!srp_dev->has_fmr || prefer_fr));
+
 	/*
 	 * Use the smallest page size supported by the HCA, down to a
 	 * minimum of 4096 bytes. We're unlikely to build large sglists
 	 * out of smaller entries.
 	 */
-	fmr_page_shift		= max(12, ffs(dev_attr->page_size_cap) - 1);
-	srp_dev->fmr_page_size	= 1 << fmr_page_shift;
-	srp_dev->fmr_page_mask	= ~((u64) srp_dev->fmr_page_size - 1);
-	srp_dev->fmr_max_size	= srp_dev->fmr_page_size * SRP_FMR_SIZE;
+	mr_page_shift		= max(12, ffs(dev_attr->page_size_cap) - 1);
+	srp_dev->mr_page_size	= 1 << mr_page_shift;
+	srp_dev->mr_page_mask	= ~((u64) srp_dev->mr_page_size - 1);
+	max_pages_per_mr	= dev_attr->max_mr_size;
+	do_div(max_pages_per_mr, srp_dev->mr_page_size);
+	srp_dev->max_pages_per_mr = min_t(u64, SRP_MAX_PAGES_PER_MR,
+					  max_pages_per_mr);
+	if (srp_dev->use_fast_reg) {
+		srp_dev->max_pages_per_mr =
+			min_t(u32, srp_dev->max_pages_per_mr,
+			      dev_attr->max_fast_reg_page_list_len);
+	}
+	srp_dev->mr_max_size	= srp_dev->mr_page_size *
+				   srp_dev->max_pages_per_mr;
+	pr_debug("%s: mr_page_shift = %d, dev_attr->max_mr_size = %#llx, dev_attr->max_fast_reg_page_list_len = %u, max_pages_per_mr = %d, mr_max_size = %#x\n",
+		 device->name, mr_page_shift, dev_attr->max_mr_size,
+		 dev_attr->max_fast_reg_page_list_len,
+		 srp_dev->max_pages_per_mr, srp_dev->mr_max_size);
 
 	INIT_LIST_HEAD(&srp_dev->dev_list);
 
@@ -2831,27 +3206,6 @@ static void srp_add_one(struct ib_device *device)
 	if (IS_ERR(srp_dev->mr))
 		goto err_pd;
 
-	for (max_pages_per_fmr = SRP_FMR_SIZE;
-			max_pages_per_fmr >= SRP_FMR_MIN_SIZE;
-			max_pages_per_fmr /= 2, srp_dev->fmr_max_size /= 2) {
-		memset(&fmr_param, 0, sizeof fmr_param);
-		fmr_param.pool_size	    = SRP_FMR_POOL_SIZE;
-		fmr_param.dirty_watermark   = SRP_FMR_DIRTY_SIZE;
-		fmr_param.cache		    = 1;
-		fmr_param.max_pages_per_fmr = max_pages_per_fmr;
-		fmr_param.page_shift	    = fmr_page_shift;
-		fmr_param.access	    = (IB_ACCESS_LOCAL_WRITE |
-					       IB_ACCESS_REMOTE_WRITE |
-					       IB_ACCESS_REMOTE_READ);
-
-		srp_dev->fmr_pool = ib_create_fmr_pool(srp_dev->pd, &fmr_param);
-		if (!IS_ERR(srp_dev->fmr_pool))
-			break;
-	}
-
-	if (IS_ERR(srp_dev->fmr_pool))
-		srp_dev->fmr_pool = NULL;
-
 	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		s = 0;
 		e = 0;
@@ -2914,8 +3268,6 @@ static void srp_remove_one(struct ib_device *device)
 		kfree(host);
 	}
 
-	if (srp_dev->fmr_pool)
-		ib_destroy_fmr_pool(srp_dev->fmr_pool);
 	ib_dereg_mr(srp_dev->mr);
 	ib_dealloc_pd(srp_dev->pd);
 
diff --git a/drivers/infiniband/ulp/srp/ib_srp.h b/drivers/infiniband/ulp/srp/ib_srp.h
index aad27b7b4a46..e46ecb15aa0d 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.h
+++ b/drivers/infiniband/ulp/srp/ib_srp.h
@@ -66,13 +66,10 @@ enum {
 	SRP_TAG_NO_REQ		= ~0U,
 	SRP_TAG_TSK_MGMT	= 1U << 31,
 
-	SRP_FMR_SIZE		= 512,
-	SRP_FMR_MIN_SIZE	= 128,
-	SRP_FMR_POOL_SIZE	= 1024,
-	SRP_FMR_DIRTY_SIZE	= SRP_FMR_POOL_SIZE / 4,
+	SRP_MAX_PAGES_PER_MR	= 512,
 
-	SRP_MAP_ALLOW_FMR	= 0,
-	SRP_MAP_NO_FMR		= 1,
+	LOCAL_INV_WR_ID_MASK	= 1,
+	FAST_REG_WR_ID_MASK	= 2,
 };
 
 enum srp_target_state {
@@ -86,15 +83,24 @@ enum srp_iu_type {
 	SRP_IU_RSP,
 };
 
+/*
+ * @mr_page_mask: HCA memory registration page mask.
+ * @mr_page_size: HCA memory registration page size.
+ * @mr_max_size: Maximum size in bytes of a single FMR / FR registration
+ *   request.
+ */
 struct srp_device {
 	struct list_head	dev_list;
 	struct ib_device       *dev;
 	struct ib_pd	       *pd;
 	struct ib_mr	       *mr;
-	struct ib_fmr_pool     *fmr_pool;
-	u64			fmr_page_mask;
-	int			fmr_page_size;
-	int			fmr_max_size;
+	u64			mr_page_mask;
+	int			mr_page_size;
+	int			mr_max_size;
+	int			max_pages_per_mr;
+	bool			has_fmr;
+	bool			has_fr;
+	bool			use_fast_reg;
 };
 
 struct srp_host {
@@ -112,11 +118,14 @@ struct srp_request {
 	struct list_head	list;
 	struct scsi_cmnd       *scmnd;
 	struct srp_iu	       *cmd;
-	struct ib_pool_fmr    **fmr_list;
+	union {
+		struct ib_pool_fmr **fmr_list;
+		struct srp_fr_desc **fr_list;
+	};
 	u64		       *map_page;
 	struct srp_direct_buf  *indirect_desc;
 	dma_addr_t		indirect_dma_addr;
-	short			nfmr;
+	short			nmdesc;
 	short			index;
 };
 
@@ -131,6 +140,10 @@ struct srp_target_port {
 	struct ib_cq	       *send_cq ____cacheline_aligned_in_smp;
 	struct ib_cq	       *recv_cq;
 	struct ib_qp	       *qp;
+	union {
+		struct ib_fmr_pool     *fmr_pool;
+		struct srp_fr_pool     *fr_pool;
+	};
 	u32			lkey;
 	u32			rkey;
 	enum srp_target_state	state;
@@ -197,15 +210,66 @@ struct srp_iu {
 	enum dma_data_direction	direction;
 };
 
+/**
+ * struct srp_fr_desc - fast registration work request arguments
+ * @entry: Entry in srp_fr_pool.free_list.
+ * @mr:    Memory region.
+ * @frpl:  Fast registration page list.
+ */
+struct srp_fr_desc {
+	struct list_head		entry;
+	struct ib_mr			*mr;
+	struct ib_fast_reg_page_list	*frpl;
+};
+
+/**
+ * struct srp_fr_pool - pool of fast registration descriptors
+ *
+ * An entry is available for allocation if and only if it occurs in @free_list.
+ *
+ * @size:      Number of descriptors in this pool.
+ * @max_page_list_len: Maximum fast registration work request page list length.
+ * @lock:      Protects free_list.
+ * @free_list: List of free descriptors.
+ * @desc:      Fast registration descriptor pool.
+ */
+struct srp_fr_pool {
+	int			size;
+	int			max_page_list_len;
+	spinlock_t		lock;
+	struct list_head	free_list;
+	struct srp_fr_desc	desc[0];
+};
+
+/**
+ * struct srp_map_state - per-request DMA memory mapping state
+ * @desc:	    Pointer to the element of the SRP buffer descriptor array
+ *		    that is being filled in.
+ * @pages:	    Array with DMA addresses of pages being considered for
+ *		    memory registration.
+ * @base_dma_addr:  DMA address of the first page that has not yet been mapped.
+ * @dma_len:	    Number of bytes that will be registered with the next
+ *		    FMR or FR memory registration call.
+ * @total_len:	    Total number of bytes in the sg-list being mapped.
+ * @npages:	    Number of page addresses in the pages[] array.
+ * @nmdesc:	    Number of FMR or FR memory descriptors used for mapping.
+ * @ndesc:	    Number of SRP buffer descriptors that have been filled in.
+ * @unmapped_sg:    First element of the sg-list that is mapped via FMR or FR.
+ * @unmapped_index: Index of the first element mapped via FMR or FR.
+ * @unmapped_addr:  DMA address of the first element mapped via FMR or FR.
+ */
 struct srp_map_state {
-	struct ib_pool_fmr    **next_fmr;
+	union {
+		struct ib_pool_fmr **next_fmr;
+		struct srp_fr_desc **next_fr;
+	};
 	struct srp_direct_buf  *desc;
 	u64		       *pages;
 	dma_addr_t		base_dma_addr;
-	u32			fmr_len;
+	u32			dma_len;
 	u32			total_len;
 	unsigned int		npages;
-	unsigned int		nfmr;
+	unsigned int		nmdesc;
 	unsigned int		ndesc;
 	struct scatterlist     *unmapped_sg;
 	int			unmapped_index;
diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index ce953d895f5b..fd325ec9f064 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -629,12 +629,10 @@ static int str_to_user(const char *str, unsigned int maxlen, void __user *p)
 	return copy_to_user(p, str, len) ? -EFAULT : len;
 }
 
-#define OLD_KEY_MAX	0x1ff
 static int handle_eviocgbit(struct input_dev *dev,
 			    unsigned int type, unsigned int size,
 			    void __user *p, int compat_mode)
 {
-	static unsigned long keymax_warn_time;
 	unsigned long *bits;
 	int len;
 
@@ -652,24 +650,8 @@ static int handle_eviocgbit(struct input_dev *dev,
 	default: return -EINVAL;
 	}
 
-	/*
-	 * Work around bugs in userspace programs that like to do
-	 * EVIOCGBIT(EV_KEY, KEY_MAX) and not realize that 'len'
-	 * should be in bytes, not in bits.
-	 */
-	if (type == EV_KEY && size == OLD_KEY_MAX) {
-		len = OLD_KEY_MAX;
-		if (printk_timed_ratelimit(&keymax_warn_time, 10 * 1000))
-			pr_warning("(EVIOCGBIT): Suspicious buffer size %u, "
-				   "limiting output to %zu bytes. See "
-				   "http://userweb.kernel.org/~dtor/eviocgbit-bug.html\n",
-				   OLD_KEY_MAX,
-				   BITS_TO_LONGS(OLD_KEY_MAX) * sizeof(long));
-	}
-
 	return bits_to_user(bits, len, size, p, compat_mode);
 }
-#undef OLD_KEY_MAX
 
 static int evdev_handle_get_keycode(struct input_dev *dev, void __user *p)
 {
diff --git a/drivers/input/input-polldev.c b/drivers/input/input-polldev.c
index 7f161d93203c..3664f81655ca 100644
--- a/drivers/input/input-polldev.c
+++ b/drivers/input/input-polldev.c
@@ -147,6 +147,11 @@ static struct attribute_group input_polldev_attribute_group = {
 	.attrs = sysfs_attrs
 };
 
+static const struct attribute_group *input_polldev_attribute_groups[] = {
+	&input_polldev_attribute_group,
+	NULL
+};
+
 /**
  * input_allocate_polled_device - allocate memory for polled device
  *
@@ -171,6 +176,91 @@ struct input_polled_dev *input_allocate_polled_device(void)
 }
 EXPORT_SYMBOL(input_allocate_polled_device);
 
+struct input_polled_devres {
+	struct input_polled_dev *polldev;
+};
+
+static int devm_input_polldev_match(struct device *dev, void *res, void *data)
+{
+	struct input_polled_devres *devres = res;
+
+	return devres->polldev == data;
+}
+
+static void devm_input_polldev_release(struct device *dev, void *res)
+{
+	struct input_polled_devres *devres = res;
+	struct input_polled_dev *polldev = devres->polldev;
+
+	dev_dbg(dev, "%s: dropping reference/freeing %s\n",
+		__func__, dev_name(&polldev->input->dev));
+
+	input_put_device(polldev->input);
+	kfree(polldev);
+}
+
+static void devm_input_polldev_unregister(struct device *dev, void *res)
+{
+	struct input_polled_devres *devres = res;
+	struct input_polled_dev *polldev = devres->polldev;
+
+	dev_dbg(dev, "%s: unregistering device %s\n",
+		__func__, dev_name(&polldev->input->dev));
+	input_unregister_device(polldev->input);
+
+	/*
+	 * Note that we are still holding extra reference to the input
+	 * device so it will stick around until devm_input_polldev_release()
+	 * is called.
+	 */
+}
+
+/**
+ * devm_input_allocate_polled_device - allocate managed polled device
+ * @dev: device owning the polled device being created
+ *
+ * Returns prepared &struct input_polled_dev or %NULL.
+ *
+ * Managed polled input devices do not need to be explicitly unregistered
+ * or freed as it will be done automatically when owner device unbinds
+ * from * its driver (or binding fails). Once such managed polled device
+ * is allocated, it is ready to be set up and registered in the same
+ * fashion as regular polled input devices (using
+ * input_register_polled_device() function).
+ *
+ * If you want to manually unregister and free such managed polled devices,
+ * it can be still done by calling input_unregister_polled_device() and
+ * input_free_polled_device(), although it is rarely needed.
+ *
+ * NOTE: the owner device is set up as parent of input device and users
+ * should not override it.
+ */
+struct input_polled_dev *devm_input_allocate_polled_device(struct device *dev)
+{
+	struct input_polled_dev *polldev;
+	struct input_polled_devres *devres;
+
+	devres = devres_alloc(devm_input_polldev_release, sizeof(*devres),
+			      GFP_KERNEL);
+	if (!devres)
+		return NULL;
+
+	polldev = input_allocate_polled_device();
+	if (!polldev) {
+		devres_free(devres);
+		return NULL;
+	}
+
+	polldev->input->dev.parent = dev;
+	polldev->devres_managed = true;
+
+	devres->polldev = polldev;
+	devres_add(dev, devres);
+
+	return polldev;
+}
+EXPORT_SYMBOL(devm_input_allocate_polled_device);
+
 /**
  * input_free_polled_device - free memory allocated for polled device
  * @dev: device to free
@@ -181,7 +271,12 @@ EXPORT_SYMBOL(input_allocate_polled_device);
 void input_free_polled_device(struct input_polled_dev *dev)
 {
 	if (dev) {
-		input_free_device(dev->input);
+		if (dev->devres_managed)
+			WARN_ON(devres_destroy(dev->input->dev.parent,
+						devm_input_polldev_release,
+						devm_input_polldev_match,
+						dev));
+		input_put_device(dev->input);
 		kfree(dev);
 	}
 }
@@ -199,26 +294,35 @@ EXPORT_SYMBOL(input_free_polled_device);
  */
 int input_register_polled_device(struct input_polled_dev *dev)
 {
+	struct input_polled_devres *devres = NULL;
 	struct input_dev *input = dev->input;
 	int error;
 
+	if (dev->devres_managed) {
+		devres = devres_alloc(devm_input_polldev_unregister,
+				      sizeof(*devres), GFP_KERNEL);
+		if (!devres)
+			return -ENOMEM;
+
+		devres->polldev = dev;
+	}
+
 	input_set_drvdata(input, dev);
 	INIT_DELAYED_WORK(&dev->work, input_polled_device_work);
+
 	if (!dev->poll_interval)
 		dev->poll_interval = 500;
 	if (!dev->poll_interval_max)
 		dev->poll_interval_max = dev->poll_interval;
+
 	input->open = input_open_polled_device;
 	input->close = input_close_polled_device;
 
-	error = input_register_device(input);
-	if (error)
-		return error;
+	input->dev.groups = input_polldev_attribute_groups;
 
-	error = sysfs_create_group(&input->dev.kobj,
-				   &input_polldev_attribute_group);
+	error = input_register_device(input);
 	if (error) {
-		input_unregister_device(input);
+		devres_free(devres);
 		return error;
 	}
 
@@ -231,6 +335,12 @@ int input_register_polled_device(struct input_polled_dev *dev)
 	 */
 	input_get_device(input);
 
+	if (dev->devres_managed) {
+		dev_dbg(input->dev.parent, "%s: registering %s with devres.\n",
+			__func__, dev_name(&input->dev));
+		devres_add(input->dev.parent, devres);
+	}
+
 	return 0;
 }
 EXPORT_SYMBOL(input_register_polled_device);
@@ -245,8 +355,11 @@ EXPORT_SYMBOL(input_register_polled_device);
  */
 void input_unregister_polled_device(struct input_polled_dev *dev)
 {
-	sysfs_remove_group(&dev->input->dev.kobj,
-			   &input_polldev_attribute_group);
+	if (dev->devres_managed)
+		WARN_ON(devres_destroy(dev->input->dev.parent,
+					devm_input_polldev_unregister,
+					devm_input_polldev_match,
+					dev));
 
 	input_unregister_device(dev->input);
 }
diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
index ffc7ad3a2c88..f7e79b481349 100644
--- a/drivers/input/keyboard/Kconfig
+++ b/drivers/input/keyboard/Kconfig
@@ -524,6 +524,17 @@ config KEYBOARD_STOWAWAY
 	  To compile this driver as a module, choose M here: the
 	  module will be called stowaway.
 
+config KEYBOARD_ST_KEYSCAN
+	tristate "STMicroelectronics keyscan support"
+	depends on ARCH_STI || COMPILE_TEST
+	select INPUT_MATRIXKMAP
+	help
+	  Say Y here if you want to use a keypad attached to the keyscan block
+	  on some STMicroelectronics SoC devices.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called st-keyscan.
+
 config KEYBOARD_SUNKBD
 	tristate "Sun Type 4 and Type 5 keyboard"
 	select SERIO
@@ -578,7 +589,7 @@ config KEYBOARD_OMAP
 
 config KEYBOARD_OMAP4
 	tristate "TI OMAP4+ keypad support"
-	depends on ARCH_OMAP2PLUS
+	depends on OF || ARCH_OMAP2PLUS
 	select INPUT_MATRIXKMAP
 	help
 	  Say Y here if you want to use the OMAP4+ keypad.
diff --git a/drivers/input/keyboard/Makefile b/drivers/input/keyboard/Makefile
index 11cff7b84b47..7504ae19049d 100644
--- a/drivers/input/keyboard/Makefile
+++ b/drivers/input/keyboard/Makefile
@@ -51,6 +51,7 @@ obj-$(CONFIG_KEYBOARD_SH_KEYSC)		+= sh_keysc.o
 obj-$(CONFIG_KEYBOARD_SPEAR)		+= spear-keyboard.o
 obj-$(CONFIG_KEYBOARD_STMPE)		+= stmpe-keypad.o
 obj-$(CONFIG_KEYBOARD_STOWAWAY)		+= stowaway.o
+obj-$(CONFIG_KEYBOARD_ST_KEYSCAN)	+= st-keyscan.o
 obj-$(CONFIG_KEYBOARD_SUNKBD)		+= sunkbd.o
 obj-$(CONFIG_KEYBOARD_TC3589X)		+= tc3589x-keypad.o
 obj-$(CONFIG_KEYBOARD_TEGRA)		+= tegra-kbc.o
diff --git a/drivers/input/keyboard/adp5520-keys.c b/drivers/input/keyboard/adp5520-keys.c
index 4cc14c2fa7d5..7f4a8b58efc1 100644
--- a/drivers/input/keyboard/adp5520-keys.c
+++ b/drivers/input/keyboard/adp5520-keys.c
@@ -12,6 +12,7 @@
 #include <linux/input.h>
 #include <linux/mfd/adp5520.h>
 #include <linux/slab.h>
+#include <linux/device.h>
 
 struct adp5520_keys {
 	struct input_dev *input;
@@ -81,7 +82,7 @@ static int adp5520_keys_probe(struct platform_device *pdev)
 		return -EINVAL;
 	}
 
-	if (pdata == NULL) {
+	if (!pdata) {
 		dev_err(&pdev->dev, "missing platform data\n");
 		return -EINVAL;
 	}
@@ -89,17 +90,15 @@ static int adp5520_keys_probe(struct platform_device *pdev)
 	if (!(pdata->rows_en_mask && pdata->cols_en_mask))
 		return -EINVAL;
 
-	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
-	if (dev == NULL) {
+	dev = devm_kzalloc(&pdev->dev, sizeof(*dev), GFP_KERNEL);
+	if (!dev) {
 		dev_err(&pdev->dev, "failed to alloc memory\n");
 		return -ENOMEM;
 	}
 
-	input = input_allocate_device();
-	if (!input) {
-		ret = -ENOMEM;
-		goto err;
-	}
+	input = devm_input_allocate_device(&pdev->dev);
+	if (!input)
+		return -ENOMEM;
 
 	dev->master = pdev->dev.parent;
 	dev->input = input;
@@ -135,7 +134,7 @@ static int adp5520_keys_probe(struct platform_device *pdev)
 	ret = input_register_device(input);
 	if (ret) {
 		dev_err(&pdev->dev, "unable to register input device\n");
-		goto err;
+		return ret;
 	}
 
 	en_mask = pdata->rows_en_mask | pdata->cols_en_mask;
@@ -157,8 +156,7 @@ static int adp5520_keys_probe(struct platform_device *pdev)
 
 	if (ret) {
 		dev_err(&pdev->dev, "failed to write\n");
-		ret = -EIO;
-		goto err1;
+		return -EIO;
 	}
 
 	dev->notifier.notifier_call = adp5520_keys_notifier;
@@ -166,19 +164,11 @@ static int adp5520_keys_probe(struct platform_device *pdev)
 			ADP5520_KP_IEN | ADP5520_KR_IEN);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to register notifier\n");
-		goto err1;
+		return ret;
 	}
 
 	platform_set_drvdata(pdev, dev);
 	return 0;
-
-err1:
-	input_unregister_device(input);
-	input = NULL;
-err:
-	input_free_device(input);
-	kfree(dev);
-	return ret;
 }
 
 static int adp5520_keys_remove(struct platform_device *pdev)
@@ -188,8 +178,6 @@ static int adp5520_keys_remove(struct platform_device *pdev)
 	adp5520_unregister_notifier(dev->master, &dev->notifier,
 				ADP5520_KP_IEN | ADP5520_KR_IEN);
 
-	input_unregister_device(dev->input);
-	kfree(dev);
 	return 0;
 }
 
diff --git a/drivers/input/keyboard/clps711x-keypad.c b/drivers/input/keyboard/clps711x-keypad.c
index 3955aecee44b..552b65c6e6b0 100644
--- a/drivers/input/keyboard/clps711x-keypad.c
+++ b/drivers/input/keyboard/clps711x-keypad.c
@@ -185,7 +185,7 @@ static int clps711x_keypad_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static struct of_device_id clps711x_keypad_of_match[] = {
+static const struct of_device_id clps711x_keypad_of_match[] = {
 	{ .compatible = "cirrus,clps711x-keypad", },
 	{ }
 };
diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c
index 2db13246eb8e..8c98e97f8e41 100644
--- a/drivers/input/keyboard/gpio_keys.c
+++ b/drivers/input/keyboard/gpio_keys.c
@@ -424,6 +424,16 @@ out:
 	return IRQ_HANDLED;
 }
 
+static void gpio_keys_quiesce_key(void *data)
+{
+	struct gpio_button_data *bdata = data;
+
+	if (bdata->timer_debounce)
+		del_timer_sync(&bdata->timer);
+
+	cancel_work_sync(&bdata->work);
+}
+
 static int gpio_keys_setup_key(struct platform_device *pdev,
 				struct input_dev *input,
 				struct gpio_button_data *bdata,
@@ -433,7 +443,8 @@ static int gpio_keys_setup_key(struct platform_device *pdev,
 	struct device *dev = &pdev->dev;
 	irq_handler_t isr;
 	unsigned long irqflags;
-	int irq, error;
+	int irq;
+	int error;
 
 	bdata->input = input;
 	bdata->button = button;
@@ -441,7 +452,8 @@ static int gpio_keys_setup_key(struct platform_device *pdev,
 
 	if (gpio_is_valid(button->gpio)) {
 
-		error = gpio_request_one(button->gpio, GPIOF_IN, desc);
+		error = devm_gpio_request_one(&pdev->dev, button->gpio,
+					      GPIOF_IN, desc);
 		if (error < 0) {
 			dev_err(dev, "Failed to request GPIO %d, error %d\n",
 				button->gpio, error);
@@ -463,7 +475,7 @@ static int gpio_keys_setup_key(struct platform_device *pdev,
 			dev_err(dev,
 				"Unable to get irq number for GPIO %d, error %d\n",
 				button->gpio, error);
-			goto fail;
+			return error;
 		}
 		bdata->irq = irq;
 
@@ -497,26 +509,33 @@ static int gpio_keys_setup_key(struct platform_device *pdev,
 	input_set_capability(input, button->type ?: EV_KEY, button->code);
 
 	/*
+	 * Install custom action to cancel debounce timer and
+	 * workqueue item.
+	 */
+	error = devm_add_action(&pdev->dev, gpio_keys_quiesce_key, bdata);
+	if (error) {
+		dev_err(&pdev->dev,
+			"failed to register quiesce action, error: %d\n",
+			error);
+		return error;
+	}
+
+	/*
 	 * If platform has specified that the button can be disabled,
 	 * we don't want it to share the interrupt line.
 	 */
 	if (!button->can_disable)
 		irqflags |= IRQF_SHARED;
 
-	error = request_any_context_irq(bdata->irq, isr, irqflags, desc, bdata);
+	error = devm_request_any_context_irq(&pdev->dev, bdata->irq,
+					     isr, irqflags, desc, bdata);
 	if (error < 0) {
 		dev_err(dev, "Unable to claim irq %d; error %d\n",
 			bdata->irq, error);
-		goto fail;
+		return error;
 	}
 
 	return 0;
-
-fail:
-	if (gpio_is_valid(button->gpio))
-		gpio_free(button->gpio);
-
-	return error;
 }
 
 static void gpio_keys_report_state(struct gpio_keys_drvdata *ddata)
@@ -578,23 +597,18 @@ gpio_keys_get_devtree_pdata(struct device *dev)
 	int i;
 
 	node = dev->of_node;
-	if (!node) {
-		error = -ENODEV;
-		goto err_out;
-	}
+	if (!node)
+		return ERR_PTR(-ENODEV);
 
 	nbuttons = of_get_child_count(node);
-	if (nbuttons == 0) {
-		error = -ENODEV;
-		goto err_out;
-	}
+	if (nbuttons == 0)
+		return ERR_PTR(-ENODEV);
 
-	pdata = kzalloc(sizeof(*pdata) + nbuttons * (sizeof *button),
-			GFP_KERNEL);
-	if (!pdata) {
-		error = -ENOMEM;
-		goto err_out;
-	}
+	pdata = devm_kzalloc(dev,
+			     sizeof(*pdata) + nbuttons * sizeof(*button),
+			     GFP_KERNEL);
+	if (!pdata)
+		return ERR_PTR(-ENOMEM);
 
 	pdata->buttons = (struct gpio_keys_button *)(pdata + 1);
 	pdata->nbuttons = nbuttons;
@@ -619,7 +633,7 @@ gpio_keys_get_devtree_pdata(struct device *dev)
 				dev_err(dev,
 					"Failed to get gpio flags, error: %d\n",
 					error);
-			goto err_free_pdata;
+			return ERR_PTR(error);
 		}
 
 		button = &pdata->buttons[i++];
@@ -630,8 +644,7 @@ gpio_keys_get_devtree_pdata(struct device *dev)
 		if (of_property_read_u32(pp, "linux,code", &button->code)) {
 			dev_err(dev, "Button without keycode: 0x%x\n",
 				button->gpio);
-			error = -EINVAL;
-			goto err_free_pdata;
+			return ERR_PTR(-EINVAL);
 		}
 
 		button->desc = of_get_property(pp, "label", NULL);
@@ -646,20 +659,13 @@ gpio_keys_get_devtree_pdata(struct device *dev)
 			button->debounce_interval = 5;
 	}
 
-	if (pdata->nbuttons == 0) {
-		error = -EINVAL;
-		goto err_free_pdata;
-	}
+	if (pdata->nbuttons == 0)
+		return ERR_PTR(-EINVAL);
 
 	return pdata;
-
-err_free_pdata:
-	kfree(pdata);
-err_out:
-	return ERR_PTR(error);
 }
 
-static struct of_device_id gpio_keys_of_match[] = {
+static const struct of_device_id gpio_keys_of_match[] = {
 	{ .compatible = "gpio-keys", },
 	{ },
 };
@@ -675,22 +681,13 @@ gpio_keys_get_devtree_pdata(struct device *dev)
 
 #endif
 
-static void gpio_remove_key(struct gpio_button_data *bdata)
-{
-	free_irq(bdata->irq, bdata);
-	if (bdata->timer_debounce)
-		del_timer_sync(&bdata->timer);
-	cancel_work_sync(&bdata->work);
-	if (gpio_is_valid(bdata->button->gpio))
-		gpio_free(bdata->button->gpio);
-}
-
 static int gpio_keys_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	const struct gpio_keys_platform_data *pdata = dev_get_platdata(dev);
 	struct gpio_keys_drvdata *ddata;
 	struct input_dev *input;
+	size_t size;
 	int i, error;
 	int wakeup = 0;
 
@@ -700,14 +697,18 @@ static int gpio_keys_probe(struct platform_device *pdev)
 			return PTR_ERR(pdata);
 	}
 
-	ddata = kzalloc(sizeof(struct gpio_keys_drvdata) +
-			pdata->nbuttons * sizeof(struct gpio_button_data),
-			GFP_KERNEL);
-	input = input_allocate_device();
-	if (!ddata || !input) {
+	size = sizeof(struct gpio_keys_drvdata) +
+			pdata->nbuttons * sizeof(struct gpio_button_data);
+	ddata = devm_kzalloc(dev, size, GFP_KERNEL);
+	if (!ddata) {
 		dev_err(dev, "failed to allocate state\n");
-		error = -ENOMEM;
-		goto fail1;
+		return -ENOMEM;
+	}
+
+	input = devm_input_allocate_device(dev);
+	if (!input) {
+		dev_err(dev, "failed to allocate input device\n");
+		return -ENOMEM;
 	}
 
 	ddata->pdata = pdata;
@@ -738,7 +739,7 @@ static int gpio_keys_probe(struct platform_device *pdev)
 
 		error = gpio_keys_setup_key(pdev, input, bdata, button);
 		if (error)
-			goto fail2;
+			return error;
 
 		if (button->wakeup)
 			wakeup = 1;
@@ -748,57 +749,31 @@ static int gpio_keys_probe(struct platform_device *pdev)
 	if (error) {
 		dev_err(dev, "Unable to export keys/switches, error: %d\n",
 			error);
-		goto fail2;
+		return error;
 	}
 
 	error = input_register_device(input);
 	if (error) {
 		dev_err(dev, "Unable to register input device, error: %d\n",
 			error);
-		goto fail3;
+		goto err_remove_group;
 	}
 
 	device_init_wakeup(&pdev->dev, wakeup);
 
 	return 0;
 
- fail3:
+err_remove_group:
 	sysfs_remove_group(&pdev->dev.kobj, &gpio_keys_attr_group);
- fail2:
-	while (--i >= 0)
-		gpio_remove_key(&ddata->data[i]);
-
- fail1:
-	input_free_device(input);
-	kfree(ddata);
-	/* If we have no platform data, we allocated pdata dynamically. */
-	if (!dev_get_platdata(&pdev->dev))
-		kfree(pdata);
-
 	return error;
 }
 
 static int gpio_keys_remove(struct platform_device *pdev)
 {
-	struct gpio_keys_drvdata *ddata = platform_get_drvdata(pdev);
-	struct input_dev *input = ddata->input;
-	int i;
-
 	sysfs_remove_group(&pdev->dev.kobj, &gpio_keys_attr_group);
 
 	device_init_wakeup(&pdev->dev, 0);
 
-	for (i = 0; i < ddata->pdata->nbuttons; i++)
-		gpio_remove_key(&ddata->data[i]);
-
-	input_unregister_device(input);
-
-	/* If we have no platform data, we allocated pdata dynamically. */
-	if (!dev_get_platdata(&pdev->dev))
-		kfree(ddata->pdata);
-
-	kfree(ddata);
-
 	return 0;
 }
 
diff --git a/drivers/input/keyboard/gpio_keys_polled.c b/drivers/input/keyboard/gpio_keys_polled.c
index e571e194ff84..432d36395f35 100644
--- a/drivers/input/keyboard/gpio_keys_polled.c
+++ b/drivers/input/keyboard/gpio_keys_polled.c
@@ -120,12 +120,10 @@ static struct gpio_keys_platform_data *gpio_keys_polled_get_devtree_pdata(struct
 	if (nbuttons == 0)
 		return NULL;
 
-	pdata = kzalloc(sizeof(*pdata) + nbuttons * (sizeof *button),
-			GFP_KERNEL);
-	if (!pdata) {
-		error = -ENOMEM;
-		goto err_out;
-	}
+	pdata = devm_kzalloc(dev, sizeof(*pdata) + nbuttons * sizeof(*button),
+			     GFP_KERNEL);
+	if (!pdata)
+		return ERR_PTR(-ENOMEM);
 
 	pdata->buttons = (struct gpio_keys_button *)(pdata + 1);
 	pdata->nbuttons = nbuttons;
@@ -151,7 +149,7 @@ static struct gpio_keys_platform_data *gpio_keys_polled_get_devtree_pdata(struct
 				dev_err(dev,
 					"Failed to get gpio flags, error: %d\n",
 					error);
-			goto err_free_pdata;
+			return ERR_PTR(error);
 		}
 
 		button = &pdata->buttons[i++];
@@ -162,8 +160,7 @@ static struct gpio_keys_platform_data *gpio_keys_polled_get_devtree_pdata(struct
 		if (of_property_read_u32(pp, "linux,code", &button->code)) {
 			dev_err(dev, "Button without keycode: 0x%x\n",
 				button->gpio);
-			error = -EINVAL;
-			goto err_free_pdata;
+			return ERR_PTR(-EINVAL);
 		}
 
 		button->desc = of_get_property(pp, "label", NULL);
@@ -178,20 +175,13 @@ static struct gpio_keys_platform_data *gpio_keys_polled_get_devtree_pdata(struct
 			button->debounce_interval = 5;
 	}
 
-	if (pdata->nbuttons == 0) {
-		error = -EINVAL;
-		goto err_free_pdata;
-	}
+	if (pdata->nbuttons == 0)
+		return ERR_PTR(-EINVAL);
 
 	return pdata;
-
-err_free_pdata:
-	kfree(pdata);
-err_out:
-	return ERR_PTR(error);
 }
 
-static struct of_device_id gpio_keys_polled_of_match[] = {
+static const struct of_device_id gpio_keys_polled_of_match[] = {
 	{ .compatible = "gpio-keys-polled", },
 	{ },
 };
@@ -213,6 +203,7 @@ static int gpio_keys_polled_probe(struct platform_device *pdev)
 	struct gpio_keys_polled_dev *bdev;
 	struct input_polled_dev *poll_dev;
 	struct input_dev *input;
+	size_t size;
 	int error;
 	int i;
 
@@ -228,24 +219,21 @@ static int gpio_keys_polled_probe(struct platform_device *pdev)
 
 	if (!pdata->poll_interval) {
 		dev_err(dev, "missing poll_interval value\n");
-		error = -EINVAL;
-		goto err_free_pdata;
+		return -EINVAL;
 	}
 
-	bdev = kzalloc(sizeof(struct gpio_keys_polled_dev) +
-		       pdata->nbuttons * sizeof(struct gpio_keys_button_data),
-		       GFP_KERNEL);
+	size = sizeof(struct gpio_keys_polled_dev) +
+			pdata->nbuttons * sizeof(struct gpio_keys_button_data);
+	bdev = devm_kzalloc(&pdev->dev, size, GFP_KERNEL);
 	if (!bdev) {
 		dev_err(dev, "no memory for private data\n");
-		error = -ENOMEM;
-		goto err_free_pdata;
+		return -ENOMEM;
 	}
 
-	poll_dev = input_allocate_polled_device();
+	poll_dev = devm_input_allocate_polled_device(&pdev->dev);
 	if (!poll_dev) {
 		dev_err(dev, "no memory for polled device\n");
-		error = -ENOMEM;
-		goto err_free_bdev;
+		return -ENOMEM;
 	}
 
 	poll_dev->private = bdev;
@@ -258,7 +246,6 @@ static int gpio_keys_polled_probe(struct platform_device *pdev)
 
 	input->name = pdev->name;
 	input->phys = DRV_NAME"/input0";
-	input->dev.parent = &pdev->dev;
 
 	input->id.bustype = BUS_HOST;
 	input->id.vendor = 0x0001;
@@ -277,16 +264,15 @@ static int gpio_keys_polled_probe(struct platform_device *pdev)
 
 		if (button->wakeup) {
 			dev_err(dev, DRV_NAME " does not support wakeup\n");
-			error = -EINVAL;
-			goto err_free_gpio;
+			return -EINVAL;
 		}
 
-		error = gpio_request_one(gpio, GPIOF_IN,
-					 button->desc ?: DRV_NAME);
+		error = devm_gpio_request_one(&pdev->dev, gpio, GPIOF_IN,
+					      button->desc ? : DRV_NAME);
 		if (error) {
 			dev_err(dev, "unable to claim gpio %u, err=%d\n",
 				gpio, error);
-			goto err_free_gpio;
+			return error;
 		}
 
 		bdata->can_sleep = gpio_cansleep(gpio);
@@ -306,7 +292,7 @@ static int gpio_keys_polled_probe(struct platform_device *pdev)
 	if (error) {
 		dev_err(dev, "unable to register polled device, err=%d\n",
 			error);
-		goto err_free_gpio;
+		return error;
 	}
 
 	/* report initial state of the buttons */
@@ -315,52 +301,10 @@ static int gpio_keys_polled_probe(struct platform_device *pdev)
 					     &bdev->data[i]);
 
 	return 0;
-
-err_free_gpio:
-	while (--i >= 0)
-		gpio_free(pdata->buttons[i].gpio);
-
-	input_free_polled_device(poll_dev);
-
-err_free_bdev:
-	kfree(bdev);
-
-err_free_pdata:
-	/* If we have no platform_data, we allocated pdata dynamically.  */
-	if (!dev_get_platdata(&pdev->dev))
-		kfree(pdata);
-
-	return error;
-}
-
-static int gpio_keys_polled_remove(struct platform_device *pdev)
-{
-	struct gpio_keys_polled_dev *bdev = platform_get_drvdata(pdev);
-	const struct gpio_keys_platform_data *pdata = bdev->pdata;
-	int i;
-
-	input_unregister_polled_device(bdev->poll_dev);
-
-	for (i = 0; i < pdata->nbuttons; i++)
-		gpio_free(pdata->buttons[i].gpio);
-
-	input_free_polled_device(bdev->poll_dev);
-
-	/*
-	 * If we had no platform_data, we allocated pdata dynamically and
-	 * must free it here.
-	 */
-	if (!dev_get_platdata(&pdev->dev))
-		kfree(pdata);
-
-	kfree(bdev);
-
-	return 0;
 }
 
 static struct platform_driver gpio_keys_polled_driver = {
 	.probe	= gpio_keys_polled_probe,
-	.remove	= gpio_keys_polled_remove,
 	.driver	= {
 		.name	= DRV_NAME,
 		.owner	= THIS_MODULE,
diff --git a/drivers/input/keyboard/imx_keypad.c b/drivers/input/keyboard/imx_keypad.c
index 97ec33572e56..8280cb16260b 100644
--- a/drivers/input/keyboard/imx_keypad.c
+++ b/drivers/input/keyboard/imx_keypad.c
@@ -415,7 +415,7 @@ open_err:
 }
 
 #ifdef CONFIG_OF
-static struct of_device_id imx_keypad_of_match[] = {
+static const struct of_device_id imx_keypad_of_match[] = {
 	{ .compatible = "fsl,imx21-kpp", },
 	{ /* sentinel */ }
 };
diff --git a/drivers/input/keyboard/jornada680_kbd.c b/drivers/input/keyboard/jornada680_kbd.c
index 69b1f002ff52..0ba4428da24a 100644
--- a/drivers/input/keyboard/jornada680_kbd.c
+++ b/drivers/input/keyboard/jornada680_kbd.c
@@ -16,6 +16,7 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/device.h>
 #include <linux/input.h>
 #include <linux/input-polldev.h>
 #include <linux/interrupt.h>
@@ -185,14 +186,15 @@ static int jornada680kbd_probe(struct platform_device *pdev)
 	struct input_dev *input_dev;
 	int i, error;
 
-	jornadakbd = kzalloc(sizeof(struct jornadakbd), GFP_KERNEL);
+	jornadakbd = devm_kzalloc(&pdev->dev, sizeof(struct jornadakbd),
+				  GFP_KERNEL);
 	if (!jornadakbd)
 		return -ENOMEM;
 
-	poll_dev = input_allocate_polled_device();
+	poll_dev = devm_input_allocate_polled_device(&pdev->dev);
 	if (!poll_dev) {
-		error = -ENOMEM;
-		goto failed;
+		dev_err(&pdev->dev, "failed to allocate polled input device\n");
+		return -ENOMEM;
 	}
 
 	platform_set_drvdata(pdev, jornadakbd);
@@ -224,27 +226,10 @@ static int jornada680kbd_probe(struct platform_device *pdev)
 	input_set_capability(input_dev, EV_MSC, MSC_SCAN);
 
 	error = input_register_polled_device(jornadakbd->poll_dev);
-	if (error)
-		goto failed;
-
-	return 0;
-
- failed:
-	printk(KERN_ERR "Jornadakbd: failed to register driver, error: %d\n",
-		error);
-	input_free_polled_device(poll_dev);
-	kfree(jornadakbd);
-	return error;
-
-}
-
-static int jornada680kbd_remove(struct platform_device *pdev)
-{
-	struct jornadakbd *jornadakbd = platform_get_drvdata(pdev);
-
-	input_unregister_polled_device(jornadakbd->poll_dev);
-	input_free_polled_device(jornadakbd->poll_dev);
-	kfree(jornadakbd);
+	if (error) {
+		dev_err(&pdev->dev, "failed to register polled input device\n");
+		return error;
+	}
 
 	return 0;
 }
@@ -255,7 +240,6 @@ static struct platform_driver jornada680kbd_driver = {
 		.owner	= THIS_MODULE,
 	},
 	.probe	= jornada680kbd_probe,
-	.remove	= jornada680kbd_remove,
 };
 module_platform_driver(jornada680kbd_driver);
 
diff --git a/drivers/input/keyboard/mcs_touchkey.c b/drivers/input/keyboard/mcs_touchkey.c
index 1da8e0b44b56..375b05ca8e2a 100644
--- a/drivers/input/keyboard/mcs_touchkey.c
+++ b/drivers/input/keyboard/mcs_touchkey.c
@@ -147,7 +147,7 @@ static int mcs_touchkey_probe(struct i2c_client *client,
 	}
 	dev_info(&client->dev, "Firmware version: %d\n", fw_ver);
 
-	input_dev->name = "MELPAS MCS Touchkey";
+	input_dev->name = "MELFAS MCS Touchkey";
 	input_dev->id.bustype = BUS_I2C;
 	input_dev->dev.parent = &client->dev;
 	input_dev->evbit[0] = BIT_MASK(EV_KEY);
diff --git a/drivers/input/keyboard/omap4-keypad.c b/drivers/input/keyboard/omap4-keypad.c
index 0400b3f2b4b9..024b7bdffe5b 100644
--- a/drivers/input/keyboard/omap4-keypad.c
+++ b/drivers/input/keyboard/omap4-keypad.c
@@ -28,11 +28,10 @@
 #include <linux/io.h>
 #include <linux/of.h>
 #include <linux/input.h>
+#include <linux/input/matrix_keypad.h>
 #include <linux/slab.h>
 #include <linux/pm_runtime.h>
 
-#include <linux/platform_data/omap4-keypad.h>
-
 /* OMAP4 registers */
 #define OMAP4_KBD_REVISION		0x00
 #define OMAP4_KBD_SYSCONFIG		0x10
@@ -218,7 +217,6 @@ static void omap4_keypad_close(struct input_dev *input)
 	pm_runtime_put_sync(input->dev.parent);
 }
 
-#ifdef CONFIG_OF
 static int omap4_keypad_parse_dt(struct device *dev,
 				 struct omap4_keypad *keypad_data)
 {
@@ -235,20 +233,9 @@ static int omap4_keypad_parse_dt(struct device *dev,
 
 	return 0;
 }
-#else
-static inline int omap4_keypad_parse_dt(struct device *dev,
-					struct omap4_keypad *keypad_data)
-{
-	return -ENOSYS;
-}
-#endif
 
 static int omap4_keypad_probe(struct platform_device *pdev)
 {
-	const struct omap4_keypad_platform_data *pdata =
-				dev_get_platdata(&pdev->dev);
-	const struct matrix_keymap_data *keymap_data =
-				pdata ? pdata->keymap_data : NULL;
 	struct omap4_keypad *keypad_data;
 	struct input_dev *input_dev;
 	struct resource *res;
@@ -277,14 +264,9 @@ static int omap4_keypad_probe(struct platform_device *pdev)
 
 	keypad_data->irq = irq;
 
-	if (pdata) {
-		keypad_data->rows = pdata->rows;
-		keypad_data->cols = pdata->cols;
-	} else {
-		error = omap4_keypad_parse_dt(&pdev->dev, keypad_data);
-		if (error)
-			return error;
-	}
+	error = omap4_keypad_parse_dt(&pdev->dev, keypad_data);
+	if (error)
+		return error;
 
 	res = request_mem_region(res->start, resource_size(res), pdev->name);
 	if (!res) {
@@ -363,7 +345,7 @@ static int omap4_keypad_probe(struct platform_device *pdev)
 		goto err_free_input;
 	}
 
-	error = matrix_keypad_build_keymap(keymap_data, NULL,
+	error = matrix_keypad_build_keymap(NULL, NULL,
 					   keypad_data->rows, keypad_data->cols,
 					   keypad_data->keymap, input_dev);
 	if (error) {
@@ -434,13 +416,11 @@ static int omap4_keypad_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_OF
 static const struct of_device_id omap_keypad_dt_match[] = {
 	{ .compatible = "ti,omap4-keypad" },
 	{},
 };
 MODULE_DEVICE_TABLE(of, omap_keypad_dt_match);
-#endif
 
 #ifdef CONFIG_PM_SLEEP
 static int omap4_keypad_suspend(struct device *dev)
@@ -482,7 +462,7 @@ static struct platform_driver omap4_keypad_driver = {
 		.name	= "omap4-keypad",
 		.owner	= THIS_MODULE,
 		.pm	= &omap4_keypad_pm_ops,
-		.of_match_table = of_match_ptr(omap_keypad_dt_match),
+		.of_match_table = omap_keypad_dt_match,
 	},
 };
 module_platform_driver(omap4_keypad_driver);
diff --git a/drivers/input/keyboard/st-keyscan.c b/drivers/input/keyboard/st-keyscan.c
new file mode 100644
index 000000000000..758b48731415
--- /dev/null
+++ b/drivers/input/keyboard/st-keyscan.c
@@ -0,0 +1,274 @@
+/*
+ * STMicroelectronics Key Scanning driver
+ *
+ * Copyright (c) 2014 STMicroelectonics Ltd.
+ * Author: Stuart Menefy <stuart.menefy@st.com>
+ *
+ * Based on sh_keysc.c, copyright 2008 Magnus Damm
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/input/matrix_keypad.h>
+
+#define ST_KEYSCAN_MAXKEYS 16
+
+#define KEYSCAN_CONFIG_OFF		0x0
+#define KEYSCAN_CONFIG_ENABLE		0x1
+#define KEYSCAN_DEBOUNCE_TIME_OFF	0x4
+#define KEYSCAN_MATRIX_STATE_OFF	0x8
+#define KEYSCAN_MATRIX_DIM_OFF		0xc
+#define KEYSCAN_MATRIX_DIM_X_SHIFT	0x0
+#define KEYSCAN_MATRIX_DIM_Y_SHIFT	0x2
+
+struct st_keyscan {
+	void __iomem *base;
+	int irq;
+	struct clk *clk;
+	struct input_dev *input_dev;
+	unsigned long last_state;
+	unsigned int n_rows;
+	unsigned int n_cols;
+	unsigned int debounce_us;
+};
+
+static irqreturn_t keyscan_isr(int irq, void *dev_id)
+{
+	struct st_keyscan *keypad = dev_id;
+	unsigned short *keycode = keypad->input_dev->keycode;
+	unsigned long state, change;
+	int bit_nr;
+
+	state = readl(keypad->base + KEYSCAN_MATRIX_STATE_OFF) & 0xffff;
+	change = keypad->last_state ^ state;
+	keypad->last_state = state;
+
+	for_each_set_bit(bit_nr, &change, BITS_PER_LONG)
+		input_report_key(keypad->input_dev,
+				 keycode[bit_nr], state & BIT(bit_nr));
+
+	input_sync(keypad->input_dev);
+
+	return IRQ_HANDLED;
+}
+
+static int keyscan_start(struct st_keyscan *keypad)
+{
+	int error;
+
+	error = clk_enable(keypad->clk);
+	if (error)
+		return error;
+
+	writel(keypad->debounce_us * (clk_get_rate(keypad->clk) / 1000000),
+	       keypad->base + KEYSCAN_DEBOUNCE_TIME_OFF);
+
+	writel(((keypad->n_cols - 1) << KEYSCAN_MATRIX_DIM_X_SHIFT) |
+	       ((keypad->n_rows - 1) << KEYSCAN_MATRIX_DIM_Y_SHIFT),
+	       keypad->base + KEYSCAN_MATRIX_DIM_OFF);
+
+	writel(KEYSCAN_CONFIG_ENABLE, keypad->base + KEYSCAN_CONFIG_OFF);
+
+	return 0;
+}
+
+static void keyscan_stop(struct st_keyscan *keypad)
+{
+	writel(0, keypad->base + KEYSCAN_CONFIG_OFF);
+
+	clk_disable(keypad->clk);
+}
+
+static int keyscan_open(struct input_dev *dev)
+{
+	struct st_keyscan *keypad = input_get_drvdata(dev);
+
+	return keyscan_start(keypad);
+}
+
+static void keyscan_close(struct input_dev *dev)
+{
+	struct st_keyscan *keypad = input_get_drvdata(dev);
+
+	keyscan_stop(keypad);
+}
+
+static int keypad_matrix_key_parse_dt(struct st_keyscan *keypad_data)
+{
+	struct device *dev = keypad_data->input_dev->dev.parent;
+	struct device_node *np = dev->of_node;
+	int error;
+
+	error = matrix_keypad_parse_of_params(dev, &keypad_data->n_rows,
+					      &keypad_data->n_cols);
+	if (error) {
+		dev_err(dev, "failed to parse keypad params\n");
+		return error;
+	}
+
+	of_property_read_u32(np, "st,debounce-us", &keypad_data->debounce_us);
+
+	dev_dbg(dev, "n_rows=%d n_col=%d debounce=%d\n",
+		keypad_data->n_rows, keypad_data->n_cols,
+		keypad_data->debounce_us);
+
+	return 0;
+}
+
+static int keyscan_probe(struct platform_device *pdev)
+{
+	struct st_keyscan *keypad_data;
+	struct input_dev *input_dev;
+	struct resource *res;
+	int error;
+
+	if (!pdev->dev.of_node) {
+		dev_err(&pdev->dev, "no DT data present\n");
+		return -EINVAL;
+	}
+
+	keypad_data = devm_kzalloc(&pdev->dev, sizeof(*keypad_data),
+				   GFP_KERNEL);
+	if (!keypad_data)
+		return -ENOMEM;
+
+	input_dev = devm_input_allocate_device(&pdev->dev);
+	if (!input_dev) {
+		dev_err(&pdev->dev, "failed to allocate the input device\n");
+		return -ENOMEM;
+	}
+
+	input_dev->name = pdev->name;
+	input_dev->phys = "keyscan-keys/input0";
+	input_dev->dev.parent = &pdev->dev;
+	input_dev->open = keyscan_open;
+	input_dev->close = keyscan_close;
+
+	input_dev->id.bustype = BUS_HOST;
+
+	error = keypad_matrix_key_parse_dt(keypad_data);
+	if (error)
+		return error;
+
+	error = matrix_keypad_build_keymap(NULL, NULL,
+					   keypad_data->n_rows,
+					   keypad_data->n_cols,
+					   NULL, input_dev);
+	if (error) {
+		dev_err(&pdev->dev, "failed to build keymap\n");
+		return error;
+	}
+
+	input_set_drvdata(input_dev, keypad_data);
+
+	keypad_data->input_dev = input_dev;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	keypad_data->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(keypad_data->base))
+		return PTR_ERR(keypad_data->base);
+
+	keypad_data->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(keypad_data->clk)) {
+		dev_err(&pdev->dev, "cannot get clock\n");
+		return PTR_ERR(keypad_data->clk);
+	}
+
+	error = clk_enable(keypad_data->clk);
+	if (error) {
+		dev_err(&pdev->dev, "failed to enable clock\n");
+		return error;
+	}
+
+	keyscan_stop(keypad_data);
+
+	keypad_data->irq = platform_get_irq(pdev, 0);
+	if (keypad_data->irq < 0) {
+		dev_err(&pdev->dev, "no IRQ specified\n");
+		return -EINVAL;
+	}
+
+	error = devm_request_irq(&pdev->dev, keypad_data->irq, keyscan_isr, 0,
+				 pdev->name, keypad_data);
+	if (error) {
+		dev_err(&pdev->dev, "failed to request IRQ\n");
+		return error;
+	}
+
+	error = input_register_device(input_dev);
+	if (error) {
+		dev_err(&pdev->dev, "failed to register input device\n");
+		return error;
+	}
+
+	platform_set_drvdata(pdev, keypad_data);
+
+	device_set_wakeup_capable(&pdev->dev, 1);
+
+	return 0;
+}
+
+static int keyscan_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct st_keyscan *keypad = platform_get_drvdata(pdev);
+	struct input_dev *input = keypad->input_dev;
+
+	mutex_lock(&input->mutex);
+
+	if (device_may_wakeup(dev))
+		enable_irq_wake(keypad->irq);
+	else if (input->users)
+		keyscan_stop(keypad);
+
+	mutex_unlock(&input->mutex);
+	return 0;
+}
+
+static int keyscan_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct st_keyscan *keypad = platform_get_drvdata(pdev);
+	struct input_dev *input = keypad->input_dev;
+	int retval = 0;
+
+	mutex_lock(&input->mutex);
+
+	if (device_may_wakeup(dev))
+		disable_irq_wake(keypad->irq);
+	else if (input->users)
+		retval = keyscan_start(keypad);
+
+	mutex_unlock(&input->mutex);
+	return retval;
+}
+
+static SIMPLE_DEV_PM_OPS(keyscan_dev_pm_ops, keyscan_suspend, keyscan_resume);
+
+static const struct of_device_id keyscan_of_match[] = {
+	{ .compatible = "st,sti-keyscan" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, keyscan_of_match);
+
+static struct platform_driver keyscan_device_driver = {
+	.probe		= keyscan_probe,
+	.driver		= {
+		.name	= "st-keyscan",
+		.pm	= &keyscan_dev_pm_ops,
+		.of_match_table = of_match_ptr(keyscan_of_match),
+	}
+};
+
+module_platform_driver(keyscan_device_driver);
+
+MODULE_AUTHOR("Stuart Menefy <stuart.menefy@st.com>");
+MODULE_DESCRIPTION("STMicroelectronics keyscan device driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/input/keyboard/tc3589x-keypad.c b/drivers/input/keyboard/tc3589x-keypad.c
index 74494a357522..ad7abae69078 100644
--- a/drivers/input/keyboard/tc3589x-keypad.c
+++ b/drivers/input/keyboard/tc3589x-keypad.c
@@ -296,6 +296,65 @@ static void tc3589x_keypad_close(struct input_dev *input)
 	tc3589x_keypad_disable(keypad);
 }
 
+#ifdef CONFIG_OF
+static const struct tc3589x_keypad_platform_data *
+tc3589x_keypad_of_probe(struct device *dev)
+{
+	struct device_node *np = dev->of_node;
+	struct tc3589x_keypad_platform_data *plat;
+	u32 cols, rows;
+	u32 debounce_ms;
+	int proplen;
+
+	if (!np)
+		return ERR_PTR(-ENODEV);
+
+	plat = devm_kzalloc(dev, sizeof(*plat), GFP_KERNEL);
+	if (!plat)
+		return ERR_PTR(-ENOMEM);
+
+	of_property_read_u32(np, "keypad,num-columns", &cols);
+	of_property_read_u32(np, "keypad,num-rows", &rows);
+	plat->kcol = (u8) cols;
+	plat->krow = (u8) rows;
+	if (!plat->krow || !plat->kcol ||
+	     plat->krow > TC_KPD_ROWS || plat->kcol > TC_KPD_COLUMNS) {
+		dev_err(dev,
+			"keypad columns/rows not properly specified (%ux%u)\n",
+			plat->kcol, plat->krow);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (!of_get_property(np, "linux,keymap", &proplen)) {
+		dev_err(dev, "property linux,keymap not found\n");
+		return ERR_PTR(-ENOENT);
+	}
+
+	plat->no_autorepeat = of_property_read_bool(np, "linux,no-autorepeat");
+	plat->enable_wakeup = of_property_read_bool(np, "linux,wakeup");
+
+	/* The custom delay format is ms/16 */
+	of_property_read_u32(np, "debounce-delay-ms", &debounce_ms);
+	if (debounce_ms)
+		plat->debounce_period = debounce_ms * 16;
+	else
+		plat->debounce_period = TC_KPD_DEBOUNCE_PERIOD;
+
+	plat->settle_time = TC_KPD_SETTLE_TIME;
+	/* FIXME: should be property of the IRQ resource? */
+	plat->irqtype = IRQF_TRIGGER_FALLING;
+
+	return plat;
+}
+#else
+static inline const struct tc3589x_keypad_platform_data *
+tc3589x_keypad_of_probe(struct device *dev)
+{
+	return ERR_PTR(-ENODEV);
+}
+#endif
+
+
 static int tc3589x_keypad_probe(struct platform_device *pdev)
 {
 	struct tc3589x *tc3589x = dev_get_drvdata(pdev->dev.parent);
@@ -306,8 +365,11 @@ static int tc3589x_keypad_probe(struct platform_device *pdev)
 
 	plat = tc3589x->pdata->keypad;
 	if (!plat) {
-		dev_err(&pdev->dev, "invalid keypad platform data\n");
-		return -EINVAL;
+		plat = tc3589x_keypad_of_probe(&pdev->dev);
+		if (IS_ERR(plat)) {
+			dev_err(&pdev->dev, "invalid keypad platform data\n");
+			return PTR_ERR(plat);
+		}
 	}
 
 	irq = platform_get_irq(pdev, 0);
diff --git a/drivers/input/misc/88pm860x_onkey.c b/drivers/input/misc/88pm860x_onkey.c
index abd8453e5212..220ce0fa15d9 100644
--- a/drivers/input/misc/88pm860x_onkey.c
+++ b/drivers/input/misc/88pm860x_onkey.c
@@ -26,6 +26,7 @@
 #include <linux/interrupt.h>
 #include <linux/mfd/88pm860x.h>
 #include <linux/slab.h>
+#include <linux/device.h>
 
 #define PM8607_WAKEUP		0x0b
 
@@ -68,7 +69,8 @@ static int pm860x_onkey_probe(struct platform_device *pdev)
 		return -EINVAL;
 	}
 
-	info = kzalloc(sizeof(struct pm860x_onkey_info), GFP_KERNEL);
+	info = devm_kzalloc(&pdev->dev, sizeof(struct pm860x_onkey_info),
+			    GFP_KERNEL);
 	if (!info)
 		return -ENOMEM;
 	info->chip = chip;
@@ -76,11 +78,10 @@ static int pm860x_onkey_probe(struct platform_device *pdev)
 	info->dev = &pdev->dev;
 	info->irq = irq;
 
-	info->idev = input_allocate_device();
+	info->idev = devm_input_allocate_device(&pdev->dev);
 	if (!info->idev) {
 		dev_err(chip->dev, "Failed to allocate input dev\n");
-		ret = -ENOMEM;
-		goto out;
+		return -ENOMEM;
 	}
 
 	info->idev->name = "88pm860x_on";
@@ -93,42 +94,22 @@ static int pm860x_onkey_probe(struct platform_device *pdev)
 	ret = input_register_device(info->idev);
 	if (ret) {
 		dev_err(chip->dev, "Can't register input device: %d\n", ret);
-		goto out_reg;
+		return ret;
 	}
 
-	ret = request_threaded_irq(info->irq, NULL, pm860x_onkey_handler,
-				   IRQF_ONESHOT, "onkey", info);
+	ret = devm_request_threaded_irq(&pdev->dev, info->irq, NULL,
+					pm860x_onkey_handler, IRQF_ONESHOT,
+					"onkey", info);
 	if (ret < 0) {
 		dev_err(chip->dev, "Failed to request IRQ: #%d: %d\n",
 			info->irq, ret);
-		goto out_irq;
+		return ret;
 	}
 
 	platform_set_drvdata(pdev, info);
 	device_init_wakeup(&pdev->dev, 1);
 
 	return 0;
-
-out_irq:
-	input_unregister_device(info->idev);
-	kfree(info);
-	return ret;
-
-out_reg:
-	input_free_device(info->idev);
-out:
-	kfree(info);
-	return ret;
-}
-
-static int pm860x_onkey_remove(struct platform_device *pdev)
-{
-	struct pm860x_onkey_info *info = platform_get_drvdata(pdev);
-
-	free_irq(info->irq, info);
-	input_unregister_device(info->idev);
-	kfree(info);
-	return 0;
 }
 
 #ifdef CONFIG_PM_SLEEP
@@ -161,7 +142,6 @@ static struct platform_driver pm860x_onkey_driver = {
 		.pm	= &pm860x_onkey_pm_ops,
 	},
 	.probe		= pm860x_onkey_probe,
-	.remove		= pm860x_onkey_remove,
 };
 module_platform_driver(pm860x_onkey_driver);
 
diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig
index 5928ea71dd69..2ff4425a893b 100644
--- a/drivers/input/misc/Kconfig
+++ b/drivers/input/misc/Kconfig
@@ -224,7 +224,7 @@ config INPUT_GP2A
 
 config INPUT_GPIO_BEEPER
 	tristate "Generic GPIO Beeper support"
-	depends on OF_GPIO
+	depends on GPIOLIB
 	help
 	  Say Y here if you have a beeper connected to a GPIO pin.
 
diff --git a/drivers/input/misc/ab8500-ponkey.c b/drivers/input/misc/ab8500-ponkey.c
index f2fbdd88ed20..95ef7dd6442d 100644
--- a/drivers/input/misc/ab8500-ponkey.c
+++ b/drivers/input/misc/ab8500-ponkey.c
@@ -7,6 +7,7 @@
  * AB8500 Power-On Key handler
  */
 
+#include <linux/device.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/platform_device.h>
@@ -65,12 +66,14 @@ static int ab8500_ponkey_probe(struct platform_device *pdev)
 		return irq_dbr;
 	}
 
-	ponkey = kzalloc(sizeof(struct ab8500_ponkey), GFP_KERNEL);
-	input = input_allocate_device();
-	if (!ponkey || !input) {
-		error = -ENOMEM;
-		goto err_free_mem;
-	}
+	ponkey = devm_kzalloc(&pdev->dev, sizeof(struct ab8500_ponkey),
+			      GFP_KERNEL);
+	if (!ponkey)
+		return -ENOMEM;
+
+	input = devm_input_allocate_device(&pdev->dev);
+	if (!input)
+		return -ENOMEM;
 
 	ponkey->idev = input;
 	ponkey->ab8500 = ab8500;
@@ -82,52 +85,32 @@ static int ab8500_ponkey_probe(struct platform_device *pdev)
 
 	input_set_capability(input, EV_KEY, KEY_POWER);
 
-	error = request_any_context_irq(ponkey->irq_dbf, ab8500_ponkey_handler,
-					0, "ab8500-ponkey-dbf", ponkey);
+	error = devm_request_any_context_irq(&pdev->dev, ponkey->irq_dbf,
+					     ab8500_ponkey_handler, 0,
+					     "ab8500-ponkey-dbf", ponkey);
 	if (error < 0) {
 		dev_err(ab8500->dev, "Failed to request dbf IRQ#%d: %d\n",
 			ponkey->irq_dbf, error);
-		goto err_free_mem;
+		return error;
 	}
 
-	error = request_any_context_irq(ponkey->irq_dbr, ab8500_ponkey_handler,
-					0, "ab8500-ponkey-dbr", ponkey);
+	error = devm_request_any_context_irq(&pdev->dev, ponkey->irq_dbr,
+					     ab8500_ponkey_handler, 0,
+					     "ab8500-ponkey-dbr", ponkey);
 	if (error < 0) {
 		dev_err(ab8500->dev, "Failed to request dbr IRQ#%d: %d\n",
 			ponkey->irq_dbr, error);
-		goto err_free_dbf_irq;
+		return error;
 	}
 
 	error = input_register_device(ponkey->idev);
 	if (error) {
 		dev_err(ab8500->dev, "Can't register input device: %d\n", error);
-		goto err_free_dbr_irq;
+		return error;
 	}
 
 	platform_set_drvdata(pdev, ponkey);
 	return 0;
-
-err_free_dbr_irq:
-	free_irq(ponkey->irq_dbr, ponkey);
-err_free_dbf_irq:
-	free_irq(ponkey->irq_dbf, ponkey);
-err_free_mem:
-	input_free_device(input);
-	kfree(ponkey);
-
-	return error;
-}
-
-static int ab8500_ponkey_remove(struct platform_device *pdev)
-{
-	struct ab8500_ponkey *ponkey = platform_get_drvdata(pdev);
-
-	free_irq(ponkey->irq_dbf, ponkey);
-	free_irq(ponkey->irq_dbr, ponkey);
-	input_unregister_device(ponkey->idev);
-	kfree(ponkey);
-
-	return 0;
 }
 
 #ifdef CONFIG_OF
@@ -144,7 +127,6 @@ static struct platform_driver ab8500_ponkey_driver = {
 		.of_match_table = of_match_ptr(ab8500_ponkey_match),
 	},
 	.probe		= ab8500_ponkey_probe,
-	.remove		= ab8500_ponkey_remove,
 };
 module_platform_driver(ab8500_ponkey_driver);
 
diff --git a/drivers/input/misc/gpio-beeper.c b/drivers/input/misc/gpio-beeper.c
index b757435e2b3d..8886af63eae3 100644
--- a/drivers/input/misc/gpio-beeper.c
+++ b/drivers/input/misc/gpio-beeper.c
@@ -1,7 +1,7 @@
 /*
  * Generic GPIO beeper driver
  *
- * Copyright (C) 2013 Alexander Shiyan <shc_work@mail.ru>
+ * Copyright (C) 2013-2014 Alexander Shiyan <shc_work@mail.ru>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -11,7 +11,8 @@
 
 #include <linux/input.h>
 #include <linux/module.h>
-#include <linux/of_gpio.h>
+#include <linux/gpio/consumer.h>
+#include <linux/of.h>
 #include <linux/workqueue.h>
 #include <linux/platform_device.h>
 
@@ -19,14 +20,13 @@
 
 struct gpio_beeper {
 	struct work_struct	work;
-	int			gpio;
-	bool			active_low;
+	struct gpio_desc	*desc;
 	bool			beeping;
 };
 
 static void gpio_beeper_toggle(struct gpio_beeper *beep, bool on)
 {
-	gpio_set_value_cansleep(beep->gpio, on ^ beep->active_low);
+	gpiod_set_value_cansleep(beep->desc, on);
 }
 
 static void gpio_beeper_work(struct work_struct *work)
@@ -65,18 +65,16 @@ static void gpio_beeper_close(struct input_dev *input)
 static int gpio_beeper_probe(struct platform_device *pdev)
 {
 	struct gpio_beeper *beep;
-	enum of_gpio_flags flags;
 	struct input_dev *input;
-	unsigned long gflags;
 	int err;
 
 	beep = devm_kzalloc(&pdev->dev, sizeof(*beep), GFP_KERNEL);
 	if (!beep)
 		return -ENOMEM;
 
-	beep->gpio = of_get_gpio_flags(pdev->dev.of_node, 0, &flags);
-	if (!gpio_is_valid(beep->gpio))
-		return beep->gpio;
+	beep->desc = devm_gpiod_get(&pdev->dev, NULL);
+	if (IS_ERR(beep->desc))
+		return PTR_ERR(beep->desc);
 
 	input = devm_input_allocate_device(&pdev->dev);
 	if (!input)
@@ -94,10 +92,7 @@ static int gpio_beeper_probe(struct platform_device *pdev)
 
 	input_set_capability(input, EV_SND, SND_BELL);
 
-	beep->active_low = flags & OF_GPIO_ACTIVE_LOW;
-	gflags = beep->active_low ? GPIOF_OUT_INIT_HIGH : GPIOF_OUT_INIT_LOW;
-
-	err = devm_gpio_request_one(&pdev->dev, beep->gpio, gflags, pdev->name);
+	err = gpiod_direction_output(beep->desc, 0);
 	if (err)
 		return err;
 
@@ -106,17 +101,19 @@ static int gpio_beeper_probe(struct platform_device *pdev)
 	return input_register_device(input);
 }
 
-static struct of_device_id gpio_beeper_of_match[] = {
+#ifdef CONFIG_OF
+static const struct of_device_id gpio_beeper_of_match[] = {
 	{ .compatible = BEEPER_MODNAME, },
 	{ }
 };
 MODULE_DEVICE_TABLE(of, gpio_beeper_of_match);
+#endif
 
 static struct platform_driver gpio_beeper_platform_driver = {
 	.driver	= {
 		.name		= BEEPER_MODNAME,
 		.owner		= THIS_MODULE,
-		.of_match_table	= gpio_beeper_of_match,
+		.of_match_table	= of_match_ptr(gpio_beeper_of_match),
 	},
 	.probe	= gpio_beeper_probe,
 };
diff --git a/drivers/input/misc/ims-pcu.c b/drivers/input/misc/ims-pcu.c
index 5a736397d9c8..719410feb84b 100644
--- a/drivers/input/misc/ims-pcu.c
+++ b/drivers/input/misc/ims-pcu.c
@@ -1566,6 +1566,7 @@ static int ims_pcu_buffers_alloc(struct ims_pcu *pcu)
 	if (!pcu->urb_ctrl_buf) {
 		dev_err(pcu->dev,
 			"Failed to allocate memory for read buffer\n");
+		error = -ENOMEM;
 		goto err_free_urb_out_buf;
 	}
 
diff --git a/drivers/input/misc/max8925_onkey.c b/drivers/input/misc/max8925_onkey.c
index eef41cfc054d..3809618e6a5d 100644
--- a/drivers/input/misc/max8925_onkey.c
+++ b/drivers/input/misc/max8925_onkey.c
@@ -26,6 +26,7 @@
 #include <linux/interrupt.h>
 #include <linux/mfd/max8925.h>
 #include <linux/slab.h>
+#include <linux/device.h>
 
 #define SW_INPUT		(1 << 7)	/* 0/1 -- up/down */
 #define HARDRESET_EN		(1 << 7)
@@ -81,12 +82,14 @@ static int max8925_onkey_probe(struct platform_device *pdev)
 		return -EINVAL;
 	}
 
-	info = kzalloc(sizeof(struct max8925_onkey_info), GFP_KERNEL);
-	input = input_allocate_device();
-	if (!info || !input) {
-		error = -ENOMEM;
-		goto err_free_mem;
-	}
+	info = devm_kzalloc(&pdev->dev, sizeof(struct max8925_onkey_info),
+			    GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+
+	input = devm_input_allocate_device(&pdev->dev);
+	if (!input)
+		return -ENOMEM;
 
 	info->idev = input;
 	info->i2c = chip->i2c;
@@ -100,55 +103,34 @@ static int max8925_onkey_probe(struct platform_device *pdev)
 	input->dev.parent = &pdev->dev;
 	input_set_capability(input, EV_KEY, KEY_POWER);
 
-	error = request_threaded_irq(irq[0], NULL, max8925_onkey_handler,
-				     IRQF_ONESHOT, "onkey-down", info);
+	error = devm_request_threaded_irq(&pdev->dev, irq[0], NULL,
+					  max8925_onkey_handler, IRQF_ONESHOT,
+					  "onkey-down", info);
 	if (error < 0) {
 		dev_err(chip->dev, "Failed to request IRQ: #%d: %d\n",
 			irq[0], error);
-		goto err_free_mem;
+		return error;
 	}
 
-	error = request_threaded_irq(irq[1], NULL, max8925_onkey_handler,
-				     IRQF_ONESHOT, "onkey-up", info);
+	error = devm_request_threaded_irq(&pdev->dev, irq[1], NULL,
+					  max8925_onkey_handler, IRQF_ONESHOT,
+					  "onkey-up", info);
 	if (error < 0) {
 		dev_err(chip->dev, "Failed to request IRQ: #%d: %d\n",
 			irq[1], error);
-		goto err_free_irq0;
+		return error;
 	}
 
 	error = input_register_device(info->idev);
 	if (error) {
 		dev_err(chip->dev, "Can't register input device: %d\n", error);
-		goto err_free_irq1;
+		return error;
 	}
 
 	platform_set_drvdata(pdev, info);
 	device_init_wakeup(&pdev->dev, 1);
 
 	return 0;
-
-err_free_irq1:
-	free_irq(irq[1], info);
-err_free_irq0:
-	free_irq(irq[0], info);
-err_free_mem:
-	input_free_device(input);
-	kfree(info);
-
-	return error;
-}
-
-static int max8925_onkey_remove(struct platform_device *pdev)
-{
-	struct max8925_onkey_info *info = platform_get_drvdata(pdev);
-	struct max8925_chip *chip = dev_get_drvdata(pdev->dev.parent);
-
-	free_irq(info->irq[0] + chip->irq_base, info);
-	free_irq(info->irq[1] + chip->irq_base, info);
-	input_unregister_device(info->idev);
-	kfree(info);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM_SLEEP
@@ -190,7 +172,6 @@ static struct platform_driver max8925_onkey_driver = {
 		.pm	= &max8925_onkey_pm_ops,
 	},
 	.probe		= max8925_onkey_probe,
-	.remove		= max8925_onkey_remove,
 };
 module_platform_driver(max8925_onkey_driver);
 
diff --git a/drivers/input/misc/max8997_haptic.c b/drivers/input/misc/max8997_haptic.c
index 1fea5484941f..a363ebbd9cc0 100644
--- a/drivers/input/misc/max8997_haptic.c
+++ b/drivers/input/misc/max8997_haptic.c
@@ -181,11 +181,21 @@ static void max8997_haptic_enable(struct max8997_haptic *chip)
 	}
 
 	if (!chip->enabled) {
-		chip->enabled = true;
-		regulator_enable(chip->regulator);
+		error = regulator_enable(chip->regulator);
+		if (error) {
+			dev_err(chip->dev, "Failed to enable regulator\n");
+			goto out;
+		}
 		max8997_haptic_configure(chip);
-		if (chip->mode == MAX8997_EXTERNAL_MODE)
-			pwm_enable(chip->pwm);
+		if (chip->mode == MAX8997_EXTERNAL_MODE) {
+			error = pwm_enable(chip->pwm);
+			if (error) {
+				dev_err(chip->dev, "Failed to enable PWM\n");
+				regulator_disable(chip->regulator);
+				goto out;
+			}
+		}
+		chip->enabled = true;
 	}
 
 out:
diff --git a/drivers/input/misc/pmic8xxx-pwrkey.c b/drivers/input/misc/pmic8xxx-pwrkey.c
index 1cb8fda7a166..c91e3d33aea9 100644
--- a/drivers/input/misc/pmic8xxx-pwrkey.c
+++ b/drivers/input/misc/pmic8xxx-pwrkey.c
@@ -92,15 +92,15 @@ static int pmic8xxx_pwrkey_probe(struct platform_device *pdev)
 	bool pull_up;
 
 	if (of_property_read_u32(pdev->dev.of_node, "debounce", &kpd_delay))
-		kpd_delay = 0;
+		kpd_delay = 15625;
 
-	pull_up = of_property_read_bool(pdev->dev.of_node, "pull-up");
-
-	if (kpd_delay > 62500) {
+	if (kpd_delay > 62500 || kpd_delay == 0) {
 		dev_err(&pdev->dev, "invalid power key trigger delay\n");
 		return -EINVAL;
 	}
 
+	pull_up = of_property_read_bool(pdev->dev.of_node, "pull-up");
+
 	regmap = dev_get_regmap(pdev->dev.parent, NULL);
 	if (!regmap) {
 		dev_err(&pdev->dev, "failed to locate regmap for the device\n");
diff --git a/drivers/input/misc/rotary_encoder.c b/drivers/input/misc/rotary_encoder.c
index 99b9e42aa748..93558a1c7f70 100644
--- a/drivers/input/misc/rotary_encoder.c
+++ b/drivers/input/misc/rotary_encoder.c
@@ -143,7 +143,7 @@ static irqreturn_t rotary_encoder_half_period_irq(int irq, void *dev_id)
 }
 
 #ifdef CONFIG_OF
-static struct of_device_id rotary_encoder_of_match[] = {
+static const struct of_device_id rotary_encoder_of_match[] = {
 	{ .compatible = "rotary-encoder", },
 	{ },
 };
diff --git a/drivers/input/misc/soc_button_array.c b/drivers/input/misc/soc_button_array.c
index 20c80f543d5e..5a6334be30b8 100644
--- a/drivers/input/misc/soc_button_array.c
+++ b/drivers/input/misc/soc_button_array.c
@@ -17,7 +17,6 @@
 #include <linux/acpi.h>
 #include <linux/gpio/consumer.h>
 #include <linux/gpio_keys.h>
-#include <linux/input.h>
 #include <linux/platform_device.h>
 #include <linux/pnp.h>
 
diff --git a/drivers/input/misc/twl6040-vibra.c b/drivers/input/misc/twl6040-vibra.c
index 77dc23b94eb1..6d26eecc278c 100644
--- a/drivers/input/misc/twl6040-vibra.c
+++ b/drivers/input/misc/twl6040-vibra.c
@@ -262,7 +262,7 @@ static int twl6040_vibra_probe(struct platform_device *pdev)
 	struct vibra_info *info;
 	int vddvibl_uV = 0;
 	int vddvibr_uV = 0;
-	int ret;
+	int error;
 
 	twl6040_core_node = of_find_node_by_name(twl6040_core_dev->of_node,
 						 "vibra");
@@ -309,12 +309,12 @@ static int twl6040_vibra_probe(struct platform_device *pdev)
 
 	mutex_init(&info->mutex);
 
-	ret = devm_request_threaded_irq(&pdev->dev, info->irq, NULL,
-					twl6040_vib_irq_handler, 0,
-					"twl6040_irq_vib", info);
-	if (ret) {
-		dev_err(info->dev, "VIB IRQ request failed: %d\n", ret);
-		return ret;
+	error = devm_request_threaded_irq(&pdev->dev, info->irq, NULL,
+					  twl6040_vib_irq_handler, 0,
+					  "twl6040_irq_vib", info);
+	if (error) {
+		dev_err(info->dev, "VIB IRQ request failed: %d\n", error);
+		return error;
 	}
 
 	info->supplies[0].supply = "vddvibl";
@@ -323,40 +323,40 @@ static int twl6040_vibra_probe(struct platform_device *pdev)
 	 * When booted with Device tree the regulators are attached to the
 	 * parent device (twl6040 MFD core)
 	 */
-	ret = regulator_bulk_get(twl6040_core_dev, ARRAY_SIZE(info->supplies),
-				 info->supplies);
-	if (ret) {
-		dev_err(info->dev, "couldn't get regulators %d\n", ret);
-		return ret;
+	error = devm_regulator_bulk_get(twl6040_core_dev,
+					ARRAY_SIZE(info->supplies),
+					info->supplies);
+	if (error) {
+		dev_err(info->dev, "couldn't get regulators %d\n", error);
+		return error;
 	}
 
 	if (vddvibl_uV) {
-		ret = regulator_set_voltage(info->supplies[0].consumer,
-					    vddvibl_uV, vddvibl_uV);
-		if (ret) {
+		error = regulator_set_voltage(info->supplies[0].consumer,
+					      vddvibl_uV, vddvibl_uV);
+		if (error) {
 			dev_err(info->dev, "failed to set VDDVIBL volt %d\n",
-				ret);
-			goto err_regulator;
+				error);
+			return error;
 		}
 	}
 
 	if (vddvibr_uV) {
-		ret = regulator_set_voltage(info->supplies[1].consumer,
-					    vddvibr_uV, vddvibr_uV);
-		if (ret) {
+		error = regulator_set_voltage(info->supplies[1].consumer,
+					      vddvibr_uV, vddvibr_uV);
+		if (error) {
 			dev_err(info->dev, "failed to set VDDVIBR volt %d\n",
-				ret);
-			goto err_regulator;
+				error);
+			return error;
 		}
 	}
 
 	INIT_WORK(&info->play_work, vibra_play_work);
 
-	info->input_dev = input_allocate_device();
-	if (info->input_dev == NULL) {
+	info->input_dev = devm_input_allocate_device(&pdev->dev);
+	if (!info->input_dev) {
 		dev_err(info->dev, "couldn't allocate input device\n");
-		ret = -ENOMEM;
-		goto err_regulator;
+		return -ENOMEM;
 	}
 
 	input_set_drvdata(info->input_dev, info);
@@ -367,44 +367,25 @@ static int twl6040_vibra_probe(struct platform_device *pdev)
 	info->input_dev->close = twl6040_vibra_close;
 	__set_bit(FF_RUMBLE, info->input_dev->ffbit);
 
-	ret = input_ff_create_memless(info->input_dev, NULL, vibra_play);
-	if (ret < 0) {
+	error = input_ff_create_memless(info->input_dev, NULL, vibra_play);
+	if (error) {
 		dev_err(info->dev, "couldn't register vibrator to FF\n");
-		goto err_ialloc;
+		return error;
 	}
 
-	ret = input_register_device(info->input_dev);
-	if (ret < 0) {
+	error = input_register_device(info->input_dev);
+	if (error) {
 		dev_err(info->dev, "couldn't register input device\n");
-		goto err_iff;
+		return error;
 	}
 
 	platform_set_drvdata(pdev, info);
 
 	return 0;
-
-err_iff:
-	input_ff_destroy(info->input_dev);
-err_ialloc:
-	input_free_device(info->input_dev);
-err_regulator:
-	regulator_bulk_free(ARRAY_SIZE(info->supplies), info->supplies);
-	return ret;
-}
-
-static int twl6040_vibra_remove(struct platform_device *pdev)
-{
-	struct vibra_info *info = platform_get_drvdata(pdev);
-
-	input_unregister_device(info->input_dev);
-	regulator_bulk_free(ARRAY_SIZE(info->supplies), info->supplies);
-
-	return 0;
 }
 
 static struct platform_driver twl6040_vibra_driver = {
 	.probe		= twl6040_vibra_probe,
-	.remove		= twl6040_vibra_remove,
 	.driver		= {
 		.name	= "twl6040-vibra",
 		.owner	= THIS_MODULE,
diff --git a/drivers/input/mouse/Kconfig b/drivers/input/mouse/Kconfig
index 6b8441f7bc32..366fc7ad5eb6 100644
--- a/drivers/input/mouse/Kconfig
+++ b/drivers/input/mouse/Kconfig
@@ -53,7 +53,7 @@ config MOUSE_PS2_LOGIPS2PP
 	default y
 	depends on MOUSE_PS2
 	help
-	  Say Y here if you have a Logictech PS/2++ mouse connected to
+	  Say Y here if you have a Logitech PS/2++ mouse connected to
 	  your system.
 
 	  If unsure, say Y.
diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c
index b96e978a37b7..ee2a04d90d20 100644
--- a/drivers/input/mouse/elantech.c
+++ b/drivers/input/mouse/elantech.c
@@ -473,8 +473,15 @@ static void elantech_report_absolute_v3(struct psmouse *psmouse,
 	input_report_key(dev, BTN_TOOL_FINGER, fingers == 1);
 	input_report_key(dev, BTN_TOOL_DOUBLETAP, fingers == 2);
 	input_report_key(dev, BTN_TOOL_TRIPLETAP, fingers == 3);
-	input_report_key(dev, BTN_LEFT, packet[0] & 0x01);
-	input_report_key(dev, BTN_RIGHT, packet[0] & 0x02);
+
+	/* For clickpads map both buttons to BTN_LEFT */
+	if (etd->fw_version & 0x001000) {
+		input_report_key(dev, BTN_LEFT, packet[0] & 0x03);
+	} else {
+		input_report_key(dev, BTN_LEFT, packet[0] & 0x01);
+		input_report_key(dev, BTN_RIGHT, packet[0] & 0x02);
+	}
+
 	input_report_abs(dev, ABS_PRESSURE, pres);
 	input_report_abs(dev, ABS_TOOL_WIDTH, width);
 
@@ -484,10 +491,17 @@ static void elantech_report_absolute_v3(struct psmouse *psmouse,
 static void elantech_input_sync_v4(struct psmouse *psmouse)
 {
 	struct input_dev *dev = psmouse->dev;
+	struct elantech_data *etd = psmouse->private;
 	unsigned char *packet = psmouse->packet;
 
-	input_report_key(dev, BTN_LEFT, packet[0] & 0x01);
-	input_report_key(dev, BTN_RIGHT, packet[0] & 0x02);
+	/* For clickpads map both buttons to BTN_LEFT */
+	if (etd->fw_version & 0x001000) {
+		input_report_key(dev, BTN_LEFT, packet[0] & 0x03);
+	} else {
+		input_report_key(dev, BTN_LEFT, packet[0] & 0x01);
+		input_report_key(dev, BTN_RIGHT, packet[0] & 0x02);
+	}
+
 	input_mt_report_pointer_emulation(dev, true);
 	input_sync(dev);
 }
@@ -835,7 +849,7 @@ static int elantech_set_absolute_mode(struct psmouse *psmouse)
 		if (etd->set_hw_resolution)
 			etd->reg_10 = 0x0b;
 		else
-			etd->reg_10 = 0x03;
+			etd->reg_10 = 0x01;
 
 		if (elantech_write_reg(psmouse, 0x10, etd->reg_10))
 			rc = -1;
@@ -1336,7 +1350,8 @@ static int elantech_reconnect(struct psmouse *psmouse)
 }
 
 /*
- * Some hw_version 3 models go into error state when we try to set bit 3 of r10
+ * Some hw_version 3 models go into error state when we try to set
+ * bit 3 and/or bit 1 of r10.
  */
 static const struct dmi_system_id no_hw_res_dmi_table[] = {
 #if defined(CONFIG_DMI) && defined(CONFIG_X86)
diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c
index c5ec703c727e..ec772d962f06 100644
--- a/drivers/input/mouse/synaptics.c
+++ b/drivers/input/mouse/synaptics.c
@@ -347,15 +347,6 @@ static int synaptics_resolution(struct psmouse *psmouse)
 	unsigned char resp[3];
 	int i;
 
-	for (i = 0; min_max_pnpid_table[i].pnp_ids; i++)
-		if (matches_pnp_id(psmouse, min_max_pnpid_table[i].pnp_ids)) {
-			priv->x_min = min_max_pnpid_table[i].x_min;
-			priv->x_max = min_max_pnpid_table[i].x_max;
-			priv->y_min = min_max_pnpid_table[i].y_min;
-			priv->y_max = min_max_pnpid_table[i].y_max;
-			return 0;
-		}
-
 	if (SYN_ID_MAJOR(priv->identity) < 4)
 		return 0;
 
@@ -366,6 +357,16 @@ static int synaptics_resolution(struct psmouse *psmouse)
 		}
 	}
 
+	for (i = 0; min_max_pnpid_table[i].pnp_ids; i++) {
+		if (matches_pnp_id(psmouse, min_max_pnpid_table[i].pnp_ids)) {
+			priv->x_min = min_max_pnpid_table[i].x_min;
+			priv->x_max = min_max_pnpid_table[i].x_max;
+			priv->y_min = min_max_pnpid_table[i].y_min;
+			priv->y_max = min_max_pnpid_table[i].y_max;
+			return 0;
+		}
+	}
+
 	if (SYN_EXT_CAP_REQUESTS(priv->capabilities) >= 5 &&
 	    SYN_CAP_MAX_DIMENSIONS(priv->ext_cap_0c)) {
 		if (synaptics_send_cmd(psmouse, SYN_QUE_EXT_MAX_COORDS, resp)) {
diff --git a/drivers/input/serio/apbps2.c b/drivers/input/serio/apbps2.c
index 17e01a807ddc..98be824544a5 100644
--- a/drivers/input/serio/apbps2.c
+++ b/drivers/input/serio/apbps2.c
@@ -203,7 +203,7 @@ static int apbps2_of_remove(struct platform_device *of_dev)
 	return 0;
 }
 
-static struct of_device_id apbps2_of_match[] = {
+static const struct of_device_id apbps2_of_match[] = {
 	{ .name = "GAISLER_APBPS2", },
 	{ .name = "01_060", },
 	{}
diff --git a/drivers/input/serio/olpc_apsp.c b/drivers/input/serio/olpc_apsp.c
index 5d2fe7ece7ca..d906f3ebc8c8 100644
--- a/drivers/input/serio/olpc_apsp.c
+++ b/drivers/input/serio/olpc_apsp.c
@@ -262,7 +262,7 @@ static int olpc_apsp_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static struct of_device_id olpc_apsp_dt_ids[] = {
+static const struct of_device_id olpc_apsp_dt_ids[] = {
 	{ .compatible = "olpc,ap-sp", },
 	{}
 };
diff --git a/drivers/input/tablet/wacom_sys.c b/drivers/input/tablet/wacom_sys.c
index 611fc3905d00..2c613cd41dd6 100644
--- a/drivers/input/tablet/wacom_sys.c
+++ b/drivers/input/tablet/wacom_sys.c
@@ -349,6 +349,7 @@ static int wacom_parse_hid(struct usb_interface *intf,
 						break;
 
 					case MTTPC:
+					case MTTPC_B:
 						features->pktlen = WACOM_PKGLEN_MTTPC;
 						break;
 
@@ -380,6 +381,16 @@ static int wacom_parse_hid(struct usb_interface *intf,
 						i += 12;
 						break;
 
+					case MTTPC_B:
+						features->x_max =
+							get_unaligned_le16(&report[i + 3]);
+						features->x_phy =
+							get_unaligned_le16(&report[i + 6]);
+						features->unit = report[i - 5];
+						features->unitExpo = report[i - 3];
+						i += 9;
+						break;
+
 					default:
 						features->x_max =
 							get_unaligned_le16(&report[i + 3]);
@@ -430,6 +441,14 @@ static int wacom_parse_hid(struct usb_interface *intf,
 						i += 12;
 						break;
 
+					case MTTPC_B:
+						features->y_max =
+							get_unaligned_le16(&report[i + 3]);
+						features->y_phy =
+							get_unaligned_le16(&report[i + 6]);
+						i += 9;
+						break;
+
 					default:
 						features->y_max =
 							features->x_max;
diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c
index 4822c57a3756..977d05cd9e2e 100644
--- a/drivers/input/tablet/wacom_wac.c
+++ b/drivers/input/tablet/wacom_wac.c
@@ -484,6 +484,8 @@ static int wacom_intuos_inout(struct wacom_wac *wacom)
 		input_report_key(input, BTN_TOUCH, 0);
 		input_report_abs(input, ABS_PRESSURE, 0);
 		input_report_abs(input, ABS_DISTANCE, wacom->features.distance_max);
+		if (features->quirks & WACOM_QUIRK_MULTI_INPUT)
+			wacom->shared->stylus_in_proximity = true;
 	}
 
 	/* Exit report */
@@ -928,12 +930,12 @@ static int wacom_24hdt_irq(struct wacom_wac *wacom)
 		input_mt_report_slot_state(input, MT_TOOL_FINGER, touch);
 
 		if (touch) {
-			int t_x = le16_to_cpup((__le16 *)&data[offset + 2]);
-			int c_x = le16_to_cpup((__le16 *)&data[offset + 4]);
-			int t_y = le16_to_cpup((__le16 *)&data[offset + 6]);
-			int c_y = le16_to_cpup((__le16 *)&data[offset + 8]);
-			int w = le16_to_cpup((__le16 *)&data[offset + 10]);
-			int h = le16_to_cpup((__le16 *)&data[offset + 12]);
+			int t_x = get_unaligned_le16(&data[offset + 2]);
+			int c_x = get_unaligned_le16(&data[offset + 4]);
+			int t_y = get_unaligned_le16(&data[offset + 6]);
+			int c_y = get_unaligned_le16(&data[offset + 8]);
+			int w = get_unaligned_le16(&data[offset + 10]);
+			int h = get_unaligned_le16(&data[offset + 12]);
 
 			input_report_abs(input, ABS_MT_POSITION_X, t_x);
 			input_report_abs(input, ABS_MT_POSITION_Y, t_y);
@@ -962,7 +964,7 @@ static int wacom_mt_touch(struct wacom_wac *wacom)
 	int x_offset = 0;
 
 	/* MTTPC does not support Height and Width */
-	if (wacom->features.type == MTTPC)
+	if (wacom->features.type == MTTPC || wacom->features.type == MTTPC_B)
 		x_offset = -4;
 
 	/*
@@ -978,7 +980,7 @@ static int wacom_mt_touch(struct wacom_wac *wacom)
 	for (i = 0; i < contacts_to_send; i++) {
 		int offset = (WACOM_BYTES_PER_MT_PACKET + x_offset) * i + 3;
 		bool touch = data[offset] & 0x1;
-		int id = le16_to_cpup((__le16 *)&data[offset + 1]);
+		int id = get_unaligned_le16(&data[offset + 1]);
 		int slot = input_mt_get_slot_by_key(input, id);
 
 		if (slot < 0)
@@ -987,8 +989,8 @@ static int wacom_mt_touch(struct wacom_wac *wacom)
 		input_mt_slot(input, slot);
 		input_mt_report_slot_state(input, MT_TOOL_FINGER, touch);
 		if (touch) {
-			int x = le16_to_cpup((__le16 *)&data[offset + x_offset + 7]);
-			int y = le16_to_cpup((__le16 *)&data[offset + x_offset + 9]);
+			int x = get_unaligned_le16(&data[offset + x_offset + 7]);
+			int y = get_unaligned_le16(&data[offset + x_offset + 9]);
 			input_report_abs(input, ABS_MT_POSITION_X, x);
 			input_report_abs(input, ABS_MT_POSITION_Y, y);
 		}
@@ -1047,6 +1049,10 @@ static int wacom_tpc_single_touch(struct wacom_wac *wacom, size_t len)
 			prox = data[0] & 0x01;
 			x = get_unaligned_le16(&data[1]);
 			y = get_unaligned_le16(&data[3]);
+		} else if (len == WACOM_PKGLEN_TPC1FG_B) {
+			prox = data[2] & 0x01;
+			x = get_unaligned_le16(&data[3]);
+			y = get_unaligned_le16(&data[5]);
 		} else {
 			prox = data[1] & 0x01;
 			x = le16_to_cpup((__le16 *)&data[2]);
@@ -1110,6 +1116,9 @@ static int wacom_tpc_irq(struct wacom_wac *wacom, size_t len)
 	case WACOM_PKGLEN_TPC2FG:
 		return wacom_tpc_mt_touch(wacom);
 
+	case WACOM_PKGLEN_PENABLED:
+		return wacom_tpc_pen(wacom);
+
 	default:
 		switch (data[0]) {
 		case WACOM_REPORT_TPC1FG:
@@ -1119,6 +1128,7 @@ static int wacom_tpc_irq(struct wacom_wac *wacom, size_t len)
 			return wacom_tpc_single_touch(wacom, len);
 
 		case WACOM_REPORT_TPCMT:
+		case WACOM_REPORT_TPCMT2:
 			return wacom_mt_touch(wacom);
 
 		case WACOM_REPORT_PENABLED:
@@ -1461,6 +1471,7 @@ void wacom_wac_irq(struct wacom_wac *wacom_wac, size_t len)
 	case TABLETPC2FG:
 	case MTSCREEN:
 	case MTTPC:
+	case MTTPC_B:
 		sync = wacom_tpc_irq(wacom_wac, len);
 		break;
 
@@ -1565,10 +1576,10 @@ static void wacom_abs_set_axis(struct input_dev *input_dev,
 	struct wacom_features *features = &wacom_wac->features;
 
 	if (features->device_type == BTN_TOOL_PEN) {
-		input_set_abs_params(input_dev, ABS_X, 0, features->x_max,
-				     features->x_fuzz, 0);
-		input_set_abs_params(input_dev, ABS_Y, 0, features->y_max,
-				     features->y_fuzz, 0);
+		input_set_abs_params(input_dev, ABS_X, features->x_min,
+				     features->x_max, features->x_fuzz, 0);
+		input_set_abs_params(input_dev, ABS_Y, features->y_min,
+				     features->y_max, features->y_fuzz, 0);
 		input_set_abs_params(input_dev, ABS_PRESSURE, 0,
 			features->pressure_max, features->pressure_fuzz, 0);
 
@@ -1802,6 +1813,7 @@ int wacom_setup_input_capabilities(struct input_dev *input_dev,
 
 	case MTSCREEN:
 	case MTTPC:
+	case MTTPC_B:
 	case TABLETPC2FG:
 		if (features->device_type == BTN_TOOL_FINGER) {
 			unsigned int flags = INPUT_MT_DIRECT;
@@ -2123,11 +2135,11 @@ static const struct wacom_features wacom_features_0x317 =
 	  63, INTUOSPL, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
 	  .touch_max = 16 };
 static const struct wacom_features wacom_features_0xF4 =
-	{ "Wacom Cintiq 24HD",       WACOM_PKGLEN_INTUOS,   104480, 65600, 2047,
-	  63, WACOM_24HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
+	{ "Wacom Cintiq 24HD",       WACOM_PKGLEN_INTUOS,   104280, 65400, 2047,
+	  63, WACOM_24HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200 };
 static const struct wacom_features wacom_features_0xF8 =
-	{ "Wacom Cintiq 24HD touch", WACOM_PKGLEN_INTUOS,   104480, 65600, 2047, /* Pen */
-	  63, WACOM_24HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	{ "Wacom Cintiq 24HD touch", WACOM_PKGLEN_INTUOS,   104280, 65400, 2047, /* Pen */
+	  63, WACOM_24HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200,
 	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0xf6 };
 static const struct wacom_features wacom_features_0xF6 =
 	{ "Wacom Cintiq 24HD touch", .type = WACOM_24HDT, /* Touch */
@@ -2142,8 +2154,8 @@ static const struct wacom_features wacom_features_0xC6 =
 	{ "Wacom Cintiq 12WX",    WACOM_PKGLEN_INTUOS,    53020, 33440, 1023,
 	  63, WACOM_BEE, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
 static const struct wacom_features wacom_features_0x304 =
-	{ "Wacom Cintiq 13HD",    WACOM_PKGLEN_INTUOS,    59552, 33848, 1023,
-	  63, WACOM_13HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
+	{ "Wacom Cintiq 13HD",    WACOM_PKGLEN_INTUOS,    59352, 33648, 1023,
+	  63, WACOM_13HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200 };
 static const struct wacom_features wacom_features_0xC7 =
 	{ "Wacom DTU1931",        WACOM_PKGLEN_GRAPHIRE,  37832, 30305,  511,
 	  0, PL, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
@@ -2157,24 +2169,24 @@ static const struct wacom_features wacom_features_0xFB =
 	{ "Wacom DTU1031",        WACOM_PKGLEN_DTUS,      22096, 13960,  511,
 	  0, DTUS, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
 static const struct wacom_features wacom_features_0x57 =
-	{ "Wacom DTK2241",        WACOM_PKGLEN_INTUOS,    95840, 54260, 2047,
-	  63, DTK, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES};
+	{ "Wacom DTK2241",        WACOM_PKGLEN_INTUOS,    95640, 54060, 2047,
+	  63, DTK, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200 };
 static const struct wacom_features wacom_features_0x59 = /* Pen */
-	{ "Wacom DTH2242",        WACOM_PKGLEN_INTUOS,    95840, 54260, 2047,
-	  63, DTK, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	{ "Wacom DTH2242",        WACOM_PKGLEN_INTUOS,    95640, 54060, 2047,
+	  63, DTK, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200,
 	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0x5D };
 static const struct wacom_features wacom_features_0x5D = /* Touch */
 	{ "Wacom DTH2242",       .type = WACOM_24HDT,
 	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0x59, .touch_max = 10 };
 static const struct wacom_features wacom_features_0xCC =
-	{ "Wacom Cintiq 21UX2",   WACOM_PKGLEN_INTUOS,    87200, 65600, 2047,
-	  63, WACOM_21UX2, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
+	{ "Wacom Cintiq 21UX2",   WACOM_PKGLEN_INTUOS,    87000, 65400, 2047,
+	  63, WACOM_21UX2, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200 };
 static const struct wacom_features wacom_features_0xFA =
-	{ "Wacom Cintiq 22HD",    WACOM_PKGLEN_INTUOS,    95840, 54260, 2047,
-	  63, WACOM_22HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
+	{ "Wacom Cintiq 22HD",    WACOM_PKGLEN_INTUOS,    95640, 54060, 2047,
+	  63, WACOM_22HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200 };
 static const struct wacom_features wacom_features_0x5B =
-	{ "Wacom Cintiq 22HDT", WACOM_PKGLEN_INTUOS,      95840, 54260, 2047,
-	  63, WACOM_22HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	{ "Wacom Cintiq 22HDT", WACOM_PKGLEN_INTUOS,      95640, 54060, 2047,
+	  63, WACOM_22HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200,
 	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0x5e };
 static const struct wacom_features wacom_features_0x5E =
 	{ "Wacom Cintiq 22HDT", .type = WACOM_24HDT,
@@ -2233,9 +2245,21 @@ static const struct wacom_features wacom_features_0x10E =
 static const struct wacom_features wacom_features_0x10F =
 	{ "Wacom ISDv4 10F",      WACOM_PKGLEN_MTTPC,     27760, 15694,  255,
 	  0, MTTPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
+static const struct wacom_features wacom_features_0x116 =
+	{ "Wacom ISDv4 116",      WACOM_PKGLEN_GRAPHIRE,  26202, 16325,  255,
+	  0, TABLETPCE, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
 static const struct wacom_features wacom_features_0x4001 =
 	{ "Wacom ISDv4 4001",      WACOM_PKGLEN_MTTPC,     26202, 16325,  255,
 	  0, MTTPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
+static const struct wacom_features wacom_features_0x4004 =
+	{ "Wacom ISDv4 4004",      WACOM_PKGLEN_MTTPC,     11060, 6220,  255,
+	  0, MTTPC_B, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
+static const struct wacom_features wacom_features_0x5000 =
+	{ "Wacom ISDv4 5000",      WACOM_PKGLEN_MTTPC,     27848, 15752,  1023,
+	  0, MTTPC_B, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
+static const struct wacom_features wacom_features_0x5002 =
+	{ "Wacom ISDv4 5002",      WACOM_PKGLEN_MTTPC,     29576, 16724,  1023,
+	  0, MTTPC_B, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
 static const struct wacom_features wacom_features_0x47 =
 	{ "Wacom Intuos2 6x8",    WACOM_PKGLEN_INTUOS,    20320, 16240, 1023,
 	  31, INTUOS, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
@@ -2316,8 +2340,8 @@ static const struct wacom_features wacom_features_0x6004 =
 	{ "ISD-V4",               WACOM_PKGLEN_GRAPHIRE,  12800,  8000,  255,
 	  0, TABLETPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
 static const struct wacom_features wacom_features_0x0307 =
-	{ "Wacom ISDv5 307", WACOM_PKGLEN_INTUOS,  59552,  33848, 2047,
-	  63, CINTIQ_HYBRID, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	{ "Wacom ISDv5 307", WACOM_PKGLEN_INTUOS,  59352,  33648, 2047,
+	  63, CINTIQ_HYBRID, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES, 200, 200,
 	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0x309 };
 static const struct wacom_features wacom_features_0x0309 =
 	{ "Wacom ISDv5 309", .type = WACOM_24HDT, /* Touch */
@@ -2447,6 +2471,7 @@ const struct usb_device_id wacom_ids[] = {
 	{ USB_DEVICE_WACOM(0x10D) },
 	{ USB_DEVICE_WACOM(0x10E) },
 	{ USB_DEVICE_WACOM(0x10F) },
+	{ USB_DEVICE_WACOM(0x116) },
 	{ USB_DEVICE_WACOM(0x300) },
 	{ USB_DEVICE_WACOM(0x301) },
 	{ USB_DEVICE_DETAILED(0x302, USB_CLASS_HID, 0, 0) },
@@ -2457,6 +2482,9 @@ const struct usb_device_id wacom_ids[] = {
 	{ USB_DEVICE_DETAILED(0x315, USB_CLASS_HID, 0, 0) },
 	{ USB_DEVICE_DETAILED(0x317, USB_CLASS_HID, 0, 0) },
 	{ USB_DEVICE_WACOM(0x4001) },
+	{ USB_DEVICE_WACOM(0x4004) },
+	{ USB_DEVICE_WACOM(0x5000) },
+	{ USB_DEVICE_WACOM(0x5002) },
 	{ USB_DEVICE_WACOM(0x47) },
 	{ USB_DEVICE_WACOM(0xF4) },
 	{ USB_DEVICE_WACOM(0xF8) },
diff --git a/drivers/input/tablet/wacom_wac.h b/drivers/input/tablet/wacom_wac.h
index f69c0ebe7fa9..b2c9a9c1b551 100644
--- a/drivers/input/tablet/wacom_wac.h
+++ b/drivers/input/tablet/wacom_wac.h
@@ -22,6 +22,7 @@
 #define WACOM_PKGLEN_BBFUN	 9
 #define WACOM_PKGLEN_INTUOS	10
 #define WACOM_PKGLEN_TPC1FG	 5
+#define WACOM_PKGLEN_TPC1FG_B	10
 #define WACOM_PKGLEN_TPC2FG	14
 #define WACOM_PKGLEN_BBTOUCH	20
 #define WACOM_PKGLEN_BBTOUCH3	64
@@ -30,6 +31,7 @@
 #define WACOM_PKGLEN_MTOUCH	62
 #define WACOM_PKGLEN_MTTPC	40
 #define WACOM_PKGLEN_DTUS	68
+#define WACOM_PKGLEN_PENABLED	 8
 
 /* wacom data size per MT contact */
 #define WACOM_BYTES_PER_MT_PACKET	11
@@ -52,6 +54,7 @@
 #define WACOM_REPORT_TPC1FG		6
 #define WACOM_REPORT_TPC2FG		13
 #define WACOM_REPORT_TPCMT		13
+#define WACOM_REPORT_TPCMT2		3
 #define WACOM_REPORT_TPCHID		15
 #define WACOM_REPORT_TPCST		16
 #define WACOM_REPORT_DTUS		17
@@ -105,6 +108,7 @@ enum {
 	TABLETPC2FG,
 	MTSCREEN,
 	MTTPC,
+	MTTPC_B,
 	MAX_TYPE
 };
 
@@ -118,6 +122,8 @@ struct wacom_features {
 	int type;
 	int x_resolution;
 	int y_resolution;
+	int x_min;
+	int y_min;
 	int device_type;
 	int x_phy;
 	int y_phy;
diff --git a/drivers/input/touchscreen/88pm860x-ts.c b/drivers/input/touchscreen/88pm860x-ts.c
index 544e20c551f8..0d4a9fad4a78 100644
--- a/drivers/input/touchscreen/88pm860x-ts.c
+++ b/drivers/input/touchscreen/88pm860x-ts.c
@@ -16,6 +16,7 @@
 #include <linux/input.h>
 #include <linux/mfd/88pm860x.h>
 #include <linux/slab.h>
+#include <linux/device.h>
 
 #define MEAS_LEN		(8)
 #define ACCURATE_BIT		(12)
@@ -234,16 +235,17 @@ static int pm860x_touch_probe(struct platform_device *pdev)
 	if (ret)
 		return ret;
 
-	touch = kzalloc(sizeof(struct pm860x_touch), GFP_KERNEL);
-	if (touch == NULL)
+	touch = devm_kzalloc(&pdev->dev, sizeof(struct pm860x_touch),
+			     GFP_KERNEL);
+	if (!touch)
 		return -ENOMEM;
+
 	platform_set_drvdata(pdev, touch);
 
-	touch->idev = input_allocate_device();
-	if (touch->idev == NULL) {
+	touch->idev = devm_input_allocate_device(&pdev->dev);
+	if (!touch->idev) {
 		dev_err(&pdev->dev, "Failed to allocate input device!\n");
-		ret = -ENOMEM;
-		goto out;
+		return -ENOMEM;
 	}
 
 	touch->idev->name = "88pm860x-touch";
@@ -258,10 +260,11 @@ static int pm860x_touch_probe(struct platform_device *pdev)
 	touch->res_x = res_x;
 	input_set_drvdata(touch->idev, touch);
 
-	ret = request_threaded_irq(touch->irq, NULL, pm860x_touch_handler,
-				   IRQF_ONESHOT, "touch", touch);
+	ret = devm_request_threaded_irq(&pdev->dev, touch->irq, NULL,
+					pm860x_touch_handler, IRQF_ONESHOT,
+					"touch", touch);
 	if (ret < 0)
-		goto out_irq;
+		return ret;
 
 	__set_bit(EV_ABS, touch->idev->evbit);
 	__set_bit(ABS_X, touch->idev->absbit);
@@ -279,28 +282,11 @@ static int pm860x_touch_probe(struct platform_device *pdev)
 	ret = input_register_device(touch->idev);
 	if (ret < 0) {
 		dev_err(chip->dev, "Failed to register touch!\n");
-		goto out_rg;
+		return ret;
 	}
 
 	platform_set_drvdata(pdev, touch);
 	return 0;
-out_rg:
-	free_irq(touch->irq, touch);
-out_irq:
-	input_free_device(touch->idev);
-out:
-	kfree(touch);
-	return ret;
-}
-
-static int pm860x_touch_remove(struct platform_device *pdev)
-{
-	struct pm860x_touch *touch = platform_get_drvdata(pdev);
-
-	input_unregister_device(touch->idev);
-	free_irq(touch->irq, touch);
-	kfree(touch);
-	return 0;
 }
 
 static struct platform_driver pm860x_touch_driver = {
@@ -309,7 +295,6 @@ static struct platform_driver pm860x_touch_driver = {
 		.owner	= THIS_MODULE,
 	},
 	.probe	= pm860x_touch_probe,
-	.remove	= pm860x_touch_remove,
 };
 module_platform_driver(pm860x_touch_driver);
 
diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig
index d4e5ab57909f..a23a94bb4bcb 100644
--- a/drivers/input/touchscreen/Kconfig
+++ b/drivers/input/touchscreen/Kconfig
@@ -11,6 +11,10 @@ menuconfig INPUT_TOUCHSCREEN
 
 if INPUT_TOUCHSCREEN
 
+config OF_TOUCHSCREEN
+	def_tristate INPUT
+	depends on INPUT && OF
+
 config TOUCHSCREEN_88PM860X
 	tristate "Marvell 88PM860x touchscreen"
 	depends on MFD_88PM860X
@@ -89,6 +93,7 @@ config TOUCHSCREEN_AD7879_SPI
 config TOUCHSCREEN_ATMEL_MXT
 	tristate "Atmel mXT I2C Touchscreen"
 	depends on I2C
+	select FW_LOADER
 	help
 	  Say Y here if you have Atmel mXT series I2C touchscreen,
 	  such as AT42QT602240/ATMXT224, connected to your system.
@@ -846,7 +851,7 @@ config TOUCHSCREEN_TSC2007
 
 config TOUCHSCREEN_W90X900
 	tristate "W90P910 touchscreen driver"
-	depends on HAVE_CLK
+	depends on ARCH_W90X900
 	help
 	  Say Y here if you have a W90P910 based touchscreen.
 
@@ -885,6 +890,17 @@ config TOUCHSCREEN_STMPE
 	  To compile this driver as a module, choose M here: the
 	  module will be called stmpe-ts.
 
+config TOUCHSCREEN_SUN4I
+	tristate "Allwinner sun4i resistive touchscreen controller support"
+	depends on ARCH_SUNXI || COMPILE_TEST
+	depends on HWMON
+	help
+	  This selects support for the resistive touchscreen controller
+	  found on Allwinner sunxi SoCs.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called sun4i-ts.
+
 config TOUCHSCREEN_SUR40
 	tristate "Samsung SUR40 (Surface 2.0/PixelSense) touchscreen"
 	depends on USB
diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile
index 03f12a1f2218..126479d8c29a 100644
--- a/drivers/input/touchscreen/Makefile
+++ b/drivers/input/touchscreen/Makefile
@@ -6,6 +6,7 @@
 
 wm97xx-ts-y := wm97xx-core.o
 
+obj-$(CONFIG_OF_TOUCHSCREEN)		+= of_touchscreen.o
 obj-$(CONFIG_TOUCHSCREEN_88PM860X)	+= 88pm860x-ts.o
 obj-$(CONFIG_TOUCHSCREEN_AD7877)	+= ad7877.o
 obj-$(CONFIG_TOUCHSCREEN_AD7879)	+= ad7879.o
@@ -53,6 +54,7 @@ obj-$(CONFIG_TOUCHSCREEN_PIXCIR)	+= pixcir_i2c_ts.o
 obj-$(CONFIG_TOUCHSCREEN_S3C2410)	+= s3c2410_ts.o
 obj-$(CONFIG_TOUCHSCREEN_ST1232)	+= st1232.o
 obj-$(CONFIG_TOUCHSCREEN_STMPE)		+= stmpe-ts.o
+obj-$(CONFIG_TOUCHSCREEN_SUN4I)		+= sun4i-ts.o
 obj-$(CONFIG_TOUCHSCREEN_SUR40)		+= sur40.o
 obj-$(CONFIG_TOUCHSCREEN_TI_AM335X_TSC)	+= ti_am335x_tsc.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213)	+= touchit213.o
diff --git a/drivers/input/touchscreen/ad7877.c b/drivers/input/touchscreen/ad7877.c
index 6793c85903ae..523865daa1d3 100644
--- a/drivers/input/touchscreen/ad7877.c
+++ b/drivers/input/touchscreen/ad7877.c
@@ -210,11 +210,6 @@ static bool gpio3;
 module_param(gpio3, bool, 0);
 MODULE_PARM_DESC(gpio3, "If gpio3 is set to 1 AUX3 acts as GPIO3");
 
-/*
- * ad7877_read/write are only used for initial setup and for sysfs controls.
- * The main traffic is done using spi_async() in the interrupt handler.
- */
-
 static int ad7877_read(struct spi_device *spi, u16 reg)
 {
 	struct ser_req *req;
diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c
index 7f8aa981500d..da201b8e37dc 100644
--- a/drivers/input/touchscreen/ads7846.c
+++ b/drivers/input/touchscreen/ads7846.c
@@ -706,7 +706,7 @@ static void ads7846_read_state(struct ads7846 *ts)
 		m = &ts->msg[msg_idx];
 		error = spi_sync(ts->spi, m);
 		if (error) {
-			dev_err(&ts->spi->dev, "spi_async --> %d\n", error);
+			dev_err(&ts->spi->dev, "spi_sync --> %d\n", error);
 			packet->tc.ignore = true;
 			return;
 		}
diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
index a70400754e92..6e0b4a2120d3 100644
--- a/drivers/input/touchscreen/atmel_mxt_ts.c
+++ b/drivers/input/touchscreen/atmel_mxt_ts.c
@@ -2,6 +2,8 @@
  * Atmel maXTouch Touchscreen driver
  *
  * Copyright (C) 2010 Samsung Electronics Co.Ltd
+ * Copyright (C) 2012 Google, Inc.
+ *
  * Author: Joonyoung Shim <jy0922.shim@samsung.com>
  *
  * This program is free software; you can redistribute  it and/or modify it
@@ -12,6 +14,8 @@
  */
 
 #include <linux/module.h>
+#include <linux/init.h>
+#include <linux/completion.h>
 #include <linux/delay.h>
 #include <linux/firmware.h>
 #include <linux/i2c.h>
@@ -25,12 +29,6 @@
 #define MXT_VER_21		21
 #define MXT_VER_22		22
 
-/* Slave addresses */
-#define MXT_APP_LOW		0x4a
-#define MXT_APP_HIGH		0x4b
-#define MXT_BOOT_LOW		0x24
-#define MXT_BOOT_HIGH		0x25
-
 /* Firmware */
 #define MXT_FW_NAME		"maxtouch.fw"
 
@@ -83,6 +81,9 @@
 #define MXT_COMMAND_REPORTALL	3
 #define MXT_COMMAND_DIAGNOSTIC	5
 
+/* Define for T6 status byte */
+#define MXT_T6_STATUS_RESET	(1 << 7)
+
 /* MXT_GEN_POWER_T7 field */
 #define MXT_POWER_IDLEACQINT	0
 #define MXT_POWER_ACTVACQINT	1
@@ -99,33 +100,26 @@
 
 /* MXT_TOUCH_MULTI_T9 field */
 #define MXT_TOUCH_CTRL		0
-#define MXT_TOUCH_XORIGIN	1
-#define MXT_TOUCH_YORIGIN	2
-#define MXT_TOUCH_XSIZE		3
-#define MXT_TOUCH_YSIZE		4
-#define MXT_TOUCH_BLEN		6
-#define MXT_TOUCH_TCHTHR	7
-#define MXT_TOUCH_TCHDI		8
-#define MXT_TOUCH_ORIENT	9
-#define MXT_TOUCH_MOVHYSTI	11
-#define MXT_TOUCH_MOVHYSTN	12
-#define MXT_TOUCH_NUMTOUCH	14
-#define MXT_TOUCH_MRGHYST	15
-#define MXT_TOUCH_MRGTHR	16
-#define MXT_TOUCH_AMPHYST	17
-#define MXT_TOUCH_XRANGE_LSB	18
-#define MXT_TOUCH_XRANGE_MSB	19
-#define MXT_TOUCH_YRANGE_LSB	20
-#define MXT_TOUCH_YRANGE_MSB	21
-#define MXT_TOUCH_XLOCLIP	22
-#define MXT_TOUCH_XHICLIP	23
-#define MXT_TOUCH_YLOCLIP	24
-#define MXT_TOUCH_YHICLIP	25
-#define MXT_TOUCH_XEDGECTRL	26
-#define MXT_TOUCH_XEDGEDIST	27
-#define MXT_TOUCH_YEDGECTRL	28
-#define MXT_TOUCH_YEDGEDIST	29
-#define MXT_TOUCH_JUMPLIMIT	30
+#define MXT_T9_ORIENT		9
+#define MXT_T9_RANGE		18
+
+/* MXT_TOUCH_MULTI_T9 status */
+#define MXT_T9_UNGRIP		(1 << 0)
+#define MXT_T9_SUPPRESS		(1 << 1)
+#define MXT_T9_AMP		(1 << 2)
+#define MXT_T9_VECTOR		(1 << 3)
+#define MXT_T9_MOVE		(1 << 4)
+#define MXT_T9_RELEASE		(1 << 5)
+#define MXT_T9_PRESS		(1 << 6)
+#define MXT_T9_DETECT		(1 << 7)
+
+struct t9_range {
+	u16 x;
+	u16 y;
+} __packed;
+
+/* MXT_TOUCH_MULTI_T9 orient */
+#define MXT_T9_ORIENT_SWITCH	(1 << 0)
 
 /* MXT_PROCI_GRIPFACE_T20 field */
 #define MXT_GRIPFACE_CTRL	0
@@ -174,17 +168,16 @@
 
 /* Define for MXT_GEN_COMMAND_T6 */
 #define MXT_BOOT_VALUE		0xa5
+#define MXT_RESET_VALUE		0x01
 #define MXT_BACKUP_VALUE	0x55
+
+/* Delay times */
 #define MXT_BACKUP_TIME		50	/* msec */
 #define MXT_RESET_TIME		200	/* msec */
-
-#define MXT_FWRESET_TIME	175	/* msec */
-
-/* MXT_SPT_GPIOPWM_T19 field */
-#define MXT_GPIO0_MASK		0x04
-#define MXT_GPIO1_MASK		0x08
-#define MXT_GPIO2_MASK		0x10
-#define MXT_GPIO3_MASK		0x20
+#define MXT_RESET_TIMEOUT	3000	/* msec */
+#define MXT_CRC_TIMEOUT		1000	/* msec */
+#define MXT_FW_RESET_TIME	3000	/* msec */
+#define MXT_FW_CHG_TIMEOUT	300	/* msec */
 
 /* Command to unlock bootloader */
 #define MXT_UNLOCK_CMD_MSB	0xaa
@@ -198,21 +191,8 @@
 #define MXT_FRAME_CRC_PASS	0x04
 #define MXT_APP_CRC_FAIL	0x40	/* valid 7 8 bit only */
 #define MXT_BOOT_STATUS_MASK	0x3f
-
-/* Touch status */
-#define MXT_UNGRIP		(1 << 0)
-#define MXT_SUPPRESS		(1 << 1)
-#define MXT_AMP			(1 << 2)
-#define MXT_VECTOR		(1 << 3)
-#define MXT_MOVE		(1 << 4)
-#define MXT_RELEASE		(1 << 5)
-#define MXT_PRESS		(1 << 6)
-#define MXT_DETECT		(1 << 7)
-
-/* Touch orient bits */
-#define MXT_XY_SWITCH		(1 << 0)
-#define MXT_X_INVERT		(1 << 1)
-#define MXT_Y_INVERT		(1 << 2)
+#define MXT_BOOT_EXTENDED_ID	(1 << 5)
+#define MXT_BOOT_ID_MASK	0x1f
 
 /* Touchscreen absolute values */
 #define MXT_MAX_AREA		0xff
@@ -232,8 +212,8 @@ struct mxt_info {
 struct mxt_object {
 	u8 type;
 	u16 start_address;
-	u8 size;		/* Size of each instance - 1 */
-	u8 instances;		/* Number of instances - 1 */
+	u8 size_minus_one;
+	u8 instances_minus_one;
 	u8 num_report_ids;
 } __packed;
 
@@ -250,19 +230,40 @@ struct mxt_data {
 	const struct mxt_platform_data *pdata;
 	struct mxt_object *object_table;
 	struct mxt_info info;
-	bool is_tp;
-
 	unsigned int irq;
 	unsigned int max_x;
 	unsigned int max_y;
+	bool in_bootloader;
+	u32 config_crc;
+	u8 bootloader_addr;
 
 	/* Cached parameters from object table */
 	u8 T6_reportid;
+	u16 T6_address;
 	u8 T9_reportid_min;
 	u8 T9_reportid_max;
 	u8 T19_reportid;
+
+	/* for fw update in bootloader */
+	struct completion bl_completion;
+
+	/* for reset handling */
+	struct completion reset_completion;
+
+	/* for config update handling */
+	struct completion crc_completion;
 };
 
+static size_t mxt_obj_size(const struct mxt_object *obj)
+{
+	return obj->size_minus_one + 1;
+}
+
+static size_t mxt_obj_instances(const struct mxt_object *obj)
+{
+	return obj->instances_minus_one + 1;
+}
+
 static bool mxt_object_readable(unsigned int type)
 {
 	switch (type) {
@@ -334,60 +335,190 @@ static void mxt_dump_message(struct device *dev,
 		message->reportid, 7, message->message);
 }
 
-static int mxt_check_bootloader(struct i2c_client *client,
-				     unsigned int state)
+static int mxt_wait_for_completion(struct mxt_data *data,
+				   struct completion *comp,
+				   unsigned int timeout_ms)
+{
+	struct device *dev = &data->client->dev;
+	unsigned long timeout = msecs_to_jiffies(timeout_ms);
+	long ret;
+
+	ret = wait_for_completion_interruptible_timeout(comp, timeout);
+	if (ret < 0) {
+		return ret;
+	} else if (ret == 0) {
+		dev_err(dev, "Wait for completion timed out.\n");
+		return -ETIMEDOUT;
+	}
+	return 0;
+}
+
+static int mxt_bootloader_read(struct mxt_data *data,
+			       u8 *val, unsigned int count)
+{
+	int ret;
+	struct i2c_msg msg;
+
+	msg.addr = data->bootloader_addr;
+	msg.flags = data->client->flags & I2C_M_TEN;
+	msg.flags |= I2C_M_RD;
+	msg.len = count;
+	msg.buf = val;
+
+	ret = i2c_transfer(data->client->adapter, &msg, 1);
+
+	if (ret == 1) {
+		ret = 0;
+	} else {
+		ret = ret < 0 ? ret : -EIO;
+		dev_err(&data->client->dev, "%s: i2c recv failed (%d)\n",
+			__func__, ret);
+	}
+
+	return ret;
+}
+
+static int mxt_bootloader_write(struct mxt_data *data,
+				const u8 * const val, unsigned int count)
+{
+	int ret;
+	struct i2c_msg msg;
+
+	msg.addr = data->bootloader_addr;
+	msg.flags = data->client->flags & I2C_M_TEN;
+	msg.len = count;
+	msg.buf = (u8 *)val;
+
+	ret = i2c_transfer(data->client->adapter, &msg, 1);
+	if (ret == 1) {
+		ret = 0;
+	} else {
+		ret = ret < 0 ? ret : -EIO;
+		dev_err(&data->client->dev, "%s: i2c send failed (%d)\n",
+			__func__, ret);
+	}
+
+	return ret;
+}
+
+static int mxt_lookup_bootloader_address(struct mxt_data *data)
+{
+	u8 appmode = data->client->addr;
+	u8 bootloader;
+
+	switch (appmode) {
+	case 0x4a:
+	case 0x4b:
+	case 0x4c:
+	case 0x4d:
+	case 0x5a:
+	case 0x5b:
+		bootloader = appmode - 0x26;
+		break;
+	default:
+		dev_err(&data->client->dev,
+			"Appmode i2c address 0x%02x not found\n",
+			appmode);
+		return -EINVAL;
+	}
+
+	data->bootloader_addr = bootloader;
+	return 0;
+}
+
+static u8 mxt_get_bootloader_version(struct mxt_data *data, u8 val)
+{
+	struct device *dev = &data->client->dev;
+	u8 buf[3];
+
+	if (val & MXT_BOOT_EXTENDED_ID) {
+		if (mxt_bootloader_read(data, &buf[0], 3) != 0) {
+			dev_err(dev, "%s: i2c failure\n", __func__);
+			return val;
+		}
+
+		dev_dbg(dev, "Bootloader ID:%d Version:%d\n", buf[1], buf[2]);
+
+		return buf[0];
+	} else {
+		dev_dbg(dev, "Bootloader ID:%d\n", val & MXT_BOOT_ID_MASK);
+
+		return val;
+	}
+}
+
+static int mxt_check_bootloader(struct mxt_data *data, unsigned int state)
 {
+	struct device *dev = &data->client->dev;
 	u8 val;
+	int ret;
 
 recheck:
-	if (i2c_master_recv(client, &val, 1) != 1) {
-		dev_err(&client->dev, "%s: i2c recv failed\n", __func__);
-		return -EIO;
+	if (state != MXT_WAITING_BOOTLOAD_CMD) {
+		/*
+		 * In application update mode, the interrupt
+		 * line signals state transitions. We must wait for the
+		 * CHG assertion before reading the status byte.
+		 * Once the status byte has been read, the line is deasserted.
+		 */
+		ret = mxt_wait_for_completion(data, &data->bl_completion,
+					      MXT_FW_CHG_TIMEOUT);
+		if (ret) {
+			/*
+			 * TODO: handle -ERESTARTSYS better by terminating
+			 * fw update process before returning to userspace
+			 * by writing length 0x000 to device (iff we are in
+			 * WAITING_FRAME_DATA state).
+			 */
+			dev_err(dev, "Update wait error %d\n", ret);
+			return ret;
+		}
 	}
 
+	ret = mxt_bootloader_read(data, &val, 1);
+	if (ret)
+		return ret;
+
+	if (state == MXT_WAITING_BOOTLOAD_CMD)
+		val = mxt_get_bootloader_version(data, val);
+
 	switch (state) {
 	case MXT_WAITING_BOOTLOAD_CMD:
 	case MXT_WAITING_FRAME_DATA:
 		val &= ~MXT_BOOT_STATUS_MASK;
 		break;
 	case MXT_FRAME_CRC_PASS:
-		if (val == MXT_FRAME_CRC_CHECK)
+		if (val == MXT_FRAME_CRC_CHECK) {
 			goto recheck;
+		} else if (val == MXT_FRAME_CRC_FAIL) {
+			dev_err(dev, "Bootloader CRC fail\n");
+			return -EINVAL;
+		}
 		break;
 	default:
 		return -EINVAL;
 	}
 
 	if (val != state) {
-		dev_err(&client->dev, "Unvalid bootloader mode state\n");
+		dev_err(dev, "Invalid bootloader state %02X != %02X\n",
+			val, state);
 		return -EINVAL;
 	}
 
 	return 0;
 }
 
-static int mxt_unlock_bootloader(struct i2c_client *client)
+static int mxt_unlock_bootloader(struct mxt_data *data)
 {
+	int ret;
 	u8 buf[2];
 
 	buf[0] = MXT_UNLOCK_CMD_LSB;
 	buf[1] = MXT_UNLOCK_CMD_MSB;
 
-	if (i2c_master_send(client, buf, 2) != 2) {
-		dev_err(&client->dev, "%s: i2c send failed\n", __func__);
-		return -EIO;
-	}
-
-	return 0;
-}
-
-static int mxt_fw_write(struct i2c_client *client,
-			     const u8 *data, unsigned int frame_size)
-{
-	if (i2c_master_send(client, data, frame_size) != frame_size) {
-		dev_err(&client->dev, "%s: i2c send failed\n", __func__);
-		return -EIO;
-	}
+	ret = mxt_bootloader_write(data, buf, 2);
+	if (ret)
+		return ret;
 
 	return 0;
 }
@@ -427,11 +558,6 @@ static int __mxt_read_reg(struct i2c_client *client,
 	return ret;
 }
 
-static int mxt_read_reg(struct i2c_client *client, u16 reg, u8 *val)
-{
-	return __mxt_read_reg(client, reg, 1, val);
-}
-
 static int __mxt_write_reg(struct i2c_client *client, u16 reg, u16 len,
 			   const void *val)
 {
@@ -479,7 +605,7 @@ mxt_get_object(struct mxt_data *data, u8 type)
 			return object;
 	}
 
-	dev_err(&data->client->dev, "Invalid object type\n");
+	dev_err(&data->client->dev, "Invalid object type T%u\n", type);
 	return NULL;
 }
 
@@ -505,7 +631,7 @@ static int mxt_write_object(struct mxt_data *data,
 	u16 reg;
 
 	object = mxt_get_object(data, type);
-	if (!object || offset >= object->size + 1)
+	if (!object || offset >= mxt_obj_size(object))
 		return -EINVAL;
 
 	reg = object->start_address;
@@ -515,18 +641,25 @@ static int mxt_write_object(struct mxt_data *data,
 static void mxt_input_button(struct mxt_data *data, struct mxt_message *message)
 {
 	struct input_dev *input = data->input_dev;
+	const struct mxt_platform_data *pdata = data->pdata;
 	bool button;
 	int i;
 
 	/* Active-low switch */
-	for (i = 0; i < MXT_NUM_GPIO; i++) {
-		if (data->pdata->key_map[i] == KEY_RESERVED)
+	for (i = 0; i < pdata->t19_num_keys; i++) {
+		if (pdata->t19_keymap[i] == KEY_RESERVED)
 			continue;
-		button = !(message->message[0] & MXT_GPIO0_MASK << i);
-		input_report_key(input, data->pdata->key_map[i], button);
+		button = !(message->message[0] & (1 << i));
+		input_report_key(input, pdata->t19_keymap[i], button);
 	}
 }
 
+static void mxt_input_sync(struct input_dev *input_dev)
+{
+	input_mt_report_pointer_emulation(input_dev, false);
+	input_sync(input_dev);
+}
+
 static void mxt_input_touchevent(struct mxt_data *data,
 				      struct mxt_message *message, int id)
 {
@@ -536,44 +669,60 @@ static void mxt_input_touchevent(struct mxt_data *data,
 	int x;
 	int y;
 	int area;
-	int pressure;
+	int amplitude;
 
 	x = (message->message[1] << 4) | ((message->message[3] >> 4) & 0xf);
 	y = (message->message[2] << 4) | ((message->message[3] & 0xf));
+
+	/* Handle 10/12 bit switching */
 	if (data->max_x < 1024)
-		x = x >> 2;
+		x >>= 2;
 	if (data->max_y < 1024)
-		y = y >> 2;
+		y >>= 2;
 
 	area = message->message[4];
-	pressure = message->message[5];
+	amplitude = message->message[5];
 
 	dev_dbg(dev,
 		"[%u] %c%c%c%c%c%c%c%c x: %5u y: %5u area: %3u amp: %3u\n",
 		id,
-		(status & MXT_DETECT) ? 'D' : '.',
-		(status & MXT_PRESS) ? 'P' : '.',
-		(status & MXT_RELEASE) ? 'R' : '.',
-		(status & MXT_MOVE) ? 'M' : '.',
-		(status & MXT_VECTOR) ? 'V' : '.',
-		(status & MXT_AMP) ? 'A' : '.',
-		(status & MXT_SUPPRESS) ? 'S' : '.',
-		(status & MXT_UNGRIP) ? 'U' : '.',
-		x, y, area, pressure);
+		(status & MXT_T9_DETECT) ? 'D' : '.',
+		(status & MXT_T9_PRESS) ? 'P' : '.',
+		(status & MXT_T9_RELEASE) ? 'R' : '.',
+		(status & MXT_T9_MOVE) ? 'M' : '.',
+		(status & MXT_T9_VECTOR) ? 'V' : '.',
+		(status & MXT_T9_AMP) ? 'A' : '.',
+		(status & MXT_T9_SUPPRESS) ? 'S' : '.',
+		(status & MXT_T9_UNGRIP) ? 'U' : '.',
+		x, y, area, amplitude);
 
 	input_mt_slot(input_dev, id);
-	input_mt_report_slot_state(input_dev, MT_TOOL_FINGER,
-				   status & MXT_DETECT);
 
-	if (status & MXT_DETECT) {
+	if (status & MXT_T9_DETECT) {
+		/*
+		 * Multiple bits may be set if the host is slow to read
+		 * the status messages, indicating all the events that
+		 * have happened.
+		 */
+		if (status & MXT_T9_RELEASE) {
+			input_mt_report_slot_state(input_dev,
+						   MT_TOOL_FINGER, 0);
+			mxt_input_sync(input_dev);
+		}
+
+		/* Touch active */
+		input_mt_report_slot_state(input_dev, MT_TOOL_FINGER, 1);
 		input_report_abs(input_dev, ABS_MT_POSITION_X, x);
 		input_report_abs(input_dev, ABS_MT_POSITION_Y, y);
-		input_report_abs(input_dev, ABS_MT_PRESSURE, pressure);
+		input_report_abs(input_dev, ABS_MT_PRESSURE, amplitude);
 		input_report_abs(input_dev, ABS_MT_TOUCH_MAJOR, area);
+	} else {
+		/* Touch no longer active, close out slot */
+		input_mt_report_slot_state(input_dev, MT_TOOL_FINGER, 0);
 	}
 }
 
-static unsigned mxt_extract_T6_csum(const u8 *csum)
+static u16 mxt_extract_T6_csum(const u8 *csum)
 {
 	return csum[0] | (csum[1] << 8) | (csum[2] << 16);
 }
@@ -584,28 +733,37 @@ static bool mxt_is_T9_message(struct mxt_data *data, struct mxt_message *msg)
 	return (id >= data->T9_reportid_min && id <= data->T9_reportid_max);
 }
 
-static irqreturn_t mxt_interrupt(int irq, void *dev_id)
+static irqreturn_t mxt_process_messages_until_invalid(struct mxt_data *data)
 {
-	struct mxt_data *data = dev_id;
 	struct mxt_message message;
 	const u8 *payload = &message.message[0];
 	struct device *dev = &data->client->dev;
 	u8 reportid;
 	bool update_input = false;
+	u32 crc;
 
 	do {
 		if (mxt_read_message(data, &message)) {
 			dev_err(dev, "Failed to read message\n");
-			goto end;
+			return IRQ_NONE;
 		}
 
 		reportid = message.reportid;
 
 		if (reportid == data->T6_reportid) {
 			u8 status = payload[0];
-			unsigned csum = mxt_extract_T6_csum(&payload[1]);
+
+			crc = mxt_extract_T6_csum(&payload[1]);
+			if (crc != data->config_crc) {
+				data->config_crc = crc;
+				complete(&data->crc_completion);
+			}
+
 			dev_dbg(dev, "Status: %02x Config Checksum: %06x\n",
-				status, csum);
+				status, data->config_crc);
+
+			if (status & MXT_T6_STATUS_RESET)
+				complete(&data->reset_completion);
 		} else if (mxt_is_T9_message(data, &message)) {
 			int id = reportid - data->T9_reportid_min;
 			mxt_input_touchevent(data, &message, id);
@@ -618,15 +776,96 @@ static irqreturn_t mxt_interrupt(int irq, void *dev_id)
 		}
 	} while (reportid != 0xff);
 
-	if (update_input) {
-		input_mt_report_pointer_emulation(data->input_dev, false);
-		input_sync(data->input_dev);
-	}
+	if (update_input)
+		mxt_input_sync(data->input_dev);
 
-end:
 	return IRQ_HANDLED;
 }
 
+static irqreturn_t mxt_interrupt(int irq, void *dev_id)
+{
+	struct mxt_data *data = dev_id;
+
+	if (data->in_bootloader) {
+		/* bootloader state transition completion */
+		complete(&data->bl_completion);
+		return IRQ_HANDLED;
+	}
+
+	return mxt_process_messages_until_invalid(data);
+}
+
+static int mxt_t6_command(struct mxt_data *data, u16 cmd_offset,
+			  u8 value, bool wait)
+{
+	u16 reg;
+	u8 command_register;
+	int timeout_counter = 0;
+	int ret;
+
+	reg = data->T6_address + cmd_offset;
+
+	ret = mxt_write_reg(data->client, reg, value);
+	if (ret)
+		return ret;
+
+	if (!wait)
+		return 0;
+
+	do {
+		msleep(20);
+		ret = __mxt_read_reg(data->client, reg, 1, &command_register);
+		if (ret)
+			return ret;
+	} while (command_register != 0 && timeout_counter++ <= 100);
+
+	if (timeout_counter > 100) {
+		dev_err(&data->client->dev, "Command failed!\n");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int mxt_soft_reset(struct mxt_data *data)
+{
+	struct device *dev = &data->client->dev;
+	int ret = 0;
+
+	dev_info(dev, "Resetting chip\n");
+
+	reinit_completion(&data->reset_completion);
+
+	ret = mxt_t6_command(data, MXT_COMMAND_RESET, MXT_RESET_VALUE, false);
+	if (ret)
+		return ret;
+
+	ret = mxt_wait_for_completion(data, &data->reset_completion,
+				      MXT_RESET_TIMEOUT);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static void mxt_update_crc(struct mxt_data *data, u8 cmd, u8 value)
+{
+	/*
+	 * On failure, CRC is set to 0 and config will always be
+	 * downloaded.
+	 */
+	data->config_crc = 0;
+	reinit_completion(&data->crc_completion);
+
+	mxt_t6_command(data, cmd, value, true);
+
+	/*
+	 * Wait for crc message. On failure, CRC is set to 0 and config will
+	 * always be downloaded.
+	 */
+	mxt_wait_for_completion(data, &data->crc_completion, MXT_CRC_TIMEOUT);
+}
+
 static int mxt_check_reg_init(struct mxt_data *data)
 {
 	const struct mxt_platform_data *pdata = data->pdata;
@@ -641,13 +880,23 @@ static int mxt_check_reg_init(struct mxt_data *data)
 		return 0;
 	}
 
+	mxt_update_crc(data, MXT_COMMAND_REPORTALL, 1);
+
+	if (data->config_crc == pdata->config_crc) {
+		dev_info(dev, "Config CRC 0x%06X: OK\n", data->config_crc);
+		return 0;
+	}
+
+	dev_info(dev, "Config CRC 0x%06X: does not match 0x%06X\n",
+		 data->config_crc, pdata->config_crc);
+
 	for (i = 0; i < data->info.object_num; i++) {
 		object = data->object_table + i;
 
 		if (!mxt_object_writable(object->type))
 			continue;
 
-		size = (object->size + 1) * (object->instances + 1);
+		size = mxt_obj_size(object) * mxt_obj_instances(object);
 		if (index + size > pdata->config_length) {
 			dev_err(dev, "Not enough config data!\n");
 			return -EINVAL;
@@ -660,6 +909,14 @@ static int mxt_check_reg_init(struct mxt_data *data)
 		index += size;
 	}
 
+	mxt_update_crc(data, MXT_COMMAND_BACKUPNV, MXT_BACKUP_VALUE);
+
+	ret = mxt_soft_reset(data);
+	if (ret)
+		return ret;
+
+	dev_info(dev, "Config successfully updated\n");
+
 	return 0;
 }
 
@@ -685,54 +942,6 @@ static int mxt_make_highchg(struct mxt_data *data)
 	return 0;
 }
 
-static void mxt_handle_pdata(struct mxt_data *data)
-{
-	const struct mxt_platform_data *pdata = data->pdata;
-	u8 voltage;
-
-	/* Set touchscreen lines */
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9, MXT_TOUCH_XSIZE,
-			pdata->x_line);
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9, MXT_TOUCH_YSIZE,
-			pdata->y_line);
-
-	/* Set touchscreen orient */
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9, MXT_TOUCH_ORIENT,
-			pdata->orient);
-
-	/* Set touchscreen burst length */
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9,
-			MXT_TOUCH_BLEN, pdata->blen);
-
-	/* Set touchscreen threshold */
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9,
-			MXT_TOUCH_TCHTHR, pdata->threshold);
-
-	/* Set touchscreen resolution */
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9,
-			MXT_TOUCH_XRANGE_LSB, (pdata->x_size - 1) & 0xff);
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9,
-			MXT_TOUCH_XRANGE_MSB, (pdata->x_size - 1) >> 8);
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9,
-			MXT_TOUCH_YRANGE_LSB, (pdata->y_size - 1) & 0xff);
-	mxt_write_object(data, MXT_TOUCH_MULTI_T9,
-			MXT_TOUCH_YRANGE_MSB, (pdata->y_size - 1) >> 8);
-
-	/* Set touchscreen voltage */
-	if (pdata->voltage) {
-		if (pdata->voltage < MXT_VOLTAGE_DEFAULT) {
-			voltage = (MXT_VOLTAGE_DEFAULT - pdata->voltage) /
-				MXT_VOLTAGE_STEP;
-			voltage = 0xff - voltage + 1;
-		} else
-			voltage = (pdata->voltage - MXT_VOLTAGE_DEFAULT) /
-				MXT_VOLTAGE_STEP;
-
-		mxt_write_object(data, MXT_SPT_CTECONFIG_T28,
-				MXT_CTE_VOLTAGE, voltage);
-	}
-}
-
 static int mxt_get_info(struct mxt_data *data)
 {
 	struct i2c_client *client = data->client;
@@ -772,7 +981,7 @@ static int mxt_get_object_table(struct mxt_data *data)
 		if (object->num_report_ids) {
 			min_id = reportid;
 			reportid += object->num_report_ids *
-					(object->instances + 1);
+					mxt_obj_instances(object);
 			max_id = reportid - 1;
 		} else {
 			min_id = 0;
@@ -780,13 +989,15 @@ static int mxt_get_object_table(struct mxt_data *data)
 		}
 
 		dev_dbg(&data->client->dev,
-			"Type %2d Start %3d Size %3d Instances %2d ReportIDs %3u : %3u\n",
-			object->type, object->start_address, object->size + 1,
-			object->instances + 1, min_id, max_id);
+			"T%u Start:%u Size:%zu Instances:%zu Report IDs:%u-%u\n",
+			object->type, object->start_address,
+			mxt_obj_size(object), mxt_obj_instances(object),
+			min_id, max_id);
 
 		switch (object->type) {
 		case MXT_GEN_COMMAND_T6:
 			data->T6_reportid = min_id;
+			data->T6_address = object->start_address;
 			break;
 		case MXT_TOUCH_MULTI_T9:
 			data->T9_reportid_min = min_id;
@@ -811,12 +1022,59 @@ static void mxt_free_object_table(struct mxt_data *data)
 	data->T19_reportid = 0;
 }
 
+static int mxt_read_t9_resolution(struct mxt_data *data)
+{
+	struct i2c_client *client = data->client;
+	int error;
+	struct t9_range range;
+	unsigned char orient;
+	struct mxt_object *object;
+
+	object = mxt_get_object(data, MXT_TOUCH_MULTI_T9);
+	if (!object)
+		return -EINVAL;
+
+	error = __mxt_read_reg(client,
+			       object->start_address + MXT_T9_RANGE,
+			       sizeof(range), &range);
+	if (error)
+		return error;
+
+	le16_to_cpus(&range.x);
+	le16_to_cpus(&range.y);
+
+	error =  __mxt_read_reg(client,
+				object->start_address + MXT_T9_ORIENT,
+				1, &orient);
+	if (error)
+		return error;
+
+	/* Handle default values */
+	if (range.x == 0)
+		range.x = 1023;
+
+	if (range.y == 0)
+		range.y = 1023;
+
+	if (orient & MXT_T9_ORIENT_SWITCH) {
+		data->max_x = range.y;
+		data->max_y = range.x;
+	} else {
+		data->max_x = range.x;
+		data->max_y = range.y;
+	}
+
+	dev_dbg(&client->dev,
+		"Touchscreen size X%uY%u\n", data->max_x, data->max_y);
+
+	return 0;
+}
+
 static int mxt_initialize(struct mxt_data *data)
 {
 	struct i2c_client *client = data->client;
 	struct mxt_info *info = &data->info;
 	int error;
-	u8 val;
 
 	error = mxt_get_info(data);
 	if (error)
@@ -832,47 +1090,29 @@ static int mxt_initialize(struct mxt_data *data)
 
 	/* Get object table information */
 	error = mxt_get_object_table(data);
-	if (error)
+	if (error) {
+		dev_err(&client->dev, "Error %d reading object table\n", error);
 		goto err_free_object_table;
+	}
 
 	/* Check register init values */
 	error = mxt_check_reg_init(data);
-	if (error)
-		goto err_free_object_table;
-
-	mxt_handle_pdata(data);
-
-	/* Backup to memory */
-	mxt_write_object(data, MXT_GEN_COMMAND_T6,
-			MXT_COMMAND_BACKUPNV,
-			MXT_BACKUP_VALUE);
-	msleep(MXT_BACKUP_TIME);
-
-	/* Soft reset */
-	mxt_write_object(data, MXT_GEN_COMMAND_T6,
-			MXT_COMMAND_RESET, 1);
-	msleep(MXT_RESET_TIME);
-
-	/* Update matrix size at info struct */
-	error = mxt_read_reg(client, MXT_MATRIX_X_SIZE, &val);
-	if (error)
+	if (error) {
+		dev_err(&client->dev, "Error %d initializing configuration\n",
+			error);
 		goto err_free_object_table;
-	info->matrix_xsize = val;
+	}
 
-	error = mxt_read_reg(client, MXT_MATRIX_Y_SIZE, &val);
-	if (error)
+	error = mxt_read_t9_resolution(data);
+	if (error) {
+		dev_err(&client->dev, "Failed to initialize T9 resolution\n");
 		goto err_free_object_table;
-	info->matrix_ysize = val;
-
-	dev_info(&client->dev,
-			"Family ID: %u Variant ID: %u Major.Minor.Build: %u.%u.%02X\n",
-			info->family_id, info->variant_id, info->version >> 4,
-			info->version & 0xf, info->build);
+	}
 
 	dev_info(&client->dev,
-			"Matrix X Size: %u Matrix Y Size: %u Object Num: %u\n",
-			info->matrix_xsize, info->matrix_ysize,
-			info->object_num);
+		 "Family: %u Variant: %u Firmware V%u.%u.%02X Objects: %u\n",
+		 info->family_id, info->variant_id, info->version >> 4,
+		 info->version & 0xf, info->build, info->object_num);
 
 	return 0;
 
@@ -881,20 +1121,6 @@ err_free_object_table:
 	return error;
 }
 
-static void mxt_calc_resolution(struct mxt_data *data)
-{
-	unsigned int max_x = data->pdata->x_size - 1;
-	unsigned int max_y = data->pdata->y_size - 1;
-
-	if (data->pdata->orient & MXT_XY_SWITCH) {
-		data->max_x = max_y;
-		data->max_y = max_x;
-	} else {
-		data->max_x = max_x;
-		data->max_y = max_y;
-	}
-}
-
 /* Firmware Version is returned as Major.Minor.Build */
 static ssize_t mxt_fw_version_show(struct device *dev,
 				   struct device_attribute *attr, char *buf)
@@ -921,11 +1147,11 @@ static ssize_t mxt_show_instance(char *buf, int count,
 {
 	int i;
 
-	if (object->instances > 0)
+	if (mxt_obj_instances(object) > 1)
 		count += scnprintf(buf + count, PAGE_SIZE - count,
 				   "Instance %u\n", instance);
 
-	for (i = 0; i < object->size + 1; i++)
+	for (i = 0; i < mxt_obj_size(object); i++)
 		count += scnprintf(buf + count, PAGE_SIZE - count,
 				"\t[%2u]: %02x (%d)\n", i, val[i], val[i]);
 	count += scnprintf(buf + count, PAGE_SIZE - count, "\n");
@@ -958,8 +1184,8 @@ static ssize_t mxt_object_show(struct device *dev,
 		count += scnprintf(buf + count, PAGE_SIZE - count,
 				"T%u:\n", object->type);
 
-		for (j = 0; j < object->instances + 1; j++) {
-			u16 size = object->size + 1;
+		for (j = 0; j < mxt_obj_instances(object); j++) {
+			u16 size = mxt_obj_size(object);
 			u16 addr = object->start_address + j * size;
 
 			error = __mxt_read_reg(data->client, addr, size, obuf);
@@ -975,13 +1201,38 @@ done:
 	return error ?: count;
 }
 
+static int mxt_check_firmware_format(struct device *dev,
+				     const struct firmware *fw)
+{
+	unsigned int pos = 0;
+	char c;
+
+	while (pos < fw->size) {
+		c = *(fw->data + pos);
+
+		if (c < '0' || (c > '9' && c < 'A') || c > 'F')
+			return 0;
+
+		pos++;
+	}
+
+	/*
+	 * To convert file try:
+	 * xxd -r -p mXTXXX__APP_VX-X-XX.enc > maxtouch.fw
+	 */
+	dev_err(dev, "Aborting: firmware file must be in binary format\n");
+
+	return -EINVAL;
+}
+
 static int mxt_load_fw(struct device *dev, const char *fn)
 {
 	struct mxt_data *data = dev_get_drvdata(dev);
-	struct i2c_client *client = data->client;
 	const struct firmware *fw = NULL;
 	unsigned int frame_size;
 	unsigned int pos = 0;
+	unsigned int retry = 0;
+	unsigned int frame = 0;
 	int ret;
 
 	ret = request_firmware(&fw, fn, dev);
@@ -990,59 +1241,91 @@ static int mxt_load_fw(struct device *dev, const char *fn)
 		return ret;
 	}
 
+	/* Check for incorrect enc file */
+	ret = mxt_check_firmware_format(dev, fw);
+	if (ret)
+		goto release_firmware;
+
+	ret = mxt_lookup_bootloader_address(data);
+	if (ret)
+		goto release_firmware;
+
 	/* Change to the bootloader mode */
-	mxt_write_object(data, MXT_GEN_COMMAND_T6,
-			MXT_COMMAND_RESET, MXT_BOOT_VALUE);
+	data->in_bootloader = true;
+
+	ret = mxt_t6_command(data, MXT_COMMAND_RESET, MXT_BOOT_VALUE, false);
+	if (ret)
+		goto release_firmware;
+
 	msleep(MXT_RESET_TIME);
 
-	/* Change to slave address of bootloader */
-	if (client->addr == MXT_APP_LOW)
-		client->addr = MXT_BOOT_LOW;
-	else
-		client->addr = MXT_BOOT_HIGH;
+	reinit_completion(&data->bl_completion);
 
-	ret = mxt_check_bootloader(client, MXT_WAITING_BOOTLOAD_CMD);
+	ret = mxt_check_bootloader(data, MXT_WAITING_BOOTLOAD_CMD);
 	if (ret)
-		goto out;
+		goto disable_irq;
 
 	/* Unlock bootloader */
-	mxt_unlock_bootloader(client);
+	mxt_unlock_bootloader(data);
 
 	while (pos < fw->size) {
-		ret = mxt_check_bootloader(client,
-						MXT_WAITING_FRAME_DATA);
+		ret = mxt_check_bootloader(data, MXT_WAITING_FRAME_DATA);
 		if (ret)
-			goto out;
+			goto disable_irq;
 
 		frame_size = ((*(fw->data + pos) << 8) | *(fw->data + pos + 1));
 
-		/* We should add 2 at frame size as the the firmware data is not
-		 * included the CRC bytes.
-		 */
+		/* Take account of CRC bytes */
 		frame_size += 2;
 
 		/* Write one frame to device */
-		mxt_fw_write(client, fw->data + pos, frame_size);
-
-		ret = mxt_check_bootloader(client,
-						MXT_FRAME_CRC_PASS);
+		ret = mxt_bootloader_write(data, fw->data + pos, frame_size);
 		if (ret)
-			goto out;
+			goto disable_irq;
 
-		pos += frame_size;
+		ret = mxt_check_bootloader(data, MXT_FRAME_CRC_PASS);
+		if (ret) {
+			retry++;
 
-		dev_dbg(dev, "Updated %d bytes / %zd bytes\n", pos, fw->size);
+			/* Back off by 20ms per retry */
+			msleep(retry * 20);
+
+			if (retry > 20) {
+				dev_err(dev, "Retry count exceeded\n");
+				goto disable_irq;
+			}
+		} else {
+			retry = 0;
+			pos += frame_size;
+			frame++;
+		}
+
+		if (frame % 50 == 0)
+			dev_dbg(dev, "Sent %d frames, %d/%zd bytes\n",
+				frame, pos, fw->size);
 	}
 
-out:
-	release_firmware(fw);
+	/* Wait for flash. */
+	ret = mxt_wait_for_completion(data, &data->bl_completion,
+				      MXT_FW_RESET_TIME);
+	if (ret)
+		goto disable_irq;
 
-	/* Change to slave address of application */
-	if (client->addr == MXT_BOOT_LOW)
-		client->addr = MXT_APP_LOW;
-	else
-		client->addr = MXT_APP_HIGH;
+	dev_dbg(dev, "Sent %d frames, %d bytes\n", frame, pos);
 
+	/*
+	 * Wait for device to reset. Some bootloader versions do not assert
+	 * the CHG line after bootloading has finished, so ignore potential
+	 * errors.
+	 */
+	mxt_wait_for_completion(data, &data->bl_completion, MXT_FW_RESET_TIME);
+
+	data->in_bootloader = false;
+
+disable_irq:
+	disable_irq(data->irq);
+release_firmware:
+	release_firmware(fw);
 	return ret;
 }
 
@@ -1053,28 +1336,23 @@ static ssize_t mxt_update_fw_store(struct device *dev,
 	struct mxt_data *data = dev_get_drvdata(dev);
 	int error;
 
-	disable_irq(data->irq);
-
 	error = mxt_load_fw(dev, MXT_FW_NAME);
 	if (error) {
 		dev_err(dev, "The firmware update failed(%d)\n", error);
 		count = error;
 	} else {
-		dev_dbg(dev, "The firmware update succeeded\n");
-
-		/* Wait for reset */
-		msleep(MXT_FWRESET_TIME);
+		dev_info(dev, "The firmware update succeeded\n");
 
 		mxt_free_object_table(data);
 
 		mxt_initialize(data);
-	}
 
-	enable_irq(data->irq);
+		enable_irq(data->irq);
 
-	error = mxt_make_highchg(data);
-	if (error)
-		return error;
+		error = mxt_make_highchg(data);
+		if (error)
+			return error;
+	}
 
 	return count;
 }
@@ -1134,6 +1412,8 @@ static int mxt_probe(struct i2c_client *client,
 	struct input_dev *input_dev;
 	int error;
 	unsigned int num_mt_slots;
+	unsigned int mt_flags = 0;
+	int i;
 
 	if (!pdata)
 		return -EINVAL;
@@ -1146,10 +1426,7 @@ static int mxt_probe(struct i2c_client *client,
 		goto err_free_mem;
 	}
 
-	data->is_tp = pdata && pdata->is_tp;
-
-	input_dev->name = (data->is_tp) ? "Atmel maXTouch Touchpad" :
-					  "Atmel maXTouch Touchscreen";
+	input_dev->name = "Atmel maXTouch Touchscreen";
 	snprintf(data->phys, sizeof(data->phys), "i2c-%u-%04x/input0",
 		 client->adapter->nr, client->addr);
 
@@ -1165,7 +1442,9 @@ static int mxt_probe(struct i2c_client *client,
 	data->pdata = pdata;
 	data->irq = client->irq;
 
-	mxt_calc_resolution(data);
+	init_completion(&data->bl_completion);
+	init_completion(&data->reset_completion);
+	init_completion(&data->crc_completion);
 
 	error = mxt_initialize(data);
 	if (error)
@@ -1175,20 +1454,15 @@ static int mxt_probe(struct i2c_client *client,
 	__set_bit(EV_KEY, input_dev->evbit);
 	__set_bit(BTN_TOUCH, input_dev->keybit);
 
-	if (data->is_tp) {
-		int i;
-		__set_bit(INPUT_PROP_POINTER, input_dev->propbit);
+	if (pdata->t19_num_keys) {
 		__set_bit(INPUT_PROP_BUTTONPAD, input_dev->propbit);
 
-		for (i = 0; i < MXT_NUM_GPIO; i++)
-			if (pdata->key_map[i] != KEY_RESERVED)
-				__set_bit(pdata->key_map[i], input_dev->keybit);
+		for (i = 0; i < pdata->t19_num_keys; i++)
+			if (pdata->t19_keymap[i] != KEY_RESERVED)
+				input_set_capability(input_dev, EV_KEY,
+						     pdata->t19_keymap[i]);
 
-		__set_bit(BTN_TOOL_FINGER, input_dev->keybit);
-		__set_bit(BTN_TOOL_DOUBLETAP, input_dev->keybit);
-		__set_bit(BTN_TOOL_TRIPLETAP, input_dev->keybit);
-		__set_bit(BTN_TOOL_QUADTAP, input_dev->keybit);
-		__set_bit(BTN_TOOL_QUINTTAP, input_dev->keybit);
+		mt_flags |= INPUT_MT_POINTER;
 
 		input_abs_set_res(input_dev, ABS_X, MXT_PIXELS_PER_MM);
 		input_abs_set_res(input_dev, ABS_Y, MXT_PIXELS_PER_MM);
@@ -1196,6 +1470,8 @@ static int mxt_probe(struct i2c_client *client,
 				  MXT_PIXELS_PER_MM);
 		input_abs_set_res(input_dev, ABS_MT_POSITION_Y,
 				  MXT_PIXELS_PER_MM);
+
+		input_dev->name = "Atmel maXTouch Touchpad";
 	}
 
 	/* For single touch */
@@ -1208,7 +1484,7 @@ static int mxt_probe(struct i2c_client *client,
 
 	/* For multi touch */
 	num_mt_slots = data->T9_reportid_max - data->T9_reportid_min + 1;
-	error = input_mt_init_slots(input_dev, num_mt_slots, 0);
+	error = input_mt_init_slots(input_dev, num_mt_slots, mt_flags);
 	if (error)
 		goto err_free_object;
 	input_set_abs_params(input_dev, ABS_MT_TOUCH_MAJOR,
@@ -1236,12 +1512,18 @@ static int mxt_probe(struct i2c_client *client,
 		goto err_free_irq;
 
 	error = input_register_device(input_dev);
-	if (error)
+	if (error) {
+		dev_err(&client->dev, "Error %d registering input device\n",
+			error);
 		goto err_free_irq;
+	}
 
 	error = sysfs_create_group(&client->dev.kobj, &mxt_attr_group);
-	if (error)
+	if (error) {
+		dev_err(&client->dev, "Failure %d creating sysfs group\n",
+			error);
 		goto err_unregister_device;
+	}
 
 	return 0;
 
@@ -1294,11 +1576,7 @@ static int mxt_resume(struct device *dev)
 	struct mxt_data *data = i2c_get_clientdata(client);
 	struct input_dev *input_dev = data->input_dev;
 
-	/* Soft reset */
-	mxt_write_object(data, MXT_GEN_COMMAND_T6,
-			MXT_COMMAND_RESET, 1);
-
-	msleep(MXT_RESET_TIME);
+	mxt_soft_reset(data);
 
 	mutex_lock(&input_dev->mutex);
 
diff --git a/drivers/input/touchscreen/auo-pixcir-ts.c b/drivers/input/touchscreen/auo-pixcir-ts.c
index d3f9f6b0f9b7..7f3c94787787 100644
--- a/drivers/input/touchscreen/auo-pixcir-ts.c
+++ b/drivers/input/touchscreen/auo-pixcir-ts.c
@@ -679,7 +679,7 @@ static const struct i2c_device_id auo_pixcir_idtable[] = {
 MODULE_DEVICE_TABLE(i2c, auo_pixcir_idtable);
 
 #ifdef CONFIG_OF
-static struct of_device_id auo_pixcir_ts_dt_idtable[] = {
+static const struct of_device_id auo_pixcir_ts_dt_idtable[] = {
 	{ .compatible = "auo,auo_pixcir_ts" },
 	{},
 };
diff --git a/drivers/input/touchscreen/da9034-ts.c b/drivers/input/touchscreen/da9034-ts.c
index 8ccf7bb4028a..cf6f4b31db4d 100644
--- a/drivers/input/touchscreen/da9034-ts.c
+++ b/drivers/input/touchscreen/da9034-ts.c
@@ -301,10 +301,11 @@ static int da9034_touch_probe(struct platform_device *pdev)
 	struct da9034_touch_pdata *pdata = dev_get_platdata(&pdev->dev);
 	struct da9034_touch *touch;
 	struct input_dev *input_dev;
-	int ret;
+	int error;
 
-	touch = kzalloc(sizeof(struct da9034_touch), GFP_KERNEL);
-	if (touch == NULL) {
+	touch = devm_kzalloc(&pdev->dev, sizeof(struct da9034_touch),
+			     GFP_KERNEL);
+	if (!touch) {
 		dev_err(&pdev->dev, "failed to allocate driver data\n");
 		return -ENOMEM;
 	}
@@ -315,18 +316,18 @@ static int da9034_touch_probe(struct platform_device *pdev)
 		touch->interval_ms	= pdata->interval_ms;
 		touch->x_inverted	= pdata->x_inverted;
 		touch->y_inverted	= pdata->y_inverted;
-	} else
+	} else {
 		/* fallback into default */
 		touch->interval_ms	= 10;
+	}
 
 	INIT_DELAYED_WORK(&touch->tsi_work, da9034_tsi_work);
 	touch->notifier.notifier_call = da9034_touch_notifier;
 
-	input_dev = input_allocate_device();
+	input_dev = devm_input_allocate_device(&pdev->dev);
 	if (!input_dev) {
 		dev_err(&pdev->dev, "failed to allocate input device\n");
-		ret = -ENOMEM;
-		goto err_free_touch;
+		return -ENOMEM;
 	}
 
 	input_dev->name		= pdev->name;
@@ -346,26 +347,9 @@ static int da9034_touch_probe(struct platform_device *pdev)
 	touch->input_dev = input_dev;
 	input_set_drvdata(input_dev, touch);
 
-	ret = input_register_device(input_dev);
-	if (ret)
-		goto err_free_input;
-
-	platform_set_drvdata(pdev, touch);
-	return 0;
-
-err_free_input:
-	input_free_device(input_dev);
-err_free_touch:
-	kfree(touch);
-	return ret;
-}
-
-static int da9034_touch_remove(struct platform_device *pdev)
-{
-	struct da9034_touch *touch = platform_get_drvdata(pdev);
-
-	input_unregister_device(touch->input_dev);
-	kfree(touch);
+	error = input_register_device(input_dev);
+	if (error)
+		return error;
 
 	return 0;
 }
@@ -376,7 +360,6 @@ static struct platform_driver da9034_touch_driver = {
 		.owner	= THIS_MODULE,
 	},
 	.probe		= da9034_touch_probe,
-	.remove		= da9034_touch_remove,
 };
 module_platform_driver(da9034_touch_driver);
 
diff --git a/drivers/input/touchscreen/edt-ft5x06.c b/drivers/input/touchscreen/edt-ft5x06.c
index f8815bebc9ef..d4f33992ad8c 100644
--- a/drivers/input/touchscreen/edt-ft5x06.c
+++ b/drivers/input/touchscreen/edt-ft5x06.c
@@ -271,7 +271,7 @@ static int edt_ft5x06_register_write(struct edt_ft5x06_ts_data *tsdata,
 		wrbuf[0] = addr;
 		wrbuf[1] = value;
 
-		return edt_ft5x06_ts_readwrite(tsdata->client, 3,
+		return edt_ft5x06_ts_readwrite(tsdata->client, 2,
 					wrbuf, 0, NULL);
 
 	default:
diff --git a/drivers/input/touchscreen/egalax_ts.c b/drivers/input/touchscreen/egalax_ts.c
index e6bcb13680b2..c8057847d71d 100644
--- a/drivers/input/touchscreen/egalax_ts.c
+++ b/drivers/input/touchscreen/egalax_ts.c
@@ -262,7 +262,7 @@ static int egalax_ts_resume(struct device *dev)
 
 static SIMPLE_DEV_PM_OPS(egalax_ts_pm_ops, egalax_ts_suspend, egalax_ts_resume);
 
-static struct of_device_id egalax_ts_dt_ids[] = {
+static const struct of_device_id egalax_ts_dt_ids[] = {
 	{ .compatible = "eeti,egalax_ts" },
 	{ /* sentinel */ }
 };
diff --git a/drivers/input/touchscreen/intel-mid-touch.c b/drivers/input/touchscreen/intel-mid-touch.c
index 4f6b156144e9..c38ca4a7e386 100644
--- a/drivers/input/touchscreen/intel-mid-touch.c
+++ b/drivers/input/touchscreen/intel-mid-touch.c
@@ -36,6 +36,7 @@
 #include <linux/irq.h>
 #include <linux/delay.h>
 #include <asm/intel_scu_ipc.h>
+#include <linux/device.h>
 
 /* PMIC Interrupt registers */
 #define PMIC_REG_ID1		0x00 /* PMIC ID1 register */
@@ -580,12 +581,17 @@ static int mrstouch_probe(struct platform_device *pdev)
 		return -EINVAL;
 	}
 
-	tsdev = kzalloc(sizeof(struct mrstouch_dev), GFP_KERNEL);
-	input = input_allocate_device();
-	if (!tsdev || !input) {
+	tsdev = devm_kzalloc(&pdev->dev, sizeof(struct mrstouch_dev),
+			     GFP_KERNEL);
+	if (!tsdev) {
 		dev_err(&pdev->dev, "unable to allocate memory\n");
-		err = -ENOMEM;
-		goto err_free_mem;
+		return -ENOMEM;
+	}
+
+	input = devm_input_allocate_device(&pdev->dev);
+	if (!input) {
+		dev_err(&pdev->dev, "unable to allocate input device\n");
+		return -ENOMEM;
 	}
 
 	tsdev->dev = &pdev->dev;
@@ -598,7 +604,7 @@ static int mrstouch_probe(struct platform_device *pdev)
 	err = mrstouch_adc_init(tsdev);
 	if (err) {
 		dev_err(&pdev->dev, "ADC initialization failed\n");
-		goto err_free_mem;
+		return err;
 	}
 
 	input->name = "mrst_touchscreen";
@@ -618,38 +624,20 @@ static int mrstouch_probe(struct platform_device *pdev)
 	input_set_abs_params(tsdev->input, ABS_PRESSURE,
 			     MRST_PRESSURE_MIN, MRST_PRESSURE_MAX, 0, 0);
 
-	err = request_threaded_irq(tsdev->irq, NULL, mrstouch_pendet_irq,
-				   IRQF_ONESHOT, "mrstouch", tsdev);
+	err = devm_request_threaded_irq(&pdev->dev, tsdev->irq, NULL,
+					mrstouch_pendet_irq, IRQF_ONESHOT,
+					"mrstouch", tsdev);
 	if (err) {
 		dev_err(tsdev->dev, "unable to allocate irq\n");
-		goto err_free_mem;
+		return err;
 	}
 
 	err = input_register_device(tsdev->input);
 	if (err) {
 		dev_err(tsdev->dev, "unable to register input device\n");
-		goto err_free_irq;
+		return err;
 	}
 
-	platform_set_drvdata(pdev, tsdev);
-	return 0;
-
-err_free_irq:
-	free_irq(tsdev->irq, tsdev);
-err_free_mem:
-	input_free_device(input);
-	kfree(tsdev);
-	return err;
-}
-
-static int mrstouch_remove(struct platform_device *pdev)
-{
-	struct mrstouch_dev *tsdev = platform_get_drvdata(pdev);
-
-	free_irq(tsdev->irq, tsdev);
-	input_unregister_device(tsdev->input);
-	kfree(tsdev);
-
 	return 0;
 }
 
@@ -659,7 +647,6 @@ static struct platform_driver mrstouch_driver = {
 		.owner	= THIS_MODULE,
 	},
 	.probe		= mrstouch_probe,
-	.remove		= mrstouch_remove,
 };
 module_platform_driver(mrstouch_driver);
 
diff --git a/drivers/input/touchscreen/lpc32xx_ts.c b/drivers/input/touchscreen/lpc32xx_ts.c
index 2058253b55d9..bb47d3442a35 100644
--- a/drivers/input/touchscreen/lpc32xx_ts.c
+++ b/drivers/input/touchscreen/lpc32xx_ts.c
@@ -384,7 +384,7 @@ static const struct dev_pm_ops lpc32xx_ts_pm_ops = {
 #endif
 
 #ifdef CONFIG_OF
-static struct of_device_id lpc32xx_tsc_of_match[] = {
+static const struct of_device_id lpc32xx_tsc_of_match[] = {
 	{ .compatible = "nxp,lpc3220-tsc", },
 	{ },
 };
diff --git a/drivers/input/touchscreen/mcs5000_ts.c b/drivers/input/touchscreen/mcs5000_ts.c
index 647e36f5930e..00510a9836b3 100644
--- a/drivers/input/touchscreen/mcs5000_ts.c
+++ b/drivers/input/touchscreen/mcs5000_ts.c
@@ -161,10 +161,9 @@ static irqreturn_t mcs5000_ts_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void mcs5000_ts_phys_init(struct mcs5000_ts_data *data)
+static void mcs5000_ts_phys_init(struct mcs5000_ts_data *data,
+				 const struct mcs_platform_data *platform_data)
 {
-	const struct mcs_platform_data *platform_data =
-		data->platform_data;
 	struct i2c_client *client = data->client;
 
 	/* Touch reset & sleep mode */
@@ -187,28 +186,32 @@ static void mcs5000_ts_phys_init(struct mcs5000_ts_data *data)
 }
 
 static int mcs5000_ts_probe(struct i2c_client *client,
-		const struct i2c_device_id *id)
+			    const struct i2c_device_id *id)
 {
+	const struct mcs_platform_data *pdata;
 	struct mcs5000_ts_data *data;
 	struct input_dev *input_dev;
-	int ret;
+	int error;
 
-	if (!dev_get_platdata(&client->dev))
+	pdata = dev_get_platdata(&client->dev);
+	if (!pdata)
 		return -EINVAL;
 
-	data = kzalloc(sizeof(struct mcs5000_ts_data), GFP_KERNEL);
-	input_dev = input_allocate_device();
-	if (!data || !input_dev) {
+	data = devm_kzalloc(&client->dev, sizeof(*data), GFP_KERNEL);
+	if (!data) {
 		dev_err(&client->dev, "Failed to allocate memory\n");
-		ret = -ENOMEM;
-		goto err_free_mem;
+		return -ENOMEM;
 	}
 
 	data->client = client;
-	data->input_dev = input_dev;
-	data->platform_data = dev_get_platdata(&client->dev);
 
-	input_dev->name = "MELPAS MCS-5000 Touchscreen";
+	input_dev = devm_input_allocate_device(&client->dev);
+	if (!input_dev) {
+		dev_err(&client->dev, "Failed to allocate input device\n");
+		return -ENOMEM;
+	}
+
+	input_dev->name = "MELFAS MCS-5000 Touchscreen";
 	input_dev->id.bustype = BUS_I2C;
 	input_dev->dev.parent = &client->dev;
 
@@ -219,44 +222,30 @@ static int mcs5000_ts_probe(struct i2c_client *client,
 	input_set_abs_params(input_dev, ABS_Y, 0, MCS5000_MAX_YC, 0, 0);
 
 	input_set_drvdata(input_dev, data);
+	data->input_dev = input_dev;
 
-	if (data->platform_data->cfg_pin)
-		data->platform_data->cfg_pin();
-
-	ret = request_threaded_irq(client->irq, NULL, mcs5000_ts_interrupt,
-			IRQF_TRIGGER_LOW | IRQF_ONESHOT, "mcs5000_ts", data);
+	if (pdata->cfg_pin)
+		pdata->cfg_pin();
 
-	if (ret < 0) {
+	error = devm_request_threaded_irq(&client->dev, client->irq,
+					  NULL, mcs5000_ts_interrupt,
+					  IRQF_TRIGGER_LOW | IRQF_ONESHOT,
+					  "mcs5000_ts", data);
+	if (error) {
 		dev_err(&client->dev, "Failed to register interrupt\n");
-		goto err_free_mem;
+		return error;
 	}
 
-	ret = input_register_device(data->input_dev);
-	if (ret < 0)
-		goto err_free_irq;
+	error = input_register_device(data->input_dev);
+	if (error) {
+		dev_err(&client->dev, "Failed to register input device\n");
+		return error;
+	}
 
-	mcs5000_ts_phys_init(data);
+	mcs5000_ts_phys_init(data, pdata);
 	i2c_set_clientdata(client, data);
 
 	return 0;
-
-err_free_irq:
-	free_irq(client->irq, data);
-err_free_mem:
-	input_free_device(input_dev);
-	kfree(data);
-	return ret;
-}
-
-static int mcs5000_ts_remove(struct i2c_client *client)
-{
-	struct mcs5000_ts_data *data = i2c_get_clientdata(client);
-
-	free_irq(client->irq, data);
-	input_unregister_device(data->input_dev);
-	kfree(data);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM
@@ -274,14 +263,15 @@ static int mcs5000_ts_resume(struct device *dev)
 {
 	struct i2c_client *client = to_i2c_client(dev);
 	struct mcs5000_ts_data *data = i2c_get_clientdata(client);
+	const struct mcs_platform_data *pdata = dev_get_platdata(dev);
 
-	mcs5000_ts_phys_init(data);
+	mcs5000_ts_phys_init(data, pdata);
 
 	return 0;
 }
+#endif
 
 static SIMPLE_DEV_PM_OPS(mcs5000_ts_pm, mcs5000_ts_suspend, mcs5000_ts_resume);
-#endif
 
 static const struct i2c_device_id mcs5000_ts_id[] = {
 	{ "mcs5000_ts", 0 },
@@ -291,12 +281,9 @@ MODULE_DEVICE_TABLE(i2c, mcs5000_ts_id);
 
 static struct i2c_driver mcs5000_ts_driver = {
 	.probe		= mcs5000_ts_probe,
-	.remove		= mcs5000_ts_remove,
 	.driver = {
 		.name = "mcs5000_ts",
-#ifdef CONFIG_PM
 		.pm   = &mcs5000_ts_pm,
-#endif
 	},
 	.id_table	= mcs5000_ts_id,
 };
diff --git a/drivers/input/touchscreen/mms114.c b/drivers/input/touchscreen/mms114.c
index 8a598c065391..372bbf7658fe 100644
--- a/drivers/input/touchscreen/mms114.c
+++ b/drivers/input/touchscreen/mms114.c
@@ -456,7 +456,7 @@ static int mms114_probe(struct i2c_client *client,
 	data->input_dev = input_dev;
 	data->pdata = pdata;
 
-	input_dev->name = "MELPAS MMS114 Touchscreen";
+	input_dev->name = "MELFAS MMS114 Touchscreen";
 	input_dev->id.bustype = BUS_I2C;
 	input_dev->dev.parent = &client->dev;
 	input_dev->open = mms114_input_open;
@@ -570,7 +570,7 @@ static const struct i2c_device_id mms114_id[] = {
 MODULE_DEVICE_TABLE(i2c, mms114_id);
 
 #ifdef CONFIG_OF
-static struct of_device_id mms114_dt_match[] = {
+static const struct of_device_id mms114_dt_match[] = {
 	{ .compatible = "melfas,mms114" },
 	{ }
 };
diff --git a/drivers/input/touchscreen/of_touchscreen.c b/drivers/input/touchscreen/of_touchscreen.c
new file mode 100644
index 000000000000..f8f9b84230b1
--- /dev/null
+++ b/drivers/input/touchscreen/of_touchscreen.c
@@ -0,0 +1,45 @@
+/*
+ *  Generic DT helper functions for touchscreen devices
+ *
+ *  Copyright (c) 2014 Sebastian Reichel <sre@kernel.org>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/of.h>
+#include <linux/input.h>
+#include <linux/input/touchscreen.h>
+
+/**
+ * touchscreen_parse_of_params - parse common touchscreen DT properties
+ * @dev: device that should be parsed
+ *
+ * This function parses common DT properties for touchscreens and setups the
+ * input device accordingly. The function keeps previously setuped default
+ * values if no value is specified via DT.
+ */
+void touchscreen_parse_of_params(struct input_dev *dev)
+{
+	struct device_node *np = dev->dev.parent->of_node;
+	struct input_absinfo *absinfo;
+
+	input_alloc_absinfo(dev);
+	if (!dev->absinfo)
+		return;
+
+	absinfo = &dev->absinfo[ABS_X];
+	of_property_read_u32(np, "touchscreen-size-x", &absinfo->maximum);
+	of_property_read_u32(np, "touchscreen-fuzz-x", &absinfo->fuzz);
+
+	absinfo = &dev->absinfo[ABS_Y];
+	of_property_read_u32(np, "touchscreen-size-y", &absinfo->maximum);
+	of_property_read_u32(np, "touchscreen-fuzz-y", &absinfo->fuzz);
+
+	absinfo = &dev->absinfo[ABS_PRESSURE];
+	of_property_read_u32(np, "touchscreen-max-pressure", &absinfo->maximum);
+	of_property_read_u32(np, "touchscreen-fuzz-pressure", &absinfo->fuzz);
+}
+EXPORT_SYMBOL(touchscreen_parse_of_params);
diff --git a/drivers/input/touchscreen/pixcir_i2c_ts.c b/drivers/input/touchscreen/pixcir_i2c_ts.c
index 02392d2061d6..19c6c0fdc94b 100644
--- a/drivers/input/touchscreen/pixcir_i2c_ts.c
+++ b/drivers/input/touchscreen/pixcir_i2c_ts.c
@@ -24,12 +24,13 @@
 #include <linux/i2c.h>
 #include <linux/input.h>
 #include <linux/input/pixcir_ts.h>
+#include <linux/gpio.h>
 
 struct pixcir_i2c_ts_data {
 	struct i2c_client *client;
 	struct input_dev *input;
 	const struct pixcir_ts_platform_data *chip;
-	bool exiting;
+	bool running;
 };
 
 static void pixcir_ts_poscheck(struct pixcir_i2c_ts_data *data)
@@ -87,11 +88,12 @@ static void pixcir_ts_poscheck(struct pixcir_i2c_ts_data *data)
 static irqreturn_t pixcir_ts_isr(int irq, void *dev_id)
 {
 	struct pixcir_i2c_ts_data *tsdata = dev_id;
+	const struct pixcir_ts_platform_data *pdata = tsdata->chip;
 
-	while (!tsdata->exiting) {
+	while (tsdata->running) {
 		pixcir_ts_poscheck(tsdata);
 
-		if (tsdata->chip->attb_read_val())
+		if (gpio_get_value(pdata->gpio_attb))
 			break;
 
 		msleep(20);
@@ -100,25 +102,221 @@ static irqreturn_t pixcir_ts_isr(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
+static int pixcir_set_power_mode(struct pixcir_i2c_ts_data *ts,
+				 enum pixcir_power_mode mode)
+{
+	struct device *dev = &ts->client->dev;
+	int ret;
+
+	ret = i2c_smbus_read_byte_data(ts->client, PIXCIR_REG_POWER_MODE);
+	if (ret < 0) {
+		dev_err(dev, "%s: can't read reg 0x%x : %d\n",
+			__func__, PIXCIR_REG_POWER_MODE, ret);
+		return ret;
+	}
+
+	ret &= ~PIXCIR_POWER_MODE_MASK;
+	ret |= mode;
+
+	/* Always AUTO_IDLE */
+	ret |= PIXCIR_POWER_ALLOW_IDLE;
+
+	ret = i2c_smbus_write_byte_data(ts->client, PIXCIR_REG_POWER_MODE, ret);
+	if (ret < 0) {
+		dev_err(dev, "%s: can't write reg 0x%x : %d\n",
+			__func__, PIXCIR_REG_POWER_MODE, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+/*
+ * Set the interrupt mode for the device i.e. ATTB line behaviour
+ *
+ * @polarity : 1 for active high, 0 for active low.
+ */
+static int pixcir_set_int_mode(struct pixcir_i2c_ts_data *ts,
+			       enum pixcir_int_mode mode, bool polarity)
+{
+	struct device *dev = &ts->client->dev;
+	int ret;
+
+	ret = i2c_smbus_read_byte_data(ts->client, PIXCIR_REG_INT_MODE);
+	if (ret < 0) {
+		dev_err(dev, "%s: can't read reg 0x%x : %d\n",
+			__func__, PIXCIR_REG_INT_MODE, ret);
+		return ret;
+	}
+
+	ret &= ~PIXCIR_INT_MODE_MASK;
+	ret |= mode;
+
+	if (polarity)
+		ret |= PIXCIR_INT_POL_HIGH;
+	else
+		ret &= ~PIXCIR_INT_POL_HIGH;
+
+	ret = i2c_smbus_write_byte_data(ts->client, PIXCIR_REG_INT_MODE, ret);
+	if (ret < 0) {
+		dev_err(dev, "%s: can't write reg 0x%x : %d\n",
+			__func__, PIXCIR_REG_INT_MODE, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+/*
+ * Enable/disable interrupt generation
+ */
+static int pixcir_int_enable(struct pixcir_i2c_ts_data *ts, bool enable)
+{
+	struct device *dev = &ts->client->dev;
+	int ret;
+
+	ret = i2c_smbus_read_byte_data(ts->client, PIXCIR_REG_INT_MODE);
+	if (ret < 0) {
+		dev_err(dev, "%s: can't read reg 0x%x : %d\n",
+			__func__, PIXCIR_REG_INT_MODE, ret);
+		return ret;
+	}
+
+	if (enable)
+		ret |= PIXCIR_INT_ENABLE;
+	else
+		ret &= ~PIXCIR_INT_ENABLE;
+
+	ret = i2c_smbus_write_byte_data(ts->client, PIXCIR_REG_INT_MODE, ret);
+	if (ret < 0) {
+		dev_err(dev, "%s: can't write reg 0x%x : %d\n",
+			__func__, PIXCIR_REG_INT_MODE, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int pixcir_start(struct pixcir_i2c_ts_data *ts)
+{
+	struct device *dev = &ts->client->dev;
+	int error;
+
+	/* LEVEL_TOUCH interrupt with active low polarity */
+	error = pixcir_set_int_mode(ts, PIXCIR_INT_LEVEL_TOUCH, 0);
+	if (error) {
+		dev_err(dev, "Failed to set interrupt mode: %d\n", error);
+		return error;
+	}
+
+	ts->running = true;
+	mb();	/* Update status before IRQ can fire */
+
+	/* enable interrupt generation */
+	error = pixcir_int_enable(ts, true);
+	if (error) {
+		dev_err(dev, "Failed to enable interrupt generation: %d\n",
+			error);
+		return error;
+	}
+
+	return 0;
+}
+
+static int pixcir_stop(struct pixcir_i2c_ts_data *ts)
+{
+	int error;
+
+	/* Disable interrupt generation */
+	error = pixcir_int_enable(ts, false);
+	if (error) {
+		dev_err(&ts->client->dev,
+			"Failed to disable interrupt generation: %d\n",
+			error);
+		return error;
+	}
+
+	/* Exit ISR if running, no more report parsing */
+	ts->running = false;
+	mb();	/* update status before we synchronize irq */
+
+	/* Wait till running ISR is complete */
+	synchronize_irq(ts->client->irq);
+
+	return 0;
+}
+
+static int pixcir_input_open(struct input_dev *dev)
+{
+	struct pixcir_i2c_ts_data *ts = input_get_drvdata(dev);
+
+	return pixcir_start(ts);
+}
+
+static void pixcir_input_close(struct input_dev *dev)
+{
+	struct pixcir_i2c_ts_data *ts = input_get_drvdata(dev);
+
+	pixcir_stop(ts);
+}
+
 #ifdef CONFIG_PM_SLEEP
 static int pixcir_i2c_ts_suspend(struct device *dev)
 {
 	struct i2c_client *client = to_i2c_client(dev);
+	struct pixcir_i2c_ts_data *ts = i2c_get_clientdata(client);
+	struct input_dev *input = ts->input;
+	int ret = 0;
+
+	mutex_lock(&input->mutex);
+
+	if (device_may_wakeup(&client->dev)) {
+		if (!input->users) {
+			ret = pixcir_start(ts);
+			if (ret) {
+				dev_err(dev, "Failed to start\n");
+				goto unlock;
+			}
+		}
 
-	if (device_may_wakeup(&client->dev))
 		enable_irq_wake(client->irq);
+	} else if (input->users) {
+		ret = pixcir_stop(ts);
+	}
 
-	return 0;
+unlock:
+	mutex_unlock(&input->mutex);
+
+	return ret;
 }
 
 static int pixcir_i2c_ts_resume(struct device *dev)
 {
 	struct i2c_client *client = to_i2c_client(dev);
+	struct pixcir_i2c_ts_data *ts = i2c_get_clientdata(client);
+	struct input_dev *input = ts->input;
+	int ret = 0;
+
+	mutex_lock(&input->mutex);
 
-	if (device_may_wakeup(&client->dev))
+	if (device_may_wakeup(&client->dev)) {
 		disable_irq_wake(client->irq);
 
-	return 0;
+		if (!input->users) {
+			ret = pixcir_stop(ts);
+			if (ret) {
+				dev_err(dev, "Failed to stop\n");
+				goto unlock;
+			}
+		}
+	} else if (input->users) {
+		ret = pixcir_start(ts);
+	}
+
+unlock:
+	mutex_unlock(&input->mutex);
+
+	return ret;
 }
 #endif
 
@@ -130,6 +328,7 @@ static int pixcir_i2c_ts_probe(struct i2c_client *client,
 {
 	const struct pixcir_ts_platform_data *pdata =
 			dev_get_platdata(&client->dev);
+	struct device *dev = &client->dev;
 	struct pixcir_i2c_ts_data *tsdata;
 	struct input_dev *input;
 	int error;
@@ -139,12 +338,19 @@ static int pixcir_i2c_ts_probe(struct i2c_client *client,
 		return -EINVAL;
 	}
 
-	tsdata = kzalloc(sizeof(*tsdata), GFP_KERNEL);
-	input = input_allocate_device();
-	if (!tsdata || !input) {
-		dev_err(&client->dev, "Failed to allocate driver data!\n");
-		error = -ENOMEM;
-		goto err_free_mem;
+	if (!gpio_is_valid(pdata->gpio_attb)) {
+		dev_err(dev, "Invalid gpio_attb in pdata\n");
+		return -EINVAL;
+	}
+
+	tsdata = devm_kzalloc(dev, sizeof(*tsdata), GFP_KERNEL);
+	if (!tsdata)
+		return -ENOMEM;
+
+	input = devm_input_allocate_device(dev);
+	if (!input) {
+		dev_err(dev, "Failed to allocate input device\n");
+		return -ENOMEM;
 	}
 
 	tsdata->client = client;
@@ -153,6 +359,8 @@ static int pixcir_i2c_ts_probe(struct i2c_client *client,
 
 	input->name = client->name;
 	input->id.bustype = BUS_I2C;
+	input->open = pixcir_input_open;
+	input->close = pixcir_input_close;
 	input->dev.parent = &client->dev;
 
 	__set_bit(EV_KEY, input->evbit);
@@ -165,44 +373,47 @@ static int pixcir_i2c_ts_probe(struct i2c_client *client,
 
 	input_set_drvdata(input, tsdata);
 
-	error = request_threaded_irq(client->irq, NULL, pixcir_ts_isr,
-				     IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
-				     client->name, tsdata);
+	error = devm_gpio_request_one(dev, pdata->gpio_attb,
+				      GPIOF_DIR_IN, "pixcir_i2c_attb");
 	if (error) {
-		dev_err(&client->dev, "Unable to request touchscreen IRQ.\n");
-		goto err_free_mem;
+		dev_err(dev, "Failed to request ATTB gpio\n");
+		return error;
 	}
 
+	error = devm_request_threaded_irq(dev, client->irq, NULL, pixcir_ts_isr,
+					  IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
+					  client->name, tsdata);
+	if (error) {
+		dev_err(dev, "failed to request irq %d\n", client->irq);
+		return error;
+	}
+
+	/* Always be in IDLE mode to save power, device supports auto wake */
+	error = pixcir_set_power_mode(tsdata, PIXCIR_POWER_IDLE);
+	if (error) {
+		dev_err(dev, "Failed to set IDLE mode\n");
+		return error;
+	}
+
+	/* Stop device till opened */
+	error = pixcir_stop(tsdata);
+	if (error)
+		return error;
+
 	error = input_register_device(input);
 	if (error)
-		goto err_free_irq;
+		return error;
 
 	i2c_set_clientdata(client, tsdata);
 	device_init_wakeup(&client->dev, 1);
 
 	return 0;
-
-err_free_irq:
-	free_irq(client->irq, tsdata);
-err_free_mem:
-	input_free_device(input);
-	kfree(tsdata);
-	return error;
 }
 
 static int pixcir_i2c_ts_remove(struct i2c_client *client)
 {
-	struct pixcir_i2c_ts_data *tsdata = i2c_get_clientdata(client);
-
 	device_init_wakeup(&client->dev, 0);
 
-	tsdata->exiting = true;
-	mb();
-	free_irq(client->irq, tsdata);
-
-	input_unregister_device(tsdata->input);
-	kfree(tsdata);
-
 	return 0;
 }
 
diff --git a/drivers/input/touchscreen/sun4i-ts.c b/drivers/input/touchscreen/sun4i-ts.c
new file mode 100644
index 000000000000..2ba826024954
--- /dev/null
+++ b/drivers/input/touchscreen/sun4i-ts.c
@@ -0,0 +1,339 @@
+/*
+ * Allwinner sunxi resistive touchscreen controller driver
+ *
+ * Copyright (C) 2013 - 2014 Hans de Goede <hdegoede@redhat.com>
+ *
+ * The hwmon parts are based on work by Corentin LABBE which is:
+ * Copyright (C) 2013 Corentin LABBE <clabbe.montjoie@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+/*
+ * The sun4i-ts controller is capable of detecting a second touch, but when a
+ * second touch is present then the accuracy becomes so bad the reported touch
+ * location is not useable.
+ *
+ * The original android driver contains some complicated heuristics using the
+ * aprox. distance between the 2 touches to see if the user is making a pinch
+ * open / close movement, and then reports emulated multi-touch events around
+ * the last touch coordinate (as the dual-touch coordinates are worthless).
+ *
+ * These kinds of heuristics are just asking for trouble (and don't belong
+ * in the kernel). So this driver offers straight forward, reliable single
+ * touch functionality only.
+ */
+
+#include <linux/err.h>
+#include <linux/hwmon.h>
+#include <linux/init.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#define TP_CTRL0		0x00
+#define TP_CTRL1		0x04
+#define TP_CTRL2		0x08
+#define TP_CTRL3		0x0c
+#define TP_INT_FIFOC		0x10
+#define TP_INT_FIFOS		0x14
+#define TP_TPR			0x18
+#define TP_CDAT			0x1c
+#define TEMP_DATA		0x20
+#define TP_DATA			0x24
+
+/* TP_CTRL0 bits */
+#define ADC_FIRST_DLY(x)	((x) << 24) /* 8 bits */
+#define ADC_FIRST_DLY_MODE(x)	((x) << 23)
+#define ADC_CLK_SEL(x)		((x) << 22)
+#define ADC_CLK_DIV(x)		((x) << 20) /* 3 bits */
+#define FS_DIV(x)		((x) << 16) /* 4 bits */
+#define T_ACQ(x)		((x) << 0) /* 16 bits */
+
+/* TP_CTRL1 bits */
+#define STYLUS_UP_DEBOUN(x)	((x) << 12) /* 8 bits */
+#define STYLUS_UP_DEBOUN_EN(x)	((x) << 9)
+#define TOUCH_PAN_CALI_EN(x)	((x) << 6)
+#define TP_DUAL_EN(x)		((x) << 5)
+#define TP_MODE_EN(x)		((x) << 4)
+#define TP_ADC_SELECT(x)	((x) << 3)
+#define ADC_CHAN_SELECT(x)	((x) << 0)  /* 3 bits */
+
+/* TP_CTRL2 bits */
+#define TP_SENSITIVE_ADJUST(x)	((x) << 28) /* 4 bits */
+#define TP_MODE_SELECT(x)	((x) << 26) /* 2 bits */
+#define PRE_MEA_EN(x)		((x) << 24)
+#define PRE_MEA_THRE_CNT(x)	((x) << 0) /* 24 bits */
+
+/* TP_CTRL3 bits */
+#define FILTER_EN(x)		((x) << 2)
+#define FILTER_TYPE(x)		((x) << 0)  /* 2 bits */
+
+/* TP_INT_FIFOC irq and fifo mask / control bits */
+#define TEMP_IRQ_EN(x)		((x) << 18)
+#define OVERRUN_IRQ_EN(x)	((x) << 17)
+#define DATA_IRQ_EN(x)		((x) << 16)
+#define TP_DATA_XY_CHANGE(x)	((x) << 13)
+#define FIFO_TRIG(x)		((x) << 8)  /* 5 bits */
+#define DATA_DRQ_EN(x)		((x) << 7)
+#define FIFO_FLUSH(x)		((x) << 4)
+#define TP_UP_IRQ_EN(x)		((x) << 1)
+#define TP_DOWN_IRQ_EN(x)	((x) << 0)
+
+/* TP_INT_FIFOS irq and fifo status bits */
+#define TEMP_DATA_PENDING	BIT(18)
+#define FIFO_OVERRUN_PENDING	BIT(17)
+#define FIFO_DATA_PENDING	BIT(16)
+#define TP_IDLE_FLG		BIT(2)
+#define TP_UP_PENDING		BIT(1)
+#define TP_DOWN_PENDING		BIT(0)
+
+/* TP_TPR bits */
+#define TEMP_ENABLE(x)		((x) << 16)
+#define TEMP_PERIOD(x)		((x) << 0)  /* t = x * 256 * 16 / clkin */
+
+struct sun4i_ts_data {
+	struct device *dev;
+	struct input_dev *input;
+	void __iomem *base;
+	unsigned int irq;
+	bool ignore_fifo_data;
+	int temp_data;
+};
+
+static void sun4i_ts_irq_handle_input(struct sun4i_ts_data *ts, u32 reg_val)
+{
+	u32 x, y;
+
+	if (reg_val & FIFO_DATA_PENDING) {
+		x = readl(ts->base + TP_DATA);
+		y = readl(ts->base + TP_DATA);
+		/* The 1st location reported after an up event is unreliable */
+		if (!ts->ignore_fifo_data) {
+			input_report_abs(ts->input, ABS_X, x);
+			input_report_abs(ts->input, ABS_Y, y);
+			/*
+			 * The hardware has a separate down status bit, but
+			 * that gets set before we get the first location,
+			 * resulting in reporting a click on the old location.
+			 */
+			input_report_key(ts->input, BTN_TOUCH, 1);
+			input_sync(ts->input);
+		} else {
+			ts->ignore_fifo_data = false;
+		}
+	}
+
+	if (reg_val & TP_UP_PENDING) {
+		ts->ignore_fifo_data = true;
+		input_report_key(ts->input, BTN_TOUCH, 0);
+		input_sync(ts->input);
+	}
+}
+
+static irqreturn_t sun4i_ts_irq(int irq, void *dev_id)
+{
+	struct sun4i_ts_data *ts = dev_id;
+	u32 reg_val;
+
+	reg_val  = readl(ts->base + TP_INT_FIFOS);
+
+	if (reg_val & TEMP_DATA_PENDING)
+		ts->temp_data = readl(ts->base + TEMP_DATA);
+
+	if (ts->input)
+		sun4i_ts_irq_handle_input(ts, reg_val);
+
+	writel(reg_val, ts->base + TP_INT_FIFOS);
+
+	return IRQ_HANDLED;
+}
+
+static int sun4i_ts_open(struct input_dev *dev)
+{
+	struct sun4i_ts_data *ts = input_get_drvdata(dev);
+
+	/* Flush, set trig level to 1, enable temp, data and up irqs */
+	writel(TEMP_IRQ_EN(1) | DATA_IRQ_EN(1) | FIFO_TRIG(1) | FIFO_FLUSH(1) |
+		TP_UP_IRQ_EN(1), ts->base + TP_INT_FIFOC);
+
+	return 0;
+}
+
+static void sun4i_ts_close(struct input_dev *dev)
+{
+	struct sun4i_ts_data *ts = input_get_drvdata(dev);
+
+	/* Deactivate all input IRQs */
+	writel(TEMP_IRQ_EN(1), ts->base + TP_INT_FIFOC);
+}
+
+static ssize_t show_temp(struct device *dev, struct device_attribute *devattr,
+			 char *buf)
+{
+	struct sun4i_ts_data *ts = dev_get_drvdata(dev);
+
+	/* No temp_data until the first irq */
+	if (ts->temp_data == -1)
+		return -EAGAIN;
+
+	return sprintf(buf, "%d\n", (ts->temp_data - 1447) * 100);
+}
+
+static ssize_t show_temp_label(struct device *dev,
+			      struct device_attribute *devattr, char *buf)
+{
+	return sprintf(buf, "SoC temperature\n");
+}
+
+static DEVICE_ATTR(temp1_input, S_IRUGO, show_temp, NULL);
+static DEVICE_ATTR(temp1_label, S_IRUGO, show_temp_label, NULL);
+
+static struct attribute *sun4i_ts_attrs[] = {
+	&dev_attr_temp1_input.attr,
+	&dev_attr_temp1_label.attr,
+	NULL
+};
+ATTRIBUTE_GROUPS(sun4i_ts);
+
+static int sun4i_ts_probe(struct platform_device *pdev)
+{
+	struct sun4i_ts_data *ts;
+	struct device *dev = &pdev->dev;
+	struct device_node *np = dev->of_node;
+	struct device *hwmon;
+	int error;
+	bool ts_attached;
+
+	ts = devm_kzalloc(dev, sizeof(struct sun4i_ts_data), GFP_KERNEL);
+	if (!ts)
+		return -ENOMEM;
+
+	ts->dev = dev;
+	ts->ignore_fifo_data = true;
+	ts->temp_data = -1;
+
+	ts_attached = of_property_read_bool(np, "allwinner,ts-attached");
+	if (ts_attached) {
+		ts->input = devm_input_allocate_device(dev);
+		if (!ts->input)
+			return -ENOMEM;
+
+		ts->input->name = pdev->name;
+		ts->input->phys = "sun4i_ts/input0";
+		ts->input->open = sun4i_ts_open;
+		ts->input->close = sun4i_ts_close;
+		ts->input->id.bustype = BUS_HOST;
+		ts->input->id.vendor = 0x0001;
+		ts->input->id.product = 0x0001;
+		ts->input->id.version = 0x0100;
+		ts->input->evbit[0] =  BIT(EV_SYN) | BIT(EV_KEY) | BIT(EV_ABS);
+		__set_bit(BTN_TOUCH, ts->input->keybit);
+		input_set_abs_params(ts->input, ABS_X, 0, 4095, 0, 0);
+		input_set_abs_params(ts->input, ABS_Y, 0, 4095, 0, 0);
+		input_set_drvdata(ts->input, ts);
+	}
+
+	ts->base = devm_ioremap_resource(dev,
+			      platform_get_resource(pdev, IORESOURCE_MEM, 0));
+	if (IS_ERR(ts->base))
+		return PTR_ERR(ts->base);
+
+	ts->irq = platform_get_irq(pdev, 0);
+	error = devm_request_irq(dev, ts->irq, sun4i_ts_irq, 0, "sun4i-ts", ts);
+	if (error)
+		return error;
+
+	/*
+	 * Select HOSC clk, clkin = clk / 6, adc samplefreq = clkin / 8192,
+	 * t_acq = clkin / (16 * 64)
+	 */
+	writel(ADC_CLK_SEL(0) | ADC_CLK_DIV(2) | FS_DIV(7) | T_ACQ(63),
+	       ts->base + TP_CTRL0);
+
+	/*
+	 * sensitive_adjust = 15 : max, which is not all that sensitive,
+	 * tp_mode = 0 : only x and y coordinates, as we don't use dual touch
+	 */
+	writel(TP_SENSITIVE_ADJUST(15) | TP_MODE_SELECT(0),
+	       ts->base + TP_CTRL2);
+
+	/* Enable median filter, type 1 : 5/3 */
+	writel(FILTER_EN(1) | FILTER_TYPE(1), ts->base + TP_CTRL3);
+
+	/* Enable temperature measurement, period 1953 (2 seconds) */
+	writel(TEMP_ENABLE(1) | TEMP_PERIOD(1953), ts->base + TP_TPR);
+
+	/*
+	 * Set stylus up debounce to aprox 10 ms, enable debounce, and
+	 * finally enable tp mode.
+	 */
+	writel(STYLUS_UP_DEBOUN(5) | STYLUS_UP_DEBOUN_EN(1) | TP_MODE_EN(1),
+	       ts->base + TP_CTRL1);
+
+	hwmon = devm_hwmon_device_register_with_groups(ts->dev, "sun4i_ts",
+						       ts, sun4i_ts_groups);
+	if (IS_ERR(hwmon))
+		return PTR_ERR(hwmon);
+
+	writel(TEMP_IRQ_EN(1), ts->base + TP_INT_FIFOC);
+
+	if (ts_attached) {
+		error = input_register_device(ts->input);
+		if (error) {
+			writel(0, ts->base + TP_INT_FIFOC);
+			return error;
+		}
+	}
+
+	platform_set_drvdata(pdev, ts);
+	return 0;
+}
+
+static int sun4i_ts_remove(struct platform_device *pdev)
+{
+	struct sun4i_ts_data *ts = platform_get_drvdata(pdev);
+
+	/* Explicit unregister to avoid open/close changing the imask later */
+	if (ts->input)
+		input_unregister_device(ts->input);
+
+	/* Deactivate all IRQs */
+	writel(0, ts->base + TP_INT_FIFOC);
+
+	return 0;
+}
+
+static const struct of_device_id sun4i_ts_of_match[] = {
+	{ .compatible = "allwinner,sun4i-a10-ts", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, sun4i_ts_of_match);
+
+static struct platform_driver sun4i_ts_driver = {
+	.driver = {
+		.owner	= THIS_MODULE,
+		.name	= "sun4i-ts",
+		.of_match_table = of_match_ptr(sun4i_ts_of_match),
+	},
+	.probe	= sun4i_ts_probe,
+	.remove	= sun4i_ts_remove,
+};
+
+module_platform_driver(sun4i_ts_driver);
+
+MODULE_DESCRIPTION("Allwinner sun4i resistive touchscreen controller driver");
+MODULE_AUTHOR("Hans de Goede <hdegoede@redhat.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/input/touchscreen/tsc2005.c b/drivers/input/touchscreen/tsc2005.c
index 550adcbbfc23..52380b68ebdf 100644
--- a/drivers/input/touchscreen/tsc2005.c
+++ b/drivers/input/touchscreen/tsc2005.c
@@ -25,11 +25,15 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/input.h>
+#include <linux/input/touchscreen.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/pm.h>
+#include <linux/of.h>
+#include <linux/of_gpio.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/tsc2005.h>
+#include <linux/regulator/consumer.h>
 
 /*
  * The touchscreen interface operates as follows:
@@ -100,6 +104,11 @@
 					 TSC2005_CFR2_AVG_7)
 
 #define MAX_12BIT			0xfff
+#define TSC2005_DEF_X_FUZZ		4
+#define TSC2005_DEF_Y_FUZZ		8
+#define TSC2005_DEF_P_FUZZ		2
+#define TSC2005_DEF_RESISTOR		280
+
 #define TSC2005_SPI_MAX_SPEED_HZ	10000000
 #define TSC2005_PENUP_TIME_MS		40
 
@@ -143,6 +152,9 @@ struct tsc2005 {
 
 	bool			pen_down;
 
+	struct regulator	*vio;
+
+	int			reset_gpio;
 	void			(*set_reset)(bool enable);
 };
 
@@ -337,6 +349,14 @@ static void tsc2005_stop_scan(struct tsc2005 *ts)
 	tsc2005_cmd(ts, TSC2005_CMD_STOP);
 }
 
+static void tsc2005_set_reset(struct tsc2005 *ts, bool enable)
+{
+	if (ts->reset_gpio >= 0)
+		gpio_set_value(ts->reset_gpio, enable);
+	else if (ts->set_reset)
+		ts->set_reset(enable);
+}
+
 /* must be called with ts->mutex held */
 static void __tsc2005_disable(struct tsc2005 *ts)
 {
@@ -355,7 +375,7 @@ static void __tsc2005_enable(struct tsc2005 *ts)
 {
 	tsc2005_start_scan(ts);
 
-	if (ts->esd_timeout && ts->set_reset) {
+	if (ts->esd_timeout && (ts->set_reset || ts->reset_gpio)) {
 		ts->last_valid_interrupt = jiffies;
 		schedule_delayed_work(&ts->esd_work,
 				round_jiffies_relative(
@@ -414,9 +434,9 @@ static ssize_t tsc2005_selftest_show(struct device *dev,
 	}
 
 	/* hardware reset */
-	ts->set_reset(false);
+	tsc2005_set_reset(ts, false);
 	usleep_range(100, 500); /* only 10us required */
-	ts->set_reset(true);
+	tsc2005_set_reset(ts, true);
 
 	if (!success)
 		goto out;
@@ -459,7 +479,7 @@ static umode_t tsc2005_attr_is_visible(struct kobject *kobj,
 	umode_t mode = attr->mode;
 
 	if (attr == &dev_attr_selftest.attr) {
-		if (!ts->set_reset)
+		if (!ts->set_reset && !ts->reset_gpio)
 			mode = 0;
 	}
 
@@ -509,9 +529,9 @@ static void tsc2005_esd_work(struct work_struct *work)
 
 	tsc2005_update_pen_state(ts, 0, 0, 0);
 
-	ts->set_reset(false);
+	tsc2005_set_reset(ts, false);
 	usleep_range(100, 500); /* only 10us required */
-	ts->set_reset(true);
+	tsc2005_set_reset(ts, true);
 
 	enable_irq(ts->spi->irq);
 	tsc2005_start_scan(ts);
@@ -572,29 +592,47 @@ static void tsc2005_setup_spi_xfer(struct tsc2005 *ts)
 static int tsc2005_probe(struct spi_device *spi)
 {
 	const struct tsc2005_platform_data *pdata = dev_get_platdata(&spi->dev);
+	struct device_node *np = spi->dev.of_node;
+
 	struct tsc2005 *ts;
 	struct input_dev *input_dev;
-	unsigned int max_x, max_y, max_p;
-	unsigned int fudge_x, fudge_y, fudge_p;
+	unsigned int max_x = MAX_12BIT;
+	unsigned int max_y = MAX_12BIT;
+	unsigned int max_p = MAX_12BIT;
+	unsigned int fudge_x = TSC2005_DEF_X_FUZZ;
+	unsigned int fudge_y = TSC2005_DEF_Y_FUZZ;
+	unsigned int fudge_p = TSC2005_DEF_P_FUZZ;
+	unsigned int x_plate_ohm = TSC2005_DEF_RESISTOR;
+	unsigned int esd_timeout;
 	int error;
 
-	if (!pdata) {
-		dev_dbg(&spi->dev, "no platform data\n");
+	if (!np && !pdata) {
+		dev_err(&spi->dev, "no platform data\n");
 		return -ENODEV;
 	}
 
-	fudge_x	= pdata->ts_x_fudge	   ? : 4;
-	fudge_y	= pdata->ts_y_fudge	   ? : 8;
-	fudge_p	= pdata->ts_pressure_fudge ? : 2;
-	max_x	= pdata->ts_x_max	   ? : MAX_12BIT;
-	max_y	= pdata->ts_y_max	   ? : MAX_12BIT;
-	max_p	= pdata->ts_pressure_max   ? : MAX_12BIT;
-
 	if (spi->irq <= 0) {
-		dev_dbg(&spi->dev, "no irq\n");
+		dev_err(&spi->dev, "no irq\n");
 		return -ENODEV;
 	}
 
+	if (pdata) {
+		fudge_x	= pdata->ts_x_fudge;
+		fudge_y	= pdata->ts_y_fudge;
+		fudge_p	= pdata->ts_pressure_fudge;
+		max_x	= pdata->ts_x_max;
+		max_y	= pdata->ts_y_max;
+		max_p	= pdata->ts_pressure_max;
+		x_plate_ohm = pdata->ts_x_plate_ohm;
+		esd_timeout = pdata->esd_timeout_ms;
+	} else {
+		x_plate_ohm = TSC2005_DEF_RESISTOR;
+		of_property_read_u32(np, "ti,x-plate-ohms", &x_plate_ohm);
+		esd_timeout = 0;
+		of_property_read_u32(np, "ti,esd-recovery-timeout-ms",
+								&esd_timeout);
+	}
+
 	spi->mode = SPI_MODE_0;
 	spi->bits_per_word = 8;
 	if (!spi->max_speed_hz)
@@ -604,19 +642,48 @@ static int tsc2005_probe(struct spi_device *spi)
 	if (error)
 		return error;
 
-	ts = kzalloc(sizeof(*ts), GFP_KERNEL);
-	input_dev = input_allocate_device();
-	if (!ts || !input_dev) {
-		error = -ENOMEM;
-		goto err_free_mem;
-	}
+	ts = devm_kzalloc(&spi->dev, sizeof(*ts), GFP_KERNEL);
+	if (!ts)
+		return -ENOMEM;
+
+	input_dev = devm_input_allocate_device(&spi->dev);
+	if (!input_dev)
+		return -ENOMEM;
 
 	ts->spi = spi;
 	ts->idev = input_dev;
 
-	ts->x_plate_ohm	= pdata->ts_x_plate_ohm	? : 280;
-	ts->esd_timeout	= pdata->esd_timeout_ms;
-	ts->set_reset	= pdata->set_reset;
+	ts->x_plate_ohm = x_plate_ohm;
+	ts->esd_timeout = esd_timeout;
+
+	if (np) {
+		ts->reset_gpio = of_get_named_gpio(np, "reset-gpios", 0);
+		if (ts->reset_gpio == -EPROBE_DEFER)
+			return ts->reset_gpio;
+		if (ts->reset_gpio < 0) {
+			dev_err(&spi->dev, "error acquiring reset gpio: %d\n",
+				ts->reset_gpio);
+			return ts->reset_gpio;
+		}
+
+		error = devm_gpio_request_one(&spi->dev, ts->reset_gpio, 0,
+					      "reset-gpios");
+		if (error) {
+			dev_err(&spi->dev, "error requesting reset gpio: %d\n",
+				error);
+			return error;
+		}
+
+		ts->vio = devm_regulator_get(&spi->dev, "vio");
+		if (IS_ERR(ts->vio)) {
+			error = PTR_ERR(ts->vio);
+			dev_err(&spi->dev, "vio regulator missing (%d)", error);
+			return error;
+		}
+	} else {
+		ts->reset_gpio = -1;
+		ts->set_reset = pdata->set_reset;
+	}
 
 	mutex_init(&ts->mutex);
 
@@ -641,6 +708,9 @@ static int tsc2005_probe(struct spi_device *spi)
 	input_set_abs_params(input_dev, ABS_Y, 0, max_y, fudge_y, 0);
 	input_set_abs_params(input_dev, ABS_PRESSURE, 0, max_p, fudge_p, 0);
 
+	if (np)
+		touchscreen_parse_of_params(input_dev);
+
 	input_dev->open = tsc2005_open;
 	input_dev->close = tsc2005_close;
 
@@ -649,12 +719,20 @@ static int tsc2005_probe(struct spi_device *spi)
 	/* Ensure the touchscreen is off */
 	tsc2005_stop_scan(ts);
 
-	error = request_threaded_irq(spi->irq, NULL, tsc2005_irq_thread,
-				     IRQF_TRIGGER_RISING | IRQF_ONESHOT,
-				     "tsc2005", ts);
+	error = devm_request_threaded_irq(&spi->dev, spi->irq, NULL,
+					  tsc2005_irq_thread,
+					  IRQF_TRIGGER_RISING | IRQF_ONESHOT,
+					  "tsc2005", ts);
 	if (error) {
 		dev_err(&spi->dev, "Failed to request irq, err: %d\n", error);
-		goto err_free_mem;
+		return error;
+	}
+
+	/* enable regulator for DT */
+	if (ts->vio) {
+		error = regulator_enable(ts->vio);
+		if (error)
+			return error;
 	}
 
 	spi_set_drvdata(spi, ts);
@@ -662,7 +740,7 @@ static int tsc2005_probe(struct spi_device *spi)
 	if (error) {
 		dev_err(&spi->dev,
 			"Failed to create sysfs attributes, err: %d\n", error);
-		goto err_clear_drvdata;
+		goto disable_regulator;
 	}
 
 	error = input_register_device(ts->idev);
@@ -677,11 +755,9 @@ static int tsc2005_probe(struct spi_device *spi)
 
 err_remove_sysfs:
 	sysfs_remove_group(&spi->dev.kobj, &tsc2005_attr_group);
-err_clear_drvdata:
-	free_irq(spi->irq, ts);
-err_free_mem:
-	input_free_device(input_dev);
-	kfree(ts);
+disable_regulator:
+	if (ts->vio)
+		regulator_disable(ts->vio);
 	return error;
 }
 
@@ -689,11 +765,10 @@ static int tsc2005_remove(struct spi_device *spi)
 {
 	struct tsc2005 *ts = spi_get_drvdata(spi);
 
-	sysfs_remove_group(&ts->spi->dev.kobj, &tsc2005_attr_group);
+	sysfs_remove_group(&spi->dev.kobj, &tsc2005_attr_group);
 
-	free_irq(ts->spi->irq, ts);
-	input_unregister_device(ts->idev);
-	kfree(ts);
+	if (ts->vio)
+		regulator_disable(ts->vio);
 
 	return 0;
 }
diff --git a/drivers/input/touchscreen/zforce_ts.c b/drivers/input/touchscreen/zforce_ts.c
index 01d30cedde46..feea85b52fa8 100644
--- a/drivers/input/touchscreen/zforce_ts.c
+++ b/drivers/input/touchscreen/zforce_ts.c
@@ -880,7 +880,7 @@ static struct i2c_device_id zforce_idtable[] = {
 MODULE_DEVICE_TABLE(i2c, zforce_idtable);
 
 #ifdef CONFIG_OF
-static struct of_device_id zforce_dt_idtable[] = {
+static const struct of_device_id zforce_dt_idtable[] = {
 	{ .compatible = "neonode,zforce" },
 	{},
 };
diff --git a/drivers/iommu/amd_iommu_v2.c b/drivers/iommu/amd_iommu_v2.c
index d4daa05efe60..499b4366a98d 100644
--- a/drivers/iommu/amd_iommu_v2.c
+++ b/drivers/iommu/amd_iommu_v2.c
@@ -45,7 +45,7 @@ struct pri_queue {
 struct pasid_state {
 	struct list_head list;			/* For global state-list */
 	atomic_t count;				/* Reference count */
-	atomic_t mmu_notifier_count;		/* Counting nested mmu_notifier
+	unsigned mmu_notifier_count;		/* Counting nested mmu_notifier
 						   calls */
 	struct task_struct *task;		/* Task bound to this PASID */
 	struct mm_struct *mm;			/* mm_struct for the faults */
@@ -53,7 +53,8 @@ struct pasid_state {
 	struct pri_queue pri[PRI_QUEUE_SIZE];	/* PRI tag states */
 	struct device_state *device_state;	/* Link to our device_state */
 	int pasid;				/* PASID index */
-	spinlock_t lock;			/* Protect pri_queues */
+	spinlock_t lock;			/* Protect pri_queues and
+						   mmu_notifer_count */
 	wait_queue_head_t wq;			/* To wait for count == 0 */
 };
 
@@ -431,15 +432,19 @@ static void mn_invalidate_range_start(struct mmu_notifier *mn,
 {
 	struct pasid_state *pasid_state;
 	struct device_state *dev_state;
+	unsigned long flags;
 
 	pasid_state = mn_to_state(mn);
 	dev_state   = pasid_state->device_state;
 
-	if (atomic_add_return(1, &pasid_state->mmu_notifier_count) == 1) {
+	spin_lock_irqsave(&pasid_state->lock, flags);
+	if (pasid_state->mmu_notifier_count == 0) {
 		amd_iommu_domain_set_gcr3(dev_state->domain,
 					  pasid_state->pasid,
 					  __pa(empty_page_table));
 	}
+	pasid_state->mmu_notifier_count += 1;
+	spin_unlock_irqrestore(&pasid_state->lock, flags);
 }
 
 static void mn_invalidate_range_end(struct mmu_notifier *mn,
@@ -448,15 +453,19 @@ static void mn_invalidate_range_end(struct mmu_notifier *mn,
 {
 	struct pasid_state *pasid_state;
 	struct device_state *dev_state;
+	unsigned long flags;
 
 	pasid_state = mn_to_state(mn);
 	dev_state   = pasid_state->device_state;
 
-	if (atomic_dec_and_test(&pasid_state->mmu_notifier_count)) {
+	spin_lock_irqsave(&pasid_state->lock, flags);
+	pasid_state->mmu_notifier_count -= 1;
+	if (pasid_state->mmu_notifier_count == 0) {
 		amd_iommu_domain_set_gcr3(dev_state->domain,
 					  pasid_state->pasid,
 					  __pa(pasid_state->mm->pgd));
 	}
+	spin_unlock_irqrestore(&pasid_state->lock, flags);
 }
 
 static void mn_release(struct mmu_notifier *mn, struct mm_struct *mm)
@@ -650,7 +659,6 @@ int amd_iommu_bind_pasid(struct pci_dev *pdev, int pasid,
 		goto out;
 
 	atomic_set(&pasid_state->count, 1);
-	atomic_set(&pasid_state->mmu_notifier_count, 0);
 	init_waitqueue_head(&pasid_state->wq);
 	spin_lock_init(&pasid_state->lock);
 
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 6bb32773c3ac..51b6b77dc3e5 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -3816,14 +3816,11 @@ int dmar_iommu_notify_scope_dev(struct dmar_pci_notify_info *info)
 				((void *)rmrr) + rmrr->header.length,
 				rmrr->segment, rmrru->devices,
 				rmrru->devices_cnt);
-			if (ret > 0)
-				break;
-			else if(ret < 0)
+			if(ret < 0)
 				return ret;
 		} else if (info->event == BUS_NOTIFY_DEL_DEVICE) {
-			if (dmar_remove_dev_scope(info, rmrr->segment,
-				rmrru->devices, rmrru->devices_cnt))
-				break;
+			dmar_remove_dev_scope(info, rmrr->segment,
+				rmrru->devices, rmrru->devices_cnt);
 		}
 	}
 
diff --git a/drivers/irqchip/irq-armada-370-xp.c b/drivers/irqchip/irq-armada-370-xp.c
index c887e6eebc41..574aba0eba4e 100644
--- a/drivers/irqchip/irq-armada-370-xp.c
+++ b/drivers/irqchip/irq-armada-370-xp.c
@@ -334,6 +334,15 @@ static void armada_mpic_send_doorbell(const struct cpumask *mask,
 
 static void armada_xp_mpic_smp_cpu_init(void)
 {
+	u32 control;
+	int nr_irqs, i;
+
+	control = readl(main_int_base + ARMADA_370_XP_INT_CONTROL);
+	nr_irqs = (control >> 2) & 0x3ff;
+
+	for (i = 0; i < nr_irqs; i++)
+		writel(i, per_cpu_int_base + ARMADA_370_XP_INT_SET_MASK_OFFS);
+
 	/* Clear pending IPIs */
 	writel(0, per_cpu_int_base + ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS);
 
@@ -474,7 +483,7 @@ static int __init armada_370_xp_mpic_of_init(struct device_node *node,
 					     struct device_node *parent)
 {
 	struct resource main_int_res, per_cpu_int_res;
-	int parent_irq;
+	int parent_irq, nr_irqs, i;
 	u32 control;
 
 	BUG_ON(of_address_to_resource(node, 0, &main_int_res));
@@ -496,9 +505,13 @@ static int __init armada_370_xp_mpic_of_init(struct device_node *node,
 	BUG_ON(!per_cpu_int_base);
 
 	control = readl(main_int_base + ARMADA_370_XP_INT_CONTROL);
+	nr_irqs = (control >> 2) & 0x3ff;
+
+	for (i = 0; i < nr_irqs; i++)
+		writel(i, main_int_base + ARMADA_370_XP_INT_CLEAR_ENABLE_OFFS);
 
 	armada_370_xp_mpic_domain =
-		irq_domain_add_linear(node, (control >> 2) & 0x3ff,
+		irq_domain_add_linear(node, nr_irqs,
 				&armada_370_xp_mpic_irq_ops, NULL);
 
 	BUG_ON(!armada_370_xp_mpic_domain);
diff --git a/drivers/irqchip/irq-brcmstb-l2.c b/drivers/irqchip/irq-brcmstb-l2.c
index 8ee2a36d5840..c15c840987d2 100644
--- a/drivers/irqchip/irq-brcmstb-l2.c
+++ b/drivers/irqchip/irq-brcmstb-l2.c
@@ -150,7 +150,7 @@ int __init brcmstb_l2_intc_of_init(struct device_node *np,
 
 	/* Allocate a single Generic IRQ chip for this node */
 	ret = irq_alloc_domain_generic_chips(data->domain, 32, 1,
-				np->full_name, handle_level_irq, clr, 0, 0);
+				np->full_name, handle_edge_irq, clr, 0, 0);
 	if (ret) {
 		pr_err("failed to allocate generic irq chip\n");
 		goto out_free_domain;
diff --git a/drivers/irqchip/spear-shirq.c b/drivers/irqchip/spear-shirq.c
index 3fdda3a40269..6ce6bd3441bf 100644
--- a/drivers/irqchip/spear-shirq.c
+++ b/drivers/irqchip/spear-shirq.c
@@ -125,7 +125,7 @@ static struct spear_shirq spear320_shirq_ras2 = {
 };
 
 static struct spear_shirq spear320_shirq_ras3 = {
-	.irq_nr = 3,
+	.irq_nr = 7,
 	.irq_bit_off = 0,
 	.invalid_irq = 1,
 	.regs = {
diff --git a/drivers/isdn/capi/Kconfig b/drivers/isdn/capi/Kconfig
index 9816c51eb5c2..7641b3096ea6 100644
--- a/drivers/isdn/capi/Kconfig
+++ b/drivers/isdn/capi/Kconfig
@@ -1,11 +1,3 @@
-config ISDN_DRV_AVMB1_VERBOSE_REASON
-	bool "Verbose reason code reporting"
-	default y
-	help
-	  If you say Y here, the CAPI drivers will give verbose reasons for
-	  disconnecting. This will increase the size of the kernel by 7 KB. If
-	  unsure, say Y.
-
 config CAPI_TRACE
 	bool "CAPI trace support"
 	default y
@@ -17,7 +9,7 @@ config CAPI_TRACE
 	  If unsure, say Y.
 
 config ISDN_CAPI_CAPI20
-	tristate "CAPI2.0 /dev/capi support"
+	tristate "CAPI2.0 /dev/capi20 support"
 	help
 	  This option will provide the CAPI 2.0 interface to userspace
 	  applications via /dev/capi20. Applications should use the
@@ -42,3 +34,11 @@ config ISDN_CAPI_CAPIDRV
 	  the legacy isdn4linux link layer.  If you have a card which is
 	  supported by a CAPI driver, but still want to use old features like
 	  ippp interfaces or ttyI emulation, say Y/M here.
+
+config ISDN_CAPI_CAPIDRV_VERBOSE
+	bool "Verbose reason code reporting"
+	depends on ISDN_CAPI_CAPIDRV
+	help
+	  If you say Y here, the capidrv interface will give verbose reasons
+	  for disconnecting. This will increase the size of the kernel by 7 KB.
+	  If unsure, say N.
diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c
index ac6f72b455d1..f9a87ed2392b 100644
--- a/drivers/isdn/capi/capi.c
+++ b/drivers/isdn/capi/capi.c
@@ -1271,7 +1271,7 @@ static int __init capinc_tty_init(void)
 		return -ENOMEM;
 	}
 	drv->driver_name = "capi_nc";
-	drv->name = "capi";
+	drv->name = "capi!";
 	drv->major = 0;
 	drv->minor_start = 0;
 	drv->type = TTY_DRIVER_TYPE_SERIAL;
@@ -1417,7 +1417,7 @@ static int __init capi_init(void)
 		return PTR_ERR(capi_class);
 	}
 
-	device_create(capi_class, NULL, MKDEV(capi_major, 0), NULL, "capi");
+	device_create(capi_class, NULL, MKDEV(capi_major, 0), NULL, "capi20");
 
 	if (capinc_tty_init() < 0) {
 		device_destroy(capi_class, MKDEV(capi_major, 0));
diff --git a/drivers/isdn/capi/capidrv.c b/drivers/isdn/capi/capidrv.c
index cc9f1927a322..fd6d28f3fc36 100644
--- a/drivers/isdn/capi/capidrv.c
+++ b/drivers/isdn/capi/capidrv.c
@@ -763,6 +763,201 @@ static inline int new_bchan(capidrv_contr *card)
 }
 
 /* ------------------------------------------------------------------- */
+static char *capi_info2str(u16 reason)
+{
+#ifndef CONFIG_ISDN_CAPI_CAPIDRV_VERBOSE
+	return "..";
+#else
+	switch (reason) {
+
+/*-- informative values (corresponding message was processed) -----*/
+	case 0x0001:
+		return "NCPI not supported by current protocol, NCPI ignored";
+	case 0x0002:
+		return "Flags not supported by current protocol, flags ignored";
+	case 0x0003:
+		return "Alert already sent by another application";
+
+/*-- error information concerning CAPI_REGISTER -----*/
+	case 0x1001:
+		return "Too many applications";
+	case 0x1002:
+		return "Logical block size too small, must be at least 128 Bytes";
+	case 0x1003:
+		return "Buffer exceeds 64 kByte";
+	case 0x1004:
+		return "Message buffer size too small, must be at least 1024 Bytes";
+	case 0x1005:
+		return "Max. number of logical connections not supported";
+	case 0x1006:
+		return "Reserved";
+	case 0x1007:
+		return "The message could not be accepted because of an internal busy condition";
+	case 0x1008:
+		return "OS resource error (no memory ?)";
+	case 0x1009:
+		return "CAPI not installed";
+	case 0x100A:
+		return "Controller does not support external equipment";
+	case 0x100B:
+		return "Controller does only support external equipment";
+
+/*-- error information concerning message exchange functions -----*/
+	case 0x1101:
+		return "Illegal application number";
+	case 0x1102:
+		return "Illegal command or subcommand or message length less than 12 bytes";
+	case 0x1103:
+		return "The message could not be accepted because of a queue full condition !! The error code does not imply that CAPI cannot receive messages directed to another controller, PLCI or NCCI";
+	case 0x1104:
+		return "Queue is empty";
+	case 0x1105:
+		return "Queue overflow, a message was lost !! This indicates a configuration error. The only recovery from this error is to perform a CAPI_RELEASE";
+	case 0x1106:
+		return "Unknown notification parameter";
+	case 0x1107:
+		return "The Message could not be accepted because of an internal busy condition";
+	case 0x1108:
+		return "OS Resource error (no memory ?)";
+	case 0x1109:
+		return "CAPI not installed";
+	case 0x110A:
+		return "Controller does not support external equipment";
+	case 0x110B:
+		return "Controller does only support external equipment";
+
+/*-- error information concerning resource / coding problems -----*/
+	case 0x2001:
+		return "Message not supported in current state";
+	case 0x2002:
+		return "Illegal Controller / PLCI / NCCI";
+	case 0x2003:
+		return "Out of PLCI";
+	case 0x2004:
+		return "Out of NCCI";
+	case 0x2005:
+		return "Out of LISTEN";
+	case 0x2006:
+		return "Out of FAX resources (protocol T.30)";
+	case 0x2007:
+		return "Illegal message parameter coding";
+
+/*-- error information concerning requested services  -----*/
+	case 0x3001:
+		return "B1 protocol not supported";
+	case 0x3002:
+		return "B2 protocol not supported";
+	case 0x3003:
+		return "B3 protocol not supported";
+	case 0x3004:
+		return "B1 protocol parameter not supported";
+	case 0x3005:
+		return "B2 protocol parameter not supported";
+	case 0x3006:
+		return "B3 protocol parameter not supported";
+	case 0x3007:
+		return "B protocol combination not supported";
+	case 0x3008:
+		return "NCPI not supported";
+	case 0x3009:
+		return "CIP Value unknown";
+	case 0x300A:
+		return "Flags not supported (reserved bits)";
+	case 0x300B:
+		return "Facility not supported";
+	case 0x300C:
+		return "Data length not supported by current protocol";
+	case 0x300D:
+		return "Reset procedure not supported by current protocol";
+
+/*-- informations about the clearing of a physical connection -----*/
+	case 0x3301:
+		return "Protocol error layer 1 (broken line or B-channel removed by signalling protocol)";
+	case 0x3302:
+		return "Protocol error layer 2";
+	case 0x3303:
+		return "Protocol error layer 3";
+	case 0x3304:
+		return "Another application got that call";
+/*-- T.30 specific reasons -----*/
+	case 0x3311:
+		return "Connecting not successful (remote station is no FAX G3 machine)";
+	case 0x3312:
+		return "Connecting not successful (training error)";
+	case 0x3313:
+		return "Disconnected before transfer (remote station does not support transfer mode, e.g. resolution)";
+	case 0x3314:
+		return "Disconnected during transfer (remote abort)";
+	case 0x3315:
+		return "Disconnected during transfer (remote procedure error, e.g. unsuccessful repetition of T.30 commands)";
+	case 0x3316:
+		return "Disconnected during transfer (local tx data underrun)";
+	case 0x3317:
+		return "Disconnected during transfer (local rx data overflow)";
+	case 0x3318:
+		return "Disconnected during transfer (local abort)";
+	case 0x3319:
+		return "Illegal parameter coding (e.g. SFF coding error)";
+
+/*-- disconnect causes from the network according to ETS 300 102-1/Q.931 -----*/
+	case 0x3481: return "Unallocated (unassigned) number";
+	case 0x3482: return "No route to specified transit network";
+	case 0x3483: return "No route to destination";
+	case 0x3486: return "Channel unacceptable";
+	case 0x3487:
+		return "Call awarded and being delivered in an established channel";
+	case 0x3490: return "Normal call clearing";
+	case 0x3491: return "User busy";
+	case 0x3492: return "No user responding";
+	case 0x3493: return "No answer from user (user alerted)";
+	case 0x3495: return "Call rejected";
+	case 0x3496: return "Number changed";
+	case 0x349A: return "Non-selected user clearing";
+	case 0x349B: return "Destination out of order";
+	case 0x349C: return "Invalid number format";
+	case 0x349D: return "Facility rejected";
+	case 0x349E: return "Response to STATUS ENQUIRY";
+	case 0x349F: return "Normal, unspecified";
+	case 0x34A2: return "No circuit / channel available";
+	case 0x34A6: return "Network out of order";
+	case 0x34A9: return "Temporary failure";
+	case 0x34AA: return "Switching equipment congestion";
+	case 0x34AB: return "Access information discarded";
+	case 0x34AC: return "Requested circuit / channel not available";
+	case 0x34AF: return "Resources unavailable, unspecified";
+	case 0x34B1: return "Quality of service unavailable";
+	case 0x34B2: return "Requested facility not subscribed";
+	case 0x34B9: return "Bearer capability not authorized";
+	case 0x34BA: return "Bearer capability not presently available";
+	case 0x34BF: return "Service or option not available, unspecified";
+	case 0x34C1: return "Bearer capability not implemented";
+	case 0x34C2: return "Channel type not implemented";
+	case 0x34C5: return "Requested facility not implemented";
+	case 0x34C6: return "Only restricted digital information bearer capability is available";
+	case 0x34CF: return "Service or option not implemented, unspecified";
+	case 0x34D1: return "Invalid call reference value";
+	case 0x34D2: return "Identified channel does not exist";
+	case 0x34D3: return "A suspended call exists, but this call identity does not";
+	case 0x34D4: return "Call identity in use";
+	case 0x34D5: return "No call suspended";
+	case 0x34D6: return "Call having the requested call identity has been cleared";
+	case 0x34D8: return "Incompatible destination";
+	case 0x34DB: return "Invalid transit network selection";
+	case 0x34DF: return "Invalid message, unspecified";
+	case 0x34E0: return "Mandatory information element is missing";
+	case 0x34E1: return "Message type non-existent or not implemented";
+	case 0x34E2: return "Message not compatible with call state or message type non-existent or not implemented";
+	case 0x34E3: return "Information element non-existent or not implemented";
+	case 0x34E4: return "Invalid information element contents";
+	case 0x34E5: return "Message not compatible with call state";
+	case 0x34E6: return "Recovery on timer expiry";
+	case 0x34EF: return "Protocol error, unspecified";
+	case 0x34FF: return "Interworking, unspecified";
+
+	default: return "No additional information";
+	}
+#endif
+}
 
 static void handle_controller(_cmsg *cmsg)
 {
diff --git a/drivers/isdn/capi/capiutil.c b/drivers/isdn/capi/capiutil.c
index d26f17033b68..6e797e502cfa 100644
--- a/drivers/isdn/capi/capiutil.c
+++ b/drivers/isdn/capi/capiutil.c
@@ -22,205 +22,6 @@
 
 /* from CAPI2.0 DDK AVM Berlin GmbH */
 
-#ifndef CONFIG_ISDN_DRV_AVMB1_VERBOSE_REASON
-char *capi_info2str(u16 reason)
-{
-	return "..";
-}
-#else
-char *capi_info2str(u16 reason)
-{
-	switch (reason) {
-
-/*-- informative values (corresponding message was processed) -----*/
-	case 0x0001:
-		return "NCPI not supported by current protocol, NCPI ignored";
-	case 0x0002:
-		return "Flags not supported by current protocol, flags ignored";
-	case 0x0003:
-		return "Alert already sent by another application";
-
-/*-- error information concerning CAPI_REGISTER -----*/
-	case 0x1001:
-		return "Too many applications";
-	case 0x1002:
-		return "Logical block size too small, must be at least 128 Bytes";
-	case 0x1003:
-		return "Buffer exceeds 64 kByte";
-	case 0x1004:
-		return "Message buffer size too small, must be at least 1024 Bytes";
-	case 0x1005:
-		return "Max. number of logical connections not supported";
-	case 0x1006:
-		return "Reserved";
-	case 0x1007:
-		return "The message could not be accepted because of an internal busy condition";
-	case 0x1008:
-		return "OS resource error (no memory ?)";
-	case 0x1009:
-		return "CAPI not installed";
-	case 0x100A:
-		return "Controller does not support external equipment";
-	case 0x100B:
-		return "Controller does only support external equipment";
-
-/*-- error information concerning message exchange functions -----*/
-	case 0x1101:
-		return "Illegal application number";
-	case 0x1102:
-		return "Illegal command or subcommand or message length less than 12 bytes";
-	case 0x1103:
-		return "The message could not be accepted because of a queue full condition !! The error code does not imply that CAPI cannot receive messages directed to another controller, PLCI or NCCI";
-	case 0x1104:
-		return "Queue is empty";
-	case 0x1105:
-		return "Queue overflow, a message was lost !! This indicates a configuration error. The only recovery from this error is to perform a CAPI_RELEASE";
-	case 0x1106:
-		return "Unknown notification parameter";
-	case 0x1107:
-		return "The Message could not be accepted because of an internal busy condition";
-	case 0x1108:
-		return "OS Resource error (no memory ?)";
-	case 0x1109:
-		return "CAPI not installed";
-	case 0x110A:
-		return "Controller does not support external equipment";
-	case 0x110B:
-		return "Controller does only support external equipment";
-
-/*-- error information concerning resource / coding problems -----*/
-	case 0x2001:
-		return "Message not supported in current state";
-	case 0x2002:
-		return "Illegal Controller / PLCI / NCCI";
-	case 0x2003:
-		return "Out of PLCI";
-	case 0x2004:
-		return "Out of NCCI";
-	case 0x2005:
-		return "Out of LISTEN";
-	case 0x2006:
-		return "Out of FAX resources (protocol T.30)";
-	case 0x2007:
-		return "Illegal message parameter coding";
-
-/*-- error information concerning requested services  -----*/
-	case 0x3001:
-		return "B1 protocol not supported";
-	case 0x3002:
-		return "B2 protocol not supported";
-	case 0x3003:
-		return "B3 protocol not supported";
-	case 0x3004:
-		return "B1 protocol parameter not supported";
-	case 0x3005:
-		return "B2 protocol parameter not supported";
-	case 0x3006:
-		return "B3 protocol parameter not supported";
-	case 0x3007:
-		return "B protocol combination not supported";
-	case 0x3008:
-		return "NCPI not supported";
-	case 0x3009:
-		return "CIP Value unknown";
-	case 0x300A:
-		return "Flags not supported (reserved bits)";
-	case 0x300B:
-		return "Facility not supported";
-	case 0x300C:
-		return "Data length not supported by current protocol";
-	case 0x300D:
-		return "Reset procedure not supported by current protocol";
-
-/*-- informations about the clearing of a physical connection -----*/
-	case 0x3301:
-		return "Protocol error layer 1 (broken line or B-channel removed by signalling protocol)";
-	case 0x3302:
-		return "Protocol error layer 2";
-	case 0x3303:
-		return "Protocol error layer 3";
-	case 0x3304:
-		return "Another application got that call";
-/*-- T.30 specific reasons -----*/
-	case 0x3311:
-		return "Connecting not successful (remote station is no FAX G3 machine)";
-	case 0x3312:
-		return "Connecting not successful (training error)";
-	case 0x3313:
-		return "Disconnected before transfer (remote station does not support transfer mode, e.g. resolution)";
-	case 0x3314:
-		return "Disconnected during transfer (remote abort)";
-	case 0x3315:
-		return "Disconnected during transfer (remote procedure error, e.g. unsuccessful repetition of T.30 commands)";
-	case 0x3316:
-		return "Disconnected during transfer (local tx data underrun)";
-	case 0x3317:
-		return "Disconnected during transfer (local rx data overflow)";
-	case 0x3318:
-		return "Disconnected during transfer (local abort)";
-	case 0x3319:
-		return "Illegal parameter coding (e.g. SFF coding error)";
-
-/*-- disconnect causes from the network according to ETS 300 102-1/Q.931 -----*/
-	case 0x3481: return "Unallocated (unassigned) number";
-	case 0x3482: return "No route to specified transit network";
-	case 0x3483: return "No route to destination";
-	case 0x3486: return "Channel unacceptable";
-	case 0x3487:
-		return "Call awarded and being delivered in an established channel";
-	case 0x3490: return "Normal call clearing";
-	case 0x3491: return "User busy";
-	case 0x3492: return "No user responding";
-	case 0x3493: return "No answer from user (user alerted)";
-	case 0x3495: return "Call rejected";
-	case 0x3496: return "Number changed";
-	case 0x349A: return "Non-selected user clearing";
-	case 0x349B: return "Destination out of order";
-	case 0x349C: return "Invalid number format";
-	case 0x349D: return "Facility rejected";
-	case 0x349E: return "Response to STATUS ENQUIRY";
-	case 0x349F: return "Normal, unspecified";
-	case 0x34A2: return "No circuit / channel available";
-	case 0x34A6: return "Network out of order";
-	case 0x34A9: return "Temporary failure";
-	case 0x34AA: return "Switching equipment congestion";
-	case 0x34AB: return "Access information discarded";
-	case 0x34AC: return "Requested circuit / channel not available";
-	case 0x34AF: return "Resources unavailable, unspecified";
-	case 0x34B1: return "Quality of service unavailable";
-	case 0x34B2: return "Requested facility not subscribed";
-	case 0x34B9: return "Bearer capability not authorized";
-	case 0x34BA: return "Bearer capability not presently available";
-	case 0x34BF: return "Service or option not available, unspecified";
-	case 0x34C1: return "Bearer capability not implemented";
-	case 0x34C2: return "Channel type not implemented";
-	case 0x34C5: return "Requested facility not implemented";
-	case 0x34C6: return "Only restricted digital information bearer capability is available";
-	case 0x34CF: return "Service or option not implemented, unspecified";
-	case 0x34D1: return "Invalid call reference value";
-	case 0x34D2: return "Identified channel does not exist";
-	case 0x34D3: return "A suspended call exists, but this call identity does not";
-	case 0x34D4: return "Call identity in use";
-	case 0x34D5: return "No call suspended";
-	case 0x34D6: return "Call having the requested call identity has been cleared";
-	case 0x34D8: return "Incompatible destination";
-	case 0x34DB: return "Invalid transit network selection";
-	case 0x34DF: return "Invalid message, unspecified";
-	case 0x34E0: return "Mandatory information element is missing";
-	case 0x34E1: return "Message type non-existent or not implemented";
-	case 0x34E2: return "Message not compatible with call state or message type non-existent or not implemented";
-	case 0x34E3: return "Information element non-existent or not implemented";
-	case 0x34E4: return "Invalid information element contents";
-	case 0x34E5: return "Message not compatible with call state";
-	case 0x34E6: return "Recovery on timer expiry";
-	case 0x34EF: return "Protocol error, unspecified";
-	case 0x34FF: return "Interworking, unspecified";
-
-	default: return "No additional information";
-	}
-}
-#endif
-
 typedef struct {
 	int typ;
 	size_t off;
@@ -1073,4 +874,3 @@ EXPORT_SYMBOL(capi_cmsg_header);
 EXPORT_SYMBOL(capi_cmd2str);
 EXPORT_SYMBOL(capi_cmsg2str);
 EXPORT_SYMBOL(capi_message2str);
-EXPORT_SYMBOL(capi_info2str);
diff --git a/drivers/isdn/hisax/Kconfig b/drivers/isdn/hisax/Kconfig
index d9edcc94c2a8..97465ac5a2d5 100644
--- a/drivers/isdn/hisax/Kconfig
+++ b/drivers/isdn/hisax/Kconfig
@@ -16,7 +16,7 @@ config ISDN_DRV_HISAX
 	  also to the configuration option of the driver for your particular
 	  card, below.
 
-if ISDN_DRV_HISAX!=n
+if ISDN_DRV_HISAX
 
 comment "D-channel protocol features"
 
@@ -348,10 +348,6 @@ config HISAX_ENTERNOW_PCI
 	  This enables HiSax support for the Formula-n enter:now PCI
 	  ISDN card.
 
-endif
-
-if ISDN_DRV_HISAX
-
 config HISAX_DEBUG
 	bool "HiSax debugging"
 	help
@@ -420,11 +416,6 @@ config HISAX_FRITZ_PCIPNP
 	  (the latter also needs you to select "ISA Plug and Play support"
 	  from the menu "Plug and Play configuration")
 
-config HISAX_AVM_A1_PCMCIA
-	bool
-	depends on HISAX_AVM_A1_CS
-	default y
-
 endif
 
 endmenu
diff --git a/drivers/isdn/hisax/hfc4s8s_l1.c b/drivers/isdn/hisax/hfc4s8s_l1.c
index 414dbf6da89a..fc9f9d03fa13 100644
--- a/drivers/isdn/hisax/hfc4s8s_l1.c
+++ b/drivers/isdn/hisax/hfc4s8s_l1.c
@@ -197,25 +197,6 @@ typedef struct _hfc4s8s_hw {
 
 
 
-/***************************/
-/* inline function defines */
-/***************************/
-#ifdef HISAX_HFC4S8S_PCIMEM	/* inline functions memory mapped */
-
-/* memory write and dummy IO read to avoid PCI byte merge problems */
-#define Write_hfc8(a, b, c) {(*((volatile u_char *)(a->membase + b)) = c); inb(a->iobase + 4);}
-/* memory write without dummy IO access for fifo data access */
-#define fWrite_hfc8(a, b, c) (*((volatile u_char *)(a->membase + b)) = c)
-#define Read_hfc8(a, b) (*((volatile u_char *)(a->membase + b)))
-#define Write_hfc16(a, b, c) (*((volatile unsigned short *)(a->membase + b)) = c)
-#define Read_hfc16(a, b) (*((volatile unsigned short *)(a->membase + b)))
-#define Write_hfc32(a, b, c) (*((volatile unsigned long *)(a->membase + b)) = c)
-#define Read_hfc32(a, b) (*((volatile unsigned long *)(a->membase + b)))
-#define wait_busy(a) {while ((Read_hfc8(a, R_STATUS) & M_BUSY));}
-#define PCI_ENA_MEMIO	0x03
-
-#else
-
 /* inline functions io mapped */
 static inline void
 SetRegAddr(hfc4s8s_hw *a, u_char b)
@@ -306,8 +287,6 @@ wait_busy(hfc4s8s_hw *a)
 
 #define PCI_ENA_REGIO	0x01
 
-#endif				/* HISAX_HFC4S8S_PCIMEM */
-
 /******************************************************/
 /* function to read critical counter registers that   */
 /* may be updated by the chip during read             */
@@ -724,26 +703,15 @@ rx_d_frame(struct hfc4s8s_l1 *l1p, int ech)
 				return;
 			} else {
 				/* read errornous D frame */
-
-#ifndef HISAX_HFC4S8S_PCIMEM
 				SetRegAddr(l1p->hw, A_FIFO_DATA0);
-#endif
 
 				while (z1 >= 4) {
-#ifdef HISAX_HFC4S8S_PCIMEM
-					Read_hfc32(l1p->hw, A_FIFO_DATA0);
-#else
 					fRead_hfc32(l1p->hw);
-#endif
 					z1 -= 4;
 				}
 
 				while (z1--)
-#ifdef HISAX_HFC4S8S_PCIMEM
-					Read_hfc8(l1p->hw, A_FIFO_DATA0);
-#else
-				fRead_hfc8(l1p->hw);
-#endif
+					fRead_hfc8(l1p->hw);
 
 				Write_hfc8(l1p->hw, A_INC_RES_FIFO, 1);
 				wait_busy(l1p->hw);
@@ -753,27 +721,16 @@ rx_d_frame(struct hfc4s8s_l1 *l1p, int ech)
 
 		cp = skb->data;
 
-#ifndef HISAX_HFC4S8S_PCIMEM
 		SetRegAddr(l1p->hw, A_FIFO_DATA0);
-#endif
 
 		while (z1 >= 4) {
-#ifdef HISAX_HFC4S8S_PCIMEM
-			*((unsigned long *) cp) =
-				Read_hfc32(l1p->hw, A_FIFO_DATA0);
-#else
 			*((unsigned long *) cp) = fRead_hfc32(l1p->hw);
-#endif
 			cp += 4;
 			z1 -= 4;
 		}
 
 		while (z1--)
-#ifdef HISAX_HFC4S8S_PCIMEM
-			*cp++ = Read_hfc8(l1p->hw, A_FIFO_DATA0);
-#else
-		*cp++ = fRead_hfc8(l1p->hw);
-#endif
+			*cp++ = fRead_hfc8(l1p->hw);
 
 		Write_hfc8(l1p->hw, A_INC_RES_FIFO, 1);	/* increment f counter */
 		wait_busy(l1p->hw);
@@ -859,28 +816,17 @@ rx_b_frame(struct hfc4s8s_btype *bch)
 			wait_busy(l1->hw);
 			return;
 		}
-#ifndef HISAX_HFC4S8S_PCIMEM
 		SetRegAddr(l1->hw, A_FIFO_DATA0);
-#endif
 
 		while (z1 >= 4) {
-#ifdef HISAX_HFC4S8S_PCIMEM
-			*((unsigned long *) bch->rx_ptr) =
-				Read_hfc32(l1->hw, A_FIFO_DATA0);
-#else
 			*((unsigned long *) bch->rx_ptr) =
 				fRead_hfc32(l1->hw);
-#endif
 			bch->rx_ptr += 4;
 			z1 -= 4;
 		}
 
 		while (z1--)
-#ifdef HISAX_HFC4S8S_PCIMEM
-			*(bch->rx_ptr++) = Read_hfc8(l1->hw, A_FIFO_DATA0);
-#else
-		*(bch->rx_ptr++) = fRead_hfc8(l1->hw);
-#endif
+			*(bch->rx_ptr++) = fRead_hfc8(l1->hw);
 
 		if (hdlc_complete) {
 			/* increment f counter */
@@ -940,29 +886,17 @@ tx_d_frame(struct hfc4s8s_l1 *l1p)
 	if ((skb = skb_dequeue(&l1p->d_tx_queue))) {
 		cp = skb->data;
 		cnt = skb->len;
-#ifndef HISAX_HFC4S8S_PCIMEM
 		SetRegAddr(l1p->hw, A_FIFO_DATA0);
-#endif
 
 		while (cnt >= 4) {
-#ifdef HISAX_HFC4S8S_PCIMEM
-			fWrite_hfc32(l1p->hw, A_FIFO_DATA0,
-				     *(unsigned long *) cp);
-#else
 			SetRegAddr(l1p->hw, A_FIFO_DATA0);
 			fWrite_hfc32(l1p->hw, *(unsigned long *) cp);
-#endif
 			cp += 4;
 			cnt -= 4;
 		}
 
-#ifdef HISAX_HFC4S8S_PCIMEM
-		while (cnt--)
-			fWrite_hfc8(l1p->hw, A_FIFO_DATA0, *cp++);
-#else
 		while (cnt--)
 			fWrite_hfc8(l1p->hw, *cp++);
-#endif
 
 		l1p->tx_cnt = skb->truesize;
 		Write_hfc8(l1p->hw, A_INC_RES_FIFO, 1);	/* increment f counter */
@@ -1037,26 +971,15 @@ tx_b_frame(struct hfc4s8s_btype *bch)
 		cp = skb->data + bch->tx_cnt;
 		bch->tx_cnt += cnt;
 
-#ifndef HISAX_HFC4S8S_PCIMEM
 		SetRegAddr(l1->hw, A_FIFO_DATA0);
-#endif
 		while (cnt >= 4) {
-#ifdef HISAX_HFC4S8S_PCIMEM
-			fWrite_hfc32(l1->hw, A_FIFO_DATA0,
-				     *(unsigned long *) cp);
-#else
 			fWrite_hfc32(l1->hw, *(unsigned long *) cp);
-#endif
 			cp += 4;
 			cnt -= 4;
 		}
 
 		while (cnt--)
-#ifdef HISAX_HFC4S8S_PCIMEM
-			fWrite_hfc8(l1->hw, A_FIFO_DATA0, *cp++);
-#else
-		fWrite_hfc8(l1->hw, *cp++);
-#endif
+			fWrite_hfc8(l1->hw, *cp++);
 
 		if (bch->tx_cnt >= skb->len) {
 			if (bch->mode == L1_MODE_HDLC) {
@@ -1281,10 +1204,8 @@ hfc4s8s_interrupt(int intno, void *dev_id)
 	if (!hw || !(hw->mr.r_irq_ctrl & M_GLOB_IRQ_EN))
 		return IRQ_NONE;
 
-#ifndef	HISAX_HFC4S8S_PCIMEM
 	/* read current selected regsister */
 	old_ioreg = GetRegAddr(hw);
-#endif
 
 	/* Layer 1 State change */
 	hw->mr.r_irq_statech |=
@@ -1292,9 +1213,7 @@ hfc4s8s_interrupt(int intno, void *dev_id)
 	if (!
 	    (b = (Read_hfc8(hw, R_STATUS) & (M_MISC_IRQSTA | M_FR_IRQSTA)))
 	    && !hw->mr.r_irq_statech) {
-#ifndef	HISAX_HFC4S8S_PCIMEM
 		SetRegAddr(hw, old_ioreg);
-#endif
 		return IRQ_NONE;
 	}
 
@@ -1322,9 +1241,7 @@ hfc4s8s_interrupt(int intno, void *dev_id)
 	/* queue the request to allow other cards to interrupt */
 	schedule_work(&hw->tqueue);
 
-#ifndef	HISAX_HFC4S8S_PCIMEM
 	SetRegAddr(hw, old_ioreg);
-#endif
 	return IRQ_HANDLED;
 }				/* hfc4s8s_interrupt */
 
@@ -1471,13 +1388,8 @@ static void
 release_pci_ports(hfc4s8s_hw *hw)
 {
 	pci_write_config_word(hw->pdev, PCI_COMMAND, 0);
-#ifdef HISAX_HFC4S8S_PCIMEM
-	if (hw->membase)
-		iounmap((void *) hw->membase);
-#else
 	if (hw->iobase)
 		release_region(hw->iobase, 8);
-#endif
 }
 
 /*****************************************/
@@ -1486,11 +1398,7 @@ release_pci_ports(hfc4s8s_hw *hw)
 static void
 enable_pci_ports(hfc4s8s_hw *hw)
 {
-#ifdef HISAX_HFC4S8S_PCIMEM
-	pci_write_config_word(hw->pdev, PCI_COMMAND, PCI_ENA_MEMIO);
-#else
 	pci_write_config_word(hw->pdev, PCI_COMMAND, PCI_ENA_REGIO);
-#endif
 }
 
 /*************************************/
@@ -1561,15 +1469,9 @@ setup_instance(hfc4s8s_hw *hw)
 		       hw->irq);
 		goto out;
 	}
-#ifdef HISAX_HFC4S8S_PCIMEM
-	printk(KERN_INFO
-	       "HFC-4S/8S: found PCI card at membase 0x%p, irq %d\n",
-	       hw->hw_membase, hw->irq);
-#else
 	printk(KERN_INFO
 	       "HFC-4S/8S: found PCI card at iobase 0x%x, irq %d\n",
 	       hw->iobase, hw->irq);
-#endif
 
 	hfc_hardware_enable(hw, 1, 0);
 
@@ -1614,17 +1516,12 @@ hfc4s8s_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	hw->irq = pdev->irq;
 	hw->iobase = pci_resource_start(pdev, 0);
 
-#ifdef HISAX_HFC4S8S_PCIMEM
-	hw->hw_membase = (u_char *) pci_resource_start(pdev, 1);
-	hw->membase = ioremap((ulong) hw->hw_membase, 256);
-#else
 	if (!request_region(hw->iobase, 8, hw->card_name)) {
 		printk(KERN_INFO
 		       "HFC-4S/8S: failed to request address space at 0x%04x\n",
 		       hw->iobase);
 		goto out;
 	}
-#endif
 
 	pci_set_drvdata(pdev, hw);
 	err = setup_instance(hw);
diff --git a/drivers/isdn/i4l/isdn_ppp.c b/drivers/isdn/i4l/isdn_ppp.c
index a5da511e3c9a..61ac63237446 100644
--- a/drivers/isdn/i4l/isdn_ppp.c
+++ b/drivers/isdn/i4l/isdn_ppp.c
@@ -634,7 +634,7 @@ isdn_ppp_ioctl(int min, struct file *file, unsigned int cmd, unsigned long arg)
 #ifdef CONFIG_IPPP_FILTER
 	case PPPIOCSPASS:
 	{
-		struct sock_fprog fprog;
+		struct sock_fprog_kern fprog;
 		struct sock_filter *code;
 		int err, len = get_filter(argp, &code);
 
@@ -653,7 +653,7 @@ isdn_ppp_ioctl(int min, struct file *file, unsigned int cmd, unsigned long arg)
 	}
 	case PPPIOCSACTIVE:
 	{
-		struct sock_fprog fprog;
+		struct sock_fprog_kern fprog;
 		struct sock_filter *code;
 		int err, len = get_filter(argp, &code);
 
diff --git a/drivers/isdn/mISDN/l1oip_core.c b/drivers/isdn/mISDN/l1oip_core.c
index 2c0d2c2bf946..9f454d76cc06 100644
--- a/drivers/isdn/mISDN/l1oip_core.c
+++ b/drivers/isdn/mISDN/l1oip_core.c
@@ -287,11 +287,9 @@ l1oip_socket_send(struct l1oip *hc, u8 localcodec, u8 channel, u32 chanmask,
 	p = frame;
 
 	/* restart timer */
-	if ((int)(hc->keep_tl.expires-jiffies) < 5 * HZ) {
-		del_timer(&hc->keep_tl);
-		hc->keep_tl.expires = jiffies + L1OIP_KEEPALIVE * HZ;
-		add_timer(&hc->keep_tl);
-	} else
+	if (time_before(hc->keep_tl.expires, jiffies + 5 * HZ))
+		mod_timer(&hc->keep_tl, jiffies + L1OIP_KEEPALIVE * HZ);
+	else
 		hc->keep_tl.expires = jiffies + L1OIP_KEEPALIVE * HZ;
 
 	if (debug & DEBUG_L1OIP_MSG)
@@ -621,11 +619,9 @@ multiframe:
 		goto multiframe;
 
 	/* restart timer */
-	if ((int)(hc->timeout_tl.expires-jiffies) < 5 * HZ || !hc->timeout_on) {
+	if (time_before(hc->timeout_tl.expires, jiffies + 5 * HZ) || !hc->timeout_on) {
 		hc->timeout_on = 1;
-		del_timer(&hc->timeout_tl);
-		hc->timeout_tl.expires = jiffies + L1OIP_TIMEOUT * HZ;
-		add_timer(&hc->timeout_tl);
+		mod_timer(&hc->timeout_tl, jiffies + L1OIP_TIMEOUT * HZ);
 	} else /* only adjust timer */
 		hc->timeout_tl.expires = jiffies + L1OIP_TIMEOUT * HZ;
 
diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 089841ca180f..a1b044e7eaad 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -300,16 +300,6 @@ config LEDS_PCA963X
 	  LED driver chip accessed via the I2C bus. Supported
 	  devices include PCA9633 and PCA9634
 
-config LEDS_PCA9685
-	tristate "LED support for PCA9685 I2C chip"
-	depends on LEDS_CLASS
-	depends on I2C
-	help
-	  This option enables support for LEDs connected to the PCA9685
-	  LED driver chip accessed via the I2C bus.
-	  The PCA9685 offers 12-bit PWM (4095 levels of brightness) on
-	  16 individual channels.
-
 config LEDS_WM831X_STATUS
 	tristate "LED support for status LEDs on WM831x PMICs"
 	depends on LEDS_CLASS
diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile
index 8b4c956e11ba..79c5155199a7 100644
--- a/drivers/leds/Makefile
+++ b/drivers/leds/Makefile
@@ -36,7 +36,6 @@ obj-$(CONFIG_LEDS_OT200)		+= leds-ot200.o
 obj-$(CONFIG_LEDS_FSG)			+= leds-fsg.o
 obj-$(CONFIG_LEDS_PCA955X)		+= leds-pca955x.o
 obj-$(CONFIG_LEDS_PCA963X)		+= leds-pca963x.o
-obj-$(CONFIG_LEDS_PCA9685)		+= leds-pca9685.o
 obj-$(CONFIG_LEDS_DA903X)		+= leds-da903x.o
 obj-$(CONFIG_LEDS_DA9052)		+= leds-da9052.o
 obj-$(CONFIG_LEDS_WM831X_STATUS)	+= leds-wm831x-status.o
diff --git a/drivers/leds/dell-led.c b/drivers/leds/dell-led.c
index e5c57389efd6..c36acaf566a6 100644
--- a/drivers/leds/dell-led.c
+++ b/drivers/leds/dell-led.c
@@ -15,12 +15,15 @@
 #include <linux/leds.h>
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/dmi.h>
+#include <linux/dell-led.h>
 
 MODULE_AUTHOR("Louis Davis/Jim Dailey");
 MODULE_DESCRIPTION("Dell LED Control Driver");
 MODULE_LICENSE("GPL");
 
 #define DELL_LED_BIOS_GUID "F6E4FE6E-909D-47cb-8BAB-C9F6F2F8D396"
+#define DELL_APP_GUID "A80593CE-A997-11DA-B012-B622A1EF5492"
 MODULE_ALIAS("wmi:" DELL_LED_BIOS_GUID);
 
 /* Error Result Codes: */
@@ -39,6 +42,149 @@ MODULE_ALIAS("wmi:" DELL_LED_BIOS_GUID);
 #define CMD_LED_OFF	17
 #define CMD_LED_BLINK	18
 
+struct app_wmi_args {
+	u16 class;
+	u16 selector;
+	u32 arg1;
+	u32 arg2;
+	u32 arg3;
+	u32 arg4;
+	u32 res1;
+	u32 res2;
+	u32 res3;
+	u32 res4;
+	char dummy[92];
+};
+
+#define GLOBAL_MIC_MUTE_ENABLE	0x364
+#define GLOBAL_MIC_MUTE_DISABLE	0x365
+
+struct dell_bios_data_token {
+	u16 tokenid;
+	u16 location;
+	u16 value;
+};
+
+struct __attribute__ ((__packed__)) dell_bios_calling_interface {
+	struct	dmi_header header;
+	u16	cmd_io_addr;
+	u8	cmd_io_code;
+	u32	supported_cmds;
+	struct	dell_bios_data_token damap[];
+};
+
+static struct dell_bios_data_token dell_mic_tokens[2];
+
+static int dell_wmi_perform_query(struct app_wmi_args *args)
+{
+	struct app_wmi_args *bios_return;
+	union acpi_object *obj;
+	struct acpi_buffer input;
+	struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
+	acpi_status status;
+	u32 rc = -EINVAL;
+
+	input.length = 128;
+	input.pointer = args;
+
+	status = wmi_evaluate_method(DELL_APP_GUID, 0, 1, &input, &output);
+	if (!ACPI_SUCCESS(status))
+		goto err_out0;
+
+	obj = output.pointer;
+	if (!obj)
+		goto err_out0;
+
+	if (obj->type != ACPI_TYPE_BUFFER)
+		goto err_out1;
+
+	bios_return = (struct app_wmi_args *)obj->buffer.pointer;
+	rc = bios_return->res1;
+	if (rc)
+		goto err_out1;
+
+	memcpy(args, bios_return, sizeof(struct app_wmi_args));
+	rc = 0;
+
+ err_out1:
+	kfree(obj);
+ err_out0:
+	return rc;
+}
+
+static void __init find_micmute_tokens(const struct dmi_header *dm, void *dummy)
+{
+	struct dell_bios_calling_interface *calling_interface;
+	struct dell_bios_data_token *token;
+	int token_size = sizeof(struct dell_bios_data_token);
+	int i = 0;
+
+	if (dm->type == 0xda && dm->length > 17) {
+		calling_interface = container_of(dm,
+				struct dell_bios_calling_interface, header);
+
+		token = &calling_interface->damap[i];
+		while (token->tokenid != 0xffff) {
+			if (token->tokenid == GLOBAL_MIC_MUTE_DISABLE)
+				memcpy(&dell_mic_tokens[0], token, token_size);
+			else if (token->tokenid == GLOBAL_MIC_MUTE_ENABLE)
+				memcpy(&dell_mic_tokens[1], token, token_size);
+
+			i++;
+			token = &calling_interface->damap[i];
+		}
+	}
+}
+
+static int dell_micmute_led_set(int state)
+{
+	struct app_wmi_args args;
+	struct dell_bios_data_token *token;
+
+	if (!wmi_has_guid(DELL_APP_GUID))
+		return -ENODEV;
+
+	if (state == 0 || state == 1)
+		token = &dell_mic_tokens[state];
+	else
+		return -EINVAL;
+
+	memset(&args, 0, sizeof(struct app_wmi_args));
+
+	args.class = 1;
+	args.arg1 = token->location;
+	args.arg2 = token->value;
+
+	dell_wmi_perform_query(&args);
+
+	return state;
+}
+
+int dell_app_wmi_led_set(int whichled, int on)
+{
+	int state = 0;
+
+	switch (whichled) {
+	case DELL_LED_MICMUTE:
+		state = dell_micmute_led_set(on);
+		break;
+	default:
+		pr_warn("led type %x is not supported\n", whichled);
+		break;
+	}
+
+	return state;
+}
+EXPORT_SYMBOL_GPL(dell_app_wmi_led_set);
+
+static int __init dell_micmute_led_init(void)
+{
+	memset(dell_mic_tokens, 0, sizeof(struct dell_bios_data_token) * 2);
+	dmi_walk(find_micmute_tokens, NULL);
+
+	return 0;
+}
+
 struct bios_args {
 	unsigned char length;
 	unsigned char result_code;
@@ -181,21 +327,32 @@ static int __init dell_led_init(void)
 {
 	int error = 0;
 
-	if (!wmi_has_guid(DELL_LED_BIOS_GUID))
+	if (!wmi_has_guid(DELL_LED_BIOS_GUID) && !wmi_has_guid(DELL_APP_GUID))
 		return -ENODEV;
 
-	error = led_off();
-	if (error != 0)
-		return -ENODEV;
+	if (wmi_has_guid(DELL_APP_GUID))
+		error = dell_micmute_led_init();
 
-	return led_classdev_register(NULL, &dell_led);
+	if (wmi_has_guid(DELL_LED_BIOS_GUID)) {
+		error = led_off();
+		if (error != 0)
+			return -ENODEV;
+
+		error = led_classdev_register(NULL, &dell_led);
+	}
+
+	return error;
 }
 
 static void __exit dell_led_exit(void)
 {
-	led_classdev_unregister(&dell_led);
+	int error = 0;
 
-	led_off();
+	if (wmi_has_guid(DELL_LED_BIOS_GUID)) {
+		error = led_off();
+		if (error == 0)
+			led_classdev_unregister(&dell_led);
+	}
 }
 
 module_init(dell_led_init);
diff --git a/drivers/leds/leds-88pm860x.c b/drivers/leds/leds-88pm860x.c
index d1e1bca90d11..c2def5551ce1 100644
--- a/drivers/leds/leds-88pm860x.c
+++ b/drivers/leds/leds-88pm860x.c
@@ -130,10 +130,9 @@ static int pm860x_led_dt_init(struct platform_device *pdev,
 	struct device_node *nproot, *np;
 	int iset = 0;
 
-	nproot = of_node_get(pdev->dev.parent->of_node);
-	if (!nproot)
+	if (!pdev->dev.parent->of_node)
 		return -ENODEV;
-	nproot = of_find_node_by_name(nproot, "leds");
+	nproot = of_get_child_by_name(pdev->dev.parent->of_node, "leds");
 	if (!nproot) {
 		dev_err(&pdev->dev, "failed to find leds node\n");
 		return -ENODEV;
diff --git a/drivers/leds/leds-adp5520.c b/drivers/leds/leds-adp5520.c
index 86b5bdb0c773..5036d7b4f82e 100644
--- a/drivers/leds/leds-adp5520.c
+++ b/drivers/leds/leds-adp5520.c
@@ -120,13 +120,10 @@ static int adp5520_led_probe(struct platform_device *pdev)
 
 	led = devm_kzalloc(&pdev->dev, sizeof(*led) * pdata->num_leds,
 				GFP_KERNEL);
-	if (led == NULL) {
-		dev_err(&pdev->dev, "failed to alloc memory\n");
+	if (!led)
 		return -ENOMEM;
-	}
 
 	ret = adp5520_led_prepare(pdev);
-
 	if (ret) {
 		dev_err(&pdev->dev, "failed to write\n");
 		return ret;
diff --git a/drivers/leds/leds-bd2802.c b/drivers/leds/leds-bd2802.c
index fb5a3472d614..6078c15d3452 100644
--- a/drivers/leds/leds-bd2802.c
+++ b/drivers/leds/leds-bd2802.c
@@ -678,10 +678,8 @@ static int bd2802_probe(struct i2c_client *client,
 	int ret, i;
 
 	led = devm_kzalloc(&client->dev, sizeof(struct bd2802_led), GFP_KERNEL);
-	if (!led) {
-		dev_err(&client->dev, "failed to allocate driver data\n");
+	if (!led)
 		return -ENOMEM;
-	}
 
 	led->client = client;
 	pdata = led->pdata = dev_get_platdata(&client->dev);
diff --git a/drivers/leds/leds-da903x.c b/drivers/leds/leds-da903x.c
index 35dffb100388..54b8b5216b8b 100644
--- a/drivers/leds/leds-da903x.c
+++ b/drivers/leds/leds-da903x.c
@@ -108,10 +108,8 @@ static int da903x_led_probe(struct platform_device *pdev)
 	}
 
 	led = devm_kzalloc(&pdev->dev, sizeof(struct da903x_led), GFP_KERNEL);
-	if (led == NULL) {
-		dev_err(&pdev->dev, "failed to alloc memory for LED%d\n", id);
+	if (!led)
 		return -ENOMEM;
-	}
 
 	led->cdev.name = pdata->name;
 	led->cdev.default_trigger = pdata->default_trigger;
diff --git a/drivers/leds/leds-da9052.c b/drivers/leds/leds-da9052.c
index 01486adc7f8b..e4da1f460ac5 100644
--- a/drivers/leds/leds-da9052.c
+++ b/drivers/leds/leds-da9052.c
@@ -126,8 +126,7 @@ static int da9052_led_probe(struct platform_device *pdev)
 	led = devm_kzalloc(&pdev->dev,
 			   sizeof(struct da9052_led) * pled->num_leds,
 			   GFP_KERNEL);
-	if (led == NULL) {
-		dev_err(&pdev->dev, "Failed to alloc memory\n");
+	if (!led) {
 		error = -ENOMEM;
 		goto err;
 	}
diff --git a/drivers/leds/leds-lp5523.c b/drivers/leds/leds-lp5523.c
index cb5ed82994ba..9e1716f8098c 100644
--- a/drivers/leds/leds-lp5523.c
+++ b/drivers/leds/leds-lp5523.c
@@ -1,5 +1,5 @@
 /*
- * lp5523.c - LP5523 LED Driver
+ * lp5523.c - LP5523, LP55231 LED Driver
  *
  * Copyright (C) 2010 Nokia Corporation
  * Copyright (C) 2012 Texas Instruments
@@ -814,6 +814,7 @@ MODULE_DEVICE_TABLE(i2c, lp5523_id);
 #ifdef CONFIG_OF
 static const struct of_device_id of_lp5523_leds_match[] = {
 	{ .compatible = "national,lp5523", },
+	{ .compatible = "ti,lp55231", },
 	{},
 };
 
diff --git a/drivers/leds/leds-pca9685.c b/drivers/leds/leds-pca9685.c
deleted file mode 100644
index 6e1ef3a9d6ef..000000000000
--- a/drivers/leds/leds-pca9685.c
+++ /dev/null
@@ -1,213 +0,0 @@
-/*
- * Copyright 2013 Maximilian Güntner <maximilian.guentner@gmail.com>
- *
- * This file is subject to the terms and conditions of version 2 of
- * the GNU General Public License.  See the file COPYING in the main
- * directory of this archive for more details.
- *
- * Based on leds-pca963x.c driver by
- * Peter Meerwald <p.meerwald@bct-electronic.com>
- *
- * Driver for the NXP PCA9685 12-Bit PWM LED driver chip.
- *
- */
-
-#include <linux/ctype.h>
-#include <linux/delay.h>
-#include <linux/err.h>
-#include <linux/i2c.h>
-#include <linux/leds.h>
-#include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/workqueue.h>
-
-#include <linux/platform_data/leds-pca9685.h>
-
-/* Register Addresses */
-#define PCA9685_MODE1 0x00
-#define PCA9685_MODE2 0x01
-#define PCA9685_LED0_ON_L 0x06
-#define PCA9685_ALL_LED_ON_L 0xFA
-
-/* MODE1 Register */
-#define PCA9685_ALLCALL 0x00
-#define PCA9685_SLEEP   0x04
-#define PCA9685_AI      0x05
-
-/* MODE2 Register */
-#define PCA9685_INVRT   0x04
-#define PCA9685_OUTDRV  0x02
-
-static const struct i2c_device_id pca9685_id[] = {
-	{ "pca9685", 0 },
-	{ }
-};
-MODULE_DEVICE_TABLE(i2c, pca9685_id);
-
-struct pca9685_led {
-	struct i2c_client *client;
-	struct work_struct work;
-	u16 brightness;
-	struct led_classdev led_cdev;
-	int led_num; /* 0-15 */
-	char name[32];
-};
-
-static void pca9685_write_msg(struct i2c_client *client, u8 *buf, u8 len)
-{
-	struct i2c_msg msg = {
-		.addr = client->addr,
-		.flags = 0x00,
-		.len = len,
-		.buf = buf
-	};
-	i2c_transfer(client->adapter, &msg, 1);
-}
-
-static void pca9685_all_off(struct i2c_client *client)
-{
-	u8 i2c_buffer[5] = {PCA9685_ALL_LED_ON_L, 0x00, 0x00, 0x00, 0x10};
-	pca9685_write_msg(client, i2c_buffer, 5);
-}
-
-static void pca9685_led_work(struct work_struct *work)
-{
-	struct pca9685_led *pca9685;
-	u8 i2c_buffer[5];
-
-	pca9685 = container_of(work, struct pca9685_led, work);
-	i2c_buffer[0] = PCA9685_LED0_ON_L + 4 * pca9685->led_num;
-	/*
-	 * 4095 is the maximum brightness, so we set the ON time to 0x1000
-	 * which disables the PWM generator for that LED
-	 */
-	if (pca9685->brightness == 4095)
-		*((__le16 *)(i2c_buffer+1)) = cpu_to_le16(0x1000);
-	else
-		*((__le16 *)(i2c_buffer+1)) = 0x0000;
-
-	if (pca9685->brightness == 0)
-		*((__le16 *)(i2c_buffer+3)) = cpu_to_le16(0x1000);
-	else if (pca9685->brightness == 4095)
-		*((__le16 *)(i2c_buffer+3)) = 0x0000;
-	else
-		*((__le16 *)(i2c_buffer+3)) = cpu_to_le16(pca9685->brightness);
-
-	pca9685_write_msg(pca9685->client, i2c_buffer, 5);
-}
-
-static void pca9685_led_set(struct led_classdev *led_cdev,
-		enum led_brightness value)
-{
-	struct pca9685_led *pca9685;
-	pca9685 = container_of(led_cdev, struct pca9685_led, led_cdev);
-	pca9685->brightness = value;
-
-	schedule_work(&pca9685->work);
-}
-
-static int pca9685_probe(struct i2c_client *client,
-		const struct i2c_device_id *id)
-{
-	struct pca9685_led *pca9685;
-	struct pca9685_platform_data *pdata;
-	int err;
-	u8 i;
-
-	pdata = dev_get_platdata(&client->dev);
-	if (pdata) {
-		if (pdata->leds.num_leds < 1 || pdata->leds.num_leds > 15) {
-			dev_err(&client->dev, "board info must claim 1-16 LEDs");
-			return -EINVAL;
-		}
-	}
-
-	pca9685 = devm_kzalloc(&client->dev, 16 * sizeof(*pca9685), GFP_KERNEL);
-	if (!pca9685)
-		return -ENOMEM;
-
-	i2c_set_clientdata(client, pca9685);
-	pca9685_all_off(client);
-
-	for (i = 0; i < 16; i++) {
-		pca9685[i].client = client;
-		pca9685[i].led_num = i;
-		pca9685[i].name[0] = '\0';
-		if (pdata && i < pdata->leds.num_leds) {
-			if (pdata->leds.leds[i].name)
-				strncpy(pca9685[i].name,
-					pdata->leds.leds[i].name,
-					sizeof(pca9685[i].name)-1);
-			if (pdata->leds.leds[i].default_trigger)
-				pca9685[i].led_cdev.default_trigger =
-					pdata->leds.leds[i].default_trigger;
-		}
-		if (strlen(pca9685[i].name) == 0) {
-			/*
-			 * Write adapter and address to the name as well.
-			 * Otherwise multiple chips attached to one host would
-			 * not work.
-			 */
-			snprintf(pca9685[i].name, sizeof(pca9685[i].name),
-					"pca9685:%d:x%.2x:%d",
-					client->adapter->nr, client->addr, i);
-		}
-		pca9685[i].led_cdev.name = pca9685[i].name;
-		pca9685[i].led_cdev.max_brightness = 0xfff;
-		pca9685[i].led_cdev.brightness_set = pca9685_led_set;
-
-		INIT_WORK(&pca9685[i].work, pca9685_led_work);
-		err = led_classdev_register(&client->dev, &pca9685[i].led_cdev);
-		if (err < 0)
-			goto exit;
-	}
-
-	if (pdata)
-		i2c_smbus_write_byte_data(client, PCA9685_MODE2,
-			pdata->outdrv << PCA9685_OUTDRV |
-			pdata->inverted << PCA9685_INVRT);
-	else
-		i2c_smbus_write_byte_data(client, PCA9685_MODE2,
-			PCA9685_TOTEM_POLE << PCA9685_OUTDRV);
-	/* Enable Auto-Increment, enable oscillator, ALLCALL/SUBADDR disabled */
-	i2c_smbus_write_byte_data(client, PCA9685_MODE1, BIT(PCA9685_AI));
-
-	return 0;
-
-exit:
-	while (i--) {
-		led_classdev_unregister(&pca9685[i].led_cdev);
-		cancel_work_sync(&pca9685[i].work);
-	}
-	return err;
-}
-
-static int pca9685_remove(struct i2c_client *client)
-{
-	struct pca9685_led *pca9685 = i2c_get_clientdata(client);
-	u8 i;
-
-	for (i = 0; i < 16; i++) {
-		led_classdev_unregister(&pca9685[i].led_cdev);
-		cancel_work_sync(&pca9685[i].work);
-	}
-	pca9685_all_off(client);
-	return 0;
-}
-
-static struct i2c_driver pca9685_driver = {
-	.driver = {
-		.name = "leds-pca9685",
-		.owner = THIS_MODULE,
-	},
-	.probe = pca9685_probe,
-	.remove = pca9685_remove,
-	.id_table = pca9685_id,
-};
-
-module_i2c_driver(pca9685_driver);
-
-MODULE_AUTHOR("Maximilian Güntner <maximilian.guentner@gmail.com>");
-MODULE_DESCRIPTION("PCA9685 LED Driver");
-MODULE_LICENSE("GPL v2");
diff --git a/drivers/leds/leds-pwm.c b/drivers/leds/leds-pwm.c
index 7d0aaed1e23a..d672bb4480f6 100644
--- a/drivers/leds/leds-pwm.c
+++ b/drivers/leds/leds-pwm.c
@@ -69,6 +69,10 @@ static void led_pwm_set(struct led_classdev *led_cdev,
 
 	duty *= brightness;
 	do_div(duty, max);
+
+	if (led_dat->active_low)
+		duty = led_dat->period - duty;
+
 	led_dat->duty = duty;
 
 	if (led_dat->can_sleep)
@@ -92,55 +96,78 @@ static void led_pwm_cleanup(struct led_pwm_priv *priv)
 	}
 }
 
-static int led_pwm_create_of(struct platform_device *pdev,
-			     struct led_pwm_priv *priv)
+static int led_pwm_add(struct device *dev, struct led_pwm_priv *priv,
+		       struct led_pwm *led, struct device_node *child)
 {
-	struct device_node *child;
+	struct led_pwm_data *led_data = &priv->leds[priv->num_leds];
 	int ret;
 
-	for_each_child_of_node(pdev->dev.of_node, child) {
-		struct led_pwm_data *led_dat = &priv->leds[priv->num_leds];
+	led_data->active_low = led->active_low;
+	led_data->cdev.name = led->name;
+	led_data->cdev.default_trigger = led->default_trigger;
+	led_data->cdev.brightness_set = led_pwm_set;
+	led_data->cdev.brightness = LED_OFF;
+	led_data->cdev.max_brightness = led->max_brightness;
+	led_data->cdev.flags = LED_CORE_SUSPENDRESUME;
 
-		led_dat->cdev.name = of_get_property(child, "label",
-						     NULL) ? : child->name;
+	if (child)
+		led_data->pwm = devm_of_pwm_get(dev, child, NULL);
+	else
+		led_data->pwm = devm_pwm_get(dev, led->name);
+	if (IS_ERR(led_data->pwm)) {
+		ret = PTR_ERR(led_data->pwm);
+		dev_err(dev, "unable to request PWM for %s: %d\n",
+			led->name, ret);
+		return ret;
+	}
 
-		led_dat->pwm = devm_of_pwm_get(&pdev->dev, child, NULL);
-		if (IS_ERR(led_dat->pwm)) {
-			dev_err(&pdev->dev, "unable to request PWM for %s\n",
-				led_dat->cdev.name);
-			ret = PTR_ERR(led_dat->pwm);
-			goto err;
-		}
-		/* Get the period from PWM core when n*/
-		led_dat->period = pwm_get_period(led_dat->pwm);
+	if (child)
+		led_data->period = pwm_get_period(led_data->pwm);
 
-		led_dat->cdev.default_trigger = of_get_property(child,
-						"linux,default-trigger", NULL);
-		of_property_read_u32(child, "max-brightness",
-				     &led_dat->cdev.max_brightness);
+	led_data->can_sleep = pwm_can_sleep(led_data->pwm);
+	if (led_data->can_sleep)
+		INIT_WORK(&led_data->work, led_pwm_work);
 
-		led_dat->cdev.brightness_set = led_pwm_set;
-		led_dat->cdev.brightness = LED_OFF;
-		led_dat->cdev.flags |= LED_CORE_SUSPENDRESUME;
+	led_data->period = pwm_get_period(led_data->pwm);
+	if (!led_data->period && (led->pwm_period_ns > 0))
+		led_data->period = led->pwm_period_ns;
 
-		led_dat->can_sleep = pwm_can_sleep(led_dat->pwm);
-		if (led_dat->can_sleep)
-			INIT_WORK(&led_dat->work, led_pwm_work);
+	ret = led_classdev_register(dev, &led_data->cdev);
+	if (ret == 0) {
+		priv->num_leds++;
+	} else {
+		dev_err(dev, "failed to register PWM led for %s: %d\n",
+			led->name, ret);
+	}
+
+	return ret;
+}
 
-		ret = led_classdev_register(&pdev->dev, &led_dat->cdev);
-		if (ret < 0) {
-			dev_err(&pdev->dev, "failed to register for %s\n",
-				led_dat->cdev.name);
+static int led_pwm_create_of(struct device *dev, struct led_pwm_priv *priv)
+{
+	struct device_node *child;
+	struct led_pwm led;
+	int ret = 0;
+
+	memset(&led, 0, sizeof(led));
+
+	for_each_child_of_node(dev->of_node, child) {
+		led.name = of_get_property(child, "label", NULL) ? :
+			   child->name;
+
+		led.default_trigger = of_get_property(child,
+						"linux,default-trigger", NULL);
+		led.active_low = of_property_read_bool(child, "active-low");
+		of_property_read_u32(child, "max-brightness",
+				     &led.max_brightness);
+
+		ret = led_pwm_add(dev, priv, &led, child);
+		if (ret) {
 			of_node_put(child);
-			goto err;
+			break;
 		}
-		priv->num_leds++;
 	}
 
-	return 0;
-err:
-	led_pwm_cleanup(priv);
-
 	return ret;
 }
 
@@ -166,51 +193,23 @@ static int led_pwm_probe(struct platform_device *pdev)
 
 	if (pdata) {
 		for (i = 0; i < count; i++) {
-			struct led_pwm *cur_led = &pdata->leds[i];
-			struct led_pwm_data *led_dat = &priv->leds[i];
-
-			led_dat->pwm = devm_pwm_get(&pdev->dev, cur_led->name);
-			if (IS_ERR(led_dat->pwm)) {
-				ret = PTR_ERR(led_dat->pwm);
-				dev_err(&pdev->dev,
-					"unable to request PWM for %s\n",
-					cur_led->name);
-				goto err;
-			}
-
-			led_dat->cdev.name = cur_led->name;
-			led_dat->cdev.default_trigger = cur_led->default_trigger;
-			led_dat->active_low = cur_led->active_low;
-			led_dat->period = cur_led->pwm_period_ns;
-			led_dat->cdev.brightness_set = led_pwm_set;
-			led_dat->cdev.brightness = LED_OFF;
-			led_dat->cdev.max_brightness = cur_led->max_brightness;
-			led_dat->cdev.flags |= LED_CORE_SUSPENDRESUME;
-
-			led_dat->can_sleep = pwm_can_sleep(led_dat->pwm);
-			if (led_dat->can_sleep)
-				INIT_WORK(&led_dat->work, led_pwm_work);
-
-			ret = led_classdev_register(&pdev->dev, &led_dat->cdev);
-			if (ret < 0)
-				goto err;
+			ret = led_pwm_add(&pdev->dev, priv, &pdata->leds[i],
+					  NULL);
+			if (ret)
+				break;
 		}
-		priv->num_leds = count;
 	} else {
-		ret = led_pwm_create_of(pdev, priv);
-		if (ret)
-			return ret;
+		ret = led_pwm_create_of(&pdev->dev, priv);
+	}
+
+	if (ret) {
+		led_pwm_cleanup(priv);
+		return ret;
 	}
 
 	platform_set_drvdata(pdev, priv);
 
 	return 0;
-
-err:
-	priv->num_leds = i;
-	led_pwm_cleanup(priv);
-
-	return ret;
 }
 
 static int led_pwm_remove(struct platform_device *pdev)
diff --git a/drivers/leds/leds-s3c24xx.c b/drivers/leds/leds-s3c24xx.c
index 28988b7b4fab..785eb53a87fc 100644
--- a/drivers/leds/leds-s3c24xx.c
+++ b/drivers/leds/leds-s3c24xx.c
@@ -76,10 +76,8 @@ static int s3c24xx_led_probe(struct platform_device *dev)
 
 	led = devm_kzalloc(&dev->dev, sizeof(struct s3c24xx_gpio_led),
 			   GFP_KERNEL);
-	if (led == NULL) {
-		dev_err(&dev->dev, "No memory for device\n");
+	if (!led)
 		return -ENOMEM;
-	}
 
 	platform_set_drvdata(dev, led);
 
diff --git a/drivers/leds/leds-sunfire.c b/drivers/leds/leds-sunfire.c
index 388632d23d44..0b8cc4a021a6 100644
--- a/drivers/leds/leds-sunfire.c
+++ b/drivers/leds/leds-sunfire.c
@@ -135,10 +135,8 @@ static int sunfire_led_generic_probe(struct platform_device *pdev,
 	}
 
 	p = devm_kzalloc(&pdev->dev, sizeof(*p), GFP_KERNEL);
-	if (!p) {
-		dev_err(&pdev->dev, "Could not allocate struct sunfire_drvdata\n");
+	if (!p)
 		return -ENOMEM;
-	}
 
 	for (i = 0; i < NUM_LEDS_PER_BOARD; i++) {
 		struct led_classdev *lp = &p->leds[i].led_cdev;
diff --git a/drivers/leds/trigger/ledtrig-cpu.c b/drivers/leds/trigger/ledtrig-cpu.c
index 1c3ee9fcaf34..aec0f02b6b3e 100644
--- a/drivers/leds/trigger/ledtrig-cpu.c
+++ b/drivers/leds/trigger/ledtrig-cpu.c
@@ -47,7 +47,7 @@ static DEFINE_PER_CPU(struct led_trigger_cpu, cpu_trig);
  */
 void ledtrig_cpu(enum cpu_led_event ledevt)
 {
-	struct led_trigger_cpu *trig = &__get_cpu_var(cpu_trig);
+	struct led_trigger_cpu *trig = this_cpu_ptr(&cpu_trig);
 
 	/* Locate the correct CPU LED */
 	switch (ledevt) {
diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c
index 23b4a3b28dbc..4eab93aa570b 100644
--- a/drivers/macintosh/smu.c
+++ b/drivers/macintosh/smu.c
@@ -1257,7 +1257,8 @@ static unsigned int smu_fpoll(struct file *file, poll_table *wait)
 		if (pp->busy && pp->cmd.status != 1)
 			mask |= POLLIN;
 		spin_unlock_irqrestore(&pp->lock, flags);
-	} if (pp->mode == smu_file_events) {
+	}
+	if (pp->mode == smu_file_events) {
 		/* Not yet implemented */
 	}
 	return mask;
diff --git a/drivers/macintosh/windfarm_pm121.c b/drivers/macintosh/windfarm_pm121.c
index 7fe58b0ae8b4..b350fb86ff08 100644
--- a/drivers/macintosh/windfarm_pm121.c
+++ b/drivers/macintosh/windfarm_pm121.c
@@ -555,8 +555,18 @@ static void pm121_create_sys_fans(int loop_id)
 	pid_param.interval	= PM121_SYS_INTERVAL;
 	pid_param.history_len	= PM121_SYS_HISTORY_SIZE;
 	pid_param.itarget	= param->itarget;
-	pid_param.min		= control->ops->get_min(control);
-	pid_param.max		= control->ops->get_max(control);
+	if(control)
+	{
+		pid_param.min		= control->ops->get_min(control);
+		pid_param.max		= control->ops->get_max(control);
+	} else {
+		/*
+		 * This is probably not the right!?
+		 * Perhaps goto fail  if control == NULL  above?
+		 */
+		pid_param.min		= 0;
+		pid_param.max		= 0;
+	}
 
 	wf_pid_init(&pm121_sys_state[loop_id]->pid, &pid_param);
 
@@ -571,7 +581,7 @@ static void pm121_create_sys_fans(int loop_id)
 	   control the same control */
 	printk(KERN_WARNING "pm121: failed to set up %s loop "
 	       "setting \"%s\" to max speed.\n",
-	       loop_names[loop_id], control->name);
+	       loop_names[loop_id], control ? control->name : "uninitialized value");
 
 	if (control)
 		wf_control_set_max(control);
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 9a8e66ae04f5..67f8b31e2054 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -669,17 +669,13 @@ static inline unsigned long file_page_offset(struct bitmap_storage *store,
 /*
  * return a pointer to the page in the filemap that contains the given bit
  *
- * this lookup is complicated by the fact that the bitmap sb might be exactly
- * 1 page (e.g., x86) or less than 1 page -- so the bitmap might start on page
- * 0 or page 1
  */
 static inline struct page *filemap_get_page(struct bitmap_storage *store,
 					    unsigned long chunk)
 {
 	if (file_page_index(store, chunk) >= store->file_pages)
 		return NULL;
-	return store->filemap[file_page_index(store, chunk)
-			      - file_page_index(store, 0)];
+	return store->filemap[file_page_index(store, chunk)];
 }
 
 static int bitmap_storage_alloc(struct bitmap_storage *store,
diff --git a/drivers/md/dm-bio-prison.c b/drivers/md/dm-bio-prison.c
index 85f0b7074257..f752d12081ff 100644
--- a/drivers/md/dm-bio-prison.c
+++ b/drivers/md/dm-bio-prison.c
@@ -14,13 +14,17 @@
 
 /*----------------------------------------------------------------*/
 
-struct dm_bio_prison {
+struct bucket {
 	spinlock_t lock;
+	struct hlist_head cells;
+};
+
+struct dm_bio_prison {
 	mempool_t *cell_pool;
 
 	unsigned nr_buckets;
 	unsigned hash_mask;
-	struct hlist_head *cells;
+	struct bucket *buckets;
 };
 
 /*----------------------------------------------------------------*/
@@ -40,6 +44,12 @@ static uint32_t calc_nr_buckets(unsigned nr_cells)
 
 static struct kmem_cache *_cell_cache;
 
+static void init_bucket(struct bucket *b)
+{
+	spin_lock_init(&b->lock);
+	INIT_HLIST_HEAD(&b->cells);
+}
+
 /*
  * @nr_cells should be the number of cells you want in use _concurrently_.
  * Don't confuse it with the number of distinct keys.
@@ -49,13 +59,12 @@ struct dm_bio_prison *dm_bio_prison_create(unsigned nr_cells)
 	unsigned i;
 	uint32_t nr_buckets = calc_nr_buckets(nr_cells);
 	size_t len = sizeof(struct dm_bio_prison) +
-		(sizeof(struct hlist_head) * nr_buckets);
+		(sizeof(struct bucket) * nr_buckets);
 	struct dm_bio_prison *prison = kmalloc(len, GFP_KERNEL);
 
 	if (!prison)
 		return NULL;
 
-	spin_lock_init(&prison->lock);
 	prison->cell_pool = mempool_create_slab_pool(nr_cells, _cell_cache);
 	if (!prison->cell_pool) {
 		kfree(prison);
@@ -64,9 +73,9 @@ struct dm_bio_prison *dm_bio_prison_create(unsigned nr_cells)
 
 	prison->nr_buckets = nr_buckets;
 	prison->hash_mask = nr_buckets - 1;
-	prison->cells = (struct hlist_head *) (prison + 1);
+	prison->buckets = (struct bucket *) (prison + 1);
 	for (i = 0; i < nr_buckets; i++)
-		INIT_HLIST_HEAD(prison->cells + i);
+		init_bucket(prison->buckets + i);
 
 	return prison;
 }
@@ -107,40 +116,44 @@ static int keys_equal(struct dm_cell_key *lhs, struct dm_cell_key *rhs)
 		       (lhs->block == rhs->block);
 }
 
-static struct dm_bio_prison_cell *__search_bucket(struct hlist_head *bucket,
+static struct bucket *get_bucket(struct dm_bio_prison *prison,
+				 struct dm_cell_key *key)
+{
+	return prison->buckets + hash_key(prison, key);
+}
+
+static struct dm_bio_prison_cell *__search_bucket(struct bucket *b,
 						  struct dm_cell_key *key)
 {
 	struct dm_bio_prison_cell *cell;
 
-	hlist_for_each_entry(cell, bucket, list)
+	hlist_for_each_entry(cell, &b->cells, list)
 		if (keys_equal(&cell->key, key))
 			return cell;
 
 	return NULL;
 }
 
-static void __setup_new_cell(struct dm_bio_prison *prison,
+static void __setup_new_cell(struct bucket *b,
 			     struct dm_cell_key *key,
 			     struct bio *holder,
-			     uint32_t hash,
 			     struct dm_bio_prison_cell *cell)
 {
 	memcpy(&cell->key, key, sizeof(cell->key));
 	cell->holder = holder;
 	bio_list_init(&cell->bios);
-	hlist_add_head(&cell->list, prison->cells + hash);
+	hlist_add_head(&cell->list, &b->cells);
 }
 
-static int __bio_detain(struct dm_bio_prison *prison,
+static int __bio_detain(struct bucket *b,
 			struct dm_cell_key *key,
 			struct bio *inmate,
 			struct dm_bio_prison_cell *cell_prealloc,
 			struct dm_bio_prison_cell **cell_result)
 {
-	uint32_t hash = hash_key(prison, key);
 	struct dm_bio_prison_cell *cell;
 
-	cell = __search_bucket(prison->cells + hash, key);
+	cell = __search_bucket(b, key);
 	if (cell) {
 		if (inmate)
 			bio_list_add(&cell->bios, inmate);
@@ -148,7 +161,7 @@ static int __bio_detain(struct dm_bio_prison *prison,
 		return 1;
 	}
 
-	__setup_new_cell(prison, key, inmate, hash, cell_prealloc);
+	__setup_new_cell(b, key, inmate, cell_prealloc);
 	*cell_result = cell_prealloc;
 	return 0;
 }
@@ -161,10 +174,11 @@ static int bio_detain(struct dm_bio_prison *prison,
 {
 	int r;
 	unsigned long flags;
+	struct bucket *b = get_bucket(prison, key);
 
-	spin_lock_irqsave(&prison->lock, flags);
-	r = __bio_detain(prison, key, inmate, cell_prealloc, cell_result);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_lock_irqsave(&b->lock, flags);
+	r = __bio_detain(b, key, inmate, cell_prealloc, cell_result);
+	spin_unlock_irqrestore(&b->lock, flags);
 
 	return r;
 }
@@ -208,10 +222,11 @@ void dm_cell_release(struct dm_bio_prison *prison,
 		     struct bio_list *bios)
 {
 	unsigned long flags;
+	struct bucket *b = get_bucket(prison, &cell->key);
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irqsave(&b->lock, flags);
 	__cell_release(cell, bios);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irqrestore(&b->lock, flags);
 }
 EXPORT_SYMBOL_GPL(dm_cell_release);
 
@@ -230,28 +245,25 @@ void dm_cell_release_no_holder(struct dm_bio_prison *prison,
 			       struct bio_list *inmates)
 {
 	unsigned long flags;
+	struct bucket *b = get_bucket(prison, &cell->key);
 
-	spin_lock_irqsave(&prison->lock, flags);
+	spin_lock_irqsave(&b->lock, flags);
 	__cell_release_no_holder(cell, inmates);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	spin_unlock_irqrestore(&b->lock, flags);
 }
 EXPORT_SYMBOL_GPL(dm_cell_release_no_holder);
 
 void dm_cell_error(struct dm_bio_prison *prison,
-		   struct dm_bio_prison_cell *cell)
+		   struct dm_bio_prison_cell *cell, int error)
 {
 	struct bio_list bios;
 	struct bio *bio;
-	unsigned long flags;
 
 	bio_list_init(&bios);
-
-	spin_lock_irqsave(&prison->lock, flags);
-	__cell_release(cell, &bios);
-	spin_unlock_irqrestore(&prison->lock, flags);
+	dm_cell_release(prison, cell, &bios);
 
 	while ((bio = bio_list_pop(&bios)))
-		bio_io_error(bio);
+		bio_endio(bio, error);
 }
 EXPORT_SYMBOL_GPL(dm_cell_error);
 
diff --git a/drivers/md/dm-bio-prison.h b/drivers/md/dm-bio-prison.h
index 3f833190eadf..6805a142b750 100644
--- a/drivers/md/dm-bio-prison.h
+++ b/drivers/md/dm-bio-prison.h
@@ -85,7 +85,7 @@ void dm_cell_release_no_holder(struct dm_bio_prison *prison,
 			       struct dm_bio_prison_cell *cell,
 			       struct bio_list *inmates);
 void dm_cell_error(struct dm_bio_prison *prison,
-		   struct dm_bio_prison_cell *cell);
+		   struct dm_bio_prison_cell *cell, int error);
 
 /*----------------------------------------------------------------*/
 
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 53b213226c01..4cba2d808afb 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2003 Christophe Saout <christophe@saout.de>
+ * Copyright (C) 2003 Jana Saout <jana@saout.de>
  * Copyright (C) 2004 Clemens Fruhwirth <clemens@endorphin.org>
  * Copyright (C) 2006-2009 Red Hat, Inc. All rights reserved.
  * Copyright (C) 2013 Milan Broz <gmazyland@gmail.com>
@@ -1996,6 +1996,6 @@ static void __exit dm_crypt_exit(void)
 module_init(dm_crypt_init);
 module_exit(dm_crypt_exit);
 
-MODULE_AUTHOR("Christophe Saout <christophe@saout.de>");
+MODULE_AUTHOR("Jana Saout <jana@saout.de>");
 MODULE_DESCRIPTION(DM_NAME " target for transparent encryption / decryption");
 MODULE_LICENSE("GPL");
diff --git a/drivers/md/dm-era-target.c b/drivers/md/dm-era-target.c
index 414dad4cb49b..ad913cd4aded 100644
--- a/drivers/md/dm-era-target.c
+++ b/drivers/md/dm-era-target.c
@@ -1391,7 +1391,8 @@ static int era_is_congested(struct dm_target_callbacks *cb, int bdi_bits)
 
 static void era_destroy(struct era *era)
 {
-	metadata_close(era->md);
+	if (era->md)
+		metadata_close(era->md);
 
 	if (era->wq)
 		destroy_workqueue(era->wq);
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 3842ac738f98..db404a0f7e2c 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -10,6 +10,7 @@
 #include <linux/device-mapper.h>
 
 #include <linux/bio.h>
+#include <linux/completion.h>
 #include <linux/mempool.h>
 #include <linux/module.h>
 #include <linux/sched.h>
@@ -32,7 +33,7 @@ struct dm_io_client {
 struct io {
 	unsigned long error_bits;
 	atomic_t count;
-	struct task_struct *sleeper;
+	struct completion *wait;
 	struct dm_io_client *client;
 	io_notify_fn callback;
 	void *context;
@@ -121,8 +122,8 @@ static void dec_count(struct io *io, unsigned int region, int error)
 			invalidate_kernel_vmap_range(io->vma_invalidate_address,
 						     io->vma_invalidate_size);
 
-		if (io->sleeper)
-			wake_up_process(io->sleeper);
+		if (io->wait)
+			complete(io->wait);
 
 		else {
 			unsigned long r = io->error_bits;
@@ -387,6 +388,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 	 */
 	volatile char io_[sizeof(struct io) + __alignof__(struct io) - 1];
 	struct io *io = (struct io *)PTR_ALIGN(&io_, __alignof__(struct io));
+	DECLARE_COMPLETION_ONSTACK(wait);
 
 	if (num_regions > 1 && (rw & RW_MASK) != WRITE) {
 		WARN_ON(1);
@@ -395,7 +397,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 
 	io->error_bits = 0;
 	atomic_set(&io->count, 1); /* see dispatch_io() */
-	io->sleeper = current;
+	io->wait = &wait;
 	io->client = client;
 
 	io->vma_invalidate_address = dp->vma_invalidate_address;
@@ -403,15 +405,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 
 	dispatch_io(rw, num_regions, where, dp, io, 1);
 
-	while (1) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
-
-		if (!atomic_read(&io->count))
-			break;
-
-		io_schedule();
-	}
-	set_current_state(TASK_RUNNING);
+	wait_for_completion_io(&wait);
 
 	if (error_bits)
 		*error_bits = io->error_bits;
@@ -434,7 +428,7 @@ static int async_io(struct dm_io_client *client, unsigned int num_regions,
 	io = mempool_alloc(client->pool, GFP_NOIO);
 	io->error_bits = 0;
 	atomic_set(&io->count, 1); /* see dispatch_io() */
-	io->sleeper = NULL;
+	io->wait = NULL;
 	io->client = client;
 	io->callback = fn;
 	io->context = context;
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index ebfa411d1a7d..f4167b013d99 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -1242,17 +1242,8 @@ static int do_end_io(struct multipath *m, struct request *clone,
 	if (!error && !clone->errors)
 		return 0;	/* I/O complete */
 
-	if (noretry_error(error)) {
-		if ((clone->cmd_flags & REQ_WRITE_SAME) &&
-		    !clone->q->limits.max_write_same_sectors) {
-			struct queue_limits *limits;
-
-			/* device doesn't really support WRITE SAME, disable it */
-			limits = dm_get_queue_limits(dm_table_get_md(m->ti->table));
-			limits->max_write_same_sectors = 0;
-		}
+	if (noretry_error(error))
 		return error;
-	}
 
 	if (mpio->pgpath)
 		fail_path(mpio->pgpath);
@@ -1620,8 +1611,9 @@ static int multipath_busy(struct dm_target *ti)
 
 	spin_lock_irqsave(&m->lock, flags);
 
-	/* pg_init in progress, requeue until done */
-	if (!pg_ready(m)) {
+	/* pg_init in progress or no paths available */
+	if (m->pg_init_in_progress ||
+	    (!m->nr_valid_paths && m->queue_if_no_path)) {
 		busy = 1;
 		goto out;
 	}
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 8e0caed0bf74..5bd2290cfb1e 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -2141,6 +2141,11 @@ static int origin_write_extent(struct dm_snapshot *merging_snap,
  * Origin: maps a linear range of a device, with hooks for snapshotting.
  */
 
+struct dm_origin {
+	struct dm_dev *dev;
+	unsigned split_boundary;
+};
+
 /*
  * Construct an origin mapping: <dev_path>
  * The context for an origin is merely a 'struct dm_dev *'
@@ -2149,41 +2154,65 @@ static int origin_write_extent(struct dm_snapshot *merging_snap,
 static int origin_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 {
 	int r;
-	struct dm_dev *dev;
+	struct dm_origin *o;
 
 	if (argc != 1) {
 		ti->error = "origin: incorrect number of arguments";
 		return -EINVAL;
 	}
 
-	r = dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &dev);
+	o = kmalloc(sizeof(struct dm_origin), GFP_KERNEL);
+	if (!o) {
+		ti->error = "Cannot allocate private origin structure";
+		r = -ENOMEM;
+		goto bad_alloc;
+	}
+
+	r = dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &o->dev);
 	if (r) {
 		ti->error = "Cannot get target device";
-		return r;
+		goto bad_open;
 	}
 
-	ti->private = dev;
+	ti->private = o;
 	ti->num_flush_bios = 1;
 
 	return 0;
+
+bad_open:
+	kfree(o);
+bad_alloc:
+	return r;
 }
 
 static void origin_dtr(struct dm_target *ti)
 {
-	struct dm_dev *dev = ti->private;
-	dm_put_device(ti, dev);
+	struct dm_origin *o = ti->private;
+	dm_put_device(ti, o->dev);
+	kfree(o);
 }
 
 static int origin_map(struct dm_target *ti, struct bio *bio)
 {
-	struct dm_dev *dev = ti->private;
-	bio->bi_bdev = dev->bdev;
+	struct dm_origin *o = ti->private;
+	unsigned available_sectors;
 
-	if (bio->bi_rw & REQ_FLUSH)
+	bio->bi_bdev = o->dev->bdev;
+
+	if (unlikely(bio->bi_rw & REQ_FLUSH))
 		return DM_MAPIO_REMAPPED;
 
+	if (bio_rw(bio) != WRITE)
+		return DM_MAPIO_REMAPPED;
+
+	available_sectors = o->split_boundary -
+		((unsigned)bio->bi_iter.bi_sector & (o->split_boundary - 1));
+
+	if (bio_sectors(bio) > available_sectors)
+		dm_accept_partial_bio(bio, available_sectors);
+
 	/* Only tell snapshots if this is a write */
-	return (bio_rw(bio) == WRITE) ? do_origin(dev, bio) : DM_MAPIO_REMAPPED;
+	return do_origin(o->dev, bio);
 }
 
 /*
@@ -2192,15 +2221,15 @@ static int origin_map(struct dm_target *ti, struct bio *bio)
  */
 static void origin_resume(struct dm_target *ti)
 {
-	struct dm_dev *dev = ti->private;
+	struct dm_origin *o = ti->private;
 
-	ti->max_io_len = get_origin_minimum_chunksize(dev->bdev);
+	o->split_boundary = get_origin_minimum_chunksize(o->dev->bdev);
 }
 
 static void origin_status(struct dm_target *ti, status_type_t type,
 			  unsigned status_flags, char *result, unsigned maxlen)
 {
-	struct dm_dev *dev = ti->private;
+	struct dm_origin *o = ti->private;
 
 	switch (type) {
 	case STATUSTYPE_INFO:
@@ -2208,7 +2237,7 @@ static void origin_status(struct dm_target *ti, status_type_t type,
 		break;
 
 	case STATUSTYPE_TABLE:
-		snprintf(result, maxlen, "%s", dev->name);
+		snprintf(result, maxlen, "%s", o->dev->name);
 		break;
 	}
 }
@@ -2216,13 +2245,13 @@ static void origin_status(struct dm_target *ti, status_type_t type,
 static int origin_merge(struct dm_target *ti, struct bvec_merge_data *bvm,
 			struct bio_vec *biovec, int max_size)
 {
-	struct dm_dev *dev = ti->private;
-	struct request_queue *q = bdev_get_queue(dev->bdev);
+	struct dm_origin *o = ti->private;
+	struct request_queue *q = bdev_get_queue(o->dev->bdev);
 
 	if (!q->merge_bvec_fn)
 		return max_size;
 
-	bvm->bi_bdev = dev->bdev;
+	bvm->bi_bdev = o->dev->bdev;
 
 	return min(max_size, q->merge_bvec_fn(q, bvm, biovec));
 }
@@ -2230,9 +2259,9 @@ static int origin_merge(struct dm_target *ti, struct bvec_merge_data *bvm,
 static int origin_iterate_devices(struct dm_target *ti,
 				  iterate_devices_callout_fn fn, void *data)
 {
-	struct dm_dev *dev = ti->private;
+	struct dm_origin *o = ti->private;
 
-	return fn(ti, dev, 0, ti->len, data);
+	return fn(ti, o->dev, 0, ti->len, data);
 }
 
 static struct target_type origin_target = {
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 50601ec7017a..5f59f1e3e5b1 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -465,8 +465,8 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 }
 EXPORT_SYMBOL(dm_get_device);
 
-int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
-			 sector_t start, sector_t len, void *data)
+static int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
+				sector_t start, sector_t len, void *data)
 {
 	struct queue_limits *limits = data;
 	struct block_device *bdev = dev->bdev;
@@ -499,7 +499,6 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
 					  (unsigned int) (PAGE_SIZE >> 9));
 	return 0;
 }
-EXPORT_SYMBOL_GPL(dm_set_device_limits);
 
 /*
  * Decrement a device's use count and remove it if necessary.
diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c
index 242ac2ea5f29..fc9c848a60c9 100644
--- a/drivers/md/dm-thin.c
+++ b/drivers/md/dm-thin.c
@@ -310,13 +310,18 @@ static void cell_defer_no_holder_no_free(struct thin_c *tc,
 	wake_worker(pool);
 }
 
-static void cell_error(struct pool *pool,
-		       struct dm_bio_prison_cell *cell)
+static void cell_error_with_code(struct pool *pool,
+				 struct dm_bio_prison_cell *cell, int error_code)
 {
-	dm_cell_error(pool->prison, cell);
+	dm_cell_error(pool->prison, cell, error_code);
 	dm_bio_prison_free_cell(pool->prison, cell);
 }
 
+static void cell_error(struct pool *pool, struct dm_bio_prison_cell *cell)
+{
+	cell_error_with_code(pool, cell, -EIO);
+}
+
 /*----------------------------------------------------------------*/
 
 /*
@@ -1027,7 +1032,7 @@ static void retry_on_resume(struct bio *bio)
 	spin_unlock_irqrestore(&tc->lock, flags);
 }
 
-static bool should_error_unserviceable_bio(struct pool *pool)
+static int should_error_unserviceable_bio(struct pool *pool)
 {
 	enum pool_mode m = get_pool_mode(pool);
 
@@ -1035,25 +1040,27 @@ static bool should_error_unserviceable_bio(struct pool *pool)
 	case PM_WRITE:
 		/* Shouldn't get here */
 		DMERR_LIMIT("bio unserviceable, yet pool is in PM_WRITE mode");
-		return true;
+		return -EIO;
 
 	case PM_OUT_OF_DATA_SPACE:
-		return pool->pf.error_if_no_space;
+		return pool->pf.error_if_no_space ? -ENOSPC : 0;
 
 	case PM_READ_ONLY:
 	case PM_FAIL:
-		return true;
+		return -EIO;
 	default:
 		/* Shouldn't get here */
 		DMERR_LIMIT("bio unserviceable, yet pool has an unknown mode");
-		return true;
+		return -EIO;
 	}
 }
 
 static void handle_unserviceable_bio(struct pool *pool, struct bio *bio)
 {
-	if (should_error_unserviceable_bio(pool))
-		bio_io_error(bio);
+	int error = should_error_unserviceable_bio(pool);
+
+	if (error)
+		bio_endio(bio, error);
 	else
 		retry_on_resume(bio);
 }
@@ -1062,18 +1069,21 @@ static void retry_bios_on_resume(struct pool *pool, struct dm_bio_prison_cell *c
 {
 	struct bio *bio;
 	struct bio_list bios;
+	int error;
 
-	if (should_error_unserviceable_bio(pool)) {
-		cell_error(pool, cell);
+	error = should_error_unserviceable_bio(pool);
+	if (error) {
+		cell_error_with_code(pool, cell, error);
 		return;
 	}
 
 	bio_list_init(&bios);
 	cell_release(pool, cell, &bios);
 
-	if (should_error_unserviceable_bio(pool))
+	error = should_error_unserviceable_bio(pool);
+	if (error)
 		while ((bio = bio_list_pop(&bios)))
-			bio_io_error(bio);
+			bio_endio(bio, error);
 	else
 		while ((bio = bio_list_pop(&bios)))
 			retry_on_resume(bio);
@@ -1610,47 +1620,63 @@ static void do_no_space_timeout(struct work_struct *ws)
 
 /*----------------------------------------------------------------*/
 
-struct noflush_work {
+struct pool_work {
 	struct work_struct worker;
-	struct thin_c *tc;
+	struct completion complete;
+};
+
+static struct pool_work *to_pool_work(struct work_struct *ws)
+{
+	return container_of(ws, struct pool_work, worker);
+}
 
-	atomic_t complete;
-	wait_queue_head_t wait;
+static void pool_work_complete(struct pool_work *pw)
+{
+	complete(&pw->complete);
+}
+
+static void pool_work_wait(struct pool_work *pw, struct pool *pool,
+			   void (*fn)(struct work_struct *))
+{
+	INIT_WORK_ONSTACK(&pw->worker, fn);
+	init_completion(&pw->complete);
+	queue_work(pool->wq, &pw->worker);
+	wait_for_completion(&pw->complete);
+}
+
+/*----------------------------------------------------------------*/
+
+struct noflush_work {
+	struct pool_work pw;
+	struct thin_c *tc;
 };
 
-static void complete_noflush_work(struct noflush_work *w)
+static struct noflush_work *to_noflush(struct work_struct *ws)
 {
-	atomic_set(&w->complete, 1);
-	wake_up(&w->wait);
+	return container_of(to_pool_work(ws), struct noflush_work, pw);
 }
 
 static void do_noflush_start(struct work_struct *ws)
 {
-	struct noflush_work *w = container_of(ws, struct noflush_work, worker);
+	struct noflush_work *w = to_noflush(ws);
 	w->tc->requeue_mode = true;
 	requeue_io(w->tc);
-	complete_noflush_work(w);
+	pool_work_complete(&w->pw);
 }
 
 static void do_noflush_stop(struct work_struct *ws)
 {
-	struct noflush_work *w = container_of(ws, struct noflush_work, worker);
+	struct noflush_work *w = to_noflush(ws);
 	w->tc->requeue_mode = false;
-	complete_noflush_work(w);
+	pool_work_complete(&w->pw);
 }
 
 static void noflush_work(struct thin_c *tc, void (*fn)(struct work_struct *))
 {
 	struct noflush_work w;
 
-	INIT_WORK_ONSTACK(&w.worker, fn);
 	w.tc = tc;
-	atomic_set(&w.complete, 0);
-	init_waitqueue_head(&w.wait);
-
-	queue_work(tc->pool->wq, &w.worker);
-
-	wait_event(w.wait, atomic_read(&w.complete));
+	pool_work_wait(&w.pw, tc->pool, fn);
 }
 
 /*----------------------------------------------------------------*/
@@ -3068,7 +3094,8 @@ static void set_discard_limits(struct pool_c *pt, struct queue_limits *limits)
 	 */
 	if (pt->adjusted_pf.discard_passdown) {
 		data_limits = &bdev_get_queue(pt->data_dev->bdev)->limits;
-		limits->discard_granularity = data_limits->discard_granularity;
+		limits->discard_granularity = max(data_limits->discard_granularity,
+						  pool->sectors_per_block << SECTOR_SHIFT);
 	} else
 		limits->discard_granularity = pool->sectors_per_block << SECTOR_SHIFT;
 }
diff --git a/drivers/md/dm-zero.c b/drivers/md/dm-zero.c
index c99003e0d47a..b9a64bbce304 100644
--- a/drivers/md/dm-zero.c
+++ b/drivers/md/dm-zero.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2003 Christophe Saout <christophe@saout.de>
+ * Copyright (C) 2003 Jana Saout <jana@saout.de>
  *
  * This file is released under the GPL.
  */
@@ -79,6 +79,6 @@ static void __exit dm_zero_exit(void)
 module_init(dm_zero_init)
 module_exit(dm_zero_exit)
 
-MODULE_AUTHOR("Christophe Saout <christophe@saout.de>");
+MODULE_AUTHOR("Jana Saout <jana@saout.de>");
 MODULE_DESCRIPTION(DM_NAME " dummy target returning zeros");
 MODULE_LICENSE("GPL");
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index aa9e093343d4..32b958dbc499 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -54,6 +54,8 @@ static void do_deferred_remove(struct work_struct *w);
 
 static DECLARE_WORK(deferred_remove_work, do_deferred_remove);
 
+static struct workqueue_struct *deferred_remove_workqueue;
+
 /*
  * For bio-based dm.
  * One of these is allocated per bio.
@@ -276,16 +278,24 @@ static int __init local_init(void)
 	if (r)
 		goto out_free_rq_tio_cache;
 
+	deferred_remove_workqueue = alloc_workqueue("kdmremove", WQ_UNBOUND, 1);
+	if (!deferred_remove_workqueue) {
+		r = -ENOMEM;
+		goto out_uevent_exit;
+	}
+
 	_major = major;
 	r = register_blkdev(_major, _name);
 	if (r < 0)
-		goto out_uevent_exit;
+		goto out_free_workqueue;
 
 	if (!_major)
 		_major = r;
 
 	return 0;
 
+out_free_workqueue:
+	destroy_workqueue(deferred_remove_workqueue);
 out_uevent_exit:
 	dm_uevent_exit();
 out_free_rq_tio_cache:
@@ -299,6 +309,7 @@ out_free_io_cache:
 static void local_exit(void)
 {
 	flush_scheduled_work();
+	destroy_workqueue(deferred_remove_workqueue);
 
 	kmem_cache_destroy(_rq_tio_cache);
 	kmem_cache_destroy(_io_cache);
@@ -407,7 +418,7 @@ static void dm_blk_close(struct gendisk *disk, fmode_t mode)
 
 	if (atomic_dec_and_test(&md->open_count) &&
 	    (test_bit(DMF_DEFERRED_REMOVE, &md->flags)))
-		schedule_work(&deferred_remove_work);
+		queue_work(deferred_remove_workqueue, &deferred_remove_work);
 
 	dm_put(md);
 
@@ -755,6 +766,14 @@ static void dec_pending(struct dm_io *io, int error)
 	}
 }
 
+static void disable_write_same(struct mapped_device *md)
+{
+	struct queue_limits *limits = dm_get_queue_limits(md);
+
+	/* device doesn't really support WRITE SAME, disable it */
+	limits->max_write_same_sectors = 0;
+}
+
 static void clone_endio(struct bio *bio, int error)
 {
 	int r = 0;
@@ -783,6 +802,10 @@ static void clone_endio(struct bio *bio, int error)
 		}
 	}
 
+	if (unlikely(r == -EREMOTEIO && (bio->bi_rw & REQ_WRITE_SAME) &&
+		     !bdev_get_queue(bio->bi_bdev)->limits.max_write_same_sectors))
+		disable_write_same(md);
+
 	free_tio(md, tio);
 	dec_pending(io, error);
 }
@@ -977,6 +1000,10 @@ static void dm_done(struct request *clone, int error, bool mapped)
 			r = rq_end_io(tio->ti, clone, error, &tio->info);
 	}
 
+	if (unlikely(r == -EREMOTEIO && (clone->cmd_flags & REQ_WRITE_SAME) &&
+		     !clone->q->limits.max_write_same_sectors))
+		disable_write_same(tio->md);
+
 	if (r <= 0)
 		/* The target wants to complete the I/O */
 		dm_end_request(clone, r);
@@ -1110,6 +1137,46 @@ int dm_set_target_max_io_len(struct dm_target *ti, sector_t len)
 }
 EXPORT_SYMBOL_GPL(dm_set_target_max_io_len);
 
+/*
+ * A target may call dm_accept_partial_bio only from the map routine.  It is
+ * allowed for all bio types except REQ_FLUSH.
+ *
+ * dm_accept_partial_bio informs the dm that the target only wants to process
+ * additional n_sectors sectors of the bio and the rest of the data should be
+ * sent in a next bio.
+ *
+ * A diagram that explains the arithmetics:
+ * +--------------------+---------------+-------+
+ * |         1          |       2       |   3   |
+ * +--------------------+---------------+-------+
+ *
+ * <-------------- *tio->len_ptr --------------->
+ *                      <------- bi_size ------->
+ *                      <-- n_sectors -->
+ *
+ * Region 1 was already iterated over with bio_advance or similar function.
+ *	(it may be empty if the target doesn't use bio_advance)
+ * Region 2 is the remaining bio size that the target wants to process.
+ *	(it may be empty if region 1 is non-empty, although there is no reason
+ *	 to make it empty)
+ * The target requires that region 3 is to be sent in the next bio.
+ *
+ * If the target wants to receive multiple copies of the bio (via num_*bios, etc),
+ * the partially processed part (the sum of regions 1+2) must be the same for all
+ * copies of the bio.
+ */
+void dm_accept_partial_bio(struct bio *bio, unsigned n_sectors)
+{
+	struct dm_target_io *tio = container_of(bio, struct dm_target_io, clone);
+	unsigned bi_size = bio->bi_iter.bi_size >> SECTOR_SHIFT;
+	BUG_ON(bio->bi_rw & REQ_FLUSH);
+	BUG_ON(bi_size > *tio->len_ptr);
+	BUG_ON(n_sectors > bi_size);
+	*tio->len_ptr -= bi_size - n_sectors;
+	bio->bi_iter.bi_size = n_sectors << SECTOR_SHIFT;
+}
+EXPORT_SYMBOL_GPL(dm_accept_partial_bio);
+
 static void __map_bio(struct dm_target_io *tio)
 {
 	int r;
@@ -1152,10 +1219,10 @@ struct clone_info {
 	struct bio *bio;
 	struct dm_io *io;
 	sector_t sector;
-	sector_t sector_count;
+	unsigned sector_count;
 };
 
-static void bio_setup_sector(struct bio *bio, sector_t sector, sector_t len)
+static void bio_setup_sector(struct bio *bio, sector_t sector, unsigned len)
 {
 	bio->bi_iter.bi_sector = sector;
 	bio->bi_iter.bi_size = to_bytes(len);
@@ -1200,11 +1267,13 @@ static struct dm_target_io *alloc_tio(struct clone_info *ci,
 
 static void __clone_and_map_simple_bio(struct clone_info *ci,
 				       struct dm_target *ti,
-				       unsigned target_bio_nr, sector_t len)
+				       unsigned target_bio_nr, unsigned *len)
 {
 	struct dm_target_io *tio = alloc_tio(ci, ti, ci->bio->bi_max_vecs, target_bio_nr);
 	struct bio *clone = &tio->clone;
 
+	tio->len_ptr = len;
+
 	/*
 	 * Discard requests require the bio's inline iovecs be initialized.
 	 * ci->bio->bi_max_vecs is BIO_INLINE_VECS anyway, for both flush
@@ -1212,13 +1281,13 @@ static void __clone_and_map_simple_bio(struct clone_info *ci,
 	 */
 	 __bio_clone_fast(clone, ci->bio);
 	if (len)
-		bio_setup_sector(clone, ci->sector, len);
+		bio_setup_sector(clone, ci->sector, *len);
 
 	__map_bio(tio);
 }
 
 static void __send_duplicate_bios(struct clone_info *ci, struct dm_target *ti,
-				  unsigned num_bios, sector_t len)
+				  unsigned num_bios, unsigned *len)
 {
 	unsigned target_bio_nr;
 
@@ -1233,13 +1302,13 @@ static int __send_empty_flush(struct clone_info *ci)
 
 	BUG_ON(bio_has_data(ci->bio));
 	while ((ti = dm_table_get_target(ci->map, target_nr++)))
-		__send_duplicate_bios(ci, ti, ti->num_flush_bios, 0);
+		__send_duplicate_bios(ci, ti, ti->num_flush_bios, NULL);
 
 	return 0;
 }
 
 static void __clone_and_map_data_bio(struct clone_info *ci, struct dm_target *ti,
-				     sector_t sector, unsigned len)
+				     sector_t sector, unsigned *len)
 {
 	struct bio *bio = ci->bio;
 	struct dm_target_io *tio;
@@ -1254,7 +1323,8 @@ static void __clone_and_map_data_bio(struct clone_info *ci, struct dm_target *ti
 
 	for (target_bio_nr = 0; target_bio_nr < num_target_bios; target_bio_nr++) {
 		tio = alloc_tio(ci, ti, 0, target_bio_nr);
-		clone_bio(tio, bio, sector, len);
+		tio->len_ptr = len;
+		clone_bio(tio, bio, sector, *len);
 		__map_bio(tio);
 	}
 }
@@ -1283,7 +1353,7 @@ static int __send_changing_extent_only(struct clone_info *ci,
 				       is_split_required_fn is_split_required)
 {
 	struct dm_target *ti;
-	sector_t len;
+	unsigned len;
 	unsigned num_bios;
 
 	do {
@@ -1302,11 +1372,11 @@ static int __send_changing_extent_only(struct clone_info *ci,
 			return -EOPNOTSUPP;
 
 		if (is_split_required && !is_split_required(ti))
-			len = min(ci->sector_count, max_io_len_target_boundary(ci->sector, ti));
+			len = min((sector_t)ci->sector_count, max_io_len_target_boundary(ci->sector, ti));
 		else
-			len = min(ci->sector_count, max_io_len(ci->sector, ti));
+			len = min((sector_t)ci->sector_count, max_io_len(ci->sector, ti));
 
-		__send_duplicate_bios(ci, ti, num_bios, len);
+		__send_duplicate_bios(ci, ti, num_bios, &len);
 
 		ci->sector += len;
 	} while (ci->sector_count -= len);
@@ -1345,7 +1415,7 @@ static int __split_and_process_non_flush(struct clone_info *ci)
 
 	len = min_t(sector_t, max_io_len(ci->sector, ti), ci->sector_count);
 
-	__clone_and_map_data_bio(ci, ti, ci->sector, len);
+	__clone_and_map_data_bio(ci, ti, ci->sector, &len);
 
 	ci->sector += len;
 	ci->sector_count -= len;
@@ -1439,7 +1509,6 @@ static int dm_merge_bvec(struct request_queue *q,
 	 * just one page.
 	 */
 	else if (queue_max_hw_sectors(q) <= PAGE_SIZE >> 9)
-
 		max_size = 0;
 
 out:
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 2382cfc9bb3f..32fc19c540d4 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3448,6 +3448,8 @@ level_store(struct mddev *mddev, const char *buf, size_t len)
 		mddev->level = LEVEL_NONE;
 		return rv;
 	}
+	if (mddev->ro)
+		return  -EROFS;
 
 	/* request to change the personality.  Need to ensure:
 	 *  - array is not engaged in resync/recovery/reshape
@@ -3634,6 +3636,8 @@ layout_store(struct mddev *mddev, const char *buf, size_t len)
 		int err;
 		if (mddev->pers->check_reshape == NULL)
 			return -EBUSY;
+		if (mddev->ro)
+			return -EROFS;
 		mddev->new_layout = n;
 		err = mddev->pers->check_reshape(mddev);
 		if (err) {
@@ -3723,6 +3727,8 @@ chunk_size_store(struct mddev *mddev, const char *buf, size_t len)
 		int err;
 		if (mddev->pers->check_reshape == NULL)
 			return -EBUSY;
+		if (mddev->ro)
+			return -EROFS;
 		mddev->new_chunk_sectors = n >> 9;
 		err = mddev->pers->check_reshape(mddev);
 		if (err) {
@@ -5593,7 +5599,7 @@ static int get_array_info(struct mddev * mddev, void __user * arg)
 	if (mddev->in_sync)
 		info.state = (1<<MD_SB_CLEAN);
 	if (mddev->bitmap && mddev->bitmap_info.offset)
-		info.state = (1<<MD_SB_BITMAP_PRESENT);
+		info.state |= (1<<MD_SB_BITMAP_PRESENT);
 	info.active_disks  = insync;
 	info.working_disks = working;
 	info.failed_disks  = failed;
@@ -6135,6 +6141,8 @@ static int update_size(struct mddev *mddev, sector_t num_sectors)
 	 */
 	if (mddev->sync_thread)
 		return -EBUSY;
+	if (mddev->ro)
+		return -EROFS;
 
 	rdev_for_each(rdev, mddev) {
 		sector_t avail = rdev->sectors;
@@ -6157,6 +6165,8 @@ static int update_raid_disks(struct mddev *mddev, int raid_disks)
 	/* change the number of raid disks */
 	if (mddev->pers->check_reshape == NULL)
 		return -EINVAL;
+	if (mddev->ro)
+		return -EROFS;
 	if (raid_disks <= 0 ||
 	    (mddev->max_disks && raid_disks >= mddev->max_disks))
 		return -EINVAL;
@@ -7491,6 +7501,19 @@ void md_do_sync(struct md_thread *thread)
 			    rdev->recovery_offset < j)
 				j = rdev->recovery_offset;
 		rcu_read_unlock();
+
+		/* If there is a bitmap, we need to make sure all
+		 * writes that started before we added a spare
+		 * complete before we start doing a recovery.
+		 * Otherwise the write might complete and (via
+		 * bitmap_endwrite) set a bit in the bitmap after the
+		 * recovery has checked that bit and skipped that
+		 * region.
+		 */
+		if (mddev->bitmap) {
+			mddev->pers->quiesce(mddev, 1);
+			mddev->pers->quiesce(mddev, 0);
+		}
 	}
 
 	printk(KERN_INFO "md: %s of RAID array %s\n", desc, mdname(mddev));
@@ -8333,7 +8356,7 @@ static int md_clear_badblocks(struct badblocks *bb, sector_t s, int sectors)
 			if (a < s) {
 				/* we need to split this range */
 				if (bb->count >= MD_MAX_BADBLOCKS) {
-					rv = 0;
+					rv = -ENOSPC;
 					goto out;
 				}
 				memmove(p+lo+1, p+lo, (bb->count - lo) * 8);
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 2afef4ec9312..6234b2e84587 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -292,9 +292,12 @@ static void do_release_stripe(struct r5conf *conf, struct stripe_head *sh,
 	BUG_ON(atomic_read(&conf->active_stripes)==0);
 	if (test_bit(STRIPE_HANDLE, &sh->state)) {
 		if (test_bit(STRIPE_DELAYED, &sh->state) &&
-		    !test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
+		    !test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
 			list_add_tail(&sh->lru, &conf->delayed_list);
-		else if (test_bit(STRIPE_BIT_DELAY, &sh->state) &&
+			if (atomic_read(&conf->preread_active_stripes)
+			    < IO_THRESHOLD)
+				md_wakeup_thread(conf->mddev->thread);
+		} else if (test_bit(STRIPE_BIT_DELAY, &sh->state) &&
 			   sh->bm_seq - conf->seq_write > 0)
 			list_add_tail(&sh->lru, &conf->bitmap_list);
 		else {
@@ -413,6 +416,11 @@ static void release_stripe(struct stripe_head *sh)
 	int hash;
 	bool wakeup;
 
+	/* Avoid release_list until the last reference.
+	 */
+	if (atomic_add_unless(&sh->count, -1, 1))
+		return;
+
 	if (unlikely(!conf->mddev->thread) ||
 		test_and_set_bit(STRIPE_ON_RELEASE_LIST, &sh->state))
 		goto slow_path;
@@ -479,6 +487,7 @@ static void shrink_buffers(struct stripe_head *sh)
 	int num = sh->raid_conf->pool_size;
 
 	for (i = 0; i < num ; i++) {
+		WARN_ON(sh->dev[i].page != sh->dev[i].orig_page);
 		p = sh->dev[i].page;
 		if (!p)
 			continue;
@@ -499,6 +508,7 @@ static int grow_buffers(struct stripe_head *sh)
 			return 1;
 		}
 		sh->dev[i].page = page;
+		sh->dev[i].orig_page = page;
 	}
 	return 0;
 }
@@ -855,6 +865,9 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 			if (test_bit(R5_ReadNoMerge, &sh->dev[i].flags))
 				bi->bi_rw |= REQ_NOMERGE;
 
+			if (test_bit(R5_SkipCopy, &sh->dev[i].flags))
+				WARN_ON(test_bit(R5_UPTODATE, &sh->dev[i].flags));
+			sh->dev[i].vec.bv_page = sh->dev[i].page;
 			bi->bi_vcnt = 1;
 			bi->bi_io_vec[0].bv_len = STRIPE_SIZE;
 			bi->bi_io_vec[0].bv_offset = 0;
@@ -899,6 +912,9 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 			else
 				rbi->bi_iter.bi_sector = (sh->sector
 						  + rrdev->data_offset);
+			if (test_bit(R5_SkipCopy, &sh->dev[i].flags))
+				WARN_ON(test_bit(R5_UPTODATE, &sh->dev[i].flags));
+			sh->dev[i].rvec.bv_page = sh->dev[i].page;
 			rbi->bi_vcnt = 1;
 			rbi->bi_io_vec[0].bv_len = STRIPE_SIZE;
 			rbi->bi_io_vec[0].bv_offset = 0;
@@ -927,8 +943,9 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 }
 
 static struct dma_async_tx_descriptor *
-async_copy_data(int frombio, struct bio *bio, struct page *page,
-	sector_t sector, struct dma_async_tx_descriptor *tx)
+async_copy_data(int frombio, struct bio *bio, struct page **page,
+	sector_t sector, struct dma_async_tx_descriptor *tx,
+	struct stripe_head *sh)
 {
 	struct bio_vec bvl;
 	struct bvec_iter iter;
@@ -965,11 +982,16 @@ async_copy_data(int frombio, struct bio *bio, struct page *page,
 		if (clen > 0) {
 			b_offset += bvl.bv_offset;
 			bio_page = bvl.bv_page;
-			if (frombio)
-				tx = async_memcpy(page, bio_page, page_offset,
+			if (frombio) {
+				if (sh->raid_conf->skip_copy &&
+				    b_offset == 0 && page_offset == 0 &&
+				    clen == STRIPE_SIZE)
+					*page = bio_page;
+				else
+					tx = async_memcpy(*page, bio_page, page_offset,
 						  b_offset, clen, &submit);
-			else
-				tx = async_memcpy(bio_page, page, b_offset,
+			} else
+				tx = async_memcpy(bio_page, *page, b_offset,
 						  page_offset, clen, &submit);
 		}
 		/* chain the operations */
@@ -1045,8 +1067,8 @@ static void ops_run_biofill(struct stripe_head *sh)
 			spin_unlock_irq(&sh->stripe_lock);
 			while (rbi && rbi->bi_iter.bi_sector <
 				dev->sector + STRIPE_SECTORS) {
-				tx = async_copy_data(0, rbi, dev->page,
-					dev->sector, tx);
+				tx = async_copy_data(0, rbi, &dev->page,
+					dev->sector, tx, sh);
 				rbi = r5_next_bio(rbi, dev->sector);
 			}
 		}
@@ -1384,6 +1406,7 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 			BUG_ON(dev->written);
 			wbi = dev->written = chosen;
 			spin_unlock_irq(&sh->stripe_lock);
+			WARN_ON(dev->page != dev->orig_page);
 
 			while (wbi && wbi->bi_iter.bi_sector <
 				dev->sector + STRIPE_SECTORS) {
@@ -1393,9 +1416,15 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 					set_bit(R5_SyncIO, &dev->flags);
 				if (wbi->bi_rw & REQ_DISCARD)
 					set_bit(R5_Discard, &dev->flags);
-				else
-					tx = async_copy_data(1, wbi, dev->page,
-						dev->sector, tx);
+				else {
+					tx = async_copy_data(1, wbi, &dev->page,
+						dev->sector, tx, sh);
+					if (dev->page != dev->orig_page) {
+						set_bit(R5_SkipCopy, &dev->flags);
+						clear_bit(R5_UPTODATE, &dev->flags);
+						clear_bit(R5_OVERWRITE, &dev->flags);
+					}
+				}
 				wbi = r5_next_bio(wbi, dev->sector);
 			}
 		}
@@ -1426,7 +1455,7 @@ static void ops_complete_reconstruct(void *stripe_head_ref)
 		struct r5dev *dev = &sh->dev[i];
 
 		if (dev->written || i == pd_idx || i == qd_idx) {
-			if (!discard)
+			if (!discard && !test_bit(R5_SkipCopy, &dev->flags))
 				set_bit(R5_UPTODATE, &dev->flags);
 			if (fua)
 				set_bit(R5_WantFUA, &dev->flags);
@@ -1839,8 +1868,10 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 		osh = get_free_stripe(conf, hash);
 		unlock_device_hash_lock(conf, hash);
 		atomic_set(&nsh->count, 1);
-		for(i=0; i<conf->pool_size; i++)
+		for(i=0; i<conf->pool_size; i++) {
 			nsh->dev[i].page = osh->dev[i].page;
+			nsh->dev[i].orig_page = osh->dev[i].page;
+		}
 		for( ; i<newsize; i++)
 			nsh->dev[i].page = NULL;
 		nsh->hash_lock_index = hash;
@@ -1896,6 +1927,7 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 			if (nsh->dev[i].page == NULL) {
 				struct page *p = alloc_page(GFP_NOIO);
 				nsh->dev[i].page = p;
+				nsh->dev[i].orig_page = p;
 				if (!p)
 					err = -ENOMEM;
 			}
@@ -2133,24 +2165,20 @@ static void raid5_end_write_request(struct bio *bi, int error)
 }
 
 static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous);
-	
+
 static void raid5_build_block(struct stripe_head *sh, int i, int previous)
 {
 	struct r5dev *dev = &sh->dev[i];
 
 	bio_init(&dev->req);
 	dev->req.bi_io_vec = &dev->vec;
-	dev->req.bi_vcnt++;
-	dev->req.bi_max_vecs++;
+	dev->req.bi_max_vecs = 1;
 	dev->req.bi_private = sh;
-	dev->vec.bv_page = dev->page;
 
 	bio_init(&dev->rreq);
 	dev->rreq.bi_io_vec = &dev->rvec;
-	dev->rreq.bi_vcnt++;
-	dev->rreq.bi_max_vecs++;
+	dev->rreq.bi_max_vecs = 1;
 	dev->rreq.bi_private = sh;
-	dev->rvec.bv_page = dev->page;
 
 	dev->flags = 0;
 	dev->sector = compute_blocknr(sh, i, previous);
@@ -2750,6 +2778,11 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 		/* and fail all 'written' */
 		bi = sh->dev[i].written;
 		sh->dev[i].written = NULL;
+		if (test_and_clear_bit(R5_SkipCopy, &sh->dev[i].flags)) {
+			WARN_ON(test_bit(R5_UPTODATE, &sh->dev[i].flags));
+			sh->dev[i].page = sh->dev[i].orig_page;
+		}
+
 		if (bi) bitmap_end = 1;
 		while (bi && bi->bi_iter.bi_sector <
 		       sh->dev[i].sector + STRIPE_SECTORS) {
@@ -2886,8 +2919,11 @@ static int fetch_block(struct stripe_head *sh, struct stripe_head_state *s,
 	     (s->failed >= 1 && fdev[0]->toread) ||
 	     (s->failed >= 2 && fdev[1]->toread) ||
 	     (sh->raid_conf->level <= 5 && s->failed && fdev[0]->towrite &&
+	      (!test_bit(R5_Insync, &dev->flags) || test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) &&
 	      !test_bit(R5_OVERWRITE, &fdev[0]->flags)) ||
-	     (sh->raid_conf->level == 6 && s->failed && s->to_write))) {
+	     (sh->raid_conf->level == 6 && s->failed && s->to_write &&
+	      s->to_write < sh->raid_conf->raid_disks - 2 &&
+	      (!test_bit(R5_Insync, &dev->flags) || test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))))) {
 		/* we would like to get this block, possibly by computing it,
 		 * otherwise read it if the backing disk is insync
 		 */
@@ -2991,12 +3027,17 @@ static void handle_stripe_clean_event(struct r5conf *conf,
 			dev = &sh->dev[i];
 			if (!test_bit(R5_LOCKED, &dev->flags) &&
 			    (test_bit(R5_UPTODATE, &dev->flags) ||
-			     test_bit(R5_Discard, &dev->flags))) {
+			     test_bit(R5_Discard, &dev->flags) ||
+			     test_bit(R5_SkipCopy, &dev->flags))) {
 				/* We can return any write requests */
 				struct bio *wbi, *wbi2;
 				pr_debug("Return write for disc %d\n", i);
 				if (test_and_clear_bit(R5_Discard, &dev->flags))
 					clear_bit(R5_UPTODATE, &dev->flags);
+				if (test_and_clear_bit(R5_SkipCopy, &dev->flags)) {
+					WARN_ON(test_bit(R5_UPTODATE, &dev->flags));
+					dev->page = dev->orig_page;
+				}
 				wbi = dev->written;
 				dev->written = NULL;
 				while (wbi && wbi->bi_iter.bi_sector <
@@ -3015,6 +3056,8 @@ static void handle_stripe_clean_event(struct r5conf *conf,
 						0);
 			} else if (test_bit(R5_Discard, &dev->flags))
 				discard_pending = 1;
+			WARN_ON(test_bit(R5_SkipCopy, &dev->flags));
+			WARN_ON(dev->page != dev->orig_page);
 		}
 	if (!discard_pending &&
 	    test_bit(R5_Discard, &sh->dev[sh->pd_idx].flags)) {
@@ -3086,7 +3129,8 @@ static void handle_stripe_dirtying(struct r5conf *conf,
 		    !test_bit(R5_LOCKED, &dev->flags) &&
 		    !(test_bit(R5_UPTODATE, &dev->flags) ||
 		    test_bit(R5_Wantcompute, &dev->flags))) {
-			if (test_bit(R5_Insync, &dev->flags)) rcw++;
+			if (test_bit(R5_Insync, &dev->flags))
+				rcw++;
 			else
 				rcw += 2*disks;
 		}
@@ -3107,10 +3151,10 @@ static void handle_stripe_dirtying(struct r5conf *conf,
 			    !(test_bit(R5_UPTODATE, &dev->flags) ||
 			    test_bit(R5_Wantcompute, &dev->flags)) &&
 			    test_bit(R5_Insync, &dev->flags)) {
-				if (
-				  test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
-					pr_debug("Read_old block "
-						 "%d for r-m-w\n", i);
+				if (test_bit(STRIPE_PREREAD_ACTIVE,
+					     &sh->state)) {
+					pr_debug("Read_old block %d for r-m-w\n",
+						 i);
 					set_bit(R5_LOCKED, &dev->flags);
 					set_bit(R5_Wantread, &dev->flags);
 					s->locked++;
@@ -3133,10 +3177,9 @@ static void handle_stripe_dirtying(struct r5conf *conf,
 			    !(test_bit(R5_UPTODATE, &dev->flags) ||
 			      test_bit(R5_Wantcompute, &dev->flags))) {
 				rcw++;
-				if (!test_bit(R5_Insync, &dev->flags))
-					continue; /* it's a failed drive */
-				if (
-				  test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+				if (test_bit(R5_Insync, &dev->flags) &&
+				    test_bit(STRIPE_PREREAD_ACTIVE,
+					     &sh->state)) {
 					pr_debug("Read_old block "
 						"%d for Reconstruct\n", i);
 					set_bit(R5_LOCKED, &dev->flags);
@@ -5031,8 +5074,8 @@ static inline sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int
 	bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks, still_degraded);
 
 	set_bit(STRIPE_SYNC_REQUESTED, &sh->state);
+	set_bit(STRIPE_HANDLE, &sh->state);
 
-	handle_stripe(sh);
 	release_stripe(sh);
 
 	return STRIPE_SECTORS;
@@ -5072,7 +5115,7 @@ static int  retry_aligned_read(struct r5conf *conf, struct bio *raid_bio)
 			/* already done this stripe */
 			continue;
 
-		sh = get_active_stripe(conf, sector, 0, 1, 0);
+		sh = get_active_stripe(conf, sector, 0, 1, 1);
 
 		if (!sh) {
 			/* failed to get a stripe - must wait */
@@ -5355,6 +5398,50 @@ raid5_preread_bypass_threshold = __ATTR(preread_bypass_threshold,
 					raid5_store_preread_threshold);
 
 static ssize_t
+raid5_show_skip_copy(struct mddev *mddev, char *page)
+{
+	struct r5conf *conf = mddev->private;
+	if (conf)
+		return sprintf(page, "%d\n", conf->skip_copy);
+	else
+		return 0;
+}
+
+static ssize_t
+raid5_store_skip_copy(struct mddev *mddev, const char *page, size_t len)
+{
+	struct r5conf *conf = mddev->private;
+	unsigned long new;
+	if (len >= PAGE_SIZE)
+		return -EINVAL;
+	if (!conf)
+		return -ENODEV;
+
+	if (kstrtoul(page, 10, &new))
+		return -EINVAL;
+	new = !!new;
+	if (new == conf->skip_copy)
+		return len;
+
+	mddev_suspend(mddev);
+	conf->skip_copy = new;
+	if (new)
+		mddev->queue->backing_dev_info.capabilities |=
+						BDI_CAP_STABLE_WRITES;
+	else
+		mddev->queue->backing_dev_info.capabilities &=
+						~BDI_CAP_STABLE_WRITES;
+	mddev_resume(mddev);
+	return len;
+}
+
+static struct md_sysfs_entry
+raid5_skip_copy = __ATTR(skip_copy, S_IRUGO | S_IWUSR,
+					raid5_show_skip_copy,
+					raid5_store_skip_copy);
+
+
+static ssize_t
 stripe_cache_active_show(struct mddev *mddev, char *page)
 {
 	struct r5conf *conf = mddev->private;
@@ -5439,6 +5526,7 @@ static struct attribute *raid5_attrs[] =  {
 	&raid5_stripecache_active.attr,
 	&raid5_preread_bypass_threshold.attr,
 	&raid5_group_thread_cnt.attr,
+	&raid5_skip_copy.attr,
 	NULL,
 };
 static struct attribute_group raid5_attrs_group = {
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 01ad8ae8f578..bc72cd4be5f8 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -232,7 +232,7 @@ struct stripe_head {
 		 */
 		struct bio	req, rreq;
 		struct bio_vec	vec, rvec;
-		struct page	*page;
+		struct page	*page, *orig_page;
 		struct bio	*toread, *read, *towrite, *written;
 		sector_t	sector;			/* sector of this page */
 		unsigned long	flags;
@@ -299,6 +299,7 @@ enum r5dev_flags {
 			 * data in, and now is a good time to write it out.
 			 */
 	R5_Discard,	/* Discard the stripe */
+	R5_SkipCopy,	/* Don't copy data from bio to stripe cache */
 };
 
 /*
@@ -436,6 +437,7 @@ struct r5conf {
 	atomic_t		pending_full_writes; /* full write backlog */
 	int			bypass_count; /* bypassed prereads */
 	int			bypass_threshold; /* preread nice */
+	int			skip_copy; /* Don't copy data from bio to stripe cache */
 	struct list_head	*last_hold; /* detect hold_list promotions */
 
 	atomic_t		reshape_stripes; /* stripes with pending writes for reshape */
diff --git a/drivers/media/platform/Kconfig b/drivers/media/platform/Kconfig
index 20f1655e6d75..8108c698b548 100644
--- a/drivers/media/platform/Kconfig
+++ b/drivers/media/platform/Kconfig
@@ -93,7 +93,9 @@ config VIDEO_M32R_AR_M64278
 
 config VIDEO_OMAP3
 	tristate "OMAP 3 Camera support"
-	depends on OMAP_IOVMM && VIDEO_V4L2 && I2C && VIDEO_V4L2_SUBDEV_API && ARCH_OMAP3
+	depends on VIDEO_V4L2 && I2C && VIDEO_V4L2_SUBDEV_API && ARCH_OMAP3
+	select ARM_DMA_USE_IOMMU
+	select OMAP_IOMMU
 	---help---
 	  Driver for an OMAP 3 camera controller.
 
diff --git a/drivers/media/platform/omap3isp/Makefile b/drivers/media/platform/omap3isp/Makefile
index e8847e79e31a..254975a9174e 100644
--- a/drivers/media/platform/omap3isp/Makefile
+++ b/drivers/media/platform/omap3isp/Makefile
@@ -3,7 +3,7 @@
 ccflags-$(CONFIG_VIDEO_OMAP3_DEBUG) += -DDEBUG
 
 omap3-isp-objs += \
-	isp.o ispqueue.o ispvideo.o \
+	isp.o ispvideo.o \
 	ispcsiphy.o ispccp2.o ispcsi2.o \
 	ispccdc.o isppreview.o ispresizer.o \
 	ispstat.o isph3a_aewb.o isph3a_af.o isphist.o
diff --git a/drivers/media/platform/omap3isp/isp.c b/drivers/media/platform/omap3isp/isp.c
index 06a0df434249..2c7aa6720569 100644
--- a/drivers/media/platform/omap3isp/isp.c
+++ b/drivers/media/platform/omap3isp/isp.c
@@ -69,6 +69,8 @@
 #include <linux/sched.h>
 #include <linux/vmalloc.h>
 
+#include <asm/dma-iommu.h>
+
 #include <media/v4l2-common.h>
 #include <media/v4l2-device.h>
 
@@ -1397,14 +1399,14 @@ int omap3isp_module_sync_idle(struct media_entity *me, wait_queue_head_t *wait,
 	if (isp_pipeline_is_last(me)) {
 		struct isp_video *video = pipe->output;
 		unsigned long flags;
-		spin_lock_irqsave(&video->queue->irqlock, flags);
+		spin_lock_irqsave(&video->irqlock, flags);
 		if (video->dmaqueue_flags & ISP_VIDEO_DMAQUEUE_UNDERRUN) {
-			spin_unlock_irqrestore(&video->queue->irqlock, flags);
+			spin_unlock_irqrestore(&video->irqlock, flags);
 			atomic_set(stopping, 0);
 			smp_mb();
 			return 0;
 		}
-		spin_unlock_irqrestore(&video->queue->irqlock, flags);
+		spin_unlock_irqrestore(&video->irqlock, flags);
 		if (!wait_event_timeout(*wait, !atomic_read(stopping),
 					msecs_to_jiffies(1000))) {
 			atomic_set(stopping, 0);
@@ -1625,7 +1627,7 @@ struct isp_device *omap3isp_get(struct isp_device *isp)
  * Decrement the reference count on the ISP. If the last reference is released,
  * power-down all submodules, disable clocks and free temporary buffers.
  */
-void omap3isp_put(struct isp_device *isp)
+static void __omap3isp_put(struct isp_device *isp, bool save_ctx)
 {
 	if (isp == NULL)
 		return;
@@ -1634,7 +1636,7 @@ void omap3isp_put(struct isp_device *isp)
 	BUG_ON(isp->ref_count == 0);
 	if (--isp->ref_count == 0) {
 		isp_disable_interrupts(isp);
-		if (isp->domain) {
+		if (save_ctx) {
 			isp_save_ctx(isp);
 			isp->has_context = 1;
 		}
@@ -1648,6 +1650,11 @@ void omap3isp_put(struct isp_device *isp)
 	mutex_unlock(&isp->isp_mutex);
 }
 
+void omap3isp_put(struct isp_device *isp)
+{
+	__omap3isp_put(isp, true);
+}
+
 /* --------------------------------------------------------------------------
  * Platform device driver
  */
@@ -2120,6 +2127,61 @@ error_csiphy:
 	return ret;
 }
 
+static void isp_detach_iommu(struct isp_device *isp)
+{
+	arm_iommu_release_mapping(isp->mapping);
+	isp->mapping = NULL;
+	iommu_group_remove_device(isp->dev);
+}
+
+static int isp_attach_iommu(struct isp_device *isp)
+{
+	struct dma_iommu_mapping *mapping;
+	struct iommu_group *group;
+	int ret;
+
+	/* Create a device group and add the device to it. */
+	group = iommu_group_alloc();
+	if (IS_ERR(group)) {
+		dev_err(isp->dev, "failed to allocate IOMMU group\n");
+		return PTR_ERR(group);
+	}
+
+	ret = iommu_group_add_device(group, isp->dev);
+	iommu_group_put(group);
+
+	if (ret < 0) {
+		dev_err(isp->dev, "failed to add device to IPMMU group\n");
+		return ret;
+	}
+
+	/*
+	 * Create the ARM mapping, used by the ARM DMA mapping core to allocate
+	 * VAs. This will allocate a corresponding IOMMU domain.
+	 */
+	mapping = arm_iommu_create_mapping(&platform_bus_type, SZ_1G, SZ_2G);
+	if (IS_ERR(mapping)) {
+		dev_err(isp->dev, "failed to create ARM IOMMU mapping\n");
+		ret = PTR_ERR(mapping);
+		goto error;
+	}
+
+	isp->mapping = mapping;
+
+	/* Attach the ARM VA mapping to the device. */
+	ret = arm_iommu_attach_device(isp->dev, mapping);
+	if (ret < 0) {
+		dev_err(isp->dev, "failed to attach device to VA mapping\n");
+		goto error;
+	}
+
+	return 0;
+
+error:
+	isp_detach_iommu(isp);
+	return ret;
+}
+
 /*
  * isp_remove - Remove ISP platform device
  * @pdev: Pointer to ISP platform device
@@ -2135,10 +2197,8 @@ static int isp_remove(struct platform_device *pdev)
 	isp_xclk_cleanup(isp);
 
 	__omap3isp_get(isp, false);
-	iommu_detach_device(isp->domain, &pdev->dev);
-	iommu_domain_free(isp->domain);
-	isp->domain = NULL;
-	omap3isp_put(isp);
+	isp_detach_iommu(isp);
+	__omap3isp_put(isp, false);
 
 	return 0;
 }
@@ -2265,39 +2325,32 @@ static int isp_probe(struct platform_device *pdev)
 		}
 	}
 
-	isp->domain = iommu_domain_alloc(pdev->dev.bus);
-	if (!isp->domain) {
-		dev_err(isp->dev, "can't alloc iommu domain\n");
-		ret = -ENOMEM;
+	/* IOMMU */
+	ret = isp_attach_iommu(isp);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "unable to attach to IOMMU\n");
 		goto error_isp;
 	}
 
-	ret = iommu_attach_device(isp->domain, &pdev->dev);
-	if (ret) {
-		dev_err(&pdev->dev, "can't attach iommu device: %d\n", ret);
-		ret = -EPROBE_DEFER;
-		goto free_domain;
-	}
-
 	/* Interrupt */
 	isp->irq_num = platform_get_irq(pdev, 0);
 	if (isp->irq_num <= 0) {
 		dev_err(isp->dev, "No IRQ resource\n");
 		ret = -ENODEV;
-		goto detach_dev;
+		goto error_iommu;
 	}
 
 	if (devm_request_irq(isp->dev, isp->irq_num, isp_isr, IRQF_SHARED,
 			     "OMAP3 ISP", isp)) {
 		dev_err(isp->dev, "Unable to request IRQ\n");
 		ret = -EINVAL;
-		goto detach_dev;
+		goto error_iommu;
 	}
 
 	/* Entities */
 	ret = isp_initialize_modules(isp);
 	if (ret < 0)
-		goto detach_dev;
+		goto error_iommu;
 
 	ret = isp_register_entities(isp);
 	if (ret < 0)
@@ -2310,14 +2363,11 @@ static int isp_probe(struct platform_device *pdev)
 
 error_modules:
 	isp_cleanup_modules(isp);
-detach_dev:
-	iommu_detach_device(isp->domain, &pdev->dev);
-free_domain:
-	iommu_domain_free(isp->domain);
-	isp->domain = NULL;
+error_iommu:
+	isp_detach_iommu(isp);
 error_isp:
 	isp_xclk_cleanup(isp);
-	omap3isp_put(isp);
+	__omap3isp_put(isp, false);
 error:
 	mutex_destroy(&isp->isp_mutex);
 
diff --git a/drivers/media/platform/omap3isp/isp.h b/drivers/media/platform/omap3isp/isp.h
index 6d5e69711907..2c314eea1252 100644
--- a/drivers/media/platform/omap3isp/isp.h
+++ b/drivers/media/platform/omap3isp/isp.h
@@ -45,8 +45,6 @@
 #include "ispcsi2.h"
 #include "ispccp2.h"
 
-#define IOMMU_FLAG (IOVMF_ENDIAN_LITTLE | IOVMF_ELSZ_8)
-
 #define ISP_TOK_TERM		0xFFFFFFFF	/*
 						 * terminating token for ISP
 						 * modules reg list
@@ -152,6 +150,7 @@ struct isp_xclk {
  *             regions.
  * @mmio_base_phys: Array with physical L4 bus addresses for ISP register
  *                  regions.
+ * @mapping: IOMMU mapping
  * @stat_lock: Spinlock for handling statistics
  * @isp_mutex: Mutex for serializing requests to ISP.
  * @stop_failure: Indicates that an entity failed to stop.
@@ -171,7 +170,6 @@ struct isp_xclk {
  * @isp_res: Pointer to current settings for ISP Resizer.
  * @isp_prev: Pointer to current settings for ISP Preview.
  * @isp_ccdc: Pointer to current settings for ISP CCDC.
- * @iommu: Pointer to requested IOMMU instance for ISP.
  * @platform_cb: ISP driver callback function pointers for platform code
  *
  * This structure is used to store the OMAP ISP Information.
@@ -189,6 +187,8 @@ struct isp_device {
 	void __iomem *mmio_base[OMAP3_ISP_IOMEM_LAST];
 	unsigned long mmio_base_phys[OMAP3_ISP_IOMEM_LAST];
 
+	struct dma_iommu_mapping *mapping;
+
 	/* ISP Obj */
 	spinlock_t stat_lock;	/* common lock for statistic drivers */
 	struct mutex isp_mutex;	/* For handling ref_count field */
@@ -219,8 +219,6 @@ struct isp_device {
 
 	unsigned int sbl_resources;
 	unsigned int subclk_resources;
-
-	struct iommu_domain *domain;
 };
 
 #define v4l2_dev_to_isp_device(dev) \
diff --git a/drivers/media/platform/omap3isp/ispccdc.c b/drivers/media/platform/omap3isp/ispccdc.c
index 4d920c800ff5..9f727d20f06d 100644
--- a/drivers/media/platform/omap3isp/ispccdc.c
+++ b/drivers/media/platform/omap3isp/ispccdc.c
@@ -30,7 +30,6 @@
 #include <linux/device.h>
 #include <linux/dma-mapping.h>
 #include <linux/mm.h>
-#include <linux/omap-iommu.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <media/v4l2-event.h>
@@ -206,7 +205,8 @@ static int ccdc_lsc_validate_config(struct isp_ccdc_device *ccdc,
  * ccdc_lsc_program_table - Program Lens Shading Compensation table address.
  * @ccdc: Pointer to ISP CCDC device.
  */
-static void ccdc_lsc_program_table(struct isp_ccdc_device *ccdc, u32 addr)
+static void ccdc_lsc_program_table(struct isp_ccdc_device *ccdc,
+				   dma_addr_t addr)
 {
 	isp_reg_writel(to_isp_device(ccdc), addr,
 		       OMAP3_ISP_IOMEM_CCDC, ISPCCDC_LSC_TABLE_BASE);
@@ -333,7 +333,7 @@ static int __ccdc_lsc_configure(struct isp_ccdc_device *ccdc,
 		return -EBUSY;
 
 	ccdc_lsc_setup_regs(ccdc, &req->config);
-	ccdc_lsc_program_table(ccdc, req->table);
+	ccdc_lsc_program_table(ccdc, req->table.dma);
 	return 0;
 }
 
@@ -368,11 +368,12 @@ static void ccdc_lsc_free_request(struct isp_ccdc_device *ccdc,
 	if (req == NULL)
 		return;
 
-	if (req->iovm)
-		dma_unmap_sg(isp->dev, req->iovm->sgt->sgl,
-			     req->iovm->sgt->nents, DMA_TO_DEVICE);
-	if (req->table)
-		omap_iommu_vfree(isp->domain, isp->dev, req->table);
+	if (req->table.addr) {
+		sg_free_table(&req->table.sgt);
+		dma_free_coherent(isp->dev, req->config.size, req->table.addr,
+				  req->table.dma);
+	}
+
 	kfree(req);
 }
 
@@ -416,7 +417,6 @@ static int ccdc_lsc_config(struct isp_ccdc_device *ccdc,
 	struct isp_device *isp = to_isp_device(ccdc);
 	struct ispccdc_lsc_config_req *req;
 	unsigned long flags;
-	void *table;
 	u16 update;
 	int ret;
 
@@ -444,38 +444,31 @@ static int ccdc_lsc_config(struct isp_ccdc_device *ccdc,
 
 		req->enable = 1;
 
-		req->table = omap_iommu_vmalloc(isp->domain, isp->dev, 0,
-					req->config.size, IOMMU_FLAG);
-		if (IS_ERR_VALUE(req->table)) {
-			req->table = 0;
-			ret = -ENOMEM;
-			goto done;
-		}
-
-		req->iovm = omap_find_iovm_area(isp->dev, req->table);
-		if (req->iovm == NULL) {
+		req->table.addr = dma_alloc_coherent(isp->dev, req->config.size,
+						     &req->table.dma,
+						     GFP_KERNEL);
+		if (req->table.addr == NULL) {
 			ret = -ENOMEM;
 			goto done;
 		}
 
-		if (!dma_map_sg(isp->dev, req->iovm->sgt->sgl,
-				req->iovm->sgt->nents, DMA_TO_DEVICE)) {
-			ret = -ENOMEM;
-			req->iovm = NULL;
+		ret = dma_get_sgtable(isp->dev, &req->table.sgt,
+				      req->table.addr, req->table.dma,
+				      req->config.size);
+		if (ret < 0)
 			goto done;
-		}
 
-		dma_sync_sg_for_cpu(isp->dev, req->iovm->sgt->sgl,
-				    req->iovm->sgt->nents, DMA_TO_DEVICE);
+		dma_sync_sg_for_cpu(isp->dev, req->table.sgt.sgl,
+				    req->table.sgt.nents, DMA_TO_DEVICE);
 
-		table = omap_da_to_va(isp->dev, req->table);
-		if (copy_from_user(table, config->lsc, req->config.size)) {
+		if (copy_from_user(req->table.addr, config->lsc,
+				   req->config.size)) {
 			ret = -EFAULT;
 			goto done;
 		}
 
-		dma_sync_sg_for_device(isp->dev, req->iovm->sgt->sgl,
-				       req->iovm->sgt->nents, DMA_TO_DEVICE);
+		dma_sync_sg_for_device(isp->dev, req->table.sgt.sgl,
+				       req->table.sgt.nents, DMA_TO_DEVICE);
 	}
 
 	spin_lock_irqsave(&ccdc->lsc.req_lock, flags);
@@ -584,7 +577,7 @@ static void ccdc_configure_fpc(struct isp_ccdc_device *ccdc)
 	if (!ccdc->fpc_en)
 		return;
 
-	isp_reg_writel(isp, ccdc->fpc.fpcaddr, OMAP3_ISP_IOMEM_CCDC,
+	isp_reg_writel(isp, ccdc->fpc.dma, OMAP3_ISP_IOMEM_CCDC,
 		       ISPCCDC_FPC_ADDR);
 	/* The FPNUM field must be set before enabling FPC. */
 	isp_reg_writel(isp, (ccdc->fpc.fpnum << ISPCCDC_FPC_FPNUM_SHIFT),
@@ -724,8 +717,9 @@ static int ccdc_config(struct isp_ccdc_device *ccdc,
 	ccdc->shadow_update = 0;
 
 	if (OMAP3ISP_CCDC_FPC & ccdc_struct->update) {
-		u32 table_old = 0;
-		u32 table_new;
+		struct omap3isp_ccdc_fpc fpc;
+		struct ispccdc_fpc fpc_old = { .addr = NULL, };
+		struct ispccdc_fpc fpc_new;
 		u32 size;
 
 		if (ccdc->state != ISP_PIPELINE_STREAM_STOPPED)
@@ -734,35 +728,39 @@ static int ccdc_config(struct isp_ccdc_device *ccdc,
 		ccdc->fpc_en = !!(OMAP3ISP_CCDC_FPC & ccdc_struct->flag);
 
 		if (ccdc->fpc_en) {
-			if (copy_from_user(&ccdc->fpc, ccdc_struct->fpc,
-					   sizeof(ccdc->fpc)))
+			if (copy_from_user(&fpc, ccdc_struct->fpc, sizeof(fpc)))
 				return -EFAULT;
 
+			size = fpc.fpnum * 4;
+
 			/*
-			 * table_new must be 64-bytes aligned, but it's
-			 * already done by omap_iommu_vmalloc().
+			 * The table address must be 64-bytes aligned, which is
+			 * guaranteed by dma_alloc_coherent().
 			 */
-			size = ccdc->fpc.fpnum * 4;
-			table_new = omap_iommu_vmalloc(isp->domain, isp->dev,
-							0, size, IOMMU_FLAG);
-			if (IS_ERR_VALUE(table_new))
+			fpc_new.fpnum = fpc.fpnum;
+			fpc_new.addr = dma_alloc_coherent(isp->dev, size,
+							  &fpc_new.dma,
+							  GFP_KERNEL);
+			if (fpc_new.addr == NULL)
 				return -ENOMEM;
 
-			if (copy_from_user(omap_da_to_va(isp->dev, table_new),
-					   (__force void __user *)
-					   ccdc->fpc.fpcaddr, size)) {
-				omap_iommu_vfree(isp->domain, isp->dev,
-								table_new);
+			if (copy_from_user(fpc_new.addr,
+					   (__force void __user *)fpc.fpcaddr,
+					   size)) {
+				dma_free_coherent(isp->dev, size, fpc_new.addr,
+						  fpc_new.dma);
 				return -EFAULT;
 			}
 
-			table_old = ccdc->fpc.fpcaddr;
-			ccdc->fpc.fpcaddr = table_new;
+			fpc_old = ccdc->fpc;
+			ccdc->fpc = fpc_new;
 		}
 
 		ccdc_configure_fpc(ccdc);
-		if (table_old != 0)
-			omap_iommu_vfree(isp->domain, isp->dev, table_old);
+
+		if (fpc_old.addr != NULL)
+			dma_free_coherent(isp->dev, fpc_old.fpnum * 4,
+					  fpc_old.addr, fpc_old.dma);
 	}
 
 	return ccdc_lsc_config(ccdc, ccdc_struct);
@@ -1523,7 +1521,7 @@ static int ccdc_isr_buffer(struct isp_ccdc_device *ccdc)
 
 	buffer = omap3isp_video_buffer_next(&ccdc->video_out);
 	if (buffer != NULL) {
-		ccdc_set_outaddr(ccdc, buffer->isp_addr);
+		ccdc_set_outaddr(ccdc, buffer->dma);
 		restart = 1;
 	}
 
@@ -1662,7 +1660,7 @@ static int ccdc_video_queue(struct isp_video *video, struct isp_buffer *buffer)
 	if (!(ccdc->output & CCDC_OUTPUT_MEMORY))
 		return -ENODEV;
 
-	ccdc_set_outaddr(ccdc, buffer->isp_addr);
+	ccdc_set_outaddr(ccdc, buffer->dma);
 
 	/* We now have a buffer queued on the output, restart the pipeline
 	 * on the next CCDC interrupt if running in continuous mode (or when
@@ -2580,8 +2578,9 @@ void omap3isp_ccdc_cleanup(struct isp_device *isp)
 	cancel_work_sync(&ccdc->lsc.table_work);
 	ccdc_lsc_free_queue(ccdc, &ccdc->lsc.free_queue);
 
-	if (ccdc->fpc.fpcaddr != 0)
-		omap_iommu_vfree(isp->domain, isp->dev, ccdc->fpc.fpcaddr);
+	if (ccdc->fpc.addr != NULL)
+		dma_free_coherent(isp->dev, ccdc->fpc.fpnum * 4, ccdc->fpc.addr,
+				  ccdc->fpc.dma);
 
 	mutex_destroy(&ccdc->ioctl_lock);
 }
diff --git a/drivers/media/platform/omap3isp/ispccdc.h b/drivers/media/platform/omap3isp/ispccdc.h
index 9d24e4107864..f65061602c71 100644
--- a/drivers/media/platform/omap3isp/ispccdc.h
+++ b/drivers/media/platform/omap3isp/ispccdc.h
@@ -46,6 +46,12 @@ enum ccdc_input_entity {
 
 #define	OMAP3ISP_CCDC_NEVENTS	16
 
+struct ispccdc_fpc {
+	void *addr;
+	dma_addr_t dma;
+	unsigned int fpnum;
+};
+
 enum ispccdc_lsc_state {
 	LSC_STATE_STOPPED = 0,
 	LSC_STATE_STOPPING = 1,
@@ -57,8 +63,12 @@ struct ispccdc_lsc_config_req {
 	struct list_head list;
 	struct omap3isp_ccdc_lsc_config config;
 	unsigned char enable;
-	u32 table;
-	struct iovm_struct *iovm;
+
+	struct {
+		void *addr;
+		dma_addr_t dma;
+		struct sg_table sgt;
+	} table;
 };
 
 /*
@@ -136,7 +146,7 @@ struct isp_ccdc_device {
 		     fpc_en:1;
 	struct omap3isp_ccdc_blcomp blcomp;
 	struct omap3isp_ccdc_bclamp clamp;
-	struct omap3isp_ccdc_fpc fpc;
+	struct ispccdc_fpc fpc;
 	struct ispccdc_lsc lsc;
 	unsigned int update;
 	unsigned int shadow_update;
diff --git a/drivers/media/platform/omap3isp/ispccp2.c b/drivers/media/platform/omap3isp/ispccp2.c
index b30b67d22a58..f3801db9095c 100644
--- a/drivers/media/platform/omap3isp/ispccp2.c
+++ b/drivers/media/platform/omap3isp/ispccp2.c
@@ -549,7 +549,7 @@ static void ccp2_isr_buffer(struct isp_ccp2_device *ccp2)
 
 	buffer = omap3isp_video_buffer_next(&ccp2->video_in);
 	if (buffer != NULL)
-		ccp2_set_inaddr(ccp2, buffer->isp_addr);
+		ccp2_set_inaddr(ccp2, buffer->dma);
 
 	pipe->state |= ISP_PIPELINE_IDLE_INPUT;
 
@@ -940,7 +940,7 @@ static int ccp2_video_queue(struct isp_video *video, struct isp_buffer *buffer)
 {
 	struct isp_ccp2_device *ccp2 = &video->isp->isp_ccp2;
 
-	ccp2_set_inaddr(ccp2, buffer->isp_addr);
+	ccp2_set_inaddr(ccp2, buffer->dma);
 	return 0;
 }
 
diff --git a/drivers/media/platform/omap3isp/ispcsi2.c b/drivers/media/platform/omap3isp/ispcsi2.c
index 620560828a48..5a2e47e58b84 100644
--- a/drivers/media/platform/omap3isp/ispcsi2.c
+++ b/drivers/media/platform/omap3isp/ispcsi2.c
@@ -695,7 +695,7 @@ static void csi2_isr_buffer(struct isp_csi2_device *csi2)
 	if (buffer == NULL)
 		return;
 
-	csi2_set_outaddr(csi2, buffer->isp_addr);
+	csi2_set_outaddr(csi2, buffer->dma);
 	csi2_ctx_enable(isp, csi2, 0, 1);
 }
 
@@ -812,7 +812,7 @@ static int csi2_queue(struct isp_video *video, struct isp_buffer *buffer)
 	struct isp_device *isp = video->isp;
 	struct isp_csi2_device *csi2 = &isp->isp_csi2a;
 
-	csi2_set_outaddr(csi2, buffer->isp_addr);
+	csi2_set_outaddr(csi2, buffer->dma);
 
 	/*
 	 * If streaming was enabled before there was a buffer queued
diff --git a/drivers/media/platform/omap3isp/isph3a_aewb.c b/drivers/media/platform/omap3isp/isph3a_aewb.c
index 75fd82b152ba..d6811ce263eb 100644
--- a/drivers/media/platform/omap3isp/isph3a_aewb.c
+++ b/drivers/media/platform/omap3isp/isph3a_aewb.c
@@ -47,7 +47,7 @@ static void h3a_aewb_setup_regs(struct ispstat *aewb, void *priv)
 	if (aewb->state == ISPSTAT_DISABLED)
 		return;
 
-	isp_reg_writel(aewb->isp, aewb->active_buf->iommu_addr,
+	isp_reg_writel(aewb->isp, aewb->active_buf->dma_addr,
 		       OMAP3_ISP_IOMEM_H3A, ISPH3A_AEWBUFST);
 
 	if (!aewb->update)
diff --git a/drivers/media/platform/omap3isp/isph3a_af.c b/drivers/media/platform/omap3isp/isph3a_af.c
index a0bf5af32438..6fc960cd30f5 100644
--- a/drivers/media/platform/omap3isp/isph3a_af.c
+++ b/drivers/media/platform/omap3isp/isph3a_af.c
@@ -51,7 +51,7 @@ static void h3a_af_setup_regs(struct ispstat *af, void *priv)
 	if (af->state == ISPSTAT_DISABLED)
 		return;
 
-	isp_reg_writel(af->isp, af->active_buf->iommu_addr, OMAP3_ISP_IOMEM_H3A,
+	isp_reg_writel(af->isp, af->active_buf->dma_addr, OMAP3_ISP_IOMEM_H3A,
 		       ISPH3A_AFBUFST);
 
 	if (!af->update)
diff --git a/drivers/media/platform/omap3isp/isppreview.c b/drivers/media/platform/omap3isp/isppreview.c
index 395b2b068c75..720809b07e75 100644
--- a/drivers/media/platform/omap3isp/isppreview.c
+++ b/drivers/media/platform/omap3isp/isppreview.c
@@ -1499,14 +1499,14 @@ static void preview_isr_buffer(struct isp_prev_device *prev)
 	if (prev->input == PREVIEW_INPUT_MEMORY) {
 		buffer = omap3isp_video_buffer_next(&prev->video_in);
 		if (buffer != NULL)
-			preview_set_inaddr(prev, buffer->isp_addr);
+			preview_set_inaddr(prev, buffer->dma);
 		pipe->state |= ISP_PIPELINE_IDLE_INPUT;
 	}
 
 	if (prev->output & PREVIEW_OUTPUT_MEMORY) {
 		buffer = omap3isp_video_buffer_next(&prev->video_out);
 		if (buffer != NULL) {
-			preview_set_outaddr(prev, buffer->isp_addr);
+			preview_set_outaddr(prev, buffer->dma);
 			restart = 1;
 		}
 		pipe->state |= ISP_PIPELINE_IDLE_OUTPUT;
@@ -1577,10 +1577,10 @@ static int preview_video_queue(struct isp_video *video,
 	struct isp_prev_device *prev = &video->isp->isp_prev;
 
 	if (video->type == V4L2_BUF_TYPE_VIDEO_OUTPUT)
-		preview_set_inaddr(prev, buffer->isp_addr);
+		preview_set_inaddr(prev, buffer->dma);
 
 	if (video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
-		preview_set_outaddr(prev, buffer->isp_addr);
+		preview_set_outaddr(prev, buffer->dma);
 
 	return 0;
 }
diff --git a/drivers/media/platform/omap3isp/ispqueue.c b/drivers/media/platform/omap3isp/ispqueue.c
deleted file mode 100644
index a5e65858e799..000000000000
--- a/drivers/media/platform/omap3isp/ispqueue.c
+++ /dev/null
@@ -1,1161 +0,0 @@
-/*
- * ispqueue.c
- *
- * TI OMAP3 ISP - Video buffers queue handling
- *
- * Copyright (C) 2010 Nokia Corporation
- *
- * Contacts: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
- *	     Sakari Ailus <sakari.ailus@iki.fi>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
- * 02110-1301 USA
- */
-
-#include <asm/cacheflush.h>
-#include <linux/dma-mapping.h>
-#include <linux/mm.h>
-#include <linux/pagemap.h>
-#include <linux/poll.h>
-#include <linux/scatterlist.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-
-#include "ispqueue.h"
-
-/* -----------------------------------------------------------------------------
- * Video buffers management
- */
-
-/*
- * isp_video_buffer_cache_sync - Keep the buffers coherent between CPU and ISP
- *
- * The typical operation required here is Cache Invalidation across
- * the (user space) buffer address range. And this _must_ be done
- * at QBUF stage (and *only* at QBUF).
- *
- * We try to use optimal cache invalidation function:
- * - dmac_map_area:
- *    - used when the number of pages are _low_.
- *    - it becomes quite slow as the number of pages increase.
- *       - for 648x492 viewfinder (150 pages) it takes 1.3 ms.
- *       - for 5 Mpix buffer (2491 pages) it takes between 25-50 ms.
- *
- * - flush_cache_all:
- *    - used when the number of pages are _high_.
- *    - time taken in the range of 500-900 us.
- *    - has a higher penalty but, as whole dcache + icache is invalidated
- */
-/*
- * FIXME: dmac_inv_range crashes randomly on the user space buffer
- *        address. Fall back to flush_cache_all for now.
- */
-#define ISP_CACHE_FLUSH_PAGES_MAX       0
-
-static void isp_video_buffer_cache_sync(struct isp_video_buffer *buf)
-{
-	if (buf->skip_cache)
-		return;
-
-	if (buf->vbuf.m.userptr == 0 || buf->npages == 0 ||
-	    buf->npages > ISP_CACHE_FLUSH_PAGES_MAX)
-		flush_cache_all();
-	else {
-		dmac_map_area((void *)buf->vbuf.m.userptr, buf->vbuf.length,
-			      DMA_FROM_DEVICE);
-		outer_inv_range(buf->vbuf.m.userptr,
-				buf->vbuf.m.userptr + buf->vbuf.length);
-	}
-}
-
-/*
- * isp_video_buffer_lock_vma - Prevent VMAs from being unmapped
- *
- * Lock the VMAs underlying the given buffer into memory. This avoids the
- * userspace buffer mapping from being swapped out, making VIPT cache handling
- * easier.
- *
- * Note that the pages will not be freed as the buffers have been locked to
- * memory using by a call to get_user_pages(), but the userspace mapping could
- * still disappear if the VMAs are not locked. This is caused by the memory
- * management code trying to be as lock-less as possible, which results in the
- * userspace mapping manager not finding out that the pages are locked under
- * some conditions.
- */
-static int isp_video_buffer_lock_vma(struct isp_video_buffer *buf, int lock)
-{
-	struct vm_area_struct *vma;
-	unsigned long start;
-	unsigned long end;
-	int ret = 0;
-
-	if (buf->vbuf.memory == V4L2_MEMORY_MMAP)
-		return 0;
-
-	/* We can be called from workqueue context if the current task dies to
-	 * unlock the VMAs. In that case there's no current memory management
-	 * context so unlocking can't be performed, but the VMAs have been or
-	 * are getting destroyed anyway so it doesn't really matter.
-	 */
-	if (!current || !current->mm)
-		return lock ? -EINVAL : 0;
-
-	start = buf->vbuf.m.userptr;
-	end = buf->vbuf.m.userptr + buf->vbuf.length - 1;
-
-	down_write(&current->mm->mmap_sem);
-	spin_lock(&current->mm->page_table_lock);
-
-	do {
-		vma = find_vma(current->mm, start);
-		if (vma == NULL) {
-			ret = -EFAULT;
-			goto out;
-		}
-
-		if (lock)
-			vma->vm_flags |= VM_LOCKED;
-		else
-			vma->vm_flags &= ~VM_LOCKED;
-
-		start = vma->vm_end + 1;
-	} while (vma->vm_end < end);
-
-	if (lock)
-		buf->vm_flags |= VM_LOCKED;
-	else
-		buf->vm_flags &= ~VM_LOCKED;
-
-out:
-	spin_unlock(&current->mm->page_table_lock);
-	up_write(&current->mm->mmap_sem);
-	return ret;
-}
-
-/*
- * isp_video_buffer_sglist_kernel - Build a scatter list for a vmalloc'ed buffer
- *
- * Iterate over the vmalloc'ed area and create a scatter list entry for every
- * page.
- */
-static int isp_video_buffer_sglist_kernel(struct isp_video_buffer *buf)
-{
-	struct scatterlist *sglist;
-	unsigned int npages;
-	unsigned int i;
-	void *addr;
-
-	addr = buf->vaddr;
-	npages = PAGE_ALIGN(buf->vbuf.length) >> PAGE_SHIFT;
-
-	sglist = vmalloc(npages * sizeof(*sglist));
-	if (sglist == NULL)
-		return -ENOMEM;
-
-	sg_init_table(sglist, npages);
-
-	for (i = 0; i < npages; ++i, addr += PAGE_SIZE) {
-		struct page *page = vmalloc_to_page(addr);
-
-		if (page == NULL || PageHighMem(page)) {
-			vfree(sglist);
-			return -EINVAL;
-		}
-
-		sg_set_page(&sglist[i], page, PAGE_SIZE, 0);
-	}
-
-	buf->sglen = npages;
-	buf->sglist = sglist;
-
-	return 0;
-}
-
-/*
- * isp_video_buffer_sglist_user - Build a scatter list for a userspace buffer
- *
- * Walk the buffer pages list and create a 1:1 mapping to a scatter list.
- */
-static int isp_video_buffer_sglist_user(struct isp_video_buffer *buf)
-{
-	struct scatterlist *sglist;
-	unsigned int offset = buf->offset;
-	unsigned int i;
-
-	sglist = vmalloc(buf->npages * sizeof(*sglist));
-	if (sglist == NULL)
-		return -ENOMEM;
-
-	sg_init_table(sglist, buf->npages);
-
-	for (i = 0; i < buf->npages; ++i) {
-		if (PageHighMem(buf->pages[i])) {
-			vfree(sglist);
-			return -EINVAL;
-		}
-
-		sg_set_page(&sglist[i], buf->pages[i], PAGE_SIZE - offset,
-			    offset);
-		offset = 0;
-	}
-
-	buf->sglen = buf->npages;
-	buf->sglist = sglist;
-
-	return 0;
-}
-
-/*
- * isp_video_buffer_sglist_pfnmap - Build a scatter list for a VM_PFNMAP buffer
- *
- * Create a scatter list of physically contiguous pages starting at the buffer
- * memory physical address.
- */
-static int isp_video_buffer_sglist_pfnmap(struct isp_video_buffer *buf)
-{
-	struct scatterlist *sglist;
-	unsigned int offset = buf->offset;
-	unsigned long pfn = buf->paddr >> PAGE_SHIFT;
-	unsigned int i;
-
-	sglist = vmalloc(buf->npages * sizeof(*sglist));
-	if (sglist == NULL)
-		return -ENOMEM;
-
-	sg_init_table(sglist, buf->npages);
-
-	for (i = 0; i < buf->npages; ++i, ++pfn) {
-		sg_set_page(&sglist[i], pfn_to_page(pfn), PAGE_SIZE - offset,
-			    offset);
-		/* PFNMAP buffers will not get DMA-mapped, set the DMA address
-		 * manually.
-		 */
-		sg_dma_address(&sglist[i]) = (pfn << PAGE_SHIFT) + offset;
-		offset = 0;
-	}
-
-	buf->sglen = buf->npages;
-	buf->sglist = sglist;
-
-	return 0;
-}
-
-/*
- * isp_video_buffer_cleanup - Release pages for a userspace VMA.
- *
- * Release pages locked by a call isp_video_buffer_prepare_user and free the
- * pages table.
- */
-static void isp_video_buffer_cleanup(struct isp_video_buffer *buf)
-{
-	enum dma_data_direction direction;
-	unsigned int i;
-
-	if (buf->queue->ops->buffer_cleanup)
-		buf->queue->ops->buffer_cleanup(buf);
-
-	if (!(buf->vm_flags & VM_PFNMAP)) {
-		direction = buf->vbuf.type == V4L2_BUF_TYPE_VIDEO_CAPTURE
-			  ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
-		dma_unmap_sg(buf->queue->dev, buf->sglist, buf->sglen,
-			     direction);
-	}
-
-	vfree(buf->sglist);
-	buf->sglist = NULL;
-	buf->sglen = 0;
-
-	if (buf->pages != NULL) {
-		isp_video_buffer_lock_vma(buf, 0);
-
-		for (i = 0; i < buf->npages; ++i)
-			page_cache_release(buf->pages[i]);
-
-		vfree(buf->pages);
-		buf->pages = NULL;
-	}
-
-	buf->npages = 0;
-	buf->skip_cache = false;
-}
-
-/*
- * isp_video_buffer_prepare_user - Pin userspace VMA pages to memory.
- *
- * This function creates a list of pages for a userspace VMA. The number of
- * pages is first computed based on the buffer size, and pages are then
- * retrieved by a call to get_user_pages.
- *
- * Pages are pinned to memory by get_user_pages, making them available for DMA
- * transfers. However, due to memory management optimization, it seems the
- * get_user_pages doesn't guarantee that the pinned pages will not be written
- * to swap and removed from the userspace mapping(s). When this happens, a page
- * fault can be generated when accessing those unmapped pages.
- *
- * If the fault is triggered by a page table walk caused by VIPT cache
- * management operations, the page fault handler might oops if the MM semaphore
- * is held, as it can't handle kernel page faults in that case. To fix that, a
- * fixup entry needs to be added to the cache management code, or the userspace
- * VMA must be locked to avoid removing pages from the userspace mapping in the
- * first place.
- *
- * If the number of pages retrieved is smaller than the number required by the
- * buffer size, the function returns -EFAULT.
- */
-static int isp_video_buffer_prepare_user(struct isp_video_buffer *buf)
-{
-	unsigned long data;
-	unsigned int first;
-	unsigned int last;
-	int ret;
-
-	data = buf->vbuf.m.userptr;
-	first = (data & PAGE_MASK) >> PAGE_SHIFT;
-	last = ((data + buf->vbuf.length - 1) & PAGE_MASK) >> PAGE_SHIFT;
-
-	buf->offset = data & ~PAGE_MASK;
-	buf->npages = last - first + 1;
-	buf->pages = vmalloc(buf->npages * sizeof(buf->pages[0]));
-	if (buf->pages == NULL)
-		return -ENOMEM;
-
-	down_read(&current->mm->mmap_sem);
-	ret = get_user_pages(current, current->mm, data & PAGE_MASK,
-			     buf->npages,
-			     buf->vbuf.type == V4L2_BUF_TYPE_VIDEO_CAPTURE, 0,
-			     buf->pages, NULL);
-	up_read(&current->mm->mmap_sem);
-
-	if (ret != buf->npages) {
-		buf->npages = ret < 0 ? 0 : ret;
-		isp_video_buffer_cleanup(buf);
-		return -EFAULT;
-	}
-
-	ret = isp_video_buffer_lock_vma(buf, 1);
-	if (ret < 0)
-		isp_video_buffer_cleanup(buf);
-
-	return ret;
-}
-
-/*
- * isp_video_buffer_prepare_pfnmap - Validate a VM_PFNMAP userspace buffer
- *
- * Userspace VM_PFNMAP buffers are supported only if they are contiguous in
- * memory and if they span a single VMA.
- *
- * Return 0 if the buffer is valid, or -EFAULT otherwise.
- */
-static int isp_video_buffer_prepare_pfnmap(struct isp_video_buffer *buf)
-{
-	struct vm_area_struct *vma;
-	unsigned long prev_pfn;
-	unsigned long this_pfn;
-	unsigned long start;
-	unsigned long end;
-	dma_addr_t pa = 0;
-	int ret = -EFAULT;
-
-	start = buf->vbuf.m.userptr;
-	end = buf->vbuf.m.userptr + buf->vbuf.length - 1;
-
-	buf->offset = start & ~PAGE_MASK;
-	buf->npages = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1;
-	buf->pages = NULL;
-
-	down_read(&current->mm->mmap_sem);
-	vma = find_vma(current->mm, start);
-	if (vma == NULL || vma->vm_end < end)
-		goto done;
-
-	for (prev_pfn = 0; start <= end; start += PAGE_SIZE) {
-		ret = follow_pfn(vma, start, &this_pfn);
-		if (ret)
-			goto done;
-
-		if (prev_pfn == 0)
-			pa = this_pfn << PAGE_SHIFT;
-		else if (this_pfn != prev_pfn + 1) {
-			ret = -EFAULT;
-			goto done;
-		}
-
-		prev_pfn = this_pfn;
-	}
-
-	buf->paddr = pa + buf->offset;
-	ret = 0;
-
-done:
-	up_read(&current->mm->mmap_sem);
-	return ret;
-}
-
-/*
- * isp_video_buffer_prepare_vm_flags - Get VMA flags for a userspace address
- *
- * This function locates the VMAs for the buffer's userspace address and checks
- * that their flags match. The only flag that we need to care for at the moment
- * is VM_PFNMAP.
- *
- * The buffer vm_flags field is set to the first VMA flags.
- *
- * Return -EFAULT if no VMA can be found for part of the buffer, or if the VMAs
- * have incompatible flags.
- */
-static int isp_video_buffer_prepare_vm_flags(struct isp_video_buffer *buf)
-{
-	struct vm_area_struct *vma;
-	pgprot_t uninitialized_var(vm_page_prot);
-	unsigned long start;
-	unsigned long end;
-	int ret = -EFAULT;
-
-	start = buf->vbuf.m.userptr;
-	end = buf->vbuf.m.userptr + buf->vbuf.length - 1;
-
-	down_read(&current->mm->mmap_sem);
-
-	do {
-		vma = find_vma(current->mm, start);
-		if (vma == NULL)
-			goto done;
-
-		if (start == buf->vbuf.m.userptr) {
-			buf->vm_flags = vma->vm_flags;
-			vm_page_prot = vma->vm_page_prot;
-		}
-
-		if ((buf->vm_flags ^ vma->vm_flags) & VM_PFNMAP)
-			goto done;
-
-		if (vm_page_prot != vma->vm_page_prot)
-			goto done;
-
-		start = vma->vm_end + 1;
-	} while (vma->vm_end < end);
-
-	/* Skip cache management to enhance performances for non-cached or
-	 * write-combining buffers.
-	 */
-	if (vm_page_prot == pgprot_noncached(vm_page_prot) ||
-	    vm_page_prot == pgprot_writecombine(vm_page_prot))
-		buf->skip_cache = true;
-
-	ret = 0;
-
-done:
-	up_read(&current->mm->mmap_sem);
-	return ret;
-}
-
-/*
- * isp_video_buffer_prepare - Make a buffer ready for operation
- *
- * Preparing a buffer involves:
- *
- * - validating VMAs (userspace buffers only)
- * - locking pages and VMAs into memory (userspace buffers only)
- * - building page and scatter-gather lists
- * - mapping buffers for DMA operation
- * - performing driver-specific preparation
- *
- * The function must be called in userspace context with a valid mm context
- * (this excludes cleanup paths such as sys_close when the userspace process
- * segfaults).
- */
-static int isp_video_buffer_prepare(struct isp_video_buffer *buf)
-{
-	enum dma_data_direction direction;
-	int ret;
-
-	switch (buf->vbuf.memory) {
-	case V4L2_MEMORY_MMAP:
-		ret = isp_video_buffer_sglist_kernel(buf);
-		break;
-
-	case V4L2_MEMORY_USERPTR:
-		ret = isp_video_buffer_prepare_vm_flags(buf);
-		if (ret < 0)
-			return ret;
-
-		if (buf->vm_flags & VM_PFNMAP) {
-			ret = isp_video_buffer_prepare_pfnmap(buf);
-			if (ret < 0)
-				return ret;
-
-			ret = isp_video_buffer_sglist_pfnmap(buf);
-		} else {
-			ret = isp_video_buffer_prepare_user(buf);
-			if (ret < 0)
-				return ret;
-
-			ret = isp_video_buffer_sglist_user(buf);
-		}
-		break;
-
-	default:
-		return -EINVAL;
-	}
-
-	if (ret < 0)
-		goto done;
-
-	if (!(buf->vm_flags & VM_PFNMAP)) {
-		direction = buf->vbuf.type == V4L2_BUF_TYPE_VIDEO_CAPTURE
-			  ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
-		ret = dma_map_sg(buf->queue->dev, buf->sglist, buf->sglen,
-				 direction);
-		if (ret != buf->sglen) {
-			ret = -EFAULT;
-			goto done;
-		}
-	}
-
-	if (buf->queue->ops->buffer_prepare)
-		ret = buf->queue->ops->buffer_prepare(buf);
-
-done:
-	if (ret < 0) {
-		isp_video_buffer_cleanup(buf);
-		return ret;
-	}
-
-	return ret;
-}
-
-/*
- * isp_video_queue_query - Query the status of a given buffer
- *
- * Locking: must be called with the queue lock held.
- */
-static void isp_video_buffer_query(struct isp_video_buffer *buf,
-				   struct v4l2_buffer *vbuf)
-{
-	memcpy(vbuf, &buf->vbuf, sizeof(*vbuf));
-
-	if (buf->vma_use_count)
-		vbuf->flags |= V4L2_BUF_FLAG_MAPPED;
-
-	switch (buf->state) {
-	case ISP_BUF_STATE_ERROR:
-		vbuf->flags |= V4L2_BUF_FLAG_ERROR;
-		/* Fallthrough */
-	case ISP_BUF_STATE_DONE:
-		vbuf->flags |= V4L2_BUF_FLAG_DONE;
-		break;
-	case ISP_BUF_STATE_QUEUED:
-	case ISP_BUF_STATE_ACTIVE:
-		vbuf->flags |= V4L2_BUF_FLAG_QUEUED;
-		break;
-	case ISP_BUF_STATE_IDLE:
-	default:
-		break;
-	}
-}
-
-/*
- * isp_video_buffer_wait - Wait for a buffer to be ready
- *
- * In non-blocking mode, return immediately with 0 if the buffer is ready or
- * -EAGAIN if the buffer is in the QUEUED or ACTIVE state.
- *
- * In blocking mode, wait (interruptibly but with no timeout) on the buffer wait
- * queue using the same condition.
- */
-static int isp_video_buffer_wait(struct isp_video_buffer *buf, int nonblocking)
-{
-	if (nonblocking) {
-		return (buf->state != ISP_BUF_STATE_QUEUED &&
-			buf->state != ISP_BUF_STATE_ACTIVE)
-			? 0 : -EAGAIN;
-	}
-
-	return wait_event_interruptible(buf->wait,
-		buf->state != ISP_BUF_STATE_QUEUED &&
-		buf->state != ISP_BUF_STATE_ACTIVE);
-}
-
-/* -----------------------------------------------------------------------------
- * Queue management
- */
-
-/*
- * isp_video_queue_free - Free video buffers memory
- *
- * Buffers can only be freed if the queue isn't streaming and if no buffer is
- * mapped to userspace. Return -EBUSY if those conditions aren't satisfied.
- *
- * This function must be called with the queue lock held.
- */
-static int isp_video_queue_free(struct isp_video_queue *queue)
-{
-	unsigned int i;
-
-	if (queue->streaming)
-		return -EBUSY;
-
-	for (i = 0; i < queue->count; ++i) {
-		if (queue->buffers[i]->vma_use_count != 0)
-			return -EBUSY;
-	}
-
-	for (i = 0; i < queue->count; ++i) {
-		struct isp_video_buffer *buf = queue->buffers[i];
-
-		isp_video_buffer_cleanup(buf);
-
-		vfree(buf->vaddr);
-		buf->vaddr = NULL;
-
-		kfree(buf);
-		queue->buffers[i] = NULL;
-	}
-
-	INIT_LIST_HEAD(&queue->queue);
-	queue->count = 0;
-	return 0;
-}
-
-/*
- * isp_video_queue_alloc - Allocate video buffers memory
- *
- * This function must be called with the queue lock held.
- */
-static int isp_video_queue_alloc(struct isp_video_queue *queue,
-				 unsigned int nbuffers,
-				 unsigned int size, enum v4l2_memory memory)
-{
-	struct isp_video_buffer *buf;
-	unsigned int i;
-	void *mem;
-	int ret;
-
-	/* Start by freeing the buffers. */
-	ret = isp_video_queue_free(queue);
-	if (ret < 0)
-		return ret;
-
-	/* Bail out if no buffers should be allocated. */
-	if (nbuffers == 0)
-		return 0;
-
-	/* Initialize the allocated buffers. */
-	for (i = 0; i < nbuffers; ++i) {
-		buf = kzalloc(queue->bufsize, GFP_KERNEL);
-		if (buf == NULL)
-			break;
-
-		if (memory == V4L2_MEMORY_MMAP) {
-			/* Allocate video buffers memory for mmap mode. Align
-			 * the size to the page size.
-			 */
-			mem = vmalloc_32_user(PAGE_ALIGN(size));
-			if (mem == NULL) {
-				kfree(buf);
-				break;
-			}
-
-			buf->vbuf.m.offset = i * PAGE_ALIGN(size);
-			buf->vaddr = mem;
-		}
-
-		buf->vbuf.index = i;
-		buf->vbuf.length = size;
-		buf->vbuf.type = queue->type;
-		buf->vbuf.flags = V4L2_BUF_FLAG_TIMESTAMP_MONOTONIC;
-		buf->vbuf.field = V4L2_FIELD_NONE;
-		buf->vbuf.memory = memory;
-
-		buf->queue = queue;
-		init_waitqueue_head(&buf->wait);
-
-		queue->buffers[i] = buf;
-	}
-
-	if (i == 0)
-		return -ENOMEM;
-
-	queue->count = i;
-	return nbuffers;
-}
-
-/**
- * omap3isp_video_queue_cleanup - Clean up the video buffers queue
- * @queue: Video buffers queue
- *
- * Free all allocated resources and clean up the video buffers queue. The queue
- * must not be busy (no ongoing video stream) and buffers must have been
- * unmapped.
- *
- * Return 0 on success or -EBUSY if the queue is busy or buffers haven't been
- * unmapped.
- */
-int omap3isp_video_queue_cleanup(struct isp_video_queue *queue)
-{
-	return isp_video_queue_free(queue);
-}
-
-/**
- * omap3isp_video_queue_init - Initialize the video buffers queue
- * @queue: Video buffers queue
- * @type: V4L2 buffer type (capture or output)
- * @ops: Driver-specific queue operations
- * @dev: Device used for DMA operations
- * @bufsize: Size of the driver-specific buffer structure
- *
- * Initialize the video buffers queue with the supplied parameters.
- *
- * The queue type must be one of V4L2_BUF_TYPE_VIDEO_CAPTURE or
- * V4L2_BUF_TYPE_VIDEO_OUTPUT. Other buffer types are not supported yet.
- *
- * Buffer objects will be allocated using the given buffer size to allow room
- * for driver-specific fields. Driver-specific buffer structures must start
- * with a struct isp_video_buffer field. Drivers with no driver-specific buffer
- * structure must pass the size of the isp_video_buffer structure in the bufsize
- * parameter.
- *
- * Return 0 on success.
- */
-int omap3isp_video_queue_init(struct isp_video_queue *queue,
-			      enum v4l2_buf_type type,
-			      const struct isp_video_queue_operations *ops,
-			      struct device *dev, unsigned int bufsize)
-{
-	INIT_LIST_HEAD(&queue->queue);
-	mutex_init(&queue->lock);
-	spin_lock_init(&queue->irqlock);
-
-	queue->type = type;
-	queue->ops = ops;
-	queue->dev = dev;
-	queue->bufsize = bufsize;
-
-	return 0;
-}
-
-/* -----------------------------------------------------------------------------
- * V4L2 operations
- */
-
-/**
- * omap3isp_video_queue_reqbufs - Allocate video buffers memory
- *
- * This function is intended to be used as a VIDIOC_REQBUFS ioctl handler. It
- * allocated video buffer objects and, for MMAP buffers, buffer memory.
- *
- * If the number of buffers is 0, all buffers are freed and the function returns
- * without performing any allocation.
- *
- * If the number of buffers is not 0, currently allocated buffers (if any) are
- * freed and the requested number of buffers are allocated. Depending on
- * driver-specific requirements and on memory availability, a number of buffer
- * smaller or bigger than requested can be allocated. This isn't considered as
- * an error.
- *
- * Return 0 on success or one of the following error codes:
- *
- * -EINVAL if the buffer type or index are invalid
- * -EBUSY if the queue is busy (streaming or buffers mapped)
- * -ENOMEM if the buffers can't be allocated due to an out-of-memory condition
- */
-int omap3isp_video_queue_reqbufs(struct isp_video_queue *queue,
-				 struct v4l2_requestbuffers *rb)
-{
-	unsigned int nbuffers = rb->count;
-	unsigned int size;
-	int ret;
-
-	if (rb->type != queue->type)
-		return -EINVAL;
-
-	queue->ops->queue_prepare(queue, &nbuffers, &size);
-	if (size == 0)
-		return -EINVAL;
-
-	nbuffers = min_t(unsigned int, nbuffers, ISP_VIDEO_MAX_BUFFERS);
-
-	mutex_lock(&queue->lock);
-
-	ret = isp_video_queue_alloc(queue, nbuffers, size, rb->memory);
-	if (ret < 0)
-		goto done;
-
-	rb->count = ret;
-	ret = 0;
-
-done:
-	mutex_unlock(&queue->lock);
-	return ret;
-}
-
-/**
- * omap3isp_video_queue_querybuf - Query the status of a buffer in a queue
- *
- * This function is intended to be used as a VIDIOC_QUERYBUF ioctl handler. It
- * returns the status of a given video buffer.
- *
- * Return 0 on success or -EINVAL if the buffer type or index are invalid.
- */
-int omap3isp_video_queue_querybuf(struct isp_video_queue *queue,
-				  struct v4l2_buffer *vbuf)
-{
-	struct isp_video_buffer *buf;
-	int ret = 0;
-
-	if (vbuf->type != queue->type)
-		return -EINVAL;
-
-	mutex_lock(&queue->lock);
-
-	if (vbuf->index >= queue->count) {
-		ret = -EINVAL;
-		goto done;
-	}
-
-	buf = queue->buffers[vbuf->index];
-	isp_video_buffer_query(buf, vbuf);
-
-done:
-	mutex_unlock(&queue->lock);
-	return ret;
-}
-
-/**
- * omap3isp_video_queue_qbuf - Queue a buffer
- *
- * This function is intended to be used as a VIDIOC_QBUF ioctl handler.
- *
- * The v4l2_buffer structure passed from userspace is first sanity tested. If
- * sane, the buffer is then processed and added to the main queue and, if the
- * queue is streaming, to the IRQ queue.
- *
- * Before being enqueued, USERPTR buffers are checked for address changes. If
- * the buffer has a different userspace address, the old memory area is unlocked
- * and the new memory area is locked.
- */
-int omap3isp_video_queue_qbuf(struct isp_video_queue *queue,
-			      struct v4l2_buffer *vbuf)
-{
-	struct isp_video_buffer *buf;
-	unsigned long flags;
-	int ret = -EINVAL;
-
-	if (vbuf->type != queue->type)
-		goto done;
-
-	mutex_lock(&queue->lock);
-
-	if (vbuf->index >= queue->count)
-		goto done;
-
-	buf = queue->buffers[vbuf->index];
-
-	if (vbuf->memory != buf->vbuf.memory)
-		goto done;
-
-	if (buf->state != ISP_BUF_STATE_IDLE)
-		goto done;
-
-	if (vbuf->memory == V4L2_MEMORY_USERPTR &&
-	    vbuf->length < buf->vbuf.length)
-		goto done;
-
-	if (vbuf->memory == V4L2_MEMORY_USERPTR &&
-	    vbuf->m.userptr != buf->vbuf.m.userptr) {
-		isp_video_buffer_cleanup(buf);
-		buf->vbuf.m.userptr = vbuf->m.userptr;
-		buf->prepared = 0;
-	}
-
-	if (!buf->prepared) {
-		ret = isp_video_buffer_prepare(buf);
-		if (ret < 0)
-			goto done;
-		buf->prepared = 1;
-	}
-
-	isp_video_buffer_cache_sync(buf);
-
-	buf->state = ISP_BUF_STATE_QUEUED;
-	list_add_tail(&buf->stream, &queue->queue);
-
-	if (queue->streaming) {
-		spin_lock_irqsave(&queue->irqlock, flags);
-		queue->ops->buffer_queue(buf);
-		spin_unlock_irqrestore(&queue->irqlock, flags);
-	}
-
-	ret = 0;
-
-done:
-	mutex_unlock(&queue->lock);
-	return ret;
-}
-
-/**
- * omap3isp_video_queue_dqbuf - Dequeue a buffer
- *
- * This function is intended to be used as a VIDIOC_DQBUF ioctl handler.
- *
- * Wait until a buffer is ready to be dequeued, remove it from the queue and
- * copy its information to the v4l2_buffer structure.
- *
- * If the nonblocking argument is not zero and no buffer is ready, return
- * -EAGAIN immediately instead of waiting.
- *
- * If no buffer has been enqueued, or if the requested buffer type doesn't match
- * the queue type, return -EINVAL.
- */
-int omap3isp_video_queue_dqbuf(struct isp_video_queue *queue,
-			       struct v4l2_buffer *vbuf, int nonblocking)
-{
-	struct isp_video_buffer *buf;
-	int ret;
-
-	if (vbuf->type != queue->type)
-		return -EINVAL;
-
-	mutex_lock(&queue->lock);
-
-	if (list_empty(&queue->queue)) {
-		ret = -EINVAL;
-		goto done;
-	}
-
-	buf = list_first_entry(&queue->queue, struct isp_video_buffer, stream);
-	ret = isp_video_buffer_wait(buf, nonblocking);
-	if (ret < 0)
-		goto done;
-
-	list_del(&buf->stream);
-
-	isp_video_buffer_query(buf, vbuf);
-	buf->state = ISP_BUF_STATE_IDLE;
-	vbuf->flags &= ~V4L2_BUF_FLAG_QUEUED;
-
-done:
-	mutex_unlock(&queue->lock);
-	return ret;
-}
-
-/**
- * omap3isp_video_queue_streamon - Start streaming
- *
- * This function is intended to be used as a VIDIOC_STREAMON ioctl handler. It
- * starts streaming on the queue and calls the buffer_queue operation for all
- * queued buffers.
- *
- * Return 0 on success.
- */
-int omap3isp_video_queue_streamon(struct isp_video_queue *queue)
-{
-	struct isp_video_buffer *buf;
-	unsigned long flags;
-
-	mutex_lock(&queue->lock);
-
-	if (queue->streaming)
-		goto done;
-
-	queue->streaming = 1;
-
-	spin_lock_irqsave(&queue->irqlock, flags);
-	list_for_each_entry(buf, &queue->queue, stream)
-		queue->ops->buffer_queue(buf);
-	spin_unlock_irqrestore(&queue->irqlock, flags);
-
-done:
-	mutex_unlock(&queue->lock);
-	return 0;
-}
-
-/**
- * omap3isp_video_queue_streamoff - Stop streaming
- *
- * This function is intended to be used as a VIDIOC_STREAMOFF ioctl handler. It
- * stops streaming on the queue and wakes up all the buffers.
- *
- * Drivers must stop the hardware and synchronize with interrupt handlers and/or
- * delayed works before calling this function to make sure no buffer will be
- * touched by the driver and/or hardware.
- */
-void omap3isp_video_queue_streamoff(struct isp_video_queue *queue)
-{
-	struct isp_video_buffer *buf;
-	unsigned long flags;
-	unsigned int i;
-
-	mutex_lock(&queue->lock);
-
-	if (!queue->streaming)
-		goto done;
-
-	queue->streaming = 0;
-
-	spin_lock_irqsave(&queue->irqlock, flags);
-	for (i = 0; i < queue->count; ++i) {
-		buf = queue->buffers[i];
-
-		if (buf->state == ISP_BUF_STATE_ACTIVE)
-			wake_up(&buf->wait);
-
-		buf->state = ISP_BUF_STATE_IDLE;
-	}
-	spin_unlock_irqrestore(&queue->irqlock, flags);
-
-	INIT_LIST_HEAD(&queue->queue);
-
-done:
-	mutex_unlock(&queue->lock);
-}
-
-/**
- * omap3isp_video_queue_discard_done - Discard all buffers marked as DONE
- *
- * This function is intended to be used with suspend/resume operations. It
- * discards all 'done' buffers as they would be too old to be requested after
- * resume.
- *
- * Drivers must stop the hardware and synchronize with interrupt handlers and/or
- * delayed works before calling this function to make sure no buffer will be
- * touched by the driver and/or hardware.
- */
-void omap3isp_video_queue_discard_done(struct isp_video_queue *queue)
-{
-	struct isp_video_buffer *buf;
-	unsigned int i;
-
-	mutex_lock(&queue->lock);
-
-	if (!queue->streaming)
-		goto done;
-
-	for (i = 0; i < queue->count; ++i) {
-		buf = queue->buffers[i];
-
-		if (buf->state == ISP_BUF_STATE_DONE)
-			buf->state = ISP_BUF_STATE_ERROR;
-	}
-
-done:
-	mutex_unlock(&queue->lock);
-}
-
-static void isp_video_queue_vm_open(struct vm_area_struct *vma)
-{
-	struct isp_video_buffer *buf = vma->vm_private_data;
-
-	buf->vma_use_count++;
-}
-
-static void isp_video_queue_vm_close(struct vm_area_struct *vma)
-{
-	struct isp_video_buffer *buf = vma->vm_private_data;
-
-	buf->vma_use_count--;
-}
-
-static const struct vm_operations_struct isp_video_queue_vm_ops = {
-	.open = isp_video_queue_vm_open,
-	.close = isp_video_queue_vm_close,
-};
-
-/**
- * omap3isp_video_queue_mmap - Map buffers to userspace
- *
- * This function is intended to be used as an mmap() file operation handler. It
- * maps a buffer to userspace based on the VMA offset.
- *
- * Only buffers of memory type MMAP are supported.
- */
-int omap3isp_video_queue_mmap(struct isp_video_queue *queue,
-			 struct vm_area_struct *vma)
-{
-	struct isp_video_buffer *uninitialized_var(buf);
-	unsigned long size;
-	unsigned int i;
-	int ret = 0;
-
-	mutex_lock(&queue->lock);
-
-	for (i = 0; i < queue->count; ++i) {
-		buf = queue->buffers[i];
-		if ((buf->vbuf.m.offset >> PAGE_SHIFT) == vma->vm_pgoff)
-			break;
-	}
-
-	if (i == queue->count) {
-		ret = -EINVAL;
-		goto done;
-	}
-
-	size = vma->vm_end - vma->vm_start;
-
-	if (buf->vbuf.memory != V4L2_MEMORY_MMAP ||
-	    size != PAGE_ALIGN(buf->vbuf.length)) {
-		ret = -EINVAL;
-		goto done;
-	}
-
-	ret = remap_vmalloc_range(vma, buf->vaddr, 0);
-	if (ret < 0)
-		goto done;
-
-	vma->vm_ops = &isp_video_queue_vm_ops;
-	vma->vm_private_data = buf;
-	isp_video_queue_vm_open(vma);
-
-done:
-	mutex_unlock(&queue->lock);
-	return ret;
-}
-
-/**
- * omap3isp_video_queue_poll - Poll video queue state
- *
- * This function is intended to be used as a poll() file operation handler. It
- * polls the state of the video buffer at the front of the queue and returns an
- * events mask.
- *
- * If no buffer is present at the front of the queue, POLLERR is returned.
- */
-unsigned int omap3isp_video_queue_poll(struct isp_video_queue *queue,
-				       struct file *file, poll_table *wait)
-{
-	struct isp_video_buffer *buf;
-	unsigned int mask = 0;
-
-	mutex_lock(&queue->lock);
-	if (list_empty(&queue->queue)) {
-		mask |= POLLERR;
-		goto done;
-	}
-	buf = list_first_entry(&queue->queue, struct isp_video_buffer, stream);
-
-	poll_wait(file, &buf->wait, wait);
-	if (buf->state == ISP_BUF_STATE_DONE ||
-	    buf->state == ISP_BUF_STATE_ERROR) {
-		if (queue->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
-			mask |= POLLIN | POLLRDNORM;
-		else
-			mask |= POLLOUT | POLLWRNORM;
-	}
-
-done:
-	mutex_unlock(&queue->lock);
-	return mask;
-}
diff --git a/drivers/media/platform/omap3isp/ispqueue.h b/drivers/media/platform/omap3isp/ispqueue.h
deleted file mode 100644
index 3e048ad65647..000000000000
--- a/drivers/media/platform/omap3isp/ispqueue.h
+++ /dev/null
@@ -1,188 +0,0 @@
-/*
- * ispqueue.h
- *
- * TI OMAP3 ISP - Video buffers queue handling
- *
- * Copyright (C) 2010 Nokia Corporation
- *
- * Contacts: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
- *	     Sakari Ailus <sakari.ailus@iki.fi>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
- * 02110-1301 USA
- */
-
-#ifndef OMAP3_ISP_QUEUE_H
-#define OMAP3_ISP_QUEUE_H
-
-#include <linux/kernel.h>
-#include <linux/list.h>
-#include <linux/mm_types.h>
-#include <linux/mutex.h>
-#include <linux/videodev2.h>
-#include <linux/wait.h>
-
-struct isp_video_queue;
-struct page;
-struct scatterlist;
-
-#define ISP_VIDEO_MAX_BUFFERS		16
-
-/**
- * enum isp_video_buffer_state - ISP video buffer state
- * @ISP_BUF_STATE_IDLE:	The buffer is under userspace control (dequeued
- *	or not queued yet).
- * @ISP_BUF_STATE_QUEUED: The buffer has been queued but isn't used by the
- *	device yet.
- * @ISP_BUF_STATE_ACTIVE: The buffer is in use for an active video transfer.
- * @ISP_BUF_STATE_ERROR: The device is done with the buffer and an error
- *	occurred. For capture device the buffer likely contains corrupted data or
- *	no data at all.
- * @ISP_BUF_STATE_DONE: The device is done with the buffer and no error occurred.
- *	For capture devices the buffer contains valid data.
- */
-enum isp_video_buffer_state {
-	ISP_BUF_STATE_IDLE,
-	ISP_BUF_STATE_QUEUED,
-	ISP_BUF_STATE_ACTIVE,
-	ISP_BUF_STATE_ERROR,
-	ISP_BUF_STATE_DONE,
-};
-
-/**
- * struct isp_video_buffer - ISP video buffer
- * @vma_use_count: Number of times the buffer is mmap'ed to userspace
- * @stream: List head for insertion into main queue
- * @queue: ISP buffers queue this buffer belongs to
- * @prepared: Whether the buffer has been prepared
- * @skip_cache: Whether to skip cache management operations for this buffer
- * @vaddr: Memory virtual address (for kernel buffers)
- * @vm_flags: Buffer VMA flags (for userspace buffers)
- * @offset: Offset inside the first page (for userspace buffers)
- * @npages: Number of pages (for userspace buffers)
- * @pages: Pages table (for userspace non-VM_PFNMAP buffers)
- * @paddr: Memory physical address (for userspace VM_PFNMAP buffers)
- * @sglen: Number of elements in the scatter list (for non-VM_PFNMAP buffers)
- * @sglist: Scatter list (for non-VM_PFNMAP buffers)
- * @vbuf: V4L2 buffer
- * @irqlist: List head for insertion into IRQ queue
- * @state: Current buffer state
- * @wait: Wait queue to signal buffer completion
- */
-struct isp_video_buffer {
-	unsigned long vma_use_count;
-	struct list_head stream;
-	struct isp_video_queue *queue;
-	unsigned int prepared:1;
-	bool skip_cache;
-
-	/* For kernel buffers. */
-	void *vaddr;
-
-	/* For userspace buffers. */
-	vm_flags_t vm_flags;
-	unsigned long offset;
-	unsigned int npages;
-	struct page **pages;
-	dma_addr_t paddr;
-
-	/* For all buffers except VM_PFNMAP. */
-	unsigned int sglen;
-	struct scatterlist *sglist;
-
-	/* Touched by the interrupt handler. */
-	struct v4l2_buffer vbuf;
-	struct list_head irqlist;
-	enum isp_video_buffer_state state;
-	wait_queue_head_t wait;
-};
-
-#define to_isp_video_buffer(vb)	container_of(vb, struct isp_video_buffer, vb)
-
-/**
- * struct isp_video_queue_operations - Driver-specific operations
- * @queue_prepare: Called before allocating buffers. Drivers should clamp the
- *	number of buffers according to their requirements, and must return the
- *	buffer size in bytes.
- * @buffer_prepare: Called the first time a buffer is queued, or after changing
- *	the userspace memory address for a USERPTR buffer, with the queue lock
- *	held. Drivers should perform device-specific buffer preparation (such as
- *	mapping the buffer memory in an IOMMU). This operation is optional.
- * @buffer_queue: Called when a buffer is being added to the queue with the
- *	queue irqlock spinlock held.
- * @buffer_cleanup: Called before freeing buffers, or before changing the
- *	userspace memory address for a USERPTR buffer, with the queue lock held.
- *	Drivers must perform cleanup operations required to undo the
- *	buffer_prepare call. This operation is optional.
- */
-struct isp_video_queue_operations {
-	void (*queue_prepare)(struct isp_video_queue *queue,
-			      unsigned int *nbuffers, unsigned int *size);
-	int  (*buffer_prepare)(struct isp_video_buffer *buf);
-	void (*buffer_queue)(struct isp_video_buffer *buf);
-	void (*buffer_cleanup)(struct isp_video_buffer *buf);
-};
-
-/**
- * struct isp_video_queue - ISP video buffers queue
- * @type: Type of video buffers handled by this queue
- * @ops: Queue operations
- * @dev: Device used for DMA operations
- * @bufsize: Size of a driver-specific buffer object
- * @count: Number of currently allocated buffers
- * @buffers: ISP video buffers
- * @lock: Mutex to protect access to the buffers, main queue and state
- * @irqlock: Spinlock to protect access to the IRQ queue
- * @streaming: Queue state, indicates whether the queue is streaming
- * @queue: List of all queued buffers
- */
-struct isp_video_queue {
-	enum v4l2_buf_type type;
-	const struct isp_video_queue_operations *ops;
-	struct device *dev;
-	unsigned int bufsize;
-
-	unsigned int count;
-	struct isp_video_buffer *buffers[ISP_VIDEO_MAX_BUFFERS];
-	struct mutex lock;
-	spinlock_t irqlock;
-
-	unsigned int streaming:1;
-
-	struct list_head queue;
-};
-
-int omap3isp_video_queue_cleanup(struct isp_video_queue *queue);
-int omap3isp_video_queue_init(struct isp_video_queue *queue,
-			      enum v4l2_buf_type type,
-			      const struct isp_video_queue_operations *ops,
-			      struct device *dev, unsigned int bufsize);
-
-int omap3isp_video_queue_reqbufs(struct isp_video_queue *queue,
-				 struct v4l2_requestbuffers *rb);
-int omap3isp_video_queue_querybuf(struct isp_video_queue *queue,
-				  struct v4l2_buffer *vbuf);
-int omap3isp_video_queue_qbuf(struct isp_video_queue *queue,
-			      struct v4l2_buffer *vbuf);
-int omap3isp_video_queue_dqbuf(struct isp_video_queue *queue,
-			       struct v4l2_buffer *vbuf, int nonblocking);
-int omap3isp_video_queue_streamon(struct isp_video_queue *queue);
-void omap3isp_video_queue_streamoff(struct isp_video_queue *queue);
-void omap3isp_video_queue_discard_done(struct isp_video_queue *queue);
-int omap3isp_video_queue_mmap(struct isp_video_queue *queue,
-			      struct vm_area_struct *vma);
-unsigned int omap3isp_video_queue_poll(struct isp_video_queue *queue,
-				       struct file *file, poll_table *wait);
-
-#endif /* OMAP3_ISP_QUEUE_H */
diff --git a/drivers/media/platform/omap3isp/ispresizer.c b/drivers/media/platform/omap3isp/ispresizer.c
index 86369df81d74..6f077c2377db 100644
--- a/drivers/media/platform/omap3isp/ispresizer.c
+++ b/drivers/media/platform/omap3isp/ispresizer.c
@@ -1040,7 +1040,7 @@ static void resizer_isr_buffer(struct isp_res_device *res)
 	 */
 	buffer = omap3isp_video_buffer_next(&res->video_out);
 	if (buffer != NULL) {
-		resizer_set_outaddr(res, buffer->isp_addr);
+		resizer_set_outaddr(res, buffer->dma);
 		restart = 1;
 	}
 
@@ -1049,7 +1049,7 @@ static void resizer_isr_buffer(struct isp_res_device *res)
 	if (res->input == RESIZER_INPUT_MEMORY) {
 		buffer = omap3isp_video_buffer_next(&res->video_in);
 		if (buffer != NULL)
-			resizer_set_inaddr(res, buffer->isp_addr);
+			resizer_set_inaddr(res, buffer->dma);
 		pipe->state |= ISP_PIPELINE_IDLE_INPUT;
 	}
 
@@ -1101,7 +1101,7 @@ static int resizer_video_queue(struct isp_video *video,
 	struct isp_res_device *res = &video->isp->isp_res;
 
 	if (video->type == V4L2_BUF_TYPE_VIDEO_OUTPUT)
-		resizer_set_inaddr(res, buffer->isp_addr);
+		resizer_set_inaddr(res, buffer->dma);
 
 	/*
 	 * We now have a buffer queued on the output. Despite what the
@@ -1116,7 +1116,7 @@ static int resizer_video_queue(struct isp_video *video,
 	 * continuous mode or when starting the stream.
 	 */
 	if (video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
-		resizer_set_outaddr(res, buffer->isp_addr);
+		resizer_set_outaddr(res, buffer->dma);
 
 	return 0;
 }
diff --git a/drivers/media/platform/omap3isp/ispstat.c b/drivers/media/platform/omap3isp/ispstat.c
index 5707f85c4cc4..e6cbc1eaf4ca 100644
--- a/drivers/media/platform/omap3isp/ispstat.c
+++ b/drivers/media/platform/omap3isp/ispstat.c
@@ -26,13 +26,12 @@
  */
 
 #include <linux/dma-mapping.h>
-#include <linux/omap-iommu.h>
 #include <linux/slab.h>
 #include <linux/uaccess.h>
 
 #include "isp.h"
 
-#define IS_COHERENT_BUF(stat)	((stat)->dma_ch >= 0)
+#define ISP_STAT_USES_DMAENGINE(stat)	((stat)->dma_ch >= 0)
 
 /*
  * MAGIC_SIZE must always be the greatest common divisor of
@@ -77,21 +76,10 @@ static void __isp_stat_buf_sync_magic(struct ispstat *stat,
 					dma_addr_t, unsigned long, size_t,
 					enum dma_data_direction))
 {
-	struct device *dev = stat->isp->dev;
-	struct page *pg;
-	dma_addr_t dma_addr;
-	u32 offset;
-
-	/* Initial magic words */
-	pg = vmalloc_to_page(buf->virt_addr);
-	dma_addr = pfn_to_dma(dev, page_to_pfn(pg));
-	dma_sync(dev, dma_addr, 0, MAGIC_SIZE, dir);
-
-	/* Final magic words */
-	pg = vmalloc_to_page(buf->virt_addr + buf_size);
-	dma_addr = pfn_to_dma(dev, page_to_pfn(pg));
-	offset = ((u32)buf->virt_addr + buf_size) & ~PAGE_MASK;
-	dma_sync(dev, dma_addr, offset, MAGIC_SIZE, dir);
+	/* Sync the initial and final magic words. */
+	dma_sync(stat->isp->dev, buf->dma_addr, 0, MAGIC_SIZE, dir);
+	dma_sync(stat->isp->dev, buf->dma_addr + (buf_size & PAGE_MASK),
+		 buf_size & ~PAGE_MASK, MAGIC_SIZE, dir);
 }
 
 static void isp_stat_buf_sync_magic_for_device(struct ispstat *stat,
@@ -99,7 +87,7 @@ static void isp_stat_buf_sync_magic_for_device(struct ispstat *stat,
 					       u32 buf_size,
 					       enum dma_data_direction dir)
 {
-	if (IS_COHERENT_BUF(stat))
+	if (ISP_STAT_USES_DMAENGINE(stat))
 		return;
 
 	__isp_stat_buf_sync_magic(stat, buf, buf_size, dir,
@@ -111,7 +99,7 @@ static void isp_stat_buf_sync_magic_for_cpu(struct ispstat *stat,
 					    u32 buf_size,
 					    enum dma_data_direction dir)
 {
-	if (IS_COHERENT_BUF(stat))
+	if (ISP_STAT_USES_DMAENGINE(stat))
 		return;
 
 	__isp_stat_buf_sync_magic(stat, buf, buf_size, dir,
@@ -180,21 +168,21 @@ static void isp_stat_buf_insert_magic(struct ispstat *stat,
 static void isp_stat_buf_sync_for_device(struct ispstat *stat,
 					 struct ispstat_buffer *buf)
 {
-	if (IS_COHERENT_BUF(stat))
+	if (ISP_STAT_USES_DMAENGINE(stat))
 		return;
 
-	dma_sync_sg_for_device(stat->isp->dev, buf->iovm->sgt->sgl,
-			       buf->iovm->sgt->nents, DMA_FROM_DEVICE);
+	dma_sync_sg_for_device(stat->isp->dev, buf->sgt.sgl,
+			       buf->sgt.nents, DMA_FROM_DEVICE);
 }
 
 static void isp_stat_buf_sync_for_cpu(struct ispstat *stat,
 				      struct ispstat_buffer *buf)
 {
-	if (IS_COHERENT_BUF(stat))
+	if (ISP_STAT_USES_DMAENGINE(stat))
 		return;
 
-	dma_sync_sg_for_cpu(stat->isp->dev, buf->iovm->sgt->sgl,
-			    buf->iovm->sgt->nents, DMA_FROM_DEVICE);
+	dma_sync_sg_for_cpu(stat->isp->dev, buf->sgt.sgl,
+			    buf->sgt.nents, DMA_FROM_DEVICE);
 }
 
 static void isp_stat_buf_clear(struct ispstat *stat)
@@ -354,29 +342,21 @@ static struct ispstat_buffer *isp_stat_buf_get(struct ispstat *stat,
 
 static void isp_stat_bufs_free(struct ispstat *stat)
 {
-	struct isp_device *isp = stat->isp;
-	int i;
+	struct device *dev = ISP_STAT_USES_DMAENGINE(stat)
+			   ? NULL : stat->isp->dev;
+	unsigned int i;
 
 	for (i = 0; i < STAT_MAX_BUFS; i++) {
 		struct ispstat_buffer *buf = &stat->buf[i];
 
-		if (!IS_COHERENT_BUF(stat)) {
-			if (IS_ERR_OR_NULL((void *)buf->iommu_addr))
-				continue;
-			if (buf->iovm)
-				dma_unmap_sg(isp->dev, buf->iovm->sgt->sgl,
-					     buf->iovm->sgt->nents,
-					     DMA_FROM_DEVICE);
-			omap_iommu_vfree(isp->domain, isp->dev,
-							buf->iommu_addr);
-		} else {
-			if (!buf->virt_addr)
-				continue;
-			dma_free_coherent(stat->isp->dev, stat->buf_alloc_size,
-					  buf->virt_addr, buf->dma_addr);
-		}
-		buf->iommu_addr = 0;
-		buf->iovm = NULL;
+		if (!buf->virt_addr)
+			continue;
+
+		sg_free_table(&buf->sgt);
+
+		dma_free_coherent(dev, stat->buf_alloc_size, buf->virt_addr,
+				  buf->dma_addr);
+
 		buf->dma_addr = 0;
 		buf->virt_addr = NULL;
 		buf->empty = 1;
@@ -389,83 +369,51 @@ static void isp_stat_bufs_free(struct ispstat *stat)
 	stat->active_buf = NULL;
 }
 
-static int isp_stat_bufs_alloc_iommu(struct ispstat *stat, unsigned int size)
-{
-	struct isp_device *isp = stat->isp;
-	int i;
-
-	stat->buf_alloc_size = size;
-
-	for (i = 0; i < STAT_MAX_BUFS; i++) {
-		struct ispstat_buffer *buf = &stat->buf[i];
-		struct iovm_struct *iovm;
-
-		WARN_ON(buf->dma_addr);
-		buf->iommu_addr = omap_iommu_vmalloc(isp->domain, isp->dev, 0,
-							size, IOMMU_FLAG);
-		if (IS_ERR((void *)buf->iommu_addr)) {
-			dev_err(stat->isp->dev,
-				 "%s: Can't acquire memory for "
-				 "buffer %d\n", stat->subdev.name, i);
-			isp_stat_bufs_free(stat);
-			return -ENOMEM;
-		}
-
-		iovm = omap_find_iovm_area(isp->dev, buf->iommu_addr);
-		if (!iovm ||
-		    !dma_map_sg(isp->dev, iovm->sgt->sgl, iovm->sgt->nents,
-				DMA_FROM_DEVICE)) {
-			isp_stat_bufs_free(stat);
-			return -ENOMEM;
-		}
-		buf->iovm = iovm;
-
-		buf->virt_addr = omap_da_to_va(stat->isp->dev,
-					  (u32)buf->iommu_addr);
-		buf->empty = 1;
-		dev_dbg(stat->isp->dev, "%s: buffer[%d] allocated."
-			"iommu_addr=0x%08lx virt_addr=0x%08lx",
-			stat->subdev.name, i, buf->iommu_addr,
-			(unsigned long)buf->virt_addr);
-	}
-
-	return 0;
-}
-
-static int isp_stat_bufs_alloc_dma(struct ispstat *stat, unsigned int size)
+static int isp_stat_bufs_alloc_one(struct device *dev,
+				   struct ispstat_buffer *buf,
+				   unsigned int size)
 {
-	int i;
-
-	stat->buf_alloc_size = size;
-
-	for (i = 0; i < STAT_MAX_BUFS; i++) {
-		struct ispstat_buffer *buf = &stat->buf[i];
-
-		WARN_ON(buf->iommu_addr);
-		buf->virt_addr = dma_alloc_coherent(stat->isp->dev, size,
-					&buf->dma_addr, GFP_KERNEL | GFP_DMA);
+	int ret;
 
-		if (!buf->virt_addr || !buf->dma_addr) {
-			dev_info(stat->isp->dev,
-				 "%s: Can't acquire memory for "
-				 "DMA buffer %d\n", stat->subdev.name, i);
-			isp_stat_bufs_free(stat);
-			return -ENOMEM;
-		}
-		buf->empty = 1;
+	buf->virt_addr = dma_alloc_coherent(dev, size, &buf->dma_addr,
+					    GFP_KERNEL | GFP_DMA);
+	if (!buf->virt_addr)
+		return -ENOMEM;
 
-		dev_dbg(stat->isp->dev, "%s: buffer[%d] allocated."
-			"dma_addr=0x%08lx virt_addr=0x%08lx\n",
-			stat->subdev.name, i, (unsigned long)buf->dma_addr,
-			(unsigned long)buf->virt_addr);
+	ret = dma_get_sgtable(dev, &buf->sgt, buf->virt_addr, buf->dma_addr,
+			      size);
+	if (ret < 0) {
+		dma_free_coherent(dev, size, buf->virt_addr, buf->dma_addr);
+		buf->virt_addr = NULL;
+		buf->dma_addr = 0;
+		return ret;
 	}
 
 	return 0;
 }
 
+/*
+ * The device passed to the DMA API depends on whether the statistics block uses
+ * ISP DMA, external DMA or PIO to transfer data.
+ *
+ * The first case (for the AEWB and AF engines) passes the ISP device, resulting
+ * in the DMA buffers being mapped through the ISP IOMMU.
+ *
+ * The second case (for the histogram engine) should pass the DMA engine device.
+ * As that device isn't accessible through the OMAP DMA engine API the driver
+ * passes NULL instead, resulting in the buffers being mapped directly as
+ * physical pages.
+ *
+ * The third case (for the histogram engine) doesn't require any mapping. The
+ * buffers could be allocated with kmalloc/vmalloc, but we still use
+ * dma_alloc_coherent() for consistency purpose.
+ */
 static int isp_stat_bufs_alloc(struct ispstat *stat, u32 size)
 {
+	struct device *dev = ISP_STAT_USES_DMAENGINE(stat)
+			   ? NULL : stat->isp->dev;
 	unsigned long flags;
+	unsigned int i;
 
 	spin_lock_irqsave(&stat->isp->stat_lock, flags);
 
@@ -489,10 +437,31 @@ static int isp_stat_bufs_alloc(struct ispstat *stat, u32 size)
 
 	isp_stat_bufs_free(stat);
 
-	if (IS_COHERENT_BUF(stat))
-		return isp_stat_bufs_alloc_dma(stat, size);
-	else
-		return isp_stat_bufs_alloc_iommu(stat, size);
+	stat->buf_alloc_size = size;
+
+	for (i = 0; i < STAT_MAX_BUFS; i++) {
+		struct ispstat_buffer *buf = &stat->buf[i];
+		int ret;
+
+		ret = isp_stat_bufs_alloc_one(dev, buf, size);
+		if (ret < 0) {
+			dev_err(stat->isp->dev,
+				"%s: Failed to allocate DMA buffer %u\n",
+				stat->subdev.name, i);
+			isp_stat_bufs_free(stat);
+			return ret;
+		}
+
+		buf->empty = 1;
+
+		dev_dbg(stat->isp->dev,
+			"%s: buffer[%u] allocated. dma=0x%08lx virt=0x%08lx",
+			stat->subdev.name, i,
+			(unsigned long)buf->dma_addr,
+			(unsigned long)buf->virt_addr);
+	}
+
+	return 0;
 }
 
 static void isp_stat_queue_event(struct ispstat *stat, int err)
diff --git a/drivers/media/platform/omap3isp/ispstat.h b/drivers/media/platform/omap3isp/ispstat.h
index 9a047c929b9f..58d6ac7cb664 100644
--- a/drivers/media/platform/omap3isp/ispstat.h
+++ b/drivers/media/platform/omap3isp/ispstat.h
@@ -46,8 +46,7 @@
 struct ispstat;
 
 struct ispstat_buffer {
-	unsigned long iommu_addr;
-	struct iovm_struct *iovm;
+	struct sg_table sgt;
 	void *virt_addr;
 	dma_addr_t dma_addr;
 	struct timespec ts;
diff --git a/drivers/media/platform/omap3isp/ispvideo.c b/drivers/media/platform/omap3isp/ispvideo.c
index 85b4036ba5e4..e36bac26476c 100644
--- a/drivers/media/platform/omap3isp/ispvideo.c
+++ b/drivers/media/platform/omap3isp/ispvideo.c
@@ -27,7 +27,6 @@
 #include <linux/clk.h>
 #include <linux/mm.h>
 #include <linux/module.h>
-#include <linux/omap-iommu.h>
 #include <linux/pagemap.h>
 #include <linux/scatterlist.h>
 #include <linux/sched.h>
@@ -35,6 +34,7 @@
 #include <linux/vmalloc.h>
 #include <media/v4l2-dev.h>
 #include <media/v4l2-ioctl.h>
+#include <media/videobuf2-dma-contig.h>
 
 #include "ispvideo.h"
 #include "isp.h"
@@ -326,90 +326,36 @@ isp_video_check_format(struct isp_video *video, struct isp_video_fh *vfh)
 }
 
 /* -----------------------------------------------------------------------------
- * IOMMU management
- */
-
-#define IOMMU_FLAG	(IOVMF_ENDIAN_LITTLE | IOVMF_ELSZ_8)
-
-/*
- * ispmmu_vmap - Wrapper for Virtual memory mapping of a scatter gather list
- * @isp: Device pointer specific to the OMAP3 ISP.
- * @sglist: Pointer to source Scatter gather list to allocate.
- * @sglen: Number of elements of the scatter-gatter list.
- *
- * Returns a resulting mapped device address by the ISP MMU, or -ENOMEM if
- * we ran out of memory.
- */
-static dma_addr_t
-ispmmu_vmap(struct isp_device *isp, const struct scatterlist *sglist, int sglen)
-{
-	struct sg_table *sgt;
-	u32 da;
-
-	sgt = kmalloc(sizeof(*sgt), GFP_KERNEL);
-	if (sgt == NULL)
-		return -ENOMEM;
-
-	sgt->sgl = (struct scatterlist *)sglist;
-	sgt->nents = sglen;
-	sgt->orig_nents = sglen;
-
-	da = omap_iommu_vmap(isp->domain, isp->dev, 0, sgt, IOMMU_FLAG);
-	if (IS_ERR_VALUE(da))
-		kfree(sgt);
-
-	return da;
-}
-
-/*
- * ispmmu_vunmap - Unmap a device address from the ISP MMU
- * @isp: Device pointer specific to the OMAP3 ISP.
- * @da: Device address generated from a ispmmu_vmap call.
- */
-static void ispmmu_vunmap(struct isp_device *isp, dma_addr_t da)
-{
-	struct sg_table *sgt;
-
-	sgt = omap_iommu_vunmap(isp->domain, isp->dev, (u32)da);
-	kfree(sgt);
-}
-
-/* -----------------------------------------------------------------------------
  * Video queue operations
  */
 
-static void isp_video_queue_prepare(struct isp_video_queue *queue,
-				    unsigned int *nbuffers, unsigned int *size)
+static int isp_video_queue_setup(struct vb2_queue *queue,
+				 const struct v4l2_format *fmt,
+				 unsigned int *count, unsigned int *num_planes,
+				 unsigned int sizes[], void *alloc_ctxs[])
 {
-	struct isp_video_fh *vfh =
-		container_of(queue, struct isp_video_fh, queue);
+	struct isp_video_fh *vfh = vb2_get_drv_priv(queue);
 	struct isp_video *video = vfh->video;
 
-	*size = vfh->format.fmt.pix.sizeimage;
-	if (*size == 0)
-		return;
+	*num_planes = 1;
 
-	*nbuffers = min(*nbuffers, video->capture_mem / PAGE_ALIGN(*size));
-}
+	sizes[0] = vfh->format.fmt.pix.sizeimage;
+	if (sizes[0] == 0)
+		return -EINVAL;
 
-static void isp_video_buffer_cleanup(struct isp_video_buffer *buf)
-{
-	struct isp_video_fh *vfh = isp_video_queue_to_isp_video_fh(buf->queue);
-	struct isp_buffer *buffer = to_isp_buffer(buf);
-	struct isp_video *video = vfh->video;
+	alloc_ctxs[0] = video->alloc_ctx;
 
-	if (buffer->isp_addr) {
-		ispmmu_vunmap(video->isp, buffer->isp_addr);
-		buffer->isp_addr = 0;
-	}
+	*count = min(*count, video->capture_mem / PAGE_ALIGN(sizes[0]));
+
+	return 0;
 }
 
-static int isp_video_buffer_prepare(struct isp_video_buffer *buf)
+static int isp_video_buffer_prepare(struct vb2_buffer *buf)
 {
-	struct isp_video_fh *vfh = isp_video_queue_to_isp_video_fh(buf->queue);
+	struct isp_video_fh *vfh = vb2_get_drv_priv(buf->vb2_queue);
 	struct isp_buffer *buffer = to_isp_buffer(buf);
 	struct isp_video *video = vfh->video;
-	unsigned long addr;
+	dma_addr_t addr;
 
 	/* Refuse to prepare the buffer is the video node has registered an
 	 * error. We don't need to take any lock here as the operation is
@@ -420,19 +366,16 @@ static int isp_video_buffer_prepare(struct isp_video_buffer *buf)
 	if (unlikely(video->error))
 		return -EIO;
 
-	addr = ispmmu_vmap(video->isp, buf->sglist, buf->sglen);
-	if (IS_ERR_VALUE(addr))
-		return -EIO;
-
+	addr = vb2_dma_contig_plane_dma_addr(buf, 0);
 	if (!IS_ALIGNED(addr, 32)) {
-		dev_dbg(video->isp->dev, "Buffer address must be "
-			"aligned to 32 bytes boundary.\n");
-		ispmmu_vunmap(video->isp, buffer->isp_addr);
+		dev_dbg(video->isp->dev,
+			"Buffer address must be aligned to 32 bytes boundary.\n");
 		return -EINVAL;
 	}
 
-	buf->vbuf.bytesused = vfh->format.fmt.pix.sizeimage;
-	buffer->isp_addr = addr;
+	vb2_set_plane_payload(&buffer->vb, 0, vfh->format.fmt.pix.sizeimage);
+	buffer->dma = addr;
+
 	return 0;
 }
 
@@ -445,9 +388,9 @@ static int isp_video_buffer_prepare(struct isp_video_buffer *buf)
  * If the pipeline is busy, it will be restarted in the output module interrupt
  * handler.
  */
-static void isp_video_buffer_queue(struct isp_video_buffer *buf)
+static void isp_video_buffer_queue(struct vb2_buffer *buf)
 {
-	struct isp_video_fh *vfh = isp_video_queue_to_isp_video_fh(buf->queue);
+	struct isp_video_fh *vfh = vb2_get_drv_priv(buf->vb2_queue);
 	struct isp_buffer *buffer = to_isp_buffer(buf);
 	struct isp_video *video = vfh->video;
 	struct isp_pipeline *pipe = to_isp_pipeline(&video->video.entity);
@@ -456,14 +399,18 @@ static void isp_video_buffer_queue(struct isp_video_buffer *buf)
 	unsigned int empty;
 	unsigned int start;
 
+	spin_lock_irqsave(&video->irqlock, flags);
+
 	if (unlikely(video->error)) {
-		buf->state = ISP_BUF_STATE_ERROR;
-		wake_up(&buf->wait);
+		vb2_buffer_done(&buffer->vb, VB2_BUF_STATE_ERROR);
+		spin_unlock_irqrestore(&video->irqlock, flags);
 		return;
 	}
 
 	empty = list_empty(&video->dmaqueue);
-	list_add_tail(&buffer->buffer.irqlist, &video->dmaqueue);
+	list_add_tail(&buffer->irqlist, &video->dmaqueue);
+
+	spin_unlock_irqrestore(&video->irqlock, flags);
 
 	if (empty) {
 		if (video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
@@ -487,23 +434,22 @@ static void isp_video_buffer_queue(struct isp_video_buffer *buf)
 	}
 }
 
-static const struct isp_video_queue_operations isp_video_queue_ops = {
-	.queue_prepare = &isp_video_queue_prepare,
-	.buffer_prepare = &isp_video_buffer_prepare,
-	.buffer_queue = &isp_video_buffer_queue,
-	.buffer_cleanup = &isp_video_buffer_cleanup,
+static const struct vb2_ops isp_video_queue_ops = {
+	.queue_setup = isp_video_queue_setup,
+	.buf_prepare = isp_video_buffer_prepare,
+	.buf_queue = isp_video_buffer_queue,
 };
 
 /*
  * omap3isp_video_buffer_next - Complete the current buffer and return the next
  * @video: ISP video object
  *
- * Remove the current video buffer from the DMA queue and fill its timestamp,
- * field count and state fields before waking up its completion handler.
+ * Remove the current video buffer from the DMA queue and fill its timestamp and
+ * field count before handing it back to videobuf2.
  *
- * For capture video nodes the buffer state is set to ISP_BUF_STATE_DONE if no
- * error has been flagged in the pipeline, or to ISP_BUF_STATE_ERROR otherwise.
- * For video output nodes the buffer state is always set to ISP_BUF_STATE_DONE.
+ * For capture video nodes the buffer state is set to VB2_BUF_STATE_DONE if no
+ * error has been flagged in the pipeline, or to VB2_BUF_STATE_ERROR otherwise.
+ * For video output nodes the buffer state is always set to VB2_BUF_STATE_DONE.
  *
  * The DMA queue is expected to contain at least one buffer.
  *
@@ -513,26 +459,25 @@ static const struct isp_video_queue_operations isp_video_queue_ops = {
 struct isp_buffer *omap3isp_video_buffer_next(struct isp_video *video)
 {
 	struct isp_pipeline *pipe = to_isp_pipeline(&video->video.entity);
-	struct isp_video_queue *queue = video->queue;
 	enum isp_pipeline_state state;
-	struct isp_video_buffer *buf;
+	struct isp_buffer *buf;
 	unsigned long flags;
 	struct timespec ts;
 
-	spin_lock_irqsave(&queue->irqlock, flags);
+	spin_lock_irqsave(&video->irqlock, flags);
 	if (WARN_ON(list_empty(&video->dmaqueue))) {
-		spin_unlock_irqrestore(&queue->irqlock, flags);
+		spin_unlock_irqrestore(&video->irqlock, flags);
 		return NULL;
 	}
 
-	buf = list_first_entry(&video->dmaqueue, struct isp_video_buffer,
+	buf = list_first_entry(&video->dmaqueue, struct isp_buffer,
 			       irqlist);
 	list_del(&buf->irqlist);
-	spin_unlock_irqrestore(&queue->irqlock, flags);
+	spin_unlock_irqrestore(&video->irqlock, flags);
 
 	ktime_get_ts(&ts);
-	buf->vbuf.timestamp.tv_sec = ts.tv_sec;
-	buf->vbuf.timestamp.tv_usec = ts.tv_nsec / NSEC_PER_USEC;
+	buf->vb.v4l2_buf.timestamp.tv_sec = ts.tv_sec;
+	buf->vb.v4l2_buf.timestamp.tv_usec = ts.tv_nsec / NSEC_PER_USEC;
 
 	/* Do frame number propagation only if this is the output video node.
 	 * Frame number either comes from the CSI receivers or it gets
@@ -541,22 +486,27 @@ struct isp_buffer *omap3isp_video_buffer_next(struct isp_video *video)
 	 * first, so the input number might lag behind by 1 in some cases.
 	 */
 	if (video == pipe->output && !pipe->do_propagation)
-		buf->vbuf.sequence = atomic_inc_return(&pipe->frame_number);
+		buf->vb.v4l2_buf.sequence =
+			atomic_inc_return(&pipe->frame_number);
 	else
-		buf->vbuf.sequence = atomic_read(&pipe->frame_number);
+		buf->vb.v4l2_buf.sequence = atomic_read(&pipe->frame_number);
 
 	/* Report pipeline errors to userspace on the capture device side. */
-	if (queue->type == V4L2_BUF_TYPE_VIDEO_CAPTURE && pipe->error) {
-		buf->state = ISP_BUF_STATE_ERROR;
+	if (video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE && pipe->error) {
+		state = VB2_BUF_STATE_ERROR;
 		pipe->error = false;
 	} else {
-		buf->state = ISP_BUF_STATE_DONE;
+		state = VB2_BUF_STATE_DONE;
 	}
 
-	wake_up(&buf->wait);
+	vb2_buffer_done(&buf->vb, state);
+
+	spin_lock_irqsave(&video->irqlock, flags);
 
 	if (list_empty(&video->dmaqueue)) {
-		if (queue->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
+		spin_unlock_irqrestore(&video->irqlock, flags);
+
+		if (video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
 			state = ISP_PIPELINE_QUEUE_OUTPUT
 			      | ISP_PIPELINE_STREAM;
 		else
@@ -571,16 +521,19 @@ struct isp_buffer *omap3isp_video_buffer_next(struct isp_video *video)
 		return NULL;
 	}
 
-	if (queue->type == V4L2_BUF_TYPE_VIDEO_CAPTURE && pipe->input != NULL) {
-		spin_lock_irqsave(&pipe->lock, flags);
+	if (video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE && pipe->input != NULL) {
+		spin_lock(&pipe->lock);
 		pipe->state &= ~ISP_PIPELINE_STREAM;
-		spin_unlock_irqrestore(&pipe->lock, flags);
+		spin_unlock(&pipe->lock);
 	}
 
-	buf = list_first_entry(&video->dmaqueue, struct isp_video_buffer,
+	buf = list_first_entry(&video->dmaqueue, struct isp_buffer,
 			       irqlist);
-	buf->state = ISP_BUF_STATE_ACTIVE;
-	return to_isp_buffer(buf);
+	buf->vb.state = VB2_BUF_STATE_ACTIVE;
+
+	spin_unlock_irqrestore(&video->irqlock, flags);
+
+	return buf;
 }
 
 /*
@@ -592,25 +545,22 @@ struct isp_buffer *omap3isp_video_buffer_next(struct isp_video *video)
  */
 void omap3isp_video_cancel_stream(struct isp_video *video)
 {
-	struct isp_video_queue *queue = video->queue;
 	unsigned long flags;
 
-	spin_lock_irqsave(&queue->irqlock, flags);
+	spin_lock_irqsave(&video->irqlock, flags);
 
 	while (!list_empty(&video->dmaqueue)) {
-		struct isp_video_buffer *buf;
+		struct isp_buffer *buf;
 
 		buf = list_first_entry(&video->dmaqueue,
-				       struct isp_video_buffer, irqlist);
+				       struct isp_buffer, irqlist);
 		list_del(&buf->irqlist);
-
-		buf->state = ISP_BUF_STATE_ERROR;
-		wake_up(&buf->wait);
+		vb2_buffer_done(&buf->vb, VB2_BUF_STATE_ERROR);
 	}
 
 	video->error = true;
 
-	spin_unlock_irqrestore(&queue->irqlock, flags);
+	spin_unlock_irqrestore(&video->irqlock, flags);
 }
 
 /*
@@ -627,12 +577,15 @@ void omap3isp_video_resume(struct isp_video *video, int continuous)
 {
 	struct isp_buffer *buf = NULL;
 
-	if (continuous && video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
-		omap3isp_video_queue_discard_done(video->queue);
+	if (continuous && video->type == V4L2_BUF_TYPE_VIDEO_CAPTURE) {
+		mutex_lock(&video->queue_lock);
+		vb2_discard_done(video->queue);
+		mutex_unlock(&video->queue_lock);
+	}
 
 	if (!list_empty(&video->dmaqueue)) {
 		buf = list_first_entry(&video->dmaqueue,
-				       struct isp_buffer, buffer.irqlist);
+				       struct isp_buffer, irqlist);
 		video->ops->queue(video, buf);
 		video->dmaqueue_flags |= ISP_VIDEO_DMAQUEUE_QUEUED;
 	} else {
@@ -840,33 +793,56 @@ static int
 isp_video_reqbufs(struct file *file, void *fh, struct v4l2_requestbuffers *rb)
 {
 	struct isp_video_fh *vfh = to_isp_video_fh(fh);
+	struct isp_video *video = video_drvdata(file);
+	int ret;
 
-	return omap3isp_video_queue_reqbufs(&vfh->queue, rb);
+	mutex_lock(&video->queue_lock);
+	ret = vb2_reqbufs(&vfh->queue, rb);
+	mutex_unlock(&video->queue_lock);
+
+	return ret;
 }
 
 static int
 isp_video_querybuf(struct file *file, void *fh, struct v4l2_buffer *b)
 {
 	struct isp_video_fh *vfh = to_isp_video_fh(fh);
+	struct isp_video *video = video_drvdata(file);
+	int ret;
+
+	mutex_lock(&video->queue_lock);
+	ret = vb2_querybuf(&vfh->queue, b);
+	mutex_unlock(&video->queue_lock);
 
-	return omap3isp_video_queue_querybuf(&vfh->queue, b);
+	return ret;
 }
 
 static int
 isp_video_qbuf(struct file *file, void *fh, struct v4l2_buffer *b)
 {
 	struct isp_video_fh *vfh = to_isp_video_fh(fh);
+	struct isp_video *video = video_drvdata(file);
+	int ret;
 
-	return omap3isp_video_queue_qbuf(&vfh->queue, b);
+	mutex_lock(&video->queue_lock);
+	ret = vb2_qbuf(&vfh->queue, b);
+	mutex_unlock(&video->queue_lock);
+
+	return ret;
 }
 
 static int
 isp_video_dqbuf(struct file *file, void *fh, struct v4l2_buffer *b)
 {
 	struct isp_video_fh *vfh = to_isp_video_fh(fh);
+	struct isp_video *video = video_drvdata(file);
+	int ret;
+
+	mutex_lock(&video->queue_lock);
+	ret = vb2_dqbuf(&vfh->queue, b, file->f_flags & O_NONBLOCK);
+	mutex_unlock(&video->queue_lock);
 
-	return omap3isp_video_queue_dqbuf(&vfh->queue, b,
-					  file->f_flags & O_NONBLOCK);
+	return ret;
 }
 
 static int isp_video_check_external_subdevs(struct isp_video *video,
@@ -1006,11 +982,6 @@ isp_video_streamon(struct file *file, void *fh, enum v4l2_buf_type type)
 
 	mutex_lock(&video->stream_lock);
 
-	if (video->streaming) {
-		mutex_unlock(&video->stream_lock);
-		return -EBUSY;
-	}
-
 	/* Start streaming on the pipeline. No link touching an entity in the
 	 * pipeline can be activated or deactivated once streaming is started.
 	 */
@@ -1069,7 +1040,9 @@ isp_video_streamon(struct file *file, void *fh, enum v4l2_buf_type type)
 	INIT_LIST_HEAD(&video->dmaqueue);
 	atomic_set(&pipe->frame_number, -1);
 
-	ret = omap3isp_video_queue_streamon(&vfh->queue);
+	mutex_lock(&video->queue_lock);
+	ret = vb2_streamon(&vfh->queue, type);
+	mutex_unlock(&video->queue_lock);
 	if (ret < 0)
 		goto err_check_format;
 
@@ -1082,19 +1055,19 @@ isp_video_streamon(struct file *file, void *fh, enum v4l2_buf_type type)
 					      ISP_PIPELINE_STREAM_CONTINUOUS);
 		if (ret < 0)
 			goto err_set_stream;
-		spin_lock_irqsave(&video->queue->irqlock, flags);
+		spin_lock_irqsave(&video->irqlock, flags);
 		if (list_empty(&video->dmaqueue))
 			video->dmaqueue_flags |= ISP_VIDEO_DMAQUEUE_UNDERRUN;
-		spin_unlock_irqrestore(&video->queue->irqlock, flags);
+		spin_unlock_irqrestore(&video->irqlock, flags);
 	}
 
-	video->streaming = 1;
-
 	mutex_unlock(&video->stream_lock);
 	return 0;
 
 err_set_stream:
-	omap3isp_video_queue_streamoff(&vfh->queue);
+	mutex_lock(&video->queue_lock);
+	vb2_streamoff(&vfh->queue, type);
+	mutex_unlock(&video->queue_lock);
 err_check_format:
 	media_entity_pipeline_stop(&video->video.entity);
 err_pipeline_start:
@@ -1130,9 +1103,9 @@ isp_video_streamoff(struct file *file, void *fh, enum v4l2_buf_type type)
 	mutex_lock(&video->stream_lock);
 
 	/* Make sure we're not streaming yet. */
-	mutex_lock(&vfh->queue.lock);
-	streaming = vfh->queue.streaming;
-	mutex_unlock(&vfh->queue.lock);
+	mutex_lock(&video->queue_lock);
+	streaming = vb2_is_streaming(&vfh->queue);
+	mutex_unlock(&video->queue_lock);
 
 	if (!streaming)
 		goto done;
@@ -1151,9 +1124,12 @@ isp_video_streamoff(struct file *file, void *fh, enum v4l2_buf_type type)
 
 	/* Stop the stream. */
 	omap3isp_pipeline_set_stream(pipe, ISP_PIPELINE_STREAM_STOPPED);
-	omap3isp_video_queue_streamoff(&vfh->queue);
+	omap3isp_video_cancel_stream(video);
+
+	mutex_lock(&video->queue_lock);
+	vb2_streamoff(&vfh->queue, type);
+	mutex_unlock(&video->queue_lock);
 	video->queue = NULL;
-	video->streaming = 0;
 	video->error = false;
 
 	if (video->isp->pdata->set_constraints)
@@ -1223,6 +1199,7 @@ static int isp_video_open(struct file *file)
 {
 	struct isp_video *video = video_drvdata(file);
 	struct isp_video_fh *handle;
+	struct vb2_queue *queue;
 	int ret = 0;
 
 	handle = kzalloc(sizeof(*handle), GFP_KERNEL);
@@ -1244,9 +1221,20 @@ static int isp_video_open(struct file *file)
 		goto done;
 	}
 
-	omap3isp_video_queue_init(&handle->queue, video->type,
-				  &isp_video_queue_ops, video->isp->dev,
-				  sizeof(struct isp_buffer));
+	queue = &handle->queue;
+	queue->type = video->type;
+	queue->io_modes = VB2_MMAP | VB2_USERPTR;
+	queue->drv_priv = handle;
+	queue->ops = &isp_video_queue_ops;
+	queue->mem_ops = &vb2_dma_contig_memops;
+	queue->buf_struct_size = sizeof(struct isp_buffer);
+	queue->timestamp_flags = V4L2_BUF_FLAG_TIMESTAMP_MONOTONIC;
+
+	ret = vb2_queue_init(&handle->queue);
+	if (ret < 0) {
+		omap3isp_put(video->isp);
+		goto done;
+	}
 
 	memset(&handle->format, 0, sizeof(handle->format));
 	handle->format.type = video->type;
@@ -1273,9 +1261,9 @@ static int isp_video_release(struct file *file)
 	/* Disable streaming and free the buffers queue resources. */
 	isp_video_streamoff(file, vfh, video->type);
 
-	mutex_lock(&handle->queue.lock);
-	omap3isp_video_queue_cleanup(&handle->queue);
-	mutex_unlock(&handle->queue.lock);
+	mutex_lock(&video->queue_lock);
+	vb2_queue_release(&handle->queue);
+	mutex_unlock(&video->queue_lock);
 
 	omap3isp_pipeline_pm_use(&video->video.entity, 0);
 
@@ -1292,16 +1280,27 @@ static int isp_video_release(struct file *file)
 static unsigned int isp_video_poll(struct file *file, poll_table *wait)
 {
 	struct isp_video_fh *vfh = to_isp_video_fh(file->private_data);
-	struct isp_video_queue *queue = &vfh->queue;
+	struct isp_video *video = video_drvdata(file);
+	int ret;
 
-	return omap3isp_video_queue_poll(queue, file, wait);
+	mutex_lock(&video->queue_lock);
+	ret = vb2_poll(&vfh->queue, file, wait);
+	mutex_unlock(&video->queue_lock);
+
+	return ret;
 }
 
 static int isp_video_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct isp_video_fh *vfh = to_isp_video_fh(file->private_data);
+	struct isp_video *video = video_drvdata(file);
+	int ret;
+
+	mutex_lock(&video->queue_lock);
+	ret = vb2_mmap(&vfh->queue, vma);
+	mutex_unlock(&video->queue_lock);
 
-	return omap3isp_video_queue_mmap(&vfh->queue, vma);
+	return ret;
 }
 
 static struct v4l2_file_operations isp_video_fops = {
@@ -1342,15 +1341,23 @@ int omap3isp_video_init(struct isp_video *video, const char *name)
 		return -EINVAL;
 	}
 
+	video->alloc_ctx = vb2_dma_contig_init_ctx(video->isp->dev);
+	if (IS_ERR(video->alloc_ctx))
+		return PTR_ERR(video->alloc_ctx);
+
 	ret = media_entity_init(&video->video.entity, 1, &video->pad, 0);
-	if (ret < 0)
+	if (ret < 0) {
+		vb2_dma_contig_cleanup_ctx(video->alloc_ctx);
 		return ret;
+	}
 
 	mutex_init(&video->mutex);
 	atomic_set(&video->active, 0);
 
 	spin_lock_init(&video->pipe.lock);
 	mutex_init(&video->stream_lock);
+	mutex_init(&video->queue_lock);
+	spin_lock_init(&video->irqlock);
 
 	/* Initialize the video device. */
 	if (video->ops == NULL)
@@ -1371,7 +1378,9 @@ int omap3isp_video_init(struct isp_video *video, const char *name)
 
 void omap3isp_video_cleanup(struct isp_video *video)
 {
+	vb2_dma_contig_cleanup_ctx(video->alloc_ctx);
 	media_entity_cleanup(&video->video.entity);
+	mutex_destroy(&video->queue_lock);
 	mutex_destroy(&video->stream_lock);
 	mutex_destroy(&video->mutex);
 }
diff --git a/drivers/media/platform/omap3isp/ispvideo.h b/drivers/media/platform/omap3isp/ispvideo.h
index 4e194076cc60..7d2e82122ecd 100644
--- a/drivers/media/platform/omap3isp/ispvideo.h
+++ b/drivers/media/platform/omap3isp/ispvideo.h
@@ -30,8 +30,7 @@
 #include <media/media-entity.h>
 #include <media/v4l2-dev.h>
 #include <media/v4l2-fh.h>
-
-#include "ispqueue.h"
+#include <media/videobuf2-core.h>
 
 #define ISP_VIDEO_DRIVER_NAME		"ispvideo"
 #define ISP_VIDEO_DRIVER_VERSION	"0.0.2"
@@ -124,17 +123,19 @@ static inline int isp_pipeline_ready(struct isp_pipeline *pipe)
 			       ISP_PIPELINE_IDLE_OUTPUT);
 }
 
-/*
- * struct isp_buffer - ISP buffer
- * @buffer: ISP video buffer
- * @isp_addr: MMU mapped address (a.k.a. device address) of the buffer.
+/**
+ * struct isp_buffer - ISP video buffer
+ * @vb: videobuf2 buffer
+ * @irqlist: List head for insertion into IRQ queue
+ * @dma: DMA address
  */
 struct isp_buffer {
-	struct isp_video_buffer buffer;
-	dma_addr_t isp_addr;
+	struct vb2_buffer vb;
+	struct list_head irqlist;
+	dma_addr_t dma;
 };
 
-#define to_isp_buffer(buf)	container_of(buf, struct isp_buffer, buffer)
+#define to_isp_buffer(buf)	container_of(buf, struct isp_buffer, vb)
 
 enum isp_video_dmaqueue_flags {
 	/* Set if DMA queue becomes empty when ISP_PIPELINE_STREAM_CONTINUOUS */
@@ -172,16 +173,16 @@ struct isp_video {
 	unsigned int bpl_value;		/* bytes per line value */
 	unsigned int bpl_padding;	/* padding at end of line */
 
-	/* Entity video node streaming */
-	unsigned int streaming:1;
-
 	/* Pipeline state */
 	struct isp_pipeline pipe;
 	struct mutex stream_lock;	/* pipeline and stream states */
 	bool error;
 
 	/* Video buffers queue */
-	struct isp_video_queue *queue;
+	void *alloc_ctx;
+	struct vb2_queue *queue;
+	struct mutex queue_lock;	/* protects the queue */
+	spinlock_t irqlock;		/* protects dmaqueue */
 	struct list_head dmaqueue;
 	enum isp_video_dmaqueue_flags dmaqueue_flags;
 
@@ -193,7 +194,7 @@ struct isp_video {
 struct isp_video_fh {
 	struct v4l2_fh vfh;
 	struct isp_video *video;
-	struct isp_video_queue queue;
+	struct vb2_queue queue;
 	struct v4l2_format format;
 	struct v4l2_fract timeperframe;
 };
diff --git a/drivers/media/v4l2-core/videobuf2-core.c b/drivers/media/v4l2-core/videobuf2-core.c
index 349e659d75fb..7c4489c42365 100644
--- a/drivers/media/v4l2-core/videobuf2-core.c
+++ b/drivers/media/v4l2-core/videobuf2-core.c
@@ -1200,6 +1200,30 @@ void vb2_buffer_done(struct vb2_buffer *vb, enum vb2_buffer_state state)
 EXPORT_SYMBOL_GPL(vb2_buffer_done);
 
 /**
+ * vb2_discard_done() - discard all buffers marked as DONE
+ * @q:		videobuf2 queue
+ *
+ * This function is intended to be used with suspend/resume operations. It
+ * discards all 'done' buffers as they would be too old to be requested after
+ * resume.
+ *
+ * Drivers must stop the hardware and synchronize with interrupt handlers and/or
+ * delayed works before calling this function to make sure no buffer will be
+ * touched by the driver and/or hardware.
+ */
+void vb2_discard_done(struct vb2_queue *q)
+{
+	struct vb2_buffer *vb;
+	unsigned long flags;
+
+	spin_lock_irqsave(&q->done_lock, flags);
+	list_for_each_entry(vb, &q->done_list, done_entry)
+		vb->state = VB2_BUF_STATE_ERROR;
+	spin_unlock_irqrestore(&q->done_lock, flags);
+}
+EXPORT_SYMBOL_GPL(vb2_discard_done);
+
+/**
  * __fill_vb2_buffer() - fill a vb2_buffer with information provided in a
  * v4l2_buffer by the userspace. The caller has already verified that struct
  * v4l2_buffer has a valid number of planes.
diff --git a/drivers/memstick/host/rtsx_pci_ms.c b/drivers/memstick/host/rtsx_pci_ms.c
index 2a635b6fdaf7..c880ba685754 100644
--- a/drivers/memstick/host/rtsx_pci_ms.c
+++ b/drivers/memstick/host/rtsx_pci_ms.c
@@ -601,6 +601,7 @@ static int rtsx_pci_ms_drv_remove(struct platform_device *pdev)
 	pcr->slots[RTSX_MS_CARD].card_event = NULL;
 	msh = host->msh;
 	host->eject = true;
+	cancel_work_sync(&host->handle_req);
 
 	mutex_lock(&host->host_mutex);
 	if (host->req) {
diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index 570b18a113ff..ebc0af7d769c 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -1037,7 +1037,7 @@ mpt_free_msg_frame(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf)
 		goto out;
 	/* signature to know if this mf is freed */
 	mf->u.frame.linkage.arg1 = cpu_to_le32(0xdeadbeaf);
-	list_add_tail(&mf->u.frame.linkage.list, &ioc->FreeQ);
+	list_add(&mf->u.frame.linkage.list, &ioc->FreeQ);
 #ifdef MFCNT
 	ioc->mfcnt--;
 #endif
diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index dcc8385adeb3..8a050e885688 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -2432,9 +2432,9 @@ mptctl_hp_hostinfo(unsigned long arg, unsigned int data_size)
 	int			rc, cim_rev;
 	ToolboxIstwiReadWriteRequest_t	*IstwiRWRequest;
 	MPT_FRAME_HDR		*mf = NULL;
-	MPIHeader_t		*mpi_hdr;
 	unsigned long		timeleft;
 	int			retval;
+	u32			msgcontext;
 
 	/* Reset long to int. Should affect IA64 and SPARC only
 	 */
@@ -2581,11 +2581,11 @@ mptctl_hp_hostinfo(unsigned long arg, unsigned int data_size)
 	}
 
 	IstwiRWRequest = (ToolboxIstwiReadWriteRequest_t *)mf;
-	mpi_hdr = (MPIHeader_t *) mf;
+	msgcontext = IstwiRWRequest->MsgContext;
 	memset(IstwiRWRequest,0,sizeof(ToolboxIstwiReadWriteRequest_t));
+	IstwiRWRequest->MsgContext = msgcontext;
 	IstwiRWRequest->Function = MPI_FUNCTION_TOOLBOX;
 	IstwiRWRequest->Tool = MPI_TOOLBOX_ISTWI_READ_WRITE_TOOL;
-	IstwiRWRequest->MsgContext = mpi_hdr->MsgContext;
 	IstwiRWRequest->Flags = MPI_TB_ISTWI_FLAGS_READ;
 	IstwiRWRequest->NumAddressBytes = 0x01;
 	IstwiRWRequest->DataLength = cpu_to_le16(0x04);
diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c
index fd75108c355e..02a3eefd6931 100644
--- a/drivers/message/fusion/mptfc.c
+++ b/drivers/message/fusion/mptfc.c
@@ -649,7 +649,7 @@ mptfc_slave_alloc(struct scsi_device *sdev)
 }
 
 static int
-mptfc_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
+mptfc_qcmd(struct Scsi_Host *shost, struct scsi_cmnd *SCpnt)
 {
 	struct mptfc_rport_info	*ri;
 	struct fc_rport	*rport = starget_to_rport(scsi_target(SCpnt->device));
@@ -658,14 +658,14 @@ mptfc_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
 
 	if (!vdevice || !vdevice->vtarget) {
 		SCpnt->result = DID_NO_CONNECT << 16;
-		done(SCpnt);
+		SCpnt->scsi_done(SCpnt);
 		return 0;
 	}
 
 	err = fc_remote_port_chkready(rport);
 	if (unlikely(err)) {
 		SCpnt->result = err;
-		done(SCpnt);
+		SCpnt->scsi_done(SCpnt);
 		return 0;
 	}
 
@@ -673,15 +673,13 @@ mptfc_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
 	ri = *((struct mptfc_rport_info **)rport->dd_data);
 	if (unlikely(!ri)) {
 		SCpnt->result = DID_IMM_RETRY << 16;
-		done(SCpnt);
+		SCpnt->scsi_done(SCpnt);
 		return 0;
 	}
 
-	return mptscsih_qcmd(SCpnt,done);
+	return mptscsih_qcmd(SCpnt);
 }
 
-static DEF_SCSI_QCMD(mptfc_qcmd)
-
 /*
  *	mptfc_display_port_link_speed - displaying link speed
  *	@ioc: Pointer to MPT_ADAPTER structure
diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c
index 00d339c361fc..711fcb5cec87 100644
--- a/drivers/message/fusion/mptsas.c
+++ b/drivers/message/fusion/mptsas.c
@@ -1896,7 +1896,7 @@ mptsas_slave_alloc(struct scsi_device *sdev)
 }
 
 static int
-mptsas_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
+mptsas_qcmd(struct Scsi_Host *shost, struct scsi_cmnd *SCpnt)
 {
 	MPT_SCSI_HOST	*hd;
 	MPT_ADAPTER	*ioc;
@@ -1904,11 +1904,11 @@ mptsas_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
 
 	if (!vdevice || !vdevice->vtarget || vdevice->vtarget->deleted) {
 		SCpnt->result = DID_NO_CONNECT << 16;
-		done(SCpnt);
+		SCpnt->scsi_done(SCpnt);
 		return 0;
 	}
 
-	hd = shost_priv(SCpnt->device->host);
+	hd = shost_priv(shost);
 	ioc = hd->ioc;
 
 	if (ioc->sas_discovery_quiesce_io)
@@ -1917,11 +1917,9 @@ mptsas_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
 	if (ioc->debug_level & MPT_DEBUG_SCSI)
 		scsi_print_command(SCpnt);
 
-	return mptscsih_qcmd(SCpnt,done);
+	return mptscsih_qcmd(SCpnt);
 }
 
-static DEF_SCSI_QCMD(mptsas_qcmd)
-
 /**
  *	mptsas_mptsas_eh_timed_out - resets the scsi_cmnd timeout
  *		if the device under question is currently in the
diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c
index 727819cc7034..2a1c6f21af27 100644
--- a/drivers/message/fusion/mptscsih.c
+++ b/drivers/message/fusion/mptscsih.c
@@ -1304,7 +1304,6 @@ int mptscsih_show_info(struct seq_file *m, struct Scsi_Host *host)
 /**
  *	mptscsih_qcmd - Primary Fusion MPT SCSI initiator IO start routine.
  *	@SCpnt: Pointer to scsi_cmnd structure
- *	@done: Pointer SCSI mid-layer IO completion function
  *
  *	(linux scsi_host_template.queuecommand routine)
  *	This is the primary SCSI IO start routine.  Create a MPI SCSIIORequest
@@ -1313,7 +1312,7 @@ int mptscsih_show_info(struct seq_file *m, struct Scsi_Host *host)
  *	Returns 0. (rtn value discarded by linux scsi mid-layer)
  */
 int
-mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
+mptscsih_qcmd(struct scsi_cmnd *SCpnt)
 {
 	MPT_SCSI_HOST		*hd;
 	MPT_FRAME_HDR		*mf;
@@ -1329,10 +1328,9 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
 
 	hd = shost_priv(SCpnt->device->host);
 	ioc = hd->ioc;
-	SCpnt->scsi_done = done;
 
-	dmfprintk(ioc, printk(MYIOC_s_DEBUG_FMT "qcmd: SCpnt=%p, done()=%p\n",
-		ioc->name, SCpnt, done));
+	dmfprintk(ioc, printk(MYIOC_s_DEBUG_FMT "qcmd: SCpnt=%p\n",
+		ioc->name, SCpnt));
 
 	if (ioc->taskmgmt_quiesce_io)
 		return SCSI_MLQUEUE_HOST_BUSY;
diff --git a/drivers/message/fusion/mptscsih.h b/drivers/message/fusion/mptscsih.h
index 83f503162f7a..99e3390807f3 100644
--- a/drivers/message/fusion/mptscsih.h
+++ b/drivers/message/fusion/mptscsih.h
@@ -113,7 +113,7 @@ extern int mptscsih_resume(struct pci_dev *pdev);
 #endif
 extern int mptscsih_show_info(struct seq_file *, struct Scsi_Host *);
 extern const char * mptscsih_info(struct Scsi_Host *SChost);
-extern int mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *));
+extern int mptscsih_qcmd(struct scsi_cmnd *SCpnt);
 extern int mptscsih_IssueTaskMgmt(MPT_SCSI_HOST *hd, u8 type, u8 channel,
 	u8 id, int lun, int ctx2abort, ulong timeout);
 extern void mptscsih_slave_destroy(struct scsi_device *device);
diff --git a/drivers/message/fusion/mptspi.c b/drivers/message/fusion/mptspi.c
index 5653e505f91f..49d11338294b 100644
--- a/drivers/message/fusion/mptspi.c
+++ b/drivers/message/fusion/mptspi.c
@@ -780,33 +780,31 @@ static int mptspi_slave_configure(struct scsi_device *sdev)
 }
 
 static int
-mptspi_qcmd_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *))
+mptspi_qcmd(struct Scsi_Host *shost, struct scsi_cmnd *SCpnt)
 {
-	struct _MPT_SCSI_HOST *hd = shost_priv(SCpnt->device->host);
+	struct _MPT_SCSI_HOST *hd = shost_priv(shost);
 	VirtDevice	*vdevice = SCpnt->device->hostdata;
 	MPT_ADAPTER *ioc = hd->ioc;
 
 	if (!vdevice || !vdevice->vtarget) {
 		SCpnt->result = DID_NO_CONNECT << 16;
-		done(SCpnt);
+		SCpnt->scsi_done(SCpnt);
 		return 0;
 	}
 
 	if (SCpnt->device->channel == 1 &&
 		mptscsih_is_phys_disk(ioc, 0, SCpnt->device->id) == 0) {
 		SCpnt->result = DID_NO_CONNECT << 16;
-		done(SCpnt);
+		SCpnt->scsi_done(SCpnt);
 		return 0;
 	}
 
 	if (spi_dv_pending(scsi_target(SCpnt->device)))
 		ddvprintk(ioc, scsi_print_command(SCpnt));
 
-	return mptscsih_qcmd(SCpnt,done);
+	return mptscsih_qcmd(SCpnt);
 }
 
-static DEF_SCSI_QCMD(mptspi_qcmd)
-
 static void mptspi_slave_destroy(struct scsi_device *sdev)
 {
 	struct scsi_target *starget = scsi_target(sdev);
diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index ee8204cc31e9..6cc4b6acc22a 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -760,6 +760,7 @@ config MFD_SYSCON
 config MFD_DAVINCI_VOICECODEC
 	tristate
 	select MFD_CORE
+	select REGMAP_MMIO
 
 config MFD_TI_AM335X_TSCADC
 	tristate "TI ADC / Touch Screen chip support"
@@ -1225,7 +1226,7 @@ config MFD_WM8994
 	  functionaltiy of the device other drivers must be enabled.
 
 config MFD_STW481X
-	bool "Support for ST Microelectronics STw481x"
+	tristate "Support for ST Microelectronics STw481x"
 	depends on I2C && ARCH_NOMADIK
 	select REGMAP_I2C
 	select MFD_CORE
@@ -1248,7 +1249,7 @@ config MCP_SA11X0
 
 # Chip drivers
 config MCP_UCB1200
-	bool "Support for UCB1200 / UCB1300"
+	tristate "Support for UCB1200 / UCB1300"
 	depends on MCP_SA11X0
 	select MCP
 
diff --git a/drivers/mfd/ab8500-core.c b/drivers/mfd/ab8500-core.c
index a8ee4a36a1d8..cf2e6a198c6b 100644
--- a/drivers/mfd/ab8500-core.c
+++ b/drivers/mfd/ab8500-core.c
@@ -591,7 +591,7 @@ static int ab8500_irq_init(struct ab8500 *ab8500, struct device_node *np)
 		num_irqs = AB8500_NR_IRQS;
 
 	/* If ->irq_base is zero this will give a linear mapping */
-	ab8500->domain = irq_domain_add_simple(NULL,
+	ab8500->domain = irq_domain_add_simple(ab8500->dev->of_node,
 			num_irqs, 0,
 			&ab8500_irq_ops, ab8500);
 
diff --git a/drivers/mfd/twl4030-power.c b/drivers/mfd/twl4030-power.c
index 96162b62f3c0..3bc969a5916b 100644
--- a/drivers/mfd/twl4030-power.c
+++ b/drivers/mfd/twl4030-power.c
@@ -29,13 +29,21 @@
 #include <linux/i2c/twl.h>
 #include <linux/platform_device.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/mach-types.h>
 
 static u8 twl4030_start_script_address = 0x2b;
 
-#define PWR_P1_SW_EVENTS	0x10
-#define PWR_DEVOFF		(1 << 0)
+/* Register bits for P1, P2 and P3_SW_EVENTS */
+#define PWR_STOPON_PRWON	BIT(6)
+#define PWR_STOPON_SYSEN	BIT(5)
+#define PWR_ENABLE_WARMRESET	BIT(4)
+#define PWR_LVL_WAKEUP		BIT(3)
+#define PWR_DEVACT		BIT(2)
+#define PWR_DEVSLP		BIT(1)
+#define PWR_DEVOFF		BIT(0)
+
 #define SEQ_OFFSYNC		(1 << 0)
 
 #define PHY_TO_OFF_PM_MASTER(p)		(p - 0x36)
@@ -52,10 +60,6 @@ static u8 twl4030_start_script_address = 0x2b;
 #define R_CFG_P2_TRANSITION	PHY_TO_OFF_PM_MASTER(0x37)
 #define R_CFG_P3_TRANSITION	PHY_TO_OFF_PM_MASTER(0x38)
 
-#define LVL_WAKEUP	0x08
-
-#define ENABLE_WARMRESET (1<<4)
-
 #define END_OF_SCRIPT		0x3f
 
 #define R_SEQ_ADD_A2S		PHY_TO_OFF_PM_MASTER(0x55)
@@ -125,6 +129,53 @@ static u8 res_config_addrs[] = {
 	[RES_MAIN_REF]	= 0x94,
 };
 
+/*
+ * Usable values for .remap_sleep and .remap_off
+ * Based on table "5.3.3 Resource Operating modes"
+ */
+enum {
+	TWL_REMAP_OFF = 0,
+	TWL_REMAP_SLEEP = 8,
+	TWL_REMAP_ACTIVE = 9,
+};
+
+/*
+ * Macros to configure the PM register states for various resources.
+ * Note that we can make MSG_SINGULAR etc private to this driver once
+ * omap3 has been made DT only.
+ */
+#define TWL_DFLT_DELAY		2	/* typically 2 32 KiHz cycles */
+#define TWL_DEV_GRP_P123	(DEV_GRP_P1 | DEV_GRP_P2 | DEV_GRP_P3)
+#define TWL_RESOURCE_SET(res, state)					\
+	{ MSG_SINGULAR(DEV_GRP_NULL, (res), (state)), TWL_DFLT_DELAY }
+#define TWL_RESOURCE_ON(res)	TWL_RESOURCE_SET(res, RES_STATE_ACTIVE)
+#define TWL_RESOURCE_OFF(res)	TWL_RESOURCE_SET(res, RES_STATE_OFF)
+#define TWL_RESOURCE_RESET(res)	TWL_RESOURCE_SET(res, RES_STATE_WRST)
+/*
+ * It seems that type1 and type2 is just the resource init order
+ * number for the type1 and type2 group.
+ */
+#define TWL_RESOURCE_SET_ACTIVE(res, state)			       	\
+	{ MSG_SINGULAR(DEV_GRP_NULL, (res), RES_STATE_ACTIVE), (state) }
+#define TWL_RESOURCE_GROUP_RESET(group, type1, type2)			\
+	{ MSG_BROADCAST(DEV_GRP_NULL, (group), (type1), (type2),	\
+		RES_STATE_WRST), TWL_DFLT_DELAY }
+#define TWL_RESOURCE_GROUP_SLEEP(group, type, type2)			\
+	{ MSG_BROADCAST(DEV_GRP_NULL, (group), (type), (type2),		\
+		RES_STATE_SLEEP), TWL_DFLT_DELAY }
+#define TWL_RESOURCE_GROUP_ACTIVE(group, type, type2)			\
+	{ MSG_BROADCAST(DEV_GRP_NULL, (group), (type), (type2),		\
+		RES_STATE_ACTIVE), TWL_DFLT_DELAY }
+#define TWL_REMAP_SLEEP(res, devgrp, typ, typ2)				\
+	{ .resource = (res), .devgroup = (devgrp),			\
+	  .type = (typ), .type2 = (typ2),				\
+	  .remap_off = TWL_REMAP_OFF,					\
+	  .remap_sleep = TWL_REMAP_SLEEP, }
+#define TWL_REMAP_OFF(res, devgrp, typ, typ2)				\
+	{ .resource = (res), .devgroup = (devgrp),			\
+	  .type = (typ), .type2 = (typ2),				\
+	  .remap_off = TWL_REMAP_OFF, .remap_sleep = TWL_REMAP_OFF, }
+
 static int twl4030_write_script_byte(u8 address, u8 byte)
 {
 	int err;
@@ -196,7 +247,7 @@ static int twl4030_config_wakeup3_sequence(u8 address)
 	err = twl_i2c_read_u8(TWL_MODULE_PM_MASTER, &data, R_P3_SW_EVENTS);
 	if (err)
 		goto out;
-	data |= LVL_WAKEUP;
+	data |= PWR_LVL_WAKEUP;
 	err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, data, R_P3_SW_EVENTS);
 out:
 	if (err)
@@ -219,7 +270,7 @@ static int twl4030_config_wakeup12_sequence(u8 address)
 	if (err)
 		goto out;
 
-	data |= LVL_WAKEUP;
+	data |= PWR_LVL_WAKEUP;
 	err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, data, R_P1_SW_EVENTS);
 	if (err)
 		goto out;
@@ -228,7 +279,7 @@ static int twl4030_config_wakeup12_sequence(u8 address)
 	if (err)
 		goto out;
 
-	data |= LVL_WAKEUP;
+	data |= PWR_LVL_WAKEUP;
 	err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, data, R_P2_SW_EVENTS);
 	if (err)
 		goto out;
@@ -281,7 +332,7 @@ static int twl4030_config_warmreset_sequence(u8 address)
 	if (err)
 		goto out;
 
-	rd_data |= ENABLE_WARMRESET;
+	rd_data |= PWR_ENABLE_WARMRESET;
 	err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, rd_data, R_P1_SW_EVENTS);
 	if (err)
 		goto out;
@@ -290,7 +341,7 @@ static int twl4030_config_warmreset_sequence(u8 address)
 	if (err)
 		goto out;
 
-	rd_data |= ENABLE_WARMRESET;
+	rd_data |= PWR_ENABLE_WARMRESET;
 	err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, rd_data, R_P2_SW_EVENTS);
 	if (err)
 		goto out;
@@ -299,7 +350,7 @@ static int twl4030_config_warmreset_sequence(u8 address)
 	if (err)
 		goto out;
 
-	rd_data |= ENABLE_WARMRESET;
+	rd_data |= PWR_ENABLE_WARMRESET;
 	err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, rd_data, R_P3_SW_EVENTS);
 out:
 	if (err)
@@ -421,6 +472,12 @@ static int load_twl4030_script(struct twl4030_script *tscript,
 			goto out;
 	}
 	if (tscript->flags & TWL4030_WAKEUP12_SCRIPT) {
+		/* Reset any existing sleep script to avoid hangs on reboot */
+		err = twl_i2c_write_u8(TWL_MODULE_PM_MASTER, END_OF_SCRIPT,
+				       R_SEQ_ADD_A2S);
+		if (err)
+			goto out;
+
 		err = twl4030_config_wakeup12_sequence(address);
 		if (err)
 			goto out;
@@ -493,7 +550,8 @@ int twl4030_remove_script(u8 flags)
 	return err;
 }
 
-static int twl4030_power_configure_scripts(struct twl4030_power_data *pdata)
+static int
+twl4030_power_configure_scripts(const struct twl4030_power_data *pdata)
 {
 	int err;
 	int i;
@@ -509,12 +567,34 @@ static int twl4030_power_configure_scripts(struct twl4030_power_data *pdata)
 	return 0;
 }
 
-static int twl4030_power_configure_resources(struct twl4030_power_data *pdata)
+static void twl4030_patch_rconfig(struct twl4030_resconfig *common,
+				  struct twl4030_resconfig *board)
+{
+	while (common->resource) {
+		struct twl4030_resconfig *b = board;
+
+		while (b->resource) {
+			if (b->resource == common->resource) {
+				*common = *b;
+				break;
+			}
+			b++;
+		}
+		common++;
+	}
+}
+
+static int
+twl4030_power_configure_resources(const struct twl4030_power_data *pdata)
 {
 	struct twl4030_resconfig *resconfig = pdata->resource_config;
+	struct twl4030_resconfig *boardconf = pdata->board_config;
 	int err;
 
 	if (resconfig) {
+		if (boardconf)
+			twl4030_patch_rconfig(resconfig, boardconf);
+
 		while (resconfig->resource) {
 			err = twl4030_configure_resource(resconfig);
 			if (err)
@@ -541,7 +621,7 @@ void twl4030_power_off(void)
 		pr_err("TWL4030 Unable to power off\n");
 }
 
-static bool twl4030_power_use_poweroff(struct twl4030_power_data *pdata,
+static bool twl4030_power_use_poweroff(const struct twl4030_power_data *pdata,
 					struct device_node *node)
 {
 	if (pdata && pdata->use_poweroff)
@@ -553,10 +633,170 @@ static bool twl4030_power_use_poweroff(struct twl4030_power_data *pdata,
 	return false;
 }
 
+#ifdef CONFIG_OF
+
+/* Generic warm reset configuration for omap3 */
+
+static struct twl4030_ins omap3_wrst_seq[] = {
+	TWL_RESOURCE_OFF(RES_NRES_PWRON),
+	TWL_RESOURCE_OFF(RES_RESET),
+	TWL_RESOURCE_RESET(RES_MAIN_REF),
+	TWL_RESOURCE_GROUP_RESET(RES_GRP_ALL, RES_TYPE_R0, RES_TYPE2_R2),
+	TWL_RESOURCE_RESET(RES_VUSB_3V1),
+	TWL_RESOURCE_GROUP_RESET(RES_GRP_ALL, RES_TYPE_R0, RES_TYPE2_R1),
+	TWL_RESOURCE_GROUP_RESET(RES_GRP_RC, RES_TYPE_ALL, RES_TYPE2_R0),
+	TWL_RESOURCE_ON(RES_RESET),
+	TWL_RESOURCE_ON(RES_NRES_PWRON),
+};
+
+static struct twl4030_script omap3_wrst_script = {
+	.script	= omap3_wrst_seq,
+	.size	= ARRAY_SIZE(omap3_wrst_seq),
+	.flags	= TWL4030_WRST_SCRIPT,
+};
+
+static struct twl4030_script *omap3_reset_scripts[] = {
+	&omap3_wrst_script,
+};
+
+static struct twl4030_resconfig omap3_rconfig[] = {
+	TWL_REMAP_SLEEP(RES_HFCLKOUT, DEV_GRP_P3, -1, -1),
+	TWL_REMAP_SLEEP(RES_VDD1, DEV_GRP_P1, -1, -1),
+	TWL_REMAP_SLEEP(RES_VDD2, DEV_GRP_P1, -1, -1),
+	{ 0, 0 },
+};
+
+static struct twl4030_power_data omap3_reset = {
+	.scripts		= omap3_reset_scripts,
+	.num			= ARRAY_SIZE(omap3_reset_scripts),
+	.resource_config	= omap3_rconfig,
+};
+
+/* Recommended generic default idle configuration for off-idle */
+
+/* Broadcast message to put res to sleep */
+static struct twl4030_ins omap3_idle_sleep_on_seq[] = {
+	TWL_RESOURCE_GROUP_SLEEP(RES_GRP_ALL, RES_TYPE_ALL, 0),
+};
+
+static struct twl4030_script omap3_idle_sleep_on_script = {
+	.script	= omap3_idle_sleep_on_seq,
+	.size	= ARRAY_SIZE(omap3_idle_sleep_on_seq),
+	.flags	= TWL4030_SLEEP_SCRIPT,
+};
+
+/* Broadcast message to put res to active */
+static struct twl4030_ins omap3_idle_wakeup_p12_seq[] = {
+	TWL_RESOURCE_GROUP_ACTIVE(RES_GRP_ALL, RES_TYPE_ALL, 0),
+};
+
+static struct twl4030_script omap3_idle_wakeup_p12_script = {
+	.script	= omap3_idle_wakeup_p12_seq,
+	.size	= ARRAY_SIZE(omap3_idle_wakeup_p12_seq),
+	.flags	= TWL4030_WAKEUP12_SCRIPT,
+};
+
+/* Broadcast message to put res to active */
+static struct twl4030_ins omap3_idle_wakeup_p3_seq[] = {
+	TWL_RESOURCE_SET_ACTIVE(RES_CLKEN, 0x37),
+	TWL_RESOURCE_GROUP_ACTIVE(RES_GRP_ALL, RES_TYPE_ALL, 0),
+};
+
+static struct twl4030_script omap3_idle_wakeup_p3_script = {
+	.script	= omap3_idle_wakeup_p3_seq,
+	.size	= ARRAY_SIZE(omap3_idle_wakeup_p3_seq),
+	.flags	= TWL4030_WAKEUP3_SCRIPT,
+};
+
+static struct twl4030_script *omap3_idle_scripts[] = {
+	&omap3_idle_wakeup_p12_script,
+	&omap3_idle_wakeup_p3_script,
+	&omap3_wrst_script,
+	&omap3_idle_sleep_on_script,
+};
+
+/*
+ * Recommended configuration based on "Recommended Sleep
+ * Sequences for the Zoom Platform":
+ * http://omappedia.com/wiki/File:Recommended_Sleep_Sequences_Zoom.pdf
+ * Note that the type1 and type2 seem to be just the init order number
+ * for type1 and type2 groups as specified in the document mentioned
+ * above.
+ */
+static struct twl4030_resconfig omap3_idle_rconfig[] = {
+	TWL_REMAP_SLEEP(RES_VAUX1, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VAUX2, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VAUX3, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VAUX4, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VMMC1, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VMMC2, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_OFF(RES_VPLL1, DEV_GRP_P1, 3, 1),
+	TWL_REMAP_SLEEP(RES_VPLL2, DEV_GRP_P1, 0, 0),
+	TWL_REMAP_SLEEP(RES_VSIM, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VDAC, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VINTANA1, TWL_DEV_GRP_P123, 1, 2),
+	TWL_REMAP_SLEEP(RES_VINTANA2, TWL_DEV_GRP_P123, 0, 2),
+	TWL_REMAP_SLEEP(RES_VINTDIG, TWL_DEV_GRP_P123, 1, 2),
+	TWL_REMAP_SLEEP(RES_VIO, TWL_DEV_GRP_P123, 2, 2),
+	TWL_REMAP_OFF(RES_VDD1, DEV_GRP_P1, 4, 1),
+	TWL_REMAP_OFF(RES_VDD2, DEV_GRP_P1, 3, 1),
+	TWL_REMAP_SLEEP(RES_VUSB_1V5, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VUSB_1V8, DEV_GRP_NULL, 0, 0),
+	TWL_REMAP_SLEEP(RES_VUSB_3V1, TWL_DEV_GRP_P123, 0, 0),
+	/* Resource #20 USB charge pump skipped */
+	TWL_REMAP_SLEEP(RES_REGEN, TWL_DEV_GRP_P123, 2, 1),
+	TWL_REMAP_SLEEP(RES_NRES_PWRON, TWL_DEV_GRP_P123, 0, 1),
+	TWL_REMAP_SLEEP(RES_CLKEN, TWL_DEV_GRP_P123, 3, 2),
+	TWL_REMAP_SLEEP(RES_SYSEN, TWL_DEV_GRP_P123, 6, 1),
+	TWL_REMAP_SLEEP(RES_HFCLKOUT, DEV_GRP_P3, 0, 2),
+	TWL_REMAP_SLEEP(RES_32KCLKOUT, TWL_DEV_GRP_P123, 0, 0),
+	TWL_REMAP_SLEEP(RES_RESET, TWL_DEV_GRP_P123, 6, 0),
+	TWL_REMAP_SLEEP(RES_MAIN_REF, TWL_DEV_GRP_P123, 0, 0),
+	{ /* Terminator */ },
+};
+
+static struct twl4030_power_data omap3_idle = {
+	.scripts		= omap3_idle_scripts,
+	.num			= ARRAY_SIZE(omap3_idle_scripts),
+	.resource_config	= omap3_idle_rconfig,
+};
+
+/* Disable 32 KiHz oscillator during idle */
+static struct twl4030_resconfig osc_off_rconfig[] = {
+	TWL_REMAP_OFF(RES_CLKEN, DEV_GRP_P1 | DEV_GRP_P3, 3, 2),
+	{ /* Terminator */ },
+};
+
+static struct twl4030_power_data osc_off_idle = {
+	.scripts		= omap3_idle_scripts,
+	.num			= ARRAY_SIZE(omap3_idle_scripts),
+	.resource_config	= omap3_idle_rconfig,
+	.board_config		= osc_off_rconfig,
+};
+
+static struct of_device_id twl4030_power_of_match[] = {
+	{
+		.compatible = "ti,twl4030-power-reset",
+		.data = &omap3_reset,
+	},
+	{
+		.compatible = "ti,twl4030-power-idle",
+		.data = &omap3_idle,
+	},
+	{
+		.compatible = "ti,twl4030-power-idle-osc-off",
+		.data = &osc_off_idle,
+	},
+	{ },
+};
+MODULE_DEVICE_TABLE(of, twl4030_power_of_match);
+#endif	/* CONFIG_OF */
+
 static int twl4030_power_probe(struct platform_device *pdev)
 {
-	struct twl4030_power_data *pdata = dev_get_platdata(&pdev->dev);
+	const struct twl4030_power_data *pdata = dev_get_platdata(&pdev->dev);
 	struct device_node *node = pdev->dev.of_node;
+	const struct of_device_id *match;
 	int err = 0;
 	int err2 = 0;
 	u8 val;
@@ -577,8 +817,12 @@ static int twl4030_power_probe(struct platform_device *pdev)
 		return err;
 	}
 
+	match = of_match_device(of_match_ptr(twl4030_power_of_match),
+				&pdev->dev);
+	if (match && match->data)
+		pdata = match->data;
+
 	if (pdata) {
-		/* TODO: convert to device tree */
 		err = twl4030_power_configure_scripts(pdata);
 		if (err) {
 			pr_err("TWL4030 failed to load scripts\n");
@@ -628,14 +872,6 @@ static int twl4030_power_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_OF
-static const struct of_device_id twl4030_power_of_match[] = {
-	{.compatible = "ti,twl4030-power", },
-	{ },
-};
-MODULE_DEVICE_TABLE(of, twl4030_power_of_match);
-#endif
-
 static struct platform_driver twl4030_power_driver = {
 	.driver = {
 		.name	= "twl4030_power",
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index a43d0c467274..ee9402324a23 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -54,7 +54,7 @@ config AD525X_DPOT_SPI
 config ATMEL_PWM
 	tristate "Atmel AT32/AT91 PWM support"
 	depends on HAVE_CLK
-	depends on AVR32 || AT91SAM9263 || AT91SAM9RL || AT91SAM9G45
+	depends on AVR32 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91SAM9G45
 	help
 	  This option enables device driver support for the PWM channels
 	  on certain Atmel processors.  Pulse Width Modulation is used for
diff --git a/drivers/misc/vexpress-syscfg.c b/drivers/misc/vexpress-syscfg.c
index 73068e50e56d..3250fc1df0aa 100644
--- a/drivers/misc/vexpress-syscfg.c
+++ b/drivers/misc/vexpress-syscfg.c
@@ -199,7 +199,7 @@ static struct regmap *vexpress_syscfg_regmap_init(struct device *dev,
 	func = kzalloc(sizeof(*func) + sizeof(*func->template) * num,
 			GFP_KERNEL);
 	if (!func)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	func->syscfg = syscfg;
 	func->num_templates = num;
@@ -231,10 +231,14 @@ static struct regmap *vexpress_syscfg_regmap_init(struct device *dev,
 	func->regmap = regmap_init(dev, NULL, func,
 			&vexpress_syscfg_regmap_config);
 
-	if (IS_ERR(func->regmap))
+	if (IS_ERR(func->regmap)) {
+		void *err = func->regmap;
+
 		kfree(func);
-	else
-		list_add(&func->list, &syscfg->funcs);
+		return err;
+	}
+
+	list_add(&func->list, &syscfg->funcs);
 
 	return func->regmap;
 }
diff --git a/drivers/misc/vmw_balloon.c b/drivers/misc/vmw_balloon.c
index 2421835d5daf..191617492181 100644
--- a/drivers/misc/vmw_balloon.c
+++ b/drivers/misc/vmw_balloon.c
@@ -17,7 +17,8 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  *
- * Maintained by: Dmitry Torokhov <dtor@vmware.com>
+ * Maintained by:	Xavier Deguillard <xdeguillard@vmware.com>
+ *			Philip Moltmann <moltmann@vmware.com>
  */
 
 /*
diff --git a/drivers/mmc/core/bus.c b/drivers/mmc/core/bus.c
index 824644875d41..d2dbf02022bd 100644
--- a/drivers/mmc/core/bus.c
+++ b/drivers/mmc/core/bus.c
@@ -341,16 +341,17 @@ int mmc_add_card(struct mmc_card *card)
 	if (mmc_host_is_spi(card->host)) {
 		pr_info("%s: new %s%s%s card on SPI\n",
 			mmc_hostname(card->host),
-			mmc_card_highspeed(card) ? "high speed " : "",
-			mmc_card_ddr_mode(card) ? "DDR " : "",
+			mmc_card_hs(card) ? "high speed " : "",
+			mmc_card_ddr52(card) ? "DDR " : "",
 			type);
 	} else {
 		pr_info("%s: new %s%s%s%s%s card at address %04x\n",
 			mmc_hostname(card->host),
 			mmc_card_uhs(card) ? "ultra high speed " :
-			(mmc_card_highspeed(card) ? "high speed " : ""),
+			(mmc_card_hs(card) ? "high speed " : ""),
+			mmc_card_hs400(card) ? "HS400 " :
 			(mmc_card_hs200(card) ? "HS200 " : ""),
-			mmc_card_ddr_mode(card) ? "DDR " : "",
+			mmc_card_ddr52(card) ? "DDR " : "",
 			uhs_bus_speed_mode, type, card->rca);
 	}
 
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index acbc3f2aaaf9..7dc0c85fdb60 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -800,6 +800,10 @@ void mmc_set_data_timeout(struct mmc_data *data, const struct mmc_card *card)
 			data->timeout_ns = limit_us * 1000;
 			data->timeout_clks = 0;
 		}
+
+		/* assign limit value if invalid */
+		if (timeout_us == 0)
+			data->timeout_ns = limit_us * 1000;
 	}
 
 	/*
@@ -1310,31 +1314,38 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
 }
 EXPORT_SYMBOL_GPL(mmc_regulator_set_ocr);
 
+#endif /* CONFIG_REGULATOR */
+
 int mmc_regulator_get_supply(struct mmc_host *mmc)
 {
 	struct device *dev = mmc_dev(mmc);
-	struct regulator *supply;
 	int ret;
 
-	supply = devm_regulator_get(dev, "vmmc");
-	mmc->supply.vmmc = supply;
+	mmc->supply.vmmc = devm_regulator_get_optional(dev, "vmmc");
 	mmc->supply.vqmmc = devm_regulator_get_optional(dev, "vqmmc");
 
-	if (IS_ERR(supply))
-		return PTR_ERR(supply);
+	if (IS_ERR(mmc->supply.vmmc)) {
+		if (PTR_ERR(mmc->supply.vmmc) == -EPROBE_DEFER)
+			return -EPROBE_DEFER;
+		dev_info(dev, "No vmmc regulator found\n");
+	} else {
+		ret = mmc_regulator_get_ocrmask(mmc->supply.vmmc);
+		if (ret > 0)
+			mmc->ocr_avail = ret;
+		else
+			dev_warn(dev, "Failed getting OCR mask: %d\n", ret);
+	}
 
-	ret = mmc_regulator_get_ocrmask(supply);
-	if (ret > 0)
-		mmc->ocr_avail = ret;
-	else
-		dev_warn(mmc_dev(mmc), "Failed getting OCR mask: %d\n", ret);
+	if (IS_ERR(mmc->supply.vqmmc)) {
+		if (PTR_ERR(mmc->supply.vqmmc) == -EPROBE_DEFER)
+			return -EPROBE_DEFER;
+		dev_info(dev, "No vqmmc regulator found\n");
+	}
 
 	return 0;
 }
 EXPORT_SYMBOL_GPL(mmc_regulator_get_supply);
 
-#endif /* CONFIG_REGULATOR */
-
 /*
  * Mask off any voltages we don't support and select
  * the lowest voltage
@@ -1533,8 +1544,13 @@ void mmc_power_up(struct mmc_host *host, u32 ocr)
 	host->ios.timing = MMC_TIMING_LEGACY;
 	mmc_set_ios(host);
 
-	/* Set signal voltage to 3.3V */
-	__mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_330);
+	/* Try to set signal voltage to 3.3V but fall back to 1.8v or 1.2v */
+	if (__mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_330) == 0)
+		dev_dbg(mmc_dev(host), "Initial signal voltage of 3.3v\n");
+	else if (__mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180) == 0)
+		dev_dbg(mmc_dev(host), "Initial signal voltage of 1.8v\n");
+	else if (__mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_120) == 0)
+		dev_dbg(mmc_dev(host), "Initial signal voltage of 1.2v\n");
 
 	/*
 	 * This delay should be sufficient to allow the power supply
@@ -2183,7 +2199,7 @@ int mmc_set_blocklen(struct mmc_card *card, unsigned int blocklen)
 {
 	struct mmc_command cmd = {0};
 
-	if (mmc_card_blockaddr(card) || mmc_card_ddr_mode(card))
+	if (mmc_card_blockaddr(card) || mmc_card_ddr52(card))
 		return 0;
 
 	cmd.opcode = MMC_SET_BLOCKLEN;
@@ -2263,7 +2279,6 @@ static int mmc_do_hw_reset(struct mmc_host *host, int check)
 		}
 	}
 
-	host->card->state &= ~(MMC_STATE_HIGHSPEED | MMC_STATE_HIGHSPEED_DDR);
 	if (mmc_host_is_spi(host)) {
 		host->ios.chip_select = MMC_CS_HIGH;
 		host->ios.bus_mode = MMC_BUSMODE_PUSHPULL;
@@ -2403,6 +2418,11 @@ void mmc_rescan(struct work_struct *work)
 		container_of(work, struct mmc_host, detect.work);
 	int i;
 
+	if (host->trigger_card_event && host->ops->card_event) {
+		host->ops->card_event(host);
+		host->trigger_card_event = false;
+	}
+
 	if (host->rescan_disable)
 		return;
 
diff --git a/drivers/mmc/core/debugfs.c b/drivers/mmc/core/debugfs.c
index 54829c0ed000..91eb16223246 100644
--- a/drivers/mmc/core/debugfs.c
+++ b/drivers/mmc/core/debugfs.c
@@ -135,8 +135,14 @@ static int mmc_ios_show(struct seq_file *s, void *data)
 	case MMC_TIMING_UHS_DDR50:
 		str = "sd uhs DDR50";
 		break;
+	case MMC_TIMING_MMC_DDR52:
+		str = "mmc DDR52";
+		break;
 	case MMC_TIMING_MMC_HS200:
-		str = "mmc high-speed SDR200";
+		str = "mmc HS200";
+		break;
+	case MMC_TIMING_MMC_HS400:
+		str = "mmc HS400";
 		break;
 	default:
 		str = "invalid";
diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
index fdea825dbb24..95cceae96944 100644
--- a/drivers/mmc/core/host.c
+++ b/drivers/mmc/core/host.c
@@ -447,6 +447,10 @@ int mmc_of_parse(struct mmc_host *host)
 		host->caps2 |= MMC_CAP2_HS200_1_8V_SDR;
 	if (of_find_property(np, "mmc-hs200-1_2v", &len))
 		host->caps2 |= MMC_CAP2_HS200_1_2V_SDR;
+	if (of_find_property(np, "mmc-hs400-1_8v", &len))
+		host->caps2 |= MMC_CAP2_HS400_1_8V | MMC_CAP2_HS200_1_8V_SDR;
+	if (of_find_property(np, "mmc-hs400-1_2v", &len))
+		host->caps2 |= MMC_CAP2_HS400_1_2V | MMC_CAP2_HS200_1_2V_SDR;
 
 	return 0;
 
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 1ab5f3a0af5b..793c6f7ddb04 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -240,31 +240,62 @@ static int mmc_get_ext_csd(struct mmc_card *card, u8 **new_ext_csd)
 static void mmc_select_card_type(struct mmc_card *card)
 {
 	struct mmc_host *host = card->host;
-	u8 card_type = card->ext_csd.raw_card_type & EXT_CSD_CARD_TYPE_MASK;
+	u8 card_type = card->ext_csd.raw_card_type;
 	u32 caps = host->caps, caps2 = host->caps2;
-	unsigned int hs_max_dtr = 0;
+	unsigned int hs_max_dtr = 0, hs200_max_dtr = 0;
+	unsigned int avail_type = 0;
 
-	if (card_type & EXT_CSD_CARD_TYPE_26)
+	if (caps & MMC_CAP_MMC_HIGHSPEED &&
+	    card_type & EXT_CSD_CARD_TYPE_HS_26) {
 		hs_max_dtr = MMC_HIGH_26_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_HS_26;
+	}
 
 	if (caps & MMC_CAP_MMC_HIGHSPEED &&
-			card_type & EXT_CSD_CARD_TYPE_52)
+	    card_type & EXT_CSD_CARD_TYPE_HS_52) {
 		hs_max_dtr = MMC_HIGH_52_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_HS_52;
+	}
+
+	if (caps & MMC_CAP_1_8V_DDR &&
+	    card_type & EXT_CSD_CARD_TYPE_DDR_1_8V) {
+		hs_max_dtr = MMC_HIGH_DDR_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_DDR_1_8V;
+	}
 
-	if ((caps & MMC_CAP_1_8V_DDR &&
-			card_type & EXT_CSD_CARD_TYPE_DDR_1_8V) ||
-	    (caps & MMC_CAP_1_2V_DDR &&
-			card_type & EXT_CSD_CARD_TYPE_DDR_1_2V))
+	if (caps & MMC_CAP_1_2V_DDR &&
+	    card_type & EXT_CSD_CARD_TYPE_DDR_1_2V) {
 		hs_max_dtr = MMC_HIGH_DDR_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_DDR_1_2V;
+	}
+
+	if (caps2 & MMC_CAP2_HS200_1_8V_SDR &&
+	    card_type & EXT_CSD_CARD_TYPE_HS200_1_8V) {
+		hs200_max_dtr = MMC_HS200_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_HS200_1_8V;
+	}
+
+	if (caps2 & MMC_CAP2_HS200_1_2V_SDR &&
+	    card_type & EXT_CSD_CARD_TYPE_HS200_1_2V) {
+		hs200_max_dtr = MMC_HS200_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_HS200_1_2V;
+	}
+
+	if (caps2 & MMC_CAP2_HS400_1_8V &&
+	    card_type & EXT_CSD_CARD_TYPE_HS400_1_8V) {
+		hs200_max_dtr = MMC_HS200_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_HS400_1_8V;
+	}
 
-	if ((caps2 & MMC_CAP2_HS200_1_8V_SDR &&
-			card_type & EXT_CSD_CARD_TYPE_SDR_1_8V) ||
-	    (caps2 & MMC_CAP2_HS200_1_2V_SDR &&
-			card_type & EXT_CSD_CARD_TYPE_SDR_1_2V))
-		hs_max_dtr = MMC_HS200_MAX_DTR;
+	if (caps2 & MMC_CAP2_HS400_1_2V &&
+	    card_type & EXT_CSD_CARD_TYPE_HS400_1_2V) {
+		hs200_max_dtr = MMC_HS200_MAX_DTR;
+		avail_type |= EXT_CSD_CARD_TYPE_HS400_1_2V;
+	}
 
 	card->ext_csd.hs_max_dtr = hs_max_dtr;
-	card->ext_csd.card_type = card_type;
+	card->ext_csd.hs200_max_dtr = hs200_max_dtr;
+	card->mmc_avail_type = avail_type;
 }
 
 /*
@@ -480,6 +511,8 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd)
 			ext_csd[EXT_CSD_PWR_CL_DDR_52_195];
 		card->ext_csd.raw_pwr_cl_ddr_52_360 =
 			ext_csd[EXT_CSD_PWR_CL_DDR_52_360];
+		card->ext_csd.raw_pwr_cl_ddr_200_360 =
+			ext_csd[EXT_CSD_PWR_CL_DDR_200_360];
 	}
 
 	if (card->ext_csd.rev >= 5) {
@@ -646,7 +679,10 @@ static int mmc_compare_ext_csds(struct mmc_card *card, unsigned bus_width)
 		(card->ext_csd.raw_pwr_cl_ddr_52_195 ==
 			bw_ext_csd[EXT_CSD_PWR_CL_DDR_52_195]) &&
 		(card->ext_csd.raw_pwr_cl_ddr_52_360 ==
-			bw_ext_csd[EXT_CSD_PWR_CL_DDR_52_360]));
+			bw_ext_csd[EXT_CSD_PWR_CL_DDR_52_360]) &&
+		(card->ext_csd.raw_pwr_cl_ddr_200_360 ==
+			bw_ext_csd[EXT_CSD_PWR_CL_DDR_200_360]));
+
 	if (err)
 		err = -EINVAL;
 
@@ -694,18 +730,10 @@ static struct attribute *mmc_std_attrs[] = {
 	&dev_attr_rel_sectors.attr,
 	NULL,
 };
-
-static struct attribute_group mmc_std_attr_group = {
-	.attrs = mmc_std_attrs,
-};
-
-static const struct attribute_group *mmc_attr_groups[] = {
-	&mmc_std_attr_group,
-	NULL,
-};
+ATTRIBUTE_GROUPS(mmc_std);
 
 static struct device_type mmc_type = {
-	.groups = mmc_attr_groups,
+	.groups = mmc_std_groups,
 };
 
 /*
@@ -714,17 +742,13 @@ static struct device_type mmc_type = {
  * extended CSD register, select it by executing the
  * mmc_switch command.
  */
-static int mmc_select_powerclass(struct mmc_card *card,
-		unsigned int bus_width)
+static int __mmc_select_powerclass(struct mmc_card *card,
+				   unsigned int bus_width)
 {
-	int err = 0;
+	struct mmc_host *host = card->host;
+	struct mmc_ext_csd *ext_csd = &card->ext_csd;
 	unsigned int pwrclass_val = 0;
-	struct mmc_host *host;
-
-	BUG_ON(!card);
-
-	host = card->host;
-	BUG_ON(!host);
+	int err = 0;
 
 	/* Power class selection is supported for versions >= 4.0 */
 	if (card->csd.mmca_vsn < CSD_SPEC_VER_4)
@@ -736,14 +760,14 @@ static int mmc_select_powerclass(struct mmc_card *card,
 
 	switch (1 << host->ios.vdd) {
 	case MMC_VDD_165_195:
-		if (host->ios.clock <= 26000000)
-			pwrclass_val = card->ext_csd.raw_pwr_cl_26_195;
-		else if	(host->ios.clock <= 52000000)
+		if (host->ios.clock <= MMC_HIGH_26_MAX_DTR)
+			pwrclass_val = ext_csd->raw_pwr_cl_26_195;
+		else if (host->ios.clock <= MMC_HIGH_52_MAX_DTR)
 			pwrclass_val = (bus_width <= EXT_CSD_BUS_WIDTH_8) ?
-				card->ext_csd.raw_pwr_cl_52_195 :
-				card->ext_csd.raw_pwr_cl_ddr_52_195;
-		else if (host->ios.clock <= 200000000)
-			pwrclass_val = card->ext_csd.raw_pwr_cl_200_195;
+				ext_csd->raw_pwr_cl_52_195 :
+				ext_csd->raw_pwr_cl_ddr_52_195;
+		else if (host->ios.clock <= MMC_HS200_MAX_DTR)
+			pwrclass_val = ext_csd->raw_pwr_cl_200_195;
 		break;
 	case MMC_VDD_27_28:
 	case MMC_VDD_28_29:
@@ -754,14 +778,16 @@ static int mmc_select_powerclass(struct mmc_card *card,
 	case MMC_VDD_33_34:
 	case MMC_VDD_34_35:
 	case MMC_VDD_35_36:
-		if (host->ios.clock <= 26000000)
-			pwrclass_val = card->ext_csd.raw_pwr_cl_26_360;
-		else if	(host->ios.clock <= 52000000)
+		if (host->ios.clock <= MMC_HIGH_26_MAX_DTR)
+			pwrclass_val = ext_csd->raw_pwr_cl_26_360;
+		else if (host->ios.clock <= MMC_HIGH_52_MAX_DTR)
 			pwrclass_val = (bus_width <= EXT_CSD_BUS_WIDTH_8) ?
-				card->ext_csd.raw_pwr_cl_52_360 :
-				card->ext_csd.raw_pwr_cl_ddr_52_360;
-		else if (host->ios.clock <= 200000000)
-			pwrclass_val = card->ext_csd.raw_pwr_cl_200_360;
+				ext_csd->raw_pwr_cl_52_360 :
+				ext_csd->raw_pwr_cl_ddr_52_360;
+		else if (host->ios.clock <= MMC_HS200_MAX_DTR)
+			pwrclass_val = (bus_width == EXT_CSD_DDR_BUS_WIDTH_8) ?
+				ext_csd->raw_pwr_cl_ddr_200_360 :
+				ext_csd->raw_pwr_cl_200_360;
 		break;
 	default:
 		pr_warning("%s: Voltage range not supported "
@@ -787,40 +813,79 @@ static int mmc_select_powerclass(struct mmc_card *card,
 	return err;
 }
 
+static int mmc_select_powerclass(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	u32 bus_width, ext_csd_bits;
+	int err, ddr;
+
+	/* Power class selection is supported for versions >= 4.0 */
+	if (card->csd.mmca_vsn < CSD_SPEC_VER_4)
+		return 0;
+
+	bus_width = host->ios.bus_width;
+	/* Power class values are defined only for 4/8 bit bus */
+	if (bus_width == MMC_BUS_WIDTH_1)
+		return 0;
+
+	ddr = card->mmc_avail_type & EXT_CSD_CARD_TYPE_DDR_52;
+	if (ddr)
+		ext_csd_bits = (bus_width == MMC_BUS_WIDTH_8) ?
+			EXT_CSD_DDR_BUS_WIDTH_8 : EXT_CSD_DDR_BUS_WIDTH_4;
+	else
+		ext_csd_bits = (bus_width == MMC_BUS_WIDTH_8) ?
+			EXT_CSD_BUS_WIDTH_8 :  EXT_CSD_BUS_WIDTH_4;
+
+	err = __mmc_select_powerclass(card, ext_csd_bits);
+	if (err)
+		pr_warn("%s: power class selection to bus width %d ddr %d failed\n",
+			mmc_hostname(host), 1 << bus_width, ddr);
+
+	return err;
+}
+
 /*
- * Selects the desired buswidth and switch to the HS200 mode
- * if bus width set without error
+ * Set the bus speed for the selected speed mode.
  */
-static int mmc_select_hs200(struct mmc_card *card)
+static void mmc_set_bus_speed(struct mmc_card *card)
+{
+	unsigned int max_dtr = (unsigned int)-1;
+
+	if ((mmc_card_hs200(card) || mmc_card_hs400(card)) &&
+	     max_dtr > card->ext_csd.hs200_max_dtr)
+		max_dtr = card->ext_csd.hs200_max_dtr;
+	else if (mmc_card_hs(card) && max_dtr > card->ext_csd.hs_max_dtr)
+		max_dtr = card->ext_csd.hs_max_dtr;
+	else if (max_dtr > card->csd.max_dtr)
+		max_dtr = card->csd.max_dtr;
+
+	mmc_set_clock(card->host, max_dtr);
+}
+
+/*
+ * Select the bus width amoung 4-bit and 8-bit(SDR).
+ * If the bus width is changed successfully, return the selected width value.
+ * Zero is returned instead of error value if the wide width is not supported.
+ */
+static int mmc_select_bus_width(struct mmc_card *card)
 {
-	int idx, err = -EINVAL;
-	struct mmc_host *host;
 	static unsigned ext_csd_bits[] = {
-		EXT_CSD_BUS_WIDTH_4,
 		EXT_CSD_BUS_WIDTH_8,
+		EXT_CSD_BUS_WIDTH_4,
 	};
 	static unsigned bus_widths[] = {
-		MMC_BUS_WIDTH_4,
 		MMC_BUS_WIDTH_8,
+		MMC_BUS_WIDTH_4,
 	};
+	struct mmc_host *host = card->host;
+	unsigned idx, bus_width = 0;
+	int err = 0;
 
-	BUG_ON(!card);
-
-	host = card->host;
-
-	if (card->ext_csd.card_type & EXT_CSD_CARD_TYPE_SDR_1_2V &&
-			host->caps2 & MMC_CAP2_HS200_1_2V_SDR)
-		err = __mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_120);
-
-	if (err && card->ext_csd.card_type & EXT_CSD_CARD_TYPE_SDR_1_8V &&
-			host->caps2 & MMC_CAP2_HS200_1_8V_SDR)
-		err = __mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180);
-
-	/* If fails try again during next card power cycle */
-	if (err)
-		goto err;
+	if ((card->csd.mmca_vsn < CSD_SPEC_VER_4) &&
+	    !(host->caps & (MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA)))
+		return 0;
 
-	idx = (host->caps & MMC_CAP_8_BIT_DATA) ? 1 : 0;
+	idx = (host->caps & MMC_CAP_8_BIT_DATA) ? 0 : 1;
 
 	/*
 	 * Unlike SD, MMC cards dont have a configuration register to notify
@@ -828,8 +893,7 @@ static int mmc_select_hs200(struct mmc_card *card)
 	 * the supported bus width or compare the ext csd values of current
 	 * bus width and ext csd values of 1 bit mode read earlier.
 	 */
-	for (; idx >= 0; idx--) {
-
+	for (; idx < ARRAY_SIZE(bus_widths); idx++) {
 		/*
 		 * Host is capable of 8bit transfer, then switch
 		 * the device to work in 8bit transfer mode. If the
@@ -844,27 +908,266 @@ static int mmc_select_hs200(struct mmc_card *card)
 		if (err)
 			continue;
 
-		mmc_set_bus_width(card->host, bus_widths[idx]);
+		bus_width = bus_widths[idx];
+		mmc_set_bus_width(host, bus_width);
 
+		/*
+		 * If controller can't handle bus width test,
+		 * compare ext_csd previously read in 1 bit mode
+		 * against ext_csd at new bus width
+		 */
 		if (!(host->caps & MMC_CAP_BUS_WIDTH_TEST))
-			err = mmc_compare_ext_csds(card, bus_widths[idx]);
+			err = mmc_compare_ext_csds(card, bus_width);
 		else
-			err = mmc_bus_test(card, bus_widths[idx]);
-		if (!err)
+			err = mmc_bus_test(card, bus_width);
+
+		if (!err) {
+			err = bus_width;
 			break;
+		} else {
+			pr_warn("%s: switch to bus width %d failed\n",
+				mmc_hostname(host), ext_csd_bits[idx]);
+		}
 	}
 
-	/* switch to HS200 mode if bus width set successfully */
+	return err;
+}
+
+/*
+ * Switch to the high-speed mode
+ */
+static int mmc_select_hs(struct mmc_card *card)
+{
+	int err;
+
+	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+			   EXT_CSD_HS_TIMING, EXT_CSD_TIMING_HS,
+			   card->ext_csd.generic_cmd6_time,
+			   true, true, true);
 	if (!err)
+		mmc_set_timing(card->host, MMC_TIMING_MMC_HS);
+
+	return err;
+}
+
+/*
+ * Activate wide bus and DDR if supported.
+ */
+static int mmc_select_hs_ddr(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	u32 bus_width, ext_csd_bits;
+	int err = 0;
+
+	if (!(card->mmc_avail_type & EXT_CSD_CARD_TYPE_DDR_52))
+		return 0;
+
+	bus_width = host->ios.bus_width;
+	if (bus_width == MMC_BUS_WIDTH_1)
+		return 0;
+
+	ext_csd_bits = (bus_width == MMC_BUS_WIDTH_8) ?
+		EXT_CSD_DDR_BUS_WIDTH_8 : EXT_CSD_DDR_BUS_WIDTH_4;
+
+	err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+			EXT_CSD_BUS_WIDTH,
+			ext_csd_bits,
+			card->ext_csd.generic_cmd6_time);
+	if (err) {
+		pr_warn("%s: switch to bus width %d ddr failed\n",
+			mmc_hostname(host), 1 << bus_width);
+		return err;
+	}
+
+	/*
+	 * eMMC cards can support 3.3V to 1.2V i/o (vccq)
+	 * signaling.
+	 *
+	 * EXT_CSD_CARD_TYPE_DDR_1_8V means 3.3V or 1.8V vccq.
+	 *
+	 * 1.8V vccq at 3.3V core voltage (vcc) is not required
+	 * in the JEDEC spec for DDR.
+	 *
+	 * Do not force change in vccq since we are obviously
+	 * working and no change to vccq is needed.
+	 *
+	 * WARNING: eMMC rules are NOT the same as SD DDR
+	 */
+	if (card->mmc_avail_type & EXT_CSD_CARD_TYPE_DDR_1_2V) {
+		err = __mmc_set_signal_voltage(host,
+				MMC_SIGNAL_VOLTAGE_120);
+		if (err)
+			return err;
+	}
+
+	mmc_set_timing(host, MMC_TIMING_MMC_DDR52);
+
+	return err;
+}
+
+static int mmc_select_hs400(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	int err = 0;
+
+	/*
+	 * HS400 mode requires 8-bit bus width
+	 */
+	if (!(card->mmc_avail_type & EXT_CSD_CARD_TYPE_HS400 &&
+	      host->ios.bus_width == MMC_BUS_WIDTH_8))
+		return 0;
+
+	/*
+	 * Before switching to dual data rate operation for HS400,
+	 * it is required to convert from HS200 mode to HS mode.
+	 */
+	mmc_set_timing(card->host, MMC_TIMING_MMC_HS);
+	mmc_set_bus_speed(card);
+
+	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+			   EXT_CSD_HS_TIMING, EXT_CSD_TIMING_HS,
+			   card->ext_csd.generic_cmd6_time,
+			   true, true, true);
+	if (err) {
+		pr_warn("%s: switch to high-speed from hs200 failed, err:%d\n",
+			mmc_hostname(host), err);
+		return err;
+	}
+
+	err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+			 EXT_CSD_BUS_WIDTH,
+			 EXT_CSD_DDR_BUS_WIDTH_8,
+			 card->ext_csd.generic_cmd6_time);
+	if (err) {
+		pr_warn("%s: switch to bus width for hs400 failed, err:%d\n",
+			mmc_hostname(host), err);
+		return err;
+	}
+
+	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+			   EXT_CSD_HS_TIMING, EXT_CSD_TIMING_HS400,
+			   card->ext_csd.generic_cmd6_time,
+			   true, true, true);
+	if (err) {
+		pr_warn("%s: switch to hs400 failed, err:%d\n",
+			 mmc_hostname(host), err);
+		return err;
+	}
+
+	mmc_set_timing(host, MMC_TIMING_MMC_HS400);
+	mmc_set_bus_speed(card);
+
+	return 0;
+}
+
+/*
+ * For device supporting HS200 mode, the following sequence
+ * should be done before executing the tuning process.
+ * 1. set the desired bus width(4-bit or 8-bit, 1-bit is not supported)
+ * 2. switch to HS200 mode
+ * 3. set the clock to > 52Mhz and <=200MHz
+ */
+static int mmc_select_hs200(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	int err = -EINVAL;
+
+	if (card->mmc_avail_type & EXT_CSD_CARD_TYPE_HS200_1_2V)
+		err = __mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_120);
+
+	if (err && card->mmc_avail_type & EXT_CSD_CARD_TYPE_HS200_1_8V)
+		err = __mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180);
+
+	/* If fails try again during next card power cycle */
+	if (err)
+		goto err;
+
+	/*
+	 * Set the bus width(4 or 8) with host's support and
+	 * switch to HS200 mode if bus width is set successfully.
+	 */
+	err = mmc_select_bus_width(card);
+	if (!IS_ERR_VALUE(err)) {
 		err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-				EXT_CSD_HS_TIMING, 2,
-				card->ext_csd.generic_cmd6_time,
-				true, true, true);
+				   EXT_CSD_HS_TIMING, EXT_CSD_TIMING_HS200,
+				   card->ext_csd.generic_cmd6_time,
+				   true, true, true);
+		if (!err)
+			mmc_set_timing(host, MMC_TIMING_MMC_HS200);
+	}
 err:
 	return err;
 }
 
 /*
+ * Activate High Speed or HS200 mode if supported.
+ */
+static int mmc_select_timing(struct mmc_card *card)
+{
+	int err = 0;
+
+	if ((card->csd.mmca_vsn < CSD_SPEC_VER_4 &&
+	     card->ext_csd.hs_max_dtr == 0))
+		goto bus_speed;
+
+	if (card->mmc_avail_type & EXT_CSD_CARD_TYPE_HS200)
+		err = mmc_select_hs200(card);
+	else if (card->mmc_avail_type & EXT_CSD_CARD_TYPE_HS)
+		err = mmc_select_hs(card);
+
+	if (err && err != -EBADMSG)
+		return err;
+
+	if (err) {
+		pr_warn("%s: switch to %s failed\n",
+			mmc_card_hs(card) ? "high-speed" :
+			(mmc_card_hs200(card) ? "hs200" : ""),
+			mmc_hostname(card->host));
+		err = 0;
+	}
+
+bus_speed:
+	/*
+	 * Set the bus speed to the selected bus timing.
+	 * If timing is not selected, backward compatible is the default.
+	 */
+	mmc_set_bus_speed(card);
+	return err;
+}
+
+/*
+ * Execute tuning sequence to seek the proper bus operating
+ * conditions for HS200 and HS400, which sends CMD21 to the device.
+ */
+static int mmc_hs200_tuning(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	int err = 0;
+
+	/*
+	 * Timing should be adjusted to the HS400 target
+	 * operation frequency for tuning process
+	 */
+	if (card->mmc_avail_type & EXT_CSD_CARD_TYPE_HS400 &&
+	    host->ios.bus_width == MMC_BUS_WIDTH_8)
+		if (host->ops->prepare_hs400_tuning)
+			host->ops->prepare_hs400_tuning(host, &host->ios);
+
+	if (host->ops->execute_tuning) {
+		mmc_host_clk_hold(host);
+		err = host->ops->execute_tuning(host,
+				MMC_SEND_TUNING_BLOCK_HS200);
+		mmc_host_clk_release(host);
+
+		if (err)
+			pr_warn("%s: tuning execution failed\n",
+				mmc_hostname(host));
+	}
+
+	return err;
+}
+
+/*
  * Handle the detection and initialisation of a card.
  *
  * In the case of a resume, "oldcard" will contain the card
@@ -874,9 +1177,8 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	struct mmc_card *oldcard)
 {
 	struct mmc_card *card;
-	int err, ddr = 0;
+	int err;
 	u32 cid[4];
-	unsigned int max_dtr;
 	u32 rocr;
 	u8 *ext_csd = NULL;
 
@@ -1068,206 +1370,34 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	}
 
 	/*
-	 * Activate high speed (if supported)
-	 */
-	if (card->ext_csd.hs_max_dtr != 0) {
-		err = 0;
-		if (card->ext_csd.hs_max_dtr > 52000000 &&
-		    host->caps2 & MMC_CAP2_HS200)
-			err = mmc_select_hs200(card);
-		else if	(host->caps & MMC_CAP_MMC_HIGHSPEED)
-			err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-					EXT_CSD_HS_TIMING, 1,
-					card->ext_csd.generic_cmd6_time,
-					true, true, true);
-
-		if (err && err != -EBADMSG)
-			goto free_card;
-
-		if (err) {
-			pr_warning("%s: switch to highspeed failed\n",
-			       mmc_hostname(card->host));
-			err = 0;
-		} else {
-			if (card->ext_csd.hs_max_dtr > 52000000 &&
-			    host->caps2 & MMC_CAP2_HS200) {
-				mmc_card_set_hs200(card);
-				mmc_set_timing(card->host,
-					       MMC_TIMING_MMC_HS200);
-			} else {
-				mmc_card_set_highspeed(card);
-				mmc_set_timing(card->host, MMC_TIMING_MMC_HS);
-			}
-		}
-	}
-
-	/*
-	 * Compute bus speed.
-	 */
-	max_dtr = (unsigned int)-1;
-
-	if (mmc_card_highspeed(card) || mmc_card_hs200(card)) {
-		if (max_dtr > card->ext_csd.hs_max_dtr)
-			max_dtr = card->ext_csd.hs_max_dtr;
-		if (mmc_card_highspeed(card) && (max_dtr > 52000000))
-			max_dtr = 52000000;
-	} else if (max_dtr > card->csd.max_dtr) {
-		max_dtr = card->csd.max_dtr;
-	}
-
-	mmc_set_clock(host, max_dtr);
-
-	/*
-	 * Indicate DDR mode (if supported).
+	 * Select timing interface
 	 */
-	if (mmc_card_highspeed(card)) {
-		if ((card->ext_csd.card_type & EXT_CSD_CARD_TYPE_DDR_1_8V)
-			&& (host->caps & MMC_CAP_1_8V_DDR))
-				ddr = MMC_1_8V_DDR_MODE;
-		else if ((card->ext_csd.card_type & EXT_CSD_CARD_TYPE_DDR_1_2V)
-			&& (host->caps & MMC_CAP_1_2V_DDR))
-				ddr = MMC_1_2V_DDR_MODE;
-	}
+	err = mmc_select_timing(card);
+	if (err)
+		goto free_card;
 
-	/*
-	 * Indicate HS200 SDR mode (if supported).
-	 */
 	if (mmc_card_hs200(card)) {
-		u32 ext_csd_bits;
-		u32 bus_width = card->host->ios.bus_width;
-
-		/*
-		 * For devices supporting HS200 mode, the bus width has
-		 * to be set before executing the tuning function. If
-		 * set before tuning, then device will respond with CRC
-		 * errors for responses on CMD line. So for HS200 the
-		 * sequence will be
-		 * 1. set bus width 4bit / 8 bit (1 bit not supported)
-		 * 2. switch to HS200 mode
-		 * 3. set the clock to > 52Mhz <=200MHz and
-		 * 4. execute tuning for HS200
-		 */
-		if ((host->caps2 & MMC_CAP2_HS200) &&
-		    card->host->ops->execute_tuning) {
-			mmc_host_clk_hold(card->host);
-			err = card->host->ops->execute_tuning(card->host,
-				MMC_SEND_TUNING_BLOCK_HS200);
-			mmc_host_clk_release(card->host);
-		}
-		if (err) {
-			pr_warning("%s: tuning execution failed\n",
-				   mmc_hostname(card->host));
+		err = mmc_hs200_tuning(card);
+		if (err)
 			goto err;
-		}
 
-		ext_csd_bits = (bus_width == MMC_BUS_WIDTH_8) ?
-				EXT_CSD_BUS_WIDTH_8 : EXT_CSD_BUS_WIDTH_4;
-		err = mmc_select_powerclass(card, ext_csd_bits);
+		err = mmc_select_hs400(card);
 		if (err)
-			pr_warning("%s: power class selection to bus width %d"
-				   " failed\n", mmc_hostname(card->host),
-				   1 << bus_width);
+			goto err;
+	} else if (mmc_card_hs(card)) {
+		/* Select the desired bus width optionally */
+		err = mmc_select_bus_width(card);
+		if (!IS_ERR_VALUE(err)) {
+			err = mmc_select_hs_ddr(card);
+			if (err)
+				goto err;
+		}
 	}
 
 	/*
-	 * Activate wide bus and DDR (if supported).
+	 * Choose the power class with selected bus interface
 	 */
-	if (!mmc_card_hs200(card) &&
-	    (card->csd.mmca_vsn >= CSD_SPEC_VER_4) &&
-	    (host->caps & (MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA))) {
-		static unsigned ext_csd_bits[][2] = {
-			{ EXT_CSD_BUS_WIDTH_8, EXT_CSD_DDR_BUS_WIDTH_8 },
-			{ EXT_CSD_BUS_WIDTH_4, EXT_CSD_DDR_BUS_WIDTH_4 },
-			{ EXT_CSD_BUS_WIDTH_1, EXT_CSD_BUS_WIDTH_1 },
-		};
-		static unsigned bus_widths[] = {
-			MMC_BUS_WIDTH_8,
-			MMC_BUS_WIDTH_4,
-			MMC_BUS_WIDTH_1
-		};
-		unsigned idx, bus_width = 0;
-
-		if (host->caps & MMC_CAP_8_BIT_DATA)
-			idx = 0;
-		else
-			idx = 1;
-		for (; idx < ARRAY_SIZE(bus_widths); idx++) {
-			bus_width = bus_widths[idx];
-			if (bus_width == MMC_BUS_WIDTH_1)
-				ddr = 0; /* no DDR for 1-bit width */
-			err = mmc_select_powerclass(card, ext_csd_bits[idx][0]);
-			if (err)
-				pr_warning("%s: power class selection to "
-					   "bus width %d failed\n",
-					   mmc_hostname(card->host),
-					   1 << bus_width);
-
-			err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-					 EXT_CSD_BUS_WIDTH,
-					 ext_csd_bits[idx][0],
-					 card->ext_csd.generic_cmd6_time);
-			if (!err) {
-				mmc_set_bus_width(card->host, bus_width);
-
-				/*
-				 * If controller can't handle bus width test,
-				 * compare ext_csd previously read in 1 bit mode
-				 * against ext_csd at new bus width
-				 */
-				if (!(host->caps & MMC_CAP_BUS_WIDTH_TEST))
-					err = mmc_compare_ext_csds(card,
-						bus_width);
-				else
-					err = mmc_bus_test(card, bus_width);
-				if (!err)
-					break;
-			}
-		}
-
-		if (!err && ddr) {
-			err = mmc_select_powerclass(card, ext_csd_bits[idx][1]);
-			if (err)
-				pr_warning("%s: power class selection to "
-					   "bus width %d ddr %d failed\n",
-					   mmc_hostname(card->host),
-					   1 << bus_width, ddr);
-
-			err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-					 EXT_CSD_BUS_WIDTH,
-					 ext_csd_bits[idx][1],
-					 card->ext_csd.generic_cmd6_time);
-		}
-		if (err) {
-			pr_warning("%s: switch to bus width %d ddr %d "
-				"failed\n", mmc_hostname(card->host),
-				1 << bus_width, ddr);
-			goto free_card;
-		} else if (ddr) {
-			/*
-			 * eMMC cards can support 3.3V to 1.2V i/o (vccq)
-			 * signaling.
-			 *
-			 * EXT_CSD_CARD_TYPE_DDR_1_8V means 3.3V or 1.8V vccq.
-			 *
-			 * 1.8V vccq at 3.3V core voltage (vcc) is not required
-			 * in the JEDEC spec for DDR.
-			 *
-			 * Do not force change in vccq since we are obviously
-			 * working and no change to vccq is needed.
-			 *
-			 * WARNING: eMMC rules are NOT the same as SD DDR
-			 */
-			if (ddr == MMC_1_2V_DDR_MODE) {
-				err = __mmc_set_signal_voltage(host,
-					MMC_SIGNAL_VOLTAGE_120);
-				if (err)
-					goto err;
-			}
-			mmc_card_set_ddr_mode(card);
-			mmc_set_timing(card->host, MMC_TIMING_UHS_DDR50);
-			mmc_set_bus_width(card->host, bus_width);
-		}
-	}
+	mmc_select_powerclass(card);
 
 	/*
 	 * Enable HPI feature (if supported)
@@ -1507,7 +1637,6 @@ static int _mmc_suspend(struct mmc_host *host, bool is_suspend)
 		err = mmc_sleep(host);
 	else if (!mmc_host_is_spi(host))
 		err = mmc_deselect_cards(host);
-	host->card->state &= ~(MMC_STATE_HIGHSPEED | MMC_STATE_HIGHSPEED_200);
 
 	if (!err) {
 		mmc_power_off(host);
@@ -1637,7 +1766,6 @@ static int mmc_power_restore(struct mmc_host *host)
 {
 	int ret;
 
-	host->card->state &= ~(MMC_STATE_HIGHSPEED | MMC_STATE_HIGHSPEED_200);
 	mmc_claim_host(host);
 	ret = mmc_init_card(host, host->card->ocr, host->card);
 	mmc_release_host(host);
diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c
index 2dd359d2242f..0c44510bf717 100644
--- a/drivers/mmc/core/sd.c
+++ b/drivers/mmc/core/sd.c
@@ -707,18 +707,10 @@ static struct attribute *sd_std_attrs[] = {
 	&dev_attr_serial.attr,
 	NULL,
 };
-
-static struct attribute_group sd_std_attr_group = {
-	.attrs = sd_std_attrs,
-};
-
-static const struct attribute_group *sd_attr_groups[] = {
-	&sd_std_attr_group,
-	NULL,
-};
+ATTRIBUTE_GROUPS(sd_std);
 
 struct device_type sd_type = {
-	.groups = sd_attr_groups,
+	.groups = sd_std_groups,
 };
 
 /*
@@ -895,7 +887,7 @@ unsigned mmc_sd_get_max_clock(struct mmc_card *card)
 {
 	unsigned max_dtr = (unsigned int)-1;
 
-	if (mmc_card_highspeed(card)) {
+	if (mmc_card_hs(card)) {
 		if (max_dtr > card->sw_caps.hs_max_dtr)
 			max_dtr = card->sw_caps.hs_max_dtr;
 	} else if (max_dtr > card->csd.max_dtr) {
@@ -905,12 +897,6 @@ unsigned mmc_sd_get_max_clock(struct mmc_card *card)
 	return max_dtr;
 }
 
-void mmc_sd_go_highspeed(struct mmc_card *card)
-{
-	mmc_card_set_highspeed(card);
-	mmc_set_timing(card->host, MMC_TIMING_SD_HS);
-}
-
 /*
  * Handle the detection and initialisation of a card.
  *
@@ -985,16 +971,13 @@ static int mmc_sd_init_card(struct mmc_host *host, u32 ocr,
 		err = mmc_sd_init_uhs_card(card);
 		if (err)
 			goto free_card;
-
-		/* Card is an ultra-high-speed card */
-		mmc_card_set_uhs(card);
 	} else {
 		/*
 		 * Attempt to change to high-speed (if supported)
 		 */
 		err = mmc_sd_switch_hs(card);
 		if (err > 0)
-			mmc_sd_go_highspeed(card);
+			mmc_set_timing(card->host, MMC_TIMING_SD_HS);
 		else if (err)
 			goto free_card;
 
@@ -1089,7 +1072,7 @@ static int _mmc_sd_suspend(struct mmc_host *host)
 
 	if (!mmc_host_is_spi(host))
 		err = mmc_deselect_cards(host);
-	host->card->state &= ~MMC_STATE_HIGHSPEED;
+
 	if (!err) {
 		mmc_power_off(host);
 		mmc_card_set_suspended(host->card);
@@ -1198,7 +1181,6 @@ static int mmc_sd_power_restore(struct mmc_host *host)
 {
 	int ret;
 
-	host->card->state &= ~MMC_STATE_HIGHSPEED;
 	mmc_claim_host(host);
 	ret = mmc_sd_init_card(host, host->card->ocr, host->card);
 	mmc_release_host(host);
diff --git a/drivers/mmc/core/sd.h b/drivers/mmc/core/sd.h
index 4b34b24f3f76..aab824a9a7f3 100644
--- a/drivers/mmc/core/sd.h
+++ b/drivers/mmc/core/sd.h
@@ -12,6 +12,5 @@ int mmc_sd_setup_card(struct mmc_host *host, struct mmc_card *card,
 	bool reinit);
 unsigned mmc_sd_get_max_clock(struct mmc_card *card);
 int mmc_sd_switch_hs(struct mmc_card *card);
-void mmc_sd_go_highspeed(struct mmc_card *card);
 
 #endif
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index 4d721c6e2af0..e636d9e99e4a 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -363,7 +363,7 @@ static unsigned mmc_sdio_get_max_clock(struct mmc_card *card)
 {
 	unsigned max_dtr;
 
-	if (mmc_card_highspeed(card)) {
+	if (mmc_card_hs(card)) {
 		/*
 		 * The SDIO specification doesn't mention how
 		 * the CIS transfer speed register relates to
@@ -733,7 +733,6 @@ try_again:
 		mmc_set_clock(host, card->cis.max_dtr);
 
 		if (card->cccr.high_speed) {
-			mmc_card_set_highspeed(card);
 			mmc_set_timing(card->host, MMC_TIMING_SD_HS);
 		}
 
@@ -792,16 +791,13 @@ try_again:
 		err = mmc_sdio_init_uhs_card(card);
 		if (err)
 			goto remove;
-
-		/* Card is an ultra-high-speed card */
-		mmc_card_set_uhs(card);
 	} else {
 		/*
 		 * Switch to high-speed (if supported).
 		 */
 		err = sdio_enable_hs(card);
 		if (err > 0)
-			mmc_sd_go_highspeed(card);
+			mmc_set_timing(card->host, MMC_TIMING_SD_HS);
 		else if (err)
 			goto remove;
 
@@ -943,40 +939,21 @@ static int mmc_sdio_pre_suspend(struct mmc_host *host)
  */
 static int mmc_sdio_suspend(struct mmc_host *host)
 {
-	int i, err = 0;
-
-	for (i = 0; i < host->card->sdio_funcs; i++) {
-		struct sdio_func *func = host->card->sdio_func[i];
-		if (func && sdio_func_present(func) && func->dev.driver) {
-			const struct dev_pm_ops *pmops = func->dev.driver->pm;
-			err = pmops->suspend(&func->dev);
-			if (err)
-				break;
-		}
-	}
-	while (err && --i >= 0) {
-		struct sdio_func *func = host->card->sdio_func[i];
-		if (func && sdio_func_present(func) && func->dev.driver) {
-			const struct dev_pm_ops *pmops = func->dev.driver->pm;
-			pmops->resume(&func->dev);
-		}
-	}
-
-	if (!err && mmc_card_keep_power(host) && mmc_card_wake_sdio_irq(host)) {
+	if (mmc_card_keep_power(host) && mmc_card_wake_sdio_irq(host)) {
 		mmc_claim_host(host);
 		sdio_disable_wide(host->card);
 		mmc_release_host(host);
 	}
 
-	if (!err && !mmc_card_keep_power(host))
+	if (!mmc_card_keep_power(host))
 		mmc_power_off(host);
 
-	return err;
+	return 0;
 }
 
 static int mmc_sdio_resume(struct mmc_host *host)
 {
-	int i, err = 0;
+	int err = 0;
 
 	BUG_ON(!host);
 	BUG_ON(!host->card);
@@ -1019,24 +996,6 @@ static int mmc_sdio_resume(struct mmc_host *host)
 		wake_up_process(host->sdio_irq_thread);
 	mmc_release_host(host);
 
-	/*
-	 * If the card looked to be the same as before suspending, then
-	 * we proceed to resume all card functions.  If one of them returns
-	 * an error then we simply return that error to the core and the
-	 * card will be redetected as new.  It is the responsibility of
-	 * the function driver to perform further tests with the extra
-	 * knowledge it has of the card to confirm the card is indeed the
-	 * same as before suspending (same MAC address for network cards,
-	 * etc.) and return an error otherwise.
-	 */
-	for (i = 0; !err && i < host->card->sdio_funcs; i++) {
-		struct sdio_func *func = host->card->sdio_func[i];
-		if (func && sdio_func_present(func) && func->dev.driver) {
-			const struct dev_pm_ops *pmops = func->dev.driver->pm;
-			err = pmops->resume(&func->dev);
-		}
-	}
-
 	host->pm_flags &= ~MMC_PM_KEEP_POWER;
 	return err;
 }
diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c
index 92d1ba8e8153..4fa8fef9147f 100644
--- a/drivers/mmc/core/sdio_bus.c
+++ b/drivers/mmc/core/sdio_bus.c
@@ -197,20 +197,8 @@ static int sdio_bus_remove(struct device *dev)
 
 #ifdef CONFIG_PM
 
-#ifdef CONFIG_PM_SLEEP
-static int pm_no_operation(struct device *dev)
-{
-	/*
-	 * Prevent the PM core from calling SDIO device drivers' suspend
-	 * callback routines, which it is not supposed to do, by using this
-	 * empty function as the bus type suspend callaback for SDIO.
-	 */
-	return 0;
-}
-#endif
-
 static const struct dev_pm_ops sdio_bus_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(pm_no_operation, pm_no_operation)
+	SET_SYSTEM_SLEEP_PM_OPS(pm_generic_suspend, pm_generic_resume)
 	SET_RUNTIME_PM_OPS(
 		pm_generic_runtime_suspend,
 		pm_generic_runtime_resume,
diff --git a/drivers/mmc/core/sdio_irq.c b/drivers/mmc/core/sdio_irq.c
index aaa90460ed23..5cc13c8d35bb 100644
--- a/drivers/mmc/core/sdio_irq.c
+++ b/drivers/mmc/core/sdio_irq.c
@@ -90,6 +90,15 @@ static int process_sdio_pending_irqs(struct mmc_host *host)
 	return ret;
 }
 
+void sdio_run_irqs(struct mmc_host *host)
+{
+	mmc_claim_host(host);
+	host->sdio_irq_pending = true;
+	process_sdio_pending_irqs(host);
+	mmc_release_host(host);
+}
+EXPORT_SYMBOL_GPL(sdio_run_irqs);
+
 static int sdio_irq_thread(void *_host)
 {
 	struct mmc_host *host = _host;
@@ -189,14 +198,20 @@ static int sdio_card_irq_get(struct mmc_card *card)
 	WARN_ON(!host->claimed);
 
 	if (!host->sdio_irqs++) {
-		atomic_set(&host->sdio_irq_thread_abort, 0);
-		host->sdio_irq_thread =
-			kthread_run(sdio_irq_thread, host, "ksdioirqd/%s",
-				mmc_hostname(host));
-		if (IS_ERR(host->sdio_irq_thread)) {
-			int err = PTR_ERR(host->sdio_irq_thread);
-			host->sdio_irqs--;
-			return err;
+		if (!(host->caps2 & MMC_CAP2_SDIO_IRQ_NOTHREAD)) {
+			atomic_set(&host->sdio_irq_thread_abort, 0);
+			host->sdio_irq_thread =
+				kthread_run(sdio_irq_thread, host,
+					    "ksdioirqd/%s", mmc_hostname(host));
+			if (IS_ERR(host->sdio_irq_thread)) {
+				int err = PTR_ERR(host->sdio_irq_thread);
+				host->sdio_irqs--;
+				return err;
+			}
+		} else {
+			mmc_host_clk_hold(host);
+			host->ops->enable_sdio_irq(host, 1);
+			mmc_host_clk_release(host);
 		}
 	}
 
@@ -211,8 +226,14 @@ static int sdio_card_irq_put(struct mmc_card *card)
 	BUG_ON(host->sdio_irqs < 1);
 
 	if (!--host->sdio_irqs) {
-		atomic_set(&host->sdio_irq_thread_abort, 1);
-		kthread_stop(host->sdio_irq_thread);
+		if (!(host->caps2 & MMC_CAP2_SDIO_IRQ_NOTHREAD)) {
+			atomic_set(&host->sdio_irq_thread_abort, 1);
+			kthread_stop(host->sdio_irq_thread);
+		} else {
+			mmc_host_clk_hold(host);
+			host->ops->enable_sdio_irq(host, 0);
+			mmc_host_clk_release(host);
+		}
 	}
 
 	return 0;
diff --git a/drivers/mmc/core/slot-gpio.c b/drivers/mmc/core/slot-gpio.c
index f7650b899e3d..5f89cb83d5f0 100644
--- a/drivers/mmc/core/slot-gpio.c
+++ b/drivers/mmc/core/slot-gpio.c
@@ -32,9 +32,7 @@ static irqreturn_t mmc_gpio_cd_irqt(int irq, void *dev_id)
 	/* Schedule a card detection after a debounce timeout */
 	struct mmc_host *host = dev_id;
 
-	if (host->ops->card_event)
-		host->ops->card_event(host);
-
+	host->trigger_card_event = true;
 	mmc_detect_change(host, msecs_to_jiffies(200));
 
 	return IRQ_HANDLED;
diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 779368b683d0..a5652548230a 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -168,7 +168,7 @@ config MMC_SDHCI_ESDHC_IMX
 
 config MMC_SDHCI_DOVE
 	tristate "SDHCI support on Marvell's Dove SoC"
-	depends on ARCH_DOVE
+	depends on ARCH_DOVE || MACH_DOVE
 	depends on MMC_SDHCI_PLTFM
 	select MMC_SDHCI_IO_ACCESSORS
 	help
@@ -216,8 +216,7 @@ config MMC_SDHCI_SIRF
 config MMC_SDHCI_PXAV3
 	tristate "Marvell MMP2 SD Host Controller support (PXAV3)"
 	depends on CLKDEV_LOOKUP
-	select MMC_SDHCI
-	select MMC_SDHCI_PLTFM
+	depends on MMC_SDHCI_PLTFM
 	default CPU_MMP2
 	help
 	  This selects the Marvell(R) PXAV3 SD Host Controller.
@@ -229,8 +228,7 @@ config MMC_SDHCI_PXAV3
 config MMC_SDHCI_PXAV2
 	tristate "Marvell PXA9XX SD Host Controller support (PXAV2)"
 	depends on CLKDEV_LOOKUP
-	select MMC_SDHCI
-	select MMC_SDHCI_PLTFM
+	depends on MMC_SDHCI_PLTFM
 	default CPU_PXA910
 	help
 	  This selects the Marvell(R) PXAV2 SD Host Controller.
@@ -264,7 +262,7 @@ config MMC_SDHCI_S3C_DMA
 config MMC_SDHCI_BCM_KONA
 	tristate "SDHCI support on Broadcom KONA platform"
 	depends on ARCH_BCM_MOBILE
-	select MMC_SDHCI_PLTFM
+	depends on MMC_SDHCI_PLTFM
 	help
 	  This selects the Broadcom Kona Secure Digital Host Controller
 	  Interface(SDHCI) support.
@@ -283,10 +281,19 @@ config MMC_SDHCI_BCM2835
 
 	  If unsure, say N.
 
+config MMC_MOXART
+	tristate "MOXART SD/MMC Host Controller support"
+	depends on ARCH_MOXART && MMC
+	help
+	  This selects support for the MOXART SD/MMC Host Controller.
+	  MOXA provides one multi-functional card reader which can
+	  be found on some embedded hardware such as UC-7112-LX.
+	  If you have a controller with this interface, say Y here.
+
 config MMC_OMAP
 	tristate "TI OMAP Multimedia Card Interface support"
 	depends on ARCH_OMAP
-	select TPS65010 if MACH_OMAP_H2
+	depends on TPS65010 || !MACH_OMAP_H2
 	help
 	  This selects the TI OMAP Multimedia card Interface.
 	  If you have an OMAP board with a Multimedia Card slot,
@@ -688,6 +695,12 @@ config MMC_WMT
 	  To compile this driver as a module, choose M here: the
 	  module will be called wmt-sdmmc.
 
+config MMC_USDHI6ROL0
+	tristate "Renesas USDHI6ROL0 SD/SDIO Host Controller support"
+	help
+	  This selects support for the Renesas USDHI6ROL0 SD/SDIO
+	  Host Controller
+
 config MMC_REALTEK_PCI
 	tristate "Realtek PCI-E SD/MMC Card Interface Driver"
 	depends on MFD_RTSX_PCI
diff --git a/drivers/mmc/host/Makefile b/drivers/mmc/host/Makefile
index 61cbc241935b..7f81ddf1dd2c 100644
--- a/drivers/mmc/host/Makefile
+++ b/drivers/mmc/host/Makefile
@@ -50,7 +50,9 @@ obj-$(CONFIG_MMC_JZ4740)	+= jz4740_mmc.o
 obj-$(CONFIG_MMC_VUB300)	+= vub300.o
 obj-$(CONFIG_MMC_USHC)		+= ushc.o
 obj-$(CONFIG_MMC_WMT)		+= wmt-sdmmc.o
+obj-$(CONFIG_MMC_MOXART)	+= moxart-mmc.o
 obj-$(CONFIG_MMC_SUNXI)		+= sunxi-mmc.o
+obj-$(CONFIG_MMC_USDHI6ROL0)	+= usdhi6rol0.o
 
 obj-$(CONFIG_MMC_REALTEK_PCI)	+= rtsx_pci_sdmmc.o
 obj-$(CONFIG_MMC_REALTEK_USB)	+= rtsx_usb_sdmmc.o
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 42706ea0ba85..bb585d940901 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -37,6 +37,7 @@
 #include <linux/atmel-mci.h>
 #include <linux/atmel_pdc.h>
 
+#include <asm/cacheflush.h>
 #include <asm/io.h>
 #include <asm/unaligned.h>
 
@@ -820,16 +821,9 @@ static void atmci_pdc_complete(struct atmel_mci *host)
 
 	atmci_pdc_cleanup(host);
 
-	/*
-	 * If the card was removed, data will be NULL. No point trying
-	 * to send the stop command or waiting for NBUSY in this case.
-	 */
-	if (host->data) {
-		dev_dbg(&host->pdev->dev,
-		        "(%s) set pending xfer complete\n", __func__);
-		atmci_set_pending(host, EVENT_XFER_COMPLETE);
-		tasklet_schedule(&host->tasklet);
-	}
+	dev_dbg(&host->pdev->dev, "(%s) set pending xfer complete\n", __func__);
+	atmci_set_pending(host, EVENT_XFER_COMPLETE);
+	tasklet_schedule(&host->tasklet);
 }
 
 static void atmci_dma_cleanup(struct atmel_mci *host)
diff --git a/drivers/mmc/host/dw_mmc-exynos.c b/drivers/mmc/host/dw_mmc-exynos.c
index 3423c5ed50c7..0fbc53ac7eae 100644
--- a/drivers/mmc/host/dw_mmc-exynos.c
+++ b/drivers/mmc/host/dw_mmc-exynos.c
@@ -187,7 +187,7 @@ static void dw_mci_exynos_set_ios(struct dw_mci *host, struct mmc_ios *ios)
 	unsigned long actual;
 	u8 div = priv->ciu_div + 1;
 
-	if (ios->timing == MMC_TIMING_UHS_DDR50) {
+	if (ios->timing == MMC_TIMING_MMC_DDR52) {
 		mci_writel(host, CLKSEL, priv->ddr_timing);
 		/* Should be double rate for DDR mode */
 		if (ios->bus_width == MMC_BUS_WIDTH_8)
@@ -386,8 +386,7 @@ static int dw_mci_exynos_execute_tuning(struct dw_mci_slot *slot, u32 opcode,
 
 /* Common capabilities of Exynos4/Exynos5 SoC */
 static unsigned long exynos_dwmmc_caps[4] = {
-	MMC_CAP_UHS_DDR50 | MMC_CAP_1_8V_DDR |
-		MMC_CAP_8_BIT_DATA | MMC_CAP_CMD23,
+	MMC_CAP_1_8V_DDR | MMC_CAP_8_BIT_DATA | MMC_CAP_CMD23,
 	MMC_CAP_CMD23,
 	MMC_CAP_CMD23,
 	MMC_CAP_CMD23,
@@ -426,7 +425,7 @@ static int dw_mci_exynos_probe(struct platform_device *pdev)
 	return dw_mci_pltfm_register(pdev, drv_data);
 }
 
-const struct dev_pm_ops dw_mci_exynos_pmops = {
+static const struct dev_pm_ops dw_mci_exynos_pmops = {
 	SET_SYSTEM_SLEEP_PM_OPS(dw_mci_exynos_suspend, dw_mci_exynos_resume)
 	.resume_noirq = dw_mci_exynos_resume_noirq,
 	.thaw_noirq = dw_mci_exynos_resume_noirq,
diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index cced599d5aeb..1ac227c603b7 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -235,12 +235,6 @@ err:
 }
 #endif /* defined(CONFIG_DEBUG_FS) */
 
-static void dw_mci_set_timeout(struct dw_mci *host)
-{
-	/* timeout (maximum) */
-	mci_writel(host, TMOUT, 0xffffffff);
-}
-
 static u32 dw_mci_prepare_command(struct mmc_host *mmc, struct mmc_command *cmd)
 {
 	struct mmc_data	*data;
@@ -257,9 +251,8 @@ static u32 dw_mci_prepare_command(struct mmc_host *mmc, struct mmc_command *cmd)
 	    (cmd->opcode == SD_IO_RW_DIRECT &&
 	     ((cmd->arg >> 9) & 0x1FFFF) == SDIO_CCCR_ABORT))
 		cmdr |= SDMMC_CMD_STOP;
-	else
-		if (cmd->opcode != MMC_SEND_STATUS && cmd->data)
-			cmdr |= SDMMC_CMD_PRV_DAT_WAIT;
+	else if (cmd->opcode != MMC_SEND_STATUS && cmd->data)
+		cmdr |= SDMMC_CMD_PRV_DAT_WAIT;
 
 	if (cmd->flags & MMC_RSP_PRESENT) {
 		/* We expect a response, so set this bit */
@@ -850,8 +843,6 @@ static void __dw_mci_start_request(struct dw_mci *host,
 	u32 cmdflags;
 
 	mrq = slot->mrq;
-	if (host->pdata->select_slot)
-		host->pdata->select_slot(slot->id);
 
 	host->cur_slot = slot;
 	host->mrq = mrq;
@@ -864,7 +855,7 @@ static void __dw_mci_start_request(struct dw_mci *host,
 
 	data = cmd->data;
 	if (data) {
-		dw_mci_set_timeout(host);
+		mci_writel(host, TMOUT, 0xFFFFFFFF);
 		mci_writel(host, BYTCNT, data->blksz*data->blocks);
 		mci_writel(host, BLKSIZ, data->blksz);
 	}
@@ -962,7 +953,7 @@ static void dw_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 	regs = mci_readl(slot->host, UHS_REG);
 
 	/* DDR mode set */
-	if (ios->timing == MMC_TIMING_UHS_DDR50)
+	if (ios->timing == MMC_TIMING_MMC_DDR52)
 		regs |= ((0x1 << slot->id) << 16);
 	else
 		regs &= ~((0x1 << slot->id) << 16);
@@ -985,17 +976,11 @@ static void dw_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 	switch (ios->power_mode) {
 	case MMC_POWER_UP:
 		set_bit(DW_MMC_CARD_NEED_INIT, &slot->flags);
-		/* Power up slot */
-		if (slot->host->pdata->setpower)
-			slot->host->pdata->setpower(slot->id, mmc->ocr_avail);
 		regs = mci_readl(slot->host, PWREN);
 		regs |= (1 << slot->id);
 		mci_writel(slot->host, PWREN, regs);
 		break;
 	case MMC_POWER_OFF:
-		/* Power down slot */
-		if (slot->host->pdata->setpower)
-			slot->host->pdata->setpower(slot->id, 0);
 		regs = mci_readl(slot->host, PWREN);
 		regs &= ~(1 << slot->id);
 		mci_writel(slot->host, PWREN, regs);
@@ -1009,15 +994,13 @@ static int dw_mci_get_ro(struct mmc_host *mmc)
 {
 	int read_only;
 	struct dw_mci_slot *slot = mmc_priv(mmc);
-	struct dw_mci_board *brd = slot->host->pdata;
+	int gpio_ro = mmc_gpio_get_ro(mmc);
 
 	/* Use platform get_ro function, else try on board write protect */
 	if (slot->quirks & DW_MCI_SLOT_QUIRK_NO_WRITE_PROTECT)
 		read_only = 0;
-	else if (brd->get_ro)
-		read_only = brd->get_ro(slot->id);
-	else if (gpio_is_valid(slot->wp_gpio))
-		read_only = gpio_get_value(slot->wp_gpio);
+	else if (!IS_ERR_VALUE(gpio_ro))
+		read_only = gpio_ro;
 	else
 		read_only =
 			mci_readl(slot->host, WRTPRT) & (1 << slot->id) ? 1 : 0;
@@ -1039,8 +1022,6 @@ static int dw_mci_get_cd(struct mmc_host *mmc)
 	/* Use platform get_cd function, else try onboard card detect */
 	if (brd->quirks & DW_MCI_QUIRK_BROKEN_CARD_DETECTION)
 		present = 1;
-	else if (brd->get_cd)
-		present = !brd->get_cd(slot->id);
 	else if (!IS_ERR_VALUE(gpio_cd))
 		present = gpio_cd;
 	else
@@ -1248,7 +1229,7 @@ static int dw_mci_data_complete(struct dw_mci *host, struct mmc_data *data)
 			data->error = -EIO;
 		}
 
-		dev_err(host->dev, "data error, status 0x%08x\n", status);
+		dev_dbg(host->dev, "data error, status 0x%08x\n", status);
 
 		/*
 		 * After an error, there may be data lingering
@@ -2045,86 +2026,15 @@ static int dw_mci_of_get_slot_quirks(struct device *dev, u8 slot)
 
 	return quirks;
 }
-
-/* find out bus-width for a given slot */
-static u32 dw_mci_of_get_bus_wd(struct device *dev, u8 slot)
-{
-	struct device_node *np = dw_mci_of_find_slot_node(dev, slot);
-	u32 bus_wd = 1;
-
-	if (!np)
-		return 1;
-
-	if (of_property_read_u32(np, "bus-width", &bus_wd))
-		dev_err(dev, "bus-width property not found, assuming width"
-			       " as 1\n");
-	return bus_wd;
-}
-
-/* find the write protect gpio for a given slot; or -1 if none specified */
-static int dw_mci_of_get_wp_gpio(struct device *dev, u8 slot)
-{
-	struct device_node *np = dw_mci_of_find_slot_node(dev, slot);
-	int gpio;
-
-	if (!np)
-		return -EINVAL;
-
-	gpio = of_get_named_gpio(np, "wp-gpios", 0);
-
-	/* Having a missing entry is valid; return silently */
-	if (!gpio_is_valid(gpio))
-		return -EINVAL;
-
-	if (devm_gpio_request(dev, gpio, "dw-mci-wp")) {
-		dev_warn(dev, "gpio [%d] request failed\n", gpio);
-		return -EINVAL;
-	}
-
-	return gpio;
-}
-
-/* find the cd gpio for a given slot */
-static void dw_mci_of_get_cd_gpio(struct device *dev, u8 slot,
-					struct mmc_host *mmc)
-{
-	struct device_node *np = dw_mci_of_find_slot_node(dev, slot);
-	int gpio;
-
-	if (!np)
-		return;
-
-	gpio = of_get_named_gpio(np, "cd-gpios", 0);
-
-	/* Having a missing entry is valid; return silently */
-	if (!gpio_is_valid(gpio))
-		return;
-
-	if (mmc_gpio_request_cd(mmc, gpio, 0))
-		dev_warn(dev, "gpio [%d] request failed\n", gpio);
-}
 #else /* CONFIG_OF */
 static int dw_mci_of_get_slot_quirks(struct device *dev, u8 slot)
 {
 	return 0;
 }
-static u32 dw_mci_of_get_bus_wd(struct device *dev, u8 slot)
-{
-	return 1;
-}
 static struct device_node *dw_mci_of_find_slot_node(struct device *dev, u8 slot)
 {
 	return NULL;
 }
-static int dw_mci_of_get_wp_gpio(struct device *dev, u8 slot)
-{
-	return -EINVAL;
-}
-static void dw_mci_of_get_cd_gpio(struct device *dev, u8 slot,
-					struct mmc_host *mmc)
-{
-	return;
-}
 #endif /* CONFIG_OF */
 
 static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
@@ -2134,7 +2044,6 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	const struct dw_mci_drv_data *drv_data = host->drv_data;
 	int ctrl_id, ret;
 	u32 freq[2];
-	u8 bus_width;
 
 	mmc = mmc_alloc_host(sizeof(struct dw_mci_slot), host->dev);
 	if (!mmc)
@@ -2158,17 +2067,7 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 		mmc->f_max = freq[1];
 	}
 
-	if (host->pdata->get_ocr)
-		mmc->ocr_avail = host->pdata->get_ocr(id);
-	else
-		mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-
-	/*
-	 * Start with slot power disabled, it will be enabled when a card
-	 * is detected.
-	 */
-	if (host->pdata->setpower)
-		host->pdata->setpower(id, 0);
+	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
 
 	if (host->pdata->caps)
 		mmc->caps = host->pdata->caps;
@@ -2189,19 +2088,7 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	if (host->pdata->caps2)
 		mmc->caps2 = host->pdata->caps2;
 
-	if (host->pdata->get_bus_wd)
-		bus_width = host->pdata->get_bus_wd(slot->id);
-	else if (host->dev->of_node)
-		bus_width = dw_mci_of_get_bus_wd(host->dev, slot->id);
-	else
-		bus_width = 1;
-
-	switch (bus_width) {
-	case 8:
-		mmc->caps |= MMC_CAP_8_BIT_DATA;
-	case 4:
-		mmc->caps |= MMC_CAP_4_BIT_DATA;
-	}
+	mmc_of_parse(mmc);
 
 	if (host->pdata->blk_settings) {
 		mmc->max_segs = host->pdata->blk_settings->max_segs;
@@ -2226,8 +2113,10 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 #endif /* CONFIG_MMC_DW_IDMAC */
 	}
 
-	slot->wp_gpio = dw_mci_of_get_wp_gpio(host->dev, slot->id);
-	dw_mci_of_get_cd_gpio(host->dev, slot->id, mmc);
+	if (dw_mci_get_cd(mmc))
+		set_bit(DW_MMC_CARD_PRESENT, &slot->flags);
+	else
+		clear_bit(DW_MMC_CARD_PRESENT, &slot->flags);
 
 	ret = mmc_add_host(mmc);
 	if (ret)
@@ -2249,10 +2138,6 @@ err_setup_bus:
 
 static void dw_mci_cleanup_slot(struct dw_mci_slot *slot, unsigned int id)
 {
-	/* Shutdown detect IRQ */
-	if (slot->host->pdata->exit)
-		slot->host->pdata->exit(id);
-
 	/* Debugfs stuff is cleaned up by mmc core */
 	mmc_remove_host(slot->mmc);
 	slot->host->slot[id] = NULL;
@@ -2399,24 +2284,9 @@ static struct dw_mci_board *dw_mci_parse_dt(struct dw_mci *host)
 			return ERR_PTR(ret);
 	}
 
-	if (of_find_property(np, "keep-power-in-suspend", NULL))
-		pdata->pm_caps |= MMC_PM_KEEP_POWER;
-
-	if (of_find_property(np, "enable-sdio-wakeup", NULL))
-		pdata->pm_caps |= MMC_PM_WAKE_SDIO_IRQ;
-
 	if (of_find_property(np, "supports-highspeed", NULL))
 		pdata->caps |= MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED;
 
-	if (of_find_property(np, "caps2-mmc-hs200-1_8v", NULL))
-		pdata->caps2 |= MMC_CAP2_HS200_1_8V_SDR;
-
-	if (of_find_property(np, "caps2-mmc-hs200-1_2v", NULL))
-		pdata->caps2 |= MMC_CAP2_HS200_1_2V_SDR;
-
-	if (of_get_property(np, "cd-inverted", NULL))
-		pdata->caps2 |= MMC_CAP2_CD_ACTIVE_HIGH;
-
 	return pdata;
 }
 
@@ -2442,9 +2312,9 @@ int dw_mci_probe(struct dw_mci *host)
 		}
 	}
 
-	if (!host->pdata->select_slot && host->pdata->num_slots > 1) {
+	if (host->pdata->num_slots > 1) {
 		dev_err(host->dev,
-			"Platform data must supply select_slot function\n");
+			"Platform data must supply num_slots.\n");
 		return -ENODEV;
 	}
 
@@ -2474,12 +2344,19 @@ int dw_mci_probe(struct dw_mci *host)
 			ret = clk_set_rate(host->ciu_clk, host->pdata->bus_hz);
 			if (ret)
 				dev_warn(host->dev,
-					 "Unable to set bus rate to %ul\n",
+					 "Unable to set bus rate to %uHz\n",
 					 host->pdata->bus_hz);
 		}
 		host->bus_hz = clk_get_rate(host->ciu_clk);
 	}
 
+	if (!host->bus_hz) {
+		dev_err(host->dev,
+			"Platform data must supply bus speed\n");
+		ret = -ENODEV;
+		goto err_clk_ciu;
+	}
+
 	if (drv_data && drv_data->init) {
 		ret = drv_data->init(host);
 		if (ret) {
@@ -2516,13 +2393,6 @@ int dw_mci_probe(struct dw_mci *host)
 		}
 	}
 
-	if (!host->bus_hz) {
-		dev_err(host->dev,
-			"Platform data must supply bus speed\n");
-		ret = -ENODEV;
-		goto err_regulator;
-	}
-
 	host->quirks = host->pdata->quirks;
 
 	spin_lock_init(&host->lock);
@@ -2666,8 +2536,6 @@ err_workqueue:
 err_dmaunmap:
 	if (host->use_dma && host->dma_ops->exit)
 		host->dma_ops->exit(host);
-
-err_regulator:
 	if (host->vmmc)
 		regulator_disable(host->vmmc);
 
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 68349779c396..738fa241d058 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -195,7 +195,6 @@ extern int dw_mci_resume(struct dw_mci *host);
  * @mmc: The mmc_host representing this slot.
  * @host: The MMC controller this slot is using.
  * @quirks: Slot-level quirks (DW_MCI_SLOT_QUIRK_XXX)
- * @wp_gpio: If gpio_is_valid() we'll use this to read write protect.
  * @ctype: Card type for this slot.
  * @mrq: mmc_request currently being processed or waiting to be
  *	processed, or NULL when the slot is idle.
@@ -214,7 +213,6 @@ struct dw_mci_slot {
 	struct dw_mci		*host;
 
 	int			quirks;
-	int			wp_gpio;
 
 	u32			ctype;
 
diff --git a/drivers/mmc/host/jz4740_mmc.c b/drivers/mmc/host/jz4740_mmc.c
index de2139cf3444..537d6c7a5ae4 100644
--- a/drivers/mmc/host/jz4740_mmc.c
+++ b/drivers/mmc/host/jz4740_mmc.c
@@ -515,10 +515,13 @@ static irqreturn_t jz_mmc_irq_worker(int irq, void *devid)
 
 		jz4740_mmc_send_command(host, req->stop);
 
-		timeout = jz4740_mmc_poll_irq(host, JZ_MMC_IRQ_PRG_DONE);
-		if (timeout) {
-			host->state = JZ4740_MMC_STATE_DONE;
-			break;
+		if (mmc_resp_type(req->stop) & MMC_RSP_BUSY) {
+			timeout = jz4740_mmc_poll_irq(host,
+						      JZ_MMC_IRQ_PRG_DONE);
+			if (timeout) {
+				host->state = JZ4740_MMC_STATE_DONE;
+				break;
+			}
 		}
 	case JZ4740_MMC_STATE_DONE:
 		break;
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 0a87e5691341..cc8d4a6099cd 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -448,7 +448,6 @@ mmc_spi_command_send(struct mmc_spi_host *host,
 {
 	struct scratch		*data = host->data;
 	u8			*cp = data->status;
-	u32			arg = cmd->arg;
 	int			status;
 	struct spi_transfer	*t;
 
@@ -465,14 +464,12 @@ mmc_spi_command_send(struct mmc_spi_host *host,
 	 * We init the whole buffer to all-ones, which is what we need
 	 * to write while we're reading (later) response data.
 	 */
-	memset(cp++, 0xff, sizeof(data->status));
+	memset(cp, 0xff, sizeof(data->status));
 
-	*cp++ = 0x40 | cmd->opcode;
-	*cp++ = (u8)(arg >> 24);
-	*cp++ = (u8)(arg >> 16);
-	*cp++ = (u8)(arg >> 8);
-	*cp++ = (u8)arg;
-	*cp++ = (crc7(0, &data->status[1], 5) << 1) | 0x01;
+	cp[1] = 0x40 | cmd->opcode;
+	put_unaligned_be32(cmd->arg, cp+2);
+	cp[6] = crc7_be(0, cp+1, 5) | 0x01;
+	cp += 7;
 
 	/* Then, read up to 13 bytes (while writing all-ones):
 	 *  - N(CR) (== 1..8) bytes of all-ones
@@ -711,10 +708,7 @@ mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t,
 	 * so we have to cope with this situation and check the response
 	 * bit-by-bit. Arggh!!!
 	 */
-	pattern  = scratch->status[0] << 24;
-	pattern |= scratch->status[1] << 16;
-	pattern |= scratch->status[2] << 8;
-	pattern |= scratch->status[3];
+	pattern = get_unaligned_be32(scratch->status);
 
 	/* First 3 bit of pattern are undefined */
 	pattern |= 0xE0000000;
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index a084edd37af5..7ad463e9741c 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -301,7 +301,8 @@ static void mmci_set_clkreg(struct mmci_host *host, unsigned int desired)
 	if (host->mmc->ios.bus_width == MMC_BUS_WIDTH_8)
 		clk |= MCI_ST_8BIT_BUS;
 
-	if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50)
+	if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
+	    host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
 		clk |= MCI_ST_UX500_NEG_EDGE;
 
 	mmci_write_clkreg(host, clk);
@@ -764,7 +765,8 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
 			mmci_write_clkreg(host, clk);
 		}
 
-	if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50)
+	if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
+	    host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
 		datactrl |= MCI_ST_DPSM_DDRMODE;
 
 	/*
diff --git a/drivers/mmc/host/moxart-mmc.c b/drivers/mmc/host/moxart-mmc.c
new file mode 100644
index 000000000000..74924a04026e
--- /dev/null
+++ b/drivers/mmc/host/moxart-mmc.c
@@ -0,0 +1,730 @@
+/*
+ * MOXA ART MMC host driver.
+ *
+ * Copyright (C) 2014 Jonas Jensen
+ *
+ * Jonas Jensen <jonas.jensen@gmail.com>
+ *
+ * Based on code from
+ * Moxa Technologies Co., Ltd. <www.moxa.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/version.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/blkdev.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmaengine.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/sd.h>
+#include <linux/sched.h>
+#include <linux/io.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/clk.h>
+#include <linux/bitops.h>
+#include <linux/of_dma.h>
+#include <linux/spinlock.h>
+
+#define REG_COMMAND		0
+#define REG_ARGUMENT		4
+#define REG_RESPONSE0		8
+#define REG_RESPONSE1		12
+#define REG_RESPONSE2		16
+#define REG_RESPONSE3		20
+#define REG_RESPONSE_COMMAND	24
+#define REG_DATA_CONTROL	28
+#define REG_DATA_TIMER		32
+#define REG_DATA_LENGTH		36
+#define REG_STATUS		40
+#define REG_CLEAR		44
+#define REG_INTERRUPT_MASK	48
+#define REG_POWER_CONTROL	52
+#define REG_CLOCK_CONTROL	56
+#define REG_BUS_WIDTH		60
+#define REG_DATA_WINDOW		64
+#define REG_FEATURE		68
+#define REG_REVISION		72
+
+/* REG_COMMAND */
+#define CMD_SDC_RESET		BIT(10)
+#define CMD_EN			BIT(9)
+#define CMD_APP_CMD		BIT(8)
+#define CMD_LONG_RSP		BIT(7)
+#define CMD_NEED_RSP		BIT(6)
+#define CMD_IDX_MASK		0x3f
+
+/* REG_RESPONSE_COMMAND */
+#define RSP_CMD_APP		BIT(6)
+#define RSP_CMD_IDX_MASK	0x3f
+
+/* REG_DATA_CONTROL */
+#define DCR_DATA_FIFO_RESET     BIT(8)
+#define DCR_DATA_THRES          BIT(7)
+#define DCR_DATA_EN		BIT(6)
+#define DCR_DMA_EN		BIT(5)
+#define DCR_DATA_WRITE		BIT(4)
+#define DCR_BLK_SIZE		0x0f
+
+/* REG_DATA_LENGTH */
+#define DATA_LEN_MASK		0xffffff
+
+/* REG_STATUS */
+#define WRITE_PROT		BIT(12)
+#define CARD_DETECT		BIT(11)
+/* 1-10 below can be sent to either registers, interrupt or clear. */
+#define CARD_CHANGE		BIT(10)
+#define FIFO_ORUN		BIT(9)
+#define FIFO_URUN		BIT(8)
+#define DATA_END		BIT(7)
+#define CMD_SENT		BIT(6)
+#define DATA_CRC_OK		BIT(5)
+#define RSP_CRC_OK		BIT(4)
+#define DATA_TIMEOUT		BIT(3)
+#define RSP_TIMEOUT		BIT(2)
+#define DATA_CRC_FAIL		BIT(1)
+#define RSP_CRC_FAIL		BIT(0)
+
+#define MASK_RSP		(RSP_TIMEOUT | RSP_CRC_FAIL | \
+				 RSP_CRC_OK  | CARD_DETECT  | CMD_SENT)
+
+#define MASK_DATA		(DATA_CRC_OK   | DATA_END | \
+				 DATA_CRC_FAIL | DATA_TIMEOUT)
+
+#define MASK_INTR_PIO		(FIFO_URUN | FIFO_ORUN | CARD_CHANGE)
+
+/* REG_POWER_CONTROL */
+#define SD_POWER_ON		BIT(4)
+#define SD_POWER_MASK		0x0f
+
+/* REG_CLOCK_CONTROL */
+#define CLK_HISPD		BIT(9)
+#define CLK_OFF			BIT(8)
+#define CLK_SD			BIT(7)
+#define CLK_DIV_MASK		0x7f
+
+/* REG_BUS_WIDTH */
+#define BUS_WIDTH_8		BIT(2)
+#define BUS_WIDTH_4		BIT(1)
+#define BUS_WIDTH_1		BIT(0)
+
+#define MMC_VDD_360		23
+#define MIN_POWER		(MMC_VDD_360 - SD_POWER_MASK)
+#define MAX_RETRIES		500000
+
+struct moxart_host {
+	spinlock_t			lock;
+
+	void __iomem			*base;
+
+	phys_addr_t			reg_phys;
+
+	struct dma_chan			*dma_chan_tx;
+	struct dma_chan                 *dma_chan_rx;
+	struct dma_async_tx_descriptor	*tx_desc;
+	struct mmc_host			*mmc;
+	struct mmc_request		*mrq;
+	struct scatterlist		*cur_sg;
+	struct completion		dma_complete;
+	struct completion		pio_complete;
+
+	u32				num_sg;
+	u32				data_remain;
+	u32				data_len;
+	u32				fifo_width;
+	u32				timeout;
+	u32				rate;
+
+	long				sysclk;
+
+	bool				have_dma;
+	bool				is_removed;
+};
+
+static inline void moxart_init_sg(struct moxart_host *host,
+				  struct mmc_data *data)
+{
+	host->cur_sg = data->sg;
+	host->num_sg = data->sg_len;
+	host->data_remain = host->cur_sg->length;
+
+	if (host->data_remain > host->data_len)
+		host->data_remain = host->data_len;
+}
+
+static inline int moxart_next_sg(struct moxart_host *host)
+{
+	int remain;
+	struct mmc_data *data = host->mrq->cmd->data;
+
+	host->cur_sg++;
+	host->num_sg--;
+
+	if (host->num_sg > 0) {
+		host->data_remain = host->cur_sg->length;
+		remain = host->data_len - data->bytes_xfered;
+		if (remain > 0 && remain < host->data_remain)
+			host->data_remain = remain;
+	}
+
+	return host->num_sg;
+}
+
+static int moxart_wait_for_status(struct moxart_host *host,
+				  u32 mask, u32 *status)
+{
+	int ret = -ETIMEDOUT;
+	u32 i;
+
+	for (i = 0; i < MAX_RETRIES; i++) {
+		*status = readl(host->base + REG_STATUS);
+		if (!(*status & mask)) {
+			udelay(5);
+			continue;
+		}
+		writel(*status & mask, host->base + REG_CLEAR);
+		ret = 0;
+		break;
+	}
+
+	if (ret)
+		dev_err(mmc_dev(host->mmc), "timed out waiting for status\n");
+
+	return ret;
+}
+
+
+static void moxart_send_command(struct moxart_host *host,
+	struct mmc_command *cmd)
+{
+	u32 status, cmdctrl;
+
+	writel(RSP_TIMEOUT  | RSP_CRC_OK |
+	       RSP_CRC_FAIL | CMD_SENT, host->base + REG_CLEAR);
+	writel(cmd->arg, host->base + REG_ARGUMENT);
+
+	cmdctrl = cmd->opcode & CMD_IDX_MASK;
+	if (cmdctrl == SD_APP_SET_BUS_WIDTH    || cmdctrl == SD_APP_OP_COND   ||
+	    cmdctrl == SD_APP_SEND_SCR         || cmdctrl == SD_APP_SD_STATUS ||
+	    cmdctrl == SD_APP_SEND_NUM_WR_BLKS)
+		cmdctrl |= CMD_APP_CMD;
+
+	if (cmd->flags & MMC_RSP_PRESENT)
+		cmdctrl |= CMD_NEED_RSP;
+
+	if (cmd->flags & MMC_RSP_136)
+		cmdctrl |= CMD_LONG_RSP;
+
+	writel(cmdctrl | CMD_EN, host->base + REG_COMMAND);
+
+	if (moxart_wait_for_status(host, MASK_RSP, &status) == -ETIMEDOUT)
+		cmd->error = -ETIMEDOUT;
+
+	if (status & RSP_TIMEOUT) {
+		cmd->error = -ETIMEDOUT;
+		return;
+	}
+	if (status & RSP_CRC_FAIL) {
+		cmd->error = -EIO;
+		return;
+	}
+	if (status & RSP_CRC_OK) {
+		if (cmd->flags & MMC_RSP_136) {
+			cmd->resp[3] = readl(host->base + REG_RESPONSE0);
+			cmd->resp[2] = readl(host->base + REG_RESPONSE1);
+			cmd->resp[1] = readl(host->base + REG_RESPONSE2);
+			cmd->resp[0] = readl(host->base + REG_RESPONSE3);
+		} else {
+			cmd->resp[0] = readl(host->base + REG_RESPONSE0);
+		}
+	}
+}
+
+static void moxart_dma_complete(void *param)
+{
+	struct moxart_host *host = param;
+
+	complete(&host->dma_complete);
+}
+
+static void moxart_transfer_dma(struct mmc_data *data, struct moxart_host *host)
+{
+	u32 len, dir_data, dir_slave;
+	unsigned long dma_time;
+	struct dma_async_tx_descriptor *desc = NULL;
+	struct dma_chan *dma_chan;
+
+	if (host->data_len == data->bytes_xfered)
+		return;
+
+	if (data->flags & MMC_DATA_WRITE) {
+		dma_chan = host->dma_chan_tx;
+		dir_data = DMA_TO_DEVICE;
+		dir_slave = DMA_MEM_TO_DEV;
+	} else {
+		dma_chan = host->dma_chan_rx;
+		dir_data = DMA_FROM_DEVICE;
+		dir_slave = DMA_DEV_TO_MEM;
+	}
+
+	len = dma_map_sg(dma_chan->device->dev, data->sg,
+			 data->sg_len, dir_data);
+
+	if (len > 0) {
+		desc = dmaengine_prep_slave_sg(dma_chan, data->sg,
+					       len, dir_slave,
+					       DMA_PREP_INTERRUPT |
+					       DMA_CTRL_ACK);
+	} else {
+		dev_err(mmc_dev(host->mmc), "dma_map_sg returned zero length\n");
+	}
+
+	if (desc) {
+		host->tx_desc = desc;
+		desc->callback = moxart_dma_complete;
+		desc->callback_param = host;
+		dmaengine_submit(desc);
+		dma_async_issue_pending(dma_chan);
+	}
+
+	data->bytes_xfered += host->data_remain;
+
+	dma_time = wait_for_completion_interruptible_timeout(
+		   &host->dma_complete, host->timeout);
+
+	dma_unmap_sg(dma_chan->device->dev,
+		     data->sg, data->sg_len,
+		     dir_data);
+}
+
+
+static void moxart_transfer_pio(struct moxart_host *host)
+{
+	struct mmc_data *data = host->mrq->cmd->data;
+	u32 *sgp, len = 0, remain, status;
+
+	if (host->data_len == data->bytes_xfered)
+		return;
+
+	sgp = sg_virt(host->cur_sg);
+	remain = host->data_remain;
+
+	if (data->flags & MMC_DATA_WRITE) {
+		while (remain > 0) {
+			if (moxart_wait_for_status(host, FIFO_URUN, &status)
+			     == -ETIMEDOUT) {
+				data->error = -ETIMEDOUT;
+				complete(&host->pio_complete);
+				return;
+			}
+			for (len = 0; len < remain && len < host->fifo_width;) {
+				iowrite32(*sgp, host->base + REG_DATA_WINDOW);
+				sgp++;
+				len += 4;
+			}
+			remain -= len;
+		}
+
+	} else {
+		while (remain > 0) {
+			if (moxart_wait_for_status(host, FIFO_ORUN, &status)
+			    == -ETIMEDOUT) {
+				data->error = -ETIMEDOUT;
+				complete(&host->pio_complete);
+				return;
+			}
+			for (len = 0; len < remain && len < host->fifo_width;) {
+				/* SCR data must be read in big endian. */
+				if (data->mrq->cmd->opcode == SD_APP_SEND_SCR)
+					*sgp = ioread32be(host->base +
+							  REG_DATA_WINDOW);
+				else
+					*sgp = ioread32(host->base +
+							REG_DATA_WINDOW);
+				sgp++;
+				len += 4;
+			}
+			remain -= len;
+		}
+	}
+
+	data->bytes_xfered += host->data_remain - remain;
+	host->data_remain = remain;
+
+	if (host->data_len != data->bytes_xfered)
+		moxart_next_sg(host);
+	else
+		complete(&host->pio_complete);
+}
+
+static void moxart_prepare_data(struct moxart_host *host)
+{
+	struct mmc_data *data = host->mrq->cmd->data;
+	u32 datactrl;
+	int blksz_bits;
+
+	if (!data)
+		return;
+
+	host->data_len = data->blocks * data->blksz;
+	blksz_bits = ffs(data->blksz) - 1;
+	BUG_ON(1 << blksz_bits != data->blksz);
+
+	moxart_init_sg(host, data);
+
+	datactrl = DCR_DATA_EN | (blksz_bits & DCR_BLK_SIZE);
+
+	if (data->flags & MMC_DATA_WRITE)
+		datactrl |= DCR_DATA_WRITE;
+
+	if ((host->data_len > host->fifo_width) && host->have_dma)
+		datactrl |= DCR_DMA_EN;
+
+	writel(DCR_DATA_FIFO_RESET, host->base + REG_DATA_CONTROL);
+	writel(MASK_DATA | FIFO_URUN | FIFO_ORUN, host->base + REG_CLEAR);
+	writel(host->rate, host->base + REG_DATA_TIMER);
+	writel(host->data_len, host->base + REG_DATA_LENGTH);
+	writel(datactrl, host->base + REG_DATA_CONTROL);
+}
+
+static void moxart_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct moxart_host *host = mmc_priv(mmc);
+	unsigned long pio_time, flags;
+	u32 status;
+
+	spin_lock_irqsave(&host->lock, flags);
+
+	init_completion(&host->dma_complete);
+	init_completion(&host->pio_complete);
+
+	host->mrq = mrq;
+
+	if (readl(host->base + REG_STATUS) & CARD_DETECT) {
+		mrq->cmd->error = -ETIMEDOUT;
+		goto request_done;
+	}
+
+	moxart_prepare_data(host);
+	moxart_send_command(host, host->mrq->cmd);
+
+	if (mrq->cmd->data) {
+		if ((host->data_len > host->fifo_width) && host->have_dma) {
+
+			writel(CARD_CHANGE, host->base + REG_INTERRUPT_MASK);
+
+			spin_unlock_irqrestore(&host->lock, flags);
+
+			moxart_transfer_dma(mrq->cmd->data, host);
+
+			spin_lock_irqsave(&host->lock, flags);
+		} else {
+
+			writel(MASK_INTR_PIO, host->base + REG_INTERRUPT_MASK);
+
+			spin_unlock_irqrestore(&host->lock, flags);
+
+			/* PIO transfers start from interrupt. */
+			pio_time = wait_for_completion_interruptible_timeout(
+				   &host->pio_complete, host->timeout);
+
+			spin_lock_irqsave(&host->lock, flags);
+		}
+
+		if (host->is_removed) {
+			dev_err(mmc_dev(host->mmc), "card removed\n");
+			mrq->cmd->error = -ETIMEDOUT;
+			goto request_done;
+		}
+
+		if (moxart_wait_for_status(host, MASK_DATA, &status)
+		    == -ETIMEDOUT) {
+			mrq->cmd->data->error = -ETIMEDOUT;
+			goto request_done;
+		}
+
+		if (status & DATA_CRC_FAIL)
+			mrq->cmd->data->error = -ETIMEDOUT;
+
+		if (mrq->cmd->data->stop)
+			moxart_send_command(host, mrq->cmd->data->stop);
+	}
+
+request_done:
+	spin_unlock_irqrestore(&host->lock, flags);
+	mmc_request_done(host->mmc, mrq);
+}
+
+static irqreturn_t moxart_irq(int irq, void *devid)
+{
+	struct moxart_host *host = (struct moxart_host *)devid;
+	u32 status;
+	unsigned long flags;
+
+	spin_lock_irqsave(&host->lock, flags);
+
+	status = readl(host->base + REG_STATUS);
+	if (status & CARD_CHANGE) {
+		host->is_removed = status & CARD_DETECT;
+		if (host->is_removed && host->have_dma) {
+			dmaengine_terminate_all(host->dma_chan_tx);
+			dmaengine_terminate_all(host->dma_chan_rx);
+		}
+		host->mrq = NULL;
+		writel(MASK_INTR_PIO, host->base + REG_CLEAR);
+		writel(CARD_CHANGE, host->base + REG_INTERRUPT_MASK);
+		mmc_detect_change(host->mmc, 0);
+	}
+	if (status & (FIFO_ORUN | FIFO_URUN) && host->mrq)
+		moxart_transfer_pio(host);
+
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+static void moxart_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
+{
+	struct moxart_host *host = mmc_priv(mmc);
+	unsigned long flags;
+	u8 power, div;
+	u32 ctrl;
+
+	spin_lock_irqsave(&host->lock, flags);
+
+	if (ios->clock) {
+		for (div = 0; div < CLK_DIV_MASK; ++div) {
+			if (ios->clock >= host->sysclk / (2 * (div + 1)))
+				break;
+		}
+		ctrl = CLK_SD | div;
+		host->rate = host->sysclk / (2 * (div + 1));
+		if (host->rate > host->sysclk)
+			ctrl |= CLK_HISPD;
+		writel(ctrl, host->base + REG_CLOCK_CONTROL);
+	}
+
+	if (ios->power_mode == MMC_POWER_OFF) {
+		writel(readl(host->base + REG_POWER_CONTROL) & ~SD_POWER_ON,
+		       host->base + REG_POWER_CONTROL);
+	} else {
+		if (ios->vdd < MIN_POWER)
+			power = 0;
+		else
+			power = ios->vdd - MIN_POWER;
+
+		writel(SD_POWER_ON | (u32) power,
+		       host->base + REG_POWER_CONTROL);
+	}
+
+	switch (ios->bus_width) {
+	case MMC_BUS_WIDTH_4:
+		writel(BUS_WIDTH_4, host->base + REG_BUS_WIDTH);
+		break;
+	case MMC_BUS_WIDTH_8:
+		writel(BUS_WIDTH_8, host->base + REG_BUS_WIDTH);
+		break;
+	default:
+		writel(BUS_WIDTH_1, host->base + REG_BUS_WIDTH);
+		break;
+	}
+
+	spin_unlock_irqrestore(&host->lock, flags);
+}
+
+
+static int moxart_get_ro(struct mmc_host *mmc)
+{
+	struct moxart_host *host = mmc_priv(mmc);
+
+	return !!(readl(host->base + REG_STATUS) & WRITE_PROT);
+}
+
+static struct mmc_host_ops moxart_ops = {
+	.request = moxart_request,
+	.set_ios = moxart_set_ios,
+	.get_ro = moxart_get_ro,
+};
+
+static int moxart_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *node = dev->of_node;
+	struct resource res_mmc;
+	struct mmc_host *mmc;
+	struct moxart_host *host = NULL;
+	struct dma_slave_config cfg;
+	struct clk *clk;
+	void __iomem *reg_mmc;
+	dma_cap_mask_t mask;
+	int irq, ret;
+	u32 i;
+
+	mmc = mmc_alloc_host(sizeof(struct moxart_host), dev);
+	if (!mmc) {
+		dev_err(dev, "mmc_alloc_host failed\n");
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = of_address_to_resource(node, 0, &res_mmc);
+	if (ret) {
+		dev_err(dev, "of_address_to_resource failed\n");
+		goto out;
+	}
+
+	irq = irq_of_parse_and_map(node, 0);
+	if (irq <= 0) {
+		dev_err(dev, "irq_of_parse_and_map failed\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	clk = of_clk_get(node, 0);
+	if (IS_ERR(clk)) {
+		dev_err(dev, "of_clk_get failed\n");
+		ret = PTR_ERR(clk);
+		goto out;
+	}
+
+	reg_mmc = devm_ioremap_resource(dev, &res_mmc);
+	if (IS_ERR(reg_mmc)) {
+		ret = PTR_ERR(reg_mmc);
+		goto out;
+	}
+
+	mmc_of_parse(mmc);
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+
+	host = mmc_priv(mmc);
+	host->mmc = mmc;
+	host->base = reg_mmc;
+	host->reg_phys = res_mmc.start;
+	host->timeout = msecs_to_jiffies(1000);
+	host->sysclk = clk_get_rate(clk);
+	host->fifo_width = readl(host->base + REG_FEATURE) << 2;
+	host->dma_chan_tx = of_dma_request_slave_channel(node, "tx");
+	host->dma_chan_rx = of_dma_request_slave_channel(node, "rx");
+
+	spin_lock_init(&host->lock);
+
+	mmc->ops = &moxart_ops;
+	mmc->f_max = DIV_ROUND_CLOSEST(host->sysclk, 2);
+	mmc->f_min = DIV_ROUND_CLOSEST(host->sysclk, CLK_DIV_MASK * 2);
+	mmc->ocr_avail = 0xffff00;	/* Support 2.0v - 3.6v power. */
+
+	if (IS_ERR(host->dma_chan_tx) || IS_ERR(host->dma_chan_rx)) {
+		dev_dbg(dev, "PIO mode transfer enabled\n");
+		host->have_dma = false;
+	} else {
+		dev_dbg(dev, "DMA channels found (%p,%p)\n",
+			 host->dma_chan_tx, host->dma_chan_rx);
+		host->have_dma = true;
+
+		cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+
+		cfg.direction = DMA_MEM_TO_DEV;
+		cfg.src_addr = 0;
+		cfg.dst_addr = host->reg_phys + REG_DATA_WINDOW;
+		dmaengine_slave_config(host->dma_chan_tx, &cfg);
+
+		cfg.direction = DMA_DEV_TO_MEM;
+		cfg.src_addr = host->reg_phys + REG_DATA_WINDOW;
+		cfg.dst_addr = 0;
+		dmaengine_slave_config(host->dma_chan_rx, &cfg);
+	}
+
+	switch ((readl(host->base + REG_BUS_WIDTH) >> 3) & 3) {
+	case 1:
+		mmc->caps |= MMC_CAP_4_BIT_DATA;
+		break;
+	case 2:
+		mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA;
+		break;
+	default:
+		break;
+	}
+
+	writel(0, host->base + REG_INTERRUPT_MASK);
+
+	writel(CMD_SDC_RESET, host->base + REG_COMMAND);
+	for (i = 0; i < MAX_RETRIES; i++) {
+		if (!(readl(host->base + REG_COMMAND) & CMD_SDC_RESET))
+			break;
+		udelay(5);
+	}
+
+	ret = devm_request_irq(dev, irq, moxart_irq, 0, "moxart-mmc", host);
+	if (ret)
+		goto out;
+
+	dev_set_drvdata(dev, mmc);
+	mmc_add_host(mmc);
+
+	dev_dbg(dev, "IRQ=%d, FIFO is %d bytes\n", irq, host->fifo_width);
+
+	return 0;
+
+out:
+	if (mmc)
+		mmc_free_host(mmc);
+	return ret;
+}
+
+static int moxart_remove(struct platform_device *pdev)
+{
+	struct mmc_host *mmc = dev_get_drvdata(&pdev->dev);
+	struct moxart_host *host = mmc_priv(mmc);
+
+	dev_set_drvdata(&pdev->dev, NULL);
+
+	if (mmc) {
+		if (!IS_ERR(host->dma_chan_tx))
+			dma_release_channel(host->dma_chan_tx);
+		if (!IS_ERR(host->dma_chan_rx))
+			dma_release_channel(host->dma_chan_rx);
+		mmc_remove_host(mmc);
+		mmc_free_host(mmc);
+
+		writel(0, host->base + REG_INTERRUPT_MASK);
+		writel(0, host->base + REG_POWER_CONTROL);
+		writel(readl(host->base + REG_CLOCK_CONTROL) | CLK_OFF,
+		       host->base + REG_CLOCK_CONTROL);
+	}
+
+	kfree(host);
+
+	return 0;
+}
+
+static const struct of_device_id moxart_mmc_match[] = {
+	{ .compatible = "moxa,moxart-mmc" },
+	{ .compatible = "faraday,ftsdc010" },
+	{ }
+};
+
+static struct platform_driver moxart_mmc_driver = {
+	.probe      = moxart_probe,
+	.remove     = moxart_remove,
+	.driver     = {
+		.name		= "mmc-moxart",
+		.owner		= THIS_MODULE,
+		.of_match_table	= moxart_mmc_match,
+	},
+};
+module_platform_driver(moxart_mmc_driver);
+
+MODULE_ALIAS("platform:mmc-moxart");
+MODULE_DESCRIPTION("MOXA ART MMC driver");
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Jonas Jensen <jonas.jensen@gmail.com>");
diff --git a/drivers/mmc/host/mvsdio.c b/drivers/mmc/host/mvsdio.c
index 45aa2206741d..6b4c5ad3b393 100644
--- a/drivers/mmc/host/mvsdio.c
+++ b/drivers/mmc/host/mvsdio.c
@@ -79,11 +79,11 @@ static int mvsd_setup_data(struct mvsd_host *host, struct mmc_data *data)
 		unsigned long t = jiffies + HZ;
 		unsigned int hw_state,  count = 0;
 		do {
+			hw_state = mvsd_read(MVSD_HW_STATE);
 			if (time_after(jiffies, t)) {
 				dev_warn(host->dev, "FIFO_EMPTY bit missing\n");
 				break;
 			}
-			hw_state = mvsd_read(MVSD_HW_STATE);
 			count++;
 		} while (!(hw_state & (1 << 13)));
 		dev_dbg(host->dev, "*** wait for FIFO_EMPTY bit "
@@ -354,6 +354,20 @@ static irqreturn_t mvsd_irq(int irq, void *dev)
 		intr_status, mvsd_read(MVSD_NOR_INTR_EN),
 		mvsd_read(MVSD_HW_STATE));
 
+	/*
+	 * It looks like, SDIO IP can issue one late, spurious irq
+	 * although all irqs should be disabled. To work around this,
+	 * bail out early, if we didn't expect any irqs to occur.
+	 */
+	if (!mvsd_read(MVSD_NOR_INTR_EN) && !mvsd_read(MVSD_ERR_INTR_EN)) {
+		dev_dbg(host->dev, "spurious irq detected intr 0x%04x intr_en 0x%04x erri 0x%04x erri_en 0x%04x\n",
+			mvsd_read(MVSD_NOR_INTR_STATUS),
+			mvsd_read(MVSD_NOR_INTR_EN),
+			mvsd_read(MVSD_ERR_INTR_STATUS),
+			mvsd_read(MVSD_ERR_INTR_EN));
+		return IRQ_HANDLED;
+	}
+
 	spin_lock(&host->lock);
 
 	/* PIO handling, if needed. Messy business... */
@@ -801,10 +815,10 @@ static int mvsd_probe(struct platform_device *pdev)
 		goto out;
 
 	if (!(mmc->caps & MMC_CAP_NEEDS_POLL))
-		dev_notice(&pdev->dev, "using GPIO for card detection\n");
+		dev_dbg(&pdev->dev, "using GPIO for card detection\n");
 	else
-		dev_notice(&pdev->dev,
-			   "lacking card detect (fall back to polling)\n");
+		dev_dbg(&pdev->dev, "lacking card detect (fall back to polling)\n");
+
 	return 0;
 
 out:
diff --git a/drivers/mmc/host/mxcmmc.c b/drivers/mmc/host/mxcmmc.c
index f7199c83f5cf..ed1cb93c3784 100644
--- a/drivers/mmc/host/mxcmmc.c
+++ b/drivers/mmc/host/mxcmmc.c
@@ -124,9 +124,8 @@ enum mxcmci_type {
 
 struct mxcmci_host {
 	struct mmc_host		*mmc;
-	struct resource		*res;
 	void __iomem		*base;
-	int			irq;
+	dma_addr_t		phys_base;
 	int			detect_irq;
 	struct dma_chan		*dma;
 	struct dma_async_tx_descriptor *desc;
@@ -154,8 +153,6 @@ struct mxcmci_host {
 	struct work_struct	datawork;
 	spinlock_t		lock;
 
-	struct regulator	*vcc;
-
 	int			burstlen;
 	int			dmareq;
 	struct dma_slave_config dma_slave_config;
@@ -241,37 +238,15 @@ static inline void mxcmci_writew(struct mxcmci_host *host, u16 val, int reg)
 
 static void mxcmci_set_clk_rate(struct mxcmci_host *host, unsigned int clk_ios);
 
-static inline void mxcmci_init_ocr(struct mxcmci_host *host)
-{
-	host->vcc = regulator_get(mmc_dev(host->mmc), "vmmc");
-
-	if (IS_ERR(host->vcc)) {
-		host->vcc = NULL;
-	} else {
-		host->mmc->ocr_avail = mmc_regulator_get_ocrmask(host->vcc);
-		if (host->pdata && host->pdata->ocr_avail)
-			dev_warn(mmc_dev(host->mmc),
-				"pdata->ocr_avail will not be used\n");
-	}
-
-	if (host->vcc == NULL) {
-		/* fall-back to platform data */
-		if (host->pdata && host->pdata->ocr_avail)
-			host->mmc->ocr_avail = host->pdata->ocr_avail;
-		else
-			host->mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-	}
-}
-
-static inline void mxcmci_set_power(struct mxcmci_host *host,
-				    unsigned char power_mode,
-				    unsigned int vdd)
+static void mxcmci_set_power(struct mxcmci_host *host, unsigned int vdd)
 {
-	if (host->vcc) {
-		if (power_mode == MMC_POWER_UP)
-			mmc_regulator_set_ocr(host->mmc, host->vcc, vdd);
-		else if (power_mode == MMC_POWER_OFF)
-			mmc_regulator_set_ocr(host->mmc, host->vcc, 0);
+	if (!IS_ERR(host->mmc->supply.vmmc)) {
+		if (host->power_mode == MMC_POWER_UP)
+			mmc_regulator_set_ocr(host->mmc,
+					      host->mmc->supply.vmmc, vdd);
+		else if (host->power_mode == MMC_POWER_OFF)
+			mmc_regulator_set_ocr(host->mmc,
+					      host->mmc->supply.vmmc, 0);
 	}
 
 	if (host->pdata && host->pdata->setpower)
@@ -299,7 +274,6 @@ static void mxcmci_softreset(struct mxcmci_host *host)
 
 	mxcmci_writew(host, 0xff, MMC_REG_RES_TO);
 }
-static int mxcmci_setup_dma(struct mmc_host *mmc);
 
 #if IS_ENABLED(CONFIG_PPC_MPC512x)
 static inline void buffer_swap32(u32 *buf, int len)
@@ -868,8 +842,8 @@ static int mxcmci_setup_dma(struct mmc_host *mmc)
 	struct mxcmci_host *host = mmc_priv(mmc);
 	struct dma_slave_config *config = &host->dma_slave_config;
 
-	config->dst_addr = host->res->start + MMC_REG_BUFFER_ACCESS;
-	config->src_addr = host->res->start + MMC_REG_BUFFER_ACCESS;
+	config->dst_addr = host->phys_base + MMC_REG_BUFFER_ACCESS;
+	config->src_addr = host->phys_base + MMC_REG_BUFFER_ACCESS;
 	config->dst_addr_width = 4;
 	config->src_addr_width = 4;
 	config->dst_maxburst = host->burstlen;
@@ -911,8 +885,8 @@ static void mxcmci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		host->cmdat &= ~CMD_DAT_CONT_BUS_WIDTH_4;
 
 	if (host->power_mode != ios->power_mode) {
-		mxcmci_set_power(host, ios->power_mode, ios->vdd);
 		host->power_mode = ios->power_mode;
+		mxcmci_set_power(host, ios->vdd);
 
 		if (ios->power_mode == MMC_POWER_ON)
 			host->cmdat |= CMD_DAT_CONT_INIT;
@@ -1040,8 +1014,8 @@ static const struct mmc_host_ops mxcmci_ops = {
 static int mxcmci_probe(struct platform_device *pdev)
 {
 	struct mmc_host *mmc;
-	struct mxcmci_host *host = NULL;
-	struct resource *iores, *r;
+	struct mxcmci_host *host;
+	struct resource *res;
 	int ret = 0, irq;
 	bool dat3_card_detect = false;
 	dma_cap_mask_t mask;
@@ -1052,21 +1026,25 @@ static int mxcmci_probe(struct platform_device *pdev)
 
 	of_id = of_match_device(mxcmci_of_match, &pdev->dev);
 
-	iores = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	irq = platform_get_irq(pdev, 0);
-	if (!iores || irq < 0)
+	if (irq < 0)
 		return -EINVAL;
 
-	r = request_mem_region(iores->start, resource_size(iores), pdev->name);
-	if (!r)
-		return -EBUSY;
+	mmc = mmc_alloc_host(sizeof(*host), &pdev->dev);
+	if (!mmc)
+		return -ENOMEM;
+
+	host = mmc_priv(mmc);
 
-	mmc = mmc_alloc_host(sizeof(struct mxcmci_host), &pdev->dev);
-	if (!mmc) {
-		ret = -ENOMEM;
-		goto out_release_mem;
+	host->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(host->base)) {
+		ret = PTR_ERR(host->base);
+		goto out_free;
 	}
 
+	host->phys_base = res->start;
+
 	ret = mmc_of_parse(mmc);
 	if (ret)
 		goto out_free;
@@ -1084,13 +1062,6 @@ static int mxcmci_probe(struct platform_device *pdev)
 	mmc->max_req_size = mmc->max_blk_size * mmc->max_blk_count;
 	mmc->max_seg_size = mmc->max_req_size;
 
-	host = mmc_priv(mmc);
-	host->base = ioremap(r->start, resource_size(r));
-	if (!host->base) {
-		ret = -ENOMEM;
-		goto out_free;
-	}
-
 	if (of_id) {
 		const struct platform_device_id *id_entry = of_id->data;
 		host->devtype = id_entry->driver_data;
@@ -1112,7 +1083,14 @@ static int mxcmci_probe(struct platform_device *pdev)
 			&& !of_property_read_bool(pdev->dev.of_node, "cd-gpios"))
 		dat3_card_detect = true;
 
-	mxcmci_init_ocr(host);
+	ret = mmc_regulator_get_supply(mmc);
+	if (ret) {
+		if (pdata && ret != -EPROBE_DEFER)
+			mmc->ocr_avail = pdata->ocr_avail ? :
+				MMC_VDD_32_33 | MMC_VDD_33_34;
+		else
+			goto out_free;
+	}
 
 	if (dat3_card_detect)
 		host->default_irq_mask =
@@ -1120,19 +1098,16 @@ static int mxcmci_probe(struct platform_device *pdev)
 	else
 		host->default_irq_mask = 0;
 
-	host->res = r;
-	host->irq = irq;
-
 	host->clk_ipg = devm_clk_get(&pdev->dev, "ipg");
 	if (IS_ERR(host->clk_ipg)) {
 		ret = PTR_ERR(host->clk_ipg);
-		goto out_iounmap;
+		goto out_free;
 	}
 
 	host->clk_per = devm_clk_get(&pdev->dev, "per");
 	if (IS_ERR(host->clk_per)) {
 		ret = PTR_ERR(host->clk_per);
-		goto out_iounmap;
+		goto out_free;
 	}
 
 	clk_prepare_enable(host->clk_per);
@@ -1159,9 +1134,9 @@ static int mxcmci_probe(struct platform_device *pdev)
 	if (!host->pdata) {
 		host->dma = dma_request_slave_channel(&pdev->dev, "rx-tx");
 	} else {
-		r = platform_get_resource(pdev, IORESOURCE_DMA, 0);
-		if (r) {
-			host->dmareq = r->start;
+		res = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+		if (res) {
+			host->dmareq = res->start;
 			host->dma_data.peripheral_type = IMX_DMATYPE_SDHC;
 			host->dma_data.priority = DMA_PRIO_LOW;
 			host->dma_data.dma_request = host->dmareq;
@@ -1178,7 +1153,8 @@ static int mxcmci_probe(struct platform_device *pdev)
 
 	INIT_WORK(&host->datawork, mxcmci_datawork);
 
-	ret = request_irq(host->irq, mxcmci_irq, 0, DRIVER_NAME, host);
+	ret = devm_request_irq(&pdev->dev, irq, mxcmci_irq, 0,
+			       dev_name(&pdev->dev), host);
 	if (ret)
 		goto out_free_dma;
 
@@ -1188,7 +1164,7 @@ static int mxcmci_probe(struct platform_device *pdev)
 		ret = host->pdata->init(&pdev->dev, mxcmci_detect_irq,
 				host->mmc);
 		if (ret)
-			goto out_free_irq;
+			goto out_free_dma;
 	}
 
 	init_timer(&host->watchdog);
@@ -1199,20 +1175,17 @@ static int mxcmci_probe(struct platform_device *pdev)
 
 	return 0;
 
-out_free_irq:
-	free_irq(host->irq, host);
 out_free_dma:
 	if (host->dma)
 		dma_release_channel(host->dma);
+
 out_clk_put:
 	clk_disable_unprepare(host->clk_per);
 	clk_disable_unprepare(host->clk_ipg);
-out_iounmap:
-	iounmap(host->base);
+
 out_free:
 	mmc_free_host(mmc);
-out_release_mem:
-	release_mem_region(iores->start, resource_size(iores));
+
 	return ret;
 }
 
@@ -1223,30 +1196,21 @@ static int mxcmci_remove(struct platform_device *pdev)
 
 	mmc_remove_host(mmc);
 
-	if (host->vcc)
-		regulator_put(host->vcc);
-
 	if (host->pdata && host->pdata->exit)
 		host->pdata->exit(&pdev->dev, mmc);
 
-	free_irq(host->irq, host);
-	iounmap(host->base);
-
 	if (host->dma)
 		dma_release_channel(host->dma);
 
 	clk_disable_unprepare(host->clk_per);
 	clk_disable_unprepare(host->clk_ipg);
 
-	release_mem_region(host->res->start, resource_size(host->res));
-
 	mmc_free_host(mmc);
 
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int mxcmci_suspend(struct device *dev)
+static int __maybe_unused mxcmci_suspend(struct device *dev)
 {
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct mxcmci_host *host = mmc_priv(mmc);
@@ -1256,7 +1220,7 @@ static int mxcmci_suspend(struct device *dev)
 	return 0;
 }
 
-static int mxcmci_resume(struct device *dev)
+static int __maybe_unused mxcmci_resume(struct device *dev)
 {
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct mxcmci_host *host = mmc_priv(mmc);
@@ -1266,11 +1230,7 @@ static int mxcmci_resume(struct device *dev)
 	return 0;
 }
 
-static const struct dev_pm_ops mxcmci_pm_ops = {
-	.suspend	= mxcmci_suspend,
-	.resume		= mxcmci_resume,
-};
-#endif
+static SIMPLE_DEV_PM_OPS(mxcmci_pm_ops, mxcmci_suspend, mxcmci_resume);
 
 static struct platform_driver mxcmci_driver = {
 	.probe		= mxcmci_probe,
@@ -1279,9 +1239,7 @@ static struct platform_driver mxcmci_driver = {
 	.driver		= {
 		.name		= DRIVER_NAME,
 		.owner		= THIS_MODULE,
-#ifdef CONFIG_PM
 		.pm	= &mxcmci_pm_ops,
-#endif
 		.of_match_table	= mxcmci_of_match,
 	}
 };
diff --git a/drivers/mmc/host/mxs-mmc.c b/drivers/mmc/host/mxs-mmc.c
index 073e871a0fc8..babfea03ba8a 100644
--- a/drivers/mmc/host/mxs-mmc.c
+++ b/drivers/mmc/host/mxs-mmc.c
@@ -70,6 +70,7 @@ struct mxs_mmc_host {
 	unsigned char			bus_width;
 	spinlock_t			lock;
 	int				sdio_irq_en;
+	bool				broken_cd;
 };
 
 static int mxs_mmc_get_cd(struct mmc_host *mmc)
@@ -78,6 +79,9 @@ static int mxs_mmc_get_cd(struct mmc_host *mmc)
 	struct mxs_ssp *ssp = &host->ssp;
 	int present, ret;
 
+	if (host->broken_cd)
+		return -ENOSYS;
+
 	ret = mmc_gpio_get_cd(mmc);
 	if (ret >= 0)
 		return ret;
@@ -568,6 +572,7 @@ static int mxs_mmc_probe(struct platform_device *pdev)
 {
 	const struct of_device_id *of_id =
 			of_match_device(mxs_mmc_dt_ids, &pdev->dev);
+	struct device_node *np = pdev->dev.of_node;
 	struct mxs_mmc_host *host;
 	struct mmc_host *mmc;
 	struct resource *iores;
@@ -634,6 +639,8 @@ static int mxs_mmc_probe(struct platform_device *pdev)
 	mmc->caps = MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED |
 		    MMC_CAP_SDIO_IRQ | MMC_CAP_NEEDS_POLL;
 
+	host->broken_cd = of_property_read_bool(np, "broken-cd");
+
 	mmc->f_min = 400000;
 	mmc->f_max = 288000000;
 
diff --git a/drivers/mmc/host/omap.c b/drivers/mmc/host/omap.c
index 5c2e58b29305..81974ecdfcbc 100644
--- a/drivers/mmc/host/omap.c
+++ b/drivers/mmc/host/omap.c
@@ -177,7 +177,7 @@ static void mmc_omap_fclk_offdelay(struct mmc_omap_slot *slot)
 	unsigned long tick_ns;
 
 	if (slot != NULL && slot->host->fclk_enabled && slot->fclk_freq > 0) {
-		tick_ns = (1000000000 + slot->fclk_freq - 1) / slot->fclk_freq;
+		tick_ns = DIV_ROUND_UP(NSEC_PER_SEC, slot->fclk_freq);
 		ndelay(8 * tick_ns);
 	}
 }
@@ -435,7 +435,7 @@ static void mmc_omap_send_stop_work(struct work_struct *work)
 	struct mmc_data *data = host->stop_data;
 	unsigned long tick_ns;
 
-	tick_ns = (1000000000 + slot->fclk_freq - 1)/slot->fclk_freq;
+	tick_ns = DIV_ROUND_UP(NSEC_PER_SEC, slot->fclk_freq);
 	ndelay(8*tick_ns);
 
 	mmc_omap_start_command(host, data->stop);
@@ -477,7 +477,7 @@ mmc_omap_send_abort(struct mmc_omap_host *host, int maxloops)
 	u16 stat = 0;
 
 	/* Sending abort takes 80 clocks. Have some extra and round up */
-	timeout = (120*1000000 + slot->fclk_freq - 1)/slot->fclk_freq;
+	timeout = DIV_ROUND_UP(120 * USEC_PER_SEC, slot->fclk_freq);
 	restarts = 0;
 	while (restarts < maxloops) {
 		OMAP_MMC_WRITE(host, STAT, 0xFFFF);
@@ -677,8 +677,8 @@ mmc_omap_xfer_data(struct mmc_omap_host *host, int write)
 	if (n > host->buffer_bytes_left)
 		n = host->buffer_bytes_left;
 
-	nwords = n / 2;
-	nwords += n & 1; /* handle odd number of bytes to transfer */
+	/* Round up to handle odd number of bytes to transfer */
+	nwords = DIV_ROUND_UP(n, 2);
 
 	host->buffer_bytes_left -= n;
 	host->total_bytes_left -= n;
diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index e91ee21549d0..6b7b75585926 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -31,7 +31,7 @@
 #include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/of_device.h>
-#include <linux/omap-dma.h>
+#include <linux/omap-dmaengine.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/core.h>
 #include <linux/mmc/mmc.h>
@@ -582,7 +582,7 @@ static void omap_hsmmc_set_clock(struct omap_hsmmc_host *host)
 	 *	- MMC/SD clock coming out of controller > 25MHz
 	 */
 	if ((mmc_slot(host).features & HSMMC_HAS_HSPE_SUPPORT) &&
-	    (ios->timing != MMC_TIMING_UHS_DDR50) &&
+	    (ios->timing != MMC_TIMING_MMC_DDR52) &&
 	    ((OMAP_HSMMC_READ(host->base, CAPA) & HSS) == HSS)) {
 		regval = OMAP_HSMMC_READ(host->base, HCTL);
 		if (clkdiv && (clk_get_rate(host->fclk)/clkdiv) > 25000000)
@@ -602,7 +602,7 @@ static void omap_hsmmc_set_bus_width(struct omap_hsmmc_host *host)
 	u32 con;
 
 	con = OMAP_HSMMC_READ(host->base, CON);
-	if (ios->timing == MMC_TIMING_UHS_DDR50)
+	if (ios->timing == MMC_TIMING_MMC_DDR52)
 		con |= DDR;	/* configure in DDR mode */
 	else
 		con &= ~DDR;
@@ -920,16 +920,17 @@ omap_hsmmc_xfer_done(struct omap_hsmmc_host *host, struct mmc_data *data)
 static void
 omap_hsmmc_cmd_done(struct omap_hsmmc_host *host, struct mmc_command *cmd)
 {
-	host->cmd = NULL;
-
 	if (host->mrq->sbc && (host->cmd == host->mrq->sbc) &&
 	    !host->mrq->sbc->error && !(host->flags & AUTO_CMD23)) {
+		host->cmd = NULL;
 		omap_hsmmc_start_dma_transfer(host);
 		omap_hsmmc_start_command(host, host->mrq->cmd,
 						host->mrq->data);
 		return;
 	}
 
+	host->cmd = NULL;
+
 	if (cmd->flags & MMC_RSP_PRESENT) {
 		if (cmd->flags & MMC_RSP_136) {
 			/* response type 2 */
@@ -1851,6 +1852,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 	unsigned tx_req, rx_req;
 	struct pinctrl *pinctrl;
 	const struct omap_mmc_of_data *data;
+	void __iomem *base;
 
 	match = of_match_device(of_match_ptr(omap_mmc_of_match), &pdev->dev);
 	if (match) {
@@ -1881,9 +1883,9 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 	if (res == NULL || irq < 0)
 		return -ENXIO;
 
-	res = request_mem_region(res->start, resource_size(res), pdev->name);
-	if (res == NULL)
-		return -EBUSY;
+	base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(base))
+		return PTR_ERR(base);
 
 	ret = omap_hsmmc_gpio_init(pdata);
 	if (ret)
@@ -1904,7 +1906,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 	host->irq	= irq;
 	host->slot_id	= 0;
 	host->mapbase	= res->start + pdata->reg_offset;
-	host->base	= ioremap(host->mapbase, SZ_4K);
+	host->base	= base + pdata->reg_offset;
 	host->power_mode = MMC_POWER_OFF;
 	host->next_data.cookie = 1;
 	host->pbias_enabled = 0;
@@ -1922,7 +1924,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 
 	spin_lock_init(&host->irq_lock);
 
-	host->fclk = clk_get(&pdev->dev, "fck");
+	host->fclk = devm_clk_get(&pdev->dev, "fck");
 	if (IS_ERR(host->fclk)) {
 		ret = PTR_ERR(host->fclk);
 		host->fclk = NULL;
@@ -1941,7 +1943,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 
 	omap_hsmmc_context_save(host);
 
-	host->dbclk = clk_get(&pdev->dev, "mmchsdb_fck");
+	host->dbclk = devm_clk_get(&pdev->dev, "mmchsdb_fck");
 	/*
 	 * MMC can still work without debounce clock.
 	 */
@@ -1949,7 +1951,6 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 		host->dbclk = NULL;
 	} else if (clk_prepare_enable(host->dbclk) != 0) {
 		dev_warn(mmc_dev(host->mmc), "Failed to enable debounce clk\n");
-		clk_put(host->dbclk);
 		host->dbclk = NULL;
 	}
 
@@ -2018,7 +2019,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 	}
 
 	/* Request IRQ for MMC operations */
-	ret = request_irq(host->irq, omap_hsmmc_irq, 0,
+	ret = devm_request_irq(&pdev->dev, host->irq, omap_hsmmc_irq, 0,
 			mmc_hostname(mmc), host);
 	if (ret) {
 		dev_err(mmc_dev(host->mmc), "Unable to grab HSMMC IRQ\n");
@@ -2029,7 +2030,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 		if (pdata->init(&pdev->dev) != 0) {
 			dev_err(mmc_dev(host->mmc),
 				"Unable to configure MMC IRQs\n");
-			goto err_irq_cd_init;
+			goto err_irq;
 		}
 	}
 
@@ -2044,9 +2045,9 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 
 	/* Request IRQ for card detect */
 	if ((mmc_slot(host).card_detect_irq)) {
-		ret = request_threaded_irq(mmc_slot(host).card_detect_irq,
-					   NULL,
-					   omap_hsmmc_detect,
+		ret = devm_request_threaded_irq(&pdev->dev,
+						mmc_slot(host).card_detect_irq,
+						NULL, omap_hsmmc_detect,
 					   IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
 					   mmc_hostname(mmc), host);
 		if (ret) {
@@ -2089,15 +2090,12 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 
 err_slot_name:
 	mmc_remove_host(mmc);
-	free_irq(mmc_slot(host).card_detect_irq, host);
 err_irq_cd:
 	if (host->use_reg)
 		omap_hsmmc_reg_put(host);
 err_reg:
 	if (host->pdata->cleanup)
 		host->pdata->cleanup(&pdev->dev);
-err_irq_cd_init:
-	free_irq(host->irq, host);
 err_irq:
 	if (host->tx_chan)
 		dma_release_channel(host->tx_chan);
@@ -2105,27 +2103,19 @@ err_irq:
 		dma_release_channel(host->rx_chan);
 	pm_runtime_put_sync(host->dev);
 	pm_runtime_disable(host->dev);
-	clk_put(host->fclk);
-	if (host->dbclk) {
+	if (host->dbclk)
 		clk_disable_unprepare(host->dbclk);
-		clk_put(host->dbclk);
-	}
 err1:
-	iounmap(host->base);
 	mmc_free_host(mmc);
 err_alloc:
 	omap_hsmmc_gpio_free(pdata);
 err:
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (res)
-		release_mem_region(res->start, resource_size(res));
 	return ret;
 }
 
 static int omap_hsmmc_remove(struct platform_device *pdev)
 {
 	struct omap_hsmmc_host *host = platform_get_drvdata(pdev);
-	struct resource *res;
 
 	pm_runtime_get_sync(host->dev);
 	mmc_remove_host(host->mmc);
@@ -2133,9 +2123,6 @@ static int omap_hsmmc_remove(struct platform_device *pdev)
 		omap_hsmmc_reg_put(host);
 	if (host->pdata->cleanup)
 		host->pdata->cleanup(&pdev->dev);
-	free_irq(host->irq, host);
-	if (mmc_slot(host).card_detect_irq)
-		free_irq(mmc_slot(host).card_detect_irq, host);
 
 	if (host->tx_chan)
 		dma_release_channel(host->tx_chan);
@@ -2144,20 +2131,12 @@ static int omap_hsmmc_remove(struct platform_device *pdev)
 
 	pm_runtime_put_sync(host->dev);
 	pm_runtime_disable(host->dev);
-	clk_put(host->fclk);
-	if (host->dbclk) {
+	if (host->dbclk)
 		clk_disable_unprepare(host->dbclk);
-		clk_put(host->dbclk);
-	}
 
 	omap_hsmmc_gpio_free(host->pdata);
-	iounmap(host->base);
 	mmc_free_host(host->mmc);
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (res)
-		release_mem_region(res->start, resource_size(res));
-
 	return 0;
 }
 
diff --git a/drivers/mmc/host/rtsx_pci_sdmmc.c b/drivers/mmc/host/rtsx_pci_sdmmc.c
index 0b9ded13a3ae..0d519649b575 100644
--- a/drivers/mmc/host/rtsx_pci_sdmmc.c
+++ b/drivers/mmc/host/rtsx_pci_sdmmc.c
@@ -236,6 +236,9 @@ static void sd_send_cmd_get_rsp(struct realtek_pci_sdmmc *host,
 	case MMC_RSP_R1:
 		rsp_type = SD_RSP_TYPE_R1;
 		break;
+	case MMC_RSP_R1 & ~MMC_RSP_CRC:
+		rsp_type = SD_RSP_TYPE_R1 | SD_NO_CHECK_CRC7;
+		break;
 	case MMC_RSP_R1B:
 		rsp_type = SD_RSP_TYPE_R1b;
 		break;
@@ -816,6 +819,7 @@ static int sd_set_timing(struct realtek_pci_sdmmc *host, unsigned char timing)
 		rtsx_pci_add_cmd(pcr, WRITE_REG_CMD, CLK_CTL, CLK_LOW_FREQ, 0);
 		break;
 
+	case MMC_TIMING_MMC_DDR52:
 	case MMC_TIMING_UHS_DDR50:
 		rtsx_pci_add_cmd(pcr, WRITE_REG_CMD, SD_CFG1,
 				0x0C | SD_ASYNC_FIFO_NOT_RST,
@@ -896,6 +900,7 @@ static void sdmmc_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		host->vpclk = true;
 		host->double_clk = false;
 		break;
+	case MMC_TIMING_MMC_DDR52:
 	case MMC_TIMING_UHS_DDR50:
 	case MMC_TIMING_UHS_SDR25:
 		host->ssc_depth = RTSX_SSC_DEPTH_1M;
diff --git a/drivers/mmc/host/rtsx_usb_sdmmc.c b/drivers/mmc/host/rtsx_usb_sdmmc.c
index e11fafa6fc6b..5d3766e792f0 100644
--- a/drivers/mmc/host/rtsx_usb_sdmmc.c
+++ b/drivers/mmc/host/rtsx_usb_sdmmc.c
@@ -34,7 +34,8 @@
 #include <linux/mfd/rtsx_usb.h>
 #include <asm/unaligned.h>
 
-#if defined(CONFIG_LEDS_CLASS) || defined(CONFIG_LEDS_CLASS_MODULE)
+#if defined(CONFIG_LEDS_CLASS) || (defined(CONFIG_LEDS_CLASS_MODULE) && \
+		defined(CONFIG_MMC_REALTEK_USB_MODULE))
 #include <linux/leds.h>
 #include <linux/workqueue.h>
 #define RTSX_USB_USE_LEDS_CLASS
@@ -59,7 +60,7 @@ struct rtsx_usb_sdmmc {
 
 	unsigned char		power_mode;
 
-#if defined(CONFIG_LEDS_CLASS) || defined(CONFIG_LEDS_CLASS_MODULE)
+#ifdef RTSX_USB_USE_LEDS_CLASS
 	struct led_classdev	led;
 	char			led_name[32];
 	struct work_struct	led_work;
diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c
index ebb3f392b589..8ce3c28cb76e 100644
--- a/drivers/mmc/host/sdhci-acpi.c
+++ b/drivers/mmc/host/sdhci-acpi.c
@@ -102,11 +102,19 @@ static void sdhci_acpi_int_hw_reset(struct sdhci_host *host)
 }
 
 static const struct sdhci_ops sdhci_acpi_ops_dflt = {
+	.set_clock = sdhci_set_clock,
 	.enable_dma = sdhci_acpi_enable_dma,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static const struct sdhci_ops sdhci_acpi_ops_int = {
+	.set_clock = sdhci_set_clock,
 	.enable_dma = sdhci_acpi_enable_dma,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 	.hw_reset   = sdhci_acpi_int_hw_reset,
 };
 
diff --git a/drivers/mmc/host/sdhci-bcm-kona.c b/drivers/mmc/host/sdhci-bcm-kona.c
index 6f166e63b817..dd780c315a63 100644
--- a/drivers/mmc/host/sdhci-bcm-kona.c
+++ b/drivers/mmc/host/sdhci-bcm-kona.c
@@ -206,9 +206,13 @@ static void sdhci_bcm_kona_init_74_clocks(struct sdhci_host *host,
 }
 
 static struct sdhci_ops sdhci_bcm_kona_ops = {
+	.set_clock = sdhci_set_clock,
 	.get_max_clock = sdhci_bcm_kona_get_max_clk,
 	.get_timeout_clock = sdhci_bcm_kona_get_timeout_clock,
 	.platform_send_init_74_clocks = sdhci_bcm_kona_init_74_clocks,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 	.card_event = sdhci_bcm_kona_card_event,
 };
 
diff --git a/drivers/mmc/host/sdhci-bcm2835.c b/drivers/mmc/host/sdhci-bcm2835.c
index f6d8d67c545f..46af9a439d7b 100644
--- a/drivers/mmc/host/sdhci-bcm2835.c
+++ b/drivers/mmc/host/sdhci-bcm2835.c
@@ -131,8 +131,12 @@ static const struct sdhci_ops bcm2835_sdhci_ops = {
 	.read_l = bcm2835_sdhci_readl,
 	.read_w = bcm2835_sdhci_readw,
 	.read_b = bcm2835_sdhci_readb,
+	.set_clock = sdhci_set_clock,
 	.get_max_clock = sdhci_pltfm_clk_get_max_clock,
 	.get_min_clock = bcm2835_sdhci_get_min_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static const struct sdhci_pltfm_data bcm2835_sdhci_pdata = {
diff --git a/drivers/mmc/host/sdhci-cns3xxx.c b/drivers/mmc/host/sdhci-cns3xxx.c
index f2cc26633cb2..14b74075589a 100644
--- a/drivers/mmc/host/sdhci-cns3xxx.c
+++ b/drivers/mmc/host/sdhci-cns3xxx.c
@@ -30,13 +30,12 @@ static void sdhci_cns3xxx_set_clock(struct sdhci_host *host, unsigned int clock)
 	u16 clk;
 	unsigned long timeout;
 
-	if (clock == host->clock)
-		return;
+	host->mmc->actual_clock = 0;
 
 	sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL);
 
 	if (clock == 0)
-		goto out;
+		return;
 
 	while (host->max_clk / div > clock) {
 		/*
@@ -75,13 +74,14 @@ static void sdhci_cns3xxx_set_clock(struct sdhci_host *host, unsigned int clock)
 
 	clk |= SDHCI_CLOCK_CARD_EN;
 	sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
-out:
-	host->clock = clock;
 }
 
 static const struct sdhci_ops sdhci_cns3xxx_ops = {
 	.get_max_clock	= sdhci_cns3xxx_get_max_clk,
 	.set_clock	= sdhci_cns3xxx_set_clock,
+	.set_bus_width	= sdhci_set_bus_width,
+	.reset          = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static const struct sdhci_pltfm_data sdhci_cns3xxx_pdata = {
@@ -90,8 +90,7 @@ static const struct sdhci_pltfm_data sdhci_cns3xxx_pdata = {
 		  SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK |
 		  SDHCI_QUIRK_INVERTED_WRITE_PROTECT |
 		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN |
-		  SDHCI_QUIRK_BROKEN_TIMEOUT_VAL |
-		  SDHCI_QUIRK_NONSTANDARD_CLOCK,
+		  SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
 };
 
 static int sdhci_cns3xxx_probe(struct platform_device *pdev)
diff --git a/drivers/mmc/host/sdhci-dove.c b/drivers/mmc/host/sdhci-dove.c
index 736d7a2eb7ec..e6278ec007d7 100644
--- a/drivers/mmc/host/sdhci-dove.c
+++ b/drivers/mmc/host/sdhci-dove.c
@@ -21,28 +21,17 @@
 
 #include <linux/clk.h>
 #include <linux/err.h>
-#include <linux/gpio.h>
 #include <linux/io.h>
 #include <linux/mmc/host.h>
 #include <linux/module.h>
 #include <linux/of.h>
-#include <linux/of_gpio.h>
 
 #include "sdhci-pltfm.h"
 
 struct sdhci_dove_priv {
 	struct clk *clk;
-	int gpio_cd;
 };
 
-static irqreturn_t sdhci_dove_carddetect_irq(int irq, void *data)
-{
-	struct sdhci_host *host = data;
-
-	tasklet_schedule(&host->card_tasklet);
-	return IRQ_HANDLED;
-}
-
 static u16 sdhci_dove_readw(struct sdhci_host *host, int reg)
 {
 	u16 ret;
@@ -60,8 +49,6 @@ static u16 sdhci_dove_readw(struct sdhci_host *host, int reg)
 
 static u32 sdhci_dove_readl(struct sdhci_host *host, int reg)
 {
-	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
-	struct sdhci_dove_priv *priv = pltfm_host->priv;
 	u32 ret;
 
 	ret = readl(host->ioaddr + reg);
@@ -71,14 +58,6 @@ static u32 sdhci_dove_readl(struct sdhci_host *host, int reg)
 		/* Mask the support for 3.0V */
 		ret &= ~SDHCI_CAN_VDD_300;
 		break;
-	case SDHCI_PRESENT_STATE:
-		if (gpio_is_valid(priv->gpio_cd)) {
-			if (gpio_get_value(priv->gpio_cd) == 0)
-				ret |= SDHCI_CARD_PRESENT;
-			else
-				ret &= ~SDHCI_CARD_PRESENT;
-		}
-		break;
 	}
 	return ret;
 }
@@ -86,6 +65,10 @@ static u32 sdhci_dove_readl(struct sdhci_host *host, int reg)
 static const struct sdhci_ops sdhci_dove_ops = {
 	.read_w	= sdhci_dove_readw,
 	.read_l	= sdhci_dove_readl,
+	.set_clock = sdhci_set_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static const struct sdhci_pltfm_data sdhci_dove_pdata = {
@@ -113,28 +96,9 @@ static int sdhci_dove_probe(struct platform_device *pdev)
 
 	priv->clk = devm_clk_get(&pdev->dev, NULL);
 
-	if (pdev->dev.of_node) {
-		priv->gpio_cd = of_get_named_gpio(pdev->dev.of_node,
-						  "cd-gpios", 0);
-	} else {
-		priv->gpio_cd = -EINVAL;
-	}
-
-	if (gpio_is_valid(priv->gpio_cd)) {
-		ret = gpio_request(priv->gpio_cd, "sdhci-cd");
-		if (ret) {
-			dev_err(&pdev->dev, "card detect gpio request failed: %d\n",
-				ret);
-			return ret;
-		}
-		gpio_direction_input(priv->gpio_cd);
-	}
-
 	host = sdhci_pltfm_init(pdev, &sdhci_dove_pdata, 0);
-	if (IS_ERR(host)) {
-		ret = PTR_ERR(host);
-		goto err_sdhci_pltfm_init;
-	}
+	if (IS_ERR(host))
+		return PTR_ERR(host);
 
 	pltfm_host = sdhci_priv(host);
 	pltfm_host->priv = priv;
@@ -142,39 +106,20 @@ static int sdhci_dove_probe(struct platform_device *pdev)
 	if (!IS_ERR(priv->clk))
 		clk_prepare_enable(priv->clk);
 
-	sdhci_get_of_property(pdev);
+	ret = mmc_of_parse(host->mmc);
+	if (ret)
+		goto err_sdhci_add;
 
 	ret = sdhci_add_host(host);
 	if (ret)
 		goto err_sdhci_add;
 
-	/*
-	 * We must request the IRQ after sdhci_add_host(), as the tasklet only
-	 * gets setup in sdhci_add_host() and we oops.
-	 */
-	if (gpio_is_valid(priv->gpio_cd)) {
-		ret = request_irq(gpio_to_irq(priv->gpio_cd),
-				  sdhci_dove_carddetect_irq,
-				  IRQF_TRIGGER_FALLING | IRQF_TRIGGER_RISING,
-				  mmc_hostname(host->mmc), host);
-		if (ret) {
-			dev_err(&pdev->dev, "card detect irq request failed: %d\n",
-				ret);
-			goto err_request_irq;
-		}
-	}
-
 	return 0;
 
-err_request_irq:
-	sdhci_remove_host(host, 0);
 err_sdhci_add:
 	if (!IS_ERR(priv->clk))
 		clk_disable_unprepare(priv->clk);
 	sdhci_pltfm_free(pdev);
-err_sdhci_pltfm_init:
-	if (gpio_is_valid(priv->gpio_cd))
-		gpio_free(priv->gpio_cd);
 	return ret;
 }
 
@@ -186,11 +131,6 @@ static int sdhci_dove_remove(struct platform_device *pdev)
 
 	sdhci_pltfm_unregister(pdev);
 
-	if (gpio_is_valid(priv->gpio_cd)) {
-		free_irq(gpio_to_irq(priv->gpio_cd), host);
-		gpio_free(priv->gpio_cd);
-	}
-
 	if (!IS_ERR(priv->clk))
 		clk_disable_unprepare(priv->clk);
 
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c
index b841bb7cd371..ccec0e32590f 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -160,7 +160,6 @@ struct pltfm_imx_data {
 		MULTIBLK_IN_PROCESS, /* exact multiblock cmd in process */
 		WAIT_FOR_INT,        /* sent CMD12, waiting for response INT */
 	} multiblock_status;
-	u32 uhs_mode;
 	u32 is_ddr;
 };
 
@@ -382,7 +381,6 @@ static u16 esdhc_readw_le(struct sdhci_host *host, int reg)
 		if (val & ESDHC_MIX_CTRL_SMPCLK_SEL)
 			ret |= SDHCI_CTRL_TUNED_CLK;
 
-		ret |= (imx_data->uhs_mode & SDHCI_CTRL_UHS_MASK);
 		ret &= ~SDHCI_CTRL_PRESET_VAL_ENABLE;
 
 		return ret;
@@ -429,7 +427,6 @@ static void esdhc_writew_le(struct sdhci_host *host, u16 val, int reg)
 		else
 			new_val &= ~ESDHC_VENDOR_SPEC_VSELECT;
 		writel(new_val, host->ioaddr + ESDHC_VENDOR_SPEC);
-		imx_data->uhs_mode = val & SDHCI_CTRL_UHS_MASK;
 		if (imx_data->socdata->flags & ESDHC_FLAG_MAN_TUNING) {
 			new_val = readl(host->ioaddr + ESDHC_MIX_CTRL);
 			if (val & SDHCI_CTRL_TUNED_CLK)
@@ -600,12 +597,14 @@ static inline void esdhc_pltfm_set_clock(struct sdhci_host *host,
 	u32 temp, val;
 
 	if (clock == 0) {
+		host->mmc->actual_clock = 0;
+
 		if (esdhc_is_usdhc(imx_data)) {
 			val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
 			writel(val & ~ESDHC_VENDOR_SPEC_FRC_SDCLK_ON,
 					host->ioaddr + ESDHC_VENDOR_SPEC);
 		}
-		goto out;
+		return;
 	}
 
 	if (esdhc_is_usdhc(imx_data) && !imx_data->is_ddr)
@@ -645,8 +644,6 @@ static inline void esdhc_pltfm_set_clock(struct sdhci_host *host,
 	}
 
 	mdelay(1);
-out:
-	host->clock = clock;
 }
 
 static unsigned int esdhc_pltfm_get_ro(struct sdhci_host *host)
@@ -668,7 +665,7 @@ static unsigned int esdhc_pltfm_get_ro(struct sdhci_host *host)
 	return -ENOSYS;
 }
 
-static int esdhc_pltfm_bus_width(struct sdhci_host *host, int width)
+static void esdhc_pltfm_set_bus_width(struct sdhci_host *host, int width)
 {
 	u32 ctrl;
 
@@ -686,8 +683,6 @@ static int esdhc_pltfm_bus_width(struct sdhci_host *host, int width)
 
 	esdhc_clrset_le(host, ESDHC_CTRL_BUSWIDTH_MASK, ctrl,
 			SDHCI_HOST_CONTROL);
-
-	return 0;
 }
 
 static void esdhc_prepare_tuning(struct sdhci_host *host, u32 val)
@@ -697,6 +692,7 @@ static void esdhc_prepare_tuning(struct sdhci_host *host, u32 val)
 	/* FIXME: delay a bit for card to be ready for next tuning due to errors */
 	mdelay(1);
 
+	/* This is balanced by the runtime put in sdhci_tasklet_finish */
 	pm_runtime_get_sync(host->mmc->parent);
 	reg = readl(host->ioaddr + ESDHC_MIX_CTRL);
 	reg |= ESDHC_MIX_CTRL_EXE_TUNE | ESDHC_MIX_CTRL_SMPCLK_SEL |
@@ -713,13 +709,12 @@ static void esdhc_request_done(struct mmc_request *mrq)
 	complete(&mrq->completion);
 }
 
-static int esdhc_send_tuning_cmd(struct sdhci_host *host, u32 opcode)
+static int esdhc_send_tuning_cmd(struct sdhci_host *host, u32 opcode,
+				 struct scatterlist *sg)
 {
 	struct mmc_command cmd = {0};
 	struct mmc_request mrq = {NULL};
 	struct mmc_data data = {0};
-	struct scatterlist sg;
-	char tuning_pattern[ESDHC_TUNING_BLOCK_PATTERN_LEN];
 
 	cmd.opcode = opcode;
 	cmd.arg = 0;
@@ -728,11 +723,9 @@ static int esdhc_send_tuning_cmd(struct sdhci_host *host, u32 opcode)
 	data.blksz = ESDHC_TUNING_BLOCK_PATTERN_LEN;
 	data.blocks = 1;
 	data.flags = MMC_DATA_READ;
-	data.sg = &sg;
+	data.sg = sg;
 	data.sg_len = 1;
 
-	sg_init_one(&sg, tuning_pattern, sizeof(tuning_pattern));
-
 	mrq.cmd = &cmd;
 	mrq.cmd->mrq = &mrq;
 	mrq.data = &data;
@@ -742,14 +735,12 @@ static int esdhc_send_tuning_cmd(struct sdhci_host *host, u32 opcode)
 	mrq.done = esdhc_request_done;
 	init_completion(&(mrq.completion));
 
-	disable_irq(host->irq);
-	spin_lock(&host->lock);
+	spin_lock_irq(&host->lock);
 	host->mrq = &mrq;
 
 	sdhci_send_command(host, mrq.cmd);
 
-	spin_unlock(&host->lock);
-	enable_irq(host->irq);
+	spin_unlock_irq(&host->lock);
 
 	wait_for_completion(&mrq.completion);
 
@@ -772,13 +763,21 @@ static void esdhc_post_tuning(struct sdhci_host *host)
 
 static int esdhc_executing_tuning(struct sdhci_host *host, u32 opcode)
 {
+	struct scatterlist sg;
+	char *tuning_pattern;
 	int min, max, avg, ret;
 
+	tuning_pattern = kmalloc(ESDHC_TUNING_BLOCK_PATTERN_LEN, GFP_KERNEL);
+	if (!tuning_pattern)
+		return -ENOMEM;
+
+	sg_init_one(&sg, tuning_pattern, ESDHC_TUNING_BLOCK_PATTERN_LEN);
+
 	/* find the mininum delay first which can pass tuning */
 	min = ESDHC_TUNE_CTRL_MIN;
 	while (min < ESDHC_TUNE_CTRL_MAX) {
 		esdhc_prepare_tuning(host, min);
-		if (!esdhc_send_tuning_cmd(host, opcode))
+		if (!esdhc_send_tuning_cmd(host, opcode, &sg))
 			break;
 		min += ESDHC_TUNE_CTRL_STEP;
 	}
@@ -787,7 +786,7 @@ static int esdhc_executing_tuning(struct sdhci_host *host, u32 opcode)
 	max = min + ESDHC_TUNE_CTRL_STEP;
 	while (max < ESDHC_TUNE_CTRL_MAX) {
 		esdhc_prepare_tuning(host, max);
-		if (esdhc_send_tuning_cmd(host, opcode)) {
+		if (esdhc_send_tuning_cmd(host, opcode, &sg)) {
 			max -= ESDHC_TUNE_CTRL_STEP;
 			break;
 		}
@@ -797,9 +796,11 @@ static int esdhc_executing_tuning(struct sdhci_host *host, u32 opcode)
 	/* use average delay to get the best timing */
 	avg = (min + max) / 2;
 	esdhc_prepare_tuning(host, avg);
-	ret = esdhc_send_tuning_cmd(host, opcode);
+	ret = esdhc_send_tuning_cmd(host, opcode, &sg);
 	esdhc_post_tuning(host);
 
+	kfree(tuning_pattern);
+
 	dev_dbg(mmc_dev(host->mmc), "tunning %s at 0x%x ret %d\n",
 		ret ? "failed" : "passed", avg, ret);
 
@@ -837,28 +838,21 @@ static int esdhc_change_pinstate(struct sdhci_host *host,
 	return pinctrl_select_state(imx_data->pinctrl, pinctrl);
 }
 
-static int esdhc_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
+static void esdhc_set_uhs_signaling(struct sdhci_host *host, unsigned timing)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct pltfm_imx_data *imx_data = pltfm_host->priv;
 	struct esdhc_platform_data *boarddata = &imx_data->boarddata;
 
-	switch (uhs) {
+	switch (timing) {
 	case MMC_TIMING_UHS_SDR12:
-		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR12;
-		break;
 	case MMC_TIMING_UHS_SDR25:
-		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR25;
-		break;
 	case MMC_TIMING_UHS_SDR50:
-		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR50;
-		break;
 	case MMC_TIMING_UHS_SDR104:
 	case MMC_TIMING_MMC_HS200:
-		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR104;
 		break;
 	case MMC_TIMING_UHS_DDR50:
-		imx_data->uhs_mode = SDHCI_CTRL_UHS_DDR50;
+	case MMC_TIMING_MMC_DDR52:
 		writel(readl(host->ioaddr + ESDHC_MIX_CTRL) |
 				ESDHC_MIX_CTRL_DDREN,
 				host->ioaddr + ESDHC_MIX_CTRL);
@@ -875,7 +869,15 @@ static int esdhc_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
 		break;
 	}
 
-	return esdhc_change_pinstate(host, uhs);
+	esdhc_change_pinstate(host, timing);
+}
+
+static void esdhc_reset(struct sdhci_host *host, u8 mask)
+{
+	sdhci_reset(host, mask);
+
+	sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+	sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
 }
 
 static struct sdhci_ops sdhci_esdhc_ops = {
@@ -888,8 +890,9 @@ static struct sdhci_ops sdhci_esdhc_ops = {
 	.get_max_clock = esdhc_pltfm_get_max_clock,
 	.get_min_clock = esdhc_pltfm_get_min_clock,
 	.get_ro = esdhc_pltfm_get_ro,
-	.platform_bus_width = esdhc_pltfm_bus_width,
+	.set_bus_width = esdhc_pltfm_set_bus_width,
 	.set_uhs_signaling = esdhc_set_uhs_signaling,
+	.reset = esdhc_reset,
 };
 
 static const struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
@@ -1170,8 +1173,10 @@ static int sdhci_esdhc_runtime_suspend(struct device *dev)
 
 	ret = sdhci_runtime_suspend_host(host);
 
-	clk_disable_unprepare(imx_data->clk_per);
-	clk_disable_unprepare(imx_data->clk_ipg);
+	if (!sdhci_sdio_irq_enabled(host)) {
+		clk_disable_unprepare(imx_data->clk_per);
+		clk_disable_unprepare(imx_data->clk_ipg);
+	}
 	clk_disable_unprepare(imx_data->clk_ahb);
 
 	return ret;
@@ -1183,8 +1188,10 @@ static int sdhci_esdhc_runtime_resume(struct device *dev)
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct pltfm_imx_data *imx_data = pltfm_host->priv;
 
-	clk_prepare_enable(imx_data->clk_per);
-	clk_prepare_enable(imx_data->clk_ipg);
+	if (!sdhci_sdio_irq_enabled(host)) {
+		clk_prepare_enable(imx_data->clk_per);
+		clk_prepare_enable(imx_data->clk_ipg);
+	}
 	clk_prepare_enable(imx_data->clk_ahb);
 
 	return sdhci_runtime_resume_host(host);
diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h
index a7d9f95a7b03..3497cfaf683c 100644
--- a/drivers/mmc/host/sdhci-esdhc.h
+++ b/drivers/mmc/host/sdhci-esdhc.h
@@ -20,10 +20,8 @@
 
 #define ESDHC_DEFAULT_QUIRKS	(SDHCI_QUIRK_FORCE_BLK_SZ_2048 | \
 				SDHCI_QUIRK_NO_BUSY_IRQ | \
-				SDHCI_QUIRK_NONSTANDARD_CLOCK | \
 				SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK | \
-				SDHCI_QUIRK_PIO_NEEDS_DELAY | \
-				SDHCI_QUIRK_RESTORE_IRQS_AFTER_RESET)
+				SDHCI_QUIRK_PIO_NEEDS_DELAY)
 
 #define ESDHC_SYSTEM_CONTROL	0x2c
 #define ESDHC_CLOCK_MASK	0x0000fff0
diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c
index acb0e9eb55f1..40573a58486a 100644
--- a/drivers/mmc/host/sdhci-msm.c
+++ b/drivers/mmc/host/sdhci-msm.c
@@ -468,6 +468,10 @@ MODULE_DEVICE_TABLE(of, sdhci_msm_dt_match);
 
 static struct sdhci_ops sdhci_msm_ops = {
 	.platform_execute_tuning = sdhci_msm_execute_tuning,
+	.reset = sdhci_reset,
+	.set_clock = sdhci_set_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static int sdhci_msm_probe(struct platform_device *pdev)
diff --git a/drivers/mmc/host/sdhci-of-arasan.c b/drivers/mmc/host/sdhci-of-arasan.c
index f7c7cf62437d..5bd1092310f2 100644
--- a/drivers/mmc/host/sdhci-of-arasan.c
+++ b/drivers/mmc/host/sdhci-of-arasan.c
@@ -52,8 +52,12 @@ static unsigned int sdhci_arasan_get_timeout_clock(struct sdhci_host *host)
 }
 
 static struct sdhci_ops sdhci_arasan_ops = {
+	.set_clock = sdhci_set_clock,
 	.get_max_clock = sdhci_pltfm_clk_get_max_clock,
 	.get_timeout_clock = sdhci_arasan_get_timeout_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static struct sdhci_pltfm_data sdhci_arasan_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c
index 0b249970b119..8be4dcfb49a0 100644
--- a/drivers/mmc/host/sdhci-of-esdhc.c
+++ b/drivers/mmc/host/sdhci-of-esdhc.c
@@ -199,13 +199,14 @@ static unsigned int esdhc_of_get_min_clock(struct sdhci_host *host)
 
 static void esdhc_of_set_clock(struct sdhci_host *host, unsigned int clock)
 {
-
 	int pre_div = 2;
 	int div = 1;
 	u32 temp;
 
+	host->mmc->actual_clock = 0;
+
 	if (clock == 0)
-		goto out;
+		return;
 
 	/* Workaround to reduce the clock frequency for p1010 esdhc */
 	if (of_find_compatible_node(NULL, NULL, "fsl,p1010-esdhc")) {
@@ -238,24 +239,8 @@ static void esdhc_of_set_clock(struct sdhci_host *host, unsigned int clock)
 		| (pre_div << ESDHC_PREDIV_SHIFT));
 	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
 	mdelay(1);
-out:
-	host->clock = clock;
-}
-
-#ifdef CONFIG_PM
-static u32 esdhc_proctl;
-static void esdhc_of_suspend(struct sdhci_host *host)
-{
-	esdhc_proctl = sdhci_be32bs_readl(host, SDHCI_HOST_CONTROL);
 }
 
-static void esdhc_of_resume(struct sdhci_host *host)
-{
-	esdhc_of_enable_dma(host);
-	sdhci_be32bs_writel(host, esdhc_proctl, SDHCI_HOST_CONTROL);
-}
-#endif
-
 static void esdhc_of_platform_init(struct sdhci_host *host)
 {
 	u32 vvn;
@@ -269,7 +254,7 @@ static void esdhc_of_platform_init(struct sdhci_host *host)
 		host->quirks &= ~SDHCI_QUIRK_NO_BUSY_IRQ;
 }
 
-static int esdhc_pltfm_bus_width(struct sdhci_host *host, int width)
+static void esdhc_pltfm_set_bus_width(struct sdhci_host *host, int width)
 {
 	u32 ctrl;
 
@@ -289,8 +274,6 @@ static int esdhc_pltfm_bus_width(struct sdhci_host *host, int width)
 
 	clrsetbits_be32(host->ioaddr + SDHCI_HOST_CONTROL,
 			ESDHC_CTRL_BUSWIDTH_MASK, ctrl);
-
-	return 0;
 }
 
 static const struct sdhci_ops sdhci_esdhc_ops = {
@@ -305,13 +288,46 @@ static const struct sdhci_ops sdhci_esdhc_ops = {
 	.get_max_clock = esdhc_of_get_max_clock,
 	.get_min_clock = esdhc_of_get_min_clock,
 	.platform_init = esdhc_of_platform_init,
-#ifdef CONFIG_PM
-	.platform_suspend = esdhc_of_suspend,
-	.platform_resume = esdhc_of_resume,
-#endif
 	.adma_workaround = esdhci_of_adma_workaround,
-	.platform_bus_width = esdhc_pltfm_bus_width,
+	.set_bus_width = esdhc_pltfm_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
+};
+
+#ifdef CONFIG_PM
+
+static u32 esdhc_proctl;
+static int esdhc_of_suspend(struct device *dev)
+{
+	struct sdhci_host *host = dev_get_drvdata(dev);
+
+	esdhc_proctl = sdhci_be32bs_readl(host, SDHCI_HOST_CONTROL);
+
+	return sdhci_suspend_host(host);
+}
+
+static int esdhc_of_resume(struct device *dev)
+{
+	struct sdhci_host *host = dev_get_drvdata(dev);
+	int ret = sdhci_resume_host(host);
+
+	if (ret == 0) {
+		/* Isn't this already done by sdhci_resume_host() ? --rmk */
+		esdhc_of_enable_dma(host);
+		sdhci_be32bs_writel(host, esdhc_proctl, SDHCI_HOST_CONTROL);
+	}
+
+	return ret;
+}
+
+static const struct dev_pm_ops esdhc_pmops = {
+	.suspend	= esdhc_of_suspend,
+	.resume		= esdhc_of_resume,
 };
+#define ESDHC_PMOPS (&esdhc_pmops)
+#else
+#define ESDHC_PMOPS NULL
+#endif
 
 static const struct sdhci_pltfm_data sdhci_esdhc_pdata = {
 	/*
@@ -374,7 +390,7 @@ static struct platform_driver sdhci_esdhc_driver = {
 		.name = "sdhci-esdhc",
 		.owner = THIS_MODULE,
 		.of_match_table = sdhci_esdhc_of_match,
-		.pm = SDHCI_PLTFM_PMOPS,
+		.pm = ESDHC_PMOPS,
 	},
 	.probe = sdhci_esdhc_probe,
 	.remove = sdhci_esdhc_remove,
diff --git a/drivers/mmc/host/sdhci-of-hlwd.c b/drivers/mmc/host/sdhci-of-hlwd.c
index 57c514a81ca5..b341661369a2 100644
--- a/drivers/mmc/host/sdhci-of-hlwd.c
+++ b/drivers/mmc/host/sdhci-of-hlwd.c
@@ -58,6 +58,10 @@ static const struct sdhci_ops sdhci_hlwd_ops = {
 	.write_l = sdhci_hlwd_writel,
 	.write_w = sdhci_hlwd_writew,
 	.write_b = sdhci_hlwd_writeb,
+	.set_clock = sdhci_set_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static const struct sdhci_pltfm_data sdhci_hlwd_pdata = {
diff --git a/drivers/mmc/host/sdhci-pci-o2micro.c b/drivers/mmc/host/sdhci-pci-o2micro.c
index f49666bcc52a..5670e381b0cf 100644
--- a/drivers/mmc/host/sdhci-pci-o2micro.c
+++ b/drivers/mmc/host/sdhci-pci-o2micro.c
@@ -21,6 +21,45 @@
 #include "sdhci-pci.h"
 #include "sdhci-pci-o2micro.h"
 
+static void o2_pci_set_baseclk(struct sdhci_pci_chip *chip, u32 value)
+{
+	u32 scratch_32;
+	pci_read_config_dword(chip->pdev,
+			      O2_SD_PLL_SETTING, &scratch_32);
+
+	scratch_32 &= 0x0000FFFF;
+	scratch_32 |= value;
+
+	pci_write_config_dword(chip->pdev,
+			       O2_SD_PLL_SETTING, scratch_32);
+}
+
+static void o2_pci_led_enable(struct sdhci_pci_chip *chip)
+{
+	int ret;
+	u32 scratch_32;
+
+	/* Set led of SD host function enable */
+	ret = pci_read_config_dword(chip->pdev,
+				    O2_SD_FUNC_REG0, &scratch_32);
+	if (ret)
+		return;
+
+	scratch_32 &= ~O2_SD_FREG0_LEDOFF;
+	pci_write_config_dword(chip->pdev,
+			       O2_SD_FUNC_REG0, scratch_32);
+
+	ret = pci_read_config_dword(chip->pdev,
+				    O2_SD_TEST_REG, &scratch_32);
+	if (ret)
+		return;
+
+	scratch_32 |= O2_SD_LED_ENABLE;
+	pci_write_config_dword(chip->pdev,
+			       O2_SD_TEST_REG, scratch_32);
+
+}
+
 void sdhci_pci_o2_fujin2_pci_init(struct sdhci_pci_chip *chip)
 {
 	u32 scratch_32;
@@ -216,6 +255,40 @@ int sdhci_pci_o2_probe(struct sdhci_pci_chip *chip)
 		scratch &= 0x7f;
 		pci_write_config_byte(chip->pdev, O2_SD_LOCK_WP, scratch);
 
+		/* DevId=8520 subId= 0x11 or 0x12  Type Chip support */
+		if (chip->pdev->device == PCI_DEVICE_ID_O2_FUJIN2) {
+			ret = pci_read_config_dword(chip->pdev,
+						    O2_SD_FUNC_REG0,
+						    &scratch_32);
+			scratch_32 = ((scratch_32 & 0xFF000000) >> 24);
+
+			/* Check Whether subId is 0x11 or 0x12 */
+			if ((scratch_32 == 0x11) || (scratch_32 == 0x12)) {
+				scratch_32 = 0x2c280000;
+
+				/* Set Base Clock to 208MZ */
+				o2_pci_set_baseclk(chip, scratch_32);
+				ret = pci_read_config_dword(chip->pdev,
+							    O2_SD_FUNC_REG4,
+							    &scratch_32);
+
+				/* Enable Base Clk setting change */
+				scratch_32 |= O2_SD_FREG4_ENABLE_CLK_SET;
+				pci_write_config_dword(chip->pdev,
+						       O2_SD_FUNC_REG4,
+						       scratch_32);
+
+				/* Set Tuning Window to 4 */
+				pci_write_config_byte(chip->pdev,
+						      O2_SD_TUNING_CTRL, 0x44);
+
+				break;
+			}
+		}
+
+		/* Enable 8520 led function */
+		o2_pci_led_enable(chip);
+
 		/* Set timeout CLK */
 		ret = pci_read_config_dword(chip->pdev,
 					    O2_SD_CLK_SETTING, &scratch_32);
@@ -276,7 +349,7 @@ int sdhci_pci_o2_probe(struct sdhci_pci_chip *chip)
 		pci_write_config_byte(chip->pdev, O2_SD_LOCK_WP, scratch);
 
 		ret = pci_read_config_dword(chip->pdev,
-					    O2_SD_FUNC_REG0, &scratch_32);
+					    O2_SD_PLL_SETTING, &scratch_32);
 
 		if ((scratch_32 & 0xff000000) == 0x01000000) {
 			scratch_32 &= 0x0000FFFF;
@@ -299,6 +372,9 @@ int sdhci_pci_o2_probe(struct sdhci_pci_chip *chip)
 					       O2_SD_FUNC_REG4, scratch_32);
 		}
 
+		/* Set Tuning Windows to 5 */
+		pci_write_config_byte(chip->pdev,
+				O2_SD_TUNING_CTRL, 0x55);
 		/* Lock WP */
 		ret = pci_read_config_byte(chip->pdev,
 					   O2_SD_LOCK_WP, &scratch);
diff --git a/drivers/mmc/host/sdhci-pci-o2micro.h b/drivers/mmc/host/sdhci-pci-o2micro.h
index dbec4c933488..f7ffc908d9a0 100644
--- a/drivers/mmc/host/sdhci-pci-o2micro.h
+++ b/drivers/mmc/host/sdhci-pci-o2micro.h
@@ -57,6 +57,9 @@
 #define O2_SD_UHS2_L1_CTRL	0x35C
 #define O2_SD_FUNC_REG3		0x3E0
 #define O2_SD_FUNC_REG4		0x3E4
+#define O2_SD_LED_ENABLE	BIT(6)
+#define O2_SD_FREG0_LEDOFF	BIT(13)
+#define O2_SD_FREG4_ENABLE_CLK_SET	BIT(22)
 
 #define O2_SD_VENDOR_SETTING	0x110
 #define O2_SD_VENDOR_SETTING2	0x1C8
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index fdc612120362..52c42fcc284c 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -1031,7 +1031,7 @@ static int sdhci_pci_enable_dma(struct sdhci_host *host)
 	return 0;
 }
 
-static int sdhci_pci_bus_width(struct sdhci_host *host, int width)
+static void sdhci_pci_set_bus_width(struct sdhci_host *host, int width)
 {
 	u8 ctrl;
 
@@ -1052,8 +1052,6 @@ static int sdhci_pci_bus_width(struct sdhci_host *host, int width)
 	}
 
 	sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
-
-	return 0;
 }
 
 static void sdhci_pci_gpio_hw_reset(struct sdhci_host *host)
@@ -1080,8 +1078,11 @@ static void sdhci_pci_hw_reset(struct sdhci_host *host)
 }
 
 static const struct sdhci_ops sdhci_pci_ops = {
+	.set_clock	= sdhci_set_clock,
 	.enable_dma	= sdhci_pci_enable_dma,
-	.platform_bus_width	= sdhci_pci_bus_width,
+	.set_bus_width	= sdhci_pci_set_bus_width,
+	.reset		= sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 	.hw_reset		= sdhci_pci_hw_reset,
 };
 
diff --git a/drivers/mmc/host/sdhci-pltfm.c b/drivers/mmc/host/sdhci-pltfm.c
index bef250e95418..7e834fb78f42 100644
--- a/drivers/mmc/host/sdhci-pltfm.c
+++ b/drivers/mmc/host/sdhci-pltfm.c
@@ -45,6 +45,10 @@ unsigned int sdhci_pltfm_clk_get_max_clock(struct sdhci_host *host)
 EXPORT_SYMBOL_GPL(sdhci_pltfm_clk_get_max_clock);
 
 static const struct sdhci_ops sdhci_pltfm_ops = {
+	.set_clock = sdhci_set_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 #ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-pxav2.c b/drivers/mmc/host/sdhci-pxav2.c
index d51e061ec576..3c0f3c0a1cc8 100644
--- a/drivers/mmc/host/sdhci-pxav2.c
+++ b/drivers/mmc/host/sdhci-pxav2.c
@@ -51,11 +51,13 @@
 #define MMC_CARD		0x1000
 #define MMC_WIDTH		0x0100
 
-static void pxav2_set_private_registers(struct sdhci_host *host, u8 mask)
+static void pxav2_reset(struct sdhci_host *host, u8 mask)
 {
 	struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
 	struct sdhci_pxa_platdata *pdata = pdev->dev.platform_data;
 
+	sdhci_reset(host, mask);
+
 	if (mask == SDHCI_RESET_ALL) {
 		u16 tmp = 0;
 
@@ -88,7 +90,7 @@ static void pxav2_set_private_registers(struct sdhci_host *host, u8 mask)
 	}
 }
 
-static int pxav2_mmc_set_width(struct sdhci_host *host, int width)
+static void pxav2_mmc_set_bus_width(struct sdhci_host *host, int width)
 {
 	u8 ctrl;
 	u16 tmp;
@@ -107,14 +109,14 @@ static int pxav2_mmc_set_width(struct sdhci_host *host, int width)
 	}
 	writew(tmp, host->ioaddr + SD_CE_ATA_2);
 	writeb(ctrl, host->ioaddr + SDHCI_HOST_CONTROL);
-
-	return 0;
 }
 
 static const struct sdhci_ops pxav2_sdhci_ops = {
+	.set_clock     = sdhci_set_clock,
 	.get_max_clock = sdhci_pltfm_clk_get_max_clock,
-	.platform_reset_exit = pxav2_set_private_registers,
-	.platform_bus_width = pxav2_mmc_set_width,
+	.set_bus_width = pxav2_mmc_set_bus_width,
+	.reset         = pxav2_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 #ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
index 2fd73b38c303..f4f128947561 100644
--- a/drivers/mmc/host/sdhci-pxav3.c
+++ b/drivers/mmc/host/sdhci-pxav3.c
@@ -112,11 +112,13 @@ static int mv_conf_mbus_windows(struct platform_device *pdev,
 	return 0;
 }
 
-static void pxav3_set_private_registers(struct sdhci_host *host, u8 mask)
+static void pxav3_reset(struct sdhci_host *host, u8 mask)
 {
 	struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
 	struct sdhci_pxa_platdata *pdata = pdev->dev.platform_data;
 
+	sdhci_reset(host, mask);
+
 	if (mask == SDHCI_RESET_ALL) {
 		/*
 		 * tune timing of read data/command when crc error happen
@@ -184,7 +186,7 @@ static void pxav3_gen_init_74_clocks(struct sdhci_host *host, u8 power_mode)
 	pxa->power_mode = power_mode;
 }
 
-static int pxav3_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
+static void pxav3_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
 {
 	u16 ctrl_2;
 
@@ -218,15 +220,16 @@ static int pxav3_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
 	dev_dbg(mmc_dev(host->mmc),
 		"%s uhs = %d, ctrl_2 = %04X\n",
 		__func__, uhs, ctrl_2);
-
-	return 0;
 }
 
 static const struct sdhci_ops pxav3_sdhci_ops = {
-	.platform_reset_exit = pxav3_set_private_registers,
+	.set_clock = sdhci_set_clock,
 	.set_uhs_signaling = pxav3_set_uhs_signaling,
 	.platform_send_init_74_clocks = pxav3_gen_init_74_clocks,
 	.get_max_clock = sdhci_pltfm_clk_get_max_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = pxav3_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static struct sdhci_pltfm_data sdhci_pxav3_pdata = {
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index d61eb5a70833..fa5954a05449 100644
--- a/drivers/mmc/host/sdhci-s3c.c
+++ b/drivers/mmc/host/sdhci-s3c.c
@@ -33,9 +33,6 @@
 
 #define MAX_BUS_CLK	(4)
 
-/* Number of gpio's used is max data bus width + command and clock lines */
-#define NUM_GPIOS(x)	(x + 2)
-
 /**
  * struct sdhci_s3c - S3C SDHCI instance
  * @host: The SDHCI host created
@@ -58,6 +55,8 @@ struct sdhci_s3c {
 	struct clk		*clk_io;
 	struct clk		*clk_bus[MAX_BUS_CLK];
 	unsigned long		clk_rates[MAX_BUS_CLK];
+
+	bool			no_divider;
 };
 
 /**
@@ -70,6 +69,7 @@ struct sdhci_s3c {
  */
 struct sdhci_s3c_drv_data {
 	unsigned int	sdhci_quirks;
+	bool		no_divider;
 };
 
 static inline struct sdhci_s3c *to_s3c(struct sdhci_host *host)
@@ -119,7 +119,7 @@ static unsigned int sdhci_s3c_consider_clock(struct sdhci_s3c *ourhost,
 	 * If controller uses a non-standard clock division, find the best clock
 	 * speed possible with selected clock source and skip the division.
 	 */
-	if (ourhost->host->quirks & SDHCI_QUIRK_NONSTANDARD_CLOCK) {
+	if (ourhost->no_divider) {
 		rate = clk_round_rate(clksrc, wanted);
 		return wanted - rate;
 	}
@@ -161,9 +161,13 @@ static void sdhci_s3c_set_clock(struct sdhci_host *host, unsigned int clock)
 	int src;
 	u32 ctrl;
 
+	host->mmc->actual_clock = 0;
+
 	/* don't bother if the clock is going off. */
-	if (clock == 0)
+	if (clock == 0) {
+		sdhci_set_clock(host, clock);
 		return;
+	}
 
 	for (src = 0; src < MAX_BUS_CLK; src++) {
 		delta = sdhci_s3c_consider_clock(ourhost, src, clock);
@@ -215,6 +219,8 @@ static void sdhci_s3c_set_clock(struct sdhci_host *host, unsigned int clock)
 	if (clock < 25 * 1000000)
 		ctrl |= (S3C_SDHCI_CTRL3_FCSEL3 | S3C_SDHCI_CTRL3_FCSEL2);
 	writel(ctrl, host->ioaddr + S3C_SDHCI_CONTROL3);
+
+	sdhci_set_clock(host, clock);
 }
 
 /**
@@ -295,10 +301,11 @@ static void sdhci_cmu_set_clock(struct sdhci_host *host, unsigned int clock)
 	unsigned long timeout;
 	u16 clk = 0;
 
+	host->mmc->actual_clock = 0;
+
 	/* If the clock is going off, set to 0 at clock control register */
 	if (clock == 0) {
 		sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL);
-		host->clock = clock;
 		return;
 	}
 
@@ -306,8 +313,6 @@ static void sdhci_cmu_set_clock(struct sdhci_host *host, unsigned int clock)
 
 	clk_set_rate(ourhost->clk_bus[ourhost->cur_clk], clock);
 
-	host->clock = clock;
-
 	clk = SDHCI_CLOCK_INT_EN;
 	sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
 
@@ -329,14 +334,14 @@ static void sdhci_cmu_set_clock(struct sdhci_host *host, unsigned int clock)
 }
 
 /**
- * sdhci_s3c_platform_bus_width - support 8bit buswidth
+ * sdhci_s3c_set_bus_width - support 8bit buswidth
  * @host: The SDHCI host being queried
  * @width: MMC_BUS_WIDTH_ macro for the bus width being requested
  *
  * We have 8-bit width support but is not a v3 controller.
  * So we add platform_bus_width() and support 8bit width.
  */
-static int sdhci_s3c_platform_bus_width(struct sdhci_host *host, int width)
+static void sdhci_s3c_set_bus_width(struct sdhci_host *host, int width)
 {
 	u8 ctrl;
 
@@ -358,93 +363,23 @@ static int sdhci_s3c_platform_bus_width(struct sdhci_host *host, int width)
 	}
 
 	sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
-
-	return 0;
 }
 
 static struct sdhci_ops sdhci_s3c_ops = {
 	.get_max_clock		= sdhci_s3c_get_max_clk,
 	.set_clock		= sdhci_s3c_set_clock,
 	.get_min_clock		= sdhci_s3c_get_min_clock,
-	.platform_bus_width	= sdhci_s3c_platform_bus_width,
+	.set_bus_width		= sdhci_s3c_set_bus_width,
+	.reset			= sdhci_reset,
+	.set_uhs_signaling	= sdhci_set_uhs_signaling,
 };
 
-static void sdhci_s3c_notify_change(struct platform_device *dev, int state)
-{
-	struct sdhci_host *host = platform_get_drvdata(dev);
-#ifdef CONFIG_PM_RUNTIME
-	struct sdhci_s3c *sc = sdhci_priv(host);
-#endif
-	unsigned long flags;
-
-	if (host) {
-		spin_lock_irqsave(&host->lock, flags);
-		if (state) {
-			dev_dbg(&dev->dev, "card inserted.\n");
-#ifdef CONFIG_PM_RUNTIME
-			clk_prepare_enable(sc->clk_io);
-#endif
-			host->flags &= ~SDHCI_DEVICE_DEAD;
-			host->quirks |= SDHCI_QUIRK_BROKEN_CARD_DETECTION;
-		} else {
-			dev_dbg(&dev->dev, "card removed.\n");
-			host->flags |= SDHCI_DEVICE_DEAD;
-			host->quirks &= ~SDHCI_QUIRK_BROKEN_CARD_DETECTION;
-#ifdef CONFIG_PM_RUNTIME
-			clk_disable_unprepare(sc->clk_io);
-#endif
-		}
-		tasklet_schedule(&host->card_tasklet);
-		spin_unlock_irqrestore(&host->lock, flags);
-	}
-}
-
-static irqreturn_t sdhci_s3c_gpio_card_detect_thread(int irq, void *dev_id)
-{
-	struct sdhci_s3c *sc = dev_id;
-	int status = gpio_get_value(sc->ext_cd_gpio);
-	if (sc->pdata->ext_cd_gpio_invert)
-		status = !status;
-	sdhci_s3c_notify_change(sc->pdev, status);
-	return IRQ_HANDLED;
-}
-
-static void sdhci_s3c_setup_card_detect_gpio(struct sdhci_s3c *sc)
-{
-	struct s3c_sdhci_platdata *pdata = sc->pdata;
-	struct device *dev = &sc->pdev->dev;
-
-	if (devm_gpio_request(dev, pdata->ext_cd_gpio, "SDHCI EXT CD") == 0) {
-		sc->ext_cd_gpio = pdata->ext_cd_gpio;
-		sc->ext_cd_irq = gpio_to_irq(pdata->ext_cd_gpio);
-		if (sc->ext_cd_irq &&
-		    request_threaded_irq(sc->ext_cd_irq, NULL,
-					 sdhci_s3c_gpio_card_detect_thread,
-					 IRQF_TRIGGER_RISING |
-					 IRQF_TRIGGER_FALLING |
-					 IRQF_ONESHOT,
-					 dev_name(dev), sc) == 0) {
-			int status = gpio_get_value(sc->ext_cd_gpio);
-			if (pdata->ext_cd_gpio_invert)
-				status = !status;
-			sdhci_s3c_notify_change(sc->pdev, status);
-		} else {
-			dev_warn(dev, "cannot request irq for card detect\n");
-			sc->ext_cd_irq = 0;
-		}
-	} else {
-		dev_err(dev, "cannot request gpio for card detect\n");
-	}
-}
-
 #ifdef CONFIG_OF
 static int sdhci_s3c_parse_dt(struct device *dev,
 		struct sdhci_host *host, struct s3c_sdhci_platdata *pdata)
 {
 	struct device_node *node = dev->of_node;
-	struct sdhci_s3c *ourhost = to_s3c(host);
 	u32 max_width;
-	int gpio;
 
 	/* if the bus-width property is not specified, assume width as 1 */
 	if (of_property_read_u32(node, "bus-width", &max_width))
@@ -462,18 +397,8 @@ static int sdhci_s3c_parse_dt(struct device *dev,
 		return 0;
 	}
 
-	gpio = of_get_named_gpio(node, "cd-gpios", 0);
-	if (gpio_is_valid(gpio)) {
-		pdata->cd_type = S3C_SDHCI_CD_GPIO;
-		pdata->ext_cd_gpio = gpio;
-		ourhost->ext_cd_gpio = -1;
-		if (of_get_property(node, "cd-inverted", NULL))
-			pdata->ext_cd_gpio_invert = 1;
+	if (of_get_named_gpio(node, "cd-gpios", 0))
 		return 0;
-	} else if (gpio != -ENOENT) {
-		dev_err(dev, "invalid card detect gpio specified\n");
-		return -EINVAL;
-	}
 
 	/* assuming internal card detect that will be configured by pinctrl */
 	pdata->cd_type = S3C_SDHCI_CD_INTERNAL;
@@ -606,8 +531,10 @@ static int sdhci_s3c_probe(struct platform_device *pdev)
 	/* Setup quirks for the controller */
 	host->quirks |= SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC;
 	host->quirks |= SDHCI_QUIRK_NO_HISPD_BIT;
-	if (drv_data)
+	if (drv_data) {
 		host->quirks |= drv_data->sdhci_quirks;
+		sc->no_divider = drv_data->no_divider;
+	}
 
 #ifndef CONFIG_MMC_SDHCI_S3C_DMA
 
@@ -656,7 +583,7 @@ static int sdhci_s3c_probe(struct platform_device *pdev)
 	 * If controller does not have internal clock divider,
 	 * we can use overriding functions instead of default.
 	 */
-	if (host->quirks & SDHCI_QUIRK_NONSTANDARD_CLOCK) {
+	if (sc->no_divider) {
 		sdhci_s3c_ops.set_clock = sdhci_cmu_set_clock;
 		sdhci_s3c_ops.get_min_clock = sdhci_cmu_get_min_clock;
 		sdhci_s3c_ops.get_max_clock = sdhci_cmu_get_max_clock;
@@ -674,6 +601,8 @@ static int sdhci_s3c_probe(struct platform_device *pdev)
 	pm_runtime_use_autosuspend(&pdev->dev);
 	pm_suspend_ignore_children(&pdev->dev, 1);
 
+	mmc_of_parse(host->mmc);
+
 	ret = sdhci_add_host(host);
 	if (ret) {
 		dev_err(dev, "sdhci_add_host() failed\n");
@@ -682,15 +611,6 @@ static int sdhci_s3c_probe(struct platform_device *pdev)
 		goto err_req_regs;
 	}
 
-	/* The following two methods of card detection might call
-	   sdhci_s3c_notify_change() immediately, so they can be called
-	   only after sdhci_add_host(). Setup errors are ignored. */
-	if (pdata->cd_type == S3C_SDHCI_CD_EXTERNAL && pdata->ext_cd_init)
-		pdata->ext_cd_init(&sdhci_s3c_notify_change);
-	if (pdata->cd_type == S3C_SDHCI_CD_GPIO &&
-	    gpio_is_valid(pdata->ext_cd_gpio))
-		sdhci_s3c_setup_card_detect_gpio(sc);
-
 #ifdef CONFIG_PM_RUNTIME
 	if (pdata->cd_type != S3C_SDHCI_CD_INTERNAL)
 		clk_disable_unprepare(sc->clk_io);
@@ -711,16 +631,12 @@ static int sdhci_s3c_remove(struct platform_device *pdev)
 {
 	struct sdhci_host *host =  platform_get_drvdata(pdev);
 	struct sdhci_s3c *sc = sdhci_priv(host);
-	struct s3c_sdhci_platdata *pdata = sc->pdata;
-
-	if (pdata->cd_type == S3C_SDHCI_CD_EXTERNAL && pdata->ext_cd_cleanup)
-		pdata->ext_cd_cleanup(&sdhci_s3c_notify_change);
 
 	if (sc->ext_cd_irq)
 		free_irq(sc->ext_cd_irq, sc);
 
 #ifdef CONFIG_PM_RUNTIME
-	if (pdata->cd_type != S3C_SDHCI_CD_INTERNAL)
+	if (sc->pdata->cd_type != S3C_SDHCI_CD_INTERNAL)
 		clk_prepare_enable(sc->clk_io);
 #endif
 	sdhci_remove_host(host, 1);
@@ -797,7 +713,7 @@ static const struct dev_pm_ops sdhci_s3c_pmops = {
 
 #if defined(CONFIG_CPU_EXYNOS4210) || defined(CONFIG_SOC_EXYNOS4212)
 static struct sdhci_s3c_drv_data exynos4_sdhci_drv_data = {
-	.sdhci_quirks = SDHCI_QUIRK_NONSTANDARD_CLOCK,
+	.no_divider = true,
 };
 #define EXYNOS4_SDHCI_DRV_DATA ((kernel_ulong_t)&exynos4_sdhci_drv_data)
 #else
diff --git a/drivers/mmc/host/sdhci-sirf.c b/drivers/mmc/host/sdhci-sirf.c
index 696122c1b468..17004531d089 100644
--- a/drivers/mmc/host/sdhci-sirf.c
+++ b/drivers/mmc/host/sdhci-sirf.c
@@ -28,7 +28,11 @@ static unsigned int sdhci_sirf_get_max_clk(struct sdhci_host *host)
 }
 
 static struct sdhci_ops sdhci_sirf_ops = {
+	.set_clock = sdhci_set_clock,
 	.get_max_clock	= sdhci_sirf_get_max_clk,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 static struct sdhci_pltfm_data sdhci_sirf_pdata = {
diff --git a/drivers/mmc/host/sdhci-spear.c b/drivers/mmc/host/sdhci-spear.c
index 0316dec3f006..9d535c7336ef 100644
--- a/drivers/mmc/host/sdhci-spear.c
+++ b/drivers/mmc/host/sdhci-spear.c
@@ -38,7 +38,10 @@ struct spear_sdhci {
 
 /* sdhci ops */
 static const struct sdhci_ops sdhci_pltfm_ops = {
-	/* Nothing to do for now. */
+	.set_clock = sdhci_set_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
 };
 
 #ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index a835898a68dd..d93a063a36f3 100644
--- a/drivers/mmc/host/sdhci-tegra.c
+++ b/drivers/mmc/host/sdhci-tegra.c
@@ -32,11 +32,17 @@
 
 /* Tegra SDHOST controller vendor register definitions */
 #define SDHCI_TEGRA_VENDOR_MISC_CTRL		0x120
+#define SDHCI_MISC_CTRL_ENABLE_SDR104		0x8
+#define SDHCI_MISC_CTRL_ENABLE_SDR50		0x10
 #define SDHCI_MISC_CTRL_ENABLE_SDHCI_SPEC_300	0x20
+#define SDHCI_MISC_CTRL_ENABLE_DDR50		0x200
 
 #define NVQUIRK_FORCE_SDHCI_SPEC_200	BIT(0)
 #define NVQUIRK_ENABLE_BLOCK_GAP_DET	BIT(1)
 #define NVQUIRK_ENABLE_SDHCI_SPEC_300	BIT(2)
+#define NVQUIRK_DISABLE_SDR50		BIT(3)
+#define NVQUIRK_DISABLE_SDR104		BIT(4)
+#define NVQUIRK_DISABLE_DDR50		BIT(5)
 
 struct sdhci_tegra_soc_data {
 	const struct sdhci_pltfm_data *pdata;
@@ -48,19 +54,6 @@ struct sdhci_tegra {
 	int power_gpio;
 };
 
-static u32 tegra_sdhci_readl(struct sdhci_host *host, int reg)
-{
-	u32 val;
-
-	if (unlikely(reg == SDHCI_PRESENT_STATE)) {
-		/* Use wp_gpio here instead? */
-		val = readl(host->ioaddr + reg);
-		return val | SDHCI_WRITE_PROTECT;
-	}
-
-	return readl(host->ioaddr + reg);
-}
-
 static u16 tegra_sdhci_readw(struct sdhci_host *host, int reg)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
@@ -108,26 +101,33 @@ static unsigned int tegra_sdhci_get_ro(struct sdhci_host *host)
 	return mmc_gpio_get_ro(host->mmc);
 }
 
-static void tegra_sdhci_reset_exit(struct sdhci_host *host, u8 mask)
+static void tegra_sdhci_reset(struct sdhci_host *host, u8 mask)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct sdhci_tegra *tegra_host = pltfm_host->priv;
 	const struct sdhci_tegra_soc_data *soc_data = tegra_host->soc_data;
+	u32 misc_ctrl;
+
+	sdhci_reset(host, mask);
 
 	if (!(mask & SDHCI_RESET_ALL))
 		return;
 
+	misc_ctrl = sdhci_readw(host, SDHCI_TEGRA_VENDOR_MISC_CTRL);
 	/* Erratum: Enable SDHCI spec v3.00 support */
-	if (soc_data->nvquirks & NVQUIRK_ENABLE_SDHCI_SPEC_300) {
-		u32 misc_ctrl;
-
-		misc_ctrl = sdhci_readb(host, SDHCI_TEGRA_VENDOR_MISC_CTRL);
+	if (soc_data->nvquirks & NVQUIRK_ENABLE_SDHCI_SPEC_300)
 		misc_ctrl |= SDHCI_MISC_CTRL_ENABLE_SDHCI_SPEC_300;
-		sdhci_writeb(host, misc_ctrl, SDHCI_TEGRA_VENDOR_MISC_CTRL);
-	}
+	/* Don't advertise UHS modes which aren't supported yet */
+	if (soc_data->nvquirks & NVQUIRK_DISABLE_SDR50)
+		misc_ctrl &= ~SDHCI_MISC_CTRL_ENABLE_SDR50;
+	if (soc_data->nvquirks & NVQUIRK_DISABLE_DDR50)
+		misc_ctrl &= ~SDHCI_MISC_CTRL_ENABLE_DDR50;
+	if (soc_data->nvquirks & NVQUIRK_DISABLE_SDR104)
+		misc_ctrl &= ~SDHCI_MISC_CTRL_ENABLE_SDR104;
+	sdhci_writew(host, misc_ctrl, SDHCI_TEGRA_VENDOR_MISC_CTRL);
 }
 
-static int tegra_sdhci_buswidth(struct sdhci_host *host, int bus_width)
+static void tegra_sdhci_set_bus_width(struct sdhci_host *host, int bus_width)
 {
 	u32 ctrl;
 
@@ -144,23 +144,25 @@ static int tegra_sdhci_buswidth(struct sdhci_host *host, int bus_width)
 			ctrl &= ~SDHCI_CTRL_4BITBUS;
 	}
 	sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
-	return 0;
 }
 
 static const struct sdhci_ops tegra_sdhci_ops = {
 	.get_ro     = tegra_sdhci_get_ro,
-	.read_l     = tegra_sdhci_readl,
 	.read_w     = tegra_sdhci_readw,
 	.write_l    = tegra_sdhci_writel,
-	.platform_bus_width = tegra_sdhci_buswidth,
-	.platform_reset_exit = tegra_sdhci_reset_exit,
+	.set_clock  = sdhci_set_clock,
+	.set_bus_width = tegra_sdhci_set_bus_width,
+	.reset      = tegra_sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
+	.get_max_clock = sdhci_pltfm_clk_get_max_clock,
 };
 
 static const struct sdhci_pltfm_data sdhci_tegra20_pdata = {
 	.quirks = SDHCI_QUIRK_BROKEN_TIMEOUT_VAL |
 		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
 		  SDHCI_QUIRK_NO_HISPD_BIT |
-		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC,
+		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC |
+		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN,
 	.ops  = &tegra_sdhci_ops,
 };
 
@@ -175,13 +177,16 @@ static const struct sdhci_pltfm_data sdhci_tegra30_pdata = {
 		  SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK |
 		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
 		  SDHCI_QUIRK_NO_HISPD_BIT |
-		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC,
+		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC |
+		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN,
 	.ops  = &tegra_sdhci_ops,
 };
 
 static struct sdhci_tegra_soc_data soc_data_tegra30 = {
 	.pdata = &sdhci_tegra30_pdata,
-	.nvquirks = NVQUIRK_ENABLE_SDHCI_SPEC_300,
+	.nvquirks = NVQUIRK_ENABLE_SDHCI_SPEC_300 |
+		    NVQUIRK_DISABLE_SDR50 |
+		    NVQUIRK_DISABLE_SDR104,
 };
 
 static const struct sdhci_pltfm_data sdhci_tegra114_pdata = {
@@ -189,12 +194,16 @@ static const struct sdhci_pltfm_data sdhci_tegra114_pdata = {
 		  SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK |
 		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
 		  SDHCI_QUIRK_NO_HISPD_BIT |
-		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC,
+		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC |
+		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN,
 	.ops  = &tegra_sdhci_ops,
 };
 
 static struct sdhci_tegra_soc_data soc_data_tegra114 = {
 	.pdata = &sdhci_tegra114_pdata,
+	.nvquirks = NVQUIRK_DISABLE_SDR50 |
+		    NVQUIRK_DISABLE_DDR50 |
+		    NVQUIRK_DISABLE_SDR104,
 };
 
 static const struct of_device_id sdhci_tegra_dt_match[] = {
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 9a79fc4b60ca..47055f3f01b8 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -44,6 +44,8 @@
 
 #define MAX_TUNING_LOOP 40
 
+#define ADMA_SIZE	((128 * 2 + 1) * 4)
+
 static unsigned int debug_quirks = 0;
 static unsigned int debug_quirks2;
 
@@ -131,43 +133,26 @@ static void sdhci_dumpregs(struct sdhci_host *host)
  *                                                                           *
 \*****************************************************************************/
 
-static void sdhci_clear_set_irqs(struct sdhci_host *host, u32 clear, u32 set)
-{
-	u32 ier;
-
-	ier = sdhci_readl(host, SDHCI_INT_ENABLE);
-	ier &= ~clear;
-	ier |= set;
-	sdhci_writel(host, ier, SDHCI_INT_ENABLE);
-	sdhci_writel(host, ier, SDHCI_SIGNAL_ENABLE);
-}
-
-static void sdhci_unmask_irqs(struct sdhci_host *host, u32 irqs)
-{
-	sdhci_clear_set_irqs(host, 0, irqs);
-}
-
-static void sdhci_mask_irqs(struct sdhci_host *host, u32 irqs)
-{
-	sdhci_clear_set_irqs(host, irqs, 0);
-}
-
 static void sdhci_set_card_detection(struct sdhci_host *host, bool enable)
 {
-	u32 present, irqs;
+	u32 present;
 
 	if ((host->quirks & SDHCI_QUIRK_BROKEN_CARD_DETECTION) ||
 	    (host->mmc->caps & MMC_CAP_NONREMOVABLE))
 		return;
 
-	present = sdhci_readl(host, SDHCI_PRESENT_STATE) &
-			      SDHCI_CARD_PRESENT;
-	irqs = present ? SDHCI_INT_CARD_REMOVE : SDHCI_INT_CARD_INSERT;
+	if (enable) {
+		present = sdhci_readl(host, SDHCI_PRESENT_STATE) &
+				      SDHCI_CARD_PRESENT;
 
-	if (enable)
-		sdhci_unmask_irqs(host, irqs);
-	else
-		sdhci_mask_irqs(host, irqs);
+		host->ier |= present ? SDHCI_INT_CARD_REMOVE :
+				       SDHCI_INT_CARD_INSERT;
+	} else {
+		host->ier &= ~(SDHCI_INT_CARD_REMOVE | SDHCI_INT_CARD_INSERT);
+	}
+
+	sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+	sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
 }
 
 static void sdhci_enable_card_detection(struct sdhci_host *host)
@@ -180,22 +165,9 @@ static void sdhci_disable_card_detection(struct sdhci_host *host)
 	sdhci_set_card_detection(host, false);
 }
 
-static void sdhci_reset(struct sdhci_host *host, u8 mask)
+void sdhci_reset(struct sdhci_host *host, u8 mask)
 {
 	unsigned long timeout;
-	u32 uninitialized_var(ier);
-
-	if (host->quirks & SDHCI_QUIRK_NO_CARD_NO_RESET) {
-		if (!(sdhci_readl(host, SDHCI_PRESENT_STATE) &
-			SDHCI_CARD_PRESENT))
-			return;
-	}
-
-	if (host->quirks & SDHCI_QUIRK_RESTORE_IRQS_AFTER_RESET)
-		ier = sdhci_readl(host, SDHCI_INT_ENABLE);
-
-	if (host->ops->platform_reset_enter)
-		host->ops->platform_reset_enter(host, mask);
 
 	sdhci_writeb(host, mask, SDHCI_SOFTWARE_RESET);
 
@@ -220,16 +192,27 @@ static void sdhci_reset(struct sdhci_host *host, u8 mask)
 		timeout--;
 		mdelay(1);
 	}
+}
+EXPORT_SYMBOL_GPL(sdhci_reset);
+
+static void sdhci_do_reset(struct sdhci_host *host, u8 mask)
+{
+	if (host->quirks & SDHCI_QUIRK_NO_CARD_NO_RESET) {
+		if (!(sdhci_readl(host, SDHCI_PRESENT_STATE) &
+			SDHCI_CARD_PRESENT))
+			return;
+	}
 
-	if (host->ops->platform_reset_exit)
-		host->ops->platform_reset_exit(host, mask);
+	host->ops->reset(host, mask);
 
-	if (host->quirks & SDHCI_QUIRK_RESTORE_IRQS_AFTER_RESET)
-		sdhci_clear_set_irqs(host, SDHCI_INT_ALL_MASK, ier);
+	if (mask & SDHCI_RESET_ALL) {
+		if (host->flags & (SDHCI_USE_SDMA | SDHCI_USE_ADMA)) {
+			if (host->ops->enable_dma)
+				host->ops->enable_dma(host);
+		}
 
-	if (host->flags & (SDHCI_USE_SDMA | SDHCI_USE_ADMA)) {
-		if ((host->ops->enable_dma) && (mask & SDHCI_RESET_ALL))
-			host->ops->enable_dma(host);
+		/* Resetting the controller clears many */
+		host->preset_enabled = false;
 	}
 }
 
@@ -238,15 +221,18 @@ static void sdhci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios);
 static void sdhci_init(struct sdhci_host *host, int soft)
 {
 	if (soft)
-		sdhci_reset(host, SDHCI_RESET_CMD|SDHCI_RESET_DATA);
+		sdhci_do_reset(host, SDHCI_RESET_CMD|SDHCI_RESET_DATA);
 	else
-		sdhci_reset(host, SDHCI_RESET_ALL);
+		sdhci_do_reset(host, SDHCI_RESET_ALL);
 
-	sdhci_clear_set_irqs(host, SDHCI_INT_ALL_MASK,
-		SDHCI_INT_BUS_POWER | SDHCI_INT_DATA_END_BIT |
-		SDHCI_INT_DATA_CRC | SDHCI_INT_DATA_TIMEOUT | SDHCI_INT_INDEX |
-		SDHCI_INT_END_BIT | SDHCI_INT_CRC | SDHCI_INT_TIMEOUT |
-		SDHCI_INT_DATA_END | SDHCI_INT_RESPONSE);
+	host->ier = SDHCI_INT_BUS_POWER | SDHCI_INT_DATA_END_BIT |
+		    SDHCI_INT_DATA_CRC | SDHCI_INT_DATA_TIMEOUT |
+		    SDHCI_INT_INDEX | SDHCI_INT_END_BIT | SDHCI_INT_CRC |
+		    SDHCI_INT_TIMEOUT | SDHCI_INT_DATA_END |
+		    SDHCI_INT_RESPONSE;
+
+	sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+	sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
 
 	if (soft) {
 		/* force clock reconfiguration */
@@ -502,11 +488,6 @@ static int sdhci_adma_table_pre(struct sdhci_host *host,
 	else
 		direction = DMA_TO_DEVICE;
 
-	/*
-	 * The ADMA descriptor table is mapped further down as we
-	 * need to fill it with data first.
-	 */
-
 	host->align_addr = dma_map_single(mmc_dev(host->mmc),
 		host->align_buffer, 128 * 4, direction);
 	if (dma_mapping_error(mmc_dev(host->mmc), host->align_addr))
@@ -567,7 +548,7 @@ static int sdhci_adma_table_pre(struct sdhci_host *host,
 		 * If this triggers then we have a calculation bug
 		 * somewhere. :/
 		 */
-		WARN_ON((desc - host->adma_desc) > (128 * 2 + 1) * 4);
+		WARN_ON((desc - host->adma_desc) > ADMA_SIZE);
 	}
 
 	if (host->quirks & SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC) {
@@ -595,17 +576,8 @@ static int sdhci_adma_table_pre(struct sdhci_host *host,
 			host->align_addr, 128 * 4, direction);
 	}
 
-	host->adma_addr = dma_map_single(mmc_dev(host->mmc),
-		host->adma_desc, (128 * 2 + 1) * 4, DMA_TO_DEVICE);
-	if (dma_mapping_error(mmc_dev(host->mmc), host->adma_addr))
-		goto unmap_entries;
-	BUG_ON(host->adma_addr & 0x3);
-
 	return 0;
 
-unmap_entries:
-	dma_unmap_sg(mmc_dev(host->mmc), data->sg,
-		data->sg_len, direction);
 unmap_align:
 	dma_unmap_single(mmc_dev(host->mmc), host->align_addr,
 		128 * 4, direction);
@@ -623,19 +595,25 @@ static void sdhci_adma_table_post(struct sdhci_host *host,
 	u8 *align;
 	char *buffer;
 	unsigned long flags;
+	bool has_unaligned;
 
 	if (data->flags & MMC_DATA_READ)
 		direction = DMA_FROM_DEVICE;
 	else
 		direction = DMA_TO_DEVICE;
 
-	dma_unmap_single(mmc_dev(host->mmc), host->adma_addr,
-		(128 * 2 + 1) * 4, DMA_TO_DEVICE);
-
 	dma_unmap_single(mmc_dev(host->mmc), host->align_addr,
 		128 * 4, direction);
 
-	if (data->flags & MMC_DATA_READ) {
+	/* Do a quick scan of the SG list for any unaligned mappings */
+	has_unaligned = false;
+	for_each_sg(data->sg, sg, host->sg_count, i)
+		if (sg_dma_address(sg) & 3) {
+			has_unaligned = true;
+			break;
+		}
+
+	if (has_unaligned && data->flags & MMC_DATA_READ) {
 		dma_sync_sg_for_cpu(mmc_dev(host->mmc), data->sg,
 			data->sg_len, direction);
 
@@ -721,9 +699,12 @@ static void sdhci_set_transfer_irqs(struct sdhci_host *host)
 	u32 dma_irqs = SDHCI_INT_DMA_END | SDHCI_INT_ADMA_ERROR;
 
 	if (host->flags & SDHCI_REQ_USE_DMA)
-		sdhci_clear_set_irqs(host, pio_irqs, dma_irqs);
+		host->ier = (host->ier & ~pio_irqs) | dma_irqs;
 	else
-		sdhci_clear_set_irqs(host, dma_irqs, pio_irqs);
+		host->ier = (host->ier & ~dma_irqs) | pio_irqs;
+
+	sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+	sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
 }
 
 static void sdhci_prepare_data(struct sdhci_host *host, struct mmc_command *cmd)
@@ -976,8 +957,8 @@ static void sdhci_finish_data(struct sdhci_host *host)
 		 * upon error conditions.
 		 */
 		if (data->error) {
-			sdhci_reset(host, SDHCI_RESET_CMD);
-			sdhci_reset(host, SDHCI_RESET_DATA);
+			sdhci_do_reset(host, SDHCI_RESET_CMD);
+			sdhci_do_reset(host, SDHCI_RESET_DATA);
 		}
 
 		sdhci_send_command(host, data->stop);
@@ -1107,24 +1088,23 @@ static void sdhci_finish_command(struct sdhci_host *host)
 
 static u16 sdhci_get_preset_value(struct sdhci_host *host)
 {
-	u16 ctrl, preset = 0;
-
-	ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
+	u16 preset = 0;
 
-	switch (ctrl & SDHCI_CTRL_UHS_MASK) {
-	case SDHCI_CTRL_UHS_SDR12:
+	switch (host->timing) {
+	case MMC_TIMING_UHS_SDR12:
 		preset = sdhci_readw(host, SDHCI_PRESET_FOR_SDR12);
 		break;
-	case SDHCI_CTRL_UHS_SDR25:
+	case MMC_TIMING_UHS_SDR25:
 		preset = sdhci_readw(host, SDHCI_PRESET_FOR_SDR25);
 		break;
-	case SDHCI_CTRL_UHS_SDR50:
+	case MMC_TIMING_UHS_SDR50:
 		preset = sdhci_readw(host, SDHCI_PRESET_FOR_SDR50);
 		break;
-	case SDHCI_CTRL_UHS_SDR104:
+	case MMC_TIMING_UHS_SDR104:
+	case MMC_TIMING_MMC_HS200:
 		preset = sdhci_readw(host, SDHCI_PRESET_FOR_SDR104);
 		break;
-	case SDHCI_CTRL_UHS_DDR50:
+	case MMC_TIMING_UHS_DDR50:
 		preset = sdhci_readw(host, SDHCI_PRESET_FOR_DDR50);
 		break;
 	default:
@@ -1136,32 +1116,22 @@ static u16 sdhci_get_preset_value(struct sdhci_host *host)
 	return preset;
 }
 
-static void sdhci_set_clock(struct sdhci_host *host, unsigned int clock)
+void sdhci_set_clock(struct sdhci_host *host, unsigned int clock)
 {
 	int div = 0; /* Initialized for compiler warning */
 	int real_div = div, clk_mul = 1;
 	u16 clk = 0;
 	unsigned long timeout;
 
-	if (clock && clock == host->clock)
-		return;
-
 	host->mmc->actual_clock = 0;
 
-	if (host->ops->set_clock) {
-		host->ops->set_clock(host, clock);
-		if (host->quirks & SDHCI_QUIRK_NONSTANDARD_CLOCK)
-			return;
-	}
-
 	sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL);
 
 	if (clock == 0)
-		goto out;
+		return;
 
 	if (host->version >= SDHCI_SPEC_300) {
-		if (sdhci_readw(host, SDHCI_HOST_CONTROL2) &
-			SDHCI_CTRL_PRESET_VAL_ENABLE) {
+		if (host->preset_enabled) {
 			u16 pre_val;
 
 			clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
@@ -1247,26 +1217,16 @@ clock_set:
 
 	clk |= SDHCI_CLOCK_CARD_EN;
 	sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
-
-out:
-	host->clock = clock;
 }
+EXPORT_SYMBOL_GPL(sdhci_set_clock);
 
-static inline void sdhci_update_clock(struct sdhci_host *host)
-{
-	unsigned int clock;
-
-	clock = host->clock;
-	host->clock = 0;
-	sdhci_set_clock(host, clock);
-}
-
-static int sdhci_set_power(struct sdhci_host *host, unsigned short power)
+static void sdhci_set_power(struct sdhci_host *host, unsigned char mode,
+			    unsigned short vdd)
 {
 	u8 pwr = 0;
 
-	if (power != (unsigned short)-1) {
-		switch (1 << power) {
+	if (mode != MMC_POWER_OFF) {
+		switch (1 << vdd) {
 		case MMC_VDD_165_195:
 			pwr = SDHCI_POWER_180;
 			break;
@@ -1284,7 +1244,7 @@ static int sdhci_set_power(struct sdhci_host *host, unsigned short power)
 	}
 
 	if (host->pwr == pwr)
-		return -1;
+		return;
 
 	host->pwr = pwr;
 
@@ -1292,38 +1252,43 @@ static int sdhci_set_power(struct sdhci_host *host, unsigned short power)
 		sdhci_writeb(host, 0, SDHCI_POWER_CONTROL);
 		if (host->quirks2 & SDHCI_QUIRK2_CARD_ON_NEEDS_BUS_ON)
 			sdhci_runtime_pm_bus_off(host);
-		return 0;
-	}
-
-	/*
-	 * Spec says that we should clear the power reg before setting
-	 * a new value. Some controllers don't seem to like this though.
-	 */
-	if (!(host->quirks & SDHCI_QUIRK_SINGLE_POWER_WRITE))
-		sdhci_writeb(host, 0, SDHCI_POWER_CONTROL);
+		vdd = 0;
+	} else {
+		/*
+		 * Spec says that we should clear the power reg before setting
+		 * a new value. Some controllers don't seem to like this though.
+		 */
+		if (!(host->quirks & SDHCI_QUIRK_SINGLE_POWER_WRITE))
+			sdhci_writeb(host, 0, SDHCI_POWER_CONTROL);
 
-	/*
-	 * At least the Marvell CaFe chip gets confused if we set the voltage
-	 * and set turn on power at the same time, so set the voltage first.
-	 */
-	if (host->quirks & SDHCI_QUIRK_NO_SIMULT_VDD_AND_POWER)
-		sdhci_writeb(host, pwr, SDHCI_POWER_CONTROL);
+		/*
+		 * At least the Marvell CaFe chip gets confused if we set the
+		 * voltage and set turn on power at the same time, so set the
+		 * voltage first.
+		 */
+		if (host->quirks & SDHCI_QUIRK_NO_SIMULT_VDD_AND_POWER)
+			sdhci_writeb(host, pwr, SDHCI_POWER_CONTROL);
 
-	pwr |= SDHCI_POWER_ON;
+		pwr |= SDHCI_POWER_ON;
 
-	sdhci_writeb(host, pwr, SDHCI_POWER_CONTROL);
+		sdhci_writeb(host, pwr, SDHCI_POWER_CONTROL);
 
-	if (host->quirks2 & SDHCI_QUIRK2_CARD_ON_NEEDS_BUS_ON)
-		sdhci_runtime_pm_bus_on(host);
+		if (host->quirks2 & SDHCI_QUIRK2_CARD_ON_NEEDS_BUS_ON)
+			sdhci_runtime_pm_bus_on(host);
 
-	/*
-	 * Some controllers need an extra 10ms delay of 10ms before they
-	 * can apply clock after applying power
-	 */
-	if (host->quirks & SDHCI_QUIRK_DELAY_AFTER_POWER)
-		mdelay(10);
+		/*
+		 * Some controllers need an extra 10ms delay of 10ms before
+		 * they can apply clock after applying power
+		 */
+		if (host->quirks & SDHCI_QUIRK_DELAY_AFTER_POWER)
+			mdelay(10);
+	}
 
-	return power;
+	if (host->vmmc) {
+		spin_unlock_irq(&host->lock);
+		mmc_regulator_set_ocr(host->mmc, host->vmmc, vdd);
+		spin_lock_irq(&host->lock);
+	}
 }
 
 /*****************************************************************************\
@@ -1427,10 +1392,53 @@ static void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq)
 	spin_unlock_irqrestore(&host->lock, flags);
 }
 
+void sdhci_set_bus_width(struct sdhci_host *host, int width)
+{
+	u8 ctrl;
+
+	ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
+	if (width == MMC_BUS_WIDTH_8) {
+		ctrl &= ~SDHCI_CTRL_4BITBUS;
+		if (host->version >= SDHCI_SPEC_300)
+			ctrl |= SDHCI_CTRL_8BITBUS;
+	} else {
+		if (host->version >= SDHCI_SPEC_300)
+			ctrl &= ~SDHCI_CTRL_8BITBUS;
+		if (width == MMC_BUS_WIDTH_4)
+			ctrl |= SDHCI_CTRL_4BITBUS;
+		else
+			ctrl &= ~SDHCI_CTRL_4BITBUS;
+	}
+	sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
+}
+EXPORT_SYMBOL_GPL(sdhci_set_bus_width);
+
+void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing)
+{
+	u16 ctrl_2;
+
+	ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
+	/* Select Bus Speed Mode for host */
+	ctrl_2 &= ~SDHCI_CTRL_UHS_MASK;
+	if ((timing == MMC_TIMING_MMC_HS200) ||
+	    (timing == MMC_TIMING_UHS_SDR104))
+		ctrl_2 |= SDHCI_CTRL_UHS_SDR104;
+	else if (timing == MMC_TIMING_UHS_SDR12)
+		ctrl_2 |= SDHCI_CTRL_UHS_SDR12;
+	else if (timing == MMC_TIMING_UHS_SDR25)
+		ctrl_2 |= SDHCI_CTRL_UHS_SDR25;
+	else if (timing == MMC_TIMING_UHS_SDR50)
+		ctrl_2 |= SDHCI_CTRL_UHS_SDR50;
+	else if ((timing == MMC_TIMING_UHS_DDR50) ||
+		 (timing == MMC_TIMING_MMC_DDR52))
+		ctrl_2 |= SDHCI_CTRL_UHS_DDR50;
+	sdhci_writew(host, ctrl_2, SDHCI_HOST_CONTROL2);
+}
+EXPORT_SYMBOL_GPL(sdhci_set_uhs_signaling);
+
 static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 {
 	unsigned long flags;
-	int vdd_bit = -1;
 	u8 ctrl;
 
 	spin_lock_irqsave(&host->lock, flags);
@@ -1456,45 +1464,17 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 		!(host->quirks2 & SDHCI_QUIRK2_PRESET_VALUE_BROKEN))
 		sdhci_enable_preset_value(host, false);
 
-	sdhci_set_clock(host, ios->clock);
-
-	if (ios->power_mode == MMC_POWER_OFF)
-		vdd_bit = sdhci_set_power(host, -1);
-	else
-		vdd_bit = sdhci_set_power(host, ios->vdd);
-
-	if (host->vmmc && vdd_bit != -1) {
-		spin_unlock_irqrestore(&host->lock, flags);
-		mmc_regulator_set_ocr(host->mmc, host->vmmc, vdd_bit);
-		spin_lock_irqsave(&host->lock, flags);
+	if (!ios->clock || ios->clock != host->clock) {
+		host->ops->set_clock(host, ios->clock);
+		host->clock = ios->clock;
 	}
 
+	sdhci_set_power(host, ios->power_mode, ios->vdd);
+
 	if (host->ops->platform_send_init_74_clocks)
 		host->ops->platform_send_init_74_clocks(host, ios->power_mode);
 
-	/*
-	 * If your platform has 8-bit width support but is not a v3 controller,
-	 * or if it requires special setup code, you should implement that in
-	 * platform_bus_width().
-	 */
-	if (host->ops->platform_bus_width) {
-		host->ops->platform_bus_width(host, ios->bus_width);
-	} else {
-		ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
-		if (ios->bus_width == MMC_BUS_WIDTH_8) {
-			ctrl &= ~SDHCI_CTRL_4BITBUS;
-			if (host->version >= SDHCI_SPEC_300)
-				ctrl |= SDHCI_CTRL_8BITBUS;
-		} else {
-			if (host->version >= SDHCI_SPEC_300)
-				ctrl &= ~SDHCI_CTRL_8BITBUS;
-			if (ios->bus_width == MMC_BUS_WIDTH_4)
-				ctrl |= SDHCI_CTRL_4BITBUS;
-			else
-				ctrl &= ~SDHCI_CTRL_4BITBUS;
-		}
-		sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
-	}
+	host->ops->set_bus_width(host, ios->bus_width);
 
 	ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
 
@@ -1510,19 +1490,20 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 
 		/* In case of UHS-I modes, set High Speed Enable */
 		if ((ios->timing == MMC_TIMING_MMC_HS200) ||
+		    (ios->timing == MMC_TIMING_MMC_DDR52) ||
 		    (ios->timing == MMC_TIMING_UHS_SDR50) ||
 		    (ios->timing == MMC_TIMING_UHS_SDR104) ||
 		    (ios->timing == MMC_TIMING_UHS_DDR50) ||
 		    (ios->timing == MMC_TIMING_UHS_SDR25))
 			ctrl |= SDHCI_CTRL_HISPD;
 
-		ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
-		if (!(ctrl_2 & SDHCI_CTRL_PRESET_VAL_ENABLE)) {
+		if (!host->preset_enabled) {
 			sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
 			/*
 			 * We only need to set Driver Strength if the
 			 * preset value enable is not set.
 			 */
+			ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
 			ctrl_2 &= ~SDHCI_CTRL_DRV_TYPE_MASK;
 			if (ios->drv_type == MMC_SET_DRIVER_TYPE_A)
 				ctrl_2 |= SDHCI_CTRL_DRV_TYPE_A;
@@ -1546,7 +1527,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 			sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
 
 			/* Re-enable SD Clock */
-			sdhci_update_clock(host);
+			host->ops->set_clock(host, host->clock);
 		}
 
 
@@ -1555,25 +1536,8 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 		clk &= ~SDHCI_CLOCK_CARD_EN;
 		sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
 
-		if (host->ops->set_uhs_signaling)
-			host->ops->set_uhs_signaling(host, ios->timing);
-		else {
-			ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
-			/* Select Bus Speed Mode for host */
-			ctrl_2 &= ~SDHCI_CTRL_UHS_MASK;
-			if ((ios->timing == MMC_TIMING_MMC_HS200) ||
-			    (ios->timing == MMC_TIMING_UHS_SDR104))
-				ctrl_2 |= SDHCI_CTRL_UHS_SDR104;
-			else if (ios->timing == MMC_TIMING_UHS_SDR12)
-				ctrl_2 |= SDHCI_CTRL_UHS_SDR12;
-			else if (ios->timing == MMC_TIMING_UHS_SDR25)
-				ctrl_2 |= SDHCI_CTRL_UHS_SDR25;
-			else if (ios->timing == MMC_TIMING_UHS_SDR50)
-				ctrl_2 |= SDHCI_CTRL_UHS_SDR50;
-			else if (ios->timing == MMC_TIMING_UHS_DDR50)
-				ctrl_2 |= SDHCI_CTRL_UHS_DDR50;
-			sdhci_writew(host, ctrl_2, SDHCI_HOST_CONTROL2);
-		}
+		host->ops->set_uhs_signaling(host, ios->timing);
+		host->timing = ios->timing;
 
 		if (!(host->quirks2 & SDHCI_QUIRK2_PRESET_VALUE_BROKEN) &&
 				((ios->timing == MMC_TIMING_UHS_SDR12) ||
@@ -1590,7 +1554,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 		}
 
 		/* Re-enable SD Clock */
-		sdhci_update_clock(host);
+		host->ops->set_clock(host, host->clock);
 	} else
 		sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
 
@@ -1600,7 +1564,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 	 * it on each ios seems to solve the problem.
 	 */
 	if(host->quirks & SDHCI_QUIRK_RESET_CMD_DATA_ON_IOS)
-		sdhci_reset(host, SDHCI_RESET_CMD | SDHCI_RESET_DATA);
+		sdhci_do_reset(host, SDHCI_RESET_CMD | SDHCI_RESET_DATA);
 
 	mmiowb();
 	spin_unlock_irqrestore(&host->lock, flags);
@@ -1709,24 +1673,16 @@ static int sdhci_get_ro(struct mmc_host *mmc)
 
 static void sdhci_enable_sdio_irq_nolock(struct sdhci_host *host, int enable)
 {
-	if (host->flags & SDHCI_DEVICE_DEAD)
-		goto out;
-
-	if (enable)
-		host->flags |= SDHCI_SDIO_IRQ_ENABLED;
-	else
-		host->flags &= ~SDHCI_SDIO_IRQ_ENABLED;
-
-	/* SDIO IRQ will be enabled as appropriate in runtime resume */
-	if (host->runtime_suspended)
-		goto out;
+	if (!(host->flags & SDHCI_DEVICE_DEAD)) {
+		if (enable)
+			host->ier |= SDHCI_INT_CARD_INT;
+		else
+			host->ier &= ~SDHCI_INT_CARD_INT;
 
-	if (enable)
-		sdhci_unmask_irqs(host, SDHCI_INT_CARD_INT);
-	else
-		sdhci_mask_irqs(host, SDHCI_INT_CARD_INT);
-out:
-	mmiowb();
+		sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+		sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
+		mmiowb();
+	}
 }
 
 static void sdhci_enable_sdio_irq(struct mmc_host *mmc, int enable)
@@ -1734,9 +1690,18 @@ static void sdhci_enable_sdio_irq(struct mmc_host *mmc, int enable)
 	struct sdhci_host *host = mmc_priv(mmc);
 	unsigned long flags;
 
+	sdhci_runtime_pm_get(host);
+
 	spin_lock_irqsave(&host->lock, flags);
+	if (enable)
+		host->flags |= SDHCI_SDIO_IRQ_ENABLED;
+	else
+		host->flags &= ~SDHCI_SDIO_IRQ_ENABLED;
+
 	sdhci_enable_sdio_irq_nolock(host, enable);
 	spin_unlock_irqrestore(&host->lock, flags);
+
+	sdhci_runtime_pm_put(host);
 }
 
 static int sdhci_do_start_signal_voltage_switch(struct sdhci_host *host,
@@ -1855,22 +1820,15 @@ static int sdhci_card_busy(struct mmc_host *mmc)
 
 static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 {
-	struct sdhci_host *host;
+	struct sdhci_host *host = mmc_priv(mmc);
 	u16 ctrl;
-	u32 ier;
 	int tuning_loop_counter = MAX_TUNING_LOOP;
-	unsigned long timeout;
 	int err = 0;
-	bool requires_tuning_nonuhs = false;
 	unsigned long flags;
 
-	host = mmc_priv(mmc);
-
 	sdhci_runtime_pm_get(host);
 	spin_lock_irqsave(&host->lock, flags);
 
-	ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
-
 	/*
 	 * The Host Controller needs tuning only in case of SDR104 mode
 	 * and for SDR50 mode when Use Tuning for SDR50 is set in the
@@ -1878,15 +1836,18 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 	 * If the Host Controller supports the HS200 mode then the
 	 * tuning function has to be executed.
 	 */
-	if (((ctrl & SDHCI_CTRL_UHS_MASK) == SDHCI_CTRL_UHS_SDR50) &&
-	    (host->flags & SDHCI_SDR50_NEEDS_TUNING ||
-	     host->flags & SDHCI_SDR104_NEEDS_TUNING))
-		requires_tuning_nonuhs = true;
-
-	if (((ctrl & SDHCI_CTRL_UHS_MASK) == SDHCI_CTRL_UHS_SDR104) ||
-	    requires_tuning_nonuhs)
-		ctrl |= SDHCI_CTRL_EXEC_TUNING;
-	else {
+	switch (host->timing) {
+	case MMC_TIMING_MMC_HS200:
+	case MMC_TIMING_UHS_SDR104:
+		break;
+
+	case MMC_TIMING_UHS_SDR50:
+		if (host->flags & SDHCI_SDR50_NEEDS_TUNING ||
+		    host->flags & SDHCI_SDR104_NEEDS_TUNING)
+			break;
+		/* FALLTHROUGH */
+
+	default:
 		spin_unlock_irqrestore(&host->lock, flags);
 		sdhci_runtime_pm_put(host);
 		return 0;
@@ -1899,6 +1860,8 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 		return err;
 	}
 
+	ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
+	ctrl |= SDHCI_CTRL_EXEC_TUNING;
 	sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
 
 	/*
@@ -1911,21 +1874,17 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 	 * to make sure we don't hit a controller bug, we _only_
 	 * enable Buffer Read Ready interrupt here.
 	 */
-	ier = sdhci_readl(host, SDHCI_INT_ENABLE);
-	sdhci_clear_set_irqs(host, ier, SDHCI_INT_DATA_AVAIL);
+	sdhci_writel(host, SDHCI_INT_DATA_AVAIL, SDHCI_INT_ENABLE);
+	sdhci_writel(host, SDHCI_INT_DATA_AVAIL, SDHCI_SIGNAL_ENABLE);
 
 	/*
 	 * Issue CMD19 repeatedly till Execute Tuning is set to 0 or the number
 	 * of loops reaches 40 times or a timeout of 150ms occurs.
 	 */
-	timeout = 150;
 	do {
 		struct mmc_command cmd = {0};
 		struct mmc_request mrq = {NULL};
 
-		if (!tuning_loop_counter && !timeout)
-			break;
-
 		cmd.opcode = opcode;
 		cmd.arg = 0;
 		cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC;
@@ -1933,6 +1892,9 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 		cmd.data = NULL;
 		cmd.error = 0;
 
+		if (tuning_loop_counter-- == 0)
+			break;
+
 		mrq.cmd = &cmd;
 		host->mrq = &mrq;
 
@@ -1990,26 +1952,25 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 		host->tuning_done = 0;
 
 		ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
-		tuning_loop_counter--;
-		timeout--;
-		mdelay(1);
+
+		/* eMMC spec does not require a delay between tuning cycles */
+		if (opcode == MMC_SEND_TUNING_BLOCK)
+			mdelay(1);
 	} while (ctrl & SDHCI_CTRL_EXEC_TUNING);
 
 	/*
 	 * The Host Driver has exhausted the maximum number of loops allowed,
 	 * so use fixed sampling frequency.
 	 */
-	if (!tuning_loop_counter || !timeout) {
+	if (tuning_loop_counter < 0) {
 		ctrl &= ~SDHCI_CTRL_TUNED_CLK;
 		sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
+	}
+	if (!(ctrl & SDHCI_CTRL_TUNED_CLK)) {
+		pr_info(DRIVER_NAME ": Tuning procedure"
+			" failed, falling back to fixed sampling"
+			" clock\n");
 		err = -EIO;
-	} else {
-		if (!(ctrl & SDHCI_CTRL_TUNED_CLK)) {
-			pr_info(DRIVER_NAME ": Tuning procedure"
-				" failed, falling back to fixed sampling"
-				" clock\n");
-			err = -EIO;
-		}
 	}
 
 out:
@@ -2044,7 +2005,8 @@ out:
 	if (err && (host->flags & SDHCI_USING_RETUNING_TIMER))
 		err = 0;
 
-	sdhci_clear_set_irqs(host, SDHCI_INT_DATA_AVAIL, ier);
+	sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+	sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
 	spin_unlock_irqrestore(&host->lock, flags);
 	sdhci_runtime_pm_put(host);
 
@@ -2054,26 +2016,30 @@ out:
 
 static void sdhci_enable_preset_value(struct sdhci_host *host, bool enable)
 {
-	u16 ctrl;
-
 	/* Host Controller v3.00 defines preset value registers */
 	if (host->version < SDHCI_SPEC_300)
 		return;
 
-	ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
-
 	/*
 	 * We only enable or disable Preset Value if they are not already
 	 * enabled or disabled respectively. Otherwise, we bail out.
 	 */
-	if (enable && !(ctrl & SDHCI_CTRL_PRESET_VAL_ENABLE)) {
-		ctrl |= SDHCI_CTRL_PRESET_VAL_ENABLE;
-		sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
-		host->flags |= SDHCI_PV_ENABLED;
-	} else if (!enable && (ctrl & SDHCI_CTRL_PRESET_VAL_ENABLE)) {
-		ctrl &= ~SDHCI_CTRL_PRESET_VAL_ENABLE;
+	if (host->preset_enabled != enable) {
+		u16 ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
+
+		if (enable)
+			ctrl |= SDHCI_CTRL_PRESET_VAL_ENABLE;
+		else
+			ctrl &= ~SDHCI_CTRL_PRESET_VAL_ENABLE;
+
 		sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
-		host->flags &= ~SDHCI_PV_ENABLED;
+
+		if (enable)
+			host->flags |= SDHCI_PV_ENABLED;
+		else
+			host->flags &= ~SDHCI_PV_ENABLED;
+
+		host->preset_enabled = enable;
 	}
 }
 
@@ -2095,8 +2061,8 @@ static void sdhci_card_event(struct mmc_host *mmc)
 		pr_err("%s: Resetting controller.\n",
 			mmc_hostname(host->mmc));
 
-		sdhci_reset(host, SDHCI_RESET_CMD);
-		sdhci_reset(host, SDHCI_RESET_DATA);
+		sdhci_do_reset(host, SDHCI_RESET_CMD);
+		sdhci_do_reset(host, SDHCI_RESET_DATA);
 
 		host->mrq->cmd->error = -ENOMEDIUM;
 		tasklet_schedule(&host->finish_tasklet);
@@ -2124,15 +2090,6 @@ static const struct mmc_host_ops sdhci_ops = {
  *                                                                           *
 \*****************************************************************************/
 
-static void sdhci_tasklet_card(unsigned long param)
-{
-	struct sdhci_host *host = (struct sdhci_host*)param;
-
-	sdhci_card_event(host->mmc);
-
-	mmc_detect_change(host->mmc, msecs_to_jiffies(200));
-}
-
 static void sdhci_tasklet_finish(unsigned long param)
 {
 	struct sdhci_host *host;
@@ -2169,12 +2126,12 @@ static void sdhci_tasklet_finish(unsigned long param)
 		/* Some controllers need this kick or reset won't work here */
 		if (host->quirks & SDHCI_QUIRK_CLOCK_BEFORE_RESET)
 			/* This is to force an update */
-			sdhci_update_clock(host);
+			host->ops->set_clock(host, host->clock);
 
 		/* Spec says we should do both at the same time, but Ricoh
 		   controllers do not like that. */
-		sdhci_reset(host, SDHCI_RESET_CMD);
-		sdhci_reset(host, SDHCI_RESET_DATA);
+		sdhci_do_reset(host, SDHCI_RESET_CMD);
+		sdhci_do_reset(host, SDHCI_RESET_DATA);
 	}
 
 	host->mrq = NULL;
@@ -2424,101 +2381,94 @@ static void sdhci_data_irq(struct sdhci_host *host, u32 intmask)
 
 static irqreturn_t sdhci_irq(int irq, void *dev_id)
 {
-	irqreturn_t result;
+	irqreturn_t result = IRQ_NONE;
 	struct sdhci_host *host = dev_id;
-	u32 intmask, unexpected = 0;
-	int cardint = 0, max_loops = 16;
+	u32 intmask, mask, unexpected = 0;
+	int max_loops = 16;
 
 	spin_lock(&host->lock);
 
-	if (host->runtime_suspended) {
+	if (host->runtime_suspended && !sdhci_sdio_irq_enabled(host)) {
 		spin_unlock(&host->lock);
 		return IRQ_NONE;
 	}
 
 	intmask = sdhci_readl(host, SDHCI_INT_STATUS);
-
 	if (!intmask || intmask == 0xffffffff) {
 		result = IRQ_NONE;
 		goto out;
 	}
 
-again:
-	DBG("*** %s got interrupt: 0x%08x\n",
-		mmc_hostname(host->mmc), intmask);
-
-	if (intmask & (SDHCI_INT_CARD_INSERT | SDHCI_INT_CARD_REMOVE)) {
-		u32 present = sdhci_readl(host, SDHCI_PRESENT_STATE) &
-			      SDHCI_CARD_PRESENT;
-
-		/*
-		 * There is a observation on i.mx esdhc.  INSERT bit will be
-		 * immediately set again when it gets cleared, if a card is
-		 * inserted.  We have to mask the irq to prevent interrupt
-		 * storm which will freeze the system.  And the REMOVE gets
-		 * the same situation.
-		 *
-		 * More testing are needed here to ensure it works for other
-		 * platforms though.
-		 */
-		sdhci_mask_irqs(host, present ? SDHCI_INT_CARD_INSERT :
-						SDHCI_INT_CARD_REMOVE);
-		sdhci_unmask_irqs(host, present ? SDHCI_INT_CARD_REMOVE :
-						  SDHCI_INT_CARD_INSERT);
-
-		sdhci_writel(host, intmask & (SDHCI_INT_CARD_INSERT |
-			     SDHCI_INT_CARD_REMOVE), SDHCI_INT_STATUS);
-		intmask &= ~(SDHCI_INT_CARD_INSERT | SDHCI_INT_CARD_REMOVE);
-		tasklet_schedule(&host->card_tasklet);
-	}
-
-	if (intmask & SDHCI_INT_CMD_MASK) {
-		sdhci_writel(host, intmask & SDHCI_INT_CMD_MASK,
-			SDHCI_INT_STATUS);
-		sdhci_cmd_irq(host, intmask & SDHCI_INT_CMD_MASK);
-	}
+	do {
+		/* Clear selected interrupts. */
+		mask = intmask & (SDHCI_INT_CMD_MASK | SDHCI_INT_DATA_MASK |
+				  SDHCI_INT_BUS_POWER);
+		sdhci_writel(host, mask, SDHCI_INT_STATUS);
 
-	if (intmask & SDHCI_INT_DATA_MASK) {
-		sdhci_writel(host, intmask & SDHCI_INT_DATA_MASK,
-			SDHCI_INT_STATUS);
-		sdhci_data_irq(host, intmask & SDHCI_INT_DATA_MASK);
-	}
+		DBG("*** %s got interrupt: 0x%08x\n",
+			mmc_hostname(host->mmc), intmask);
 
-	intmask &= ~(SDHCI_INT_CMD_MASK | SDHCI_INT_DATA_MASK);
+		if (intmask & (SDHCI_INT_CARD_INSERT | SDHCI_INT_CARD_REMOVE)) {
+			u32 present = sdhci_readl(host, SDHCI_PRESENT_STATE) &
+				      SDHCI_CARD_PRESENT;
 
-	intmask &= ~SDHCI_INT_ERROR;
+			/*
+			 * There is a observation on i.mx esdhc.  INSERT
+			 * bit will be immediately set again when it gets
+			 * cleared, if a card is inserted.  We have to mask
+			 * the irq to prevent interrupt storm which will
+			 * freeze the system.  And the REMOVE gets the
+			 * same situation.
+			 *
+			 * More testing are needed here to ensure it works
+			 * for other platforms though.
+			 */
+			host->ier &= ~(SDHCI_INT_CARD_INSERT |
+				       SDHCI_INT_CARD_REMOVE);
+			host->ier |= present ? SDHCI_INT_CARD_REMOVE :
+					       SDHCI_INT_CARD_INSERT;
+			sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+			sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
+
+			sdhci_writel(host, intmask & (SDHCI_INT_CARD_INSERT |
+				     SDHCI_INT_CARD_REMOVE), SDHCI_INT_STATUS);
+
+			host->thread_isr |= intmask & (SDHCI_INT_CARD_INSERT |
+						       SDHCI_INT_CARD_REMOVE);
+			result = IRQ_WAKE_THREAD;
+		}
 
-	if (intmask & SDHCI_INT_BUS_POWER) {
-		pr_err("%s: Card is consuming too much power!\n",
-			mmc_hostname(host->mmc));
-		sdhci_writel(host, SDHCI_INT_BUS_POWER, SDHCI_INT_STATUS);
-	}
+		if (intmask & SDHCI_INT_CMD_MASK)
+			sdhci_cmd_irq(host, intmask & SDHCI_INT_CMD_MASK);
 
-	intmask &= ~SDHCI_INT_BUS_POWER;
+		if (intmask & SDHCI_INT_DATA_MASK)
+			sdhci_data_irq(host, intmask & SDHCI_INT_DATA_MASK);
 
-	if (intmask & SDHCI_INT_CARD_INT)
-		cardint = 1;
+		if (intmask & SDHCI_INT_BUS_POWER)
+			pr_err("%s: Card is consuming too much power!\n",
+				mmc_hostname(host->mmc));
 
-	intmask &= ~SDHCI_INT_CARD_INT;
+		if (intmask & SDHCI_INT_CARD_INT) {
+			sdhci_enable_sdio_irq_nolock(host, false);
+			host->thread_isr |= SDHCI_INT_CARD_INT;
+			result = IRQ_WAKE_THREAD;
+		}
 
-	if (intmask) {
-		unexpected |= intmask;
-		sdhci_writel(host, intmask, SDHCI_INT_STATUS);
-	}
+		intmask &= ~(SDHCI_INT_CARD_INSERT | SDHCI_INT_CARD_REMOVE |
+			     SDHCI_INT_CMD_MASK | SDHCI_INT_DATA_MASK |
+			     SDHCI_INT_ERROR | SDHCI_INT_BUS_POWER |
+			     SDHCI_INT_CARD_INT);
 
-	result = IRQ_HANDLED;
+		if (intmask) {
+			unexpected |= intmask;
+			sdhci_writel(host, intmask, SDHCI_INT_STATUS);
+		}
 
-	intmask = sdhci_readl(host, SDHCI_INT_STATUS);
+		if (result == IRQ_NONE)
+			result = IRQ_HANDLED;
 
-	/*
-	 * If we know we'll call the driver to signal SDIO IRQ, disregard
-	 * further indications of Card Interrupt in the status to avoid a
-	 * needless loop.
-	 */
-	if (cardint)
-		intmask &= ~SDHCI_INT_CARD_INT;
-	if (intmask && --max_loops)
-		goto again;
+		intmask = sdhci_readl(host, SDHCI_INT_STATUS);
+	} while (intmask && --max_loops);
 out:
 	spin_unlock(&host->lock);
 
@@ -2527,15 +2477,38 @@ out:
 			   mmc_hostname(host->mmc), unexpected);
 		sdhci_dumpregs(host);
 	}
-	/*
-	 * We have to delay this as it calls back into the driver.
-	 */
-	if (cardint)
-		mmc_signal_sdio_irq(host->mmc);
 
 	return result;
 }
 
+static irqreturn_t sdhci_thread_irq(int irq, void *dev_id)
+{
+	struct sdhci_host *host = dev_id;
+	unsigned long flags;
+	u32 isr;
+
+	spin_lock_irqsave(&host->lock, flags);
+	isr = host->thread_isr;
+	host->thread_isr = 0;
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	if (isr & (SDHCI_INT_CARD_INSERT | SDHCI_INT_CARD_REMOVE)) {
+		sdhci_card_event(host->mmc);
+		mmc_detect_change(host->mmc, msecs_to_jiffies(200));
+	}
+
+	if (isr & SDHCI_INT_CARD_INT) {
+		sdio_run_irqs(host->mmc);
+
+		spin_lock_irqsave(&host->lock, flags);
+		if (host->flags & SDHCI_SDIO_IRQ_ENABLED)
+			sdhci_enable_sdio_irq_nolock(host, true);
+		spin_unlock_irqrestore(&host->lock, flags);
+	}
+
+	return isr ? IRQ_HANDLED : IRQ_NONE;
+}
+
 /*****************************************************************************\
  *                                                                           *
  * Suspend/resume                                                            *
@@ -2572,9 +2545,6 @@ EXPORT_SYMBOL_GPL(sdhci_disable_irq_wakeups);
 
 int sdhci_suspend_host(struct sdhci_host *host)
 {
-	if (host->ops->platform_suspend)
-		host->ops->platform_suspend(host);
-
 	sdhci_disable_card_detection(host);
 
 	/* Disable tuning since we are suspending */
@@ -2584,7 +2554,9 @@ int sdhci_suspend_host(struct sdhci_host *host)
 	}
 
 	if (!device_may_wakeup(mmc_dev(host->mmc))) {
-		sdhci_mask_irqs(host, SDHCI_INT_ALL_MASK);
+		host->ier = 0;
+		sdhci_writel(host, 0, SDHCI_INT_ENABLE);
+		sdhci_writel(host, 0, SDHCI_SIGNAL_ENABLE);
 		free_irq(host->irq, host);
 	} else {
 		sdhci_enable_irq_wakeups(host);
@@ -2605,8 +2577,9 @@ int sdhci_resume_host(struct sdhci_host *host)
 	}
 
 	if (!device_may_wakeup(mmc_dev(host->mmc))) {
-		ret = request_irq(host->irq, sdhci_irq, IRQF_SHARED,
-				  mmc_hostname(host->mmc), host);
+		ret = request_threaded_irq(host->irq, sdhci_irq,
+					   sdhci_thread_irq, IRQF_SHARED,
+					   mmc_hostname(host->mmc), host);
 		if (ret)
 			return ret;
 	} else {
@@ -2628,9 +2601,6 @@ int sdhci_resume_host(struct sdhci_host *host)
 
 	sdhci_enable_card_detection(host);
 
-	if (host->ops->platform_resume)
-		host->ops->platform_resume(host);
-
 	/* Set the re-tuning expiration flag */
 	if (host->flags & SDHCI_USING_RETUNING_TIMER)
 		host->flags |= SDHCI_NEEDS_RETUNING;
@@ -2682,10 +2652,12 @@ int sdhci_runtime_suspend_host(struct sdhci_host *host)
 	}
 
 	spin_lock_irqsave(&host->lock, flags);
-	sdhci_mask_irqs(host, SDHCI_INT_ALL_MASK);
+	host->ier &= SDHCI_INT_CARD_INT;
+	sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
+	sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
 	spin_unlock_irqrestore(&host->lock, flags);
 
-	synchronize_irq(host->irq);
+	synchronize_hardirq(host->irq);
 
 	spin_lock_irqsave(&host->lock, flags);
 	host->runtime_suspended = true;
@@ -2729,7 +2701,7 @@ int sdhci_runtime_resume_host(struct sdhci_host *host)
 	host->runtime_suspended = false;
 
 	/* Enable SDIO IRQ */
-	if ((host->flags & SDHCI_SDIO_IRQ_ENABLED))
+	if (host->flags & SDHCI_SDIO_IRQ_ENABLED)
 		sdhci_enable_sdio_irq_nolock(host, true);
 
 	/* Enable Card Detection */
@@ -2788,7 +2760,7 @@ int sdhci_add_host(struct sdhci_host *host)
 	if (debug_quirks2)
 		host->quirks2 = debug_quirks2;
 
-	sdhci_reset(host, SDHCI_RESET_ALL);
+	sdhci_do_reset(host, SDHCI_RESET_ALL);
 
 	host->version = sdhci_readw(host, SDHCI_HOST_VERSION);
 	host->version = (host->version & SDHCI_SPEC_VER_MASK)
@@ -2848,15 +2820,29 @@ int sdhci_add_host(struct sdhci_host *host)
 		 * (128) and potentially one alignment transfer for
 		 * each of those entries.
 		 */
-		host->adma_desc = kmalloc((128 * 2 + 1) * 4, GFP_KERNEL);
+		host->adma_desc = dma_alloc_coherent(mmc_dev(host->mmc),
+						     ADMA_SIZE, &host->adma_addr,
+						     GFP_KERNEL);
 		host->align_buffer = kmalloc(128 * 4, GFP_KERNEL);
 		if (!host->adma_desc || !host->align_buffer) {
-			kfree(host->adma_desc);
+			dma_free_coherent(mmc_dev(host->mmc), ADMA_SIZE,
+					  host->adma_desc, host->adma_addr);
 			kfree(host->align_buffer);
 			pr_warning("%s: Unable to allocate ADMA "
 				"buffers. Falling back to standard DMA.\n",
 				mmc_hostname(mmc));
 			host->flags &= ~SDHCI_USE_ADMA;
+			host->adma_desc = NULL;
+			host->align_buffer = NULL;
+		} else if (host->adma_addr & 3) {
+			pr_warning("%s: unable to allocate aligned ADMA descriptor\n",
+				   mmc_hostname(mmc));
+			host->flags &= ~SDHCI_USE_ADMA;
+			dma_free_coherent(mmc_dev(host->mmc), ADMA_SIZE,
+					  host->adma_desc, host->adma_addr);
+			kfree(host->align_buffer);
+			host->adma_desc = NULL;
+			host->align_buffer = NULL;
 		}
 	}
 
@@ -2941,6 +2927,7 @@ int sdhci_add_host(struct sdhci_host *host)
 	mmc->max_busy_timeout = (1 << 27) / host->timeout_clk;
 
 	mmc->caps |= MMC_CAP_SDIO_IRQ | MMC_CAP_ERASE | MMC_CAP_CMD23;
+	mmc->caps2 |= MMC_CAP2_SDIO_IRQ_NOTHREAD;
 
 	if (host->quirks & SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12)
 		host->flags |= SDHCI_AUTO_CMD12;
@@ -3212,8 +3199,6 @@ int sdhci_add_host(struct sdhci_host *host)
 	/*
 	 * Init tasklets.
 	 */
-	tasklet_init(&host->card_tasklet,
-		sdhci_tasklet_card, (unsigned long)host);
 	tasklet_init(&host->finish_tasklet,
 		sdhci_tasklet_finish, (unsigned long)host);
 
@@ -3230,8 +3215,8 @@ int sdhci_add_host(struct sdhci_host *host)
 
 	sdhci_init(host, 0);
 
-	ret = request_irq(host->irq, sdhci_irq, IRQF_SHARED,
-		mmc_hostname(mmc), host);
+	ret = request_threaded_irq(host->irq, sdhci_irq, sdhci_thread_irq,
+				   IRQF_SHARED,	mmc_hostname(mmc), host);
 	if (ret) {
 		pr_err("%s: Failed to request IRQ %d: %d\n",
 		       mmc_hostname(mmc), host->irq, ret);
@@ -3273,12 +3258,12 @@ int sdhci_add_host(struct sdhci_host *host)
 
 #ifdef SDHCI_USE_LEDS_CLASS
 reset:
-	sdhci_reset(host, SDHCI_RESET_ALL);
-	sdhci_mask_irqs(host, SDHCI_INT_ALL_MASK);
+	sdhci_do_reset(host, SDHCI_RESET_ALL);
+	sdhci_writel(host, 0, SDHCI_INT_ENABLE);
+	sdhci_writel(host, 0, SDHCI_SIGNAL_ENABLE);
 	free_irq(host->irq, host);
 #endif
 untasklet:
-	tasklet_kill(&host->card_tasklet);
 	tasklet_kill(&host->finish_tasklet);
 
 	return ret;
@@ -3315,14 +3300,14 @@ void sdhci_remove_host(struct sdhci_host *host, int dead)
 #endif
 
 	if (!dead)
-		sdhci_reset(host, SDHCI_RESET_ALL);
+		sdhci_do_reset(host, SDHCI_RESET_ALL);
 
-	sdhci_mask_irqs(host, SDHCI_INT_ALL_MASK);
+	sdhci_writel(host, 0, SDHCI_INT_ENABLE);
+	sdhci_writel(host, 0, SDHCI_SIGNAL_ENABLE);
 	free_irq(host->irq, host);
 
 	del_timer_sync(&host->timer);
 
-	tasklet_kill(&host->card_tasklet);
 	tasklet_kill(&host->finish_tasklet);
 
 	if (host->vmmc) {
@@ -3335,7 +3320,9 @@ void sdhci_remove_host(struct sdhci_host *host, int dead)
 		regulator_put(host->vqmmc);
 	}
 
-	kfree(host->adma_desc);
+	if (host->adma_desc)
+		dma_free_coherent(mmc_dev(host->mmc), ADMA_SIZE,
+				  host->adma_desc, host->adma_addr);
 	kfree(host->align_buffer);
 
 	host->adma_desc = NULL;
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 0a3ed01887db..4a5cd5e3fa3e 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -281,18 +281,14 @@ struct sdhci_ops {
 	unsigned int	(*get_max_clock)(struct sdhci_host *host);
 	unsigned int	(*get_min_clock)(struct sdhci_host *host);
 	unsigned int	(*get_timeout_clock)(struct sdhci_host *host);
-	int		(*platform_bus_width)(struct sdhci_host *host,
-					       int width);
+	void		(*set_bus_width)(struct sdhci_host *host, int width);
 	void (*platform_send_init_74_clocks)(struct sdhci_host *host,
 					     u8 power_mode);
 	unsigned int    (*get_ro)(struct sdhci_host *host);
-	void	(*platform_reset_enter)(struct sdhci_host *host, u8 mask);
-	void	(*platform_reset_exit)(struct sdhci_host *host, u8 mask);
+	void		(*reset)(struct sdhci_host *host, u8 mask);
 	int	(*platform_execute_tuning)(struct sdhci_host *host, u32 opcode);
-	int	(*set_uhs_signaling)(struct sdhci_host *host, unsigned int uhs);
+	void	(*set_uhs_signaling)(struct sdhci_host *host, unsigned int uhs);
 	void	(*hw_reset)(struct sdhci_host *host);
-	void	(*platform_suspend)(struct sdhci_host *host);
-	void	(*platform_resume)(struct sdhci_host *host);
 	void    (*adma_workaround)(struct sdhci_host *host, u32 intmask);
 	void	(*platform_init)(struct sdhci_host *host);
 	void    (*card_event)(struct sdhci_host *host);
@@ -397,6 +393,16 @@ extern void sdhci_remove_host(struct sdhci_host *host, int dead);
 extern void sdhci_send_command(struct sdhci_host *host,
 				struct mmc_command *cmd);
 
+static inline bool sdhci_sdio_irq_enabled(struct sdhci_host *host)
+{
+	return !!(host->flags & SDHCI_SDIO_IRQ_ENABLED);
+}
+
+void sdhci_set_clock(struct sdhci_host *host, unsigned int clock);
+void sdhci_set_bus_width(struct sdhci_host *host, int width);
+void sdhci_reset(struct sdhci_host *host, u8 mask);
+void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing);
+
 #ifdef CONFIG_PM
 extern int sdhci_suspend_host(struct sdhci_host *host);
 extern int sdhci_resume_host(struct sdhci_host *host);
diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c
index 54730f4aac87..656fbba4c422 100644
--- a/drivers/mmc/host/sh_mmcif.c
+++ b/drivers/mmc/host/sh_mmcif.c
@@ -803,12 +803,13 @@ static u32 sh_mmcif_set_cmd(struct sh_mmcif_host *host,
 			break;
 		}
 		switch (host->timing) {
-		case MMC_TIMING_UHS_DDR50:
+		case MMC_TIMING_MMC_DDR52:
 			/*
 			 * MMC core will only set this timing, if the host
-			 * advertises the MMC_CAP_UHS_DDR50 capability. MMCIF
-			 * implementations with this capability, e.g. sh73a0,
-			 * will have to set it in their platform data.
+			 * advertises the MMC_CAP_1_8V_DDR/MMC_CAP_1_2V_DDR
+			 * capability. MMCIF implementations with this
+			 * capability, e.g. sh73a0, will have to set it
+			 * in their platform data.
 			 */
 			tmp |= CMD_SET_DARS;
 			break;
diff --git a/drivers/mmc/host/usdhi6rol0.c b/drivers/mmc/host/usdhi6rol0.c
new file mode 100644
index 000000000000..f0a39eb049af
--- /dev/null
+++ b/drivers/mmc/host/usdhi6rol0.c
@@ -0,0 +1,1847 @@
+/*
+ * Copyright (C) 2013-2014 Renesas Electronics Europe Ltd.
+ * Author: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmaengine.h>
+#include <linux/highmem.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/log2.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/mmc.h>
+#include <linux/mmc/sd.h>
+#include <linux/mmc/sdio.h>
+#include <linux/module.h>
+#include <linux/pagemap.h>
+#include <linux/platform_device.h>
+#include <linux/scatterlist.h>
+#include <linux/string.h>
+#include <linux/time.h>
+#include <linux/virtio.h>
+#include <linux/workqueue.h>
+
+#define USDHI6_SD_CMD		0x0000
+#define USDHI6_SD_PORT_SEL	0x0004
+#define USDHI6_SD_ARG		0x0008
+#define USDHI6_SD_STOP		0x0010
+#define USDHI6_SD_SECCNT	0x0014
+#define USDHI6_SD_RSP10		0x0018
+#define USDHI6_SD_RSP32		0x0020
+#define USDHI6_SD_RSP54		0x0028
+#define USDHI6_SD_RSP76		0x0030
+#define USDHI6_SD_INFO1		0x0038
+#define USDHI6_SD_INFO2		0x003c
+#define USDHI6_SD_INFO1_MASK	0x0040
+#define USDHI6_SD_INFO2_MASK	0x0044
+#define USDHI6_SD_CLK_CTRL	0x0048
+#define USDHI6_SD_SIZE		0x004c
+#define USDHI6_SD_OPTION	0x0050
+#define USDHI6_SD_ERR_STS1	0x0058
+#define USDHI6_SD_ERR_STS2	0x005c
+#define USDHI6_SD_BUF0		0x0060
+#define USDHI6_SDIO_MODE	0x0068
+#define USDHI6_SDIO_INFO1	0x006c
+#define USDHI6_SDIO_INFO1_MASK	0x0070
+#define USDHI6_CC_EXT_MODE	0x01b0
+#define USDHI6_SOFT_RST		0x01c0
+#define USDHI6_VERSION		0x01c4
+#define USDHI6_HOST_MODE	0x01c8
+#define USDHI6_SDIF_MODE	0x01cc
+
+#define USDHI6_SD_CMD_APP		0x0040
+#define USDHI6_SD_CMD_MODE_RSP_AUTO	0x0000
+#define USDHI6_SD_CMD_MODE_RSP_NONE	0x0300
+#define USDHI6_SD_CMD_MODE_RSP_R1	0x0400	/* Also R5, R6, R7 */
+#define USDHI6_SD_CMD_MODE_RSP_R1B	0x0500	/* R1b */
+#define USDHI6_SD_CMD_MODE_RSP_R2	0x0600
+#define USDHI6_SD_CMD_MODE_RSP_R3	0x0700	/* Also R4 */
+#define USDHI6_SD_CMD_DATA		0x0800
+#define USDHI6_SD_CMD_READ		0x1000
+#define USDHI6_SD_CMD_MULTI		0x2000
+#define USDHI6_SD_CMD_CMD12_AUTO_OFF	0x4000
+
+#define USDHI6_CC_EXT_MODE_SDRW		BIT(1)
+
+#define USDHI6_SD_INFO1_RSP_END		BIT(0)
+#define USDHI6_SD_INFO1_ACCESS_END	BIT(2)
+#define USDHI6_SD_INFO1_CARD_OUT	BIT(3)
+#define USDHI6_SD_INFO1_CARD_IN		BIT(4)
+#define USDHI6_SD_INFO1_CD		BIT(5)
+#define USDHI6_SD_INFO1_WP		BIT(7)
+#define USDHI6_SD_INFO1_D3_CARD_OUT	BIT(8)
+#define USDHI6_SD_INFO1_D3_CARD_IN	BIT(9)
+
+#define USDHI6_SD_INFO2_CMD_ERR		BIT(0)
+#define USDHI6_SD_INFO2_CRC_ERR		BIT(1)
+#define USDHI6_SD_INFO2_END_ERR		BIT(2)
+#define USDHI6_SD_INFO2_TOUT		BIT(3)
+#define USDHI6_SD_INFO2_IWA_ERR		BIT(4)
+#define USDHI6_SD_INFO2_IRA_ERR		BIT(5)
+#define USDHI6_SD_INFO2_RSP_TOUT	BIT(6)
+#define USDHI6_SD_INFO2_SDDAT0		BIT(7)
+#define USDHI6_SD_INFO2_BRE		BIT(8)
+#define USDHI6_SD_INFO2_BWE		BIT(9)
+#define USDHI6_SD_INFO2_SCLKDIVEN	BIT(13)
+#define USDHI6_SD_INFO2_CBSY		BIT(14)
+#define USDHI6_SD_INFO2_ILA		BIT(15)
+
+#define USDHI6_SD_INFO1_CARD_INSERT (USDHI6_SD_INFO1_CARD_IN | USDHI6_SD_INFO1_D3_CARD_IN)
+#define USDHI6_SD_INFO1_CARD_EJECT (USDHI6_SD_INFO1_CARD_OUT | USDHI6_SD_INFO1_D3_CARD_OUT)
+#define USDHI6_SD_INFO1_CARD (USDHI6_SD_INFO1_CARD_INSERT | USDHI6_SD_INFO1_CARD_EJECT)
+#define USDHI6_SD_INFO1_CARD_CD (USDHI6_SD_INFO1_CARD_IN | USDHI6_SD_INFO1_CARD_OUT)
+
+#define USDHI6_SD_INFO2_ERR	(USDHI6_SD_INFO2_CMD_ERR |	\
+	USDHI6_SD_INFO2_CRC_ERR | USDHI6_SD_INFO2_END_ERR |	\
+	USDHI6_SD_INFO2_TOUT | USDHI6_SD_INFO2_IWA_ERR |	\
+	USDHI6_SD_INFO2_IRA_ERR | USDHI6_SD_INFO2_RSP_TOUT |	\
+	USDHI6_SD_INFO2_ILA)
+
+#define USDHI6_SD_INFO1_IRQ	(USDHI6_SD_INFO1_RSP_END | USDHI6_SD_INFO1_ACCESS_END | \
+				 USDHI6_SD_INFO1_CARD)
+
+#define USDHI6_SD_INFO2_IRQ	(USDHI6_SD_INFO2_ERR | USDHI6_SD_INFO2_BRE | \
+				 USDHI6_SD_INFO2_BWE | 0x0800 | USDHI6_SD_INFO2_ILA)
+
+#define USDHI6_SD_CLK_CTRL_SCLKEN	BIT(8)
+
+#define USDHI6_SD_STOP_STP		BIT(0)
+#define USDHI6_SD_STOP_SEC		BIT(8)
+
+#define USDHI6_SDIO_INFO1_IOIRQ		BIT(0)
+#define USDHI6_SDIO_INFO1_EXPUB52	BIT(14)
+#define USDHI6_SDIO_INFO1_EXWT		BIT(15)
+
+#define USDHI6_SD_ERR_STS1_CRC_NO_ERROR	BIT(13)
+
+#define USDHI6_SOFT_RST_RESERVED	(BIT(1) | BIT(2))
+#define USDHI6_SOFT_RST_RESET		BIT(0)
+
+#define USDHI6_SD_OPTION_TIMEOUT_SHIFT	4
+#define USDHI6_SD_OPTION_TIMEOUT_MASK	(0xf << USDHI6_SD_OPTION_TIMEOUT_SHIFT)
+#define USDHI6_SD_OPTION_WIDTH_1	BIT(15)
+
+#define USDHI6_SD_PORT_SEL_PORTS_SHIFT	8
+
+#define USDHI6_SD_CLK_CTRL_DIV_MASK	0xff
+
+#define USDHI6_SDIO_INFO1_IRQ	(USDHI6_SDIO_INFO1_IOIRQ | 3 | \
+				 USDHI6_SDIO_INFO1_EXPUB52 | USDHI6_SDIO_INFO1_EXWT)
+
+#define USDHI6_MIN_DMA 64
+
+enum usdhi6_wait_for {
+	USDHI6_WAIT_FOR_REQUEST,
+	USDHI6_WAIT_FOR_CMD,
+	USDHI6_WAIT_FOR_MREAD,
+	USDHI6_WAIT_FOR_MWRITE,
+	USDHI6_WAIT_FOR_READ,
+	USDHI6_WAIT_FOR_WRITE,
+	USDHI6_WAIT_FOR_DATA_END,
+	USDHI6_WAIT_FOR_STOP,
+	USDHI6_WAIT_FOR_DMA,
+};
+
+struct usdhi6_page {
+	struct page *page;
+	void *mapped;		/* mapped page */
+};
+
+struct usdhi6_host {
+	struct mmc_host *mmc;
+	struct mmc_request *mrq;
+	void __iomem *base;
+	struct clk *clk;
+
+	/* SG memory handling */
+
+	/* Common for multiple and single block requests */
+	struct usdhi6_page pg;	/* current page from an SG */
+	void *blk_page;		/* either a mapped page, or the bounce buffer */
+	size_t offset;		/* offset within a page, including sg->offset */
+
+	/* Blocks, crossing a page boundary */
+	size_t head_len;
+	struct usdhi6_page head_pg;
+
+	/* A bounce buffer for unaligned blocks or blocks, crossing a page boundary */
+	struct scatterlist bounce_sg;
+	u8 bounce_buf[512];
+
+	/* Multiple block requests only */
+	struct scatterlist *sg;	/* current SG segment */
+	int page_idx;		/* page index within an SG segment */
+
+	enum usdhi6_wait_for wait;
+	u32 status_mask;
+	u32 status2_mask;
+	u32 sdio_mask;
+	u32 io_error;
+	u32 irq_status;
+	unsigned long imclk;
+	unsigned long rate;
+	bool app_cmd;
+
+	/* Timeout handling */
+	struct delayed_work timeout_work;
+	unsigned long timeout;
+
+	/* DMA support */
+	struct dma_chan *chan_rx;
+	struct dma_chan *chan_tx;
+	bool dma_active;
+};
+
+/*			I/O primitives					*/
+
+static void usdhi6_write(struct usdhi6_host *host, u32 reg, u32 data)
+{
+	iowrite32(data, host->base + reg);
+	dev_vdbg(mmc_dev(host->mmc), "%s(0x%p + 0x%x) = 0x%x\n", __func__,
+		host->base, reg, data);
+}
+
+static void usdhi6_write16(struct usdhi6_host *host, u32 reg, u16 data)
+{
+	iowrite16(data, host->base + reg);
+	dev_vdbg(mmc_dev(host->mmc), "%s(0x%p + 0x%x) = 0x%x\n", __func__,
+		host->base, reg, data);
+}
+
+static u32 usdhi6_read(struct usdhi6_host *host, u32 reg)
+{
+	u32 data = ioread32(host->base + reg);
+	dev_vdbg(mmc_dev(host->mmc), "%s(0x%p + 0x%x) = 0x%x\n", __func__,
+		host->base, reg, data);
+	return data;
+}
+
+static u16 usdhi6_read16(struct usdhi6_host *host, u32 reg)
+{
+	u16 data = ioread16(host->base + reg);
+	dev_vdbg(mmc_dev(host->mmc), "%s(0x%p + 0x%x) = 0x%x\n", __func__,
+		host->base, reg, data);
+	return data;
+}
+
+static void usdhi6_irq_enable(struct usdhi6_host *host, u32 info1, u32 info2)
+{
+	host->status_mask = USDHI6_SD_INFO1_IRQ & ~info1;
+	host->status2_mask = USDHI6_SD_INFO2_IRQ & ~info2;
+	usdhi6_write(host, USDHI6_SD_INFO1_MASK, host->status_mask);
+	usdhi6_write(host, USDHI6_SD_INFO2_MASK, host->status2_mask);
+}
+
+static void usdhi6_wait_for_resp(struct usdhi6_host *host)
+{
+	usdhi6_irq_enable(host, USDHI6_SD_INFO1_RSP_END |
+			  USDHI6_SD_INFO1_ACCESS_END | USDHI6_SD_INFO1_CARD_CD,
+			  USDHI6_SD_INFO2_ERR);
+}
+
+static void usdhi6_wait_for_brwe(struct usdhi6_host *host, bool read)
+{
+	usdhi6_irq_enable(host, USDHI6_SD_INFO1_ACCESS_END |
+			  USDHI6_SD_INFO1_CARD_CD, USDHI6_SD_INFO2_ERR |
+			  (read ? USDHI6_SD_INFO2_BRE : USDHI6_SD_INFO2_BWE));
+}
+
+static void usdhi6_only_cd(struct usdhi6_host *host)
+{
+	/* Mask all except card hotplug */
+	usdhi6_irq_enable(host, USDHI6_SD_INFO1_CARD_CD, 0);
+}
+
+static void usdhi6_mask_all(struct usdhi6_host *host)
+{
+	usdhi6_irq_enable(host, 0, 0);
+}
+
+static int usdhi6_error_code(struct usdhi6_host *host)
+{
+	u32 err;
+
+	usdhi6_write(host, USDHI6_SD_STOP, USDHI6_SD_STOP_STP);
+
+	if (host->io_error &
+	    (USDHI6_SD_INFO2_RSP_TOUT | USDHI6_SD_INFO2_TOUT)) {
+		u32 rsp54 = usdhi6_read(host, USDHI6_SD_RSP54);
+		int opc = host->mrq ? host->mrq->cmd->opcode : -1;
+
+		err = usdhi6_read(host, USDHI6_SD_ERR_STS2);
+		/* Response timeout is often normal, don't spam the log */
+		if (host->wait == USDHI6_WAIT_FOR_CMD)
+			dev_dbg(mmc_dev(host->mmc),
+				"T-out sts 0x%x, resp 0x%x, state %u, CMD%d\n",
+				err, rsp54, host->wait, opc);
+		else
+			dev_warn(mmc_dev(host->mmc),
+				 "T-out sts 0x%x, resp 0x%x, state %u, CMD%d\n",
+				 err, rsp54, host->wait, opc);
+		return -ETIMEDOUT;
+	}
+
+	err = usdhi6_read(host, USDHI6_SD_ERR_STS1);
+	if (err != USDHI6_SD_ERR_STS1_CRC_NO_ERROR)
+		dev_warn(mmc_dev(host->mmc), "Err sts 0x%x, state %u, CMD%d\n",
+			 err, host->wait, host->mrq ? host->mrq->cmd->opcode : -1);
+	if (host->io_error & USDHI6_SD_INFO2_ILA)
+		return -EILSEQ;
+
+	return -EIO;
+}
+
+/*			Scatter-Gather management			*/
+
+/*
+ * In PIO mode we have to map each page separately, using kmap(). That way
+ * adjacent pages are mapped to non-adjacent virtual addresses. That's why we
+ * have to use a bounce buffer for blocks, crossing page boundaries. Such blocks
+ * have been observed with an SDIO WiFi card (b43 driver).
+ */
+static void usdhi6_blk_bounce(struct usdhi6_host *host,
+			      struct scatterlist *sg)
+{
+	struct mmc_data *data = host->mrq->data;
+	size_t blk_head = host->head_len;
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): CMD%u of %u SG: %ux%u @ 0x%x\n",
+		__func__, host->mrq->cmd->opcode, data->sg_len,
+		data->blksz, data->blocks, sg->offset);
+
+	host->head_pg.page	= host->pg.page;
+	host->head_pg.mapped	= host->pg.mapped;
+	host->pg.page		= nth_page(host->pg.page, 1);
+	host->pg.mapped		= kmap(host->pg.page);
+
+	host->blk_page = host->bounce_buf;
+	host->offset = 0;
+
+	if (data->flags & MMC_DATA_READ)
+		return;
+
+	memcpy(host->bounce_buf, host->head_pg.mapped + PAGE_SIZE - blk_head,
+	       blk_head);
+	memcpy(host->bounce_buf + blk_head, host->pg.mapped,
+	       data->blksz - blk_head);
+}
+
+/* Only called for multiple block IO */
+static void usdhi6_sg_prep(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+	struct mmc_data *data = mrq->data;
+
+	usdhi6_write(host, USDHI6_SD_SECCNT, data->blocks);
+
+	host->sg = data->sg;
+	/* TODO: if we always map, this is redundant */
+	host->offset = host->sg->offset;
+}
+
+/* Map the first page in an SG segment: common for multiple and single block IO */
+static void *usdhi6_sg_map(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+	struct scatterlist *sg = data->sg_len > 1 ? host->sg : data->sg;
+	size_t head = PAGE_SIZE - sg->offset;
+	size_t blk_head = head % data->blksz;
+
+	WARN(host->pg.page, "%p not properly unmapped!\n", host->pg.page);
+	if (WARN(sg_dma_len(sg) % data->blksz,
+		 "SG size %u isn't a multiple of block size %u\n",
+		 sg_dma_len(sg), data->blksz))
+		return NULL;
+
+	host->pg.page = sg_page(sg);
+	host->pg.mapped = kmap(host->pg.page);
+	host->offset = sg->offset;
+
+	/*
+	 * Block size must be a power of 2 for multi-block transfers,
+	 * therefore blk_head is equal for all pages in this SG
+	 */
+	host->head_len = blk_head;
+
+	if (head < data->blksz)
+		/*
+		 * The first block in the SG crosses a page boundary.
+		 * Max blksz = 512, so blocks can only span 2 pages
+		 */
+		usdhi6_blk_bounce(host, sg);
+	else
+		host->blk_page = host->pg.mapped;
+
+	dev_dbg(mmc_dev(host->mmc), "Mapped %p (%lx) at %p + %u for CMD%u @ 0x%p\n",
+		host->pg.page, page_to_pfn(host->pg.page), host->pg.mapped,
+		sg->offset, host->mrq->cmd->opcode, host->mrq);
+
+	return host->blk_page + host->offset;
+}
+
+/* Unmap the current page: common for multiple and single block IO */
+static void usdhi6_sg_unmap(struct usdhi6_host *host, bool force)
+{
+	struct mmc_data *data = host->mrq->data;
+	struct page *page = host->head_pg.page;
+
+	if (page) {
+		/* Previous block was cross-page boundary */
+		struct scatterlist *sg = data->sg_len > 1 ?
+			host->sg : data->sg;
+		size_t blk_head = host->head_len;
+
+		if (!data->error && data->flags & MMC_DATA_READ) {
+			memcpy(host->head_pg.mapped + PAGE_SIZE - blk_head,
+			       host->bounce_buf, blk_head);
+			memcpy(host->pg.mapped, host->bounce_buf + blk_head,
+			       data->blksz - blk_head);
+		}
+
+		flush_dcache_page(page);
+		kunmap(page);
+
+		host->head_pg.page = NULL;
+
+		if (!force && sg_dma_len(sg) + sg->offset >
+		    (host->page_idx << PAGE_SHIFT) + data->blksz - blk_head)
+			/* More blocks in this SG, don't unmap the next page */
+			return;
+	}
+
+	page = host->pg.page;
+	if (!page)
+		return;
+
+	flush_dcache_page(page);
+	kunmap(page);
+
+	host->pg.page = NULL;
+}
+
+/* Called from MMC_WRITE_MULTIPLE_BLOCK or MMC_READ_MULTIPLE_BLOCK */
+static void usdhi6_sg_advance(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+	size_t done, total;
+
+	/* New offset: set at the end of the previous block */
+	if (host->head_pg.page) {
+		/* Finished a cross-page block, jump to the new page */
+		host->page_idx++;
+		host->offset = data->blksz - host->head_len;
+		host->blk_page = host->pg.mapped;
+		usdhi6_sg_unmap(host, false);
+	} else {
+		host->offset += data->blksz;
+		/* The completed block didn't cross a page boundary */
+		if (host->offset == PAGE_SIZE) {
+			/* If required, we'll map the page below */
+			host->offset = 0;
+			host->page_idx++;
+		}
+	}
+
+	/*
+	 * Now host->blk_page + host->offset point at the end of our last block
+	 * and host->page_idx is the index of the page, in which our new block
+	 * is located, if any
+	 */
+
+	done = (host->page_idx << PAGE_SHIFT) + host->offset;
+	total = host->sg->offset + sg_dma_len(host->sg);
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): %zu of %zu @ %zu\n", __func__,
+		done, total, host->offset);
+
+	if (done < total && host->offset) {
+		/* More blocks in this page */
+		if (host->offset + data->blksz > PAGE_SIZE)
+			/* We approached at a block, that spans 2 pages */
+			usdhi6_blk_bounce(host, host->sg);
+
+		return;
+	}
+
+	/* Finished current page or an SG segment */
+	usdhi6_sg_unmap(host, false);
+
+	if (done == total) {
+		/*
+		 * End of an SG segment or the complete SG: jump to the next
+		 * segment, we'll map it later in usdhi6_blk_read() or
+		 * usdhi6_blk_write()
+		 */
+		struct scatterlist *next = sg_next(host->sg);
+
+		host->page_idx = 0;
+
+		if (!next)
+			host->wait = USDHI6_WAIT_FOR_DATA_END;
+		host->sg = next;
+
+		if (WARN(next && sg_dma_len(next) % data->blksz,
+			 "SG size %u isn't a multiple of block size %u\n",
+			 sg_dma_len(next), data->blksz))
+			data->error = -EINVAL;
+
+		return;
+	}
+
+	/* We cannot get here after crossing a page border */
+
+	/* Next page in the same SG */
+	host->pg.page = nth_page(sg_page(host->sg), host->page_idx);
+	host->pg.mapped = kmap(host->pg.page);
+	host->blk_page = host->pg.mapped;
+
+	dev_dbg(mmc_dev(host->mmc), "Mapped %p (%lx) at %p for CMD%u @ 0x%p\n",
+		host->pg.page, page_to_pfn(host->pg.page), host->pg.mapped,
+		host->mrq->cmd->opcode, host->mrq);
+}
+
+/*			DMA handling					*/
+
+static void usdhi6_dma_release(struct usdhi6_host *host)
+{
+	host->dma_active = false;
+	if (host->chan_tx) {
+		struct dma_chan *chan = host->chan_tx;
+		host->chan_tx = NULL;
+		dma_release_channel(chan);
+	}
+	if (host->chan_rx) {
+		struct dma_chan *chan = host->chan_rx;
+		host->chan_rx = NULL;
+		dma_release_channel(chan);
+	}
+}
+
+static void usdhi6_dma_stop_unmap(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+
+	if (!host->dma_active)
+		return;
+
+	usdhi6_write(host, USDHI6_CC_EXT_MODE, 0);
+	host->dma_active = false;
+
+	if (data->flags & MMC_DATA_READ)
+		dma_unmap_sg(host->chan_rx->device->dev, data->sg,
+			     data->sg_len, DMA_FROM_DEVICE);
+	else
+		dma_unmap_sg(host->chan_tx->device->dev, data->sg,
+			     data->sg_len, DMA_TO_DEVICE);
+}
+
+static void usdhi6_dma_complete(void *arg)
+{
+	struct usdhi6_host *host = arg;
+	struct mmc_request *mrq = host->mrq;
+
+	if (WARN(!mrq || !mrq->data, "%s: NULL data in DMA completion for %p!\n",
+		 dev_name(mmc_dev(host->mmc)), mrq))
+		return;
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): CMD%u DMA completed\n", __func__,
+		mrq->cmd->opcode);
+
+	usdhi6_dma_stop_unmap(host);
+	usdhi6_wait_for_brwe(host, mrq->data->flags & MMC_DATA_READ);
+}
+
+static int usdhi6_dma_setup(struct usdhi6_host *host, struct dma_chan *chan,
+			    enum dma_transfer_direction dir)
+{
+	struct mmc_data *data = host->mrq->data;
+	struct scatterlist *sg = data->sg;
+	struct dma_async_tx_descriptor *desc = NULL;
+	dma_cookie_t cookie = -EINVAL;
+	enum dma_data_direction data_dir;
+	int ret;
+
+	switch (dir) {
+	case DMA_MEM_TO_DEV:
+		data_dir = DMA_TO_DEVICE;
+		break;
+	case DMA_DEV_TO_MEM:
+		data_dir = DMA_FROM_DEVICE;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ret = dma_map_sg(chan->device->dev, sg, data->sg_len, data_dir);
+	if (ret > 0) {
+		host->dma_active = true;
+		desc = dmaengine_prep_slave_sg(chan, sg, ret, dir,
+					DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	}
+
+	if (desc) {
+		desc->callback = usdhi6_dma_complete;
+		desc->callback_param = host;
+		cookie = dmaengine_submit(desc);
+	}
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): mapped %d -> %d, cookie %d @ %p\n",
+		__func__, data->sg_len, ret, cookie, desc);
+
+	if (cookie < 0) {
+		/* DMA failed, fall back to PIO */
+		if (ret >= 0)
+			ret = cookie;
+		usdhi6_dma_release(host);
+		dev_warn(mmc_dev(host->mmc),
+			 "DMA failed: %d, falling back to PIO\n", ret);
+	}
+
+	return cookie;
+}
+
+static int usdhi6_dma_start(struct usdhi6_host *host)
+{
+	if (!host->chan_rx || !host->chan_tx)
+		return -ENODEV;
+
+	if (host->mrq->data->flags & MMC_DATA_READ)
+		return usdhi6_dma_setup(host, host->chan_rx, DMA_DEV_TO_MEM);
+
+	return usdhi6_dma_setup(host, host->chan_tx, DMA_MEM_TO_DEV);
+}
+
+static void usdhi6_dma_kill(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): SG of %u: %ux%u\n",
+		__func__, data->sg_len, data->blocks, data->blksz);
+	/* Abort DMA */
+	if (data->flags & MMC_DATA_READ)
+		dmaengine_terminate_all(host->chan_rx);
+	else
+		dmaengine_terminate_all(host->chan_tx);
+}
+
+static void usdhi6_dma_check_error(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): IO error %d, status 0x%x\n",
+		__func__, host->io_error, usdhi6_read(host, USDHI6_SD_INFO1));
+
+	if (host->io_error) {
+		data->error = usdhi6_error_code(host);
+		data->bytes_xfered = 0;
+		usdhi6_dma_kill(host);
+		usdhi6_dma_release(host);
+		dev_warn(mmc_dev(host->mmc),
+			 "DMA failed: %d, falling back to PIO\n", data->error);
+		return;
+	}
+
+	/*
+	 * The datasheet tells us to check a response from the card, whereas
+	 * responses only come after the command phase, not after the data
+	 * phase. Let's check anyway.
+	 */
+	if (host->irq_status & USDHI6_SD_INFO1_RSP_END)
+		dev_warn(mmc_dev(host->mmc), "Unexpected response received!\n");
+}
+
+static void usdhi6_dma_kick(struct usdhi6_host *host)
+{
+	if (host->mrq->data->flags & MMC_DATA_READ)
+		dma_async_issue_pending(host->chan_rx);
+	else
+		dma_async_issue_pending(host->chan_tx);
+}
+
+static void usdhi6_dma_request(struct usdhi6_host *host, phys_addr_t start)
+{
+	struct dma_slave_config cfg = {
+		.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES,
+		.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES,
+	};
+	int ret;
+
+	host->chan_tx = dma_request_slave_channel(mmc_dev(host->mmc), "tx");
+	dev_dbg(mmc_dev(host->mmc), "%s: TX: got channel %p\n", __func__,
+		host->chan_tx);
+
+	if (!host->chan_tx)
+		return;
+
+	cfg.direction = DMA_MEM_TO_DEV;
+	cfg.dst_addr = start + USDHI6_SD_BUF0;
+	cfg.dst_maxburst = 128;	/* 128 words * 4 bytes = 512 bytes */
+	cfg.src_addr = 0;
+	ret = dmaengine_slave_config(host->chan_tx, &cfg);
+	if (ret < 0)
+		goto e_release_tx;
+
+	host->chan_rx = dma_request_slave_channel(mmc_dev(host->mmc), "rx");
+	dev_dbg(mmc_dev(host->mmc), "%s: RX: got channel %p\n", __func__,
+		host->chan_rx);
+
+	if (!host->chan_rx)
+		goto e_release_tx;
+
+	cfg.direction = DMA_DEV_TO_MEM;
+	cfg.src_addr = cfg.dst_addr;
+	cfg.src_maxburst = 128;	/* 128 words * 4 bytes = 512 bytes */
+	cfg.dst_addr = 0;
+	ret = dmaengine_slave_config(host->chan_rx, &cfg);
+	if (ret < 0)
+		goto e_release_rx;
+
+	return;
+
+e_release_rx:
+	dma_release_channel(host->chan_rx);
+	host->chan_rx = NULL;
+e_release_tx:
+	dma_release_channel(host->chan_tx);
+	host->chan_tx = NULL;
+}
+
+/*			API helpers					*/
+
+static void usdhi6_clk_set(struct usdhi6_host *host, struct mmc_ios *ios)
+{
+	unsigned long rate = ios->clock;
+	u32 val;
+	unsigned int i;
+
+	for (i = 1000; i; i--) {
+		if (usdhi6_read(host, USDHI6_SD_INFO2) & USDHI6_SD_INFO2_SCLKDIVEN)
+			break;
+		usleep_range(10, 100);
+	}
+
+	if (!i) {
+		dev_err(mmc_dev(host->mmc), "SD bus busy, clock set aborted\n");
+		return;
+	}
+
+	val = usdhi6_read(host, USDHI6_SD_CLK_CTRL) & ~USDHI6_SD_CLK_CTRL_DIV_MASK;
+
+	if (rate) {
+		unsigned long new_rate;
+
+		if (host->imclk <= rate) {
+			if (ios->timing != MMC_TIMING_UHS_DDR50) {
+				/* Cannot have 1-to-1 clock in DDR mode */
+				new_rate = host->imclk;
+				val |= 0xff;
+			} else {
+				new_rate = host->imclk / 2;
+			}
+		} else {
+			unsigned long div =
+				roundup_pow_of_two(DIV_ROUND_UP(host->imclk, rate));
+			val |= div >> 2;
+			new_rate = host->imclk / div;
+		}
+
+		if (host->rate == new_rate)
+			return;
+
+		host->rate = new_rate;
+
+		dev_dbg(mmc_dev(host->mmc), "target %lu, div %u, set %lu\n",
+			rate, (val & 0xff) << 2, new_rate);
+	}
+
+	/*
+	 * if old or new rate is equal to input rate, have to switch the clock
+	 * off before changing and on after
+	 */
+	if (host->imclk == rate || host->imclk == host->rate || !rate)
+		usdhi6_write(host, USDHI6_SD_CLK_CTRL,
+			     val & ~USDHI6_SD_CLK_CTRL_SCLKEN);
+
+	if (!rate) {
+		host->rate = 0;
+		return;
+	}
+
+	usdhi6_write(host, USDHI6_SD_CLK_CTRL, val);
+
+	if (host->imclk == rate || host->imclk == host->rate ||
+	    !(val & USDHI6_SD_CLK_CTRL_SCLKEN))
+		usdhi6_write(host, USDHI6_SD_CLK_CTRL,
+			     val | USDHI6_SD_CLK_CTRL_SCLKEN);
+}
+
+static void usdhi6_set_power(struct usdhi6_host *host, struct mmc_ios *ios)
+{
+	struct mmc_host *mmc = host->mmc;
+
+	if (!IS_ERR(mmc->supply.vmmc))
+		/* Errors ignored... */
+		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc,
+				      ios->power_mode ? ios->vdd : 0);
+}
+
+static int usdhi6_reset(struct usdhi6_host *host)
+{
+	int i;
+
+	usdhi6_write(host, USDHI6_SOFT_RST, USDHI6_SOFT_RST_RESERVED);
+	cpu_relax();
+	usdhi6_write(host, USDHI6_SOFT_RST, USDHI6_SOFT_RST_RESERVED | USDHI6_SOFT_RST_RESET);
+	for (i = 1000; i; i--)
+		if (usdhi6_read(host, USDHI6_SOFT_RST) & USDHI6_SOFT_RST_RESET)
+			break;
+
+	return i ? 0 : -ETIMEDOUT;
+}
+
+static void usdhi6_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
+{
+	struct usdhi6_host *host = mmc_priv(mmc);
+	u32 option, mode;
+	int ret;
+
+	dev_dbg(mmc_dev(mmc), "%uHz, OCR: %u, power %u, bus-width %u, timing %u\n",
+		ios->clock, ios->vdd, ios->power_mode, ios->bus_width, ios->timing);
+
+	switch (ios->power_mode) {
+	case MMC_POWER_OFF:
+		usdhi6_set_power(host, ios);
+		usdhi6_only_cd(host);
+		break;
+	case MMC_POWER_UP:
+		/*
+		 * We only also touch USDHI6_SD_OPTION from .request(), which
+		 * cannot race with MMC_POWER_UP
+		 */
+		ret = usdhi6_reset(host);
+		if (ret < 0) {
+			dev_err(mmc_dev(mmc), "Cannot reset the interface!\n");
+		} else {
+			usdhi6_set_power(host, ios);
+			usdhi6_only_cd(host);
+		}
+		break;
+	case MMC_POWER_ON:
+		option = usdhi6_read(host, USDHI6_SD_OPTION);
+		/*
+		 * The eMMC standard only allows 4 or 8 bits in the DDR mode,
+		 * the same probably holds for SD cards. We check here anyway,
+		 * since the datasheet explicitly requires 4 bits for DDR.
+		 */
+		if (ios->bus_width == MMC_BUS_WIDTH_1) {
+			if (ios->timing == MMC_TIMING_UHS_DDR50)
+				dev_err(mmc_dev(mmc),
+					"4 bits are required for DDR\n");
+			option |= USDHI6_SD_OPTION_WIDTH_1;
+			mode = 0;
+		} else {
+			option &= ~USDHI6_SD_OPTION_WIDTH_1;
+			mode = ios->timing == MMC_TIMING_UHS_DDR50;
+		}
+		usdhi6_write(host, USDHI6_SD_OPTION, option);
+		usdhi6_write(host, USDHI6_SDIF_MODE, mode);
+		break;
+	}
+
+	if (host->rate != ios->clock)
+		usdhi6_clk_set(host, ios);
+}
+
+/* This is data timeout. Response timeout is fixed to 640 clock cycles */
+static void usdhi6_timeout_set(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+	u32 val;
+	unsigned long ticks;
+
+	if (!mrq->data)
+		ticks = host->rate / 1000 * mrq->cmd->busy_timeout;
+	else
+		ticks = host->rate / 1000000 * (mrq->data->timeout_ns / 1000) +
+			mrq->data->timeout_clks;
+
+	if (!ticks || ticks > 1 << 27)
+		/* Max timeout */
+		val = 14;
+	else if (ticks < 1 << 13)
+		/* Min timeout */
+		val = 0;
+	else
+		val = order_base_2(ticks) - 13;
+
+	dev_dbg(mmc_dev(host->mmc), "Set %s timeout %lu ticks @ %lu Hz\n",
+		mrq->data ? "data" : "cmd", ticks, host->rate);
+
+	/* Timeout Counter mask: 0xf0 */
+	usdhi6_write(host, USDHI6_SD_OPTION, (val << USDHI6_SD_OPTION_TIMEOUT_SHIFT) |
+		     (usdhi6_read(host, USDHI6_SD_OPTION) & ~USDHI6_SD_OPTION_TIMEOUT_MASK));
+}
+
+static void usdhi6_request_done(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+	struct mmc_data *data = mrq->data;
+
+	if (WARN(host->pg.page || host->head_pg.page,
+		 "Page %p or %p not unmapped: wait %u, CMD%d(%c) @ +0x%zx %ux%u in SG%u!\n",
+		 host->pg.page, host->head_pg.page, host->wait, mrq->cmd->opcode,
+		 data ? (data->flags & MMC_DATA_READ ? 'R' : 'W') : '-',
+		 data ? host->offset : 0, data ? data->blocks : 0,
+		 data ? data->blksz : 0, data ? data->sg_len : 0))
+		usdhi6_sg_unmap(host, true);
+
+	if (mrq->cmd->error ||
+	    (data && data->error) ||
+	    (mrq->stop && mrq->stop->error))
+		dev_dbg(mmc_dev(host->mmc), "%s(CMD%d: %ux%u): err %d %d %d\n",
+			__func__, mrq->cmd->opcode, data ? data->blocks : 0,
+			data ? data->blksz : 0,
+			mrq->cmd->error,
+			data ? data->error : 1,
+			mrq->stop ? mrq->stop->error : 1);
+
+	/* Disable DMA */
+	usdhi6_write(host, USDHI6_CC_EXT_MODE, 0);
+	host->wait = USDHI6_WAIT_FOR_REQUEST;
+	host->mrq = NULL;
+
+	mmc_request_done(host->mmc, mrq);
+}
+
+static int usdhi6_cmd_flags(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+	struct mmc_command *cmd = mrq->cmd;
+	u16 opc = cmd->opcode;
+
+	if (host->app_cmd) {
+		host->app_cmd = false;
+		opc |= USDHI6_SD_CMD_APP;
+	}
+
+	if (mrq->data) {
+		opc |= USDHI6_SD_CMD_DATA;
+
+		if (mrq->data->flags & MMC_DATA_READ)
+			opc |= USDHI6_SD_CMD_READ;
+
+		if (cmd->opcode == MMC_READ_MULTIPLE_BLOCK ||
+		    cmd->opcode == MMC_WRITE_MULTIPLE_BLOCK ||
+		    (cmd->opcode == SD_IO_RW_EXTENDED &&
+		     mrq->data->blocks > 1)) {
+			opc |= USDHI6_SD_CMD_MULTI;
+			if (!mrq->stop)
+				opc |= USDHI6_SD_CMD_CMD12_AUTO_OFF;
+		}
+
+		switch (mmc_resp_type(cmd)) {
+		case MMC_RSP_NONE:
+			opc |= USDHI6_SD_CMD_MODE_RSP_NONE;
+			break;
+		case MMC_RSP_R1:
+			opc |= USDHI6_SD_CMD_MODE_RSP_R1;
+			break;
+		case MMC_RSP_R1B:
+			opc |= USDHI6_SD_CMD_MODE_RSP_R1B;
+			break;
+		case MMC_RSP_R2:
+			opc |= USDHI6_SD_CMD_MODE_RSP_R2;
+			break;
+		case MMC_RSP_R3:
+			opc |= USDHI6_SD_CMD_MODE_RSP_R3;
+			break;
+		default:
+			dev_warn(mmc_dev(host->mmc),
+				 "Unknown response type %d\n",
+				 mmc_resp_type(cmd));
+			return -EINVAL;
+		}
+	}
+
+	return opc;
+}
+
+static int usdhi6_rq_start(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+	struct mmc_command *cmd = mrq->cmd;
+	struct mmc_data *data = mrq->data;
+	int opc = usdhi6_cmd_flags(host);
+	int i;
+
+	if (opc < 0)
+		return opc;
+
+	for (i = 1000; i; i--) {
+		if (!(usdhi6_read(host, USDHI6_SD_INFO2) & USDHI6_SD_INFO2_CBSY))
+			break;
+		usleep_range(10, 100);
+	}
+
+	if (!i) {
+		dev_dbg(mmc_dev(host->mmc), "Command active, request aborted\n");
+		return -EAGAIN;
+	}
+
+	if (data) {
+		bool use_dma;
+		int ret = 0;
+
+		host->page_idx = 0;
+
+		if (cmd->opcode == SD_IO_RW_EXTENDED && data->blocks > 1) {
+			switch (data->blksz) {
+			case 512:
+				break;
+			case 32:
+			case 64:
+			case 128:
+			case 256:
+				if (mrq->stop)
+					ret = -EINVAL;
+				break;
+			default:
+				ret = -EINVAL;
+			}
+		} else if ((cmd->opcode == MMC_READ_MULTIPLE_BLOCK ||
+			    cmd->opcode == MMC_WRITE_MULTIPLE_BLOCK) &&
+			   data->blksz != 512) {
+			ret = -EINVAL;
+		}
+
+		if (ret < 0) {
+			dev_warn(mmc_dev(host->mmc), "%s(): %u blocks of %u bytes\n",
+				 __func__, data->blocks, data->blksz);
+			return -EINVAL;
+		}
+
+		if (cmd->opcode == MMC_READ_MULTIPLE_BLOCK ||
+		    cmd->opcode == MMC_WRITE_MULTIPLE_BLOCK ||
+		    (cmd->opcode == SD_IO_RW_EXTENDED &&
+		     data->blocks > 1))
+			usdhi6_sg_prep(host);
+
+		usdhi6_write(host, USDHI6_SD_SIZE, data->blksz);
+
+		if ((data->blksz >= USDHI6_MIN_DMA ||
+		     data->blocks > 1) &&
+		    (data->blksz % 4 ||
+		     data->sg->offset % 4))
+			dev_dbg(mmc_dev(host->mmc),
+				"Bad SG of %u: %ux%u @ %u\n", data->sg_len,
+				data->blksz, data->blocks, data->sg->offset);
+
+		/* Enable DMA for USDHI6_MIN_DMA bytes or more */
+		use_dma = data->blksz >= USDHI6_MIN_DMA &&
+			!(data->blksz % 4) &&
+			usdhi6_dma_start(host) >= DMA_MIN_COOKIE;
+
+		if (use_dma)
+			usdhi6_write(host, USDHI6_CC_EXT_MODE, USDHI6_CC_EXT_MODE_SDRW);
+
+		dev_dbg(mmc_dev(host->mmc),
+			"%s(): request opcode %u, %u blocks of %u bytes in %u segments, %s %s @+0x%x%s\n",
+			__func__, cmd->opcode, data->blocks, data->blksz,
+			data->sg_len, use_dma ? "DMA" : "PIO",
+			data->flags & MMC_DATA_READ ? "read" : "write",
+			data->sg->offset, mrq->stop ? " + stop" : "");
+	} else {
+		dev_dbg(mmc_dev(host->mmc), "%s(): request opcode %u\n",
+			__func__, cmd->opcode);
+	}
+
+	/* We have to get a command completion interrupt with DMA too */
+	usdhi6_wait_for_resp(host);
+
+	host->wait = USDHI6_WAIT_FOR_CMD;
+	schedule_delayed_work(&host->timeout_work, host->timeout);
+
+	/* SEC bit is required to enable block counting by the core */
+	usdhi6_write(host, USDHI6_SD_STOP,
+		     data && data->blocks > 1 ? USDHI6_SD_STOP_SEC : 0);
+	usdhi6_write(host, USDHI6_SD_ARG, cmd->arg);
+
+	/* Kick command execution */
+	usdhi6_write(host, USDHI6_SD_CMD, opc);
+
+	return 0;
+}
+
+static void usdhi6_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct usdhi6_host *host = mmc_priv(mmc);
+	int ret;
+
+	cancel_delayed_work_sync(&host->timeout_work);
+
+	host->mrq = mrq;
+	host->sg = NULL;
+
+	usdhi6_timeout_set(host);
+	ret = usdhi6_rq_start(host);
+	if (ret < 0) {
+		mrq->cmd->error = ret;
+		usdhi6_request_done(host);
+	}
+}
+
+static int usdhi6_get_cd(struct mmc_host *mmc)
+{
+	struct usdhi6_host *host = mmc_priv(mmc);
+	/* Read is atomic, no need to lock */
+	u32 status = usdhi6_read(host, USDHI6_SD_INFO1) & USDHI6_SD_INFO1_CD;
+
+/*
+ *	level	status.CD	CD_ACTIVE_HIGH	card present
+ *	1	0		0		0
+ *	1	0		1		1
+ *	0	1		0		1
+ *	0	1		1		0
+ */
+	return !status ^ !(mmc->caps2 & MMC_CAP2_CD_ACTIVE_HIGH);
+}
+
+static int usdhi6_get_ro(struct mmc_host *mmc)
+{
+	struct usdhi6_host *host = mmc_priv(mmc);
+	/* No locking as above */
+	u32 status = usdhi6_read(host, USDHI6_SD_INFO1) & USDHI6_SD_INFO1_WP;
+
+/*
+ *	level	status.WP	RO_ACTIVE_HIGH	card read-only
+ *	1	0		0		0
+ *	1	0		1		1
+ *	0	1		0		1
+ *	0	1		1		0
+ */
+	return !status ^ !(mmc->caps2 & MMC_CAP2_RO_ACTIVE_HIGH);
+}
+
+static void usdhi6_enable_sdio_irq(struct mmc_host *mmc, int enable)
+{
+	struct usdhi6_host *host = mmc_priv(mmc);
+
+	dev_dbg(mmc_dev(mmc), "%s(): %sable\n", __func__, enable ? "en" : "dis");
+
+	if (enable) {
+		host->sdio_mask = USDHI6_SDIO_INFO1_IRQ & ~USDHI6_SDIO_INFO1_IOIRQ;
+		usdhi6_write(host, USDHI6_SDIO_INFO1_MASK, host->sdio_mask);
+		usdhi6_write(host, USDHI6_SDIO_MODE, 1);
+	} else {
+		usdhi6_write(host, USDHI6_SDIO_MODE, 0);
+		usdhi6_write(host, USDHI6_SDIO_INFO1_MASK, USDHI6_SDIO_INFO1_IRQ);
+		host->sdio_mask = USDHI6_SDIO_INFO1_IRQ;
+	}
+}
+
+static struct mmc_host_ops usdhi6_ops = {
+	.request	= usdhi6_request,
+	.set_ios	= usdhi6_set_ios,
+	.get_cd		= usdhi6_get_cd,
+	.get_ro		= usdhi6_get_ro,
+	.enable_sdio_irq = usdhi6_enable_sdio_irq,
+};
+
+/*			State machine handlers				*/
+
+static void usdhi6_resp_cmd12(struct usdhi6_host *host)
+{
+	struct mmc_command *cmd = host->mrq->stop;
+	cmd->resp[0] = usdhi6_read(host, USDHI6_SD_RSP10);
+}
+
+static void usdhi6_resp_read(struct usdhi6_host *host)
+{
+	struct mmc_command *cmd = host->mrq->cmd;
+	u32 *rsp = cmd->resp, tmp = 0;
+	int i;
+
+/*
+ * RSP10	39-8
+ * RSP32	71-40
+ * RSP54	103-72
+ * RSP76	127-104
+ * R2-type response:
+ * resp[0]	= r[127..96]
+ * resp[1]	= r[95..64]
+ * resp[2]	= r[63..32]
+ * resp[3]	= r[31..0]
+ * Other responses:
+ * resp[0]	= r[39..8]
+ */
+
+	if (mmc_resp_type(cmd) == MMC_RSP_NONE)
+		return;
+
+	if (!(host->irq_status & USDHI6_SD_INFO1_RSP_END)) {
+		dev_err(mmc_dev(host->mmc),
+			"CMD%d: response expected but is missing!\n", cmd->opcode);
+		return;
+	}
+
+	if (mmc_resp_type(cmd) & MMC_RSP_136)
+		for (i = 0; i < 4; i++) {
+			if (i)
+				rsp[3 - i] = tmp >> 24;
+			tmp = usdhi6_read(host, USDHI6_SD_RSP10 + i * 8);
+			rsp[3 - i] |= tmp << 8;
+		}
+	else if (cmd->opcode == MMC_READ_MULTIPLE_BLOCK ||
+		 cmd->opcode == MMC_WRITE_MULTIPLE_BLOCK)
+		/* Read RSP54 to avoid conflict with auto CMD12 */
+		rsp[0] = usdhi6_read(host, USDHI6_SD_RSP54);
+	else
+		rsp[0] = usdhi6_read(host, USDHI6_SD_RSP10);
+
+	dev_dbg(mmc_dev(host->mmc), "Response 0x%x\n", rsp[0]);
+}
+
+static int usdhi6_blk_read(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+	u32 *p;
+	int i, rest;
+
+	if (host->io_error) {
+		data->error = usdhi6_error_code(host);
+		goto error;
+	}
+
+	if (host->pg.page) {
+		p = host->blk_page + host->offset;
+	} else {
+		p = usdhi6_sg_map(host);
+		if (!p) {
+			data->error = -ENOMEM;
+			goto error;
+		}
+	}
+
+	for (i = 0; i < data->blksz / 4; i++, p++)
+		*p = usdhi6_read(host, USDHI6_SD_BUF0);
+
+	rest = data->blksz % 4;
+	for (i = 0; i < (rest + 1) / 2; i++) {
+		u16 d = usdhi6_read16(host, USDHI6_SD_BUF0);
+		((u8 *)p)[2 * i] = ((u8 *)&d)[0];
+		if (rest > 1 && !i)
+			((u8 *)p)[2 * i + 1] = ((u8 *)&d)[1];
+	}
+
+	return 0;
+
+error:
+	dev_dbg(mmc_dev(host->mmc), "%s(): %d\n", __func__, data->error);
+	host->wait = USDHI6_WAIT_FOR_REQUEST;
+	return data->error;
+}
+
+static int usdhi6_blk_write(struct usdhi6_host *host)
+{
+	struct mmc_data *data = host->mrq->data;
+	u32 *p;
+	int i, rest;
+
+	if (host->io_error) {
+		data->error = usdhi6_error_code(host);
+		goto error;
+	}
+
+	if (host->pg.page) {
+		p = host->blk_page + host->offset;
+	} else {
+		p = usdhi6_sg_map(host);
+		if (!p) {
+			data->error = -ENOMEM;
+			goto error;
+		}
+	}
+
+	for (i = 0; i < data->blksz / 4; i++, p++)
+		usdhi6_write(host, USDHI6_SD_BUF0, *p);
+
+	rest = data->blksz % 4;
+	for (i = 0; i < (rest + 1) / 2; i++) {
+		u16 d;
+		((u8 *)&d)[0] = ((u8 *)p)[2 * i];
+		if (rest > 1 && !i)
+			((u8 *)&d)[1] = ((u8 *)p)[2 * i + 1];
+		else
+			((u8 *)&d)[1] = 0;
+		usdhi6_write16(host, USDHI6_SD_BUF0, d);
+	}
+
+	return 0;
+
+error:
+	dev_dbg(mmc_dev(host->mmc), "%s(): %d\n", __func__, data->error);
+	host->wait = USDHI6_WAIT_FOR_REQUEST;
+	return data->error;
+}
+
+static int usdhi6_stop_cmd(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+
+	switch (mrq->cmd->opcode) {
+	case MMC_READ_MULTIPLE_BLOCK:
+	case MMC_WRITE_MULTIPLE_BLOCK:
+		if (mrq->stop->opcode == MMC_STOP_TRANSMISSION) {
+			host->wait = USDHI6_WAIT_FOR_STOP;
+			return 0;
+		}
+		/* Unsupported STOP command */
+	default:
+		dev_err(mmc_dev(host->mmc),
+			"unsupported stop CMD%d for CMD%d\n",
+			mrq->stop->opcode, mrq->cmd->opcode);
+		mrq->stop->error = -EOPNOTSUPP;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static bool usdhi6_end_cmd(struct usdhi6_host *host)
+{
+	struct mmc_request *mrq = host->mrq;
+	struct mmc_command *cmd = mrq->cmd;
+
+	if (host->io_error) {
+		cmd->error = usdhi6_error_code(host);
+		return false;
+	}
+
+	usdhi6_resp_read(host);
+
+	if (!mrq->data)
+		return false;
+
+	if (host->dma_active) {
+		usdhi6_dma_kick(host);
+		if (!mrq->stop)
+			host->wait = USDHI6_WAIT_FOR_DMA;
+		else if (usdhi6_stop_cmd(host) < 0)
+			return false;
+	} else if (mrq->data->flags & MMC_DATA_READ) {
+		if (cmd->opcode == MMC_READ_MULTIPLE_BLOCK ||
+		    (cmd->opcode == SD_IO_RW_EXTENDED &&
+		     mrq->data->blocks > 1))
+			host->wait = USDHI6_WAIT_FOR_MREAD;
+		else
+			host->wait = USDHI6_WAIT_FOR_READ;
+	} else {
+		if (cmd->opcode == MMC_WRITE_MULTIPLE_BLOCK ||
+		    (cmd->opcode == SD_IO_RW_EXTENDED &&
+		     mrq->data->blocks > 1))
+			host->wait = USDHI6_WAIT_FOR_MWRITE;
+		else
+			host->wait = USDHI6_WAIT_FOR_WRITE;
+	}
+
+	return true;
+}
+
+static bool usdhi6_read_block(struct usdhi6_host *host)
+{
+	/* ACCESS_END IRQ is already unmasked */
+	int ret = usdhi6_blk_read(host);
+
+	/*
+	 * Have to force unmapping both pages: the single block could have been
+	 * cross-page, in which case for single-block IO host->page_idx == 0.
+	 * So, if we don't force, the second page won't be unmapped.
+	 */
+	usdhi6_sg_unmap(host, true);
+
+	if (ret < 0)
+		return false;
+
+	host->wait = USDHI6_WAIT_FOR_DATA_END;
+	return true;
+}
+
+static bool usdhi6_mread_block(struct usdhi6_host *host)
+{
+	int ret = usdhi6_blk_read(host);
+
+	if (ret < 0)
+		return false;
+
+	usdhi6_sg_advance(host);
+
+	return !host->mrq->data->error &&
+		(host->wait != USDHI6_WAIT_FOR_DATA_END || !host->mrq->stop);
+}
+
+static bool usdhi6_write_block(struct usdhi6_host *host)
+{
+	int ret = usdhi6_blk_write(host);
+
+	/* See comment in usdhi6_read_block() */
+	usdhi6_sg_unmap(host, true);
+
+	if (ret < 0)
+		return false;
+
+	host->wait = USDHI6_WAIT_FOR_DATA_END;
+	return true;
+}
+
+static bool usdhi6_mwrite_block(struct usdhi6_host *host)
+{
+	int ret = usdhi6_blk_write(host);
+
+	if (ret < 0)
+		return false;
+
+	usdhi6_sg_advance(host);
+
+	return !host->mrq->data->error &&
+		(host->wait != USDHI6_WAIT_FOR_DATA_END || !host->mrq->stop);
+}
+
+/*			Interrupt & timeout handlers			*/
+
+static irqreturn_t usdhi6_sd_bh(int irq, void *dev_id)
+{
+	struct usdhi6_host *host = dev_id;
+	struct mmc_request *mrq;
+	struct mmc_command *cmd;
+	struct mmc_data *data;
+	bool io_wait = false;
+
+	cancel_delayed_work_sync(&host->timeout_work);
+
+	mrq = host->mrq;
+	if (!mrq)
+		return IRQ_HANDLED;
+
+	cmd = mrq->cmd;
+	data = mrq->data;
+
+	switch (host->wait) {
+	case USDHI6_WAIT_FOR_REQUEST:
+		/* We're too late, the timeout has already kicked in */
+		return IRQ_HANDLED;
+	case USDHI6_WAIT_FOR_CMD:
+		/* Wait for data? */
+		io_wait = usdhi6_end_cmd(host);
+		break;
+	case USDHI6_WAIT_FOR_MREAD:
+		/* Wait for more data? */
+		io_wait = usdhi6_mread_block(host);
+		break;
+	case USDHI6_WAIT_FOR_READ:
+		/* Wait for data end? */
+		io_wait = usdhi6_read_block(host);
+		break;
+	case USDHI6_WAIT_FOR_MWRITE:
+		/* Wait data to write? */
+		io_wait = usdhi6_mwrite_block(host);
+		break;
+	case USDHI6_WAIT_FOR_WRITE:
+		/* Wait for data end? */
+		io_wait = usdhi6_write_block(host);
+		break;
+	case USDHI6_WAIT_FOR_DMA:
+		usdhi6_dma_check_error(host);
+		break;
+	case USDHI6_WAIT_FOR_STOP:
+		usdhi6_write(host, USDHI6_SD_STOP, 0);
+		if (host->io_error) {
+			int ret = usdhi6_error_code(host);
+			if (mrq->stop)
+				mrq->stop->error = ret;
+			else
+				mrq->data->error = ret;
+			dev_warn(mmc_dev(host->mmc), "%s(): %d\n", __func__, ret);
+			break;
+		}
+		usdhi6_resp_cmd12(host);
+		mrq->stop->error = 0;
+		break;
+	case USDHI6_WAIT_FOR_DATA_END:
+		if (host->io_error) {
+			mrq->data->error = usdhi6_error_code(host);
+			dev_warn(mmc_dev(host->mmc), "%s(): %d\n", __func__,
+				 mrq->data->error);
+		}
+		break;
+	default:
+		cmd->error = -EFAULT;
+		dev_err(mmc_dev(host->mmc), "Invalid state %u\n", host->wait);
+		usdhi6_request_done(host);
+		return IRQ_HANDLED;
+	}
+
+	if (io_wait) {
+		schedule_delayed_work(&host->timeout_work, host->timeout);
+		/* Wait for more data or ACCESS_END */
+		if (!host->dma_active)
+			usdhi6_wait_for_brwe(host, mrq->data->flags & MMC_DATA_READ);
+		return IRQ_HANDLED;
+	}
+
+	if (!cmd->error) {
+		if (data) {
+			if (!data->error) {
+				if (host->wait != USDHI6_WAIT_FOR_STOP &&
+				    host->mrq->stop &&
+				    !host->mrq->stop->error &&
+				    !usdhi6_stop_cmd(host)) {
+					/* Sending STOP */
+					usdhi6_wait_for_resp(host);
+
+					schedule_delayed_work(&host->timeout_work,
+							      host->timeout);
+
+					return IRQ_HANDLED;
+				}
+
+				data->bytes_xfered = data->blocks * data->blksz;
+			} else {
+				/* Data error: might need to unmap the last page */
+				dev_warn(mmc_dev(host->mmc), "%s(): data error %d\n",
+					 __func__, data->error);
+				usdhi6_sg_unmap(host, true);
+			}
+		} else if (cmd->opcode == MMC_APP_CMD) {
+			host->app_cmd = true;
+		}
+	}
+
+	usdhi6_request_done(host);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t usdhi6_sd(int irq, void *dev_id)
+{
+	struct usdhi6_host *host = dev_id;
+	u16 status, status2, error;
+
+	status = usdhi6_read(host, USDHI6_SD_INFO1) & ~host->status_mask &
+		~USDHI6_SD_INFO1_CARD;
+	status2 = usdhi6_read(host, USDHI6_SD_INFO2) & ~host->status2_mask;
+
+	usdhi6_only_cd(host);
+
+	dev_dbg(mmc_dev(host->mmc),
+		"IRQ status = 0x%08x, status2 = 0x%08x\n", status, status2);
+
+	if (!status && !status2)
+		return IRQ_NONE;
+
+	error = status2 & USDHI6_SD_INFO2_ERR;
+
+	/* Ack / clear interrupts */
+	if (USDHI6_SD_INFO1_IRQ & status)
+		usdhi6_write(host, USDHI6_SD_INFO1,
+			     0xffff & ~(USDHI6_SD_INFO1_IRQ & status));
+
+	if (USDHI6_SD_INFO2_IRQ & status2) {
+		if (error)
+			/* In error cases BWE and BRE aren't cleared automatically */
+			status2 |= USDHI6_SD_INFO2_BWE | USDHI6_SD_INFO2_BRE;
+
+		usdhi6_write(host, USDHI6_SD_INFO2,
+			     0xffff & ~(USDHI6_SD_INFO2_IRQ & status2));
+	}
+
+	host->io_error = error;
+	host->irq_status = status;
+
+	if (error) {
+		/* Don't pollute the log with unsupported command timeouts */
+		if (host->wait != USDHI6_WAIT_FOR_CMD ||
+		    error != USDHI6_SD_INFO2_RSP_TOUT)
+			dev_warn(mmc_dev(host->mmc),
+				 "%s(): INFO2 error bits 0x%08x\n",
+				 __func__, error);
+		else
+			dev_dbg(mmc_dev(host->mmc),
+				"%s(): INFO2 error bits 0x%08x\n",
+				__func__, error);
+	}
+
+	return IRQ_WAKE_THREAD;
+}
+
+static irqreturn_t usdhi6_sdio(int irq, void *dev_id)
+{
+	struct usdhi6_host *host = dev_id;
+	u32 status = usdhi6_read(host, USDHI6_SDIO_INFO1) & ~host->sdio_mask;
+
+	dev_dbg(mmc_dev(host->mmc), "%s(): status 0x%x\n", __func__, status);
+
+	if (!status)
+		return IRQ_NONE;
+
+	usdhi6_write(host, USDHI6_SDIO_INFO1, ~status);
+
+	mmc_signal_sdio_irq(host->mmc);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t usdhi6_cd(int irq, void *dev_id)
+{
+	struct usdhi6_host *host = dev_id;
+	struct mmc_host *mmc = host->mmc;
+	u16 status;
+
+	/* We're only interested in hotplug events here */
+	status = usdhi6_read(host, USDHI6_SD_INFO1) & ~host->status_mask &
+		USDHI6_SD_INFO1_CARD;
+
+	if (!status)
+		return IRQ_NONE;
+
+	/* Ack */
+	usdhi6_write(host, USDHI6_SD_INFO1, !status);
+
+	if (!work_pending(&mmc->detect.work) &&
+	    (((status & USDHI6_SD_INFO1_CARD_INSERT) &&
+	      !mmc->card) ||
+	     ((status & USDHI6_SD_INFO1_CARD_EJECT) &&
+	      mmc->card)))
+		mmc_detect_change(mmc, msecs_to_jiffies(100));
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * Actually this should not be needed, if the built-in timeout works reliably in
+ * the both PIO cases and DMA never fails. But if DMA does fail, a timeout
+ * handler might be the only way to catch the error.
+ */
+static void usdhi6_timeout_work(struct work_struct *work)
+{
+	struct delayed_work *d = container_of(work, struct delayed_work, work);
+	struct usdhi6_host *host = container_of(d, struct usdhi6_host, timeout_work);
+	struct mmc_request *mrq = host->mrq;
+	struct mmc_data *data = mrq ? mrq->data : NULL;
+
+	dev_warn(mmc_dev(host->mmc),
+		 "%s timeout wait %u CMD%d: IRQ 0x%08x:0x%08x, last IRQ 0x%08x\n",
+		 host->dma_active ? "DMA" : "PIO",
+		 host->wait, mrq ? mrq->cmd->opcode : -1,
+		 usdhi6_read(host, USDHI6_SD_INFO1),
+		 usdhi6_read(host, USDHI6_SD_INFO2), host->irq_status);
+
+	if (host->dma_active) {
+		usdhi6_dma_kill(host);
+		usdhi6_dma_stop_unmap(host);
+	}
+
+	switch (host->wait) {
+	default:
+		dev_err(mmc_dev(host->mmc), "Invalid state %u\n", host->wait);
+		/* mrq can be NULL in this actually impossible case */
+	case USDHI6_WAIT_FOR_CMD:
+		usdhi6_error_code(host);
+		if (mrq)
+			mrq->cmd->error = -ETIMEDOUT;
+		break;
+	case USDHI6_WAIT_FOR_STOP:
+		usdhi6_error_code(host);
+		mrq->stop->error = -ETIMEDOUT;
+		break;
+	case USDHI6_WAIT_FOR_DMA:
+	case USDHI6_WAIT_FOR_MREAD:
+	case USDHI6_WAIT_FOR_MWRITE:
+	case USDHI6_WAIT_FOR_READ:
+	case USDHI6_WAIT_FOR_WRITE:
+		dev_dbg(mmc_dev(host->mmc),
+			"%c: page #%u @ +0x%zx %ux%u in SG%u. Current SG %u bytes @ %u\n",
+			data->flags & MMC_DATA_READ ? 'R' : 'W', host->page_idx,
+			host->offset, data->blocks, data->blksz, data->sg_len,
+			sg_dma_len(host->sg), host->sg->offset);
+		usdhi6_sg_unmap(host, true);
+		/*
+		 * If USDHI6_WAIT_FOR_DATA_END times out, we have already unmapped
+		 * the page
+		 */
+	case USDHI6_WAIT_FOR_DATA_END:
+		usdhi6_error_code(host);
+		data->error = -ETIMEDOUT;
+	}
+
+	if (mrq)
+		usdhi6_request_done(host);
+}
+
+/*			 Probe / release				*/
+
+static const struct of_device_id usdhi6_of_match[] = {
+	{.compatible = "renesas,usdhi6rol0"},
+	{}
+};
+MODULE_DEVICE_TABLE(of, usdhi6_of_match);
+
+static int usdhi6_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct mmc_host *mmc;
+	struct usdhi6_host *host;
+	struct resource *res;
+	int irq_cd, irq_sd, irq_sdio;
+	u32 version;
+	int ret;
+
+	if (!dev->of_node)
+		return -ENODEV;
+
+	irq_cd = platform_get_irq_byname(pdev, "card detect");
+	irq_sd = platform_get_irq_byname(pdev, "data");
+	irq_sdio = platform_get_irq_byname(pdev, "SDIO");
+	if (irq_sd < 0 || irq_sdio < 0)
+		return -ENODEV;
+
+	mmc = mmc_alloc_host(sizeof(struct usdhi6_host), dev);
+	if (!mmc)
+		return -ENOMEM;
+
+	ret = mmc_of_parse(mmc);
+	if (ret < 0)
+		goto e_free_mmc;
+
+	mmc_regulator_get_supply(mmc);
+
+	host		= mmc_priv(mmc);
+	host->mmc	= mmc;
+	host->wait	= USDHI6_WAIT_FOR_REQUEST;
+	host->timeout	= msecs_to_jiffies(4000);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	host->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(host->base)) {
+		ret = PTR_ERR(host->base);
+		goto e_free_mmc;
+	}
+
+	host->clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(host->clk))
+		goto e_free_mmc;
+
+	host->imclk = clk_get_rate(host->clk);
+
+	ret = clk_prepare_enable(host->clk);
+	if (ret < 0)
+		goto e_free_mmc;
+
+	version = usdhi6_read(host, USDHI6_VERSION);
+	if ((version & 0xfff) != 0xa0d) {
+		dev_err(dev, "Version not recognized %x\n", version);
+		goto e_clk_off;
+	}
+
+	dev_info(dev, "A USDHI6ROL0 SD host detected with %d ports\n",
+		 usdhi6_read(host, USDHI6_SD_PORT_SEL) >> USDHI6_SD_PORT_SEL_PORTS_SHIFT);
+
+	usdhi6_mask_all(host);
+
+	if (irq_cd >= 0) {
+		ret = devm_request_irq(dev, irq_cd, usdhi6_cd, 0,
+				       dev_name(dev), host);
+		if (ret < 0)
+			goto e_clk_off;
+	} else {
+		mmc->caps |= MMC_CAP_NEEDS_POLL;
+	}
+
+	ret = devm_request_threaded_irq(dev, irq_sd, usdhi6_sd, usdhi6_sd_bh, 0,
+			       dev_name(dev), host);
+	if (ret < 0)
+		goto e_clk_off;
+
+	ret = devm_request_irq(dev, irq_sdio, usdhi6_sdio, 0,
+			       dev_name(dev), host);
+	if (ret < 0)
+		goto e_clk_off;
+
+	INIT_DELAYED_WORK(&host->timeout_work, usdhi6_timeout_work);
+
+	usdhi6_dma_request(host, res->start);
+
+	mmc->ops = &usdhi6_ops;
+	mmc->caps |= MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED |
+		MMC_CAP_UHS_SDR50 | MMC_CAP_UHS_DDR50 | MMC_CAP_SDIO_IRQ;
+	/* Set .max_segs to some random number. Feel free to adjust. */
+	mmc->max_segs = 32;
+	mmc->max_blk_size = 512;
+	mmc->max_req_size = PAGE_CACHE_SIZE * mmc->max_segs;
+	mmc->max_blk_count = mmc->max_req_size / mmc->max_blk_size;
+	/*
+	 * Setting .max_seg_size to 1 page would simplify our page-mapping code,
+	 * But OTOH, having large segments makes DMA more efficient. We could
+	 * check, whether we managed to get DMA and fall back to 1 page
+	 * segments, but if we do manage to obtain DMA and then it fails at
+	 * run-time and we fall back to PIO, we will continue getting large
+	 * segments. So, we wouldn't be able to get rid of the code anyway.
+	 */
+	mmc->max_seg_size = mmc->max_req_size;
+	if (!mmc->f_max)
+		mmc->f_max = host->imclk;
+	mmc->f_min = host->imclk / 512;
+
+	platform_set_drvdata(pdev, host);
+
+	ret = mmc_add_host(mmc);
+	if (ret < 0)
+		goto e_clk_off;
+
+	return 0;
+
+e_clk_off:
+	clk_disable_unprepare(host->clk);
+e_free_mmc:
+	mmc_free_host(mmc);
+
+	return ret;
+}
+
+static int usdhi6_remove(struct platform_device *pdev)
+{
+	struct usdhi6_host *host = platform_get_drvdata(pdev);
+
+	mmc_remove_host(host->mmc);
+
+	usdhi6_mask_all(host);
+	cancel_delayed_work_sync(&host->timeout_work);
+	usdhi6_dma_release(host);
+	clk_disable_unprepare(host->clk);
+	mmc_free_host(host->mmc);
+
+	return 0;
+}
+
+static struct platform_driver usdhi6_driver = {
+	.probe		= usdhi6_probe,
+	.remove		= usdhi6_remove,
+	.driver		= {
+		.name	= "usdhi6rol0",
+		.owner	= THIS_MODULE,
+		.of_match_table = usdhi6_of_match,
+	},
+};
+
+module_platform_driver(usdhi6_driver);
+
+MODULE_DESCRIPTION("Renesas usdhi6rol0 SD/SDIO host driver");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:usdhi6rol0");
+MODULE_AUTHOR("Guennadi Liakhovetski <g.liakhovetski@gmx.de>");
diff --git a/drivers/mmc/host/wmt-sdmmc.c b/drivers/mmc/host/wmt-sdmmc.c
index 498d1f799085..282891a8e451 100644
--- a/drivers/mmc/host/wmt-sdmmc.c
+++ b/drivers/mmc/host/wmt-sdmmc.c
@@ -840,7 +840,7 @@ static int wmt_mci_probe(struct platform_device *pdev)
 	priv->dma_desc_buffer = dma_alloc_coherent(&pdev->dev,
 						   mmc->max_blk_count * 16,
 						   &priv->dma_desc_device_addr,
-						   208);
+						   GFP_KERNEL);
 	if (!priv->dma_desc_buffer) {
 		dev_err(&pdev->dev, "DMA alloc fail\n");
 		ret = -EPERM;
diff --git a/drivers/mtd/Kconfig b/drivers/mtd/Kconfig
index 5d49a2129618..94b821042d9d 100644
--- a/drivers/mtd/Kconfig
+++ b/drivers/mtd/Kconfig
@@ -321,6 +321,8 @@ source "drivers/mtd/onenand/Kconfig"
 
 source "drivers/mtd/lpddr/Kconfig"
 
+source "drivers/mtd/spi-nor/Kconfig"
+
 source "drivers/mtd/ubi/Kconfig"
 
 endif # MTD
diff --git a/drivers/mtd/Makefile b/drivers/mtd/Makefile
index 4cfb31e6c966..99bb9a1f6e16 100644
--- a/drivers/mtd/Makefile
+++ b/drivers/mtd/Makefile
@@ -32,4 +32,5 @@ inftl-objs		:= inftlcore.o inftlmount.o
 
 obj-y		+= chips/ lpddr/ maps/ devices/ nand/ onenand/ tests/
 
+obj-$(CONFIG_MTD_SPI_NOR)	+= spi-nor/
 obj-$(CONFIG_MTD_UBI)		+= ubi/
diff --git a/drivers/mtd/chips/Kconfig b/drivers/mtd/chips/Kconfig
index e4696b37f3de..9f02c28c0204 100644
--- a/drivers/mtd/chips/Kconfig
+++ b/drivers/mtd/chips/Kconfig
@@ -169,33 +169,33 @@ config MTD_OTP
 	  in the programming of OTP bits will waste them.
 
 config MTD_CFI_INTELEXT
-	tristate "Support for Intel/Sharp flash chips"
+	tristate "Support for CFI command set 0001 (Intel/Sharp chips)"
 	depends on MTD_GEN_PROBE
 	select MTD_CFI_UTIL
 	help
 	  The Common Flash Interface defines a number of different command
 	  sets which a CFI-compliant chip may claim to implement. This code
-	  provides support for one of those command sets, used on Intel
-	  StrataFlash and other parts.
+	  provides support for command set 0001, used on Intel StrataFlash
+	  and other parts.
 
 config MTD_CFI_AMDSTD
-	tristate "Support for AMD/Fujitsu/Spansion flash chips"
+	tristate "Support for CFI command set 0002 (AMD/Fujitsu/Spansion chips)"
 	depends on MTD_GEN_PROBE
 	select MTD_CFI_UTIL
 	help
 	  The Common Flash Interface defines a number of different command
 	  sets which a CFI-compliant chip may claim to implement. This code
-	  provides support for one of those command sets, used on chips
-	  including the AMD Am29LV320.
+	  provides support for command set 0002, used on chips including
+	  the AMD Am29LV320.
 
 config MTD_CFI_STAA
-	tristate "Support for ST (Advanced Architecture) flash chips"
+	tristate "Support for CFI command set 0020 (ST (Advanced Architecture) chips)"
 	depends on MTD_GEN_PROBE
 	select MTD_CFI_UTIL
 	help
 	  The Common Flash Interface defines a number of different command
 	  sets which a CFI-compliant chip may claim to implement. This code
-	  provides support for one of those command sets.
+	  provides support for command set 0020.
 
 config MTD_CFI_UTIL
 	tristate
diff --git a/drivers/mtd/chips/cfi_cmdset_0020.c b/drivers/mtd/chips/cfi_cmdset_0020.c
index 6293855fb5ee..423666b51efb 100644
--- a/drivers/mtd/chips/cfi_cmdset_0020.c
+++ b/drivers/mtd/chips/cfi_cmdset_0020.c
@@ -961,7 +961,7 @@ static int cfi_staa_erase_varsize(struct mtd_info *mtd,
 			chipnum++;
 
 			if (chipnum >= cfi->numchips)
-			break;
+				break;
 		}
 	}
 
@@ -1170,7 +1170,7 @@ static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 			chipnum++;
 
 			if (chipnum >= cfi->numchips)
-			break;
+				break;
 		}
 	}
 	return 0;
diff --git a/drivers/mtd/chips/cfi_util.c b/drivers/mtd/chips/cfi_util.c
index 08049f6eea60..09c79bd0b4f4 100644
--- a/drivers/mtd/chips/cfi_util.c
+++ b/drivers/mtd/chips/cfi_util.c
@@ -239,7 +239,7 @@ int cfi_varsize_frob(struct mtd_info *mtd, varsize_frob_t frob,
 			chipnum++;
 
 			if (chipnum >= cfi->numchips)
-			break;
+				break;
 		}
 	}
 
diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig
index 1210bc2923b7..c49d0b127fef 100644
--- a/drivers/mtd/devices/Kconfig
+++ b/drivers/mtd/devices/Kconfig
@@ -80,7 +80,7 @@ config MTD_DATAFLASH_OTP
 
 config MTD_M25P80
 	tristate "Support most SPI Flash chips (AT26DF, M25P, W25X, ...)"
-	depends on SPI_MASTER
+	depends on SPI_MASTER && MTD_SPI_NOR
 	help
 	  This enables access to most modern SPI flash chips, used for
 	  program and data storage.   Series supported include Atmel AT26DF,
@@ -212,7 +212,7 @@ config MTD_DOCG3
 
 config MTD_ST_SPI_FSM
 	tristate "ST Microelectronics SPI FSM Serial Flash Controller"
-	depends on ARM || SH
+	depends on ARCH_STI
 	help
 	  This provides an MTD device driver for the ST Microelectronics
 	  SPI Fast Sequence Mode (FSM) Serial Flash Controller and support
diff --git a/drivers/mtd/devices/docg3.c b/drivers/mtd/devices/docg3.c
index dd5e1018d37b..91a169c44b39 100644
--- a/drivers/mtd/devices/docg3.c
+++ b/drivers/mtd/devices/docg3.c
@@ -1608,8 +1608,8 @@ static ssize_t dps1_insert_key(struct device *dev,
 #define FLOOR_SYSFS(id) { \
 	__ATTR(f##id##_dps0_is_keylocked, S_IRUGO, dps0_is_key_locked, NULL), \
 	__ATTR(f##id##_dps1_is_keylocked, S_IRUGO, dps1_is_key_locked, NULL), \
-	__ATTR(f##id##_dps0_protection_key, S_IWUGO, NULL, dps0_insert_key), \
-	__ATTR(f##id##_dps1_protection_key, S_IWUGO, NULL, dps1_insert_key), \
+	__ATTR(f##id##_dps0_protection_key, S_IWUSR|S_IWGRP, NULL, dps0_insert_key), \
+	__ATTR(f##id##_dps1_protection_key, S_IWUSR|S_IWGRP, NULL, dps1_insert_key), \
 }
 
 static struct device_attribute doc_sys_attrs[DOC_MAX_NBFLOORS][4] = {
diff --git a/drivers/mtd/devices/elm.c b/drivers/mtd/devices/elm.c
index 1fd4a0f77967..7df86948e6d4 100644
--- a/drivers/mtd/devices/elm.c
+++ b/drivers/mtd/devices/elm.c
@@ -213,6 +213,28 @@ static void elm_load_syndrome(struct elm_info *info,
 				val = cpu_to_be32(*(u32 *) &ecc[0]) >> 12;
 				elm_write_reg(info, offset, val);
 				break;
+			case BCH16_ECC:
+				val = cpu_to_be32(*(u32 *) &ecc[22]);
+				elm_write_reg(info, offset, val);
+				offset += 4;
+				val = cpu_to_be32(*(u32 *) &ecc[18]);
+				elm_write_reg(info, offset, val);
+				offset += 4;
+				val = cpu_to_be32(*(u32 *) &ecc[14]);
+				elm_write_reg(info, offset, val);
+				offset += 4;
+				val = cpu_to_be32(*(u32 *) &ecc[10]);
+				elm_write_reg(info, offset, val);
+				offset += 4;
+				val = cpu_to_be32(*(u32 *) &ecc[6]);
+				elm_write_reg(info, offset, val);
+				offset += 4;
+				val = cpu_to_be32(*(u32 *) &ecc[2]);
+				elm_write_reg(info, offset, val);
+				offset += 4;
+				val = cpu_to_be32(*(u32 *) &ecc[0]) >> 16;
+				elm_write_reg(info, offset, val);
+				break;
 			default:
 				pr_err("invalid config bch_type\n");
 			}
@@ -418,6 +440,7 @@ static int elm_remove(struct platform_device *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
 /**
  * elm_context_save
  * saves ELM configurations to preserve them across Hardware powered-down
@@ -435,6 +458,13 @@ static int elm_context_save(struct elm_info *info)
 	for (i = 0; i < ERROR_VECTOR_MAX; i++) {
 		offset = i * SYNDROME_FRAGMENT_REG_SIZE;
 		switch (bch_type) {
+		case BCH16_ECC:
+			regs->elm_syndrome_fragment_6[i] = elm_read_reg(info,
+					ELM_SYNDROME_FRAGMENT_6 + offset);
+			regs->elm_syndrome_fragment_5[i] = elm_read_reg(info,
+					ELM_SYNDROME_FRAGMENT_5 + offset);
+			regs->elm_syndrome_fragment_4[i] = elm_read_reg(info,
+					ELM_SYNDROME_FRAGMENT_4 + offset);
 		case BCH8_ECC:
 			regs->elm_syndrome_fragment_3[i] = elm_read_reg(info,
 					ELM_SYNDROME_FRAGMENT_3 + offset);
@@ -473,6 +503,13 @@ static int elm_context_restore(struct elm_info *info)
 	for (i = 0; i < ERROR_VECTOR_MAX; i++) {
 		offset = i * SYNDROME_FRAGMENT_REG_SIZE;
 		switch (bch_type) {
+		case BCH16_ECC:
+			elm_write_reg(info, ELM_SYNDROME_FRAGMENT_6 + offset,
+					regs->elm_syndrome_fragment_6[i]);
+			elm_write_reg(info, ELM_SYNDROME_FRAGMENT_5 + offset,
+					regs->elm_syndrome_fragment_5[i]);
+			elm_write_reg(info, ELM_SYNDROME_FRAGMENT_4 + offset,
+					regs->elm_syndrome_fragment_4[i]);
 		case BCH8_ECC:
 			elm_write_reg(info, ELM_SYNDROME_FRAGMENT_3 + offset,
 					regs->elm_syndrome_fragment_3[i]);
@@ -509,6 +546,7 @@ static int elm_resume(struct device *dev)
 	elm_context_restore(info);
 	return 0;
 }
+#endif
 
 static SIMPLE_DEV_PM_OPS(elm_pm_ops, elm_suspend, elm_resume);
 
diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index 524dab3ac938..ed7e0a1bed3c 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -19,485 +19,98 @@
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/device.h>
-#include <linux/interrupt.h>
-#include <linux/mutex.h>
-#include <linux/math64.h>
-#include <linux/slab.h>
-#include <linux/sched.h>
-#include <linux/mod_devicetable.h>
 
-#include <linux/mtd/cfi.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
-#include <linux/of_platform.h>
 
 #include <linux/spi/spi.h>
 #include <linux/spi/flash.h>
+#include <linux/mtd/spi-nor.h>
 
-/* Flash opcodes. */
-#define	OPCODE_WREN		0x06	/* Write enable */
-#define	OPCODE_RDSR		0x05	/* Read status register */
-#define	OPCODE_WRSR		0x01	/* Write status register 1 byte */
-#define	OPCODE_NORM_READ	0x03	/* Read data bytes (low frequency) */
-#define	OPCODE_FAST_READ	0x0b	/* Read data bytes (high frequency) */
-#define	OPCODE_DUAL_READ        0x3b    /* Read data bytes (Dual SPI) */
-#define	OPCODE_QUAD_READ        0x6b    /* Read data bytes (Quad SPI) */
-#define	OPCODE_PP		0x02	/* Page program (up to 256 bytes) */
-#define	OPCODE_BE_4K		0x20	/* Erase 4KiB block */
-#define	OPCODE_BE_4K_PMC	0xd7	/* Erase 4KiB block on PMC chips */
-#define	OPCODE_BE_32K		0x52	/* Erase 32KiB block */
-#define	OPCODE_CHIP_ERASE	0xc7	/* Erase whole flash chip */
-#define	OPCODE_SE		0xd8	/* Sector erase (usually 64KiB) */
-#define	OPCODE_RDID		0x9f	/* Read JEDEC ID */
-#define	OPCODE_RDCR             0x35    /* Read configuration register */
-
-/* 4-byte address opcodes - used on Spansion and some Macronix flashes. */
-#define	OPCODE_NORM_READ_4B	0x13	/* Read data bytes (low frequency) */
-#define	OPCODE_FAST_READ_4B	0x0c	/* Read data bytes (high frequency) */
-#define	OPCODE_DUAL_READ_4B	0x3c    /* Read data bytes (Dual SPI) */
-#define	OPCODE_QUAD_READ_4B	0x6c    /* Read data bytes (Quad SPI) */
-#define	OPCODE_PP_4B		0x12	/* Page program (up to 256 bytes) */
-#define	OPCODE_SE_4B		0xdc	/* Sector erase (usually 64KiB) */
-
-/* Used for SST flashes only. */
-#define	OPCODE_BP		0x02	/* Byte program */
-#define	OPCODE_WRDI		0x04	/* Write disable */
-#define	OPCODE_AAI_WP		0xad	/* Auto address increment word program */
-
-/* Used for Macronix and Winbond flashes. */
-#define	OPCODE_EN4B		0xb7	/* Enter 4-byte mode */
-#define	OPCODE_EX4B		0xe9	/* Exit 4-byte mode */
-
-/* Used for Spansion flashes only. */
-#define	OPCODE_BRWR		0x17	/* Bank register write */
-
-/* Status Register bits. */
-#define	SR_WIP			1	/* Write in progress */
-#define	SR_WEL			2	/* Write enable latch */
-/* meaning of other SR_* bits may differ between vendors */
-#define	SR_BP0			4	/* Block protect 0 */
-#define	SR_BP1			8	/* Block protect 1 */
-#define	SR_BP2			0x10	/* Block protect 2 */
-#define	SR_SRWD			0x80	/* SR write protect */
-
-#define SR_QUAD_EN_MX           0x40    /* Macronix Quad I/O */
-
-/* Configuration Register bits. */
-#define CR_QUAD_EN_SPAN		0x2     /* Spansion Quad I/O */
-
-/* Define max times to check status register before we give up. */
-#define	MAX_READY_WAIT_JIFFIES	(40 * HZ)	/* M25P16 specs 40s max chip erase */
 #define	MAX_CMD_SIZE		6
-
-#define JEDEC_MFR(_jedec_id)	((_jedec_id) >> 16)
-
-/****************************************************************************/
-
-enum read_type {
-	M25P80_NORMAL = 0,
-	M25P80_FAST,
-	M25P80_DUAL,
-	M25P80_QUAD,
-};
-
 struct m25p {
 	struct spi_device	*spi;
-	struct mutex		lock;
+	struct spi_nor		spi_nor;
 	struct mtd_info		mtd;
-	u16			page_size;
-	u16			addr_width;
-	u8			erase_opcode;
-	u8			read_opcode;
-	u8			program_opcode;
-	u8			*command;
-	enum read_type		flash_read;
+	u8			command[MAX_CMD_SIZE];
 };
 
-static inline struct m25p *mtd_to_m25p(struct mtd_info *mtd)
-{
-	return container_of(mtd, struct m25p, mtd);
-}
-
-/****************************************************************************/
-
-/*
- * Internal helper functions
- */
-
-/*
- * Read the status register, returning its value in the location
- * Return the status register value.
- * Returns negative if error occurred.
- */
-static int read_sr(struct m25p *flash)
-{
-	ssize_t retval;
-	u8 code = OPCODE_RDSR;
-	u8 val;
-
-	retval = spi_write_then_read(flash->spi, &code, 1, &val, 1);
-
-	if (retval < 0) {
-		dev_err(&flash->spi->dev, "error %d reading SR\n",
-				(int) retval);
-		return retval;
-	}
-
-	return val;
-}
-
-/*
- * Read configuration register, returning its value in the
- * location. Return the configuration register value.
- * Returns negative if error occured.
- */
-static int read_cr(struct m25p *flash)
-{
-	u8 code = OPCODE_RDCR;
-	int ret;
-	u8 val;
-
-	ret = spi_write_then_read(flash->spi, &code, 1, &val, 1);
-	if (ret < 0) {
-		dev_err(&flash->spi->dev, "error %d reading CR\n", ret);
-		return ret;
-	}
-
-	return val;
-}
-
-/*
- * Write status register 1 byte
- * Returns negative if error occurred.
- */
-static int write_sr(struct m25p *flash, u8 val)
-{
-	flash->command[0] = OPCODE_WRSR;
-	flash->command[1] = val;
-
-	return spi_write(flash->spi, flash->command, 2);
-}
-
-/*
- * Set write enable latch with Write Enable command.
- * Returns negative if error occurred.
- */
-static inline int write_enable(struct m25p *flash)
-{
-	u8	code = OPCODE_WREN;
-
-	return spi_write_then_read(flash->spi, &code, 1, NULL, 0);
-}
-
-/*
- * Send write disble instruction to the chip.
- */
-static inline int write_disable(struct m25p *flash)
-{
-	u8	code = OPCODE_WRDI;
-
-	return spi_write_then_read(flash->spi, &code, 1, NULL, 0);
-}
-
-/*
- * Enable/disable 4-byte addressing mode.
- */
-static inline int set_4byte(struct m25p *flash, u32 jedec_id, int enable)
-{
-	int status;
-	bool need_wren = false;
-
-	switch (JEDEC_MFR(jedec_id)) {
-	case CFI_MFR_ST: /* Micron, actually */
-		/* Some Micron need WREN command; all will accept it */
-		need_wren = true;
-	case CFI_MFR_MACRONIX:
-	case 0xEF /* winbond */:
-		if (need_wren)
-			write_enable(flash);
-
-		flash->command[0] = enable ? OPCODE_EN4B : OPCODE_EX4B;
-		status = spi_write(flash->spi, flash->command, 1);
-
-		if (need_wren)
-			write_disable(flash);
-
-		return status;
-	default:
-		/* Spansion style */
-		flash->command[0] = OPCODE_BRWR;
-		flash->command[1] = enable << 7;
-		return spi_write(flash->spi, flash->command, 2);
-	}
-}
-
-/*
- * Service routine to read status register until ready, or timeout occurs.
- * Returns non-zero if error.
- */
-static int wait_till_ready(struct m25p *flash)
-{
-	unsigned long deadline;
-	int sr;
-
-	deadline = jiffies + MAX_READY_WAIT_JIFFIES;
-
-	do {
-		if ((sr = read_sr(flash)) < 0)
-			break;
-		else if (!(sr & SR_WIP))
-			return 0;
-
-		cond_resched();
-
-	} while (!time_after_eq(jiffies, deadline));
-
-	return 1;
-}
-
-/*
- * Write status Register and configuration register with 2 bytes
- * The first byte will be written to the status register, while the
- * second byte will be written to the configuration register.
- * Return negative if error occured.
- */
-static int write_sr_cr(struct m25p *flash, u16 val)
-{
-	flash->command[0] = OPCODE_WRSR;
-	flash->command[1] = val & 0xff;
-	flash->command[2] = (val >> 8);
-
-	return spi_write(flash->spi, flash->command, 3);
-}
-
-static int macronix_quad_enable(struct m25p *flash)
-{
-	int ret, val;
-	u8 cmd[2];
-	cmd[0] = OPCODE_WRSR;
-
-	val = read_sr(flash);
-	cmd[1] = val | SR_QUAD_EN_MX;
-	write_enable(flash);
-
-	spi_write(flash->spi, &cmd, 2);
-
-	if (wait_till_ready(flash))
-		return 1;
-
-	ret = read_sr(flash);
-	if (!(ret > 0 && (ret & SR_QUAD_EN_MX))) {
-		dev_err(&flash->spi->dev, "Macronix Quad bit not set\n");
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int spansion_quad_enable(struct m25p *flash)
+static int m25p80_read_reg(struct spi_nor *nor, u8 code, u8 *val, int len)
 {
+	struct m25p *flash = nor->priv;
+	struct spi_device *spi = flash->spi;
 	int ret;
-	int quad_en = CR_QUAD_EN_SPAN << 8;
-
-	write_enable(flash);
 
-	ret = write_sr_cr(flash, quad_en);
-	if (ret < 0) {
-		dev_err(&flash->spi->dev,
-			"error while writing configuration register\n");
-		return -EINVAL;
-	}
-
-	/* read back and check it */
-	ret = read_cr(flash);
-	if (!(ret > 0 && (ret & CR_QUAD_EN_SPAN))) {
-		dev_err(&flash->spi->dev, "Spansion Quad bit not set\n");
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int set_quad_mode(struct m25p *flash, u32 jedec_id)
-{
-	int status;
-
-	switch (JEDEC_MFR(jedec_id)) {
-	case CFI_MFR_MACRONIX:
-		status = macronix_quad_enable(flash);
-		if (status) {
-			dev_err(&flash->spi->dev,
-				"Macronix quad-read not enabled\n");
-			return -EINVAL;
-		}
-		return status;
-	default:
-		status = spansion_quad_enable(flash);
-		if (status) {
-			dev_err(&flash->spi->dev,
-				"Spansion quad-read not enabled\n");
-			return -EINVAL;
-		}
-		return status;
-	}
-}
-
-/*
- * Erase the whole flash memory
- *
- * Returns 0 if successful, non-zero otherwise.
- */
-static int erase_chip(struct m25p *flash)
-{
-	pr_debug("%s: %s %lldKiB\n", dev_name(&flash->spi->dev), __func__,
-			(long long)(flash->mtd.size >> 10));
+	ret = spi_write_then_read(spi, &code, 1, val, len);
+	if (ret < 0)
+		dev_err(&spi->dev, "error %d reading %x\n", ret, code);
 
-	/* Wait until finished previous write command. */
-	if (wait_till_ready(flash))
-		return 1;
-
-	/* Send write enable, then erase commands. */
-	write_enable(flash);
-
-	/* Set up command buffer. */
-	flash->command[0] = OPCODE_CHIP_ERASE;
-
-	spi_write(flash->spi, flash->command, 1);
-
-	return 0;
+	return ret;
 }
 
-static void m25p_addr2cmd(struct m25p *flash, unsigned int addr, u8 *cmd)
+static void m25p_addr2cmd(struct spi_nor *nor, unsigned int addr, u8 *cmd)
 {
 	/* opcode is in cmd[0] */
-	cmd[1] = addr >> (flash->addr_width * 8 -  8);
-	cmd[2] = addr >> (flash->addr_width * 8 - 16);
-	cmd[3] = addr >> (flash->addr_width * 8 - 24);
-	cmd[4] = addr >> (flash->addr_width * 8 - 32);
+	cmd[1] = addr >> (nor->addr_width * 8 -  8);
+	cmd[2] = addr >> (nor->addr_width * 8 - 16);
+	cmd[3] = addr >> (nor->addr_width * 8 - 24);
+	cmd[4] = addr >> (nor->addr_width * 8 - 32);
 }
 
-static int m25p_cmdsz(struct m25p *flash)
+static int m25p_cmdsz(struct spi_nor *nor)
 {
-	return 1 + flash->addr_width;
+	return 1 + nor->addr_width;
 }
 
-/*
- * Erase one sector of flash memory at offset ``offset'' which is any
- * address within the sector which should be erased.
- *
- * Returns 0 if successful, non-zero otherwise.
- */
-static int erase_sector(struct m25p *flash, u32 offset)
+static int m25p80_write_reg(struct spi_nor *nor, u8 opcode, u8 *buf, int len,
+			int wr_en)
 {
-	pr_debug("%s: %s %dKiB at 0x%08x\n", dev_name(&flash->spi->dev),
-			__func__, flash->mtd.erasesize / 1024, offset);
-
-	/* Wait until finished previous write command. */
-	if (wait_till_ready(flash))
-		return 1;
+	struct m25p *flash = nor->priv;
+	struct spi_device *spi = flash->spi;
 
-	/* Send write enable, then erase commands. */
-	write_enable(flash);
-
-	/* Set up command buffer. */
-	flash->command[0] = flash->erase_opcode;
-	m25p_addr2cmd(flash, offset, flash->command);
-
-	spi_write(flash->spi, flash->command, m25p_cmdsz(flash));
+	flash->command[0] = opcode;
+	if (buf)
+		memcpy(&flash->command[1], buf, len);
 
-	return 0;
+	return spi_write(spi, flash->command, len + 1);
 }
 
-/****************************************************************************/
-
-/*
- * MTD implementation
- */
-
-/*
- * Erase an address range on the flash chip.  The address range may extend
- * one or more erase sectors.  Return an error is there is a problem erasing.
- */
-static int m25p80_erase(struct mtd_info *mtd, struct erase_info *instr)
+static void m25p80_write(struct spi_nor *nor, loff_t to, size_t len,
+			size_t *retlen, const u_char *buf)
 {
-	struct m25p *flash = mtd_to_m25p(mtd);
-	u32 addr,len;
-	uint32_t rem;
-
-	pr_debug("%s: %s at 0x%llx, len %lld\n", dev_name(&flash->spi->dev),
-			__func__, (long long)instr->addr,
-			(long long)instr->len);
-
-	div_u64_rem(instr->len, mtd->erasesize, &rem);
-	if (rem)
-		return -EINVAL;
-
-	addr = instr->addr;
-	len = instr->len;
-
-	mutex_lock(&flash->lock);
-
-	/* whole-chip erase? */
-	if (len == flash->mtd.size) {
-		if (erase_chip(flash)) {
-			instr->state = MTD_ERASE_FAILED;
-			mutex_unlock(&flash->lock);
-			return -EIO;
-		}
+	struct m25p *flash = nor->priv;
+	struct spi_device *spi = flash->spi;
+	struct spi_transfer t[2] = {};
+	struct spi_message m;
+	int cmd_sz = m25p_cmdsz(nor);
 
-	/* REVISIT in some cases we could speed up erasing large regions
-	 * by using OPCODE_SE instead of OPCODE_BE_4K.  We may have set up
-	 * to use "small sector erase", but that's not always optimal.
-	 */
+	spi_message_init(&m);
 
-	/* "sector"-at-a-time erase */
-	} else {
-		while (len) {
-			if (erase_sector(flash, addr)) {
-				instr->state = MTD_ERASE_FAILED;
-				mutex_unlock(&flash->lock);
-				return -EIO;
-			}
+	if (nor->program_opcode == SPINOR_OP_AAI_WP && nor->sst_write_second)
+		cmd_sz = 1;
 
-			addr += mtd->erasesize;
-			len -= mtd->erasesize;
-		}
-	}
+	flash->command[0] = nor->program_opcode;
+	m25p_addr2cmd(nor, to, flash->command);
 
-	mutex_unlock(&flash->lock);
+	t[0].tx_buf = flash->command;
+	t[0].len = cmd_sz;
+	spi_message_add_tail(&t[0], &m);
 
-	instr->state = MTD_ERASE_DONE;
-	mtd_erase_callback(instr);
+	t[1].tx_buf = buf;
+	t[1].len = len;
+	spi_message_add_tail(&t[1], &m);
 
-	return 0;
-}
+	spi_sync(spi, &m);
 
-/*
- * Dummy Cycle calculation for different type of read.
- * It can be used to support more commands with
- * different dummy cycle requirements.
- */
-static inline int m25p80_dummy_cycles_read(struct m25p *flash)
-{
-	switch (flash->flash_read) {
-	case M25P80_FAST:
-	case M25P80_DUAL:
-	case M25P80_QUAD:
-		return 1;
-	case M25P80_NORMAL:
-		return 0;
-	default:
-		dev_err(&flash->spi->dev, "No valid read type supported\n");
-		return -1;
-	}
+	*retlen += m.actual_length - cmd_sz;
 }
 
-static inline unsigned int m25p80_rx_nbits(const struct m25p *flash)
+static inline unsigned int m25p80_rx_nbits(struct spi_nor *nor)
 {
-	switch (flash->flash_read) {
-	case M25P80_DUAL:
+	switch (nor->flash_read) {
+	case SPI_NOR_DUAL:
 		return 2;
-	case M25P80_QUAD:
+	case SPI_NOR_QUAD:
 		return 4;
 	default:
 		return 0;
@@ -505,590 +118,72 @@ static inline unsigned int m25p80_rx_nbits(const struct m25p *flash)
 }
 
 /*
- * Read an address range from the flash chip.  The address range
+ * Read an address range from the nor chip.  The address range
  * may be any size provided it is within the physical boundaries.
  */
-static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len,
-	size_t *retlen, u_char *buf)
+static int m25p80_read(struct spi_nor *nor, loff_t from, size_t len,
+			size_t *retlen, u_char *buf)
 {
-	struct m25p *flash = mtd_to_m25p(mtd);
+	struct m25p *flash = nor->priv;
+	struct spi_device *spi = flash->spi;
 	struct spi_transfer t[2];
 	struct spi_message m;
-	uint8_t opcode;
-	int dummy;
+	int dummy = nor->read_dummy;
+	int ret;
 
-	pr_debug("%s: %s from 0x%08x, len %zd\n", dev_name(&flash->spi->dev),
-			__func__, (u32)from, len);
+	/* Wait till previous write/erase is done. */
+	ret = nor->wait_till_ready(nor);
+	if (ret)
+		return ret;
 
 	spi_message_init(&m);
 	memset(t, 0, (sizeof t));
 
-	dummy =  m25p80_dummy_cycles_read(flash);
-	if (dummy < 0) {
-		dev_err(&flash->spi->dev, "No valid read command supported\n");
-		return -EINVAL;
-	}
+	flash->command[0] = nor->read_opcode;
+	m25p_addr2cmd(nor, from, flash->command);
 
 	t[0].tx_buf = flash->command;
-	t[0].len = m25p_cmdsz(flash) + dummy;
+	t[0].len = m25p_cmdsz(nor) + dummy;
 	spi_message_add_tail(&t[0], &m);
 
 	t[1].rx_buf = buf;
-	t[1].rx_nbits = m25p80_rx_nbits(flash);
+	t[1].rx_nbits = m25p80_rx_nbits(nor);
 	t[1].len = len;
 	spi_message_add_tail(&t[1], &m);
 
-	mutex_lock(&flash->lock);
-
-	/* Wait till previous write/erase is done. */
-	if (wait_till_ready(flash)) {
-		/* REVISIT status return?? */
-		mutex_unlock(&flash->lock);
-		return 1;
-	}
-
-	/* Set up the write data buffer. */
-	opcode = flash->read_opcode;
-	flash->command[0] = opcode;
-	m25p_addr2cmd(flash, from, flash->command);
-
-	spi_sync(flash->spi, &m);
-
-	*retlen = m.actual_length - m25p_cmdsz(flash) - dummy;
-
-	mutex_unlock(&flash->lock);
+	spi_sync(spi, &m);
 
+	*retlen = m.actual_length - m25p_cmdsz(nor) - dummy;
 	return 0;
 }
 
-/*
- * Write an address range to the flash chip.  Data must be written in
- * FLASH_PAGESIZE chunks.  The address range may be any size provided
- * it is within the physical boundaries.
- */
-static int m25p80_write(struct mtd_info *mtd, loff_t to, size_t len,
-	size_t *retlen, const u_char *buf)
+static int m25p80_erase(struct spi_nor *nor, loff_t offset)
 {
-	struct m25p *flash = mtd_to_m25p(mtd);
-	u32 page_offset, page_size;
-	struct spi_transfer t[2];
-	struct spi_message m;
-
-	pr_debug("%s: %s to 0x%08x, len %zd\n", dev_name(&flash->spi->dev),
-			__func__, (u32)to, len);
-
-	spi_message_init(&m);
-	memset(t, 0, (sizeof t));
-
-	t[0].tx_buf = flash->command;
-	t[0].len = m25p_cmdsz(flash);
-	spi_message_add_tail(&t[0], &m);
-
-	t[1].tx_buf = buf;
-	spi_message_add_tail(&t[1], &m);
-
-	mutex_lock(&flash->lock);
-
-	/* Wait until finished previous write command. */
-	if (wait_till_ready(flash)) {
-		mutex_unlock(&flash->lock);
-		return 1;
-	}
-
-	write_enable(flash);
-
-	/* Set up the opcode in the write buffer. */
-	flash->command[0] = flash->program_opcode;
-	m25p_addr2cmd(flash, to, flash->command);
-
-	page_offset = to & (flash->page_size - 1);
-
-	/* do all the bytes fit onto one page? */
-	if (page_offset + len <= flash->page_size) {
-		t[1].len = len;
-
-		spi_sync(flash->spi, &m);
-
-		*retlen = m.actual_length - m25p_cmdsz(flash);
-	} else {
-		u32 i;
-
-		/* the size of data remaining on the first page */
-		page_size = flash->page_size - page_offset;
-
-		t[1].len = page_size;
-		spi_sync(flash->spi, &m);
-
-		*retlen = m.actual_length - m25p_cmdsz(flash);
-
-		/* write everything in flash->page_size chunks */
-		for (i = page_size; i < len; i += page_size) {
-			page_size = len - i;
-			if (page_size > flash->page_size)
-				page_size = flash->page_size;
-
-			/* write the next page to flash */
-			m25p_addr2cmd(flash, to + i, flash->command);
-
-			t[1].tx_buf = buf + i;
-			t[1].len = page_size;
-
-			wait_till_ready(flash);
-
-			write_enable(flash);
-
-			spi_sync(flash->spi, &m);
-
-			*retlen += m.actual_length - m25p_cmdsz(flash);
-		}
-	}
-
-	mutex_unlock(&flash->lock);
-
-	return 0;
-}
-
-static int sst_write(struct mtd_info *mtd, loff_t to, size_t len,
-		size_t *retlen, const u_char *buf)
-{
-	struct m25p *flash = mtd_to_m25p(mtd);
-	struct spi_transfer t[2];
-	struct spi_message m;
-	size_t actual;
-	int cmd_sz, ret;
-
-	pr_debug("%s: %s to 0x%08x, len %zd\n", dev_name(&flash->spi->dev),
-			__func__, (u32)to, len);
-
-	spi_message_init(&m);
-	memset(t, 0, (sizeof t));
-
-	t[0].tx_buf = flash->command;
-	t[0].len = m25p_cmdsz(flash);
-	spi_message_add_tail(&t[0], &m);
-
-	t[1].tx_buf = buf;
-	spi_message_add_tail(&t[1], &m);
+	struct m25p *flash = nor->priv;
+	int ret;
 
-	mutex_lock(&flash->lock);
+	dev_dbg(nor->dev, "%dKiB at 0x%08x\n",
+		flash->mtd.erasesize / 1024, (u32)offset);
 
 	/* Wait until finished previous write command. */
-	ret = wait_till_ready(flash);
+	ret = nor->wait_till_ready(nor);
 	if (ret)
-		goto time_out;
-
-	write_enable(flash);
-
-	actual = to % 2;
-	/* Start write from odd address. */
-	if (actual) {
-		flash->command[0] = OPCODE_BP;
-		m25p_addr2cmd(flash, to, flash->command);
-
-		/* write one byte. */
-		t[1].len = 1;
-		spi_sync(flash->spi, &m);
-		ret = wait_till_ready(flash);
-		if (ret)
-			goto time_out;
-		*retlen += m.actual_length - m25p_cmdsz(flash);
-	}
-	to += actual;
-
-	flash->command[0] = OPCODE_AAI_WP;
-	m25p_addr2cmd(flash, to, flash->command);
-
-	/* Write out most of the data here. */
-	cmd_sz = m25p_cmdsz(flash);
-	for (; actual < len - 1; actual += 2) {
-		t[0].len = cmd_sz;
-		/* write two bytes. */
-		t[1].len = 2;
-		t[1].tx_buf = buf + actual;
+		return ret;
 
-		spi_sync(flash->spi, &m);
-		ret = wait_till_ready(flash);
-		if (ret)
-			goto time_out;
-		*retlen += m.actual_length - cmd_sz;
-		cmd_sz = 1;
-		to += 2;
-	}
-	write_disable(flash);
-	ret = wait_till_ready(flash);
+	/* Send write enable, then erase commands. */
+	ret = nor->write_reg(nor, SPINOR_OP_WREN, NULL, 0, 0);
 	if (ret)
-		goto time_out;
-
-	/* Write out trailing byte if it exists. */
-	if (actual != len) {
-		write_enable(flash);
-		flash->command[0] = OPCODE_BP;
-		m25p_addr2cmd(flash, to, flash->command);
-		t[0].len = m25p_cmdsz(flash);
-		t[1].len = 1;
-		t[1].tx_buf = buf + actual;
-
-		spi_sync(flash->spi, &m);
-		ret = wait_till_ready(flash);
-		if (ret)
-			goto time_out;
-		*retlen += m.actual_length - m25p_cmdsz(flash);
-		write_disable(flash);
-	}
-
-time_out:
-	mutex_unlock(&flash->lock);
-	return ret;
-}
-
-static int m25p80_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
-{
-	struct m25p *flash = mtd_to_m25p(mtd);
-	uint32_t offset = ofs;
-	uint8_t status_old, status_new;
-	int res = 0;
-
-	mutex_lock(&flash->lock);
-	/* Wait until finished previous command */
-	if (wait_till_ready(flash)) {
-		res = 1;
-		goto err;
-	}
-
-	status_old = read_sr(flash);
-
-	if (offset < flash->mtd.size-(flash->mtd.size/2))
-		status_new = status_old | SR_BP2 | SR_BP1 | SR_BP0;
-	else if (offset < flash->mtd.size-(flash->mtd.size/4))
-		status_new = (status_old & ~SR_BP0) | SR_BP2 | SR_BP1;
-	else if (offset < flash->mtd.size-(flash->mtd.size/8))
-		status_new = (status_old & ~SR_BP1) | SR_BP2 | SR_BP0;
-	else if (offset < flash->mtd.size-(flash->mtd.size/16))
-		status_new = (status_old & ~(SR_BP0|SR_BP1)) | SR_BP2;
-	else if (offset < flash->mtd.size-(flash->mtd.size/32))
-		status_new = (status_old & ~SR_BP2) | SR_BP1 | SR_BP0;
-	else if (offset < flash->mtd.size-(flash->mtd.size/64))
-		status_new = (status_old & ~(SR_BP2|SR_BP0)) | SR_BP1;
-	else
-		status_new = (status_old & ~(SR_BP2|SR_BP1)) | SR_BP0;
-
-	/* Only modify protection if it will not unlock other areas */
-	if ((status_new&(SR_BP2|SR_BP1|SR_BP0)) >
-					(status_old&(SR_BP2|SR_BP1|SR_BP0))) {
-		write_enable(flash);
-		if (write_sr(flash, status_new) < 0) {
-			res = 1;
-			goto err;
-		}
-	}
-
-err:	mutex_unlock(&flash->lock);
-	return res;
-}
-
-static int m25p80_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
-{
-	struct m25p *flash = mtd_to_m25p(mtd);
-	uint32_t offset = ofs;
-	uint8_t status_old, status_new;
-	int res = 0;
-
-	mutex_lock(&flash->lock);
-	/* Wait until finished previous command */
-	if (wait_till_ready(flash)) {
-		res = 1;
-		goto err;
-	}
-
-	status_old = read_sr(flash);
-
-	if (offset+len > flash->mtd.size-(flash->mtd.size/64))
-		status_new = status_old & ~(SR_BP2|SR_BP1|SR_BP0);
-	else if (offset+len > flash->mtd.size-(flash->mtd.size/32))
-		status_new = (status_old & ~(SR_BP2|SR_BP1)) | SR_BP0;
-	else if (offset+len > flash->mtd.size-(flash->mtd.size/16))
-		status_new = (status_old & ~(SR_BP2|SR_BP0)) | SR_BP1;
-	else if (offset+len > flash->mtd.size-(flash->mtd.size/8))
-		status_new = (status_old & ~SR_BP2) | SR_BP1 | SR_BP0;
-	else if (offset+len > flash->mtd.size-(flash->mtd.size/4))
-		status_new = (status_old & ~(SR_BP0|SR_BP1)) | SR_BP2;
-	else if (offset+len > flash->mtd.size-(flash->mtd.size/2))
-		status_new = (status_old & ~SR_BP1) | SR_BP2 | SR_BP0;
-	else
-		status_new = (status_old & ~SR_BP0) | SR_BP2 | SR_BP1;
-
-	/* Only modify protection if it will not lock other areas */
-	if ((status_new&(SR_BP2|SR_BP1|SR_BP0)) <
-					(status_old&(SR_BP2|SR_BP1|SR_BP0))) {
-		write_enable(flash);
-		if (write_sr(flash, status_new) < 0) {
-			res = 1;
-			goto err;
-		}
-	}
-
-err:	mutex_unlock(&flash->lock);
-	return res;
-}
-
-/****************************************************************************/
-
-/*
- * SPI device driver setup and teardown
- */
-
-struct flash_info {
-	/* JEDEC id zero means "no ID" (most older chips); otherwise it has
-	 * a high byte of zero plus three data bytes: the manufacturer id,
-	 * then a two byte device id.
-	 */
-	u32		jedec_id;
-	u16             ext_id;
-
-	/* The size listed here is what works with OPCODE_SE, which isn't
-	 * necessarily called a "sector" by the vendor.
-	 */
-	unsigned	sector_size;
-	u16		n_sectors;
-
-	u16		page_size;
-	u16		addr_width;
-
-	u16		flags;
-#define	SECT_4K		0x01		/* OPCODE_BE_4K works uniformly */
-#define	M25P_NO_ERASE	0x02		/* No erase command needed */
-#define	SST_WRITE	0x04		/* use SST byte programming */
-#define	M25P_NO_FR	0x08		/* Can't do fastread */
-#define	SECT_4K_PMC	0x10		/* OPCODE_BE_4K_PMC works uniformly */
-#define	M25P80_DUAL_READ	0x20    /* Flash supports Dual Read */
-#define	M25P80_QUAD_READ	0x40    /* Flash supports Quad Read */
-};
-
-#define INFO(_jedec_id, _ext_id, _sector_size, _n_sectors, _flags)	\
-	((kernel_ulong_t)&(struct flash_info) {				\
-		.jedec_id = (_jedec_id),				\
-		.ext_id = (_ext_id),					\
-		.sector_size = (_sector_size),				\
-		.n_sectors = (_n_sectors),				\
-		.page_size = 256,					\
-		.flags = (_flags),					\
-	})
-
-#define CAT25_INFO(_sector_size, _n_sectors, _page_size, _addr_width, _flags)	\
-	((kernel_ulong_t)&(struct flash_info) {				\
-		.sector_size = (_sector_size),				\
-		.n_sectors = (_n_sectors),				\
-		.page_size = (_page_size),				\
-		.addr_width = (_addr_width),				\
-		.flags = (_flags),					\
-	})
-
-/* NOTE: double check command sets and memory organization when you add
- * more flash chips.  This current list focusses on newer chips, which
- * have been converging on command sets which including JEDEC ID.
- */
-static const struct spi_device_id m25p_ids[] = {
-	/* Atmel -- some are (confusingly) marketed as "DataFlash" */
-	{ "at25fs010",  INFO(0x1f6601, 0, 32 * 1024,   4, SECT_4K) },
-	{ "at25fs040",  INFO(0x1f6604, 0, 64 * 1024,   8, SECT_4K) },
-
-	{ "at25df041a", INFO(0x1f4401, 0, 64 * 1024,   8, SECT_4K) },
-	{ "at25df321a", INFO(0x1f4701, 0, 64 * 1024,  64, SECT_4K) },
-	{ "at25df641",  INFO(0x1f4800, 0, 64 * 1024, 128, SECT_4K) },
-
-	{ "at26f004",   INFO(0x1f0400, 0, 64 * 1024,  8, SECT_4K) },
-	{ "at26df081a", INFO(0x1f4501, 0, 64 * 1024, 16, SECT_4K) },
-	{ "at26df161a", INFO(0x1f4601, 0, 64 * 1024, 32, SECT_4K) },
-	{ "at26df321",  INFO(0x1f4700, 0, 64 * 1024, 64, SECT_4K) },
-
-	{ "at45db081d", INFO(0x1f2500, 0, 64 * 1024, 16, SECT_4K) },
-
-	/* EON -- en25xxx */
-	{ "en25f32",    INFO(0x1c3116, 0, 64 * 1024,   64, SECT_4K) },
-	{ "en25p32",    INFO(0x1c2016, 0, 64 * 1024,   64, 0) },
-	{ "en25q32b",   INFO(0x1c3016, 0, 64 * 1024,   64, 0) },
-	{ "en25p64",    INFO(0x1c2017, 0, 64 * 1024,  128, 0) },
-	{ "en25q64",    INFO(0x1c3017, 0, 64 * 1024,  128, SECT_4K) },
-	{ "en25qh256",  INFO(0x1c7019, 0, 64 * 1024,  512, 0) },
-
-	/* ESMT */
-	{ "f25l32pa", INFO(0x8c2016, 0, 64 * 1024, 64, SECT_4K) },
-
-	/* Everspin */
-	{ "mr25h256", CAT25_INFO( 32 * 1024, 1, 256, 2, M25P_NO_ERASE | M25P_NO_FR) },
-	{ "mr25h10",  CAT25_INFO(128 * 1024, 1, 256, 3, M25P_NO_ERASE | M25P_NO_FR) },
-
-	/* GigaDevice */
-	{ "gd25q32", INFO(0xc84016, 0, 64 * 1024,  64, SECT_4K) },
-	{ "gd25q64", INFO(0xc84017, 0, 64 * 1024, 128, SECT_4K) },
-
-	/* Intel/Numonyx -- xxxs33b */
-	{ "160s33b",  INFO(0x898911, 0, 64 * 1024,  32, 0) },
-	{ "320s33b",  INFO(0x898912, 0, 64 * 1024,  64, 0) },
-	{ "640s33b",  INFO(0x898913, 0, 64 * 1024, 128, 0) },
-
-	/* Macronix */
-	{ "mx25l2005a",  INFO(0xc22012, 0, 64 * 1024,   4, SECT_4K) },
-	{ "mx25l4005a",  INFO(0xc22013, 0, 64 * 1024,   8, SECT_4K) },
-	{ "mx25l8005",   INFO(0xc22014, 0, 64 * 1024,  16, 0) },
-	{ "mx25l1606e",  INFO(0xc22015, 0, 64 * 1024,  32, SECT_4K) },
-	{ "mx25l3205d",  INFO(0xc22016, 0, 64 * 1024,  64, 0) },
-	{ "mx25l3255e",  INFO(0xc29e16, 0, 64 * 1024,  64, SECT_4K) },
-	{ "mx25l6405d",  INFO(0xc22017, 0, 64 * 1024, 128, 0) },
-	{ "mx25l12805d", INFO(0xc22018, 0, 64 * 1024, 256, 0) },
-	{ "mx25l12855e", INFO(0xc22618, 0, 64 * 1024, 256, 0) },
-	{ "mx25l25635e", INFO(0xc22019, 0, 64 * 1024, 512, 0) },
-	{ "mx25l25655e", INFO(0xc22619, 0, 64 * 1024, 512, 0) },
-	{ "mx66l51235l", INFO(0xc2201a, 0, 64 * 1024, 1024, M25P80_QUAD_READ) },
-	{ "mx66l1g55g",  INFO(0xc2261b, 0, 64 * 1024, 2048, M25P80_QUAD_READ) },
-
-	/* Micron */
-	{ "n25q064",     INFO(0x20ba17, 0, 64 * 1024,  128, 0) },
-	{ "n25q128a11",  INFO(0x20bb18, 0, 64 * 1024,  256, 0) },
-	{ "n25q128a13",  INFO(0x20ba18, 0, 64 * 1024,  256, 0) },
-	{ "n25q256a",    INFO(0x20ba19, 0, 64 * 1024,  512, SECT_4K) },
-	{ "n25q512a",    INFO(0x20bb20, 0, 64 * 1024, 1024, SECT_4K) },
-
-	/* PMC */
-	{ "pm25lv512",   INFO(0,        0, 32 * 1024,    2, SECT_4K_PMC) },
-	{ "pm25lv010",   INFO(0,        0, 32 * 1024,    4, SECT_4K_PMC) },
-	{ "pm25lq032",   INFO(0x7f9d46, 0, 64 * 1024,   64, SECT_4K) },
-
-	/* Spansion -- single (large) sector size only, at least
-	 * for the chips listed here (without boot sectors).
-	 */
-	{ "s25sl032p",  INFO(0x010215, 0x4d00,  64 * 1024,  64, 0) },
-	{ "s25sl064p",  INFO(0x010216, 0x4d00,  64 * 1024, 128, 0) },
-	{ "s25fl256s0", INFO(0x010219, 0x4d00, 256 * 1024, 128, 0) },
-	{ "s25fl256s1", INFO(0x010219, 0x4d01,  64 * 1024, 512, M25P80_DUAL_READ | M25P80_QUAD_READ) },
-	{ "s25fl512s",  INFO(0x010220, 0x4d00, 256 * 1024, 256, M25P80_DUAL_READ | M25P80_QUAD_READ) },
-	{ "s70fl01gs",  INFO(0x010221, 0x4d00, 256 * 1024, 256, 0) },
-	{ "s25sl12800", INFO(0x012018, 0x0300, 256 * 1024,  64, 0) },
-	{ "s25sl12801", INFO(0x012018, 0x0301,  64 * 1024, 256, 0) },
-	{ "s25fl129p0", INFO(0x012018, 0x4d00, 256 * 1024,  64, 0) },
-	{ "s25fl129p1", INFO(0x012018, 0x4d01,  64 * 1024, 256, 0) },
-	{ "s25sl004a",  INFO(0x010212,      0,  64 * 1024,   8, 0) },
-	{ "s25sl008a",  INFO(0x010213,      0,  64 * 1024,  16, 0) },
-	{ "s25sl016a",  INFO(0x010214,      0,  64 * 1024,  32, 0) },
-	{ "s25sl032a",  INFO(0x010215,      0,  64 * 1024,  64, 0) },
-	{ "s25sl064a",  INFO(0x010216,      0,  64 * 1024, 128, 0) },
-	{ "s25fl008k",  INFO(0xef4014,      0,  64 * 1024,  16, SECT_4K) },
-	{ "s25fl016k",  INFO(0xef4015,      0,  64 * 1024,  32, SECT_4K) },
-	{ "s25fl064k",  INFO(0xef4017,      0,  64 * 1024, 128, SECT_4K) },
-
-	/* SST -- large erase sizes are "overlays", "sectors" are 4K */
-	{ "sst25vf040b", INFO(0xbf258d, 0, 64 * 1024,  8, SECT_4K | SST_WRITE) },
-	{ "sst25vf080b", INFO(0xbf258e, 0, 64 * 1024, 16, SECT_4K | SST_WRITE) },
-	{ "sst25vf016b", INFO(0xbf2541, 0, 64 * 1024, 32, SECT_4K | SST_WRITE) },
-	{ "sst25vf032b", INFO(0xbf254a, 0, 64 * 1024, 64, SECT_4K | SST_WRITE) },
-	{ "sst25vf064c", INFO(0xbf254b, 0, 64 * 1024, 128, SECT_4K) },
-	{ "sst25wf512",  INFO(0xbf2501, 0, 64 * 1024,  1, SECT_4K | SST_WRITE) },
-	{ "sst25wf010",  INFO(0xbf2502, 0, 64 * 1024,  2, SECT_4K | SST_WRITE) },
-	{ "sst25wf020",  INFO(0xbf2503, 0, 64 * 1024,  4, SECT_4K | SST_WRITE) },
-	{ "sst25wf040",  INFO(0xbf2504, 0, 64 * 1024,  8, SECT_4K | SST_WRITE) },
-
-	/* ST Microelectronics -- newer production may have feature updates */
-	{ "m25p05",  INFO(0x202010,  0,  32 * 1024,   2, 0) },
-	{ "m25p10",  INFO(0x202011,  0,  32 * 1024,   4, 0) },
-	{ "m25p20",  INFO(0x202012,  0,  64 * 1024,   4, 0) },
-	{ "m25p40",  INFO(0x202013,  0,  64 * 1024,   8, 0) },
-	{ "m25p80",  INFO(0x202014,  0,  64 * 1024,  16, 0) },
-	{ "m25p16",  INFO(0x202015,  0,  64 * 1024,  32, 0) },
-	{ "m25p32",  INFO(0x202016,  0,  64 * 1024,  64, 0) },
-	{ "m25p64",  INFO(0x202017,  0,  64 * 1024, 128, 0) },
-	{ "m25p128", INFO(0x202018,  0, 256 * 1024,  64, 0) },
-	{ "n25q032", INFO(0x20ba16,  0,  64 * 1024,  64, 0) },
-
-	{ "m25p05-nonjedec",  INFO(0, 0,  32 * 1024,   2, 0) },
-	{ "m25p10-nonjedec",  INFO(0, 0,  32 * 1024,   4, 0) },
-	{ "m25p20-nonjedec",  INFO(0, 0,  64 * 1024,   4, 0) },
-	{ "m25p40-nonjedec",  INFO(0, 0,  64 * 1024,   8, 0) },
-	{ "m25p80-nonjedec",  INFO(0, 0,  64 * 1024,  16, 0) },
-	{ "m25p16-nonjedec",  INFO(0, 0,  64 * 1024,  32, 0) },
-	{ "m25p32-nonjedec",  INFO(0, 0,  64 * 1024,  64, 0) },
-	{ "m25p64-nonjedec",  INFO(0, 0,  64 * 1024, 128, 0) },
-	{ "m25p128-nonjedec", INFO(0, 0, 256 * 1024,  64, 0) },
-
-	{ "m45pe10", INFO(0x204011,  0, 64 * 1024,    2, 0) },
-	{ "m45pe80", INFO(0x204014,  0, 64 * 1024,   16, 0) },
-	{ "m45pe16", INFO(0x204015,  0, 64 * 1024,   32, 0) },
-
-	{ "m25pe20", INFO(0x208012,  0, 64 * 1024,  4,       0) },
-	{ "m25pe80", INFO(0x208014,  0, 64 * 1024, 16,       0) },
-	{ "m25pe16", INFO(0x208015,  0, 64 * 1024, 32, SECT_4K) },
-
-	{ "m25px16",    INFO(0x207115,  0, 64 * 1024, 32, SECT_4K) },
-	{ "m25px32",    INFO(0x207116,  0, 64 * 1024, 64, SECT_4K) },
-	{ "m25px32-s0", INFO(0x207316,  0, 64 * 1024, 64, SECT_4K) },
-	{ "m25px32-s1", INFO(0x206316,  0, 64 * 1024, 64, SECT_4K) },
-	{ "m25px64",    INFO(0x207117,  0, 64 * 1024, 128, 0) },
-
-	/* Winbond -- w25x "blocks" are 64K, "sectors" are 4KiB */
-	{ "w25x10", INFO(0xef3011, 0, 64 * 1024,  2,  SECT_4K) },
-	{ "w25x20", INFO(0xef3012, 0, 64 * 1024,  4,  SECT_4K) },
-	{ "w25x40", INFO(0xef3013, 0, 64 * 1024,  8,  SECT_4K) },
-	{ "w25x80", INFO(0xef3014, 0, 64 * 1024,  16, SECT_4K) },
-	{ "w25x16", INFO(0xef3015, 0, 64 * 1024,  32, SECT_4K) },
-	{ "w25x32", INFO(0xef3016, 0, 64 * 1024,  64, SECT_4K) },
-	{ "w25q32", INFO(0xef4016, 0, 64 * 1024,  64, SECT_4K) },
-	{ "w25q32dw", INFO(0xef6016, 0, 64 * 1024,  64, SECT_4K) },
-	{ "w25x64", INFO(0xef3017, 0, 64 * 1024, 128, SECT_4K) },
-	{ "w25q64", INFO(0xef4017, 0, 64 * 1024, 128, SECT_4K) },
-	{ "w25q128", INFO(0xef4018, 0, 64 * 1024, 256, SECT_4K) },
-	{ "w25q80", INFO(0xef5014, 0, 64 * 1024,  16, SECT_4K) },
-	{ "w25q80bl", INFO(0xef4014, 0, 64 * 1024,  16, SECT_4K) },
-	{ "w25q128", INFO(0xef4018, 0, 64 * 1024, 256, SECT_4K) },
-	{ "w25q256", INFO(0xef4019, 0, 64 * 1024, 512, SECT_4K) },
-
-	/* Catalyst / On Semiconductor -- non-JEDEC */
-	{ "cat25c11", CAT25_INFO(  16, 8, 16, 1, M25P_NO_ERASE | M25P_NO_FR) },
-	{ "cat25c03", CAT25_INFO(  32, 8, 16, 2, M25P_NO_ERASE | M25P_NO_FR) },
-	{ "cat25c09", CAT25_INFO( 128, 8, 32, 2, M25P_NO_ERASE | M25P_NO_FR) },
-	{ "cat25c17", CAT25_INFO( 256, 8, 32, 2, M25P_NO_ERASE | M25P_NO_FR) },
-	{ "cat25128", CAT25_INFO(2048, 8, 64, 2, M25P_NO_ERASE | M25P_NO_FR) },
-	{ },
-};
-MODULE_DEVICE_TABLE(spi, m25p_ids);
-
-static const struct spi_device_id *jedec_probe(struct spi_device *spi)
-{
-	int			tmp;
-	u8			code = OPCODE_RDID;
-	u8			id[5];
-	u32			jedec;
-	u16                     ext_jedec;
-	struct flash_info	*info;
+		return ret;
 
-	/* JEDEC also defines an optional "extended device information"
-	 * string for after vendor-specific data, after the three bytes
-	 * we use here.  Supporting some chips might require using it.
-	 */
-	tmp = spi_write_then_read(spi, &code, 1, id, 5);
-	if (tmp < 0) {
-		pr_debug("%s: error %d reading JEDEC ID\n",
-				dev_name(&spi->dev), tmp);
-		return ERR_PTR(tmp);
-	}
-	jedec = id[0];
-	jedec = jedec << 8;
-	jedec |= id[1];
-	jedec = jedec << 8;
-	jedec |= id[2];
+	/* Set up command buffer. */
+	flash->command[0] = nor->erase_opcode;
+	m25p_addr2cmd(nor, offset, flash->command);
 
-	ext_jedec = id[3] << 8 | id[4];
+	spi_write(flash->spi, flash->command, m25p_cmdsz(nor));
 
-	for (tmp = 0; tmp < ARRAY_SIZE(m25p_ids) - 1; tmp++) {
-		info = (void *)m25p_ids[tmp].driver_data;
-		if (info->jedec_id == jedec) {
-			if (info->ext_id == 0 || info->ext_id == ext_jedec)
-				return &m25p_ids[tmp];
-		}
-	}
-	dev_err(&spi->dev, "unrecognized JEDEC id %06x\n", jedec);
-	return ERR_PTR(-ENODEV);
+	return 0;
 }
 
-
 /*
  * board specific setup should have ensured the SPI clock used here
  * matches what the READ command supports, at least until this driver
@@ -1096,231 +191,45 @@ static const struct spi_device_id *jedec_probe(struct spi_device *spi)
  */
 static int m25p_probe(struct spi_device *spi)
 {
-	const struct spi_device_id	*id = spi_get_device_id(spi);
-	struct flash_platform_data	*data;
-	struct m25p			*flash;
-	struct flash_info		*info;
-	unsigned			i;
 	struct mtd_part_parser_data	ppdata;
-	struct device_node *np = spi->dev.of_node;
+	struct flash_platform_data	*data;
+	struct m25p *flash;
+	struct spi_nor *nor;
+	enum read_mode mode = SPI_NOR_NORMAL;
 	int ret;
 
-	/* Platform data helps sort out which chip type we have, as
-	 * well as how this board partitions it.  If we don't have
-	 * a chip ID, try the JEDEC id commands; they'll work for most
-	 * newer chips, even if we don't recognize the particular chip.
-	 */
-	data = dev_get_platdata(&spi->dev);
-	if (data && data->type) {
-		const struct spi_device_id *plat_id;
-
-		for (i = 0; i < ARRAY_SIZE(m25p_ids) - 1; i++) {
-			plat_id = &m25p_ids[i];
-			if (strcmp(data->type, plat_id->name))
-				continue;
-			break;
-		}
-
-		if (i < ARRAY_SIZE(m25p_ids) - 1)
-			id = plat_id;
-		else
-			dev_warn(&spi->dev, "unrecognized id %s\n", data->type);
-	}
-
-	info = (void *)id->driver_data;
-
-	if (info->jedec_id) {
-		const struct spi_device_id *jid;
-
-		jid = jedec_probe(spi);
-		if (IS_ERR(jid)) {
-			return PTR_ERR(jid);
-		} else if (jid != id) {
-			/*
-			 * JEDEC knows better, so overwrite platform ID. We
-			 * can't trust partitions any longer, but we'll let
-			 * mtd apply them anyway, since some partitions may be
-			 * marked read-only, and we don't want to lose that
-			 * information, even if it's not 100% accurate.
-			 */
-			dev_warn(&spi->dev, "found %s, expected %s\n",
-				 jid->name, id->name);
-			id = jid;
-			info = (void *)jid->driver_data;
-		}
-	}
-
 	flash = devm_kzalloc(&spi->dev, sizeof(*flash), GFP_KERNEL);
 	if (!flash)
 		return -ENOMEM;
 
-	flash->command = devm_kzalloc(&spi->dev, MAX_CMD_SIZE, GFP_KERNEL);
-	if (!flash->command)
-		return -ENOMEM;
-
-	flash->spi = spi;
-	mutex_init(&flash->lock);
-	spi_set_drvdata(spi, flash);
-
-	/*
-	 * Atmel, SST and Intel/Numonyx serial flash tend to power
-	 * up with the software protection bits set
-	 */
+	nor = &flash->spi_nor;
 
-	if (JEDEC_MFR(info->jedec_id) == CFI_MFR_ATMEL ||
-	    JEDEC_MFR(info->jedec_id) == CFI_MFR_INTEL ||
-	    JEDEC_MFR(info->jedec_id) == CFI_MFR_SST) {
-		write_enable(flash);
-		write_sr(flash, 0);
-	}
-
-	if (data && data->name)
-		flash->mtd.name = data->name;
-	else
-		flash->mtd.name = dev_name(&spi->dev);
-
-	flash->mtd.type = MTD_NORFLASH;
-	flash->mtd.writesize = 1;
-	flash->mtd.flags = MTD_CAP_NORFLASH;
-	flash->mtd.size = info->sector_size * info->n_sectors;
-	flash->mtd._erase = m25p80_erase;
-	flash->mtd._read = m25p80_read;
-
-	/* flash protection support for STmicro chips */
-	if (JEDEC_MFR(info->jedec_id) == CFI_MFR_ST) {
-		flash->mtd._lock = m25p80_lock;
-		flash->mtd._unlock = m25p80_unlock;
-	}
+	/* install the hooks */
+	nor->read = m25p80_read;
+	nor->write = m25p80_write;
+	nor->erase = m25p80_erase;
+	nor->write_reg = m25p80_write_reg;
+	nor->read_reg = m25p80_read_reg;
 
-	/* sst flash chips use AAI word program */
-	if (info->flags & SST_WRITE)
-		flash->mtd._write = sst_write;
-	else
-		flash->mtd._write = m25p80_write;
+	nor->dev = &spi->dev;
+	nor->mtd = &flash->mtd;
+	nor->priv = flash;
 
-	/* prefer "small sector" erase if possible */
-	if (info->flags & SECT_4K) {
-		flash->erase_opcode = OPCODE_BE_4K;
-		flash->mtd.erasesize = 4096;
-	} else if (info->flags & SECT_4K_PMC) {
-		flash->erase_opcode = OPCODE_BE_4K_PMC;
-		flash->mtd.erasesize = 4096;
-	} else {
-		flash->erase_opcode = OPCODE_SE;
-		flash->mtd.erasesize = info->sector_size;
-	}
+	spi_set_drvdata(spi, flash);
+	flash->mtd.priv = nor;
+	flash->spi = spi;
 
-	if (info->flags & M25P_NO_ERASE)
-		flash->mtd.flags |= MTD_NO_ERASE;
+	if (spi->mode & SPI_RX_QUAD)
+		mode = SPI_NOR_QUAD;
+	else if (spi->mode & SPI_RX_DUAL)
+		mode = SPI_NOR_DUAL;
+	ret = spi_nor_scan(nor, spi_get_device_id(spi), mode);
+	if (ret)
+		return ret;
 
+	data = dev_get_platdata(&spi->dev);
 	ppdata.of_node = spi->dev.of_node;
-	flash->mtd.dev.parent = &spi->dev;
-	flash->page_size = info->page_size;
-	flash->mtd.writebufsize = flash->page_size;
-
-	if (np) {
-		/* If we were instantiated by DT, use it */
-		if (of_property_read_bool(np, "m25p,fast-read"))
-			flash->flash_read = M25P80_FAST;
-		else
-			flash->flash_read = M25P80_NORMAL;
-	} else {
-		/* If we weren't instantiated by DT, default to fast-read */
-		flash->flash_read = M25P80_FAST;
-	}
-
-	/* Some devices cannot do fast-read, no matter what DT tells us */
-	if (info->flags & M25P_NO_FR)
-		flash->flash_read = M25P80_NORMAL;
-
-	/* Quad/Dual-read mode takes precedence over fast/normal */
-	if (spi->mode & SPI_RX_QUAD && info->flags & M25P80_QUAD_READ) {
-		ret = set_quad_mode(flash, info->jedec_id);
-		if (ret) {
-			dev_err(&flash->spi->dev, "quad mode not supported\n");
-			return ret;
-		}
-		flash->flash_read = M25P80_QUAD;
-	} else if (spi->mode & SPI_RX_DUAL && info->flags & M25P80_DUAL_READ) {
-		flash->flash_read = M25P80_DUAL;
-	}
 
-	/* Default commands */
-	switch (flash->flash_read) {
-	case M25P80_QUAD:
-		flash->read_opcode = OPCODE_QUAD_READ;
-		break;
-	case M25P80_DUAL:
-		flash->read_opcode = OPCODE_DUAL_READ;
-		break;
-	case M25P80_FAST:
-		flash->read_opcode = OPCODE_FAST_READ;
-		break;
-	case M25P80_NORMAL:
-		flash->read_opcode = OPCODE_NORM_READ;
-		break;
-	default:
-		dev_err(&flash->spi->dev, "No Read opcode defined\n");
-		return -EINVAL;
-	}
-
-	flash->program_opcode = OPCODE_PP;
-
-	if (info->addr_width)
-		flash->addr_width = info->addr_width;
-	else if (flash->mtd.size > 0x1000000) {
-		/* enable 4-byte addressing if the device exceeds 16MiB */
-		flash->addr_width = 4;
-		if (JEDEC_MFR(info->jedec_id) == CFI_MFR_AMD) {
-			/* Dedicated 4-byte command set */
-			switch (flash->flash_read) {
-			case M25P80_QUAD:
-				flash->read_opcode = OPCODE_QUAD_READ_4B;
-				break;
-			case M25P80_DUAL:
-				flash->read_opcode = OPCODE_DUAL_READ_4B;
-				break;
-			case M25P80_FAST:
-				flash->read_opcode = OPCODE_FAST_READ_4B;
-				break;
-			case M25P80_NORMAL:
-				flash->read_opcode = OPCODE_NORM_READ_4B;
-				break;
-			}
-			flash->program_opcode = OPCODE_PP_4B;
-			/* No small sector erase for 4-byte command set */
-			flash->erase_opcode = OPCODE_SE_4B;
-			flash->mtd.erasesize = info->sector_size;
-		} else
-			set_4byte(flash, info->jedec_id, 1);
-	} else {
-		flash->addr_width = 3;
-	}
-
-	dev_info(&spi->dev, "%s (%lld Kbytes)\n", id->name,
-			(long long)flash->mtd.size >> 10);
-
-	pr_debug("mtd .name = %s, .size = 0x%llx (%lldMiB) "
-			".erasesize = 0x%.8x (%uKiB) .numeraseregions = %d\n",
-		flash->mtd.name,
-		(long long)flash->mtd.size, (long long)(flash->mtd.size >> 20),
-		flash->mtd.erasesize, flash->mtd.erasesize / 1024,
-		flash->mtd.numeraseregions);
-
-	if (flash->mtd.numeraseregions)
-		for (i = 0; i < flash->mtd.numeraseregions; i++)
-			pr_debug("mtd.eraseregions[%d] = { .offset = 0x%llx, "
-				".erasesize = 0x%.8x (%uKiB), "
-				".numblocks = %d }\n",
-				i, (long long)flash->mtd.eraseregions[i].offset,
-				flash->mtd.eraseregions[i].erasesize,
-				flash->mtd.eraseregions[i].erasesize / 1024,
-				flash->mtd.eraseregions[i].numblocks);
-
-
-	/* partitions should match sector boundaries; and it may be good to
-	 * use readonly partitions for writeprotected sectors (BP2..BP0).
-	 */
 	return mtd_device_parse_register(&flash->mtd, NULL, &ppdata,
 			data ? data->parts : NULL,
 			data ? data->nr_parts : 0);
@@ -1341,7 +250,7 @@ static struct spi_driver m25p80_driver = {
 		.name	= "m25p80",
 		.owner	= THIS_MODULE,
 	},
-	.id_table	= m25p_ids,
+	.id_table	= spi_nor_ids,
 	.probe	= m25p_probe,
 	.remove	= m25p_remove,
 
diff --git a/drivers/mtd/devices/serial_flash_cmds.h b/drivers/mtd/devices/serial_flash_cmds.h
index 4f0c2c7c898e..f59a125295d0 100644
--- a/drivers/mtd/devices/serial_flash_cmds.h
+++ b/drivers/mtd/devices/serial_flash_cmds.h
@@ -13,43 +13,23 @@
 #define _MTD_SERIAL_FLASH_CMDS_H
 
 /* Generic Flash Commands/OPCODEs */
-#define FLASH_CMD_WREN		0x06
-#define FLASH_CMD_WRDI		0x04
-#define FLASH_CMD_RDID		0x9f
-#define FLASH_CMD_RDSR		0x05
-#define FLASH_CMD_RDSR2		0x35
-#define FLASH_CMD_WRSR		0x01
-#define FLASH_CMD_SE_4K		0x20
-#define FLASH_CMD_SE_32K	0x52
-#define FLASH_CMD_SE		0xd8
-#define FLASH_CMD_CHIPERASE	0xc7
-#define FLASH_CMD_WRVCR		0x81
-#define FLASH_CMD_RDVCR		0x85
+#define SPINOR_OP_RDSR2		0x35
+#define SPINOR_OP_WRVCR		0x81
+#define SPINOR_OP_RDVCR		0x85
 
 /* JEDEC Standard - Serial Flash Discoverable Parmeters (SFDP) Commands */
-#define FLASH_CMD_READ		0x03	/* READ */
-#define FLASH_CMD_READ_FAST	0x0b	/* FAST READ */
-#define FLASH_CMD_READ_1_1_2	0x3b	/* DUAL OUTPUT READ */
-#define FLASH_CMD_READ_1_2_2	0xbb	/* DUAL I/O READ */
-#define FLASH_CMD_READ_1_1_4	0x6b	/* QUAD OUTPUT READ */
-#define FLASH_CMD_READ_1_4_4	0xeb	/* QUAD I/O READ */
+#define SPINOR_OP_READ_1_2_2	0xbb	/* DUAL I/O READ */
+#define SPINOR_OP_READ_1_4_4	0xeb	/* QUAD I/O READ */
 
-#define FLASH_CMD_WRITE		0x02	/* PAGE PROGRAM */
-#define FLASH_CMD_WRITE_1_1_2	0xa2	/* DUAL INPUT PROGRAM */
-#define FLASH_CMD_WRITE_1_2_2	0xd2	/* DUAL INPUT EXT PROGRAM */
-#define FLASH_CMD_WRITE_1_1_4	0x32	/* QUAD INPUT PROGRAM */
-#define FLASH_CMD_WRITE_1_4_4	0x12	/* QUAD INPUT EXT PROGRAM */
-
-#define FLASH_CMD_EN4B_ADDR	0xb7	/* Enter 4-byte address mode */
-#define FLASH_CMD_EX4B_ADDR	0xe9	/* Exit 4-byte address mode */
+#define SPINOR_OP_WRITE		0x02	/* PAGE PROGRAM */
+#define SPINOR_OP_WRITE_1_1_2	0xa2	/* DUAL INPUT PROGRAM */
+#define SPINOR_OP_WRITE_1_2_2	0xd2	/* DUAL INPUT EXT PROGRAM */
+#define SPINOR_OP_WRITE_1_1_4	0x32	/* QUAD INPUT PROGRAM */
+#define SPINOR_OP_WRITE_1_4_4	0x12	/* QUAD INPUT EXT PROGRAM */
 
 /* READ commands with 32-bit addressing */
-#define FLASH_CMD_READ4		0x13
-#define FLASH_CMD_READ4_FAST	0x0c
-#define FLASH_CMD_READ4_1_1_2	0x3c
-#define FLASH_CMD_READ4_1_2_2	0xbc
-#define FLASH_CMD_READ4_1_1_4	0x6c
-#define FLASH_CMD_READ4_1_4_4	0xec
+#define SPINOR_OP_READ4_1_2_2	0xbc
+#define SPINOR_OP_READ4_1_4_4	0xec
 
 /* Configuration flags */
 #define FLASH_FLAG_SINGLE	0x000000ff
diff --git a/drivers/mtd/devices/slram.c b/drivers/mtd/devices/slram.c
index 5a5cd2ace4a6..2fc4957cbe7f 100644
--- a/drivers/mtd/devices/slram.c
+++ b/drivers/mtd/devices/slram.c
@@ -280,14 +280,11 @@ __setup("slram=", mtd_slram_setup);
 static int __init init_slram(void)
 {
 	char *devname;
-	int i;
 
 #ifndef MODULE
 	char *devstart;
 	char *devlength;
 
-	i = 0;
-
 	if (!map) {
 		E("slram: not enough parameters.\n");
 		return(-EINVAL);
@@ -314,6 +311,7 @@ static int __init init_slram(void)
 	}
 #else
 	int count;
+	int i;
 
 	for (count = 0; count < SLRAM_MAX_DEVICES_PARAMS && map[count];
 			count++) {
diff --git a/drivers/mtd/devices/st_spi_fsm.c b/drivers/mtd/devices/st_spi_fsm.c
index 1957d7c8e185..d252514d3e98 100644
--- a/drivers/mtd/devices/st_spi_fsm.c
+++ b/drivers/mtd/devices/st_spi_fsm.c
@@ -19,6 +19,7 @@
 #include <linux/mfd/syscon.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
+#include <linux/mtd/spi-nor.h>
 #include <linux/sched.h>
 #include <linux/delay.h>
 #include <linux/io.h>
@@ -201,44 +202,6 @@
 
 #define STFSM_MAX_WAIT_SEQ_MS  1000     /* FSM execution time */
 
-/* Flash Commands */
-#define FLASH_CMD_WREN         0x06
-#define FLASH_CMD_WRDI         0x04
-#define FLASH_CMD_RDID         0x9f
-#define FLASH_CMD_RDSR         0x05
-#define FLASH_CMD_RDSR2                0x35
-#define FLASH_CMD_WRSR         0x01
-#define FLASH_CMD_SE_4K                0x20
-#define FLASH_CMD_SE_32K       0x52
-#define FLASH_CMD_SE           0xd8
-#define FLASH_CMD_CHIPERASE    0xc7
-#define FLASH_CMD_WRVCR                0x81
-#define FLASH_CMD_RDVCR                0x85
-
-#define FLASH_CMD_READ         0x03    /* READ */
-#define FLASH_CMD_READ_FAST    0x0b    /* FAST READ */
-#define FLASH_CMD_READ_1_1_2   0x3b    /* DUAL OUTPUT READ */
-#define FLASH_CMD_READ_1_2_2   0xbb    /* DUAL I/O READ */
-#define FLASH_CMD_READ_1_1_4   0x6b    /* QUAD OUTPUT READ */
-#define FLASH_CMD_READ_1_4_4   0xeb    /* QUAD I/O READ */
-
-#define FLASH_CMD_WRITE                0x02    /* PAGE PROGRAM */
-#define FLASH_CMD_WRITE_1_1_2  0xa2    /* DUAL INPUT PROGRAM */
-#define FLASH_CMD_WRITE_1_2_2  0xd2    /* DUAL INPUT EXT PROGRAM */
-#define FLASH_CMD_WRITE_1_1_4  0x32    /* QUAD INPUT PROGRAM */
-#define FLASH_CMD_WRITE_1_4_4  0x12    /* QUAD INPUT EXT PROGRAM */
-
-#define FLASH_CMD_EN4B_ADDR    0xb7    /* Enter 4-byte address mode */
-#define FLASH_CMD_EX4B_ADDR    0xe9    /* Exit 4-byte address mode */
-
-/* READ commands with 32-bit addressing (N25Q256 and S25FLxxxS) */
-#define FLASH_CMD_READ4                0x13
-#define FLASH_CMD_READ4_FAST   0x0c
-#define FLASH_CMD_READ4_1_1_2  0x3c
-#define FLASH_CMD_READ4_1_2_2  0xbc
-#define FLASH_CMD_READ4_1_1_4  0x6c
-#define FLASH_CMD_READ4_1_4_4  0xec
-
 /* S25FLxxxS commands */
 #define S25FL_CMD_WRITE4_1_1_4 0x34
 #define S25FL_CMD_SE4          0xdc
@@ -246,7 +209,7 @@
 #define S25FL_CMD_DYBWR                0xe1
 #define S25FL_CMD_DYBRD                0xe0
 #define S25FL_CMD_WRITE4       0x12    /* Note, opcode clashes with
-					* 'FLASH_CMD_WRITE_1_4_4'
+					* 'SPINOR_OP_WRITE_1_4_4'
 					* as found on N25Qxxx devices! */
 
 /* Status register */
@@ -261,6 +224,12 @@
 #define S25FL_STATUS_E_ERR     0x20
 #define S25FL_STATUS_P_ERR     0x40
 
+#define N25Q_CMD_WRVCR         0x81
+#define N25Q_CMD_RDVCR         0x85
+#define N25Q_CMD_RDVECR        0x65
+#define N25Q_CMD_RDNVCR        0xb5
+#define N25Q_CMD_WRNVCR        0xb1
+
 #define FLASH_PAGESIZE         256			/* In Bytes    */
 #define FLASH_PAGESIZE_32      (FLASH_PAGESIZE / 4)	/* In uint32_t */
 #define FLASH_MAX_BUSY_WAIT    (300 * HZ)	/* Maximum 'CHIPERASE' time */
@@ -270,7 +239,6 @@
  */
 #define CFG_READ_TOGGLE_32BIT_ADDR     0x00000001
 #define CFG_WRITE_TOGGLE_32BIT_ADDR    0x00000002
-#define CFG_WRITE_EX_32BIT_ADDR_DELAY  0x00000004
 #define CFG_ERASESEC_TOGGLE_32BIT_ADDR 0x00000008
 #define CFG_S25FL_CHECK_ERROR_FLAGS    0x00000010
 
@@ -329,7 +297,7 @@ struct flash_info {
 	u32             jedec_id;
 	u16             ext_id;
 	/*
-	 * The size listed here is what works with FLASH_CMD_SE, which isn't
+	 * The size listed here is what works with SPINOR_OP_SE, which isn't
 	 * necessarily called a "sector" by the vendor.
 	 */
 	unsigned        sector_size;
@@ -369,17 +337,26 @@ static struct flash_info flash_types[] = {
 	{ "m25px32", 0x207116, 0,  64 * 1024,  64, M25PX_FLAG, 75, NULL },
 	{ "m25px64", 0x207117, 0,  64 * 1024, 128, M25PX_FLAG, 75, NULL },
 
+	/* Macronix MX25xxx
+	 *     - Support for 'FLASH_FLAG_WRITE_1_4_4' is omitted for devices
+	 *       where operating frequency must be reduced.
+	 */
 #define MX25_FLAG (FLASH_FLAG_READ_WRITE       |	\
 		   FLASH_FLAG_READ_FAST         |	\
 		   FLASH_FLAG_READ_1_1_2        |	\
 		   FLASH_FLAG_READ_1_2_2        |	\
 		   FLASH_FLAG_READ_1_1_4        |	\
-		   FLASH_FLAG_READ_1_4_4        |	\
 		   FLASH_FLAG_SE_4K             |	\
 		   FLASH_FLAG_SE_32K)
+	{ "mx25l3255e",  0xc29e16, 0, 64 * 1024, 64,
+	  (MX25_FLAG | FLASH_FLAG_WRITE_1_4_4), 86,
+	  stfsm_mx25_config},
 	{ "mx25l25635e", 0xc22019, 0, 64*1024, 512,
 	  (MX25_FLAG | FLASH_FLAG_32BIT_ADDR | FLASH_FLAG_RESET), 70,
 	  stfsm_mx25_config },
+	{ "mx25l25655e", 0xc22619, 0, 64*1024, 512,
+	  (MX25_FLAG | FLASH_FLAG_32BIT_ADDR | FLASH_FLAG_RESET), 70,
+	  stfsm_mx25_config},
 
 #define N25Q_FLAG (FLASH_FLAG_READ_WRITE       |	\
 		   FLASH_FLAG_READ_FAST         |	\
@@ -407,6 +384,8 @@ static struct flash_info flash_types[] = {
 			FLASH_FLAG_READ_1_4_4   |	\
 			FLASH_FLAG_WRITE_1_1_4  |	\
 			FLASH_FLAG_READ_FAST)
+	{ "s25fl032p",  0x010215, 0x4d00,  64 * 1024,  64, S25FLXXXP_FLAG, 80,
+	  stfsm_s25fl_config},
 	{ "s25fl129p0", 0x012018, 0x4d00, 256 * 1024,  64, S25FLXXXP_FLAG, 80,
 	  stfsm_s25fl_config },
 	{ "s25fl129p1", 0x012018, 0x4d01,  64 * 1024, 256, S25FLXXXP_FLAG, 80,
@@ -473,22 +452,22 @@ static struct flash_info flash_types[] = {
 
 /* Default READ configurations, in order of preference */
 static struct seq_rw_config default_read_configs[] = {
-	{FLASH_FLAG_READ_1_4_4, FLASH_CMD_READ_1_4_4,	0, 4, 4, 0x00, 2, 4},
-	{FLASH_FLAG_READ_1_1_4, FLASH_CMD_READ_1_1_4,	0, 1, 4, 0x00, 4, 0},
-	{FLASH_FLAG_READ_1_2_2, FLASH_CMD_READ_1_2_2,	0, 2, 2, 0x00, 4, 0},
-	{FLASH_FLAG_READ_1_1_2, FLASH_CMD_READ_1_1_2,	0, 1, 2, 0x00, 0, 8},
-	{FLASH_FLAG_READ_FAST,	FLASH_CMD_READ_FAST,	0, 1, 1, 0x00, 0, 8},
-	{FLASH_FLAG_READ_WRITE, FLASH_CMD_READ,		0, 1, 1, 0x00, 0, 0},
+	{FLASH_FLAG_READ_1_4_4, SPINOR_OP_READ_1_4_4,	0, 4, 4, 0x00, 2, 4},
+	{FLASH_FLAG_READ_1_1_4, SPINOR_OP_READ_1_1_4,	0, 1, 4, 0x00, 4, 0},
+	{FLASH_FLAG_READ_1_2_2, SPINOR_OP_READ_1_2_2,	0, 2, 2, 0x00, 4, 0},
+	{FLASH_FLAG_READ_1_1_2, SPINOR_OP_READ_1_1_2,	0, 1, 2, 0x00, 0, 8},
+	{FLASH_FLAG_READ_FAST,	SPINOR_OP_READ_FAST,	0, 1, 1, 0x00, 0, 8},
+	{FLASH_FLAG_READ_WRITE, SPINOR_OP_READ,		0, 1, 1, 0x00, 0, 0},
 	{0x00,			0,			0, 0, 0, 0x00, 0, 0},
 };
 
 /* Default WRITE configurations */
 static struct seq_rw_config default_write_configs[] = {
-	{FLASH_FLAG_WRITE_1_4_4, FLASH_CMD_WRITE_1_4_4, 1, 4, 4, 0x00, 0, 0},
-	{FLASH_FLAG_WRITE_1_1_4, FLASH_CMD_WRITE_1_1_4, 1, 1, 4, 0x00, 0, 0},
-	{FLASH_FLAG_WRITE_1_2_2, FLASH_CMD_WRITE_1_2_2, 1, 2, 2, 0x00, 0, 0},
-	{FLASH_FLAG_WRITE_1_1_2, FLASH_CMD_WRITE_1_1_2, 1, 1, 2, 0x00, 0, 0},
-	{FLASH_FLAG_READ_WRITE,  FLASH_CMD_WRITE,       1, 1, 1, 0x00, 0, 0},
+	{FLASH_FLAG_WRITE_1_4_4, SPINOR_OP_WRITE_1_4_4, 1, 4, 4, 0x00, 0, 0},
+	{FLASH_FLAG_WRITE_1_1_4, SPINOR_OP_WRITE_1_1_4, 1, 1, 4, 0x00, 0, 0},
+	{FLASH_FLAG_WRITE_1_2_2, SPINOR_OP_WRITE_1_2_2, 1, 2, 2, 0x00, 0, 0},
+	{FLASH_FLAG_WRITE_1_1_2, SPINOR_OP_WRITE_1_1_2, 1, 1, 2, 0x00, 0, 0},
+	{FLASH_FLAG_READ_WRITE,  SPINOR_OP_WRITE,       1, 1, 1, 0x00, 0, 0},
 	{0x00,			 0,			0, 0, 0, 0x00, 0, 0},
 };
 
@@ -511,12 +490,12 @@ static struct seq_rw_config default_write_configs[] = {
  * cycles.
  */
 static struct seq_rw_config n25q_read3_configs[] = {
-	{FLASH_FLAG_READ_1_4_4, FLASH_CMD_READ_1_4_4,	0, 4, 4, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_1_4, FLASH_CMD_READ_1_1_4,	0, 1, 4, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_2_2, FLASH_CMD_READ_1_2_2,	0, 2, 2, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_1_2, FLASH_CMD_READ_1_1_2,	0, 1, 2, 0x00, 0, 8},
-	{FLASH_FLAG_READ_FAST,	FLASH_CMD_READ_FAST,	0, 1, 1, 0x00, 0, 8},
-	{FLASH_FLAG_READ_WRITE, FLASH_CMD_READ,	        0, 1, 1, 0x00, 0, 0},
+	{FLASH_FLAG_READ_1_4_4, SPINOR_OP_READ_1_4_4,	0, 4, 4, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_1_4, SPINOR_OP_READ_1_1_4,	0, 1, 4, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_2_2, SPINOR_OP_READ_1_2_2,	0, 2, 2, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_1_2, SPINOR_OP_READ_1_1_2,	0, 1, 2, 0x00, 0, 8},
+	{FLASH_FLAG_READ_FAST,	SPINOR_OP_READ_FAST,	0, 1, 1, 0x00, 0, 8},
+	{FLASH_FLAG_READ_WRITE, SPINOR_OP_READ,	        0, 1, 1, 0x00, 0, 0},
 	{0x00,			0,			0, 0, 0, 0x00, 0, 0},
 };
 
@@ -526,12 +505,12 @@ static struct seq_rw_config n25q_read3_configs[] = {
  *	- 'FAST' variants configured for 8 dummy cycles (see note above.)
  */
 static struct seq_rw_config n25q_read4_configs[] = {
-	{FLASH_FLAG_READ_1_4_4, FLASH_CMD_READ4_1_4_4,	0, 4, 4, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_1_4, FLASH_CMD_READ4_1_1_4,	0, 1, 4, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_2_2, FLASH_CMD_READ4_1_2_2,	0, 2, 2, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_1_2, FLASH_CMD_READ4_1_1_2,	0, 1, 2, 0x00, 0, 8},
-	{FLASH_FLAG_READ_FAST,	FLASH_CMD_READ4_FAST,	0, 1, 1, 0x00, 0, 8},
-	{FLASH_FLAG_READ_WRITE, FLASH_CMD_READ4,	0, 1, 1, 0x00, 0, 0},
+	{FLASH_FLAG_READ_1_4_4, SPINOR_OP_READ4_1_4_4,	0, 4, 4, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_1_4, SPINOR_OP_READ4_1_1_4,	0, 1, 4, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_2_2, SPINOR_OP_READ4_1_2_2,	0, 2, 2, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_1_2, SPINOR_OP_READ4_1_1_2,	0, 1, 2, 0x00, 0, 8},
+	{FLASH_FLAG_READ_FAST,	SPINOR_OP_READ4_FAST,	0, 1, 1, 0x00, 0, 8},
+	{FLASH_FLAG_READ_WRITE, SPINOR_OP_READ4,	0, 1, 1, 0x00, 0, 0},
 	{0x00,			0,			0, 0, 0, 0x00, 0, 0},
 };
 
@@ -544,7 +523,7 @@ static int stfsm_mx25_en_32bit_addr_seq(struct stfsm_seq *seq)
 {
 	seq->seq_opc[0] = (SEQ_OPC_PADS_1 |
 			   SEQ_OPC_CYCLES(8) |
-			   SEQ_OPC_OPCODE(FLASH_CMD_EN4B_ADDR) |
+			   SEQ_OPC_OPCODE(SPINOR_OP_EN4B) |
 			   SEQ_OPC_CSDEASSERT);
 
 	seq->seq[0] = STFSM_INST_CMD1;
@@ -572,12 +551,12 @@ static int stfsm_mx25_en_32bit_addr_seq(struct stfsm_seq *seq)
  * entering a state that is incompatible with the SPIBoot Controller.
  */
 static struct seq_rw_config stfsm_s25fl_read4_configs[] = {
-	{FLASH_FLAG_READ_1_4_4,  FLASH_CMD_READ4_1_4_4,  0, 4, 4, 0x00, 2, 4},
-	{FLASH_FLAG_READ_1_1_4,  FLASH_CMD_READ4_1_1_4,  0, 1, 4, 0x00, 0, 8},
-	{FLASH_FLAG_READ_1_2_2,  FLASH_CMD_READ4_1_2_2,  0, 2, 2, 0x00, 4, 0},
-	{FLASH_FLAG_READ_1_1_2,  FLASH_CMD_READ4_1_1_2,  0, 1, 2, 0x00, 0, 8},
-	{FLASH_FLAG_READ_FAST,   FLASH_CMD_READ4_FAST,   0, 1, 1, 0x00, 0, 8},
-	{FLASH_FLAG_READ_WRITE,  FLASH_CMD_READ4,        0, 1, 1, 0x00, 0, 0},
+	{FLASH_FLAG_READ_1_4_4,  SPINOR_OP_READ4_1_4_4,  0, 4, 4, 0x00, 2, 4},
+	{FLASH_FLAG_READ_1_1_4,  SPINOR_OP_READ4_1_1_4,  0, 1, 4, 0x00, 0, 8},
+	{FLASH_FLAG_READ_1_2_2,  SPINOR_OP_READ4_1_2_2,  0, 2, 2, 0x00, 4, 0},
+	{FLASH_FLAG_READ_1_1_2,  SPINOR_OP_READ4_1_1_2,  0, 1, 2, 0x00, 0, 8},
+	{FLASH_FLAG_READ_FAST,   SPINOR_OP_READ4_FAST,   0, 1, 1, 0x00, 0, 8},
+	{FLASH_FLAG_READ_WRITE,  SPINOR_OP_READ4,        0, 1, 1, 0x00, 0, 0},
 	{0x00,                   0,                      0, 0, 0, 0x00, 0, 0},
 };
 
@@ -590,13 +569,13 @@ static struct seq_rw_config stfsm_s25fl_write4_configs[] = {
 /*
  * [W25Qxxx] Configuration
  */
-#define W25Q_STATUS_QE			(0x1 << 9)
+#define W25Q_STATUS_QE			(0x1 << 1)
 
 static struct stfsm_seq stfsm_seq_read_jedec = {
 	.data_size = TRANSFER_SIZE(8),
 	.seq_opc[0] = (SEQ_OPC_PADS_1 |
 		       SEQ_OPC_CYCLES(8) |
-		       SEQ_OPC_OPCODE(FLASH_CMD_RDID)),
+		       SEQ_OPC_OPCODE(SPINOR_OP_RDID)),
 	.seq = {
 		STFSM_INST_CMD1,
 		STFSM_INST_DATA_READ,
@@ -612,7 +591,7 @@ static struct stfsm_seq stfsm_seq_read_status_fifo = {
 	.data_size = TRANSFER_SIZE(4),
 	.seq_opc[0] = (SEQ_OPC_PADS_1 |
 		       SEQ_OPC_CYCLES(8) |
-		       SEQ_OPC_OPCODE(FLASH_CMD_RDSR)),
+		       SEQ_OPC_OPCODE(SPINOR_OP_RDSR)),
 	.seq = {
 		STFSM_INST_CMD1,
 		STFSM_INST_DATA_READ,
@@ -628,10 +607,10 @@ static struct stfsm_seq stfsm_seq_erase_sector = {
 	/* 'addr_cfg' configured during initialisation */
 	.seq_opc = {
 		(SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		 SEQ_OPC_OPCODE(FLASH_CMD_WREN) | SEQ_OPC_CSDEASSERT),
+		 SEQ_OPC_OPCODE(SPINOR_OP_WREN) | SEQ_OPC_CSDEASSERT),
 
 		(SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		 SEQ_OPC_OPCODE(FLASH_CMD_SE)),
+		 SEQ_OPC_OPCODE(SPINOR_OP_SE)),
 	},
 	.seq = {
 		STFSM_INST_CMD1,
@@ -649,10 +628,10 @@ static struct stfsm_seq stfsm_seq_erase_sector = {
 static struct stfsm_seq stfsm_seq_erase_chip = {
 	.seq_opc = {
 		(SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		 SEQ_OPC_OPCODE(FLASH_CMD_WREN) | SEQ_OPC_CSDEASSERT),
+		 SEQ_OPC_OPCODE(SPINOR_OP_WREN) | SEQ_OPC_CSDEASSERT),
 
 		(SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		 SEQ_OPC_OPCODE(FLASH_CMD_CHIPERASE) | SEQ_OPC_CSDEASSERT),
+		 SEQ_OPC_OPCODE(SPINOR_OP_CHIP_ERASE) | SEQ_OPC_CSDEASSERT),
 	},
 	.seq = {
 		STFSM_INST_CMD1,
@@ -669,26 +648,9 @@ static struct stfsm_seq stfsm_seq_erase_chip = {
 
 static struct stfsm_seq stfsm_seq_write_status = {
 	.seq_opc[0] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		       SEQ_OPC_OPCODE(FLASH_CMD_WREN) | SEQ_OPC_CSDEASSERT),
-	.seq_opc[1] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		       SEQ_OPC_OPCODE(FLASH_CMD_WRSR)),
-	.seq = {
-		STFSM_INST_CMD1,
-		STFSM_INST_CMD2,
-		STFSM_INST_STA_WR1,
-		STFSM_INST_STOP,
-	},
-	.seq_cfg = (SEQ_CFG_PADS_1 |
-		    SEQ_CFG_READNOTWRITE |
-		    SEQ_CFG_CSDEASSERT |
-		    SEQ_CFG_STARTSEQ),
-};
-
-static struct stfsm_seq stfsm_seq_wrvcr = {
-	.seq_opc[0] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		       SEQ_OPC_OPCODE(FLASH_CMD_WREN) | SEQ_OPC_CSDEASSERT),
+		       SEQ_OPC_OPCODE(SPINOR_OP_WREN) | SEQ_OPC_CSDEASSERT),
 	.seq_opc[1] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-		       SEQ_OPC_OPCODE(FLASH_CMD_WRVCR)),
+		       SEQ_OPC_OPCODE(SPINOR_OP_WRSR)),
 	.seq = {
 		STFSM_INST_CMD1,
 		STFSM_INST_CMD2,
@@ -704,9 +666,9 @@ static struct stfsm_seq stfsm_seq_wrvcr = {
 static int stfsm_n25q_en_32bit_addr_seq(struct stfsm_seq *seq)
 {
 	seq->seq_opc[0] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-			   SEQ_OPC_OPCODE(FLASH_CMD_EN4B_ADDR));
+			   SEQ_OPC_OPCODE(SPINOR_OP_EN4B));
 	seq->seq_opc[1] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-			   SEQ_OPC_OPCODE(FLASH_CMD_WREN) |
+			   SEQ_OPC_OPCODE(SPINOR_OP_WREN) |
 			   SEQ_OPC_CSDEASSERT);
 
 	seq->seq[0] = STFSM_INST_CMD2;
@@ -793,7 +755,7 @@ static void stfsm_read_fifo(struct stfsm *fsm, uint32_t *buf, uint32_t size)
 
 	dev_dbg(fsm->dev, "Reading %d bytes from FIFO\n", size);
 
-	BUG_ON((((uint32_t)buf) & 0x3) || (size & 0x3));
+	BUG_ON((((uintptr_t)buf) & 0x3) || (size & 0x3));
 
 	while (remaining) {
 		for (;;) {
@@ -817,7 +779,7 @@ static int stfsm_write_fifo(struct stfsm *fsm, const uint32_t *buf,
 
 	dev_dbg(fsm->dev, "writing %d bytes to FIFO\n", size);
 
-	BUG_ON((((uint32_t)buf) & 0x3) || (size & 0x3));
+	BUG_ON((((uintptr_t)buf) & 0x3) || (size & 0x3));
 
 	writesl(fsm->base + SPI_FAST_SEQ_DATA_REG, buf, words);
 
@@ -827,7 +789,7 @@ static int stfsm_write_fifo(struct stfsm *fsm, const uint32_t *buf,
 static int stfsm_enter_32bit_addr(struct stfsm *fsm, int enter)
 {
 	struct stfsm_seq *seq = &fsm->stfsm_seq_en_32bit_addr;
-	uint32_t cmd = enter ? FLASH_CMD_EN4B_ADDR : FLASH_CMD_EX4B_ADDR;
+	uint32_t cmd = enter ? SPINOR_OP_EN4B : SPINOR_OP_EX4B;
 
 	seq->seq_opc[0] = (SEQ_OPC_PADS_1 |
 			   SEQ_OPC_CYCLES(8) |
@@ -851,7 +813,7 @@ static uint8_t stfsm_wait_busy(struct stfsm *fsm)
 	/* Use RDRS1 */
 	seq->seq_opc[0] = (SEQ_OPC_PADS_1 |
 			   SEQ_OPC_CYCLES(8) |
-			   SEQ_OPC_OPCODE(FLASH_CMD_RDSR));
+			   SEQ_OPC_OPCODE(SPINOR_OP_RDSR));
 
 	/* Load read_status sequence */
 	stfsm_load_seq(fsm, seq);
@@ -889,60 +851,57 @@ static uint8_t stfsm_wait_busy(struct stfsm *fsm)
 }
 
 static int stfsm_read_status(struct stfsm *fsm, uint8_t cmd,
-			   uint8_t *status)
+			     uint8_t *data, int bytes)
 {
 	struct stfsm_seq *seq = &stfsm_seq_read_status_fifo;
 	uint32_t tmp;
+	uint8_t *t = (uint8_t *)&tmp;
+	int i;
 
-	dev_dbg(fsm->dev, "reading STA[%s]\n",
-		(cmd == FLASH_CMD_RDSR) ? "1" : "2");
+	dev_dbg(fsm->dev, "read 'status' register [0x%02x], %d byte(s)\n",
+		cmd, bytes);
 
-	seq->seq_opc[0] = (SEQ_OPC_PADS_1 |
-			   SEQ_OPC_CYCLES(8) |
+	BUG_ON(bytes != 1 && bytes != 2);
+
+	seq->seq_opc[0] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
 			   SEQ_OPC_OPCODE(cmd)),
 
 	stfsm_load_seq(fsm, seq);
 
 	stfsm_read_fifo(fsm, &tmp, 4);
 
-	*status = (uint8_t)(tmp >> 24);
+	for (i = 0; i < bytes; i++)
+		data[i] = t[i];
 
 	stfsm_wait_seq(fsm);
 
 	return 0;
 }
 
-static int stfsm_write_status(struct stfsm *fsm, uint16_t status,
-			       int sta_bytes)
+static int stfsm_write_status(struct stfsm *fsm, uint8_t cmd,
+			    uint16_t data, int bytes, int wait_busy)
 {
 	struct stfsm_seq *seq = &stfsm_seq_write_status;
 
-	dev_dbg(fsm->dev, "writing STA[%s] 0x%04x\n",
-		(sta_bytes == 1) ? "1" : "1+2", status);
-
-	seq->status = (uint32_t)status | STA_PADS_1 | STA_CSDEASSERT;
-	seq->seq[2] = (sta_bytes == 1) ?
-		STFSM_INST_STA_WR1 : STFSM_INST_STA_WR1_2;
-
-	stfsm_load_seq(fsm, seq);
-
-	stfsm_wait_seq(fsm);
+	dev_dbg(fsm->dev,
+		"write 'status' register [0x%02x], %d byte(s), 0x%04x\n"
+		" %s wait-busy\n", cmd, bytes, data, wait_busy ? "with" : "no");
 
-	return 0;
-};
+	BUG_ON(bytes != 1 && bytes != 2);
 
-static int stfsm_wrvcr(struct stfsm *fsm, uint8_t data)
-{
-	struct stfsm_seq *seq = &stfsm_seq_wrvcr;
-
-	dev_dbg(fsm->dev, "writing VCR 0x%02x\n", data);
+	seq->seq_opc[1] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
+			   SEQ_OPC_OPCODE(cmd));
 
-	seq->status = (STA_DATA_BYTE1(data) | STA_PADS_1 | STA_CSDEASSERT);
+	seq->status = (uint32_t)data | STA_PADS_1 | STA_CSDEASSERT;
+	seq->seq[2] = (bytes == 1) ? STFSM_INST_STA_WR1 : STFSM_INST_STA_WR1_2;
 
 	stfsm_load_seq(fsm, seq);
 
 	stfsm_wait_seq(fsm);
 
+	if (wait_busy)
+		stfsm_wait_busy(fsm);
+
 	return 0;
 }
 
@@ -1027,7 +986,7 @@ static void stfsm_prepare_rw_seq(struct stfsm *fsm,
 	if (cfg->write)
 		seq->seq_opc[i++] = (SEQ_OPC_PADS_1 |
 				     SEQ_OPC_CYCLES(8) |
-				     SEQ_OPC_OPCODE(FLASH_CMD_WREN) |
+				     SEQ_OPC_OPCODE(SPINOR_OP_WREN) |
 				     SEQ_OPC_CSDEASSERT);
 
 	/* Address configuration (24 or 32-bit addresses) */
@@ -1149,31 +1108,36 @@ static int stfsm_mx25_config(struct stfsm *fsm)
 		stfsm_mx25_en_32bit_addr_seq(&fsm->stfsm_seq_en_32bit_addr);
 
 		soc_reset = stfsm_can_handle_soc_reset(fsm);
-		if (soc_reset || !fsm->booted_from_spi) {
+		if (soc_reset || !fsm->booted_from_spi)
 			/* If we can handle SoC resets, we enable 32-bit address
 			 * mode pervasively */
 			stfsm_enter_32bit_addr(fsm, 1);
 
-		} else {
+		else
 			/* Else, enable/disable 32-bit addressing before/after
 			 * each operation */
 			fsm->configuration = (CFG_READ_TOGGLE_32BIT_ADDR |
 					      CFG_WRITE_TOGGLE_32BIT_ADDR |
 					      CFG_ERASESEC_TOGGLE_32BIT_ADDR);
-			/* It seems a small delay is required after exiting
-			 * 32-bit mode following a write operation.  The issue
-			 * is under investigation.
-			 */
-			fsm->configuration |= CFG_WRITE_EX_32BIT_ADDR_DELAY;
-		}
 	}
 
-	/* For QUAD mode, set 'QE' STATUS bit */
+	/* Check status of 'QE' bit, update if required. */
+	stfsm_read_status(fsm, SPINOR_OP_RDSR, &sta, 1);
 	data_pads = ((fsm->stfsm_seq_read.seq_cfg >> 16) & 0x3) + 1;
 	if (data_pads == 4) {
-		stfsm_read_status(fsm, FLASH_CMD_RDSR, &sta);
-		sta |= MX25_STATUS_QE;
-		stfsm_write_status(fsm, sta, 1);
+		if (!(sta & MX25_STATUS_QE)) {
+			/* Set 'QE' */
+			sta |= MX25_STATUS_QE;
+
+			stfsm_write_status(fsm, SPINOR_OP_WRSR, sta, 1, 1);
+		}
+	} else {
+		if (sta & MX25_STATUS_QE) {
+			/* Clear 'QE' */
+			sta &= ~MX25_STATUS_QE;
+
+			stfsm_write_status(fsm, SPINOR_OP_WRSR, sta, 1, 1);
+		}
 	}
 
 	return 0;
@@ -1239,7 +1203,7 @@ static int stfsm_n25q_config(struct stfsm *fsm)
 	 */
 	vcr = (N25Q_VCR_DUMMY_CYCLES(8) | N25Q_VCR_XIP_DISABLED |
 	       N25Q_VCR_WRAP_CONT);
-	stfsm_wrvcr(fsm, vcr);
+	stfsm_write_status(fsm, N25Q_CMD_WRVCR, vcr, 1, 0);
 
 	return 0;
 }
@@ -1297,7 +1261,7 @@ static void stfsm_s25fl_write_dyb(struct stfsm *fsm, uint32_t offs, uint8_t dby)
 {
 	struct stfsm_seq seq = {
 		.seq_opc[0] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
-			       SEQ_OPC_OPCODE(FLASH_CMD_WREN) |
+			       SEQ_OPC_OPCODE(SPINOR_OP_WREN) |
 			       SEQ_OPC_CSDEASSERT),
 		.seq_opc[1] = (SEQ_OPC_PADS_1 | SEQ_OPC_CYCLES(8) |
 			       SEQ_OPC_OPCODE(S25FL_CMD_DYBWR)),
@@ -1337,7 +1301,7 @@ static int stfsm_s25fl_clear_status_reg(struct stfsm *fsm)
 			       SEQ_OPC_CSDEASSERT),
 		.seq_opc[1] = (SEQ_OPC_PADS_1 |
 			       SEQ_OPC_CYCLES(8) |
-			       SEQ_OPC_OPCODE(FLASH_CMD_WRDI) |
+			       SEQ_OPC_OPCODE(SPINOR_OP_WRDI) |
 			       SEQ_OPC_CSDEASSERT),
 		.seq = {
 			STFSM_INST_CMD1,
@@ -1367,6 +1331,7 @@ static int stfsm_s25fl_config(struct stfsm *fsm)
 	uint32_t offs;
 	uint16_t sta_wr;
 	uint8_t sr1, cr1, dyb;
+	int update_sr = 0;
 	int ret;
 
 	if (flags & FLASH_FLAG_32BIT_ADDR) {
@@ -1414,34 +1379,28 @@ static int stfsm_s25fl_config(struct stfsm *fsm)
 		}
 	}
 
-	/* Check status of 'QE' bit */
+	/* Check status of 'QE' bit, update if required. */
+	stfsm_read_status(fsm, SPINOR_OP_RDSR2, &cr1, 1);
 	data_pads = ((fsm->stfsm_seq_read.seq_cfg >> 16) & 0x3) + 1;
-	stfsm_read_status(fsm, FLASH_CMD_RDSR2, &cr1);
 	if (data_pads == 4) {
 		if (!(cr1 & STFSM_S25FL_CONFIG_QE)) {
 			/* Set 'QE' */
 			cr1 |= STFSM_S25FL_CONFIG_QE;
 
-			stfsm_read_status(fsm, FLASH_CMD_RDSR, &sr1);
-			sta_wr = ((uint16_t)cr1  << 8) | sr1;
-
-			stfsm_write_status(fsm, sta_wr, 2);
-
-			stfsm_wait_busy(fsm);
+			update_sr = 1;
 		}
 	} else {
-		if ((cr1 & STFSM_S25FL_CONFIG_QE)) {
+		if (cr1 & STFSM_S25FL_CONFIG_QE) {
 			/* Clear 'QE' */
 			cr1 &= ~STFSM_S25FL_CONFIG_QE;
 
-			stfsm_read_status(fsm, FLASH_CMD_RDSR, &sr1);
-			sta_wr = ((uint16_t)cr1  << 8) | sr1;
-
-			stfsm_write_status(fsm, sta_wr, 2);
-
-			stfsm_wait_busy(fsm);
+			update_sr = 1;
 		}
-
+	}
+	if (update_sr) {
+		stfsm_read_status(fsm, SPINOR_OP_RDSR, &sr1, 1);
+		sta_wr = ((uint16_t)cr1  << 8) | sr1;
+		stfsm_write_status(fsm, SPINOR_OP_WRSR, sta_wr, 2, 1);
 	}
 
 	/*
@@ -1456,27 +1415,36 @@ static int stfsm_s25fl_config(struct stfsm *fsm)
 static int stfsm_w25q_config(struct stfsm *fsm)
 {
 	uint32_t data_pads;
-	uint16_t sta_wr;
-	uint8_t sta1, sta2;
+	uint8_t sr1, sr2;
+	uint16_t sr_wr;
+	int update_sr = 0;
 	int ret;
 
 	ret = stfsm_prepare_rwe_seqs_default(fsm);
 	if (ret)
 		return ret;
 
-	/* If using QUAD mode, set QE STATUS bit */
+	/* Check status of 'QE' bit, update if required. */
+	stfsm_read_status(fsm, SPINOR_OP_RDSR2, &sr2, 1);
 	data_pads = ((fsm->stfsm_seq_read.seq_cfg >> 16) & 0x3) + 1;
 	if (data_pads == 4) {
-		stfsm_read_status(fsm, FLASH_CMD_RDSR, &sta1);
-		stfsm_read_status(fsm, FLASH_CMD_RDSR2, &sta2);
-
-		sta_wr = ((uint16_t)sta2 << 8) | sta1;
-
-		sta_wr |= W25Q_STATUS_QE;
-
-		stfsm_write_status(fsm, sta_wr, 2);
-
-		stfsm_wait_busy(fsm);
+		if (!(sr2 & W25Q_STATUS_QE)) {
+			/* Set 'QE' */
+			sr2 |= W25Q_STATUS_QE;
+			update_sr = 1;
+		}
+	} else {
+		if (sr2 & W25Q_STATUS_QE) {
+			/* Clear 'QE' */
+			sr2 &= ~W25Q_STATUS_QE;
+			update_sr = 1;
+		}
+	}
+	if (update_sr) {
+		/* Write status register */
+		stfsm_read_status(fsm, SPINOR_OP_RDSR, &sr1, 1);
+		sr_wr = ((uint16_t)sr2 << 8) | sr1;
+		stfsm_write_status(fsm, SPINOR_OP_WRSR, sr_wr, 2, 1);
 	}
 
 	return 0;
@@ -1506,7 +1474,7 @@ static int stfsm_read(struct stfsm *fsm, uint8_t *buf, uint32_t size,
 	read_mask = (data_pads << 2) - 1;
 
 	/* Handle non-aligned buf */
-	p = ((uint32_t)buf & 0x3) ? (uint8_t *)page_buf : buf;
+	p = ((uintptr_t)buf & 0x3) ? (uint8_t *)page_buf : buf;
 
 	/* Handle non-aligned size */
 	size_ub = (size + read_mask) & ~read_mask;
@@ -1528,7 +1496,7 @@ static int stfsm_read(struct stfsm *fsm, uint8_t *buf, uint32_t size,
 	}
 
 	/* Handle non-aligned buf */
-	if ((uint32_t)buf & 0x3)
+	if ((uintptr_t)buf & 0x3)
 		memcpy(buf, page_buf, size);
 
 	/* Wait for sequence to finish */
@@ -1570,7 +1538,7 @@ static int stfsm_write(struct stfsm *fsm, const uint8_t *buf,
 	write_mask = (data_pads << 2) - 1;
 
 	/* Handle non-aligned buf */
-	if ((uint32_t)buf & 0x3) {
+	if ((uintptr_t)buf & 0x3) {
 		memcpy(page_buf, buf, size);
 		p = (uint8_t *)page_buf;
 	} else {
@@ -1628,11 +1596,8 @@ static int stfsm_write(struct stfsm *fsm, const uint8_t *buf,
 		stfsm_s25fl_clear_status_reg(fsm);
 
 	/* Exit 32-bit address mode, if required */
-	if (fsm->configuration & CFG_WRITE_TOGGLE_32BIT_ADDR) {
+	if (fsm->configuration & CFG_WRITE_TOGGLE_32BIT_ADDR)
 		stfsm_enter_32bit_addr(fsm, 0);
-		if (fsm->configuration & CFG_WRITE_EX_32BIT_ADDR_DELAY)
-			udelay(1);
-	}
 
 	return 0;
 }
@@ -1736,7 +1701,7 @@ static int stfsm_mtd_write(struct mtd_info *mtd, loff_t to, size_t len,
 
 	while (len) {
 		/* Write up to page boundary */
-		bytes = min(FLASH_PAGESIZE - page_offs, len);
+		bytes = min_t(size_t, FLASH_PAGESIZE - page_offs, len);
 
 		ret = stfsm_write(fsm, b, bytes, to);
 		if (ret)
@@ -1935,6 +1900,13 @@ static int stfsm_init(struct stfsm *fsm)
 	       fsm->base + SPI_CONFIGDATA);
 	writel(STFSM_DEFAULT_WR_TIME, fsm->base + SPI_STATUS_WR_TIME_REG);
 
+	/*
+	 * Set the FSM 'WAIT' delay to the minimum workable value.  Note, for
+	 * our purposes, the WAIT instruction is used purely to achieve
+	 * "sequence validity" rather than actually implement a delay.
+	 */
+	writel(0x00000001, fsm->base + SPI_PROGRAM_ERASE_TIME);
+
 	/* Clear FIFO, just in case */
 	stfsm_clear_fifo(fsm);
 
@@ -2086,7 +2058,7 @@ static int stfsm_remove(struct platform_device *pdev)
 	return mtd_device_unregister(&fsm->mtd);
 }
 
-static struct of_device_id stfsm_match[] = {
+static const struct of_device_id stfsm_match[] = {
 	{ .compatible = "st,spi-fsm", },
 	{},
 };
diff --git a/drivers/mtd/lpddr/Kconfig b/drivers/mtd/lpddr/Kconfig
index 265f969817e3..3a19cbee24d7 100644
--- a/drivers/mtd/lpddr/Kconfig
+++ b/drivers/mtd/lpddr/Kconfig
@@ -1,5 +1,5 @@
-menu "LPDDR flash memory drivers"
-	depends on MTD!=n
+menu "LPDDR & LPDDR2 PCM memory drivers"
+	depends on MTD
 
 config MTD_LPDDR
 	tristate "Support for LPDDR flash chips"
@@ -17,4 +17,13 @@ config MTD_QINFO_PROBE
 	    Window QINFO interface, permits software to be used for entire
 	    families of devices. This serves similar purpose of CFI on legacy
 	    Flash products
+
+config MTD_LPDDR2_NVM
+	# ARM dependency is only for writel_relaxed()
+	depends on MTD && ARM
+	tristate "Support for LPDDR2-NVM flash chips"
+	help
+	  This option enables support of PCM memories with a LPDDR2-NVM
+	  (Low power double data rate 2) interface.
+
 endmenu
diff --git a/drivers/mtd/lpddr/Makefile b/drivers/mtd/lpddr/Makefile
index da48e46b5812..881d440d483e 100644
--- a/drivers/mtd/lpddr/Makefile
+++ b/drivers/mtd/lpddr/Makefile
@@ -4,3 +4,4 @@
 
 obj-$(CONFIG_MTD_QINFO_PROBE)	+= qinfo_probe.o
 obj-$(CONFIG_MTD_LPDDR)	+= lpddr_cmds.o
+obj-$(CONFIG_MTD_LPDDR2_NVM) += lpddr2_nvm.o
diff --git a/drivers/mtd/lpddr/lpddr2_nvm.c b/drivers/mtd/lpddr/lpddr2_nvm.c
new file mode 100644
index 000000000000..063cec40d0ae
--- /dev/null
+++ b/drivers/mtd/lpddr/lpddr2_nvm.c
@@ -0,0 +1,507 @@
+/*
+ * LPDDR2-NVM MTD driver. This module provides read, write, erase, lock/unlock
+ * support for LPDDR2-NVM PCM memories
+ *
+ * Copyright © 2012 Micron Technology, Inc.
+ *
+ * Vincenzo Aliberti <vincenzo.aliberti@gmail.com>
+ * Domenico Manna <domenico.manna@gmail.com>
+ * Many thanks to Andrea Vigilante for initial enabling
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": %s: " fmt, __func__
+
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/mtd/map.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/partitions.h>
+#include <linux/slab.h>
+#include <linux/platform_device.h>
+#include <linux/ioport.h>
+#include <linux/err.h>
+
+/* Parameters */
+#define ERASE_BLOCKSIZE			(0x00020000/2)	/* in Word */
+#define WRITE_BUFFSIZE			(0x00000400/2)	/* in Word */
+#define OW_BASE_ADDRESS			0x00000000	/* OW offset */
+#define BUS_WIDTH			0x00000020	/* x32 devices */
+
+/* PFOW symbols address offset */
+#define PFOW_QUERY_STRING_P		(0x0000/2)	/* in Word */
+#define PFOW_QUERY_STRING_F		(0x0002/2)	/* in Word */
+#define PFOW_QUERY_STRING_O		(0x0004/2)	/* in Word */
+#define PFOW_QUERY_STRING_W		(0x0006/2)	/* in Word */
+
+/* OW registers address */
+#define CMD_CODE_OFS			(0x0080/2)	/* in Word */
+#define CMD_DATA_OFS			(0x0084/2)	/* in Word */
+#define CMD_ADD_L_OFS			(0x0088/2)	/* in Word */
+#define CMD_ADD_H_OFS			(0x008A/2)	/* in Word */
+#define MPR_L_OFS			(0x0090/2)	/* in Word */
+#define MPR_H_OFS			(0x0092/2)	/* in Word */
+#define CMD_EXEC_OFS			(0x00C0/2)	/* in Word */
+#define STATUS_REG_OFS			(0x00CC/2)	/* in Word */
+#define PRG_BUFFER_OFS			(0x0010/2)	/* in Word */
+
+/* Datamask */
+#define MR_CFGMASK			0x8000
+#define SR_OK_DATAMASK			0x0080
+
+/* LPDDR2-NVM Commands */
+#define LPDDR2_NVM_LOCK			0x0061
+#define LPDDR2_NVM_UNLOCK		0x0062
+#define LPDDR2_NVM_SW_PROGRAM		0x0041
+#define LPDDR2_NVM_SW_OVERWRITE		0x0042
+#define LPDDR2_NVM_BUF_PROGRAM		0x00E9
+#define LPDDR2_NVM_BUF_OVERWRITE	0x00EA
+#define LPDDR2_NVM_ERASE		0x0020
+
+/* LPDDR2-NVM Registers offset */
+#define LPDDR2_MODE_REG_DATA		0x0040
+#define LPDDR2_MODE_REG_CFG		0x0050
+
+/*
+ * Internal Type Definitions
+ * pcm_int_data contains memory controller details:
+ * @reg_data : LPDDR2_MODE_REG_DATA register address after remapping
+ * @reg_cfg  : LPDDR2_MODE_REG_CFG register address after remapping
+ * &bus_width: memory bus-width (eg: x16 2 Bytes, x32 4 Bytes)
+ */
+struct pcm_int_data {
+	void __iomem *ctl_regs;
+	int bus_width;
+};
+
+static DEFINE_MUTEX(lpdd2_nvm_mutex);
+
+/*
+ * Build a map_word starting from an u_long
+ */
+static inline map_word build_map_word(u_long myword)
+{
+	map_word val = { {0} };
+	val.x[0] = myword;
+	return val;
+}
+
+/*
+ * Build Mode Register Configuration DataMask based on device bus-width
+ */
+static inline u_int build_mr_cfgmask(u_int bus_width)
+{
+	u_int val = MR_CFGMASK;
+
+	if (bus_width == 0x0004)		/* x32 device */
+		val = val << 16;
+
+	return val;
+}
+
+/*
+ * Build Status Register OK DataMask based on device bus-width
+ */
+static inline u_int build_sr_ok_datamask(u_int bus_width)
+{
+	u_int val = SR_OK_DATAMASK;
+
+	if (bus_width == 0x0004)		/* x32 device */
+		val = (val << 16)+val;
+
+	return val;
+}
+
+/*
+ * Evaluates Overlay Window Control Registers address
+ */
+static inline u_long ow_reg_add(struct map_info *map, u_long offset)
+{
+	u_long val = 0;
+	struct pcm_int_data *pcm_data = map->fldrv_priv;
+
+	val = map->pfow_base + offset*pcm_data->bus_width;
+
+	return val;
+}
+
+/*
+ * Enable lpddr2-nvm Overlay Window
+ * Overlay Window is a memory mapped area containing all LPDDR2-NVM registers
+ * used by device commands as well as uservisible resources like Device Status
+ * Register, Device ID, etc
+ */
+static inline void ow_enable(struct map_info *map)
+{
+	struct pcm_int_data *pcm_data = map->fldrv_priv;
+
+	writel_relaxed(build_mr_cfgmask(pcm_data->bus_width) | 0x18,
+		pcm_data->ctl_regs + LPDDR2_MODE_REG_CFG);
+	writel_relaxed(0x01, pcm_data->ctl_regs + LPDDR2_MODE_REG_DATA);
+}
+
+/*
+ * Disable lpddr2-nvm Overlay Window
+ * Overlay Window is a memory mapped area containing all LPDDR2-NVM registers
+ * used by device commands as well as uservisible resources like Device Status
+ * Register, Device ID, etc
+ */
+static inline void ow_disable(struct map_info *map)
+{
+	struct pcm_int_data *pcm_data = map->fldrv_priv;
+
+	writel_relaxed(build_mr_cfgmask(pcm_data->bus_width) | 0x18,
+		pcm_data->ctl_regs + LPDDR2_MODE_REG_CFG);
+	writel_relaxed(0x02, pcm_data->ctl_regs + LPDDR2_MODE_REG_DATA);
+}
+
+/*
+ * Execute lpddr2-nvm operations
+ */
+static int lpddr2_nvm_do_op(struct map_info *map, u_long cmd_code,
+	u_long cmd_data, u_long cmd_add, u_long cmd_mpr, u_char *buf)
+{
+	map_word add_l = { {0} }, add_h = { {0} }, mpr_l = { {0} },
+		mpr_h = { {0} }, data_l = { {0} }, cmd = { {0} },
+		exec_cmd = { {0} }, sr;
+	map_word data_h = { {0} };	/* only for 2x x16 devices stacked */
+	u_long i, status_reg, prg_buff_ofs;
+	struct pcm_int_data *pcm_data = map->fldrv_priv;
+	u_int sr_ok_datamask = build_sr_ok_datamask(pcm_data->bus_width);
+
+	/* Builds low and high words for OW Control Registers */
+	add_l.x[0]	= cmd_add & 0x0000FFFF;
+	add_h.x[0]	= (cmd_add >> 16) & 0x0000FFFF;
+	mpr_l.x[0]	= cmd_mpr & 0x0000FFFF;
+	mpr_h.x[0]	= (cmd_mpr >> 16) & 0x0000FFFF;
+	cmd.x[0]	= cmd_code & 0x0000FFFF;
+	exec_cmd.x[0]	= 0x0001;
+	data_l.x[0]	= cmd_data & 0x0000FFFF;
+	data_h.x[0]	= (cmd_data >> 16) & 0x0000FFFF; /* only for 2x x16 */
+
+	/* Set Overlay Window Control Registers */
+	map_write(map, cmd, ow_reg_add(map, CMD_CODE_OFS));
+	map_write(map, data_l, ow_reg_add(map, CMD_DATA_OFS));
+	map_write(map, add_l, ow_reg_add(map, CMD_ADD_L_OFS));
+	map_write(map, add_h, ow_reg_add(map, CMD_ADD_H_OFS));
+	map_write(map, mpr_l, ow_reg_add(map, MPR_L_OFS));
+	map_write(map, mpr_h, ow_reg_add(map, MPR_H_OFS));
+	if (pcm_data->bus_width == 0x0004) {	/* 2x16 devices stacked */
+		map_write(map, cmd, ow_reg_add(map, CMD_CODE_OFS) + 2);
+		map_write(map, data_h, ow_reg_add(map, CMD_DATA_OFS) + 2);
+		map_write(map, add_l, ow_reg_add(map, CMD_ADD_L_OFS) + 2);
+		map_write(map, add_h, ow_reg_add(map, CMD_ADD_H_OFS) + 2);
+		map_write(map, mpr_l, ow_reg_add(map, MPR_L_OFS) + 2);
+		map_write(map, mpr_h, ow_reg_add(map, MPR_H_OFS) + 2);
+	}
+
+	/* Fill Program Buffer */
+	if ((cmd_code == LPDDR2_NVM_BUF_PROGRAM) ||
+		(cmd_code == LPDDR2_NVM_BUF_OVERWRITE)) {
+		prg_buff_ofs = (map_read(map,
+			ow_reg_add(map, PRG_BUFFER_OFS))).x[0];
+		for (i = 0; i < cmd_mpr; i++) {
+			map_write(map, build_map_word(buf[i]), map->pfow_base +
+			prg_buff_ofs + i);
+		}
+	}
+
+	/* Command Execute */
+	map_write(map, exec_cmd, ow_reg_add(map, CMD_EXEC_OFS));
+	if (pcm_data->bus_width == 0x0004)	/* 2x16 devices stacked */
+		map_write(map, exec_cmd, ow_reg_add(map, CMD_EXEC_OFS) + 2);
+
+	/* Status Register Check */
+	do {
+		sr = map_read(map, ow_reg_add(map, STATUS_REG_OFS));
+		status_reg = sr.x[0];
+		if (pcm_data->bus_width == 0x0004) {/* 2x16 devices stacked */
+			sr = map_read(map, ow_reg_add(map,
+				STATUS_REG_OFS) + 2);
+			status_reg += sr.x[0] << 16;
+		}
+	} while ((status_reg & sr_ok_datamask) != sr_ok_datamask);
+
+	return (((status_reg & sr_ok_datamask) == sr_ok_datamask) ? 0 : -EIO);
+}
+
+/*
+ * Execute lpddr2-nvm operations @ block level
+ */
+static int lpddr2_nvm_do_block_op(struct mtd_info *mtd, loff_t start_add,
+	uint64_t len, u_char block_op)
+{
+	struct map_info *map = mtd->priv;
+	u_long add, end_add;
+	int ret = 0;
+
+	mutex_lock(&lpdd2_nvm_mutex);
+
+	ow_enable(map);
+
+	add = start_add;
+	end_add = add + len;
+
+	do {
+		ret = lpddr2_nvm_do_op(map, block_op, 0x00, add, add, NULL);
+		if (ret)
+			goto out;
+		add += mtd->erasesize;
+	} while (add < end_add);
+
+out:
+	ow_disable(map);
+	mutex_unlock(&lpdd2_nvm_mutex);
+	return ret;
+}
+
+/*
+ * verify presence of PFOW string
+ */
+static int lpddr2_nvm_pfow_present(struct map_info *map)
+{
+	map_word pfow_val[4];
+	unsigned int found = 1;
+
+	mutex_lock(&lpdd2_nvm_mutex);
+
+	ow_enable(map);
+
+	/* Load string from array */
+	pfow_val[0] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_P));
+	pfow_val[1] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_F));
+	pfow_val[2] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_O));
+	pfow_val[3] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_W));
+
+	/* Verify the string loaded vs expected */
+	if (!map_word_equal(map, build_map_word('P'), pfow_val[0]))
+		found = 0;
+	if (!map_word_equal(map, build_map_word('F'), pfow_val[1]))
+		found = 0;
+	if (!map_word_equal(map, build_map_word('O'), pfow_val[2]))
+		found = 0;
+	if (!map_word_equal(map, build_map_word('W'), pfow_val[3]))
+		found = 0;
+
+	ow_disable(map);
+
+	mutex_unlock(&lpdd2_nvm_mutex);
+
+	return found;
+}
+
+/*
+ * lpddr2_nvm driver read method
+ */
+static int lpddr2_nvm_read(struct mtd_info *mtd, loff_t start_add,
+				size_t len, size_t *retlen, u_char *buf)
+{
+	struct map_info *map = mtd->priv;
+
+	mutex_lock(&lpdd2_nvm_mutex);
+
+	*retlen = len;
+
+	map_copy_from(map, buf, start_add, *retlen);
+
+	mutex_unlock(&lpdd2_nvm_mutex);
+	return 0;
+}
+
+/*
+ * lpddr2_nvm driver write method
+ */
+static int lpddr2_nvm_write(struct mtd_info *mtd, loff_t start_add,
+				size_t len, size_t *retlen, const u_char *buf)
+{
+	struct map_info *map = mtd->priv;
+	struct pcm_int_data *pcm_data = map->fldrv_priv;
+	u_long add, current_len, tot_len, target_len, my_data;
+	u_char *write_buf = (u_char *)buf;
+	int ret = 0;
+
+	mutex_lock(&lpdd2_nvm_mutex);
+
+	ow_enable(map);
+
+	/* Set start value for the variables */
+	add = start_add;
+	target_len = len;
+	tot_len = 0;
+
+	while (tot_len < target_len) {
+		if (!(IS_ALIGNED(add, mtd->writesize))) { /* do sw program */
+			my_data = write_buf[tot_len];
+			my_data += (write_buf[tot_len+1]) << 8;
+			if (pcm_data->bus_width == 0x0004) {/* 2x16 devices */
+				my_data += (write_buf[tot_len+2]) << 16;
+				my_data += (write_buf[tot_len+3]) << 24;
+			}
+			ret = lpddr2_nvm_do_op(map, LPDDR2_NVM_SW_OVERWRITE,
+				my_data, add, 0x00, NULL);
+			if (ret)
+				goto out;
+
+			add += pcm_data->bus_width;
+			tot_len += pcm_data->bus_width;
+		} else {		/* do buffer program */
+			current_len = min(target_len - tot_len,
+				(u_long) mtd->writesize);
+			ret = lpddr2_nvm_do_op(map, LPDDR2_NVM_BUF_OVERWRITE,
+				0x00, add, current_len, write_buf + tot_len);
+			if (ret)
+				goto out;
+
+			add += current_len;
+			tot_len += current_len;
+		}
+	}
+
+out:
+	*retlen = tot_len;
+	ow_disable(map);
+	mutex_unlock(&lpdd2_nvm_mutex);
+	return ret;
+}
+
+/*
+ * lpddr2_nvm driver erase method
+ */
+static int lpddr2_nvm_erase(struct mtd_info *mtd, struct erase_info *instr)
+{
+	int ret = lpddr2_nvm_do_block_op(mtd, instr->addr, instr->len,
+		LPDDR2_NVM_ERASE);
+	if (!ret) {
+		instr->state = MTD_ERASE_DONE;
+		mtd_erase_callback(instr);
+	}
+
+	return ret;
+}
+
+/*
+ * lpddr2_nvm driver unlock method
+ */
+static int lpddr2_nvm_unlock(struct mtd_info *mtd, loff_t start_add,
+	uint64_t len)
+{
+	return lpddr2_nvm_do_block_op(mtd, start_add, len, LPDDR2_NVM_UNLOCK);
+}
+
+/*
+ * lpddr2_nvm driver lock method
+ */
+static int lpddr2_nvm_lock(struct mtd_info *mtd, loff_t start_add,
+	uint64_t len)
+{
+	return lpddr2_nvm_do_block_op(mtd, start_add, len, LPDDR2_NVM_LOCK);
+}
+
+/*
+ * lpddr2_nvm driver probe method
+ */
+static int lpddr2_nvm_probe(struct platform_device *pdev)
+{
+	struct map_info *map;
+	struct mtd_info *mtd;
+	struct resource *add_range;
+	struct resource *control_regs;
+	struct pcm_int_data *pcm_data;
+
+	/* Allocate memory control_regs data structures */
+	pcm_data = devm_kzalloc(&pdev->dev, sizeof(*pcm_data), GFP_KERNEL);
+	if (!pcm_data)
+		return -ENOMEM;
+
+	pcm_data->bus_width = BUS_WIDTH;
+
+	/* Allocate memory for map_info & mtd_info data structures */
+	map = devm_kzalloc(&pdev->dev, sizeof(*map), GFP_KERNEL);
+	if (!map)
+		return -ENOMEM;
+
+	mtd = devm_kzalloc(&pdev->dev, sizeof(*mtd), GFP_KERNEL);
+	if (!mtd)
+		return -ENOMEM;
+
+	/* lpddr2_nvm address range */
+	add_range = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+	/* Populate map_info data structure */
+	*map = (struct map_info) {
+		.virt		= devm_ioremap_resource(&pdev->dev, add_range),
+		.name		= pdev->dev.init_name,
+		.phys		= add_range->start,
+		.size		= resource_size(add_range),
+		.bankwidth	= pcm_data->bus_width / 2,
+		.pfow_base	= OW_BASE_ADDRESS,
+		.fldrv_priv	= pcm_data,
+	};
+	if (IS_ERR(map->virt))
+		return PTR_ERR(map->virt);
+
+	simple_map_init(map);	/* fill with default methods */
+
+	control_regs = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	pcm_data->ctl_regs = devm_ioremap_resource(&pdev->dev, control_regs);
+	if (IS_ERR(pcm_data->ctl_regs))
+		return PTR_ERR(pcm_data->ctl_regs);
+
+	/* Populate mtd_info data structure */
+	*mtd = (struct mtd_info) {
+		.name		= pdev->dev.init_name,
+		.type		= MTD_RAM,
+		.priv		= map,
+		.size		= resource_size(add_range),
+		.erasesize	= ERASE_BLOCKSIZE * pcm_data->bus_width,
+		.writesize	= 1,
+		.writebufsize	= WRITE_BUFFSIZE * pcm_data->bus_width,
+		.flags		= (MTD_CAP_NVRAM | MTD_POWERUP_LOCK),
+		._read		= lpddr2_nvm_read,
+		._write		= lpddr2_nvm_write,
+		._erase		= lpddr2_nvm_erase,
+		._unlock	= lpddr2_nvm_unlock,
+		._lock		= lpddr2_nvm_lock,
+	};
+
+	/* Verify the presence of the device looking for PFOW string */
+	if (!lpddr2_nvm_pfow_present(map)) {
+		pr_err("device not recognized\n");
+		return -EINVAL;
+	}
+	/* Parse partitions and register the MTD device */
+	return mtd_device_parse_register(mtd, NULL, NULL, NULL, 0);
+}
+
+/*
+ * lpddr2_nvm driver remove method
+ */
+static int lpddr2_nvm_remove(struct platform_device *pdev)
+{
+	return mtd_device_unregister(dev_get_drvdata(&pdev->dev));
+}
+
+/* Initialize platform_driver data structure for lpddr2_nvm */
+static struct platform_driver lpddr2_nvm_drv = {
+	.driver		= {
+		.name	= "lpddr2_nvm",
+	},
+	.probe		= lpddr2_nvm_probe,
+	.remove		= lpddr2_nvm_remove,
+};
+
+module_platform_driver(lpddr2_nvm_drv);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Vincenzo Aliberti <vincenzo.aliberti@gmail.com>");
+MODULE_DESCRIPTION("MTD driver for LPDDR2-NVM PCM memories");
diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index fce23fe043f7..21b2874a303b 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -108,7 +108,7 @@ config MTD_SUN_UFLASH
 
 config MTD_SC520CDP
 	tristate "CFI Flash device mapped on AMD SC520 CDP"
-	depends on X86 && MTD_CFI
+	depends on (MELAN || COMPILE_TEST) && MTD_CFI
 	help
 	  The SC520 CDP board has two banks of CFI-compliant chips and one
 	  Dual-in-line JEDEC chip. This 'mapping' driver supports that
@@ -116,7 +116,7 @@ config MTD_SC520CDP
 
 config MTD_NETSC520
 	tristate "CFI Flash device mapped on AMD NetSc520"
-	depends on X86 && MTD_CFI
+	depends on (MELAN || COMPILE_TEST) && MTD_CFI
 	help
 	  This enables access routines for the flash chips on the AMD NetSc520
 	  demonstration board. If you have one of these boards and would like
diff --git a/drivers/mtd/maps/sc520cdp.c b/drivers/mtd/maps/sc520cdp.c
index 8fead8e46bce..093edd51bdc7 100644
--- a/drivers/mtd/maps/sc520cdp.c
+++ b/drivers/mtd/maps/sc520cdp.c
@@ -183,7 +183,7 @@ static const struct sc520_par_table par_table[NUM_FLASH_BANKS] =
 
 static void sc520cdp_setup_par(void)
 {
-	volatile unsigned long __iomem *mmcr;
+	unsigned long __iomem *mmcr;
 	unsigned long mmcr_val;
 	int i, j;
 
@@ -203,11 +203,11 @@ static void sc520cdp_setup_par(void)
 	*/
 	for(i = 0; i < NUM_FLASH_BANKS; i++) {		/* for each par_table entry  */
 		for(j = 0; j < NUM_SC520_PAR; j++) {	/* for each PAR register     */
-			mmcr_val = mmcr[SC520_PAR(j)];
+			mmcr_val = readl(&mmcr[SC520_PAR(j)]);
 			/* if target device field matches, reprogram the PAR */
 			if((mmcr_val & SC520_PAR_TRGDEV) == par_table[i].trgdev)
 			{
-				mmcr[SC520_PAR(j)] = par_table[i].new_par;
+				writel(par_table[i].new_par, &mmcr[SC520_PAR(j)]);
 				break;
 			}
 		}
diff --git a/drivers/mtd/maps/solutionengine.c b/drivers/mtd/maps/solutionengine.c
index 83a7a7091562..bb580bc16445 100644
--- a/drivers/mtd/maps/solutionengine.c
+++ b/drivers/mtd/maps/solutionengine.c
@@ -33,28 +33,6 @@ struct map_info soleng_flash_map = {
 
 static const char * const probes[] = { "RedBoot", "cmdlinepart", NULL };
 
-#ifdef CONFIG_MTD_SUPERH_RESERVE
-static struct mtd_partition superh_se_partitions[] = {
-	/* Reserved for boot code, read-only */
-	{
-		.name = "flash_boot",
-		.offset = 0x00000000,
-		.size = CONFIG_MTD_SUPERH_RESERVE,
-		.mask_flags = MTD_WRITEABLE,
-	},
-	/* All else is writable (e.g. JFFS) */
-	{
-		.name = "Flash FS",
-		.offset = MTDPART_OFS_NXTBLK,
-		.size = MTDPART_SIZ_FULL,
-	}
-};
-#define NUM_PARTITIONS ARRAY_SIZE(superh_se_partitions)
-#else
-#define superh_se_partitions NULL
-#define NUM_PARTITIONS 0
-#endif /* CONFIG_MTD_SUPERH_RESERVE */
-
 static int __init init_soleng_maps(void)
 {
 	/* First probe at offset 0 */
@@ -92,8 +70,7 @@ static int __init init_soleng_maps(void)
 		mtd_device_register(eprom_mtd, NULL, 0);
 	}
 
-	mtd_device_parse_register(flash_mtd, probes, NULL,
-				  superh_se_partitions, NUM_PARTITIONS);
+	mtd_device_parse_register(flash_mtd, probes, NULL, NULL, 0);
 
 	return 0;
 }
diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
index 4dbfaee9aa95..43e30992a369 100644
--- a/drivers/mtd/mtd_blkdevs.c
+++ b/drivers/mtd/mtd_blkdevs.c
@@ -87,6 +87,9 @@ static int do_blktrans_request(struct mtd_blktrans_ops *tr,
 	if (req->cmd_type != REQ_TYPE_FS)
 		return -EIO;
 
+	if (req->cmd_flags & REQ_FLUSH)
+		return tr->flush(dev);
+
 	if (blk_rq_pos(req) + blk_rq_cur_sectors(req) >
 	    get_capacity(req->rq_disk))
 		return -EIO;
@@ -407,6 +410,9 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
 	if (!new->rq)
 		goto error3;
 
+	if (tr->flush)
+		blk_queue_flush(new->rq, REQ_FLUSH);
+
 	new->rq->queuedata = new;
 	blk_queue_logical_block_size(new->rq, tr->blksize);
 
diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 7d4e7b9da3a1..a0f54e80670c 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -568,13 +568,18 @@ static int mtdchar_write_ioctl(struct mtd_info *mtd,
 {
 	struct mtd_write_req req;
 	struct mtd_oob_ops ops;
-	void __user *usr_data, *usr_oob;
+	const void __user *usr_data, *usr_oob;
 	int ret;
 
-	if (copy_from_user(&req, argp, sizeof(req)) ||
-			!access_ok(VERIFY_READ, req.usr_data, req.len) ||
-			!access_ok(VERIFY_READ, req.usr_oob, req.ooblen))
+	if (copy_from_user(&req, argp, sizeof(req)))
 		return -EFAULT;
+
+	usr_data = (const void __user *)(uintptr_t)req.usr_data;
+	usr_oob = (const void __user *)(uintptr_t)req.usr_oob;
+	if (!access_ok(VERIFY_READ, usr_data, req.len) ||
+	    !access_ok(VERIFY_READ, usr_oob, req.ooblen))
+		return -EFAULT;
+
 	if (!mtd->_write_oob)
 		return -EOPNOTSUPP;
 
@@ -583,10 +588,7 @@ static int mtdchar_write_ioctl(struct mtd_info *mtd,
 	ops.ooblen = (size_t)req.ooblen;
 	ops.ooboffs = 0;
 
-	usr_data = (void __user *)(uintptr_t)req.usr_data;
-	usr_oob = (void __user *)(uintptr_t)req.usr_oob;
-
-	if (req.usr_data) {
+	if (usr_data) {
 		ops.datbuf = memdup_user(usr_data, ops.len);
 		if (IS_ERR(ops.datbuf))
 			return PTR_ERR(ops.datbuf);
@@ -594,7 +596,7 @@ static int mtdchar_write_ioctl(struct mtd_info *mtd,
 		ops.datbuf = NULL;
 	}
 
-	if (req.usr_oob) {
+	if (usr_oob) {
 		ops.oobbuf = memdup_user(usr_oob, ops.ooblen);
 		if (IS_ERR(ops.oobbuf)) {
 			kfree(ops.datbuf);
diff --git a/drivers/mtd/nand/bf5xx_nand.c b/drivers/mtd/nand/bf5xx_nand.c
index b7a24946ca26..722898aea7a6 100644
--- a/drivers/mtd/nand/bf5xx_nand.c
+++ b/drivers/mtd/nand/bf5xx_nand.c
@@ -679,9 +679,6 @@ static int bf5xx_nand_remove(struct platform_device *pdev)
 	peripheral_free_list(bfin_nfc_pin_req);
 	bf5xx_nand_dma_remove(info);
 
-	/* free the common resources */
-	kfree(info);
-
 	return 0;
 }
 
@@ -742,10 +739,10 @@ static int bf5xx_nand_probe(struct platform_device *pdev)
 		return -EFAULT;
 	}
 
-	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	info = devm_kzalloc(&pdev->dev, sizeof(*info), GFP_KERNEL);
 	if (info == NULL) {
 		err = -ENOMEM;
-		goto out_err_kzalloc;
+		goto out_err;
 	}
 
 	platform_set_drvdata(pdev, info);
@@ -790,7 +787,7 @@ static int bf5xx_nand_probe(struct platform_device *pdev)
 	/* initialise the hardware */
 	err = bf5xx_nand_hw_init(info);
 	if (err)
-		goto out_err_hw_init;
+		goto out_err;
 
 	/* setup hardware ECC data struct */
 	if (hardware_ecc) {
@@ -827,9 +824,7 @@ static int bf5xx_nand_probe(struct platform_device *pdev)
 
 out_err_nand_scan:
 	bf5xx_nand_dma_remove(info);
-out_err_hw_init:
-	kfree(info);
-out_err_kzalloc:
+out_err:
 	peripheral_free_list(bfin_nfc_pin_req);
 
 	return err;
diff --git a/drivers/mtd/nand/denali.c b/drivers/mtd/nand/denali.c
index c07cd573ad3a..9f2012a3e764 100644
--- a/drivers/mtd/nand/denali.c
+++ b/drivers/mtd/nand/denali.c
@@ -1233,7 +1233,7 @@ static int denali_waitfunc(struct mtd_info *mtd, struct nand_chip *chip)
 	return status;
 }
 
-static void denali_erase(struct mtd_info *mtd, int page)
+static int denali_erase(struct mtd_info *mtd, int page)
 {
 	struct denali_nand_info *denali = mtd_to_denali(mtd);
 
@@ -1250,8 +1250,7 @@ static void denali_erase(struct mtd_info *mtd, int page)
 	irq_status = wait_for_irq(denali, INTR_STATUS__ERASE_COMP |
 					INTR_STATUS__ERASE_FAIL);
 
-	denali->status = (irq_status & INTR_STATUS__ERASE_FAIL) ?
-						NAND_STATUS_FAIL : PASS;
+	return (irq_status & INTR_STATUS__ERASE_FAIL) ? NAND_STATUS_FAIL : PASS;
 }
 
 static void denali_cmdfunc(struct mtd_info *mtd, unsigned int cmd, int col,
@@ -1584,7 +1583,7 @@ int denali_init(struct denali_nand_info *denali)
 	denali->nand.ecc.write_page_raw = denali_write_page_raw;
 	denali->nand.ecc.read_oob = denali_read_oob;
 	denali->nand.ecc.write_oob = denali_write_oob;
-	denali->nand.erase_cmd = denali_erase;
+	denali->nand.erase = denali_erase;
 
 	if (nand_scan_tail(&denali->mtd)) {
 		ret = -ENXIO;
diff --git a/drivers/mtd/nand/docg4.c b/drivers/mtd/nand/docg4.c
index 1b0265e85a06..ce24637e14f1 100644
--- a/drivers/mtd/nand/docg4.c
+++ b/drivers/mtd/nand/docg4.c
@@ -872,7 +872,7 @@ static int docg4_read_oob(struct mtd_info *mtd, struct nand_chip *nand,
 	return 0;
 }
 
-static void docg4_erase_block(struct mtd_info *mtd, int page)
+static int docg4_erase_block(struct mtd_info *mtd, int page)
 {
 	struct nand_chip *nand = mtd->priv;
 	struct docg4_priv *doc = nand->priv;
@@ -916,6 +916,8 @@ static void docg4_erase_block(struct mtd_info *mtd, int page)
 	write_nop(docptr);
 	poll_status(doc);
 	write_nop(docptr);
+
+	return nand->waitfunc(mtd, nand);
 }
 
 static int write_page(struct mtd_info *mtd, struct nand_chip *nand,
@@ -1236,7 +1238,7 @@ static void __init init_mtd_structs(struct mtd_info *mtd)
 	nand->block_markbad = docg4_block_markbad;
 	nand->read_buf = docg4_read_buf;
 	nand->write_buf = docg4_write_buf16;
-	nand->erase_cmd = docg4_erase_block;
+	nand->erase = docg4_erase_block;
 	nand->ecc.read_page = docg4_read_page;
 	nand->ecc.write_page = docg4_write_page;
 	nand->ecc.read_page_raw = docg4_read_page_raw;
diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c
index ec549cd9849f..545a5c002f09 100644
--- a/drivers/mtd/nand/fsl_elbc_nand.c
+++ b/drivers/mtd/nand/fsl_elbc_nand.c
@@ -723,6 +723,19 @@ static int fsl_elbc_write_page(struct mtd_info *mtd, struct nand_chip *chip,
 	return 0;
 }
 
+/* ECC will be calculated automatically, and errors will be detected in
+ * waitfunc.
+ */
+static int fsl_elbc_write_subpage(struct mtd_info *mtd, struct nand_chip *chip,
+				uint32_t offset, uint32_t data_len,
+				const uint8_t *buf, int oob_required)
+{
+	fsl_elbc_write_buf(mtd, buf, mtd->writesize);
+	fsl_elbc_write_buf(mtd, chip->oob_poi, mtd->oobsize);
+
+	return 0;
+}
+
 static int fsl_elbc_chip_init(struct fsl_elbc_mtd *priv)
 {
 	struct fsl_lbc_ctrl *ctrl = priv->ctrl;
@@ -761,6 +774,7 @@ static int fsl_elbc_chip_init(struct fsl_elbc_mtd *priv)
 
 	chip->ecc.read_page = fsl_elbc_read_page;
 	chip->ecc.write_page = fsl_elbc_write_page;
+	chip->ecc.write_subpage = fsl_elbc_write_subpage;
 
 	/* If CS Base Register selects full hardware ECC then use it */
 	if ((in_be32(&lbc->bank[priv->bank].br) & BR_DECC) ==
diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c
index cb45d2f8e208..2338124dd05f 100644
--- a/drivers/mtd/nand/fsl_ifc_nand.c
+++ b/drivers/mtd/nand/fsl_ifc_nand.c
@@ -56,7 +56,7 @@ struct fsl_ifc_nand_ctrl {
 	struct nand_hw_control controller;
 	struct fsl_ifc_mtd *chips[FSL_IFC_BANK_COUNT];
 
-	u8 __iomem *addr;	/* Address of assigned IFC buffer	*/
+	void __iomem *addr;	/* Address of assigned IFC buffer	*/
 	unsigned int page;	/* Last page written to / read from	*/
 	unsigned int read_bytes;/* Number of bytes read during command	*/
 	unsigned int column;	/* Saved column from SEQIN		*/
@@ -591,7 +591,10 @@ static void fsl_ifc_cmdfunc(struct mtd_info *mtd, unsigned int command,
 		 * The chip always seems to report that it is
 		 * write-protected, even when it is not.
 		 */
-		setbits8(ifc_nand_ctrl->addr, NAND_STATUS_WP);
+		if (chip->options & NAND_BUSWIDTH_16)
+			setbits16(ifc_nand_ctrl->addr, NAND_STATUS_WP);
+		else
+			setbits8(ifc_nand_ctrl->addr, NAND_STATUS_WP);
 		return;
 
 	case NAND_CMD_RESET:
@@ -636,7 +639,7 @@ static void fsl_ifc_write_buf(struct mtd_info *mtd, const u8 *buf, int len)
 		len = bufsize - ifc_nand_ctrl->index;
 	}
 
-	memcpy_toio(&ifc_nand_ctrl->addr[ifc_nand_ctrl->index], buf, len);
+	memcpy_toio(ifc_nand_ctrl->addr + ifc_nand_ctrl->index, buf, len);
 	ifc_nand_ctrl->index += len;
 }
 
@@ -648,13 +651,16 @@ static uint8_t fsl_ifc_read_byte(struct mtd_info *mtd)
 {
 	struct nand_chip *chip = mtd->priv;
 	struct fsl_ifc_mtd *priv = chip->priv;
+	unsigned int offset;
 
 	/*
 	 * If there are still bytes in the IFC buffer, then use the
 	 * next byte.
 	 */
-	if (ifc_nand_ctrl->index < ifc_nand_ctrl->read_bytes)
-		return in_8(&ifc_nand_ctrl->addr[ifc_nand_ctrl->index++]);
+	if (ifc_nand_ctrl->index < ifc_nand_ctrl->read_bytes) {
+		offset = ifc_nand_ctrl->index++;
+		return in_8(ifc_nand_ctrl->addr + offset);
+	}
 
 	dev_err(priv->dev, "%s: beyond end of buffer\n", __func__);
 	return ERR_BYTE;
@@ -675,8 +681,7 @@ static uint8_t fsl_ifc_read_byte16(struct mtd_info *mtd)
 	 * next byte.
 	 */
 	if (ifc_nand_ctrl->index < ifc_nand_ctrl->read_bytes) {
-		data = in_be16((uint16_t __iomem *)&ifc_nand_ctrl->
-			       addr[ifc_nand_ctrl->index]);
+		data = in_be16(ifc_nand_ctrl->addr + ifc_nand_ctrl->index);
 		ifc_nand_ctrl->index += 2;
 		return (uint8_t) data;
 	}
@@ -701,7 +706,7 @@ static void fsl_ifc_read_buf(struct mtd_info *mtd, u8 *buf, int len)
 
 	avail = min((unsigned int)len,
 			ifc_nand_ctrl->read_bytes - ifc_nand_ctrl->index);
-	memcpy_fromio(buf, &ifc_nand_ctrl->addr[ifc_nand_ctrl->index], avail);
+	memcpy_fromio(buf, ifc_nand_ctrl->addr + ifc_nand_ctrl->index, avail);
 	ifc_nand_ctrl->index += avail;
 
 	if (len > avail)
diff --git a/drivers/mtd/nand/gpmi-nand/bch-regs.h b/drivers/mtd/nand/gpmi-nand/bch-regs.h
index 588f5374047c..05bb91f2f4c4 100644
--- a/drivers/mtd/nand/gpmi-nand/bch-regs.h
+++ b/drivers/mtd/nand/gpmi-nand/bch-regs.h
@@ -54,7 +54,7 @@
 #define MX6Q_BP_BCH_FLASH0LAYOUT0_ECC0		11
 #define MX6Q_BM_BCH_FLASH0LAYOUT0_ECC0	(0x1f << MX6Q_BP_BCH_FLASH0LAYOUT0_ECC0)
 #define BF_BCH_FLASH0LAYOUT0_ECC0(v, x)				\
-	(GPMI_IS_MX6Q(x)					\
+	(GPMI_IS_MX6(x)					\
 		? (((v) << MX6Q_BP_BCH_FLASH0LAYOUT0_ECC0)	\
 			& MX6Q_BM_BCH_FLASH0LAYOUT0_ECC0)	\
 		: (((v) << BP_BCH_FLASH0LAYOUT0_ECC0)		\
@@ -65,7 +65,7 @@
 #define MX6Q_BM_BCH_FLASH0LAYOUT0_GF_13_14			\
 				(0x1 << MX6Q_BP_BCH_FLASH0LAYOUT0_GF_13_14)
 #define BF_BCH_FLASH0LAYOUT0_GF(v, x)				\
-	((GPMI_IS_MX6Q(x) && ((v) == 14))			\
+	((GPMI_IS_MX6(x) && ((v) == 14))			\
 		? (((1) << MX6Q_BP_BCH_FLASH0LAYOUT0_GF_13_14)	\
 			& MX6Q_BM_BCH_FLASH0LAYOUT0_GF_13_14)	\
 		: 0						\
@@ -77,7 +77,7 @@
 #define MX6Q_BM_BCH_FLASH0LAYOUT0_DATA0_SIZE	\
 			(0x3ff << BP_BCH_FLASH0LAYOUT0_DATA0_SIZE)
 #define BF_BCH_FLASH0LAYOUT0_DATA0_SIZE(v, x)				\
-	(GPMI_IS_MX6Q(x)						\
+	(GPMI_IS_MX6(x)						\
 		? (((v) >> 2) & MX6Q_BM_BCH_FLASH0LAYOUT0_DATA0_SIZE)	\
 		: ((v) & BM_BCH_FLASH0LAYOUT0_DATA0_SIZE)		\
 	)
@@ -96,7 +96,7 @@
 #define MX6Q_BP_BCH_FLASH0LAYOUT1_ECCN		11
 #define MX6Q_BM_BCH_FLASH0LAYOUT1_ECCN	(0x1f << MX6Q_BP_BCH_FLASH0LAYOUT1_ECCN)
 #define BF_BCH_FLASH0LAYOUT1_ECCN(v, x)				\
-	(GPMI_IS_MX6Q(x)					\
+	(GPMI_IS_MX6(x)					\
 		? (((v) << MX6Q_BP_BCH_FLASH0LAYOUT1_ECCN)	\
 			& MX6Q_BM_BCH_FLASH0LAYOUT1_ECCN)	\
 		: (((v) << BP_BCH_FLASH0LAYOUT1_ECCN)		\
@@ -107,7 +107,7 @@
 #define MX6Q_BM_BCH_FLASH0LAYOUT1_GF_13_14			\
 				(0x1 << MX6Q_BP_BCH_FLASH0LAYOUT1_GF_13_14)
 #define BF_BCH_FLASH0LAYOUT1_GF(v, x)				\
-	((GPMI_IS_MX6Q(x) && ((v) == 14))			\
+	((GPMI_IS_MX6(x) && ((v) == 14))			\
 		? (((1) << MX6Q_BP_BCH_FLASH0LAYOUT1_GF_13_14)	\
 			& MX6Q_BM_BCH_FLASH0LAYOUT1_GF_13_14)	\
 		: 0						\
@@ -119,7 +119,7 @@
 #define MX6Q_BM_BCH_FLASH0LAYOUT1_DATAN_SIZE	\
 			(0x3ff << BP_BCH_FLASH0LAYOUT1_DATAN_SIZE)
 #define BF_BCH_FLASH0LAYOUT1_DATAN_SIZE(v, x)				\
-	(GPMI_IS_MX6Q(x)						\
+	(GPMI_IS_MX6(x)						\
 		? (((v) >> 2) & MX6Q_BM_BCH_FLASH0LAYOUT1_DATAN_SIZE)	\
 		: ((v) & BM_BCH_FLASH0LAYOUT1_DATAN_SIZE)		\
 	)
diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-lib.c b/drivers/mtd/nand/gpmi-nand/gpmi-lib.c
index dd1df605a1d6..87e658ce23ef 100644
--- a/drivers/mtd/nand/gpmi-nand/gpmi-lib.c
+++ b/drivers/mtd/nand/gpmi-nand/gpmi-lib.c
@@ -861,7 +861,7 @@ static void gpmi_compute_edo_timing(struct gpmi_nand_data *this,
 	struct resources *r = &this->resources;
 	unsigned long rate = clk_get_rate(r->clock[0]);
 	int mode = this->timing_mode;
-	int dll_threshold = 16; /* in ns */
+	int dll_threshold = this->devdata->max_chain_delay;
 	unsigned long delay;
 	unsigned long clk_period;
 	int t_rea;
@@ -886,9 +886,6 @@ static void gpmi_compute_edo_timing(struct gpmi_nand_data *this,
 	/* [3] for GPMI_HW_GPMI_CTRL1 */
 	hw->wrn_dly_sel = BV_GPMI_CTRL1_WRN_DLY_SEL_NO_DELAY;
 
-	if (GPMI_IS_MX6Q(this))
-		dll_threshold = 12;
-
 	/*
 	 * Enlarge 10 times for the numerator and denominator in {3}.
 	 * This make us to get more accurate result.
@@ -974,7 +971,7 @@ int gpmi_extra_init(struct gpmi_nand_data *this)
 	struct nand_chip *chip = &this->nand;
 
 	/* Enable the asynchronous EDO feature. */
-	if (GPMI_IS_MX6Q(this) && chip->onfi_version) {
+	if (GPMI_IS_MX6(this) && chip->onfi_version) {
 		int mode = onfi_get_async_timing_mode(chip);
 
 		/* We only support the timing mode 4 and mode 5. */
@@ -1096,12 +1093,12 @@ int gpmi_is_ready(struct gpmi_nand_data *this, unsigned chip)
 	if (GPMI_IS_MX23(this)) {
 		mask = MX23_BM_GPMI_DEBUG_READY0 << chip;
 		reg = readl(r->gpmi_regs + HW_GPMI_DEBUG);
-	} else if (GPMI_IS_MX28(this) || GPMI_IS_MX6Q(this)) {
+	} else if (GPMI_IS_MX28(this) || GPMI_IS_MX6(this)) {
 		/*
 		 * In the imx6, all the ready/busy pins are bound
 		 * together. So we only need to check chip 0.
 		 */
-		if (GPMI_IS_MX6Q(this))
+		if (GPMI_IS_MX6(this))
 			chip = 0;
 
 		/* MX28 shares the same R/B register as MX6Q. */
diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-nand.c b/drivers/mtd/nand/gpmi-nand/gpmi-nand.c
index bb77f750e75a..f638cd8077ca 100644
--- a/drivers/mtd/nand/gpmi-nand/gpmi-nand.c
+++ b/drivers/mtd/nand/gpmi-nand/gpmi-nand.c
@@ -53,6 +53,30 @@ static struct nand_ecclayout gpmi_hw_ecclayout = {
 	.oobfree = { {.offset = 0, .length = 0} }
 };
 
+static const struct gpmi_devdata gpmi_devdata_imx23 = {
+	.type = IS_MX23,
+	.bch_max_ecc_strength = 20,
+	.max_chain_delay = 16,
+};
+
+static const struct gpmi_devdata gpmi_devdata_imx28 = {
+	.type = IS_MX28,
+	.bch_max_ecc_strength = 20,
+	.max_chain_delay = 16,
+};
+
+static const struct gpmi_devdata gpmi_devdata_imx6q = {
+	.type = IS_MX6Q,
+	.bch_max_ecc_strength = 40,
+	.max_chain_delay = 12,
+};
+
+static const struct gpmi_devdata gpmi_devdata_imx6sx = {
+	.type = IS_MX6SX,
+	.bch_max_ecc_strength = 62,
+	.max_chain_delay = 12,
+};
+
 static irqreturn_t bch_irq(int irq, void *cookie)
 {
 	struct gpmi_nand_data *this = cookie;
@@ -102,14 +126,8 @@ static inline bool gpmi_check_ecc(struct gpmi_nand_data *this)
 		/* The mx23/mx28 only support the GF13. */
 		if (geo->gf_len == 14)
 			return false;
-
-		if (geo->ecc_strength > MXS_ECC_STRENGTH_MAX)
-			return false;
-	} else if (GPMI_IS_MX6Q(this)) {
-		if (geo->ecc_strength > MX6_ECC_STRENGTH_MAX)
-			return false;
 	}
-	return true;
+	return geo->ecc_strength <= this->devdata->bch_max_ecc_strength;
 }
 
 /*
@@ -270,8 +288,7 @@ static int legacy_set_geometry(struct gpmi_nand_data *this)
 			"We can not support this nand chip."
 			" Its required ecc strength(%d) is beyond our"
 			" capability(%d).\n", geo->ecc_strength,
-			(GPMI_IS_MX6Q(this) ? MX6_ECC_STRENGTH_MAX
-					: MXS_ECC_STRENGTH_MAX));
+			this->devdata->bch_max_ecc_strength);
 		return -EINVAL;
 	}
 
@@ -572,7 +589,7 @@ static int gpmi_get_clks(struct gpmi_nand_data *this)
 	}
 
 	/* Get extra clocks */
-	if (GPMI_IS_MX6Q(this))
+	if (GPMI_IS_MX6(this))
 		extra_clks = extra_clks_for_mx6q;
 	if (!extra_clks)
 		return 0;
@@ -590,9 +607,9 @@ static int gpmi_get_clks(struct gpmi_nand_data *this)
 		r->clock[i] = clk;
 	}
 
-	if (GPMI_IS_MX6Q(this))
+	if (GPMI_IS_MX6(this))
 		/*
-		 * Set the default value for the gpmi clock in mx6q:
+		 * Set the default value for the gpmi clock.
 		 *
 		 * If you want to use the ONFI nand which is in the
 		 * Synchronous Mode, you should change the clock as you need.
@@ -1655,7 +1672,7 @@ static int gpmi_init_last(struct gpmi_nand_data *this)
 	 *  (1) the chip is imx6, and
 	 *  (2) the size of the ECC parity is byte aligned.
 	 */
-	if (GPMI_IS_MX6Q(this) &&
+	if (GPMI_IS_MX6(this) &&
 		((bch_geo->gf_len * bch_geo->ecc_strength) % 8) == 0) {
 		ecc->read_subpage = gpmi_ecc_read_subpage;
 		chip->options |= NAND_SUBPAGE_READ;
@@ -1711,7 +1728,7 @@ static int gpmi_nand_init(struct gpmi_nand_data *this)
 	if (ret)
 		goto err_out;
 
-	ret = nand_scan_ident(mtd, GPMI_IS_MX6Q(this) ? 2 : 1, NULL);
+	ret = nand_scan_ident(mtd, GPMI_IS_MX6(this) ? 2 : 1, NULL);
 	if (ret)
 		goto err_out;
 
@@ -1740,23 +1757,19 @@ err_out:
 	return ret;
 }
 
-static const struct platform_device_id gpmi_ids[] = {
-	{ .name = "imx23-gpmi-nand", .driver_data = IS_MX23, },
-	{ .name = "imx28-gpmi-nand", .driver_data = IS_MX28, },
-	{ .name = "imx6q-gpmi-nand", .driver_data = IS_MX6Q, },
-	{}
-};
-
 static const struct of_device_id gpmi_nand_id_table[] = {
 	{
 		.compatible = "fsl,imx23-gpmi-nand",
-		.data = (void *)&gpmi_ids[IS_MX23],
+		.data = (void *)&gpmi_devdata_imx23,
 	}, {
 		.compatible = "fsl,imx28-gpmi-nand",
-		.data = (void *)&gpmi_ids[IS_MX28],
+		.data = (void *)&gpmi_devdata_imx28,
 	}, {
 		.compatible = "fsl,imx6q-gpmi-nand",
-		.data = (void *)&gpmi_ids[IS_MX6Q],
+		.data = (void *)&gpmi_devdata_imx6q,
+	}, {
+		.compatible = "fsl,imx6sx-gpmi-nand",
+		.data = (void *)&gpmi_devdata_imx6sx,
 	}, {}
 };
 MODULE_DEVICE_TABLE(of, gpmi_nand_id_table);
@@ -1767,18 +1780,18 @@ static int gpmi_nand_probe(struct platform_device *pdev)
 	const struct of_device_id *of_id;
 	int ret;
 
+	this = devm_kzalloc(&pdev->dev, sizeof(*this), GFP_KERNEL);
+	if (!this)
+		return -ENOMEM;
+
 	of_id = of_match_device(gpmi_nand_id_table, &pdev->dev);
 	if (of_id) {
-		pdev->id_entry = of_id->data;
+		this->devdata = of_id->data;
 	} else {
 		dev_err(&pdev->dev, "Failed to find the right device id.\n");
 		return -ENODEV;
 	}
 
-	this = devm_kzalloc(&pdev->dev, sizeof(*this), GFP_KERNEL);
-	if (!this)
-		return -ENOMEM;
-
 	platform_set_drvdata(pdev, this);
 	this->pdev  = pdev;
 	this->dev   = &pdev->dev;
@@ -1823,7 +1836,6 @@ static struct platform_driver gpmi_nand_driver = {
 	},
 	.probe   = gpmi_nand_probe,
 	.remove  = gpmi_nand_remove,
-	.id_table = gpmi_ids,
 };
 module_platform_driver(gpmi_nand_driver);
 
diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-nand.h b/drivers/mtd/nand/gpmi-nand/gpmi-nand.h
index 4c801fa18725..32c6ba49f986 100644
--- a/drivers/mtd/nand/gpmi-nand/gpmi-nand.h
+++ b/drivers/mtd/nand/gpmi-nand/gpmi-nand.h
@@ -119,11 +119,25 @@ struct nand_timing {
 	int8_t  tRHOH_in_ns;
 };
 
+enum gpmi_type {
+	IS_MX23,
+	IS_MX28,
+	IS_MX6Q,
+	IS_MX6SX
+};
+
+struct gpmi_devdata {
+	enum gpmi_type type;
+	int bch_max_ecc_strength;
+	int max_chain_delay; /* See the async EDO mode */
+};
+
 struct gpmi_nand_data {
 	/* flags */
 #define GPMI_ASYNC_EDO_ENABLED	(1 << 0)
 #define GPMI_TIMING_INIT_OK	(1 << 1)
 	int			flags;
+	const struct gpmi_devdata *devdata;
 
 	/* System Interface */
 	struct device		*dev;
@@ -281,15 +295,11 @@ extern int gpmi_read_page(struct gpmi_nand_data *,
 #define STATUS_ERASED		0xff
 #define STATUS_UNCORRECTABLE	0xfe
 
-/* BCH's bit correction capability. */
-#define MXS_ECC_STRENGTH_MAX	20	/* mx23 and mx28 */
-#define MX6_ECC_STRENGTH_MAX	40
-
-/* Use the platform_id to distinguish different Archs. */
-#define IS_MX23			0x0
-#define IS_MX28			0x1
-#define IS_MX6Q			0x2
-#define GPMI_IS_MX23(x)		((x)->pdev->id_entry->driver_data == IS_MX23)
-#define GPMI_IS_MX28(x)		((x)->pdev->id_entry->driver_data == IS_MX28)
-#define GPMI_IS_MX6Q(x)		((x)->pdev->id_entry->driver_data == IS_MX6Q)
+/* Use the devdata to distinguish different Archs. */
+#define GPMI_IS_MX23(x)		((x)->devdata->type == IS_MX23)
+#define GPMI_IS_MX28(x)		((x)->devdata->type == IS_MX28)
+#define GPMI_IS_MX6Q(x)		((x)->devdata->type == IS_MX6Q)
+#define GPMI_IS_MX6SX(x)	((x)->devdata->type == IS_MX6SX)
+
+#define GPMI_IS_MX6(x)		(GPMI_IS_MX6Q(x) || GPMI_IS_MX6SX(x))
 #endif
diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index 9d01c4df838c..41167e9e991e 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -37,6 +37,7 @@
 #include <linux/err.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
+#include <linux/mm.h>
 #include <linux/types.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
@@ -1204,8 +1205,7 @@ static int nand_read_subpage(struct mtd_info *mtd, struct nand_chip *chip,
 	 * ecc.pos. Let's make sure that there are no gaps in ECC positions.
 	 */
 	for (i = 0; i < eccfrag_len - 1; i++) {
-		if (eccpos[i + start_step * chip->ecc.bytes] + 1 !=
-			eccpos[i + start_step * chip->ecc.bytes + 1]) {
+		if (eccpos[i + index] + 1 != eccpos[i + index + 1]) {
 			gaps = 1;
 			break;
 		}
@@ -1501,6 +1501,7 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
 		mtd->oobavail : mtd->oobsize;
 
 	uint8_t *bufpoi, *oob, *buf;
+	int use_bufpoi;
 	unsigned int max_bitflips = 0;
 	int retry_mode = 0;
 	bool ecc_fail = false;
@@ -1523,9 +1524,20 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
 		bytes = min(mtd->writesize - col, readlen);
 		aligned = (bytes == mtd->writesize);
 
+		if (!aligned)
+			use_bufpoi = 1;
+		else if (chip->options & NAND_USE_BOUNCE_BUFFER)
+			use_bufpoi = !virt_addr_valid(buf);
+		else
+			use_bufpoi = 0;
+
 		/* Is the current page in the buffer? */
 		if (realpage != chip->pagebuf || oob) {
-			bufpoi = aligned ? buf : chip->buffers->databuf;
+			bufpoi = use_bufpoi ? chip->buffers->databuf : buf;
+
+			if (use_bufpoi && aligned)
+				pr_debug("%s: using read bounce buffer for buf@%p\n",
+						 __func__, buf);
 
 read_retry:
 			chip->cmdfunc(mtd, NAND_CMD_READ0, 0x00, page);
@@ -1547,7 +1559,7 @@ read_retry:
 				ret = chip->ecc.read_page(mtd, chip, bufpoi,
 							  oob_required, page);
 			if (ret < 0) {
-				if (!aligned)
+				if (use_bufpoi)
 					/* Invalidate page cache */
 					chip->pagebuf = -1;
 				break;
@@ -1556,7 +1568,7 @@ read_retry:
 			max_bitflips = max_t(unsigned int, max_bitflips, ret);
 
 			/* Transfer not aligned data */
-			if (!aligned) {
+			if (use_bufpoi) {
 				if (!NAND_HAS_SUBPAGE_READ(chip) && !oob &&
 				    !(mtd->ecc_stats.failed - ecc_failures) &&
 				    (ops->mode != MTD_OPS_RAW)) {
@@ -2376,11 +2388,23 @@ static int nand_do_write_ops(struct mtd_info *mtd, loff_t to,
 		int bytes = mtd->writesize;
 		int cached = writelen > bytes && page != blockmask;
 		uint8_t *wbuf = buf;
+		int use_bufpoi;
+		int part_pagewr = (column || writelen < (mtd->writesize - 1));
+
+		if (part_pagewr)
+			use_bufpoi = 1;
+		else if (chip->options & NAND_USE_BOUNCE_BUFFER)
+			use_bufpoi = !virt_addr_valid(buf);
+		else
+			use_bufpoi = 0;
 
-		/* Partial page write? */
-		if (unlikely(column || writelen < (mtd->writesize - 1))) {
+		/* Partial page write?, or need to use bounce buffer */
+		if (use_bufpoi) {
+			pr_debug("%s: using write bounce buffer for buf@%p\n",
+					 __func__, buf);
 			cached = 0;
-			bytes = min_t(int, bytes - column, (int) writelen);
+			if (part_pagewr)
+				bytes = min_t(int, bytes - column, writelen);
 			chip->pagebuf = -1;
 			memset(chip->buffers->databuf, 0xff, mtd->writesize);
 			memcpy(&chip->buffers->databuf[column], buf, bytes);
@@ -2618,18 +2642,20 @@ out:
 }
 
 /**
- * single_erase_cmd - [GENERIC] NAND standard block erase command function
+ * single_erase - [GENERIC] NAND standard block erase command function
  * @mtd: MTD device structure
  * @page: the page address of the block which will be erased
  *
- * Standard erase command for NAND chips.
+ * Standard erase command for NAND chips. Returns NAND status.
  */
-static void single_erase_cmd(struct mtd_info *mtd, int page)
+static int single_erase(struct mtd_info *mtd, int page)
 {
 	struct nand_chip *chip = mtd->priv;
 	/* Send commands to erase a block */
 	chip->cmdfunc(mtd, NAND_CMD_ERASE1, -1, page);
 	chip->cmdfunc(mtd, NAND_CMD_ERASE2, -1, -1);
+
+	return chip->waitfunc(mtd, chip);
 }
 
 /**
@@ -2710,9 +2736,7 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr,
 		    (page + pages_per_block))
 			chip->pagebuf = -1;
 
-		chip->erase_cmd(mtd, page & chip->pagemask);
-
-		status = chip->waitfunc(mtd, chip);
+		status = chip->erase(mtd, page & chip->pagemask);
 
 		/*
 		 * See if operation failed and additional status checks are
@@ -3607,7 +3631,7 @@ static struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd,
 
 	chip->onfi_version = 0;
 	if (!type->name || !type->pagesize) {
-		/* Check is chip is ONFI compliant */
+		/* Check if the chip is ONFI compliant */
 		if (nand_flash_detect_onfi(mtd, chip, &busw))
 			goto ident_done;
 
@@ -3685,7 +3709,7 @@ ident_done:
 	}
 
 	chip->badblockbits = 8;
-	chip->erase_cmd = single_erase_cmd;
+	chip->erase = single_erase;
 
 	/* Do not replace user supplied command function! */
 	if (mtd->writesize > 512 && chip->cmdfunc == nand_command)
@@ -3770,6 +3794,39 @@ int nand_scan_ident(struct mtd_info *mtd, int maxchips,
 }
 EXPORT_SYMBOL(nand_scan_ident);
 
+/*
+ * Check if the chip configuration meet the datasheet requirements.
+
+ * If our configuration corrects A bits per B bytes and the minimum
+ * required correction level is X bits per Y bytes, then we must ensure
+ * both of the following are true:
+ *
+ * (1) A / B >= X / Y
+ * (2) A >= X
+ *
+ * Requirement (1) ensures we can correct for the required bitflip density.
+ * Requirement (2) ensures we can correct even when all bitflips are clumped
+ * in the same sector.
+ */
+static bool nand_ecc_strength_good(struct mtd_info *mtd)
+{
+	struct nand_chip *chip = mtd->priv;
+	struct nand_ecc_ctrl *ecc = &chip->ecc;
+	int corr, ds_corr;
+
+	if (ecc->size == 0 || chip->ecc_step_ds == 0)
+		/* Not enough information */
+		return true;
+
+	/*
+	 * We get the number of corrected bits per page to compare
+	 * the correction density.
+	 */
+	corr = (mtd->writesize * ecc->strength) / ecc->size;
+	ds_corr = (mtd->writesize * chip->ecc_strength_ds) / chip->ecc_step_ds;
+
+	return corr >= ds_corr && ecc->strength >= chip->ecc_strength_ds;
+}
 
 /**
  * nand_scan_tail - [NAND Interface] Scan for the NAND device
@@ -3990,6 +4047,9 @@ int nand_scan_tail(struct mtd_info *mtd)
 		ecc->layout->oobavail += ecc->layout->oobfree[i].length;
 	mtd->oobavail = ecc->layout->oobavail;
 
+	/* ECC sanity check: warn noisily if it's too weak */
+	WARN_ON(!nand_ecc_strength_good(mtd));
+
 	/*
 	 * Set the number of read / write steps for one page depending on ECC
 	 * mode.
@@ -4023,8 +4083,16 @@ int nand_scan_tail(struct mtd_info *mtd)
 	chip->pagebuf = -1;
 
 	/* Large page NAND with SOFT_ECC should support subpage reads */
-	if ((ecc->mode == NAND_ECC_SOFT) && (chip->page_shift > 9))
-		chip->options |= NAND_SUBPAGE_READ;
+	switch (ecc->mode) {
+	case NAND_ECC_SOFT:
+	case NAND_ECC_SOFT_BCH:
+		if (chip->page_shift > 9)
+			chip->options |= NAND_SUBPAGE_READ;
+		break;
+
+	default:
+		break;
+	}
 
 	/* Fill in remaining MTD driver data */
 	mtd->type = nand_is_slc(chip) ? MTD_NANDFLASH : MTD_MLCNANDFLASH;
diff --git a/drivers/mtd/nand/nand_bbt.c b/drivers/mtd/nand/nand_bbt.c
index c0615d1526f9..7f0c3b4c2a4f 100644
--- a/drivers/mtd/nand/nand_bbt.c
+++ b/drivers/mtd/nand/nand_bbt.c
@@ -528,7 +528,7 @@ static int search_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 {
 	struct nand_chip *this = mtd->priv;
 	int i, chips;
-	int bits, startblock, block, dir;
+	int startblock, block, dir;
 	int scanlen = mtd->writesize + mtd->oobsize;
 	int bbtblocks;
 	int blocktopage = this->bbt_erase_shift - this->page_shift;
@@ -552,9 +552,6 @@ static int search_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 		bbtblocks = mtd->size >> this->bbt_erase_shift;
 	}
 
-	/* Number of bits for each erase block in the bbt */
-	bits = td->options & NAND_BBT_NRBITS_MSK;
-
 	for (i = 0; i < chips; i++) {
 		/* Reset version information */
 		td->version[i] = 0;
@@ -1285,6 +1282,7 @@ static int nand_create_badblock_pattern(struct nand_chip *this)
 int nand_default_bbt(struct mtd_info *mtd)
 {
 	struct nand_chip *this = mtd->priv;
+	int ret;
 
 	/* Is a flash based bad block table requested? */
 	if (this->bbt_options & NAND_BBT_USE_FLASH) {
@@ -1303,8 +1301,11 @@ int nand_default_bbt(struct mtd_info *mtd)
 		this->bbt_md = NULL;
 	}
 
-	if (!this->badblock_pattern)
-		nand_create_badblock_pattern(this);
+	if (!this->badblock_pattern) {
+		ret = nand_create_badblock_pattern(this);
+		if (ret)
+			return ret;
+	}
 
 	return nand_scan_bbt(mtd, this->badblock_pattern);
 }
diff --git a/drivers/mtd/nand/nand_ecc.c b/drivers/mtd/nand/nand_ecc.c
index 053c9a2d47c3..97c4c0216c90 100644
--- a/drivers/mtd/nand/nand_ecc.c
+++ b/drivers/mtd/nand/nand_ecc.c
@@ -506,7 +506,7 @@ int __nand_correct_data(unsigned char *buf,
 	if ((bitsperbyte[b0] + bitsperbyte[b1] + bitsperbyte[b2]) == 1)
 		return 1;	/* error in ECC data; no action needed */
 
-	pr_err("%s: uncorrectable ECC error", __func__);
+	pr_err("%s: uncorrectable ECC error\n", __func__);
 	return -1;
 }
 EXPORT_SYMBOL(__nand_correct_data);
diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c
index 42e8a770e631..4f0d83648e5a 100644
--- a/drivers/mtd/nand/nandsim.c
+++ b/drivers/mtd/nand/nandsim.c
@@ -575,12 +575,12 @@ static int alloc_device(struct nandsim *ns)
 		cfile = filp_open(cache_file, O_CREAT | O_RDWR | O_LARGEFILE, 0600);
 		if (IS_ERR(cfile))
 			return PTR_ERR(cfile);
-		if (!cfile->f_op->read && !cfile->f_op->aio_read) {
+		if (!(cfile->f_mode & FMODE_CAN_READ)) {
 			NS_ERR("alloc_device: cache file not readable\n");
 			err = -EINVAL;
 			goto err_close;
 		}
-		if (!cfile->f_op->write && !cfile->f_op->aio_write) {
+		if (!(cfile->f_mode & FMODE_CAN_WRITE)) {
 			NS_ERR("alloc_device: cache file not writeable\n");
 			err = -EINVAL;
 			goto err_close;
diff --git a/drivers/mtd/nand/omap2.c b/drivers/mtd/nand/omap2.c
index 1ff49b80bdaf..f0ed92e210a1 100644
--- a/drivers/mtd/nand/omap2.c
+++ b/drivers/mtd/nand/omap2.c
@@ -137,6 +137,10 @@
 #define BADBLOCK_MARKER_LENGTH		2
 
 #ifdef CONFIG_MTD_NAND_OMAP_BCH
+static u_char bch16_vector[] = {0xf5, 0x24, 0x1c, 0xd0, 0x61, 0xb3, 0xf1, 0x55,
+				0x2e, 0x2c, 0x86, 0xa3, 0xed, 0x36, 0x1b, 0x78,
+				0x48, 0x76, 0xa9, 0x3b, 0x97, 0xd1, 0x7a, 0x93,
+				0x07, 0x0e};
 static u_char bch8_vector[] = {0xf3, 0xdb, 0x14, 0x16, 0x8b, 0xd2, 0xbe, 0xcc,
 	0xac, 0x6b, 0xff, 0x99, 0x7b};
 static u_char bch4_vector[] = {0x00, 0x6b, 0x31, 0xdd, 0x41, 0xbc, 0x10};
@@ -1114,6 +1118,19 @@ static void __maybe_unused omap_enable_hwecc_bch(struct mtd_info *mtd, int mode)
 			ecc_size1 = BCH_ECC_SIZE1;
 		}
 		break;
+	case OMAP_ECC_BCH16_CODE_HW:
+		bch_type = 0x2;
+		nsectors = chip->ecc.steps;
+		if (mode == NAND_ECC_READ) {
+			wr_mode	  = 0x01;
+			ecc_size0 = 52; /* ECC bits in nibbles per sector */
+			ecc_size1 = 0;  /* non-ECC bits in nibbles per sector */
+		} else {
+			wr_mode	  = 0x01;
+			ecc_size0 = 0;  /* extra bits in nibbles per sector */
+			ecc_size1 = 52; /* OOB bits in nibbles per sector */
+		}
+		break;
 	default:
 		return;
 	}
@@ -1162,7 +1179,8 @@ static int __maybe_unused omap_calculate_ecc_bch(struct mtd_info *mtd,
 	struct gpmc_nand_regs	*gpmc_regs = &info->reg;
 	u8 *ecc_code;
 	unsigned long nsectors, bch_val1, bch_val2, bch_val3, bch_val4;
-	int i;
+	u32 val;
+	int i, j;
 
 	nsectors = ((readl(info->reg.gpmc_ecc_config) >> 4) & 0x7) + 1;
 	for (i = 0; i < nsectors; i++) {
@@ -1201,6 +1219,41 @@ static int __maybe_unused omap_calculate_ecc_bch(struct mtd_info *mtd,
 			*ecc_code++ = ((bch_val1 >> 4) & 0xFF);
 			*ecc_code++ = ((bch_val1 & 0xF) << 4);
 			break;
+		case OMAP_ECC_BCH16_CODE_HW:
+			val = readl(gpmc_regs->gpmc_bch_result6[i]);
+			ecc_code[0]  = ((val >>  8) & 0xFF);
+			ecc_code[1]  = ((val >>  0) & 0xFF);
+			val = readl(gpmc_regs->gpmc_bch_result5[i]);
+			ecc_code[2]  = ((val >> 24) & 0xFF);
+			ecc_code[3]  = ((val >> 16) & 0xFF);
+			ecc_code[4]  = ((val >>  8) & 0xFF);
+			ecc_code[5]  = ((val >>  0) & 0xFF);
+			val = readl(gpmc_regs->gpmc_bch_result4[i]);
+			ecc_code[6]  = ((val >> 24) & 0xFF);
+			ecc_code[7]  = ((val >> 16) & 0xFF);
+			ecc_code[8]  = ((val >>  8) & 0xFF);
+			ecc_code[9]  = ((val >>  0) & 0xFF);
+			val = readl(gpmc_regs->gpmc_bch_result3[i]);
+			ecc_code[10] = ((val >> 24) & 0xFF);
+			ecc_code[11] = ((val >> 16) & 0xFF);
+			ecc_code[12] = ((val >>  8) & 0xFF);
+			ecc_code[13] = ((val >>  0) & 0xFF);
+			val = readl(gpmc_regs->gpmc_bch_result2[i]);
+			ecc_code[14] = ((val >> 24) & 0xFF);
+			ecc_code[15] = ((val >> 16) & 0xFF);
+			ecc_code[16] = ((val >>  8) & 0xFF);
+			ecc_code[17] = ((val >>  0) & 0xFF);
+			val = readl(gpmc_regs->gpmc_bch_result1[i]);
+			ecc_code[18] = ((val >> 24) & 0xFF);
+			ecc_code[19] = ((val >> 16) & 0xFF);
+			ecc_code[20] = ((val >>  8) & 0xFF);
+			ecc_code[21] = ((val >>  0) & 0xFF);
+			val = readl(gpmc_regs->gpmc_bch_result0[i]);
+			ecc_code[22] = ((val >> 24) & 0xFF);
+			ecc_code[23] = ((val >> 16) & 0xFF);
+			ecc_code[24] = ((val >>  8) & 0xFF);
+			ecc_code[25] = ((val >>  0) & 0xFF);
+			break;
 		default:
 			return -EINVAL;
 		}
@@ -1210,8 +1263,8 @@ static int __maybe_unused omap_calculate_ecc_bch(struct mtd_info *mtd,
 		case OMAP_ECC_BCH4_CODE_HW_DETECTION_SW:
 			/* Add constant polynomial to remainder, so that
 			 * ECC of blank pages results in 0x0 on reading back */
-			for (i = 0; i < eccbytes; i++)
-				ecc_calc[i] ^= bch4_polynomial[i];
+			for (j = 0; j < eccbytes; j++)
+				ecc_calc[j] ^= bch4_polynomial[j];
 			break;
 		case OMAP_ECC_BCH4_CODE_HW:
 			/* Set  8th ECC byte as 0x0 for ROM compatibility */
@@ -1220,13 +1273,15 @@ static int __maybe_unused omap_calculate_ecc_bch(struct mtd_info *mtd,
 		case OMAP_ECC_BCH8_CODE_HW_DETECTION_SW:
 			/* Add constant polynomial to remainder, so that
 			 * ECC of blank pages results in 0x0 on reading back */
-			for (i = 0; i < eccbytes; i++)
-				ecc_calc[i] ^= bch8_polynomial[i];
+			for (j = 0; j < eccbytes; j++)
+				ecc_calc[j] ^= bch8_polynomial[j];
 			break;
 		case OMAP_ECC_BCH8_CODE_HW:
 			/* Set 14th ECC byte as 0x0 for ROM compatibility */
 			ecc_calc[eccbytes - 1] = 0x0;
 			break;
+		case OMAP_ECC_BCH16_CODE_HW:
+			break;
 		default:
 			return -EINVAL;
 		}
@@ -1237,6 +1292,7 @@ static int __maybe_unused omap_calculate_ecc_bch(struct mtd_info *mtd,
 	return 0;
 }
 
+#ifdef CONFIG_MTD_NAND_OMAP_BCH
 /**
  * erased_sector_bitflips - count bit flips
  * @data:	data sector buffer
@@ -1276,7 +1332,6 @@ static int erased_sector_bitflips(u_char *data, u_char *oob,
 	return flip_bits;
 }
 
-#ifdef CONFIG_MTD_NAND_OMAP_BCH
 /**
  * omap_elm_correct_data - corrects page data area in case error reported
  * @mtd:	MTD device structure
@@ -1318,6 +1373,10 @@ static int omap_elm_correct_data(struct mtd_info *mtd, u_char *data,
 		actual_eccbytes = ecc->bytes - 1;
 		erased_ecc_vec = bch8_vector;
 		break;
+	case OMAP_ECC_BCH16_CODE_HW:
+		actual_eccbytes = ecc->bytes;
+		erased_ecc_vec = bch16_vector;
+		break;
 	default:
 		pr_err("invalid driver configuration\n");
 		return -EINVAL;
@@ -1382,7 +1441,7 @@ static int omap_elm_correct_data(struct mtd_info *mtd, u_char *data,
 
 	/* Check if any error reported */
 	if (!is_error_reported)
-		return 0;
+		return stat;
 
 	/* Decode BCH error using ELM module */
 	elm_decode_bch_error_page(info->elm_dev, ecc_vec, err_vec);
@@ -1401,6 +1460,7 @@ static int omap_elm_correct_data(struct mtd_info *mtd, u_char *data,
 						BCH4_BIT_PAD;
 					break;
 				case OMAP_ECC_BCH8_CODE_HW:
+				case OMAP_ECC_BCH16_CODE_HW:
 					pos = err_vec[i].error_loc[j];
 					break;
 				default:
@@ -1912,6 +1972,40 @@ static int omap_nand_probe(struct platform_device *pdev)
 		goto return_error;
 #endif
 
+	case OMAP_ECC_BCH16_CODE_HW:
+#ifdef CONFIG_MTD_NAND_OMAP_BCH
+		pr_info("using OMAP_ECC_BCH16_CODE_HW ECC scheme\n");
+		nand_chip->ecc.mode		= NAND_ECC_HW;
+		nand_chip->ecc.size		= 512;
+		nand_chip->ecc.bytes		= 26;
+		nand_chip->ecc.strength		= 16;
+		nand_chip->ecc.hwctl		= omap_enable_hwecc_bch;
+		nand_chip->ecc.correct		= omap_elm_correct_data;
+		nand_chip->ecc.calculate	= omap_calculate_ecc_bch;
+		nand_chip->ecc.read_page	= omap_read_page_bch;
+		nand_chip->ecc.write_page	= omap_write_page_bch;
+		/* This ECC scheme requires ELM H/W block */
+		err = is_elm_present(info, pdata->elm_of_node, BCH16_ECC);
+		if (err < 0) {
+			pr_err("ELM is required for this ECC scheme\n");
+			goto return_error;
+		}
+		/* define ECC layout */
+		ecclayout->eccbytes		= nand_chip->ecc.bytes *
+							(mtd->writesize /
+							nand_chip->ecc.size);
+		oob_index			= BADBLOCK_MARKER_LENGTH;
+		for (i = 0; i < ecclayout->eccbytes; i++, oob_index++)
+			ecclayout->eccpos[i]	= oob_index;
+		/* reserved marker already included in ecclayout->eccbytes */
+		ecclayout->oobfree->offset	=
+				ecclayout->eccpos[ecclayout->eccbytes - 1] + 1;
+		break;
+#else
+		pr_err("nand: error: CONFIG_MTD_NAND_OMAP_BCH not enabled\n");
+		err = -EINVAL;
+		goto return_error;
+#endif
 	default:
 		pr_err("nand: error: invalid or unsupported ECC scheme\n");
 		err = -EINVAL;
diff --git a/drivers/mtd/nand/orion_nand.c b/drivers/mtd/nand/orion_nand.c
index dd7fe817eafb..471b4df3a5ac 100644
--- a/drivers/mtd/nand/orion_nand.c
+++ b/drivers/mtd/nand/orion_nand.c
@@ -214,7 +214,7 @@ static int orion_nand_remove(struct platform_device *pdev)
 }
 
 #ifdef CONFIG_OF
-static struct of_device_id orion_nand_of_match_table[] = {
+static const struct of_device_id orion_nand_of_match_table[] = {
 	{ .compatible = "marvell,orion-nand", },
 	{},
 };
diff --git a/drivers/mtd/nand/pxa3xx_nand.c b/drivers/mtd/nand/pxa3xx_nand.c
index 7588fe2c127f..96b0b1d27df1 100644
--- a/drivers/mtd/nand/pxa3xx_nand.c
+++ b/drivers/mtd/nand/pxa3xx_nand.c
@@ -127,10 +127,10 @@
 
 /* macros for registers read/write */
 #define nand_writel(info, off, val)	\
-	__raw_writel((val), (info)->mmio_base + (off))
+	writel_relaxed((val), (info)->mmio_base + (off))
 
 #define nand_readl(info, off)		\
-	__raw_readl((info)->mmio_base + (off))
+	readl_relaxed((info)->mmio_base + (off))
 
 /* error code and state */
 enum {
@@ -337,7 +337,7 @@ static struct nand_ecclayout ecc_layout_4KB_bch8bit = {
 /* convert nano-seconds to nand flash controller clock cycles */
 #define ns2cycle(ns, clk)	(int)((ns) * (clk / 1000000) / 1000)
 
-static struct of_device_id pxa3xx_nand_dt_ids[] = {
+static const struct of_device_id pxa3xx_nand_dt_ids[] = {
 	{
 		.compatible = "marvell,pxa3xx-nand",
 		.data       = (void *)PXA3XX_NAND_VARIANT_PXA,
@@ -1354,7 +1354,6 @@ static int pxa_ecc_init(struct pxa3xx_nand_info *info,
 		ecc->mode = NAND_ECC_HW;
 		ecc->size = 512;
 		ecc->strength = 1;
-		return 1;
 
 	} else if (strength == 1 && ecc_stepsize == 512 && page_size == 512) {
 		info->chunk_size = 512;
@@ -1363,7 +1362,6 @@ static int pxa_ecc_init(struct pxa3xx_nand_info *info,
 		ecc->mode = NAND_ECC_HW;
 		ecc->size = 512;
 		ecc->strength = 1;
-		return 1;
 
 	/*
 	 * Required ECC: 4-bit correction per 512 bytes
@@ -1378,7 +1376,6 @@ static int pxa_ecc_init(struct pxa3xx_nand_info *info,
 		ecc->size = info->chunk_size;
 		ecc->layout = &ecc_layout_2KB_bch4bit;
 		ecc->strength = 16;
-		return 1;
 
 	} else if (strength == 4 && ecc_stepsize == 512 && page_size == 4096) {
 		info->ecc_bch = 1;
@@ -1389,7 +1386,6 @@ static int pxa_ecc_init(struct pxa3xx_nand_info *info,
 		ecc->size = info->chunk_size;
 		ecc->layout = &ecc_layout_4KB_bch4bit;
 		ecc->strength = 16;
-		return 1;
 
 	/*
 	 * Required ECC: 8-bit correction per 512 bytes
@@ -1404,8 +1400,15 @@ static int pxa_ecc_init(struct pxa3xx_nand_info *info,
 		ecc->size = info->chunk_size;
 		ecc->layout = &ecc_layout_4KB_bch8bit;
 		ecc->strength = 16;
-		return 1;
+	} else {
+		dev_err(&info->pdev->dev,
+			"ECC strength %d at page size %d is not supported\n",
+			strength, page_size);
+		return -ENODEV;
 	}
+
+	dev_info(&info->pdev->dev, "ECC strength %d, ECC step size %d\n",
+		 ecc->strength, ecc->size);
 	return 0;
 }
 
@@ -1516,8 +1519,13 @@ KEEP_CONFIG:
 		}
 	}
 
-	ecc_strength = chip->ecc_strength_ds;
-	ecc_step = chip->ecc_step_ds;
+	if (pdata->ecc_strength && pdata->ecc_step_size) {
+		ecc_strength = pdata->ecc_strength;
+		ecc_step = pdata->ecc_step_size;
+	} else {
+		ecc_strength = chip->ecc_strength_ds;
+		ecc_step = chip->ecc_step_ds;
+	}
 
 	/* Set default ECC strength requirements on non-ONFI devices */
 	if (ecc_strength < 1 && ecc_step < 1) {
@@ -1527,12 +1535,8 @@ KEEP_CONFIG:
 
 	ret = pxa_ecc_init(info, &chip->ecc, ecc_strength,
 			   ecc_step, mtd->writesize);
-	if (!ret) {
-		dev_err(&info->pdev->dev,
-			"ECC strength %d at page size %d is not supported\n",
-			ecc_strength, mtd->writesize);
-		return -ENODEV;
-	}
+	if (ret)
+		return ret;
 
 	/* calculate addressing information */
 	if (mtd->writesize >= 2048)
@@ -1730,6 +1734,14 @@ static int pxa3xx_nand_probe_dt(struct platform_device *pdev)
 	of_property_read_u32(np, "num-cs", &pdata->num_cs);
 	pdata->flash_bbt = of_get_nand_on_flash_bbt(np);
 
+	pdata->ecc_strength = of_get_nand_ecc_strength(np);
+	if (pdata->ecc_strength < 0)
+		pdata->ecc_strength = 0;
+
+	pdata->ecc_step_size = of_get_nand_ecc_step_size(np);
+	if (pdata->ecc_step_size < 0)
+		pdata->ecc_step_size = 0;
+
 	pdev->dev.platform_data = pdata;
 
 	return 0;
diff --git a/drivers/mtd/nand/r852.c b/drivers/mtd/nand/r852.c
index 325930db3f04..baea83f4dea8 100644
--- a/drivers/mtd/nand/r852.c
+++ b/drivers/mtd/nand/r852.c
@@ -245,7 +245,7 @@ static void r852_write_buf(struct mtd_info *mtd, const uint8_t *buf, int len)
 	}
 
 	/* write DWORD chinks - faster */
-	while (len) {
+	while (len >= 4) {
 		reg = buf[0] | buf[1] << 8 | buf[2] << 16 | buf[3] << 24;
 		r852_write_reg_dword(dev, R852_DATALINE, reg);
 		buf += 4;
@@ -254,8 +254,10 @@ static void r852_write_buf(struct mtd_info *mtd, const uint8_t *buf, int len)
 	}
 
 	/* write rest */
-	while (len)
+	while (len > 0) {
 		r852_write_reg(dev, R852_DATALINE, *buf++);
+		len--;
+	}
 }
 
 /*
diff --git a/drivers/mtd/onenand/samsung.c b/drivers/mtd/onenand/samsung.c
index b1a792fd1c23..efb819c3df2f 100644
--- a/drivers/mtd/onenand/samsung.c
+++ b/drivers/mtd/onenand/samsung.c
@@ -537,9 +537,9 @@ static int onenand_write_bufferram(struct mtd_info *mtd, int area,
 	return 0;
 }
 
-static int (*s5pc110_dma_ops)(void *dst, void *src, size_t count, int direction);
+static int (*s5pc110_dma_ops)(dma_addr_t dst, dma_addr_t src, size_t count, int direction);
 
-static int s5pc110_dma_poll(void *dst, void *src, size_t count, int direction)
+static int s5pc110_dma_poll(dma_addr_t dst, dma_addr_t src, size_t count, int direction)
 {
 	void __iomem *base = onenand->dma_addr;
 	int status;
@@ -605,7 +605,7 @@ static irqreturn_t s5pc110_onenand_irq(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
-static int s5pc110_dma_irq(void *dst, void *src, size_t count, int direction)
+static int s5pc110_dma_irq(dma_addr_t dst, dma_addr_t src, size_t count, int direction)
 {
 	void __iomem *base = onenand->dma_addr;
 	int status;
@@ -686,7 +686,7 @@ static int s5pc110_read_bufferram(struct mtd_info *mtd, int area,
 		dev_err(dev, "Couldn't map a %d byte buffer for DMA\n", count);
 		goto normal;
 	}
-	err = s5pc110_dma_ops((void *) dma_dst, (void *) dma_src,
+	err = s5pc110_dma_ops(dma_dst, dma_src,
 			count, S5PC110_DMA_DIR_READ);
 
 	if (page_dma)
diff --git a/drivers/mtd/spi-nor/Kconfig b/drivers/mtd/spi-nor/Kconfig
new file mode 100644
index 000000000000..f8acfa4310ef
--- /dev/null
+++ b/drivers/mtd/spi-nor/Kconfig
@@ -0,0 +1,17 @@
+menuconfig MTD_SPI_NOR
+	tristate "SPI-NOR device support"
+	depends on MTD
+	help
+	  This is the framework for the SPI NOR which can be used by the SPI
+	  device drivers and the SPI-NOR device driver.
+
+if MTD_SPI_NOR
+
+config SPI_FSL_QUADSPI
+	tristate "Freescale Quad SPI controller"
+	depends on ARCH_MXC
+	help
+	  This enables support for the Quad SPI controller in master mode.
+	  We only connect the NOR to this controller now.
+
+endif # MTD_SPI_NOR
diff --git a/drivers/mtd/spi-nor/Makefile b/drivers/mtd/spi-nor/Makefile
new file mode 100644
index 000000000000..6a7ce1462247
--- /dev/null
+++ b/drivers/mtd/spi-nor/Makefile
@@ -0,0 +1,2 @@
+obj-$(CONFIG_MTD_SPI_NOR)	+= spi-nor.o
+obj-$(CONFIG_SPI_FSL_QUADSPI)	+= fsl-quadspi.o
diff --git a/drivers/mtd/spi-nor/fsl-quadspi.c b/drivers/mtd/spi-nor/fsl-quadspi.c
new file mode 100644
index 000000000000..8d659a2888d5
--- /dev/null
+++ b/drivers/mtd/spi-nor/fsl-quadspi.c
@@ -0,0 +1,1009 @@
+/*
+ * Freescale QuadSPI driver.
+ *
+ * Copyright (C) 2013 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/errno.h>
+#include <linux/platform_device.h>
+#include <linux/sched.h>
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/timer.h>
+#include <linux/jiffies.h>
+#include <linux/completion.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mtd/spi-nor.h>
+
+/* The registers */
+#define QUADSPI_MCR			0x00
+#define QUADSPI_MCR_RESERVED_SHIFT	16
+#define QUADSPI_MCR_RESERVED_MASK	(0xF << QUADSPI_MCR_RESERVED_SHIFT)
+#define QUADSPI_MCR_MDIS_SHIFT		14
+#define QUADSPI_MCR_MDIS_MASK		(1 << QUADSPI_MCR_MDIS_SHIFT)
+#define QUADSPI_MCR_CLR_TXF_SHIFT	11
+#define QUADSPI_MCR_CLR_TXF_MASK	(1 << QUADSPI_MCR_CLR_TXF_SHIFT)
+#define QUADSPI_MCR_CLR_RXF_SHIFT	10
+#define QUADSPI_MCR_CLR_RXF_MASK	(1 << QUADSPI_MCR_CLR_RXF_SHIFT)
+#define QUADSPI_MCR_DDR_EN_SHIFT	7
+#define QUADSPI_MCR_DDR_EN_MASK		(1 << QUADSPI_MCR_DDR_EN_SHIFT)
+#define QUADSPI_MCR_END_CFG_SHIFT	2
+#define QUADSPI_MCR_END_CFG_MASK	(3 << QUADSPI_MCR_END_CFG_SHIFT)
+#define QUADSPI_MCR_SWRSTHD_SHIFT	1
+#define QUADSPI_MCR_SWRSTHD_MASK	(1 << QUADSPI_MCR_SWRSTHD_SHIFT)
+#define QUADSPI_MCR_SWRSTSD_SHIFT	0
+#define QUADSPI_MCR_SWRSTSD_MASK	(1 << QUADSPI_MCR_SWRSTSD_SHIFT)
+
+#define QUADSPI_IPCR			0x08
+#define QUADSPI_IPCR_SEQID_SHIFT	24
+#define QUADSPI_IPCR_SEQID_MASK		(0xF << QUADSPI_IPCR_SEQID_SHIFT)
+
+#define QUADSPI_BUF0CR			0x10
+#define QUADSPI_BUF1CR			0x14
+#define QUADSPI_BUF2CR			0x18
+#define QUADSPI_BUFXCR_INVALID_MSTRID	0xe
+
+#define QUADSPI_BUF3CR			0x1c
+#define QUADSPI_BUF3CR_ALLMST_SHIFT	31
+#define QUADSPI_BUF3CR_ALLMST		(1 << QUADSPI_BUF3CR_ALLMST_SHIFT)
+
+#define QUADSPI_BFGENCR			0x20
+#define QUADSPI_BFGENCR_PAR_EN_SHIFT	16
+#define QUADSPI_BFGENCR_PAR_EN_MASK	(1 << (QUADSPI_BFGENCR_PAR_EN_SHIFT))
+#define QUADSPI_BFGENCR_SEQID_SHIFT	12
+#define QUADSPI_BFGENCR_SEQID_MASK	(0xF << QUADSPI_BFGENCR_SEQID_SHIFT)
+
+#define QUADSPI_BUF0IND			0x30
+#define QUADSPI_BUF1IND			0x34
+#define QUADSPI_BUF2IND			0x38
+#define QUADSPI_SFAR			0x100
+
+#define QUADSPI_SMPR			0x108
+#define QUADSPI_SMPR_DDRSMP_SHIFT	16
+#define QUADSPI_SMPR_DDRSMP_MASK	(7 << QUADSPI_SMPR_DDRSMP_SHIFT)
+#define QUADSPI_SMPR_FSDLY_SHIFT	6
+#define QUADSPI_SMPR_FSDLY_MASK		(1 << QUADSPI_SMPR_FSDLY_SHIFT)
+#define QUADSPI_SMPR_FSPHS_SHIFT	5
+#define QUADSPI_SMPR_FSPHS_MASK		(1 << QUADSPI_SMPR_FSPHS_SHIFT)
+#define QUADSPI_SMPR_HSENA_SHIFT	0
+#define QUADSPI_SMPR_HSENA_MASK		(1 << QUADSPI_SMPR_HSENA_SHIFT)
+
+#define QUADSPI_RBSR			0x10c
+#define QUADSPI_RBSR_RDBFL_SHIFT	8
+#define QUADSPI_RBSR_RDBFL_MASK		(0x3F << QUADSPI_RBSR_RDBFL_SHIFT)
+
+#define QUADSPI_RBCT			0x110
+#define QUADSPI_RBCT_WMRK_MASK		0x1F
+#define QUADSPI_RBCT_RXBRD_SHIFT	8
+#define QUADSPI_RBCT_RXBRD_USEIPS	(0x1 << QUADSPI_RBCT_RXBRD_SHIFT)
+
+#define QUADSPI_TBSR			0x150
+#define QUADSPI_TBDR			0x154
+#define QUADSPI_SR			0x15c
+#define QUADSPI_SR_IP_ACC_SHIFT		1
+#define QUADSPI_SR_IP_ACC_MASK		(0x1 << QUADSPI_SR_IP_ACC_SHIFT)
+#define QUADSPI_SR_AHB_ACC_SHIFT	2
+#define QUADSPI_SR_AHB_ACC_MASK		(0x1 << QUADSPI_SR_AHB_ACC_SHIFT)
+
+#define QUADSPI_FR			0x160
+#define QUADSPI_FR_TFF_MASK		0x1
+
+#define QUADSPI_SFA1AD			0x180
+#define QUADSPI_SFA2AD			0x184
+#define QUADSPI_SFB1AD			0x188
+#define QUADSPI_SFB2AD			0x18c
+#define QUADSPI_RBDR			0x200
+
+#define QUADSPI_LUTKEY			0x300
+#define QUADSPI_LUTKEY_VALUE		0x5AF05AF0
+
+#define QUADSPI_LCKCR			0x304
+#define QUADSPI_LCKER_LOCK		0x1
+#define QUADSPI_LCKER_UNLOCK		0x2
+
+#define QUADSPI_RSER			0x164
+#define QUADSPI_RSER_TFIE		(0x1 << 0)
+
+#define QUADSPI_LUT_BASE		0x310
+
+/*
+ * The definition of the LUT register shows below:
+ *
+ *  ---------------------------------------------------
+ *  | INSTR1 | PAD1 | OPRND1 | INSTR0 | PAD0 | OPRND0 |
+ *  ---------------------------------------------------
+ */
+#define OPRND0_SHIFT		0
+#define PAD0_SHIFT		8
+#define INSTR0_SHIFT		10
+#define OPRND1_SHIFT		16
+
+/* Instruction set for the LUT register. */
+#define LUT_STOP		0
+#define LUT_CMD			1
+#define LUT_ADDR		2
+#define LUT_DUMMY		3
+#define LUT_MODE		4
+#define LUT_MODE2		5
+#define LUT_MODE4		6
+#define LUT_READ		7
+#define LUT_WRITE		8
+#define LUT_JMP_ON_CS		9
+#define LUT_ADDR_DDR		10
+#define LUT_MODE_DDR		11
+#define LUT_MODE2_DDR		12
+#define LUT_MODE4_DDR		13
+#define LUT_READ_DDR		14
+#define LUT_WRITE_DDR		15
+#define LUT_DATA_LEARN		16
+
+/*
+ * The PAD definitions for LUT register.
+ *
+ * The pad stands for the lines number of IO[0:3].
+ * For example, the Quad read need four IO lines, so you should
+ * set LUT_PAD4 which means we use four IO lines.
+ */
+#define LUT_PAD1		0
+#define LUT_PAD2		1
+#define LUT_PAD4		2
+
+/* Oprands for the LUT register. */
+#define ADDR24BIT		0x18
+#define ADDR32BIT		0x20
+
+/* Macros for constructing the LUT register. */
+#define LUT0(ins, pad, opr)						\
+		(((opr) << OPRND0_SHIFT) | ((LUT_##pad) << PAD0_SHIFT) | \
+		((LUT_##ins) << INSTR0_SHIFT))
+
+#define LUT1(ins, pad, opr)	(LUT0(ins, pad, opr) << OPRND1_SHIFT)
+
+/* other macros for LUT register. */
+#define QUADSPI_LUT(x)          (QUADSPI_LUT_BASE + (x) * 4)
+#define QUADSPI_LUT_NUM		64
+
+/* SEQID -- we can have 16 seqids at most. */
+#define SEQID_QUAD_READ		0
+#define SEQID_WREN		1
+#define SEQID_WRDI		2
+#define SEQID_RDSR		3
+#define SEQID_SE		4
+#define SEQID_CHIP_ERASE	5
+#define SEQID_PP		6
+#define SEQID_RDID		7
+#define SEQID_WRSR		8
+#define SEQID_RDCR		9
+#define SEQID_EN4B		10
+#define SEQID_BRWR		11
+
+enum fsl_qspi_devtype {
+	FSL_QUADSPI_VYBRID,
+	FSL_QUADSPI_IMX6SX,
+};
+
+struct fsl_qspi_devtype_data {
+	enum fsl_qspi_devtype devtype;
+	int rxfifo;
+	int txfifo;
+};
+
+static struct fsl_qspi_devtype_data vybrid_data = {
+	.devtype = FSL_QUADSPI_VYBRID,
+	.rxfifo = 128,
+	.txfifo = 64
+};
+
+static struct fsl_qspi_devtype_data imx6sx_data = {
+	.devtype = FSL_QUADSPI_IMX6SX,
+	.rxfifo = 128,
+	.txfifo = 512
+};
+
+#define FSL_QSPI_MAX_CHIP	4
+struct fsl_qspi {
+	struct mtd_info mtd[FSL_QSPI_MAX_CHIP];
+	struct spi_nor nor[FSL_QSPI_MAX_CHIP];
+	void __iomem *iobase;
+	void __iomem *ahb_base; /* Used when read from AHB bus */
+	u32 memmap_phy;
+	struct clk *clk, *clk_en;
+	struct device *dev;
+	struct completion c;
+	struct fsl_qspi_devtype_data *devtype_data;
+	u32 nor_size;
+	u32 nor_num;
+	u32 clk_rate;
+	unsigned int chip_base_addr; /* We may support two chips. */
+};
+
+static inline int is_vybrid_qspi(struct fsl_qspi *q)
+{
+	return q->devtype_data->devtype == FSL_QUADSPI_VYBRID;
+}
+
+static inline int is_imx6sx_qspi(struct fsl_qspi *q)
+{
+	return q->devtype_data->devtype == FSL_QUADSPI_IMX6SX;
+}
+
+/*
+ * An IC bug makes us to re-arrange the 32-bit data.
+ * The following chips, such as IMX6SLX, have fixed this bug.
+ */
+static inline u32 fsl_qspi_endian_xchg(struct fsl_qspi *q, u32 a)
+{
+	return is_vybrid_qspi(q) ? __swab32(a) : a;
+}
+
+static inline void fsl_qspi_unlock_lut(struct fsl_qspi *q)
+{
+	writel(QUADSPI_LUTKEY_VALUE, q->iobase + QUADSPI_LUTKEY);
+	writel(QUADSPI_LCKER_UNLOCK, q->iobase + QUADSPI_LCKCR);
+}
+
+static inline void fsl_qspi_lock_lut(struct fsl_qspi *q)
+{
+	writel(QUADSPI_LUTKEY_VALUE, q->iobase + QUADSPI_LUTKEY);
+	writel(QUADSPI_LCKER_LOCK, q->iobase + QUADSPI_LCKCR);
+}
+
+static irqreturn_t fsl_qspi_irq_handler(int irq, void *dev_id)
+{
+	struct fsl_qspi *q = dev_id;
+	u32 reg;
+
+	/* clear interrupt */
+	reg = readl(q->iobase + QUADSPI_FR);
+	writel(reg, q->iobase + QUADSPI_FR);
+
+	if (reg & QUADSPI_FR_TFF_MASK)
+		complete(&q->c);
+
+	dev_dbg(q->dev, "QUADSPI_FR : 0x%.8x:0x%.8x\n", q->chip_base_addr, reg);
+	return IRQ_HANDLED;
+}
+
+static void fsl_qspi_init_lut(struct fsl_qspi *q)
+{
+	void __iomem *base = q->iobase;
+	int rxfifo = q->devtype_data->rxfifo;
+	u32 lut_base;
+	u8 cmd, addrlen, dummy;
+	int i;
+
+	fsl_qspi_unlock_lut(q);
+
+	/* Clear all the LUT table */
+	for (i = 0; i < QUADSPI_LUT_NUM; i++)
+		writel(0, base + QUADSPI_LUT_BASE + i * 4);
+
+	/* Quad Read */
+	lut_base = SEQID_QUAD_READ * 4;
+
+	if (q->nor_size <= SZ_16M) {
+		cmd = SPINOR_OP_READ_1_1_4;
+		addrlen = ADDR24BIT;
+		dummy = 8;
+	} else {
+		/* use the 4-byte address */
+		cmd = SPINOR_OP_READ_1_1_4;
+		addrlen = ADDR32BIT;
+		dummy = 8;
+	}
+
+	writel(LUT0(CMD, PAD1, cmd) | LUT1(ADDR, PAD1, addrlen),
+			base + QUADSPI_LUT(lut_base));
+	writel(LUT0(DUMMY, PAD1, dummy) | LUT1(READ, PAD4, rxfifo),
+			base + QUADSPI_LUT(lut_base + 1));
+
+	/* Write enable */
+	lut_base = SEQID_WREN * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_WREN), base + QUADSPI_LUT(lut_base));
+
+	/* Page Program */
+	lut_base = SEQID_PP * 4;
+
+	if (q->nor_size <= SZ_16M) {
+		cmd = SPINOR_OP_PP;
+		addrlen = ADDR24BIT;
+	} else {
+		/* use the 4-byte address */
+		cmd = SPINOR_OP_PP;
+		addrlen = ADDR32BIT;
+	}
+
+	writel(LUT0(CMD, PAD1, cmd) | LUT1(ADDR, PAD1, addrlen),
+			base + QUADSPI_LUT(lut_base));
+	writel(LUT0(WRITE, PAD1, 0), base + QUADSPI_LUT(lut_base + 1));
+
+	/* Read Status */
+	lut_base = SEQID_RDSR * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_RDSR) | LUT1(READ, PAD1, 0x1),
+			base + QUADSPI_LUT(lut_base));
+
+	/* Erase a sector */
+	lut_base = SEQID_SE * 4;
+
+	if (q->nor_size <= SZ_16M) {
+		cmd = SPINOR_OP_SE;
+		addrlen = ADDR24BIT;
+	} else {
+		/* use the 4-byte address */
+		cmd = SPINOR_OP_SE;
+		addrlen = ADDR32BIT;
+	}
+
+	writel(LUT0(CMD, PAD1, cmd) | LUT1(ADDR, PAD1, addrlen),
+			base + QUADSPI_LUT(lut_base));
+
+	/* Erase the whole chip */
+	lut_base = SEQID_CHIP_ERASE * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_CHIP_ERASE),
+			base + QUADSPI_LUT(lut_base));
+
+	/* READ ID */
+	lut_base = SEQID_RDID * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_RDID) | LUT1(READ, PAD1, 0x8),
+			base + QUADSPI_LUT(lut_base));
+
+	/* Write Register */
+	lut_base = SEQID_WRSR * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_WRSR) | LUT1(WRITE, PAD1, 0x2),
+			base + QUADSPI_LUT(lut_base));
+
+	/* Read Configuration Register */
+	lut_base = SEQID_RDCR * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_RDCR) | LUT1(READ, PAD1, 0x1),
+			base + QUADSPI_LUT(lut_base));
+
+	/* Write disable */
+	lut_base = SEQID_WRDI * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_WRDI), base + QUADSPI_LUT(lut_base));
+
+	/* Enter 4 Byte Mode (Micron) */
+	lut_base = SEQID_EN4B * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_EN4B), base + QUADSPI_LUT(lut_base));
+
+	/* Enter 4 Byte Mode (Spansion) */
+	lut_base = SEQID_BRWR * 4;
+	writel(LUT0(CMD, PAD1, SPINOR_OP_BRWR), base + QUADSPI_LUT(lut_base));
+
+	fsl_qspi_lock_lut(q);
+}
+
+/* Get the SEQID for the command */
+static int fsl_qspi_get_seqid(struct fsl_qspi *q, u8 cmd)
+{
+	switch (cmd) {
+	case SPINOR_OP_READ_1_1_4:
+		return SEQID_QUAD_READ;
+	case SPINOR_OP_WREN:
+		return SEQID_WREN;
+	case SPINOR_OP_WRDI:
+		return SEQID_WRDI;
+	case SPINOR_OP_RDSR:
+		return SEQID_RDSR;
+	case SPINOR_OP_SE:
+		return SEQID_SE;
+	case SPINOR_OP_CHIP_ERASE:
+		return SEQID_CHIP_ERASE;
+	case SPINOR_OP_PP:
+		return SEQID_PP;
+	case SPINOR_OP_RDID:
+		return SEQID_RDID;
+	case SPINOR_OP_WRSR:
+		return SEQID_WRSR;
+	case SPINOR_OP_RDCR:
+		return SEQID_RDCR;
+	case SPINOR_OP_EN4B:
+		return SEQID_EN4B;
+	case SPINOR_OP_BRWR:
+		return SEQID_BRWR;
+	default:
+		dev_err(q->dev, "Unsupported cmd 0x%.2x\n", cmd);
+		break;
+	}
+	return -EINVAL;
+}
+
+static int
+fsl_qspi_runcmd(struct fsl_qspi *q, u8 cmd, unsigned int addr, int len)
+{
+	void __iomem *base = q->iobase;
+	int seqid;
+	u32 reg, reg2;
+	int err;
+
+	init_completion(&q->c);
+	dev_dbg(q->dev, "to 0x%.8x:0x%.8x, len:%d, cmd:%.2x\n",
+			q->chip_base_addr, addr, len, cmd);
+
+	/* save the reg */
+	reg = readl(base + QUADSPI_MCR);
+
+	writel(q->memmap_phy + q->chip_base_addr + addr, base + QUADSPI_SFAR);
+	writel(QUADSPI_RBCT_WMRK_MASK | QUADSPI_RBCT_RXBRD_USEIPS,
+			base + QUADSPI_RBCT);
+	writel(reg | QUADSPI_MCR_CLR_RXF_MASK, base + QUADSPI_MCR);
+
+	do {
+		reg2 = readl(base + QUADSPI_SR);
+		if (reg2 & (QUADSPI_SR_IP_ACC_MASK | QUADSPI_SR_AHB_ACC_MASK)) {
+			udelay(1);
+			dev_dbg(q->dev, "The controller is busy, 0x%x\n", reg2);
+			continue;
+		}
+		break;
+	} while (1);
+
+	/* trigger the LUT now */
+	seqid = fsl_qspi_get_seqid(q, cmd);
+	writel((seqid << QUADSPI_IPCR_SEQID_SHIFT) | len, base + QUADSPI_IPCR);
+
+	/* Wait for the interrupt. */
+	err = wait_for_completion_timeout(&q->c, msecs_to_jiffies(1000));
+	if (!err) {
+		dev_err(q->dev,
+			"cmd 0x%.2x timeout, addr@%.8x, FR:0x%.8x, SR:0x%.8x\n",
+			cmd, addr, readl(base + QUADSPI_FR),
+			readl(base + QUADSPI_SR));
+		err = -ETIMEDOUT;
+	} else {
+		err = 0;
+	}
+
+	/* restore the MCR */
+	writel(reg, base + QUADSPI_MCR);
+
+	return err;
+}
+
+/* Read out the data from the QUADSPI_RBDR buffer registers. */
+static void fsl_qspi_read_data(struct fsl_qspi *q, int len, u8 *rxbuf)
+{
+	u32 tmp;
+	int i = 0;
+
+	while (len > 0) {
+		tmp = readl(q->iobase + QUADSPI_RBDR + i * 4);
+		tmp = fsl_qspi_endian_xchg(q, tmp);
+		dev_dbg(q->dev, "chip addr:0x%.8x, rcv:0x%.8x\n",
+				q->chip_base_addr, tmp);
+
+		if (len >= 4) {
+			*((u32 *)rxbuf) = tmp;
+			rxbuf += 4;
+		} else {
+			memcpy(rxbuf, &tmp, len);
+			break;
+		}
+
+		len -= 4;
+		i++;
+	}
+}
+
+/*
+ * If we have changed the content of the flash by writing or erasing,
+ * we need to invalidate the AHB buffer. If we do not do so, we may read out
+ * the wrong data. The spec tells us reset the AHB domain and Serial Flash
+ * domain at the same time.
+ */
+static inline void fsl_qspi_invalid(struct fsl_qspi *q)
+{
+	u32 reg;
+
+	reg = readl(q->iobase + QUADSPI_MCR);
+	reg |= QUADSPI_MCR_SWRSTHD_MASK | QUADSPI_MCR_SWRSTSD_MASK;
+	writel(reg, q->iobase + QUADSPI_MCR);
+
+	/*
+	 * The minimum delay : 1 AHB + 2 SFCK clocks.
+	 * Delay 1 us is enough.
+	 */
+	udelay(1);
+
+	reg &= ~(QUADSPI_MCR_SWRSTHD_MASK | QUADSPI_MCR_SWRSTSD_MASK);
+	writel(reg, q->iobase + QUADSPI_MCR);
+}
+
+static int fsl_qspi_nor_write(struct fsl_qspi *q, struct spi_nor *nor,
+				u8 opcode, unsigned int to, u32 *txbuf,
+				unsigned count, size_t *retlen)
+{
+	int ret, i, j;
+	u32 tmp;
+
+	dev_dbg(q->dev, "to 0x%.8x:0x%.8x, len : %d\n",
+		q->chip_base_addr, to, count);
+
+	/* clear the TX FIFO. */
+	tmp = readl(q->iobase + QUADSPI_MCR);
+	writel(tmp | QUADSPI_MCR_CLR_RXF_MASK, q->iobase + QUADSPI_MCR);
+
+	/* fill the TX data to the FIFO */
+	for (j = 0, i = ((count + 3) / 4); j < i; j++) {
+		tmp = fsl_qspi_endian_xchg(q, *txbuf);
+		writel(tmp, q->iobase + QUADSPI_TBDR);
+		txbuf++;
+	}
+
+	/* Trigger it */
+	ret = fsl_qspi_runcmd(q, opcode, to, count);
+
+	if (ret == 0 && retlen)
+		*retlen += count;
+
+	return ret;
+}
+
+static void fsl_qspi_set_map_addr(struct fsl_qspi *q)
+{
+	int nor_size = q->nor_size;
+	void __iomem *base = q->iobase;
+
+	writel(nor_size + q->memmap_phy, base + QUADSPI_SFA1AD);
+	writel(nor_size * 2 + q->memmap_phy, base + QUADSPI_SFA2AD);
+	writel(nor_size * 3 + q->memmap_phy, base + QUADSPI_SFB1AD);
+	writel(nor_size * 4 + q->memmap_phy, base + QUADSPI_SFB2AD);
+}
+
+/*
+ * There are two different ways to read out the data from the flash:
+ *  the "IP Command Read" and the "AHB Command Read".
+ *
+ * The IC guy suggests we use the "AHB Command Read" which is faster
+ * then the "IP Command Read". (What's more is that there is a bug in
+ * the "IP Command Read" in the Vybrid.)
+ *
+ * After we set up the registers for the "AHB Command Read", we can use
+ * the memcpy to read the data directly. A "missed" access to the buffer
+ * causes the controller to clear the buffer, and use the sequence pointed
+ * by the QUADSPI_BFGENCR[SEQID] to initiate a read from the flash.
+ */
+static void fsl_qspi_init_abh_read(struct fsl_qspi *q)
+{
+	void __iomem *base = q->iobase;
+	int seqid;
+
+	/* AHB configuration for access buffer 0/1/2 .*/
+	writel(QUADSPI_BUFXCR_INVALID_MSTRID, base + QUADSPI_BUF0CR);
+	writel(QUADSPI_BUFXCR_INVALID_MSTRID, base + QUADSPI_BUF1CR);
+	writel(QUADSPI_BUFXCR_INVALID_MSTRID, base + QUADSPI_BUF2CR);
+	writel(QUADSPI_BUF3CR_ALLMST, base + QUADSPI_BUF3CR);
+
+	/* We only use the buffer3 */
+	writel(0, base + QUADSPI_BUF0IND);
+	writel(0, base + QUADSPI_BUF1IND);
+	writel(0, base + QUADSPI_BUF2IND);
+
+	/* Set the default lut sequence for AHB Read. */
+	seqid = fsl_qspi_get_seqid(q, q->nor[0].read_opcode);
+	writel(seqid << QUADSPI_BFGENCR_SEQID_SHIFT,
+		q->iobase + QUADSPI_BFGENCR);
+}
+
+/* We use this function to do some basic init for spi_nor_scan(). */
+static int fsl_qspi_nor_setup(struct fsl_qspi *q)
+{
+	void __iomem *base = q->iobase;
+	u32 reg;
+	int ret;
+
+	/* the default frequency, we will change it in the future.*/
+	ret = clk_set_rate(q->clk, 66000000);
+	if (ret)
+		return ret;
+
+	/* Init the LUT table. */
+	fsl_qspi_init_lut(q);
+
+	/* Disable the module */
+	writel(QUADSPI_MCR_MDIS_MASK | QUADSPI_MCR_RESERVED_MASK,
+			base + QUADSPI_MCR);
+
+	reg = readl(base + QUADSPI_SMPR);
+	writel(reg & ~(QUADSPI_SMPR_FSDLY_MASK
+			| QUADSPI_SMPR_FSPHS_MASK
+			| QUADSPI_SMPR_HSENA_MASK
+			| QUADSPI_SMPR_DDRSMP_MASK), base + QUADSPI_SMPR);
+
+	/* Enable the module */
+	writel(QUADSPI_MCR_RESERVED_MASK | QUADSPI_MCR_END_CFG_MASK,
+			base + QUADSPI_MCR);
+
+	/* enable the interrupt */
+	writel(QUADSPI_RSER_TFIE, q->iobase + QUADSPI_RSER);
+
+	return 0;
+}
+
+static int fsl_qspi_nor_setup_last(struct fsl_qspi *q)
+{
+	unsigned long rate = q->clk_rate;
+	int ret;
+
+	if (is_imx6sx_qspi(q))
+		rate *= 4;
+
+	ret = clk_set_rate(q->clk, rate);
+	if (ret)
+		return ret;
+
+	/* Init the LUT table again. */
+	fsl_qspi_init_lut(q);
+
+	/* Init for AHB read */
+	fsl_qspi_init_abh_read(q);
+
+	return 0;
+}
+
+static struct of_device_id fsl_qspi_dt_ids[] = {
+	{ .compatible = "fsl,vf610-qspi", .data = (void *)&vybrid_data, },
+	{ .compatible = "fsl,imx6sx-qspi", .data = (void *)&imx6sx_data, },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, fsl_qspi_dt_ids);
+
+static void fsl_qspi_set_base_addr(struct fsl_qspi *q, struct spi_nor *nor)
+{
+	q->chip_base_addr = q->nor_size * (nor - q->nor);
+}
+
+static int fsl_qspi_read_reg(struct spi_nor *nor, u8 opcode, u8 *buf, int len)
+{
+	int ret;
+	struct fsl_qspi *q = nor->priv;
+
+	ret = fsl_qspi_runcmd(q, opcode, 0, len);
+	if (ret)
+		return ret;
+
+	fsl_qspi_read_data(q, len, buf);
+	return 0;
+}
+
+static int fsl_qspi_write_reg(struct spi_nor *nor, u8 opcode, u8 *buf, int len,
+			int write_enable)
+{
+	struct fsl_qspi *q = nor->priv;
+	int ret;
+
+	if (!buf) {
+		ret = fsl_qspi_runcmd(q, opcode, 0, 1);
+		if (ret)
+			return ret;
+
+		if (opcode == SPINOR_OP_CHIP_ERASE)
+			fsl_qspi_invalid(q);
+
+	} else if (len > 0) {
+		ret = fsl_qspi_nor_write(q, nor, opcode, 0,
+					(u32 *)buf, len, NULL);
+	} else {
+		dev_err(q->dev, "invalid cmd %d\n", opcode);
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static void fsl_qspi_write(struct spi_nor *nor, loff_t to,
+		size_t len, size_t *retlen, const u_char *buf)
+{
+	struct fsl_qspi *q = nor->priv;
+
+	fsl_qspi_nor_write(q, nor, nor->program_opcode, to,
+				(u32 *)buf, len, retlen);
+
+	/* invalid the data in the AHB buffer. */
+	fsl_qspi_invalid(q);
+}
+
+static int fsl_qspi_read(struct spi_nor *nor, loff_t from,
+		size_t len, size_t *retlen, u_char *buf)
+{
+	struct fsl_qspi *q = nor->priv;
+	u8 cmd = nor->read_opcode;
+	int ret;
+
+	dev_dbg(q->dev, "cmd [%x],read from (0x%p, 0x%.8x, 0x%.8x),len:%d\n",
+		cmd, q->ahb_base, q->chip_base_addr, (unsigned int)from, len);
+
+	/* Wait until the previous command is finished. */
+	ret = nor->wait_till_ready(nor);
+	if (ret)
+		return ret;
+
+	/* Read out the data directly from the AHB buffer.*/
+	memcpy(buf, q->ahb_base + q->chip_base_addr + from, len);
+
+	*retlen += len;
+	return 0;
+}
+
+static int fsl_qspi_erase(struct spi_nor *nor, loff_t offs)
+{
+	struct fsl_qspi *q = nor->priv;
+	int ret;
+
+	dev_dbg(nor->dev, "%dKiB at 0x%08x:0x%08x\n",
+		nor->mtd->erasesize / 1024, q->chip_base_addr, (u32)offs);
+
+	/* Wait until finished previous write command. */
+	ret = nor->wait_till_ready(nor);
+	if (ret)
+		return ret;
+
+	/* Send write enable, then erase commands. */
+	ret = nor->write_reg(nor, SPINOR_OP_WREN, NULL, 0, 0);
+	if (ret)
+		return ret;
+
+	ret = fsl_qspi_runcmd(q, nor->erase_opcode, offs, 0);
+	if (ret)
+		return ret;
+
+	fsl_qspi_invalid(q);
+	return 0;
+}
+
+static int fsl_qspi_prep(struct spi_nor *nor, enum spi_nor_ops ops)
+{
+	struct fsl_qspi *q = nor->priv;
+	int ret;
+
+	ret = clk_enable(q->clk_en);
+	if (ret)
+		return ret;
+
+	ret = clk_enable(q->clk);
+	if (ret) {
+		clk_disable(q->clk_en);
+		return ret;
+	}
+
+	fsl_qspi_set_base_addr(q, nor);
+	return 0;
+}
+
+static void fsl_qspi_unprep(struct spi_nor *nor, enum spi_nor_ops ops)
+{
+	struct fsl_qspi *q = nor->priv;
+
+	clk_disable(q->clk);
+	clk_disable(q->clk_en);
+}
+
+static int fsl_qspi_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct mtd_part_parser_data ppdata;
+	struct device *dev = &pdev->dev;
+	struct fsl_qspi *q;
+	struct resource *res;
+	struct spi_nor *nor;
+	struct mtd_info *mtd;
+	int ret, i = 0;
+	bool has_second_chip = false;
+	const struct of_device_id *of_id =
+			of_match_device(fsl_qspi_dt_ids, &pdev->dev);
+
+	q = devm_kzalloc(dev, sizeof(*q), GFP_KERNEL);
+	if (!q)
+		return -ENOMEM;
+
+	q->nor_num = of_get_child_count(dev->of_node);
+	if (!q->nor_num || q->nor_num > FSL_QSPI_MAX_CHIP)
+		return -ENODEV;
+
+	/* find the resources */
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "QuadSPI");
+	q->iobase = devm_ioremap_resource(dev, res);
+	if (IS_ERR(q->iobase)) {
+		ret = PTR_ERR(q->iobase);
+		goto map_failed;
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
+					"QuadSPI-memory");
+	q->ahb_base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(q->ahb_base)) {
+		ret = PTR_ERR(q->ahb_base);
+		goto map_failed;
+	}
+	q->memmap_phy = res->start;
+
+	/* find the clocks */
+	q->clk_en = devm_clk_get(dev, "qspi_en");
+	if (IS_ERR(q->clk_en)) {
+		ret = PTR_ERR(q->clk_en);
+		goto map_failed;
+	}
+
+	q->clk = devm_clk_get(dev, "qspi");
+	if (IS_ERR(q->clk)) {
+		ret = PTR_ERR(q->clk);
+		goto map_failed;
+	}
+
+	ret = clk_prepare_enable(q->clk_en);
+	if (ret) {
+		dev_err(dev, "can not enable the qspi_en clock\n");
+		goto map_failed;
+	}
+
+	ret = clk_prepare_enable(q->clk);
+	if (ret) {
+		clk_disable_unprepare(q->clk_en);
+		dev_err(dev, "can not enable the qspi clock\n");
+		goto map_failed;
+	}
+
+	/* find the irq */
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0) {
+		dev_err(dev, "failed to get the irq\n");
+		goto irq_failed;
+	}
+
+	ret = devm_request_irq(dev, ret,
+			fsl_qspi_irq_handler, 0, pdev->name, q);
+	if (ret) {
+		dev_err(dev, "failed to request irq.\n");
+		goto irq_failed;
+	}
+
+	q->dev = dev;
+	q->devtype_data = (struct fsl_qspi_devtype_data *)of_id->data;
+	platform_set_drvdata(pdev, q);
+
+	ret = fsl_qspi_nor_setup(q);
+	if (ret)
+		goto irq_failed;
+
+	if (of_get_property(np, "fsl,qspi-has-second-chip", NULL))
+		has_second_chip = true;
+
+	/* iterate the subnodes. */
+	for_each_available_child_of_node(dev->of_node, np) {
+		const struct spi_device_id *id;
+		char modalias[40];
+
+		/* skip the holes */
+		if (!has_second_chip)
+			i *= 2;
+
+		nor = &q->nor[i];
+		mtd = &q->mtd[i];
+
+		nor->mtd = mtd;
+		nor->dev = dev;
+		nor->priv = q;
+		mtd->priv = nor;
+
+		/* fill the hooks */
+		nor->read_reg = fsl_qspi_read_reg;
+		nor->write_reg = fsl_qspi_write_reg;
+		nor->read = fsl_qspi_read;
+		nor->write = fsl_qspi_write;
+		nor->erase = fsl_qspi_erase;
+
+		nor->prepare = fsl_qspi_prep;
+		nor->unprepare = fsl_qspi_unprep;
+
+		if (of_modalias_node(np, modalias, sizeof(modalias)) < 0)
+			goto map_failed;
+
+		id = spi_nor_match_id(modalias);
+		if (!id)
+			goto map_failed;
+
+		ret = of_property_read_u32(np, "spi-max-frequency",
+				&q->clk_rate);
+		if (ret < 0)
+			goto map_failed;
+
+		/* set the chip address for READID */
+		fsl_qspi_set_base_addr(q, nor);
+
+		ret = spi_nor_scan(nor, id, SPI_NOR_QUAD);
+		if (ret)
+			goto map_failed;
+
+		ppdata.of_node = np;
+		ret = mtd_device_parse_register(mtd, NULL, &ppdata, NULL, 0);
+		if (ret)
+			goto map_failed;
+
+		/* Set the correct NOR size now. */
+		if (q->nor_size == 0) {
+			q->nor_size = mtd->size;
+
+			/* Map the SPI NOR to accessiable address */
+			fsl_qspi_set_map_addr(q);
+		}
+
+		/*
+		 * The TX FIFO is 64 bytes in the Vybrid, but the Page Program
+		 * may writes 265 bytes per time. The write is working in the
+		 * unit of the TX FIFO, not in the unit of the SPI NOR's page
+		 * size.
+		 *
+		 * So shrink the spi_nor->page_size if it is larger then the
+		 * TX FIFO.
+		 */
+		if (nor->page_size > q->devtype_data->txfifo)
+			nor->page_size = q->devtype_data->txfifo;
+
+		i++;
+	}
+
+	/* finish the rest init. */
+	ret = fsl_qspi_nor_setup_last(q);
+	if (ret)
+		goto last_init_failed;
+
+	clk_disable(q->clk);
+	clk_disable(q->clk_en);
+	dev_info(dev, "QuadSPI SPI NOR flash driver\n");
+	return 0;
+
+last_init_failed:
+	for (i = 0; i < q->nor_num; i++)
+		mtd_device_unregister(&q->mtd[i]);
+
+irq_failed:
+	clk_disable_unprepare(q->clk);
+	clk_disable_unprepare(q->clk_en);
+map_failed:
+	dev_err(dev, "Freescale QuadSPI probe failed\n");
+	return ret;
+}
+
+static int fsl_qspi_remove(struct platform_device *pdev)
+{
+	struct fsl_qspi *q = platform_get_drvdata(pdev);
+	int i;
+
+	for (i = 0; i < q->nor_num; i++)
+		mtd_device_unregister(&q->mtd[i]);
+
+	/* disable the hardware */
+	writel(QUADSPI_MCR_MDIS_MASK, q->iobase + QUADSPI_MCR);
+	writel(0x0, q->iobase + QUADSPI_RSER);
+
+	clk_unprepare(q->clk);
+	clk_unprepare(q->clk_en);
+	return 0;
+}
+
+static struct platform_driver fsl_qspi_driver = {
+	.driver = {
+		.name	= "fsl-quadspi",
+		.bus	= &platform_bus_type,
+		.owner	= THIS_MODULE,
+		.of_match_table = fsl_qspi_dt_ids,
+	},
+	.probe          = fsl_qspi_probe,
+	.remove		= fsl_qspi_remove,
+};
+module_platform_driver(fsl_qspi_driver);
+
+MODULE_DESCRIPTION("Freescale QuadSPI Controller Driver");
+MODULE_AUTHOR("Freescale Semiconductor Inc.");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/mtd/spi-nor/spi-nor.c b/drivers/mtd/spi-nor/spi-nor.c
new file mode 100644
index 000000000000..c713c8656710
--- /dev/null
+++ b/drivers/mtd/spi-nor/spi-nor.c
@@ -0,0 +1,1107 @@
+/*
+ * Based on m25p80.c, by Mike Lavender (mike@steroidmicros.com), with
+ * influence from lart.c (Abraham Van Der Merwe) and mtd_dataflash.c
+ *
+ * Copyright (C) 2005, Intec Automation Inc.
+ * Copyright (C) 2014, Freescale Semiconductor, Inc.
+ *
+ * This code is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/mutex.h>
+#include <linux/math64.h>
+
+#include <linux/mtd/cfi.h>
+#include <linux/mtd/mtd.h>
+#include <linux/of_platform.h>
+#include <linux/spi/flash.h>
+#include <linux/mtd/spi-nor.h>
+
+/* Define max times to check status register before we give up. */
+#define	MAX_READY_WAIT_JIFFIES	(40 * HZ) /* M25P16 specs 40s max chip erase */
+
+#define JEDEC_MFR(_jedec_id)	((_jedec_id) >> 16)
+
+/*
+ * Read the status register, returning its value in the location
+ * Return the status register value.
+ * Returns negative if error occurred.
+ */
+static int read_sr(struct spi_nor *nor)
+{
+	int ret;
+	u8 val;
+
+	ret = nor->read_reg(nor, SPINOR_OP_RDSR, &val, 1);
+	if (ret < 0) {
+		pr_err("error %d reading SR\n", (int) ret);
+		return ret;
+	}
+
+	return val;
+}
+
+/*
+ * Read configuration register, returning its value in the
+ * location. Return the configuration register value.
+ * Returns negative if error occured.
+ */
+static int read_cr(struct spi_nor *nor)
+{
+	int ret;
+	u8 val;
+
+	ret = nor->read_reg(nor, SPINOR_OP_RDCR, &val, 1);
+	if (ret < 0) {
+		dev_err(nor->dev, "error %d reading CR\n", ret);
+		return ret;
+	}
+
+	return val;
+}
+
+/*
+ * Dummy Cycle calculation for different type of read.
+ * It can be used to support more commands with
+ * different dummy cycle requirements.
+ */
+static inline int spi_nor_read_dummy_cycles(struct spi_nor *nor)
+{
+	switch (nor->flash_read) {
+	case SPI_NOR_FAST:
+	case SPI_NOR_DUAL:
+	case SPI_NOR_QUAD:
+		return 1;
+	case SPI_NOR_NORMAL:
+		return 0;
+	}
+	return 0;
+}
+
+/*
+ * Write status register 1 byte
+ * Returns negative if error occurred.
+ */
+static inline int write_sr(struct spi_nor *nor, u8 val)
+{
+	nor->cmd_buf[0] = val;
+	return nor->write_reg(nor, SPINOR_OP_WRSR, nor->cmd_buf, 1, 0);
+}
+
+/*
+ * Set write enable latch with Write Enable command.
+ * Returns negative if error occurred.
+ */
+static inline int write_enable(struct spi_nor *nor)
+{
+	return nor->write_reg(nor, SPINOR_OP_WREN, NULL, 0, 0);
+}
+
+/*
+ * Send write disble instruction to the chip.
+ */
+static inline int write_disable(struct spi_nor *nor)
+{
+	return nor->write_reg(nor, SPINOR_OP_WRDI, NULL, 0, 0);
+}
+
+static inline struct spi_nor *mtd_to_spi_nor(struct mtd_info *mtd)
+{
+	return mtd->priv;
+}
+
+/* Enable/disable 4-byte addressing mode. */
+static inline int set_4byte(struct spi_nor *nor, u32 jedec_id, int enable)
+{
+	int status;
+	bool need_wren = false;
+	u8 cmd;
+
+	switch (JEDEC_MFR(jedec_id)) {
+	case CFI_MFR_ST: /* Micron, actually */
+		/* Some Micron need WREN command; all will accept it */
+		need_wren = true;
+	case CFI_MFR_MACRONIX:
+	case 0xEF /* winbond */:
+		if (need_wren)
+			write_enable(nor);
+
+		cmd = enable ? SPINOR_OP_EN4B : SPINOR_OP_EX4B;
+		status = nor->write_reg(nor, cmd, NULL, 0, 0);
+		if (need_wren)
+			write_disable(nor);
+
+		return status;
+	default:
+		/* Spansion style */
+		nor->cmd_buf[0] = enable << 7;
+		return nor->write_reg(nor, SPINOR_OP_BRWR, nor->cmd_buf, 1, 0);
+	}
+}
+
+static int spi_nor_wait_till_ready(struct spi_nor *nor)
+{
+	unsigned long deadline;
+	int sr;
+
+	deadline = jiffies + MAX_READY_WAIT_JIFFIES;
+
+	do {
+		cond_resched();
+
+		sr = read_sr(nor);
+		if (sr < 0)
+			break;
+		else if (!(sr & SR_WIP))
+			return 0;
+	} while (!time_after_eq(jiffies, deadline));
+
+	return -ETIMEDOUT;
+}
+
+/*
+ * Service routine to read status register until ready, or timeout occurs.
+ * Returns non-zero if error.
+ */
+static int wait_till_ready(struct spi_nor *nor)
+{
+	return nor->wait_till_ready(nor);
+}
+
+/*
+ * Erase the whole flash memory
+ *
+ * Returns 0 if successful, non-zero otherwise.
+ */
+static int erase_chip(struct spi_nor *nor)
+{
+	int ret;
+
+	dev_dbg(nor->dev, " %lldKiB\n", (long long)(nor->mtd->size >> 10));
+
+	/* Wait until finished previous write command. */
+	ret = wait_till_ready(nor);
+	if (ret)
+		return ret;
+
+	/* Send write enable, then erase commands. */
+	write_enable(nor);
+
+	return nor->write_reg(nor, SPINOR_OP_CHIP_ERASE, NULL, 0, 0);
+}
+
+static int spi_nor_lock_and_prep(struct spi_nor *nor, enum spi_nor_ops ops)
+{
+	int ret = 0;
+
+	mutex_lock(&nor->lock);
+
+	if (nor->prepare) {
+		ret = nor->prepare(nor, ops);
+		if (ret) {
+			dev_err(nor->dev, "failed in the preparation.\n");
+			mutex_unlock(&nor->lock);
+			return ret;
+		}
+	}
+	return ret;
+}
+
+static void spi_nor_unlock_and_unprep(struct spi_nor *nor, enum spi_nor_ops ops)
+{
+	if (nor->unprepare)
+		nor->unprepare(nor, ops);
+	mutex_unlock(&nor->lock);
+}
+
+/*
+ * Erase an address range on the nor chip.  The address range may extend
+ * one or more erase sectors.  Return an error is there is a problem erasing.
+ */
+static int spi_nor_erase(struct mtd_info *mtd, struct erase_info *instr)
+{
+	struct spi_nor *nor = mtd_to_spi_nor(mtd);
+	u32 addr, len;
+	uint32_t rem;
+	int ret;
+
+	dev_dbg(nor->dev, "at 0x%llx, len %lld\n", (long long)instr->addr,
+			(long long)instr->len);
+
+	div_u64_rem(instr->len, mtd->erasesize, &rem);
+	if (rem)
+		return -EINVAL;
+
+	addr = instr->addr;
+	len = instr->len;
+
+	ret = spi_nor_lock_and_prep(nor, SPI_NOR_OPS_ERASE);
+	if (ret)
+		return ret;
+
+	/* whole-chip erase? */
+	if (len == mtd->size) {
+		if (erase_chip(nor)) {
+			ret = -EIO;
+			goto erase_err;
+		}
+
+	/* REVISIT in some cases we could speed up erasing large regions
+	 * by using SPINOR_OP_SE instead of SPINOR_OP_BE_4K.  We may have set up
+	 * to use "small sector erase", but that's not always optimal.
+	 */
+
+	/* "sector"-at-a-time erase */
+	} else {
+		while (len) {
+			if (nor->erase(nor, addr)) {
+				ret = -EIO;
+				goto erase_err;
+			}
+
+			addr += mtd->erasesize;
+			len -= mtd->erasesize;
+		}
+	}
+
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_ERASE);
+
+	instr->state = MTD_ERASE_DONE;
+	mtd_erase_callback(instr);
+
+	return ret;
+
+erase_err:
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_ERASE);
+	instr->state = MTD_ERASE_FAILED;
+	return ret;
+}
+
+static int spi_nor_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
+{
+	struct spi_nor *nor = mtd_to_spi_nor(mtd);
+	uint32_t offset = ofs;
+	uint8_t status_old, status_new;
+	int ret = 0;
+
+	ret = spi_nor_lock_and_prep(nor, SPI_NOR_OPS_LOCK);
+	if (ret)
+		return ret;
+
+	/* Wait until finished previous command */
+	ret = wait_till_ready(nor);
+	if (ret)
+		goto err;
+
+	status_old = read_sr(nor);
+
+	if (offset < mtd->size - (mtd->size / 2))
+		status_new = status_old | SR_BP2 | SR_BP1 | SR_BP0;
+	else if (offset < mtd->size - (mtd->size / 4))
+		status_new = (status_old & ~SR_BP0) | SR_BP2 | SR_BP1;
+	else if (offset < mtd->size - (mtd->size / 8))
+		status_new = (status_old & ~SR_BP1) | SR_BP2 | SR_BP0;
+	else if (offset < mtd->size - (mtd->size / 16))
+		status_new = (status_old & ~(SR_BP0 | SR_BP1)) | SR_BP2;
+	else if (offset < mtd->size - (mtd->size / 32))
+		status_new = (status_old & ~SR_BP2) | SR_BP1 | SR_BP0;
+	else if (offset < mtd->size - (mtd->size / 64))
+		status_new = (status_old & ~(SR_BP2 | SR_BP0)) | SR_BP1;
+	else
+		status_new = (status_old & ~(SR_BP2 | SR_BP1)) | SR_BP0;
+
+	/* Only modify protection if it will not unlock other areas */
+	if ((status_new & (SR_BP2 | SR_BP1 | SR_BP0)) >
+				(status_old & (SR_BP2 | SR_BP1 | SR_BP0))) {
+		write_enable(nor);
+		ret = write_sr(nor, status_new);
+		if (ret)
+			goto err;
+	}
+
+err:
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_LOCK);
+	return ret;
+}
+
+static int spi_nor_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
+{
+	struct spi_nor *nor = mtd_to_spi_nor(mtd);
+	uint32_t offset = ofs;
+	uint8_t status_old, status_new;
+	int ret = 0;
+
+	ret = spi_nor_lock_and_prep(nor, SPI_NOR_OPS_UNLOCK);
+	if (ret)
+		return ret;
+
+	/* Wait until finished previous command */
+	ret = wait_till_ready(nor);
+	if (ret)
+		goto err;
+
+	status_old = read_sr(nor);
+
+	if (offset+len > mtd->size - (mtd->size / 64))
+		status_new = status_old & ~(SR_BP2 | SR_BP1 | SR_BP0);
+	else if (offset+len > mtd->size - (mtd->size / 32))
+		status_new = (status_old & ~(SR_BP2 | SR_BP1)) | SR_BP0;
+	else if (offset+len > mtd->size - (mtd->size / 16))
+		status_new = (status_old & ~(SR_BP2 | SR_BP0)) | SR_BP1;
+	else if (offset+len > mtd->size - (mtd->size / 8))
+		status_new = (status_old & ~SR_BP2) | SR_BP1 | SR_BP0;
+	else if (offset+len > mtd->size - (mtd->size / 4))
+		status_new = (status_old & ~(SR_BP0 | SR_BP1)) | SR_BP2;
+	else if (offset+len > mtd->size - (mtd->size / 2))
+		status_new = (status_old & ~SR_BP1) | SR_BP2 | SR_BP0;
+	else
+		status_new = (status_old & ~SR_BP0) | SR_BP2 | SR_BP1;
+
+	/* Only modify protection if it will not lock other areas */
+	if ((status_new & (SR_BP2 | SR_BP1 | SR_BP0)) <
+				(status_old & (SR_BP2 | SR_BP1 | SR_BP0))) {
+		write_enable(nor);
+		ret = write_sr(nor, status_new);
+		if (ret)
+			goto err;
+	}
+
+err:
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_UNLOCK);
+	return ret;
+}
+
+struct flash_info {
+	/* JEDEC id zero means "no ID" (most older chips); otherwise it has
+	 * a high byte of zero plus three data bytes: the manufacturer id,
+	 * then a two byte device id.
+	 */
+	u32		jedec_id;
+	u16             ext_id;
+
+	/* The size listed here is what works with SPINOR_OP_SE, which isn't
+	 * necessarily called a "sector" by the vendor.
+	 */
+	unsigned	sector_size;
+	u16		n_sectors;
+
+	u16		page_size;
+	u16		addr_width;
+
+	u16		flags;
+#define	SECT_4K			0x01	/* SPINOR_OP_BE_4K works uniformly */
+#define	SPI_NOR_NO_ERASE	0x02	/* No erase command needed */
+#define	SST_WRITE		0x04	/* use SST byte programming */
+#define	SPI_NOR_NO_FR		0x08	/* Can't do fastread */
+#define	SECT_4K_PMC		0x10	/* SPINOR_OP_BE_4K_PMC works uniformly */
+#define	SPI_NOR_DUAL_READ	0x20    /* Flash supports Dual Read */
+#define	SPI_NOR_QUAD_READ	0x40    /* Flash supports Quad Read */
+};
+
+#define INFO(_jedec_id, _ext_id, _sector_size, _n_sectors, _flags)	\
+	((kernel_ulong_t)&(struct flash_info) {				\
+		.jedec_id = (_jedec_id),				\
+		.ext_id = (_ext_id),					\
+		.sector_size = (_sector_size),				\
+		.n_sectors = (_n_sectors),				\
+		.page_size = 256,					\
+		.flags = (_flags),					\
+	})
+
+#define CAT25_INFO(_sector_size, _n_sectors, _page_size, _addr_width, _flags)	\
+	((kernel_ulong_t)&(struct flash_info) {				\
+		.sector_size = (_sector_size),				\
+		.n_sectors = (_n_sectors),				\
+		.page_size = (_page_size),				\
+		.addr_width = (_addr_width),				\
+		.flags = (_flags),					\
+	})
+
+/* NOTE: double check command sets and memory organization when you add
+ * more nor chips.  This current list focusses on newer chips, which
+ * have been converging on command sets which including JEDEC ID.
+ */
+const struct spi_device_id spi_nor_ids[] = {
+	/* Atmel -- some are (confusingly) marketed as "DataFlash" */
+	{ "at25fs010",  INFO(0x1f6601, 0, 32 * 1024,   4, SECT_4K) },
+	{ "at25fs040",  INFO(0x1f6604, 0, 64 * 1024,   8, SECT_4K) },
+
+	{ "at25df041a", INFO(0x1f4401, 0, 64 * 1024,   8, SECT_4K) },
+	{ "at25df321a", INFO(0x1f4701, 0, 64 * 1024,  64, SECT_4K) },
+	{ "at25df641",  INFO(0x1f4800, 0, 64 * 1024, 128, SECT_4K) },
+
+	{ "at26f004",   INFO(0x1f0400, 0, 64 * 1024,  8, SECT_4K) },
+	{ "at26df081a", INFO(0x1f4501, 0, 64 * 1024, 16, SECT_4K) },
+	{ "at26df161a", INFO(0x1f4601, 0, 64 * 1024, 32, SECT_4K) },
+	{ "at26df321",  INFO(0x1f4700, 0, 64 * 1024, 64, SECT_4K) },
+
+	{ "at45db081d", INFO(0x1f2500, 0, 64 * 1024, 16, SECT_4K) },
+
+	/* EON -- en25xxx */
+	{ "en25f32",    INFO(0x1c3116, 0, 64 * 1024,   64, SECT_4K) },
+	{ "en25p32",    INFO(0x1c2016, 0, 64 * 1024,   64, 0) },
+	{ "en25q32b",   INFO(0x1c3016, 0, 64 * 1024,   64, 0) },
+	{ "en25p64",    INFO(0x1c2017, 0, 64 * 1024,  128, 0) },
+	{ "en25q64",    INFO(0x1c3017, 0, 64 * 1024,  128, SECT_4K) },
+	{ "en25qh256",  INFO(0x1c7019, 0, 64 * 1024,  512, 0) },
+
+	/* ESMT */
+	{ "f25l32pa", INFO(0x8c2016, 0, 64 * 1024, 64, SECT_4K) },
+
+	/* Everspin */
+	{ "mr25h256", CAT25_INFO( 32 * 1024, 1, 256, 2, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+	{ "mr25h10",  CAT25_INFO(128 * 1024, 1, 256, 3, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+
+	/* GigaDevice */
+	{ "gd25q32", INFO(0xc84016, 0, 64 * 1024,  64, SECT_4K) },
+	{ "gd25q64", INFO(0xc84017, 0, 64 * 1024, 128, SECT_4K) },
+
+	/* Intel/Numonyx -- xxxs33b */
+	{ "160s33b",  INFO(0x898911, 0, 64 * 1024,  32, 0) },
+	{ "320s33b",  INFO(0x898912, 0, 64 * 1024,  64, 0) },
+	{ "640s33b",  INFO(0x898913, 0, 64 * 1024, 128, 0) },
+
+	/* Macronix */
+	{ "mx25l2005a",  INFO(0xc22012, 0, 64 * 1024,   4, SECT_4K) },
+	{ "mx25l4005a",  INFO(0xc22013, 0, 64 * 1024,   8, SECT_4K) },
+	{ "mx25l8005",   INFO(0xc22014, 0, 64 * 1024,  16, 0) },
+	{ "mx25l1606e",  INFO(0xc22015, 0, 64 * 1024,  32, SECT_4K) },
+	{ "mx25l3205d",  INFO(0xc22016, 0, 64 * 1024,  64, 0) },
+	{ "mx25l3255e",  INFO(0xc29e16, 0, 64 * 1024,  64, SECT_4K) },
+	{ "mx25l6405d",  INFO(0xc22017, 0, 64 * 1024, 128, 0) },
+	{ "mx25l12805d", INFO(0xc22018, 0, 64 * 1024, 256, 0) },
+	{ "mx25l12855e", INFO(0xc22618, 0, 64 * 1024, 256, 0) },
+	{ "mx25l25635e", INFO(0xc22019, 0, 64 * 1024, 512, 0) },
+	{ "mx25l25655e", INFO(0xc22619, 0, 64 * 1024, 512, 0) },
+	{ "mx66l51235l", INFO(0xc2201a, 0, 64 * 1024, 1024, SPI_NOR_QUAD_READ) },
+	{ "mx66l1g55g",  INFO(0xc2261b, 0, 64 * 1024, 2048, SPI_NOR_QUAD_READ) },
+
+	/* Micron */
+	{ "n25q064",     INFO(0x20ba17, 0, 64 * 1024,  128, 0) },
+	{ "n25q128a11",  INFO(0x20bb18, 0, 64 * 1024,  256, 0) },
+	{ "n25q128a13",  INFO(0x20ba18, 0, 64 * 1024,  256, 0) },
+	{ "n25q256a",    INFO(0x20ba19, 0, 64 * 1024,  512, SECT_4K) },
+	{ "n25q512a",    INFO(0x20bb20, 0, 64 * 1024, 1024, SECT_4K) },
+
+	/* PMC */
+	{ "pm25lv512",   INFO(0,        0, 32 * 1024,    2, SECT_4K_PMC) },
+	{ "pm25lv010",   INFO(0,        0, 32 * 1024,    4, SECT_4K_PMC) },
+	{ "pm25lq032",   INFO(0x7f9d46, 0, 64 * 1024,   64, SECT_4K) },
+
+	/* Spansion -- single (large) sector size only, at least
+	 * for the chips listed here (without boot sectors).
+	 */
+	{ "s25sl032p",  INFO(0x010215, 0x4d00,  64 * 1024,  64, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ) },
+	{ "s25sl064p",  INFO(0x010216, 0x4d00,  64 * 1024, 128, 0) },
+	{ "s25fl256s0", INFO(0x010219, 0x4d00, 256 * 1024, 128, 0) },
+	{ "s25fl256s1", INFO(0x010219, 0x4d01,  64 * 1024, 512, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ) },
+	{ "s25fl512s",  INFO(0x010220, 0x4d00, 256 * 1024, 256, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ) },
+	{ "s70fl01gs",  INFO(0x010221, 0x4d00, 256 * 1024, 256, 0) },
+	{ "s25sl12800", INFO(0x012018, 0x0300, 256 * 1024,  64, 0) },
+	{ "s25sl12801", INFO(0x012018, 0x0301,  64 * 1024, 256, 0) },
+	{ "s25fl129p0", INFO(0x012018, 0x4d00, 256 * 1024,  64, 0) },
+	{ "s25fl129p1", INFO(0x012018, 0x4d01,  64 * 1024, 256, 0) },
+	{ "s25sl004a",  INFO(0x010212,      0,  64 * 1024,   8, 0) },
+	{ "s25sl008a",  INFO(0x010213,      0,  64 * 1024,  16, 0) },
+	{ "s25sl016a",  INFO(0x010214,      0,  64 * 1024,  32, 0) },
+	{ "s25sl032a",  INFO(0x010215,      0,  64 * 1024,  64, 0) },
+	{ "s25sl064a",  INFO(0x010216,      0,  64 * 1024, 128, 0) },
+	{ "s25fl008k",  INFO(0xef4014,      0,  64 * 1024,  16, SECT_4K) },
+	{ "s25fl016k",  INFO(0xef4015,      0,  64 * 1024,  32, SECT_4K) },
+	{ "s25fl064k",  INFO(0xef4017,      0,  64 * 1024, 128, SECT_4K) },
+
+	/* SST -- large erase sizes are "overlays", "sectors" are 4K */
+	{ "sst25vf040b", INFO(0xbf258d, 0, 64 * 1024,  8, SECT_4K | SST_WRITE) },
+	{ "sst25vf080b", INFO(0xbf258e, 0, 64 * 1024, 16, SECT_4K | SST_WRITE) },
+	{ "sst25vf016b", INFO(0xbf2541, 0, 64 * 1024, 32, SECT_4K | SST_WRITE) },
+	{ "sst25vf032b", INFO(0xbf254a, 0, 64 * 1024, 64, SECT_4K | SST_WRITE) },
+	{ "sst25vf064c", INFO(0xbf254b, 0, 64 * 1024, 128, SECT_4K) },
+	{ "sst25wf512",  INFO(0xbf2501, 0, 64 * 1024,  1, SECT_4K | SST_WRITE) },
+	{ "sst25wf010",  INFO(0xbf2502, 0, 64 * 1024,  2, SECT_4K | SST_WRITE) },
+	{ "sst25wf020",  INFO(0xbf2503, 0, 64 * 1024,  4, SECT_4K | SST_WRITE) },
+	{ "sst25wf040",  INFO(0xbf2504, 0, 64 * 1024,  8, SECT_4K | SST_WRITE) },
+
+	/* ST Microelectronics -- newer production may have feature updates */
+	{ "m25p05",  INFO(0x202010,  0,  32 * 1024,   2, 0) },
+	{ "m25p10",  INFO(0x202011,  0,  32 * 1024,   4, 0) },
+	{ "m25p20",  INFO(0x202012,  0,  64 * 1024,   4, 0) },
+	{ "m25p40",  INFO(0x202013,  0,  64 * 1024,   8, 0) },
+	{ "m25p80",  INFO(0x202014,  0,  64 * 1024,  16, 0) },
+	{ "m25p16",  INFO(0x202015,  0,  64 * 1024,  32, 0) },
+	{ "m25p32",  INFO(0x202016,  0,  64 * 1024,  64, 0) },
+	{ "m25p64",  INFO(0x202017,  0,  64 * 1024, 128, 0) },
+	{ "m25p128", INFO(0x202018,  0, 256 * 1024,  64, 0) },
+	{ "n25q032", INFO(0x20ba16,  0,  64 * 1024,  64, 0) },
+
+	{ "m25p05-nonjedec",  INFO(0, 0,  32 * 1024,   2, 0) },
+	{ "m25p10-nonjedec",  INFO(0, 0,  32 * 1024,   4, 0) },
+	{ "m25p20-nonjedec",  INFO(0, 0,  64 * 1024,   4, 0) },
+	{ "m25p40-nonjedec",  INFO(0, 0,  64 * 1024,   8, 0) },
+	{ "m25p80-nonjedec",  INFO(0, 0,  64 * 1024,  16, 0) },
+	{ "m25p16-nonjedec",  INFO(0, 0,  64 * 1024,  32, 0) },
+	{ "m25p32-nonjedec",  INFO(0, 0,  64 * 1024,  64, 0) },
+	{ "m25p64-nonjedec",  INFO(0, 0,  64 * 1024, 128, 0) },
+	{ "m25p128-nonjedec", INFO(0, 0, 256 * 1024,  64, 0) },
+
+	{ "m45pe10", INFO(0x204011,  0, 64 * 1024,    2, 0) },
+	{ "m45pe80", INFO(0x204014,  0, 64 * 1024,   16, 0) },
+	{ "m45pe16", INFO(0x204015,  0, 64 * 1024,   32, 0) },
+
+	{ "m25pe20", INFO(0x208012,  0, 64 * 1024,  4,       0) },
+	{ "m25pe80", INFO(0x208014,  0, 64 * 1024, 16,       0) },
+	{ "m25pe16", INFO(0x208015,  0, 64 * 1024, 32, SECT_4K) },
+
+	{ "m25px16",    INFO(0x207115,  0, 64 * 1024, 32, SECT_4K) },
+	{ "m25px32",    INFO(0x207116,  0, 64 * 1024, 64, SECT_4K) },
+	{ "m25px32-s0", INFO(0x207316,  0, 64 * 1024, 64, SECT_4K) },
+	{ "m25px32-s1", INFO(0x206316,  0, 64 * 1024, 64, SECT_4K) },
+	{ "m25px64",    INFO(0x207117,  0, 64 * 1024, 128, 0) },
+
+	/* Winbond -- w25x "blocks" are 64K, "sectors" are 4KiB */
+	{ "w25x10", INFO(0xef3011, 0, 64 * 1024,  2,  SECT_4K) },
+	{ "w25x20", INFO(0xef3012, 0, 64 * 1024,  4,  SECT_4K) },
+	{ "w25x40", INFO(0xef3013, 0, 64 * 1024,  8,  SECT_4K) },
+	{ "w25x80", INFO(0xef3014, 0, 64 * 1024,  16, SECT_4K) },
+	{ "w25x16", INFO(0xef3015, 0, 64 * 1024,  32, SECT_4K) },
+	{ "w25x32", INFO(0xef3016, 0, 64 * 1024,  64, SECT_4K) },
+	{ "w25q32", INFO(0xef4016, 0, 64 * 1024,  64, SECT_4K) },
+	{ "w25q32dw", INFO(0xef6016, 0, 64 * 1024,  64, SECT_4K) },
+	{ "w25x64", INFO(0xef3017, 0, 64 * 1024, 128, SECT_4K) },
+	{ "w25q64", INFO(0xef4017, 0, 64 * 1024, 128, SECT_4K) },
+	{ "w25q128", INFO(0xef4018, 0, 64 * 1024, 256, SECT_4K) },
+	{ "w25q80", INFO(0xef5014, 0, 64 * 1024,  16, SECT_4K) },
+	{ "w25q80bl", INFO(0xef4014, 0, 64 * 1024,  16, SECT_4K) },
+	{ "w25q128", INFO(0xef4018, 0, 64 * 1024, 256, SECT_4K) },
+	{ "w25q256", INFO(0xef4019, 0, 64 * 1024, 512, SECT_4K) },
+
+	/* Catalyst / On Semiconductor -- non-JEDEC */
+	{ "cat25c11", CAT25_INFO(  16, 8, 16, 1, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+	{ "cat25c03", CAT25_INFO(  32, 8, 16, 2, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+	{ "cat25c09", CAT25_INFO( 128, 8, 32, 2, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+	{ "cat25c17", CAT25_INFO( 256, 8, 32, 2, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+	{ "cat25128", CAT25_INFO(2048, 8, 64, 2, SPI_NOR_NO_ERASE | SPI_NOR_NO_FR) },
+	{ },
+};
+EXPORT_SYMBOL_GPL(spi_nor_ids);
+
+static const struct spi_device_id *spi_nor_read_id(struct spi_nor *nor)
+{
+	int			tmp;
+	u8			id[5];
+	u32			jedec;
+	u16                     ext_jedec;
+	struct flash_info	*info;
+
+	tmp = nor->read_reg(nor, SPINOR_OP_RDID, id, 5);
+	if (tmp < 0) {
+		dev_dbg(nor->dev, " error %d reading JEDEC ID\n", tmp);
+		return ERR_PTR(tmp);
+	}
+	jedec = id[0];
+	jedec = jedec << 8;
+	jedec |= id[1];
+	jedec = jedec << 8;
+	jedec |= id[2];
+
+	ext_jedec = id[3] << 8 | id[4];
+
+	for (tmp = 0; tmp < ARRAY_SIZE(spi_nor_ids) - 1; tmp++) {
+		info = (void *)spi_nor_ids[tmp].driver_data;
+		if (info->jedec_id == jedec) {
+			if (info->ext_id == 0 || info->ext_id == ext_jedec)
+				return &spi_nor_ids[tmp];
+		}
+	}
+	dev_err(nor->dev, "unrecognized JEDEC id %06x\n", jedec);
+	return ERR_PTR(-ENODEV);
+}
+
+static const struct spi_device_id *jedec_probe(struct spi_nor *nor)
+{
+	return nor->read_id(nor);
+}
+
+static int spi_nor_read(struct mtd_info *mtd, loff_t from, size_t len,
+			size_t *retlen, u_char *buf)
+{
+	struct spi_nor *nor = mtd_to_spi_nor(mtd);
+	int ret;
+
+	dev_dbg(nor->dev, "from 0x%08x, len %zd\n", (u32)from, len);
+
+	ret = spi_nor_lock_and_prep(nor, SPI_NOR_OPS_READ);
+	if (ret)
+		return ret;
+
+	ret = nor->read(nor, from, len, retlen, buf);
+
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_READ);
+	return ret;
+}
+
+static int sst_write(struct mtd_info *mtd, loff_t to, size_t len,
+		size_t *retlen, const u_char *buf)
+{
+	struct spi_nor *nor = mtd_to_spi_nor(mtd);
+	size_t actual;
+	int ret;
+
+	dev_dbg(nor->dev, "to 0x%08x, len %zd\n", (u32)to, len);
+
+	ret = spi_nor_lock_and_prep(nor, SPI_NOR_OPS_WRITE);
+	if (ret)
+		return ret;
+
+	/* Wait until finished previous write command. */
+	ret = wait_till_ready(nor);
+	if (ret)
+		goto time_out;
+
+	write_enable(nor);
+
+	nor->sst_write_second = false;
+
+	actual = to % 2;
+	/* Start write from odd address. */
+	if (actual) {
+		nor->program_opcode = SPINOR_OP_BP;
+
+		/* write one byte. */
+		nor->write(nor, to, 1, retlen, buf);
+		ret = wait_till_ready(nor);
+		if (ret)
+			goto time_out;
+	}
+	to += actual;
+
+	/* Write out most of the data here. */
+	for (; actual < len - 1; actual += 2) {
+		nor->program_opcode = SPINOR_OP_AAI_WP;
+
+		/* write two bytes. */
+		nor->write(nor, to, 2, retlen, buf + actual);
+		ret = wait_till_ready(nor);
+		if (ret)
+			goto time_out;
+		to += 2;
+		nor->sst_write_second = true;
+	}
+	nor->sst_write_second = false;
+
+	write_disable(nor);
+	ret = wait_till_ready(nor);
+	if (ret)
+		goto time_out;
+
+	/* Write out trailing byte if it exists. */
+	if (actual != len) {
+		write_enable(nor);
+
+		nor->program_opcode = SPINOR_OP_BP;
+		nor->write(nor, to, 1, retlen, buf + actual);
+
+		ret = wait_till_ready(nor);
+		if (ret)
+			goto time_out;
+		write_disable(nor);
+	}
+time_out:
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_WRITE);
+	return ret;
+}
+
+/*
+ * Write an address range to the nor chip.  Data must be written in
+ * FLASH_PAGESIZE chunks.  The address range may be any size provided
+ * it is within the physical boundaries.
+ */
+static int spi_nor_write(struct mtd_info *mtd, loff_t to, size_t len,
+	size_t *retlen, const u_char *buf)
+{
+	struct spi_nor *nor = mtd_to_spi_nor(mtd);
+	u32 page_offset, page_size, i;
+	int ret;
+
+	dev_dbg(nor->dev, "to 0x%08x, len %zd\n", (u32)to, len);
+
+	ret = spi_nor_lock_and_prep(nor, SPI_NOR_OPS_WRITE);
+	if (ret)
+		return ret;
+
+	/* Wait until finished previous write command. */
+	ret = wait_till_ready(nor);
+	if (ret)
+		goto write_err;
+
+	write_enable(nor);
+
+	page_offset = to & (nor->page_size - 1);
+
+	/* do all the bytes fit onto one page? */
+	if (page_offset + len <= nor->page_size) {
+		nor->write(nor, to, len, retlen, buf);
+	} else {
+		/* the size of data remaining on the first page */
+		page_size = nor->page_size - page_offset;
+		nor->write(nor, to, page_size, retlen, buf);
+
+		/* write everything in nor->page_size chunks */
+		for (i = page_size; i < len; i += page_size) {
+			page_size = len - i;
+			if (page_size > nor->page_size)
+				page_size = nor->page_size;
+
+			wait_till_ready(nor);
+			write_enable(nor);
+
+			nor->write(nor, to + i, page_size, retlen, buf + i);
+		}
+	}
+
+write_err:
+	spi_nor_unlock_and_unprep(nor, SPI_NOR_OPS_WRITE);
+	return 0;
+}
+
+static int macronix_quad_enable(struct spi_nor *nor)
+{
+	int ret, val;
+
+	val = read_sr(nor);
+	write_enable(nor);
+
+	nor->cmd_buf[0] = val | SR_QUAD_EN_MX;
+	nor->write_reg(nor, SPINOR_OP_WRSR, nor->cmd_buf, 1, 0);
+
+	if (wait_till_ready(nor))
+		return 1;
+
+	ret = read_sr(nor);
+	if (!(ret > 0 && (ret & SR_QUAD_EN_MX))) {
+		dev_err(nor->dev, "Macronix Quad bit not set\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/*
+ * Write status Register and configuration register with 2 bytes
+ * The first byte will be written to the status register, while the
+ * second byte will be written to the configuration register.
+ * Return negative if error occured.
+ */
+static int write_sr_cr(struct spi_nor *nor, u16 val)
+{
+	nor->cmd_buf[0] = val & 0xff;
+	nor->cmd_buf[1] = (val >> 8);
+
+	return nor->write_reg(nor, SPINOR_OP_WRSR, nor->cmd_buf, 2, 0);
+}
+
+static int spansion_quad_enable(struct spi_nor *nor)
+{
+	int ret;
+	int quad_en = CR_QUAD_EN_SPAN << 8;
+
+	write_enable(nor);
+
+	ret = write_sr_cr(nor, quad_en);
+	if (ret < 0) {
+		dev_err(nor->dev,
+			"error while writing configuration register\n");
+		return -EINVAL;
+	}
+
+	/* read back and check it */
+	ret = read_cr(nor);
+	if (!(ret > 0 && (ret & CR_QUAD_EN_SPAN))) {
+		dev_err(nor->dev, "Spansion Quad bit not set\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int set_quad_mode(struct spi_nor *nor, u32 jedec_id)
+{
+	int status;
+
+	switch (JEDEC_MFR(jedec_id)) {
+	case CFI_MFR_MACRONIX:
+		status = macronix_quad_enable(nor);
+		if (status) {
+			dev_err(nor->dev, "Macronix quad-read not enabled\n");
+			return -EINVAL;
+		}
+		return status;
+	default:
+		status = spansion_quad_enable(nor);
+		if (status) {
+			dev_err(nor->dev, "Spansion quad-read not enabled\n");
+			return -EINVAL;
+		}
+		return status;
+	}
+}
+
+static int spi_nor_check(struct spi_nor *nor)
+{
+	if (!nor->dev || !nor->read || !nor->write ||
+		!nor->read_reg || !nor->write_reg || !nor->erase) {
+		pr_err("spi-nor: please fill all the necessary fields!\n");
+		return -EINVAL;
+	}
+
+	if (!nor->read_id)
+		nor->read_id = spi_nor_read_id;
+	if (!nor->wait_till_ready)
+		nor->wait_till_ready = spi_nor_wait_till_ready;
+
+	return 0;
+}
+
+int spi_nor_scan(struct spi_nor *nor, const struct spi_device_id *id,
+			enum read_mode mode)
+{
+	struct flash_info		*info;
+	struct flash_platform_data	*data;
+	struct device *dev = nor->dev;
+	struct mtd_info *mtd = nor->mtd;
+	struct device_node *np = dev->of_node;
+	int ret;
+	int i;
+
+	ret = spi_nor_check(nor);
+	if (ret)
+		return ret;
+
+	/* Platform data helps sort out which chip type we have, as
+	 * well as how this board partitions it.  If we don't have
+	 * a chip ID, try the JEDEC id commands; they'll work for most
+	 * newer chips, even if we don't recognize the particular chip.
+	 */
+	data = dev_get_platdata(dev);
+	if (data && data->type) {
+		const struct spi_device_id *plat_id;
+
+		for (i = 0; i < ARRAY_SIZE(spi_nor_ids) - 1; i++) {
+			plat_id = &spi_nor_ids[i];
+			if (strcmp(data->type, plat_id->name))
+				continue;
+			break;
+		}
+
+		if (i < ARRAY_SIZE(spi_nor_ids) - 1)
+			id = plat_id;
+		else
+			dev_warn(dev, "unrecognized id %s\n", data->type);
+	}
+
+	info = (void *)id->driver_data;
+
+	if (info->jedec_id) {
+		const struct spi_device_id *jid;
+
+		jid = jedec_probe(nor);
+		if (IS_ERR(jid)) {
+			return PTR_ERR(jid);
+		} else if (jid != id) {
+			/*
+			 * JEDEC knows better, so overwrite platform ID. We
+			 * can't trust partitions any longer, but we'll let
+			 * mtd apply them anyway, since some partitions may be
+			 * marked read-only, and we don't want to lose that
+			 * information, even if it's not 100% accurate.
+			 */
+			dev_warn(dev, "found %s, expected %s\n",
+				 jid->name, id->name);
+			id = jid;
+			info = (void *)jid->driver_data;
+		}
+	}
+
+	mutex_init(&nor->lock);
+
+	/*
+	 * Atmel, SST and Intel/Numonyx serial nor tend to power
+	 * up with the software protection bits set
+	 */
+
+	if (JEDEC_MFR(info->jedec_id) == CFI_MFR_ATMEL ||
+	    JEDEC_MFR(info->jedec_id) == CFI_MFR_INTEL ||
+	    JEDEC_MFR(info->jedec_id) == CFI_MFR_SST) {
+		write_enable(nor);
+		write_sr(nor, 0);
+	}
+
+	if (data && data->name)
+		mtd->name = data->name;
+	else
+		mtd->name = dev_name(dev);
+
+	mtd->type = MTD_NORFLASH;
+	mtd->writesize = 1;
+	mtd->flags = MTD_CAP_NORFLASH;
+	mtd->size = info->sector_size * info->n_sectors;
+	mtd->_erase = spi_nor_erase;
+	mtd->_read = spi_nor_read;
+
+	/* nor protection support for STmicro chips */
+	if (JEDEC_MFR(info->jedec_id) == CFI_MFR_ST) {
+		mtd->_lock = spi_nor_lock;
+		mtd->_unlock = spi_nor_unlock;
+	}
+
+	/* sst nor chips use AAI word program */
+	if (info->flags & SST_WRITE)
+		mtd->_write = sst_write;
+	else
+		mtd->_write = spi_nor_write;
+
+	/* prefer "small sector" erase if possible */
+	if (info->flags & SECT_4K) {
+		nor->erase_opcode = SPINOR_OP_BE_4K;
+		mtd->erasesize = 4096;
+	} else if (info->flags & SECT_4K_PMC) {
+		nor->erase_opcode = SPINOR_OP_BE_4K_PMC;
+		mtd->erasesize = 4096;
+	} else {
+		nor->erase_opcode = SPINOR_OP_SE;
+		mtd->erasesize = info->sector_size;
+	}
+
+	if (info->flags & SPI_NOR_NO_ERASE)
+		mtd->flags |= MTD_NO_ERASE;
+
+	mtd->dev.parent = dev;
+	nor->page_size = info->page_size;
+	mtd->writebufsize = nor->page_size;
+
+	if (np) {
+		/* If we were instantiated by DT, use it */
+		if (of_property_read_bool(np, "m25p,fast-read"))
+			nor->flash_read = SPI_NOR_FAST;
+		else
+			nor->flash_read = SPI_NOR_NORMAL;
+	} else {
+		/* If we weren't instantiated by DT, default to fast-read */
+		nor->flash_read = SPI_NOR_FAST;
+	}
+
+	/* Some devices cannot do fast-read, no matter what DT tells us */
+	if (info->flags & SPI_NOR_NO_FR)
+		nor->flash_read = SPI_NOR_NORMAL;
+
+	/* Quad/Dual-read mode takes precedence over fast/normal */
+	if (mode == SPI_NOR_QUAD && info->flags & SPI_NOR_QUAD_READ) {
+		ret = set_quad_mode(nor, info->jedec_id);
+		if (ret) {
+			dev_err(dev, "quad mode not supported\n");
+			return ret;
+		}
+		nor->flash_read = SPI_NOR_QUAD;
+	} else if (mode == SPI_NOR_DUAL && info->flags & SPI_NOR_DUAL_READ) {
+		nor->flash_read = SPI_NOR_DUAL;
+	}
+
+	/* Default commands */
+	switch (nor->flash_read) {
+	case SPI_NOR_QUAD:
+		nor->read_opcode = SPINOR_OP_READ_1_1_4;
+		break;
+	case SPI_NOR_DUAL:
+		nor->read_opcode = SPINOR_OP_READ_1_1_2;
+		break;
+	case SPI_NOR_FAST:
+		nor->read_opcode = SPINOR_OP_READ_FAST;
+		break;
+	case SPI_NOR_NORMAL:
+		nor->read_opcode = SPINOR_OP_READ;
+		break;
+	default:
+		dev_err(dev, "No Read opcode defined\n");
+		return -EINVAL;
+	}
+
+	nor->program_opcode = SPINOR_OP_PP;
+
+	if (info->addr_width)
+		nor->addr_width = info->addr_width;
+	else if (mtd->size > 0x1000000) {
+		/* enable 4-byte addressing if the device exceeds 16MiB */
+		nor->addr_width = 4;
+		if (JEDEC_MFR(info->jedec_id) == CFI_MFR_AMD) {
+			/* Dedicated 4-byte command set */
+			switch (nor->flash_read) {
+			case SPI_NOR_QUAD:
+				nor->read_opcode = SPINOR_OP_READ4_1_1_4;
+				break;
+			case SPI_NOR_DUAL:
+				nor->read_opcode = SPINOR_OP_READ4_1_1_2;
+				break;
+			case SPI_NOR_FAST:
+				nor->read_opcode = SPINOR_OP_READ4_FAST;
+				break;
+			case SPI_NOR_NORMAL:
+				nor->read_opcode = SPINOR_OP_READ4;
+				break;
+			}
+			nor->program_opcode = SPINOR_OP_PP_4B;
+			/* No small sector erase for 4-byte command set */
+			nor->erase_opcode = SPINOR_OP_SE_4B;
+			mtd->erasesize = info->sector_size;
+		} else
+			set_4byte(nor, info->jedec_id, 1);
+	} else {
+		nor->addr_width = 3;
+	}
+
+	nor->read_dummy = spi_nor_read_dummy_cycles(nor);
+
+	dev_info(dev, "%s (%lld Kbytes)\n", id->name,
+			(long long)mtd->size >> 10);
+
+	dev_dbg(dev,
+		"mtd .name = %s, .size = 0x%llx (%lldMiB), "
+		".erasesize = 0x%.8x (%uKiB) .numeraseregions = %d\n",
+		mtd->name, (long long)mtd->size, (long long)(mtd->size >> 20),
+		mtd->erasesize, mtd->erasesize / 1024, mtd->numeraseregions);
+
+	if (mtd->numeraseregions)
+		for (i = 0; i < mtd->numeraseregions; i++)
+			dev_dbg(dev,
+				"mtd.eraseregions[%d] = { .offset = 0x%llx, "
+				".erasesize = 0x%.8x (%uKiB), "
+				".numblocks = %d }\n",
+				i, (long long)mtd->eraseregions[i].offset,
+				mtd->eraseregions[i].erasesize,
+				mtd->eraseregions[i].erasesize / 1024,
+				mtd->eraseregions[i].numblocks);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(spi_nor_scan);
+
+const struct spi_device_id *spi_nor_match_id(char *name)
+{
+	const struct spi_device_id *id = spi_nor_ids;
+
+	while (id->name[0]) {
+		if (!strcmp(name, id->name))
+			return id;
+		id++;
+	}
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(spi_nor_match_id);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Huang Shijie <shijie8@gmail.com>");
+MODULE_AUTHOR("Mike Lavender");
+MODULE_DESCRIPTION("framework for SPI NOR");
diff --git a/drivers/mtd/tests/oobtest.c b/drivers/mtd/tests/oobtest.c
index 2e9e2d11f204..f19ab1acde1f 100644
--- a/drivers/mtd/tests/oobtest.c
+++ b/drivers/mtd/tests/oobtest.c
@@ -69,8 +69,8 @@ static int write_eraseblock(int ebnum)
 	int err = 0;
 	loff_t addr = ebnum * mtd->erasesize;
 
+	prandom_bytes_state(&rnd_state, writebuf, use_len_max * pgcnt);
 	for (i = 0; i < pgcnt; ++i, addr += mtd->writesize) {
-		prandom_bytes_state(&rnd_state, writebuf, use_len);
 		ops.mode      = MTD_OPS_AUTO_OOB;
 		ops.len       = 0;
 		ops.retlen    = 0;
@@ -78,7 +78,7 @@ static int write_eraseblock(int ebnum)
 		ops.oobretlen = 0;
 		ops.ooboffs   = use_offset;
 		ops.datbuf    = NULL;
-		ops.oobbuf    = writebuf;
+		ops.oobbuf    = writebuf + (use_len_max * i) + use_offset;
 		err = mtd_write_oob(mtd, addr, &ops);
 		if (err || ops.oobretlen != use_len) {
 			pr_err("error: writeoob failed at %#llx\n",
@@ -122,8 +122,8 @@ static int verify_eraseblock(int ebnum)
 	int err = 0;
 	loff_t addr = ebnum * mtd->erasesize;
 
+	prandom_bytes_state(&rnd_state, writebuf, use_len_max * pgcnt);
 	for (i = 0; i < pgcnt; ++i, addr += mtd->writesize) {
-		prandom_bytes_state(&rnd_state, writebuf, use_len);
 		ops.mode      = MTD_OPS_AUTO_OOB;
 		ops.len       = 0;
 		ops.retlen    = 0;
@@ -139,7 +139,8 @@ static int verify_eraseblock(int ebnum)
 			errcnt += 1;
 			return err ? err : -1;
 		}
-		if (memcmp(readbuf, writebuf, use_len)) {
+		if (memcmp(readbuf, writebuf + (use_len_max * i) + use_offset,
+			   use_len)) {
 			pr_err("error: verify failed at %#llx\n",
 			       (long long)addr);
 			errcnt += 1;
@@ -166,7 +167,9 @@ static int verify_eraseblock(int ebnum)
 				errcnt += 1;
 				return err ? err : -1;
 			}
-			if (memcmp(readbuf + use_offset, writebuf, use_len)) {
+			if (memcmp(readbuf + use_offset,
+				   writebuf + (use_len_max * i) + use_offset,
+				   use_len)) {
 				pr_err("error: verify failed at %#llx\n",
 						(long long)addr);
 				errcnt += 1;
@@ -566,8 +569,8 @@ static int __init mtd_oobtest_init(void)
 		if (bbt[i] || bbt[i + 1])
 			continue;
 		addr = (i + 1) * mtd->erasesize - mtd->writesize;
+		prandom_bytes_state(&rnd_state, writebuf, sz * cnt);
 		for (pg = 0; pg < cnt; ++pg) {
-			prandom_bytes_state(&rnd_state, writebuf, sz);
 			ops.mode      = MTD_OPS_AUTO_OOB;
 			ops.len       = 0;
 			ops.retlen    = 0;
@@ -575,7 +578,7 @@ static int __init mtd_oobtest_init(void)
 			ops.oobretlen = 0;
 			ops.ooboffs   = 0;
 			ops.datbuf    = NULL;
-			ops.oobbuf    = writebuf;
+			ops.oobbuf    = writebuf + pg * sz;
 			err = mtd_write_oob(mtd, addr, &ops);
 			if (err)
 				goto out;
diff --git a/drivers/mtd/ubi/block.c b/drivers/mtd/ubi/block.c
index 20a667c95da4..8457df7ec5af 100644
--- a/drivers/mtd/ubi/block.c
+++ b/drivers/mtd/ubi/block.c
@@ -432,8 +432,10 @@ int ubiblock_create(struct ubi_volume_info *vi)
 	 * Rembember workqueues are cheap, they're not threads.
 	 */
 	dev->wq = alloc_workqueue("%s", 0, 0, gd->disk_name);
-	if (!dev->wq)
+	if (!dev->wq) {
+		ret = -ENOMEM;
 		goto out_free_queue;
+	}
 	INIT_WORK(&dev->work, ubiblock_do_work);
 
 	mutex_lock(&devices_mutex);
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index f54562a5998e..7646220ca6e2 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -731,7 +731,7 @@ static int rename_volumes(struct ubi_device *ubi,
 			goto out_free;
 		}
 
-		re->desc = ubi_open_volume(ubi->ubi_num, vol_id, UBI_EXCLUSIVE);
+		re->desc = ubi_open_volume(ubi->ubi_num, vol_id, UBI_READWRITE);
 		if (IS_ERR(re->desc)) {
 			err = PTR_ERR(re->desc);
 			ubi_err("cannot open volume %d, error %d", vol_id, err);
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index c5dad652614d..b04e7d059888 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -125,9 +125,9 @@ static struct ubi_ainf_volume *add_vol(struct ubi_attach_info *ai, int vol_id,
 		parent = *p;
 		av = rb_entry(parent, struct ubi_ainf_volume, rb);
 
-		if (vol_id > av->vol_id)
+		if (vol_id < av->vol_id)
 			p = &(*p)->rb_left;
-		else if (vol_id > av->vol_id)
+		else
 			p = &(*p)->rb_right;
 	}
 
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index b667a51ed215..0dfeaf5da3f2 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -157,7 +157,7 @@ static inline struct aggregator *__get_first_agg(struct port *port)
 
 	rcu_read_lock();
 	first_slave = bond_first_slave_rcu(bond);
-	agg = first_slave ? &(SLAVE_AD_INFO(first_slave).aggregator) : NULL;
+	agg = first_slave ? &(SLAVE_AD_INFO(first_slave)->aggregator) : NULL;
 	rcu_read_unlock();
 
 	return agg;
@@ -192,7 +192,7 @@ static inline void __enable_port(struct port *port)
 {
 	struct slave *slave = port->slave;
 
-	if ((slave->link == BOND_LINK_UP) && IS_UP(slave->dev))
+	if ((slave->link == BOND_LINK_UP) && bond_slave_is_up(slave))
 		bond_set_slave_active_flags(slave, BOND_SLAVE_NOTIFY_LATER);
 }
 
@@ -241,7 +241,7 @@ static inline int __check_agg_selection_timer(struct port *port)
  */
 static inline void __get_state_machine_lock(struct port *port)
 {
-	spin_lock_bh(&(SLAVE_AD_INFO(port->slave).state_machine_lock));
+	spin_lock_bh(&(SLAVE_AD_INFO(port->slave)->state_machine_lock));
 }
 
 /**
@@ -250,7 +250,7 @@ static inline void __get_state_machine_lock(struct port *port)
  */
 static inline void __release_state_machine_lock(struct port *port)
 {
-	spin_unlock_bh(&(SLAVE_AD_INFO(port->slave).state_machine_lock));
+	spin_unlock_bh(&(SLAVE_AD_INFO(port->slave)->state_machine_lock));
 }
 
 /**
@@ -350,7 +350,7 @@ static u8 __get_duplex(struct port *port)
 static inline void __initialize_port_locks(struct slave *slave)
 {
 	/* make sure it isn't called twice */
-	spin_lock_init(&(SLAVE_AD_INFO(slave).state_machine_lock));
+	spin_lock_init(&(SLAVE_AD_INFO(slave)->state_machine_lock));
 }
 
 /* Conversions */
@@ -688,8 +688,8 @@ static struct aggregator *__get_active_agg(struct aggregator *aggregator)
 	struct slave *slave;
 
 	bond_for_each_slave_rcu(bond, slave, iter)
-		if (SLAVE_AD_INFO(slave).aggregator.is_active)
-			return &(SLAVE_AD_INFO(slave).aggregator);
+		if (SLAVE_AD_INFO(slave)->aggregator.is_active)
+			return &(SLAVE_AD_INFO(slave)->aggregator);
 
 	return NULL;
 }
@@ -1293,7 +1293,7 @@ static void ad_port_selection_logic(struct port *port)
 	}
 	/* search on all aggregators for a suitable aggregator for this port */
 	bond_for_each_slave(bond, slave, iter) {
-		aggregator = &(SLAVE_AD_INFO(slave).aggregator);
+		aggregator = &(SLAVE_AD_INFO(slave)->aggregator);
 
 		/* keep a free aggregator for later use(if needed) */
 		if (!aggregator->lag_ports) {
@@ -1504,7 +1504,7 @@ static void ad_agg_selection_logic(struct aggregator *agg)
 	best = (active && agg_device_up(active)) ? active : NULL;
 
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		agg = &(SLAVE_AD_INFO(slave).aggregator);
+		agg = &(SLAVE_AD_INFO(slave)->aggregator);
 
 		agg->is_active = 0;
 
@@ -1549,7 +1549,7 @@ static void ad_agg_selection_logic(struct aggregator *agg)
 			 best->slave ? best->slave->dev->name : "NULL");
 
 		bond_for_each_slave_rcu(bond, slave, iter) {
-			agg = &(SLAVE_AD_INFO(slave).aggregator);
+			agg = &(SLAVE_AD_INFO(slave)->aggregator);
 
 			pr_debug("Agg=%d; P=%d; a k=%d; p k=%d; Ind=%d; Act=%d\n",
 				 agg->aggregator_identifier, agg->num_of_ports,
@@ -1840,16 +1840,16 @@ void bond_3ad_bind_slave(struct slave *slave)
 	struct aggregator *aggregator;
 
 	/* check that the slave has not been initialized yet. */
-	if (SLAVE_AD_INFO(slave).port.slave != slave) {
+	if (SLAVE_AD_INFO(slave)->port.slave != slave) {
 
 		/* port initialization */
-		port = &(SLAVE_AD_INFO(slave).port);
+		port = &(SLAVE_AD_INFO(slave)->port);
 
 		ad_initialize_port(port, bond->params.lacp_fast);
 
 		__initialize_port_locks(slave);
 		port->slave = slave;
-		port->actor_port_number = SLAVE_AD_INFO(slave).id;
+		port->actor_port_number = SLAVE_AD_INFO(slave)->id;
 		/* key is determined according to the link speed, duplex and user key(which
 		 * is yet not supported)
 		 */
@@ -1874,7 +1874,7 @@ void bond_3ad_bind_slave(struct slave *slave)
 		__disable_port(port);
 
 		/* aggregator initialization */
-		aggregator = &(SLAVE_AD_INFO(slave).aggregator);
+		aggregator = &(SLAVE_AD_INFO(slave)->aggregator);
 
 		ad_initialize_agg(aggregator);
 
@@ -1903,8 +1903,8 @@ void bond_3ad_unbind_slave(struct slave *slave)
 	struct slave *slave_iter;
 	struct list_head *iter;
 
-	aggregator = &(SLAVE_AD_INFO(slave).aggregator);
-	port = &(SLAVE_AD_INFO(slave).port);
+	aggregator = &(SLAVE_AD_INFO(slave)->aggregator);
+	port = &(SLAVE_AD_INFO(slave)->port);
 
 	/* if slave is null, the whole port is not initialized */
 	if (!port->slave) {
@@ -1932,7 +1932,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
 		    (aggregator->lag_ports->next_port_in_aggregator)) {
 			/* find new aggregator for the related port(s) */
 			bond_for_each_slave(bond, slave_iter, iter) {
-				new_aggregator = &(SLAVE_AD_INFO(slave_iter).aggregator);
+				new_aggregator = &(SLAVE_AD_INFO(slave_iter)->aggregator);
 				/* if the new aggregator is empty, or it is
 				 * connected to our port only
 				 */
@@ -2010,7 +2010,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
 
 	/* find the aggregator that this port is connected to */
 	bond_for_each_slave(bond, slave_iter, iter) {
-		temp_aggregator = &(SLAVE_AD_INFO(slave_iter).aggregator);
+		temp_aggregator = &(SLAVE_AD_INFO(slave_iter)->aggregator);
 		prev_port = NULL;
 		/* search the port in the aggregator's related ports */
 		for (temp_port = temp_aggregator->lag_ports; temp_port;
@@ -2076,7 +2076,7 @@ void bond_3ad_state_machine_handler(struct work_struct *work)
 	if (BOND_AD_INFO(bond).agg_select_timer &&
 	    !(--BOND_AD_INFO(bond).agg_select_timer)) {
 		slave = bond_first_slave_rcu(bond);
-		port = slave ? &(SLAVE_AD_INFO(slave).port) : NULL;
+		port = slave ? &(SLAVE_AD_INFO(slave)->port) : NULL;
 
 		/* select the active aggregator for the bond */
 		if (port) {
@@ -2094,7 +2094,7 @@ void bond_3ad_state_machine_handler(struct work_struct *work)
 
 	/* for each port run the state machines */
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		port = &(SLAVE_AD_INFO(slave).port);
+		port = &(SLAVE_AD_INFO(slave)->port);
 		if (!port->slave) {
 			pr_warn_ratelimited("%s: Warning: Found an uninitialized port\n",
 					    bond->dev->name);
@@ -2155,7 +2155,7 @@ static int bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave,
 
 	if (length >= sizeof(struct lacpdu)) {
 
-		port = &(SLAVE_AD_INFO(slave).port);
+		port = &(SLAVE_AD_INFO(slave)->port);
 
 		if (!port->slave) {
 			pr_warn_ratelimited("%s: Warning: port of slave %s is uninitialized\n",
@@ -2212,7 +2212,7 @@ void bond_3ad_adapter_speed_changed(struct slave *slave)
 {
 	struct port *port;
 
-	port = &(SLAVE_AD_INFO(slave).port);
+	port = &(SLAVE_AD_INFO(slave)->port);
 
 	/* if slave is null, the whole port is not initialized */
 	if (!port->slave) {
@@ -2245,7 +2245,7 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave)
 {
 	struct port *port;
 
-	port = &(SLAVE_AD_INFO(slave).port);
+	port = &(SLAVE_AD_INFO(slave)->port);
 
 	/* if slave is null, the whole port is not initialized */
 	if (!port->slave) {
@@ -2279,7 +2279,7 @@ void bond_3ad_handle_link_change(struct slave *slave, char link)
 {
 	struct port *port;
 
-	port = &(SLAVE_AD_INFO(slave).port);
+	port = &(SLAVE_AD_INFO(slave)->port);
 
 	/* if slave is null, the whole port is not initialized */
 	if (!port->slave) {
@@ -2347,7 +2347,7 @@ int bond_3ad_set_carrier(struct bonding *bond)
 		ret = 0;
 		goto out;
 	}
-	active = __get_active_agg(&(SLAVE_AD_INFO(first_slave).aggregator));
+	active = __get_active_agg(&(SLAVE_AD_INFO(first_slave)->aggregator));
 	if (active) {
 		/* are enough slaves available to consider link up? */
 		if (active->num_of_ports < bond->params.min_links) {
@@ -2384,7 +2384,7 @@ int __bond_3ad_get_active_agg_info(struct bonding *bond,
 	struct port *port;
 
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		port = &(SLAVE_AD_INFO(slave).port);
+		port = &(SLAVE_AD_INFO(slave)->port);
 		if (port->aggregator && port->aggregator->is_active) {
 			aggregator = port->aggregator;
 			break;
@@ -2440,22 +2440,22 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
 		goto err_free;
 	}
 
-	slave_agg_no = bond_xmit_hash(bond, skb, slaves_in_agg);
+	slave_agg_no = bond_xmit_hash(bond, skb) % slaves_in_agg;
 	first_ok_slave = NULL;
 
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		agg = SLAVE_AD_INFO(slave).port.aggregator;
+		agg = SLAVE_AD_INFO(slave)->port.aggregator;
 		if (!agg || agg->aggregator_identifier != agg_id)
 			continue;
 
 		if (slave_agg_no >= 0) {
-			if (!first_ok_slave && SLAVE_IS_OK(slave))
+			if (!first_ok_slave && bond_slave_can_tx(slave))
 				first_ok_slave = slave;
 			slave_agg_no--;
 			continue;
 		}
 
-		if (SLAVE_IS_OK(slave)) {
+		if (bond_slave_can_tx(slave)) {
 			bond_dev_queue_xmit(bond, skb, slave->dev);
 			goto out;
 		}
@@ -2522,7 +2522,7 @@ void bond_3ad_update_lacp_rate(struct bonding *bond)
 
 	lacp_fast = bond->params.lacp_fast;
 	bond_for_each_slave(bond, slave, iter) {
-		port = &(SLAVE_AD_INFO(slave).port);
+		port = &(SLAVE_AD_INFO(slave)->port);
 		__get_state_machine_lock(port);
 		if (lacp_fast)
 			port->actor_oper_port_state |= AD_STATE_LACP_TIMEOUT;
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index 93580a47cc54..76c0dade233f 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -229,7 +229,7 @@ static struct slave *tlb_get_least_loaded_slave(struct bonding *bond)
 
 	/* Find the slave with the largest gap */
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		if (SLAVE_IS_OK(slave)) {
+		if (bond_slave_can_tx(slave)) {
 			long long gap = compute_gap(slave);
 
 			if (max_gap < gap) {
@@ -384,7 +384,7 @@ static struct slave *rlb_next_rx_slave(struct bonding *bond)
 	bool found = false;
 
 	bond_for_each_slave(bond, slave, iter) {
-		if (!SLAVE_IS_OK(slave))
+		if (!bond_slave_can_tx(slave))
 			continue;
 		if (!found) {
 			if (!before || before->speed < slave->speed)
@@ -417,7 +417,7 @@ static struct slave *__rlb_next_rx_slave(struct bonding *bond)
 	bool found = false;
 
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		if (!SLAVE_IS_OK(slave))
+		if (!bond_slave_can_tx(slave))
 			continue;
 		if (!found) {
 			if (!before || before->speed < slave->speed)
@@ -755,7 +755,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
 	/* Don't modify or load balance ARPs that do not originate locally
 	 * (e.g.,arrive via a bridge).
 	 */
-	if (!bond_slave_has_mac_rcu(bond, arp->mac_src))
+	if (!bond_slave_has_mac_rx(bond, arp->mac_src))
 		return NULL;
 
 	if (arp->op_code == htons(ARPOP_REPLY)) {
@@ -1039,11 +1039,14 @@ static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[],
 	struct bonding *bond = bond_get_bond_by_slave(slave);
 	struct net_device *upper;
 	struct list_head *iter;
+	struct bond_vlan_tag tags[BOND_MAX_VLAN_ENCAP];
 
 	/* send untagged */
 	alb_send_lp_vid(slave, mac_addr, 0, 0);
 
-	/* loop through vlans and send one packet for each */
+	/* loop through all devices and see if we need to send a packet
+	 * for that device.
+	 */
 	rcu_read_lock();
 	netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
 		if (is_vlan_dev(upper) && vlan_get_encap_level(upper) == 0) {
@@ -1059,6 +1062,16 @@ static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[],
 						vlan_dev_vlan_id(upper));
 			}
 		}
+
+		/* If this is a macvlan device, then only send updates
+		 * when strict_match is turned off.
+		 */
+		if (netif_is_macvlan(upper) && !strict_match) {
+			memset(tags, 0, sizeof(tags));
+			bond_verify_device_path(bond->dev, upper, tags);
+			alb_send_lp_vid(slave, upper->dev_addr,
+					tags[0].vlan_proto, tags[0].vlan_id);
+		}
 	}
 	rcu_read_unlock();
 }
@@ -1068,7 +1081,7 @@ static int alb_set_slave_mac_addr(struct slave *slave, u8 addr[])
 	struct net_device *dev = slave->dev;
 	struct sockaddr s_addr;
 
-	if (slave->bond->params.mode == BOND_MODE_TLB) {
+	if (BOND_MODE(slave->bond) == BOND_MODE_TLB) {
 		memcpy(dev->dev_addr, addr, dev->addr_len);
 		return 0;
 	}
@@ -1111,13 +1124,13 @@ static void alb_swap_mac_addr(struct slave *slave1, struct slave *slave2)
 static void alb_fasten_mac_swap(struct bonding *bond, struct slave *slave1,
 				struct slave *slave2)
 {
-	int slaves_state_differ = (SLAVE_IS_OK(slave1) != SLAVE_IS_OK(slave2));
+	int slaves_state_differ = (bond_slave_can_tx(slave1) != bond_slave_can_tx(slave2));
 	struct slave *disabled_slave = NULL;
 
 	ASSERT_RTNL();
 
 	/* fasten the change in the switch */
-	if (SLAVE_IS_OK(slave1)) {
+	if (bond_slave_can_tx(slave1)) {
 		alb_send_learning_packets(slave1, slave1->dev->dev_addr, false);
 		if (bond->alb_info.rlb_enabled) {
 			/* inform the clients that the mac address
@@ -1129,7 +1142,7 @@ static void alb_fasten_mac_swap(struct bonding *bond, struct slave *slave1,
 		disabled_slave = slave1;
 	}
 
-	if (SLAVE_IS_OK(slave2)) {
+	if (bond_slave_can_tx(slave2)) {
 		alb_send_learning_packets(slave2, slave2->dev->dev_addr, false);
 		if (bond->alb_info.rlb_enabled) {
 			/* inform the clients that the mac address
@@ -1358,6 +1371,77 @@ void bond_alb_deinitialize(struct bonding *bond)
 		rlb_deinitialize(bond);
 }
 
+static int bond_do_alb_xmit(struct sk_buff *skb, struct bonding *bond,
+		struct slave *tx_slave)
+{
+	struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
+	struct ethhdr *eth_data = eth_hdr(skb);
+
+	if (!tx_slave) {
+		/* unbalanced or unassigned, send through primary */
+		tx_slave = rcu_dereference(bond->curr_active_slave);
+		if (bond->params.tlb_dynamic_lb)
+			bond_info->unbalanced_load += skb->len;
+	}
+
+	if (tx_slave && bond_slave_can_tx(tx_slave)) {
+		if (tx_slave != rcu_dereference(bond->curr_active_slave)) {
+			ether_addr_copy(eth_data->h_source,
+					tx_slave->dev->dev_addr);
+		}
+
+		bond_dev_queue_xmit(bond, skb, tx_slave->dev);
+		goto out;
+	}
+
+	if (tx_slave && bond->params.tlb_dynamic_lb) {
+		_lock_tx_hashtbl(bond);
+		__tlb_clear_slave(bond, tx_slave, 0);
+		_unlock_tx_hashtbl(bond);
+	}
+
+	/* no suitable interface, frame not sent */
+	dev_kfree_skb_any(skb);
+out:
+	return NETDEV_TX_OK;
+}
+
+int bond_tlb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
+{
+	struct bonding *bond = netdev_priv(bond_dev);
+	struct ethhdr *eth_data;
+	struct slave *tx_slave = NULL;
+	u32 hash_index;
+
+	skb_reset_mac_header(skb);
+	eth_data = eth_hdr(skb);
+
+	/* Do not TX balance any multicast or broadcast */
+	if (!is_multicast_ether_addr(eth_data->h_dest)) {
+		switch (skb->protocol) {
+		case htons(ETH_P_IP):
+		case htons(ETH_P_IPX):
+		    /* In case of IPX, it will falback to L2 hash */
+		case htons(ETH_P_IPV6):
+			hash_index = bond_xmit_hash(bond, skb);
+			if (bond->params.tlb_dynamic_lb) {
+				tx_slave = tlb_choose_channel(bond,
+							      hash_index & 0xFF,
+							      skb->len);
+			} else {
+				struct list_head *iter;
+				int idx = hash_index % bond->slave_cnt;
+
+				bond_for_each_slave_rcu(bond, tx_slave, iter)
+					if (--idx < 0)
+						break;
+			}
+			break;
+		}
+	}
+	return bond_do_alb_xmit(skb, bond, tx_slave);
+}
+
 int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 {
 	struct bonding *bond = netdev_priv(bond_dev);
@@ -1366,7 +1450,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 	struct slave *tx_slave = NULL;
 	static const __be32 ip_bcast = htonl(0xffffffff);
 	int hash_size = 0;
-	int do_tx_balance = 1;
+	bool do_tx_balance = true;
 	u32 hash_index = 0;
 	const u8 *hash_start = NULL;
 	struct ipv6hdr *ip6hdr;
@@ -1381,7 +1465,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 		if (ether_addr_equal_64bits(eth_data->h_dest, mac_bcast) ||
 		    (iph->daddr == ip_bcast) ||
 		    (iph->protocol == IPPROTO_IGMP)) {
-			do_tx_balance = 0;
+			do_tx_balance = false;
 			break;
 		}
 		hash_start = (char *)&(iph->daddr);
@@ -1393,7 +1477,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 		 * that here just in case.
 		 */
 		if (ether_addr_equal_64bits(eth_data->h_dest, mac_bcast)) {
-			do_tx_balance = 0;
+			do_tx_balance = false;
 			break;
 		}
 
@@ -1401,7 +1485,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 		 * broadcasts in IPv4.
 		 */
 		if (ether_addr_equal_64bits(eth_data->h_dest, mac_v6_allmcast)) {
-			do_tx_balance = 0;
+			do_tx_balance = false;
 			break;
 		}
 
@@ -1411,7 +1495,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 		 */
 		ip6hdr = ipv6_hdr(skb);
 		if (ipv6_addr_any(&ip6hdr->saddr)) {
-			do_tx_balance = 0;
+			do_tx_balance = false;
 			break;
 		}
 
@@ -1421,7 +1505,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 	case ETH_P_IPX:
 		if (ipx_hdr(skb)->ipx_checksum != IPX_NO_CHECKSUM) {
 			/* something is wrong with this packet */
-			do_tx_balance = 0;
+			do_tx_balance = false;
 			break;
 		}
 
@@ -1430,7 +1514,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 			 * this family since it has an "ARP" like
 			 * mechanism
 			 */
-			do_tx_balance = 0;
+			do_tx_balance = false;
 			break;
 		}
 
@@ -1438,12 +1522,12 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 		hash_size = ETH_ALEN;
 		break;
 	case ETH_P_ARP:
-		do_tx_balance = 0;
+		do_tx_balance = false;
 		if (bond_info->rlb_enabled)
 			tx_slave = rlb_arp_xmit(skb, bond);
 		break;
 	default:
-		do_tx_balance = 0;
+		do_tx_balance = false;
 		break;
 	}
 
@@ -1452,32 +1536,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 		tx_slave = tlb_choose_channel(bond, hash_index, skb->len);
 	}
 
-	if (!tx_slave) {
-		/* unbalanced or unassigned, send through primary */
-		tx_slave = rcu_dereference(bond->curr_active_slave);
-		bond_info->unbalanced_load += skb->len;
-	}
-
-	if (tx_slave && SLAVE_IS_OK(tx_slave)) {
-		if (tx_slave != rcu_dereference(bond->curr_active_slave)) {
-			ether_addr_copy(eth_data->h_source,
-					tx_slave->dev->dev_addr);
-		}
-
-		bond_dev_queue_xmit(bond, skb, tx_slave->dev);
-		goto out;
-	}
-
-	if (tx_slave) {
-		_lock_tx_hashtbl(bond);
-		__tlb_clear_slave(bond, tx_slave, 0);
-		_unlock_tx_hashtbl(bond);
-	}
-
-	/* no suitable interface, frame not sent */
-	dev_kfree_skb_any(skb);
-out:
-	return NETDEV_TX_OK;
+	return bond_do_alb_xmit(skb, bond, tx_slave);
 }
 
 void bond_alb_monitor(struct work_struct *work)
@@ -1514,8 +1573,10 @@ void bond_alb_monitor(struct work_struct *work)
 			/* If updating current_active, use all currently
 			 * user mac addreses (!strict_match).  Otherwise, only
 			 * use mac of the slave device.
+			 * In RLB mode, we always use strict matches.
 			 */
-			strict_match = (slave != bond->curr_active_slave);
+			strict_match = (slave != bond->curr_active_slave ||
+					bond_info->rlb_enabled);
 			alb_send_learning_packets(slave, slave->dev->dev_addr,
 						  strict_match);
 		}
@@ -1719,7 +1780,7 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave
 	/* in TLB mode, the slave might flip down/up with the old dev_addr,
 	 * and thus filter bond->dev_addr's packets, so force bond's mac
 	 */
-	if (bond->params.mode == BOND_MODE_TLB) {
+	if (BOND_MODE(bond) == BOND_MODE_TLB) {
 		struct sockaddr sa;
 		u8 tmp_addr[ETH_ALEN];
 
diff --git a/drivers/net/bonding/bond_alb.h b/drivers/net/bonding/bond_alb.h
index e09dd4bfafff..5fc76c01636c 100644
--- a/drivers/net/bonding/bond_alb.h
+++ b/drivers/net/bonding/bond_alb.h
@@ -175,6 +175,7 @@ void bond_alb_deinit_slave(struct bonding *bond, struct slave *slave);
 void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char link);
 void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave);
 int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev);
+int bond_tlb_xmit(struct sk_buff *skb, struct net_device *bond_dev);
 void bond_alb_monitor(struct work_struct *);
 int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr);
 void bond_alb_clear_vlan(struct bonding *bond, unsigned short vlan_id);
diff --git a/drivers/net/bonding/bond_debugfs.c b/drivers/net/bonding/bond_debugfs.c
index 2d3f7fa541ff..658e761c4568 100644
--- a/drivers/net/bonding/bond_debugfs.c
+++ b/drivers/net/bonding/bond_debugfs.c
@@ -23,7 +23,7 @@ static int bond_debug_rlb_hash_show(struct seq_file *m, void *v)
 	struct rlb_client_info *client_info;
 	u32 hash_index;
 
-	if (bond->params.mode != BOND_MODE_ALB)
+	if (BOND_MODE(bond) != BOND_MODE_ALB)
 		return 0;
 
 	seq_printf(m, "SourceIP        DestinationIP   "
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d3a67896d435..3a451b6cd3d5 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -343,7 +343,7 @@ static int bond_set_carrier(struct bonding *bond)
 	if (!bond_has_slaves(bond))
 		goto down;
 
-	if (bond->params.mode == BOND_MODE_8023AD)
+	if (BOND_MODE(bond) == BOND_MODE_8023AD)
 		return bond_3ad_set_carrier(bond);
 
 	bond_for_each_slave(bond, slave, iter) {
@@ -497,7 +497,7 @@ static int bond_set_promiscuity(struct bonding *bond, int inc)
 	struct list_head *iter;
 	int err = 0;
 
-	if (USES_PRIMARY(bond->params.mode)) {
+	if (bond_uses_primary(bond)) {
 		/* write lock already acquired */
 		if (bond->curr_active_slave) {
 			err = dev_set_promiscuity(bond->curr_active_slave->dev,
@@ -523,7 +523,7 @@ static int bond_set_allmulti(struct bonding *bond, int inc)
 	struct list_head *iter;
 	int err = 0;
 
-	if (USES_PRIMARY(bond->params.mode)) {
+	if (bond_uses_primary(bond)) {
 		/* write lock already acquired */
 		if (bond->curr_active_slave) {
 			err = dev_set_allmulti(bond->curr_active_slave->dev,
@@ -574,7 +574,7 @@ static void bond_hw_addr_flush(struct net_device *bond_dev,
 	dev_uc_unsync(slave_dev, bond_dev);
 	dev_mc_unsync(slave_dev, bond_dev);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		/* del lacpdu mc addr from mc list */
 		u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR;
 
@@ -585,8 +585,8 @@ static void bond_hw_addr_flush(struct net_device *bond_dev,
 /*--------------------------- Active slave change ---------------------------*/
 
 /* Update the hardware address list and promisc/allmulti for the new and
- * old active slaves (if any).  Modes that are !USES_PRIMARY keep all
- * slaves up date at all times; only the USES_PRIMARY modes need to call
+ * old active slaves (if any).  Modes that are not using primary keep all
+ * slaves up date at all times; only the modes that use primary need to call
  * this function to swap these settings during a failover.
  */
 static void bond_hw_addr_swap(struct bonding *bond, struct slave *new_active,
@@ -747,7 +747,7 @@ static struct slave *bond_find_best_slave(struct bonding *bond)
 	bond_for_each_slave(bond, slave, iter) {
 		if (slave->link == BOND_LINK_UP)
 			return slave;
-		if (slave->link == BOND_LINK_BACK && IS_UP(slave->dev) &&
+		if (slave->link == BOND_LINK_BACK && bond_slave_is_up(slave) &&
 		    slave->delay < mintime) {
 			mintime = slave->delay;
 			bestslave = slave;
@@ -801,7 +801,7 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
 		new_active->last_link_up = jiffies;
 
 		if (new_active->link == BOND_LINK_BACK) {
-			if (USES_PRIMARY(bond->params.mode)) {
+			if (bond_uses_primary(bond)) {
 				pr_info("%s: making interface %s the new active one %d ms earlier\n",
 					bond->dev->name, new_active->dev->name,
 					(bond->params.updelay - new_active->delay) * bond->params.miimon);
@@ -810,20 +810,20 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
 			new_active->delay = 0;
 			new_active->link = BOND_LINK_UP;
 
-			if (bond->params.mode == BOND_MODE_8023AD)
+			if (BOND_MODE(bond) == BOND_MODE_8023AD)
 				bond_3ad_handle_link_change(new_active, BOND_LINK_UP);
 
 			if (bond_is_lb(bond))
 				bond_alb_handle_link_change(bond, new_active, BOND_LINK_UP);
 		} else {
-			if (USES_PRIMARY(bond->params.mode)) {
+			if (bond_uses_primary(bond)) {
 				pr_info("%s: making interface %s the new active one\n",
 					bond->dev->name, new_active->dev->name);
 			}
 		}
 	}
 
-	if (USES_PRIMARY(bond->params.mode))
+	if (bond_uses_primary(bond))
 		bond_hw_addr_swap(bond, new_active, old_active);
 
 	if (bond_is_lb(bond)) {
@@ -838,7 +838,7 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
 		rcu_assign_pointer(bond->curr_active_slave, new_active);
 	}
 
-	if (bond->params.mode == BOND_MODE_ACTIVEBACKUP) {
+	if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) {
 		if (old_active)
 			bond_set_slave_inactive_flags(old_active,
 						      BOND_SLAVE_NOTIFY_NOW);
@@ -876,8 +876,8 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
 	 * resend only if bond is brought up with the affected
 	 * bonding modes and the retransmission is enabled */
 	if (netif_running(bond->dev) && (bond->params.resend_igmp > 0) &&
-	    ((USES_PRIMARY(bond->params.mode) && new_active) ||
-	     bond->params.mode == BOND_MODE_ROUNDROBIN)) {
+	    ((bond_uses_primary(bond) && new_active) ||
+	     BOND_MODE(bond) == BOND_MODE_ROUNDROBIN)) {
 		bond->igmp_retrans = bond->params.resend_igmp;
 		queue_delayed_work(bond->wq, &bond->mcast_work, 1);
 	}
@@ -958,7 +958,7 @@ static void bond_netpoll_cleanup(struct net_device *bond_dev)
 	struct slave *slave;
 
 	bond_for_each_slave(bond, slave, iter)
-		if (IS_UP(slave->dev))
+		if (bond_slave_is_up(slave))
 			slave_disable_netpoll(slave);
 }
 
@@ -1025,10 +1025,14 @@ static netdev_features_t bond_fix_features(struct net_device *dev,
 				 NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \
 				 NETIF_F_HIGHDMA | NETIF_F_LRO)
 
+#define BOND_ENC_FEATURES	(NETIF_F_ALL_CSUM | NETIF_F_SG | NETIF_F_RXCSUM |\
+				 NETIF_F_TSO | NETIF_F_GSO_UDP_TUNNEL)
+
 static void bond_compute_features(struct bonding *bond)
 {
 	unsigned int flags, dst_release_flag = IFF_XMIT_DST_RELEASE;
 	netdev_features_t vlan_features = BOND_VLAN_FEATURES;
+	netdev_features_t enc_features  = BOND_ENC_FEATURES;
 	struct net_device *bond_dev = bond->dev;
 	struct list_head *iter;
 	struct slave *slave;
@@ -1038,11 +1042,15 @@ static void bond_compute_features(struct bonding *bond)
 
 	if (!bond_has_slaves(bond))
 		goto done;
+	vlan_features &= NETIF_F_ALL_FOR_ALL;
 
 	bond_for_each_slave(bond, slave, iter) {
 		vlan_features = netdev_increment_features(vlan_features,
 			slave->dev->vlan_features, BOND_VLAN_FEATURES);
 
+		enc_features = netdev_increment_features(enc_features,
+							 slave->dev->hw_enc_features,
+							 BOND_ENC_FEATURES);
 		dst_release_flag &= slave->dev->priv_flags;
 		if (slave->dev->hard_header_len > max_hard_header_len)
 			max_hard_header_len = slave->dev->hard_header_len;
@@ -1053,6 +1061,7 @@ static void bond_compute_features(struct bonding *bond)
 
 done:
 	bond_dev->vlan_features = vlan_features;
+	bond_dev->hw_enc_features = enc_features;
 	bond_dev->hard_header_len = max_hard_header_len;
 	bond_dev->gso_max_segs = gso_max_segs;
 	netif_set_gso_max_size(bond_dev, gso_max_size);
@@ -1084,7 +1093,7 @@ static bool bond_should_deliver_exact_match(struct sk_buff *skb,
 					    struct bonding *bond)
 {
 	if (bond_is_slave_inactive(slave)) {
-		if (bond->params.mode == BOND_MODE_ALB &&
+		if (BOND_MODE(bond) == BOND_MODE_ALB &&
 		    skb->pkt_type != PACKET_BROADCAST &&
 		    skb->pkt_type != PACKET_MULTICAST)
 			return false;
@@ -1126,7 +1135,7 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb)
 
 	skb->dev = bond->dev;
 
-	if (bond->params.mode == BOND_MODE_ALB &&
+	if (BOND_MODE(bond) == BOND_MODE_ALB &&
 	    bond->dev->priv_flags & IFF_BRIDGE_PORT &&
 	    skb->pkt_type == PACKET_HOST) {
 
@@ -1163,6 +1172,35 @@ static void bond_upper_dev_unlink(struct net_device *bond_dev,
 	rtmsg_ifinfo(RTM_NEWLINK, slave_dev, IFF_SLAVE, GFP_KERNEL);
 }
 
+static struct slave *bond_alloc_slave(struct bonding *bond)
+{
+	struct slave *slave = NULL;
+
+	slave = kzalloc(sizeof(struct slave), GFP_KERNEL);
+	if (!slave)
+		return NULL;
+
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
+		SLAVE_AD_INFO(slave) = kzalloc(sizeof(struct ad_slave_info),
+					       GFP_KERNEL);
+		if (!SLAVE_AD_INFO(slave)) {
+			kfree(slave);
+			return NULL;
+		}
+	}
+	return slave;
+}
+
+static void bond_free_slave(struct slave *slave)
+{
+	struct bonding *bond = bond_get_bond_by_slave(slave);
+
+	if (BOND_MODE(bond) == BOND_MODE_8023AD)
+		kfree(SLAVE_AD_INFO(slave));
+
+	kfree(slave);
+}
+
 /* enslave device <slave> to bond device <master> */
 int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 {
@@ -1269,7 +1307,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		if (!bond_has_slaves(bond)) {
 			pr_warn("%s: Warning: The first slave device specified does not support setting the MAC address\n",
 				bond_dev->name);
-			if (bond->params.mode == BOND_MODE_ACTIVEBACKUP) {
+			if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) {
 				bond->params.fail_over_mac = BOND_FOM_ACTIVE;
 				pr_warn("%s: Setting fail_over_mac to active for active-backup mode\n",
 					bond_dev->name);
@@ -1290,11 +1328,14 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	    bond->dev->addr_assign_type == NET_ADDR_RANDOM)
 		bond_set_dev_addr(bond->dev, slave_dev);
 
-	new_slave = kzalloc(sizeof(struct slave), GFP_KERNEL);
+	new_slave = bond_alloc_slave(bond);
 	if (!new_slave) {
 		res = -ENOMEM;
 		goto err_undo_flags;
 	}
+
+	new_slave->bond = bond;
+	new_slave->dev = slave_dev;
 	/*
 	 * Set the new_slave's queue_id to be zero.  Queue ID mapping
 	 * is set via sysfs or module option if desired.
@@ -1317,7 +1358,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	ether_addr_copy(new_slave->perm_hwaddr, slave_dev->dev_addr);
 
 	if (!bond->params.fail_over_mac ||
-	    bond->params.mode != BOND_MODE_ACTIVEBACKUP) {
+	    BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) {
 		/*
 		 * Set slave to master's mac address.  The application already
 		 * set the master's mac address to that of the first slave
@@ -1338,8 +1379,6 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		goto err_restore_mac;
 	}
 
-	new_slave->bond = bond;
-	new_slave->dev = slave_dev;
 	slave_dev->priv_flags |= IFF_BONDING;
 
 	if (bond_is_lb(bond)) {
@@ -1351,10 +1390,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 			goto err_close;
 	}
 
-	/* If the mode USES_PRIMARY, then the following is handled by
+	/* If the mode uses primary, then the following is handled by
 	 * bond_change_active_slave().
 	 */
-	if (!USES_PRIMARY(bond->params.mode)) {
+	if (!bond_uses_primary(bond)) {
 		/* set promiscuity level to new slave */
 		if (bond_dev->flags & IFF_PROMISC) {
 			res = dev_set_promiscuity(slave_dev, 1);
@@ -1377,7 +1416,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		netif_addr_unlock_bh(bond_dev);
 	}
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		/* add lacpdu mc addr to mc list */
 		u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR;
 
@@ -1450,7 +1489,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		 new_slave->link == BOND_LINK_DOWN ? "DOWN" :
 		 (new_slave->link == BOND_LINK_UP ? "UP" : "BACK"));
 
-	if (USES_PRIMARY(bond->params.mode) && bond->params.primary[0]) {
+	if (bond_uses_primary(bond) && bond->params.primary[0]) {
 		/* if there is a primary slave, remember it */
 		if (strcmp(bond->params.primary, new_slave->dev->name) == 0) {
 			bond->primary_slave = new_slave;
@@ -1458,7 +1497,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		}
 	}
 
-	switch (bond->params.mode) {
+	switch (BOND_MODE(bond)) {
 	case BOND_MODE_ACTIVEBACKUP:
 		bond_set_slave_inactive_flags(new_slave,
 					      BOND_SLAVE_NOTIFY_NOW);
@@ -1471,14 +1510,14 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		bond_set_slave_inactive_flags(new_slave, BOND_SLAVE_NOTIFY_NOW);
 		/* if this is the first slave */
 		if (!prev_slave) {
-			SLAVE_AD_INFO(new_slave).id = 1;
+			SLAVE_AD_INFO(new_slave)->id = 1;
 			/* Initialize AD with the number of times that the AD timer is called in 1 second
 			 * can be called only after the mac address of the bond is set
 			 */
 			bond_3ad_initialize(bond, 1000/AD_TIMER_INTERVAL);
 		} else {
-			SLAVE_AD_INFO(new_slave).id =
-				SLAVE_AD_INFO(prev_slave).id + 1;
+			SLAVE_AD_INFO(new_slave)->id =
+				SLAVE_AD_INFO(prev_slave)->id + 1;
 		}
 
 		bond_3ad_bind_slave(new_slave);
@@ -1539,7 +1578,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	bond_compute_features(bond);
 	bond_set_carrier(bond);
 
-	if (USES_PRIMARY(bond->params.mode)) {
+	if (bond_uses_primary(bond)) {
 		block_netpoll_tx();
 		write_lock_bh(&bond->curr_slave_lock);
 		bond_select_active_slave(bond);
@@ -1563,7 +1602,7 @@ err_unregister:
 	netdev_rx_handler_unregister(slave_dev);
 
 err_detach:
-	if (!USES_PRIMARY(bond->params.mode))
+	if (!bond_uses_primary(bond))
 		bond_hw_addr_flush(bond_dev, slave_dev);
 
 	vlan_vids_del_by_dev(slave_dev, bond_dev);
@@ -1585,7 +1624,7 @@ err_close:
 
 err_restore_mac:
 	if (!bond->params.fail_over_mac ||
-	    bond->params.mode != BOND_MODE_ACTIVEBACKUP) {
+	    BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) {
 		/* XXX TODO - fom follow mode needs to change master's
 		 * MAC if this slave's MAC is in use by the bond, or at
 		 * least print a warning.
@@ -1599,7 +1638,7 @@ err_restore_mtu:
 	dev_set_mtu(slave_dev, new_slave->original_mtu);
 
 err_free:
-	kfree(new_slave);
+	bond_free_slave(new_slave);
 
 err_undo_flags:
 	/* Enslave of first slave has failed and we need to fix master's mac */
@@ -1661,7 +1700,7 @@ static int __bond_release_one(struct net_device *bond_dev,
 	write_lock_bh(&bond->lock);
 
 	/* Inform AD package of unbinding of slave. */
-	if (bond->params.mode == BOND_MODE_8023AD)
+	if (BOND_MODE(bond) == BOND_MODE_8023AD)
 		bond_3ad_unbind_slave(slave);
 
 	write_unlock_bh(&bond->lock);
@@ -1676,7 +1715,7 @@ static int __bond_release_one(struct net_device *bond_dev,
 	bond->current_arp_slave = NULL;
 
 	if (!all && (!bond->params.fail_over_mac ||
-		     bond->params.mode != BOND_MODE_ACTIVEBACKUP)) {
+		     BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP)) {
 		if (ether_addr_equal_64bits(bond_dev->dev_addr, slave->perm_hwaddr) &&
 		    bond_has_slaves(bond))
 			pr_warn("%s: Warning: the permanent HWaddr of %s - %pM - is still in use by %s - set the HWaddr of %s to a different address to avoid conflicts\n",
@@ -1748,10 +1787,10 @@ static int __bond_release_one(struct net_device *bond_dev,
 	/* must do this from outside any spinlocks */
 	vlan_vids_del_by_dev(slave_dev, bond_dev);
 
-	/* If the mode USES_PRIMARY, then this cases was handled above by
+	/* If the mode uses primary, then this cases was handled above by
 	 * bond_change_active_slave(..., NULL)
 	 */
-	if (!USES_PRIMARY(bond->params.mode)) {
+	if (!bond_uses_primary(bond)) {
 		/* unset promiscuity level from slave
 		 * NOTE: The NETDEV_CHANGEADDR call above may change the value
 		 * of the IFF_PROMISC flag in the bond_dev, but we need the
@@ -1775,7 +1814,7 @@ static int __bond_release_one(struct net_device *bond_dev,
 	dev_close(slave_dev);
 
 	if (bond->params.fail_over_mac != BOND_FOM_ACTIVE ||
-	    bond->params.mode != BOND_MODE_ACTIVEBACKUP) {
+	    BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) {
 		/* restore original ("permanent") mac address */
 		ether_addr_copy(addr.sa_data, slave->perm_hwaddr);
 		addr.sa_family = slave_dev->type;
@@ -1786,7 +1825,7 @@ static int __bond_release_one(struct net_device *bond_dev,
 
 	slave_dev->priv_flags &= ~IFF_BONDING;
 
-	kfree(slave);
+	bond_free_slave(slave);
 
 	return 0;  /* deletion OK */
 }
@@ -1821,7 +1860,7 @@ static int bond_info_query(struct net_device *bond_dev, struct ifbond *info)
 {
 	struct bonding *bond = netdev_priv(bond_dev);
 
-	info->bond_mode = bond->params.mode;
+	info->bond_mode = BOND_MODE(bond);
 	info->miimon = bond->params.miimon;
 
 	info->num_slaves = bond->slave_cnt;
@@ -1877,7 +1916,7 @@ static int bond_miimon_inspect(struct bonding *bond)
 			if (slave->delay) {
 				pr_info("%s: link status down for %sinterface %s, disabling it in %d ms\n",
 					bond->dev->name,
-					(bond->params.mode ==
+					(BOND_MODE(bond) ==
 					 BOND_MODE_ACTIVEBACKUP) ?
 					(bond_is_active_slave(slave) ?
 					 "active " : "backup ") : "",
@@ -1968,10 +2007,10 @@ static void bond_miimon_commit(struct bonding *bond)
 			slave->link = BOND_LINK_UP;
 			slave->last_link_up = jiffies;
 
-			if (bond->params.mode == BOND_MODE_8023AD) {
+			if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 				/* prevent it from being the active one */
 				bond_set_backup_slave(slave);
-			} else if (bond->params.mode != BOND_MODE_ACTIVEBACKUP) {
+			} else if (BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) {
 				/* make it immediately active */
 				bond_set_active_slave(slave);
 			} else if (slave != bond->primary_slave) {
@@ -1985,7 +2024,7 @@ static void bond_miimon_commit(struct bonding *bond)
 				slave->duplex ? "full" : "half");
 
 			/* notify ad that the link status has changed */
-			if (bond->params.mode == BOND_MODE_8023AD)
+			if (BOND_MODE(bond) == BOND_MODE_8023AD)
 				bond_3ad_handle_link_change(slave, BOND_LINK_UP);
 
 			if (bond_is_lb(bond))
@@ -2004,15 +2043,15 @@ static void bond_miimon_commit(struct bonding *bond)
 
 			slave->link = BOND_LINK_DOWN;
 
-			if (bond->params.mode == BOND_MODE_ACTIVEBACKUP ||
-			    bond->params.mode == BOND_MODE_8023AD)
+			if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP ||
+			    BOND_MODE(bond) == BOND_MODE_8023AD)
 				bond_set_slave_inactive_flags(slave,
 							      BOND_SLAVE_NOTIFY_NOW);
 
 			pr_info("%s: link status definitely down for interface %s, disabling it\n",
 				bond->dev->name, slave->dev->name);
 
-			if (bond->params.mode == BOND_MODE_8023AD)
+			if (BOND_MODE(bond) == BOND_MODE_8023AD)
 				bond_3ad_handle_link_change(slave,
 							    BOND_LINK_DOWN);
 
@@ -2175,9 +2214,9 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
  * When the path is validated, collect any vlan information in the
  * path.
  */
-static bool bond_verify_device_path(struct net_device *start_dev,
-				    struct net_device *end_dev,
-				    struct bond_vlan_tag *tags)
+bool bond_verify_device_path(struct net_device *start_dev,
+			     struct net_device *end_dev,
+			     struct bond_vlan_tag *tags)
 {
 	struct net_device *upper;
 	struct list_head  *iter;
@@ -2287,8 +2326,8 @@ int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond,
 	int alen, is_arp = skb->protocol == __cpu_to_be16(ETH_P_ARP);
 
 	if (!slave_do_arp_validate(bond, slave)) {
-		if ((slave_do_arp_validate_only(bond, slave) && is_arp) ||
-		    !slave_do_arp_validate_only(bond, slave))
+		if ((slave_do_arp_validate_only(bond) && is_arp) ||
+		    !slave_do_arp_validate_only(bond))
 			slave->last_rx = jiffies;
 		return RX_HANDLER_ANOTHER;
 	} else if (!is_arp) {
@@ -2456,7 +2495,7 @@ static void bond_loadbalance_arp_mon(struct work_struct *work)
 		 * do - all replies will be rx'ed on same link causing slaves
 		 * to be unstable during low/no traffic periods
 		 */
-		if (IS_UP(slave->dev))
+		if (bond_slave_is_up(slave))
 			bond_arp_send_all(bond, slave);
 	}
 
@@ -2678,10 +2717,10 @@ static bool bond_ab_arp_probe(struct bonding *bond)
 	bond_set_slave_inactive_flags(curr_arp_slave, BOND_SLAVE_NOTIFY_LATER);
 
 	bond_for_each_slave_rcu(bond, slave, iter) {
-		if (!found && !before && IS_UP(slave->dev))
+		if (!found && !before && bond_slave_is_up(slave))
 			before = slave;
 
-		if (found && !new_slave && IS_UP(slave->dev))
+		if (found && !new_slave && bond_slave_is_up(slave))
 			new_slave = slave;
 		/* if the link state is up at this point, we
 		 * mark it down - this can happen if we have
@@ -2690,7 +2729,7 @@ static bool bond_ab_arp_probe(struct bonding *bond)
 		 * one the current slave so it is still marked
 		 * up when it is actually down
 		 */
-		if (!IS_UP(slave->dev) && slave->link == BOND_LINK_UP) {
+		if (!bond_slave_is_up(slave) && slave->link == BOND_LINK_UP) {
 			slave->link = BOND_LINK_DOWN;
 			if (slave->link_failure_count < UINT_MAX)
 				slave->link_failure_count++;
@@ -2853,7 +2892,7 @@ static int bond_slave_netdev_event(unsigned long event,
 
 		bond_update_speed_duplex(slave);
 
-		if (bond->params.mode == BOND_MODE_8023AD) {
+		if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 			if (old_speed != slave->speed)
 				bond_3ad_adapter_speed_changed(slave);
 			if (old_duplex != slave->duplex)
@@ -2881,7 +2920,7 @@ static int bond_slave_netdev_event(unsigned long event,
 		break;
 	case NETDEV_CHANGENAME:
 		/* we don't care if we don't have primary set */
-		if (!USES_PRIMARY(bond->params.mode) ||
+		if (!bond_uses_primary(bond) ||
 		    !bond->params.primary[0])
 			break;
 
@@ -3011,20 +3050,18 @@ static bool bond_flow_dissect(struct bonding *bond, struct sk_buff *skb,
  * bond_xmit_hash - generate a hash value based on the xmit policy
  * @bond: bonding device
  * @skb: buffer to use for headers
- * @count: modulo value
  *
  * This function will extract the necessary headers from the skb buffer and use
  * them to generate a hash based on the xmit_policy set in the bonding device
- * which will be reduced modulo count before returning.
  */
-int bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, int count)
+u32 bond_xmit_hash(struct bonding *bond, struct sk_buff *skb)
 {
 	struct flow_keys flow;
 	u32 hash;
 
 	if (bond->params.xmit_policy == BOND_XMIT_POLICY_LAYER2 ||
 	    !bond_flow_dissect(bond, skb, &flow))
-		return bond_eth_hash(skb) % count;
+		return bond_eth_hash(skb);
 
 	if (bond->params.xmit_policy == BOND_XMIT_POLICY_LAYER23 ||
 	    bond->params.xmit_policy == BOND_XMIT_POLICY_ENCAP23)
@@ -3035,7 +3072,7 @@ int bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, int count)
 	hash ^= (hash >> 16);
 	hash ^= (hash >> 8);
 
-	return hash % count;
+	return hash;
 }
 
 /*-------------------------- Device entry points ----------------------------*/
@@ -3046,7 +3083,7 @@ static void bond_work_init_all(struct bonding *bond)
 			  bond_resend_igmp_join_requests_delayed);
 	INIT_DELAYED_WORK(&bond->alb_work, bond_alb_monitor);
 	INIT_DELAYED_WORK(&bond->mii_work, bond_mii_monitor);
-	if (bond->params.mode == BOND_MODE_ACTIVEBACKUP)
+	if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP)
 		INIT_DELAYED_WORK(&bond->arp_work, bond_activebackup_arp_mon);
 	else
 		INIT_DELAYED_WORK(&bond->arp_work, bond_loadbalance_arp_mon);
@@ -3073,7 +3110,7 @@ static int bond_open(struct net_device *bond_dev)
 	if (bond_has_slaves(bond)) {
 		read_lock(&bond->curr_slave_lock);
 		bond_for_each_slave(bond, slave, iter) {
-			if (USES_PRIMARY(bond->params.mode)
+			if (bond_uses_primary(bond)
 				&& (slave != bond->curr_active_slave)) {
 				bond_set_slave_inactive_flags(slave,
 							      BOND_SLAVE_NOTIFY_NOW);
@@ -3092,9 +3129,10 @@ static int bond_open(struct net_device *bond_dev)
 		/* bond_alb_initialize must be called before the timer
 		 * is started.
 		 */
-		if (bond_alb_initialize(bond, (bond->params.mode == BOND_MODE_ALB)))
+		if (bond_alb_initialize(bond, (BOND_MODE(bond) == BOND_MODE_ALB)))
 			return -ENOMEM;
-		queue_delayed_work(bond->wq, &bond->alb_work, 0);
+		if (bond->params.tlb_dynamic_lb)
+			queue_delayed_work(bond->wq, &bond->alb_work, 0);
 	}
 
 	if (bond->params.miimon)  /* link check interval, in milliseconds. */
@@ -3105,7 +3143,7 @@ static int bond_open(struct net_device *bond_dev)
 		bond->recv_probe = bond_arp_rcv;
 	}
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		queue_delayed_work(bond->wq, &bond->ad_work, 0);
 		/* register to receive LACPDUs */
 		bond->recv_probe = bond_3ad_lacpdu_recv;
@@ -3310,7 +3348,7 @@ static void bond_set_rx_mode(struct net_device *bond_dev)
 
 
 	rcu_read_lock();
-	if (USES_PRIMARY(bond->params.mode)) {
+	if (bond_uses_primary(bond)) {
 		slave = rcu_dereference(bond->curr_active_slave);
 		if (slave) {
 			dev_uc_sync(slave->dev, bond_dev);
@@ -3464,7 +3502,7 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr)
 	struct list_head *iter;
 	int res = 0;
 
-	if (bond->params.mode == BOND_MODE_ALB)
+	if (BOND_MODE(bond) == BOND_MODE_ALB)
 		return bond_alb_set_mac_address(bond_dev, addr);
 
 
@@ -3475,7 +3513,7 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr)
 	 * Returning an error causes ifenslave to fail.
 	 */
 	if (bond->params.fail_over_mac &&
-	    bond->params.mode == BOND_MODE_ACTIVEBACKUP)
+	    BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP)
 		return 0;
 
 	if (!is_valid_ether_addr(sa->sa_data))
@@ -3555,7 +3593,7 @@ static void bond_xmit_slave_id(struct bonding *bond, struct sk_buff *skb, int sl
 	/* Here we start from the slave with slave_id */
 	bond_for_each_slave_rcu(bond, slave, iter) {
 		if (--i < 0) {
-			if (slave_can_tx(slave)) {
+			if (bond_slave_can_tx(slave)) {
 				bond_dev_queue_xmit(bond, skb, slave->dev);
 				return;
 			}
@@ -3567,7 +3605,7 @@ static void bond_xmit_slave_id(struct bonding *bond, struct sk_buff *skb, int sl
 	bond_for_each_slave_rcu(bond, slave, iter) {
 		if (--i < 0)
 			break;
-		if (slave_can_tx(slave)) {
+		if (bond_slave_can_tx(slave)) {
 			bond_dev_queue_xmit(bond, skb, slave->dev);
 			return;
 		}
@@ -3624,7 +3662,7 @@ static int bond_xmit_roundrobin(struct sk_buff *skb, struct net_device *bond_dev
 	 */
 	if (iph->protocol == IPPROTO_IGMP && skb->protocol == htons(ETH_P_IP)) {
 		slave = rcu_dereference(bond->curr_active_slave);
-		if (slave && slave_can_tx(slave))
+		if (slave && bond_slave_can_tx(slave))
 			bond_dev_queue_xmit(bond, skb, slave->dev);
 		else
 			bond_xmit_slave_id(bond, skb, 0);
@@ -3662,7 +3700,7 @@ static int bond_xmit_xor(struct sk_buff *skb, struct net_device *bond_dev)
 {
 	struct bonding *bond = netdev_priv(bond_dev);
 
-	bond_xmit_slave_id(bond, skb, bond_xmit_hash(bond, skb, bond->slave_cnt));
+	bond_xmit_slave_id(bond, skb, bond_xmit_hash(bond, skb) % bond->slave_cnt);
 
 	return NETDEV_TX_OK;
 }
@@ -3677,7 +3715,7 @@ static int bond_xmit_broadcast(struct sk_buff *skb, struct net_device *bond_dev)
 	bond_for_each_slave_rcu(bond, slave, iter) {
 		if (bond_is_last_slave(bond, slave))
 			break;
-		if (IS_UP(slave->dev) && slave->link == BOND_LINK_UP) {
+		if (bond_slave_is_up(slave) && slave->link == BOND_LINK_UP) {
 			struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
 
 			if (!skb2) {
@@ -3689,7 +3727,7 @@ static int bond_xmit_broadcast(struct sk_buff *skb, struct net_device *bond_dev)
 			bond_dev_queue_xmit(bond, skb2, slave->dev);
 		}
 	}
-	if (slave && IS_UP(slave->dev) && slave->link == BOND_LINK_UP)
+	if (slave && bond_slave_is_up(slave) && slave->link == BOND_LINK_UP)
 		bond_dev_queue_xmit(bond, skb, slave->dev);
 	else
 		dev_kfree_skb_any(skb);
@@ -3714,7 +3752,7 @@ static inline int bond_slave_override(struct bonding *bond,
 	/* Find out if any slaves have the same mapping as this skb. */
 	bond_for_each_slave_rcu(bond, slave, iter) {
 		if (slave->queue_id == skb->queue_mapping) {
-			if (slave_can_tx(slave)) {
+			if (bond_slave_can_tx(slave)) {
 				bond_dev_queue_xmit(bond, skb, slave->dev);
 				return 0;
 			}
@@ -3755,12 +3793,11 @@ static netdev_tx_t __bond_start_xmit(struct sk_buff *skb, struct net_device *dev
 {
 	struct bonding *bond = netdev_priv(dev);
 
-	if (TX_QUEUE_OVERRIDE(bond->params.mode)) {
-		if (!bond_slave_override(bond, skb))
-			return NETDEV_TX_OK;
-	}
+	if (bond_should_override_tx_queue(bond) &&
+	    !bond_slave_override(bond, skb))
+		return NETDEV_TX_OK;
 
-	switch (bond->params.mode) {
+	switch (BOND_MODE(bond)) {
 	case BOND_MODE_ROUNDROBIN:
 		return bond_xmit_roundrobin(skb, dev);
 	case BOND_MODE_ACTIVEBACKUP:
@@ -3772,12 +3809,13 @@ static netdev_tx_t __bond_start_xmit(struct sk_buff *skb, struct net_device *dev
 	case BOND_MODE_8023AD:
 		return bond_3ad_xmit_xor(skb, dev);
 	case BOND_MODE_ALB:
-	case BOND_MODE_TLB:
 		return bond_alb_xmit(skb, dev);
+	case BOND_MODE_TLB:
+		return bond_tlb_xmit(skb, dev);
 	default:
 		/* Should never happen, mode already checked */
 		pr_err("%s: Error: Unknown bonding mode %d\n",
-		       dev->name, bond->params.mode);
+		       dev->name, BOND_MODE(bond));
 		WARN_ON_ONCE(1);
 		dev_kfree_skb_any(skb);
 		return NETDEV_TX_OK;
@@ -3817,14 +3855,14 @@ static int bond_ethtool_get_settings(struct net_device *bond_dev,
 	ecmd->duplex = DUPLEX_UNKNOWN;
 	ecmd->port = PORT_OTHER;
 
-	/* Since SLAVE_IS_OK returns false for all inactive or down slaves, we
+	/* Since bond_slave_can_tx returns false for all inactive or down slaves, we
 	 * do not need to check mode.  Though link speed might not represent
 	 * the true receive or transmit bandwidth (not all modes are symmetric)
 	 * this is an accurate maximum.
 	 */
 	read_lock(&bond->lock);
 	bond_for_each_slave(bond, slave, iter) {
-		if (SLAVE_IS_OK(slave)) {
+		if (bond_slave_can_tx(slave)) {
 			if (slave->speed != SPEED_UNKNOWN)
 				speed += slave->speed;
 			if (ecmd->duplex == DUPLEX_UNKNOWN &&
@@ -3915,7 +3953,7 @@ void bond_setup(struct net_device *bond_dev)
 	/* Initialize the device options */
 	bond_dev->tx_queue_len = 0;
 	bond_dev->flags |= IFF_MASTER|IFF_MULTICAST;
-	bond_dev->priv_flags |= IFF_BONDING;
+	bond_dev->priv_flags |= IFF_BONDING | IFF_UNICAST_FLT;
 	bond_dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
 
 	/* At first, we block adding VLANs. That's the only way to
@@ -3945,6 +3983,7 @@ void bond_setup(struct net_device *bond_dev)
 				NETIF_F_HW_VLAN_CTAG_FILTER;
 
 	bond_dev->hw_features &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_HW_CSUM);
+	bond_dev->hw_features |= NETIF_F_GSO_UDP_TUNNEL;
 	bond_dev->features |= bond_dev->hw_features;
 }
 
@@ -3994,7 +4033,8 @@ static int bond_check_params(struct bond_params *params)
 
 	if (xmit_hash_policy) {
 		if ((bond_mode != BOND_MODE_XOR) &&
-		    (bond_mode != BOND_MODE_8023AD)) {
+		    (bond_mode != BOND_MODE_8023AD) &&
+		    (bond_mode != BOND_MODE_TLB)) {
 			pr_info("xmit_hash_policy param is irrelevant in mode %s\n",
 				bond_mode_name(bond_mode));
 		} else {
@@ -4079,7 +4119,7 @@ static int bond_check_params(struct bond_params *params)
 	}
 
 	/* reset values for 802.3ad/TLB/ALB */
-	if (BOND_NO_USES_ARP(bond_mode)) {
+	if (!bond_mode_uses_arp(bond_mode)) {
 		if (!miimon) {
 			pr_warn("Warning: miimon must be specified, otherwise bonding will not detect link failure, speed and duplex which are essential for 802.3ad operation\n");
 			pr_warn("Forcing miimon to 100msec\n");
@@ -4161,7 +4201,7 @@ static int bond_check_params(struct bond_params *params)
 		   catch mistakes */
 		__be32 ip;
 		if (!in4_pton(arp_ip_target[i], -1, (u8 *)&ip, -1, NULL) ||
-		    IS_IP_TARGET_UNUSABLE_ADDRESS(ip)) {
+		    !bond_is_ip_target_ok(ip)) {
 			pr_warn("Warning: bad arp_ip_target module parameter (%s), ARP monitoring will not be performed\n",
 				arp_ip_target[i]);
 			arp_interval = 0;
@@ -4234,7 +4274,7 @@ static int bond_check_params(struct bond_params *params)
 		pr_debug("Warning: either miimon or arp_interval and arp_ip_target module parameters must be specified, otherwise bonding will not detect link failures! see bonding.txt for details\n");
 	}
 
-	if (primary && !USES_PRIMARY(bond_mode)) {
+	if (primary && !bond_mode_uses_primary(bond_mode)) {
 		/* currently, using a primary only makes sense
 		 * in active backup, TLB or ALB modes
 		 */
@@ -4300,6 +4340,7 @@ static int bond_check_params(struct bond_params *params)
 	params->min_links = min_links;
 	params->lp_interval = lp_interval;
 	params->packets_per_slave = packets_per_slave;
+	params->tlb_dynamic_lb = 1; /* Default value */
 	if (packets_per_slave > 0) {
 		params->reciprocal_packets_per_slave =
 			reciprocal_value(packets_per_slave);
diff --git a/drivers/net/bonding/bond_netlink.c b/drivers/net/bonding/bond_netlink.c
index f847e165d252..5ab3c1847e67 100644
--- a/drivers/net/bonding/bond_netlink.c
+++ b/drivers/net/bonding/bond_netlink.c
@@ -56,10 +56,10 @@ static int bond_fill_slave_info(struct sk_buff *skb,
 	if (nla_put_u16(skb, IFLA_BOND_SLAVE_QUEUE_ID, slave->queue_id))
 		goto nla_put_failure;
 
-	if (slave->bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(slave->bond) == BOND_MODE_8023AD) {
 		const struct aggregator *agg;
 
-		agg = SLAVE_AD_INFO(slave).port.aggregator;
+		agg = SLAVE_AD_INFO(slave)->port.aggregator;
 		if (agg)
 			if (nla_put_u16(skb, IFLA_BOND_SLAVE_AD_AGGREGATOR_ID,
 					agg->aggregator_identifier))
@@ -407,7 +407,7 @@ static int bond_fill_info(struct sk_buff *skb,
 	unsigned int packets_per_slave;
 	int i, targets_added;
 
-	if (nla_put_u8(skb, IFLA_BOND_MODE, bond->params.mode))
+	if (nla_put_u8(skb, IFLA_BOND_MODE, BOND_MODE(bond)))
 		goto nla_put_failure;
 
 	if (slave_dev &&
@@ -505,7 +505,7 @@ static int bond_fill_info(struct sk_buff *skb,
 		       bond->params.ad_select))
 		goto nla_put_failure;
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info info;
 
 		if (!bond_3ad_get_active_agg_info(bond, &info)) {
diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index 832070298446..540e0167bf24 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -70,6 +70,8 @@ static int bond_option_mode_set(struct bonding *bond,
 				const struct bond_opt_value *newval);
 static int bond_option_slaves_set(struct bonding *bond,
 				  const struct bond_opt_value *newval);
+static int bond_option_tlb_dynamic_lb_set(struct bonding *bond,
+				  const struct bond_opt_value *newval);
 
 
 static const struct bond_opt_value bond_mode_tbl[] = {
@@ -180,6 +182,12 @@ static const struct bond_opt_value bond_lp_interval_tbl[] = {
 	{ NULL,      -1,      0},
 };
 
+static const struct bond_opt_value bond_tlb_dynamic_lb_tbl[] = {
+	{ "off", 0,  0},
+	{ "on",  1,  BOND_VALFLAG_DEFAULT},
+	{ NULL,  -1, 0}
+};
+
 static const struct bond_option bond_opts[] = {
 	[BOND_OPT_MODE] = {
 		.id = BOND_OPT_MODE,
@@ -200,7 +208,7 @@ static const struct bond_option bond_opts[] = {
 	[BOND_OPT_XMIT_HASH] = {
 		.id = BOND_OPT_XMIT_HASH,
 		.name = "xmit_hash_policy",
-		.desc = "balance-xor and 802.3ad hashing method",
+		.desc = "balance-xor, 802.3ad, and tlb hashing method",
 		.values = bond_xmit_hashtype_tbl,
 		.set = bond_option_xmit_hash_policy_set
 	},
@@ -365,9 +373,33 @@ static const struct bond_option bond_opts[] = {
 		.flags = BOND_OPTFLAG_RAWVAL,
 		.set = bond_option_slaves_set
 	},
+	[BOND_OPT_TLB_DYNAMIC_LB] = {
+		.id = BOND_OPT_TLB_DYNAMIC_LB,
+		.name = "tlb_dynamic_lb",
+		.desc = "Enable dynamic flow shuffling",
+		.unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_TLB)),
+		.values = bond_tlb_dynamic_lb_tbl,
+		.flags = BOND_OPTFLAG_IFDOWN,
+		.set = bond_option_tlb_dynamic_lb_set,
+	},
 	{ }
 };
 
+/* Searches for an option by name */
+const struct bond_option *bond_opt_get_by_name(const char *name)
+{
+	const struct bond_option *opt;
+	int option;
+
+	for (option = 0; option < BOND_OPT_LAST; option++) {
+		opt = bond_opt_get(option);
+		if (opt && !strcmp(opt->name, name))
+			return opt;
+	}
+
+	return NULL;
+}
+
 /* Searches for a value in opt's values[] table */
 const struct bond_opt_value *bond_opt_get_val(unsigned int option, u64 val)
 {
@@ -641,7 +673,7 @@ const struct bond_option *bond_opt_get(unsigned int option)
 
 int bond_option_mode_set(struct bonding *bond, const struct bond_opt_value *newval)
 {
-	if (BOND_NO_USES_ARP(newval->value) && bond->params.arp_interval) {
+	if (!bond_mode_uses_arp(newval->value) && bond->params.arp_interval) {
 		pr_info("%s: %s mode is incompatible with arp monitoring, start mii monitoring\n",
 			bond->dev->name, newval->string);
 		/* disable arp monitoring */
@@ -662,7 +694,7 @@ int bond_option_mode_set(struct bonding *bond, const struct bond_opt_value *newv
 static struct net_device *__bond_option_active_slave_get(struct bonding *bond,
 							 struct slave *slave)
 {
-	return USES_PRIMARY(bond->params.mode) && slave ? slave->dev : NULL;
+	return bond_uses_primary(bond) && slave ? slave->dev : NULL;
 }
 
 struct net_device *bond_option_active_slave_get_rcu(struct bonding *bond)
@@ -727,7 +759,7 @@ static int bond_option_active_slave_set(struct bonding *bond,
 				bond->dev->name, new_active->dev->name);
 		} else {
 			if (old_active && (new_active->link == BOND_LINK_UP) &&
-			    IS_UP(new_active->dev)) {
+			    bond_slave_is_up(new_active)) {
 				pr_info("%s: Setting %s as active slave\n",
 					bond->dev->name, new_active->dev->name);
 				bond_change_active_slave(bond, new_active);
@@ -746,6 +778,10 @@ static int bond_option_active_slave_set(struct bonding *bond,
 	return ret;
 }
 
+/* There are two tricky bits here.  First, if MII monitoring is activated, then
+ * we must disable ARP monitoring.  Second, if the timer isn't running, we must
+ * start it.
+ */
 static int bond_option_miimon_set(struct bonding *bond,
 				  const struct bond_opt_value *newval)
 {
@@ -784,6 +820,10 @@ static int bond_option_miimon_set(struct bonding *bond,
 	return 0;
 }
 
+/* Set up and down delays. These must be multiples of the
+ * MII monitoring value, and are stored internally as the multiplier.
+ * Thus, we must translate to MS for the real world.
+ */
 static int bond_option_updelay_set(struct bonding *bond,
 				   const struct bond_opt_value *newval)
 {
@@ -842,6 +882,10 @@ static int bond_option_use_carrier_set(struct bonding *bond,
 	return 0;
 }
 
+/* There are two tricky bits here.  First, if ARP monitoring is activated, then
+ * we must disable MII monitoring.  Second, if the ARP timer isn't running,
+ * we must start it.
+ */
 static int bond_option_arp_interval_set(struct bonding *bond,
 					const struct bond_opt_value *newval)
 {
@@ -899,7 +943,7 @@ static int _bond_option_arp_ip_target_add(struct bonding *bond, __be32 target)
 	__be32 *targets = bond->params.arp_targets;
 	int ind;
 
-	if (IS_IP_TARGET_UNUSABLE_ADDRESS(target)) {
+	if (!bond_is_ip_target_ok(target)) {
 		pr_err("%s: invalid ARP target %pI4 specified for addition\n",
 		       bond->dev->name, &target);
 		return -EINVAL;
@@ -944,7 +988,7 @@ static int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target)
 	unsigned long *targets_rx;
 	int ind, i;
 
-	if (IS_IP_TARGET_UNUSABLE_ADDRESS(target)) {
+	if (!bond_is_ip_target_ok(target)) {
 		pr_err("%s: invalid ARP target %pI4 specified for removal\n",
 		       bond->dev->name, &target);
 		return -EINVAL;
@@ -1338,3 +1382,13 @@ err_no_cmd:
 	ret = -EPERM;
 	goto out;
 }
+
+static int bond_option_tlb_dynamic_lb_set(struct bonding *bond,
+					  const struct bond_opt_value *newval)
+{
+	pr_info("%s: Setting dynamic-lb to %s (%llu)\n",
+		bond->dev->name, newval->string, newval->value);
+	bond->params.tlb_dynamic_lb = newval->value;
+
+	return 0;
+}
diff --git a/drivers/net/bonding/bond_options.h b/drivers/net/bonding/bond_options.h
index 12be9e1bfb0c..17ded5b29176 100644
--- a/drivers/net/bonding/bond_options.h
+++ b/drivers/net/bonding/bond_options.h
@@ -62,6 +62,7 @@ enum {
 	BOND_OPT_RESEND_IGMP,
 	BOND_OPT_LP_INTERVAL,
 	BOND_OPT_SLAVES,
+	BOND_OPT_TLB_DYNAMIC_LB,
 	BOND_OPT_LAST
 };
 
@@ -104,6 +105,7 @@ int bond_opt_tryset_rtnl(struct bonding *bond, unsigned int option, char *buf);
 const struct bond_opt_value *bond_opt_parse(const struct bond_option *opt,
 					    struct bond_opt_value *val);
 const struct bond_option *bond_opt_get(unsigned int option);
+const struct bond_option *bond_opt_get_by_name(const char *name);
 const struct bond_opt_value *bond_opt_get_val(unsigned int option, u64 val);
 
 /* This helper is used to initialize a bond_opt_value structure for parameter
diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c
index 013fdd0f45e9..b215b479bb3a 100644
--- a/drivers/net/bonding/bond_procfs.c
+++ b/drivers/net/bonding/bond_procfs.c
@@ -72,9 +72,9 @@ static void bond_info_show_master(struct seq_file *seq)
 	curr = rcu_dereference(bond->curr_active_slave);
 
 	seq_printf(seq, "Bonding Mode: %s",
-		   bond_mode_name(bond->params.mode));
+		   bond_mode_name(BOND_MODE(bond)));
 
-	if (bond->params.mode == BOND_MODE_ACTIVEBACKUP &&
+	if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP &&
 	    bond->params.fail_over_mac) {
 		optval = bond_opt_get_val(BOND_OPT_FAIL_OVER_MAC,
 					  bond->params.fail_over_mac);
@@ -83,15 +83,15 @@ static void bond_info_show_master(struct seq_file *seq)
 
 	seq_printf(seq, "\n");
 
-	if (bond->params.mode == BOND_MODE_XOR ||
-		bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_XOR ||
+		BOND_MODE(bond) == BOND_MODE_8023AD) {
 		optval = bond_opt_get_val(BOND_OPT_XMIT_HASH,
 					  bond->params.xmit_policy);
 		seq_printf(seq, "Transmit Hash Policy: %s (%d)\n",
 			   optval->string, bond->params.xmit_policy);
 	}
 
-	if (USES_PRIMARY(bond->params.mode)) {
+	if (bond_uses_primary(bond)) {
 		seq_printf(seq, "Primary Slave: %s",
 			   (bond->primary_slave) ?
 			   bond->primary_slave->dev->name : "None");
@@ -134,7 +134,7 @@ static void bond_info_show_master(struct seq_file *seq)
 		seq_printf(seq, "\n");
 	}
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info ad_info;
 
 		seq_puts(seq, "\n802.3ad info\n");
@@ -188,9 +188,9 @@ static void bond_info_show_slave(struct seq_file *seq,
 
 	seq_printf(seq, "Permanent HW addr: %pM\n", slave->perm_hwaddr);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		const struct aggregator *agg
-			= SLAVE_AD_INFO(slave).port.aggregator;
+			= SLAVE_AD_INFO(slave)->port.aggregator;
 
 		if (agg)
 			seq_printf(seq, "Aggregator ID: %d\n",
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index 5f6babcfc26e..daed52f68ce1 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -45,8 +45,7 @@
 #define to_dev(obj)	container_of(obj, struct device, kobj)
 #define to_bond(cd)	((struct bonding *)(netdev_priv(to_net_dev(cd))))
 
-/*
- * "show" function for the bond_masters attribute.
+/* "show" function for the bond_masters attribute.
  * The class parameter is ignored.
  */
 static ssize_t bonding_show_bonds(struct class *cls,
@@ -88,14 +87,12 @@ static struct net_device *bond_get_by_name(struct bond_net *bn, const char *ifna
 	return NULL;
 }
 
-/*
- * "store" function for the bond_masters attribute.  This is what
+/* "store" function for the bond_masters attribute.  This is what
  * creates and deletes entire bonds.
  *
  * The class parameter is ignored.
  *
  */
-
 static ssize_t bonding_store_bonds(struct class *cls,
 				   struct class_attribute *attr,
 				   const char *buffer, size_t count)
@@ -158,9 +155,26 @@ static const struct class_attribute class_attr_bonding_masters = {
 	.store = bonding_store_bonds,
 };
 
-/*
- * Show the slaves in the current bond.
- */
+/* Generic "store" method for bonding sysfs option setting */
+static ssize_t bonding_sysfs_store_option(struct device *d,
+					  struct device_attribute *attr,
+					  const char *buffer, size_t count)
+{
+	struct bonding *bond = to_bond(d);
+	const struct bond_option *opt;
+	int ret;
+
+	opt = bond_opt_get_by_name(attr->attr.name);
+	if (WARN_ON(!opt))
+		return -ENOENT;
+	ret = bond_opt_tryset_rtnl(bond, opt->id, (char *)buffer);
+	if (!ret)
+		ret = count;
+
+	return ret;
+}
+
+/* Show the slaves in the current bond. */
 static ssize_t bonding_show_slaves(struct device *d,
 				   struct device_attribute *attr, char *buf)
 {
@@ -190,62 +204,24 @@ static ssize_t bonding_show_slaves(struct device *d,
 
 	return res;
 }
-
-/*
- * Set the slaves in the current bond.
- * This is supposed to be only thin wrapper for bond_enslave and bond_release.
- * All hard work should be done there.
- */
-static ssize_t bonding_store_slaves(struct device *d,
-				    struct device_attribute *attr,
-				    const char *buffer, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_SLAVES, (char *)buffer);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(slaves, S_IRUGO | S_IWUSR, bonding_show_slaves,
-		   bonding_store_slaves);
+		   bonding_sysfs_store_option);
 
-/*
- * Show and set the bonding mode.  The bond interface must be down to
- * change the mode.
- */
+/* Show the bonding mode. */
 static ssize_t bonding_show_mode(struct device *d,
 				 struct device_attribute *attr, char *buf)
 {
 	struct bonding *bond = to_bond(d);
 	const struct bond_opt_value *val;
 
-	val = bond_opt_get_val(BOND_OPT_MODE, bond->params.mode);
+	val = bond_opt_get_val(BOND_OPT_MODE, BOND_MODE(bond));
 
-	return sprintf(buf, "%s %d\n", val->string, bond->params.mode);
-}
-
-static ssize_t bonding_store_mode(struct device *d,
-				  struct device_attribute *attr,
-				  const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_MODE, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
+	return sprintf(buf, "%s %d\n", val->string, BOND_MODE(bond));
 }
 static DEVICE_ATTR(mode, S_IRUGO | S_IWUSR,
-		   bonding_show_mode, bonding_store_mode);
+		   bonding_show_mode, bonding_sysfs_store_option);
 
-/*
- * Show and set the bonding transmit hash method.
- */
+/* Show the bonding transmit hash method. */
 static ssize_t bonding_show_xmit_hash(struct device *d,
 				      struct device_attribute *attr,
 				      char *buf)
@@ -257,26 +233,10 @@ static ssize_t bonding_show_xmit_hash(struct device *d,
 
 	return sprintf(buf, "%s %d\n", val->string, bond->params.xmit_policy);
 }
-
-static ssize_t bonding_store_xmit_hash(struct device *d,
-				       struct device_attribute *attr,
-				       const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_XMIT_HASH, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(xmit_hash_policy, S_IRUGO | S_IWUSR,
-		   bonding_show_xmit_hash, bonding_store_xmit_hash);
+		   bonding_show_xmit_hash, bonding_sysfs_store_option);
 
-/*
- * Show and set arp_validate.
- */
+/* Show arp_validate. */
 static ssize_t bonding_show_arp_validate(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
@@ -289,26 +249,10 @@ static ssize_t bonding_show_arp_validate(struct device *d,
 
 	return sprintf(buf, "%s %d\n", val->string, bond->params.arp_validate);
 }
-
-static ssize_t bonding_store_arp_validate(struct device *d,
-					  struct device_attribute *attr,
-					  const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_ARP_VALIDATE, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
-
 static DEVICE_ATTR(arp_validate, S_IRUGO | S_IWUSR, bonding_show_arp_validate,
-		   bonding_store_arp_validate);
-/*
- * Show and set arp_all_targets.
- */
+		   bonding_sysfs_store_option);
+
+/* Show arp_all_targets. */
 static ssize_t bonding_show_arp_all_targets(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
@@ -321,28 +265,10 @@ static ssize_t bonding_show_arp_all_targets(struct device *d,
 	return sprintf(buf, "%s %d\n",
 		       val->string, bond->params.arp_all_targets);
 }
-
-static ssize_t bonding_store_arp_all_targets(struct device *d,
-					  struct device_attribute *attr,
-					  const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_ARP_ALL_TARGETS, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
-
 static DEVICE_ATTR(arp_all_targets, S_IRUGO | S_IWUSR,
-		   bonding_show_arp_all_targets, bonding_store_arp_all_targets);
+		   bonding_show_arp_all_targets, bonding_sysfs_store_option);
 
-/*
- * Show and store fail_over_mac.  User only allowed to change the
- * value when there are no slaves.
- */
+/* Show fail_over_mac. */
 static ssize_t bonding_show_fail_over_mac(struct device *d,
 					  struct device_attribute *attr,
 					  char *buf)
@@ -355,30 +281,10 @@ static ssize_t bonding_show_fail_over_mac(struct device *d,
 
 	return sprintf(buf, "%s %d\n", val->string, bond->params.fail_over_mac);
 }
-
-static ssize_t bonding_store_fail_over_mac(struct device *d,
-					   struct device_attribute *attr,
-					   const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_FAIL_OVER_MAC, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
-
 static DEVICE_ATTR(fail_over_mac, S_IRUGO | S_IWUSR,
-		   bonding_show_fail_over_mac, bonding_store_fail_over_mac);
+		   bonding_show_fail_over_mac, bonding_sysfs_store_option);
 
-/*
- * Show and set the arp timer interval.  There are two tricky bits
- * here.  First, if ARP monitoring is activated, then we must disable
- * MII monitoring.  Second, if the ARP timer isn't running, we must
- * start it.
- */
+/* Show the arp timer interval. */
 static ssize_t bonding_show_arp_interval(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
@@ -387,26 +293,10 @@ static ssize_t bonding_show_arp_interval(struct device *d,
 
 	return sprintf(buf, "%d\n", bond->params.arp_interval);
 }
-
-static ssize_t bonding_store_arp_interval(struct device *d,
-					  struct device_attribute *attr,
-					  const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_ARP_INTERVAL, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(arp_interval, S_IRUGO | S_IWUSR,
-		   bonding_show_arp_interval, bonding_store_arp_interval);
+		   bonding_show_arp_interval, bonding_sysfs_store_option);
 
-/*
- * Show and set the arp targets.
- */
+/* Show the arp targets. */
 static ssize_t bonding_show_arp_targets(struct device *d,
 					struct device_attribute *attr,
 					char *buf)
@@ -424,27 +314,10 @@ static ssize_t bonding_show_arp_targets(struct device *d,
 
 	return res;
 }
+static DEVICE_ATTR(arp_ip_target, S_IRUGO | S_IWUSR,
+		   bonding_show_arp_targets, bonding_sysfs_store_option);
 
-static ssize_t bonding_store_arp_targets(struct device *d,
-					 struct device_attribute *attr,
-					 const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_ARP_TARGETS, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
-static DEVICE_ATTR(arp_ip_target, S_IRUGO | S_IWUSR , bonding_show_arp_targets, bonding_store_arp_targets);
-
-/*
- * Show and set the up and down delays.  These must be multiples of the
- * MII monitoring value, and are stored internally as the multiplier.
- * Thus, we must translate to MS for the real world.
- */
+/* Show the up and down delays. */
 static ssize_t bonding_show_downdelay(struct device *d,
 				      struct device_attribute *attr,
 				      char *buf)
@@ -453,22 +326,8 @@ static ssize_t bonding_show_downdelay(struct device *d,
 
 	return sprintf(buf, "%d\n", bond->params.downdelay * bond->params.miimon);
 }
-
-static ssize_t bonding_store_downdelay(struct device *d,
-				       struct device_attribute *attr,
-				       const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_DOWNDELAY, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(downdelay, S_IRUGO | S_IWUSR,
-		   bonding_show_downdelay, bonding_store_downdelay);
+		   bonding_show_downdelay, bonding_sysfs_store_option);
 
 static ssize_t bonding_show_updelay(struct device *d,
 				    struct device_attribute *attr,
@@ -479,27 +338,10 @@ static ssize_t bonding_show_updelay(struct device *d,
 	return sprintf(buf, "%d\n", bond->params.updelay * bond->params.miimon);
 
 }
-
-static ssize_t bonding_store_updelay(struct device *d,
-				     struct device_attribute *attr,
-				     const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_UPDELAY, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(updelay, S_IRUGO | S_IWUSR,
-		   bonding_show_updelay, bonding_store_updelay);
+		   bonding_show_updelay, bonding_sysfs_store_option);
 
-/*
- * Show and set the LACP interval.  Interface must be down, and the mode
- * must be set to 802.3ad mode.
- */
+/* Show the LACP interval. */
 static ssize_t bonding_show_lacp(struct device *d,
 				 struct device_attribute *attr,
 				 char *buf)
@@ -511,22 +353,8 @@ static ssize_t bonding_show_lacp(struct device *d,
 
 	return sprintf(buf, "%s %d\n", val->string, bond->params.lacp_fast);
 }
-
-static ssize_t bonding_store_lacp(struct device *d,
-				  struct device_attribute *attr,
-				  const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_LACP_RATE, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(lacp_rate, S_IRUGO | S_IWUSR,
-		   bonding_show_lacp, bonding_store_lacp);
+		   bonding_show_lacp, bonding_sysfs_store_option);
 
 static ssize_t bonding_show_min_links(struct device *d,
 				      struct device_attribute *attr,
@@ -536,22 +364,8 @@ static ssize_t bonding_show_min_links(struct device *d,
 
 	return sprintf(buf, "%u\n", bond->params.min_links);
 }
-
-static ssize_t bonding_store_min_links(struct device *d,
-				       struct device_attribute *attr,
-				       const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_MINLINKS, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(min_links, S_IRUGO | S_IWUSR,
-		   bonding_show_min_links, bonding_store_min_links);
+		   bonding_show_min_links, bonding_sysfs_store_option);
 
 static ssize_t bonding_show_ad_select(struct device *d,
 				      struct device_attribute *attr,
@@ -564,27 +378,10 @@ static ssize_t bonding_show_ad_select(struct device *d,
 
 	return sprintf(buf, "%s %d\n", val->string, bond->params.ad_select);
 }
-
-
-static ssize_t bonding_store_ad_select(struct device *d,
-				       struct device_attribute *attr,
-				       const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_AD_SELECT, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(ad_select, S_IRUGO | S_IWUSR,
-		   bonding_show_ad_select, bonding_store_ad_select);
+		   bonding_show_ad_select, bonding_sysfs_store_option);
 
-/*
- * Show and set the number of peer notifications to send after a failover event.
- */
+/* Show and set the number of peer notifications to send after a failover event. */
 static ssize_t bonding_show_num_peer_notif(struct device *d,
 					   struct device_attribute *attr,
 					   char *buf)
@@ -611,12 +408,7 @@ static DEVICE_ATTR(num_grat_arp, S_IRUGO | S_IWUSR,
 static DEVICE_ATTR(num_unsol_na, S_IRUGO | S_IWUSR,
 		   bonding_show_num_peer_notif, bonding_store_num_peer_notif);
 
-/*
- * Show and set the MII monitor interval.  There are two tricky bits
- * here.  First, if MII monitoring is activated, then we must disable
- * ARP monitoring.  Second, if the timer isn't running, we must
- * start it.
- */
+/* Show the MII monitor interval. */
 static ssize_t bonding_show_miimon(struct device *d,
 				   struct device_attribute *attr,
 				   char *buf)
@@ -625,30 +417,10 @@ static ssize_t bonding_show_miimon(struct device *d,
 
 	return sprintf(buf, "%d\n", bond->params.miimon);
 }
-
-static ssize_t bonding_store_miimon(struct device *d,
-				    struct device_attribute *attr,
-				    const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_MIIMON, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(miimon, S_IRUGO | S_IWUSR,
-		   bonding_show_miimon, bonding_store_miimon);
+		   bonding_show_miimon, bonding_sysfs_store_option);
 
-/*
- * Show and set the primary slave.  The store function is much
- * simpler than bonding_store_slaves function because it only needs to
- * handle one interface name.
- * The bond must be a mode that supports a primary for this be
- * set.
- */
+/* Show the primary slave. */
 static ssize_t bonding_show_primary(struct device *d,
 				    struct device_attribute *attr,
 				    char *buf)
@@ -661,26 +433,10 @@ static ssize_t bonding_show_primary(struct device *d,
 
 	return count;
 }
-
-static ssize_t bonding_store_primary(struct device *d,
-				     struct device_attribute *attr,
-				     const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_PRIMARY, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(primary, S_IRUGO | S_IWUSR,
-		   bonding_show_primary, bonding_store_primary);
+		   bonding_show_primary, bonding_sysfs_store_option);
 
-/*
- * Show and set the primary_reselect flag.
- */
+/* Show the primary_reselect flag. */
 static ssize_t bonding_show_primary_reselect(struct device *d,
 					     struct device_attribute *attr,
 					     char *buf)
@@ -694,28 +450,10 @@ static ssize_t bonding_show_primary_reselect(struct device *d,
 	return sprintf(buf, "%s %d\n",
 		       val->string, bond->params.primary_reselect);
 }
-
-static ssize_t bonding_store_primary_reselect(struct device *d,
-					      struct device_attribute *attr,
-					      const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_PRIMARY_RESELECT,
-				   (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(primary_reselect, S_IRUGO | S_IWUSR,
-		   bonding_show_primary_reselect,
-		   bonding_store_primary_reselect);
+		   bonding_show_primary_reselect, bonding_sysfs_store_option);
 
-/*
- * Show and set the use_carrier flag.
- */
+/* Show the use_carrier flag. */
 static ssize_t bonding_show_carrier(struct device *d,
 				    struct device_attribute *attr,
 				    char *buf)
@@ -724,27 +462,11 @@ static ssize_t bonding_show_carrier(struct device *d,
 
 	return sprintf(buf, "%d\n", bond->params.use_carrier);
 }
-
-static ssize_t bonding_store_carrier(struct device *d,
-				     struct device_attribute *attr,
-				     const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_USE_CARRIER, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(use_carrier, S_IRUGO | S_IWUSR,
-		   bonding_show_carrier, bonding_store_carrier);
+		   bonding_show_carrier, bonding_sysfs_store_option);
 
 
-/*
- * Show and set currently active_slave.
- */
+/* Show currently active_slave. */
 static ssize_t bonding_show_active_slave(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
@@ -761,27 +483,10 @@ static ssize_t bonding_show_active_slave(struct device *d,
 
 	return count;
 }
-
-static ssize_t bonding_store_active_slave(struct device *d,
-					  struct device_attribute *attr,
-					  const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_ACTIVE_SLAVE, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(active_slave, S_IRUGO | S_IWUSR,
-		   bonding_show_active_slave, bonding_store_active_slave);
-
+		   bonding_show_active_slave, bonding_sysfs_store_option);
 
-/*
- * Show link status of the bond interface.
- */
+/* Show link status of the bond interface. */
 static ssize_t bonding_show_mii_status(struct device *d,
 				       struct device_attribute *attr,
 				       char *buf)
@@ -792,9 +497,7 @@ static ssize_t bonding_show_mii_status(struct device *d,
 }
 static DEVICE_ATTR(mii_status, S_IRUGO, bonding_show_mii_status, NULL);
 
-/*
- * Show current 802.3ad aggregator ID.
- */
+/* Show current 802.3ad aggregator ID. */
 static ssize_t bonding_show_ad_aggregator(struct device *d,
 					  struct device_attribute *attr,
 					  char *buf)
@@ -802,7 +505,7 @@ static ssize_t bonding_show_ad_aggregator(struct device *d,
 	int count = 0;
 	struct bonding *bond = to_bond(d);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info ad_info;
 		count = sprintf(buf, "%d\n",
 				bond_3ad_get_active_agg_info(bond, &ad_info)
@@ -814,9 +517,7 @@ static ssize_t bonding_show_ad_aggregator(struct device *d,
 static DEVICE_ATTR(ad_aggregator, S_IRUGO, bonding_show_ad_aggregator, NULL);
 
 
-/*
- * Show number of active 802.3ad ports.
- */
+/* Show number of active 802.3ad ports. */
 static ssize_t bonding_show_ad_num_ports(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
@@ -824,7 +525,7 @@ static ssize_t bonding_show_ad_num_ports(struct device *d,
 	int count = 0;
 	struct bonding *bond = to_bond(d);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info ad_info;
 		count = sprintf(buf, "%d\n",
 				bond_3ad_get_active_agg_info(bond, &ad_info)
@@ -836,9 +537,7 @@ static ssize_t bonding_show_ad_num_ports(struct device *d,
 static DEVICE_ATTR(ad_num_ports, S_IRUGO, bonding_show_ad_num_ports, NULL);
 
 
-/*
- * Show current 802.3ad actor key.
- */
+/* Show current 802.3ad actor key. */
 static ssize_t bonding_show_ad_actor_key(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
@@ -846,7 +545,7 @@ static ssize_t bonding_show_ad_actor_key(struct device *d,
 	int count = 0;
 	struct bonding *bond = to_bond(d);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info ad_info;
 		count = sprintf(buf, "%d\n",
 				bond_3ad_get_active_agg_info(bond, &ad_info)
@@ -858,9 +557,7 @@ static ssize_t bonding_show_ad_actor_key(struct device *d,
 static DEVICE_ATTR(ad_actor_key, S_IRUGO, bonding_show_ad_actor_key, NULL);
 
 
-/*
- * Show current 802.3ad partner key.
- */
+/* Show current 802.3ad partner key. */
 static ssize_t bonding_show_ad_partner_key(struct device *d,
 					   struct device_attribute *attr,
 					   char *buf)
@@ -868,7 +565,7 @@ static ssize_t bonding_show_ad_partner_key(struct device *d,
 	int count = 0;
 	struct bonding *bond = to_bond(d);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info ad_info;
 		count = sprintf(buf, "%d\n",
 				bond_3ad_get_active_agg_info(bond, &ad_info)
@@ -880,9 +577,7 @@ static ssize_t bonding_show_ad_partner_key(struct device *d,
 static DEVICE_ATTR(ad_partner_key, S_IRUGO, bonding_show_ad_partner_key, NULL);
 
 
-/*
- * Show current 802.3ad partner mac.
- */
+/* Show current 802.3ad partner mac. */
 static ssize_t bonding_show_ad_partner_mac(struct device *d,
 					   struct device_attribute *attr,
 					   char *buf)
@@ -890,7 +585,7 @@ static ssize_t bonding_show_ad_partner_mac(struct device *d,
 	int count = 0;
 	struct bonding *bond = to_bond(d);
 
-	if (bond->params.mode == BOND_MODE_8023AD) {
+	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info ad_info;
 		if (!bond_3ad_get_active_agg_info(bond, &ad_info))
 			count = sprintf(buf, "%pM\n", ad_info.partner_system);
@@ -900,9 +595,7 @@ static ssize_t bonding_show_ad_partner_mac(struct device *d,
 }
 static DEVICE_ATTR(ad_partner_mac, S_IRUGO, bonding_show_ad_partner_mac, NULL);
 
-/*
- * Show the queue_ids of the slaves in the current bond.
- */
+/* Show the queue_ids of the slaves in the current bond. */
 static ssize_t bonding_show_queue_id(struct device *d,
 				     struct device_attribute *attr,
 				     char *buf)
@@ -933,31 +626,11 @@ static ssize_t bonding_show_queue_id(struct device *d,
 
 	return res;
 }
-
-/*
- * Set the queue_ids of the  slaves in the current bond.  The bond
- * interface must be enslaved for this to work.
- */
-static ssize_t bonding_store_queue_id(struct device *d,
-				      struct device_attribute *attr,
-				      const char *buffer, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_QUEUE_ID, (char *)buffer);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(queue_id, S_IRUGO | S_IWUSR, bonding_show_queue_id,
-		   bonding_store_queue_id);
+		   bonding_sysfs_store_option);
 
 
-/*
- * Show and set the all_slaves_active flag.
- */
+/* Show the all_slaves_active flag. */
 static ssize_t bonding_show_slaves_active(struct device *d,
 					  struct device_attribute *attr,
 					  char *buf)
@@ -966,27 +639,10 @@ static ssize_t bonding_show_slaves_active(struct device *d,
 
 	return sprintf(buf, "%d\n", bond->params.all_slaves_active);
 }
-
-static ssize_t bonding_store_slaves_active(struct device *d,
-					   struct device_attribute *attr,
-					   const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_ALL_SLAVES_ACTIVE,
-				   (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
 static DEVICE_ATTR(all_slaves_active, S_IRUGO | S_IWUSR,
-		   bonding_show_slaves_active, bonding_store_slaves_active);
+		   bonding_show_slaves_active, bonding_sysfs_store_option);
 
-/*
- * Show and set the number of IGMP membership reports to send on link failure
- */
+/* Show the number of IGMP membership reports to send on link failure */
 static ssize_t bonding_show_resend_igmp(struct device *d,
 					struct device_attribute *attr,
 					char *buf)
@@ -995,23 +651,8 @@ static ssize_t bonding_show_resend_igmp(struct device *d,
 
 	return sprintf(buf, "%d\n", bond->params.resend_igmp);
 }
-
-static ssize_t bonding_store_resend_igmp(struct device *d,
-					 struct device_attribute *attr,
-					 const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_RESEND_IGMP, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
-}
-
 static DEVICE_ATTR(resend_igmp, S_IRUGO | S_IWUSR,
-		   bonding_show_resend_igmp, bonding_store_resend_igmp);
+		   bonding_show_resend_igmp, bonding_sysfs_store_option);
 
 
 static ssize_t bonding_show_lp_interval(struct device *d,
@@ -1019,25 +660,21 @@ static ssize_t bonding_show_lp_interval(struct device *d,
 					char *buf)
 {
 	struct bonding *bond = to_bond(d);
+
 	return sprintf(buf, "%d\n", bond->params.lp_interval);
 }
+static DEVICE_ATTR(lp_interval, S_IRUGO | S_IWUSR,
+		   bonding_show_lp_interval, bonding_sysfs_store_option);
 
-static ssize_t bonding_store_lp_interval(struct device *d,
-					 struct device_attribute *attr,
-					 const char *buf, size_t count)
+static ssize_t bonding_show_tlb_dynamic_lb(struct device *d,
+					   struct device_attribute *attr,
+					   char *buf)
 {
 	struct bonding *bond = to_bond(d);
-	int ret;
-
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_LP_INTERVAL, (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
+	return sprintf(buf, "%d\n", bond->params.tlb_dynamic_lb);
 }
-
-static DEVICE_ATTR(lp_interval, S_IRUGO | S_IWUSR,
-		   bonding_show_lp_interval, bonding_store_lp_interval);
+static DEVICE_ATTR(tlb_dynamic_lb, S_IRUGO | S_IWUSR,
+		   bonding_show_tlb_dynamic_lb, bonding_sysfs_store_option);
 
 static ssize_t bonding_show_packets_per_slave(struct device *d,
 					      struct device_attribute *attr,
@@ -1045,27 +682,11 @@ static ssize_t bonding_show_packets_per_slave(struct device *d,
 {
 	struct bonding *bond = to_bond(d);
 	unsigned int packets_per_slave = bond->params.packets_per_slave;
-	return sprintf(buf, "%u\n", packets_per_slave);
-}
-
-static ssize_t bonding_store_packets_per_slave(struct device *d,
-					       struct device_attribute *attr,
-					       const char *buf, size_t count)
-{
-	struct bonding *bond = to_bond(d);
-	int ret;
 
-	ret = bond_opt_tryset_rtnl(bond, BOND_OPT_PACKETS_PER_SLAVE,
-				   (char *)buf);
-	if (!ret)
-		ret = count;
-
-	return ret;
+	return sprintf(buf, "%u\n", packets_per_slave);
 }
-
 static DEVICE_ATTR(packets_per_slave, S_IRUGO | S_IWUSR,
-		   bonding_show_packets_per_slave,
-		   bonding_store_packets_per_slave);
+		   bonding_show_packets_per_slave, bonding_sysfs_store_option);
 
 static struct attribute *per_bond_attrs[] = {
 	&dev_attr_slaves.attr,
@@ -1099,6 +720,7 @@ static struct attribute *per_bond_attrs[] = {
 	&dev_attr_min_links.attr,
 	&dev_attr_lp_interval.attr,
 	&dev_attr_packets_per_slave.attr,
+	&dev_attr_tlb_dynamic_lb.attr,
 	NULL,
 };
 
@@ -1107,8 +729,7 @@ static struct attribute_group bonding_group = {
 	.attrs = per_bond_attrs,
 };
 
-/*
- * Initialize sysfs.  This sets up the bonding_masters file in
+/* Initialize sysfs.  This sets up the bonding_masters file in
  * /sys/class/net.
  */
 int bond_create_sysfs(struct bond_net *bn)
@@ -1120,8 +741,7 @@ int bond_create_sysfs(struct bond_net *bn)
 
 	ret = netdev_class_create_file_ns(&bn->class_attr_bonding_masters,
 					  bn->net);
-	/*
-	 * Permit multiple loads of the module by ignoring failures to
+	/* Permit multiple loads of the module by ignoring failures to
 	 * create the bonding_masters sysfs file.  Bonding devices
 	 * created by second or subsequent loads of the module will
 	 * not be listed in, or controllable by, bonding_masters, but
@@ -1144,16 +764,13 @@ int bond_create_sysfs(struct bond_net *bn)
 
 }
 
-/*
- * Remove /sys/class/net/bonding_masters.
- */
+/* Remove /sys/class/net/bonding_masters. */
 void bond_destroy_sysfs(struct bond_net *bn)
 {
 	netdev_class_remove_file_ns(&bn->class_attr_bonding_masters, bn->net);
 }
 
-/*
- * Initialize sysfs for each bond.  This sets up and registers
+/* Initialize sysfs for each bond.  This sets up and registers
  * the 'bondctl' directory for each individual bond under /sys/class/net.
  */
 void bond_prepare_sysfs_group(struct bonding *bond)
diff --git a/drivers/net/bonding/bond_sysfs_slave.c b/drivers/net/bonding/bond_sysfs_slave.c
index 2e4eec5450c8..198677f58ce0 100644
--- a/drivers/net/bonding/bond_sysfs_slave.c
+++ b/drivers/net/bonding/bond_sysfs_slave.c
@@ -69,8 +69,8 @@ static ssize_t ad_aggregator_id_show(struct slave *slave, char *buf)
 {
 	const struct aggregator *agg;
 
-	if (slave->bond->params.mode == BOND_MODE_8023AD) {
-		agg = SLAVE_AD_INFO(slave).port.aggregator;
+	if (BOND_MODE(slave->bond) == BOND_MODE_8023AD) {
+		agg = SLAVE_AD_INFO(slave)->port.aggregator;
 		if (agg)
 			return sprintf(buf, "%d\n",
 				       agg->aggregator_identifier);
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 00bea320e3b5..0b4d9cde0b05 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -41,42 +41,6 @@
 
 #define BOND_DEFAULT_MIIMON	100
 
-#define IS_UP(dev)					   \
-	      ((((dev)->flags & IFF_UP) == IFF_UP)	&& \
-	       netif_running(dev)			&& \
-	       netif_carrier_ok(dev))
-
-/*
- * Checks whether slave is ready for transmit.
- */
-#define SLAVE_IS_OK(slave)			        \
-		    (((slave)->dev->flags & IFF_UP)  && \
-		     netif_running((slave)->dev)     && \
-		     ((slave)->link == BOND_LINK_UP) && \
-		     bond_is_active_slave(slave))
-
-
-#define USES_PRIMARY(mode)				\
-		(((mode) == BOND_MODE_ACTIVEBACKUP) ||	\
-		 ((mode) == BOND_MODE_TLB)          ||	\
-		 ((mode) == BOND_MODE_ALB))
-
-#define BOND_NO_USES_ARP(mode)				\
-		(((mode) == BOND_MODE_8023AD)	||	\
-		 ((mode) == BOND_MODE_TLB)	||	\
-		 ((mode) == BOND_MODE_ALB))
-
-#define TX_QUEUE_OVERRIDE(mode)				\
-			(((mode) == BOND_MODE_ACTIVEBACKUP) ||	\
-			 ((mode) == BOND_MODE_ROUNDROBIN))
-
-#define BOND_MODE_IS_LB(mode)			\
-		(((mode) == BOND_MODE_TLB) ||	\
-		 ((mode) == BOND_MODE_ALB))
-
-#define IS_IP_TARGET_UNUSABLE_ADDRESS(a)	\
-	((htonl(INADDR_BROADCAST) == a) ||	\
-	 ipv4_is_zeronet(a))
 /*
  * Less bad way to call ioctl from within the kernel; this needs to be
  * done some other way to get the call out of interrupt context.
@@ -90,6 +54,8 @@
 	set_fs(fs);			\
 	res; })
 
+#define BOND_MODE(bond) ((bond)->params.mode)
+
 /* slave list primitives */
 #define bond_slave_list(bond) (&(bond)->dev->adj_list.lower)
 
@@ -175,6 +141,7 @@ struct bond_params {
 	int resend_igmp;
 	int lp_interval;
 	int packets_per_slave;
+	int tlb_dynamic_lb;
 	struct reciprocal_value reciprocal_packets_per_slave;
 };
 
@@ -183,8 +150,6 @@ struct bond_parm_tbl {
 	int mode;
 };
 
-#define BOND_MAX_MODENAME_LEN 20
-
 struct slave {
 	struct net_device *dev; /* first - useful for panic debug */
 	struct bonding *bond; /* our master */
@@ -205,7 +170,7 @@ struct slave {
 	u32    speed;
 	u16    queue_id;
 	u8     perm_hwaddr[ETH_ALEN];
-	struct ad_slave_info ad_info; /* HUGE - better to dynamically alloc */
+	struct ad_slave_info *ad_info;
 	struct tlb_slave_info tlb_info;
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	struct netpoll *np;
@@ -285,14 +250,41 @@ static inline struct slave *bond_get_slave_by_dev(struct bonding *bond,
 
 static inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
 {
-	if (!slave || !slave->bond)
-		return NULL;
 	return slave->bond;
 }
 
+static inline bool bond_should_override_tx_queue(struct bonding *bond)
+{
+	return BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP ||
+	       BOND_MODE(bond) == BOND_MODE_ROUNDROBIN;
+}
+
 static inline bool bond_is_lb(const struct bonding *bond)
 {
-	return BOND_MODE_IS_LB(bond->params.mode);
+	return BOND_MODE(bond) == BOND_MODE_TLB ||
+	       BOND_MODE(bond) == BOND_MODE_ALB;
+}
+
+static inline bool bond_mode_uses_arp(int mode)
+{
+	return mode != BOND_MODE_8023AD && mode != BOND_MODE_TLB &&
+	       mode != BOND_MODE_ALB;
+}
+
+static inline bool bond_mode_uses_primary(int mode)
+{
+	return mode == BOND_MODE_ACTIVEBACKUP || mode == BOND_MODE_TLB ||
+	       mode == BOND_MODE_ALB;
+}
+
+static inline bool bond_uses_primary(struct bonding *bond)
+{
+	return bond_mode_uses_primary(BOND_MODE(bond));
+}
+
+static inline bool bond_slave_is_up(struct slave *slave)
+{
+	return netif_running(slave->dev) && netif_carrier_ok(slave->dev);
 }
 
 static inline void bond_set_active_slave(struct slave *slave)
@@ -365,6 +357,12 @@ static inline bool bond_is_active_slave(struct slave *slave)
 	return !bond_slave_state(slave);
 }
 
+static inline bool bond_slave_can_tx(struct slave *slave)
+{
+	return bond_slave_is_up(slave) && slave->link == BOND_LINK_UP &&
+	       bond_is_active_slave(slave);
+}
+
 #define BOND_PRI_RESELECT_ALWAYS	0
 #define BOND_PRI_RESELECT_BETTER	1
 #define BOND_PRI_RESELECT_FAILURE	2
@@ -396,12 +394,16 @@ static inline int slave_do_arp_validate(struct bonding *bond,
 	return bond->params.arp_validate & (1 << bond_slave_state(slave));
 }
 
-static inline int slave_do_arp_validate_only(struct bonding *bond,
-					     struct slave *slave)
+static inline int slave_do_arp_validate_only(struct bonding *bond)
 {
 	return bond->params.arp_validate & BOND_ARP_FILTER;
 }
 
+static inline int bond_is_ip_target_ok(__be32 addr)
+{
+	return !ipv4_is_lbcast(addr) && !ipv4_is_zeronet(addr);
+}
+
 /* Get the oldest arp which we've received on this slave for bond's
  * arp_targets.
  */
@@ -479,16 +481,14 @@ static inline __be32 bond_confirm_addr(struct net_device *dev, __be32 dst, __be3
 	return addr;
 }
 
-static inline bool slave_can_tx(struct slave *slave)
-{
-	if (IS_UP(slave->dev) && slave->link == BOND_LINK_UP &&
-	    bond_is_active_slave(slave))
-		return true;
-	else
-		return false;
-}
-
-struct bond_net;
+struct bond_net {
+	struct net		*net;	/* Associated network namespace */
+	struct list_head	dev_list;
+#ifdef CONFIG_PROC_FS
+	struct proc_dir_entry	*proc_dir;
+#endif
+	struct class_attribute	class_attr_bonding_masters;
+};
 
 int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, struct slave *slave);
 void bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_device *slave_dev);
@@ -500,7 +500,7 @@ int bond_sysfs_slave_add(struct slave *slave);
 void bond_sysfs_slave_del(struct slave *slave);
 int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev);
 int bond_release(struct net_device *bond_dev, struct net_device *slave_dev);
-int bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, int count);
+u32 bond_xmit_hash(struct bonding *bond, struct sk_buff *skb);
 void bond_select_active_slave(struct bonding *bond);
 void bond_change_active_slave(struct bonding *bond, struct slave *new_active);
 void bond_create_debugfs(void);
@@ -516,15 +516,9 @@ void bond_netlink_fini(void);
 struct net_device *bond_option_active_slave_get_rcu(struct bonding *bond);
 struct net_device *bond_option_active_slave_get(struct bonding *bond);
 const char *bond_slave_link_status(s8 link);
-
-struct bond_net {
-	struct net *		net;	/* Associated network namespace */
-	struct list_head	dev_list;
-#ifdef CONFIG_PROC_FS
-	struct proc_dir_entry *	proc_dir;
-#endif
-	struct class_attribute	class_attr_bonding_masters;
-};
+bool bond_verify_device_path(struct net_device *start_dev,
+			     struct net_device *end_dev,
+			     struct bond_vlan_tag *tags);
 
 #ifdef CONFIG_PROC_FS
 void bond_create_proc_entry(struct bonding *bond);
@@ -576,6 +570,27 @@ static inline struct slave *bond_slave_has_mac_rcu(struct bonding *bond,
 	return NULL;
 }
 
+/* Caller must hold rcu_read_lock() for read */
+static inline bool bond_slave_has_mac_rx(struct bonding *bond, const u8 *mac)
+{
+	struct list_head *iter;
+	struct slave *tmp;
+	struct netdev_hw_addr *ha;
+
+	bond_for_each_slave_rcu(bond, tmp, iter)
+		if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
+			return true;
+
+	if (netdev_uc_empty(bond->dev))
+		return false;
+
+	netdev_for_each_uc_addr(ha, bond->dev)
+		if (ether_addr_equal_64bits(mac, ha->addr))
+			return true;
+
+	return false;
+}
+
 /* Check if the ip is present in arp ip list, or first free slot if ip == 0
  * Returns -1 if not found, index if found
  */
diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index 9e7d95dae2c7..41688229c570 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -65,7 +65,7 @@ config CAN_LEDS
 
 config CAN_AT91
 	tristate "Atmel AT91 onchip CAN controller"
-	depends on ARM
+	depends on ARCH_AT91 || COMPILE_TEST
 	---help---
 	  This is a driver for the SoC CAN controller in Atmel's AT91SAM9263
 	  and AT91SAM9X5 processors.
@@ -77,12 +77,6 @@ config CAN_TI_HECC
 	  Driver for TI HECC (High End CAN Controller) module found on many
 	  TI devices. The device specifications are available from www.ti.com
 
-config CAN_MCP251X
-	tristate "Microchip MCP251x SPI CAN controllers"
-	depends on SPI && HAS_DMA
-	---help---
-	  Driver for the Microchip MCP251x SPI CAN controllers.
-
 config CAN_BFIN
 	depends on BF534 || BF536 || BF537 || BF538 || BF539 || BF54x
 	tristate "Analog Devices Blackfin on-chip CAN"
@@ -110,7 +104,7 @@ config CAN_FLEXCAN
 
 config PCH_CAN
 	tristate "Intel EG20T PCH CAN controller"
-	depends on PCI
+	depends on PCI && (X86_32 || COMPILE_TEST)
 	---help---
 	  This driver is for PCH CAN of Topcliff (Intel EG20T PCH) which
 	  is an IOH for x86 embedded processor (Intel Atom E6xx series).
@@ -125,6 +119,24 @@ config CAN_GRCAN
 	  endian syntheses of the cores would need some modifications on
 	  the hardware level to work.
 
+config CAN_RCAR
+	tristate "Renesas R-Car CAN controller"
+	depends on ARM
+	---help---
+	  Say Y here if you want to use CAN controller found on Renesas R-Car
+	  SoCs.
+
+	  To compile this driver as a module, choose M here: the module will
+	  be called rcar_can.
+
+config CAN_XILINXCAN
+	tristate "Xilinx CAN"
+	depends on ARCH_ZYNQ || MICROBLAZE || COMPILE_TEST
+	depends on COMMON_CLK && HAS_IOMEM
+	---help---
+	  Xilinx CAN driver. This driver supports both soft AXI CAN IP and
+	  Zynq CANPS IP.
+
 source "drivers/net/can/mscan/Kconfig"
 
 source "drivers/net/can/sja1000/Kconfig"
@@ -133,6 +145,8 @@ source "drivers/net/can/c_can/Kconfig"
 
 source "drivers/net/can/cc770/Kconfig"
 
+source "drivers/net/can/spi/Kconfig"
+
 source "drivers/net/can/usb/Kconfig"
 
 source "drivers/net/can/softing/Kconfig"
diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile
index c7440392adbb..1697f22353a9 100644
--- a/drivers/net/can/Makefile
+++ b/drivers/net/can/Makefile
@@ -10,6 +10,7 @@ can-dev-y			:= dev.o
 
 can-dev-$(CONFIG_CAN_LEDS)	+= led.o
 
+obj-y				+= spi/
 obj-y				+= usb/
 obj-y				+= softing/
 
@@ -19,11 +20,12 @@ obj-$(CONFIG_CAN_C_CAN)		+= c_can/
 obj-$(CONFIG_CAN_CC770)		+= cc770/
 obj-$(CONFIG_CAN_AT91)		+= at91_can.o
 obj-$(CONFIG_CAN_TI_HECC)	+= ti_hecc.o
-obj-$(CONFIG_CAN_MCP251X)	+= mcp251x.o
 obj-$(CONFIG_CAN_BFIN)		+= bfin_can.o
 obj-$(CONFIG_CAN_JANZ_ICAN3)	+= janz-ican3.o
 obj-$(CONFIG_CAN_FLEXCAN)	+= flexcan.o
 obj-$(CONFIG_PCH_CAN)		+= pch_can.o
 obj-$(CONFIG_CAN_GRCAN)		+= grcan.o
+obj-$(CONFIG_CAN_RCAR)		+= rcar_can.o
+obj-$(CONFIG_CAN_XILINXCAN)	+= xilinx_can.o
 
 ccflags-$(CONFIG_CAN_DEBUG_DEVICES) := -DDEBUG
diff --git a/drivers/net/can/c_can/c_can.c b/drivers/net/can/c_can/c_can.c
index 95e04e2002da..8e78bb48f5a4 100644
--- a/drivers/net/can/c_can/c_can.c
+++ b/drivers/net/can/c_can/c_can.c
@@ -252,8 +252,7 @@ static void c_can_obj_update(struct net_device *dev, int iface, u32 cmd, u32 obj
 	struct c_can_priv *priv = netdev_priv(dev);
 	int cnt, reg = C_CAN_IFACE(COMREQ_REG, iface);
 
-	priv->write_reg(priv, reg + 1, cmd);
-	priv->write_reg(priv, reg, obj);
+	priv->write_reg32(priv, reg, (cmd << 16) | obj);
 
 	for (cnt = MIN_TIMEOUT_VALUE; cnt; cnt--) {
 		if (!(priv->read_reg(priv, reg) & IF_COMR_BUSY))
@@ -328,8 +327,7 @@ static void c_can_setup_tx_object(struct net_device *dev, int iface,
 		change_bit(idx, &priv->tx_dir);
 	}
 
-	priv->write_reg(priv, C_CAN_IFACE(ARB1_REG, iface), arb);
-	priv->write_reg(priv, C_CAN_IFACE(ARB2_REG, iface), arb >> 16);
+	priv->write_reg32(priv, C_CAN_IFACE(ARB1_REG, iface), arb);
 
 	priv->write_reg(priv, C_CAN_IFACE(MSGCTRL_REG, iface), ctrl);
 
@@ -391,8 +389,7 @@ static int c_can_read_msg_object(struct net_device *dev, int iface, u32 ctrl)
 
 	frame->can_dlc = get_can_dlc(ctrl & 0x0F);
 
-	arb = priv->read_reg(priv, C_CAN_IFACE(ARB1_REG, iface));
-	arb |= priv->read_reg(priv, C_CAN_IFACE(ARB2_REG, iface)) << 16;
+	arb = priv->read_reg32(priv, C_CAN_IFACE(ARB1_REG, iface));
 
 	if (arb & IF_ARB_MSGXTD)
 		frame->can_id = (arb & CAN_EFF_MASK) | CAN_EFF_FLAG;
@@ -424,12 +421,10 @@ static void c_can_setup_receive_object(struct net_device *dev, int iface,
 	struct c_can_priv *priv = netdev_priv(dev);
 
 	mask |= BIT(29);
-	priv->write_reg(priv, C_CAN_IFACE(MASK1_REG, iface), mask);
-	priv->write_reg(priv, C_CAN_IFACE(MASK2_REG, iface), mask >> 16);
+	priv->write_reg32(priv, C_CAN_IFACE(MASK1_REG, iface), mask);
 
 	id |= IF_ARB_MSGVAL;
-	priv->write_reg(priv, C_CAN_IFACE(ARB1_REG, iface), id);
-	priv->write_reg(priv, C_CAN_IFACE(ARB2_REG, iface), id >> 16);
+	priv->write_reg32(priv, C_CAN_IFACE(ARB1_REG, iface), id);
 
 	priv->write_reg(priv, C_CAN_IFACE(MSGCTRL_REG, iface), mcont);
 	c_can_object_put(dev, iface, obj, IF_COMM_RCV_SETUP);
diff --git a/drivers/net/can/c_can/c_can.h b/drivers/net/can/c_can/c_can.h
index c56f1b1c11ca..99ad1aa576b0 100644
--- a/drivers/net/can/c_can/c_can.h
+++ b/drivers/net/can/c_can/c_can.h
@@ -78,6 +78,7 @@ enum reg {
 	C_CAN_INTPND2_REG,
 	C_CAN_MSGVAL1_REG,
 	C_CAN_MSGVAL2_REG,
+	C_CAN_FUNCTION_REG,
 };
 
 static const u16 reg_map_c_can[] = {
@@ -129,6 +130,7 @@ static const u16 reg_map_d_can[] = {
 	[C_CAN_BRPEXT_REG]	= 0x0E,
 	[C_CAN_INT_REG]		= 0x10,
 	[C_CAN_TEST_REG]	= 0x14,
+	[C_CAN_FUNCTION_REG]	= 0x18,
 	[C_CAN_TXRQST1_REG]	= 0x88,
 	[C_CAN_TXRQST2_REG]	= 0x8A,
 	[C_CAN_NEWDAT1_REG]	= 0x9C,
@@ -176,8 +178,10 @@ struct c_can_priv {
 	atomic_t tx_active;
 	unsigned long tx_dir;
 	int last_status;
-	u16 (*read_reg) (struct c_can_priv *priv, enum reg index);
-	void (*write_reg) (struct c_can_priv *priv, enum reg index, u16 val);
+	u16 (*read_reg) (const struct c_can_priv *priv, enum reg index);
+	void (*write_reg) (const struct c_can_priv *priv, enum reg index, u16 val);
+	u32 (*read_reg32) (const struct c_can_priv *priv, enum reg index);
+	void (*write_reg32) (const struct c_can_priv *priv, enum reg index, u32 val);
 	void __iomem *base;
 	const u16 *regs;
 	void *priv;		/* for board-specific data */
diff --git a/drivers/net/can/c_can/c_can_pci.c b/drivers/net/can/c_can/c_can_pci.c
index fe5f6303b584..5d11e0e4225b 100644
--- a/drivers/net/can/c_can/c_can_pci.c
+++ b/drivers/net/can/c_can/c_can_pci.c
@@ -19,9 +19,13 @@
 
 #include "c_can.h"
 
+#define PCI_DEVICE_ID_PCH_CAN	0x8818
+#define PCH_PCI_SOFT_RESET	0x01fc
+
 enum c_can_pci_reg_align {
 	C_CAN_REG_ALIGN_16,
 	C_CAN_REG_ALIGN_32,
+	C_CAN_REG_32,
 };
 
 struct c_can_pci_data {
@@ -31,6 +35,10 @@ struct c_can_pci_data {
 	enum c_can_pci_reg_align reg_align;
 	/* Set the frequency */
 	unsigned int freq;
+	/* PCI bar number */
+	int bar;
+	/* Callback for reset */
+	void (*init)(const struct c_can_priv *priv, bool enable);
 };
 
 /*
@@ -39,30 +47,70 @@ struct c_can_pci_data {
  * registers can be aligned to a 16-bit boundary or 32-bit boundary etc.
  * Handle the same by providing a common read/write interface.
  */
-static u16 c_can_pci_read_reg_aligned_to_16bit(struct c_can_priv *priv,
+static u16 c_can_pci_read_reg_aligned_to_16bit(const struct c_can_priv *priv,
 						enum reg index)
 {
 	return readw(priv->base + priv->regs[index]);
 }
 
-static void c_can_pci_write_reg_aligned_to_16bit(struct c_can_priv *priv,
+static void c_can_pci_write_reg_aligned_to_16bit(const struct c_can_priv *priv,
 						enum reg index, u16 val)
 {
 	writew(val, priv->base + priv->regs[index]);
 }
 
-static u16 c_can_pci_read_reg_aligned_to_32bit(struct c_can_priv *priv,
+static u16 c_can_pci_read_reg_aligned_to_32bit(const struct c_can_priv *priv,
 						enum reg index)
 {
 	return readw(priv->base + 2 * priv->regs[index]);
 }
 
-static void c_can_pci_write_reg_aligned_to_32bit(struct c_can_priv *priv,
+static void c_can_pci_write_reg_aligned_to_32bit(const struct c_can_priv *priv,
 						enum reg index, u16 val)
 {
 	writew(val, priv->base + 2 * priv->regs[index]);
 }
 
+static u16 c_can_pci_read_reg_32bit(const struct c_can_priv *priv,
+				    enum reg index)
+{
+	return (u16)ioread32(priv->base + 2 * priv->regs[index]);
+}
+
+static void c_can_pci_write_reg_32bit(const struct c_can_priv *priv,
+				      enum reg index, u16 val)
+{
+	iowrite32((u32)val, priv->base + 2 * priv->regs[index]);
+}
+
+static u32 c_can_pci_read_reg32(const struct c_can_priv *priv, enum reg index)
+{
+	u32 val;
+
+	val = priv->read_reg(priv, index);
+	val |= ((u32) priv->read_reg(priv, index + 1)) << 16;
+
+	return val;
+}
+
+static void c_can_pci_write_reg32(const struct c_can_priv *priv, enum reg index,
+		u32 val)
+{
+	priv->write_reg(priv, index + 1, val >> 16);
+	priv->write_reg(priv, index, val);
+}
+
+static void c_can_pci_reset_pch(const struct c_can_priv *priv, bool enable)
+{
+	if (enable) {
+		u32 __iomem *addr = priv->base + PCH_PCI_SOFT_RESET;
+
+		/* write to sw reset register */
+		iowrite32(1, addr);
+		iowrite32(0, addr);
+	}
+}
+
 static int c_can_pci_probe(struct pci_dev *pdev,
 			   const struct pci_device_id *ent)
 {
@@ -90,7 +138,8 @@ static int c_can_pci_probe(struct pci_dev *pdev,
 		pci_set_master(pdev);
 	}
 
-	addr = pci_iomap(pdev, 0, pci_resource_len(pdev, 0));
+	addr = pci_iomap(pdev, c_can_pci_data->bar,
+			 pci_resource_len(pdev, c_can_pci_data->bar));
 	if (!addr) {
 		dev_err(&pdev->dev,
 			"device has no PCI memory resources, "
@@ -147,10 +196,18 @@ static int c_can_pci_probe(struct pci_dev *pdev,
 		priv->read_reg = c_can_pci_read_reg_aligned_to_16bit;
 		priv->write_reg = c_can_pci_write_reg_aligned_to_16bit;
 		break;
+	case C_CAN_REG_32:
+		priv->read_reg = c_can_pci_read_reg_32bit;
+		priv->write_reg = c_can_pci_write_reg_32bit;
+		break;
 	default:
 		ret = -EINVAL;
 		goto out_free_c_can;
 	}
+	priv->read_reg32 = c_can_pci_read_reg32;
+	priv->write_reg32 = c_can_pci_write_reg32;
+
+	priv->raminit = c_can_pci_data->init;
 
 	ret = register_c_can_dev(dev);
 	if (ret) {
@@ -198,6 +255,15 @@ static struct c_can_pci_data c_can_sta2x11= {
 	.type = BOSCH_C_CAN,
 	.reg_align = C_CAN_REG_ALIGN_32,
 	.freq = 52000000, /* 52 Mhz */
+	.bar = 0,
+};
+
+static struct c_can_pci_data c_can_pch = {
+	.type = BOSCH_C_CAN,
+	.reg_align = C_CAN_REG_32,
+	.freq = 50000000, /* 50 MHz */
+	.init = c_can_pci_reset_pch,
+	.bar = 1,
 };
 
 #define C_CAN_ID(_vend, _dev, _driverdata) {		\
@@ -207,6 +273,8 @@ static struct c_can_pci_data c_can_sta2x11= {
 static DEFINE_PCI_DEVICE_TABLE(c_can_pci_tbl) = {
 	C_CAN_ID(PCI_VENDOR_ID_STMICRO, PCI_DEVICE_ID_STMICRO_CAN,
 		 c_can_sta2x11),
+	C_CAN_ID(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_PCH_CAN,
+		 c_can_pch),
 	{},
 };
 static struct pci_driver c_can_pci_driver = {
diff --git a/drivers/net/can/c_can/c_can_platform.c b/drivers/net/can/c_can/c_can_platform.c
index 1df0b322d1e4..824108cd9fd5 100644
--- a/drivers/net/can/c_can/c_can_platform.c
+++ b/drivers/net/can/c_can/c_can_platform.c
@@ -40,6 +40,7 @@
 #define CAN_RAMINIT_START_MASK(i)	(0x001 << (i))
 #define CAN_RAMINIT_DONE_MASK(i)	(0x100 << (i))
 #define CAN_RAMINIT_ALL_MASK(i)		(0x101 << (i))
+#define DCAN_RAM_INIT_BIT		(1 << 3)
 static DEFINE_SPINLOCK(raminit_lock);
 /*
  * 16-bit c_can registers can be arranged differently in the memory
@@ -47,31 +48,31 @@ static DEFINE_SPINLOCK(raminit_lock);
  * registers can be aligned to a 16-bit boundary or 32-bit boundary etc.
  * Handle the same by providing a common read/write interface.
  */
-static u16 c_can_plat_read_reg_aligned_to_16bit(struct c_can_priv *priv,
+static u16 c_can_plat_read_reg_aligned_to_16bit(const struct c_can_priv *priv,
 						enum reg index)
 {
 	return readw(priv->base + priv->regs[index]);
 }
 
-static void c_can_plat_write_reg_aligned_to_16bit(struct c_can_priv *priv,
+static void c_can_plat_write_reg_aligned_to_16bit(const struct c_can_priv *priv,
 						enum reg index, u16 val)
 {
 	writew(val, priv->base + priv->regs[index]);
 }
 
-static u16 c_can_plat_read_reg_aligned_to_32bit(struct c_can_priv *priv,
+static u16 c_can_plat_read_reg_aligned_to_32bit(const struct c_can_priv *priv,
 						enum reg index)
 {
 	return readw(priv->base + 2 * priv->regs[index]);
 }
 
-static void c_can_plat_write_reg_aligned_to_32bit(struct c_can_priv *priv,
+static void c_can_plat_write_reg_aligned_to_32bit(const struct c_can_priv *priv,
 						enum reg index, u16 val)
 {
 	writew(val, priv->base + 2 * priv->regs[index]);
 }
 
-static void c_can_hw_raminit_wait(const struct c_can_priv *priv, u32 mask,
+static void c_can_hw_raminit_wait_ti(const struct c_can_priv *priv, u32 mask,
 				  u32 val)
 {
 	/* We look only at the bits of our instance. */
@@ -80,7 +81,7 @@ static void c_can_hw_raminit_wait(const struct c_can_priv *priv, u32 mask,
 		udelay(1);
 }
 
-static void c_can_hw_raminit(const struct c_can_priv *priv, bool enable)
+static void c_can_hw_raminit_ti(const struct c_can_priv *priv, bool enable)
 {
 	u32 mask = CAN_RAMINIT_ALL_MASK(priv->instance);
 	u32 ctrl;
@@ -96,18 +97,68 @@ static void c_can_hw_raminit(const struct c_can_priv *priv, bool enable)
 	ctrl |= CAN_RAMINIT_DONE_MASK(priv->instance);
 	writel(ctrl, priv->raminit_ctrlreg);
 	ctrl &= ~CAN_RAMINIT_DONE_MASK(priv->instance);
-	c_can_hw_raminit_wait(priv, ctrl, mask);
+	c_can_hw_raminit_wait_ti(priv, ctrl, mask);
 
 	if (enable) {
 		/* Set start bit and wait for the done bit. */
 		ctrl |= CAN_RAMINIT_START_MASK(priv->instance);
 		writel(ctrl, priv->raminit_ctrlreg);
 		ctrl |= CAN_RAMINIT_DONE_MASK(priv->instance);
-		c_can_hw_raminit_wait(priv, ctrl, mask);
+		c_can_hw_raminit_wait_ti(priv, ctrl, mask);
 	}
 	spin_unlock(&raminit_lock);
 }
 
+static u32 c_can_plat_read_reg32(const struct c_can_priv *priv, enum reg index)
+{
+	u32 val;
+
+	val = priv->read_reg(priv, index);
+	val |= ((u32) priv->read_reg(priv, index + 1)) << 16;
+
+	return val;
+}
+
+static void c_can_plat_write_reg32(const struct c_can_priv *priv, enum reg index,
+		u32 val)
+{
+	priv->write_reg(priv, index + 1, val >> 16);
+	priv->write_reg(priv, index, val);
+}
+
+static u32 d_can_plat_read_reg32(const struct c_can_priv *priv, enum reg index)
+{
+	return readl(priv->base + priv->regs[index]);
+}
+
+static void d_can_plat_write_reg32(const struct c_can_priv *priv, enum reg index,
+		u32 val)
+{
+	writel(val, priv->base + priv->regs[index]);
+}
+
+static void c_can_hw_raminit_wait(const struct c_can_priv *priv, u32 mask)
+{
+	while (priv->read_reg32(priv, C_CAN_FUNCTION_REG) & mask)
+		udelay(1);
+}
+
+static void c_can_hw_raminit(const struct c_can_priv *priv, bool enable)
+{
+	u32 ctrl;
+
+	ctrl = priv->read_reg32(priv, C_CAN_FUNCTION_REG);
+	ctrl &= ~DCAN_RAM_INIT_BIT;
+	priv->write_reg32(priv, C_CAN_FUNCTION_REG, ctrl);
+	c_can_hw_raminit_wait(priv, ctrl);
+
+	if (enable) {
+		ctrl |= DCAN_RAM_INIT_BIT;
+		priv->write_reg32(priv, C_CAN_FUNCTION_REG, ctrl);
+		c_can_hw_raminit_wait(priv, ctrl);
+	}
+}
+
 static struct platform_device_id c_can_id_table[] = {
 	[BOSCH_C_CAN_PLATFORM] = {
 		.name = KBUILD_MODNAME,
@@ -201,11 +252,15 @@ static int c_can_plat_probe(struct platform_device *pdev)
 		case IORESOURCE_MEM_32BIT:
 			priv->read_reg = c_can_plat_read_reg_aligned_to_32bit;
 			priv->write_reg = c_can_plat_write_reg_aligned_to_32bit;
+			priv->read_reg32 = c_can_plat_read_reg32;
+			priv->write_reg32 = c_can_plat_write_reg32;
 			break;
 		case IORESOURCE_MEM_16BIT:
 		default:
 			priv->read_reg = c_can_plat_read_reg_aligned_to_16bit;
 			priv->write_reg = c_can_plat_write_reg_aligned_to_16bit;
+			priv->read_reg32 = c_can_plat_read_reg32;
+			priv->write_reg32 = c_can_plat_write_reg32;
 			break;
 		}
 		break;
@@ -214,6 +269,8 @@ static int c_can_plat_probe(struct platform_device *pdev)
 		priv->can.ctrlmode_supported |= CAN_CTRLMODE_3_SAMPLES;
 		priv->read_reg = c_can_plat_read_reg_aligned_to_16bit;
 		priv->write_reg = c_can_plat_write_reg_aligned_to_16bit;
+		priv->read_reg32 = d_can_plat_read_reg32;
+		priv->write_reg32 = d_can_plat_write_reg32;
 
 		if (pdev->dev.of_node)
 			priv->instance = of_alias_get_id(pdev->dev.of_node, "d_can");
@@ -221,11 +278,20 @@ static int c_can_plat_probe(struct platform_device *pdev)
 			priv->instance = pdev->id;
 
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+		/* Not all D_CAN modules have a separate register for the D_CAN
+		 * RAM initialization. Use default RAM init bit in D_CAN module
+		 * if not specified in DT.
+		 */
+		if (!res) {
+			priv->raminit = c_can_hw_raminit;
+			break;
+		}
+
 		priv->raminit_ctrlreg = devm_ioremap_resource(&pdev->dev, res);
 		if (IS_ERR(priv->raminit_ctrlreg) || priv->instance < 0)
 			dev_info(&pdev->dev, "control memory is not used for raminit\n");
 		else
-			priv->raminit = c_can_hw_raminit;
+			priv->raminit = c_can_hw_raminit_ti;
 		break;
 	default:
 		ret = -EINVAL;
diff --git a/drivers/net/can/mscan/Kconfig b/drivers/net/can/mscan/Kconfig
index f19be5269e7b..81c711719490 100644
--- a/drivers/net/can/mscan/Kconfig
+++ b/drivers/net/can/mscan/Kconfig
@@ -1,5 +1,5 @@
 config CAN_MSCAN
-	depends on PPC || M68K
+	depends on PPC
 	tristate "Support for Freescale MSCAN based chips"
 	---help---
 	  The Motorola Scalable Controller Area Network (MSCAN) definition
diff --git a/drivers/net/can/rcar_can.c b/drivers/net/can/rcar_can.c
new file mode 100644
index 000000000000..5268d216ecfa
--- /dev/null
+++ b/drivers/net/can/rcar_can.c
@@ -0,0 +1,876 @@
+/* Renesas R-Car CAN device driver
+ *
+ * Copyright (C) 2013 Cogent Embedded, Inc. <source@cogentembedded.com>
+ * Copyright (C) 2013 Renesas Solutions Corp.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/errno.h>
+#include <linux/netdevice.h>
+#include <linux/platform_device.h>
+#include <linux/can/led.h>
+#include <linux/can/dev.h>
+#include <linux/clk.h>
+#include <linux/can/platform/rcar_can.h>
+
+#define RCAR_CAN_DRV_NAME	"rcar_can"
+
+/* Mailbox configuration:
+ * mailbox 60 - 63 - Rx FIFO mailboxes
+ * mailbox 56 - 59 - Tx FIFO mailboxes
+ * non-FIFO mailboxes are not used
+ */
+#define RCAR_CAN_N_MBX		64 /* Number of mailboxes in non-FIFO mode */
+#define RCAR_CAN_RX_FIFO_MBX	60 /* Mailbox - window to Rx FIFO */
+#define RCAR_CAN_TX_FIFO_MBX	56 /* Mailbox - window to Tx FIFO */
+#define RCAR_CAN_FIFO_DEPTH	4
+
+/* Mailbox registers structure */
+struct rcar_can_mbox_regs {
+	u32 id;		/* IDE and RTR bits, SID and EID */
+	u8 stub;	/* Not used */
+	u8 dlc;		/* Data Length Code - bits [0..3] */
+	u8 data[8];	/* Data Bytes */
+	u8 tsh;		/* Time Stamp Higher Byte */
+	u8 tsl;		/* Time Stamp Lower Byte */
+};
+
+struct rcar_can_regs {
+	struct rcar_can_mbox_regs mb[RCAR_CAN_N_MBX]; /* Mailbox registers */
+	u32 mkr_2_9[8];	/* Mask Registers 2-9 */
+	u32 fidcr[2];	/* FIFO Received ID Compare Register */
+	u32 mkivlr1;	/* Mask Invalid Register 1 */
+	u32 mier1;	/* Mailbox Interrupt Enable Register 1 */
+	u32 mkr_0_1[2];	/* Mask Registers 0-1 */
+	u32 mkivlr0;    /* Mask Invalid Register 0*/
+	u32 mier0;      /* Mailbox Interrupt Enable Register 0 */
+	u8 pad_440[0x3c0];
+	u8 mctl[64];	/* Message Control Registers */
+	u16 ctlr;	/* Control Register */
+	u16 str;	/* Status register */
+	u8 bcr[3];	/* Bit Configuration Register */
+	u8 clkr;	/* Clock Select Register */
+	u8 rfcr;	/* Receive FIFO Control Register */
+	u8 rfpcr;	/* Receive FIFO Pointer Control Register */
+	u8 tfcr;	/* Transmit FIFO Control Register */
+	u8 tfpcr;       /* Transmit FIFO Pointer Control Register */
+	u8 eier;	/* Error Interrupt Enable Register */
+	u8 eifr;	/* Error Interrupt Factor Judge Register */
+	u8 recr;	/* Receive Error Count Register */
+	u8 tecr;        /* Transmit Error Count Register */
+	u8 ecsr;	/* Error Code Store Register */
+	u8 cssr;	/* Channel Search Support Register */
+	u8 mssr;	/* Mailbox Search Status Register */
+	u8 msmr;	/* Mailbox Search Mode Register */
+	u16 tsr;	/* Time Stamp Register */
+	u8 afsr;	/* Acceptance Filter Support Register */
+	u8 pad_857;
+	u8 tcr;		/* Test Control Register */
+	u8 pad_859[7];
+	u8 ier;		/* Interrupt Enable Register */
+	u8 isr;		/* Interrupt Status Register */
+	u8 pad_862;
+	u8 mbsmr;	/* Mailbox Search Mask Register */
+};
+
+struct rcar_can_priv {
+	struct can_priv can;	/* Must be the first member! */
+	struct net_device *ndev;
+	struct napi_struct napi;
+	struct rcar_can_regs __iomem *regs;
+	struct clk *clk;
+	u8 tx_dlc[RCAR_CAN_FIFO_DEPTH];
+	u32 tx_head;
+	u32 tx_tail;
+	u8 clock_select;
+	u8 ier;
+};
+
+static const struct can_bittiming_const rcar_can_bittiming_const = {
+	.name = RCAR_CAN_DRV_NAME,
+	.tseg1_min = 4,
+	.tseg1_max = 16,
+	.tseg2_min = 2,
+	.tseg2_max = 8,
+	.sjw_max = 4,
+	.brp_min = 1,
+	.brp_max = 1024,
+	.brp_inc = 1,
+};
+
+/* Control Register bits */
+#define RCAR_CAN_CTLR_BOM	(3 << 11) /* Bus-Off Recovery Mode Bits */
+#define RCAR_CAN_CTLR_BOM_ENT	(1 << 11) /* Entry to halt mode */
+					/* at bus-off entry */
+#define RCAR_CAN_CTLR_SLPM	(1 << 10)
+#define RCAR_CAN_CTLR_CANM	(3 << 8) /* Operating Mode Select Bit */
+#define RCAR_CAN_CTLR_CANM_HALT	(1 << 9)
+#define RCAR_CAN_CTLR_CANM_RESET (1 << 8)
+#define RCAR_CAN_CTLR_CANM_FORCE_RESET (3 << 8)
+#define RCAR_CAN_CTLR_MLM	(1 << 3) /* Message Lost Mode Select */
+#define RCAR_CAN_CTLR_IDFM	(3 << 1) /* ID Format Mode Select Bits */
+#define RCAR_CAN_CTLR_IDFM_MIXED (1 << 2) /* Mixed ID mode */
+#define RCAR_CAN_CTLR_MBM	(1 << 0) /* Mailbox Mode select */
+
+/* Status Register bits */
+#define RCAR_CAN_STR_RSTST	(1 << 8) /* Reset Status Bit */
+
+/* FIFO Received ID Compare Registers 0 and 1 bits */
+#define RCAR_CAN_FIDCR_IDE	(1 << 31) /* ID Extension Bit */
+#define RCAR_CAN_FIDCR_RTR	(1 << 30) /* Remote Transmission Request Bit */
+
+/* Receive FIFO Control Register bits */
+#define RCAR_CAN_RFCR_RFEST	(1 << 7) /* Receive FIFO Empty Status Flag */
+#define RCAR_CAN_RFCR_RFE	(1 << 0) /* Receive FIFO Enable */
+
+/* Transmit FIFO Control Register bits */
+#define RCAR_CAN_TFCR_TFUST	(7 << 1) /* Transmit FIFO Unsent Message */
+					/* Number Status Bits */
+#define RCAR_CAN_TFCR_TFUST_SHIFT 1	/* Offset of Transmit FIFO Unsent */
+					/* Message Number Status Bits */
+#define RCAR_CAN_TFCR_TFE	(1 << 0) /* Transmit FIFO Enable */
+
+#define RCAR_CAN_N_RX_MKREGS1	2	/* Number of mask registers */
+					/* for Rx mailboxes 0-31 */
+#define RCAR_CAN_N_RX_MKREGS2	8
+
+/* Bit Configuration Register settings */
+#define RCAR_CAN_BCR_TSEG1(x)	(((x) & 0x0f) << 20)
+#define RCAR_CAN_BCR_BPR(x)	(((x) & 0x3ff) << 8)
+#define RCAR_CAN_BCR_SJW(x)	(((x) & 0x3) << 4)
+#define RCAR_CAN_BCR_TSEG2(x)	((x) & 0x07)
+
+/* Mailbox and Mask Registers bits */
+#define RCAR_CAN_IDE		(1 << 31)
+#define RCAR_CAN_RTR		(1 << 30)
+#define RCAR_CAN_SID_SHIFT	18
+
+/* Mailbox Interrupt Enable Register 1 bits */
+#define RCAR_CAN_MIER1_RXFIE	(1 << 28) /* Receive  FIFO Interrupt Enable */
+#define RCAR_CAN_MIER1_TXFIE	(1 << 24) /* Transmit FIFO Interrupt Enable */
+
+/* Interrupt Enable Register bits */
+#define RCAR_CAN_IER_ERSIE	(1 << 5) /* Error (ERS) Interrupt Enable Bit */
+#define RCAR_CAN_IER_RXFIE	(1 << 4) /* Reception FIFO Interrupt */
+					/* Enable Bit */
+#define RCAR_CAN_IER_TXFIE	(1 << 3) /* Transmission FIFO Interrupt */
+					/* Enable Bit */
+/* Interrupt Status Register bits */
+#define RCAR_CAN_ISR_ERSF	(1 << 5) /* Error (ERS) Interrupt Status Bit */
+#define RCAR_CAN_ISR_RXFF	(1 << 4) /* Reception FIFO Interrupt */
+					/* Status Bit */
+#define RCAR_CAN_ISR_TXFF	(1 << 3) /* Transmission FIFO Interrupt */
+					/* Status Bit */
+
+/* Error Interrupt Enable Register bits */
+#define RCAR_CAN_EIER_BLIE	(1 << 7) /* Bus Lock Interrupt Enable */
+#define RCAR_CAN_EIER_OLIE	(1 << 6) /* Overload Frame Transmit */
+					/* Interrupt Enable */
+#define RCAR_CAN_EIER_ORIE	(1 << 5) /* Receive Overrun  Interrupt Enable */
+#define RCAR_CAN_EIER_BORIE	(1 << 4) /* Bus-Off Recovery Interrupt Enable */
+#define RCAR_CAN_EIER_BOEIE	(1 << 3) /* Bus-Off Entry Interrupt Enable */
+#define RCAR_CAN_EIER_EPIE	(1 << 2) /* Error Passive Interrupt Enable */
+#define RCAR_CAN_EIER_EWIE	(1 << 1) /* Error Warning Interrupt Enable */
+#define RCAR_CAN_EIER_BEIE	(1 << 0) /* Bus Error Interrupt Enable */
+
+/* Error Interrupt Factor Judge Register bits */
+#define RCAR_CAN_EIFR_BLIF	(1 << 7) /* Bus Lock Detect Flag */
+#define RCAR_CAN_EIFR_OLIF	(1 << 6) /* Overload Frame Transmission */
+					 /* Detect Flag */
+#define RCAR_CAN_EIFR_ORIF	(1 << 5) /* Receive Overrun Detect Flag */
+#define RCAR_CAN_EIFR_BORIF	(1 << 4) /* Bus-Off Recovery Detect Flag */
+#define RCAR_CAN_EIFR_BOEIF	(1 << 3) /* Bus-Off Entry Detect Flag */
+#define RCAR_CAN_EIFR_EPIF	(1 << 2) /* Error Passive Detect Flag */
+#define RCAR_CAN_EIFR_EWIF	(1 << 1) /* Error Warning Detect Flag */
+#define RCAR_CAN_EIFR_BEIF	(1 << 0) /* Bus Error Detect Flag */
+
+/* Error Code Store Register bits */
+#define RCAR_CAN_ECSR_EDPM	(1 << 7) /* Error Display Mode Select Bit */
+#define RCAR_CAN_ECSR_ADEF	(1 << 6) /* ACK Delimiter Error Flag */
+#define RCAR_CAN_ECSR_BE0F	(1 << 5) /* Bit Error (dominant) Flag */
+#define RCAR_CAN_ECSR_BE1F	(1 << 4) /* Bit Error (recessive) Flag */
+#define RCAR_CAN_ECSR_CEF	(1 << 3) /* CRC Error Flag */
+#define RCAR_CAN_ECSR_AEF	(1 << 2) /* ACK Error Flag */
+#define RCAR_CAN_ECSR_FEF	(1 << 1) /* Form Error Flag */
+#define RCAR_CAN_ECSR_SEF	(1 << 0) /* Stuff Error Flag */
+
+#define RCAR_CAN_NAPI_WEIGHT	4
+#define MAX_STR_READS		0x100
+
+static void tx_failure_cleanup(struct net_device *ndev)
+{
+	int i;
+
+	for (i = 0; i < RCAR_CAN_FIFO_DEPTH; i++)
+		can_free_echo_skb(ndev, i);
+}
+
+static void rcar_can_error(struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+	u8 eifr, txerr = 0, rxerr = 0;
+
+	/* Propagate the error condition to the CAN stack */
+	skb = alloc_can_err_skb(ndev, &cf);
+
+	eifr = readb(&priv->regs->eifr);
+	if (eifr & (RCAR_CAN_EIFR_EWIF | RCAR_CAN_EIFR_EPIF)) {
+		txerr = readb(&priv->regs->tecr);
+		rxerr = readb(&priv->regs->recr);
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[6] = txerr;
+			cf->data[7] = rxerr;
+		}
+	}
+	if (eifr & RCAR_CAN_EIFR_BEIF) {
+		int rx_errors = 0, tx_errors = 0;
+		u8 ecsr;
+
+		netdev_dbg(priv->ndev, "Bus error interrupt:\n");
+		if (skb) {
+			cf->can_id |= CAN_ERR_BUSERROR | CAN_ERR_PROT;
+			cf->data[2] = CAN_ERR_PROT_UNSPEC;
+		}
+		ecsr = readb(&priv->regs->ecsr);
+		if (ecsr & RCAR_CAN_ECSR_ADEF) {
+			netdev_dbg(priv->ndev, "ACK Delimiter Error\n");
+			tx_errors++;
+			writeb(~RCAR_CAN_ECSR_ADEF, &priv->regs->ecsr);
+			if (skb)
+				cf->data[3] |= CAN_ERR_PROT_LOC_ACK_DEL;
+		}
+		if (ecsr & RCAR_CAN_ECSR_BE0F) {
+			netdev_dbg(priv->ndev, "Bit Error (dominant)\n");
+			tx_errors++;
+			writeb(~RCAR_CAN_ECSR_BE0F, &priv->regs->ecsr);
+			if (skb)
+				cf->data[2] |= CAN_ERR_PROT_BIT0;
+		}
+		if (ecsr & RCAR_CAN_ECSR_BE1F) {
+			netdev_dbg(priv->ndev, "Bit Error (recessive)\n");
+			tx_errors++;
+			writeb(~RCAR_CAN_ECSR_BE1F, &priv->regs->ecsr);
+			if (skb)
+				cf->data[2] |= CAN_ERR_PROT_BIT1;
+		}
+		if (ecsr & RCAR_CAN_ECSR_CEF) {
+			netdev_dbg(priv->ndev, "CRC Error\n");
+			rx_errors++;
+			writeb(~RCAR_CAN_ECSR_CEF, &priv->regs->ecsr);
+			if (skb)
+				cf->data[3] |= CAN_ERR_PROT_LOC_CRC_SEQ;
+		}
+		if (ecsr & RCAR_CAN_ECSR_AEF) {
+			netdev_dbg(priv->ndev, "ACK Error\n");
+			tx_errors++;
+			writeb(~RCAR_CAN_ECSR_AEF, &priv->regs->ecsr);
+			if (skb) {
+				cf->can_id |= CAN_ERR_ACK;
+				cf->data[3] |= CAN_ERR_PROT_LOC_ACK;
+			}
+		}
+		if (ecsr & RCAR_CAN_ECSR_FEF) {
+			netdev_dbg(priv->ndev, "Form Error\n");
+			rx_errors++;
+			writeb(~RCAR_CAN_ECSR_FEF, &priv->regs->ecsr);
+			if (skb)
+				cf->data[2] |= CAN_ERR_PROT_FORM;
+		}
+		if (ecsr & RCAR_CAN_ECSR_SEF) {
+			netdev_dbg(priv->ndev, "Stuff Error\n");
+			rx_errors++;
+			writeb(~RCAR_CAN_ECSR_SEF, &priv->regs->ecsr);
+			if (skb)
+				cf->data[2] |= CAN_ERR_PROT_STUFF;
+		}
+
+		priv->can.can_stats.bus_error++;
+		ndev->stats.rx_errors += rx_errors;
+		ndev->stats.tx_errors += tx_errors;
+		writeb(~RCAR_CAN_EIFR_BEIF, &priv->regs->eifr);
+	}
+	if (eifr & RCAR_CAN_EIFR_EWIF) {
+		netdev_dbg(priv->ndev, "Error warning interrupt\n");
+		priv->can.state = CAN_STATE_ERROR_WARNING;
+		priv->can.can_stats.error_warning++;
+		/* Clear interrupt condition */
+		writeb(~RCAR_CAN_EIFR_EWIF, &priv->regs->eifr);
+		if (skb)
+			cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_WARNING :
+					      CAN_ERR_CRTL_RX_WARNING;
+	}
+	if (eifr & RCAR_CAN_EIFR_EPIF) {
+		netdev_dbg(priv->ndev, "Error passive interrupt\n");
+		priv->can.state = CAN_STATE_ERROR_PASSIVE;
+		priv->can.can_stats.error_passive++;
+		/* Clear interrupt condition */
+		writeb(~RCAR_CAN_EIFR_EPIF, &priv->regs->eifr);
+		if (skb)
+			cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_PASSIVE :
+					      CAN_ERR_CRTL_RX_PASSIVE;
+	}
+	if (eifr & RCAR_CAN_EIFR_BOEIF) {
+		netdev_dbg(priv->ndev, "Bus-off entry interrupt\n");
+		tx_failure_cleanup(ndev);
+		priv->ier = RCAR_CAN_IER_ERSIE;
+		writeb(priv->ier, &priv->regs->ier);
+		priv->can.state = CAN_STATE_BUS_OFF;
+		/* Clear interrupt condition */
+		writeb(~RCAR_CAN_EIFR_BOEIF, &priv->regs->eifr);
+		can_bus_off(ndev);
+		if (skb)
+			cf->can_id |= CAN_ERR_BUSOFF;
+	}
+	if (eifr & RCAR_CAN_EIFR_ORIF) {
+		netdev_dbg(priv->ndev, "Receive overrun error interrupt\n");
+		ndev->stats.rx_over_errors++;
+		ndev->stats.rx_errors++;
+		writeb(~RCAR_CAN_EIFR_ORIF, &priv->regs->eifr);
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+		}
+	}
+	if (eifr & RCAR_CAN_EIFR_OLIF) {
+		netdev_dbg(priv->ndev,
+			   "Overload Frame Transmission error interrupt\n");
+		ndev->stats.rx_over_errors++;
+		ndev->stats.rx_errors++;
+		writeb(~RCAR_CAN_EIFR_OLIF, &priv->regs->eifr);
+		if (skb) {
+			cf->can_id |= CAN_ERR_PROT;
+			cf->data[2] |= CAN_ERR_PROT_OVERLOAD;
+		}
+	}
+
+	if (skb) {
+		stats->rx_packets++;
+		stats->rx_bytes += cf->can_dlc;
+		netif_rx(skb);
+	}
+}
+
+static void rcar_can_tx_done(struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	u8 isr;
+
+	while (1) {
+		u8 unsent = readb(&priv->regs->tfcr);
+
+		unsent = (unsent & RCAR_CAN_TFCR_TFUST) >>
+			  RCAR_CAN_TFCR_TFUST_SHIFT;
+		if (priv->tx_head - priv->tx_tail <= unsent)
+			break;
+		stats->tx_packets++;
+		stats->tx_bytes += priv->tx_dlc[priv->tx_tail %
+						RCAR_CAN_FIFO_DEPTH];
+		priv->tx_dlc[priv->tx_tail % RCAR_CAN_FIFO_DEPTH] = 0;
+		can_get_echo_skb(ndev, priv->tx_tail % RCAR_CAN_FIFO_DEPTH);
+		priv->tx_tail++;
+		netif_wake_queue(ndev);
+	}
+	/* Clear interrupt */
+	isr = readb(&priv->regs->isr);
+	writeb(isr & ~RCAR_CAN_ISR_TXFF, &priv->regs->isr);
+	can_led_event(ndev, CAN_LED_EVENT_TX);
+}
+
+static irqreturn_t rcar_can_interrupt(int irq, void *dev_id)
+{
+	struct net_device *ndev = dev_id;
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	u8 isr;
+
+	isr = readb(&priv->regs->isr);
+	if (!(isr & priv->ier))
+		return IRQ_NONE;
+
+	if (isr & RCAR_CAN_ISR_ERSF)
+		rcar_can_error(ndev);
+
+	if (isr & RCAR_CAN_ISR_TXFF)
+		rcar_can_tx_done(ndev);
+
+	if (isr & RCAR_CAN_ISR_RXFF) {
+		if (napi_schedule_prep(&priv->napi)) {
+			/* Disable Rx FIFO interrupts */
+			priv->ier &= ~RCAR_CAN_IER_RXFIE;
+			writeb(priv->ier, &priv->regs->ier);
+			__napi_schedule(&priv->napi);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static void rcar_can_set_bittiming(struct net_device *dev)
+{
+	struct rcar_can_priv *priv = netdev_priv(dev);
+	struct can_bittiming *bt = &priv->can.bittiming;
+	u32 bcr;
+
+	bcr = RCAR_CAN_BCR_TSEG1(bt->phase_seg1 + bt->prop_seg - 1) |
+	      RCAR_CAN_BCR_BPR(bt->brp - 1) | RCAR_CAN_BCR_SJW(bt->sjw - 1) |
+	      RCAR_CAN_BCR_TSEG2(bt->phase_seg2 - 1);
+	/* Don't overwrite CLKR with 32-bit BCR access; CLKR has 8-bit access.
+	 * All the registers are big-endian but they get byte-swapped on 32-bit
+	 * read/write (but not on 8-bit, contrary to the manuals)...
+	 */
+	writel((bcr << 8) | priv->clock_select, &priv->regs->bcr);
+}
+
+static void rcar_can_start(struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	u16 ctlr;
+	int i;
+
+	/* Set controller to known mode:
+	 * - FIFO mailbox mode
+	 * - accept all messages
+	 * - overrun mode
+	 * CAN is in sleep mode after MCU hardware or software reset.
+	 */
+	ctlr = readw(&priv->regs->ctlr);
+	ctlr &= ~RCAR_CAN_CTLR_SLPM;
+	writew(ctlr, &priv->regs->ctlr);
+	/* Go to reset mode */
+	ctlr |= RCAR_CAN_CTLR_CANM_FORCE_RESET;
+	writew(ctlr, &priv->regs->ctlr);
+	for (i = 0; i < MAX_STR_READS; i++) {
+		if (readw(&priv->regs->str) & RCAR_CAN_STR_RSTST)
+			break;
+	}
+	rcar_can_set_bittiming(ndev);
+	ctlr |= RCAR_CAN_CTLR_IDFM_MIXED; /* Select mixed ID mode */
+	ctlr |= RCAR_CAN_CTLR_BOM_ENT;	/* Entry to halt mode automatically */
+					/* at bus-off */
+	ctlr |= RCAR_CAN_CTLR_MBM;	/* Select FIFO mailbox mode */
+	ctlr |= RCAR_CAN_CTLR_MLM;	/* Overrun mode */
+	writew(ctlr, &priv->regs->ctlr);
+
+	/* Accept all SID and EID */
+	writel(0, &priv->regs->mkr_2_9[6]);
+	writel(0, &priv->regs->mkr_2_9[7]);
+	/* In FIFO mailbox mode, write "0" to bits 24 to 31 */
+	writel(0, &priv->regs->mkivlr1);
+	/* Accept all frames */
+	writel(0, &priv->regs->fidcr[0]);
+	writel(RCAR_CAN_FIDCR_IDE | RCAR_CAN_FIDCR_RTR, &priv->regs->fidcr[1]);
+	/* Enable and configure FIFO mailbox interrupts */
+	writel(RCAR_CAN_MIER1_RXFIE | RCAR_CAN_MIER1_TXFIE, &priv->regs->mier1);
+
+	priv->ier = RCAR_CAN_IER_ERSIE | RCAR_CAN_IER_RXFIE |
+		    RCAR_CAN_IER_TXFIE;
+	writeb(priv->ier, &priv->regs->ier);
+
+	/* Accumulate error codes */
+	writeb(RCAR_CAN_ECSR_EDPM, &priv->regs->ecsr);
+	/* Enable error interrupts */
+	writeb(RCAR_CAN_EIER_EWIE | RCAR_CAN_EIER_EPIE | RCAR_CAN_EIER_BOEIE |
+	       (priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING ?
+	       RCAR_CAN_EIER_BEIE : 0) | RCAR_CAN_EIER_ORIE |
+	       RCAR_CAN_EIER_OLIE, &priv->regs->eier);
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+
+	/* Go to operation mode */
+	writew(ctlr & ~RCAR_CAN_CTLR_CANM, &priv->regs->ctlr);
+	for (i = 0; i < MAX_STR_READS; i++) {
+		if (!(readw(&priv->regs->str) & RCAR_CAN_STR_RSTST))
+			break;
+	}
+	/* Enable Rx and Tx FIFO */
+	writeb(RCAR_CAN_RFCR_RFE, &priv->regs->rfcr);
+	writeb(RCAR_CAN_TFCR_TFE, &priv->regs->tfcr);
+}
+
+static int rcar_can_open(struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	int err;
+
+	err = clk_prepare_enable(priv->clk);
+	if (err) {
+		netdev_err(ndev, "clk_prepare_enable() failed, error %d\n",
+			   err);
+		goto out;
+	}
+	err = open_candev(ndev);
+	if (err) {
+		netdev_err(ndev, "open_candev() failed, error %d\n", err);
+		goto out_clock;
+	}
+	napi_enable(&priv->napi);
+	err = request_irq(ndev->irq, rcar_can_interrupt, 0, ndev->name, ndev);
+	if (err) {
+		netdev_err(ndev, "error requesting interrupt %x\n", ndev->irq);
+		goto out_close;
+	}
+	can_led_event(ndev, CAN_LED_EVENT_OPEN);
+	rcar_can_start(ndev);
+	netif_start_queue(ndev);
+	return 0;
+out_close:
+	napi_disable(&priv->napi);
+	close_candev(ndev);
+out_clock:
+	clk_disable_unprepare(priv->clk);
+out:
+	return err;
+}
+
+static void rcar_can_stop(struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	u16 ctlr;
+	int i;
+
+	/* Go to (force) reset mode */
+	ctlr = readw(&priv->regs->ctlr);
+	ctlr |= RCAR_CAN_CTLR_CANM_FORCE_RESET;
+	writew(ctlr, &priv->regs->ctlr);
+	for (i = 0; i < MAX_STR_READS; i++) {
+		if (readw(&priv->regs->str) & RCAR_CAN_STR_RSTST)
+			break;
+	}
+	writel(0, &priv->regs->mier0);
+	writel(0, &priv->regs->mier1);
+	writeb(0, &priv->regs->ier);
+	writeb(0, &priv->regs->eier);
+	/* Go to sleep mode */
+	ctlr |= RCAR_CAN_CTLR_SLPM;
+	writew(ctlr, &priv->regs->ctlr);
+	priv->can.state = CAN_STATE_STOPPED;
+}
+
+static int rcar_can_close(struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+
+	netif_stop_queue(ndev);
+	rcar_can_stop(ndev);
+	free_irq(ndev->irq, ndev);
+	napi_disable(&priv->napi);
+	clk_disable_unprepare(priv->clk);
+	close_candev(ndev);
+	can_led_event(ndev, CAN_LED_EVENT_STOP);
+	return 0;
+}
+
+static netdev_tx_t rcar_can_start_xmit(struct sk_buff *skb,
+				       struct net_device *ndev)
+{
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	struct can_frame *cf = (struct can_frame *)skb->data;
+	u32 data, i;
+
+	if (can_dropped_invalid_skb(ndev, skb))
+		return NETDEV_TX_OK;
+
+	if (cf->can_id & CAN_EFF_FLAG)	/* Extended frame format */
+		data = (cf->can_id & CAN_EFF_MASK) | RCAR_CAN_IDE;
+	else				/* Standard frame format */
+		data = (cf->can_id & CAN_SFF_MASK) << RCAR_CAN_SID_SHIFT;
+
+	if (cf->can_id & CAN_RTR_FLAG) { /* Remote transmission request */
+		data |= RCAR_CAN_RTR;
+	} else {
+		for (i = 0; i < cf->can_dlc; i++)
+			writeb(cf->data[i],
+			       &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].data[i]);
+	}
+
+	writel(data, &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].id);
+
+	writeb(cf->can_dlc, &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].dlc);
+
+	priv->tx_dlc[priv->tx_head % RCAR_CAN_FIFO_DEPTH] = cf->can_dlc;
+	can_put_echo_skb(skb, ndev, priv->tx_head % RCAR_CAN_FIFO_DEPTH);
+	priv->tx_head++;
+	/* Start Tx: write 0xff to the TFPCR register to increment
+	 * the CPU-side pointer for the transmit FIFO to the next
+	 * mailbox location
+	 */
+	writeb(0xff, &priv->regs->tfpcr);
+	/* Stop the queue if we've filled all FIFO entries */
+	if (priv->tx_head - priv->tx_tail >= RCAR_CAN_FIFO_DEPTH)
+		netif_stop_queue(ndev);
+
+	return NETDEV_TX_OK;
+}
+
+static const struct net_device_ops rcar_can_netdev_ops = {
+	.ndo_open = rcar_can_open,
+	.ndo_stop = rcar_can_close,
+	.ndo_start_xmit = rcar_can_start_xmit,
+};
+
+static void rcar_can_rx_pkt(struct rcar_can_priv *priv)
+{
+	struct net_device_stats *stats = &priv->ndev->stats;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+	u32 data;
+	u8 dlc;
+
+	skb = alloc_can_skb(priv->ndev, &cf);
+	if (!skb) {
+		stats->rx_dropped++;
+		return;
+	}
+
+	data = readl(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].id);
+	if (data & RCAR_CAN_IDE)
+		cf->can_id = (data & CAN_EFF_MASK) | CAN_EFF_FLAG;
+	else
+		cf->can_id = (data >> RCAR_CAN_SID_SHIFT) & CAN_SFF_MASK;
+
+	dlc = readb(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].dlc);
+	cf->can_dlc = get_can_dlc(dlc);
+	if (data & RCAR_CAN_RTR) {
+		cf->can_id |= CAN_RTR_FLAG;
+	} else {
+		for (dlc = 0; dlc < cf->can_dlc; dlc++)
+			cf->data[dlc] =
+			readb(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].data[dlc]);
+	}
+
+	can_led_event(priv->ndev, CAN_LED_EVENT_RX);
+
+	stats->rx_bytes += cf->can_dlc;
+	stats->rx_packets++;
+	netif_receive_skb(skb);
+}
+
+static int rcar_can_rx_poll(struct napi_struct *napi, int quota)
+{
+	struct rcar_can_priv *priv = container_of(napi,
+						  struct rcar_can_priv, napi);
+	int num_pkts;
+
+	for (num_pkts = 0; num_pkts < quota; num_pkts++) {
+		u8 rfcr, isr;
+
+		isr = readb(&priv->regs->isr);
+		/* Clear interrupt bit */
+		if (isr & RCAR_CAN_ISR_RXFF)
+			writeb(isr & ~RCAR_CAN_ISR_RXFF, &priv->regs->isr);
+		rfcr = readb(&priv->regs->rfcr);
+		if (rfcr & RCAR_CAN_RFCR_RFEST)
+			break;
+		rcar_can_rx_pkt(priv);
+		/* Write 0xff to the RFPCR register to increment
+		 * the CPU-side pointer for the receive FIFO
+		 * to the next mailbox location
+		 */
+		writeb(0xff, &priv->regs->rfpcr);
+	}
+	/* All packets processed */
+	if (num_pkts < quota) {
+		napi_complete(napi);
+		priv->ier |= RCAR_CAN_IER_RXFIE;
+		writeb(priv->ier, &priv->regs->ier);
+	}
+	return num_pkts;
+}
+
+static int rcar_can_do_set_mode(struct net_device *ndev, enum can_mode mode)
+{
+	switch (mode) {
+	case CAN_MODE_START:
+		rcar_can_start(ndev);
+		netif_wake_queue(ndev);
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int rcar_can_get_berr_counter(const struct net_device *dev,
+				     struct can_berr_counter *bec)
+{
+	struct rcar_can_priv *priv = netdev_priv(dev);
+	int err;
+
+	err = clk_prepare_enable(priv->clk);
+	if (err)
+		return err;
+	bec->txerr = readb(&priv->regs->tecr);
+	bec->rxerr = readb(&priv->regs->recr);
+	clk_disable_unprepare(priv->clk);
+	return 0;
+}
+
+static int rcar_can_probe(struct platform_device *pdev)
+{
+	struct rcar_can_platform_data *pdata;
+	struct rcar_can_priv *priv;
+	struct net_device *ndev;
+	struct resource *mem;
+	void __iomem *addr;
+	int err = -ENODEV;
+	int irq;
+
+	pdata = dev_get_platdata(&pdev->dev);
+	if (!pdata) {
+		dev_err(&pdev->dev, "No platform data provided!\n");
+		goto fail;
+	}
+
+	irq = platform_get_irq(pdev, 0);
+	if (!irq) {
+		dev_err(&pdev->dev, "No IRQ resource\n");
+		goto fail;
+	}
+
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	addr = devm_ioremap_resource(&pdev->dev, mem);
+	if (IS_ERR(addr)) {
+		err = PTR_ERR(addr);
+		goto fail;
+	}
+
+	ndev = alloc_candev(sizeof(struct rcar_can_priv), RCAR_CAN_FIFO_DEPTH);
+	if (!ndev) {
+		dev_err(&pdev->dev, "alloc_candev() failed\n");
+		err = -ENOMEM;
+		goto fail;
+	}
+
+	priv = netdev_priv(ndev);
+
+	priv->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(priv->clk)) {
+		err = PTR_ERR(priv->clk);
+		dev_err(&pdev->dev, "cannot get clock: %d\n", err);
+		goto fail_clk;
+	}
+
+	ndev->netdev_ops = &rcar_can_netdev_ops;
+	ndev->irq = irq;
+	ndev->flags |= IFF_ECHO;
+	priv->ndev = ndev;
+	priv->regs = addr;
+	priv->clock_select = pdata->clock_select;
+	priv->can.clock.freq = clk_get_rate(priv->clk);
+	priv->can.bittiming_const = &rcar_can_bittiming_const;
+	priv->can.do_set_mode = rcar_can_do_set_mode;
+	priv->can.do_get_berr_counter = rcar_can_get_berr_counter;
+	priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING;
+	platform_set_drvdata(pdev, ndev);
+	SET_NETDEV_DEV(ndev, &pdev->dev);
+
+	netif_napi_add(ndev, &priv->napi, rcar_can_rx_poll,
+		       RCAR_CAN_NAPI_WEIGHT);
+	err = register_candev(ndev);
+	if (err) {
+		dev_err(&pdev->dev, "register_candev() failed, error %d\n",
+			err);
+		goto fail_candev;
+	}
+
+	devm_can_led_init(ndev);
+
+	dev_info(&pdev->dev, "device registered (reg_base=%p, irq=%u)\n",
+		 priv->regs, ndev->irq);
+
+	return 0;
+fail_candev:
+	netif_napi_del(&priv->napi);
+fail_clk:
+	free_candev(ndev);
+fail:
+	return err;
+}
+
+static int rcar_can_remove(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+
+	unregister_candev(ndev);
+	netif_napi_del(&priv->napi);
+	free_candev(ndev);
+	return 0;
+}
+
+static int __maybe_unused rcar_can_suspend(struct device *dev)
+{
+	struct net_device *ndev = dev_get_drvdata(dev);
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	u16 ctlr;
+
+	if (netif_running(ndev)) {
+		netif_stop_queue(ndev);
+		netif_device_detach(ndev);
+	}
+	ctlr = readw(&priv->regs->ctlr);
+	ctlr |= RCAR_CAN_CTLR_CANM_HALT;
+	writew(ctlr, &priv->regs->ctlr);
+	ctlr |= RCAR_CAN_CTLR_SLPM;
+	writew(ctlr, &priv->regs->ctlr);
+	priv->can.state = CAN_STATE_SLEEPING;
+
+	clk_disable(priv->clk);
+	return 0;
+}
+
+static int __maybe_unused rcar_can_resume(struct device *dev)
+{
+	struct net_device *ndev = dev_get_drvdata(dev);
+	struct rcar_can_priv *priv = netdev_priv(ndev);
+	u16 ctlr;
+	int err;
+
+	err = clk_enable(priv->clk);
+	if (err) {
+		netdev_err(ndev, "clk_enable() failed, error %d\n", err);
+		return err;
+	}
+
+	ctlr = readw(&priv->regs->ctlr);
+	ctlr &= ~RCAR_CAN_CTLR_SLPM;
+	writew(ctlr, &priv->regs->ctlr);
+	ctlr &= ~RCAR_CAN_CTLR_CANM;
+	writew(ctlr, &priv->regs->ctlr);
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+
+	if (netif_running(ndev)) {
+		netif_device_attach(ndev);
+		netif_start_queue(ndev);
+	}
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(rcar_can_pm_ops, rcar_can_suspend, rcar_can_resume);
+
+static struct platform_driver rcar_can_driver = {
+	.driver = {
+		.name = RCAR_CAN_DRV_NAME,
+		.owner = THIS_MODULE,
+		.pm = &rcar_can_pm_ops,
+	},
+	.probe = rcar_can_probe,
+	.remove = rcar_can_remove,
+};
+
+module_platform_driver(rcar_can_driver);
+
+MODULE_AUTHOR("Cogent Embedded, Inc.");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("CAN driver for Renesas R-Car SoC");
+MODULE_ALIAS("platform:" RCAR_CAN_DRV_NAME);
diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c
index dcf9196f6316..ea4d4f1a6411 100644
--- a/drivers/net/can/slcan.c
+++ b/drivers/net/can/slcan.c
@@ -52,6 +52,7 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/workqueue.h>
 #include <linux/can.h>
 #include <linux/can/skb.h>
 
@@ -85,6 +86,7 @@ struct slcan {
 	struct tty_struct	*tty;		/* ptr to TTY structure	     */
 	struct net_device	*dev;		/* easy for intr handling    */
 	spinlock_t		lock;
+	struct work_struct	tx_work;	/* Flushes transmit buffer   */
 
 	/* These are pointers to the malloc()ed frame buffers. */
 	unsigned char		rbuff[SLC_MTU];	/* receiver buffer	     */
@@ -309,36 +311,46 @@ static void slc_encaps(struct slcan *sl, struct can_frame *cf)
 	sl->dev->stats.tx_bytes += cf->can_dlc;
 }
 
-/*
- * Called by the driver when there's room for more data.  If we have
- * more packets to send, we send them here.
- */
-static void slcan_write_wakeup(struct tty_struct *tty)
+/* Write out any remaining transmit buffer. Scheduled when tty is writable */
+static void slcan_transmit(struct work_struct *work)
 {
+	struct slcan *sl = container_of(work, struct slcan, tx_work);
 	int actual;
-	struct slcan *sl = (struct slcan *) tty->disc_data;
 
+	spin_lock_bh(&sl->lock);
 	/* First make sure we're connected. */
-	if (!sl || sl->magic != SLCAN_MAGIC || !netif_running(sl->dev))
+	if (!sl->tty || sl->magic != SLCAN_MAGIC || !netif_running(sl->dev)) {
+		spin_unlock_bh(&sl->lock);
 		return;
+	}
 
-	spin_lock_bh(&sl->lock);
 	if (sl->xleft <= 0)  {
 		/* Now serial buffer is almost free & we can start
 		 * transmission of another packet */
 		sl->dev->stats.tx_packets++;
-		clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
+		clear_bit(TTY_DO_WRITE_WAKEUP, &sl->tty->flags);
 		spin_unlock_bh(&sl->lock);
 		netif_wake_queue(sl->dev);
 		return;
 	}
 
-	actual = tty->ops->write(tty, sl->xhead, sl->xleft);
+	actual = sl->tty->ops->write(sl->tty, sl->xhead, sl->xleft);
 	sl->xleft -= actual;
 	sl->xhead += actual;
 	spin_unlock_bh(&sl->lock);
 }
 
+/*
+ * Called by the driver when there's room for more data.
+ * Schedule the transmit.
+ */
+static void slcan_write_wakeup(struct tty_struct *tty)
+{
+	struct slcan *sl = tty->disc_data;
+
+	schedule_work(&sl->tx_work);
+}
+
 /* Send a can_frame to a TTY queue. */
 static netdev_tx_t slc_xmit(struct sk_buff *skb, struct net_device *dev)
 {
@@ -528,6 +540,7 @@ static struct slcan *slc_alloc(dev_t line)
 	sl->magic = SLCAN_MAGIC;
 	sl->dev	= dev;
 	spin_lock_init(&sl->lock);
+	INIT_WORK(&sl->tx_work, slcan_transmit);
 	slcan_devs[i] = dev;
 
 	return sl;
@@ -626,8 +639,12 @@ static void slcan_close(struct tty_struct *tty)
 	if (!sl || sl->magic != SLCAN_MAGIC || sl->tty != tty)
 		return;
 
+	spin_lock_bh(&sl->lock);
 	tty->disc_data = NULL;
 	sl->tty = NULL;
+	spin_unlock_bh(&sl->lock);
+
+	flush_work(&sl->tx_work);
 
 	/* Flush network side */
 	unregister_netdev(sl->dev);
diff --git a/drivers/net/can/softing/softing_main.c b/drivers/net/can/softing/softing_main.c
index 7d8c8f3672dd..bacd236ce306 100644
--- a/drivers/net/can/softing/softing_main.c
+++ b/drivers/net/can/softing/softing_main.c
@@ -556,15 +556,6 @@ failed:
 /*
  * netdev sysfs
  */
-static ssize_t show_channel(struct device *dev, struct device_attribute *attr,
-		char *buf)
-{
-	struct net_device *ndev = to_net_dev(dev);
-	struct softing_priv *priv = netdev2softing(ndev);
-
-	return sprintf(buf, "%i\n", priv->index);
-}
-
 static ssize_t show_chip(struct device *dev, struct device_attribute *attr,
 		char *buf)
 {
@@ -609,12 +600,10 @@ static ssize_t store_output(struct device *dev, struct device_attribute *attr,
 	return count;
 }
 
-static const DEVICE_ATTR(channel, S_IRUGO, show_channel, NULL);
 static const DEVICE_ATTR(chip, S_IRUGO, show_chip, NULL);
 static const DEVICE_ATTR(output, S_IRUGO | S_IWUSR, show_output, store_output);
 
 static const struct attribute *const netdev_sysfs_attrs[] = {
-	&dev_attr_channel.attr,
 	&dev_attr_chip.attr,
 	&dev_attr_output.attr,
 	NULL,
@@ -679,17 +668,20 @@ static int softing_netdev_register(struct net_device *netdev)
 {
 	int ret;
 
-	netdev->sysfs_groups[0] = &netdev_sysfs_group;
 	ret = register_candev(netdev);
 	if (ret) {
 		dev_alert(&netdev->dev, "register failed\n");
 		return ret;
 	}
+	if (sysfs_create_group(&netdev->dev.kobj, &netdev_sysfs_group) < 0)
+		netdev_alert(netdev, "sysfs group failed\n");
+
 	return 0;
 }
 
 static void softing_netdev_cleanup(struct net_device *netdev)
 {
+	sysfs_remove_group(&netdev->dev.kobj, &netdev_sysfs_group);
 	unregister_candev(netdev);
 	free_candev(netdev);
 }
@@ -721,8 +713,6 @@ DEV_ATTR_RO(firmware_version, id.fw_version);
 DEV_ATTR_RO_STR(hardware, pdat->name);
 DEV_ATTR_RO(hardware_version, id.hw_version);
 DEV_ATTR_RO(license, id.license);
-DEV_ATTR_RO(frequency, id.freq);
-DEV_ATTR_RO(txpending, tx.pending);
 
 static struct attribute *softing_pdev_attrs[] = {
 	&dev_attr_serial.attr,
@@ -731,8 +721,6 @@ static struct attribute *softing_pdev_attrs[] = {
 	&dev_attr_hardware.attr,
 	&dev_attr_hardware_version.attr,
 	&dev_attr_license.attr,
-	&dev_attr_frequency.attr,
-	&dev_attr_txpending.attr,
 	NULL,
 };
 
diff --git a/drivers/net/can/spi/Kconfig b/drivers/net/can/spi/Kconfig
new file mode 100644
index 000000000000..148cae5871a6
--- /dev/null
+++ b/drivers/net/can/spi/Kconfig
@@ -0,0 +1,10 @@
+menu "CAN SPI interfaces"
+	depends on SPI
+
+config CAN_MCP251X
+	tristate "Microchip MCP251x SPI CAN controllers"
+	depends on HAS_DMA
+	---help---
+	  Driver for the Microchip MCP251x SPI CAN controllers.
+
+endmenu
diff --git a/drivers/net/can/spi/Makefile b/drivers/net/can/spi/Makefile
new file mode 100644
index 000000000000..90bcacffbc65
--- /dev/null
+++ b/drivers/net/can/spi/Makefile
@@ -0,0 +1,8 @@
+#
+#  Makefile for the Linux Controller Area Network SPI drivers.
+#
+
+
+obj-$(CONFIG_CAN_MCP251X)	+= mcp251x.o
+
+ccflags-$(CONFIG_CAN_DEBUG_DEVICES) := -DDEBUG
diff --git a/drivers/net/can/mcp251x.c b/drivers/net/can/spi/mcp251x.c
index 28c11f815245..5df239e68812 100644
--- a/drivers/net/can/mcp251x.c
+++ b/drivers/net/can/spi/mcp251x.c
@@ -214,6 +214,8 @@
 
 #define TX_ECHO_SKB_MAX	1
 
+#define MCP251X_OST_DELAY_MS	(5)
+
 #define DEVICE_NAME "mcp251x"
 
 static int mcp251x_enable_dma; /* Enable SPI DMA. Default: 0 (Off) */
@@ -624,50 +626,45 @@ static int mcp251x_setup(struct net_device *net, struct mcp251x_priv *priv,
 static int mcp251x_hw_reset(struct spi_device *spi)
 {
 	struct mcp251x_priv *priv = spi_get_drvdata(spi);
+	u8 reg;
 	int ret;
-	unsigned long timeout;
+
+	/* Wait for oscillator startup timer after power up */
+	mdelay(MCP251X_OST_DELAY_MS);
 
 	priv->spi_tx_buf[0] = INSTRUCTION_RESET;
-	ret = spi_write(spi, priv->spi_tx_buf, 1);
-	if (ret) {
-		dev_err(&spi->dev, "reset failed: ret = %d\n", ret);
-		return -EIO;
-	}
+	ret = mcp251x_spi_trans(spi, 1);
+	if (ret)
+		return ret;
+
+	/* Wait for oscillator startup timer after reset */
+	mdelay(MCP251X_OST_DELAY_MS);
+	
+	reg = mcp251x_read_reg(spi, CANSTAT);
+	if ((reg & CANCTRL_REQOP_MASK) != CANCTRL_REQOP_CONF)
+		return -ENODEV;
 
-	/* Wait for reset to finish */
-	timeout = jiffies + HZ;
-	mdelay(10);
-	while ((mcp251x_read_reg(spi, CANSTAT) & CANCTRL_REQOP_MASK)
-	       != CANCTRL_REQOP_CONF) {
-		schedule();
-		if (time_after(jiffies, timeout)) {
-			dev_err(&spi->dev, "MCP251x didn't"
-				" enter in conf mode after reset\n");
-			return -EBUSY;
-		}
-	}
 	return 0;
 }
 
 static int mcp251x_hw_probe(struct spi_device *spi)
 {
-	int st1, st2;
+	u8 ctrl;
+	int ret;
 
-	mcp251x_hw_reset(spi);
+	ret = mcp251x_hw_reset(spi);
+	if (ret)
+		return ret;
 
-	/*
-	 * Please note that these are "magic values" based on after
-	 * reset defaults taken from data sheet which allows us to see
-	 * if we really have a chip on the bus (we avoid common all
-	 * zeroes or all ones situations)
-	 */
-	st1 = mcp251x_read_reg(spi, CANSTAT) & 0xEE;
-	st2 = mcp251x_read_reg(spi, CANCTRL) & 0x17;
+	ctrl = mcp251x_read_reg(spi, CANCTRL);
+
+	dev_dbg(&spi->dev, "CANCTRL 0x%02x\n", ctrl);
 
-	dev_dbg(&spi->dev, "CANSTAT 0x%02x CANCTRL 0x%02x\n", st1, st2);
+	/* Check for power up default value */
+	if ((ctrl & 0x17) != 0x07)
+		return -ENODEV;
 
-	/* Check for power up default values */
-	return (st1 == 0x80 && st2 == 0x07) ? 1 : 0;
+	return 0;
 }
 
 static int mcp251x_power_enable(struct regulator *reg, int enable)
@@ -776,7 +773,6 @@ static void mcp251x_restart_work_handler(struct work_struct *ws)
 
 	mutex_lock(&priv->mcp_lock);
 	if (priv->after_suspend) {
-		mdelay(10);
 		mcp251x_hw_reset(spi);
 		mcp251x_setup(net, priv, spi);
 		if (priv->after_suspend & AFTER_SUSPEND_RESTART) {
@@ -955,7 +951,7 @@ static int mcp251x_open(struct net_device *net)
 	priv->tx_len = 0;
 
 	ret = request_threaded_irq(spi->irq, NULL, mcp251x_can_ist,
-				   flags, DEVICE_NAME, priv);
+				   flags | IRQF_ONESHOT, DEVICE_NAME, priv);
 	if (ret) {
 		dev_err(&spi->dev, "failed to acquire irq %d\n", spi->irq);
 		mcp251x_power_enable(priv->transceiver, 0);
@@ -1032,8 +1028,8 @@ static int mcp251x_can_probe(struct spi_device *spi)
 	struct mcp251x_platform_data *pdata = dev_get_platdata(&spi->dev);
 	struct net_device *net;
 	struct mcp251x_priv *priv;
-	int freq, ret = -ENODEV;
 	struct clk *clk;
+	int freq, ret;
 
 	clk = devm_clk_get(&spi->dev, NULL);
 	if (IS_ERR(clk)) {
@@ -1076,6 +1072,18 @@ static int mcp251x_can_probe(struct spi_device *spi)
 	priv->net = net;
 	priv->clk = clk;
 
+	spi_set_drvdata(spi, priv);
+
+	/* Configure the SPI bus */
+	spi->bits_per_word = 8;
+	if (mcp251x_is_2510(spi))
+		spi->max_speed_hz = spi->max_speed_hz ? : 5 * 1000 * 1000;
+	else
+		spi->max_speed_hz = spi->max_speed_hz ? : 10 * 1000 * 1000;
+	ret = spi_setup(spi);
+	if (ret)
+		goto out_clk;
+
 	priv->power = devm_regulator_get(&spi->dev, "vdd");
 	priv->transceiver = devm_regulator_get(&spi->dev, "xceiver");
 	if ((PTR_ERR(priv->power) == -EPROBE_DEFER) ||
@@ -1088,8 +1096,6 @@ static int mcp251x_can_probe(struct spi_device *spi)
 	if (ret)
 		goto out_clk;
 
-	spi_set_drvdata(spi, priv);
-
 	priv->spi = spi;
 	mutex_init(&priv->mcp_lock);
 
@@ -1134,20 +1140,11 @@ static int mcp251x_can_probe(struct spi_device *spi)
 
 	SET_NETDEV_DEV(net, &spi->dev);
 
-	/* Configure the SPI bus */
-	spi->mode = spi->mode ? : SPI_MODE_0;
-	if (mcp251x_is_2510(spi))
-		spi->max_speed_hz = spi->max_speed_hz ? : 5 * 1000 * 1000;
-	else
-		spi->max_speed_hz = spi->max_speed_hz ? : 10 * 1000 * 1000;
-	spi->bits_per_word = 8;
-	spi_setup(spi);
-
 	/* Here is OK to not lock the MCP, no one knows about it yet */
-	if (!mcp251x_hw_probe(spi)) {
-		ret = -ENODEV;
+	ret = mcp251x_hw_probe(spi);
+	if (ret)
 		goto error_probe;
-	}
+
 	mcp251x_hw_sleep(spi);
 
 	ret = register_candev(net);
@@ -1156,7 +1153,7 @@ static int mcp251x_can_probe(struct spi_device *spi)
 
 	devm_can_led_init(net);
 
-	return ret;
+	return 0;
 
 error_probe:
 	if (mcp251x_enable_dma)
diff --git a/drivers/net/can/usb/Kconfig b/drivers/net/can/usb/Kconfig
index fc96a3d83ebe..a77db919363c 100644
--- a/drivers/net/can/usb/Kconfig
+++ b/drivers/net/can/usb/Kconfig
@@ -13,13 +13,21 @@ config CAN_ESD_USB2
           This driver supports the CAN-USB/2 interface
           from esd electronic system design gmbh (http://www.esd.eu).
 
+config CAN_GS_USB
+	tristate "Geschwister Schneider UG interfaces"
+	---help---
+	  This driver supports the Geschwister Schneider USB/CAN devices.
+	  If unsure choose N,
+	  choose Y for built in support,
+	  M to compile as module (module will be named: gs_usb).
+
 config CAN_KVASER_USB
 	tristate "Kvaser CAN/USB interface"
 	---help---
 	  This driver adds support for Kvaser CAN/USB devices like Kvaser
 	  Leaf Light.
 
-	  The driver gives support for the following devices:
+	  The driver provides support for the following devices:
 	    - Kvaser Leaf Light
 	    - Kvaser Leaf Professional HS
 	    - Kvaser Leaf SemiPro HS
@@ -36,6 +44,8 @@ config CAN_KVASER_USB
 	    - Kvaser Leaf Light "China"
 	    - Kvaser BlackBird SemiPro
 	    - Kvaser USBcan R
+	    - Kvaser Leaf Light v2
+	    - Kvaser Mini PCI Express HS
 
 	  If unsure, say N.
 
diff --git a/drivers/net/can/usb/Makefile b/drivers/net/can/usb/Makefile
index becef460a91a..7b9a393b1ac8 100644
--- a/drivers/net/can/usb/Makefile
+++ b/drivers/net/can/usb/Makefile
@@ -4,6 +4,7 @@
 
 obj-$(CONFIG_CAN_EMS_USB) += ems_usb.o
 obj-$(CONFIG_CAN_ESD_USB2) += esd_usb2.o
+obj-$(CONFIG_CAN_GS_USB) += gs_usb.o
 obj-$(CONFIG_CAN_KVASER_USB) += kvaser_usb.o
 obj-$(CONFIG_CAN_PEAK_USB) += peak_usb/
 obj-$(CONFIG_CAN_8DEV_USB) += usb_8dev.o
diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c
new file mode 100644
index 000000000000..04b0f84612f0
--- /dev/null
+++ b/drivers/net/can/usb/gs_usb.c
@@ -0,0 +1,971 @@
+/* CAN driver for Geschwister Schneider USB/CAN devices.
+ *
+ * Copyright (C) 2013 Geschwister Schneider Technologie-,
+ * Entwicklungs- und Vertriebs UG (Haftungsbeschränkt).
+ *
+ * Many thanks to all socketcan devs!
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published
+ * by the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/signal.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/usb.h>
+
+#include <linux/can.h>
+#include <linux/can/dev.h>
+#include <linux/can/error.h>
+
+/* Device specific constants */
+#define USB_GSUSB_1_VENDOR_ID      0x1d50
+#define USB_GSUSB_1_PRODUCT_ID     0x606f
+
+#define GSUSB_ENDPOINT_IN          1
+#define GSUSB_ENDPOINT_OUT         2
+
+/* Device specific constants */
+enum gs_usb_breq {
+	GS_USB_BREQ_HOST_FORMAT = 0,
+	GS_USB_BREQ_BITTIMING,
+	GS_USB_BREQ_MODE,
+	GS_USB_BREQ_BERR,
+	GS_USB_BREQ_BT_CONST,
+	GS_USB_BREQ_DEVICE_CONFIG
+};
+
+enum gs_can_mode {
+	/* reset a channel. turns it off */
+	GS_CAN_MODE_RESET = 0,
+	/* starts a channel */
+	GS_CAN_MODE_START
+};
+
+enum gs_can_state {
+	GS_CAN_STATE_ERROR_ACTIVE = 0,
+	GS_CAN_STATE_ERROR_WARNING,
+	GS_CAN_STATE_ERROR_PASSIVE,
+	GS_CAN_STATE_BUS_OFF,
+	GS_CAN_STATE_STOPPED,
+	GS_CAN_STATE_SLEEPING
+};
+
+/* data types passed between host and device */
+struct gs_host_config {
+	u32 byte_order;
+} __packed;
+/* All data exchanged between host and device is exchanged in host byte order,
+ * thanks to the struct gs_host_config byte_order member, which is sent first
+ * to indicate the desired byte order.
+ */
+
+struct gs_device_config {
+	u8 reserved1;
+	u8 reserved2;
+	u8 reserved3;
+	u8 icount;
+	u32 sw_version;
+	u32 hw_version;
+} __packed;
+
+#define GS_CAN_MODE_NORMAL               0
+#define GS_CAN_MODE_LISTEN_ONLY          (1<<0)
+#define GS_CAN_MODE_LOOP_BACK            (1<<1)
+#define GS_CAN_MODE_TRIPLE_SAMPLE        (1<<2)
+#define GS_CAN_MODE_ONE_SHOT             (1<<3)
+
+struct gs_device_mode {
+	u32 mode;
+	u32 flags;
+} __packed;
+
+struct gs_device_state {
+	u32 state;
+	u32 rxerr;
+	u32 txerr;
+} __packed;
+
+struct gs_device_bittiming {
+	u32 prop_seg;
+	u32 phase_seg1;
+	u32 phase_seg2;
+	u32 sjw;
+	u32 brp;
+} __packed;
+
+#define GS_CAN_FEATURE_LISTEN_ONLY      (1<<0)
+#define GS_CAN_FEATURE_LOOP_BACK        (1<<1)
+#define GS_CAN_FEATURE_TRIPLE_SAMPLE    (1<<2)
+#define GS_CAN_FEATURE_ONE_SHOT         (1<<3)
+
+struct gs_device_bt_const {
+	u32 feature;
+	u32 fclk_can;
+	u32 tseg1_min;
+	u32 tseg1_max;
+	u32 tseg2_min;
+	u32 tseg2_max;
+	u32 sjw_max;
+	u32 brp_min;
+	u32 brp_max;
+	u32 brp_inc;
+} __packed;
+
+#define GS_CAN_FLAG_OVERFLOW 1
+
+struct gs_host_frame {
+	u32 echo_id;
+	u32 can_id;
+
+	u8 can_dlc;
+	u8 channel;
+	u8 flags;
+	u8 reserved;
+
+	u8 data[8];
+} __packed;
+/* The GS USB devices make use of the same flags and masks as in
+ * linux/can.h and linux/can/error.h, and no additional mapping is necessary.
+ */
+
+/* Only send a max of GS_MAX_TX_URBS frames per channel at a time. */
+#define GS_MAX_TX_URBS 10
+/* Only launch a max of GS_MAX_RX_URBS usb requests at a time. */
+#define GS_MAX_RX_URBS 30
+/* Maximum number of interfaces the driver supports per device.
+ * Current hardware only supports 2 interfaces. The future may vary.
+ */
+#define GS_MAX_INTF 2
+
+struct gs_tx_context {
+	struct gs_can *dev;
+	unsigned int echo_id;
+};
+
+struct gs_can {
+	struct can_priv can; /* must be the first member */
+
+	struct gs_usb *parent;
+
+	struct net_device *netdev;
+	struct usb_device *udev;
+	struct usb_interface *iface;
+
+	struct can_bittiming_const bt_const;
+	unsigned int channel;	/* channel number */
+
+	/* This lock prevents a race condition between xmit and recieve. */
+	spinlock_t tx_ctx_lock;
+	struct gs_tx_context tx_context[GS_MAX_TX_URBS];
+
+	struct usb_anchor tx_submitted;
+	atomic_t active_tx_urbs;
+};
+
+/* usb interface struct */
+struct gs_usb {
+	struct gs_can *canch[GS_MAX_INTF];
+	struct usb_anchor rx_submitted;
+	atomic_t active_channels;
+	struct usb_device *udev;
+};
+
+/* 'allocate' a tx context.
+ * returns a valid tx context or NULL if there is no space.
+ */
+static struct gs_tx_context *gs_alloc_tx_context(struct gs_can *dev)
+{
+	int i = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->tx_ctx_lock, flags);
+
+	for (; i < GS_MAX_TX_URBS; i++) {
+		if (dev->tx_context[i].echo_id == GS_MAX_TX_URBS) {
+			dev->tx_context[i].echo_id = i;
+			spin_unlock_irqrestore(&dev->tx_ctx_lock, flags);
+			return &dev->tx_context[i];
+		}
+	}
+
+	spin_unlock_irqrestore(&dev->tx_ctx_lock, flags);
+	return NULL;
+}
+
+/* releases a tx context
+ */
+static void gs_free_tx_context(struct gs_tx_context *txc)
+{
+	txc->echo_id = GS_MAX_TX_URBS;
+}
+
+/* Get a tx context by id.
+ */
+static struct gs_tx_context *gs_get_tx_context(struct gs_can *dev, unsigned int id)
+{
+	unsigned long flags;
+
+	if (id < GS_MAX_TX_URBS) {
+		spin_lock_irqsave(&dev->tx_ctx_lock, flags);
+		if (dev->tx_context[id].echo_id == id) {
+			spin_unlock_irqrestore(&dev->tx_ctx_lock, flags);
+			return &dev->tx_context[id];
+		}
+		spin_unlock_irqrestore(&dev->tx_ctx_lock, flags);
+	}
+	return NULL;
+}
+
+static int gs_cmd_reset(struct gs_usb *gsusb, struct gs_can *gsdev)
+{
+	struct gs_device_mode *dm;
+	struct usb_interface *intf = gsdev->iface;
+	int rc;
+
+	dm = kzalloc(sizeof(*dm), GFP_KERNEL);
+	if (!dm)
+		return -ENOMEM;
+
+	dm->mode = GS_CAN_MODE_RESET;
+
+	rc = usb_control_msg(interface_to_usbdev(intf),
+			     usb_sndctrlpipe(interface_to_usbdev(intf), 0),
+			     GS_USB_BREQ_MODE,
+			     USB_DIR_OUT|USB_TYPE_VENDOR|USB_RECIP_INTERFACE,
+			     gsdev->channel,
+			     0,
+			     dm,
+			     sizeof(*dm),
+			     1000);
+
+	return rc;
+}
+
+static void gs_update_state(struct gs_can *dev, struct can_frame *cf)
+{
+	struct can_device_stats *can_stats = &dev->can.can_stats;
+
+	if (cf->can_id & CAN_ERR_RESTARTED) {
+		dev->can.state = CAN_STATE_ERROR_ACTIVE;
+		can_stats->restarts++;
+	} else if (cf->can_id & CAN_ERR_BUSOFF) {
+		dev->can.state = CAN_STATE_BUS_OFF;
+		can_stats->bus_off++;
+	} else if (cf->can_id & CAN_ERR_CRTL) {
+		if ((cf->data[1] & CAN_ERR_CRTL_TX_WARNING) ||
+		    (cf->data[1] & CAN_ERR_CRTL_RX_WARNING)) {
+			dev->can.state = CAN_STATE_ERROR_WARNING;
+			can_stats->error_warning++;
+		} else if ((cf->data[1] & CAN_ERR_CRTL_TX_PASSIVE) ||
+			   (cf->data[1] & CAN_ERR_CRTL_RX_PASSIVE)) {
+			dev->can.state = CAN_STATE_ERROR_PASSIVE;
+			can_stats->error_passive++;
+		} else {
+			dev->can.state = CAN_STATE_ERROR_ACTIVE;
+		}
+	}
+}
+
+static void gs_usb_recieve_bulk_callback(struct urb *urb)
+{
+	struct gs_usb *usbcan = urb->context;
+	struct gs_can *dev;
+	struct net_device *netdev;
+	int rc;
+	struct net_device_stats *stats;
+	struct gs_host_frame *hf = urb->transfer_buffer;
+	struct gs_tx_context *txc;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+
+	BUG_ON(!usbcan);
+
+	switch (urb->status) {
+	case 0: /* success */
+		break;
+	case -ENOENT:
+	case -ESHUTDOWN:
+		return;
+	default:
+		/* do not resubmit aborted urbs. eg: when device goes down */
+		return;
+	}
+
+	/* device reports out of range channel id */
+	if (hf->channel >= GS_MAX_INTF)
+		goto resubmit_urb;
+
+	dev = usbcan->canch[hf->channel];
+
+	netdev = dev->netdev;
+	stats = &netdev->stats;
+
+	if (!netif_device_present(netdev))
+		return;
+
+	if (hf->echo_id == -1) { /* normal rx */
+		skb = alloc_can_skb(dev->netdev, &cf);
+		if (!skb)
+			return;
+
+		cf->can_id = hf->can_id;
+
+		cf->can_dlc = get_can_dlc(hf->can_dlc);
+		memcpy(cf->data, hf->data, 8);
+
+		/* ERROR frames tell us information about the controller */
+		if (hf->can_id & CAN_ERR_FLAG)
+			gs_update_state(dev, cf);
+
+		netdev->stats.rx_packets++;
+		netdev->stats.rx_bytes += hf->can_dlc;
+
+		netif_rx(skb);
+	} else { /* echo_id == hf->echo_id */
+		if (hf->echo_id >= GS_MAX_TX_URBS) {
+			netdev_err(netdev,
+				   "Unexpected out of range echo id %d\n",
+				   hf->echo_id);
+			goto resubmit_urb;
+		}
+
+		netdev->stats.tx_packets++;
+		netdev->stats.tx_bytes += hf->can_dlc;
+
+		txc = gs_get_tx_context(dev, hf->echo_id);
+
+		/* bad devices send bad echo_ids. */
+		if (!txc) {
+			netdev_err(netdev,
+				   "Unexpected unused echo id %d\n",
+				   hf->echo_id);
+			goto resubmit_urb;
+		}
+
+		can_get_echo_skb(netdev, hf->echo_id);
+
+		gs_free_tx_context(txc);
+
+		netif_wake_queue(netdev);
+	}
+
+	if (hf->flags & GS_CAN_FLAG_OVERFLOW) {
+		skb = alloc_can_err_skb(netdev, &cf);
+		if (!skb)
+			goto resubmit_urb;
+
+		cf->can_id |= CAN_ERR_CRTL;
+		cf->can_dlc = CAN_ERR_DLC;
+		cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+		stats->rx_over_errors++;
+		stats->rx_errors++;
+		netif_rx(skb);
+	}
+
+ resubmit_urb:
+	usb_fill_bulk_urb(urb,
+			  usbcan->udev,
+			  usb_rcvbulkpipe(usbcan->udev, GSUSB_ENDPOINT_IN),
+			  hf,
+			  sizeof(struct gs_host_frame),
+			  gs_usb_recieve_bulk_callback,
+			  usbcan
+			  );
+
+	rc = usb_submit_urb(urb, GFP_ATOMIC);
+
+	/* USB failure take down all interfaces */
+	if (rc == -ENODEV) {
+		for (rc = 0; rc < GS_MAX_INTF; rc++) {
+			if (usbcan->canch[rc])
+				netif_device_detach(usbcan->canch[rc]->netdev);
+		}
+	}
+}
+
+static int gs_usb_set_bittiming(struct net_device *netdev)
+{
+	struct gs_can *dev = netdev_priv(netdev);
+	struct can_bittiming *bt = &dev->can.bittiming;
+	struct usb_interface *intf = dev->iface;
+	int rc;
+	struct gs_device_bittiming *dbt;
+
+	dbt = kmalloc(sizeof(*dbt), GFP_KERNEL);
+	if (!dbt)
+		return -ENOMEM;
+
+	dbt->prop_seg = bt->prop_seg;
+	dbt->phase_seg1 = bt->phase_seg1;
+	dbt->phase_seg2 = bt->phase_seg2;
+	dbt->sjw = bt->sjw;
+	dbt->brp = bt->brp;
+
+	/* request bit timings */
+	rc = usb_control_msg(interface_to_usbdev(intf),
+			     usb_sndctrlpipe(interface_to_usbdev(intf), 0),
+			     GS_USB_BREQ_BITTIMING,
+			     USB_DIR_OUT|USB_TYPE_VENDOR|USB_RECIP_INTERFACE,
+			     dev->channel,
+			     0,
+			     dbt,
+			     sizeof(*dbt),
+			     1000);
+
+	kfree(dbt);
+
+	if (rc < 0)
+		dev_err(netdev->dev.parent, "Couldn't set bittimings (err=%d)",
+			rc);
+
+	return rc;
+}
+
+static void gs_usb_xmit_callback(struct urb *urb)
+{
+	struct gs_tx_context *txc = urb->context;
+	struct gs_can *dev = txc->dev;
+	struct net_device *netdev = dev->netdev;
+
+	if (urb->status)
+		netdev_info(netdev, "usb xmit fail %d\n", txc->echo_id);
+
+	usb_free_coherent(urb->dev,
+			  urb->transfer_buffer_length,
+			  urb->transfer_buffer,
+			  urb->transfer_dma);
+
+	atomic_dec(&dev->active_tx_urbs);
+
+	if (!netif_device_present(netdev))
+		return;
+
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+}
+
+static netdev_tx_t gs_can_start_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct gs_can *dev = netdev_priv(netdev);
+	struct net_device_stats *stats = &dev->netdev->stats;
+	struct urb *urb;
+	struct gs_host_frame *hf;
+	struct can_frame *cf;
+	int rc;
+	unsigned int idx;
+	struct gs_tx_context *txc;
+
+	if (can_dropped_invalid_skb(netdev, skb))
+		return NETDEV_TX_OK;
+
+	/* find an empty context to keep track of transmission */
+	txc = gs_alloc_tx_context(dev);
+	if (!txc)
+		return NETDEV_TX_BUSY;
+
+	/* create a URB, and a buffer for it */
+	urb = usb_alloc_urb(0, GFP_ATOMIC);
+	if (!urb) {
+		netdev_err(netdev, "No memory left for URB\n");
+		goto nomem_urb;
+	}
+
+	hf = usb_alloc_coherent(dev->udev, sizeof(*hf), GFP_ATOMIC,
+				&urb->transfer_dma);
+	if (!hf) {
+		netdev_err(netdev, "No memory left for USB buffer\n");
+		goto nomem_hf;
+	}
+
+	idx = txc->echo_id;
+
+	if (idx >= GS_MAX_TX_URBS) {
+		netdev_err(netdev, "Invalid tx context %d\n", idx);
+		goto badidx;
+	}
+
+	hf->echo_id = idx;
+	hf->channel = dev->channel;
+
+	cf = (struct can_frame *)skb->data;
+
+	hf->can_id = cf->can_id;
+	hf->can_dlc = cf->can_dlc;
+	memcpy(hf->data, cf->data, cf->can_dlc);
+
+	usb_fill_bulk_urb(urb, dev->udev,
+			  usb_sndbulkpipe(dev->udev, GSUSB_ENDPOINT_OUT),
+			  hf,
+			  sizeof(*hf),
+			  gs_usb_xmit_callback,
+			  txc);
+
+	urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+	usb_anchor_urb(urb, &dev->tx_submitted);
+
+	can_put_echo_skb(skb, netdev, idx);
+
+	atomic_inc(&dev->active_tx_urbs);
+
+	rc = usb_submit_urb(urb, GFP_ATOMIC);
+	if (unlikely(rc)) {			/* usb send failed */
+		atomic_dec(&dev->active_tx_urbs);
+
+		can_free_echo_skb(netdev, idx);
+		gs_free_tx_context(txc);
+
+		usb_unanchor_urb(urb);
+		usb_free_coherent(dev->udev,
+				  sizeof(*hf),
+				  hf,
+				  urb->transfer_dma);
+
+
+		if (rc == -ENODEV) {
+			netif_device_detach(netdev);
+		} else {
+			netdev_err(netdev, "usb_submit failed (err=%d)\n", rc);
+			stats->tx_dropped++;
+		}
+	} else {
+		/* Slow down tx path */
+		if (atomic_read(&dev->active_tx_urbs) >= GS_MAX_TX_URBS)
+			netif_stop_queue(netdev);
+	}
+
+	/* let usb core take care of this urb */
+	usb_free_urb(urb);
+
+	return NETDEV_TX_OK;
+
+ badidx:
+	usb_free_coherent(dev->udev,
+			  sizeof(*hf),
+			  hf,
+			  urb->transfer_dma);
+ nomem_hf:
+	usb_free_urb(urb);
+
+ nomem_urb:
+	gs_free_tx_context(txc);
+	dev_kfree_skb(skb);
+	stats->tx_dropped++;
+	return NETDEV_TX_OK;
+}
+
+static int gs_can_open(struct net_device *netdev)
+{
+	struct gs_can *dev = netdev_priv(netdev);
+	struct gs_usb *parent = dev->parent;
+	int rc, i;
+	struct gs_device_mode *dm;
+	u32 ctrlmode;
+
+	rc = open_candev(netdev);
+	if (rc)
+		return rc;
+
+	if (atomic_add_return(1, &parent->active_channels) == 1) {
+		for (i = 0; i < GS_MAX_RX_URBS; i++) {
+			struct urb *urb;
+			u8 *buf;
+
+			/* alloc rx urb */
+			urb = usb_alloc_urb(0, GFP_KERNEL);
+			if (!urb) {
+				netdev_err(netdev,
+					   "No memory left for URB\n");
+				return -ENOMEM;
+			}
+
+			/* alloc rx buffer */
+			buf = usb_alloc_coherent(dev->udev,
+						 sizeof(struct gs_host_frame),
+						 GFP_KERNEL,
+						 &urb->transfer_dma);
+			if (!buf) {
+				netdev_err(netdev,
+					   "No memory left for USB buffer\n");
+				usb_free_urb(urb);
+				return -ENOMEM;
+			}
+
+			/* fill, anchor, and submit rx urb */
+			usb_fill_bulk_urb(urb,
+					  dev->udev,
+					  usb_rcvbulkpipe(dev->udev,
+							  GSUSB_ENDPOINT_IN),
+					  buf,
+					  sizeof(struct gs_host_frame),
+					  gs_usb_recieve_bulk_callback,
+					  parent);
+			urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+			usb_anchor_urb(urb, &parent->rx_submitted);
+
+			rc = usb_submit_urb(urb, GFP_KERNEL);
+			if (rc) {
+				if (rc == -ENODEV)
+					netif_device_detach(dev->netdev);
+
+				netdev_err(netdev,
+					   "usb_submit failed (err=%d)\n",
+					   rc);
+
+				usb_unanchor_urb(urb);
+				break;
+			}
+
+			/* Drop reference,
+			 * USB core will take care of freeing it
+			 */
+			usb_free_urb(urb);
+		}
+	}
+
+	dm = kmalloc(sizeof(*dm), GFP_KERNEL);
+	if (!dm)
+		return -ENOMEM;
+
+	/* flags */
+	ctrlmode = dev->can.ctrlmode;
+	dm->flags = 0;
+
+	if (ctrlmode & CAN_CTRLMODE_LOOPBACK)
+		dm->flags |= GS_CAN_MODE_LOOP_BACK;
+	else if (ctrlmode & CAN_CTRLMODE_LISTENONLY)
+		dm->flags |= GS_CAN_MODE_LISTEN_ONLY;
+
+	/* Controller is not allowed to retry TX
+	 * this mode is unavailable on atmels uc3c hardware
+	 */
+	if (ctrlmode & CAN_CTRLMODE_ONE_SHOT)
+		dm->flags |= GS_CAN_MODE_ONE_SHOT;
+
+	if (ctrlmode & CAN_CTRLMODE_3_SAMPLES)
+		dm->flags |= GS_CAN_MODE_TRIPLE_SAMPLE;
+
+	/* finally start device */
+	dm->mode = GS_CAN_MODE_START;
+	rc = usb_control_msg(interface_to_usbdev(dev->iface),
+			     usb_sndctrlpipe(interface_to_usbdev(dev->iface), 0),
+			     GS_USB_BREQ_MODE,
+			     USB_DIR_OUT|USB_TYPE_VENDOR|USB_RECIP_INTERFACE,
+			     dev->channel,
+			     0,
+			     dm,
+			     sizeof(*dm),
+			     1000);
+
+	if (rc < 0) {
+		netdev_err(netdev, "Couldn't start device (err=%d)\n", rc);
+		kfree(dm);
+		return rc;
+	}
+
+	kfree(dm);
+
+	dev->can.state = CAN_STATE_ERROR_ACTIVE;
+
+	if (!(dev->can.ctrlmode & CAN_CTRLMODE_LISTENONLY))
+		netif_start_queue(netdev);
+
+	return 0;
+}
+
+static int gs_can_close(struct net_device *netdev)
+{
+	int rc;
+	struct gs_can *dev = netdev_priv(netdev);
+	struct gs_usb *parent = dev->parent;
+
+	netif_stop_queue(netdev);
+
+	/* Stop polling */
+	if (atomic_dec_and_test(&parent->active_channels))
+		usb_kill_anchored_urbs(&parent->rx_submitted);
+
+	/* Stop sending URBs */
+	usb_kill_anchored_urbs(&dev->tx_submitted);
+	atomic_set(&dev->active_tx_urbs, 0);
+
+	/* reset the device */
+	rc = gs_cmd_reset(parent, dev);
+	if (rc < 0)
+		netdev_warn(netdev, "Couldn't shutdown device (err=%d)", rc);
+
+	/* reset tx contexts */
+	for (rc = 0; rc < GS_MAX_TX_URBS; rc++) {
+		dev->tx_context[rc].dev = dev;
+		dev->tx_context[rc].echo_id = GS_MAX_TX_URBS;
+	}
+
+	/* close the netdev */
+	close_candev(netdev);
+
+	return 0;
+}
+
+static const struct net_device_ops gs_usb_netdev_ops = {
+	.ndo_open = gs_can_open,
+	.ndo_stop = gs_can_close,
+	.ndo_start_xmit = gs_can_start_xmit,
+};
+
+static struct gs_can *gs_make_candev(unsigned int channel, struct usb_interface *intf)
+{
+	struct gs_can *dev;
+	struct net_device *netdev;
+	int rc;
+	struct gs_device_bt_const *bt_const;
+
+	bt_const = kmalloc(sizeof(*bt_const), GFP_KERNEL);
+	if (!bt_const)
+		return ERR_PTR(-ENOMEM);
+
+	/* fetch bit timing constants */
+	rc = usb_control_msg(interface_to_usbdev(intf),
+			     usb_rcvctrlpipe(interface_to_usbdev(intf), 0),
+			     GS_USB_BREQ_BT_CONST,
+			     USB_DIR_IN|USB_TYPE_VENDOR|USB_RECIP_INTERFACE,
+			     channel,
+			     0,
+			     bt_const,
+			     sizeof(*bt_const),
+			     1000);
+
+	if (rc < 0) {
+		dev_err(&intf->dev,
+			"Couldn't get bit timing const for channel (err=%d)\n",
+			rc);
+		kfree(bt_const);
+		return ERR_PTR(rc);
+	}
+
+	/* create netdev */
+	netdev = alloc_candev(sizeof(struct gs_can), GS_MAX_TX_URBS);
+	if (!netdev) {
+		dev_err(&intf->dev, "Couldn't allocate candev\n");
+		kfree(bt_const);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	dev = netdev_priv(netdev);
+
+	netdev->netdev_ops = &gs_usb_netdev_ops;
+
+	netdev->flags |= IFF_ECHO; /* we support full roundtrip echo */
+
+	/* dev settup */
+	strcpy(dev->bt_const.name, "gs_usb");
+	dev->bt_const.tseg1_min = bt_const->tseg1_min;
+	dev->bt_const.tseg1_max = bt_const->tseg1_max;
+	dev->bt_const.tseg2_min = bt_const->tseg2_min;
+	dev->bt_const.tseg2_max = bt_const->tseg2_max;
+	dev->bt_const.sjw_max = bt_const->sjw_max;
+	dev->bt_const.brp_min = bt_const->brp_min;
+	dev->bt_const.brp_max = bt_const->brp_max;
+	dev->bt_const.brp_inc = bt_const->brp_inc;
+
+	dev->udev = interface_to_usbdev(intf);
+	dev->iface = intf;
+	dev->netdev = netdev;
+	dev->channel = channel;
+
+	init_usb_anchor(&dev->tx_submitted);
+	atomic_set(&dev->active_tx_urbs, 0);
+	spin_lock_init(&dev->tx_ctx_lock);
+	for (rc = 0; rc < GS_MAX_TX_URBS; rc++) {
+		dev->tx_context[rc].dev = dev;
+		dev->tx_context[rc].echo_id = GS_MAX_TX_URBS;
+	}
+
+	/* can settup */
+	dev->can.state = CAN_STATE_STOPPED;
+	dev->can.clock.freq = bt_const->fclk_can;
+	dev->can.bittiming_const = &dev->bt_const;
+	dev->can.do_set_bittiming = gs_usb_set_bittiming;
+
+	dev->can.ctrlmode_supported = 0;
+
+	if (bt_const->feature & GS_CAN_FEATURE_LISTEN_ONLY)
+		dev->can.ctrlmode_supported |= CAN_CTRLMODE_LISTENONLY;
+
+	if (bt_const->feature & GS_CAN_FEATURE_LOOP_BACK)
+		dev->can.ctrlmode_supported |= CAN_CTRLMODE_LOOPBACK;
+
+	if (bt_const->feature & GS_CAN_FEATURE_TRIPLE_SAMPLE)
+		dev->can.ctrlmode_supported |= CAN_CTRLMODE_3_SAMPLES;
+
+	if (bt_const->feature & GS_CAN_FEATURE_ONE_SHOT)
+		dev->can.ctrlmode_supported |= CAN_CTRLMODE_ONE_SHOT;
+
+	kfree(bt_const);
+
+	SET_NETDEV_DEV(netdev, &intf->dev);
+
+	rc = register_candev(dev->netdev);
+	if (rc) {
+		free_candev(dev->netdev);
+		dev_err(&intf->dev, "Couldn't register candev (err=%d)\n", rc);
+		return ERR_PTR(rc);
+	}
+
+	return dev;
+}
+
+static void gs_destroy_candev(struct gs_can *dev)
+{
+	unregister_candev(dev->netdev);
+	free_candev(dev->netdev);
+	usb_kill_anchored_urbs(&dev->tx_submitted);
+	kfree(dev);
+}
+
+static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
+{
+	struct gs_usb *dev;
+	int rc = -ENOMEM;
+	unsigned int icount, i;
+	struct gs_host_config *hconf;
+	struct gs_device_config *dconf;
+
+	hconf = kmalloc(sizeof(*hconf), GFP_KERNEL);
+	if (!hconf)
+		return -ENOMEM;
+
+	hconf->byte_order = 0x0000beef;
+
+	/* send host config */
+	rc = usb_control_msg(interface_to_usbdev(intf),
+			     usb_sndctrlpipe(interface_to_usbdev(intf), 0),
+			     GS_USB_BREQ_HOST_FORMAT,
+			     USB_DIR_OUT|USB_TYPE_VENDOR|USB_RECIP_INTERFACE,
+			     1,
+			     intf->altsetting[0].desc.bInterfaceNumber,
+			     hconf,
+			     sizeof(*hconf),
+			     1000);
+
+	kfree(hconf);
+
+	if (rc < 0) {
+		dev_err(&intf->dev, "Couldn't send data format (err=%d)\n",
+			rc);
+		return rc;
+	}
+
+	dconf = kmalloc(sizeof(*dconf), GFP_KERNEL);
+	if (!dconf)
+		return -ENOMEM;
+
+	/* read device config */
+	rc = usb_control_msg(interface_to_usbdev(intf),
+			     usb_rcvctrlpipe(interface_to_usbdev(intf), 0),
+			     GS_USB_BREQ_DEVICE_CONFIG,
+			     USB_DIR_IN|USB_TYPE_VENDOR|USB_RECIP_INTERFACE,
+			     1,
+			     intf->altsetting[0].desc.bInterfaceNumber,
+			     dconf,
+			     sizeof(*dconf),
+			     1000);
+	if (rc < 0) {
+		dev_err(&intf->dev, "Couldn't get device config: (err=%d)\n",
+			rc);
+
+		kfree(dconf);
+
+		return rc;
+	}
+
+	icount = dconf->icount+1;
+
+	kfree(dconf);
+
+	dev_info(&intf->dev, "Configuring for %d interfaces\n", icount);
+
+	if (icount > GS_MAX_INTF) {
+		dev_err(&intf->dev,
+			"Driver cannot handle more that %d CAN interfaces\n",
+			GS_MAX_INTF);
+		return -EINVAL;
+	}
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	init_usb_anchor(&dev->rx_submitted);
+
+	atomic_set(&dev->active_channels, 0);
+
+	usb_set_intfdata(intf, dev);
+	dev->udev = interface_to_usbdev(intf);
+
+	for (i = 0; i < icount; i++) {
+		dev->canch[i] = gs_make_candev(i, intf);
+		if (IS_ERR_OR_NULL(dev->canch[i])) {
+			/* on failure destroy previously created candevs */
+			icount = i;
+			for (i = 0; i < icount; i++) {
+				gs_destroy_candev(dev->canch[i]);
+				dev->canch[i] = NULL;
+			}
+			kfree(dev);
+			return rc;
+		}
+		dev->canch[i]->parent = dev;
+	}
+
+	return 0;
+}
+
+static void gs_usb_disconnect(struct usb_interface *intf)
+{
+	unsigned i;
+	struct gs_usb *dev = usb_get_intfdata(intf);
+	usb_set_intfdata(intf, NULL);
+
+	if (!dev) {
+		dev_err(&intf->dev, "Disconnect (nodata)\n");
+		return;
+	}
+
+	for (i = 0; i < GS_MAX_INTF; i++) {
+		struct gs_can *can = dev->canch[i];
+
+		if (!can)
+			continue;
+
+		gs_destroy_candev(can);
+	}
+
+	usb_kill_anchored_urbs(&dev->rx_submitted);
+}
+
+static const struct usb_device_id gs_usb_table[] = {
+	{USB_DEVICE(USB_GSUSB_1_VENDOR_ID, USB_GSUSB_1_PRODUCT_ID)},
+	{} /* Terminating entry */
+};
+
+MODULE_DEVICE_TABLE(usb, gs_usb_table);
+
+static struct usb_driver gs_usb_driver = {
+	.name       = "gs_usb",
+	.probe      = gs_usb_probe,
+	.disconnect = gs_usb_disconnect,
+	.id_table   = gs_usb_table,
+};
+
+module_usb_driver(gs_usb_driver);
+
+MODULE_AUTHOR("Maximilian Schneider <mws@schneidersoft.net>");
+MODULE_DESCRIPTION(
+"Socket CAN device driver for Geschwister Schneider Technologie-, "
+"Entwicklungs- und Vertriebs UG. USB2.0 to CAN interfaces.");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/can/usb/kvaser_usb.c b/drivers/net/can/usb/kvaser_usb.c
index 4ca46edc061d..541fb7a05625 100644
--- a/drivers/net/can/usb/kvaser_usb.c
+++ b/drivers/net/can/usb/kvaser_usb.c
@@ -53,6 +53,8 @@
 #define USB_OEM_MERCURY_PRODUCT_ID	34
 #define USB_OEM_LEAF_PRODUCT_ID		35
 #define USB_CAN_R_PRODUCT_ID		39
+#define USB_LEAF_LITE_V2_PRODUCT_ID	288
+#define USB_MINI_PCIE_HS_PRODUCT_ID	289
 
 /* USB devices features */
 #define KVASER_HAS_SILENT_MODE		BIT(0)
@@ -356,6 +358,8 @@ static const struct usb_device_id kvaser_usb_table[] = {
 		.driver_info = KVASER_HAS_TXRX_ERRORS },
 	{ USB_DEVICE(KVASER_VENDOR_ID, USB_CAN_R_PRODUCT_ID),
 		.driver_info = KVASER_HAS_TXRX_ERRORS },
+	{ USB_DEVICE(KVASER_VENDOR_ID, USB_LEAF_LITE_V2_PRODUCT_ID) },
+	{ USB_DEVICE(KVASER_VENDOR_ID, USB_MINI_PCIE_HS_PRODUCT_ID) },
 	{ }
 };
 MODULE_DEVICE_TABLE(usb, kvaser_usb_table);
@@ -379,38 +383,43 @@ static int kvaser_usb_wait_msg(const struct kvaser_usb *dev, u8 id,
 	void *buf;
 	int actual_len;
 	int err;
-	int pos = 0;
+	int pos;
+	unsigned long to = jiffies + msecs_to_jiffies(USB_RECV_TIMEOUT);
 
 	buf = kzalloc(RX_BUFFER_SIZE, GFP_KERNEL);
 	if (!buf)
 		return -ENOMEM;
 
-	err = usb_bulk_msg(dev->udev,
-			   usb_rcvbulkpipe(dev->udev,
-					   dev->bulk_in->bEndpointAddress),
-			   buf, RX_BUFFER_SIZE, &actual_len,
-			   USB_RECV_TIMEOUT);
-	if (err < 0)
-		goto end;
+	do {
+		err = usb_bulk_msg(dev->udev,
+				   usb_rcvbulkpipe(dev->udev,
+					dev->bulk_in->bEndpointAddress),
+				   buf, RX_BUFFER_SIZE, &actual_len,
+				   USB_RECV_TIMEOUT);
+		if (err < 0)
+			goto end;
 
-	while (pos <= actual_len - MSG_HEADER_LEN) {
-		tmp = buf + pos;
+		pos = 0;
+		while (pos <= actual_len - MSG_HEADER_LEN) {
+			tmp = buf + pos;
 
-		if (!tmp->len)
-			break;
+			if (!tmp->len)
+				break;
 
-		if (pos + tmp->len > actual_len) {
-			dev_err(dev->udev->dev.parent, "Format error\n");
-			break;
-		}
+			if (pos + tmp->len > actual_len) {
+				dev_err(dev->udev->dev.parent,
+					"Format error\n");
+				break;
+			}
 
-		if (tmp->id == id) {
-			memcpy(msg, tmp, tmp->len);
-			goto end;
-		}
+			if (tmp->id == id) {
+				memcpy(msg, tmp, tmp->len);
+				goto end;
+			}
 
-		pos += tmp->len;
-	}
+			pos += tmp->len;
+		}
+	} while (time_before(jiffies, to));
 
 	err = -EINVAL;
 
diff --git a/drivers/net/can/xilinx_can.c b/drivers/net/can/xilinx_can.c
new file mode 100644
index 000000000000..5e8b5609c067
--- /dev/null
+++ b/drivers/net/can/xilinx_can.c
@@ -0,0 +1,1208 @@
+/* Xilinx CAN device driver
+ *
+ * Copyright (C) 2012 - 2014 Xilinx, Inc.
+ * Copyright (C) 2009 PetaLogix. All rights reserved.
+ *
+ * Description:
+ * This driver is developed for Axi CAN IP and for Zynq CANPS Controller.
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clk.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/skbuff.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/can/dev.h>
+#include <linux/can/error.h>
+#include <linux/can/led.h>
+
+#define DRIVER_NAME	"xilinx_can"
+
+/* CAN registers set */
+enum xcan_reg {
+	XCAN_SRR_OFFSET		= 0x00, /* Software reset */
+	XCAN_MSR_OFFSET		= 0x04, /* Mode select */
+	XCAN_BRPR_OFFSET	= 0x08, /* Baud rate prescaler */
+	XCAN_BTR_OFFSET		= 0x0C, /* Bit timing */
+	XCAN_ECR_OFFSET		= 0x10, /* Error counter */
+	XCAN_ESR_OFFSET		= 0x14, /* Error status */
+	XCAN_SR_OFFSET		= 0x18, /* Status */
+	XCAN_ISR_OFFSET		= 0x1C, /* Interrupt status */
+	XCAN_IER_OFFSET		= 0x20, /* Interrupt enable */
+	XCAN_ICR_OFFSET		= 0x24, /* Interrupt clear */
+	XCAN_TXFIFO_ID_OFFSET	= 0x30,/* TX FIFO ID */
+	XCAN_TXFIFO_DLC_OFFSET	= 0x34, /* TX FIFO DLC */
+	XCAN_TXFIFO_DW1_OFFSET	= 0x38, /* TX FIFO Data Word 1 */
+	XCAN_TXFIFO_DW2_OFFSET	= 0x3C, /* TX FIFO Data Word 2 */
+	XCAN_RXFIFO_ID_OFFSET	= 0x50, /* RX FIFO ID */
+	XCAN_RXFIFO_DLC_OFFSET	= 0x54, /* RX FIFO DLC */
+	XCAN_RXFIFO_DW1_OFFSET	= 0x58, /* RX FIFO Data Word 1 */
+	XCAN_RXFIFO_DW2_OFFSET	= 0x5C, /* RX FIFO Data Word 2 */
+};
+
+/* CAN register bit masks - XCAN_<REG>_<BIT>_MASK */
+#define XCAN_SRR_CEN_MASK		0x00000002 /* CAN enable */
+#define XCAN_SRR_RESET_MASK		0x00000001 /* Soft Reset the CAN core */
+#define XCAN_MSR_LBACK_MASK		0x00000002 /* Loop back mode select */
+#define XCAN_MSR_SLEEP_MASK		0x00000001 /* Sleep mode select */
+#define XCAN_BRPR_BRP_MASK		0x000000FF /* Baud rate prescaler */
+#define XCAN_BTR_SJW_MASK		0x00000180 /* Synchronous jump width */
+#define XCAN_BTR_TS2_MASK		0x00000070 /* Time segment 2 */
+#define XCAN_BTR_TS1_MASK		0x0000000F /* Time segment 1 */
+#define XCAN_ECR_REC_MASK		0x0000FF00 /* Receive error counter */
+#define XCAN_ECR_TEC_MASK		0x000000FF /* Transmit error counter */
+#define XCAN_ESR_ACKER_MASK		0x00000010 /* ACK error */
+#define XCAN_ESR_BERR_MASK		0x00000008 /* Bit error */
+#define XCAN_ESR_STER_MASK		0x00000004 /* Stuff error */
+#define XCAN_ESR_FMER_MASK		0x00000002 /* Form error */
+#define XCAN_ESR_CRCER_MASK		0x00000001 /* CRC error */
+#define XCAN_SR_TXFLL_MASK		0x00000400 /* TX FIFO is full */
+#define XCAN_SR_ESTAT_MASK		0x00000180 /* Error status */
+#define XCAN_SR_ERRWRN_MASK		0x00000040 /* Error warning */
+#define XCAN_SR_NORMAL_MASK		0x00000008 /* Normal mode */
+#define XCAN_SR_LBACK_MASK		0x00000002 /* Loop back mode */
+#define XCAN_SR_CONFIG_MASK		0x00000001 /* Configuration mode */
+#define XCAN_IXR_TXFEMP_MASK		0x00004000 /* TX FIFO Empty */
+#define XCAN_IXR_WKUP_MASK		0x00000800 /* Wake up interrupt */
+#define XCAN_IXR_SLP_MASK		0x00000400 /* Sleep interrupt */
+#define XCAN_IXR_BSOFF_MASK		0x00000200 /* Bus off interrupt */
+#define XCAN_IXR_ERROR_MASK		0x00000100 /* Error interrupt */
+#define XCAN_IXR_RXNEMP_MASK		0x00000080 /* RX FIFO NotEmpty intr */
+#define XCAN_IXR_RXOFLW_MASK		0x00000040 /* RX FIFO Overflow intr */
+#define XCAN_IXR_RXOK_MASK		0x00000010 /* Message received intr */
+#define XCAN_IXR_TXFLL_MASK		0x00000004 /* Tx FIFO Full intr */
+#define XCAN_IXR_TXOK_MASK		0x00000002 /* TX successful intr */
+#define XCAN_IXR_ARBLST_MASK		0x00000001 /* Arbitration lost intr */
+#define XCAN_IDR_ID1_MASK		0xFFE00000 /* Standard msg identifier */
+#define XCAN_IDR_SRR_MASK		0x00100000 /* Substitute remote TXreq */
+#define XCAN_IDR_IDE_MASK		0x00080000 /* Identifier extension */
+#define XCAN_IDR_ID2_MASK		0x0007FFFE /* Extended message ident */
+#define XCAN_IDR_RTR_MASK		0x00000001 /* Remote TX request */
+#define XCAN_DLCR_DLC_MASK		0xF0000000 /* Data length code */
+
+#define XCAN_INTR_ALL		(XCAN_IXR_TXOK_MASK | XCAN_IXR_BSOFF_MASK |\
+				 XCAN_IXR_WKUP_MASK | XCAN_IXR_SLP_MASK | \
+				 XCAN_IXR_RXNEMP_MASK | XCAN_IXR_ERROR_MASK | \
+				 XCAN_IXR_ARBLST_MASK | XCAN_IXR_RXOK_MASK)
+
+/* CAN register bit shift - XCAN_<REG>_<BIT>_SHIFT */
+#define XCAN_BTR_SJW_SHIFT		7  /* Synchronous jump width */
+#define XCAN_BTR_TS2_SHIFT		4  /* Time segment 2 */
+#define XCAN_IDR_ID1_SHIFT		21 /* Standard Messg Identifier */
+#define XCAN_IDR_ID2_SHIFT		1  /* Extended Message Identifier */
+#define XCAN_DLCR_DLC_SHIFT		28 /* Data length code */
+#define XCAN_ESR_REC_SHIFT		8  /* Rx Error Count */
+
+/* CAN frame length constants */
+#define XCAN_FRAME_MAX_DATA_LEN		8
+#define XCAN_TIMEOUT			(1 * HZ)
+
+/**
+ * struct xcan_priv - This definition define CAN driver instance
+ * @can:			CAN private data structure.
+ * @tx_head:			Tx CAN packets ready to send on the queue
+ * @tx_tail:			Tx CAN packets successfully sended on the queue
+ * @tx_max:			Maximum number packets the driver can send
+ * @napi:			NAPI structure
+ * @read_reg:			For reading data from CAN registers
+ * @write_reg:			For writing data to CAN registers
+ * @dev:			Network device data structure
+ * @reg_base:			Ioremapped address to registers
+ * @irq_flags:			For request_irq()
+ * @bus_clk:			Pointer to struct clk
+ * @can_clk:			Pointer to struct clk
+ */
+struct xcan_priv {
+	struct can_priv can;
+	unsigned int tx_head;
+	unsigned int tx_tail;
+	unsigned int tx_max;
+	struct napi_struct napi;
+	u32 (*read_reg)(const struct xcan_priv *priv, enum xcan_reg reg);
+	void (*write_reg)(const struct xcan_priv *priv, enum xcan_reg reg,
+			u32 val);
+	struct net_device *dev;
+	void __iomem *reg_base;
+	unsigned long irq_flags;
+	struct clk *bus_clk;
+	struct clk *can_clk;
+};
+
+/* CAN Bittiming constants as per Xilinx CAN specs */
+static const struct can_bittiming_const xcan_bittiming_const = {
+	.name = DRIVER_NAME,
+	.tseg1_min = 1,
+	.tseg1_max = 16,
+	.tseg2_min = 1,
+	.tseg2_max = 8,
+	.sjw_max = 4,
+	.brp_min = 1,
+	.brp_max = 256,
+	.brp_inc = 1,
+};
+
+/**
+ * xcan_write_reg_le - Write a value to the device register little endian
+ * @priv:	Driver private data structure
+ * @reg:	Register offset
+ * @val:	Value to write at the Register offset
+ *
+ * Write data to the paricular CAN register
+ */
+static void xcan_write_reg_le(const struct xcan_priv *priv, enum xcan_reg reg,
+			u32 val)
+{
+	iowrite32(val, priv->reg_base + reg);
+}
+
+/**
+ * xcan_read_reg_le - Read a value from the device register little endian
+ * @priv:	Driver private data structure
+ * @reg:	Register offset
+ *
+ * Read data from the particular CAN register
+ * Return: value read from the CAN register
+ */
+static u32 xcan_read_reg_le(const struct xcan_priv *priv, enum xcan_reg reg)
+{
+	return ioread32(priv->reg_base + reg);
+}
+
+/**
+ * xcan_write_reg_be - Write a value to the device register big endian
+ * @priv:	Driver private data structure
+ * @reg:	Register offset
+ * @val:	Value to write at the Register offset
+ *
+ * Write data to the paricular CAN register
+ */
+static void xcan_write_reg_be(const struct xcan_priv *priv, enum xcan_reg reg,
+			u32 val)
+{
+	iowrite32be(val, priv->reg_base + reg);
+}
+
+/**
+ * xcan_read_reg_be - Read a value from the device register big endian
+ * @priv:	Driver private data structure
+ * @reg:	Register offset
+ *
+ * Read data from the particular CAN register
+ * Return: value read from the CAN register
+ */
+static u32 xcan_read_reg_be(const struct xcan_priv *priv, enum xcan_reg reg)
+{
+	return ioread32be(priv->reg_base + reg);
+}
+
+/**
+ * set_reset_mode - Resets the CAN device mode
+ * @ndev:	Pointer to net_device structure
+ *
+ * This is the driver reset mode routine.The driver
+ * enters into configuration mode.
+ *
+ * Return: 0 on success and failure value on error
+ */
+static int set_reset_mode(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	unsigned long timeout;
+
+	priv->write_reg(priv, XCAN_SRR_OFFSET, XCAN_SRR_RESET_MASK);
+
+	timeout = jiffies + XCAN_TIMEOUT;
+	while (!(priv->read_reg(priv, XCAN_SR_OFFSET) & XCAN_SR_CONFIG_MASK)) {
+		if (time_after(jiffies, timeout)) {
+			netdev_warn(ndev, "timed out for config mode\n");
+			return -ETIMEDOUT;
+		}
+		usleep_range(500, 10000);
+	}
+
+	return 0;
+}
+
+/**
+ * xcan_set_bittiming - CAN set bit timing routine
+ * @ndev:	Pointer to net_device structure
+ *
+ * This is the driver set bittiming  routine.
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_set_bittiming(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	struct can_bittiming *bt = &priv->can.bittiming;
+	u32 btr0, btr1;
+	u32 is_config_mode;
+
+	/* Check whether Xilinx CAN is in configuration mode.
+	 * It cannot set bit timing if Xilinx CAN is not in configuration mode.
+	 */
+	is_config_mode = priv->read_reg(priv, XCAN_SR_OFFSET) &
+				XCAN_SR_CONFIG_MASK;
+	if (!is_config_mode) {
+		netdev_alert(ndev,
+		     "BUG! Cannot set bittiming - CAN is not in config mode\n");
+		return -EPERM;
+	}
+
+	/* Setting Baud Rate prescalar value in BRPR Register */
+	btr0 = (bt->brp - 1);
+
+	/* Setting Time Segment 1 in BTR Register */
+	btr1 = (bt->prop_seg + bt->phase_seg1 - 1);
+
+	/* Setting Time Segment 2 in BTR Register */
+	btr1 |= (bt->phase_seg2 - 1) << XCAN_BTR_TS2_SHIFT;
+
+	/* Setting Synchronous jump width in BTR Register */
+	btr1 |= (bt->sjw - 1) << XCAN_BTR_SJW_SHIFT;
+
+	priv->write_reg(priv, XCAN_BRPR_OFFSET, btr0);
+	priv->write_reg(priv, XCAN_BTR_OFFSET, btr1);
+
+	netdev_dbg(ndev, "BRPR=0x%08x, BTR=0x%08x\n",
+			priv->read_reg(priv, XCAN_BRPR_OFFSET),
+			priv->read_reg(priv, XCAN_BTR_OFFSET));
+
+	return 0;
+}
+
+/**
+ * xcan_chip_start - This the drivers start routine
+ * @ndev:	Pointer to net_device structure
+ *
+ * This is the drivers start routine.
+ * Based on the State of the CAN device it puts
+ * the CAN device into a proper mode.
+ *
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_chip_start(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	u32 err, reg_msr, reg_sr_mask;
+	unsigned long timeout;
+
+	/* Check if it is in reset mode */
+	err = set_reset_mode(ndev);
+	if (err < 0)
+		return err;
+
+	err = xcan_set_bittiming(ndev);
+	if (err < 0)
+		return err;
+
+	/* Enable interrupts */
+	priv->write_reg(priv, XCAN_IER_OFFSET, XCAN_INTR_ALL);
+
+	/* Check whether it is loopback mode or normal mode  */
+	if (priv->can.ctrlmode & CAN_CTRLMODE_LOOPBACK) {
+		reg_msr = XCAN_MSR_LBACK_MASK;
+		reg_sr_mask = XCAN_SR_LBACK_MASK;
+	} else {
+		reg_msr = 0x0;
+		reg_sr_mask = XCAN_SR_NORMAL_MASK;
+	}
+
+	priv->write_reg(priv, XCAN_MSR_OFFSET, reg_msr);
+	priv->write_reg(priv, XCAN_SRR_OFFSET, XCAN_SRR_CEN_MASK);
+
+	timeout = jiffies + XCAN_TIMEOUT;
+	while (!(priv->read_reg(priv, XCAN_SR_OFFSET) & reg_sr_mask)) {
+		if (time_after(jiffies, timeout)) {
+			netdev_warn(ndev,
+				"timed out for correct mode\n");
+			return -ETIMEDOUT;
+		}
+	}
+	netdev_dbg(ndev, "status:#x%08x\n",
+			priv->read_reg(priv, XCAN_SR_OFFSET));
+
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+	return 0;
+}
+
+/**
+ * xcan_do_set_mode - This sets the mode of the driver
+ * @ndev:	Pointer to net_device structure
+ * @mode:	Tells the mode of the driver
+ *
+ * This check the drivers state and calls the
+ * the corresponding modes to set.
+ *
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_do_set_mode(struct net_device *ndev, enum can_mode mode)
+{
+	int ret;
+
+	switch (mode) {
+	case CAN_MODE_START:
+		ret = xcan_chip_start(ndev);
+		if (ret < 0) {
+			netdev_err(ndev, "xcan_chip_start failed!\n");
+			return ret;
+		}
+		netif_wake_queue(ndev);
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	return ret;
+}
+
+/**
+ * xcan_start_xmit - Starts the transmission
+ * @skb:	sk_buff pointer that contains data to be Txed
+ * @ndev:	Pointer to net_device structure
+ *
+ * This function is invoked from upper layers to initiate transmission. This
+ * function uses the next available free txbuff and populates their fields to
+ * start the transmission.
+ *
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	struct can_frame *cf = (struct can_frame *)skb->data;
+	u32 id, dlc, data[2] = {0, 0};
+
+	if (can_dropped_invalid_skb(ndev, skb))
+		return NETDEV_TX_OK;
+
+	/* Check if the TX buffer is full */
+	if (unlikely(priv->read_reg(priv, XCAN_SR_OFFSET) &
+			XCAN_SR_TXFLL_MASK)) {
+		netif_stop_queue(ndev);
+		netdev_err(ndev, "BUG!, TX FIFO full when queue awake!\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	/* Watch carefully on the bit sequence */
+	if (cf->can_id & CAN_EFF_FLAG) {
+		/* Extended CAN ID format */
+		id = ((cf->can_id & CAN_EFF_MASK) << XCAN_IDR_ID2_SHIFT) &
+			XCAN_IDR_ID2_MASK;
+		id |= (((cf->can_id & CAN_EFF_MASK) >>
+			(CAN_EFF_ID_BITS-CAN_SFF_ID_BITS)) <<
+			XCAN_IDR_ID1_SHIFT) & XCAN_IDR_ID1_MASK;
+
+		/* The substibute remote TX request bit should be "1"
+		 * for extended frames as in the Xilinx CAN datasheet
+		 */
+		id |= XCAN_IDR_IDE_MASK | XCAN_IDR_SRR_MASK;
+
+		if (cf->can_id & CAN_RTR_FLAG)
+			/* Extended frames remote TX request */
+			id |= XCAN_IDR_RTR_MASK;
+	} else {
+		/* Standard CAN ID format */
+		id = ((cf->can_id & CAN_SFF_MASK) << XCAN_IDR_ID1_SHIFT) &
+			XCAN_IDR_ID1_MASK;
+
+		if (cf->can_id & CAN_RTR_FLAG)
+			/* Standard frames remote TX request */
+			id |= XCAN_IDR_SRR_MASK;
+	}
+
+	dlc = cf->can_dlc << XCAN_DLCR_DLC_SHIFT;
+
+	if (cf->can_dlc > 0)
+		data[0] = be32_to_cpup((__be32 *)(cf->data + 0));
+	if (cf->can_dlc > 4)
+		data[1] = be32_to_cpup((__be32 *)(cf->data + 4));
+
+	can_put_echo_skb(skb, ndev, priv->tx_head % priv->tx_max);
+	priv->tx_head++;
+
+	/* Write the Frame to Xilinx CAN TX FIFO */
+	priv->write_reg(priv, XCAN_TXFIFO_ID_OFFSET, id);
+	/* If the CAN frame is RTR frame this write triggers tranmission */
+	priv->write_reg(priv, XCAN_TXFIFO_DLC_OFFSET, dlc);
+	if (!(cf->can_id & CAN_RTR_FLAG)) {
+		priv->write_reg(priv, XCAN_TXFIFO_DW1_OFFSET, data[0]);
+		/* If the CAN frame is Standard/Extended frame this
+		 * write triggers tranmission
+		 */
+		priv->write_reg(priv, XCAN_TXFIFO_DW2_OFFSET, data[1]);
+		stats->tx_bytes += cf->can_dlc;
+	}
+
+	/* Check if the TX buffer is full */
+	if ((priv->tx_head - priv->tx_tail) == priv->tx_max)
+		netif_stop_queue(ndev);
+
+	return NETDEV_TX_OK;
+}
+
+/**
+ * xcan_rx -  Is called from CAN isr to complete the received
+ *		frame  processing
+ * @ndev:	Pointer to net_device structure
+ *
+ * This function is invoked from the CAN isr(poll) to process the Rx frames. It
+ * does minimal processing and invokes "netif_receive_skb" to complete further
+ * processing.
+ * Return: 1 on success and 0 on failure.
+ */
+static int xcan_rx(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+	u32 id_xcan, dlc, data[2] = {0, 0};
+
+	skb = alloc_can_skb(ndev, &cf);
+	if (unlikely(!skb)) {
+		stats->rx_dropped++;
+		return 0;
+	}
+
+	/* Read a frame from Xilinx zynq CANPS */
+	id_xcan = priv->read_reg(priv, XCAN_RXFIFO_ID_OFFSET);
+	dlc = priv->read_reg(priv, XCAN_RXFIFO_DLC_OFFSET) >>
+				XCAN_DLCR_DLC_SHIFT;
+
+	/* Change Xilinx CAN data length format to socketCAN data format */
+	cf->can_dlc = get_can_dlc(dlc);
+
+	/* Change Xilinx CAN ID format to socketCAN ID format */
+	if (id_xcan & XCAN_IDR_IDE_MASK) {
+		/* The received frame is an Extended format frame */
+		cf->can_id = (id_xcan & XCAN_IDR_ID1_MASK) >> 3;
+		cf->can_id |= (id_xcan & XCAN_IDR_ID2_MASK) >>
+				XCAN_IDR_ID2_SHIFT;
+		cf->can_id |= CAN_EFF_FLAG;
+		if (id_xcan & XCAN_IDR_RTR_MASK)
+			cf->can_id |= CAN_RTR_FLAG;
+	} else {
+		/* The received frame is a standard format frame */
+		cf->can_id = (id_xcan & XCAN_IDR_ID1_MASK) >>
+				XCAN_IDR_ID1_SHIFT;
+		if (id_xcan & XCAN_IDR_SRR_MASK)
+			cf->can_id |= CAN_RTR_FLAG;
+	}
+
+	if (!(id_xcan & XCAN_IDR_SRR_MASK)) {
+		data[0] = priv->read_reg(priv, XCAN_RXFIFO_DW1_OFFSET);
+		data[1] = priv->read_reg(priv, XCAN_RXFIFO_DW2_OFFSET);
+
+		/* Change Xilinx CAN data format to socketCAN data format */
+		if (cf->can_dlc > 0)
+			*(__be32 *)(cf->data) = cpu_to_be32(data[0]);
+		if (cf->can_dlc > 4)
+			*(__be32 *)(cf->data + 4) = cpu_to_be32(data[1]);
+	}
+
+	stats->rx_bytes += cf->can_dlc;
+	stats->rx_packets++;
+	netif_receive_skb(skb);
+
+	return 1;
+}
+
+/**
+ * xcan_err_interrupt - error frame Isr
+ * @ndev:	net_device pointer
+ * @isr:	interrupt status register value
+ *
+ * This is the CAN error interrupt and it will
+ * check the the type of error and forward the error
+ * frame to upper layers.
+ */
+static void xcan_err_interrupt(struct net_device *ndev, u32 isr)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+	u32 err_status, status, txerr = 0, rxerr = 0;
+
+	skb = alloc_can_err_skb(ndev, &cf);
+
+	err_status = priv->read_reg(priv, XCAN_ESR_OFFSET);
+	priv->write_reg(priv, XCAN_ESR_OFFSET, err_status);
+	txerr = priv->read_reg(priv, XCAN_ECR_OFFSET) & XCAN_ECR_TEC_MASK;
+	rxerr = ((priv->read_reg(priv, XCAN_ECR_OFFSET) &
+			XCAN_ECR_REC_MASK) >> XCAN_ESR_REC_SHIFT);
+	status = priv->read_reg(priv, XCAN_SR_OFFSET);
+
+	if (isr & XCAN_IXR_BSOFF_MASK) {
+		priv->can.state = CAN_STATE_BUS_OFF;
+		priv->can.can_stats.bus_off++;
+		/* Leave device in Config Mode in bus-off state */
+		priv->write_reg(priv, XCAN_SRR_OFFSET, XCAN_SRR_RESET_MASK);
+		can_bus_off(ndev);
+		if (skb)
+			cf->can_id |= CAN_ERR_BUSOFF;
+	} else if ((status & XCAN_SR_ESTAT_MASK) == XCAN_SR_ESTAT_MASK) {
+		priv->can.state = CAN_STATE_ERROR_PASSIVE;
+		priv->can.can_stats.error_passive++;
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[1] = (rxerr > 127) ?
+					CAN_ERR_CRTL_RX_PASSIVE :
+					CAN_ERR_CRTL_TX_PASSIVE;
+			cf->data[6] = txerr;
+			cf->data[7] = rxerr;
+		}
+	} else if (status & XCAN_SR_ERRWRN_MASK) {
+		priv->can.state = CAN_STATE_ERROR_WARNING;
+		priv->can.can_stats.error_warning++;
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[1] |= (txerr > rxerr) ?
+					CAN_ERR_CRTL_TX_WARNING :
+					CAN_ERR_CRTL_RX_WARNING;
+			cf->data[6] = txerr;
+			cf->data[7] = rxerr;
+		}
+	}
+
+	/* Check for Arbitration lost interrupt */
+	if (isr & XCAN_IXR_ARBLST_MASK) {
+		priv->can.can_stats.arbitration_lost++;
+		if (skb) {
+			cf->can_id |= CAN_ERR_LOSTARB;
+			cf->data[0] = CAN_ERR_LOSTARB_UNSPEC;
+		}
+	}
+
+	/* Check for RX FIFO Overflow interrupt */
+	if (isr & XCAN_IXR_RXOFLW_MASK) {
+		stats->rx_over_errors++;
+		stats->rx_errors++;
+		priv->write_reg(priv, XCAN_SRR_OFFSET, XCAN_SRR_RESET_MASK);
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[1] |= CAN_ERR_CRTL_RX_OVERFLOW;
+		}
+	}
+
+	/* Check for error interrupt */
+	if (isr & XCAN_IXR_ERROR_MASK) {
+		if (skb) {
+			cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
+			cf->data[2] |= CAN_ERR_PROT_UNSPEC;
+		}
+
+		/* Check for Ack error interrupt */
+		if (err_status & XCAN_ESR_ACKER_MASK) {
+			stats->tx_errors++;
+			if (skb) {
+				cf->can_id |= CAN_ERR_ACK;
+				cf->data[3] |= CAN_ERR_PROT_LOC_ACK;
+			}
+		}
+
+		/* Check for Bit error interrupt */
+		if (err_status & XCAN_ESR_BERR_MASK) {
+			stats->tx_errors++;
+			if (skb) {
+				cf->can_id |= CAN_ERR_PROT;
+				cf->data[2] = CAN_ERR_PROT_BIT;
+			}
+		}
+
+		/* Check for Stuff error interrupt */
+		if (err_status & XCAN_ESR_STER_MASK) {
+			stats->rx_errors++;
+			if (skb) {
+				cf->can_id |= CAN_ERR_PROT;
+				cf->data[2] = CAN_ERR_PROT_STUFF;
+			}
+		}
+
+		/* Check for Form error interrupt */
+		if (err_status & XCAN_ESR_FMER_MASK) {
+			stats->rx_errors++;
+			if (skb) {
+				cf->can_id |= CAN_ERR_PROT;
+				cf->data[2] = CAN_ERR_PROT_FORM;
+			}
+		}
+
+		/* Check for CRC error interrupt */
+		if (err_status & XCAN_ESR_CRCER_MASK) {
+			stats->rx_errors++;
+			if (skb) {
+				cf->can_id |= CAN_ERR_PROT;
+				cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ |
+						CAN_ERR_PROT_LOC_CRC_DEL;
+			}
+		}
+			priv->can.can_stats.bus_error++;
+	}
+
+	if (skb) {
+		stats->rx_packets++;
+		stats->rx_bytes += cf->can_dlc;
+		netif_rx(skb);
+	}
+
+	netdev_dbg(ndev, "%s: error status register:0x%x\n",
+			__func__, priv->read_reg(priv, XCAN_ESR_OFFSET));
+}
+
+/**
+ * xcan_state_interrupt - It will check the state of the CAN device
+ * @ndev:	net_device pointer
+ * @isr:	interrupt status register value
+ *
+ * This will checks the state of the CAN device
+ * and puts the device into appropriate state.
+ */
+static void xcan_state_interrupt(struct net_device *ndev, u32 isr)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+
+	/* Check for Sleep interrupt if set put CAN device in sleep state */
+	if (isr & XCAN_IXR_SLP_MASK)
+		priv->can.state = CAN_STATE_SLEEPING;
+
+	/* Check for Wake up interrupt if set put CAN device in Active state */
+	if (isr & XCAN_IXR_WKUP_MASK)
+		priv->can.state = CAN_STATE_ERROR_ACTIVE;
+}
+
+/**
+ * xcan_rx_poll - Poll routine for rx packets (NAPI)
+ * @napi:	napi structure pointer
+ * @quota:	Max number of rx packets to be processed.
+ *
+ * This is the poll routine for rx part.
+ * It will process the packets maximux quota value.
+ *
+ * Return: number of packets received
+ */
+static int xcan_rx_poll(struct napi_struct *napi, int quota)
+{
+	struct net_device *ndev = napi->dev;
+	struct xcan_priv *priv = netdev_priv(ndev);
+	u32 isr, ier;
+	int work_done = 0;
+
+	isr = priv->read_reg(priv, XCAN_ISR_OFFSET);
+	while ((isr & XCAN_IXR_RXNEMP_MASK) && (work_done < quota)) {
+		if (isr & XCAN_IXR_RXOK_MASK) {
+			priv->write_reg(priv, XCAN_ICR_OFFSET,
+				XCAN_IXR_RXOK_MASK);
+			work_done += xcan_rx(ndev);
+		} else {
+			priv->write_reg(priv, XCAN_ICR_OFFSET,
+				XCAN_IXR_RXNEMP_MASK);
+			break;
+		}
+		priv->write_reg(priv, XCAN_ICR_OFFSET, XCAN_IXR_RXNEMP_MASK);
+		isr = priv->read_reg(priv, XCAN_ISR_OFFSET);
+	}
+
+	if (work_done)
+		can_led_event(ndev, CAN_LED_EVENT_RX);
+
+	if (work_done < quota) {
+		napi_complete(napi);
+		ier = priv->read_reg(priv, XCAN_IER_OFFSET);
+		ier |= (XCAN_IXR_RXOK_MASK | XCAN_IXR_RXNEMP_MASK);
+		priv->write_reg(priv, XCAN_IER_OFFSET, ier);
+	}
+	return work_done;
+}
+
+/**
+ * xcan_tx_interrupt - Tx Done Isr
+ * @ndev:	net_device pointer
+ * @isr:	Interrupt status register value
+ */
+static void xcan_tx_interrupt(struct net_device *ndev, u32 isr)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+
+	while ((priv->tx_head - priv->tx_tail > 0) &&
+			(isr & XCAN_IXR_TXOK_MASK)) {
+		priv->write_reg(priv, XCAN_ICR_OFFSET, XCAN_IXR_TXOK_MASK);
+		can_get_echo_skb(ndev, priv->tx_tail %
+					priv->tx_max);
+		priv->tx_tail++;
+		stats->tx_packets++;
+		isr = priv->read_reg(priv, XCAN_ISR_OFFSET);
+	}
+	can_led_event(ndev, CAN_LED_EVENT_TX);
+	netif_wake_queue(ndev);
+}
+
+/**
+ * xcan_interrupt - CAN Isr
+ * @irq:	irq number
+ * @dev_id:	device id poniter
+ *
+ * This is the xilinx CAN Isr. It checks for the type of interrupt
+ * and invokes the corresponding ISR.
+ *
+ * Return:
+ * IRQ_NONE - If CAN device is in sleep mode, IRQ_HANDLED otherwise
+ */
+static irqreturn_t xcan_interrupt(int irq, void *dev_id)
+{
+	struct net_device *ndev = (struct net_device *)dev_id;
+	struct xcan_priv *priv = netdev_priv(ndev);
+	u32 isr, ier;
+
+	/* Get the interrupt status from Xilinx CAN */
+	isr = priv->read_reg(priv, XCAN_ISR_OFFSET);
+	if (!isr)
+		return IRQ_NONE;
+
+	/* Check for the type of interrupt and Processing it */
+	if (isr & (XCAN_IXR_SLP_MASK | XCAN_IXR_WKUP_MASK)) {
+		priv->write_reg(priv, XCAN_ICR_OFFSET, (XCAN_IXR_SLP_MASK |
+				XCAN_IXR_WKUP_MASK));
+		xcan_state_interrupt(ndev, isr);
+	}
+
+	/* Check for Tx interrupt and Processing it */
+	if (isr & XCAN_IXR_TXOK_MASK)
+		xcan_tx_interrupt(ndev, isr);
+
+	/* Check for the type of error interrupt and Processing it */
+	if (isr & (XCAN_IXR_ERROR_MASK | XCAN_IXR_RXOFLW_MASK |
+			XCAN_IXR_BSOFF_MASK | XCAN_IXR_ARBLST_MASK)) {
+		priv->write_reg(priv, XCAN_ICR_OFFSET, (XCAN_IXR_ERROR_MASK |
+				XCAN_IXR_RXOFLW_MASK | XCAN_IXR_BSOFF_MASK |
+				XCAN_IXR_ARBLST_MASK));
+		xcan_err_interrupt(ndev, isr);
+	}
+
+	/* Check for the type of receive interrupt and Processing it */
+	if (isr & (XCAN_IXR_RXNEMP_MASK | XCAN_IXR_RXOK_MASK)) {
+		ier = priv->read_reg(priv, XCAN_IER_OFFSET);
+		ier &= ~(XCAN_IXR_RXNEMP_MASK | XCAN_IXR_RXOK_MASK);
+		priv->write_reg(priv, XCAN_IER_OFFSET, ier);
+		napi_schedule(&priv->napi);
+	}
+	return IRQ_HANDLED;
+}
+
+/**
+ * xcan_chip_stop - Driver stop routine
+ * @ndev:	Pointer to net_device structure
+ *
+ * This is the drivers stop routine. It will disable the
+ * interrupts and put the device into configuration mode.
+ */
+static void xcan_chip_stop(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	u32 ier;
+
+	/* Disable interrupts and leave the can in configuration mode */
+	ier = priv->read_reg(priv, XCAN_IER_OFFSET);
+	ier &= ~XCAN_INTR_ALL;
+	priv->write_reg(priv, XCAN_IER_OFFSET, ier);
+	priv->write_reg(priv, XCAN_SRR_OFFSET, XCAN_SRR_RESET_MASK);
+	priv->can.state = CAN_STATE_STOPPED;
+}
+
+/**
+ * xcan_open - Driver open routine
+ * @ndev:	Pointer to net_device structure
+ *
+ * This is the driver open routine.
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_open(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	int ret;
+
+	ret = request_irq(ndev->irq, xcan_interrupt, priv->irq_flags,
+			ndev->name, ndev);
+	if (ret < 0) {
+		netdev_err(ndev, "irq allocation for CAN failed\n");
+		goto err;
+	}
+
+	ret = clk_prepare_enable(priv->can_clk);
+	if (ret) {
+		netdev_err(ndev, "unable to enable device clock\n");
+		goto err_irq;
+	}
+
+	ret = clk_prepare_enable(priv->bus_clk);
+	if (ret) {
+		netdev_err(ndev, "unable to enable bus clock\n");
+		goto err_can_clk;
+	}
+
+	/* Set chip into reset mode */
+	ret = set_reset_mode(ndev);
+	if (ret < 0) {
+		netdev_err(ndev, "mode resetting failed!\n");
+		goto err_bus_clk;
+	}
+
+	/* Common open */
+	ret = open_candev(ndev);
+	if (ret)
+		goto err_bus_clk;
+
+	ret = xcan_chip_start(ndev);
+	if (ret < 0) {
+		netdev_err(ndev, "xcan_chip_start failed!\n");
+		goto err_candev;
+	}
+
+	can_led_event(ndev, CAN_LED_EVENT_OPEN);
+	napi_enable(&priv->napi);
+	netif_start_queue(ndev);
+
+	return 0;
+
+err_candev:
+	close_candev(ndev);
+err_bus_clk:
+	clk_disable_unprepare(priv->bus_clk);
+err_can_clk:
+	clk_disable_unprepare(priv->can_clk);
+err_irq:
+	free_irq(ndev->irq, ndev);
+err:
+	return ret;
+}
+
+/**
+ * xcan_close - Driver close routine
+ * @ndev:	Pointer to net_device structure
+ *
+ * Return: 0 always
+ */
+static int xcan_close(struct net_device *ndev)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+
+	netif_stop_queue(ndev);
+	napi_disable(&priv->napi);
+	xcan_chip_stop(ndev);
+	clk_disable_unprepare(priv->bus_clk);
+	clk_disable_unprepare(priv->can_clk);
+	free_irq(ndev->irq, ndev);
+	close_candev(ndev);
+
+	can_led_event(ndev, CAN_LED_EVENT_STOP);
+
+	return 0;
+}
+
+/**
+ * xcan_get_berr_counter - error counter routine
+ * @ndev:	Pointer to net_device structure
+ * @bec:	Pointer to can_berr_counter structure
+ *
+ * This is the driver error counter routine.
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_get_berr_counter(const struct net_device *ndev,
+					struct can_berr_counter *bec)
+{
+	struct xcan_priv *priv = netdev_priv(ndev);
+	int ret;
+
+	ret = clk_prepare_enable(priv->can_clk);
+	if (ret)
+		goto err;
+
+	ret = clk_prepare_enable(priv->bus_clk);
+	if (ret)
+		goto err_clk;
+
+	bec->txerr = priv->read_reg(priv, XCAN_ECR_OFFSET) & XCAN_ECR_TEC_MASK;
+	bec->rxerr = ((priv->read_reg(priv, XCAN_ECR_OFFSET) &
+			XCAN_ECR_REC_MASK) >> XCAN_ESR_REC_SHIFT);
+
+	clk_disable_unprepare(priv->bus_clk);
+	clk_disable_unprepare(priv->can_clk);
+
+	return 0;
+
+err_clk:
+	clk_disable_unprepare(priv->can_clk);
+err:
+	return ret;
+}
+
+
+static const struct net_device_ops xcan_netdev_ops = {
+	.ndo_open	= xcan_open,
+	.ndo_stop	= xcan_close,
+	.ndo_start_xmit	= xcan_start_xmit,
+};
+
+/**
+ * xcan_suspend - Suspend method for the driver
+ * @dev:	Address of the platform_device structure
+ *
+ * Put the driver into low power mode.
+ * Return: 0 always
+ */
+static int __maybe_unused xcan_suspend(struct device *dev)
+{
+	struct platform_device *pdev = dev_get_drvdata(dev);
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct xcan_priv *priv = netdev_priv(ndev);
+
+	if (netif_running(ndev)) {
+		netif_stop_queue(ndev);
+		netif_device_detach(ndev);
+	}
+
+	priv->write_reg(priv, XCAN_MSR_OFFSET, XCAN_MSR_SLEEP_MASK);
+	priv->can.state = CAN_STATE_SLEEPING;
+
+	clk_disable(priv->bus_clk);
+	clk_disable(priv->can_clk);
+
+	return 0;
+}
+
+/**
+ * xcan_resume - Resume from suspend
+ * @dev:	Address of the platformdevice structure
+ *
+ * Resume operation after suspend.
+ * Return: 0 on success and failure value on error
+ */
+static int __maybe_unused xcan_resume(struct device *dev)
+{
+	struct platform_device *pdev = dev_get_drvdata(dev);
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct xcan_priv *priv = netdev_priv(ndev);
+	int ret;
+
+	ret = clk_enable(priv->bus_clk);
+	if (ret) {
+		dev_err(dev, "Cannot enable clock.\n");
+		return ret;
+	}
+	ret = clk_enable(priv->can_clk);
+	if (ret) {
+		dev_err(dev, "Cannot enable clock.\n");
+		clk_disable_unprepare(priv->bus_clk);
+		return ret;
+	}
+
+	priv->write_reg(priv, XCAN_MSR_OFFSET, 0);
+	priv->write_reg(priv, XCAN_SRR_OFFSET, XCAN_SRR_CEN_MASK);
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+
+	if (netif_running(ndev)) {
+		netif_device_attach(ndev);
+		netif_start_queue(ndev);
+	}
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(xcan_dev_pm_ops, xcan_suspend, xcan_resume);
+
+/**
+ * xcan_probe - Platform registration call
+ * @pdev:	Handle to the platform device structure
+ *
+ * This function does all the memory allocation and registration for the CAN
+ * device.
+ *
+ * Return: 0 on success and failure value on error
+ */
+static int xcan_probe(struct platform_device *pdev)
+{
+	struct resource *res; /* IO mem resources */
+	struct net_device *ndev;
+	struct xcan_priv *priv;
+	void __iomem *addr;
+	int ret, rx_max, tx_max;
+
+	/* Get the virtual base address for the device */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	addr = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(addr)) {
+		ret = PTR_ERR(addr);
+		goto err;
+	}
+
+	ret = of_property_read_u32(pdev->dev.of_node, "tx-fifo-depth", &tx_max);
+	if (ret < 0)
+		goto err;
+
+	ret = of_property_read_u32(pdev->dev.of_node, "rx-fifo-depth", &rx_max);
+	if (ret < 0)
+		goto err;
+
+	/* Create a CAN device instance */
+	ndev = alloc_candev(sizeof(struct xcan_priv), tx_max);
+	if (!ndev)
+		return -ENOMEM;
+
+	priv = netdev_priv(ndev);
+	priv->dev = ndev;
+	priv->can.bittiming_const = &xcan_bittiming_const;
+	priv->can.do_set_mode = xcan_do_set_mode;
+	priv->can.do_get_berr_counter = xcan_get_berr_counter;
+	priv->can.ctrlmode_supported = CAN_CTRLMODE_LOOPBACK |
+					CAN_CTRLMODE_BERR_REPORTING;
+	priv->reg_base = addr;
+	priv->tx_max = tx_max;
+
+	/* Get IRQ for the device */
+	ndev->irq = platform_get_irq(pdev, 0);
+	ndev->flags |= IFF_ECHO;	/* We support local echo */
+
+	platform_set_drvdata(pdev, ndev);
+	SET_NETDEV_DEV(ndev, &pdev->dev);
+	ndev->netdev_ops = &xcan_netdev_ops;
+
+	/* Getting the CAN can_clk info */
+	priv->can_clk = devm_clk_get(&pdev->dev, "can_clk");
+	if (IS_ERR(priv->can_clk)) {
+		dev_err(&pdev->dev, "Device clock not found.\n");
+		ret = PTR_ERR(priv->can_clk);
+		goto err_free;
+	}
+	/* Check for type of CAN device */
+	if (of_device_is_compatible(pdev->dev.of_node,
+				    "xlnx,zynq-can-1.0")) {
+		priv->bus_clk = devm_clk_get(&pdev->dev, "pclk");
+		if (IS_ERR(priv->bus_clk)) {
+			dev_err(&pdev->dev, "bus clock not found\n");
+			ret = PTR_ERR(priv->bus_clk);
+			goto err_free;
+		}
+	} else {
+		priv->bus_clk = devm_clk_get(&pdev->dev, "s_axi_aclk");
+		if (IS_ERR(priv->bus_clk)) {
+			dev_err(&pdev->dev, "bus clock not found\n");
+			ret = PTR_ERR(priv->bus_clk);
+			goto err_free;
+		}
+	}
+
+	ret = clk_prepare_enable(priv->can_clk);
+	if (ret) {
+		dev_err(&pdev->dev, "unable to enable device clock\n");
+		goto err_free;
+	}
+
+	ret = clk_prepare_enable(priv->bus_clk);
+	if (ret) {
+		dev_err(&pdev->dev, "unable to enable bus clock\n");
+		goto err_unprepare_disable_dev;
+	}
+
+	priv->write_reg = xcan_write_reg_le;
+	priv->read_reg = xcan_read_reg_le;
+
+	if (priv->read_reg(priv, XCAN_SR_OFFSET) != XCAN_SR_CONFIG_MASK) {
+		priv->write_reg = xcan_write_reg_be;
+		priv->read_reg = xcan_read_reg_be;
+	}
+
+	priv->can.clock.freq = clk_get_rate(priv->can_clk);
+
+	netif_napi_add(ndev, &priv->napi, xcan_rx_poll, rx_max);
+
+	ret = register_candev(ndev);
+	if (ret) {
+		dev_err(&pdev->dev, "fail to register failed (err=%d)\n", ret);
+		goto err_unprepare_disable_busclk;
+	}
+
+	devm_can_led_init(ndev);
+	clk_disable_unprepare(priv->bus_clk);
+	clk_disable_unprepare(priv->can_clk);
+	netdev_dbg(ndev, "reg_base=0x%p irq=%d clock=%d, tx fifo depth:%d\n",
+			priv->reg_base, ndev->irq, priv->can.clock.freq,
+			priv->tx_max);
+
+	return 0;
+
+err_unprepare_disable_busclk:
+	clk_disable_unprepare(priv->bus_clk);
+err_unprepare_disable_dev:
+	clk_disable_unprepare(priv->can_clk);
+err_free:
+	free_candev(ndev);
+err:
+	return ret;
+}
+
+/**
+ * xcan_remove - Unregister the device after releasing the resources
+ * @pdev:	Handle to the platform device structure
+ *
+ * This function frees all the resources allocated to the device.
+ * Return: 0 always
+ */
+static int xcan_remove(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct xcan_priv *priv = netdev_priv(ndev);
+
+	if (set_reset_mode(ndev) < 0)
+		netdev_err(ndev, "mode resetting failed!\n");
+
+	unregister_candev(ndev);
+	netif_napi_del(&priv->napi);
+	free_candev(ndev);
+
+	return 0;
+}
+
+/* Match table for OF platform binding */
+static struct of_device_id xcan_of_match[] = {
+	{ .compatible = "xlnx,zynq-can-1.0", },
+	{ .compatible = "xlnx,axi-can-1.00.a", },
+	{ /* end of list */ },
+};
+MODULE_DEVICE_TABLE(of, xcan_of_match);
+
+static struct platform_driver xcan_driver = {
+	.probe = xcan_probe,
+	.remove	= xcan_remove,
+	.driver	= {
+		.owner = THIS_MODULE,
+		.name = DRIVER_NAME,
+		.pm = &xcan_dev_pm_ops,
+		.of_match_table	= xcan_of_match,
+	},
+};
+
+module_platform_driver(xcan_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Xilinx Inc");
+MODULE_DESCRIPTION("Xilinx CAN interface");
diff --git a/drivers/net/dsa/mv88e6123_61_65.c b/drivers/net/dsa/mv88e6123_61_65.c
index 41ee5b6ae917..69c42513dd72 100644
--- a/drivers/net/dsa/mv88e6123_61_65.c
+++ b/drivers/net/dsa/mv88e6123_61_65.c
@@ -289,7 +289,7 @@ static int mv88e6123_61_65_setup_port(struct dsa_switch *ds, int p)
 
 static int mv88e6123_61_65_setup(struct dsa_switch *ds)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int i;
 	int ret;
 
diff --git a/drivers/net/dsa/mv88e6131.c b/drivers/net/dsa/mv88e6131.c
index dadfafba64e9..953bc6a49e59 100644
--- a/drivers/net/dsa/mv88e6131.c
+++ b/drivers/net/dsa/mv88e6131.c
@@ -155,7 +155,7 @@ static int mv88e6131_setup_global(struct dsa_switch *ds)
 
 static int mv88e6131_setup_port(struct dsa_switch *ds, int p)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int addr = REG_PORT(p);
 	u16 val;
 
@@ -274,7 +274,7 @@ static int mv88e6131_setup_port(struct dsa_switch *ds, int p)
 
 static int mv88e6131_setup(struct dsa_switch *ds)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int i;
 	int ret;
 
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 17314ed9456d..9ce2146346b6 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -74,7 +74,7 @@ int __mv88e6xxx_reg_read(struct mii_bus *bus, int sw_addr, int addr, int reg)
 
 int mv88e6xxx_reg_read(struct dsa_switch *ds, int addr, int reg)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int ret;
 
 	mutex_lock(&ps->smi_mutex);
@@ -118,7 +118,7 @@ int __mv88e6xxx_reg_write(struct mii_bus *bus, int sw_addr, int addr,
 
 int mv88e6xxx_reg_write(struct dsa_switch *ds, int addr, int reg, u16 val)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int ret;
 
 	mutex_lock(&ps->smi_mutex);
@@ -256,7 +256,7 @@ static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps)
 
 static int mv88e6xxx_ppu_access_get(struct dsa_switch *ds)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int ret;
 
 	mutex_lock(&ps->ppu_mutex);
@@ -283,7 +283,7 @@ static int mv88e6xxx_ppu_access_get(struct dsa_switch *ds)
 
 static void mv88e6xxx_ppu_access_put(struct dsa_switch *ds)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 
 	/* Schedule a timer to re-enable the PHY polling unit. */
 	mod_timer(&ps->ppu_timer, jiffies + msecs_to_jiffies(10));
@@ -292,7 +292,7 @@ static void mv88e6xxx_ppu_access_put(struct dsa_switch *ds)
 
 void mv88e6xxx_ppu_state_init(struct dsa_switch *ds)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 
 	mutex_init(&ps->ppu_mutex);
 	INIT_WORK(&ps->ppu_work, mv88e6xxx_ppu_reenable_work);
@@ -463,7 +463,7 @@ void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds,
 				 int nr_stats, struct mv88e6xxx_hw_stat *stats,
 				 int port, uint64_t *data)
 {
-	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int ret;
 	int i;
 
diff --git a/drivers/net/ethernet/3com/3c509.c b/drivers/net/ethernet/3com/3c509.c
index 35df0b9e6848..a968654b631d 100644
--- a/drivers/net/ethernet/3com/3c509.c
+++ b/drivers/net/ethernet/3com/3c509.c
@@ -534,7 +534,7 @@ static int el3_common_init(struct net_device *dev)
 	/* The EL3-specific entries in the device structure. */
 	dev->netdev_ops = &netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
-	SET_ETHTOOL_OPS(dev, &ethtool_ops);
+	dev->ethtool_ops = &ethtool_ops;
 
 	err = register_netdev(dev);
 	if (err) {
diff --git a/drivers/net/ethernet/3com/3c589_cs.c b/drivers/net/ethernet/3com/3c589_cs.c
index 063557e037f2..f18647c23559 100644
--- a/drivers/net/ethernet/3com/3c589_cs.c
+++ b/drivers/net/ethernet/3com/3c589_cs.c
@@ -218,7 +218,7 @@ static int tc589_probe(struct pcmcia_device *link)
 	dev->netdev_ops = &el3_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
-	SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
+	dev->ethtool_ops = &netdev_ethtool_ops;
 
 	return tc589_config(link);
 }
diff --git a/drivers/net/ethernet/3com/typhoon.c b/drivers/net/ethernet/3com/typhoon.c
index 465cc7108d8a..e13b04624ded 100644
--- a/drivers/net/ethernet/3com/typhoon.c
+++ b/drivers/net/ethernet/3com/typhoon.c
@@ -2435,7 +2435,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netif_napi_add(dev, &tp->napi, typhoon_poll, 16);
 	dev->watchdog_timeo	= TX_TIMEOUT;
 
-	SET_ETHTOOL_OPS(dev, &typhoon_ethtool_ops);
+	dev->ethtool_ops = &typhoon_ethtool_ops;
 
 	/* We can handle scatter gather, up to 16 entries, and
 	 * we can do IP checksumming (only version 4, doh...)
diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c
index 455d4c399b52..1d162ccb4733 100644
--- a/drivers/net/ethernet/8390/ax88796.c
+++ b/drivers/net/ethernet/8390/ax88796.c
@@ -157,7 +157,7 @@ static void ax_reset_8390(struct net_device *dev)
 
 	/* This check _should_not_ be necessary, omit eventually. */
 	while ((ei_inb(addr + EN0_ISR) & ENISR_RESET) == 0) {
-		if (jiffies - reset_start_time > 2 * HZ / 100) {
+		if (time_after(jiffies, reset_start_time + 2 * HZ / 100)) {
 			netdev_warn(dev, "%s: did not complete.\n", __func__);
 			break;
 		}
@@ -293,7 +293,7 @@ static void ax_block_output(struct net_device *dev, int count,
 	dma_start = jiffies;
 
 	while ((ei_inb(nic_base + EN0_ISR) & ENISR_RDC) == 0) {
-		if (jiffies - dma_start > 2 * HZ / 100) {		/* 20ms */
+		if (time_after(jiffies, dma_start + 2 * HZ / 100)) { /* 20ms */
 			netdev_warn(dev, "timeout waiting for Tx RDC.\n");
 			ax_reset_8390(dev);
 			ax_NS8390_init(dev, 1);
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 051349458462..edb718661850 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -68,6 +68,7 @@ source "drivers/net/ethernet/neterion/Kconfig"
 source "drivers/net/ethernet/faraday/Kconfig"
 source "drivers/net/ethernet/freescale/Kconfig"
 source "drivers/net/ethernet/fujitsu/Kconfig"
+source "drivers/net/ethernet/hisilicon/Kconfig"
 source "drivers/net/ethernet/hp/Kconfig"
 source "drivers/net/ethernet/ibm/Kconfig"
 source "drivers/net/ethernet/intel/Kconfig"
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 35190e36c456..58de3339ab3c 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_NET_VENDOR_EXAR) += neterion/
 obj-$(CONFIG_NET_VENDOR_FARADAY) += faraday/
 obj-$(CONFIG_NET_VENDOR_FREESCALE) += freescale/
 obj-$(CONFIG_NET_VENDOR_FUJITSU) += fujitsu/
+obj-$(CONFIG_NET_VENDOR_HISILICON) += hisilicon/
 obj-$(CONFIG_NET_VENDOR_HP) += hp/
 obj-$(CONFIG_NET_VENDOR_IBM) += ibm/
 obj-$(CONFIG_NET_VENDOR_INTEL) += intel/
diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c
index 171d73c1d3c2..40dbbf740331 100644
--- a/drivers/net/ethernet/adaptec/starfire.c
+++ b/drivers/net/ethernet/adaptec/starfire.c
@@ -784,7 +784,7 @@ static int starfire_init_one(struct pci_dev *pdev,
 
 	dev->netdev_ops = &netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
-	SET_ETHTOOL_OPS(dev, &ethtool_ops);
+	dev->ethtool_ops = &ethtool_ops;
 
 	netif_napi_add(dev, &np->napi, netdev_poll, max_interrupt_work);
 
diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c
index 28460676b8ca..d81e7167a8b5 100644
--- a/drivers/net/ethernet/allwinner/sun4i-emac.c
+++ b/drivers/net/ethernet/allwinner/sun4i-emac.c
@@ -736,6 +736,7 @@ static int emac_open(struct net_device *dev)
 
 	ret = emac_mdio_probe(dev);
 	if (ret < 0) {
+		free_irq(dev->irq, dev);
 		netdev_err(dev, "cannot probe MDIO bus\n");
 		return ret;
 	}
diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c
index 1517e9df5ba1..9a6991be9749 100644
--- a/drivers/net/ethernet/alteon/acenic.c
+++ b/drivers/net/ethernet/alteon/acenic.c
@@ -476,7 +476,7 @@ static int acenic_probe_one(struct pci_dev *pdev,
 	dev->watchdog_timeo = 5*HZ;
 
 	dev->netdev_ops = &ace_netdev_ops;
-	SET_ETHTOOL_OPS(dev, &ace_ethtool_ops);
+	dev->ethtool_ops = &ace_ethtool_ops;
 
 	/* we only display this string ONCE */
 	if (!boards_found)
diff --git a/drivers/net/ethernet/altera/altera_sgdma.c b/drivers/net/ethernet/altera/altera_sgdma.c
index 99cc56f451cf..580553d42d34 100644
--- a/drivers/net/ethernet/altera/altera_sgdma.c
+++ b/drivers/net/ethernet/altera/altera_sgdma.c
@@ -353,7 +353,6 @@ static int sgdma_async_read(struct altera_tse_private *priv)
 
 	struct sgdma_descrip __iomem *cdesc = &descbase[0];
 	struct sgdma_descrip __iomem *ndesc = &descbase[1];
-
 	struct tse_buffer *rxbuffer = NULL;
 
 	if (!sgdma_rxbusy(priv)) {
diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c
index 54c25eff7952..be72e1e64525 100644
--- a/drivers/net/ethernet/altera/altera_tse_ethtool.c
+++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c
@@ -271,5 +271,5 @@ static const struct ethtool_ops tse_ethtool_ops = {
 
 void altera_tse_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &tse_ethtool_ops);
+	netdev->ethtool_ops = &tse_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig
index 562df46e0a82..bbaf36d9f5e1 100644
--- a/drivers/net/ethernet/amd/Kconfig
+++ b/drivers/net/ethernet/amd/Kconfig
@@ -7,7 +7,7 @@ config NET_VENDOR_AMD
 	default y
 	depends on DIO || MACH_DECSTATION || MVME147 || ATARI || SUN3 || \
 		   SUN3X || SBUS || PCI || ZORRO || (ISA && ISA_DMA_API) || \
-		   (ARM && ARCH_EBSA110) || ISA || EISA || PCMCIA
+		   (ARM && ARCH_EBSA110) || ISA || EISA || PCMCIA || ARM64
 	---help---
 	  If you have a network (Ethernet) chipset belonging to this class,
 	  say Y.
@@ -177,4 +177,16 @@ config SUNLANCE
 	  To compile this driver as a module, choose M here: the module
 	  will be called sunlance.
 
+config AMD_XGBE
+	tristate "AMD 10GbE Ethernet driver"
+	depends on OF_NET
+	select PHYLIB
+	select AMD_XGBE_PHY
+	---help---
+	  This driver supports the AMD 10GbE Ethernet device found on an
+	  AMD SoC.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called amd-xgbe.
+
 endif # NET_VENDOR_AMD
diff --git a/drivers/net/ethernet/amd/Makefile b/drivers/net/ethernet/amd/Makefile
index cdd4301a973d..a38a2dce3eb3 100644
--- a/drivers/net/ethernet/amd/Makefile
+++ b/drivers/net/ethernet/amd/Makefile
@@ -17,3 +17,4 @@ obj-$(CONFIG_NI65) += ni65.o
 obj-$(CONFIG_PCNET32) += pcnet32.o
 obj-$(CONFIG_SUN3LANCE) += sun3lance.o
 obj-$(CONFIG_SUNLANCE) += sunlance.o
+obj-$(CONFIG_AMD_XGBE) += xgbe/
diff --git a/drivers/net/ethernet/amd/amd8111e.c b/drivers/net/ethernet/amd/amd8111e.c
index 26efaaa5e73f..068dc7cad5fa 100644
--- a/drivers/net/ethernet/amd/amd8111e.c
+++ b/drivers/net/ethernet/amd/amd8111e.c
@@ -1900,7 +1900,7 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 
 	/* Initialize driver entry points */
 	dev->netdev_ops = &amd8111e_netdev_ops;
-	SET_ETHTOOL_OPS(dev, &ops);
+	dev->ethtool_ops = &ops;
 	dev->irq =pdev->irq;
 	dev->watchdog_timeo = AMD8111E_TX_TIMEOUT;
 	netif_napi_add(dev, &lp->napi, amd8111e_rx_poll, 32);
diff --git a/drivers/net/ethernet/amd/ariadne.c b/drivers/net/ethernet/amd/ariadne.c
index b08101b31b8b..968b7bfac8fc 100644
--- a/drivers/net/ethernet/amd/ariadne.c
+++ b/drivers/net/ethernet/amd/ariadne.c
@@ -718,7 +718,6 @@ static int ariadne_init_one(struct zorro_dev *z,
 	unsigned long mem_start = board + ARIADNE_RAM;
 	struct resource *r1, *r2;
 	struct net_device *dev;
-	struct ariadne_private *priv;
 	u32 serial;
 	int err;
 
@@ -738,8 +737,6 @@ static int ariadne_init_one(struct zorro_dev *z,
 		return -ENOMEM;
 	}
 
-	priv = netdev_priv(dev);
-
 	r1->name = dev->name;
 	r2->name = dev->name;
 
diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c
index a2bd91e3d302..a78e4c136959 100644
--- a/drivers/net/ethernet/amd/au1000_eth.c
+++ b/drivers/net/ethernet/amd/au1000_eth.c
@@ -1229,7 +1229,7 @@ static int au1000_probe(struct platform_device *pdev)
 	dev->base_addr = base->start;
 	dev->irq = irq;
 	dev->netdev_ops = &au1000_netdev_ops;
-	SET_ETHTOOL_OPS(dev, &au1000_ethtool_ops);
+	dev->ethtool_ops = &au1000_ethtool_ops;
 	dev->watchdog_timeo = ETH_TX_TIMEOUT;
 
 	/*
diff --git a/drivers/net/ethernet/amd/hplance.c b/drivers/net/ethernet/amd/hplance.c
index 47ce57c2c893..6c9de117ffc6 100644
--- a/drivers/net/ethernet/amd/hplance.c
+++ b/drivers/net/ethernet/amd/hplance.c
@@ -27,9 +27,9 @@
 
 #include "hplance.h"
 
-/* We have 16834 bytes of RAM for the init block and buffers. This places
+/* We have 16392 bytes of RAM for the init block and buffers. This places
  * an upper limit on the number of buffers we can use. NetBSD uses 8 Rx
- * buffers and 2 Tx buffers.
+ * buffers and 2 Tx buffers, it takes (8 + 2) * 1544 bytes.
  */
 #define LANCE_LOG_TX_BUFFERS 1
 #define LANCE_LOG_RX_BUFFERS 3
diff --git a/drivers/net/ethernet/amd/mvme147.c b/drivers/net/ethernet/amd/mvme147.c
index 0e8399dec054..0660ac5846bb 100644
--- a/drivers/net/ethernet/amd/mvme147.c
+++ b/drivers/net/ethernet/amd/mvme147.c
@@ -26,9 +26,9 @@
 #include <asm/pgtable.h>
 #include <asm/mvme147hw.h>
 
-/* We have 16834 bytes of RAM for the init block and buffers. This places
+/* We have 32K of RAM for the init block and buffers. This places
  * an upper limit on the number of buffers we can use. NetBSD uses 8 Rx
- * buffers and 2 Tx buffers.
+ * buffers and 2 Tx buffers, it takes (8 + 2) * 1544 bytes.
  */
 #define LANCE_LOG_TX_BUFFERS 1
 #define LANCE_LOG_RX_BUFFERS 3
@@ -111,7 +111,7 @@ struct net_device * __init mvme147lance_probe(int unit)
 	       dev->dev_addr);
 
 	lp = netdev_priv(dev);
-	lp->ram = __get_dma_pages(GFP_ATOMIC, 3);	/* 16K */
+	lp->ram = __get_dma_pages(GFP_ATOMIC, 3);	/* 32K */
 	if (!lp->ram) {
 		printk("%s: No memory for LANCE buffers\n", dev->name);
 		free_netdev(dev);
diff --git a/drivers/net/ethernet/amd/nmclan_cs.c b/drivers/net/ethernet/amd/nmclan_cs.c
index 08569fe2b182..abf3b1581c82 100644
--- a/drivers/net/ethernet/amd/nmclan_cs.c
+++ b/drivers/net/ethernet/amd/nmclan_cs.c
@@ -457,7 +457,7 @@ static int nmclan_probe(struct pcmcia_device *link)
     lp->tx_free_frames=AM2150_MAX_TX_FRAMES;
 
     dev->netdev_ops = &mace_netdev_ops;
-    SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
+    dev->ethtool_ops = &netdev_ethtool_ops;
     dev->watchdog_timeo = TX_TIMEOUT;
 
     return nmclan_config(link);
diff --git a/drivers/net/ethernet/amd/xgbe/Makefile b/drivers/net/ethernet/amd/xgbe/Makefile
new file mode 100644
index 000000000000..26cf9af1642f
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/Makefile
@@ -0,0 +1,6 @@
+obj-$(CONFIG_AMD_XGBE) += amd-xgbe.o
+
+amd-xgbe-objs := xgbe-main.o xgbe-drv.o xgbe-dev.o \
+		 xgbe-desc.o xgbe-ethtool.o xgbe-mdio.o
+
+amd-xgbe-$(CONFIG_DEBUG_FS) += xgbe-debugfs.o
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
new file mode 100644
index 000000000000..bf462ee86f5c
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
@@ -0,0 +1,1007 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __XGBE_COMMON_H__
+#define __XGBE_COMMON_H__
+
+/* DMA register offsets */
+#define DMA_MR				0x3000
+#define DMA_SBMR			0x3004
+#define DMA_ISR				0x3008
+#define DMA_AXIARCR			0x3010
+#define DMA_AXIAWCR			0x3018
+#define DMA_DSR0			0x3020
+#define DMA_DSR1			0x3024
+#define DMA_DSR2			0x3028
+#define DMA_DSR3			0x302c
+#define DMA_DSR4			0x3030
+
+/* DMA register entry bit positions and sizes */
+#define DMA_AXIARCR_DRC_INDEX		0
+#define DMA_AXIARCR_DRC_WIDTH		4
+#define DMA_AXIARCR_DRD_INDEX		4
+#define DMA_AXIARCR_DRD_WIDTH		2
+#define DMA_AXIARCR_TEC_INDEX		8
+#define DMA_AXIARCR_TEC_WIDTH		4
+#define DMA_AXIARCR_TED_INDEX		12
+#define DMA_AXIARCR_TED_WIDTH		2
+#define DMA_AXIARCR_THC_INDEX		16
+#define DMA_AXIARCR_THC_WIDTH		4
+#define DMA_AXIARCR_THD_INDEX		20
+#define DMA_AXIARCR_THD_WIDTH		2
+#define DMA_AXIAWCR_DWC_INDEX		0
+#define DMA_AXIAWCR_DWC_WIDTH		4
+#define DMA_AXIAWCR_DWD_INDEX		4
+#define DMA_AXIAWCR_DWD_WIDTH		2
+#define DMA_AXIAWCR_RPC_INDEX		8
+#define DMA_AXIAWCR_RPC_WIDTH		4
+#define DMA_AXIAWCR_RPD_INDEX		12
+#define DMA_AXIAWCR_RPD_WIDTH		2
+#define DMA_AXIAWCR_RHC_INDEX		16
+#define DMA_AXIAWCR_RHC_WIDTH		4
+#define DMA_AXIAWCR_RHD_INDEX		20
+#define DMA_AXIAWCR_RHD_WIDTH		2
+#define DMA_AXIAWCR_TDC_INDEX		24
+#define DMA_AXIAWCR_TDC_WIDTH		4
+#define DMA_AXIAWCR_TDD_INDEX		28
+#define DMA_AXIAWCR_TDD_WIDTH		2
+#define DMA_DSR0_RPS_INDEX		8
+#define DMA_DSR0_RPS_WIDTH		4
+#define DMA_DSR0_TPS_INDEX		12
+#define DMA_DSR0_TPS_WIDTH		4
+#define DMA_ISR_MACIS_INDEX		17
+#define DMA_ISR_MACIS_WIDTH		1
+#define DMA_ISR_MTLIS_INDEX		16
+#define DMA_ISR_MTLIS_WIDTH		1
+#define DMA_MR_SWR_INDEX		0
+#define DMA_MR_SWR_WIDTH		1
+#define DMA_SBMR_EAME_INDEX		11
+#define DMA_SBMR_EAME_WIDTH		1
+#define DMA_SBMR_UNDEF_INDEX		0
+#define DMA_SBMR_UNDEF_WIDTH		1
+
+/* DMA channel register offsets
+ *   Multiple channels can be active.  The first channel has registers
+ *   that begin at 0x3100.  Each subsequent channel has registers that
+ *   are accessed using an offset of 0x80 from the previous channel.
+ */
+#define DMA_CH_BASE			0x3100
+#define DMA_CH_INC			0x80
+
+#define DMA_CH_CR			0x00
+#define DMA_CH_TCR			0x04
+#define DMA_CH_RCR			0x08
+#define DMA_CH_TDLR_HI			0x10
+#define DMA_CH_TDLR_LO			0x14
+#define DMA_CH_RDLR_HI			0x18
+#define DMA_CH_RDLR_LO			0x1c
+#define DMA_CH_TDTR_LO			0x24
+#define DMA_CH_RDTR_LO			0x2c
+#define DMA_CH_TDRLR			0x30
+#define DMA_CH_RDRLR			0x34
+#define DMA_CH_IER			0x38
+#define DMA_CH_RIWT			0x3c
+#define DMA_CH_CATDR_LO			0x44
+#define DMA_CH_CARDR_LO			0x4c
+#define DMA_CH_CATBR_HI			0x50
+#define DMA_CH_CATBR_LO			0x54
+#define DMA_CH_CARBR_HI			0x58
+#define DMA_CH_CARBR_LO			0x5c
+#define DMA_CH_SR			0x60
+
+/* DMA channel register entry bit positions and sizes */
+#define DMA_CH_CR_PBLX8_INDEX		16
+#define DMA_CH_CR_PBLX8_WIDTH		1
+#define DMA_CH_IER_AIE_INDEX		15
+#define DMA_CH_IER_AIE_WIDTH		1
+#define DMA_CH_IER_FBEE_INDEX		12
+#define DMA_CH_IER_FBEE_WIDTH		1
+#define DMA_CH_IER_NIE_INDEX		16
+#define DMA_CH_IER_NIE_WIDTH		1
+#define DMA_CH_IER_RBUE_INDEX		7
+#define DMA_CH_IER_RBUE_WIDTH		1
+#define DMA_CH_IER_RIE_INDEX		6
+#define DMA_CH_IER_RIE_WIDTH		1
+#define DMA_CH_IER_RSE_INDEX		8
+#define DMA_CH_IER_RSE_WIDTH		1
+#define DMA_CH_IER_TBUE_INDEX		2
+#define DMA_CH_IER_TBUE_WIDTH		1
+#define DMA_CH_IER_TIE_INDEX		0
+#define DMA_CH_IER_TIE_WIDTH		1
+#define DMA_CH_IER_TXSE_INDEX		1
+#define DMA_CH_IER_TXSE_WIDTH		1
+#define DMA_CH_RCR_PBL_INDEX		16
+#define DMA_CH_RCR_PBL_WIDTH		6
+#define DMA_CH_RCR_RBSZ_INDEX		1
+#define DMA_CH_RCR_RBSZ_WIDTH		14
+#define DMA_CH_RCR_SR_INDEX		0
+#define DMA_CH_RCR_SR_WIDTH		1
+#define DMA_CH_RIWT_RWT_INDEX		0
+#define DMA_CH_RIWT_RWT_WIDTH		8
+#define DMA_CH_SR_FBE_INDEX		12
+#define DMA_CH_SR_FBE_WIDTH		1
+#define DMA_CH_SR_RBU_INDEX		7
+#define DMA_CH_SR_RBU_WIDTH		1
+#define DMA_CH_SR_RI_INDEX		6
+#define DMA_CH_SR_RI_WIDTH		1
+#define DMA_CH_SR_RPS_INDEX		8
+#define DMA_CH_SR_RPS_WIDTH		1
+#define DMA_CH_SR_TBU_INDEX		2
+#define DMA_CH_SR_TBU_WIDTH		1
+#define DMA_CH_SR_TI_INDEX		0
+#define DMA_CH_SR_TI_WIDTH		1
+#define DMA_CH_SR_TPS_INDEX		1
+#define DMA_CH_SR_TPS_WIDTH		1
+#define DMA_CH_TCR_OSP_INDEX		4
+#define DMA_CH_TCR_OSP_WIDTH		1
+#define DMA_CH_TCR_PBL_INDEX		16
+#define DMA_CH_TCR_PBL_WIDTH		6
+#define DMA_CH_TCR_ST_INDEX		0
+#define DMA_CH_TCR_ST_WIDTH		1
+#define DMA_CH_TCR_TSE_INDEX		12
+#define DMA_CH_TCR_TSE_WIDTH		1
+
+/* DMA channel register values */
+#define DMA_OSP_DISABLE			0x00
+#define DMA_OSP_ENABLE			0x01
+#define DMA_PBL_1			1
+#define DMA_PBL_2			2
+#define DMA_PBL_4			4
+#define DMA_PBL_8			8
+#define DMA_PBL_16			16
+#define DMA_PBL_32			32
+#define DMA_PBL_64			64      /* 8 x 8 */
+#define DMA_PBL_128			128     /* 8 x 16 */
+#define DMA_PBL_256			256     /* 8 x 32 */
+#define DMA_PBL_X8_DISABLE		0x00
+#define DMA_PBL_X8_ENABLE		0x01
+
+
+/* MAC register offsets */
+#define MAC_TCR				0x0000
+#define MAC_RCR				0x0004
+#define MAC_PFR				0x0008
+#define MAC_WTR				0x000c
+#define MAC_HTR0			0x0010
+#define MAC_HTR1			0x0014
+#define MAC_HTR2			0x0018
+#define MAC_HTR3			0x001c
+#define MAC_HTR4			0x0020
+#define MAC_HTR5			0x0024
+#define MAC_HTR6			0x0028
+#define MAC_HTR7			0x002c
+#define MAC_VLANTR			0x0050
+#define MAC_VLANHTR			0x0058
+#define MAC_VLANIR			0x0060
+#define MAC_IVLANIR			0x0064
+#define MAC_RETMR			0x006c
+#define MAC_Q0TFCR			0x0070
+#define MAC_RFCR			0x0090
+#define MAC_RQC0R			0x00a0
+#define MAC_RQC1R			0x00a4
+#define MAC_RQC2R			0x00a8
+#define MAC_RQC3R			0x00ac
+#define MAC_ISR				0x00b0
+#define MAC_IER				0x00b4
+#define MAC_RTSR			0x00b8
+#define MAC_PMTCSR			0x00c0
+#define MAC_RWKPFR			0x00c4
+#define MAC_LPICSR			0x00d0
+#define MAC_LPITCR			0x00d4
+#define MAC_VR				0x0110
+#define MAC_DR				0x0114
+#define MAC_HWF0R			0x011c
+#define MAC_HWF1R			0x0120
+#define MAC_HWF2R			0x0124
+#define MAC_GPIOCR			0x0278
+#define MAC_GPIOSR			0x027c
+#define MAC_MACA0HR			0x0300
+#define MAC_MACA0LR			0x0304
+#define MAC_MACA1HR			0x0308
+#define MAC_MACA1LR			0x030c
+
+#define MAC_QTFCR_INC			4
+#define MAC_MACA_INC			4
+
+/* MAC register entry bit positions and sizes */
+#define MAC_HWF0R_ADDMACADRSEL_INDEX	18
+#define MAC_HWF0R_ADDMACADRSEL_WIDTH	5
+#define MAC_HWF0R_ARPOFFSEL_INDEX	9
+#define MAC_HWF0R_ARPOFFSEL_WIDTH	1
+#define MAC_HWF0R_EEESEL_INDEX		13
+#define MAC_HWF0R_EEESEL_WIDTH		1
+#define MAC_HWF0R_GMIISEL_INDEX		1
+#define MAC_HWF0R_GMIISEL_WIDTH		1
+#define MAC_HWF0R_MGKSEL_INDEX		7
+#define MAC_HWF0R_MGKSEL_WIDTH		1
+#define MAC_HWF0R_MMCSEL_INDEX		8
+#define MAC_HWF0R_MMCSEL_WIDTH		1
+#define MAC_HWF0R_RWKSEL_INDEX		6
+#define MAC_HWF0R_RWKSEL_WIDTH		1
+#define MAC_HWF0R_RXCOESEL_INDEX	16
+#define MAC_HWF0R_RXCOESEL_WIDTH	1
+#define MAC_HWF0R_SAVLANINS_INDEX	27
+#define MAC_HWF0R_SAVLANINS_WIDTH	1
+#define MAC_HWF0R_SMASEL_INDEX		5
+#define MAC_HWF0R_SMASEL_WIDTH		1
+#define MAC_HWF0R_TSSEL_INDEX		12
+#define MAC_HWF0R_TSSEL_WIDTH		1
+#define MAC_HWF0R_TSSTSSEL_INDEX	25
+#define MAC_HWF0R_TSSTSSEL_WIDTH	2
+#define MAC_HWF0R_TXCOESEL_INDEX	14
+#define MAC_HWF0R_TXCOESEL_WIDTH	1
+#define MAC_HWF0R_VLHASH_INDEX		4
+#define MAC_HWF0R_VLHASH_WIDTH		1
+#define MAC_HWF1R_ADVTHWORD_INDEX	13
+#define MAC_HWF1R_ADVTHWORD_WIDTH	1
+#define MAC_HWF1R_DBGMEMA_INDEX		19
+#define MAC_HWF1R_DBGMEMA_WIDTH		1
+#define MAC_HWF1R_DCBEN_INDEX		16
+#define MAC_HWF1R_DCBEN_WIDTH		1
+#define MAC_HWF1R_HASHTBLSZ_INDEX	24
+#define MAC_HWF1R_HASHTBLSZ_WIDTH	3
+#define MAC_HWF1R_L3L4FNUM_INDEX	27
+#define MAC_HWF1R_L3L4FNUM_WIDTH	4
+#define MAC_HWF1R_RSSEN_INDEX		20
+#define MAC_HWF1R_RSSEN_WIDTH		1
+#define MAC_HWF1R_RXFIFOSIZE_INDEX	0
+#define MAC_HWF1R_RXFIFOSIZE_WIDTH	5
+#define MAC_HWF1R_SPHEN_INDEX		17
+#define MAC_HWF1R_SPHEN_WIDTH		1
+#define MAC_HWF1R_TSOEN_INDEX		18
+#define MAC_HWF1R_TSOEN_WIDTH		1
+#define MAC_HWF1R_TXFIFOSIZE_INDEX	6
+#define MAC_HWF1R_TXFIFOSIZE_WIDTH	5
+#define MAC_HWF2R_AUXSNAPNUM_INDEX	28
+#define MAC_HWF2R_AUXSNAPNUM_WIDTH	3
+#define MAC_HWF2R_PPSOUTNUM_INDEX	24
+#define MAC_HWF2R_PPSOUTNUM_WIDTH	3
+#define MAC_HWF2R_RXCHCNT_INDEX		12
+#define MAC_HWF2R_RXCHCNT_WIDTH		4
+#define MAC_HWF2R_RXQCNT_INDEX		0
+#define MAC_HWF2R_RXQCNT_WIDTH		4
+#define MAC_HWF2R_TXCHCNT_INDEX		18
+#define MAC_HWF2R_TXCHCNT_WIDTH		4
+#define MAC_HWF2R_TXQCNT_INDEX		6
+#define MAC_HWF2R_TXQCNT_WIDTH		4
+#define MAC_ISR_MMCRXIS_INDEX		9
+#define MAC_ISR_MMCRXIS_WIDTH		1
+#define MAC_ISR_MMCTXIS_INDEX		10
+#define MAC_ISR_MMCTXIS_WIDTH		1
+#define MAC_ISR_PMTIS_INDEX		4
+#define MAC_ISR_PMTIS_WIDTH		1
+#define MAC_MACA1HR_AE_INDEX		31
+#define MAC_MACA1HR_AE_WIDTH		1
+#define MAC_PFR_HMC_INDEX		2
+#define MAC_PFR_HMC_WIDTH		1
+#define MAC_PFR_HUC_INDEX		1
+#define MAC_PFR_HUC_WIDTH		1
+#define MAC_PFR_PM_INDEX		4
+#define MAC_PFR_PM_WIDTH		1
+#define MAC_PFR_PR_INDEX		0
+#define MAC_PFR_PR_WIDTH		1
+#define MAC_PMTCSR_MGKPKTEN_INDEX	1
+#define MAC_PMTCSR_MGKPKTEN_WIDTH	1
+#define MAC_PMTCSR_PWRDWN_INDEX		0
+#define MAC_PMTCSR_PWRDWN_WIDTH		1
+#define MAC_PMTCSR_RWKFILTRST_INDEX	31
+#define MAC_PMTCSR_RWKFILTRST_WIDTH	1
+#define MAC_PMTCSR_RWKPKTEN_INDEX	2
+#define MAC_PMTCSR_RWKPKTEN_WIDTH	1
+#define MAC_Q0TFCR_PT_INDEX		16
+#define MAC_Q0TFCR_PT_WIDTH		16
+#define MAC_Q0TFCR_TFE_INDEX		1
+#define MAC_Q0TFCR_TFE_WIDTH		1
+#define MAC_RCR_ACS_INDEX		1
+#define MAC_RCR_ACS_WIDTH		1
+#define MAC_RCR_CST_INDEX		2
+#define MAC_RCR_CST_WIDTH		1
+#define MAC_RCR_DCRCC_INDEX		3
+#define MAC_RCR_DCRCC_WIDTH		1
+#define MAC_RCR_IPC_INDEX		9
+#define MAC_RCR_IPC_WIDTH		1
+#define MAC_RCR_JE_INDEX		8
+#define MAC_RCR_JE_WIDTH		1
+#define MAC_RCR_LM_INDEX		10
+#define MAC_RCR_LM_WIDTH		1
+#define MAC_RCR_RE_INDEX		0
+#define MAC_RCR_RE_WIDTH		1
+#define MAC_RFCR_RFE_INDEX		0
+#define MAC_RFCR_RFE_WIDTH		1
+#define MAC_RQC0R_RXQ0EN_INDEX		0
+#define MAC_RQC0R_RXQ0EN_WIDTH		2
+#define MAC_TCR_SS_INDEX		29
+#define MAC_TCR_SS_WIDTH		2
+#define MAC_TCR_TE_INDEX		0
+#define MAC_TCR_TE_WIDTH		1
+#define MAC_VLANTR_DOVLTC_INDEX		20
+#define MAC_VLANTR_DOVLTC_WIDTH		1
+#define MAC_VLANTR_ERSVLM_INDEX		19
+#define MAC_VLANTR_ERSVLM_WIDTH		1
+#define MAC_VLANTR_ESVL_INDEX		18
+#define MAC_VLANTR_ESVL_WIDTH		1
+#define MAC_VLANTR_EVLS_INDEX		21
+#define MAC_VLANTR_EVLS_WIDTH		2
+#define MAC_VLANTR_EVLRXS_INDEX		24
+#define MAC_VLANTR_EVLRXS_WIDTH		1
+#define MAC_VR_DEVID_INDEX		8
+#define MAC_VR_DEVID_WIDTH		8
+#define MAC_VR_SNPSVER_INDEX		0
+#define MAC_VR_SNPSVER_WIDTH		8
+#define MAC_VR_USERVER_INDEX		16
+#define MAC_VR_USERVER_WIDTH		8
+
+/* MMC register offsets */
+#define MMC_CR				0x0800
+#define MMC_RISR			0x0804
+#define MMC_TISR			0x0808
+#define MMC_RIER			0x080c
+#define MMC_TIER			0x0810
+#define MMC_TXOCTETCOUNT_GB_LO		0x0814
+#define MMC_TXOCTETCOUNT_GB_HI		0x0818
+#define MMC_TXFRAMECOUNT_GB_LO		0x081c
+#define MMC_TXFRAMECOUNT_GB_HI		0x0820
+#define MMC_TXBROADCASTFRAMES_G_LO	0x0824
+#define MMC_TXBROADCASTFRAMES_G_HI	0x0828
+#define MMC_TXMULTICASTFRAMES_G_LO	0x082c
+#define MMC_TXMULTICASTFRAMES_G_HI	0x0830
+#define MMC_TX64OCTETS_GB_LO		0x0834
+#define MMC_TX64OCTETS_GB_HI		0x0838
+#define MMC_TX65TO127OCTETS_GB_LO	0x083c
+#define MMC_TX65TO127OCTETS_GB_HI	0x0840
+#define MMC_TX128TO255OCTETS_GB_LO	0x0844
+#define MMC_TX128TO255OCTETS_GB_HI	0x0848
+#define MMC_TX256TO511OCTETS_GB_LO	0x084c
+#define MMC_TX256TO511OCTETS_GB_HI	0x0850
+#define MMC_TX512TO1023OCTETS_GB_LO	0x0854
+#define MMC_TX512TO1023OCTETS_GB_HI	0x0858
+#define MMC_TX1024TOMAXOCTETS_GB_LO	0x085c
+#define MMC_TX1024TOMAXOCTETS_GB_HI	0x0860
+#define MMC_TXUNICASTFRAMES_GB_LO	0x0864
+#define MMC_TXUNICASTFRAMES_GB_HI	0x0868
+#define MMC_TXMULTICASTFRAMES_GB_LO	0x086c
+#define MMC_TXMULTICASTFRAMES_GB_HI	0x0870
+#define MMC_TXBROADCASTFRAMES_GB_LO	0x0874
+#define MMC_TXBROADCASTFRAMES_GB_HI	0x0878
+#define MMC_TXUNDERFLOWERROR_LO		0x087c
+#define MMC_TXUNDERFLOWERROR_HI		0x0880
+#define MMC_TXOCTETCOUNT_G_LO		0x0884
+#define MMC_TXOCTETCOUNT_G_HI		0x0888
+#define MMC_TXFRAMECOUNT_G_LO		0x088c
+#define MMC_TXFRAMECOUNT_G_HI		0x0890
+#define MMC_TXPAUSEFRAMES_LO		0x0894
+#define MMC_TXPAUSEFRAMES_HI		0x0898
+#define MMC_TXVLANFRAMES_G_LO		0x089c
+#define MMC_TXVLANFRAMES_G_HI		0x08a0
+#define MMC_RXFRAMECOUNT_GB_LO		0x0900
+#define MMC_RXFRAMECOUNT_GB_HI		0x0904
+#define MMC_RXOCTETCOUNT_GB_LO		0x0908
+#define MMC_RXOCTETCOUNT_GB_HI		0x090c
+#define MMC_RXOCTETCOUNT_G_LO		0x0910
+#define MMC_RXOCTETCOUNT_G_HI		0x0914
+#define MMC_RXBROADCASTFRAMES_G_LO	0x0918
+#define MMC_RXBROADCASTFRAMES_G_HI	0x091c
+#define MMC_RXMULTICASTFRAMES_G_LO	0x0920
+#define MMC_RXMULTICASTFRAMES_G_HI	0x0924
+#define MMC_RXCRCERROR_LO		0x0928
+#define MMC_RXCRCERROR_HI		0x092c
+#define MMC_RXRUNTERROR			0x0930
+#define MMC_RXJABBERERROR		0x0934
+#define MMC_RXUNDERSIZE_G		0x0938
+#define MMC_RXOVERSIZE_G		0x093c
+#define MMC_RX64OCTETS_GB_LO		0x0940
+#define MMC_RX64OCTETS_GB_HI		0x0944
+#define MMC_RX65TO127OCTETS_GB_LO	0x0948
+#define MMC_RX65TO127OCTETS_GB_HI	0x094c
+#define MMC_RX128TO255OCTETS_GB_LO	0x0950
+#define MMC_RX128TO255OCTETS_GB_HI	0x0954
+#define MMC_RX256TO511OCTETS_GB_LO	0x0958
+#define MMC_RX256TO511OCTETS_GB_HI	0x095c
+#define MMC_RX512TO1023OCTETS_GB_LO	0x0960
+#define MMC_RX512TO1023OCTETS_GB_HI	0x0964
+#define MMC_RX1024TOMAXOCTETS_GB_LO	0x0968
+#define MMC_RX1024TOMAXOCTETS_GB_HI	0x096c
+#define MMC_RXUNICASTFRAMES_G_LO	0x0970
+#define MMC_RXUNICASTFRAMES_G_HI	0x0974
+#define MMC_RXLENGTHERROR_LO		0x0978
+#define MMC_RXLENGTHERROR_HI		0x097c
+#define MMC_RXOUTOFRANGETYPE_LO		0x0980
+#define MMC_RXOUTOFRANGETYPE_HI		0x0984
+#define MMC_RXPAUSEFRAMES_LO		0x0988
+#define MMC_RXPAUSEFRAMES_HI		0x098c
+#define MMC_RXFIFOOVERFLOW_LO		0x0990
+#define MMC_RXFIFOOVERFLOW_HI		0x0994
+#define MMC_RXVLANFRAMES_GB_LO		0x0998
+#define MMC_RXVLANFRAMES_GB_HI		0x099c
+#define MMC_RXWATCHDOGERROR		0x09a0
+
+/* MMC register entry bit positions and sizes */
+#define MMC_CR_CR_INDEX				0
+#define MMC_CR_CR_WIDTH				1
+#define MMC_CR_CSR_INDEX			1
+#define MMC_CR_CSR_WIDTH			1
+#define MMC_CR_ROR_INDEX			2
+#define MMC_CR_ROR_WIDTH			1
+#define MMC_CR_MCF_INDEX			3
+#define MMC_CR_MCF_WIDTH			1
+#define MMC_CR_MCT_INDEX			4
+#define MMC_CR_MCT_WIDTH			2
+#define MMC_RIER_ALL_INTERRUPTS_INDEX		0
+#define MMC_RIER_ALL_INTERRUPTS_WIDTH		23
+#define MMC_RISR_RXFRAMECOUNT_GB_INDEX		0
+#define MMC_RISR_RXFRAMECOUNT_GB_WIDTH		1
+#define MMC_RISR_RXOCTETCOUNT_GB_INDEX		1
+#define MMC_RISR_RXOCTETCOUNT_GB_WIDTH		1
+#define MMC_RISR_RXOCTETCOUNT_G_INDEX		2
+#define MMC_RISR_RXOCTETCOUNT_G_WIDTH		1
+#define MMC_RISR_RXBROADCASTFRAMES_G_INDEX	3
+#define MMC_RISR_RXBROADCASTFRAMES_G_WIDTH	1
+#define MMC_RISR_RXMULTICASTFRAMES_G_INDEX	4
+#define MMC_RISR_RXMULTICASTFRAMES_G_WIDTH	1
+#define MMC_RISR_RXCRCERROR_INDEX		5
+#define MMC_RISR_RXCRCERROR_WIDTH		1
+#define MMC_RISR_RXRUNTERROR_INDEX		6
+#define MMC_RISR_RXRUNTERROR_WIDTH		1
+#define MMC_RISR_RXJABBERERROR_INDEX		7
+#define MMC_RISR_RXJABBERERROR_WIDTH		1
+#define MMC_RISR_RXUNDERSIZE_G_INDEX		8
+#define MMC_RISR_RXUNDERSIZE_G_WIDTH		1
+#define MMC_RISR_RXOVERSIZE_G_INDEX		9
+#define MMC_RISR_RXOVERSIZE_G_WIDTH		1
+#define MMC_RISR_RX64OCTETS_GB_INDEX		10
+#define MMC_RISR_RX64OCTETS_GB_WIDTH		1
+#define MMC_RISR_RX65TO127OCTETS_GB_INDEX	11
+#define MMC_RISR_RX65TO127OCTETS_GB_WIDTH	1
+#define MMC_RISR_RX128TO255OCTETS_GB_INDEX	12
+#define MMC_RISR_RX128TO255OCTETS_GB_WIDTH	1
+#define MMC_RISR_RX256TO511OCTETS_GB_INDEX	13
+#define MMC_RISR_RX256TO511OCTETS_GB_WIDTH	1
+#define MMC_RISR_RX512TO1023OCTETS_GB_INDEX	14
+#define MMC_RISR_RX512TO1023OCTETS_GB_WIDTH	1
+#define MMC_RISR_RX1024TOMAXOCTETS_GB_INDEX	15
+#define MMC_RISR_RX1024TOMAXOCTETS_GB_WIDTH	1
+#define MMC_RISR_RXUNICASTFRAMES_G_INDEX	16
+#define MMC_RISR_RXUNICASTFRAMES_G_WIDTH	1
+#define MMC_RISR_RXLENGTHERROR_INDEX		17
+#define MMC_RISR_RXLENGTHERROR_WIDTH		1
+#define MMC_RISR_RXOUTOFRANGETYPE_INDEX		18
+#define MMC_RISR_RXOUTOFRANGETYPE_WIDTH		1
+#define MMC_RISR_RXPAUSEFRAMES_INDEX		19
+#define MMC_RISR_RXPAUSEFRAMES_WIDTH		1
+#define MMC_RISR_RXFIFOOVERFLOW_INDEX		20
+#define MMC_RISR_RXFIFOOVERFLOW_WIDTH		1
+#define MMC_RISR_RXVLANFRAMES_GB_INDEX		21
+#define MMC_RISR_RXVLANFRAMES_GB_WIDTH		1
+#define MMC_RISR_RXWATCHDOGERROR_INDEX		22
+#define MMC_RISR_RXWATCHDOGERROR_WIDTH		1
+#define MMC_TIER_ALL_INTERRUPTS_INDEX		0
+#define MMC_TIER_ALL_INTERRUPTS_WIDTH		18
+#define MMC_TISR_TXOCTETCOUNT_GB_INDEX		0
+#define MMC_TISR_TXOCTETCOUNT_GB_WIDTH		1
+#define MMC_TISR_TXFRAMECOUNT_GB_INDEX		1
+#define MMC_TISR_TXFRAMECOUNT_GB_WIDTH		1
+#define MMC_TISR_TXBROADCASTFRAMES_G_INDEX	2
+#define MMC_TISR_TXBROADCASTFRAMES_G_WIDTH	1
+#define MMC_TISR_TXMULTICASTFRAMES_G_INDEX	3
+#define MMC_TISR_TXMULTICASTFRAMES_G_WIDTH	1
+#define MMC_TISR_TX64OCTETS_GB_INDEX		4
+#define MMC_TISR_TX64OCTETS_GB_WIDTH		1
+#define MMC_TISR_TX65TO127OCTETS_GB_INDEX	5
+#define MMC_TISR_TX65TO127OCTETS_GB_WIDTH	1
+#define MMC_TISR_TX128TO255OCTETS_GB_INDEX	6
+#define MMC_TISR_TX128TO255OCTETS_GB_WIDTH	1
+#define MMC_TISR_TX256TO511OCTETS_GB_INDEX	7
+#define MMC_TISR_TX256TO511OCTETS_GB_WIDTH	1
+#define MMC_TISR_TX512TO1023OCTETS_GB_INDEX	8
+#define MMC_TISR_TX512TO1023OCTETS_GB_WIDTH	1
+#define MMC_TISR_TX1024TOMAXOCTETS_GB_INDEX	9
+#define MMC_TISR_TX1024TOMAXOCTETS_GB_WIDTH	1
+#define MMC_TISR_TXUNICASTFRAMES_GB_INDEX	10
+#define MMC_TISR_TXUNICASTFRAMES_GB_WIDTH	1
+#define MMC_TISR_TXMULTICASTFRAMES_GB_INDEX	11
+#define MMC_TISR_TXMULTICASTFRAMES_GB_WIDTH	1
+#define MMC_TISR_TXBROADCASTFRAMES_GB_INDEX	12
+#define MMC_TISR_TXBROADCASTFRAMES_GB_WIDTH	1
+#define MMC_TISR_TXUNDERFLOWERROR_INDEX		13
+#define MMC_TISR_TXUNDERFLOWERROR_WIDTH		1
+#define MMC_TISR_TXOCTETCOUNT_G_INDEX		14
+#define MMC_TISR_TXOCTETCOUNT_G_WIDTH		1
+#define MMC_TISR_TXFRAMECOUNT_G_INDEX		15
+#define MMC_TISR_TXFRAMECOUNT_G_WIDTH		1
+#define MMC_TISR_TXPAUSEFRAMES_INDEX		16
+#define MMC_TISR_TXPAUSEFRAMES_WIDTH		1
+#define MMC_TISR_TXVLANFRAMES_G_INDEX		17
+#define MMC_TISR_TXVLANFRAMES_G_WIDTH		1
+
+/* MTL register offsets */
+#define MTL_OMR				0x1000
+#define MTL_FDCR			0x1008
+#define MTL_FDSR			0x100c
+#define MTL_FDDR			0x1010
+#define MTL_ISR				0x1020
+#define MTL_RQDCM0R			0x1030
+#define MTL_TCPM0R			0x1040
+#define MTL_TCPM1R			0x1044
+
+#define MTL_RQDCM_INC			4
+#define MTL_RQDCM_Q_PER_REG		4
+
+/* MTL register entry bit positions and sizes */
+#define MTL_OMR_ETSALG_INDEX		5
+#define MTL_OMR_ETSALG_WIDTH		2
+#define MTL_OMR_RAA_INDEX		2
+#define MTL_OMR_RAA_WIDTH		1
+
+/* MTL queue register offsets
+ *   Multiple queues can be active.  The first queue has registers
+ *   that begin at 0x1100.  Each subsequent queue has registers that
+ *   are accessed using an offset of 0x80 from the previous queue.
+ */
+#define MTL_Q_BASE			0x1100
+#define MTL_Q_INC			0x80
+
+#define MTL_Q_TQOMR			0x00
+#define MTL_Q_TQUR			0x04
+#define MTL_Q_TQDR			0x08
+#define MTL_Q_TCECR			0x10
+#define MTL_Q_TCESR			0x14
+#define MTL_Q_TCQWR			0x18
+#define MTL_Q_RQOMR			0x40
+#define MTL_Q_RQMPOCR			0x44
+#define MTL_Q_RQDR			0x4c
+#define MTL_Q_IER			0x70
+#define MTL_Q_ISR			0x74
+
+/* MTL queue register entry bit positions and sizes */
+#define MTL_Q_TCQWR_QW_INDEX		0
+#define MTL_Q_TCQWR_QW_WIDTH		21
+#define MTL_Q_RQOMR_EHFC_INDEX		7
+#define MTL_Q_RQOMR_EHFC_WIDTH		1
+#define MTL_Q_RQOMR_RFA_INDEX		8
+#define MTL_Q_RQOMR_RFA_WIDTH		3
+#define MTL_Q_RQOMR_RFD_INDEX		13
+#define MTL_Q_RQOMR_RFD_WIDTH		3
+#define MTL_Q_RQOMR_RQS_INDEX		16
+#define MTL_Q_RQOMR_RQS_WIDTH		9
+#define MTL_Q_RQOMR_RSF_INDEX		5
+#define MTL_Q_RQOMR_RSF_WIDTH		1
+#define MTL_Q_RQOMR_RTC_INDEX		0
+#define MTL_Q_RQOMR_RTC_WIDTH		2
+#define MTL_Q_TQOMR_FTQ_INDEX		0
+#define MTL_Q_TQOMR_FTQ_WIDTH		1
+#define MTL_Q_TQOMR_TQS_INDEX		16
+#define MTL_Q_TQOMR_TQS_WIDTH		10
+#define MTL_Q_TQOMR_TSF_INDEX		1
+#define MTL_Q_TQOMR_TSF_WIDTH		1
+#define MTL_Q_TQOMR_TTC_INDEX		4
+#define MTL_Q_TQOMR_TTC_WIDTH		3
+#define MTL_Q_TQOMR_TXQEN_INDEX		2
+#define MTL_Q_TQOMR_TXQEN_WIDTH		2
+
+/* MTL queue register value */
+#define MTL_RSF_DISABLE			0x00
+#define MTL_RSF_ENABLE			0x01
+#define MTL_TSF_DISABLE			0x00
+#define MTL_TSF_ENABLE			0x01
+
+#define MTL_RX_THRESHOLD_64		0x00
+#define MTL_RX_THRESHOLD_96		0x02
+#define MTL_RX_THRESHOLD_128		0x03
+#define MTL_TX_THRESHOLD_32		0x01
+#define MTL_TX_THRESHOLD_64		0x00
+#define MTL_TX_THRESHOLD_96		0x02
+#define MTL_TX_THRESHOLD_128		0x03
+#define MTL_TX_THRESHOLD_192		0x04
+#define MTL_TX_THRESHOLD_256		0x05
+#define MTL_TX_THRESHOLD_384		0x06
+#define MTL_TX_THRESHOLD_512		0x07
+
+#define MTL_ETSALG_WRR			0x00
+#define MTL_ETSALG_WFQ			0x01
+#define MTL_ETSALG_DWRR			0x02
+#define MTL_RAA_SP			0x00
+#define MTL_RAA_WSP			0x01
+
+#define MTL_Q_DISABLED			0x00
+#define MTL_Q_ENABLED			0x02
+
+
+/* MTL traffic class register offsets
+ *   Multiple traffic classes can be active.  The first class has registers
+ *   that begin at 0x1100.  Each subsequent queue has registers that
+ *   are accessed using an offset of 0x80 from the previous queue.
+ */
+#define MTL_TC_BASE			MTL_Q_BASE
+#define MTL_TC_INC			MTL_Q_INC
+
+#define MTL_TC_ETSCR			0x10
+
+/* MTL traffic class register entry bit positions and sizes */
+#define MTL_TC_ETSCR_TSA_INDEX		0
+#define MTL_TC_ETSCR_TSA_WIDTH		2
+
+/* MTL traffic class register value */
+#define MTL_TSA_SP			0x00
+#define MTL_TSA_ETS			0x02
+
+
+/* PCS MMD select register offset
+ *  The MMD select register is used for accessing PCS registers
+ *  when the underlying APB3 interface is using indirect addressing.
+ *  Indirect addressing requires accessing registers in two phases,
+ *  an address phase and a data phase.  The address phases requires
+ *  writing an address selection value to the MMD select regiesters.
+ */
+#define PCS_MMD_SELECT			0xff
+
+
+/* Descriptor/Packet entry bit positions and sizes */
+#define RX_PACKET_ERRORS_CRC_INDEX		2
+#define RX_PACKET_ERRORS_CRC_WIDTH		1
+#define RX_PACKET_ERRORS_FRAME_INDEX		3
+#define RX_PACKET_ERRORS_FRAME_WIDTH		1
+#define RX_PACKET_ERRORS_LENGTH_INDEX		0
+#define RX_PACKET_ERRORS_LENGTH_WIDTH		1
+#define RX_PACKET_ERRORS_OVERRUN_INDEX		1
+#define RX_PACKET_ERRORS_OVERRUN_WIDTH		1
+
+#define RX_PACKET_ATTRIBUTES_CSUM_DONE_INDEX	0
+#define RX_PACKET_ATTRIBUTES_CSUM_DONE_WIDTH	1
+#define RX_PACKET_ATTRIBUTES_VLAN_CTAG_INDEX	1
+#define RX_PACKET_ATTRIBUTES_VLAN_CTAG_WIDTH	1
+#define RX_PACKET_ATTRIBUTES_INCOMPLETE_INDEX	2
+#define RX_PACKET_ATTRIBUTES_INCOMPLETE_WIDTH	1
+
+#define RX_NORMAL_DESC0_OVT_INDEX		0
+#define RX_NORMAL_DESC0_OVT_WIDTH		16
+#define RX_NORMAL_DESC3_ES_INDEX		15
+#define RX_NORMAL_DESC3_ES_WIDTH		1
+#define RX_NORMAL_DESC3_ETLT_INDEX		16
+#define RX_NORMAL_DESC3_ETLT_WIDTH		4
+#define RX_NORMAL_DESC3_INTE_INDEX		30
+#define RX_NORMAL_DESC3_INTE_WIDTH		1
+#define RX_NORMAL_DESC3_LD_INDEX		28
+#define RX_NORMAL_DESC3_LD_WIDTH		1
+#define RX_NORMAL_DESC3_OWN_INDEX		31
+#define RX_NORMAL_DESC3_OWN_WIDTH		1
+#define RX_NORMAL_DESC3_PL_INDEX		0
+#define RX_NORMAL_DESC3_PL_WIDTH		14
+
+#define TX_PACKET_ATTRIBUTES_CSUM_ENABLE_INDEX	0
+#define TX_PACKET_ATTRIBUTES_CSUM_ENABLE_WIDTH	1
+#define TX_PACKET_ATTRIBUTES_TSO_ENABLE_INDEX	1
+#define TX_PACKET_ATTRIBUTES_TSO_ENABLE_WIDTH	1
+#define TX_PACKET_ATTRIBUTES_VLAN_CTAG_INDEX	2
+#define TX_PACKET_ATTRIBUTES_VLAN_CTAG_WIDTH	1
+
+#define TX_CONTEXT_DESC2_MSS_INDEX		0
+#define TX_CONTEXT_DESC2_MSS_WIDTH		15
+#define TX_CONTEXT_DESC3_CTXT_INDEX		30
+#define TX_CONTEXT_DESC3_CTXT_WIDTH		1
+#define TX_CONTEXT_DESC3_TCMSSV_INDEX		26
+#define TX_CONTEXT_DESC3_TCMSSV_WIDTH		1
+#define TX_CONTEXT_DESC3_VLTV_INDEX		16
+#define TX_CONTEXT_DESC3_VLTV_WIDTH		1
+#define TX_CONTEXT_DESC3_VT_INDEX		0
+#define TX_CONTEXT_DESC3_VT_WIDTH		16
+
+#define TX_NORMAL_DESC2_HL_B1L_INDEX		0
+#define TX_NORMAL_DESC2_HL_B1L_WIDTH		14
+#define TX_NORMAL_DESC2_IC_INDEX		31
+#define TX_NORMAL_DESC2_IC_WIDTH		1
+#define TX_NORMAL_DESC2_VTIR_INDEX		14
+#define TX_NORMAL_DESC2_VTIR_WIDTH		2
+#define TX_NORMAL_DESC3_CIC_INDEX		16
+#define TX_NORMAL_DESC3_CIC_WIDTH		2
+#define TX_NORMAL_DESC3_CPC_INDEX		26
+#define TX_NORMAL_DESC3_CPC_WIDTH		2
+#define TX_NORMAL_DESC3_CTXT_INDEX		30
+#define TX_NORMAL_DESC3_CTXT_WIDTH		1
+#define TX_NORMAL_DESC3_FD_INDEX		29
+#define TX_NORMAL_DESC3_FD_WIDTH		1
+#define TX_NORMAL_DESC3_FL_INDEX		0
+#define TX_NORMAL_DESC3_FL_WIDTH		15
+#define TX_NORMAL_DESC3_LD_INDEX		28
+#define TX_NORMAL_DESC3_LD_WIDTH		1
+#define TX_NORMAL_DESC3_OWN_INDEX		31
+#define TX_NORMAL_DESC3_OWN_WIDTH		1
+#define TX_NORMAL_DESC3_TCPHDRLEN_INDEX		19
+#define TX_NORMAL_DESC3_TCPHDRLEN_WIDTH		4
+#define TX_NORMAL_DESC3_TCPPL_INDEX		0
+#define TX_NORMAL_DESC3_TCPPL_WIDTH		18
+#define TX_NORMAL_DESC3_TSE_INDEX		18
+#define TX_NORMAL_DESC3_TSE_WIDTH		1
+
+#define TX_NORMAL_DESC2_VLAN_INSERT		0x2
+
+/* MDIO undefined or vendor specific registers */
+#ifndef MDIO_AN_COMP_STAT
+#define MDIO_AN_COMP_STAT		0x0030
+#endif
+
+
+/* Bit setting and getting macros
+ *  The get macro will extract the current bit field value from within
+ *  the variable
+ *
+ *  The set macro will clear the current bit field value within the
+ *  variable and then set the bit field of the variable to the
+ *  specified value
+ */
+#define GET_BITS(_var, _index, _width)					\
+	(((_var) >> (_index)) & ((0x1 << (_width)) - 1))
+
+#define SET_BITS(_var, _index, _width, _val)				\
+do {									\
+	(_var) &= ~(((0x1 << (_width)) - 1) << (_index));		\
+	(_var) |= (((_val) & ((0x1 << (_width)) - 1)) << (_index));	\
+} while (0)
+
+#define GET_BITS_LE(_var, _index, _width)				\
+	((le32_to_cpu((_var)) >> (_index)) & ((0x1 << (_width)) - 1))
+
+#define SET_BITS_LE(_var, _index, _width, _val)				\
+do {									\
+	(_var) &= cpu_to_le32(~(((0x1 << (_width)) - 1) << (_index)));	\
+	(_var) |= cpu_to_le32((((_val) &				\
+			      ((0x1 << (_width)) - 1)) << (_index)));	\
+} while (0)
+
+
+/* Bit setting and getting macros based on register fields
+ *  The get macro uses the bit field definitions formed using the input
+ *  names to extract the current bit field value from within the
+ *  variable
+ *
+ *  The set macro uses the bit field definitions formed using the input
+ *  names to set the bit field of the variable to the specified value
+ */
+#define XGMAC_GET_BITS(_var, _prefix, _field)				\
+	GET_BITS((_var),						\
+		 _prefix##_##_field##_INDEX,				\
+		 _prefix##_##_field##_WIDTH)
+
+#define XGMAC_SET_BITS(_var, _prefix, _field, _val)			\
+	SET_BITS((_var),						\
+		 _prefix##_##_field##_INDEX,				\
+		 _prefix##_##_field##_WIDTH, (_val))
+
+#define XGMAC_GET_BITS_LE(_var, _prefix, _field)			\
+	GET_BITS_LE((_var),						\
+		 _prefix##_##_field##_INDEX,				\
+		 _prefix##_##_field##_WIDTH)
+
+#define XGMAC_SET_BITS_LE(_var, _prefix, _field, _val)			\
+	SET_BITS_LE((_var),						\
+		 _prefix##_##_field##_INDEX,				\
+		 _prefix##_##_field##_WIDTH, (_val))
+
+
+/* Macros for reading or writing registers
+ *  The ioread macros will get bit fields or full values using the
+ *  register definitions formed using the input names
+ *
+ *  The iowrite macros will set bit fields or full values using the
+ *  register definitions formed using the input names
+ */
+#define XGMAC_IOREAD(_pdata, _reg)					\
+	ioread32((_pdata)->xgmac_regs + _reg)
+
+#define XGMAC_IOREAD_BITS(_pdata, _reg, _field)				\
+	GET_BITS(XGMAC_IOREAD((_pdata), _reg),				\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH)
+
+#define XGMAC_IOWRITE(_pdata, _reg, _val)				\
+	iowrite32((_val), (_pdata)->xgmac_regs + _reg)
+
+#define XGMAC_IOWRITE_BITS(_pdata, _reg, _field, _val)			\
+do {									\
+	u32 reg_val = XGMAC_IOREAD((_pdata), _reg);			\
+	SET_BITS(reg_val,						\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH, (_val));			\
+	XGMAC_IOWRITE((_pdata), _reg, reg_val);				\
+} while (0)
+
+
+/* Macros for reading or writing MTL queue or traffic class registers
+ *  Similar to the standard read and write macros except that the
+ *  base register value is calculated by the queue or traffic class number
+ */
+#define XGMAC_MTL_IOREAD(_pdata, _n, _reg)				\
+	ioread32((_pdata)->xgmac_regs +					\
+		 MTL_Q_BASE + ((_n) * MTL_Q_INC) + _reg)
+
+#define XGMAC_MTL_IOREAD_BITS(_pdata, _n, _reg, _field)			\
+	GET_BITS(XGMAC_MTL_IOREAD((_pdata), (_n), _reg),		\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH)
+
+#define XGMAC_MTL_IOWRITE(_pdata, _n, _reg, _val)			\
+	iowrite32((_val), (_pdata)->xgmac_regs +			\
+		  MTL_Q_BASE + ((_n) * MTL_Q_INC) + _reg)
+
+#define XGMAC_MTL_IOWRITE_BITS(_pdata, _n, _reg, _field, _val)		\
+do {									\
+	u32 reg_val = XGMAC_MTL_IOREAD((_pdata), (_n), _reg);		\
+	SET_BITS(reg_val,						\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH, (_val));			\
+	XGMAC_MTL_IOWRITE((_pdata), (_n), _reg, reg_val);		\
+} while (0)
+
+
+/* Macros for reading or writing DMA channel registers
+ *  Similar to the standard read and write macros except that the
+ *  base register value is obtained from the ring
+ */
+#define XGMAC_DMA_IOREAD(_channel, _reg)				\
+	ioread32((_channel)->dma_regs + _reg)
+
+#define XGMAC_DMA_IOREAD_BITS(_channel, _reg, _field)			\
+	GET_BITS(XGMAC_DMA_IOREAD((_channel), _reg),			\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH)
+
+#define XGMAC_DMA_IOWRITE(_channel, _reg, _val)				\
+	iowrite32((_val), (_channel)->dma_regs + _reg)
+
+#define XGMAC_DMA_IOWRITE_BITS(_channel, _reg, _field, _val)		\
+do {									\
+	u32 reg_val = XGMAC_DMA_IOREAD((_channel), _reg);		\
+	SET_BITS(reg_val,						\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH, (_val));			\
+	XGMAC_DMA_IOWRITE((_channel), _reg, reg_val);			\
+} while (0)
+
+
+/* Macros for building, reading or writing register values or bits
+ * within the register values of XPCS registers.
+ */
+#define XPCS_IOWRITE(_pdata, _off, _val)				\
+	iowrite32(_val, (_pdata)->xpcs_regs + (_off))
+
+#define XPCS_IOREAD(_pdata, _off)					\
+	ioread32((_pdata)->xpcs_regs + (_off))
+
+
+/* Macros for building, reading or writing register values or bits
+ * using MDIO.  Different from above because of the use of standardized
+ * Linux include values.  No shifting is performed with the bit
+ * operations, everything works on mask values.
+ */
+#define XMDIO_READ(_pdata, _mmd, _reg)					\
+	((_pdata)->hw_if.read_mmd_regs((_pdata), 0,			\
+		MII_ADDR_C45 | (_mmd << 16) | ((_reg) & 0xffff)))
+
+#define XMDIO_READ_BITS(_pdata, _mmd, _reg, _mask)			\
+	(XMDIO_READ((_pdata), _mmd, _reg) & _mask)
+
+#define XMDIO_WRITE(_pdata, _mmd, _reg, _val)				\
+	((_pdata)->hw_if.write_mmd_regs((_pdata), 0,			\
+		MII_ADDR_C45 | (_mmd << 16) | ((_reg) & 0xffff), (_val)))
+
+#define XMDIO_WRITE_BITS(_pdata, _mmd, _reg, _mask, _val)		\
+do {									\
+	u32 mmd_val = XMDIO_READ((_pdata), _mmd, _reg);			\
+	mmd_val &= ~_mask;						\
+	mmd_val |= (_val);						\
+	XMDIO_WRITE((_pdata), _mmd, _reg, mmd_val);			\
+} while (0)
+
+#endif
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c b/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c
new file mode 100644
index 000000000000..6bb76d5c817b
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-debugfs.c
@@ -0,0 +1,375 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/debugfs.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+static ssize_t xgbe_common_read(char __user *buffer, size_t count,
+				loff_t *ppos, unsigned int value)
+{
+	char *buf;
+	ssize_t len;
+
+	if (*ppos != 0)
+		return 0;
+
+	buf = kasprintf(GFP_KERNEL, "0x%08x\n", value);
+	if (!buf)
+		return -ENOMEM;
+
+	if (count < strlen(buf)) {
+		kfree(buf);
+		return -ENOSPC;
+	}
+
+	len = simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf));
+	kfree(buf);
+
+	return len;
+}
+
+static ssize_t xgbe_common_write(const char __user *buffer, size_t count,
+				 loff_t *ppos, unsigned int *value)
+{
+	char workarea[32];
+	ssize_t len;
+	unsigned int scan_value;
+
+	if (*ppos != 0)
+		return 0;
+
+	if (count >= sizeof(workarea))
+		return -ENOSPC;
+
+	len = simple_write_to_buffer(workarea, sizeof(workarea) - 1, ppos,
+				     buffer, count);
+	if (len < 0)
+		return len;
+
+	workarea[len] = '\0';
+	if (sscanf(workarea, "%x", &scan_value) == 1)
+		*value = scan_value;
+	else
+		return -EIO;
+
+	return len;
+}
+
+static ssize_t xgmac_reg_addr_read(struct file *filp, char __user *buffer,
+				   size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+
+	return xgbe_common_read(buffer, count, ppos, pdata->debugfs_xgmac_reg);
+}
+
+static ssize_t xgmac_reg_addr_write(struct file *filp,
+				    const char __user *buffer,
+				    size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+
+	return xgbe_common_write(buffer, count, ppos,
+				 &pdata->debugfs_xgmac_reg);
+}
+
+static ssize_t xgmac_reg_value_read(struct file *filp, char __user *buffer,
+				    size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+	unsigned int value;
+
+	value = XGMAC_IOREAD(pdata, pdata->debugfs_xgmac_reg);
+
+	return xgbe_common_read(buffer, count, ppos, value);
+}
+
+static ssize_t xgmac_reg_value_write(struct file *filp,
+				     const char __user *buffer,
+				     size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+	unsigned int value;
+	ssize_t len;
+
+	len = xgbe_common_write(buffer, count, ppos, &value);
+	if (len < 0)
+		return len;
+
+	XGMAC_IOWRITE(pdata, pdata->debugfs_xgmac_reg, value);
+
+	return len;
+}
+
+static const struct file_operations xgmac_reg_addr_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read =  xgmac_reg_addr_read,
+	.write = xgmac_reg_addr_write,
+};
+
+static const struct file_operations xgmac_reg_value_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read =  xgmac_reg_value_read,
+	.write = xgmac_reg_value_write,
+};
+
+static ssize_t xpcs_mmd_read(struct file *filp, char __user *buffer,
+			     size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+
+	return xgbe_common_read(buffer, count, ppos, pdata->debugfs_xpcs_mmd);
+}
+
+static ssize_t xpcs_mmd_write(struct file *filp, const char __user *buffer,
+			      size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+
+	return xgbe_common_write(buffer, count, ppos,
+				 &pdata->debugfs_xpcs_mmd);
+}
+
+static ssize_t xpcs_reg_addr_read(struct file *filp, char __user *buffer,
+				  size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+
+	return xgbe_common_read(buffer, count, ppos, pdata->debugfs_xpcs_reg);
+}
+
+static ssize_t xpcs_reg_addr_write(struct file *filp, const char __user *buffer,
+				   size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+
+	return xgbe_common_write(buffer, count, ppos,
+				 &pdata->debugfs_xpcs_reg);
+}
+
+static ssize_t xpcs_reg_value_read(struct file *filp, char __user *buffer,
+				   size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+	unsigned int value;
+
+	value = pdata->hw_if.read_mmd_regs(pdata, pdata->debugfs_xpcs_mmd,
+					   pdata->debugfs_xpcs_reg);
+
+	return xgbe_common_read(buffer, count, ppos, value);
+}
+
+static ssize_t xpcs_reg_value_write(struct file *filp,
+				    const char __user *buffer,
+				    size_t count, loff_t *ppos)
+{
+	struct xgbe_prv_data *pdata = filp->private_data;
+	unsigned int value;
+	ssize_t len;
+
+	len = xgbe_common_write(buffer, count, ppos, &value);
+	if (len < 0)
+		return len;
+
+	pdata->hw_if.write_mmd_regs(pdata, pdata->debugfs_xpcs_mmd,
+				    pdata->debugfs_xpcs_reg, value);
+
+	return len;
+}
+
+static const struct file_operations xpcs_mmd_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read =  xpcs_mmd_read,
+	.write = xpcs_mmd_write,
+};
+
+static const struct file_operations xpcs_reg_addr_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read =  xpcs_reg_addr_read,
+	.write = xpcs_reg_addr_write,
+};
+
+static const struct file_operations xpcs_reg_value_fops = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.read =  xpcs_reg_value_read,
+	.write = xpcs_reg_value_write,
+};
+
+void xgbe_debugfs_init(struct xgbe_prv_data *pdata)
+{
+	struct dentry *pfile;
+	char *buf;
+
+	/* Set defaults */
+	pdata->debugfs_xgmac_reg = 0;
+	pdata->debugfs_xpcs_mmd = 1;
+	pdata->debugfs_xpcs_reg = 0;
+
+	buf = kasprintf(GFP_KERNEL, "amd-xgbe-%s", pdata->netdev->name);
+	pdata->xgbe_debugfs = debugfs_create_dir(buf, NULL);
+	if (pdata->xgbe_debugfs == NULL) {
+		netdev_err(pdata->netdev, "debugfs_create_dir failed\n");
+		return;
+	}
+
+	pfile = debugfs_create_file("xgmac_register", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xgmac_reg_addr_fops);
+	if (!pfile)
+		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+
+	pfile = debugfs_create_file("xgmac_register_value", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xgmac_reg_value_fops);
+	if (!pfile)
+		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+
+	pfile = debugfs_create_file("xpcs_mmd", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xpcs_mmd_fops);
+	if (!pfile)
+		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+
+	pfile = debugfs_create_file("xpcs_register", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xpcs_reg_addr_fops);
+	if (!pfile)
+		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+
+	pfile = debugfs_create_file("xpcs_register_value", 0600,
+				    pdata->xgbe_debugfs, pdata,
+				    &xpcs_reg_value_fops);
+	if (!pfile)
+		netdev_err(pdata->netdev, "debugfs_create_file failed\n");
+
+	kfree(buf);
+}
+
+void xgbe_debugfs_exit(struct xgbe_prv_data *pdata)
+{
+	debugfs_remove_recursive(pdata->xgbe_debugfs);
+	pdata->xgbe_debugfs = NULL;
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
new file mode 100644
index 000000000000..6f1c85956d50
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
@@ -0,0 +1,556 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+static void xgbe_unmap_skb(struct xgbe_prv_data *, struct xgbe_ring_data *);
+
+static void xgbe_free_ring(struct xgbe_prv_data *pdata,
+			   struct xgbe_ring *ring)
+{
+	struct xgbe_ring_data *rdata;
+	unsigned int i;
+
+	if (!ring)
+		return;
+
+	if (ring->rdata) {
+		for (i = 0; i < ring->rdesc_count; i++) {
+			rdata = GET_DESC_DATA(ring, i);
+			xgbe_unmap_skb(pdata, rdata);
+		}
+
+		kfree(ring->rdata);
+		ring->rdata = NULL;
+	}
+
+	if (ring->rdesc) {
+		dma_free_coherent(pdata->dev,
+				  (sizeof(struct xgbe_ring_desc) *
+				   ring->rdesc_count),
+				  ring->rdesc, ring->rdesc_dma);
+		ring->rdesc = NULL;
+	}
+}
+
+static void xgbe_free_ring_resources(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	DBGPR("-->xgbe_free_ring_resources\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		xgbe_free_ring(pdata, channel->tx_ring);
+		xgbe_free_ring(pdata, channel->rx_ring);
+	}
+
+	DBGPR("<--xgbe_free_ring_resources\n");
+}
+
+static int xgbe_init_ring(struct xgbe_prv_data *pdata,
+			  struct xgbe_ring *ring, unsigned int rdesc_count)
+{
+	DBGPR("-->xgbe_init_ring\n");
+
+	if (!ring)
+		return 0;
+
+	/* Descriptors */
+	ring->rdesc_count = rdesc_count;
+	ring->rdesc = dma_alloc_coherent(pdata->dev,
+					 (sizeof(struct xgbe_ring_desc) *
+					  rdesc_count), &ring->rdesc_dma,
+					 GFP_KERNEL);
+	if (!ring->rdesc)
+		return -ENOMEM;
+
+	/* Descriptor information */
+	ring->rdata = kcalloc(rdesc_count, sizeof(struct xgbe_ring_data),
+			      GFP_KERNEL);
+	if (!ring->rdata)
+		return -ENOMEM;
+
+	DBGPR("    rdesc=0x%p, rdesc_dma=0x%llx, rdata=0x%p\n",
+	      ring->rdesc, ring->rdesc_dma, ring->rdata);
+
+	DBGPR("<--xgbe_init_ring\n");
+
+	return 0;
+}
+
+static int xgbe_alloc_ring_resources(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+	int ret;
+
+	DBGPR("-->xgbe_alloc_ring_resources\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		DBGPR("  %s - tx_ring:\n", channel->name);
+		ret = xgbe_init_ring(pdata, channel->tx_ring,
+				     pdata->tx_desc_count);
+		if (ret) {
+			netdev_alert(pdata->netdev,
+				     "error initializing Tx ring\n");
+			goto err_ring;
+		}
+
+		DBGPR("  %s - rx_ring:\n", channel->name);
+		ret = xgbe_init_ring(pdata, channel->rx_ring,
+				     pdata->rx_desc_count);
+		if (ret) {
+			netdev_alert(pdata->netdev,
+				     "error initializing Tx ring\n");
+			goto err_ring;
+		}
+	}
+
+	DBGPR("<--xgbe_alloc_ring_resources\n");
+
+	return 0;
+
+err_ring:
+	xgbe_free_ring_resources(pdata);
+
+	return ret;
+}
+
+static void xgbe_wrapper_tx_descriptor_init(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_channel *channel;
+	struct xgbe_ring *ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+	dma_addr_t rdesc_dma;
+	unsigned int i, j;
+
+	DBGPR("-->xgbe_wrapper_tx_descriptor_init\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		ring = channel->tx_ring;
+		if (!ring)
+			break;
+
+		rdesc = ring->rdesc;
+		rdesc_dma = ring->rdesc_dma;
+
+		for (j = 0; j < ring->rdesc_count; j++) {
+			rdata = GET_DESC_DATA(ring, j);
+
+			rdata->rdesc = rdesc;
+			rdata->rdesc_dma = rdesc_dma;
+
+			rdesc++;
+			rdesc_dma += sizeof(struct xgbe_ring_desc);
+		}
+
+		ring->cur = 0;
+		ring->dirty = 0;
+		ring->tx.queue_stopped = 0;
+
+		hw_if->tx_desc_init(channel);
+	}
+
+	DBGPR("<--xgbe_wrapper_tx_descriptor_init\n");
+}
+
+static void xgbe_wrapper_rx_descriptor_init(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_channel *channel;
+	struct xgbe_ring *ring;
+	struct xgbe_ring_desc *rdesc;
+	struct xgbe_ring_data *rdata;
+	dma_addr_t rdesc_dma, skb_dma;
+	struct sk_buff *skb = NULL;
+	unsigned int i, j;
+
+	DBGPR("-->xgbe_wrapper_rx_descriptor_init\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		ring = channel->rx_ring;
+		if (!ring)
+			break;
+
+		rdesc = ring->rdesc;
+		rdesc_dma = ring->rdesc_dma;
+
+		for (j = 0; j < ring->rdesc_count; j++) {
+			rdata = GET_DESC_DATA(ring, j);
+
+			rdata->rdesc = rdesc;
+			rdata->rdesc_dma = rdesc_dma;
+
+			/* Allocate skb & assign to each rdesc */
+			skb = dev_alloc_skb(pdata->rx_buf_size);
+			if (skb == NULL)
+				break;
+			skb_dma = dma_map_single(pdata->dev, skb->data,
+						 pdata->rx_buf_size,
+						 DMA_FROM_DEVICE);
+			if (dma_mapping_error(pdata->dev, skb_dma)) {
+				netdev_alert(pdata->netdev,
+					     "failed to do the dma map\n");
+				dev_kfree_skb_any(skb);
+				break;
+			}
+			rdata->skb = skb;
+			rdata->skb_dma = skb_dma;
+			rdata->skb_dma_len = pdata->rx_buf_size;
+
+			rdesc++;
+			rdesc_dma += sizeof(struct xgbe_ring_desc);
+		}
+
+		ring->cur = 0;
+		ring->dirty = 0;
+		ring->rx.realloc_index = 0;
+		ring->rx.realloc_threshold = 0;
+
+		hw_if->rx_desc_init(channel);
+	}
+
+	DBGPR("<--xgbe_wrapper_rx_descriptor_init\n");
+}
+
+static void xgbe_unmap_skb(struct xgbe_prv_data *pdata,
+			   struct xgbe_ring_data *rdata)
+{
+	if (rdata->skb_dma) {
+		if (rdata->mapped_as_page) {
+			dma_unmap_page(pdata->dev, rdata->skb_dma,
+				       rdata->skb_dma_len, DMA_TO_DEVICE);
+		} else {
+			dma_unmap_single(pdata->dev, rdata->skb_dma,
+					 rdata->skb_dma_len, DMA_TO_DEVICE);
+		}
+		rdata->skb_dma = 0;
+		rdata->skb_dma_len = 0;
+	}
+
+	if (rdata->skb) {
+		dev_kfree_skb_any(rdata->skb);
+		rdata->skb = NULL;
+	}
+
+	rdata->tso_header = 0;
+	rdata->len = 0;
+	rdata->interrupt = 0;
+	rdata->mapped_as_page = 0;
+}
+
+static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_ring *ring = channel->tx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_packet_data *packet;
+	struct skb_frag_struct *frag;
+	dma_addr_t skb_dma;
+	unsigned int start_index, cur_index;
+	unsigned int offset, tso, vlan, datalen, len;
+	unsigned int i;
+
+	DBGPR("-->xgbe_map_tx_skb: cur = %d\n", ring->cur);
+
+	offset = 0;
+	start_index = ring->cur;
+	cur_index = ring->cur;
+
+	packet = &ring->packet_data;
+	packet->rdesc_count = 0;
+	packet->length = 0;
+
+	tso = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			     TSO_ENABLE);
+	vlan = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			      VLAN_CTAG);
+
+	/* Save space for a context descriptor if needed */
+	if ((tso && (packet->mss != ring->tx.cur_mss)) ||
+	    (vlan && (packet->vlan_ctag != ring->tx.cur_vlan_ctag)))
+		cur_index++;
+	rdata = GET_DESC_DATA(ring, cur_index);
+
+	if (tso) {
+		DBGPR("  TSO packet\n");
+
+		/* Map the TSO header */
+		skb_dma = dma_map_single(pdata->dev, skb->data,
+					 packet->header_len, DMA_TO_DEVICE);
+		if (dma_mapping_error(pdata->dev, skb_dma)) {
+			netdev_alert(pdata->netdev, "dma_map_single failed\n");
+			goto err_out;
+		}
+		rdata->skb_dma = skb_dma;
+		rdata->skb_dma_len = packet->header_len;
+		rdata->tso_header = 1;
+
+		offset = packet->header_len;
+
+		packet->length += packet->header_len;
+
+		cur_index++;
+		rdata = GET_DESC_DATA(ring, cur_index);
+	}
+
+	/* Map the (remainder of the) packet */
+	for (datalen = skb_headlen(skb) - offset; datalen; ) {
+		len = min_t(unsigned int, datalen, TX_MAX_BUF_SIZE);
+
+		skb_dma = dma_map_single(pdata->dev, skb->data + offset, len,
+					 DMA_TO_DEVICE);
+		if (dma_mapping_error(pdata->dev, skb_dma)) {
+			netdev_alert(pdata->netdev, "dma_map_single failed\n");
+			goto err_out;
+		}
+		rdata->skb_dma = skb_dma;
+		rdata->skb_dma_len = len;
+		DBGPR("  skb data: index=%u, dma=0x%llx, len=%u\n",
+		      cur_index, skb_dma, len);
+
+		datalen -= len;
+		offset += len;
+
+		packet->length += len;
+
+		cur_index++;
+		rdata = GET_DESC_DATA(ring, cur_index);
+	}
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		DBGPR("  mapping frag %u\n", i);
+
+		frag = &skb_shinfo(skb)->frags[i];
+		offset = 0;
+
+		for (datalen = skb_frag_size(frag); datalen; ) {
+			len = min_t(unsigned int, datalen, TX_MAX_BUF_SIZE);
+
+			skb_dma = skb_frag_dma_map(pdata->dev, frag, offset,
+						   len, DMA_TO_DEVICE);
+			if (dma_mapping_error(pdata->dev, skb_dma)) {
+				netdev_alert(pdata->netdev,
+					     "skb_frag_dma_map failed\n");
+				goto err_out;
+			}
+			rdata->skb_dma = skb_dma;
+			rdata->skb_dma_len = len;
+			rdata->mapped_as_page = 1;
+			DBGPR("  skb data: index=%u, dma=0x%llx, len=%u\n",
+			      cur_index, skb_dma, len);
+
+			datalen -= len;
+			offset += len;
+
+			packet->length += len;
+
+			cur_index++;
+			rdata = GET_DESC_DATA(ring, cur_index);
+		}
+	}
+
+	/* Save the skb address in the last entry */
+	rdata->skb = skb;
+
+	/* Save the number of descriptor entries used */
+	packet->rdesc_count = cur_index - start_index;
+
+	DBGPR("<--xgbe_map_tx_skb: count=%u\n", packet->rdesc_count);
+
+	return packet->rdesc_count;
+
+err_out:
+	while (start_index < cur_index) {
+		rdata = GET_DESC_DATA(ring, start_index++);
+		xgbe_unmap_skb(pdata, rdata);
+	}
+
+	DBGPR("<--xgbe_map_tx_skb: count=0\n");
+
+	return 0;
+}
+
+static void xgbe_realloc_skb(struct xgbe_channel *channel)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_ring *ring = channel->rx_ring;
+	struct xgbe_ring_data *rdata;
+	struct sk_buff *skb = NULL;
+	dma_addr_t skb_dma;
+	int i;
+
+	DBGPR("-->xgbe_realloc_skb: rx_ring->rx.realloc_index = %u\n",
+	      ring->rx.realloc_index);
+
+	for (i = 0; i < ring->dirty; i++) {
+		rdata = GET_DESC_DATA(ring, ring->rx.realloc_index);
+
+		/* Reset rdata values */
+		xgbe_unmap_skb(pdata, rdata);
+
+		/* Allocate skb & assign to each rdesc */
+		skb = dev_alloc_skb(pdata->rx_buf_size);
+		if (skb == NULL) {
+			netdev_alert(pdata->netdev,
+				     "failed to allocate skb\n");
+			break;
+		}
+		skb_dma = dma_map_single(pdata->dev, skb->data,
+					 pdata->rx_buf_size, DMA_FROM_DEVICE);
+		if (dma_mapping_error(pdata->dev, skb_dma)) {
+			netdev_alert(pdata->netdev,
+				     "failed to do the dma map\n");
+			dev_kfree_skb_any(skb);
+			break;
+		}
+		rdata->skb = skb;
+		rdata->skb_dma = skb_dma;
+		rdata->skb_dma_len = pdata->rx_buf_size;
+
+		hw_if->rx_desc_reset(rdata);
+
+		ring->rx.realloc_index++;
+	}
+	ring->dirty = 0;
+
+	DBGPR("<--xgbe_realloc_skb\n");
+}
+
+void xgbe_init_function_ptrs_desc(struct xgbe_desc_if *desc_if)
+{
+	DBGPR("-->xgbe_init_function_ptrs_desc\n");
+
+	desc_if->alloc_ring_resources = xgbe_alloc_ring_resources;
+	desc_if->free_ring_resources = xgbe_free_ring_resources;
+	desc_if->map_tx_skb = xgbe_map_tx_skb;
+	desc_if->realloc_skb = xgbe_realloc_skb;
+	desc_if->unmap_skb = xgbe_unmap_skb;
+	desc_if->wrapper_tx_desc_init = xgbe_wrapper_tx_descriptor_init;
+	desc_if->wrapper_rx_desc_init = xgbe_wrapper_rx_descriptor_init;
+
+	DBGPR("<--xgbe_init_function_ptrs_desc\n");
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
new file mode 100644
index 000000000000..002293b0819d
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -0,0 +1,2182 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/phy.h>
+#include <linux/clk.h>
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+static unsigned int xgbe_usec_to_riwt(struct xgbe_prv_data *pdata,
+				      unsigned int usec)
+{
+	unsigned long rate;
+	unsigned int ret;
+
+	DBGPR("-->xgbe_usec_to_riwt\n");
+
+	rate = clk_get_rate(pdata->sysclock);
+
+	/*
+	 * Convert the input usec value to the watchdog timer value. Each
+	 * watchdog timer value is equivalent to 256 clock cycles.
+	 * Calculate the required value as:
+	 *   ( usec * ( system_clock_mhz / 10^6 ) / 256
+	 */
+	ret = (usec * (rate / 1000000)) / 256;
+
+	DBGPR("<--xgbe_usec_to_riwt\n");
+
+	return ret;
+}
+
+static unsigned int xgbe_riwt_to_usec(struct xgbe_prv_data *pdata,
+				      unsigned int riwt)
+{
+	unsigned long rate;
+	unsigned int ret;
+
+	DBGPR("-->xgbe_riwt_to_usec\n");
+
+	rate = clk_get_rate(pdata->sysclock);
+
+	/*
+	 * Convert the input watchdog timer value to the usec value. Each
+	 * watchdog timer value is equivalent to 256 clock cycles.
+	 * Calculate the required value as:
+	 *   ( riwt * 256 ) / ( system_clock_mhz / 10^6 )
+	 */
+	ret = (riwt * 256) / (rate / 1000000);
+
+	DBGPR("<--xgbe_riwt_to_usec\n");
+
+	return ret;
+}
+
+static int xgbe_config_pblx8(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++)
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_CR, PBLX8,
+				       pdata->pblx8);
+
+	return 0;
+}
+
+static int xgbe_get_tx_pbl_val(struct xgbe_prv_data *pdata)
+{
+	return XGMAC_DMA_IOREAD_BITS(pdata->channel, DMA_CH_TCR, PBL);
+}
+
+static int xgbe_config_tx_pbl_val(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, PBL,
+				       pdata->tx_pbl);
+	}
+
+	return 0;
+}
+
+static int xgbe_get_rx_pbl_val(struct xgbe_prv_data *pdata)
+{
+	return XGMAC_DMA_IOREAD_BITS(pdata->channel, DMA_CH_RCR, PBL);
+}
+
+static int xgbe_config_rx_pbl_val(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, PBL,
+				       pdata->rx_pbl);
+	}
+
+	return 0;
+}
+
+static int xgbe_config_osp_mode(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, OSP,
+				       pdata->tx_osp_mode);
+	}
+
+	return 0;
+}
+
+static int xgbe_config_rsf_mode(struct xgbe_prv_data *pdata, unsigned int val)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, RSF, val);
+
+	return 0;
+}
+
+static int xgbe_config_tsf_mode(struct xgbe_prv_data *pdata, unsigned int val)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TSF, val);
+
+	return 0;
+}
+
+static int xgbe_config_rx_threshold(struct xgbe_prv_data *pdata,
+				    unsigned int val)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, RTC, val);
+
+	return 0;
+}
+
+static int xgbe_config_tx_threshold(struct xgbe_prv_data *pdata,
+				    unsigned int val)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TTC, val);
+
+	return 0;
+}
+
+static int xgbe_config_rx_coalesce(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RIWT, RWT,
+				       pdata->rx_riwt);
+	}
+
+	return 0;
+}
+
+static int xgbe_config_tx_coalesce(struct xgbe_prv_data *pdata)
+{
+	return 0;
+}
+
+static void xgbe_config_rx_buffer_size(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, RBSZ,
+				       pdata->rx_buf_size);
+	}
+}
+
+static void xgbe_config_tso_mode(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, TSE, 1);
+	}
+}
+
+static int xgbe_disable_tx_flow_control(struct xgbe_prv_data *pdata)
+{
+	unsigned int max_q_count, q_count;
+	unsigned int reg, reg_val;
+	unsigned int i;
+
+	/* Clear MTL flow control */
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, EHFC, 0);
+
+	/* Clear MAC flow control */
+	max_q_count = XGMAC_MAX_FLOW_CONTROL_QUEUES;
+	q_count = min_t(unsigned int, pdata->hw_feat.rx_q_cnt, max_q_count);
+	reg = MAC_Q0TFCR;
+	for (i = 0; i < q_count; i++) {
+		reg_val = XGMAC_IOREAD(pdata, reg);
+		XGMAC_SET_BITS(reg_val, MAC_Q0TFCR, TFE, 0);
+		XGMAC_IOWRITE(pdata, reg, reg_val);
+
+		reg += MAC_QTFCR_INC;
+	}
+
+	return 0;
+}
+
+static int xgbe_enable_tx_flow_control(struct xgbe_prv_data *pdata)
+{
+	unsigned int max_q_count, q_count;
+	unsigned int reg, reg_val;
+	unsigned int i;
+
+	/* Set MTL flow control */
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, EHFC, 1);
+
+	/* Set MAC flow control */
+	max_q_count = XGMAC_MAX_FLOW_CONTROL_QUEUES;
+	q_count = min_t(unsigned int, pdata->hw_feat.rx_q_cnt, max_q_count);
+	reg = MAC_Q0TFCR;
+	for (i = 0; i < q_count; i++) {
+		reg_val = XGMAC_IOREAD(pdata, reg);
+
+		/* Enable transmit flow control */
+		XGMAC_SET_BITS(reg_val, MAC_Q0TFCR, TFE, 1);
+		/* Set pause time */
+		XGMAC_SET_BITS(reg_val, MAC_Q0TFCR, PT, 0xffff);
+
+		XGMAC_IOWRITE(pdata, reg, reg_val);
+
+		reg += MAC_QTFCR_INC;
+	}
+
+	return 0;
+}
+
+static int xgbe_disable_rx_flow_control(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_RFCR, RFE, 0);
+
+	return 0;
+}
+
+static int xgbe_enable_rx_flow_control(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_RFCR, RFE, 1);
+
+	return 0;
+}
+
+static int xgbe_config_tx_flow_control(struct xgbe_prv_data *pdata)
+{
+	if (pdata->tx_pause)
+		xgbe_enable_tx_flow_control(pdata);
+	else
+		xgbe_disable_tx_flow_control(pdata);
+
+	return 0;
+}
+
+static int xgbe_config_rx_flow_control(struct xgbe_prv_data *pdata)
+{
+	if (pdata->rx_pause)
+		xgbe_enable_rx_flow_control(pdata);
+	else
+		xgbe_disable_rx_flow_control(pdata);
+
+	return 0;
+}
+
+static void xgbe_config_flow_control(struct xgbe_prv_data *pdata)
+{
+	xgbe_config_tx_flow_control(pdata);
+	xgbe_config_rx_flow_control(pdata);
+}
+
+static void xgbe_enable_dma_interrupts(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int dma_ch_isr, dma_ch_ier;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		/* Clear all the interrupts which are set */
+		dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR);
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_SR, dma_ch_isr);
+
+		/* Clear all interrupt enable bits */
+		dma_ch_ier = 0;
+
+		/* Enable following interrupts
+		 *   NIE  - Normal Interrupt Summary Enable
+		 *   AIE  - Abnormal Interrupt Summary Enable
+		 *   FBEE - Fatal Bus Error Enable
+		 */
+		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, NIE, 1);
+		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, AIE, 1);
+		XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, FBEE, 1);
+
+		if (channel->tx_ring) {
+			/* Enable the following Tx interrupts
+			 *   TIE  - Transmit Interrupt Enable (unless polling)
+			 */
+			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 1);
+		}
+		if (channel->rx_ring) {
+			/* Enable following Rx interrupts
+			 *   RBUE - Receive Buffer Unavailable Enable
+			 *   RIE  - Receive Interrupt Enable
+			 */
+			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RBUE, 1);
+			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 1);
+		}
+
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
+	}
+}
+
+static void xgbe_enable_mtl_interrupts(struct xgbe_prv_data *pdata)
+{
+	unsigned int mtl_q_isr;
+	unsigned int q_count, i;
+
+	q_count = max(pdata->hw_feat.tx_q_cnt, pdata->hw_feat.rx_q_cnt);
+	for (i = 0; i < q_count; i++) {
+		/* Clear all the interrupts which are set */
+		mtl_q_isr = XGMAC_MTL_IOREAD(pdata, i, MTL_Q_ISR);
+		XGMAC_MTL_IOWRITE(pdata, i, MTL_Q_ISR, mtl_q_isr);
+
+		/* No MTL interrupts to be enabled */
+		XGMAC_MTL_IOWRITE(pdata, i, MTL_Q_ISR, 0);
+	}
+}
+
+static void xgbe_enable_mac_interrupts(struct xgbe_prv_data *pdata)
+{
+	/* No MAC interrupts to be enabled */
+	XGMAC_IOWRITE(pdata, MAC_IER, 0);
+
+	/* Enable all counter interrupts */
+	XGMAC_IOWRITE_BITS(pdata, MMC_RIER, ALL_INTERRUPTS, 0xff);
+	XGMAC_IOWRITE_BITS(pdata, MMC_TIER, ALL_INTERRUPTS, 0xff);
+}
+
+static int xgbe_set_gmii_speed(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, SS, 0x3);
+
+	return 0;
+}
+
+static int xgbe_set_gmii_2500_speed(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, SS, 0x2);
+
+	return 0;
+}
+
+static int xgbe_set_xgmii_speed(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, SS, 0);
+
+	return 0;
+}
+
+static int xgbe_set_promiscuous_mode(struct xgbe_prv_data *pdata,
+				     unsigned int enable)
+{
+	unsigned int val = enable ? 1 : 0;
+
+	if (XGMAC_IOREAD_BITS(pdata, MAC_PFR, PR) == val)
+		return 0;
+
+	DBGPR("  %s promiscuous mode\n", enable ? "entering" : "leaving");
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, PR, val);
+
+	return 0;
+}
+
+static int xgbe_set_all_multicast_mode(struct xgbe_prv_data *pdata,
+				       unsigned int enable)
+{
+	unsigned int val = enable ? 1 : 0;
+
+	if (XGMAC_IOREAD_BITS(pdata, MAC_PFR, PM) == val)
+		return 0;
+
+	DBGPR("  %s allmulti mode\n", enable ? "entering" : "leaving");
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, PM, val);
+
+	return 0;
+}
+
+static int xgbe_set_addn_mac_addrs(struct xgbe_prv_data *pdata,
+				   unsigned int am_mode)
+{
+	struct netdev_hw_addr *ha;
+	unsigned int mac_reg;
+	unsigned int mac_addr_hi, mac_addr_lo;
+	u8 *mac_addr;
+	unsigned int i;
+
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, HUC, 0);
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, HMC, 0);
+
+	i = 0;
+	mac_reg = MAC_MACA1HR;
+
+	netdev_for_each_uc_addr(ha, pdata->netdev) {
+		mac_addr_lo = 0;
+		mac_addr_hi = 0;
+		mac_addr = (u8 *)&mac_addr_lo;
+		mac_addr[0] = ha->addr[0];
+		mac_addr[1] = ha->addr[1];
+		mac_addr[2] = ha->addr[2];
+		mac_addr[3] = ha->addr[3];
+		mac_addr = (u8 *)&mac_addr_hi;
+		mac_addr[0] = ha->addr[4];
+		mac_addr[1] = ha->addr[5];
+
+		DBGPR("  adding unicast address %pM at 0x%04x\n",
+		      ha->addr, mac_reg);
+
+		XGMAC_SET_BITS(mac_addr_hi, MAC_MACA1HR, AE, 1);
+
+		XGMAC_IOWRITE(pdata, mac_reg, mac_addr_hi);
+		mac_reg += MAC_MACA_INC;
+		XGMAC_IOWRITE(pdata, mac_reg, mac_addr_lo);
+		mac_reg += MAC_MACA_INC;
+
+		i++;
+	}
+
+	if (!am_mode) {
+		netdev_for_each_mc_addr(ha, pdata->netdev) {
+			mac_addr_lo = 0;
+			mac_addr_hi = 0;
+			mac_addr = (u8 *)&mac_addr_lo;
+			mac_addr[0] = ha->addr[0];
+			mac_addr[1] = ha->addr[1];
+			mac_addr[2] = ha->addr[2];
+			mac_addr[3] = ha->addr[3];
+			mac_addr = (u8 *)&mac_addr_hi;
+			mac_addr[0] = ha->addr[4];
+			mac_addr[1] = ha->addr[5];
+
+			DBGPR("  adding multicast address %pM at 0x%04x\n",
+			      ha->addr, mac_reg);
+
+			XGMAC_SET_BITS(mac_addr_hi, MAC_MACA1HR, AE, 1);
+
+			XGMAC_IOWRITE(pdata, mac_reg, mac_addr_hi);
+			mac_reg += MAC_MACA_INC;
+			XGMAC_IOWRITE(pdata, mac_reg, mac_addr_lo);
+			mac_reg += MAC_MACA_INC;
+
+			i++;
+		}
+	}
+
+	/* Clear remaining additional MAC address entries */
+	for (; i < pdata->hw_feat.addn_mac; i++) {
+		XGMAC_IOWRITE(pdata, mac_reg, 0);
+		mac_reg += MAC_MACA_INC;
+		XGMAC_IOWRITE(pdata, mac_reg, 0);
+		mac_reg += MAC_MACA_INC;
+	}
+
+	return 0;
+}
+
+static int xgbe_set_mac_address(struct xgbe_prv_data *pdata, u8 *addr)
+{
+	unsigned int mac_addr_hi, mac_addr_lo;
+
+	mac_addr_hi = (addr[5] <<  8) | (addr[4] <<  0);
+	mac_addr_lo = (addr[3] << 24) | (addr[2] << 16) |
+		      (addr[1] <<  8) | (addr[0] <<  0);
+
+	XGMAC_IOWRITE(pdata, MAC_MACA0HR, mac_addr_hi);
+	XGMAC_IOWRITE(pdata, MAC_MACA0LR, mac_addr_lo);
+
+	return 0;
+}
+
+static int xgbe_read_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
+			      int mmd_reg)
+{
+	unsigned int mmd_address;
+	int mmd_data;
+
+	if (mmd_reg & MII_ADDR_C45)
+		mmd_address = mmd_reg & ~MII_ADDR_C45;
+	else
+		mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+
+	/* The PCS registers are accessed using mmio. The underlying APB3
+	 * management interface uses indirect addressing to access the MMD
+	 * register sets. This requires accessing of the PCS register in two
+	 * phases, an address phase and a data phase.
+	 *
+	 * The mmio interface is based on 32-bit offsets and values. All
+	 * register offsets must therefore be adjusted by left shifting the
+	 * offset 2 bits and reading 32 bits of data.
+	 */
+	mutex_lock(&pdata->xpcs_mutex);
+	XPCS_IOWRITE(pdata, PCS_MMD_SELECT << 2, mmd_address >> 8);
+	mmd_data = XPCS_IOREAD(pdata, (mmd_address & 0xff) << 2);
+	mutex_unlock(&pdata->xpcs_mutex);
+
+	return mmd_data;
+}
+
+static void xgbe_write_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
+				int mmd_reg, int mmd_data)
+{
+	unsigned int mmd_address;
+
+	if (mmd_reg & MII_ADDR_C45)
+		mmd_address = mmd_reg & ~MII_ADDR_C45;
+	else
+		mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff);
+
+	/* The PCS registers are accessed using mmio. The underlying APB3
+	 * management interface uses indirect addressing to access the MMD
+	 * register sets. This requires accessing of the PCS register in two
+	 * phases, an address phase and a data phase.
+	 *
+	 * The mmio interface is based on 32-bit offsets and values. All
+	 * register offsets must therefore be adjusted by left shifting the
+	 * offset 2 bits and reading 32 bits of data.
+	 */
+	mutex_lock(&pdata->xpcs_mutex);
+	XPCS_IOWRITE(pdata, PCS_MMD_SELECT << 2, mmd_address >> 8);
+	XPCS_IOWRITE(pdata, (mmd_address & 0xff) << 2, mmd_data);
+	mutex_unlock(&pdata->xpcs_mutex);
+}
+
+static int xgbe_tx_complete(struct xgbe_ring_desc *rdesc)
+{
+	return !XGMAC_GET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, OWN);
+}
+
+static int xgbe_disable_rx_csum(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, IPC, 0);
+
+	return 0;
+}
+
+static int xgbe_enable_rx_csum(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, IPC, 1);
+
+	return 0;
+}
+
+static int xgbe_enable_rx_vlan_stripping(struct xgbe_prv_data *pdata)
+{
+	/* Put the VLAN tag in the Rx descriptor */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLRXS, 1);
+
+	/* Don't check the VLAN type */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, DOVLTC, 1);
+
+	/* Check only C-TAG (0x8100) packets */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ERSVLM, 0);
+
+	/* Don't consider an S-TAG (0x88A8) packet as a VLAN packet */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ESVL, 0);
+
+	/* Enable VLAN tag stripping */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLS, 0x3);
+
+	return 0;
+}
+
+static int xgbe_disable_rx_vlan_stripping(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLS, 0);
+
+	return 0;
+}
+
+static void xgbe_tx_desc_reset(struct xgbe_ring_data *rdata)
+{
+	struct xgbe_ring_desc *rdesc = rdata->rdesc;
+
+	/* Reset the Tx descriptor
+	 *   Set buffer 1 (lo) address to zero
+	 *   Set buffer 1 (hi) address to zero
+	 *   Reset all other control bits (IC, TTSE, B2L & B1L)
+	 *   Reset all other control bits (OWN, CTXT, FD, LD, CPC, CIC, etc)
+	 */
+	rdesc->desc0 = 0;
+	rdesc->desc1 = 0;
+	rdesc->desc2 = 0;
+	rdesc->desc3 = 0;
+}
+
+static void xgbe_tx_desc_init(struct xgbe_channel *channel)
+{
+	struct xgbe_ring *ring = channel->tx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+	int i;
+	int start_index = ring->cur;
+
+	DBGPR("-->tx_desc_init\n");
+
+	/* Initialze all descriptors */
+	for (i = 0; i < ring->rdesc_count; i++) {
+		rdata = GET_DESC_DATA(ring, i);
+		rdesc = rdata->rdesc;
+
+		/* Initialize Tx descriptor
+		 *   Set buffer 1 (lo) address to zero
+		 *   Set buffer 1 (hi) address to zero
+		 *   Reset all other control bits (IC, TTSE, B2L & B1L)
+		 *   Reset all other control bits (OWN, CTXT, FD, LD, CPC, CIC,
+		 *     etc)
+		 */
+		rdesc->desc0 = 0;
+		rdesc->desc1 = 0;
+		rdesc->desc2 = 0;
+		rdesc->desc3 = 0;
+	}
+
+	/* Make sure everything is written to the descriptor(s) before
+	 * telling the device about them
+	 */
+	wmb();
+
+	/* Update the total number of Tx descriptors */
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDRLR, ring->rdesc_count - 1);
+
+	/* Update the starting address of descriptor ring */
+	rdata = GET_DESC_DATA(ring, start_index);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDLR_HI,
+			  upper_32_bits(rdata->rdesc_dma));
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDLR_LO,
+			  lower_32_bits(rdata->rdesc_dma));
+
+	DBGPR("<--tx_desc_init\n");
+}
+
+static void xgbe_rx_desc_reset(struct xgbe_ring_data *rdata)
+{
+	struct xgbe_ring_desc *rdesc = rdata->rdesc;
+
+	/* Reset the Rx descriptor
+	 *   Set buffer 1 (lo) address to dma address (lo)
+	 *   Set buffer 1 (hi) address to dma address (hi)
+	 *   Set buffer 2 (lo) address to zero
+	 *   Set buffer 2 (hi) address to zero and set control bits
+	 *     OWN and INTE
+	 */
+	rdesc->desc0 = cpu_to_le32(lower_32_bits(rdata->skb_dma));
+	rdesc->desc1 = cpu_to_le32(upper_32_bits(rdata->skb_dma));
+	rdesc->desc2 = 0;
+
+	rdesc->desc3 = 0;
+	if (rdata->interrupt)
+		XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE, 1);
+
+	/* Since the Rx DMA engine is likely running, make sure everything
+	 * is written to the descriptor(s) before setting the OWN bit
+	 * for the descriptor
+	 */
+	wmb();
+
+	XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, OWN, 1);
+
+	/* Make sure ownership is written to the descriptor */
+	wmb();
+}
+
+static void xgbe_rx_desc_init(struct xgbe_channel *channel)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_ring *ring = channel->rx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+	unsigned int start_index = ring->cur;
+	unsigned int rx_coalesce, rx_frames;
+	unsigned int i;
+
+	DBGPR("-->rx_desc_init\n");
+
+	rx_coalesce = (pdata->rx_riwt || pdata->rx_frames) ? 1 : 0;
+	rx_frames = pdata->rx_frames;
+
+	/* Initialize all descriptors */
+	for (i = 0; i < ring->rdesc_count; i++) {
+		rdata = GET_DESC_DATA(ring, i);
+		rdesc = rdata->rdesc;
+
+		/* Initialize Rx descriptor
+		 *   Set buffer 1 (lo) address to dma address (lo)
+		 *   Set buffer 1 (hi) address to dma address (hi)
+		 *   Set buffer 2 (lo) address to zero
+		 *   Set buffer 2 (hi) address to zero and set control
+		 *     bits OWN and INTE appropriateley
+		 */
+		rdesc->desc0 = cpu_to_le32(lower_32_bits(rdata->skb_dma));
+		rdesc->desc1 = cpu_to_le32(upper_32_bits(rdata->skb_dma));
+		rdesc->desc2 = 0;
+		rdesc->desc3 = 0;
+		XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, OWN, 1);
+		XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE, 1);
+		rdata->interrupt = 1;
+		if (rx_coalesce && (!rx_frames || ((i + 1) % rx_frames))) {
+			/* Clear interrupt on completion bit */
+			XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE,
+					  0);
+			rdata->interrupt = 0;
+		}
+	}
+
+	/* Make sure everything is written to the descriptors before
+	 * telling the device about them
+	 */
+	wmb();
+
+	/* Update the total number of Rx descriptors */
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_RDRLR, ring->rdesc_count - 1);
+
+	/* Update the starting address of descriptor ring */
+	rdata = GET_DESC_DATA(ring, start_index);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_RDLR_HI,
+			  upper_32_bits(rdata->rdesc_dma));
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_RDLR_LO,
+			  lower_32_bits(rdata->rdesc_dma));
+
+	/* Update the Rx Descriptor Tail Pointer */
+	rdata = GET_DESC_DATA(ring, start_index + ring->rdesc_count - 1);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_RDTR_LO,
+			  lower_32_bits(rdata->rdesc_dma));
+
+	DBGPR("<--rx_desc_init\n");
+}
+
+static void xgbe_pre_xmit(struct xgbe_channel *channel)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_ring *ring = channel->tx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+	struct xgbe_packet_data *packet = &ring->packet_data;
+	unsigned int csum, tso, vlan;
+	unsigned int tso_context, vlan_context;
+	unsigned int tx_coalesce, tx_frames;
+	int start_index = ring->cur;
+	int i;
+
+	DBGPR("-->xgbe_pre_xmit\n");
+
+	csum = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			      CSUM_ENABLE);
+	tso = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			     TSO_ENABLE);
+	vlan = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			      VLAN_CTAG);
+
+	if (tso && (packet->mss != ring->tx.cur_mss))
+		tso_context = 1;
+	else
+		tso_context = 0;
+
+	if (vlan && (packet->vlan_ctag != ring->tx.cur_vlan_ctag))
+		vlan_context = 1;
+	else
+		vlan_context = 0;
+
+	tx_coalesce = (pdata->tx_usecs || pdata->tx_frames) ? 1 : 0;
+	tx_frames = pdata->tx_frames;
+	if (tx_coalesce && !channel->tx_timer_active)
+		ring->coalesce_count = 0;
+
+	rdata = GET_DESC_DATA(ring, ring->cur);
+	rdesc = rdata->rdesc;
+
+	/* Create a context descriptor if this is a TSO packet */
+	if (tso_context || vlan_context) {
+		if (tso_context) {
+			DBGPR("  TSO context descriptor, mss=%u\n",
+			      packet->mss);
+
+			/* Set the MSS size */
+			XGMAC_SET_BITS_LE(rdesc->desc2, TX_CONTEXT_DESC2,
+					  MSS, packet->mss);
+
+			/* Mark it as a CONTEXT descriptor */
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_CONTEXT_DESC3,
+					  CTXT, 1);
+
+			/* Indicate this descriptor contains the MSS */
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_CONTEXT_DESC3,
+					  TCMSSV, 1);
+
+			ring->tx.cur_mss = packet->mss;
+		}
+
+		if (vlan_context) {
+			DBGPR("  VLAN context descriptor, ctag=%u\n",
+			      packet->vlan_ctag);
+
+			/* Mark it as a CONTEXT descriptor */
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_CONTEXT_DESC3,
+					  CTXT, 1);
+
+			/* Set the VLAN tag */
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_CONTEXT_DESC3,
+					  VT, packet->vlan_ctag);
+
+			/* Indicate this descriptor contains the VLAN tag */
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_CONTEXT_DESC3,
+					  VLTV, 1);
+
+			ring->tx.cur_vlan_ctag = packet->vlan_ctag;
+		}
+
+		ring->cur++;
+		rdata = GET_DESC_DATA(ring, ring->cur);
+		rdesc = rdata->rdesc;
+	}
+
+	/* Update buffer address (for TSO this is the header) */
+	rdesc->desc0 =  cpu_to_le32(lower_32_bits(rdata->skb_dma));
+	rdesc->desc1 =  cpu_to_le32(upper_32_bits(rdata->skb_dma));
+
+	/* Update the buffer length */
+	XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, HL_B1L,
+			  rdata->skb_dma_len);
+
+	/* VLAN tag insertion check */
+	if (vlan)
+		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, VTIR,
+				  TX_NORMAL_DESC2_VLAN_INSERT);
+
+	/* Set IC bit based on Tx coalescing settings */
+	XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 1);
+	if (tx_coalesce && (!tx_frames ||
+			    (++ring->coalesce_count % tx_frames)))
+		/* Clear IC bit */
+		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 0);
+
+	/* Mark it as First Descriptor */
+	XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, FD, 1);
+
+	/* Mark it as a NORMAL descriptor */
+	XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, CTXT, 0);
+
+	/* Set OWN bit if not the first descriptor */
+	if (ring->cur != start_index)
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, OWN, 1);
+
+	if (tso) {
+		/* Enable TSO */
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, TSE, 1);
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, TCPPL,
+				  packet->tcp_payload_len);
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, TCPHDRLEN,
+				  packet->tcp_header_len / 4);
+	} else {
+		/* Enable CRC and Pad Insertion */
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, CPC, 0);
+
+		/* Enable HW CSUM */
+		if (csum)
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3,
+					  CIC, 0x3);
+
+		/* Set the total length to be transmitted */
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, FL,
+				  packet->length);
+	}
+
+	for (i = ring->cur - start_index + 1; i < packet->rdesc_count; i++) {
+		ring->cur++;
+		rdata = GET_DESC_DATA(ring, ring->cur);
+		rdesc = rdata->rdesc;
+
+		/* Update buffer address */
+		rdesc->desc0 = cpu_to_le32(lower_32_bits(rdata->skb_dma));
+		rdesc->desc1 = cpu_to_le32(upper_32_bits(rdata->skb_dma));
+
+		/* Update the buffer length */
+		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, HL_B1L,
+				  rdata->skb_dma_len);
+
+		/* Set IC bit based on Tx coalescing settings */
+		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 1);
+		if (tx_coalesce && (!tx_frames ||
+				    (++ring->coalesce_count % tx_frames)))
+			/* Clear IC bit */
+			XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 0);
+
+		/* Set OWN bit */
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, OWN, 1);
+
+		/* Mark it as NORMAL descriptor */
+		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, CTXT, 0);
+
+		/* Enable HW CSUM */
+		if (csum)
+			XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3,
+					  CIC, 0x3);
+	}
+
+	/* Set LAST bit for the last descriptor */
+	XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, LD, 1);
+
+	/* In case the Tx DMA engine is running, make sure everything
+	 * is written to the descriptor(s) before setting the OWN bit
+	 * for the first descriptor
+	 */
+	wmb();
+
+	/* Set OWN bit for the first descriptor */
+	rdata = GET_DESC_DATA(ring, start_index);
+	rdesc = rdata->rdesc;
+	XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, OWN, 1);
+
+#ifdef XGMAC_ENABLE_TX_DESC_DUMP
+	xgbe_dump_tx_desc(ring, start_index, packet->rdesc_count, 1);
+#endif
+
+	/* Make sure ownership is written to the descriptor */
+	wmb();
+
+	/* Issue a poll command to Tx DMA by writing address
+	 * of next immediate free descriptor */
+	ring->cur++;
+	rdata = GET_DESC_DATA(ring, ring->cur);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDTR_LO,
+			  lower_32_bits(rdata->rdesc_dma));
+
+	/* Start the Tx coalescing timer */
+	if (tx_coalesce && !channel->tx_timer_active) {
+		channel->tx_timer_active = 1;
+		hrtimer_start(&channel->tx_timer,
+			      ktime_set(0, pdata->tx_usecs * NSEC_PER_USEC),
+			      HRTIMER_MODE_REL);
+	}
+
+	DBGPR("  %s: descriptors %u to %u written\n",
+	      channel->name, start_index & (ring->rdesc_count - 1),
+	      (ring->cur - 1) & (ring->rdesc_count - 1));
+
+	DBGPR("<--xgbe_pre_xmit\n");
+}
+
+static int xgbe_dev_read(struct xgbe_channel *channel)
+{
+	struct xgbe_ring *ring = channel->rx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+	struct xgbe_packet_data *packet = &ring->packet_data;
+	unsigned int err, etlt;
+
+	DBGPR("-->xgbe_dev_read: cur = %d\n", ring->cur);
+
+	rdata = GET_DESC_DATA(ring, ring->cur);
+	rdesc = rdata->rdesc;
+
+	/* Check for data availability */
+	if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, OWN))
+		return 1;
+
+#ifdef XGMAC_ENABLE_RX_DESC_DUMP
+	xgbe_dump_rx_desc(ring, rdesc, ring->cur);
+#endif
+
+	/* Get the packet length */
+	rdata->len = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, PL);
+
+	if (!XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, LD)) {
+		/* Not all the data has been transferred for this packet */
+		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+			       INCOMPLETE, 1);
+		return 0;
+	}
+
+	/* This is the last of the data for this packet */
+	XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+		       INCOMPLETE, 0);
+
+	/* Set checksum done indicator as appropriate */
+	if (channel->pdata->netdev->features & NETIF_F_RXCSUM)
+		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+			       CSUM_DONE, 1);
+
+	/* Check for errors (only valid in last descriptor) */
+	err = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, ES);
+	etlt = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, ETLT);
+	DBGPR("  err=%u, etlt=%#x\n", err, etlt);
+
+	if (!err || (err && !etlt)) {
+		if (etlt == 0x09) {
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       VLAN_CTAG, 1);
+			packet->vlan_ctag = XGMAC_GET_BITS_LE(rdesc->desc0,
+							      RX_NORMAL_DESC0,
+							      OVT);
+			DBGPR("  vlan-ctag=0x%04x\n", packet->vlan_ctag);
+		}
+	} else {
+		if ((etlt == 0x05) || (etlt == 0x06))
+			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+				       CSUM_DONE, 0);
+		else
+			XGMAC_SET_BITS(packet->errors, RX_PACKET_ERRORS,
+				       FRAME, 1);
+	}
+
+	DBGPR("<--xgbe_dev_read: %s - descriptor=%u (cur=%d)\n", channel->name,
+	      ring->cur & (ring->rdesc_count - 1), ring->cur);
+
+	return 0;
+}
+
+static int xgbe_is_context_desc(struct xgbe_ring_desc *rdesc)
+{
+	/* Rx and Tx share CTXT bit, so check TDES3.CTXT bit */
+	return XGMAC_GET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, CTXT);
+}
+
+static int xgbe_is_last_desc(struct xgbe_ring_desc *rdesc)
+{
+	/* Rx and Tx share LD bit, so check TDES3.LD bit */
+	return XGMAC_GET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, LD);
+}
+
+static void xgbe_save_interrupt_status(struct xgbe_channel *channel,
+				       enum xgbe_int_state int_state)
+{
+	unsigned int dma_ch_ier;
+
+	if (int_state == XGMAC_INT_STATE_SAVE) {
+		channel->saved_ier = XGMAC_DMA_IOREAD(channel, DMA_CH_IER);
+		channel->saved_ier &= DMA_INTERRUPT_MASK;
+	} else {
+		dma_ch_ier = XGMAC_DMA_IOREAD(channel, DMA_CH_IER);
+		dma_ch_ier |= channel->saved_ier;
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
+	}
+}
+
+static int xgbe_enable_int(struct xgbe_channel *channel,
+			   enum xgbe_int int_id)
+{
+	switch (int_id) {
+	case XGMAC_INT_DMA_ISR_DC0IS:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TIE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_TI:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TIE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_TPS:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TXSE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_TBU:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TBUE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_RI:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, RIE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_RBU:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, RBUE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_RPS:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, RSE, 1);
+		break;
+	case XGMAC_INT_DMA_CH_SR_FBE:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, FBEE, 1);
+		break;
+	case XGMAC_INT_DMA_ALL:
+		xgbe_save_interrupt_status(channel, XGMAC_INT_STATE_RESTORE);
+		break;
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+static int xgbe_disable_int(struct xgbe_channel *channel,
+			    enum xgbe_int int_id)
+{
+	unsigned int dma_ch_ier;
+
+	switch (int_id) {
+	case XGMAC_INT_DMA_ISR_DC0IS:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TIE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_TI:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TIE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_TPS:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TXSE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_TBU:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, TBUE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_RI:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, RIE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_RBU:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, RBUE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_RPS:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, RSE, 0);
+		break;
+	case XGMAC_INT_DMA_CH_SR_FBE:
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_IER, FBEE, 0);
+		break;
+	case XGMAC_INT_DMA_ALL:
+		xgbe_save_interrupt_status(channel, XGMAC_INT_STATE_SAVE);
+
+		dma_ch_ier = XGMAC_DMA_IOREAD(channel, DMA_CH_IER);
+		dma_ch_ier &= ~DMA_INTERRUPT_MASK;
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
+		break;
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+static int xgbe_exit(struct xgbe_prv_data *pdata)
+{
+	unsigned int count = 2000;
+
+	DBGPR("-->xgbe_exit\n");
+
+	/* Issue a software reset */
+	XGMAC_IOWRITE_BITS(pdata, DMA_MR, SWR, 1);
+	usleep_range(10, 15);
+
+	/* Poll Until Poll Condition */
+	while (count-- && XGMAC_IOREAD_BITS(pdata, DMA_MR, SWR))
+		usleep_range(500, 600);
+
+	if (!count)
+		return -EBUSY;
+
+	DBGPR("<--xgbe_exit\n");
+
+	return 0;
+}
+
+static int xgbe_flush_tx_queues(struct xgbe_prv_data *pdata)
+{
+	unsigned int i, count;
+
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, FTQ, 1);
+
+	/* Poll Until Poll Condition */
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++) {
+		count = 2000;
+		while (count-- && XGMAC_MTL_IOREAD_BITS(pdata, i,
+							MTL_Q_TQOMR, FTQ))
+			usleep_range(500, 600);
+
+		if (!count)
+			return -EBUSY;
+	}
+
+	return 0;
+}
+
+static void xgbe_config_dma_bus(struct xgbe_prv_data *pdata)
+{
+	/* Set enhanced addressing mode */
+	XGMAC_IOWRITE_BITS(pdata, DMA_SBMR, EAME, 1);
+
+	/* Set the System Bus mode */
+	XGMAC_IOWRITE_BITS(pdata, DMA_SBMR, UNDEF, 1);
+}
+
+static void xgbe_config_dma_cache(struct xgbe_prv_data *pdata)
+{
+	unsigned int arcache, awcache;
+
+	arcache = 0;
+	XGMAC_SET_BITS(arcache, DMA_AXIARCR, DRC, DMA_ARCACHE_SETTING);
+	XGMAC_SET_BITS(arcache, DMA_AXIARCR, DRD, DMA_ARDOMAIN_SETTING);
+	XGMAC_SET_BITS(arcache, DMA_AXIARCR, TEC, DMA_ARCACHE_SETTING);
+	XGMAC_SET_BITS(arcache, DMA_AXIARCR, TED, DMA_ARDOMAIN_SETTING);
+	XGMAC_SET_BITS(arcache, DMA_AXIARCR, THC, DMA_ARCACHE_SETTING);
+	XGMAC_SET_BITS(arcache, DMA_AXIARCR, THD, DMA_ARDOMAIN_SETTING);
+	XGMAC_IOWRITE(pdata, DMA_AXIARCR, arcache);
+
+	awcache = 0;
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, DWC, DMA_AWCACHE_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, DWD, DMA_AWDOMAIN_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RPC, DMA_AWCACHE_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RPD, DMA_AWDOMAIN_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RHC, DMA_AWCACHE_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, RHD, DMA_AWDOMAIN_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, TDC, DMA_AWCACHE_SETTING);
+	XGMAC_SET_BITS(awcache, DMA_AXIAWCR, TDD, DMA_AWDOMAIN_SETTING);
+	XGMAC_IOWRITE(pdata, DMA_AXIAWCR, awcache);
+}
+
+static void xgbe_config_mtl_mode(struct xgbe_prv_data *pdata)
+{
+	unsigned int i;
+
+	/* Set Tx to weighted round robin scheduling algorithm (when
+	 * traffic class is using ETS algorithm)
+	 */
+	XGMAC_IOWRITE_BITS(pdata, MTL_OMR, ETSALG, MTL_ETSALG_WRR);
+
+	/* Set Tx traffic classes to strict priority algorithm */
+	for (i = 0; i < XGBE_TC_CNT; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_TC_ETSCR, TSA, MTL_TSA_SP);
+
+	/* Set Rx to strict priority algorithm */
+	XGMAC_IOWRITE_BITS(pdata, MTL_OMR, RAA, MTL_RAA_SP);
+}
+
+static unsigned int xgbe_calculate_per_queue_fifo(unsigned long fifo_size,
+						  unsigned char queue_count)
+{
+	unsigned int q_fifo_size = 0;
+	enum xgbe_mtl_fifo_size p_fifo = XGMAC_MTL_FIFO_SIZE_256;
+
+	/* Calculate Tx/Rx fifo share per queue */
+	switch (fifo_size) {
+	case 0:
+		q_fifo_size = FIFO_SIZE_B(128);
+		break;
+	case 1:
+		q_fifo_size = FIFO_SIZE_B(256);
+		break;
+	case 2:
+		q_fifo_size = FIFO_SIZE_B(512);
+		break;
+	case 3:
+		q_fifo_size = FIFO_SIZE_KB(1);
+		break;
+	case 4:
+		q_fifo_size = FIFO_SIZE_KB(2);
+		break;
+	case 5:
+		q_fifo_size = FIFO_SIZE_KB(4);
+		break;
+	case 6:
+		q_fifo_size = FIFO_SIZE_KB(8);
+		break;
+	case 7:
+		q_fifo_size = FIFO_SIZE_KB(16);
+		break;
+	case 8:
+		q_fifo_size = FIFO_SIZE_KB(32);
+		break;
+	case 9:
+		q_fifo_size = FIFO_SIZE_KB(64);
+		break;
+	case 10:
+		q_fifo_size = FIFO_SIZE_KB(128);
+		break;
+	case 11:
+		q_fifo_size = FIFO_SIZE_KB(256);
+		break;
+	}
+	q_fifo_size = q_fifo_size / queue_count;
+
+	/* Set the queue fifo size programmable value */
+	if (q_fifo_size >= FIFO_SIZE_KB(256))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_256K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(128))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_128K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(64))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_64K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(32))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_32K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(16))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_16K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(8))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_8K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(4))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_4K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(2))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_2K;
+	else if (q_fifo_size >= FIFO_SIZE_KB(1))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_1K;
+	else if (q_fifo_size >= FIFO_SIZE_B(512))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_512;
+	else if (q_fifo_size >= FIFO_SIZE_B(256))
+		p_fifo = XGMAC_MTL_FIFO_SIZE_256;
+
+	return p_fifo;
+}
+
+static void xgbe_config_tx_fifo_size(struct xgbe_prv_data *pdata)
+{
+	enum xgbe_mtl_fifo_size fifo_size;
+	unsigned int i;
+
+	fifo_size = xgbe_calculate_per_queue_fifo(pdata->hw_feat.tx_fifo_size,
+						  pdata->hw_feat.tx_q_cnt);
+
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TQS, fifo_size);
+
+	netdev_notice(pdata->netdev, "%d Tx queues, %d byte fifo per queue\n",
+		      pdata->hw_feat.tx_q_cnt, ((fifo_size + 1) * 256));
+}
+
+static void xgbe_config_rx_fifo_size(struct xgbe_prv_data *pdata)
+{
+	enum xgbe_mtl_fifo_size fifo_size;
+	unsigned int i;
+
+	fifo_size = xgbe_calculate_per_queue_fifo(pdata->hw_feat.rx_fifo_size,
+						  pdata->hw_feat.rx_q_cnt);
+
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, RQS, fifo_size);
+
+	netdev_notice(pdata->netdev, "%d Rx queues, %d byte fifo per queue\n",
+		      pdata->hw_feat.rx_q_cnt, ((fifo_size + 1) * 256));
+}
+
+static void xgbe_config_rx_queue_mapping(struct xgbe_prv_data *pdata)
+{
+	unsigned int i, reg, reg_val;
+	unsigned int q_count = pdata->hw_feat.rx_q_cnt;
+
+	/* Select dynamic mapping of MTL Rx queue to DMA Rx channel */
+	reg = MTL_RQDCM0R;
+	reg_val = 0;
+	for (i = 0; i < q_count;) {
+		reg_val |= (0x80 << ((i++ % MTL_RQDCM_Q_PER_REG) << 3));
+
+		if ((i % MTL_RQDCM_Q_PER_REG) && (i != q_count))
+			continue;
+
+		XGMAC_IOWRITE(pdata, reg, reg_val);
+
+		reg += MTL_RQDCM_INC;
+		reg_val = 0;
+	}
+}
+
+static void xgbe_config_flow_control_threshold(struct xgbe_prv_data *pdata)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++) {
+		/* Activate flow control when less than 4k left in fifo */
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, RFA, 2);
+
+		/* De-activate flow control when more than 6k left in fifo */
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, RFD, 4);
+	}
+}
+
+static void xgbe_config_mac_address(struct xgbe_prv_data *pdata)
+{
+	xgbe_set_mac_address(pdata, pdata->netdev->dev_addr);
+}
+
+static void xgbe_config_jumbo_enable(struct xgbe_prv_data *pdata)
+{
+	unsigned int val;
+
+	val = (pdata->netdev->mtu > XGMAC_STD_PACKET_MTU) ? 1 : 0;
+
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, JE, val);
+}
+
+static void xgbe_config_checksum_offload(struct xgbe_prv_data *pdata)
+{
+	if (pdata->netdev->features & NETIF_F_RXCSUM)
+		xgbe_enable_rx_csum(pdata);
+	else
+		xgbe_disable_rx_csum(pdata);
+}
+
+static void xgbe_config_vlan_support(struct xgbe_prv_data *pdata)
+{
+	if (pdata->netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
+		xgbe_enable_rx_vlan_stripping(pdata);
+	else
+		xgbe_disable_rx_vlan_stripping(pdata);
+}
+
+static void xgbe_tx_mmc_int(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_mmc_stats *stats = &pdata->mmc_stats;
+	unsigned int mmc_isr = XGMAC_IOREAD(pdata, MMC_TISR);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXOCTETCOUNT_GB))
+		stats->txoctetcount_gb +=
+			XGMAC_IOREAD(pdata, MMC_TXOCTETCOUNT_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXFRAMECOUNT_GB))
+		stats->txframecount_gb +=
+			XGMAC_IOREAD(pdata, MMC_TXFRAMECOUNT_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXBROADCASTFRAMES_G))
+		stats->txbroadcastframes_g +=
+			XGMAC_IOREAD(pdata, MMC_TXBROADCASTFRAMES_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXMULTICASTFRAMES_G))
+		stats->txmulticastframes_g +=
+			XGMAC_IOREAD(pdata, MMC_TXMULTICASTFRAMES_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TX64OCTETS_GB))
+		stats->tx64octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_TX64OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TX65TO127OCTETS_GB))
+		stats->tx65to127octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_TX65TO127OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TX128TO255OCTETS_GB))
+		stats->tx128to255octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_TX128TO255OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TX256TO511OCTETS_GB))
+		stats->tx256to511octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_TX256TO511OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TX512TO1023OCTETS_GB))
+		stats->tx512to1023octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_TX512TO1023OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TX1024TOMAXOCTETS_GB))
+		stats->tx1024tomaxoctets_gb +=
+			XGMAC_IOREAD(pdata, MMC_TX1024TOMAXOCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXUNICASTFRAMES_GB))
+		stats->txunicastframes_gb +=
+			XGMAC_IOREAD(pdata, MMC_TXUNICASTFRAMES_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXMULTICASTFRAMES_GB))
+		stats->txmulticastframes_gb +=
+			XGMAC_IOREAD(pdata, MMC_TXMULTICASTFRAMES_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXBROADCASTFRAMES_GB))
+		stats->txbroadcastframes_g +=
+			XGMAC_IOREAD(pdata, MMC_TXBROADCASTFRAMES_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXUNDERFLOWERROR))
+		stats->txunderflowerror +=
+			XGMAC_IOREAD(pdata, MMC_TXUNDERFLOWERROR_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXOCTETCOUNT_G))
+		stats->txoctetcount_g +=
+			XGMAC_IOREAD(pdata, MMC_TXOCTETCOUNT_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXFRAMECOUNT_G))
+		stats->txframecount_g +=
+			XGMAC_IOREAD(pdata, MMC_TXFRAMECOUNT_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXPAUSEFRAMES))
+		stats->txpauseframes +=
+			XGMAC_IOREAD(pdata, MMC_TXPAUSEFRAMES_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_TISR, TXVLANFRAMES_G))
+		stats->txvlanframes_g +=
+			XGMAC_IOREAD(pdata, MMC_TXVLANFRAMES_G_LO);
+}
+
+static void xgbe_rx_mmc_int(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_mmc_stats *stats = &pdata->mmc_stats;
+	unsigned int mmc_isr = XGMAC_IOREAD(pdata, MMC_RISR);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXFRAMECOUNT_GB))
+		stats->rxframecount_gb +=
+			XGMAC_IOREAD(pdata, MMC_RXFRAMECOUNT_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXOCTETCOUNT_GB))
+		stats->rxoctetcount_gb +=
+			XGMAC_IOREAD(pdata, MMC_RXOCTETCOUNT_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXOCTETCOUNT_G))
+		stats->rxoctetcount_g +=
+			XGMAC_IOREAD(pdata, MMC_RXOCTETCOUNT_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXBROADCASTFRAMES_G))
+		stats->rxbroadcastframes_g +=
+			XGMAC_IOREAD(pdata, MMC_RXBROADCASTFRAMES_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXMULTICASTFRAMES_G))
+		stats->rxmulticastframes_g +=
+			XGMAC_IOREAD(pdata, MMC_RXMULTICASTFRAMES_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXCRCERROR))
+		stats->rxcrcerror +=
+			XGMAC_IOREAD(pdata, MMC_RXCRCERROR_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXRUNTERROR))
+		stats->rxrunterror +=
+			XGMAC_IOREAD(pdata, MMC_RXRUNTERROR);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXJABBERERROR))
+		stats->rxjabbererror +=
+			XGMAC_IOREAD(pdata, MMC_RXJABBERERROR);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXUNDERSIZE_G))
+		stats->rxundersize_g +=
+			XGMAC_IOREAD(pdata, MMC_RXUNDERSIZE_G);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXOVERSIZE_G))
+		stats->rxoversize_g +=
+			XGMAC_IOREAD(pdata, MMC_RXOVERSIZE_G);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RX64OCTETS_GB))
+		stats->rx64octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_RX64OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RX65TO127OCTETS_GB))
+		stats->rx65to127octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_RX65TO127OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RX128TO255OCTETS_GB))
+		stats->rx128to255octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_RX128TO255OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RX256TO511OCTETS_GB))
+		stats->rx256to511octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_RX256TO511OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RX512TO1023OCTETS_GB))
+		stats->rx512to1023octets_gb +=
+			XGMAC_IOREAD(pdata, MMC_RX512TO1023OCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RX1024TOMAXOCTETS_GB))
+		stats->rx1024tomaxoctets_gb +=
+			XGMAC_IOREAD(pdata, MMC_RX1024TOMAXOCTETS_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXUNICASTFRAMES_G))
+		stats->rxunicastframes_g +=
+			XGMAC_IOREAD(pdata, MMC_RXUNICASTFRAMES_G_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXLENGTHERROR))
+		stats->rxlengtherror +=
+			XGMAC_IOREAD(pdata, MMC_RXLENGTHERROR_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXOUTOFRANGETYPE))
+		stats->rxoutofrangetype +=
+			XGMAC_IOREAD(pdata, MMC_RXOUTOFRANGETYPE_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXPAUSEFRAMES))
+		stats->rxpauseframes +=
+			XGMAC_IOREAD(pdata, MMC_RXPAUSEFRAMES_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXFIFOOVERFLOW))
+		stats->rxfifooverflow +=
+			XGMAC_IOREAD(pdata, MMC_RXFIFOOVERFLOW_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXVLANFRAMES_GB))
+		stats->rxvlanframes_gb +=
+			XGMAC_IOREAD(pdata, MMC_RXVLANFRAMES_GB_LO);
+
+	if (XGMAC_GET_BITS(mmc_isr, MMC_RISR, RXWATCHDOGERROR))
+		stats->rxwatchdogerror +=
+			XGMAC_IOREAD(pdata, MMC_RXWATCHDOGERROR);
+}
+
+static void xgbe_read_mmc_stats(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_mmc_stats *stats = &pdata->mmc_stats;
+
+	/* Freeze counters */
+	XGMAC_IOWRITE_BITS(pdata, MMC_CR, MCF, 1);
+
+	stats->txoctetcount_gb +=
+		XGMAC_IOREAD(pdata, MMC_TXOCTETCOUNT_GB_LO);
+
+	stats->txframecount_gb +=
+		XGMAC_IOREAD(pdata, MMC_TXFRAMECOUNT_GB_LO);
+
+	stats->txbroadcastframes_g +=
+		XGMAC_IOREAD(pdata, MMC_TXBROADCASTFRAMES_G_LO);
+
+	stats->txmulticastframes_g +=
+		XGMAC_IOREAD(pdata, MMC_TXMULTICASTFRAMES_G_LO);
+
+	stats->tx64octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_TX64OCTETS_GB_LO);
+
+	stats->tx65to127octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_TX65TO127OCTETS_GB_LO);
+
+	stats->tx128to255octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_TX128TO255OCTETS_GB_LO);
+
+	stats->tx256to511octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_TX256TO511OCTETS_GB_LO);
+
+	stats->tx512to1023octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_TX512TO1023OCTETS_GB_LO);
+
+	stats->tx1024tomaxoctets_gb +=
+		XGMAC_IOREAD(pdata, MMC_TX1024TOMAXOCTETS_GB_LO);
+
+	stats->txunicastframes_gb +=
+		XGMAC_IOREAD(pdata, MMC_TXUNICASTFRAMES_GB_LO);
+
+	stats->txmulticastframes_gb +=
+		XGMAC_IOREAD(pdata, MMC_TXMULTICASTFRAMES_GB_LO);
+
+	stats->txbroadcastframes_g +=
+		XGMAC_IOREAD(pdata, MMC_TXBROADCASTFRAMES_GB_LO);
+
+	stats->txunderflowerror +=
+		XGMAC_IOREAD(pdata, MMC_TXUNDERFLOWERROR_LO);
+
+	stats->txoctetcount_g +=
+		XGMAC_IOREAD(pdata, MMC_TXOCTETCOUNT_G_LO);
+
+	stats->txframecount_g +=
+		XGMAC_IOREAD(pdata, MMC_TXFRAMECOUNT_G_LO);
+
+	stats->txpauseframes +=
+		XGMAC_IOREAD(pdata, MMC_TXPAUSEFRAMES_LO);
+
+	stats->txvlanframes_g +=
+		XGMAC_IOREAD(pdata, MMC_TXVLANFRAMES_G_LO);
+
+	stats->rxframecount_gb +=
+		XGMAC_IOREAD(pdata, MMC_RXFRAMECOUNT_GB_LO);
+
+	stats->rxoctetcount_gb +=
+		XGMAC_IOREAD(pdata, MMC_RXOCTETCOUNT_GB_LO);
+
+	stats->rxoctetcount_g +=
+		XGMAC_IOREAD(pdata, MMC_RXOCTETCOUNT_G_LO);
+
+	stats->rxbroadcastframes_g +=
+		XGMAC_IOREAD(pdata, MMC_RXBROADCASTFRAMES_G_LO);
+
+	stats->rxmulticastframes_g +=
+		XGMAC_IOREAD(pdata, MMC_RXMULTICASTFRAMES_G_LO);
+
+	stats->rxcrcerror +=
+		XGMAC_IOREAD(pdata, MMC_RXCRCERROR_LO);
+
+	stats->rxrunterror +=
+		XGMAC_IOREAD(pdata, MMC_RXRUNTERROR);
+
+	stats->rxjabbererror +=
+		XGMAC_IOREAD(pdata, MMC_RXJABBERERROR);
+
+	stats->rxundersize_g +=
+		XGMAC_IOREAD(pdata, MMC_RXUNDERSIZE_G);
+
+	stats->rxoversize_g +=
+		XGMAC_IOREAD(pdata, MMC_RXOVERSIZE_G);
+
+	stats->rx64octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_RX64OCTETS_GB_LO);
+
+	stats->rx65to127octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_RX65TO127OCTETS_GB_LO);
+
+	stats->rx128to255octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_RX128TO255OCTETS_GB_LO);
+
+	stats->rx256to511octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_RX256TO511OCTETS_GB_LO);
+
+	stats->rx512to1023octets_gb +=
+		XGMAC_IOREAD(pdata, MMC_RX512TO1023OCTETS_GB_LO);
+
+	stats->rx1024tomaxoctets_gb +=
+		XGMAC_IOREAD(pdata, MMC_RX1024TOMAXOCTETS_GB_LO);
+
+	stats->rxunicastframes_g +=
+		XGMAC_IOREAD(pdata, MMC_RXUNICASTFRAMES_G_LO);
+
+	stats->rxlengtherror +=
+		XGMAC_IOREAD(pdata, MMC_RXLENGTHERROR_LO);
+
+	stats->rxoutofrangetype +=
+		XGMAC_IOREAD(pdata, MMC_RXOUTOFRANGETYPE_LO);
+
+	stats->rxpauseframes +=
+		XGMAC_IOREAD(pdata, MMC_RXPAUSEFRAMES_LO);
+
+	stats->rxfifooverflow +=
+		XGMAC_IOREAD(pdata, MMC_RXFIFOOVERFLOW_LO);
+
+	stats->rxvlanframes_gb +=
+		XGMAC_IOREAD(pdata, MMC_RXVLANFRAMES_GB_LO);
+
+	stats->rxwatchdogerror +=
+		XGMAC_IOREAD(pdata, MMC_RXWATCHDOGERROR);
+
+	/* Un-freeze counters */
+	XGMAC_IOWRITE_BITS(pdata, MMC_CR, MCF, 0);
+}
+
+static void xgbe_config_mmc(struct xgbe_prv_data *pdata)
+{
+	/* Set counters to reset on read */
+	XGMAC_IOWRITE_BITS(pdata, MMC_CR, ROR, 1);
+
+	/* Reset the counters */
+	XGMAC_IOWRITE_BITS(pdata, MMC_CR, CR, 1);
+}
+
+static void xgbe_enable_tx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Enable each Tx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 1);
+	}
+
+	/* Enable each Tx queue */
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TXQEN,
+				       MTL_Q_ENABLED);
+
+	/* Enable MAC Tx */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 1);
+}
+
+static void xgbe_disable_tx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Disable MAC Tx */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 0);
+
+	/* Disable each Tx queue */
+	for (i = 0; i < pdata->hw_feat.tx_q_cnt; i++)
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TXQEN, 0);
+
+	/* Disable each Tx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 0);
+	}
+}
+
+static void xgbe_enable_rx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int reg_val, i;
+
+	/* Enable each Rx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 1);
+	}
+
+	/* Enable each Rx queue */
+	reg_val = 0;
+	for (i = 0; i < pdata->hw_feat.rx_q_cnt; i++)
+		reg_val |= (0x02 << (i << 1));
+	XGMAC_IOWRITE(pdata, MAC_RQC0R, reg_val);
+
+	/* Enable MAC Rx */
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, DCRCC, 1);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, CST, 1);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, ACS, 1);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, RE, 1);
+}
+
+static void xgbe_disable_rx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Disable MAC Rx */
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, DCRCC, 0);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, CST, 0);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, ACS, 0);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, RE, 0);
+
+	/* Disable each Rx queue */
+	XGMAC_IOWRITE(pdata, MAC_RQC0R, 0);
+
+	/* Disable each Rx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 0);
+	}
+}
+
+static void xgbe_powerup_tx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Enable each Tx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 1);
+	}
+
+	/* Enable MAC Tx */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 1);
+}
+
+static void xgbe_powerdown_tx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Disable MAC Tx */
+	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 0);
+
+	/* Disable each Tx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_TCR, ST, 0);
+	}
+}
+
+static void xgbe_powerup_rx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Enable each Rx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 1);
+	}
+}
+
+static void xgbe_powerdown_rx(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	/* Disable each Rx DMA channel */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_RCR, SR, 0);
+	}
+}
+
+static int xgbe_init(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	int ret;
+
+	DBGPR("-->xgbe_init\n");
+
+	/* Flush Tx queues */
+	ret = xgbe_flush_tx_queues(pdata);
+	if (ret)
+		return ret;
+
+	/*
+	 * Initialize DMA related features
+	 */
+	xgbe_config_dma_bus(pdata);
+	xgbe_config_dma_cache(pdata);
+	xgbe_config_osp_mode(pdata);
+	xgbe_config_pblx8(pdata);
+	xgbe_config_tx_pbl_val(pdata);
+	xgbe_config_rx_pbl_val(pdata);
+	xgbe_config_rx_coalesce(pdata);
+	xgbe_config_tx_coalesce(pdata);
+	xgbe_config_rx_buffer_size(pdata);
+	xgbe_config_tso_mode(pdata);
+	desc_if->wrapper_tx_desc_init(pdata);
+	desc_if->wrapper_rx_desc_init(pdata);
+	xgbe_enable_dma_interrupts(pdata);
+
+	/*
+	 * Initialize MTL related features
+	 */
+	xgbe_config_mtl_mode(pdata);
+	xgbe_config_rx_queue_mapping(pdata);
+	/*TODO: Program the priorities mapped to the Selected Traffic Classes
+		in MTL_TC_Prty_Map0-3 registers */
+	xgbe_config_tsf_mode(pdata, pdata->tx_sf_mode);
+	xgbe_config_rsf_mode(pdata, pdata->rx_sf_mode);
+	xgbe_config_tx_threshold(pdata, pdata->tx_threshold);
+	xgbe_config_rx_threshold(pdata, pdata->rx_threshold);
+	xgbe_config_tx_fifo_size(pdata);
+	xgbe_config_rx_fifo_size(pdata);
+	xgbe_config_flow_control_threshold(pdata);
+	/*TODO: Queue to Traffic Class Mapping (Q2TCMAP) */
+	/*TODO: Error Packet and undersized good Packet forwarding enable
+		(FEP and FUP)
+	 */
+	xgbe_enable_mtl_interrupts(pdata);
+
+	/* Transmit Class Weight */
+	XGMAC_IOWRITE_BITS(pdata, MTL_Q_TCQWR, QW, 0x10);
+
+	/*
+	 * Initialize MAC related features
+	 */
+	xgbe_config_mac_address(pdata);
+	xgbe_config_jumbo_enable(pdata);
+	xgbe_config_flow_control(pdata);
+	xgbe_config_checksum_offload(pdata);
+	xgbe_config_vlan_support(pdata);
+	xgbe_config_mmc(pdata);
+	xgbe_enable_mac_interrupts(pdata);
+
+	DBGPR("<--xgbe_init\n");
+
+	return 0;
+}
+
+void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
+{
+	DBGPR("-->xgbe_init_function_ptrs\n");
+
+	hw_if->tx_complete = xgbe_tx_complete;
+
+	hw_if->set_promiscuous_mode = xgbe_set_promiscuous_mode;
+	hw_if->set_all_multicast_mode = xgbe_set_all_multicast_mode;
+	hw_if->set_addn_mac_addrs = xgbe_set_addn_mac_addrs;
+	hw_if->set_mac_address = xgbe_set_mac_address;
+
+	hw_if->enable_rx_csum = xgbe_enable_rx_csum;
+	hw_if->disable_rx_csum = xgbe_disable_rx_csum;
+
+	hw_if->enable_rx_vlan_stripping = xgbe_enable_rx_vlan_stripping;
+	hw_if->disable_rx_vlan_stripping = xgbe_disable_rx_vlan_stripping;
+
+	hw_if->read_mmd_regs = xgbe_read_mmd_regs;
+	hw_if->write_mmd_regs = xgbe_write_mmd_regs;
+
+	hw_if->set_gmii_speed = xgbe_set_gmii_speed;
+	hw_if->set_gmii_2500_speed = xgbe_set_gmii_2500_speed;
+	hw_if->set_xgmii_speed = xgbe_set_xgmii_speed;
+
+	hw_if->enable_tx = xgbe_enable_tx;
+	hw_if->disable_tx = xgbe_disable_tx;
+	hw_if->enable_rx = xgbe_enable_rx;
+	hw_if->disable_rx = xgbe_disable_rx;
+
+	hw_if->powerup_tx = xgbe_powerup_tx;
+	hw_if->powerdown_tx = xgbe_powerdown_tx;
+	hw_if->powerup_rx = xgbe_powerup_rx;
+	hw_if->powerdown_rx = xgbe_powerdown_rx;
+
+	hw_if->pre_xmit = xgbe_pre_xmit;
+	hw_if->dev_read = xgbe_dev_read;
+	hw_if->enable_int = xgbe_enable_int;
+	hw_if->disable_int = xgbe_disable_int;
+	hw_if->init = xgbe_init;
+	hw_if->exit = xgbe_exit;
+
+	/* Descriptor related Sequences have to be initialized here */
+	hw_if->tx_desc_init = xgbe_tx_desc_init;
+	hw_if->rx_desc_init = xgbe_rx_desc_init;
+	hw_if->tx_desc_reset = xgbe_tx_desc_reset;
+	hw_if->rx_desc_reset = xgbe_rx_desc_reset;
+	hw_if->is_last_desc = xgbe_is_last_desc;
+	hw_if->is_context_desc = xgbe_is_context_desc;
+
+	/* For FLOW ctrl */
+	hw_if->config_tx_flow_control = xgbe_config_tx_flow_control;
+	hw_if->config_rx_flow_control = xgbe_config_rx_flow_control;
+
+	/* For RX coalescing */
+	hw_if->config_rx_coalesce = xgbe_config_rx_coalesce;
+	hw_if->config_tx_coalesce = xgbe_config_tx_coalesce;
+	hw_if->usec_to_riwt = xgbe_usec_to_riwt;
+	hw_if->riwt_to_usec = xgbe_riwt_to_usec;
+
+	/* For RX and TX threshold config */
+	hw_if->config_rx_threshold = xgbe_config_rx_threshold;
+	hw_if->config_tx_threshold = xgbe_config_tx_threshold;
+
+	/* For RX and TX Store and Forward Mode config */
+	hw_if->config_rsf_mode = xgbe_config_rsf_mode;
+	hw_if->config_tsf_mode = xgbe_config_tsf_mode;
+
+	/* For TX DMA Operating on Second Frame config */
+	hw_if->config_osp_mode = xgbe_config_osp_mode;
+
+	/* For RX and TX PBL config */
+	hw_if->config_rx_pbl_val = xgbe_config_rx_pbl_val;
+	hw_if->get_rx_pbl_val = xgbe_get_rx_pbl_val;
+	hw_if->config_tx_pbl_val = xgbe_config_tx_pbl_val;
+	hw_if->get_tx_pbl_val = xgbe_get_tx_pbl_val;
+	hw_if->config_pblx8 = xgbe_config_pblx8;
+
+	/* For MMC statistics support */
+	hw_if->tx_mmc_int = xgbe_tx_mmc_int;
+	hw_if->rx_mmc_int = xgbe_rx_mmc_int;
+	hw_if->read_mmc_stats = xgbe_read_mmc_stats;
+
+	DBGPR("<--xgbe_init_function_ptrs\n");
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
new file mode 100644
index 000000000000..cfe3d93b5f52
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -0,0 +1,1351 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/tcp.h>
+#include <linux/if_vlan.h>
+#include <linux/phy.h>
+#include <net/busy_poll.h>
+#include <linux/clk.h>
+#include <linux/if_ether.h>
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+static int xgbe_poll(struct napi_struct *, int);
+static void xgbe_set_rx_mode(struct net_device *);
+
+static inline unsigned int xgbe_tx_avail_desc(struct xgbe_ring *ring)
+{
+	return (ring->rdesc_count - (ring->cur - ring->dirty));
+}
+
+static int xgbe_calc_rx_buf_size(struct net_device *netdev, unsigned int mtu)
+{
+	unsigned int rx_buf_size;
+
+	if (mtu > XGMAC_JUMBO_PACKET_MTU) {
+		netdev_alert(netdev, "MTU exceeds maximum supported value\n");
+		return -EINVAL;
+	}
+
+	rx_buf_size = mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
+	if (rx_buf_size < RX_MIN_BUF_SIZE)
+		rx_buf_size = RX_MIN_BUF_SIZE;
+	rx_buf_size = (rx_buf_size + RX_BUF_ALIGN - 1) & ~(RX_BUF_ALIGN - 1);
+
+	return rx_buf_size;
+}
+
+static void xgbe_enable_rx_tx_ints(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (channel->tx_ring)
+			hw_if->enable_int(channel,
+					  XGMAC_INT_DMA_CH_SR_TI);
+		if (channel->rx_ring)
+			hw_if->enable_int(channel,
+					  XGMAC_INT_DMA_CH_SR_RI);
+	}
+}
+
+static void xgbe_disable_rx_tx_ints(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (channel->tx_ring)
+			hw_if->disable_int(channel,
+					   XGMAC_INT_DMA_CH_SR_TI);
+		if (channel->rx_ring)
+			hw_if->disable_int(channel,
+					   XGMAC_INT_DMA_CH_SR_RI);
+	}
+}
+
+static irqreturn_t xgbe_isr(int irq, void *data)
+{
+	struct xgbe_prv_data *pdata = data;
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_channel *channel;
+	unsigned int dma_isr, dma_ch_isr;
+	unsigned int mac_isr;
+	unsigned int i;
+
+	/* The DMA interrupt status register also reports MAC and MTL
+	 * interrupts. So for polling mode, we just need to check for
+	 * this register to be non-zero
+	 */
+	dma_isr = XGMAC_IOREAD(pdata, DMA_ISR);
+	if (!dma_isr)
+		goto isr_done;
+
+	DBGPR("-->xgbe_isr\n");
+
+	DBGPR("  DMA_ISR = %08x\n", dma_isr);
+	DBGPR("  DMA_DS0 = %08x\n", XGMAC_IOREAD(pdata, DMA_DSR0));
+	DBGPR("  DMA_DS1 = %08x\n", XGMAC_IOREAD(pdata, DMA_DSR1));
+
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!(dma_isr & (1 << i)))
+			continue;
+
+		channel = pdata->channel + i;
+
+		dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR);
+		DBGPR("  DMA_CH%u_ISR = %08x\n", i, dma_ch_isr);
+
+		if (XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, TI) ||
+		    XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, RI)) {
+			if (napi_schedule_prep(&pdata->napi)) {
+				/* Disable Tx and Rx interrupts */
+				xgbe_disable_rx_tx_ints(pdata);
+
+				/* Turn on polling */
+				__napi_schedule(&pdata->napi);
+			}
+		}
+
+		/* Restart the device on a Fatal Bus Error */
+		if (XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, FBE))
+			schedule_work(&pdata->restart_work);
+
+		/* Clear all interrupt signals */
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_SR, dma_ch_isr);
+	}
+
+	if (XGMAC_GET_BITS(dma_isr, DMA_ISR, MACIS)) {
+		mac_isr = XGMAC_IOREAD(pdata, MAC_ISR);
+
+		if (XGMAC_GET_BITS(mac_isr, MAC_ISR, MMCTXIS))
+			hw_if->tx_mmc_int(pdata);
+
+		if (XGMAC_GET_BITS(mac_isr, MAC_ISR, MMCRXIS))
+			hw_if->rx_mmc_int(pdata);
+	}
+
+	DBGPR("  DMA_ISR = %08x\n", XGMAC_IOREAD(pdata, DMA_ISR));
+
+	DBGPR("<--xgbe_isr\n");
+
+isr_done:
+	return IRQ_HANDLED;
+}
+
+static enum hrtimer_restart xgbe_tx_timer(struct hrtimer *timer)
+{
+	struct xgbe_channel *channel = container_of(timer,
+						    struct xgbe_channel,
+						    tx_timer);
+	struct xgbe_ring *ring = channel->tx_ring;
+	struct xgbe_prv_data *pdata = channel->pdata;
+	unsigned long flags;
+
+	DBGPR("-->xgbe_tx_timer\n");
+
+	spin_lock_irqsave(&ring->lock, flags);
+
+	if (napi_schedule_prep(&pdata->napi)) {
+		/* Disable Tx and Rx interrupts */
+		xgbe_disable_rx_tx_ints(pdata);
+
+		/* Turn on polling */
+		__napi_schedule(&pdata->napi);
+	}
+
+	channel->tx_timer_active = 0;
+
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	DBGPR("<--xgbe_tx_timer\n");
+
+	return HRTIMER_NORESTART;
+}
+
+static void xgbe_init_tx_timers(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	DBGPR("-->xgbe_init_tx_timers\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		DBGPR("  %s adding tx timer\n", channel->name);
+		hrtimer_init(&channel->tx_timer, CLOCK_MONOTONIC,
+			     HRTIMER_MODE_REL);
+		channel->tx_timer.function = xgbe_tx_timer;
+	}
+
+	DBGPR("<--xgbe_init_tx_timers\n");
+}
+
+static void xgbe_stop_tx_timers(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	DBGPR("-->xgbe_stop_tx_timers\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		DBGPR("  %s deleting tx timer\n", channel->name);
+		channel->tx_timer_active = 0;
+		hrtimer_cancel(&channel->tx_timer);
+	}
+
+	DBGPR("<--xgbe_stop_tx_timers\n");
+}
+
+void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata)
+{
+	unsigned int mac_hfr0, mac_hfr1, mac_hfr2;
+	struct xgbe_hw_features *hw_feat = &pdata->hw_feat;
+
+	DBGPR("-->xgbe_get_all_hw_features\n");
+
+	mac_hfr0 = XGMAC_IOREAD(pdata, MAC_HWF0R);
+	mac_hfr1 = XGMAC_IOREAD(pdata, MAC_HWF1R);
+	mac_hfr2 = XGMAC_IOREAD(pdata, MAC_HWF2R);
+
+	memset(hw_feat, 0, sizeof(*hw_feat));
+
+	/* Hardware feature register 0 */
+	hw_feat->gmii        = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, GMIISEL);
+	hw_feat->vlhash      = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, VLHASH);
+	hw_feat->sma         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, SMASEL);
+	hw_feat->rwk         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, RWKSEL);
+	hw_feat->mgk         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, MGKSEL);
+	hw_feat->mmc         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, MMCSEL);
+	hw_feat->aoe         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, ARPOFFSEL);
+	hw_feat->ts          = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, TSSEL);
+	hw_feat->eee         = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, EEESEL);
+	hw_feat->tx_coe      = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, TXCOESEL);
+	hw_feat->rx_coe      = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, RXCOESEL);
+	hw_feat->addn_mac    = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R,
+					      ADDMACADRSEL);
+	hw_feat->ts_src      = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, TSSTSSEL);
+	hw_feat->sa_vlan_ins = XGMAC_GET_BITS(mac_hfr0, MAC_HWF0R, SAVLANINS);
+
+	/* Hardware feature register 1 */
+	hw_feat->rx_fifo_size  = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R,
+						RXFIFOSIZE);
+	hw_feat->tx_fifo_size  = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R,
+						TXFIFOSIZE);
+	hw_feat->dcb           = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, DCBEN);
+	hw_feat->sph           = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, SPHEN);
+	hw_feat->tso           = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, TSOEN);
+	hw_feat->dma_debug     = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, DBGMEMA);
+	hw_feat->hash_table_size = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R,
+						  HASHTBLSZ);
+	hw_feat->l3l4_filter_num = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R,
+						  L3L4FNUM);
+
+	/* Hardware feature register 2 */
+	hw_feat->rx_q_cnt     = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, RXQCNT);
+	hw_feat->tx_q_cnt     = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, TXQCNT);
+	hw_feat->rx_ch_cnt    = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, RXCHCNT);
+	hw_feat->tx_ch_cnt    = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, TXCHCNT);
+	hw_feat->pps_out_num  = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, PPSOUTNUM);
+	hw_feat->aux_snap_num = XGMAC_GET_BITS(mac_hfr2, MAC_HWF2R, AUXSNAPNUM);
+
+	/* The Queue and Channel counts are zero based so increment them
+	 * to get the actual number
+	 */
+	hw_feat->rx_q_cnt++;
+	hw_feat->tx_q_cnt++;
+	hw_feat->rx_ch_cnt++;
+	hw_feat->tx_ch_cnt++;
+
+	DBGPR("<--xgbe_get_all_hw_features\n");
+}
+
+static void xgbe_napi_enable(struct xgbe_prv_data *pdata, unsigned int add)
+{
+	if (add)
+		netif_napi_add(pdata->netdev, &pdata->napi, xgbe_poll,
+			       NAPI_POLL_WEIGHT);
+	napi_enable(&pdata->napi);
+}
+
+static void xgbe_napi_disable(struct xgbe_prv_data *pdata)
+{
+	napi_disable(&pdata->napi);
+}
+
+void xgbe_init_tx_coalesce(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+
+	DBGPR("-->xgbe_init_tx_coalesce\n");
+
+	pdata->tx_usecs = XGMAC_INIT_DMA_TX_USECS;
+	pdata->tx_frames = XGMAC_INIT_DMA_TX_FRAMES;
+
+	hw_if->config_tx_coalesce(pdata);
+
+	DBGPR("<--xgbe_init_tx_coalesce\n");
+}
+
+void xgbe_init_rx_coalesce(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+
+	DBGPR("-->xgbe_init_rx_coalesce\n");
+
+	pdata->rx_riwt = hw_if->usec_to_riwt(pdata, XGMAC_INIT_DMA_RX_USECS);
+	pdata->rx_frames = XGMAC_INIT_DMA_RX_FRAMES;
+
+	hw_if->config_rx_coalesce(pdata);
+
+	DBGPR("<--xgbe_init_rx_coalesce\n");
+}
+
+static void xgbe_free_tx_skbuff(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_channel *channel;
+	struct xgbe_ring *ring;
+	struct xgbe_ring_data *rdata;
+	unsigned int i, j;
+
+	DBGPR("-->xgbe_free_tx_skbuff\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		ring = channel->tx_ring;
+		if (!ring)
+			break;
+
+		for (j = 0; j < ring->rdesc_count; j++) {
+			rdata = GET_DESC_DATA(ring, j);
+			desc_if->unmap_skb(pdata, rdata);
+		}
+	}
+
+	DBGPR("<--xgbe_free_tx_skbuff\n");
+}
+
+static void xgbe_free_rx_skbuff(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_channel *channel;
+	struct xgbe_ring *ring;
+	struct xgbe_ring_data *rdata;
+	unsigned int i, j;
+
+	DBGPR("-->xgbe_free_rx_skbuff\n");
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		ring = channel->rx_ring;
+		if (!ring)
+			break;
+
+		for (j = 0; j < ring->rdesc_count; j++) {
+			rdata = GET_DESC_DATA(ring, j);
+			desc_if->unmap_skb(pdata, rdata);
+		}
+	}
+
+	DBGPR("<--xgbe_free_rx_skbuff\n");
+}
+
+int xgbe_powerdown(struct net_device *netdev, unsigned int caller)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned long flags;
+
+	DBGPR("-->xgbe_powerdown\n");
+
+	if (!netif_running(netdev) ||
+	    (caller == XGMAC_IOCTL_CONTEXT && pdata->power_down)) {
+		netdev_alert(netdev, "Device is already powered down\n");
+		DBGPR("<--xgbe_powerdown\n");
+		return -EINVAL;
+	}
+
+	phy_stop(pdata->phydev);
+
+	spin_lock_irqsave(&pdata->lock, flags);
+
+	if (caller == XGMAC_DRIVER_CONTEXT)
+		netif_device_detach(netdev);
+
+	netif_tx_stop_all_queues(netdev);
+	xgbe_napi_disable(pdata);
+
+	/* Powerdown Tx/Rx */
+	hw_if->powerdown_tx(pdata);
+	hw_if->powerdown_rx(pdata);
+
+	pdata->power_down = 1;
+
+	spin_unlock_irqrestore(&pdata->lock, flags);
+
+	DBGPR("<--xgbe_powerdown\n");
+
+	return 0;
+}
+
+int xgbe_powerup(struct net_device *netdev, unsigned int caller)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned long flags;
+
+	DBGPR("-->xgbe_powerup\n");
+
+	if (!netif_running(netdev) ||
+	    (caller == XGMAC_IOCTL_CONTEXT && !pdata->power_down)) {
+		netdev_alert(netdev, "Device is already powered up\n");
+		DBGPR("<--xgbe_powerup\n");
+		return -EINVAL;
+	}
+
+	spin_lock_irqsave(&pdata->lock, flags);
+
+	pdata->power_down = 0;
+
+	phy_start(pdata->phydev);
+
+	/* Enable Tx/Rx */
+	hw_if->powerup_tx(pdata);
+	hw_if->powerup_rx(pdata);
+
+	if (caller == XGMAC_DRIVER_CONTEXT)
+		netif_device_attach(netdev);
+
+	xgbe_napi_enable(pdata, 0);
+	netif_tx_start_all_queues(netdev);
+
+	spin_unlock_irqrestore(&pdata->lock, flags);
+
+	DBGPR("<--xgbe_powerup\n");
+
+	return 0;
+}
+
+static int xgbe_start(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct net_device *netdev = pdata->netdev;
+
+	DBGPR("-->xgbe_start\n");
+
+	xgbe_set_rx_mode(netdev);
+
+	hw_if->init(pdata);
+
+	phy_start(pdata->phydev);
+
+	hw_if->enable_tx(pdata);
+	hw_if->enable_rx(pdata);
+
+	xgbe_init_tx_timers(pdata);
+
+	xgbe_napi_enable(pdata, 1);
+	netif_tx_start_all_queues(netdev);
+
+	DBGPR("<--xgbe_start\n");
+
+	return 0;
+}
+
+static void xgbe_stop(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct net_device *netdev = pdata->netdev;
+
+	DBGPR("-->xgbe_stop\n");
+
+	phy_stop(pdata->phydev);
+
+	netif_tx_stop_all_queues(netdev);
+	xgbe_napi_disable(pdata);
+
+	xgbe_stop_tx_timers(pdata);
+
+	hw_if->disable_tx(pdata);
+	hw_if->disable_rx(pdata);
+
+	DBGPR("<--xgbe_stop\n");
+}
+
+static void xgbe_restart_dev(struct xgbe_prv_data *pdata, unsigned int reset)
+{
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+
+	DBGPR("-->xgbe_restart_dev\n");
+
+	/* If not running, "restart" will happen on open */
+	if (!netif_running(pdata->netdev))
+		return;
+
+	xgbe_stop(pdata);
+	synchronize_irq(pdata->irq_number);
+
+	xgbe_free_tx_skbuff(pdata);
+	xgbe_free_rx_skbuff(pdata);
+
+	/* Issue software reset to device if requested */
+	if (reset)
+		hw_if->exit(pdata);
+
+	xgbe_start(pdata);
+
+	DBGPR("<--xgbe_restart_dev\n");
+}
+
+static void xgbe_restart(struct work_struct *work)
+{
+	struct xgbe_prv_data *pdata = container_of(work,
+						   struct xgbe_prv_data,
+						   restart_work);
+
+	rtnl_lock();
+
+	xgbe_restart_dev(pdata, 1);
+
+	rtnl_unlock();
+}
+
+static void xgbe_prep_vlan(struct sk_buff *skb, struct xgbe_packet_data *packet)
+{
+	if (vlan_tx_tag_present(skb))
+		packet->vlan_ctag = vlan_tx_tag_get(skb);
+}
+
+static int xgbe_prep_tso(struct sk_buff *skb, struct xgbe_packet_data *packet)
+{
+	int ret;
+
+	if (!XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			    TSO_ENABLE))
+		return 0;
+
+	ret = skb_cow_head(skb, 0);
+	if (ret)
+		return ret;
+
+	packet->header_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	packet->tcp_header_len = tcp_hdrlen(skb);
+	packet->tcp_payload_len = skb->len - packet->header_len;
+	packet->mss = skb_shinfo(skb)->gso_size;
+	DBGPR("  packet->header_len=%u\n", packet->header_len);
+	DBGPR("  packet->tcp_header_len=%u, packet->tcp_payload_len=%u\n",
+	      packet->tcp_header_len, packet->tcp_payload_len);
+	DBGPR("  packet->mss=%u\n", packet->mss);
+
+	return 0;
+}
+
+static int xgbe_is_tso(struct sk_buff *skb)
+{
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
+	if (!skb_is_gso(skb))
+		return 0;
+
+	DBGPR("  TSO packet to be processed\n");
+
+	return 1;
+}
+
+static void xgbe_packet_info(struct xgbe_ring *ring, struct sk_buff *skb,
+			     struct xgbe_packet_data *packet)
+{
+	struct skb_frag_struct *frag;
+	unsigned int context_desc;
+	unsigned int len;
+	unsigned int i;
+
+	context_desc = 0;
+	packet->rdesc_count = 0;
+
+	if (xgbe_is_tso(skb)) {
+		/* TSO requires an extra desriptor if mss is different */
+		if (skb_shinfo(skb)->gso_size != ring->tx.cur_mss) {
+			context_desc = 1;
+			packet->rdesc_count++;
+		}
+
+		/* TSO requires an extra desriptor for TSO header */
+		packet->rdesc_count++;
+
+		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			       TSO_ENABLE, 1);
+		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			       CSUM_ENABLE, 1);
+	} else if (skb->ip_summed == CHECKSUM_PARTIAL)
+		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			       CSUM_ENABLE, 1);
+
+	if (vlan_tx_tag_present(skb)) {
+		/* VLAN requires an extra descriptor if tag is different */
+		if (vlan_tx_tag_get(skb) != ring->tx.cur_vlan_ctag)
+			/* We can share with the TSO context descriptor */
+			if (!context_desc) {
+				context_desc = 1;
+				packet->rdesc_count++;
+			}
+
+		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
+			       VLAN_CTAG, 1);
+	}
+
+	for (len = skb_headlen(skb); len;) {
+		packet->rdesc_count++;
+		len -= min_t(unsigned int, len, TX_MAX_BUF_SIZE);
+	}
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		frag = &skb_shinfo(skb)->frags[i];
+		for (len = skb_frag_size(frag); len; ) {
+			packet->rdesc_count++;
+			len -= min_t(unsigned int, len, TX_MAX_BUF_SIZE);
+		}
+	}
+}
+
+static int xgbe_open(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	int ret;
+
+	DBGPR("-->xgbe_open\n");
+
+	/* Enable the clock */
+	ret = clk_prepare_enable(pdata->sysclock);
+	if (ret) {
+		netdev_alert(netdev, "clk_prepare_enable failed\n");
+		return ret;
+	}
+
+	/* Calculate the Rx buffer size before allocating rings */
+	ret = xgbe_calc_rx_buf_size(netdev, netdev->mtu);
+	if (ret < 0)
+		goto err_clk;
+	pdata->rx_buf_size = ret;
+
+	/* Allocate the ring descriptors and buffers */
+	ret = desc_if->alloc_ring_resources(pdata);
+	if (ret)
+		goto err_clk;
+
+	/* Initialize the device restart work struct */
+	INIT_WORK(&pdata->restart_work, xgbe_restart);
+
+	/* Request interrupts */
+	ret = devm_request_irq(pdata->dev, netdev->irq, xgbe_isr, 0,
+			       netdev->name, pdata);
+	if (ret) {
+		netdev_alert(netdev, "error requesting irq %d\n",
+			     pdata->irq_number);
+		goto err_irq;
+	}
+	pdata->irq_number = netdev->irq;
+
+	ret = xgbe_start(pdata);
+	if (ret)
+		goto err_start;
+
+	DBGPR("<--xgbe_open\n");
+
+	return 0;
+
+err_start:
+	hw_if->exit(pdata);
+
+	devm_free_irq(pdata->dev, pdata->irq_number, pdata);
+	pdata->irq_number = 0;
+
+err_irq:
+	desc_if->free_ring_resources(pdata);
+
+err_clk:
+	clk_disable_unprepare(pdata->sysclock);
+
+	return ret;
+}
+
+static int xgbe_close(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+
+	DBGPR("-->xgbe_close\n");
+
+	/* Stop the device */
+	xgbe_stop(pdata);
+
+	/* Issue software reset to device */
+	hw_if->exit(pdata);
+
+	/* Free all the ring data */
+	desc_if->free_ring_resources(pdata);
+
+	/* Release the interrupt */
+	if (pdata->irq_number != 0) {
+		devm_free_irq(pdata->dev, pdata->irq_number, pdata);
+		pdata->irq_number = 0;
+	}
+
+	/* Disable the clock */
+	clk_disable_unprepare(pdata->sysclock);
+
+	DBGPR("<--xgbe_close\n");
+
+	return 0;
+}
+
+static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_channel *channel;
+	struct xgbe_ring *ring;
+	struct xgbe_packet_data *packet;
+	unsigned long flags;
+	int ret;
+
+	DBGPR("-->xgbe_xmit: skb->len = %d\n", skb->len);
+
+	channel = pdata->channel + skb->queue_mapping;
+	ring = channel->tx_ring;
+	packet = &ring->packet_data;
+
+	ret = NETDEV_TX_OK;
+
+	spin_lock_irqsave(&ring->lock, flags);
+
+	if (skb->len == 0) {
+		netdev_err(netdev, "empty skb received from stack\n");
+		dev_kfree_skb_any(skb);
+		goto tx_netdev_return;
+	}
+
+	/* Calculate preliminary packet info */
+	memset(packet, 0, sizeof(*packet));
+	xgbe_packet_info(ring, skb, packet);
+
+	/* Check that there are enough descriptors available */
+	if (packet->rdesc_count > xgbe_tx_avail_desc(ring)) {
+		DBGPR("  Tx queue stopped, not enough descriptors available\n");
+		netif_stop_subqueue(netdev, channel->queue_index);
+		ring->tx.queue_stopped = 1;
+		ret = NETDEV_TX_BUSY;
+		goto tx_netdev_return;
+	}
+
+	ret = xgbe_prep_tso(skb, packet);
+	if (ret) {
+		netdev_err(netdev, "error processing TSO packet\n");
+		dev_kfree_skb_any(skb);
+		goto tx_netdev_return;
+	}
+	xgbe_prep_vlan(skb, packet);
+
+	if (!desc_if->map_tx_skb(channel, skb)) {
+		dev_kfree_skb_any(skb);
+		goto tx_netdev_return;
+	}
+
+	/* Configure required descriptor fields for transmission */
+	hw_if->pre_xmit(channel);
+
+#ifdef XGMAC_ENABLE_TX_PKT_DUMP
+	xgbe_print_pkt(netdev, skb, true);
+#endif
+
+tx_netdev_return:
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	DBGPR("<--xgbe_xmit\n");
+
+	return ret;
+}
+
+static void xgbe_set_rx_mode(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned int pr_mode, am_mode;
+
+	DBGPR("-->xgbe_set_rx_mode\n");
+
+	pr_mode = ((netdev->flags & IFF_PROMISC) != 0);
+	am_mode = ((netdev->flags & IFF_ALLMULTI) != 0);
+
+	if (netdev_uc_count(netdev) > pdata->hw_feat.addn_mac)
+		pr_mode = 1;
+	if (netdev_mc_count(netdev) > pdata->hw_feat.addn_mac)
+		am_mode = 1;
+	if ((netdev_uc_count(netdev) + netdev_mc_count(netdev)) >
+	     pdata->hw_feat.addn_mac)
+		pr_mode = 1;
+
+	hw_if->set_promiscuous_mode(pdata, pr_mode);
+	hw_if->set_all_multicast_mode(pdata, am_mode);
+	if (!pr_mode)
+		hw_if->set_addn_mac_addrs(pdata, am_mode);
+
+	DBGPR("<--xgbe_set_rx_mode\n");
+}
+
+static int xgbe_set_mac_address(struct net_device *netdev, void *addr)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct sockaddr *saddr = addr;
+
+	DBGPR("-->xgbe_set_mac_address\n");
+
+	if (!is_valid_ether_addr(saddr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	memcpy(netdev->dev_addr, saddr->sa_data, netdev->addr_len);
+
+	hw_if->set_mac_address(pdata, netdev->dev_addr);
+
+	DBGPR("<--xgbe_set_mac_address\n");
+
+	return 0;
+}
+
+static int xgbe_change_mtu(struct net_device *netdev, int mtu)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	int ret;
+
+	DBGPR("-->xgbe_change_mtu\n");
+
+	ret = xgbe_calc_rx_buf_size(netdev, mtu);
+	if (ret < 0)
+		return ret;
+
+	pdata->rx_buf_size = ret;
+	netdev->mtu = mtu;
+
+	xgbe_restart_dev(pdata, 0);
+
+	DBGPR("<--xgbe_change_mtu\n");
+
+	return 0;
+}
+
+static struct rtnl_link_stats64 *xgbe_get_stats64(struct net_device *netdev,
+						  struct rtnl_link_stats64 *s)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_mmc_stats *pstats = &pdata->mmc_stats;
+
+	DBGPR("-->%s\n", __func__);
+
+	pdata->hw_if.read_mmc_stats(pdata);
+
+	s->rx_packets = pstats->rxframecount_gb;
+	s->rx_bytes = pstats->rxoctetcount_gb;
+	s->rx_errors = pstats->rxframecount_gb -
+		       pstats->rxbroadcastframes_g -
+		       pstats->rxmulticastframes_g -
+		       pstats->rxunicastframes_g;
+	s->multicast = pstats->rxmulticastframes_g;
+	s->rx_length_errors = pstats->rxlengtherror;
+	s->rx_crc_errors = pstats->rxcrcerror;
+	s->rx_fifo_errors = pstats->rxfifooverflow;
+
+	s->tx_packets = pstats->txframecount_gb;
+	s->tx_bytes = pstats->txoctetcount_gb;
+	s->tx_errors = pstats->txframecount_gb - pstats->txframecount_g;
+	s->tx_dropped = netdev->stats.tx_dropped;
+
+	DBGPR("<--%s\n", __func__);
+
+	return s;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void xgbe_poll_controller(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	DBGPR("-->xgbe_poll_controller\n");
+
+	disable_irq(pdata->irq_number);
+
+	xgbe_isr(pdata->irq_number, pdata);
+
+	enable_irq(pdata->irq_number);
+
+	DBGPR("<--xgbe_poll_controller\n");
+}
+#endif /* End CONFIG_NET_POLL_CONTROLLER */
+
+static int xgbe_set_features(struct net_device *netdev,
+			     netdev_features_t features)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned int rxcsum_enabled, rxvlan_enabled;
+
+	rxcsum_enabled = !!(pdata->netdev_features & NETIF_F_RXCSUM);
+	rxvlan_enabled = !!(pdata->netdev_features & NETIF_F_HW_VLAN_CTAG_RX);
+
+	if ((features & NETIF_F_RXCSUM) && !rxcsum_enabled) {
+		hw_if->enable_rx_csum(pdata);
+		netdev_alert(netdev, "state change - rxcsum enabled\n");
+	} else if (!(features & NETIF_F_RXCSUM) && rxcsum_enabled) {
+		hw_if->disable_rx_csum(pdata);
+		netdev_alert(netdev, "state change - rxcsum disabled\n");
+	}
+
+	if ((features & NETIF_F_HW_VLAN_CTAG_RX) && !rxvlan_enabled) {
+		hw_if->enable_rx_vlan_stripping(pdata);
+		netdev_alert(netdev, "state change - rxvlan enabled\n");
+	} else if (!(features & NETIF_F_HW_VLAN_CTAG_RX) && rxvlan_enabled) {
+		hw_if->disable_rx_vlan_stripping(pdata);
+		netdev_alert(netdev, "state change - rxvlan disabled\n");
+	}
+
+	pdata->netdev_features = features;
+
+	DBGPR("<--xgbe_set_features\n");
+
+	return 0;
+}
+
+static const struct net_device_ops xgbe_netdev_ops = {
+	.ndo_open		= xgbe_open,
+	.ndo_stop		= xgbe_close,
+	.ndo_start_xmit		= xgbe_xmit,
+	.ndo_set_rx_mode	= xgbe_set_rx_mode,
+	.ndo_set_mac_address	= xgbe_set_mac_address,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_change_mtu		= xgbe_change_mtu,
+	.ndo_get_stats64	= xgbe_get_stats64,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= xgbe_poll_controller,
+#endif
+	.ndo_set_features	= xgbe_set_features,
+};
+
+struct net_device_ops *xgbe_get_netdev_ops(void)
+{
+	return (struct net_device_ops *)&xgbe_netdev_ops;
+}
+
+static int xgbe_tx_poll(struct xgbe_channel *channel)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_ring *ring = channel->tx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+	struct net_device *netdev = pdata->netdev;
+	unsigned long flags;
+	int processed = 0;
+
+	DBGPR("-->xgbe_tx_poll\n");
+
+	/* Nothing to do if there isn't a Tx ring for this channel */
+	if (!ring)
+		return 0;
+
+	spin_lock_irqsave(&ring->lock, flags);
+
+	while ((processed < TX_DESC_MAX_PROC) && (ring->dirty < ring->cur)) {
+		rdata = GET_DESC_DATA(ring, ring->dirty);
+		rdesc = rdata->rdesc;
+
+		if (!hw_if->tx_complete(rdesc))
+			break;
+
+#ifdef XGMAC_ENABLE_TX_DESC_DUMP
+		xgbe_dump_tx_desc(ring, ring->dirty, 1, 0);
+#endif
+
+		/* Free the SKB and reset the descriptor for re-use */
+		desc_if->unmap_skb(pdata, rdata);
+		hw_if->tx_desc_reset(rdata);
+
+		processed++;
+		ring->dirty++;
+	}
+
+	if ((ring->tx.queue_stopped == 1) &&
+	    (xgbe_tx_avail_desc(ring) > TX_DESC_MIN_FREE)) {
+		ring->tx.queue_stopped = 0;
+		netif_wake_subqueue(netdev, channel->queue_index);
+	}
+
+	DBGPR("<--xgbe_tx_poll: processed=%d\n", processed);
+
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	return processed;
+}
+
+static int xgbe_rx_poll(struct xgbe_channel *channel, int budget)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_ring *ring = channel->rx_ring;
+	struct xgbe_ring_data *rdata;
+	struct xgbe_packet_data *packet;
+	struct net_device *netdev = pdata->netdev;
+	struct sk_buff *skb;
+	unsigned int incomplete, error;
+	unsigned int cur_len, put_len, max_len;
+	int received = 0;
+
+	DBGPR("-->xgbe_rx_poll: budget=%d\n", budget);
+
+	/* Nothing to do if there isn't a Rx ring for this channel */
+	if (!ring)
+		return 0;
+
+	packet = &ring->packet_data;
+	while (received < budget) {
+		DBGPR("  cur = %d\n", ring->cur);
+
+		/* Clear the packet data information */
+		memset(packet, 0, sizeof(*packet));
+		skb = NULL;
+		error = 0;
+		cur_len = 0;
+
+read_again:
+		rdata = GET_DESC_DATA(ring, ring->cur);
+
+		if (hw_if->dev_read(channel))
+			break;
+
+		received++;
+		ring->cur++;
+		ring->dirty++;
+
+		dma_unmap_single(pdata->dev, rdata->skb_dma,
+				 rdata->skb_dma_len, DMA_FROM_DEVICE);
+		rdata->skb_dma = 0;
+
+		incomplete = XGMAC_GET_BITS(packet->attributes,
+					    RX_PACKET_ATTRIBUTES,
+					    INCOMPLETE);
+
+		/* Earlier error, just drain the remaining data */
+		if (incomplete && error)
+			goto read_again;
+
+		if (error || packet->errors) {
+			if (packet->errors)
+				DBGPR("Error in received packet\n");
+			dev_kfree_skb(skb);
+			continue;
+		}
+
+		put_len = rdata->len - cur_len;
+		if (skb) {
+			if (pskb_expand_head(skb, 0, put_len, GFP_ATOMIC)) {
+				DBGPR("pskb_expand_head error\n");
+				if (incomplete) {
+					error = 1;
+					goto read_again;
+				}
+
+				dev_kfree_skb(skb);
+				continue;
+			}
+			memcpy(skb_tail_pointer(skb), rdata->skb->data,
+			       put_len);
+		} else {
+			skb = rdata->skb;
+			rdata->skb = NULL;
+		}
+		skb_put(skb, put_len);
+		cur_len += put_len;
+
+		if (incomplete)
+			goto read_again;
+
+		/* Be sure we don't exceed the configured MTU */
+		max_len = netdev->mtu + ETH_HLEN;
+		if (!(netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+		    (skb->protocol == htons(ETH_P_8021Q)))
+			max_len += VLAN_HLEN;
+
+		if (skb->len > max_len) {
+			DBGPR("packet length exceeds configured MTU\n");
+			dev_kfree_skb(skb);
+			continue;
+		}
+
+#ifdef XGMAC_ENABLE_RX_PKT_DUMP
+		xgbe_print_pkt(netdev, skb, false);
+#endif
+
+		skb_checksum_none_assert(skb);
+		if (XGMAC_GET_BITS(packet->attributes,
+				   RX_PACKET_ATTRIBUTES, CSUM_DONE))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		if (XGMAC_GET_BITS(packet->attributes,
+				   RX_PACKET_ATTRIBUTES, VLAN_CTAG))
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+					       packet->vlan_ctag);
+
+		skb->dev = netdev;
+		skb->protocol = eth_type_trans(skb, netdev);
+		skb_record_rx_queue(skb, channel->queue_index);
+		skb_mark_napi_id(skb, &pdata->napi);
+
+		netdev->last_rx = jiffies;
+		napi_gro_receive(&pdata->napi, skb);
+	}
+
+	if (received) {
+		desc_if->realloc_skb(channel);
+
+		/* Update the Rx Tail Pointer Register with address of
+		 * the last cleaned entry */
+		rdata = GET_DESC_DATA(ring, ring->rx.realloc_index - 1);
+		XGMAC_DMA_IOWRITE(channel, DMA_CH_RDTR_LO,
+				  lower_32_bits(rdata->rdesc_dma));
+	}
+
+	DBGPR("<--xgbe_rx_poll: received = %d\n", received);
+
+	return received;
+}
+
+static int xgbe_poll(struct napi_struct *napi, int budget)
+{
+	struct xgbe_prv_data *pdata = container_of(napi, struct xgbe_prv_data,
+						   napi);
+	struct xgbe_channel *channel;
+	int processed;
+	unsigned int i;
+
+	DBGPR("-->xgbe_poll: budget=%d\n", budget);
+
+	/* Cleanup Tx ring first */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++)
+		xgbe_tx_poll(channel);
+
+	/* Process Rx ring next */
+	processed = 0;
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++)
+		processed += xgbe_rx_poll(channel, budget - processed);
+
+	/* If we processed everything, we are done */
+	if (processed < budget) {
+		/* Turn off polling */
+		napi_complete(napi);
+
+		/* Enable Tx and Rx interrupts */
+		xgbe_enable_rx_tx_ints(pdata);
+	}
+
+	DBGPR("<--xgbe_poll: received = %d\n", processed);
+
+	return processed;
+}
+
+void xgbe_dump_tx_desc(struct xgbe_ring *ring, unsigned int idx,
+		       unsigned int count, unsigned int flag)
+{
+	struct xgbe_ring_data *rdata;
+	struct xgbe_ring_desc *rdesc;
+
+	while (count--) {
+		rdata = GET_DESC_DATA(ring, idx);
+		rdesc = rdata->rdesc;
+		DBGPR("TX_NORMAL_DESC[%d %s] = %08x:%08x:%08x:%08x\n", idx,
+		      (flag == 1) ? "QUEUED FOR TX" : "TX BY DEVICE",
+		      le32_to_cpu(rdesc->desc0), le32_to_cpu(rdesc->desc1),
+		      le32_to_cpu(rdesc->desc2), le32_to_cpu(rdesc->desc3));
+		idx++;
+	}
+}
+
+void xgbe_dump_rx_desc(struct xgbe_ring *ring, struct xgbe_ring_desc *desc,
+		       unsigned int idx)
+{
+	DBGPR("RX_NORMAL_DESC[%d RX BY DEVICE] = %08x:%08x:%08x:%08x\n", idx,
+	      le32_to_cpu(desc->desc0), le32_to_cpu(desc->desc1),
+	      le32_to_cpu(desc->desc2), le32_to_cpu(desc->desc3));
+}
+
+void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx)
+{
+	struct ethhdr *eth = (struct ethhdr *)skb->data;
+	unsigned char *buf = skb->data;
+	unsigned char buffer[128];
+	unsigned int i, j;
+
+	netdev_alert(netdev, "\n************** SKB dump ****************\n");
+
+	netdev_alert(netdev, "%s packet of %d bytes\n",
+		     (tx_rx ? "TX" : "RX"), skb->len);
+
+	netdev_alert(netdev, "Dst MAC addr: %pM\n", eth->h_dest);
+	netdev_alert(netdev, "Src MAC addr: %pM\n", eth->h_source);
+	netdev_alert(netdev, "Protocol: 0x%04hx\n", ntohs(eth->h_proto));
+
+	for (i = 0, j = 0; i < skb->len;) {
+		j += snprintf(buffer + j, sizeof(buffer) - j, "%02hhx",
+			      buf[i++]);
+
+		if ((i % 32) == 0) {
+			netdev_alert(netdev, "  0x%04x: %s\n", i - 32, buffer);
+			j = 0;
+		} else if ((i % 16) == 0) {
+			buffer[j++] = ' ';
+			buffer[j++] = ' ';
+		} else if ((i % 4) == 0) {
+			buffer[j++] = ' ';
+		}
+	}
+	if (i % 32)
+		netdev_alert(netdev, "  0x%04x: %s\n", i - (i % 32), buffer);
+
+	netdev_alert(netdev, "\n************** SKB dump ****************\n");
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
new file mode 100644
index 000000000000..8909f2b51af1
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
@@ -0,0 +1,510 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/phy.h>
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+struct xgbe_stats {
+	char stat_string[ETH_GSTRING_LEN];
+	int stat_size;
+	int stat_offset;
+};
+
+#define XGMAC_MMC_STAT(_string, _var)				\
+	{ _string,						\
+	  FIELD_SIZEOF(struct xgbe_mmc_stats, _var),		\
+	  offsetof(struct xgbe_prv_data, mmc_stats._var),	\
+	}
+
+static const struct xgbe_stats xgbe_gstring_stats[] = {
+	XGMAC_MMC_STAT("tx_bytes", txoctetcount_gb),
+	XGMAC_MMC_STAT("tx_packets", txframecount_gb),
+	XGMAC_MMC_STAT("tx_unicast_packets", txunicastframes_gb),
+	XGMAC_MMC_STAT("tx_broadcast_packets", txbroadcastframes_gb),
+	XGMAC_MMC_STAT("tx_multicast_packets", txmulticastframes_gb),
+	XGMAC_MMC_STAT("tx_vlan_packets", txvlanframes_g),
+	XGMAC_MMC_STAT("tx_64_byte_packets", tx64octets_gb),
+	XGMAC_MMC_STAT("tx_65_to_127_byte_packets", tx65to127octets_gb),
+	XGMAC_MMC_STAT("tx_128_to_255_byte_packets", tx128to255octets_gb),
+	XGMAC_MMC_STAT("tx_256_to_511_byte_packets", tx256to511octets_gb),
+	XGMAC_MMC_STAT("tx_512_to_1023_byte_packets", tx512to1023octets_gb),
+	XGMAC_MMC_STAT("tx_1024_to_max_byte_packets", tx1024tomaxoctets_gb),
+	XGMAC_MMC_STAT("tx_underflow_errors", txunderflowerror),
+	XGMAC_MMC_STAT("tx_pause_frames", txpauseframes),
+
+	XGMAC_MMC_STAT("rx_bytes", rxoctetcount_gb),
+	XGMAC_MMC_STAT("rx_packets", rxframecount_gb),
+	XGMAC_MMC_STAT("rx_unicast_packets", rxunicastframes_g),
+	XGMAC_MMC_STAT("rx_broadcast_packets", rxbroadcastframes_g),
+	XGMAC_MMC_STAT("rx_multicast_packets", rxmulticastframes_g),
+	XGMAC_MMC_STAT("rx_vlan_packets", rxvlanframes_gb),
+	XGMAC_MMC_STAT("rx_64_byte_packets", rx64octets_gb),
+	XGMAC_MMC_STAT("rx_65_to_127_byte_packets", rx65to127octets_gb),
+	XGMAC_MMC_STAT("rx_128_to_255_byte_packets", rx128to255octets_gb),
+	XGMAC_MMC_STAT("rx_256_to_511_byte_packets", rx256to511octets_gb),
+	XGMAC_MMC_STAT("rx_512_to_1023_byte_packets", rx512to1023octets_gb),
+	XGMAC_MMC_STAT("rx_1024_to_max_byte_packets", rx1024tomaxoctets_gb),
+	XGMAC_MMC_STAT("rx_undersize_packets", rxundersize_g),
+	XGMAC_MMC_STAT("rx_oversize_packets", rxoversize_g),
+	XGMAC_MMC_STAT("rx_crc_errors", rxcrcerror),
+	XGMAC_MMC_STAT("rx_crc_errors_small_packets", rxrunterror),
+	XGMAC_MMC_STAT("rx_crc_errors_giant_packets", rxjabbererror),
+	XGMAC_MMC_STAT("rx_length_errors", rxlengtherror),
+	XGMAC_MMC_STAT("rx_out_of_range_errors", rxoutofrangetype),
+	XGMAC_MMC_STAT("rx_fifo_overflow_errors", rxfifooverflow),
+	XGMAC_MMC_STAT("rx_watchdog_errors", rxwatchdogerror),
+	XGMAC_MMC_STAT("rx_pause_frames", rxpauseframes),
+};
+#define XGBE_STATS_COUNT	ARRAY_SIZE(xgbe_gstring_stats)
+
+static void xgbe_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+{
+	int i;
+
+	DBGPR("-->%s\n", __func__);
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < XGBE_STATS_COUNT; i++) {
+			memcpy(data, xgbe_gstring_stats[i].stat_string,
+			       ETH_GSTRING_LEN);
+			data += ETH_GSTRING_LEN;
+		}
+		break;
+	}
+
+	DBGPR("<--%s\n", __func__);
+}
+
+static void xgbe_get_ethtool_stats(struct net_device *netdev,
+				   struct ethtool_stats *stats, u64 *data)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	u8 *stat;
+	int i;
+
+	DBGPR("-->%s\n", __func__);
+
+	pdata->hw_if.read_mmc_stats(pdata);
+	for (i = 0; i < XGBE_STATS_COUNT; i++) {
+		stat = (u8 *)pdata + xgbe_gstring_stats[i].stat_offset;
+		*data++ = *(u64 *)stat;
+	}
+
+	DBGPR("<--%s\n", __func__);
+}
+
+static int xgbe_get_sset_count(struct net_device *netdev, int stringset)
+{
+	int ret;
+
+	DBGPR("-->%s\n", __func__);
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		ret = XGBE_STATS_COUNT;
+		break;
+
+	default:
+		ret = -EOPNOTSUPP;
+	}
+
+	DBGPR("<--%s\n", __func__);
+
+	return ret;
+}
+
+static void xgbe_get_pauseparam(struct net_device *netdev,
+				struct ethtool_pauseparam *pause)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	DBGPR("-->xgbe_get_pauseparam\n");
+
+	pause->autoneg = pdata->pause_autoneg;
+	pause->tx_pause = pdata->tx_pause;
+	pause->rx_pause = pdata->rx_pause;
+
+	DBGPR("<--xgbe_get_pauseparam\n");
+}
+
+static int xgbe_set_pauseparam(struct net_device *netdev,
+			       struct ethtool_pauseparam *pause)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct phy_device *phydev = pdata->phydev;
+	int ret = 0;
+
+	DBGPR("-->xgbe_set_pauseparam\n");
+
+	DBGPR("  autoneg = %d, tx_pause = %d, rx_pause = %d\n",
+	      pause->autoneg, pause->tx_pause, pause->rx_pause);
+
+	pdata->pause_autoneg = pause->autoneg;
+	if (pause->autoneg) {
+		phydev->advertising |= ADVERTISED_Pause;
+		phydev->advertising |= ADVERTISED_Asym_Pause;
+
+	} else {
+		phydev->advertising &= ~ADVERTISED_Pause;
+		phydev->advertising &= ~ADVERTISED_Asym_Pause;
+
+		pdata->tx_pause = pause->tx_pause;
+		pdata->rx_pause = pause->rx_pause;
+	}
+
+	if (netif_running(netdev))
+		ret = phy_start_aneg(phydev);
+
+	DBGPR("<--xgbe_set_pauseparam\n");
+
+	return ret;
+}
+
+static int xgbe_get_settings(struct net_device *netdev,
+			     struct ethtool_cmd *cmd)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	int ret;
+
+	DBGPR("-->xgbe_get_settings\n");
+
+	if (!pdata->phydev)
+		return -ENODEV;
+
+	spin_lock_irq(&pdata->lock);
+
+	ret = phy_ethtool_gset(pdata->phydev, cmd);
+	cmd->transceiver = XCVR_EXTERNAL;
+
+	spin_unlock_irq(&pdata->lock);
+
+	DBGPR("<--xgbe_get_settings\n");
+
+	return ret;
+}
+
+static int xgbe_set_settings(struct net_device *netdev,
+			     struct ethtool_cmd *cmd)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct phy_device *phydev = pdata->phydev;
+	u32 speed;
+	int ret;
+
+	DBGPR("-->xgbe_set_settings\n");
+
+	if (!pdata->phydev)
+		return -ENODEV;
+
+	spin_lock_irq(&pdata->lock);
+
+	speed = ethtool_cmd_speed(cmd);
+
+	ret = -EINVAL;
+	if (cmd->phy_address != phydev->addr)
+		goto unlock;
+
+	if ((cmd->autoneg != AUTONEG_ENABLE) &&
+	    (cmd->autoneg != AUTONEG_DISABLE))
+		goto unlock;
+
+	if ((cmd->autoneg == AUTONEG_DISABLE) &&
+	    (((speed != SPEED_10000) && (speed != SPEED_1000)) ||
+	     (cmd->duplex != DUPLEX_FULL)))
+		goto unlock;
+
+	if (cmd->autoneg == AUTONEG_ENABLE) {
+		/* Clear settings needed to force speeds */
+		phydev->supported &= ~SUPPORTED_1000baseT_Full;
+		phydev->supported &= ~SUPPORTED_10000baseT_Full;
+	} else {
+		/* Add settings needed to force speed */
+		phydev->supported |= SUPPORTED_1000baseT_Full;
+		phydev->supported |= SUPPORTED_10000baseT_Full;
+	}
+
+	cmd->advertising &= phydev->supported;
+	if ((cmd->autoneg == AUTONEG_ENABLE) && !cmd->advertising)
+		goto unlock;
+
+	ret = 0;
+	phydev->autoneg = cmd->autoneg;
+	phydev->speed = speed;
+	phydev->duplex = cmd->duplex;
+	phydev->advertising = cmd->advertising;
+
+	if (cmd->autoneg == AUTONEG_ENABLE)
+		phydev->advertising |= ADVERTISED_Autoneg;
+	else
+		phydev->advertising &= ~ADVERTISED_Autoneg;
+
+	if (netif_running(netdev))
+		ret = phy_start_aneg(phydev);
+
+unlock:
+	spin_unlock_irq(&pdata->lock);
+
+	DBGPR("<--xgbe_set_settings\n");
+
+	return ret;
+}
+
+static void xgbe_get_drvinfo(struct net_device *netdev,
+			     struct ethtool_drvinfo *drvinfo)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	strlcpy(drvinfo->driver, XGBE_DRV_NAME, sizeof(drvinfo->driver));
+	strlcpy(drvinfo->version, XGBE_DRV_VERSION, sizeof(drvinfo->version));
+	strlcpy(drvinfo->bus_info, dev_name(pdata->dev),
+		sizeof(drvinfo->bus_info));
+	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%d.%d.%d",
+		 XGMAC_IOREAD_BITS(pdata, MAC_VR, USERVER),
+		 XGMAC_IOREAD_BITS(pdata, MAC_VR, DEVID),
+		 XGMAC_IOREAD_BITS(pdata, MAC_VR, SNPSVER));
+	drvinfo->n_stats = XGBE_STATS_COUNT;
+}
+
+static int xgbe_get_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *ec)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned int riwt;
+
+	DBGPR("-->xgbe_get_coalesce\n");
+
+	memset(ec, 0, sizeof(struct ethtool_coalesce));
+
+	riwt = pdata->rx_riwt;
+	ec->rx_coalesce_usecs = hw_if->riwt_to_usec(pdata, riwt);
+	ec->rx_max_coalesced_frames = pdata->rx_frames;
+
+	ec->tx_coalesce_usecs = pdata->tx_usecs;
+	ec->tx_max_coalesced_frames = pdata->tx_frames;
+
+	DBGPR("<--xgbe_get_coalesce\n");
+
+	return 0;
+}
+
+static int xgbe_set_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *ec)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned int rx_frames, rx_riwt, rx_usecs;
+	unsigned int tx_frames, tx_usecs;
+
+	DBGPR("-->xgbe_set_coalesce\n");
+
+	/* Check for not supported parameters  */
+	if ((ec->rx_coalesce_usecs_irq) ||
+	    (ec->rx_max_coalesced_frames_irq) ||
+	    (ec->tx_coalesce_usecs_irq) ||
+	    (ec->tx_max_coalesced_frames_irq) ||
+	    (ec->stats_block_coalesce_usecs) ||
+	    (ec->use_adaptive_rx_coalesce) ||
+	    (ec->use_adaptive_tx_coalesce) ||
+	    (ec->pkt_rate_low) ||
+	    (ec->rx_coalesce_usecs_low) ||
+	    (ec->rx_max_coalesced_frames_low) ||
+	    (ec->tx_coalesce_usecs_low) ||
+	    (ec->tx_max_coalesced_frames_low) ||
+	    (ec->pkt_rate_high) ||
+	    (ec->rx_coalesce_usecs_high) ||
+	    (ec->rx_max_coalesced_frames_high) ||
+	    (ec->tx_coalesce_usecs_high) ||
+	    (ec->tx_max_coalesced_frames_high) ||
+	    (ec->rate_sample_interval))
+		return -EOPNOTSUPP;
+
+	/* Can only change rx-frames when interface is down (see
+	 * rx_descriptor_init in xgbe-dev.c)
+	 */
+	rx_frames = pdata->rx_frames;
+	if (rx_frames != ec->rx_max_coalesced_frames && netif_running(netdev)) {
+		netdev_alert(netdev,
+			     "interface must be down to change rx-frames\n");
+		return -EINVAL;
+	}
+
+	rx_riwt = hw_if->usec_to_riwt(pdata, ec->rx_coalesce_usecs);
+	rx_frames = ec->rx_max_coalesced_frames;
+
+	/* Use smallest possible value if conversion resulted in zero */
+	if (ec->rx_coalesce_usecs && !rx_riwt)
+		rx_riwt = 1;
+
+	/* Check the bounds of values for Rx */
+	if (rx_riwt > XGMAC_MAX_DMA_RIWT) {
+		rx_usecs = hw_if->riwt_to_usec(pdata, XGMAC_MAX_DMA_RIWT);
+		netdev_alert(netdev, "rx-usec is limited to %d usecs\n",
+			     rx_usecs);
+		return -EINVAL;
+	}
+	if (rx_frames > pdata->channel->rx_ring->rdesc_count) {
+		netdev_alert(netdev, "rx-frames is limited to %d frames\n",
+			     pdata->channel->rx_ring->rdesc_count);
+		return -EINVAL;
+	}
+
+	tx_usecs = ec->tx_coalesce_usecs;
+	tx_frames = ec->tx_max_coalesced_frames;
+
+	/* Check the bounds of values for Tx */
+	if (tx_frames > pdata->channel->tx_ring->rdesc_count) {
+		netdev_alert(netdev, "tx-frames is limited to %d frames\n",
+			     pdata->channel->tx_ring->rdesc_count);
+		return -EINVAL;
+	}
+
+	pdata->rx_riwt = rx_riwt;
+	pdata->rx_frames = rx_frames;
+	hw_if->config_rx_coalesce(pdata);
+
+	pdata->tx_usecs = tx_usecs;
+	pdata->tx_frames = tx_frames;
+	hw_if->config_tx_coalesce(pdata);
+
+	DBGPR("<--xgbe_set_coalesce\n");
+
+	return 0;
+}
+
+static const struct ethtool_ops xgbe_ethtool_ops = {
+	.get_settings = xgbe_get_settings,
+	.set_settings = xgbe_set_settings,
+	.get_drvinfo = xgbe_get_drvinfo,
+	.get_link = ethtool_op_get_link,
+	.get_coalesce = xgbe_get_coalesce,
+	.set_coalesce = xgbe_set_coalesce,
+	.get_pauseparam = xgbe_get_pauseparam,
+	.set_pauseparam = xgbe_set_pauseparam,
+	.get_strings = xgbe_get_strings,
+	.get_ethtool_stats = xgbe_get_ethtool_stats,
+	.get_sset_count = xgbe_get_sset_count,
+};
+
+struct ethtool_ops *xgbe_get_ethtool_ops(void)
+{
+	return (struct ethtool_ops *)&xgbe_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-main.c b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
new file mode 100644
index 000000000000..c83584a26713
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
@@ -0,0 +1,512 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_net.h>
+#include <linux/clk.h>
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+MODULE_AUTHOR("Tom Lendacky <thomas.lendacky@amd.com>");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_VERSION(XGBE_DRV_VERSION);
+MODULE_DESCRIPTION(XGBE_DRV_DESC);
+
+static struct xgbe_channel *xgbe_alloc_rings(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel_mem, *channel;
+	struct xgbe_ring *tx_ring, *rx_ring;
+	unsigned int count, i;
+
+	DBGPR("-->xgbe_alloc_rings\n");
+
+	count = max_t(unsigned int, pdata->tx_ring_count, pdata->rx_ring_count);
+
+	channel_mem = devm_kcalloc(pdata->dev, count,
+				   sizeof(struct xgbe_channel), GFP_KERNEL);
+	if (!channel_mem)
+		return NULL;
+
+	tx_ring = devm_kcalloc(pdata->dev, pdata->tx_ring_count,
+			       sizeof(struct xgbe_ring), GFP_KERNEL);
+	if (!tx_ring)
+		return NULL;
+
+	rx_ring = devm_kcalloc(pdata->dev, pdata->rx_ring_count,
+			       sizeof(struct xgbe_ring), GFP_KERNEL);
+	if (!rx_ring)
+		return NULL;
+
+	for (i = 0, channel = channel_mem; i < count; i++, channel++) {
+		snprintf(channel->name, sizeof(channel->name), "channel-%d", i);
+		channel->pdata = pdata;
+		channel->queue_index = i;
+		channel->dma_regs = pdata->xgmac_regs + DMA_CH_BASE +
+				    (DMA_CH_INC * i);
+
+		if (i < pdata->tx_ring_count) {
+			spin_lock_init(&tx_ring->lock);
+			channel->tx_ring = tx_ring++;
+		}
+
+		if (i < pdata->rx_ring_count) {
+			spin_lock_init(&tx_ring->lock);
+			channel->rx_ring = rx_ring++;
+		}
+
+		DBGPR("  %s - queue_index=%u, dma_regs=%p, tx=%p, rx=%p\n",
+		      channel->name, channel->queue_index, channel->dma_regs,
+		      channel->tx_ring, channel->rx_ring);
+	}
+
+	pdata->channel_count = count;
+
+	DBGPR("<--xgbe_alloc_rings\n");
+
+	return channel_mem;
+}
+
+static void xgbe_default_config(struct xgbe_prv_data *pdata)
+{
+	DBGPR("-->xgbe_default_config\n");
+
+	pdata->pblx8 = DMA_PBL_X8_ENABLE;
+	pdata->tx_sf_mode = MTL_TSF_ENABLE;
+	pdata->tx_threshold = MTL_TX_THRESHOLD_64;
+	pdata->tx_pbl = DMA_PBL_16;
+	pdata->tx_osp_mode = DMA_OSP_ENABLE;
+	pdata->rx_sf_mode = MTL_RSF_DISABLE;
+	pdata->rx_threshold = MTL_RX_THRESHOLD_64;
+	pdata->rx_pbl = DMA_PBL_16;
+	pdata->pause_autoneg = 1;
+	pdata->tx_pause = 1;
+	pdata->rx_pause = 1;
+	pdata->power_down = 0;
+	pdata->default_autoneg = AUTONEG_ENABLE;
+	pdata->default_speed = SPEED_10000;
+
+	DBGPR("<--xgbe_default_config\n");
+}
+
+static void xgbe_init_all_fptrs(struct xgbe_prv_data *pdata)
+{
+	xgbe_init_function_ptrs_dev(&pdata->hw_if);
+	xgbe_init_function_ptrs_desc(&pdata->desc_if);
+}
+
+static int xgbe_probe(struct platform_device *pdev)
+{
+	struct xgbe_prv_data *pdata;
+	struct xgbe_hw_if *hw_if;
+	struct xgbe_desc_if *desc_if;
+	struct net_device *netdev;
+	struct device *dev = &pdev->dev;
+	struct resource *res;
+	const u8 *mac_addr;
+	int ret;
+
+	DBGPR("--> xgbe_probe\n");
+
+	netdev = alloc_etherdev_mq(sizeof(struct xgbe_prv_data),
+				   XGBE_MAX_DMA_CHANNELS);
+	if (!netdev) {
+		dev_err(dev, "alloc_etherdev failed\n");
+		ret = -ENOMEM;
+		goto err_alloc;
+	}
+	SET_NETDEV_DEV(netdev, dev);
+	pdata = netdev_priv(netdev);
+	pdata->netdev = netdev;
+	pdata->pdev = pdev;
+	pdata->dev = dev;
+	platform_set_drvdata(pdev, netdev);
+
+	spin_lock_init(&pdata->lock);
+	mutex_init(&pdata->xpcs_mutex);
+
+	/* Set and validate the number of descriptors for a ring */
+	BUILD_BUG_ON_NOT_POWER_OF_2(TX_DESC_CNT);
+	pdata->tx_desc_count = TX_DESC_CNT;
+	if (pdata->tx_desc_count & (pdata->tx_desc_count - 1)) {
+		dev_err(dev, "tx descriptor count (%d) is not valid\n",
+			pdata->tx_desc_count);
+		ret = -EINVAL;
+		goto err_io;
+	}
+	BUILD_BUG_ON_NOT_POWER_OF_2(RX_DESC_CNT);
+	pdata->rx_desc_count = RX_DESC_CNT;
+	if (pdata->rx_desc_count & (pdata->rx_desc_count - 1)) {
+		dev_err(dev, "rx descriptor count (%d) is not valid\n",
+			pdata->rx_desc_count);
+		ret = -EINVAL;
+		goto err_io;
+	}
+
+	/* Obtain the system clock setting */
+	pdata->sysclock = devm_clk_get(dev, NULL);
+	if (IS_ERR(pdata->sysclock)) {
+		dev_err(dev, "devm_clk_get failed\n");
+		ret = PTR_ERR(pdata->sysclock);
+		goto err_io;
+	}
+
+	/* Obtain the mmio areas for the device */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	pdata->xgmac_regs = devm_ioremap_resource(dev, res);
+	if (IS_ERR(pdata->xgmac_regs)) {
+		dev_err(dev, "xgmac ioremap failed\n");
+		ret = PTR_ERR(pdata->xgmac_regs);
+		goto err_io;
+	}
+	DBGPR("  xgmac_regs = %p\n", pdata->xgmac_regs);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	pdata->xpcs_regs = devm_ioremap_resource(dev, res);
+	if (IS_ERR(pdata->xpcs_regs)) {
+		dev_err(dev, "xpcs ioremap failed\n");
+		ret = PTR_ERR(pdata->xpcs_regs);
+		goto err_io;
+	}
+	DBGPR("  xpcs_regs  = %p\n", pdata->xpcs_regs);
+
+	/* Set the DMA mask */
+	if (!dev->dma_mask)
+		dev->dma_mask = &dev->coherent_dma_mask;
+	*(dev->dma_mask) = DMA_BIT_MASK(40);
+	dev->coherent_dma_mask = DMA_BIT_MASK(40);
+
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0) {
+		dev_err(dev, "platform_get_irq failed\n");
+		goto err_io;
+	}
+	netdev->irq = ret;
+	netdev->base_addr = (unsigned long)pdata->xgmac_regs;
+
+	/* Set all the function pointers */
+	xgbe_init_all_fptrs(pdata);
+	hw_if = &pdata->hw_if;
+	desc_if = &pdata->desc_if;
+
+	/* Issue software reset to device */
+	hw_if->exit(pdata);
+
+	/* Populate the hardware features */
+	xgbe_get_all_hw_features(pdata);
+
+	/* Retrieve the MAC address */
+	mac_addr = of_get_mac_address(dev->of_node);
+	if (!mac_addr) {
+		dev_err(dev, "invalid mac address for this device\n");
+		ret = -EINVAL;
+		goto err_io;
+	}
+	memcpy(netdev->dev_addr, mac_addr, netdev->addr_len);
+
+	/* Retrieve the PHY mode - it must be "xgmii" */
+	pdata->phy_mode = of_get_phy_mode(dev->of_node);
+	if (pdata->phy_mode != PHY_INTERFACE_MODE_XGMII) {
+		dev_err(dev, "invalid phy-mode specified for this device\n");
+		ret = -EINVAL;
+		goto err_io;
+	}
+
+	/* Set default configuration data */
+	xgbe_default_config(pdata);
+
+	/* Calculate the number of Tx and Rx rings to be created */
+	pdata->tx_ring_count = min_t(unsigned int, num_online_cpus(),
+				     pdata->hw_feat.tx_ch_cnt);
+	if (netif_set_real_num_tx_queues(netdev, pdata->tx_ring_count)) {
+		dev_err(dev, "error setting real tx queue count\n");
+		goto err_io;
+	}
+
+	pdata->rx_ring_count = min_t(unsigned int,
+				     netif_get_num_default_rss_queues(),
+				     pdata->hw_feat.rx_ch_cnt);
+	ret = netif_set_real_num_rx_queues(netdev, pdata->rx_ring_count);
+	if (ret) {
+		dev_err(dev, "error setting real rx queue count\n");
+		goto err_io;
+	}
+
+	/* Allocate the rings for the DMA channels */
+	pdata->channel = xgbe_alloc_rings(pdata);
+	if (!pdata->channel) {
+		dev_err(dev, "ring allocation failed\n");
+		ret = -ENOMEM;
+		goto err_io;
+	}
+
+	/* Prepare to regsiter with MDIO */
+	pdata->mii_bus_id = kasprintf(GFP_KERNEL, "%s", pdev->name);
+	if (!pdata->mii_bus_id) {
+		dev_err(dev, "failed to allocate mii bus id\n");
+		ret = -ENOMEM;
+		goto err_io;
+	}
+	ret = xgbe_mdio_register(pdata);
+	if (ret)
+		goto err_bus_id;
+
+	/* Set network and ethtool operations */
+	netdev->netdev_ops = xgbe_get_netdev_ops();
+	netdev->ethtool_ops = xgbe_get_ethtool_ops();
+
+	/* Set device features */
+	netdev->hw_features = NETIF_F_SG |
+			      NETIF_F_IP_CSUM |
+			      NETIF_F_IPV6_CSUM |
+			      NETIF_F_RXCSUM |
+			      NETIF_F_TSO |
+			      NETIF_F_TSO6 |
+			      NETIF_F_GRO |
+			      NETIF_F_HW_VLAN_CTAG_RX |
+			      NETIF_F_HW_VLAN_CTAG_TX;
+
+	netdev->vlan_features |= NETIF_F_SG |
+				 NETIF_F_IP_CSUM |
+				 NETIF_F_IPV6_CSUM |
+				 NETIF_F_TSO |
+				 NETIF_F_TSO6;
+
+	netdev->features |= netdev->hw_features;
+	pdata->netdev_features = netdev->features;
+
+	xgbe_init_rx_coalesce(pdata);
+	xgbe_init_tx_coalesce(pdata);
+
+	netif_carrier_off(netdev);
+	ret = register_netdev(netdev);
+	if (ret) {
+		dev_err(dev, "net device registration failed\n");
+		goto err_reg_netdev;
+	}
+
+	xgbe_debugfs_init(pdata);
+
+	netdev_notice(netdev, "net device enabled\n");
+
+	DBGPR("<-- xgbe_probe\n");
+
+	return 0;
+
+err_reg_netdev:
+	xgbe_mdio_unregister(pdata);
+
+err_bus_id:
+	kfree(pdata->mii_bus_id);
+
+err_io:
+	free_netdev(netdev);
+
+err_alloc:
+	dev_notice(dev, "net device not enabled\n");
+
+	return ret;
+}
+
+static int xgbe_remove(struct platform_device *pdev)
+{
+	struct net_device *netdev = platform_get_drvdata(pdev);
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	DBGPR("-->xgbe_remove\n");
+
+	xgbe_debugfs_exit(pdata);
+
+	unregister_netdev(netdev);
+
+	xgbe_mdio_unregister(pdata);
+
+	kfree(pdata->mii_bus_id);
+
+	free_netdev(netdev);
+
+	DBGPR("<--xgbe_remove\n");
+
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int xgbe_suspend(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	int ret;
+
+	DBGPR("-->xgbe_suspend\n");
+
+	if (!netif_running(netdev)) {
+		DBGPR("<--xgbe_dev_suspend\n");
+		return -EINVAL;
+	}
+
+	ret = xgbe_powerdown(netdev, XGMAC_DRIVER_CONTEXT);
+
+	DBGPR("<--xgbe_suspend\n");
+
+	return ret;
+}
+
+static int xgbe_resume(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	int ret;
+
+	DBGPR("-->xgbe_resume\n");
+
+	if (!netif_running(netdev)) {
+		DBGPR("<--xgbe_dev_resume\n");
+		return -EINVAL;
+	}
+
+	ret = xgbe_powerup(netdev, XGMAC_DRIVER_CONTEXT);
+
+	DBGPR("<--xgbe_resume\n");
+
+	return ret;
+}
+#endif /* CONFIG_PM */
+
+static const struct of_device_id xgbe_of_match[] = {
+	{ .compatible = "amd,xgbe-seattle-v1a", },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, xgbe_of_match);
+static SIMPLE_DEV_PM_OPS(xgbe_pm_ops, xgbe_suspend, xgbe_resume);
+
+static struct platform_driver xgbe_driver = {
+	.driver = {
+		.name = "amd-xgbe",
+		.of_match_table = xgbe_of_match,
+		.pm = &xgbe_pm_ops,
+	},
+	.probe = xgbe_probe,
+	.remove = xgbe_remove,
+};
+
+module_platform_driver(xgbe_driver);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
new file mode 100644
index 000000000000..ea7a5d6750ea
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
@@ -0,0 +1,433 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/module.h>
+#include <linux/kmod.h>
+#include <linux/spinlock.h>
+#include <linux/mdio.h>
+#include <linux/phy.h>
+#include <linux/of.h>
+
+#include "xgbe.h"
+#include "xgbe-common.h"
+
+
+static int xgbe_mdio_read(struct mii_bus *mii, int prtad, int mmd_reg)
+{
+	struct xgbe_prv_data *pdata = mii->priv;
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	int mmd_data;
+
+	DBGPR_MDIO("-->xgbe_mdio_read: prtad=%#x mmd_reg=%#x\n",
+		   prtad, mmd_reg);
+
+	mmd_data = hw_if->read_mmd_regs(pdata, prtad, mmd_reg);
+
+	DBGPR_MDIO("<--xgbe_mdio_read: mmd_data=%#x\n", mmd_data);
+
+	return mmd_data;
+}
+
+static int xgbe_mdio_write(struct mii_bus *mii, int prtad, int mmd_reg,
+			   u16 mmd_val)
+{
+	struct xgbe_prv_data *pdata = mii->priv;
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	int mmd_data = mmd_val;
+
+	DBGPR_MDIO("-->xgbe_mdio_write: prtad=%#x mmd_reg=%#x mmd_data=%#x\n",
+		   prtad, mmd_reg, mmd_data);
+
+	hw_if->write_mmd_regs(pdata, prtad, mmd_reg, mmd_data);
+
+	DBGPR_MDIO("<--xgbe_mdio_write\n");
+
+	return 0;
+}
+
+static void xgbe_adjust_link(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct phy_device *phydev = pdata->phydev;
+	unsigned long flags;
+	int new_state = 0;
+
+	if (phydev == NULL)
+		return;
+
+	DBGPR_MDIO("-->xgbe_adjust_link: address=%d, newlink=%d, curlink=%d\n",
+		   phydev->addr, phydev->link, pdata->phy_link);
+
+	spin_lock_irqsave(&pdata->lock, flags);
+
+	if (phydev->link) {
+		/* Flow control support */
+		if (pdata->pause_autoneg) {
+			if (phydev->pause || phydev->asym_pause) {
+				pdata->tx_pause = 1;
+				pdata->rx_pause = 1;
+			} else {
+				pdata->tx_pause = 0;
+				pdata->rx_pause = 0;
+			}
+		}
+
+		if (pdata->tx_pause != pdata->phy_tx_pause) {
+			hw_if->config_tx_flow_control(pdata);
+			pdata->phy_tx_pause = pdata->tx_pause;
+		}
+
+		if (pdata->rx_pause != pdata->phy_rx_pause) {
+			hw_if->config_rx_flow_control(pdata);
+			pdata->phy_rx_pause = pdata->rx_pause;
+		}
+
+		/* Speed support */
+		if (phydev->speed != pdata->phy_speed) {
+			new_state = 1;
+
+			switch (phydev->speed) {
+			case SPEED_10000:
+				hw_if->set_xgmii_speed(pdata);
+				break;
+
+			case SPEED_2500:
+				hw_if->set_gmii_2500_speed(pdata);
+				break;
+
+			case SPEED_1000:
+				hw_if->set_gmii_speed(pdata);
+				break;
+			}
+			pdata->phy_speed = phydev->speed;
+		}
+
+		if (phydev->link != pdata->phy_link) {
+			new_state = 1;
+			pdata->phy_link = 1;
+		}
+	} else if (pdata->phy_link) {
+		new_state = 1;
+		pdata->phy_link = 0;
+		pdata->phy_speed = SPEED_UNKNOWN;
+	}
+
+	if (new_state)
+		phy_print_status(phydev);
+
+	spin_unlock_irqrestore(&pdata->lock, flags);
+
+	DBGPR_MDIO("<--xgbe_adjust_link\n");
+}
+
+void xgbe_dump_phy_registers(struct xgbe_prv_data *pdata)
+{
+	struct device *dev = pdata->dev;
+	struct phy_device *phydev = pdata->mii->phy_map[XGBE_PRTAD];
+	int i;
+
+	dev_alert(dev, "\n************* PHY Reg dump **********************\n");
+
+	dev_alert(dev, "PCS Control Reg (%#04x) = %#04x\n", MDIO_CTRL1,
+		  XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1));
+	dev_alert(dev, "PCS Status Reg (%#04x) = %#04x\n", MDIO_STAT1,
+		  XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1));
+	dev_alert(dev, "Phy Id (PHYS ID 1 %#04x)= %#04x\n", MDIO_DEVID1,
+		  XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVID1));
+	dev_alert(dev, "Phy Id (PHYS ID 2 %#04x)= %#04x\n", MDIO_DEVID2,
+		  XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVID2));
+	dev_alert(dev, "Devices in Package (%#04x)= %#04x\n", MDIO_DEVS1,
+		  XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVS1));
+	dev_alert(dev, "Devices in Package (%#04x)= %#04x\n", MDIO_DEVS2,
+		  XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVS2));
+
+	dev_alert(dev, "Auto-Neg Control Reg (%#04x) = %#04x\n", MDIO_CTRL1,
+		  XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1));
+	dev_alert(dev, "Auto-Neg Status Reg (%#04x) = %#04x\n", MDIO_STAT1,
+		  XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_STAT1));
+	dev_alert(dev, "Auto-Neg Ad Reg 1 (%#04x) = %#04x\n",
+		  MDIO_AN_ADVERTISE,
+		  XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE));
+	dev_alert(dev, "Auto-Neg Ad Reg 2 (%#04x) = %#04x\n",
+		  MDIO_AN_ADVERTISE + 1,
+		  XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1));
+	dev_alert(dev, "Auto-Neg Ad Reg 3 (%#04x) = %#04x\n",
+		  MDIO_AN_ADVERTISE + 2,
+		  XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2));
+	dev_alert(dev, "Auto-Neg Completion Reg (%#04x) = %#04x\n",
+		  MDIO_AN_COMP_STAT,
+		  XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_COMP_STAT));
+
+	dev_alert(dev, "MMD Device Mask = %#x\n",
+		  phydev->c45_ids.devices_in_package);
+	for (i = 0; i < ARRAY_SIZE(phydev->c45_ids.device_ids); i++)
+		dev_alert(dev, "  MMD %d: ID = %#08x\n", i,
+			  phydev->c45_ids.device_ids[i]);
+
+	dev_alert(dev, "\n*************************************************\n");
+}
+
+int xgbe_mdio_register(struct xgbe_prv_data *pdata)
+{
+	struct net_device *netdev = pdata->netdev;
+	struct device_node *phy_node;
+	struct mii_bus *mii;
+	struct phy_device *phydev;
+	int ret = 0;
+
+	DBGPR("-->xgbe_mdio_register\n");
+
+	/* Retrieve the phy-handle */
+	phy_node = of_parse_phandle(pdata->dev->of_node, "phy-handle", 0);
+	if (!phy_node) {
+		dev_err(pdata->dev, "unable to parse phy-handle\n");
+		return -EINVAL;
+	}
+
+	/* Register with the MDIO bus */
+	mii = mdiobus_alloc();
+	if (mii == NULL) {
+		dev_err(pdata->dev, "mdiobus_alloc failed\n");
+		ret = -ENOMEM;
+		goto err_node_get;
+	}
+
+	/* Register on the MDIO bus (don't probe any PHYs) */
+	mii->name = XGBE_PHY_NAME;
+	mii->read = xgbe_mdio_read;
+	mii->write = xgbe_mdio_write;
+	snprintf(mii->id, sizeof(mii->id), "%s", pdata->mii_bus_id);
+	mii->priv = pdata;
+	mii->phy_mask = ~0;
+	mii->parent = pdata->dev;
+	ret = mdiobus_register(mii);
+	if (ret) {
+		dev_err(pdata->dev, "mdiobus_register failed\n");
+		goto err_mdiobus_alloc;
+	}
+	DBGPR("  mdiobus_register succeeded for %s\n", pdata->mii_bus_id);
+
+	/* Probe the PCS using Clause 45 */
+	phydev = get_phy_device(mii, XGBE_PRTAD, true);
+	if (IS_ERR(phydev) || !phydev ||
+	    !phydev->c45_ids.device_ids[MDIO_MMD_PCS]) {
+		dev_err(pdata->dev, "get_phy_device failed\n");
+		ret = phydev ? PTR_ERR(phydev) : -ENOLINK;
+		goto err_mdiobus_register;
+	}
+	request_module(MDIO_MODULE_PREFIX MDIO_ID_FMT,
+		       MDIO_ID_ARGS(phydev->c45_ids.device_ids[MDIO_MMD_PCS]));
+
+	of_node_get(phy_node);
+	phydev->dev.of_node = phy_node;
+	ret = phy_device_register(phydev);
+	if (ret) {
+		dev_err(pdata->dev, "phy_device_register failed\n");
+		of_node_put(phy_node);
+		goto err_phy_device;
+	}
+
+	/* Add a reference to the PHY driver so it can't be unloaded */
+	pdata->phy_module = phydev->dev.driver ?
+			    phydev->dev.driver->owner : NULL;
+	if (!try_module_get(pdata->phy_module)) {
+		dev_err(pdata->dev, "try_module_get failed\n");
+		ret = -EIO;
+		goto err_phy_device;
+	}
+
+	pdata->mii = mii;
+	pdata->mdio_mmd = MDIO_MMD_PCS;
+
+	pdata->phy_link = -1;
+	pdata->phy_speed = SPEED_UNKNOWN;
+	pdata->phy_tx_pause = pdata->tx_pause;
+	pdata->phy_rx_pause = pdata->rx_pause;
+
+	ret = phy_connect_direct(netdev, phydev, &xgbe_adjust_link,
+				 pdata->phy_mode);
+	if (ret) {
+		netdev_err(netdev, "phy_connect_direct failed\n");
+		goto err_phy_device;
+	}
+
+	if (!phydev->drv || (phydev->drv->phy_id == 0)) {
+		netdev_err(netdev, "phy_id not valid\n");
+		ret = -ENODEV;
+		goto err_phy_connect;
+	}
+	DBGPR("  phy_connect_direct succeeded for PHY %s, link=%d\n",
+	      dev_name(&phydev->dev), phydev->link);
+
+	phydev->autoneg = pdata->default_autoneg;
+	if (phydev->autoneg == AUTONEG_DISABLE) {
+		/* Add settings needed to force speed */
+		phydev->supported |= SUPPORTED_1000baseT_Full;
+		phydev->supported |= SUPPORTED_10000baseT_Full;
+
+		phydev->speed = pdata->default_speed;
+		phydev->duplex = DUPLEX_FULL;
+
+		phydev->advertising &= ~ADVERTISED_Autoneg;
+	}
+
+	pdata->phydev = phydev;
+
+	of_node_put(phy_node);
+
+	DBGPHY_REGS(pdata);
+
+	DBGPR("<--xgbe_mdio_register\n");
+
+	return 0;
+
+err_phy_connect:
+	phy_disconnect(phydev);
+
+err_phy_device:
+	phy_device_free(phydev);
+
+err_mdiobus_register:
+	mdiobus_unregister(mii);
+
+err_mdiobus_alloc:
+	mdiobus_free(mii);
+
+err_node_get:
+	of_node_put(phy_node);
+
+	return ret;
+}
+
+void xgbe_mdio_unregister(struct xgbe_prv_data *pdata)
+{
+	DBGPR("-->xgbe_mdio_unregister\n");
+
+	phy_disconnect(pdata->phydev);
+	pdata->phydev = NULL;
+
+	module_put(pdata->phy_module);
+	pdata->phy_module = NULL;
+
+	mdiobus_unregister(pdata->mii);
+	pdata->mii->priv = NULL;
+
+	mdiobus_free(pdata->mii);
+	pdata->mii = NULL;
+
+	DBGPR("<--xgbe_mdio_unregister\n");
+}
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h
new file mode 100644
index 000000000000..ab0627162c01
--- /dev/null
+++ b/drivers/net/ethernet/amd/xgbe/xgbe.h
@@ -0,0 +1,676 @@
+/*
+ * AMD 10Gb Ethernet driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * This file incorporates work covered by the following copyright and
+ * permission notice:
+ *     The Synopsys DWC ETHER XGMAC Software Driver and documentation
+ *     (hereinafter "Software") is an unsupported proprietary work of Synopsys,
+ *     Inc. unless otherwise expressly agreed to in writing between Synopsys
+ *     and you.
+ *
+ *     The Software IS NOT an item of Licensed Software or Licensed Product
+ *     under any End User Software License Agreement or Agreement for Licensed
+ *     Product with Synopsys or any supplement thereto.  Permission is hereby
+ *     granted, free of charge, to any person obtaining a copy of this software
+ *     annotated with this license and the Software, to deal in the Software
+ *     without restriction, including without limitation the rights to use,
+ *     copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+ *     of the Software, and to permit persons to whom the Software is furnished
+ *     to do so, subject to the following conditions:
+ *
+ *     The above copyright notice and this permission notice shall be included
+ *     in all copies or substantial portions of the Software.
+ *
+ *     THIS SOFTWARE IS BEING DISTRIBUTED BY SYNOPSYS SOLELY ON AN "AS IS"
+ *     BASIS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+ *     TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+ *     PARTICULAR PURPOSE ARE HEREBY DISCLAIMED. IN NO EVENT SHALL SYNOPSYS
+ *     BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *     CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *     SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ *     INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ *     CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ *     ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ *     THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __XGBE_H__
+#define __XGBE_H__
+
+#include <linux/dma-mapping.h>
+#include <linux/netdevice.h>
+#include <linux/workqueue.h>
+#include <linux/phy.h>
+
+
+#define XGBE_DRV_NAME		"amd-xgbe"
+#define XGBE_DRV_VERSION	"1.0.0-a"
+#define XGBE_DRV_DESC		"AMD 10 Gigabit Ethernet Driver"
+
+/* Descriptor related defines */
+#define TX_DESC_CNT		512
+#define TX_DESC_MIN_FREE	(TX_DESC_CNT >> 3)
+#define TX_DESC_MAX_PROC	(TX_DESC_CNT >> 1)
+#define RX_DESC_CNT		512
+
+#define TX_MAX_BUF_SIZE		(0x3fff & ~(64 - 1))
+
+#define RX_MIN_BUF_SIZE		(ETH_FRAME_LEN + ETH_FCS_LEN + VLAN_HLEN)
+#define RX_BUF_ALIGN		64
+
+#define XGBE_MAX_DMA_CHANNELS	16
+#define DMA_ARDOMAIN_SETTING	0x2
+#define DMA_ARCACHE_SETTING	0xb
+#define DMA_AWDOMAIN_SETTING	0x2
+#define DMA_AWCACHE_SETTING	0x7
+#define DMA_INTERRUPT_MASK	0x31c7
+
+#define XGMAC_MIN_PACKET	60
+#define XGMAC_STD_PACKET_MTU	1500
+#define XGMAC_MAX_STD_PACKET	1518
+#define XGMAC_JUMBO_PACKET_MTU	9000
+#define XGMAC_MAX_JUMBO_PACKET	9018
+
+#define MAX_MULTICAST_LIST	14
+#define TX_FLAGS_IP_PKT		0x00000001
+#define TX_FLAGS_TCP_PKT	0x00000002
+
+/* MDIO bus phy name */
+#define XGBE_PHY_NAME		"amd_xgbe_phy"
+#define XGBE_PRTAD		0
+
+/* Driver PMT macros */
+#define XGMAC_DRIVER_CONTEXT	1
+#define XGMAC_IOCTL_CONTEXT	2
+
+#define FIFO_SIZE_B(x)		(x)
+#define FIFO_SIZE_KB(x)		(x * 1024)
+
+#define XGBE_TC_CNT		2
+
+/* Helper macro for descriptor handling
+ *  Always use GET_DESC_DATA to access the descriptor data
+ *  since the index is free-running and needs to be and-ed
+ *  with the descriptor count value of the ring to index to
+ *  the proper descriptor data.
+ */
+#define GET_DESC_DATA(_ring, _idx)				\
+	((_ring)->rdata +					\
+	 ((_idx) & ((_ring)->rdesc_count - 1)))
+
+
+/* Default coalescing parameters */
+#define XGMAC_INIT_DMA_TX_USECS		100
+#define XGMAC_INIT_DMA_TX_FRAMES	16
+
+#define XGMAC_MAX_DMA_RIWT		0xff
+#define XGMAC_INIT_DMA_RX_USECS		100
+#define XGMAC_INIT_DMA_RX_FRAMES	16
+
+/* Flow control queue count */
+#define XGMAC_MAX_FLOW_CONTROL_QUEUES	8
+
+
+struct xgbe_prv_data;
+
+struct xgbe_packet_data {
+	unsigned int attributes;
+
+	unsigned int errors;
+
+	unsigned int rdesc_count;
+	unsigned int length;
+
+	unsigned int header_len;
+	unsigned int tcp_header_len;
+	unsigned int tcp_payload_len;
+	unsigned short mss;
+
+	unsigned short vlan_ctag;
+};
+
+/* Common Rx and Tx descriptor mapping */
+struct xgbe_ring_desc {
+	unsigned int desc0;
+	unsigned int desc1;
+	unsigned int desc2;
+	unsigned int desc3;
+};
+
+/* Structure used to hold information related to the descriptor
+ * and the packet associated with the descriptor (always use
+ * use the GET_DESC_DATA macro to access this data from the ring)
+ */
+struct xgbe_ring_data {
+	struct xgbe_ring_desc *rdesc;	/* Virtual address of descriptor */
+	dma_addr_t rdesc_dma;		/* DMA address of descriptor */
+
+	struct sk_buff *skb;		/* Virtual address of SKB */
+	dma_addr_t skb_dma;		/* DMA address of SKB data */
+	unsigned int skb_dma_len;	/* Length of SKB DMA area */
+	unsigned int tso_header;        /* TSO header indicator */
+
+	unsigned short len;		/* Length of received Rx packet */
+
+	unsigned int interrupt;		/* Interrupt indicator */
+
+	unsigned int mapped_as_page;
+};
+
+struct xgbe_ring {
+	/* Ring lock - used just for TX rings at the moment */
+	spinlock_t lock;
+
+	/* Per packet related information */
+	struct xgbe_packet_data packet_data;
+
+	/* Virtual/DMA addresses and count of allocated descriptor memory */
+	struct xgbe_ring_desc *rdesc;
+	dma_addr_t rdesc_dma;
+	unsigned int rdesc_count;
+
+	/* Array of descriptor data corresponding the descriptor memory
+	 * (always use the GET_DESC_DATA macro to access this data)
+	 */
+	struct xgbe_ring_data *rdata;
+
+	/* Ring index values
+	 *  cur   - Tx: index of descriptor to be used for current transfer
+	 *          Rx: index of descriptor to check for packet availability
+	 *  dirty - Tx: index of descriptor to check for transfer complete
+	 *          Rx: count of descriptors in which a packet has been received
+	 *              (used with skb_realloc_index to refresh the ring)
+	 */
+	unsigned int cur;
+	unsigned int dirty;
+
+	/* Coalesce frame count used for interrupt bit setting */
+	unsigned int coalesce_count;
+
+	union {
+		struct {
+			unsigned int queue_stopped;
+			unsigned short cur_mss;
+			unsigned short cur_vlan_ctag;
+		} tx;
+
+		struct {
+			unsigned int realloc_index;
+			unsigned int realloc_threshold;
+		} rx;
+	};
+} ____cacheline_aligned;
+
+/* Structure used to describe the descriptor rings associated with
+ * a DMA channel.
+ */
+struct xgbe_channel {
+	char name[16];
+
+	/* Address of private data area for device */
+	struct xgbe_prv_data *pdata;
+
+	/* Queue index and base address of queue's DMA registers */
+	unsigned int queue_index;
+	void __iomem *dma_regs;
+
+	unsigned int saved_ier;
+
+	unsigned int tx_timer_active;
+	struct hrtimer tx_timer;
+
+	struct xgbe_ring *tx_ring;
+	struct xgbe_ring *rx_ring;
+} ____cacheline_aligned;
+
+enum xgbe_int {
+	XGMAC_INT_DMA_ISR_DC0IS,
+	XGMAC_INT_DMA_CH_SR_TI,
+	XGMAC_INT_DMA_CH_SR_TPS,
+	XGMAC_INT_DMA_CH_SR_TBU,
+	XGMAC_INT_DMA_CH_SR_RI,
+	XGMAC_INT_DMA_CH_SR_RBU,
+	XGMAC_INT_DMA_CH_SR_RPS,
+	XGMAC_INT_DMA_CH_SR_FBE,
+	XGMAC_INT_DMA_ALL,
+};
+
+enum xgbe_int_state {
+	XGMAC_INT_STATE_SAVE,
+	XGMAC_INT_STATE_RESTORE,
+};
+
+enum xgbe_mtl_fifo_size {
+	XGMAC_MTL_FIFO_SIZE_256  = 0x00,
+	XGMAC_MTL_FIFO_SIZE_512  = 0x01,
+	XGMAC_MTL_FIFO_SIZE_1K   = 0x03,
+	XGMAC_MTL_FIFO_SIZE_2K   = 0x07,
+	XGMAC_MTL_FIFO_SIZE_4K   = 0x0f,
+	XGMAC_MTL_FIFO_SIZE_8K   = 0x1f,
+	XGMAC_MTL_FIFO_SIZE_16K  = 0x3f,
+	XGMAC_MTL_FIFO_SIZE_32K  = 0x7f,
+	XGMAC_MTL_FIFO_SIZE_64K  = 0xff,
+	XGMAC_MTL_FIFO_SIZE_128K = 0x1ff,
+	XGMAC_MTL_FIFO_SIZE_256K = 0x3ff,
+};
+
+struct xgbe_mmc_stats {
+	/* Tx Stats */
+	u64 txoctetcount_gb;
+	u64 txframecount_gb;
+	u64 txbroadcastframes_g;
+	u64 txmulticastframes_g;
+	u64 tx64octets_gb;
+	u64 tx65to127octets_gb;
+	u64 tx128to255octets_gb;
+	u64 tx256to511octets_gb;
+	u64 tx512to1023octets_gb;
+	u64 tx1024tomaxoctets_gb;
+	u64 txunicastframes_gb;
+	u64 txmulticastframes_gb;
+	u64 txbroadcastframes_gb;
+	u64 txunderflowerror;
+	u64 txoctetcount_g;
+	u64 txframecount_g;
+	u64 txpauseframes;
+	u64 txvlanframes_g;
+
+	/* Rx Stats */
+	u64 rxframecount_gb;
+	u64 rxoctetcount_gb;
+	u64 rxoctetcount_g;
+	u64 rxbroadcastframes_g;
+	u64 rxmulticastframes_g;
+	u64 rxcrcerror;
+	u64 rxrunterror;
+	u64 rxjabbererror;
+	u64 rxundersize_g;
+	u64 rxoversize_g;
+	u64 rx64octets_gb;
+	u64 rx65to127octets_gb;
+	u64 rx128to255octets_gb;
+	u64 rx256to511octets_gb;
+	u64 rx512to1023octets_gb;
+	u64 rx1024tomaxoctets_gb;
+	u64 rxunicastframes_g;
+	u64 rxlengtherror;
+	u64 rxoutofrangetype;
+	u64 rxpauseframes;
+	u64 rxfifooverflow;
+	u64 rxvlanframes_gb;
+	u64 rxwatchdogerror;
+};
+
+struct xgbe_hw_if {
+	int (*tx_complete)(struct xgbe_ring_desc *);
+
+	int (*set_promiscuous_mode)(struct xgbe_prv_data *, unsigned int);
+	int (*set_all_multicast_mode)(struct xgbe_prv_data *, unsigned int);
+	int (*set_addn_mac_addrs)(struct xgbe_prv_data *, unsigned int);
+	int (*set_mac_address)(struct xgbe_prv_data *, u8 *addr);
+
+	int (*enable_rx_csum)(struct xgbe_prv_data *);
+	int (*disable_rx_csum)(struct xgbe_prv_data *);
+
+	int (*enable_rx_vlan_stripping)(struct xgbe_prv_data *);
+	int (*disable_rx_vlan_stripping)(struct xgbe_prv_data *);
+
+	int (*read_mmd_regs)(struct xgbe_prv_data *, int, int);
+	void (*write_mmd_regs)(struct xgbe_prv_data *, int, int, int);
+	int (*set_gmii_speed)(struct xgbe_prv_data *);
+	int (*set_gmii_2500_speed)(struct xgbe_prv_data *);
+	int (*set_xgmii_speed)(struct xgbe_prv_data *);
+
+	void (*enable_tx)(struct xgbe_prv_data *);
+	void (*disable_tx)(struct xgbe_prv_data *);
+	void (*enable_rx)(struct xgbe_prv_data *);
+	void (*disable_rx)(struct xgbe_prv_data *);
+
+	void (*powerup_tx)(struct xgbe_prv_data *);
+	void (*powerdown_tx)(struct xgbe_prv_data *);
+	void (*powerup_rx)(struct xgbe_prv_data *);
+	void (*powerdown_rx)(struct xgbe_prv_data *);
+
+	int (*init)(struct xgbe_prv_data *);
+	int (*exit)(struct xgbe_prv_data *);
+
+	int (*enable_int)(struct xgbe_channel *, enum xgbe_int);
+	int (*disable_int)(struct xgbe_channel *, enum xgbe_int);
+	void (*pre_xmit)(struct xgbe_channel *);
+	int (*dev_read)(struct xgbe_channel *);
+	void (*tx_desc_init)(struct xgbe_channel *);
+	void (*rx_desc_init)(struct xgbe_channel *);
+	void (*rx_desc_reset)(struct xgbe_ring_data *);
+	void (*tx_desc_reset)(struct xgbe_ring_data *);
+	int (*is_last_desc)(struct xgbe_ring_desc *);
+	int (*is_context_desc)(struct xgbe_ring_desc *);
+
+	/* For FLOW ctrl */
+	int (*config_tx_flow_control)(struct xgbe_prv_data *);
+	int (*config_rx_flow_control)(struct xgbe_prv_data *);
+
+	/* For RX coalescing */
+	int (*config_rx_coalesce)(struct xgbe_prv_data *);
+	int (*config_tx_coalesce)(struct xgbe_prv_data *);
+	unsigned int (*usec_to_riwt)(struct xgbe_prv_data *, unsigned int);
+	unsigned int (*riwt_to_usec)(struct xgbe_prv_data *, unsigned int);
+
+	/* For RX and TX threshold config */
+	int (*config_rx_threshold)(struct xgbe_prv_data *, unsigned int);
+	int (*config_tx_threshold)(struct xgbe_prv_data *, unsigned int);
+
+	/* For RX and TX Store and Forward Mode config */
+	int (*config_rsf_mode)(struct xgbe_prv_data *, unsigned int);
+	int (*config_tsf_mode)(struct xgbe_prv_data *, unsigned int);
+
+	/* For TX DMA Operate on Second Frame config */
+	int (*config_osp_mode)(struct xgbe_prv_data *);
+
+	/* For RX and TX PBL config */
+	int (*config_rx_pbl_val)(struct xgbe_prv_data *);
+	int (*get_rx_pbl_val)(struct xgbe_prv_data *);
+	int (*config_tx_pbl_val)(struct xgbe_prv_data *);
+	int (*get_tx_pbl_val)(struct xgbe_prv_data *);
+	int (*config_pblx8)(struct xgbe_prv_data *);
+
+	/* For MMC statistics */
+	void (*rx_mmc_int)(struct xgbe_prv_data *);
+	void (*tx_mmc_int)(struct xgbe_prv_data *);
+	void (*read_mmc_stats)(struct xgbe_prv_data *);
+};
+
+struct xgbe_desc_if {
+	int (*alloc_ring_resources)(struct xgbe_prv_data *);
+	void (*free_ring_resources)(struct xgbe_prv_data *);
+	int (*map_tx_skb)(struct xgbe_channel *, struct sk_buff *);
+	void (*realloc_skb)(struct xgbe_channel *);
+	void (*unmap_skb)(struct xgbe_prv_data *, struct xgbe_ring_data *);
+	void (*wrapper_tx_desc_init)(struct xgbe_prv_data *);
+	void (*wrapper_rx_desc_init)(struct xgbe_prv_data *);
+};
+
+/* This structure contains flags that indicate what hardware features
+ * or configurations are present in the device.
+ */
+struct xgbe_hw_features {
+	/* HW Feature Register0 */
+	unsigned int gmii;		/* 1000 Mbps support */
+	unsigned int vlhash;		/* VLAN Hash Filter */
+	unsigned int sma;		/* SMA(MDIO) Interface */
+	unsigned int rwk;		/* PMT remote wake-up packet */
+	unsigned int mgk;		/* PMT magic packet */
+	unsigned int mmc;		/* RMON module */
+	unsigned int aoe;		/* ARP Offload */
+	unsigned int ts;		/* IEEE 1588-2008 Adavanced Timestamp */
+	unsigned int eee;		/* Energy Efficient Ethernet */
+	unsigned int tx_coe;		/* Tx Checksum Offload */
+	unsigned int rx_coe;		/* Rx Checksum Offload */
+	unsigned int addn_mac;		/* Additional MAC Addresses */
+	unsigned int ts_src;		/* Timestamp Source */
+	unsigned int sa_vlan_ins;	/* Source Address or VLAN Insertion */
+
+	/* HW Feature Register1 */
+	unsigned int rx_fifo_size;	/* MTL Receive FIFO Size */
+	unsigned int tx_fifo_size;	/* MTL Transmit FIFO Size */
+	unsigned int adv_ts_hi;		/* Advance Timestamping High Word */
+	unsigned int dcb;		/* DCB Feature */
+	unsigned int sph;		/* Split Header Feature */
+	unsigned int tso;		/* TCP Segmentation Offload */
+	unsigned int dma_debug;		/* DMA Debug Registers */
+	unsigned int rss;		/* Receive Side Scaling */
+	unsigned int hash_table_size;	/* Hash Table Size */
+	unsigned int l3l4_filter_num;	/* Number of L3-L4 Filters */
+
+	/* HW Feature Register2 */
+	unsigned int rx_q_cnt;		/* Number of MTL Receive Queues */
+	unsigned int tx_q_cnt;		/* Number of MTL Transmit Queues */
+	unsigned int rx_ch_cnt;		/* Number of DMA Receive Channels */
+	unsigned int tx_ch_cnt;		/* Number of DMA Transmit Channels */
+	unsigned int pps_out_num;	/* Number of PPS outputs */
+	unsigned int aux_snap_num;	/* Number of Aux snapshot inputs */
+};
+
+struct xgbe_prv_data {
+	struct net_device *netdev;
+	struct platform_device *pdev;
+	struct device *dev;
+
+	/* XGMAC/XPCS related mmio registers */
+	void __iomem *xgmac_regs;	/* XGMAC CSRs */
+	void __iomem *xpcs_regs;	/* XPCS MMD registers */
+
+	/* Overall device lock */
+	spinlock_t lock;
+
+	/* XPCS indirect addressing mutex */
+	struct mutex xpcs_mutex;
+
+	int irq_number;
+
+	struct xgbe_hw_if hw_if;
+	struct xgbe_desc_if desc_if;
+
+	/* Rings for Tx/Rx on a DMA channel */
+	struct xgbe_channel *channel;
+	unsigned int channel_count;
+	unsigned int tx_ring_count;
+	unsigned int tx_desc_count;
+	unsigned int rx_ring_count;
+	unsigned int rx_desc_count;
+
+	/* Tx/Rx common settings */
+	unsigned int pblx8;
+
+	/* Tx settings */
+	unsigned int tx_sf_mode;
+	unsigned int tx_threshold;
+	unsigned int tx_pbl;
+	unsigned int tx_osp_mode;
+
+	/* Rx settings */
+	unsigned int rx_sf_mode;
+	unsigned int rx_threshold;
+	unsigned int rx_pbl;
+
+	/* Tx coalescing settings */
+	unsigned int tx_usecs;
+	unsigned int tx_frames;
+
+	/* Rx coalescing settings */
+	unsigned int rx_riwt;
+	unsigned int rx_frames;
+
+	/* Current MTU */
+	unsigned int rx_buf_size;
+
+	/* Flow control settings */
+	unsigned int pause_autoneg;
+	unsigned int tx_pause;
+	unsigned int rx_pause;
+
+	/* MDIO settings */
+	struct module *phy_module;
+	char *mii_bus_id;
+	struct mii_bus *mii;
+	int mdio_mmd;
+	struct phy_device *phydev;
+	int default_autoneg;
+	int default_speed;
+
+	/* Current PHY settings */
+	phy_interface_t phy_mode;
+	int phy_link;
+	int phy_speed;
+	unsigned int phy_tx_pause;
+	unsigned int phy_rx_pause;
+
+	/* Netdev related settings */
+	netdev_features_t netdev_features;
+	struct napi_struct napi;
+	struct xgbe_mmc_stats mmc_stats;
+
+	/* System clock value used for Rx watchdog */
+	struct clk *sysclock;
+
+	/* Hardware features of the device */
+	struct xgbe_hw_features hw_feat;
+
+	/* Device restart work structure */
+	struct work_struct restart_work;
+
+	/* Keeps track of power mode */
+	unsigned int power_down;
+
+#ifdef CONFIG_DEBUG_FS
+	struct dentry *xgbe_debugfs;
+
+	unsigned int debugfs_xgmac_reg;
+
+	unsigned int debugfs_xpcs_mmd;
+	unsigned int debugfs_xpcs_reg;
+#endif
+};
+
+/* Function prototypes*/
+
+void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *);
+void xgbe_init_function_ptrs_desc(struct xgbe_desc_if *);
+struct net_device_ops *xgbe_get_netdev_ops(void);
+struct ethtool_ops *xgbe_get_ethtool_ops(void);
+
+int xgbe_mdio_register(struct xgbe_prv_data *);
+void xgbe_mdio_unregister(struct xgbe_prv_data *);
+void xgbe_dump_phy_registers(struct xgbe_prv_data *);
+void xgbe_dump_tx_desc(struct xgbe_ring *, unsigned int, unsigned int,
+		       unsigned int);
+void xgbe_dump_rx_desc(struct xgbe_ring *, struct xgbe_ring_desc *,
+		       unsigned int);
+void xgbe_print_pkt(struct net_device *, struct sk_buff *, bool);
+void xgbe_get_all_hw_features(struct xgbe_prv_data *);
+int xgbe_powerup(struct net_device *, unsigned int);
+int xgbe_powerdown(struct net_device *, unsigned int);
+void xgbe_init_rx_coalesce(struct xgbe_prv_data *);
+void xgbe_init_tx_coalesce(struct xgbe_prv_data *);
+
+#ifdef CONFIG_DEBUG_FS
+void xgbe_debugfs_init(struct xgbe_prv_data *);
+void xgbe_debugfs_exit(struct xgbe_prv_data *);
+#else
+static inline void xgbe_debugfs_init(struct xgbe_prv_data *pdata) {}
+static inline void xgbe_debugfs_exit(struct xgbe_prv_data *pdata) {}
+#endif /* CONFIG_DEBUG_FS */
+
+/* NOTE: Uncomment for TX and RX DESCRIPTOR DUMP in KERNEL LOG */
+#if 0
+#define XGMAC_ENABLE_TX_DESC_DUMP
+#define XGMAC_ENABLE_RX_DESC_DUMP
+#endif
+
+/* NOTE: Uncomment for TX and RX PACKET DUMP in KERNEL LOG */
+#if 0
+#define XGMAC_ENABLE_TX_PKT_DUMP
+#define XGMAC_ENABLE_RX_PKT_DUMP
+#endif
+
+/* NOTE: Uncomment for function trace log messages in KERNEL LOG */
+#if 0
+#define YDEBUG
+#define YDEBUG_MDIO
+#endif
+
+/* For debug prints */
+#ifdef YDEBUG
+#define DBGPR(x...) pr_alert(x)
+#define DBGPHY_REGS(x...) xgbe_dump_phy_registers(x)
+#else
+#define DBGPR(x...) do { } while (0)
+#define DBGPHY_REGS(x...) do { } while (0)
+#endif
+
+#ifdef YDEBUG_MDIO
+#define DBGPR_MDIO(x...) pr_alert(x)
+#else
+#define DBGPR_MDIO(x...) do { } while (0)
+#endif
+
+#endif
diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c
index d647a7d115ac..18e2faccebb0 100644
--- a/drivers/net/ethernet/arc/emac_main.c
+++ b/drivers/net/ethernet/arc/emac_main.c
@@ -13,6 +13,7 @@
  *		Vineet Gupta
  */
 
+#include <linux/crc32.h>
 #include <linux/etherdevice.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
@@ -362,6 +363,15 @@ static irqreturn_t arc_emac_intr(int irq, void *dev_instance)
 	return IRQ_HANDLED;
 }
 
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void arc_emac_poll_controller(struct net_device *dev)
+{
+	disable_irq(dev->irq);
+	arc_emac_intr(dev->irq, dev);
+	enable_irq(dev->irq);
+}
+#endif
+
 /**
  * arc_emac_open - Open the network device.
  * @ndev:	Pointer to the network device.
@@ -451,6 +461,41 @@ static int arc_emac_open(struct net_device *ndev)
 }
 
 /**
+ * arc_emac_set_rx_mode - Change the receive filtering mode.
+ * @ndev:	Pointer to the network device.
+ *
+ * This function enables/disables promiscuous or all-multicast mode
+ * and updates the multicast filtering list of the network device.
+ */
+static void arc_emac_set_rx_mode(struct net_device *ndev)
+{
+	struct arc_emac_priv *priv = netdev_priv(ndev);
+
+	if (ndev->flags & IFF_PROMISC) {
+		arc_reg_or(priv, R_CTRL, PROM_MASK);
+	} else {
+		arc_reg_clr(priv, R_CTRL, PROM_MASK);
+
+		if (ndev->flags & IFF_ALLMULTI) {
+			arc_reg_set(priv, R_LAFL, ~0);
+			arc_reg_set(priv, R_LAFH, ~0);
+		} else {
+			struct netdev_hw_addr *ha;
+			unsigned int filter[2] = { 0, 0 };
+			int bit;
+
+			netdev_for_each_mc_addr(ha, ndev) {
+				bit = ether_crc_le(ETH_ALEN, ha->addr) >> 26;
+				filter[bit >> 5] |= 1 << (bit & 31);
+			}
+
+			arc_reg_set(priv, R_LAFL, filter[0]);
+			arc_reg_set(priv, R_LAFH, filter[1]);
+		}
+	}
+}
+
+/**
  * arc_emac_stop - Close the network device.
  * @ndev:	Pointer to the network device.
  *
@@ -620,6 +665,10 @@ static const struct net_device_ops arc_emac_netdev_ops = {
 	.ndo_start_xmit		= arc_emac_tx,
 	.ndo_set_mac_address	= arc_emac_set_address,
 	.ndo_get_stats		= arc_emac_stats,
+	.ndo_set_rx_mode	= arc_emac_set_rx_mode,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= arc_emac_poll_controller,
+#endif
 };
 
 static int arc_emac_probe(struct platform_device *pdev)
diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index 17bb9ce96260..49faa97a30c3 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -1302,7 +1302,7 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	netdev->netdev_ops = &alx_netdev_ops;
-	SET_ETHTOOL_OPS(netdev, &alx_ethtool_ops);
+	netdev->ethtool_ops = &alx_ethtool_ops;
 	netdev->irq = pdev->irq;
 	netdev->watchdog_timeo = ALX_WATCHDOG_TIME;
 
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c b/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c
index 859ea844ba0f..48694c239d5c 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c
@@ -56,8 +56,8 @@ static int atl1c_get_settings(struct net_device *netdev,
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = AUTONEG_ENABLE;
@@ -305,5 +305,5 @@ static const struct ethtool_ops atl1c_ethtool_ops = {
 
 void atl1c_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &atl1c_ethtool_ops);
+	netdev->ethtool_ops = &atl1c_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c
index 82b23861bf55..1be072f4afc2 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c
@@ -57,8 +57,8 @@ static int atl1e_get_settings(struct net_device *netdev,
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = AUTONEG_ENABLE;
@@ -388,5 +388,5 @@ static const struct ethtool_ops atl1e_ethtool_ops = {
 
 void atl1e_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &atl1e_ethtool_ops);
+	netdev->ethtool_ops = &atl1e_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c
index dfd0e91fa726..b460db7919a2 100644
--- a/drivers/net/ethernet/atheros/atlx/atl1.c
+++ b/drivers/net/ethernet/atheros/atlx/atl1.c
@@ -3258,8 +3258,8 @@ static int atl1_get_settings(struct net_device *netdev,
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 	if (hw->media_type == MEDIA_TYPE_AUTO_SENSOR ||
 	    hw->media_type == MEDIA_TYPE_1000M_FULL)
diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c
index 78befb522a52..6746bd717146 100644
--- a/drivers/net/ethernet/atheros/atlx/atl2.c
+++ b/drivers/net/ethernet/atheros/atlx/atl2.c
@@ -1396,7 +1396,7 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	atl2_setup_pcicmd(pdev);
 
 	netdev->netdev_ops = &atl2_netdev_ops;
-	SET_ETHTOOL_OPS(netdev, &atl2_ethtool_ops);
+	netdev->ethtool_ops = &atl2_ethtool_ops;
 	netdev->watchdog_timeo = 5 * HZ;
 	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
 
@@ -1769,8 +1769,8 @@ static int atl2_get_settings(struct net_device *netdev,
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = AUTONEG_ENABLE;
diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig
index 85dbddd03722..3e488094b073 100644
--- a/drivers/net/ethernet/broadcom/Kconfig
+++ b/drivers/net/ethernet/broadcom/Kconfig
@@ -150,4 +150,15 @@ config BGMAC
 	  In case of using this driver on BCM4706 it's also requires to enable
 	  BCMA_DRIVER_GMAC_CMN to make it work.
 
+config SYSTEMPORT
+	tristate "Broadcom SYSTEMPORT internal MAC support"
+	depends on OF
+	select MII
+	select PHYLIB
+	select FIXED_PHY if SYSTEMPORT=y
+	help
+	  This driver supports the built-in Ethernet MACs found in the
+	  Broadcom BCM7xxx Set Top Box family chipset using an internal
+	  Ethernet switch.
+
 endif # NET_VENDOR_BROADCOM
diff --git a/drivers/net/ethernet/broadcom/Makefile b/drivers/net/ethernet/broadcom/Makefile
index fd639a0d4c7d..e2a958a657e0 100644
--- a/drivers/net/ethernet/broadcom/Makefile
+++ b/drivers/net/ethernet/broadcom/Makefile
@@ -11,3 +11,4 @@ obj-$(CONFIG_BNX2X) += bnx2x/
 obj-$(CONFIG_SB1250_MAC) += sb1250-mac.o
 obj-$(CONFIG_TIGON3) += tg3.o
 obj-$(CONFIG_BGMAC) += bgmac.o
+obj-$(CONFIG_SYSTEMPORT) += bcmsysport.o
diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c
index 05ba62589017..ca5a20a48b14 100644
--- a/drivers/net/ethernet/broadcom/b44.c
+++ b/drivers/net/ethernet/broadcom/b44.c
@@ -2380,7 +2380,7 @@ static int b44_init_one(struct ssb_device *sdev,
 	netif_napi_add(dev, &bp->napi, b44_poll, 64);
 	dev->watchdog_timeo = B44_TX_TIMEOUT;
 	dev->irq = sdev->irq;
-	SET_ETHTOOL_OPS(dev, &b44_ethtool_ops);
+	dev->ethtool_ops = &b44_ethtool_ops;
 
 	err = ssb_bus_powerup(sdev->bus, 0);
 	if (err) {
diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
index a7d11f5565d6..3e8d1a88ed3d 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
@@ -1315,8 +1315,7 @@ static const struct bcm_enet_stats bcm_enet_gstrings_stats[] = {
 
 };
 
-#define BCM_ENET_STATS_LEN	\
-	(sizeof(bcm_enet_gstrings_stats) / sizeof(struct bcm_enet_stats))
+#define BCM_ENET_STATS_LEN	ARRAY_SIZE(bcm_enet_gstrings_stats)
 
 static const u32 unused_mib_regs[] = {
 	ETH_MIB_TX_ALL_OCTETS,
@@ -1898,7 +1897,7 @@ static int bcm_enet_probe(struct platform_device *pdev)
 	dev->netdev_ops = &bcm_enet_ops;
 	netif_napi_add(dev, &priv->napi, bcm_enet_poll, 16);
 
-	SET_ETHTOOL_OPS(dev, &bcm_enet_ethtool_ops);
+	dev->ethtool_ops = &bcm_enet_ethtool_ops;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	ret = register_netdev(dev);
@@ -2784,7 +2783,7 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 	/* register netdevice */
 	dev->netdev_ops = &bcm_enetsw_ops;
 	netif_napi_add(dev, &priv->napi, bcm_enet_poll, 16);
-	SET_ETHTOOL_OPS(dev, &bcm_enetsw_ethtool_ops);
+	dev->ethtool_ops = &bcm_enetsw_ethtool_ops;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	spin_lock_init(&priv->enetsw_mdio_lock);
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
new file mode 100644
index 000000000000..141160ef249a
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -0,0 +1,1654 @@
+/*
+ * Broadcom BCM7xxx System Port Ethernet MAC driver
+ *
+ * Copyright (C) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#define pr_fmt(fmt)	KBUILD_MODNAME ": " fmt
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+#include <linux/of_net.h>
+#include <linux/of_mdio.h>
+#include <linux/phy.h>
+#include <linux/phy_fixed.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+
+#include "bcmsysport.h"
+
+/* I/O accessors register helpers */
+#define BCM_SYSPORT_IO_MACRO(name, offset) \
+static inline u32 name##_readl(struct bcm_sysport_priv *priv, u32 off)	\
+{									\
+	u32 reg = __raw_readl(priv->base + offset + off);		\
+	return reg;							\
+}									\
+static inline void name##_writel(struct bcm_sysport_priv *priv,		\
+				  u32 val, u32 off)			\
+{									\
+	__raw_writel(val, priv->base + offset + off);			\
+}									\
+
+BCM_SYSPORT_IO_MACRO(intrl2_0, SYS_PORT_INTRL2_0_OFFSET);
+BCM_SYSPORT_IO_MACRO(intrl2_1, SYS_PORT_INTRL2_1_OFFSET);
+BCM_SYSPORT_IO_MACRO(umac, SYS_PORT_UMAC_OFFSET);
+BCM_SYSPORT_IO_MACRO(tdma, SYS_PORT_TDMA_OFFSET);
+BCM_SYSPORT_IO_MACRO(rdma, SYS_PORT_RDMA_OFFSET);
+BCM_SYSPORT_IO_MACRO(rxchk, SYS_PORT_RXCHK_OFFSET);
+BCM_SYSPORT_IO_MACRO(txchk, SYS_PORT_TXCHK_OFFSET);
+BCM_SYSPORT_IO_MACRO(rbuf, SYS_PORT_RBUF_OFFSET);
+BCM_SYSPORT_IO_MACRO(tbuf, SYS_PORT_TBUF_OFFSET);
+BCM_SYSPORT_IO_MACRO(topctrl, SYS_PORT_TOPCTRL_OFFSET);
+
+/* L2-interrupt masking/unmasking helpers, does automatic saving of the applied
+ * mask in a software copy to avoid CPU_MASK_STATUS reads in hot-paths.
+  */
+#define BCM_SYSPORT_INTR_L2(which)	\
+static inline void intrl2_##which##_mask_clear(struct bcm_sysport_priv *priv, \
+						u32 mask)		\
+{									\
+	intrl2_##which##_writel(priv, mask, INTRL2_CPU_MASK_CLEAR);	\
+	priv->irq##which##_mask &= ~(mask);				\
+}									\
+static inline void intrl2_##which##_mask_set(struct bcm_sysport_priv *priv, \
+						u32 mask)		\
+{									\
+	intrl2_## which##_writel(priv, mask, INTRL2_CPU_MASK_SET);	\
+	priv->irq##which##_mask |= (mask);				\
+}									\
+
+BCM_SYSPORT_INTR_L2(0)
+BCM_SYSPORT_INTR_L2(1)
+
+/* Register accesses to GISB/RBUS registers are expensive (few hundred
+ * nanoseconds), so keep the check for 64-bits explicit here to save
+ * one register write per-packet on 32-bits platforms.
+ */
+static inline void dma_desc_set_addr(struct bcm_sysport_priv *priv,
+				     void __iomem *d,
+				     dma_addr_t addr)
+{
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+	__raw_writel(upper_32_bits(addr) & DESC_ADDR_HI_MASK,
+			d + DESC_ADDR_HI_STATUS_LEN);
+#endif
+	__raw_writel(lower_32_bits(addr), d + DESC_ADDR_LO);
+}
+
+static inline void tdma_port_write_desc_addr(struct bcm_sysport_priv *priv,
+						struct dma_desc *desc,
+						unsigned int port)
+{
+	/* Ports are latched, so write upper address first */
+	tdma_writel(priv, desc->addr_status_len, TDMA_WRITE_PORT_HI(port));
+	tdma_writel(priv, desc->addr_lo, TDMA_WRITE_PORT_LO(port));
+}
+
+/* Ethtool operations */
+static int bcm_sysport_set_settings(struct net_device *dev,
+				    struct ethtool_cmd *cmd)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	return phy_ethtool_sset(priv->phydev, cmd);
+}
+
+static int bcm_sysport_get_settings(struct net_device *dev,
+					struct ethtool_cmd *cmd)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+
+	if (!netif_running(dev))
+		return -EINVAL;
+
+	return phy_ethtool_gset(priv->phydev, cmd);
+}
+
+static int bcm_sysport_set_rx_csum(struct net_device *dev,
+					netdev_features_t wanted)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	u32 reg;
+
+	priv->rx_csum_en = !!(wanted & NETIF_F_RXCSUM);
+	reg = rxchk_readl(priv, RXCHK_CONTROL);
+	if (priv->rx_csum_en)
+		reg |= RXCHK_EN;
+	else
+		reg &= ~RXCHK_EN;
+
+	/* If UniMAC forwards CRC, we need to skip over it to get
+	 * a valid CHK bit to be set in the per-packet status word
+	 */
+	if (priv->rx_csum_en && priv->crc_fwd)
+		reg |= RXCHK_SKIP_FCS;
+	else
+		reg &= ~RXCHK_SKIP_FCS;
+
+	rxchk_writel(priv, reg, RXCHK_CONTROL);
+
+	return 0;
+}
+
+static int bcm_sysport_set_tx_csum(struct net_device *dev,
+					netdev_features_t wanted)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	u32 reg;
+
+	/* Hardware transmit checksum requires us to enable the Transmit status
+	 * block prepended to the packet contents
+	 */
+	priv->tsb_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
+	reg = tdma_readl(priv, TDMA_CONTROL);
+	if (priv->tsb_en)
+		reg |= TSB_EN;
+	else
+		reg &= ~TSB_EN;
+	tdma_writel(priv, reg, TDMA_CONTROL);
+
+	return 0;
+}
+
+static int bcm_sysport_set_features(struct net_device *dev,
+					netdev_features_t features)
+{
+	netdev_features_t changed = features ^ dev->features;
+	netdev_features_t wanted = dev->wanted_features;
+	int ret = 0;
+
+	if (changed & NETIF_F_RXCSUM)
+		ret = bcm_sysport_set_rx_csum(dev, wanted);
+	if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
+		ret = bcm_sysport_set_tx_csum(dev, wanted);
+
+	return ret;
+}
+
+/* Hardware counters must be kept in sync because the order/offset
+ * is important here (order in structure declaration = order in hardware)
+ */
+static const struct bcm_sysport_stats bcm_sysport_gstrings_stats[] = {
+	/* general stats */
+	STAT_NETDEV(rx_packets),
+	STAT_NETDEV(tx_packets),
+	STAT_NETDEV(rx_bytes),
+	STAT_NETDEV(tx_bytes),
+	STAT_NETDEV(rx_errors),
+	STAT_NETDEV(tx_errors),
+	STAT_NETDEV(rx_dropped),
+	STAT_NETDEV(tx_dropped),
+	STAT_NETDEV(multicast),
+	/* UniMAC RSV counters */
+	STAT_MIB_RX("rx_64_octets", mib.rx.pkt_cnt.cnt_64),
+	STAT_MIB_RX("rx_65_127_oct", mib.rx.pkt_cnt.cnt_127),
+	STAT_MIB_RX("rx_128_255_oct", mib.rx.pkt_cnt.cnt_255),
+	STAT_MIB_RX("rx_256_511_oct", mib.rx.pkt_cnt.cnt_511),
+	STAT_MIB_RX("rx_512_1023_oct", mib.rx.pkt_cnt.cnt_1023),
+	STAT_MIB_RX("rx_1024_1518_oct", mib.rx.pkt_cnt.cnt_1518),
+	STAT_MIB_RX("rx_vlan_1519_1522_oct", mib.rx.pkt_cnt.cnt_mgv),
+	STAT_MIB_RX("rx_1522_2047_oct", mib.rx.pkt_cnt.cnt_2047),
+	STAT_MIB_RX("rx_2048_4095_oct", mib.rx.pkt_cnt.cnt_4095),
+	STAT_MIB_RX("rx_4096_9216_oct", mib.rx.pkt_cnt.cnt_9216),
+	STAT_MIB_RX("rx_pkts", mib.rx.pkt),
+	STAT_MIB_RX("rx_bytes", mib.rx.bytes),
+	STAT_MIB_RX("rx_multicast", mib.rx.mca),
+	STAT_MIB_RX("rx_broadcast", mib.rx.bca),
+	STAT_MIB_RX("rx_fcs", mib.rx.fcs),
+	STAT_MIB_RX("rx_control", mib.rx.cf),
+	STAT_MIB_RX("rx_pause", mib.rx.pf),
+	STAT_MIB_RX("rx_unknown", mib.rx.uo),
+	STAT_MIB_RX("rx_align", mib.rx.aln),
+	STAT_MIB_RX("rx_outrange", mib.rx.flr),
+	STAT_MIB_RX("rx_code", mib.rx.cde),
+	STAT_MIB_RX("rx_carrier", mib.rx.fcr),
+	STAT_MIB_RX("rx_oversize", mib.rx.ovr),
+	STAT_MIB_RX("rx_jabber", mib.rx.jbr),
+	STAT_MIB_RX("rx_mtu_err", mib.rx.mtue),
+	STAT_MIB_RX("rx_good_pkts", mib.rx.pok),
+	STAT_MIB_RX("rx_unicast", mib.rx.uc),
+	STAT_MIB_RX("rx_ppp", mib.rx.ppp),
+	STAT_MIB_RX("rx_crc", mib.rx.rcrc),
+	/* UniMAC TSV counters */
+	STAT_MIB_TX("tx_64_octets", mib.tx.pkt_cnt.cnt_64),
+	STAT_MIB_TX("tx_65_127_oct", mib.tx.pkt_cnt.cnt_127),
+	STAT_MIB_TX("tx_128_255_oct", mib.tx.pkt_cnt.cnt_255),
+	STAT_MIB_TX("tx_256_511_oct", mib.tx.pkt_cnt.cnt_511),
+	STAT_MIB_TX("tx_512_1023_oct", mib.tx.pkt_cnt.cnt_1023),
+	STAT_MIB_TX("tx_1024_1518_oct", mib.tx.pkt_cnt.cnt_1518),
+	STAT_MIB_TX("tx_vlan_1519_1522_oct", mib.tx.pkt_cnt.cnt_mgv),
+	STAT_MIB_TX("tx_1522_2047_oct", mib.tx.pkt_cnt.cnt_2047),
+	STAT_MIB_TX("tx_2048_4095_oct", mib.tx.pkt_cnt.cnt_4095),
+	STAT_MIB_TX("tx_4096_9216_oct", mib.tx.pkt_cnt.cnt_9216),
+	STAT_MIB_TX("tx_pkts", mib.tx.pkts),
+	STAT_MIB_TX("tx_multicast", mib.tx.mca),
+	STAT_MIB_TX("tx_broadcast", mib.tx.bca),
+	STAT_MIB_TX("tx_pause", mib.tx.pf),
+	STAT_MIB_TX("tx_control", mib.tx.cf),
+	STAT_MIB_TX("tx_fcs_err", mib.tx.fcs),
+	STAT_MIB_TX("tx_oversize", mib.tx.ovr),
+	STAT_MIB_TX("tx_defer", mib.tx.drf),
+	STAT_MIB_TX("tx_excess_defer", mib.tx.edf),
+	STAT_MIB_TX("tx_single_col", mib.tx.scl),
+	STAT_MIB_TX("tx_multi_col", mib.tx.mcl),
+	STAT_MIB_TX("tx_late_col", mib.tx.lcl),
+	STAT_MIB_TX("tx_excess_col", mib.tx.ecl),
+	STAT_MIB_TX("tx_frags", mib.tx.frg),
+	STAT_MIB_TX("tx_total_col", mib.tx.ncl),
+	STAT_MIB_TX("tx_jabber", mib.tx.jbr),
+	STAT_MIB_TX("tx_bytes", mib.tx.bytes),
+	STAT_MIB_TX("tx_good_pkts", mib.tx.pok),
+	STAT_MIB_TX("tx_unicast", mib.tx.uc),
+	/* UniMAC RUNT counters */
+	STAT_RUNT("rx_runt_pkts", mib.rx_runt_cnt),
+	STAT_RUNT("rx_runt_valid_fcs", mib.rx_runt_fcs),
+	STAT_RUNT("rx_runt_inval_fcs_align", mib.rx_runt_fcs_align),
+	STAT_RUNT("rx_runt_bytes", mib.rx_runt_bytes),
+	/* RXCHK misc statistics */
+	STAT_RXCHK("rxchk_bad_csum", mib.rxchk_bad_csum, RXCHK_BAD_CSUM_CNTR),
+	STAT_RXCHK("rxchk_other_pkt_disc", mib.rxchk_other_pkt_disc,
+			RXCHK_OTHER_DISC_CNTR),
+	/* RBUF misc statistics */
+	STAT_RBUF("rbuf_ovflow_cnt", mib.rbuf_ovflow_cnt, RBUF_OVFL_DISC_CNTR),
+	STAT_RBUF("rbuf_err_cnt", mib.rbuf_err_cnt, RBUF_ERR_PKT_CNTR),
+};
+
+#define BCM_SYSPORT_STATS_LEN	ARRAY_SIZE(bcm_sysport_gstrings_stats)
+
+static void bcm_sysport_get_drvinfo(struct net_device *dev,
+					struct ethtool_drvinfo *info)
+{
+	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strlcpy(info->version, "0.1", sizeof(info->version));
+	strlcpy(info->bus_info, "platform", sizeof(info->bus_info));
+	info->n_stats = BCM_SYSPORT_STATS_LEN;
+}
+
+static u32 bcm_sysport_get_msglvl(struct net_device *dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+
+	return priv->msg_enable;
+}
+
+static void bcm_sysport_set_msglvl(struct net_device *dev, u32 enable)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+
+	priv->msg_enable = enable;
+}
+
+static int bcm_sysport_get_sset_count(struct net_device *dev, int string_set)
+{
+	switch (string_set) {
+	case ETH_SS_STATS:
+		return BCM_SYSPORT_STATS_LEN;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void bcm_sysport_get_strings(struct net_device *dev,
+					u32 stringset, u8 *data)
+{
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < BCM_SYSPORT_STATS_LEN; i++) {
+			memcpy(data + i * ETH_GSTRING_LEN,
+				bcm_sysport_gstrings_stats[i].stat_string,
+				ETH_GSTRING_LEN);
+		}
+		break;
+	default:
+		break;
+	}
+}
+
+static void bcm_sysport_update_mib_counters(struct bcm_sysport_priv *priv)
+{
+	int i, j = 0;
+
+	for (i = 0; i < BCM_SYSPORT_STATS_LEN; i++) {
+		const struct bcm_sysport_stats *s;
+		u8 offset = 0;
+		u32 val = 0;
+		char *p;
+
+		s = &bcm_sysport_gstrings_stats[i];
+		switch (s->type) {
+		case BCM_SYSPORT_STAT_NETDEV:
+			continue;
+		case BCM_SYSPORT_STAT_MIB_RX:
+		case BCM_SYSPORT_STAT_MIB_TX:
+		case BCM_SYSPORT_STAT_RUNT:
+			if (s->type != BCM_SYSPORT_STAT_MIB_RX)
+				offset = UMAC_MIB_STAT_OFFSET;
+			val = umac_readl(priv, UMAC_MIB_START + j + offset);
+			break;
+		case BCM_SYSPORT_STAT_RXCHK:
+			val = rxchk_readl(priv, s->reg_offset);
+			if (val == ~0)
+				rxchk_writel(priv, 0, s->reg_offset);
+			break;
+		case BCM_SYSPORT_STAT_RBUF:
+			val = rbuf_readl(priv, s->reg_offset);
+			if (val == ~0)
+				rbuf_writel(priv, 0, s->reg_offset);
+			break;
+		}
+
+		j += s->stat_sizeof;
+		p = (char *)priv + s->stat_offset;
+		*(u32 *)p = val;
+	}
+
+	netif_dbg(priv, hw, priv->netdev, "updated MIB counters\n");
+}
+
+static void bcm_sysport_get_stats(struct net_device *dev,
+					struct ethtool_stats *stats, u64 *data)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	int i;
+
+	if (netif_running(dev))
+		bcm_sysport_update_mib_counters(priv);
+
+	for (i =  0; i < BCM_SYSPORT_STATS_LEN; i++) {
+		const struct bcm_sysport_stats *s;
+		char *p;
+
+		s = &bcm_sysport_gstrings_stats[i];
+		if (s->type == BCM_SYSPORT_STAT_NETDEV)
+			p = (char *)&dev->stats;
+		else
+			p = (char *)priv;
+		p += s->stat_offset;
+		data[i] = *(u32 *)p;
+	}
+}
+
+static void bcm_sysport_free_cb(struct bcm_sysport_cb *cb)
+{
+	dev_kfree_skb_any(cb->skb);
+	cb->skb = NULL;
+	dma_unmap_addr_set(cb, dma_addr, 0);
+}
+
+static int bcm_sysport_rx_refill(struct bcm_sysport_priv *priv,
+				 struct bcm_sysport_cb *cb)
+{
+	struct device *kdev = &priv->pdev->dev;
+	struct net_device *ndev = priv->netdev;
+	dma_addr_t mapping;
+	int ret;
+
+	cb->skb = netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH);
+	if (!cb->skb) {
+		netif_err(priv, rx_err, ndev, "SKB alloc failed\n");
+		return -ENOMEM;
+	}
+
+	mapping = dma_map_single(kdev, cb->skb->data,
+				RX_BUF_LENGTH, DMA_FROM_DEVICE);
+	ret = dma_mapping_error(kdev, mapping);
+	if (ret) {
+		bcm_sysport_free_cb(cb);
+		netif_err(priv, rx_err, ndev, "DMA mapping failure\n");
+		return ret;
+	}
+
+	dma_unmap_addr_set(cb, dma_addr, mapping);
+	dma_desc_set_addr(priv, priv->rx_bd_assign_ptr, mapping);
+
+	priv->rx_bd_assign_index++;
+	priv->rx_bd_assign_index &= (priv->num_rx_bds - 1);
+	priv->rx_bd_assign_ptr = priv->rx_bds +
+		(priv->rx_bd_assign_index * DESC_SIZE);
+
+	netif_dbg(priv, rx_status, ndev, "RX refill\n");
+
+	return 0;
+}
+
+static int bcm_sysport_alloc_rx_bufs(struct bcm_sysport_priv *priv)
+{
+	struct bcm_sysport_cb *cb;
+	int ret = 0;
+	unsigned int i;
+
+	for (i = 0; i < priv->num_rx_bds; i++) {
+		cb = &priv->rx_cbs[priv->rx_bd_assign_index];
+		if (cb->skb)
+			continue;
+
+		ret = bcm_sysport_rx_refill(priv, cb);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+/* Poll the hardware for up to budget packets to process */
+static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv,
+					unsigned int budget)
+{
+	struct device *kdev = &priv->pdev->dev;
+	struct net_device *ndev = priv->netdev;
+	unsigned int processed = 0, to_process;
+	struct bcm_sysport_cb *cb;
+	struct sk_buff *skb;
+	unsigned int p_index;
+	u16 len, status;
+	struct bcm_rsb *rsb;
+
+	/* Determine how much we should process since last call */
+	p_index = rdma_readl(priv, RDMA_PROD_INDEX);
+	p_index &= RDMA_PROD_INDEX_MASK;
+
+	if (p_index < priv->rx_c_index)
+		to_process = (RDMA_CONS_INDEX_MASK + 1) -
+			priv->rx_c_index + p_index;
+	else
+		to_process = p_index - priv->rx_c_index;
+
+	netif_dbg(priv, rx_status, ndev,
+			"p_index=%d rx_c_index=%d to_process=%d\n",
+			p_index, priv->rx_c_index, to_process);
+
+	while ((processed < to_process) &&
+		(processed < budget)) {
+
+		cb = &priv->rx_cbs[priv->rx_read_ptr];
+		skb = cb->skb;
+		dma_unmap_single(kdev, dma_unmap_addr(cb, dma_addr),
+				RX_BUF_LENGTH, DMA_FROM_DEVICE);
+
+		/* Extract the Receive Status Block prepended */
+		rsb = (struct bcm_rsb *)skb->data;
+		len = (rsb->rx_status_len >> DESC_LEN_SHIFT) & DESC_LEN_MASK;
+		status = (rsb->rx_status_len >> DESC_STATUS_SHIFT) &
+			DESC_STATUS_MASK;
+
+		processed++;
+		priv->rx_read_ptr++;
+		if (priv->rx_read_ptr == priv->num_rx_bds)
+			priv->rx_read_ptr = 0;
+
+		netif_dbg(priv, rx_status, ndev,
+				"p=%d, c=%d, rd_ptr=%d, len=%d, flag=0x%04x\n",
+				p_index, priv->rx_c_index, priv->rx_read_ptr,
+				len, status);
+
+		if (unlikely(!skb)) {
+			netif_err(priv, rx_err, ndev, "out of memory!\n");
+			ndev->stats.rx_dropped++;
+			ndev->stats.rx_errors++;
+			goto refill;
+		}
+
+		if (unlikely(!(status & DESC_EOP) || !(status & DESC_SOP))) {
+			netif_err(priv, rx_status, ndev, "fragmented packet!\n");
+			ndev->stats.rx_dropped++;
+			ndev->stats.rx_errors++;
+			bcm_sysport_free_cb(cb);
+			goto refill;
+		}
+
+		if (unlikely(status & (RX_STATUS_ERR | RX_STATUS_OVFLOW))) {
+			netif_err(priv, rx_err, ndev, "error packet\n");
+			if (status & RX_STATUS_OVFLOW)
+				ndev->stats.rx_over_errors++;
+			ndev->stats.rx_dropped++;
+			ndev->stats.rx_errors++;
+			bcm_sysport_free_cb(cb);
+			goto refill;
+		}
+
+		skb_put(skb, len);
+
+		/* Hardware validated our checksum */
+		if (likely(status & DESC_L4_CSUM))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		/* Hardware pre-pends packets with 2bytes before Ethernet
+		 * header plus we have the Receive Status Block, strip off all
+		 * of this from the SKB.
+		 */
+		skb_pull(skb, sizeof(*rsb) + 2);
+		len -= (sizeof(*rsb) + 2);
+
+		/* UniMAC may forward CRC */
+		if (priv->crc_fwd) {
+			skb_trim(skb, len - ETH_FCS_LEN);
+			len -= ETH_FCS_LEN;
+		}
+
+		skb->protocol = eth_type_trans(skb, ndev);
+		ndev->stats.rx_packets++;
+		ndev->stats.rx_bytes += len;
+
+		napi_gro_receive(&priv->napi, skb);
+refill:
+		bcm_sysport_rx_refill(priv, cb);
+	}
+
+	return processed;
+}
+
+static void bcm_sysport_tx_reclaim_one(struct bcm_sysport_priv *priv,
+					struct bcm_sysport_cb *cb,
+					unsigned int *bytes_compl,
+					unsigned int *pkts_compl)
+{
+	struct device *kdev = &priv->pdev->dev;
+	struct net_device *ndev = priv->netdev;
+
+	if (cb->skb) {
+		ndev->stats.tx_bytes += cb->skb->len;
+		*bytes_compl += cb->skb->len;
+		dma_unmap_single(kdev, dma_unmap_addr(cb, dma_addr),
+				dma_unmap_len(cb, dma_len),
+				DMA_TO_DEVICE);
+		ndev->stats.tx_packets++;
+		(*pkts_compl)++;
+		bcm_sysport_free_cb(cb);
+	/* SKB fragment */
+	} else if (dma_unmap_addr(cb, dma_addr)) {
+		ndev->stats.tx_bytes += dma_unmap_len(cb, dma_len);
+		dma_unmap_page(kdev, dma_unmap_addr(cb, dma_addr),
+				dma_unmap_len(cb, dma_len), DMA_TO_DEVICE);
+		dma_unmap_addr_set(cb, dma_addr, 0);
+	}
+}
+
+/* Reclaim queued SKBs for transmission completion, lockless version */
+static unsigned int __bcm_sysport_tx_reclaim(struct bcm_sysport_priv *priv,
+					     struct bcm_sysport_tx_ring *ring)
+{
+	struct net_device *ndev = priv->netdev;
+	unsigned int c_index, last_c_index, last_tx_cn, num_tx_cbs;
+	unsigned int pkts_compl = 0, bytes_compl = 0;
+	struct bcm_sysport_cb *cb;
+	struct netdev_queue *txq;
+	u32 hw_ind;
+
+	txq = netdev_get_tx_queue(ndev, ring->index);
+
+	/* Compute how many descriptors have been processed since last call */
+	hw_ind = tdma_readl(priv, TDMA_DESC_RING_PROD_CONS_INDEX(ring->index));
+	c_index = (hw_ind >> RING_CONS_INDEX_SHIFT) & RING_CONS_INDEX_MASK;
+	ring->p_index = (hw_ind & RING_PROD_INDEX_MASK);
+
+	last_c_index = ring->c_index;
+	num_tx_cbs = ring->size;
+
+	c_index &= (num_tx_cbs - 1);
+
+	if (c_index >= last_c_index)
+		last_tx_cn = c_index - last_c_index;
+	else
+		last_tx_cn = num_tx_cbs - last_c_index + c_index;
+
+	netif_dbg(priv, tx_done, ndev,
+			"ring=%d c_index=%d last_tx_cn=%d last_c_index=%d\n",
+			ring->index, c_index, last_tx_cn, last_c_index);
+
+	while (last_tx_cn-- > 0) {
+		cb = ring->cbs + last_c_index;
+		bcm_sysport_tx_reclaim_one(priv, cb, &bytes_compl, &pkts_compl);
+
+		ring->desc_count++;
+		last_c_index++;
+		last_c_index &= (num_tx_cbs - 1);
+	}
+
+	ring->c_index = c_index;
+
+	if (netif_tx_queue_stopped(txq) && pkts_compl)
+		netif_tx_wake_queue(txq);
+
+	netif_dbg(priv, tx_done, ndev,
+			"ring=%d c_index=%d pkts_compl=%d, bytes_compl=%d\n",
+			ring->index, ring->c_index, pkts_compl, bytes_compl);
+
+	return pkts_compl;
+}
+
+/* Locked version of the per-ring TX reclaim routine */
+static unsigned int bcm_sysport_tx_reclaim(struct bcm_sysport_priv *priv,
+					   struct bcm_sysport_tx_ring *ring)
+{
+	unsigned int released;
+	unsigned long flags;
+
+	spin_lock_irqsave(&ring->lock, flags);
+	released = __bcm_sysport_tx_reclaim(priv, ring);
+	spin_unlock_irqrestore(&ring->lock, flags);
+
+	return released;
+}
+
+static int bcm_sysport_tx_poll(struct napi_struct *napi, int budget)
+{
+	struct bcm_sysport_tx_ring *ring =
+		container_of(napi, struct bcm_sysport_tx_ring, napi);
+	unsigned int work_done = 0;
+
+	work_done = bcm_sysport_tx_reclaim(ring->priv, ring);
+
+	if (work_done < budget) {
+		napi_complete(napi);
+		/* re-enable TX interrupt */
+		intrl2_1_mask_clear(ring->priv, BIT(ring->index));
+	}
+
+	return work_done;
+}
+
+static void bcm_sysport_tx_reclaim_all(struct bcm_sysport_priv *priv)
+{
+	unsigned int q;
+
+	for (q = 0; q < priv->netdev->num_tx_queues; q++)
+		bcm_sysport_tx_reclaim(priv, &priv->tx_rings[q]);
+}
+
+static int bcm_sysport_poll(struct napi_struct *napi, int budget)
+{
+	struct bcm_sysport_priv *priv =
+		container_of(napi, struct bcm_sysport_priv, napi);
+	unsigned int work_done = 0;
+
+	work_done = bcm_sysport_desc_rx(priv, budget);
+
+	priv->rx_c_index += work_done;
+	priv->rx_c_index &= RDMA_CONS_INDEX_MASK;
+	rdma_writel(priv, priv->rx_c_index, RDMA_CONS_INDEX);
+
+	if (work_done < budget) {
+		napi_complete(napi);
+		/* re-enable RX interrupts */
+		intrl2_0_mask_clear(priv, INTRL2_0_RDMA_MBDONE);
+	}
+
+	return work_done;
+}
+
+
+/* RX and misc interrupt routine */
+static irqreturn_t bcm_sysport_rx_isr(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+
+	priv->irq0_stat = intrl2_0_readl(priv, INTRL2_CPU_STATUS) &
+			  ~intrl2_0_readl(priv, INTRL2_CPU_MASK_STATUS);
+	intrl2_0_writel(priv, priv->irq0_stat, INTRL2_CPU_CLEAR);
+
+	if (unlikely(priv->irq0_stat == 0)) {
+		netdev_warn(priv->netdev, "spurious RX interrupt\n");
+		return IRQ_NONE;
+	}
+
+	if (priv->irq0_stat & INTRL2_0_RDMA_MBDONE) {
+		if (likely(napi_schedule_prep(&priv->napi))) {
+			/* disable RX interrupts */
+			intrl2_0_mask_set(priv, INTRL2_0_RDMA_MBDONE);
+			__napi_schedule(&priv->napi);
+		}
+	}
+
+	/* TX ring is full, perform a full reclaim since we do not know
+	 * which one would trigger this interrupt
+	 */
+	if (priv->irq0_stat & INTRL2_0_TX_RING_FULL)
+		bcm_sysport_tx_reclaim_all(priv);
+
+	return IRQ_HANDLED;
+}
+
+/* TX interrupt service routine */
+static irqreturn_t bcm_sysport_tx_isr(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct bcm_sysport_tx_ring *txr;
+	unsigned int ring;
+
+	priv->irq1_stat = intrl2_1_readl(priv, INTRL2_CPU_STATUS) &
+				~intrl2_1_readl(priv, INTRL2_CPU_MASK_STATUS);
+	intrl2_1_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
+
+	if (unlikely(priv->irq1_stat == 0)) {
+		netdev_warn(priv->netdev, "spurious TX interrupt\n");
+		return IRQ_NONE;
+	}
+
+	for (ring = 0; ring < dev->num_tx_queues; ring++) {
+		if (!(priv->irq1_stat & BIT(ring)))
+			continue;
+
+		txr = &priv->tx_rings[ring];
+
+		if (likely(napi_schedule_prep(&txr->napi))) {
+			intrl2_1_mask_set(priv, BIT(ring));
+			__napi_schedule(&txr->napi);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int bcm_sysport_insert_tsb(struct sk_buff *skb, struct net_device *dev)
+{
+	struct sk_buff *nskb;
+	struct bcm_tsb *tsb;
+	u32 csum_info;
+	u8 ip_proto;
+	u16 csum_start;
+	u16 ip_ver;
+
+	/* Re-allocate SKB if needed */
+	if (unlikely(skb_headroom(skb) < sizeof(*tsb))) {
+		nskb = skb_realloc_headroom(skb, sizeof(*tsb));
+		dev_kfree_skb(skb);
+		if (!nskb) {
+			dev->stats.tx_errors++;
+			dev->stats.tx_dropped++;
+			return -ENOMEM;
+		}
+		skb = nskb;
+	}
+
+	tsb = (struct bcm_tsb *)skb_push(skb, sizeof(*tsb));
+	/* Zero-out TSB by default */
+	memset(tsb, 0, sizeof(*tsb));
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		ip_ver = htons(skb->protocol);
+		switch (ip_ver) {
+		case ETH_P_IP:
+			ip_proto = ip_hdr(skb)->protocol;
+			break;
+		case ETH_P_IPV6:
+			ip_proto = ipv6_hdr(skb)->nexthdr;
+			break;
+		default:
+			return 0;
+		}
+
+		/* Get the checksum offset and the L4 (transport) offset */
+		csum_start = skb_checksum_start_offset(skb) - sizeof(*tsb);
+		csum_info = (csum_start + skb->csum_offset) & L4_CSUM_PTR_MASK;
+		csum_info |= (csum_start << L4_PTR_SHIFT);
+
+		if (ip_proto == IPPROTO_TCP || ip_proto == IPPROTO_UDP) {
+			csum_info |= L4_LENGTH_VALID;
+			if (ip_proto == IPPROTO_UDP && ip_ver == ETH_P_IP)
+				csum_info |= L4_UDP;
+		} else
+			csum_info = 0;
+
+		tsb->l4_ptr_dest_map = csum_info;
+	}
+
+	return 0;
+}
+
+static netdev_tx_t bcm_sysport_xmit(struct sk_buff *skb,
+				    struct net_device *dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct device *kdev = &priv->pdev->dev;
+	struct bcm_sysport_tx_ring *ring;
+	struct bcm_sysport_cb *cb;
+	struct netdev_queue *txq;
+	struct dma_desc *desc;
+	unsigned int skb_len;
+	unsigned long flags;
+	dma_addr_t mapping;
+	u32 len_status;
+	u16 queue;
+	int ret;
+
+	queue = skb_get_queue_mapping(skb);
+	txq = netdev_get_tx_queue(dev, queue);
+	ring = &priv->tx_rings[queue];
+
+	/* lock against tx reclaim in BH context and TX ring full interrupt */
+	spin_lock_irqsave(&ring->lock, flags);
+	if (unlikely(ring->desc_count == 0)) {
+		netif_tx_stop_queue(txq);
+		netdev_err(dev, "queue %d awake and ring full!\n", queue);
+		ret = NETDEV_TX_BUSY;
+		goto out;
+	}
+
+	/* Insert TSB and checksum infos */
+	if (priv->tsb_en) {
+		ret = bcm_sysport_insert_tsb(skb, dev);
+		if (ret) {
+			ret = NETDEV_TX_OK;
+			goto out;
+		}
+	}
+
+	/* The Ethernet switch we are interfaced with needs packets to be at
+	 * least 64 bytes (including FCS) otherwise they will be discarded when
+	 * they enter the switch port logic. When Broadcom tags are enabled, we
+	 * need to make sure that packets are at least 68 bytes
+	 * (including FCS and tag) because the length verification is done after
+	 * the Broadcom tag is stripped off the ingress packet.
+	 */
+	if (skb_padto(skb, ETH_ZLEN + ENET_BRCM_TAG_LEN)) {
+		ret = NETDEV_TX_OK;
+		goto out;
+	}
+
+	skb_len = skb->len < ETH_ZLEN + ENET_BRCM_TAG_LEN ?
+			ETH_ZLEN + ENET_BRCM_TAG_LEN : skb->len;
+
+	mapping = dma_map_single(kdev, skb->data, skb_len, DMA_TO_DEVICE);
+	if (dma_mapping_error(kdev, mapping)) {
+		netif_err(priv, tx_err, dev, "DMA map failed at %p (len=%d)\n",
+				skb->data, skb_len);
+		ret = NETDEV_TX_OK;
+		goto out;
+	}
+
+	/* Remember the SKB for future freeing */
+	cb = &ring->cbs[ring->curr_desc];
+	cb->skb = skb;
+	dma_unmap_addr_set(cb, dma_addr, mapping);
+	dma_unmap_len_set(cb, dma_len, skb_len);
+
+	/* Fetch a descriptor entry from our pool */
+	desc = ring->desc_cpu;
+
+	desc->addr_lo = lower_32_bits(mapping);
+	len_status = upper_32_bits(mapping) & DESC_ADDR_HI_MASK;
+	len_status |= (skb_len << DESC_LEN_SHIFT);
+	len_status |= (DESC_SOP | DESC_EOP | TX_STATUS_APP_CRC) <<
+			DESC_STATUS_SHIFT;
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		len_status |= (DESC_L4_CSUM << DESC_STATUS_SHIFT);
+
+	ring->curr_desc++;
+	if (ring->curr_desc == ring->size)
+		ring->curr_desc = 0;
+	ring->desc_count--;
+
+	/* Ensure write completion of the descriptor status/length
+	 * in DRAM before the System Port WRITE_PORT register latches
+	 * the value
+	 */
+	wmb();
+	desc->addr_status_len = len_status;
+	wmb();
+
+	/* Write this descriptor address to the RING write port */
+	tdma_port_write_desc_addr(priv, desc, ring->index);
+
+	/* Check ring space and update SW control flow */
+	if (ring->desc_count == 0)
+		netif_tx_stop_queue(txq);
+
+	netif_dbg(priv, tx_queued, dev, "ring=%d desc_count=%d, curr_desc=%d\n",
+			ring->index, ring->desc_count, ring->curr_desc);
+
+	ret = NETDEV_TX_OK;
+out:
+	spin_unlock_irqrestore(&ring->lock, flags);
+	return ret;
+}
+
+static void bcm_sysport_tx_timeout(struct net_device *dev)
+{
+	netdev_warn(dev, "transmit timeout!\n");
+
+	dev->trans_start = jiffies;
+	dev->stats.tx_errors++;
+
+	netif_tx_wake_all_queues(dev);
+}
+
+/* phylib adjust link callback */
+static void bcm_sysport_adj_link(struct net_device *dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	struct phy_device *phydev = priv->phydev;
+	unsigned int changed = 0;
+	u32 cmd_bits = 0, reg;
+
+	if (priv->old_link != phydev->link) {
+		changed = 1;
+		priv->old_link = phydev->link;
+	}
+
+	if (priv->old_duplex != phydev->duplex) {
+		changed = 1;
+		priv->old_duplex = phydev->duplex;
+	}
+
+	switch (phydev->speed) {
+	case SPEED_2500:
+		cmd_bits = CMD_SPEED_2500;
+		break;
+	case SPEED_1000:
+		cmd_bits = CMD_SPEED_1000;
+		break;
+	case SPEED_100:
+		cmd_bits = CMD_SPEED_100;
+		break;
+	case SPEED_10:
+		cmd_bits = CMD_SPEED_10;
+		break;
+	default:
+		break;
+	}
+	cmd_bits <<= CMD_SPEED_SHIFT;
+
+	if (phydev->duplex == DUPLEX_HALF)
+		cmd_bits |= CMD_HD_EN;
+
+	if (priv->old_pause != phydev->pause) {
+		changed = 1;
+		priv->old_pause = phydev->pause;
+	}
+
+	if (!phydev->pause)
+		cmd_bits |= CMD_RX_PAUSE_IGNORE | CMD_TX_PAUSE_IGNORE;
+
+	if (changed) {
+		reg = umac_readl(priv, UMAC_CMD);
+		reg &= ~((CMD_SPEED_MASK << CMD_SPEED_SHIFT) |
+			CMD_HD_EN | CMD_RX_PAUSE_IGNORE |
+			CMD_TX_PAUSE_IGNORE);
+		reg |= cmd_bits;
+		umac_writel(priv, reg, UMAC_CMD);
+
+		phy_print_status(priv->phydev);
+	}
+}
+
+static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv,
+				    unsigned int index)
+{
+	struct bcm_sysport_tx_ring *ring = &priv->tx_rings[index];
+	struct device *kdev = &priv->pdev->dev;
+	size_t size;
+	void *p;
+	u32 reg;
+
+	/* Simple descriptors partitioning for now */
+	size = 256;
+
+	/* We just need one DMA descriptor which is DMA-able, since writing to
+	 * the port will allocate a new descriptor in its internal linked-list
+	 */
+	p = dma_zalloc_coherent(kdev, 1, &ring->desc_dma, GFP_KERNEL);
+	if (!p) {
+		netif_err(priv, hw, priv->netdev, "DMA alloc failed\n");
+		return -ENOMEM;
+	}
+
+	ring->cbs = kzalloc(sizeof(struct bcm_sysport_cb) * size, GFP_KERNEL);
+	if (!ring->cbs) {
+		netif_err(priv, hw, priv->netdev, "CB allocation failed\n");
+		return -ENOMEM;
+	}
+
+	/* Initialize SW view of the ring */
+	spin_lock_init(&ring->lock);
+	ring->priv = priv;
+	netif_napi_add(priv->netdev, &ring->napi, bcm_sysport_tx_poll, 64);
+	ring->index = index;
+	ring->size = size;
+	ring->alloc_size = ring->size;
+	ring->desc_cpu = p;
+	ring->desc_count = ring->size;
+	ring->curr_desc = 0;
+
+	/* Initialize HW ring */
+	tdma_writel(priv, RING_EN, TDMA_DESC_RING_HEAD_TAIL_PTR(index));
+	tdma_writel(priv, 0, TDMA_DESC_RING_COUNT(index));
+	tdma_writel(priv, 1, TDMA_DESC_RING_INTR_CONTROL(index));
+	tdma_writel(priv, 0, TDMA_DESC_RING_PROD_CONS_INDEX(index));
+	tdma_writel(priv, RING_IGNORE_STATUS, TDMA_DESC_RING_MAPPING(index));
+	tdma_writel(priv, 0, TDMA_DESC_RING_PCP_DEI_VID(index));
+
+	/* Program the number of descriptors as MAX_THRESHOLD and half of
+	 * its size for the hysteresis trigger
+	 */
+	tdma_writel(priv, ring->size |
+			1 << RING_HYST_THRESH_SHIFT,
+			TDMA_DESC_RING_MAX_HYST(index));
+
+	/* Enable the ring queue in the arbiter */
+	reg = tdma_readl(priv, TDMA_TIER1_ARB_0_QUEUE_EN);
+	reg |= (1 << index);
+	tdma_writel(priv, reg, TDMA_TIER1_ARB_0_QUEUE_EN);
+
+	napi_enable(&ring->napi);
+
+	netif_dbg(priv, hw, priv->netdev,
+			"TDMA cfg, size=%d, desc_cpu=%p\n",
+			ring->size, ring->desc_cpu);
+
+	return 0;
+}
+
+static void bcm_sysport_fini_tx_ring(struct bcm_sysport_priv *priv,
+					unsigned int index)
+{
+	struct bcm_sysport_tx_ring *ring = &priv->tx_rings[index];
+	struct device *kdev = &priv->pdev->dev;
+	u32 reg;
+
+	/* Caller should stop the TDMA engine */
+	reg = tdma_readl(priv, TDMA_STATUS);
+	if (!(reg & TDMA_DISABLED))
+		netdev_warn(priv->netdev, "TDMA not stopped!\n");
+
+	napi_disable(&ring->napi);
+	netif_napi_del(&ring->napi);
+
+	bcm_sysport_tx_reclaim(priv, ring);
+
+	kfree(ring->cbs);
+	ring->cbs = NULL;
+
+	if (ring->desc_dma) {
+		dma_free_coherent(kdev, 1, ring->desc_cpu, ring->desc_dma);
+		ring->desc_dma = 0;
+	}
+	ring->size = 0;
+	ring->alloc_size = 0;
+
+	netif_dbg(priv, hw, priv->netdev, "TDMA fini done\n");
+}
+
+/* RDMA helper */
+static inline int rdma_enable_set(struct bcm_sysport_priv *priv,
+					unsigned int enable)
+{
+	unsigned int timeout = 1000;
+	u32 reg;
+
+	reg = rdma_readl(priv, RDMA_CONTROL);
+	if (enable)
+		reg |= RDMA_EN;
+	else
+		reg &= ~RDMA_EN;
+	rdma_writel(priv, reg, RDMA_CONTROL);
+
+	/* Poll for RMDA disabling completion */
+	do {
+		reg = rdma_readl(priv, RDMA_STATUS);
+		if (!!(reg & RDMA_DISABLED) == !enable)
+			return 0;
+		usleep_range(1000, 2000);
+	} while (timeout-- > 0);
+
+	netdev_err(priv->netdev, "timeout waiting for RDMA to finish\n");
+
+	return -ETIMEDOUT;
+}
+
+/* TDMA helper */
+static inline int tdma_enable_set(struct bcm_sysport_priv *priv,
+					unsigned int enable)
+{
+	unsigned int timeout = 1000;
+	u32 reg;
+
+	reg = tdma_readl(priv, TDMA_CONTROL);
+	if (enable)
+		reg |= TDMA_EN;
+	else
+		reg &= ~TDMA_EN;
+	tdma_writel(priv, reg, TDMA_CONTROL);
+
+	/* Poll for TMDA disabling completion */
+	do {
+		reg = tdma_readl(priv, TDMA_STATUS);
+		if (!!(reg & TDMA_DISABLED) == !enable)
+			return 0;
+
+		usleep_range(1000, 2000);
+	} while (timeout-- > 0);
+
+	netdev_err(priv->netdev, "timeout waiting for TDMA to finish\n");
+
+	return -ETIMEDOUT;
+}
+
+static int bcm_sysport_init_rx_ring(struct bcm_sysport_priv *priv)
+{
+	u32 reg;
+	int ret;
+
+	/* Initialize SW view of the RX ring */
+	priv->num_rx_bds = NUM_RX_DESC;
+	priv->rx_bds = priv->base + SYS_PORT_RDMA_OFFSET;
+	priv->rx_bd_assign_ptr = priv->rx_bds;
+	priv->rx_bd_assign_index = 0;
+	priv->rx_c_index = 0;
+	priv->rx_read_ptr = 0;
+	priv->rx_cbs = kzalloc(priv->num_rx_bds *
+				sizeof(struct bcm_sysport_cb), GFP_KERNEL);
+	if (!priv->rx_cbs) {
+		netif_err(priv, hw, priv->netdev, "CB allocation failed\n");
+		return -ENOMEM;
+	}
+
+	ret = bcm_sysport_alloc_rx_bufs(priv);
+	if (ret) {
+		netif_err(priv, hw, priv->netdev, "SKB allocation failed\n");
+		return ret;
+	}
+
+	/* Initialize HW, ensure RDMA is disabled */
+	reg = rdma_readl(priv, RDMA_STATUS);
+	if (!(reg & RDMA_DISABLED))
+		rdma_enable_set(priv, 0);
+
+	rdma_writel(priv, 0, RDMA_WRITE_PTR_LO);
+	rdma_writel(priv, 0, RDMA_WRITE_PTR_HI);
+	rdma_writel(priv, 0, RDMA_PROD_INDEX);
+	rdma_writel(priv, 0, RDMA_CONS_INDEX);
+	rdma_writel(priv, priv->num_rx_bds << RDMA_RING_SIZE_SHIFT |
+			  RX_BUF_LENGTH, RDMA_RING_BUF_SIZE);
+	/* Operate the queue in ring mode */
+	rdma_writel(priv, 0, RDMA_START_ADDR_HI);
+	rdma_writel(priv, 0, RDMA_START_ADDR_LO);
+	rdma_writel(priv, 0, RDMA_END_ADDR_HI);
+	rdma_writel(priv, NUM_HW_RX_DESC_WORDS - 1, RDMA_END_ADDR_LO);
+
+	rdma_writel(priv, 1, RDMA_MBDONE_INTR);
+
+	netif_dbg(priv, hw, priv->netdev,
+			"RDMA cfg, num_rx_bds=%d, rx_bds=%p\n",
+			priv->num_rx_bds, priv->rx_bds);
+
+	return 0;
+}
+
+static void bcm_sysport_fini_rx_ring(struct bcm_sysport_priv *priv)
+{
+	struct bcm_sysport_cb *cb;
+	unsigned int i;
+	u32 reg;
+
+	/* Caller should ensure RDMA is disabled */
+	reg = rdma_readl(priv, RDMA_STATUS);
+	if (!(reg & RDMA_DISABLED))
+		netdev_warn(priv->netdev, "RDMA not stopped!\n");
+
+	for (i = 0; i < priv->num_rx_bds; i++) {
+		cb = &priv->rx_cbs[i];
+		if (dma_unmap_addr(cb, dma_addr))
+			dma_unmap_single(&priv->pdev->dev,
+					dma_unmap_addr(cb, dma_addr),
+					RX_BUF_LENGTH, DMA_FROM_DEVICE);
+		bcm_sysport_free_cb(cb);
+	}
+
+	kfree(priv->rx_cbs);
+	priv->rx_cbs = NULL;
+
+	netif_dbg(priv, hw, priv->netdev, "RDMA fini done\n");
+}
+
+static void bcm_sysport_set_rx_mode(struct net_device *dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	u32 reg;
+
+	reg = umac_readl(priv, UMAC_CMD);
+	if (dev->flags & IFF_PROMISC)
+		reg |= CMD_PROMISC;
+	else
+		reg &= ~CMD_PROMISC;
+	umac_writel(priv, reg, UMAC_CMD);
+
+	/* No support for ALLMULTI */
+	if (dev->flags & IFF_ALLMULTI)
+		return;
+}
+
+static inline void umac_enable_set(struct bcm_sysport_priv *priv,
+					unsigned int enable)
+{
+	u32 reg;
+
+	reg = umac_readl(priv, UMAC_CMD);
+	if (enable)
+		reg |= CMD_RX_EN | CMD_TX_EN;
+	else
+		reg &= ~(CMD_RX_EN | CMD_TX_EN);
+	umac_writel(priv, reg, UMAC_CMD);
+
+	/* UniMAC stops on a packet boundary, wait for a full-sized packet
+	 * to be processed (1 msec).
+	 */
+	if (enable == 0)
+		usleep_range(1000, 2000);
+}
+
+static inline int umac_reset(struct bcm_sysport_priv *priv)
+{
+	unsigned int timeout = 0;
+	u32 reg;
+	int ret = 0;
+
+	umac_writel(priv, 0, UMAC_CMD);
+	while (timeout++ < 1000) {
+		reg = umac_readl(priv, UMAC_CMD);
+		if (!(reg & CMD_SW_RESET))
+			break;
+
+		udelay(1);
+	}
+
+	if (timeout == 1000) {
+		dev_err(&priv->pdev->dev,
+			"timeout waiting for MAC to come out of reset\n");
+		ret = -ETIMEDOUT;
+	}
+
+	return ret;
+}
+
+static void umac_set_hw_addr(struct bcm_sysport_priv *priv,
+				unsigned char *addr)
+{
+	umac_writel(priv, (addr[0] << 24) | (addr[1] << 16) |
+			(addr[2] << 8) | addr[3], UMAC_MAC0);
+	umac_writel(priv, (addr[4] << 8) | addr[5], UMAC_MAC1);
+}
+
+static void topctrl_flush(struct bcm_sysport_priv *priv)
+{
+	topctrl_writel(priv, RX_FLUSH, RX_FLUSH_CNTL);
+	topctrl_writel(priv, TX_FLUSH, TX_FLUSH_CNTL);
+	mdelay(1);
+	topctrl_writel(priv, 0, RX_FLUSH_CNTL);
+	topctrl_writel(priv, 0, TX_FLUSH_CNTL);
+}
+
+static int bcm_sysport_open(struct net_device *dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	unsigned int i;
+	u32 reg;
+	int ret;
+
+	/* Reset UniMAC */
+	ret = umac_reset(priv);
+	if (ret) {
+		netdev_err(dev, "UniMAC reset failed\n");
+		return ret;
+	}
+
+	/* Flush TX and RX FIFOs at TOPCTRL level */
+	topctrl_flush(priv);
+
+	/* Disable the UniMAC RX/TX */
+	umac_enable_set(priv, 0);
+
+	/* Enable RBUF 2bytes alignment and Receive Status Block */
+	reg = rbuf_readl(priv, RBUF_CONTROL);
+	reg |= RBUF_4B_ALGN | RBUF_RSB_EN;
+	rbuf_writel(priv, reg, RBUF_CONTROL);
+
+	/* Set maximum frame length */
+	umac_writel(priv, UMAC_MAX_MTU_SIZE, UMAC_MAX_FRAME_LEN);
+
+	/* Set MAC address */
+	umac_set_hw_addr(priv, dev->dev_addr);
+
+	/* Read CRC forward */
+	priv->crc_fwd = !!(umac_readl(priv, UMAC_CMD) & CMD_CRC_FWD);
+
+	priv->phydev = of_phy_connect(dev, priv->phy_dn, bcm_sysport_adj_link,
+					0, priv->phy_interface);
+	if (!priv->phydev) {
+		netdev_err(dev, "could not attach to PHY\n");
+		return -ENODEV;
+	}
+
+	/* Reset house keeping link status */
+	priv->old_duplex = -1;
+	priv->old_link = -1;
+	priv->old_pause = -1;
+
+	/* mask all interrupts and request them */
+	intrl2_0_writel(priv, 0xffffffff, INTRL2_CPU_MASK_SET);
+	intrl2_0_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
+	intrl2_0_writel(priv, 0, INTRL2_CPU_MASK_CLEAR);
+	intrl2_1_writel(priv, 0xffffffff, INTRL2_CPU_MASK_SET);
+	intrl2_1_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
+	intrl2_1_writel(priv, 0, INTRL2_CPU_MASK_CLEAR);
+
+	ret = request_irq(priv->irq0, bcm_sysport_rx_isr, 0, dev->name, dev);
+	if (ret) {
+		netdev_err(dev, "failed to request RX interrupt\n");
+		goto out_phy_disconnect;
+	}
+
+	ret = request_irq(priv->irq1, bcm_sysport_tx_isr, 0, dev->name, dev);
+	if (ret) {
+		netdev_err(dev, "failed to request TX interrupt\n");
+		goto out_free_irq0;
+	}
+
+	/* Initialize both hardware and software ring */
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		ret = bcm_sysport_init_tx_ring(priv, i);
+		if (ret) {
+			netdev_err(dev, "failed to initialize TX ring %d\n",
+					i);
+			goto out_free_tx_ring;
+		}
+	}
+
+	/* Initialize linked-list */
+	tdma_writel(priv, TDMA_LL_RAM_INIT_BUSY, TDMA_STATUS);
+
+	/* Initialize RX ring */
+	ret = bcm_sysport_init_rx_ring(priv);
+	if (ret) {
+		netdev_err(dev, "failed to initialize RX ring\n");
+		goto out_free_rx_ring;
+	}
+
+	/* Turn on RDMA */
+	ret = rdma_enable_set(priv, 1);
+	if (ret)
+		goto out_free_rx_ring;
+
+	/* Enable RX interrupt and TX ring full interrupt */
+	intrl2_0_mask_clear(priv, INTRL2_0_RDMA_MBDONE | INTRL2_0_TX_RING_FULL);
+
+	/* Turn on TDMA */
+	ret = tdma_enable_set(priv, 1);
+	if (ret)
+		goto out_clear_rx_int;
+
+	/* Enable NAPI */
+	napi_enable(&priv->napi);
+
+	/* Turn on UniMAC TX/RX */
+	umac_enable_set(priv, 1);
+
+	phy_start(priv->phydev);
+
+	/* Enable TX interrupts for the 32 TXQs */
+	intrl2_1_mask_clear(priv, 0xffffffff);
+
+	/* Last call before we start the real business */
+	netif_tx_start_all_queues(dev);
+
+	return 0;
+
+out_clear_rx_int:
+	intrl2_0_mask_set(priv, INTRL2_0_RDMA_MBDONE | INTRL2_0_TX_RING_FULL);
+out_free_rx_ring:
+	bcm_sysport_fini_rx_ring(priv);
+out_free_tx_ring:
+	for (i = 0; i < dev->num_tx_queues; i++)
+		bcm_sysport_fini_tx_ring(priv, i);
+	free_irq(priv->irq1, dev);
+out_free_irq0:
+	free_irq(priv->irq0, dev);
+out_phy_disconnect:
+	phy_disconnect(priv->phydev);
+	return ret;
+}
+
+static int bcm_sysport_stop(struct net_device *dev)
+{
+	struct bcm_sysport_priv *priv = netdev_priv(dev);
+	unsigned int i;
+	u32 reg;
+	int ret;
+
+	/* stop all software from updating hardware */
+	netif_tx_stop_all_queues(dev);
+	napi_disable(&priv->napi);
+	phy_stop(priv->phydev);
+
+	/* mask all interrupts */
+	intrl2_0_mask_set(priv, 0xffffffff);
+	intrl2_0_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
+	intrl2_1_mask_set(priv, 0xffffffff);
+	intrl2_1_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
+
+	/* Disable UniMAC RX */
+	reg = umac_readl(priv, UMAC_CMD);
+	reg &= ~CMD_RX_EN;
+	umac_writel(priv, reg, UMAC_CMD);
+
+	ret = tdma_enable_set(priv, 0);
+	if (ret) {
+		netdev_err(dev, "timeout disabling RDMA\n");
+		return ret;
+	}
+
+	/* Wait for a maximum packet size to be drained */
+	usleep_range(2000, 3000);
+
+	ret = rdma_enable_set(priv, 0);
+	if (ret) {
+		netdev_err(dev, "timeout disabling TDMA\n");
+		return ret;
+	}
+
+	/* Disable UniMAC TX */
+	reg = umac_readl(priv, UMAC_CMD);
+	reg &= ~CMD_TX_EN;
+	umac_writel(priv, reg, UMAC_CMD);
+
+	/* Free RX/TX rings SW structures */
+	for (i = 0; i < dev->num_tx_queues; i++)
+		bcm_sysport_fini_tx_ring(priv, i);
+	bcm_sysport_fini_rx_ring(priv);
+
+	free_irq(priv->irq0, dev);
+	free_irq(priv->irq1, dev);
+
+	/* Disconnect from PHY */
+	phy_disconnect(priv->phydev);
+
+	return 0;
+}
+
+static struct ethtool_ops bcm_sysport_ethtool_ops = {
+	.get_settings		= bcm_sysport_get_settings,
+	.set_settings		= bcm_sysport_set_settings,
+	.get_drvinfo		= bcm_sysport_get_drvinfo,
+	.get_msglevel		= bcm_sysport_get_msglvl,
+	.set_msglevel		= bcm_sysport_set_msglvl,
+	.get_link		= ethtool_op_get_link,
+	.get_strings		= bcm_sysport_get_strings,
+	.get_ethtool_stats	= bcm_sysport_get_stats,
+	.get_sset_count		= bcm_sysport_get_sset_count,
+};
+
+static const struct net_device_ops bcm_sysport_netdev_ops = {
+	.ndo_start_xmit		= bcm_sysport_xmit,
+	.ndo_tx_timeout		= bcm_sysport_tx_timeout,
+	.ndo_open		= bcm_sysport_open,
+	.ndo_stop		= bcm_sysport_stop,
+	.ndo_set_features	= bcm_sysport_set_features,
+	.ndo_set_rx_mode	= bcm_sysport_set_rx_mode,
+};
+
+#define REV_FMT	"v%2x.%02x"
+
+static int bcm_sysport_probe(struct platform_device *pdev)
+{
+	struct bcm_sysport_priv *priv;
+	struct device_node *dn;
+	struct net_device *dev;
+	const void *macaddr;
+	struct resource *r;
+	u32 txq, rxq;
+	int ret;
+
+	dn = pdev->dev.of_node;
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+	/* Read the Transmit/Receive Queue properties */
+	if (of_property_read_u32(dn, "systemport,num-txq", &txq))
+		txq = TDMA_NUM_RINGS;
+	if (of_property_read_u32(dn, "systemport,num-rxq", &rxq))
+		rxq = 1;
+
+	dev = alloc_etherdev_mqs(sizeof(*priv), txq, rxq);
+	if (!dev)
+		return -ENOMEM;
+
+	/* Initialize private members */
+	priv = netdev_priv(dev);
+
+	priv->irq0 = platform_get_irq(pdev, 0);
+	priv->irq1 = platform_get_irq(pdev, 1);
+	if (priv->irq0 <= 0 || priv->irq1 <= 0) {
+		dev_err(&pdev->dev, "invalid interrupts\n");
+		ret = -EINVAL;
+		goto err;
+	}
+
+	priv->base = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(priv->base)) {
+		ret = PTR_ERR(priv->base);
+		goto err;
+	}
+
+	priv->netdev = dev;
+	priv->pdev = pdev;
+
+	priv->phy_interface = of_get_phy_mode(dn);
+	/* Default to GMII interface mode */
+	if (priv->phy_interface < 0)
+		priv->phy_interface = PHY_INTERFACE_MODE_GMII;
+
+	/* In the case of a fixed PHY, the DT node associated
+	 * to the PHY is the Ethernet MAC DT node.
+	 */
+	if (of_phy_is_fixed_link(dn)) {
+		ret = of_phy_register_fixed_link(dn);
+		if (ret) {
+			dev_err(&pdev->dev, "failed to register fixed PHY\n");
+			goto err;
+		}
+
+		priv->phy_dn = dn;
+	}
+
+	/* Initialize netdevice members */
+	macaddr = of_get_mac_address(dn);
+	if (!macaddr || !is_valid_ether_addr(macaddr)) {
+		dev_warn(&pdev->dev, "using random Ethernet MAC\n");
+		random_ether_addr(dev->dev_addr);
+	} else {
+		ether_addr_copy(dev->dev_addr, macaddr);
+	}
+
+	SET_NETDEV_DEV(dev, &pdev->dev);
+	dev_set_drvdata(&pdev->dev, dev);
+	dev->ethtool_ops = &bcm_sysport_ethtool_ops;
+	dev->netdev_ops = &bcm_sysport_netdev_ops;
+	netif_napi_add(dev, &priv->napi, bcm_sysport_poll, 64);
+
+	/* HW supported features, none enabled by default */
+	dev->hw_features |= NETIF_F_RXCSUM | NETIF_F_HIGHDMA |
+				NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
+
+	/* Set the needed headroom once and for all */
+	BUILD_BUG_ON(sizeof(struct bcm_tsb) != 8);
+	dev->needed_headroom += sizeof(struct bcm_tsb);
+
+	/* We are interfaced to a switch which handles the multicast
+	 * filtering for us, so we do not support programming any
+	 * multicast hash table in this Ethernet MAC.
+	 */
+	dev->flags &= ~IFF_MULTICAST;
+
+	/* libphy will adjust the link state accordingly */
+	netif_carrier_off(dev);
+
+	ret = register_netdev(dev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register net_device\n");
+		goto err;
+	}
+
+	priv->rev = topctrl_readl(priv, REV_CNTL) & REV_MASK;
+	dev_info(&pdev->dev,
+		"Broadcom SYSTEMPORT" REV_FMT
+		" at 0x%p (irqs: %d, %d, TXQs: %d, RXQs: %d)\n",
+		(priv->rev >> 8) & 0xff, priv->rev & 0xff,
+		priv->base, priv->irq0, priv->irq1, txq, rxq);
+
+	return 0;
+err:
+	free_netdev(dev);
+	return ret;
+}
+
+static int bcm_sysport_remove(struct platform_device *pdev)
+{
+	struct net_device *dev = dev_get_drvdata(&pdev->dev);
+
+	/* Not much to do, ndo_close has been called
+	 * and we use managed allocations
+	 */
+	unregister_netdev(dev);
+	free_netdev(dev);
+	dev_set_drvdata(&pdev->dev, NULL);
+
+	return 0;
+}
+
+static const struct of_device_id bcm_sysport_of_match[] = {
+	{ .compatible = "brcm,systemport-v1.00" },
+	{ .compatible = "brcm,systemport" },
+	{ /* sentinel */ }
+};
+
+static struct platform_driver bcm_sysport_driver = {
+	.probe	= bcm_sysport_probe,
+	.remove	= bcm_sysport_remove,
+	.driver =  {
+		.name = "brcm-systemport",
+		.owner = THIS_MODULE,
+		.of_match_table = bcm_sysport_of_match,
+	},
+};
+module_platform_driver(bcm_sysport_driver);
+
+MODULE_AUTHOR("Broadcom Corporation");
+MODULE_DESCRIPTION("Broadcom System Port Ethernet MAC driver");
+MODULE_ALIAS("platform:brcm-systemport");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h
new file mode 100644
index 000000000000..281c08246037
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bcmsysport.h
@@ -0,0 +1,678 @@
+/*
+ * Broadcom BCM7xxx System Port Ethernet MAC driver
+ *
+ * Copyright (C) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __BCM_SYSPORT_H
+#define __BCM_SYSPORT_H
+
+#include <linux/if_vlan.h>
+
+/* Receive/transmit descriptor format */
+#define DESC_ADDR_HI_STATUS_LEN	0x00
+#define  DESC_ADDR_HI_SHIFT	0
+#define  DESC_ADDR_HI_MASK	0xff
+#define  DESC_STATUS_SHIFT	8
+#define  DESC_STATUS_MASK	0x3ff
+#define  DESC_LEN_SHIFT		18
+#define  DESC_LEN_MASK		0x7fff
+#define DESC_ADDR_LO		0x04
+
+/* HW supports 40-bit addressing hence the */
+#define DESC_SIZE		(WORDS_PER_DESC * sizeof(u32))
+
+/* Default RX buffer allocation size */
+#define RX_BUF_LENGTH		2048
+
+/* Body(1500) + EH_SIZE(14) + VLANTAG(4) + BRCMTAG(4) + FCS(4) = 1526.
+ * 1536 is multiple of 256 bytes
+ */
+#define ENET_BRCM_TAG_LEN	4
+#define ENET_PAD		10
+#define UMAC_MAX_MTU_SIZE	(ETH_DATA_LEN + ETH_HLEN + VLAN_HLEN + \
+				 ENET_BRCM_TAG_LEN + ETH_FCS_LEN + ENET_PAD)
+
+/* Transmit status block */
+struct bcm_tsb {
+	u32 pcp_dei_vid;
+#define PCP_DEI_MASK		0xf
+#define VID_SHIFT		4
+#define VID_MASK		0xfff
+	u32 l4_ptr_dest_map;
+#define L4_CSUM_PTR_MASK	0x1ff
+#define L4_PTR_SHIFT		9
+#define L4_PTR_MASK		0x1ff
+#define L4_UDP			(1 << 18)
+#define L4_LENGTH_VALID		(1 << 19)
+#define DEST_MAP_SHIFT		20
+#define DEST_MAP_MASK		0x1ff
+};
+
+/* Receive status block uses the same
+ * definitions as the DMA descriptor
+ */
+struct bcm_rsb {
+	u32 rx_status_len;
+	u32 brcm_egress_tag;
+};
+
+/* Common Receive/Transmit status bits */
+#define DESC_L4_CSUM		(1 << 7)
+#define DESC_SOP		(1 << 8)
+#define DESC_EOP		(1 << 9)
+
+/* Receive Status bits */
+#define RX_STATUS_UCAST			0
+#define RX_STATUS_BCAST			0x04
+#define RX_STATUS_MCAST			0x08
+#define RX_STATUS_L2_MCAST		0x0c
+#define RX_STATUS_ERR			(1 << 4)
+#define RX_STATUS_OVFLOW		(1 << 5)
+#define RX_STATUS_PARSE_FAIL		(1 << 6)
+
+/* Transmit Status bits */
+#define TX_STATUS_VLAN_NO_ACT		0x00
+#define TX_STATUS_VLAN_PCP_TSB		0x01
+#define TX_STATUS_VLAN_QUEUE		0x02
+#define TX_STATUS_VLAN_VID_TSB		0x03
+#define TX_STATUS_OWR_CRC		(1 << 2)
+#define TX_STATUS_APP_CRC		(1 << 3)
+#define TX_STATUS_BRCM_TAG_NO_ACT	0
+#define TX_STATUS_BRCM_TAG_ZERO		0x10
+#define TX_STATUS_BRCM_TAG_ONE_QUEUE	0x20
+#define TX_STATUS_BRCM_TAG_ONE_TSB	0x30
+#define TX_STATUS_SKIP_BYTES		(1 << 6)
+
+/* Specific register definitions */
+#define SYS_PORT_TOPCTRL_OFFSET		0
+#define REV_CNTL			0x00
+#define  REV_MASK			0xffff
+
+#define RX_FLUSH_CNTL			0x04
+#define  RX_FLUSH			(1 << 0)
+
+#define TX_FLUSH_CNTL			0x08
+#define  TX_FLUSH			(1 << 0)
+
+#define MISC_CNTL			0x0c
+#define  SYS_CLK_SEL			(1 << 0)
+#define  TDMA_EOP_SEL			(1 << 1)
+
+/* Level-2 Interrupt controller offsets and defines */
+#define SYS_PORT_INTRL2_0_OFFSET	0x200
+#define SYS_PORT_INTRL2_1_OFFSET	0x240
+#define INTRL2_CPU_STATUS		0x00
+#define INTRL2_CPU_SET			0x04
+#define INTRL2_CPU_CLEAR		0x08
+#define INTRL2_CPU_MASK_STATUS		0x0c
+#define INTRL2_CPU_MASK_SET		0x10
+#define INTRL2_CPU_MASK_CLEAR		0x14
+
+/* Level-2 instance 0 interrupt bits */
+#define INTRL2_0_GISB_ERR		(1 << 0)
+#define INTRL2_0_RBUF_OVFLOW		(1 << 1)
+#define INTRL2_0_TBUF_UNDFLOW		(1 << 2)
+#define INTRL2_0_MPD			(1 << 3)
+#define INTRL2_0_BRCM_MATCH_TAG		(1 << 4)
+#define INTRL2_0_RDMA_MBDONE		(1 << 5)
+#define INTRL2_0_OVER_MAX_THRESH	(1 << 6)
+#define INTRL2_0_BELOW_HYST_THRESH	(1 << 7)
+#define INTRL2_0_FREE_LIST_EMPTY	(1 << 8)
+#define INTRL2_0_TX_RING_FULL		(1 << 9)
+#define INTRL2_0_DESC_ALLOC_ERR		(1 << 10)
+#define INTRL2_0_UNEXP_PKTSIZE_ACK	(1 << 11)
+
+/* RXCHK offset and defines */
+#define SYS_PORT_RXCHK_OFFSET		0x300
+
+#define RXCHK_CONTROL			0x00
+#define  RXCHK_EN			(1 << 0)
+#define  RXCHK_SKIP_FCS			(1 << 1)
+#define  RXCHK_BAD_CSUM_DIS		(1 << 2)
+#define  RXCHK_BRCM_TAG_EN		(1 << 3)
+#define  RXCHK_BRCM_TAG_MATCH_SHIFT	4
+#define  RXCHK_BRCM_TAG_MATCH_MASK	0xff
+#define  RXCHK_PARSE_TNL		(1 << 12)
+#define  RXCHK_VIOL_EN			(1 << 13)
+#define  RXCHK_VIOL_DIS			(1 << 14)
+#define  RXCHK_INCOM_PKT		(1 << 15)
+#define  RXCHK_V6_DUPEXT_EN		(1 << 16)
+#define  RXCHK_V6_DUPEXT_DIS		(1 << 17)
+#define  RXCHK_ETHERTYPE_DIS		(1 << 18)
+#define  RXCHK_L2_HDR_DIS		(1 << 19)
+#define  RXCHK_L3_HDR_DIS		(1 << 20)
+#define  RXCHK_MAC_RX_ERR_DIS		(1 << 21)
+#define  RXCHK_PARSE_AUTH		(1 << 22)
+
+#define RXCHK_BRCM_TAG0			0x04
+#define RXCHK_BRCM_TAG(i)		((i) * RXCHK_BRCM_TAG0)
+#define RXCHK_BRCM_TAG0_MASK		0x24
+#define RXCHK_BRCM_TAG_MASK(i)		((i) * RXCHK_BRCM_TAG0_MASK)
+#define RXCHK_BRCM_TAG_MATCH_STATUS	0x44
+#define RXCHK_ETHERTYPE			0x48
+#define RXCHK_BAD_CSUM_CNTR		0x4C
+#define RXCHK_OTHER_DISC_CNTR		0x50
+
+/* TXCHCK offsets and defines */
+#define SYS_PORT_TXCHK_OFFSET		0x380
+#define TXCHK_PKT_RDY_THRESH		0x00
+
+/* Receive buffer offset and defines */
+#define SYS_PORT_RBUF_OFFSET		0x400
+
+#define RBUF_CONTROL			0x00
+#define  RBUF_RSB_EN			(1 << 0)
+#define  RBUF_4B_ALGN			(1 << 1)
+#define  RBUF_BRCM_TAG_STRIP		(1 << 2)
+#define  RBUF_BAD_PKT_DISC		(1 << 3)
+#define  RBUF_RESUME_THRESH_SHIFT	4
+#define  RBUF_RESUME_THRESH_MASK	0xff
+#define  RBUF_OK_TO_SEND_SHIFT		12
+#define  RBUF_OK_TO_SEND_MASK		0xff
+#define  RBUF_CRC_REPLACE		(1 << 20)
+#define  RBUF_OK_TO_SEND_MODE		(1 << 21)
+#define  RBUF_RSB_SWAP			(1 << 22)
+#define  RBUF_ACPI_EN			(1 << 23)
+
+#define RBUF_PKT_RDY_THRESH		0x04
+
+#define RBUF_STATUS			0x08
+#define  RBUF_WOL_MODE			(1 << 0)
+#define  RBUF_MPD			(1 << 1)
+#define  RBUF_ACPI			(1 << 2)
+
+#define RBUF_OVFL_DISC_CNTR		0x0c
+#define RBUF_ERR_PKT_CNTR		0x10
+
+/* Transmit buffer offset and defines */
+#define SYS_PORT_TBUF_OFFSET		0x600
+
+#define TBUF_CONTROL			0x00
+#define  TBUF_BP_EN			(1 << 0)
+#define  TBUF_MAX_PKT_THRESH_SHIFT	1
+#define  TBUF_MAX_PKT_THRESH_MASK	0x1f
+#define  TBUF_FULL_THRESH_SHIFT		8
+#define  TBUF_FULL_THRESH_MASK		0x1f
+
+/* UniMAC offset and defines */
+#define SYS_PORT_UMAC_OFFSET		0x800
+
+#define UMAC_CMD			0x008
+#define  CMD_TX_EN			(1 << 0)
+#define  CMD_RX_EN			(1 << 1)
+#define  CMD_SPEED_SHIFT		2
+#define  CMD_SPEED_10			0
+#define  CMD_SPEED_100			1
+#define  CMD_SPEED_1000			2
+#define  CMD_SPEED_2500			3
+#define  CMD_SPEED_MASK			3
+#define  CMD_PROMISC			(1 << 4)
+#define  CMD_PAD_EN			(1 << 5)
+#define  CMD_CRC_FWD			(1 << 6)
+#define  CMD_PAUSE_FWD			(1 << 7)
+#define  CMD_RX_PAUSE_IGNORE		(1 << 8)
+#define  CMD_TX_ADDR_INS		(1 << 9)
+#define  CMD_HD_EN			(1 << 10)
+#define  CMD_SW_RESET			(1 << 13)
+#define  CMD_LCL_LOOP_EN		(1 << 15)
+#define  CMD_AUTO_CONFIG		(1 << 22)
+#define  CMD_CNTL_FRM_EN		(1 << 23)
+#define  CMD_NO_LEN_CHK			(1 << 24)
+#define  CMD_RMT_LOOP_EN		(1 << 25)
+#define  CMD_PRBL_EN			(1 << 27)
+#define  CMD_TX_PAUSE_IGNORE		(1 << 28)
+#define  CMD_TX_RX_EN			(1 << 29)
+#define  CMD_RUNT_FILTER_DIS		(1 << 30)
+
+#define UMAC_MAC0			0x00c
+#define UMAC_MAC1			0x010
+#define UMAC_MAX_FRAME_LEN		0x014
+
+#define UMAC_TX_FLUSH			0x334
+
+#define UMAC_MIB_START			0x400
+
+/* There is a 0xC gap between the end of RX and beginning of TX stats and then
+ * between the end of TX stats and the beginning of the RX RUNT
+ */
+#define UMAC_MIB_STAT_OFFSET		0xc
+
+#define UMAC_MIB_CTRL			0x580
+#define  MIB_RX_CNT_RST			(1 << 0)
+#define  MIB_RUNT_CNT_RST		(1 << 1)
+#define  MIB_TX_CNT_RST			(1 << 2)
+#define UMAC_MDF_CTRL			0x650
+#define UMAC_MDF_ADDR			0x654
+
+/* Receive DMA offset and defines */
+#define SYS_PORT_RDMA_OFFSET		0x2000
+
+#define RDMA_CONTROL			0x1000
+#define  RDMA_EN			(1 << 0)
+#define  RDMA_RING_CFG			(1 << 1)
+#define  RDMA_DISC_EN			(1 << 2)
+#define  RDMA_BUF_DATA_OFFSET_SHIFT	4
+#define  RDMA_BUF_DATA_OFFSET_MASK	0x3ff
+
+#define RDMA_STATUS			0x1004
+#define  RDMA_DISABLED			(1 << 0)
+#define  RDMA_DESC_RAM_INIT_BUSY	(1 << 1)
+#define  RDMA_BP_STATUS			(1 << 2)
+
+#define RDMA_SCB_BURST_SIZE		0x1008
+
+#define RDMA_RING_BUF_SIZE		0x100c
+#define  RDMA_RING_SIZE_SHIFT		16
+
+#define RDMA_WRITE_PTR_HI		0x1010
+#define RDMA_WRITE_PTR_LO		0x1014
+#define RDMA_PROD_INDEX			0x1018
+#define  RDMA_PROD_INDEX_MASK		0xffff
+
+#define RDMA_CONS_INDEX			0x101c
+#define  RDMA_CONS_INDEX_MASK		0xffff
+
+#define RDMA_START_ADDR_HI		0x1020
+#define RDMA_START_ADDR_LO		0x1024
+#define RDMA_END_ADDR_HI		0x1028
+#define RDMA_END_ADDR_LO		0x102c
+
+#define RDMA_MBDONE_INTR		0x1030
+#define  RDMA_INTR_THRESH_MASK		0xff
+#define  RDMA_TIMEOUT_SHIFT		16
+#define  RDMA_TIMEOUT_MASK		0xffff
+
+#define RDMA_XON_XOFF_THRESH		0x1034
+#define  RDMA_XON_XOFF_THRESH_MASK	0xffff
+#define  RDMA_XOFF_THRESH_SHIFT		16
+
+#define RDMA_READ_PTR_HI		0x1038
+#define RDMA_READ_PTR_LO		0x103c
+
+#define RDMA_OVERRIDE			0x1040
+#define  RDMA_LE_MODE			(1 << 0)
+#define  RDMA_REG_MODE			(1 << 1)
+
+#define RDMA_TEST			0x1044
+#define  RDMA_TP_OUT_SEL		(1 << 0)
+#define  RDMA_MEM_SEL			(1 << 1)
+
+#define RDMA_DEBUG			0x1048
+
+/* Transmit DMA offset and defines */
+#define TDMA_NUM_RINGS			32	/* rings = queues */
+#define TDMA_PORT_SIZE			DESC_SIZE /* two 32-bits words */
+
+#define SYS_PORT_TDMA_OFFSET		0x4000
+#define TDMA_WRITE_PORT_OFFSET		0x0000
+#define TDMA_WRITE_PORT_HI(i)		(TDMA_WRITE_PORT_OFFSET + \
+					(i) * TDMA_PORT_SIZE)
+#define TDMA_WRITE_PORT_LO(i)		(TDMA_WRITE_PORT_OFFSET + \
+					sizeof(u32) + (i) * TDMA_PORT_SIZE)
+
+#define TDMA_READ_PORT_OFFSET		(TDMA_WRITE_PORT_OFFSET + \
+					(TDMA_NUM_RINGS * TDMA_PORT_SIZE))
+#define TDMA_READ_PORT_HI(i)		(TDMA_READ_PORT_OFFSET + \
+					(i) * TDMA_PORT_SIZE)
+#define TDMA_READ_PORT_LO(i)		(TDMA_READ_PORT_OFFSET + \
+					sizeof(u32) + (i) * TDMA_PORT_SIZE)
+
+#define TDMA_READ_PORT_CMD_OFFSET	(TDMA_READ_PORT_OFFSET + \
+					(TDMA_NUM_RINGS * TDMA_PORT_SIZE))
+#define TDMA_READ_PORT_CMD(i)		(TDMA_READ_PORT_CMD_OFFSET + \
+					(i) * sizeof(u32))
+
+#define TDMA_DESC_RING_00_BASE		(TDMA_READ_PORT_CMD_OFFSET + \
+					(TDMA_NUM_RINGS * sizeof(u32)))
+
+/* Register offsets and defines relatives to a specific ring number */
+#define RING_HEAD_TAIL_PTR		0x00
+#define  RING_HEAD_MASK			0x7ff
+#define  RING_TAIL_SHIFT		11
+#define  RING_TAIL_MASK			0x7ff
+#define  RING_FLUSH			(1 << 24)
+#define  RING_EN			(1 << 25)
+
+#define RING_COUNT			0x04
+#define  RING_COUNT_MASK		0x7ff
+#define  RING_BUFF_DONE_SHIFT		11
+#define  RING_BUFF_DONE_MASK		0x7ff
+
+#define RING_MAX_HYST			0x08
+#define  RING_MAX_THRESH_MASK		0x7ff
+#define  RING_HYST_THRESH_SHIFT		11
+#define  RING_HYST_THRESH_MASK		0x7ff
+
+#define RING_INTR_CONTROL		0x0c
+#define  RING_INTR_THRESH_MASK		0x7ff
+#define  RING_EMPTY_INTR_EN		(1 << 15)
+#define  RING_TIMEOUT_SHIFT		16
+#define  RING_TIMEOUT_MASK		0xffff
+
+#define RING_PROD_CONS_INDEX		0x10
+#define  RING_PROD_INDEX_MASK		0xffff
+#define  RING_CONS_INDEX_SHIFT		16
+#define  RING_CONS_INDEX_MASK		0xffff
+
+#define RING_MAPPING			0x14
+#define  RING_QID_MASK			0x3
+#define  RING_PORT_ID_SHIFT		3
+#define  RING_PORT_ID_MASK		0x7
+#define  RING_IGNORE_STATUS		(1 << 6)
+#define  RING_FAILOVER_EN		(1 << 7)
+#define  RING_CREDIT_SHIFT		8
+#define  RING_CREDIT_MASK		0xffff
+
+#define RING_PCP_DEI_VID		0x18
+#define  RING_VID_MASK			0x7ff
+#define  RING_DEI			(1 << 12)
+#define  RING_PCP_SHIFT			13
+#define  RING_PCP_MASK			0x7
+#define  RING_PKT_SIZE_ADJ_SHIFT	16
+#define  RING_PKT_SIZE_ADJ_MASK		0xf
+
+#define TDMA_DESC_RING_SIZE		28
+
+/* Defininition for a given TX ring base address */
+#define TDMA_DESC_RING_BASE(i)		(TDMA_DESC_RING_00_BASE + \
+					((i) * TDMA_DESC_RING_SIZE))
+
+/* Ring indexed register addreses */
+#define TDMA_DESC_RING_HEAD_TAIL_PTR(i)	(TDMA_DESC_RING_BASE(i) + \
+					RING_HEAD_TAIL_PTR)
+#define TDMA_DESC_RING_COUNT(i)		(TDMA_DESC_RING_BASE(i) + \
+					RING_COUNT)
+#define TDMA_DESC_RING_MAX_HYST(i)	(TDMA_DESC_RING_BASE(i) + \
+					RING_MAX_HYST)
+#define TDMA_DESC_RING_INTR_CONTROL(i)	(TDMA_DESC_RING_BASE(i) + \
+					RING_INTR_CONTROL)
+#define TDMA_DESC_RING_PROD_CONS_INDEX(i) \
+					(TDMA_DESC_RING_BASE(i) + \
+					RING_PROD_CONS_INDEX)
+#define TDMA_DESC_RING_MAPPING(i)	(TDMA_DESC_RING_BASE(i) + \
+					RING_MAPPING)
+#define TDMA_DESC_RING_PCP_DEI_VID(i)	(TDMA_DESC_RING_BASE(i) + \
+					RING_PCP_DEI_VID)
+
+#define TDMA_CONTROL			0x600
+#define  TDMA_EN			(1 << 0)
+#define  TSB_EN				(1 << 1)
+#define  TSB_SWAP			(1 << 2)
+#define  ACB_ALGO			(1 << 3)
+#define  BUF_DATA_OFFSET_SHIFT		4
+#define  BUF_DATA_OFFSET_MASK		0x3ff
+#define  VLAN_EN			(1 << 14)
+#define  SW_BRCM_TAG			(1 << 15)
+#define  WNC_KPT_SIZE_UPDATE		(1 << 16)
+#define  SYNC_PKT_SIZE			(1 << 17)
+#define  ACH_TXDONE_DELAY_SHIFT		18
+#define  ACH_TXDONE_DELAY_MASK		0xff
+
+#define TDMA_STATUS			0x604
+#define  TDMA_DISABLED			(1 << 0)
+#define  TDMA_LL_RAM_INIT_BUSY		(1 << 1)
+
+#define TDMA_SCB_BURST_SIZE		0x608
+#define TDMA_OVER_MAX_THRESH_STATUS	0x60c
+#define TDMA_OVER_HYST_THRESH_STATUS	0x610
+#define TDMA_TPID			0x614
+
+#define TDMA_FREE_LIST_HEAD_TAIL_PTR	0x618
+#define  TDMA_FREE_HEAD_MASK		0x7ff
+#define  TDMA_FREE_TAIL_SHIFT		11
+#define  TDMA_FREE_TAIL_MASK		0x7ff
+
+#define TDMA_FREE_LIST_COUNT		0x61c
+#define  TDMA_FREE_LIST_COUNT_MASK	0x7ff
+
+#define TDMA_TIER2_ARB_CTRL		0x620
+#define  TDMA_ARB_MODE_RR		0
+#define  TDMA_ARB_MODE_WEIGHT_RR	0x1
+#define  TDMA_ARB_MODE_STRICT		0x2
+#define  TDMA_ARB_MODE_DEFICIT_RR	0x3
+#define  TDMA_CREDIT_SHIFT		4
+#define  TDMA_CREDIT_MASK		0xffff
+
+#define TDMA_TIER1_ARB_0_CTRL		0x624
+#define  TDMA_ARB_EN			(1 << 0)
+
+#define TDMA_TIER1_ARB_0_QUEUE_EN	0x628
+#define TDMA_TIER1_ARB_1_CTRL		0x62c
+#define TDMA_TIER1_ARB_1_QUEUE_EN	0x630
+#define TDMA_TIER1_ARB_2_CTRL		0x634
+#define TDMA_TIER1_ARB_2_QUEUE_EN	0x638
+#define TDMA_TIER1_ARB_3_CTRL		0x63c
+#define TDMA_TIER1_ARB_3_QUEUE_EN	0x640
+
+#define TDMA_SCB_ENDIAN_OVERRIDE	0x644
+#define  TDMA_LE_MODE			(1 << 0)
+#define  TDMA_REG_MODE			(1 << 1)
+
+#define TDMA_TEST			0x648
+#define  TDMA_TP_OUT_SEL		(1 << 0)
+#define  TDMA_MEM_TM			(1 << 1)
+
+#define TDMA_DEBUG			0x64c
+
+/* Transmit/Receive descriptor */
+struct dma_desc {
+	u32	addr_status_len;
+	u32	addr_lo;
+};
+
+/* Number of Receive hardware descriptor words */
+#define NUM_HW_RX_DESC_WORDS		1024
+/* Real number of usable descriptors */
+#define NUM_RX_DESC			(NUM_HW_RX_DESC_WORDS / WORDS_PER_DESC)
+
+/* Internal linked-list RAM has up to 1536 entries */
+#define NUM_TX_DESC			1536
+
+#define WORDS_PER_DESC			(sizeof(struct dma_desc) / sizeof(u32))
+
+/* Rx/Tx common counter group.*/
+struct bcm_sysport_pkt_counters {
+	u32	cnt_64;		/* RO Received/Transmited 64 bytes packet */
+	u32	cnt_127;	/* RO Rx/Tx 127 bytes packet */
+	u32	cnt_255;	/* RO Rx/Tx 65-255 bytes packet */
+	u32	cnt_511;	/* RO Rx/Tx 256-511 bytes packet */
+	u32	cnt_1023;	/* RO Rx/Tx 512-1023 bytes packet */
+	u32	cnt_1518;	/* RO Rx/Tx 1024-1518 bytes packet */
+	u32	cnt_mgv;	/* RO Rx/Tx 1519-1522 good VLAN packet */
+	u32	cnt_2047;	/* RO Rx/Tx 1522-2047 bytes packet*/
+	u32	cnt_4095;	/* RO Rx/Tx 2048-4095 bytes packet*/
+	u32	cnt_9216;	/* RO Rx/Tx 4096-9216 bytes packet*/
+};
+
+/* RSV, Receive Status Vector */
+struct bcm_sysport_rx_counters {
+	struct  bcm_sysport_pkt_counters pkt_cnt;
+	u32	pkt;		/* RO (0x428) Received pkt count*/
+	u32	bytes;		/* RO Received byte count */
+	u32	mca;		/* RO # of Received multicast pkt */
+	u32	bca;		/* RO # of Receive broadcast pkt */
+	u32	fcs;		/* RO # of Received FCS error  */
+	u32	cf;		/* RO # of Received control frame pkt*/
+	u32	pf;		/* RO # of Received pause frame pkt */
+	u32	uo;		/* RO # of unknown op code pkt */
+	u32	aln;		/* RO # of alignment error count */
+	u32	flr;		/* RO # of frame length out of range count */
+	u32	cde;		/* RO # of code error pkt */
+	u32	fcr;		/* RO # of carrier sense error pkt */
+	u32	ovr;		/* RO # of oversize pkt*/
+	u32	jbr;		/* RO # of jabber count */
+	u32	mtue;		/* RO # of MTU error pkt*/
+	u32	pok;		/* RO # of Received good pkt */
+	u32	uc;		/* RO # of unicast pkt */
+	u32	ppp;		/* RO # of PPP pkt */
+	u32	rcrc;		/* RO (0x470),# of CRC match pkt */
+};
+
+/* TSV, Transmit Status Vector */
+struct bcm_sysport_tx_counters {
+	struct bcm_sysport_pkt_counters pkt_cnt;
+	u32	pkts;		/* RO (0x4a8) Transmited pkt */
+	u32	mca;		/* RO # of xmited multicast pkt */
+	u32	bca;		/* RO # of xmited broadcast pkt */
+	u32	pf;		/* RO # of xmited pause frame count */
+	u32	cf;		/* RO # of xmited control frame count */
+	u32	fcs;		/* RO # of xmited FCS error count */
+	u32	ovr;		/* RO # of xmited oversize pkt */
+	u32	drf;		/* RO # of xmited deferral pkt */
+	u32	edf;		/* RO # of xmited Excessive deferral pkt*/
+	u32	scl;		/* RO # of xmited single collision pkt */
+	u32	mcl;		/* RO # of xmited multiple collision pkt*/
+	u32	lcl;		/* RO # of xmited late collision pkt */
+	u32	ecl;		/* RO # of xmited excessive collision pkt*/
+	u32	frg;		/* RO # of xmited fragments pkt*/
+	u32	ncl;		/* RO # of xmited total collision count */
+	u32	jbr;		/* RO # of xmited jabber count*/
+	u32	bytes;		/* RO # of xmited byte count */
+	u32	pok;		/* RO # of xmited good pkt */
+	u32	uc;		/* RO (0x0x4f0)# of xmited unitcast pkt */
+};
+
+struct bcm_sysport_mib {
+	struct bcm_sysport_rx_counters rx;
+	struct bcm_sysport_tx_counters tx;
+	u32 rx_runt_cnt;
+	u32 rx_runt_fcs;
+	u32 rx_runt_fcs_align;
+	u32 rx_runt_bytes;
+	u32 rxchk_bad_csum;
+	u32 rxchk_other_pkt_disc;
+	u32 rbuf_ovflow_cnt;
+	u32 rbuf_err_cnt;
+};
+
+/* HW maintains a large list of counters */
+enum bcm_sysport_stat_type {
+	BCM_SYSPORT_STAT_NETDEV = -1,
+	BCM_SYSPORT_STAT_MIB_RX,
+	BCM_SYSPORT_STAT_MIB_TX,
+	BCM_SYSPORT_STAT_RUNT,
+	BCM_SYSPORT_STAT_RXCHK,
+	BCM_SYSPORT_STAT_RBUF,
+};
+
+/* Macros to help define ethtool statistics */
+#define STAT_NETDEV(m) { \
+	.stat_string = __stringify(m), \
+	.stat_sizeof = sizeof(((struct net_device_stats *)0)->m), \
+	.stat_offset = offsetof(struct net_device_stats, m), \
+	.type = BCM_SYSPORT_STAT_NETDEV, \
+}
+
+#define STAT_MIB(str, m, _type) { \
+	.stat_string = str, \
+	.stat_sizeof = sizeof(((struct bcm_sysport_priv *)0)->m), \
+	.stat_offset = offsetof(struct bcm_sysport_priv, m), \
+	.type = _type, \
+}
+
+#define STAT_MIB_RX(str, m) STAT_MIB(str, m, BCM_SYSPORT_STAT_MIB_RX)
+#define STAT_MIB_TX(str, m) STAT_MIB(str, m, BCM_SYSPORT_STAT_MIB_TX)
+#define STAT_RUNT(str, m) STAT_MIB(str, m, BCM_SYSPORT_STAT_RUNT)
+
+#define STAT_RXCHK(str, m, ofs) { \
+	.stat_string = str, \
+	.stat_sizeof = sizeof(((struct bcm_sysport_priv *)0)->m), \
+	.stat_offset = offsetof(struct bcm_sysport_priv, m), \
+	.type = BCM_SYSPORT_STAT_RXCHK, \
+	.reg_offset = ofs, \
+}
+
+#define STAT_RBUF(str, m, ofs) { \
+	.stat_string = str, \
+	.stat_sizeof = sizeof(((struct bcm_sysport_priv *)0)->m), \
+	.stat_offset = offsetof(struct bcm_sysport_priv, m), \
+	.type = BCM_SYSPORT_STAT_RBUF, \
+	.reg_offset = ofs, \
+}
+
+struct bcm_sysport_stats {
+	char stat_string[ETH_GSTRING_LEN];
+	int stat_sizeof;
+	int stat_offset;
+	enum bcm_sysport_stat_type type;
+	/* reg offset from UMAC base for misc counters */
+	u16 reg_offset;
+};
+
+/* Software house keeping helper structure */
+struct bcm_sysport_cb {
+	struct sk_buff	*skb;		/* SKB for RX packets */
+	void __iomem	*bd_addr;	/* Buffer descriptor PHYS addr */
+
+	DEFINE_DMA_UNMAP_ADDR(dma_addr);
+	DEFINE_DMA_UNMAP_LEN(dma_len);
+};
+
+/* Software view of the TX ring */
+struct bcm_sysport_tx_ring {
+	spinlock_t	lock;		/* Ring lock for tx reclaim/xmit */
+	struct napi_struct napi;	/* NAPI per tx queue */
+	dma_addr_t	desc_dma;	/* DMA cookie */
+	unsigned int	index;		/* Ring index */
+	unsigned int	size;		/* Ring current size */
+	unsigned int	alloc_size;	/* Ring one-time allocated size */
+	unsigned int	desc_count;	/* Number of descriptors */
+	unsigned int	curr_desc;	/* Current descriptor */
+	unsigned int	c_index;	/* Last consumer index */
+	unsigned int	p_index;	/* Current producer index */
+	struct bcm_sysport_cb *cbs;	/* Transmit control blocks */
+	struct dma_desc	*desc_cpu;	/* CPU view of the descriptor */
+	struct bcm_sysport_priv *priv;	/* private context backpointer */
+};
+
+/* Driver private structure */
+struct bcm_sysport_priv {
+	void __iomem		*base;
+	u32			irq0_stat;
+	u32			irq0_mask;
+	u32			irq1_stat;
+	u32			irq1_mask;
+	struct napi_struct	napi ____cacheline_aligned;
+	struct net_device	*netdev;
+	struct platform_device	*pdev;
+	int			irq0;
+	int			irq1;
+
+	/* Transmit rings */
+	struct bcm_sysport_tx_ring tx_rings[TDMA_NUM_RINGS];
+
+	/* Receive queue */
+	void __iomem		*rx_bds;
+	void __iomem		*rx_bd_assign_ptr;
+	unsigned int		rx_bd_assign_index;
+	struct bcm_sysport_cb	*rx_cbs;
+	unsigned int		num_rx_bds;
+	unsigned int		rx_read_ptr;
+	unsigned int		rx_c_index;
+
+	/* PHY device */
+	struct device_node	*phy_dn;
+	struct phy_device	*phydev;
+	phy_interface_t		phy_interface;
+	int			old_pause;
+	int			old_link;
+	int			old_duplex;
+
+	/* Misc fields */
+	unsigned int		rx_csum_en:1;
+	unsigned int		tsb_en:1;
+	unsigned int		crc_fwd:1;
+	u16			rev;
+
+	/* MIB related fields */
+	struct bcm_sysport_mib	mib;
+
+	/* Ethtool */
+	u32			msg_enable;
+};
+#endif /* __BCM_SYSPORT_H */
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index 0297a79a38e1..05c6af6c418f 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -1436,7 +1436,7 @@ static int bgmac_probe(struct bcma_device *core)
 		return -ENOMEM;
 	net_dev->netdev_ops = &bgmac_netdev_ops;
 	net_dev->irq = core->irq;
-	SET_ETHTOOL_OPS(net_dev, &bgmac_ethtool_ops);
+	net_dev->ethtool_ops = &bgmac_ethtool_ops;
 	bgmac = netdev_priv(net_dev);
 	bgmac->net_dev = net_dev;
 	bgmac->core = core;
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index 0ab83708b6a1..67d2b0047371 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -6916,8 +6916,8 @@ bnx2_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		}
 	}
 	else {
-		ethtool_cmd_speed_set(cmd, -1);
-		cmd->duplex = -1;
+		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 	spin_unlock_bh(&bp->phy_lock);
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index 4d8f8aba0ea5..4cab09d3f807 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index dd57c7c5a3da..47c5814114e1 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  * UDP CSUM errata workaround by Arik Gendelman
@@ -906,6 +906,18 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 		bd_prod = RX_BD(bd_prod);
 		bd_cons = RX_BD(bd_cons);
 
+		/* A rmb() is required to ensure that the CQE is not read
+		 * before it is written by the adapter DMA.  PCI ordering
+		 * rules will make sure the other fields are written before
+		 * the marker at the end of struct eth_fast_path_rx_cqe
+		 * but without rmb() a weakly ordered processor can process
+		 * stale data.  Without the barrier TPA state-machine might
+		 * enter inconsistent state and kernel stack might be
+		 * provided with incorrect packet description - these lead
+		 * to various kernel crashed.
+		 */
+		rmb();
+
 		cqe_fp_flags = cqe_fp->type_error_flags;
 		cqe_fp_type = cqe_fp_flags & ETH_FAST_PATH_RX_CQE_TYPE;
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 3448cc033ca5..571427c7226b 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  * UDP CSUM errata workaround by Arik Gendelman
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
index 97ea5421dd96..51a952c51cb1 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
@@ -12,7 +12,7 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Dmitry Kravkov
  *
  */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h
index 804b8f64463e..c6939ecb02c5 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h
@@ -12,7 +12,7 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Dmitry Kravkov
  *
  */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index b6de05e3149b..bd0600cf7266 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  * UDP CSUM errata workaround by Arik Gendelman
@@ -3316,7 +3316,7 @@ static u32 bnx2x_get_rxfh_indir_size(struct net_device *dev)
 	return T_ETH_INDIRECTION_TABLE_SIZE;
 }
 
-static int bnx2x_get_rxfh_indir(struct net_device *dev, u32 *indir)
+static int bnx2x_get_rxfh(struct net_device *dev, u32 *indir, u8 *key)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 	u8 ind_table[T_ETH_INDIRECTION_TABLE_SIZE] = {0};
@@ -3340,14 +3340,15 @@ static int bnx2x_get_rxfh_indir(struct net_device *dev, u32 *indir)
 	return 0;
 }
 
-static int bnx2x_set_rxfh_indir(struct net_device *dev, const u32 *indir)
+static int bnx2x_set_rxfh(struct net_device *dev, const u32 *indir,
+			  const u8 *key)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 	size_t i;
 
 	for (i = 0; i < T_ETH_INDIRECTION_TABLE_SIZE; i++) {
 		/*
-		 * The same as in bnx2x_get_rxfh_indir: we can't use a memcpy()
+		 * The same as in bnx2x_get_rxfh: we can't use a memcpy()
 		 * as an internal storage of an indirection table is a u8 array
 		 * while indir->ring_index points to an array of u32.
 		 *
@@ -3471,8 +3472,8 @@ static const struct ethtool_ops bnx2x_ethtool_ops = {
 	.get_rxnfc		= bnx2x_get_rxnfc,
 	.set_rxnfc		= bnx2x_set_rxnfc,
 	.get_rxfh_indir_size	= bnx2x_get_rxfh_indir_size,
-	.get_rxfh_indir		= bnx2x_get_rxfh_indir,
-	.set_rxfh_indir		= bnx2x_set_rxfh_indir,
+	.get_rxfh		= bnx2x_get_rxfh,
+	.set_rxfh		= bnx2x_set_rxfh,
 	.get_channels		= bnx2x_get_channels,
 	.set_channels		= bnx2x_set_channels,
 	.get_module_info	= bnx2x_get_module_info,
@@ -3498,16 +3499,14 @@ static const struct ethtool_ops bnx2x_vf_ethtool_ops = {
 	.get_rxnfc		= bnx2x_get_rxnfc,
 	.set_rxnfc		= bnx2x_set_rxnfc,
 	.get_rxfh_indir_size	= bnx2x_get_rxfh_indir_size,
-	.get_rxfh_indir		= bnx2x_get_rxfh_indir,
-	.set_rxfh_indir		= bnx2x_set_rxfh_indir,
+	.get_rxfh		= bnx2x_get_rxfh,
+	.set_rxfh		= bnx2x_set_rxfh,
 	.get_channels		= bnx2x_get_channels,
 	.set_channels		= bnx2x_set_channels,
 };
 
 void bnx2x_set_ethtool_ops(struct bnx2x *bp, struct net_device *netdev)
 {
-	if (IS_PF(bp))
-		SET_ETHTOOL_OPS(netdev, &bnx2x_ethtool_ops);
-	else /* vf */
-		SET_ETHTOOL_OPS(netdev, &bnx2x_vf_ethtool_ops);
+	netdev->ethtool_ops = (IS_PF(bp)) ?
+		&bnx2x_ethtool_ops : &bnx2x_vf_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_file_hdr.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_file_hdr.h
index f572ae164fce..8aafd9b5d6a2 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_file_hdr.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_file_hdr.h
@@ -6,8 +6,8 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
- * Written by: Vladislav Zolotarov <vladz@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
+ * Written by: Vladislav Zolotarov
  * Based on the original idea of John Wright <john.wright@hp.com>.
  */
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
index c2dfea7968f4..bd90e50bd8e6 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
@@ -7,9 +7,9 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
- * Modified by: Vladislav Zolotarov <vladz@broadcom.com>
+ * Modified by: Vladislav Zolotarov
  */
 
 #ifndef BNX2X_INIT_H
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h
index 8ab0dd900960..5669ed2e87d0 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init_ops.h
@@ -8,8 +8,8 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
- * Written by: Vladislav Zolotarov <vladz@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
+ * Written by: Vladislav Zolotarov
  */
 
 #ifndef BNX2X_INIT_OPS_H
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
index 9b6b3d7304b6..53fb4fa61b40 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
@@ -2218,7 +2218,6 @@ int bnx2x_update_pfc(struct link_params *params,
 	 */
 	u32 val;
 	struct bnx2x *bp = params->bp;
-	int bnx2x_status = 0;
 	u8 bmac_loopback = (params->loopback_mode == LOOPBACK_BMAC);
 
 	if (params->feature_config_flags & FEATURE_CONFIG_PFC_ENABLED)
@@ -2232,7 +2231,7 @@ int bnx2x_update_pfc(struct link_params *params,
 	bnx2x_update_pfc_nig(params, vars, pfc_params);
 
 	if (!vars->link_up)
-		return bnx2x_status;
+		return 0;
 
 	DP(NETIF_MSG_LINK, "About to update PFC in BMAC\n");
 
@@ -2246,7 +2245,7 @@ int bnx2x_update_pfc(struct link_params *params,
 		    == 0) {
 			DP(NETIF_MSG_LINK, "About to update PFC in EMAC\n");
 			bnx2x_emac_enable(params, vars, 0);
-			return bnx2x_status;
+			return 0;
 		}
 		if (CHIP_IS_E2(bp))
 			bnx2x_update_pfc_bmac2(params, vars, bmac_loopback);
@@ -2260,7 +2259,7 @@ int bnx2x_update_pfc(struct link_params *params,
 			val = 1;
 		REG_WR(bp, NIG_REG_BMAC0_PAUSE_OUT_EN + params->port*4, val);
 	}
-	return bnx2x_status;
+	return 0;
 }
 
 static int bnx2x_bmac1_enable(struct link_params *params,
@@ -3703,7 +3702,8 @@ static void bnx2x_warpcore_restart_AN_KR(struct bnx2x_phy *phy,
 static void bnx2x_warpcore_enable_AN_KR(struct bnx2x_phy *phy,
 					struct link_params *params,
 					struct link_vars *vars) {
-	u16 lane, i, cl72_ctrl, an_adv = 0;
+	u16 lane, i, cl72_ctrl, an_adv = 0, val;
+	u32 wc_lane_config;
 	struct bnx2x *bp = params->bp;
 	static struct bnx2x_reg_set reg_set[] = {
 		{MDIO_WC_DEVAD, MDIO_WC_REG_SERDESDIGITAL_CONTROL1000X2, 0x7},
@@ -3822,15 +3822,27 @@ static void bnx2x_warpcore_enable_AN_KR(struct bnx2x_phy *phy,
 		/* Enable Auto-Detect to support 1G over CL37 as well */
 		bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
 				 MDIO_WC_REG_SERDESDIGITAL_CONTROL1000X1, 0x10);
-
+		wc_lane_config = REG_RD(bp, params->shmem_base +
+					offsetof(struct shmem_region, dev_info.
+					shared_hw_config.wc_lane_config));
+		bnx2x_cl45_read(bp, phy, MDIO_WC_DEVAD,
+				MDIO_WC_REG_RX0_PCI_CTRL + (lane << 4), &val);
 		/* Force cl48 sync_status LOW to avoid getting stuck in CL73
 		 * parallel-detect loop when CL73 and CL37 are enabled.
 		 */
-		CL22_WR_OVER_CL45(bp, phy, MDIO_REG_BANK_AER_BLOCK,
-				  MDIO_AER_BLOCK_AER_REG, 0);
+		val |= 1 << 11;
+
+		/* Restore Polarity settings in case it was run over by
+		 * previous link owner
+		 */
+		if (wc_lane_config &
+		    (SHARED_HW_CFG_RX_LANE0_POL_FLIP_ENABLED << lane))
+			val |= 3 << 2;
+		else
+			val &= ~(3 << 2);
 		bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
-				 MDIO_WC_REG_RXB_ANA_RX_CONTROL_PCI, 0x0800);
-		bnx2x_set_aer_mmd(params, phy);
+				 MDIO_WC_REG_RX0_PCI_CTRL + (lane << 4),
+				 val);
 
 		bnx2x_disable_kr2(params, vars, phy);
 	}
@@ -6473,7 +6485,6 @@ int bnx2x_test_link(struct link_params *params, struct link_vars *vars,
 static int bnx2x_link_initialize(struct link_params *params,
 				 struct link_vars *vars)
 {
-	int rc = 0;
 	u8 phy_index, non_ext_phy;
 	struct bnx2x *bp = params->bp;
 	/* In case of external phy existence, the line speed would be the
@@ -6546,7 +6557,7 @@ static int bnx2x_link_initialize(struct link_params *params,
 			NIG_STATUS_XGXS0_LINK_STATUS |
 			NIG_STATUS_SERDES0_LINK_STATUS |
 			NIG_MASK_MI_INT));
-	return rc;
+	return 0;
 }
 
 static void bnx2x_int_link_reset(struct bnx2x_phy *phy,
@@ -12461,6 +12472,7 @@ static int bnx2x_avoid_link_flap(struct link_params *params,
 	u32 dont_clear_stat, lfa_sts;
 	struct bnx2x *bp = params->bp;
 
+	bnx2x_set_mdio_emac_per_phy(bp, params);
 	/* Sync the link parameters */
 	bnx2x_link_status_update(params, vars);
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 3a8e51ed5bec..2887034523e0 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  * UDP CSUM errata workaround by Arik Gendelman
@@ -10053,6 +10053,24 @@ static void bnx2x_prev_unload_close_mac(struct bnx2x *bp,
 #define BCM_5710_UNDI_FW_MF_VERS	(0x05)
 #define BNX2X_PREV_UNDI_MF_PORT(p) (BAR_TSTRORM_INTMEM + 0x150c + ((p) << 4))
 #define BNX2X_PREV_UNDI_MF_FUNC(f) (BAR_TSTRORM_INTMEM + 0x184c + ((f) << 4))
+
+static bool bnx2x_prev_is_after_undi(struct bnx2x *bp)
+{
+	/* UNDI marks its presence in DORQ -
+	 * it initializes CID offset for normal bell to 0x7
+	 */
+	if (!(REG_RD(bp, MISC_REG_RESET_REG_1) &
+	    MISC_REGISTERS_RESET_REG_1_RST_DORQ))
+		return false;
+
+	if (REG_RD(bp, DORQ_REG_NORM_CID_OFST) == 0x7) {
+		BNX2X_DEV_INFO("UNDI previously loaded\n");
+		return true;
+	}
+
+	return false;
+}
+
 static bool bnx2x_prev_unload_undi_fw_supports_mf(struct bnx2x *bp)
 {
 	u8 major, minor, version;
@@ -10302,6 +10320,10 @@ static int bnx2x_prev_unload_uncommon(struct bnx2x *bp)
 
 	BNX2X_DEV_INFO("Path is unmarked\n");
 
+	/* Cannot proceed with FLR if UNDI is loaded, since FW does not match */
+	if (bnx2x_prev_is_after_undi(bp))
+		goto out;
+
 	/* If function has FLR capabilities, and existing FW version matches
 	 * the one required, then FLR will be sufficient to clean any residue
 	 * left by previous driver
@@ -10322,6 +10344,7 @@ static int bnx2x_prev_unload_uncommon(struct bnx2x *bp)
 
 	BNX2X_DEV_INFO("Could not FLR\n");
 
+out:
 	/* Close the MCP request, return failure*/
 	rc = bnx2x_prev_mcp_done(bp);
 	if (!rc)
@@ -10360,19 +10383,13 @@ static int bnx2x_prev_unload_common(struct bnx2x *bp)
 		/* close LLH filters towards the BRB */
 		bnx2x_set_rx_filter(&bp->link_params, 0);
 
-		/* Check if the UNDI driver was previously loaded
-		 * UNDI driver initializes CID offset for normal bell to 0x7
-		 */
-		if (reset_reg & MISC_REGISTERS_RESET_REG_1_RST_DORQ) {
-			tmp_reg = REG_RD(bp, DORQ_REG_NORM_CID_OFST);
-			if (tmp_reg == 0x7) {
-				BNX2X_DEV_INFO("UNDI previously loaded\n");
-				prev_undi = true;
-				/* clear the UNDI indication */
-				REG_WR(bp, DORQ_REG_NORM_CID_OFST, 0);
-				/* clear possible idle check errors */
-				REG_RD(bp, NIG_REG_NIG_INT_STS_CLR_0);
-			}
+		/* Check if the UNDI driver was previously loaded */
+		if (bnx2x_prev_is_after_undi(bp)) {
+			prev_undi = true;
+			/* clear the UNDI indication */
+			REG_WR(bp, DORQ_REG_NORM_CID_OFST, 0);
+			/* clear possible idle check errors */
+			REG_RD(bp, NIG_REG_NIG_INT_STS_CLR_0);
 		}
 		if (!CHIP_IS_E1x(bp))
 			/* block FW from writing to host */
@@ -13283,8 +13300,8 @@ static int bnx2x_eeh_nic_unload(struct bnx2x *bp)
 	netdev_reset_tc(bp->dev);
 
 	del_timer_sync(&bp->timer);
-	cancel_delayed_work(&bp->sp_task);
-	cancel_delayed_work(&bp->period_task);
+	cancel_delayed_work_sync(&bp->sp_task);
+	cancel_delayed_work_sync(&bp->period_task);
 
 	spin_lock_bh(&bp->stats_lock);
 	bp->stats_state = STATS_STATE_DISABLED;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
index d725317c4277..b1936044767a 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
@@ -12,7 +12,7 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Vladislav Zolotarov
  *
  */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
index 80f6c790ed88..718ecd294661 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
@@ -12,7 +12,7 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Vladislav Zolotarov
  *
  */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
index faf01488d26e..eda8583f6fc0 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
@@ -12,9 +12,9 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
- * Written by: Shmulik Ravid <shmulikr@broadcom.com>
- *	       Ariel Elior <ariele@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
+ * Written by: Shmulik Ravid
+ *	       Ariel Elior <ariel.elior@qlogic.com>
  *
  */
 #include "bnx2x.h"
@@ -1071,8 +1071,10 @@ void bnx2x_iov_init_dq(struct bnx2x *bp)
 	REG_WR(bp, DORQ_REG_VF_TYPE_MIN_MCID_0, 0);
 	REG_WR(bp, DORQ_REG_VF_TYPE_MAX_MCID_0, 0x1ffff);
 
-	/* set the VF doorbell threshold */
-	REG_WR(bp, DORQ_REG_VF_USAGE_CT_LIMIT, 4);
+	/* set the VF doorbell threshold. This threshold represents the amount
+	 * of doorbells allowed in the main DORQ fifo for a specific VF.
+	 */
+	REG_WR(bp, DORQ_REG_VF_USAGE_CT_LIMIT, 64);
 }
 
 void bnx2x_iov_init_dmae(struct bnx2x *bp)
@@ -2576,7 +2578,8 @@ int bnx2x_get_vf_config(struct net_device *dev, int vfidx,
 
 	ivi->vf = vfidx;
 	ivi->qos = 0;
-	ivi->tx_rate = 10000; /* always 10G. TBA take from link struct */
+	ivi->max_tx_rate = 10000; /* always 10G. TBA take from link struct */
+	ivi->min_tx_rate = 0;
 	ivi->spoofchk = 1; /*always enabled */
 	if (vf->state == VF_ENABLED) {
 		/* mac and vlan are in vlan_mac objects */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
index 6929adba52f9..96c575e147a5 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
@@ -12,9 +12,9 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
- * Written by: Shmulik Ravid <shmulikr@broadcom.com>
- *	       Ariel Elior <ariele@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
+ * Written by: Shmulik Ravid
+ *	       Ariel Elior <ariel.elior@qlogic.com>
  */
 #ifndef BNX2X_SRIOV_H
 #define BNX2X_SRIOV_H
@@ -571,7 +571,7 @@ static inline void __iomem *bnx2x_vf_doorbells(struct bnx2x *bp)
 	return NULL;
 }
 
-static inline void bnx2x_vf_pci_dealloc(struct bnx2 *bp) {return 0; }
+static inline void bnx2x_vf_pci_dealloc(struct bnx2x *bp) {}
 static inline int bnx2x_vf_pci_alloc(struct bnx2x *bp) {return 0; }
 static inline void bnx2x_pf_set_vfs_vlan(struct bnx2x *bp) {}
 static inline int bnx2x_sriov_configure(struct pci_dev *dev, int num_vfs) {return 0; }
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
index 3b75070411aa..ca47665f94bf 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  * UDP CSUM errata workaround by Arik Gendelman
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
index f35845006cdd..2beceaefdeea 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
@@ -6,7 +6,7 @@
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
  * Written by: Eliezer Tamir
  * Based on code from Michael Chan's bnx2 driver
  * UDP CSUM errata workaround by Arik Gendelman
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
index 784c7155b98a..d712d0ddd719 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
@@ -12,9 +12,9 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
- * Written by: Shmulik Ravid <shmulikr@broadcom.com>
- *	       Ariel Elior <ariele@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
+ * Written by: Shmulik Ravid
+ *	       Ariel Elior <ariel.elior@qlogic.com>
  */
 
 #include "bnx2x.h"
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
index c922b81170e5..e21e706762c9 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
@@ -12,8 +12,8 @@
  * license other than the GPL, without Broadcom's express prior written
  * consent.
  *
- * Maintained by: Eilon Greenstein <eilong@broadcom.com>
- * Written by: Ariel Elior <ariele@broadcom.com>
+ * Maintained by: Ariel Elior <ariel.elior@qlogic.com>
+ * Written by: Ariel Elior <ariel.elior@qlogic.com>
  */
 #ifndef VF_PF_IF_H
 #define VF_PF_IF_H
diff --git a/drivers/net/ethernet/broadcom/cnic.c b/drivers/net/ethernet/broadcom/cnic.c
index 4dd48d2fa804..8244e2b14bb4 100644
--- a/drivers/net/ethernet/broadcom/cnic.c
+++ b/drivers/net/ethernet/broadcom/cnic.c
@@ -608,6 +608,10 @@ static int cnic_unregister_device(struct cnic_dev *dev, int ulp_type)
 		pr_err("%s: Bad type %d\n", __func__, ulp_type);
 		return -EINVAL;
 	}
+
+	if (ulp_type == CNIC_ULP_ISCSI)
+		cnic_send_nlmsg(cp, ISCSI_KEVENT_IF_DOWN, NULL);
+
 	mutex_lock(&cnic_lock);
 	if (rcu_dereference(cp->ulp_ops[ulp_type])) {
 		RCU_INIT_POINTER(cp->ulp_ops[ulp_type], NULL);
@@ -620,9 +624,7 @@ static int cnic_unregister_device(struct cnic_dev *dev, int ulp_type)
 	}
 	mutex_unlock(&cnic_lock);
 
-	if (ulp_type == CNIC_ULP_ISCSI)
-		cnic_send_nlmsg(cp, ISCSI_KEVENT_IF_DOWN, NULL);
-	else if (ulp_type == CNIC_ULP_FCOE)
+	if (ulp_type == CNIC_ULP_FCOE)
 		dev->fcoe_cap = NULL;
 
 	synchronize_rcu();
@@ -1039,21 +1041,17 @@ static int cnic_alloc_uio_rings(struct cnic_dev *dev, int pages)
 	struct cnic_local *cp = dev->cnic_priv;
 	struct cnic_uio_dev *udev;
 
-	read_lock(&cnic_dev_lock);
 	list_for_each_entry(udev, &cnic_udev_list, list) {
 		if (udev->pdev == dev->pcidev) {
 			udev->dev = dev;
 			if (__cnic_alloc_uio_rings(udev, pages)) {
 				udev->dev = NULL;
-				read_unlock(&cnic_dev_lock);
 				return -ENOMEM;
 			}
 			cp->udev = udev;
-			read_unlock(&cnic_dev_lock);
 			return 0;
 		}
 	}
-	read_unlock(&cnic_dev_lock);
 
 	udev = kzalloc(sizeof(struct cnic_uio_dev), GFP_ATOMIC);
 	if (!udev)
@@ -1067,9 +1065,7 @@ static int cnic_alloc_uio_rings(struct cnic_dev *dev, int pages)
 	if (__cnic_alloc_uio_rings(udev, pages))
 		goto err_udev;
 
-	write_lock(&cnic_dev_lock);
 	list_add(&udev->list, &cnic_udev_list);
-	write_unlock(&cnic_dev_lock);
 
 	pci_dev_get(udev->pdev);
 
@@ -5624,20 +5620,27 @@ static void cnic_rcv_netevent(struct cnic_local *cp, unsigned long event,
 {
 	int if_type;
 
-	rcu_read_lock();
 	for (if_type = 0; if_type < MAX_CNIC_ULP_TYPE; if_type++) {
 		struct cnic_ulp_ops *ulp_ops;
 		void *ctx;
 
-		ulp_ops = rcu_dereference(cp->ulp_ops[if_type]);
-		if (!ulp_ops || !ulp_ops->indicate_netevent)
+		mutex_lock(&cnic_lock);
+		ulp_ops = rcu_dereference_protected(cp->ulp_ops[if_type],
+						lockdep_is_held(&cnic_lock));
+		if (!ulp_ops || !ulp_ops->indicate_netevent) {
+			mutex_unlock(&cnic_lock);
 			continue;
+		}
 
 		ctx = cp->ulp_handle[if_type];
 
+		set_bit(ULP_F_CALL_PENDING, &cp->ulp_flags[if_type]);
+		mutex_unlock(&cnic_lock);
+
 		ulp_ops->indicate_netevent(ctx, event, vlan_id);
+
+		clear_bit(ULP_F_CALL_PENDING, &cp->ulp_flags[if_type]);
 	}
-	rcu_read_unlock();
 }
 
 /* netdev event handler */
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 0966bd04375f..5ba1cfbd60da 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -2481,7 +2481,7 @@ static int bcmgenet_probe(struct platform_device *pdev)
 	dev_set_drvdata(&pdev->dev, dev);
 	ether_addr_copy(dev->dev_addr, macaddr);
 	dev->watchdog_timeo = 2 * HZ;
-	SET_ETHTOOL_OPS(dev, &bcmgenet_ethtool_ops);
+	dev->ethtool_ops = &bcmgenet_ethtool_ops;
 	dev->netdev_ops = &bcmgenet_netdev_ops;
 	netif_napi_add(dev, &priv->napi, bcmgenet_poll, 64);
 
diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c
index 4608673beaff..add8d8596084 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmmii.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c
@@ -298,6 +298,7 @@ int bcmgenet_mii_config(struct net_device *dev)
 static int bcmgenet_mii_probe(struct net_device *dev)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
+	struct device_node *dn = priv->pdev->dev.of_node;
 	struct phy_device *phydev;
 	unsigned int phy_flags;
 	int ret;
@@ -307,15 +308,19 @@ static int bcmgenet_mii_probe(struct net_device *dev)
 		return 0;
 	}
 
-	if (priv->phy_dn)
-		phydev = of_phy_connect(dev, priv->phy_dn,
-					bcmgenet_mii_setup, 0,
-					priv->phy_interface);
-	else
-		phydev = of_phy_connect_fixed_link(dev,
-					bcmgenet_mii_setup,
-					priv->phy_interface);
+	/* In the case of a fixed PHY, the DT node associated
+	 * to the PHY is the Ethernet MAC DT node.
+	 */
+	if (of_phy_is_fixed_link(dn)) {
+		ret = of_phy_register_fixed_link(dn);
+		if (ret)
+			return ret;
+
+		priv->phy_dn = dn;
+	}
 
+	phydev = of_phy_connect(dev, priv->phy_dn, bcmgenet_mii_setup, 0,
+				priv->phy_interface);
 	if (!phydev) {
 		pr_err("could not attach to PHY\n");
 		return -ENODEV;
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index e5d95c5ce1ad..8afa579e7c40 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -4,7 +4,7 @@
  * Copyright (C) 2001, 2002, 2003, 2004 David S. Miller (davem@redhat.com)
  * Copyright (C) 2001, 2002, 2003 Jeff Garzik (jgarzik@pobox.com)
  * Copyright (C) 2004 Sun Microsystems Inc.
- * Copyright (C) 2005-2013 Broadcom Corporation.
+ * Copyright (C) 2005-2014 Broadcom Corporation.
  *
  * Firmware is:
  *	Derived from proprietary unpublished source code,
@@ -94,10 +94,10 @@ static inline void _tg3_flag_clear(enum TG3_FLAGS flag, unsigned long *bits)
 
 #define DRV_MODULE_NAME		"tg3"
 #define TG3_MAJ_NUM			3
-#define TG3_MIN_NUM			136
+#define TG3_MIN_NUM			137
 #define DRV_MODULE_VERSION	\
 	__stringify(TG3_MAJ_NUM) "." __stringify(TG3_MIN_NUM)
-#define DRV_MODULE_RELDATE	"Jan 03, 2014"
+#define DRV_MODULE_RELDATE	"May 11, 2014"
 
 #define RESET_KIND_SHUTDOWN	0
 #define RESET_KIND_INIT		1
@@ -3224,7 +3224,7 @@ static int tg3_nvram_read_using_eeprom(struct tg3 *tp,
 	return 0;
 }
 
-#define NVRAM_CMD_TIMEOUT 10000
+#define NVRAM_CMD_TIMEOUT 5000
 
 static int tg3_nvram_exec_cmd(struct tg3 *tp, u32 nvram_cmd)
 {
@@ -3232,7 +3232,7 @@ static int tg3_nvram_exec_cmd(struct tg3 *tp, u32 nvram_cmd)
 
 	tw32(NVRAM_CMD, nvram_cmd);
 	for (i = 0; i < NVRAM_CMD_TIMEOUT; i++) {
-		udelay(10);
+		usleep_range(10, 40);
 		if (tr32(NVRAM_CMD) & NVRAM_CMD_DONE) {
 			udelay(10);
 			break;
@@ -7854,8 +7854,8 @@ static int tg3_tso_bug(struct tg3 *tp, struct sk_buff *skb)
 		netif_wake_queue(tp->dev);
 	}
 
-	segs = skb_gso_segment(skb, tp->dev->features & ~NETIF_F_TSO);
-	if (IS_ERR(segs))
+	segs = skb_gso_segment(skb, tp->dev->features & ~(NETIF_F_TSO | NETIF_F_TSO6));
+	if (IS_ERR(segs) || !segs)
 		goto tg3_tso_bug_end;
 
 	do {
@@ -7871,9 +7871,7 @@ tg3_tso_bug_end:
 	return NETDEV_TX_OK;
 }
 
-/* hard_start_xmit for devices that have the 4G bug and/or 40-bit bug and
- * support TG3_FLAG_HW_TSO_1 or firmware TSO only.
- */
+/* hard_start_xmit for all devices */
 static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct tg3 *tp = netdev_priv(dev);
@@ -7884,6 +7882,10 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct tg3_napi *tnapi;
 	struct netdev_queue *txq;
 	unsigned int last;
+	struct iphdr *iph = NULL;
+	struct tcphdr *tcph = NULL;
+	__sum16 tcp_csum = 0, ip_csum = 0;
+	__be16 ip_tot_len = 0;
 
 	txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
 	tnapi = &tp->napi[skb_get_queue_mapping(skb)];
@@ -7915,7 +7917,6 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	mss = skb_shinfo(skb)->gso_size;
 	if (mss) {
-		struct iphdr *iph;
 		u32 tcp_opt_len, hdr_len;
 
 		if (skb_cow_head(skb, 0))
@@ -7927,27 +7928,31 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb) - ETH_HLEN;
 
 		if (!skb_is_gso_v6(skb)) {
+			if (unlikely((ETH_HLEN + hdr_len) > 80) &&
+			    tg3_flag(tp, TSO_BUG))
+				return tg3_tso_bug(tp, skb);
+
+			ip_csum = iph->check;
+			ip_tot_len = iph->tot_len;
 			iph->check = 0;
 			iph->tot_len = htons(mss + hdr_len);
 		}
 
-		if (unlikely((ETH_HLEN + hdr_len) > 80) &&
-		    tg3_flag(tp, TSO_BUG))
-			return tg3_tso_bug(tp, skb);
-
 		base_flags |= (TXD_FLAG_CPU_PRE_DMA |
 			       TXD_FLAG_CPU_POST_DMA);
 
+		tcph = tcp_hdr(skb);
+		tcp_csum = tcph->check;
+
 		if (tg3_flag(tp, HW_TSO_1) ||
 		    tg3_flag(tp, HW_TSO_2) ||
 		    tg3_flag(tp, HW_TSO_3)) {
-			tcp_hdr(skb)->check = 0;
+			tcph->check = 0;
 			base_flags &= ~TXD_FLAG_TCPUDP_CSUM;
-		} else
-			tcp_hdr(skb)->check = ~csum_tcpudp_magic(iph->saddr,
-								 iph->daddr, 0,
-								 IPPROTO_TCP,
-								 0);
+		} else {
+			tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
+							 0, IPPROTO_TCP, 0);
+		}
 
 		if (tg3_flag(tp, HW_TSO_3)) {
 			mss |= (hdr_len & 0xc) << 12;
@@ -8047,6 +8052,18 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (would_hit_hwbug) {
 		tg3_tx_skb_unmap(tnapi, tnapi->tx_prod, i);
 
+		if (mss) {
+			/* If it's a TSO packet, do GSO instead of
+			 * allocating and copying to a large linear SKB
+			 */
+			if (ip_tot_len) {
+				iph->check = ip_csum;
+				iph->tot_len = ip_tot_len;
+			}
+			tcph->check = tcp_csum;
+			return tg3_tso_bug(tp, skb);
+		}
+
 		/* If the workaround fails due to memory/mapping
 		 * failure, silently drop this packet.
 		 */
@@ -11876,9 +11893,9 @@ static int tg3_get_eeprom_len(struct net_device *dev)
 static int tg3_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *data)
 {
 	struct tg3 *tp = netdev_priv(dev);
-	int ret;
+	int ret, cpmu_restore = 0;
 	u8  *pd;
-	u32 i, offset, len, b_offset, b_count;
+	u32 i, offset, len, b_offset, b_count, cpmu_val = 0;
 	__be32 val;
 
 	if (tg3_flag(tp, NO_NVRAM))
@@ -11890,6 +11907,19 @@ static int tg3_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 
 	eeprom->magic = TG3_EEPROM_MAGIC;
 
+	/* Override clock, link aware and link idle modes */
+	if (tg3_flag(tp, CPMU_PRESENT)) {
+		cpmu_val = tr32(TG3_CPMU_CTRL);
+		if (cpmu_val & (CPMU_CTRL_LINK_AWARE_MODE |
+				CPMU_CTRL_LINK_IDLE_MODE)) {
+			tw32(TG3_CPMU_CTRL, cpmu_val &
+					    ~(CPMU_CTRL_LINK_AWARE_MODE |
+					     CPMU_CTRL_LINK_IDLE_MODE));
+			cpmu_restore = 1;
+		}
+	}
+	tg3_override_clk(tp);
+
 	if (offset & 3) {
 		/* adjustments to start on required 4 byte boundary */
 		b_offset = offset & 3;
@@ -11900,7 +11930,7 @@ static int tg3_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 		}
 		ret = tg3_nvram_read_be32(tp, offset-b_offset, &val);
 		if (ret)
-			return ret;
+			goto eeprom_done;
 		memcpy(data, ((char *)&val) + b_offset, b_count);
 		len -= b_count;
 		offset += b_count;
@@ -11912,10 +11942,20 @@ static int tg3_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 	for (i = 0; i < (len - (len & 3)); i += 4) {
 		ret = tg3_nvram_read_be32(tp, offset + i, &val);
 		if (ret) {
+			if (i)
+				i -= 4;
 			eeprom->len += i;
-			return ret;
+			goto eeprom_done;
 		}
 		memcpy(pd + i, &val, 4);
+		if (need_resched()) {
+			if (signal_pending(current)) {
+				eeprom->len += i;
+				ret = -EINTR;
+				goto eeprom_done;
+			}
+			cond_resched();
+		}
 	}
 	eeprom->len += i;
 
@@ -11926,11 +11966,19 @@ static int tg3_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
 		b_offset = offset + len - b_count;
 		ret = tg3_nvram_read_be32(tp, b_offset, &val);
 		if (ret)
-			return ret;
+			goto eeprom_done;
 		memcpy(pd, &val, b_count);
 		eeprom->len += b_count;
 	}
-	return 0;
+	ret = 0;
+
+eeprom_done:
+	/* Restore clock, link aware and link idle modes */
+	tg3_restore_clk(tp);
+	if (cpmu_restore)
+		tw32(TG3_CPMU_CTRL, cpmu_val);
+
+	return ret;
 }
 
 static int tg3_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *data)
@@ -12484,7 +12532,7 @@ static u32 tg3_get_rxfh_indir_size(struct net_device *dev)
 	return size;
 }
 
-static int tg3_get_rxfh_indir(struct net_device *dev, u32 *indir)
+static int tg3_get_rxfh(struct net_device *dev, u32 *indir, u8 *key)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	int i;
@@ -12495,7 +12543,7 @@ static int tg3_get_rxfh_indir(struct net_device *dev, u32 *indir)
 	return 0;
 }
 
-static int tg3_set_rxfh_indir(struct net_device *dev, const u32 *indir)
+static int tg3_set_rxfh(struct net_device *dev, const u32 *indir, const u8 *key)
 {
 	struct tg3 *tp = netdev_priv(dev);
 	size_t i;
@@ -14027,8 +14075,8 @@ static const struct ethtool_ops tg3_ethtool_ops = {
 	.get_sset_count		= tg3_get_sset_count,
 	.get_rxnfc		= tg3_get_rxnfc,
 	.get_rxfh_indir_size    = tg3_get_rxfh_indir_size,
-	.get_rxfh_indir		= tg3_get_rxfh_indir,
-	.set_rxfh_indir		= tg3_set_rxfh_indir,
+	.get_rxfh		= tg3_get_rxfh,
+	.set_rxfh		= tg3_set_rxfh,
 	.get_channels		= tg3_get_channels,
 	.set_channels		= tg3_set_channels,
 	.get_ts_info		= tg3_get_ts_info,
diff --git a/drivers/net/ethernet/broadcom/tg3.h b/drivers/net/ethernet/broadcom/tg3.h
index 04321e5a356e..461accaf0aa4 100644
--- a/drivers/net/ethernet/broadcom/tg3.h
+++ b/drivers/net/ethernet/broadcom/tg3.h
@@ -4,7 +4,7 @@
  * Copyright (C) 2001, 2002, 2003, 2004 David S. Miller (davem@redhat.com)
  * Copyright (C) 2001 Jeff Garzik (jgarzik@pobox.com)
  * Copyright (C) 2004 Sun Microsystems Inc.
- * Copyright (C) 2007-2013 Broadcom Corporation.
+ * Copyright (C) 2007-2014 Broadcom Corporation.
  */
 
 #ifndef _T3_H
diff --git a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c
index f9e150825bb5..882cad71ad62 100644
--- a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c
+++ b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c
@@ -266,8 +266,8 @@ bnad_get_settings(struct net_device *netdev, struct ethtool_cmd *cmd)
 		ethtool_cmd_speed_set(cmd, SPEED_10000);
 		cmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(cmd, -1);
-		cmd->duplex = -1;
+		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 	cmd->transceiver = XCVR_EXTERNAL;
 	cmd->maxtxpkt = 0;
@@ -1137,5 +1137,5 @@ static const struct ethtool_ops bnad_ethtool_ops = {
 void
 bnad_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &bnad_ethtool_ops);
+	netdev->ethtool_ops = &bnad_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c
index 521dfea44b83..25d6b2a10e4e 100644
--- a/drivers/net/ethernet/calxeda/xgmac.c
+++ b/drivers/net/ethernet/calxeda/xgmac.c
@@ -1737,7 +1737,7 @@ static int xgmac_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, ndev);
 	ether_setup(ndev);
 	ndev->netdev_ops = &xgmac_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &xgmac_ethtool_ops);
+	ndev->ethtool_ops = &xgmac_ethtool_ops;
 	spin_lock_init(&priv->stats_lock);
 	INIT_WORK(&priv->tx_timeout_work, xgmac_tx_timeout_work);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
index 05613a85ce61..186566bfdbc8 100644
--- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
+++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
@@ -580,8 +580,8 @@ static int get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		ethtool_cmd_speed_set(cmd, p->link_config.speed);
 		cmd->duplex = p->link_config.duplex;
 	} else {
-		ethtool_cmd_speed_set(cmd, -1);
-		cmd->duplex = -1;
+		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	cmd->port = (cmd->supported & SUPPORTED_TP) ? PORT_TP : PORT_FIBRE;
@@ -1100,7 +1100,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 		netif_napi_add(netdev, &adapter->napi, t1_poll, 64);
 
-		SET_ETHTOOL_OPS(netdev, &t1_ethtool_ops);
+		netdev->ethtool_ops = &t1_ethtool_ops;
 	}
 
 	if (t1_init_sw_modules(adapter, bi) < 0) {
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
index 07bbb711b7e5..5d9cce053cc9 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
@@ -1809,8 +1809,8 @@ static int get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		ethtool_cmd_speed_set(cmd, p->link_config.speed);
 		cmd->duplex = p->link_config.duplex;
 	} else {
-		ethtool_cmd_speed_set(cmd, -1);
-		cmd->duplex = -1;
+		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	cmd->port = (cmd->supported & SUPPORTED_TP) ? PORT_TP : PORT_FIBRE;
@@ -3291,7 +3291,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			netdev->features |= NETIF_F_HIGHDMA;
 
 		netdev->netdev_ops = &cxgb_netdev_ops;
-		SET_ETHTOOL_OPS(netdev, &cxgb_ethtool_ops);
+		netdev->ethtool_ops = &cxgb_ethtool_ops;
 	}
 
 	pci_set_drvdata(pdev, adapter);
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
index c0a9dd55f4e5..b0cbb2b7fd48 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
@@ -185,7 +185,7 @@ static struct net_device *get_iff_from_mac(struct adapter *adapter,
 		if (ether_addr_equal(dev->dev_addr, mac)) {
 			rcu_read_lock();
 			if (vlan && vlan != VLAN_VID_MASK) {
-				dev = __vlan_find_dev_deep(dev, htons(ETH_P_8021Q), vlan);
+				dev = __vlan_find_dev_deep_rcu(dev, htons(ETH_P_8021Q), vlan);
 			} else if (netif_is_bond_slave(dev)) {
 				struct net_device *upper_dev;
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 32db37709263..f503dce4ab17 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -357,11 +357,17 @@ enum {
 	MAX_OFLD_QSETS = 16,          /* # of offload Tx/Rx queue sets */
 	MAX_CTRL_QUEUES = NCHAN,      /* # of control Tx queues */
 	MAX_RDMA_QUEUES = NCHAN,      /* # of streaming RDMA Rx queues */
+	MAX_RDMA_CIQS = NCHAN,        /* # of  RDMA concentrator IQs */
+	MAX_ISCSI_QUEUES = NCHAN,     /* # of streaming iSCSI Rx queues */
 };
 
 enum {
-	MAX_EGRQ = 128,         /* max # of egress queues, including FLs */
-	MAX_INGQ = 64           /* max # of interrupt-capable ingress queues */
+	INGQ_EXTRAS = 2,        /* firmware event queue and */
+				/*   forwarded interrupts */
+	MAX_EGRQ = MAX_ETH_QSETS*2 + MAX_OFLD_QSETS*2
+		   + MAX_CTRL_QUEUES + MAX_RDMA_QUEUES + MAX_ISCSI_QUEUES,
+	MAX_INGQ = MAX_ETH_QSETS + MAX_OFLD_QSETS + MAX_RDMA_QUEUES
+		   + MAX_RDMA_CIQS + MAX_ISCSI_QUEUES + INGQ_EXTRAS,
 };
 
 struct adapter;
@@ -538,6 +544,7 @@ struct sge {
 	struct sge_eth_rxq ethrxq[MAX_ETH_QSETS];
 	struct sge_ofld_rxq ofldrxq[MAX_OFLD_QSETS];
 	struct sge_ofld_rxq rdmarxq[MAX_RDMA_QUEUES];
+	struct sge_ofld_rxq rdmaciq[MAX_RDMA_CIQS];
 	struct sge_rspq fw_evtq ____cacheline_aligned_in_smp;
 
 	struct sge_rspq intrq ____cacheline_aligned_in_smp;
@@ -548,8 +555,10 @@ struct sge {
 	u16 ethtxq_rover;           /* Tx queue to clean up next */
 	u16 ofldqsets;              /* # of active offload queue sets */
 	u16 rdmaqs;                 /* # of available RDMA Rx queues */
+	u16 rdmaciqs;               /* # of available RDMA concentrator IQs */
 	u16 ofld_rxq[MAX_OFLD_QSETS];
 	u16 rdma_rxq[NCHAN];
+	u16 rdma_ciq[NCHAN];
 	u16 timer_val[SGE_NTIMERS];
 	u8 counter_val[SGE_NCOUNTERS];
 	u32 fl_pg_order;            /* large page allocation size */
@@ -577,6 +586,7 @@ struct sge {
 #define for_each_ethrxq(sge, i) for (i = 0; i < (sge)->ethqsets; i++)
 #define for_each_ofldrxq(sge, i) for (i = 0; i < (sge)->ofldqsets; i++)
 #define for_each_rdmarxq(sge, i) for (i = 0; i < (sge)->rdmaqs; i++)
+#define for_each_rdmaciq(sge, i) for (i = 0; i < (sge)->rdmaciqs; i++)
 
 struct l2t_data;
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 24e16e3301e0..a83271cf17c3 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -818,12 +818,17 @@ static void name_msix_vecs(struct adapter *adap)
 	for_each_rdmarxq(&adap->sge, i)
 		snprintf(adap->msix_info[msi_idx++].desc, n, "%s-rdma%d",
 			 adap->port[0]->name, i);
+
+	for_each_rdmaciq(&adap->sge, i)
+		snprintf(adap->msix_info[msi_idx++].desc, n, "%s-rdma-ciq%d",
+			 adap->port[0]->name, i);
 }
 
 static int request_msix_queue_irqs(struct adapter *adap)
 {
 	struct sge *s = &adap->sge;
-	int err, ethqidx, ofldqidx = 0, rdmaqidx = 0, msi_index = 2;
+	int err, ethqidx, ofldqidx = 0, rdmaqidx = 0, rdmaciqqidx = 0;
+	int msi_index = 2;
 
 	err = request_irq(adap->msix_info[1].vec, t4_sge_intr_msix, 0,
 			  adap->msix_info[1].desc, &s->fw_evtq);
@@ -857,9 +862,21 @@ static int request_msix_queue_irqs(struct adapter *adap)
 			goto unwind;
 		msi_index++;
 	}
+	for_each_rdmaciq(s, rdmaciqqidx) {
+		err = request_irq(adap->msix_info[msi_index].vec,
+				  t4_sge_intr_msix, 0,
+				  adap->msix_info[msi_index].desc,
+				  &s->rdmaciq[rdmaciqqidx].rspq);
+		if (err)
+			goto unwind;
+		msi_index++;
+	}
 	return 0;
 
 unwind:
+	while (--rdmaciqqidx >= 0)
+		free_irq(adap->msix_info[--msi_index].vec,
+			 &s->rdmaciq[rdmaciqqidx].rspq);
 	while (--rdmaqidx >= 0)
 		free_irq(adap->msix_info[--msi_index].vec,
 			 &s->rdmarxq[rdmaqidx].rspq);
@@ -885,6 +902,8 @@ static void free_msix_queue_irqs(struct adapter *adap)
 		free_irq(adap->msix_info[msi_index++].vec, &s->ofldrxq[i].rspq);
 	for_each_rdmarxq(s, i)
 		free_irq(adap->msix_info[msi_index++].vec, &s->rdmarxq[i].rspq);
+	for_each_rdmaciq(s, i)
+		free_irq(adap->msix_info[msi_index++].vec, &s->rdmaciq[i].rspq);
 }
 
 /**
@@ -1047,7 +1066,8 @@ freeout:	t4_free_sge_resources(adap);
 		if (msi_idx > 0)
 			msi_idx++;
 		err = t4_sge_alloc_rxq(adap, &q->rspq, false, dev, msi_idx,
-				       &q->fl, uldrx_handler);
+				       q->fl.size ? &q->fl : NULL,
+				       uldrx_handler);
 		if (err)
 			goto freeout;
 		memset(&q->stats, 0, sizeof(q->stats));
@@ -1064,13 +1084,28 @@ freeout:	t4_free_sge_resources(adap);
 		if (msi_idx > 0)
 			msi_idx++;
 		err = t4_sge_alloc_rxq(adap, &q->rspq, false, adap->port[i],
-				       msi_idx, &q->fl, uldrx_handler);
+				       msi_idx, q->fl.size ? &q->fl : NULL,
+				       uldrx_handler);
 		if (err)
 			goto freeout;
 		memset(&q->stats, 0, sizeof(q->stats));
 		s->rdma_rxq[i] = q->rspq.abs_id;
 	}
 
+	for_each_rdmaciq(s, i) {
+		struct sge_ofld_rxq *q = &s->rdmaciq[i];
+
+		if (msi_idx > 0)
+			msi_idx++;
+		err = t4_sge_alloc_rxq(adap, &q->rspq, false, adap->port[i],
+				       msi_idx, q->fl.size ? &q->fl : NULL,
+				       uldrx_handler);
+		if (err)
+			goto freeout;
+		memset(&q->stats, 0, sizeof(q->stats));
+		s->rdma_ciq[i] = q->rspq.abs_id;
+	}
+
 	for_each_port(adap, i) {
 		/*
 		 * Note that ->rdmarxq[i].rspq.cntxt_id below is 0 if we don't
@@ -2252,12 +2287,19 @@ static int get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	else if (p->port_type == FW_PORT_TYPE_FIBER_XFI ||
 		 p->port_type == FW_PORT_TYPE_FIBER_XAUI)
 		cmd->port = PORT_FIBRE;
-	else if (p->port_type == FW_PORT_TYPE_SFP) {
-		if (p->mod_type == FW_PORT_MOD_TYPE_TWINAX_PASSIVE ||
-		    p->mod_type == FW_PORT_MOD_TYPE_TWINAX_ACTIVE)
+	else if (p->port_type == FW_PORT_TYPE_SFP ||
+		 p->port_type == FW_PORT_TYPE_QSFP_10G ||
+		 p->port_type == FW_PORT_TYPE_QSFP) {
+		if (p->mod_type == FW_PORT_MOD_TYPE_LR ||
+		    p->mod_type == FW_PORT_MOD_TYPE_SR ||
+		    p->mod_type == FW_PORT_MOD_TYPE_ER ||
+		    p->mod_type == FW_PORT_MOD_TYPE_LRM)
+			cmd->port = PORT_FIBRE;
+		else if (p->mod_type == FW_PORT_MOD_TYPE_TWINAX_PASSIVE ||
+			 p->mod_type == FW_PORT_MOD_TYPE_TWINAX_ACTIVE)
 			cmd->port = PORT_DA;
 		else
-			cmd->port = PORT_FIBRE;
+			cmd->port = PORT_OTHER;
 	} else
 		cmd->port = PORT_OTHER;
 
@@ -2461,8 +2503,7 @@ static unsigned int qtimer_val(const struct adapter *adap,
 }
 
 /**
- *	set_rxq_intr_params - set a queue's interrupt holdoff parameters
- *	@adap: the adapter
+ *	set_rspq_intr_params - set a queue's interrupt holdoff parameters
  *	@q: the Rx queue
  *	@us: the hold-off time in us, or 0 to disable timer
  *	@cnt: the hold-off packet count, or 0 to disable counter
@@ -2470,9 +2511,11 @@ static unsigned int qtimer_val(const struct adapter *adap,
  *	Sets an Rx queue's interrupt hold-off time and packet count.  At least
  *	one of the two needs to be enabled for the queue to generate interrupts.
  */
-static int set_rxq_intr_params(struct adapter *adap, struct sge_rspq *q,
-			       unsigned int us, unsigned int cnt)
+static int set_rspq_intr_params(struct sge_rspq *q,
+				unsigned int us, unsigned int cnt)
 {
+	struct adapter *adap = q->adap;
+
 	if ((us | cnt) == 0)
 		cnt = 1;
 
@@ -2499,24 +2542,34 @@ static int set_rxq_intr_params(struct adapter *adap, struct sge_rspq *q,
 	return 0;
 }
 
-static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+/**
+ * set_rx_intr_params - set a net devices's RX interrupt holdoff paramete!
+ * @dev: the network device
+ * @us: the hold-off time in us, or 0 to disable timer
+ * @cnt: the hold-off packet count, or 0 to disable counter
+ *
+ * Set the RX interrupt hold-off parameters for a network device.
+ */
+static int set_rx_intr_params(struct net_device *dev,
+			      unsigned int us, unsigned int cnt)
 {
-	const struct port_info *pi = netdev_priv(dev);
+	int i, err;
+	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adap = pi->adapter;
-	struct sge_rspq *q;
-	int i;
-	int r = 0;
-
-	for (i = pi->first_qset; i < pi->first_qset + pi->nqsets; i++) {
-		q = &adap->sge.ethrxq[i].rspq;
-		r = set_rxq_intr_params(adap, q, c->rx_coalesce_usecs,
-			c->rx_max_coalesced_frames);
-		if (r) {
-			dev_err(&dev->dev, "failed to set coalesce %d\n", r);
-			break;
-		}
+	struct sge_eth_rxq *q = &adap->sge.ethrxq[pi->first_qset];
+
+	for (i = 0; i < pi->nqsets; i++, q++) {
+		err = set_rspq_intr_params(&q->rspq, us, cnt);
+		if (err)
+			return err;
 	}
-	return r;
+	return 0;
+}
+
+static int set_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
+{
+	return set_rx_intr_params(dev, c->rx_coalesce_usecs,
+				  c->rx_max_coalesced_frames);
 }
 
 static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
@@ -2732,7 +2785,7 @@ static u32 get_rss_table_size(struct net_device *dev)
 	return pi->rss_size;
 }
 
-static int get_rss_table(struct net_device *dev, u32 *p)
+static int get_rss_table(struct net_device *dev, u32 *p, u8 *key)
 {
 	const struct port_info *pi = netdev_priv(dev);
 	unsigned int n = pi->rss_size;
@@ -2742,7 +2795,7 @@ static int get_rss_table(struct net_device *dev, u32 *p)
 	return 0;
 }
 
-static int set_rss_table(struct net_device *dev, const u32 *p)
+static int set_rss_table(struct net_device *dev, const u32 *p, const u8 *key)
 {
 	unsigned int i;
 	struct port_info *pi = netdev_priv(dev);
@@ -2844,8 +2897,8 @@ static const struct ethtool_ops cxgb_ethtool_ops = {
 	.set_wol           = set_wol,
 	.get_rxnfc         = get_rxnfc,
 	.get_rxfh_indir_size = get_rss_table_size,
-	.get_rxfh_indir    = get_rss_table,
-	.set_rxfh_indir    = set_rss_table,
+	.get_rxfh	   = get_rss_table,
+	.set_rxfh	   = set_rss_table,
 	.flash_device      = set_flash,
 };
 
@@ -3386,6 +3439,77 @@ unsigned int cxgb4_best_mtu(const unsigned short *mtus, unsigned short mtu,
 EXPORT_SYMBOL(cxgb4_best_mtu);
 
 /**
+ *     cxgb4_best_aligned_mtu - find best MTU, [hopefully] data size aligned
+ *     @mtus: the HW MTU table
+ *     @header_size: Header Size
+ *     @data_size_max: maximum Data Segment Size
+ *     @data_size_align: desired Data Segment Size Alignment (2^N)
+ *     @mtu_idxp: HW MTU Table Index return value pointer (possibly NULL)
+ *
+ *     Similar to cxgb4_best_mtu() but instead of searching the Hardware
+ *     MTU Table based solely on a Maximum MTU parameter, we break that
+ *     parameter up into a Header Size and Maximum Data Segment Size, and
+ *     provide a desired Data Segment Size Alignment.  If we find an MTU in
+ *     the Hardware MTU Table which will result in a Data Segment Size with
+ *     the requested alignment _and_ that MTU isn't "too far" from the
+ *     closest MTU, then we'll return that rather than the closest MTU.
+ */
+unsigned int cxgb4_best_aligned_mtu(const unsigned short *mtus,
+				    unsigned short header_size,
+				    unsigned short data_size_max,
+				    unsigned short data_size_align,
+				    unsigned int *mtu_idxp)
+{
+	unsigned short max_mtu = header_size + data_size_max;
+	unsigned short data_size_align_mask = data_size_align - 1;
+	int mtu_idx, aligned_mtu_idx;
+
+	/* Scan the MTU Table till we find an MTU which is larger than our
+	 * Maximum MTU or we reach the end of the table.  Along the way,
+	 * record the last MTU found, if any, which will result in a Data
+	 * Segment Length matching the requested alignment.
+	 */
+	for (mtu_idx = 0, aligned_mtu_idx = -1; mtu_idx < NMTUS; mtu_idx++) {
+		unsigned short data_size = mtus[mtu_idx] - header_size;
+
+		/* If this MTU minus the Header Size would result in a
+		 * Data Segment Size of the desired alignment, remember it.
+		 */
+		if ((data_size & data_size_align_mask) == 0)
+			aligned_mtu_idx = mtu_idx;
+
+		/* If we're not at the end of the Hardware MTU Table and the
+		 * next element is larger than our Maximum MTU, drop out of
+		 * the loop.
+		 */
+		if (mtu_idx+1 < NMTUS && mtus[mtu_idx+1] > max_mtu)
+			break;
+	}
+
+	/* If we fell out of the loop because we ran to the end of the table,
+	 * then we just have to use the last [largest] entry.
+	 */
+	if (mtu_idx == NMTUS)
+		mtu_idx--;
+
+	/* If we found an MTU which resulted in the requested Data Segment
+	 * Length alignment and that's "not far" from the largest MTU which is
+	 * less than or equal to the maximum MTU, then use that.
+	 */
+	if (aligned_mtu_idx >= 0 &&
+	    mtu_idx - aligned_mtu_idx <= 1)
+		mtu_idx = aligned_mtu_idx;
+
+	/* If the caller has passed in an MTU Index pointer, pass the
+	 * MTU Index back.  Return the MTU value.
+	 */
+	if (mtu_idxp)
+		*mtu_idxp = mtu_idx;
+	return mtus[mtu_idx];
+}
+EXPORT_SYMBOL(cxgb4_best_aligned_mtu);
+
+/**
  *	cxgb4_port_chan - get the HW channel of a port
  *	@dev: the net device for the port
  *
@@ -3782,7 +3906,9 @@ static void uld_attach(struct adapter *adap, unsigned int uld)
 	lli.mtus = adap->params.mtus;
 	if (uld == CXGB4_ULD_RDMA) {
 		lli.rxq_ids = adap->sge.rdma_rxq;
+		lli.ciq_ids = adap->sge.rdma_ciq;
 		lli.nrxq = adap->sge.rdmaqs;
+		lli.nciq = adap->sge.rdmaciqs;
 	} else if (uld == CXGB4_ULD_ISCSI) {
 		lli.rxq_ids = adap->sge.ofld_rxq;
 		lli.nrxq = adap->sge.ofldqsets;
@@ -3931,22 +4057,19 @@ int cxgb4_unregister_uld(enum cxgb4_uld type)
 EXPORT_SYMBOL(cxgb4_unregister_uld);
 
 /* Check if netdev on which event is occured belongs to us or not. Return
- * suceess (1) if it belongs otherwise failure (0).
+ * success (true) if it belongs otherwise failure (false).
+ * Called with rcu_read_lock() held.
  */
-static int cxgb4_netdev(struct net_device *netdev)
+static bool cxgb4_netdev(const struct net_device *netdev)
 {
 	struct adapter *adap;
 	int i;
 
-	spin_lock(&adap_rcu_lock);
 	list_for_each_entry_rcu(adap, &adap_rcu_list, rcu_node)
 		for (i = 0; i < MAX_NPORTS; i++)
-			if (adap->port[i] == netdev) {
-				spin_unlock(&adap_rcu_lock);
-				return 1;
-			}
-	spin_unlock(&adap_rcu_lock);
-	return 0;
+			if (adap->port[i] == netdev)
+				return true;
+	return false;
 }
 
 static int clip_add(struct net_device *event_dev, struct inet6_ifaddr *ifa,
@@ -4061,7 +4184,7 @@ static int update_root_dev_clip(struct net_device *dev)
 
 	/* Parse all bond and vlan devices layered on top of the physical dev */
 	for (i = 0; i < VLAN_N_VID; i++) {
-		root_dev = __vlan_find_dev_deep(dev, htons(ETH_P_8021Q), i);
+		root_dev = __vlan_find_dev_deep_rcu(dev, htons(ETH_P_8021Q), i);
 		if (!root_dev)
 			continue;
 
@@ -5528,13 +5651,41 @@ static int adap_init0(struct adapter *adap)
 #undef FW_PARAM_PFVF
 #undef FW_PARAM_DEV
 
-	/*
-	 * These are finalized by FW initialization, load their values now.
+	/* The MTU/MSS Table is initialized by now, so load their values.  If
+	 * we're initializing the adapter, then we'll make any modifications
+	 * we want to the MTU/MSS Table and also initialize the congestion
+	 * parameters.
 	 */
 	t4_read_mtu_tbl(adap, adap->params.mtus, NULL);
-	t4_load_mtus(adap, adap->params.mtus, adap->params.a_wnd,
-		     adap->params.b_wnd);
+	if (state != DEV_STATE_INIT) {
+		int i;
 
+		/* The default MTU Table contains values 1492 and 1500.
+		 * However, for TCP, it's better to have two values which are
+		 * a multiple of 8 +/- 4 bytes apart near this popular MTU.
+		 * This allows us to have a TCP Data Payload which is a
+		 * multiple of 8 regardless of what combination of TCP Options
+		 * are in use (always a multiple of 4 bytes) which is
+		 * important for performance reasons.  For instance, if no
+		 * options are in use, then we have a 20-byte IP header and a
+		 * 20-byte TCP header.  In this case, a 1500-byte MSS would
+		 * result in a TCP Data Payload of 1500 - 40 == 1460 bytes
+		 * which is not a multiple of 8.  So using an MSS of 1488 in
+		 * this case results in a TCP Data Payload of 1448 bytes which
+		 * is a multiple of 8.  On the other hand, if 12-byte TCP Time
+		 * Stamps have been negotiated, then an MTU of 1500 bytes
+		 * results in a TCP Data Payload of 1448 bytes which, as
+		 * above, is a multiple of 8 bytes ...
+		 */
+		for (i = 0; i < NMTUS; i++)
+			if (adap->params.mtus[i] == 1492) {
+				adap->params.mtus[i] = 1488;
+				break;
+			}
+
+		t4_load_mtus(adap, adap->params.mtus, adap->params.a_wnd,
+			     adap->params.b_wnd);
+	}
 	t4_init_tp_params(adap);
 	adap->flags |= FW_OK;
 	return 0;
@@ -5669,12 +5820,12 @@ static inline bool is_x_10g_port(const struct link_config *lc)
 	       (lc->supported & FW_PORT_CAP_SPEED_40G) != 0;
 }
 
-static inline void init_rspq(struct sge_rspq *q, u8 timer_idx, u8 pkt_cnt_idx,
+static inline void init_rspq(struct adapter *adap, struct sge_rspq *q,
+			     unsigned int us, unsigned int cnt,
 			     unsigned int size, unsigned int iqe_size)
 {
-	q->intr_params = QINTR_TIMER_IDX(timer_idx) |
-			 (pkt_cnt_idx < SGE_NCOUNTERS ? QINTR_CNT_EN : 0);
-	q->pktcnt_idx = pkt_cnt_idx < SGE_NCOUNTERS ? pkt_cnt_idx : 0;
+	q->adap = adap;
+	set_rspq_intr_params(q, us, cnt);
 	q->iqe_len = iqe_size;
 	q->size = size;
 }
@@ -5688,6 +5839,7 @@ static void cfg_queues(struct adapter *adap)
 {
 	struct sge *s = &adap->sge;
 	int i, q10g = 0, n10g = 0, qidx = 0;
+	int ciq_size;
 
 	for_each_port(adap, i)
 		n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg);
@@ -5726,12 +5878,13 @@ static void cfg_queues(struct adapter *adap)
 			s->ofldqsets = adap->params.nports;
 		/* For RDMA one Rx queue per channel suffices */
 		s->rdmaqs = adap->params.nports;
+		s->rdmaciqs = adap->params.nports;
 	}
 
 	for (i = 0; i < ARRAY_SIZE(s->ethrxq); i++) {
 		struct sge_eth_rxq *r = &s->ethrxq[i];
 
-		init_rspq(&r->rspq, 0, 0, 1024, 64);
+		init_rspq(adap, &r->rspq, 5, 10, 1024, 64);
 		r->fl.size = 72;
 	}
 
@@ -5747,7 +5900,7 @@ static void cfg_queues(struct adapter *adap)
 	for (i = 0; i < ARRAY_SIZE(s->ofldrxq); i++) {
 		struct sge_ofld_rxq *r = &s->ofldrxq[i];
 
-		init_rspq(&r->rspq, 0, 0, 1024, 64);
+		init_rspq(adap, &r->rspq, 5, 1, 1024, 64);
 		r->rspq.uld = CXGB4_ULD_ISCSI;
 		r->fl.size = 72;
 	}
@@ -5755,13 +5908,26 @@ static void cfg_queues(struct adapter *adap)
 	for (i = 0; i < ARRAY_SIZE(s->rdmarxq); i++) {
 		struct sge_ofld_rxq *r = &s->rdmarxq[i];
 
-		init_rspq(&r->rspq, 0, 0, 511, 64);
+		init_rspq(adap, &r->rspq, 5, 1, 511, 64);
 		r->rspq.uld = CXGB4_ULD_RDMA;
 		r->fl.size = 72;
 	}
 
-	init_rspq(&s->fw_evtq, 6, 0, 512, 64);
-	init_rspq(&s->intrq, 6, 0, 2 * MAX_INGQ, 64);
+	ciq_size = 64 + adap->vres.cq.size + adap->tids.nftids;
+	if (ciq_size > SGE_MAX_IQ_SIZE) {
+		CH_WARN(adap, "CIQ size too small for available IQs\n");
+		ciq_size = SGE_MAX_IQ_SIZE;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(s->rdmaciq); i++) {
+		struct sge_ofld_rxq *r = &s->rdmaciq[i];
+
+		init_rspq(adap, &r->rspq, 5, 1, ciq_size, 64);
+		r->rspq.uld = CXGB4_ULD_RDMA;
+	}
+
+	init_rspq(adap, &s->fw_evtq, 0, 1, 1024, 64);
+	init_rspq(adap, &s->intrq, 0, 1, 2 * MAX_INGQ, 64);
 }
 
 /*
@@ -5808,9 +5974,9 @@ static int enable_msix(struct adapter *adap)
 
 	want = s->max_ethqsets + EXTRA_VECS;
 	if (is_offload(adap)) {
-		want += s->rdmaqs + s->ofldqsets;
+		want += s->rdmaqs + s->rdmaciqs + s->ofldqsets;
 		/* need nchan for each possible ULD */
-		ofld_need = 2 * nchan;
+		ofld_need = 3 * nchan;
 	}
 	need = adap->params.nports + EXTRA_VECS + ofld_need;
 
@@ -6076,7 +6242,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->priv_flags |= IFF_UNICAST_FLT;
 
 		netdev->netdev_ops = &cxgb4_netdev_ops;
-		SET_ETHTOOL_OPS(netdev, &cxgb_ethtool_ops);
+		netdev->ethtool_ops = &cxgb_ethtool_ops;
 	}
 
 	pci_set_drvdata(pdev, adapter);
@@ -6227,6 +6393,7 @@ static void remove_one(struct pci_dev *pdev)
 			adapter->flags &= ~DEV_ENABLED;
 		}
 		pci_release_regions(pdev);
+		synchronize_rcu();
 		kfree(adapter);
 	} else
 		pci_release_regions(pdev);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index e274a047528f..55e9daf7f9d4 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -232,8 +232,10 @@ struct cxgb4_lld_info {
 	const struct cxgb4_virt_res *vr;     /* assorted HW resources */
 	const unsigned short *mtus;          /* MTU table */
 	const unsigned short *rxq_ids;       /* the ULD's Rx queue ids */
+	const unsigned short *ciq_ids;       /* the ULD's concentrator IQ ids */
 	unsigned short nrxq;                 /* # of Rx queues */
 	unsigned short ntxq;                 /* # of Tx queues */
+	unsigned short nciq;		     /* # of concentrator IQ */
 	unsigned char nchan:4;               /* # of channels */
 	unsigned char nports:4;              /* # of ports */
 	unsigned char wr_cred;               /* WR 16-byte credits */
@@ -274,6 +276,11 @@ unsigned int cxgb4_port_viid(const struct net_device *dev);
 unsigned int cxgb4_port_idx(const struct net_device *dev);
 unsigned int cxgb4_best_mtu(const unsigned short *mtus, unsigned short mtu,
 			    unsigned int *idx);
+unsigned int cxgb4_best_aligned_mtu(const unsigned short *mtus,
+				    unsigned short header_size,
+				    unsigned short data_size_max,
+				    unsigned short data_size_align,
+				    unsigned int *mtu_idxp);
 void cxgb4_get_tcp_stats(struct pci_dev *pdev, struct tp_tcp_stats *v4,
 			 struct tp_tcp_stats *v6);
 void cxgb4_iscsi_init(struct net_device *dev, unsigned int tag_mask,
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index e249528c8e60..dd4355d248e4 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -1697,7 +1697,8 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 		return handle_trace_pkt(q->adap, si);
 
 	pkt = (const struct cpl_rx_pkt *)rsp;
-	csum_ok = pkt->csum_calc && !pkt->err_vec;
+	csum_ok = pkt->csum_calc && !pkt->err_vec &&
+		  (q->netdev->features & NETIF_F_RXCSUM);
 	if ((pkt->l2info & htonl(RXF_TCP)) &&
 	    (q->netdev->features & NETIF_F_GRO) && csum_ok && !pkt->ip_frag) {
 		do_gro(rxq, si, pkt);
@@ -1720,8 +1721,7 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 
 	rxq->stats.pkts++;
 
-	if (csum_ok && (q->netdev->features & NETIF_F_RXCSUM) &&
-	    (pkt->l2info & htonl(RXF_UDP | RXF_TCP))) {
+	if (csum_ok && (pkt->l2info & htonl(RXF_UDP | RXF_TCP))) {
 		if (!pkt->ip_frag) {
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 			rxq->stats.rx_cso++;
@@ -2215,7 +2215,6 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 	iq->cntxt_id = ntohs(c.iqid);
 	iq->abs_id = ntohs(c.physiqid);
 	iq->size--;                           /* subtract status entry */
-	iq->adap = adap;
 	iq->netdev = dev;
 	iq->handler = hnd;
 
@@ -2515,6 +2514,10 @@ void t4_free_sge_resources(struct adapter *adap)
 		if (oq->rspq.desc)
 			free_rspq_fl(adap, &oq->rspq, &oq->fl);
 	}
+	for (i = 0, oq = adap->sge.rdmaciq; i < adap->sge.rdmaciqs; i++, oq++) {
+		if (oq->rspq.desc)
+			free_rspq_fl(adap, &oq->rspq, &oq->fl);
+	}
 
 	/* clean up offload Tx queues */
 	for (i = 0; i < ARRAY_SIZE(adap->sge.ofldtxq); i++) {
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index bba67681aeaa..931478e7bd28 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -3962,6 +3962,7 @@ int t4_port_init(struct adapter *adap, int mbox, int pf, int vf)
 		p->lport = j;
 		p->rss_size = rss_size;
 		memcpy(adap->port[i]->dev_addr, addr, ETH_ALEN);
+		adap->port[i]->dev_port = j;
 
 		ret = ntohl(c.u.info.lstatus_to_modtype);
 		p->mdio_addr = (ret & FW_PORT_CMD_MDIOCAP) ?
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
index 1d1623be9f1e..71b799b5b0f4 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
@@ -68,6 +68,7 @@ enum {
 	SGE_MAX_WR_LEN = 512,     /* max WR size in bytes */
 	SGE_NTIMERS = 6,          /* # of interrupt holdoff timer values */
 	SGE_NCOUNTERS = 4,        /* # of interrupt packet counter values */
+	SGE_MAX_IQ_SIZE = 65520,
 
 	SGE_TIMER_RSTRT_CNTR = 6, /* restart RX packet threshold counter */
 	SGE_TIMER_UPD_CIDX = 7,   /* update cidx only */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
index f2738c710789..973eb11aa98a 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
@@ -227,6 +227,7 @@ struct cpl_pass_open_req {
 #define DELACK(x)     ((x) << 5)
 #define ULP_MODE(x)   ((x) << 8)
 #define RCV_BUFSIZ(x) ((x) << 12)
+#define RCV_BUFSIZ_MASK 0x3FFU
 #define DSCP(x)       ((x) << 22)
 #define SMAC_SEL(x)   ((u64)(x) << 28)
 #define L2T_IDX(x)    ((u64)(x) << 36)
@@ -278,6 +279,15 @@ struct cpl_pass_accept_rpl {
 	__be64 opt0;
 };
 
+struct cpl_t5_pass_accept_rpl {
+	WR_HDR;
+	union opcode_tid ot;
+	__be32 opt2;
+	__be64 opt0;
+	__be32 iss;
+	__be32 rsvd;
+};
+
 struct cpl_act_open_req {
 	WR_HDR;
 	union opcode_tid ot;
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 52859288de7b..ff1cdd1788b5 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -2664,7 +2664,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 		netdev->priv_flags |= IFF_UNICAST_FLT;
 
 		netdev->netdev_ops = &cxgb4vf_netdev_ops;
-		SET_ETHTOOL_OPS(netdev, &cxgb4vf_ethtool_ops);
+		netdev->ethtool_ops = &cxgb4vf_ethtool_ops;
 
 		/*
 		 * Initialize the hardware/software state for the port.
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index 9d88c1d50b49..bdfa80ca5e31 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -1510,7 +1510,8 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp,
 {
 	struct sk_buff *skb;
 	const struct cpl_rx_pkt *pkt = (void *)rsp;
-	bool csum_ok = pkt->csum_calc && !pkt->err_vec;
+	bool csum_ok = pkt->csum_calc && !pkt->err_vec &&
+		       (rspq->netdev->features & NETIF_F_RXCSUM);
 	struct sge_eth_rxq *rxq = container_of(rspq, struct sge_eth_rxq, rspq);
 
 	/*
@@ -1538,8 +1539,8 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp,
 	skb_record_rx_queue(skb, rspq->idx);
 	rxq->stats.pkts++;
 
-	if (csum_ok && (rspq->netdev->features & NETIF_F_RXCSUM) &&
-	    !pkt->err_vec && (be32_to_cpu(pkt->l2info) & (RXF_UDP|RXF_TCP))) {
+	if (csum_ok && !pkt->err_vec &&
+	    (be32_to_cpu(pkt->l2info) & (RXF_UDP|RXF_TCP))) {
 		if (!pkt->ip_frag)
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else {
diff --git a/drivers/net/ethernet/cisco/enic/enic.h b/drivers/net/ethernet/cisco/enic/enic.h
index e35c8e0202ad..14f465f239d6 100644
--- a/drivers/net/ethernet/cisco/enic/enic.h
+++ b/drivers/net/ethernet/cisco/enic/enic.h
@@ -43,6 +43,8 @@
 #define ENIC_CQ_MAX		(ENIC_WQ_MAX + ENIC_RQ_MAX)
 #define ENIC_INTR_MAX		(ENIC_CQ_MAX + 2)
 
+#define ENIC_AIC_LARGE_PKT_DIFF	3
+
 struct enic_msix_entry {
 	int requested;
 	char devname[IFNAMSIZ];
@@ -50,6 +52,33 @@ struct enic_msix_entry {
 	void *devid;
 };
 
+/* Store only the lower range.  Higher range is given by fw. */
+struct enic_intr_mod_range {
+	u32 small_pkt_range_start;
+	u32 large_pkt_range_start;
+};
+
+struct enic_intr_mod_table {
+	u32 rx_rate;
+	u32 range_percent;
+};
+
+#define ENIC_MAX_LINK_SPEEDS		3
+#define ENIC_LINK_SPEED_10G		10000
+#define ENIC_LINK_SPEED_4G		4000
+#define ENIC_LINK_40G_INDEX		2
+#define ENIC_LINK_10G_INDEX		1
+#define ENIC_LINK_4G_INDEX		0
+#define ENIC_RX_COALESCE_RANGE_END	125
+#define ENIC_AIC_TS_BREAK		100
+
+struct enic_rx_coal {
+	u32 small_pkt_range_start;
+	u32 large_pkt_range_start;
+	u32 range_end;
+	u32 use_adaptive_rx_coalesce;
+};
+
 /* priv_flags */
 #define ENIC_SRIOV_ENABLED		(1 << 0)
 
@@ -85,13 +114,12 @@ struct enic {
 	u32 msg_enable;
 	spinlock_t devcmd_lock;
 	u8 mac_addr[ETH_ALEN];
-	u8 mc_addr[ENIC_MULTICAST_PERFECT_FILTERS][ETH_ALEN];
-	u8 uc_addr[ENIC_UNICAST_PERFECT_FILTERS][ETH_ALEN];
 	unsigned int flags;
 	unsigned int priv_flags;
 	unsigned int mc_count;
 	unsigned int uc_count;
 	u32 port_mtu;
+	struct enic_rx_coal rx_coalesce_setting;
 	u32 rx_coalesce_usecs;
 	u32 tx_coalesce_usecs;
 #ifdef CONFIG_PCI_IOV
diff --git a/drivers/net/ethernet/cisco/enic/enic_dev.c b/drivers/net/ethernet/cisco/enic/enic_dev.c
index 4b6e5695b263..3e27df522847 100644
--- a/drivers/net/ethernet/cisco/enic/enic_dev.c
+++ b/drivers/net/ethernet/cisco/enic/enic_dev.c
@@ -88,7 +88,7 @@ int enic_dev_packet_filter(struct enic *enic, int directed, int multicast,
 	return err;
 }
 
-int enic_dev_add_addr(struct enic *enic, u8 *addr)
+int enic_dev_add_addr(struct enic *enic, const u8 *addr)
 {
 	int err;
 
@@ -99,7 +99,7 @@ int enic_dev_add_addr(struct enic *enic, u8 *addr)
 	return err;
 }
 
-int enic_dev_del_addr(struct enic *enic, u8 *addr)
+int enic_dev_del_addr(struct enic *enic, const u8 *addr)
 {
 	int err;
 
diff --git a/drivers/net/ethernet/cisco/enic/enic_dev.h b/drivers/net/ethernet/cisco/enic/enic_dev.h
index 129b14a4efb0..36ea1ab25f6a 100644
--- a/drivers/net/ethernet/cisco/enic/enic_dev.h
+++ b/drivers/net/ethernet/cisco/enic/enic_dev.h
@@ -45,8 +45,8 @@ int enic_dev_add_station_addr(struct enic *enic);
 int enic_dev_del_station_addr(struct enic *enic);
 int enic_dev_packet_filter(struct enic *enic, int directed, int multicast,
 	int broadcast, int promisc, int allmulti);
-int enic_dev_add_addr(struct enic *enic, u8 *addr);
-int enic_dev_del_addr(struct enic *enic, u8 *addr);
+int enic_dev_add_addr(struct enic *enic, const u8 *addr);
+int enic_dev_del_addr(struct enic *enic, const u8 *addr);
 int enic_vlan_rx_add_vid(struct net_device *netdev, __be16 proto, u16 vid);
 int enic_vlan_rx_kill_vid(struct net_device *netdev, __be16 proto, u16 vid);
 int enic_dev_notify_unset(struct enic *enic);
diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
index 47e3562f4866..2e50b5489d20 100644
--- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c
+++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
@@ -79,6 +79,17 @@ static const struct enic_stat enic_rx_stats[] = {
 static const unsigned int enic_n_tx_stats = ARRAY_SIZE(enic_tx_stats);
 static const unsigned int enic_n_rx_stats = ARRAY_SIZE(enic_rx_stats);
 
+void enic_intr_coal_set_rx(struct enic *enic, u32 timer)
+{
+	int i;
+	int intr;
+
+	for (i = 0; i < enic->rq_count; i++) {
+		intr = enic_msix_rq_intr(enic, i);
+		vnic_intr_coalescing_timer_set(&enic->intr[intr], timer);
+	}
+}
+
 static int enic_get_settings(struct net_device *netdev,
 	struct ethtool_cmd *ecmd)
 {
@@ -93,8 +104,8 @@ static int enic_get_settings(struct net_device *netdev,
 		ethtool_cmd_speed_set(ecmd, vnic_dev_port_speed(enic->vdev));
 		ecmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = AUTONEG_DISABLE;
@@ -178,9 +189,14 @@ static int enic_get_coalesce(struct net_device *netdev,
 	struct ethtool_coalesce *ecmd)
 {
 	struct enic *enic = netdev_priv(netdev);
+	struct enic_rx_coal *rxcoal = &enic->rx_coalesce_setting;
 
 	ecmd->tx_coalesce_usecs = enic->tx_coalesce_usecs;
 	ecmd->rx_coalesce_usecs = enic->rx_coalesce_usecs;
+	if (rxcoal->use_adaptive_rx_coalesce)
+		ecmd->use_adaptive_rx_coalesce = 1;
+	ecmd->rx_coalesce_usecs_low = rxcoal->small_pkt_range_start;
+	ecmd->rx_coalesce_usecs_high = rxcoal->range_end;
 
 	return 0;
 }
@@ -191,17 +207,31 @@ static int enic_set_coalesce(struct net_device *netdev,
 	struct enic *enic = netdev_priv(netdev);
 	u32 tx_coalesce_usecs;
 	u32 rx_coalesce_usecs;
+	u32 rx_coalesce_usecs_low;
+	u32 rx_coalesce_usecs_high;
+	u32 coalesce_usecs_max;
 	unsigned int i, intr;
+	struct enic_rx_coal *rxcoal = &enic->rx_coalesce_setting;
 
+	coalesce_usecs_max = vnic_dev_get_intr_coal_timer_max(enic->vdev);
 	tx_coalesce_usecs = min_t(u32, ecmd->tx_coalesce_usecs,
-		vnic_dev_get_intr_coal_timer_max(enic->vdev));
+				  coalesce_usecs_max);
 	rx_coalesce_usecs = min_t(u32, ecmd->rx_coalesce_usecs,
-		vnic_dev_get_intr_coal_timer_max(enic->vdev));
+				  coalesce_usecs_max);
+
+	rx_coalesce_usecs_low = min_t(u32, ecmd->rx_coalesce_usecs_low,
+				      coalesce_usecs_max);
+	rx_coalesce_usecs_high = min_t(u32, ecmd->rx_coalesce_usecs_high,
+				       coalesce_usecs_max);
 
 	switch (vnic_dev_get_intr_mode(enic->vdev)) {
 	case VNIC_DEV_INTR_MODE_INTX:
 		if (tx_coalesce_usecs != rx_coalesce_usecs)
 			return -EINVAL;
+		if (ecmd->use_adaptive_rx_coalesce	||
+		    ecmd->rx_coalesce_usecs_low		||
+		    ecmd->rx_coalesce_usecs_high)
+			return -EOPNOTSUPP;
 
 		intr = enic_legacy_io_intr();
 		vnic_intr_coalescing_timer_set(&enic->intr[intr],
@@ -210,6 +240,10 @@ static int enic_set_coalesce(struct net_device *netdev,
 	case VNIC_DEV_INTR_MODE_MSI:
 		if (tx_coalesce_usecs != rx_coalesce_usecs)
 			return -EINVAL;
+		if (ecmd->use_adaptive_rx_coalesce	||
+		    ecmd->rx_coalesce_usecs_low		||
+		    ecmd->rx_coalesce_usecs_high)
+			return -EOPNOTSUPP;
 
 		vnic_intr_coalescing_timer_set(&enic->intr[0],
 			tx_coalesce_usecs);
@@ -221,12 +255,27 @@ static int enic_set_coalesce(struct net_device *netdev,
 				tx_coalesce_usecs);
 		}
 
-		for (i = 0; i < enic->rq_count; i++) {
-			intr = enic_msix_rq_intr(enic, i);
-			vnic_intr_coalescing_timer_set(&enic->intr[intr],
-				rx_coalesce_usecs);
+		if (rxcoal->use_adaptive_rx_coalesce) {
+			if (!ecmd->use_adaptive_rx_coalesce) {
+				rxcoal->use_adaptive_rx_coalesce = 0;
+				enic_intr_coal_set_rx(enic, rx_coalesce_usecs);
+			}
+		} else {
+			if (ecmd->use_adaptive_rx_coalesce)
+				rxcoal->use_adaptive_rx_coalesce = 1;
+			else
+				enic_intr_coal_set_rx(enic, rx_coalesce_usecs);
 		}
 
+		if (ecmd->rx_coalesce_usecs_high) {
+			if (rx_coalesce_usecs_high <
+			    (rx_coalesce_usecs_low + ENIC_AIC_LARGE_PKT_DIFF))
+				return -EINVAL;
+			rxcoal->range_end = rx_coalesce_usecs_high;
+			rxcoal->small_pkt_range_start = rx_coalesce_usecs_low;
+			rxcoal->large_pkt_range_start = rx_coalesce_usecs_low +
+							ENIC_AIC_LARGE_PKT_DIFF;
+		}
 		break;
 	default:
 		break;
@@ -253,5 +302,5 @@ static const struct ethtool_ops enic_ethtool_ops = {
 
 void enic_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &enic_ethtool_ops);
+	netdev->ethtool_ops = &enic_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
index 2945718ce806..f32f828b7f3d 100644
--- a/drivers/net/ethernet/cisco/enic/enic_main.c
+++ b/drivers/net/ethernet/cisco/enic/enic_main.c
@@ -38,6 +38,7 @@
 #include <linux/rtnetlink.h>
 #include <linux/prefetch.h>
 #include <net/ip6_checksum.h>
+#include <linux/ktime.h>
 
 #include "cq_enet_desc.h"
 #include "vnic_dev.h"
@@ -72,6 +73,35 @@ MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, enic_id_table);
 
+#define ENIC_LARGE_PKT_THRESHOLD		1000
+#define ENIC_MAX_COALESCE_TIMERS		10
+/*  Interrupt moderation table, which will be used to decide the
+ *  coalescing timer values
+ *  {rx_rate in Mbps, mapping percentage of the range}
+ */
+struct enic_intr_mod_table mod_table[ENIC_MAX_COALESCE_TIMERS + 1] = {
+	{4000,  0},
+	{4400, 10},
+	{5060, 20},
+	{5230, 30},
+	{5540, 40},
+	{5820, 50},
+	{6120, 60},
+	{6435, 70},
+	{6745, 80},
+	{7000, 90},
+	{0xFFFFFFFF, 100}
+};
+
+/* This table helps the driver to pick different ranges for rx coalescing
+ * timer depending on the link speed.
+ */
+struct enic_intr_mod_range mod_range[ENIC_MAX_LINK_SPEEDS] = {
+	{0,  0}, /* 0  - 4  Gbps */
+	{0,  3}, /* 4  - 10 Gbps */
+	{3,  6}, /* 10 - 40 Gbps */
+};
+
 int enic_is_dynamic(struct enic *enic)
 {
 	return enic->pdev->device == PCI_DEVICE_ID_CISCO_VIC_ENET_DYN;
@@ -586,8 +616,71 @@ static struct rtnl_link_stats64 *enic_get_stats(struct net_device *netdev,
 	return net_stats;
 }
 
+static int enic_mc_sync(struct net_device *netdev, const u8 *mc_addr)
+{
+	struct enic *enic = netdev_priv(netdev);
+
+	if (enic->mc_count == ENIC_MULTICAST_PERFECT_FILTERS) {
+		unsigned int mc_count = netdev_mc_count(netdev);
+
+		netdev_warn(netdev, "Registering only %d out of %d multicast addresses\n",
+			    ENIC_MULTICAST_PERFECT_FILTERS, mc_count);
+
+		return -ENOSPC;
+	}
+
+	enic_dev_add_addr(enic, mc_addr);
+	enic->mc_count++;
+
+	return 0;
+}
+
+static int enic_mc_unsync(struct net_device *netdev, const u8 *mc_addr)
+{
+	struct enic *enic = netdev_priv(netdev);
+
+	enic_dev_del_addr(enic, mc_addr);
+	enic->mc_count--;
+
+	return 0;
+}
+
+static int enic_uc_sync(struct net_device *netdev, const u8 *uc_addr)
+{
+	struct enic *enic = netdev_priv(netdev);
+
+	if (enic->uc_count == ENIC_UNICAST_PERFECT_FILTERS) {
+		unsigned int uc_count = netdev_uc_count(netdev);
+
+		netdev_warn(netdev, "Registering only %d out of %d unicast addresses\n",
+			    ENIC_UNICAST_PERFECT_FILTERS, uc_count);
+
+		return -ENOSPC;
+	}
+
+	enic_dev_add_addr(enic, uc_addr);
+	enic->uc_count++;
+
+	return 0;
+}
+
+static int enic_uc_unsync(struct net_device *netdev, const u8 *uc_addr)
+{
+	struct enic *enic = netdev_priv(netdev);
+
+	enic_dev_del_addr(enic, uc_addr);
+	enic->uc_count--;
+
+	return 0;
+}
+
 void enic_reset_addr_lists(struct enic *enic)
 {
+	struct net_device *netdev = enic->netdev;
+
+	__dev_uc_unsync(netdev, NULL);
+	__dev_mc_unsync(netdev, NULL);
+
 	enic->mc_count = 0;
 	enic->uc_count = 0;
 	enic->flags = 0;
@@ -654,112 +747,6 @@ static int enic_set_mac_address(struct net_device *netdev, void *p)
 	return enic_dev_add_station_addr(enic);
 }
 
-static void enic_update_multicast_addr_list(struct enic *enic)
-{
-	struct net_device *netdev = enic->netdev;
-	struct netdev_hw_addr *ha;
-	unsigned int mc_count = netdev_mc_count(netdev);
-	u8 mc_addr[ENIC_MULTICAST_PERFECT_FILTERS][ETH_ALEN];
-	unsigned int i, j;
-
-	if (mc_count > ENIC_MULTICAST_PERFECT_FILTERS) {
-		netdev_warn(netdev, "Registering only %d out of %d "
-			"multicast addresses\n",
-			ENIC_MULTICAST_PERFECT_FILTERS, mc_count);
-		mc_count = ENIC_MULTICAST_PERFECT_FILTERS;
-	}
-
-	/* Is there an easier way?  Trying to minimize to
-	 * calls to add/del multicast addrs.  We keep the
-	 * addrs from the last call in enic->mc_addr and
-	 * look for changes to add/del.
-	 */
-
-	i = 0;
-	netdev_for_each_mc_addr(ha, netdev) {
-		if (i == mc_count)
-			break;
-		memcpy(mc_addr[i++], ha->addr, ETH_ALEN);
-	}
-
-	for (i = 0; i < enic->mc_count; i++) {
-		for (j = 0; j < mc_count; j++)
-			if (ether_addr_equal(enic->mc_addr[i], mc_addr[j]))
-				break;
-		if (j == mc_count)
-			enic_dev_del_addr(enic, enic->mc_addr[i]);
-	}
-
-	for (i = 0; i < mc_count; i++) {
-		for (j = 0; j < enic->mc_count; j++)
-			if (ether_addr_equal(mc_addr[i], enic->mc_addr[j]))
-				break;
-		if (j == enic->mc_count)
-			enic_dev_add_addr(enic, mc_addr[i]);
-	}
-
-	/* Save the list to compare against next time
-	 */
-
-	for (i = 0; i < mc_count; i++)
-		memcpy(enic->mc_addr[i], mc_addr[i], ETH_ALEN);
-
-	enic->mc_count = mc_count;
-}
-
-static void enic_update_unicast_addr_list(struct enic *enic)
-{
-	struct net_device *netdev = enic->netdev;
-	struct netdev_hw_addr *ha;
-	unsigned int uc_count = netdev_uc_count(netdev);
-	u8 uc_addr[ENIC_UNICAST_PERFECT_FILTERS][ETH_ALEN];
-	unsigned int i, j;
-
-	if (uc_count > ENIC_UNICAST_PERFECT_FILTERS) {
-		netdev_warn(netdev, "Registering only %d out of %d "
-			"unicast addresses\n",
-			ENIC_UNICAST_PERFECT_FILTERS, uc_count);
-		uc_count = ENIC_UNICAST_PERFECT_FILTERS;
-	}
-
-	/* Is there an easier way?  Trying to minimize to
-	 * calls to add/del unicast addrs.  We keep the
-	 * addrs from the last call in enic->uc_addr and
-	 * look for changes to add/del.
-	 */
-
-	i = 0;
-	netdev_for_each_uc_addr(ha, netdev) {
-		if (i == uc_count)
-			break;
-		memcpy(uc_addr[i++], ha->addr, ETH_ALEN);
-	}
-
-	for (i = 0; i < enic->uc_count; i++) {
-		for (j = 0; j < uc_count; j++)
-			if (ether_addr_equal(enic->uc_addr[i], uc_addr[j]))
-				break;
-		if (j == uc_count)
-			enic_dev_del_addr(enic, enic->uc_addr[i]);
-	}
-
-	for (i = 0; i < uc_count; i++) {
-		for (j = 0; j < enic->uc_count; j++)
-			if (ether_addr_equal(uc_addr[i], enic->uc_addr[j]))
-				break;
-		if (j == enic->uc_count)
-			enic_dev_add_addr(enic, uc_addr[i]);
-	}
-
-	/* Save the list to compare against next time
-	 */
-
-	for (i = 0; i < uc_count; i++)
-		memcpy(enic->uc_addr[i], uc_addr[i], ETH_ALEN);
-
-	enic->uc_count = uc_count;
-}
-
 /* netif_tx_lock held, BHs disabled */
 static void enic_set_rx_mode(struct net_device *netdev)
 {
@@ -782,9 +769,9 @@ static void enic_set_rx_mode(struct net_device *netdev)
 	}
 
 	if (!promisc) {
-		enic_update_unicast_addr_list(enic);
+		__dev_uc_sync(netdev, enic_uc_sync, enic_uc_unsync);
 		if (!allmulti)
-			enic_update_multicast_addr_list(enic);
+			__dev_mc_sync(netdev, enic_mc_sync, enic_mc_unsync);
 	}
 }
 
@@ -979,6 +966,15 @@ static int enic_rq_alloc_buf(struct vnic_rq *rq)
 	return 0;
 }
 
+static void enic_intr_update_pkt_size(struct vnic_rx_bytes_counter *pkt_size,
+				      u32 pkt_len)
+{
+	if (ENIC_LARGE_PKT_THRESHOLD <= pkt_len)
+		pkt_size->large_pkt_bytes_cnt += pkt_len;
+	else
+		pkt_size->small_pkt_bytes_cnt += pkt_len;
+}
+
 static void enic_rq_indicate_buf(struct vnic_rq *rq,
 	struct cq_desc *cq_desc, struct vnic_rq_buf *buf,
 	int skipped, void *opaque)
@@ -986,6 +982,7 @@ static void enic_rq_indicate_buf(struct vnic_rq *rq,
 	struct enic *enic = vnic_dev_priv(rq->vdev);
 	struct net_device *netdev = enic->netdev;
 	struct sk_buff *skb;
+	struct vnic_cq *cq = &enic->cq[enic_cq_rq(enic, rq->index)];
 
 	u8 type, color, eop, sop, ingress_port, vlan_stripped;
 	u8 fcoe, fcoe_sof, fcoe_fc_crc_ok, fcoe_enc_error, fcoe_eof;
@@ -1056,6 +1053,9 @@ static void enic_rq_indicate_buf(struct vnic_rq *rq,
 			napi_gro_receive(&enic->napi[q_number], skb);
 		else
 			netif_receive_skb(skb);
+		if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
+			enic_intr_update_pkt_size(&cq->pkt_size_counter,
+						  bytes_written);
 	} else {
 
 		/* Buffer overflow
@@ -1134,6 +1134,64 @@ static int enic_poll(struct napi_struct *napi, int budget)
 	return rq_work_done;
 }
 
+static void enic_set_int_moderation(struct enic *enic, struct vnic_rq *rq)
+{
+	unsigned int intr = enic_msix_rq_intr(enic, rq->index);
+	struct vnic_cq *cq = &enic->cq[enic_cq_rq(enic, rq->index)];
+	u32 timer = cq->tobe_rx_coal_timeval;
+
+	if (cq->tobe_rx_coal_timeval != cq->cur_rx_coal_timeval) {
+		vnic_intr_coalescing_timer_set(&enic->intr[intr], timer);
+		cq->cur_rx_coal_timeval = cq->tobe_rx_coal_timeval;
+	}
+}
+
+static void enic_calc_int_moderation(struct enic *enic, struct vnic_rq *rq)
+{
+	struct enic_rx_coal *rx_coal = &enic->rx_coalesce_setting;
+	struct vnic_cq *cq = &enic->cq[enic_cq_rq(enic, rq->index)];
+	struct vnic_rx_bytes_counter *pkt_size_counter = &cq->pkt_size_counter;
+	int index;
+	u32 timer;
+	u32 range_start;
+	u32 traffic;
+	u64 delta;
+	ktime_t now = ktime_get();
+
+	delta = ktime_us_delta(now, cq->prev_ts);
+	if (delta < ENIC_AIC_TS_BREAK)
+		return;
+	cq->prev_ts = now;
+
+	traffic = pkt_size_counter->large_pkt_bytes_cnt +
+		  pkt_size_counter->small_pkt_bytes_cnt;
+	/* The table takes Mbps
+	 * traffic *= 8    => bits
+	 * traffic *= (10^6 / delta)    => bps
+	 * traffic /= 10^6     => Mbps
+	 *
+	 * Combining, traffic *= (8 / delta)
+	 */
+
+	traffic <<= 3;
+	traffic = delta > UINT_MAX ? 0 : traffic / (u32)delta;
+
+	for (index = 0; index < ENIC_MAX_COALESCE_TIMERS; index++)
+		if (traffic < mod_table[index].rx_rate)
+			break;
+	range_start = (pkt_size_counter->small_pkt_bytes_cnt >
+		       pkt_size_counter->large_pkt_bytes_cnt << 1) ?
+		      rx_coal->small_pkt_range_start :
+		      rx_coal->large_pkt_range_start;
+	timer = range_start + ((rx_coal->range_end - range_start) *
+			       mod_table[index].range_percent / 100);
+	/* Damping */
+	cq->tobe_rx_coal_timeval = (timer + cq->tobe_rx_coal_timeval) >> 1;
+
+	pkt_size_counter->large_pkt_bytes_cnt = 0;
+	pkt_size_counter->small_pkt_bytes_cnt = 0;
+}
+
 static int enic_poll_msix(struct napi_struct *napi, int budget)
 {
 	struct net_device *netdev = napi->dev;
@@ -1171,6 +1229,13 @@ static int enic_poll_msix(struct napi_struct *napi, int budget)
 
 	if (err)
 		work_done = work_to_do;
+	if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
+		/* Call the function which refreshes
+		 * the intr coalescing timer value based on
+		 * the traffic.  This is supported only in
+		 * the case of MSI-x mode
+		 */
+		enic_calc_int_moderation(enic, &enic->rq[rq]);
 
 	if (work_done < work_to_do) {
 
@@ -1179,6 +1244,8 @@ static int enic_poll_msix(struct napi_struct *napi, int budget)
 		 */
 
 		napi_complete(napi);
+		if (enic->rx_coalesce_setting.use_adaptive_rx_coalesce)
+			enic_set_int_moderation(enic, &enic->rq[rq]);
 		vnic_intr_unmask(&enic->intr[intr]);
 	}
 
@@ -1314,6 +1381,42 @@ static void enic_synchronize_irqs(struct enic *enic)
 	}
 }
 
+static void enic_set_rx_coal_setting(struct enic *enic)
+{
+	unsigned int speed;
+	int index = -1;
+	struct enic_rx_coal *rx_coal = &enic->rx_coalesce_setting;
+
+	/* If intr mode is not MSIX, do not do adaptive coalescing */
+	if (VNIC_DEV_INTR_MODE_MSIX != vnic_dev_get_intr_mode(enic->vdev)) {
+		netdev_info(enic->netdev, "INTR mode is not MSIX, Not initializing adaptive coalescing");
+		return;
+	}
+
+	/* 1. Read the link speed from fw
+	 * 2. Pick the default range for the speed
+	 * 3. Update it in enic->rx_coalesce_setting
+	 */
+	speed = vnic_dev_port_speed(enic->vdev);
+	if (ENIC_LINK_SPEED_10G < speed)
+		index = ENIC_LINK_40G_INDEX;
+	else if (ENIC_LINK_SPEED_4G < speed)
+		index = ENIC_LINK_10G_INDEX;
+	else
+		index = ENIC_LINK_4G_INDEX;
+
+	rx_coal->small_pkt_range_start = mod_range[index].small_pkt_range_start;
+	rx_coal->large_pkt_range_start = mod_range[index].large_pkt_range_start;
+	rx_coal->range_end = ENIC_RX_COALESCE_RANGE_END;
+
+	/* Start with the value provided by UCSM */
+	for (index = 0; index < enic->rq_count; index++)
+		enic->cq[index].cur_rx_coal_timeval =
+				enic->config.intr_timer_usec;
+
+	rx_coal->use_adaptive_rx_coalesce = 1;
+}
+
 static int enic_dev_notify_set(struct enic *enic)
 {
 	int err;
@@ -2231,6 +2334,7 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	enic->notify_timer.function = enic_notify_timer;
 	enic->notify_timer.data = (unsigned long)enic;
 
+	enic_set_rx_coal_setting(enic);
 	INIT_WORK(&enic->reset, enic_reset);
 	INIT_WORK(&enic->change_mtu_work, enic_change_mtu_work);
 
@@ -2250,6 +2354,9 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	enic->tx_coalesce_usecs = enic->config.intr_timer_usec;
+	/* rx coalesce time already got initialized. This gets used
+	 * if adaptive coal is turned off
+	 */
 	enic->rx_coalesce_usecs = enic->tx_coalesce_usecs;
 
 	if (enic_is_dynamic(enic) || enic_is_sriov_vf(enic))
diff --git a/drivers/net/ethernet/cisco/enic/vnic_cq.h b/drivers/net/ethernet/cisco/enic/vnic_cq.h
index 579315cbe803..4e6aa65857f7 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_cq.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_cq.h
@@ -50,6 +50,11 @@ struct vnic_cq_ctrl {
 	u32 pad10;
 };
 
+struct vnic_rx_bytes_counter {
+	unsigned int small_pkt_bytes_cnt;
+	unsigned int large_pkt_bytes_cnt;
+};
+
 struct vnic_cq {
 	unsigned int index;
 	struct vnic_dev *vdev;
@@ -58,6 +63,10 @@ struct vnic_cq {
 	unsigned int to_clean;
 	unsigned int last_color;
 	unsigned int interrupt_offset;
+	struct vnic_rx_bytes_counter pkt_size_counter;
+	unsigned int cur_rx_coal_timeval;
+	unsigned int tobe_rx_coal_timeval;
+	ktime_t prev_ts;
 };
 
 static inline unsigned int vnic_cq_service(struct vnic_cq *cq,
diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.c b/drivers/net/ethernet/cisco/enic/vnic_dev.c
index 69dd92598b7e..e86a45cb9e68 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_dev.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_dev.c
@@ -657,7 +657,7 @@ int vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast,
 	return err;
 }
 
-int vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr)
+int vnic_dev_add_addr(struct vnic_dev *vdev, const u8 *addr)
 {
 	u64 a0 = 0, a1 = 0;
 	int wait = 1000;
@@ -674,7 +674,7 @@ int vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr)
 	return err;
 }
 
-int vnic_dev_del_addr(struct vnic_dev *vdev, u8 *addr)
+int vnic_dev_del_addr(struct vnic_dev *vdev, const u8 *addr)
 {
 	u64 a0 = 0, a1 = 0;
 	int wait = 1000;
diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.h b/drivers/net/ethernet/cisco/enic/vnic_dev.h
index e670029862a1..1f3b301f8225 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_dev.h
+++ b/drivers/net/ethernet/cisco/enic/vnic_dev.h
@@ -95,8 +95,8 @@ int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats);
 int vnic_dev_hang_notify(struct vnic_dev *vdev);
 int vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast,
 	int broadcast, int promisc, int allmulti);
-int vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr);
-int vnic_dev_del_addr(struct vnic_dev *vdev, u8 *addr);
+int vnic_dev_add_addr(struct vnic_dev *vdev, const u8 *addr);
+int vnic_dev_del_addr(struct vnic_dev *vdev, const u8 *addr);
 int vnic_dev_get_mac_addr(struct vnic_dev *vdev, u8 *mac_addr);
 int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr);
 int vnic_dev_notify_unset(struct vnic_dev *vdev);
diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index 8c4b93be333b..13723c96d1a2 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -109,6 +109,7 @@ typedef struct board_info {
 	u8		imr_all;
 
 	unsigned int	flags;
+	unsigned int	in_timeout:1;
 	unsigned int	in_suspend:1;
 	unsigned int	wake_supported:1;
 
@@ -187,13 +188,13 @@ dm9000_reset(board_info_t *db)
 	 * The essential point is that we have to do a double reset, and the
 	 * instruction is to set LBK into MAC internal loopback mode.
 	 */
-	iow(db, DM9000_NCR, 0x03);
+	iow(db, DM9000_NCR, NCR_RST | NCR_MAC_LBK);
 	udelay(100); /* Application note says at least 20 us */
 	if (ior(db, DM9000_NCR) & 1)
 		dev_err(db->dev, "dm9000 did not respond to first reset\n");
 
 	iow(db, DM9000_NCR, 0);
-	iow(db, DM9000_NCR, 0x03);
+	iow(db, DM9000_NCR, NCR_RST | NCR_MAC_LBK);
 	udelay(100);
 	if (ior(db, DM9000_NCR) & 1)
 		dev_err(db->dev, "dm9000 did not respond to second reset\n");
@@ -273,7 +274,7 @@ static void dm9000_dumpblk_32bit(void __iomem *reg, int count)
  */
 static void dm9000_msleep(board_info_t *db, unsigned int ms)
 {
-	if (db->in_suspend)
+	if (db->in_suspend || db->in_timeout)
 		mdelay(ms);
 	else
 		msleep(ms);
@@ -334,7 +335,8 @@ dm9000_phy_write(struct net_device *dev,
 	unsigned long reg_save;
 
 	dm9000_dbg(db, 5, "phy_write[%02x] = %04x\n", reg, value);
-	mutex_lock(&db->addr_lock);
+	if (!db->in_timeout)
+		mutex_lock(&db->addr_lock);
 
 	spin_lock_irqsave(&db->lock, flags);
 
@@ -365,7 +367,8 @@ dm9000_phy_write(struct net_device *dev,
 	writeb(reg_save, db->io_addr);
 
 	spin_unlock_irqrestore(&db->lock, flags);
-	mutex_unlock(&db->addr_lock);
+	if (!db->in_timeout)
+		mutex_unlock(&db->addr_lock);
 }
 
 /* dm9000_set_io
@@ -882,6 +885,18 @@ dm9000_hash_table(struct net_device *dev)
 	spin_unlock_irqrestore(&db->lock, flags);
 }
 
+static void
+dm9000_mask_interrupts(board_info_t *db)
+{
+	iow(db, DM9000_IMR, IMR_PAR);
+}
+
+static void
+dm9000_unmask_interrupts(board_info_t *db)
+{
+	iow(db, DM9000_IMR, db->imr_all);
+}
+
 /*
  * Initialize dm9000 board
  */
@@ -894,6 +909,9 @@ dm9000_init_dm9000(struct net_device *dev)
 
 	dm9000_dbg(db, 1, "entering %s\n", __func__);
 
+	dm9000_reset(db);
+	dm9000_mask_interrupts(db);
+
 	/* I/O mode */
 	db->io_mode = ior(db, DM9000_ISR) >> 6;	/* ISR bit7:6 keeps I/O mode */
 
@@ -941,9 +959,6 @@ dm9000_init_dm9000(struct net_device *dev)
 
 	db->imr_all = imr;
 
-	/* Enable TX/RX interrupt mask */
-	iow(db, DM9000_IMR, imr);
-
 	/* Init Driver variable */
 	db->tx_pkt_cnt = 0;
 	db->queue_pkt_len = 0;
@@ -959,17 +974,19 @@ static void dm9000_timeout(struct net_device *dev)
 
 	/* Save previous register address */
 	spin_lock_irqsave(&db->lock, flags);
+	db->in_timeout = 1;
 	reg_save = readb(db->io_addr);
 
 	netif_stop_queue(dev);
-	dm9000_reset(db);
 	dm9000_init_dm9000(dev);
+	dm9000_unmask_interrupts(db);
 	/* We can accept TX packets again */
 	dev->trans_start = jiffies; /* prevent tx timeout */
 	netif_wake_queue(dev);
 
 	/* Restore previous register address */
 	writeb(reg_save, db->io_addr);
+	db->in_timeout = 0;
 	spin_unlock_irqrestore(&db->lock, flags);
 }
 
@@ -1093,7 +1110,6 @@ dm9000_rx(struct net_device *dev)
 		if (rxbyte & DM9000_PKT_ERR) {
 			dev_warn(db->dev, "status check fail: %d\n", rxbyte);
 			iow(db, DM9000_RCR, 0x00);	/* Stop Device */
-			iow(db, DM9000_ISR, IMR_PAR);	/* Stop INT request */
 			return;
 		}
 
@@ -1193,9 +1209,7 @@ static irqreturn_t dm9000_interrupt(int irq, void *dev_id)
 	/* Save previous register address */
 	reg_save = readb(db->io_addr);
 
-	/* Disable all interrupts */
-	iow(db, DM9000_IMR, IMR_PAR);
-
+	dm9000_mask_interrupts(db);
 	/* Got DM9000 interrupt status */
 	int_status = ior(db, DM9000_ISR);	/* Got ISR */
 	iow(db, DM9000_ISR, int_status);	/* Clear ISR status */
@@ -1218,9 +1232,7 @@ static irqreturn_t dm9000_interrupt(int irq, void *dev_id)
 		}
 	}
 
-	/* Re-enable interrupt mask */
-	iow(db, DM9000_IMR, db->imr_all);
-
+	dm9000_unmask_interrupts(db);
 	/* Restore previous register address */
 	writeb(reg_save, db->io_addr);
 
@@ -1292,6 +1304,9 @@ dm9000_open(struct net_device *dev)
 	 * may work, and tell the user that this is a problem */
 
 	if (irqflags == IRQF_TRIGGER_NONE)
+		irqflags = irq_get_trigger_type(dev->irq);
+
+	if (irqflags == IRQF_TRIGGER_NONE)
 		dev_warn(db->dev, "WARNING: no IRQ resource flags set.\n");
 
 	irqflags |= IRQF_SHARED;
@@ -1301,11 +1316,14 @@ dm9000_open(struct net_device *dev)
 	mdelay(1); /* delay needs by DM9000B */
 
 	/* Initialize DM9000 board */
-	dm9000_reset(db);
 	dm9000_init_dm9000(dev);
 
 	if (request_irq(dev->irq, dm9000_interrupt, irqflags, dev->name, dev))
 		return -EAGAIN;
+	/* Now that we have an interrupt handler hooked up we can unmask
+	 * our interrupts
+	 */
+	dm9000_unmask_interrupts(db);
 
 	/* Init driver variable */
 	db->dbug_cnt = 0;
@@ -1313,7 +1331,8 @@ dm9000_open(struct net_device *dev)
 	mii_check_media(&db->mii, netif_msg_link(db), 1);
 	netif_start_queue(dev);
 
-	dm9000_schedule_poll(db);
+	/* Poll initial link status */
+	schedule_delayed_work(&db->phy_poll, 1);
 
 	return 0;
 }
@@ -1326,7 +1345,7 @@ dm9000_shutdown(struct net_device *dev)
 	/* RESET device */
 	dm9000_phy_write(dev, 0, MII_BMCR, BMCR_RESET);	/* PHY RESET */
 	iow(db, DM9000_GPR, 0x01);	/* Power-Down PHY */
-	iow(db, DM9000_IMR, IMR_PAR);	/* Disable all interrupt */
+	dm9000_mask_interrupts(db);
 	iow(db, DM9000_RCR, 0x00);	/* Disable RX */
 }
 
@@ -1547,12 +1566,7 @@ dm9000_probe(struct platform_device *pdev)
 	db->flags |= DM9000_PLATF_SIMPLE_PHY;
 #endif
 
-	/* Fixing bug on dm9000_probe, takeover dm9000_reset(db),
-	 * Need 'NCR_MAC_LBK' bit to indeed stable our DM9000 fifo
-	 * while probe stage.
-	 */
-
-	iow(db, DM9000_NCR, NCR_MAC_LBK | NCR_RST);
+	dm9000_reset(db);
 
 	/* try multiple times, DM9000 sometimes gets the read wrong */
 	for (i = 0; i < 8; i++) {
@@ -1695,8 +1709,8 @@ dm9000_drv_resume(struct device *dev)
 			/* reset if we were not in wake mode to ensure if
 			 * the device was powered off it is in a known state */
 			if (!db->wake_state) {
-				dm9000_reset(db);
 				dm9000_init_dm9000(ndev);
+				dm9000_unmask_interrupts(db);
 			}
 
 			netif_device_attach(ndev);
diff --git a/drivers/net/ethernet/dec/tulip/timer.c b/drivers/net/ethernet/dec/tulip/timer.c
index 768379b8aee9..523d9dde50a2 100644
--- a/drivers/net/ethernet/dec/tulip/timer.c
+++ b/drivers/net/ethernet/dec/tulip/timer.c
@@ -158,7 +158,7 @@ void comet_timer(unsigned long data)
 {
 	struct net_device *dev = (struct net_device *)data;
 	struct tulip_private *tp = netdev_priv(dev);
-	int next_tick = 60*HZ;
+	int next_tick = 2*HZ;
 
 	if (tulip_debug > 1)
 		netdev_dbg(dev, "Comet link status %04x partner capability %04x\n",
diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c
index 1642de78aac8..861660841ce2 100644
--- a/drivers/net/ethernet/dec/tulip/tulip_core.c
+++ b/drivers/net/ethernet/dec/tulip/tulip_core.c
@@ -1703,7 +1703,7 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 #ifdef CONFIG_TULIP_NAPI
 	netif_napi_add(dev, &tp->napi, tulip_poll, 16);
 #endif
-	SET_ETHTOOL_OPS(dev, &ops);
+	dev->ethtool_ops = &ops;
 
 	if (register_netdev(dev))
 		goto err_out_free_ring;
diff --git a/drivers/net/ethernet/dec/tulip/uli526x.c b/drivers/net/ethernet/dec/tulip/uli526x.c
index aa801a6af7b9..80afec335a11 100644
--- a/drivers/net/ethernet/dec/tulip/uli526x.c
+++ b/drivers/net/ethernet/dec/tulip/uli526x.c
@@ -962,8 +962,8 @@ ULi_ethtool_gset(struct uli526x_board_info *db, struct ethtool_cmd *ecmd)
 	}
 	if(db->link_failed)
 	{
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	if (db->media_mode & ULI526X_AUTO)
diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c
index 4fb756d219f7..1274b6fdac8a 100644
--- a/drivers/net/ethernet/dlink/dl2k.c
+++ b/drivers/net/ethernet/dlink/dl2k.c
@@ -227,7 +227,7 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 	dev->netdev_ops = &netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
-	SET_ETHTOOL_OPS(dev, &ethtool_ops);
+	dev->ethtool_ops = &ethtool_ops;
 #if 0
 	dev->features = NETIF_F_IP_CSUM;
 #endif
@@ -1185,8 +1185,8 @@ static int rio_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		ethtool_cmd_speed_set(cmd, np->speed);
 		cmd->duplex = np->full_duplex ? DUPLEX_FULL : DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(cmd, -1);
-		cmd->duplex = -1;
+		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 	if ( np->an_enable)
 		cmd->autoneg = AUTONEG_ENABLE;
diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c
index d9e5ca0d48c1..433c1e185442 100644
--- a/drivers/net/ethernet/dlink/sundance.c
+++ b/drivers/net/ethernet/dlink/sundance.c
@@ -577,7 +577,7 @@ static int sundance_probe1(struct pci_dev *pdev,
 
 	/* The chip-specific entries in the device structure. */
 	dev->netdev_ops = &netdev_ops;
-	SET_ETHTOOL_OPS(dev, &ethtool_ops);
+	dev->ethtool_ops = &ethtool_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
 	pci_set_drvdata(pdev, dev);
diff --git a/drivers/net/ethernet/ec_bhf.c b/drivers/net/ethernet/ec_bhf.c
index 4884205e56ee..056b44b93477 100644
--- a/drivers/net/ethernet/ec_bhf.c
+++ b/drivers/net/ethernet/ec_bhf.c
@@ -134,17 +134,17 @@ struct ec_bhf_priv {
 
 	struct pci_dev *dev;
 
-	void * __iomem io;
-	void * __iomem dma_io;
+	void __iomem *io;
+	void __iomem *dma_io;
 
 	struct hrtimer hrtimer;
 
 	int tx_dma_chan;
 	int rx_dma_chan;
-	void * __iomem ec_io;
-	void * __iomem fifo_io;
-	void * __iomem mii_io;
-	void * __iomem mac_io;
+	void __iomem *ec_io;
+	void __iomem *fifo_io;
+	void __iomem *mii_io;
+	void __iomem *mac_io;
 
 	struct bhf_dma rx_buf;
 	struct rx_desc *rx_descs;
@@ -297,7 +297,7 @@ static int ec_bhf_setup_offsets(struct ec_bhf_priv *priv)
 {
 	struct device *dev = PRIV_TO_DEV(priv);
 	unsigned block_count, i;
-	void * __iomem ec_info;
+	void __iomem *ec_info;
 
 	dev_dbg(dev, "Info block:\n");
 	dev_dbg(dev, "Type of function: %x\n", (unsigned)ioread16(priv->io));
@@ -569,8 +569,8 @@ static int ec_bhf_probe(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	struct net_device *net_dev;
 	struct ec_bhf_priv *priv;
-	void * __iomem dma_io;
-	void * __iomem io;
+	void __iomem *dma_io;
+	void __iomem *io;
 	int err = 0;
 
 	err = pci_enable_device(dev);
@@ -615,7 +615,7 @@ static int ec_bhf_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	}
 
 	net_dev = alloc_etherdev(sizeof(struct ec_bhf_priv));
-	if (net_dev == 0) {
+	if (net_dev == NULL) {
 		err = -ENOMEM;
 		goto err_unmap_dma_io;
 	}
diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index 97db5a7179df..c2f5d2d3b932 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -120,6 +120,9 @@ static inline char *nic_name(struct pci_dev *pdev)
 #define MAX_VFS			30 /* Max VFs supported by BE3 FW */
 #define FW_VER_LEN		32
 
+#define	RSS_INDIR_TABLE_LEN	128
+#define RSS_HASH_KEY_LEN	40
+
 struct be_dma_mem {
 	void *va;
 	dma_addr_t dma;
@@ -371,6 +374,7 @@ enum vf_state {
 #define BE_FLAGS_LINK_STATUS_INIT		1
 #define BE_FLAGS_WORKER_SCHEDULED		(1 << 3)
 #define BE_FLAGS_VLAN_PROMISC			(1 << 4)
+#define BE_FLAGS_MCAST_PROMISC			(1 << 5)
 #define BE_FLAGS_NAPI_ENABLED			(1 << 9)
 #define BE_FLAGS_QNQ_ASYNC_EVT_RCVD		(1 << 11)
 #define BE_FLAGS_VXLAN_OFFLOADS			(1 << 12)
@@ -409,6 +413,13 @@ struct be_resources {
 	u32 if_cap_flags;
 };
 
+struct rss_info {
+	u64 rss_flags;
+	u8 rsstable[RSS_INDIR_TABLE_LEN];
+	u8 rss_queue[RSS_INDIR_TABLE_LEN];
+	u8 rss_hkey[RSS_HASH_KEY_LEN];
+};
+
 struct be_adapter {
 	struct pci_dev *pdev;
 	struct net_device *netdev;
@@ -445,7 +456,7 @@ struct be_adapter {
 	struct be_drv_stats drv_stats;
 	struct be_aic_obj aic_obj[MAX_EVT_QS];
 	u16 vlans_added;
-	u8 vlan_tag[VLAN_N_VID];
+	unsigned long vids[BITS_TO_LONGS(VLAN_N_VID)];
 	u8 vlan_prio_bmap;	/* Available Priority BitMap */
 	u16 recommended_prio;	/* Recommended Priority */
 	struct be_dma_mem rx_filter; /* Cmd DMA mem for rx-filter */
@@ -507,7 +518,7 @@ struct be_adapter {
 	u32 msg_enable;
 	int be_get_temp_freq;
 	u8 pf_number;
-	u64 rss_flags;
+	struct rss_info rss_info;
 };
 
 #define be_physfn(adapter)		(!adapter->virtfn)
@@ -546,9 +557,7 @@ static inline u16 be_max_qs(struct be_adapter *adapter)
 #define be_pvid_tagging_enabled(adapter)	(adapter->pvid)
 
 /* Is BE in QNQ multi-channel mode */
-#define be_is_qnq_mode(adapter)		(adapter->mc_type == FLEX10 ||  \
-					 adapter->mc_type == vNIC1 ||	\
-					 adapter->mc_type == UFP)
+#define be_is_qnq_mode(adapter)		(adapter->function_mode & QNQ_MODE)
 
 #define lancer_chip(adapter)	(adapter->pdev->device == OC_DEVICE_ID3 || \
 				 adapter->pdev->device == OC_DEVICE_ID4)
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index d1ec15af0d24..f4ea3490f446 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -52,8 +52,7 @@ static struct be_cmd_priv_map cmd_priv_map[] = {
 	}
 };
 
-static bool be_cmd_allowed(struct be_adapter *adapter, u8 opcode,
-			   u8 subsystem)
+static bool be_cmd_allowed(struct be_adapter *adapter, u8 opcode, u8 subsystem)
 {
 	int i;
 	int num_entries = sizeof(cmd_priv_map)/sizeof(struct be_cmd_priv_map);
@@ -120,21 +119,28 @@ static struct be_cmd_resp_hdr *be_decode_resp_hdr(u32 tag0, u32 tag1)
 	return (void *)addr;
 }
 
-static int be_mcc_compl_process(struct be_adapter *adapter,
-				struct be_mcc_compl *compl)
+static bool be_skip_err_log(u8 opcode, u16 base_status, u16 addl_status)
 {
-	u16 compl_status, extd_status;
-	struct be_cmd_resp_hdr *resp_hdr;
-	u8 opcode = 0, subsystem = 0;
-
-	/* Just swap the status to host endian; mcc tag is opaquely copied
-	 * from mcc_wrb */
-	be_dws_le_to_cpu(compl, 4);
-
-	compl_status = (compl->status >> CQE_STATUS_COMPL_SHIFT) &
-				CQE_STATUS_COMPL_MASK;
+	if (base_status == MCC_STATUS_NOT_SUPPORTED ||
+	    base_status == MCC_STATUS_ILLEGAL_REQUEST ||
+	    addl_status == MCC_ADDL_STATUS_TOO_MANY_INTERFACES ||
+	    (opcode == OPCODE_COMMON_WRITE_FLASHROM &&
+	    (base_status == MCC_STATUS_ILLEGAL_FIELD ||
+	     addl_status == MCC_ADDL_STATUS_FLASH_IMAGE_CRC_MISMATCH)))
+		return true;
+	else
+		return false;
+}
 
-	resp_hdr = be_decode_resp_hdr(compl->tag0, compl->tag1);
+/* Place holder for all the async MCC cmds wherein the caller is not in a busy
+ * loop (has not issued be_mcc_notify_wait())
+ */
+static void be_async_cmd_process(struct be_adapter *adapter,
+				 struct be_mcc_compl *compl,
+				 struct be_cmd_resp_hdr *resp_hdr)
+{
+	enum mcc_base_status base_status = base_status(compl->status);
+	u8 opcode = 0, subsystem = 0;
 
 	if (resp_hdr) {
 		opcode = resp_hdr->opcode;
@@ -144,61 +150,86 @@ static int be_mcc_compl_process(struct be_adapter *adapter,
 	if (opcode == OPCODE_LOWLEVEL_LOOPBACK_TEST &&
 	    subsystem == CMD_SUBSYSTEM_LOWLEVEL) {
 		complete(&adapter->et_cmd_compl);
-		return 0;
+		return;
 	}
 
-	if (((opcode == OPCODE_COMMON_WRITE_FLASHROM) ||
-	     (opcode == OPCODE_COMMON_WRITE_OBJECT)) &&
-	    (subsystem == CMD_SUBSYSTEM_COMMON)) {
-		adapter->flash_status = compl_status;
+	if ((opcode == OPCODE_COMMON_WRITE_FLASHROM ||
+	     opcode == OPCODE_COMMON_WRITE_OBJECT) &&
+	    subsystem == CMD_SUBSYSTEM_COMMON) {
+		adapter->flash_status = compl->status;
 		complete(&adapter->et_cmd_compl);
+		return;
 	}
 
-	if (compl_status == MCC_STATUS_SUCCESS) {
-		if (((opcode == OPCODE_ETH_GET_STATISTICS) ||
-		     (opcode == OPCODE_ETH_GET_PPORT_STATS)) &&
-		    (subsystem == CMD_SUBSYSTEM_ETH)) {
-			be_parse_stats(adapter);
-			adapter->stats_cmd_sent = false;
-		}
-		if (opcode == OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES &&
-		    subsystem == CMD_SUBSYSTEM_COMMON) {
+	if ((opcode == OPCODE_ETH_GET_STATISTICS ||
+	     opcode == OPCODE_ETH_GET_PPORT_STATS) &&
+	    subsystem == CMD_SUBSYSTEM_ETH &&
+	    base_status == MCC_STATUS_SUCCESS) {
+		be_parse_stats(adapter);
+		adapter->stats_cmd_sent = false;
+		return;
+	}
+
+	if (opcode == OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES &&
+	    subsystem == CMD_SUBSYSTEM_COMMON) {
+		if (base_status == MCC_STATUS_SUCCESS) {
 			struct be_cmd_resp_get_cntl_addnl_attribs *resp =
-				(void *)resp_hdr;
+							(void *)resp_hdr;
 			adapter->drv_stats.be_on_die_temperature =
-				resp->on_die_temperature;
-		}
-	} else {
-		if (opcode == OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES)
+						resp->on_die_temperature;
+		} else {
 			adapter->be_get_temp_freq = 0;
+		}
+		return;
+	}
+}
+
+static int be_mcc_compl_process(struct be_adapter *adapter,
+				struct be_mcc_compl *compl)
+{
+	enum mcc_base_status base_status;
+	enum mcc_addl_status addl_status;
+	struct be_cmd_resp_hdr *resp_hdr;
+	u8 opcode = 0, subsystem = 0;
+
+	/* Just swap the status to host endian; mcc tag is opaquely copied
+	 * from mcc_wrb */
+	be_dws_le_to_cpu(compl, 4);
+
+	base_status = base_status(compl->status);
+	addl_status = addl_status(compl->status);
+
+	resp_hdr = be_decode_resp_hdr(compl->tag0, compl->tag1);
+	if (resp_hdr) {
+		opcode = resp_hdr->opcode;
+		subsystem = resp_hdr->subsystem;
+	}
+
+	be_async_cmd_process(adapter, compl, resp_hdr);
 
-		if (compl_status == MCC_STATUS_NOT_SUPPORTED ||
-			compl_status == MCC_STATUS_ILLEGAL_REQUEST)
-			goto done;
+	if (base_status != MCC_STATUS_SUCCESS &&
+	    !be_skip_err_log(opcode, base_status, addl_status)) {
 
-		if (compl_status == MCC_STATUS_UNAUTHORIZED_REQUEST) {
+		if (base_status == MCC_STATUS_UNAUTHORIZED_REQUEST) {
 			dev_warn(&adapter->pdev->dev,
 				 "VF is not privileged to issue opcode %d-%d\n",
 				 opcode, subsystem);
 		} else {
-			extd_status = (compl->status >> CQE_STATUS_EXTD_SHIFT) &
-					CQE_STATUS_EXTD_MASK;
 			dev_err(&adapter->pdev->dev,
 				"opcode %d-%d failed:status %d-%d\n",
-				opcode, subsystem, compl_status, extd_status);
-
-			if (extd_status == MCC_ADDL_STS_INSUFFICIENT_RESOURCES)
-				return extd_status;
+				opcode, subsystem, base_status, addl_status);
 		}
 	}
-done:
-	return compl_status;
+	return compl->status;
 }
 
 /* Link state evt is a string of bytes; no need for endian swapping */
 static void be_async_link_state_process(struct be_adapter *adapter,
-		struct be_async_event_link_state *evt)
+					struct be_mcc_compl *compl)
 {
+	struct be_async_event_link_state *evt =
+			(struct be_async_event_link_state *)compl;
+
 	/* When link status changes, link speed must be re-queried from FW */
 	adapter->phy.link_speed = -1;
 
@@ -221,8 +252,11 @@ static void be_async_link_state_process(struct be_adapter *adapter,
 
 /* Grp5 CoS Priority evt */
 static void be_async_grp5_cos_priority_process(struct be_adapter *adapter,
-		struct be_async_event_grp5_cos_priority *evt)
+					       struct be_mcc_compl *compl)
 {
+	struct be_async_event_grp5_cos_priority *evt =
+			(struct be_async_event_grp5_cos_priority *)compl;
+
 	if (evt->valid) {
 		adapter->vlan_prio_bmap = evt->available_priority_bmap;
 		adapter->recommended_prio &= ~VLAN_PRIO_MASK;
@@ -233,8 +267,11 @@ static void be_async_grp5_cos_priority_process(struct be_adapter *adapter,
 
 /* Grp5 QOS Speed evt: qos_link_speed is in units of 10 Mbps */
 static void be_async_grp5_qos_speed_process(struct be_adapter *adapter,
-		struct be_async_event_grp5_qos_link_speed *evt)
+					    struct be_mcc_compl *compl)
 {
+	struct be_async_event_grp5_qos_link_speed *evt =
+			(struct be_async_event_grp5_qos_link_speed *)compl;
+
 	if (adapter->phy.link_speed >= 0 &&
 	    evt->physical_port == adapter->port_num)
 		adapter->phy.link_speed = le16_to_cpu(evt->qos_link_speed) * 10;
@@ -242,8 +279,11 @@ static void be_async_grp5_qos_speed_process(struct be_adapter *adapter,
 
 /*Grp5 PVID evt*/
 static void be_async_grp5_pvid_state_process(struct be_adapter *adapter,
-		struct be_async_event_grp5_pvid_state *evt)
+					     struct be_mcc_compl *compl)
 {
+	struct be_async_event_grp5_pvid_state *evt =
+			(struct be_async_event_grp5_pvid_state *)compl;
+
 	if (evt->enabled) {
 		adapter->pvid = le16_to_cpu(evt->tag) & VLAN_VID_MASK;
 		dev_info(&adapter->pdev->dev, "LPVID: %d\n", adapter->pvid);
@@ -253,26 +293,21 @@ static void be_async_grp5_pvid_state_process(struct be_adapter *adapter,
 }
 
 static void be_async_grp5_evt_process(struct be_adapter *adapter,
-		u32 trailer, struct be_mcc_compl *evt)
+				      struct be_mcc_compl *compl)
 {
-	u8 event_type = 0;
-
-	event_type = (trailer >> ASYNC_TRAILER_EVENT_TYPE_SHIFT) &
-		ASYNC_TRAILER_EVENT_TYPE_MASK;
+	u8 event_type = (compl->flags >> ASYNC_EVENT_TYPE_SHIFT) &
+				ASYNC_EVENT_TYPE_MASK;
 
 	switch (event_type) {
 	case ASYNC_EVENT_COS_PRIORITY:
-		be_async_grp5_cos_priority_process(adapter,
-		(struct be_async_event_grp5_cos_priority *)evt);
-	break;
+		be_async_grp5_cos_priority_process(adapter, compl);
+		break;
 	case ASYNC_EVENT_QOS_SPEED:
-		be_async_grp5_qos_speed_process(adapter,
-		(struct be_async_event_grp5_qos_link_speed *)evt);
-	break;
+		be_async_grp5_qos_speed_process(adapter, compl);
+		break;
 	case ASYNC_EVENT_PVID_STATE:
-		be_async_grp5_pvid_state_process(adapter,
-		(struct be_async_event_grp5_pvid_state *)evt);
-	break;
+		be_async_grp5_pvid_state_process(adapter, compl);
+		break;
 	default:
 		dev_warn(&adapter->pdev->dev, "Unknown grp5 event 0x%x!\n",
 			 event_type);
@@ -281,13 +316,13 @@ static void be_async_grp5_evt_process(struct be_adapter *adapter,
 }
 
 static void be_async_dbg_evt_process(struct be_adapter *adapter,
-		u32 trailer, struct be_mcc_compl *cmp)
+				     struct be_mcc_compl *cmp)
 {
 	u8 event_type = 0;
 	struct be_async_event_qnq *evt = (struct be_async_event_qnq *) cmp;
 
-	event_type = (trailer >> ASYNC_TRAILER_EVENT_TYPE_SHIFT) &
-		ASYNC_TRAILER_EVENT_TYPE_MASK;
+	event_type = (cmp->flags >> ASYNC_EVENT_TYPE_SHIFT) &
+			ASYNC_EVENT_TYPE_MASK;
 
 	switch (event_type) {
 	case ASYNC_DEBUG_EVENT_TYPE_QNQ:
@@ -302,25 +337,33 @@ static void be_async_dbg_evt_process(struct be_adapter *adapter,
 	}
 }
 
-static inline bool is_link_state_evt(u32 trailer)
+static inline bool is_link_state_evt(u32 flags)
 {
-	return ((trailer >> ASYNC_TRAILER_EVENT_CODE_SHIFT) &
-		ASYNC_TRAILER_EVENT_CODE_MASK) ==
-				ASYNC_EVENT_CODE_LINK_STATE;
+	return ((flags >> ASYNC_EVENT_CODE_SHIFT) & ASYNC_EVENT_CODE_MASK) ==
+			ASYNC_EVENT_CODE_LINK_STATE;
 }
 
-static inline bool is_grp5_evt(u32 trailer)
+static inline bool is_grp5_evt(u32 flags)
 {
-	return (((trailer >> ASYNC_TRAILER_EVENT_CODE_SHIFT) &
-		ASYNC_TRAILER_EVENT_CODE_MASK) ==
-				ASYNC_EVENT_CODE_GRP_5);
+	return ((flags >> ASYNC_EVENT_CODE_SHIFT) & ASYNC_EVENT_CODE_MASK) ==
+			ASYNC_EVENT_CODE_GRP_5;
 }
 
-static inline bool is_dbg_evt(u32 trailer)
+static inline bool is_dbg_evt(u32 flags)
 {
-	return (((trailer >> ASYNC_TRAILER_EVENT_CODE_SHIFT) &
-		ASYNC_TRAILER_EVENT_CODE_MASK) ==
-				ASYNC_EVENT_CODE_QNQ);
+	return ((flags >> ASYNC_EVENT_CODE_SHIFT) & ASYNC_EVENT_CODE_MASK) ==
+			ASYNC_EVENT_CODE_QNQ;
+}
+
+static void be_mcc_event_process(struct be_adapter *adapter,
+				 struct be_mcc_compl *compl)
+{
+	if (is_link_state_evt(compl->flags))
+		be_async_link_state_process(adapter, compl);
+	else if (is_grp5_evt(compl->flags))
+		be_async_grp5_evt_process(adapter, compl);
+	else if (is_dbg_evt(compl->flags))
+		be_async_dbg_evt_process(adapter, compl);
 }
 
 static struct be_mcc_compl *be_mcc_compl_get(struct be_adapter *adapter)
@@ -362,21 +405,13 @@ int be_process_mcc(struct be_adapter *adapter)
 	struct be_mcc_obj *mcc_obj = &adapter->mcc_obj;
 
 	spin_lock(&adapter->mcc_cq_lock);
+
 	while ((compl = be_mcc_compl_get(adapter))) {
 		if (compl->flags & CQE_FLAGS_ASYNC_MASK) {
-			/* Interpret flags as an async trailer */
-			if (is_link_state_evt(compl->flags))
-				be_async_link_state_process(adapter,
-				(struct be_async_event_link_state *) compl);
-			else if (is_grp5_evt(compl->flags))
-				be_async_grp5_evt_process(adapter,
-				compl->flags, compl);
-			else if (is_dbg_evt(compl->flags))
-				be_async_dbg_evt_process(adapter,
-				compl->flags, compl);
+			be_mcc_event_process(adapter, compl);
 		} else if (compl->flags & CQE_FLAGS_COMPLETED_MASK) {
-				status = be_mcc_compl_process(adapter, compl);
-				atomic_dec(&mcc_obj->q.used);
+			status = be_mcc_compl_process(adapter, compl);
+			atomic_dec(&mcc_obj->q.used);
 		}
 		be_mcc_compl_use(compl);
 		num++;
@@ -436,7 +471,9 @@ static int be_mcc_notify_wait(struct be_adapter *adapter)
 	if (status == -EIO)
 		goto out;
 
-	status = resp->status;
+	status = (resp->base_status |
+		  ((resp->addl_status & CQE_ADDL_STATUS_MASK) <<
+		   CQE_ADDL_STATUS_SHIFT));
 out:
 	return status;
 }
@@ -560,10 +597,8 @@ static bool lancer_provisioning_error(struct be_adapter *adapter)
 	u32 sliport_status = 0, sliport_err1 = 0, sliport_err2 = 0;
 	sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET);
 	if (sliport_status & SLIPORT_STATUS_ERR_MASK) {
-		sliport_err1 = ioread32(adapter->db +
-					SLIPORT_ERROR1_OFFSET);
-		sliport_err2 = ioread32(adapter->db +
-					SLIPORT_ERROR2_OFFSET);
+		sliport_err1 = ioread32(adapter->db + SLIPORT_ERROR1_OFFSET);
+		sliport_err2 = ioread32(adapter->db + SLIPORT_ERROR2_OFFSET);
 
 		if (sliport_err1 == SLIPORT_ERROR_NO_RESOURCE1 &&
 		    sliport_err2 == SLIPORT_ERROR_NO_RESOURCE2)
@@ -630,8 +665,7 @@ int be_fw_wait_ready(struct be_adapter *adapter)
 		if (stage == POST_STAGE_ARMFW_RDY)
 			return 0;
 
-		dev_info(dev, "Waiting for POST, %ds elapsed\n",
-			 timeout);
+		dev_info(dev, "Waiting for POST, %ds elapsed\n", timeout);
 		if (msleep_interruptible(2000)) {
 			dev_err(dev, "Waiting for POST aborted\n");
 			return -EINTR;
@@ -649,8 +683,7 @@ static inline struct be_sge *nonembedded_sgl(struct be_mcc_wrb *wrb)
 	return &wrb->payload.sgl[0];
 }
 
-static inline void fill_wrb_tags(struct be_mcc_wrb *wrb,
-				 unsigned long addr)
+static inline void fill_wrb_tags(struct be_mcc_wrb *wrb, unsigned long addr)
 {
 	wrb->tag0 = addr & 0xFFFFFFFF;
 	wrb->tag1 = upper_32_bits(addr);
@@ -659,8 +692,9 @@ static inline void fill_wrb_tags(struct be_mcc_wrb *wrb,
 /* Don't touch the hdr after it's prepared */
 /* mem will be NULL for embedded commands */
 static void be_wrb_cmd_hdr_prepare(struct be_cmd_req_hdr *req_hdr,
-				u8 subsystem, u8 opcode, int cmd_len,
-				struct be_mcc_wrb *wrb, struct be_dma_mem *mem)
+				   u8 subsystem, u8 opcode, int cmd_len,
+				   struct be_mcc_wrb *wrb,
+				   struct be_dma_mem *mem)
 {
 	struct be_sge *sge;
 
@@ -683,7 +717,7 @@ static void be_wrb_cmd_hdr_prepare(struct be_cmd_req_hdr *req_hdr,
 }
 
 static void be_cmd_page_addrs_prepare(struct phys_addr *pages, u32 max_pages,
-			struct be_dma_mem *mem)
+				      struct be_dma_mem *mem)
 {
 	int i, buf_pages = min(PAGES_4K_SPANNED(mem->va, mem->size), max_pages);
 	u64 dma = (u64)mem->dma;
@@ -868,7 +902,8 @@ int be_cmd_eq_create(struct be_adapter *adapter, struct be_eq_obj *eqo)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_EQ_CREATE, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_EQ_CREATE, sizeof(*req), wrb,
+			       NULL);
 
 	/* Support for EQ_CREATEv2 available only SH-R onwards */
 	if (!(BEx_chip(adapter) || lancer_chip(adapter)))
@@ -917,7 +952,8 @@ int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_NTWK_MAC_QUERY, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_NTWK_MAC_QUERY, sizeof(*req), wrb,
+			       NULL);
 	req->type = MAC_ADDRESS_TYPE_NETWORK;
 	if (permanent) {
 		req->permanent = 1;
@@ -940,7 +976,7 @@ err:
 
 /* Uses synchronous MCCQ */
 int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr,
-		u32 if_id, u32 *pmac_id, u32 domain)
+		    u32 if_id, u32 *pmac_id, u32 domain)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_pmac_add *req;
@@ -956,7 +992,8 @@ int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_NTWK_PMAC_ADD, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_NTWK_PMAC_ADD, sizeof(*req), wrb,
+			       NULL);
 
 	req->hdr.domain = domain;
 	req->if_id = cpu_to_le32(if_id);
@@ -1012,7 +1049,7 @@ err:
 
 /* Uses Mbox */
 int be_cmd_cq_create(struct be_adapter *adapter, struct be_queue_info *cq,
-		struct be_queue_info *eq, bool no_delay, int coalesce_wm)
+		     struct be_queue_info *eq, bool no_delay, int coalesce_wm)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_cq_create *req;
@@ -1028,17 +1065,18 @@ int be_cmd_cq_create(struct be_adapter *adapter, struct be_queue_info *cq,
 	ctxt = &req->context;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_CQ_CREATE, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_CQ_CREATE, sizeof(*req), wrb,
+			       NULL);
 
 	req->num_pages =  cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size));
 
 	if (BEx_chip(adapter)) {
 		AMAP_SET_BITS(struct amap_cq_context_be, coalescwm, ctxt,
-								coalesce_wm);
+			      coalesce_wm);
 		AMAP_SET_BITS(struct amap_cq_context_be, nodelay,
-								ctxt, no_delay);
+			      ctxt, no_delay);
 		AMAP_SET_BITS(struct amap_cq_context_be, count, ctxt,
-						__ilog2_u32(cq->len/256));
+			      __ilog2_u32(cq->len / 256));
 		AMAP_SET_BITS(struct amap_cq_context_be, valid, ctxt, 1);
 		AMAP_SET_BITS(struct amap_cq_context_be, eventable, ctxt, 1);
 		AMAP_SET_BITS(struct amap_cq_context_be, eqid, ctxt, eq->id);
@@ -1053,14 +1091,12 @@ int be_cmd_cq_create(struct be_adapter *adapter, struct be_queue_info *cq,
 			AMAP_SET_BITS(struct amap_cq_context_v2, coalescwm,
 				      ctxt, coalesce_wm);
 		AMAP_SET_BITS(struct amap_cq_context_v2, nodelay, ctxt,
-								no_delay);
+			      no_delay);
 		AMAP_SET_BITS(struct amap_cq_context_v2, count, ctxt,
-						__ilog2_u32(cq->len/256));
+			      __ilog2_u32(cq->len / 256));
 		AMAP_SET_BITS(struct amap_cq_context_v2, valid, ctxt, 1);
-		AMAP_SET_BITS(struct amap_cq_context_v2, eventable,
-								ctxt, 1);
-		AMAP_SET_BITS(struct amap_cq_context_v2, eqid,
-								ctxt, eq->id);
+		AMAP_SET_BITS(struct amap_cq_context_v2, eventable, ctxt, 1);
+		AMAP_SET_BITS(struct amap_cq_context_v2, eqid, ctxt, eq->id);
 	}
 
 	be_dws_cpu_to_le(ctxt, sizeof(req->context));
@@ -1088,8 +1124,8 @@ static u32 be_encoded_q_len(int q_len)
 }
 
 static int be_cmd_mccq_ext_create(struct be_adapter *adapter,
-				struct be_queue_info *mccq,
-				struct be_queue_info *cq)
+				  struct be_queue_info *mccq,
+				  struct be_queue_info *cq)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_mcc_ext_create *req;
@@ -1105,13 +1141,14 @@ static int be_cmd_mccq_ext_create(struct be_adapter *adapter,
 	ctxt = &req->context;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_MCC_CREATE_EXT, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_MCC_CREATE_EXT, sizeof(*req), wrb,
+			       NULL);
 
 	req->num_pages = cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size));
 	if (BEx_chip(adapter)) {
 		AMAP_SET_BITS(struct amap_mcc_context_be, valid, ctxt, 1);
 		AMAP_SET_BITS(struct amap_mcc_context_be, ring_size, ctxt,
-						be_encoded_q_len(mccq->len));
+			      be_encoded_q_len(mccq->len));
 		AMAP_SET_BITS(struct amap_mcc_context_be, cq_id, ctxt, cq->id);
 	} else {
 		req->hdr.version = 1;
@@ -1145,8 +1182,8 @@ static int be_cmd_mccq_ext_create(struct be_adapter *adapter,
 }
 
 static int be_cmd_mccq_org_create(struct be_adapter *adapter,
-				struct be_queue_info *mccq,
-				struct be_queue_info *cq)
+				  struct be_queue_info *mccq,
+				  struct be_queue_info *cq)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_mcc_create *req;
@@ -1162,13 +1199,14 @@ static int be_cmd_mccq_org_create(struct be_adapter *adapter,
 	ctxt = &req->context;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_MCC_CREATE, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_MCC_CREATE, sizeof(*req), wrb,
+			       NULL);
 
 	req->num_pages = cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size));
 
 	AMAP_SET_BITS(struct amap_mcc_context_be, valid, ctxt, 1);
 	AMAP_SET_BITS(struct amap_mcc_context_be, ring_size, ctxt,
-			be_encoded_q_len(mccq->len));
+		      be_encoded_q_len(mccq->len));
 	AMAP_SET_BITS(struct amap_mcc_context_be, cq_id, ctxt, cq->id);
 
 	be_dws_cpu_to_le(ctxt, sizeof(req->context));
@@ -1187,8 +1225,7 @@ static int be_cmd_mccq_org_create(struct be_adapter *adapter,
 }
 
 int be_cmd_mccq_create(struct be_adapter *adapter,
-			struct be_queue_info *mccq,
-			struct be_queue_info *cq)
+		       struct be_queue_info *mccq, struct be_queue_info *cq)
 {
 	int status;
 
@@ -1213,7 +1250,7 @@ int be_cmd_txq_create(struct be_adapter *adapter, struct be_tx_obj *txo)
 
 	req = embedded_payload(&wrb);
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
-				OPCODE_ETH_TX_CREATE, sizeof(*req), &wrb, NULL);
+			       OPCODE_ETH_TX_CREATE, sizeof(*req), &wrb, NULL);
 
 	if (lancer_chip(adapter)) {
 		req->hdr.version = 1;
@@ -1250,8 +1287,8 @@ int be_cmd_txq_create(struct be_adapter *adapter, struct be_tx_obj *txo)
 
 /* Uses MCC */
 int be_cmd_rxq_create(struct be_adapter *adapter,
-		struct be_queue_info *rxq, u16 cq_id, u16 frag_size,
-		u32 if_id, u32 rss, u8 *rss_id)
+		      struct be_queue_info *rxq, u16 cq_id, u16 frag_size,
+		      u32 if_id, u32 rss, u8 *rss_id)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_eth_rx_create *req;
@@ -1268,7 +1305,7 @@ int be_cmd_rxq_create(struct be_adapter *adapter,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
-				OPCODE_ETH_RX_CREATE, sizeof(*req), wrb, NULL);
+			       OPCODE_ETH_RX_CREATE, sizeof(*req), wrb, NULL);
 
 	req->cq_id = cpu_to_le16(cq_id);
 	req->frag_size = fls(frag_size) - 1;
@@ -1295,7 +1332,7 @@ err:
  * Uses Mbox
  */
 int be_cmd_q_destroy(struct be_adapter *adapter, struct be_queue_info *q,
-		int queue_type)
+		     int queue_type)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_q_destroy *req;
@@ -1334,7 +1371,7 @@ int be_cmd_q_destroy(struct be_adapter *adapter, struct be_queue_info *q,
 	}
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, subsys, opcode, sizeof(*req), wrb,
-				NULL);
+			       NULL);
 	req->id = cpu_to_le16(q->id);
 
 	status = be_mbox_notify_wait(adapter);
@@ -1361,7 +1398,7 @@ int be_cmd_rxq_destroy(struct be_adapter *adapter, struct be_queue_info *q)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
-			OPCODE_ETH_RX_DESTROY, sizeof(*req), wrb, NULL);
+			       OPCODE_ETH_RX_DESTROY, sizeof(*req), wrb, NULL);
 	req->id = cpu_to_le16(q->id);
 
 	status = be_mcc_notify_wait(adapter);
@@ -1384,7 +1421,8 @@ int be_cmd_if_create(struct be_adapter *adapter, u32 cap_flags, u32 en_flags,
 
 	req = embedded_payload(&wrb);
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_NTWK_INTERFACE_CREATE, sizeof(*req), &wrb, NULL);
+			       OPCODE_COMMON_NTWK_INTERFACE_CREATE,
+			       sizeof(*req), &wrb, NULL);
 	req->hdr.domain = domain;
 	req->capability_flags = cpu_to_le32(cap_flags);
 	req->enable_flags = cpu_to_le32(en_flags);
@@ -1422,7 +1460,8 @@ int be_cmd_if_destroy(struct be_adapter *adapter, int interface_id, u32 domain)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_NTWK_INTERFACE_DESTROY, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_NTWK_INTERFACE_DESTROY,
+			       sizeof(*req), wrb, NULL);
 	req->hdr.domain = domain;
 	req->interface_id = cpu_to_le32(interface_id);
 
@@ -1452,7 +1491,8 @@ int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd)
 	hdr = nonemb_cmd->va;
 
 	be_wrb_cmd_hdr_prepare(hdr, CMD_SUBSYSTEM_ETH,
-		OPCODE_ETH_GET_STATISTICS, nonemb_cmd->size, wrb, nonemb_cmd);
+			       OPCODE_ETH_GET_STATISTICS, nonemb_cmd->size, wrb,
+			       nonemb_cmd);
 
 	/* version 1 of the cmd is not supported only by BE2 */
 	if (BE2_chip(adapter))
@@ -1472,7 +1512,7 @@ err:
 
 /* Lancer Stats */
 int lancer_cmd_get_pport_stats(struct be_adapter *adapter,
-				struct be_dma_mem *nonemb_cmd)
+			       struct be_dma_mem *nonemb_cmd)
 {
 
 	struct be_mcc_wrb *wrb;
@@ -1493,8 +1533,8 @@ int lancer_cmd_get_pport_stats(struct be_adapter *adapter,
 	req = nonemb_cmd->va;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
-			OPCODE_ETH_GET_PPORT_STATS, nonemb_cmd->size, wrb,
-			nonemb_cmd);
+			       OPCODE_ETH_GET_PPORT_STATS, nonemb_cmd->size,
+			       wrb, nonemb_cmd);
 
 	req->cmd_params.params.pport_num = cpu_to_le16(adapter->hba_port_num);
 	req->cmd_params.params.reset_stats = 0;
@@ -1553,7 +1593,8 @@ int be_cmd_link_status_query(struct be_adapter *adapter, u16 *link_speed,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_NTWK_LINK_STATUS_QUERY, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_NTWK_LINK_STATUS_QUERY,
+			       sizeof(*req), wrb, NULL);
 
 	/* version 1 of the cmd is not supported only by BE2 */
 	if (!BE2_chip(adapter))
@@ -1598,8 +1639,8 @@ int be_cmd_get_die_temperature(struct be_adapter *adapter)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES, sizeof(*req),
-		wrb, NULL);
+			       OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES,
+			       sizeof(*req), wrb, NULL);
 
 	be_mcc_notify(adapter);
 
@@ -1625,7 +1666,8 @@ int be_cmd_get_reg_len(struct be_adapter *adapter, u32 *log_size)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_MANAGE_FAT, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_MANAGE_FAT, sizeof(*req), wrb,
+			       NULL);
 	req->fat_operation = cpu_to_le32(QUERY_FAT);
 	status = be_mcc_notify_wait(adapter);
 	if (!status) {
@@ -1655,8 +1697,8 @@ void be_cmd_get_regs(struct be_adapter *adapter, u32 buf_len, void *buf)
 
 	get_fat_cmd.size = sizeof(struct be_cmd_req_get_fat) + 60*1024;
 	get_fat_cmd.va = pci_alloc_consistent(adapter->pdev,
-			get_fat_cmd.size,
-			&get_fat_cmd.dma);
+					      get_fat_cmd.size,
+					      &get_fat_cmd.dma);
 	if (!get_fat_cmd.va) {
 		status = -ENOMEM;
 		dev_err(&adapter->pdev->dev,
@@ -1679,8 +1721,8 @@ void be_cmd_get_regs(struct be_adapter *adapter, u32 buf_len, void *buf)
 
 		payload_len = sizeof(struct be_cmd_req_get_fat) + buf_size;
 		be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-				OPCODE_COMMON_MANAGE_FAT, payload_len, wrb,
-				&get_fat_cmd);
+				       OPCODE_COMMON_MANAGE_FAT, payload_len,
+				       wrb, &get_fat_cmd);
 
 		req->fat_operation = cpu_to_le32(RETRIEVE_FAT);
 		req->read_log_offset = cpu_to_le32(log_offset);
@@ -1691,8 +1733,8 @@ void be_cmd_get_regs(struct be_adapter *adapter, u32 buf_len, void *buf)
 		if (!status) {
 			struct be_cmd_resp_get_fat *resp = get_fat_cmd.va;
 			memcpy(buf + offset,
-				resp->data_buffer,
-				le32_to_cpu(resp->read_log_length));
+			       resp->data_buffer,
+			       le32_to_cpu(resp->read_log_length));
 		} else {
 			dev_err(&adapter->pdev->dev, "FAT Table Retrieve error\n");
 			goto err;
@@ -1702,14 +1744,13 @@ void be_cmd_get_regs(struct be_adapter *adapter, u32 buf_len, void *buf)
 	}
 err:
 	pci_free_consistent(adapter->pdev, get_fat_cmd.size,
-			get_fat_cmd.va,
-			get_fat_cmd.dma);
+			    get_fat_cmd.va, get_fat_cmd.dma);
 	spin_unlock_bh(&adapter->mcc_lock);
 }
 
 /* Uses synchronous mcc */
 int be_cmd_get_fw_ver(struct be_adapter *adapter, char *fw_ver,
-			char *fw_on_flash)
+		      char *fw_on_flash)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_get_fw_version *req;
@@ -1726,7 +1767,8 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter, char *fw_ver,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_GET_FW_VERSION, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_GET_FW_VERSION, sizeof(*req), wrb,
+			       NULL);
 	status = be_mcc_notify_wait(adapter);
 	if (!status) {
 		struct be_cmd_resp_get_fw_version *resp = embedded_payload(wrb);
@@ -1759,7 +1801,8 @@ int be_cmd_modify_eqd(struct be_adapter *adapter, struct be_set_eqd *set_eqd,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_MODIFY_EQ_DELAY, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_MODIFY_EQ_DELAY, sizeof(*req), wrb,
+			       NULL);
 
 	req->num_eq = cpu_to_le32(num);
 	for (i = 0; i < num; i++) {
@@ -1777,7 +1820,7 @@ err:
 
 /* Uses sycnhronous mcc */
 int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
-		       u32 num, bool promiscuous)
+		       u32 num)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_vlan_config *req;
@@ -1793,19 +1836,16 @@ int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_NTWK_VLAN_CONFIG, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_NTWK_VLAN_CONFIG, sizeof(*req),
+			       wrb, NULL);
 
 	req->interface_id = if_id;
-	req->promiscuous = promiscuous;
 	req->untagged = BE_IF_FLAGS_UNTAGGED & be_if_cap_flags(adapter) ? 1 : 0;
 	req->num_vlan = num;
-	if (!promiscuous) {
-		memcpy(req->normal_vlan, vtag_array,
-			req->num_vlan * sizeof(vtag_array[0]));
-	}
+	memcpy(req->normal_vlan, vtag_array,
+	       req->num_vlan * sizeof(vtag_array[0]));
 
 	status = be_mcc_notify_wait(adapter);
-
 err:
 	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
@@ -1827,18 +1867,19 @@ int be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value)
 	}
 	memset(req, 0, sizeof(*req));
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-				OPCODE_COMMON_NTWK_RX_FILTER, sizeof(*req),
-				wrb, mem);
+			       OPCODE_COMMON_NTWK_RX_FILTER, sizeof(*req),
+			       wrb, mem);
 
 	req->if_id = cpu_to_le32(adapter->if_handle);
 	if (flags & IFF_PROMISC) {
 		req->if_flags_mask = cpu_to_le32(BE_IF_FLAGS_PROMISCUOUS |
-					BE_IF_FLAGS_VLAN_PROMISCUOUS |
-					BE_IF_FLAGS_MCAST_PROMISCUOUS);
+						 BE_IF_FLAGS_VLAN_PROMISCUOUS |
+						 BE_IF_FLAGS_MCAST_PROMISCUOUS);
 		if (value == ON)
-			req->if_flags = cpu_to_le32(BE_IF_FLAGS_PROMISCUOUS |
-						BE_IF_FLAGS_VLAN_PROMISCUOUS |
-						BE_IF_FLAGS_MCAST_PROMISCUOUS);
+			req->if_flags =
+				cpu_to_le32(BE_IF_FLAGS_PROMISCUOUS |
+					    BE_IF_FLAGS_VLAN_PROMISCUOUS |
+					    BE_IF_FLAGS_MCAST_PROMISCUOUS);
 	} else if (flags & IFF_ALLMULTI) {
 		req->if_flags_mask = req->if_flags =
 				cpu_to_le32(BE_IF_FLAGS_MCAST_PROMISCUOUS);
@@ -1867,7 +1908,7 @@ int be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value)
 	}
 
 	if ((req->if_flags_mask & cpu_to_le32(be_if_cap_flags(adapter))) !=
-	     req->if_flags_mask) {
+	    req->if_flags_mask) {
 		dev_warn(&adapter->pdev->dev,
 			 "Cannot set rx filter flags 0x%x\n",
 			 req->if_flags_mask);
@@ -1905,7 +1946,8 @@ int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_SET_FLOW_CONTROL, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_SET_FLOW_CONTROL, sizeof(*req),
+			       wrb, NULL);
 
 	req->tx_flow_control = cpu_to_le16((u16)tx_fc);
 	req->rx_flow_control = cpu_to_le16((u16)rx_fc);
@@ -1938,7 +1980,8 @@ int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_GET_FLOW_CONTROL, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_GET_FLOW_CONTROL, sizeof(*req),
+			       wrb, NULL);
 
 	status = be_mcc_notify_wait(adapter);
 	if (!status) {
@@ -1968,7 +2011,8 @@ int be_cmd_query_fw_cfg(struct be_adapter *adapter, u32 *port_num,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_QUERY_FIRMWARE_CONFIG, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_QUERY_FIRMWARE_CONFIG,
+			       sizeof(*req), wrb, NULL);
 
 	status = be_mbox_notify_wait(adapter);
 	if (!status) {
@@ -2011,7 +2055,8 @@ int be_cmd_reset_function(struct be_adapter *adapter)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(req, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_FUNCTION_RESET, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_FUNCTION_RESET, sizeof(*req), wrb,
+			       NULL);
 
 	status = be_mbox_notify_wait(adapter);
 
@@ -2020,47 +2065,47 @@ int be_cmd_reset_function(struct be_adapter *adapter)
 }
 
 int be_cmd_rss_config(struct be_adapter *adapter, u8 *rsstable,
-			u32 rss_hash_opts, u16 table_size)
+		      u32 rss_hash_opts, u16 table_size, const u8 *rss_hkey)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_rss_config *req;
-	u32 myhash[10] = {0x15d43fa5, 0x2534685a, 0x5f87693a, 0x5668494e,
-			0x33cf6a53, 0x383334c6, 0x76ac4257, 0x59b242b2,
-			0x3ea83c02, 0x4a110304};
 	int status;
 
 	if (!(be_if_cap_flags(adapter) & BE_IF_FLAGS_RSS))
 		return 0;
 
-	if (mutex_lock_interruptible(&adapter->mbox_lock))
-		return -1;
+	spin_lock_bh(&adapter->mcc_lock);
 
-	wrb = wrb_from_mbox(adapter);
+	wrb = wrb_from_mccq(adapter);
+	if (!wrb) {
+		status = -EBUSY;
+		goto err;
+	}
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
-		OPCODE_ETH_RSS_CONFIG, sizeof(*req), wrb, NULL);
+			       OPCODE_ETH_RSS_CONFIG, sizeof(*req), wrb, NULL);
 
 	req->if_id = cpu_to_le32(adapter->if_handle);
 	req->enable_rss = cpu_to_le16(rss_hash_opts);
 	req->cpu_table_size_log2 = cpu_to_le16(fls(table_size) - 1);
 
-	if (lancer_chip(adapter) || skyhawk_chip(adapter))
+	if (!BEx_chip(adapter))
 		req->hdr.version = 1;
 
 	memcpy(req->cpu_table, rsstable, table_size);
-	memcpy(req->hash, myhash, sizeof(myhash));
+	memcpy(req->hash, rss_hkey, RSS_HASH_KEY_LEN);
 	be_dws_cpu_to_le(req->hash, sizeof(req->hash));
 
-	status = be_mbox_notify_wait(adapter);
-
-	mutex_unlock(&adapter->mbox_lock);
+	status = be_mcc_notify_wait(adapter);
+err:
+	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
 
 /* Uses sync mcc */
 int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num,
-			u8 bcn, u8 sts, u8 state)
+			    u8 bcn, u8 sts, u8 state)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_enable_disable_beacon *req;
@@ -2076,7 +2121,8 @@ int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_ENABLE_DISABLE_BEACON, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_ENABLE_DISABLE_BEACON,
+			       sizeof(*req), wrb, NULL);
 
 	req->port_num = port_num;
 	req->beacon_state = state;
@@ -2107,7 +2153,8 @@ int be_cmd_get_beacon_state(struct be_adapter *adapter, u8 port_num, u32 *state)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_GET_BEACON_STATE, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_GET_BEACON_STATE, sizeof(*req),
+			       wrb, NULL);
 
 	req->port_num = port_num;
 
@@ -2146,20 +2193,20 @@ int lancer_cmd_write_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-				OPCODE_COMMON_WRITE_OBJECT,
-				sizeof(struct lancer_cmd_req_write_object), wrb,
-				NULL);
+			       OPCODE_COMMON_WRITE_OBJECT,
+			       sizeof(struct lancer_cmd_req_write_object), wrb,
+			       NULL);
 
 	ctxt = &req->context;
 	AMAP_SET_BITS(struct amap_lancer_write_obj_context,
-			write_length, ctxt, data_size);
+		      write_length, ctxt, data_size);
 
 	if (data_size == 0)
 		AMAP_SET_BITS(struct amap_lancer_write_obj_context,
-				eof, ctxt, 1);
+			      eof, ctxt, 1);
 	else
 		AMAP_SET_BITS(struct amap_lancer_write_obj_context,
-				eof, ctxt, 0);
+			      eof, ctxt, 0);
 
 	be_dws_cpu_to_le(ctxt, sizeof(req->context));
 	req->write_offset = cpu_to_le32(data_offset);
@@ -2167,8 +2214,8 @@ int lancer_cmd_write_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 	req->descriptor_count = cpu_to_le32(1);
 	req->buf_len = cpu_to_le32(data_size);
 	req->addr_low = cpu_to_le32((cmd->dma +
-				sizeof(struct lancer_cmd_req_write_object))
-				& 0xFFFFFFFF);
+				     sizeof(struct lancer_cmd_req_write_object))
+				    & 0xFFFFFFFF);
 	req->addr_high = cpu_to_le32(upper_32_bits(cmd->dma +
 				sizeof(struct lancer_cmd_req_write_object)));
 
@@ -2197,8 +2244,8 @@ err_unlock:
 }
 
 int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
-		u32 data_size, u32 data_offset, const char *obj_name,
-		u32 *data_read, u32 *eof, u8 *addn_status)
+			   u32 data_size, u32 data_offset, const char *obj_name,
+			   u32 *data_read, u32 *eof, u8 *addn_status)
 {
 	struct be_mcc_wrb *wrb;
 	struct lancer_cmd_req_read_object *req;
@@ -2216,9 +2263,9 @@ int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_READ_OBJECT,
-			sizeof(struct lancer_cmd_req_read_object), wrb,
-			NULL);
+			       OPCODE_COMMON_READ_OBJECT,
+			       sizeof(struct lancer_cmd_req_read_object), wrb,
+			       NULL);
 
 	req->desired_read_len = cpu_to_le32(data_size);
 	req->read_offset = cpu_to_le32(data_offset);
@@ -2244,7 +2291,7 @@ err_unlock:
 }
 
 int be_cmd_write_flashrom(struct be_adapter *adapter, struct be_dma_mem *cmd,
-			u32 flash_type, u32 flash_opcode, u32 buf_size)
+			  u32 flash_type, u32 flash_opcode, u32 buf_size)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_write_flashrom *req;
@@ -2261,7 +2308,8 @@ int be_cmd_write_flashrom(struct be_adapter *adapter, struct be_dma_mem *cmd,
 	req = cmd->va;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_WRITE_FLASHROM, cmd->size, wrb, cmd);
+			       OPCODE_COMMON_WRITE_FLASHROM, cmd->size, wrb,
+			       cmd);
 
 	req->params.op_type = cpu_to_le32(flash_type);
 	req->params.op_code = cpu_to_le32(flash_opcode);
@@ -2284,7 +2332,7 @@ err_unlock:
 }
 
 int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc,
-			 int offset)
+			  u16 optype, int offset)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_read_flash_crc *req;
@@ -2303,7 +2351,7 @@ int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc,
 			       OPCODE_COMMON_READ_FLASHROM, sizeof(*req),
 			       wrb, NULL);
 
-	req->params.op_type = cpu_to_le32(OPTYPE_REDBOOT);
+	req->params.op_type = cpu_to_le32(optype);
 	req->params.op_code = cpu_to_le32(FLASHROM_OPER_REPORT);
 	req->params.offset = cpu_to_le32(offset);
 	req->params.data_buf_size = cpu_to_le32(0x4);
@@ -2318,7 +2366,7 @@ err:
 }
 
 int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac,
-				struct be_dma_mem *nonemb_cmd)
+			    struct be_dma_mem *nonemb_cmd)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_acpi_wol_magic_config *req;
@@ -2334,8 +2382,8 @@ int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac,
 	req = nonemb_cmd->va;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
-		OPCODE_ETH_ACPI_WOL_MAGIC_CONFIG, sizeof(*req), wrb,
-		nonemb_cmd);
+			       OPCODE_ETH_ACPI_WOL_MAGIC_CONFIG, sizeof(*req),
+			       wrb, nonemb_cmd);
 	memcpy(req->magic_mac, mac, ETH_ALEN);
 
 	status = be_mcc_notify_wait(adapter);
@@ -2363,8 +2411,8 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_LOWLEVEL,
-			OPCODE_LOWLEVEL_SET_LOOPBACK_MODE, sizeof(*req), wrb,
-			NULL);
+			       OPCODE_LOWLEVEL_SET_LOOPBACK_MODE, sizeof(*req),
+			       wrb, NULL);
 
 	req->src_port = port_num;
 	req->dest_port = port_num;
@@ -2378,7 +2426,8 @@ err:
 }
 
 int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num,
-		u32 loopback_type, u32 pkt_size, u32 num_pkts, u64 pattern)
+			 u32 loopback_type, u32 pkt_size, u32 num_pkts,
+			 u64 pattern)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_loopback_test *req;
@@ -2396,7 +2445,8 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num,
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_LOWLEVEL,
-			OPCODE_LOWLEVEL_LOOPBACK_TEST, sizeof(*req), wrb, NULL);
+			       OPCODE_LOWLEVEL_LOOPBACK_TEST, sizeof(*req), wrb,
+			       NULL);
 
 	req->hdr.timeout = cpu_to_le32(15);
 	req->pattern = cpu_to_le64(pattern);
@@ -2421,7 +2471,7 @@ err:
 }
 
 int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
-				u32 byte_cnt, struct be_dma_mem *cmd)
+			u32 byte_cnt, struct be_dma_mem *cmd)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_ddrdma_test *req;
@@ -2437,7 +2487,8 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
 	}
 	req = cmd->va;
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_LOWLEVEL,
-			OPCODE_LOWLEVEL_HOST_DDR_DMA, cmd->size, wrb, cmd);
+			       OPCODE_LOWLEVEL_HOST_DDR_DMA, cmd->size, wrb,
+			       cmd);
 
 	req->pattern = cpu_to_le64(pattern);
 	req->byte_count = cpu_to_le32(byte_cnt);
@@ -2465,7 +2516,7 @@ err:
 }
 
 int be_cmd_get_seeprom_data(struct be_adapter *adapter,
-				struct be_dma_mem *nonemb_cmd)
+			    struct be_dma_mem *nonemb_cmd)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_seeprom_read *req;
@@ -2481,8 +2532,8 @@ int be_cmd_get_seeprom_data(struct be_adapter *adapter,
 	req = nonemb_cmd->va;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_SEEPROM_READ, sizeof(*req), wrb,
-			nonemb_cmd);
+			       OPCODE_COMMON_SEEPROM_READ, sizeof(*req), wrb,
+			       nonemb_cmd);
 
 	status = be_mcc_notify_wait(adapter);
 
@@ -2510,8 +2561,7 @@ int be_cmd_get_phy_info(struct be_adapter *adapter)
 		goto err;
 	}
 	cmd.size = sizeof(struct be_cmd_req_get_phy_info);
-	cmd.va = pci_alloc_consistent(adapter->pdev, cmd.size,
-					&cmd.dma);
+	cmd.va = pci_alloc_consistent(adapter->pdev, cmd.size, &cmd.dma);
 	if (!cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory alloc failure\n");
 		status = -ENOMEM;
@@ -2521,8 +2571,8 @@ int be_cmd_get_phy_info(struct be_adapter *adapter)
 	req = cmd.va;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_GET_PHY_DETAILS, sizeof(*req),
-			wrb, &cmd);
+			       OPCODE_COMMON_GET_PHY_DETAILS, sizeof(*req),
+			       wrb, &cmd);
 
 	status = be_mcc_notify_wait(adapter);
 	if (!status) {
@@ -2544,8 +2594,7 @@ int be_cmd_get_phy_info(struct be_adapter *adapter)
 				BE_SUPPORTED_SPEED_1GBPS;
 		}
 	}
-	pci_free_consistent(adapter->pdev, cmd.size,
-				cmd.va, cmd.dma);
+	pci_free_consistent(adapter->pdev, cmd.size, cmd.va, cmd.dma);
 err:
 	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
@@ -2568,7 +2617,7 @@ int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_SET_QOS, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_SET_QOS, sizeof(*req), wrb, NULL);
 
 	req->hdr.domain = domain;
 	req->valid_bits = cpu_to_le32(BE_QOS_BITS_NIC);
@@ -2597,10 +2646,9 @@ int be_cmd_get_cntl_attributes(struct be_adapter *adapter)
 	memset(&attribs_cmd, 0, sizeof(struct be_dma_mem));
 	attribs_cmd.size = sizeof(struct be_cmd_resp_cntl_attribs);
 	attribs_cmd.va = pci_alloc_consistent(adapter->pdev, attribs_cmd.size,
-						&attribs_cmd.dma);
+					      &attribs_cmd.dma);
 	if (!attribs_cmd.va) {
-		dev_err(&adapter->pdev->dev,
-				"Memory allocation failure\n");
+		dev_err(&adapter->pdev->dev, "Memory allocation failure\n");
 		status = -ENOMEM;
 		goto err;
 	}
@@ -2613,8 +2661,8 @@ int be_cmd_get_cntl_attributes(struct be_adapter *adapter)
 	req = attribs_cmd.va;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			 OPCODE_COMMON_GET_CNTL_ATTRIBUTES, payload_len, wrb,
-			&attribs_cmd);
+			       OPCODE_COMMON_GET_CNTL_ATTRIBUTES, payload_len,
+			       wrb, &attribs_cmd);
 
 	status = be_mbox_notify_wait(adapter);
 	if (!status) {
@@ -2649,7 +2697,8 @@ int be_cmd_req_native_mode(struct be_adapter *adapter)
 	req = embedded_payload(wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-		OPCODE_COMMON_SET_DRIVER_FUNCTION_CAP, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_SET_DRIVER_FUNCTION_CAP,
+			       sizeof(*req), wrb, NULL);
 
 	req->valid_cap_flags = cpu_to_le32(CAPABILITY_SW_TIMESTAMPS |
 				CAPABILITY_BE3_NATIVE_ERX_API);
@@ -2762,12 +2811,12 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac,
 	memset(&get_mac_list_cmd, 0, sizeof(struct be_dma_mem));
 	get_mac_list_cmd.size = sizeof(struct be_cmd_resp_get_mac_list);
 	get_mac_list_cmd.va = pci_alloc_consistent(adapter->pdev,
-			get_mac_list_cmd.size,
-			&get_mac_list_cmd.dma);
+						   get_mac_list_cmd.size,
+						   &get_mac_list_cmd.dma);
 
 	if (!get_mac_list_cmd.va) {
 		dev_err(&adapter->pdev->dev,
-				"Memory allocation failure during GET_MAC_LIST\n");
+			"Memory allocation failure during GET_MAC_LIST\n");
 		return -ENOMEM;
 	}
 
@@ -2831,18 +2880,18 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac,
 		/* If no active mac_id found, return first mac addr */
 		*pmac_id_valid = false;
 		memcpy(mac, resp->macaddr_list[0].mac_addr_id.macaddr,
-								ETH_ALEN);
+		       ETH_ALEN);
 	}
 
 out:
 	spin_unlock_bh(&adapter->mcc_lock);
 	pci_free_consistent(adapter->pdev, get_mac_list_cmd.size,
-			get_mac_list_cmd.va, get_mac_list_cmd.dma);
+			    get_mac_list_cmd.va, get_mac_list_cmd.dma);
 	return status;
 }
 
-int be_cmd_get_active_mac(struct be_adapter *adapter, u32 curr_pmac_id, u8 *mac,
-			  u32 if_handle, bool active, u32 domain)
+int be_cmd_get_active_mac(struct be_adapter *adapter, u32 curr_pmac_id,
+			  u8 *mac, u32 if_handle, bool active, u32 domain)
 {
 
 	if (!active)
@@ -2892,7 +2941,7 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array,
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_req_set_mac_list);
 	cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size,
-			&cmd.dma, GFP_KERNEL);
+				    &cmd.dma, GFP_KERNEL);
 	if (!cmd.va)
 		return -ENOMEM;
 
@@ -2906,8 +2955,8 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array,
 
 	req = cmd.va;
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-				OPCODE_COMMON_SET_MAC_LIST, sizeof(*req),
-				wrb, &cmd);
+			       OPCODE_COMMON_SET_MAC_LIST, sizeof(*req),
+			       wrb, &cmd);
 
 	req->hdr.domain = domain;
 	req->mac_count = mac_count;
@@ -2917,8 +2966,7 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array,
 	status = be_mcc_notify_wait(adapter);
 
 err:
-	dma_free_coherent(&adapter->pdev->dev, cmd.size,
-				cmd.va, cmd.dma);
+	dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma);
 	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
 }
@@ -2963,7 +3011,8 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid,
 	ctxt = &req->context;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_SET_HSW_CONFIG, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_SET_HSW_CONFIG, sizeof(*req), wrb,
+			       NULL);
 
 	req->hdr.domain = domain;
 	AMAP_SET_BITS(struct amap_set_hsw_context, interface_id, ctxt, intf_id);
@@ -3009,7 +3058,8 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid,
 	ctxt = &req->context;
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
-			OPCODE_COMMON_GET_HSW_CONFIG, sizeof(*req), wrb, NULL);
+			       OPCODE_COMMON_GET_HSW_CONFIG, sizeof(*req), wrb,
+			       NULL);
 
 	req->hdr.domain = domain;
 	AMAP_SET_BITS(struct amap_get_hsw_req_context, interface_id,
@@ -3027,10 +3077,9 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid,
 	if (!status) {
 		struct be_cmd_resp_get_hsw_config *resp =
 						embedded_payload(wrb);
-		be_dws_le_to_cpu(&resp->context,
-						sizeof(resp->context));
+		be_dws_le_to_cpu(&resp->context, sizeof(resp->context));
 		vid = AMAP_GET_BITS(struct amap_get_hsw_resp_context,
-							pvid, &resp->context);
+				    pvid, &resp->context);
 		if (pvid)
 			*pvid = le16_to_cpu(vid);
 		if (mode)
@@ -3062,11 +3111,9 @@ int be_cmd_get_acpi_wol_cap(struct be_adapter *adapter)
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_resp_acpi_wol_magic_config_v1);
-	cmd.va = pci_alloc_consistent(adapter->pdev, cmd.size,
-					       &cmd.dma);
+	cmd.va = pci_alloc_consistent(adapter->pdev, cmd.size, &cmd.dma);
 	if (!cmd.va) {
-		dev_err(&adapter->pdev->dev,
-				"Memory allocation failure\n");
+		dev_err(&adapter->pdev->dev, "Memory allocation failure\n");
 		status = -ENOMEM;
 		goto err;
 	}
@@ -3349,8 +3396,7 @@ int be_cmd_get_func_config(struct be_adapter *adapter, struct be_resources *res)
 
 	memset(&cmd, 0, sizeof(struct be_dma_mem));
 	cmd.size = sizeof(struct be_cmd_resp_get_func_config);
-	cmd.va = pci_alloc_consistent(adapter->pdev, cmd.size,
-				      &cmd.dma);
+	cmd.va = pci_alloc_consistent(adapter->pdev, cmd.size, &cmd.dma);
 	if (!cmd.va) {
 		dev_err(&adapter->pdev->dev, "Memory alloc failure\n");
 		status = -ENOMEM;
@@ -3396,7 +3442,7 @@ err:
 
 /* Uses mbox */
 static int be_cmd_get_profile_config_mbox(struct be_adapter *adapter,
-					u8 domain, struct be_dma_mem *cmd)
+					  u8 domain, struct be_dma_mem *cmd)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_get_profile_config *req;
@@ -3424,7 +3470,7 @@ static int be_cmd_get_profile_config_mbox(struct be_adapter *adapter,
 
 /* Uses sync mcc */
 static int be_cmd_get_profile_config_mccq(struct be_adapter *adapter,
-					u8 domain, struct be_dma_mem *cmd)
+					  u8 domain, struct be_dma_mem *cmd)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_get_profile_config *req;
@@ -3484,8 +3530,8 @@ int be_cmd_get_profile_config(struct be_adapter *adapter,
 	resp = cmd.va;
 	desc_count = le32_to_cpu(resp->desc_count);
 
-	pcie =  be_get_pcie_desc(adapter->pdev->devfn, resp->func_param,
-				 desc_count);
+	pcie = be_get_pcie_desc(adapter->pdev->devfn, resp->func_param,
+				desc_count);
 	if (pcie)
 		res->max_vfs = le16_to_cpu(pcie->num_vfs);
 
@@ -3548,33 +3594,47 @@ void be_reset_nic_desc(struct be_nic_res_desc *nic)
 	nic->cq_count = 0xFFFF;
 	nic->toe_conn_count = 0xFFFF;
 	nic->eq_count = 0xFFFF;
+	nic->iface_count = 0xFFFF;
 	nic->link_param = 0xFF;
+	nic->channel_id_param = cpu_to_le16(0xF000);
 	nic->acpi_params = 0xFF;
 	nic->wol_param = 0x0F;
-	nic->bw_min = 0xFFFFFFFF;
+	nic->tunnel_iface_count = 0xFFFF;
+	nic->direct_tenant_iface_count = 0xFFFF;
 	nic->bw_max = 0xFFFFFFFF;
 }
 
-int be_cmd_config_qos(struct be_adapter *adapter, u32 bps, u8 domain)
+int be_cmd_config_qos(struct be_adapter *adapter, u32 max_rate, u16 link_speed,
+		      u8 domain)
 {
-	if (lancer_chip(adapter)) {
-		struct be_nic_res_desc nic_desc;
+	struct be_nic_res_desc nic_desc;
+	u32 bw_percent;
+	u16 version = 0;
+
+	if (BE3_chip(adapter))
+		return be_cmd_set_qos(adapter, max_rate / 10, domain);
 
-		be_reset_nic_desc(&nic_desc);
+	be_reset_nic_desc(&nic_desc);
+	nic_desc.pf_num = adapter->pf_number;
+	nic_desc.vf_num = domain;
+	if (lancer_chip(adapter)) {
 		nic_desc.hdr.desc_type = NIC_RESOURCE_DESC_TYPE_V0;
 		nic_desc.hdr.desc_len = RESOURCE_DESC_SIZE_V0;
 		nic_desc.flags = (1 << QUN_SHIFT) | (1 << IMM_SHIFT) |
 					(1 << NOSV_SHIFT);
-		nic_desc.pf_num = adapter->pf_number;
-		nic_desc.vf_num = domain;
-		nic_desc.bw_max = cpu_to_le32(bps);
-
-		return be_cmd_set_profile_config(adapter, &nic_desc,
-						 RESOURCE_DESC_SIZE_V0,
-						 0, domain);
+		nic_desc.bw_max = cpu_to_le32(max_rate / 10);
 	} else {
-		return be_cmd_set_qos(adapter, bps, domain);
+		version = 1;
+		nic_desc.hdr.desc_type = NIC_RESOURCE_DESC_TYPE_V1;
+		nic_desc.hdr.desc_len = RESOURCE_DESC_SIZE_V1;
+		nic_desc.flags = (1 << IMM_SHIFT) | (1 << NOSV_SHIFT);
+		bw_percent = max_rate ? (max_rate * 100) / link_speed : 100;
+		nic_desc.bw_max = cpu_to_le32(bw_percent);
 	}
+
+	return be_cmd_set_profile_config(adapter, &nic_desc,
+					 nic_desc.hdr.desc_len,
+					 version, domain);
 }
 
 int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op)
@@ -3859,7 +3919,7 @@ err:
 }
 
 int be_roce_mcc_cmd(void *netdev_handle, void *wrb_payload,
-			int wrb_payload_size, u16 *cmd_status, u16 *ext_status)
+		    int wrb_payload_size, u16 *cmd_status, u16 *ext_status)
 {
 	struct be_adapter *adapter = netdev_priv(netdev_handle);
 	struct be_mcc_wrb *wrb;
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index b60e4d53c1c9..59b3c056f329 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -50,7 +50,7 @@ struct be_mcc_wrb {
 #define CQE_FLAGS_CONSUMED_MASK 	(1 << 27)
 
 /* Completion Status */
-enum {
+enum mcc_base_status {
 	MCC_STATUS_SUCCESS = 0,
 	MCC_STATUS_FAILED = 1,
 	MCC_STATUS_ILLEGAL_REQUEST = 2,
@@ -60,12 +60,25 @@ enum {
 	MCC_STATUS_NOT_SUPPORTED = 66
 };
 
-#define MCC_ADDL_STS_INSUFFICIENT_RESOURCES	0x16
+/* Additional status */
+enum mcc_addl_status {
+	MCC_ADDL_STATUS_INSUFFICIENT_RESOURCES = 0x16,
+	MCC_ADDL_STATUS_FLASH_IMAGE_CRC_MISMATCH = 0x4d,
+	MCC_ADDL_STATUS_TOO_MANY_INTERFACES = 0x4a
+};
+
+#define CQE_BASE_STATUS_MASK		0xFFFF
+#define CQE_BASE_STATUS_SHIFT		0	/* bits 0 - 15 */
+#define CQE_ADDL_STATUS_MASK		0xFF
+#define CQE_ADDL_STATUS_SHIFT		16	/* bits 16 - 31 */
 
-#define CQE_STATUS_COMPL_MASK		0xFFFF
-#define CQE_STATUS_COMPL_SHIFT		0	/* bits 0 - 15 */
-#define CQE_STATUS_EXTD_MASK		0xFFFF
-#define CQE_STATUS_EXTD_SHIFT		16	/* bits 16 - 31 */
+#define base_status(status)		\
+		((enum mcc_base_status)	\
+			(status > 0 ? (status & CQE_BASE_STATUS_MASK) : 0))
+#define addl_status(status)		\
+		((enum mcc_addl_status)	\
+			(status > 0 ? (status >> CQE_ADDL_STATUS_SHIFT) & \
+					CQE_ADDL_STATUS_MASK : 0))
 
 struct be_mcc_compl {
 	u32 status;		/* dword 0 */
@@ -74,13 +87,13 @@ struct be_mcc_compl {
 	u32 flags;		/* dword 3 */
 };
 
-/* When the async bit of mcc_compl is set, the last 4 bytes of
- * mcc_compl is interpreted as follows:
+/* When the async bit of mcc_compl flags is set, flags
+ * is interpreted as follows:
  */
-#define ASYNC_TRAILER_EVENT_CODE_SHIFT	8	/* bits 8 - 15 */
-#define ASYNC_TRAILER_EVENT_CODE_MASK	0xFF
-#define ASYNC_TRAILER_EVENT_TYPE_SHIFT	16
-#define ASYNC_TRAILER_EVENT_TYPE_MASK	0xFF
+#define ASYNC_EVENT_CODE_SHIFT		8	/* bits 8 - 15 */
+#define ASYNC_EVENT_CODE_MASK		0xFF
+#define ASYNC_EVENT_TYPE_SHIFT		16
+#define ASYNC_EVENT_TYPE_MASK		0xFF
 #define ASYNC_EVENT_CODE_LINK_STATE	0x1
 #define ASYNC_EVENT_CODE_GRP_5		0x5
 #define ASYNC_EVENT_QOS_SPEED		0x1
@@ -89,10 +102,6 @@ struct be_mcc_compl {
 #define ASYNC_EVENT_CODE_QNQ		0x6
 #define ASYNC_DEBUG_EVENT_TYPE_QNQ	1
 
-struct be_async_event_trailer {
-	u32 code;
-};
-
 enum {
 	LINK_DOWN	= 0x0,
 	LINK_UP		= 0x1
@@ -100,7 +109,7 @@ enum {
 #define LINK_STATUS_MASK			0x1
 #define LOGICAL_LINK_STATUS_MASK		0x2
 
-/* When the event code of an async trailer is link-state, the mcc_compl
+/* When the event code of compl->flags is link-state, the mcc_compl
  * must be interpreted as follows
  */
 struct be_async_event_link_state {
@@ -110,10 +119,10 @@ struct be_async_event_link_state {
 	u8 port_speed;
 	u8 port_fault;
 	u8 rsvd0[7];
-	struct be_async_event_trailer trailer;
+	u32 flags;
 } __packed;
 
-/* When the event code of an async trailer is GRP-5 and event_type is QOS_SPEED
+/* When the event code of compl->flags is GRP-5 and event_type is QOS_SPEED
  * the mcc_compl must be interpreted as follows
  */
 struct be_async_event_grp5_qos_link_speed {
@@ -121,10 +130,10 @@ struct be_async_event_grp5_qos_link_speed {
 	u8 rsvd[5];
 	u16 qos_link_speed;
 	u32 event_tag;
-	struct be_async_event_trailer trailer;
+	u32 flags;
 } __packed;
 
-/* When the event code of an async trailer is GRP5 and event type is
+/* When the event code of compl->flags is GRP5 and event type is
  * CoS-Priority, the mcc_compl must be interpreted as follows
  */
 struct be_async_event_grp5_cos_priority {
@@ -134,10 +143,10 @@ struct be_async_event_grp5_cos_priority {
 	u8 valid;
 	u8 rsvd0;
 	u8 event_tag;
-	struct be_async_event_trailer trailer;
+	u32 flags;
 } __packed;
 
-/* When the event code of an async trailer is GRP5 and event type is
+/* When the event code of compl->flags is GRP5 and event type is
  * PVID state, the mcc_compl must be interpreted as follows
  */
 struct be_async_event_grp5_pvid_state {
@@ -146,7 +155,7 @@ struct be_async_event_grp5_pvid_state {
 	u16 tag;
 	u32 event_tag;
 	u32 rsvd1;
-	struct be_async_event_trailer trailer;
+	u32 flags;
 } __packed;
 
 /* async event indicating outer VLAN tag in QnQ */
@@ -156,7 +165,7 @@ struct be_async_event_qnq {
 	u16 vlan_tag;
 	u32 event_tag;
 	u8 rsvd1[4];
-	struct be_async_event_trailer trailer;
+	u32 flags;
 } __packed;
 
 struct be_mcc_mailbox {
@@ -258,8 +267,8 @@ struct be_cmd_resp_hdr {
 	u8 opcode;		/* dword 0 */
 	u8 subsystem;		/* dword 0 */
 	u8 rsvd[2];		/* dword 0 */
-	u8 status;		/* dword 1 */
-	u8 add_status;		/* dword 1 */
+	u8 base_status;		/* dword 1 */
+	u8 addl_status;		/* dword 1 */
 	u8 rsvd1[2];		/* dword 1 */
 	u32 response_length;	/* dword 2 */
 	u32 actual_resp_len;	/* dword 3 */
@@ -1082,7 +1091,7 @@ struct be_cmd_resp_modify_eq_delay {
  * based on the skew/IPL.
  */
 #define RDMA_ENABLED				0x4
-#define FLEX10_MODE				0x400
+#define QNQ_MODE				0x400
 #define VNIC_MODE				0x20000
 #define UMC_ENABLED				0x1000000
 struct be_cmd_req_query_fw_cfg {
@@ -1186,7 +1195,8 @@ struct be_cmd_read_flash_crc {
 	struct flashrom_params params;
 	u8 crc[4];
 	u8 rsvd[4];
-};
+} __packed;
+
 /**************** Lancer Firmware Flash ************/
 struct amap_lancer_write_obj_context {
 	u8 write_length[24];
@@ -1891,16 +1901,20 @@ struct be_nic_res_desc {
 	u16 cq_count;
 	u16 toe_conn_count;
 	u16 eq_count;
-	u32 rsvd5;
+	u16 vlan_id;
+	u16 iface_count;
 	u32 cap_flags;
 	u8 link_param;
-	u8 rsvd6[3];
+	u8 rsvd6;
+	u16 channel_id_param;
 	u32 bw_min;
 	u32 bw_max;
 	u8 acpi_params;
 	u8 wol_param;
 	u16 rsvd7;
-	u32 rsvd8[7];
+	u16 tunnel_iface_count;
+	u16 direct_tenant_iface_count;
+	u32 rsvd8[6];
 } __packed;
 
 /************ Multi-Channel type ***********/
@@ -2060,7 +2074,7 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter, char *fw_ver,
 		      char *fw_on_flash);
 int be_cmd_modify_eqd(struct be_adapter *adapter, struct be_set_eqd *, int num);
 int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
-		       u32 num, bool promiscuous);
+		       u32 num);
 int be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 status);
 int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc);
 int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc);
@@ -2068,7 +2082,7 @@ int be_cmd_query_fw_cfg(struct be_adapter *adapter, u32 *port_num,
 			u32 *function_mode, u32 *function_caps, u16 *asic_rev);
 int be_cmd_reset_function(struct be_adapter *adapter);
 int be_cmd_rss_config(struct be_adapter *adapter, u8 *rsstable,
-		      u32 rss_hash_opts, u16 table_size);
+		      u32 rss_hash_opts, u16 table_size, const u8 *rss_hkey);
 int be_process_mcc(struct be_adapter *adapter);
 int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num, u8 beacon,
 			    u8 status, u8 state);
@@ -2084,7 +2098,7 @@ int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 			   u32 data_size, u32 data_offset, const char *obj_name,
 			   u32 *data_read, u32 *eof, u8 *addn_status);
 int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc,
-			 int offset);
+			  u16 optype, int offset);
 int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac,
 			    struct be_dma_mem *nonemb_cmd);
 int be_cmd_fw_init(struct be_adapter *adapter);
@@ -2101,7 +2115,8 @@ int be_cmd_get_seeprom_data(struct be_adapter *adapter,
 int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num,
 			u8 loopback_type, u8 enable);
 int be_cmd_get_phy_info(struct be_adapter *adapter);
-int be_cmd_config_qos(struct be_adapter *adapter, u32 bps, u8 domain);
+int be_cmd_config_qos(struct be_adapter *adapter, u32 max_rate,
+		      u16 link_speed, u8 domain);
 void be_detect_error(struct be_adapter *adapter);
 int be_cmd_get_die_temperature(struct be_adapter *adapter);
 int be_cmd_get_cntl_attributes(struct be_adapter *adapter);
diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index 15ba96cba65d..e2da4d20dd3d 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -132,6 +132,7 @@ static const struct be_ethtool_stat et_rx_stats[] = {
 	{DRVSTAT_RX_INFO(rx_bytes)},/* If moving this member see above note */
 	{DRVSTAT_RX_INFO(rx_pkts)}, /* If moving this member see above note */
 	{DRVSTAT_RX_INFO(rx_compl)},
+	{DRVSTAT_RX_INFO(rx_compl_err)},
 	{DRVSTAT_RX_INFO(rx_mcast_pkts)},
 	/* Number of page allocation failures while posting receive buffers
 	 * to HW.
@@ -181,7 +182,7 @@ static const char et_self_tests[][ETH_GSTRING_LEN] = {
 #define BE_NO_LOOPBACK 0xff
 
 static void be_get_drvinfo(struct net_device *netdev,
-				struct ethtool_drvinfo *drvinfo)
+			   struct ethtool_drvinfo *drvinfo)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -201,8 +202,7 @@ static void be_get_drvinfo(struct net_device *netdev,
 	drvinfo->eedump_len = 0;
 }
 
-static u32
-lancer_cmd_get_file_len(struct be_adapter *adapter, u8 *file_name)
+static u32 lancer_cmd_get_file_len(struct be_adapter *adapter, u8 *file_name)
 {
 	u32 data_read = 0, eof;
 	u8 addn_status;
@@ -212,14 +212,14 @@ lancer_cmd_get_file_len(struct be_adapter *adapter, u8 *file_name)
 	memset(&data_len_cmd, 0, sizeof(data_len_cmd));
 	/* data_offset and data_size should be 0 to get reg len */
 	status = lancer_cmd_read_object(adapter, &data_len_cmd, 0, 0,
-				file_name, &data_read, &eof, &addn_status);
+					file_name, &data_read, &eof,
+					&addn_status);
 
 	return data_read;
 }
 
-static int
-lancer_cmd_read_file(struct be_adapter *adapter, u8 *file_name,
-		u32 buf_len, void *buf)
+static int lancer_cmd_read_file(struct be_adapter *adapter, u8 *file_name,
+				u32 buf_len, void *buf)
 {
 	struct be_dma_mem read_cmd;
 	u32 read_len = 0, total_read_len = 0, chunk_size;
@@ -229,11 +229,11 @@ lancer_cmd_read_file(struct be_adapter *adapter, u8 *file_name,
 
 	read_cmd.size = LANCER_READ_FILE_CHUNK;
 	read_cmd.va = pci_alloc_consistent(adapter->pdev, read_cmd.size,
-			&read_cmd.dma);
+					   &read_cmd.dma);
 
 	if (!read_cmd.va) {
 		dev_err(&adapter->pdev->dev,
-				"Memory allocation failure while reading dump\n");
+			"Memory allocation failure while reading dump\n");
 		return -ENOMEM;
 	}
 
@@ -242,8 +242,8 @@ lancer_cmd_read_file(struct be_adapter *adapter, u8 *file_name,
 				LANCER_READ_FILE_CHUNK);
 		chunk_size = ALIGN(chunk_size, 4);
 		status = lancer_cmd_read_object(adapter, &read_cmd, chunk_size,
-				total_read_len, file_name, &read_len,
-				&eof, &addn_status);
+						total_read_len, file_name,
+						&read_len, &eof, &addn_status);
 		if (!status) {
 			memcpy(buf + total_read_len, read_cmd.va, read_len);
 			total_read_len += read_len;
@@ -254,13 +254,12 @@ lancer_cmd_read_file(struct be_adapter *adapter, u8 *file_name,
 		}
 	}
 	pci_free_consistent(adapter->pdev, read_cmd.size, read_cmd.va,
-			read_cmd.dma);
+			    read_cmd.dma);
 
 	return status;
 }
 
-static int
-be_get_reg_len(struct net_device *netdev)
+static int be_get_reg_len(struct net_device *netdev)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	u32 log_size = 0;
@@ -271,7 +270,7 @@ be_get_reg_len(struct net_device *netdev)
 	if (be_physfn(adapter)) {
 		if (lancer_chip(adapter))
 			log_size = lancer_cmd_get_file_len(adapter,
-					LANCER_FW_DUMP_FILE);
+							   LANCER_FW_DUMP_FILE);
 		else
 			be_cmd_get_reg_len(adapter, &log_size);
 	}
@@ -287,7 +286,7 @@ be_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *buf)
 		memset(buf, 0, regs->len);
 		if (lancer_chip(adapter))
 			lancer_cmd_read_file(adapter, LANCER_FW_DUMP_FILE,
-					regs->len, buf);
+					     regs->len, buf);
 		else
 			be_cmd_get_regs(adapter, regs->len, buf);
 	}
@@ -337,9 +336,8 @@ static int be_set_coalesce(struct net_device *netdev,
 	return 0;
 }
 
-static void
-be_get_ethtool_stats(struct net_device *netdev,
-		struct ethtool_stats *stats, uint64_t *data)
+static void be_get_ethtool_stats(struct net_device *netdev,
+				 struct ethtool_stats *stats, uint64_t *data)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_rx_obj *rxo;
@@ -390,9 +388,8 @@ be_get_ethtool_stats(struct net_device *netdev,
 	}
 }
 
-static void
-be_get_stat_strings(struct net_device *netdev, uint32_t stringset,
-		uint8_t *data)
+static void be_get_stat_strings(struct net_device *netdev, uint32_t stringset,
+				uint8_t *data)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	int i, j;
@@ -642,16 +639,15 @@ be_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *ecmd)
 	adapter->rx_fc = ecmd->rx_pause;
 
 	status = be_cmd_set_flow_control(adapter,
-					adapter->tx_fc, adapter->rx_fc);
+					 adapter->tx_fc, adapter->rx_fc);
 	if (status)
 		dev_warn(&adapter->pdev->dev, "Pause param set failed.\n");
 
 	return status;
 }
 
-static int
-be_set_phys_id(struct net_device *netdev,
-	       enum ethtool_phys_id_state state)
+static int be_set_phys_id(struct net_device *netdev,
+			  enum ethtool_phys_id_state state)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -708,8 +704,7 @@ static int be_set_dump(struct net_device *netdev, struct ethtool_dump *dump)
 	return status;
 }
 
-static void
-be_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+static void be_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -723,8 +718,7 @@ be_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 	memset(&wol->sopass, 0, sizeof(wol->sopass));
 }
 
-static int
-be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+static int be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -744,8 +738,7 @@ be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 	return 0;
 }
 
-static int
-be_test_ddr_dma(struct be_adapter *adapter)
+static int be_test_ddr_dma(struct be_adapter *adapter)
 {
 	int ret, i;
 	struct be_dma_mem ddrdma_cmd;
@@ -761,7 +754,7 @@ be_test_ddr_dma(struct be_adapter *adapter)
 
 	for (i = 0; i < 2; i++) {
 		ret = be_cmd_ddr_dma_test(adapter, pattern[i],
-					4096, &ddrdma_cmd);
+					  4096, &ddrdma_cmd);
 		if (ret != 0)
 			goto err;
 	}
@@ -773,20 +766,17 @@ err:
 }
 
 static u64 be_loopback_test(struct be_adapter *adapter, u8 loopback_type,
-				u64 *status)
+			    u64 *status)
 {
-	be_cmd_set_loopback(adapter, adapter->hba_port_num,
-				loopback_type, 1);
+	be_cmd_set_loopback(adapter, adapter->hba_port_num, loopback_type, 1);
 	*status = be_cmd_loopback_test(adapter, adapter->hba_port_num,
-				loopback_type, 1500,
-				2, 0xabc);
-	be_cmd_set_loopback(adapter, adapter->hba_port_num,
-				BE_NO_LOOPBACK, 1);
+				       loopback_type, 1500, 2, 0xabc);
+	be_cmd_set_loopback(adapter, adapter->hba_port_num, BE_NO_LOOPBACK, 1);
 	return *status;
 }
 
-static void
-be_self_test(struct net_device *netdev, struct ethtool_test *test, u64 *data)
+static void be_self_test(struct net_device *netdev, struct ethtool_test *test,
+			 u64 *data)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	int status;
@@ -801,12 +791,10 @@ be_self_test(struct net_device *netdev, struct ethtool_test *test, u64 *data)
 	memset(data, 0, sizeof(u64) * ETHTOOL_TESTS_NUM);
 
 	if (test->flags & ETH_TEST_FL_OFFLINE) {
-		if (be_loopback_test(adapter, BE_MAC_LOOPBACK,
-				     &data[0]) != 0)
+		if (be_loopback_test(adapter, BE_MAC_LOOPBACK, &data[0]) != 0)
 			test->flags |= ETH_TEST_FL_FAILED;
 
-		if (be_loopback_test(adapter, BE_PHY_LOOPBACK,
-				     &data[1]) != 0)
+		if (be_loopback_test(adapter, BE_PHY_LOOPBACK, &data[1]) != 0)
 			test->flags |= ETH_TEST_FL_FAILED;
 
 		if (test->flags & ETH_TEST_FL_EXTERNAL_LB) {
@@ -832,16 +820,14 @@ be_self_test(struct net_device *netdev, struct ethtool_test *test, u64 *data)
 	}
 }
 
-static int
-be_do_flash(struct net_device *netdev, struct ethtool_flash *efl)
+static int be_do_flash(struct net_device *netdev, struct ethtool_flash *efl)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
 	return be_load_fw(adapter, efl->data);
 }
 
-static int
-be_get_eeprom_len(struct net_device *netdev)
+static int be_get_eeprom_len(struct net_device *netdev)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -851,18 +837,17 @@ be_get_eeprom_len(struct net_device *netdev)
 	if (lancer_chip(adapter)) {
 		if (be_physfn(adapter))
 			return lancer_cmd_get_file_len(adapter,
-					LANCER_VPD_PF_FILE);
+						       LANCER_VPD_PF_FILE);
 		else
 			return lancer_cmd_get_file_len(adapter,
-					LANCER_VPD_VF_FILE);
+						       LANCER_VPD_VF_FILE);
 	} else {
 		return BE_READ_SEEPROM_LEN;
 	}
 }
 
-static int
-be_read_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom,
-			uint8_t *data)
+static int be_read_eeprom(struct net_device *netdev,
+			  struct ethtool_eeprom *eeprom, uint8_t *data)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_dma_mem eeprom_cmd;
@@ -875,10 +860,10 @@ be_read_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom,
 	if (lancer_chip(adapter)) {
 		if (be_physfn(adapter))
 			return lancer_cmd_read_file(adapter, LANCER_VPD_PF_FILE,
-					eeprom->len, data);
+						    eeprom->len, data);
 		else
 			return lancer_cmd_read_file(adapter, LANCER_VPD_VF_FILE,
-					eeprom->len, data);
+						    eeprom->len, data);
 	}
 
 	eeprom->magic = BE_VENDOR_ID | (adapter->pdev->device<<16);
@@ -933,27 +918,27 @@ static u64 be_get_rss_hash_opts(struct be_adapter *adapter, u64 flow_type)
 
 	switch (flow_type) {
 	case TCP_V4_FLOW:
-		if (adapter->rss_flags & RSS_ENABLE_IPV4)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_IPV4)
 			data |= RXH_IP_DST | RXH_IP_SRC;
-		if (adapter->rss_flags & RSS_ENABLE_TCP_IPV4)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_TCP_IPV4)
 			data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 	case UDP_V4_FLOW:
-		if (adapter->rss_flags & RSS_ENABLE_IPV4)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_IPV4)
 			data |= RXH_IP_DST | RXH_IP_SRC;
-		if (adapter->rss_flags & RSS_ENABLE_UDP_IPV4)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_UDP_IPV4)
 			data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 	case TCP_V6_FLOW:
-		if (adapter->rss_flags & RSS_ENABLE_IPV6)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_IPV6)
 			data |= RXH_IP_DST | RXH_IP_SRC;
-		if (adapter->rss_flags & RSS_ENABLE_TCP_IPV6)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_TCP_IPV6)
 			data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 	case UDP_V6_FLOW:
-		if (adapter->rss_flags & RSS_ENABLE_IPV6)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_IPV6)
 			data |= RXH_IP_DST | RXH_IP_SRC;
-		if (adapter->rss_flags & RSS_ENABLE_UDP_IPV6)
+		if (adapter->rss_info.rss_flags & RSS_ENABLE_UDP_IPV6)
 			data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 	}
@@ -962,7 +947,7 @@ static u64 be_get_rss_hash_opts(struct be_adapter *adapter, u64 flow_type)
 }
 
 static int be_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
-		      u32 *rule_locs)
+			u32 *rule_locs)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 
@@ -992,7 +977,7 @@ static int be_set_rss_hash_opts(struct be_adapter *adapter,
 	struct be_rx_obj *rxo;
 	int status = 0, i, j;
 	u8 rsstable[128];
-	u32 rss_flags = adapter->rss_flags;
+	u32 rss_flags = adapter->rss_info.rss_flags;
 
 	if (cmd->data != L3_RSS_FLAGS &&
 	    cmd->data != (L3_RSS_FLAGS | L4_RSS_FLAGS))
@@ -1039,7 +1024,7 @@ static int be_set_rss_hash_opts(struct be_adapter *adapter,
 		return -EINVAL;
 	}
 
-	if (rss_flags == adapter->rss_flags)
+	if (rss_flags == adapter->rss_info.rss_flags)
 		return status;
 
 	if (be_multi_rxq(adapter)) {
@@ -1051,9 +1036,11 @@ static int be_set_rss_hash_opts(struct be_adapter *adapter,
 			}
 		}
 	}
-	status = be_cmd_rss_config(adapter, rsstable, rss_flags, 128);
+
+	status = be_cmd_rss_config(adapter, adapter->rss_info.rsstable,
+				   rss_flags, 128, adapter->rss_info.rss_hkey);
 	if (!status)
-		adapter->rss_flags = rss_flags;
+		adapter->rss_info.rss_flags = rss_flags;
 
 	return status;
 }
@@ -1103,6 +1090,69 @@ static int be_set_channels(struct net_device  *netdev,
 	return be_update_queues(adapter);
 }
 
+static u32 be_get_rxfh_indir_size(struct net_device *netdev)
+{
+	return RSS_INDIR_TABLE_LEN;
+}
+
+static u32 be_get_rxfh_key_size(struct net_device *netdev)
+{
+	return RSS_HASH_KEY_LEN;
+}
+
+static int be_get_rxfh(struct net_device *netdev, u32 *indir, u8 *hkey)
+{
+	struct be_adapter *adapter = netdev_priv(netdev);
+	int i;
+	struct rss_info *rss = &adapter->rss_info;
+
+	if (indir) {
+		for (i = 0; i < RSS_INDIR_TABLE_LEN; i++)
+			indir[i] = rss->rss_queue[i];
+	}
+
+	if (hkey)
+		memcpy(hkey, rss->rss_hkey, RSS_HASH_KEY_LEN);
+
+	return 0;
+}
+
+static int be_set_rxfh(struct net_device *netdev, const u32 *indir,
+		       const u8 *hkey)
+{
+	int rc = 0, i, j;
+	struct be_adapter *adapter = netdev_priv(netdev);
+	u8 rsstable[RSS_INDIR_TABLE_LEN];
+
+	if (indir) {
+		struct be_rx_obj *rxo;
+		for (i = 0; i < RSS_INDIR_TABLE_LEN; i++) {
+			j = indir[i];
+			rxo = &adapter->rx_obj[j];
+			rsstable[i] = rxo->rss_id;
+			adapter->rss_info.rss_queue[i] = j;
+		}
+	} else {
+		memcpy(rsstable, adapter->rss_info.rsstable,
+		       RSS_INDIR_TABLE_LEN);
+	}
+
+	if (!hkey)
+		hkey =  adapter->rss_info.rss_hkey;
+
+	rc = be_cmd_rss_config(adapter, rsstable,
+			adapter->rss_info.rss_flags,
+			RSS_INDIR_TABLE_LEN, hkey);
+	if (rc) {
+		adapter->rss_info.rss_flags = RSS_ENABLE_NONE;
+		return -EIO;
+	}
+	memcpy(adapter->rss_info.rss_hkey, hkey, RSS_HASH_KEY_LEN);
+	memcpy(adapter->rss_info.rsstable, rsstable,
+	       RSS_INDIR_TABLE_LEN);
+	return 0;
+}
+
 const struct ethtool_ops be_ethtool_ops = {
 	.get_settings = be_get_settings,
 	.get_drvinfo = be_get_drvinfo,
@@ -1129,6 +1179,10 @@ const struct ethtool_ops be_ethtool_ops = {
 	.self_test = be_self_test,
 	.get_rxnfc = be_get_rxnfc,
 	.set_rxnfc = be_set_rxnfc,
+	.get_rxfh_indir_size = be_get_rxfh_indir_size,
+	.get_rxfh_key_size = be_get_rxfh_key_size,
+	.get_rxfh = be_get_rxfh,
+	.set_rxfh = be_set_rxfh,
 	.get_channels = be_get_channels,
 	.set_channels = be_set_channels
 };
diff --git a/drivers/net/ethernet/emulex/benet/be_hw.h b/drivers/net/ethernet/emulex/benet/be_hw.h
index 3bd198550edb..8840c64aaeca 100644
--- a/drivers/net/ethernet/emulex/benet/be_hw.h
+++ b/drivers/net/ethernet/emulex/benet/be_hw.h
@@ -188,10 +188,14 @@
 #define OPTYPE_FCOE_FW_ACTIVE		10
 #define OPTYPE_FCOE_FW_BACKUP		11
 #define OPTYPE_NCSI_FW			13
+#define OPTYPE_REDBOOT_DIR		18
+#define OPTYPE_REDBOOT_CONFIG		19
+#define OPTYPE_SH_PHY_FW		21
+#define OPTYPE_FLASHISM_JUMPVECTOR	22
+#define OPTYPE_UFI_DIR			23
 #define OPTYPE_PHY_FW			99
 #define TN_8022				13
 
-#define ILLEGAL_IOCTL_REQ		2
 #define FLASHROM_OPER_PHY_FLASH		9
 #define FLASHROM_OPER_PHY_SAVE		10
 #define FLASHROM_OPER_FLASH		1
@@ -250,6 +254,9 @@
 #define IMAGE_FIRMWARE_BACKUP_FCoE	178
 #define IMAGE_FIRMWARE_BACKUP_COMP_FCoE 179
 #define IMAGE_FIRMWARE_PHY		192
+#define IMAGE_REDBOOT_DIR		208
+#define IMAGE_REDBOOT_CONFIG		209
+#define IMAGE_UFI_DIR			210
 #define IMAGE_BOOT_CODE			224
 
 /************* Rx Packet Type Encoding **************/
@@ -534,7 +541,8 @@ struct flash_section_entry {
 	u32 image_size;
 	u32 cksum;
 	u32 entry_point;
-	u32 rsvd0;
+	u16 optype;
+	u16 rsvd0;
 	u32 rsvd1;
 	u8 ver_data[32];
 } __packed;
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index dc19bc5dec77..34a26e42f19d 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -134,7 +134,7 @@ static void be_queue_free(struct be_adapter *adapter, struct be_queue_info *q)
 }
 
 static int be_queue_alloc(struct be_adapter *adapter, struct be_queue_info *q,
-		u16 len, u16 entry_size)
+			  u16 len, u16 entry_size)
 {
 	struct be_dma_mem *mem = &q->dma_mem;
 
@@ -154,7 +154,7 @@ static void be_reg_intr_set(struct be_adapter *adapter, bool enable)
 	u32 reg, enabled;
 
 	pci_read_config_dword(adapter->pdev, PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET,
-				&reg);
+			      &reg);
 	enabled = reg & MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK;
 
 	if (!enabled && enable)
@@ -165,7 +165,7 @@ static void be_reg_intr_set(struct be_adapter *adapter, bool enable)
 		return;
 
 	pci_write_config_dword(adapter->pdev,
-			PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET, reg);
+			       PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET, reg);
 }
 
 static void be_intr_set(struct be_adapter *adapter, bool enable)
@@ -206,12 +206,11 @@ static void be_txq_notify(struct be_adapter *adapter, struct be_tx_obj *txo,
 }
 
 static void be_eq_notify(struct be_adapter *adapter, u16 qid,
-		bool arm, bool clear_int, u16 num_popped)
+			 bool arm, bool clear_int, u16 num_popped)
 {
 	u32 val = 0;
 	val |= qid & DB_EQ_RING_ID_MASK;
-	val |= ((qid & DB_EQ_RING_ID_EXT_MASK) <<
-			DB_EQ_RING_ID_EXT_MASK_SHIFT);
+	val |= ((qid & DB_EQ_RING_ID_EXT_MASK) << DB_EQ_RING_ID_EXT_MASK_SHIFT);
 
 	if (adapter->eeh_error)
 		return;
@@ -477,7 +476,7 @@ static void populate_be_v2_stats(struct be_adapter *adapter)
 	drvs->rx_drops_no_tpre_descr = rxf_stats->rx_drops_no_tpre_descr;
 	drvs->rx_drops_too_many_frags = rxf_stats->rx_drops_too_many_frags;
 	adapter->drv_stats.eth_red_drops = pmem_sts->eth_red_drops;
-	if (be_roce_supported(adapter))  {
+	if (be_roce_supported(adapter)) {
 		drvs->rx_roce_bytes_lsd = port_stats->roce_bytes_received_lsd;
 		drvs->rx_roce_bytes_msd = port_stats->roce_bytes_received_msd;
 		drvs->rx_roce_frames = port_stats->roce_frames_received;
@@ -491,8 +490,7 @@ static void populate_lancer_stats(struct be_adapter *adapter)
 {
 
 	struct be_drv_stats *drvs = &adapter->drv_stats;
-	struct lancer_pport_stats *pport_stats =
-					pport_stats_from_cmd(adapter);
+	struct lancer_pport_stats *pport_stats = pport_stats_from_cmd(adapter);
 
 	be_dws_le_to_cpu(pport_stats, sizeof(*pport_stats));
 	drvs->rx_pause_frames = pport_stats->rx_pause_frames_lo;
@@ -539,8 +537,7 @@ static void accumulate_16bit_val(u32 *acc, u16 val)
 }
 
 static void populate_erx_stats(struct be_adapter *adapter,
-			struct be_rx_obj *rxo,
-			u32 erx_stat)
+			       struct be_rx_obj *rxo, u32 erx_stat)
 {
 	if (!BEx_chip(adapter))
 		rx_stats(rxo)->rx_drops_no_frags = erx_stat;
@@ -579,7 +576,7 @@ void be_parse_stats(struct be_adapter *adapter)
 }
 
 static struct rtnl_link_stats64 *be_get_stats64(struct net_device *netdev,
-					struct rtnl_link_stats64 *stats)
+						struct rtnl_link_stats64 *stats)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_drv_stats *drvs = &adapter->drv_stats;
@@ -660,7 +657,8 @@ void be_link_status_update(struct be_adapter *adapter, u8 link_status)
 }
 
 static void be_tx_stats_update(struct be_tx_obj *txo,
-			u32 wrb_cnt, u32 copied, u32 gso_segs, bool stopped)
+			       u32 wrb_cnt, u32 copied, u32 gso_segs,
+			       bool stopped)
 {
 	struct be_tx_stats *stats = tx_stats(txo);
 
@@ -676,7 +674,7 @@ static void be_tx_stats_update(struct be_tx_obj *txo,
 
 /* Determine number of WRB entries needed to xmit data in an skb */
 static u32 wrb_cnt_for_skb(struct be_adapter *adapter, struct sk_buff *skb,
-								bool *dummy)
+			   bool *dummy)
 {
 	int cnt = (skb->len > skb->data_len);
 
@@ -704,7 +702,7 @@ static inline void wrb_fill(struct be_eth_wrb *wrb, u64 addr, int len)
 }
 
 static inline u16 be_get_tx_vlan_tag(struct be_adapter *adapter,
-					struct sk_buff *skb)
+				     struct sk_buff *skb)
 {
 	u8 vlan_prio;
 	u16 vlan_tag;
@@ -733,7 +731,8 @@ static u16 skb_ip_proto(struct sk_buff *skb)
 }
 
 static void wrb_fill_hdr(struct be_adapter *adapter, struct be_eth_hdr_wrb *hdr,
-		struct sk_buff *skb, u32 wrb_cnt, u32 len, bool skip_hw_vlan)
+			 struct sk_buff *skb, u32 wrb_cnt, u32 len,
+			 bool skip_hw_vlan)
 {
 	u16 vlan_tag, proto;
 
@@ -774,7 +773,7 @@ static void wrb_fill_hdr(struct be_adapter *adapter, struct be_eth_hdr_wrb *hdr,
 }
 
 static void unmap_tx_frag(struct device *dev, struct be_eth_wrb *wrb,
-		bool unmap_single)
+			  bool unmap_single)
 {
 	dma_addr_t dma;
 
@@ -791,8 +790,8 @@ static void unmap_tx_frag(struct device *dev, struct be_eth_wrb *wrb,
 }
 
 static int make_tx_wrbs(struct be_adapter *adapter, struct be_queue_info *txq,
-		struct sk_buff *skb, u32 wrb_cnt, bool dummy_wrb,
-		bool skip_hw_vlan)
+			struct sk_buff *skb, u32 wrb_cnt, bool dummy_wrb,
+			bool skip_hw_vlan)
 {
 	dma_addr_t busaddr;
 	int i, copied = 0;
@@ -821,8 +820,7 @@ static int make_tx_wrbs(struct be_adapter *adapter, struct be_queue_info *txq,
 	}
 
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		const struct skb_frag_struct *frag =
-			&skb_shinfo(skb)->frags[i];
+		const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
 		busaddr = skb_frag_dma_map(dev, frag, 0,
 					   skb_frag_size(frag), DMA_TO_DEVICE);
 		if (dma_mapping_error(dev, busaddr))
@@ -927,8 +925,7 @@ static int be_vlan_tag_tx_chk(struct be_adapter *adapter, struct sk_buff *skb)
 	return vlan_tx_tag_present(skb) || adapter->pvid || adapter->qnq_vid;
 }
 
-static int be_ipv6_tx_stall_chk(struct be_adapter *adapter,
-				struct sk_buff *skb)
+static int be_ipv6_tx_stall_chk(struct be_adapter *adapter, struct sk_buff *skb)
 {
 	return BE3_chip(adapter) && be_ipv6_exthdr_check(skb);
 }
@@ -959,7 +956,7 @@ static struct sk_buff *be_lancer_xmit_workarounds(struct be_adapter *adapter,
 	 */
 	if (be_pvid_tagging_enabled(adapter) &&
 	    veh->h_vlan_proto == htons(ETH_P_8021Q))
-			*skip_hw_vlan = true;
+		*skip_hw_vlan = true;
 
 	/* HW has a bug wherein it will calculate CSUM for VLAN
 	 * pkts even though it is disabled.
@@ -1077,16 +1074,15 @@ static int be_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	if (new_mtu < BE_MIN_MTU ||
-			new_mtu > (BE_MAX_JUMBO_FRAME_SIZE -
-					(ETH_HLEN + ETH_FCS_LEN))) {
+	    new_mtu > (BE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN))) {
 		dev_info(&adapter->pdev->dev,
-			"MTU must be between %d and %d bytes\n",
-			BE_MIN_MTU,
-			(BE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN)));
+			 "MTU must be between %d and %d bytes\n",
+			 BE_MIN_MTU,
+			 (BE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN)));
 		return -EINVAL;
 	}
 	dev_info(&adapter->pdev->dev, "MTU changed from %d to %d bytes\n",
-			netdev->mtu, new_mtu);
+		 netdev->mtu, new_mtu);
 	netdev->mtu = new_mtu;
 	return 0;
 }
@@ -1098,7 +1094,7 @@ static int be_change_mtu(struct net_device *netdev, int new_mtu)
 static int be_vid_config(struct be_adapter *adapter)
 {
 	u16 vids[BE_NUM_VLANS_SUPPORTED];
-	u16 num = 0, i;
+	u16 num = 0, i = 0;
 	int status = 0;
 
 	/* No need to further configure vids if in promiscuous mode */
@@ -1109,16 +1105,14 @@ static int be_vid_config(struct be_adapter *adapter)
 		goto set_vlan_promisc;
 
 	/* Construct VLAN Table to give to HW */
-	for (i = 0; i < VLAN_N_VID; i++)
-		if (adapter->vlan_tag[i])
-			vids[num++] = cpu_to_le16(i);
-
-	status = be_cmd_vlan_config(adapter, adapter->if_handle,
-				    vids, num, 0);
+	for_each_set_bit(i, adapter->vids, VLAN_N_VID)
+		vids[num++] = cpu_to_le16(i);
 
+	status = be_cmd_vlan_config(adapter, adapter->if_handle, vids, num);
 	if (status) {
 		/* Set to VLAN promisc mode as setting VLAN filter failed */
-		if (status == MCC_ADDL_STS_INSUFFICIENT_RESOURCES)
+		if (addl_status(status) ==
+				MCC_ADDL_STATUS_INSUFFICIENT_RESOURCES)
 			goto set_vlan_promisc;
 		dev_err(&adapter->pdev->dev,
 			"Setting HW VLAN filtering failed.\n");
@@ -1160,16 +1154,16 @@ static int be_vlan_add_vid(struct net_device *netdev, __be16 proto, u16 vid)
 	if (lancer_chip(adapter) && vid == 0)
 		return status;
 
-	if (adapter->vlan_tag[vid])
+	if (test_bit(vid, adapter->vids))
 		return status;
 
-	adapter->vlan_tag[vid] = 1;
+	set_bit(vid, adapter->vids);
 	adapter->vlans_added++;
 
 	status = be_vid_config(adapter);
 	if (status) {
 		adapter->vlans_added--;
-		adapter->vlan_tag[vid] = 0;
+		clear_bit(vid, adapter->vids);
 	}
 
 	return status;
@@ -1184,12 +1178,12 @@ static int be_vlan_rem_vid(struct net_device *netdev, __be16 proto, u16 vid)
 	if (lancer_chip(adapter) && vid == 0)
 		goto ret;
 
-	adapter->vlan_tag[vid] = 0;
+	clear_bit(vid, adapter->vids);
 	status = be_vid_config(adapter);
 	if (!status)
 		adapter->vlans_added--;
 	else
-		adapter->vlan_tag[vid] = 1;
+		set_bit(vid, adapter->vids);
 ret:
 	return status;
 }
@@ -1197,7 +1191,7 @@ ret:
 static void be_clear_promisc(struct be_adapter *adapter)
 {
 	adapter->promiscuous = false;
-	adapter->flags &= ~BE_FLAGS_VLAN_PROMISC;
+	adapter->flags &= ~(BE_FLAGS_VLAN_PROMISC | BE_FLAGS_MCAST_PROMISC);
 
 	be_cmd_rx_filter(adapter, IFF_PROMISC, OFF);
 }
@@ -1222,10 +1216,8 @@ static void be_set_rx_mode(struct net_device *netdev)
 
 	/* Enable multicast promisc if num configured exceeds what we support */
 	if (netdev->flags & IFF_ALLMULTI ||
-	    netdev_mc_count(netdev) > be_max_mc(adapter)) {
-		be_cmd_rx_filter(adapter, IFF_ALLMULTI, ON);
-		goto done;
-	}
+	    netdev_mc_count(netdev) > be_max_mc(adapter))
+		goto set_mcast_promisc;
 
 	if (netdev_uc_count(netdev) != adapter->uc_macs) {
 		struct netdev_hw_addr *ha;
@@ -1251,13 +1243,22 @@ static void be_set_rx_mode(struct net_device *netdev)
 	}
 
 	status = be_cmd_rx_filter(adapter, IFF_MULTICAST, ON);
-
-	/* Set to MCAST promisc mode if setting MULTICAST address fails */
-	if (status) {
-		dev_info(&adapter->pdev->dev, "Exhausted multicast HW filters.\n");
-		dev_info(&adapter->pdev->dev, "Disabling HW multicast filtering.\n");
-		be_cmd_rx_filter(adapter, IFF_ALLMULTI, ON);
+	if (!status) {
+		if (adapter->flags & BE_FLAGS_MCAST_PROMISC)
+			adapter->flags &= ~BE_FLAGS_MCAST_PROMISC;
+		goto done;
 	}
+
+set_mcast_promisc:
+	if (adapter->flags & BE_FLAGS_MCAST_PROMISC)
+		return;
+
+	/* Set to MCAST promisc mode if setting MULTICAST address fails
+	 * or if num configured exceeds what we support
+	 */
+	status = be_cmd_rx_filter(adapter, IFF_ALLMULTI, ON);
+	if (!status)
+		adapter->flags |= BE_FLAGS_MCAST_PROMISC;
 done:
 	return;
 }
@@ -1287,7 +1288,7 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 
 	if (status)
 		dev_err(&adapter->pdev->dev, "MAC %pM set on VF %d Failed\n",
-				mac, vf);
+			mac, vf);
 	else
 		memcpy(vf_cfg->mac_addr, mac, ETH_ALEN);
 
@@ -1295,7 +1296,7 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 }
 
 static int be_get_vf_config(struct net_device *netdev, int vf,
-			struct ifla_vf_info *vi)
+			    struct ifla_vf_info *vi)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_vf_cfg *vf_cfg = &adapter->vf_cfg[vf];
@@ -1307,7 +1308,8 @@ static int be_get_vf_config(struct net_device *netdev, int vf,
 		return -EINVAL;
 
 	vi->vf = vf;
-	vi->tx_rate = vf_cfg->tx_rate;
+	vi->max_tx_rate = vf_cfg->tx_rate;
+	vi->min_tx_rate = 0;
 	vi->vlan = vf_cfg->vlan_tag & VLAN_VID_MASK;
 	vi->qos = vf_cfg->vlan_tag >> VLAN_PRIO_SHIFT;
 	memcpy(&vi->mac, vf_cfg->mac_addr, ETH_ALEN);
@@ -1316,8 +1318,7 @@ static int be_get_vf_config(struct net_device *netdev, int vf,
 	return 0;
 }
 
-static int be_set_vf_vlan(struct net_device *netdev,
-			int vf, u16 vlan, u8 qos)
+static int be_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct be_vf_cfg *vf_cfg = &adapter->vf_cfg[vf];
@@ -1348,11 +1349,14 @@ static int be_set_vf_vlan(struct net_device *netdev,
 	return status;
 }
 
-static int be_set_vf_tx_rate(struct net_device *netdev,
-			int vf, int rate)
+static int be_set_vf_tx_rate(struct net_device *netdev, int vf,
+			     int min_tx_rate, int max_tx_rate)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
-	int status = 0;
+	struct device *dev = &adapter->pdev->dev;
+	int percent_rate, status = 0;
+	u16 link_speed = 0;
+	u8 link_status;
 
 	if (!sriov_enabled(adapter))
 		return -EPERM;
@@ -1360,18 +1364,50 @@ static int be_set_vf_tx_rate(struct net_device *netdev,
 	if (vf >= adapter->num_vfs)
 		return -EINVAL;
 
-	if (rate < 100 || rate > 10000) {
-		dev_err(&adapter->pdev->dev,
-			"tx rate must be between 100 and 10000 Mbps\n");
+	if (min_tx_rate)
 		return -EINVAL;
+
+	if (!max_tx_rate)
+		goto config_qos;
+
+	status = be_cmd_link_status_query(adapter, &link_speed,
+					  &link_status, 0);
+	if (status)
+		goto err;
+
+	if (!link_status) {
+		dev_err(dev, "TX-rate setting not allowed when link is down\n");
+		status = -EPERM;
+		goto err;
+	}
+
+	if (max_tx_rate < 100 || max_tx_rate > link_speed) {
+		dev_err(dev, "TX-rate must be between 100 and %d Mbps\n",
+			link_speed);
+		status = -EINVAL;
+		goto err;
+	}
+
+	/* On Skyhawk the QOS setting must be done only as a % value */
+	percent_rate = link_speed / 100;
+	if (skyhawk_chip(adapter) && (max_tx_rate % percent_rate)) {
+		dev_err(dev, "TX-rate must be a multiple of %d Mbps\n",
+			percent_rate);
+		status = -EINVAL;
+		goto err;
 	}
 
-	status = be_cmd_config_qos(adapter, rate / 10, vf + 1);
+config_qos:
+	status = be_cmd_config_qos(adapter, max_tx_rate, link_speed, vf + 1);
 	if (status)
-		dev_err(&adapter->pdev->dev,
-				"tx rate %d on VF %d failed\n", rate, vf);
-	else
-		adapter->vf_cfg[vf].tx_rate = rate;
+		goto err;
+
+	adapter->vf_cfg[vf].tx_rate = max_tx_rate;
+	return 0;
+
+err:
+	dev_err(dev, "TX-rate setting of %dMbps on VF%d failed\n",
+		max_tx_rate, vf);
 	return status;
 }
 static int be_set_vf_link_state(struct net_device *netdev, int vf,
@@ -1469,7 +1505,7 @@ modify_eqd:
 }
 
 static void be_rx_stats_update(struct be_rx_obj *rxo,
-		struct be_rx_compl_info *rxcp)
+			       struct be_rx_compl_info *rxcp)
 {
 	struct be_rx_stats *stats = rx_stats(rxo);
 
@@ -1566,7 +1602,8 @@ static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb,
 		skb_frag_set_page(skb, 0, page_info->page);
 		skb_shinfo(skb)->frags[0].page_offset =
 					page_info->page_offset + hdr_len;
-		skb_frag_size_set(&skb_shinfo(skb)->frags[0], curr_frag_len - hdr_len);
+		skb_frag_size_set(&skb_shinfo(skb)->frags[0],
+				  curr_frag_len - hdr_len);
 		skb->data_len = curr_frag_len - hdr_len;
 		skb->truesize += rx_frag_size;
 		skb->tail += hdr_len;
@@ -1725,8 +1762,8 @@ static void be_parse_rx_compl_v1(struct be_eth_rx_compl *compl,
 	if (rxcp->vlanf) {
 		rxcp->qnq = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, qnq,
 					  compl);
-		rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, vlan_tag,
-					       compl);
+		rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v1,
+					       vlan_tag, compl);
 	}
 	rxcp->port = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, port, compl);
 	rxcp->tunneled =
@@ -1757,8 +1794,8 @@ static void be_parse_rx_compl_v0(struct be_eth_rx_compl *compl,
 	if (rxcp->vlanf) {
 		rxcp->qnq = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, qnq,
 					  compl);
-		rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, vlan_tag,
-					       compl);
+		rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v0,
+					       vlan_tag, compl);
 	}
 	rxcp->port = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, port, compl);
 	rxcp->ip_frag = AMAP_GET_BITS(struct amap_eth_rx_compl_v0,
@@ -1799,7 +1836,7 @@ static struct be_rx_compl_info *be_rx_compl_get(struct be_rx_obj *rxo)
 			rxcp->vlan_tag = swab16(rxcp->vlan_tag);
 
 		if (adapter->pvid == (rxcp->vlan_tag & VLAN_VID_MASK) &&
-		    !adapter->vlan_tag[rxcp->vlan_tag])
+		    !test_bit(rxcp->vlan_tag, adapter->vids))
 			rxcp->vlanf = 0;
 	}
 
@@ -1915,7 +1952,7 @@ static struct be_eth_tx_compl *be_tx_compl_get(struct be_queue_info *tx_cq)
 }
 
 static u16 be_tx_compl_process(struct be_adapter *adapter,
-		struct be_tx_obj *txo, u16 last_index)
+			       struct be_tx_obj *txo, u16 last_index)
 {
 	struct be_queue_info *txq = &txo->q;
 	struct be_eth_wrb *wrb;
@@ -2122,7 +2159,7 @@ static int be_evt_queues_create(struct be_adapter *adapter)
 
 		eq = &eqo->q;
 		rc = be_queue_alloc(adapter, eq, EVNT_Q_LEN,
-					sizeof(struct be_eq_entry));
+				    sizeof(struct be_eq_entry));
 		if (rc)
 			return rc;
 
@@ -2155,7 +2192,7 @@ static int be_mcc_queues_create(struct be_adapter *adapter)
 
 	cq = &adapter->mcc_obj.cq;
 	if (be_queue_alloc(adapter, cq, MCC_CQ_LEN,
-			sizeof(struct be_mcc_compl)))
+			   sizeof(struct be_mcc_compl)))
 		goto err;
 
 	/* Use the default EQ for MCC completions */
@@ -2275,7 +2312,7 @@ static int be_rx_cqs_create(struct be_adapter *adapter)
 		rxo->adapter = adapter;
 		cq = &rxo->cq;
 		rc = be_queue_alloc(adapter, cq, RX_CQ_LEN,
-				sizeof(struct be_eth_rx_compl));
+				    sizeof(struct be_eth_rx_compl));
 		if (rc)
 			return rc;
 
@@ -2339,7 +2376,7 @@ static inline bool do_gro(struct be_rx_compl_info *rxcp)
 }
 
 static int be_process_rx(struct be_rx_obj *rxo, struct napi_struct *napi,
-			int budget, int polling)
+			 int budget, int polling)
 {
 	struct be_adapter *adapter = rxo->adapter;
 	struct be_queue_info *rx_cq = &rxo->cq;
@@ -2365,7 +2402,7 @@ static int be_process_rx(struct be_rx_obj *rxo, struct napi_struct *napi,
 		 * promiscuous mode on some skews
 		 */
 		if (unlikely(rxcp->port != adapter->port_num &&
-				!lancer_chip(adapter))) {
+			     !lancer_chip(adapter))) {
 			be_rx_compl_discard(rxo, rxcp);
 			goto loop_continue;
 		}
@@ -2405,8 +2442,9 @@ static bool be_process_tx(struct be_adapter *adapter, struct be_tx_obj *txo,
 		if (!txcp)
 			break;
 		num_wrbs += be_tx_compl_process(adapter, txo,
-				AMAP_GET_BITS(struct amap_eth_tx_compl,
-					wrb_index, txcp));
+						AMAP_GET_BITS(struct
+							      amap_eth_tx_compl,
+							      wrb_index, txcp));
 	}
 
 	if (work_done) {
@@ -2416,7 +2454,7 @@ static bool be_process_tx(struct be_adapter *adapter, struct be_tx_obj *txo,
 		/* As Tx wrbs have been freed up, wake up netdev queue
 		 * if it was stopped due to lack of tx wrbs.  */
 		if (__netif_subqueue_stopped(adapter->netdev, idx) &&
-			atomic_read(&txo->q.used) < txo->q.len / 2) {
+		    atomic_read(&txo->q.used) < txo->q.len / 2) {
 			netif_wake_subqueue(adapter->netdev, idx);
 		}
 
@@ -2510,9 +2548,9 @@ void be_detect_error(struct be_adapter *adapter)
 		sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET);
 		if (sliport_status & SLIPORT_STATUS_ERR_MASK) {
 			sliport_err1 = ioread32(adapter->db +
-					SLIPORT_ERROR1_OFFSET);
+						SLIPORT_ERROR1_OFFSET);
 			sliport_err2 = ioread32(adapter->db +
-					SLIPORT_ERROR2_OFFSET);
+						SLIPORT_ERROR2_OFFSET);
 			adapter->hw_error = true;
 			/* Do not log error messages if its a FW reset */
 			if (sliport_err1 == SLIPORT_ERROR_FW_RESET1 &&
@@ -2531,13 +2569,13 @@ void be_detect_error(struct be_adapter *adapter)
 		}
 	} else {
 		pci_read_config_dword(adapter->pdev,
-				PCICFG_UE_STATUS_LOW, &ue_lo);
+				      PCICFG_UE_STATUS_LOW, &ue_lo);
 		pci_read_config_dword(adapter->pdev,
-				PCICFG_UE_STATUS_HIGH, &ue_hi);
+				      PCICFG_UE_STATUS_HIGH, &ue_hi);
 		pci_read_config_dword(adapter->pdev,
-				PCICFG_UE_STATUS_LOW_MASK, &ue_lo_mask);
+				      PCICFG_UE_STATUS_LOW_MASK, &ue_lo_mask);
 		pci_read_config_dword(adapter->pdev,
-				PCICFG_UE_STATUS_HI_MASK, &ue_hi_mask);
+				      PCICFG_UE_STATUS_HI_MASK, &ue_hi_mask);
 
 		ue_lo = (ue_lo & ~ue_lo_mask);
 		ue_hi = (ue_hi & ~ue_hi_mask);
@@ -2624,7 +2662,7 @@ fail:
 }
 
 static inline int be_msix_vec_get(struct be_adapter *adapter,
-				struct be_eq_obj *eqo)
+				  struct be_eq_obj *eqo)
 {
 	return adapter->msix_entries[eqo->msix_idx].vector;
 }
@@ -2648,7 +2686,7 @@ err_msix:
 	for (i--, eqo = &adapter->eq_obj[i]; i >= 0; i--, eqo--)
 		free_irq(be_msix_vec_get(adapter, eqo), eqo);
 	dev_warn(&adapter->pdev->dev, "MSIX Request IRQ failed - err %d\n",
-		status);
+		 status);
 	be_msix_disable(adapter);
 	return status;
 }
@@ -2774,7 +2812,8 @@ static int be_rx_qs_create(struct be_adapter *adapter)
 {
 	struct be_rx_obj *rxo;
 	int rc, i, j;
-	u8 rsstable[128];
+	u8 rss_hkey[RSS_HASH_KEY_LEN];
+	struct rss_info *rss = &adapter->rss_info;
 
 	for_all_rx_queues(adapter, rxo, i) {
 		rc = be_queue_alloc(adapter, &rxo->q, RX_Q_LEN,
@@ -2799,31 +2838,36 @@ static int be_rx_qs_create(struct be_adapter *adapter)
 	}
 
 	if (be_multi_rxq(adapter)) {
-		for (j = 0; j < 128; j += adapter->num_rx_qs - 1) {
+		for (j = 0; j < RSS_INDIR_TABLE_LEN;
+			j += adapter->num_rx_qs - 1) {
 			for_all_rss_queues(adapter, rxo, i) {
-				if ((j + i) >= 128)
+				if ((j + i) >= RSS_INDIR_TABLE_LEN)
 					break;
-				rsstable[j + i] = rxo->rss_id;
+				rss->rsstable[j + i] = rxo->rss_id;
+				rss->rss_queue[j + i] = i;
 			}
 		}
-		adapter->rss_flags = RSS_ENABLE_TCP_IPV4 | RSS_ENABLE_IPV4 |
-					RSS_ENABLE_TCP_IPV6 | RSS_ENABLE_IPV6;
+		rss->rss_flags = RSS_ENABLE_TCP_IPV4 | RSS_ENABLE_IPV4 |
+			RSS_ENABLE_TCP_IPV6 | RSS_ENABLE_IPV6;
 
 		if (!BEx_chip(adapter))
-			adapter->rss_flags |= RSS_ENABLE_UDP_IPV4 |
-						RSS_ENABLE_UDP_IPV6;
+			rss->rss_flags |= RSS_ENABLE_UDP_IPV4 |
+				RSS_ENABLE_UDP_IPV6;
 	} else {
 		/* Disable RSS, if only default RX Q is created */
-		adapter->rss_flags = RSS_ENABLE_NONE;
+		rss->rss_flags = RSS_ENABLE_NONE;
 	}
 
-	rc = be_cmd_rss_config(adapter, rsstable, adapter->rss_flags,
-			       128);
+	get_random_bytes(rss_hkey, RSS_HASH_KEY_LEN);
+	rc = be_cmd_rss_config(adapter, rss->rsstable, rss->rss_flags,
+			       128, rss_hkey);
 	if (rc) {
-		adapter->rss_flags = RSS_ENABLE_NONE;
+		rss->rss_flags = RSS_ENABLE_NONE;
 		return rc;
 	}
 
+	memcpy(rss->rss_hkey, rss_hkey, RSS_HASH_KEY_LEN);
+
 	/* First time posting */
 	for_all_rx_queues(adapter, rxo, i)
 		be_post_rx_frags(rxo, GFP_KERNEL);
@@ -2896,7 +2940,8 @@ static int be_setup_wol(struct be_adapter *adapter, bool enable)
 
 	if (enable) {
 		status = pci_write_config_dword(adapter->pdev,
-			PCICFG_PM_CONTROL_OFFSET, PCICFG_PM_CONTROL_MASK);
+						PCICFG_PM_CONTROL_OFFSET,
+						PCICFG_PM_CONTROL_MASK);
 		if (status) {
 			dev_err(&adapter->pdev->dev,
 				"Could not enable Wake-on-lan\n");
@@ -2905,7 +2950,8 @@ static int be_setup_wol(struct be_adapter *adapter, bool enable)
 			return status;
 		}
 		status = be_cmd_enable_magic_wol(adapter,
-				adapter->netdev->dev_addr, &cmd);
+						 adapter->netdev->dev_addr,
+						 &cmd);
 		pci_enable_wake(adapter->pdev, PCI_D3hot, 1);
 		pci_enable_wake(adapter->pdev, PCI_D3cold, 1);
 	} else {
@@ -2944,7 +2990,8 @@ static int be_vf_eth_addr_config(struct be_adapter *adapter)
 
 		if (status)
 			dev_err(&adapter->pdev->dev,
-			"Mac address assignment failed for VF %d\n", vf);
+				"Mac address assignment failed for VF %d\n",
+				vf);
 		else
 			memcpy(vf_cfg->mac_addr, mac, ETH_ALEN);
 
@@ -3086,9 +3133,11 @@ static int be_vfs_if_create(struct be_adapter *adapter)
 
 		/* If a FW profile exists, then cap_flags are updated */
 		en_flags = cap_flags & (BE_IF_FLAGS_UNTAGGED |
-			   BE_IF_FLAGS_BROADCAST | BE_IF_FLAGS_MULTICAST);
-		status = be_cmd_if_create(adapter, cap_flags, en_flags,
-					  &vf_cfg->if_handle, vf + 1);
+					BE_IF_FLAGS_BROADCAST |
+					BE_IF_FLAGS_MULTICAST);
+		status =
+		    be_cmd_if_create(adapter, cap_flags, en_flags,
+				     &vf_cfg->if_handle, vf + 1);
 		if (status)
 			goto err;
 	}
@@ -3119,7 +3168,6 @@ static int be_vf_setup(struct be_adapter *adapter)
 	struct be_vf_cfg *vf_cfg;
 	int status, old_vfs, vf;
 	u32 privileges;
-	u16 lnk_speed;
 
 	old_vfs = pci_num_vf(adapter->pdev);
 	if (old_vfs) {
@@ -3175,16 +3223,9 @@ static int be_vf_setup(struct be_adapter *adapter)
 					 vf);
 		}
 
-		/* BE3 FW, by default, caps VF TX-rate to 100mbps.
-		 * Allow full available bandwidth
-		 */
-		if (BE3_chip(adapter) && !old_vfs)
-			be_cmd_config_qos(adapter, 1000, vf + 1);
-
-		status = be_cmd_link_status_query(adapter, &lnk_speed,
-						  NULL, vf + 1);
-		if (!status)
-			vf_cfg->tx_rate = lnk_speed;
+		/* Allow full available bandwidth */
+		if (!old_vfs)
+			be_cmd_config_qos(adapter, 0, 0, vf + 1);
 
 		if (!old_vfs) {
 			be_cmd_enable_vf(adapter, vf + 1);
@@ -3213,9 +3254,9 @@ err:
 
 static u8 be_convert_mc_type(u32 function_mode)
 {
-	if (function_mode & VNIC_MODE && function_mode & FLEX10_MODE)
+	if (function_mode & VNIC_MODE && function_mode & QNQ_MODE)
 		return vNIC1;
-	else if (function_mode & FLEX10_MODE)
+	else if (function_mode & QNQ_MODE)
 		return FLEX10;
 	else if (function_mode & VNIC_MODE)
 		return vNIC2;
@@ -3590,35 +3631,7 @@ static void be_netpoll(struct net_device *netdev)
 }
 #endif
 
-#define FW_FILE_HDR_SIGN 	"ServerEngines Corp. "
-static char flash_cookie[2][16] =      {"*** SE FLAS", "H DIRECTORY *** "};
-
-static bool be_flash_redboot(struct be_adapter *adapter,
-			const u8 *p, u32 img_start, int image_size,
-			int hdr_size)
-{
-	u32 crc_offset;
-	u8 flashed_crc[4];
-	int status;
-
-	crc_offset = hdr_size + img_start + image_size - 4;
-
-	p += crc_offset;
-
-	status = be_cmd_get_flash_crc(adapter, flashed_crc,
-			(image_size - 4));
-	if (status) {
-		dev_err(&adapter->pdev->dev,
-		"could not get crc from flash, not flashing redboot\n");
-		return false;
-	}
-
-	/*update redboot only if crc does not match*/
-	if (!memcmp(flashed_crc, p, 4))
-		return false;
-	else
-		return true;
-}
+static char flash_cookie[2][16] = {"*** SE FLAS", "H DIRECTORY *** "};
 
 static bool phy_flashing_required(struct be_adapter *adapter)
 {
@@ -3649,8 +3662,8 @@ static bool is_comp_in_ufi(struct be_adapter *adapter,
 }
 
 static struct flash_section_info *get_fsec_info(struct be_adapter *adapter,
-					 int header_size,
-					 const struct firmware *fw)
+						int header_size,
+						const struct firmware *fw)
 {
 	struct flash_section_info *fsec = NULL;
 	const u8 *p = fw->data;
@@ -3665,12 +3678,35 @@ static struct flash_section_info *get_fsec_info(struct be_adapter *adapter,
 	return NULL;
 }
 
+static int be_check_flash_crc(struct be_adapter *adapter, const u8 *p,
+			      u32 img_offset, u32 img_size, int hdr_size,
+			      u16 img_optype, bool *crc_match)
+{
+	u32 crc_offset;
+	int status;
+	u8 crc[4];
+
+	status = be_cmd_get_flash_crc(adapter, crc, img_optype, img_size - 4);
+	if (status)
+		return status;
+
+	crc_offset = hdr_size + img_offset + img_size - 4;
+
+	/* Skip flashing, if crc of flashed region matches */
+	if (!memcmp(crc, p + crc_offset, 4))
+		*crc_match = true;
+	else
+		*crc_match = false;
+
+	return status;
+}
+
 static int be_flash(struct be_adapter *adapter, const u8 *img,
-		struct be_dma_mem *flash_cmd, int optype, int img_size)
+		    struct be_dma_mem *flash_cmd, int optype, int img_size)
 {
-	u32 total_bytes = 0, flash_op, num_bytes = 0;
-	int status = 0;
 	struct be_cmd_write_flashrom *req = flash_cmd->va;
+	u32 total_bytes, flash_op, num_bytes;
+	int status;
 
 	total_bytes = img_size;
 	while (total_bytes) {
@@ -3693,32 +3729,28 @@ static int be_flash(struct be_adapter *adapter, const u8 *img,
 		memcpy(req->data_buf, img, num_bytes);
 		img += num_bytes;
 		status = be_cmd_write_flashrom(adapter, flash_cmd, optype,
-						flash_op, num_bytes);
-		if (status) {
-			if (status == ILLEGAL_IOCTL_REQ &&
-			    optype == OPTYPE_PHY_FW)
-				break;
-			dev_err(&adapter->pdev->dev,
-				"cmd to write to flash rom failed.\n");
+					       flash_op, num_bytes);
+		if (base_status(status) == MCC_STATUS_ILLEGAL_REQUEST &&
+		    optype == OPTYPE_PHY_FW)
+			break;
+		else if (status)
 			return status;
-		}
 	}
 	return 0;
 }
 
 /* For BE2, BE3 and BE3-R */
 static int be_flash_BEx(struct be_adapter *adapter,
-			 const struct firmware *fw,
-			 struct be_dma_mem *flash_cmd,
-			 int num_of_images)
-
+			const struct firmware *fw,
+			struct be_dma_mem *flash_cmd, int num_of_images)
 {
-	int status = 0, i, filehdr_size = 0;
 	int img_hdrs_size = (num_of_images * sizeof(struct image_hdr));
-	const u8 *p = fw->data;
-	const struct flash_comp *pflashcomp;
-	int num_comp, redboot;
+	struct device *dev = &adapter->pdev->dev;
 	struct flash_section_info *fsec = NULL;
+	int status, i, filehdr_size, num_comp;
+	const struct flash_comp *pflashcomp;
+	bool crc_match;
+	const u8 *p;
 
 	struct flash_comp gen3_flash_types[] = {
 		{ FLASH_iSCSI_PRIMARY_IMAGE_START_g3, OPTYPE_ISCSI_ACTIVE,
@@ -3775,8 +3807,7 @@ static int be_flash_BEx(struct be_adapter *adapter,
 	/* Get flash section info*/
 	fsec = get_fsec_info(adapter, filehdr_size + img_hdrs_size, fw);
 	if (!fsec) {
-		dev_err(&adapter->pdev->dev,
-			"Invalid Cookie. UFI corrupted ?\n");
+		dev_err(dev, "Invalid Cookie. FW image may be corrupted\n");
 		return -1;
 	}
 	for (i = 0; i < num_comp; i++) {
@@ -3792,23 +3823,32 @@ static int be_flash_BEx(struct be_adapter *adapter,
 				continue;
 
 		if (pflashcomp[i].optype == OPTYPE_REDBOOT) {
-			redboot = be_flash_redboot(adapter, fw->data,
-				pflashcomp[i].offset, pflashcomp[i].size,
-				filehdr_size + img_hdrs_size);
-			if (!redboot)
+			status = be_check_flash_crc(adapter, fw->data,
+						    pflashcomp[i].offset,
+						    pflashcomp[i].size,
+						    filehdr_size +
+						    img_hdrs_size,
+						    OPTYPE_REDBOOT, &crc_match);
+			if (status) {
+				dev_err(dev,
+					"Could not get CRC for 0x%x region\n",
+					pflashcomp[i].optype);
+				continue;
+			}
+
+			if (crc_match)
 				continue;
 		}
 
-		p = fw->data;
-		p += filehdr_size + pflashcomp[i].offset + img_hdrs_size;
+		p = fw->data + filehdr_size + pflashcomp[i].offset +
+			img_hdrs_size;
 		if (p + pflashcomp[i].size > fw->data + fw->size)
 			return -1;
 
 		status = be_flash(adapter, p, flash_cmd, pflashcomp[i].optype,
-					pflashcomp[i].size);
+				  pflashcomp[i].size);
 		if (status) {
-			dev_err(&adapter->pdev->dev,
-				"Flashing section type %d failed.\n",
+			dev_err(dev, "Flashing section type 0x%x failed\n",
 				pflashcomp[i].img_type);
 			return status;
 		}
@@ -3816,80 +3856,142 @@ static int be_flash_BEx(struct be_adapter *adapter,
 	return 0;
 }
 
+static u16 be_get_img_optype(struct flash_section_entry fsec_entry)
+{
+	u32 img_type = le32_to_cpu(fsec_entry.type);
+	u16 img_optype = le16_to_cpu(fsec_entry.optype);
+
+	if (img_optype != 0xFFFF)
+		return img_optype;
+
+	switch (img_type) {
+	case IMAGE_FIRMWARE_iSCSI:
+		img_optype = OPTYPE_ISCSI_ACTIVE;
+		break;
+	case IMAGE_BOOT_CODE:
+		img_optype = OPTYPE_REDBOOT;
+		break;
+	case IMAGE_OPTION_ROM_ISCSI:
+		img_optype = OPTYPE_BIOS;
+		break;
+	case IMAGE_OPTION_ROM_PXE:
+		img_optype = OPTYPE_PXE_BIOS;
+		break;
+	case IMAGE_OPTION_ROM_FCoE:
+		img_optype = OPTYPE_FCOE_BIOS;
+		break;
+	case IMAGE_FIRMWARE_BACKUP_iSCSI:
+		img_optype = OPTYPE_ISCSI_BACKUP;
+		break;
+	case IMAGE_NCSI:
+		img_optype = OPTYPE_NCSI_FW;
+		break;
+	case IMAGE_FLASHISM_JUMPVECTOR:
+		img_optype = OPTYPE_FLASHISM_JUMPVECTOR;
+		break;
+	case IMAGE_FIRMWARE_PHY:
+		img_optype = OPTYPE_SH_PHY_FW;
+		break;
+	case IMAGE_REDBOOT_DIR:
+		img_optype = OPTYPE_REDBOOT_DIR;
+		break;
+	case IMAGE_REDBOOT_CONFIG:
+		img_optype = OPTYPE_REDBOOT_CONFIG;
+		break;
+	case IMAGE_UFI_DIR:
+		img_optype = OPTYPE_UFI_DIR;
+		break;
+	default:
+		break;
+	}
+
+	return img_optype;
+}
+
 static int be_flash_skyhawk(struct be_adapter *adapter,
-		const struct firmware *fw,
-		struct be_dma_mem *flash_cmd, int num_of_images)
+			    const struct firmware *fw,
+			    struct be_dma_mem *flash_cmd, int num_of_images)
 {
-	int status = 0, i, filehdr_size = 0;
-	int img_offset, img_size, img_optype, redboot;
 	int img_hdrs_size = num_of_images * sizeof(struct image_hdr);
-	const u8 *p = fw->data;
+	struct device *dev = &adapter->pdev->dev;
 	struct flash_section_info *fsec = NULL;
+	u32 img_offset, img_size, img_type;
+	int status, i, filehdr_size;
+	bool crc_match, old_fw_img;
+	u16 img_optype;
+	const u8 *p;
 
 	filehdr_size = sizeof(struct flash_file_hdr_g3);
 	fsec = get_fsec_info(adapter, filehdr_size + img_hdrs_size, fw);
 	if (!fsec) {
-		dev_err(&adapter->pdev->dev,
-			"Invalid Cookie. UFI corrupted ?\n");
+		dev_err(dev, "Invalid Cookie. FW image may be corrupted\n");
 		return -1;
 	}
 
 	for (i = 0; i < le32_to_cpu(fsec->fsec_hdr.num_images); i++) {
 		img_offset = le32_to_cpu(fsec->fsec_entry[i].offset);
 		img_size   = le32_to_cpu(fsec->fsec_entry[i].pad_size);
+		img_type   = le32_to_cpu(fsec->fsec_entry[i].type);
+		img_optype = be_get_img_optype(fsec->fsec_entry[i]);
+		old_fw_img = fsec->fsec_entry[i].optype == 0xFFFF;
 
-		switch (le32_to_cpu(fsec->fsec_entry[i].type)) {
-		case IMAGE_FIRMWARE_iSCSI:
-			img_optype = OPTYPE_ISCSI_ACTIVE;
-			break;
-		case IMAGE_BOOT_CODE:
-			img_optype = OPTYPE_REDBOOT;
-			break;
-		case IMAGE_OPTION_ROM_ISCSI:
-			img_optype = OPTYPE_BIOS;
-			break;
-		case IMAGE_OPTION_ROM_PXE:
-			img_optype = OPTYPE_PXE_BIOS;
-			break;
-		case IMAGE_OPTION_ROM_FCoE:
-			img_optype = OPTYPE_FCOE_BIOS;
-			break;
-		case IMAGE_FIRMWARE_BACKUP_iSCSI:
-			img_optype = OPTYPE_ISCSI_BACKUP;
-			break;
-		case IMAGE_NCSI:
-			img_optype = OPTYPE_NCSI_FW;
-			break;
-		default:
+		if (img_optype == 0xFFFF)
 			continue;
+		/* Don't bother verifying CRC if an old FW image is being
+		 * flashed
+		 */
+		if (old_fw_img)
+			goto flash;
+
+		status = be_check_flash_crc(adapter, fw->data, img_offset,
+					    img_size, filehdr_size +
+					    img_hdrs_size, img_optype,
+					    &crc_match);
+		/* The current FW image on the card does not recognize the new
+		 * FLASH op_type. The FW download is partially complete.
+		 * Reboot the server now to enable FW image to recognize the
+		 * new FLASH op_type. To complete the remaining process,
+		 * download the same FW again after the reboot.
+		 */
+		if (base_status(status) == MCC_STATUS_ILLEGAL_REQUEST ||
+		    base_status(status) == MCC_STATUS_ILLEGAL_FIELD) {
+			dev_err(dev, "Flash incomplete. Reset the server\n");
+			dev_err(dev, "Download FW image again after reset\n");
+			return -EAGAIN;
+		} else if (status) {
+			dev_err(dev, "Could not get CRC for 0x%x region\n",
+				img_optype);
+			return -EFAULT;
 		}
 
-		if (img_optype == OPTYPE_REDBOOT) {
-			redboot = be_flash_redboot(adapter, fw->data,
-					img_offset, img_size,
-					filehdr_size + img_hdrs_size);
-			if (!redboot)
-				continue;
-		}
+		if (crc_match)
+			continue;
 
-		p = fw->data;
-		p += filehdr_size + img_offset + img_hdrs_size;
+flash:
+		p = fw->data + filehdr_size + img_offset + img_hdrs_size;
 		if (p + img_size > fw->data + fw->size)
 			return -1;
 
 		status = be_flash(adapter, p, flash_cmd, img_optype, img_size);
-		if (status) {
-			dev_err(&adapter->pdev->dev,
-				"Flashing section type %d failed.\n",
-				fsec->fsec_entry[i].type);
-			return status;
+		/* For old FW images ignore ILLEGAL_FIELD error or errors on
+		 * UFI_DIR region
+		 */
+		if (old_fw_img &&
+		    (base_status(status) == MCC_STATUS_ILLEGAL_FIELD ||
+		     (img_optype == OPTYPE_UFI_DIR &&
+		      base_status(status) == MCC_STATUS_FAILED))) {
+			continue;
+		} else if (status) {
+			dev_err(dev, "Flashing section type 0x%x failed\n",
+				img_type);
+			return -EFAULT;
 		}
 	}
 	return 0;
 }
 
 static int lancer_fw_download(struct be_adapter *adapter,
-				const struct firmware *fw)
+			      const struct firmware *fw)
 {
 #define LANCER_FW_DOWNLOAD_CHUNK      (32 * 1024)
 #define LANCER_FW_DOWNLOAD_LOCATION   "/prg"
@@ -3955,7 +4057,7 @@ static int lancer_fw_download(struct be_adapter *adapter,
 	}
 
 	dma_free_coherent(&adapter->pdev->dev, flash_cmd.size, flash_cmd.va,
-				flash_cmd.dma);
+			  flash_cmd.dma);
 	if (status) {
 		dev_err(&adapter->pdev->dev,
 			"Firmware load error. "
@@ -3976,9 +4078,8 @@ static int lancer_fw_download(struct be_adapter *adapter,
 			goto lancer_fw_exit;
 		}
 	} else if (change_status != LANCER_NO_RESET_NEEDED) {
-			dev_err(&adapter->pdev->dev,
-				"System reboot required for new FW"
-				" to be active\n");
+		dev_err(&adapter->pdev->dev,
+			"System reboot required for new FW to be active\n");
 	}
 
 	dev_info(&adapter->pdev->dev, "Firmware flashed successfully\n");
@@ -4042,7 +4143,7 @@ static int be_fw_download(struct be_adapter *adapter, const struct firmware* fw)
 			switch (ufi_type) {
 			case UFI_TYPE4:
 				status = be_flash_skyhawk(adapter, fw,
-							&flash_cmd, num_imgs);
+							  &flash_cmd, num_imgs);
 				break;
 			case UFI_TYPE3R:
 				status = be_flash_BEx(adapter, fw, &flash_cmd,
@@ -4112,8 +4213,7 @@ fw_exit:
 	return status;
 }
 
-static int be_ndo_bridge_setlink(struct net_device *dev,
-				    struct nlmsghdr *nlh)
+static int be_ndo_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh)
 {
 	struct be_adapter *adapter = netdev_priv(dev);
 	struct nlattr *attr, *br_spec;
@@ -4155,8 +4255,7 @@ err:
 }
 
 static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
-				    struct net_device *dev,
-				    u32 filter_mask)
+				 struct net_device *dev, u32 filter_mask)
 {
 	struct be_adapter *adapter = netdev_priv(dev);
 	int status = 0;
@@ -4254,7 +4353,7 @@ static const struct net_device_ops be_netdev_ops = {
 	.ndo_vlan_rx_kill_vid	= be_vlan_rem_vid,
 	.ndo_set_vf_mac		= be_set_vf_mac,
 	.ndo_set_vf_vlan	= be_set_vf_vlan,
-	.ndo_set_vf_tx_rate	= be_set_vf_tx_rate,
+	.ndo_set_vf_rate	= be_set_vf_tx_rate,
 	.ndo_get_vf_config	= be_get_vf_config,
 	.ndo_set_vf_link_state  = be_set_vf_link_state,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -4301,7 +4400,7 @@ static void be_netdev_init(struct net_device *netdev)
 
 	netdev->netdev_ops = &be_netdev_ops;
 
-	SET_ETHTOOL_OPS(netdev, &be_ethtool_ops);
+	netdev->ethtool_ops = &be_ethtool_ops;
 }
 
 static void be_unmap_pci_bars(struct be_adapter *adapter)
@@ -4870,7 +4969,7 @@ static void be_shutdown(struct pci_dev *pdev)
 }
 
 static pci_ers_result_t be_eeh_err_detected(struct pci_dev *pdev,
-				pci_channel_state_t state)
+					    pci_channel_state_t state)
 {
 	struct be_adapter *adapter = pci_get_drvdata(pdev);
 	struct net_device *netdev =  adapter->netdev;
diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c
index 8b70ca7e342b..f3658bdb64cc 100644
--- a/drivers/net/ethernet/ethoc.c
+++ b/drivers/net/ethernet/ethoc.c
@@ -769,11 +769,6 @@ static int ethoc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return phy_mii_ioctl(phy, ifr, cmd);
 }
 
-static int ethoc_config(struct net_device *dev, struct ifmap *map)
-{
-	return -ENOSYS;
-}
-
 static void ethoc_do_set_mac_address(struct net_device *dev)
 {
 	struct ethoc *priv = netdev_priv(dev);
@@ -995,7 +990,6 @@ static const struct net_device_ops ethoc_netdev_ops = {
 	.ndo_open = ethoc_open,
 	.ndo_stop = ethoc_stop,
 	.ndo_do_ioctl = ethoc_ioctl,
-	.ndo_set_config = ethoc_config,
 	.ndo_set_mac_address = ethoc_set_mac_address,
 	.ndo_set_rx_mode = ethoc_set_multicast_list,
 	.ndo_change_mtu = ethoc_change_mtu,
diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c
index 68069eabc4f8..c77fa4a69844 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.c
+++ b/drivers/net/ethernet/faraday/ftgmac100.c
@@ -1210,7 +1210,7 @@ static int ftgmac100_probe(struct platform_device *pdev)
 
 	SET_NETDEV_DEV(netdev, &pdev->dev);
 
-	SET_ETHTOOL_OPS(netdev, &ftgmac100_ethtool_ops);
+	netdev->ethtool_ops = &ftgmac100_ethtool_ops;
 	netdev->netdev_ops = &ftgmac100_netdev_ops;
 	netdev->features = NETIF_F_IP_CSUM | NETIF_F_GRO;
 
diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c
index 8be5b40c0a12..4ff1adc6bfca 100644
--- a/drivers/net/ethernet/faraday/ftmac100.c
+++ b/drivers/net/ethernet/faraday/ftmac100.c
@@ -1085,7 +1085,7 @@ static int ftmac100_probe(struct platform_device *pdev)
 	}
 
 	SET_NETDEV_DEV(netdev, &pdev->dev);
-	SET_ETHTOOL_OPS(netdev, &ftmac100_ethtool_ops);
+	netdev->ethtool_ops = &ftmac100_ethtool_ops;
 	netdev->netdev_ops = &ftmac100_netdev_ops;
 
 	platform_set_drvdata(pdev, netdev);
diff --git a/drivers/net/ethernet/freescale/Kconfig b/drivers/net/ethernet/freescale/Kconfig
index 6048dc8604ee..270308315d43 100644
--- a/drivers/net/ethernet/freescale/Kconfig
+++ b/drivers/net/ethernet/freescale/Kconfig
@@ -67,6 +67,7 @@ config FSL_XGMAC_MDIO
 	tristate "Freescale XGMAC MDIO"
 	depends on FSL_SOC
 	select PHYLIB
+	select OF_MDIO
 	---help---
 	  This driver supports the MDIO bus on the Fman 10G Ethernet MACs.
 
diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index 3b8d6d19ff05..671d080105a7 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -221,7 +221,7 @@ struct bufdesc_ex {
 #define BD_ENET_TX_RCMASK       ((ushort)0x003c)
 #define BD_ENET_TX_UN           ((ushort)0x0002)
 #define BD_ENET_TX_CSL          ((ushort)0x0001)
-#define BD_ENET_TX_STATS        ((ushort)0x03ff)        /* All status bits */
+#define BD_ENET_TX_STATS        ((ushort)0x0fff)        /* All status bits */
 
 /*enhanced buffer descriptor control/status used by Ethernet transmit*/
 #define BD_ENET_TX_INT          0x40000000
@@ -246,8 +246,8 @@ struct bufdesc_ex {
 #define RX_RING_SIZE		(FEC_ENET_RX_FRPPG * FEC_ENET_RX_PAGES)
 #define FEC_ENET_TX_FRSIZE	2048
 #define FEC_ENET_TX_FRPPG	(PAGE_SIZE / FEC_ENET_TX_FRSIZE)
-#define TX_RING_SIZE		16	/* Must be power of two */
-#define TX_RING_MOD_MASK	15	/*   for this to work */
+#define TX_RING_SIZE		512	/* Must be power of two */
+#define TX_RING_MOD_MASK	511	/*   for this to work */
 
 #define BD_ENET_RX_INT          0x00800000
 #define BD_ENET_RX_PTP          ((ushort)0x0400)
@@ -296,8 +296,15 @@ struct fec_enet_private {
 	/* The ring entries to be free()ed */
 	struct bufdesc	*dirty_tx;
 
+	unsigned short bufdesc_size;
 	unsigned short tx_ring_size;
 	unsigned short rx_ring_size;
+	unsigned short tx_stop_threshold;
+	unsigned short tx_wake_threshold;
+
+	/* Software TSO */
+	char *tso_hdrs;
+	dma_addr_t tso_hdrs_dma;
 
 	struct	platform_device *pdev;
 
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 8d69e439f0c5..77037fd377b8 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -36,6 +36,7 @@
 #include <linux/in.h>
 #include <linux/ip.h>
 #include <net/ip.h>
+#include <net/tso.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
 #include <linux/icmp.h>
@@ -54,6 +55,7 @@
 #include <linux/of_net.h>
 #include <linux/regulator/consumer.h>
 #include <linux/if_vlan.h>
+#include <linux/pinctrl/consumer.h>
 
 #include <asm/cacheflush.h>
 
@@ -172,10 +174,6 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address");
 #endif
 #endif /* CONFIG_M5272 */
 
-#if (((RX_RING_SIZE + TX_RING_SIZE) * 32) > PAGE_SIZE)
-#error "FEC: descriptor ring size constants too large"
-#endif
-
 /* Interrupt events/masks. */
 #define FEC_ENET_HBERR	((uint)0x80000000)	/* Heartbeat error */
 #define FEC_ENET_BABR	((uint)0x40000000)	/* Babbling receiver */
@@ -231,6 +229,15 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address");
 #define FEC_PAUSE_FLAG_AUTONEG	0x1
 #define FEC_PAUSE_FLAG_ENABLE	0x2
 
+#define TSO_HEADER_SIZE		128
+/* Max number of allowed TCP segments for software TSO */
+#define FEC_MAX_TSO_SEGS	100
+#define FEC_MAX_SKB_DESCS	(FEC_MAX_TSO_SEGS * 2 + MAX_SKB_FRAGS)
+
+#define IS_TSO_HEADER(txq, addr) \
+	((addr >= txq->tso_hdrs_dma) && \
+	(addr < txq->tso_hdrs_dma + txq->tx_ring_size * TSO_HEADER_SIZE))
+
 static int mii_cnt;
 
 static inline
@@ -286,6 +293,22 @@ struct bufdesc *fec_enet_get_prevdesc(struct bufdesc *bdp, struct fec_enet_priva
 		return (new_bd < base) ? (new_bd + ring_size) : new_bd;
 }
 
+static int fec_enet_get_bd_index(struct bufdesc *base, struct bufdesc *bdp,
+				struct fec_enet_private *fep)
+{
+	return ((const char *)bdp - (const char *)base) / fep->bufdesc_size;
+}
+
+static int fec_enet_get_free_txdesc_num(struct fec_enet_private *fep)
+{
+	int entries;
+
+	entries = ((const char *)fep->dirty_tx -
+			(const char *)fep->cur_tx) / fep->bufdesc_size - 1;
+
+	return entries > 0 ? entries : entries + fep->tx_ring_size;
+}
+
 static void *swap_buffer(void *bufaddr, int len)
 {
 	int i;
@@ -297,6 +320,11 @@ static void *swap_buffer(void *bufaddr, int len)
 	return bufaddr;
 }
 
+static inline bool is_ipv4_pkt(struct sk_buff *skb)
+{
+	return skb->protocol == htons(ETH_P_IP) && ip_hdr(skb)->version == 4;
+}
+
 static int
 fec_enet_clear_csum(struct sk_buff *skb, struct net_device *ndev)
 {
@@ -307,33 +335,134 @@ fec_enet_clear_csum(struct sk_buff *skb, struct net_device *ndev)
 	if (unlikely(skb_cow_head(skb, 0)))
 		return -1;
 
+	if (is_ipv4_pkt(skb))
+		ip_hdr(skb)->check = 0;
 	*(__sum16 *)(skb->head + skb->csum_start + skb->csum_offset) = 0;
 
 	return 0;
 }
 
-static netdev_tx_t
-fec_enet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+static void
+fec_enet_submit_work(struct bufdesc *bdp, struct fec_enet_private *fep)
+{
+	const struct platform_device_id *id_entry =
+				platform_get_device_id(fep->pdev);
+	struct bufdesc *bdp_pre;
+
+	bdp_pre = fec_enet_get_prevdesc(bdp, fep);
+	if ((id_entry->driver_data & FEC_QUIRK_ERR006358) &&
+	    !(bdp_pre->cbd_sc & BD_ENET_TX_READY)) {
+		fep->delay_work.trig_tx = true;
+		schedule_delayed_work(&(fep->delay_work.delay_work),
+					msecs_to_jiffies(1));
+	}
+}
+
+static int
+fec_enet_txq_submit_frag_skb(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	const struct platform_device_id *id_entry =
 				platform_get_device_id(fep->pdev);
-	struct bufdesc *bdp, *bdp_pre;
-	void *bufaddr;
-	unsigned short	status;
+	struct bufdesc *bdp = fep->cur_tx;
+	struct bufdesc_ex *ebdp;
+	int nr_frags = skb_shinfo(skb)->nr_frags;
+	int frag, frag_len;
+	unsigned short status;
+	unsigned int estatus = 0;
+	skb_frag_t *this_frag;
 	unsigned int index;
+	void *bufaddr;
+	int i;
 
-	/* Fill in a Tx ring entry */
+	for (frag = 0; frag < nr_frags; frag++) {
+		this_frag = &skb_shinfo(skb)->frags[frag];
+		bdp = fec_enet_get_nextdesc(bdp, fep);
+		ebdp = (struct bufdesc_ex *)bdp;
+
+		status = bdp->cbd_sc;
+		status &= ~BD_ENET_TX_STATS;
+		status |= (BD_ENET_TX_TC | BD_ENET_TX_READY);
+		frag_len = skb_shinfo(skb)->frags[frag].size;
+
+		/* Handle the last BD specially */
+		if (frag == nr_frags - 1) {
+			status |= (BD_ENET_TX_INTR | BD_ENET_TX_LAST);
+			if (fep->bufdesc_ex) {
+				estatus |= BD_ENET_TX_INT;
+				if (unlikely(skb_shinfo(skb)->tx_flags &
+					SKBTX_HW_TSTAMP && fep->hwts_tx_en))
+					estatus |= BD_ENET_TX_TS;
+			}
+		}
+
+		if (fep->bufdesc_ex) {
+			if (skb->ip_summed == CHECKSUM_PARTIAL)
+				estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
+			ebdp->cbd_bdu = 0;
+			ebdp->cbd_esc = estatus;
+		}
+
+		bufaddr = page_address(this_frag->page.p) + this_frag->page_offset;
+
+		index = fec_enet_get_bd_index(fep->tx_bd_base, bdp, fep);
+		if (((unsigned long) bufaddr) & FEC_ALIGNMENT ||
+			id_entry->driver_data & FEC_QUIRK_SWAP_FRAME) {
+			memcpy(fep->tx_bounce[index], bufaddr, frag_len);
+			bufaddr = fep->tx_bounce[index];
+
+			if (id_entry->driver_data & FEC_QUIRK_SWAP_FRAME)
+				swap_buffer(bufaddr, frag_len);
+		}
+
+		bdp->cbd_bufaddr = dma_map_single(&fep->pdev->dev, bufaddr,
+						frag_len, DMA_TO_DEVICE);
+		if (dma_mapping_error(&fep->pdev->dev, bdp->cbd_bufaddr)) {
+			dev_kfree_skb_any(skb);
+			if (net_ratelimit())
+				netdev_err(ndev, "Tx DMA memory map failed\n");
+			goto dma_mapping_error;
+		}
+
+		bdp->cbd_datlen = frag_len;
+		bdp->cbd_sc = status;
+	}
+
+	fep->cur_tx = bdp;
+
+	return 0;
+
+dma_mapping_error:
 	bdp = fep->cur_tx;
+	for (i = 0; i < frag; i++) {
+		bdp = fec_enet_get_nextdesc(bdp, fep);
+		dma_unmap_single(&fep->pdev->dev, bdp->cbd_bufaddr,
+				bdp->cbd_datlen, DMA_TO_DEVICE);
+	}
+	return NETDEV_TX_OK;
+}
 
-	status = bdp->cbd_sc;
+static int fec_enet_txq_submit_skb(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	const struct platform_device_id *id_entry =
+				platform_get_device_id(fep->pdev);
+	int nr_frags = skb_shinfo(skb)->nr_frags;
+	struct bufdesc *bdp, *last_bdp;
+	void *bufaddr;
+	unsigned short status;
+	unsigned short buflen;
+	unsigned int estatus = 0;
+	unsigned int index;
+	int entries_free;
+	int ret;
 
-	if (status & BD_ENET_TX_READY) {
-		/* Ooops.  All transmit buffers are full.  Bail out.
-		 * This should not happen, since ndev->tbusy should be set.
-		 */
-		netdev_err(ndev, "tx queue full!\n");
-		return NETDEV_TX_BUSY;
+	entries_free = fec_enet_get_free_txdesc_num(fep);
+	if (entries_free < MAX_SKB_FRAGS + 1) {
+		dev_kfree_skb_any(skb);
+		if (net_ratelimit())
+			netdev_err(ndev, "NOT enough BD for SG!\n");
+		return NETDEV_TX_OK;
 	}
 
 	/* Protocol checksum off-load for TCP and UDP. */
@@ -342,102 +471,300 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 		return NETDEV_TX_OK;
 	}
 
-	/* Clear all of the status flags */
+	/* Fill in a Tx ring entry */
+	bdp = fep->cur_tx;
+	status = bdp->cbd_sc;
 	status &= ~BD_ENET_TX_STATS;
 
 	/* Set buffer length and buffer pointer */
 	bufaddr = skb->data;
-	bdp->cbd_datlen = skb->len;
+	buflen = skb_headlen(skb);
 
-	/*
-	 * On some FEC implementations data must be aligned on
-	 * 4-byte boundaries. Use bounce buffers to copy data
-	 * and get it aligned. Ugh.
-	 */
-	if (fep->bufdesc_ex)
-		index = (struct bufdesc_ex *)bdp -
-			(struct bufdesc_ex *)fep->tx_bd_base;
-	else
-		index = bdp - fep->tx_bd_base;
-
-	if (((unsigned long) bufaddr) & FEC_ALIGNMENT) {
-		memcpy(fep->tx_bounce[index], skb->data, skb->len);
+	index = fec_enet_get_bd_index(fep->tx_bd_base, bdp, fep);
+	if (((unsigned long) bufaddr) & FEC_ALIGNMENT ||
+		id_entry->driver_data & FEC_QUIRK_SWAP_FRAME) {
+		memcpy(fep->tx_bounce[index], skb->data, buflen);
 		bufaddr = fep->tx_bounce[index];
-	}
 
-	/*
-	 * Some design made an incorrect assumption on endian mode of
-	 * the system that it's running on. As the result, driver has to
-	 * swap every frame going to and coming from the controller.
-	 */
-	if (id_entry->driver_data & FEC_QUIRK_SWAP_FRAME)
-		swap_buffer(bufaddr, skb->len);
-
-	/* Save skb pointer */
-	fep->tx_skbuff[index] = skb;
+		if (id_entry->driver_data & FEC_QUIRK_SWAP_FRAME)
+			swap_buffer(bufaddr, buflen);
+	}
 
 	/* Push the data cache so the CPM does not get stale memory
 	 * data.
 	 */
 	bdp->cbd_bufaddr = dma_map_single(&fep->pdev->dev, bufaddr,
-			skb->len, DMA_TO_DEVICE);
+					buflen, DMA_TO_DEVICE);
 	if (dma_mapping_error(&fep->pdev->dev, bdp->cbd_bufaddr)) {
-		bdp->cbd_bufaddr = 0;
-		fep->tx_skbuff[index] = NULL;
 		dev_kfree_skb_any(skb);
 		if (net_ratelimit())
 			netdev_err(ndev, "Tx DMA memory map failed\n");
 		return NETDEV_TX_OK;
 	}
 
+	if (nr_frags) {
+		ret = fec_enet_txq_submit_frag_skb(skb, ndev);
+		if (ret)
+			return ret;
+	} else {
+		status |= (BD_ENET_TX_INTR | BD_ENET_TX_LAST);
+		if (fep->bufdesc_ex) {
+			estatus = BD_ENET_TX_INT;
+			if (unlikely(skb_shinfo(skb)->tx_flags &
+				SKBTX_HW_TSTAMP && fep->hwts_tx_en))
+				estatus |= BD_ENET_TX_TS;
+		}
+	}
+
 	if (fep->bufdesc_ex) {
 
 		struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
-		ebdp->cbd_bdu = 0;
+
 		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP &&
-			fep->hwts_tx_en)) {
-			ebdp->cbd_esc = (BD_ENET_TX_TS | BD_ENET_TX_INT);
+			fep->hwts_tx_en))
 			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-		} else {
-			ebdp->cbd_esc = BD_ENET_TX_INT;
 
-			/* Enable protocol checksum flags
-			 * We do not bother with the IP Checksum bits as they
-			 * are done by the kernel
-			 */
-			if (skb->ip_summed == CHECKSUM_PARTIAL)
-				ebdp->cbd_esc |= BD_ENET_TX_PINS;
-		}
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
+			estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
+
+		ebdp->cbd_bdu = 0;
+		ebdp->cbd_esc = estatus;
 	}
 
+	last_bdp = fep->cur_tx;
+	index = fec_enet_get_bd_index(fep->tx_bd_base, last_bdp, fep);
+	/* Save skb pointer */
+	fep->tx_skbuff[index] = skb;
+
+	bdp->cbd_datlen = buflen;
+
 	/* Send it on its way.  Tell FEC it's ready, interrupt when done,
 	 * it's the last BD of the frame, and to put the CRC on the end.
 	 */
-	status |= (BD_ENET_TX_READY | BD_ENET_TX_INTR
-			| BD_ENET_TX_LAST | BD_ENET_TX_TC);
+	status |= (BD_ENET_TX_READY | BD_ENET_TX_TC);
 	bdp->cbd_sc = status;
 
-	bdp_pre = fec_enet_get_prevdesc(bdp, fep);
-	if ((id_entry->driver_data & FEC_QUIRK_ERR006358) &&
-	    !(bdp_pre->cbd_sc & BD_ENET_TX_READY)) {
-		fep->delay_work.trig_tx = true;
-		schedule_delayed_work(&(fep->delay_work.delay_work),
-					msecs_to_jiffies(1));
-	}
+	fec_enet_submit_work(bdp, fep);
 
 	/* If this was the last BD in the ring, start at the beginning again. */
-	bdp = fec_enet_get_nextdesc(bdp, fep);
+	bdp = fec_enet_get_nextdesc(last_bdp, fep);
 
 	skb_tx_timestamp(skb);
 
 	fep->cur_tx = bdp;
 
-	if (fep->cur_tx == fep->dirty_tx)
-		netif_stop_queue(ndev);
+	/* Trigger transmission start */
+	writel(0, fep->hwp + FEC_X_DES_ACTIVE);
+
+	return 0;
+}
+
+static int
+fec_enet_txq_put_data_tso(struct sk_buff *skb, struct net_device *ndev,
+			struct bufdesc *bdp, int index, char *data,
+			int size, bool last_tcp, bool is_last)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	const struct platform_device_id *id_entry =
+				platform_get_device_id(fep->pdev);
+	struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
+	unsigned short status;
+	unsigned int estatus = 0;
+
+	status = bdp->cbd_sc;
+	status &= ~BD_ENET_TX_STATS;
+
+	status |= (BD_ENET_TX_TC | BD_ENET_TX_READY);
+	bdp->cbd_datlen = size;
+
+	if (((unsigned long) data) & FEC_ALIGNMENT ||
+		id_entry->driver_data & FEC_QUIRK_SWAP_FRAME) {
+		memcpy(fep->tx_bounce[index], data, size);
+		data = fep->tx_bounce[index];
+
+		if (id_entry->driver_data & FEC_QUIRK_SWAP_FRAME)
+			swap_buffer(data, size);
+	}
+
+	bdp->cbd_bufaddr = dma_map_single(&fep->pdev->dev, data,
+					size, DMA_TO_DEVICE);
+	if (dma_mapping_error(&fep->pdev->dev, bdp->cbd_bufaddr)) {
+		dev_kfree_skb_any(skb);
+		if (net_ratelimit())
+			netdev_err(ndev, "Tx DMA memory map failed\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	if (fep->bufdesc_ex) {
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
+			estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
+		ebdp->cbd_bdu = 0;
+		ebdp->cbd_esc = estatus;
+	}
+
+	/* Handle the last BD specially */
+	if (last_tcp)
+		status |= (BD_ENET_TX_LAST | BD_ENET_TX_TC);
+	if (is_last) {
+		status |= BD_ENET_TX_INTR;
+		if (fep->bufdesc_ex)
+			ebdp->cbd_esc |= BD_ENET_TX_INT;
+	}
+
+	bdp->cbd_sc = status;
+
+	return 0;
+}
+
+static int
+fec_enet_txq_put_hdr_tso(struct sk_buff *skb, struct net_device *ndev,
+			struct bufdesc *bdp, int index)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	const struct platform_device_id *id_entry =
+				platform_get_device_id(fep->pdev);
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
+	void *bufaddr;
+	unsigned long dmabuf;
+	unsigned short status;
+	unsigned int estatus = 0;
+
+	status = bdp->cbd_sc;
+	status &= ~BD_ENET_TX_STATS;
+	status |= (BD_ENET_TX_TC | BD_ENET_TX_READY);
+
+	bufaddr = fep->tso_hdrs + index * TSO_HEADER_SIZE;
+	dmabuf = fep->tso_hdrs_dma + index * TSO_HEADER_SIZE;
+	if (((unsigned long) bufaddr) & FEC_ALIGNMENT ||
+		id_entry->driver_data & FEC_QUIRK_SWAP_FRAME) {
+		memcpy(fep->tx_bounce[index], skb->data, hdr_len);
+		bufaddr = fep->tx_bounce[index];
+
+		if (id_entry->driver_data & FEC_QUIRK_SWAP_FRAME)
+			swap_buffer(bufaddr, hdr_len);
+
+		dmabuf = dma_map_single(&fep->pdev->dev, bufaddr,
+					hdr_len, DMA_TO_DEVICE);
+		if (dma_mapping_error(&fep->pdev->dev, dmabuf)) {
+			dev_kfree_skb_any(skb);
+			if (net_ratelimit())
+				netdev_err(ndev, "Tx DMA memory map failed\n");
+			return NETDEV_TX_BUSY;
+		}
+	}
+
+	bdp->cbd_bufaddr = dmabuf;
+	bdp->cbd_datlen = hdr_len;
+
+	if (fep->bufdesc_ex) {
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
+			estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
+		ebdp->cbd_bdu = 0;
+		ebdp->cbd_esc = estatus;
+	}
+
+	bdp->cbd_sc = status;
+
+	return 0;
+}
+
+static int fec_enet_txq_submit_tso(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	int total_len, data_left;
+	struct bufdesc *bdp = fep->cur_tx;
+	struct tso_t tso;
+	unsigned int index = 0;
+	int ret;
+
+	if (tso_count_descs(skb) >= fec_enet_get_free_txdesc_num(fep)) {
+		dev_kfree_skb_any(skb);
+		if (net_ratelimit())
+			netdev_err(ndev, "NOT enough BD for TSO!\n");
+		return NETDEV_TX_OK;
+	}
+
+	/* Protocol checksum off-load for TCP and UDP. */
+	if (fec_enet_clear_csum(skb, ndev)) {
+		dev_kfree_skb_any(skb);
+		return NETDEV_TX_OK;
+	}
+
+	/* Initialize the TSO handler, and prepare the first payload */
+	tso_start(skb, &tso);
+
+	total_len = skb->len - hdr_len;
+	while (total_len > 0) {
+		char *hdr;
+
+		index = fec_enet_get_bd_index(fep->tx_bd_base, bdp, fep);
+		data_left = min_t(int, skb_shinfo(skb)->gso_size, total_len);
+		total_len -= data_left;
+
+		/* prepare packet headers: MAC + IP + TCP */
+		hdr = fep->tso_hdrs + index * TSO_HEADER_SIZE;
+		tso_build_hdr(skb, hdr, &tso, data_left, total_len == 0);
+		ret = fec_enet_txq_put_hdr_tso(skb, ndev, bdp, index);
+		if (ret)
+			goto err_release;
+
+		while (data_left > 0) {
+			int size;
+
+			size = min_t(int, tso.size, data_left);
+			bdp = fec_enet_get_nextdesc(bdp, fep);
+			index = fec_enet_get_bd_index(fep->tx_bd_base, bdp, fep);
+			ret = fec_enet_txq_put_data_tso(skb, ndev, bdp, index, tso.data,
+							size, size == data_left,
+							total_len == 0);
+			if (ret)
+				goto err_release;
+
+			data_left -= size;
+			tso_build_data(skb, &tso, size);
+		}
+
+		bdp = fec_enet_get_nextdesc(bdp, fep);
+	}
+
+	/* Save skb pointer */
+	fep->tx_skbuff[index] = skb;
+
+	fec_enet_submit_work(bdp, fep);
+
+	skb_tx_timestamp(skb);
+	fep->cur_tx = bdp;
 
 	/* Trigger transmission start */
 	writel(0, fep->hwp + FEC_X_DES_ACTIVE);
 
+	return 0;
+
+err_release:
+	/* TODO: Release all used data descriptors for TSO */
+	return ret;
+}
+
+static netdev_tx_t
+fec_enet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	int entries_free;
+	int ret;
+
+	if (skb_is_gso(skb))
+		ret = fec_enet_txq_submit_tso(skb, ndev);
+	else
+		ret = fec_enet_txq_submit_skb(skb, ndev);
+	if (ret)
+		return ret;
+
+	entries_free = fec_enet_get_free_txdesc_num(fep);
+	if (entries_free <= fep->tx_stop_threshold)
+		netif_stop_queue(ndev);
+
 	return NETDEV_TX_OK;
 }
 
@@ -756,6 +1083,7 @@ fec_enet_tx(struct net_device *ndev)
 	unsigned short status;
 	struct	sk_buff	*skb;
 	int	index = 0;
+	int	entries_free;
 
 	fep = netdev_priv(ndev);
 	bdp = fep->dirty_tx;
@@ -769,16 +1097,17 @@ fec_enet_tx(struct net_device *ndev)
 		if (bdp == fep->cur_tx)
 			break;
 
-		if (fep->bufdesc_ex)
-			index = (struct bufdesc_ex *)bdp -
-				(struct bufdesc_ex *)fep->tx_bd_base;
-		else
-			index = bdp - fep->tx_bd_base;
+		index = fec_enet_get_bd_index(fep->tx_bd_base, bdp, fep);
 
 		skb = fep->tx_skbuff[index];
-		dma_unmap_single(&fep->pdev->dev, bdp->cbd_bufaddr, skb->len,
-				DMA_TO_DEVICE);
+		if (!IS_TSO_HEADER(fep, bdp->cbd_bufaddr))
+			dma_unmap_single(&fep->pdev->dev, bdp->cbd_bufaddr,
+					bdp->cbd_datlen, DMA_TO_DEVICE);
 		bdp->cbd_bufaddr = 0;
+		if (!skb) {
+			bdp = fec_enet_get_nextdesc(bdp, fep);
+			continue;
+		}
 
 		/* Check for errors. */
 		if (status & (BD_ENET_TX_HB | BD_ENET_TX_LC |
@@ -797,7 +1126,7 @@ fec_enet_tx(struct net_device *ndev)
 				ndev->stats.tx_carrier_errors++;
 		} else {
 			ndev->stats.tx_packets++;
-			ndev->stats.tx_bytes += bdp->cbd_datlen;
+			ndev->stats.tx_bytes += skb->len;
 		}
 
 		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS) &&
@@ -834,15 +1163,15 @@ fec_enet_tx(struct net_device *ndev)
 
 		/* Since we have freed up a buffer, the ring is no longer full
 		 */
-		if (fep->dirty_tx != fep->cur_tx) {
-			if (netif_queue_stopped(ndev))
+		if (netif_queue_stopped(ndev)) {
+			entries_free = fec_enet_get_free_txdesc_num(fep);
+			if (entries_free >= fep->tx_wake_threshold)
 				netif_wake_queue(ndev);
 		}
 	}
 	return;
 }
 
-
 /* During a receive, the cur_rx points to the current incoming buffer.
  * When we update through the ring, if the next incoming buffer has
  * not been given to the system, we just set the empty indicator,
@@ -920,11 +1249,7 @@ fec_enet_rx(struct net_device *ndev, int budget)
 		pkt_len = bdp->cbd_datlen;
 		ndev->stats.rx_bytes += pkt_len;
 
-		if (fep->bufdesc_ex)
-			index = (struct bufdesc_ex *)bdp -
-				(struct bufdesc_ex *)fep->rx_bd_base;
-		else
-			index = bdp - fep->rx_bd_base;
+		index = fec_enet_get_bd_index(fep->rx_bd_base, bdp, fep);
 		data = fep->rx_skbuff[index]->data;
 		dma_sync_single_for_cpu(&fep->pdev->dev, bdp->cbd_bufaddr,
 					FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE);
@@ -1255,6 +1580,49 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum,
 	return 0;
 }
 
+static int fec_enet_clk_enable(struct net_device *ndev, bool enable)
+{
+	struct fec_enet_private *fep = netdev_priv(ndev);
+	int ret;
+
+	if (enable) {
+		ret = clk_prepare_enable(fep->clk_ahb);
+		if (ret)
+			return ret;
+		ret = clk_prepare_enable(fep->clk_ipg);
+		if (ret)
+			goto failed_clk_ipg;
+		if (fep->clk_enet_out) {
+			ret = clk_prepare_enable(fep->clk_enet_out);
+			if (ret)
+				goto failed_clk_enet_out;
+		}
+		if (fep->clk_ptp) {
+			ret = clk_prepare_enable(fep->clk_ptp);
+			if (ret)
+				goto failed_clk_ptp;
+		}
+	} else {
+		clk_disable_unprepare(fep->clk_ahb);
+		clk_disable_unprepare(fep->clk_ipg);
+		if (fep->clk_enet_out)
+			clk_disable_unprepare(fep->clk_enet_out);
+		if (fep->clk_ptp)
+			clk_disable_unprepare(fep->clk_ptp);
+	}
+
+	return 0;
+failed_clk_ptp:
+	if (fep->clk_enet_out)
+		clk_disable_unprepare(fep->clk_enet_out);
+failed_clk_enet_out:
+		clk_disable_unprepare(fep->clk_ipg);
+failed_clk_ipg:
+		clk_disable_unprepare(fep->clk_ahb);
+
+	return ret;
+}
+
 static int fec_enet_mii_probe(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
@@ -1364,7 +1732,7 @@ static int fec_enet_mii_init(struct platform_device *pdev)
 	 * Reference Manual has an error on this, and gets fixed on i.MX6Q
 	 * document.
 	 */
-	fep->phy_speed = DIV_ROUND_UP(clk_get_rate(fep->clk_ahb), 5000000);
+	fep->phy_speed = DIV_ROUND_UP(clk_get_rate(fep->clk_ipg), 5000000);
 	if (id_entry->driver_data & FEC_QUIRK_ENET_MAC)
 		fep->phy_speed--;
 	fep->phy_speed <<= 1;
@@ -1773,6 +2141,11 @@ fec_enet_open(struct net_device *ndev)
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	int ret;
 
+	pinctrl_pm_select_default_state(&fep->pdev->dev);
+	ret = fec_enet_clk_enable(ndev, true);
+	if (ret)
+		return ret;
+
 	/* I should reset the ring buffers here, but I don't yet know
 	 * a simple way to do that.
 	 */
@@ -1811,6 +2184,8 @@ fec_enet_close(struct net_device *ndev)
 		phy_disconnect(fep->phy_dev);
 	}
 
+	fec_enet_clk_enable(ndev, false);
+	pinctrl_pm_select_sleep_state(&fep->pdev->dev);
 	fec_enet_free_buffers(ndev);
 
 	return 0;
@@ -1988,13 +2363,35 @@ static int fec_enet_init(struct net_device *ndev)
 	const struct platform_device_id *id_entry =
 				platform_get_device_id(fep->pdev);
 	struct bufdesc *cbd_base;
+	int bd_size;
+
+	/* init the tx & rx ring size */
+	fep->tx_ring_size = TX_RING_SIZE;
+	fep->rx_ring_size = RX_RING_SIZE;
+
+	fep->tx_stop_threshold = FEC_MAX_SKB_DESCS;
+	fep->tx_wake_threshold = (fep->tx_ring_size - fep->tx_stop_threshold) / 2;
+
+	if (fep->bufdesc_ex)
+		fep->bufdesc_size = sizeof(struct bufdesc_ex);
+	else
+		fep->bufdesc_size = sizeof(struct bufdesc);
+	bd_size = (fep->tx_ring_size + fep->rx_ring_size) *
+			fep->bufdesc_size;
 
 	/* Allocate memory for buffer descriptors. */
-	cbd_base = dma_alloc_coherent(NULL, PAGE_SIZE, &fep->bd_dma,
+	cbd_base = dma_alloc_coherent(NULL, bd_size, &fep->bd_dma,
 				      GFP_KERNEL);
 	if (!cbd_base)
 		return -ENOMEM;
 
+	fep->tso_hdrs = dma_alloc_coherent(NULL, fep->tx_ring_size * TSO_HEADER_SIZE,
+						&fep->tso_hdrs_dma, GFP_KERNEL);
+	if (!fep->tso_hdrs) {
+		dma_free_coherent(NULL, bd_size, cbd_base, fep->bd_dma);
+		return -ENOMEM;
+	}
+
 	memset(cbd_base, 0, PAGE_SIZE);
 
 	fep->netdev = ndev;
@@ -2004,10 +2401,6 @@ static int fec_enet_init(struct net_device *ndev)
 	/* make sure MAC we just acquired is programmed into the hw */
 	fec_set_mac_address(ndev, NULL);
 
-	/* init the tx & rx ring size */
-	fep->tx_ring_size = TX_RING_SIZE;
-	fep->rx_ring_size = RX_RING_SIZE;
-
 	/* Set receive and transmit descriptor base. */
 	fep->rx_bd_base = cbd_base;
 	if (fep->bufdesc_ex)
@@ -2024,21 +2417,21 @@ static int fec_enet_init(struct net_device *ndev)
 	writel(FEC_RX_DISABLED_IMASK, fep->hwp + FEC_IMASK);
 	netif_napi_add(ndev, &fep->napi, fec_enet_rx_napi, NAPI_POLL_WEIGHT);
 
-	if (id_entry->driver_data & FEC_QUIRK_HAS_VLAN) {
+	if (id_entry->driver_data & FEC_QUIRK_HAS_VLAN)
 		/* enable hw VLAN support */
 		ndev->features |= NETIF_F_HW_VLAN_CTAG_RX;
-		ndev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX;
-	}
 
 	if (id_entry->driver_data & FEC_QUIRK_HAS_CSUM) {
+		ndev->gso_max_segs = FEC_MAX_TSO_SEGS;
+
 		/* enable hw accelerator */
 		ndev->features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM
-				| NETIF_F_RXCSUM);
-		ndev->hw_features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM
-				| NETIF_F_RXCSUM);
+				| NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_TSO);
 		fep->csum_flags |= FLAG_RX_CSUM_ENABLED;
 	}
 
+	ndev->hw_features = ndev->features;
+
 	fec_restart(ndev, 0);
 
 	return 0;
@@ -2114,6 +2507,9 @@ fec_probe(struct platform_device *pdev)
 		fep->pause_flag |= FEC_PAUSE_FLAG_AUTONEG;
 #endif
 
+	/* Select default pin state */
+	pinctrl_pm_select_default_state(&pdev->dev);
+
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	fep->hwp = devm_ioremap_resource(&pdev->dev, r);
 	if (IS_ERR(fep->hwp)) {
@@ -2164,26 +2560,10 @@ fec_probe(struct platform_device *pdev)
 		fep->bufdesc_ex = 0;
 	}
 
-	ret = clk_prepare_enable(fep->clk_ahb);
+	ret = fec_enet_clk_enable(ndev, true);
 	if (ret)
 		goto failed_clk;
 
-	ret = clk_prepare_enable(fep->clk_ipg);
-	if (ret)
-		goto failed_clk_ipg;
-
-	if (fep->clk_enet_out) {
-		ret = clk_prepare_enable(fep->clk_enet_out);
-		if (ret)
-			goto failed_clk_enet_out;
-	}
-
-	if (fep->clk_ptp) {
-		ret = clk_prepare_enable(fep->clk_ptp);
-		if (ret)
-			goto failed_clk_ptp;
-	}
-
 	fep->reg_phy = devm_regulator_get(&pdev->dev, "phy");
 	if (!IS_ERR(fep->reg_phy)) {
 		ret = regulator_enable(fep->reg_phy);
@@ -2225,6 +2605,8 @@ fec_probe(struct platform_device *pdev)
 
 	/* Carrier starts down, phylib will bring it up */
 	netif_carrier_off(ndev);
+	fec_enet_clk_enable(ndev, false);
+	pinctrl_pm_select_sleep_state(&pdev->dev);
 
 	ret = register_netdev(ndev);
 	if (ret)
@@ -2244,15 +2626,7 @@ failed_init:
 	if (fep->reg_phy)
 		regulator_disable(fep->reg_phy);
 failed_regulator:
-	if (fep->clk_ptp)
-		clk_disable_unprepare(fep->clk_ptp);
-failed_clk_ptp:
-	if (fep->clk_enet_out)
-		clk_disable_unprepare(fep->clk_enet_out);
-failed_clk_enet_out:
-	clk_disable_unprepare(fep->clk_ipg);
-failed_clk_ipg:
-	clk_disable_unprepare(fep->clk_ahb);
+	fec_enet_clk_enable(ndev, false);
 failed_clk:
 failed_ioremap:
 	free_netdev(ndev);
@@ -2272,14 +2646,9 @@ fec_drv_remove(struct platform_device *pdev)
 	del_timer_sync(&fep->time_keep);
 	if (fep->reg_phy)
 		regulator_disable(fep->reg_phy);
-	if (fep->clk_ptp)
-		clk_disable_unprepare(fep->clk_ptp);
 	if (fep->ptp_clock)
 		ptp_clock_unregister(fep->ptp_clock);
-	if (fep->clk_enet_out)
-		clk_disable_unprepare(fep->clk_enet_out);
-	clk_disable_unprepare(fep->clk_ipg);
-	clk_disable_unprepare(fep->clk_ahb);
+	fec_enet_clk_enable(ndev, false);
 	free_netdev(ndev);
 
 	return 0;
@@ -2296,12 +2665,8 @@ fec_suspend(struct device *dev)
 		fec_stop(ndev);
 		netif_device_detach(ndev);
 	}
-	if (fep->clk_ptp)
-		clk_disable_unprepare(fep->clk_ptp);
-	if (fep->clk_enet_out)
-		clk_disable_unprepare(fep->clk_enet_out);
-	clk_disable_unprepare(fep->clk_ipg);
-	clk_disable_unprepare(fep->clk_ahb);
+	fec_enet_clk_enable(ndev, false);
+	pinctrl_pm_select_sleep_state(&fep->pdev->dev);
 
 	if (fep->reg_phy)
 		regulator_disable(fep->reg_phy);
@@ -2322,25 +2687,10 @@ fec_resume(struct device *dev)
 			return ret;
 	}
 
-	ret = clk_prepare_enable(fep->clk_ahb);
+	pinctrl_pm_select_default_state(&fep->pdev->dev);
+	ret = fec_enet_clk_enable(ndev, true);
 	if (ret)
-		goto failed_clk_ahb;
-
-	ret = clk_prepare_enable(fep->clk_ipg);
-	if (ret)
-		goto failed_clk_ipg;
-
-	if (fep->clk_enet_out) {
-		ret = clk_prepare_enable(fep->clk_enet_out);
-		if (ret)
-			goto failed_clk_enet_out;
-	}
-
-	if (fep->clk_ptp) {
-		ret = clk_prepare_enable(fep->clk_ptp);
-		if (ret)
-			goto failed_clk_ptp;
-	}
+		goto failed_clk;
 
 	if (netif_running(ndev)) {
 		fec_restart(ndev, fep->full_duplex);
@@ -2349,14 +2699,7 @@ fec_resume(struct device *dev)
 
 	return 0;
 
-failed_clk_ptp:
-	if (fep->clk_enet_out)
-		clk_disable_unprepare(fep->clk_enet_out);
-failed_clk_enet_out:
-	clk_disable_unprepare(fep->clk_ipg);
-failed_clk_ipg:
-	clk_disable_unprepare(fep->clk_ahb);
-failed_clk_ahb:
+failed_clk:
 	if (fep->reg_phy)
 		regulator_disable(fep->reg_phy);
 	return ret;
diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
index dc80db41d6b3..cfaf17b70f3f 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
+++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
@@ -792,10 +792,6 @@ static int fs_init_phy(struct net_device *dev)
 	phydev = of_phy_connect(dev, fep->fpi->phy_node, &fs_adjust_link, 0,
 				iface);
 	if (!phydev) {
-		phydev = of_phy_connect_fixed_link(dev, &fs_adjust_link,
-						   iface);
-	}
-	if (!phydev) {
 		dev_err(&dev->dev, "Could not attach to PHY\n");
 		return -ENODEV;
 	}
@@ -1029,9 +1025,16 @@ static int fs_enet_probe(struct platform_device *ofdev)
 	fpi->use_napi = 1;
 	fpi->napi_weight = 17;
 	fpi->phy_node = of_parse_phandle(ofdev->dev.of_node, "phy-handle", 0);
-	if ((!fpi->phy_node) && (!of_get_property(ofdev->dev.of_node, "fixed-link",
-						  NULL)))
-		goto out_free_fpi;
+	if (!fpi->phy_node && of_phy_is_fixed_link(ofdev->dev.of_node)) {
+		err = of_phy_register_fixed_link(ofdev->dev.of_node);
+		if (err)
+			goto out_free_fpi;
+
+		/* In the case of a fixed PHY, the DT node associated
+		 * to the PHY is the Ethernet MAC DT node.
+		 */
+		fpi->phy_node = ofdev->dev.of_node;
+	}
 
 	if (of_device_is_compatible(ofdev->dev.of_node, "fsl,mpc5125-fec")) {
 		phy_connection_type = of_get_property(ofdev->dev.of_node,
diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index ee6ddbd4f252..a6cf40e62f3a 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -889,6 +889,17 @@ static int gfar_of_init(struct platform_device *ofdev, struct net_device **pdev)
 
 	priv->phy_node = of_parse_phandle(np, "phy-handle", 0);
 
+	/* In the case of a fixed PHY, the DT node associated
+	 * to the PHY is the Ethernet MAC DT node.
+	 */
+	if (of_phy_is_fixed_link(np)) {
+		err = of_phy_register_fixed_link(np);
+		if (err)
+			goto err_grp_init;
+
+		priv->phy_node = np;
+	}
+
 	/* Find the TBI PHY.  If it's not there, we don't support SGMII */
 	priv->tbi_node = of_parse_phandle(np, "tbi-handle", 0);
 
@@ -1231,7 +1242,7 @@ static void gfar_hw_init(struct gfar_private *priv)
 		gfar_write_isrg(priv);
 }
 
-static void __init gfar_init_addr_hash_table(struct gfar_private *priv)
+static void gfar_init_addr_hash_table(struct gfar_private *priv)
 {
 	struct gfar __iomem *regs = priv->gfargrp[0].regs;
 
@@ -1373,6 +1384,9 @@ static int gfar_probe(struct platform_device *ofdev)
 
 	gfar_hw_init(priv);
 
+	/* Carrier starts down, phylib will bring it up */
+	netif_carrier_off(dev);
+
 	err = register_netdev(dev);
 
 	if (err) {
@@ -1380,9 +1394,6 @@ static int gfar_probe(struct platform_device *ofdev)
 		goto register_fail;
 	}
 
-	/* Carrier starts down, phylib will bring it up */
-	netif_carrier_off(dev);
-
 	device_init_wakeup(&dev->dev,
 			   priv->device_flags &
 			   FSL_GIANFAR_DEV_HAS_MAGIC_PACKET);
@@ -1660,9 +1671,6 @@ static int init_phy(struct net_device *dev)
 
 	priv->phydev = of_phy_connect(dev, priv->phy_node, &adjust_link, 0,
 				      interface);
-	if (!priv->phydev)
-		priv->phydev = of_phy_connect_fixed_link(dev, &adjust_link,
-							 interface);
 	if (!priv->phydev) {
 		dev_err(&dev->dev, "could not attach to PHY\n");
 		return -ENODEV;
diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c
index c8299c31b21f..fab39e295441 100644
--- a/drivers/net/ethernet/freescale/ucc_geth.c
+++ b/drivers/net/ethernet/freescale/ucc_geth.c
@@ -1728,9 +1728,6 @@ static int init_phy(struct net_device *dev)
 
 	phydev = of_phy_connect(dev, ug_info->phy_node, &adjust_link, 0,
 				priv->phy_interface);
-	if (!phydev)
-		phydev = of_phy_connect_fixed_link(dev, &adjust_link,
-						   priv->phy_interface);
 	if (!phydev) {
 		dev_err(&dev->dev, "Could not attach to PHY\n");
 		return -ENODEV;
@@ -3790,6 +3787,17 @@ static int ucc_geth_probe(struct platform_device* ofdev)
 	ug_info->uf_info.irq = irq_of_parse_and_map(np, 0);
 
 	ug_info->phy_node = of_parse_phandle(np, "phy-handle", 0);
+	if (!ug_info->phy_node) {
+		/* In the case of a fixed PHY, the DT node associated
+		 * to the PHY is the Ethernet MAC DT node.
+		 */
+		if (of_phy_is_fixed_link(np)) {
+			err = of_phy_register_fixed_link(np);
+			if (err)
+				return err;
+		}
+		ug_info->phy_node = np;
+	}
 
 	/* Find the TBI PHY node.  If it's not there, we don't support SGMII */
 	ug_info->tbi_node = of_parse_phandle(np, "tbi-handle", 0);
diff --git a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
index 413329eff2ff..cc83350d56ba 100644
--- a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
+++ b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
@@ -417,5 +417,5 @@ static const struct ethtool_ops uec_ethtool_ops = {
 
 void uec_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &uec_ethtool_ops);
+	netdev->ethtool_ops = &uec_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/freescale/xgmac_mdio.c b/drivers/net/ethernet/freescale/xgmac_mdio.c
index d449fcb90199..0c9d55c862ae 100644
--- a/drivers/net/ethernet/freescale/xgmac_mdio.c
+++ b/drivers/net/ethernet/freescale/xgmac_mdio.c
@@ -162,7 +162,9 @@ static int xgmac_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
 
 	/* Return all Fs if nothing was there */
 	if (in_be32(&regs->mdio_stat) & MDIO_STAT_RD_ER) {
-		dev_err(&bus->dev, "MDIO read error\n");
+		dev_err(&bus->dev,
+			"Error while reading PHY%d reg at %d.%d\n",
+			phy_id, dev_addr, regnum);
 		return 0xffff;
 	}
 
diff --git a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
index 7becab1aa3e4..cfe7a7431730 100644
--- a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
+++ b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
@@ -256,7 +256,7 @@ static int fmvj18x_probe(struct pcmcia_device *link)
     dev->netdev_ops = &fjn_netdev_ops;
     dev->watchdog_timeo = TX_TIMEOUT;
 
-    SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
+    dev->ethtool_ops = &netdev_ethtool_ops;
 
     return fmvj18x_config(link);
 } /* fmvj18x_attach */
diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig
new file mode 100644
index 000000000000..e9421731b05e
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/Kconfig
@@ -0,0 +1,27 @@
+#
+# HISILICON device configuration
+#
+
+config NET_VENDOR_HISILICON
+	bool "Hisilicon devices"
+	default y
+	depends on ARM
+	---help---
+	  If you have a network (Ethernet) card belonging to this class, say Y
+	  and read the Ethernet-HOWTO, available from
+	  <http://www.tldp.org/docs.html#howto>.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Hisilicon devices. If you say Y, you will be asked
+	  for your specific card in the following questions.
+
+if NET_VENDOR_HISILICON
+
+config HIX5HD2_GMAC
+	tristate "Hisilicon HIX5HD2 Family Network Device Support"
+	select PHYLIB
+	help
+	  This selects the hix5hd2 mac family network device.
+
+endif # NET_VENDOR_HISILICON
diff --git a/drivers/net/ethernet/hisilicon/Makefile b/drivers/net/ethernet/hisilicon/Makefile
new file mode 100644
index 000000000000..9175e84622d4
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the HISILICON network device drivers.
+#
+
+obj-$(CONFIG_HIX5HD2_GMAC) += hix5hd2_gmac.o
diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
new file mode 100644
index 000000000000..0ffdcd381fdd
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
@@ -0,0 +1,1066 @@
+/* Copyright (c) 2014 Linaro Ltd.
+ * Copyright (c) 2014 Hisilicon Limited.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/etherdevice.h>
+#include <linux/platform_device.h>
+#include <linux/of_net.h>
+#include <linux/of_mdio.h>
+#include <linux/clk.h>
+#include <linux/circ_buf.h>
+
+#define STATION_ADDR_LOW		0x0000
+#define STATION_ADDR_HIGH		0x0004
+#define MAC_DUPLEX_HALF_CTRL		0x0008
+#define MAX_FRM_SIZE			0x003c
+#define PORT_MODE			0x0040
+#define PORT_EN				0x0044
+#define BITS_TX_EN			BIT(2)
+#define BITS_RX_EN			BIT(1)
+#define REC_FILT_CONTROL		0x0064
+#define BIT_CRC_ERR_PASS		BIT(5)
+#define BIT_PAUSE_FRM_PASS		BIT(4)
+#define BIT_VLAN_DROP_EN		BIT(3)
+#define BIT_BC_DROP_EN			BIT(2)
+#define BIT_MC_MATCH_EN			BIT(1)
+#define BIT_UC_MATCH_EN			BIT(0)
+#define PORT_MC_ADDR_LOW		0x0068
+#define PORT_MC_ADDR_HIGH		0x006C
+#define CF_CRC_STRIP			0x01b0
+#define MODE_CHANGE_EN			0x01b4
+#define BIT_MODE_CHANGE_EN		BIT(0)
+#define COL_SLOT_TIME			0x01c0
+#define RECV_CONTROL			0x01e0
+#define BIT_STRIP_PAD_EN		BIT(3)
+#define BIT_RUNT_PKT_EN			BIT(4)
+#define CONTROL_WORD			0x0214
+#define MDIO_SINGLE_CMD			0x03c0
+#define MDIO_SINGLE_DATA		0x03c4
+#define MDIO_CTRL			0x03cc
+#define MDIO_RDATA_STATUS		0x03d0
+
+#define MDIO_START			BIT(20)
+#define MDIO_R_VALID			BIT(0)
+#define MDIO_READ			(BIT(17) | MDIO_START)
+#define MDIO_WRITE			(BIT(16) | MDIO_START)
+
+#define RX_FQ_START_ADDR		0x0500
+#define RX_FQ_DEPTH			0x0504
+#define RX_FQ_WR_ADDR			0x0508
+#define RX_FQ_RD_ADDR			0x050c
+#define RX_FQ_VLDDESC_CNT		0x0510
+#define RX_FQ_ALEMPTY_TH		0x0514
+#define RX_FQ_REG_EN			0x0518
+#define BITS_RX_FQ_START_ADDR_EN	BIT(2)
+#define BITS_RX_FQ_DEPTH_EN		BIT(1)
+#define BITS_RX_FQ_RD_ADDR_EN		BIT(0)
+#define RX_FQ_ALFULL_TH			0x051c
+#define RX_BQ_START_ADDR		0x0520
+#define RX_BQ_DEPTH			0x0524
+#define RX_BQ_WR_ADDR			0x0528
+#define RX_BQ_RD_ADDR			0x052c
+#define RX_BQ_FREE_DESC_CNT		0x0530
+#define RX_BQ_ALEMPTY_TH		0x0534
+#define RX_BQ_REG_EN			0x0538
+#define BITS_RX_BQ_START_ADDR_EN	BIT(2)
+#define BITS_RX_BQ_DEPTH_EN		BIT(1)
+#define BITS_RX_BQ_WR_ADDR_EN		BIT(0)
+#define RX_BQ_ALFULL_TH			0x053c
+#define TX_BQ_START_ADDR		0x0580
+#define TX_BQ_DEPTH			0x0584
+#define TX_BQ_WR_ADDR			0x0588
+#define TX_BQ_RD_ADDR			0x058c
+#define TX_BQ_VLDDESC_CNT		0x0590
+#define TX_BQ_ALEMPTY_TH		0x0594
+#define TX_BQ_REG_EN			0x0598
+#define BITS_TX_BQ_START_ADDR_EN	BIT(2)
+#define BITS_TX_BQ_DEPTH_EN		BIT(1)
+#define BITS_TX_BQ_RD_ADDR_EN		BIT(0)
+#define TX_BQ_ALFULL_TH			0x059c
+#define TX_RQ_START_ADDR		0x05a0
+#define TX_RQ_DEPTH			0x05a4
+#define TX_RQ_WR_ADDR			0x05a8
+#define TX_RQ_RD_ADDR			0x05ac
+#define TX_RQ_FREE_DESC_CNT		0x05b0
+#define TX_RQ_ALEMPTY_TH		0x05b4
+#define TX_RQ_REG_EN			0x05b8
+#define BITS_TX_RQ_START_ADDR_EN	BIT(2)
+#define BITS_TX_RQ_DEPTH_EN		BIT(1)
+#define BITS_TX_RQ_WR_ADDR_EN		BIT(0)
+#define TX_RQ_ALFULL_TH			0x05bc
+#define RAW_PMU_INT			0x05c0
+#define ENA_PMU_INT			0x05c4
+#define STATUS_PMU_INT			0x05c8
+#define MAC_FIFO_ERR_IN			BIT(30)
+#define TX_RQ_IN_TIMEOUT_INT		BIT(29)
+#define RX_BQ_IN_TIMEOUT_INT		BIT(28)
+#define TXOUTCFF_FULL_INT		BIT(27)
+#define TXOUTCFF_EMPTY_INT		BIT(26)
+#define TXCFF_FULL_INT			BIT(25)
+#define TXCFF_EMPTY_INT			BIT(24)
+#define RXOUTCFF_FULL_INT		BIT(23)
+#define RXOUTCFF_EMPTY_INT		BIT(22)
+#define RXCFF_FULL_INT			BIT(21)
+#define RXCFF_EMPTY_INT			BIT(20)
+#define TX_RQ_IN_INT			BIT(19)
+#define TX_BQ_OUT_INT			BIT(18)
+#define RX_BQ_IN_INT			BIT(17)
+#define RX_FQ_OUT_INT			BIT(16)
+#define TX_RQ_EMPTY_INT			BIT(15)
+#define TX_RQ_FULL_INT			BIT(14)
+#define TX_RQ_ALEMPTY_INT		BIT(13)
+#define TX_RQ_ALFULL_INT		BIT(12)
+#define TX_BQ_EMPTY_INT			BIT(11)
+#define TX_BQ_FULL_INT			BIT(10)
+#define TX_BQ_ALEMPTY_INT		BIT(9)
+#define TX_BQ_ALFULL_INT		BIT(8)
+#define RX_BQ_EMPTY_INT			BIT(7)
+#define RX_BQ_FULL_INT			BIT(6)
+#define RX_BQ_ALEMPTY_INT		BIT(5)
+#define RX_BQ_ALFULL_INT		BIT(4)
+#define RX_FQ_EMPTY_INT			BIT(3)
+#define RX_FQ_FULL_INT			BIT(2)
+#define RX_FQ_ALEMPTY_INT		BIT(1)
+#define RX_FQ_ALFULL_INT		BIT(0)
+
+#define DEF_INT_MASK			(RX_BQ_IN_INT | RX_BQ_IN_TIMEOUT_INT | \
+					TX_RQ_IN_INT | TX_RQ_IN_TIMEOUT_INT)
+
+#define DESC_WR_RD_ENA			0x05cc
+#define IN_QUEUE_TH			0x05d8
+#define OUT_QUEUE_TH			0x05dc
+#define QUEUE_TX_BQ_SHIFT		16
+#define RX_BQ_IN_TIMEOUT_TH		0x05e0
+#define TX_RQ_IN_TIMEOUT_TH		0x05e4
+#define STOP_CMD			0x05e8
+#define BITS_TX_STOP			BIT(1)
+#define BITS_RX_STOP			BIT(0)
+#define FLUSH_CMD			0x05eC
+#define BITS_TX_FLUSH_CMD		BIT(5)
+#define BITS_RX_FLUSH_CMD		BIT(4)
+#define BITS_TX_FLUSH_FLAG_DOWN		BIT(3)
+#define BITS_TX_FLUSH_FLAG_UP		BIT(2)
+#define BITS_RX_FLUSH_FLAG_DOWN		BIT(1)
+#define BITS_RX_FLUSH_FLAG_UP		BIT(0)
+#define RX_CFF_NUM_REG			0x05f0
+#define PMU_FSM_REG			0x05f8
+#define RX_FIFO_PKT_IN_NUM		0x05fc
+#define RX_FIFO_PKT_OUT_NUM		0x0600
+
+#define RGMII_SPEED_1000		0x2c
+#define RGMII_SPEED_100			0x2f
+#define RGMII_SPEED_10			0x2d
+#define MII_SPEED_100			0x0f
+#define MII_SPEED_10			0x0d
+#define GMAC_SPEED_1000			0x05
+#define GMAC_SPEED_100			0x01
+#define GMAC_SPEED_10			0x00
+#define GMAC_FULL_DUPLEX		BIT(4)
+
+#define RX_BQ_INT_THRESHOLD		0x01
+#define TX_RQ_INT_THRESHOLD		0x01
+#define RX_BQ_IN_TIMEOUT		0x10000
+#define TX_RQ_IN_TIMEOUT		0x50000
+
+#define MAC_MAX_FRAME_SIZE		1600
+#define DESC_SIZE			32
+#define RX_DESC_NUM			1024
+#define TX_DESC_NUM			1024
+
+#define DESC_VLD_FREE			0
+#define DESC_VLD_BUSY			0x80000000
+#define DESC_FL_MID			0
+#define DESC_FL_LAST			0x20000000
+#define DESC_FL_FIRST			0x40000000
+#define DESC_FL_FULL			0x60000000
+#define DESC_DATA_LEN_OFF		16
+#define DESC_BUFF_LEN_OFF		0
+#define DESC_DATA_MASK			0x7ff
+
+/* DMA descriptor ring helpers */
+#define dma_ring_incr(n, s)		(((n) + 1) & ((s) - 1))
+#define dma_cnt(n)			((n) >> 5)
+#define dma_byte(n)			((n) << 5)
+
+struct hix5hd2_desc {
+	__le32 buff_addr;
+	__le32 cmd;
+} __aligned(32);
+
+struct hix5hd2_desc_sw {
+	struct hix5hd2_desc *desc;
+	dma_addr_t	phys_addr;
+	unsigned int	count;
+	unsigned int	size;
+};
+
+#define QUEUE_NUMS	4
+struct hix5hd2_priv {
+	struct hix5hd2_desc_sw pool[QUEUE_NUMS];
+#define rx_fq		pool[0]
+#define rx_bq		pool[1]
+#define tx_bq		pool[2]
+#define tx_rq		pool[3]
+
+	void __iomem *base;
+	void __iomem *ctrl_base;
+
+	struct sk_buff *tx_skb[TX_DESC_NUM];
+	struct sk_buff *rx_skb[RX_DESC_NUM];
+
+	struct device *dev;
+	struct net_device *netdev;
+
+	struct phy_device *phy;
+	struct device_node *phy_node;
+	phy_interface_t	phy_mode;
+
+	unsigned int speed;
+	unsigned int duplex;
+
+	struct clk *clk;
+	struct mii_bus *bus;
+	struct napi_struct napi;
+	struct work_struct tx_timeout_task;
+};
+
+static void hix5hd2_config_port(struct net_device *dev, u32 speed, u32 duplex)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	u32 val;
+
+	priv->speed = speed;
+	priv->duplex = duplex;
+
+	switch (priv->phy_mode) {
+	case PHY_INTERFACE_MODE_RGMII:
+		if (speed == SPEED_1000)
+			val = RGMII_SPEED_1000;
+		else if (speed == SPEED_100)
+			val = RGMII_SPEED_100;
+		else
+			val = RGMII_SPEED_10;
+		break;
+	case PHY_INTERFACE_MODE_MII:
+		if (speed == SPEED_100)
+			val = MII_SPEED_100;
+		else
+			val = MII_SPEED_10;
+		break;
+	default:
+		netdev_warn(dev, "not supported mode\n");
+		val = MII_SPEED_10;
+		break;
+	}
+
+	if (duplex)
+		val |= GMAC_FULL_DUPLEX;
+	writel_relaxed(val, priv->ctrl_base);
+
+	writel_relaxed(BIT_MODE_CHANGE_EN, priv->base + MODE_CHANGE_EN);
+	if (speed == SPEED_1000)
+		val = GMAC_SPEED_1000;
+	else if (speed == SPEED_100)
+		val = GMAC_SPEED_100;
+	else
+		val = GMAC_SPEED_10;
+	writel_relaxed(val, priv->base + PORT_MODE);
+	writel_relaxed(0, priv->base + MODE_CHANGE_EN);
+	writel_relaxed(duplex, priv->base + MAC_DUPLEX_HALF_CTRL);
+}
+
+static void hix5hd2_set_desc_depth(struct hix5hd2_priv *priv, int rx, int tx)
+{
+	writel_relaxed(BITS_RX_FQ_DEPTH_EN, priv->base + RX_FQ_REG_EN);
+	writel_relaxed(rx << 3, priv->base + RX_FQ_DEPTH);
+	writel_relaxed(0, priv->base + RX_FQ_REG_EN);
+
+	writel_relaxed(BITS_RX_BQ_DEPTH_EN, priv->base + RX_BQ_REG_EN);
+	writel_relaxed(rx << 3, priv->base + RX_BQ_DEPTH);
+	writel_relaxed(0, priv->base + RX_BQ_REG_EN);
+
+	writel_relaxed(BITS_TX_BQ_DEPTH_EN, priv->base + TX_BQ_REG_EN);
+	writel_relaxed(tx << 3, priv->base + TX_BQ_DEPTH);
+	writel_relaxed(0, priv->base + TX_BQ_REG_EN);
+
+	writel_relaxed(BITS_TX_RQ_DEPTH_EN, priv->base + TX_RQ_REG_EN);
+	writel_relaxed(tx << 3, priv->base + TX_RQ_DEPTH);
+	writel_relaxed(0, priv->base + TX_RQ_REG_EN);
+}
+
+static void hix5hd2_set_rx_fq(struct hix5hd2_priv *priv, dma_addr_t phy_addr)
+{
+	writel_relaxed(BITS_RX_FQ_START_ADDR_EN, priv->base + RX_FQ_REG_EN);
+	writel_relaxed(phy_addr, priv->base + RX_FQ_START_ADDR);
+	writel_relaxed(0, priv->base + RX_FQ_REG_EN);
+}
+
+static void hix5hd2_set_rx_bq(struct hix5hd2_priv *priv, dma_addr_t phy_addr)
+{
+	writel_relaxed(BITS_RX_BQ_START_ADDR_EN, priv->base + RX_BQ_REG_EN);
+	writel_relaxed(phy_addr, priv->base + RX_BQ_START_ADDR);
+	writel_relaxed(0, priv->base + RX_BQ_REG_EN);
+}
+
+static void hix5hd2_set_tx_bq(struct hix5hd2_priv *priv, dma_addr_t phy_addr)
+{
+	writel_relaxed(BITS_TX_BQ_START_ADDR_EN, priv->base + TX_BQ_REG_EN);
+	writel_relaxed(phy_addr, priv->base + TX_BQ_START_ADDR);
+	writel_relaxed(0, priv->base + TX_BQ_REG_EN);
+}
+
+static void hix5hd2_set_tx_rq(struct hix5hd2_priv *priv, dma_addr_t phy_addr)
+{
+	writel_relaxed(BITS_TX_RQ_START_ADDR_EN, priv->base + TX_RQ_REG_EN);
+	writel_relaxed(phy_addr, priv->base + TX_RQ_START_ADDR);
+	writel_relaxed(0, priv->base + TX_RQ_REG_EN);
+}
+
+static void hix5hd2_set_desc_addr(struct hix5hd2_priv *priv)
+{
+	hix5hd2_set_rx_fq(priv, priv->rx_fq.phys_addr);
+	hix5hd2_set_rx_bq(priv, priv->rx_bq.phys_addr);
+	hix5hd2_set_tx_rq(priv, priv->tx_rq.phys_addr);
+	hix5hd2_set_tx_bq(priv, priv->tx_bq.phys_addr);
+}
+
+static void hix5hd2_hw_init(struct hix5hd2_priv *priv)
+{
+	u32 val;
+
+	/* disable and clear all interrupts */
+	writel_relaxed(0, priv->base + ENA_PMU_INT);
+	writel_relaxed(~0, priv->base + RAW_PMU_INT);
+
+	writel_relaxed(BIT_CRC_ERR_PASS, priv->base + REC_FILT_CONTROL);
+	writel_relaxed(MAC_MAX_FRAME_SIZE, priv->base + CONTROL_WORD);
+	writel_relaxed(0, priv->base + COL_SLOT_TIME);
+
+	val = RX_BQ_INT_THRESHOLD | TX_RQ_INT_THRESHOLD << QUEUE_TX_BQ_SHIFT;
+	writel_relaxed(val, priv->base + IN_QUEUE_TH);
+
+	writel_relaxed(RX_BQ_IN_TIMEOUT, priv->base + RX_BQ_IN_TIMEOUT_TH);
+	writel_relaxed(TX_RQ_IN_TIMEOUT, priv->base + TX_RQ_IN_TIMEOUT_TH);
+
+	hix5hd2_set_desc_depth(priv, RX_DESC_NUM, TX_DESC_NUM);
+	hix5hd2_set_desc_addr(priv);
+}
+
+static void hix5hd2_irq_enable(struct hix5hd2_priv *priv)
+{
+	writel_relaxed(DEF_INT_MASK, priv->base + ENA_PMU_INT);
+}
+
+static void hix5hd2_irq_disable(struct hix5hd2_priv *priv)
+{
+	writel_relaxed(0, priv->base + ENA_PMU_INT);
+}
+
+static void hix5hd2_port_enable(struct hix5hd2_priv *priv)
+{
+	writel_relaxed(0xf, priv->base + DESC_WR_RD_ENA);
+	writel_relaxed(BITS_RX_EN | BITS_TX_EN, priv->base + PORT_EN);
+}
+
+static void hix5hd2_port_disable(struct hix5hd2_priv *priv)
+{
+	writel_relaxed(~(BITS_RX_EN | BITS_TX_EN), priv->base + PORT_EN);
+	writel_relaxed(0, priv->base + DESC_WR_RD_ENA);
+}
+
+static void hix5hd2_hw_set_mac_addr(struct net_device *dev)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	unsigned char *mac = dev->dev_addr;
+	u32 val;
+
+	val = mac[1] | (mac[0] << 8);
+	writel_relaxed(val, priv->base + STATION_ADDR_HIGH);
+
+	val = mac[5] | (mac[4] << 8) | (mac[3] << 16) | (mac[2] << 24);
+	writel_relaxed(val, priv->base + STATION_ADDR_LOW);
+}
+
+static int hix5hd2_net_set_mac_address(struct net_device *dev, void *p)
+{
+	int ret;
+
+	ret = eth_mac_addr(dev, p);
+	if (!ret)
+		hix5hd2_hw_set_mac_addr(dev);
+
+	return ret;
+}
+
+static void hix5hd2_adjust_link(struct net_device *dev)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	struct phy_device *phy = priv->phy;
+
+	if ((priv->speed != phy->speed) || (priv->duplex != phy->duplex)) {
+		hix5hd2_config_port(dev, phy->speed, phy->duplex);
+		phy_print_status(phy);
+	}
+}
+
+static void hix5hd2_rx_refill(struct hix5hd2_priv *priv)
+{
+	struct hix5hd2_desc *desc;
+	struct sk_buff *skb;
+	u32 start, end, num, pos, i;
+	u32 len = MAC_MAX_FRAME_SIZE;
+	dma_addr_t addr;
+
+	/* software write pointer */
+	start = dma_cnt(readl_relaxed(priv->base + RX_FQ_WR_ADDR));
+	/* logic read pointer */
+	end = dma_cnt(readl_relaxed(priv->base + RX_FQ_RD_ADDR));
+	num = CIRC_SPACE(start, end, RX_DESC_NUM);
+
+	for (i = 0, pos = start; i < num; i++) {
+		if (priv->rx_skb[pos]) {
+			break;
+		} else {
+			skb = netdev_alloc_skb_ip_align(priv->netdev, len);
+			if (unlikely(skb == NULL))
+				break;
+		}
+
+		addr = dma_map_single(priv->dev, skb->data, len, DMA_FROM_DEVICE);
+		if (dma_mapping_error(priv->dev, addr)) {
+			dev_kfree_skb_any(skb);
+			break;
+		}
+
+		desc = priv->rx_fq.desc + pos;
+		desc->buff_addr = cpu_to_le32(addr);
+		priv->rx_skb[pos] = skb;
+		desc->cmd = cpu_to_le32(DESC_VLD_FREE |
+					(len - 1) << DESC_BUFF_LEN_OFF);
+		pos = dma_ring_incr(pos, RX_DESC_NUM);
+	}
+
+	/* ensure desc updated */
+	wmb();
+
+	if (pos != start)
+		writel_relaxed(dma_byte(pos), priv->base + RX_FQ_WR_ADDR);
+}
+
+static int hix5hd2_rx(struct net_device *dev, int limit)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	struct sk_buff *skb;
+	struct hix5hd2_desc *desc;
+	dma_addr_t addr;
+	u32 start, end, num, pos, i, len;
+
+	/* software read pointer */
+	start = dma_cnt(readl_relaxed(priv->base + RX_BQ_RD_ADDR));
+	/* logic write pointer */
+	end = dma_cnt(readl_relaxed(priv->base + RX_BQ_WR_ADDR));
+	num = CIRC_CNT(end, start, RX_DESC_NUM);
+	if (num > limit)
+		num = limit;
+
+	/* ensure get updated desc */
+	rmb();
+	for (i = 0, pos = start; i < num; i++) {
+		skb = priv->rx_skb[pos];
+		if (unlikely(!skb)) {
+			netdev_err(dev, "inconsistent rx_skb\n");
+			break;
+		}
+		priv->rx_skb[pos] = NULL;
+
+		desc = priv->rx_bq.desc + pos;
+		len = (le32_to_cpu(desc->cmd) >> DESC_DATA_LEN_OFF) &
+		       DESC_DATA_MASK;
+		addr = le32_to_cpu(desc->buff_addr);
+		dma_unmap_single(priv->dev, addr, MAC_MAX_FRAME_SIZE,
+				 DMA_FROM_DEVICE);
+
+		skb_put(skb, len);
+		if (skb->len > MAC_MAX_FRAME_SIZE) {
+			netdev_err(dev, "rcv len err, len = %d\n", skb->len);
+			dev->stats.rx_errors++;
+			dev->stats.rx_length_errors++;
+			dev_kfree_skb_any(skb);
+			goto next;
+		}
+
+		skb->protocol = eth_type_trans(skb, dev);
+		napi_gro_receive(&priv->napi, skb);
+		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += skb->len;
+		dev->last_rx = jiffies;
+next:
+		pos = dma_ring_incr(pos, RX_DESC_NUM);
+	}
+
+	if (pos != start)
+		writel_relaxed(dma_byte(pos), priv->base + RX_BQ_RD_ADDR);
+
+	hix5hd2_rx_refill(priv);
+
+	return num;
+}
+
+static void hix5hd2_xmit_reclaim(struct net_device *dev)
+{
+	struct sk_buff *skb;
+	struct hix5hd2_desc *desc;
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	unsigned int bytes_compl = 0, pkts_compl = 0;
+	u32 start, end, num, pos, i;
+	dma_addr_t addr;
+
+	netif_tx_lock(dev);
+
+	/* software read */
+	start = dma_cnt(readl_relaxed(priv->base + TX_RQ_RD_ADDR));
+	/* logic write */
+	end = dma_cnt(readl_relaxed(priv->base + TX_RQ_WR_ADDR));
+	num = CIRC_CNT(end, start, TX_DESC_NUM);
+
+	for (i = 0, pos = start; i < num; i++) {
+		skb = priv->tx_skb[pos];
+		if (unlikely(!skb)) {
+			netdev_err(dev, "inconsistent tx_skb\n");
+			break;
+		}
+
+		pkts_compl++;
+		bytes_compl += skb->len;
+		desc = priv->tx_rq.desc + pos;
+		addr = le32_to_cpu(desc->buff_addr);
+		dma_unmap_single(priv->dev, addr, skb->len, DMA_TO_DEVICE);
+		priv->tx_skb[pos] = NULL;
+		dev_consume_skb_any(skb);
+		pos = dma_ring_incr(pos, TX_DESC_NUM);
+	}
+
+	if (pos != start)
+		writel_relaxed(dma_byte(pos), priv->base + TX_RQ_RD_ADDR);
+
+	netif_tx_unlock(dev);
+
+	if (pkts_compl || bytes_compl)
+		netdev_completed_queue(dev, pkts_compl, bytes_compl);
+
+	if (unlikely(netif_queue_stopped(priv->netdev)) && pkts_compl)
+		netif_wake_queue(priv->netdev);
+}
+
+static int hix5hd2_poll(struct napi_struct *napi, int budget)
+{
+	struct hix5hd2_priv *priv = container_of(napi,
+				struct hix5hd2_priv, napi);
+	struct net_device *dev = priv->netdev;
+	int work_done = 0, task = budget;
+	int ints, num;
+
+	do {
+		hix5hd2_xmit_reclaim(dev);
+		num = hix5hd2_rx(dev, task);
+		work_done += num;
+		task -= num;
+		if ((work_done >= budget) || (num == 0))
+			break;
+
+		ints = readl_relaxed(priv->base + RAW_PMU_INT);
+		writel_relaxed(ints, priv->base + RAW_PMU_INT);
+	} while (ints & DEF_INT_MASK);
+
+	if (work_done < budget) {
+		napi_complete(napi);
+		hix5hd2_irq_enable(priv);
+	}
+
+	return work_done;
+}
+
+static irqreturn_t hix5hd2_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = (struct net_device *)dev_id;
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	int ints = readl_relaxed(priv->base + RAW_PMU_INT);
+
+	writel_relaxed(ints, priv->base + RAW_PMU_INT);
+	if (likely(ints & DEF_INT_MASK)) {
+		hix5hd2_irq_disable(priv);
+		napi_schedule(&priv->napi);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int hix5hd2_net_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	struct hix5hd2_desc *desc;
+	dma_addr_t addr;
+	u32 pos;
+
+	/* software write pointer */
+	pos = dma_cnt(readl_relaxed(priv->base + TX_BQ_WR_ADDR));
+	if (unlikely(priv->tx_skb[pos])) {
+		dev->stats.tx_dropped++;
+		dev->stats.tx_fifo_errors++;
+		netif_stop_queue(dev);
+		return NETDEV_TX_BUSY;
+	}
+
+	addr = dma_map_single(priv->dev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(priv->dev, addr)) {
+		dev_kfree_skb_any(skb);
+		return NETDEV_TX_OK;
+	}
+
+	desc = priv->tx_bq.desc + pos;
+	desc->buff_addr = cpu_to_le32(addr);
+	priv->tx_skb[pos] = skb;
+	desc->cmd = cpu_to_le32(DESC_VLD_BUSY | DESC_FL_FULL |
+				(skb->len & DESC_DATA_MASK) << DESC_DATA_LEN_OFF |
+				(skb->len & DESC_DATA_MASK) << DESC_BUFF_LEN_OFF);
+
+	/* ensure desc updated */
+	wmb();
+
+	pos = dma_ring_incr(pos, TX_DESC_NUM);
+	writel_relaxed(dma_byte(pos), priv->base + TX_BQ_WR_ADDR);
+
+	dev->trans_start = jiffies;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
+	netdev_sent_queue(dev, skb->len);
+
+	return NETDEV_TX_OK;
+}
+
+static void hix5hd2_free_dma_desc_rings(struct hix5hd2_priv *priv)
+{
+	struct hix5hd2_desc *desc;
+	dma_addr_t addr;
+	int i;
+
+	for (i = 0; i < RX_DESC_NUM; i++) {
+		struct sk_buff *skb = priv->rx_skb[i];
+		if (skb == NULL)
+			continue;
+
+		desc = priv->rx_fq.desc + i;
+		addr = le32_to_cpu(desc->buff_addr);
+		dma_unmap_single(priv->dev, addr,
+				 MAC_MAX_FRAME_SIZE, DMA_FROM_DEVICE);
+		dev_kfree_skb_any(skb);
+		priv->rx_skb[i] = NULL;
+	}
+
+	for (i = 0; i < TX_DESC_NUM; i++) {
+		struct sk_buff *skb = priv->tx_skb[i];
+		if (skb == NULL)
+			continue;
+
+		desc = priv->tx_rq.desc + i;
+		addr = le32_to_cpu(desc->buff_addr);
+		dma_unmap_single(priv->dev, addr, skb->len, DMA_TO_DEVICE);
+		dev_kfree_skb_any(skb);
+		priv->tx_skb[i] = NULL;
+	}
+}
+
+static int hix5hd2_net_open(struct net_device *dev)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+	int ret;
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret < 0) {
+		netdev_err(dev, "failed to enable clk %d\n", ret);
+		return ret;
+	}
+
+	priv->phy = of_phy_connect(dev, priv->phy_node,
+				   &hix5hd2_adjust_link, 0, priv->phy_mode);
+	if (!priv->phy)
+		return -ENODEV;
+
+	phy_start(priv->phy);
+	hix5hd2_hw_init(priv);
+	hix5hd2_rx_refill(priv);
+
+	netdev_reset_queue(dev);
+	netif_start_queue(dev);
+	napi_enable(&priv->napi);
+
+	hix5hd2_port_enable(priv);
+	hix5hd2_irq_enable(priv);
+
+	return 0;
+}
+
+static int hix5hd2_net_close(struct net_device *dev)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+
+	hix5hd2_port_disable(priv);
+	hix5hd2_irq_disable(priv);
+	napi_disable(&priv->napi);
+	netif_stop_queue(dev);
+	hix5hd2_free_dma_desc_rings(priv);
+
+	if (priv->phy) {
+		phy_stop(priv->phy);
+		phy_disconnect(priv->phy);
+	}
+
+	clk_disable_unprepare(priv->clk);
+
+	return 0;
+}
+
+static void hix5hd2_tx_timeout_task(struct work_struct *work)
+{
+	struct hix5hd2_priv *priv;
+
+	priv = container_of(work, struct hix5hd2_priv, tx_timeout_task);
+	hix5hd2_net_close(priv->netdev);
+	hix5hd2_net_open(priv->netdev);
+}
+
+static void hix5hd2_net_timeout(struct net_device *dev)
+{
+	struct hix5hd2_priv *priv = netdev_priv(dev);
+
+	schedule_work(&priv->tx_timeout_task);
+}
+
+static const struct net_device_ops hix5hd2_netdev_ops = {
+	.ndo_open		= hix5hd2_net_open,
+	.ndo_stop		= hix5hd2_net_close,
+	.ndo_start_xmit		= hix5hd2_net_xmit,
+	.ndo_tx_timeout		= hix5hd2_net_timeout,
+	.ndo_set_mac_address	= hix5hd2_net_set_mac_address,
+};
+
+static int hix5hd2_get_settings(struct net_device *net_dev,
+				struct ethtool_cmd *cmd)
+{
+	struct hix5hd2_priv *priv = netdev_priv(net_dev);
+
+	if (!priv->phy)
+		return -ENODEV;
+
+	return phy_ethtool_gset(priv->phy, cmd);
+}
+
+static int hix5hd2_set_settings(struct net_device *net_dev,
+				struct ethtool_cmd *cmd)
+{
+	struct hix5hd2_priv *priv = netdev_priv(net_dev);
+
+	if (!priv->phy)
+		return -ENODEV;
+
+	return phy_ethtool_sset(priv->phy, cmd);
+}
+
+static struct ethtool_ops hix5hd2_ethtools_ops = {
+	.get_link		= ethtool_op_get_link,
+	.get_settings		= hix5hd2_get_settings,
+	.set_settings		= hix5hd2_set_settings,
+};
+
+static int hix5hd2_mdio_wait_ready(struct mii_bus *bus)
+{
+	struct hix5hd2_priv *priv = bus->priv;
+	void __iomem *base = priv->base;
+	int i, timeout = 10000;
+
+	for (i = 0; readl_relaxed(base + MDIO_SINGLE_CMD) & MDIO_START; i++) {
+		if (i == timeout)
+			return -ETIMEDOUT;
+		usleep_range(10, 20);
+	}
+
+	return 0;
+}
+
+static int hix5hd2_mdio_read(struct mii_bus *bus, int phy, int reg)
+{
+	struct hix5hd2_priv *priv = bus->priv;
+	void __iomem *base = priv->base;
+	int val, ret;
+
+	ret = hix5hd2_mdio_wait_ready(bus);
+	if (ret < 0)
+		goto out;
+
+	writel_relaxed(MDIO_READ | phy << 8 | reg, base + MDIO_SINGLE_CMD);
+	ret = hix5hd2_mdio_wait_ready(bus);
+	if (ret < 0)
+		goto out;
+
+	val = readl_relaxed(base + MDIO_RDATA_STATUS);
+	if (val & MDIO_R_VALID) {
+		dev_err(bus->parent, "SMI bus read not valid\n");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	val = readl_relaxed(priv->base + MDIO_SINGLE_DATA);
+	ret = (val >> 16) & 0xFFFF;
+out:
+	return ret;
+}
+
+static int hix5hd2_mdio_write(struct mii_bus *bus, int phy, int reg, u16 val)
+{
+	struct hix5hd2_priv *priv = bus->priv;
+	void __iomem *base = priv->base;
+	int ret;
+
+	ret = hix5hd2_mdio_wait_ready(bus);
+	if (ret < 0)
+		goto out;
+
+	writel_relaxed(val, base + MDIO_SINGLE_DATA);
+	writel_relaxed(MDIO_WRITE | phy << 8 | reg, base + MDIO_SINGLE_CMD);
+	ret = hix5hd2_mdio_wait_ready(bus);
+out:
+	return ret;
+}
+
+static void hix5hd2_destroy_hw_desc_queue(struct hix5hd2_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < QUEUE_NUMS; i++) {
+		if (priv->pool[i].desc) {
+			dma_free_coherent(priv->dev, priv->pool[i].size,
+					  priv->pool[i].desc,
+					  priv->pool[i].phys_addr);
+			priv->pool[i].desc = NULL;
+		}
+	}
+}
+
+static int hix5hd2_init_hw_desc_queue(struct hix5hd2_priv *priv)
+{
+	struct device *dev = priv->dev;
+	struct hix5hd2_desc *virt_addr;
+	dma_addr_t phys_addr;
+	int size, i;
+
+	priv->rx_fq.count = RX_DESC_NUM;
+	priv->rx_bq.count = RX_DESC_NUM;
+	priv->tx_bq.count = TX_DESC_NUM;
+	priv->tx_rq.count = TX_DESC_NUM;
+
+	for (i = 0; i < QUEUE_NUMS; i++) {
+		size = priv->pool[i].count * sizeof(struct hix5hd2_desc);
+		virt_addr = dma_alloc_coherent(dev, size, &phys_addr,
+					       GFP_KERNEL);
+		if (virt_addr == NULL)
+			goto error_free_pool;
+
+		memset(virt_addr, 0, size);
+		priv->pool[i].size = size;
+		priv->pool[i].desc = virt_addr;
+		priv->pool[i].phys_addr = phys_addr;
+	}
+	return 0;
+
+error_free_pool:
+	hix5hd2_destroy_hw_desc_queue(priv);
+
+	return -ENOMEM;
+}
+
+static int hix5hd2_dev_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *node = dev->of_node;
+	struct net_device *ndev;
+	struct hix5hd2_priv *priv;
+	struct resource *res;
+	struct mii_bus *bus;
+	const char *mac_addr;
+	int ret;
+
+	ndev = alloc_etherdev(sizeof(struct hix5hd2_priv));
+	if (!ndev)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, ndev);
+
+	priv = netdev_priv(ndev);
+	priv->dev = dev;
+	priv->netdev = ndev;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	priv->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(priv->base)) {
+		ret = PTR_ERR(priv->base);
+		goto out_free_netdev;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	priv->ctrl_base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(priv->ctrl_base)) {
+		ret = PTR_ERR(priv->ctrl_base);
+		goto out_free_netdev;
+	}
+
+	priv->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(priv->clk)) {
+		netdev_err(ndev, "failed to get clk\n");
+		ret = -ENODEV;
+		goto out_free_netdev;
+	}
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret < 0) {
+		netdev_err(ndev, "failed to enable clk %d\n", ret);
+		goto out_free_netdev;
+	}
+
+	bus = mdiobus_alloc();
+	if (bus == NULL) {
+		ret = -ENOMEM;
+		goto out_free_netdev;
+	}
+
+	bus->priv = priv;
+	bus->name = "hix5hd2_mii_bus";
+	bus->read = hix5hd2_mdio_read;
+	bus->write = hix5hd2_mdio_write;
+	bus->parent = &pdev->dev;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mii", dev_name(&pdev->dev));
+	priv->bus = bus;
+
+	ret = of_mdiobus_register(bus, node);
+	if (ret)
+		goto err_free_mdio;
+
+	priv->phy_mode = of_get_phy_mode(node);
+	if (priv->phy_mode < 0) {
+		netdev_err(ndev, "not find phy-mode\n");
+		ret = -EINVAL;
+		goto err_mdiobus;
+	}
+
+	priv->phy_node = of_parse_phandle(node, "phy-handle", 0);
+	if (!priv->phy_node) {
+		netdev_err(ndev, "not find phy-handle\n");
+		ret = -EINVAL;
+		goto err_mdiobus;
+	}
+
+	ndev->irq = platform_get_irq(pdev, 0);
+	if (ndev->irq <= 0) {
+		netdev_err(ndev, "No irq resource\n");
+		ret = -EINVAL;
+		goto out_phy_node;
+	}
+
+	ret = devm_request_irq(dev, ndev->irq, hix5hd2_interrupt,
+			       0, pdev->name, ndev);
+	if (ret) {
+		netdev_err(ndev, "devm_request_irq failed\n");
+		goto out_phy_node;
+	}
+
+	mac_addr = of_get_mac_address(node);
+	if (mac_addr)
+		ether_addr_copy(ndev->dev_addr, mac_addr);
+	if (!is_valid_ether_addr(ndev->dev_addr)) {
+		eth_hw_addr_random(ndev);
+		netdev_warn(ndev, "using random MAC address %pM\n",
+			    ndev->dev_addr);
+	}
+
+	INIT_WORK(&priv->tx_timeout_task, hix5hd2_tx_timeout_task);
+	ndev->watchdog_timeo = 6 * HZ;
+	ndev->priv_flags |= IFF_UNICAST_FLT;
+	ndev->netdev_ops = &hix5hd2_netdev_ops;
+	ndev->ethtool_ops = &hix5hd2_ethtools_ops;
+	SET_NETDEV_DEV(ndev, dev);
+
+	ret = hix5hd2_init_hw_desc_queue(priv);
+	if (ret)
+		goto out_phy_node;
+
+	netif_napi_add(ndev, &priv->napi, hix5hd2_poll, NAPI_POLL_WEIGHT);
+	ret = register_netdev(priv->netdev);
+	if (ret) {
+		netdev_err(ndev, "register_netdev failed!");
+		goto out_destroy_queue;
+	}
+
+	clk_disable_unprepare(priv->clk);
+
+	return ret;
+
+out_destroy_queue:
+	netif_napi_del(&priv->napi);
+	hix5hd2_destroy_hw_desc_queue(priv);
+out_phy_node:
+	of_node_put(priv->phy_node);
+err_mdiobus:
+	mdiobus_unregister(bus);
+err_free_mdio:
+	mdiobus_free(bus);
+out_free_netdev:
+	free_netdev(ndev);
+
+	return ret;
+}
+
+static int hix5hd2_dev_remove(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct hix5hd2_priv *priv = netdev_priv(ndev);
+
+	netif_napi_del(&priv->napi);
+	unregister_netdev(ndev);
+	mdiobus_unregister(priv->bus);
+	mdiobus_free(priv->bus);
+
+	hix5hd2_destroy_hw_desc_queue(priv);
+	of_node_put(priv->phy_node);
+	cancel_work_sync(&priv->tx_timeout_task);
+	free_netdev(ndev);
+
+	return 0;
+}
+
+static const struct of_device_id hix5hd2_of_match[] = {
+	{.compatible = "hisilicon,hix5hd2-gmac",},
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, hix5hd2_of_match);
+
+static struct platform_driver hix5hd2_dev_driver = {
+	.driver = {
+		.name = "hix5hd2-gmac",
+		.of_match_table = hix5hd2_of_match,
+	},
+	.probe = hix5hd2_dev_probe,
+	.remove = hix5hd2_dev_remove,
+};
+
+module_platform_driver(hix5hd2_dev_driver);
+
+MODULE_DESCRIPTION("HISILICON HIX5HD2 Ethernet driver");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:hix5hd2-gmac");
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c b/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
index 95837b99a464..85a3866459cf 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_ethtool.c
@@ -63,8 +63,8 @@ static int ehea_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		cmd->duplex = port->full_duplex == 1 ?
 						     DUPLEX_FULL : DUPLEX_HALF;
 	} else {
-		speed = ~0;
-		cmd->duplex = -1;
+		speed = SPEED_UNKNOWN;
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 	ethtool_cmd_speed_set(cmd, speed);
 
@@ -278,5 +278,5 @@ static const struct ethtool_ops ehea_ethtool_ops = {
 
 void ehea_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &ehea_ethtool_ops);
+	netdev->ethtool_ops = &ehea_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c
index 538903bf13bc..a0b418e007a0 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -28,6 +28,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/device.h>
 #include <linux/in.h>
 #include <linux/ip.h>
 #include <linux/tcp.h>
@@ -3273,7 +3274,7 @@ static int ehea_probe_adapter(struct platform_device *dev)
 		return -EINVAL;
 	}
 
-	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL);
+	adapter = devm_kzalloc(&dev->dev, sizeof(*adapter), GFP_KERNEL);
 	if (!adapter) {
 		ret = -ENOMEM;
 		dev_err(&dev->dev, "no mem for ehea_adapter\n");
@@ -3359,7 +3360,6 @@ out_kill_eq:
 
 out_free_ad:
 	list_del(&adapter->list);
-	kfree(adapter);
 
 out:
 	ehea_update_firmware_handles();
@@ -3386,7 +3386,6 @@ static int ehea_remove(struct platform_device *dev)
 	ehea_destroy_eq(adapter->neq);
 	ehea_remove_adapter_mr(adapter);
 	list_del(&adapter->list);
-	kfree(adapter);
 
 	ehea_update_firmware_handles();
 
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_qmr.c b/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
index 9b03033bb557..a0820f72b25c 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_qmr.c
@@ -103,12 +103,14 @@ out_nomem:
 
 static void hw_queue_dtor(struct hw_queue *queue)
 {
-	int pages_per_kpage = PAGE_SIZE / queue->pagesize;
+	int pages_per_kpage;
 	int i, nr_pages;
 
 	if (!queue || !queue->queue_pages)
 		return;
 
+	pages_per_kpage = PAGE_SIZE / queue->pagesize;
+
 	nr_pages = queue->queue_length / queue->pagesize;
 
 	for (i = 0; i < nr_pages; i += pages_per_kpage)
diff --git a/drivers/net/ethernet/ibm/emac/core.c b/drivers/net/ethernet/ibm/emac/core.c
index ae342fdb42c8..87bd953cc2ee 100644
--- a/drivers/net/ethernet/ibm/emac/core.c
+++ b/drivers/net/ethernet/ibm/emac/core.c
@@ -2879,7 +2879,7 @@ static int emac_probe(struct platform_device *ofdev)
 		dev->commac.ops = &emac_commac_sg_ops;
 	} else
 		ndev->netdev_ops = &emac_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &emac_ethtool_ops);
+	ndev->ethtool_ops = &emac_ethtool_ops;
 
 	netif_carrier_off(ndev);
 
diff --git a/drivers/net/ethernet/icplus/ipg.c b/drivers/net/ethernet/icplus/ipg.c
index 25045ae07171..5727779a7df2 100644
--- a/drivers/net/ethernet/icplus/ipg.c
+++ b/drivers/net/ethernet/icplus/ipg.c
@@ -2245,7 +2245,7 @@ static int ipg_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	 */
 	dev->netdev_ops = &ipg_netdev_ops;
 	SET_NETDEV_DEV(dev, &pdev->dev);
-	SET_ETHTOOL_OPS(dev, &ipg_ethtool_ops);
+	dev->ethtool_ops = &ipg_ethtool_ops;
 
 	rc = pci_request_regions(pdev, DRV_NAME);
 	if (rc)
diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c
index b56461ce674c..9d979d7debef 100644
--- a/drivers/net/ethernet/intel/e100.c
+++ b/drivers/net/ethernet/intel/e100.c
@@ -2854,7 +2854,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->hw_features |= NETIF_F_RXALL;
 
 	netdev->netdev_ops = &e100_netdev_ops;
-	SET_ETHTOOL_OPS(netdev, &e100_ethtool_ops);
+	netdev->ethtool_ops = &e100_ethtool_ops;
 	netdev->watchdog_timeo = E100_WATCHDOG_PERIOD;
 	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
 
diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index 73a8aeefb92a..d50f78afb56d 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -168,8 +168,8 @@ static int e1000_get_settings(struct net_device *netdev,
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = ((hw->media_type == e1000_media_type_fiber) ||
@@ -1460,7 +1460,8 @@ static int e1000_run_loopback_test(struct e1000_adapter *adapter)
 			 * enough time to complete the receives, if it's
 			 * exceeded, break and error off
 			 */
-		} while (good_cnt < 64 && jiffies < (time + 20));
+		} while (good_cnt < 64 && time_after(time + 20, jiffies));
+
 		if (good_cnt != 64) {
 			ret_val = 13; /* ret_val is the same as mis-compare */
 			break;
@@ -1905,5 +1906,5 @@ static const struct ethtool_ops e1000_ethtool_ops = {
 
 void e1000_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &e1000_ethtool_ops);
+	netdev->ethtool_ops = &e1000_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/e1000/e1000_hw.c b/drivers/net/ethernet/intel/e1000/e1000_hw.c
index c1d3fdb296a0..e9b07ccc0eba 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_hw.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_hw.c
@@ -4877,10 +4877,10 @@ void e1000_tbi_adjust_stats(struct e1000_hw *hw, struct e1000_hw_stats *stats,
 	 * since the test for a multicast frame will test positive on
 	 * a broadcast frame.
 	 */
-	if ((mac_addr[0] == (u8) 0xff) && (mac_addr[1] == (u8) 0xff))
+	if (is_broadcast_ether_addr(mac_addr))
 		/* Broadcast packet */
 		stats->bprc++;
-	else if (*mac_addr & 0x01)
+	else if (is_multicast_ether_addr(mac_addr))
 		/* Multicast packet */
 		stats->mprc++;
 
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index 27058dfe418b..660971f304b2 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -3105,11 +3105,6 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
 	 */
 	tx_ring = adapter->tx_ring;
 
-	if (unlikely(skb->len <= 0)) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
 	/* On PCI/PCI-X HW, if packet size is less than ETH_ZLEN,
 	 * packets may get corrupted during padding by HW.
 	 * To WA this issue, pad all small packets manually.
diff --git a/drivers/net/ethernet/intel/e1000e/80003es2lan.c b/drivers/net/ethernet/intel/e1000e/80003es2lan.c
index a5f6b11d6992..08f22f348800 100644
--- a/drivers/net/ethernet/intel/e1000e/80003es2lan.c
+++ b/drivers/net/ethernet/intel/e1000e/80003es2lan.c
@@ -1365,6 +1365,7 @@ static const struct e1000_mac_operations es2_mac_ops = {
 	.setup_led		= e1000e_setup_led_generic,
 	.config_collision_dist	= e1000e_config_collision_dist_generic,
 	.rar_set		= e1000e_rar_set_generic,
+	.rar_get_count		= e1000e_rar_get_count_generic,
 };
 
 static const struct e1000_phy_operations es2_phy_ops = {
diff --git a/drivers/net/ethernet/intel/e1000e/82571.c b/drivers/net/ethernet/intel/e1000e/82571.c
index e0aa7f1efb08..218481e509f9 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.c
+++ b/drivers/net/ethernet/intel/e1000e/82571.c
@@ -1896,6 +1896,7 @@ static const struct e1000_mac_operations e82571_mac_ops = {
 	.config_collision_dist	= e1000e_config_collision_dist_generic,
 	.read_mac_addr		= e1000_read_mac_addr_82571,
 	.rar_set		= e1000e_rar_set_generic,
+	.rar_get_count		= e1000e_rar_get_count_generic,
 };
 
 static const struct e1000_phy_operations e82_phy_ops_igp = {
diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h
index 1471c5464a89..7785240a0da1 100644
--- a/drivers/net/ethernet/intel/e1000e/e1000.h
+++ b/drivers/net/ethernet/intel/e1000e/e1000.h
@@ -265,10 +265,10 @@ struct e1000_adapter {
 	u32 tx_hwtstamp_timeouts;
 
 	/* Rx */
-	bool (*clean_rx) (struct e1000_ring *ring, int *work_done,
-			  int work_to_do) ____cacheline_aligned_in_smp;
-	void (*alloc_rx_buf) (struct e1000_ring *ring, int cleaned_count,
-			      gfp_t gfp);
+	bool (*clean_rx)(struct e1000_ring *ring, int *work_done,
+			 int work_to_do) ____cacheline_aligned_in_smp;
+	void (*alloc_rx_buf)(struct e1000_ring *ring, int cleaned_count,
+			     gfp_t gfp);
 	struct e1000_ring *rx_ring;
 
 	u32 rx_int_delay;
@@ -391,6 +391,8 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca);
  * 25MHz	46-bit	2^46 / 10^9 / 3600 = 19.55 hours
  */
 #define E1000_SYSTIM_OVERFLOW_PERIOD	(HZ * 60 * 60 * 4)
+#define E1000_MAX_82574_SYSTIM_REREADS	50
+#define E1000_82574_SYSTIM_EPSILON	(1ULL << 35ULL)
 
 /* hardware capability, feature, and workaround flags */
 #define FLAG_HAS_AMT                      (1 << 0)
@@ -573,35 +575,8 @@ static inline u32 __er32(struct e1000_hw *hw, unsigned long reg)
 
 #define er32(reg)	__er32(hw, E1000_##reg)
 
-/**
- * __ew32_prepare - prepare to write to MAC CSR register on certain parts
- * @hw: pointer to the HW structure
- *
- * When updating the MAC CSR registers, the Manageability Engine (ME) could
- * be accessing the registers at the same time.  Normally, this is handled in
- * h/w by an arbiter but on some parts there is a bug that acknowledges Host
- * accesses later than it should which could result in the register to have
- * an incorrect value.  Workaround this by checking the FWSM register which
- * has bit 24 set while ME is accessing MAC CSR registers, wait if it is set
- * and try again a number of times.
- **/
-static inline s32 __ew32_prepare(struct e1000_hw *hw)
-{
-	s32 i = E1000_ICH_FWSM_PCIM2PCI_COUNT;
-
-	while ((er32(FWSM) & E1000_ICH_FWSM_PCIM2PCI) && --i)
-		udelay(50);
-
-	return i;
-}
-
-static inline void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val)
-{
-	if (hw->adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
-		__ew32_prepare(hw);
-
-	writel(val, hw->hw_addr + reg);
-}
+s32 __ew32_prepare(struct e1000_hw *hw);
+void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val);
 
 #define ew32(reg, val)	__ew32(hw, E1000_##reg, (val))
 
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index cad250bc1b99..815e26c6d34b 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -159,8 +159,8 @@ static int e1000_get_settings(struct net_device *netdev,
 		ecmd->transceiver = XCVR_EXTERNAL;
 	}
 
-	speed = -1;
-	ecmd->duplex = -1;
+	speed = SPEED_UNKNOWN;
+	ecmd->duplex = DUPLEX_UNKNOWN;
 
 	if (netif_running(netdev)) {
 		if (netif_carrier_ok(netdev)) {
@@ -169,6 +169,7 @@ static int e1000_get_settings(struct net_device *netdev,
 		}
 	} else if (!pm_runtime_suspended(netdev->dev.parent)) {
 		u32 status = er32(STATUS);
+
 		if (status & E1000_STATUS_LU) {
 			if (status & E1000_STATUS_SPEED_1000)
 				speed = SPEED_1000;
@@ -783,25 +784,26 @@ static bool reg_pattern_test(struct e1000_adapter *adapter, u64 *data,
 			      reg + (offset << 2), val,
 			      (test[pat] & write & mask));
 			*data = reg;
-			return 1;
+			return true;
 		}
 	}
-	return 0;
+	return false;
 }
 
 static bool reg_set_and_check(struct e1000_adapter *adapter, u64 *data,
 			      int reg, u32 mask, u32 write)
 {
 	u32 val;
+
 	__ew32(&adapter->hw, reg, write & mask);
 	val = __er32(&adapter->hw, reg);
 	if ((write & mask) != (val & mask)) {
 		e_err("set/check test failed (reg 0x%05X): got 0x%08X expected 0x%08X\n",
 		      reg, (val & mask), (write & mask));
 		*data = reg;
-		return 1;
+		return true;
 	}
-	return 0;
+	return false;
 }
 
 #define REG_PATTERN_TEST_ARRAY(reg, offset, mask, write)                       \
@@ -1717,6 +1719,7 @@ static int e1000_link_test(struct e1000_adapter *adapter, u64 *data)
 	*data = 0;
 	if (hw->phy.media_type == e1000_media_type_internal_serdes) {
 		int i = 0;
+
 		hw->mac.serdes_has_link = false;
 
 		/* On some blade server designs, link establishment
@@ -2315,5 +2318,5 @@ static const struct ethtool_ops e1000_ethtool_ops = {
 
 void e1000e_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &e1000_ethtool_ops);
+	netdev->ethtool_ops = &e1000_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index 6b3de5f39a97..72f5475c4b90 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -469,8 +469,9 @@ struct e1000_mac_operations {
 	s32  (*setup_led)(struct e1000_hw *);
 	void (*write_vfta)(struct e1000_hw *, u32, u32);
 	void (*config_collision_dist)(struct e1000_hw *);
-	void (*rar_set)(struct e1000_hw *, u8 *, u32);
+	int  (*rar_set)(struct e1000_hw *, u8 *, u32);
 	s32  (*read_mac_addr)(struct e1000_hw *);
+	u32  (*rar_get_count)(struct e1000_hw *);
 };
 
 /* When to use various PHY register access functions:
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index f0bbd4246d71..8894ab8ed6bd 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -139,8 +139,9 @@ static s32 e1000_k1_gig_workaround_hv(struct e1000_hw *hw, bool link);
 static s32 e1000_set_mdio_slow_mode_hv(struct e1000_hw *hw);
 static bool e1000_check_mng_mode_ich8lan(struct e1000_hw *hw);
 static bool e1000_check_mng_mode_pchlan(struct e1000_hw *hw);
-static void e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index);
-static void e1000_rar_set_pch_lpt(struct e1000_hw *hw, u8 *addr, u32 index);
+static int e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index);
+static int e1000_rar_set_pch_lpt(struct e1000_hw *hw, u8 *addr, u32 index);
+static u32 e1000_rar_get_count_pch_lpt(struct e1000_hw *hw);
 static s32 e1000_k1_workaround_lv(struct e1000_hw *hw);
 static void e1000_gate_hw_phy_config_ich8lan(struct e1000_hw *hw, bool gate);
 static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force);
@@ -704,6 +705,7 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw)
 		mac->ops.rar_set = e1000_rar_set_pch_lpt;
 		mac->ops.setup_physical_interface =
 		    e1000_setup_copper_link_pch_lpt;
+		mac->ops.rar_get_count = e1000_rar_get_count_pch_lpt;
 	}
 
 	/* Enable PCS Lock-loss workaround for ICH8 */
@@ -1334,6 +1336,7 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 	if (((hw->mac.type == e1000_pch2lan) ||
 	     (hw->mac.type == e1000_pch_lpt)) && link) {
 		u32 reg;
+
 		reg = er32(STATUS);
 		if (!(reg & (E1000_STATUS_FD | E1000_STATUS_SPEED_MASK))) {
 			u16 emi_addr;
@@ -1634,9 +1637,9 @@ static bool e1000_check_mng_mode_ich8lan(struct e1000_hw *hw)
 	u32 fwsm;
 
 	fwsm = er32(FWSM);
-	return ((fwsm & E1000_ICH_FWSM_FW_VALID) &&
+	return (fwsm & E1000_ICH_FWSM_FW_VALID) &&
 		((fwsm & E1000_FWSM_MODE_MASK) ==
-		 (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT)));
+		 (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT));
 }
 
 /**
@@ -1667,7 +1670,7 @@ static bool e1000_check_mng_mode_pchlan(struct e1000_hw *hw)
  *  contain the MAC address but RAR[1-6] are reserved for manageability (ME).
  *  Use SHRA[0-3] in place of those reserved for ME.
  **/
-static void e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index)
+static int e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index)
 {
 	u32 rar_low, rar_high;
 
@@ -1689,7 +1692,7 @@ static void e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index)
 		e1e_flush();
 		ew32(RAH(index), rar_high);
 		e1e_flush();
-		return;
+		return 0;
 	}
 
 	/* RAR[1-6] are owned by manageability.  Skip those and program the
@@ -1712,7 +1715,7 @@ static void e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index)
 		/* verify the register updates */
 		if ((er32(SHRAL(index - 1)) == rar_low) &&
 		    (er32(SHRAH(index - 1)) == rar_high))
-			return;
+			return 0;
 
 		e_dbg("SHRA[%d] might be locked by ME - FWSM=0x%8.8x\n",
 		      (index - 1), er32(FWSM));
@@ -1720,6 +1723,43 @@ static void e1000_rar_set_pch2lan(struct e1000_hw *hw, u8 *addr, u32 index)
 
 out:
 	e_dbg("Failed to write receive address at index %d\n", index);
+	return -E1000_ERR_CONFIG;
+}
+
+/**
+ *  e1000_rar_get_count_pch_lpt - Get the number of available SHRA
+ *  @hw: pointer to the HW structure
+ *
+ *  Get the number of available receive registers that the Host can
+ *  program. SHRA[0-10] are the shared receive address registers
+ *  that are shared between the Host and manageability engine (ME).
+ *  ME can reserve any number of addresses and the host needs to be
+ *  able to tell how many available registers it has access to.
+ **/
+static u32 e1000_rar_get_count_pch_lpt(struct e1000_hw *hw)
+{
+	u32 wlock_mac;
+	u32 num_entries;
+
+	wlock_mac = er32(FWSM) & E1000_FWSM_WLOCK_MAC_MASK;
+	wlock_mac >>= E1000_FWSM_WLOCK_MAC_SHIFT;
+
+	switch (wlock_mac) {
+	case 0:
+		/* All SHRA[0..10] and RAR[0] available */
+		num_entries = hw->mac.rar_entry_count;
+		break;
+	case 1:
+		/* Only RAR[0] available */
+		num_entries = 1;
+		break;
+	default:
+		/* SHRA[0..(wlock_mac - 1)] available + RAR[0] */
+		num_entries = wlock_mac + 1;
+		break;
+	}
+
+	return num_entries;
 }
 
 /**
@@ -1733,7 +1773,7 @@ out:
  *  contain the MAC address. SHRA[0-10] are the shared receive address
  *  registers that are shared between the Host and manageability engine (ME).
  **/
-static void e1000_rar_set_pch_lpt(struct e1000_hw *hw, u8 *addr, u32 index)
+static int e1000_rar_set_pch_lpt(struct e1000_hw *hw, u8 *addr, u32 index)
 {
 	u32 rar_low, rar_high;
 	u32 wlock_mac;
@@ -1755,7 +1795,7 @@ static void e1000_rar_set_pch_lpt(struct e1000_hw *hw, u8 *addr, u32 index)
 		e1e_flush();
 		ew32(RAH(index), rar_high);
 		e1e_flush();
-		return;
+		return 0;
 	}
 
 	/* The manageability engine (ME) can lock certain SHRAR registers that
@@ -1787,12 +1827,13 @@ static void e1000_rar_set_pch_lpt(struct e1000_hw *hw, u8 *addr, u32 index)
 			/* verify the register updates */
 			if ((er32(SHRAL_PCH_LPT(index - 1)) == rar_low) &&
 			    (er32(SHRAH_PCH_LPT(index - 1)) == rar_high))
-				return;
+				return 0;
 		}
 	}
 
 out:
 	e_dbg("Failed to write receive address at index %d\n", index);
+	return -E1000_ERR_CONFIG;
 }
 
 /**
@@ -4976,6 +5017,7 @@ static const struct e1000_mac_operations ich8_mac_ops = {
 	/* id_led_init dependent on mac type */
 	.config_collision_dist	= e1000e_config_collision_dist_generic,
 	.rar_set		= e1000e_rar_set_generic,
+	.rar_get_count		= e1000e_rar_get_count_generic,
 };
 
 static const struct e1000_phy_operations ich8_phy_ops = {
diff --git a/drivers/net/ethernet/intel/e1000e/mac.c b/drivers/net/ethernet/intel/e1000e/mac.c
index baa0a466d1d0..8c386f3a15eb 100644
--- a/drivers/net/ethernet/intel/e1000e/mac.c
+++ b/drivers/net/ethernet/intel/e1000e/mac.c
@@ -211,6 +211,11 @@ s32 e1000_check_alt_mac_addr_generic(struct e1000_hw *hw)
 	return 0;
 }
 
+u32 e1000e_rar_get_count_generic(struct e1000_hw *hw)
+{
+	return hw->mac.rar_entry_count;
+}
+
 /**
  *  e1000e_rar_set_generic - Set receive address register
  *  @hw: pointer to the HW structure
@@ -220,7 +225,7 @@ s32 e1000_check_alt_mac_addr_generic(struct e1000_hw *hw)
  *  Sets the receive address array register at index to the address passed
  *  in by addr.
  **/
-void e1000e_rar_set_generic(struct e1000_hw *hw, u8 *addr, u32 index)
+int e1000e_rar_set_generic(struct e1000_hw *hw, u8 *addr, u32 index)
 {
 	u32 rar_low, rar_high;
 
@@ -244,6 +249,8 @@ void e1000e_rar_set_generic(struct e1000_hw *hw, u8 *addr, u32 index)
 	e1e_flush();
 	ew32(RAH(index), rar_high);
 	e1e_flush();
+
+	return 0;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/e1000e/mac.h b/drivers/net/ethernet/intel/e1000e/mac.h
index 4e81c2825b7a..0513d90cdeea 100644
--- a/drivers/net/ethernet/intel/e1000e/mac.h
+++ b/drivers/net/ethernet/intel/e1000e/mac.h
@@ -61,7 +61,8 @@ void e1000e_update_adaptive(struct e1000_hw *hw);
 void e1000_write_vfta_generic(struct e1000_hw *hw, u32 offset, u32 value);
 
 void e1000_set_lan_id_multi_port_pcie(struct e1000_hw *hw);
-void e1000e_rar_set_generic(struct e1000_hw *hw, u8 *addr, u32 index);
+u32 e1000e_rar_get_count_generic(struct e1000_hw *hw);
+int e1000e_rar_set_generic(struct e1000_hw *hw, u8 *addr, u32 index);
 void e1000e_config_collision_dist_generic(struct e1000_hw *hw);
 
 #endif
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 3e69386add04..201cc93f3625 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -124,6 +124,36 @@ static const struct e1000_reg_info e1000_reg_info_tbl[] = {
 };
 
 /**
+ * __ew32_prepare - prepare to write to MAC CSR register on certain parts
+ * @hw: pointer to the HW structure
+ *
+ * When updating the MAC CSR registers, the Manageability Engine (ME) could
+ * be accessing the registers at the same time.  Normally, this is handled in
+ * h/w by an arbiter but on some parts there is a bug that acknowledges Host
+ * accesses later than it should which could result in the register to have
+ * an incorrect value.  Workaround this by checking the FWSM register which
+ * has bit 24 set while ME is accessing MAC CSR registers, wait if it is set
+ * and try again a number of times.
+ **/
+s32 __ew32_prepare(struct e1000_hw *hw)
+{
+	s32 i = E1000_ICH_FWSM_PCIM2PCI_COUNT;
+
+	while ((er32(FWSM) & E1000_ICH_FWSM_PCIM2PCI) && --i)
+		udelay(50);
+
+	return i;
+}
+
+void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val)
+{
+	if (hw->adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
+		__ew32_prepare(hw);
+
+	writel(val, hw->hw_addr + reg);
+}
+
+/**
  * e1000_regdump - register printout routine
  * @hw: pointer to the HW structure
  * @reginfo: pointer to the register info table
@@ -599,6 +629,7 @@ static void e1000e_update_rdt_wa(struct e1000_ring *rx_ring, unsigned int i)
 
 	if (unlikely(!ret_val && (i != readl(rx_ring->tail)))) {
 		u32 rctl = er32(RCTL);
+
 		ew32(RCTL, rctl & ~E1000_RCTL_EN);
 		e_err("ME firmware caused invalid RDT - resetting\n");
 		schedule_work(&adapter->reset_task);
@@ -615,6 +646,7 @@ static void e1000e_update_tdt_wa(struct e1000_ring *tx_ring, unsigned int i)
 
 	if (unlikely(!ret_val && (i != readl(tx_ring->tail)))) {
 		u32 tctl = er32(TCTL);
+
 		ew32(TCTL, tctl & ~E1000_TCTL_EN);
 		e_err("ME firmware caused invalid TDT - resetting\n");
 		schedule_work(&adapter->reset_task);
@@ -1198,6 +1230,7 @@ static bool e1000_clean_tx_irq(struct e1000_ring *tx_ring)
 	while ((eop_desc->upper.data & cpu_to_le32(E1000_TXD_STAT_DD)) &&
 	       (count < tx_ring->count)) {
 		bool cleaned = false;
+
 		rmb();		/* read buffer_info after eop_desc */
 		for (; !cleaned; count++) {
 			tx_desc = E1000_TX_DESC(*tx_ring, i);
@@ -1753,6 +1786,7 @@ static irqreturn_t e1000_intr_msi(int __always_unused irq, void *data)
 		    adapter->flags & FLAG_RX_NEEDS_RESTART) {
 			/* disable receives */
 			u32 rctl = er32(RCTL);
+
 			ew32(RCTL, rctl & ~E1000_RCTL_EN);
 			adapter->flags |= FLAG_RESTART_NOW;
 		}
@@ -1960,6 +1994,7 @@ static void e1000_configure_msix(struct e1000_adapter *adapter)
 	/* Workaround issue with spurious interrupts on 82574 in MSI-X mode */
 	if (hw->mac.type == e1000_82574) {
 		u32 rfctl = er32(RFCTL);
+
 		rfctl |= E1000_RFCTL_ACK_DIS;
 		ew32(RFCTL, rfctl);
 	}
@@ -2204,6 +2239,7 @@ static void e1000_irq_disable(struct e1000_adapter *adapter)
 
 	if (adapter->msix_entries) {
 		int i;
+
 		for (i = 0; i < adapter->num_vectors; i++)
 			synchronize_irq(adapter->msix_entries[i].vector);
 	} else {
@@ -2921,6 +2957,7 @@ static void e1000_configure_tx(struct e1000_adapter *adapter)
 
 	if (adapter->flags2 & FLAG2_DMA_BURST) {
 		u32 txdctl = er32(TXDCTL(0));
+
 		txdctl &= ~(E1000_TXDCTL_PTHRESH | E1000_TXDCTL_HTHRESH |
 			    E1000_TXDCTL_WTHRESH);
 		/* set up some performance related parameters to encourage the
@@ -3239,6 +3276,7 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 
 		if (adapter->flags & FLAG_IS_ICH) {
 			u32 rxdctl = er32(RXDCTL(0));
+
 			ew32(RXDCTL(0), rxdctl | 0x3);
 		}
 
@@ -3303,9 +3341,11 @@ static int e1000e_write_uc_addr_list(struct net_device *netdev)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
-	unsigned int rar_entries = hw->mac.rar_entry_count;
+	unsigned int rar_entries;
 	int count = 0;
 
+	rar_entries = hw->mac.ops.rar_get_count(hw);
+
 	/* save a rar entry for our hardware address */
 	rar_entries--;
 
@@ -3324,9 +3364,13 @@ static int e1000e_write_uc_addr_list(struct net_device *netdev)
 		 * combining
 		 */
 		netdev_for_each_uc_addr(ha, netdev) {
+			int rval;
+
 			if (!rar_entries)
 				break;
-			hw->mac.ops.rar_set(hw, ha->addr, rar_entries--);
+			rval = hw->mac.ops.rar_set(hw, ha->addr, rar_entries--);
+			if (rval < 0)
+				return -ENOMEM;
 			count++;
 		}
 	}
@@ -4085,12 +4129,37 @@ static cycle_t e1000e_cyclecounter_read(const struct cyclecounter *cc)
 	struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter,
 						     cc);
 	struct e1000_hw *hw = &adapter->hw;
-	cycle_t systim;
+	cycle_t systim, systim_next;
 
 	/* latch SYSTIMH on read of SYSTIML */
 	systim = (cycle_t)er32(SYSTIML);
 	systim |= (cycle_t)er32(SYSTIMH) << 32;
 
+	if ((hw->mac.type == e1000_82574) || (hw->mac.type == e1000_82583)) {
+		u64 incvalue, time_delta, rem, temp;
+		int i;
+
+		/* errata for 82574/82583 possible bad bits read from SYSTIMH/L
+		 * check to see that the time is incrementing at a reasonable
+		 * rate and is a multiple of incvalue
+		 */
+		incvalue = er32(TIMINCA) & E1000_TIMINCA_INCVALUE_MASK;
+		for (i = 0; i < E1000_MAX_82574_SYSTIM_REREADS; i++) {
+			/* latch SYSTIMH on read of SYSTIML */
+			systim_next = (cycle_t)er32(SYSTIML);
+			systim_next |= (cycle_t)er32(SYSTIMH) << 32;
+
+			time_delta = systim_next - systim;
+			temp = time_delta;
+			rem = do_div(temp, incvalue);
+
+			systim = systim_next;
+
+			if ((time_delta < E1000_82574_SYSTIM_EPSILON) &&
+			    (rem == 0))
+				break;
+		}
+	}
 	return systim;
 }
 
@@ -4491,7 +4560,7 @@ static void e1000e_update_phy_task(struct work_struct *work)
 	e1000_get_phy_info(hw);
 
 	/* Enable EEE on 82579 after link up */
-	if (hw->phy.type == e1000_phy_82579)
+	if (hw->phy.type >= e1000_phy_82579)
 		e1000_set_eee_pchlan(hw);
 }
 
@@ -4695,6 +4764,7 @@ static void e1000e_update_stats(struct e1000_adapter *adapter)
 	/* Correctable ECC Errors */
 	if (hw->mac.type == e1000_pch_lpt) {
 		u32 pbeccsts = er32(PBECCSTS);
+
 		adapter->corr_errors +=
 		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
 		adapter->uncorr_errors +=
@@ -4808,6 +4878,7 @@ static void e1000e_enable_receives(struct e1000_adapter *adapter)
 	    (adapter->flags & FLAG_RESTART_NOW)) {
 		struct e1000_hw *hw = &adapter->hw;
 		u32 rctl = er32(RCTL);
+
 		ew32(RCTL, rctl | E1000_RCTL_EN);
 		adapter->flags &= ~FLAG_RESTART_NOW;
 	}
@@ -4930,6 +5001,7 @@ static void e1000_watchdog_task(struct work_struct *work)
 			if ((adapter->flags & FLAG_TARC_SPEED_MODE_BIT) &&
 			    !txb2b) {
 				u32 tarc0;
+
 				tarc0 = er32(TARC(0));
 				tarc0 &= ~SPEED_MODE_BIT;
 				ew32(TARC(0), tarc0);
@@ -5170,7 +5242,7 @@ static bool e1000_tx_csum(struct e1000_ring *tx_ring, struct sk_buff *skb)
 	__be16 protocol;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
-		return 0;
+		return false;
 
 	if (skb->protocol == cpu_to_be16(ETH_P_8021Q))
 		protocol = vlan_eth_hdr(skb)->h_vlan_encapsulated_proto;
@@ -5215,7 +5287,7 @@ static bool e1000_tx_csum(struct e1000_ring *tx_ring, struct sk_buff *skb)
 		i = 0;
 	tx_ring->next_to_use = i;
 
-	return 1;
+	return true;
 }
 
 static int e1000_tx_map(struct e1000_ring *tx_ring, struct sk_buff *skb,
@@ -6209,6 +6281,7 @@ static int __e1000_resume(struct pci_dev *pdev)
 		e1e_wphy(&adapter->hw, BM_WUS, ~0);
 	} else {
 		u32 wus = er32(WUS);
+
 		if (wus) {
 			e_info("MAC Wakeup cause - %s\n",
 			       wus & E1000_WUS_EX ? "Unicast Packet" :
@@ -7027,7 +7100,7 @@ static const struct pci_error_handlers e1000_err_handler = {
 	.resume = e1000_io_resume,
 };
 
-static DEFINE_PCI_DEVICE_TABLE(e1000_pci_tbl) = {
+static const struct pci_device_id e1000_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_COPPER), board_82571 },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_FIBER), board_82571 },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_COPPER), board_82571 },
@@ -7144,6 +7217,7 @@ static struct pci_driver e1000_driver = {
 static int __init e1000_init_module(void)
 {
 	int ret;
+
 	pr_info("Intel(R) PRO/1000 Network Driver - %s\n",
 		e1000e_driver_version);
 	pr_info("Copyright(c) 1999 - 2014 Intel Corporation.\n");
diff --git a/drivers/net/ethernet/intel/e1000e/nvm.c b/drivers/net/ethernet/intel/e1000e/nvm.c
index a9a976f04bff..b1f212b7baf7 100644
--- a/drivers/net/ethernet/intel/e1000e/nvm.c
+++ b/drivers/net/ethernet/intel/e1000e/nvm.c
@@ -398,6 +398,7 @@ s32 e1000e_write_nvm_spi(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
 		/* Loop to allow for up to whole page write of eeprom */
 		while (widx < words) {
 			u16 word_out = data[widx];
+
 			word_out = (word_out >> 8) | (word_out << 8);
 			e1000_shift_out_eec_bits(hw, word_out, 16);
 			widx++;
diff --git a/drivers/net/ethernet/intel/e1000e/param.c b/drivers/net/ethernet/intel/e1000e/param.c
index d0ac0f3249c8..aa1923f7ebdd 100644
--- a/drivers/net/ethernet/intel/e1000e/param.c
+++ b/drivers/net/ethernet/intel/e1000e/param.c
@@ -436,6 +436,7 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 
 		if (num_IntMode > bd) {
 			unsigned int int_mode = IntMode[bd];
+
 			e1000_validate_option(&int_mode, &opt, adapter);
 			adapter->int_mode = int_mode;
 		} else {
@@ -457,6 +458,7 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 
 		if (num_SmartPowerDownEnable > bd) {
 			unsigned int spd = SmartPowerDownEnable[bd];
+
 			e1000_validate_option(&spd, &opt, adapter);
 			if ((adapter->flags & FLAG_HAS_SMART_POWER_DOWN) && spd)
 				adapter->flags |= FLAG_SMART_POWER_DOWN;
@@ -473,6 +475,7 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 
 		if (num_CrcStripping > bd) {
 			unsigned int crc_stripping = CrcStripping[bd];
+
 			e1000_validate_option(&crc_stripping, &opt, adapter);
 			if (crc_stripping == OPTION_ENABLED) {
 				adapter->flags2 |= FLAG2_CRC_STRIPPING;
@@ -495,6 +498,7 @@ void e1000e_check_options(struct e1000_adapter *adapter)
 
 		if (num_KumeranLockLoss > bd) {
 			unsigned int kmrn_lock_loss = KumeranLockLoss[bd];
+
 			e1000_validate_option(&kmrn_lock_loss, &opt, adapter);
 			enabled = kmrn_lock_loss;
 		}
diff --git a/drivers/net/ethernet/intel/e1000e/phy.c b/drivers/net/ethernet/intel/e1000e/phy.c
index 00b3fc98bf30..b2005e13fb01 100644
--- a/drivers/net/ethernet/intel/e1000e/phy.c
+++ b/drivers/net/ethernet/intel/e1000e/phy.c
@@ -2896,6 +2896,7 @@ static s32 __e1000_write_phy_reg_hv(struct e1000_hw *hw, u32 offset, u16 data,
 		    (hw->phy.addr == 2) &&
 		    !(MAX_PHY_REG_ADDRESS & reg) && (data & (1 << 11))) {
 			u16 data2 = 0x7EFF;
+
 			ret_val = e1000_access_phy_debug_regs_hv(hw,
 								 (1 << 6) | 0x3,
 								 &data2, false);
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index beb7b4393a6c..65985846345d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -72,6 +72,7 @@
 #define I40E_MIN_NUM_DESCRIPTORS      64
 #define I40E_MIN_MSIX                 2
 #define I40E_DEFAULT_NUM_VMDQ_VSI     8 /* max 256 VSIs */
+#define I40E_MIN_VSI_ALLOC            51 /* LAN, ATR, FCOE, 32 VF, 16 VMDQ */
 #define I40E_DEFAULT_QUEUES_PER_VMDQ  2 /* max 16 qps */
 #define I40E_DEFAULT_QUEUES_PER_VF    4
 #define I40E_DEFAULT_QUEUES_PER_TC    1 /* should be a power of 2 */
@@ -97,10 +98,6 @@
 #define STRINGIFY(foo)  #foo
 #define XSTRINGIFY(bar) STRINGIFY(bar)
 
-#ifndef ARCH_HAS_PREFETCH
-#define prefetch(X)
-#endif
-
 #define I40E_RX_DESC(R, i)			\
 	((ring_is_16byte_desc_enabled(R))	\
 		? (union i40e_32byte_rx_desc *)	\
@@ -157,11 +154,23 @@ struct i40e_lump_tracking {
 #define I40E_FDIR_BUFFER_FULL_MARGIN	10
 #define I40E_FDIR_BUFFER_HEAD_ROOM	200
 
+enum i40e_fd_stat_idx {
+	I40E_FD_STAT_ATR,
+	I40E_FD_STAT_SB,
+	I40E_FD_STAT_PF_COUNT
+};
+#define I40E_FD_STAT_PF_IDX(pf_id) ((pf_id) * I40E_FD_STAT_PF_COUNT)
+#define I40E_FD_ATR_STAT_IDX(pf_id) \
+			(I40E_FD_STAT_PF_IDX(pf_id) + I40E_FD_STAT_ATR)
+#define I40E_FD_SB_STAT_IDX(pf_id)  \
+			(I40E_FD_STAT_PF_IDX(pf_id) + I40E_FD_STAT_SB)
+
 struct i40e_fdir_filter {
 	struct hlist_node fdir_node;
 	/* filter ipnut set */
 	u8 flow_type;
 	u8 ip4_proto;
+	/* TX packet view of src and dst */
 	__be32 dst_ip[4];
 	__be32 src_ip[4];
 	__be16 src_port;
@@ -205,7 +214,6 @@ struct i40e_pf {
 	unsigned long state;
 	unsigned long link_check_timeout;
 	struct msix_entry *msix_entries;
-	u16 num_msix_entries;
 	bool fc_autoneg_status;
 
 	u16 eeprom_version;
@@ -220,11 +228,14 @@ struct i40e_pf {
 	u16 rss_size;              /* num queues in the RSS array */
 	u16 rss_size_max;          /* HW defined max RSS queues */
 	u16 fdir_pf_filter_count;  /* num of guaranteed filters for this PF */
+	u16 num_alloc_vsi;         /* num VSIs this driver supports */
 	u8 atr_sample_rate;
 	bool wol_en;
 
 	struct hlist_head fdir_filter_list;
 	u16 fdir_pf_active_filters;
+	u16 fd_sb_cnt_idx;
+	u16 fd_atr_cnt_idx;
 
 #ifdef CONFIG_I40E_VXLAN
 	__be16  vxlan_ports[I40E_MAX_PF_UDP_OFFLOAD_PORTS];
@@ -266,6 +277,7 @@ struct i40e_pf {
 #ifdef CONFIG_I40E_VXLAN
 #define I40E_FLAG_VXLAN_FILTER_SYNC            (u64)(1 << 27)
 #endif
+#define I40E_FLAG_DCB_CAPABLE                  (u64)(1 << 29)
 
 	/* tracks features that get auto disabled by errors */
 	u64 auto_disable_flags;
@@ -300,7 +312,6 @@ struct i40e_pf {
 	u16 pf_seid;
 	u16 main_vsi_seid;
 	u16 mac_seid;
-	struct i40e_aqc_get_switch_config_data *sw_config;
 	struct kobject *switch_kobj;
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *i40e_dbg_pf;
@@ -329,9 +340,7 @@ struct i40e_pf {
 	struct ptp_clock *ptp_clock;
 	struct ptp_clock_info ptp_caps;
 	struct sk_buff *ptp_tx_skb;
-	struct work_struct ptp_tx_work;
 	struct hwtstamp_config tstamp_config;
-	unsigned long ptp_tx_start;
 	unsigned long last_rx_ptp_check;
 	spinlock_t tmreg_lock; /* Used to protect the device time registers. */
 	u64 ptp_base_adj;
@@ -420,6 +429,7 @@ struct i40e_vsi {
 	struct i40e_q_vector **q_vectors;
 	int num_q_vectors;
 	int base_vector;
+	bool irqs_ready;
 
 	u16 seid;            /* HW index of this VSI (absolute index) */
 	u16 id;              /* VSI number */
@@ -540,6 +550,15 @@ static inline bool i40e_rx_is_programming_status(u64 qw)
 		(qw >> I40E_RX_PROG_STATUS_DESC_LENGTH_SHIFT);
 }
 
+/**
+ * i40e_get_fd_cnt_all - get the total FD filter space available
+ * @pf: pointer to the pf struct
+ **/
+static inline int i40e_get_fd_cnt_all(struct i40e_pf *pf)
+{
+	return pf->hw.fdir_shared_filter_count + pf->fdir_pf_filter_count;
+}
+
 /* needed by i40e_ethtool.c */
 int i40e_up(struct i40e_vsi *vsi);
 void i40e_down(struct i40e_vsi *vsi);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c
index ed3902bf249b..7a027499fc57 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c
@@ -33,6 +33,16 @@
 static void i40e_resume_aq(struct i40e_hw *hw);
 
 /**
+ * i40e_is_nvm_update_op - return true if this is an NVM update operation
+ * @desc: API request descriptor
+ **/
+static inline bool i40e_is_nvm_update_op(struct i40e_aq_desc *desc)
+{
+	return (desc->opcode == i40e_aqc_opc_nvm_erase) ||
+	       (desc->opcode == i40e_aqc_opc_nvm_update);
+}
+
+/**
  *  i40e_adminq_init_regs - Initialize AdminQ registers
  *  @hw: pointer to the hardware structure
  *
@@ -281,8 +291,11 @@ static void i40e_free_asq_bufs(struct i40e_hw *hw)
  *
  *  Configure base address and length registers for the transmit queue
  **/
-static void i40e_config_asq_regs(struct i40e_hw *hw)
+static i40e_status i40e_config_asq_regs(struct i40e_hw *hw)
 {
+	i40e_status ret_code = 0;
+	u32 reg = 0;
+
 	if (hw->mac.type == I40E_MAC_VF) {
 		/* configure the transmit queue */
 		wr32(hw, I40E_VF_ATQBAH1,
@@ -291,6 +304,7 @@ static void i40e_config_asq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.asq.desc_buf.pa));
 		wr32(hw, I40E_VF_ATQLEN1, (hw->aq.num_asq_entries |
 					  I40E_VF_ATQLEN1_ATQENABLE_MASK));
+		reg = rd32(hw, I40E_VF_ATQBAL1);
 	} else {
 		/* configure the transmit queue */
 		wr32(hw, I40E_PF_ATQBAH,
@@ -299,7 +313,14 @@ static void i40e_config_asq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.asq.desc_buf.pa));
 		wr32(hw, I40E_PF_ATQLEN, (hw->aq.num_asq_entries |
 					  I40E_PF_ATQLEN_ATQENABLE_MASK));
+		reg = rd32(hw, I40E_PF_ATQBAL);
 	}
+
+	/* Check one register to verify that config was applied */
+	if (reg != lower_32_bits(hw->aq.asq.desc_buf.pa))
+		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+
+	return ret_code;
 }
 
 /**
@@ -308,8 +329,11 @@ static void i40e_config_asq_regs(struct i40e_hw *hw)
  *
  * Configure base address and length registers for the receive (event queue)
  **/
-static void i40e_config_arq_regs(struct i40e_hw *hw)
+static i40e_status i40e_config_arq_regs(struct i40e_hw *hw)
 {
+	i40e_status ret_code = 0;
+	u32 reg = 0;
+
 	if (hw->mac.type == I40E_MAC_VF) {
 		/* configure the receive queue */
 		wr32(hw, I40E_VF_ARQBAH1,
@@ -318,6 +342,7 @@ static void i40e_config_arq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.arq.desc_buf.pa));
 		wr32(hw, I40E_VF_ARQLEN1, (hw->aq.num_arq_entries |
 					  I40E_VF_ARQLEN1_ARQENABLE_MASK));
+		reg = rd32(hw, I40E_VF_ARQBAL1);
 	} else {
 		/* configure the receive queue */
 		wr32(hw, I40E_PF_ARQBAH,
@@ -326,10 +351,17 @@ static void i40e_config_arq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.arq.desc_buf.pa));
 		wr32(hw, I40E_PF_ARQLEN, (hw->aq.num_arq_entries |
 					  I40E_PF_ARQLEN_ARQENABLE_MASK));
+		reg = rd32(hw, I40E_PF_ARQBAL);
 	}
 
 	/* Update tail in the HW to post pre-allocated buffers */
 	wr32(hw, hw->aq.arq.tail, hw->aq.num_arq_entries - 1);
+
+	/* Check one register to verify that config was applied */
+	if (reg != lower_32_bits(hw->aq.arq.desc_buf.pa))
+		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+
+	return ret_code;
 }
 
 /**
@@ -377,7 +409,9 @@ static i40e_status i40e_init_asq(struct i40e_hw *hw)
 		goto init_adminq_free_rings;
 
 	/* initialize base registers */
-	i40e_config_asq_regs(hw);
+	ret_code = i40e_config_asq_regs(hw);
+	if (ret_code)
+		goto init_adminq_free_rings;
 
 	/* success! */
 	goto init_adminq_exit;
@@ -434,7 +468,9 @@ static i40e_status i40e_init_arq(struct i40e_hw *hw)
 		goto init_adminq_free_rings;
 
 	/* initialize base registers */
-	i40e_config_arq_regs(hw);
+	ret_code = i40e_config_arq_regs(hw);
+	if (ret_code)
+		goto init_adminq_free_rings;
 
 	/* success! */
 	goto init_adminq_exit;
@@ -577,14 +613,14 @@ i40e_status i40e_init_adminq(struct i40e_hw *hw)
 	i40e_read_nvm_word(hw, I40E_SR_NVM_EETRACK_HI, &eetrack_hi);
 	hw->nvm.eetrack = (eetrack_hi << 16) | eetrack_lo;
 
-	if (hw->aq.api_maj_ver != I40E_FW_API_VERSION_MAJOR ||
-	    hw->aq.api_min_ver > I40E_FW_API_VERSION_MINOR) {
+	if (hw->aq.api_maj_ver > I40E_FW_API_VERSION_MAJOR) {
 		ret_code = I40E_ERR_FIRMWARE_API_VERSION;
 		goto init_adminq_free_arq;
 	}
 
 	/* pre-emptive resource lock release */
 	i40e_aq_release_resource(hw, I40E_NVM_RESOURCE_ID, 0, NULL);
+	hw->aq.nvm_busy = false;
 
 	ret_code = i40e_aq_set_hmc_resource_profile(hw,
 						    I40E_HMC_PROFILE_DEFAULT,
@@ -708,6 +744,12 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 		goto asq_send_command_exit;
 	}
 
+	if (i40e_is_nvm_update_op(desc) && hw->aq.nvm_busy) {
+		i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, "AQTX: NVM busy.\n");
+		status = I40E_ERR_NVM;
+		goto asq_send_command_exit;
+	}
+
 	details = I40E_ADMINQ_DETAILS(hw->aq.asq, hw->aq.asq.next_to_use);
 	if (cmd_details) {
 		*details = *cmd_details;
@@ -835,6 +877,9 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw,
 		hw->aq.asq_last_status = (enum i40e_admin_queue_err)retval;
 	}
 
+	if (i40e_is_nvm_update_op(desc))
+		hw->aq.nvm_busy = true;
+
 	/* update the error if time out occurred */
 	if ((!cmd_completed) &&
 	    (!details->async && !details->postpone)) {
@@ -929,6 +974,9 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 			       e->msg_size);
 	}
 
+	if (i40e_is_nvm_update_op(&e->desc))
+		hw->aq.nvm_busy = false;
+
 	/* Restore the original datalen and buffer address in the desc,
 	 * FW updates datalen to indicate the event message
 	 * size
diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.h b/drivers/net/ethernet/intel/i40e/i40e_adminq.h
index 993f7685a911..b1552fbc48a0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.h
@@ -90,6 +90,7 @@ struct i40e_adminq_info {
 	u16 fw_min_ver;                 /* firmware minor version */
 	u16 api_maj_ver;                /* api major version */
 	u16 api_min_ver;                /* api minor version */
+	bool nvm_busy;
 
 	struct mutex asq_mutex; /* Send queue lock */
 	struct mutex arq_mutex; /* Receive queue lock */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 7b6374a8f8da..15f289f2917f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -34,7 +34,7 @@
  */
 
 #define I40E_FW_API_VERSION_MAJOR  0x0001
-#define I40E_FW_API_VERSION_MINOR  0x0001
+#define I40E_FW_API_VERSION_MINOR  0x0002
 
 struct i40e_aq_desc {
 	__le16 flags;
@@ -123,6 +123,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_version      = 0x0001,
 	i40e_aqc_opc_driver_version   = 0x0002,
 	i40e_aqc_opc_queue_shutdown   = 0x0003,
+	i40e_aqc_opc_set_pf_context   = 0x0004,
 
 	/* resource ownership */
 	i40e_aqc_opc_request_resource = 0x0008,
@@ -182,9 +183,6 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_add_mirror_rule    = 0x0260,
 	i40e_aqc_opc_delete_mirror_rule = 0x0261,
 
-	i40e_aqc_opc_set_storm_control_config = 0x0280,
-	i40e_aqc_opc_get_storm_control_config = 0x0281,
-
 	/* DCB commands */
 	i40e_aqc_opc_dcb_ignore_pfc = 0x0301,
 	i40e_aqc_opc_dcb_updated    = 0x0302,
@@ -207,6 +205,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_query_switching_comp_bw_config        = 0x041A,
 	i40e_aqc_opc_suspend_port_tx                       = 0x041B,
 	i40e_aqc_opc_resume_port_tx                        = 0x041C,
+	i40e_aqc_opc_configure_partition_bw                = 0x041D,
 
 	/* hmc */
 	i40e_aqc_opc_query_hmc_resource_profile = 0x0500,
@@ -224,13 +223,15 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_partner_advt    = 0x0616,
 	i40e_aqc_opc_set_lb_modes        = 0x0618,
 	i40e_aqc_opc_get_phy_wol_caps    = 0x0621,
-	i40e_aqc_opc_set_phy_reset       = 0x0622,
+	i40e_aqc_opc_set_phy_debug	 = 0x0622,
 	i40e_aqc_opc_upload_ext_phy_fm   = 0x0625,
 
 	/* NVM commands */
-	i40e_aqc_opc_nvm_read   = 0x0701,
-	i40e_aqc_opc_nvm_erase  = 0x0702,
-	i40e_aqc_opc_nvm_update = 0x0703,
+	i40e_aqc_opc_nvm_read         = 0x0701,
+	i40e_aqc_opc_nvm_erase        = 0x0702,
+	i40e_aqc_opc_nvm_update       = 0x0703,
+	i40e_aqc_opc_nvm_config_read  = 0x0704,
+	i40e_aqc_opc_nvm_config_write = 0x0705,
 
 	/* virtualization commands */
 	i40e_aqc_opc_send_msg_to_pf   = 0x0801,
@@ -272,8 +273,6 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_debug_set_mode         = 0xFF01,
 	i40e_aqc_opc_debug_read_reg         = 0xFF03,
 	i40e_aqc_opc_debug_write_reg        = 0xFF04,
-	i40e_aqc_opc_debug_read_reg_sg      = 0xFF05,
-	i40e_aqc_opc_debug_write_reg_sg     = 0xFF06,
 	i40e_aqc_opc_debug_modify_reg       = 0xFF07,
 	i40e_aqc_opc_debug_dump_internals   = 0xFF08,
 	i40e_aqc_opc_debug_modify_internals = 0xFF09,
@@ -341,6 +340,14 @@ struct i40e_aqc_queue_shutdown {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_queue_shutdown);
 
+/* Set PF context (0x0004, direct) */
+struct i40e_aqc_set_pf_context {
+	u8	pf_id;
+	u8	reserved[15];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_pf_context);
+
 /* Request resource ownership (direct 0x0008)
  * Release resource ownership (direct 0x0009)
  */
@@ -1289,27 +1296,6 @@ struct i40e_aqc_add_delete_mirror_rule_completion {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_add_delete_mirror_rule_completion);
 
-/* Set Storm Control Configuration (direct 0x0280)
- * Get Storm Control Configuration (direct 0x0281)
- *    the command and response use the same descriptor structure
- */
-struct i40e_aqc_set_get_storm_control_config {
-	__le32 broadcast_threshold;
-	__le32 multicast_threshold;
-	__le32 control_flags;
-#define I40E_AQC_STORM_CONTROL_MDIPW            0x01
-#define I40E_AQC_STORM_CONTROL_MDICW            0x02
-#define I40E_AQC_STORM_CONTROL_BDIPW            0x04
-#define I40E_AQC_STORM_CONTROL_BDICW            0x08
-#define I40E_AQC_STORM_CONTROL_BIDU             0x10
-#define I40E_AQC_STORM_CONTROL_INTERVAL_SHIFT   8
-#define I40E_AQC_STORM_CONTROL_INTERVAL_MASK    (0x3FF << \
-					I40E_AQC_STORM_CONTROL_INTERVAL_SHIFT)
-	u8     reserved[4];
-};
-
-I40E_CHECK_CMD_LENGTH(i40e_aqc_set_get_storm_control_config);
-
 /* DCB 0x03xx*/
 
 /* PFC Ignore (direct 0x0301)
@@ -1427,11 +1413,12 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_configure_switching_comp_bw_limit);
 struct i40e_aqc_configure_switching_comp_ets_data {
 	u8     reserved[4];
 	u8     tc_valid_bits;
-	u8     reserved1;
+	u8     seepage;
+#define I40E_AQ_ETS_SEEPAGE_EN_MASK     0x1
 	u8     tc_strict_priority_flags;
-	u8     reserved2[17];
+	u8     reserved1[17];
 	u8     tc_bw_share_credits[8];
-	u8     reserved3[96];
+	u8     reserved2[96];
 };
 
 /* Configure Switching Component Bandwidth Limits per Tc (indirect 0x0416) */
@@ -1499,6 +1486,15 @@ struct i40e_aqc_query_switching_comp_bw_config_resp {
  * (direct 0x041B and 0x041C) uses the generic SEID struct
  */
 
+/* Configure partition BW
+ * (indirect 0x041D)
+ */
+struct i40e_aqc_configure_partition_bw_data {
+	__le16 pf_valid_bits;
+	u8     min_bw[16];      /* guaranteed bandwidth */
+	u8     max_bw[16];      /* bandwidth limit */
+};
+
 /* Get and set the active HMC resource profile and status.
  * (direct 0x0500) and (direct 0x0501)
  */
@@ -1539,6 +1535,8 @@ enum i40e_aq_phy_type {
 	I40E_PHY_TYPE_XLPPI			= 0x9,
 	I40E_PHY_TYPE_40GBASE_CR4_CU		= 0xA,
 	I40E_PHY_TYPE_10GBASE_CR1_CU		= 0xB,
+	I40E_PHY_TYPE_10GBASE_AOC		= 0xC,
+	I40E_PHY_TYPE_40GBASE_AOC		= 0xD,
 	I40E_PHY_TYPE_100BASE_TX		= 0x11,
 	I40E_PHY_TYPE_1000BASE_T		= 0x12,
 	I40E_PHY_TYPE_10GBASE_T			= 0x13,
@@ -1549,7 +1547,10 @@ enum i40e_aq_phy_type {
 	I40E_PHY_TYPE_40GBASE_CR4		= 0x18,
 	I40E_PHY_TYPE_40GBASE_SR4		= 0x19,
 	I40E_PHY_TYPE_40GBASE_LR4		= 0x1A,
-	I40E_PHY_TYPE_20GBASE_KR2		= 0x1B,
+	I40E_PHY_TYPE_1000BASE_SX		= 0x1B,
+	I40E_PHY_TYPE_1000BASE_LX		= 0x1C,
+	I40E_PHY_TYPE_1000BASE_T_OPTICAL	= 0x1D,
+	I40E_PHY_TYPE_20GBASE_KR2		= 0x1E,
 	I40E_PHY_TYPE_MAX
 };
 
@@ -1583,11 +1584,8 @@ struct i40e_aq_get_phy_abilities_resp {
 #define I40E_AQ_PHY_FLAG_PAUSE_TX         0x01
 #define I40E_AQ_PHY_FLAG_PAUSE_RX         0x02
 #define I40E_AQ_PHY_FLAG_LOW_POWER        0x04
-#define I40E_AQ_PHY_FLAG_AN_SHIFT         3
-#define I40E_AQ_PHY_FLAG_AN_MASK          (0x3 << I40E_AQ_PHY_FLAG_AN_SHIFT)
-#define I40E_AQ_PHY_FLAG_AN_OFF           0x00 /* link forced on */
-#define I40E_AQ_PHY_FLAG_AN_OFF_LINK_DOWN 0x01
-#define I40E_AQ_PHY_FLAG_AN_ON            0x02
+#define I40E_AQ_PHY_LINK_ENABLED		  0x08
+#define I40E_AQ_PHY_AN_ENABLED			  0x10
 #define I40E_AQ_PHY_FLAG_MODULE_QUAL      0x20
 	__le16 eee_capability;
 #define I40E_AQ_EEE_100BASE_TX       0x0002
@@ -1696,6 +1694,7 @@ struct i40e_aqc_get_link_status {
 #define I40E_AQ_LINK_TX_ACTIVE       0x00
 #define I40E_AQ_LINK_TX_DRAINED      0x01
 #define I40E_AQ_LINK_TX_FLUSHED      0x03
+#define I40E_AQ_LINK_FORCED_40G      0x10
 	u8     loopback;         /* use defines from i40e_aqc_set_lb_mode */
 	__le16 max_frame_size;
 	u8     config;
@@ -1747,14 +1746,21 @@ struct i40e_aqc_set_lb_mode {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_lb_mode);
 
-/* Set PHY Reset command (0x0622) */
-struct i40e_aqc_set_phy_reset {
-	u8     reset_flags;
-#define I40E_AQ_PHY_RESET_REQUEST  0x02
+/* Set PHY Debug command (0x0622) */
+struct i40e_aqc_set_phy_debug {
+	u8     command_flags;
+#define I40E_AQ_PHY_DEBUG_RESET_INTERNAL	0x02
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SHIFT	2
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_MASK	(0x03 << \
+					I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SHIFT)
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_NONE	0x00
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_HARD	0x01
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SOFT	0x02
+#define I40E_AQ_PHY_DEBUG_DISABLE_LINK_FW	0x10
 	u8     reserved[15];
 };
 
-I40E_CHECK_CMD_LENGTH(i40e_aqc_set_phy_reset);
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_phy_debug);
 
 enum i40e_aq_phy_reg_type {
 	I40E_AQC_PHY_REG_INTERNAL         = 0x1,
@@ -1779,6 +1785,47 @@ struct i40e_aqc_nvm_update {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_update);
 
+/* NVM Config Read (indirect 0x0704) */
+struct i40e_aqc_nvm_config_read {
+	__le16 cmd_flags;
+#define ANVM_SINGLE_OR_MULTIPLE_FEATURES_MASK	1
+#define ANVM_READ_SINGLE_FEATURE		0
+#define ANVM_READ_MULTIPLE_FEATURES		1
+	__le16 element_count;
+	__le16 element_id;		/* Feature/field ID */
+	u8     reserved[2];
+	__le32 address_high;
+	__le32 address_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_config_read);
+
+/* NVM Config Write (indirect 0x0705) */
+struct i40e_aqc_nvm_config_write {
+	__le16 cmd_flags;
+	__le16 element_count;
+	u8     reserved[4];
+	__le32 address_high;
+	__le32 address_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_config_write);
+
+struct i40e_aqc_nvm_config_data_feature {
+	__le16 feature_id;
+	__le16 instance_id;
+	__le16 feature_options;
+	__le16 feature_selection;
+};
+
+struct i40e_aqc_nvm_config_data_immediate_field {
+#define ANVM_FEATURE_OR_IMMEDIATE_MASK	0x2
+	__le16 field_id;
+	__le16 instance_id;
+	__le16 field_options;
+	__le16 field_value;
+};
+
 /* Send to PF command (indirect 0x0801) id is only used by PF
  * Send to VF command (indirect 0x0802) id is only used by PF
  * Send to Peer PF command (indirect 0x0803)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 922cdcc45c54..6e65f19dd6e5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -43,12 +43,10 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 	if (hw->vendor_id == PCI_VENDOR_ID_INTEL) {
 		switch (hw->device_id) {
 		case I40E_DEV_ID_SFP_XL710:
-		case I40E_DEV_ID_SFP_X710:
 		case I40E_DEV_ID_QEMU:
 		case I40E_DEV_ID_KX_A:
 		case I40E_DEV_ID_KX_B:
 		case I40E_DEV_ID_KX_C:
-		case I40E_DEV_ID_KX_D:
 		case I40E_DEV_ID_QSFP_A:
 		case I40E_DEV_ID_QSFP_B:
 		case I40E_DEV_ID_QSFP_C:
@@ -133,7 +131,11 @@ void i40e_debug_aq(struct i40e_hw *hw, enum i40e_debug_mask mask, void *desc,
  **/
 bool i40e_check_asq_alive(struct i40e_hw *hw)
 {
-	return !!(rd32(hw, hw->aq.asq.len) & I40E_PF_ATQLEN_ATQENABLE_MASK);
+	if (hw->aq.asq.len)
+		return !!(rd32(hw, hw->aq.asq.len) &
+			  I40E_PF_ATQLEN_ATQENABLE_MASK);
+	else
+		return false;
 }
 
 /**
@@ -653,6 +655,36 @@ i40e_status i40e_get_mac_addr(struct i40e_hw *hw, u8 *mac_addr)
 }
 
 /**
+ * i40e_pre_tx_queue_cfg - pre tx queue configure
+ * @hw: pointer to the HW structure
+ * @queue: target pf queue index
+ * @enable: state change request
+ *
+ * Handles hw requirement to indicate intention to enable
+ * or disable target queue.
+ **/
+void i40e_pre_tx_queue_cfg(struct i40e_hw *hw, u32 queue, bool enable)
+{
+	u32 abs_queue_idx = hw->func_caps.base_queue + queue;
+	u32 reg_block = 0;
+	u32 reg_val;
+
+	if (abs_queue_idx >= 128)
+		reg_block = abs_queue_idx / 128;
+
+	reg_val = rd32(hw, I40E_GLLAN_TXPRE_QDIS(reg_block));
+	reg_val &= ~I40E_GLLAN_TXPRE_QDIS_QINDX_MASK;
+	reg_val |= (abs_queue_idx << I40E_GLLAN_TXPRE_QDIS_QINDX_SHIFT);
+
+	if (enable)
+		reg_val |= I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_MASK;
+	else
+		reg_val |= I40E_GLLAN_TXPRE_QDIS_SET_QDIS_MASK;
+
+	wr32(hw, I40E_GLLAN_TXPRE_QDIS(reg_block), reg_val);
+}
+
+/**
  * i40e_get_media_type - Gets media type
  * @hw: pointer to the hardware structure
  **/
@@ -699,7 +731,7 @@ static enum i40e_media_type i40e_get_media_type(struct i40e_hw *hw)
 }
 
 #define I40E_PF_RESET_WAIT_COUNT_A0	200
-#define I40E_PF_RESET_WAIT_COUNT	10
+#define I40E_PF_RESET_WAIT_COUNT	100
 /**
  * i40e_pf_reset - Reset the PF
  * @hw: pointer to the hardware structure
@@ -789,6 +821,9 @@ void i40e_clear_pxe_mode(struct i40e_hw *hw)
 {
 	u32 reg;
 
+	if (i40e_check_asq_alive(hw))
+		i40e_aq_clear_pxe_mode(hw, NULL);
+
 	/* Clear single descriptor fetch/write-back mode */
 	reg = rd32(hw, I40E_GLLAN_RCTL_0);
 
@@ -907,6 +942,33 @@ void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink)
 /* Admin command wrappers */
 
 /**
+ * i40e_aq_clear_pxe_mode
+ * @hw: pointer to the hw struct
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Tell the firmware that the driver is taking over from PXE
+ **/
+i40e_status i40e_aq_clear_pxe_mode(struct i40e_hw *hw,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	i40e_status status;
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_clear_pxe *cmd =
+		(struct i40e_aqc_clear_pxe *)&desc.params.raw;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_clear_pxe_mode);
+
+	cmd->rx_cnt = 0x2;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	wr32(hw, I40E_GLLAN_RCTL_0, 0x1);
+
+	return status;
+}
+
+/**
  * i40e_aq_set_link_restart_an
  * @hw: pointer to the hw struct
  * @cmd_details: pointer to command details structure or NULL
@@ -975,6 +1037,13 @@ i40e_status i40e_aq_get_link_info(struct i40e_hw *hw,
 	hw_link_info->an_info = resp->an_info;
 	hw_link_info->ext_info = resp->ext_info;
 	hw_link_info->loopback = resp->loopback;
+	hw_link_info->max_frame_size = le16_to_cpu(resp->max_frame_size);
+	hw_link_info->pacing = resp->config & I40E_AQ_CONFIG_PACING_MASK;
+
+	if (resp->config & I40E_AQ_CONFIG_CRC_ENA)
+		hw_link_info->crc_enable = true;
+	else
+		hw_link_info->crc_enable = false;
 
 	if (resp->command_flags & cpu_to_le16(I40E_AQ_LSE_ENABLE))
 		hw_link_info->lse_enable = true;
@@ -1021,8 +1090,6 @@ i40e_status i40e_aq_add_vsi(struct i40e_hw *hw,
 	cmd->vsi_flags = cpu_to_le16(vsi_ctx->flags);
 
 	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
-	if (sizeof(vsi_ctx->info) > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info,
 				    sizeof(vsi_ctx->info), cmd_details);
@@ -1163,8 +1230,6 @@ i40e_status i40e_aq_get_vsi_params(struct i40e_hw *hw,
 	cmd->uplink_seid = cpu_to_le16(vsi_ctx->seid);
 
 	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
-	if (sizeof(vsi_ctx->info) > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info,
 				    sizeof(vsi_ctx->info), NULL);
@@ -1203,8 +1268,6 @@ i40e_status i40e_aq_update_vsi_params(struct i40e_hw *hw,
 	cmd->uplink_seid = cpu_to_le16(vsi_ctx->seid);
 
 	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
-	if (sizeof(vsi_ctx->info) > I40E_AQ_LARGE_BUF)
-		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info,
 				    sizeof(vsi_ctx->info), cmd_details);
@@ -1300,6 +1363,7 @@ i40e_status i40e_aq_send_driver_version(struct i40e_hw *hw,
 	struct i40e_aqc_driver_version *cmd =
 		(struct i40e_aqc_driver_version *)&desc.params.raw;
 	i40e_status status;
+	u16 len;
 
 	if (dv == NULL)
 		return I40E_ERR_PARAM;
@@ -1311,7 +1375,14 @@ i40e_status i40e_aq_send_driver_version(struct i40e_hw *hw,
 	cmd->driver_minor_ver = dv->minor_version;
 	cmd->driver_build_ver = dv->build_version;
 	cmd->driver_subbuild_ver = dv->subbuild_version;
-	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	len = 0;
+	while (len < sizeof(dv->driver_string) &&
+	       (dv->driver_string[len] < 0x80) &&
+	       dv->driver_string[len])
+		len++;
+	status = i40e_asq_send_command(hw, &desc, dv->driver_string,
+				       len, cmd_details);
 
 	return status;
 }
@@ -1900,6 +1971,12 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 		}
 	}
 
+	/* Software override ensuring FCoE is disabled if npar or mfp
+	 * mode because it is not supported in these modes.
+	 */
+	if (p->npar_enable || p->mfp_mode_1)
+		p->fcoe = false;
+
 	/* additional HW specific goodies that might
 	 * someday be HW version specific
 	 */
@@ -2094,8 +2171,8 @@ i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
  * @cmd_details: pointer to command details structure or NULL
  **/
 i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
-				u16 udp_port, u8 header_len,
-				u8 protocol_index, u8 *filter_index,
+				u16 udp_port, u8 protocol_index,
+				u8 *filter_index,
 				struct i40e_asq_cmd_details *cmd_details)
 {
 	struct i40e_aq_desc desc;
@@ -2253,6 +2330,35 @@ static i40e_status i40e_aq_tx_sched_cmd(struct i40e_hw *hw, u16 seid,
 }
 
 /**
+ * i40e_aq_config_vsi_bw_limit - Configure VSI BW Limit
+ * @hw: pointer to the hw struct
+ * @seid: VSI seid
+ * @credit: BW limit credits (0 = disabled)
+ * @max_credit: Max BW limit credits
+ * @cmd_details: pointer to command details structure or NULL
+ **/
+i40e_status i40e_aq_config_vsi_bw_limit(struct i40e_hw *hw,
+				u16 seid, u16 credit, u8 max_credit,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_configure_vsi_bw_limit *cmd =
+		(struct i40e_aqc_configure_vsi_bw_limit *)&desc.params.raw;
+	i40e_status status;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_configure_vsi_bw_limit);
+
+	cmd->vsi_seid = cpu_to_le16(seid);
+	cmd->credit = cpu_to_le16(credit);
+	cmd->max_credit = max_credit;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
  * i40e_aq_config_vsi_tc_bw - Config VSI BW Allocation per TC
  * @hw: pointer to the hw struct
  * @seid: VSI seid
@@ -2405,7 +2511,7 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
 {
 	u32 fcoe_cntx_size, fcoe_filt_size;
 	u32 pe_cntx_size, pe_filt_size;
-	u32 fcoe_fmax, pe_fmax;
+	u32 fcoe_fmax;
 	u32 val;
 
 	/* Validate FCoE settings passed */
@@ -2480,13 +2586,6 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw,
 	if (fcoe_filt_size + fcoe_cntx_size >  fcoe_fmax)
 		return I40E_ERR_INVALID_SIZE;
 
-	/* PEHSIZE + PEDSIZE should not be greater than PMPEXFMAX */
-	val = rd32(hw, I40E_GLHMC_PEXFMAX);
-	pe_fmax = (val & I40E_GLHMC_PEXFMAX_PMPEXFMAX_MASK)
-		   >> I40E_GLHMC_PEXFMAX_PMPEXFMAX_SHIFT;
-	if (pe_filt_size + pe_cntx_size >  pe_fmax)
-		return I40E_ERR_INVALID_SIZE;
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c b/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
index 6e8103abfd0d..00bc0cdb3a03 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
@@ -232,7 +232,7 @@ static void i40e_dcbnl_del_app(struct i40e_pf *pf,
 			      struct i40e_ieee_app_priority_table *app)
 {
 	int v, err;
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v] && pf->vsi[v]->netdev) {
 			err = i40e_dcbnl_vsi_del_app(pf->vsi[v], app);
 			if (err)
@@ -302,8 +302,8 @@ void i40e_dcbnl_setup(struct i40e_vsi *vsi)
 	struct net_device *dev = vsi->netdev;
 	struct i40e_pf *pf = i40e_netdev_to_pf(dev);
 
-	/* DCB not enabled */
-	if (!(pf->flags & I40E_FLAG_DCB_ENABLED))
+	/* Not DCB capable */
+	if (!(pf->flags & I40E_FLAG_DCB_CAPABLE))
 		return;
 
 	/* Do not setup DCB NL ops for MFP mode */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 3c37386fd138..cffdfc21290f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -45,7 +45,7 @@ static struct i40e_vsi *i40e_dbg_find_vsi(struct i40e_pf *pf, int seid)
 	if (seid < 0)
 		dev_info(&pf->pdev->dev, "%d: bad seid\n", seid);
 	else
-		for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+		for (i = 0; i < pf->num_alloc_vsi; i++)
 			if (pf->vsi[i] && (pf->vsi[i]->seid == seid))
 				return pf->vsi[i];
 
@@ -843,7 +843,7 @@ static void i40e_dbg_dump_vsi_no_seid(struct i40e_pf *pf)
 {
 	int i;
 
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i])
 			dev_info(&pf->pdev->dev, "dump vsi[%d]: %d\n",
 				 i, pf->vsi[i]->seid);
@@ -862,12 +862,11 @@ static void i40e_dbg_dump_eth_stats(struct i40e_pf *pf,
 		 "    rx_bytes = \t%lld \trx_unicast = \t\t%lld \trx_multicast = \t%lld\n",
 		estats->rx_bytes, estats->rx_unicast, estats->rx_multicast);
 	dev_info(&pf->pdev->dev,
-		 "    rx_broadcast = \t%lld \trx_discards = \t\t%lld \trx_errors = \t%lld\n",
-		 estats->rx_broadcast, estats->rx_discards, estats->rx_errors);
+		 "    rx_broadcast = \t%lld \trx_discards = \t\t%lld\n",
+		 estats->rx_broadcast, estats->rx_discards);
 	dev_info(&pf->pdev->dev,
-		 "    rx_missed = \t%lld \trx_unknown_protocol = \t%lld \ttx_bytes = \t%lld\n",
-		 estats->rx_missed, estats->rx_unknown_protocol,
-		 estats->tx_bytes);
+		 "    rx_unknown_protocol = \t%lld \ttx_bytes = \t%lld\n",
+		 estats->rx_unknown_protocol, estats->tx_bytes);
 	dev_info(&pf->pdev->dev,
 		 "    tx_unicast = \t%lld \ttx_multicast = \t\t%lld \ttx_broadcast = \t%lld\n",
 		 estats->tx_unicast, estats->tx_multicast, estats->tx_broadcast);
@@ -1527,7 +1526,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			cnt = sscanf(&cmd_buf[15], "%i", &vsi_seid);
 			if (cnt == 0) {
 				int i;
-				for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+				for (i = 0; i < pf->num_alloc_vsi; i++)
 					i40e_vsi_reset_stats(pf->vsi[i]);
 				dev_info(&pf->pdev->dev, "vsi clear stats called for all vsi's\n");
 			} else if (cnt == 1) {
@@ -1744,10 +1743,6 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		i40e_dbg_cmd_fd_ctrl(pf, I40E_FLAG_FD_ATR_ENABLED, false);
 	} else if (strncmp(cmd_buf, "fd-atr on", 9) == 0) {
 		i40e_dbg_cmd_fd_ctrl(pf, I40E_FLAG_FD_ATR_ENABLED, true);
-	} else if (strncmp(cmd_buf, "fd-sb off", 9) == 0) {
-		i40e_dbg_cmd_fd_ctrl(pf, I40E_FLAG_FD_SB_ENABLED, false);
-	} else if (strncmp(cmd_buf, "fd-sb on", 8) == 0) {
-		i40e_dbg_cmd_fd_ctrl(pf, I40E_FLAG_FD_SB_ENABLED, true);
 	} else if (strncmp(cmd_buf, "lldp", 4) == 0) {
 		if (strncmp(&cmd_buf[5], "stop", 4) == 0) {
 			int ret;
@@ -1967,8 +1962,6 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 		dev_info(&pf->pdev->dev, "  rem fd_filter <dest q_index> <flex_off> <pctype> <dest_vsi> <dest_ctl> <fd_status> <cnt_index> <fd_id> <packet_len> <packet>\n");
 		dev_info(&pf->pdev->dev, "  fd-atr off\n");
 		dev_info(&pf->pdev->dev, "  fd-atr on\n");
-		dev_info(&pf->pdev->dev, "  fd-sb off\n");
-		dev_info(&pf->pdev->dev, "  fd-sb on\n");
 		dev_info(&pf->pdev->dev, "  lldp start\n");
 		dev_info(&pf->pdev->dev, "  lldp stop\n");
 		dev_info(&pf->pdev->dev, "  lldp get local\n");
diff --git a/drivers/net/ethernet/intel/i40e/i40e_diag.c b/drivers/net/ethernet/intel/i40e/i40e_diag.c
index b2380daef8c1..56438bd579e6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_diag.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_diag.c
@@ -67,17 +67,25 @@ static i40e_status i40e_diag_reg_pattern_test(struct i40e_hw *hw,
 
 struct i40e_diag_reg_test_info i40e_reg_list[] = {
 	/* offset               mask         elements   stride */
-	{I40E_QTX_CTL(0),       0x0000FFBF,   4, I40E_QTX_CTL(1) - I40E_QTX_CTL(0)},
-	{I40E_PFINT_ITR0(0),    0x00000FFF,   3, I40E_PFINT_ITR0(1) - I40E_PFINT_ITR0(0)},
-	{I40E_PFINT_ITRN(0, 0), 0x00000FFF,   8, I40E_PFINT_ITRN(0, 1) - I40E_PFINT_ITRN(0, 0)},
-	{I40E_PFINT_ITRN(1, 0), 0x00000FFF,   8, I40E_PFINT_ITRN(1, 1) - I40E_PFINT_ITRN(1, 0)},
-	{I40E_PFINT_ITRN(2, 0), 0x00000FFF,   8, I40E_PFINT_ITRN(2, 1) - I40E_PFINT_ITRN(2, 0)},
-	{I40E_PFINT_STAT_CTL0,  0x0000000C,   1, 0},
-	{I40E_PFINT_LNKLST0,    0x00001FFF,   1, 0},
-	{I40E_PFINT_LNKLSTN(0), 0x000007FF,  64, I40E_PFINT_LNKLSTN(1) - I40E_PFINT_LNKLSTN(0)},
-	{I40E_QINT_TQCTL(0),    0x000000FF,  64, I40E_QINT_TQCTL(1) - I40E_QINT_TQCTL(0)},
-	{I40E_QINT_RQCTL(0),    0x000000FF,  64, I40E_QINT_RQCTL(1) - I40E_QINT_RQCTL(0)},
-	{I40E_PFINT_ICR0_ENA,   0xF7F20000,   1, 0},
+	{I40E_QTX_CTL(0),       0x0000FFBF, 1,
+		I40E_QTX_CTL(1) - I40E_QTX_CTL(0)},
+	{I40E_PFINT_ITR0(0),    0x00000FFF, 3,
+		I40E_PFINT_ITR0(1) - I40E_PFINT_ITR0(0)},
+	{I40E_PFINT_ITRN(0, 0), 0x00000FFF, 1,
+		I40E_PFINT_ITRN(0, 1) - I40E_PFINT_ITRN(0, 0)},
+	{I40E_PFINT_ITRN(1, 0), 0x00000FFF, 1,
+		I40E_PFINT_ITRN(1, 1) - I40E_PFINT_ITRN(1, 0)},
+	{I40E_PFINT_ITRN(2, 0), 0x00000FFF, 1,
+		I40E_PFINT_ITRN(2, 1) - I40E_PFINT_ITRN(2, 0)},
+	{I40E_PFINT_STAT_CTL0,  0x0000000C, 1, 0},
+	{I40E_PFINT_LNKLST0,    0x00001FFF, 1, 0},
+	{I40E_PFINT_LNKLSTN(0), 0x000007FF, 1,
+		I40E_PFINT_LNKLSTN(1) - I40E_PFINT_LNKLSTN(0)},
+	{I40E_QINT_TQCTL(0),    0x000000FF, 1,
+		I40E_QINT_TQCTL(1) - I40E_QINT_TQCTL(0)},
+	{I40E_QINT_RQCTL(0),    0x000000FF, 1,
+		I40E_QINT_RQCTL(1) - I40E_QINT_RQCTL(0)},
+	{I40E_PFINT_ICR0_ENA,   0xF7F20000, 1, 0},
 	{ 0 }
 };
 
@@ -93,9 +101,25 @@ i40e_status i40e_diag_reg_test(struct i40e_hw *hw)
 	u32 reg, mask;
 	u32 i, j;
 
-	for (i = 0; (i40e_reg_list[i].offset != 0) && !ret_code; i++) {
+	for (i = 0; i40e_reg_list[i].offset != 0 &&
+					     !ret_code; i++) {
+
+		/* set actual reg range for dynamically allocated resources */
+		if (i40e_reg_list[i].offset == I40E_QTX_CTL(0) &&
+		    hw->func_caps.num_tx_qp != 0)
+			i40e_reg_list[i].elements = hw->func_caps.num_tx_qp;
+		if ((i40e_reg_list[i].offset == I40E_PFINT_ITRN(0, 0) ||
+		     i40e_reg_list[i].offset == I40E_PFINT_ITRN(1, 0) ||
+		     i40e_reg_list[i].offset == I40E_PFINT_ITRN(2, 0) ||
+		     i40e_reg_list[i].offset == I40E_QINT_TQCTL(0) ||
+		     i40e_reg_list[i].offset == I40E_QINT_RQCTL(0)) &&
+		    hw->func_caps.num_msix_vectors != 0)
+			i40e_reg_list[i].elements =
+				hw->func_caps.num_msix_vectors - 1;
+
+		/* test register access */
 		mask = i40e_reg_list[i].mask;
-		for (j = 0; (j < i40e_reg_list[i].elements) && !ret_code; j++) {
+		for (j = 0; j < i40e_reg_list[i].elements && !ret_code; j++) {
 			reg = i40e_reg_list[i].offset +
 			      (j * i40e_reg_list[i].stride);
 			ret_code = i40e_diag_reg_pattern_test(hw, reg, mask);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 03d99cbc5c25..4a488ffcd6b0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -46,6 +46,8 @@ struct i40e_stats {
 		I40E_STAT(struct i40e_pf, _name, _stat)
 #define I40E_VSI_STAT(_name, _stat) \
 		I40E_STAT(struct i40e_vsi, _name, _stat)
+#define I40E_VEB_STAT(_name, _stat) \
+		I40E_STAT(struct i40e_veb, _name, _stat)
 
 static const struct i40e_stats i40e_gstrings_net_stats[] = {
 	I40E_NETDEV_STAT(rx_packets),
@@ -56,12 +58,36 @@ static const struct i40e_stats i40e_gstrings_net_stats[] = {
 	I40E_NETDEV_STAT(tx_errors),
 	I40E_NETDEV_STAT(rx_dropped),
 	I40E_NETDEV_STAT(tx_dropped),
-	I40E_NETDEV_STAT(multicast),
 	I40E_NETDEV_STAT(collisions),
 	I40E_NETDEV_STAT(rx_length_errors),
 	I40E_NETDEV_STAT(rx_crc_errors),
 };
 
+static const struct i40e_stats i40e_gstrings_veb_stats[] = {
+	I40E_VEB_STAT("rx_bytes", stats.rx_bytes),
+	I40E_VEB_STAT("tx_bytes", stats.tx_bytes),
+	I40E_VEB_STAT("rx_unicast", stats.rx_unicast),
+	I40E_VEB_STAT("tx_unicast", stats.tx_unicast),
+	I40E_VEB_STAT("rx_multicast", stats.rx_multicast),
+	I40E_VEB_STAT("tx_multicast", stats.tx_multicast),
+	I40E_VEB_STAT("rx_broadcast", stats.rx_broadcast),
+	I40E_VEB_STAT("tx_broadcast", stats.tx_broadcast),
+	I40E_VEB_STAT("rx_discards", stats.rx_discards),
+	I40E_VEB_STAT("tx_discards", stats.tx_discards),
+	I40E_VEB_STAT("tx_errors", stats.tx_errors),
+	I40E_VEB_STAT("rx_unknown_protocol", stats.rx_unknown_protocol),
+};
+
+static const struct i40e_stats i40e_gstrings_misc_stats[] = {
+	I40E_VSI_STAT("rx_unicast", eth_stats.rx_unicast),
+	I40E_VSI_STAT("tx_unicast", eth_stats.tx_unicast),
+	I40E_VSI_STAT("rx_multicast", eth_stats.rx_multicast),
+	I40E_VSI_STAT("tx_multicast", eth_stats.tx_multicast),
+	I40E_VSI_STAT("rx_broadcast", eth_stats.rx_broadcast),
+	I40E_VSI_STAT("tx_broadcast", eth_stats.tx_broadcast),
+	I40E_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol),
+};
+
 static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 				 struct ethtool_rxnfc *cmd);
 
@@ -78,7 +104,12 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 static struct i40e_stats i40e_gstrings_stats[] = {
 	I40E_PF_STAT("rx_bytes", stats.eth.rx_bytes),
 	I40E_PF_STAT("tx_bytes", stats.eth.tx_bytes),
-	I40E_PF_STAT("rx_errors", stats.eth.rx_errors),
+	I40E_PF_STAT("rx_unicast", stats.eth.rx_unicast),
+	I40E_PF_STAT("tx_unicast", stats.eth.tx_unicast),
+	I40E_PF_STAT("rx_multicast", stats.eth.rx_multicast),
+	I40E_PF_STAT("tx_multicast", stats.eth.tx_multicast),
+	I40E_PF_STAT("rx_broadcast", stats.eth.rx_broadcast),
+	I40E_PF_STAT("tx_broadcast", stats.eth.tx_broadcast),
 	I40E_PF_STAT("tx_errors", stats.eth.tx_errors),
 	I40E_PF_STAT("rx_dropped", stats.eth.rx_discards),
 	I40E_PF_STAT("tx_dropped", stats.eth.tx_discards),
@@ -88,6 +119,7 @@ static struct i40e_stats i40e_gstrings_stats[] = {
 	I40E_PF_STAT("mac_local_faults", stats.mac_local_faults),
 	I40E_PF_STAT("mac_remote_faults", stats.mac_remote_faults),
 	I40E_PF_STAT("tx_timeout", tx_timeout_count),
+	I40E_PF_STAT("rx_csum_bad", hw_csum_rx_error),
 	I40E_PF_STAT("rx_length_errors", stats.rx_length_errors),
 	I40E_PF_STAT("link_xon_rx", stats.link_xon_rx),
 	I40E_PF_STAT("link_xoff_rx", stats.link_xoff_rx),
@@ -112,8 +144,10 @@ static struct i40e_stats i40e_gstrings_stats[] = {
 	I40E_PF_STAT("rx_oversize", stats.rx_oversize),
 	I40E_PF_STAT("rx_jabber", stats.rx_jabber),
 	I40E_PF_STAT("VF_admin_queue_requests", vf_aq_requests),
-	I40E_PF_STAT("tx_hwtstamp_timeouts", tx_hwtstamp_timeouts),
 	I40E_PF_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared),
+	I40E_PF_STAT("fdir_atr_match", stats.fd_atr_match),
+	I40E_PF_STAT("fdir_sb_match", stats.fd_sb_match),
+
 	/* LPI stats */
 	I40E_PF_STAT("tx_lpi_status", stats.tx_lpi_status),
 	I40E_PF_STAT("rx_lpi_status", stats.rx_lpi_status),
@@ -122,11 +156,14 @@ static struct i40e_stats i40e_gstrings_stats[] = {
 };
 
 #define I40E_QUEUE_STATS_LEN(n) \
-  ((((struct i40e_netdev_priv *)netdev_priv((n)))->vsi->num_queue_pairs + \
-    ((struct i40e_netdev_priv *)netdev_priv((n)))->vsi->num_queue_pairs) * 2)
+	(((struct i40e_netdev_priv *)netdev_priv((n)))->vsi->num_queue_pairs \
+	    * 2 /* Tx and Rx together */                                     \
+	    * (sizeof(struct i40e_queue_stats) / sizeof(u64)))
 #define I40E_GLOBAL_STATS_LEN	ARRAY_SIZE(i40e_gstrings_stats)
 #define I40E_NETDEV_STATS_LEN   ARRAY_SIZE(i40e_gstrings_net_stats)
+#define I40E_MISC_STATS_LEN	ARRAY_SIZE(i40e_gstrings_misc_stats)
 #define I40E_VSI_STATS_LEN(n)   (I40E_NETDEV_STATS_LEN + \
+				 I40E_MISC_STATS_LEN + \
 				 I40E_QUEUE_STATS_LEN((n)))
 #define I40E_PFC_STATS_LEN ( \
 		(FIELD_SIZEOF(struct i40e_pf, stats.priority_xoff_rx) + \
@@ -135,6 +172,7 @@ static struct i40e_stats i40e_gstrings_stats[] = {
 		 FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_tx) + \
 		 FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_2_xoff)) \
 		 / sizeof(u64))
+#define I40E_VEB_STATS_LEN	ARRAY_SIZE(i40e_gstrings_veb_stats)
 #define I40E_PF_STATS_LEN(n)	(I40E_GLOBAL_STATS_LEN + \
 				 I40E_PFC_STATS_LEN + \
 				 I40E_VSI_STATS_LEN((n)))
@@ -620,10 +658,15 @@ static int i40e_get_sset_count(struct net_device *netdev, int sset)
 	case ETH_SS_TEST:
 		return I40E_TEST_LEN;
 	case ETH_SS_STATS:
-		if (vsi == pf->vsi[pf->lan_vsi])
-			return I40E_PF_STATS_LEN(netdev);
-		else
+		if (vsi == pf->vsi[pf->lan_vsi]) {
+			int len = I40E_PF_STATS_LEN(netdev);
+
+			if (pf->lan_veb != I40E_NO_VEB)
+				len += I40E_VEB_STATS_LEN;
+			return len;
+		} else {
 			return I40E_VSI_STATS_LEN(netdev);
+		}
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -633,6 +676,7 @@ static void i40e_get_ethtool_stats(struct net_device *netdev,
 				   struct ethtool_stats *stats, u64 *data)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_ring *tx_ring, *rx_ring;
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
 	int i = 0;
@@ -648,10 +692,14 @@ static void i40e_get_ethtool_stats(struct net_device *netdev,
 		data[i++] = (i40e_gstrings_net_stats[j].sizeof_stat ==
 			sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
 	}
+	for (j = 0; j < I40E_MISC_STATS_LEN; j++) {
+		p = (char *)vsi + i40e_gstrings_misc_stats[j].stat_offset;
+		data[i++] = (i40e_gstrings_misc_stats[j].sizeof_stat ==
+			    sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+	}
 	rcu_read_lock();
-	for (j = 0; j < vsi->num_queue_pairs; j++, i += 4) {
-		struct i40e_ring *tx_ring = ACCESS_ONCE(vsi->tx_rings[j]);
-		struct i40e_ring *rx_ring;
+	for (j = 0; j < vsi->num_queue_pairs; j++) {
+		tx_ring = ACCESS_ONCE(vsi->tx_rings[j]);
 
 		if (!tx_ring)
 			continue;
@@ -662,33 +710,45 @@ static void i40e_get_ethtool_stats(struct net_device *netdev,
 			data[i] = tx_ring->stats.packets;
 			data[i + 1] = tx_ring->stats.bytes;
 		} while (u64_stats_fetch_retry_irq(&tx_ring->syncp, start));
+		i += 2;
 
 		/* Rx ring is the 2nd half of the queue pair */
 		rx_ring = &tx_ring[1];
 		do {
 			start = u64_stats_fetch_begin_irq(&rx_ring->syncp);
-			data[i + 2] = rx_ring->stats.packets;
-			data[i + 3] = rx_ring->stats.bytes;
+			data[i] = rx_ring->stats.packets;
+			data[i + 1] = rx_ring->stats.bytes;
 		} while (u64_stats_fetch_retry_irq(&rx_ring->syncp, start));
+		i += 2;
 	}
 	rcu_read_unlock();
-	if (vsi == pf->vsi[pf->lan_vsi]) {
-		for (j = 0; j < I40E_GLOBAL_STATS_LEN; j++) {
-			p = (char *)pf + i40e_gstrings_stats[j].stat_offset;
-			data[i++] = (i40e_gstrings_stats[j].sizeof_stat ==
-				   sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
-		}
-		for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
-			data[i++] = pf->stats.priority_xon_tx[j];
-			data[i++] = pf->stats.priority_xoff_tx[j];
-		}
-		for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
-			data[i++] = pf->stats.priority_xon_rx[j];
-			data[i++] = pf->stats.priority_xoff_rx[j];
+	if (vsi != pf->vsi[pf->lan_vsi])
+		return;
+
+	if (pf->lan_veb != I40E_NO_VEB) {
+		struct i40e_veb *veb = pf->veb[pf->lan_veb];
+		for (j = 0; j < I40E_VEB_STATS_LEN; j++) {
+			p = (char *)veb;
+			p += i40e_gstrings_veb_stats[j].stat_offset;
+			data[i++] = (i40e_gstrings_veb_stats[j].sizeof_stat ==
+				     sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
 		}
-		for (j = 0; j < I40E_MAX_USER_PRIORITY; j++)
-			data[i++] = pf->stats.priority_xon_2_xoff[j];
 	}
+	for (j = 0; j < I40E_GLOBAL_STATS_LEN; j++) {
+		p = (char *)pf + i40e_gstrings_stats[j].stat_offset;
+		data[i++] = (i40e_gstrings_stats[j].sizeof_stat ==
+			     sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+	}
+	for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
+		data[i++] = pf->stats.priority_xon_tx[j];
+		data[i++] = pf->stats.priority_xoff_tx[j];
+	}
+	for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
+		data[i++] = pf->stats.priority_xon_rx[j];
+		data[i++] = pf->stats.priority_xoff_rx[j];
+	}
+	for (j = 0; j < I40E_MAX_USER_PRIORITY; j++)
+		data[i++] = pf->stats.priority_xon_2_xoff[j];
 }
 
 static void i40e_get_strings(struct net_device *netdev, u32 stringset,
@@ -713,6 +773,11 @@ static void i40e_get_strings(struct net_device *netdev, u32 stringset,
 				 i40e_gstrings_net_stats[i].stat_string);
 			p += ETH_GSTRING_LEN;
 		}
+		for (i = 0; i < I40E_MISC_STATS_LEN; i++) {
+			snprintf(p, ETH_GSTRING_LEN, "%s",
+				 i40e_gstrings_misc_stats[i].stat_string);
+			p += ETH_GSTRING_LEN;
+		}
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
 			snprintf(p, ETH_GSTRING_LEN, "tx-%u.tx_packets", i);
 			p += ETH_GSTRING_LEN;
@@ -723,34 +788,42 @@ static void i40e_get_strings(struct net_device *netdev, u32 stringset,
 			snprintf(p, ETH_GSTRING_LEN, "rx-%u.rx_bytes", i);
 			p += ETH_GSTRING_LEN;
 		}
-		if (vsi == pf->vsi[pf->lan_vsi]) {
-			for (i = 0; i < I40E_GLOBAL_STATS_LEN; i++) {
-				snprintf(p, ETH_GSTRING_LEN, "port.%s",
-					 i40e_gstrings_stats[i].stat_string);
-				p += ETH_GSTRING_LEN;
-			}
-			for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-				snprintf(p, ETH_GSTRING_LEN,
-					 "port.tx_priority_%u_xon", i);
-				p += ETH_GSTRING_LEN;
-				snprintf(p, ETH_GSTRING_LEN,
-					 "port.tx_priority_%u_xoff", i);
-				p += ETH_GSTRING_LEN;
-			}
-			for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-				snprintf(p, ETH_GSTRING_LEN,
-					 "port.rx_priority_%u_xon", i);
-				p += ETH_GSTRING_LEN;
-				snprintf(p, ETH_GSTRING_LEN,
-					 "port.rx_priority_%u_xoff", i);
-				p += ETH_GSTRING_LEN;
-			}
-			for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-				snprintf(p, ETH_GSTRING_LEN,
-					 "port.rx_priority_%u_xon_2_xoff", i);
+		if (vsi != pf->vsi[pf->lan_vsi])
+			return;
+
+		if (pf->lan_veb != I40E_NO_VEB) {
+			for (i = 0; i < I40E_VEB_STATS_LEN; i++) {
+				snprintf(p, ETH_GSTRING_LEN, "veb.%s",
+					i40e_gstrings_veb_stats[i].stat_string);
 				p += ETH_GSTRING_LEN;
 			}
 		}
+		for (i = 0; i < I40E_GLOBAL_STATS_LEN; i++) {
+			snprintf(p, ETH_GSTRING_LEN, "port.%s",
+				 i40e_gstrings_stats[i].stat_string);
+			p += ETH_GSTRING_LEN;
+		}
+		for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+			snprintf(p, ETH_GSTRING_LEN,
+				 "port.tx_priority_%u_xon", i);
+			p += ETH_GSTRING_LEN;
+			snprintf(p, ETH_GSTRING_LEN,
+				 "port.tx_priority_%u_xoff", i);
+			p += ETH_GSTRING_LEN;
+		}
+		for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+			snprintf(p, ETH_GSTRING_LEN,
+				 "port.rx_priority_%u_xon", i);
+			p += ETH_GSTRING_LEN;
+			snprintf(p, ETH_GSTRING_LEN,
+				 "port.rx_priority_%u_xoff", i);
+			p += ETH_GSTRING_LEN;
+		}
+		for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+			snprintf(p, ETH_GSTRING_LEN,
+				 "port.rx_priority_%u_xon_2_xoff", i);
+			p += ETH_GSTRING_LEN;
+		}
 		/* BUG_ON(p - data != I40E_STATS_LEN * ETH_GSTRING_LEN); */
 		break;
 	}
@@ -1007,14 +1080,13 @@ static int i40e_get_coalesce(struct net_device *netdev,
 	ec->rx_max_coalesced_frames_irq = vsi->work_limit;
 
 	if (ITR_IS_DYNAMIC(vsi->rx_itr_setting))
-		ec->rx_coalesce_usecs = 1;
-	else
-		ec->rx_coalesce_usecs = vsi->rx_itr_setting;
+		ec->use_adaptive_rx_coalesce = 1;
 
 	if (ITR_IS_DYNAMIC(vsi->tx_itr_setting))
-		ec->tx_coalesce_usecs = 1;
-	else
-		ec->tx_coalesce_usecs = vsi->tx_itr_setting;
+		ec->use_adaptive_tx_coalesce = 1;
+
+	ec->rx_coalesce_usecs = vsi->rx_itr_setting & ~I40E_ITR_DYNAMIC;
+	ec->tx_coalesce_usecs = vsi->tx_itr_setting & ~I40E_ITR_DYNAMIC;
 
 	return 0;
 }
@@ -1033,37 +1105,27 @@ static int i40e_set_coalesce(struct net_device *netdev,
 	if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq)
 		vsi->work_limit = ec->tx_max_coalesced_frames_irq;
 
-	switch (ec->rx_coalesce_usecs) {
-	case 0:
-		vsi->rx_itr_setting = 0;
-		break;
-	case 1:
-		vsi->rx_itr_setting = (I40E_ITR_DYNAMIC |
-				       ITR_REG_TO_USEC(I40E_ITR_RX_DEF));
-		break;
-	default:
-		if ((ec->rx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
-		    (ec->rx_coalesce_usecs > (I40E_MAX_ITR << 1)))
-			return -EINVAL;
+	if ((ec->rx_coalesce_usecs >= (I40E_MIN_ITR << 1)) &&
+	    (ec->rx_coalesce_usecs <= (I40E_MAX_ITR << 1)))
 		vsi->rx_itr_setting = ec->rx_coalesce_usecs;
-		break;
-	}
+	else
+		return -EINVAL;
 
-	switch (ec->tx_coalesce_usecs) {
-	case 0:
-		vsi->tx_itr_setting = 0;
-		break;
-	case 1:
-		vsi->tx_itr_setting = (I40E_ITR_DYNAMIC |
-				       ITR_REG_TO_USEC(I40E_ITR_TX_DEF));
-		break;
-	default:
-		if ((ec->tx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
-		    (ec->tx_coalesce_usecs > (I40E_MAX_ITR << 1)))
-			return -EINVAL;
+	if ((ec->tx_coalesce_usecs >= (I40E_MIN_ITR << 1)) &&
+	    (ec->tx_coalesce_usecs <= (I40E_MAX_ITR << 1)))
 		vsi->tx_itr_setting = ec->tx_coalesce_usecs;
-		break;
-	}
+	else
+		return -EINVAL;
+
+	if (ec->use_adaptive_rx_coalesce)
+		vsi->rx_itr_setting |= I40E_ITR_DYNAMIC;
+	else
+		vsi->rx_itr_setting &= ~I40E_ITR_DYNAMIC;
+
+	if (ec->use_adaptive_tx_coalesce)
+		vsi->tx_itr_setting |= I40E_ITR_DYNAMIC;
+	else
+		vsi->tx_itr_setting &= ~I40E_ITR_DYNAMIC;
 
 	vector = vsi->base_vector;
 	for (i = 0; i < vsi->num_q_vectors; i++, vector++) {
@@ -1140,8 +1202,7 @@ static int i40e_get_ethtool_fdir_all(struct i40e_pf *pf,
 	int cnt = 0;
 
 	/* report total rule count */
-	cmd->data = pf->hw.fdir_shared_filter_count +
-		    pf->fdir_pf_filter_count;
+	cmd->data = i40e_get_fd_cnt_all(pf);
 
 	hlist_for_each_entry_safe(rule, node2,
 				  &pf->fdir_filter_list, fdir_node) {
@@ -1175,10 +1236,6 @@ static int i40e_get_ethtool_fdir_entry(struct i40e_pf *pf,
 	struct i40e_fdir_filter *rule = NULL;
 	struct hlist_node *node2;
 
-	/* report total rule count */
-	cmd->data = pf->hw.fdir_shared_filter_count +
-		    pf->fdir_pf_filter_count;
-
 	hlist_for_each_entry_safe(rule, node2,
 				  &pf->fdir_filter_list, fdir_node) {
 		if (fsp->location <= rule->fd_id)
@@ -1189,11 +1246,24 @@ static int i40e_get_ethtool_fdir_entry(struct i40e_pf *pf,
 		return -EINVAL;
 
 	fsp->flow_type = rule->flow_type;
-	fsp->h_u.tcp_ip4_spec.psrc = rule->src_port;
-	fsp->h_u.tcp_ip4_spec.pdst = rule->dst_port;
-	fsp->h_u.tcp_ip4_spec.ip4src = rule->src_ip[0];
-	fsp->h_u.tcp_ip4_spec.ip4dst = rule->dst_ip[0];
-	fsp->ring_cookie = rule->q_index;
+	if (fsp->flow_type == IP_USER_FLOW) {
+		fsp->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4;
+		fsp->h_u.usr_ip4_spec.proto = 0;
+		fsp->m_u.usr_ip4_spec.proto = 0;
+	}
+
+	/* Reverse the src and dest notion, since the HW views them from
+	 * Tx perspective where as the user expects it from Rx filter view.
+	 */
+	fsp->h_u.tcp_ip4_spec.psrc = rule->dst_port;
+	fsp->h_u.tcp_ip4_spec.pdst = rule->src_port;
+	fsp->h_u.tcp_ip4_spec.ip4src = rule->dst_ip[0];
+	fsp->h_u.tcp_ip4_spec.ip4dst = rule->src_ip[0];
+
+	if (rule->dest_ctl == I40E_FILTER_PROGRAM_DESC_DEST_DROP_PACKET)
+		fsp->ring_cookie = RX_CLS_FLOW_DISC;
+	else
+		fsp->ring_cookie = rule->q_index;
 
 	return 0;
 }
@@ -1223,6 +1293,8 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 		break;
 	case ETHTOOL_GRXCLSRLCNT:
 		cmd->rule_cnt = pf->fdir_pf_active_filters;
+		/* report total rule count */
+		cmd->data = i40e_get_fd_cnt_all(pf);
 		ret = 0;
 		break;
 	case ETHTOOL_GRXCLSRULE:
@@ -1291,16 +1363,12 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 	case UDP_V4_FLOW:
 		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 		case 0:
-			hena &=
-			~(((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
+			hena &= ~(((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
+				  ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
 			break;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
-			hena |=
-			(((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP)  |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
+			hena |= (((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
+				  ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
 			break;
 		default:
 			return -EINVAL;
@@ -1309,16 +1377,12 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 	case UDP_V6_FLOW:
 		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 		case 0:
-			hena &=
-			~(((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
+			hena &= ~(((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
+				  ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
 			break;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
-			hena |=
-			(((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP)  |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
+			hena |= (((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
+				 ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
 			break;
 		default:
 			return -EINVAL;
@@ -1503,7 +1567,8 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 		return -EINVAL;
 	}
 
-	if (fsp->ring_cookie >= vsi->num_queue_pairs)
+	if ((fsp->ring_cookie != RX_CLS_FLOW_DISC) &&
+	    (fsp->ring_cookie >= vsi->num_queue_pairs))
 		return -EINVAL;
 
 	input = kzalloc(sizeof(*input), GFP_KERNEL);
@@ -1524,13 +1589,17 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 	input->pctype = 0;
 	input->dest_vsi = vsi->id;
 	input->fd_status = I40E_FILTER_PROGRAM_DESC_FD_STATUS_FD_ID;
-	input->cnt_index = 0;
+	input->cnt_index  = pf->fd_sb_cnt_idx;
 	input->flow_type = fsp->flow_type;
 	input->ip4_proto = fsp->h_u.usr_ip4_spec.proto;
-	input->src_port = fsp->h_u.tcp_ip4_spec.psrc;
-	input->dst_port = fsp->h_u.tcp_ip4_spec.pdst;
-	input->src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
-	input->dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
+
+	/* Reverse the src and dest notion, since the HW expects them to be from
+	 * Tx perspective where as the input from user is from Rx filter view.
+	 */
+	input->dst_port = fsp->h_u.tcp_ip4_spec.psrc;
+	input->src_port = fsp->h_u.tcp_ip4_spec.pdst;
+	input->dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
+	input->src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
 
 	ret = i40e_add_del_fdir(vsi, input, true);
 	if (ret)
@@ -1692,5 +1761,5 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 
 void i40e_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &i40e_ethtool_ops);
+	netdev->ethtool_ops = &i40e_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_hmc.c b/drivers/net/ethernet/intel/i40e/i40e_hmc.c
index bf2d4cc5b569..9b987ccc9e82 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_hmc.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_hmc.c
@@ -201,7 +201,7 @@ exit:
  **/
 i40e_status i40e_remove_pd_bp(struct i40e_hw *hw,
 					struct i40e_hmc_info *hmc_info,
-					u32 idx, bool is_pf)
+					u32 idx)
 {
 	i40e_status ret_code = 0;
 	struct i40e_hmc_pd_entry *pd_entry;
@@ -237,10 +237,7 @@ i40e_status i40e_remove_pd_bp(struct i40e_hw *hw,
 	pd_addr = (u64 *)pd_table->pd_page_addr.va;
 	pd_addr += rel_pd_idx;
 	memset(pd_addr, 0, sizeof(u64));
-	if (is_pf)
-		I40E_INVALIDATE_PF_HMC_PD(hw, sd_idx, idx);
-	else
-		I40E_INVALIDATE_VF_HMC_PD(hw, sd_idx, idx, hmc_info->hmc_fn_id);
+	I40E_INVALIDATE_PF_HMC_PD(hw, sd_idx, idx);
 
 	/* free memory here */
 	ret_code = i40e_free_dma_mem(hw, &(pd_entry->bp.addr));
diff --git a/drivers/net/ethernet/intel/i40e/i40e_hmc.h b/drivers/net/ethernet/intel/i40e/i40e_hmc.h
index 0cd4701234f8..b45d8fedc5e7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_hmc.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_hmc.h
@@ -163,11 +163,6 @@ struct i40e_hmc_info {
 	    (((sd_idx) << I40E_PFHMC_PDINV_PMSDIDX_SHIFT) |		\
 	     ((pd_idx) << I40E_PFHMC_PDINV_PMPDIDX_SHIFT)))
 
-#define I40E_INVALIDATE_VF_HMC_PD(hw, sd_idx, pd_idx, hmc_fn_id)	   \
-	wr32((hw), I40E_GLHMC_VFPDINV((hmc_fn_id) - I40E_FIRST_VF_FPM_ID), \
-	     (((sd_idx) << I40E_PFHMC_PDINV_PMSDIDX_SHIFT) |		   \
-	      ((pd_idx) << I40E_PFHMC_PDINV_PMPDIDX_SHIFT)))
-
 /**
  * I40E_FIND_SD_INDEX_LIMIT - finds segment descriptor index limit
  * @hmc_info: pointer to the HMC configuration information structure
@@ -226,7 +221,7 @@ i40e_status i40e_add_pd_table_entry(struct i40e_hw *hw,
 					      u32 pd_index);
 i40e_status i40e_remove_pd_bp(struct i40e_hw *hw,
 					struct i40e_hmc_info *hmc_info,
-					u32 idx, bool is_pf);
+					u32 idx);
 i40e_status i40e_prep_remove_sd_bp(struct i40e_hmc_info *hmc_info,
 					     u32 idx);
 i40e_status i40e_remove_sd_bp_new(struct i40e_hw *hw,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c
index d5d98fe2691d..870ab1ee072c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.c
@@ -397,7 +397,7 @@ static i40e_status i40e_create_lan_hmc_object(struct i40e_hw *hw,
 				/* remove the backing pages from pd_idx1 to i */
 				while (i && (i > pd_idx1)) {
 					i40e_remove_pd_bp(hw, info->hmc_info,
-							  (i - 1), true);
+							  (i - 1));
 					i--;
 				}
 			}
@@ -433,11 +433,7 @@ exit_sd_error:
 				      ((j - 1) * I40E_HMC_MAX_BP_COUNT));
 			pd_lmt1 = min(pd_lmt, (j * I40E_HMC_MAX_BP_COUNT));
 			for (i = pd_idx1; i < pd_lmt1; i++) {
-				i40e_remove_pd_bp(
-					hw,
-					info->hmc_info,
-					i,
-					true);
+				i40e_remove_pd_bp(hw, info->hmc_info, i);
 			}
 			i40e_remove_pd_page(hw, info->hmc_info, (j - 1));
 			break;
@@ -616,8 +612,7 @@ static i40e_status i40e_delete_lan_hmc_object(struct i40e_hw *hw,
 		pd_table =
 			&info->hmc_info->sd_table.sd_entry[sd_idx].u.pd_table;
 		if (pd_table->pd_entry[rel_pd_idx].valid) {
-			ret_code = i40e_remove_pd_bp(hw, info->hmc_info,
-						     j, true);
+			ret_code = i40e_remove_pd_bp(hw, info->hmc_info, j);
 			if (ret_code)
 				goto exit;
 		}
@@ -747,6 +742,7 @@ static struct i40e_context_ele i40e_hmc_rxq_ce_info[] = {
 	{ I40E_HMC_STORE(i40e_hmc_obj_rxq, tphdata_ena),  1,	195 },
 	{ I40E_HMC_STORE(i40e_hmc_obj_rxq, tphhead_ena),  1,	196 },
 	{ I40E_HMC_STORE(i40e_hmc_obj_rxq, lrxqthresh),   3,	198 },
+	{ I40E_HMC_STORE(i40e_hmc_obj_rxq, prefena),      1,	201 },
 	{ 0 }
 };
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h
index 341de925a298..eb65fe23c4a7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_lan_hmc.h
@@ -56,6 +56,7 @@ struct i40e_hmc_obj_rxq {
 	u8  tphdata_ena;
 	u8  tphhead_ena;
 	u8  lrxqthresh;
+	u8  prefena;	/* NOTE: normally must be set to 1 at init */
 };
 
 /* Tx queue context data */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 2e72449f1265..275ca9a1719e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -38,8 +38,8 @@ static const char i40e_driver_string[] =
 #define DRV_KERN "-k"
 
 #define DRV_VERSION_MAJOR 0
-#define DRV_VERSION_MINOR 3
-#define DRV_VERSION_BUILD 36
+#define DRV_VERSION_MINOR 4
+#define DRV_VERSION_BUILD 10
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
@@ -67,12 +67,10 @@ static int i40e_veb_get_bw_info(struct i40e_veb *veb);
  */
 static DEFINE_PCI_DEVICE_TABLE(i40e_pci_tbl) = {
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_XL710), 0},
-	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X710), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QEMU), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_A), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_B), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_C), 0},
-	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_D), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_A), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_B), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_C), 0},
@@ -356,6 +354,7 @@ static struct rtnl_link_stats64 *i40e_get_netdev_stats_struct(
 					     struct rtnl_link_stats64 *stats)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_ring *tx_ring, *rx_ring;
 	struct i40e_vsi *vsi = np->vsi;
 	struct rtnl_link_stats64 *vsi_stats = i40e_get_vsi_stats_struct(vsi);
 	int i;
@@ -368,7 +367,6 @@ static struct rtnl_link_stats64 *i40e_get_netdev_stats_struct(
 
 	rcu_read_lock();
 	for (i = 0; i < vsi->num_queue_pairs; i++) {
-		struct i40e_ring *tx_ring, *rx_ring;
 		u64 bytes, packets;
 		unsigned int start;
 
@@ -397,7 +395,7 @@ static struct rtnl_link_stats64 *i40e_get_netdev_stats_struct(
 	}
 	rcu_read_unlock();
 
-	/* following stats updated by ixgbe_watchdog_task() */
+	/* following stats updated by i40e_watchdog_subtask() */
 	stats->multicast	= vsi_stats->multicast;
 	stats->tx_errors	= vsi_stats->tx_errors;
 	stats->tx_dropped	= vsi_stats->tx_dropped;
@@ -530,6 +528,12 @@ void i40e_update_eth_stats(struct i40e_vsi *vsi)
 	i40e_stat_update32(hw, I40E_GLV_RDPC(stat_idx),
 			   vsi->stat_offsets_loaded,
 			   &oes->rx_discards, &es->rx_discards);
+	i40e_stat_update32(hw, I40E_GLV_RUPP(stat_idx),
+			   vsi->stat_offsets_loaded,
+			   &oes->rx_unknown_protocol, &es->rx_unknown_protocol);
+	i40e_stat_update32(hw, I40E_GLV_TEPC(stat_idx),
+			   vsi->stat_offsets_loaded,
+			   &oes->tx_errors, &es->tx_errors);
 
 	i40e_stat_update48(hw, I40E_GLV_GORCH(stat_idx),
 			   I40E_GLV_GORCL(stat_idx),
@@ -648,10 +652,10 @@ static void i40e_update_link_xoff_rx(struct i40e_pf *pf)
 		return;
 
 	/* Clear the __I40E_HANG_CHECK_ARMED bit for all Tx rings */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 
-		if (!vsi)
+		if (!vsi || !vsi->tx_rings[0])
 			continue;
 
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
@@ -702,10 +706,10 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf)
 	}
 
 	/* Clear the __I40E_HANG_CHECK_ARMED bit for Tx rings */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 
-		if (!vsi)
+		if (!vsi || !vsi->tx_rings[0])
 			continue;
 
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
@@ -720,19 +724,18 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf)
 }
 
 /**
- * i40e_update_stats - Update the board statistics counters.
+ * i40e_update_vsi_stats - Update the vsi statistics counters.
  * @vsi: the VSI to be updated
  *
  * There are a few instances where we store the same stat in a
  * couple of different structs.  This is partly because we have
  * the netdev stats that need to be filled out, which is slightly
  * different from the "eth_stats" defined by the chip and used in
- * VF communications.  We sort it all out here in a central place.
+ * VF communications.  We sort it out here.
  **/
-void i40e_update_stats(struct i40e_vsi *vsi)
+static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 {
 	struct i40e_pf *pf = vsi->back;
-	struct i40e_hw *hw = &pf->hw;
 	struct rtnl_link_stats64 *ons;
 	struct rtnl_link_stats64 *ns;   /* netdev stats */
 	struct i40e_eth_stats *oes;
@@ -741,8 +744,6 @@ void i40e_update_stats(struct i40e_vsi *vsi)
 	u32 rx_page, rx_buf;
 	u64 rx_p, rx_b;
 	u64 tx_p, tx_b;
-	u32 val;
-	int i;
 	u16 q;
 
 	if (test_bit(__I40E_DOWN, &vsi->state) ||
@@ -804,196 +805,256 @@ void i40e_update_stats(struct i40e_vsi *vsi)
 	ns->tx_packets = tx_p;
 	ns->tx_bytes = tx_b;
 
-	i40e_update_eth_stats(vsi);
 	/* update netdev stats from eth stats */
-	ons->rx_errors = oes->rx_errors;
-	ns->rx_errors = es->rx_errors;
+	i40e_update_eth_stats(vsi);
 	ons->tx_errors = oes->tx_errors;
 	ns->tx_errors = es->tx_errors;
 	ons->multicast = oes->rx_multicast;
 	ns->multicast = es->rx_multicast;
+	ons->rx_dropped = oes->rx_discards;
+	ns->rx_dropped = es->rx_discards;
 	ons->tx_dropped = oes->tx_discards;
 	ns->tx_dropped = es->tx_discards;
 
-	/* Get the port data only if this is the main PF VSI */
+	/* pull in a couple PF stats if this is the main vsi */
 	if (vsi == pf->vsi[pf->lan_vsi]) {
-		struct i40e_hw_port_stats *nsd = &pf->stats;
-		struct i40e_hw_port_stats *osd = &pf->stats_offsets;
+		ns->rx_crc_errors = pf->stats.crc_errors;
+		ns->rx_errors = pf->stats.crc_errors + pf->stats.illegal_bytes;
+		ns->rx_length_errors = pf->stats.rx_length_errors;
+	}
+}
 
-		i40e_stat_update48(hw, I40E_GLPRT_GORCH(hw->port),
-				   I40E_GLPRT_GORCL(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->eth.rx_bytes, &nsd->eth.rx_bytes);
-		i40e_stat_update48(hw, I40E_GLPRT_GOTCH(hw->port),
-				   I40E_GLPRT_GOTCL(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->eth.tx_bytes, &nsd->eth.tx_bytes);
-		i40e_stat_update32(hw, I40E_GLPRT_RDPC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->eth.rx_discards,
-				   &nsd->eth.rx_discards);
-		i40e_stat_update32(hw, I40E_GLPRT_TDPC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->eth.tx_discards,
-				   &nsd->eth.tx_discards);
-		i40e_stat_update48(hw, I40E_GLPRT_MPRCH(hw->port),
-				   I40E_GLPRT_MPRCL(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->eth.rx_multicast,
-				   &nsd->eth.rx_multicast);
+/**
+ * i40e_update_pf_stats - Update the pf statistics counters.
+ * @pf: the PF to be updated
+ **/
+static void i40e_update_pf_stats(struct i40e_pf *pf)
+{
+	struct i40e_hw_port_stats *osd = &pf->stats_offsets;
+	struct i40e_hw_port_stats *nsd = &pf->stats;
+	struct i40e_hw *hw = &pf->hw;
+	u32 val;
+	int i;
 
-		i40e_stat_update32(hw, I40E_GLPRT_TDOLD(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_dropped_link_down,
-				   &nsd->tx_dropped_link_down);
+	i40e_stat_update48(hw, I40E_GLPRT_GORCH(hw->port),
+			   I40E_GLPRT_GORCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.rx_bytes, &nsd->eth.rx_bytes);
+	i40e_stat_update48(hw, I40E_GLPRT_GOTCH(hw->port),
+			   I40E_GLPRT_GOTCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.tx_bytes, &nsd->eth.tx_bytes);
+	i40e_stat_update32(hw, I40E_GLPRT_RDPC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.rx_discards,
+			   &nsd->eth.rx_discards);
+	i40e_stat_update32(hw, I40E_GLPRT_TDPC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.tx_discards,
+			   &nsd->eth.tx_discards);
 
-		i40e_stat_update32(hw, I40E_GLPRT_CRCERRS(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->crc_errors, &nsd->crc_errors);
-		ns->rx_crc_errors = nsd->crc_errors;
+	i40e_stat_update48(hw, I40E_GLPRT_UPRCH(hw->port),
+			   I40E_GLPRT_UPRCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.rx_unicast,
+			   &nsd->eth.rx_unicast);
+	i40e_stat_update48(hw, I40E_GLPRT_MPRCH(hw->port),
+			   I40E_GLPRT_MPRCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.rx_multicast,
+			   &nsd->eth.rx_multicast);
+	i40e_stat_update48(hw, I40E_GLPRT_BPRCH(hw->port),
+			   I40E_GLPRT_BPRCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.rx_broadcast,
+			   &nsd->eth.rx_broadcast);
+	i40e_stat_update48(hw, I40E_GLPRT_UPTCH(hw->port),
+			   I40E_GLPRT_UPTCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.tx_unicast,
+			   &nsd->eth.tx_unicast);
+	i40e_stat_update48(hw, I40E_GLPRT_MPTCH(hw->port),
+			   I40E_GLPRT_MPTCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.tx_multicast,
+			   &nsd->eth.tx_multicast);
+	i40e_stat_update48(hw, I40E_GLPRT_BPTCH(hw->port),
+			   I40E_GLPRT_BPTCL(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->eth.tx_broadcast,
+			   &nsd->eth.tx_broadcast);
 
-		i40e_stat_update32(hw, I40E_GLPRT_ILLERRC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->illegal_bytes, &nsd->illegal_bytes);
-		ns->rx_errors = nsd->crc_errors
-				+ nsd->illegal_bytes;
+	i40e_stat_update32(hw, I40E_GLPRT_TDOLD(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_dropped_link_down,
+			   &nsd->tx_dropped_link_down);
 
-		i40e_stat_update32(hw, I40E_GLPRT_MLFC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->mac_local_faults,
-				   &nsd->mac_local_faults);
-		i40e_stat_update32(hw, I40E_GLPRT_MRFC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->mac_remote_faults,
-				   &nsd->mac_remote_faults);
+	i40e_stat_update32(hw, I40E_GLPRT_CRCERRS(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->crc_errors, &nsd->crc_errors);
 
-		i40e_stat_update32(hw, I40E_GLPRT_RLEC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_length_errors,
-				   &nsd->rx_length_errors);
-		ns->rx_length_errors = nsd->rx_length_errors;
+	i40e_stat_update32(hw, I40E_GLPRT_ILLERRC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->illegal_bytes, &nsd->illegal_bytes);
 
-		i40e_stat_update32(hw, I40E_GLPRT_LXONRXC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->link_xon_rx, &nsd->link_xon_rx);
-		i40e_stat_update32(hw, I40E_GLPRT_LXONTXC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->link_xon_tx, &nsd->link_xon_tx);
-		i40e_update_prio_xoff_rx(pf);  /* handles I40E_GLPRT_LXOFFRXC */
-		i40e_stat_update32(hw, I40E_GLPRT_LXOFFTXC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->link_xoff_tx, &nsd->link_xoff_tx);
-
-		for (i = 0; i < 8; i++) {
-			i40e_stat_update32(hw, I40E_GLPRT_PXONRXC(hw->port, i),
-					   pf->stat_offsets_loaded,
-					   &osd->priority_xon_rx[i],
-					   &nsd->priority_xon_rx[i]);
-			i40e_stat_update32(hw, I40E_GLPRT_PXONTXC(hw->port, i),
-					   pf->stat_offsets_loaded,
-					   &osd->priority_xon_tx[i],
-					   &nsd->priority_xon_tx[i]);
-			i40e_stat_update32(hw, I40E_GLPRT_PXOFFTXC(hw->port, i),
-					   pf->stat_offsets_loaded,
-					   &osd->priority_xoff_tx[i],
-					   &nsd->priority_xoff_tx[i]);
-			i40e_stat_update32(hw,
-					   I40E_GLPRT_RXON2OFFCNT(hw->port, i),
-					   pf->stat_offsets_loaded,
-					   &osd->priority_xon_2_xoff[i],
-					   &nsd->priority_xon_2_xoff[i]);
-		}
+	i40e_stat_update32(hw, I40E_GLPRT_MLFC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->mac_local_faults,
+			   &nsd->mac_local_faults);
+	i40e_stat_update32(hw, I40E_GLPRT_MRFC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->mac_remote_faults,
+			   &nsd->mac_remote_faults);
 
-		i40e_stat_update48(hw, I40E_GLPRT_PRC64H(hw->port),
-				   I40E_GLPRT_PRC64L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_64, &nsd->rx_size_64);
-		i40e_stat_update48(hw, I40E_GLPRT_PRC127H(hw->port),
-				   I40E_GLPRT_PRC127L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_127, &nsd->rx_size_127);
-		i40e_stat_update48(hw, I40E_GLPRT_PRC255H(hw->port),
-				   I40E_GLPRT_PRC255L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_255, &nsd->rx_size_255);
-		i40e_stat_update48(hw, I40E_GLPRT_PRC511H(hw->port),
-				   I40E_GLPRT_PRC511L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_511, &nsd->rx_size_511);
-		i40e_stat_update48(hw, I40E_GLPRT_PRC1023H(hw->port),
-				   I40E_GLPRT_PRC1023L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_1023, &nsd->rx_size_1023);
-		i40e_stat_update48(hw, I40E_GLPRT_PRC1522H(hw->port),
-				   I40E_GLPRT_PRC1522L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_1522, &nsd->rx_size_1522);
-		i40e_stat_update48(hw, I40E_GLPRT_PRC9522H(hw->port),
-				   I40E_GLPRT_PRC9522L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_size_big, &nsd->rx_size_big);
+	i40e_stat_update32(hw, I40E_GLPRT_RLEC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_length_errors,
+			   &nsd->rx_length_errors);
 
-		i40e_stat_update48(hw, I40E_GLPRT_PTC64H(hw->port),
-				   I40E_GLPRT_PTC64L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_64, &nsd->tx_size_64);
-		i40e_stat_update48(hw, I40E_GLPRT_PTC127H(hw->port),
-				   I40E_GLPRT_PTC127L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_127, &nsd->tx_size_127);
-		i40e_stat_update48(hw, I40E_GLPRT_PTC255H(hw->port),
-				   I40E_GLPRT_PTC255L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_255, &nsd->tx_size_255);
-		i40e_stat_update48(hw, I40E_GLPRT_PTC511H(hw->port),
-				   I40E_GLPRT_PTC511L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_511, &nsd->tx_size_511);
-		i40e_stat_update48(hw, I40E_GLPRT_PTC1023H(hw->port),
-				   I40E_GLPRT_PTC1023L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_1023, &nsd->tx_size_1023);
-		i40e_stat_update48(hw, I40E_GLPRT_PTC1522H(hw->port),
-				   I40E_GLPRT_PTC1522L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_1522, &nsd->tx_size_1522);
-		i40e_stat_update48(hw, I40E_GLPRT_PTC9522H(hw->port),
-				   I40E_GLPRT_PTC9522L(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->tx_size_big, &nsd->tx_size_big);
+	i40e_stat_update32(hw, I40E_GLPRT_LXONRXC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->link_xon_rx, &nsd->link_xon_rx);
+	i40e_stat_update32(hw, I40E_GLPRT_LXONTXC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->link_xon_tx, &nsd->link_xon_tx);
+	i40e_update_prio_xoff_rx(pf);  /* handles I40E_GLPRT_LXOFFRXC */
+	i40e_stat_update32(hw, I40E_GLPRT_LXOFFTXC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->link_xoff_tx, &nsd->link_xoff_tx);
 
-		i40e_stat_update32(hw, I40E_GLPRT_RUC(hw->port),
-				   pf->stat_offsets_loaded,
-				   &osd->rx_undersize, &nsd->rx_undersize);
-		i40e_stat_update32(hw, I40E_GLPRT_RFC(hw->port),
+	for (i = 0; i < 8; i++) {
+		i40e_stat_update32(hw, I40E_GLPRT_PXONRXC(hw->port, i),
 				   pf->stat_offsets_loaded,
-				   &osd->rx_fragments, &nsd->rx_fragments);
-		i40e_stat_update32(hw, I40E_GLPRT_ROC(hw->port),
+				   &osd->priority_xon_rx[i],
+				   &nsd->priority_xon_rx[i]);
+		i40e_stat_update32(hw, I40E_GLPRT_PXONTXC(hw->port, i),
 				   pf->stat_offsets_loaded,
-				   &osd->rx_oversize, &nsd->rx_oversize);
-		i40e_stat_update32(hw, I40E_GLPRT_RJC(hw->port),
+				   &osd->priority_xon_tx[i],
+				   &nsd->priority_xon_tx[i]);
+		i40e_stat_update32(hw, I40E_GLPRT_PXOFFTXC(hw->port, i),
 				   pf->stat_offsets_loaded,
-				   &osd->rx_jabber, &nsd->rx_jabber);
-
-		val = rd32(hw, I40E_PRTPM_EEE_STAT);
-		nsd->tx_lpi_status =
-			       (val & I40E_PRTPM_EEE_STAT_TX_LPI_STATUS_MASK) >>
-				I40E_PRTPM_EEE_STAT_TX_LPI_STATUS_SHIFT;
-		nsd->rx_lpi_status =
-			       (val & I40E_PRTPM_EEE_STAT_RX_LPI_STATUS_MASK) >>
-				I40E_PRTPM_EEE_STAT_RX_LPI_STATUS_SHIFT;
-		i40e_stat_update32(hw, I40E_PRTPM_TLPIC,
+				   &osd->priority_xoff_tx[i],
+				   &nsd->priority_xoff_tx[i]);
+		i40e_stat_update32(hw,
+				   I40E_GLPRT_RXON2OFFCNT(hw->port, i),
 				   pf->stat_offsets_loaded,
-				   &osd->tx_lpi_count, &nsd->tx_lpi_count);
-		i40e_stat_update32(hw, I40E_PRTPM_RLPIC,
-				   pf->stat_offsets_loaded,
-				   &osd->rx_lpi_count, &nsd->rx_lpi_count);
+				   &osd->priority_xon_2_xoff[i],
+				   &nsd->priority_xon_2_xoff[i]);
 	}
 
+	i40e_stat_update48(hw, I40E_GLPRT_PRC64H(hw->port),
+			   I40E_GLPRT_PRC64L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_64, &nsd->rx_size_64);
+	i40e_stat_update48(hw, I40E_GLPRT_PRC127H(hw->port),
+			   I40E_GLPRT_PRC127L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_127, &nsd->rx_size_127);
+	i40e_stat_update48(hw, I40E_GLPRT_PRC255H(hw->port),
+			   I40E_GLPRT_PRC255L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_255, &nsd->rx_size_255);
+	i40e_stat_update48(hw, I40E_GLPRT_PRC511H(hw->port),
+			   I40E_GLPRT_PRC511L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_511, &nsd->rx_size_511);
+	i40e_stat_update48(hw, I40E_GLPRT_PRC1023H(hw->port),
+			   I40E_GLPRT_PRC1023L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_1023, &nsd->rx_size_1023);
+	i40e_stat_update48(hw, I40E_GLPRT_PRC1522H(hw->port),
+			   I40E_GLPRT_PRC1522L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_1522, &nsd->rx_size_1522);
+	i40e_stat_update48(hw, I40E_GLPRT_PRC9522H(hw->port),
+			   I40E_GLPRT_PRC9522L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_size_big, &nsd->rx_size_big);
+
+	i40e_stat_update48(hw, I40E_GLPRT_PTC64H(hw->port),
+			   I40E_GLPRT_PTC64L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_64, &nsd->tx_size_64);
+	i40e_stat_update48(hw, I40E_GLPRT_PTC127H(hw->port),
+			   I40E_GLPRT_PTC127L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_127, &nsd->tx_size_127);
+	i40e_stat_update48(hw, I40E_GLPRT_PTC255H(hw->port),
+			   I40E_GLPRT_PTC255L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_255, &nsd->tx_size_255);
+	i40e_stat_update48(hw, I40E_GLPRT_PTC511H(hw->port),
+			   I40E_GLPRT_PTC511L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_511, &nsd->tx_size_511);
+	i40e_stat_update48(hw, I40E_GLPRT_PTC1023H(hw->port),
+			   I40E_GLPRT_PTC1023L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_1023, &nsd->tx_size_1023);
+	i40e_stat_update48(hw, I40E_GLPRT_PTC1522H(hw->port),
+			   I40E_GLPRT_PTC1522L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_1522, &nsd->tx_size_1522);
+	i40e_stat_update48(hw, I40E_GLPRT_PTC9522H(hw->port),
+			   I40E_GLPRT_PTC9522L(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->tx_size_big, &nsd->tx_size_big);
+
+	i40e_stat_update32(hw, I40E_GLPRT_RUC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_undersize, &nsd->rx_undersize);
+	i40e_stat_update32(hw, I40E_GLPRT_RFC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_fragments, &nsd->rx_fragments);
+	i40e_stat_update32(hw, I40E_GLPRT_ROC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_oversize, &nsd->rx_oversize);
+	i40e_stat_update32(hw, I40E_GLPRT_RJC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->rx_jabber, &nsd->rx_jabber);
+
+	/* FDIR stats */
+	i40e_stat_update32(hw, I40E_GLQF_PCNT(pf->fd_atr_cnt_idx),
+			   pf->stat_offsets_loaded,
+			   &osd->fd_atr_match, &nsd->fd_atr_match);
+	i40e_stat_update32(hw, I40E_GLQF_PCNT(pf->fd_sb_cnt_idx),
+			   pf->stat_offsets_loaded,
+			   &osd->fd_sb_match, &nsd->fd_sb_match);
+
+	val = rd32(hw, I40E_PRTPM_EEE_STAT);
+	nsd->tx_lpi_status =
+		       (val & I40E_PRTPM_EEE_STAT_TX_LPI_STATUS_MASK) >>
+			I40E_PRTPM_EEE_STAT_TX_LPI_STATUS_SHIFT;
+	nsd->rx_lpi_status =
+		       (val & I40E_PRTPM_EEE_STAT_RX_LPI_STATUS_MASK) >>
+			I40E_PRTPM_EEE_STAT_RX_LPI_STATUS_SHIFT;
+	i40e_stat_update32(hw, I40E_PRTPM_TLPIC,
+			   pf->stat_offsets_loaded,
+			   &osd->tx_lpi_count, &nsd->tx_lpi_count);
+	i40e_stat_update32(hw, I40E_PRTPM_RLPIC,
+			   pf->stat_offsets_loaded,
+			   &osd->rx_lpi_count, &nsd->rx_lpi_count);
+
 	pf->stat_offsets_loaded = true;
 }
 
 /**
+ * i40e_update_stats - Update the various statistics counters.
+ * @vsi: the VSI to be updated
+ *
+ * Update the various stats for this VSI and its related entities.
+ **/
+void i40e_update_stats(struct i40e_vsi *vsi)
+{
+	struct i40e_pf *pf = vsi->back;
+
+	if (vsi == pf->vsi[pf->lan_vsi])
+		i40e_update_pf_stats(pf);
+
+	i40e_update_vsi_stats(vsi);
+}
+
+/**
  * i40e_find_filter - Search VSI filter list for specific mac/vlan filter
  * @vsi: the VSI to be searched
  * @macaddr: the MAC address
@@ -1101,6 +1162,30 @@ struct i40e_mac_filter *i40e_put_mac_in_vlan(struct i40e_vsi *vsi, u8 *macaddr,
 }
 
 /**
+ * i40e_rm_default_mac_filter - Remove the default MAC filter set by NVM
+ * @vsi: the PF Main VSI - inappropriate for any other VSI
+ * @macaddr: the MAC address
+ **/
+static void i40e_rm_default_mac_filter(struct i40e_vsi *vsi, u8 *macaddr)
+{
+	struct i40e_aqc_remove_macvlan_element_data element;
+	struct i40e_pf *pf = vsi->back;
+	i40e_status aq_ret;
+
+	/* Only appropriate for the PF main VSI */
+	if (vsi->type != I40E_VSI_MAIN)
+		return;
+
+	ether_addr_copy(element.mac_addr, macaddr);
+	element.vlan_tag = 0;
+	element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH |
+			I40E_AQC_MACVLAN_DEL_IGNORE_VLAN;
+	aq_ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL);
+	if (aq_ret)
+		dev_err(&pf->pdev->dev, "Could not remove default MAC-VLAN\n");
+}
+
+/**
  * i40e_add_filter - Add a mac/vlan filter to the VSI
  * @vsi: the VSI to be searched
  * @macaddr: the MAC address
@@ -1125,7 +1210,7 @@ struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi,
 		if (!f)
 			goto add_filter_out;
 
-		memcpy(f->macaddr, macaddr, ETH_ALEN);
+		ether_addr_copy(f->macaddr, macaddr);
 		f->vlan = vlan;
 		f->changed = true;
 
@@ -1249,7 +1334,7 @@ static int i40e_set_mac(struct net_device *netdev, void *p)
 			return -EADDRNOTAVAIL;
 		}
 
-		memcpy(vsi->back->hw.mac.addr, addr->sa_data, netdev->addr_len);
+		ether_addr_copy(vsi->back->hw.mac.addr, addr->sa_data);
 	}
 
 	/* In order to be sure to not drop any packets, add the new address
@@ -1263,7 +1348,7 @@ static int i40e_set_mac(struct net_device *netdev, void *p)
 	i40e_del_filter(vsi, netdev->dev_addr, I40E_VLAN_ANY, false, false);
 	i40e_sync_vsi_filters(vsi);
 
-	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	ether_addr_copy(netdev->dev_addr, addr->sa_data);
 
 	return 0;
 }
@@ -1313,7 +1398,7 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi,
 	vsi->tc_config.numtc = numtc;
 	vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1;
 	/* Number of queues per enabled TC */
-	num_tc_qps = rounddown_pow_of_two(vsi->alloc_queue_pairs/numtc);
+	num_tc_qps = vsi->alloc_queue_pairs/numtc;
 	num_tc_qps = min_t(int, num_tc_qps, I40E_MAX_QUEUES_PER_TC);
 
 	/* Setup queue offset/count for all TCs for given VSI */
@@ -1520,8 +1605,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 			cmd_flags = 0;
 
 			/* add to delete list */
-			memcpy(del_list[num_del].mac_addr,
-			       f->macaddr, ETH_ALEN);
+			ether_addr_copy(del_list[num_del].mac_addr, f->macaddr);
 			del_list[num_del].vlan_tag =
 				cpu_to_le16((u16)(f->vlan ==
 					    I40E_VLAN_ANY ? 0 : f->vlan));
@@ -1542,7 +1626,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 				num_del = 0;
 				memset(del_list, 0, sizeof(*del_list));
 
-				if (aq_ret)
+				if (aq_ret &&
+				    pf->hw.aq.asq_last_status !=
+							      I40E_AQ_RC_ENOENT)
 					dev_info(&pf->pdev->dev,
 						 "ignoring delete macvlan error, err %d, aq_err %d while flushing a full buffer\n",
 						 aq_ret,
@@ -1554,7 +1640,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 						     del_list, num_del, NULL);
 			num_del = 0;
 
-			if (aq_ret)
+			if (aq_ret &&
+			    pf->hw.aq.asq_last_status != I40E_AQ_RC_ENOENT)
 				dev_info(&pf->pdev->dev,
 					 "ignoring delete macvlan error, err %d, aq_err %d\n",
 					 aq_ret, pf->hw.aq.asq_last_status);
@@ -1583,8 +1670,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 			cmd_flags = 0;
 
 			/* add to add array */
-			memcpy(add_list[num_add].mac_addr,
-			       f->macaddr, ETH_ALEN);
+			ether_addr_copy(add_list[num_add].mac_addr, f->macaddr);
 			add_list[num_add].vlan_tag =
 				cpu_to_le16(
 				 (u16)(f->vlan == I40E_VLAN_ANY ? 0 : f->vlan));
@@ -1681,7 +1767,7 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf)
 		return;
 	pf->flags &= ~I40E_FLAG_FILTER_SYNC;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v] &&
 		    (pf->vsi[v]->flags & I40E_VSI_FLAG_FILTER_CHANGED))
 			i40e_sync_vsi_filters(pf->vsi[v]);
@@ -1698,7 +1784,7 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf)
 static int i40e_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN;
+	int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
 	struct i40e_vsi *vsi = np->vsi;
 
 	/* MTU < 68 is an error and causes problems on some kernels */
@@ -2312,6 +2398,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring)
 	rx_ctx.crcstrip = 1;
 	rx_ctx.l2tsel = 1;
 	rx_ctx.showiv = 1;
+	/* set the prefena field to 1 because the manual says to */
+	rx_ctx.prefena = 1;
 
 	/* clear the context in the HMC */
 	err = i40e_clear_lan_rx_queue_context(hw, pf_q);
@@ -2413,6 +2501,7 @@ static int i40e_vsi_configure_rx(struct i40e_vsi *vsi)
  **/
 static void i40e_vsi_config_dcb_rings(struct i40e_vsi *vsi)
 {
+	struct i40e_ring *tx_ring, *rx_ring;
 	u16 qoffset, qcount;
 	int i, n;
 
@@ -2426,8 +2515,8 @@ static void i40e_vsi_config_dcb_rings(struct i40e_vsi *vsi)
 		qoffset = vsi->tc_config.tc_info[n].qoffset;
 		qcount = vsi->tc_config.tc_info[n].qcount;
 		for (i = qoffset; i < (qoffset + qcount); i++) {
-			struct i40e_ring *rx_ring = vsi->rx_rings[i];
-			struct i40e_ring *tx_ring = vsi->tx_rings[i];
+			rx_ring = vsi->rx_rings[i];
+			tx_ring = vsi->tx_rings[i];
 			rx_ring->dcb_tc = n;
 			tx_ring->dcb_tc = n;
 		}
@@ -2565,7 +2654,6 @@ static void i40e_enable_misc_int_causes(struct i40e_hw *hw)
 	      I40E_PFINT_ICR0_ENA_PCI_EXCEPTION_MASK |
 	      I40E_PFINT_ICR0_ENA_GPIO_MASK          |
 	      I40E_PFINT_ICR0_ENA_TIMESYNC_MASK      |
-	      I40E_PFINT_ICR0_ENA_STORM_DETECT_MASK  |
 	      I40E_PFINT_ICR0_ENA_HMC_ERR_MASK       |
 	      I40E_PFINT_ICR0_ENA_VFLR_MASK          |
 	      I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
@@ -2733,6 +2821,7 @@ static int i40e_vsi_request_irq_msix(struct i40e_vsi *vsi, char *basename)
 				      &q_vector->affinity_mask);
 	}
 
+	vsi->irqs_ready = true;
 	return 0;
 
 free_queue_irqs:
@@ -3152,6 +3241,12 @@ static int i40e_vsi_control_tx(struct i40e_vsi *vsi, bool enable)
 
 	pf_q = vsi->base_queue;
 	for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
+
+		/* warn the TX unit of coming changes */
+		i40e_pre_tx_queue_cfg(&pf->hw, pf_q, enable);
+		if (!enable)
+			udelay(10);
+
 		for (j = 0; j < 50; j++) {
 			tx_reg = rd32(hw, I40E_QTX_ENA(pf_q));
 			if (((tx_reg >> I40E_QTX_ENA_QENA_REQ_SHIFT) & 1) ==
@@ -3160,9 +3255,7 @@ static int i40e_vsi_control_tx(struct i40e_vsi *vsi, bool enable)
 			usleep_range(1000, 2000);
 		}
 		/* Skip if the queue is already in the requested state */
-		if (enable && (tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
-			continue;
-		if (!enable && !(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
+		if (enable == !!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
 			continue;
 
 		/* turn on/off the queue */
@@ -3178,13 +3271,8 @@ static int i40e_vsi_control_tx(struct i40e_vsi *vsi, bool enable)
 		/* wait for the change to finish */
 		for (j = 0; j < 10; j++) {
 			tx_reg = rd32(hw, I40E_QTX_ENA(pf_q));
-			if (enable) {
-				if ((tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
-					break;
-			} else {
-				if (!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
-					break;
-			}
+			if (enable == !!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
+				break;
 
 			udelay(10);
 		}
@@ -3223,15 +3311,9 @@ static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable)
 			usleep_range(1000, 2000);
 		}
 
-		if (enable) {
-			/* is STAT set ? */
-			if ((rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
-				continue;
-		} else {
-			/* is !STAT set ? */
-			if (!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
-				continue;
-		}
+		/* Skip if the queue is already in the requested state */
+		if (enable == !!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
+			continue;
 
 		/* turn on/off the queue */
 		if (enable)
@@ -3244,13 +3326,8 @@ static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable)
 		for (j = 0; j < 10; j++) {
 			rx_reg = rd32(hw, I40E_QRX_ENA(pf_q));
 
-			if (enable) {
-				if ((rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
-					break;
-			} else {
-				if (!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
-					break;
-			}
+			if (enable == !!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
+				break;
 
 			udelay(10);
 		}
@@ -3304,6 +3381,10 @@ static void i40e_vsi_free_irq(struct i40e_vsi *vsi)
 		if (!vsi->q_vectors)
 			return;
 
+		if (!vsi->irqs_ready)
+			return;
+
+		vsi->irqs_ready = false;
 		for (i = 0; i < vsi->num_q_vectors; i++) {
 			u16 vector = i + base;
 
@@ -3476,7 +3557,7 @@ static void i40e_clear_interrupt_scheme(struct i40e_pf *pf)
 	int i;
 
 	i40e_put_lump(pf->irq_pile, 0, I40E_PILE_VALID_BIT-1);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i])
 			i40e_vsi_free_q_vectors(pf->vsi[i]);
 	i40e_reset_interrupt_capability(pf);
@@ -3513,6 +3594,19 @@ static void i40e_napi_disable_all(struct i40e_vsi *vsi)
 }
 
 /**
+ * i40e_vsi_close - Shut down a VSI
+ * @vsi: the vsi to be quelled
+ **/
+static void i40e_vsi_close(struct i40e_vsi *vsi)
+{
+	if (!test_and_set_bit(__I40E_DOWN, &vsi->state))
+		i40e_down(vsi);
+	i40e_vsi_free_irq(vsi);
+	i40e_vsi_free_tx_resources(vsi);
+	i40e_vsi_free_rx_resources(vsi);
+}
+
+/**
  * i40e_quiesce_vsi - Pause a given VSI
  * @vsi: the VSI being paused
  **/
@@ -3525,8 +3619,7 @@ static void i40e_quiesce_vsi(struct i40e_vsi *vsi)
 	if (vsi->netdev && netif_running(vsi->netdev)) {
 		vsi->netdev->netdev_ops->ndo_stop(vsi->netdev);
 	} else {
-		set_bit(__I40E_DOWN, &vsi->state);
-		i40e_down(vsi);
+		i40e_vsi_close(vsi);
 	}
 }
 
@@ -3543,7 +3636,7 @@ static void i40e_unquiesce_vsi(struct i40e_vsi *vsi)
 	if (vsi->netdev && netif_running(vsi->netdev))
 		vsi->netdev->netdev_ops->ndo_open(vsi->netdev);
 	else
-		i40e_up(vsi);   /* this clears the DOWN bit */
+		i40e_vsi_open(vsi);   /* this clears the DOWN bit */
 }
 
 /**
@@ -3554,7 +3647,7 @@ static void i40e_pf_quiesce_all_vsi(struct i40e_pf *pf)
 {
 	int v;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			i40e_quiesce_vsi(pf->vsi[v]);
 	}
@@ -3568,7 +3661,7 @@ static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf)
 {
 	int v;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			i40e_unquiesce_vsi(pf->vsi[v]);
 	}
@@ -4009,7 +4102,7 @@ static void i40e_dcb_reconfigure(struct i40e_pf *pf)
 	}
 
 	/* Update each VSI */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (!pf->vsi[v])
 			continue;
 
@@ -4028,6 +4121,8 @@ static void i40e_dcb_reconfigure(struct i40e_pf *pf)
 				 pf->vsi[v]->seid);
 			/* Will try to configure as many components */
 		} else {
+			/* Re-configure VSI vectors based on updated TC map */
+			i40e_vsi_map_rings_to_vectors(pf->vsi[v]);
 			if (pf->vsi[v]->netdev)
 				i40e_dcbnl_set_all(pf->vsi[v]);
 		}
@@ -4065,14 +4160,69 @@ static int i40e_init_pf_dcb(struct i40e_pf *pf)
 			/* When status is not DISABLED then DCBX in FW */
 			pf->dcbx_cap = DCB_CAP_DCBX_LLD_MANAGED |
 				       DCB_CAP_DCBX_VER_IEEE;
-			pf->flags |= I40E_FLAG_DCB_ENABLED;
+
+			pf->flags |= I40E_FLAG_DCB_CAPABLE;
+			/* Enable DCB tagging only when more than one TC */
+			if (i40e_dcb_get_num_tc(&hw->local_dcbx_config) > 1)
+				pf->flags |= I40E_FLAG_DCB_ENABLED;
 		}
+	} else {
+		dev_info(&pf->pdev->dev, "AQ Querying DCB configuration failed: %d\n",
+			 pf->hw.aq.asq_last_status);
 	}
 
 out:
 	return err;
 }
 #endif /* CONFIG_I40E_DCB */
+#define SPEED_SIZE 14
+#define FC_SIZE 8
+/**
+ * i40e_print_link_message - print link up or down
+ * @vsi: the VSI for which link needs a message
+ */
+static void i40e_print_link_message(struct i40e_vsi *vsi, bool isup)
+{
+	char speed[SPEED_SIZE] = "Unknown";
+	char fc[FC_SIZE] = "RX/TX";
+
+	if (!isup) {
+		netdev_info(vsi->netdev, "NIC Link is Down\n");
+		return;
+	}
+
+	switch (vsi->back->hw.phy.link_info.link_speed) {
+	case I40E_LINK_SPEED_40GB:
+		strncpy(speed, "40 Gbps", SPEED_SIZE);
+		break;
+	case I40E_LINK_SPEED_10GB:
+		strncpy(speed, "10 Gbps", SPEED_SIZE);
+		break;
+	case I40E_LINK_SPEED_1GB:
+		strncpy(speed, "1000 Mbps", SPEED_SIZE);
+		break;
+	default:
+		break;
+	}
+
+	switch (vsi->back->hw.fc.current_mode) {
+	case I40E_FC_FULL:
+		strncpy(fc, "RX/TX", FC_SIZE);
+		break;
+	case I40E_FC_TX_PAUSE:
+		strncpy(fc, "TX", FC_SIZE);
+		break;
+	case I40E_FC_RX_PAUSE:
+		strncpy(fc, "RX", FC_SIZE);
+		break;
+	default:
+		strncpy(fc, "None", FC_SIZE);
+		break;
+	}
+
+	netdev_info(vsi->netdev, "NIC Link is Up %s Full Duplex, Flow Control: %s\n",
+		    speed, fc);
+}
 
 /**
  * i40e_up_complete - Finish the last steps of bringing up a connection
@@ -4099,11 +4249,11 @@ static int i40e_up_complete(struct i40e_vsi *vsi)
 
 	if ((pf->hw.phy.link_info.link_info & I40E_AQ_LINK_UP) &&
 	    (vsi->netdev)) {
-		netdev_info(vsi->netdev, "NIC Link is Up\n");
+		i40e_print_link_message(vsi, true);
 		netif_tx_start_all_queues(vsi->netdev);
 		netif_carrier_on(vsi->netdev);
 	} else if (vsi->netdev) {
-		netdev_info(vsi->netdev, "NIC Link is Down\n");
+		i40e_print_link_message(vsi, false);
 	}
 
 	/* replay FDIR SB filters */
@@ -4309,24 +4459,32 @@ int i40e_vsi_open(struct i40e_vsi *vsi)
 	if (err)
 		goto err_setup_rx;
 
-	if (!vsi->netdev) {
-		err = EINVAL;
-		goto err_setup_rx;
-	}
-	snprintf(int_name, sizeof(int_name) - 1, "%s-%s",
-		 dev_driver_string(&pf->pdev->dev), vsi->netdev->name);
-	err = i40e_vsi_request_irq(vsi, int_name);
-	if (err)
-		goto err_setup_rx;
+	if (vsi->netdev) {
+		snprintf(int_name, sizeof(int_name) - 1, "%s-%s",
+			 dev_driver_string(&pf->pdev->dev), vsi->netdev->name);
+		err = i40e_vsi_request_irq(vsi, int_name);
+		if (err)
+			goto err_setup_rx;
 
-	/* Notify the stack of the actual queue counts. */
-	err = netif_set_real_num_tx_queues(vsi->netdev, vsi->num_queue_pairs);
-	if (err)
-		goto err_set_queues;
+		/* Notify the stack of the actual queue counts. */
+		err = netif_set_real_num_tx_queues(vsi->netdev,
+						   vsi->num_queue_pairs);
+		if (err)
+			goto err_set_queues;
 
-	err = netif_set_real_num_rx_queues(vsi->netdev, vsi->num_queue_pairs);
-	if (err)
-		goto err_set_queues;
+		err = netif_set_real_num_rx_queues(vsi->netdev,
+						   vsi->num_queue_pairs);
+		if (err)
+			goto err_set_queues;
+
+	} else if (vsi->type == I40E_VSI_FDIR) {
+		snprintf(int_name, sizeof(int_name) - 1, "%s-fdir",
+			 dev_driver_string(&pf->pdev->dev));
+		err = i40e_vsi_request_irq(vsi, int_name);
+	} else {
+		err = -EINVAL;
+		goto err_setup_rx;
+	}
 
 	err = i40e_up_complete(vsi);
 	if (err)
@@ -4383,14 +4541,7 @@ static int i40e_close(struct net_device *netdev)
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
 
-	if (test_and_set_bit(__I40E_DOWN, &vsi->state))
-		return 0;
-
-	i40e_down(vsi);
-	i40e_vsi_free_irq(vsi);
-
-	i40e_vsi_free_tx_resources(vsi);
-	i40e_vsi_free_rx_resources(vsi);
+	i40e_vsi_close(vsi);
 
 	return 0;
 }
@@ -4410,6 +4561,9 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags)
 
 	WARN_ON(in_interrupt());
 
+	if (i40e_check_asq_alive(&pf->hw))
+		i40e_vc_notify_reset(pf);
+
 	/* do the biggest reset indicated */
 	if (reset_flags & (1 << __I40E_GLOBAL_RESET_REQUESTED)) {
 
@@ -4475,7 +4629,7 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags)
 		/* Find the VSI(s) that requested a re-init */
 		dev_info(&pf->pdev->dev,
 			 "VSI reinit requested\n");
-		for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+		for (v = 0; v < pf->num_alloc_vsi; v++) {
 			struct i40e_vsi *vsi = pf->vsi[v];
 			if (vsi != NULL &&
 			    test_bit(__I40E_REINIT_REQUESTED, &vsi->state)) {
@@ -4565,6 +4719,10 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf,
 	int ret = 0;
 	u8 type;
 
+	/* Not DCB capable or capability disabled */
+	if (!(pf->flags & I40E_FLAG_DCB_CAPABLE))
+		return ret;
+
 	/* Ignore if event is not for Nearest Bridge */
 	type = ((mib->type >> I40E_AQ_LLDP_BRIDGE_TYPE_SHIFT)
 		& I40E_AQ_LLDP_BRIDGE_TYPE_MASK);
@@ -4606,6 +4764,12 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf,
 	if (!need_reconfig)
 		goto exit;
 
+	/* Enable DCB tagging only when more than one TC */
+	if (i40e_dcb_get_num_tc(dcbx_cfg) > 1)
+		pf->flags |= I40E_FLAG_DCB_ENABLED;
+	else
+		pf->flags &= ~I40E_FLAG_DCB_ENABLED;
+
 	/* Reconfiguration needed quiesce all VSIs */
 	i40e_pf_quiesce_all_vsi(pf);
 
@@ -4709,8 +4873,7 @@ void i40e_fdir_check_and_reenable(struct i40e_pf *pf)
 	    (pf->flags & I40E_FLAG_FD_SB_ENABLED))
 		return;
 	fcnt_prog = i40e_get_current_fd_count(pf);
-	fcnt_avail = pf->hw.fdir_shared_filter_count +
-					       pf->fdir_pf_filter_count;
+	fcnt_avail = i40e_get_fd_cnt_all(pf);
 	if (fcnt_prog < (fcnt_avail - I40E_FDIR_BUFFER_HEAD_ROOM)) {
 		if ((pf->flags & I40E_FLAG_FD_SB_ENABLED) &&
 		    (pf->auto_disable_flags & I40E_FLAG_FD_SB_ENABLED)) {
@@ -4803,7 +4966,7 @@ static void i40e_veb_link_event(struct i40e_veb *veb, bool link_up)
 			i40e_veb_link_event(pf->veb[i], link_up);
 
 	/* ... now the local VSIs */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && (pf->vsi[i]->uplink_seid == veb->seid))
 			i40e_vsi_link_event(pf->vsi[i], link_up);
 }
@@ -4821,10 +4984,8 @@ static void i40e_link_event(struct i40e_pf *pf)
 
 	if (new_link == old_link)
 		return;
-
 	if (!test_bit(__I40E_DOWN, &pf->vsi[pf->lan_vsi]->state))
-		netdev_info(pf->vsi[pf->lan_vsi]->netdev,
-			    "NIC Link is %s\n", (new_link ? "Up" : "Down"));
+		i40e_print_link_message(pf->vsi[pf->lan_vsi], new_link);
 
 	/* Notify the base of the switch tree connected to
 	 * the link.  Floating VEBs are not notified.
@@ -4862,7 +5023,7 @@ static void i40e_check_hang_subtask(struct i40e_pf *pf)
 	 *     for each q_vector
 	 *         force an interrupt
 	 */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 		int armed = 0;
 
@@ -4912,7 +5073,7 @@ static void i40e_watchdog_subtask(struct i40e_pf *pf)
 	/* Update the stats for active netdevs so the network stack
 	 * can look at updated numbers whenever it cares to
 	 */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && pf->vsi[i]->netdev)
 			i40e_update_stats(pf->vsi[i]);
 
@@ -5018,11 +5179,47 @@ static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
 	u16 pending, i = 0;
 	i40e_status ret;
 	u16 opcode;
+	u32 oldval;
 	u32 val;
 
 	if (!test_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state))
 		return;
 
+	/* check for error indications */
+	val = rd32(&pf->hw, pf->hw.aq.arq.len);
+	oldval = val;
+	if (val & I40E_PF_ARQLEN_ARQVFE_MASK) {
+		dev_info(&pf->pdev->dev, "ARQ VF Error detected\n");
+		val &= ~I40E_PF_ARQLEN_ARQVFE_MASK;
+	}
+	if (val & I40E_PF_ARQLEN_ARQOVFL_MASK) {
+		dev_info(&pf->pdev->dev, "ARQ Overflow Error detected\n");
+		val &= ~I40E_PF_ARQLEN_ARQOVFL_MASK;
+	}
+	if (val & I40E_PF_ARQLEN_ARQCRIT_MASK) {
+		dev_info(&pf->pdev->dev, "ARQ Critical Error detected\n");
+		val &= ~I40E_PF_ARQLEN_ARQCRIT_MASK;
+	}
+	if (oldval != val)
+		wr32(&pf->hw, pf->hw.aq.arq.len, val);
+
+	val = rd32(&pf->hw, pf->hw.aq.asq.len);
+	oldval = val;
+	if (val & I40E_PF_ATQLEN_ATQVFE_MASK) {
+		dev_info(&pf->pdev->dev, "ASQ VF Error detected\n");
+		val &= ~I40E_PF_ATQLEN_ATQVFE_MASK;
+	}
+	if (val & I40E_PF_ATQLEN_ATQOVFL_MASK) {
+		dev_info(&pf->pdev->dev, "ASQ Overflow Error detected\n");
+		val &= ~I40E_PF_ATQLEN_ATQOVFL_MASK;
+	}
+	if (val & I40E_PF_ATQLEN_ATQCRIT_MASK) {
+		dev_info(&pf->pdev->dev, "ASQ Critical Error detected\n");
+		val &= ~I40E_PF_ATQLEN_ATQCRIT_MASK;
+	}
+	if (oldval != val)
+		wr32(&pf->hw, pf->hw.aq.asq.len, val);
+
 	event.msg_size = I40E_MAX_AQ_BUF_SIZE;
 	event.msg_buf = kzalloc(event.msg_size, GFP_KERNEL);
 	if (!event.msg_buf)
@@ -5128,7 +5325,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb)
 	int ret;
 
 	/* build VSI that owns this VEB, temporarily attached to base VEB */
-	for (v = 0; v < pf->hw.func_caps.num_vsis && !ctl_vsi; v++) {
+	for (v = 0; v < pf->num_alloc_vsi && !ctl_vsi; v++) {
 		if (pf->vsi[v] &&
 		    pf->vsi[v]->veb_idx == veb->idx &&
 		    pf->vsi[v]->flags & I40E_VSI_FLAG_VEB_OWNER) {
@@ -5158,7 +5355,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb)
 		goto end_reconstitute;
 
 	/* create the remaining VSIs attached to this VEB */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (!pf->vsi[v] || pf->vsi[v] == ctl_vsi)
 			continue;
 
@@ -5226,9 +5423,6 @@ static int i40e_get_capabilities(struct i40e_pf *pf)
 		}
 	} while (err);
 
-	/* increment MSI-X count because current FW skips one */
-	pf->hw.func_caps.num_msix_vectors++;
-
 	if (((pf->hw.aq.fw_maj_ver == 2) && (pf->hw.aq.fw_min_ver < 22)) ||
 	    (pf->hw.aq.fw_maj_ver < 2)) {
 		pf->hw.func_caps.num_msix_vectors++;
@@ -5267,15 +5461,14 @@ static int i40e_vsi_clear(struct i40e_vsi *vsi);
 static void i40e_fdir_sb_setup(struct i40e_pf *pf)
 {
 	struct i40e_vsi *vsi;
-	bool new_vsi = false;
-	int err, i;
+	int i;
 
 	if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED))
 		return;
 
 	/* find existing VSI and see if it needs configuring */
 	vsi = NULL;
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			vsi = pf->vsi[i];
 			break;
@@ -5288,47 +5481,12 @@ static void i40e_fdir_sb_setup(struct i40e_pf *pf)
 				     pf->vsi[pf->lan_vsi]->seid, 0);
 		if (!vsi) {
 			dev_info(&pf->pdev->dev, "Couldn't create FDir VSI\n");
-			goto err_vsi;
+			pf->flags &= ~I40E_FLAG_FD_SB_ENABLED;
+			return;
 		}
-		new_vsi = true;
-	}
-	i40e_vsi_setup_irqhandler(vsi, i40e_fdir_clean_ring);
-
-	err = i40e_vsi_setup_tx_resources(vsi);
-	if (err)
-		goto err_setup_tx;
-	err = i40e_vsi_setup_rx_resources(vsi);
-	if (err)
-		goto err_setup_rx;
-
-	if (new_vsi) {
-		char int_name[IFNAMSIZ + 9];
-		err = i40e_vsi_configure(vsi);
-		if (err)
-			goto err_setup_rx;
-		snprintf(int_name, sizeof(int_name) - 1, "%s-fdir",
-			 dev_driver_string(&pf->pdev->dev));
-		err = i40e_vsi_request_irq(vsi, int_name);
-		if (err)
-			goto err_setup_rx;
-		err = i40e_up_complete(vsi);
-		if (err)
-			goto err_up_complete;
-		clear_bit(__I40E_NEEDS_RESTART, &vsi->state);
 	}
 
-	return;
-
-err_up_complete:
-	i40e_down(vsi);
-	i40e_vsi_free_irq(vsi);
-err_setup_rx:
-	i40e_vsi_free_rx_resources(vsi);
-err_setup_tx:
-	i40e_vsi_free_tx_resources(vsi);
-err_vsi:
-	pf->flags &= ~I40E_FLAG_FD_SB_ENABLED;
-	i40e_vsi_clear(vsi);
+	i40e_vsi_setup_irqhandler(vsi, i40e_fdir_clean_ring);
 }
 
 /**
@@ -5340,7 +5498,7 @@ static void i40e_fdir_teardown(struct i40e_pf *pf)
 	int i;
 
 	i40e_fdir_filter_exit(pf);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			i40e_vsi_release(pf->vsi[i]);
 			break;
@@ -5357,7 +5515,7 @@ static void i40e_fdir_teardown(struct i40e_pf *pf)
 static int i40e_prep_for_reset(struct i40e_pf *pf)
 {
 	struct i40e_hw *hw = &pf->hw;
-	i40e_status ret;
+	i40e_status ret = 0;
 	u32 v;
 
 	clear_bit(__I40E_RESET_INTR_RECEIVED, &pf->state);
@@ -5366,13 +5524,10 @@ static int i40e_prep_for_reset(struct i40e_pf *pf)
 
 	dev_dbg(&pf->pdev->dev, "Tearing down internal switch for reset\n");
 
-	if (i40e_check_asq_alive(hw))
-		i40e_vc_notify_reset(pf);
-
 	/* quiesce the VSIs and their queues that are not already DOWN */
 	i40e_pf_quiesce_all_vsi(pf);
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			pf->vsi[v]->seid = 0;
 	}
@@ -5380,22 +5535,40 @@ static int i40e_prep_for_reset(struct i40e_pf *pf)
 	i40e_shutdown_adminq(&pf->hw);
 
 	/* call shutdown HMC */
-	ret = i40e_shutdown_lan_hmc(hw);
-	if (ret) {
-		dev_info(&pf->pdev->dev, "shutdown_lan_hmc failed: %d\n", ret);
-		clear_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state);
+	if (hw->hmc.hmc_obj) {
+		ret = i40e_shutdown_lan_hmc(hw);
+		if (ret) {
+			dev_warn(&pf->pdev->dev,
+				 "shutdown_lan_hmc failed: %d\n", ret);
+			clear_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state);
+		}
 	}
 	return ret;
 }
 
 /**
+ * i40e_send_version - update firmware with driver version
+ * @pf: PF struct
+ */
+static void i40e_send_version(struct i40e_pf *pf)
+{
+	struct i40e_driver_version dv;
+
+	dv.major_version = DRV_VERSION_MAJOR;
+	dv.minor_version = DRV_VERSION_MINOR;
+	dv.build_version = DRV_VERSION_BUILD;
+	dv.subbuild_version = 0;
+	strncpy(dv.driver_string, DRV_VERSION, sizeof(dv.driver_string));
+	i40e_aq_send_driver_version(&pf->hw, &dv, NULL);
+}
+
+/**
  * i40e_reset_and_rebuild - reset and rebuild using a saved config
  * @pf: board private structure
  * @reinit: if the Main VSI needs to re-initialized.
  **/
 static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit)
 {
-	struct i40e_driver_version dv;
 	struct i40e_hw *hw = &pf->hw;
 	i40e_status ret;
 	u32 v;
@@ -5405,8 +5578,10 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit)
 	 * because the reset will make them disappear.
 	 */
 	ret = i40e_pf_reset(hw);
-	if (ret)
+	if (ret) {
 		dev_info(&pf->pdev->dev, "PF reset failed, %d\n", ret);
+		goto end_core_reset;
+	}
 	pf->pfr_count++;
 
 	if (test_bit(__I40E_DOWN, &pf->state))
@@ -5426,6 +5601,7 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit)
 		i40e_verify_eeprom(pf);
 	}
 
+	i40e_clear_pxe_mode(hw);
 	ret = i40e_get_capabilities(pf);
 	if (ret) {
 		dev_info(&pf->pdev->dev, "i40e_get_capabilities failed, %d\n",
@@ -5526,13 +5702,7 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit)
 	}
 
 	/* tell the firmware that we're starting */
-	dv.major_version = DRV_VERSION_MAJOR;
-	dv.minor_version = DRV_VERSION_MINOR;
-	dv.build_version = DRV_VERSION_BUILD;
-	dv.subbuild_version = 0;
-	i40e_aq_send_driver_version(&pf->hw, &dv, NULL);
-
-	dev_info(&pf->pdev->dev, "reset complete\n");
+	i40e_send_version(pf);
 
 end_core_reset:
 	clear_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state);
@@ -5642,7 +5812,6 @@ static void i40e_handle_mdd_event(struct i40e_pf *pf)
  **/
 static void i40e_sync_vxlan_filters_subtask(struct i40e_pf *pf)
 {
-	const int vxlan_hdr_qwords = 4;
 	struct i40e_hw *hw = &pf->hw;
 	i40e_status ret;
 	u8 filter_index;
@@ -5660,7 +5829,6 @@ static void i40e_sync_vxlan_filters_subtask(struct i40e_pf *pf)
 			port = pf->vxlan_ports[i];
 			ret = port ?
 			      i40e_aq_add_udp_tunnel(hw, ntohs(port),
-						     vxlan_hdr_qwords,
 						     I40E_AQC_TUNNEL_TYPE_VXLAN,
 						     &filter_index, NULL)
 			      : i40e_aq_del_udp_tunnel(hw, i, NULL);
@@ -5839,15 +6007,15 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type)
 	 * find next empty vsi slot, looping back around if necessary
 	 */
 	i = pf->next_vsi;
-	while (i < pf->hw.func_caps.num_vsis && pf->vsi[i])
+	while (i < pf->num_alloc_vsi && pf->vsi[i])
 		i++;
-	if (i >= pf->hw.func_caps.num_vsis) {
+	if (i >= pf->num_alloc_vsi) {
 		i = 0;
 		while (i < pf->next_vsi && pf->vsi[i])
 			i++;
 	}
 
-	if (i < pf->hw.func_caps.num_vsis && !pf->vsi[i]) {
+	if (i < pf->num_alloc_vsi && !pf->vsi[i]) {
 		vsi_idx = i;             /* Found one! */
 	} else {
 		ret = -ENODEV;
@@ -5870,6 +6038,7 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type)
 	vsi->netdev_registered = false;
 	vsi->work_limit = I40E_DEFAULT_IRQ_WORK;
 	INIT_LIST_HEAD(&vsi->mac_filter_list);
+	vsi->irqs_ready = false;
 
 	ret = i40e_set_num_rings_in_vsi(vsi);
 	if (ret)
@@ -5987,14 +6156,12 @@ static void i40e_vsi_clear_rings(struct i40e_vsi *vsi)
  **/
 static int i40e_alloc_rings(struct i40e_vsi *vsi)
 {
+	struct i40e_ring *tx_ring, *rx_ring;
 	struct i40e_pf *pf = vsi->back;
 	int i;
 
 	/* Set basic values in the rings to be used later during open() */
 	for (i = 0; i < vsi->alloc_queue_pairs; i++) {
-		struct i40e_ring *tx_ring;
-		struct i40e_ring *rx_ring;
-
 		/* allocate space for both Tx and Rx in one shot */
 		tx_ring = kzalloc(sizeof(struct i40e_ring) * 2, GFP_KERNEL);
 		if (!tx_ring)
@@ -6052,8 +6219,6 @@ static int i40e_reserve_msix_vectors(struct i40e_pf *pf, int vectors)
 		vectors = 0;
 	}
 
-	pf->num_msix_entries = vectors;
-
 	return vectors;
 }
 
@@ -6107,6 +6272,16 @@ static int i40e_init_msix(struct i40e_pf *pf)
 	for (i = 0; i < v_budget; i++)
 		pf->msix_entries[i].entry = i;
 	vec = i40e_reserve_msix_vectors(pf, v_budget);
+
+	if (vec != v_budget) {
+		/* If we have limited resources, we will start with no vectors
+		 * for the special features and then allocate vectors to some
+		 * of these features based on the policy and at the end disable
+		 * the features that did not get any vectors.
+		 */
+		pf->num_vmdq_msix = 0;
+	}
+
 	if (vec < I40E_MIN_MSIX) {
 		pf->flags &= ~I40E_FLAG_MSIX_ENABLED;
 		kfree(pf->msix_entries);
@@ -6115,27 +6290,25 @@ static int i40e_init_msix(struct i40e_pf *pf)
 
 	} else if (vec == I40E_MIN_MSIX) {
 		/* Adjust for minimal MSIX use */
-		dev_info(&pf->pdev->dev, "Features disabled, not enough MSI-X vectors\n");
-		pf->flags &= ~I40E_FLAG_VMDQ_ENABLED;
 		pf->num_vmdq_vsis = 0;
 		pf->num_vmdq_qps = 0;
-		pf->num_vmdq_msix = 0;
 		pf->num_lan_qps = 1;
 		pf->num_lan_msix = 1;
 
 	} else if (vec != v_budget) {
+		/* reserve the misc vector */
+		vec--;
+
 		/* Scale vector usage down */
 		pf->num_vmdq_msix = 1;    /* force VMDqs to only one vector */
-		vec--;                    /* reserve the misc vector */
+		pf->num_vmdq_vsis = 1;
 
 		/* partition out the remaining vectors */
 		switch (vec) {
 		case 2:
-			pf->num_vmdq_vsis = 1;
 			pf->num_lan_msix = 1;
 			break;
 		case 3:
-			pf->num_vmdq_vsis = 1;
 			pf->num_lan_msix = 2;
 			break;
 		default:
@@ -6147,6 +6320,11 @@ static int i40e_init_msix(struct i40e_pf *pf)
 		}
 	}
 
+	if ((pf->flags & I40E_FLAG_VMDQ_ENABLED) &&
+	    (pf->num_vmdq_msix == 0)) {
+		dev_info(&pf->pdev->dev, "VMDq disabled, not enough MSI-X vectors\n");
+		pf->flags &= ~I40E_FLAG_VMDQ_ENABLED;
+	}
 	return err;
 }
 
@@ -6171,7 +6349,7 @@ static int i40e_vsi_alloc_q_vector(struct i40e_vsi *vsi, int v_idx)
 	cpumask_set_cpu(v_idx, &q_vector->affinity_mask);
 	if (vsi->netdev)
 		netif_napi_add(vsi->netdev, &q_vector->napi,
-			       i40e_napi_poll, vsi->work_limit);
+			       i40e_napi_poll, NAPI_POLL_WEIGHT);
 
 	q_vector->rx.latency_range = I40E_LOW_LATENCY;
 	q_vector->tx.latency_range = I40E_LOW_LATENCY;
@@ -6231,7 +6409,7 @@ static void i40e_init_interrupt_scheme(struct i40e_pf *pf)
 		if (err) {
 			pf->flags &= ~(I40E_FLAG_MSIX_ENABLED	|
 				       I40E_FLAG_RSS_ENABLED	|
-				       I40E_FLAG_DCB_ENABLED	|
+				       I40E_FLAG_DCB_CAPABLE	|
 				       I40E_FLAG_SRIOV_ENABLED	|
 				       I40E_FLAG_FD_SB_ENABLED	|
 				       I40E_FLAG_FD_ATR_ENABLED	|
@@ -6364,7 +6542,6 @@ int i40e_reconfig_rss_queues(struct i40e_pf *pf, int queue_count)
 		return 0;
 
 	queue_count = min_t(int, queue_count, pf->rss_size_max);
-	queue_count = rounddown_pow_of_two(queue_count);
 
 	if (queue_count != pf->rss_size) {
 		i40e_prep_for_reset(pf);
@@ -6407,6 +6584,10 @@ static int i40e_sw_init(struct i40e_pf *pf)
 		    I40E_FLAG_MSIX_ENABLED    |
 		    I40E_FLAG_RX_1BUF_ENABLED;
 
+	/* Set default ITR */
+	pf->rx_itr_default = I40E_ITR_DYNAMIC | I40E_ITR_RX_DEF;
+	pf->tx_itr_default = I40E_ITR_DYNAMIC | I40E_ITR_TX_DEF;
+
 	/* Depending on PF configurations, it is possible that the RSS
 	 * maximum might end up larger than the available queues
 	 */
@@ -6416,7 +6597,6 @@ static int i40e_sw_init(struct i40e_pf *pf)
 	if (pf->hw.func_caps.rss) {
 		pf->flags |= I40E_FLAG_RSS_ENABLED;
 		pf->rss_size = min_t(int, pf->rss_size_max, num_online_cpus());
-		pf->rss_size = rounddown_pow_of_two(pf->rss_size);
 	} else {
 		pf->rss_size = 1;
 	}
@@ -6432,8 +6612,12 @@ static int i40e_sw_init(struct i40e_pf *pf)
 	    (pf->hw.func_caps.fd_filters_best_effort > 0)) {
 		pf->flags |= I40E_FLAG_FD_ATR_ENABLED;
 		pf->atr_sample_rate = I40E_DEFAULT_ATR_SAMPLE_RATE;
+		/* Setup a counter for fd_atr per pf */
+		pf->fd_atr_cnt_idx = I40E_FD_ATR_STAT_IDX(pf->hw.pf_id);
 		if (!(pf->flags & I40E_FLAG_MFP_ENABLED)) {
 			pf->flags |= I40E_FLAG_FD_SB_ENABLED;
+			/* Setup a counter for fd_sb per pf */
+			pf->fd_sb_cnt_idx = I40E_FD_SB_STAT_IDX(pf->hw.pf_id);
 		} else {
 			dev_info(&pf->pdev->dev,
 				 "Flow Director Sideband mode Disabled in MFP mode\n");
@@ -6649,6 +6833,96 @@ static void i40e_del_vxlan_port(struct net_device *netdev,
 }
 
 #endif
+#ifdef HAVE_FDB_OPS
+#ifdef USE_CONST_DEV_UC_CHAR
+static int i40e_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
+			    struct net_device *dev,
+			    const unsigned char *addr,
+			    u16 flags)
+#else
+static int i40e_ndo_fdb_add(struct ndmsg *ndm,
+			    struct net_device *dev,
+			    unsigned char *addr,
+			    u16 flags)
+#endif
+{
+	struct i40e_netdev_priv *np = netdev_priv(dev);
+	struct i40e_pf *pf = np->vsi->back;
+	int err = 0;
+
+	if (!(pf->flags & I40E_FLAG_SRIOV_ENABLED))
+		return -EOPNOTSUPP;
+
+	/* Hardware does not support aging addresses so if a
+	 * ndm_state is given only allow permanent addresses
+	 */
+	if (ndm->ndm_state && !(ndm->ndm_state & NUD_PERMANENT)) {
+		netdev_info(dev, "FDB only supports static addresses\n");
+		return -EINVAL;
+	}
+
+	if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
+		err = dev_uc_add_excl(dev, addr);
+	else if (is_multicast_ether_addr(addr))
+		err = dev_mc_add_excl(dev, addr);
+	else
+		err = -EINVAL;
+
+	/* Only return duplicate errors if NLM_F_EXCL is set */
+	if (err == -EEXIST && !(flags & NLM_F_EXCL))
+		err = 0;
+
+	return err;
+}
+
+#ifndef USE_DEFAULT_FDB_DEL_DUMP
+#ifdef USE_CONST_DEV_UC_CHAR
+static int i40e_ndo_fdb_del(struct ndmsg *ndm,
+			    struct net_device *dev,
+			    const unsigned char *addr)
+#else
+static int i40e_ndo_fdb_del(struct ndmsg *ndm,
+			    struct net_device *dev,
+			    unsigned char *addr)
+#endif
+{
+	struct i40e_netdev_priv *np = netdev_priv(dev);
+	struct i40e_pf *pf = np->vsi->back;
+	int err = -EOPNOTSUPP;
+
+	if (ndm->ndm_state & NUD_PERMANENT) {
+		netdev_info(dev, "FDB only supports static addresses\n");
+		return -EINVAL;
+	}
+
+	if (pf->flags & I40E_FLAG_SRIOV_ENABLED) {
+		if (is_unicast_ether_addr(addr))
+			err = dev_uc_del(dev, addr);
+		else if (is_multicast_ether_addr(addr))
+			err = dev_mc_del(dev, addr);
+		else
+			err = -EINVAL;
+	}
+
+	return err;
+}
+
+static int i40e_ndo_fdb_dump(struct sk_buff *skb,
+			     struct netlink_callback *cb,
+			     struct net_device *dev,
+			     int idx)
+{
+	struct i40e_netdev_priv *np = netdev_priv(dev);
+	struct i40e_pf *pf = np->vsi->back;
+
+	if (pf->flags & I40E_FLAG_SRIOV_ENABLED)
+		idx = ndo_dflt_fdb_dump(skb, cb, dev, idx);
+
+	return idx;
+}
+
+#endif /* USE_DEFAULT_FDB_DEL_DUMP */
+#endif /* HAVE_FDB_OPS */
 static const struct net_device_ops i40e_netdev_ops = {
 	.ndo_open		= i40e_open,
 	.ndo_stop		= i40e_close,
@@ -6669,13 +6943,21 @@ static const struct net_device_ops i40e_netdev_ops = {
 	.ndo_set_features	= i40e_set_features,
 	.ndo_set_vf_mac		= i40e_ndo_set_vf_mac,
 	.ndo_set_vf_vlan	= i40e_ndo_set_vf_port_vlan,
-	.ndo_set_vf_tx_rate	= i40e_ndo_set_vf_bw,
+	.ndo_set_vf_rate	= i40e_ndo_set_vf_bw,
 	.ndo_get_vf_config	= i40e_ndo_get_vf_config,
 	.ndo_set_vf_link_state	= i40e_ndo_set_vf_link_state,
+	.ndo_set_vf_spoofchk	= i40e_ndo_set_vf_spoofck,
 #ifdef CONFIG_I40E_VXLAN
 	.ndo_add_vxlan_port	= i40e_add_vxlan_port,
 	.ndo_del_vxlan_port	= i40e_del_vxlan_port,
 #endif
+#ifdef HAVE_FDB_OPS
+	.ndo_fdb_add		= i40e_ndo_fdb_add,
+#ifndef USE_DEFAULT_FDB_DEL_DUMP
+	.ndo_fdb_del		= i40e_ndo_fdb_del,
+	.ndo_fdb_dump		= i40e_ndo_fdb_dump,
+#endif
+#endif
 };
 
 /**
@@ -6720,16 +7002,26 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 			   NETIF_F_TSO_ECN	       |
 			   NETIF_F_TSO6		       |
 			   NETIF_F_RXCSUM	       |
-			   NETIF_F_NTUPLE	       |
 			   NETIF_F_RXHASH	       |
 			   0;
 
+	if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
+		netdev->features |= NETIF_F_NTUPLE;
+
 	/* copy netdev features into list of user selectable features */
 	netdev->hw_features |= netdev->features;
 
 	if (vsi->type == I40E_VSI_MAIN) {
 		SET_NETDEV_DEV(netdev, &pf->pdev->dev);
-		memcpy(mac_addr, hw->mac.perm_addr, ETH_ALEN);
+		ether_addr_copy(mac_addr, hw->mac.perm_addr);
+		/* The following two steps are necessary to prevent reception
+		 * of tagged packets - by default the NVM loads a MAC-VLAN
+		 * filter that will accept any tagged packet.  This is to
+		 * prevent that during normal operations until a specific
+		 * VLAN tag filter has been set.
+		 */
+		i40e_rm_default_mac_filter(vsi, mac_addr);
+		i40e_add_filter(vsi, mac_addr, I40E_VLAN_ANY, false, true);
 	} else {
 		/* relate the VSI_VMDQ name to the VSI_MAIN name */
 		snprintf(netdev->name, IFNAMSIZ, "%sv%%d",
@@ -6739,8 +7031,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	}
 	i40e_add_filter(vsi, brdcast, I40E_VLAN_ANY, false, false);
 
-	memcpy(netdev->dev_addr, mac_addr, ETH_ALEN);
-	memcpy(netdev->perm_addr, mac_addr, ETH_ALEN);
+	ether_addr_copy(netdev->dev_addr, mac_addr);
+	ether_addr_copy(netdev->perm_addr, mac_addr);
 	/* vlan gets same features (except vlan offload)
 	 * after any tweaks for specific VSI types
 	 */
@@ -6772,7 +7064,6 @@ static void i40e_vsi_delete(struct i40e_vsi *vsi)
 		return;
 
 	i40e_aq_delete_element(&vsi->back->hw, vsi->seid, NULL);
-	return;
 }
 
 /**
@@ -6898,6 +7189,13 @@ static int i40e_add_vsi(struct i40e_vsi *vsi)
 
 		ctxt.info.valid_sections |= cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
 		ctxt.info.port_vlan_flags |= I40E_AQ_VSI_PVLAN_MODE_ALL;
+		if (pf->vf[vsi->vf_id].spoofchk) {
+			ctxt.info.valid_sections |=
+				cpu_to_le16(I40E_AQ_VSI_PROP_SECURITY_VALID);
+			ctxt.info.sec_flags |=
+				(I40E_AQ_VSI_SEC_FLAG_ENABLE_VLAN_CHK |
+				 I40E_AQ_VSI_SEC_FLAG_ENABLE_MAC_CHK);
+		}
 		/* Setup the VSI tx/rx queue map for TC0 only for now */
 		i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, true);
 		break;
@@ -6982,11 +7280,7 @@ int i40e_vsi_release(struct i40e_vsi *vsi)
 				unregister_netdev(vsi->netdev);
 			}
 		} else {
-			if (!test_and_set_bit(__I40E_DOWN, &vsi->state))
-				i40e_down(vsi);
-			i40e_vsi_free_irq(vsi);
-			i40e_vsi_free_tx_resources(vsi);
-			i40e_vsi_free_rx_resources(vsi);
+			i40e_vsi_close(vsi);
 		}
 		i40e_vsi_disable_irq(vsi);
 	}
@@ -7013,7 +7307,7 @@ int i40e_vsi_release(struct i40e_vsi *vsi)
 	 * the orphan VEBs yet.  We'll wait for an explicit remove request
 	 * from up the network stack.
 	 */
-	for (n = 0, i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (n = 0, i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] &&
 		    pf->vsi[i]->uplink_seid == uplink_seid &&
 		    (pf->vsi[i]->flags & I40E_VSI_FLAG_VEB_OWNER) == 0) {
@@ -7192,7 +7486,7 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
 
 	if (!veb && uplink_seid != pf->mac_seid) {
 
-		for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+		for (i = 0; i < pf->num_alloc_vsi; i++) {
 			if (pf->vsi[i] && pf->vsi[i]->seid == uplink_seid) {
 				vsi = pf->vsi[i];
 				break;
@@ -7435,7 +7729,7 @@ static void i40e_switch_branch_release(struct i40e_veb *branch)
 	 * NOTE: Removing the last VSI on a VEB has the SIDE EFFECT of removing
 	 *       the VEB itself, so don't use (*branch) after this loop.
 	 */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (!pf->vsi[i])
 			continue;
 		if (pf->vsi[i]->uplink_seid == branch_seid &&
@@ -7487,7 +7781,7 @@ void i40e_veb_release(struct i40e_veb *veb)
 	pf = veb->pf;
 
 	/* find the remaining VSI and check for extras */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->uplink_seid == veb->seid) {
 			n++;
 			vsi = pf->vsi[i];
@@ -7516,8 +7810,6 @@ void i40e_veb_release(struct i40e_veb *veb)
 
 	i40e_aq_delete_element(&pf->hw, veb->seid, NULL);
 	i40e_veb_clear(veb);
-
-	return;
 }
 
 /**
@@ -7601,10 +7893,10 @@ struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags,
 	}
 
 	/* make sure there is such a vsi and uplink */
-	for (vsi_idx = 0; vsi_idx < pf->hw.func_caps.num_vsis; vsi_idx++)
+	for (vsi_idx = 0; vsi_idx < pf->num_alloc_vsi; vsi_idx++)
 		if (pf->vsi[vsi_idx] && pf->vsi[vsi_idx]->seid == vsi_seid)
 			break;
-	if (vsi_idx >= pf->hw.func_caps.num_vsis && vsi_seid != 0) {
+	if (vsi_idx >= pf->num_alloc_vsi && vsi_seid != 0) {
 		dev_info(&pf->pdev->dev, "vsi seid %d not found\n",
 			 vsi_seid);
 		return NULL;
@@ -7639,6 +7931,8 @@ struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags,
 	ret = i40e_add_veb(veb, pf->vsi[vsi_idx]);
 	if (ret)
 		goto err_veb;
+	if (vsi_idx == pf->lan_vsi)
+		pf->lan_veb = veb->idx;
 
 	return veb;
 
@@ -7774,15 +8068,6 @@ int i40e_fetch_switch_configuration(struct i40e_pf *pf, bool printconfig)
 				 "header: %d reported %d total\n",
 				 num_reported, num_total);
 
-		if (num_reported) {
-			int sz = sizeof(*sw_config) * num_reported;
-
-			kfree(pf->sw_config);
-			pf->sw_config = kzalloc(sz, GFP_KERNEL);
-			if (pf->sw_config)
-				memcpy(pf->sw_config, sw_config, sz);
-		}
-
 		for (i = 0; i < num_reported; i++) {
 			struct i40e_aqc_switch_config_element_resp *ele =
 				&sw_config->element[i];
@@ -7949,9 +8234,7 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf)
 	queues_left = pf->hw.func_caps.num_tx_qp;
 
 	if ((queues_left == 1) ||
-	    !(pf->flags & I40E_FLAG_MSIX_ENABLED) ||
-	    !(pf->flags & (I40E_FLAG_RSS_ENABLED | I40E_FLAG_FD_SB_ENABLED |
-			   I40E_FLAG_DCB_ENABLED))) {
+	    !(pf->flags & I40E_FLAG_MSIX_ENABLED)) {
 		/* one qp for PF, no queues for anything else */
 		queues_left = 0;
 		pf->rss_size = pf->num_lan_qps = 1;
@@ -7960,14 +8243,27 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf)
 		pf->flags &= ~(I40E_FLAG_RSS_ENABLED	|
 			       I40E_FLAG_FD_SB_ENABLED	|
 			       I40E_FLAG_FD_ATR_ENABLED	|
-			       I40E_FLAG_DCB_ENABLED	|
+			       I40E_FLAG_DCB_CAPABLE	|
 			       I40E_FLAG_SRIOV_ENABLED	|
 			       I40E_FLAG_VMDQ_ENABLED);
+	} else if (!(pf->flags & (I40E_FLAG_RSS_ENABLED |
+				  I40E_FLAG_FD_SB_ENABLED |
+				  I40E_FLAG_FD_ATR_ENABLED |
+				  I40E_FLAG_DCB_CAPABLE))) {
+		/* one qp for PF */
+		pf->rss_size = pf->num_lan_qps = 1;
+		queues_left -= pf->num_lan_qps;
+
+		pf->flags &= ~(I40E_FLAG_RSS_ENABLED	|
+			       I40E_FLAG_FD_SB_ENABLED	|
+			       I40E_FLAG_FD_ATR_ENABLED	|
+			       I40E_FLAG_DCB_ENABLED	|
+			       I40E_FLAG_VMDQ_ENABLED);
 	} else {
 		/* Not enough queues for all TCs */
-		if ((pf->flags & I40E_FLAG_DCB_ENABLED) &&
+		if ((pf->flags & I40E_FLAG_DCB_CAPABLE) &&
 		    (queues_left < I40E_MAX_TRAFFIC_CLASS)) {
-			pf->flags &= ~I40E_FLAG_DCB_ENABLED;
+			pf->flags &= ~I40E_FLAG_DCB_CAPABLE;
 			dev_info(&pf->pdev->dev, "not enough queues for DCB. DCB is disabled.\n");
 		}
 		pf->num_lan_qps = pf->rss_size_max;
@@ -7998,7 +8294,6 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf)
 	}
 
 	pf->queues_left = queues_left;
-	return;
 }
 
 /**
@@ -8055,12 +8350,13 @@ static void i40e_print_features(struct i40e_pf *pf)
 
 	if (pf->flags & I40E_FLAG_RSS_ENABLED)
 		buf += sprintf(buf, "RSS ");
-	buf += sprintf(buf, "FDir ");
 	if (pf->flags & I40E_FLAG_FD_ATR_ENABLED)
-		buf += sprintf(buf, "ATR ");
-	if (pf->flags & I40E_FLAG_FD_SB_ENABLED)
+		buf += sprintf(buf, "FD_ATR ");
+	if (pf->flags & I40E_FLAG_FD_SB_ENABLED) {
+		buf += sprintf(buf, "FD_SB ");
 		buf += sprintf(buf, "NTUPLE ");
-	if (pf->flags & I40E_FLAG_DCB_ENABLED)
+	}
+	if (pf->flags & I40E_FLAG_DCB_CAPABLE)
 		buf += sprintf(buf, "DCB ");
 	if (pf->flags & I40E_FLAG_PTP)
 		buf += sprintf(buf, "PTP ");
@@ -8083,13 +8379,13 @@ static void i40e_print_features(struct i40e_pf *pf)
  **/
 static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	struct i40e_driver_version dv;
 	struct i40e_pf *pf;
 	struct i40e_hw *hw;
 	static u16 pfs_found;
 	u16 link_status;
 	int err = 0;
 	u32 len;
+	u32 i;
 
 	err = pci_enable_device_mem(pdev);
 	if (err)
@@ -8201,6 +8497,10 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	i40e_verify_eeprom(pf);
 
+	/* Rev 0 hardware was never productized */
+	if (hw->revision_id < 1)
+		dev_warn(&pdev->dev, "This device is a pre-production adapter/LOM. Please be aware there may be issues with your hardware. If you are experiencing problems please contact your Intel or hardware representative who provided you with this hardware.\n");
+
 	i40e_clear_pxe_mode(hw);
 	err = i40e_get_capabilities(pf);
 	if (err)
@@ -8234,7 +8534,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_mac_addr;
 	}
 	dev_info(&pdev->dev, "MAC address: %pM\n", hw->mac.addr);
-	memcpy(hw->mac.perm_addr, hw->mac.addr, ETH_ALEN);
+	ether_addr_copy(hw->mac.perm_addr, hw->mac.addr);
 
 	pci_set_drvdata(pdev, pf);
 	pci_save_state(pdev);
@@ -8242,8 +8542,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	err = i40e_init_pf_dcb(pf);
 	if (err) {
 		dev_info(&pdev->dev, "init_pf_dcb failed: %d\n", err);
-		pf->flags &= ~I40E_FLAG_DCB_ENABLED;
-		goto err_init_dcb;
+		pf->flags &= ~I40E_FLAG_DCB_CAPABLE;
+		/* Continue without DCB enabled */
 	}
 #endif /* CONFIG_I40E_DCB */
 
@@ -8264,10 +8564,18 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	i40e_determine_queue_usage(pf);
 	i40e_init_interrupt_scheme(pf);
 
-	/* Set up the *vsi struct based on the number of VSIs in the HW,
-	 * and set up our local tracking of the MAIN PF vsi.
+	/* The number of VSIs reported by the FW is the minimum guaranteed
+	 * to us; HW supports far more and we share the remaining pool with
+	 * the other PFs. We allocate space for more than the guarantee with
+	 * the understanding that we might not get them all later.
 	 */
-	len = sizeof(struct i40e_vsi *) * pf->hw.func_caps.num_vsis;
+	if (pf->hw.func_caps.num_vsis < I40E_MIN_VSI_ALLOC)
+		pf->num_alloc_vsi = I40E_MIN_VSI_ALLOC;
+	else
+		pf->num_alloc_vsi = pf->hw.func_caps.num_vsis;
+
+	/* Set up the *vsi struct and our local tracking of the MAIN PF vsi. */
+	len = sizeof(struct i40e_vsi *) * pf->num_alloc_vsi;
 	pf->vsi = kzalloc(len, GFP_KERNEL);
 	if (!pf->vsi) {
 		err = -ENOMEM;
@@ -8279,6 +8587,13 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_info(&pdev->dev, "setup_pf_switch failed: %d\n", err);
 		goto err_vsis;
 	}
+	/* if FDIR VSI was set up, start it now */
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
+		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
+			i40e_vsi_open(pf->vsi[i]);
+			break;
+		}
+	}
 
 	/* The main driver is (mostly) up and happy. We need to set this state
 	 * before setting up the misc vector or we get a race and the vector
@@ -8300,6 +8615,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		}
 	}
 
+#ifdef CONFIG_PCI_IOV
 	/* prep for VF support */
 	if ((pf->flags & I40E_FLAG_SRIOV_ENABLED) &&
 	    (pf->flags & I40E_FLAG_MSIX_ENABLED) &&
@@ -8322,17 +8638,14 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 					 err);
 		}
 	}
+#endif /* CONFIG_PCI_IOV */
 
 	pfs_found++;
 
 	i40e_dbg_pf_init(pf);
 
 	/* tell the firmware that we're starting */
-	dv.major_version = DRV_VERSION_MAJOR;
-	dv.minor_version = DRV_VERSION_MINOR;
-	dv.build_version = DRV_VERSION_BUILD;
-	dv.subbuild_version = 0;
-	i40e_aq_send_driver_version(&pf->hw, &dv, NULL);
+	i40e_send_version(pf);
 
 	/* since everything's happy, start the service_task timer */
 	mod_timer(&pf->service_timer,
@@ -8373,9 +8686,6 @@ err_vsis:
 err_switch_setup:
 	i40e_reset_interrupt_capability(pf);
 	del_timer_sync(&pf->service_timer);
-#ifdef CONFIG_I40E_DCB
-err_init_dcb:
-#endif /* CONFIG_I40E_DCB */
 err_mac_addr:
 err_configure_lan_hmc:
 	(void)i40e_shutdown_lan_hmc(hw);
@@ -8456,10 +8766,13 @@ static void i40e_remove(struct pci_dev *pdev)
 	}
 
 	/* shutdown and destroy the HMC */
-	ret_code = i40e_shutdown_lan_hmc(&pf->hw);
-	if (ret_code)
-		dev_warn(&pdev->dev,
-			 "Failed to destroy the HMC resources: %d\n", ret_code);
+	if (pf->hw.hmc.hmc_obj) {
+		ret_code = i40e_shutdown_lan_hmc(&pf->hw);
+		if (ret_code)
+			dev_warn(&pdev->dev,
+				 "Failed to destroy the HMC resources: %d\n",
+				 ret_code);
+	}
 
 	/* shutdown the adminq */
 	ret_code = i40e_shutdown_adminq(&pf->hw);
@@ -8470,7 +8783,7 @@ static void i40e_remove(struct pci_dev *pdev)
 
 	/* Clear all dynamic memory lists of rings, q_vectors, and VSIs */
 	i40e_clear_interrupt_scheme(pf);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i]) {
 			i40e_vsi_clear_rings(pf->vsi[i]);
 			i40e_vsi_clear(pf->vsi[i]);
@@ -8485,7 +8798,6 @@ static void i40e_remove(struct pci_dev *pdev)
 
 	kfree(pf->qp_pile);
 	kfree(pf->irq_pile);
-	kfree(pf->sw_config);
 	kfree(pf->vsi);
 
 	/* force a PF reset to clean anything leftover */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index 9cd57e617959..a430699c41d5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -70,10 +70,12 @@ i40e_status i40e_aq_get_firmware_version(struct i40e_hw *hw,
 				u16 *fw_major_version, u16 *fw_minor_version,
 				u16 *api_major_version, u16 *api_minor_version,
 				struct i40e_asq_cmd_details *cmd_details);
-i40e_status i40e_aq_set_phy_reset(struct i40e_hw *hw,
+i40e_status i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw, u16 vsi_id,
 				struct i40e_asq_cmd_details *cmd_details);
+i40e_status i40e_aq_clear_pxe_mode(struct i40e_hw *hw,
+				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_set_link_restart_an(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_get_link_info(struct i40e_hw *hw,
@@ -157,8 +159,8 @@ i40e_status i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent,
 i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
-				u16 udp_port, u8 header_len,
-				u8 protocol_index, u8 *filter_index,
+				u16 udp_port, u8 protocol_index,
+				u8 *filter_index,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
 				struct i40e_asq_cmd_details *cmd_details);
@@ -167,6 +169,9 @@ i40e_status i40e_aq_delete_element(struct i40e_hw *hw, u16 seid,
 i40e_status i40e_aq_mac_address_write(struct i40e_hw *hw,
 				    u16 flags, u8 *mac_addr,
 				    struct i40e_asq_cmd_details *cmd_details);
+i40e_status i40e_aq_config_vsi_bw_limit(struct i40e_hw *hw,
+				u16 seid, u16 credit, u8 max_credit,
+				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_dcb_updated(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_set_hmc_resource_profile(struct i40e_hw *hw,
@@ -216,6 +221,7 @@ bool i40e_get_link_status(struct i40e_hw *hw);
 i40e_status i40e_get_mac_addr(struct i40e_hw *hw,
 						u8 *mac_addr);
 i40e_status i40e_validate_mac_addr(u8 *mac_addr);
+void i40e_pre_tx_queue_cfg(struct i40e_hw *hw, u32 queue, bool enable);
 /* prototype for functions used for NVM access */
 i40e_status i40e_init_nvm(struct i40e_hw *hw);
 i40e_status i40e_acquire_nvm(struct i40e_hw *hw,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ptp.c b/drivers/net/ethernet/intel/i40e/i40e_ptp.c
index e61e63720800..101f439acda6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ptp.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ptp.c
@@ -48,7 +48,6 @@
 					I40E_PRTTSYN_CTL1_TSYNTYPE_SHIFT)
 #define I40E_PRTTSYN_CTL1_TSYNTYPE_V2  (0x2 << \
 					I40E_PRTTSYN_CTL1_TSYNTYPE_SHIFT)
-#define I40E_PTP_TX_TIMEOUT  (HZ * 15)
 
 /**
  * i40e_ptp_read - Read the PHC time from the device
@@ -217,40 +216,6 @@ static int i40e_ptp_settime(struct ptp_clock_info *ptp,
 }
 
 /**
- * i40e_ptp_tx_work
- * @work: pointer to work struct
- *
- * This work function polls the PRTTSYN_STAT_0.TXTIME bit to determine when a
- * Tx timestamp event has occurred, in order to pass the Tx timestamp value up
- * the stack in the skb.
- */
-static void i40e_ptp_tx_work(struct work_struct *work)
-{
-	struct i40e_pf *pf = container_of(work, struct i40e_pf,
-					  ptp_tx_work);
-	struct i40e_hw *hw = &pf->hw;
-	u32 prttsyn_stat_0;
-
-	if (!pf->ptp_tx_skb)
-		return;
-
-	if (time_is_before_jiffies(pf->ptp_tx_start +
-				   I40E_PTP_TX_TIMEOUT)) {
-		dev_kfree_skb_any(pf->ptp_tx_skb);
-		pf->ptp_tx_skb = NULL;
-		pf->tx_hwtstamp_timeouts++;
-		dev_warn(&pf->pdev->dev, "clearing Tx timestamp hang\n");
-		return;
-	}
-
-	prttsyn_stat_0 = rd32(hw, I40E_PRTTSYN_STAT_0);
-	if (prttsyn_stat_0 & I40E_PRTTSYN_STAT_0_TXTIME_MASK)
-		i40e_ptp_tx_hwtstamp(pf);
-	else
-		schedule_work(&pf->ptp_tx_work);
-}
-
-/**
  * i40e_ptp_enable - Enable/disable ancillary features of the PHC subsystem
  * @ptp: The PTP clock structure
  * @rq: The requested feature to change
@@ -608,7 +573,6 @@ void i40e_ptp_init(struct i40e_pf *pf)
 		u32 regval;
 
 		spin_lock_init(&pf->tmreg_lock);
-		INIT_WORK(&pf->ptp_tx_work, i40e_ptp_tx_work);
 
 		dev_info(&pf->pdev->dev, "%s: added PHC on %s\n", __func__,
 			 netdev->name);
@@ -647,7 +611,6 @@ void i40e_ptp_stop(struct i40e_pf *pf)
 	pf->ptp_tx = false;
 	pf->ptp_rx = false;
 
-	cancel_work_sync(&pf->ptp_tx_work);
 	if (pf->ptp_tx_skb) {
 		dev_kfree_skb_any(pf->ptp_tx_skb);
 		pf->ptp_tx_skb = NULL;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_register.h b/drivers/net/ethernet/intel/i40e/i40e_register.h
index 1d40f425acf1..947de98500f3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_register.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_register.h
@@ -1340,8 +1340,6 @@
 #define I40E_PFINT_ICR0_GPIO_MASK (0x1 << I40E_PFINT_ICR0_GPIO_SHIFT)
 #define I40E_PFINT_ICR0_TIMESYNC_SHIFT 23
 #define I40E_PFINT_ICR0_TIMESYNC_MASK (0x1 << I40E_PFINT_ICR0_TIMESYNC_SHIFT)
-#define I40E_PFINT_ICR0_STORM_DETECT_SHIFT 24
-#define I40E_PFINT_ICR0_STORM_DETECT_MASK (0x1 << I40E_PFINT_ICR0_STORM_DETECT_SHIFT)
 #define I40E_PFINT_ICR0_LINK_STAT_CHANGE_SHIFT 25
 #define I40E_PFINT_ICR0_LINK_STAT_CHANGE_MASK (0x1 << I40E_PFINT_ICR0_LINK_STAT_CHANGE_SHIFT)
 #define I40E_PFINT_ICR0_HMC_ERR_SHIFT 26
@@ -1367,8 +1365,6 @@
 #define I40E_PFINT_ICR0_ENA_GPIO_MASK (0x1 << I40E_PFINT_ICR0_ENA_GPIO_SHIFT)
 #define I40E_PFINT_ICR0_ENA_TIMESYNC_SHIFT 23
 #define I40E_PFINT_ICR0_ENA_TIMESYNC_MASK (0x1 << I40E_PFINT_ICR0_ENA_TIMESYNC_SHIFT)
-#define I40E_PFINT_ICR0_ENA_STORM_DETECT_SHIFT 24
-#define I40E_PFINT_ICR0_ENA_STORM_DETECT_MASK (0x1 << I40E_PFINT_ICR0_ENA_STORM_DETECT_SHIFT)
 #define I40E_PFINT_ICR0_ENA_LINK_STAT_CHANGE_SHIFT 25
 #define I40E_PFINT_ICR0_ENA_LINK_STAT_CHANGE_MASK (0x1 << I40E_PFINT_ICR0_ENA_LINK_STAT_CHANGE_SHIFT)
 #define I40E_PFINT_ICR0_ENA_HMC_ERR_SHIFT 26
@@ -1589,6 +1585,14 @@
 #define I40E_GLLAN_TSOMSK_M 0x000442DC
 #define I40E_GLLAN_TSOMSK_M_TCPMSKM_SHIFT 0
 #define I40E_GLLAN_TSOMSK_M_TCPMSKM_MASK (0xFFF << I40E_GLLAN_TSOMSK_M_TCPMSKM_SHIFT)
+#define I40E_GLLAN_TXPRE_QDIS(_i) (0x000E6500 + ((_i) * 4)) /* i=0..11 */
+#define I40E_GLLAN_TXPRE_QDIS_QINDX_SHIFT 0
+#define I40E_GLLAN_TXPRE_QDIS_QINDX_MASK (0x7FF << I40E_GLLAN_TXPRE_QDIS_QINDX_SHIFT)
+#define I40E_GLLAN_TXPRE_QDIS_SET_QDIS_SHIFT 30
+#define I40E_GLLAN_TXPRE_QDIS_SET_QDIS_MASK (0x1 << I40E_GLLAN_TXPRE_QDIS_SET_QDIS_SHIFT)
+#define I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_SHIFT 31
+#define I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_MASK (0x1 << I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_SHIFT)
+
 #define I40E_PFLAN_QALLOC 0x001C0400
 #define I40E_PFLAN_QALLOC_FIRSTQ_SHIFT 0
 #define I40E_PFLAN_QALLOC_FIRSTQ_MASK (0x7FF << I40E_PFLAN_QALLOC_FIRSTQ_SHIFT)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 9478ddc66caf..e49f31dbd5d8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -24,6 +24,7 @@
  *
  ******************************************************************************/
 
+#include <linux/prefetch.h>
 #include "i40e.h"
 #include "i40e_prototype.h"
 
@@ -61,7 +62,7 @@ int i40e_program_fdir_filter(struct i40e_fdir_filter *fdir_data, u8 *raw_packet,
 
 	/* find existing FDIR VSI */
 	vsi = NULL;
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR)
 			vsi = pf->vsi[i];
 	if (!vsi)
@@ -120,7 +121,7 @@ int i40e_program_fdir_filter(struct i40e_fdir_filter *fdir_data, u8 *raw_packet,
 		dcc |= I40E_TXD_FLTR_QW1_CNT_ENA_MASK;
 		dcc |= ((u32)fdir_data->cnt_index <<
 			I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) &
-		       I40E_TXD_FLTR_QW1_CNTINDEX_MASK;
+			I40E_TXD_FLTR_QW1_CNTINDEX_MASK;
 	}
 
 	fdir_desc->dtype_cmd_cntindex = cpu_to_le32(dcc);
@@ -183,7 +184,6 @@ static int i40e_add_del_fdir_udpv4(struct i40e_vsi *vsi,
 	struct iphdr *ip;
 	bool err = false;
 	int ret;
-	int i;
 	static char packet[] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x08, 0,
 		0x45, 0, 0, 0x1c, 0, 0, 0x40, 0, 0x40, 0x11, 0, 0, 0, 0, 0, 0,
 		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
@@ -199,21 +199,17 @@ static int i40e_add_del_fdir_udpv4(struct i40e_vsi *vsi,
 	ip->saddr = fd_data->src_ip[0];
 	udp->source = fd_data->src_port;
 
-	for (i = I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP;
-	     i <= I40E_FILTER_PCTYPE_NONF_IPV4_UDP; i++) {
-		fd_data->pctype = i;
-		ret = i40e_program_fdir_filter(fd_data, raw_packet, pf, add);
-
-		if (ret) {
-			dev_info(&pf->pdev->dev,
-				 "Filter command send failed for PCTYPE %d (ret = %d)\n",
-				 fd_data->pctype, ret);
-			err = true;
-		} else {
-			dev_info(&pf->pdev->dev,
-				 "Filter OK for PCTYPE %d (ret = %d)\n",
-				 fd_data->pctype, ret);
-		}
+	fd_data->pctype = I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
+	ret = i40e_program_fdir_filter(fd_data, raw_packet, pf, add);
+	if (ret) {
+		dev_info(&pf->pdev->dev,
+			 "Filter command send failed for PCTYPE %d (ret = %d)\n",
+			 fd_data->pctype, ret);
+		err = true;
+	} else {
+		dev_info(&pf->pdev->dev,
+			 "Filter OK for PCTYPE %d (ret = %d)\n",
+			 fd_data->pctype, ret);
 	}
 
 	return err ? -EOPNOTSUPP : 0;
@@ -262,7 +258,7 @@ static int i40e_add_del_fdir_tcpv4(struct i40e_vsi *vsi,
 		}
 	}
 
-	fd_data->pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN;
+	fd_data->pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
 	ret = i40e_program_fdir_filter(fd_data, raw_packet, pf, add);
 
 	if (ret) {
@@ -455,22 +451,20 @@ static void i40e_fd_handle_status(struct i40e_ring *rx_ring,
 
 		/* filter programming failed most likely due to table full */
 		fcnt_prog = i40e_get_current_fd_count(pf);
-		fcnt_avail = pf->hw.fdir_shared_filter_count +
-						       pf->fdir_pf_filter_count;
-
+		fcnt_avail = i40e_get_fd_cnt_all(pf);
 		/* If ATR is running fcnt_prog can quickly change,
 		 * if we are very close to full, it makes sense to disable
 		 * FD ATR/SB and then re-enable it when there is room.
 		 */
 		if (fcnt_prog >= (fcnt_avail - I40E_FDIR_BUFFER_FULL_MARGIN)) {
 			/* Turn off ATR first */
-			if (pf->flags | I40E_FLAG_FD_ATR_ENABLED) {
+			if (pf->flags & I40E_FLAG_FD_ATR_ENABLED) {
 				pf->flags &= ~I40E_FLAG_FD_ATR_ENABLED;
 				dev_warn(&pdev->dev, "FD filter space full, ATR for further flows will be turned off\n");
 				pf->auto_disable_flags |=
 						       I40E_FLAG_FD_ATR_ENABLED;
 				pf->flags |= I40E_FLAG_FDIR_REQUIRES_REINIT;
-			} else if (pf->flags | I40E_FLAG_FD_SB_ENABLED) {
+			} else if (pf->flags & I40E_FLAG_FD_SB_ENABLED) {
 				pf->flags &= ~I40E_FLAG_FD_SB_ENABLED;
 				dev_warn(&pdev->dev, "FD filter space full, new ntuple rules will not be added\n");
 				pf->auto_disable_flags |=
@@ -1199,10 +1193,12 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    u32 rx_error,
 				    u16 rx_ptype)
 {
+	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
+	bool ipv4 = false, ipv6 = false;
 	bool ipv4_tunnel, ipv6_tunnel;
 	__wsum rx_udp_csum;
-	__sum16 csum;
 	struct iphdr *iph;
+	__sum16 csum;
 
 	ipv4_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
 		      (rx_ptype < I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
@@ -1213,29 +1209,57 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	skb->ip_summed = CHECKSUM_NONE;
 
 	/* Rx csum enabled and ip headers found? */
-	if (!(vsi->netdev->features & NETIF_F_RXCSUM &&
-	      rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+	if (!(vsi->netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	/* did the hardware decode the packet and checksum? */
+	if (!(rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+		return;
+
+	/* both known and outer_ip must be set for the below code to work */
+	if (!(decoded.known && decoded.outer_ip))
 		return;
 
+	if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+	    decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4)
+		ipv4 = true;
+	else if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+		 decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6)
+		ipv6 = true;
+
+	if (ipv4 &&
+	    (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
+			 (1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))))
+		goto checksum_fail;
+
 	/* likely incorrect csum if alternate IP extension headers found */
-	if (rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+	if (ipv6 &&
+	    decoded.inner_prot == I40E_RX_PTYPE_INNER_PROT_TCP &&
+	    rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT) &&
+	    rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+		/* don't increment checksum err here, non-fatal err */
 		return;
 
-	/* IP or L4 or outmost IP checksum error */
-	if (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_L4E_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))) {
-		vsi->back->hw_csum_rx_error++;
+	/* there was some L4 error, count error and punt packet to the stack */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT))
+		goto checksum_fail;
+
+	/* handle packets that were not able to be checksummed due
+	 * to arrival speed, in this case the stack can compute
+	 * the csum.
+	 */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_PPRS_SHIFT))
 		return;
-	}
 
+	/* If VXLAN traffic has an outer UDPv4 checksum we need to check
+	 * it in the driver, hardware does not do it for us.
+	 * Since L3L4P bit was set we assume a valid IHL value (>=5)
+	 * so the total length of IPv4 header is IHL*4 bytes
+	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
+	 */
 	if (ipv4_tunnel &&
+	    (decoded.inner_prot != I40E_RX_PTYPE_INNER_PROT_UDP) &&
 	    !(rx_status & (1 << I40E_RX_DESC_STATUS_UDP_0_SHIFT))) {
-		/* If VXLAN traffic has an outer UDPv4 checksum we need to check
-		 * it in the driver, hardware does not do it for us.
-		 * Since L3L4P bit was set we assume a valid IHL value (>=5)
-		 * so the total length of IPv4 header is IHL*4 bytes
-		 */
 		skb->transport_header = skb->mac_header +
 					sizeof(struct ethhdr) +
 					(ip_hdr(skb)->ihl * 4);
@@ -1252,13 +1276,16 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				(skb->len - skb_transport_offset(skb)),
 				IPPROTO_UDP, rx_udp_csum);
 
-		if (udp_hdr(skb)->check != csum) {
-			vsi->back->hw_csum_rx_error++;
-			return;
-		}
+		if (udp_hdr(skb)->check != csum)
+			goto checksum_fail;
 	}
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	return;
+
+checksum_fail:
+	vsi->back->hw_csum_rx_error++;
 }
 
 /**
@@ -1435,6 +1462,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 		/* ERR_MASK will only have valid bits if EOP set */
 		if (unlikely(rx_error & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) {
 			dev_kfree_skb_any(skb);
+			/* TODO: shouldn't we increment a counter indicating the
+			 * drop?
+			 */
 			goto next_desc;
 		}
 
@@ -1665,6 +1695,11 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	dtype_cmd |= I40E_FILTER_PROGRAM_DESC_FD_STATUS_FD_ID <<
 		     I40E_TXD_FLTR_QW1_FD_STATUS_SHIFT;
 
+	dtype_cmd |= I40E_TXD_FLTR_QW1_CNT_ENA_MASK;
+	dtype_cmd |=
+		((u32)pf->fd_atr_cnt_idx << I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) &
+		I40E_TXD_FLTR_QW1_CNTINDEX_MASK;
+
 	fdir_desc->qindex_flex_ptype_vsi = cpu_to_le32(flex_ptype);
 	fdir_desc->dtype_cmd_cntindex = cpu_to_le32(dtype_cmd);
 }
@@ -1825,9 +1860,6 @@ static int i40e_tsyn(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	*cd_type_cmd_tso_mss |= (u64)I40E_TX_CTX_DESC_TSYN <<
 				I40E_TXD_CTX_QW1_CMD_SHIFT;
 
-	pf->ptp_tx_start = jiffies;
-	schedule_work(&pf->ptp_tx_work);
-
 	return 1;
 }
 
@@ -2179,9 +2211,7 @@ static int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 				      struct i40e_ring *tx_ring)
 {
-#if PAGE_SIZE > I40E_MAX_DATA_PER_TXD
 	unsigned int f;
-#endif
 	int count = 0;
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
@@ -2190,12 +2220,9 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 	 *       + 1 desc for context descriptor,
 	 * otherwise try next time
 	 */
-#if PAGE_SIZE > I40E_MAX_DATA_PER_TXD
 	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
 		count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
-#else
-	count += skb_shinfo(skb)->nr_frags;
-#endif
+
 	count += TXD_USE_COUNT(skb_headlen(skb));
 	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
 		tx_ring->tx_stats.tx_busy++;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index d5349698e513..0277894fe1c4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -27,7 +27,7 @@
 #ifndef _I40E_TXRX_H_
 #define _I40E_TXRX_H_
 
-/* Interrupt Throttling and Rate Limiting (storm control) Goodies */
+/* Interrupt Throttling and Rate Limiting Goodies */
 
 #define I40E_MAX_ITR               0x0FF0  /* reg uses 2 usec resolution */
 #define I40E_MIN_ITR               0x0004  /* reg uses 2 usec resolution */
@@ -69,16 +69,11 @@ enum i40e_dyn_idx_t {
 
 /* Supported RSS offloads */
 #define I40E_DEFAULT_RSS_HENA ( \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_SCTP) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) | \
 	((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP) | \
@@ -122,11 +117,11 @@ enum i40e_dyn_idx_t {
 #define i40e_rx_desc i40e_32byte_rx_desc
 
 #define I40E_MIN_TX_LEN		17
-#define I40E_MAX_DATA_PER_TXD	16383	/* aka 16kB - 1 */
+#define I40E_MAX_DATA_PER_TXD	8192
 
 /* Tx Descriptors needed, worst case */
 #define TXD_USE_COUNT(S) DIV_ROUND_UP((S), I40E_MAX_DATA_PER_TXD)
-#define DESC_NEEDED ((MAX_SKB_FRAGS * TXD_USE_COUNT(PAGE_SIZE)) + 4)
+#define DESC_NEEDED (MAX_SKB_FRAGS + 4)
 
 #define I40E_TX_FLAGS_CSUM		(u32)(1)
 #define I40E_TX_FLAGS_HW_VLAN		(u32)(1 << 1)
@@ -184,7 +179,6 @@ enum i40e_ring_state_t {
 	__I40E_TX_DETECT_HANG,
 	__I40E_HANG_CHECK_ARMED,
 	__I40E_RX_PS_ENABLED,
-	__I40E_RX_LRO_ENABLED,
 	__I40E_RX_16BYTE_DESC_ENABLED,
 };
 
@@ -200,12 +194,6 @@ enum i40e_ring_state_t {
 	set_bit(__I40E_TX_DETECT_HANG, &(ring)->state)
 #define clear_check_for_tx_hang(ring) \
 	clear_bit(__I40E_TX_DETECT_HANG, &(ring)->state)
-#define ring_is_lro_enabled(ring) \
-	test_bit(__I40E_RX_LRO_ENABLED, &(ring)->state)
-#define set_ring_lro_enabled(ring) \
-	set_bit(__I40E_RX_LRO_ENABLED, &(ring)->state)
-#define clear_ring_lro_enabled(ring) \
-	clear_bit(__I40E_RX_LRO_ENABLED, &(ring)->state)
 #define ring_is_16byte_desc_enabled(ring) \
 	test_bit(__I40E_RX_16BYTE_DESC_ENABLED, &(ring)->state)
 #define set_ring_16byte_desc_enabled(ring) \
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index 71a968fe557f..9d39ff23c5fb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -36,12 +36,10 @@
 
 /* Device IDs */
 #define I40E_DEV_ID_SFP_XL710		0x1572
-#define I40E_DEV_ID_SFP_X710		0x1573
 #define I40E_DEV_ID_QEMU		0x1574
 #define I40E_DEV_ID_KX_A		0x157F
 #define I40E_DEV_ID_KX_B		0x1580
 #define I40E_DEV_ID_KX_C		0x1581
-#define I40E_DEV_ID_KX_D		0x1582
 #define I40E_DEV_ID_QSFP_A		0x1583
 #define I40E_DEV_ID_QSFP_B		0x1584
 #define I40E_DEV_ID_QSFP_C		0x1585
@@ -60,8 +58,8 @@
 /* Max default timeout in ms, */
 #define I40E_MAX_NVM_TIMEOUT		18000
 
-/* Switch from mc to the 2usec global time (this is the GTIME resolution) */
-#define I40E_MS_TO_GTIME(time)		(((time) * 1000) / 2)
+/* Switch from ms to the 1usec global time (this is the GTIME resolution) */
+#define I40E_MS_TO_GTIME(time)		((time) * 1000)
 
 /* forward declaration */
 struct i40e_hw;
@@ -167,6 +165,9 @@ struct i40e_link_status {
 	u8 loopback;
 	/* is Link Status Event notification to SW enabled */
 	bool lse_enable;
+	u16 max_frame_size;
+	bool crc_enable;
+	u8 pacing;
 };
 
 struct i40e_phy_info {
@@ -409,6 +410,7 @@ struct i40e_driver_version {
 	u8 minor_version;
 	u8 build_version;
 	u8 subbuild_version;
+	u8 driver_string[32];
 };
 
 /* RX Descriptors */
@@ -488,9 +490,6 @@ union i40e_32byte_rx_desc {
 	} wb;  /* writeback */
 };
 
-#define I40E_RXD_QW1_STATUS_SHIFT	0
-#define I40E_RXD_QW1_STATUS_MASK	(0x7FFFUL << I40E_RXD_QW1_STATUS_SHIFT)
-
 enum i40e_rx_desc_status_bits {
 	/* Note: These are predefined bit offsets */
 	I40E_RX_DESC_STATUS_DD_SHIFT		= 0,
@@ -507,9 +506,14 @@ enum i40e_rx_desc_status_bits {
 	I40E_RX_DESC_STATUS_LPBK_SHIFT		= 14,
 	I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT	= 15,
 	I40E_RX_DESC_STATUS_RESERVED_SHIFT	= 16, /* 2 BITS */
-	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 18
+	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 18,
+	I40E_RX_DESC_STATUS_LAST /* this entry must be last!!! */
 };
 
+#define I40E_RXD_QW1_STATUS_SHIFT	0
+#define I40E_RXD_QW1_STATUS_MASK	(((1 << I40E_RX_DESC_STATUS_LAST) - 1) \
+					 << I40E_RXD_QW1_STATUS_SHIFT)
+
 #define I40E_RXD_QW1_STATUS_TSYNINDX_SHIFT   I40E_RX_DESC_STATUS_TSYNINDX_SHIFT
 #define I40E_RXD_QW1_STATUS_TSYNINDX_MASK	(0x3UL << \
 					     I40E_RXD_QW1_STATUS_TSYNINDX_SHIFT)
@@ -537,7 +541,8 @@ enum i40e_rx_desc_error_bits {
 	I40E_RX_DESC_ERROR_IPE_SHIFT		= 3,
 	I40E_RX_DESC_ERROR_L4E_SHIFT		= 4,
 	I40E_RX_DESC_ERROR_EIPE_SHIFT		= 5,
-	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6
+	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6,
+	I40E_RX_DESC_ERROR_PPRS_SHIFT		= 7
 };
 
 enum i40e_rx_desc_error_l3l4e_fcoe_masks {
@@ -658,7 +663,6 @@ enum i40e_rx_desc_ext_status_bits {
 	I40E_RX_DESC_EXT_STATUS_L2TAG3P_SHIFT	= 1,
 	I40E_RX_DESC_EXT_STATUS_FLEXBL_SHIFT	= 2, /* 2 BITS */
 	I40E_RX_DESC_EXT_STATUS_FLEXBH_SHIFT	= 4, /* 2 BITS */
-	I40E_RX_DESC_EXT_STATUS_FTYPE_SHIFT	= 6, /* 3 BITS */
 	I40E_RX_DESC_EXT_STATUS_FDLONGB_SHIFT	= 9,
 	I40E_RX_DESC_EXT_STATUS_FCOELONGB_SHIFT	= 10,
 	I40E_RX_DESC_EXT_STATUS_PELONGB_SHIFT	= 11,
@@ -862,18 +866,14 @@ struct i40e_filter_program_desc {
 
 /* Packet Classifier Types for filters */
 enum i40e_filter_pctype {
-	/* Note: Values 0-28 are reserved for future use */
-	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP	= 29,
-	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP	= 30,
+	/* Note: Values 0-30 are reserved for future use */
 	I40E_FILTER_PCTYPE_NONF_IPV4_UDP		= 31,
-	I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN		= 32,
+	/* Note: Value 32 is reserved for future use */
 	I40E_FILTER_PCTYPE_NONF_IPV4_TCP		= 33,
 	I40E_FILTER_PCTYPE_NONF_IPV4_SCTP		= 34,
 	I40E_FILTER_PCTYPE_NONF_IPV4_OTHER		= 35,
 	I40E_FILTER_PCTYPE_FRAG_IPV4			= 36,
-	/* Note: Values 37-38 are reserved for future use */
-	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP	= 39,
-	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP	= 40,
+	/* Note: Values 37-40 are reserved for future use */
 	I40E_FILTER_PCTYPE_NONF_IPV6_UDP		= 41,
 	I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN		= 42,
 	I40E_FILTER_PCTYPE_NONF_IPV6_TCP		= 43,
@@ -955,6 +955,16 @@ struct i40e_vsi_context {
 	struct i40e_aqc_vsi_properties_data info;
 };
 
+struct i40e_veb_context {
+	u16 seid;
+	u16 uplink_seid;
+	u16 veb_number;
+	u16 vebs_allocated;
+	u16 vebs_unallocated;
+	u16 flags;
+	struct i40e_aqc_get_veb_parameters_completion info;
+};
+
 /* Statistics collected by each port, VSI, VEB, and S-channel */
 struct i40e_eth_stats {
 	u64 rx_bytes;			/* gorc */
@@ -962,8 +972,6 @@ struct i40e_eth_stats {
 	u64 rx_multicast;		/* mprc */
 	u64 rx_broadcast;		/* bprc */
 	u64 rx_discards;		/* rdpc */
-	u64 rx_errors;			/* repc */
-	u64 rx_missed;			/* rmpc */
 	u64 rx_unknown_protocol;	/* rupp */
 	u64 tx_bytes;			/* gotc */
 	u64 tx_unicast;			/* uptc */
@@ -1015,9 +1023,12 @@ struct i40e_hw_port_stats {
 	u64 tx_size_big;		/* ptc9522 */
 	u64 mac_short_packet_dropped;	/* mspdc */
 	u64 checksum_error;		/* xec */
+	/* flow director stats */
+	u64 fd_atr_match;
+	u64 fd_sb_match;
 	/* EEE LPI */
-	bool tx_lpi_status;
-	bool rx_lpi_status;
+	u32 tx_lpi_status;
+	u32 rx_lpi_status;
 	u64 tx_lpi_count;		/* etlpic */
 	u64 rx_lpi_count;		/* erlpic */
 };
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h
index 22a1b69cd646..70951d2edcad 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h
@@ -341,10 +341,6 @@ struct i40e_virtchnl_pf_event {
 	int severity;
 };
 
-/* The following are TBD, not necessary for LAN functionality.
- * I40E_VIRTCHNL_OP_FCOE
- */
-
 /* VF reset states - these are written into the RSTAT register:
  * I40E_VFGEN_RSTAT1 on the PF
  * I40E_VFGEN_RSTAT on the VF
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 02c11a7f7d29..f5b9d2062573 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -29,6 +29,24 @@
 /***********************misc routines*****************************/
 
 /**
+ * i40e_vc_disable_vf
+ * @pf: pointer to the pf info
+ * @vf: pointer to the vf info
+ *
+ * Disable the VF through a SW reset
+ **/
+static inline void i40e_vc_disable_vf(struct i40e_pf *pf, struct i40e_vf *vf)
+{
+	struct i40e_hw *hw = &pf->hw;
+	u32 reg;
+
+	reg = rd32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id));
+	reg |= I40E_VPGEN_VFRTRIG_VFSWR_MASK;
+	wr32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id), reg);
+	i40e_flush(hw);
+}
+
+/**
  * i40e_vc_isvalid_vsi_id
  * @vf: pointer to the vf info
  * @vsi_id: vf relative vsi id
@@ -230,9 +248,8 @@ static int i40e_config_vsi_tx_queue(struct i40e_vf *vf, u16 vsi_idx,
 	tx_ctx.qlen = info->ring_len;
 	tx_ctx.rdylist = le16_to_cpu(pf->vsi[vsi_idx]->info.qs_handle[0]);
 	tx_ctx.rdylist_act = 0;
-	tx_ctx.head_wb_ena = 1;
-	tx_ctx.head_wb_addr = info->dma_ring_addr +
-			      (info->ring_len * sizeof(struct i40e_tx_desc));
+	tx_ctx.head_wb_ena = info->headwb_enabled;
+	tx_ctx.head_wb_addr = info->dma_headwb_addr;
 
 	/* clear the context in the HMC */
 	ret = i40e_clear_lan_tx_queue_context(hw, pf_queue_id);
@@ -336,6 +353,7 @@ static int i40e_config_vsi_rx_queue(struct i40e_vf *vf, u16 vsi_idx,
 	rx_ctx.tphhead_ena = 1;
 	rx_ctx.lrxqthresh = 2;
 	rx_ctx.crcstrip = 1;
+	rx_ctx.prefena = 1;
 
 	/* clear the context in the HMC */
 	ret = i40e_clear_lan_rx_queue_context(hw, pf_queue_id);
@@ -416,6 +434,15 @@ static int i40e_alloc_vsi_res(struct i40e_vf *vf, enum i40e_vsi_type type)
 	if (ret)
 		dev_err(&pf->pdev->dev, "Unable to program ucast filters\n");
 
+	/* Set VF bandwidth if specified */
+	if (vf->tx_rate) {
+		ret = i40e_aq_config_vsi_bw_limit(&pf->hw, vsi->seid,
+						  vf->tx_rate / 50, 0, NULL);
+		if (ret)
+			dev_err(&pf->pdev->dev, "Unable to set tx rate, VF %d, error code %d.\n",
+				vf->vf_id, ret);
+	}
+
 error_alloc_vsi_res:
 	return ret;
 }
@@ -815,6 +842,10 @@ void i40e_free_vfs(struct i40e_pf *pf)
 	kfree(pf->vf);
 	pf->vf = NULL;
 
+	/* This check is for when the driver is unloaded while VFs are
+	 * assigned. Setting the number of VFs to 0 through sysfs is caught
+	 * before this function ever gets called.
+	 */
 	if (!i40e_vfs_are_assigned(pf)) {
 		pci_disable_sriov(pf->pdev);
 		/* Acknowledge VFLR for all VFS. Without this, VFs will fail to
@@ -867,6 +898,7 @@ int i40e_alloc_vfs(struct i40e_pf *pf, u16 num_alloc_vfs)
 		ret = -ENOMEM;
 		goto err_alloc;
 	}
+	pf->vf = vfs;
 
 	/* apply default profile */
 	for (i = 0; i < num_alloc_vfs; i++) {
@@ -876,13 +908,13 @@ int i40e_alloc_vfs(struct i40e_pf *pf, u16 num_alloc_vfs)
 
 		/* assign default capabilities */
 		set_bit(I40E_VIRTCHNL_VF_CAP_L2, &vfs[i].vf_caps);
+		vfs[i].spoofchk = true;
 		/* vf resources get allocated during reset */
 		i40e_reset_vf(&vfs[i], false);
 
 		/* enable vf vplan_qtable mappings */
 		i40e_enable_vf_mappings(&vfs[i]);
 	}
-	pf->vf = vfs;
 	pf->num_alloc_vfs = num_alloc_vfs;
 
 	i40e_enable_pf_switch_lb(pf);
@@ -951,7 +983,12 @@ int i40e_pci_sriov_configure(struct pci_dev *pdev, int num_vfs)
 	if (num_vfs)
 		return i40e_pci_sriov_enable(pdev, num_vfs);
 
-	i40e_free_vfs(pf);
+	if (!i40e_vfs_are_assigned(pf)) {
+		i40e_free_vfs(pf);
+	} else {
+		dev_warn(&pdev->dev, "Unable to free VFs because some are assigned to VMs.\n");
+		return -EINVAL;
+	}
 	return 0;
 }
 
@@ -2022,16 +2059,14 @@ int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
 	}
 
 	/* delete the temporary mac address */
-	i40e_del_filter(vsi, vf->default_lan_addr.addr, 0, true, false);
+	i40e_del_filter(vsi, vf->default_lan_addr.addr, vf->port_vlan_id,
+			true, false);
 
-	/* add the new mac address */
-	f = i40e_add_filter(vsi, mac, 0, true, false);
-	if (!f) {
-		dev_err(&pf->pdev->dev,
-			"Unable to add VF ucast filter\n");
-		ret = -ENOMEM;
-		goto error_param;
-	}
+	/* Delete all the filters for this VSI - we're going to kill it
+	 * anyway.
+	 */
+	list_for_each_entry(f, &vsi->mac_filter_list, list)
+		i40e_del_filter(vsi, f->macaddr, f->vlan, true, false);
 
 	dev_info(&pf->pdev->dev, "Setting MAC %pM on VF %d\n", mac, vf_id);
 	/* program mac filter */
@@ -2040,7 +2075,7 @@ int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
 		ret = -EIO;
 		goto error_param;
 	}
-	memcpy(vf->default_lan_addr.addr, mac, ETH_ALEN);
+	ether_addr_copy(vf->default_lan_addr.addr, mac);
 	vf->pf_set_mac = true;
 	dev_info(&pf->pdev->dev, "Reload the VF driver to make this change effective.\n");
 	ret = 0;
@@ -2088,18 +2123,28 @@ int i40e_ndo_set_vf_port_vlan(struct net_device *netdev,
 		goto error_pvid;
 	}
 
-	if (vsi->info.pvid == 0 && i40e_is_vsi_in_vlan(vsi))
+	if (vsi->info.pvid == 0 && i40e_is_vsi_in_vlan(vsi)) {
 		dev_err(&pf->pdev->dev,
 			"VF %d has already configured VLAN filters and the administrator is requesting a port VLAN override.\nPlease unload and reload the VF driver for this change to take effect.\n",
 			vf_id);
+		/* Administrator Error - knock the VF offline until he does
+		 * the right thing by reconfiguring his network correctly
+		 * and then reloading the VF driver.
+		 */
+		i40e_vc_disable_vf(pf, vf);
+	}
 
 	/* Check for condition where there was already a port VLAN ID
 	 * filter set and now it is being deleted by setting it to zero.
+	 * Additionally check for the condition where there was a port
+	 * VLAN but now there is a new and different port VLAN being set.
 	 * Before deleting all the old VLAN filters we must add new ones
 	 * with -1 (I40E_VLAN_ANY) or otherwise we're left with all our
 	 * MAC addresses deleted.
 	 */
-	if (!(vlan_id || qos) && vsi->info.pvid)
+	if ((!(vlan_id || qos) ||
+	    (vlan_id | qos) != le16_to_cpu(vsi->info.pvid)) &&
+	    vsi->info.pvid)
 		ret = i40e_vsi_add_vlan(vsi, I40E_VLAN_ANY);
 
 	if (vsi->info.pvid) {
@@ -2150,6 +2195,8 @@ error_pvid:
 	return ret;
 }
 
+#define I40E_BW_CREDIT_DIVISOR 50     /* 50Mbps per BW credit */
+#define I40E_MAX_BW_INACTIVE_ACCUM 4  /* device can accumulate 4 credits max */
 /**
  * i40e_ndo_set_vf_bw
  * @netdev: network interface device structure
@@ -2158,9 +2205,76 @@ error_pvid:
  *
  * configure vf tx rate
  **/
-int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate)
+int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate,
+		       int max_tx_rate)
 {
-	return -EOPNOTSUPP;
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_pf *pf = np->vsi->back;
+	struct i40e_vsi *vsi;
+	struct i40e_vf *vf;
+	int speed = 0;
+	int ret = 0;
+
+	/* validate the request */
+	if (vf_id >= pf->num_alloc_vfs) {
+		dev_err(&pf->pdev->dev, "Invalid VF Identifier %d.\n", vf_id);
+		ret = -EINVAL;
+		goto error;
+	}
+
+	if (min_tx_rate) {
+		dev_err(&pf->pdev->dev, "Invalid min tx rate (%d) (greater than 0) specified for vf %d.\n",
+			min_tx_rate, vf_id);
+		return -EINVAL;
+	}
+
+	vf = &(pf->vf[vf_id]);
+	vsi = pf->vsi[vf->lan_vsi_index];
+	if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states)) {
+		dev_err(&pf->pdev->dev, "Uninitialized VF %d.\n", vf_id);
+		ret = -EINVAL;
+		goto error;
+	}
+
+	switch (pf->hw.phy.link_info.link_speed) {
+	case I40E_LINK_SPEED_40GB:
+		speed = 40000;
+		break;
+	case I40E_LINK_SPEED_10GB:
+		speed = 10000;
+		break;
+	case I40E_LINK_SPEED_1GB:
+		speed = 1000;
+		break;
+	default:
+		break;
+	}
+
+	if (max_tx_rate > speed) {
+		dev_err(&pf->pdev->dev, "Invalid max tx rate %d specified for vf %d.",
+			max_tx_rate, vf->vf_id);
+		ret = -EINVAL;
+		goto error;
+	}
+
+	if ((max_tx_rate < 50) && (max_tx_rate > 0)) {
+		dev_warn(&pf->pdev->dev, "Setting max Tx rate to minimum usable value of 50Mbps.\n");
+		max_tx_rate = 50;
+	}
+
+	/* Tx rate credits are in values of 50Mbps, 0 is disabled*/
+	ret = i40e_aq_config_vsi_bw_limit(&pf->hw, vsi->seid,
+					  max_tx_rate / I40E_BW_CREDIT_DIVISOR,
+					  I40E_MAX_BW_INACTIVE_ACCUM, NULL);
+	if (ret) {
+		dev_err(&pf->pdev->dev, "Unable to set max tx rate, error code %d.\n",
+			ret);
+		ret = -EIO;
+		goto error;
+	}
+	vf->tx_rate = max_tx_rate;
+error:
+	return ret;
 }
 
 /**
@@ -2200,10 +2314,18 @@ int i40e_ndo_get_vf_config(struct net_device *netdev,
 
 	memcpy(&ivi->mac, vf->default_lan_addr.addr, ETH_ALEN);
 
-	ivi->tx_rate = 0;
+	ivi->max_tx_rate = vf->tx_rate;
+	ivi->min_tx_rate = 0;
 	ivi->vlan = le16_to_cpu(vsi->info.pvid) & I40E_VLAN_MASK;
 	ivi->qos = (le16_to_cpu(vsi->info.pvid) & I40E_PRIORITY_MASK) >>
 		   I40E_VLAN_PRIORITY_SHIFT;
+	if (vf->link_forced == false)
+		ivi->linkstate = IFLA_VF_LINK_STATE_AUTO;
+	else if (vf->link_up == true)
+		ivi->linkstate = IFLA_VF_LINK_STATE_ENABLE;
+	else
+		ivi->linkstate = IFLA_VF_LINK_STATE_DISABLE;
+	ivi->spoofchk = vf->spoofchk;
 	ret = 0;
 
 error_param:
@@ -2270,3 +2392,50 @@ int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link)
 error_out:
 	return ret;
 }
+
+/**
+ * i40e_ndo_set_vf_spoofchk
+ * @netdev: network interface device structure
+ * @vf_id: vf identifier
+ * @enable: flag to enable or disable feature
+ *
+ * Enable or disable VF spoof checking
+ **/
+int i40e_ndo_set_vf_spoofck(struct net_device *netdev, int vf_id, bool enable)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_vsi_context ctxt;
+	struct i40e_hw *hw = &pf->hw;
+	struct i40e_vf *vf;
+	int ret = 0;
+
+	/* validate the request */
+	if (vf_id >= pf->num_alloc_vfs) {
+		dev_err(&pf->pdev->dev, "Invalid VF Identifier %d\n", vf_id);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	vf = &(pf->vf[vf_id]);
+
+	if (enable == vf->spoofchk)
+		goto out;
+
+	vf->spoofchk = enable;
+	memset(&ctxt, 0, sizeof(ctxt));
+	ctxt.seid = pf->vsi[vf->lan_vsi_index]->seid;
+	ctxt.pf_num = pf->hw.pf_id;
+	ctxt.info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_SECURITY_VALID);
+	if (enable)
+		ctxt.info.sec_flags |= I40E_AQ_VSI_SEC_FLAG_ENABLE_MAC_CHK;
+	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
+	if (ret) {
+		dev_err(&pf->pdev->dev, "Error %d updating VSI parameters\n",
+			ret);
+		ret = -EIO;
+	}
+out:
+	return ret;
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
index 389c47f396d5..63e7e0d81ad2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
@@ -98,8 +98,10 @@ struct i40e_vf {
 
 	unsigned long vf_caps;	/* vf's adv. capabilities */
 	unsigned long vf_states;	/* vf's runtime states */
+	unsigned int tx_rate;	/* Tx bandwidth limit in Mbps */
 	bool link_forced;
 	bool link_up;		/* only valid if vf link is forced */
+	bool spoofchk;
 };
 
 void i40e_free_vfs(struct i40e_pf *pf);
@@ -115,10 +117,12 @@ void i40e_vc_notify_vf_reset(struct i40e_vf *vf);
 int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac);
 int i40e_ndo_set_vf_port_vlan(struct net_device *netdev,
 			      int vf_id, u16 vlan_id, u8 qos);
-int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate);
+int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate,
+		       int max_tx_rate);
 int i40e_ndo_get_vf_config(struct net_device *netdev,
 			   int vf_id, struct ifla_vf_info *ivi);
 int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link);
+int i40e_ndo_set_vf_spoofck(struct net_device *netdev, int vf_id, bool enable);
 
 void i40e_vc_notify_link_state(struct i40e_pf *pf);
 void i40e_vc_notify_reset(struct i40e_pf *pf);
diff --git a/drivers/net/ethernet/intel/i40evf/Makefile b/drivers/net/ethernet/intel/i40evf/Makefile
index e09be37a07a8..3a423836a565 100644
--- a/drivers/net/ethernet/intel/i40evf/Makefile
+++ b/drivers/net/ethernet/intel/i40evf/Makefile
@@ -1,7 +1,7 @@
 ################################################################################
 #
 # Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
-# Copyright(c) 2013 Intel Corporation.
+# Copyright(c) 2013 - 2014 Intel Corporation.
 #
 # This program is free software; you can redistribute it and/or modify it
 # under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
 # FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 # more details.
 #
+# You should have received a copy of the GNU General Public License along
+# with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
 # The full GNU General Public License is included in this distribution in
 # the file called "COPYING".
 #
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c
index 5470ce95936e..eb67cce3e8f9 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -28,6 +31,16 @@
 #include "i40e_prototype.h"
 
 /**
+ * i40e_is_nvm_update_op - return true if this is an NVM update operation
+ * @desc: API request descriptor
+ **/
+static inline bool i40e_is_nvm_update_op(struct i40e_aq_desc *desc)
+{
+	return (desc->opcode == i40e_aqc_opc_nvm_erase) ||
+	       (desc->opcode == i40e_aqc_opc_nvm_update);
+}
+
+/**
  *  i40e_adminq_init_regs - Initialize AdminQ registers
  *  @hw: pointer to the hardware structure
  *
@@ -276,8 +289,11 @@ static void i40e_free_asq_bufs(struct i40e_hw *hw)
  *
  *  Configure base address and length registers for the transmit queue
  **/
-static void i40e_config_asq_regs(struct i40e_hw *hw)
+static i40e_status i40e_config_asq_regs(struct i40e_hw *hw)
 {
+	i40e_status ret_code = 0;
+	u32 reg = 0;
+
 	if (hw->mac.type == I40E_MAC_VF) {
 		/* configure the transmit queue */
 		wr32(hw, I40E_VF_ATQBAH1,
@@ -286,6 +302,7 @@ static void i40e_config_asq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.asq.desc_buf.pa));
 		wr32(hw, I40E_VF_ATQLEN1, (hw->aq.num_asq_entries |
 					  I40E_VF_ATQLEN1_ATQENABLE_MASK));
+		reg = rd32(hw, I40E_VF_ATQBAL1);
 	} else {
 		/* configure the transmit queue */
 		wr32(hw, I40E_PF_ATQBAH,
@@ -294,7 +311,14 @@ static void i40e_config_asq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.asq.desc_buf.pa));
 		wr32(hw, I40E_PF_ATQLEN, (hw->aq.num_asq_entries |
 					  I40E_PF_ATQLEN_ATQENABLE_MASK));
+		reg = rd32(hw, I40E_PF_ATQBAL);
 	}
+
+	/* Check one register to verify that config was applied */
+	if (reg != lower_32_bits(hw->aq.asq.desc_buf.pa))
+		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+
+	return ret_code;
 }
 
 /**
@@ -303,8 +327,11 @@ static void i40e_config_asq_regs(struct i40e_hw *hw)
  *
  * Configure base address and length registers for the receive (event queue)
  **/
-static void i40e_config_arq_regs(struct i40e_hw *hw)
+static i40e_status i40e_config_arq_regs(struct i40e_hw *hw)
 {
+	i40e_status ret_code = 0;
+	u32 reg = 0;
+
 	if (hw->mac.type == I40E_MAC_VF) {
 		/* configure the receive queue */
 		wr32(hw, I40E_VF_ARQBAH1,
@@ -313,6 +340,7 @@ static void i40e_config_arq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.arq.desc_buf.pa));
 		wr32(hw, I40E_VF_ARQLEN1, (hw->aq.num_arq_entries |
 					  I40E_VF_ARQLEN1_ARQENABLE_MASK));
+		reg = rd32(hw, I40E_VF_ARQBAL1);
 	} else {
 		/* configure the receive queue */
 		wr32(hw, I40E_PF_ARQBAH,
@@ -321,10 +349,17 @@ static void i40e_config_arq_regs(struct i40e_hw *hw)
 		    lower_32_bits(hw->aq.arq.desc_buf.pa));
 		wr32(hw, I40E_PF_ARQLEN, (hw->aq.num_arq_entries |
 					  I40E_PF_ARQLEN_ARQENABLE_MASK));
+		reg = rd32(hw, I40E_PF_ARQBAL);
 	}
 
 	/* Update tail in the HW to post pre-allocated buffers */
 	wr32(hw, hw->aq.arq.tail, hw->aq.num_arq_entries - 1);
+
+	/* Check one register to verify that config was applied */
+	if (reg != lower_32_bits(hw->aq.arq.desc_buf.pa))
+		ret_code = I40E_ERR_ADMIN_QUEUE_ERROR;
+
+	return ret_code;
 }
 
 /**
@@ -372,7 +407,9 @@ static i40e_status i40e_init_asq(struct i40e_hw *hw)
 		goto init_adminq_free_rings;
 
 	/* initialize base registers */
-	i40e_config_asq_regs(hw);
+	ret_code = i40e_config_asq_regs(hw);
+	if (ret_code)
+		goto init_adminq_free_rings;
 
 	/* success! */
 	goto init_adminq_exit;
@@ -429,7 +466,9 @@ static i40e_status i40e_init_arq(struct i40e_hw *hw)
 		goto init_adminq_free_rings;
 
 	/* initialize base registers */
-	i40e_config_arq_regs(hw);
+	ret_code = i40e_config_arq_regs(hw);
+	if (ret_code)
+		goto init_adminq_free_rings;
 
 	/* success! */
 	goto init_adminq_exit;
@@ -659,6 +698,12 @@ i40e_status i40evf_asq_send_command(struct i40e_hw *hw,
 		goto asq_send_command_exit;
 	}
 
+	if (i40e_is_nvm_update_op(desc) && hw->aq.nvm_busy) {
+		i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, "AQTX: NVM busy.\n");
+		status = I40E_ERR_NVM;
+		goto asq_send_command_exit;
+	}
+
 	details = I40E_ADMINQ_DETAILS(hw->aq.asq, hw->aq.asq.next_to_use);
 	if (cmd_details) {
 		*details = *cmd_details;
@@ -786,6 +831,9 @@ i40e_status i40evf_asq_send_command(struct i40e_hw *hw,
 		hw->aq.asq_last_status = (enum i40e_admin_queue_err)retval;
 	}
 
+	if (i40e_is_nvm_update_op(desc))
+		hw->aq.nvm_busy = true;
+
 	/* update the error if time out occurred */
 	if ((!cmd_completed) &&
 	    (!details->async && !details->postpone)) {
@@ -880,6 +928,9 @@ i40e_status i40evf_clean_arq_element(struct i40e_hw *hw,
 			       e->msg_size);
 	}
 
+	if (i40e_is_nvm_update_op(&e->desc))
+		hw->aq.nvm_busy = false;
+
 	/* Restore the original datalen and buffer address in the desc,
 	 * FW updates datalen to indicate the event message
 	 * size
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq.h
index 8f72c31d95cc..e3472c62e155 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -87,6 +90,7 @@ struct i40e_adminq_info {
 	u16 fw_min_ver;                 /* firmware minor version */
 	u16 api_maj_ver;                /* api major version */
 	u16 api_min_ver;                /* api minor version */
+	bool nvm_busy;
 
 	struct mutex asq_mutex; /* Send queue lock */
 	struct mutex arq_mutex; /* Receive queue lock */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 97662b6bd98a..e656ea7a7920 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -31,7 +34,7 @@
  */
 
 #define I40E_FW_API_VERSION_MAJOR  0x0001
-#define I40E_FW_API_VERSION_MINOR  0x0001
+#define I40E_FW_API_VERSION_MINOR  0x0002
 #define I40E_FW_API_VERSION_A0_MINOR  0x0000
 
 struct i40e_aq_desc {
@@ -121,6 +124,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_version      = 0x0001,
 	i40e_aqc_opc_driver_version   = 0x0002,
 	i40e_aqc_opc_queue_shutdown   = 0x0003,
+	i40e_aqc_opc_set_pf_context   = 0x0004,
 
 	/* resource ownership */
 	i40e_aqc_opc_request_resource = 0x0008,
@@ -180,9 +184,6 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_add_mirror_rule    = 0x0260,
 	i40e_aqc_opc_delete_mirror_rule = 0x0261,
 
-	i40e_aqc_opc_set_storm_control_config = 0x0280,
-	i40e_aqc_opc_get_storm_control_config = 0x0281,
-
 	/* DCB commands */
 	i40e_aqc_opc_dcb_ignore_pfc = 0x0301,
 	i40e_aqc_opc_dcb_updated    = 0x0302,
@@ -205,6 +206,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_query_switching_comp_bw_config        = 0x041A,
 	i40e_aqc_opc_suspend_port_tx                       = 0x041B,
 	i40e_aqc_opc_resume_port_tx                        = 0x041C,
+	i40e_aqc_opc_configure_partition_bw                = 0x041D,
 
 	/* hmc */
 	i40e_aqc_opc_query_hmc_resource_profile = 0x0500,
@@ -222,13 +224,15 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_partner_advt    = 0x0616,
 	i40e_aqc_opc_set_lb_modes        = 0x0618,
 	i40e_aqc_opc_get_phy_wol_caps    = 0x0621,
-	i40e_aqc_opc_set_phy_reset       = 0x0622,
+	i40e_aqc_opc_set_phy_debug	 = 0x0622,
 	i40e_aqc_opc_upload_ext_phy_fm   = 0x0625,
 
 	/* NVM commands */
-	i40e_aqc_opc_nvm_read   = 0x0701,
-	i40e_aqc_opc_nvm_erase  = 0x0702,
-	i40e_aqc_opc_nvm_update = 0x0703,
+	i40e_aqc_opc_nvm_read         = 0x0701,
+	i40e_aqc_opc_nvm_erase        = 0x0702,
+	i40e_aqc_opc_nvm_update       = 0x0703,
+	i40e_aqc_opc_nvm_config_read  = 0x0704,
+	i40e_aqc_opc_nvm_config_write = 0x0705,
 
 	/* virtualization commands */
 	i40e_aqc_opc_send_msg_to_pf   = 0x0801,
@@ -270,8 +274,6 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_debug_set_mode         = 0xFF01,
 	i40e_aqc_opc_debug_read_reg         = 0xFF03,
 	i40e_aqc_opc_debug_write_reg        = 0xFF04,
-	i40e_aqc_opc_debug_read_reg_sg      = 0xFF05,
-	i40e_aqc_opc_debug_write_reg_sg     = 0xFF06,
 	i40e_aqc_opc_debug_modify_reg       = 0xFF07,
 	i40e_aqc_opc_debug_dump_internals   = 0xFF08,
 	i40e_aqc_opc_debug_modify_internals = 0xFF09,
@@ -339,6 +341,14 @@ struct i40e_aqc_queue_shutdown {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_queue_shutdown);
 
+/* Set PF context (0x0004, direct) */
+struct i40e_aqc_set_pf_context {
+	u8	pf_id;
+	u8	reserved[15];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_pf_context);
+
 /* Request resource ownership (direct 0x0008)
  * Release resource ownership (direct 0x0009)
  */
@@ -678,7 +688,6 @@ struct i40e_aqc_add_get_update_vsi {
 #define I40E_AQ_VSI_TYPE_PF             0x2
 #define I40E_AQ_VSI_TYPE_EMP_MNG        0x3
 #define I40E_AQ_VSI_FLAG_CASCADED_PV    0x4
-#define I40E_AQ_VSI_FLAG_CLOUD_VSI      0x8
 	__le32 addr_high;
 	__le32 addr_low;
 };
@@ -1040,7 +1049,9 @@ struct i40e_aqc_set_vsi_promiscuous_modes {
 #define I40E_AQC_SET_VSI_PROMISC_VLAN        0x10
 	__le16 seid;
 #define I40E_AQC_VSI_PROM_CMD_SEID_MASK      0x3FF
-	u8     reserved[10];
+	__le16 vlan_tag;
+#define I40E_AQC_SET_VSI_VLAN_VALID          0x8000
+	u8     reserved[8];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_vsi_promiscuous_modes);
@@ -1289,27 +1300,6 @@ struct i40e_aqc_add_delete_mirror_rule_completion {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_add_delete_mirror_rule_completion);
 
-/* Set Storm Control Configuration (direct 0x0280)
- * Get Storm Control Configuration (direct 0x0281)
- *    the command and response use the same descriptor structure
- */
-struct i40e_aqc_set_get_storm_control_config {
-	__le32 broadcast_threshold;
-	__le32 multicast_threshold;
-	__le32 control_flags;
-#define I40E_AQC_STORM_CONTROL_MDIPW            0x01
-#define I40E_AQC_STORM_CONTROL_MDICW            0x02
-#define I40E_AQC_STORM_CONTROL_BDIPW            0x04
-#define I40E_AQC_STORM_CONTROL_BDICW            0x08
-#define I40E_AQC_STORM_CONTROL_BIDU             0x10
-#define I40E_AQC_STORM_CONTROL_INTERVAL_SHIFT   8
-#define I40E_AQC_STORM_CONTROL_INTERVAL_MASK    (0x3FF << \
-					I40E_AQC_STORM_CONTROL_INTERVAL_SHIFT)
-	u8     reserved[4];
-};
-
-I40E_CHECK_CMD_LENGTH(i40e_aqc_set_get_storm_control_config);
-
 /* DCB 0x03xx*/
 
 /* PFC Ignore (direct 0x0301)
@@ -1427,11 +1417,12 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_configure_switching_comp_bw_limit);
 struct i40e_aqc_configure_switching_comp_ets_data {
 	u8     reserved[4];
 	u8     tc_valid_bits;
-	u8     reserved1;
+	u8     seepage;
+#define I40E_AQ_ETS_SEEPAGE_EN_MASK     0x1
 	u8     tc_strict_priority_flags;
-	u8     reserved2[17];
+	u8     reserved1[17];
 	u8     tc_bw_share_credits[8];
-	u8     reserved3[96];
+	u8     reserved2[96];
 };
 
 /* Configure Switching Component Bandwidth Limits per Tc (indirect 0x0416) */
@@ -1499,6 +1490,15 @@ struct i40e_aqc_query_switching_comp_bw_config_resp {
  * (direct 0x041B and 0x041C) uses the generic SEID struct
  */
 
+/* Configure partition BW
+ * (indirect 0x041D)
+ */
+struct i40e_aqc_configure_partition_bw_data {
+	__le16 pf_valid_bits;
+	u8     min_bw[16];      /* guaranteed bandwidth */
+	u8     max_bw[16];      /* bandwidth limit */
+};
+
 /* Get and set the active HMC resource profile and status.
  * (direct 0x0500) and (direct 0x0501)
  */
@@ -1539,6 +1539,8 @@ enum i40e_aq_phy_type {
 	I40E_PHY_TYPE_XLPPI			= 0x9,
 	I40E_PHY_TYPE_40GBASE_CR4_CU		= 0xA,
 	I40E_PHY_TYPE_10GBASE_CR1_CU		= 0xB,
+	I40E_PHY_TYPE_10GBASE_AOC		= 0xC,
+	I40E_PHY_TYPE_40GBASE_AOC		= 0xD,
 	I40E_PHY_TYPE_100BASE_TX		= 0x11,
 	I40E_PHY_TYPE_1000BASE_T		= 0x12,
 	I40E_PHY_TYPE_10GBASE_T			= 0x13,
@@ -1549,7 +1551,10 @@ enum i40e_aq_phy_type {
 	I40E_PHY_TYPE_40GBASE_CR4		= 0x18,
 	I40E_PHY_TYPE_40GBASE_SR4		= 0x19,
 	I40E_PHY_TYPE_40GBASE_LR4		= 0x1A,
-	I40E_PHY_TYPE_20GBASE_KR2		= 0x1B,
+	I40E_PHY_TYPE_1000BASE_SX		= 0x1B,
+	I40E_PHY_TYPE_1000BASE_LX		= 0x1C,
+	I40E_PHY_TYPE_1000BASE_T_OPTICAL	= 0x1D,
+	I40E_PHY_TYPE_20GBASE_KR2		= 0x1E,
 	I40E_PHY_TYPE_MAX
 };
 
@@ -1583,11 +1588,8 @@ struct i40e_aq_get_phy_abilities_resp {
 #define I40E_AQ_PHY_FLAG_PAUSE_TX         0x01
 #define I40E_AQ_PHY_FLAG_PAUSE_RX         0x02
 #define I40E_AQ_PHY_FLAG_LOW_POWER        0x04
-#define I40E_AQ_PHY_FLAG_AN_SHIFT         3
-#define I40E_AQ_PHY_FLAG_AN_MASK          (0x3 << I40E_AQ_PHY_FLAG_AN_SHIFT)
-#define I40E_AQ_PHY_FLAG_AN_OFF           0x00 /* link forced on */
-#define I40E_AQ_PHY_FLAG_AN_OFF_LINK_DOWN 0x01
-#define I40E_AQ_PHY_FLAG_AN_ON            0x02
+#define I40E_AQ_PHY_LINK_ENABLED		  0x08
+#define I40E_AQ_PHY_AN_ENABLED			  0x10
 #define I40E_AQ_PHY_FLAG_MODULE_QUAL      0x20
 	__le16 eee_capability;
 #define I40E_AQ_EEE_100BASE_TX       0x0002
@@ -1696,6 +1698,7 @@ struct i40e_aqc_get_link_status {
 #define I40E_AQ_LINK_TX_ACTIVE       0x00
 #define I40E_AQ_LINK_TX_DRAINED      0x01
 #define I40E_AQ_LINK_TX_FLUSHED      0x03
+#define I40E_AQ_LINK_FORCED_40G      0x10
 	u8     loopback;         /* use defines from i40e_aqc_set_lb_mode */
 	__le16 max_frame_size;
 	u8     config;
@@ -1747,14 +1750,21 @@ struct i40e_aqc_set_lb_mode {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_lb_mode);
 
-/* Set PHY Reset command (0x0622) */
-struct i40e_aqc_set_phy_reset {
-	u8     reset_flags;
-#define I40E_AQ_PHY_RESET_REQUEST  0x02
+/* Set PHY Debug command (0x0622) */
+struct i40e_aqc_set_phy_debug {
+	u8     command_flags;
+#define I40E_AQ_PHY_DEBUG_RESET_INTERNAL	0x02
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SHIFT	2
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_MASK	(0x03 << \
+					I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SHIFT)
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_NONE	0x00
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_HARD	0x01
+#define I40E_AQ_PHY_DEBUG_RESET_EXTERNAL_SOFT	0x02
+#define I40E_AQ_PHY_DEBUG_DISABLE_LINK_FW	0x10
 	u8     reserved[15];
 };
 
-I40E_CHECK_CMD_LENGTH(i40e_aqc_set_phy_reset);
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_phy_debug);
 
 enum i40e_aq_phy_reg_type {
 	I40E_AQC_PHY_REG_INTERNAL         = 0x1,
@@ -1779,6 +1789,47 @@ struct i40e_aqc_nvm_update {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_update);
 
+/* NVM Config Read (indirect 0x0704) */
+struct i40e_aqc_nvm_config_read {
+	__le16 cmd_flags;
+#define ANVM_SINGLE_OR_MULTIPLE_FEATURES_MASK	1
+#define ANVM_READ_SINGLE_FEATURE		0
+#define ANVM_READ_MULTIPLE_FEATURES		1
+	__le16 element_count;
+	__le16 element_id;		/* Feature/field ID */
+	u8     reserved[2];
+	__le32 address_high;
+	__le32 address_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_config_read);
+
+/* NVM Config Write (indirect 0x0705) */
+struct i40e_aqc_nvm_config_write {
+	__le16 cmd_flags;
+	__le16 element_count;
+	u8     reserved[4];
+	__le32 address_high;
+	__le32 address_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_nvm_config_write);
+
+struct i40e_aqc_nvm_config_data_feature {
+	__le16 feature_id;
+	__le16 instance_id;
+	__le16 feature_options;
+	__le16 feature_selection;
+};
+
+struct i40e_aqc_nvm_config_data_immediate_field {
+#define ANVM_FEATURE_OR_IMMEDIATE_MASK	0x2
+	__le16 field_id;
+	__le16 instance_id;
+	__le16 field_options;
+	__le16 field_value;
+};
+
 /* Send to PF command (indirect 0x0801) id is only used by PF
  * Send to VF command (indirect 0x0802) id is only used by PF
  * Send to Peer PF command (indirect 0x0803)
@@ -1948,19 +1999,12 @@ I40E_CHECK_CMD_LENGTH(i40e_aqc_lldp_start);
 /* Add Udp Tunnel command and completion (direct 0x0B00) */
 struct i40e_aqc_add_udp_tunnel {
 	__le16 udp_port;
-	u8     header_len; /* in DWords, 1 to 15 */
+	u8     reserved0[3];
 	u8     protocol_type;
-#define I40E_AQC_TUNNEL_TYPE_TEREDO	0x0
-#define I40E_AQC_TUNNEL_TYPE_VXLAN	0x2
-#define I40E_AQC_TUNNEL_TYPE_NGE	0x3
-	u8     variable_udp_length;
-#define I40E_AQC_TUNNEL_FIXED_UDP_LENGTH	0x0
-#define I40E_AQC_TUNNEL_VARIABLE_UDP_LENGTH	0x1
-	u8		udp_key_index;
-#define I40E_AQC_TUNNEL_KEY_INDEX_VXLAN			0x0
-#define I40E_AQC_TUNNEL_KEY_INDEX_NGE			0x1
-#define I40E_AQC_TUNNEL_KEY_INDEX_PROPRIETARY_UDP	0x2
-	u8		reserved[10];
+#define I40E_AQC_TUNNEL_TYPE_VXLAN	0x00
+#define I40E_AQC_TUNNEL_TYPE_NGE	0x01
+#define I40E_AQC_TUNNEL_TYPE_TEREDO	0x10
+	u8     reserved1[10];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_add_udp_tunnel);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_alloc.h b/drivers/net/ethernet/intel/i40evf/i40e_alloc.h
index d8654fb9e525..8e6a6dd9212b 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_alloc.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_alloc.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_common.c b/drivers/net/ethernet/intel/i40evf/i40e_common.c
index ae084378faab..a43155afdbe2 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_common.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -40,12 +43,10 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 	if (hw->vendor_id == PCI_VENDOR_ID_INTEL) {
 		switch (hw->device_id) {
 		case I40E_DEV_ID_SFP_XL710:
-		case I40E_DEV_ID_SFP_X710:
 		case I40E_DEV_ID_QEMU:
 		case I40E_DEV_ID_KX_A:
 		case I40E_DEV_ID_KX_B:
 		case I40E_DEV_ID_KX_C:
-		case I40E_DEV_ID_KX_D:
 		case I40E_DEV_ID_QSFP_A:
 		case I40E_DEV_ID_QSFP_B:
 		case I40E_DEV_ID_QSFP_C:
@@ -130,7 +131,11 @@ void i40evf_debug_aq(struct i40e_hw *hw, enum i40e_debug_mask mask, void *desc,
  **/
 bool i40evf_check_asq_alive(struct i40e_hw *hw)
 {
-	return !!(rd32(hw, hw->aq.asq.len) & I40E_PF_ATQLEN_ATQENABLE_MASK);
+	if (hw->aq.asq.len)
+		return !!(rd32(hw, hw->aq.asq.len) &
+			  I40E_PF_ATQLEN_ATQENABLE_MASK);
+	else
+		return false;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_hmc.h b/drivers/net/ethernet/intel/i40evf/i40e_hmc.h
index cb97b3eed440..a2ad9a4e399d 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_hmc.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_hmc.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -160,11 +163,6 @@ struct i40e_hmc_info {
 	    (((sd_idx) << I40E_PFHMC_PDINV_PMSDIDX_SHIFT) |		\
 	     ((pd_idx) << I40E_PFHMC_PDINV_PMPDIDX_SHIFT)))
 
-#define I40E_INVALIDATE_VF_HMC_PD(hw, sd_idx, pd_idx, hmc_fn_id)	   \
-	wr32((hw), I40E_GLHMC_VFPDINV((hmc_fn_id) - I40E_FIRST_VF_FPM_ID), \
-	     (((sd_idx) << I40E_PFHMC_PDINV_PMSDIDX_SHIFT) |		   \
-	      ((pd_idx) << I40E_PFHMC_PDINV_PMPDIDX_SHIFT)))
-
 /**
  * I40E_FIND_SD_INDEX_LIMIT - finds segment descriptor index limit
  * @hmc_info: pointer to the HMC configuration information structure
@@ -223,7 +221,7 @@ i40e_status i40e_add_pd_table_entry(struct i40e_hw *hw,
 					      u32 pd_index);
 i40e_status i40e_remove_pd_bp(struct i40e_hw *hw,
 					struct i40e_hmc_info *hmc_info,
-					u32 idx, bool is_pf);
+					u32 idx);
 i40e_status i40e_prep_remove_sd_bp(struct i40e_hmc_info *hmc_info,
 					     u32 idx);
 i40e_status i40e_remove_sd_bp_new(struct i40e_hw *hw,
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_lan_hmc.h b/drivers/net/ethernet/intel/i40evf/i40e_lan_hmc.h
index 17e42ca26d0b..d6f762241537 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_lan_hmc.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_lan_hmc.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -53,6 +56,7 @@ struct i40e_hmc_obj_rxq {
 	u8  tphdata_ena;
 	u8  tphhead_ena;
 	u8  lrxqthresh;
+	u8  prefena;	/* NOTE: normally must be set to 1 at init */
 };
 
 /* Tx queue context data */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_osdep.h b/drivers/net/ethernet/intel/i40evf/i40e_osdep.h
index 622f373b745d..21a91b14bf81 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_osdep.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_osdep.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
index 97ab8c2b76f8..849edcc2e398 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_register.h b/drivers/net/ethernet/intel/i40evf/i40e_register.h
index 30af953cf106..369839655818 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_register.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_register.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -1337,8 +1340,6 @@
 #define I40E_PFINT_ICR0_GPIO_MASK (0x1 << I40E_PFINT_ICR0_GPIO_SHIFT)
 #define I40E_PFINT_ICR0_TIMESYNC_SHIFT 23
 #define I40E_PFINT_ICR0_TIMESYNC_MASK (0x1 << I40E_PFINT_ICR0_TIMESYNC_SHIFT)
-#define I40E_PFINT_ICR0_STORM_DETECT_SHIFT 24
-#define I40E_PFINT_ICR0_STORM_DETECT_MASK (0x1 << I40E_PFINT_ICR0_STORM_DETECT_SHIFT)
 #define I40E_PFINT_ICR0_LINK_STAT_CHANGE_SHIFT 25
 #define I40E_PFINT_ICR0_LINK_STAT_CHANGE_MASK (0x1 << I40E_PFINT_ICR0_LINK_STAT_CHANGE_SHIFT)
 #define I40E_PFINT_ICR0_HMC_ERR_SHIFT 26
@@ -1364,8 +1365,6 @@
 #define I40E_PFINT_ICR0_ENA_GPIO_MASK (0x1 << I40E_PFINT_ICR0_ENA_GPIO_SHIFT)
 #define I40E_PFINT_ICR0_ENA_TIMESYNC_SHIFT 23
 #define I40E_PFINT_ICR0_ENA_TIMESYNC_MASK (0x1 << I40E_PFINT_ICR0_ENA_TIMESYNC_SHIFT)
-#define I40E_PFINT_ICR0_ENA_STORM_DETECT_SHIFT 24
-#define I40E_PFINT_ICR0_ENA_STORM_DETECT_MASK (0x1 << I40E_PFINT_ICR0_ENA_STORM_DETECT_SHIFT)
 #define I40E_PFINT_ICR0_ENA_LINK_STAT_CHANGE_SHIFT 25
 #define I40E_PFINT_ICR0_ENA_LINK_STAT_CHANGE_MASK (0x1 << I40E_PFINT_ICR0_ENA_LINK_STAT_CHANGE_SHIFT)
 #define I40E_PFINT_ICR0_ENA_HMC_ERR_SHIFT 26
@@ -1586,6 +1585,14 @@
 #define I40E_GLLAN_TSOMSK_M 0x000442DC
 #define I40E_GLLAN_TSOMSK_M_TCPMSKM_SHIFT 0
 #define I40E_GLLAN_TSOMSK_M_TCPMSKM_MASK (0xFFF << I40E_GLLAN_TSOMSK_M_TCPMSKM_SHIFT)
+#define I40E_GLLAN_TXPRE_QDIS(_i) (0x000E6500 + ((_i) * 4)) /* i=0..11 */
+#define I40E_GLLAN_TXPRE_QDIS_QINDX_SHIFT 0
+#define I40E_GLLAN_TXPRE_QDIS_QINDX_MASK (0x7FF << I40E_GLLAN_TXPRE_QDIS_QINDX_SHIFT)
+#define I40E_GLLAN_TXPRE_QDIS_SET_QDIS_SHIFT 30
+#define I40E_GLLAN_TXPRE_QDIS_SET_QDIS_MASK (0x1 << I40E_GLLAN_TXPRE_QDIS_SET_QDIS_SHIFT)
+#define I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_SHIFT 31
+#define I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_MASK (0x1 << I40E_GLLAN_TXPRE_QDIS_CLEAR_QDIS_SHIFT)
+
 #define I40E_PFLAN_QALLOC 0x001C0400
 #define I40E_PFLAN_QALLOC_FIRSTQ_SHIFT 0
 #define I40E_PFLAN_QALLOC_FIRSTQ_MASK (0x7FF << I40E_PFLAN_QALLOC_FIRSTQ_SHIFT)
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_status.h b/drivers/net/ethernet/intel/i40evf/i40e_status.h
index 7c08cc2e339b..7fa7a41915c1 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_status.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_status.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index b9f50f40abe1..48ebb6cd69f2 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -725,10 +728,12 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    u32 rx_error,
 				    u16 rx_ptype)
 {
+	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
+	bool ipv4 = false, ipv6 = false;
 	bool ipv4_tunnel, ipv6_tunnel;
 	__wsum rx_udp_csum;
-	__sum16 csum;
 	struct iphdr *iph;
+	__sum16 csum;
 
 	ipv4_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
 		      (rx_ptype < I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
@@ -739,29 +744,57 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	skb->ip_summed = CHECKSUM_NONE;
 
 	/* Rx csum enabled and ip headers found? */
-	if (!(vsi->netdev->features & NETIF_F_RXCSUM &&
-	      rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+	if (!(vsi->netdev->features & NETIF_F_RXCSUM))
 		return;
 
+	/* did the hardware decode the packet and checksum? */
+	if (!(rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+		return;
+
+	/* both known and outer_ip must be set for the below code to work */
+	if (!(decoded.known && decoded.outer_ip))
+		return;
+
+	if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+	    decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4)
+		ipv4 = true;
+	else if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+		 decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6)
+		ipv6 = true;
+
+	if (ipv4 &&
+	    (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
+			 (1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))))
+		goto checksum_fail;
+
 	/* likely incorrect csum if alternate IP extension headers found */
-	if (rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+	if (ipv6 &&
+	    decoded.inner_prot == I40E_RX_PTYPE_INNER_PROT_TCP &&
+	    rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT) &&
+	    rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+		/* don't increment checksum err here, non-fatal err */
 		return;
 
-	/* IP or L4 or outmost IP checksum error */
-	if (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_L4E_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))) {
-		vsi->back->hw_csum_rx_error++;
+	/* there was some L4 error, count error and punt packet to the stack */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT))
+		goto checksum_fail;
+
+	/* handle packets that were not able to be checksummed due
+	 * to arrival speed, in this case the stack can compute
+	 * the csum.
+	 */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_PPRS_SHIFT))
 		return;
-	}
 
+	/* If VXLAN traffic has an outer UDPv4 checksum we need to check
+	 * it in the driver, hardware does not do it for us.
+	 * Since L3L4P bit was set we assume a valid IHL value (>=5)
+	 * so the total length of IPv4 header is IHL*4 bytes
+	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
+	 */
 	if (ipv4_tunnel &&
+	    (decoded.inner_prot != I40E_RX_PTYPE_INNER_PROT_UDP) &&
 	    !(rx_status & (1 << I40E_RX_DESC_STATUS_UDP_0_SHIFT))) {
-		/* If VXLAN traffic has an outer UDPv4 checksum we need to check
-		 * it in the driver, hardware does not do it for us.
-		 * Since L3L4P bit was set we assume a valid IHL value (>=5)
-		 * so the total length of IPv4 header is IHL*4 bytes
-		 */
 		skb->transport_header = skb->mac_header +
 					sizeof(struct ethhdr) +
 					(ip_hdr(skb)->ihl * 4);
@@ -778,13 +811,16 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				(skb->len - skb_transport_offset(skb)),
 				IPPROTO_UDP, rx_udp_csum);
 
-		if (udp_hdr(skb)->check != csum) {
-			vsi->back->hw_csum_rx_error++;
-			return;
-		}
+		if (udp_hdr(skb)->check != csum)
+			goto checksum_fail;
 	}
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	return;
+
+checksum_fail:
+	vsi->back->hw_csum_rx_error++;
 }
 
 /**
@@ -953,6 +989,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 		/* ERR_MASK will only have valid bits if EOP set */
 		if (unlikely(rx_error & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) {
 			dev_kfree_skb_any(skb);
+			/* TODO: shouldn't we increment a counter indicating the
+			 * drop?
+			 */
 			goto next_desc;
 		}
 
@@ -1508,9 +1547,7 @@ static int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 				      struct i40e_ring *tx_ring)
 {
-#if PAGE_SIZE > I40E_MAX_DATA_PER_TXD
 	unsigned int f;
-#endif
 	int count = 0;
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
@@ -1519,12 +1556,9 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 	 *       + 1 desc for context descriptor,
 	 * otherwise try next time
 	 */
-#if PAGE_SIZE > I40E_MAX_DATA_PER_TXD
 	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
 		count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
-#else
-	count += skb_shinfo(skb)->nr_frags;
-#endif
+
 	count += TXD_USE_COUNT(skb_headlen(skb));
 	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
 		tx_ring->tx_stats.tx_busy++;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 10bf49e18d7f..30d248bc5d19 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -24,7 +27,7 @@
 #ifndef _I40E_TXRX_H_
 #define _I40E_TXRX_H_
 
-/* Interrupt Throttling and Rate Limiting (storm control) Goodies */
+/* Interrupt Throttling and Rate Limiting Goodies */
 
 #define I40E_MAX_ITR               0x0FF0  /* reg uses 2 usec resolution */
 #define I40E_MIN_ITR               0x0004  /* reg uses 2 usec resolution */
@@ -66,16 +69,11 @@ enum i40e_dyn_idx_t {
 
 /* Supported RSS offloads */
 #define I40E_DEFAULT_RSS_HENA ( \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_SCTP) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) | \
 	((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) | \
-	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN) | \
 	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP) | \
@@ -119,11 +117,11 @@ enum i40e_dyn_idx_t {
 #define i40e_rx_desc i40e_32byte_rx_desc
 
 #define I40E_MIN_TX_LEN		17
-#define I40E_MAX_DATA_PER_TXD	16383	/* aka 16kB - 1 */
+#define I40E_MAX_DATA_PER_TXD	8192
 
 /* Tx Descriptors needed, worst case */
 #define TXD_USE_COUNT(S) DIV_ROUND_UP((S), I40E_MAX_DATA_PER_TXD)
-#define DESC_NEEDED ((MAX_SKB_FRAGS * TXD_USE_COUNT(PAGE_SIZE)) + 4)
+#define DESC_NEEDED (MAX_SKB_FRAGS + 4)
 
 #define I40E_TX_FLAGS_CSUM		(u32)(1)
 #define I40E_TX_FLAGS_HW_VLAN		(u32)(1 << 1)
@@ -180,7 +178,6 @@ enum i40e_ring_state_t {
 	__I40E_TX_DETECT_HANG,
 	__I40E_HANG_CHECK_ARMED,
 	__I40E_RX_PS_ENABLED,
-	__I40E_RX_LRO_ENABLED,
 	__I40E_RX_16BYTE_DESC_ENABLED,
 };
 
@@ -196,12 +193,6 @@ enum i40e_ring_state_t {
 	set_bit(__I40E_TX_DETECT_HANG, &(ring)->state)
 #define clear_check_for_tx_hang(ring) \
 	clear_bit(__I40E_TX_DETECT_HANG, &(ring)->state)
-#define ring_is_lro_enabled(ring) \
-	test_bit(__I40E_RX_LRO_ENABLED, &(ring)->state)
-#define set_ring_lro_enabled(ring) \
-	set_bit(__I40E_RX_LRO_ENABLED, &(ring)->state)
-#define clear_ring_lro_enabled(ring) \
-	clear_bit(__I40E_RX_LRO_ENABLED, &(ring)->state)
 #define ring_is_16byte_desc_enabled(ring) \
 	test_bit(__I40E_RX_16BYTE_DESC_ENABLED, &(ring)->state)
 #define set_ring_16byte_desc_enabled(ring) \
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_type.h b/drivers/net/ethernet/intel/i40evf/i40e_type.h
index 4673b3381edd..d3cf5a69de54 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_type.h
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -32,13 +35,11 @@
 #include "i40e_lan_hmc.h"
 
 /* Device IDs */
-#define I40E_DEV_ID_SFP_XL710	0x1572
-#define I40E_DEV_ID_SFP_X710		0x1573
+#define I40E_DEV_ID_SFP_XL710		0x1572
 #define I40E_DEV_ID_QEMU		0x1574
 #define I40E_DEV_ID_KX_A		0x157F
 #define I40E_DEV_ID_KX_B		0x1580
 #define I40E_DEV_ID_KX_C		0x1581
-#define I40E_DEV_ID_KX_D		0x1582
 #define I40E_DEV_ID_QSFP_A		0x1583
 #define I40E_DEV_ID_QSFP_B		0x1584
 #define I40E_DEV_ID_QSFP_C		0x1585
@@ -57,8 +58,8 @@
 /* Max default timeout in ms, */
 #define I40E_MAX_NVM_TIMEOUT		18000
 
-/* Switch from mc to the 2usec global time (this is the GTIME resolution) */
-#define I40E_MS_TO_GTIME(time)		(((time) * 1000) / 2)
+/* Switch from ms to the 1usec global time (this is the GTIME resolution) */
+#define I40E_MS_TO_GTIME(time)		((time) * 1000)
 
 /* forward declaration */
 struct i40e_hw;
@@ -101,15 +102,6 @@ enum i40e_debug_mask {
 	I40E_DEBUG_ALL			= 0xFFFFFFFF
 };
 
-/* PCI Bus Info */
-#define I40E_PCI_LINK_WIDTH_1		0x10
-#define I40E_PCI_LINK_WIDTH_2		0x20
-#define I40E_PCI_LINK_WIDTH_4		0x40
-#define I40E_PCI_LINK_WIDTH_8		0x80
-#define I40E_PCI_LINK_SPEED_2500	0x1
-#define I40E_PCI_LINK_SPEED_5000	0x2
-#define I40E_PCI_LINK_SPEED_8000	0x3
-
 /* These are structs for managing the hardware information and the operations.
  * The structures of function pointers are filled out at init time when we
  * know for sure exactly which hardware we're working with.  This gives us the
@@ -173,6 +165,9 @@ struct i40e_link_status {
 	u8 loopback;
 	/* is Link Status Event notification to SW enabled */
 	bool lse_enable;
+	u16 max_frame_size;
+	bool crc_enable;
+	u8 pacing;
 };
 
 struct i40e_phy_info {
@@ -415,6 +410,7 @@ struct i40e_driver_version {
 	u8 minor_version;
 	u8 build_version;
 	u8 subbuild_version;
+	u8 driver_string[32];
 };
 
 /* RX Descriptors */
@@ -494,9 +490,6 @@ union i40e_32byte_rx_desc {
 	} wb;  /* writeback */
 };
 
-#define I40E_RXD_QW1_STATUS_SHIFT	0
-#define I40E_RXD_QW1_STATUS_MASK	(0x7FFFUL << I40E_RXD_QW1_STATUS_SHIFT)
-
 enum i40e_rx_desc_status_bits {
 	/* Note: These are predefined bit offsets */
 	I40E_RX_DESC_STATUS_DD_SHIFT		= 0,
@@ -513,9 +506,14 @@ enum i40e_rx_desc_status_bits {
 	I40E_RX_DESC_STATUS_LPBK_SHIFT		= 14,
 	I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT	= 15,
 	I40E_RX_DESC_STATUS_RESERVED_SHIFT	= 16, /* 2 BITS */
-	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 18
+	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 18,
+	I40E_RX_DESC_STATUS_LAST /* this entry must be last!!! */
 };
 
+#define I40E_RXD_QW1_STATUS_SHIFT	0
+#define I40E_RXD_QW1_STATUS_MASK	(((1 << I40E_RX_DESC_STATUS_LAST) - 1) \
+					 << I40E_RXD_QW1_STATUS_SHIFT)
+
 #define I40E_RXD_QW1_STATUS_TSYNINDX_SHIFT   I40E_RX_DESC_STATUS_TSYNINDX_SHIFT
 #define I40E_RXD_QW1_STATUS_TSYNINDX_MASK	(0x3UL << \
 					     I40E_RXD_QW1_STATUS_TSYNINDX_SHIFT)
@@ -543,7 +541,8 @@ enum i40e_rx_desc_error_bits {
 	I40E_RX_DESC_ERROR_IPE_SHIFT		= 3,
 	I40E_RX_DESC_ERROR_L4E_SHIFT		= 4,
 	I40E_RX_DESC_ERROR_EIPE_SHIFT		= 5,
-	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6
+	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6,
+	I40E_RX_DESC_ERROR_PPRS_SHIFT		= 7
 };
 
 enum i40e_rx_desc_error_l3l4e_fcoe_masks {
@@ -664,7 +663,6 @@ enum i40e_rx_desc_ext_status_bits {
 	I40E_RX_DESC_EXT_STATUS_L2TAG3P_SHIFT	= 1,
 	I40E_RX_DESC_EXT_STATUS_FLEXBL_SHIFT	= 2, /* 2 BITS */
 	I40E_RX_DESC_EXT_STATUS_FLEXBH_SHIFT	= 4, /* 2 BITS */
-	I40E_RX_DESC_EXT_STATUS_FTYPE_SHIFT	= 6, /* 3 BITS */
 	I40E_RX_DESC_EXT_STATUS_FDLONGB_SHIFT	= 9,
 	I40E_RX_DESC_EXT_STATUS_FCOELONGB_SHIFT	= 10,
 	I40E_RX_DESC_EXT_STATUS_PELONGB_SHIFT	= 11,
@@ -868,18 +866,14 @@ struct i40e_filter_program_desc {
 
 /* Packet Classifier Types for filters */
 enum i40e_filter_pctype {
-	/* Note: Values 0-28 are reserved for future use */
-	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP	= 29,
-	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP	= 30,
+	/* Note: Values 0-30 are reserved for future use */
 	I40E_FILTER_PCTYPE_NONF_IPV4_UDP		= 31,
-	I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN		= 32,
+	/* Note: Value 32 is reserved for future use */
 	I40E_FILTER_PCTYPE_NONF_IPV4_TCP		= 33,
 	I40E_FILTER_PCTYPE_NONF_IPV4_SCTP		= 34,
 	I40E_FILTER_PCTYPE_NONF_IPV4_OTHER		= 35,
 	I40E_FILTER_PCTYPE_FRAG_IPV4			= 36,
-	/* Note: Values 37-38 are reserved for future use */
-	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP	= 39,
-	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP	= 40,
+	/* Note: Values 37-40 are reserved for future use */
 	I40E_FILTER_PCTYPE_NONF_IPV6_UDP		= 41,
 	I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN		= 42,
 	I40E_FILTER_PCTYPE_NONF_IPV6_TCP		= 43,
@@ -961,6 +955,16 @@ struct i40e_vsi_context {
 	struct i40e_aqc_vsi_properties_data info;
 };
 
+struct i40e_veb_context {
+	u16 seid;
+	u16 uplink_seid;
+	u16 veb_number;
+	u16 vebs_allocated;
+	u16 vebs_unallocated;
+	u16 flags;
+	struct i40e_aqc_get_veb_parameters_completion info;
+};
+
 /* Statistics collected by each port, VSI, VEB, and S-channel */
 struct i40e_eth_stats {
 	u64 rx_bytes;			/* gorc */
@@ -968,8 +972,6 @@ struct i40e_eth_stats {
 	u64 rx_multicast;		/* mprc */
 	u64 rx_broadcast;		/* bprc */
 	u64 rx_discards;		/* rdpc */
-	u64 rx_errors;			/* repc */
-	u64 rx_missed;			/* rmpc */
 	u64 rx_unknown_protocol;	/* rupp */
 	u64 tx_bytes;			/* gotc */
 	u64 tx_unicast;			/* uptc */
@@ -1021,9 +1023,12 @@ struct i40e_hw_port_stats {
 	u64 tx_size_big;		/* ptc9522 */
 	u64 mac_short_packet_dropped;	/* mspdc */
 	u64 checksum_error;		/* xec */
+	/* flow director stats */
+	u64 fd_atr_match;
+	u64 fd_sb_match;
 	/* EEE LPI */
-	bool tx_lpi_status;
-	bool rx_lpi_status;
+	u32 tx_lpi_status;
+	u32 rx_lpi_status;
 	u64 tx_lpi_count;		/* etlpic */
 	u64 rx_lpi_count;		/* erlpic */
 };
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h b/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h
index ccf45d04b7ef..cd18d5689006 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2013 - 2014 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -338,10 +341,6 @@ struct i40e_virtchnl_pf_event {
 	int severity;
 };
 
-/* The following are TBD, not necessary for LAN functionality.
- * I40E_VIRTCHNL_OP_FCOE
- */
-
 /* VF reset states - these are written into the RSTAT register:
  * I40E_VFGEN_RSTAT1 on the PF
  * I40E_VFGEN_RSTAT on the VF
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index 807807d62387..30ef519d4b91 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -77,7 +80,7 @@ struct i40e_vsi {
 #define I40EVF_MIN_TXD       64
 #define I40EVF_MAX_RXD       4096
 #define I40EVF_MIN_RXD       64
-#define I40EVF_REQ_DESCRIPTOR_MULTIPLE  8
+#define I40EVF_REQ_DESCRIPTOR_MULTIPLE  32
 
 /* Supported Rx Buffer Sizes */
 #define I40EVF_RXBUFFER_64    64     /* Used for packet split */
@@ -193,10 +196,12 @@ struct i40evf_adapter {
 	struct i40e_ring *tx_rings[I40E_MAX_VSI_QP];
 	u32 tx_timeout_count;
 	struct list_head mac_filter_list;
+	u32 tx_desc_count;
 
 	/* RX */
 	struct i40e_ring *rx_rings[I40E_MAX_VSI_QP];
 	u64 hw_csum_rx_error;
+	u32 rx_desc_count;
 	int num_msix_vectors;
 	struct msix_entry *msix_entries;
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
index 8b0db1ce179c..60407a9df0c1 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -44,8 +47,6 @@ static const struct i40evf_stats i40evf_gstrings_stats[] = {
 	I40EVF_STAT("rx_multicast", current_stats.rx_multicast),
 	I40EVF_STAT("rx_broadcast", current_stats.rx_broadcast),
 	I40EVF_STAT("rx_discards", current_stats.rx_discards),
-	I40EVF_STAT("rx_errors", current_stats.rx_errors),
-	I40EVF_STAT("rx_missed", current_stats.rx_missed),
 	I40EVF_STAT("rx_unknown_protocol", current_stats.rx_unknown_protocol),
 	I40EVF_STAT("tx_bytes", current_stats.tx_bytes),
 	I40EVF_STAT("tx_unicast", current_stats.tx_unicast),
@@ -56,10 +57,12 @@ static const struct i40evf_stats i40evf_gstrings_stats[] = {
 };
 
 #define I40EVF_GLOBAL_STATS_LEN ARRAY_SIZE(i40evf_gstrings_stats)
-#define I40EVF_QUEUE_STATS_LEN \
+#define I40EVF_QUEUE_STATS_LEN(_dev) \
 	(((struct i40evf_adapter *) \
-		netdev_priv(netdev))->vsi_res->num_queue_pairs * 4)
-#define I40EVF_STATS_LEN (I40EVF_GLOBAL_STATS_LEN + I40EVF_QUEUE_STATS_LEN)
+		netdev_priv(_dev))->vsi_res->num_queue_pairs \
+		  * 2 * (sizeof(struct i40e_queue_stats) / sizeof(u64)))
+#define I40EVF_STATS_LEN(_dev) \
+	(I40EVF_GLOBAL_STATS_LEN + I40EVF_QUEUE_STATS_LEN(_dev))
 
 /**
  * i40evf_get_settings - Get Link Speed and Duplex settings
@@ -75,7 +78,7 @@ static int i40evf_get_settings(struct net_device *netdev,
 	/* In the future the VF will be able to query the PF for
 	 * some information - for now use a dummy value
 	 */
-	ecmd->supported = SUPPORTED_10000baseT_Full;
+	ecmd->supported = 0;
 	ecmd->autoneg = AUTONEG_DISABLE;
 	ecmd->transceiver = XCVR_DUMMY1;
 	ecmd->port = PORT_NONE;
@@ -94,9 +97,9 @@ static int i40evf_get_settings(struct net_device *netdev,
 static int i40evf_get_sset_count(struct net_device *netdev, int sset)
 {
 	if (sset == ETH_SS_STATS)
-		return I40EVF_STATS_LEN;
+		return I40EVF_STATS_LEN(netdev);
 	else
-		return -ENOTSUPP;
+		return -EINVAL;
 }
 
 /**
@@ -219,13 +222,11 @@ static void i40evf_get_ringparam(struct net_device *netdev,
 				  struct ethtool_ringparam *ring)
 {
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
-	struct i40e_ring *tx_ring = adapter->tx_rings[0];
-	struct i40e_ring *rx_ring = adapter->rx_rings[0];
 
 	ring->rx_max_pending = I40EVF_MAX_RXD;
 	ring->tx_max_pending = I40EVF_MAX_TXD;
-	ring->rx_pending = rx_ring->count;
-	ring->tx_pending = tx_ring->count;
+	ring->rx_pending = adapter->rx_desc_count;
+	ring->tx_pending = adapter->tx_desc_count;
 }
 
 /**
@@ -241,7 +242,6 @@ static int i40evf_set_ringparam(struct net_device *netdev,
 {
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
 	u32 new_rx_count, new_tx_count;
-	int i;
 
 	if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))
 		return -EINVAL;
@@ -257,17 +257,16 @@ static int i40evf_set_ringparam(struct net_device *netdev,
 	new_rx_count = ALIGN(new_rx_count, I40EVF_REQ_DESCRIPTOR_MULTIPLE);
 
 	/* if nothing to do return success */
-	if ((new_tx_count == adapter->tx_rings[0]->count) &&
-	    (new_rx_count == adapter->rx_rings[0]->count))
+	if ((new_tx_count == adapter->tx_desc_count) &&
+	    (new_rx_count == adapter->rx_desc_count))
 		return 0;
 
-	for (i = 0; i < adapter->vsi_res->num_queue_pairs; i++) {
-		adapter->tx_rings[0]->count = new_tx_count;
-		adapter->rx_rings[0]->count = new_rx_count;
-	}
+	adapter->tx_desc_count = new_tx_count;
+	adapter->rx_desc_count = new_rx_count;
 
 	if (netif_running(netdev))
 		i40evf_reinit_locked(adapter);
+
 	return 0;
 }
 
@@ -290,14 +289,13 @@ static int i40evf_get_coalesce(struct net_device *netdev,
 	ec->rx_max_coalesced_frames = vsi->work_limit;
 
 	if (ITR_IS_DYNAMIC(vsi->rx_itr_setting))
-		ec->rx_coalesce_usecs = 1;
-	else
-		ec->rx_coalesce_usecs = vsi->rx_itr_setting;
+		ec->use_adaptive_rx_coalesce = 1;
 
 	if (ITR_IS_DYNAMIC(vsi->tx_itr_setting))
-		ec->tx_coalesce_usecs = 1;
-	else
-		ec->tx_coalesce_usecs = vsi->tx_itr_setting;
+		ec->use_adaptive_tx_coalesce = 1;
+
+	ec->rx_coalesce_usecs = vsi->rx_itr_setting & ~I40E_ITR_DYNAMIC;
+	ec->tx_coalesce_usecs = vsi->tx_itr_setting & ~I40E_ITR_DYNAMIC;
 
 	return 0;
 }
@@ -318,54 +316,361 @@ static int i40evf_set_coalesce(struct net_device *netdev,
 	struct i40e_q_vector *q_vector;
 	int i;
 
-	if (ec->tx_max_coalesced_frames || ec->rx_max_coalesced_frames)
-		vsi->work_limit = ec->tx_max_coalesced_frames;
+	if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq)
+		vsi->work_limit = ec->tx_max_coalesced_frames_irq;
+
+	if ((ec->rx_coalesce_usecs >= (I40E_MIN_ITR << 1)) &&
+	    (ec->rx_coalesce_usecs <= (I40E_MAX_ITR << 1)))
+		vsi->rx_itr_setting = ec->rx_coalesce_usecs;
+
+	else
+		return -EINVAL;
+
+	if ((ec->tx_coalesce_usecs >= (I40E_MIN_ITR << 1)) &&
+	    (ec->tx_coalesce_usecs <= (I40E_MAX_ITR << 1)))
+		vsi->tx_itr_setting = ec->tx_coalesce_usecs;
+	else if (ec->use_adaptive_tx_coalesce)
+		vsi->tx_itr_setting = (I40E_ITR_DYNAMIC |
+				       ITR_REG_TO_USEC(I40E_ITR_RX_DEF));
+	else
+		return -EINVAL;
+
+	if (ec->use_adaptive_rx_coalesce)
+		vsi->rx_itr_setting |= I40E_ITR_DYNAMIC;
+	else
+		vsi->rx_itr_setting &= ~I40E_ITR_DYNAMIC;
+
+	if (ec->use_adaptive_tx_coalesce)
+		vsi->tx_itr_setting |= I40E_ITR_DYNAMIC;
+	else
+		vsi->tx_itr_setting &= ~I40E_ITR_DYNAMIC;
 
-	switch (ec->rx_coalesce_usecs) {
-	case 0:
-		vsi->rx_itr_setting = 0;
+	for (i = 0; i < adapter->num_msix_vectors - NONQ_VECS; i++) {
+		q_vector = adapter->q_vector[i];
+		q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
+		wr32(hw, I40E_VFINT_ITRN1(0, i), q_vector->rx.itr);
+		q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
+		wr32(hw, I40E_VFINT_ITRN1(1, i), q_vector->tx.itr);
+		i40e_flush(hw);
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_get_rss_hash_opts - Get RSS hash Input Set for each flow type
+ * @adapter: board private structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the flow is supported, else Invalid Input.
+ **/
+static int i40evf_get_rss_hash_opts(struct i40evf_adapter *adapter,
+				    struct ethtool_rxnfc *cmd)
+{
+	struct i40e_hw *hw = &adapter->hw;
+	u64 hena = (u64)rd32(hw, I40E_VFQF_HENA(0)) |
+		   ((u64)rd32(hw, I40E_VFQF_HENA(1)) << 32);
+
+	/* We always hash on IP src and dest addresses */
+	cmd->data = RXH_IP_SRC | RXH_IP_DST;
+
+	switch (cmd->flow_type) {
+	case TCP_V4_FLOW:
+		if (hena & ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP))
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
-	case 1:
-		vsi->rx_itr_setting = (I40E_ITR_DYNAMIC
-				       | ITR_REG_TO_USEC(I40E_ITR_RX_DEF));
+	case UDP_V4_FLOW:
+		if (hena & ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP))
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
-	default:
-		if ((ec->rx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
-		    (ec->rx_coalesce_usecs > (I40E_MAX_ITR << 1)))
-			return -EINVAL;
-		vsi->rx_itr_setting = ec->rx_coalesce_usecs;
+
+	case SCTP_V4_FLOW:
+	case AH_ESP_V4_FLOW:
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+	case IPV4_FLOW:
+		break;
+
+	case TCP_V6_FLOW:
+		if (hena & ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP))
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
+	case UDP_V6_FLOW:
+		if (hena & ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP))
+			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+
+	case SCTP_V6_FLOW:
+	case AH_ESP_V6_FLOW:
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+	case IPV6_FLOW:
+		break;
+	default:
+		cmd->data = 0;
+		return -EINVAL;
 	}
 
-	switch (ec->tx_coalesce_usecs) {
-	case 0:
-		vsi->tx_itr_setting = 0;
+	return 0;
+}
+
+/**
+ * i40evf_get_rxnfc - command to get RX flow classification rules
+ * @netdev: network interface device structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the command is supported.
+ **/
+static int i40evf_get_rxnfc(struct net_device *netdev,
+			    struct ethtool_rxnfc *cmd,
+			    u32 *rule_locs)
+{
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+	int ret = -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = adapter->vsi_res->num_queue_pairs;
+		ret = 0;
 		break;
-	case 1:
-		vsi->tx_itr_setting = (I40E_ITR_DYNAMIC
-				       | ITR_REG_TO_USEC(I40E_ITR_TX_DEF));
+	case ETHTOOL_GRXFH:
+		ret = i40evf_get_rss_hash_opts(adapter, cmd);
 		break;
 	default:
-		if ((ec->tx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
-		    (ec->tx_coalesce_usecs > (I40E_MAX_ITR << 1)))
+		break;
+	}
+
+	return ret;
+}
+
+/**
+ * i40evf_set_rss_hash_opt - Enable/Disable flow types for RSS hash
+ * @adapter: board private structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the flow input set is supported.
+ **/
+static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
+				   struct ethtool_rxnfc *nfc)
+{
+	struct i40e_hw *hw = &adapter->hw;
+
+	u64 hena = (u64)rd32(hw, I40E_VFQF_HENA(0)) |
+		   ((u64)rd32(hw, I40E_VFQF_HENA(1)) << 32);
+
+	/* RSS does not support anything other than hashing
+	 * to queues on src and dst IPs and ports
+	 */
+	if (nfc->data & ~(RXH_IP_SRC | RXH_IP_DST |
+			  RXH_L4_B_0_1 | RXH_L4_B_2_3))
+		return -EINVAL;
+
+	/* We need at least the IP SRC and DEST fields for hashing */
+	if (!(nfc->data & RXH_IP_SRC) ||
+	    !(nfc->data & RXH_IP_DST))
+		return -EINVAL;
+
+	switch (nfc->flow_type) {
+	case TCP_V4_FLOW:
+		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+		case 0:
+			hena &= ~((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
+			break;
+		default:
 			return -EINVAL;
-		vsi->tx_itr_setting = ec->tx_coalesce_usecs;
+		}
+		break;
+	case TCP_V6_FLOW:
+		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+		case 0:
+			hena &= ~((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case UDP_V4_FLOW:
+		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+		case 0:
+			hena &= ~(((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
+				  ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			hena |= (((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
+				 ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
+			break;
+		default:
+			return -EINVAL;
+		}
 		break;
+	case UDP_V6_FLOW:
+		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+		case 0:
+			hena &= ~(((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
+				  ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			hena |= (((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
+				 ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case AH_ESP_V4_FLOW:
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+	case SCTP_V4_FLOW:
+		if ((nfc->data & RXH_L4_B_0_1) ||
+		    (nfc->data & RXH_L4_B_2_3))
+			return -EINVAL;
+		hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER);
+		break;
+	case AH_ESP_V6_FLOW:
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+	case SCTP_V6_FLOW:
+		if ((nfc->data & RXH_L4_B_0_1) ||
+		    (nfc->data & RXH_L4_B_2_3))
+			return -EINVAL;
+		hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_OTHER);
+		break;
+	case IPV4_FLOW:
+		hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) |
+			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4);
+		break;
+	case IPV6_FLOW:
+		hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) |
+			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6);
+		break;
+	default:
+		return -EINVAL;
 	}
 
-	for (i = 0; i < adapter->num_msix_vectors - NONQ_VECS; i++) {
-		q_vector = adapter->q_vector[i];
-		q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
-		wr32(hw, I40E_VFINT_ITRN1(0, i), q_vector->rx.itr);
-		q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
-		wr32(hw, I40E_VFINT_ITRN1(1, i), q_vector->tx.itr);
-		i40e_flush(hw);
+	wr32(hw, I40E_VFQF_HENA(0), (u32)hena);
+	wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32));
+	i40e_flush(hw);
+
+	return 0;
+}
+
+/**
+ * i40evf_set_rxnfc - command to set RX flow classification rules
+ * @netdev: network interface device structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the command is supported.
+ **/
+static int i40evf_set_rxnfc(struct net_device *netdev,
+			    struct ethtool_rxnfc *cmd)
+{
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+	int ret = -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXFH:
+		ret = i40evf_set_rss_hash_opt(adapter, cmd);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+/**
+ * i40evf_get_channels: get the number of channels supported by the device
+ * @netdev: network interface device structure
+ * @ch: channel information structure
+ *
+ * For the purposes of our device, we only use combined channels, i.e. a tx/rx
+ * queue pair. Report one extra channel to match our "other" MSI-X vector.
+ **/
+static void i40evf_get_channels(struct net_device *netdev,
+				struct ethtool_channels *ch)
+{
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+
+	/* Report maximum channels */
+	ch->max_combined = adapter->vsi_res->num_queue_pairs;
+
+	ch->max_other = NONQ_VECS;
+	ch->other_count = NONQ_VECS;
+
+	ch->combined_count = adapter->vsi_res->num_queue_pairs;
+}
+
+/**
+ * i40evf_get_rxfh_indir_size - get the rx flow hash indirection table size
+ * @netdev: network interface device structure
+ *
+ * Returns the table size.
+ **/
+static u32 i40evf_get_rxfh_indir_size(struct net_device *netdev)
+{
+	return (I40E_VFQF_HLUT_MAX_INDEX + 1) * 4;
+}
+
+/**
+ * i40evf_get_rxfh - get the rx flow hash indirection table
+ * @netdev: network interface device structure
+ * @indir: indirection table
+ * @key: hash key (will be %NULL until get_rxfh_key_size is implemented)
+ *
+ * Reads the indirection table directly from the hardware. Always returns 0.
+ **/
+static int i40evf_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key)
+{
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+	struct i40e_hw *hw = &adapter->hw;
+	u32 hlut_val;
+	int i, j;
+
+	for (i = 0, j = 0; i < I40E_VFQF_HLUT_MAX_INDEX; i++) {
+		hlut_val = rd32(hw, I40E_VFQF_HLUT(i));
+		indir[j++] = hlut_val & 0xff;
+		indir[j++] = (hlut_val >> 8) & 0xff;
+		indir[j++] = (hlut_val >> 16) & 0xff;
+		indir[j++] = (hlut_val >> 24) & 0xff;
+	}
+	return 0;
+}
+
+/**
+ * i40evf_set_rxfh - set the rx flow hash indirection table
+ * @netdev: network interface device structure
+ * @indir: indirection table
+ * @key: hash key (will be %NULL until get_rxfh_key_size is implemented)
+ *
+ * Returns -EINVAL if the table specifies an inavlid queue id, otherwise
+ * returns 0 after programming the table.
+ **/
+static int i40evf_set_rxfh(struct net_device *netdev, const u32 *indir,
+			   const u8 *key)
+{
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+	struct i40e_hw *hw = &adapter->hw;
+	u32 hlut_val;
+	int i, j;
+
+	for (i = 0, j = 0; i < I40E_VFQF_HLUT_MAX_INDEX + 1; i++) {
+		hlut_val = indir[j++];
+		hlut_val |= indir[j++] << 8;
+		hlut_val |= indir[j++] << 16;
+		hlut_val |= indir[j++] << 24;
+		wr32(hw, I40E_VFQF_HLUT(i), hlut_val);
 	}
 
 	return 0;
 }
 
-static struct ethtool_ops i40evf_ethtool_ops = {
+static const struct ethtool_ops i40evf_ethtool_ops = {
 	.get_settings		= i40evf_get_settings,
 	.get_drvinfo		= i40evf_get_drvinfo,
 	.get_link		= ethtool_op_get_link,
@@ -378,6 +683,12 @@ static struct ethtool_ops i40evf_ethtool_ops = {
 	.set_msglevel		= i40evf_set_msglevel,
 	.get_coalesce		= i40evf_get_coalesce,
 	.set_coalesce		= i40evf_set_coalesce,
+	.get_rxnfc		= i40evf_get_rxnfc,
+	.set_rxnfc		= i40evf_set_rxnfc,
+	.get_rxfh_indir_size	= i40evf_get_rxfh_indir_size,
+	.get_rxfh		= i40evf_get_rxfh,
+	.set_rxfh		= i40evf_set_rxfh,
+	.get_channels		= i40evf_get_channels,
 };
 
 /**
@@ -389,5 +700,5 @@ static struct ethtool_ops i40evf_ethtool_ops = {
  **/
 void i40evf_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &i40evf_ethtool_ops);
+	netdev->ethtool_ops = &i40evf_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 2797548fde0d..7fc5f3b5d6bf 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -25,13 +28,15 @@
 #include "i40e_prototype.h"
 static int i40evf_setup_all_tx_resources(struct i40evf_adapter *adapter);
 static int i40evf_setup_all_rx_resources(struct i40evf_adapter *adapter);
+static void i40evf_free_all_tx_resources(struct i40evf_adapter *adapter);
+static void i40evf_free_all_rx_resources(struct i40evf_adapter *adapter);
 static int i40evf_close(struct net_device *netdev);
 
 char i40evf_driver_name[] = "i40evf";
 static const char i40evf_driver_string[] =
 	"Intel(R) XL710 X710 Virtual Function Network Driver";
 
-#define DRV_VERSION "0.9.16"
+#define DRV_VERSION "0.9.34"
 const char i40evf_driver_version[] = DRV_VERSION;
 static const char i40evf_copyright[] =
 	"Copyright (c) 2013 - 2014 Intel Corporation.";
@@ -167,7 +172,6 @@ static void i40evf_tx_timeout(struct net_device *netdev)
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
 
 	adapter->tx_timeout_count++;
-	dev_info(&adapter->pdev->dev, "TX timeout detected.\n");
 	if (!(adapter->flags & I40EVF_FLAG_RESET_PENDING)) {
 		adapter->flags |= I40EVF_FLAG_RESET_NEEDED;
 		schedule_work(&adapter->reset_task);
@@ -657,12 +661,9 @@ i40evf_vlan_filter *i40evf_add_vlan(struct i40evf_adapter *adapter, u16 vlan)
 	f = i40evf_find_vlan(adapter, vlan);
 	if (NULL == f) {
 		f = kzalloc(sizeof(*f), GFP_ATOMIC);
-		if (NULL == f) {
-			dev_info(&adapter->pdev->dev,
-				 "%s: no memory for new VLAN filter\n",
-				 __func__);
+		if (NULL == f)
 			return NULL;
-		}
+
 		f->vlan = vlan;
 
 		INIT_LIST_HEAD(&f->list);
@@ -688,7 +689,6 @@ static void i40evf_del_vlan(struct i40evf_adapter *adapter, u16 vlan)
 		f->remove = true;
 		adapter->aq_required |= I40EVF_FLAG_AQ_DEL_VLAN_FILTER;
 	}
-	return;
 }
 
 /**
@@ -767,14 +767,12 @@ i40evf_mac_filter *i40evf_add_filter(struct i40evf_adapter *adapter,
 	if (NULL == f) {
 		f = kzalloc(sizeof(*f), GFP_ATOMIC);
 		if (NULL == f) {
-			dev_info(&adapter->pdev->dev,
-				 "%s: no memory for new filter\n", __func__);
 			clear_bit(__I40EVF_IN_CRITICAL_TASK,
 				  &adapter->crit_section);
 			return NULL;
 		}
 
-		memcpy(f->macaddr, macaddr, ETH_ALEN);
+		ether_addr_copy(f->macaddr, macaddr);
 
 		list_add(&f->list, &adapter->mac_filter_list);
 		f->add = true;
@@ -807,9 +805,8 @@ static int i40evf_set_mac(struct net_device *netdev, void *p)
 
 	f = i40evf_add_filter(adapter, addr->sa_data);
 	if (f) {
-		memcpy(hw->mac.addr, addr->sa_data, netdev->addr_len);
-		memcpy(netdev->dev_addr, adapter->hw.mac.addr,
-		       netdev->addr_len);
+		ether_addr_copy(hw->mac.addr, addr->sa_data);
+		ether_addr_copy(netdev->dev_addr, adapter->hw.mac.addr);
 	}
 
 	return (f == NULL) ? -ENOMEM : 0;
@@ -841,7 +838,7 @@ static void i40evf_set_rx_mode(struct net_device *netdev)
 	list_for_each_entry_safe(f, ftmp, &adapter->mac_filter_list, list) {
 		bool found = false;
 
-		if (f->macaddr[0] & 0x01) {
+		if (is_multicast_ether_addr(f->macaddr)) {
 			netdev_for_each_mc_addr(mca, netdev) {
 				if (ether_addr_equal(mca->addr, f->macaddr)) {
 					found = true;
@@ -970,6 +967,9 @@ void i40evf_down(struct i40evf_adapter *adapter)
 	struct net_device *netdev = adapter->netdev;
 	struct i40evf_mac_filter *f;
 
+	if (adapter->state == __I40EVF_DOWN)
+		return;
+
 	/* remove all MAC filters */
 	list_for_each_entry(f, &adapter->mac_filter_list, list) {
 		f->remove = true;
@@ -1027,30 +1027,21 @@ i40evf_acquire_msix_vectors(struct i40evf_adapter *adapter, int vectors)
 	 * Right now, we simply care about how many we'll get; we'll
 	 * set them up later while requesting irq's.
 	 */
-	while (vectors >= vector_threshold) {
-		err = pci_enable_msix(adapter->pdev, adapter->msix_entries,
-				      vectors);
-		if (!err) /* Success in acquiring all requested vectors. */
-			break;
-		else if (err < 0)
-			vectors = 0; /* Nasty failure, quit now */
-		else /* err == number of vectors we should try again with */
-			vectors = err;
-	}
-
-	if (vectors < vector_threshold) {
-		dev_err(&adapter->pdev->dev, "Unable to allocate MSI-X interrupts.\n");
+	err = pci_enable_msix_range(adapter->pdev, adapter->msix_entries,
+				    vector_threshold, vectors);
+	if (err < 0) {
+		dev_err(&adapter->pdev->dev, "Unable to allocate MSI-X interrupts\n");
 		kfree(adapter->msix_entries);
 		adapter->msix_entries = NULL;
-		err = -EIO;
-	} else {
-		/* Adjust for only the vectors we'll use, which is minimum
-		 * of max_msix_q_vectors + NONQ_VECS, or the number of
-		 * vectors we were allocated.
-		 */
-		adapter->num_msix_vectors = vectors;
+		return err;
 	}
-	return err;
+
+	/* Adjust for only the vectors we'll use, which is minimum
+	 * of max_msix_q_vectors + NONQ_VECS, or the number of
+	 * vectors we were allocated.
+	 */
+	adapter->num_msix_vectors = err;
+	return 0;
 }
 
 /**
@@ -1096,14 +1087,14 @@ static int i40evf_alloc_queues(struct i40evf_adapter *adapter)
 		tx_ring->queue_index = i;
 		tx_ring->netdev = adapter->netdev;
 		tx_ring->dev = &adapter->pdev->dev;
-		tx_ring->count = I40EVF_DEFAULT_TXD;
+		tx_ring->count = adapter->tx_desc_count;
 		adapter->tx_rings[i] = tx_ring;
 
 		rx_ring = &tx_ring[1];
 		rx_ring->queue_index = i;
 		rx_ring->netdev = adapter->netdev;
 		rx_ring->dev = &adapter->pdev->dev;
-		rx_ring->count = I40EVF_DEFAULT_RXD;
+		rx_ring->count = adapter->rx_desc_count;
 		adapter->rx_rings[i] = rx_ring;
 	}
 
@@ -1141,9 +1132,6 @@ static int i40evf_set_interrupt_capability(struct i40evf_adapter *adapter)
 	v_budget = min_t(int, pairs, (int)(num_online_cpus() * 2)) + NONQ_VECS;
 	v_budget = min_t(int, v_budget, (int)adapter->vf_res->max_vectors);
 
-	/* A failure in MSI-X entry allocation isn't fatal, but it does
-	 * mean we disable MSI-X capabilities of the adapter.
-	 */
 	adapter->msix_entries = kcalloc(v_budget,
 					sizeof(struct msix_entry), GFP_KERNEL);
 	if (!adapter->msix_entries) {
@@ -1183,7 +1171,7 @@ static int i40evf_alloc_q_vectors(struct i40evf_adapter *adapter)
 		q_vector->vsi = &adapter->vsi;
 		q_vector->v_idx = q_idx;
 		netif_napi_add(adapter->netdev, &q_vector->napi,
-				       i40evf_napi_poll, 64);
+				       i40evf_napi_poll, NAPI_POLL_WEIGHT);
 		adapter->q_vector[q_idx] = q_vector;
 	}
 
@@ -1236,8 +1224,6 @@ void i40evf_reset_interrupt_capability(struct i40evf_adapter *adapter)
 	pci_disable_msix(adapter->pdev);
 	kfree(adapter->msix_entries);
 	adapter->msix_entries = NULL;
-
-	return;
 }
 
 /**
@@ -1309,7 +1295,6 @@ static void i40evf_watchdog_task(struct work_struct *work)
 		goto restart_watchdog;
 
 	if (adapter->flags & I40EVF_FLAG_PF_COMMS_FAILED) {
-		dev_info(&adapter->pdev->dev, "Checking for redemption\n");
 		if ((rd32(hw, I40E_VFGEN_RSTAT) & 0x3) == I40E_VFR_VFACTIVE) {
 			/* A chance for redemption! */
 			dev_err(&adapter->pdev->dev, "Hardware came out of reset. Attempting reinit.\n");
@@ -1340,8 +1325,7 @@ static void i40evf_watchdog_task(struct work_struct *work)
 	    (rd32(hw, I40E_VFGEN_RSTAT) & 0x3) != I40E_VFR_VFACTIVE) {
 		adapter->state = __I40EVF_RESETTING;
 		adapter->flags |= I40EVF_FLAG_RESET_PENDING;
-		dev_err(&adapter->pdev->dev, "Hardware reset detected.\n");
-		dev_info(&adapter->pdev->dev, "Scheduling reset task\n");
+		dev_err(&adapter->pdev->dev, "Hardware reset detected\n");
 		schedule_work(&adapter->reset_task);
 		adapter->aq_pending = 0;
 		adapter->aq_required = 0;
@@ -1413,7 +1397,7 @@ restart_watchdog:
 }
 
 /**
- * i40evf_configure_rss - increment to next available tx queue
+ * next_queue - increment to next available tx queue
  * @adapter: board private structure
  * @j: queue counter
  *
@@ -1504,15 +1488,12 @@ static void i40evf_reset_task(struct work_struct *work)
 	for (i = 0; i < I40EVF_RESET_WAIT_COUNT; i++) {
 		rstat_val = rd32(hw, I40E_VFGEN_RSTAT) &
 			    I40E_VFGEN_RSTAT_VFR_STATE_MASK;
-		if (rstat_val != I40E_VFR_VFACTIVE) {
-			dev_info(&adapter->pdev->dev, "Reset now occurring\n");
+		if (rstat_val != I40E_VFR_VFACTIVE)
 			break;
-		} else {
+		else
 			msleep(I40EVF_RESET_WAIT_MS);
-		}
 	}
 	if (i == I40EVF_RESET_WAIT_COUNT) {
-		dev_err(&adapter->pdev->dev, "Reset was not detected\n");
 		adapter->flags &= ~I40EVF_FLAG_RESET_PENDING;
 		goto continue_reset; /* act like the reset happened */
 	}
@@ -1521,22 +1502,24 @@ static void i40evf_reset_task(struct work_struct *work)
 	for (i = 0; i < I40EVF_RESET_WAIT_COUNT; i++) {
 		rstat_val = rd32(hw, I40E_VFGEN_RSTAT) &
 			    I40E_VFGEN_RSTAT_VFR_STATE_MASK;
-		if (rstat_val == I40E_VFR_VFACTIVE) {
-			dev_info(&adapter->pdev->dev, "Reset is complete. Reinitializing.\n");
+		if (rstat_val == I40E_VFR_VFACTIVE)
 			break;
-		} else {
+		else
 			msleep(I40EVF_RESET_WAIT_MS);
-		}
 	}
 	if (i == I40EVF_RESET_WAIT_COUNT) {
 		/* reset never finished */
-		dev_err(&adapter->pdev->dev, "Reset never finished (%x). PF driver is dead, and so am I.\n",
+		dev_err(&adapter->pdev->dev, "Reset never finished (%x)\n",
 			rstat_val);
 		adapter->flags |= I40EVF_FLAG_PF_COMMS_FAILED;
 
-		if (netif_running(adapter->netdev))
-			i40evf_close(adapter->netdev);
-
+		if (netif_running(adapter->netdev)) {
+			set_bit(__I40E_DOWN, &adapter->vsi.state);
+			i40evf_down(adapter);
+			i40evf_free_traffic_irqs(adapter);
+			i40evf_free_all_tx_resources(adapter);
+			i40evf_free_all_rx_resources(adapter);
+		}
 		i40evf_free_misc_irq(adapter);
 		i40evf_reset_interrupt_capability(adapter);
 		i40evf_free_queues(adapter);
@@ -1591,7 +1574,7 @@ continue_reset:
 	}
 	return;
 reset_err:
-	dev_err(&adapter->pdev->dev, "failed to allocate resources during reinit.\n");
+	dev_err(&adapter->pdev->dev, "failed to allocate resources during reinit\n");
 	i40evf_close(adapter->netdev);
 }
 
@@ -1607,6 +1590,7 @@ static void i40evf_adminq_task(struct work_struct *work)
 	struct i40e_arq_event_info event;
 	struct i40e_virtchnl_msg *v_msg;
 	i40e_status ret;
+	u32 val, oldval;
 	u16 pending;
 
 	if (adapter->flags & I40EVF_FLAG_PF_COMMS_FAILED)
@@ -1614,11 +1598,9 @@ static void i40evf_adminq_task(struct work_struct *work)
 
 	event.msg_size = I40EVF_MAX_AQ_BUF_SIZE;
 	event.msg_buf = kzalloc(event.msg_size, GFP_KERNEL);
-	if (!event.msg_buf) {
-		dev_info(&adapter->pdev->dev, "%s: no memory for ARQ clean\n",
-				 __func__);
+	if (!event.msg_buf)
 		return;
-	}
+
 	v_msg = (struct i40e_virtchnl_msg *)&event.desc;
 	do {
 		ret = i40evf_clean_arq_element(hw, &event, &pending);
@@ -1636,6 +1618,41 @@ static void i40evf_adminq_task(struct work_struct *work)
 		}
 	} while (pending);
 
+	/* check for error indications */
+	val = rd32(hw, hw->aq.arq.len);
+	oldval = val;
+	if (val & I40E_VF_ARQLEN_ARQVFE_MASK) {
+		dev_info(&adapter->pdev->dev, "ARQ VF Error detected\n");
+		val &= ~I40E_VF_ARQLEN_ARQVFE_MASK;
+	}
+	if (val & I40E_VF_ARQLEN_ARQOVFL_MASK) {
+		dev_info(&adapter->pdev->dev, "ARQ Overflow Error detected\n");
+		val &= ~I40E_VF_ARQLEN_ARQOVFL_MASK;
+	}
+	if (val & I40E_VF_ARQLEN_ARQCRIT_MASK) {
+		dev_info(&adapter->pdev->dev, "ARQ Critical Error detected\n");
+		val &= ~I40E_VF_ARQLEN_ARQCRIT_MASK;
+	}
+	if (oldval != val)
+		wr32(hw, hw->aq.arq.len, val);
+
+	val = rd32(hw, hw->aq.asq.len);
+	oldval = val;
+	if (val & I40E_VF_ATQLEN_ATQVFE_MASK) {
+		dev_info(&adapter->pdev->dev, "ASQ VF Error detected\n");
+		val &= ~I40E_VF_ATQLEN_ATQVFE_MASK;
+	}
+	if (val & I40E_VF_ATQLEN_ATQOVFL_MASK) {
+		dev_info(&adapter->pdev->dev, "ASQ Overflow Error detected\n");
+		val &= ~I40E_VF_ATQLEN_ATQOVFL_MASK;
+	}
+	if (val & I40E_VF_ATQLEN_ATQCRIT_MASK) {
+		dev_info(&adapter->pdev->dev, "ASQ Critical Error detected\n");
+		val &= ~I40E_VF_ATQLEN_ATQCRIT_MASK;
+	}
+	if (oldval != val)
+		wr32(hw, hw->aq.asq.len, val);
+
 	/* re-enable Admin queue interrupt cause */
 	i40evf_misc_irq_enable(adapter);
 
@@ -1673,6 +1690,7 @@ static int i40evf_setup_all_tx_resources(struct i40evf_adapter *adapter)
 	int i, err = 0;
 
 	for (i = 0; i < adapter->vsi_res->num_queue_pairs; i++) {
+		adapter->tx_rings[i]->count = adapter->tx_desc_count;
 		err = i40evf_setup_tx_descriptors(adapter->tx_rings[i]);
 		if (!err)
 			continue;
@@ -1700,6 +1718,7 @@ static int i40evf_setup_all_rx_resources(struct i40evf_adapter *adapter)
 	int i, err = 0;
 
 	for (i = 0; i < adapter->vsi_res->num_queue_pairs; i++) {
+		adapter->rx_rings[i]->count = adapter->rx_desc_count;
 		err = i40evf_setup_rx_descriptors(adapter->rx_rings[i]);
 		if (!err)
 			continue;
@@ -1804,12 +1823,11 @@ static int i40evf_close(struct net_device *netdev)
 	if (adapter->state <= __I40EVF_DOWN)
 		return 0;
 
-	/* signal that we are down to the interrupt handler */
-	adapter->state = __I40EVF_DOWN;
 
 	set_bit(__I40E_DOWN, &adapter->vsi.state);
 
 	i40evf_down(adapter);
+	adapter->state = __I40EVF_DOWN;
 	i40evf_free_traffic_irqs(adapter);
 
 	i40evf_free_all_tx_resources(adapter);
@@ -1848,8 +1866,6 @@ void i40evf_reinit_locked(struct i40evf_adapter *adapter)
 
 	WARN_ON(in_interrupt());
 
-	adapter->state = __I40EVF_RESETTING;
-
 	i40evf_down(adapter);
 
 	/* allocate transmit descriptors */
@@ -1872,7 +1888,7 @@ void i40evf_reinit_locked(struct i40evf_adapter *adapter)
 	return;
 
 err_reinit:
-	dev_err(&adapter->pdev->dev, "failed to allocate resources during reinit.\n");
+	dev_err(&adapter->pdev->dev, "failed to allocate resources during reinit\n");
 	i40evf_close(netdev);
 }
 
@@ -1967,7 +1983,7 @@ static void i40evf_init_task(struct work_struct *work)
 		}
 		err = i40evf_check_reset_complete(hw);
 		if (err) {
-			dev_err(&pdev->dev, "Device is still in reset (%d)\n",
+			dev_info(&pdev->dev, "Device is still in reset (%d), retrying\n",
 				err);
 			goto err;
 		}
@@ -1993,14 +2009,14 @@ static void i40evf_init_task(struct work_struct *work)
 		break;
 	case __I40EVF_INIT_VERSION_CHECK:
 		if (!i40evf_asq_done(hw)) {
-			dev_err(&pdev->dev, "Admin queue command never completed.\n");
+			dev_err(&pdev->dev, "Admin queue command never completed\n");
 			goto err;
 		}
 
 		/* aq msg sent, awaiting reply */
 		err = i40evf_verify_api_ver(adapter);
 		if (err) {
-			dev_err(&pdev->dev, "Unable to verify API version (%d)\n",
+			dev_info(&pdev->dev, "Unable to verify API version (%d), retrying\n",
 				err);
 			goto err;
 		}
@@ -2074,12 +2090,12 @@ static void i40evf_init_task(struct work_struct *work)
 	netdev->hw_features &= ~NETIF_F_RXCSUM;
 
 	if (!is_valid_ether_addr(adapter->hw.mac.addr)) {
-		dev_info(&pdev->dev, "Invalid MAC address %pMAC, using random\n",
+		dev_info(&pdev->dev, "Invalid MAC address %pM, using random\n",
 			 adapter->hw.mac.addr);
 		random_ether_addr(adapter->hw.mac.addr);
 	}
-	memcpy(netdev->dev_addr, adapter->hw.mac.addr, netdev->addr_len);
-	memcpy(netdev->perm_addr, adapter->hw.mac.addr, netdev->addr_len);
+	ether_addr_copy(netdev->dev_addr, adapter->hw.mac.addr);
+	ether_addr_copy(netdev->perm_addr, adapter->hw.mac.addr);
 
 	INIT_LIST_HEAD(&adapter->mac_filter_list);
 	INIT_LIST_HEAD(&adapter->vlan_filter_list);
@@ -2087,7 +2103,7 @@ static void i40evf_init_task(struct work_struct *work)
 	if (NULL == f)
 		goto err_sw_init;
 
-	memcpy(f->macaddr, adapter->hw.mac.addr, ETH_ALEN);
+	ether_addr_copy(f->macaddr, adapter->hw.mac.addr);
 	f->add = true;
 	adapter->aq_required |= I40EVF_FLAG_AQ_ADD_MAC_FILTER;
 
@@ -2098,6 +2114,8 @@ static void i40evf_init_task(struct work_struct *work)
 	adapter->watchdog_timer.data = (unsigned long)adapter;
 	mod_timer(&adapter->watchdog_timer, jiffies + 1);
 
+	adapter->tx_desc_count = I40EVF_DEFAULT_TXD;
+	adapter->rx_desc_count = I40EVF_DEFAULT_RXD;
 	err = i40evf_init_interrupt_scheme(adapter);
 	if (err)
 		goto err_sw_init;
@@ -2114,8 +2132,10 @@ static void i40evf_init_task(struct work_struct *work)
 	adapter->vsi.back = adapter;
 	adapter->vsi.base_vector = 1;
 	adapter->vsi.work_limit = I40E_DEFAULT_IRQ_WORK;
-	adapter->vsi.rx_itr_setting = I40E_ITR_DYNAMIC;
-	adapter->vsi.tx_itr_setting = I40E_ITR_DYNAMIC;
+	adapter->vsi.rx_itr_setting = (I40E_ITR_DYNAMIC |
+				       ITR_REG_TO_USEC(I40E_ITR_RX_DEF));
+	adapter->vsi.tx_itr_setting = (I40E_ITR_DYNAMIC |
+				       ITR_REG_TO_USEC(I40E_ITR_TX_DEF));
 	adapter->vsi.netdev = adapter->netdev;
 
 	if (!adapter->netdev_registered) {
@@ -2128,7 +2148,7 @@ static void i40evf_init_task(struct work_struct *work)
 
 	netif_tx_stop_all_queues(netdev);
 
-	dev_info(&pdev->dev, "MAC address: %pMAC\n", adapter->hw.mac.addr);
+	dev_info(&pdev->dev, "MAC address: %pM\n", adapter->hw.mac.addr);
 	if (netdev->features & NETIF_F_GRO)
 		dev_info(&pdev->dev, "GRO is enabled\n");
 
@@ -2152,12 +2172,11 @@ err_alloc:
 err:
 	/* Things went into the weeds, so try again later */
 	if (++adapter->aq_wait_count > I40EVF_AQ_MAX_ERR) {
-		dev_err(&pdev->dev, "Failed to communicate with PF; giving up.\n");
+		dev_err(&pdev->dev, "Failed to communicate with PF; giving up\n");
 		adapter->flags |= I40EVF_FLAG_PF_COMMS_FAILED;
 		return; /* do not reschedule */
 	}
 	schedule_delayed_work(&adapter->init_task, HZ * 3);
-	return;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index e294f012647d..2dc0bac76717 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -12,6 +12,9 @@
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
  * more details.
  *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
  * The full GNU General Public License is included in this distribution in
  * the file called "COPYING".
  *
@@ -216,11 +219,9 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter)
 	len = sizeof(struct i40e_virtchnl_vsi_queue_config_info) +
 		       (sizeof(struct i40e_virtchnl_queue_pair_info) * pairs);
 	vqci = kzalloc(len, GFP_ATOMIC);
-	if (!vqci) {
-		dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
-			__func__);
+	if (!vqci)
 		return;
-	}
+
 	vqci->vsi_id = adapter->vsi_res->vsi_id;
 	vqci->num_queue_pairs = pairs;
 	vqpi = vqci->qpair;
@@ -232,6 +233,9 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter)
 		vqpi->txq.queue_id = i;
 		vqpi->txq.ring_len = adapter->tx_rings[i]->count;
 		vqpi->txq.dma_ring_addr = adapter->tx_rings[i]->dma;
+		vqpi->txq.headwb_enabled = 1;
+		vqpi->txq.dma_headwb_addr = vqpi->txq.dma_ring_addr +
+		    (vqpi->txq.ring_len * sizeof(struct i40e_tx_desc));
 
 		vqpi->rxq.vsi_id = vqci->vsi_id;
 		vqpi->rxq.queue_id = i;
@@ -329,11 +333,8 @@ void i40evf_map_queues(struct i40evf_adapter *adapter)
 	      (adapter->num_msix_vectors *
 		sizeof(struct i40e_virtchnl_vector_map));
 	vimi = kzalloc(len, GFP_ATOMIC);
-	if (!vimi) {
-		dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
-			__func__);
+	if (!vimi)
 		return;
-	}
 
 	vimi->num_vectors = adapter->num_msix_vectors;
 	/* Queue vectors first */
@@ -390,7 +391,7 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter)
 	len = sizeof(struct i40e_virtchnl_ether_addr_list) +
 	      (count * sizeof(struct i40e_virtchnl_ether_addr));
 	if (len > I40EVF_MAX_AQ_BUF_SIZE) {
-		dev_warn(&adapter->pdev->dev, "%s: Too many MAC address changes in one request.\n",
+		dev_warn(&adapter->pdev->dev, "%s: Too many MAC address changes in one request\n",
 			__func__);
 		count = (I40EVF_MAX_AQ_BUF_SIZE -
 			 sizeof(struct i40e_virtchnl_ether_addr_list)) /
@@ -399,16 +400,14 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter)
 	}
 
 	veal = kzalloc(len, GFP_ATOMIC);
-	if (!veal) {
-		dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
-			__func__);
+	if (!veal)
 		return;
-	}
+
 	veal->vsi_id = adapter->vsi_res->vsi_id;
 	veal->num_elements = count;
 	list_for_each_entry(f, &adapter->mac_filter_list, list) {
 		if (f->add) {
-			memcpy(veal->list[i].addr, f->macaddr, ETH_ALEN);
+			ether_addr_copy(veal->list[i].addr, f->macaddr);
 			i++;
 			f->add = false;
 		}
@@ -454,7 +453,7 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter)
 	len = sizeof(struct i40e_virtchnl_ether_addr_list) +
 	      (count * sizeof(struct i40e_virtchnl_ether_addr));
 	if (len > I40EVF_MAX_AQ_BUF_SIZE) {
-		dev_warn(&adapter->pdev->dev, "%s: Too many MAC address changes in one request.\n",
+		dev_warn(&adapter->pdev->dev, "%s: Too many MAC address changes in one request\n",
 			__func__);
 		count = (I40EVF_MAX_AQ_BUF_SIZE -
 			 sizeof(struct i40e_virtchnl_ether_addr_list)) /
@@ -462,16 +461,14 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter)
 		len = I40EVF_MAX_AQ_BUF_SIZE;
 	}
 	veal = kzalloc(len, GFP_ATOMIC);
-	if (!veal) {
-		dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
-			__func__);
+	if (!veal)
 		return;
-	}
+
 	veal->vsi_id = adapter->vsi_res->vsi_id;
 	veal->num_elements = count;
 	list_for_each_entry_safe(f, ftmp, &adapter->mac_filter_list, list) {
 		if (f->remove) {
-			memcpy(veal->list[i].addr, f->macaddr, ETH_ALEN);
+			ether_addr_copy(veal->list[i].addr, f->macaddr);
 			i++;
 			list_del(&f->list);
 			kfree(f);
@@ -518,7 +515,7 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter)
 	len = sizeof(struct i40e_virtchnl_vlan_filter_list) +
 	      (count * sizeof(u16));
 	if (len > I40EVF_MAX_AQ_BUF_SIZE) {
-		dev_warn(&adapter->pdev->dev, "%s: Too many VLAN changes in one request.\n",
+		dev_warn(&adapter->pdev->dev, "%s: Too many VLAN changes in one request\n",
 			__func__);
 		count = (I40EVF_MAX_AQ_BUF_SIZE -
 			 sizeof(struct i40e_virtchnl_vlan_filter_list)) /
@@ -526,11 +523,9 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter)
 		len = I40EVF_MAX_AQ_BUF_SIZE;
 	}
 	vvfl = kzalloc(len, GFP_ATOMIC);
-	if (!vvfl) {
-		dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
-			__func__);
+	if (!vvfl)
 		return;
-	}
+
 	vvfl->vsi_id = adapter->vsi_res->vsi_id;
 	vvfl->num_elements = count;
 	list_for_each_entry(f, &adapter->vlan_filter_list, list) {
@@ -580,7 +575,7 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter)
 	len = sizeof(struct i40e_virtchnl_vlan_filter_list) +
 	      (count * sizeof(u16));
 	if (len > I40EVF_MAX_AQ_BUF_SIZE) {
-		dev_warn(&adapter->pdev->dev, "%s: Too many VLAN changes in one request.\n",
+		dev_warn(&adapter->pdev->dev, "%s: Too many VLAN changes in one request\n",
 			__func__);
 		count = (I40EVF_MAX_AQ_BUF_SIZE -
 			 sizeof(struct i40e_virtchnl_vlan_filter_list)) /
@@ -588,11 +583,9 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter)
 		len = I40EVF_MAX_AQ_BUF_SIZE;
 	}
 	vvfl = kzalloc(len, GFP_ATOMIC);
-	if (!vvfl) {
-		dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
-			__func__);
+	if (!vvfl)
 		return;
-	}
+
 	vvfl->vsi_id = adapter->vsi_res->vsi_id;
 	vvfl->num_elements = count;
 	list_for_each_entry_safe(f, ftmp, &adapter->vlan_filter_list, list) {
@@ -721,7 +714,7 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 		return;
 	}
 	if (v_opcode != adapter->current_op) {
-		dev_err(&adapter->pdev->dev, "%s: Pending op is %d, received %d.\n",
+		dev_err(&adapter->pdev->dev, "%s: Pending op is %d, received %d\n",
 			__func__, adapter->current_op, v_opcode);
 		/* We're probably completely screwed at this point, but clear
 		 * the current op and try to carry on....
@@ -730,7 +723,7 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 		return;
 	}
 	if (v_retval) {
-		dev_err(&adapter->pdev->dev, "%s: PF returned error %d to our request %d!\n",
+		dev_err(&adapter->pdev->dev, "%s: PF returned error %d to our request %d\n",
 			__func__, v_retval, v_opcode);
 	}
 	switch (v_opcode) {
@@ -745,9 +738,8 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 						 stats->tx_broadcast;
 		adapter->net_stats.rx_bytes = stats->rx_bytes;
 		adapter->net_stats.tx_bytes = stats->tx_bytes;
-		adapter->net_stats.rx_errors = stats->rx_errors;
 		adapter->net_stats.tx_errors = stats->tx_errors;
-		adapter->net_stats.rx_dropped = stats->rx_missed;
+		adapter->net_stats.rx_dropped = stats->rx_discards;
 		adapter->net_stats.tx_dropped = stats->tx_discards;
 		adapter->current_stats = *stats;
 		}
@@ -781,7 +773,7 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 		adapter->aq_pending &= ~(I40EVF_FLAG_AQ_MAP_VECTORS);
 		break;
 	default:
-		dev_warn(&adapter->pdev->dev, "%s: Received unexpected message %d from PF.\n",
+		dev_warn(&adapter->pdev->dev, "%s: Received unexpected message %d from PF\n",
 			__func__, v_opcode);
 		break;
 	} /* switch v_opcode */
diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.c b/drivers/net/ethernet/intel/igb/e1000_82575.c
index fa36fe12e775..a2db388cc31e 100644
--- a/drivers/net/ethernet/intel/igb/e1000_82575.c
+++ b/drivers/net/ethernet/intel/igb/e1000_82575.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 /* e1000_82575
  * e1000_82576
@@ -73,9 +70,8 @@ static s32  igb_validate_nvm_checksum_82580(struct e1000_hw *hw);
 static s32  igb_update_nvm_checksum_82580(struct e1000_hw *hw);
 static s32 igb_validate_nvm_checksum_i350(struct e1000_hw *hw);
 static s32 igb_update_nvm_checksum_i350(struct e1000_hw *hw);
-static const u16 e1000_82580_rxpbs_table[] =
-	{ 36, 72, 144, 1, 2, 4, 8, 16,
-	  35, 70, 140 };
+static const u16 e1000_82580_rxpbs_table[] = {
+	36, 72, 144, 1, 2, 4, 8, 16, 35, 70, 140 };
 
 /**
  *  igb_sgmii_uses_mdio_82575 - Determine if I2C pins are for external MDIO
@@ -159,7 +155,7 @@ static s32 igb_check_for_link_media_swap(struct e1000_hw *hw)
 		ret_val = igb_check_for_link_82575(hw);
 	}
 
-	return E1000_SUCCESS;
+	return 0;
 }
 
 /**
@@ -526,7 +522,7 @@ out:
 static s32 igb_get_invariants_82575(struct e1000_hw *hw)
 {
 	struct e1000_mac_info *mac = &hw->mac;
-	struct e1000_dev_spec_82575 * dev_spec = &hw->dev_spec._82575;
+	struct e1000_dev_spec_82575 *dev_spec = &hw->dev_spec._82575;
 	s32 ret_val;
 	u32 ctrl_ext = 0;
 	u32 link_mode = 0;
@@ -1008,7 +1004,6 @@ out:
 static s32 igb_set_d0_lplu_state_82580(struct e1000_hw *hw, bool active)
 {
 	struct e1000_phy_info *phy = &hw->phy;
-	s32 ret_val = 0;
 	u16 data;
 
 	data = rd32(E1000_82580_PHY_POWER_MGMT);
@@ -1032,7 +1027,7 @@ static s32 igb_set_d0_lplu_state_82580(struct e1000_hw *hw, bool active)
 			data &= ~E1000_82580_PM_SPD; }
 
 	wr32(E1000_82580_PHY_POWER_MGMT, data);
-	return ret_val;
+	return 0;
 }
 
 /**
@@ -1052,7 +1047,6 @@ static s32 igb_set_d0_lplu_state_82580(struct e1000_hw *hw, bool active)
 static s32 igb_set_d3_lplu_state_82580(struct e1000_hw *hw, bool active)
 {
 	struct e1000_phy_info *phy = &hw->phy;
-	s32 ret_val = 0;
 	u16 data;
 
 	data = rd32(E1000_82580_PHY_POWER_MGMT);
@@ -1077,7 +1071,7 @@ static s32 igb_set_d3_lplu_state_82580(struct e1000_hw *hw, bool active)
 	}
 
 	wr32(E1000_82580_PHY_POWER_MGMT, data);
-	return ret_val;
+	return 0;
 }
 
 /**
@@ -1180,8 +1174,8 @@ static void igb_release_swfw_sync_82575(struct e1000_hw *hw, u16 mask)
 {
 	u32 swfw_sync;
 
-	while (igb_get_hw_semaphore(hw) != 0);
-	/* Empty */
+	while (igb_get_hw_semaphore(hw) != 0)
+		; /* Empty */
 
 	swfw_sync = rd32(E1000_SW_FW_SYNC);
 	swfw_sync &= ~mask;
@@ -1203,7 +1197,6 @@ static void igb_release_swfw_sync_82575(struct e1000_hw *hw, u16 mask)
 static s32 igb_get_cfg_done_82575(struct e1000_hw *hw)
 {
 	s32 timeout = PHY_CFG_TIMEOUT;
-	s32 ret_val = 0;
 	u32 mask = E1000_NVM_CFG_DONE_PORT_0;
 
 	if (hw->bus.func == 1)
@@ -1216,7 +1209,7 @@ static s32 igb_get_cfg_done_82575(struct e1000_hw *hw)
 	while (timeout) {
 		if (rd32(E1000_EEMNGCTL) & mask)
 			break;
-		msleep(1);
+		usleep_range(1000, 2000);
 		timeout--;
 	}
 	if (!timeout)
@@ -1227,7 +1220,7 @@ static s32 igb_get_cfg_done_82575(struct e1000_hw *hw)
 	    (hw->phy.type == e1000_phy_igp_3))
 		igb_phy_init_script_igp3(hw);
 
-	return ret_val;
+	return 0;
 }
 
 /**
@@ -1269,7 +1262,7 @@ static s32 igb_check_for_link_82575(struct e1000_hw *hw)
 
 	if (hw->phy.media_type != e1000_media_type_copper) {
 		ret_val = igb_get_pcs_speed_and_duplex_82575(hw, &speed,
-		                                             &duplex);
+							     &duplex);
 		/* Use this flag to determine if link needs to be checked or
 		 * not.  If  we have link clear the flag so that we do not
 		 * continue to check for link.
@@ -1316,7 +1309,7 @@ void igb_power_up_serdes_link_82575(struct e1000_hw *hw)
 
 	/* flush the write to verify completion */
 	wrfl();
-	msleep(1);
+	usleep_range(1000, 2000);
 }
 
 /**
@@ -1411,7 +1404,7 @@ void igb_shutdown_serdes_link_82575(struct e1000_hw *hw)
 
 		/* flush the write to verify completion */
 		wrfl();
-		msleep(1);
+		usleep_range(1000, 2000);
 	}
 }
 
@@ -1436,9 +1429,8 @@ static s32 igb_reset_hw_82575(struct e1000_hw *hw)
 
 	/* set the completion timeout for interface */
 	ret_val = igb_set_pcie_completion_timeout(hw);
-	if (ret_val) {
+	if (ret_val)
 		hw_dbg("PCI-E Set completion timeout has failed.\n");
-	}
 
 	hw_dbg("Masking off all interrupts\n");
 	wr32(E1000_IMC, 0xffffffff);
@@ -1447,7 +1439,7 @@ static s32 igb_reset_hw_82575(struct e1000_hw *hw)
 	wr32(E1000_TCTL, E1000_TCTL_PSP);
 	wrfl();
 
-	msleep(10);
+	usleep_range(10000, 20000);
 
 	ctrl = rd32(E1000_CTRL);
 
@@ -1622,7 +1614,7 @@ static s32 igb_setup_serdes_link_82575(struct e1000_hw *hw)
 {
 	u32 ctrl_ext, ctrl_reg, reg, anadv_reg;
 	bool pcs_autoneg;
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 	u16 data;
 
 	if ((hw->phy.media_type != e1000_media_type_internal_serdes) &&
@@ -1676,7 +1668,7 @@ static s32 igb_setup_serdes_link_82575(struct e1000_hw *hw)
 		    hw->mac.type == e1000_82576) {
 			ret_val = hw->nvm.ops.read(hw, NVM_COMPAT, 1, &data);
 			if (ret_val) {
-				printk(KERN_DEBUG "NVM Read Error\n\n");
+				hw_dbg(KERN_DEBUG "NVM Read Error\n\n");
 				return ret_val;
 			}
 
@@ -1689,7 +1681,7 @@ static s32 igb_setup_serdes_link_82575(struct e1000_hw *hw)
 		 * link either autoneg or be forced to 1000/Full
 		 */
 		ctrl_reg |= E1000_CTRL_SPD_1000 | E1000_CTRL_FRCSPD |
-		            E1000_CTRL_FD | E1000_CTRL_FRCDPX;
+				E1000_CTRL_FD | E1000_CTRL_FRCDPX;
 
 		/* set speed of 1000/Full if speed/duplex is forced */
 		reg |= E1000_PCS_LCTL_FSV_1000 | E1000_PCS_LCTL_FDV_FULL;
@@ -1925,7 +1917,7 @@ void igb_rx_fifo_flush_82575(struct e1000_hw *hw)
 	}
 	/* Poll all queues to verify they have shut down */
 	for (ms_wait = 0; ms_wait < 10; ms_wait++) {
-		msleep(1);
+		usleep_range(1000, 2000);
 		rx_enabled = 0;
 		for (i = 0; i < 4; i++)
 			rx_enabled |= rd32(E1000_RXDCTL(i));
@@ -1953,7 +1945,7 @@ void igb_rx_fifo_flush_82575(struct e1000_hw *hw)
 	wr32(E1000_RCTL, temp_rctl);
 	wr32(E1000_RCTL, temp_rctl | E1000_RCTL_EN);
 	wrfl();
-	msleep(2);
+	usleep_range(2000, 3000);
 
 	/* Enable RX queues that were previously enabled and restore our
 	 * previous state
@@ -2005,14 +1997,14 @@ static s32 igb_set_pcie_completion_timeout(struct e1000_hw *hw)
 	 * 16ms to 55ms
 	 */
 	ret_val = igb_read_pcie_cap_reg(hw, PCIE_DEVICE_CONTROL2,
-	                                &pcie_devctl2);
+					&pcie_devctl2);
 	if (ret_val)
 		goto out;
 
 	pcie_devctl2 |= PCIE_DEVICE_CONTROL2_16ms;
 
 	ret_val = igb_write_pcie_cap_reg(hw, PCIE_DEVICE_CONTROL2,
-	                                 &pcie_devctl2);
+					 &pcie_devctl2);
 out:
 	/* disable completion timeout resend */
 	gcr &= ~E1000_GCR_CMPL_TMOUT_RESEND;
@@ -2241,7 +2233,7 @@ static s32 igb_reset_hw_82580(struct e1000_hw *hw)
 	wr32(E1000_TCTL, E1000_TCTL_PSP);
 	wrfl();
 
-	msleep(10);
+	usleep_range(10000, 11000);
 
 	/* Determine whether or not a global dev reset is requested */
 	if (global_device_reset &&
@@ -2259,7 +2251,7 @@ static s32 igb_reset_hw_82580(struct e1000_hw *hw)
 
 	/* Add delay to insure DEV_RST has time to complete */
 	if (global_device_reset)
-		msleep(5);
+		usleep_range(5000, 6000);
 
 	ret_val = igb_get_auto_rd_done(hw);
 	if (ret_val) {
@@ -2436,8 +2428,7 @@ static s32 igb_update_nvm_checksum_82580(struct e1000_hw *hw)
 
 	ret_val = hw->nvm.ops.read(hw, NVM_COMPATIBILITY_REG_3, 1, &nvm_data);
 	if (ret_val) {
-		hw_dbg("NVM Read Error while updating checksum"
-			" compatibility bit.\n");
+		hw_dbg("NVM Read Error while updating checksum compatibility bit.\n");
 		goto out;
 	}
 
@@ -2447,8 +2438,7 @@ static s32 igb_update_nvm_checksum_82580(struct e1000_hw *hw)
 		ret_val = hw->nvm.ops.write(hw, NVM_COMPATIBILITY_REG_3, 1,
 					&nvm_data);
 		if (ret_val) {
-			hw_dbg("NVM Write Error while updating checksum"
-				" compatibility bit.\n");
+			hw_dbg("NVM Write Error while updating checksum compatibility bit.\n");
 			goto out;
 		}
 	}
@@ -2525,7 +2515,7 @@ out:
 static s32 __igb_access_emi_reg(struct e1000_hw *hw, u16 address,
 				  u16 *data, bool read)
 {
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 
 	ret_val = hw->phy.ops.write_reg(hw, E1000_EMIADD, address);
 	if (ret_val)
@@ -2559,7 +2549,6 @@ s32 igb_read_emi_reg(struct e1000_hw *hw, u16 addr, u16 *data)
  **/
 s32 igb_set_eee_i350(struct e1000_hw *hw)
 {
-	s32 ret_val = 0;
 	u32 ipcnfg, eeer;
 
 	if ((hw->mac.type < e1000_i350) ||
@@ -2593,7 +2582,7 @@ s32 igb_set_eee_i350(struct e1000_hw *hw)
 	rd32(E1000_EEER);
 out:
 
-	return ret_val;
+	return 0;
 }
 
 /**
@@ -2720,7 +2709,6 @@ static const u8 e1000_emc_therm_limit[4] = {
  **/
 static s32 igb_get_thermal_sensor_data_generic(struct e1000_hw *hw)
 {
-	s32 status = E1000_SUCCESS;
 	u16 ets_offset;
 	u16 ets_cfg;
 	u16 ets_sensor;
@@ -2738,7 +2726,7 @@ static s32 igb_get_thermal_sensor_data_generic(struct e1000_hw *hw)
 	/* Return the internal sensor only if ETS is unsupported */
 	hw->nvm.ops.read(hw, NVM_ETS_CFG, 1, &ets_offset);
 	if ((ets_offset == 0x0000) || (ets_offset == 0xFFFF))
-		return status;
+		return 0;
 
 	hw->nvm.ops.read(hw, ets_offset, 1, &ets_cfg);
 	if (((ets_cfg & NVM_ETS_TYPE_MASK) >> NVM_ETS_TYPE_SHIFT)
@@ -2762,7 +2750,7 @@ static s32 igb_get_thermal_sensor_data_generic(struct e1000_hw *hw)
 					E1000_I2C_THERMAL_SENSOR_ADDR,
 					&data->sensor[i].temp);
 	}
-	return status;
+	return 0;
 }
 
 /**
@@ -2774,7 +2762,6 @@ static s32 igb_get_thermal_sensor_data_generic(struct e1000_hw *hw)
  **/
 static s32 igb_init_thermal_sensor_thresh_generic(struct e1000_hw *hw)
 {
-	s32 status = E1000_SUCCESS;
 	u16 ets_offset;
 	u16 ets_cfg;
 	u16 ets_sensor;
@@ -2800,7 +2787,7 @@ static s32 igb_init_thermal_sensor_thresh_generic(struct e1000_hw *hw)
 	/* Return the internal sensor only if ETS is unsupported */
 	hw->nvm.ops.read(hw, NVM_ETS_CFG, 1, &ets_offset);
 	if ((ets_offset == 0x0000) || (ets_offset == 0xFFFF))
-		return status;
+		return 0;
 
 	hw->nvm.ops.read(hw, ets_offset, 1, &ets_cfg);
 	if (((ets_cfg & NVM_ETS_TYPE_MASK) >> NVM_ETS_TYPE_SHIFT)
@@ -2831,7 +2818,7 @@ static s32 igb_init_thermal_sensor_thresh_generic(struct e1000_hw *hw)
 							low_thresh_delta;
 		}
 	}
-	return status;
+	return 0;
 }
 
 #endif
diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.h b/drivers/net/ethernet/intel/igb/e1000_82575.h
index 09d78be72416..b407c55738fa 100644
--- a/drivers/net/ethernet/intel/igb/e1000_82575.h
+++ b/drivers/net/ethernet/intel/igb/e1000_82575.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_82575_H_
 #define _E1000_82575_H_
@@ -37,9 +34,9 @@ s32 igb_write_i2c_byte(struct e1000_hw *hw, u8 byte_offset, u8 dev_addr,
 		       u8 data);
 
 #define ID_LED_DEFAULT_82575_SERDES ((ID_LED_DEF1_DEF2 << 12) | \
-                                     (ID_LED_DEF1_DEF2 <<  8) | \
-                                     (ID_LED_DEF1_DEF2 <<  4) | \
-                                     (ID_LED_OFF1_ON2))
+				     (ID_LED_DEF1_DEF2 <<  8) | \
+				     (ID_LED_DEF1_DEF2 <<  4) | \
+				     (ID_LED_OFF1_ON2))
 
 #define E1000_RAR_ENTRIES_82575        16
 #define E1000_RAR_ENTRIES_82576        24
@@ -67,16 +64,16 @@ s32 igb_write_i2c_byte(struct e1000_hw *hw, u8 byte_offset, u8 dev_addr,
 #define E1000_MRQC_RSS_FIELD_IPV6_UDP_EX    0x01000000
 
 #define E1000_EICR_TX_QUEUE ( \
-    E1000_EICR_TX_QUEUE0 |    \
-    E1000_EICR_TX_QUEUE1 |    \
-    E1000_EICR_TX_QUEUE2 |    \
-    E1000_EICR_TX_QUEUE3)
+	E1000_EICR_TX_QUEUE0 |    \
+	E1000_EICR_TX_QUEUE1 |    \
+	E1000_EICR_TX_QUEUE2 |    \
+	E1000_EICR_TX_QUEUE3)
 
 #define E1000_EICR_RX_QUEUE ( \
-    E1000_EICR_RX_QUEUE0 |    \
-    E1000_EICR_RX_QUEUE1 |    \
-    E1000_EICR_RX_QUEUE2 |    \
-    E1000_EICR_RX_QUEUE3)
+	E1000_EICR_RX_QUEUE0 |    \
+	E1000_EICR_RX_QUEUE1 |    \
+	E1000_EICR_RX_QUEUE2 |    \
+	E1000_EICR_RX_QUEUE3)
 
 /* Immediate Interrupt Rx (A.K.A. Low Latency Interrupt) */
 #define E1000_IMIREXT_SIZE_BP     0x00001000  /* Packet size bypass */
@@ -92,8 +89,7 @@ union e1000_adv_rx_desc {
 		struct {
 			struct {
 				__le16 pkt_info;   /* RSS type, Packet type */
-				__le16 hdr_info;   /* Split Header,
-						    * header buffer length */
+				__le16 hdr_info;   /* Split Head, buf len */
 			} lo_dword;
 			union {
 				__le32 rss;          /* RSS Hash */
diff --git a/drivers/net/ethernet/intel/igb/e1000_defines.h b/drivers/net/ethernet/intel/igb/e1000_defines.h
index b05bf925ac72..2a8bb35c2df2 100644
--- a/drivers/net/ethernet/intel/igb/e1000_defines.h
+++ b/drivers/net/ethernet/intel/igb/e1000_defines.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_DEFINES_H_
 #define _E1000_DEFINES_H_
@@ -101,11 +98,11 @@
 
 /* Same mask, but for extended and packet split descriptors */
 #define E1000_RXDEXT_ERR_FRAME_ERR_MASK ( \
-    E1000_RXDEXT_STATERR_CE  |            \
-    E1000_RXDEXT_STATERR_SE  |            \
-    E1000_RXDEXT_STATERR_SEQ |            \
-    E1000_RXDEXT_STATERR_CXE |            \
-    E1000_RXDEXT_STATERR_RXE)
+	E1000_RXDEXT_STATERR_CE  |            \
+	E1000_RXDEXT_STATERR_SE  |            \
+	E1000_RXDEXT_STATERR_SEQ |            \
+	E1000_RXDEXT_STATERR_CXE |            \
+	E1000_RXDEXT_STATERR_RXE)
 
 #define E1000_MRQC_RSS_FIELD_IPV4_TCP          0x00010000
 #define E1000_MRQC_RSS_FIELD_IPV4              0x00020000
@@ -307,39 +304,34 @@
 #define E1000_TCTL_RTLC   0x01000000    /* Re-transmit on late collision */
 
 /* DMA Coalescing register fields */
-#define E1000_DMACR_DMACWT_MASK         0x00003FFF /* DMA Coalescing
-							* Watchdog Timer */
-#define E1000_DMACR_DMACTHR_MASK        0x00FF0000 /* DMA Coalescing Receive
-							* Threshold */
+#define E1000_DMACR_DMACWT_MASK         0x00003FFF /* DMA Coal Watchdog Timer */
+#define E1000_DMACR_DMACTHR_MASK        0x00FF0000 /* DMA Coal Rx Threshold */
 #define E1000_DMACR_DMACTHR_SHIFT       16
-#define E1000_DMACR_DMAC_LX_MASK        0x30000000 /* Lx when no PCIe
-							* transactions */
+#define E1000_DMACR_DMAC_LX_MASK        0x30000000 /* Lx when no PCIe trans */
 #define E1000_DMACR_DMAC_LX_SHIFT       28
 #define E1000_DMACR_DMAC_EN             0x80000000 /* Enable DMA Coalescing */
 /* DMA Coalescing BMC-to-OS Watchdog Enable */
 #define E1000_DMACR_DC_BMC2OSW_EN	0x00008000
 
-#define E1000_DMCTXTH_DMCTTHR_MASK      0x00000FFF /* DMA Coalescing Transmit
-							* Threshold */
+#define E1000_DMCTXTH_DMCTTHR_MASK      0x00000FFF /* DMA Coal Tx Threshold */
 
 #define E1000_DMCTLX_TTLX_MASK          0x00000FFF /* Time to LX request */
 
-#define E1000_DMCRTRH_UTRESH_MASK       0x0007FFFF /* Receive Traffic Rate
-							* Threshold */
-#define E1000_DMCRTRH_LRPRCW            0x80000000 /* Rcv packet rate in
-							* current window */
+#define E1000_DMCRTRH_UTRESH_MASK       0x0007FFFF /* Rx Traffic Rate Thresh */
+#define E1000_DMCRTRH_LRPRCW            0x80000000 /* Rx pkt rate curr window */
 
-#define E1000_DMCCNT_CCOUNT_MASK        0x01FFFFFF /* DMA Coal Rcv Traffic
-							* Current Cnt */
+#define E1000_DMCCNT_CCOUNT_MASK        0x01FFFFFF /* DMA Coal Rx Current Cnt */
 
-#define E1000_FCRTC_RTH_COAL_MASK       0x0003FFF0 /* Flow ctrl Rcv Threshold
-							* High val */
+#define E1000_FCRTC_RTH_COAL_MASK       0x0003FFF0 /* FC Rx Thresh High val */
 #define E1000_FCRTC_RTH_COAL_SHIFT      4
 #define E1000_PCIEMISC_LX_DECISION      0x00000080 /* Lx power decision */
 
 /* Timestamp in Rx buffer */
 #define E1000_RXPBS_CFG_TS_EN           0x80000000
 
+#define I210_RXPBSIZE_DEFAULT		0x000000A2 /* RXPBSIZE default */
+#define I210_TXPBSIZE_DEFAULT		0x04000014 /* TXPBSIZE default */
+
 /* SerDes Control */
 #define E1000_SCTL_DISABLE_SERDES_LOOPBACK 0x0400
 
@@ -406,12 +398,12 @@
  *   o LSC    = Link Status Change
  */
 #define IMS_ENABLE_MASK ( \
-    E1000_IMS_RXT0   |    \
-    E1000_IMS_TXDW   |    \
-    E1000_IMS_RXDMT0 |    \
-    E1000_IMS_RXSEQ  |    \
-    E1000_IMS_LSC    |    \
-    E1000_IMS_DOUTSYNC)
+	E1000_IMS_RXT0   |    \
+	E1000_IMS_TXDW   |    \
+	E1000_IMS_RXDMT0 |    \
+	E1000_IMS_RXSEQ  |    \
+	E1000_IMS_LSC    |    \
+	E1000_IMS_DOUTSYNC)
 
 /* Interrupt Mask Set */
 #define E1000_IMS_TXDW      E1000_ICR_TXDW      /* Transmit desc written back */
@@ -467,7 +459,6 @@
 #define E1000_RAH_POOL_1 0x00040000
 
 /* Error Codes */
-#define E1000_SUCCESS      0
 #define E1000_ERR_NVM      1
 #define E1000_ERR_PHY      2
 #define E1000_ERR_CONFIG   3
@@ -1011,8 +1002,7 @@
 #define E1000_VFTA_ENTRY_BIT_SHIFT_MASK      0x1F
 
 /* DMA Coalescing register fields */
-#define E1000_PCIEMISC_LX_DECISION      0x00000080 /* Lx power decision based
-                                                      on DMA coal */
+#define E1000_PCIEMISC_LX_DECISION      0x00000080 /* Lx power on DMA coal */
 
 /* Tx Rate-Scheduler Config fields */
 #define E1000_RTTBCNRC_RS_ENA		0x80000000
diff --git a/drivers/net/ethernet/intel/igb/e1000_hw.h b/drivers/net/ethernet/intel/igb/e1000_hw.h
index 10741d170f2d..89925e405849 100644
--- a/drivers/net/ethernet/intel/igb/e1000_hw.h
+++ b/drivers/net/ethernet/intel/igb/e1000_hw.h
@@ -1,28 +1,24 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_HW_H_
 #define _E1000_HW_H_
@@ -320,15 +316,15 @@ struct e1000_host_mng_command_info {
 #include "e1000_mbx.h"
 
 struct e1000_mac_operations {
-	s32  (*check_for_link)(struct e1000_hw *);
-	s32  (*reset_hw)(struct e1000_hw *);
-	s32  (*init_hw)(struct e1000_hw *);
+	s32 (*check_for_link)(struct e1000_hw *);
+	s32 (*reset_hw)(struct e1000_hw *);
+	s32 (*init_hw)(struct e1000_hw *);
 	bool (*check_mng_mode)(struct e1000_hw *);
-	s32  (*setup_physical_interface)(struct e1000_hw *);
+	s32 (*setup_physical_interface)(struct e1000_hw *);
 	void (*rar_set)(struct e1000_hw *, u8 *, u32);
-	s32  (*read_mac_addr)(struct e1000_hw *);
-	s32  (*get_speed_and_duplex)(struct e1000_hw *, u16 *, u16 *);
-	s32  (*acquire_swfw_sync)(struct e1000_hw *, u16);
+	s32 (*read_mac_addr)(struct e1000_hw *);
+	s32 (*get_speed_and_duplex)(struct e1000_hw *, u16 *, u16 *);
+	s32 (*acquire_swfw_sync)(struct e1000_hw *, u16);
 	void (*release_swfw_sync)(struct e1000_hw *, u16);
 #ifdef CONFIG_IGB_HWMON
 	s32 (*get_thermal_sensor_data)(struct e1000_hw *);
@@ -338,31 +334,31 @@ struct e1000_mac_operations {
 };
 
 struct e1000_phy_operations {
-	s32  (*acquire)(struct e1000_hw *);
-	s32  (*check_polarity)(struct e1000_hw *);
-	s32  (*check_reset_block)(struct e1000_hw *);
-	s32  (*force_speed_duplex)(struct e1000_hw *);
-	s32  (*get_cfg_done)(struct e1000_hw *hw);
-	s32  (*get_cable_length)(struct e1000_hw *);
-	s32  (*get_phy_info)(struct e1000_hw *);
-	s32  (*read_reg)(struct e1000_hw *, u32, u16 *);
+	s32 (*acquire)(struct e1000_hw *);
+	s32 (*check_polarity)(struct e1000_hw *);
+	s32 (*check_reset_block)(struct e1000_hw *);
+	s32 (*force_speed_duplex)(struct e1000_hw *);
+	s32 (*get_cfg_done)(struct e1000_hw *hw);
+	s32 (*get_cable_length)(struct e1000_hw *);
+	s32 (*get_phy_info)(struct e1000_hw *);
+	s32 (*read_reg)(struct e1000_hw *, u32, u16 *);
 	void (*release)(struct e1000_hw *);
-	s32  (*reset)(struct e1000_hw *);
-	s32  (*set_d0_lplu_state)(struct e1000_hw *, bool);
-	s32  (*set_d3_lplu_state)(struct e1000_hw *, bool);
-	s32  (*write_reg)(struct e1000_hw *, u32, u16);
+	s32 (*reset)(struct e1000_hw *);
+	s32 (*set_d0_lplu_state)(struct e1000_hw *, bool);
+	s32 (*set_d3_lplu_state)(struct e1000_hw *, bool);
+	s32 (*write_reg)(struct e1000_hw *, u32, u16);
 	s32 (*read_i2c_byte)(struct e1000_hw *, u8, u8, u8 *);
 	s32 (*write_i2c_byte)(struct e1000_hw *, u8, u8, u8);
 };
 
 struct e1000_nvm_operations {
-	s32  (*acquire)(struct e1000_hw *);
-	s32  (*read)(struct e1000_hw *, u16, u16, u16 *);
+	s32 (*acquire)(struct e1000_hw *);
+	s32 (*read)(struct e1000_hw *, u16, u16, u16 *);
 	void (*release)(struct e1000_hw *);
-	s32  (*write)(struct e1000_hw *, u16, u16, u16 *);
-	s32  (*update)(struct e1000_hw *);
-	s32  (*validate)(struct e1000_hw *);
-	s32  (*valid_led_default)(struct e1000_hw *, u16 *);
+	s32 (*write)(struct e1000_hw *, u16, u16, u16 *);
+	s32 (*update)(struct e1000_hw *);
+	s32 (*validate)(struct e1000_hw *);
+	s32 (*valid_led_default)(struct e1000_hw *, u16 *);
 };
 
 #define E1000_MAX_SENSORS		3
diff --git a/drivers/net/ethernet/intel/igb/e1000_i210.c b/drivers/net/ethernet/intel/igb/e1000_i210.c
index f67f8a170b90..337161f440dd 100644
--- a/drivers/net/ethernet/intel/igb/e1000_i210.c
+++ b/drivers/net/ethernet/intel/igb/e1000_i210.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 /* e1000_i210
  * e1000_i211
@@ -100,7 +97,7 @@ static s32 igb_get_hw_semaphore_i210(struct e1000_hw *hw)
 		return -E1000_ERR_NVM;
 	}
 
-	return E1000_SUCCESS;
+	return 0;
 }
 
 /**
@@ -142,7 +139,7 @@ s32 igb_acquire_swfw_sync_i210(struct e1000_hw *hw, u16 mask)
 	u32 swfw_sync;
 	u32 swmask = mask;
 	u32 fwmask = mask << 16;
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 	s32 i = 0, timeout = 200; /* FIXME: find real value to use here */
 
 	while (i < timeout) {
@@ -187,7 +184,7 @@ void igb_release_swfw_sync_i210(struct e1000_hw *hw, u16 mask)
 {
 	u32 swfw_sync;
 
-	while (igb_get_hw_semaphore_i210(hw) != E1000_SUCCESS)
+	while (igb_get_hw_semaphore_i210(hw))
 		; /* Empty */
 
 	swfw_sync = rd32(E1000_SW_FW_SYNC);
@@ -210,7 +207,7 @@ void igb_release_swfw_sync_i210(struct e1000_hw *hw, u16 mask)
 static s32 igb_read_nvm_srrd_i210(struct e1000_hw *hw, u16 offset, u16 words,
 				  u16 *data)
 {
-	s32 status = E1000_SUCCESS;
+	s32 status = 0;
 	u16 i, count;
 
 	/* We cannot hold synchronization semaphores for too long,
@@ -220,7 +217,7 @@ static s32 igb_read_nvm_srrd_i210(struct e1000_hw *hw, u16 offset, u16 words,
 	for (i = 0; i < words; i += E1000_EERD_EEWR_MAX_COUNT) {
 		count = (words - i) / E1000_EERD_EEWR_MAX_COUNT > 0 ?
 			E1000_EERD_EEWR_MAX_COUNT : (words - i);
-		if (hw->nvm.ops.acquire(hw) == E1000_SUCCESS) {
+		if (!(hw->nvm.ops.acquire(hw))) {
 			status = igb_read_nvm_eerd(hw, offset, count,
 						     data + i);
 			hw->nvm.ops.release(hw);
@@ -228,7 +225,7 @@ static s32 igb_read_nvm_srrd_i210(struct e1000_hw *hw, u16 offset, u16 words,
 			status = E1000_ERR_SWFW_SYNC;
 		}
 
-		if (status != E1000_SUCCESS)
+		if (status)
 			break;
 	}
 
@@ -253,7 +250,7 @@ static s32 igb_write_nvm_srwr(struct e1000_hw *hw, u16 offset, u16 words,
 	struct e1000_nvm_info *nvm = &hw->nvm;
 	u32 i, k, eewr = 0;
 	u32 attempts = 100000;
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 
 	/* A check for invalid values:  offset too large, too many words,
 	 * too many words for the offset, and not enough words.
@@ -275,13 +272,13 @@ static s32 igb_write_nvm_srwr(struct e1000_hw *hw, u16 offset, u16 words,
 		for (k = 0; k < attempts; k++) {
 			if (E1000_NVM_RW_REG_DONE &
 			    rd32(E1000_SRWR)) {
-				ret_val = E1000_SUCCESS;
+				ret_val = 0;
 				break;
 			}
 			udelay(5);
 	}
 
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			hw_dbg("Shadow RAM write EEWR timed out\n");
 			break;
 		}
@@ -310,7 +307,7 @@ out:
 static s32 igb_write_nvm_srwr_i210(struct e1000_hw *hw, u16 offset, u16 words,
 				   u16 *data)
 {
-	s32 status = E1000_SUCCESS;
+	s32 status = 0;
 	u16 i, count;
 
 	/* We cannot hold synchronization semaphores for too long,
@@ -320,7 +317,7 @@ static s32 igb_write_nvm_srwr_i210(struct e1000_hw *hw, u16 offset, u16 words,
 	for (i = 0; i < words; i += E1000_EERD_EEWR_MAX_COUNT) {
 		count = (words - i) / E1000_EERD_EEWR_MAX_COUNT > 0 ?
 			E1000_EERD_EEWR_MAX_COUNT : (words - i);
-		if (hw->nvm.ops.acquire(hw) == E1000_SUCCESS) {
+		if (!(hw->nvm.ops.acquire(hw))) {
 			status = igb_write_nvm_srwr(hw, offset, count,
 						      data + i);
 			hw->nvm.ops.release(hw);
@@ -328,7 +325,7 @@ static s32 igb_write_nvm_srwr_i210(struct e1000_hw *hw, u16 offset, u16 words,
 			status = E1000_ERR_SWFW_SYNC;
 		}
 
-		if (status != E1000_SUCCESS)
+		if (status)
 			break;
 	}
 
@@ -367,12 +364,12 @@ static s32 igb_read_invm_word_i210(struct e1000_hw *hw, u8 address, u16 *data)
 				*data = INVM_DWORD_TO_WORD_DATA(invm_dword);
 				hw_dbg("Read INVM Word 0x%02x = %x\n",
 					  address, *data);
-				status = E1000_SUCCESS;
+				status = 0;
 				break;
 			}
 		}
 	}
-	if (status != E1000_SUCCESS)
+	if (status)
 		hw_dbg("Requested word 0x%02x not found in OTP\n", address);
 	return status;
 }
@@ -388,7 +385,7 @@ static s32 igb_read_invm_word_i210(struct e1000_hw *hw, u8 address, u16 *data)
 static s32 igb_read_invm_i210(struct e1000_hw *hw, u16 offset,
 				u16 words __always_unused, u16 *data)
 {
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 
 	/* Only the MAC addr is required to be present in the iNVM */
 	switch (offset) {
@@ -398,43 +395,44 @@ static s32 igb_read_invm_i210(struct e1000_hw *hw, u16 offset,
 						     &data[1]);
 		ret_val |= igb_read_invm_word_i210(hw, (u8)offset+2,
 						     &data[2]);
-		if (ret_val != E1000_SUCCESS)
+		if (ret_val)
 			hw_dbg("MAC Addr not found in iNVM\n");
 		break;
 	case NVM_INIT_CTRL_2:
 		ret_val = igb_read_invm_word_i210(hw, (u8)offset, data);
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			*data = NVM_INIT_CTRL_2_DEFAULT_I211;
-			ret_val = E1000_SUCCESS;
+			ret_val = 0;
 		}
 		break;
 	case NVM_INIT_CTRL_4:
 		ret_val = igb_read_invm_word_i210(hw, (u8)offset, data);
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			*data = NVM_INIT_CTRL_4_DEFAULT_I211;
-			ret_val = E1000_SUCCESS;
+			ret_val = 0;
 		}
 		break;
 	case NVM_LED_1_CFG:
 		ret_val = igb_read_invm_word_i210(hw, (u8)offset, data);
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			*data = NVM_LED_1_CFG_DEFAULT_I211;
-			ret_val = E1000_SUCCESS;
+			ret_val = 0;
 		}
 		break;
 	case NVM_LED_0_2_CFG:
 		ret_val = igb_read_invm_word_i210(hw, (u8)offset, data);
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			*data = NVM_LED_0_2_CFG_DEFAULT_I211;
-			ret_val = E1000_SUCCESS;
+			ret_val = 0;
 		}
 		break;
 	case NVM_ID_LED_SETTINGS:
 		ret_val = igb_read_invm_word_i210(hw, (u8)offset, data);
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			*data = ID_LED_RESERVED_FFFF;
-			ret_val = E1000_SUCCESS;
+			ret_val = 0;
 		}
+		break;
 	case NVM_SUB_DEV_ID:
 		*data = hw->subsystem_device_id;
 		break;
@@ -488,14 +486,14 @@ s32 igb_read_invm_version(struct e1000_hw *hw,
 		/* Check if we have first version location used */
 		if ((i == 1) && ((*record & E1000_INVM_VER_FIELD_ONE) == 0)) {
 			version = 0;
-			status = E1000_SUCCESS;
+			status = 0;
 			break;
 		}
 		/* Check if we have second version location used */
 		else if ((i == 1) &&
 			 ((*record & E1000_INVM_VER_FIELD_TWO) == 0)) {
 			version = (*record & E1000_INVM_VER_FIELD_ONE) >> 3;
-			status = E1000_SUCCESS;
+			status = 0;
 			break;
 		}
 		/* Check if we have odd version location
@@ -506,7 +504,7 @@ s32 igb_read_invm_version(struct e1000_hw *hw,
 			 (i != 1))) {
 			version = (*next_record & E1000_INVM_VER_FIELD_TWO)
 				  >> 13;
-			status = E1000_SUCCESS;
+			status = 0;
 			break;
 		}
 		/* Check if we have even version location
@@ -515,12 +513,12 @@ s32 igb_read_invm_version(struct e1000_hw *hw,
 		else if (((*record & E1000_INVM_VER_FIELD_TWO) == 0) &&
 			 ((*record & 0x3) == 0)) {
 			version = (*record & E1000_INVM_VER_FIELD_ONE) >> 3;
-			status = E1000_SUCCESS;
+			status = 0;
 			break;
 		}
 	}
 
-	if (status == E1000_SUCCESS) {
+	if (!status) {
 		invm_ver->invm_major = (version & E1000_INVM_MAJOR_MASK)
 					>> E1000_INVM_MAJOR_SHIFT;
 		invm_ver->invm_minor = version & E1000_INVM_MINOR_MASK;
@@ -533,7 +531,7 @@ s32 igb_read_invm_version(struct e1000_hw *hw,
 		/* Check if we have image type in first location used */
 		if ((i == 1) && ((*record & E1000_INVM_IMGTYPE_FIELD) == 0)) {
 			invm_ver->invm_img_type = 0;
-			status = E1000_SUCCESS;
+			status = 0;
 			break;
 		}
 		/* Check if we have image type in first location used */
@@ -542,7 +540,7 @@ s32 igb_read_invm_version(struct e1000_hw *hw,
 			 ((((*record & 0x3) != 0) && (i != 1)))) {
 			invm_ver->invm_img_type =
 				(*next_record & E1000_INVM_IMGTYPE_FIELD) >> 23;
-			status = E1000_SUCCESS;
+			status = 0;
 			break;
 		}
 	}
@@ -558,10 +556,10 @@ s32 igb_read_invm_version(struct e1000_hw *hw,
  **/
 static s32 igb_validate_nvm_checksum_i210(struct e1000_hw *hw)
 {
-	s32 status = E1000_SUCCESS;
+	s32 status = 0;
 	s32 (*read_op_ptr)(struct e1000_hw *, u16, u16, u16 *);
 
-	if (hw->nvm.ops.acquire(hw) == E1000_SUCCESS) {
+	if (!(hw->nvm.ops.acquire(hw))) {
 
 		/* Replace the read function with semaphore grabbing with
 		 * the one that skips this for a while.
@@ -593,7 +591,7 @@ static s32 igb_validate_nvm_checksum_i210(struct e1000_hw *hw)
  **/
 static s32 igb_update_nvm_checksum_i210(struct e1000_hw *hw)
 {
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 	u16 checksum = 0;
 	u16 i, nvm_data;
 
@@ -602,12 +600,12 @@ static s32 igb_update_nvm_checksum_i210(struct e1000_hw *hw)
 	 * EEPROM read fails
 	 */
 	ret_val = igb_read_nvm_eerd(hw, 0, 1, &nvm_data);
-	if (ret_val != E1000_SUCCESS) {
+	if (ret_val) {
 		hw_dbg("EEPROM read failed\n");
 		goto out;
 	}
 
-	if (hw->nvm.ops.acquire(hw) == E1000_SUCCESS) {
+	if (!(hw->nvm.ops.acquire(hw))) {
 		/* Do not use hw->nvm.ops.write, hw->nvm.ops.read
 		 * because we do not want to take the synchronization
 		 * semaphores twice here.
@@ -625,7 +623,7 @@ static s32 igb_update_nvm_checksum_i210(struct e1000_hw *hw)
 		checksum = (u16) NVM_SUM - checksum;
 		ret_val = igb_write_nvm_srwr(hw, NVM_CHECKSUM_REG, 1,
 						&checksum);
-		if (ret_val != E1000_SUCCESS) {
+		if (ret_val) {
 			hw->nvm.ops.release(hw);
 			hw_dbg("NVM Write Error while updating checksum.\n");
 			goto out;
@@ -654,7 +652,7 @@ static s32 igb_pool_flash_update_done_i210(struct e1000_hw *hw)
 	for (i = 0; i < E1000_FLUDONE_ATTEMPTS; i++) {
 		reg = rd32(E1000_EECD);
 		if (reg & E1000_EECD_FLUDONE_I210) {
-			ret_val = E1000_SUCCESS;
+			ret_val = 0;
 			break;
 		}
 		udelay(5);
@@ -687,7 +685,7 @@ bool igb_get_flash_presence_i210(struct e1000_hw *hw)
  **/
 static s32 igb_update_flash_i210(struct e1000_hw *hw)
 {
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 	u32 flup;
 
 	ret_val = igb_pool_flash_update_done_i210(hw);
@@ -700,7 +698,7 @@ static s32 igb_update_flash_i210(struct e1000_hw *hw)
 	wr32(E1000_EECD, flup);
 
 	ret_val = igb_pool_flash_update_done_i210(hw);
-	if (ret_val == E1000_SUCCESS)
+	if (ret_val)
 		hw_dbg("Flash update complete\n");
 	else
 		hw_dbg("Flash update time out\n");
@@ -753,7 +751,7 @@ out:
 static s32 __igb_access_xmdio_reg(struct e1000_hw *hw, u16 address,
 				  u8 dev_addr, u16 *data, bool read)
 {
-	s32 ret_val = E1000_SUCCESS;
+	s32 ret_val = 0;
 
 	ret_val = hw->phy.ops.write_reg(hw, E1000_MMDAC, dev_addr);
 	if (ret_val)
diff --git a/drivers/net/ethernet/intel/igb/e1000_i210.h b/drivers/net/ethernet/intel/igb/e1000_i210.h
index 907fe99a9813..9f34976687ba 100644
--- a/drivers/net/ethernet/intel/igb/e1000_i210.h
+++ b/drivers/net/ethernet/intel/igb/e1000_i210.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_I210_H_
 #define _E1000_I210_H_
diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c
index 1e0c404db81a..2a88595f956c 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.c
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #include <linux/if_ether.h>
 #include <linux/delay.h>
@@ -442,7 +439,7 @@ static u32 igb_hash_mc_addr(struct e1000_hw *hw, u8 *mc_addr)
  *  The caller must have a packed mc_addr_list of multicast addresses.
  **/
 void igb_update_mc_addr_list(struct e1000_hw *hw,
-                             u8 *mc_addr_list, u32 mc_addr_count)
+			     u8 *mc_addr_list, u32 mc_addr_count)
 {
 	u32 hash_value, hash_bit, hash_reg;
 	int i;
@@ -866,8 +863,7 @@ s32 igb_config_fc_after_link_up(struct e1000_hw *hw)
 			goto out;
 
 		if (!(mii_status_reg & MII_SR_AUTONEG_COMPLETE)) {
-			hw_dbg("Copper PHY and Auto Neg "
-				 "has not completed.\n");
+			hw_dbg("Copper PHY and Auto Neg has not completed.\n");
 			goto out;
 		}
 
@@ -1265,7 +1261,7 @@ s32 igb_get_auto_rd_done(struct e1000_hw *hw)
 	while (i < AUTO_READ_DONE_TIMEOUT) {
 		if (rd32(E1000_EECD) & E1000_EECD_AUTO_RD)
 			break;
-		msleep(1);
+		usleep_range(1000, 2000);
 		i++;
 	}
 
@@ -1298,7 +1294,7 @@ static s32 igb_valid_led_default(struct e1000_hw *hw, u16 *data)
 	}
 
 	if (*data == ID_LED_RESERVED_0000 || *data == ID_LED_RESERVED_FFFF) {
-		switch(hw->phy.media_type) {
+		switch (hw->phy.media_type) {
 		case e1000_media_type_internal_serdes:
 			*data = ID_LED_DEFAULT_82575_SERDES;
 			break;
diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.h b/drivers/net/ethernet/intel/igb/e1000_mac.h
index 99299ba8ee3a..ea24961b0d70 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.h
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_MAC_H_
 #define _E1000_MAC_H_
diff --git a/drivers/net/ethernet/intel/igb/e1000_mbx.c b/drivers/net/ethernet/intel/igb/e1000_mbx.c
index d5b121771c31..162cc49345d0 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mbx.c
+++ b/drivers/net/ethernet/intel/igb/e1000_mbx.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #include "e1000_mbx.h"
 
diff --git a/drivers/net/ethernet/intel/igb/e1000_mbx.h b/drivers/net/ethernet/intel/igb/e1000_mbx.h
index f52f5515e5a8..d20af6b2f581 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mbx.h
+++ b/drivers/net/ethernet/intel/igb/e1000_mbx.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_MBX_H_
 #define _E1000_MBX_H_
diff --git a/drivers/net/ethernet/intel/igb/e1000_nvm.c b/drivers/net/ethernet/intel/igb/e1000_nvm.c
index 9abf82919c65..e8280d0d7f02 100644
--- a/drivers/net/ethernet/intel/igb/e1000_nvm.c
+++ b/drivers/net/ethernet/intel/igb/e1000_nvm.c
@@ -1,28 +1,24 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #include <linux/if_ether.h>
 #include <linux/delay.h>
@@ -480,6 +476,7 @@ s32 igb_write_nvm_spi(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
 		/* Loop to allow for up to whole page write of eeprom */
 		while (widx < words) {
 			u16 word_out = data[widx];
+
 			word_out = (word_out >> 8) | (word_out << 8);
 			igb_shift_out_eec_bits(hw, word_out, 16);
 			widx++;
@@ -801,5 +798,4 @@ etrack_id:
 		fw_vers->etrack_id = (eeprom_verh << NVM_ETRACK_SHIFT)
 			| eeprom_verl;
 	}
-	return;
 }
diff --git a/drivers/net/ethernet/intel/igb/e1000_nvm.h b/drivers/net/ethernet/intel/igb/e1000_nvm.h
index 5b101170b17e..febc9cdb7391 100644
--- a/drivers/net/ethernet/intel/igb/e1000_nvm.h
+++ b/drivers/net/ethernet/intel/igb/e1000_nvm.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_NVM_H_
 #define _E1000_NVM_H_
@@ -32,7 +29,7 @@ void igb_release_nvm(struct e1000_hw *hw);
 s32  igb_read_mac_addr(struct e1000_hw *hw);
 s32  igb_read_part_num(struct e1000_hw *hw, u32 *part_num);
 s32  igb_read_part_string(struct e1000_hw *hw, u8 *part_num,
-                          u32 part_num_size);
+			  u32 part_num_size);
 s32  igb_read_nvm_eerd(struct e1000_hw *hw, u16 offset, u16 words, u16 *data);
 s32  igb_read_nvm_spi(struct e1000_hw *hw, u16 offset, u16 words, u16 *data);
 s32  igb_write_nvm_spi(struct e1000_hw *hw, u16 offset, u16 words, u16 *data);
diff --git a/drivers/net/ethernet/intel/igb/e1000_phy.c b/drivers/net/ethernet/intel/igb/e1000_phy.c
index 4009bbab7407..c1bb64d8366f 100644
--- a/drivers/net/ethernet/intel/igb/e1000_phy.c
+++ b/drivers/net/ethernet/intel/igb/e1000_phy.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #include <linux/if_ether.h>
 #include <linux/delay.h>
@@ -924,8 +921,7 @@ static s32 igb_copper_link_autoneg(struct e1000_hw *hw)
 	if (phy->autoneg_wait_to_complete) {
 		ret_val = igb_wait_autoneg(hw);
 		if (ret_val) {
-			hw_dbg("Error while waiting for "
-			       "autoneg to complete\n");
+			hw_dbg("Error while waiting for autoneg to complete\n");
 			goto out;
 		}
 	}
@@ -2208,16 +2204,10 @@ s32 igb_phy_init_script_igp3(struct e1000_hw *hw)
 void igb_power_up_phy_copper(struct e1000_hw *hw)
 {
 	u16 mii_reg = 0;
-	u16 power_reg = 0;
 
 	/* The PHY will retain its settings across a power down/up cycle */
 	hw->phy.ops.read_reg(hw, PHY_CONTROL, &mii_reg);
 	mii_reg &= ~MII_CR_POWER_DOWN;
-	if (hw->phy.type == e1000_phy_i210) {
-		hw->phy.ops.read_reg(hw, GS40G_COPPER_SPEC, &power_reg);
-		power_reg &= ~GS40G_CS_POWER_DOWN;
-		hw->phy.ops.write_reg(hw, GS40G_COPPER_SPEC, power_reg);
-	}
 	hw->phy.ops.write_reg(hw, PHY_CONTROL, mii_reg);
 }
 
@@ -2231,20 +2221,12 @@ void igb_power_up_phy_copper(struct e1000_hw *hw)
 void igb_power_down_phy_copper(struct e1000_hw *hw)
 {
 	u16 mii_reg = 0;
-	u16 power_reg = 0;
 
 	/* The PHY will retain its settings across a power down/up cycle */
 	hw->phy.ops.read_reg(hw, PHY_CONTROL, &mii_reg);
 	mii_reg |= MII_CR_POWER_DOWN;
-
-	/* i210 Phy requires an additional bit for power up/down */
-	if (hw->phy.type == e1000_phy_i210) {
-		hw->phy.ops.read_reg(hw, GS40G_COPPER_SPEC, &power_reg);
-		power_reg |= GS40G_CS_POWER_DOWN;
-		hw->phy.ops.write_reg(hw, GS40G_COPPER_SPEC, power_reg);
-	}
 	hw->phy.ops.write_reg(hw, PHY_CONTROL, mii_reg);
-	msleep(1);
+	usleep_range(1000, 2000);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/igb/e1000_phy.h b/drivers/net/ethernet/intel/igb/e1000_phy.h
index 4c2c36c46a73..7af4ffab0285 100644
--- a/drivers/net/ethernet/intel/igb/e1000_phy.h
+++ b/drivers/net/ethernet/intel/igb/e1000_phy.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_PHY_H_
 #define _E1000_PHY_H_
@@ -154,7 +151,6 @@ s32  igb_check_polarity_m88(struct e1000_hw *hw);
 #define GS40G_MAC_LB			0x4140
 #define GS40G_MAC_SPEED_1G		0X0006
 #define GS40G_COPPER_SPEC		0x0010
-#define GS40G_CS_POWER_DOWN		0x0002
 #define GS40G_LINE_LB			0x4000
 
 /* SFP modules ID memory locations */
diff --git a/drivers/net/ethernet/intel/igb/e1000_regs.h b/drivers/net/ethernet/intel/igb/e1000_regs.h
index bdb246e848e1..1cc4b1a7e597 100644
--- a/drivers/net/ethernet/intel/igb/e1000_regs.h
+++ b/drivers/net/ethernet/intel/igb/e1000_regs.h
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #ifndef _E1000_REGS_H_
 #define _E1000_REGS_H_
@@ -195,6 +192,10 @@
 				    : (0x0E038 + ((_n) * 0x40)))
 #define E1000_TDWBAH(_n)  ((_n) < 4 ? (0x0383C + ((_n) * 0x100)) \
 				    : (0x0E03C + ((_n) * 0x40)))
+
+#define E1000_RXPBS	0x02404  /* Rx Packet Buffer Size - RW */
+#define E1000_TXPBS	0x03404  /* Tx Packet Buffer Size - RW */
+
 #define E1000_TDFH     0x03410  /* TX Data FIFO Head - RW */
 #define E1000_TDFT     0x03418  /* TX Data FIFO Tail - RW */
 #define E1000_TDFHS    0x03420  /* TX Data FIFO Head Saved - RW */
@@ -301,9 +302,9 @@
 #define E1000_RA2      0x054E0  /* 2nd half of Rx address array - RW Array */
 #define E1000_PSRTYPE(_i)       (0x05480 + ((_i) * 4))
 #define E1000_RAL(_i)  (((_i) <= 15) ? (0x05400 + ((_i) * 8)) : \
-                                       (0x054E0 + ((_i - 16) * 8)))
+					(0x054E0 + ((_i - 16) * 8)))
 #define E1000_RAH(_i)  (((_i) <= 15) ? (0x05404 + ((_i) * 8)) : \
-                                       (0x054E4 + ((_i - 16) * 8)))
+					(0x054E4 + ((_i - 16) * 8)))
 #define E1000_IP4AT_REG(_i)     (0x05840 + ((_i) * 8))
 #define E1000_IP6AT_REG(_i)     (0x05880 + ((_i) * 4))
 #define E1000_WUPM_REG(_i)      (0x05A00 + ((_i) * 4))
@@ -358,8 +359,7 @@
 #define E1000_VMBMEM(_n)       (0x00800 + (64 * (_n)))
 #define E1000_VMOLR(_n)        (0x05AD0 + (4 * (_n)))
 #define E1000_DVMOLR(_n)       (0x0C038 + (64 * (_n)))
-#define E1000_VLVF(_n)         (0x05D00 + (4 * (_n))) /* VLAN Virtual Machine
-                                                       * Filter - RW */
+#define E1000_VLVF(_n)         (0x05D00 + (4 * (_n))) /* VLAN VM Filter */
 #define E1000_VMVIR(_n)        (0x03700 + (4 * (_n)))
 
 struct e1000_hw;
diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index 27130065d92a..06102d1f7c03 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -1,29 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
-
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 /* Linux PRO/1000 Ethernet Driver main header file */
 
@@ -198,6 +194,7 @@ struct igb_tx_buffer {
 	unsigned int bytecount;
 	u16 gso_segs;
 	__be16 protocol;
+
 	DEFINE_DMA_UNMAP_ADDR(dma);
 	DEFINE_DMA_UNMAP_LEN(len);
 	u32 tx_flags;
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index e5570acbeea8..c737d1f40838 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 /* ethtool support for igb */
 
@@ -144,6 +141,7 @@ static int igb_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
 	struct e1000_dev_spec_82575 *dev_spec = &hw->dev_spec._82575;
 	struct e1000_sfp_flags *eth_flags = &dev_spec->eth_flags;
 	u32 status;
+	u32 speed;
 
 	status = rd32(E1000_STATUS);
 	if (hw->phy.media_type == e1000_media_type_copper) {
@@ -218,13 +216,13 @@ static int igb_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
 	if (status & E1000_STATUS_LU) {
 		if ((status & E1000_STATUS_2P5_SKU) &&
 		    !(status & E1000_STATUS_2P5_SKU_OVER)) {
-			ecmd->speed = SPEED_2500;
+			speed = SPEED_2500;
 		} else if (status & E1000_STATUS_SPEED_1000) {
-			ecmd->speed = SPEED_1000;
+			speed = SPEED_1000;
 		} else if (status & E1000_STATUS_SPEED_100) {
-			ecmd->speed = SPEED_100;
+			speed = SPEED_100;
 		} else {
-			ecmd->speed = SPEED_10;
+			speed = SPEED_10;
 		}
 		if ((status & E1000_STATUS_FD) ||
 		    hw->phy.media_type != e1000_media_type_copper)
@@ -232,9 +230,10 @@ static int igb_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ecmd->speed = -1;
-		ecmd->duplex = -1;
+		speed = SPEED_UNKNOWN;
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
+	ethtool_cmd_speed_set(ecmd, speed);
 	if ((hw->phy.media_type == e1000_media_type_fiber) ||
 	    hw->mac.autoneg)
 		ecmd->autoneg = AUTONEG_ENABLE;
@@ -286,7 +285,7 @@ static int igb_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
 	}
 
 	while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
-		msleep(1);
+		usleep_range(1000, 2000);
 
 	if (ecmd->autoneg == AUTONEG_ENABLE) {
 		hw->mac.autoneg = 1;
@@ -399,7 +398,7 @@ static int igb_set_pauseparam(struct net_device *netdev,
 	adapter->fc_autoneg = pause->autoneg;
 
 	while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
-		msleep(1);
+		usleep_range(1000, 2000);
 
 	if (adapter->fc_autoneg == AUTONEG_ENABLE) {
 		hw->fc.requested_mode = e1000_fc_default;
@@ -886,7 +885,7 @@ static int igb_set_ringparam(struct net_device *netdev,
 	}
 
 	while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
-		msleep(1);
+		usleep_range(1000, 2000);
 
 	if (!netif_running(adapter->netdev)) {
 		for (i = 0; i < adapter->num_tx_queues; i++)
@@ -1060,8 +1059,8 @@ static struct igb_reg_test reg_test_i350[] = {
 	{ E1000_TDT(0),	   0x100, 4,  PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
 	{ E1000_TDT(4),	   0x40,  4,  PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
 	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
-	{ E1000_RCTL, 	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
-	{ E1000_RCTL, 	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
+	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
+	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
 	{ E1000_TCTL,	   0x100, 1,  SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
 	{ E1000_RA,	   0, 16, TABLE64_TEST_LO,
 						0xFFFFFFFF, 0xFFFFFFFF },
@@ -1103,8 +1102,8 @@ static struct igb_reg_test reg_test_82580[] = {
 	{ E1000_TDT(0),	   0x100, 4,  PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
 	{ E1000_TDT(4),	   0x40,  4,  PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
 	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
-	{ E1000_RCTL, 	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
-	{ E1000_RCTL, 	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
+	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
+	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
 	{ E1000_TCTL,	   0x100, 1,  SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
 	{ E1000_RA,	   0, 16, TABLE64_TEST_LO,
 						0xFFFFFFFF, 0xFFFFFFFF },
@@ -1132,8 +1131,10 @@ static struct igb_reg_test reg_test_82576[] = {
 	{ E1000_RDBAH(4),  0x40, 12, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
 	{ E1000_RDLEN(4),  0x40, 12, PATTERN_TEST, 0x000FFFF0, 0x000FFFFF },
 	/* Enable all RX queues before testing. */
-	{ E1000_RXDCTL(0), 0x100, 4,  WRITE_NO_TEST, 0, E1000_RXDCTL_QUEUE_ENABLE },
-	{ E1000_RXDCTL(4), 0x40, 12,  WRITE_NO_TEST, 0, E1000_RXDCTL_QUEUE_ENABLE },
+	{ E1000_RXDCTL(0), 0x100, 4, WRITE_NO_TEST, 0,
+	  E1000_RXDCTL_QUEUE_ENABLE },
+	{ E1000_RXDCTL(4), 0x40, 12, WRITE_NO_TEST, 0,
+	  E1000_RXDCTL_QUEUE_ENABLE },
 	/* RDH is read-only for 82576, only test RDT. */
 	{ E1000_RDT(0),	   0x100, 4,  PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
 	{ E1000_RDT(4),	   0x40, 12,  PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
@@ -1149,14 +1150,14 @@ static struct igb_reg_test reg_test_82576[] = {
 	{ E1000_TDBAH(4),  0x40, 12,  PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
 	{ E1000_TDLEN(4),  0x40, 12,  PATTERN_TEST, 0x000FFFF0, 0x000FFFFF },
 	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
-	{ E1000_RCTL, 	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
-	{ E1000_RCTL, 	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
+	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
+	{ E1000_RCTL,	   0x100, 1,  SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
 	{ E1000_TCTL,	   0x100, 1,  SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
 	{ E1000_RA,	   0, 16, TABLE64_TEST_LO, 0xFFFFFFFF, 0xFFFFFFFF },
 	{ E1000_RA,	   0, 16, TABLE64_TEST_HI, 0x83FFFFFF, 0xFFFFFFFF },
 	{ E1000_RA2,	   0, 8, TABLE64_TEST_LO, 0xFFFFFFFF, 0xFFFFFFFF },
 	{ E1000_RA2,	   0, 8, TABLE64_TEST_HI, 0x83FFFFFF, 0xFFFFFFFF },
-	{ E1000_MTA,	   0, 128,TABLE32_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
+	{ E1000_MTA,	   0, 128, TABLE32_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
 	{ 0, 0, 0, 0 }
 };
 
@@ -1170,7 +1171,8 @@ static struct igb_reg_test reg_test_82575[] = {
 	{ E1000_RDBAH(0),  0x100, 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
 	{ E1000_RDLEN(0),  0x100, 4, PATTERN_TEST, 0x000FFF80, 0x000FFFFF },
 	/* Enable all four RX queues before testing. */
-	{ E1000_RXDCTL(0), 0x100, 4, WRITE_NO_TEST, 0, E1000_RXDCTL_QUEUE_ENABLE },
+	{ E1000_RXDCTL(0), 0x100, 4, WRITE_NO_TEST, 0,
+	  E1000_RXDCTL_QUEUE_ENABLE },
 	/* RDH is read-only for 82575, only test RDT. */
 	{ E1000_RDT(0),    0x100, 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
 	{ E1000_RXDCTL(0), 0x100, 4, WRITE_NO_TEST, 0, 0 },
@@ -1196,8 +1198,8 @@ static bool reg_pattern_test(struct igb_adapter *adapter, u64 *data,
 {
 	struct e1000_hw *hw = &adapter->hw;
 	u32 pat, val;
-	static const u32 _test[] =
-		{0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF};
+	static const u32 _test[] = {
+		0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF};
 	for (pat = 0; pat < ARRAY_SIZE(_test); pat++) {
 		wr32(reg, (_test[pat] & write));
 		val = rd32(reg) & mask;
@@ -1206,11 +1208,11 @@ static bool reg_pattern_test(struct igb_adapter *adapter, u64 *data,
 				"pattern test reg %04X failed: got 0x%08X expected 0x%08X\n",
 				reg, val, (_test[pat] & write & mask));
 			*data = reg;
-			return 1;
+			return true;
 		}
 	}
 
-	return 0;
+	return false;
 }
 
 static bool reg_set_and_check(struct igb_adapter *adapter, u64 *data,
@@ -1218,17 +1220,18 @@ static bool reg_set_and_check(struct igb_adapter *adapter, u64 *data,
 {
 	struct e1000_hw *hw = &adapter->hw;
 	u32 val;
+
 	wr32(reg, write & mask);
 	val = rd32(reg);
 	if ((write & mask) != (val & mask)) {
 		dev_err(&adapter->pdev->dev,
-			"set/check reg %04X test failed: got 0x%08X expected 0x%08X\n", reg,
-			(val & mask), (write & mask));
+			"set/check reg %04X test failed: got 0x%08X expected 0x%08X\n",
+			reg, (val & mask), (write & mask));
 		*data = reg;
-		return 1;
+		return true;
 	}
 
-	return 0;
+	return false;
 }
 
 #define REG_PATTERN_TEST(reg, mask, write) \
@@ -1387,14 +1390,14 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
 	/* Hook up test interrupt handler just for this test */
 	if (adapter->flags & IGB_FLAG_HAS_MSIX) {
 		if (request_irq(adapter->msix_entries[0].vector,
-		                igb_test_intr, 0, netdev->name, adapter)) {
+				igb_test_intr, 0, netdev->name, adapter)) {
 			*data = 1;
 			return -1;
 		}
 	} else if (adapter->flags & IGB_FLAG_HAS_MSI) {
 		shared_int = false;
 		if (request_irq(irq,
-		                igb_test_intr, 0, netdev->name, adapter)) {
+				igb_test_intr, 0, netdev->name, adapter)) {
 			*data = 1;
 			return -1;
 		}
@@ -1412,7 +1415,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
 	/* Disable all the interrupts */
 	wr32(E1000_IMC, ~0);
 	wrfl();
-	msleep(10);
+	usleep_range(10000, 11000);
 
 	/* Define all writable bits for ICS */
 	switch (hw->mac.type) {
@@ -1459,7 +1462,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
 			wr32(E1000_IMC, mask);
 			wr32(E1000_ICS, mask);
 			wrfl();
-			msleep(10);
+			usleep_range(10000, 11000);
 
 			if (adapter->test_icr & mask) {
 				*data = 3;
@@ -1481,7 +1484,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
 		wr32(E1000_IMS, mask);
 		wr32(E1000_ICS, mask);
 		wrfl();
-		msleep(10);
+		usleep_range(10000, 11000);
 
 		if (!(adapter->test_icr & mask)) {
 			*data = 4;
@@ -1503,7 +1506,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
 			wr32(E1000_IMC, ~mask);
 			wr32(E1000_ICS, ~mask);
 			wrfl();
-			msleep(10);
+			usleep_range(10000, 11000);
 
 			if (adapter->test_icr & mask) {
 				*data = 5;
@@ -1515,7 +1518,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
 	/* Disable all the interrupts */
 	wr32(E1000_IMC, ~0);
 	wrfl();
-	msleep(10);
+	usleep_range(10000, 11000);
 
 	/* Unhook test interrupt handler */
 	if (adapter->flags & IGB_FLAG_HAS_MSIX)
@@ -1664,8 +1667,8 @@ static int igb_setup_loopback_test(struct igb_adapter *adapter)
 		(hw->device_id == E1000_DEV_ID_DH89XXCC_SERDES) ||
 		(hw->device_id == E1000_DEV_ID_DH89XXCC_BACKPLANE) ||
 		(hw->device_id == E1000_DEV_ID_DH89XXCC_SFP) ||
-		(hw->device_id == E1000_DEV_ID_I354_SGMII)) {
-
+		(hw->device_id == E1000_DEV_ID_I354_SGMII) ||
+		(hw->device_id == E1000_DEV_ID_I354_BACKPLANE_2_5GBPS)) {
 			/* Enable DH89xxCC MPHY for near end loopback */
 			reg = rd32(E1000_MPHY_ADDR_CTL);
 			reg = (reg & E1000_MPHY_ADDR_CTL_OFFSET_MASK) |
@@ -1949,6 +1952,7 @@ static int igb_link_test(struct igb_adapter *adapter, u64 *data)
 	*data = 0;
 	if (hw->phy.media_type == e1000_media_type_internal_serdes) {
 		int i = 0;
+
 		hw->mac.serdes_has_link = false;
 
 		/* On some blade server designs, link establishment
@@ -2413,9 +2417,11 @@ static int igb_get_rss_hash_opts(struct igb_adapter *adapter,
 	switch (cmd->flow_type) {
 	case TCP_V4_FLOW:
 		cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* Fall through */
 	case UDP_V4_FLOW:
 		if (adapter->flags & IGB_FLAG_RSS_FIELD_IPV4_UDP)
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* Fall through */
 	case SCTP_V4_FLOW:
 	case AH_ESP_V4_FLOW:
 	case AH_V4_FLOW:
@@ -2425,9 +2431,11 @@ static int igb_get_rss_hash_opts(struct igb_adapter *adapter,
 		break;
 	case TCP_V6_FLOW:
 		cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* Fall through */
 	case UDP_V6_FLOW:
 		if (adapter->flags & IGB_FLAG_RSS_FIELD_IPV6_UDP)
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* Fall through */
 	case SCTP_V6_FLOW:
 	case AH_ESP_V6_FLOW:
 	case AH_V6_FLOW:
@@ -2730,7 +2738,7 @@ static int igb_get_module_info(struct net_device *netdev,
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
-	u32 status = E1000_SUCCESS;
+	u32 status = 0;
 	u16 sff8472_rev, addr_mode;
 	bool page_swap = false;
 
@@ -2740,12 +2748,12 @@ static int igb_get_module_info(struct net_device *netdev,
 
 	/* Check whether we support SFF-8472 or not */
 	status = igb_read_phy_reg_i2c(hw, IGB_SFF_8472_COMP, &sff8472_rev);
-	if (status != E1000_SUCCESS)
+	if (status)
 		return -EIO;
 
 	/* addressing mode is not supported */
 	status = igb_read_phy_reg_i2c(hw, IGB_SFF_8472_SWAP, &addr_mode);
-	if (status != E1000_SUCCESS)
+	if (status)
 		return -EIO;
 
 	/* addressing mode is not supported */
@@ -2772,7 +2780,7 @@ static int igb_get_module_eeprom(struct net_device *netdev,
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
-	u32 status = E1000_SUCCESS;
+	u32 status = 0;
 	u16 *dataword;
 	u16 first_word, last_word;
 	int i = 0;
@@ -2791,7 +2799,7 @@ static int igb_get_module_eeprom(struct net_device *netdev,
 	/* Read EEPROM block, SFF-8079/SFF-8472, word at a time */
 	for (i = 0; i < last_word - first_word + 1; i++) {
 		status = igb_read_phy_reg_i2c(hw, first_word + i, &dataword[i]);
-		if (status != E1000_SUCCESS) {
+		if (status) {
 			/* Error occurred while reading module */
 			kfree(dataword);
 			return -EIO;
@@ -2824,7 +2832,7 @@ static u32 igb_get_rxfh_indir_size(struct net_device *netdev)
 	return IGB_RETA_SIZE;
 }
 
-static int igb_get_rxfh_indir(struct net_device *netdev, u32 *indir)
+static int igb_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key)
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	int i;
@@ -2870,7 +2878,8 @@ void igb_write_rss_indir_tbl(struct igb_adapter *adapter)
 	}
 }
 
-static int igb_set_rxfh_indir(struct net_device *netdev, const u32 *indir)
+static int igb_set_rxfh(struct net_device *netdev, const u32 *indir,
+			const u8 *key)
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
@@ -3019,8 +3028,8 @@ static const struct ethtool_ops igb_ethtool_ops = {
 	.get_module_info	= igb_get_module_info,
 	.get_module_eeprom	= igb_get_module_eeprom,
 	.get_rxfh_indir_size	= igb_get_rxfh_indir_size,
-	.get_rxfh_indir		= igb_get_rxfh_indir,
-	.set_rxfh_indir		= igb_set_rxfh_indir,
+	.get_rxfh		= igb_get_rxfh,
+	.set_rxfh		= igb_set_rxfh,
 	.get_channels		= igb_get_channels,
 	.set_channels		= igb_set_channels,
 	.begin			= igb_ethtool_begin,
@@ -3029,5 +3038,5 @@ static const struct ethtool_ops igb_ethtool_ops = {
 
 void igb_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &igb_ethtool_ops);
+	netdev->ethtool_ops = &igb_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/igb/igb_hwmon.c b/drivers/net/ethernet/intel/igb/igb_hwmon.c
index 8333f67acf96..44b6a68f1af7 100644
--- a/drivers/net/ethernet/intel/igb/igb_hwmon.c
+++ b/drivers/net/ethernet/intel/igb/igb_hwmon.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #include "igb.h"
 #include "e1000_82575.h"
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 16430a8440fa..f145adbb55ac 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -1,28 +1,25 @@
-/*******************************************************************************
-
-  Intel(R) Gigabit Ethernet Linux driver
-  Copyright(c) 2007-2014 Intel Corporation.
-
-  This program is free software; you can redistribute it and/or modify it
-  under the terms and conditions of the GNU General Public License,
-  version 2, as published by the Free Software Foundation.
-
-  This program is distributed in the hope it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, see <http://www.gnu.org/licenses/>.
-
-  The full GNU General Public License is included in this distribution in
-  the file called "COPYING".
-
-  Contact Information:
-  e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
-  Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
-
-*******************************************************************************/
+/* Intel(R) Gigabit Ethernet Linux driver
+ * Copyright(c) 2007-2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
@@ -75,7 +72,7 @@ static const struct e1000_info *igb_info_tbl[] = {
 	[board_82575] = &e1000_82575_info,
 };
 
-static DEFINE_PCI_DEVICE_TABLE(igb_pci_tbl) = {
+static const struct pci_device_id igb_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_BACKPLANE_1GBPS) },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_SGMII) },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_BACKPLANE_2_5GBPS) },
@@ -117,7 +114,6 @@ static DEFINE_PCI_DEVICE_TABLE(igb_pci_tbl) = {
 
 MODULE_DEVICE_TABLE(pci, igb_pci_tbl);
 
-void igb_reset(struct igb_adapter *);
 static int igb_setup_all_tx_resources(struct igb_adapter *);
 static int igb_setup_all_rx_resources(struct igb_adapter *);
 static void igb_free_all_tx_resources(struct igb_adapter *);
@@ -141,7 +137,7 @@ static void igb_watchdog(unsigned long);
 static void igb_watchdog_task(struct work_struct *);
 static netdev_tx_t igb_xmit_frame(struct sk_buff *skb, struct net_device *);
 static struct rtnl_link_stats64 *igb_get_stats64(struct net_device *dev,
-						 struct rtnl_link_stats64 *stats);
+					  struct rtnl_link_stats64 *stats);
 static int igb_change_mtu(struct net_device *, int);
 static int igb_set_mac(struct net_device *, void *);
 static void igb_set_uta(struct igb_adapter *adapter);
@@ -159,7 +155,8 @@ static bool igb_clean_rx_irq(struct igb_q_vector *, int);
 static int igb_ioctl(struct net_device *, struct ifreq *, int cmd);
 static void igb_tx_timeout(struct net_device *);
 static void igb_reset_task(struct work_struct *);
-static void igb_vlan_mode(struct net_device *netdev, netdev_features_t features);
+static void igb_vlan_mode(struct net_device *netdev,
+			  netdev_features_t features);
 static int igb_vlan_rx_add_vid(struct net_device *, __be16, u16);
 static int igb_vlan_rx_kill_vid(struct net_device *, __be16, u16);
 static void igb_restore_vlan(struct igb_adapter *);
@@ -172,7 +169,7 @@ static void igb_restore_vf_multicasts(struct igb_adapter *adapter);
 static int igb_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac);
 static int igb_ndo_set_vf_vlan(struct net_device *netdev,
 			       int vf, u16 vlan, u8 qos);
-static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate);
+static int igb_ndo_set_vf_bw(struct net_device *, int, int, int);
 static int igb_ndo_set_vf_spoofchk(struct net_device *netdev, int vf,
 				   bool setting);
 static int igb_ndo_get_vf_config(struct net_device *netdev, int vf,
@@ -215,10 +212,9 @@ static struct notifier_block dca_notifier = {
 static void igb_netpoll(struct net_device *);
 #endif
 #ifdef CONFIG_PCI_IOV
-static unsigned int max_vfs = 0;
+static unsigned int max_vfs;
 module_param(max_vfs, uint, 0);
-MODULE_PARM_DESC(max_vfs, "Maximum number of virtual functions to allocate "
-                 "per physical function");
+MODULE_PARM_DESC(max_vfs, "Maximum number of virtual functions to allocate per physical function");
 #endif /* CONFIG_PCI_IOV */
 
 static pci_ers_result_t igb_io_error_detected(struct pci_dev *,
@@ -384,8 +380,7 @@ static void igb_dump(struct igb_adapter *adapter)
 	/* Print netdevice Info */
 	if (netdev) {
 		dev_info(&adapter->pdev->dev, "Net device Info\n");
-		pr_info("Device Name     state            trans_start      "
-			"last_rx\n");
+		pr_info("Device Name     state            trans_start      last_rx\n");
 		pr_info("%-15s %016lX %016lX %016lX\n", netdev->name,
 			netdev->state, netdev->trans_start, netdev->last_rx);
 	}
@@ -438,9 +433,7 @@ static void igb_dump(struct igb_adapter *adapter)
 		pr_info("------------------------------------\n");
 		pr_info("TX QUEUE INDEX = %d\n", tx_ring->queue_index);
 		pr_info("------------------------------------\n");
-		pr_info("T [desc]     [address 63:0  ] [PlPOCIStDDM Ln] "
-			"[bi->dma       ] leng  ntw timestamp        "
-			"bi->skb\n");
+		pr_info("T [desc]     [address 63:0  ] [PlPOCIStDDM Ln] [bi->dma       ] leng  ntw timestamp        bi->skb\n");
 
 		for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) {
 			const char *next_desc;
@@ -458,9 +451,8 @@ static void igb_dump(struct igb_adapter *adapter)
 			else
 				next_desc = "";
 
-			pr_info("T [0x%03X]    %016llX %016llX %016llX"
-				" %04X  %p %016llX %p%s\n", i,
-				le64_to_cpu(u0->a),
+			pr_info("T [0x%03X]    %016llX %016llX %016llX %04X  %p %016llX %p%s\n",
+				i, le64_to_cpu(u0->a),
 				le64_to_cpu(u0->b),
 				(u64)dma_unmap_addr(buffer_info, dma),
 				dma_unmap_len(buffer_info, len),
@@ -519,10 +511,8 @@ rx_ring_summary:
 		pr_info("------------------------------------\n");
 		pr_info("RX QUEUE INDEX = %d\n", rx_ring->queue_index);
 		pr_info("------------------------------------\n");
-		pr_info("R  [desc]      [ PktBuf     A0] [  HeadBuf   DD] "
-			"[bi->dma       ] [bi->skb] <-- Adv Rx Read format\n");
-		pr_info("RWB[desc]      [PcsmIpSHl PtRs] [vl er S cks ln] -----"
-			"----------- [bi->skb] <-- Adv Rx Write-Back format\n");
+		pr_info("R  [desc]      [ PktBuf     A0] [  HeadBuf   DD] [bi->dma       ] [bi->skb] <-- Adv Rx Read format\n");
+		pr_info("RWB[desc]      [PcsmIpSHl PtRs] [vl er S cks ln] ---------------- [bi->skb] <-- Adv Rx Write-Back format\n");
 
 		for (i = 0; i < rx_ring->count; i++) {
 			const char *next_desc;
@@ -584,7 +574,7 @@ static int igb_get_i2c_data(void *data)
 	struct e1000_hw *hw = &adapter->hw;
 	s32 i2cctl = rd32(E1000_I2CPARAMS);
 
-	return ((i2cctl & E1000_I2C_DATA_IN) != 0);
+	return !!(i2cctl & E1000_I2C_DATA_IN);
 }
 
 /**
@@ -648,7 +638,7 @@ static int igb_get_i2c_clk(void *data)
 	struct e1000_hw *hw = &adapter->hw;
 	s32 i2cctl = rd32(E1000_I2CPARAMS);
 
-	return ((i2cctl & E1000_I2C_CLK_IN) != 0);
+	return !!(i2cctl & E1000_I2C_CLK_IN);
 }
 
 static const struct i2c_algo_bit_data igb_i2c_algo = {
@@ -681,9 +671,9 @@ struct net_device *igb_get_hw_dev(struct e1000_hw *hw)
 static int __init igb_init_module(void)
 {
 	int ret;
+
 	pr_info("%s - version %s\n",
 	       igb_driver_string, igb_driver_version);
-
 	pr_info("%s\n", igb_copyright);
 
 #ifdef CONFIG_IGB_DCA
@@ -736,12 +726,14 @@ static void igb_cache_ring_register(struct igb_adapter *adapter)
 				adapter->rx_ring[i]->reg_idx = rbase_offset +
 							       Q_IDX_82576(i);
 		}
+		/* Fall through */
 	case e1000_82575:
 	case e1000_82580:
 	case e1000_i350:
 	case e1000_i354:
 	case e1000_i210:
 	case e1000_i211:
+		/* Fall through */
 	default:
 		for (; i < adapter->num_rx_queues; i++)
 			adapter->rx_ring[i]->reg_idx = rbase_offset + i;
@@ -1292,8 +1284,7 @@ static int igb_alloc_q_vector(struct igb_adapter *adapter,
 		if (adapter->hw.mac.type >= e1000_82576)
 			set_bit(IGB_RING_FLAG_RX_SCTP_CSUM, &ring->flags);
 
-		/*
-		 * On i350, i354, i210, and i211, loopback VLAN packets
+		/* On i350, i354, i210, and i211, loopback VLAN packets
 		 * have the tag byte-swapped.
 		 */
 		if (adapter->hw.mac.type >= e1000_i350)
@@ -1345,6 +1336,7 @@ static int igb_alloc_q_vectors(struct igb_adapter *adapter)
 	for (; v_idx < q_vectors; v_idx++) {
 		int rqpv = DIV_ROUND_UP(rxr_remaining, q_vectors - v_idx);
 		int tqpv = DIV_ROUND_UP(txr_remaining, q_vectors - v_idx);
+
 		err = igb_alloc_q_vector(adapter, q_vectors, v_idx,
 					 tqpv, txr_idx, rqpv, rxr_idx);
 
@@ -1484,6 +1476,7 @@ static void igb_irq_disable(struct igb_adapter *adapter)
 	 */
 	if (adapter->flags & IGB_FLAG_HAS_MSIX) {
 		u32 regval = rd32(E1000_EIAM);
+
 		wr32(E1000_EIAM, regval & ~adapter->eims_enable_mask);
 		wr32(E1000_EIMC, adapter->eims_enable_mask);
 		regval = rd32(E1000_EIAC);
@@ -1495,6 +1488,7 @@ static void igb_irq_disable(struct igb_adapter *adapter)
 	wrfl();
 	if (adapter->flags & IGB_FLAG_HAS_MSIX) {
 		int i;
+
 		for (i = 0; i < adapter->num_q_vectors; i++)
 			synchronize_irq(adapter->msix_entries[i].vector);
 	} else {
@@ -1513,6 +1507,7 @@ static void igb_irq_enable(struct igb_adapter *adapter)
 	if (adapter->flags & IGB_FLAG_HAS_MSIX) {
 		u32 ims = E1000_IMS_LSC | E1000_IMS_DOUTSYNC | E1000_IMS_DRSTA;
 		u32 regval = rd32(E1000_EIAC);
+
 		wr32(E1000_EIAC, regval | adapter->eims_enable_mask);
 		regval = rd32(E1000_EIAM);
 		wr32(E1000_EIAM, regval | adapter->eims_enable_mask);
@@ -1745,6 +1740,7 @@ int igb_up(struct igb_adapter *adapter)
 	/* notify VFs that reset has been completed */
 	if (adapter->vfs_allocated_count) {
 		u32 reg_data = rd32(E1000_CTRL_EXT);
+
 		reg_data |= E1000_CTRL_EXT_PFRSTD;
 		wr32(E1000_CTRL_EXT, reg_data);
 	}
@@ -1787,7 +1783,7 @@ void igb_down(struct igb_adapter *adapter)
 	wr32(E1000_TCTL, tctl);
 	/* flush both disables and wait for them to finish */
 	wrfl();
-	msleep(10);
+	usleep_range(10000, 11000);
 
 	igb_irq_disable(adapter);
 
@@ -1827,7 +1823,7 @@ void igb_reinit_locked(struct igb_adapter *adapter)
 {
 	WARN_ON(in_interrupt());
 	while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
-		msleep(1);
+		usleep_range(1000, 2000);
 	igb_down(adapter);
 	igb_up(adapter);
 	clear_bit(__IGB_RESETTING, &adapter->state);
@@ -1960,6 +1956,7 @@ void igb_reset(struct igb_adapter *adapter)
 	/* disable receive for all VFs and wait one second */
 	if (adapter->vfs_allocated_count) {
 		int i;
+
 		for (i = 0 ; i < adapter->vfs_allocated_count; i++)
 			adapter->vf_data[i].flags &= IGB_VF_FLAG_PF_SET_MAC;
 
@@ -2087,7 +2084,7 @@ static const struct net_device_ops igb_netdev_ops = {
 	.ndo_vlan_rx_kill_vid	= igb_vlan_rx_kill_vid,
 	.ndo_set_vf_mac		= igb_ndo_set_vf_mac,
 	.ndo_set_vf_vlan	= igb_ndo_set_vf_vlan,
-	.ndo_set_vf_tx_rate	= igb_ndo_set_vf_bw,
+	.ndo_set_vf_rate	= igb_ndo_set_vf_bw,
 	.ndo_set_vf_spoofchk	= igb_ndo_set_vf_spoofchk,
 	.ndo_get_vf_config	= igb_ndo_get_vf_config,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -2142,7 +2139,6 @@ void igb_set_fw_version(struct igb_adapter *adapter)
 		}
 		break;
 	}
-	return;
 }
 
 /**
@@ -2203,11 +2199,11 @@ static void igb_init_mas(struct igb_adapter *adapter)
  **/
 static s32 igb_init_i2c(struct igb_adapter *adapter)
 {
-	s32 status = E1000_SUCCESS;
+	s32 status = 0;
 
 	/* I2C interface supported on i350 devices */
 	if (adapter->hw.mac.type != e1000_i350)
-		return E1000_SUCCESS;
+		return 0;
 
 	/* Initialize the i2c bus which is controlled by the registers.
 	 * This bus will use the i2c_algo_bit structue that implements
@@ -2437,6 +2433,12 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* get firmware version for ethtool -i */
 	igb_set_fw_version(adapter);
 
+	/* configure RXPBSIZE and TXPBSIZE */
+	if (hw->mac.type == e1000_i210) {
+		wr32(E1000_RXPBS, I210_RXPBSIZE_DEFAULT);
+		wr32(E1000_TXPBS, I210_TXPBSIZE_DEFAULT);
+	}
+
 	setup_timer(&adapter->watchdog_timer, igb_watchdog,
 		    (unsigned long) adapter);
 	setup_timer(&adapter->phy_info_timer, igb_update_phy_info,
@@ -2529,7 +2531,8 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	/* let the f/w know that the h/w is now under the control of the
-	 * driver. */
+	 * driver.
+	 */
 	igb_get_hw_control(adapter);
 
 	strcpy(netdev->name, "eth%d");
@@ -3077,6 +3080,7 @@ static int __igb_open(struct net_device *netdev, bool resuming)
 	/* notify VFs that reset has been completed */
 	if (adapter->vfs_allocated_count) {
 		u32 reg_data = rd32(E1000_CTRL_EXT);
+
 		reg_data |= E1000_CTRL_EXT_PFRSTD;
 		wr32(E1000_CTRL_EXT, reg_data);
 	}
@@ -3248,7 +3252,7 @@ void igb_setup_tctl(struct igb_adapter *adapter)
  *  Configure a transmit ring after a reset.
  **/
 void igb_configure_tx_ring(struct igb_adapter *adapter,
-                           struct igb_ring *ring)
+			   struct igb_ring *ring)
 {
 	struct e1000_hw *hw = &adapter->hw;
 	u32 txdctl = 0;
@@ -3389,7 +3393,8 @@ static void igb_setup_mrqc(struct igb_adapter *adapter)
 
 	if (adapter->rss_indir_tbl_init != num_rx_queues) {
 		for (j = 0; j < IGB_RETA_SIZE; j++)
-			adapter->rss_indir_tbl[j] = (j * num_rx_queues) / IGB_RETA_SIZE;
+			adapter->rss_indir_tbl[j] =
+			(j * num_rx_queues) / IGB_RETA_SIZE;
 		adapter->rss_indir_tbl_init = num_rx_queues;
 	}
 	igb_write_rss_indir_tbl(adapter);
@@ -3430,6 +3435,7 @@ static void igb_setup_mrqc(struct igb_adapter *adapter)
 		if (hw->mac.type > e1000_82575) {
 			/* Set the default pool for the PF's first queue */
 			u32 vtctl = rd32(E1000_VT_CTL);
+
 			vtctl &= ~(E1000_VT_CTL_DEFAULT_POOL_MASK |
 				   E1000_VT_CTL_DISABLE_DEF_POOL);
 			vtctl |= adapter->vfs_allocated_count <<
@@ -3511,7 +3517,7 @@ void igb_setup_rctl(struct igb_adapter *adapter)
 }
 
 static inline int igb_set_vf_rlpml(struct igb_adapter *adapter, int size,
-                                   int vfn)
+				   int vfn)
 {
 	struct e1000_hw *hw = &adapter->hw;
 	u32 vmolr;
@@ -4058,7 +4064,8 @@ static void igb_check_wvbr(struct igb_adapter *adapter)
 	switch (hw->mac.type) {
 	case e1000_82576:
 	case e1000_i350:
-		if (!(wvbr = rd32(E1000_WVBR)))
+		wvbr = rd32(E1000_WVBR);
+		if (!wvbr)
 			return;
 		break;
 	default:
@@ -4077,7 +4084,7 @@ static void igb_spoof_check(struct igb_adapter *adapter)
 	if (!adapter->wvbr)
 		return;
 
-	for(j = 0; j < adapter->vfs_allocated_count; j++) {
+	for (j = 0; j < adapter->vfs_allocated_count; j++) {
 		if (adapter->wvbr & (1 << j) ||
 		    adapter->wvbr & (1 << (j + IGB_STAGGERED_QUEUE_OFFSET))) {
 			dev_warn(&adapter->pdev->dev,
@@ -4209,14 +4216,15 @@ static void igb_watchdog_task(struct work_struct *work)
 
 		if (!netif_carrier_ok(netdev)) {
 			u32 ctrl;
+
 			hw->mac.ops.get_speed_and_duplex(hw,
 							 &adapter->link_speed,
 							 &adapter->link_duplex);
 
 			ctrl = rd32(E1000_CTRL);
 			/* Links status message must follow this format */
-			printk(KERN_INFO "igb: %s NIC Link is Up %d Mbps %s "
-			       "Duplex, Flow Control: %s\n",
+			netdev_info(netdev,
+			       "igb: %s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
 			       netdev->name,
 			       adapter->link_speed,
 			       adapter->link_duplex == FULL_DUPLEX ?
@@ -4242,11 +4250,8 @@ static void igb_watchdog_task(struct work_struct *work)
 
 			/* check for thermal sensor event */
 			if (igb_thermal_sensor_event(hw,
-			    E1000_THSTAT_LINK_THROTTLE)) {
-				netdev_info(netdev, "The network adapter link "
-					    "speed was downshifted because it "
-					    "overheated\n");
-			}
+			    E1000_THSTAT_LINK_THROTTLE))
+				netdev_info(netdev, "The network adapter link speed was downshifted because it overheated\n");
 
 			/* adjust timeout factor according to speed/duplex */
 			adapter->tx_timeout_factor = 1;
@@ -4277,12 +4282,11 @@ static void igb_watchdog_task(struct work_struct *work)
 			/* check for thermal sensor event */
 			if (igb_thermal_sensor_event(hw,
 			    E1000_THSTAT_PWR_DOWN)) {
-				netdev_err(netdev, "The network adapter was "
-					   "stopped because it overheated\n");
+				netdev_err(netdev, "The network adapter was stopped because it overheated\n");
 			}
 
 			/* Links status message must follow this format */
-			printk(KERN_INFO "igb: %s NIC Link is Down\n",
+			netdev_info(netdev, "igb: %s NIC Link is Down\n",
 			       netdev->name);
 			netif_carrier_off(netdev);
 
@@ -4344,6 +4348,7 @@ static void igb_watchdog_task(struct work_struct *work)
 	/* Cause software interrupt to ensure Rx ring is cleaned */
 	if (adapter->flags & IGB_FLAG_HAS_MSIX) {
 		u32 eics = 0;
+
 		for (i = 0; i < adapter->num_q_vectors; i++)
 			eics |= adapter->q_vector[i]->eims_value;
 		wr32(E1000_EICS, eics);
@@ -4483,13 +4488,12 @@ static void igb_update_itr(struct igb_q_vector *q_vector,
 	case low_latency:  /* 50 usec aka 20000 ints/s */
 		if (bytes > 10000) {
 			/* this if handles the TSO accounting */
-			if (bytes/packets > 8000) {
+			if (bytes/packets > 8000)
 				itrval = bulk_latency;
-			} else if ((packets < 10) || ((bytes/packets) > 1200)) {
+			else if ((packets < 10) || ((bytes/packets) > 1200))
 				itrval = bulk_latency;
-			} else if ((packets > 35)) {
+			else if ((packets > 35))
 				itrval = lowest_latency;
-			}
 		} else if (bytes/packets > 2000) {
 			itrval = bulk_latency;
 		} else if (packets <= 2 && bytes < 512) {
@@ -4675,6 +4679,7 @@ static void igb_tx_csum(struct igb_ring *tx_ring, struct igb_tx_buffer *first)
 			return;
 	} else {
 		u8 l4_hdr = 0;
+
 		switch (first->protocol) {
 		case htons(ETH_P_IP):
 			vlan_macip_lens |= skb_network_header_len(skb);
@@ -4962,6 +4967,7 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb,
 	 */
 	if (NETDEV_FRAG_PAGE_MAX_SIZE > IGB_MAX_DATA_PER_TXD) {
 		unsigned short f;
+
 		for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
 			count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
 	} else {
@@ -5140,7 +5146,7 @@ static int igb_change_mtu(struct net_device *netdev, int new_mtu)
 		max_frame = ETH_FRAME_LEN + ETH_FCS_LEN;
 
 	while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
-		msleep(1);
+		usleep_range(1000, 2000);
 
 	/* igb_down has a dependency on max_frame_size */
 	adapter->max_frame_size = max_frame;
@@ -5621,6 +5627,7 @@ static int igb_set_vf_promisc(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
 			vmolr |= E1000_VMOLR_MPME;
 		} else if (vf_data->num_vf_mc_hashes) {
 			int j;
+
 			vmolr |= E1000_VMOLR_ROMPE;
 			for (j = 0; j < vf_data->num_vf_mc_hashes; j++)
 				igb_mta_set(hw, vf_data->vf_mc_hashes[j]);
@@ -5672,6 +5679,7 @@ static void igb_restore_vf_multicasts(struct igb_adapter *adapter)
 
 	for (i = 0; i < adapter->vfs_allocated_count; i++) {
 		u32 vmolr = rd32(E1000_VMOLR(i));
+
 		vmolr &= ~(E1000_VMOLR_ROMPE | E1000_VMOLR_MPME);
 
 		vf_data = &adapter->vf_data[i];
@@ -5770,6 +5778,7 @@ static s32 igb_vlvf_set(struct igb_adapter *adapter, u32 vid, bool add, u32 vf)
 
 			if (!adapter->vf_data[vf].vlans_enabled) {
 				u32 size;
+
 				reg = rd32(E1000_VMOLR(vf));
 				size = reg & E1000_VMOLR_RLPML_MASK;
 				size += 4;
@@ -5798,6 +5807,7 @@ static s32 igb_vlvf_set(struct igb_adapter *adapter, u32 vid, bool add, u32 vf)
 			adapter->vf_data[vf].vlans_enabled--;
 			if (!adapter->vf_data[vf].vlans_enabled) {
 				u32 size;
+
 				reg = rd32(E1000_VMOLR(vf));
 				size = reg & E1000_VMOLR_RLPML_MASK;
 				size -= 4;
@@ -5902,8 +5912,8 @@ static int igb_set_vf_vlan(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
 	 */
 	if (!add && (adapter->netdev->flags & IFF_PROMISC)) {
 		u32 vlvf, bits;
-
 		int regndx = igb_find_vlvf_entry(adapter, vid);
+
 		if (regndx < 0)
 			goto out;
 		/* See if any other pools are set for this VLAN filter
@@ -6494,7 +6504,7 @@ static void igb_reuse_rx_page(struct igb_ring *rx_ring,
 	rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
 
 	/* transfer page from old buffer to new buffer */
-	memcpy(new_buff, old_buff, sizeof(struct igb_rx_buffer));
+	*new_buff = *old_buff;
 
 	/* sync the buffer for use by the device */
 	dma_sync_single_range_for_device(rx_ring->dev, old_buff->dma,
@@ -6963,6 +6973,7 @@ static void igb_process_skb_fields(struct igb_ring *rx_ring,
 	if ((dev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
 	    igb_test_staterr(rx_desc, E1000_RXD_STAT_VP)) {
 		u16 vid;
+
 		if (igb_test_staterr(rx_desc, E1000_RXDEXT_STATERR_LB) &&
 		    test_bit(IGB_RING_FLAG_RX_LB_VLAN_BSWAP, &rx_ring->flags))
 			vid = be16_to_cpu(rx_desc->wb.upper.vlan);
@@ -7051,7 +7062,7 @@ static bool igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget)
 	if (cleaned_count)
 		igb_alloc_rx_buffers(rx_ring, cleaned_count);
 
-	return (total_packets < budget);
+	return total_packets < budget;
 }
 
 static bool igb_alloc_mapped_page(struct igb_ring *rx_ring,
@@ -7172,7 +7183,7 @@ static int igb_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 		break;
 	case SIOCGMIIREG:
 		if (igb_read_phy_reg(&adapter->hw, data->reg_num & 0x1F,
-		                     &data->val_out))
+				     &data->val_out))
 			return -EIO;
 		break;
 	case SIOCSMIIREG:
@@ -7873,7 +7884,8 @@ static void igb_check_vf_rate_limit(struct igb_adapter *adapter)
 	}
 }
 
-static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate)
+static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf,
+			     int min_tx_rate, int max_tx_rate)
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
@@ -7882,15 +7894,19 @@ static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate)
 	if (hw->mac.type != e1000_82576)
 		return -EOPNOTSUPP;
 
+	if (min_tx_rate)
+		return -EINVAL;
+
 	actual_link_speed = igb_link_mbps(adapter->link_speed);
 	if ((vf >= adapter->vfs_allocated_count) ||
 	    (!(rd32(E1000_STATUS) & E1000_STATUS_LU)) ||
-	    (tx_rate < 0) || (tx_rate > actual_link_speed))
+	    (max_tx_rate < 0) ||
+	    (max_tx_rate > actual_link_speed))
 		return -EINVAL;
 
 	adapter->vf_rate_link_speed = actual_link_speed;
-	adapter->vf_data[vf].tx_rate = (u16)tx_rate;
-	igb_set_vf_rate_limit(hw, vf, tx_rate, actual_link_speed);
+	adapter->vf_data[vf].tx_rate = (u16)max_tx_rate;
+	igb_set_vf_rate_limit(hw, vf, max_tx_rate, actual_link_speed);
 
 	return 0;
 }
@@ -7919,7 +7935,7 @@ static int igb_ndo_set_vf_spoofchk(struct net_device *netdev, int vf,
 	wr32(reg_offset, reg_val);
 
 	adapter->vf_data[vf].spoofchk_enabled = setting;
-	return E1000_SUCCESS;
+	return 0;
 }
 
 static int igb_ndo_get_vf_config(struct net_device *netdev,
@@ -7930,7 +7946,8 @@ static int igb_ndo_get_vf_config(struct net_device *netdev,
 		return -EINVAL;
 	ivi->vf = vf;
 	memcpy(&ivi->mac, adapter->vf_data[vf].vf_mac_addresses, ETH_ALEN);
-	ivi->tx_rate = adapter->vf_data[vf].tx_rate;
+	ivi->max_tx_rate = adapter->vf_data[vf].tx_rate;
+	ivi->min_tx_rate = 0;
 	ivi->vlan = adapter->vf_data[vf].pf_vlan;
 	ivi->qos = adapter->vf_data[vf].pf_qos;
 	ivi->spoofchk = adapter->vf_data[vf].spoofchk_enabled;
@@ -7955,11 +7972,13 @@ static void igb_vmm_control(struct igb_adapter *adapter)
 		reg = rd32(E1000_DTXCTL);
 		reg |= E1000_DTXCTL_VLAN_ADDED;
 		wr32(E1000_DTXCTL, reg);
+		/* Fall through */
 	case e1000_82580:
 		/* enable replication vlan tag stripping */
 		reg = rd32(E1000_RPLOLR);
 		reg |= E1000_RPLOLR_STRVLAN;
 		wr32(E1000_RPLOLR, reg);
+		/* Fall through */
 	case e1000_i350:
 		/* none of the above registers are supported by i350 */
 		break;
@@ -8049,6 +8068,7 @@ static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
 		} /* endif adapter->dmac is not disabled */
 	} else if (hw->mac.type == e1000_82580) {
 		u32 reg = rd32(E1000_PCIEMISC);
+
 		wr32(E1000_PCIEMISC, reg & ~E1000_PCIEMISC_LX_DECISION);
 		wr32(E1000_DMACR, 0);
 	}
@@ -8077,8 +8097,7 @@ s32 igb_read_i2c_byte(struct e1000_hw *hw, u8 byte_offset,
 
 	swfw_mask = E1000_SWFW_PHY0_SM;
 
-	if (hw->mac.ops.acquire_swfw_sync(hw, swfw_mask)
-	    != E1000_SUCCESS)
+	if (hw->mac.ops.acquire_swfw_sync(hw, swfw_mask))
 		return E1000_ERR_SWFW_SYNC;
 
 	status = i2c_smbus_read_byte_data(this_client, byte_offset);
@@ -8088,7 +8107,7 @@ s32 igb_read_i2c_byte(struct e1000_hw *hw, u8 byte_offset,
 		return E1000_ERR_I2C;
 	else {
 		*data = status;
-		return E1000_SUCCESS;
+		return 0;
 	}
 }
 
@@ -8113,7 +8132,7 @@ s32 igb_write_i2c_byte(struct e1000_hw *hw, u8 byte_offset,
 	if (!this_client)
 		return E1000_ERR_I2C;
 
-	if (hw->mac.ops.acquire_swfw_sync(hw, swfw_mask) != E1000_SUCCESS)
+	if (hw->mac.ops.acquire_swfw_sync(hw, swfw_mask))
 		return E1000_ERR_SWFW_SYNC;
 	status = i2c_smbus_write_byte_data(this_client, byte_offset, data);
 	hw->mac.ops.release_swfw_sync(hw, swfw_mask);
@@ -8121,7 +8140,7 @@ s32 igb_write_i2c_byte(struct e1000_hw *hw, u8 byte_offset,
 	if (status)
 		return E1000_ERR_I2C;
 	else
-		return E1000_SUCCESS;
+		return 0;
 
 }
 
diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index ab25e49365f7..794c139f0cc0 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -360,8 +360,8 @@ static int igb_ptp_settime_i210(struct ptp_clock_info *ptp,
 	return 0;
 }
 
-static int igb_ptp_enable(struct ptp_clock_info *ptp,
-			  struct ptp_clock_request *rq, int on)
+static int igb_ptp_feature_enable(struct ptp_clock_info *ptp,
+				  struct ptp_clock_request *rq, int on)
 {
 	return -EOPNOTSUPP;
 }
@@ -559,10 +559,11 @@ int igb_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr)
 	return copy_to_user(ifr->ifr_data, config, sizeof(*config)) ?
 		-EFAULT : 0;
 }
+
 /**
- * igb_ptp_set_ts_config - control hardware time stamping
- * @netdev:
- * @ifreq:
+ * igb_ptp_set_timestamp_mode - setup hardware for timestamping
+ * @adapter: networking device structure
+ * @config: hwtstamp configuration
  *
  * Outgoing time stamping can be enabled and disabled. Play nice and
  * disable it when requested, although it shouldn't case any overhead
@@ -575,12 +576,11 @@ int igb_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr)
  * type has to be specified. Matching the kind of event packet is
  * not supported, with the exception of "all V2 events regardless of
  * level 2 or 4".
- **/
-int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr)
+ */
+static int igb_ptp_set_timestamp_mode(struct igb_adapter *adapter,
+				      struct hwtstamp_config *config)
 {
-	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
-	struct hwtstamp_config *config = &adapter->tstamp_config;
 	u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED;
 	u32 tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED;
 	u32 tsync_rx_cfg = 0;
@@ -588,9 +588,6 @@ int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr)
 	bool is_l2 = false;
 	u32 regval;
 
-	if (copy_from_user(config, ifr->ifr_data, sizeof(*config)))
-		return -EFAULT;
-
 	/* reserved for future extensions */
 	if (config->flags)
 		return -EINVAL;
@@ -725,7 +722,33 @@ int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr)
 	regval = rd32(E1000_RXSTMPL);
 	regval = rd32(E1000_RXSTMPH);
 
-	return copy_to_user(ifr->ifr_data, config, sizeof(*config)) ?
+	return 0;
+}
+
+/**
+ * igb_ptp_set_ts_config - set hardware time stamping config
+ * @netdev:
+ * @ifreq:
+ *
+ **/
+int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr)
+{
+	struct igb_adapter *adapter = netdev_priv(netdev);
+	struct hwtstamp_config config;
+	int err;
+
+	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	err = igb_ptp_set_timestamp_mode(adapter, &config);
+	if (err)
+		return err;
+
+	/* save these settings for future reference */
+	memcpy(&adapter->tstamp_config, &config,
+	       sizeof(adapter->tstamp_config));
+
+	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
 		-EFAULT : 0;
 }
 
@@ -745,7 +768,7 @@ void igb_ptp_init(struct igb_adapter *adapter)
 		adapter->ptp_caps.adjtime = igb_ptp_adjtime_82576;
 		adapter->ptp_caps.gettime = igb_ptp_gettime_82576;
 		adapter->ptp_caps.settime = igb_ptp_settime_82576;
-		adapter->ptp_caps.enable = igb_ptp_enable;
+		adapter->ptp_caps.enable = igb_ptp_feature_enable;
 		adapter->cc.read = igb_ptp_read_82576;
 		adapter->cc.mask = CLOCKSOURCE_MASK(64);
 		adapter->cc.mult = 1;
@@ -765,7 +788,7 @@ void igb_ptp_init(struct igb_adapter *adapter)
 		adapter->ptp_caps.adjtime = igb_ptp_adjtime_82576;
 		adapter->ptp_caps.gettime = igb_ptp_gettime_82576;
 		adapter->ptp_caps.settime = igb_ptp_settime_82576;
-		adapter->ptp_caps.enable = igb_ptp_enable;
+		adapter->ptp_caps.enable = igb_ptp_feature_enable;
 		adapter->cc.read = igb_ptp_read_82580;
 		adapter->cc.mask = CLOCKSOURCE_MASK(IGB_NBITS_82580);
 		adapter->cc.mult = 1;
@@ -784,7 +807,7 @@ void igb_ptp_init(struct igb_adapter *adapter)
 		adapter->ptp_caps.adjtime = igb_ptp_adjtime_i210;
 		adapter->ptp_caps.gettime = igb_ptp_gettime_i210;
 		adapter->ptp_caps.settime = igb_ptp_settime_i210;
-		adapter->ptp_caps.enable = igb_ptp_enable;
+		adapter->ptp_caps.enable = igb_ptp_feature_enable;
 		/* Enable the timer functions by clearing bit 31. */
 		wr32(E1000_TSAUXC, 0x0);
 		break;
@@ -820,6 +843,9 @@ void igb_ptp_init(struct igb_adapter *adapter)
 		wr32(E1000_IMS, E1000_IMS_TS);
 	}
 
+	adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
+	adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF;
+
 	adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps,
 						&adapter->pdev->dev);
 	if (IS_ERR(adapter->ptp_clock)) {
@@ -884,7 +910,7 @@ void igb_ptp_reset(struct igb_adapter *adapter)
 		return;
 
 	/* reset the tstamp_config */
-	memset(&adapter->tstamp_config, 0, sizeof(adapter->tstamp_config));
+	igb_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config);
 
 	switch (adapter->hw.mac.type) {
 	case e1000_82576:
diff --git a/drivers/net/ethernet/intel/igbvf/ethtool.c b/drivers/net/ethernet/intel/igbvf/ethtool.c
index 90eef07943f4..2178f87e9f61 100644
--- a/drivers/net/ethernet/intel/igbvf/ethtool.c
+++ b/drivers/net/ethernet/intel/igbvf/ethtool.c
@@ -101,8 +101,8 @@ static int igbvf_get_settings(struct net_device *netdev,
 		else
 			ecmd->duplex = DUPLEX_HALF;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = AUTONEG_DISABLE;
@@ -119,7 +119,6 @@ static int igbvf_set_settings(struct net_device *netdev,
 static void igbvf_get_pauseparam(struct net_device *netdev,
                                  struct ethtool_pauseparam *pause)
 {
-	return;
 }
 
 static int igbvf_set_pauseparam(struct net_device *netdev,
@@ -476,5 +475,5 @@ static const struct ethtool_ops igbvf_ethtool_ops = {
 
 void igbvf_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &igbvf_ethtool_ops);
+	netdev->ethtool_ops = &igbvf_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c b/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c
index dbb7dd2f8e36..b311e9e710d2 100644
--- a/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c
@@ -107,8 +107,8 @@ ixgb_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
 		ethtool_cmd_speed_set(ecmd, SPEED_10000);
 		ecmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	ecmd->autoneg = AUTONEG_DISABLE;
@@ -656,5 +656,5 @@ static const struct ethtool_ops ixgb_ethtool_ops = {
 
 void ixgb_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &ixgb_ethtool_ops);
+	netdev->ethtool_ops = &ixgb_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index c6c4ca7d68e6..ac9f2148cdc5 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -155,7 +155,6 @@ struct vf_data_storage {
 struct vf_macvlans {
 	struct list_head l;
 	int vf;
-	int rar_entry;
 	bool free;
 	bool is_macvlan;
 	u8 vf_macvlan[ETH_ALEN];
@@ -363,7 +362,7 @@ struct ixgbe_ring_container {
 	for (pos = (head).ring; pos != NULL; pos = pos->next)
 
 #define MAX_RX_PACKET_BUFFERS ((adapter->flags & IXGBE_FLAG_DCB_ENABLED) \
-                              ? 8 : 1)
+			      ? 8 : 1)
 #define MAX_TX_PACKET_BUFFERS MAX_RX_PACKET_BUFFERS
 
 /* MAX_Q_VECTORS of these are allocated,
@@ -613,6 +612,15 @@ static inline void ixgbe_write_tail(struct ixgbe_ring *ring, u32 value)
 #define MAX_MSIX_VECTORS_82598 18
 #define MAX_Q_VECTORS_82598 16
 
+struct ixgbe_mac_addr {
+	u8 addr[ETH_ALEN];
+	u16 queue;
+	u16 state; /* bitmask */
+};
+#define IXGBE_MAC_STATE_DEFAULT		0x1
+#define IXGBE_MAC_STATE_MODIFIED	0x2
+#define IXGBE_MAC_STATE_IN_USE		0x4
+
 #define MAX_Q_VECTORS MAX_Q_VECTORS_82599
 #define MAX_MSIX_COUNT MAX_MSIX_VECTORS_82599
 
@@ -785,6 +793,7 @@ struct ixgbe_adapter {
 
 	u32 timer_event_accumulator;
 	u32 vferr_refcount;
+	struct ixgbe_mac_addr *mac_table;
 	struct kobject *info_kobj;
 #ifdef CONFIG_IXGBE_HWMON
 	struct hwmon_buff *ixgbe_hwmon_buff;
@@ -863,6 +872,13 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter);
 int ixgbe_init_interrupt_scheme(struct ixgbe_adapter *adapter);
 int ixgbe_wol_supported(struct ixgbe_adapter *adapter, u16 device_id,
 			       u16 subdevice_id);
+#ifdef CONFIG_PCI_IOV
+void ixgbe_full_sync_mac_table(struct ixgbe_adapter *adapter);
+#endif
+int ixgbe_add_mac_filter(struct ixgbe_adapter *adapter,
+			 u8 *addr, u16 queue);
+int ixgbe_del_mac_filter(struct ixgbe_adapter *adapter,
+			 u8 *addr, u16 queue);
 void ixgbe_clear_interrupt_scheme(struct ixgbe_adapter *adapter);
 netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *, struct ixgbe_adapter *,
 				  struct ixgbe_ring *);
@@ -941,6 +957,7 @@ static inline struct netdev_queue *txring_txq(const struct ixgbe_ring *ring)
 }
 
 void ixgbe_ptp_init(struct ixgbe_adapter *adapter);
+void ixgbe_ptp_suspend(struct ixgbe_adapter *adapter);
 void ixgbe_ptp_stop(struct ixgbe_adapter *adapter);
 void ixgbe_ptp_overflow_check(struct ixgbe_adapter *adapter);
 void ixgbe_ptp_rx_hang(struct ixgbe_adapter *adapter);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_82598.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_82598.c
index 4c78ea8946c1..15609331ec17 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_82598.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_82598.c
@@ -41,10 +41,10 @@
 #define IXGBE_82598_RX_PB_SIZE	 512
 
 static s32 ixgbe_setup_copper_link_82598(struct ixgbe_hw *hw,
-                                         ixgbe_link_speed speed,
-                                         bool autoneg_wait_to_complete);
+					 ixgbe_link_speed speed,
+					 bool autoneg_wait_to_complete);
 static s32 ixgbe_read_i2c_eeprom_82598(struct ixgbe_hw *hw, u8 byte_offset,
-                                       u8 *eeprom_data);
+				       u8 *eeprom_data);
 
 /**
  *  ixgbe_set_pcie_completion_timeout - set pci-e completion timeout
@@ -140,7 +140,7 @@ static s32 ixgbe_init_phy_ops_82598(struct ixgbe_hw *hw)
 		phy->ops.setup_link = &ixgbe_setup_phy_link_tnx;
 		phy->ops.check_link = &ixgbe_check_phy_link_tnx;
 		phy->ops.get_firmware_version =
-		             &ixgbe_get_phy_firmware_version_tnx;
+			     &ixgbe_get_phy_firmware_version_tnx;
 		break;
 	case ixgbe_phy_nl:
 		phy->ops.reset = &ixgbe_reset_phy_nl;
@@ -156,8 +156,8 @@ static s32 ixgbe_init_phy_ops_82598(struct ixgbe_hw *hw)
 
 		/* Check to see if SFP+ module is supported */
 		ret_val = ixgbe_get_sfp_init_sequence_offsets(hw,
-		                                            &list_offset,
-		                                            &data_offset);
+							    &list_offset,
+							    &data_offset);
 		if (ret_val != 0) {
 			ret_val = IXGBE_ERR_SFP_NOT_SUPPORTED;
 			goto out;
@@ -219,8 +219,8 @@ static s32 ixgbe_start_hw_82598(struct ixgbe_hw *hw)
  *  Determines the link capabilities by reading the AUTOC register.
  **/
 static s32 ixgbe_get_link_capabilities_82598(struct ixgbe_hw *hw,
-                                             ixgbe_link_speed *speed,
-                                             bool *autoneg)
+					     ixgbe_link_speed *speed,
+					     bool *autoneg)
 {
 	s32 status = 0;
 	u32 autoc = 0;
@@ -337,19 +337,25 @@ static s32 ixgbe_fc_enable_82598(struct ixgbe_hw *hw)
 	int i;
 	bool link_up;
 
-	/*
-	 * Validate the water mark configuration for packet buffer 0.  Zero
-	 * water marks indicate that the packet buffer was not configured
-	 * and the watermarks for packet buffer 0 should always be configured.
-	 */
-	if (!hw->fc.low_water ||
-	    !hw->fc.high_water[0] ||
-	    !hw->fc.pause_time) {
-		hw_dbg(hw, "Invalid water mark configuration\n");
+	/* Validate the water mark configuration */
+	if (!hw->fc.pause_time) {
 		ret_val = IXGBE_ERR_INVALID_LINK_SETTINGS;
 		goto out;
 	}
 
+	/* Low water mark of zero causes XOFF floods */
+	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+		if ((hw->fc.current_mode & ixgbe_fc_tx_pause) &&
+		    hw->fc.high_water[i]) {
+			if (!hw->fc.low_water[i] ||
+			    hw->fc.low_water[i] >= hw->fc.high_water[i]) {
+				hw_dbg(hw, "Invalid water mark configuration\n");
+				ret_val = IXGBE_ERR_INVALID_LINK_SETTINGS;
+				goto out;
+			}
+		}
+	}
+
 	/*
 	 * On 82598 having Rx FC on causes resets while doing 1G
 	 * so if it's on turn it off once we know link_speed. For
@@ -432,12 +438,11 @@ static s32 ixgbe_fc_enable_82598(struct ixgbe_hw *hw)
 	IXGBE_WRITE_REG(hw, IXGBE_FCTRL, fctrl_reg);
 	IXGBE_WRITE_REG(hw, IXGBE_RMCS, rmcs_reg);
 
-	fcrtl = (hw->fc.low_water << 10) | IXGBE_FCRTL_XONE;
-
 	/* Set up and enable Rx high/low water mark thresholds, enable XON. */
 	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
 		if ((hw->fc.current_mode & ixgbe_fc_tx_pause) &&
 		    hw->fc.high_water[i]) {
+			fcrtl = (hw->fc.low_water[i] << 10) | IXGBE_FCRTL_XONE;
 			fcrth = (hw->fc.high_water[i] << 10) | IXGBE_FCRTH_FCEN;
 			IXGBE_WRITE_REG(hw, IXGBE_FCRTL(i), fcrtl);
 			IXGBE_WRITE_REG(hw, IXGBE_FCRTH(i), fcrth);
@@ -468,7 +473,7 @@ out:
  *  Restarts the link.  Performs autonegotiation if needed.
  **/
 static s32 ixgbe_start_mac_link_82598(struct ixgbe_hw *hw,
-                                      bool autoneg_wait_to_complete)
+				      bool autoneg_wait_to_complete)
 {
 	u32 autoc_reg;
 	u32 links_reg;
@@ -550,8 +555,8 @@ static s32 ixgbe_validate_link_ready(struct ixgbe_hw *hw)
  *  Reads the links register to determine if link is up and the current speed
  **/
 static s32 ixgbe_check_mac_link_82598(struct ixgbe_hw *hw,
-                                      ixgbe_link_speed *speed, bool *link_up,
-                                      bool link_up_wait_to_complete)
+				      ixgbe_link_speed *speed, bool *link_up,
+				      bool link_up_wait_to_complete)
 {
 	u32 links_reg;
 	u32 i;
@@ -567,7 +572,7 @@ static s32 ixgbe_check_mac_link_82598(struct ixgbe_hw *hw,
 		hw->phy.ops.read_reg(hw, 0xC79F, MDIO_MMD_PMAPMD, &link_reg);
 		hw->phy.ops.read_reg(hw, 0xC79F, MDIO_MMD_PMAPMD, &link_reg);
 		hw->phy.ops.read_reg(hw, 0xC00C, MDIO_MMD_PMAPMD,
-		                     &adapt_comp_reg);
+				     &adapt_comp_reg);
 		if (link_up_wait_to_complete) {
 			for (i = 0; i < IXGBE_LINK_UP_TIME; i++) {
 				if ((link_reg & 1) &&
@@ -579,11 +584,11 @@ static s32 ixgbe_check_mac_link_82598(struct ixgbe_hw *hw,
 				}
 				msleep(100);
 				hw->phy.ops.read_reg(hw, 0xC79F,
-				                     MDIO_MMD_PMAPMD,
-				                     &link_reg);
+						     MDIO_MMD_PMAPMD,
+						     &link_reg);
 				hw->phy.ops.read_reg(hw, 0xC00C,
-				                     MDIO_MMD_PMAPMD,
-				                     &adapt_comp_reg);
+						     MDIO_MMD_PMAPMD,
+						     &adapt_comp_reg);
 			}
 		} else {
 			if ((link_reg & 1) && ((adapt_comp_reg & 1) == 0))
@@ -656,7 +661,7 @@ static s32 ixgbe_setup_mac_link_82598(struct ixgbe_hw *hw,
 
 	/* Set KX4/KX support according to speed requested */
 	else if (link_mode == IXGBE_AUTOC_LMS_KX4_AN ||
-	         link_mode == IXGBE_AUTOC_LMS_KX4_AN_1G_AN) {
+		 link_mode == IXGBE_AUTOC_LMS_KX4_AN_1G_AN) {
 		autoc &= ~IXGBE_AUTOC_KX4_KX_SUPP_MASK;
 		if (speed & IXGBE_LINK_SPEED_10GB_FULL)
 			autoc |= IXGBE_AUTOC_KX4_SUPP;
@@ -689,14 +694,14 @@ static s32 ixgbe_setup_mac_link_82598(struct ixgbe_hw *hw,
  *  Sets the link speed in the AUTOC register in the MAC and restarts link.
  **/
 static s32 ixgbe_setup_copper_link_82598(struct ixgbe_hw *hw,
-                                               ixgbe_link_speed speed,
-                                               bool autoneg_wait_to_complete)
+					       ixgbe_link_speed speed,
+					       bool autoneg_wait_to_complete)
 {
 	s32 status;
 
 	/* Setup the PHY according to input speed */
 	status = hw->phy.ops.setup_link_speed(hw, speed,
-	                                      autoneg_wait_to_complete);
+					      autoneg_wait_to_complete);
 	/* Set up MAC */
 	ixgbe_start_mac_link_82598(hw, autoneg_wait_to_complete);
 
@@ -735,28 +740,28 @@ static s32 ixgbe_reset_hw_82598(struct ixgbe_hw *hw)
 	if (analog_val & IXGBE_ATLAS_PDN_TX_REG_EN) {
 		/* Enable Tx Atlas so packets can be transmitted again */
 		hw->mac.ops.read_analog_reg8(hw, IXGBE_ATLAS_PDN_LPBK,
-		                             &analog_val);
+					     &analog_val);
 		analog_val &= ~IXGBE_ATLAS_PDN_TX_REG_EN;
 		hw->mac.ops.write_analog_reg8(hw, IXGBE_ATLAS_PDN_LPBK,
-		                              analog_val);
+					      analog_val);
 
 		hw->mac.ops.read_analog_reg8(hw, IXGBE_ATLAS_PDN_10G,
-		                             &analog_val);
+					     &analog_val);
 		analog_val &= ~IXGBE_ATLAS_PDN_TX_10G_QL_ALL;
 		hw->mac.ops.write_analog_reg8(hw, IXGBE_ATLAS_PDN_10G,
-		                              analog_val);
+					      analog_val);
 
 		hw->mac.ops.read_analog_reg8(hw, IXGBE_ATLAS_PDN_1G,
-		                             &analog_val);
+					     &analog_val);
 		analog_val &= ~IXGBE_ATLAS_PDN_TX_1G_QL_ALL;
 		hw->mac.ops.write_analog_reg8(hw, IXGBE_ATLAS_PDN_1G,
-		                              analog_val);
+					      analog_val);
 
 		hw->mac.ops.read_analog_reg8(hw, IXGBE_ATLAS_PDN_AN,
-		                             &analog_val);
+					     &analog_val);
 		analog_val &= ~IXGBE_ATLAS_PDN_TX_AN_QL_ALL;
 		hw->mac.ops.write_analog_reg8(hw, IXGBE_ATLAS_PDN_AN,
-		                              analog_val);
+					      analog_val);
 	}
 
 	/* Reset PHY */
@@ -955,7 +960,7 @@ static s32 ixgbe_clear_vfta_82598(struct ixgbe_hw *hw)
 	for (vlanbyte = 0; vlanbyte < 4; vlanbyte++)
 		for (offset = 0; offset < hw->mac.vft_size; offset++)
 			IXGBE_WRITE_REG(hw, IXGBE_VFTAVIND(vlanbyte, offset),
-			                0);
+					0);
 
 	return 0;
 }
@@ -973,7 +978,7 @@ static s32 ixgbe_read_analog_reg8_82598(struct ixgbe_hw *hw, u32 reg, u8 *val)
 	u32  atlas_ctl;
 
 	IXGBE_WRITE_REG(hw, IXGBE_ATLASCTL,
-	                IXGBE_ATLASCTL_WRITE_CMD | (reg << 8));
+			IXGBE_ATLASCTL_WRITE_CMD | (reg << 8));
 	IXGBE_WRITE_FLUSH(hw);
 	udelay(10);
 	atlas_ctl = IXGBE_READ_REG(hw, IXGBE_ATLASCTL);
@@ -1273,8 +1278,6 @@ static void ixgbe_set_rxpba_82598(struct ixgbe_hw *hw, int num_pb,
 	/* Setup Tx packet buffer sizes */
 	for (i = 0; i < IXGBE_MAX_PACKET_BUFFERS; i++)
 		IXGBE_WRITE_REG(hw, IXGBE_TXPBSIZE(i), IXGBE_TXPBSIZE_40KB);
-
-	return;
 }
 
 static struct ixgbe_mac_operations mac_ops_82598 = {
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c
index f32b3dd1ba8e..bc7c924240a5 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c
@@ -48,17 +48,17 @@ static s32 ixgbe_setup_mac_link_multispeed_fiber(struct ixgbe_hw *hw,
 						 ixgbe_link_speed speed,
 						 bool autoneg_wait_to_complete);
 static s32 ixgbe_setup_mac_link_smartspeed(struct ixgbe_hw *hw,
-                                           ixgbe_link_speed speed,
-                                           bool autoneg_wait_to_complete);
+					   ixgbe_link_speed speed,
+					   bool autoneg_wait_to_complete);
 static void ixgbe_stop_mac_link_on_d3_82599(struct ixgbe_hw *hw);
 static s32 ixgbe_start_mac_link_82599(struct ixgbe_hw *hw,
 				      bool autoneg_wait_to_complete);
 static s32 ixgbe_setup_mac_link_82599(struct ixgbe_hw *hw,
-                               ixgbe_link_speed speed,
-                               bool autoneg_wait_to_complete);
+			       ixgbe_link_speed speed,
+			       bool autoneg_wait_to_complete);
 static s32 ixgbe_setup_copper_link_82599(struct ixgbe_hw *hw,
-                                         ixgbe_link_speed speed,
-                                         bool autoneg_wait_to_complete);
+					 ixgbe_link_speed speed,
+					 bool autoneg_wait_to_complete);
 static s32 ixgbe_verify_fw_version_82599(struct ixgbe_hw *hw);
 static s32 ixgbe_read_i2c_byte_82599(struct ixgbe_hw *hw, u8 byte_offset,
 				     u8 dev_addr, u8 *data);
@@ -96,9 +96,9 @@ static void ixgbe_init_mac_link_ops_82599(struct ixgbe_hw *hw)
 	if ((mac->ops.get_media_type(hw) == ixgbe_media_type_fiber) &&
 	    !ixgbe_mng_enabled(hw)) {
 		mac->ops.disable_tx_laser =
-		                       &ixgbe_disable_tx_laser_multispeed_fiber;
+				       &ixgbe_disable_tx_laser_multispeed_fiber;
 		mac->ops.enable_tx_laser =
-		                        &ixgbe_enable_tx_laser_multispeed_fiber;
+					&ixgbe_enable_tx_laser_multispeed_fiber;
 		mac->ops.flap_tx_laser = &ixgbe_flap_tx_laser_multispeed_fiber;
 	} else {
 		mac->ops.disable_tx_laser = NULL;
@@ -132,13 +132,13 @@ static s32 ixgbe_setup_sfp_modules_82599(struct ixgbe_hw *hw)
 		hw->phy.ops.reset = NULL;
 
 		ret_val = ixgbe_get_sfp_init_sequence_offsets(hw, &list_offset,
-		                                              &data_offset);
+							      &data_offset);
 		if (ret_val != 0)
 			goto setup_sfp_out;
 
 		/* PHY config will finish before releasing the semaphore */
 		ret_val = hw->mac.ops.acquire_swfw_sync(hw,
-		                                        IXGBE_GSSR_MAC_CSR_SM);
+							IXGBE_GSSR_MAC_CSR_SM);
 		if (ret_val != 0) {
 			ret_val = IXGBE_ERR_SWFW_SYNC;
 			goto setup_sfp_out;
@@ -334,7 +334,7 @@ static s32 ixgbe_init_phy_ops_82599(struct ixgbe_hw *hw)
 		phy->ops.check_link = &ixgbe_check_phy_link_tnx;
 		phy->ops.setup_link = &ixgbe_setup_phy_link_tnx;
 		phy->ops.get_firmware_version =
-		             &ixgbe_get_phy_firmware_version_tnx;
+			     &ixgbe_get_phy_firmware_version_tnx;
 		break;
 	default:
 		break;
@@ -352,7 +352,7 @@ static s32 ixgbe_init_phy_ops_82599(struct ixgbe_hw *hw)
  *  Determines the link capabilities by reading the AUTOC register.
  **/
 static s32 ixgbe_get_link_capabilities_82599(struct ixgbe_hw *hw,
-                                             ixgbe_link_speed *speed,
+					     ixgbe_link_speed *speed,
 					     bool *autoneg)
 {
 	s32 status = 0;
@@ -543,7 +543,7 @@ static void ixgbe_stop_mac_link_on_d3_82599(struct ixgbe_hw *hw)
  *  Restarts the link.  Performs autonegotiation if needed.
  **/
 static s32 ixgbe_start_mac_link_82599(struct ixgbe_hw *hw,
-                               bool autoneg_wait_to_complete)
+			       bool autoneg_wait_to_complete)
 {
 	u32 autoc_reg;
 	u32 links_reg;
@@ -672,8 +672,8 @@ static void ixgbe_flap_tx_laser_multispeed_fiber(struct ixgbe_hw *hw)
  *  Set the link speed in the AUTOC register and restarts link.
  **/
 static s32 ixgbe_setup_mac_link_multispeed_fiber(struct ixgbe_hw *hw,
-                                          ixgbe_link_speed speed,
-                                          bool autoneg_wait_to_complete)
+					  ixgbe_link_speed speed,
+					  bool autoneg_wait_to_complete)
 {
 	s32 status = 0;
 	ixgbe_link_speed link_speed = IXGBE_LINK_SPEED_UNKNOWN;
@@ -820,8 +820,8 @@ static s32 ixgbe_setup_mac_link_multispeed_fiber(struct ixgbe_hw *hw,
 	 */
 	if (speedcnt > 1)
 		status = ixgbe_setup_mac_link_multispeed_fiber(hw,
-		                                               highest_link_speed,
-		                                               autoneg_wait_to_complete);
+							       highest_link_speed,
+							       autoneg_wait_to_complete);
 
 out:
 	/* Set autoneg_advertised value based on input link speed */
@@ -1009,8 +1009,8 @@ static s32 ixgbe_setup_mac_link_82599(struct ixgbe_hw *hw,
 		if (speed & IXGBE_LINK_SPEED_1GB_FULL)
 			autoc |= IXGBE_AUTOC_KX_SUPP;
 	} else if ((pma_pmd_1g == IXGBE_AUTOC_1G_SFI) &&
-	           (link_mode == IXGBE_AUTOC_LMS_1G_LINK_NO_AN ||
-	            link_mode == IXGBE_AUTOC_LMS_1G_AN)) {
+		   (link_mode == IXGBE_AUTOC_LMS_1G_LINK_NO_AN ||
+		    link_mode == IXGBE_AUTOC_LMS_1G_AN)) {
 		/* Switch from 1G SFI to 10G SFI if requested */
 		if ((speed == IXGBE_LINK_SPEED_10GB_FULL) &&
 		    (pma_pmd_10g_serial == IXGBE_AUTOC2_10G_SFI)) {
@@ -1018,7 +1018,7 @@ static s32 ixgbe_setup_mac_link_82599(struct ixgbe_hw *hw,
 			autoc |= IXGBE_AUTOC_LMS_10G_SERIAL;
 		}
 	} else if ((pma_pmd_10g_serial == IXGBE_AUTOC2_10G_SFI) &&
-	           (link_mode == IXGBE_AUTOC_LMS_10G_SERIAL)) {
+		   (link_mode == IXGBE_AUTOC_LMS_10G_SERIAL)) {
 		/* Switch from 10G SFI to 1G SFI if requested */
 		if ((speed == IXGBE_LINK_SPEED_1GB_FULL) &&
 		    (pma_pmd_1g == IXGBE_AUTOC_1G_SFI)) {
@@ -1051,7 +1051,7 @@ static s32 ixgbe_setup_mac_link_82599(struct ixgbe_hw *hw,
 				}
 				if (!(links_reg & IXGBE_LINKS_KX_AN_COMP)) {
 					status =
-					        IXGBE_ERR_AUTONEG_NOT_COMPLETE;
+						IXGBE_ERR_AUTONEG_NOT_COMPLETE;
 					hw_dbg(hw, "Autoneg did not complete.\n");
 				}
 			}
@@ -1074,14 +1074,14 @@ out:
  *  Restarts link on PHY and MAC based on settings passed in.
  **/
 static s32 ixgbe_setup_copper_link_82599(struct ixgbe_hw *hw,
-                                         ixgbe_link_speed speed,
-                                         bool autoneg_wait_to_complete)
+					 ixgbe_link_speed speed,
+					 bool autoneg_wait_to_complete)
 {
 	s32 status;
 
 	/* Setup the PHY according to input speed */
 	status = hw->phy.ops.setup_link_speed(hw, speed,
-	                                      autoneg_wait_to_complete);
+					      autoneg_wait_to_complete);
 	/* Set up MAC */
 	ixgbe_start_mac_link_82599(hw, autoneg_wait_to_complete);
 
@@ -1224,7 +1224,7 @@ mac_reset_top:
 		    (hw->mac.orig_autoc2 & IXGBE_AUTOC2_UPPER_MASK)) {
 			autoc2 &= ~IXGBE_AUTOC2_UPPER_MASK;
 			autoc2 |= (hw->mac.orig_autoc2 &
-			           IXGBE_AUTOC2_UPPER_MASK);
+				   IXGBE_AUTOC2_UPPER_MASK);
 			IXGBE_WRITE_REG(hw, IXGBE_AUTOC2, autoc2);
 		}
 	}
@@ -1246,7 +1246,7 @@ mac_reset_top:
 	/* Add the SAN MAC address to the RAR only if it's a valid address */
 	if (is_valid_ether_addr(hw->mac.san_addr)) {
 		hw->mac.ops.set_rar(hw, hw->mac.num_rar_entries - 1,
-		                    hw->mac.san_addr, 0, IXGBE_RAH_AV);
+				    hw->mac.san_addr, 0, IXGBE_RAH_AV);
 
 		/* Save the SAN MAC RAR index */
 		hw->mac.san_mac_rar_index = hw->mac.num_rar_entries - 1;
@@ -1257,7 +1257,7 @@ mac_reset_top:
 
 	/* Store the alternative WWNN/WWPN prefix */
 	hw->mac.ops.get_wwn_prefix(hw, &hw->mac.wwnn_prefix,
-	                               &hw->mac.wwpn_prefix);
+				       &hw->mac.wwpn_prefix);
 
 reset_hw_out:
 	return status;
@@ -1271,6 +1271,7 @@ s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw)
 {
 	int i;
 	u32 fdirctrl = IXGBE_READ_REG(hw, IXGBE_FDIRCTRL);
+
 	fdirctrl &= ~IXGBE_FDIRCTRL_INIT_DONE;
 
 	/*
@@ -1284,8 +1285,7 @@ s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw)
 		udelay(10);
 	}
 	if (i >= IXGBE_FDIRCMD_CMD_POLL) {
-		hw_dbg(hw, "Flow Director previous command isn't complete, "
-		       "aborting table re-initialization.\n");
+		hw_dbg(hw, "Flow Director previous command isn't complete, aborting table re-initialization.\n");
 		return IXGBE_ERR_FDIR_REINIT_FAILED;
 	}
 
@@ -1299,12 +1299,12 @@ s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw)
 	 * - write 0 to bit 8 of FDIRCMD register
 	 */
 	IXGBE_WRITE_REG(hw, IXGBE_FDIRCMD,
-	                (IXGBE_READ_REG(hw, IXGBE_FDIRCMD) |
-	                 IXGBE_FDIRCMD_CLEARHT));
+			(IXGBE_READ_REG(hw, IXGBE_FDIRCMD) |
+			 IXGBE_FDIRCMD_CLEARHT));
 	IXGBE_WRITE_FLUSH(hw);
 	IXGBE_WRITE_REG(hw, IXGBE_FDIRCMD,
-	                (IXGBE_READ_REG(hw, IXGBE_FDIRCMD) &
-	                 ~IXGBE_FDIRCMD_CLEARHT));
+			(IXGBE_READ_REG(hw, IXGBE_FDIRCMD) &
+			 ~IXGBE_FDIRCMD_CLEARHT));
 	IXGBE_WRITE_FLUSH(hw);
 	/*
 	 * Clear FDIR Hash register to clear any leftover hashes
@@ -1319,7 +1319,7 @@ s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw)
 	/* Poll init-done after we write FDIRCTRL register */
 	for (i = 0; i < IXGBE_FDIR_INIT_DONE_POLL; i++) {
 		if (IXGBE_READ_REG(hw, IXGBE_FDIRCTRL) &
-		                   IXGBE_FDIRCTRL_INIT_DONE)
+				   IXGBE_FDIRCTRL_INIT_DONE)
 			break;
 		usleep_range(1000, 2000);
 	}
@@ -1368,7 +1368,7 @@ static void ixgbe_fdir_enable_82599(struct ixgbe_hw *hw, u32 fdirctrl)
 	IXGBE_WRITE_FLUSH(hw);
 	for (i = 0; i < IXGBE_FDIR_INIT_DONE_POLL; i++) {
 		if (IXGBE_READ_REG(hw, IXGBE_FDIRCTRL) &
-		                   IXGBE_FDIRCTRL_INIT_DONE)
+				   IXGBE_FDIRCTRL_INIT_DONE)
 			break;
 		usleep_range(1000, 2000);
 	}
@@ -1453,7 +1453,7 @@ do { \
 		bucket_hash ^= hi_hash_dword >> n; \
 	else if (IXGBE_ATR_SIGNATURE_HASH_KEY & (0x01 << (n + 16))) \
 		sig_hash ^= hi_hash_dword << (16 - n); \
-} while (0);
+} while (0)
 
 /**
  *  ixgbe_atr_compute_sig_hash_82599 - Compute the signature hash
@@ -1529,9 +1529,9 @@ static u32 ixgbe_atr_compute_sig_hash_82599(union ixgbe_atr_hash_dword input,
  *  @queue: queue index to direct traffic to
  **/
 s32 ixgbe_fdir_add_signature_filter_82599(struct ixgbe_hw *hw,
-                                          union ixgbe_atr_hash_dword input,
-                                          union ixgbe_atr_hash_dword common,
-                                          u8 queue)
+					  union ixgbe_atr_hash_dword input,
+					  union ixgbe_atr_hash_dword common,
+					  u8 queue)
 {
 	u64  fdirhashcmd;
 	u32  fdircmd;
@@ -1555,7 +1555,7 @@ s32 ixgbe_fdir_add_signature_filter_82599(struct ixgbe_hw *hw,
 
 	/* configure FDIRCMD register */
 	fdircmd = IXGBE_FDIRCMD_CMD_ADD_FLOW | IXGBE_FDIRCMD_FILTER_UPDATE |
-	          IXGBE_FDIRCMD_LAST | IXGBE_FDIRCMD_QUEUE_EN;
+		  IXGBE_FDIRCMD_LAST | IXGBE_FDIRCMD_QUEUE_EN;
 	fdircmd |= input.formatted.flow_type << IXGBE_FDIRCMD_FLOW_TYPE_SHIFT;
 	fdircmd |= (u32)queue << IXGBE_FDIRCMD_RX_QUEUE_SHIFT;
 
@@ -1579,7 +1579,7 @@ do { \
 		bucket_hash ^= lo_hash_dword >> n; \
 	if (IXGBE_ATR_BUCKET_HASH_KEY & (0x01 << (n + 16))) \
 		bucket_hash ^= hi_hash_dword >> n; \
-} while (0);
+} while (0)
 
 /**
  *  ixgbe_atr_compute_perfect_hash_82599 - Compute the perfect filter hash
@@ -1651,6 +1651,7 @@ void ixgbe_atr_compute_perfect_hash_82599(union ixgbe_atr_input *input,
 static u32 ixgbe_get_fdirtcpm_82599(union ixgbe_atr_input *input_mask)
 {
 	u32 mask = ntohs(input_mask->formatted.dst_port);
+
 	mask <<= IXGBE_FDIRTCPM_DPORTM_SHIFT;
 	mask |= ntohs(input_mask->formatted.src_port);
 	mask = ((mask & 0x55555555) << 1) | ((mask & 0xAAAAAAAA) >> 1);
@@ -1885,7 +1886,7 @@ static s32 ixgbe_read_analog_reg8_82599(struct ixgbe_hw *hw, u32 reg, u8 *val)
 	u32  core_ctl;
 
 	IXGBE_WRITE_REG(hw, IXGBE_CORECTL, IXGBE_CORECTL_WRITE_CMD |
-	                (reg << 8));
+			(reg << 8));
 	IXGBE_WRITE_FLUSH(hw);
 	udelay(10);
 	core_ctl = IXGBE_READ_REG(hw, IXGBE_CORECTL);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c
index 981b8a7b100d..4e5385a2a465 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c
@@ -41,7 +41,7 @@ static void ixgbe_release_eeprom_semaphore(struct ixgbe_hw *hw);
 static s32 ixgbe_ready_eeprom(struct ixgbe_hw *hw);
 static void ixgbe_standby_eeprom(struct ixgbe_hw *hw);
 static void ixgbe_shift_out_eeprom_bits(struct ixgbe_hw *hw, u16 data,
-                                        u16 count);
+					u16 count);
 static u16 ixgbe_shift_in_eeprom_bits(struct ixgbe_hw *hw, u16 count);
 static void ixgbe_raise_eeprom_clk(struct ixgbe_hw *hw, u32 *eec);
 static void ixgbe_lower_eeprom_clk(struct ixgbe_hw *hw, u32 *eec);
@@ -271,6 +271,7 @@ out:
  **/
 s32 ixgbe_start_hw_generic(struct ixgbe_hw *hw)
 {
+	s32 ret_val;
 	u32 ctrl_ext;
 
 	/* Set the media type */
@@ -292,12 +293,15 @@ s32 ixgbe_start_hw_generic(struct ixgbe_hw *hw)
 	IXGBE_WRITE_FLUSH(hw);
 
 	/* Setup flow control */
-	ixgbe_setup_fc(hw);
+	ret_val = ixgbe_setup_fc(hw);
+	if (!ret_val)
+		goto out;
 
 	/* Clear adapter stopped flag */
 	hw->adapter_stopped = false;
 
-	return 0;
+out:
+	return ret_val;
 }
 
 /**
@@ -481,7 +485,7 @@ s32 ixgbe_clear_hw_cntrs_generic(struct ixgbe_hw *hw)
  *  Reads the part number string from the EEPROM.
  **/
 s32 ixgbe_read_pba_string_generic(struct ixgbe_hw *hw, u8 *pba_num,
-                                  u32 pba_num_size)
+				  u32 pba_num_size)
 {
 	s32 ret_val;
 	u16 data;
@@ -814,9 +818,8 @@ s32 ixgbe_init_eeprom_params_generic(struct ixgbe_hw *hw)
 			eeprom->address_bits = 16;
 		else
 			eeprom->address_bits = 8;
-		hw_dbg(hw, "Eeprom params: type = %d, size = %d, address bits: "
-			  "%d\n", eeprom->type, eeprom->word_size,
-			  eeprom->address_bits);
+		hw_dbg(hw, "Eeprom params: type = %d, size = %d, address bits: %d\n",
+		       eeprom->type, eeprom->word_size, eeprom->address_bits);
 	}
 
 	return 0;
@@ -1388,8 +1391,7 @@ static s32 ixgbe_get_eeprom_semaphore(struct ixgbe_hw *hw)
 	}
 
 	if (i == timeout) {
-		hw_dbg(hw, "Driver can't access the Eeprom - SMBI Semaphore "
-		       "not granted.\n");
+		hw_dbg(hw, "Driver can't access the Eeprom - SMBI Semaphore not granted.\n");
 		/*
 		 * this release is particularly important because our attempts
 		 * above to get the semaphore may have succeeded, and if there
@@ -1434,14 +1436,12 @@ static s32 ixgbe_get_eeprom_semaphore(struct ixgbe_hw *hw)
 		 * was not granted because we don't have access to the EEPROM
 		 */
 		if (i >= timeout) {
-			hw_dbg(hw, "SWESMBI Software EEPROM semaphore "
-			       "not granted.\n");
+			hw_dbg(hw, "SWESMBI Software EEPROM semaphore not granted.\n");
 			ixgbe_release_eeprom_semaphore(hw);
 			status = IXGBE_ERR_EEPROM;
 		}
 	} else {
-		hw_dbg(hw, "Software semaphore SMBI between device drivers "
-		       "not granted.\n");
+		hw_dbg(hw, "Software semaphore SMBI between device drivers not granted.\n");
 	}
 
 	return status;
@@ -1483,7 +1483,7 @@ static s32 ixgbe_ready_eeprom(struct ixgbe_hw *hw)
 	 */
 	for (i = 0; i < IXGBE_EEPROM_MAX_RETRY_SPI; i += 5) {
 		ixgbe_shift_out_eeprom_bits(hw, IXGBE_EEPROM_RDSR_OPCODE_SPI,
-		                            IXGBE_EEPROM_OPCODE_BITS);
+					    IXGBE_EEPROM_OPCODE_BITS);
 		spi_stat_reg = (u8)ixgbe_shift_in_eeprom_bits(hw, 8);
 		if (!(spi_stat_reg & IXGBE_EEPROM_STATUS_RDY_SPI))
 			break;
@@ -1532,7 +1532,7 @@ static void ixgbe_standby_eeprom(struct ixgbe_hw *hw)
  *  @count: number of bits to shift out
  **/
 static void ixgbe_shift_out_eeprom_bits(struct ixgbe_hw *hw, u16 data,
-                                        u16 count)
+					u16 count)
 {
 	u32 eec;
 	u32 mask;
@@ -1736,7 +1736,7 @@ u16 ixgbe_calc_eeprom_checksum_generic(struct ixgbe_hw *hw)
  *  caller does not need checksum_val, the value can be NULL.
  **/
 s32 ixgbe_validate_eeprom_checksum_generic(struct ixgbe_hw *hw,
-                                           u16 *checksum_val)
+					   u16 *checksum_val)
 {
 	s32 status;
 	u16 checksum;
@@ -1809,7 +1809,7 @@ s32 ixgbe_update_eeprom_checksum_generic(struct ixgbe_hw *hw)
  *  Puts an ethernet address into a receive address register.
  **/
 s32 ixgbe_set_rar_generic(struct ixgbe_hw *hw, u32 index, u8 *addr, u32 vmdq,
-                          u32 enable_addr)
+			  u32 enable_addr)
 {
 	u32 rar_low, rar_high;
 	u32 rar_entries = hw->mac.num_rar_entries;
@@ -2053,7 +2053,7 @@ s32 ixgbe_update_mc_addr_list_generic(struct ixgbe_hw *hw,
 
 	if (hw->addr_ctrl.mta_in_use > 0)
 		IXGBE_WRITE_REG(hw, IXGBE_MCSTCTRL,
-		                IXGBE_MCSTCTRL_MFE | hw->mac.mc_filter_type);
+				IXGBE_MCSTCTRL_MFE | hw->mac.mc_filter_type);
 
 	hw_dbg(hw, "ixgbe_update_mc_addr_list_generic Complete\n");
 	return 0;
@@ -2071,7 +2071,7 @@ s32 ixgbe_enable_mc_generic(struct ixgbe_hw *hw)
 
 	if (a->mta_in_use > 0)
 		IXGBE_WRITE_REG(hw, IXGBE_MCSTCTRL, IXGBE_MCSTCTRL_MFE |
-		                hw->mac.mc_filter_type);
+				hw->mac.mc_filter_type);
 
 	return 0;
 }
@@ -2106,19 +2106,25 @@ s32 ixgbe_fc_enable_generic(struct ixgbe_hw *hw)
 	u32 fcrtl, fcrth;
 	int i;
 
-	/*
-	 * Validate the water mark configuration for packet buffer 0.  Zero
-	 * water marks indicate that the packet buffer was not configured
-	 * and the watermarks for packet buffer 0 should always be configured.
-	 */
-	if (!hw->fc.low_water ||
-	    !hw->fc.high_water[0] ||
-	    !hw->fc.pause_time) {
-		hw_dbg(hw, "Invalid water mark configuration\n");
+	/* Validate the water mark configuration. */
+	if (!hw->fc.pause_time) {
 		ret_val = IXGBE_ERR_INVALID_LINK_SETTINGS;
 		goto out;
 	}
 
+	/* Low water mark of zero causes XOFF floods */
+	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+		if ((hw->fc.current_mode & ixgbe_fc_tx_pause) &&
+		    hw->fc.high_water[i]) {
+			if (!hw->fc.low_water[i] ||
+			    hw->fc.low_water[i] >= hw->fc.high_water[i]) {
+				hw_dbg(hw, "Invalid water mark configuration\n");
+				ret_val = IXGBE_ERR_INVALID_LINK_SETTINGS;
+				goto out;
+			}
+		}
+	}
+
 	/* Negotiate the fc mode to use */
 	ixgbe_fc_autoneg(hw);
 
@@ -2181,12 +2187,11 @@ s32 ixgbe_fc_enable_generic(struct ixgbe_hw *hw)
 	IXGBE_WRITE_REG(hw, IXGBE_MFLCN, mflcn_reg);
 	IXGBE_WRITE_REG(hw, IXGBE_FCCFG, fccfg_reg);
 
-	fcrtl = (hw->fc.low_water << 10) | IXGBE_FCRTL_XONE;
-
 	/* Set up and enable Rx high/low water mark thresholds, enable XON. */
 	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
 		if ((hw->fc.current_mode & ixgbe_fc_tx_pause) &&
 		    hw->fc.high_water[i]) {
+			fcrtl = (hw->fc.low_water[i] << 10) | IXGBE_FCRTL_XONE;
 			IXGBE_WRITE_REG(hw, IXGBE_FCRTL_82599(i), fcrtl);
 			fcrth = (hw->fc.high_water[i] << 10) | IXGBE_FCRTH_FCEN;
 		} else {
@@ -2654,8 +2659,7 @@ s32 ixgbe_disable_rx_buff_generic(struct ixgbe_hw *hw)
 
 	/* For informational purposes only */
 	if (i >= IXGBE_MAX_SECRX_POLL)
-		hw_dbg(hw, "Rx unit being enabled before security "
-		       "path fully disabled.  Continuing with init.\n");
+		hw_dbg(hw, "Rx unit being enabled before security path fully disabled. Continuing with init.\n");
 
 	return 0;
 
@@ -2782,7 +2786,7 @@ out:
  *  get and set mac_addr routines.
  **/
 static s32 ixgbe_get_san_mac_addr_offset(struct ixgbe_hw *hw,
-                                        u16 *san_mac_offset)
+					u16 *san_mac_offset)
 {
 	s32 ret_val;
 
@@ -2828,7 +2832,7 @@ s32 ixgbe_get_san_mac_addr_generic(struct ixgbe_hw *hw, u8 *san_mac_addr)
 	hw->mac.ops.set_lan_id(hw);
 	/* apply the port offset to the address offset */
 	(hw->bus.func) ? (san_mac_offset += IXGBE_SAN_MAC_ADDR_PORT1_OFFSET) :
-	                 (san_mac_offset += IXGBE_SAN_MAC_ADDR_PORT0_OFFSET);
+			 (san_mac_offset += IXGBE_SAN_MAC_ADDR_PORT0_OFFSET);
 	for (i = 0; i < 3; i++) {
 		ret_val = hw->eeprom.ops.read(hw, san_mac_offset,
 					      &san_mac_data);
@@ -3068,7 +3072,7 @@ static s32 ixgbe_find_vlvf_slot(struct ixgbe_hw *hw, u32 vlan)
  *  Turn on/off specified VLAN in the VLAN filter table.
  **/
 s32 ixgbe_set_vfta_generic(struct ixgbe_hw *hw, u32 vlan, u32 vind,
-                           bool vlan_on)
+			   bool vlan_on)
 {
 	s32 regindex;
 	u32 bitindex;
@@ -3190,9 +3194,9 @@ s32 ixgbe_set_vfta_generic(struct ixgbe_hw *hw, u32 vlan, u32 vind,
 				 * Ignore it. */
 				vfta_changed = false;
 			}
-		}
-		else
+		} else {
 			IXGBE_WRITE_REG(hw, IXGBE_VLVF(vlvf_index), 0);
+		}
 	}
 
 	if (vfta_changed)
@@ -3292,7 +3296,7 @@ s32 ixgbe_check_mac_link_generic(struct ixgbe_hw *hw, ixgbe_link_speed *speed,
  *  block to check the support for the alternative WWNN/WWPN prefix support.
  **/
 s32 ixgbe_get_wwn_prefix_generic(struct ixgbe_hw *hw, u16 *wwnn_prefix,
-                                        u16 *wwpn_prefix)
+					u16 *wwpn_prefix)
 {
 	u16 offset, caps;
 	u16 alt_san_mac_blk_offset;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.h
index f12c40fb5537..2ae5d4b8fc93 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.h
@@ -39,7 +39,7 @@ s32 ixgbe_start_hw_generic(struct ixgbe_hw *hw);
 s32 ixgbe_start_hw_gen2(struct ixgbe_hw *hw);
 s32 ixgbe_clear_hw_cntrs_generic(struct ixgbe_hw *hw);
 s32 ixgbe_read_pba_string_generic(struct ixgbe_hw *hw, u8 *pba_num,
-                                  u32 pba_num_size);
+				  u32 pba_num_size);
 s32 ixgbe_get_mac_addr_generic(struct ixgbe_hw *hw, u8 *mac_addr);
 enum ixgbe_bus_width ixgbe_convert_bus_width(u16 link_status);
 enum ixgbe_bus_speed ixgbe_convert_bus_speed(u16 link_status);
@@ -61,16 +61,16 @@ s32 ixgbe_write_eewr_generic(struct ixgbe_hw *hw, u16 offset, u16 data);
 s32 ixgbe_write_eewr_buffer_generic(struct ixgbe_hw *hw, u16 offset,
 				    u16 words, u16 *data);
 s32 ixgbe_read_eeprom_bit_bang_generic(struct ixgbe_hw *hw, u16 offset,
-                                       u16 *data);
+				       u16 *data);
 s32 ixgbe_read_eeprom_buffer_bit_bang_generic(struct ixgbe_hw *hw, u16 offset,
 					      u16 words, u16 *data);
 u16 ixgbe_calc_eeprom_checksum_generic(struct ixgbe_hw *hw);
 s32 ixgbe_validate_eeprom_checksum_generic(struct ixgbe_hw *hw,
-                                           u16 *checksum_val);
+					   u16 *checksum_val);
 s32 ixgbe_update_eeprom_checksum_generic(struct ixgbe_hw *hw);
 
 s32 ixgbe_set_rar_generic(struct ixgbe_hw *hw, u32 index, u8 *addr, u32 vmdq,
-                          u32 enable_addr);
+			  u32 enable_addr);
 s32 ixgbe_clear_rar_generic(struct ixgbe_hw *hw, u32 index);
 s32 ixgbe_init_rx_addrs_generic(struct ixgbe_hw *hw);
 s32 ixgbe_update_mc_addr_list_generic(struct ixgbe_hw *hw,
@@ -92,13 +92,13 @@ s32 ixgbe_set_vmdq_san_mac_generic(struct ixgbe_hw *hw, u32 vmdq);
 s32 ixgbe_clear_vmdq_generic(struct ixgbe_hw *hw, u32 rar, u32 vmdq);
 s32 ixgbe_init_uta_tables_generic(struct ixgbe_hw *hw);
 s32 ixgbe_set_vfta_generic(struct ixgbe_hw *hw, u32 vlan,
-                           u32 vind, bool vlan_on);
+			   u32 vind, bool vlan_on);
 s32 ixgbe_clear_vfta_generic(struct ixgbe_hw *hw);
 s32 ixgbe_check_mac_link_generic(struct ixgbe_hw *hw,
-                                 ixgbe_link_speed *speed,
-                                 bool *link_up, bool link_up_wait_to_complete);
+				 ixgbe_link_speed *speed,
+				 bool *link_up, bool link_up_wait_to_complete);
 s32 ixgbe_get_wwn_prefix_generic(struct ixgbe_hw *hw, u16 *wwnn_prefix,
-                                 u16 *wwpn_prefix);
+				 u16 *wwpn_prefix);
 
 s32 prot_autoc_read_generic(struct ixgbe_hw *hw, bool *, u32 *reg_val);
 s32 prot_autoc_write_generic(struct ixgbe_hw *hw, u32 reg_val, bool locked);
@@ -141,8 +141,6 @@ static inline bool ixgbe_removed(void __iomem *addr)
 	return unlikely(!addr);
 }
 
-void ixgbe_check_remove(struct ixgbe_hw *hw, u32 reg);
-
 static inline void ixgbe_write_reg(struct ixgbe_hw *hw, u32 reg, u32 value)
 {
 	u8 __iomem *reg_addr = ACCESS_ONCE(hw->hw_addr);
@@ -172,18 +170,7 @@ static inline void ixgbe_write_reg64(struct ixgbe_hw *hw, u32 reg, u64 value)
 }
 #define IXGBE_WRITE_REG64(a, reg, value) ixgbe_write_reg64((a), (reg), (value))
 
-static inline u32 ixgbe_read_reg(struct ixgbe_hw *hw, u32 reg)
-{
-	u8 __iomem *reg_addr = ACCESS_ONCE(hw->hw_addr);
-	u32 value;
-
-	if (ixgbe_removed(reg_addr))
-		return IXGBE_FAILED_READ_REG;
-	value = readl(reg_addr + reg);
-	if (unlikely(value == IXGBE_FAILED_READ_REG))
-		ixgbe_check_remove(hw, reg);
-	return value;
-}
+u32 ixgbe_read_reg(struct ixgbe_hw *hw, u32 reg);
 #define IXGBE_READ_REG(a, reg) ixgbe_read_reg((a), (reg))
 
 #define IXGBE_WRITE_REG_ARRAY(a, reg, offset, value) \
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
index e055e000131b..a689ee0d4bed 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
@@ -267,7 +267,7 @@ void ixgbe_dcb_unpack_map(struct ixgbe_dcb_config *cfg, int direction, u8 *map)
  * Configure dcb settings and enable dcb mode.
  */
 s32 ixgbe_dcb_hw_config(struct ixgbe_hw *hw,
-                        struct ixgbe_dcb_config *dcb_config)
+			struct ixgbe_dcb_config *dcb_config)
 {
 	s32 ret = 0;
 	u8 pfc_en;
@@ -389,7 +389,6 @@ static void ixgbe_dcb_read_rtrup2tc_82599(struct ixgbe_hw *hw, u8 *map)
 	for (i = 0; i < MAX_USER_PRIORITY; i++)
 		map[i] = IXGBE_RTRUP2TC_UP_MASK &
 			(reg >> (i * IXGBE_RTRUP2TC_UP_SHIFT));
-	return;
 }
 
 void ixgbe_dcb_read_rtrup2tc(struct ixgbe_hw *hw, u8 *map)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82598.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82598.c
index 7a77f37a7cbc..d3ba63f9ad37 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82598.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82598.c
@@ -208,7 +208,6 @@ s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *hw, u8 pfc_en)
 
 	IXGBE_WRITE_REG(hw, IXGBE_FCTRL, reg);
 
-	fcrtl = (hw->fc.low_water << 10) | IXGBE_FCRTL_XONE;
 	/* Configure PFC Tx thresholds per TC */
 	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
 		if (!(pfc_en & (1 << i))) {
@@ -217,6 +216,7 @@ s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *hw, u8 pfc_en)
 			continue;
 		}
 
+		fcrtl = (hw->fc.low_water[i] << 10) | IXGBE_FCRTL_XONE;
 		reg = (hw->fc.high_water[i] << 10) | IXGBE_FCRTH_FCEN;
 		IXGBE_WRITE_REG(hw, IXGBE_FCRTL(i), fcrtl);
 		IXGBE_WRITE_REG(hw, IXGBE_FCRTH(i), reg);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.c
index bdb99b3b0f30..3b932fe64ab6 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.c
@@ -242,7 +242,6 @@ s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw, u8 pfc_en, u8 *prio_tc)
 			max_tc = prio_tc[i];
 	}
 
-	fcrtl = (hw->fc.low_water << 10) | IXGBE_FCRTL_XONE;
 
 	/* Configure PFC Tx thresholds per TC */
 	for (i = 0; i <= max_tc; i++) {
@@ -257,6 +256,7 @@ s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw, u8 pfc_en, u8 *prio_tc)
 
 		if (enabled) {
 			reg = (hw->fc.high_water[i] << 10) | IXGBE_FCRTH_FCEN;
+			fcrtl = (hw->fc.low_water[i] << 10) | IXGBE_FCRTL_XONE;
 			IXGBE_WRITE_REG(hw, IXGBE_FCRTL_82599(i), fcrtl);
 		} else {
 			reg = IXGBE_READ_REG(hw, IXGBE_RXPBSIZE(i)) - 32;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.h
index d5a1e3db0774..90c370230e20 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_82599.h
@@ -31,17 +31,17 @@
 
 /* DCB register definitions */
 #define IXGBE_RTTDCS_TDPAC      0x00000001 /* 0 Round Robin,
-                                            * 1 WSP - Weighted Strict Priority
-                                            */
+					    * 1 WSP - Weighted Strict Priority
+					    */
 #define IXGBE_RTTDCS_VMPAC      0x00000002 /* 0 Round Robin,
-                                            * 1 WRR - Weighted Round Robin
-                                            */
+					    * 1 WRR - Weighted Round Robin
+					    */
 #define IXGBE_RTTDCS_TDRM       0x00000010 /* Transmit Recycle Mode */
 #define IXGBE_RTTDCS_ARBDIS     0x00000040 /* DCB arbiter disable */
 #define IXGBE_RTTDCS_BDPM       0x00400000 /* Bypass Data Pipe - must clear! */
 #define IXGBE_RTTDCS_BPBFSM     0x00800000 /* Bypass PB Free Space - must
-                                             * clear!
-                                             */
+					     * clear!
+					     */
 #define IXGBE_RTTDCS_SPEED_CHG  0x80000000 /* Link speed change */
 
 /* Receive UP2TC mapping */
@@ -56,11 +56,11 @@
 #define IXGBE_RTRPT4C_LSP       0x80000000 /* LSP enable bit */
 
 #define IXGBE_RDRXCTL_MPBEN     0x00000010 /* DMA config for multiple packet
-                                            * buffers enable
-                                            */
+					    * buffers enable
+					    */
 #define IXGBE_RDRXCTL_MCEN      0x00000040 /* DMA config for multiple cores
-                                            * (RSS) enable
-                                            */
+					    * (RSS) enable
+					    */
 
 /* RTRPCS Bit Masks */
 #define IXGBE_RTRPCS_RRM        0x00000002 /* Receive Recycle Mode enable */
@@ -81,8 +81,8 @@
 
 /* RTTPCS Bit Masks */
 #define IXGBE_RTTPCS_TPPAC      0x00000020 /* 0 Round Robin,
-                                            * 1 SP - Strict Priority
-                                            */
+					    * 1 SP - Strict Priority
+					    */
 #define IXGBE_RTTPCS_ARBDIS     0x00000040 /* Arbiter disable */
 #define IXGBE_RTTPCS_TPRM       0x00000100 /* Transmit Recycle Mode enable */
 #define IXGBE_RTTPCS_ARBD_SHIFT 22
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c
index edd89a1ef27f..5172b6b12c09 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c
@@ -192,8 +192,8 @@ static void ixgbe_dcbnl_get_perm_hw_addr(struct net_device *netdev,
 }
 
 static void ixgbe_dcbnl_set_pg_tc_cfg_tx(struct net_device *netdev, int tc,
-                                         u8 prio, u8 bwg_id, u8 bw_pct,
-                                         u8 up_map)
+					 u8 prio, u8 bwg_id, u8 bw_pct,
+					 u8 up_map)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -210,7 +210,7 @@ static void ixgbe_dcbnl_set_pg_tc_cfg_tx(struct net_device *netdev, int tc,
 }
 
 static void ixgbe_dcbnl_set_pg_bwg_cfg_tx(struct net_device *netdev, int bwg_id,
-                                          u8 bw_pct)
+					  u8 bw_pct)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -218,8 +218,8 @@ static void ixgbe_dcbnl_set_pg_bwg_cfg_tx(struct net_device *netdev, int bwg_id,
 }
 
 static void ixgbe_dcbnl_set_pg_tc_cfg_rx(struct net_device *netdev, int tc,
-                                         u8 prio, u8 bwg_id, u8 bw_pct,
-                                         u8 up_map)
+					 u8 prio, u8 bwg_id, u8 bw_pct,
+					 u8 up_map)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -236,7 +236,7 @@ static void ixgbe_dcbnl_set_pg_tc_cfg_rx(struct net_device *netdev, int tc,
 }
 
 static void ixgbe_dcbnl_set_pg_bwg_cfg_rx(struct net_device *netdev, int bwg_id,
-                                          u8 bw_pct)
+					  u8 bw_pct)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -244,8 +244,8 @@ static void ixgbe_dcbnl_set_pg_bwg_cfg_rx(struct net_device *netdev, int bwg_id,
 }
 
 static void ixgbe_dcbnl_get_pg_tc_cfg_tx(struct net_device *netdev, int tc,
-                                         u8 *prio, u8 *bwg_id, u8 *bw_pct,
-                                         u8 *up_map)
+					 u8 *prio, u8 *bwg_id, u8 *bw_pct,
+					 u8 *up_map)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -256,7 +256,7 @@ static void ixgbe_dcbnl_get_pg_tc_cfg_tx(struct net_device *netdev, int tc,
 }
 
 static void ixgbe_dcbnl_get_pg_bwg_cfg_tx(struct net_device *netdev, int bwg_id,
-                                          u8 *bw_pct)
+					  u8 *bw_pct)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -264,8 +264,8 @@ static void ixgbe_dcbnl_get_pg_bwg_cfg_tx(struct net_device *netdev, int bwg_id,
 }
 
 static void ixgbe_dcbnl_get_pg_tc_cfg_rx(struct net_device *netdev, int tc,
-                                         u8 *prio, u8 *bwg_id, u8 *bw_pct,
-                                         u8 *up_map)
+					 u8 *prio, u8 *bwg_id, u8 *bw_pct,
+					 u8 *up_map)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -276,7 +276,7 @@ static void ixgbe_dcbnl_get_pg_tc_cfg_rx(struct net_device *netdev, int tc,
 }
 
 static void ixgbe_dcbnl_get_pg_bwg_cfg_rx(struct net_device *netdev, int bwg_id,
-                                          u8 *bw_pct)
+					  u8 *bw_pct)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -284,7 +284,7 @@ static void ixgbe_dcbnl_get_pg_bwg_cfg_rx(struct net_device *netdev, int bwg_id,
 }
 
 static void ixgbe_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority,
-                                    u8 setting)
+				    u8 setting)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -295,7 +295,7 @@ static void ixgbe_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority,
 }
 
 static void ixgbe_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority,
-                                    u8 *setting)
+				    u8 *setting)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c
index 472b0f450bf9..5e2c1e35e517 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c
@@ -253,8 +253,7 @@ void ixgbe_dbg_adapter_init(struct ixgbe_adapter *adapter)
  **/
 void ixgbe_dbg_adapter_exit(struct ixgbe_adapter *adapter)
 {
-	if (adapter->ixgbe_dbg_adapter)
-		debugfs_remove_recursive(adapter->ixgbe_dbg_adapter);
+	debugfs_remove_recursive(adapter->ixgbe_dbg_adapter);
 	adapter->ixgbe_dbg_adapter = NULL;
 }
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 6c55c14d082a..a452730a3278 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -141,8 +141,8 @@ static const struct ixgbe_stats ixgbe_gstrings_stats[] = {
 			 sizeof(((struct ixgbe_adapter *)0)->stats.pxofftxc)) \
 			/ sizeof(u64))
 #define IXGBE_STATS_LEN (IXGBE_GLOBAL_STATS_LEN + \
-                         IXGBE_PB_STATS_LEN + \
-                         IXGBE_QUEUE_STATS_LEN)
+			 IXGBE_PB_STATS_LEN + \
+			 IXGBE_QUEUE_STATS_LEN)
 
 static const char ixgbe_gstrings_test[][ETH_GSTRING_LEN] = {
 	"Register test  (offline)", "Eeprom test    (offline)",
@@ -152,7 +152,7 @@ static const char ixgbe_gstrings_test[][ETH_GSTRING_LEN] = {
 #define IXGBE_TEST_LEN sizeof(ixgbe_gstrings_test) / ETH_GSTRING_LEN
 
 static int ixgbe_get_settings(struct net_device *netdev,
-                              struct ethtool_cmd *ecmd)
+			      struct ethtool_cmd *ecmd)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -161,13 +161,6 @@ static int ixgbe_get_settings(struct net_device *netdev,
 	bool autoneg = false;
 	bool link_up;
 
-	/* SFP type is needed for get_link_capabilities */
-	if (hw->phy.media_type & (ixgbe_media_type_fiber |
-				  ixgbe_media_type_fiber_qsfp)) {
-		if (hw->phy.sfp_type == ixgbe_sfp_type_not_present)
-				hw->phy.ops.identify_sfp(hw);
-	}
-
 	hw->mac.ops.get_link_capabilities(hw, &supported_link, &autoneg);
 
 	/* set the supported link speeds */
@@ -303,15 +296,15 @@ static int ixgbe_get_settings(struct net_device *netdev,
 		}
 		ecmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	return 0;
 }
 
 static int ixgbe_set_settings(struct net_device *netdev,
-                              struct ethtool_cmd *ecmd)
+			      struct ethtool_cmd *ecmd)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -368,7 +361,7 @@ static int ixgbe_set_settings(struct net_device *netdev,
 }
 
 static void ixgbe_get_pauseparam(struct net_device *netdev,
-                                 struct ethtool_pauseparam *pause)
+				 struct ethtool_pauseparam *pause)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -390,7 +383,7 @@ static void ixgbe_get_pauseparam(struct net_device *netdev,
 }
 
 static int ixgbe_set_pauseparam(struct net_device *netdev,
-                                struct ethtool_pauseparam *pause)
+				struct ethtool_pauseparam *pause)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -450,7 +443,7 @@ static int ixgbe_get_regs_len(struct net_device *netdev)
 #define IXGBE_GET_STAT(_A_, _R_) _A_->stats._R_
 
 static void ixgbe_get_regs(struct net_device *netdev,
-                           struct ethtool_regs *regs, void *p)
+			   struct ethtool_regs *regs, void *p)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -812,7 +805,7 @@ static int ixgbe_get_eeprom_len(struct net_device *netdev)
 }
 
 static int ixgbe_get_eeprom(struct net_device *netdev,
-                            struct ethtool_eeprom *eeprom, u8 *bytes)
+			    struct ethtool_eeprom *eeprom, u8 *bytes)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -918,7 +911,7 @@ err:
 }
 
 static void ixgbe_get_drvinfo(struct net_device *netdev,
-                              struct ethtool_drvinfo *drvinfo)
+			      struct ethtool_drvinfo *drvinfo)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	u32 nvm_track_id;
@@ -940,7 +933,7 @@ static void ixgbe_get_drvinfo(struct net_device *netdev,
 }
 
 static void ixgbe_get_ringparam(struct net_device *netdev,
-                                struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_ring *tx_ring = adapter->tx_ring[0];
@@ -953,7 +946,7 @@ static void ixgbe_get_ringparam(struct net_device *netdev,
 }
 
 static int ixgbe_set_ringparam(struct net_device *netdev,
-                               struct ethtool_ringparam *ring)
+			       struct ethtool_ringparam *ring)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_ring *temp_ring;
@@ -1082,7 +1075,7 @@ static int ixgbe_get_sset_count(struct net_device *netdev, int sset)
 }
 
 static void ixgbe_get_ethtool_stats(struct net_device *netdev,
-                                    struct ethtool_stats *stats, u64 *data)
+				    struct ethtool_stats *stats, u64 *data)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct rtnl_link_stats64 temp;
@@ -1110,7 +1103,7 @@ static void ixgbe_get_ethtool_stats(struct net_device *netdev,
 		}
 
 		data[i] = (ixgbe_gstrings_stats[i].sizeof_stat ==
-		           sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+			   sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
 	}
 	for (j = 0; j < netdev->num_tx_queues; j++) {
 		ring = adapter->tx_ring[j];
@@ -1180,7 +1173,7 @@ static void ixgbe_get_ethtool_stats(struct net_device *netdev,
 }
 
 static void ixgbe_get_strings(struct net_device *netdev, u32 stringset,
-                              u8 *data)
+			      u8 *data)
 {
 	char *p = (char *)data;
 	int i;
@@ -1357,8 +1350,7 @@ static bool reg_pattern_test(struct ixgbe_adapter *adapter, u64 *data, int reg,
 		ixgbe_write_reg(&adapter->hw, reg, test_pattern[pat] & write);
 		val = ixgbe_read_reg(&adapter->hw, reg);
 		if (val != (test_pattern[pat] & write & mask)) {
-			e_err(drv, "pattern test reg %04X failed: got "
-			      "0x%08X expected 0x%08X\n",
+			e_err(drv, "pattern test reg %04X failed: got 0x%08X expected 0x%08X\n",
 			      reg, val, (test_pattern[pat] & write & mask));
 			*data = reg;
 			ixgbe_write_reg(&adapter->hw, reg, before);
@@ -1382,8 +1374,8 @@ static bool reg_set_and_check(struct ixgbe_adapter *adapter, u64 *data, int reg,
 	ixgbe_write_reg(&adapter->hw, reg, write & mask);
 	val = ixgbe_read_reg(&adapter->hw, reg);
 	if ((write & mask) != (val & mask)) {
-		e_err(drv, "set/check reg %04X test failed: got 0x%08X "
-		      "expected 0x%08X\n", reg, (val & mask), (write & mask));
+		e_err(drv, "set/check reg %04X test failed: got 0x%08X expected 0x%08X\n",
+		      reg, (val & mask), (write & mask));
 		*data = reg;
 		ixgbe_write_reg(&adapter->hw, reg, before);
 		return true;
@@ -1430,8 +1422,8 @@ static int ixgbe_reg_test(struct ixgbe_adapter *adapter, u64 *data)
 	ixgbe_write_reg(&adapter->hw, IXGBE_STATUS, toggle);
 	after = ixgbe_read_reg(&adapter->hw, IXGBE_STATUS) & toggle;
 	if (value != after) {
-		e_err(drv, "failed STATUS register test got: 0x%08X "
-		      "expected: 0x%08X\n", after, value);
+		e_err(drv, "failed STATUS register test got: 0x%08X expected: 0x%08X\n",
+		      after, value);
 		*data = 1;
 		return 1;
 	}
@@ -1533,10 +1525,10 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data)
 			return -1;
 		}
 	} else if (!request_irq(irq, ixgbe_test_intr, IRQF_PROBE_SHARED,
-	                        netdev->name, netdev)) {
+				netdev->name, netdev)) {
 		shared_int = false;
 	} else if (request_irq(irq, ixgbe_test_intr, IRQF_SHARED,
-	                       netdev->name, netdev)) {
+			       netdev->name, netdev)) {
 		*data = 1;
 		return -1;
 	}
@@ -1563,9 +1555,9 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data)
 			 */
 			adapter->test_icr = 0;
 			IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC,
-			                ~mask & 0x00007FFF);
+					~mask & 0x00007FFF);
 			IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS,
-			                ~mask & 0x00007FFF);
+					~mask & 0x00007FFF);
 			IXGBE_WRITE_FLUSH(&adapter->hw);
 			usleep_range(10000, 20000);
 
@@ -1587,7 +1579,7 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data)
 		IXGBE_WRITE_FLUSH(&adapter->hw);
 		usleep_range(10000, 20000);
 
-		if (!(adapter->test_icr &mask)) {
+		if (!(adapter->test_icr & mask)) {
 			*data = 4;
 			break;
 		}
@@ -1602,9 +1594,9 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data)
 			 */
 			adapter->test_icr = 0;
 			IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC,
-			                ~mask & 0x00007FFF);
+					~mask & 0x00007FFF);
 			IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS,
-			                ~mask & 0x00007FFF);
+					~mask & 0x00007FFF);
 			IXGBE_WRITE_FLUSH(&adapter->hw);
 			usleep_range(10000, 20000);
 
@@ -1964,7 +1956,7 @@ out:
 }
 
 static void ixgbe_diag_test(struct net_device *netdev,
-                            struct ethtool_test *eth_test, u64 *data)
+			    struct ethtool_test *eth_test, u64 *data)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	bool if_running = netif_running(netdev);
@@ -1987,10 +1979,7 @@ static void ixgbe_diag_test(struct net_device *netdev,
 			int i;
 			for (i = 0; i < adapter->num_vfs; i++) {
 				if (adapter->vfinfo[i].clear_to_send) {
-					netdev_warn(netdev, "%s",
-						    "offline diagnostic is not "
-						    "supported when VFs are "
-						    "present\n");
+					netdev_warn(netdev, "offline diagnostic is not supported when VFs are present\n");
 					data[0] = 1;
 					data[1] = 1;
 					data[2] = 1;
@@ -2037,8 +2026,7 @@ static void ixgbe_diag_test(struct net_device *netdev,
 		 * loopback diagnostic. */
 		if (adapter->flags & (IXGBE_FLAG_SRIOV_ENABLED |
 				      IXGBE_FLAG_VMDQ_ENABLED)) {
-			e_info(hw, "Skip MAC loopback diagnostic in VT "
-			       "mode\n");
+			e_info(hw, "Skip MAC loopback diagnostic in VT mode\n");
 			data[3] = 0;
 			goto skip_loopback;
 		}
@@ -2078,7 +2066,7 @@ skip_ol_tests:
 }
 
 static int ixgbe_wol_exclusion(struct ixgbe_adapter *adapter,
-                               struct ethtool_wolinfo *wol)
+			       struct ethtool_wolinfo *wol)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	int retval = 0;
@@ -2094,12 +2082,12 @@ static int ixgbe_wol_exclusion(struct ixgbe_adapter *adapter,
 }
 
 static void ixgbe_get_wol(struct net_device *netdev,
-                          struct ethtool_wolinfo *wol)
+			  struct ethtool_wolinfo *wol)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
 	wol->supported = WAKE_UCAST | WAKE_MCAST |
-	                 WAKE_BCAST | WAKE_MAGIC;
+			 WAKE_BCAST | WAKE_MAGIC;
 	wol->wolopts = 0;
 
 	if (ixgbe_wol_exclusion(adapter, wol) ||
@@ -2181,7 +2169,7 @@ static int ixgbe_set_phys_id(struct net_device *netdev,
 }
 
 static int ixgbe_get_coalesce(struct net_device *netdev,
-                              struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
@@ -2222,8 +2210,7 @@ static bool ixgbe_update_rsc(struct ixgbe_adapter *adapter)
 	    adapter->rx_itr_setting > IXGBE_MIN_RSC_ITR) {
 		if (!(adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)) {
 			adapter->flags2 |= IXGBE_FLAG2_RSC_ENABLED;
-			e_info(probe, "rx-usecs value high enough "
-				      "to re-enable RSC\n");
+			e_info(probe, "rx-usecs value high enough to re-enable RSC\n");
 			return true;
 		}
 	/* if interrupt rate is too high then disable RSC */
@@ -2236,7 +2223,7 @@ static bool ixgbe_update_rsc(struct ixgbe_adapter *adapter)
 }
 
 static int ixgbe_set_coalesce(struct net_device *netdev,
-                              struct ethtool_coalesce *ec)
+			      struct ethtool_coalesce *ec)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_q_vector *q_vector;
@@ -2421,9 +2408,11 @@ static int ixgbe_get_rss_hash_opts(struct ixgbe_adapter *adapter,
 	switch (cmd->flow_type) {
 	case TCP_V4_FLOW:
 		cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* fallthrough */
 	case UDP_V4_FLOW:
 		if (adapter->flags2 & IXGBE_FLAG2_RSS_FIELD_IPV4_UDP)
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* fallthrough */
 	case SCTP_V4_FLOW:
 	case AH_ESP_V4_FLOW:
 	case AH_V4_FLOW:
@@ -2433,9 +2422,11 @@ static int ixgbe_get_rss_hash_opts(struct ixgbe_adapter *adapter,
 		break;
 	case TCP_V6_FLOW:
 		cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* fallthrough */
 	case UDP_V6_FLOW:
 		if (adapter->flags2 & IXGBE_FLAG2_RSS_FIELD_IPV6_UDP)
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		/* fallthrough */
 	case SCTP_V6_FLOW:
 	case AH_ESP_V6_FLOW:
 	case AH_V6_FLOW:
@@ -2787,8 +2778,7 @@ static int ixgbe_set_rss_hash_opt(struct ixgbe_adapter *adapter,
 
 		if ((flags2 & UDP_RSS_FLAGS) &&
 		    !(adapter->flags2 & UDP_RSS_FLAGS))
-			e_warn(drv, "enabling UDP RSS: fragmented packets"
-			       " may arrive out of order to the stack above\n");
+			e_warn(drv, "enabling UDP RSS: fragmented packets may arrive out of order to the stack above\n");
 
 		adapter->flags2 = flags2;
 
@@ -3099,5 +3089,5 @@ static const struct ethtool_ops ixgbe_ethtool_ops = {
 
 void ixgbe_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &ixgbe_ethtool_ops);
+	netdev->ethtool_ops = &ixgbe_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.h
index b16cc786750d..0772b7730fce 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.h
@@ -81,9 +81,7 @@ struct ixgbe_fcoe {
 	void *extra_ddp_buffer;
 	dma_addr_t extra_ddp_buffer_dma;
 	unsigned long mode;
-#ifdef CONFIG_IXGBE_DCB
 	u8 up;
-#endif
 };
 
 #endif /* _IXGBE_FCOE_H */
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
index 2067d392cc3d..2d9451e39686 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
@@ -1113,8 +1113,8 @@ static void ixgbe_set_interrupt_capability(struct ixgbe_adapter *adapter)
 	err = pci_enable_msi(adapter->pdev);
 	if (err) {
 		netif_printk(adapter, hw, KERN_DEBUG, adapter->netdev,
-			     "Unable to allocate MSI interrupt, "
-			     "falling back to legacy.  Error: %d\n", err);
+			     "Unable to allocate MSI interrupt, falling back to legacy.  Error: %d\n",
+			     err);
 		return;
 	}
 	adapter->flags |= IXGBE_FLAG_MSI_ENABLED;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index c047c3ef8d71..f5aa3311ea28 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -301,7 +301,7 @@ static void ixgbe_remove_adapter(struct ixgbe_hw *hw)
 		ixgbe_service_event_schedule(adapter);
 }
 
-void ixgbe_check_remove(struct ixgbe_hw *hw, u32 reg)
+static void ixgbe_check_remove(struct ixgbe_hw *hw, u32 reg)
 {
 	u32 value;
 
@@ -320,6 +320,32 @@ void ixgbe_check_remove(struct ixgbe_hw *hw, u32 reg)
 		ixgbe_remove_adapter(hw);
 }
 
+/**
+ * ixgbe_read_reg - Read from device register
+ * @hw: hw specific details
+ * @reg: offset of register to read
+ *
+ * Returns : value read or IXGBE_FAILED_READ_REG if removed
+ *
+ * This function is used to read device registers. It checks for device
+ * removal by confirming any read that returns all ones by checking the
+ * status register value for all ones. This function avoids reading from
+ * the hardware if a removal was previously detected in which case it
+ * returns IXGBE_FAILED_READ_REG (all ones).
+ */
+u32 ixgbe_read_reg(struct ixgbe_hw *hw, u32 reg)
+{
+	u8 __iomem *reg_addr = ACCESS_ONCE(hw->hw_addr);
+	u32 value;
+
+	if (ixgbe_removed(reg_addr))
+		return IXGBE_FAILED_READ_REG;
+	value = readl(reg_addr + reg);
+	if (unlikely(value == IXGBE_FAILED_READ_REG))
+		ixgbe_check_remove(hw, reg);
+	return value;
+}
+
 static bool ixgbe_check_cfg_remove(struct ixgbe_hw *hw, struct pci_dev *pdev)
 {
 	u16 value;
@@ -3743,35 +3769,6 @@ static int ixgbe_vlan_rx_kill_vid(struct net_device *netdev,
 }
 
 /**
- * ixgbe_vlan_filter_disable - helper to disable hw vlan filtering
- * @adapter: driver data
- */
-static void ixgbe_vlan_filter_disable(struct ixgbe_adapter *adapter)
-{
-	struct ixgbe_hw *hw = &adapter->hw;
-	u32 vlnctrl;
-
-	vlnctrl = IXGBE_READ_REG(hw, IXGBE_VLNCTRL);
-	vlnctrl &= ~(IXGBE_VLNCTRL_VFE | IXGBE_VLNCTRL_CFIEN);
-	IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl);
-}
-
-/**
- * ixgbe_vlan_filter_enable - helper to enable hw vlan filtering
- * @adapter: driver data
- */
-static void ixgbe_vlan_filter_enable(struct ixgbe_adapter *adapter)
-{
-	struct ixgbe_hw *hw = &adapter->hw;
-	u32 vlnctrl;
-
-	vlnctrl = IXGBE_READ_REG(hw, IXGBE_VLNCTRL);
-	vlnctrl |= IXGBE_VLNCTRL_VFE;
-	vlnctrl &= ~IXGBE_VLNCTRL_CFIEN;
-	IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl);
-}
-
-/**
  * ixgbe_vlan_strip_disable - helper to disable hw vlan stripping
  * @adapter: driver data
  */
@@ -3850,6 +3847,158 @@ static void ixgbe_restore_vlan(struct ixgbe_adapter *adapter)
 }
 
 /**
+ * ixgbe_write_mc_addr_list - write multicast addresses to MTA
+ * @netdev: network interface device structure
+ *
+ * Writes multicast address list to the MTA hash table.
+ * Returns: -ENOMEM on failure
+ *                0 on no addresses written
+ *                X on writing X addresses to MTA
+ **/
+static int ixgbe_write_mc_addr_list(struct net_device *netdev)
+{
+	struct ixgbe_adapter *adapter = netdev_priv(netdev);
+	struct ixgbe_hw *hw = &adapter->hw;
+
+	if (!netif_running(netdev))
+		return 0;
+
+	if (hw->mac.ops.update_mc_addr_list)
+		hw->mac.ops.update_mc_addr_list(hw, netdev);
+	else
+		return -ENOMEM;
+
+#ifdef CONFIG_PCI_IOV
+	ixgbe_restore_vf_multicasts(adapter);
+#endif
+
+	return netdev_mc_count(netdev);
+}
+
+#ifdef CONFIG_PCI_IOV
+void ixgbe_full_sync_mac_table(struct ixgbe_adapter *adapter)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int i;
+	for (i = 0; i < hw->mac.num_rar_entries; i++) {
+		if (adapter->mac_table[i].state & IXGBE_MAC_STATE_IN_USE)
+			hw->mac.ops.set_rar(hw, i, adapter->mac_table[i].addr,
+					    adapter->mac_table[i].queue,
+					    IXGBE_RAH_AV);
+		else
+			hw->mac.ops.clear_rar(hw, i);
+
+		adapter->mac_table[i].state &= ~(IXGBE_MAC_STATE_MODIFIED);
+	}
+}
+#endif
+
+static void ixgbe_sync_mac_table(struct ixgbe_adapter *adapter)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int i;
+	for (i = 0; i < hw->mac.num_rar_entries; i++) {
+		if (adapter->mac_table[i].state & IXGBE_MAC_STATE_MODIFIED) {
+			if (adapter->mac_table[i].state &
+			    IXGBE_MAC_STATE_IN_USE)
+				hw->mac.ops.set_rar(hw, i,
+						adapter->mac_table[i].addr,
+						adapter->mac_table[i].queue,
+						IXGBE_RAH_AV);
+			else
+				hw->mac.ops.clear_rar(hw, i);
+
+			adapter->mac_table[i].state &=
+						~(IXGBE_MAC_STATE_MODIFIED);
+		}
+	}
+}
+
+static void ixgbe_flush_sw_mac_table(struct ixgbe_adapter *adapter)
+{
+	int i;
+	struct ixgbe_hw *hw = &adapter->hw;
+
+	for (i = 0; i < hw->mac.num_rar_entries; i++) {
+		adapter->mac_table[i].state |= IXGBE_MAC_STATE_MODIFIED;
+		adapter->mac_table[i].state &= ~IXGBE_MAC_STATE_IN_USE;
+		memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
+		adapter->mac_table[i].queue = 0;
+	}
+	ixgbe_sync_mac_table(adapter);
+}
+
+static int ixgbe_available_rars(struct ixgbe_adapter *adapter)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int i, count = 0;
+
+	for (i = 0; i < hw->mac.num_rar_entries; i++) {
+		if (adapter->mac_table[i].state == 0)
+			count++;
+	}
+	return count;
+}
+
+/* this function destroys the first RAR entry */
+static void ixgbe_mac_set_default_filter(struct ixgbe_adapter *adapter,
+					 u8 *addr)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+
+	memcpy(&adapter->mac_table[0].addr, addr, ETH_ALEN);
+	adapter->mac_table[0].queue = VMDQ_P(0);
+	adapter->mac_table[0].state = (IXGBE_MAC_STATE_DEFAULT |
+				       IXGBE_MAC_STATE_IN_USE);
+	hw->mac.ops.set_rar(hw, 0, adapter->mac_table[0].addr,
+			    adapter->mac_table[0].queue,
+			    IXGBE_RAH_AV);
+}
+
+int ixgbe_add_mac_filter(struct ixgbe_adapter *adapter, u8 *addr, u16 queue)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int i;
+
+	if (is_zero_ether_addr(addr))
+		return -EINVAL;
+
+	for (i = 0; i < hw->mac.num_rar_entries; i++) {
+		if (adapter->mac_table[i].state & IXGBE_MAC_STATE_IN_USE)
+			continue;
+		adapter->mac_table[i].state |= (IXGBE_MAC_STATE_MODIFIED |
+						IXGBE_MAC_STATE_IN_USE);
+		ether_addr_copy(adapter->mac_table[i].addr, addr);
+		adapter->mac_table[i].queue = queue;
+		ixgbe_sync_mac_table(adapter);
+		return i;
+	}
+	return -ENOMEM;
+}
+
+int ixgbe_del_mac_filter(struct ixgbe_adapter *adapter, u8 *addr, u16 queue)
+{
+	/* search table for addr, if found, set to 0 and sync */
+	int i;
+	struct ixgbe_hw *hw = &adapter->hw;
+
+	if (is_zero_ether_addr(addr))
+		return -EINVAL;
+
+	for (i = 0; i < hw->mac.num_rar_entries; i++) {
+		if (ether_addr_equal(addr, adapter->mac_table[i].addr) &&
+		    adapter->mac_table[i].queue == queue) {
+			adapter->mac_table[i].state |= IXGBE_MAC_STATE_MODIFIED;
+			adapter->mac_table[i].state &= ~IXGBE_MAC_STATE_IN_USE;
+			memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
+			adapter->mac_table[i].queue = 0;
+			ixgbe_sync_mac_table(adapter);
+			return 0;
+		}
+	}
+	return -ENOMEM;
+}
+/**
  * ixgbe_write_uc_addr_list - write unicast addresses to RAR table
  * @netdev: network interface device structure
  *
@@ -3858,39 +4007,23 @@ static void ixgbe_restore_vlan(struct ixgbe_adapter *adapter)
  *                0 on no addresses written
  *                X on writing X addresses to the RAR table
  **/
-static int ixgbe_write_uc_addr_list(struct net_device *netdev)
+static int ixgbe_write_uc_addr_list(struct net_device *netdev, int vfn)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
-	struct ixgbe_hw *hw = &adapter->hw;
-	unsigned int rar_entries = hw->mac.num_rar_entries - 1;
 	int count = 0;
 
-	/* In SR-IOV/VMDQ modes significantly less RAR entries are available */
-	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
-		rar_entries = IXGBE_MAX_PF_MACVLANS - 1;
-
 	/* return ENOMEM indicating insufficient memory for addresses */
-	if (netdev_uc_count(netdev) > rar_entries)
+	if (netdev_uc_count(netdev) > ixgbe_available_rars(adapter))
 		return -ENOMEM;
 
 	if (!netdev_uc_empty(netdev)) {
 		struct netdev_hw_addr *ha;
-		/* return error if we do not support writing to RAR table */
-		if (!hw->mac.ops.set_rar)
-			return -ENOMEM;
-
 		netdev_for_each_uc_addr(ha, netdev) {
-			if (!rar_entries)
-				break;
-			hw->mac.ops.set_rar(hw, rar_entries--, ha->addr,
-					    VMDQ_P(0), IXGBE_RAH_AV);
+			ixgbe_del_mac_filter(adapter, ha->addr, vfn);
+			ixgbe_add_mac_filter(adapter, ha->addr, vfn);
 			count++;
 		}
 	}
-	/* write the addresses in reverse order to avoid write combining */
-	for (; rar_entries > 0 ; rar_entries--)
-		hw->mac.ops.clear_rar(hw, rar_entries);
-
 	return count;
 }
 
@@ -3908,11 +4041,12 @@ void ixgbe_set_rx_mode(struct net_device *netdev)
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
 	u32 fctrl, vmolr = IXGBE_VMOLR_BAM | IXGBE_VMOLR_AUPE;
+	u32 vlnctrl;
 	int count;
 
 	/* Check for Promiscuous and All Multicast modes */
-
 	fctrl = IXGBE_READ_REG(hw, IXGBE_FCTRL);
+	vlnctrl = IXGBE_READ_REG(hw, IXGBE_VLNCTRL);
 
 	/* set all bits that we expect to always be set */
 	fctrl &= ~IXGBE_FCTRL_SBP; /* disable store-bad-packets */
@@ -3922,26 +4056,24 @@ void ixgbe_set_rx_mode(struct net_device *netdev)
 
 	/* clear the bits we are changing the status of */
 	fctrl &= ~(IXGBE_FCTRL_UPE | IXGBE_FCTRL_MPE);
-
+	vlnctrl &= ~(IXGBE_VLNCTRL_VFE | IXGBE_VLNCTRL_CFIEN);
 	if (netdev->flags & IFF_PROMISC) {
 		hw->addr_ctrl.user_set_promisc = true;
 		fctrl |= (IXGBE_FCTRL_UPE | IXGBE_FCTRL_MPE);
-		vmolr |= (IXGBE_VMOLR_ROPE | IXGBE_VMOLR_MPE);
+		vmolr |= IXGBE_VMOLR_MPE;
 		/* Only disable hardware filter vlans in promiscuous mode
 		 * if SR-IOV and VMDQ are disabled - otherwise ensure
 		 * that hardware VLAN filters remain enabled.
 		 */
 		if (!(adapter->flags & (IXGBE_FLAG_VMDQ_ENABLED |
 					IXGBE_FLAG_SRIOV_ENABLED)))
-			ixgbe_vlan_filter_disable(adapter);
-		else
-			ixgbe_vlan_filter_enable(adapter);
+			vlnctrl |= (IXGBE_VLNCTRL_VFE | IXGBE_VLNCTRL_CFIEN);
 	} else {
 		if (netdev->flags & IFF_ALLMULTI) {
 			fctrl |= IXGBE_FCTRL_MPE;
 			vmolr |= IXGBE_VMOLR_MPE;
 		}
-		ixgbe_vlan_filter_enable(adapter);
+		vlnctrl |= IXGBE_VLNCTRL_VFE;
 		hw->addr_ctrl.user_set_promisc = false;
 	}
 
@@ -3950,7 +4082,7 @@ void ixgbe_set_rx_mode(struct net_device *netdev)
 	 * sufficient space to store all the addresses then enable
 	 * unicast promiscuous mode
 	 */
-	count = ixgbe_write_uc_addr_list(netdev);
+	count = ixgbe_write_uc_addr_list(netdev, VMDQ_P(0));
 	if (count < 0) {
 		fctrl |= IXGBE_FCTRL_UPE;
 		vmolr |= IXGBE_VMOLR_ROPE;
@@ -3960,11 +4092,13 @@ void ixgbe_set_rx_mode(struct net_device *netdev)
 	 * then we should just turn on promiscuous mode so
 	 * that we can at least receive multicast traffic
 	 */
-	hw->mac.ops.update_mc_addr_list(hw, netdev);
-	vmolr |= IXGBE_VMOLR_ROMPE;
-
-	if (adapter->num_vfs)
-		ixgbe_restore_vf_multicasts(adapter);
+	count = ixgbe_write_mc_addr_list(netdev);
+	if (count < 0) {
+		fctrl |= IXGBE_FCTRL_MPE;
+		vmolr |= IXGBE_VMOLR_MPE;
+	} else if (count) {
+		vmolr |= IXGBE_VMOLR_ROMPE;
+	}
 
 	if (hw->mac.type != ixgbe_mac_82598EB) {
 		vmolr |= IXGBE_READ_REG(hw, IXGBE_VMOLR(VMDQ_P(0))) &
@@ -3985,6 +4119,7 @@ void ixgbe_set_rx_mode(struct net_device *netdev)
 		/* NOTE:  VLAN filtering is disabled by setting PROMISC */
 	}
 
+	IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl);
 	IXGBE_WRITE_REG(hw, IXGBE_FCTRL, fctrl);
 
 	if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
@@ -4101,8 +4236,8 @@ static int ixgbe_hpbthresh(struct ixgbe_adapter *adapter, int pb)
 	    (tc < IXGBE_FCOE_JUMBO_FRAME_SIZE) &&
 	    (pb == ixgbe_fcoe_get_tc(adapter)))
 		tc = IXGBE_FCOE_JUMBO_FRAME_SIZE;
-
 #endif
+
 	/* Calculate delay value for device */
 	switch (hw->mac.type) {
 	case ixgbe_mac_X540:
@@ -4143,7 +4278,7 @@ static int ixgbe_hpbthresh(struct ixgbe_adapter *adapter, int pb)
  * @adapter: board private structure to calculate for
  * @pb: packet buffer to calculate
  */
-static int ixgbe_lpbthresh(struct ixgbe_adapter *adapter)
+static int ixgbe_lpbthresh(struct ixgbe_adapter *adapter, int pb)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	struct net_device *dev = adapter->netdev;
@@ -4153,6 +4288,14 @@ static int ixgbe_lpbthresh(struct ixgbe_adapter *adapter)
 	/* Calculate max LAN frame size */
 	tc = dev->mtu + ETH_HLEN + ETH_FCS_LEN;
 
+#ifdef IXGBE_FCOE
+	/* FCoE traffic class uses FCOE jumbo frames */
+	if ((dev->features & NETIF_F_FCOE_MTU) &&
+	    (tc < IXGBE_FCOE_JUMBO_FRAME_SIZE) &&
+	    (pb == netdev_get_prio_tc_map(dev, adapter->fcoe.up)))
+		tc = IXGBE_FCOE_JUMBO_FRAME_SIZE;
+#endif
+
 	/* Calculate delay value for device */
 	switch (hw->mac.type) {
 	case ixgbe_mac_X540:
@@ -4179,15 +4322,17 @@ static void ixgbe_pbthresh_setup(struct ixgbe_adapter *adapter)
 	if (!num_tc)
 		num_tc = 1;
 
-	hw->fc.low_water = ixgbe_lpbthresh(adapter);
-
 	for (i = 0; i < num_tc; i++) {
 		hw->fc.high_water[i] = ixgbe_hpbthresh(adapter, i);
+		hw->fc.low_water[i] = ixgbe_lpbthresh(adapter, i);
 
 		/* Low water marks must not be larger than high water marks */
-		if (hw->fc.low_water > hw->fc.high_water[i])
-			hw->fc.low_water = 0;
+		if (hw->fc.low_water[i] > hw->fc.high_water[i])
+			hw->fc.low_water[i] = 0;
 	}
+
+	for (; i < MAX_TRAFFIC_CLASS; i++)
+		hw->fc.high_water[i] = 0;
 }
 
 static void ixgbe_configure_pb(struct ixgbe_adapter *adapter)
@@ -4249,20 +4394,10 @@ static void ixgbe_macvlan_set_rx_mode(struct net_device *dev, unsigned int pool,
 		vmolr |= IXGBE_VMOLR_ROMPE;
 		hw->mac.ops.update_mc_addr_list(hw, dev);
 	}
-	ixgbe_write_uc_addr_list(adapter->netdev);
+	ixgbe_write_uc_addr_list(adapter->netdev, pool);
 	IXGBE_WRITE_REG(hw, IXGBE_VMOLR(pool), vmolr);
 }
 
-static void ixgbe_add_mac_filter(struct ixgbe_adapter *adapter,
-				 u8 *addr, u16 pool)
-{
-	struct ixgbe_hw *hw = &adapter->hw;
-	unsigned int entry;
-
-	entry = hw->mac.num_rar_entries - pool;
-	hw->mac.ops.set_rar(hw, entry, addr, VMDQ_P(pool), IXGBE_RAH_AV);
-}
-
 static void ixgbe_fwd_psrtype(struct ixgbe_fwd_adapter *vadapter)
 {
 	struct ixgbe_adapter *adapter = vadapter->real_adapter;
@@ -4521,6 +4656,8 @@ static inline bool ixgbe_is_sfp(struct ixgbe_hw *hw)
 	case ixgbe_phy_qsfp_active_unknown:
 	case ixgbe_phy_qsfp_intel:
 	case ixgbe_phy_qsfp_unknown:
+	/* ixgbe_phy_none is set when no SFP module is present */
+	case ixgbe_phy_none:
 		return true;
 	case ixgbe_phy_nl:
 		if (hw->mac.type == ixgbe_mac_82598EB)
@@ -4742,7 +4879,9 @@ void ixgbe_up(struct ixgbe_adapter *adapter)
 void ixgbe_reset(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
+	struct net_device *netdev = adapter->netdev;
 	int err;
+	u8 old_addr[ETH_ALEN];
 
 	if (ixgbe_removed(hw->hw_addr))
 		return;
@@ -4778,9 +4917,10 @@ void ixgbe_reset(struct ixgbe_adapter *adapter)
 	}
 
 	clear_bit(__IXGBE_IN_SFP_INIT, &adapter->state);
-
-	/* reprogram the RAR[0] in case user changed it. */
-	hw->mac.ops.set_rar(hw, 0, hw->mac.addr, VMDQ_P(0), IXGBE_RAH_AV);
+	/* do not flush user set addresses */
+	memcpy(old_addr, &adapter->mac_table[0].addr, netdev->addr_len);
+	ixgbe_flush_sw_mac_table(adapter);
+	ixgbe_mac_set_default_filter(adapter, old_addr);
 
 	/* update SAN MAC vmdq pool selection */
 	if (hw->mac.san_mac_rar_index)
@@ -5026,6 +5166,10 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter)
 #endif /* CONFIG_IXGBE_DCB */
 #endif /* IXGBE_FCOE */
 
+	adapter->mac_table = kzalloc(sizeof(struct ixgbe_mac_addr) *
+				     hw->mac.num_rar_entries,
+				     GFP_ATOMIC);
+
 	/* Set MAC specific capability flags and exceptions */
 	switch (hw->mac.type) {
 	case ixgbe_mac_82598EB:
@@ -5517,6 +5661,17 @@ err_setup_tx:
 	return err;
 }
 
+static void ixgbe_close_suspend(struct ixgbe_adapter *adapter)
+{
+	ixgbe_ptp_suspend(adapter);
+
+	ixgbe_down(adapter);
+	ixgbe_free_irq(adapter);
+
+	ixgbe_free_all_tx_resources(adapter);
+	ixgbe_free_all_rx_resources(adapter);
+}
+
 /**
  * ixgbe_close - Disables a network interface
  * @netdev: network interface device structure
@@ -5534,14 +5689,10 @@ static int ixgbe_close(struct net_device *netdev)
 
 	ixgbe_ptp_stop(adapter);
 
-	ixgbe_down(adapter);
-	ixgbe_free_irq(adapter);
+	ixgbe_close_suspend(adapter);
 
 	ixgbe_fdir_filter_exit(adapter);
 
-	ixgbe_free_all_tx_resources(adapter);
-	ixgbe_free_all_rx_resources(adapter);
-
 	ixgbe_release_hw_control(adapter);
 
 	return 0;
@@ -5608,12 +5759,8 @@ static int __ixgbe_shutdown(struct pci_dev *pdev, bool *enable_wake)
 	netif_device_detach(netdev);
 
 	rtnl_lock();
-	if (netif_running(netdev)) {
-		ixgbe_down(adapter);
-		ixgbe_free_irq(adapter);
-		ixgbe_free_all_tx_resources(adapter);
-		ixgbe_free_all_rx_resources(adapter);
-	}
+	if (netif_running(netdev))
+		ixgbe_close_suspend(adapter);
 	rtnl_unlock();
 
 	ixgbe_clear_interrupt_scheme(adapter);
@@ -5945,7 +6092,7 @@ static void ixgbe_fdir_reinit_subtask(struct ixgbe_adapter *adapter)
 	if (ixgbe_reinit_fdir_tables_82599(hw) == 0) {
 		for (i = 0; i < adapter->num_tx_queues; i++)
 			set_bit(__IXGBE_TX_FDIR_INIT_DONE,
-			        &(adapter->tx_ring[i]->state));
+				&(adapter->tx_ring[i]->state));
 		/* re-enable flow director interrupts */
 		IXGBE_WRITE_REG(hw, IXGBE_EIMS, IXGBE_EIMS_FLOW_DIR);
 	} else {
@@ -7172,16 +7319,17 @@ static int ixgbe_set_mac(struct net_device *netdev, void *p)
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
 	struct sockaddr *addr = p;
+	int ret;
 
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
+	ixgbe_del_mac_filter(adapter, hw->mac.addr, VMDQ_P(0));
 	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
 	memcpy(hw->mac.addr, addr->sa_data, netdev->addr_len);
 
-	hw->mac.ops.set_rar(hw, 0, hw->mac.addr, VMDQ_P(0), IXGBE_RAH_AV);
-
-	return 0;
+	ret = ixgbe_add_mac_filter(adapter, hw->mac.addr, VMDQ_P(0));
+	return ret > 0 ? 0 : ret;
 }
 
 static int
@@ -7783,7 +7931,7 @@ static const struct net_device_ops ixgbe_netdev_ops = {
 	.ndo_do_ioctl		= ixgbe_ioctl,
 	.ndo_set_vf_mac		= ixgbe_ndo_set_vf_mac,
 	.ndo_set_vf_vlan	= ixgbe_ndo_set_vf_vlan,
-	.ndo_set_vf_tx_rate	= ixgbe_ndo_set_vf_bw,
+	.ndo_set_vf_rate	= ixgbe_ndo_set_vf_bw,
 	.ndo_set_vf_spoofchk	= ixgbe_ndo_set_vf_spoofchk,
 	.ndo_get_vf_config	= ixgbe_ndo_get_vf_config,
 	.ndo_get_stats64	= ixgbe_get_stats64,
@@ -8187,6 +8335,8 @@ skip_sriov:
 		goto err_sw_init;
 	}
 
+	ixgbe_mac_set_default_filter(adapter, hw->mac.perm_addr);
+
 	setup_timer(&adapter->service_timer, &ixgbe_service_timer,
 		    (unsigned long) adapter);
 
@@ -8242,7 +8392,7 @@ skip_sriov:
 	if (ixgbe_is_sfp(hw) && hw->phy.sfp_type != ixgbe_sfp_type_not_present)
 		e_dev_info("MAC: %d, PHY: %d, SFP+: %d, PBA No: %s\n",
 			   hw->mac.type, hw->phy.type, hw->phy.sfp_type,
-		           part_str);
+			   part_str);
 	else
 		e_dev_info("MAC: %d, PHY: %d, PBA No: %s\n",
 			   hw->mac.type, hw->phy.type, part_str);
@@ -8304,8 +8454,8 @@ skip_sriov:
 
 	ixgbe_dbg_adapter_init(adapter);
 
-	/* Need link setup for MNG FW, else wait for IXGBE_UP */
-	if (ixgbe_mng_enabled(hw) && hw->mac.ops.setup_link)
+	/* setup link for SFP devices with MNG FW, else wait for IXGBE_UP */
+	if (ixgbe_mng_enabled(hw) && ixgbe_is_sfp(hw) && hw->mac.ops.setup_link)
 		hw->mac.ops.setup_link(hw,
 			IXGBE_LINK_SPEED_10GB_FULL | IXGBE_LINK_SPEED_1GB_FULL,
 			true);
@@ -8319,6 +8469,7 @@ err_sw_init:
 	ixgbe_disable_sriov(adapter);
 	adapter->flags2 &= ~IXGBE_FLAG2_SEARCH_FOR_SFP;
 	iounmap(adapter->io_addr);
+	kfree(adapter->mac_table);
 err_ioremap:
 	free_netdev(netdev);
 err_alloc_etherdev:
@@ -8392,6 +8543,7 @@ static void ixgbe_remove(struct pci_dev *pdev)
 
 	e_dev_info("complete\n");
 
+	kfree(adapter->mac_table);
 	free_netdev(netdev);
 
 	pci_disable_pcie_error_reporting(pdev);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.c
index f5c6af2b891b..1918e0abf734 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.c
@@ -223,7 +223,7 @@ out:
  *  received an ack to that message within delay * timeout period
  **/
 static s32 ixgbe_write_posted_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size,
-                           u16 mbx_id)
+			   u16 mbx_id)
 {
 	struct ixgbe_mbx_info *mbx = &hw->mbx;
 	s32 ret_val = IXGBE_ERR_MBX;
@@ -269,7 +269,7 @@ static s32 ixgbe_check_for_msg_pf(struct ixgbe_hw *hw, u16 vf_number)
 	u32 vf_bit = vf_number % 16;
 
 	if (!ixgbe_check_for_bit_pf(hw, IXGBE_MBVFICR_VFREQ_VF1 << vf_bit,
-	                            index)) {
+				    index)) {
 		ret_val = 0;
 		hw->mbx.stats.reqs++;
 	}
@@ -291,7 +291,7 @@ static s32 ixgbe_check_for_ack_pf(struct ixgbe_hw *hw, u16 vf_number)
 	u32 vf_bit = vf_number % 16;
 
 	if (!ixgbe_check_for_bit_pf(hw, IXGBE_MBVFICR_VFACK_VF1 << vf_bit,
-	                            index)) {
+				    index)) {
 		ret_val = 0;
 		hw->mbx.stats.acks++;
 	}
@@ -366,7 +366,7 @@ static s32 ixgbe_obtain_mbx_lock_pf(struct ixgbe_hw *hw, u16 vf_number)
  *  returns SUCCESS if it successfully copied message into the buffer
  **/
 static s32 ixgbe_write_mbx_pf(struct ixgbe_hw *hw, u32 *msg, u16 size,
-                              u16 vf_number)
+			      u16 vf_number)
 {
 	s32 ret_val;
 	u16 i;
@@ -407,7 +407,7 @@ out_no_write:
  *  a message due to a VF request so no polling for message is needed.
  **/
 static s32 ixgbe_read_mbx_pf(struct ixgbe_hw *hw, u32 *msg, u16 size,
-                             u16 vf_number)
+			     u16 vf_number)
 {
 	s32 ret_val;
 	u16 i;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.h
index a9b9ad69ed0e..a5cb755de3a9 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_mbx.h
@@ -54,11 +54,11 @@
  * Message ACK's are the value or'd with 0xF0000000
  */
 #define IXGBE_VT_MSGTYPE_ACK      0x80000000  /* Messages below or'd with
-                                               * this are the ACK */
+					       * this are the ACK */
 #define IXGBE_VT_MSGTYPE_NACK     0x40000000  /* Messages below or'd with
-                                               * this are the NACK */
+					       * this are the NACK */
 #define IXGBE_VT_MSGTYPE_CTS      0x20000000  /* Indicates that VF is still
-                                                 clear to send requests */
+						 clear to send requests */
 #define IXGBE_VT_MSGINFO_SHIFT    16
 /* bits 23:16 are used for exra info for certain messages */
 #define IXGBE_VT_MSGINFO_MASK     (0xFF << IXGBE_VT_MSGINFO_SHIFT)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
index a76af8e28a04..ff68b7a9deff 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
@@ -67,7 +67,7 @@ s32 ixgbe_identify_phy_generic(struct ixgbe_hw *hw)
 			if (mdio45_probe(&hw->phy.mdio, phy_addr) == 0) {
 				ixgbe_get_phy_id(hw);
 				hw->phy.type =
-				        ixgbe_get_phy_type_from_id(hw->phy.id);
+					ixgbe_get_phy_type_from_id(hw->phy.id);
 
 				if (hw->phy.type == ixgbe_phy_unknown) {
 					hw->phy.ops.read_reg(hw,
@@ -136,12 +136,12 @@ static s32 ixgbe_get_phy_id(struct ixgbe_hw *hw)
 	u16 phy_id_low = 0;
 
 	status = hw->phy.ops.read_reg(hw, MDIO_DEVID1, MDIO_MMD_PMAPMD,
-	                              &phy_id_high);
+				      &phy_id_high);
 
 	if (status == 0) {
 		hw->phy.id = (u32)(phy_id_high << 16);
 		status = hw->phy.ops.read_reg(hw, MDIO_DEVID2, MDIO_MMD_PMAPMD,
-		                              &phy_id_low);
+					      &phy_id_low);
 		hw->phy.id |= (u32)(phy_id_low & IXGBE_PHY_REVISION_MASK);
 		hw->phy.revision = (u32)(phy_id_low & ~IXGBE_PHY_REVISION_MASK);
 	}
@@ -318,7 +318,7 @@ s32 ixgbe_read_phy_reg_mdi(struct ixgbe_hw *hw, u32 reg_addr, u32 device_type,
  *  @phy_data: Pointer to read data from PHY register
  **/
 s32 ixgbe_read_phy_reg_generic(struct ixgbe_hw *hw, u32 reg_addr,
-                               u32 device_type, u16 *phy_data)
+			       u32 device_type, u16 *phy_data)
 {
 	s32 status;
 	u16 gssr;
@@ -421,7 +421,7 @@ s32 ixgbe_write_phy_reg_mdi(struct ixgbe_hw *hw, u32 reg_addr,
  *  @phy_data: Data to write to the PHY register
  **/
 s32 ixgbe_write_phy_reg_generic(struct ixgbe_hw *hw, u32 reg_addr,
-                                u32 device_type, u16 phy_data)
+				u32 device_type, u16 phy_data)
 {
 	s32 status;
 	u16 gssr;
@@ -548,8 +548,8 @@ s32 ixgbe_setup_phy_link_generic(struct ixgbe_hw *hw)
  *  @speed: new link speed
  **/
 s32 ixgbe_setup_phy_link_speed_generic(struct ixgbe_hw *hw,
-                                       ixgbe_link_speed speed,
-                                       bool autoneg_wait_to_complete)
+				       ixgbe_link_speed speed,
+				       bool autoneg_wait_to_complete)
 {
 
 	/*
@@ -582,8 +582,8 @@ s32 ixgbe_setup_phy_link_speed_generic(struct ixgbe_hw *hw,
  * Determines the link capabilities by reading the AUTOC register.
  */
 s32 ixgbe_get_copper_link_capabilities_generic(struct ixgbe_hw *hw,
-                                               ixgbe_link_speed *speed,
-                                               bool *autoneg)
+					       ixgbe_link_speed *speed,
+					       bool *autoneg)
 {
 	s32 status = IXGBE_ERR_LINK_SETUP;
 	u16 speed_ability;
@@ -592,7 +592,7 @@ s32 ixgbe_get_copper_link_capabilities_generic(struct ixgbe_hw *hw,
 	*autoneg = true;
 
 	status = hw->phy.ops.read_reg(hw, MDIO_SPEED, MDIO_MMD_PMAPMD,
-	                              &speed_ability);
+				      &speed_ability);
 
 	if (status == 0) {
 		if (speed_ability & MDIO_SPEED_10G)
@@ -806,11 +806,11 @@ s32 ixgbe_reset_phy_nl(struct ixgbe_hw *hw)
 
 	/* reset the PHY and poll for completion */
 	hw->phy.ops.write_reg(hw, MDIO_CTRL1, MDIO_MMD_PHYXS,
-	                      (phy_data | MDIO_CTRL1_RESET));
+			      (phy_data | MDIO_CTRL1_RESET));
 
 	for (i = 0; i < 100; i++) {
 		hw->phy.ops.read_reg(hw, MDIO_CTRL1, MDIO_MMD_PHYXS,
-		                     &phy_data);
+				     &phy_data);
 		if ((phy_data & MDIO_CTRL1_RESET) == 0)
 			break;
 		usleep_range(10000, 20000);
@@ -824,7 +824,7 @@ s32 ixgbe_reset_phy_nl(struct ixgbe_hw *hw)
 
 	/* Get init offsets */
 	ret_val = ixgbe_get_sfp_init_sequence_offsets(hw, &list_offset,
-	                                              &data_offset);
+						      &data_offset);
 	if (ret_val != 0)
 		goto out;
 
@@ -838,7 +838,7 @@ s32 ixgbe_reset_phy_nl(struct ixgbe_hw *hw)
 		if (ret_val)
 			goto err_eeprom;
 		control = (eword & IXGBE_CONTROL_MASK_NL) >>
-		           IXGBE_CONTROL_SHIFT_NL;
+			   IXGBE_CONTROL_SHIFT_NL;
 		edata = eword & IXGBE_DATA_MASK_NL;
 		switch (control) {
 		case IXGBE_DELAY_NL:
@@ -859,7 +859,7 @@ s32 ixgbe_reset_phy_nl(struct ixgbe_hw *hw)
 				if (ret_val)
 					goto err_eeprom;
 				hw->phy.ops.write_reg(hw, phy_offset,
-				                      MDIO_MMD_PMAPMD, eword);
+						      MDIO_MMD_PMAPMD, eword);
 				hw_dbg(hw, "Wrote %4.4x to %4.4x\n", eword,
 				       phy_offset);
 				data_offset++;
@@ -1010,10 +1010,10 @@ s32 ixgbe_identify_sfp_module_generic(struct ixgbe_hw *hw)
 			if (cable_tech & IXGBE_SFF_DA_PASSIVE_CABLE) {
 				if (hw->bus.lan_id == 0)
 					hw->phy.sfp_type =
-					             ixgbe_sfp_type_da_cu_core0;
+						     ixgbe_sfp_type_da_cu_core0;
 				else
 					hw->phy.sfp_type =
-					             ixgbe_sfp_type_da_cu_core1;
+						     ixgbe_sfp_type_da_cu_core1;
 			} else if (cable_tech & IXGBE_SFF_DA_ACTIVE_CABLE) {
 				hw->phy.ops.read_i2c_eeprom(
 						hw, IXGBE_SFF_CABLE_SPEC_COMP,
@@ -1035,10 +1035,10 @@ s32 ixgbe_identify_sfp_module_generic(struct ixgbe_hw *hw)
 				    IXGBE_SFF_10GBASELR_CAPABLE)) {
 				if (hw->bus.lan_id == 0)
 					hw->phy.sfp_type =
-					              ixgbe_sfp_type_srlr_core0;
+						      ixgbe_sfp_type_srlr_core0;
 				else
 					hw->phy.sfp_type =
-					              ixgbe_sfp_type_srlr_core1;
+						      ixgbe_sfp_type_srlr_core1;
 			} else if (comp_codes_1g & IXGBE_SFF_1GBASET_CAPABLE) {
 				if (hw->bus.lan_id == 0)
 					hw->phy.sfp_type =
@@ -1087,15 +1087,15 @@ s32 ixgbe_identify_sfp_module_generic(struct ixgbe_hw *hw)
 				goto err_read_i2c_eeprom;
 
 			status = hw->phy.ops.read_i2c_eeprom(hw,
-			                            IXGBE_SFF_VENDOR_OUI_BYTE1,
-			                            &oui_bytes[1]);
+						    IXGBE_SFF_VENDOR_OUI_BYTE1,
+						    &oui_bytes[1]);
 
 			if (status != 0)
 				goto err_read_i2c_eeprom;
 
 			status = hw->phy.ops.read_i2c_eeprom(hw,
-			                            IXGBE_SFF_VENDOR_OUI_BYTE2,
-			                            &oui_bytes[2]);
+						    IXGBE_SFF_VENDOR_OUI_BYTE2,
+						    &oui_bytes[2]);
 
 			if (status != 0)
 				goto err_read_i2c_eeprom;
@@ -1403,8 +1403,8 @@ err_read_i2c_eeprom:
  *  so it returns the offsets to the phy init sequence block.
  **/
 s32 ixgbe_get_sfp_init_sequence_offsets(struct ixgbe_hw *hw,
-                                        u16 *list_offset,
-                                        u16 *data_offset)
+					u16 *list_offset,
+					u16 *data_offset)
 {
 	u16 sfp_id;
 	u16 sfp_type = hw->phy.sfp_type;
@@ -1493,11 +1493,11 @@ err_phy:
  *  Performs byte read operation to SFP module's EEPROM over I2C interface.
  **/
 s32 ixgbe_read_i2c_eeprom_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                  u8 *eeprom_data)
+				  u8 *eeprom_data)
 {
 	return hw->phy.ops.read_i2c_byte(hw, byte_offset,
-	                                 IXGBE_I2C_EEPROM_DEV_ADDR,
-	                                 eeprom_data);
+					 IXGBE_I2C_EEPROM_DEV_ADDR,
+					 eeprom_data);
 }
 
 /**
@@ -1525,11 +1525,11 @@ s32 ixgbe_read_i2c_sff8472_generic(struct ixgbe_hw *hw, u8 byte_offset,
  *  Performs byte write operation to SFP module's EEPROM over I2C interface.
  **/
 s32 ixgbe_write_i2c_eeprom_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                   u8 eeprom_data)
+				   u8 eeprom_data)
 {
 	return hw->phy.ops.write_i2c_byte(hw, byte_offset,
-	                                  IXGBE_I2C_EEPROM_DEV_ADDR,
-	                                  eeprom_data);
+					  IXGBE_I2C_EEPROM_DEV_ADDR,
+					  eeprom_data);
 }
 
 /**
@@ -1542,7 +1542,7 @@ s32 ixgbe_write_i2c_eeprom_generic(struct ixgbe_hw *hw, u8 byte_offset,
  *  a specified device address.
  **/
 s32 ixgbe_read_i2c_byte_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                u8 dev_addr, u8 *data)
+				u8 dev_addr, u8 *data)
 {
 	s32 status = 0;
 	u32 max_retry = 10;
@@ -1631,7 +1631,7 @@ read_byte_out:
  *  a specified device address.
  **/
 s32 ixgbe_write_i2c_byte_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                 u8 dev_addr, u8 data)
+				 u8 dev_addr, u8 data)
 {
 	s32 status = 0;
 	u32 max_retry = 1;
@@ -2046,7 +2046,7 @@ s32 ixgbe_tn_check_overtemp(struct ixgbe_hw *hw)
 
 	/* Check that the LASI temp alarm status was triggered */
 	hw->phy.ops.read_reg(hw, IXGBE_TN_LASI_STATUS_REG,
-	                     MDIO_MMD_PMAPMD, &phy_data);
+			     MDIO_MMD_PMAPMD, &phy_data);
 
 	if (!(phy_data & IXGBE_TN_LASI_STATUS_TEMP_ALARM))
 		goto out;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.h
index 0bb047f751c2..54071ed17e3b 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.h
@@ -114,47 +114,47 @@ s32 ixgbe_init_phy_ops_generic(struct ixgbe_hw *hw);
 s32 ixgbe_identify_phy_generic(struct ixgbe_hw *hw);
 s32 ixgbe_reset_phy_generic(struct ixgbe_hw *hw);
 s32 ixgbe_read_phy_reg_generic(struct ixgbe_hw *hw, u32 reg_addr,
-                               u32 device_type, u16 *phy_data);
+			       u32 device_type, u16 *phy_data);
 s32 ixgbe_write_phy_reg_generic(struct ixgbe_hw *hw, u32 reg_addr,
-                                u32 device_type, u16 phy_data);
+				u32 device_type, u16 phy_data);
 s32 ixgbe_read_phy_reg_mdi(struct ixgbe_hw *hw, u32 reg_addr,
 			   u32 device_type, u16 *phy_data);
 s32 ixgbe_write_phy_reg_mdi(struct ixgbe_hw *hw, u32 reg_addr,
 			    u32 device_type, u16 phy_data);
 s32 ixgbe_setup_phy_link_generic(struct ixgbe_hw *hw);
 s32 ixgbe_setup_phy_link_speed_generic(struct ixgbe_hw *hw,
-                                       ixgbe_link_speed speed,
-                                       bool autoneg_wait_to_complete);
+				       ixgbe_link_speed speed,
+				       bool autoneg_wait_to_complete);
 s32 ixgbe_get_copper_link_capabilities_generic(struct ixgbe_hw *hw,
-                                               ixgbe_link_speed *speed,
-                                               bool *autoneg);
+					       ixgbe_link_speed *speed,
+					       bool *autoneg);
 bool ixgbe_check_reset_blocked(struct ixgbe_hw *hw);
 
 /* PHY specific */
 s32 ixgbe_check_phy_link_tnx(struct ixgbe_hw *hw,
-                             ixgbe_link_speed *speed,
-                             bool *link_up);
+			     ixgbe_link_speed *speed,
+			     bool *link_up);
 s32 ixgbe_setup_phy_link_tnx(struct ixgbe_hw *hw);
 s32 ixgbe_get_phy_firmware_version_tnx(struct ixgbe_hw *hw,
-                                       u16 *firmware_version);
+				       u16 *firmware_version);
 s32 ixgbe_get_phy_firmware_version_generic(struct ixgbe_hw *hw,
-                                           u16 *firmware_version);
+					   u16 *firmware_version);
 
 s32 ixgbe_reset_phy_nl(struct ixgbe_hw *hw);
 s32 ixgbe_identify_module_generic(struct ixgbe_hw *hw);
 s32 ixgbe_identify_sfp_module_generic(struct ixgbe_hw *hw);
 s32 ixgbe_get_sfp_init_sequence_offsets(struct ixgbe_hw *hw,
-                                        u16 *list_offset,
-                                        u16 *data_offset);
+					u16 *list_offset,
+					u16 *data_offset);
 s32 ixgbe_tn_check_overtemp(struct ixgbe_hw *hw);
 s32 ixgbe_read_i2c_byte_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                u8 dev_addr, u8 *data);
+				u8 dev_addr, u8 *data);
 s32 ixgbe_write_i2c_byte_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                 u8 dev_addr, u8 data);
+				 u8 dev_addr, u8 data);
 s32 ixgbe_read_i2c_eeprom_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                  u8 *eeprom_data);
+				  u8 *eeprom_data);
 s32 ixgbe_read_i2c_sff8472_generic(struct ixgbe_hw *hw, u8 byte_offset,
 				   u8 *sff8472_data);
 s32 ixgbe_write_i2c_eeprom_generic(struct ixgbe_hw *hw, u8 byte_offset,
-                                   u8 eeprom_data);
+				   u8 eeprom_data);
 #endif /* _IXGBE_PHY_H_ */
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c
index 8902ae683457..68f87ecb8a76 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c
@@ -26,7 +26,6 @@
 
 *******************************************************************************/
 #include "ixgbe.h"
-#include <linux/export.h>
 #include <linux/ptp_classify.h>
 
 /*
@@ -334,7 +333,7 @@ static int ixgbe_ptp_settime(struct ptp_clock_info *ptp,
 }
 
 /**
- * ixgbe_ptp_enable
+ * ixgbe_ptp_feature_enable
  * @ptp: the ptp clock structure
  * @rq: the requested feature to change
  * @on: whether to enable or disable the feature
@@ -342,8 +341,8 @@ static int ixgbe_ptp_settime(struct ptp_clock_info *ptp,
  * enable (or disable) ancillary features of the phc subsystem.
  * our driver only supports the PPS feature on the X540
  */
-static int ixgbe_ptp_enable(struct ptp_clock_info *ptp,
-			    struct ptp_clock_request *rq, int on)
+static int ixgbe_ptp_feature_enable(struct ptp_clock_info *ptp,
+				    struct ptp_clock_request *rq, int on)
 {
 	struct ixgbe_adapter *adapter =
 		container_of(ptp, struct ixgbe_adapter, ptp_caps);
@@ -570,9 +569,9 @@ int ixgbe_ptp_get_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
 }
 
 /**
- * ixgbe_ptp_set_ts_config - control hardware time stamping
- * @adapter: pointer to adapter struct
- * @ifreq: ioctl data
+ * ixgbe_ptp_set_timestamp_mode - setup the hardware for the requested mode
+ * @adapter: the private ixgbe adapter structure
+ * @config: the hwtstamp configuration requested
  *
  * Outgoing time stamping can be enabled and disabled. Play nice and
  * disable it when requested, although it shouldn't cause any overhead
@@ -590,25 +589,25 @@ int ixgbe_ptp_get_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
  * packets, regardless of the type specified in the register, only use V2
  * Event mode. This more accurately tells the user what the hardware is going
  * to do anyways.
+ *
+ * Note: this may modify the hwtstamp configuration towards a more general
+ * mode, if required to support the specifically requested mode.
  */
-int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
+static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
+				 struct hwtstamp_config *config)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	struct hwtstamp_config config;
 	u32 tsync_tx_ctl = IXGBE_TSYNCTXCTL_ENABLED;
 	u32 tsync_rx_ctl = IXGBE_TSYNCRXCTL_ENABLED;
 	u32 tsync_rx_mtrl = PTP_EV_PORT << 16;
 	bool is_l2 = false;
 	u32 regval;
 
-	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
-		return -EFAULT;
-
 	/* reserved for future extensions */
-	if (config.flags)
+	if (config->flags)
 		return -EINVAL;
 
-	switch (config.tx_type) {
+	switch (config->tx_type) {
 	case HWTSTAMP_TX_OFF:
 		tsync_tx_ctl = 0;
 	case HWTSTAMP_TX_ON:
@@ -617,7 +616,7 @@ int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
 		return -ERANGE;
 	}
 
-	switch (config.rx_filter) {
+	switch (config->rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
 		tsync_rx_ctl = 0;
 		tsync_rx_mtrl = 0;
@@ -641,7 +640,7 @@ int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
 	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
 		tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_EVENT_V2;
 		is_l2 = true;
-		config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
 		break;
 	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
 	case HWTSTAMP_FILTER_ALL:
@@ -652,7 +651,7 @@ int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
 		 * Delay_Req messages and hardware does not support
 		 * timestamping all packets => return error
 		 */
-		config.rx_filter = HWTSTAMP_FILTER_NONE;
+		config->rx_filter = HWTSTAMP_FILTER_NONE;
 		return -ERANGE;
 	}
 
@@ -671,7 +670,6 @@ int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
 	else
 		IXGBE_WRITE_REG(hw, IXGBE_ETQF(IXGBE_ETQF_FILTER_1588), 0);
 
-
 	/* enable/disable TX */
 	regval = IXGBE_READ_REG(hw, IXGBE_TSYNCTXCTL);
 	regval &= ~IXGBE_TSYNCTXCTL_ENABLED;
@@ -693,6 +691,29 @@ int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
 	regval = IXGBE_READ_REG(hw, IXGBE_TXSTMPH);
 	regval = IXGBE_READ_REG(hw, IXGBE_RXSTMPH);
 
+	return 0;
+}
+
+/**
+ * ixgbe_ptp_set_ts_config - user entry point for timestamp mode
+ * @adapter: pointer to adapter struct
+ * @ifreq: ioctl data
+ *
+ * Set hardware to requested mode. If unsupported, return an error with no
+ * changes. Otherwise, store the mode for future reference.
+ */
+int ixgbe_ptp_set_ts_config(struct ixgbe_adapter *adapter, struct ifreq *ifr)
+{
+	struct hwtstamp_config config;
+	int err;
+
+	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	err = ixgbe_ptp_set_timestamp_mode(adapter, &config);
+	if (err)
+		return err;
+
 	/* save these settings for future reference */
 	memcpy(&adapter->tstamp_config, &config,
 	       sizeof(adapter->tstamp_config));
@@ -790,9 +811,13 @@ void ixgbe_ptp_start_cyclecounter(struct ixgbe_adapter *adapter)
  * ixgbe_ptp_reset
  * @adapter: the ixgbe private board structure
  *
- * When the MAC resets, all timesync features are reset. This function should be
- * called to re-enable the PTP clock structure. It will re-init the timecounter
- * structure based on the kernel time as well as setup the cycle counter data.
+ * When the MAC resets, all the hardware bits for timesync are reset. This
+ * function is used to re-enable the device for PTP based on current settings.
+ * We do lose the current clock time, so just reset the cyclecounter to the
+ * system real clock time.
+ *
+ * This function will maintain hwtstamp_config settings, and resets the SDP
+ * output if it was enabled.
  */
 void ixgbe_ptp_reset(struct ixgbe_adapter *adapter)
 {
@@ -804,8 +829,8 @@ void ixgbe_ptp_reset(struct ixgbe_adapter *adapter)
 	IXGBE_WRITE_REG(hw, IXGBE_SYSTIMH, 0x00000000);
 	IXGBE_WRITE_FLUSH(hw);
 
-	/* Reset the saved tstamp_config */
-	memset(&adapter->tstamp_config, 0, sizeof(adapter->tstamp_config));
+	/* reset the hardware timestamping mode */
+	ixgbe_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config);
 
 	ixgbe_ptp_start_cyclecounter(adapter);
 
@@ -825,16 +850,23 @@ void ixgbe_ptp_reset(struct ixgbe_adapter *adapter)
 }
 
 /**
- * ixgbe_ptp_init
+ * ixgbe_ptp_create_clock
  * @adapter: the ixgbe private adapter structure
  *
- * This function performs the required steps for enabling ptp
- * support. If ptp support has already been loaded it simply calls the
- * cyclecounter init routine and exits.
+ * This function performs setup of the user entry point function table and
+ * initializes the PTP clock device, which is used to access the clock-like
+ * features of the PTP core. It will be called by ixgbe_ptp_init, only if
+ * there isn't already a clock device (such as after a suspend/resume cycle,
+ * where the clock device wasn't destroyed).
  */
-void ixgbe_ptp_init(struct ixgbe_adapter *adapter)
+static int ixgbe_ptp_create_clock(struct ixgbe_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
+	long err;
+
+	/* do nothing if we already have a clock device */
+	if (!IS_ERR_OR_NULL(adapter->ptp_clock))
+		return 0;
 
 	switch (adapter->hw.mac.type) {
 	case ixgbe_mac_X540:
@@ -851,7 +883,7 @@ void ixgbe_ptp_init(struct ixgbe_adapter *adapter)
 		adapter->ptp_caps.adjtime = ixgbe_ptp_adjtime;
 		adapter->ptp_caps.gettime = ixgbe_ptp_gettime;
 		adapter->ptp_caps.settime = ixgbe_ptp_settime;
-		adapter->ptp_caps.enable = ixgbe_ptp_enable;
+		adapter->ptp_caps.enable = ixgbe_ptp_feature_enable;
 		break;
 	case ixgbe_mac_82599EB:
 		snprintf(adapter->ptp_caps.name,
@@ -867,24 +899,57 @@ void ixgbe_ptp_init(struct ixgbe_adapter *adapter)
 		adapter->ptp_caps.adjtime = ixgbe_ptp_adjtime;
 		adapter->ptp_caps.gettime = ixgbe_ptp_gettime;
 		adapter->ptp_caps.settime = ixgbe_ptp_settime;
-		adapter->ptp_caps.enable = ixgbe_ptp_enable;
+		adapter->ptp_caps.enable = ixgbe_ptp_feature_enable;
 		break;
 	default:
 		adapter->ptp_clock = NULL;
-		return;
+		return -EOPNOTSUPP;
 	}
 
-	spin_lock_init(&adapter->tmreg_lock);
-	INIT_WORK(&adapter->ptp_tx_work, ixgbe_ptp_tx_hwtstamp_work);
-
 	adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps,
 						&adapter->pdev->dev);
 	if (IS_ERR(adapter->ptp_clock)) {
+		err = PTR_ERR(adapter->ptp_clock);
 		adapter->ptp_clock = NULL;
 		e_dev_err("ptp_clock_register failed\n");
+		return err;
 	} else
 		e_dev_info("registered PHC device on %s\n", netdev->name);
 
+	/* set default timestamp mode to disabled here. We do this in
+	 * create_clock instead of init, because we don't want to override the
+	 * previous settings during a resume cycle.
+	 */
+	adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
+	adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF;
+
+	return 0;
+}
+
+/**
+ * ixgbe_ptp_init
+ * @adapter: the ixgbe private adapter structure
+ *
+ * This function performs the required steps for enabling PTP
+ * support. If PTP support has already been loaded it simply calls the
+ * cyclecounter init routine and exits.
+ */
+void ixgbe_ptp_init(struct ixgbe_adapter *adapter)
+{
+	/* initialize the spin lock first since we can't control when a user
+	 * will call the entry functions once we have initialized the clock
+	 * device
+	 */
+	spin_lock_init(&adapter->tmreg_lock);
+
+	/* obtain a PTP device, or re-use an existing device */
+	if (ixgbe_ptp_create_clock(adapter))
+		return;
+
+	/* we have a clock so we can initialize work now */
+	INIT_WORK(&adapter->ptp_tx_work, ixgbe_ptp_tx_hwtstamp_work);
+
+	/* reset the PTP related hardware bits */
 	ixgbe_ptp_reset(adapter);
 
 	/* enter the IXGBE_PTP_RUNNING state */
@@ -894,28 +959,45 @@ void ixgbe_ptp_init(struct ixgbe_adapter *adapter)
 }
 
 /**
- * ixgbe_ptp_stop - disable ptp device and stop the overflow check
- * @adapter: pointer to adapter struct
+ * ixgbe_ptp_suspend - stop PTP work items
+ * @ adapter: pointer to adapter struct
  *
- * this function stops the ptp support, and cancels the delayed work.
+ * this function suspends PTP activity, and prevents more PTP work from being
+ * generated, but does not destroy the PTP clock device.
  */
-void ixgbe_ptp_stop(struct ixgbe_adapter *adapter)
+void ixgbe_ptp_suspend(struct ixgbe_adapter *adapter)
 {
 	/* Leave the IXGBE_PTP_RUNNING state. */
 	if (!test_and_clear_bit(__IXGBE_PTP_RUNNING, &adapter->state))
 		return;
 
-	/* stop the PPS signal */
-	adapter->flags2 &= ~IXGBE_FLAG2_PTP_PPS_ENABLED;
-	ixgbe_ptp_setup_sdp(adapter);
+	/* since this might be called in suspend, we don't clear the state,
+	 * but simply reset the auxiliary PPS signal control register
+	 */
+	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TSAUXC, 0x0);
 
+	/* ensure that we cancel any pending PTP Tx work item in progress */
 	cancel_work_sync(&adapter->ptp_tx_work);
 	if (adapter->ptp_tx_skb) {
 		dev_kfree_skb_any(adapter->ptp_tx_skb);
 		adapter->ptp_tx_skb = NULL;
 		clear_bit_unlock(__IXGBE_PTP_TX_IN_PROGRESS, &adapter->state);
 	}
+}
+
+/**
+ * ixgbe_ptp_stop - close the PTP device
+ * @adapter: pointer to adapter struct
+ *
+ * completely destroy the PTP device, should only be called when the device is
+ * being fully closed.
+ */
+void ixgbe_ptp_stop(struct ixgbe_adapter *adapter)
+{
+	/* first, suspend PTP activity */
+	ixgbe_ptp_suspend(adapter);
 
+	/* disable the PTP clock device */
 	if (adapter->ptp_clock) {
 		ptp_clock_unregister(adapter->ptp_clock);
 		adapter->ptp_clock = NULL;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index e6c68d396c99..16b3a1cd9db6 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -72,8 +72,6 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 		for (i = 0; i < num_vf_macvlans; i++) {
 			mv_list->vf = -1;
 			mv_list->free = true;
-			mv_list->rar_entry = hw->mac.num_rar_entries -
-				(i + adapter->num_vfs + 1);
 			list_add(&mv_list->l, &adapter->vf_mvs.l);
 			mv_list++;
 		}
@@ -327,6 +325,7 @@ static int ixgbe_set_vf_multicasts(struct ixgbe_adapter *adapter,
 	u32 vector_bit;
 	u32 vector_reg;
 	u32 mta_reg;
+	u32 vmolr = IXGBE_READ_REG(hw, IXGBE_VMOLR(vf));
 
 	/* only so many hash values supported */
 	entries = min(entries, IXGBE_MAX_VF_MC_ENTRIES);
@@ -353,25 +352,13 @@ static int ixgbe_set_vf_multicasts(struct ixgbe_adapter *adapter,
 		mta_reg |= (1 << vector_bit);
 		IXGBE_WRITE_REG(hw, IXGBE_MTA(vector_reg), mta_reg);
 	}
+	vmolr |= IXGBE_VMOLR_ROMPE;
+	IXGBE_WRITE_REG(hw, IXGBE_VMOLR(vf), vmolr);
 
 	return 0;
 }
 
-static void ixgbe_restore_vf_macvlans(struct ixgbe_adapter *adapter)
-{
-	struct ixgbe_hw *hw = &adapter->hw;
-	struct list_head *pos;
-	struct vf_macvlans *entry;
-
-	list_for_each(pos, &adapter->vf_mvs.l) {
-		entry = list_entry(pos, struct vf_macvlans, l);
-		if (!entry->free)
-			hw->mac.ops.set_rar(hw, entry->rar_entry,
-					    entry->vf_macvlan,
-					    entry->vf, IXGBE_RAH_AV);
-	}
-}
-
+#ifdef CONFIG_PCI_IOV
 void ixgbe_restore_vf_multicasts(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -382,6 +369,7 @@ void ixgbe_restore_vf_multicasts(struct ixgbe_adapter *adapter)
 	u32 mta_reg;
 
 	for (i = 0; i < adapter->num_vfs; i++) {
+		u32 vmolr = IXGBE_READ_REG(hw, IXGBE_VMOLR(i));
 		vfinfo = &adapter->vfinfo[i];
 		for (j = 0; j < vfinfo->num_vf_mc_hashes; j++) {
 			hw->addr_ctrl.mta_in_use++;
@@ -391,11 +379,18 @@ void ixgbe_restore_vf_multicasts(struct ixgbe_adapter *adapter)
 			mta_reg |= (1 << vector_bit);
 			IXGBE_WRITE_REG(hw, IXGBE_MTA(vector_reg), mta_reg);
 		}
+
+		if (vfinfo->num_vf_mc_hashes)
+			vmolr |= IXGBE_VMOLR_ROMPE;
+		else
+			vmolr &= ~IXGBE_VMOLR_ROMPE;
+		IXGBE_WRITE_REG(hw, IXGBE_VMOLR(i), vmolr);
 	}
 
 	/* Restore any VF macvlans */
-	ixgbe_restore_vf_macvlans(adapter);
+	ixgbe_full_sync_mac_table(adapter);
 }
+#endif
 
 static int ixgbe_set_vf_vlan(struct ixgbe_adapter *adapter, int add, int vid,
 			     u32 vf)
@@ -495,8 +490,7 @@ static s32 ixgbe_set_vf_lpe(struct ixgbe_adapter *adapter, u32 *msgbuf, u32 vf)
 static void ixgbe_set_vmolr(struct ixgbe_hw *hw, u32 vf, bool aupe)
 {
 	u32 vmolr = IXGBE_READ_REG(hw, IXGBE_VMOLR(vf));
-	vmolr |= (IXGBE_VMOLR_ROMPE |
-		  IXGBE_VMOLR_BAM);
+	vmolr |= IXGBE_VMOLR_BAM;
 	if (aupe)
 		vmolr |= IXGBE_VMOLR_AUPE;
 	else
@@ -514,7 +508,6 @@ static inline void ixgbe_vf_reset_event(struct ixgbe_adapter *adapter, u32 vf)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	struct vf_data_storage *vfinfo = &adapter->vfinfo[vf];
-	int rar_entry = hw->mac.num_rar_entries - (vf + 1);
 	u8 num_tcs = netdev_get_num_tc(adapter->netdev);
 
 	/* add PF assigned VLAN or VLAN 0 */
@@ -544,7 +537,7 @@ static inline void ixgbe_vf_reset_event(struct ixgbe_adapter *adapter, u32 vf)
 	/* Flush and reset the mta with the new values */
 	ixgbe_set_rx_mode(adapter->netdev);
 
-	hw->mac.ops.clear_rar(hw, rar_entry);
+	ixgbe_del_mac_filter(adapter, adapter->vfinfo[vf].vf_mac_addresses, vf);
 
 	/* reset VF api back to unknown */
 	adapter->vfinfo[vf].vf_api = ixgbe_mbox_api_10;
@@ -553,11 +546,9 @@ static inline void ixgbe_vf_reset_event(struct ixgbe_adapter *adapter, u32 vf)
 static int ixgbe_set_vf_mac(struct ixgbe_adapter *adapter,
 			    int vf, unsigned char *mac_addr)
 {
-	struct ixgbe_hw *hw = &adapter->hw;
-	int rar_entry = hw->mac.num_rar_entries - (vf + 1);
-
+	ixgbe_del_mac_filter(adapter, adapter->vfinfo[vf].vf_mac_addresses, vf);
 	memcpy(adapter->vfinfo[vf].vf_mac_addresses, mac_addr, ETH_ALEN);
-	hw->mac.ops.set_rar(hw, rar_entry, mac_addr, vf, IXGBE_RAH_AV);
+	ixgbe_add_mac_filter(adapter, adapter->vfinfo[vf].vf_mac_addresses, vf);
 
 	return 0;
 }
@@ -565,7 +556,6 @@ static int ixgbe_set_vf_mac(struct ixgbe_adapter *adapter,
 static int ixgbe_set_vf_macvlan(struct ixgbe_adapter *adapter,
 				int vf, int index, unsigned char *mac_addr)
 {
-	struct ixgbe_hw *hw = &adapter->hw;
 	struct list_head *pos;
 	struct vf_macvlans *entry;
 
@@ -576,7 +566,8 @@ static int ixgbe_set_vf_macvlan(struct ixgbe_adapter *adapter,
 				entry->vf = -1;
 				entry->free = true;
 				entry->is_macvlan = false;
-				hw->mac.ops.clear_rar(hw, entry->rar_entry);
+				ixgbe_del_mac_filter(adapter,
+						     entry->vf_macvlan, vf);
 			}
 		}
 	}
@@ -612,7 +603,7 @@ static int ixgbe_set_vf_macvlan(struct ixgbe_adapter *adapter,
 	entry->vf = vf;
 	memcpy(entry->vf_macvlan, mac_addr, ETH_ALEN);
 
-	hw->mac.ops.set_rar(hw, entry->rar_entry, mac_addr, vf, IXGBE_RAH_AV);
+	ixgbe_add_mac_filter(adapter, mac_addr, vf);
 
 	return 0;
 }
@@ -1138,9 +1129,9 @@ int ixgbe_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos)
 			adapter->vfinfo[vf].vlan_count--;
 		adapter->vfinfo[vf].pf_vlan = 0;
 		adapter->vfinfo[vf].pf_qos = 0;
-       }
+	}
 out:
-       return err;
+	return err;
 }
 
 static int ixgbe_link_mbps(struct ixgbe_adapter *adapter)
@@ -1231,7 +1222,8 @@ void ixgbe_check_vf_rate_limit(struct ixgbe_adapter *adapter)
 	}
 }
 
-int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate)
+int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int min_tx_rate,
+			int max_tx_rate)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	int link_speed;
@@ -1249,13 +1241,16 @@ int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate)
 	if (link_speed != 10000)
 		return -EINVAL;
 
+	if (min_tx_rate)
+		return -EINVAL;
+
 	/* rate limit cannot be less than 10Mbs or greater than link speed */
-	if (tx_rate && ((tx_rate <= 10) || (tx_rate > link_speed)))
+	if (max_tx_rate && ((max_tx_rate <= 10) || (max_tx_rate > link_speed)))
 		return -EINVAL;
 
 	/* store values */
 	adapter->vf_rate_link_speed = link_speed;
-	adapter->vfinfo[vf].tx_rate = tx_rate;
+	adapter->vfinfo[vf].tx_rate = max_tx_rate;
 
 	/* update hardware configuration */
 	ixgbe_set_vf_rate_limit(adapter, vf);
@@ -1297,7 +1292,8 @@ int ixgbe_ndo_get_vf_config(struct net_device *netdev,
 		return -EINVAL;
 	ivi->vf = vf;
 	memcpy(&ivi->mac, adapter->vfinfo[vf].vf_mac_addresses, ETH_ALEN);
-	ivi->tx_rate = adapter->vfinfo[vf].tx_rate;
+	ivi->max_tx_rate = adapter->vfinfo[vf].tx_rate;
+	ivi->min_tx_rate = 0;
 	ivi->vlan = adapter->vfinfo[vf].pf_vlan;
 	ivi->qos = adapter->vfinfo[vf].pf_qos;
 	ivi->spoofchk = adapter->vfinfo[vf].spoofchk_enabled;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
index 139eaddfb2ed..32c26d586c01 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
@@ -34,7 +34,9 @@
  */
 #define IXGBE_MAX_VFS_DRV_LIMIT  (IXGBE_MAX_VF_FUNCTIONS - 1)
 
+#ifdef CONFIG_PCI_IOV
 void ixgbe_restore_vf_multicasts(struct ixgbe_adapter *adapter);
+#endif
 void ixgbe_msg_task(struct ixgbe_adapter *adapter);
 int ixgbe_vf_configuration(struct pci_dev *pdev, unsigned int event_mask);
 void ixgbe_disable_tx_rx(struct ixgbe_adapter *adapter);
@@ -42,7 +44,8 @@ void ixgbe_ping_all_vfs(struct ixgbe_adapter *adapter);
 int ixgbe_ndo_set_vf_mac(struct net_device *netdev, int queue, u8 *mac);
 int ixgbe_ndo_set_vf_vlan(struct net_device *netdev, int queue, u16 vlan,
 			   u8 qos);
-int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate);
+int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int min_tx_rate,
+			int max_tx_rate);
 int ixgbe_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting);
 int ixgbe_ndo_get_vf_config(struct net_device *netdev,
 			    int vf, struct ifla_vf_info *ivi);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
index 8a6ff2423f07..9a89f98b35f0 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
@@ -160,7 +160,7 @@ struct ixgbe_thermal_sensor_data {
 #define IXGBE_MAX_EITR     0x00000FF8
 #define IXGBE_MIN_EITR     8
 #define IXGBE_EITR(_i)  (((_i) <= 23) ? (0x00820 + ((_i) * 4)) : \
-                         (0x012300 + (((_i) - 24) * 4)))
+			 (0x012300 + (((_i) - 24) * 4)))
 #define IXGBE_EITR_ITR_INT_MASK 0x00000FF8
 #define IXGBE_EITR_LLI_MOD      0x00008000
 #define IXGBE_EITR_CNT_WDIS     0x80000000
@@ -213,7 +213,7 @@ struct ixgbe_thermal_sensor_data {
  * 64-127: 0x0D014 + (n-64)*0x40
  */
 #define IXGBE_SRRCTL(_i) (((_i) <= 15) ? (0x02100 + ((_i) * 4)) : \
-                          (((_i) < 64) ? (0x01014 + ((_i) * 0x40)) : \
+			  (((_i) < 64) ? (0x01014 + ((_i) * 0x40)) : \
 			  (0x0D014 + (((_i) - 64) * 0x40))))
 /*
  * Rx DCA Control Register:
@@ -222,11 +222,11 @@ struct ixgbe_thermal_sensor_data {
  * 64-127: 0x0D00C + (n-64)*0x40
  */
 #define IXGBE_DCA_RXCTRL(_i)    (((_i) <= 15) ? (0x02200 + ((_i) * 4)) : \
-                                 (((_i) < 64) ? (0x0100C + ((_i) * 0x40)) : \
+				 (((_i) < 64) ? (0x0100C + ((_i) * 0x40)) : \
 				 (0x0D00C + (((_i) - 64) * 0x40))))
 #define IXGBE_RDRXCTL           0x02F00
 #define IXGBE_RXPBSIZE(_i)      (0x03C00 + ((_i) * 4))
-                                             /* 8 of these 0x03C00 - 0x03C1C */
+					     /* 8 of these 0x03C00 - 0x03C1C */
 #define IXGBE_RXCTRL    0x03000
 #define IXGBE_DROPEN    0x03D04
 #define IXGBE_RXPBSIZE_SHIFT 10
@@ -239,14 +239,14 @@ struct ixgbe_thermal_sensor_data {
 /* Multicast Table Array - 128 entries */
 #define IXGBE_MTA(_i)   (0x05200 + ((_i) * 4))
 #define IXGBE_RAL(_i)   (((_i) <= 15) ? (0x05400 + ((_i) * 8)) : \
-                         (0x0A200 + ((_i) * 8)))
+			 (0x0A200 + ((_i) * 8)))
 #define IXGBE_RAH(_i)   (((_i) <= 15) ? (0x05404 + ((_i) * 8)) : \
-                         (0x0A204 + ((_i) * 8)))
+			 (0x0A204 + ((_i) * 8)))
 #define IXGBE_MPSAR_LO(_i) (0x0A600 + ((_i) * 8))
 #define IXGBE_MPSAR_HI(_i) (0x0A604 + ((_i) * 8))
 /* Packet split receive type */
 #define IXGBE_PSRTYPE(_i)    (((_i) <= 15) ? (0x05480 + ((_i) * 4)) : \
-                              (0x0EA00 + ((_i) * 4)))
+			      (0x0EA00 + ((_i) * 4)))
 /* array of 4096 1-bit vlan filters */
 #define IXGBE_VFTA(_i)  (0x0A000 + ((_i) * 4))
 /*array of 4096 4-bit vlan vmdq indices */
@@ -696,7 +696,7 @@ struct ixgbe_thermal_sensor_data {
 
 #define IXGBE_RQSMR(_i) (0x02300 + ((_i) * 4))
 #define IXGBE_TQSMR(_i) (((_i) <= 7) ? (0x07300 + ((_i) * 4)) : \
-                         (0x08600 + ((_i) * 4)))
+			 (0x08600 + ((_i) * 4)))
 #define IXGBE_TQSM(_i)  (0x08600 + ((_i) * 4))
 
 #define IXGBE_QPRC(_i) (0x01030 + ((_i) * 0x40)) /* 16 of these */
@@ -820,7 +820,7 @@ struct ixgbe_thermal_sensor_data {
 #define IXGBE_GCR_EXT_VT_MODE_32        0x00000002
 #define IXGBE_GCR_EXT_VT_MODE_64        0x00000003
 #define IXGBE_GCR_EXT_SRIOV             (IXGBE_GCR_EXT_MSIX_EN | \
-                                         IXGBE_GCR_EXT_VT_MODE_64)
+					 IXGBE_GCR_EXT_VT_MODE_64)
 
 /* Time Sync Registers */
 #define IXGBE_TSYNCRXCTL 0x05188 /* Rx Time Sync Control register - RW */
@@ -1396,10 +1396,10 @@ enum {
 #define IXGBE_EIMC_OTHER        IXGBE_EICR_OTHER     /* INT Cause Active */
 
 #define IXGBE_EIMS_ENABLE_MASK ( \
-                                IXGBE_EIMS_RTX_QUEUE       | \
-                                IXGBE_EIMS_LSC             | \
-                                IXGBE_EIMS_TCP_TIMER       | \
-                                IXGBE_EIMS_OTHER)
+				IXGBE_EIMS_RTX_QUEUE       | \
+				IXGBE_EIMS_LSC             | \
+				IXGBE_EIMS_TCP_TIMER       | \
+				IXGBE_EIMS_OTHER)
 
 /* Immediate Interrupt Rx (A.K.A. Low Latency Interrupt) */
 #define IXGBE_IMIR_PORT_IM_EN     0x00010000  /* TCP port enable */
@@ -2161,18 +2161,18 @@ enum {
 
 /* Masks to determine if packets should be dropped due to frame errors */
 #define IXGBE_RXD_ERR_FRAME_ERR_MASK ( \
-                                      IXGBE_RXD_ERR_CE | \
-                                      IXGBE_RXD_ERR_LE | \
-                                      IXGBE_RXD_ERR_PE | \
-                                      IXGBE_RXD_ERR_OSE | \
-                                      IXGBE_RXD_ERR_USE)
+				      IXGBE_RXD_ERR_CE | \
+				      IXGBE_RXD_ERR_LE | \
+				      IXGBE_RXD_ERR_PE | \
+				      IXGBE_RXD_ERR_OSE | \
+				      IXGBE_RXD_ERR_USE)
 
 #define IXGBE_RXDADV_ERR_FRAME_ERR_MASK ( \
-                                      IXGBE_RXDADV_ERR_CE | \
-                                      IXGBE_RXDADV_ERR_LE | \
-                                      IXGBE_RXDADV_ERR_PE | \
-                                      IXGBE_RXDADV_ERR_OSE | \
-                                      IXGBE_RXDADV_ERR_USE)
+				      IXGBE_RXDADV_ERR_CE | \
+				      IXGBE_RXDADV_ERR_LE | \
+				      IXGBE_RXDADV_ERR_PE | \
+				      IXGBE_RXDADV_ERR_OSE | \
+				      IXGBE_RXDADV_ERR_USE)
 
 /* Multicast bit mask */
 #define IXGBE_MCSTCTRL_MFE      0x4
@@ -2393,9 +2393,9 @@ struct ixgbe_adv_tx_context_desc {
 #define IXGBE_ADVTXD_CC         0x00000080 /* Check Context */
 #define IXGBE_ADVTXD_POPTS_SHIFT      8  /* Adv desc POPTS shift */
 #define IXGBE_ADVTXD_POPTS_IXSM (IXGBE_TXD_POPTS_IXSM << \
-                                 IXGBE_ADVTXD_POPTS_SHIFT)
+				 IXGBE_ADVTXD_POPTS_SHIFT)
 #define IXGBE_ADVTXD_POPTS_TXSM (IXGBE_TXD_POPTS_TXSM << \
-                                 IXGBE_ADVTXD_POPTS_SHIFT)
+				 IXGBE_ADVTXD_POPTS_SHIFT)
 #define IXGBE_ADVTXD_POPTS_ISCO_1ST  0x00000000 /* 1st TSO of iSCSI PDU */
 #define IXGBE_ADVTXD_POPTS_ISCO_MDL  0x00000800 /* Middle TSO of iSCSI PDU */
 #define IXGBE_ADVTXD_POPTS_ISCO_LAST 0x00001000 /* Last TSO of iSCSI PDU */
@@ -2435,10 +2435,10 @@ typedef u32 ixgbe_link_speed;
 #define IXGBE_LINK_SPEED_1GB_FULL  0x0020
 #define IXGBE_LINK_SPEED_10GB_FULL 0x0080
 #define IXGBE_LINK_SPEED_82598_AUTONEG (IXGBE_LINK_SPEED_1GB_FULL | \
-                                        IXGBE_LINK_SPEED_10GB_FULL)
+					IXGBE_LINK_SPEED_10GB_FULL)
 #define IXGBE_LINK_SPEED_82599_AUTONEG (IXGBE_LINK_SPEED_100_FULL | \
-                                        IXGBE_LINK_SPEED_1GB_FULL | \
-                                        IXGBE_LINK_SPEED_10GB_FULL)
+					IXGBE_LINK_SPEED_1GB_FULL | \
+					IXGBE_LINK_SPEED_10GB_FULL)
 
 
 /* Physical layer type */
@@ -2746,7 +2746,7 @@ struct ixgbe_bus_info {
 /* Flow control parameters */
 struct ixgbe_fc_info {
 	u32 high_water[MAX_TRAFFIC_CLASS]; /* Flow Control High-water */
-	u32 low_water; /* Flow Control Low-water */
+	u32 low_water[MAX_TRAFFIC_CLASS]; /* Flow Control Low-water */
 	u16 pause_time; /* Flow Control Pause timer */
 	bool send_xon; /* Flow control send XON */
 	bool strict_ieee; /* Strict IEEE mode */
@@ -2840,7 +2840,7 @@ struct ixgbe_hw;
 
 /* iterator type for walking multicast address lists */
 typedef u8* (*ixgbe_mc_addr_itr) (struct ixgbe_hw *hw, u8 **mc_addr_ptr,
-                                  u32 *vmdq);
+				  u32 *vmdq);
 
 /* Function pointer table */
 struct ixgbe_eeprom_operations {
@@ -2887,7 +2887,7 @@ struct ixgbe_mac_operations {
 	s32 (*setup_link)(struct ixgbe_hw *, ixgbe_link_speed, bool);
 	s32 (*check_link)(struct ixgbe_hw *, ixgbe_link_speed *, bool *, bool);
 	s32 (*get_link_capabilities)(struct ixgbe_hw *, ixgbe_link_speed *,
-	                             bool *);
+				     bool *);
 
 	/* Packet Buffer Manipulation */
 	void (*set_rxpba)(struct ixgbe_hw *, int, u32, int);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c
index 188a5974b85c..40dd798e1290 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c
@@ -81,7 +81,7 @@ static s32 ixgbe_setup_mac_link_X540(struct ixgbe_hw *hw,
 				     bool autoneg_wait_to_complete)
 {
 	return hw->phy.ops.setup_link_speed(hw, speed,
-	                                    autoneg_wait_to_complete);
+					    autoneg_wait_to_complete);
 }
 
 /**
@@ -155,7 +155,7 @@ mac_reset_top:
 	/* Add the SAN MAC address to the RAR only if it's a valid address */
 	if (is_valid_ether_addr(hw->mac.san_addr)) {
 		hw->mac.ops.set_rar(hw, hw->mac.num_rar_entries - 1,
-		                    hw->mac.san_addr, 0, IXGBE_RAH_AV);
+				    hw->mac.san_addr, 0, IXGBE_RAH_AV);
 
 		/* Save the SAN MAC RAR index */
 		hw->mac.san_mac_rar_index = hw->mac.num_rar_entries - 1;
@@ -166,7 +166,7 @@ mac_reset_top:
 
 	/* Store the alternative WWNN/WWPN prefix */
 	hw->mac.ops.get_wwn_prefix(hw, &hw->mac.wwnn_prefix,
-	                           &hw->mac.wwpn_prefix);
+				   &hw->mac.wwpn_prefix);
 
 reset_hw_out:
 	return status;
@@ -237,9 +237,9 @@ static s32 ixgbe_init_eeprom_params_X540(struct ixgbe_hw *hw)
 
 		eec = IXGBE_READ_REG(hw, IXGBE_EEC);
 		eeprom_size = (u16)((eec & IXGBE_EEC_SIZE) >>
-		                    IXGBE_EEC_SIZE_SHIFT);
+				    IXGBE_EEC_SIZE_SHIFT);
 		eeprom->word_size = 1 << (eeprom_size +
-		                          IXGBE_EEPROM_WORD_SIZE_SHIFT);
+					  IXGBE_EEPROM_WORD_SIZE_SHIFT);
 
 		hw_dbg(hw, "Eeprom params: type = %d, size = %d\n",
 		       eeprom->type, eeprom->word_size);
@@ -712,8 +712,7 @@ static s32 ixgbe_get_swfw_sync_semaphore(struct ixgbe_hw *hw)
 			udelay(50);
 		}
 	} else {
-		hw_dbg(hw, "Software semaphore SMBI between device drivers "
-		           "not granted.\n");
+		hw_dbg(hw, "Software semaphore SMBI between device drivers not granted.\n");
 	}
 
 	return status;
@@ -813,7 +812,7 @@ static struct ixgbe_mac_operations mac_ops_X540 = {
 	.clear_hw_cntrs         = &ixgbe_clear_hw_cntrs_generic,
 	.get_media_type         = &ixgbe_get_media_type_X540,
 	.get_supported_physical_layer =
-                                  &ixgbe_get_supported_physical_layer_X540,
+				  &ixgbe_get_supported_physical_layer_X540,
 	.enable_rx_dma          = &ixgbe_enable_rx_dma_generic,
 	.get_mac_addr           = &ixgbe_get_mac_addr_generic,
 	.get_san_mac_addr       = &ixgbe_get_san_mac_addr_generic,
diff --git a/drivers/net/ethernet/intel/ixgbevf/ethtool.c b/drivers/net/ethernet/intel/ixgbevf/ethtool.c
index 1baecb60f065..d420f124633f 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ethtool.c
@@ -135,8 +135,8 @@ static int ixgbevf_get_settings(struct net_device *netdev,
 		ethtool_cmd_speed_set(ecmd, speed);
 		ecmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	return 0;
@@ -813,5 +813,5 @@ static const struct ethtool_ops ixgbevf_ethtool_ops = {
 
 void ixgbevf_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &ixgbevf_ethtool_ops);
+	netdev->ethtool_ops = &ixgbevf_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index de2793b06305..75467f83772c 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -85,7 +85,7 @@ static DEFINE_PCI_DEVICE_TABLE(ixgbevf_pci_tbl) = {
 MODULE_DEVICE_TABLE(pci, ixgbevf_pci_tbl);
 
 MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
-MODULE_DESCRIPTION("Intel(R) 82599 Virtual Function Driver");
+MODULE_DESCRIPTION("Intel(R) 10 Gigabit Virtual Function Network Driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 
diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c
index b7b8d74c22d9..b151a949f352 100644
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -42,6 +42,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/in.h>
 #include <linux/ip.h>
+#include <net/tso.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
 #include <linux/etherdevice.h>
@@ -179,10 +180,18 @@ static char mv643xx_eth_driver_version[] = "1.4";
  * Misc definitions.
  */
 #define DEFAULT_RX_QUEUE_SIZE	128
-#define DEFAULT_TX_QUEUE_SIZE	256
+#define DEFAULT_TX_QUEUE_SIZE	512
 #define SKB_DMA_REALIGN		((PAGE_SIZE - NET_SKB_PAD) % SMP_CACHE_BYTES)
 
+#define TSO_HEADER_SIZE		128
 
+/* Max number of allowed TCP segments for software TSO */
+#define MV643XX_MAX_TSO_SEGS 100
+#define MV643XX_MAX_SKB_DESCS (MV643XX_MAX_TSO_SEGS * 2 + MAX_SKB_FRAGS)
+
+#define IS_TSO_HEADER(txq, addr) \
+	((addr >= txq->tso_hdrs_dma) && \
+	 (addr < txq->tso_hdrs_dma + txq->tx_ring_size * TSO_HEADER_SIZE))
 /*
  * RX/TX descriptors.
  */
@@ -250,6 +259,7 @@ struct tx_desc {
 #define GEN_TCP_UDP_CHECKSUM		0x00020000
 #define UDP_FRAME			0x00010000
 #define MAC_HDR_EXTRA_4_BYTES		0x00008000
+#define GEN_TCP_UDP_CHK_FULL		0x00000400
 #define MAC_HDR_EXTRA_8_BYTES		0x00000200
 
 #define TX_IHL_SHIFT			11
@@ -345,6 +355,12 @@ struct tx_queue {
 	int tx_curr_desc;
 	int tx_used_desc;
 
+	int tx_stop_threshold;
+	int tx_wake_threshold;
+
+	char *tso_hdrs;
+	dma_addr_t tso_hdrs_dma;
+
 	struct tx_desc *tx_desc_area;
 	dma_addr_t tx_desc_dma;
 	int tx_desc_area_size;
@@ -491,7 +507,7 @@ static void txq_maybe_wake(struct tx_queue *txq)
 
 	if (netif_tx_queue_stopped(nq)) {
 		__netif_tx_lock(nq, smp_processor_id());
-		if (txq->tx_ring_size - txq->tx_desc_count >= MAX_SKB_FRAGS + 1)
+		if (txq->tx_desc_count <= txq->tx_wake_threshold)
 			netif_tx_wake_queue(nq);
 		__netif_tx_unlock(nq);
 	}
@@ -661,6 +677,198 @@ static inline unsigned int has_tiny_unaligned_frags(struct sk_buff *skb)
 	return 0;
 }
 
+static inline __be16 sum16_as_be(__sum16 sum)
+{
+	return (__force __be16)sum;
+}
+
+static int skb_tx_csum(struct mv643xx_eth_private *mp, struct sk_buff *skb,
+		       u16 *l4i_chk, u32 *command, int length)
+{
+	int ret;
+	u32 cmd = 0;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		int hdr_len;
+		int tag_bytes;
+
+		BUG_ON(skb->protocol != htons(ETH_P_IP) &&
+		       skb->protocol != htons(ETH_P_8021Q));
+
+		hdr_len = (void *)ip_hdr(skb) - (void *)skb->data;
+		tag_bytes = hdr_len - ETH_HLEN;
+
+		if (length - hdr_len > mp->shared->tx_csum_limit ||
+		    unlikely(tag_bytes & ~12)) {
+			ret = skb_checksum_help(skb);
+			if (!ret)
+				goto no_csum;
+			return ret;
+		}
+
+		if (tag_bytes & 4)
+			cmd |= MAC_HDR_EXTRA_4_BYTES;
+		if (tag_bytes & 8)
+			cmd |= MAC_HDR_EXTRA_8_BYTES;
+
+		cmd |= GEN_TCP_UDP_CHECKSUM | GEN_TCP_UDP_CHK_FULL |
+			   GEN_IP_V4_CHECKSUM   |
+			   ip_hdr(skb)->ihl << TX_IHL_SHIFT;
+
+		/* TODO: Revisit this. With the usage of GEN_TCP_UDP_CHK_FULL
+		 * it seems we don't need to pass the initial checksum. */
+		switch (ip_hdr(skb)->protocol) {
+		case IPPROTO_UDP:
+			cmd |= UDP_FRAME;
+			*l4i_chk = 0;
+			break;
+		case IPPROTO_TCP:
+			*l4i_chk = 0;
+			break;
+		default:
+			WARN(1, "protocol not supported");
+		}
+	} else {
+no_csum:
+		/* Errata BTS #50, IHL must be 5 if no HW checksum */
+		cmd |= 5 << TX_IHL_SHIFT;
+	}
+	*command = cmd;
+	return 0;
+}
+
+static inline int
+txq_put_data_tso(struct net_device *dev, struct tx_queue *txq,
+		 struct sk_buff *skb, char *data, int length,
+		 bool last_tcp, bool is_last)
+{
+	int tx_index;
+	u32 cmd_sts;
+	struct tx_desc *desc;
+
+	tx_index = txq->tx_curr_desc++;
+	if (txq->tx_curr_desc == txq->tx_ring_size)
+		txq->tx_curr_desc = 0;
+	desc = &txq->tx_desc_area[tx_index];
+
+	desc->l4i_chk = 0;
+	desc->byte_cnt = length;
+	desc->buf_ptr = dma_map_single(dev->dev.parent, data,
+				       length, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(dev->dev.parent, desc->buf_ptr))) {
+		WARN(1, "dma_map_single failed!\n");
+		return -ENOMEM;
+	}
+
+	cmd_sts = BUFFER_OWNED_BY_DMA;
+	if (last_tcp) {
+		/* last descriptor in the TCP packet */
+		cmd_sts |= ZERO_PADDING | TX_LAST_DESC;
+		/* last descriptor in SKB */
+		if (is_last)
+			cmd_sts |= TX_ENABLE_INTERRUPT;
+	}
+	desc->cmd_sts = cmd_sts;
+	return 0;
+}
+
+static inline void
+txq_put_hdr_tso(struct sk_buff *skb, struct tx_queue *txq, int length)
+{
+	struct mv643xx_eth_private *mp = txq_to_mp(txq);
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	int tx_index;
+	struct tx_desc *desc;
+	int ret;
+	u32 cmd_csum = 0;
+	u16 l4i_chk = 0;
+
+	tx_index = txq->tx_curr_desc;
+	desc = &txq->tx_desc_area[tx_index];
+
+	ret = skb_tx_csum(mp, skb, &l4i_chk, &cmd_csum, length);
+	if (ret)
+		WARN(1, "failed to prepare checksum!");
+
+	/* Should we set this? Can't use the value from skb_tx_csum()
+	 * as it's not the correct initial L4 checksum to use. */
+	desc->l4i_chk = 0;
+
+	desc->byte_cnt = hdr_len;
+	desc->buf_ptr = txq->tso_hdrs_dma +
+			txq->tx_curr_desc * TSO_HEADER_SIZE;
+	desc->cmd_sts = cmd_csum | BUFFER_OWNED_BY_DMA  | TX_FIRST_DESC |
+				   GEN_CRC;
+
+	txq->tx_curr_desc++;
+	if (txq->tx_curr_desc == txq->tx_ring_size)
+		txq->tx_curr_desc = 0;
+}
+
+static int txq_submit_tso(struct tx_queue *txq, struct sk_buff *skb,
+			  struct net_device *dev)
+{
+	struct mv643xx_eth_private *mp = txq_to_mp(txq);
+	int total_len, data_left, ret;
+	int desc_count = 0;
+	struct tso_t tso;
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+
+	/* Count needed descriptors */
+	if ((txq->tx_desc_count + tso_count_descs(skb)) >= txq->tx_ring_size) {
+		netdev_dbg(dev, "not enough descriptors for TSO!\n");
+		return -EBUSY;
+	}
+
+	/* Initialize the TSO handler, and prepare the first payload */
+	tso_start(skb, &tso);
+
+	total_len = skb->len - hdr_len;
+	while (total_len > 0) {
+		char *hdr;
+
+		data_left = min_t(int, skb_shinfo(skb)->gso_size, total_len);
+		total_len -= data_left;
+		desc_count++;
+
+		/* prepare packet headers: MAC + IP + TCP */
+		hdr = txq->tso_hdrs + txq->tx_curr_desc * TSO_HEADER_SIZE;
+		tso_build_hdr(skb, hdr, &tso, data_left, total_len == 0);
+		txq_put_hdr_tso(skb, txq, data_left);
+
+		while (data_left > 0) {
+			int size;
+			desc_count++;
+
+			size = min_t(int, tso.size, data_left);
+			ret = txq_put_data_tso(dev, txq, skb, tso.data, size,
+					       size == data_left,
+					       total_len == 0);
+			if (ret)
+				goto err_release;
+			data_left -= size;
+			tso_build_data(skb, &tso, size);
+		}
+	}
+
+	__skb_queue_tail(&txq->tx_skb, skb);
+	skb_tx_timestamp(skb);
+
+	/* clear TX_END status */
+	mp->work_tx_end &= ~(1 << txq->index);
+
+	/* ensure all descriptors are written before poking hardware */
+	wmb();
+	txq_enable(txq);
+	txq->tx_desc_count += desc_count;
+	return 0;
+err_release:
+	/* TODO: Release all used data descriptors; header descriptors must not
+	 * be DMA-unmapped.
+	 */
+	return ret;
+}
+
 static void txq_submit_frag_skb(struct tx_queue *txq, struct sk_buff *skb)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
@@ -671,8 +879,10 @@ static void txq_submit_frag_skb(struct tx_queue *txq, struct sk_buff *skb)
 		skb_frag_t *this_frag;
 		int tx_index;
 		struct tx_desc *desc;
+		void *addr;
 
 		this_frag = &skb_shinfo(skb)->frags[frag];
+		addr = page_address(this_frag->page.p) + this_frag->page_offset;
 		tx_index = txq->tx_curr_desc++;
 		if (txq->tx_curr_desc == txq->tx_ring_size)
 			txq->tx_curr_desc = 0;
@@ -692,19 +902,13 @@ static void txq_submit_frag_skb(struct tx_queue *txq, struct sk_buff *skb)
 
 		desc->l4i_chk = 0;
 		desc->byte_cnt = skb_frag_size(this_frag);
-		desc->buf_ptr = skb_frag_dma_map(mp->dev->dev.parent,
-						 this_frag, 0,
-						 skb_frag_size(this_frag),
-						 DMA_TO_DEVICE);
+		desc->buf_ptr = dma_map_single(mp->dev->dev.parent, addr,
+					       desc->byte_cnt, DMA_TO_DEVICE);
 	}
 }
 
-static inline __be16 sum16_as_be(__sum16 sum)
-{
-	return (__force __be16)sum;
-}
-
-static int txq_submit_skb(struct tx_queue *txq, struct sk_buff *skb)
+static int txq_submit_skb(struct tx_queue *txq, struct sk_buff *skb,
+			  struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
 	int nr_frags = skb_shinfo(skb)->nr_frags;
@@ -712,54 +916,22 @@ static int txq_submit_skb(struct tx_queue *txq, struct sk_buff *skb)
 	struct tx_desc *desc;
 	u32 cmd_sts;
 	u16 l4i_chk;
-	int length;
+	int length, ret;
 
-	cmd_sts = TX_FIRST_DESC | GEN_CRC | BUFFER_OWNED_BY_DMA;
+	cmd_sts = 0;
 	l4i_chk = 0;
 
-	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		int hdr_len;
-		int tag_bytes;
-
-		BUG_ON(skb->protocol != htons(ETH_P_IP) &&
-		       skb->protocol != htons(ETH_P_8021Q));
-
-		hdr_len = (void *)ip_hdr(skb) - (void *)skb->data;
-		tag_bytes = hdr_len - ETH_HLEN;
-		if (skb->len - hdr_len > mp->shared->tx_csum_limit ||
-		    unlikely(tag_bytes & ~12)) {
-			if (skb_checksum_help(skb) == 0)
-				goto no_csum;
-			dev_kfree_skb_any(skb);
-			return 1;
-		}
-
-		if (tag_bytes & 4)
-			cmd_sts |= MAC_HDR_EXTRA_4_BYTES;
-		if (tag_bytes & 8)
-			cmd_sts |= MAC_HDR_EXTRA_8_BYTES;
-
-		cmd_sts |= GEN_TCP_UDP_CHECKSUM |
-			   GEN_IP_V4_CHECKSUM   |
-			   ip_hdr(skb)->ihl << TX_IHL_SHIFT;
-
-		switch (ip_hdr(skb)->protocol) {
-		case IPPROTO_UDP:
-			cmd_sts |= UDP_FRAME;
-			l4i_chk = ntohs(sum16_as_be(udp_hdr(skb)->check));
-			break;
-		case IPPROTO_TCP:
-			l4i_chk = ntohs(sum16_as_be(tcp_hdr(skb)->check));
-			break;
-		default:
-			BUG();
-		}
-	} else {
-no_csum:
-		/* Errata BTS #50, IHL must be 5 if no HW checksum */
-		cmd_sts |= 5 << TX_IHL_SHIFT;
+	if (txq->tx_ring_size - txq->tx_desc_count < MAX_SKB_FRAGS + 1) {
+		if (net_ratelimit())
+			netdev_err(dev, "tx queue full?!\n");
+		return -EBUSY;
 	}
 
+	ret = skb_tx_csum(mp, skb, &l4i_chk, &cmd_sts, skb->len);
+	if (ret)
+		return ret;
+	cmd_sts |= TX_FIRST_DESC | GEN_CRC | BUFFER_OWNED_BY_DMA;
+
 	tx_index = txq->tx_curr_desc++;
 	if (txq->tx_curr_desc == txq->tx_ring_size)
 		txq->tx_curr_desc = 0;
@@ -801,7 +973,7 @@ no_csum:
 static netdev_tx_t mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	int length, queue;
+	int length, queue, ret;
 	struct tx_queue *txq;
 	struct netdev_queue *nq;
 
@@ -810,30 +982,26 @@ static netdev_tx_t mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	nq = netdev_get_tx_queue(dev, queue);
 
 	if (has_tiny_unaligned_frags(skb) && __skb_linearize(skb)) {
-		txq->tx_dropped++;
 		netdev_printk(KERN_DEBUG, dev,
 			      "failed to linearize skb with tiny unaligned fragment\n");
 		return NETDEV_TX_BUSY;
 	}
 
-	if (txq->tx_ring_size - txq->tx_desc_count < MAX_SKB_FRAGS + 1) {
-		if (net_ratelimit())
-			netdev_err(dev, "tx queue full?!\n");
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
 	length = skb->len;
 
-	if (!txq_submit_skb(txq, skb)) {
-		int entries_left;
-
+	if (skb_is_gso(skb))
+		ret = txq_submit_tso(txq, skb, dev);
+	else
+		ret = txq_submit_skb(txq, skb, dev);
+	if (!ret) {
 		txq->tx_bytes += length;
 		txq->tx_packets++;
 
-		entries_left = txq->tx_ring_size - txq->tx_desc_count;
-		if (entries_left < MAX_SKB_FRAGS + 1)
+		if (txq->tx_desc_count >= txq->tx_stop_threshold)
 			netif_tx_stop_queue(nq);
+	} else {
+		txq->tx_dropped++;
+		dev_kfree_skb_any(skb);
 	}
 
 	return NETDEV_TX_OK;
@@ -907,14 +1075,9 @@ static int txq_reclaim(struct tx_queue *txq, int budget, int force)
 			mp->dev->stats.tx_errors++;
 		}
 
-		if (cmd_sts & TX_FIRST_DESC) {
+		if (!IS_TSO_HEADER(txq, desc->buf_ptr))
 			dma_unmap_single(mp->dev->dev.parent, desc->buf_ptr,
 					 desc->byte_cnt, DMA_TO_DEVICE);
-		} else {
-			dma_unmap_page(mp->dev->dev.parent, desc->buf_ptr,
-				       desc->byte_cnt, DMA_TO_DEVICE);
-		}
-
 		dev_kfree_skb(skb);
 	}
 
@@ -1010,8 +1173,9 @@ static void txq_set_fixed_prio_mode(struct tx_queue *txq)
 
 
 /* mii management interface *************************************************/
-static void mv643xx_adjust_pscr(struct mv643xx_eth_private *mp)
+static void mv643xx_eth_adjust_link(struct net_device *dev)
 {
+	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	u32 pscr = rdlp(mp, PORT_SERIAL_CONTROL);
 	u32 autoneg_disable = FORCE_LINK_PASS |
 	             DISABLE_AUTO_NEG_SPEED_GMII |
@@ -1387,7 +1551,7 @@ mv643xx_eth_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 
 	ret = phy_ethtool_sset(mp->phy, cmd);
 	if (!ret)
-		mv643xx_adjust_pscr(mp);
+		mv643xx_eth_adjust_link(dev);
 	return ret;
 }
 
@@ -1456,7 +1620,11 @@ mv643xx_eth_set_ringparam(struct net_device *dev, struct ethtool_ringparam *er)
 		return -EINVAL;
 
 	mp->rx_ring_size = er->rx_pending < 4096 ? er->rx_pending : 4096;
-	mp->tx_ring_size = er->tx_pending < 4096 ? er->tx_pending : 4096;
+	mp->tx_ring_size = clamp_t(unsigned int, er->tx_pending,
+				   MV643XX_MAX_SKB_DESCS * 2, 4096);
+	if (mp->tx_ring_size != er->tx_pending)
+		netdev_warn(dev, "TX queue size set to %u (requested %u)\n",
+			    mp->tx_ring_size, er->tx_pending);
 
 	if (netif_running(dev)) {
 		mv643xx_eth_stop(dev);
@@ -1832,6 +2000,13 @@ static int txq_init(struct mv643xx_eth_private *mp, int index)
 
 	txq->tx_ring_size = mp->tx_ring_size;
 
+	/* A queue must always have room for at least one skb.
+	 * Therefore, stop the queue when the free entries reaches
+	 * the maximum number of descriptors per skb.
+	 */
+	txq->tx_stop_threshold = txq->tx_ring_size - MV643XX_MAX_SKB_DESCS;
+	txq->tx_wake_threshold = txq->tx_stop_threshold / 2;
+
 	txq->tx_desc_count = 0;
 	txq->tx_curr_desc = 0;
 	txq->tx_used_desc = 0;
@@ -1871,6 +2046,15 @@ static int txq_init(struct mv643xx_eth_private *mp, int index)
 					nexti * sizeof(struct tx_desc);
 	}
 
+	/* Allocate DMA buffers for TSO MAC/IP/TCP headers */
+	txq->tso_hdrs = dma_alloc_coherent(mp->dev->dev.parent,
+					   txq->tx_ring_size * TSO_HEADER_SIZE,
+					   &txq->tso_hdrs_dma, GFP_KERNEL);
+	if (txq->tso_hdrs == NULL) {
+		dma_free_coherent(mp->dev->dev.parent, txq->tx_desc_area_size,
+				  txq->tx_desc_area, txq->tx_desc_dma);
+		return -ENOMEM;
+	}
 	skb_queue_head_init(&txq->tx_skb);
 
 	return 0;
@@ -1891,6 +2075,10 @@ static void txq_deinit(struct tx_queue *txq)
 	else
 		dma_free_coherent(mp->dev->dev.parent, txq->tx_desc_area_size,
 				  txq->tx_desc_area, txq->tx_desc_dma);
+	if (txq->tso_hdrs)
+		dma_free_coherent(mp->dev->dev.parent,
+				  txq->tx_ring_size * TSO_HEADER_SIZE,
+				  txq->tso_hdrs, txq->tso_hdrs_dma);
 }
 
 
@@ -2303,7 +2491,7 @@ static int mv643xx_eth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 	ret = phy_mii_ioctl(mp->phy, ifr, cmd);
 	if (!ret)
-		mv643xx_adjust_pscr(mp);
+		mv643xx_eth_adjust_link(dev);
 	return ret;
 }
 
@@ -2678,6 +2866,7 @@ static void set_params(struct mv643xx_eth_private *mp,
 		       struct mv643xx_eth_platform_data *pd)
 {
 	struct net_device *dev = mp->dev;
+	unsigned int tx_ring_size;
 
 	if (is_valid_ether_addr(pd->mac_addr))
 		memcpy(dev->dev_addr, pd->mac_addr, ETH_ALEN);
@@ -2692,22 +2881,22 @@ static void set_params(struct mv643xx_eth_private *mp,
 
 	mp->rxq_count = pd->rx_queue_count ? : 1;
 
-	mp->tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
+	tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
 	if (pd->tx_queue_size)
-		mp->tx_ring_size = pd->tx_queue_size;
+		tx_ring_size = pd->tx_queue_size;
+
+	mp->tx_ring_size = clamp_t(unsigned int, tx_ring_size,
+				   MV643XX_MAX_SKB_DESCS * 2, 4096);
+	if (mp->tx_ring_size != tx_ring_size)
+		netdev_warn(dev, "TX queue size set to %u (requested %u)\n",
+			    mp->tx_ring_size, tx_ring_size);
+
 	mp->tx_desc_sram_addr = pd->tx_sram_addr;
 	mp->tx_desc_sram_size = pd->tx_sram_size;
 
 	mp->txq_count = pd->tx_queue_count ? : 1;
 }
 
-static void mv643xx_eth_adjust_link(struct net_device *dev)
-{
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
-
-	mv643xx_adjust_pscr(mp);
-}
-
 static struct phy_device *phy_scan(struct mv643xx_eth_private *mp,
 				   int phy_addr)
 {
@@ -2889,7 +3078,7 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 	if (err)
 		goto out;
 
-	SET_ETHTOOL_OPS(dev, &mv643xx_eth_ethtool_ops);
+	dev->ethtool_ops = &mv643xx_eth_ethtool_ops;
 
 	init_pscr(mp, pd->speed, pd->duplex);
 
@@ -2921,11 +3110,14 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 	dev->watchdog_timeo = 2 * HZ;
 	dev->base_addr = 0;
 
-	dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_RXCSUM;
-	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_RXCSUM;
-	dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM;
+	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO;
+	dev->vlan_features = dev->features;
+
+	dev->features |= NETIF_F_RXCSUM;
+	dev->hw_features = dev->features;
 
 	dev->priv_flags |= IFF_UNICAST_FLT;
+	dev->gso_max_segs = MV643XX_MAX_TSO_SEGS;
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
diff --git a/drivers/net/ethernet/marvell/mvmdio.c b/drivers/net/ethernet/marvell/mvmdio.c
index 9d5ced263a5e..fc2fb25343f4 100644
--- a/drivers/net/ethernet/marvell/mvmdio.c
+++ b/drivers/net/ethernet/marvell/mvmdio.c
@@ -195,11 +195,10 @@ static int orion_mdio_probe(struct platform_device *pdev)
 		return -ENODEV;
 	}
 
-	bus = mdiobus_alloc_size(sizeof(struct orion_mdio_dev));
-	if (!bus) {
-		dev_err(&pdev->dev, "Cannot allocate MDIO bus\n");
+	bus = devm_mdiobus_alloc_size(&pdev->dev,
+				      sizeof(struct orion_mdio_dev));
+	if (!bus)
 		return -ENOMEM;
-	}
 
 	bus->name = "orion_mdio_bus";
 	bus->read = orion_mdio_read;
@@ -208,11 +207,10 @@ static int orion_mdio_probe(struct platform_device *pdev)
 		 dev_name(&pdev->dev));
 	bus->parent = &pdev->dev;
 
-	bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL);
-	if (!bus->irq) {
-		mdiobus_free(bus);
+	bus->irq = devm_kmalloc_array(&pdev->dev, PHY_MAX_ADDR, sizeof(int),
+				      GFP_KERNEL);
+	if (!bus->irq)
 		return -ENOMEM;
-	}
 
 	for (i = 0; i < PHY_MAX_ADDR; i++)
 		bus->irq[i] = PHY_POLL;
@@ -264,8 +262,6 @@ static int orion_mdio_probe(struct platform_device *pdev)
 out_mdio:
 	if (!IS_ERR(dev->clk))
 		clk_disable_unprepare(dev->clk);
-	kfree(bus->irq);
-	mdiobus_free(bus);
 	return ret;
 }
 
@@ -276,8 +272,6 @@ static int orion_mdio_remove(struct platform_device *pdev)
 
 	writel(0, dev->regs + MVMDIO_ERR_INT_MASK);
 	mdiobus_unregister(bus);
-	kfree(bus->irq);
-	mdiobus_free(bus);
 	if (!IS_ERR(dev->clk))
 		clk_disable_unprepare(dev->clk);
 
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 14786c8bf99e..45beca17fa50 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -23,6 +23,7 @@
 #include <net/ip.h>
 #include <net/ipv6.h>
 #include <linux/io.h>
+#include <net/tso.h>
 #include <linux/of.h>
 #include <linux/of_irq.h>
 #include <linux/of_mdio.h>
@@ -218,9 +219,6 @@
 #define MVNETA_RX_COAL_PKTS		32
 #define MVNETA_RX_COAL_USEC		100
 
-/* Napi polling weight */
-#define MVNETA_RX_POLL_WEIGHT		64
-
 /* The two bytes Marvell header. Either contains a special value used
  * by Marvell switches when a specific hardware mode is enabled (not
  * supported by this driver) or is filled automatically by zeroes on
@@ -244,12 +242,20 @@
 
 #define MVNETA_TX_MTU_MAX		0x3ffff
 
+/* TSO header size */
+#define TSO_HEADER_SIZE 128
+
 /* Max number of Rx descriptors */
 #define MVNETA_MAX_RXD 128
 
 /* Max number of Tx descriptors */
 #define MVNETA_MAX_TXD 532
 
+/* Max number of allowed TCP segments for software TSO */
+#define MVNETA_MAX_TSO_SEGS 100
+
+#define MVNETA_MAX_SKB_DESCS (MVNETA_MAX_TSO_SEGS * 2 + MAX_SKB_FRAGS)
+
 /* descriptor aligned size */
 #define MVNETA_DESC_ALIGNED_SIZE	32
 
@@ -258,6 +264,10 @@
 	      ETH_HLEN + ETH_FCS_LEN,			     \
 	      MVNETA_CPU_D_CACHE_LINE_SIZE)
 
+#define IS_TSO_HEADER(txq, addr) \
+	((addr >= txq->tso_hdrs_phys) && \
+	 (addr < txq->tso_hdrs_phys + txq->size * TSO_HEADER_SIZE))
+
 #define MVNETA_RX_BUF_SIZE(pkt_size)   ((pkt_size) + NET_SKB_PAD)
 
 struct mvneta_pcpu_stats {
@@ -279,9 +289,6 @@ struct mvneta_port {
 	u32 cause_rx_tx;
 	struct napi_struct napi;
 
-	/* Napi weight */
-	int weight;
-
 	/* Core clock */
 	struct clk *clk;
 	u8 mcast_count[256];
@@ -390,6 +397,8 @@ struct mvneta_tx_queue {
 	 * descriptor ring
 	 */
 	int count;
+	int tx_stop_threshold;
+	int tx_wake_threshold;
 
 	/* Array of transmitted skb */
 	struct sk_buff **tx_skb;
@@ -413,6 +422,12 @@ struct mvneta_tx_queue {
 
 	/* Index of the next TX DMA descriptor to process */
 	int next_desc_to_proc;
+
+	/* DMA buffers for TSO headers */
+	char *tso_hdrs;
+
+	/* DMA address of TSO headers */
+	dma_addr_t tso_hdrs_phys;
 };
 
 struct mvneta_rx_queue {
@@ -441,7 +456,10 @@ struct mvneta_rx_queue {
 	int next_desc_to_proc;
 };
 
-static int rxq_number = 8;
+/* The hardware supports eight (8) rx queues, but we are only allowing
+ * the first one to be used. Therefore, let's just allocate one queue.
+ */
+static int rxq_number = 1;
 static int txq_number = 8;
 
 static int rxq_def;
@@ -1277,11 +1295,12 @@ static void mvneta_txq_bufs_free(struct mvneta_port *pp,
 
 		mvneta_txq_inc_get(txq);
 
+		if (!IS_TSO_HEADER(txq, tx_desc->buf_phys_addr))
+			dma_unmap_single(pp->dev->dev.parent,
+					 tx_desc->buf_phys_addr,
+					 tx_desc->data_size, DMA_TO_DEVICE);
 		if (!skb)
 			continue;
-
-		dma_unmap_single(pp->dev->dev.parent, tx_desc->buf_phys_addr,
-				 tx_desc->data_size, DMA_TO_DEVICE);
 		dev_kfree_skb_any(skb);
 	}
 }
@@ -1302,7 +1321,7 @@ static void mvneta_txq_done(struct mvneta_port *pp,
 	txq->count -= tx_done;
 
 	if (netif_tx_queue_stopped(nq)) {
-		if (txq->size - txq->count >= MAX_SKB_FRAGS + 1)
+		if (txq->count <= txq->tx_wake_threshold)
 			netif_tx_wake_queue(nq);
 	}
 }
@@ -1519,14 +1538,134 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
 	return rx_done;
 }
 
+static inline void
+mvneta_tso_put_hdr(struct sk_buff *skb,
+		   struct mvneta_port *pp, struct mvneta_tx_queue *txq)
+{
+	struct mvneta_tx_desc *tx_desc;
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+
+	txq->tx_skb[txq->txq_put_index] = NULL;
+	tx_desc = mvneta_txq_next_desc_get(txq);
+	tx_desc->data_size = hdr_len;
+	tx_desc->command = mvneta_skb_tx_csum(pp, skb);
+	tx_desc->command |= MVNETA_TXD_F_DESC;
+	tx_desc->buf_phys_addr = txq->tso_hdrs_phys +
+				 txq->txq_put_index * TSO_HEADER_SIZE;
+	mvneta_txq_inc_put(txq);
+}
+
+static inline int
+mvneta_tso_put_data(struct net_device *dev, struct mvneta_tx_queue *txq,
+		    struct sk_buff *skb, char *data, int size,
+		    bool last_tcp, bool is_last)
+{
+	struct mvneta_tx_desc *tx_desc;
+
+	tx_desc = mvneta_txq_next_desc_get(txq);
+	tx_desc->data_size = size;
+	tx_desc->buf_phys_addr = dma_map_single(dev->dev.parent, data,
+						size, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(dev->dev.parent,
+		     tx_desc->buf_phys_addr))) {
+		mvneta_txq_desc_put(txq);
+		return -ENOMEM;
+	}
+
+	tx_desc->command = 0;
+	txq->tx_skb[txq->txq_put_index] = NULL;
+
+	if (last_tcp) {
+		/* last descriptor in the TCP packet */
+		tx_desc->command = MVNETA_TXD_L_DESC;
+
+		/* last descriptor in SKB */
+		if (is_last)
+			txq->tx_skb[txq->txq_put_index] = skb;
+	}
+	mvneta_txq_inc_put(txq);
+	return 0;
+}
+
+static int mvneta_tx_tso(struct sk_buff *skb, struct net_device *dev,
+			 struct mvneta_tx_queue *txq)
+{
+	int total_len, data_left;
+	int desc_count = 0;
+	struct mvneta_port *pp = netdev_priv(dev);
+	struct tso_t tso;
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	int i;
+
+	/* Count needed descriptors */
+	if ((txq->count + tso_count_descs(skb)) >= txq->size)
+		return 0;
+
+	if (skb_headlen(skb) < (skb_transport_offset(skb) + tcp_hdrlen(skb))) {
+		pr_info("*** Is this even  possible???!?!?\n");
+		return 0;
+	}
+
+	/* Initialize the TSO handler, and prepare the first payload */
+	tso_start(skb, &tso);
+
+	total_len = skb->len - hdr_len;
+	while (total_len > 0) {
+		char *hdr;
+
+		data_left = min_t(int, skb_shinfo(skb)->gso_size, total_len);
+		total_len -= data_left;
+		desc_count++;
+
+		/* prepare packet headers: MAC + IP + TCP */
+		hdr = txq->tso_hdrs + txq->txq_put_index * TSO_HEADER_SIZE;
+		tso_build_hdr(skb, hdr, &tso, data_left, total_len == 0);
+
+		mvneta_tso_put_hdr(skb, pp, txq);
+
+		while (data_left > 0) {
+			int size;
+			desc_count++;
+
+			size = min_t(int, tso.size, data_left);
+
+			if (mvneta_tso_put_data(dev, txq, skb,
+						 tso.data, size,
+						 size == data_left,
+						 total_len == 0))
+				goto err_release;
+			data_left -= size;
+
+			tso_build_data(skb, &tso, size);
+		}
+	}
+
+	return desc_count;
+
+err_release:
+	/* Release all used data descriptors; header descriptors must not
+	 * be DMA-unmapped.
+	 */
+	for (i = desc_count - 1; i >= 0; i--) {
+		struct mvneta_tx_desc *tx_desc = txq->descs + i;
+		if (!IS_TSO_HEADER(txq, tx_desc->buf_phys_addr))
+			dma_unmap_single(pp->dev->dev.parent,
+					 tx_desc->buf_phys_addr,
+					 tx_desc->data_size,
+					 DMA_TO_DEVICE);
+		mvneta_txq_desc_put(txq);
+	}
+	return 0;
+}
+
 /* Handle tx fragmentation processing */
 static int mvneta_tx_frag_process(struct mvneta_port *pp, struct sk_buff *skb,
 				  struct mvneta_tx_queue *txq)
 {
 	struct mvneta_tx_desc *tx_desc;
-	int i;
+	int i, nr_frags = skb_shinfo(skb)->nr_frags;
 
-	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+	for (i = 0; i < nr_frags; i++) {
 		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		void *addr = page_address(frag->page.p) + frag->page_offset;
 
@@ -1543,20 +1682,16 @@ static int mvneta_tx_frag_process(struct mvneta_port *pp, struct sk_buff *skb,
 			goto error;
 		}
 
-		if (i == (skb_shinfo(skb)->nr_frags - 1)) {
+		if (i == nr_frags - 1) {
 			/* Last descriptor */
 			tx_desc->command = MVNETA_TXD_L_DESC | MVNETA_TXD_Z_PAD;
-
 			txq->tx_skb[txq->txq_put_index] = skb;
-
-			mvneta_txq_inc_put(txq);
 		} else {
 			/* Descriptor in the middle: Not First, Not Last */
 			tx_desc->command = 0;
-
 			txq->tx_skb[txq->txq_put_index] = NULL;
-			mvneta_txq_inc_put(txq);
 		}
+		mvneta_txq_inc_put(txq);
 	}
 
 	return 0;
@@ -1584,15 +1719,18 @@ static int mvneta_tx(struct sk_buff *skb, struct net_device *dev)
 	u16 txq_id = skb_get_queue_mapping(skb);
 	struct mvneta_tx_queue *txq = &pp->txqs[txq_id];
 	struct mvneta_tx_desc *tx_desc;
-	struct netdev_queue *nq;
 	int frags = 0;
 	u32 tx_cmd;
 
 	if (!netif_running(dev))
 		goto out;
 
+	if (skb_is_gso(skb)) {
+		frags = mvneta_tx_tso(skb, dev, txq);
+		goto out;
+	}
+
 	frags = skb_shinfo(skb)->nr_frags + 1;
-	nq    = netdev_get_tx_queue(dev, txq_id);
 
 	/* Get a descriptor for the first part of the packet */
 	tx_desc = mvneta_txq_next_desc_get(txq);
@@ -1635,15 +1773,16 @@ static int mvneta_tx(struct sk_buff *skb, struct net_device *dev)
 		}
 	}
 
-	txq->count += frags;
-	mvneta_txq_pend_desc_add(pp, txq, frags);
-
-	if (txq->size - txq->count < MAX_SKB_FRAGS + 1)
-		netif_tx_stop_queue(nq);
-
 out:
 	if (frags > 0) {
 		struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats);
+		struct netdev_queue *nq = netdev_get_tx_queue(dev, txq_id);
+
+		txq->count += frags;
+		mvneta_txq_pend_desc_add(pp, txq, frags);
+
+		if (txq->count >= txq->tx_stop_threshold)
+			netif_tx_stop_queue(nq);
 
 		u64_stats_update_begin(&stats->syncp);
 		stats->tx_packets++;
@@ -2003,7 +2142,7 @@ static void mvneta_tx_reset(struct mvneta_port *pp)
 {
 	int queue;
 
-	/* free the skb's in the hal tx ring */
+	/* free the skb's in the tx ring */
 	for (queue = 0; queue < txq_number; queue++)
 		mvneta_txq_done_force(pp, &pp->txqs[queue]);
 
@@ -2081,6 +2220,14 @@ static int mvneta_txq_init(struct mvneta_port *pp,
 {
 	txq->size = pp->tx_ring_size;
 
+	/* A queue must always have room for at least one skb.
+	 * Therefore, stop the queue when the free entries reaches
+	 * the maximum number of descriptors per skb.
+	 */
+	txq->tx_stop_threshold = txq->size - MVNETA_MAX_SKB_DESCS;
+	txq->tx_wake_threshold = txq->tx_stop_threshold / 2;
+
+
 	/* Allocate memory for TX descriptors */
 	txq->descs = dma_alloc_coherent(pp->dev->dev.parent,
 					txq->size * MVNETA_DESC_ALIGNED_SIZE,
@@ -2109,6 +2256,18 @@ static int mvneta_txq_init(struct mvneta_port *pp,
 				  txq->descs, txq->descs_phys);
 		return -ENOMEM;
 	}
+
+	/* Allocate DMA buffers for TSO MAC/IP/TCP headers */
+	txq->tso_hdrs = dma_alloc_coherent(pp->dev->dev.parent,
+					   txq->size * TSO_HEADER_SIZE,
+					   &txq->tso_hdrs_phys, GFP_KERNEL);
+	if (txq->tso_hdrs == NULL) {
+		kfree(txq->tx_skb);
+		dma_free_coherent(pp->dev->dev.parent,
+				  txq->size * MVNETA_DESC_ALIGNED_SIZE,
+				  txq->descs, txq->descs_phys);
+		return -ENOMEM;
+	}
 	mvneta_tx_done_pkts_coal_set(pp, txq, txq->done_pkts_coal);
 
 	return 0;
@@ -2120,6 +2279,10 @@ static void mvneta_txq_deinit(struct mvneta_port *pp,
 {
 	kfree(txq->tx_skb);
 
+	if (txq->tso_hdrs)
+		dma_free_coherent(pp->dev->dev.parent,
+				  txq->size * TSO_HEADER_SIZE,
+				  txq->tso_hdrs, txq->tso_hdrs_phys);
 	if (txq->descs)
 		dma_free_coherent(pp->dev->dev.parent,
 				  txq->size * MVNETA_DESC_ALIGNED_SIZE,
@@ -2279,24 +2442,28 @@ static int mvneta_change_mtu(struct net_device *dev, int mtu)
 		return 0;
 
 	/* The interface is running, so we have to force a
-	 * reallocation of the RXQs
+	 * reallocation of the queues
 	 */
 	mvneta_stop_dev(pp);
 
 	mvneta_cleanup_txqs(pp);
 	mvneta_cleanup_rxqs(pp);
 
-	pp->pkt_size = MVNETA_RX_PKT_SIZE(pp->dev->mtu);
+	pp->pkt_size = MVNETA_RX_PKT_SIZE(dev->mtu);
 	pp->frag_size = SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(pp->pkt_size)) +
 	                SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 
 	ret = mvneta_setup_rxqs(pp);
 	if (ret) {
-		netdev_err(pp->dev, "unable to setup rxqs after MTU change\n");
+		netdev_err(dev, "unable to setup rxqs after MTU change\n");
 		return ret;
 	}
 
-	mvneta_setup_txqs(pp);
+	ret = mvneta_setup_txqs(pp);
+	if (ret) {
+		netdev_err(dev, "unable to setup txqs after MTU change\n");
+		return ret;
+	}
 
 	mvneta_start_dev(pp);
 	mvneta_port_up(pp);
@@ -2323,22 +2490,19 @@ static void mvneta_get_mac_addr(struct mvneta_port *pp, unsigned char *addr)
 static int mvneta_set_mac_addr(struct net_device *dev, void *addr)
 {
 	struct mvneta_port *pp = netdev_priv(dev);
-	u8 *mac = addr + 2;
-	int i;
-
-	if (netif_running(dev))
-		return -EBUSY;
+	struct sockaddr *sockaddr = addr;
+	int ret;
 
+	ret = eth_prepare_mac_addr_change(dev, addr);
+	if (ret < 0)
+		return ret;
 	/* Remove previous address table entry */
 	mvneta_mac_addr_set(pp, dev->dev_addr, -1);
 
 	/* Set new addr in hw */
-	mvneta_mac_addr_set(pp, mac, rxq_def);
-
-	/* Set addr in the device */
-	for (i = 0; i < ETH_ALEN; i++)
-		dev->dev_addr[i] = mac[i];
+	mvneta_mac_addr_set(pp, sockaddr->sa_data, rxq_def);
 
+	eth_commit_mac_addr_change(dev, addr);
 	return 0;
 }
 
@@ -2433,8 +2597,6 @@ static int mvneta_open(struct net_device *dev)
 	struct mvneta_port *pp = netdev_priv(dev);
 	int ret;
 
-	mvneta_mac_addr_set(pp, dev->dev_addr, rxq_def);
-
 	pp->pkt_size = MVNETA_RX_PKT_SIZE(pp->dev->mtu);
 	pp->frag_size = SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(pp->pkt_size)) +
 	                SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
@@ -2600,8 +2762,12 @@ static int mvneta_ethtool_set_ringparam(struct net_device *dev,
 		return -EINVAL;
 	pp->rx_ring_size = ring->rx_pending < MVNETA_MAX_RXD ?
 		ring->rx_pending : MVNETA_MAX_RXD;
-	pp->tx_ring_size = ring->tx_pending < MVNETA_MAX_TXD ?
-		ring->tx_pending : MVNETA_MAX_TXD;
+
+	pp->tx_ring_size = clamp_t(u16, ring->tx_pending,
+				   MVNETA_MAX_SKB_DESCS * 2, MVNETA_MAX_TXD);
+	if (pp->tx_ring_size != ring->tx_pending)
+		netdev_warn(dev, "TX queue size set to %u (requested %u)\n",
+			    pp->tx_ring_size, ring->tx_pending);
 
 	if (netif_running(dev)) {
 		mvneta_stop(dev);
@@ -2638,7 +2804,7 @@ const struct ethtool_ops mvneta_eth_tool_ops = {
 };
 
 /* Initialize hw */
-static int mvneta_init(struct mvneta_port *pp, int phy_addr)
+static int mvneta_init(struct device *dev, struct mvneta_port *pp)
 {
 	int queue;
 
@@ -2648,8 +2814,8 @@ static int mvneta_init(struct mvneta_port *pp, int phy_addr)
 	/* Set port default values */
 	mvneta_defaults_set(pp);
 
-	pp->txqs = kzalloc(txq_number * sizeof(struct mvneta_tx_queue),
-			   GFP_KERNEL);
+	pp->txqs = devm_kcalloc(dev, txq_number, sizeof(struct mvneta_tx_queue),
+				GFP_KERNEL);
 	if (!pp->txqs)
 		return -ENOMEM;
 
@@ -2661,12 +2827,10 @@ static int mvneta_init(struct mvneta_port *pp, int phy_addr)
 		txq->done_pkts_coal = MVNETA_TXDONE_COAL_PKTS;
 	}
 
-	pp->rxqs = kzalloc(rxq_number * sizeof(struct mvneta_rx_queue),
-			   GFP_KERNEL);
-	if (!pp->rxqs) {
-		kfree(pp->txqs);
+	pp->rxqs = devm_kcalloc(dev, rxq_number, sizeof(struct mvneta_rx_queue),
+				GFP_KERNEL);
+	if (!pp->rxqs)
 		return -ENOMEM;
-	}
 
 	/* Create Rx descriptor rings */
 	for (queue = 0; queue < rxq_number; queue++) {
@@ -2680,12 +2844,6 @@ static int mvneta_init(struct mvneta_port *pp, int phy_addr)
 	return 0;
 }
 
-static void mvneta_deinit(struct mvneta_port *pp)
-{
-	kfree(pp->txqs);
-	kfree(pp->rxqs);
-}
-
 /* platform glue : initialize decoding windows */
 static void mvneta_conf_mbus_windows(struct mvneta_port *pp,
 				     const struct mbus_dram_target_info *dram)
@@ -2768,7 +2926,6 @@ static int mvneta_probe(struct platform_device *pdev)
 	struct resource *res;
 	struct device_node *dn = pdev->dev.of_node;
 	struct device_node *phy_node;
-	u32 phy_addr;
 	struct mvneta_port *pp;
 	struct net_device *dev;
 	const char *dt_mac_addr;
@@ -2797,9 +2954,22 @@ static int mvneta_probe(struct platform_device *pdev)
 
 	phy_node = of_parse_phandle(dn, "phy", 0);
 	if (!phy_node) {
-		dev_err(&pdev->dev, "no associated PHY\n");
-		err = -ENODEV;
-		goto err_free_irq;
+		if (!of_phy_is_fixed_link(dn)) {
+			dev_err(&pdev->dev, "no PHY specified\n");
+			err = -ENODEV;
+			goto err_free_irq;
+		}
+
+		err = of_phy_register_fixed_link(dn);
+		if (err < 0) {
+			dev_err(&pdev->dev, "cannot register fixed PHY\n");
+			goto err_free_irq;
+		}
+
+		/* In the case of a fixed PHY, the DT node associated
+		 * to the PHY is the Ethernet MAC DT node.
+		 */
+		phy_node = dn;
 	}
 
 	phy_mode = of_get_phy_mode(dn);
@@ -2813,11 +2983,9 @@ static int mvneta_probe(struct platform_device *pdev)
 	dev->watchdog_timeo = 5 * HZ;
 	dev->netdev_ops = &mvneta_netdev_ops;
 
-	SET_ETHTOOL_OPS(dev, &mvneta_eth_tool_ops);
+	dev->ethtool_ops = &mvneta_eth_tool_ops;
 
 	pp = netdev_priv(dev);
-
-	pp->weight = MVNETA_RX_POLL_WEIGHT;
 	pp->phy_node = phy_node;
 	pp->phy_interface = phy_mode;
 
@@ -2864,33 +3032,32 @@ static int mvneta_probe(struct platform_device *pdev)
 	pp->dev = dev;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
-	err = mvneta_init(pp, phy_addr);
-	if (err < 0) {
-		dev_err(&pdev->dev, "can't init eth hal\n");
+	err = mvneta_init(&pdev->dev, pp);
+	if (err < 0)
 		goto err_free_stats;
-	}
 
 	err = mvneta_port_power_up(pp, phy_mode);
 	if (err < 0) {
 		dev_err(&pdev->dev, "can't power up port\n");
-		goto err_deinit;
+		goto err_free_stats;
 	}
 
 	dram_target_info = mv_mbus_dram_info();
 	if (dram_target_info)
 		mvneta_conf_mbus_windows(pp, dram_target_info);
 
-	netif_napi_add(dev, &pp->napi, mvneta_poll, pp->weight);
+	netif_napi_add(dev, &pp->napi, mvneta_poll, NAPI_POLL_WEIGHT);
 
-	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM;
-	dev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM;
-	dev->vlan_features |= NETIF_F_SG | NETIF_F_IP_CSUM;
+	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO;
+	dev->hw_features |= dev->features;
+	dev->vlan_features |= dev->features;
 	dev->priv_flags |= IFF_UNICAST_FLT;
+	dev->gso_max_segs = MVNETA_MAX_TSO_SEGS;
 
 	err = register_netdev(dev);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register\n");
-		goto err_deinit;
+		goto err_free_stats;
 	}
 
 	netdev_info(dev, "Using %s mac address %pM\n", mac_from,
@@ -2900,8 +3067,6 @@ static int mvneta_probe(struct platform_device *pdev)
 
 	return 0;
 
-err_deinit:
-	mvneta_deinit(pp);
 err_free_stats:
 	free_percpu(pp->stats);
 err_clk:
@@ -2920,7 +3085,6 @@ static int mvneta_remove(struct platform_device *pdev)
 	struct mvneta_port *pp = netdev_priv(dev);
 
 	unregister_netdev(dev);
-	mvneta_deinit(pp);
 	clk_disable_unprepare(pp->clk);
 	free_percpu(pp->stats);
 	irq_dispose_mapping(dev->irq);
diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c
index b358c2f6f4bd..8f5aa7c62b18 100644
--- a/drivers/net/ethernet/marvell/pxa168_eth.c
+++ b/drivers/net/ethernet/marvell/pxa168_eth.c
@@ -1488,7 +1488,7 @@ static int pxa168_eth_probe(struct platform_device *pdev)
 	dev->netdev_ops = &pxa168_eth_netdev_ops;
 	dev->watchdog_timeo = 2 * HZ;
 	dev->base_addr = 0;
-	SET_ETHTOOL_OPS(dev, &pxa168_ethtool_ops);
+	dev->ethtool_ops = &pxa168_ethtool_ops;
 
 	INIT_WORK(&pep->tx_timeout_task, pxa168_eth_tx_timeout_task);
 
diff --git a/drivers/net/ethernet/marvell/skge.c b/drivers/net/ethernet/marvell/skge.c
index 7f81ae66cc89..e912b6887d40 100644
--- a/drivers/net/ethernet/marvell/skge.c
+++ b/drivers/net/ethernet/marvell/skge.c
@@ -4199,6 +4199,13 @@ static struct dmi_system_id skge_32bit_dma_boards[] = {
 			DMI_MATCH(DMI_BOARD_NAME, "P5NSLI")
 		},
 	},
+	{
+		.ident = "FUJITSU SIEMENS A8NE-FM",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTek Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "A8NE-FM")
+		},
+	},
 	{}
 };
 
diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c
index b81106451a0a..69693384b58c 100644
--- a/drivers/net/ethernet/marvell/sky2.c
+++ b/drivers/net/ethernet/marvell/sky2.c
@@ -4760,7 +4760,7 @@ static struct net_device *sky2_init_netdev(struct sky2_hw *hw, unsigned port,
 
 	SET_NETDEV_DEV(dev, &hw->pdev->dev);
 	dev->irq = hw->pdev->irq;
-	SET_ETHTOOL_OPS(dev, &sky2_ethtool_ops);
+	dev->ethtool_ops = &sky2_ethtool_ops;
 	dev->watchdog_timeo = TX_WATCHDOG;
 	dev->netdev_ops = &sky2_netdev_ops[port];
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/alloc.c b/drivers/net/ethernet/mellanox/mlx4/alloc.c
index c3ad464d0627..b0297da50304 100644
--- a/drivers/net/ethernet/mellanox/mlx4/alloc.c
+++ b/drivers/net/ethernet/mellanox/mlx4/alloc.c
@@ -171,7 +171,7 @@ void mlx4_bitmap_cleanup(struct mlx4_bitmap *bitmap)
  */
 
 int mlx4_buf_alloc(struct mlx4_dev *dev, int size, int max_direct,
-		   struct mlx4_buf *buf)
+		   struct mlx4_buf *buf, gfp_t gfp)
 {
 	dma_addr_t t;
 
@@ -180,7 +180,7 @@ int mlx4_buf_alloc(struct mlx4_dev *dev, int size, int max_direct,
 		buf->npages       = 1;
 		buf->page_shift   = get_order(size) + PAGE_SHIFT;
 		buf->direct.buf   = dma_alloc_coherent(&dev->pdev->dev,
-						       size, &t, GFP_KERNEL);
+						       size, &t, gfp);
 		if (!buf->direct.buf)
 			return -ENOMEM;
 
@@ -200,14 +200,14 @@ int mlx4_buf_alloc(struct mlx4_dev *dev, int size, int max_direct,
 		buf->npages      = buf->nbufs;
 		buf->page_shift  = PAGE_SHIFT;
 		buf->page_list   = kcalloc(buf->nbufs, sizeof(*buf->page_list),
-					   GFP_KERNEL);
+					   gfp);
 		if (!buf->page_list)
 			return -ENOMEM;
 
 		for (i = 0; i < buf->nbufs; ++i) {
 			buf->page_list[i].buf =
 				dma_alloc_coherent(&dev->pdev->dev, PAGE_SIZE,
-						   &t, GFP_KERNEL);
+						   &t, gfp);
 			if (!buf->page_list[i].buf)
 				goto err_free;
 
@@ -218,7 +218,7 @@ int mlx4_buf_alloc(struct mlx4_dev *dev, int size, int max_direct,
 
 		if (BITS_PER_LONG == 64) {
 			struct page **pages;
-			pages = kmalloc(sizeof *pages * buf->nbufs, GFP_KERNEL);
+			pages = kmalloc(sizeof *pages * buf->nbufs, gfp);
 			if (!pages)
 				goto err_free;
 			for (i = 0; i < buf->nbufs; ++i)
@@ -260,11 +260,12 @@ void mlx4_buf_free(struct mlx4_dev *dev, int size, struct mlx4_buf *buf)
 }
 EXPORT_SYMBOL_GPL(mlx4_buf_free);
 
-static struct mlx4_db_pgdir *mlx4_alloc_db_pgdir(struct device *dma_device)
+static struct mlx4_db_pgdir *mlx4_alloc_db_pgdir(struct device *dma_device,
+						 gfp_t gfp)
 {
 	struct mlx4_db_pgdir *pgdir;
 
-	pgdir = kzalloc(sizeof *pgdir, GFP_KERNEL);
+	pgdir = kzalloc(sizeof *pgdir, gfp);
 	if (!pgdir)
 		return NULL;
 
@@ -272,7 +273,7 @@ static struct mlx4_db_pgdir *mlx4_alloc_db_pgdir(struct device *dma_device)
 	pgdir->bits[0] = pgdir->order0;
 	pgdir->bits[1] = pgdir->order1;
 	pgdir->db_page = dma_alloc_coherent(dma_device, PAGE_SIZE,
-					    &pgdir->db_dma, GFP_KERNEL);
+					    &pgdir->db_dma, gfp);
 	if (!pgdir->db_page) {
 		kfree(pgdir);
 		return NULL;
@@ -312,7 +313,7 @@ found:
 	return 0;
 }
 
-int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, int order)
+int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, int order, gfp_t gfp)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_db_pgdir *pgdir;
@@ -324,7 +325,7 @@ int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, int order)
 		if (!mlx4_alloc_db_from_pgdir(pgdir, db, order))
 			goto out;
 
-	pgdir = mlx4_alloc_db_pgdir(&(dev->pdev->dev));
+	pgdir = mlx4_alloc_db_pgdir(&(dev->pdev->dev), gfp);
 	if (!pgdir) {
 		ret = -ENOMEM;
 		goto out;
@@ -376,13 +377,13 @@ int mlx4_alloc_hwq_res(struct mlx4_dev *dev, struct mlx4_hwq_resources *wqres,
 {
 	int err;
 
-	err = mlx4_db_alloc(dev, &wqres->db, 1);
+	err = mlx4_db_alloc(dev, &wqres->db, 1, GFP_KERNEL);
 	if (err)
 		return err;
 
 	*wqres->db.db = 0;
 
-	err = mlx4_buf_alloc(dev, size, max_direct, &wqres->buf);
+	err = mlx4_buf_alloc(dev, size, max_direct, &wqres->buf, GFP_KERNEL);
 	if (err)
 		goto err_db;
 
@@ -391,7 +392,7 @@ int mlx4_alloc_hwq_res(struct mlx4_dev *dev, struct mlx4_hwq_resources *wqres,
 	if (err)
 		goto err_buf;
 
-	err = mlx4_buf_write_mtt(dev, &wqres->mtt, &wqres->buf);
+	err = mlx4_buf_write_mtt(dev, &wqres->mtt, &wqres->buf, GFP_KERNEL);
 	if (err)
 		goto err_mtt;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index 92d3249f63f1..5d940a26055c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -212,8 +212,7 @@ static int mlx4_comm_cmd_poll(struct mlx4_dev *dev, u8 cmd, u16 param,
 
 	/* First, verify that the master reports correct status */
 	if (comm_pending(dev)) {
-		mlx4_warn(dev, "Communication channel is not idle."
-			  "my toggle is %d (cmd:0x%x)\n",
+		mlx4_warn(dev, "Communication channel is not idle - my toggle is %d (cmd:0x%x)\n",
 			  priv->cmd.comm_toggle, cmd);
 		return -EAGAIN;
 	}
@@ -422,9 +421,8 @@ static int mlx4_slave_cmd(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
 					*out_param =
 						be64_to_cpu(vhcr->out_param);
 				else {
-					mlx4_err(dev, "response expected while"
-						 "output mailbox is NULL for "
-						 "command 0x%x\n", op);
+					mlx4_err(dev, "response expected while output mailbox is NULL for command 0x%x\n",
+						 op);
 					vhcr->status = CMD_STAT_BAD_PARAM;
 				}
 			}
@@ -439,16 +437,15 @@ static int mlx4_slave_cmd(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
 					*out_param =
 						be64_to_cpu(vhcr->out_param);
 				else {
-					mlx4_err(dev, "response expected while"
-						 "output mailbox is NULL for "
-						 "command 0x%x\n", op);
+					mlx4_err(dev, "response expected while output mailbox is NULL for command 0x%x\n",
+						 op);
 					vhcr->status = CMD_STAT_BAD_PARAM;
 				}
 			}
 			ret = mlx4_status_to_errno(vhcr->status);
 		} else
-			mlx4_err(dev, "failed execution of VHCR_POST command"
-				 "opcode 0x%x\n", op);
+			mlx4_err(dev, "failed execution of VHCR_POST command opcode 0x%x\n",
+				 op);
 	}
 
 	mutex_unlock(&priv->cmd.slave_cmd_mutex);
@@ -476,6 +473,13 @@ static int mlx4_cmd_poll(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
 		goto out;
 	}
 
+	if (out_is_imm && !out_param) {
+		mlx4_err(dev, "response expected while output mailbox is NULL for command 0x%x\n",
+			 op);
+		err = -EINVAL;
+		goto out;
+	}
+
 	err = mlx4_cmd_post(dev, in_param, out_param ? *out_param : 0,
 			    in_modifier, op_modifier, op, CMD_POLL_TOKEN, 0);
 	if (err)
@@ -554,6 +558,13 @@ static int mlx4_cmd_wait(struct mlx4_dev *dev, u64 in_param, u64 *out_param,
 	cmd->free_head = context->next;
 	spin_unlock(&cmd->context_lock);
 
+	if (out_is_imm && !out_param) {
+		mlx4_err(dev, "response expected while output mailbox is NULL for command 0x%x\n",
+			 op);
+		err = -EINVAL;
+		goto out;
+	}
+
 	init_completion(&context->done);
 
 	mlx4_cmd_post(dev, in_param, out_param ? *out_param : 0,
@@ -625,9 +636,8 @@ static int mlx4_ACCESS_MEM(struct mlx4_dev *dev, u64 master_addr,
 
 	if ((slave_addr & 0xfff) | (master_addr & 0xfff) |
 	    (slave & ~0x7f) | (size & 0xff)) {
-		mlx4_err(dev, "Bad access mem params - slave_addr:0x%llx "
-			      "master_addr:0x%llx slave_id:%d size:%d\n",
-			      slave_addr, master_addr, slave, size);
+		mlx4_err(dev, "Bad access mem params - slave_addr:0x%llx master_addr:0x%llx slave_id:%d size:%d\n",
+			 slave_addr, master_addr, slave, size);
 		return -EINVAL;
 	}
 
@@ -705,20 +715,28 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave,
 	struct ib_smp *smp = inbox->buf;
 	u32 index;
 	u8 port;
+	u8 opcode_modifier;
 	u16 *table;
 	int err;
 	int vidx, pidx;
+	int network_view;
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct ib_smp *outsmp = outbox->buf;
 	__be16 *outtab = (__be16 *)(outsmp->data);
 	__be32 slave_cap_mask;
 	__be64 slave_node_guid;
+
 	port = vhcr->in_modifier;
 
+	/* network-view bit is for driver use only, and should not be passed to FW */
+	opcode_modifier = vhcr->op_modifier & ~0x8; /* clear netw view bit */
+	network_view = !!(vhcr->op_modifier & 0x8);
+
 	if (smp->base_version == 1 &&
 	    smp->mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED &&
 	    smp->class_version == 1) {
-		if (smp->method	== IB_MGMT_METHOD_GET) {
+		/* host view is paravirtualized */
+		if (!network_view && smp->method == IB_MGMT_METHOD_GET) {
 			if (smp->attr_id == IB_SMP_ATTR_PKEY_TABLE) {
 				index = be32_to_cpu(smp->attr_mod);
 				if (port < 1 || port > dev->caps.num_ports)
@@ -743,7 +761,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave,
 				/*get the slave specific caps:*/
 				/*do the command */
 				err = mlx4_cmd_box(dev, inbox->dma, outbox->dma,
-					    vhcr->in_modifier, vhcr->op_modifier,
+					    vhcr->in_modifier, opcode_modifier,
 					    vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE);
 				/* modify the response for slaves */
 				if (!err && slave != mlx4_master_func_num(dev)) {
@@ -760,7 +778,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave,
 				smp->attr_mod = cpu_to_be32(slave / 8);
 				/* execute cmd */
 				err = mlx4_cmd_box(dev, inbox->dma, outbox->dma,
-					     vhcr->in_modifier, vhcr->op_modifier,
+					     vhcr->in_modifier, opcode_modifier,
 					     vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE);
 				if (!err) {
 					/* if needed, move slave gid to index 0 */
@@ -774,7 +792,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave,
 			}
 			if (smp->attr_id == IB_SMP_ATTR_NODE_INFO) {
 				err = mlx4_cmd_box(dev, inbox->dma, outbox->dma,
-					     vhcr->in_modifier, vhcr->op_modifier,
+					     vhcr->in_modifier, opcode_modifier,
 					     vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE);
 				if (!err) {
 					slave_node_guid =  mlx4_get_slave_node_guid(dev, slave);
@@ -784,19 +802,24 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave,
 			}
 		}
 	}
+
+	/* Non-privileged VFs are only allowed "host" view LID-routed 'Get' MADs.
+	 * These are the MADs used by ib verbs (such as ib_query_gids).
+	 */
 	if (slave != mlx4_master_func_num(dev) &&
-	    ((smp->mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) ||
-	     (smp->mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED &&
-	      smp->method == IB_MGMT_METHOD_SET))) {
-		mlx4_err(dev, "slave %d is trying to execute a Subnet MGMT MAD, "
-			 "class 0x%x, method 0x%x for attr 0x%x. Rejecting\n",
-			 slave, smp->method, smp->mgmt_class,
-			 be16_to_cpu(smp->attr_id));
-		return -EPERM;
+	    !mlx4_vf_smi_enabled(dev, slave, port)) {
+		if (!(smp->mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED &&
+		      smp->method == IB_MGMT_METHOD_GET) || network_view) {
+			mlx4_err(dev, "Unprivileged slave %d is trying to execute a Subnet MGMT MAD, class 0x%x, method 0x%x, view=%s for attr 0x%x. Rejecting\n",
+				 slave, smp->method, smp->mgmt_class,
+				 network_view ? "Network" : "Host",
+				 be16_to_cpu(smp->attr_id));
+			return -EPERM;
+		}
 	}
-	/*default:*/
+
 	return mlx4_cmd_box(dev, inbox->dma, outbox->dma,
-				    vhcr->in_modifier, vhcr->op_modifier,
+				    vhcr->in_modifier, opcode_modifier,
 				    vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE);
 }
 
@@ -1409,8 +1432,8 @@ static int mlx4_master_process_vhcr(struct mlx4_dev *dev, int slave,
 				      ALIGN(sizeof(struct mlx4_vhcr_cmd),
 					    MLX4_ACCESS_MEM_ALIGN), 1);
 		if (ret) {
-			mlx4_err(dev, "%s:Failed reading vhcr"
-				 "ret: 0x%x\n", __func__, ret);
+			mlx4_err(dev, "%s: Failed reading vhcr ret: 0x%x\n",
+				 __func__, ret);
 			kfree(vhcr);
 			return ret;
 		}
@@ -1461,9 +1484,8 @@ static int mlx4_master_process_vhcr(struct mlx4_dev *dev, int slave,
 
 	/* Apply permission and bound checks if applicable */
 	if (cmd->verify && cmd->verify(dev, slave, vhcr, inbox)) {
-		mlx4_warn(dev, "Command:0x%x from slave: %d failed protection "
-			  "checks for resource_id:%d\n", vhcr->op, slave,
-			  vhcr->in_modifier);
+		mlx4_warn(dev, "Command:0x%x from slave: %d failed protection checks for resource_id:%d\n",
+			  vhcr->op, slave, vhcr->in_modifier);
 		vhcr_cmd->status = CMD_STAT_BAD_OP;
 		goto out_status;
 	}
@@ -1502,8 +1524,7 @@ static int mlx4_master_process_vhcr(struct mlx4_dev *dev, int slave,
 	}
 
 	if (err) {
-		mlx4_warn(dev, "vhcr command:0x%x slave:%d failed with"
-			  " error:%d, status %d\n",
+		mlx4_warn(dev, "vhcr command:0x%x slave:%d failed with error:%d, status %d\n",
 			  vhcr->op, slave, vhcr->errno, err);
 		vhcr_cmd->status = mlx4_errno_to_status(err);
 		goto out_status;
@@ -1537,8 +1558,8 @@ out_status:
 				 __func__);
 		else if (vhcr->e_bit &&
 			 mlx4_GEN_EQE(dev, slave, &priv->mfunc.master.cmd_eqe))
-				mlx4_warn(dev, "Failed to generate command completion "
-					  "eqe for slave %d\n", slave);
+				mlx4_warn(dev, "Failed to generate command completion eqe for slave %d\n",
+					  slave);
 	}
 
 out:
@@ -1577,8 +1598,9 @@ static int mlx4_master_immediate_activate_vlan_qos(struct mlx4_priv *priv,
 
 	mlx4_dbg(dev, "updating immediately admin params slave %d port %d\n",
 		 slave, port);
-	mlx4_dbg(dev, "vlan %d QoS %d link down %d\n", vp_admin->default_vlan,
-		 vp_admin->default_qos, vp_admin->link_state);
+	mlx4_dbg(dev, "vlan %d QoS %d link down %d\n",
+		 vp_admin->default_vlan, vp_admin->default_qos,
+		 vp_admin->link_state);
 
 	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (!work)
@@ -1591,7 +1613,7 @@ static int mlx4_master_immediate_activate_vlan_qos(struct mlx4_priv *priv,
 						   &admin_vlan_ix);
 			if (err) {
 				kfree(work);
-				mlx4_warn((&priv->dev),
+				mlx4_warn(&priv->dev,
 					  "No vlan resources slave %d, port %d\n",
 					  slave, port);
 				return err;
@@ -1600,7 +1622,7 @@ static int mlx4_master_immediate_activate_vlan_qos(struct mlx4_priv *priv,
 			admin_vlan_ix = NO_INDX;
 		}
 		work->flags |= MLX4_VF_IMMED_VLAN_FLAG_VLAN;
-		mlx4_dbg((&(priv->dev)),
+		mlx4_dbg(&priv->dev,
 			 "alloc vlan %d idx  %d slave %d port %d\n",
 			 (int)(vp_admin->default_vlan),
 			 admin_vlan_ix, slave, port);
@@ -1653,6 +1675,8 @@ static int mlx4_master_activate_admin_state(struct mlx4_priv *priv, int slave)
 	for (port = min_port; port <= max_port; port++) {
 		if (!test_bit(port - 1, actv_ports.ports))
 			continue;
+		priv->mfunc.master.vf_oper[slave].smi_enabled[port] =
+			priv->mfunc.master.vf_admin[slave].enable_smi[port];
 		vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
 		vp_admin = &priv->mfunc.master.vf_admin[slave].vport[port];
 		vp_oper->state = *vp_admin;
@@ -1661,12 +1685,12 @@ static int mlx4_master_activate_admin_state(struct mlx4_priv *priv, int slave)
 						   vp_admin->default_vlan, &(vp_oper->vlan_idx));
 			if (err) {
 				vp_oper->vlan_idx = NO_INDX;
-				mlx4_warn((&priv->dev),
+				mlx4_warn(&priv->dev,
 					  "No vlan resorces slave %d, port %d\n",
 					  slave, port);
 				return err;
 			}
-			mlx4_dbg((&(priv->dev)), "alloc vlan %d idx  %d slave %d port %d\n",
+			mlx4_dbg(&priv->dev, "alloc vlan %d idx  %d slave %d port %d\n",
 				 (int)(vp_oper->state.default_vlan),
 				 vp_oper->vlan_idx, slave, port);
 		}
@@ -1677,12 +1701,12 @@ static int mlx4_master_activate_admin_state(struct mlx4_priv *priv, int slave)
 			if (0 > vp_oper->mac_idx) {
 				err = vp_oper->mac_idx;
 				vp_oper->mac_idx = NO_INDX;
-				mlx4_warn((&priv->dev),
+				mlx4_warn(&priv->dev,
 					  "No mac resorces slave %d, port %d\n",
 					  slave, port);
 				return err;
 			}
-			mlx4_dbg((&(priv->dev)), "alloc mac %llx idx  %d slave %d port %d\n",
+			mlx4_dbg(&priv->dev, "alloc mac %llx idx  %d slave %d port %d\n",
 				 vp_oper->state.mac, vp_oper->mac_idx, slave, port);
 		}
 	}
@@ -1704,6 +1728,8 @@ static void mlx4_master_deactivate_admin_state(struct mlx4_priv *priv, int slave
 	for (port = min_port; port <= max_port; port++) {
 		if (!test_bit(port - 1, actv_ports.ports))
 			continue;
+		priv->mfunc.master.vf_oper[slave].smi_enabled[port] =
+			MLX4_VF_SMI_DISABLED;
 		vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
 		if (NO_INDX != vp_oper->vlan_idx) {
 			__mlx4_unregister_vlan(&priv->dev,
@@ -1731,8 +1757,8 @@ static void mlx4_master_do_cmd(struct mlx4_dev *dev, int slave, u8 cmd,
 	slave_state[slave].comm_toggle ^= 1;
 	reply = (u32) slave_state[slave].comm_toggle << 31;
 	if (toggle != slave_state[slave].comm_toggle) {
-		mlx4_warn(dev, "Incorrect toggle %d from slave %d. *** MASTER"
-			  "STATE COMPROMISIED ***\n", toggle, slave);
+		mlx4_warn(dev, "Incorrect toggle %d from slave %d. *** MASTER STATE COMPROMISED ***\n",
+			  toggle, slave);
 		goto reset_slave;
 	}
 	if (cmd == MLX4_COMM_CMD_RESET) {
@@ -1759,8 +1785,8 @@ static void mlx4_master_do_cmd(struct mlx4_dev *dev, int slave, u8 cmd,
 	/*command from slave in the middle of FLR*/
 	if (cmd != MLX4_COMM_CMD_RESET &&
 	    MLX4_COMM_CMD_FLR == slave_state[slave].last_cmd) {
-		mlx4_warn(dev, "slave:%d is Trying to run cmd(0x%x) "
-			  "in the middle of FLR\n", slave, cmd);
+		mlx4_warn(dev, "slave:%d is Trying to run cmd(0x%x) in the middle of FLR\n",
+			  slave, cmd);
 		return;
 	}
 
@@ -1798,8 +1824,8 @@ static void mlx4_master_do_cmd(struct mlx4_dev *dev, int slave, u8 cmd,
 
 		mutex_lock(&priv->cmd.slave_cmd_mutex);
 		if (mlx4_master_process_vhcr(dev, slave, NULL)) {
-			mlx4_err(dev, "Failed processing vhcr for slave:%d,"
-				 " resetting slave.\n", slave);
+			mlx4_err(dev, "Failed processing vhcr for slave:%d, resetting slave\n",
+				 slave);
 			mutex_unlock(&priv->cmd.slave_cmd_mutex);
 			goto reset_slave;
 		}
@@ -1816,8 +1842,7 @@ static void mlx4_master_do_cmd(struct mlx4_dev *dev, int slave, u8 cmd,
 		is_going_down = 1;
 	spin_unlock_irqrestore(&priv->mfunc.master.slave_state_lock, flags);
 	if (is_going_down) {
-		mlx4_warn(dev, "Slave is going down aborting command(%d)"
-			  " executing from slave:%d\n",
+		mlx4_warn(dev, "Slave is going down aborting command(%d) executing from slave:%d\n",
 			  cmd, slave);
 		return;
 	}
@@ -1880,10 +1905,9 @@ void mlx4_master_comm_channel(struct work_struct *work)
 			if (toggle != slt) {
 				if (master->slave_state[slave].comm_toggle
 				    != slt) {
-					printk(KERN_INFO "slave %d out of sync."
-					       " read toggle %d, state toggle %d. "
-					       "Resynching.\n", slave, slt,
-					       master->slave_state[slave].comm_toggle);
+					pr_info("slave %d out of sync. read toggle %d, state toggle %d. Resynching.\n",
+						slave, slt,
+						master->slave_state[slave].comm_toggle);
 					master->slave_state[slave].comm_toggle =
 						slt;
 				}
@@ -1896,8 +1920,7 @@ void mlx4_master_comm_channel(struct work_struct *work)
 	}
 
 	if (reported && reported != served)
-		mlx4_warn(dev, "Got command event with bitmask from %d slaves"
-			  " but %d were served\n",
+		mlx4_warn(dev, "Got command event with bitmask from %d slaves but %d were served\n",
 			  reported, served);
 
 	if (mlx4_ARM_COMM_CHANNEL(dev))
@@ -1953,7 +1976,7 @@ int mlx4_multi_func_init(struct mlx4_dev *dev)
 		ioremap(pci_resource_start(dev->pdev, 2) +
 			MLX4_SLAVE_COMM_BASE, MLX4_COMM_PAGESIZE);
 	if (!priv->mfunc.comm) {
-		mlx4_err(dev, "Couldn't map communication vector.\n");
+		mlx4_err(dev, "Couldn't map communication vector\n");
 		goto err_vhcr;
 	}
 
@@ -2080,7 +2103,7 @@ int mlx4_cmd_init(struct mlx4_dev *dev)
 		priv->cmd.hcr = ioremap(pci_resource_start(dev->pdev, 0) +
 					MLX4_HCR_BASE, MLX4_HCR_SIZE);
 		if (!priv->cmd.hcr) {
-			mlx4_err(dev, "Couldn't map command register.\n");
+			mlx4_err(dev, "Couldn't map command register\n");
 			return -ENOMEM;
 		}
 	}
@@ -2481,11 +2504,12 @@ int mlx4_get_vf_config(struct mlx4_dev *dev, int port, int vf, struct ifla_vf_in
 	ivf->mac[4] = ((s_info->mac >> (1*8)) & 0xff);
 	ivf->mac[5] = ((s_info->mac)  & 0xff);
 
-	ivf->vlan	= s_info->default_vlan;
-	ivf->qos	= s_info->default_qos;
-	ivf->tx_rate	= s_info->tx_rate;
-	ivf->spoofchk	= s_info->spoofchk;
-	ivf->linkstate	= s_info->link_state;
+	ivf->vlan		= s_info->default_vlan;
+	ivf->qos		= s_info->default_qos;
+	ivf->max_tx_rate	= s_info->tx_rate;
+	ivf->min_tx_rate	= 0;
+	ivf->spoofchk		= s_info->spoofchk;
+	ivf->linkstate		= s_info->link_state;
 
 	return 0;
 }
@@ -2537,3 +2561,50 @@ int mlx4_set_vf_link_state(struct mlx4_dev *dev, int port, int vf, int link_stat
 	return 0;
 }
 EXPORT_SYMBOL_GPL(mlx4_set_vf_link_state);
+
+int mlx4_vf_smi_enabled(struct mlx4_dev *dev, int slave, int port)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+
+	if (slave < 1 || slave >= dev->num_slaves ||
+	    port < 1 || port > MLX4_MAX_PORTS)
+		return 0;
+
+	return priv->mfunc.master.vf_oper[slave].smi_enabled[port] ==
+		MLX4_VF_SMI_ENABLED;
+}
+EXPORT_SYMBOL_GPL(mlx4_vf_smi_enabled);
+
+int mlx4_vf_get_enable_smi_admin(struct mlx4_dev *dev, int slave, int port)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+
+	if (slave == mlx4_master_func_num(dev))
+		return 1;
+
+	if (slave < 1 || slave >= dev->num_slaves ||
+	    port < 1 || port > MLX4_MAX_PORTS)
+		return 0;
+
+	return priv->mfunc.master.vf_admin[slave].enable_smi[port] ==
+		MLX4_VF_SMI_ENABLED;
+}
+EXPORT_SYMBOL_GPL(mlx4_vf_get_enable_smi_admin);
+
+int mlx4_vf_set_enable_smi_admin(struct mlx4_dev *dev, int slave, int port,
+				 int enabled)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+
+	if (slave == mlx4_master_func_num(dev))
+		return 0;
+
+	if (slave < 1 || slave >= dev->num_slaves ||
+	    port < 1 || port > MLX4_MAX_PORTS ||
+	    enabled < 0 || enabled > 1)
+		return -EINVAL;
+
+	priv->mfunc.master.vf_admin[slave].enable_smi[port] = enabled;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx4_vf_set_enable_smi_admin);
diff --git a/drivers/net/ethernet/mellanox/mlx4/cq.c b/drivers/net/ethernet/mellanox/mlx4/cq.c
index 0487121e4a0f..80f725228f5b 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cq.c
@@ -173,11 +173,11 @@ int __mlx4_cq_alloc_icm(struct mlx4_dev *dev, int *cqn)
 	if (*cqn == -1)
 		return -ENOMEM;
 
-	err = mlx4_table_get(dev, &cq_table->table, *cqn);
+	err = mlx4_table_get(dev, &cq_table->table, *cqn, GFP_KERNEL);
 	if (err)
 		goto err_out;
 
-	err = mlx4_table_get(dev, &cq_table->cmpt_table, *cqn);
+	err = mlx4_table_get(dev, &cq_table->cmpt_table, *cqn, GFP_KERNEL);
 	if (err)
 		goto err_put;
 	return 0;
@@ -293,6 +293,9 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent,
 	atomic_set(&cq->refcount, 1);
 	init_completion(&cq->free);
 
+	cq->irq = priv->eq_table.eq[cq->vector].irq;
+	cq->irq_affinity_change = false;
+
 	return 0;
 
 err_radix:
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
index c2cd8d31bcad..4b2130760eed 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
@@ -125,8 +125,7 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
 						   &cq->vector)) {
 					cq->vector = (cq->ring + 1 + priv->port)
 					    % mdev->dev->caps.num_comp_vectors;
-					mlx4_warn(mdev, "Failed Assigning an EQ to "
-						  "%s ,Falling back to legacy EQ's\n",
+					mlx4_warn(mdev, "Failed assigning an EQ to %s, falling back to legacy EQ's\n",
 						  name);
 				}
 			}
@@ -164,6 +163,13 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
 		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq,
 			       NAPI_POLL_WEIGHT);
 	} else {
+		struct mlx4_en_rx_ring *ring = priv->rx_ring[cq->ring];
+
+		err = irq_set_affinity_hint(cq->mcq.irq,
+					    ring->affinity_mask);
+		if (err)
+			mlx4_warn(mdev, "Failed setting affinity hint\n");
+
 		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq, 64);
 		napi_hash_add(&cq->napi);
 	}
@@ -180,8 +186,11 @@ void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq **pcq)
 
 	mlx4_en_unmap_buffer(&cq->wqres.buf);
 	mlx4_free_hwq_res(mdev->dev, &cq->wqres, cq->buf_size);
-	if (priv->mdev->dev->caps.comp_pool && cq->vector)
+	if (priv->mdev->dev->caps.comp_pool && cq->vector) {
+		if (!cq->is_tx)
+			irq_set_affinity_hint(cq->mcq.irq, NULL);
 		mlx4_release_eq(priv->mdev->dev, cq->vector);
+	}
 	cq->vector = 0;
 	cq->buf_size = 0;
 	cq->buf = NULL;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 3e8d33605fe7..fa1a069e14e6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -378,8 +378,8 @@ static int mlx4_en_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		ethtool_cmd_speed_set(cmd, priv->port_state.link_speed);
 		cmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(cmd, -1);
-		cmd->duplex = -1;
+		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
+		cmd->duplex = DUPLEX_UNKNOWN;
 	}
 
 	if (trans_type > 0 && trans_type <= 0xC) {
@@ -564,7 +564,7 @@ static u32 mlx4_en_get_rxfh_indir_size(struct net_device *dev)
 	return priv->rx_ring_num;
 }
 
-static int mlx4_en_get_rxfh_indir(struct net_device *dev, u32 *ring_index)
+static int mlx4_en_get_rxfh(struct net_device *dev, u32 *ring_index, u8 *key)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_rss_map *rss_map = &priv->rss_map;
@@ -582,8 +582,8 @@ static int mlx4_en_get_rxfh_indir(struct net_device *dev, u32 *ring_index)
 	return err;
 }
 
-static int mlx4_en_set_rxfh_indir(struct net_device *dev,
-		const u32 *ring_index)
+static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index,
+			    const u8 *key)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
@@ -925,13 +925,13 @@ static int mlx4_en_flow_replace(struct net_device *dev,
 		qpn = cmd->fs.ring_cookie & (EN_ETHTOOL_QP_ATTACH - 1);
 	} else {
 		if (cmd->fs.ring_cookie >= priv->rx_ring_num) {
-			en_warn(priv, "rxnfc: RX ring (%llu) doesn't exist.\n",
+			en_warn(priv, "rxnfc: RX ring (%llu) doesn't exist\n",
 				cmd->fs.ring_cookie);
 			return -EINVAL;
 		}
 		qpn = priv->rss_map.qps[cmd->fs.ring_cookie].qpn;
 		if (!qpn) {
-			en_warn(priv, "rxnfc: RX ring (%llu) is inactive.\n",
+			en_warn(priv, "rxnfc: RX ring (%llu) is inactive\n",
 				cmd->fs.ring_cookie);
 			return -EINVAL;
 		}
@@ -956,7 +956,7 @@ static int mlx4_en_flow_replace(struct net_device *dev,
 	}
 	err = mlx4_flow_attach(priv->mdev->dev, &rule, &reg_id);
 	if (err) {
-		en_err(priv, "Fail to attach network rule at location %d.\n",
+		en_err(priv, "Fail to attach network rule at location %d\n",
 		       cmd->fs.location);
 		goto out_free_list;
 	}
@@ -1121,7 +1121,7 @@ static int mlx4_en_set_channels(struct net_device *dev,
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
-	int port_up;
+	int port_up = 0;
 	int err = 0;
 
 	if (channel->other_count || channel->combined_count ||
@@ -1151,7 +1151,8 @@ static int mlx4_en_set_channels(struct net_device *dev,
 	netif_set_real_num_tx_queues(dev, priv->tx_ring_num);
 	netif_set_real_num_rx_queues(dev, priv->rx_ring_num);
 
-	mlx4_en_setup_tc(dev, MLX4_EN_NUM_UP);
+	if (dev->num_tc)
+		mlx4_en_setup_tc(dev, MLX4_EN_NUM_UP);
 
 	en_warn(priv, "Using %d TX rings\n", priv->tx_ring_num);
 	en_warn(priv, "Using %d RX rings\n", priv->rx_ring_num);
@@ -1223,8 +1224,8 @@ const struct ethtool_ops mlx4_en_ethtool_ops = {
 	.get_rxnfc = mlx4_en_get_rxnfc,
 	.set_rxnfc = mlx4_en_set_rxnfc,
 	.get_rxfh_indir_size = mlx4_en_get_rxfh_indir_size,
-	.get_rxfh_indir = mlx4_en_get_rxfh_indir,
-	.set_rxfh_indir = mlx4_en_set_rxfh_indir,
+	.get_rxfh = mlx4_en_get_rxfh,
+	.set_rxfh = mlx4_en_set_rxfh,
 	.get_channels = mlx4_en_get_channels,
 	.set_channels = mlx4_en_set_channels,
 	.get_ts_info = mlx4_en_get_ts_info,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c
index 0c59d4fe7e3a..f953c1d7eae6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c
@@ -133,7 +133,7 @@ static int mlx4_en_get_profile(struct mlx4_en_dev *mdev)
 			MLX4_EN_MAX_TX_RING_P_UP);
 	if (params->udp_rss && !(mdev->dev->caps.flags
 					& MLX4_DEV_CAP_FLAG_UDP_RSS)) {
-		mlx4_warn(mdev, "UDP RSS is not supported on this device.\n");
+		mlx4_warn(mdev, "UDP RSS is not supported on this device\n");
 		params->udp_rss = 0;
 	}
 	for (i = 1; i <= MLX4_MAX_PORTS; i++) {
@@ -251,8 +251,7 @@ static void *mlx4_en_add(struct mlx4_dev *dev)
 
 	mdev->LSO_support = !!(dev->caps.flags & (1 << 15));
 	if (!mdev->LSO_support)
-		mlx4_warn(mdev, "LSO not supported, please upgrade to later "
-				"FW version to enable LSO\n");
+		mlx4_warn(mdev, "LSO not supported, please upgrade to later FW version to enable LSO\n");
 
 	if (mlx4_mr_alloc(mdev->dev, mdev->priv_pdn, 0, ~0ull,
 			 MLX4_PERM_LOCAL_WRITE |  MLX4_PERM_LOCAL_READ,
@@ -268,7 +267,7 @@ static void *mlx4_en_add(struct mlx4_dev *dev)
 	/* Build device profile according to supplied module parameters */
 	err = mlx4_en_get_profile(mdev);
 	if (err) {
-		mlx4_err(mdev, "Bad module parameters, aborting.\n");
+		mlx4_err(mdev, "Bad module parameters, aborting\n");
 		goto err_mr;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 7e4b1720c3d1..7d4fb7bf2593 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -130,7 +130,7 @@ static enum mlx4_net_trans_rule_id mlx4_ip_proto_to_trans_rule_id(u8 ip_proto)
 	case IPPROTO_TCP:
 		return MLX4_NET_TRANS_RULE_ID_TCP;
 	default:
-		return -EPROTONOSUPPORT;
+		return MLX4_NET_TRANS_RULE_NUM;
 	}
 };
 
@@ -177,7 +177,7 @@ static void mlx4_en_filter_work(struct work_struct *work)
 	int rc;
 	__be64 mac_mask = cpu_to_be64(MLX4_MAC_MASK << 16);
 
-	if (spec_tcp_udp.id < 0) {
+	if (spec_tcp_udp.id >= MLX4_NET_TRANS_RULE_NUM) {
 		en_warn(priv, "RFS: ignoring unsupported ip protocol (%d)\n",
 			filter->ip_proto);
 		goto ignore;
@@ -770,11 +770,12 @@ static int mlx4_en_do_set_mac(struct mlx4_en_priv *priv)
 					  priv->dev->dev_addr, priv->prev_mac);
 		if (err)
 			en_err(priv, "Failed changing HW MAC address\n");
-		memcpy(priv->prev_mac, priv->dev->dev_addr,
-		       sizeof(priv->prev_mac));
 	} else
 		en_dbg(HW, priv, "Port is down while registering mac, exiting...\n");
 
+	memcpy(priv->prev_mac, priv->dev->dev_addr,
+	       sizeof(priv->prev_mac));
+
 	return err;
 }
 
@@ -788,9 +789,8 @@ static int mlx4_en_set_mac(struct net_device *dev, void *addr)
 	if (!is_valid_ether_addr(saddr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	memcpy(dev->dev_addr, saddr->sa_data, ETH_ALEN);
-
 	mutex_lock(&mdev->state_lock);
+	memcpy(dev->dev_addr, saddr->sa_data, ETH_ALEN);
 	err = mlx4_en_do_set_mac(priv);
 	mutex_unlock(&mdev->state_lock);
 
@@ -1526,6 +1526,27 @@ static void mlx4_en_linkstate(struct work_struct *work)
 	mutex_unlock(&mdev->state_lock);
 }
 
+static int mlx4_en_init_affinity_hint(struct mlx4_en_priv *priv, int ring_idx)
+{
+	struct mlx4_en_rx_ring *ring = priv->rx_ring[ring_idx];
+	int numa_node = priv->mdev->dev->numa_node;
+	int ret = 0;
+
+	if (!zalloc_cpumask_var(&ring->affinity_mask, GFP_KERNEL))
+		return -ENOMEM;
+
+	ret = cpumask_set_cpu_local_first(ring_idx, numa_node,
+					  ring->affinity_mask);
+	if (ret)
+		free_cpumask_var(ring->affinity_mask);
+
+	return ret;
+}
+
+static void mlx4_en_free_affinity_hint(struct mlx4_en_priv *priv, int ring_idx)
+{
+	free_cpumask_var(priv->rx_ring[ring_idx]->affinity_mask);
+}
 
 int mlx4_en_start_port(struct net_device *dev)
 {
@@ -1567,17 +1588,25 @@ int mlx4_en_start_port(struct net_device *dev)
 
 		mlx4_en_cq_init_lock(cq);
 
+		err = mlx4_en_init_affinity_hint(priv, i);
+		if (err) {
+			en_err(priv, "Failed preparing IRQ affinity hint\n");
+			goto cq_err;
+		}
+
 		err = mlx4_en_activate_cq(priv, cq, i);
 		if (err) {
 			en_err(priv, "Failed activating Rx CQ\n");
+			mlx4_en_free_affinity_hint(priv, i);
 			goto cq_err;
 		}
 		for (j = 0; j < cq->size; j++)
 			cq->buf[j].owner_sr_opcode = MLX4_CQE_OWNER_MASK;
 		err = mlx4_en_set_cq_moder(priv, cq);
 		if (err) {
-			en_err(priv, "Failed setting cq moderation parameters");
+			en_err(priv, "Failed setting cq moderation parameters\n");
 			mlx4_en_deactivate_cq(priv, cq);
+			mlx4_en_free_affinity_hint(priv, i);
 			goto cq_err;
 		}
 		mlx4_en_arm_cq(priv, cq);
@@ -1615,7 +1644,7 @@ int mlx4_en_start_port(struct net_device *dev)
 		}
 		err = mlx4_en_set_cq_moder(priv, cq);
 		if (err) {
-			en_err(priv, "Failed setting cq moderation parameters");
+			en_err(priv, "Failed setting cq moderation parameters\n");
 			mlx4_en_deactivate_cq(priv, cq);
 			goto tx_err;
 		}
@@ -1715,8 +1744,10 @@ rss_err:
 mac_err:
 	mlx4_en_put_qp(priv);
 cq_err:
-	while (rx_index--)
+	while (rx_index--) {
 		mlx4_en_deactivate_cq(priv, priv->rx_cq[rx_index]);
+		mlx4_en_free_affinity_hint(priv, i);
+	}
 	for (i = 0; i < priv->rx_ring_num; i++)
 		mlx4_en_deactivate_rx_ring(priv, priv->rx_ring[i]);
 
@@ -1847,6 +1878,8 @@ void mlx4_en_stop_port(struct net_device *dev, int detach)
 			msleep(1);
 		mlx4_en_deactivate_rx_ring(priv, priv->rx_ring[i]);
 		mlx4_en_deactivate_cq(priv, cq);
+
+		mlx4_en_free_affinity_hint(priv, i);
 	}
 }
 
@@ -2539,7 +2572,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	netif_set_real_num_tx_queues(dev, priv->tx_ring_num);
 	netif_set_real_num_rx_queues(dev, priv->rx_ring_num);
 
-	SET_ETHTOOL_OPS(dev, &mlx4_en_ethtool_ops);
+	dev->ethtool_ops = &mlx4_en_ethtool_ops;
 
 	/*
 	 * Set driver features
@@ -2594,8 +2627,8 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 				    prof->tx_pause, prof->tx_ppp,
 				    prof->rx_pause, prof->rx_ppp);
 	if (err) {
-		en_err(priv, "Failed setting port general configurations "
-		       "for port %d, with error %d\n", priv->port, err);
+		en_err(priv, "Failed setting port general configurations for port %d, with error %d\n",
+		       priv->port, err);
 		goto out;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index ba049ae88749..d2d415732d99 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -270,13 +270,11 @@ static int mlx4_en_fill_rx_buffers(struct mlx4_en_priv *priv)
 						    ring->actual_size,
 						    GFP_KERNEL)) {
 				if (ring->actual_size < MLX4_EN_MIN_RX_SIZE) {
-					en_err(priv, "Failed to allocate "
-						     "enough rx buffers\n");
+					en_err(priv, "Failed to allocate enough rx buffers\n");
 					return -ENOMEM;
 				} else {
 					new_size = rounddown_pow_of_two(ring->actual_size);
-					en_warn(priv, "Only %d buffers allocated "
-						      "reducing ring size to %d",
+					en_warn(priv, "Only %d buffers allocated reducing ring size to %d\n",
 						ring->actual_size, new_size);
 					goto reduce_rings;
 				}
@@ -685,10 +683,9 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 		/* Drop packet on bad receive or bad checksum */
 		if (unlikely((cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) ==
 						MLX4_CQE_OPCODE_ERROR)) {
-			en_err(priv, "CQE completed in error - vendor "
-				  "syndrom:%d syndrom:%d\n",
-				  ((struct mlx4_err_cqe *) cqe)->vendor_err_syndrome,
-				  ((struct mlx4_err_cqe *) cqe)->syndrome);
+			en_err(priv, "CQE completed in error - vendor syndrom:%d syndrom:%d\n",
+			       ((struct mlx4_err_cqe *)cqe)->vendor_err_syndrome,
+			       ((struct mlx4_err_cqe *)cqe)->syndrome);
 			goto next;
 		}
 		if (unlikely(cqe->badfcs_enc & MLX4_CQE_BAD_FCS)) {
@@ -898,10 +895,17 @@ int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget)
 	mlx4_en_cq_unlock_napi(cq);
 
 	/* If we used up all the quota - we're probably not done yet... */
-	if (done == budget)
+	if (done == budget) {
 		INC_PERF_COUNTER(priv->pstats.napi_quota);
-	else {
+		if (unlikely(cq->mcq.irq_affinity_change)) {
+			cq->mcq.irq_affinity_change = false;
+			napi_complete(napi);
+			mlx4_en_arm_cq(priv, cq);
+			return 0;
+		}
+	} else {
 		/* Done for now */
+		cq->mcq.irq_affinity_change = false;
 		napi_complete(napi);
 		mlx4_en_arm_cq(priv, cq);
 	}
@@ -944,8 +948,8 @@ void mlx4_en_calc_rx_buf(struct net_device *dev)
 	priv->rx_skb_size = eff_mtu;
 	priv->log_rx_info = ROUNDUP_LOG2(i * sizeof(struct mlx4_en_rx_alloc));
 
-	en_dbg(DRV, priv, "Rx buffer scatter-list (effective-mtu:%d "
-		  "num_frags:%d):\n", eff_mtu, priv->num_frags);
+	en_dbg(DRV, priv, "Rx buffer scatter-list (effective-mtu:%d num_frags:%d):\n",
+	       eff_mtu, priv->num_frags);
 	for (i = 0; i < priv->num_frags; i++) {
 		en_err(priv,
 		       "  frag:%d - size:%d prefix:%d align:%d stride:%d\n",
@@ -972,7 +976,7 @@ static int mlx4_en_config_rss_qp(struct mlx4_en_priv *priv, int qpn,
 	if (!context)
 		return -ENOMEM;
 
-	err = mlx4_qp_alloc(mdev->dev, qpn, qp);
+	err = mlx4_qp_alloc(mdev->dev, qpn, qp, GFP_KERNEL);
 	if (err) {
 		en_err(priv, "Failed to allocate qp #%x\n", qpn);
 		goto out;
@@ -1012,7 +1016,7 @@ int mlx4_en_create_drop_qp(struct mlx4_en_priv *priv)
 		en_err(priv, "Failed reserving drop qpn\n");
 		return err;
 	}
-	err = mlx4_qp_alloc(priv->mdev->dev, qpn, &priv->drop_qp);
+	err = mlx4_qp_alloc(priv->mdev->dev, qpn, &priv->drop_qp, GFP_KERNEL);
 	if (err) {
 		en_err(priv, "Failed allocating drop qp\n");
 		mlx4_qp_release_range(priv->mdev->dev, qpn, 1);
@@ -1071,7 +1075,7 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
 	}
 
 	/* Configure RSS indirection qp */
-	err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, &rss_map->indir_qp);
+	err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, &rss_map->indir_qp, GFP_KERNEL);
 	if (err) {
 		en_err(priv, "Failed to allocate RSS indirection QP\n");
 		goto rss_err;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index dd1f6d346459..8be7483f8236 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -108,12 +108,12 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 
 	ring->buf = ring->wqres.buf.direct.buf;
 
-	en_dbg(DRV, priv, "Allocated TX ring (addr:%p) - buf:%p size:%d "
-	       "buf_size:%d dma:%llx\n", ring, ring->buf, ring->size,
-	       ring->buf_size, (unsigned long long) ring->wqres.buf.direct.map);
+	en_dbg(DRV, priv, "Allocated TX ring (addr:%p) - buf:%p size:%d buf_size:%d dma:%llx\n",
+	       ring, ring->buf, ring->size, ring->buf_size,
+	       (unsigned long long) ring->wqres.buf.direct.map);
 
 	ring->qpn = qpn;
-	err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp);
+	err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp, GFP_KERNEL);
 	if (err) {
 		en_err(priv, "Failed allocating qp %d\n", ring->qpn);
 		goto err_map;
@@ -122,7 +122,7 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
 
 	err = mlx4_bf_alloc(mdev->dev, &ring->bf, node);
 	if (err) {
-		en_dbg(DRV, priv, "working without blueflame (%d)", err);
+		en_dbg(DRV, priv, "working without blueflame (%d)\n", err);
 		ring->bf.uar = &mdev->priv_uar;
 		ring->bf.uar->map = mdev->uar_map;
 		ring->bf_enabled = false;
@@ -474,9 +474,15 @@ int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget)
 	/* If we used up all the quota - we're probably not done yet... */
 	if (done < budget) {
 		/* Done for now */
+		cq->mcq.irq_affinity_change = false;
 		napi_complete(napi);
 		mlx4_en_arm_cq(priv, cq);
 		return done;
+	} else if (unlikely(cq->mcq.irq_affinity_change)) {
+		cq->mcq.irq_affinity_change = false;
+		napi_complete(napi);
+		mlx4_en_arm_cq(priv, cq);
+		return 0;
 	}
 	return budget;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index d501a2b0fb79..d954ec1eac17 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -53,6 +53,11 @@ enum {
 	MLX4_EQ_ENTRY_SIZE	= 0x20
 };
 
+struct mlx4_irq_notify {
+	void *arg;
+	struct irq_affinity_notify notify;
+};
+
 #define MLX4_EQ_STATUS_OK	   ( 0 << 28)
 #define MLX4_EQ_STATUS_WRITE_FAIL  (10 << 28)
 #define MLX4_EQ_OWNER_SW	   ( 0 << 24)
@@ -152,14 +157,13 @@ void mlx4_gen_slave_eqe(struct work_struct *work)
 				if (i != dev->caps.function &&
 				    master->slave_state[i].active)
 					if (mlx4_GEN_EQE(dev, i, eqe))
-						mlx4_warn(dev, "Failed to "
-							  " generate event "
-							  "for slave %d\n", i);
+						mlx4_warn(dev, "Failed to generate event for slave %d\n",
+							  i);
 			}
 		} else {
 			if (mlx4_GEN_EQE(dev, slave, eqe))
-				mlx4_warn(dev, "Failed to generate event "
-					       "for slave %d\n", slave);
+				mlx4_warn(dev, "Failed to generate event for slave %d\n",
+					  slave);
 		}
 		++slave_eq->cons;
 	}
@@ -177,8 +181,8 @@ static void slave_event(struct mlx4_dev *dev, u8 slave, struct mlx4_eqe *eqe)
 	s_eqe = &slave_eq->event_eqe[slave_eq->prod & (SLAVE_EVENT_EQ_SIZE - 1)];
 	if ((!!(s_eqe->owner & 0x80)) ^
 	    (!!(slave_eq->prod & SLAVE_EVENT_EQ_SIZE))) {
-		mlx4_warn(dev, "Master failed to generate an EQE for slave: %d. "
-			  "No free EQE on slave events queue\n", slave);
+		mlx4_warn(dev, "Master failed to generate an EQE for slave: %d. No free EQE on slave events queue\n",
+			  slave);
 		spin_unlock_irqrestore(&slave_eq->event_lock, flags);
 		return;
 	}
@@ -375,9 +379,9 @@ int set_and_calc_slave_port_state(struct mlx4_dev *dev, int slave,
 		}
 		break;
 	default:
-		pr_err("%s: BUG!!! UNKNOWN state: "
-		       "slave:%d, port:%d\n", __func__, slave, port);
-			goto out;
+		pr_err("%s: BUG!!! UNKNOWN state: slave:%d, port:%d\n",
+		       __func__, slave, port);
+		goto out;
 	}
 	ret = mlx4_get_slave_port_state(dev, slave, port);
 
@@ -425,8 +429,8 @@ void mlx4_master_handle_slave_flr(struct work_struct *work)
 	for (i = 0 ; i < dev->num_slaves; i++) {
 
 		if (MLX4_COMM_CMD_FLR == slave_state[i].last_cmd) {
-			mlx4_dbg(dev, "mlx4_handle_slave_flr: "
-				 "clean slave: %d\n", i);
+			mlx4_dbg(dev, "mlx4_handle_slave_flr: clean slave: %d\n",
+				 i);
 
 			mlx4_delete_all_resources_for_slave(dev, i);
 			/*return the slave to running mode*/
@@ -438,8 +442,8 @@ void mlx4_master_handle_slave_flr(struct work_struct *work)
 			err = mlx4_cmd(dev, 0, i, 0, MLX4_CMD_INFORM_FLR_DONE,
 				       MLX4_CMD_TIME_CLASS_A, MLX4_CMD_WRAPPED);
 			if (err)
-				mlx4_warn(dev, "Failed to notify FW on "
-					  "FLR done (slave:%d)\n", i);
+				mlx4_warn(dev, "Failed to notify FW on FLR done (slave:%d)\n",
+					  i);
 		}
 	}
 }
@@ -490,9 +494,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 						be32_to_cpu(eqe->event.qp.qpn)
 						& 0xffffff, &slave);
 				if (ret && ret != -ENOENT) {
-					mlx4_dbg(dev, "QP event %02x(%02x) on "
-						 "EQ %d at index %u: could "
-						 "not get slave id (%d)\n",
+					mlx4_dbg(dev, "QP event %02x(%02x) on EQ %d at index %u: could not get slave id (%d)\n",
 						 eqe->type, eqe->subtype,
 						 eq->eqn, eq->cons_index, ret);
 					break;
@@ -520,23 +522,19 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 						& 0xffffff,
 						&slave);
 				if (ret && ret != -ENOENT) {
-					mlx4_warn(dev, "SRQ event %02x(%02x) "
-						  "on EQ %d at index %u: could"
-						  " not get slave id (%d)\n",
+					mlx4_warn(dev, "SRQ event %02x(%02x) on EQ %d at index %u: could not get slave id (%d)\n",
 						  eqe->type, eqe->subtype,
 						  eq->eqn, eq->cons_index, ret);
 					break;
 				}
-				mlx4_warn(dev, "%s: slave:%d, srq_no:0x%x,"
-					  " event: %02x(%02x)\n", __func__,
-					  slave,
+				mlx4_warn(dev, "%s: slave:%d, srq_no:0x%x, event: %02x(%02x)\n",
+					  __func__, slave,
 					  be32_to_cpu(eqe->event.srq.srqn),
 					  eqe->type, eqe->subtype);
 
 				if (!ret && slave != dev->caps.function) {
-					mlx4_warn(dev, "%s: sending event "
-						  "%02x(%02x) to slave:%d\n",
-						   __func__, eqe->type,
+					mlx4_warn(dev, "%s: sending event %02x(%02x) to slave:%d\n",
+						  __func__, eqe->type,
 						  eqe->subtype, slave);
 					mlx4_slave_event(dev, slave, eqe);
 					break;
@@ -569,8 +567,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 					if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) {
 						if (i == mlx4_master_func_num(dev))
 							continue;
-						mlx4_dbg(dev, "%s: Sending MLX4_PORT_CHANGE_SUBTYPE_DOWN"
-							 " to slave: %d, port:%d\n",
+						mlx4_dbg(dev, "%s: Sending MLX4_PORT_CHANGE_SUBTYPE_DOWN to slave: %d, port:%d\n",
 							 __func__, i, port);
 						s_info = &priv->mfunc.master.vf_oper[slave].vport[port].state;
 						if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state) {
@@ -634,11 +631,9 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 					be32_to_cpu(eqe->event.cq_err.cqn)
 					& 0xffffff, &slave);
 				if (ret && ret != -ENOENT) {
-					mlx4_dbg(dev, "CQ event %02x(%02x) on "
-						 "EQ %d at index %u: could "
-						  "not get slave id (%d)\n",
-						  eqe->type, eqe->subtype,
-						  eq->eqn, eq->cons_index, ret);
+					mlx4_dbg(dev, "CQ event %02x(%02x) on EQ %d at index %u: could not get slave id (%d)\n",
+						 eqe->type, eqe->subtype,
+						 eq->eqn, eq->cons_index, ret);
 					break;
 				}
 
@@ -667,8 +662,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 
 		case MLX4_EVENT_TYPE_COMM_CHANNEL:
 			if (!mlx4_is_master(dev)) {
-				mlx4_warn(dev, "Received comm channel event "
-					       "for non master device\n");
+				mlx4_warn(dev, "Received comm channel event for non master device\n");
 				break;
 			}
 			memcpy(&priv->mfunc.master.comm_arm_bit_vector,
@@ -681,8 +675,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 		case MLX4_EVENT_TYPE_FLR_EVENT:
 			flr_slave = be32_to_cpu(eqe->event.flr_event.slave_id);
 			if (!mlx4_is_master(dev)) {
-				mlx4_warn(dev, "Non-master function received"
-					       "FLR event\n");
+				mlx4_warn(dev, "Non-master function received FLR event\n");
 				break;
 			}
 
@@ -711,22 +704,17 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 			if (eqe->subtype == MLX4_FATAL_WARNING_SUBTYPE_WARMING) {
 				if (mlx4_is_master(dev))
 					for (i = 0; i < dev->num_slaves; i++) {
-						mlx4_dbg(dev, "%s: Sending "
-							"MLX4_FATAL_WARNING_SUBTYPE_WARMING"
-							" to slave: %d\n", __func__, i);
+						mlx4_dbg(dev, "%s: Sending MLX4_FATAL_WARNING_SUBTYPE_WARMING to slave: %d\n",
+							 __func__, i);
 						if (i == dev->caps.function)
 							continue;
 						mlx4_slave_event(dev, i, eqe);
 					}
-				mlx4_err(dev, "Temperature Threshold was reached! "
-					"Threshold: %d celsius degrees; "
-					"Current Temperature: %d\n",
-					be16_to_cpu(eqe->event.warming.warning_threshold),
-					be16_to_cpu(eqe->event.warming.current_temperature));
+				mlx4_err(dev, "Temperature Threshold was reached! Threshold: %d celsius degrees; Current Temperature: %d\n",
+					 be16_to_cpu(eqe->event.warming.warning_threshold),
+					 be16_to_cpu(eqe->event.warming.current_temperature));
 			} else
-				mlx4_warn(dev, "Unhandled event FATAL WARNING (%02x), "
-					  "subtype %02x on EQ %d at index %u. owner=%x, "
-					  "nent=0x%x, slave=%x, ownership=%s\n",
+				mlx4_warn(dev, "Unhandled event FATAL WARNING (%02x), subtype %02x on EQ %d at index %u. owner=%x, nent=0x%x, slave=%x, ownership=%s\n",
 					  eqe->type, eqe->subtype, eq->eqn,
 					  eq->cons_index, eqe->owner, eq->nent,
 					  eqe->slave_id,
@@ -743,9 +731,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 		case MLX4_EVENT_TYPE_EEC_CATAS_ERROR:
 		case MLX4_EVENT_TYPE_ECC_DETECT:
 		default:
-			mlx4_warn(dev, "Unhandled event %02x(%02x) on EQ %d at "
-				  "index %u. owner=%x, nent=0x%x, slave=%x, "
-				  "ownership=%s\n",
+			mlx4_warn(dev, "Unhandled event %02x(%02x) on EQ %d at index %u. owner=%x, nent=0x%x, slave=%x, ownership=%s\n",
 				  eqe->type, eqe->subtype, eq->eqn,
 				  eq->cons_index, eqe->owner, eq->nent,
 				  eqe->slave_id,
@@ -1088,7 +1074,7 @@ static int mlx4_map_clr_int(struct mlx4_dev *dev)
 	priv->clr_base = ioremap(pci_resource_start(dev->pdev, priv->fw.clr_int_bar) +
 				 priv->fw.clr_int_base, MLX4_CLR_INT_SIZE);
 	if (!priv->clr_base) {
-		mlx4_err(dev, "Couldn't map interrupt clear register, aborting.\n");
+		mlx4_err(dev, "Couldn't map interrupt clear register, aborting\n");
 		return -ENOMEM;
 	}
 
@@ -1102,6 +1088,57 @@ static void mlx4_unmap_clr_int(struct mlx4_dev *dev)
 	iounmap(priv->clr_base);
 }
 
+static void mlx4_irq_notifier_notify(struct irq_affinity_notify *notify,
+				     const cpumask_t *mask)
+{
+	struct mlx4_irq_notify *n = container_of(notify,
+						 struct mlx4_irq_notify,
+						 notify);
+	struct mlx4_priv *priv = (struct mlx4_priv *)n->arg;
+	struct radix_tree_iter iter;
+	void **slot;
+
+	radix_tree_for_each_slot(slot, &priv->cq_table.tree, &iter, 0) {
+		struct mlx4_cq *cq = (struct mlx4_cq *)(*slot);
+
+		if (cq->irq == notify->irq)
+			cq->irq_affinity_change = true;
+	}
+}
+
+static void mlx4_release_irq_notifier(struct kref *ref)
+{
+	struct mlx4_irq_notify *n = container_of(ref, struct mlx4_irq_notify,
+						 notify.kref);
+	kfree(n);
+}
+
+static void mlx4_assign_irq_notifier(struct mlx4_priv *priv,
+				     struct mlx4_dev *dev, int irq)
+{
+	struct mlx4_irq_notify *irq_notifier = NULL;
+	int err = 0;
+
+	irq_notifier = kzalloc(sizeof(*irq_notifier), GFP_KERNEL);
+	if (!irq_notifier) {
+		mlx4_warn(dev, "Failed to allocate irq notifier. irq %d\n",
+			  irq);
+		return;
+	}
+
+	irq_notifier->notify.irq = irq;
+	irq_notifier->notify.notify = mlx4_irq_notifier_notify;
+	irq_notifier->notify.release = mlx4_release_irq_notifier;
+	irq_notifier->arg = priv;
+	err = irq_set_affinity_notifier(irq, &irq_notifier->notify);
+	if (err) {
+		kfree(irq_notifier);
+		irq_notifier = NULL;
+		mlx4_warn(dev, "Failed to set irq notifier. irq %d\n", irq);
+	}
+}
+
+
 int mlx4_alloc_eq_table(struct mlx4_dev *dev)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
@@ -1372,6 +1409,9 @@ int mlx4_assign_eq(struct mlx4_dev *dev, char *name, struct cpu_rmap *rmap,
 				continue;
 				/*we dont want to break here*/
 			}
+			mlx4_assign_irq_notifier(priv, dev,
+						 priv->eq_table.eq[vec].irq);
+
 			eq_set_ci(&priv->eq_table.eq[vec], 1);
 		}
 	}
@@ -1398,6 +1438,9 @@ void mlx4_release_eq(struct mlx4_dev *dev, int vec)
 		  Belonging to a legacy EQ*/
 		mutex_lock(&priv->msix_ctl.pool_lock);
 		if (priv->msix_ctl.pool_bm & 1ULL << i) {
+			irq_set_affinity_notifier(
+				priv->eq_table.eq[vec].irq,
+				NULL);
 			free_irq(priv->eq_table.eq[vec].irq,
 				 &priv->eq_table.eq[vec]);
 			priv->msix_ctl.pool_bm &= ~(1ULL << i);
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index d16a4d118903..688e1eabab29 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -178,8 +178,8 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 				struct mlx4_cmd_info *cmd)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
-	u8	field;
-	u32	size;
+	u8	field, port;
+	u32	size, proxy_qp, qkey;
 	int	err = 0;
 
 #define QUERY_FUNC_CAP_FLAGS_OFFSET		0x0
@@ -209,6 +209,7 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 
 /* when opcode modifier = 1 */
 #define QUERY_FUNC_CAP_PHYS_PORT_OFFSET		0x3
+#define QUERY_FUNC_CAP_PRIV_VF_QKEY_OFFSET	0x4
 #define QUERY_FUNC_CAP_FLAGS0_OFFSET		0x8
 #define QUERY_FUNC_CAP_FLAGS1_OFFSET		0xc
 
@@ -221,6 +222,7 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 #define QUERY_FUNC_CAP_FLAGS1_FORCE_MAC		0x40
 #define QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN	0x80
 #define QUERY_FUNC_CAP_FLAGS1_NIC_INFO			0x10
+#define QUERY_FUNC_CAP_VF_ENABLE_QP0		0x08
 
 #define QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID 0x80
 
@@ -234,28 +236,35 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
 			return -EINVAL;
 
 		vhcr->in_modifier = converted_port;
-		/* Set nic_info bit to mark new fields support */
-		field  = QUERY_FUNC_CAP_FLAGS1_NIC_INFO;
-		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS1_OFFSET);
-
 		/* phys-port = logical-port */
 		field = vhcr->in_modifier -
 			find_first_bit(actv_ports.ports, dev->caps.num_ports);
 		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_PHYS_PORT_OFFSET);
 
-		field = vhcr->in_modifier;
+		port = vhcr->in_modifier;
+		proxy_qp = dev->phys_caps.base_proxy_sqpn + 8 * slave + port - 1;
+
+		/* Set nic_info bit to mark new fields support */
+		field  = QUERY_FUNC_CAP_FLAGS1_NIC_INFO;
+
+		if (mlx4_vf_smi_enabled(dev, slave, port) &&
+		    !mlx4_get_parav_qkey(dev, proxy_qp, &qkey)) {
+			field |= QUERY_FUNC_CAP_VF_ENABLE_QP0;
+			MLX4_PUT(outbox->buf, qkey,
+				 QUERY_FUNC_CAP_PRIV_VF_QKEY_OFFSET);
+		}
+		MLX4_PUT(outbox->buf, field, QUERY_FUNC_CAP_FLAGS1_OFFSET);
+
 		/* size is now the QP number */
-		size = dev->phys_caps.base_tunnel_sqpn + 8 * slave + field - 1;
+		size = dev->phys_caps.base_tunnel_sqpn + 8 * slave + port - 1;
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP0_TUNNEL);
 
 		size += 2;
 		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP1_TUNNEL);
 
-		size = dev->phys_caps.base_proxy_sqpn + 8 * slave + field - 1;
-		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP0_PROXY);
-
-		size += 2;
-		MLX4_PUT(outbox->buf, size, QUERY_FUNC_CAP_QP1_PROXY);
+		MLX4_PUT(outbox->buf, proxy_qp, QUERY_FUNC_CAP_QP0_PROXY);
+		proxy_qp += 2;
+		MLX4_PUT(outbox->buf, proxy_qp, QUERY_FUNC_CAP_QP1_PROXY);
 
 		MLX4_PUT(outbox->buf, dev->caps.phys_port_id[vhcr->in_modifier],
 			 QUERY_FUNC_CAP_PHYS_PORT_ID);
@@ -326,7 +335,7 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 	struct mlx4_cmd_mailbox *mailbox;
 	u32			*outbox;
 	u8			field, op_modifier;
-	u32			size;
+	u32			size, qkey;
 	int			err = 0, quotas = 0;
 
 	op_modifier = !!gen_or_port; /* 0 = general, 1 = logical port */
@@ -414,7 +423,7 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 
 	MLX4_GET(func_cap->flags1, outbox, QUERY_FUNC_CAP_FLAGS1_OFFSET);
 	if (dev->caps.port_type[gen_or_port] == MLX4_PORT_TYPE_ETH) {
-		if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_OFFSET) {
+		if (func_cap->flags1 & QUERY_FUNC_CAP_FLAGS1_FORCE_VLAN) {
 			mlx4_err(dev, "VLAN is enforced on this port\n");
 			err = -EPROTONOSUPPORT;
 			goto out;
@@ -428,8 +437,7 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 	} else if (dev->caps.port_type[gen_or_port] == MLX4_PORT_TYPE_IB) {
 		MLX4_GET(field, outbox, QUERY_FUNC_CAP_FLAGS0_OFFSET);
 		if (field & QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID) {
-			mlx4_err(dev, "phy_wqe_gid is "
-				 "enforced on this ib port\n");
+			mlx4_err(dev, "phy_wqe_gid is enforced on this ib port\n");
 			err = -EPROTONOSUPPORT;
 			goto out;
 		}
@@ -442,6 +450,13 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u32 gen_or_port,
 		goto out;
 	}
 
+	if (func_cap->flags1 & QUERY_FUNC_CAP_VF_ENABLE_QP0) {
+		MLX4_GET(qkey, outbox, QUERY_FUNC_CAP_PRIV_VF_QKEY_OFFSET);
+		func_cap->qp0_qkey = qkey;
+	} else {
+		func_cap->qp0_qkey = 0;
+	}
+
 	MLX4_GET(size, outbox, QUERY_FUNC_CAP_QP0_TUNNEL);
 	func_cap->qp0_tunnel_qpn = size & 0xFFFFFF;
 
@@ -1054,10 +1069,10 @@ int mlx4_map_cmd(struct mlx4_dev *dev, u16 op, struct mlx4_icm *icm, u64 virt)
 		 */
 		lg = ffs(mlx4_icm_addr(&iter) | mlx4_icm_size(&iter)) - 1;
 		if (lg < MLX4_ICM_PAGE_SHIFT) {
-			mlx4_warn(dev, "Got FW area not aligned to %d (%llx/%lx).\n",
-				   MLX4_ICM_PAGE_SIZE,
-				   (unsigned long long) mlx4_icm_addr(&iter),
-				   mlx4_icm_size(&iter));
+			mlx4_warn(dev, "Got FW area not aligned to %d (%llx/%lx)\n",
+				  MLX4_ICM_PAGE_SIZE,
+				  (unsigned long long) mlx4_icm_addr(&iter),
+				  mlx4_icm_size(&iter));
 			err = -EINVAL;
 			goto out;
 		}
@@ -1093,14 +1108,14 @@ int mlx4_map_cmd(struct mlx4_dev *dev, u16 op, struct mlx4_icm *icm, u64 virt)
 
 	switch (op) {
 	case MLX4_CMD_MAP_FA:
-		mlx4_dbg(dev, "Mapped %d chunks/%d KB for FW.\n", tc, ts);
+		mlx4_dbg(dev, "Mapped %d chunks/%d KB for FW\n", tc, ts);
 		break;
 	case MLX4_CMD_MAP_ICM_AUX:
-		mlx4_dbg(dev, "Mapped %d chunks/%d KB for ICM aux.\n", tc, ts);
+		mlx4_dbg(dev, "Mapped %d chunks/%d KB for ICM aux\n", tc, ts);
 		break;
 	case MLX4_CMD_MAP_ICM:
-		mlx4_dbg(dev, "Mapped %d chunks/%d KB at %llx for ICM.\n",
-			  tc, ts, (unsigned long long) virt - (ts << 10));
+		mlx4_dbg(dev, "Mapped %d chunks/%d KB at %llx for ICM\n",
+			 tc, ts, (unsigned long long) virt - (ts << 10));
 		break;
 	}
 
@@ -1186,14 +1201,13 @@ int mlx4_QUERY_FW(struct mlx4_dev *dev)
 	MLX4_GET(cmd_if_rev, outbox, QUERY_FW_CMD_IF_REV_OFFSET);
 	if (cmd_if_rev < MLX4_COMMAND_INTERFACE_MIN_REV ||
 	    cmd_if_rev > MLX4_COMMAND_INTERFACE_MAX_REV) {
-		mlx4_err(dev, "Installed FW has unsupported "
-			 "command interface revision %d.\n",
+		mlx4_err(dev, "Installed FW has unsupported command interface revision %d\n",
 			 cmd_if_rev);
 		mlx4_err(dev, "(Installed FW version is %d.%d.%03d)\n",
 			 (int) (dev->caps.fw_ver >> 32),
 			 (int) (dev->caps.fw_ver >> 16) & 0xffff,
 			 (int) dev->caps.fw_ver & 0xffff);
-		mlx4_err(dev, "This driver version supports only revisions %d to %d.\n",
+		mlx4_err(dev, "This driver version supports only revisions %d to %d\n",
 			 MLX4_COMMAND_INTERFACE_MIN_REV, MLX4_COMMAND_INTERFACE_MAX_REV);
 		err = -ENODEV;
 		goto out;
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h
index 6811ee00ba7c..1fce03ebe5c4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.h
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.h
@@ -134,6 +134,7 @@ struct mlx4_func_cap {
 	int	max_eq;
 	int	reserved_eq;
 	int	mcg_quota;
+	u32	qp0_qkey;
 	u32	qp0_tunnel_qpn;
 	u32	qp0_proxy_qpn;
 	u32	qp1_tunnel_qpn;
diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.c b/drivers/net/ethernet/mellanox/mlx4/icm.c
index 5fbf4924c272..97c9b1db1d27 100644
--- a/drivers/net/ethernet/mellanox/mlx4/icm.c
+++ b/drivers/net/ethernet/mellanox/mlx4/icm.c
@@ -245,7 +245,8 @@ int mlx4_UNMAP_ICM_AUX(struct mlx4_dev *dev)
 			MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE);
 }
 
-int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, u32 obj)
+int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, u32 obj,
+		   gfp_t gfp)
 {
 	u32 i = (obj & (table->num_obj - 1)) /
 			(MLX4_TABLE_CHUNK_SIZE / table->obj_size);
@@ -259,7 +260,7 @@ int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, u32 obj)
 	}
 
 	table->icm[i] = mlx4_alloc_icm(dev, MLX4_TABLE_CHUNK_SIZE >> PAGE_SHIFT,
-				       (table->lowmem ? GFP_KERNEL : GFP_HIGHUSER) |
+				       (table->lowmem ? gfp : GFP_HIGHUSER) |
 				       __GFP_NOWARN, table->coherent);
 	if (!table->icm[i]) {
 		ret = -ENOMEM;
@@ -356,7 +357,7 @@ int mlx4_table_get_range(struct mlx4_dev *dev, struct mlx4_icm_table *table,
 	u32 i;
 
 	for (i = start; i <= end; i += inc) {
-		err = mlx4_table_get(dev, table, i);
+		err = mlx4_table_get(dev, table, i, GFP_KERNEL);
 		if (err)
 			goto fail;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/icm.h b/drivers/net/ethernet/mellanox/mlx4/icm.h
index dee67fa39107..0c7364550150 100644
--- a/drivers/net/ethernet/mellanox/mlx4/icm.h
+++ b/drivers/net/ethernet/mellanox/mlx4/icm.h
@@ -71,7 +71,8 @@ struct mlx4_icm *mlx4_alloc_icm(struct mlx4_dev *dev, int npages,
 				gfp_t gfp_mask, int coherent);
 void mlx4_free_icm(struct mlx4_dev *dev, struct mlx4_icm *icm, int coherent);
 
-int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, u32 obj);
+int mlx4_table_get(struct mlx4_dev *dev, struct mlx4_icm_table *table, u32 obj,
+		   gfp_t gfp);
 void mlx4_table_put(struct mlx4_dev *dev, struct mlx4_icm_table *table, u32 obj);
 int mlx4_table_get_range(struct mlx4_dev *dev, struct mlx4_icm_table *table,
 			 u32 start, u32 end);
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index c187d748115f..82ab427290c3 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -104,8 +104,6 @@ module_param(enable_64b_cqe_eqe, bool, 0444);
 MODULE_PARM_DESC(enable_64b_cqe_eqe,
 		 "Enable 64 byte CQEs/EQEs when the FW supports this (default: True)");
 
-#define HCA_GLOBAL_CAP_MASK            0
-
 #define PF_CONTEXT_BEHAVIOUR_MASK	MLX4_FUNC_CAP_64B_EQE_CQE
 
 static char mlx4_version[] =
@@ -134,8 +132,7 @@ MODULE_PARM_DESC(log_num_vlan, "Log2 max number of VLANs per ETH port (0-7)");
 
 static bool use_prio;
 module_param_named(use_prio, use_prio, bool, 0444);
-MODULE_PARM_DESC(use_prio, "Enable steering by VLAN priority on ETH ports "
-		  "(0/1, default 0)");
+MODULE_PARM_DESC(use_prio, "Enable steering by VLAN priority on ETH ports (deprecated)");
 
 int log_mtts_per_seg = ilog2(MLX4_MTT_ENTRY_PER_SEG);
 module_param_named(log_mtts_per_seg, log_mtts_per_seg, int, 0444);
@@ -163,8 +160,7 @@ int mlx4_check_port_params(struct mlx4_dev *dev,
 	for (i = 0; i < dev->caps.num_ports - 1; i++) {
 		if (port_type[i] != port_type[i + 1]) {
 			if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP)) {
-				mlx4_err(dev, "Only same port types supported "
-					 "on this HCA, aborting.\n");
+				mlx4_err(dev, "Only same port types supported on this HCA, aborting\n");
 				return -EINVAL;
 			}
 		}
@@ -172,8 +168,8 @@ int mlx4_check_port_params(struct mlx4_dev *dev,
 
 	for (i = 0; i < dev->caps.num_ports; i++) {
 		if (!(port_type[i] & dev->caps.supported_type[i+1])) {
-			mlx4_err(dev, "Requested port type for port %d is not "
-				      "supported on this HCA\n", i + 1);
+			mlx4_err(dev, "Requested port type for port %d is not supported on this HCA\n",
+				 i + 1);
 			return -EINVAL;
 		}
 	}
@@ -195,26 +191,23 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 
 	err = mlx4_QUERY_DEV_CAP(dev, dev_cap);
 	if (err) {
-		mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n");
+		mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting\n");
 		return err;
 	}
 
 	if (dev_cap->min_page_sz > PAGE_SIZE) {
-		mlx4_err(dev, "HCA minimum page size of %d bigger than "
-			 "kernel PAGE_SIZE of %ld, aborting.\n",
+		mlx4_err(dev, "HCA minimum page size of %d bigger than kernel PAGE_SIZE of %ld, aborting\n",
 			 dev_cap->min_page_sz, PAGE_SIZE);
 		return -ENODEV;
 	}
 	if (dev_cap->num_ports > MLX4_MAX_PORTS) {
-		mlx4_err(dev, "HCA has %d ports, but we only support %d, "
-			 "aborting.\n",
+		mlx4_err(dev, "HCA has %d ports, but we only support %d, aborting\n",
 			 dev_cap->num_ports, MLX4_MAX_PORTS);
 		return -ENODEV;
 	}
 
 	if (dev_cap->uar_size > pci_resource_len(dev->pdev, 2)) {
-		mlx4_err(dev, "HCA reported UAR size of 0x%x bigger than "
-			 "PCI resource 2 size of 0x%llx, aborting.\n",
+		mlx4_err(dev, "HCA reported UAR size of 0x%x bigger than PCI resource 2 size of 0x%llx, aborting\n",
 			 dev_cap->uar_size,
 			 (unsigned long long) pci_resource_len(dev->pdev, 2));
 		return -ENODEV;
@@ -296,7 +289,6 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 
 	dev->caps.log_num_macs  = log_num_mac;
 	dev->caps.log_num_vlans = MLX4_LOG_NUM_VLANS;
-	dev->caps.log_num_prios = use_prio ? 3 : 0;
 
 	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		dev->caps.port_type[i] = MLX4_PORT_TYPE_NONE;
@@ -347,14 +339,12 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 
 		if (dev->caps.log_num_macs > dev_cap->log_max_macs[i]) {
 			dev->caps.log_num_macs = dev_cap->log_max_macs[i];
-			mlx4_warn(dev, "Requested number of MACs is too much "
-				  "for port %d, reducing to %d.\n",
+			mlx4_warn(dev, "Requested number of MACs is too much for port %d, reducing to %d\n",
 				  i, 1 << dev->caps.log_num_macs);
 		}
 		if (dev->caps.log_num_vlans > dev_cap->log_max_vlans[i]) {
 			dev->caps.log_num_vlans = dev_cap->log_max_vlans[i];
-			mlx4_warn(dev, "Requested number of VLANs is too much "
-				  "for port %d, reducing to %d.\n",
+			mlx4_warn(dev, "Requested number of VLANs is too much for port %d, reducing to %d\n",
 				  i, 1 << dev->caps.log_num_vlans);
 		}
 	}
@@ -366,7 +356,6 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_ADDR] =
 		(1 << dev->caps.log_num_macs) *
 		(1 << dev->caps.log_num_vlans) *
-		(1 << dev->caps.log_num_prios) *
 		dev->caps.num_ports;
 	dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_EXCH] = MLX4_NUM_FEXCH;
 
@@ -584,13 +573,14 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	memset(&hca_param, 0, sizeof(hca_param));
 	err = mlx4_QUERY_HCA(dev, &hca_param);
 	if (err) {
-		mlx4_err(dev, "QUERY_HCA command failed, aborting.\n");
+		mlx4_err(dev, "QUERY_HCA command failed, aborting\n");
 		return err;
 	}
 
-	/*fail if the hca has an unknown capability */
-	if ((hca_param.global_caps | HCA_GLOBAL_CAP_MASK) !=
-	    HCA_GLOBAL_CAP_MASK) {
+	/* fail if the hca has an unknown global capability
+	 * at this time global_caps should be always zeroed
+	 */
+	if (hca_param.global_caps) {
 		mlx4_err(dev, "Unknown hca global capabilities\n");
 		return -ENOSYS;
 	}
@@ -603,19 +593,18 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	dev->caps.max_qp_dest_rdma = 1 << hca_param.log_rd_per_qp;
 	err = mlx4_dev_cap(dev, &dev_cap);
 	if (err) {
-		mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n");
+		mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting\n");
 		return err;
 	}
 
 	err = mlx4_QUERY_FW(dev);
 	if (err)
-		mlx4_err(dev, "QUERY_FW command failed: could not get FW version.\n");
+		mlx4_err(dev, "QUERY_FW command failed: could not get FW version\n");
 
 	page_size = ~dev->caps.page_size_cap + 1;
 	mlx4_warn(dev, "HCA minimum page size:%d\n", page_size);
 	if (page_size > PAGE_SIZE) {
-		mlx4_err(dev, "HCA minimum page size of %d bigger than "
-			 "kernel PAGE_SIZE of %ld, aborting.\n",
+		mlx4_err(dev, "HCA minimum page size of %d bigger than kernel PAGE_SIZE of %ld, aborting\n",
 			 page_size, PAGE_SIZE);
 		return -ENODEV;
 	}
@@ -633,8 +622,8 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	memset(&func_cap, 0, sizeof(func_cap));
 	err = mlx4_QUERY_FUNC_CAP(dev, 0, &func_cap);
 	if (err) {
-		mlx4_err(dev, "QUERY_FUNC_CAP general command failed, aborting (%d).\n",
-			  err);
+		mlx4_err(dev, "QUERY_FUNC_CAP general command failed, aborting (%d)\n",
+			 err);
 		return err;
 	}
 
@@ -661,18 +650,20 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	dev->caps.num_amgms             = 0;
 
 	if (dev->caps.num_ports > MLX4_MAX_PORTS) {
-		mlx4_err(dev, "HCA has %d ports, but we only support %d, "
-			 "aborting.\n", dev->caps.num_ports, MLX4_MAX_PORTS);
+		mlx4_err(dev, "HCA has %d ports, but we only support %d, aborting\n",
+			 dev->caps.num_ports, MLX4_MAX_PORTS);
 		return -ENODEV;
 	}
 
+	dev->caps.qp0_qkey = kcalloc(dev->caps.num_ports, sizeof(u32), GFP_KERNEL);
 	dev->caps.qp0_tunnel = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
 	dev->caps.qp0_proxy = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
 	dev->caps.qp1_tunnel = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
 	dev->caps.qp1_proxy = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
 
 	if (!dev->caps.qp0_tunnel || !dev->caps.qp0_proxy ||
-	    !dev->caps.qp1_tunnel || !dev->caps.qp1_proxy) {
+	    !dev->caps.qp1_tunnel || !dev->caps.qp1_proxy ||
+	    !dev->caps.qp0_qkey) {
 		err = -ENOMEM;
 		goto err_mem;
 	}
@@ -680,10 +671,11 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		err = mlx4_QUERY_FUNC_CAP(dev, (u32) i, &func_cap);
 		if (err) {
-			mlx4_err(dev, "QUERY_FUNC_CAP port command failed for"
-				 " port %d, aborting (%d).\n", i, err);
+			mlx4_err(dev, "QUERY_FUNC_CAP port command failed for port %d, aborting (%d)\n",
+				 i, err);
 			goto err_mem;
 		}
+		dev->caps.qp0_qkey[i - 1] = func_cap.qp0_qkey;
 		dev->caps.qp0_tunnel[i - 1] = func_cap.qp0_tunnel_qpn;
 		dev->caps.qp0_proxy[i - 1] = func_cap.qp0_proxy_qpn;
 		dev->caps.qp1_tunnel[i - 1] = func_cap.qp1_tunnel_qpn;
@@ -699,8 +691,7 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	if (dev->caps.uar_page_size * (dev->caps.num_uars -
 				       dev->caps.reserved_uars) >
 				       pci_resource_len(dev->pdev, 2)) {
-		mlx4_err(dev, "HCA reported UAR region size of 0x%x bigger than "
-			 "PCI resource 2 size of 0x%llx, aborting.\n",
+		mlx4_err(dev, "HCA reported UAR region size of 0x%x bigger than PCI resource 2 size of 0x%llx, aborting\n",
 			 dev->caps.uar_page_size * dev->caps.num_uars,
 			 (unsigned long long) pci_resource_len(dev->pdev, 2));
 		goto err_mem;
@@ -722,19 +713,23 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 	}
 
 	dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
-	mlx4_warn(dev, "Timestamping is not supported in slave mode.\n");
+	mlx4_warn(dev, "Timestamping is not supported in slave mode\n");
 
 	slave_adjust_steering_mode(dev, &dev_cap, &hca_param);
 
 	return 0;
 
 err_mem:
+	kfree(dev->caps.qp0_qkey);
 	kfree(dev->caps.qp0_tunnel);
 	kfree(dev->caps.qp0_proxy);
 	kfree(dev->caps.qp1_tunnel);
 	kfree(dev->caps.qp1_proxy);
-	dev->caps.qp0_tunnel = dev->caps.qp0_proxy =
-		dev->caps.qp1_tunnel = dev->caps.qp1_proxy = NULL;
+	dev->caps.qp0_qkey = NULL;
+	dev->caps.qp0_tunnel = NULL;
+	dev->caps.qp0_proxy = NULL;
+	dev->caps.qp1_tunnel = NULL;
+	dev->caps.qp1_proxy = NULL;
 
 	return err;
 }
@@ -784,8 +779,8 @@ int mlx4_change_port_types(struct mlx4_dev *dev,
 			dev->caps.port_type[port] = port_types[port - 1];
 			err = mlx4_SET_PORT(dev, port, -1);
 			if (err) {
-				mlx4_err(dev, "Failed to set port %d, "
-					      "aborting\n", port);
+				mlx4_err(dev, "Failed to set port %d, aborting\n",
+					 port);
 				goto out;
 			}
 		}
@@ -868,9 +863,7 @@ static ssize_t set_port_type(struct device *dev,
 		}
 	}
 	if (err) {
-		mlx4_err(mdev, "Auto sensing is not supported on this HCA. "
-			       "Set only 'eth' or 'ib' for both ports "
-			       "(should be the same)\n");
+		mlx4_err(mdev, "Auto sensing is not supported on this HCA. Set only 'eth' or 'ib' for both ports (should be the same)\n");
 		goto out;
 	}
 
@@ -975,8 +968,8 @@ static ssize_t set_port_ib_mtu(struct device *dev,
 		mlx4_CLOSE_PORT(mdev, port);
 		err = mlx4_SET_PORT(mdev, port, -1);
 		if (err) {
-			mlx4_err(mdev, "Failed to set port %d, "
-				      "aborting\n", port);
+			mlx4_err(mdev, "Failed to set port %d, aborting\n",
+				 port);
 			goto err_set_port;
 		}
 	}
@@ -995,19 +988,19 @@ static int mlx4_load_fw(struct mlx4_dev *dev)
 	priv->fw.fw_icm = mlx4_alloc_icm(dev, priv->fw.fw_pages,
 					 GFP_HIGHUSER | __GFP_NOWARN, 0);
 	if (!priv->fw.fw_icm) {
-		mlx4_err(dev, "Couldn't allocate FW area, aborting.\n");
+		mlx4_err(dev, "Couldn't allocate FW area, aborting\n");
 		return -ENOMEM;
 	}
 
 	err = mlx4_MAP_FA(dev, priv->fw.fw_icm);
 	if (err) {
-		mlx4_err(dev, "MAP_FA command failed, aborting.\n");
+		mlx4_err(dev, "MAP_FA command failed, aborting\n");
 		goto err_free;
 	}
 
 	err = mlx4_RUN_FW(dev);
 	if (err) {
-		mlx4_err(dev, "RUN_FW command failed, aborting.\n");
+		mlx4_err(dev, "RUN_FW command failed, aborting\n");
 		goto err_unmap_fa;
 	}
 
@@ -1091,30 +1084,30 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 
 	err = mlx4_SET_ICM_SIZE(dev, icm_size, &aux_pages);
 	if (err) {
-		mlx4_err(dev, "SET_ICM_SIZE command failed, aborting.\n");
+		mlx4_err(dev, "SET_ICM_SIZE command failed, aborting\n");
 		return err;
 	}
 
-	mlx4_dbg(dev, "%lld KB of HCA context requires %lld KB aux memory.\n",
+	mlx4_dbg(dev, "%lld KB of HCA context requires %lld KB aux memory\n",
 		 (unsigned long long) icm_size >> 10,
 		 (unsigned long long) aux_pages << 2);
 
 	priv->fw.aux_icm = mlx4_alloc_icm(dev, aux_pages,
 					  GFP_HIGHUSER | __GFP_NOWARN, 0);
 	if (!priv->fw.aux_icm) {
-		mlx4_err(dev, "Couldn't allocate aux memory, aborting.\n");
+		mlx4_err(dev, "Couldn't allocate aux memory, aborting\n");
 		return -ENOMEM;
 	}
 
 	err = mlx4_MAP_ICM_AUX(dev, priv->fw.aux_icm);
 	if (err) {
-		mlx4_err(dev, "MAP_ICM_AUX command failed, aborting.\n");
+		mlx4_err(dev, "MAP_ICM_AUX command failed, aborting\n");
 		goto err_free_aux;
 	}
 
 	err = mlx4_init_cmpt_table(dev, init_hca->cmpt_base, dev_cap->cmpt_entry_sz);
 	if (err) {
-		mlx4_err(dev, "Failed to map cMPT context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map cMPT context memory, aborting\n");
 		goto err_unmap_aux;
 	}
 
@@ -1125,7 +1118,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  init_hca->eqc_base, dev_cap->eqc_entry_sz,
 				  num_eqs, num_eqs, 0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map EQ context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map EQ context memory, aborting\n");
 		goto err_unmap_cmpt;
 	}
 
@@ -1146,7 +1139,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.num_mtts,
 				  dev->caps.reserved_mtts, 1, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map MTT context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map MTT context memory, aborting\n");
 		goto err_unmap_eq;
 	}
 
@@ -1156,7 +1149,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.num_mpts,
 				  dev->caps.reserved_mrws, 1, 1);
 	if (err) {
-		mlx4_err(dev, "Failed to map dMPT context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map dMPT context memory, aborting\n");
 		goto err_unmap_mtt;
 	}
 
@@ -1167,7 +1160,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
 				  0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map QP context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map QP context memory, aborting\n");
 		goto err_unmap_dmpt;
 	}
 
@@ -1178,7 +1171,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
 				  0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map AUXC context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map AUXC context memory, aborting\n");
 		goto err_unmap_qp;
 	}
 
@@ -1189,7 +1182,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
 				  0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map ALTC context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map ALTC context memory, aborting\n");
 		goto err_unmap_auxc;
 	}
 
@@ -1210,7 +1203,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.num_cqs,
 				  dev->caps.reserved_cqs, 0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map CQ context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map CQ context memory, aborting\n");
 		goto err_unmap_rdmarc;
 	}
 
@@ -1220,7 +1213,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.num_srqs,
 				  dev->caps.reserved_srqs, 0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map SRQ context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map SRQ context memory, aborting\n");
 		goto err_unmap_cq;
 	}
 
@@ -1238,7 +1231,7 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  dev->caps.num_mgms + dev->caps.num_amgms,
 				  0, 0);
 	if (err) {
-		mlx4_err(dev, "Failed to map MCG context memory, aborting.\n");
+		mlx4_err(dev, "Failed to map MCG context memory, aborting\n");
 		goto err_unmap_srq;
 	}
 
@@ -1315,7 +1308,7 @@ static void mlx4_slave_exit(struct mlx4_dev *dev)
 
 	mutex_lock(&priv->cmd.slave_cmd_mutex);
 	if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, MLX4_COMM_TIME))
-		mlx4_warn(dev, "Failed to close slave function.\n");
+		mlx4_warn(dev, "Failed to close slave function\n");
 	mutex_unlock(&priv->cmd.slave_cmd_mutex);
 }
 
@@ -1413,7 +1406,7 @@ static int mlx4_init_slave(struct mlx4_dev *dev)
 	u32 cmd_channel_ver;
 
 	if (atomic_read(&pf_loading)) {
-		mlx4_warn(dev, "PF is not ready. Deferring probe\n");
+		mlx4_warn(dev, "PF is not ready - Deferring probe\n");
 		return -EPROBE_DEFER;
 	}
 
@@ -1426,8 +1419,7 @@ static int mlx4_init_slave(struct mlx4_dev *dev)
 	 * NUM_OF_RESET_RETRIES times before leaving.*/
 	if (ret_from_reset) {
 		if (MLX4_DELAY_RESET_SLAVE == ret_from_reset) {
-			mlx4_warn(dev, "slave is currently in the "
-				  "middle of FLR. Deferring probe.\n");
+			mlx4_warn(dev, "slave is currently in the middle of FLR - Deferring probe\n");
 			mutex_unlock(&priv->cmd.slave_cmd_mutex);
 			return -EPROBE_DEFER;
 		} else
@@ -1441,8 +1433,7 @@ static int mlx4_init_slave(struct mlx4_dev *dev)
 
 	if (MLX4_COMM_GET_IF_REV(cmd_channel_ver) !=
 		MLX4_COMM_GET_IF_REV(slave_read)) {
-		mlx4_err(dev, "slave driver version is not supported"
-			 " by the master\n");
+		mlx4_err(dev, "slave driver version is not supported by the master\n");
 		goto err;
 	}
 
@@ -1520,8 +1511,7 @@ static void choose_steering_mode(struct mlx4_dev *dev,
 
 			if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER ||
 			    dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER)
-				mlx4_warn(dev, "Must have both UC_STEER and MC_STEER flags "
-					  "set to use B0 steering. Falling back to A0 steering mode.\n");
+				mlx4_warn(dev, "Must have both UC_STEER and MC_STEER flags set to use B0 steering - falling back to A0 steering mode\n");
 		}
 		dev->oper_log_mgm_entry_size =
 			mlx4_log_num_mgm_entry_size > 0 ?
@@ -1529,8 +1519,7 @@ static void choose_steering_mode(struct mlx4_dev *dev,
 			MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE;
 		dev->caps.num_qp_per_mgm = mlx4_get_qp_per_mgm(dev);
 	}
-	mlx4_dbg(dev, "Steering mode is: %s, oper_log_mgm_entry_size = %d, "
-		 "modparam log_num_mgm_entry_size = %d\n",
+	mlx4_dbg(dev, "Steering mode is: %s, oper_log_mgm_entry_size = %d, modparam log_num_mgm_entry_size = %d\n",
 		 mlx4_steering_mode_str(dev->caps.steering_mode),
 		 dev->oper_log_mgm_entry_size,
 		 mlx4_log_num_mgm_entry_size);
@@ -1564,15 +1553,15 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 		err = mlx4_QUERY_FW(dev);
 		if (err) {
 			if (err == -EACCES)
-				mlx4_info(dev, "non-primary physical function, skipping.\n");
+				mlx4_info(dev, "non-primary physical function, skipping\n");
 			else
-				mlx4_err(dev, "QUERY_FW command failed, aborting.\n");
+				mlx4_err(dev, "QUERY_FW command failed, aborting\n");
 			return err;
 		}
 
 		err = mlx4_load_fw(dev);
 		if (err) {
-			mlx4_err(dev, "Failed to start FW, aborting.\n");
+			mlx4_err(dev, "Failed to start FW, aborting\n");
 			return err;
 		}
 
@@ -1584,7 +1573,7 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		err = mlx4_dev_cap(dev, &dev_cap);
 		if (err) {
-			mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n");
+			mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting\n");
 			goto err_stop_fw;
 		}
 
@@ -1625,7 +1614,7 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		err = mlx4_INIT_HCA(dev, &init_hca);
 		if (err) {
-			mlx4_err(dev, "INIT_HCA command failed, aborting.\n");
+			mlx4_err(dev, "INIT_HCA command failed, aborting\n");
 			goto err_free_icm;
 		}
 		/*
@@ -1636,7 +1625,7 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 			memset(&init_hca, 0, sizeof(init_hca));
 			err = mlx4_QUERY_HCA(dev, &init_hca);
 			if (err) {
-				mlx4_err(dev, "QUERY_HCA command failed, disable timestamp.\n");
+				mlx4_err(dev, "QUERY_HCA command failed, disable timestamp\n");
 				dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
 			} else {
 				dev->caps.hca_core_clock =
@@ -1649,14 +1638,14 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 			if (!dev->caps.hca_core_clock) {
 				dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
 				mlx4_err(dev,
-					 "HCA frequency is 0. Timestamping is not supported.");
+					 "HCA frequency is 0 - timestamping is not supported\n");
 			} else if (map_internal_clock(dev)) {
 				/*
 				 * Map internal clock,
 				 * in case of failure disable timestamping
 				 */
 				dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
-				mlx4_err(dev, "Failed to map internal clock. Timestamping is not supported.\n");
+				mlx4_err(dev, "Failed to map internal clock. Timestamping is not supported\n");
 			}
 		}
 	} else {
@@ -1683,7 +1672,7 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 	err = mlx4_QUERY_ADAPTER(dev, &adapter);
 	if (err) {
-		mlx4_err(dev, "QUERY_ADAPTER command failed, aborting.\n");
+		mlx4_err(dev, "QUERY_ADAPTER command failed, aborting\n");
 		goto unmap_bf;
 	}
 
@@ -1696,6 +1685,14 @@ unmap_bf:
 	unmap_internal_clock(dev);
 	unmap_bf_area(dev);
 
+	if (mlx4_is_slave(dev)) {
+		kfree(dev->caps.qp0_qkey);
+		kfree(dev->caps.qp0_tunnel);
+		kfree(dev->caps.qp0_proxy);
+		kfree(dev->caps.qp1_tunnel);
+		kfree(dev->caps.qp1_proxy);
+	}
+
 err_close:
 	if (mlx4_is_slave(dev))
 		mlx4_slave_exit(dev);
@@ -1793,79 +1790,69 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 
 	err = mlx4_init_uar_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "user access region table, aborting.\n");
-		return err;
+		mlx4_err(dev, "Failed to initialize user access region table, aborting\n");
+		 return err;
 	}
 
 	err = mlx4_uar_alloc(dev, &priv->driver_uar);
 	if (err) {
-		mlx4_err(dev, "Failed to allocate driver access region, "
-			 "aborting.\n");
+		mlx4_err(dev, "Failed to allocate driver access region, aborting\n");
 		goto err_uar_table_free;
 	}
 
 	priv->kar = ioremap((phys_addr_t) priv->driver_uar.pfn << PAGE_SHIFT, PAGE_SIZE);
 	if (!priv->kar) {
-		mlx4_err(dev, "Couldn't map kernel access region, "
-			 "aborting.\n");
+		mlx4_err(dev, "Couldn't map kernel access region, aborting\n");
 		err = -ENOMEM;
 		goto err_uar_free;
 	}
 
 	err = mlx4_init_pd_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "protection domain table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize protection domain table, aborting\n");
 		goto err_kar_unmap;
 	}
 
 	err = mlx4_init_xrcd_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "reliable connection domain table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize reliable connection domain table, aborting\n");
 		goto err_pd_table_free;
 	}
 
 	err = mlx4_init_mr_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "memory region table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize memory region table, aborting\n");
 		goto err_xrcd_table_free;
 	}
 
 	if (!mlx4_is_slave(dev)) {
 		err = mlx4_init_mcg_table(dev);
 		if (err) {
-			mlx4_err(dev, "Failed to initialize multicast group table, aborting.\n");
+			mlx4_err(dev, "Failed to initialize multicast group table, aborting\n");
 			goto err_mr_table_free;
 		}
 	}
 
 	err = mlx4_init_eq_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "event queue table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize event queue table, aborting\n");
 		goto err_mcg_table_free;
 	}
 
 	err = mlx4_cmd_use_events(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to switch to event-driven "
-			 "firmware commands, aborting.\n");
+		mlx4_err(dev, "Failed to switch to event-driven firmware commands, aborting\n");
 		goto err_eq_table_free;
 	}
 
 	err = mlx4_NOP(dev);
 	if (err) {
 		if (dev->flags & MLX4_FLAG_MSI_X) {
-			mlx4_warn(dev, "NOP command failed to generate MSI-X "
-				  "interrupt IRQ %d).\n",
+			mlx4_warn(dev, "NOP command failed to generate MSI-X interrupt IRQ %d)\n",
 				  priv->eq_table.eq[dev->caps.num_comp_vectors].irq);
-			mlx4_warn(dev, "Trying again without MSI-X.\n");
+			mlx4_warn(dev, "Trying again without MSI-X\n");
 		} else {
-			mlx4_err(dev, "NOP command failed to generate interrupt "
-				 "(IRQ %d), aborting.\n",
+			mlx4_err(dev, "NOP command failed to generate interrupt (IRQ %d), aborting\n",
 				 priv->eq_table.eq[dev->caps.num_comp_vectors].irq);
 			mlx4_err(dev, "BIOS or ACPI interrupt routing problem?\n");
 		}
@@ -1877,28 +1864,25 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 
 	err = mlx4_init_cq_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "completion queue table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize completion queue table, aborting\n");
 		goto err_cmd_poll;
 	}
 
 	err = mlx4_init_srq_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "shared receive queue table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize shared receive queue table, aborting\n");
 		goto err_cq_table_free;
 	}
 
 	err = mlx4_init_qp_table(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to initialize "
-			 "queue pair table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize queue pair table, aborting\n");
 		goto err_srq_table_free;
 	}
 
 	err = mlx4_init_counters_table(dev);
 	if (err && err != -ENOENT) {
-		mlx4_err(dev, "Failed to initialize counters table, aborting.\n");
+		mlx4_err(dev, "Failed to initialize counters table, aborting\n");
 		goto err_qp_table_free;
 	}
 
@@ -1908,9 +1892,8 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 			err = mlx4_get_port_ib_caps(dev, port,
 						    &ib_port_default_caps);
 			if (err)
-				mlx4_warn(dev, "failed to get port %d default "
-					  "ib capabilities (%d). Continuing "
-					  "with caps = 0\n", port, err);
+				mlx4_warn(dev, "failed to get port %d default ib capabilities (%d). Continuing with caps = 0\n",
+					  port, err);
 			dev->caps.ib_port_def_cap[port] = ib_port_default_caps;
 
 			/* initialize per-slave default ib port capabilities */
@@ -1920,7 +1903,7 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 					if (i == mlx4_master_func_num(dev))
 						continue;
 					priv->mfunc.master.slave_state[i].ib_cap_mask[port] =
-							ib_port_default_caps;
+						ib_port_default_caps;
 				}
 			}
 
@@ -1933,7 +1916,7 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 					    dev->caps.pkey_table_len[port] : -1);
 			if (err) {
 				mlx4_err(dev, "Failed to set port %d, aborting\n",
-					port);
+					 port);
 				goto err_counters_table_free;
 			}
 		}
@@ -2009,7 +1992,7 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev)
 			kfree(entries);
 			goto no_msi;
 		} else if (nreq < MSIX_LEGACY_SZ +
-				  dev->caps.num_ports * MIN_MSIX_P_PORT) {
+			   dev->caps.num_ports * MIN_MSIX_P_PORT) {
 			/*Working in legacy mode , all EQ's shared*/
 			dev->caps.comp_pool           = 0;
 			dev->caps.num_comp_vectors = nreq - 1;
@@ -2210,8 +2193,7 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 
 	err = pci_enable_device(pdev);
 	if (err) {
-		dev_err(&pdev->dev, "Cannot enable PCI device, "
-			"aborting.\n");
+		dev_err(&pdev->dev, "Cannot enable PCI device, aborting\n");
 		return err;
 	}
 
@@ -2258,14 +2240,13 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 	 */
 	if (!(pci_dev_data & MLX4_PCI_DEV_IS_VF) &&
 	    !(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
-		dev_err(&pdev->dev, "Missing DCS, aborting."
-			"(driver_data: 0x%x, pci_resource_flags(pdev, 0):0x%lx)\n",
+		dev_err(&pdev->dev, "Missing DCS, aborting (driver_data: 0x%x, pci_resource_flags(pdev, 0):0x%lx)\n",
 			pci_dev_data, pci_resource_flags(pdev, 0));
 		err = -ENODEV;
 		goto err_disable_pdev;
 	}
 	if (!(pci_resource_flags(pdev, 2) & IORESOURCE_MEM)) {
-		dev_err(&pdev->dev, "Missing UAR, aborting.\n");
+		dev_err(&pdev->dev, "Missing UAR, aborting\n");
 		err = -ENODEV;
 		goto err_disable_pdev;
 	}
@@ -2280,21 +2261,19 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 
 	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
 	if (err) {
-		dev_warn(&pdev->dev, "Warning: couldn't set 64-bit PCI DMA mask.\n");
+		dev_warn(&pdev->dev, "Warning: couldn't set 64-bit PCI DMA mask\n");
 		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (err) {
-			dev_err(&pdev->dev, "Can't set PCI DMA mask, aborting.\n");
+			dev_err(&pdev->dev, "Can't set PCI DMA mask, aborting\n");
 			goto err_release_regions;
 		}
 	}
 	err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
 	if (err) {
-		dev_warn(&pdev->dev, "Warning: couldn't set 64-bit "
-			 "consistent PCI DMA mask.\n");
+		dev_warn(&pdev->dev, "Warning: couldn't set 64-bit consistent PCI DMA mask\n");
 		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (err) {
-			dev_err(&pdev->dev, "Can't set consistent PCI DMA mask, "
-				"aborting.\n");
+			dev_err(&pdev->dev, "Can't set consistent PCI DMA mask, aborting\n");
 			goto err_release_regions;
 		}
 	}
@@ -2325,7 +2304,7 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 		if (total_vfs) {
 			unsigned vfs_offset = 0;
 			for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]) &&
-			     vfs_offset + nvfs[i] < extended_func_num(pdev);
+				     vfs_offset + nvfs[i] < extended_func_num(pdev);
 			     vfs_offset += nvfs[i], i++)
 				;
 			if (i == sizeof(nvfs)/sizeof(nvfs[0])) {
@@ -2351,8 +2330,7 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 			if (err < 0)
 				goto err_free_dev;
 			else {
-				mlx4_warn(dev, "Multiple PFs not yet supported."
-					  " Skipping PF.\n");
+				mlx4_warn(dev, "Multiple PFs not yet supported - Skipping PF\n");
 				err = -EINVAL;
 				goto err_free_dev;
 			}
@@ -2362,8 +2340,8 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 			mlx4_warn(dev, "Enabling SR-IOV with %d VFs\n",
 				  total_vfs);
 			dev->dev_vfs = kzalloc(
-					total_vfs * sizeof(*dev->dev_vfs),
-					GFP_KERNEL);
+				total_vfs * sizeof(*dev->dev_vfs),
+				GFP_KERNEL);
 			if (NULL == dev->dev_vfs) {
 				mlx4_err(dev, "Failed to allocate memory for VFs\n");
 				err = 0;
@@ -2371,14 +2349,14 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 				atomic_inc(&pf_loading);
 				err = pci_enable_sriov(pdev, total_vfs);
 				if (err) {
-					mlx4_err(dev, "Failed to enable SR-IOV, continuing without SR-IOV (err = %d).\n",
+					mlx4_err(dev, "Failed to enable SR-IOV, continuing without SR-IOV (err = %d)\n",
 						 err);
 					atomic_dec(&pf_loading);
 					err = 0;
 				} else {
 					mlx4_warn(dev, "Running in master mode\n");
 					dev->flags |= MLX4_FLAG_SRIOV |
-						      MLX4_FLAG_MASTER;
+						MLX4_FLAG_MASTER;
 					dev->num_vfs = total_vfs;
 					sriov_initialized = 1;
 				}
@@ -2395,7 +2373,7 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 		 */
 		err = mlx4_reset(dev);
 		if (err) {
-			mlx4_err(dev, "Failed to reset HCA, aborting.\n");
+			mlx4_err(dev, "Failed to reset HCA, aborting\n");
 			goto err_rel_own;
 		}
 	}
@@ -2403,7 +2381,7 @@ static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data)
 slave_start:
 	err = mlx4_cmd_init(dev);
 	if (err) {
-		mlx4_err(dev, "Failed to init command interface, aborting.\n");
+		mlx4_err(dev, "Failed to init command interface, aborting\n");
 		goto err_sriov;
 	}
 
@@ -2417,8 +2395,7 @@ slave_start:
 			dev->num_slaves = 0;
 			err = mlx4_multi_func_init(dev);
 			if (err) {
-				mlx4_err(dev, "Failed to init slave mfunc"
-					 " interface, aborting.\n");
+				mlx4_err(dev, "Failed to init slave mfunc interface, aborting\n");
 				goto err_cmd;
 			}
 		}
@@ -2450,8 +2427,7 @@ slave_start:
 		unsigned sum = 0;
 		err = mlx4_multi_func_init(dev);
 		if (err) {
-			mlx4_err(dev, "Failed to init master mfunc"
-				 "interface, aborting.\n");
+			mlx4_err(dev, "Failed to init master mfunc interface, aborting\n");
 			goto err_close;
 		}
 		if (sriov_initialized) {
@@ -2462,11 +2438,9 @@ slave_start:
 			if (ib_ports &&
 			    (num_vfs_argc > 1 || probe_vfs_argc > 1)) {
 				mlx4_err(dev,
-					 "Invalid syntax of num_vfs/probe_vfs "
-					 "with IB port. Single port VFs syntax"
-					 " is only supported when all ports "
-					 "are configured as ethernet\n");
-				goto err_close;
+					 "Invalid syntax of num_vfs/probe_vfs with IB port - single port VFs syntax is only supported when all ports are configured as ethernet\n");
+				err = -EINVAL;
+				goto err_master_mfunc;
 			}
 			for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]); i++) {
 				unsigned j;
@@ -2491,8 +2465,7 @@ slave_start:
 	if ((mlx4_is_mfunc(dev)) &&
 	    !(dev->flags & MLX4_FLAG_MSI_X)) {
 		err = -ENOSYS;
-		mlx4_err(dev, "INTx is not supported in multi-function mode."
-			 " aborting.\n");
+		mlx4_err(dev, "INTx is not supported in multi-function mode, aborting\n");
 		goto err_free_eq;
 	}
 
@@ -2566,6 +2539,14 @@ err_master_mfunc:
 	if (mlx4_is_master(dev))
 		mlx4_multi_func_cleanup(dev);
 
+	if (mlx4_is_slave(dev)) {
+		kfree(dev->caps.qp0_qkey);
+		kfree(dev->caps.qp0_tunnel);
+		kfree(dev->caps.qp0_proxy);
+		kfree(dev->caps.qp1_tunnel);
+		kfree(dev->caps.qp1_proxy);
+	}
+
 err_close:
 	if (dev->flags & MLX4_FLAG_MSI_X)
 		pci_disable_msix(pdev);
@@ -2637,7 +2618,7 @@ static void __mlx4_remove_one(struct pci_dev *pdev)
 	/* in SRIOV it is not allowed to unload the pf's
 	 * driver while there are alive vf's */
 	if (mlx4_is_master(dev) && mlx4_how_many_lives_vf(dev))
-		printk(KERN_ERR "Removing PF when there are assigned VF's !!!\n");
+		pr_warn("Removing PF when there are assigned VF's !!!\n");
 	mlx4_stop_sense(dev);
 	mlx4_unregister_device(dev);
 
@@ -2689,6 +2670,7 @@ static void __mlx4_remove_one(struct pci_dev *pdev)
 	if (!mlx4_is_slave(dev))
 		mlx4_free_ownership(dev);
 
+	kfree(dev->caps.qp0_qkey);
 	kfree(dev->caps.qp0_tunnel);
 	kfree(dev->caps.qp0_proxy);
 	kfree(dev->caps.qp1_tunnel);
@@ -2800,7 +2782,7 @@ static struct pci_driver mlx4_driver = {
 	.name		= DRV_NAME,
 	.id_table	= mlx4_pci_table,
 	.probe		= mlx4_init_one,
-	.shutdown	= mlx4_remove_one,
+	.shutdown	= __mlx4_remove_one,
 	.remove		= mlx4_remove_one,
 	.err_handler    = &mlx4_err_handler,
 };
@@ -2808,33 +2790,36 @@ static struct pci_driver mlx4_driver = {
 static int __init mlx4_verify_params(void)
 {
 	if ((log_num_mac < 0) || (log_num_mac > 7)) {
-		pr_warning("mlx4_core: bad num_mac: %d\n", log_num_mac);
+		pr_warn("mlx4_core: bad num_mac: %d\n", log_num_mac);
 		return -1;
 	}
 
 	if (log_num_vlan != 0)
-		pr_warning("mlx4_core: log_num_vlan - obsolete module param, using %d\n",
-			   MLX4_LOG_NUM_VLANS);
+		pr_warn("mlx4_core: log_num_vlan - obsolete module param, using %d\n",
+			MLX4_LOG_NUM_VLANS);
+
+	if (use_prio != 0)
+		pr_warn("mlx4_core: use_prio - obsolete module param, ignored\n");
 
 	if ((log_mtts_per_seg < 1) || (log_mtts_per_seg > 7)) {
-		pr_warning("mlx4_core: bad log_mtts_per_seg: %d\n", log_mtts_per_seg);
+		pr_warn("mlx4_core: bad log_mtts_per_seg: %d\n",
+			log_mtts_per_seg);
 		return -1;
 	}
 
 	/* Check if module param for ports type has legal combination */
 	if (port_type_array[0] == false && port_type_array[1] == true) {
-		printk(KERN_WARNING "Module parameter configuration ETH/IB is not supported. Switching to default configuration IB/IB\n");
+		pr_warn("Module parameter configuration ETH/IB is not supported. Switching to default configuration IB/IB\n");
 		port_type_array[0] = true;
 	}
 
 	if (mlx4_log_num_mgm_entry_size != -1 &&
 	    (mlx4_log_num_mgm_entry_size < MLX4_MIN_MGM_LOG_ENTRY_SIZE ||
 	     mlx4_log_num_mgm_entry_size > MLX4_MAX_MGM_LOG_ENTRY_SIZE)) {
-		pr_warning("mlx4_core: mlx4_log_num_mgm_entry_size (%d) not "
-			   "in legal range (-1 or %d..%d)\n",
-			   mlx4_log_num_mgm_entry_size,
-			   MLX4_MIN_MGM_LOG_ENTRY_SIZE,
-			   MLX4_MAX_MGM_LOG_ENTRY_SIZE);
+		pr_warn("mlx4_core: mlx4_log_num_mgm_entry_size (%d) not in legal range (-1 or %d..%d)\n",
+			mlx4_log_num_mgm_entry_size,
+			MLX4_MIN_MGM_LOG_ENTRY_SIZE,
+			MLX4_MAX_MGM_LOG_ENTRY_SIZE);
 		return -1;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c
index 80ccb4edf825..4c36def8e10f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mcg.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c
@@ -638,7 +638,7 @@ static int find_entry(struct mlx4_dev *dev, u8 port,
 
 		if (!(be32_to_cpu(mgm->members_count) & 0xffffff)) {
 			if (*index != hash) {
-				mlx4_err(dev, "Found zero MGID in AMGM.\n");
+				mlx4_err(dev, "Found zero MGID in AMGM\n");
 				err = -EINVAL;
 			}
 			return err;
@@ -874,7 +874,7 @@ static void mlx4_err_rule(struct mlx4_dev *dev, char *str,
 	mlx4_err(dev, "%s", buf);
 
 	if (len >= BUF_SIZE)
-		mlx4_err(dev, "Network rule error message was truncated, print buffer is too small.\n");
+		mlx4_err(dev, "Network rule error message was truncated, print buffer is too small\n");
 }
 
 int mlx4_flow_attach(struct mlx4_dev *dev,
@@ -897,7 +897,7 @@ int mlx4_flow_attach(struct mlx4_dev *dev,
 		ret = parse_trans_rule(dev, cur, mailbox->buf + size);
 		if (ret < 0) {
 			mlx4_free_cmd_mailbox(dev, mailbox);
-			return -EINVAL;
+			return ret;
 		}
 		size += ret;
 	}
@@ -905,10 +905,10 @@ int mlx4_flow_attach(struct mlx4_dev *dev,
 	ret = mlx4_QP_FLOW_STEERING_ATTACH(dev, mailbox, size >> 2, reg_id);
 	if (ret == -ENOMEM)
 		mlx4_err_rule(dev,
-			      "mcg table is full. Fail to register network rule.\n",
+			      "mcg table is full. Fail to register network rule\n",
 			      rule);
 	else if (ret)
-		mlx4_err_rule(dev, "Fail to register network rule.\n", rule);
+		mlx4_err_rule(dev, "Fail to register network rule\n", rule);
 
 	mlx4_free_cmd_mailbox(dev, mailbox);
 
@@ -994,7 +994,7 @@ int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 
 	members_count = be32_to_cpu(mgm->members_count) & 0xffffff;
 	if (members_count == dev->caps.num_qp_per_mgm) {
-		mlx4_err(dev, "MGM at index %x is full.\n", index);
+		mlx4_err(dev, "MGM at index %x is full\n", index);
 		err = -ENOMEM;
 		goto out;
 	}
@@ -1042,7 +1042,7 @@ out:
 	}
 	if (err && link && index != -1) {
 		if (index < dev->caps.num_mgms)
-			mlx4_warn(dev, "Got AMGM index %d < %d",
+			mlx4_warn(dev, "Got AMGM index %d < %d\n",
 				  index, dev->caps.num_mgms);
 		else
 			mlx4_bitmap_free(&priv->mcg_table.bitmap,
@@ -1133,7 +1133,7 @@ int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 
 		if (amgm_index) {
 			if (amgm_index < dev->caps.num_mgms)
-				mlx4_warn(dev, "MGM entry %d had AMGM index %d < %d",
+				mlx4_warn(dev, "MGM entry %d had AMGM index %d < %d\n",
 					  index, amgm_index, dev->caps.num_mgms);
 			else
 				mlx4_bitmap_free(&priv->mcg_table.bitmap,
@@ -1153,7 +1153,7 @@ int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			goto out;
 
 		if (index < dev->caps.num_mgms)
-			mlx4_warn(dev, "entry %d had next AMGM index %d < %d",
+			mlx4_warn(dev, "entry %d had next AMGM index %d < %d\n",
 				  prev, index, dev->caps.num_mgms);
 		else
 			mlx4_bitmap_free(&priv->mcg_table.bitmap,
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 8e9eb02e09cb..1d8af7336807 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -133,6 +133,11 @@ enum {
 	MLX4_COMM_CMD_FLR = 254
 };
 
+enum {
+	MLX4_VF_SMI_DISABLED,
+	MLX4_VF_SMI_ENABLED
+};
+
 /*The flag indicates that the slave should delay the RESET cmd*/
 #define MLX4_DELAY_RESET_SLAVE 0xbbbbbbb
 /*indicates how many retries will be done if we are in the middle of FLR*/
@@ -216,18 +221,19 @@ extern int mlx4_debug_level;
 #define mlx4_debug_level	(0)
 #endif /* CONFIG_MLX4_DEBUG */
 
-#define mlx4_dbg(mdev, format, arg...)					\
+#define mlx4_dbg(mdev, format, ...)					\
 do {									\
 	if (mlx4_debug_level)						\
-		dev_printk(KERN_DEBUG, &mdev->pdev->dev, format, ##arg); \
+		dev_printk(KERN_DEBUG, &(mdev)->pdev->dev, format,	\
+			   ##__VA_ARGS__);				\
 } while (0)
 
-#define mlx4_err(mdev, format, arg...) \
-	dev_err(&mdev->pdev->dev, format, ##arg)
-#define mlx4_info(mdev, format, arg...) \
-	dev_info(&mdev->pdev->dev, format, ##arg)
-#define mlx4_warn(mdev, format, arg...) \
-	dev_warn(&mdev->pdev->dev, format, ##arg)
+#define mlx4_err(mdev, format, ...)					\
+	dev_err(&(mdev)->pdev->dev, format, ##__VA_ARGS__)
+#define mlx4_info(mdev, format, ...)					\
+	dev_info(&(mdev)->pdev->dev, format, ##__VA_ARGS__)
+#define mlx4_warn(mdev, format, ...)					\
+	dev_warn(&(mdev)->pdev->dev, format, ##__VA_ARGS__)
 
 extern int mlx4_log_num_mgm_entry_size;
 extern int log_mtts_per_seg;
@@ -488,6 +494,7 @@ struct mlx4_vport_state {
 
 struct mlx4_vf_admin_state {
 	struct mlx4_vport_state vport[MLX4_MAX_PORTS + 1];
+	u8 enable_smi[MLX4_MAX_PORTS + 1];
 };
 
 struct mlx4_vport_oper_state {
@@ -495,8 +502,10 @@ struct mlx4_vport_oper_state {
 	int mac_idx;
 	int vlan_idx;
 };
+
 struct mlx4_vf_oper_state {
 	struct mlx4_vport_oper_state vport[MLX4_MAX_PORTS + 1];
+	u8 smi_enabled[MLX4_MAX_PORTS + 1];
 };
 
 struct slave_list {
@@ -895,7 +904,7 @@ void mlx4_cleanup_cq_table(struct mlx4_dev *dev);
 void mlx4_cleanup_qp_table(struct mlx4_dev *dev);
 void mlx4_cleanup_srq_table(struct mlx4_dev *dev);
 void mlx4_cleanup_mcg_table(struct mlx4_dev *dev);
-int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn);
+int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn, gfp_t gfp);
 void __mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn);
 int __mlx4_cq_alloc_icm(struct mlx4_dev *dev, int *cqn);
 void __mlx4_cq_free_icm(struct mlx4_dev *dev, int cqn);
@@ -903,7 +912,7 @@ int __mlx4_srq_alloc_icm(struct mlx4_dev *dev, int *srqn);
 void __mlx4_srq_free_icm(struct mlx4_dev *dev, int srqn);
 int __mlx4_mpt_reserve(struct mlx4_dev *dev);
 void __mlx4_mpt_release(struct mlx4_dev *dev, u32 index);
-int __mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index);
+int __mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index, gfp_t gfp);
 void __mlx4_mpt_free_icm(struct mlx4_dev *dev, u32 index);
 u32 __mlx4_alloc_mtt_range(struct mlx4_dev *dev, int order);
 void __mlx4_free_mtt_range(struct mlx4_dev *dev, u32 first_seg, int order);
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 04d9b6fe3e80..0e15295bedd6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -313,6 +313,7 @@ struct mlx4_en_rx_ring {
 	unsigned long csum_ok;
 	unsigned long csum_none;
 	int hwtstamp_rx_filter;
+	cpumask_var_t affinity_mask;
 };
 
 struct mlx4_en_cq {
@@ -830,26 +831,26 @@ __printf(3, 4)
 int en_print(const char *level, const struct mlx4_en_priv *priv,
 	     const char *format, ...);
 
-#define en_dbg(mlevel, priv, format, arg...)			\
-do {								\
-	if (NETIF_MSG_##mlevel & priv->msg_enable)		\
-		en_print(KERN_DEBUG, priv, format, ##arg);	\
+#define en_dbg(mlevel, priv, format, ...)				\
+do {									\
+	if (NETIF_MSG_##mlevel & (priv)->msg_enable)			\
+		en_print(KERN_DEBUG, priv, format, ##__VA_ARGS__);	\
 } while (0)
-#define en_warn(priv, format, arg...)			\
-	en_print(KERN_WARNING, priv, format, ##arg)
-#define en_err(priv, format, arg...)			\
-	en_print(KERN_ERR, priv, format, ##arg)
-#define en_info(priv, format, arg...)			\
-	en_print(KERN_INFO, priv, format, ## arg)
-
-#define mlx4_err(mdev, format, arg...)			\
-	pr_err("%s %s: " format, DRV_NAME,		\
-	       dev_name(&mdev->pdev->dev), ##arg)
-#define mlx4_info(mdev, format, arg...)			\
-	pr_info("%s %s: " format, DRV_NAME,		\
-		dev_name(&mdev->pdev->dev), ##arg)
-#define mlx4_warn(mdev, format, arg...)			\
-	pr_warning("%s %s: " format, DRV_NAME,		\
-		   dev_name(&mdev->pdev->dev), ##arg)
+#define en_warn(priv, format, ...)					\
+	en_print(KERN_WARNING, priv, format, ##__VA_ARGS__)
+#define en_err(priv, format, ...)					\
+	en_print(KERN_ERR, priv, format, ##__VA_ARGS__)
+#define en_info(priv, format, ...)					\
+	en_print(KERN_INFO, priv, format, ##__VA_ARGS__)
+
+#define mlx4_err(mdev, format, ...)					\
+	pr_err(DRV_NAME " %s: " format,					\
+	       dev_name(&(mdev)->pdev->dev), ##__VA_ARGS__)
+#define mlx4_info(mdev, format, ...)					\
+	pr_info(DRV_NAME " %s: " format,				\
+		dev_name(&(mdev)->pdev->dev), ##__VA_ARGS__)
+#define mlx4_warn(mdev, format, ...)					\
+	pr_warn(DRV_NAME " %s: " format,				\
+		dev_name(&(mdev)->pdev->dev), ##__VA_ARGS__)
 
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx4/mr.c b/drivers/net/ethernet/mellanox/mlx4/mr.c
index 24835853b753..2839abb878a6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mr.c
+++ b/drivers/net/ethernet/mellanox/mlx4/mr.c
@@ -250,8 +250,8 @@ static void mlx4_free_mtt_range(struct mlx4_dev *dev, u32 offset, int order)
 						       MLX4_CMD_TIME_CLASS_A,
 						       MLX4_CMD_WRAPPED);
 		if (err)
-			mlx4_warn(dev, "Failed to free mtt range at:"
-				  "%d order:%d\n", offset, order);
+			mlx4_warn(dev, "Failed to free mtt range at:%d order:%d\n",
+				  offset, order);
 		return;
 	}
 	 __mlx4_free_mtt_range(dev, offset, order);
@@ -364,14 +364,14 @@ static void mlx4_mpt_release(struct mlx4_dev *dev, u32 index)
 	__mlx4_mpt_release(dev, index);
 }
 
-int __mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index)
+int __mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index, gfp_t gfp)
 {
 	struct mlx4_mr_table *mr_table = &mlx4_priv(dev)->mr_table;
 
-	return mlx4_table_get(dev, &mr_table->dmpt_table, index);
+	return mlx4_table_get(dev, &mr_table->dmpt_table, index, gfp);
 }
 
-static int mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index)
+static int mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index, gfp_t gfp)
 {
 	u64 param = 0;
 
@@ -382,7 +382,7 @@ static int mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index)
 							MLX4_CMD_TIME_CLASS_A,
 							MLX4_CMD_WRAPPED);
 	}
-	return __mlx4_mpt_alloc_icm(dev, index);
+	return __mlx4_mpt_alloc_icm(dev, index, gfp);
 }
 
 void __mlx4_mpt_free_icm(struct mlx4_dev *dev, u32 index)
@@ -436,8 +436,8 @@ static int mlx4_mr_free_reserved(struct mlx4_dev *dev, struct mlx4_mr *mr)
 				     key_to_hw_index(mr->key) &
 				     (dev->caps.num_mpts - 1));
 		if (err) {
-			mlx4_warn(dev, "HW2SW_MPT failed (%d),", err);
-			mlx4_warn(dev, "MR has MWs bound to it.\n");
+			mlx4_warn(dev, "HW2SW_MPT failed (%d), MR has MWs bound to it\n",
+				  err);
 			return err;
 		}
 
@@ -469,7 +469,7 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
 	struct mlx4_mpt_entry *mpt_entry;
 	int err;
 
-	err = mlx4_mpt_alloc_icm(dev, key_to_hw_index(mr->key));
+	err = mlx4_mpt_alloc_icm(dev, key_to_hw_index(mr->key), GFP_KERNEL);
 	if (err)
 		return err;
 
@@ -627,13 +627,14 @@ int mlx4_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 EXPORT_SYMBOL_GPL(mlx4_write_mtt);
 
 int mlx4_buf_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
-		       struct mlx4_buf *buf)
+		       struct mlx4_buf *buf, gfp_t gfp)
 {
 	u64 *page_list;
 	int err;
 	int i;
 
-	page_list = kmalloc(buf->npages * sizeof *page_list, GFP_KERNEL);
+	page_list = kmalloc(buf->npages * sizeof *page_list,
+			    gfp);
 	if (!page_list)
 		return -ENOMEM;
 
@@ -680,7 +681,7 @@ int mlx4_mw_enable(struct mlx4_dev *dev, struct mlx4_mw *mw)
 	struct mlx4_mpt_entry *mpt_entry;
 	int err;
 
-	err = mlx4_mpt_alloc_icm(dev, key_to_hw_index(mw->key));
+	err = mlx4_mpt_alloc_icm(dev, key_to_hw_index(mw->key), GFP_KERNEL);
 	if (err)
 		return err;
 
@@ -773,7 +774,7 @@ int mlx4_init_mr_table(struct mlx4_dev *dev)
 			mlx4_alloc_mtt_range(dev,
 					     fls(dev->caps.reserved_mtts - 1));
 		if (priv->reserved_mtts < 0) {
-			mlx4_warn(dev, "MTT table of order %u is too small.\n",
+			mlx4_warn(dev, "MTT table of order %u is too small\n",
 				  mr_table->mtt_buddy.max_order);
 			err = -ENOMEM;
 			goto err_reserve_mtts;
@@ -954,8 +955,7 @@ void mlx4_fmr_unmap(struct mlx4_dev *dev, struct mlx4_fmr *fmr,
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox)) {
 		err = PTR_ERR(mailbox);
-		printk(KERN_WARNING "mlx4_ib: mlx4_alloc_cmd_mailbox"
-		       " failed (%d)\n", err);
+		pr_warn("mlx4_ib: mlx4_alloc_cmd_mailbox failed (%d)\n", err);
 		return;
 	}
 
@@ -964,8 +964,7 @@ void mlx4_fmr_unmap(struct mlx4_dev *dev, struct mlx4_fmr *fmr,
 			     (dev->caps.num_mpts - 1));
 	mlx4_free_cmd_mailbox(dev, mailbox);
 	if (err) {
-		printk(KERN_WARNING "mlx4_ib: mlx4_HW2SW_MPT failed (%d)\n",
-		       err);
+		pr_warn("mlx4_ib: mlx4_HW2SW_MPT failed (%d)\n", err);
 		return;
 	}
 	fmr->mr.enabled = MLX4_MPT_EN_SW;
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index 5ec6f203c6e6..7ab97174886d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -254,8 +254,8 @@ void __mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 	if (validate_index(dev, table, index))
 		goto out;
 	if (--table->refs[index]) {
-		mlx4_dbg(dev, "Have more references for index %d,"
-			 "no need to modify mac table\n", index);
+		mlx4_dbg(dev, "Have more references for index %d, no need to modify mac table\n",
+			 index);
 		goto out;
 	}
 
@@ -453,9 +453,8 @@ void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan)
 	}
 
 	if (--table->refs[index]) {
-		mlx4_dbg(dev, "Have %d more references for index %d,"
-			 "no need to modify vlan table\n", table->refs[index],
-			 index);
+		mlx4_dbg(dev, "Have %d more references for index %d, no need to modify vlan table\n",
+			 table->refs[index], index);
 		goto out;
 	}
 	table->entries[index] = 0;
@@ -796,8 +795,7 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod,
 					if (!memcmp(gid_entry_mbox->raw, gid_entry_tbl->raw,
 						    sizeof(gid_entry_tbl->raw))) {
 						/* found duplicate */
-						mlx4_warn(dev, "requested gid entry for slave:%d "
-							  "is a duplicate of gid at index %d\n",
+						mlx4_warn(dev, "requested gid entry for slave:%d is a duplicate of gid at index %d\n",
 							  slave, i);
 						mutex_unlock(&(priv->port[port].gid_table.mutex));
 						return -EINVAL;
diff --git a/drivers/net/ethernet/mellanox/mlx4/profile.c b/drivers/net/ethernet/mellanox/mlx4/profile.c
index 8e0c3cc2a1ec..14089d9e1667 100644
--- a/drivers/net/ethernet/mellanox/mlx4/profile.c
+++ b/drivers/net/ethernet/mellanox/mlx4/profile.c
@@ -164,18 +164,17 @@ u64 mlx4_make_profile(struct mlx4_dev *dev,
 		}
 
 		if (total_size > dev_cap->max_icm_sz) {
-			mlx4_err(dev, "Profile requires 0x%llx bytes; "
-				  "won't fit in 0x%llx bytes of context memory.\n",
-				  (unsigned long long) total_size,
-				  (unsigned long long) dev_cap->max_icm_sz);
+			mlx4_err(dev, "Profile requires 0x%llx bytes; won't fit in 0x%llx bytes of context memory\n",
+				 (unsigned long long) total_size,
+				 (unsigned long long) dev_cap->max_icm_sz);
 			kfree(profile);
 			return -ENOMEM;
 		}
 
 		if (profile[i].size)
-			mlx4_dbg(dev, "  profile[%2d] (%6s): 2^%02d entries @ 0x%10llx, "
-				  "size 0x%10llx\n",
-				 i, res_name[profile[i].type], profile[i].log_num,
+			mlx4_dbg(dev, "  profile[%2d] (%6s): 2^%02d entries @ 0x%10llx, size 0x%10llx\n",
+				 i, res_name[profile[i].type],
+				 profile[i].log_num,
 				 (unsigned long long) profile[i].start,
 				 (unsigned long long) profile[i].size);
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c
index fbd32af89c7c..0dc31d85fc3b 100644
--- a/drivers/net/ethernet/mellanox/mlx4/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx4/qp.c
@@ -264,37 +264,37 @@ void mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt)
 			       MLX4_CMD_FREE_RES,
 			       MLX4_CMD_TIME_CLASS_A, MLX4_CMD_WRAPPED);
 		if (err) {
-			mlx4_warn(dev, "Failed to release qp range"
-				  " base:%d cnt:%d\n", base_qpn, cnt);
+			mlx4_warn(dev, "Failed to release qp range base:%d cnt:%d\n",
+				  base_qpn, cnt);
 		}
 	} else
 		 __mlx4_qp_release_range(dev, base_qpn, cnt);
 }
 EXPORT_SYMBOL_GPL(mlx4_qp_release_range);
 
-int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn)
+int __mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn, gfp_t gfp)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_qp_table *qp_table = &priv->qp_table;
 	int err;
 
-	err = mlx4_table_get(dev, &qp_table->qp_table, qpn);
+	err = mlx4_table_get(dev, &qp_table->qp_table, qpn, gfp);
 	if (err)
 		goto err_out;
 
-	err = mlx4_table_get(dev, &qp_table->auxc_table, qpn);
+	err = mlx4_table_get(dev, &qp_table->auxc_table, qpn, gfp);
 	if (err)
 		goto err_put_qp;
 
-	err = mlx4_table_get(dev, &qp_table->altc_table, qpn);
+	err = mlx4_table_get(dev, &qp_table->altc_table, qpn, gfp);
 	if (err)
 		goto err_put_auxc;
 
-	err = mlx4_table_get(dev, &qp_table->rdmarc_table, qpn);
+	err = mlx4_table_get(dev, &qp_table->rdmarc_table, qpn, gfp);
 	if (err)
 		goto err_put_altc;
 
-	err = mlx4_table_get(dev, &qp_table->cmpt_table, qpn);
+	err = mlx4_table_get(dev, &qp_table->cmpt_table, qpn, gfp);
 	if (err)
 		goto err_put_rdmarc;
 
@@ -316,7 +316,7 @@ err_out:
 	return err;
 }
 
-static int mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn)
+static int mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn, gfp_t gfp)
 {
 	u64 param = 0;
 
@@ -326,7 +326,7 @@ static int mlx4_qp_alloc_icm(struct mlx4_dev *dev, int qpn)
 				    MLX4_CMD_ALLOC_RES, MLX4_CMD_TIME_CLASS_A,
 				    MLX4_CMD_WRAPPED);
 	}
-	return __mlx4_qp_alloc_icm(dev, qpn);
+	return __mlx4_qp_alloc_icm(dev, qpn, gfp);
 }
 
 void __mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn)
@@ -355,7 +355,7 @@ static void mlx4_qp_free_icm(struct mlx4_dev *dev, int qpn)
 		__mlx4_qp_free_icm(dev, qpn);
 }
 
-int mlx4_qp_alloc(struct mlx4_dev *dev, int qpn, struct mlx4_qp *qp)
+int mlx4_qp_alloc(struct mlx4_dev *dev, int qpn, struct mlx4_qp *qp, gfp_t gfp)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_qp_table *qp_table = &priv->qp_table;
@@ -366,7 +366,7 @@ int mlx4_qp_alloc(struct mlx4_dev *dev, int qpn, struct mlx4_qp *qp)
 
 	qp->qpn = qpn;
 
-	err = mlx4_qp_alloc_icm(dev, qpn);
+	err = mlx4_qp_alloc_icm(dev, qpn, gfp);
 	if (err)
 		return err;
 
@@ -612,8 +612,7 @@ int mlx4_qp_to_ready(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 		err = mlx4_qp_modify(dev, mtt, states[i], states[i + 1],
 				     context, 0, 0, qp);
 		if (err) {
-			mlx4_err(dev, "Failed to bring QP to state: "
-				 "%d with error: %d\n",
+			mlx4_err(dev, "Failed to bring QP to state: %d with error: %d\n",
 				 states[i + 1], err);
 			return err;
 		}
diff --git a/drivers/net/ethernet/mellanox/mlx4/reset.c b/drivers/net/ethernet/mellanox/mlx4/reset.c
index dd1b5093d8b1..ea1c6d092145 100644
--- a/drivers/net/ethernet/mellanox/mlx4/reset.c
+++ b/drivers/net/ethernet/mellanox/mlx4/reset.c
@@ -72,8 +72,7 @@ int mlx4_reset(struct mlx4_dev *dev)
 	hca_header = kmalloc(256, GFP_KERNEL);
 	if (!hca_header) {
 		err = -ENOMEM;
-		mlx4_err(dev, "Couldn't allocate memory to save HCA "
-			  "PCI header, aborting.\n");
+		mlx4_err(dev, "Couldn't allocate memory to save HCA PCI header, aborting\n");
 		goto out;
 	}
 
@@ -84,8 +83,7 @@ int mlx4_reset(struct mlx4_dev *dev)
 			continue;
 		if (pci_read_config_dword(dev->pdev, i * 4, hca_header + i)) {
 			err = -ENODEV;
-			mlx4_err(dev, "Couldn't save HCA "
-				  "PCI header, aborting.\n");
+			mlx4_err(dev, "Couldn't save HCA PCI header, aborting\n");
 			goto out;
 		}
 	}
@@ -94,7 +92,7 @@ int mlx4_reset(struct mlx4_dev *dev)
 			MLX4_RESET_SIZE);
 	if (!reset) {
 		err = -ENOMEM;
-		mlx4_err(dev, "Couldn't map HCA reset register, aborting.\n");
+		mlx4_err(dev, "Couldn't map HCA reset register, aborting\n");
 		goto out;
 	}
 
@@ -133,8 +131,7 @@ int mlx4_reset(struct mlx4_dev *dev)
 
 	if (vendor == 0xffff) {
 		err = -ENODEV;
-		mlx4_err(dev, "PCI device did not come back after reset, "
-			  "aborting.\n");
+		mlx4_err(dev, "PCI device did not come back after reset, aborting\n");
 		goto out;
 	}
 
@@ -144,16 +141,14 @@ int mlx4_reset(struct mlx4_dev *dev)
 		if (pcie_capability_write_word(dev->pdev, PCI_EXP_DEVCTL,
 					       devctl)) {
 			err = -ENODEV;
-			mlx4_err(dev, "Couldn't restore HCA PCI Express "
-				 "Device Control register, aborting.\n");
+			mlx4_err(dev, "Couldn't restore HCA PCI Express Device Control register, aborting\n");
 			goto out;
 		}
 		linkctl = hca_header[(pcie_cap + PCI_EXP_LNKCTL) / 4];
 		if (pcie_capability_write_word(dev->pdev, PCI_EXP_LNKCTL,
 					       linkctl)) {
 			err = -ENODEV;
-			mlx4_err(dev, "Couldn't restore HCA PCI Express "
-				 "Link control register, aborting.\n");
+			mlx4_err(dev, "Couldn't restore HCA PCI Express Link control register, aborting\n");
 			goto out;
 		}
 	}
@@ -164,8 +159,8 @@ int mlx4_reset(struct mlx4_dev *dev)
 
 		if (pci_write_config_dword(dev->pdev, i * 4, hca_header[i])) {
 			err = -ENODEV;
-			mlx4_err(dev, "Couldn't restore HCA reg %x, "
-				  "aborting.\n", i);
+			mlx4_err(dev, "Couldn't restore HCA reg %x, aborting\n",
+				 i);
 			goto out;
 		}
 	}
@@ -173,8 +168,7 @@ int mlx4_reset(struct mlx4_dev *dev)
 	if (pci_write_config_dword(dev->pdev, PCI_COMMAND,
 				   hca_header[PCI_COMMAND / 4])) {
 		err = -ENODEV;
-		mlx4_err(dev, "Couldn't restore HCA COMMAND, "
-			  "aborting.\n");
+		mlx4_err(dev, "Couldn't restore HCA COMMAND, aborting\n");
 		goto out;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index f16e539749c4..0efc1368e5a8 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -279,7 +279,7 @@ enum qp_transition {
 };
 
 /* For Debug uses */
-static const char *ResourceType(enum mlx4_resource rt)
+static const char *resource_str(enum mlx4_resource rt)
 {
 	switch (rt) {
 	case RES_QP: return "RES_QP";
@@ -307,6 +307,7 @@ static inline int mlx4_grant_resource(struct mlx4_dev *dev, int slave,
 		&priv->mfunc.master.res_tracker.res_alloc[res_type];
 	int err = -EINVAL;
 	int allocated, free, reserved, guaranteed, from_free;
+	int from_rsvd;
 
 	if (slave > dev->num_vfs)
 		return -EINVAL;
@@ -321,11 +322,16 @@ static inline int mlx4_grant_resource(struct mlx4_dev *dev, int slave,
 		res_alloc->res_reserved;
 	guaranteed = res_alloc->guaranteed[slave];
 
-	if (allocated + count > res_alloc->quota[slave])
+	if (allocated + count > res_alloc->quota[slave]) {
+		mlx4_warn(dev, "VF %d port %d res %s: quota exceeded, count %d alloc %d quota %d\n",
+			  slave, port, resource_str(res_type), count,
+			  allocated, res_alloc->quota[slave]);
 		goto out;
+	}
 
 	if (allocated + count <= guaranteed) {
 		err = 0;
+		from_rsvd = count;
 	} else {
 		/* portion may need to be obtained from free area */
 		if (guaranteed - allocated > 0)
@@ -333,8 +339,14 @@ static inline int mlx4_grant_resource(struct mlx4_dev *dev, int slave,
 		else
 			from_free = count;
 
-		if (free - from_free > reserved)
+		from_rsvd = count - from_free;
+
+		if (free - from_free >= reserved)
 			err = 0;
+		else
+			mlx4_warn(dev, "VF %d port %d res %s: free pool empty, free %d from_free %d rsvd %d\n",
+				  slave, port, resource_str(res_type), free,
+				  from_free, reserved);
 	}
 
 	if (!err) {
@@ -342,9 +354,11 @@ static inline int mlx4_grant_resource(struct mlx4_dev *dev, int slave,
 		if (port > 0) {
 			res_alloc->allocated[(port - 1) * (dev->num_vfs + 1) + slave] += count;
 			res_alloc->res_port_free[port - 1] -= count;
+			res_alloc->res_port_rsvd[port - 1] -= from_rsvd;
 		} else {
 			res_alloc->allocated[slave] += count;
 			res_alloc->res_free -= count;
+			res_alloc->res_reserved -= from_rsvd;
 		}
 	}
 
@@ -360,17 +374,36 @@ static inline void mlx4_release_resource(struct mlx4_dev *dev, int slave,
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct resource_allocator *res_alloc =
 		&priv->mfunc.master.res_tracker.res_alloc[res_type];
+	int allocated, guaranteed, from_rsvd;
 
 	if (slave > dev->num_vfs)
 		return;
 
 	spin_lock(&res_alloc->alloc_lock);
+
+	allocated = (port > 0) ?
+		res_alloc->allocated[(port - 1) * (dev->num_vfs + 1) + slave] :
+		res_alloc->allocated[slave];
+	guaranteed = res_alloc->guaranteed[slave];
+
+	if (allocated - count >= guaranteed) {
+		from_rsvd = 0;
+	} else {
+		/* portion may need to be returned to reserved area */
+		if (allocated - guaranteed > 0)
+			from_rsvd = count - (allocated - guaranteed);
+		else
+			from_rsvd = count;
+	}
+
 	if (port > 0) {
 		res_alloc->allocated[(port - 1) * (dev->num_vfs + 1) + slave] -= count;
 		res_alloc->res_port_free[port - 1] += count;
+		res_alloc->res_port_rsvd[port - 1] += from_rsvd;
 	} else {
 		res_alloc->allocated[slave] -= count;
 		res_alloc->res_free += count;
+		res_alloc->res_reserved += from_rsvd;
 	}
 
 	spin_unlock(&res_alloc->alloc_lock);
@@ -963,7 +996,7 @@ static struct res_common *alloc_tr(u64 id, enum mlx4_resource type, int slave,
 		ret = alloc_srq_tr(id);
 		break;
 	case RES_MAC:
-		printk(KERN_ERR "implementation missing\n");
+		pr_err("implementation missing\n");
 		return NULL;
 	case RES_COUNTER:
 		ret = alloc_counter_tr(id);
@@ -1057,10 +1090,10 @@ static int remove_mtt_ok(struct res_mtt *res, int order)
 {
 	if (res->com.state == RES_MTT_BUSY ||
 	    atomic_read(&res->ref_count)) {
-		printk(KERN_DEBUG "%s-%d: state %s, ref_count %d\n",
-		       __func__, __LINE__,
-		       mtt_states_str(res->com.state),
-		       atomic_read(&res->ref_count));
+		pr_devel("%s-%d: state %s, ref_count %d\n",
+			 __func__, __LINE__,
+			 mtt_states_str(res->com.state),
+			 atomic_read(&res->ref_count));
 		return -EBUSY;
 	} else if (res->com.state != RES_MTT_ALLOCATED)
 		return -EPERM;
@@ -1533,7 +1566,7 @@ static int qp_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 			return err;
 
 		if (!fw_reserved(dev, qpn)) {
-			err = __mlx4_qp_alloc_icm(dev, qpn);
+			err = __mlx4_qp_alloc_icm(dev, qpn, GFP_KERNEL);
 			if (err) {
 				res_abort_move(dev, slave, RES_QP, qpn);
 				return err;
@@ -1620,7 +1653,7 @@ static int mpt_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 		if (err)
 			return err;
 
-		err = __mlx4_mpt_alloc_icm(dev, mpt->key);
+		err = __mlx4_mpt_alloc_icm(dev, mpt->key, GFP_KERNEL);
 		if (err) {
 			res_abort_move(dev, slave, RES_MPT, id);
 			return err;
@@ -2828,10 +2861,12 @@ static int get_containing_mtt(struct mlx4_dev *dev, int slave, int start,
 }
 
 static int verify_qp_parameters(struct mlx4_dev *dev,
+				struct mlx4_vhcr *vhcr,
 				struct mlx4_cmd_mailbox *inbox,
 				enum qp_transition transition, u8 slave)
 {
 	u32			qp_type;
+	u32			qpn;
 	struct mlx4_qp_context	*qp_ctx;
 	enum mlx4_qp_optpar	optpar;
 	int port;
@@ -2874,8 +2909,22 @@ static int verify_qp_parameters(struct mlx4_dev *dev,
 		default:
 			break;
 		}
+		break;
 
+	case MLX4_QP_ST_MLX:
+		qpn = vhcr->in_modifier & 0x7fffff;
+		port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
+		if (transition == QP_TRANS_INIT2RTR &&
+		    slave != mlx4_master_func_num(dev) &&
+		    mlx4_is_qp_reserved(dev, qpn) &&
+		    !mlx4_vf_smi_enabled(dev, slave, port)) {
+			/* only enabled VFs may create MLX proxy QPs */
+			mlx4_err(dev, "%s: unprivileged slave %d attempting to create an MLX proxy special QP on port %d\n",
+				 __func__, slave, port);
+			return -EPERM;
+		}
 		break;
+
 	default:
 		break;
 	}
@@ -3455,7 +3504,7 @@ int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave,
 	err = adjust_qp_sched_queue(dev, slave, qpc, inbox);
 	if (err)
 		return err;
-	err = verify_qp_parameters(dev, inbox, QP_TRANS_INIT2RTR, slave);
+	err = verify_qp_parameters(dev, vhcr, inbox, QP_TRANS_INIT2RTR, slave);
 	if (err)
 		return err;
 
@@ -3509,7 +3558,7 @@ int mlx4_RTR2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 	err = adjust_qp_sched_queue(dev, slave, context, inbox);
 	if (err)
 		return err;
-	err = verify_qp_parameters(dev, inbox, QP_TRANS_RTR2RTS, slave);
+	err = verify_qp_parameters(dev, vhcr, inbox, QP_TRANS_RTR2RTS, slave);
 	if (err)
 		return err;
 
@@ -3531,7 +3580,7 @@ int mlx4_RTS2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 	err = adjust_qp_sched_queue(dev, slave, context, inbox);
 	if (err)
 		return err;
-	err = verify_qp_parameters(dev, inbox, QP_TRANS_RTS2RTS, slave);
+	err = verify_qp_parameters(dev, vhcr, inbox, QP_TRANS_RTS2RTS, slave);
 	if (err)
 		return err;
 
@@ -3568,7 +3617,7 @@ int mlx4_SQD2SQD_QP_wrapper(struct mlx4_dev *dev, int slave,
 	err = adjust_qp_sched_queue(dev, slave, context, inbox);
 	if (err)
 		return err;
-	err = verify_qp_parameters(dev, inbox, QP_TRANS_SQD2SQD, slave);
+	err = verify_qp_parameters(dev, vhcr, inbox, QP_TRANS_SQD2SQD, slave);
 	if (err)
 		return err;
 
@@ -3590,7 +3639,7 @@ int mlx4_SQD2RTS_QP_wrapper(struct mlx4_dev *dev, int slave,
 	err = adjust_qp_sched_queue(dev, slave, context, inbox);
 	if (err)
 		return err;
-	err = verify_qp_parameters(dev, inbox, QP_TRANS_SQD2RTS, slave);
+	err = verify_qp_parameters(dev, vhcr, inbox, QP_TRANS_SQD2RTS, slave);
 	if (err)
 		return err;
 
@@ -3881,7 +3930,7 @@ static int add_eth_header(struct mlx4_dev *dev, int slave,
 		}
 	}
 	if (!be_mac) {
-		pr_err("Failed adding eth header to FS rule, Can't find matching MAC for port %d .\n",
+		pr_err("Failed adding eth header to FS rule, Can't find matching MAC for port %d\n",
 		       port);
 		return -EINVAL;
 	}
@@ -3978,7 +4027,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	qpn = be32_to_cpu(ctrl->qpn) & 0xffffff;
 	err = get_res(dev, slave, qpn, RES_QP, &rqp);
 	if (err) {
-		pr_err("Steering rule with qpn 0x%x rejected.\n", qpn);
+		pr_err("Steering rule with qpn 0x%x rejected\n", qpn);
 		return err;
 	}
 	rule_header = (struct _rule_hw *)(ctrl + 1);
@@ -3996,7 +4045,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	case MLX4_NET_TRANS_RULE_ID_IPV4:
 	case MLX4_NET_TRANS_RULE_ID_TCP:
 	case MLX4_NET_TRANS_RULE_ID_UDP:
-		pr_warn("Can't attach FS rule without L2 headers, adding L2 header.\n");
+		pr_warn("Can't attach FS rule without L2 headers, adding L2 header\n");
 		if (add_eth_header(dev, slave, inbox, rlist, header_id)) {
 			err = -EINVAL;
 			goto err_put;
@@ -4005,7 +4054,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 			sizeof(struct mlx4_net_trans_rule_hw_eth) >> 2;
 		break;
 	default:
-		pr_err("Corrupted mailbox.\n");
+		pr_err("Corrupted mailbox\n");
 		err = -EINVAL;
 		goto err_put;
 	}
@@ -4019,7 +4068,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 
 	err = add_res_range(dev, slave, vhcr->out_param, 1, RES_FS_RULE, qpn);
 	if (err) {
-		mlx4_err(dev, "Fail to add flow steering resources.\n ");
+		mlx4_err(dev, "Fail to add flow steering resources\n");
 		/* detach rule*/
 		mlx4_cmd(dev, vhcr->out_param, 0, 0,
 			 MLX4_QP_FLOW_STEERING_DETACH, MLX4_CMD_TIME_CLASS_A,
@@ -4057,7 +4106,7 @@ int mlx4_QP_FLOW_STEERING_DETACH_wrapper(struct mlx4_dev *dev, int slave,
 
 	err = rem_res_range(dev, slave, vhcr->in_param, 1, RES_FS_RULE, 0);
 	if (err) {
-		mlx4_err(dev, "Fail to remove flow steering resources.\n ");
+		mlx4_err(dev, "Fail to remove flow steering resources\n");
 		goto out;
 	}
 
@@ -4135,7 +4184,7 @@ static int _move_all_busy(struct mlx4_dev *dev, int slave,
 					if (print)
 						mlx4_dbg(dev,
 							 "%s id 0x%llx is busy\n",
-							  ResourceType(type),
+							  resource_str(type),
 							  r->res_id);
 					++busy;
 				} else {
@@ -4186,8 +4235,8 @@ static void rem_slave_qps(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_QP);
 	if (err)
-		mlx4_warn(dev, "rem_slave_qps: Could not move all qps to busy"
-			  "for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_qps: Could not move all qps to busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(qp, tmp, qp_list, com.list) {
@@ -4225,10 +4274,8 @@ static void rem_slave_qps(struct mlx4_dev *dev, int slave)
 						       MLX4_CMD_TIME_CLASS_A,
 						       MLX4_CMD_NATIVE);
 					if (err)
-						mlx4_dbg(dev, "rem_slave_qps: failed"
-							 " to move slave %d qpn %d to"
-							 " reset\n", slave,
-							 qp->local_qpn);
+						mlx4_dbg(dev, "rem_slave_qps: failed to move slave %d qpn %d to reset\n",
+							 slave, qp->local_qpn);
 					atomic_dec(&qp->rcq->ref_count);
 					atomic_dec(&qp->scq->ref_count);
 					atomic_dec(&qp->mtt->ref_count);
@@ -4262,8 +4309,8 @@ static void rem_slave_srqs(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_SRQ);
 	if (err)
-		mlx4_warn(dev, "rem_slave_srqs: Could not move all srqs to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_srqs: Could not move all srqs - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(srq, tmp, srq_list, com.list) {
@@ -4293,9 +4340,7 @@ static void rem_slave_srqs(struct mlx4_dev *dev, int slave)
 						       MLX4_CMD_TIME_CLASS_A,
 						       MLX4_CMD_NATIVE);
 					if (err)
-						mlx4_dbg(dev, "rem_slave_srqs: failed"
-							 " to move slave %d srq %d to"
-							 " SW ownership\n",
+						mlx4_dbg(dev, "rem_slave_srqs: failed to move slave %d srq %d to SW ownership\n",
 							 slave, srqn);
 
 					atomic_dec(&srq->mtt->ref_count);
@@ -4330,8 +4375,8 @@ static void rem_slave_cqs(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_CQ);
 	if (err)
-		mlx4_warn(dev, "rem_slave_cqs: Could not move all cqs to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_cqs: Could not move all cqs - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(cq, tmp, cq_list, com.list) {
@@ -4361,9 +4406,7 @@ static void rem_slave_cqs(struct mlx4_dev *dev, int slave)
 						       MLX4_CMD_TIME_CLASS_A,
 						       MLX4_CMD_NATIVE);
 					if (err)
-						mlx4_dbg(dev, "rem_slave_cqs: failed"
-							 " to move slave %d cq %d to"
-							 " SW ownership\n",
+						mlx4_dbg(dev, "rem_slave_cqs: failed to move slave %d cq %d to SW ownership\n",
 							 slave, cqn);
 					atomic_dec(&cq->mtt->ref_count);
 					state = RES_CQ_ALLOCATED;
@@ -4395,8 +4438,8 @@ static void rem_slave_mrs(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_MPT);
 	if (err)
-		mlx4_warn(dev, "rem_slave_mrs: Could not move all mpts to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_mrs: Could not move all mpts - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(mpt, tmp, mpt_list, com.list) {
@@ -4431,9 +4474,7 @@ static void rem_slave_mrs(struct mlx4_dev *dev, int slave)
 						     MLX4_CMD_TIME_CLASS_A,
 						     MLX4_CMD_NATIVE);
 					if (err)
-						mlx4_dbg(dev, "rem_slave_mrs: failed"
-							 " to move slave %d mpt %d to"
-							 " SW ownership\n",
+						mlx4_dbg(dev, "rem_slave_mrs: failed to move slave %d mpt %d to SW ownership\n",
 							 slave, mptn);
 					if (mpt->mtt)
 						atomic_dec(&mpt->mtt->ref_count);
@@ -4465,8 +4506,8 @@ static void rem_slave_mtts(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_MTT);
 	if (err)
-		mlx4_warn(dev, "rem_slave_mtts: Could not move all mtts to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_mtts: Could not move all mtts  - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(mtt, tmp, mtt_list, com.list) {
@@ -4568,8 +4609,8 @@ static void rem_slave_eqs(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_EQ);
 	if (err)
-		mlx4_warn(dev, "rem_slave_eqs: Could not move all eqs to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_eqs: Could not move all eqs - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(eq, tmp, eq_list, com.list) {
@@ -4601,9 +4642,8 @@ static void rem_slave_eqs(struct mlx4_dev *dev, int slave)
 							   MLX4_CMD_TIME_CLASS_A,
 							   MLX4_CMD_NATIVE);
 					if (err)
-						mlx4_dbg(dev, "rem_slave_eqs: failed"
-							 " to move slave %d eqs %d to"
-							 " SW ownership\n", slave, eqn);
+						mlx4_dbg(dev, "rem_slave_eqs: failed to move slave %d eqs %d to SW ownership\n",
+							 slave, eqn);
 					mlx4_free_cmd_mailbox(dev, mailbox);
 					atomic_dec(&eq->mtt->ref_count);
 					state = RES_EQ_RESERVED;
@@ -4632,8 +4672,8 @@ static void rem_slave_counters(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_COUNTER);
 	if (err)
-		mlx4_warn(dev, "rem_slave_counters: Could not move all counters to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_counters: Could not move all counters - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(counter, tmp, counter_list, com.list) {
@@ -4663,8 +4703,8 @@ static void rem_slave_xrcdns(struct mlx4_dev *dev, int slave)
 
 	err = move_all_busy(dev, slave, RES_XRCD);
 	if (err)
-		mlx4_warn(dev, "rem_slave_xrcdns: Could not move all xrcdns to "
-			  "busy for slave %d\n", slave);
+		mlx4_warn(dev, "rem_slave_xrcdns: Could not move all xrcdns - too busy for slave %d\n",
+			  slave);
 
 	spin_lock_irq(mlx4_tlock(dev));
 	list_for_each_entry_safe(xrcd, tmp, xrcdn_list, com.list) {
@@ -4809,10 +4849,8 @@ void mlx4_vf_immed_vlan_work_handler(struct work_struct *_work)
 				       0, MLX4_CMD_UPDATE_QP,
 				       MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE);
 			if (err) {
-				mlx4_info(dev, "UPDATE_QP failed for slave %d, "
-					  "port %d, qpn %d (%d)\n",
-					  work->slave, port, qp->local_qpn,
-					  err);
+				mlx4_info(dev, "UPDATE_QP failed for slave %d, port %d, qpn %d (%d)\n",
+					  work->slave, port, qp->local_qpn, err);
 				errors++;
 			}
 		}
diff --git a/drivers/net/ethernet/mellanox/mlx4/srq.c b/drivers/net/ethernet/mellanox/mlx4/srq.c
index 98faf870b0b0..67146624eb58 100644
--- a/drivers/net/ethernet/mellanox/mlx4/srq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/srq.c
@@ -103,11 +103,11 @@ int __mlx4_srq_alloc_icm(struct mlx4_dev *dev, int *srqn)
 	if (*srqn == -1)
 		return -ENOMEM;
 
-	err = mlx4_table_get(dev, &srq_table->table, *srqn);
+	err = mlx4_table_get(dev, &srq_table->table, *srqn, GFP_KERNEL);
 	if (err)
 		goto err_out;
 
-	err = mlx4_table_get(dev, &srq_table->cmpt_table, *srqn);
+	err = mlx4_table_get(dev, &srq_table->cmpt_table, *srqn, GFP_KERNEL);
 	if (err)
 		goto err_put;
 	return 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 405c4fbcd0ad..87d1b018a9c3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -620,8 +620,8 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent)
 			       mlx5_command_str(msg_to_opcode(ent->in)),
 			       msg_to_opcode(ent->in));
 	}
-	mlx5_core_dbg(dev, "err %d, delivery status %s(%d)\n", err,
-		      deliv_status_to_str(ent->status), ent->status);
+	mlx5_core_dbg(dev, "err %d, delivery status %s(%d)\n",
+		      err, deliv_status_to_str(ent->status), ent->status);
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 64a61b286b2c..7f39ebcd6ad0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -208,7 +208,8 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 		 */
 		rmb();
 
-		mlx5_core_dbg(eq->dev, "eqn %d, eqe type %s\n", eq->eqn, eqe_type_str(eqe->type));
+		mlx5_core_dbg(eq->dev, "eqn %d, eqe type %s\n",
+			      eq->eqn, eqe_type_str(eqe->type));
 		switch (eqe->type) {
 		case MLX5_EVENT_TYPE_COMP:
 			cqn = be32_to_cpu(eqe->data.comp.cqn) & 0xffffff;
@@ -270,14 +271,16 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 				u16 func_id = be16_to_cpu(eqe->data.req_pages.func_id);
 				s32 npages = be32_to_cpu(eqe->data.req_pages.num_pages);
 
-				mlx5_core_dbg(dev, "page request for func 0x%x, napges %d\n", func_id, npages);
+				mlx5_core_dbg(dev, "page request for func 0x%x, npages %d\n",
+					      func_id, npages);
 				mlx5_core_req_pages_handler(dev, func_id, npages);
 			}
 			break;
 
 
 		default:
-			mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n", eqe->type, eq->eqn);
+			mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n",
+				       eqe->type, eq->eqn);
 			break;
 		}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index c3eee5f70051..ee24f132e319 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -66,10 +66,10 @@ static int set_dma_caps(struct pci_dev *pdev)
 
 	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
 	if (err) {
-		dev_warn(&pdev->dev, "Warning: couldn't set 64-bit PCI DMA mask.\n");
+		dev_warn(&pdev->dev, "Warning: couldn't set 64-bit PCI DMA mask\n");
 		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (err) {
-			dev_err(&pdev->dev, "Can't set PCI DMA mask, aborting.\n");
+			dev_err(&pdev->dev, "Can't set PCI DMA mask, aborting\n");
 			return err;
 		}
 	}
@@ -77,11 +77,11 @@ static int set_dma_caps(struct pci_dev *pdev)
 	err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
 	if (err) {
 		dev_warn(&pdev->dev,
-			 "Warning: couldn't set 64-bit consistent PCI DMA mask.\n");
+			 "Warning: couldn't set 64-bit consistent PCI DMA mask\n");
 		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (err) {
 			dev_err(&pdev->dev,
-				"Can't set consistent PCI DMA mask, aborting.\n");
+				"Can't set consistent PCI DMA mask, aborting\n");
 			return err;
 		}
 	}
@@ -95,7 +95,7 @@ static int request_bar(struct pci_dev *pdev)
 	int err = 0;
 
 	if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
-		dev_err(&pdev->dev, "Missing registers BAR, aborting.\n");
+		dev_err(&pdev->dev, "Missing registers BAR, aborting\n");
 		return -ENODEV;
 	}
 
@@ -319,13 +319,13 @@ int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev)
 
 	err = pci_enable_device(pdev);
 	if (err) {
-		dev_err(&pdev->dev, "Cannot enable PCI device, aborting.\n");
+		dev_err(&pdev->dev, "Cannot enable PCI device, aborting\n");
 		goto err_dbg;
 	}
 
 	err = request_bar(pdev);
 	if (err) {
-		dev_err(&pdev->dev, "error requesting BARs, aborting.\n");
+		dev_err(&pdev->dev, "error requesting BARs, aborting\n");
 		goto err_disable;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 68b74e1ae1b0..f0c9f9a7a361 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -39,24 +39,26 @@
 
 extern int mlx5_core_debug_mask;
 
-#define mlx5_core_dbg(dev, format, arg...)				       \
-pr_debug("%s:%s:%d:(pid %d): " format, (dev)->priv.name, __func__, __LINE__,   \
-	 current->pid, ##arg)
+#define mlx5_core_dbg(dev, format, ...)					\
+	pr_debug("%s:%s:%d:(pid %d): " format,				\
+		 (dev)->priv.name, __func__, __LINE__, current->pid,	\
+		 ##__VA_ARGS__)
 
-#define mlx5_core_dbg_mask(dev, mask, format, arg...)			       \
-do {									       \
-	if ((mask) & mlx5_core_debug_mask)				       \
-		pr_debug("%s:%s:%d:(pid %d): " format, (dev)->priv.name,       \
-			 __func__, __LINE__, current->pid, ##arg);	       \
+#define mlx5_core_dbg_mask(dev, mask, format, ...)			\
+do {									\
+	if ((mask) & mlx5_core_debug_mask)				\
+		mlx5_core_dbg(dev, format, ##__VA_ARGS__);		\
 } while (0)
 
-#define mlx5_core_err(dev, format, arg...) \
-pr_err("%s:%s:%d:(pid %d): " format, (dev)->priv.name, __func__, __LINE__,     \
-	current->pid, ##arg)
+#define mlx5_core_err(dev, format, ...)					\
+	pr_err("%s:%s:%d:(pid %d): " format,				\
+	       (dev)->priv.name, __func__, __LINE__, current->pid,	\
+	       ##__VA_ARGS__)
 
-#define mlx5_core_warn(dev, format, arg...) \
-pr_warn("%s:%s:%d:(pid %d): " format, (dev)->priv.name, __func__, __LINE__,    \
-	current->pid, ##arg)
+#define mlx5_core_warn(dev, format, ...)				\
+	pr_warn("%s:%s:%d:(pid %d): " format,				\
+		(dev)->priv.name, __func__, __LINE__, current->pid,	\
+		##__VA_ARGS__)
 
 enum {
 	MLX5_CMD_DATA, /* print command payload only */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c
index 4cc927649404..ba0401d4af50 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c
@@ -73,7 +73,7 @@ int mlx5_core_create_mkey(struct mlx5_core_dev *dev, struct mlx5_core_mr *mr,
 	}
 
 	if (err) {
-		mlx5_core_dbg(dev, "cmd exec faile %d\n", err);
+		mlx5_core_dbg(dev, "cmd exec failed %d\n", err);
 		return err;
 	}
 
@@ -82,7 +82,11 @@ int mlx5_core_create_mkey(struct mlx5_core_dev *dev, struct mlx5_core_mr *mr,
 		return mlx5_cmd_status_to_err(&lout.hdr);
 	}
 
+	mr->iova = be64_to_cpu(in->seg.start_addr);
+	mr->size = be64_to_cpu(in->seg.len);
 	mr->key = mlx5_idx_to_mkey(be32_to_cpu(lout.mkey) & 0xffffff) | key;
+	mr->pd = be32_to_cpu(in->seg.flags_pd) & 0xffffff;
+
 	mlx5_core_dbg(dev, "out 0x%x, key 0x%x, mkey 0x%x\n",
 		      be32_to_cpu(lout.mkey), key, mr->key);
 
@@ -191,7 +195,8 @@ int mlx5_core_create_psv(struct mlx5_core_dev *dev, u32 pdn,
 	}
 
 	if (out.hdr.status) {
-		mlx5_core_err(dev, "create_psv bad status %d\n", out.hdr.status);
+		mlx5_core_err(dev, "create_psv bad status %d\n",
+			      out.hdr.status);
 		return mlx5_cmd_status_to_err(&out.hdr);
 	}
 
@@ -220,7 +225,8 @@ int mlx5_core_destroy_psv(struct mlx5_core_dev *dev, int psv_num)
 	}
 
 	if (out.hdr.status) {
-		mlx5_core_err(dev, "destroy_psv bad status %d\n", out.hdr.status);
+		mlx5_core_err(dev, "destroy_psv bad status %d\n",
+			      out.hdr.status);
 		err = mlx5_cmd_status_to_err(&out.hdr);
 		goto out;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
index d59790a82bc3..c2a953ef0e67 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
@@ -311,7 +311,8 @@ retry:
 	in->num_entries = cpu_to_be32(npages);
 	err = mlx5_cmd_exec(dev, in, inlen, &out, sizeof(out));
 	if (err) {
-		mlx5_core_warn(dev, "func_id 0x%x, npages %d, err %d\n", func_id, npages, err);
+		mlx5_core_warn(dev, "func_id 0x%x, npages %d, err %d\n",
+			       func_id, npages, err);
 		goto out_alloc;
 	}
 	dev->priv.fw_pages += npages;
@@ -319,7 +320,8 @@ retry:
 	if (out.hdr.status) {
 		err = mlx5_cmd_status_to_err(&out.hdr);
 		if (err) {
-			mlx5_core_warn(dev, "func_id 0x%x, npages %d, status %d\n", func_id, npages, out.hdr.status);
+			mlx5_core_warn(dev, "func_id 0x%x, npages %d, status %d\n",
+				       func_id, npages, out.hdr.status);
 			goto out_alloc;
 		}
 	}
@@ -378,7 +380,7 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
 	mlx5_core_dbg(dev, "npages %d, outlen %d\n", npages, outlen);
 	err = mlx5_cmd_exec(dev, &in, sizeof(in), out, outlen);
 	if (err) {
-		mlx5_core_err(dev, "failed recliaming pages\n");
+		mlx5_core_err(dev, "failed reclaiming pages\n");
 		goto out_free;
 	}
 	dev->priv.fw_pages -= npages;
@@ -414,8 +416,8 @@ static void pages_work_handler(struct work_struct *work)
 		err = give_pages(dev, req->func_id, req->npages, 1);
 
 	if (err)
-		mlx5_core_warn(dev, "%s fail %d\n", req->npages < 0 ?
-			       "reclaim" : "give", err);
+		mlx5_core_warn(dev, "%s fail %d\n",
+			       req->npages < 0 ? "reclaim" : "give", err);
 
 	kfree(req);
 }
@@ -487,7 +489,8 @@ int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev)
 					    optimal_reclaimed_pages(),
 					    &nclaimed);
 			if (err) {
-				mlx5_core_warn(dev, "failed reclaiming pages (%d)\n", err);
+				mlx5_core_warn(dev, "failed reclaiming pages (%d)\n",
+					       err);
 				return err;
 			}
 			if (nclaimed)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
index 510576213dd0..8145b4668229 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
@@ -79,7 +79,7 @@ int mlx5_core_create_qp(struct mlx5_core_dev *dev,
 
 	err = mlx5_cmd_exec(dev, in, inlen, &out, sizeof(out));
 	if (err) {
-		mlx5_core_warn(dev, "ret %d", err);
+		mlx5_core_warn(dev, "ret %d\n", err);
 		return err;
 	}
 
@@ -96,7 +96,7 @@ int mlx5_core_create_qp(struct mlx5_core_dev *dev,
 	err = radix_tree_insert(&table->tree, qp->qpn, qp);
 	spin_unlock_irq(&table->lock);
 	if (err) {
-		mlx5_core_warn(dev, "err %d", err);
+		mlx5_core_warn(dev, "err %d\n", err);
 		goto err_cmd;
 	}
 
diff --git a/drivers/net/ethernet/micrel/ks8695net.c b/drivers/net/ethernet/micrel/ks8695net.c
index 16435b3cfa9f..6c7c78baedca 100644
--- a/drivers/net/ethernet/micrel/ks8695net.c
+++ b/drivers/net/ethernet/micrel/ks8695net.c
@@ -1504,15 +1504,15 @@ ks8695_probe(struct platform_device *pdev)
 	if (ksp->phyiface_regs && ksp->link_irq == -1) {
 		ks8695_init_switch(ksp);
 		ksp->dtype = KS8695_DTYPE_LAN;
-		SET_ETHTOOL_OPS(ndev, &ks8695_ethtool_ops);
+		ndev->ethtool_ops = &ks8695_ethtool_ops;
 	} else if (ksp->phyiface_regs && ksp->link_irq != -1) {
 		ks8695_init_wan_phy(ksp);
 		ksp->dtype = KS8695_DTYPE_WAN;
-		SET_ETHTOOL_OPS(ndev, &ks8695_wan_ethtool_ops);
+		ndev->ethtool_ops = &ks8695_wan_ethtool_ops;
 	} else {
 		/* No initialisation since HPNA does not have a PHY */
 		ksp->dtype = KS8695_DTYPE_HPNA;
-		SET_ETHTOOL_OPS(ndev, &ks8695_ethtool_ops);
+		ndev->ethtool_ops = &ks8695_ethtool_ops;
 	}
 
 	/* And bring up the net_device with the net core */
diff --git a/drivers/net/ethernet/micrel/ks8851.c b/drivers/net/ethernet/micrel/ks8851.c
index e0c92e0e5e1d..66d4ab703f45 100644
--- a/drivers/net/ethernet/micrel/ks8851.c
+++ b/drivers/net/ethernet/micrel/ks8851.c
@@ -26,6 +26,8 @@
 #include <linux/regulator/consumer.h>
 
 #include <linux/spi/spi.h>
+#include <linux/gpio.h>
+#include <linux/of_gpio.h>
 
 #include "ks8851.h"
 
@@ -85,6 +87,8 @@ union ks8851_tx_hdr {
  * @eeprom_size: Companion eeprom size in Bytes, 0 if no eeprom
  * @eeprom: 93CX6 EEPROM state for accessing on-board EEPROM.
  * @vdd_reg:	Optional regulator supplying the chip
+ * @vdd_io: Optional digital power supply for IO
+ * @gpio: Optional reset_n gpio
  *
  * The @lock ensures that the chip is protected when certain operations are
  * in progress. When the read or write packet transfer is in progress, most
@@ -133,6 +137,8 @@ struct ks8851_net {
 
 	struct eeprom_93cx6	eeprom;
 	struct regulator	*vdd_reg;
+	struct regulator	*vdd_io;
+	int			gpio;
 };
 
 static int msg_enable;
@@ -1404,6 +1410,7 @@ static int ks8851_probe(struct spi_device *spi)
 	struct ks8851_net *ks;
 	int ret;
 	unsigned cider;
+	int gpio;
 
 	ndev = alloc_etherdev(sizeof(struct ks8851_net));
 	if (!ndev)
@@ -1417,20 +1424,53 @@ static int ks8851_probe(struct spi_device *spi)
 	ks->spidev = spi;
 	ks->tx_space = 6144;
 
-	ks->vdd_reg = regulator_get_optional(&spi->dev, "vdd");
-	if (IS_ERR(ks->vdd_reg)) {
-		ret = PTR_ERR(ks->vdd_reg);
-		if (ret == -EPROBE_DEFER)
-			goto err_reg;
-	} else {
-		ret = regulator_enable(ks->vdd_reg);
+	gpio = of_get_named_gpio_flags(spi->dev.of_node, "reset-gpios",
+				       0, NULL);
+	if (gpio == -EPROBE_DEFER) {
+		ret = gpio;
+		goto err_gpio;
+	}
+
+	ks->gpio = gpio;
+	if (gpio_is_valid(gpio)) {
+		ret = devm_gpio_request_one(&spi->dev, gpio,
+					    GPIOF_OUT_INIT_LOW, "ks8851_rst_n");
 		if (ret) {
-			dev_err(&spi->dev, "regulator enable fail: %d\n",
-				ret);
-			goto err_reg_en;
+			dev_err(&spi->dev, "reset gpio request failed\n");
+			goto err_gpio;
 		}
 	}
 
+	ks->vdd_io = devm_regulator_get(&spi->dev, "vdd-io");
+	if (IS_ERR(ks->vdd_io)) {
+		ret = PTR_ERR(ks->vdd_io);
+		goto err_reg_io;
+	}
+
+	ret = regulator_enable(ks->vdd_io);
+	if (ret) {
+		dev_err(&spi->dev, "regulator vdd_io enable fail: %d\n",
+			ret);
+		goto err_reg_io;
+	}
+
+	ks->vdd_reg = devm_regulator_get(&spi->dev, "vdd");
+	if (IS_ERR(ks->vdd_reg)) {
+		ret = PTR_ERR(ks->vdd_reg);
+		goto err_reg;
+	}
+
+	ret = regulator_enable(ks->vdd_reg);
+	if (ret) {
+		dev_err(&spi->dev, "regulator vdd enable fail: %d\n",
+			ret);
+		goto err_reg;
+	}
+
+	if (gpio_is_valid(gpio)) {
+		usleep_range(10000, 11000);
+		gpio_set_value(gpio, 1);
+	}
 
 	mutex_init(&ks->lock);
 	spin_lock_init(&ks->statelock);
@@ -1471,7 +1511,7 @@ static int ks8851_probe(struct spi_device *spi)
 
 	skb_queue_head_init(&ks->txq);
 
-	SET_ETHTOOL_OPS(ndev, &ks8851_ethtool_ops);
+	ndev->ethtool_ops = &ks8851_ethtool_ops;
 	SET_NETDEV_DEV(ndev, &spi->dev);
 
 	spi_set_drvdata(spi, ks);
@@ -1527,13 +1567,14 @@ err_netdev:
 	free_irq(ndev->irq, ks);
 
 err_irq:
+	if (gpio_is_valid(gpio))
+		gpio_set_value(gpio, 0);
 err_id:
-	if (!IS_ERR(ks->vdd_reg))
-		regulator_disable(ks->vdd_reg);
-err_reg_en:
-	if (!IS_ERR(ks->vdd_reg))
-		regulator_put(ks->vdd_reg);
+	regulator_disable(ks->vdd_reg);
 err_reg:
+	regulator_disable(ks->vdd_io);
+err_reg_io:
+err_gpio:
 	free_netdev(ndev);
 	return ret;
 }
@@ -1547,18 +1588,24 @@ static int ks8851_remove(struct spi_device *spi)
 
 	unregister_netdev(priv->netdev);
 	free_irq(spi->irq, priv);
-	if (!IS_ERR(priv->vdd_reg)) {
-		regulator_disable(priv->vdd_reg);
-		regulator_put(priv->vdd_reg);
-	}
+	if (gpio_is_valid(priv->gpio))
+		gpio_set_value(priv->gpio, 0);
+	regulator_disable(priv->vdd_reg);
+	regulator_disable(priv->vdd_io);
 	free_netdev(priv->netdev);
 
 	return 0;
 }
 
+static const struct of_device_id ks8851_match_table[] = {
+	{ .compatible = "micrel,ks8851" },
+	{ }
+};
+
 static struct spi_driver ks8851_driver = {
 	.driver = {
 		.name = "ks8851",
+		.of_match_table = ks8851_match_table,
 		.owner = THIS_MODULE,
 		.pm = &ks8851_pm_ops,
 	},
diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c
index 14ac0e2bc09f..064a48d0c368 100644
--- a/drivers/net/ethernet/micrel/ksz884x.c
+++ b/drivers/net/ethernet/micrel/ksz884x.c
@@ -4930,7 +4930,7 @@ static void netdev_tx_timeout(struct net_device *dev)
 		 * Only reset the hardware if time between calls is long
 		 * enough.
 		 */
-		if (jiffies - last_reset <= dev->watchdog_timeo)
+		if (time_before_eq(jiffies, last_reset + dev->watchdog_timeo))
 			hw_priv = NULL;
 	}
 
@@ -7072,6 +7072,7 @@ static int pcidev_init(struct pci_dev *pdev, const struct pci_device_id *id)
 		dev = alloc_etherdev(sizeof(struct dev_priv));
 		if (!dev)
 			goto pcidev_init_reg_err;
+		SET_NETDEV_DEV(dev, &pdev->dev);
 		info->netdev[i] = dev;
 
 		priv = netdev_priv(dev);
@@ -7106,7 +7107,7 @@ static int pcidev_init(struct pci_dev *pdev, const struct pci_device_id *id)
 		}
 
 		dev->netdev_ops = &netdev_ops;
-		SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
+		dev->ethtool_ops = &netdev_ethtool_ops;
 		if (register_netdev(dev))
 			goto pcidev_init_reg_err;
 		port_set_power_saving(port, true);
diff --git a/drivers/net/ethernet/microchip/enc28j60.c b/drivers/net/ethernet/microchip/enc28j60.c
index c7b40aa21f22..b1b5f66b8b69 100644
--- a/drivers/net/ethernet/microchip/enc28j60.c
+++ b/drivers/net/ethernet/microchip/enc28j60.c
@@ -1593,7 +1593,7 @@ static int enc28j60_probe(struct spi_device *spi)
 	dev->irq = spi->irq;
 	dev->netdev_ops = &enc28j60_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
-	SET_ETHTOOL_OPS(dev, &enc28j60_ethtool_ops);
+	dev->ethtool_ops = &enc28j60_ethtool_ops;
 
 	enc28j60_lowpower(priv, true);
 
diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index 130f6b204efa..f3d5d79f1cd1 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -4112,7 +4112,7 @@ static int myri10ge_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	setup_timer(&mgp->watchdog_timer, myri10ge_watchdog_timer,
 		    (unsigned long)mgp);
 
-	SET_ETHTOOL_OPS(netdev, &myri10ge_ethtool_ops);
+	netdev->ethtool_ops = &myri10ge_ethtool_ops;
 	INIT_WORK(&mgp->watchdog_work, myri10ge_watchdog);
 	status = register_netdev(netdev);
 	if (status != 0) {
diff --git a/drivers/net/ethernet/natsemi/natsemi.c b/drivers/net/ethernet/natsemi/natsemi.c
index 64ec2a437f46..291fba8b9f07 100644
--- a/drivers/net/ethernet/natsemi/natsemi.c
+++ b/drivers/net/ethernet/natsemi/natsemi.c
@@ -927,7 +927,7 @@ static int natsemi_probe1(struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->netdev_ops = &natsemi_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
-	SET_ETHTOOL_OPS(dev, &ethtool_ops);
+	dev->ethtool_ops = &ethtool_ops;
 
 	if (mtu)
 		dev->mtu = mtu;
diff --git a/drivers/net/ethernet/natsemi/ns83820.c b/drivers/net/ethernet/natsemi/ns83820.c
index dbccf1de49ec..19bb8244b9e3 100644
--- a/drivers/net/ethernet/natsemi/ns83820.c
+++ b/drivers/net/ethernet/natsemi/ns83820.c
@@ -2030,7 +2030,7 @@ static int ns83820_init_one(struct pci_dev *pci_dev,
 		pci_dev->subsystem_vendor, pci_dev->subsystem_device);
 
 	ndev->netdev_ops = &netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &ops);
+	ndev->ethtool_ops = &ops;
 	ndev->watchdog_timeo = 5 * HZ;
 	pci_set_drvdata(pci_dev, ndev);
 
diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c
index a2844ff322c4..be587647c706 100644
--- a/drivers/net/ethernet/neterion/s2io.c
+++ b/drivers/net/ethernet/neterion/s2io.c
@@ -534,15 +534,6 @@ static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 	netif_tx_start_all_queues(sp->dev);
 }
 
-static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
-{
-	if (!sp->config.multiq)
-		sp->mac_control.fifos[fifo_no].queue_state =
-			FIFO_QUEUE_START;
-
-	netif_tx_start_all_queues(sp->dev);
-}
-
 static inline void s2io_wake_all_tx_queue(struct s2io_nic *sp)
 {
 	if (!sp->config.multiq) {
@@ -5369,8 +5360,8 @@ static int s2io_ethtool_gset(struct net_device *dev, struct ethtool_cmd *info)
 		ethtool_cmd_speed_set(info, SPEED_10000);
 		info->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(info, -1);
-		info->duplex = -1;
+		ethtool_cmd_speed_set(info, SPEED_UNKNOWN);
+		info->duplex = DUPLEX_UNKNOWN;
 	}
 
 	info->autoneg = AUTONEG_DISABLE;
@@ -7919,7 +7910,7 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 
 	/*  Driver entry points */
 	dev->netdev_ops = &s2io_netdev_ops;
-	SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
+	dev->ethtool_ops = &netdev_ethtool_ops;
 	dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM |
 		NETIF_F_TSO | NETIF_F_TSO6 |
 		NETIF_F_RXCSUM | NETIF_F_LRO;
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-config.c b/drivers/net/ethernet/neterion/vxge/vxge-config.c
index 089b713b9f7b..2bbd01fcb9b0 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-config.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-config.c
@@ -120,7 +120,6 @@ __vxge_hw_device_register_poll(void __iomem *reg, u64 mask, u32 max_millis)
 {
 	u64 val64;
 	u32 i = 0;
-	enum vxge_hw_status ret = VXGE_HW_FAIL;
 
 	udelay(10);
 
@@ -139,7 +138,7 @@ __vxge_hw_device_register_poll(void __iomem *reg, u64 mask, u32 max_millis)
 		mdelay(1);
 	} while (++i <= max_millis);
 
-	return ret;
+	return VXGE_HW_FAIL;
 }
 
 static inline enum vxge_hw_status
@@ -1682,12 +1681,10 @@ enum vxge_hw_status vxge_hw_driver_stats_get(
 			struct __vxge_hw_device *hldev,
 			struct vxge_hw_device_stats_sw_info *sw_stats)
 {
-	enum vxge_hw_status status = VXGE_HW_OK;
-
 	memcpy(sw_stats, &hldev->stats.sw_dev_info_stats,
 		sizeof(struct vxge_hw_device_stats_sw_info));
 
-	return status;
+	return VXGE_HW_OK;
 }
 
 /*
@@ -3228,7 +3225,6 @@ enum vxge_hw_status
 vxge_hw_vpath_strip_fcs_check(struct __vxge_hw_device *hldev, u64 vpath_mask)
 {
 	struct vxge_hw_vpmgmt_reg       __iomem *vpmgmt_reg;
-	enum vxge_hw_status status = VXGE_HW_OK;
 	int i = 0, j = 0;
 
 	for (i = 0; i < VXGE_HW_MAX_VIRTUAL_PATHS; i++) {
@@ -3241,7 +3237,7 @@ vxge_hw_vpath_strip_fcs_check(struct __vxge_hw_device *hldev, u64 vpath_mask)
 				return VXGE_HW_FAIL;
 		}
 	}
-	return status;
+	return VXGE_HW_OK;
 }
 /*
  * vxge_hw_mgmt_reg_Write - Write Titan register.
@@ -3979,7 +3975,6 @@ __vxge_hw_vpath_mgmt_read(
 {
 	u32 i, mtu = 0, max_pyld = 0;
 	u64 val64;
-	enum vxge_hw_status status = VXGE_HW_OK;
 
 	for (i = 0; i < VXGE_HW_MAC_MAX_MAC_PORT_ID; i++) {
 
@@ -4009,7 +4004,7 @@ __vxge_hw_vpath_mgmt_read(
 	else
 		VXGE_HW_DEVICE_LINK_STATE_SET(vpath->hldev, VXGE_HW_LINK_DOWN);
 
-	return status;
+	return VXGE_HW_OK;
 }
 
 /*
@@ -4039,14 +4034,13 @@ static enum vxge_hw_status
 __vxge_hw_vpath_reset(struct __vxge_hw_device *hldev, u32 vp_id)
 {
 	u64 val64;
-	enum vxge_hw_status status = VXGE_HW_OK;
 
 	val64 = VXGE_HW_CMN_RSTHDLR_CFG0_SW_RESET_VPATH(1 << (16 - vp_id));
 
 	__vxge_hw_pio_mem_write32_upper((u32)vxge_bVALn(val64, 0, 32),
 				&hldev->common_reg->cmn_rsthdlr_cfg0);
 
-	return status;
+	return VXGE_HW_OK;
 }
 
 /*
@@ -4227,7 +4221,6 @@ static enum vxge_hw_status
 __vxge_hw_vpath_mac_configure(struct __vxge_hw_device *hldev, u32 vp_id)
 {
 	u64 val64;
-	enum vxge_hw_status status = VXGE_HW_OK;
 	struct __vxge_hw_virtualpath *vpath;
 	struct vxge_hw_vp_config *vp_config;
 	struct vxge_hw_vpath_reg __iomem *vp_reg;
@@ -4283,7 +4276,7 @@ __vxge_hw_vpath_mac_configure(struct __vxge_hw_device *hldev, u32 vp_id)
 
 		writeq(val64, &vp_reg->rxmac_vcfg1);
 	}
-	return status;
+	return VXGE_HW_OK;
 }
 
 /*
@@ -4295,7 +4288,6 @@ static enum vxge_hw_status
 __vxge_hw_vpath_tim_configure(struct __vxge_hw_device *hldev, u32 vp_id)
 {
 	u64 val64;
-	enum vxge_hw_status status = VXGE_HW_OK;
 	struct __vxge_hw_virtualpath *vpath;
 	struct vxge_hw_vpath_reg __iomem *vp_reg;
 	struct vxge_hw_vp_config *config;
@@ -4545,7 +4537,7 @@ __vxge_hw_vpath_tim_configure(struct __vxge_hw_device *hldev, u32 vp_id)
 	val64 |= VXGE_HW_TIM_WRKLD_CLC_CNT_RX_TX(3);
 	writeq(val64, &vp_reg->tim_wrkld_clc);
 
-	return status;
+	return VXGE_HW_OK;
 }
 
 /*
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c b/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c
index f8f073880f84..b07d552a27d4 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c
@@ -62,8 +62,8 @@ static int vxge_ethtool_gset(struct net_device *dev, struct ethtool_cmd *info)
 		ethtool_cmd_speed_set(info, SPEED_10000);
 		info->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(info, -1);
-		info->duplex = -1;
+		ethtool_cmd_speed_set(info, SPEED_UNKNOWN);
+		info->duplex = DUPLEX_UNKNOWN;
 	}
 
 	info->autoneg = AUTONEG_DISABLE;
@@ -1128,5 +1128,5 @@ static const struct ethtool_ops vxge_ethtool_ops = {
 
 void vxge_initialize_ethtool_ops(struct net_device *ndev)
 {
-	SET_ETHTOOL_OPS(ndev, &vxge_ethtool_ops);
+	ndev->ethtool_ops = &vxge_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c
index d107bcbb8543..7a0deadd53bf 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-main.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c
@@ -2122,7 +2122,7 @@ static int vxge_open_vpaths(struct vxgedev *vdev)
 static void adaptive_coalesce_tx_interrupts(struct vxge_fifo *fifo)
 {
 	fifo->interrupt_count++;
-	if (jiffies > fifo->jiffies + HZ / 100) {
+	if (time_before(fifo->jiffies + HZ / 100, jiffies)) {
 		struct __vxge_hw_fifo *hw_fifo = fifo->handle;
 
 		fifo->jiffies = jiffies;
@@ -2150,7 +2150,7 @@ static void adaptive_coalesce_tx_interrupts(struct vxge_fifo *fifo)
 static void adaptive_coalesce_rx_interrupts(struct vxge_ring *ring)
 {
 	ring->interrupt_count++;
-	if (jiffies > ring->jiffies + HZ / 100) {
+	if (time_before(ring->jiffies + HZ / 100, jiffies)) {
 		struct __vxge_hw_ring *hw_ring = ring->handle;
 
 		ring->jiffies = jiffies;
diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index fddb464aeab3..9afc536c5734 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -406,7 +406,7 @@ union ring_type {
 
 #define NV_RX_DESCRIPTORVALID	(1<<16)
 #define NV_RX_MISSEDFRAME	(1<<17)
-#define NV_RX_SUBSTRACT1	(1<<18)
+#define NV_RX_SUBTRACT1		(1<<18)
 #define NV_RX_ERROR1		(1<<23)
 #define NV_RX_ERROR2		(1<<24)
 #define NV_RX_ERROR3		(1<<25)
@@ -423,7 +423,7 @@ union ring_type {
 #define NV_RX2_CHECKSUM_IP_TCP	(0x14000000)
 #define NV_RX2_CHECKSUM_IP_UDP	(0x18000000)
 #define NV_RX2_DESCRIPTORVALID	(1<<29)
-#define NV_RX2_SUBSTRACT1	(1<<25)
+#define NV_RX2_SUBTRACT1	(1<<25)
 #define NV_RX2_ERROR1		(1<<18)
 #define NV_RX2_ERROR2		(1<<19)
 #define NV_RX2_ERROR3		(1<<20)
@@ -2832,7 +2832,7 @@ static int nv_rx_process(struct net_device *dev, int limit)
 					}
 					/* framing errors are soft errors */
 					else if ((flags & NV_RX_ERROR_MASK) == NV_RX_FRAMINGERR) {
-						if (flags & NV_RX_SUBSTRACT1)
+						if (flags & NV_RX_SUBTRACT1)
 							len--;
 					}
 					/* the rest are hard errors */
@@ -2863,7 +2863,7 @@ static int nv_rx_process(struct net_device *dev, int limit)
 					}
 					/* framing errors are soft errors */
 					else if ((flags & NV_RX2_ERROR_MASK) == NV_RX2_FRAMINGERR) {
-						if (flags & NV_RX2_SUBSTRACT1)
+						if (flags & NV_RX2_SUBTRACT1)
 							len--;
 					}
 					/* the rest are hard errors */
@@ -2937,7 +2937,7 @@ static int nv_rx_process_optimized(struct net_device *dev, int limit)
 				}
 				/* framing errors are soft errors */
 				else if ((flags & NV_RX2_ERROR_MASK) == NV_RX2_FRAMINGERR) {
-					if (flags & NV_RX2_SUBSTRACT1)
+					if (flags & NV_RX2_SUBTRACT1)
 						len--;
 				}
 				/* the rest are hard errors */
@@ -4285,8 +4285,8 @@ static int nv_get_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
 		if (np->duplex)
 			ecmd->duplex = DUPLEX_FULL;
 	} else {
-		speed = -1;
-		ecmd->duplex = -1;
+		speed = SPEED_UNKNOWN;
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 	ethtool_cmd_speed_set(ecmd, speed);
 	ecmd->autoneg = np->autoneg;
@@ -5766,7 +5766,7 @@ static int nv_probe(struct pci_dev *pci_dev, const struct pci_device_id *id)
 		dev->netdev_ops = &nv_netdev_ops_optimized;
 
 	netif_napi_add(dev, &np->napi, nv_napi_poll, RX_WORK_PER_LOOP);
-	SET_ETHTOOL_OPS(dev, &ops);
+	dev->ethtool_ops = &ops;
 	dev->watchdog_timeo = NV_WATCHDOG_TIMEO;
 
 	pci_set_drvdata(pci_dev, dev);
diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c
index 422d9b51ac24..8706c0dbd0c3 100644
--- a/drivers/net/ethernet/nxp/lpc_eth.c
+++ b/drivers/net/ethernet/nxp/lpc_eth.c
@@ -1361,7 +1361,7 @@ static int lpc_eth_drv_probe(struct platform_device *pdev)
 	__lpc_eth_clock_enable(pldat, true);
 
 	/* Map IO space */
-	pldat->net_base = ioremap(res->start, res->end - res->start + 1);
+	pldat->net_base = ioremap(res->start, resource_size(res));
 	if (!pldat->net_base) {
 		dev_err(&pdev->dev, "failed to map registers\n");
 		ret = -ENOMEM;
@@ -1417,10 +1417,8 @@ static int lpc_eth_drv_probe(struct platform_device *pdev)
 	}
 	pldat->dma_buff_base_p = dma_handle;
 
-	netdev_dbg(ndev, "IO address start     :0x%08x\n",
-			res->start);
-	netdev_dbg(ndev, "IO address size      :%d\n",
-			res->end - res->start + 1);
+	netdev_dbg(ndev, "IO address space     :%pR\n", res);
+	netdev_dbg(ndev, "IO address size      :%d\n", resource_size(res));
 	netdev_dbg(ndev, "IO address (mapped)  :0x%p\n",
 			pldat->net_base);
 	netdev_dbg(ndev, "IRQ number           :%d\n", ndev->irq);
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/Kconfig b/drivers/net/ethernet/oki-semi/pch_gbe/Kconfig
index a588ffde9700..44c8be1c6805 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/Kconfig
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/Kconfig
@@ -4,7 +4,7 @@
 
 config PCH_GBE
 	tristate "OKI SEMICONDUCTOR IOH(ML7223/ML7831) GbE"
-	depends on PCI && (X86 || COMPILE_TEST)
+	depends on PCI && (X86_32 || COMPILE_TEST)
 	select MII
 	select PTP_1588_CLOCK_PCH
 	---help---
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c
index 826f0ccdc23c..4fe8ea96bd25 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c
@@ -91,7 +91,7 @@ static int pch_gbe_get_settings(struct net_device *netdev,
 	ecmd->advertising &= ~(ADVERTISED_TP | ADVERTISED_1000baseT_Half);
 
 	if (!netif_carrier_ok(adapter->netdev))
-		ethtool_cmd_speed_set(ecmd, -1);
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
 	return ret;
 }
 
@@ -508,5 +508,5 @@ static const struct ethtool_ops pch_gbe_ethtool_ops = {
 
 void pch_gbe_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &pch_gbe_ethtool_ops);
+	netdev->ethtool_ops = &pch_gbe_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/packetengines/hamachi.c b/drivers/net/ethernet/packetengines/hamachi.c
index b6bdeb3c1971..9a997e4c3e08 100644
--- a/drivers/net/ethernet/packetengines/hamachi.c
+++ b/drivers/net/ethernet/packetengines/hamachi.c
@@ -724,10 +724,8 @@ static int hamachi_init_one(struct pci_dev *pdev,
 
 	/* The Hamachi-specific entries in the device structure. */
 	dev->netdev_ops = &hamachi_netdev_ops;
-	if (chip_tbl[hmp->chip_id].flags & CanHaveMII)
-		SET_ETHTOOL_OPS(dev, &ethtool_ops);
-	else
-		SET_ETHTOOL_OPS(dev, &ethtool_ops_no_mii);
+	dev->ethtool_ops = (chip_tbl[hmp->chip_id].flags & CanHaveMII) ?
+		&ethtool_ops : &ethtool_ops_no_mii;
 	dev->watchdog_timeo = TX_TIMEOUT;
 	if (mtu)
 		dev->mtu = mtu;
diff --git a/drivers/net/ethernet/packetengines/yellowfin.c b/drivers/net/ethernet/packetengines/yellowfin.c
index 9a6cb482dcd0..69a8dc095072 100644
--- a/drivers/net/ethernet/packetengines/yellowfin.c
+++ b/drivers/net/ethernet/packetengines/yellowfin.c
@@ -472,7 +472,7 @@ static int yellowfin_init_one(struct pci_dev *pdev,
 
 	/* The Yellowfin-specific entries in the device structure. */
 	dev->netdev_ops = &netdev_ops;
-	SET_ETHTOOL_OPS(dev, &ethtool_ops);
+	dev->ethtool_ops = &ethtool_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
 	if (mtu)
diff --git a/drivers/net/ethernet/qlogic/Kconfig b/drivers/net/ethernet/qlogic/Kconfig
index c14bd3116e45..d49cba129081 100644
--- a/drivers/net/ethernet/qlogic/Kconfig
+++ b/drivers/net/ethernet/qlogic/Kconfig
@@ -66,6 +66,17 @@ config QLCNIC_VXLAN
 	  Say Y here if you want to enable hardware offload support for
 	  Virtual eXtensible Local Area Network (VXLAN) in the driver.
 
+config QLCNIC_HWMON
+	bool "QLOGIC QLCNIC 82XX and 83XX family HWMON support"
+	depends on QLCNIC && HWMON && !(QLCNIC=y && HWMON=m)
+	default y
+	---help---
+	  This configuration parameter can be used to read the
+	  board temperature in Converged Ethernet devices
+	  supported by qlcnic.
+
+	  This data is available via the hwmon sysfs interface.
+
 config QLGE
 	tristate "QLogic QLGE 10Gb Ethernet Driver Support"
 	depends on PCI
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index f09c35d669b3..5bf05818a12c 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -1373,7 +1373,7 @@ netxen_setup_netdev(struct netxen_adapter *adapter,
 
 	netxen_nic_change_mtu(netdev, netdev->mtu);
 
-	SET_ETHTOOL_OPS(netdev, &netxen_nic_ethtool_ops);
+	netdev->ethtool_ops = &netxen_nic_ethtool_ops;
 
 	netdev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
 	                      NETIF_F_RXCSUM;
diff --git a/drivers/net/ethernet/qlogic/qla3xxx.c b/drivers/net/ethernet/qlogic/qla3xxx.c
index 2eabd44f8914..b5d6bc1a8b00 100644
--- a/drivers/net/ethernet/qlogic/qla3xxx.c
+++ b/drivers/net/ethernet/qlogic/qla3xxx.c
@@ -3838,7 +3838,7 @@ static int ql3xxx_probe(struct pci_dev *pdev,
 
 	/* Set driver entry points */
 	ndev->netdev_ops = &ql3xxx_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &ql3xxx_ethtool_ops);
+	ndev->ethtool_ops = &ql3xxx_ethtool_ops;
 	ndev->watchdog_timeo = 5 * HZ;
 
 	netif_napi_add(ndev, &qdev->napi, ql_poll, 64);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
index f785d01c7d12..be618b9e874f 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
@@ -39,8 +39,8 @@
 
 #define _QLCNIC_LINUX_MAJOR 5
 #define _QLCNIC_LINUX_MINOR 3
-#define _QLCNIC_LINUX_SUBVERSION 57
-#define QLCNIC_LINUX_VERSIONID  "5.3.57"
+#define _QLCNIC_LINUX_SUBVERSION 60
+#define QLCNIC_LINUX_VERSIONID  "5.3.60"
 #define QLCNIC_DRV_IDC_VER  0x01
 #define QLCNIC_DRIVER_VERSION  ((_QLCNIC_LINUX_MAJOR << 16) |\
 		 (_QLCNIC_LINUX_MINOR << 8) | (_QLCNIC_LINUX_SUBVERSION))
@@ -441,6 +441,8 @@ struct qlcnic_82xx_dump_template_hdr {
 	u32	rsvd1[0];
 };
 
+#define QLC_PEX_DMA_READ_SIZE	(PAGE_SIZE * 16)
+
 struct qlcnic_fw_dump {
 	u8	clr;	/* flag to indicate if dump is cleared */
 	bool	enable; /* enable/disable dump */
@@ -537,6 +539,7 @@ struct qlcnic_hardware_context {
 	u8 phys_port_id[ETH_ALEN];
 	u8 lb_mode;
 	u16 vxlan_port;
+	struct device *hwmon_dev;
 };
 
 struct qlcnic_adapter_stats {
@@ -1018,6 +1021,8 @@ struct qlcnic_ipaddr {
 #define QLCNIC_DEL_VXLAN_PORT		0x200000
 #endif
 
+#define QLCNIC_VLAN_FILTERING		0x800000
+
 #define QLCNIC_IS_MSI_FAMILY(adapter) \
 	((adapter)->flags & (QLCNIC_MSI_ENABLED | QLCNIC_MSIX_ENABLED))
 #define QLCNIC_IS_TSO_CAPABLE(adapter)  \
@@ -1316,6 +1321,7 @@ struct qlcnic_eswitch {
 #define QL_STATUS_INVALID_PARAM	-1
 
 #define MAX_BW			100	/* % of link speed */
+#define MIN_BW			1	/* % of link speed */
 #define MAX_VLAN_ID		4095
 #define MIN_VLAN_ID		2
 #define DEFAULT_MAC_LEARN	1
@@ -1692,7 +1698,7 @@ int qlcnic_read_mac_addr(struct qlcnic_adapter *);
 int qlcnic_setup_netdev(struct qlcnic_adapter *, struct net_device *, int);
 void qlcnic_set_netdev_features(struct qlcnic_adapter *,
 				struct qlcnic_esw_func_cfg *);
-void qlcnic_sriov_vf_schedule_multi(struct net_device *);
+void qlcnic_sriov_vf_set_multi(struct net_device *);
 int qlcnic_is_valid_nic_func(struct qlcnic_adapter *, u8);
 int qlcnic_get_pci_func_type(struct qlcnic_adapter *, u16, u16 *, u16 *,
 			     u16 *);
@@ -2338,6 +2344,16 @@ static inline bool qlcnic_83xx_vf_check(struct qlcnic_adapter *adapter)
 	return (device == PCI_DEVICE_ID_QLOGIC_VF_QLE834X) ? true : false;
 }
 
+static inline bool qlcnic_sriov_check(struct qlcnic_adapter *adapter)
+{
+	bool status;
+
+	status = (qlcnic_sriov_pf_check(adapter) ||
+		  qlcnic_sriov_vf_check(adapter)) ? true : false;
+
+	return status;
+}
+
 static inline u32 qlcnic_get_vnic_func_count(struct qlcnic_adapter *adapter)
 {
 	if (qlcnic_84xx_check(adapter))
@@ -2345,4 +2361,18 @@ static inline u32 qlcnic_get_vnic_func_count(struct qlcnic_adapter *adapter)
 	else
 		return QLC_DEFAULT_VNIC_COUNT;
 }
+
+#ifdef CONFIG_QLCNIC_HWMON
+void qlcnic_register_hwmon_dev(struct qlcnic_adapter *);
+void qlcnic_unregister_hwmon_dev(struct qlcnic_adapter *);
+#else
+static inline void qlcnic_register_hwmon_dev(struct qlcnic_adapter *adapter)
+{
+	return;
+}
+static inline void qlcnic_unregister_hwmon_dev(struct qlcnic_adapter *adapter)
+{
+	return;
+}
+#endif
 #endif				/* __QLCNIC_H_ */
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
index b7cffb46a75d..a4a4ec0b68f8 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
@@ -33,6 +33,7 @@ static void qlcnic_83xx_get_beacon_state(struct qlcnic_adapter *);
 #define RSS_HASHTYPE_IP_TCP		0x3
 #define QLC_83XX_FW_MBX_CMD		0
 #define QLC_SKIP_INACTIVE_PCI_REGS	7
+#define QLC_MAX_LEGACY_FUNC_SUPP	8
 
 static const struct qlcnic_mailbox_metadata qlcnic_83xx_mbx_tbl[] = {
 	{QLCNIC_CMD_CONFIGURE_IP_ADDR, 6, 1},
@@ -357,8 +358,15 @@ int qlcnic_83xx_setup_intr(struct qlcnic_adapter *adapter)
 	if (!ahw->intr_tbl)
 		return -ENOMEM;
 
-	if (!(adapter->flags & QLCNIC_MSIX_ENABLED))
+	if (!(adapter->flags & QLCNIC_MSIX_ENABLED)) {
+		if (adapter->ahw->pci_func >= QLC_MAX_LEGACY_FUNC_SUPP) {
+			dev_err(&adapter->pdev->dev, "PCI function number 8 and higher are not supported with legacy interrupt, func 0x%x\n",
+				ahw->pci_func);
+			return -EOPNOTSUPP;
+		}
+
 		qlcnic_83xx_enable_legacy(adapter);
+	}
 
 	for (i = 0; i < num_msix; i++) {
 		if (adapter->flags & QLCNIC_MSIX_ENABLED)
@@ -879,6 +887,9 @@ int qlcnic_83xx_alloc_mbx_args(struct qlcnic_cmd_args *mbx,
 			return 0;
 		}
 	}
+
+	dev_err(&adapter->pdev->dev, "%s: Invalid mailbox command opcode 0x%x\n",
+		__func__, type);
 	return -EINVAL;
 }
 
@@ -3026,19 +3037,18 @@ void qlcnic_83xx_unlock_driver(struct qlcnic_adapter *adapter)
 	QLCRDX(adapter->ahw, QLC_83XX_DRV_UNLOCK);
 }
 
-int qlcnic_83xx_ms_mem_write128(struct qlcnic_adapter *adapter, u64 addr,
+int qlcnic_ms_mem_write128(struct qlcnic_adapter *adapter, u64 addr,
 				u32 *data, u32 count)
 {
 	int i, j, ret = 0;
 	u32 temp;
-	int err = 0;
 
 	/* Check alignment */
 	if (addr & 0xF)
 		return -EIO;
 
 	mutex_lock(&adapter->ahw->mem_lock);
-	qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_ADDR_HI, 0);
+	qlcnic_ind_wr(adapter, QLCNIC_MS_ADDR_HI, 0);
 
 	for (i = 0; i < count; i++, addr += 16) {
 		if (!((ADDR_IN_RANGE(addr, QLCNIC_ADDR_QDR_NET,
@@ -3049,26 +3059,16 @@ int qlcnic_83xx_ms_mem_write128(struct qlcnic_adapter *adapter, u64 addr,
 			return -EIO;
 		}
 
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_ADDR_LO, addr);
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_WRTDATA_LO,
-					     *data++);
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_WRTDATA_HI,
-					     *data++);
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_WRTDATA_ULO,
-					     *data++);
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_WRTDATA_UHI,
-					     *data++);
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_CTRL,
-					     QLCNIC_TA_WRITE_ENABLE);
-		qlcnic_83xx_wrt_reg_indirect(adapter, QLCNIC_MS_CTRL,
-					     QLCNIC_TA_WRITE_START);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_ADDR_LO, addr);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_WRTDATA_LO, *data++);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_WRTDATA_HI, *data++);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_WRTDATA_ULO, *data++);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_WRTDATA_UHI, *data++);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_CTRL, QLCNIC_TA_WRITE_ENABLE);
+		qlcnic_ind_wr(adapter, QLCNIC_MS_CTRL, QLCNIC_TA_WRITE_START);
 
 		for (j = 0; j < MAX_CTL_CHECK; j++) {
-			temp = QLCRD32(adapter, QLCNIC_MS_CTRL, &err);
-			if (err == -EIO) {
-				mutex_unlock(&adapter->ahw->mem_lock);
-				return err;
-			}
+			temp = qlcnic_ind_rd(adapter, QLCNIC_MS_CTRL);
 
 			if ((temp & TA_CTL_BUSY) == 0)
 				break;
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h
index 88d809c35633..2bf101a47d02 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h
@@ -418,7 +418,6 @@ enum qlcnic_83xx_states {
 #define QLC_83XX_GET_FUNC_MODE_FROM_NPAR_INFO(val)	(val & 0x80000000)
 #define QLC_83XX_GET_LRO_CAPABILITY(val)		(val & 0x20)
 #define QLC_83XX_GET_LSO_CAPABILITY(val)		(val & 0x40)
-#define QLC_83XX_GET_LSO_CAPABILITY(val)		(val & 0x40)
 #define QLC_83XX_GET_HW_LRO_CAPABILITY(val)		(val & 0x400)
 #define QLC_83XX_GET_VLAN_ALIGN_CAPABILITY(val)	(val & 0x4000)
 #define QLC_83XX_GET_FW_LRO_MSS_CAPABILITY(val)	(val & 0x20000)
@@ -560,7 +559,7 @@ void qlcnic_83xx_napi_del(struct qlcnic_adapter *);
 void qlcnic_83xx_napi_enable(struct qlcnic_adapter *);
 void qlcnic_83xx_napi_disable(struct qlcnic_adapter *);
 int qlcnic_83xx_config_led(struct qlcnic_adapter *, u32, u32);
-void qlcnic_ind_wr(struct qlcnic_adapter *, u32, u32);
+int qlcnic_ind_wr(struct qlcnic_adapter *, u32, u32);
 int qlcnic_ind_rd(struct qlcnic_adapter *, u32);
 int qlcnic_83xx_create_rx_ctx(struct qlcnic_adapter *);
 int qlcnic_83xx_create_tx_ctx(struct qlcnic_adapter *,
@@ -617,7 +616,6 @@ void qlcnic_83xx_idc_request_reset(struct qlcnic_adapter *, u32);
 int qlcnic_83xx_lock_driver(struct qlcnic_adapter *);
 void qlcnic_83xx_unlock_driver(struct qlcnic_adapter *);
 int qlcnic_83xx_set_default_offload_settings(struct qlcnic_adapter *);
-int qlcnic_83xx_ms_mem_write128(struct qlcnic_adapter *, u64, u32 *, u32);
 int qlcnic_83xx_idc_vnic_pf_entry(struct qlcnic_adapter *);
 int qlcnic_83xx_disable_vnic_mode(struct qlcnic_adapter *, int);
 int qlcnic_83xx_config_vnic_opmode(struct qlcnic_adapter *);
@@ -659,4 +657,5 @@ void qlcnic_83xx_cache_tmpl_hdr_values(struct qlcnic_fw_dump *);
 u32 qlcnic_83xx_get_cap_size(void *, int);
 void qlcnic_83xx_set_sys_info(void *, int, u32);
 void qlcnic_83xx_store_cap_mask(void *, u32);
+int qlcnic_ms_mem_write128(struct qlcnic_adapter *, u64, u32 *, u32);
 #endif
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c
index ba20c721ee97..f33559b72528 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c
@@ -1363,8 +1363,8 @@ static int qlcnic_83xx_copy_bootloader(struct qlcnic_adapter *adapter)
 		return ret;
 	}
 	/* 16 byte write to MS memory */
-	ret = qlcnic_83xx_ms_mem_write128(adapter, dest, (u32 *)p_cache,
-					  size / 16);
+	ret = qlcnic_ms_mem_write128(adapter, dest, (u32 *)p_cache,
+				     size / 16);
 	if (ret) {
 		vfree(p_cache);
 		return ret;
@@ -1389,8 +1389,8 @@ static int qlcnic_83xx_copy_fw_file(struct qlcnic_adapter *adapter)
 	p_cache = (u32 *)fw->data;
 	addr = (u64)dest;
 
-	ret = qlcnic_83xx_ms_mem_write128(adapter, addr,
-					  p_cache, size / 16);
+	ret = qlcnic_ms_mem_write128(adapter, addr,
+				     p_cache, size / 16);
 	if (ret) {
 		dev_err(&adapter->pdev->dev, "MS memory write failed\n");
 		release_firmware(fw);
@@ -1405,8 +1405,8 @@ static int qlcnic_83xx_copy_fw_file(struct qlcnic_adapter *adapter)
 			data[i] = fw->data[size + i];
 		for (; i < 16; i++)
 			data[i] = 0;
-		ret = qlcnic_83xx_ms_mem_write128(adapter, addr,
-						  (u32 *)data, 1);
+		ret = qlcnic_ms_mem_write128(adapter, addr,
+					     (u32 *)data, 1);
 		if (ret) {
 			dev_err(&adapter->pdev->dev,
 				"MS memory write failed\n");
@@ -2181,6 +2181,8 @@ int qlcnic_83xx_configure_opmode(struct qlcnic_adapter *adapter)
 		max_sds_rings = QLCNIC_MAX_SDS_RINGS;
 		max_tx_rings = QLCNIC_MAX_TX_RINGS;
 	} else {
+		dev_err(&adapter->pdev->dev, "%s: Invalid opmode %d\n",
+			__func__, ret);
 		return -EIO;
 	}
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c
index c1e11f5715b0..304e247bdf33 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c
@@ -1027,8 +1027,11 @@ int qlcnic_config_port_mirroring(struct qlcnic_adapter *adapter, u8 id,
 	u32 arg1;
 
 	if (adapter->ahw->op_mode != QLCNIC_MGMT_FUNC ||
-	    !(adapter->eswitch[id].flags & QLCNIC_SWITCH_ENABLE))
+	    !(adapter->eswitch[id].flags & QLCNIC_SWITCH_ENABLE)) {
+		dev_err(&adapter->pdev->dev, "%s: Not a management function\n",
+			__func__);
 		return err;
+	}
 
 	arg1 = id | (enable_mirroring ? BIT_4 : 0);
 	arg1 |= pci_func << 8;
@@ -1318,8 +1321,12 @@ int qlcnic_config_switch_port(struct qlcnic_adapter *adapter,
 	u32 arg1, arg2 = 0;
 	u8 pci_func;
 
-	if (adapter->ahw->op_mode != QLCNIC_MGMT_FUNC)
+	if (adapter->ahw->op_mode != QLCNIC_MGMT_FUNC) {
+		dev_err(&adapter->pdev->dev, "%s: Not a management function\n",
+			__func__);
 		return err;
+	}
+
 	pci_func = esw_cfg->pci_func;
 	index = qlcnic_is_valid_nic_func(adapter, pci_func);
 	if (index < 0)
@@ -1363,6 +1370,8 @@ int qlcnic_config_switch_port(struct qlcnic_adapter *adapter,
 			arg1 &= ~(0x0ffff << 16);
 			break;
 	default:
+		dev_err(&adapter->pdev->dev, "%s: Invalid opmode 0x%x\n",
+			__func__, esw_cfg->op_mode);
 		return err;
 	}
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
index 5bacf5210aed..1b7f3dbae289 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
@@ -726,6 +726,11 @@ static int qlcnic_set_channels(struct net_device *dev,
 	struct qlcnic_adapter *adapter = netdev_priv(dev);
 	int err;
 
+	if (!(adapter->flags & QLCNIC_MSIX_ENABLED)) {
+		netdev_err(dev, "No RSS/TSS support in non MSI-X mode\n");
+		return -EINVAL;
+	}
+
 	if (channel->other_count || channel->combined_count)
 		return -EINVAL;
 
@@ -734,7 +739,7 @@ static int qlcnic_set_channels(struct net_device *dev,
 	if (err)
 		return err;
 
-	if (channel->rx_count) {
+	if (adapter->drv_sds_rings != channel->rx_count) {
 		err = qlcnic_validate_rings(adapter, channel->rx_count,
 					    QLCNIC_RX_QUEUE);
 		if (err) {
@@ -745,7 +750,7 @@ static int qlcnic_set_channels(struct net_device *dev,
 		adapter->drv_rss_rings = channel->rx_count;
 	}
 
-	if (channel->tx_count) {
+	if (adapter->drv_tx_rings != channel->tx_count) {
 		err = qlcnic_validate_rings(adapter, channel->tx_count,
 					    QLCNIC_TX_QUEUE);
 		if (err) {
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
index 9f3adf4e70b5..851cb4a80d50 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
@@ -373,12 +373,16 @@ int qlcnic_ind_rd(struct qlcnic_adapter *adapter, u32 addr)
 	return data;
 }
 
-void qlcnic_ind_wr(struct qlcnic_adapter *adapter, u32 addr, u32 data)
+int qlcnic_ind_wr(struct qlcnic_adapter *adapter, u32 addr, u32 data)
 {
+	int ret = 0;
+
 	if (qlcnic_82xx_check(adapter))
 		qlcnic_write_window_reg(addr, adapter->ahw->pci_base0, data);
 	else
-		qlcnic_83xx_wrt_reg_indirect(adapter, addr, data);
+		ret = qlcnic_83xx_wrt_reg_indirect(adapter, addr, data);
+
+	return ret;
 }
 
 static int
@@ -567,28 +571,14 @@ static void __qlcnic_set_multi(struct net_device *netdev, u16 vlan)
 void qlcnic_set_multi(struct net_device *netdev)
 {
 	struct qlcnic_adapter *adapter = netdev_priv(netdev);
-	struct qlcnic_mac_vlan_list *cur;
-	struct netdev_hw_addr *ha;
-	size_t temp;
 
 	if (!test_bit(__QLCNIC_FW_ATTACHED, &adapter->state))
 		return;
-	if (qlcnic_sriov_vf_check(adapter)) {
-		if (!netdev_mc_empty(netdev)) {
-			netdev_for_each_mc_addr(ha, netdev) {
-				temp = sizeof(struct qlcnic_mac_vlan_list);
-				cur = kzalloc(temp, GFP_ATOMIC);
-				if (cur == NULL)
-					break;
-				memcpy(cur->mac_addr,
-				       ha->addr, ETH_ALEN);
-				list_add_tail(&cur->list, &adapter->vf_mc_list);
-			}
-		}
-		qlcnic_sriov_vf_schedule_multi(adapter->netdev);
-		return;
-	}
-	__qlcnic_set_multi(netdev, 0);
+
+	if (qlcnic_sriov_vf_check(adapter))
+		qlcnic_sriov_vf_set_multi(netdev);
+	else
+		__qlcnic_set_multi(netdev, 0);
 }
 
 int qlcnic_82xx_nic_set_promisc(struct qlcnic_adapter *adapter, u32 mode)
@@ -630,7 +620,7 @@ void qlcnic_prune_lb_filters(struct qlcnic_adapter *adapter)
 	struct hlist_node *n;
 	struct hlist_head *head;
 	int i;
-	unsigned long time;
+	unsigned long expires;
 	u8 cmd;
 
 	for (i = 0; i < adapter->fhash.fbucket_size; i++) {
@@ -638,8 +628,8 @@ void qlcnic_prune_lb_filters(struct qlcnic_adapter *adapter)
 		hlist_for_each_entry_safe(tmp_fil, n, head, fnode) {
 			cmd =  tmp_fil->vlan_id ? QLCNIC_MAC_VLAN_DEL :
 						  QLCNIC_MAC_DEL;
-			time = tmp_fil->ftime;
-			if (jiffies > (QLCNIC_FILTER_AGE * HZ + time)) {
+			expires = tmp_fil->ftime + QLCNIC_FILTER_AGE * HZ;
+			if (time_before(expires, jiffies)) {
 				qlcnic_sre_macaddr_change(adapter,
 							  tmp_fil->faddr,
 							  tmp_fil->vlan_id,
@@ -657,8 +647,8 @@ void qlcnic_prune_lb_filters(struct qlcnic_adapter *adapter)
 
 		hlist_for_each_entry_safe(tmp_fil, n, head, fnode)
 		{
-			time = tmp_fil->ftime;
-			if (jiffies > (QLCNIC_FILTER_AGE * HZ + time)) {
+			expires = tmp_fil->ftime + QLCNIC_FILTER_AGE * HZ;
+			if (time_before(expires, jiffies)) {
 				spin_lock_bh(&adapter->rx_mac_learn_lock);
 				adapter->rx_fhash.fnum--;
 				hlist_del(&tmp_fil->fnode);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
index 173b3d12991f..e45bf09af0c9 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
@@ -305,7 +305,6 @@ static void qlcnic_send_filter(struct qlcnic_adapter *adapter,
 {
 	struct vlan_ethhdr *vh = (struct vlan_ethhdr *)(skb->data);
 	struct ethhdr *phdr = (struct ethhdr *)(skb->data);
-	struct net_device *netdev = adapter->netdev;
 	u16 protocol = ntohs(skb->protocol);
 	struct qlcnic_filter *fil, *tmp_fil;
 	struct hlist_head *head;
@@ -314,27 +313,16 @@ static void qlcnic_send_filter(struct qlcnic_adapter *adapter,
 	u16 vlan_id = 0;
 	u8 hindex, hval;
 
-	if (!qlcnic_sriov_pf_check(adapter)) {
-		if (ether_addr_equal(phdr->h_source, adapter->mac_addr))
-			return;
-	} else {
+	if (ether_addr_equal(phdr->h_source, adapter->mac_addr))
+		return;
+
+	if (adapter->flags & QLCNIC_VLAN_FILTERING) {
 		if (protocol == ETH_P_8021Q) {
 			vh = (struct vlan_ethhdr *)skb->data;
 			vlan_id = ntohs(vh->h_vlan_TCI);
 		} else if (vlan_tx_tag_present(skb)) {
 			vlan_id = vlan_tx_tag_get(skb);
 		}
-
-		if (ether_addr_equal(phdr->h_source, adapter->mac_addr) &&
-		    !vlan_id)
-			return;
-	}
-
-	if (adapter->fhash.fnum >= adapter->fhash.fmax) {
-		adapter->stats.mac_filter_limit_overrun++;
-		netdev_info(netdev, "Can not add more than %d mac-vlan filters, configured %d\n",
-			    adapter->fhash.fmax, adapter->fhash.fnum);
-		return;
 	}
 
 	memcpy(&src_addr, phdr->h_source, ETH_ALEN);
@@ -353,6 +341,11 @@ static void qlcnic_send_filter(struct qlcnic_adapter *adapter,
 		}
 	}
 
+	if (unlikely(adapter->fhash.fnum >= adapter->fhash.fmax)) {
+		adapter->stats.mac_filter_limit_overrun++;
+		return;
+	}
+
 	fil = kzalloc(sizeof(struct qlcnic_filter), GFP_ATOMIC);
 	if (!fil)
 		return;
@@ -1216,8 +1209,7 @@ qlcnic_process_rcv(struct qlcnic_adapter *adapter,
 	if (!skb)
 		return buffer;
 
-	if (adapter->drv_mac_learn &&
-	    (adapter->flags & QLCNIC_ESWITCH_ENABLED)) {
+	if (adapter->rx_mac_learn) {
 		t_vid = 0;
 		is_lb_pkt = qlcnic_82xx_is_lb_pkt(sts_data0);
 		qlcnic_add_lb_filter(adapter, skb, is_lb_pkt, t_vid);
@@ -1293,8 +1285,7 @@ qlcnic_process_lro(struct qlcnic_adapter *adapter,
 	if (!skb)
 		return buffer;
 
-	if (adapter->drv_mac_learn &&
-	    (adapter->flags & QLCNIC_ESWITCH_ENABLED)) {
+	if (adapter->rx_mac_learn) {
 		t_vid = 0;
 		is_lb_pkt = qlcnic_82xx_is_lb_pkt(sts_data0);
 		qlcnic_add_lb_filter(adapter, skb, is_lb_pkt, t_vid);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 7e55e88a81bf..4fc186713b66 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -378,7 +378,8 @@ static int qlcnic_fdb_del(struct ndmsg *ndm, struct nlattr *tb[],
 	if (!adapter->fdb_mac_learn)
 		return ndo_dflt_fdb_del(ndm, tb, netdev, addr);
 
-	if (adapter->flags & QLCNIC_ESWITCH_ENABLED) {
+	if ((adapter->flags & QLCNIC_ESWITCH_ENABLED) ||
+	    qlcnic_sriov_check(adapter)) {
 		if (is_unicast_ether_addr(addr)) {
 			err = dev_uc_del(netdev, addr);
 			if (!err)
@@ -402,7 +403,8 @@ static int qlcnic_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
 	if (!adapter->fdb_mac_learn)
 		return ndo_dflt_fdb_add(ndm, tb, netdev, addr, flags);
 
-	if (!(adapter->flags & QLCNIC_ESWITCH_ENABLED)) {
+	if (!(adapter->flags & QLCNIC_ESWITCH_ENABLED) &&
+	    !qlcnic_sriov_check(adapter)) {
 		pr_info("%s: FDB e-switch is not enabled\n", __func__);
 		return -EOPNOTSUPP;
 	}
@@ -432,7 +434,8 @@ static int qlcnic_fdb_dump(struct sk_buff *skb, struct netlink_callback *ncb,
 	if (!adapter->fdb_mac_learn)
 		return ndo_dflt_fdb_dump(skb, ncb, netdev, idx);
 
-	if (adapter->flags & QLCNIC_ESWITCH_ENABLED)
+	if ((adapter->flags & QLCNIC_ESWITCH_ENABLED) ||
+	    qlcnic_sriov_check(adapter))
 		idx = ndo_dflt_fdb_dump(skb, ncb, netdev, idx);
 
 	return idx;
@@ -522,7 +525,7 @@ static const struct net_device_ops qlcnic_netdev_ops = {
 #endif
 #ifdef CONFIG_QLCNIC_SRIOV
 	.ndo_set_vf_mac		= qlcnic_sriov_set_vf_mac,
-	.ndo_set_vf_tx_rate	= qlcnic_sriov_set_vf_tx_rate,
+	.ndo_set_vf_rate	= qlcnic_sriov_set_vf_tx_rate,
 	.ndo_get_vf_config	= qlcnic_sriov_get_vf_config,
 	.ndo_set_vf_vlan	= qlcnic_sriov_set_vf_vlan,
 	.ndo_set_vf_spoofchk	= qlcnic_sriov_set_vf_spoofchk,
@@ -690,10 +693,10 @@ int qlcnic_setup_tss_rss_intr(struct qlcnic_adapter *adapter)
 		adapter->msix_entries[vector].entry = vector;
 
 restore:
-	err = pci_enable_msix(pdev, adapter->msix_entries, num_msix);
-	if (err > 0) {
+	err = pci_enable_msix_exact(pdev, adapter->msix_entries, num_msix);
+	if (err == -ENOSPC) {
 		if (!adapter->drv_tss_rings && !adapter->drv_rss_rings)
-			return -ENOSPC;
+			return err;
 
 		netdev_info(adapter->netdev,
 			    "Unable to allocate %d MSI-X vectors, Available vectors %d\n",
@@ -1014,6 +1017,8 @@ int qlcnic_init_pci_info(struct qlcnic_adapter *adapter)
 
 		if (pfn >= ahw->max_vnic_func) {
 			ret = QL_STATUS_INVALID_PARAM;
+			dev_err(&adapter->pdev->dev, "%s: Invalid function 0x%x, max 0x%x\n",
+				__func__, pfn, ahw->max_vnic_func);
 			goto err_eswitch;
 		}
 
@@ -1915,8 +1920,6 @@ void __qlcnic_down(struct qlcnic_adapter *adapter, struct net_device *netdev)
 	if (!test_and_clear_bit(__QLCNIC_DEV_UP, &adapter->state))
 		return;
 
-	if (qlcnic_sriov_vf_check(adapter))
-		qlcnic_sriov_cleanup_async_list(&adapter->ahw->sriov->bc);
 	smp_mb();
 	netif_carrier_off(netdev);
 	adapter->ahw->linkup = 0;
@@ -1928,6 +1931,8 @@ void __qlcnic_down(struct qlcnic_adapter *adapter, struct net_device *netdev)
 		qlcnic_delete_lb_filters(adapter);
 
 	qlcnic_nic_set_promisc(adapter, QLCNIC_NIU_NON_PROMISC_MODE);
+	if (qlcnic_sriov_vf_check(adapter))
+		qlcnic_sriov_cleanup_async_list(&adapter->ahw->sriov->bc);
 
 	qlcnic_napi_disable(adapter);
 
@@ -2052,6 +2057,7 @@ out:
 
 static int qlcnic_alloc_adapter_resources(struct qlcnic_adapter *adapter)
 {
+	struct qlcnic_hardware_context *ahw = adapter->ahw;
 	int err = 0;
 
 	adapter->recv_ctx = kzalloc(sizeof(struct qlcnic_recv_context),
@@ -2061,6 +2067,18 @@ static int qlcnic_alloc_adapter_resources(struct qlcnic_adapter *adapter)
 		goto err_out;
 	}
 
+	if (qlcnic_83xx_check(adapter)) {
+		ahw->coal.type = QLCNIC_INTR_COAL_TYPE_RX_TX;
+		ahw->coal.tx_time_us = QLCNIC_DEF_INTR_COALESCE_TX_TIME_US;
+		ahw->coal.tx_packets = QLCNIC_DEF_INTR_COALESCE_TX_PACKETS;
+		ahw->coal.rx_time_us = QLCNIC_DEF_INTR_COALESCE_RX_TIME_US;
+		ahw->coal.rx_packets = QLCNIC_DEF_INTR_COALESCE_RX_PACKETS;
+	} else {
+		ahw->coal.type = QLCNIC_INTR_COAL_TYPE_RX;
+		ahw->coal.rx_time_us = QLCNIC_DEF_INTR_COALESCE_RX_TIME_US;
+		ahw->coal.rx_packets = QLCNIC_DEF_INTR_COALESCE_RX_PACKETS;
+	}
+
 	/* clear stats */
 	memset(&adapter->stats, 0, sizeof(adapter->stats));
 err_out:
@@ -2069,12 +2087,20 @@ err_out:
 
 static void qlcnic_free_adapter_resources(struct qlcnic_adapter *adapter)
 {
+	struct qlcnic_fw_dump *fw_dump = &adapter->ahw->fw_dump;
+
 	kfree(adapter->recv_ctx);
 	adapter->recv_ctx = NULL;
 
-	if (adapter->ahw->fw_dump.tmpl_hdr) {
-		vfree(adapter->ahw->fw_dump.tmpl_hdr);
-		adapter->ahw->fw_dump.tmpl_hdr = NULL;
+	if (fw_dump->tmpl_hdr) {
+		vfree(fw_dump->tmpl_hdr);
+		fw_dump->tmpl_hdr = NULL;
+	}
+
+	if (fw_dump->dma_buffer) {
+		dma_free_coherent(&adapter->pdev->dev, QLC_PEX_DMA_READ_SIZE,
+				  fw_dump->dma_buffer, fw_dump->phys_addr);
+		fw_dump->dma_buffer = NULL;
 	}
 
 	kfree(adapter->ahw->reset.buff);
@@ -2247,10 +2273,8 @@ qlcnic_setup_netdev(struct qlcnic_adapter *adapter, struct net_device *netdev,
 
 	qlcnic_change_mtu(netdev, netdev->mtu);
 
-	if (qlcnic_sriov_vf_check(adapter))
-		SET_ETHTOOL_OPS(netdev, &qlcnic_sriov_vf_ethtool_ops);
-	else
-		SET_ETHTOOL_OPS(netdev, &qlcnic_ethtool_ops);
+	netdev->ethtool_ops = (qlcnic_sriov_vf_check(adapter)) ?
+		&qlcnic_sriov_vf_ethtool_ops : &qlcnic_ethtool_ops;
 
 	netdev->features |= (NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
 			     NETIF_F_IPV6_CSUM | NETIF_F_GRO |
@@ -2417,9 +2441,6 @@ qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int err, pci_using_dac = -1;
 	char board_name[QLCNIC_MAX_BOARD_NAME_LEN + 19]; /* MAC + ": " + name */
 
-	if (pdev->is_virtfn)
-		return -ENODEV;
-
 	err = pci_enable_device(pdev);
 	if (err)
 		return err;
@@ -2552,9 +2573,11 @@ qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			case -ENOMEM:
 				dev_err(&pdev->dev, "Adapter initialization failed. Please reboot\n");
 				goto err_out_free_hw;
+			case -EOPNOTSUPP:
+				dev_err(&pdev->dev, "Adapter initialization failed\n");
+				goto err_out_free_hw;
 			default:
-				dev_err(&pdev->dev, "Adapter initialization failed. A reboot may be required to recover from this failure\n");
-				dev_err(&pdev->dev, "If reboot does not help to recover from this failure, try a flash update of the adapter\n");
+				dev_err(&pdev->dev, "Adapter initialization failed. Driver will load in maintenance mode to recover the adapter using the application\n");
 				goto err_out_maintenance_mode;
 			}
 		}
@@ -2628,7 +2651,7 @@ qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		qlcnic_alloc_lb_filters_mem(adapter);
 
 	qlcnic_add_sysfs(adapter);
-
+	qlcnic_register_hwmon_dev(adapter);
 	return 0;
 
 err_out_disable_mbx_intr:
@@ -2665,7 +2688,7 @@ err_out_disable_pdev:
 err_out_maintenance_mode:
 	set_bit(__QLCNIC_MAINTENANCE_MODE, &adapter->state);
 	netdev->netdev_ops = &qlcnic_netdev_failed_ops;
-	SET_ETHTOOL_OPS(netdev, &qlcnic_ethtool_failed_ops);
+	netdev->ethtool_ops = &qlcnic_ethtool_failed_ops;
 	ahw->port_type = QLCNIC_XGBE;
 
 	if (qlcnic_83xx_check(adapter))
@@ -2698,9 +2721,9 @@ static void qlcnic_remove(struct pci_dev *pdev)
 		return;
 
 	netdev = adapter->netdev;
-	qlcnic_sriov_pf_disable(adapter);
 
 	qlcnic_cancel_idc_work(adapter);
+	qlcnic_sriov_pf_disable(adapter);
 	ahw = adapter->ahw;
 
 	unregister_netdev(netdev);
@@ -2735,6 +2758,8 @@ static void qlcnic_remove(struct pci_dev *pdev)
 
 	qlcnic_remove_sysfs(adapter);
 
+	qlcnic_unregister_hwmon_dev(adapter);
+
 	qlcnic_cleanup_pci_map(adapter->ahw);
 
 	qlcnic_release_firmware(adapter);
@@ -2828,6 +2853,8 @@ static int qlcnic_close(struct net_device *netdev)
 	return 0;
 }
 
+#define QLCNIC_VF_LB_BUCKET_SIZE 1
+
 void qlcnic_alloc_lb_filters_mem(struct qlcnic_adapter *adapter)
 {
 	void *head;
@@ -2843,7 +2870,10 @@ void qlcnic_alloc_lb_filters_mem(struct qlcnic_adapter *adapter)
 	spin_lock_init(&adapter->mac_learn_lock);
 	spin_lock_init(&adapter->rx_mac_learn_lock);
 
-	if (qlcnic_82xx_check(adapter)) {
+	if (qlcnic_sriov_vf_check(adapter)) {
+		filter_size = QLCNIC_83XX_SRIOV_VF_MAX_MAC - 1;
+		adapter->fhash.fbucket_size = QLCNIC_VF_LB_BUCKET_SIZE;
+	} else if (qlcnic_82xx_check(adapter)) {
 		filter_size = QLCNIC_LB_MAX_FILTERS;
 		adapter->fhash.fbucket_size = QLCNIC_LB_BUCKET_SIZE;
 	} else {
@@ -3973,16 +4003,6 @@ int qlcnic_validate_rings(struct qlcnic_adapter *adapter, __u32 ring_cnt,
 		strcpy(buf, "Tx");
 	}
 
-	if (!QLCNIC_IS_MSI_FAMILY(adapter)) {
-		netdev_err(netdev, "No RSS/TSS support in INT-x mode\n");
-		return -EINVAL;
-	}
-
-	if (adapter->flags & QLCNIC_MSI_ENABLED) {
-		netdev_err(netdev, "No RSS/TSS support in MSI mode\n");
-		return -EINVAL;
-	}
-
 	if (!is_power_of_2(ring_cnt)) {
 		netdev_err(netdev, "%s rings value should be a power of 2\n",
 			   buf);
@@ -4122,7 +4142,7 @@ void qlcnic_restore_indev_addr(struct net_device *netdev, unsigned long event)
 
 	rcu_read_lock();
 	for_each_set_bit(vid, adapter->vlans, VLAN_N_VID) {
-		dev = __vlan_find_dev_deep(netdev, htons(ETH_P_8021Q), vid);
+		dev = __vlan_find_dev_deep_rcu(netdev, htons(ETH_P_8021Q), vid);
 		if (!dev)
 			continue;
 		qlcnic_config_indev_addr(adapter, dev, event);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
index 37b979b1266b..e46fc39d425d 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
@@ -238,6 +238,8 @@ void qlcnic_82xx_cache_tmpl_hdr_values(struct qlcnic_fw_dump *fw_dump)
 
 	hdr->drv_cap_mask = hdr->cap_mask;
 	fw_dump->cap_mask = hdr->cap_mask;
+
+	fw_dump->use_pex_dma = (hdr->capabilities & BIT_0) ? true : false;
 }
 
 inline u32 qlcnic_82xx_get_cap_size(void *t_hdr, int index)
@@ -276,6 +278,8 @@ inline void qlcnic_83xx_set_saved_state(void *t_hdr, u32 index,
 	hdr->saved_state[index] = value;
 }
 
+#define QLCNIC_TEMPLATE_VERSION (0x20001)
+
 void qlcnic_83xx_cache_tmpl_hdr_values(struct qlcnic_fw_dump *fw_dump)
 {
 	struct qlcnic_83xx_dump_template_hdr *hdr;
@@ -288,6 +292,9 @@ void qlcnic_83xx_cache_tmpl_hdr_values(struct qlcnic_fw_dump *fw_dump)
 
 	hdr->drv_cap_mask = hdr->cap_mask;
 	fw_dump->cap_mask = hdr->cap_mask;
+
+	fw_dump->use_pex_dma = (fw_dump->version & 0xfffff) >=
+			       QLCNIC_TEMPLATE_VERSION;
 }
 
 inline u32 qlcnic_83xx_get_cap_size(void *t_hdr, int index)
@@ -653,34 +660,31 @@ out:
 #define QLC_DMA_CMD_BUFF_ADDR_HI	4
 #define QLC_DMA_CMD_STATUS_CTRL		8
 
-#define QLC_PEX_DMA_READ_SIZE		(PAGE_SIZE * 16)
-
 static int qlcnic_start_pex_dma(struct qlcnic_adapter *adapter,
 				struct __mem *mem)
 {
-	struct qlcnic_83xx_dump_template_hdr *tmpl_hdr;
 	struct device *dev = &adapter->pdev->dev;
 	u32 dma_no, dma_base_addr, temp_addr;
 	int i, ret, dma_sts;
+	void *tmpl_hdr;
 
 	tmpl_hdr = adapter->ahw->fw_dump.tmpl_hdr;
-	dma_no = tmpl_hdr->saved_state[QLC_83XX_DMA_ENGINE_INDEX];
+	dma_no = qlcnic_get_saved_state(adapter, tmpl_hdr,
+					QLC_83XX_DMA_ENGINE_INDEX);
 	dma_base_addr = QLC_DMA_REG_BASE_ADDR(dma_no);
 
 	temp_addr = dma_base_addr + QLC_DMA_CMD_BUFF_ADDR_LOW;
-	ret = qlcnic_83xx_wrt_reg_indirect(adapter, temp_addr,
-					   mem->desc_card_addr);
+	ret = qlcnic_ind_wr(adapter, temp_addr, mem->desc_card_addr);
 	if (ret)
 		return ret;
 
 	temp_addr = dma_base_addr + QLC_DMA_CMD_BUFF_ADDR_HI;
-	ret = qlcnic_83xx_wrt_reg_indirect(adapter, temp_addr, 0);
+	ret = qlcnic_ind_wr(adapter, temp_addr, 0);
 	if (ret)
 		return ret;
 
 	temp_addr = dma_base_addr + QLC_DMA_CMD_STATUS_CTRL;
-	ret = qlcnic_83xx_wrt_reg_indirect(adapter, temp_addr,
-					   mem->start_dma_cmd);
+	ret = qlcnic_ind_wr(adapter, temp_addr, mem->start_dma_cmd);
 	if (ret)
 		return ret;
 
@@ -710,15 +714,16 @@ static u32 qlcnic_read_memory_pexdma(struct qlcnic_adapter *adapter,
 	struct qlcnic_fw_dump *fw_dump = &adapter->ahw->fw_dump;
 	u32 temp, dma_base_addr, size = 0, read_size = 0;
 	struct qlcnic_pex_dma_descriptor *dma_descr;
-	struct qlcnic_83xx_dump_template_hdr *tmpl_hdr;
 	struct device *dev = &adapter->pdev->dev;
 	dma_addr_t dma_phys_addr;
 	void *dma_buffer;
+	void *tmpl_hdr;
 
 	tmpl_hdr = fw_dump->tmpl_hdr;
 
 	/* Check if DMA engine is available */
-	temp = tmpl_hdr->saved_state[QLC_83XX_DMA_ENGINE_INDEX];
+	temp = qlcnic_get_saved_state(adapter, tmpl_hdr,
+				      QLC_83XX_DMA_ENGINE_INDEX);
 	dma_base_addr = QLC_DMA_REG_BASE_ADDR(temp);
 	temp = qlcnic_ind_rd(adapter,
 			     dma_base_addr + QLC_DMA_CMD_STATUS_CTRL);
@@ -764,8 +769,8 @@ static u32 qlcnic_read_memory_pexdma(struct qlcnic_adapter *adapter,
 
 		/* Write DMA descriptor to MS memory*/
 		temp = sizeof(struct qlcnic_pex_dma_descriptor) / 16;
-		*ret = qlcnic_83xx_ms_mem_write128(adapter, mem->desc_card_addr,
-						   (u32 *)dma_descr, temp);
+		*ret = qlcnic_ms_mem_write128(adapter, mem->desc_card_addr,
+					      (u32 *)dma_descr, temp);
 		if (*ret) {
 			dev_info(dev, "Failed to write DMA descriptor to MS memory at address 0x%x\n",
 				 mem->desc_card_addr);
@@ -1141,8 +1146,6 @@ free_mem:
 	return err;
 }
 
-#define QLCNIC_TEMPLATE_VERSION (0x20001)
-
 int qlcnic_fw_cmd_get_minidump_temp(struct qlcnic_adapter *adapter)
 {
 	struct qlcnic_hardware_context *ahw;
@@ -1150,6 +1153,7 @@ int qlcnic_fw_cmd_get_minidump_temp(struct qlcnic_adapter *adapter)
 	u32 version, csum, *tmp_buf;
 	u8 use_flash_temp = 0;
 	u32 temp_size = 0;
+	void *temp_buffer;
 	int err;
 
 	ahw = adapter->ahw;
@@ -1199,16 +1203,23 @@ flash_temp:
 
 	qlcnic_cache_tmpl_hdr_values(adapter, fw_dump);
 
+	if (fw_dump->use_pex_dma) {
+		fw_dump->dma_buffer = NULL;
+		temp_buffer = dma_alloc_coherent(&adapter->pdev->dev,
+						 QLC_PEX_DMA_READ_SIZE,
+						 &fw_dump->phys_addr,
+						 GFP_KERNEL);
+		if (!temp_buffer)
+			fw_dump->use_pex_dma = false;
+		else
+			fw_dump->dma_buffer = temp_buffer;
+	}
+
+
 	dev_info(&adapter->pdev->dev,
 		 "Default minidump capture mask 0x%x\n",
 		 fw_dump->cap_mask);
 
-	if (qlcnic_83xx_check(adapter) &&
-	    (fw_dump->version & 0xfffff) >= QLCNIC_TEMPLATE_VERSION)
-		fw_dump->use_pex_dma = true;
-	else
-		fw_dump->use_pex_dma = false;
-
 	qlcnic_enable_fw_dump_state(adapter);
 
 	return 0;
@@ -1224,7 +1235,7 @@ int qlcnic_dump_fw(struct qlcnic_adapter *adapter)
 	struct device *dev = &adapter->pdev->dev;
 	struct qlcnic_hardware_context *ahw;
 	struct qlcnic_dump_entry *entry;
-	void *temp_buffer, *tmpl_hdr;
+	void *tmpl_hdr;
 	u32 ocm_window;
 	__le32 *buffer;
 	char mesg[64];
@@ -1268,16 +1279,6 @@ int qlcnic_dump_fw(struct qlcnic_adapter *adapter)
 	qlcnic_set_sys_info(adapter, tmpl_hdr, 0, QLCNIC_DRIVER_VERSION);
 	qlcnic_set_sys_info(adapter, tmpl_hdr, 1, adapter->fw_version);
 
-	if (fw_dump->use_pex_dma) {
-		temp_buffer = dma_alloc_coherent(dev, QLC_PEX_DMA_READ_SIZE,
-						 &fw_dump->phys_addr,
-						 GFP_KERNEL);
-		if (!temp_buffer)
-			fw_dump->use_pex_dma = false;
-		else
-			fw_dump->dma_buffer = temp_buffer;
-	}
-
 	if (qlcnic_82xx_check(adapter)) {
 		ops_cnt = ARRAY_SIZE(qlcnic_fw_dump_ops);
 		fw_dump_ops = qlcnic_fw_dump_ops;
@@ -1335,10 +1336,6 @@ int qlcnic_dump_fw(struct qlcnic_adapter *adapter)
 	/* Send a udev event to notify availability of FW dump */
 	kobject_uevent_env(&dev->kobj, KOBJ_CHANGE, msg);
 
-	if (fw_dump->use_pex_dma)
-		dma_free_coherent(dev, QLC_PEX_DMA_READ_SIZE,
-				  fw_dump->dma_buffer, fw_dump->phys_addr);
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h
index 396bd1fd1d27..4677b2edccca 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h
@@ -52,6 +52,7 @@ enum qlcnic_bc_commands {
 	QLCNIC_BC_CMD_CFG_GUEST_VLAN = 0x3,
 };
 
+#define QLCNIC_83XX_SRIOV_VF_MAX_MAC 2
 #define QLC_BC_CMD 1
 
 struct qlcnic_trans_list {
@@ -151,13 +152,14 @@ struct qlcnic_vf_info {
 	struct qlcnic_trans_list	rcv_pend;
 	struct qlcnic_adapter		*adapter;
 	struct qlcnic_vport		*vp;
-	struct mutex			vlan_list_lock;	/* Lock for VLAN list */
+	spinlock_t			vlan_list_lock;	/* Lock for VLAN list */
 };
 
 struct qlcnic_async_work_list {
 	struct list_head	list;
 	struct work_struct	work;
 	void			*ptr;
+	struct qlcnic_cmd_args	*cmd;
 };
 
 struct qlcnic_back_channel {
@@ -231,7 +233,7 @@ bool qlcnic_sriov_soft_flr_check(struct qlcnic_adapter *,
 void qlcnic_sriov_pf_reset(struct qlcnic_adapter *);
 int qlcnic_sriov_pf_reinit(struct qlcnic_adapter *);
 int qlcnic_sriov_set_vf_mac(struct net_device *, int, u8 *);
-int qlcnic_sriov_set_vf_tx_rate(struct net_device *, int, int);
+int qlcnic_sriov_set_vf_tx_rate(struct net_device *, int, int, int);
 int qlcnic_sriov_get_vf_config(struct net_device *, int ,
 			       struct ifla_vf_info *);
 int qlcnic_sriov_set_vf_vlan(struct net_device *, int, u16, u8);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c
index 6afe9c1f5ab9..1659c804f1d5 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c
@@ -39,6 +39,8 @@ static int qlcnic_sriov_channel_cfg_cmd(struct qlcnic_adapter *, u8);
 static void qlcnic_sriov_process_bc_cmd(struct work_struct *);
 static int qlcnic_sriov_vf_shutdown(struct pci_dev *);
 static int qlcnic_sriov_vf_resume(struct qlcnic_adapter *);
+static int qlcnic_sriov_async_issue_cmd(struct qlcnic_adapter *,
+					struct qlcnic_cmd_args *);
 
 static struct qlcnic_hardware_ops qlcnic_sriov_vf_hw_ops = {
 	.read_crb			= qlcnic_83xx_read_crb,
@@ -181,7 +183,7 @@ int qlcnic_sriov_init(struct qlcnic_adapter *adapter, int num_vfs)
 		vf->adapter = adapter;
 		vf->pci_func = qlcnic_sriov_virtid_fn(adapter, i);
 		mutex_init(&vf->send_cmd_lock);
-		mutex_init(&vf->vlan_list_lock);
+		spin_lock_init(&vf->vlan_list_lock);
 		INIT_LIST_HEAD(&vf->rcv_act.wait_list);
 		INIT_LIST_HEAD(&vf->rcv_pend.wait_list);
 		spin_lock_init(&vf->rcv_act.lock);
@@ -197,8 +199,10 @@ int qlcnic_sriov_init(struct qlcnic_adapter *adapter, int num_vfs)
 				goto qlcnic_destroy_async_wq;
 			}
 			sriov->vf_info[i].vp = vp;
+			vp->vlan_mode = QLC_GUEST_VLAN_MODE;
 			vp->max_tx_bw = MAX_BW;
-			vp->spoofchk = true;
+			vp->min_tx_bw = MIN_BW;
+			vp->spoofchk = false;
 			random_ether_addr(vp->mac);
 			dev_info(&adapter->pdev->dev,
 				 "MAC Address %pM is configured for VF %d\n",
@@ -454,6 +458,7 @@ static int qlcnic_sriov_get_vf_acl(struct qlcnic_adapter *adapter)
 	struct qlcnic_cmd_args cmd;
 	int ret = 0;
 
+	memset(&cmd, 0, sizeof(cmd));
 	ret = qlcnic_sriov_alloc_bc_mbx_args(&cmd, QLCNIC_BC_CMD_GET_ACL);
 	if (ret)
 		return ret;
@@ -515,6 +520,8 @@ static int qlcnic_sriov_setup_vf(struct qlcnic_adapter *adapter,
 {
 	int err;
 
+	adapter->flags |= QLCNIC_VLAN_FILTERING;
+	adapter->ahw->total_nic_func = 1;
 	INIT_LIST_HEAD(&adapter->vf_mc_list);
 	if (!qlcnic_use_msi_x && !!qlcnic_use_msi)
 		dev_warn(&adapter->pdev->dev,
@@ -770,6 +777,7 @@ static int qlcnic_sriov_prepare_bc_hdr(struct qlcnic_bc_trans *trans,
 		cmd->req.arg = (u32 *)trans->req_pay;
 		cmd->rsp.arg = (u32 *)trans->rsp_pay;
 		cmd_op = cmd->req.arg[0] & 0xff;
+		cmd->cmd_op = cmd_op;
 		remainder = (trans->rsp_pay_size) % (bc_pay_sz);
 		num_frags = (trans->rsp_pay_size) / (bc_pay_sz);
 		if (remainder)
@@ -1356,7 +1364,7 @@ static int qlcnic_sriov_retry_bc_cmd(struct qlcnic_adapter *adapter,
 	return -EIO;
 }
 
-static int qlcnic_sriov_issue_cmd(struct qlcnic_adapter *adapter,
+static int __qlcnic_sriov_issue_cmd(struct qlcnic_adapter *adapter,
 				  struct qlcnic_cmd_args *cmd)
 {
 	struct qlcnic_hardware_context *ahw = adapter->ahw;
@@ -1408,12 +1416,17 @@ retry:
 	    (mbx_err_code == QLCNIC_MBX_PORT_RSP_OK)) {
 		rsp = QLCNIC_RCODE_SUCCESS;
 	} else {
-		rsp = mbx_err_code;
-		if (!rsp)
-			rsp = 1;
-		dev_err(dev,
-			"MBX command 0x%x failed with err:0x%x for VF %d\n",
-			opcode, mbx_err_code, func);
+		if (cmd->type == QLC_83XX_MBX_CMD_NO_WAIT) {
+			rsp = QLCNIC_RCODE_SUCCESS;
+		} else {
+			rsp = mbx_err_code;
+			if (!rsp)
+				rsp = 1;
+
+			dev_err(dev,
+				"MBX command 0x%x failed with err:0x%x for VF %d\n",
+				opcode, mbx_err_code, func);
+		}
 	}
 
 err_out:
@@ -1435,12 +1448,23 @@ free_cmd:
 	return rsp;
 }
 
+
+static int qlcnic_sriov_issue_cmd(struct qlcnic_adapter *adapter,
+				  struct qlcnic_cmd_args *cmd)
+{
+	if (cmd->type == QLC_83XX_MBX_CMD_NO_WAIT)
+		return qlcnic_sriov_async_issue_cmd(adapter, cmd);
+	else
+		return __qlcnic_sriov_issue_cmd(adapter, cmd);
+}
+
 static int qlcnic_sriov_channel_cfg_cmd(struct qlcnic_adapter *adapter, u8 cmd_op)
 {
 	struct qlcnic_cmd_args cmd;
 	struct qlcnic_vf_info *vf = &adapter->ahw->sriov->vf_info[0];
 	int ret;
 
+	memset(&cmd, 0, sizeof(cmd));
 	if (qlcnic_sriov_alloc_bc_mbx_args(&cmd, cmd_op))
 		return -ENOMEM;
 
@@ -1465,58 +1489,28 @@ out:
 	return ret;
 }
 
-static void qlcnic_vf_add_mc_list(struct net_device *netdev)
+static void qlcnic_vf_add_mc_list(struct net_device *netdev, const u8 *mac)
 {
 	struct qlcnic_adapter *adapter = netdev_priv(netdev);
 	struct qlcnic_sriov *sriov = adapter->ahw->sriov;
-	struct qlcnic_mac_vlan_list *cur;
-	struct list_head *head, tmp_list;
 	struct qlcnic_vf_info *vf;
 	u16 vlan_id;
 	int i;
 
-	static const u8 bcast_addr[ETH_ALEN] = {
-		0xff, 0xff, 0xff, 0xff, 0xff, 0xff
-	};
-
 	vf = &adapter->ahw->sriov->vf_info[0];
-	INIT_LIST_HEAD(&tmp_list);
-	head = &adapter->vf_mc_list;
-	netif_addr_lock_bh(netdev);
 
-	while (!list_empty(head)) {
-		cur = list_entry(head->next, struct qlcnic_mac_vlan_list, list);
-		list_move(&cur->list, &tmp_list);
-	}
-
-	netif_addr_unlock_bh(netdev);
-
-	while (!list_empty(&tmp_list)) {
-		cur = list_entry((&tmp_list)->next,
-				 struct qlcnic_mac_vlan_list, list);
-		if (!qlcnic_sriov_check_any_vlan(vf)) {
-			qlcnic_nic_add_mac(adapter, bcast_addr, 0);
-			qlcnic_nic_add_mac(adapter, cur->mac_addr, 0);
-		} else {
-			mutex_lock(&vf->vlan_list_lock);
-			for (i = 0; i < sriov->num_allowed_vlans; i++) {
-				vlan_id = vf->sriov_vlans[i];
-				if (vlan_id) {
-					qlcnic_nic_add_mac(adapter, bcast_addr,
-							   vlan_id);
-					qlcnic_nic_add_mac(adapter,
-							   cur->mac_addr,
-							   vlan_id);
-				}
-			}
-			mutex_unlock(&vf->vlan_list_lock);
-			if (qlcnic_84xx_check(adapter)) {
-				qlcnic_nic_add_mac(adapter, bcast_addr, 0);
-				qlcnic_nic_add_mac(adapter, cur->mac_addr, 0);
-			}
+	if (!qlcnic_sriov_check_any_vlan(vf)) {
+		qlcnic_nic_add_mac(adapter, mac, 0);
+	} else {
+		spin_lock(&vf->vlan_list_lock);
+		for (i = 0; i < sriov->num_allowed_vlans; i++) {
+			vlan_id = vf->sriov_vlans[i];
+			if (vlan_id)
+				qlcnic_nic_add_mac(adapter, mac, vlan_id);
 		}
-		list_del(&cur->list);
-		kfree(cur);
+		spin_unlock(&vf->vlan_list_lock);
+		if (qlcnic_84xx_check(adapter))
+			qlcnic_nic_add_mac(adapter, mac, 0);
 	}
 }
 
@@ -1525,6 +1519,7 @@ void qlcnic_sriov_cleanup_async_list(struct qlcnic_back_channel *bc)
 	struct list_head *head = &bc->async_list;
 	struct qlcnic_async_work_list *entry;
 
+	flush_workqueue(bc->bc_async_wq);
 	while (!list_empty(head)) {
 		entry = list_entry(head->next, struct qlcnic_async_work_list,
 				   list);
@@ -1534,10 +1529,14 @@ void qlcnic_sriov_cleanup_async_list(struct qlcnic_back_channel *bc)
 	}
 }
 
-static void qlcnic_sriov_vf_set_multi(struct net_device *netdev)
+void qlcnic_sriov_vf_set_multi(struct net_device *netdev)
 {
 	struct qlcnic_adapter *adapter = netdev_priv(netdev);
 	struct qlcnic_hardware_context *ahw = adapter->ahw;
+	static const u8 bcast_addr[ETH_ALEN] = {
+		0xff, 0xff, 0xff, 0xff, 0xff, 0xff
+	};
+	struct netdev_hw_addr *ha;
 	u32 mode = VPORT_MISS_MODE_DROP;
 
 	if (!test_bit(__QLCNIC_FW_ATTACHED, &adapter->state))
@@ -1549,23 +1548,49 @@ static void qlcnic_sriov_vf_set_multi(struct net_device *netdev)
 	} else if ((netdev->flags & IFF_ALLMULTI) ||
 		   (netdev_mc_count(netdev) > ahw->max_mc_count)) {
 		mode = VPORT_MISS_MODE_ACCEPT_MULTI;
+	} else {
+		qlcnic_vf_add_mc_list(netdev, bcast_addr);
+		if (!netdev_mc_empty(netdev)) {
+			netdev_for_each_mc_addr(ha, netdev)
+				qlcnic_vf_add_mc_list(netdev, ha->addr);
+		}
 	}
 
-	if (qlcnic_sriov_vf_check(adapter))
-		qlcnic_vf_add_mc_list(netdev);
+	/* configure unicast MAC address, if there is not sufficient space
+	 * to store all the unicast addresses then enable promiscuous mode
+	 */
+	if (netdev_uc_count(netdev) > ahw->max_uc_count) {
+		mode = VPORT_MISS_MODE_ACCEPT_ALL;
+	} else if (!netdev_uc_empty(netdev)) {
+		netdev_for_each_uc_addr(ha, netdev)
+			qlcnic_vf_add_mc_list(netdev, ha->addr);
+	}
+
+	if (adapter->pdev->is_virtfn) {
+		if (mode == VPORT_MISS_MODE_ACCEPT_ALL &&
+		    !adapter->fdb_mac_learn) {
+			qlcnic_alloc_lb_filters_mem(adapter);
+			adapter->drv_mac_learn = 1;
+			adapter->rx_mac_learn = true;
+		} else {
+			adapter->drv_mac_learn = 0;
+			adapter->rx_mac_learn = false;
+		}
+	}
 
 	qlcnic_nic_set_promisc(adapter, mode);
 }
 
-static void qlcnic_sriov_handle_async_multi(struct work_struct *work)
+static void qlcnic_sriov_handle_async_issue_cmd(struct work_struct *work)
 {
 	struct qlcnic_async_work_list *entry;
-	struct net_device *netdev;
+	struct qlcnic_adapter *adapter;
+	struct qlcnic_cmd_args *cmd;
 
 	entry = container_of(work, struct qlcnic_async_work_list, work);
-	netdev = (struct net_device *)entry->ptr;
-
-	qlcnic_sriov_vf_set_multi(netdev);
+	adapter = entry->ptr;
+	cmd = entry->cmd;
+	__qlcnic_sriov_issue_cmd(adapter, cmd);
 	return;
 }
 
@@ -1595,8 +1620,9 @@ qlcnic_sriov_get_free_node_async_work(struct qlcnic_back_channel *bc)
 	return entry;
 }
 
-static void qlcnic_sriov_schedule_bc_async_work(struct qlcnic_back_channel *bc,
-						work_func_t func, void *data)
+static void qlcnic_sriov_schedule_async_cmd(struct qlcnic_back_channel *bc,
+					    work_func_t func, void *data,
+					    struct qlcnic_cmd_args *cmd)
 {
 	struct qlcnic_async_work_list *entry = NULL;
 
@@ -1605,21 +1631,23 @@ static void qlcnic_sriov_schedule_bc_async_work(struct qlcnic_back_channel *bc,
 		return;
 
 	entry->ptr = data;
+	entry->cmd = cmd;
 	INIT_WORK(&entry->work, func);
 	queue_work(bc->bc_async_wq, &entry->work);
 }
 
-void qlcnic_sriov_vf_schedule_multi(struct net_device *netdev)
+static int qlcnic_sriov_async_issue_cmd(struct qlcnic_adapter *adapter,
+					struct qlcnic_cmd_args *cmd)
 {
 
-	struct qlcnic_adapter *adapter = netdev_priv(netdev);
 	struct qlcnic_back_channel *bc = &adapter->ahw->sriov->bc;
 
 	if (adapter->need_fw_reset)
-		return;
+		return -EIO;
 
-	qlcnic_sriov_schedule_bc_async_work(bc, qlcnic_sriov_handle_async_multi,
-					    netdev);
+	qlcnic_sriov_schedule_async_cmd(bc, qlcnic_sriov_handle_async_issue_cmd,
+					adapter, cmd);
+	return 0;
 }
 
 static int qlcnic_sriov_vf_reinit_driver(struct qlcnic_adapter *adapter)
@@ -1843,6 +1871,12 @@ static int qlcnic_sriov_vf_idc_unknown_state(struct qlcnic_adapter *adapter)
 	return 0;
 }
 
+static void qlcnic_sriov_vf_periodic_tasks(struct qlcnic_adapter *adapter)
+{
+	if (adapter->fhash.fnum)
+		qlcnic_prune_lb_filters(adapter);
+}
+
 static void qlcnic_sriov_vf_poll_dev_state(struct work_struct *work)
 {
 	struct qlcnic_adapter *adapter;
@@ -1874,6 +1908,8 @@ static void qlcnic_sriov_vf_poll_dev_state(struct work_struct *work)
 	}
 
 	idc->prev_state = idc->curr_state;
+	qlcnic_sriov_vf_periodic_tasks(adapter);
+
 	if (!ret && test_bit(QLC_83XX_MODULE_LOADED, &idc->status))
 		qlcnic_schedule_work(adapter, qlcnic_sriov_vf_poll_dev_state,
 				     idc->delay);
@@ -1897,7 +1933,7 @@ static int qlcnic_sriov_check_vlan_id(struct qlcnic_sriov *sriov,
 	if (!vf->sriov_vlans)
 		return err;
 
-	mutex_lock(&vf->vlan_list_lock);
+	spin_lock_bh(&vf->vlan_list_lock);
 
 	for (i = 0; i < sriov->num_allowed_vlans; i++) {
 		if (vf->sriov_vlans[i] == vlan_id) {
@@ -1906,7 +1942,7 @@ static int qlcnic_sriov_check_vlan_id(struct qlcnic_sriov *sriov,
 		}
 	}
 
-	mutex_unlock(&vf->vlan_list_lock);
+	spin_unlock_bh(&vf->vlan_list_lock);
 	return err;
 }
 
@@ -1915,12 +1951,12 @@ static int qlcnic_sriov_validate_num_vlans(struct qlcnic_sriov *sriov,
 {
 	int err = 0;
 
-	mutex_lock(&vf->vlan_list_lock);
+	spin_lock_bh(&vf->vlan_list_lock);
 
 	if (vf->num_vlan >= sriov->num_allowed_vlans)
 		err = -EINVAL;
 
-	mutex_unlock(&vf->vlan_list_lock);
+	spin_unlock_bh(&vf->vlan_list_lock);
 	return err;
 }
 
@@ -1973,7 +2009,7 @@ static void qlcnic_sriov_vlan_operation(struct qlcnic_vf_info *vf, u16 vlan_id,
 	if (!vf->sriov_vlans)
 		return;
 
-	mutex_lock(&vf->vlan_list_lock);
+	spin_lock_bh(&vf->vlan_list_lock);
 
 	switch (opcode) {
 	case QLC_VLAN_ADD:
@@ -1986,7 +2022,7 @@ static void qlcnic_sriov_vlan_operation(struct qlcnic_vf_info *vf, u16 vlan_id,
 		netdev_err(adapter->netdev, "Invalid VLAN operation\n");
 	}
 
-	mutex_unlock(&vf->vlan_list_lock);
+	spin_unlock_bh(&vf->vlan_list_lock);
 	return;
 }
 
@@ -1994,10 +2030,12 @@ int qlcnic_sriov_cfg_vf_guest_vlan(struct qlcnic_adapter *adapter,
 				   u16 vid, u8 enable)
 {
 	struct qlcnic_sriov *sriov = adapter->ahw->sriov;
+	struct net_device *netdev = adapter->netdev;
 	struct qlcnic_vf_info *vf;
 	struct qlcnic_cmd_args cmd;
 	int ret;
 
+	memset(&cmd, 0, sizeof(cmd));
 	if (vid == 0)
 		return 0;
 
@@ -2019,14 +2057,18 @@ int qlcnic_sriov_cfg_vf_guest_vlan(struct qlcnic_adapter *adapter,
 		dev_err(&adapter->pdev->dev,
 			"Failed to configure guest VLAN, err=%d\n", ret);
 	} else {
+		netif_addr_lock_bh(netdev);
 		qlcnic_free_mac_list(adapter);
+		netif_addr_unlock_bh(netdev);
 
 		if (enable)
 			qlcnic_sriov_vlan_operation(vf, vid, QLC_VLAN_ADD);
 		else
 			qlcnic_sriov_vlan_operation(vf, vid, QLC_VLAN_DELETE);
 
-		qlcnic_set_multi(adapter->netdev);
+		netif_addr_lock_bh(netdev);
+		qlcnic_set_multi(netdev);
+		netif_addr_unlock_bh(netdev);
 	}
 
 	qlcnic_free_mbx_args(&cmd);
@@ -2157,11 +2199,11 @@ bool qlcnic_sriov_check_any_vlan(struct qlcnic_vf_info *vf)
 {
 	bool err = false;
 
-	mutex_lock(&vf->vlan_list_lock);
+	spin_lock_bh(&vf->vlan_list_lock);
 
 	if (vf->num_vlan)
 		err = true;
 
-	mutex_unlock(&vf->vlan_list_lock);
+	spin_unlock_bh(&vf->vlan_list_lock);
 	return err;
 }
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c
index 280137991544..a29538b86edf 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c
@@ -16,6 +16,7 @@
 #define QLC_VF_FLOOD_BIT	BIT_16
 #define QLC_FLOOD_MODE		0x5
 #define QLC_SRIOV_ALLOW_VLAN0	BIT_19
+#define QLC_INTR_COAL_TYPE_MASK	0x7
 
 static int qlcnic_sriov_pf_get_vport_handle(struct qlcnic_adapter *, u8);
 
@@ -83,7 +84,7 @@ static int qlcnic_sriov_pf_cal_res_limit(struct qlcnic_adapter *adapter,
 	info->max_tx_ques = res->num_tx_queues / max;
 
 	if (qlcnic_83xx_pf_check(adapter))
-		num_macs = 1;
+		num_macs = QLCNIC_83XX_SRIOV_VF_MAX_MAC;
 
 	info->max_rx_mcast_mac_filters = res->num_rx_mcast_mac_filters;
 
@@ -337,9 +338,12 @@ static int qlcnic_sriov_pf_cfg_vlan_filtering(struct qlcnic_adapter *adapter,
 
 	cmd.req.arg[1] = 0x4;
 	if (enable) {
+		adapter->flags |= QLCNIC_VLAN_FILTERING;
 		cmd.req.arg[1] |= BIT_16;
 		if (qlcnic_84xx_check(adapter))
 			cmd.req.arg[1] |= QLC_SRIOV_ALLOW_VLAN0;
+	} else {
+		adapter->flags &= ~QLCNIC_VLAN_FILTERING;
 	}
 
 	err = qlcnic_issue_cmd(adapter, &cmd);
@@ -471,12 +475,12 @@ static int qlcnic_pci_sriov_disable(struct qlcnic_adapter *adapter)
 		return -EPERM;
 	}
 
+	qlcnic_sriov_pf_disable(adapter);
+
 	rtnl_lock();
 	if (netif_running(netdev))
 		__qlcnic_down(adapter, netdev);
 
-	qlcnic_sriov_pf_disable(adapter);
-
 	qlcnic_sriov_free_vlans(adapter);
 
 	qlcnic_sriov_pf_cleanup(adapter);
@@ -595,7 +599,6 @@ static int __qlcnic_pci_sriov_enable(struct qlcnic_adapter *adapter,
 
 	qlcnic_sriov_alloc_vlans(adapter);
 
-	err = qlcnic_sriov_pf_enable(adapter, num_vfs);
 	return err;
 
 del_flr_queue:
@@ -626,25 +629,36 @@ static int qlcnic_pci_sriov_enable(struct qlcnic_adapter *adapter, int num_vfs)
 		__qlcnic_down(adapter, netdev);
 
 	err = __qlcnic_pci_sriov_enable(adapter, num_vfs);
-	if (err) {
-		netdev_info(netdev, "Failed to enable SR-IOV on port %d\n",
-			    adapter->portnum);
+	if (err)
+		goto error;
 
-		err = -EIO;
-		if (qlcnic_83xx_configure_opmode(adapter))
-			goto error;
-	} else {
+	if (netif_running(netdev))
+		__qlcnic_up(adapter, netdev);
+
+	rtnl_unlock();
+	err = qlcnic_sriov_pf_enable(adapter, num_vfs);
+	if (!err) {
 		netdev_info(netdev,
 			    "SR-IOV is enabled successfully on port %d\n",
 			    adapter->portnum);
 		/* Return number of vfs enabled */
-		err = num_vfs;
+		return num_vfs;
 	}
+
+	rtnl_lock();
 	if (netif_running(netdev))
-		__qlcnic_up(adapter, netdev);
+		__qlcnic_down(adapter, netdev);
 
 error:
+	if (!qlcnic_83xx_configure_opmode(adapter)) {
+		if (netif_running(netdev))
+			__qlcnic_up(adapter, netdev);
+	}
+
 	rtnl_unlock();
+	netdev_info(netdev, "Failed to enable SR-IOV on port %d\n",
+		    adapter->portnum);
+
 	return err;
 }
 
@@ -773,7 +787,7 @@ static int qlcnic_sriov_cfg_vf_def_mac(struct qlcnic_adapter *adapter,
 				       struct qlcnic_vf_info *vf,
 				       u16 vlan, u8 op)
 {
-	struct qlcnic_cmd_args cmd;
+	struct qlcnic_cmd_args *cmd;
 	struct qlcnic_macvlan_mbx mv;
 	struct qlcnic_vport *vp;
 	u8 *addr;
@@ -783,21 +797,27 @@ static int qlcnic_sriov_cfg_vf_def_mac(struct qlcnic_adapter *adapter,
 
 	vp = vf->vp;
 
-	if (qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_CONFIG_MAC_VLAN))
+	cmd = kzalloc(sizeof(*cmd), GFP_ATOMIC);
+	if (!cmd)
 		return -ENOMEM;
 
+	err = qlcnic_alloc_mbx_args(cmd, adapter, QLCNIC_CMD_CONFIG_MAC_VLAN);
+	if (err)
+		goto free_cmd;
+
+	cmd->type = QLC_83XX_MBX_CMD_NO_WAIT;
 	vpid = qlcnic_sriov_pf_get_vport_handle(adapter, vf->pci_func);
 	if (vpid < 0) {
 		err = -EINVAL;
-		goto out;
+		goto free_args;
 	}
 
 	if (vlan)
 		op = ((op == QLCNIC_MAC_ADD || op == QLCNIC_MAC_VLAN_ADD) ?
 		      QLCNIC_MAC_VLAN_ADD : QLCNIC_MAC_VLAN_DEL);
 
-	cmd.req.arg[1] = op | (1 << 8) | (3 << 6);
-	cmd.req.arg[1] |= ((vpid & 0xffff) << 16) | BIT_31;
+	cmd->req.arg[1] = op | (1 << 8) | (3 << 6);
+	cmd->req.arg[1] |= ((vpid & 0xffff) << 16) | BIT_31;
 
 	addr = vp->mac;
 	mv.vlan = vlan;
@@ -807,18 +827,18 @@ static int qlcnic_sriov_cfg_vf_def_mac(struct qlcnic_adapter *adapter,
 	mv.mac_addr3 = addr[3];
 	mv.mac_addr4 = addr[4];
 	mv.mac_addr5 = addr[5];
-	buf = &cmd.req.arg[2];
+	buf = &cmd->req.arg[2];
 	memcpy(buf, &mv, sizeof(struct qlcnic_macvlan_mbx));
 
-	err = qlcnic_issue_cmd(adapter, &cmd);
+	err = qlcnic_issue_cmd(adapter, cmd);
 
-	if (err)
-		dev_err(&adapter->pdev->dev,
-			"MAC-VLAN %s to CAM failed, err=%d.\n",
-			((op == 1) ? "add " : "delete "), err);
+	if (!err)
+		return err;
 
-out:
-	qlcnic_free_mbx_args(&cmd);
+free_args:
+	qlcnic_free_mbx_args(cmd);
+free_cmd:
+	kfree(cmd);
 	return err;
 }
 
@@ -840,7 +860,7 @@ static void qlcnic_83xx_cfg_default_mac_vlan(struct qlcnic_adapter *adapter,
 
 	sriov = adapter->ahw->sriov;
 
-	mutex_lock(&vf->vlan_list_lock);
+	spin_lock_bh(&vf->vlan_list_lock);
 	if (vf->num_vlan) {
 		for (i = 0; i < sriov->num_allowed_vlans; i++) {
 			vlan = vf->sriov_vlans[i];
@@ -849,7 +869,7 @@ static void qlcnic_83xx_cfg_default_mac_vlan(struct qlcnic_adapter *adapter,
 							    opcode);
 		}
 	}
-	mutex_unlock(&vf->vlan_list_lock);
+	spin_unlock_bh(&vf->vlan_list_lock);
 
 	if (vf->vp->vlan_mode != QLC_PVID_MODE) {
 		if (qlcnic_83xx_pf_check(adapter) &&
@@ -1178,19 +1198,41 @@ static int qlcnic_sriov_validate_cfg_intrcoal(struct qlcnic_adapter *adapter,
 {
 	struct qlcnic_nic_intr_coalesce *coal = &adapter->ahw->coal;
 	u16 ctx_id, pkts, time;
+	int err = -EINVAL;
+	u8 type;
 
+	type = cmd->req.arg[1] & QLC_INTR_COAL_TYPE_MASK;
 	ctx_id = cmd->req.arg[1] >> 16;
 	pkts = cmd->req.arg[2] & 0xffff;
 	time = cmd->req.arg[2] >> 16;
 
-	if (ctx_id != vf->rx_ctx_id)
-		return -EINVAL;
-	if (pkts > coal->rx_packets)
-		return -EINVAL;
-	if (time < coal->rx_time_us)
-		return -EINVAL;
+	switch (type) {
+	case QLCNIC_INTR_COAL_TYPE_RX:
+		if (ctx_id != vf->rx_ctx_id || pkts > coal->rx_packets ||
+		    time < coal->rx_time_us)
+			goto err_label;
+		break;
+	case QLCNIC_INTR_COAL_TYPE_TX:
+		if (ctx_id != vf->tx_ctx_id || pkts > coal->tx_packets ||
+		    time < coal->tx_time_us)
+			goto err_label;
+		break;
+	default:
+		netdev_err(adapter->netdev, "Invalid coalescing type 0x%x received\n",
+			   type);
+		return err;
+	}
 
 	return 0;
+
+err_label:
+	netdev_err(adapter->netdev, "Expected: rx_ctx_id 0x%x rx_packets 0x%x rx_time_us 0x%x tx_ctx_id 0x%x tx_packets 0x%x tx_time_us 0x%x\n",
+		   vf->rx_ctx_id, coal->rx_packets, coal->rx_time_us,
+		   vf->tx_ctx_id, coal->tx_packets, coal->tx_time_us);
+	netdev_err(adapter->netdev, "Received: ctx_id 0x%x packets 0x%x time_us 0x%x type 0x%x\n",
+		   ctx_id, pkts, time, type);
+
+	return err;
 }
 
 static int qlcnic_sriov_pf_cfg_intrcoal_cmd(struct qlcnic_bc_trans *tran,
@@ -1214,7 +1256,6 @@ static int qlcnic_sriov_validate_cfg_macvlan(struct qlcnic_adapter *adapter,
 					     struct qlcnic_vf_info *vf,
 					     struct qlcnic_cmd_args *cmd)
 {
-	struct qlcnic_macvlan_mbx *macvlan;
 	struct qlcnic_vport *vp = vf->vp;
 	u8 op, new_op;
 
@@ -1224,14 +1265,6 @@ static int qlcnic_sriov_validate_cfg_macvlan(struct qlcnic_adapter *adapter,
 	cmd->req.arg[1] |= (vf->vp->handle << 16);
 	cmd->req.arg[1] |= BIT_31;
 
-	macvlan = (struct qlcnic_macvlan_mbx *)&cmd->req.arg[2];
-	if (!(macvlan->mac_addr0 & BIT_0)) {
-		dev_err(&adapter->pdev->dev,
-			"MAC address change is not allowed from VF %d",
-			vf->pci_func);
-		return -EINVAL;
-	}
-
 	if (vp->vlan_mode == QLC_PVID_MODE) {
 		op = cmd->req.arg[1] & 0x7;
 		cmd->req.arg[1] &= ~0x7;
@@ -1815,7 +1848,8 @@ int qlcnic_sriov_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 	return 0;
 }
 
-int qlcnic_sriov_set_vf_tx_rate(struct net_device *netdev, int vf, int tx_rate)
+int qlcnic_sriov_set_vf_tx_rate(struct net_device *netdev, int vf,
+				int min_tx_rate, int max_tx_rate)
 {
 	struct qlcnic_adapter *adapter = netdev_priv(netdev);
 	struct qlcnic_sriov *sriov = adapter->ahw->sriov;
@@ -1830,35 +1864,52 @@ int qlcnic_sriov_set_vf_tx_rate(struct net_device *netdev, int vf, int tx_rate)
 	if (vf >= sriov->num_vfs)
 		return -EINVAL;
 
-	if (tx_rate >= 10000 || tx_rate < 100) {
+	vf_info = &sriov->vf_info[vf];
+	vp = vf_info->vp;
+	vpid = vp->handle;
+
+	if (!min_tx_rate)
+		min_tx_rate = QLC_VF_MIN_TX_RATE;
+
+	if (max_tx_rate &&
+	    (max_tx_rate >= 10000 || max_tx_rate < min_tx_rate)) {
 		netdev_err(netdev,
-			   "Invalid Tx rate, allowed range is [%d - %d]",
-			   QLC_VF_MIN_TX_RATE, QLC_VF_MAX_TX_RATE);
+			   "Invalid max Tx rate, allowed range is [%d - %d]",
+			   min_tx_rate, QLC_VF_MAX_TX_RATE);
 		return -EINVAL;
 	}
 
-	if (tx_rate == 0)
-		tx_rate = 10000;
+	if (!max_tx_rate)
+		max_tx_rate = 10000;
 
-	vf_info = &sriov->vf_info[vf];
-	vp = vf_info->vp;
-	vpid = vp->handle;
+	if (min_tx_rate &&
+	    (min_tx_rate > max_tx_rate || min_tx_rate < QLC_VF_MIN_TX_RATE)) {
+		netdev_err(netdev,
+			   "Invalid min Tx rate, allowed range is [%d - %d]",
+			   QLC_VF_MIN_TX_RATE, max_tx_rate);
+		return -EINVAL;
+	}
 
 	if (test_bit(QLC_BC_VF_STATE, &vf_info->state)) {
 		if (qlcnic_sriov_get_vf_vport_info(adapter, &nic_info, vpid))
 			return -EIO;
 
-		nic_info.max_tx_bw = tx_rate / 100;
+		nic_info.max_tx_bw = max_tx_rate / 100;
+		nic_info.min_tx_bw = min_tx_rate / 100;
 		nic_info.bit_offsets = BIT_0;
 
 		if (qlcnic_sriov_pf_set_vport_info(adapter, &nic_info, vpid))
 			return -EIO;
 	}
 
-	vp->max_tx_bw = tx_rate / 100;
+	vp->max_tx_bw = max_tx_rate / 100;
 	netdev_info(netdev,
-		    "Setting Tx rate %d (Mbps), %d %% of PF bandwidth, for VF %d\n",
-		    tx_rate, vp->max_tx_bw, vf);
+		    "Setting Max Tx rate %d (Mbps), %d %% of PF bandwidth, for VF %d\n",
+		    max_tx_rate, vp->max_tx_bw, vf);
+	vp->min_tx_bw = min_tx_rate / 100;
+	netdev_info(netdev,
+		    "Setting Min Tx rate %d (Mbps), %d %% of PF bandwidth, for VF %d\n",
+		    min_tx_rate, vp->min_tx_bw, vf);
 	return 0;
 }
 
@@ -1957,9 +2008,13 @@ int qlcnic_sriov_get_vf_config(struct net_device *netdev,
 	ivi->qos = vp->qos;
 	ivi->spoofchk = vp->spoofchk;
 	if (vp->max_tx_bw == MAX_BW)
-		ivi->tx_rate = 0;
+		ivi->max_tx_rate = 0;
+	else
+		ivi->max_tx_rate = vp->max_tx_bw * 100;
+	if (vp->min_tx_bw == MIN_BW)
+		ivi->min_tx_rate = 0;
 	else
-		ivi->tx_rate = vp->max_tx_bw * 100;
+		ivi->min_tx_rate = vp->min_tx_bw * 100;
 
 	ivi->vf = vf;
 	return 0;
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
index cd346e27f2e1..f5786d5792df 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
@@ -19,6 +19,10 @@
 #include <linux/sysfs.h>
 #include <linux/aer.h>
 #include <linux/log2.h>
+#ifdef CONFIG_QLCNIC_HWMON
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#endif
 
 #define QLC_STATUS_UNSUPPORTED_CMD	-2
 
@@ -358,6 +362,8 @@ int qlcnic_is_valid_nic_func(struct qlcnic_adapter *adapter, u8 pci_func)
 		if (adapter->npars[i].pci_func == pci_func)
 			return i;
 	}
+
+	dev_err(&adapter->pdev->dev, "%s: Invalid nic function\n", __func__);
 	return -EINVAL;
 }
 
@@ -1243,6 +1249,68 @@ static struct bin_attribute bin_attr_flash = {
 	.write = qlcnic_83xx_sysfs_flash_write_handler,
 };
 
+#ifdef CONFIG_QLCNIC_HWMON
+
+static ssize_t qlcnic_hwmon_show_temp(struct device *dev,
+				      struct device_attribute *dev_attr,
+				      char *buf)
+{
+	struct qlcnic_adapter *adapter = dev_get_drvdata(dev);
+	unsigned int temperature = 0, value = 0;
+
+	if (qlcnic_83xx_check(adapter))
+		value = QLCRDX(adapter->ahw, QLC_83XX_ASIC_TEMP);
+	else if (qlcnic_82xx_check(adapter))
+		value = QLC_SHARED_REG_RD32(adapter, QLCNIC_ASIC_TEMP);
+
+	temperature = qlcnic_get_temp_val(value);
+	/* display millidegree celcius */
+	temperature *= 1000;
+	return sprintf(buf, "%u\n", temperature);
+}
+
+/* hwmon-sysfs attributes */
+static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO,
+			  qlcnic_hwmon_show_temp, NULL, 1);
+
+static struct attribute *qlcnic_hwmon_attrs[] = {
+	&sensor_dev_attr_temp1_input.dev_attr.attr,
+	NULL
+};
+
+ATTRIBUTE_GROUPS(qlcnic_hwmon);
+
+void qlcnic_register_hwmon_dev(struct qlcnic_adapter *adapter)
+{
+	struct device *dev = &adapter->pdev->dev;
+	struct device *hwmon_dev;
+
+	/* Skip hwmon registration for a VF device */
+	if (qlcnic_sriov_vf_check(adapter)) {
+		adapter->ahw->hwmon_dev = NULL;
+		return;
+	}
+	hwmon_dev = hwmon_device_register_with_groups(dev, qlcnic_driver_name,
+						      adapter,
+						      qlcnic_hwmon_groups);
+	if (IS_ERR(hwmon_dev)) {
+		dev_err(dev, "Cannot register with hwmon, err=%ld\n",
+			PTR_ERR(hwmon_dev));
+		hwmon_dev = NULL;
+	}
+	adapter->ahw->hwmon_dev = hwmon_dev;
+}
+
+void qlcnic_unregister_hwmon_dev(struct qlcnic_adapter *adapter)
+{
+	struct device *hwmon_dev = adapter->ahw->hwmon_dev;
+	if (hwmon_dev) {
+		hwmon_device_unregister(hwmon_dev);
+		adapter->ahw->hwmon_dev = NULL;
+	}
+}
+#endif
+
 void qlcnic_create_sysfs_entries(struct qlcnic_adapter *adapter)
 {
 	struct device *dev = &adapter->pdev->dev;
diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_main.c b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
index 0a1d76acab81..b40050e03a56 100644
--- a/drivers/net/ethernet/qlogic/qlge/qlge_main.c
+++ b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
@@ -3595,7 +3595,7 @@ static int ql_request_irq(struct ql_adapter *qdev)
 	}
 	return status;
 err_irq:
-	netif_err(qdev, ifup, qdev->ndev, "Failed to get the interrupts!!!/n");
+	netif_err(qdev, ifup, qdev->ndev, "Failed to get the interrupts!!!\n");
 	ql_free_irq(qdev);
 	return status;
 }
@@ -4770,7 +4770,7 @@ static int qlge_probe(struct pci_dev *pdev,
 	ndev->irq = pdev->irq;
 
 	ndev->netdev_ops = &qlge_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &qlge_ethtool_ops);
+	ndev->ethtool_ops = &qlge_ethtool_ops;
 	ndev->watchdog_timeo = 10 * HZ;
 
 	err = register_netdev(ndev);
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index aa1c079f231d..be425ad5e824 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -7125,7 +7125,7 @@ rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	for (i = 0; i < ETH_ALEN; i++)
 		dev->dev_addr[i] = RTL_R8(MAC0 + i);
 
-	SET_ETHTOOL_OPS(dev, &rtl8169_ethtool_ops);
+	dev->ethtool_ops = &rtl8169_ethtool_ops;
 	dev->watchdog_timeo = RTL8169_TX_TIMEOUT;
 
 	netif_napi_add(dev, &tp->napi, rtl8169_poll, R8169_NAPI_WEIGHT);
diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 6a9509ccd33b..7622213beef1 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -307,6 +307,27 @@ static const u16 sh_eth_offset_fast_sh4[SH_ETH_MAX_REGISTER_OFFSET] = {
 };
 
 static const u16 sh_eth_offset_fast_sh3_sh2[SH_ETH_MAX_REGISTER_OFFSET] = {
+	[EDMR]		= 0x0000,
+	[EDTRR]		= 0x0004,
+	[EDRRR]		= 0x0008,
+	[TDLAR]		= 0x000c,
+	[RDLAR]		= 0x0010,
+	[EESR]		= 0x0014,
+	[EESIPR]	= 0x0018,
+	[TRSCER]	= 0x001c,
+	[RMFCR]		= 0x0020,
+	[TFTR]		= 0x0024,
+	[FDR]		= 0x0028,
+	[RMCR]		= 0x002c,
+	[EDOCR]		= 0x0030,
+	[FCFTR]		= 0x0034,
+	[RPADIR]	= 0x0038,
+	[TRIMD]		= 0x003c,
+	[RBWAR]		= 0x0040,
+	[RDFAR]		= 0x0044,
+	[TBRAR]		= 0x004c,
+	[TDFAR]		= 0x0050,
+
 	[ECMR]		= 0x0160,
 	[ECSR]		= 0x0164,
 	[ECSIPR]	= 0x0168,
@@ -546,7 +567,6 @@ static struct sh_eth_cpu_data sh7757_data = {
 	.register_type	= SH_ETH_REG_FAST_SH4,
 
 	.eesipr_value	= DMAC_M_RFRMER | DMAC_M_ECI | 0x003fffff,
-	.rmcr_value	= RMCR_RNC,
 
 	.tx_check	= EESR_FTC | EESR_CND | EESR_DLC | EESR_CD | EESR_RTO,
 	.eesr_err_check	= EESR_TWB | EESR_TABT | EESR_RABT | EESR_RFE |
@@ -624,7 +644,6 @@ static struct sh_eth_cpu_data sh7757_data_giga = {
 			  EESR_RFE | EESR_RDE | EESR_RFRMER | EESR_TFE |
 			  EESR_TDE | EESR_ECI,
 	.fdr_value	= 0x0000072f,
-	.rmcr_value	= RMCR_RNC,
 
 	.irq_flags	= IRQF_SHARED,
 	.apr		= 1,
@@ -752,7 +771,6 @@ static struct sh_eth_cpu_data r8a7740_data = {
 			  EESR_RFE | EESR_RDE | EESR_RFRMER | EESR_TFE |
 			  EESR_TDE | EESR_ECI,
 	.fdr_value	= 0x0000070f,
-	.rmcr_value	= RMCR_RNC,
 
 	.apr		= 1,
 	.mpr		= 1,
@@ -784,7 +802,6 @@ static struct sh_eth_cpu_data r7s72100_data = {
 			  EESR_RFE | EESR_RDE | EESR_RFRMER | EESR_TFE |
 			  EESR_TDE | EESR_ECI,
 	.fdr_value	= 0x0000070f,
-	.rmcr_value	= RMCR_RNC,
 
 	.no_psr		= 1,
 	.apr		= 1,
@@ -833,9 +850,6 @@ static void sh_eth_set_default_cpu_data(struct sh_eth_cpu_data *cd)
 	if (!cd->fdr_value)
 		cd->fdr_value = DEFAULT_FDR_INIT;
 
-	if (!cd->rmcr_value)
-		cd->rmcr_value = DEFAULT_RMCR_VALUE;
-
 	if (!cd->tx_check)
 		cd->tx_check = DEFAULT_TX_CHECK;
 
@@ -1287,8 +1301,8 @@ static int sh_eth_dev_init(struct net_device *ndev, bool start)
 	sh_eth_write(ndev, mdp->cd->fdr_value, FDR);
 	sh_eth_write(ndev, 0, TFTR);
 
-	/* Frame recv control */
-	sh_eth_write(ndev, mdp->cd->rmcr_value, RMCR);
+	/* Frame recv control (enable multiple-packets per rx irq) */
+	sh_eth_write(ndev, RMCR_RNC, RMCR);
 
 	sh_eth_write(ndev, DESC_I_RINT8 | DESC_I_RINT5 | DESC_I_TINT2, TRSCER);
 
@@ -1385,7 +1399,6 @@ static int sh_eth_rx(struct net_device *ndev, u32 intr_status, int *quota)
 	int entry = mdp->cur_rx % mdp->num_rx_ring;
 	int boguscnt = (mdp->dirty_rx + mdp->num_rx_ring) - mdp->cur_rx;
 	struct sk_buff *skb;
-	int exceeded = 0;
 	u16 pkt_len = 0;
 	u32 desc_status;
 
@@ -1397,10 +1410,9 @@ static int sh_eth_rx(struct net_device *ndev, u32 intr_status, int *quota)
 		if (--boguscnt < 0)
 			break;
 
-		if (*quota <= 0) {
-			exceeded = 1;
+		if (*quota <= 0)
 			break;
-		}
+
 		(*quota)--;
 
 		if (!(desc_status & RDFEND))
@@ -1448,7 +1460,6 @@ static int sh_eth_rx(struct net_device *ndev, u32 intr_status, int *quota)
 			ndev->stats.rx_packets++;
 			ndev->stats.rx_bytes += pkt_len;
 		}
-		rxdesc->status |= cpu_to_edmac(mdp, RD_RACT);
 		entry = (++mdp->cur_rx) % mdp->num_rx_ring;
 		rxdesc = &mdp->rx_ring[entry];
 	}
@@ -1494,7 +1505,7 @@ static int sh_eth_rx(struct net_device *ndev, u32 intr_status, int *quota)
 		sh_eth_write(ndev, EDRRR_R, EDRRR);
 	}
 
-	return exceeded;
+	return *quota <= 0;
 }
 
 static void sh_eth_rcv_snd_disable(struct net_device *ndev)
@@ -2627,8 +2638,8 @@ static int sh_mdio_init(struct sh_eth_private *mdp,
 		 pdev->name, pdev->id);
 
 	/* PHY IRQ */
-	mdp->mii_bus->irq = devm_kzalloc(dev, sizeof(int) * PHY_MAX_ADDR,
-					 GFP_KERNEL);
+	mdp->mii_bus->irq = devm_kmalloc_array(dev, PHY_MAX_ADDR, sizeof(int),
+					       GFP_KERNEL);
 	if (!mdp->mii_bus->irq) {
 		ret = -ENOMEM;
 		goto out_free_bus;
@@ -2843,7 +2854,7 @@ static int sh_eth_drv_probe(struct platform_device *pdev)
 		ndev->netdev_ops = &sh_eth_netdev_ops_tsu;
 	else
 		ndev->netdev_ops = &sh_eth_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &sh_eth_ethtool_ops);
+	ndev->ethtool_ops = &sh_eth_ethtool_ops;
 	ndev->watchdog_timeo = TX_TIMEOUT;
 
 	/* debug message level */
diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h
index d55e37cd5fec..b37c427144ee 100644
--- a/drivers/net/ethernet/renesas/sh_eth.h
+++ b/drivers/net/ethernet/renesas/sh_eth.h
@@ -319,7 +319,6 @@ enum TD_STS_BIT {
 enum RMCR_BIT {
 	RMCR_RNC = 0x00000001,
 };
-#define DEFAULT_RMCR_VALUE	0x00000000
 
 /* ECMR */
 enum FELIC_MODE_BIT {
@@ -466,7 +465,6 @@ struct sh_eth_cpu_data {
 	unsigned long fdr_value;
 	unsigned long fcftr_value;
 	unsigned long rpadir_value;
-	unsigned long rmcr_value;
 
 	/* interrupt checking mask */
 	unsigned long tx_check;
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c
index 0415fa50eeb7..c0981ae45874 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c
@@ -520,5 +520,5 @@ static const struct ethtool_ops sxgbe_ethtool_ops = {
 
 void sxgbe_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &sxgbe_ethtool_ops);
+	netdev->ethtool_ops = &sxgbe_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
index 82a9a983869f..698494481d18 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
@@ -425,8 +425,8 @@ dmamem_err:
  * @rx_rsize: ring size
  * Description:  this function initializes the DMA RX descriptor
  */
-void free_rx_ring(struct device *dev, struct sxgbe_rx_queue *rx_ring,
-		  int rx_rsize)
+static void free_rx_ring(struct device *dev, struct sxgbe_rx_queue *rx_ring,
+			 int rx_rsize)
 {
 	dma_free_coherent(dev, rx_rsize * sizeof(struct sxgbe_rx_norm_desc),
 			  rx_ring->dma_rx, rx_ring->dma_rx_phy);
@@ -519,8 +519,8 @@ error:
  * @tx_rsize: ring size
  * Description:  this function initializes the DMA TX descriptor
  */
-void free_tx_ring(struct device *dev, struct sxgbe_tx_queue *tx_ring,
-		  int tx_rsize)
+static void free_tx_ring(struct device *dev, struct sxgbe_tx_queue *tx_ring,
+			 int tx_rsize)
 {
 	dma_free_coherent(dev, tx_rsize * sizeof(struct sxgbe_tx_norm_desc),
 			  tx_ring->dma_tx, tx_ring->dma_tx_phy);
@@ -1221,11 +1221,10 @@ static int sxgbe_release(struct net_device *dev)
 
 	return 0;
 }
-
 /* Prepare first Tx descriptor for doing TSO operation */
-void sxgbe_tso_prepare(struct sxgbe_priv_data *priv,
-		       struct sxgbe_tx_norm_desc *first_desc,
-		       struct sk_buff *skb)
+static void sxgbe_tso_prepare(struct sxgbe_priv_data *priv,
+			      struct sxgbe_tx_norm_desc *first_desc,
+			      struct sk_buff *skb)
 {
 	unsigned int total_hdr_len, tcp_hdr_len;
 
@@ -1914,40 +1913,6 @@ static void sxgbe_set_rx_mode(struct net_device *dev)
 		   readl(ioaddr + SXGBE_HASH_LOW));
 }
 
-/**
- * sxgbe_config - entry point for changing configuration mode passed on by
- * ifconfig
- * @dev : pointer to the device structure
- * @map : pointer to the device mapping structure
- * Description:
- * This function is a driver entry point which gets called by the kernel
- * whenever some device configuration is changed.
- * Return value:
- * This function returns 0 if success and appropriate error otherwise.
- */
-static int sxgbe_config(struct net_device *dev, struct ifmap *map)
-{
-	struct sxgbe_priv_data *priv = netdev_priv(dev);
-
-	/* Can't act on a running interface */
-	if (dev->flags & IFF_UP)
-		return -EBUSY;
-
-	/* Don't allow changing the I/O address */
-	if (map->base_addr != (unsigned long)priv->ioaddr) {
-		netdev_warn(dev, "can't change I/O address\n");
-		return -EOPNOTSUPP;
-	}
-
-	/* Don't allow changing the IRQ */
-	if (map->irq != priv->irq) {
-		netdev_warn(dev, "not change IRQ number %d\n", priv->irq);
-		return -EOPNOTSUPP;
-	}
-
-	return 0;
-}
-
 #ifdef CONFIG_NET_POLL_CONTROLLER
 /**
  * sxgbe_poll_controller - entry point for polling receive by device
@@ -2009,7 +1974,6 @@ static const struct net_device_ops sxgbe_netdev_ops = {
 	.ndo_set_rx_mode	= sxgbe_set_rx_mode,
 	.ndo_tx_timeout		= sxgbe_tx_timeout,
 	.ndo_do_ioctl		= sxgbe_ioctl,
-	.ndo_set_config		= sxgbe_config,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= sxgbe_poll_controller,
 #endif
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_reg.h b/drivers/net/ethernet/samsung/sxgbe/sxgbe_reg.h
index 56f8bf5a3f1b..81437d91df99 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_reg.h
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_reg.h
@@ -188,7 +188,6 @@
 
 /* L3/L4 function registers */
 #define SXGBE_CORE_L34_ADDCTL_REG	0x0C00
-#define SXGBE_CORE_L34_ADDCTL_REG	0x0C00
 #define SXGBE_CORE_L34_DATA_REG		0x0C04
 
 /* ARP registers */
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 63d595fd3cc5..1e274045970f 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -2248,7 +2248,7 @@ static int efx_register_netdev(struct efx_nic *efx)
 	} else {
 		net_dev->netdev_ops = &efx_farch_netdev_ops;
 	}
-	SET_ETHTOOL_OPS(net_dev, &efx_ethtool_ops);
+	net_dev->ethtool_ops = &efx_ethtool_ops;
 	net_dev->gso_max_segs = EFX_TSO_MAX_SEGS;
 
 	rtnl_lock();
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 0de8b07c24c2..74739c4b9997 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -1033,7 +1033,7 @@ static u32 efx_ethtool_get_rxfh_indir_size(struct net_device *net_dev)
 		0 : ARRAY_SIZE(efx->rx_indir_table));
 }
 
-static int efx_ethtool_get_rxfh_indir(struct net_device *net_dev, u32 *indir)
+static int efx_ethtool_get_rxfh(struct net_device *net_dev, u32 *indir, u8 *key)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
 
@@ -1041,8 +1041,8 @@ static int efx_ethtool_get_rxfh_indir(struct net_device *net_dev, u32 *indir)
 	return 0;
 }
 
-static int efx_ethtool_set_rxfh_indir(struct net_device *net_dev,
-				      const u32 *indir)
+static int efx_ethtool_set_rxfh(struct net_device *net_dev,
+				const u32 *indir, const u8 *key)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
 
@@ -1125,8 +1125,8 @@ const struct ethtool_ops efx_ethtool_ops = {
 	.get_rxnfc		= efx_ethtool_get_rxnfc,
 	.set_rxnfc		= efx_ethtool_set_rxnfc,
 	.get_rxfh_indir_size	= efx_ethtool_get_rxfh_indir_size,
-	.get_rxfh_indir		= efx_ethtool_get_rxfh_indir,
-	.set_rxfh_indir		= efx_ethtool_set_rxfh_indir,
+	.get_rxfh		= efx_ethtool_get_rxfh,
+	.set_rxfh		= efx_ethtool_set_rxfh,
 	.get_ts_info		= efx_ethtool_get_ts_info,
 	.get_module_info	= efx_ethtool_get_module_info,
 	.get_module_eeprom	= efx_ethtool_get_module_eeprom,
diff --git a/drivers/net/ethernet/sfc/io.h b/drivers/net/ethernet/sfc/io.h
index 4d3f119b67b3..afb94aa2c15e 100644
--- a/drivers/net/ethernet/sfc/io.h
+++ b/drivers/net/ethernet/sfc/io.h
@@ -66,10 +66,17 @@
 #define EFX_USE_QWORD_IO 1
 #endif
 
+/* Hardware issue requires that only 64-bit naturally aligned writes
+ * are seen by hardware. Its not strictly necessary to restrict to
+ * x86_64 arch, but done for safety since unusual write combining behaviour
+ * can break PIO.
+ */
+#ifdef CONFIG_X86_64
 /* PIO is a win only if write-combining is possible */
 #ifdef ARCH_HAS_IOREMAP_WC
 #define EFX_USE_PIO 1
 #endif
+#endif
 
 #ifdef EFX_USE_QWORD_IO
 static inline void _efx_writeq(struct efx_nic *efx, __le64 value,
diff --git a/drivers/net/ethernet/sfc/siena_sriov.c b/drivers/net/ethernet/sfc/siena_sriov.c
index 9a9205e77896..43d2e64546ed 100644
--- a/drivers/net/ethernet/sfc/siena_sriov.c
+++ b/drivers/net/ethernet/sfc/siena_sriov.c
@@ -1633,7 +1633,8 @@ int efx_sriov_get_vf_config(struct net_device *net_dev, int vf_i,
 
 	ivi->vf = vf_i;
 	ether_addr_copy(ivi->mac, vf->addr.mac_addr);
-	ivi->tx_rate = 0;
+	ivi->max_tx_rate = 0;
+	ivi->min_tx_rate = 0;
 	tci = ntohs(vf->addr.tci);
 	ivi->vlan = tci & VLAN_VID_MASK;
 	ivi->qos = (tci >> VLAN_PRIO_SHIFT) & 0x7;
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index fa9475300411..ede8dcca0ff3 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -189,6 +189,18 @@ struct efx_short_copy_buffer {
 	u8 buf[L1_CACHE_BYTES];
 };
 
+/* Copy in explicit 64-bit writes. */
+static void efx_memcpy_64(void __iomem *dest, void *src, size_t len)
+{
+	u64 *src64 = src;
+	u64 __iomem *dest64 = dest;
+	size_t l64 = len / 8;
+	size_t i;
+
+	for (i = 0; i < l64; i++)
+		writeq(src64[i], &dest64[i]);
+}
+
 /* Copy to PIO, respecting that writes to PIO buffers must be dword aligned.
  * Advances piobuf pointer. Leaves additional data in the copy buffer.
  */
@@ -198,7 +210,7 @@ static void efx_memcpy_toio_aligned(struct efx_nic *efx, u8 __iomem **piobuf,
 {
 	int block_len = len & ~(sizeof(copy_buf->buf) - 1);
 
-	memcpy_toio(*piobuf, data, block_len);
+	efx_memcpy_64(*piobuf, data, block_len);
 	*piobuf += block_len;
 	len -= block_len;
 
@@ -230,7 +242,7 @@ static void efx_memcpy_toio_aligned_cb(struct efx_nic *efx, u8 __iomem **piobuf,
 		if (copy_buf->used < sizeof(copy_buf->buf))
 			return;
 
-		memcpy_toio(*piobuf, copy_buf->buf, sizeof(copy_buf->buf));
+		efx_memcpy_64(*piobuf, copy_buf->buf, sizeof(copy_buf->buf));
 		*piobuf += sizeof(copy_buf->buf);
 		data += copy_to_buf;
 		len -= copy_to_buf;
@@ -245,7 +257,7 @@ static void efx_flush_copy_buffer(struct efx_nic *efx, u8 __iomem *piobuf,
 {
 	/* if there's anything in it, write the whole buffer, including junk */
 	if (copy_buf->used)
-		memcpy_toio(piobuf, copy_buf->buf, sizeof(copy_buf->buf));
+		efx_memcpy_64(piobuf, copy_buf->buf, sizeof(copy_buf->buf));
 }
 
 /* Traverse skb structure and copy fragments in to PIO buffer.
@@ -304,8 +316,8 @@ efx_enqueue_skb_pio(struct efx_tx_queue *tx_queue, struct sk_buff *skb)
 		 */
 		BUILD_BUG_ON(L1_CACHE_BYTES >
 			     SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
-		memcpy_toio(tx_queue->piobuf, skb->data,
-			    ALIGN(skb->len, L1_CACHE_BYTES));
+		efx_memcpy_64(tx_queue->piobuf, skb->data,
+			      ALIGN(skb->len, L1_CACHE_BYTES));
 	}
 
 	EFX_POPULATE_QWORD_5(buffer->option,
diff --git a/drivers/net/ethernet/sis/sis190.c b/drivers/net/ethernet/sis/sis190.c
index acbbe48a519c..a86339903b9b 100644
--- a/drivers/net/ethernet/sis/sis190.c
+++ b/drivers/net/ethernet/sis/sis190.c
@@ -1877,7 +1877,7 @@ static int sis190_init_one(struct pci_dev *pdev,
 
 	dev->netdev_ops = &sis190_netdev_ops;
 
-	SET_ETHTOOL_OPS(dev, &sis190_ethtool_ops);
+	dev->ethtool_ops = &sis190_ethtool_ops;
 	dev->watchdog_timeo = SIS190_TX_TIMEOUT;
 
 	spin_lock_init(&tp->lock);
diff --git a/drivers/net/ethernet/smsc/smc91c92_cs.c b/drivers/net/ethernet/smsc/smc91c92_cs.c
index c7a4868571f9..6b33127ab352 100644
--- a/drivers/net/ethernet/smsc/smc91c92_cs.c
+++ b/drivers/net/ethernet/smsc/smc91c92_cs.c
@@ -318,7 +318,7 @@ static int smc91c92_probe(struct pcmcia_device *link)
 
     /* The SMC91c92-specific entries in the device structure. */
     dev->netdev_ops = &smc_netdev_ops;
-    SET_ETHTOOL_OPS(dev, &ethtool_ops);
+    dev->ethtool_ops = &ethtool_ops;
     dev->watchdog_timeo = TX_TIMEOUT;
 
     smc->mii_if.dev = dev;
diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c
index a0fc151da40d..5e13fa5524ae 100644
--- a/drivers/net/ethernet/smsc/smsc911x.c
+++ b/drivers/net/ethernet/smsc/smsc911x.c
@@ -2477,6 +2477,8 @@ static int smsc911x_drv_probe(struct platform_device *pdev)
 		goto out_disable_resources;
 	}
 
+	netif_carrier_off(dev);
+
 	retval = register_netdev(dev);
 	if (retval) {
 		SMSC_WARN(pdata, probe, "Error %i registering device", retval);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index c5f9cb85c8ef..c62e67f3c2f0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -322,9 +322,7 @@ static int stmmac_ethtool_getsettings(struct net_device *dev,
 		return -EBUSY;
 	}
 	cmd->transceiver = XCVR_INTERNAL;
-	spin_lock_irq(&priv->lock);
 	rc = phy_ethtool_gset(phy, cmd);
-	spin_unlock_irq(&priv->lock);
 	return rc;
 }
 
@@ -431,8 +429,6 @@ stmmac_get_pauseparam(struct net_device *netdev,
 	if (priv->pcs)	/* FIXME */
 		return;
 
-	spin_lock(&priv->lock);
-
 	pause->rx_pause = 0;
 	pause->tx_pause = 0;
 	pause->autoneg = priv->phydev->autoneg;
@@ -442,7 +438,6 @@ stmmac_get_pauseparam(struct net_device *netdev,
 	if (priv->flow_ctrl & FLOW_TX)
 		pause->tx_pause = 1;
 
-	spin_unlock(&priv->lock);
 }
 
 static int
@@ -457,8 +452,6 @@ stmmac_set_pauseparam(struct net_device *netdev,
 	if (priv->pcs)	/* FIXME */
 		return -EOPNOTSUPP;
 
-	spin_lock(&priv->lock);
-
 	if (pause->rx_pause)
 		new_pause |= FLOW_RX;
 	if (pause->tx_pause)
@@ -473,7 +466,6 @@ stmmac_set_pauseparam(struct net_device *netdev,
 	} else
 		priv->hw->mac->flow_ctrl(priv->ioaddr, phy->duplex,
 					 priv->flow_ctrl, priv->pause);
-	spin_unlock(&priv->lock);
 	return ret;
 }
 
@@ -784,5 +776,5 @@ static const struct ethtool_ops stmmac_ethtool_ops = {
 
 void stmmac_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &stmmac_ethtool_ops);
+	netdev->ethtool_ops = &stmmac_ethtool_ops;
 }
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 0f4841d2e8dc..057a1208e594 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1753,7 +1753,7 @@ static int stmmac_open(struct net_device *dev)
 	}
 
 	/* Request the IRQ lines */
-	if (priv->lpi_irq != -ENXIO) {
+	if (priv->lpi_irq > 0) {
 		ret = request_irq(priv->lpi_irq, stmmac_interrupt, IRQF_SHARED,
 				  dev->name, dev);
 		if (unlikely(ret < 0)) {
@@ -1813,7 +1813,7 @@ static int stmmac_release(struct net_device *dev)
 	free_irq(dev->irq, dev);
 	if (priv->wol_irq != dev->irq)
 		free_irq(priv->wol_irq, dev);
-	if (priv->lpi_irq != -ENXIO)
+	if (priv->lpi_irq > 0)
 		free_irq(priv->lpi_irq, dev);
 
 	/* Stop TX/RX DMA and clear the descriptors */
@@ -2212,27 +2212,6 @@ static void stmmac_tx_timeout(struct net_device *dev)
 	stmmac_tx_err(priv);
 }
 
-/* Configuration changes (passed on by ifconfig) */
-static int stmmac_config(struct net_device *dev, struct ifmap *map)
-{
-	if (dev->flags & IFF_UP)	/* can't act on a running interface */
-		return -EBUSY;
-
-	/* Don't allow changing the I/O address */
-	if (map->base_addr != dev->base_addr) {
-		pr_warn("%s: can't change I/O address\n", dev->name);
-		return -EOPNOTSUPP;
-	}
-
-	/* Don't allow changing the IRQ */
-	if (map->irq != dev->irq) {
-		pr_warn("%s: not change IRQ number %d\n", dev->name, dev->irq);
-		return -EOPNOTSUPP;
-	}
-
-	return 0;
-}
-
 /**
  *  stmmac_set_rx_mode - entry point for multicast addressing
  *  @dev : pointer to the device structure
@@ -2598,7 +2577,6 @@ static const struct net_device_ops stmmac_netdev_ops = {
 	.ndo_set_rx_mode = stmmac_set_rx_mode,
 	.ndo_tx_timeout = stmmac_tx_timeout,
 	.ndo_do_ioctl = stmmac_ioctl,
-	.ndo_set_config = stmmac_config,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller = stmmac_poll_controller,
 #endif
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index a468eb107823..a5b1e1b776fe 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -205,10 +205,13 @@ int stmmac_mdio_register(struct net_device *ndev)
 	if (new_bus == NULL)
 		return -ENOMEM;
 
-	if (mdio_bus_data->irqs)
+	if (mdio_bus_data->irqs) {
 		irqlist = mdio_bus_data->irqs;
-	else
+	} else {
+		for (addr = 0; addr < PHY_MAX_ADDR; addr++)
+			priv->mii_irq[addr] = PHY_POLL;
 		irqlist = priv->mii_irq;
+	}
 
 #ifdef CONFIG_OF
 	if (priv->device->of_node)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
index 46aef5108bea..ea7a65be1f9a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
@@ -237,10 +237,12 @@ static int stmmac_pltfr_probe(struct platform_device *pdev)
 
 	/* Get the MAC information */
 	priv->dev->irq = platform_get_irq_byname(pdev, "macirq");
-	if (priv->dev->irq == -ENXIO) {
-		pr_err("%s: ERROR: MAC IRQ configuration "
-		       "information not found\n", __func__);
-		return -ENXIO;
+	if (priv->dev->irq < 0) {
+		if (priv->dev->irq != -EPROBE_DEFER) {
+			netdev_err(priv->dev,
+				   "MAC IRQ configuration information not found\n");
+		}
+		return priv->dev->irq;
 	}
 
 	/*
@@ -252,10 +254,15 @@ static int stmmac_pltfr_probe(struct platform_device *pdev)
 	 * so the driver will continue to use the mac irq (ndev->irq)
 	 */
 	priv->wol_irq = platform_get_irq_byname(pdev, "eth_wake_irq");
-	if (priv->wol_irq == -ENXIO)
+	if (priv->wol_irq < 0) {
+		if (priv->wol_irq == -EPROBE_DEFER)
+			return -EPROBE_DEFER;
 		priv->wol_irq = priv->dev->irq;
+	}
 
 	priv->lpi_irq = platform_get_irq_byname(pdev, "eth_lpi");
+	if (priv->lpi_irq == -EPROBE_DEFER)
+		return -EPROBE_DEFER;
 
 	platform_set_drvdata(pdev, priv->dev);
 
diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c
index 2ead87759ab4..38da73a2a886 100644
--- a/drivers/net/ethernet/tehuti/tehuti.c
+++ b/drivers/net/ethernet/tehuti/tehuti.c
@@ -2413,7 +2413,7 @@ static void bdx_set_ethtool_ops(struct net_device *netdev)
 		.get_ethtool_stats = bdx_get_ethtool_stats,
 	};
 
-	SET_ETHTOOL_OPS(netdev, &bdx_ethtool_ops);
+	netdev->ethtool_ops = &bdx_ethtool_ops;
 }
 
 /**
diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c
index 73f74f369437..7399a52f7c26 100644
--- a/drivers/net/ethernet/ti/cpmac.c
+++ b/drivers/net/ethernet/ti/cpmac.c
@@ -313,19 +313,6 @@ static int mii_irqs[PHY_MAX_ADDR] = { PHY_POLL, };
 
 static struct mii_bus *cpmac_mii;
 
-static int cpmac_config(struct net_device *dev, struct ifmap *map)
-{
-	if (dev->flags & IFF_UP)
-		return -EBUSY;
-
-	/* Don't allow changing the I/O address */
-	if (map->base_addr != dev->base_addr)
-		return -EOPNOTSUPP;
-
-	/* ignore other fields */
-	return 0;
-}
-
 static void cpmac_set_multicast_list(struct net_device *dev)
 {
 	struct netdev_hw_addr *ha;
@@ -1100,7 +1087,6 @@ static const struct net_device_ops cpmac_netdev_ops = {
 	.ndo_tx_timeout		= cpmac_tx_timeout,
 	.ndo_set_rx_mode	= cpmac_set_multicast_list,
 	.ndo_do_ioctl		= cpmac_ioctl,
-	.ndo_set_config		= cpmac_config,
 	.ndo_change_mtu		= eth_change_mtu,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
diff --git a/drivers/net/ethernet/ti/cpsw-phy-sel.c b/drivers/net/ethernet/ti/cpsw-phy-sel.c
index 148da9ae8366..aa8bf45e53dc 100644
--- a/drivers/net/ethernet/ti/cpsw-phy-sel.c
+++ b/drivers/net/ethernet/ti/cpsw-phy-sel.c
@@ -29,6 +29,8 @@
 #define AM33XX_GMII_SEL_RMII2_IO_CLK_EN	BIT(7)
 #define AM33XX_GMII_SEL_RMII1_IO_CLK_EN	BIT(6)
 
+#define GMII_SEL_MODE_MASK		0x3
+
 struct cpsw_phy_sel_priv {
 	struct device	*dev;
 	u32 __iomem	*gmii_sel;
@@ -65,7 +67,7 @@ static void cpsw_gmii_sel_am3352(struct cpsw_phy_sel_priv *priv,
 		break;
 	};
 
-	mask = 0x3 << (slave * 2) | BIT(slave + 6);
+	mask = GMII_SEL_MODE_MASK << (slave * 2) | BIT(slave + 6);
 	mode <<= slave * 2;
 
 	if (priv->rmii_clock_external) {
@@ -81,6 +83,55 @@ static void cpsw_gmii_sel_am3352(struct cpsw_phy_sel_priv *priv,
 	writel(reg, priv->gmii_sel);
 }
 
+static void cpsw_gmii_sel_dra7xx(struct cpsw_phy_sel_priv *priv,
+				 phy_interface_t phy_mode, int slave)
+{
+	u32 reg;
+	u32 mask;
+	u32 mode = 0;
+
+	reg = readl(priv->gmii_sel);
+
+	switch (phy_mode) {
+	case PHY_INTERFACE_MODE_RMII:
+		mode = AM33XX_GMII_SEL_MODE_RMII;
+		break;
+
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		mode = AM33XX_GMII_SEL_MODE_RGMII;
+		break;
+
+	case PHY_INTERFACE_MODE_MII:
+	default:
+		mode = AM33XX_GMII_SEL_MODE_MII;
+		break;
+	};
+
+	switch (slave) {
+	case 0:
+		mask = GMII_SEL_MODE_MASK;
+		break;
+	case 1:
+		mask = GMII_SEL_MODE_MASK << 4;
+		mode <<= 4;
+		break;
+	default:
+		dev_err(priv->dev, "invalid slave number...\n");
+		return;
+	}
+
+	if (priv->rmii_clock_external)
+		dev_err(priv->dev, "RMII External clock is not supported\n");
+
+	reg &= ~mask;
+	reg |= mode;
+
+	writel(reg, priv->gmii_sel);
+}
+
 static struct platform_driver cpsw_phy_sel_driver;
 static int match(struct device *dev, void *data)
 {
@@ -112,6 +163,14 @@ static const struct of_device_id cpsw_phy_sel_id_table[] = {
 		.compatible	= "ti,am3352-cpsw-phy-sel",
 		.data		= &cpsw_gmii_sel_am3352,
 	},
+	{
+		.compatible	= "ti,dra7xx-cpsw-phy-sel",
+		.data		= &cpsw_gmii_sel_dra7xx,
+	},
+	{
+		.compatible	= "ti,am43xx-cpsw-phy-sel",
+		.data		= &cpsw_gmii_sel_am3352,
+	},
 	{}
 };
 MODULE_DEVICE_TABLE(of, cpsw_phy_sel_id_table);
@@ -132,6 +191,7 @@ static int cpsw_phy_sel_probe(struct platform_device *pdev)
 		return -ENOMEM;
 	}
 
+	priv->dev = &pdev->dev;
 	priv->cpsw_phy_sel = of_id->data;
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "gmii-sel");
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index c331b7ebc812..b988d16cd34e 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -143,13 +143,13 @@ do {								\
 		u32 i;		\
 		for (i = 0; i < priv->num_irqs; i++) \
 			enable_irq(priv->irqs_table[i]); \
-	} while (0);
+	} while (0)
 #define cpsw_disable_irq(priv)	\
 	do {			\
 		u32 i;		\
 		for (i = 0; i < priv->num_irqs; i++) \
 			disable_irq_nosync(priv->irqs_table[i]); \
-	} while (0);
+	} while (0)
 
 #define cpsw_slave_index(priv)				\
 		((priv->data.dual_emac) ? priv->emac_port :	\
@@ -248,20 +248,31 @@ struct cpsw_ss_regs {
 #define TS_131              (1<<11) /* Time Sync Dest IP Addr 131 enable */
 #define TS_130              (1<<10) /* Time Sync Dest IP Addr 130 enable */
 #define TS_129              (1<<9)  /* Time Sync Dest IP Addr 129 enable */
-#define TS_BIT8             (1<<8)  /* ts_ttl_nonzero? */
+#define TS_TTL_NONZERO      (1<<8)  /* Time Sync Time To Live Non-zero enable */
+#define TS_ANNEX_F_EN       (1<<6)  /* Time Sync Annex F enable */
 #define TS_ANNEX_D_EN       (1<<4)  /* Time Sync Annex D enable */
 #define TS_LTYPE2_EN        (1<<3)  /* Time Sync LTYPE 2 enable */
 #define TS_LTYPE1_EN        (1<<2)  /* Time Sync LTYPE 1 enable */
 #define TS_TX_EN            (1<<1)  /* Time Sync Transmit Enable */
 #define TS_RX_EN            (1<<0)  /* Time Sync Receive Enable */
 
-#define CTRL_TS_BITS \
-	(TS_320 | TS_319 | TS_132 | TS_131 | TS_130 | TS_129 | TS_BIT8 | \
-	 TS_ANNEX_D_EN | TS_LTYPE1_EN)
+#define CTRL_V2_TS_BITS \
+	(TS_320 | TS_319 | TS_132 | TS_131 | TS_130 | TS_129 |\
+	 TS_TTL_NONZERO  | TS_ANNEX_D_EN | TS_LTYPE1_EN)
+
+#define CTRL_V2_ALL_TS_MASK (CTRL_V2_TS_BITS | TS_TX_EN | TS_RX_EN)
+#define CTRL_V2_TX_TS_BITS  (CTRL_V2_TS_BITS | TS_TX_EN)
+#define CTRL_V2_RX_TS_BITS  (CTRL_V2_TS_BITS | TS_RX_EN)
+
+
+#define CTRL_V3_TS_BITS \
+	(TS_320 | TS_319 | TS_132 | TS_131 | TS_130 | TS_129 |\
+	 TS_TTL_NONZERO | TS_ANNEX_F_EN | TS_ANNEX_D_EN |\
+	 TS_LTYPE1_EN)
 
-#define CTRL_ALL_TS_MASK (CTRL_TS_BITS | TS_TX_EN | TS_RX_EN)
-#define CTRL_TX_TS_BITS  (CTRL_TS_BITS | TS_TX_EN)
-#define CTRL_RX_TS_BITS  (CTRL_TS_BITS | TS_RX_EN)
+#define CTRL_V3_ALL_TS_MASK (CTRL_V3_TS_BITS | TS_TX_EN | TS_RX_EN)
+#define CTRL_V3_TX_TS_BITS  (CTRL_V3_TS_BITS | TS_TX_EN)
+#define CTRL_V3_RX_TS_BITS  (CTRL_V3_TS_BITS | TS_RX_EN)
 
 /* Bit definitions for the CPSW2_TS_SEQ_MTYPE register */
 #define TS_SEQ_ID_OFFSET_SHIFT   (16)    /* Time Sync Sequence ID Offset */
@@ -1201,7 +1212,12 @@ static int cpsw_ndo_open(struct net_device *ndev)
 	for_each_slave(priv, cpsw_slave_open, priv);
 
 	/* Add default VLAN */
-	cpsw_add_default_vlan(priv);
+	if (!priv->data.dual_emac)
+		cpsw_add_default_vlan(priv);
+	else
+		cpsw_ale_add_vlan(priv->ale, priv->data.default_vlan,
+				  ALE_ALL_PORTS << priv->host_port,
+				  ALE_ALL_PORTS << priv->host_port, 0, 0);
 
 	if (!cpsw_common_res_usage_state(priv)) {
 		/* setup tx dma to fixed prio and zero offset */
@@ -1376,13 +1392,27 @@ static void cpsw_hwtstamp_v2(struct cpsw_priv *priv)
 		slave = &priv->slaves[priv->data.active_slave];
 
 	ctrl = slave_read(slave, CPSW2_CONTROL);
-	ctrl &= ~CTRL_ALL_TS_MASK;
+	switch (priv->version) {
+	case CPSW_VERSION_2:
+		ctrl &= ~CTRL_V2_ALL_TS_MASK;
 
-	if (priv->cpts->tx_enable)
-		ctrl |= CTRL_TX_TS_BITS;
+		if (priv->cpts->tx_enable)
+			ctrl |= CTRL_V2_TX_TS_BITS;
 
-	if (priv->cpts->rx_enable)
-		ctrl |= CTRL_RX_TS_BITS;
+		if (priv->cpts->rx_enable)
+			ctrl |= CTRL_V2_RX_TS_BITS;
+	break;
+	case CPSW_VERSION_3:
+	default:
+		ctrl &= ~CTRL_V3_ALL_TS_MASK;
+
+		if (priv->cpts->tx_enable)
+			ctrl |= CTRL_V3_TX_TS_BITS;
+
+		if (priv->cpts->rx_enable)
+			ctrl |= CTRL_V3_RX_TS_BITS;
+	break;
+	}
 
 	mtype = (30 << TS_SEQ_ID_OFFSET_SHIFT) | EVENT_MSG_BITS;
 
@@ -1398,7 +1428,8 @@ static int cpsw_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
 	struct hwtstamp_config cfg;
 
 	if (priv->version != CPSW_VERSION_1 &&
-	    priv->version != CPSW_VERSION_2)
+	    priv->version != CPSW_VERSION_2 &&
+	    priv->version != CPSW_VERSION_3)
 		return -EOPNOTSUPP;
 
 	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
@@ -1443,6 +1474,7 @@ static int cpsw_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
 		cpsw_hwtstamp_v1(priv);
 		break;
 	case CPSW_VERSION_2:
+	case CPSW_VERSION_3:
 		cpsw_hwtstamp_v2(priv);
 		break;
 	default:
@@ -1459,7 +1491,8 @@ static int cpsw_hwtstamp_get(struct net_device *dev, struct ifreq *ifr)
 	struct hwtstamp_config cfg;
 
 	if (priv->version != CPSW_VERSION_1 &&
-	    priv->version != CPSW_VERSION_2)
+	    priv->version != CPSW_VERSION_2 &&
+	    priv->version != CPSW_VERSION_3)
 		return -EOPNOTSUPP;
 
 	cfg.flags = 0;
@@ -1780,25 +1813,25 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 		return -EINVAL;
 
 	if (of_property_read_u32(node, "slaves", &prop)) {
-		pr_err("Missing slaves property in the DT.\n");
+		dev_err(&pdev->dev, "Missing slaves property in the DT.\n");
 		return -EINVAL;
 	}
 	data->slaves = prop;
 
 	if (of_property_read_u32(node, "active_slave", &prop)) {
-		pr_err("Missing active_slave property in the DT.\n");
+		dev_err(&pdev->dev, "Missing active_slave property in the DT.\n");
 		return -EINVAL;
 	}
 	data->active_slave = prop;
 
 	if (of_property_read_u32(node, "cpts_clock_mult", &prop)) {
-		pr_err("Missing cpts_clock_mult property in the DT.\n");
+		dev_err(&pdev->dev, "Missing cpts_clock_mult property in the DT.\n");
 		return -EINVAL;
 	}
 	data->cpts_clock_mult = prop;
 
 	if (of_property_read_u32(node, "cpts_clock_shift", &prop)) {
-		pr_err("Missing cpts_clock_shift property in the DT.\n");
+		dev_err(&pdev->dev, "Missing cpts_clock_shift property in the DT.\n");
 		return -EINVAL;
 	}
 	data->cpts_clock_shift = prop;
@@ -1810,31 +1843,31 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 		return -ENOMEM;
 
 	if (of_property_read_u32(node, "cpdma_channels", &prop)) {
-		pr_err("Missing cpdma_channels property in the DT.\n");
+		dev_err(&pdev->dev, "Missing cpdma_channels property in the DT.\n");
 		return -EINVAL;
 	}
 	data->channels = prop;
 
 	if (of_property_read_u32(node, "ale_entries", &prop)) {
-		pr_err("Missing ale_entries property in the DT.\n");
+		dev_err(&pdev->dev, "Missing ale_entries property in the DT.\n");
 		return -EINVAL;
 	}
 	data->ale_entries = prop;
 
 	if (of_property_read_u32(node, "bd_ram_size", &prop)) {
-		pr_err("Missing bd_ram_size property in the DT.\n");
+		dev_err(&pdev->dev, "Missing bd_ram_size property in the DT.\n");
 		return -EINVAL;
 	}
 	data->bd_ram_size = prop;
 
 	if (of_property_read_u32(node, "rx_descs", &prop)) {
-		pr_err("Missing rx_descs property in the DT.\n");
+		dev_err(&pdev->dev, "Missing rx_descs property in the DT.\n");
 		return -EINVAL;
 	}
 	data->rx_descs = prop;
 
 	if (of_property_read_u32(node, "mac_control", &prop)) {
-		pr_err("Missing mac_control property in the DT.\n");
+		dev_err(&pdev->dev, "Missing mac_control property in the DT.\n");
 		return -EINVAL;
 	}
 	data->mac_control = prop;
@@ -1848,7 +1881,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 	ret = of_platform_populate(node, NULL, NULL, &pdev->dev);
 	/* We do not want to force this, as in some cases may not have child */
 	if (ret)
-		pr_warn("Doesn't have any child node\n");
+		dev_warn(&pdev->dev, "Doesn't have any child node\n");
 
 	for_each_child_of_node(node, slave_node) {
 		struct cpsw_slave_data *slave_data = data->slave_data + i;
@@ -1865,7 +1898,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 
 		parp = of_get_property(slave_node, "phy_id", &lenp);
 		if ((parp == NULL) || (lenp != (sizeof(void *) * 2))) {
-			pr_err("Missing slave[%d] phy_id property\n", i);
+			dev_err(&pdev->dev, "Missing slave[%d] phy_id property\n", i);
 			return -EINVAL;
 		}
 		mdio_node = of_find_node_by_phandle(be32_to_cpup(parp));
@@ -1885,18 +1918,18 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 
 		slave_data->phy_if = of_get_phy_mode(slave_node);
 		if (slave_data->phy_if < 0) {
-			pr_err("Missing or malformed slave[%d] phy-mode property\n",
-			       i);
+			dev_err(&pdev->dev, "Missing or malformed slave[%d] phy-mode property\n",
+				i);
 			return slave_data->phy_if;
 		}
 
 		if (data->dual_emac) {
 			if (of_property_read_u32(slave_node, "dual_emac_res_vlan",
 						 &prop)) {
-				pr_err("Missing dual_emac_res_vlan in DT.\n");
+				dev_err(&pdev->dev, "Missing dual_emac_res_vlan in DT.\n");
 				slave_data->dual_emac_res_vlan = i+1;
-				pr_err("Using %d as Reserved VLAN for %d slave\n",
-				       slave_data->dual_emac_res_vlan, i);
+				dev_err(&pdev->dev, "Using %d as Reserved VLAN for %d slave\n",
+					slave_data->dual_emac_res_vlan, i);
 			} else {
 				slave_data->dual_emac_res_vlan = prop;
 			}
@@ -1920,7 +1953,7 @@ static int cpsw_probe_dual_emac(struct platform_device *pdev,
 
 	ndev = alloc_etherdev(sizeof(struct cpsw_priv));
 	if (!ndev) {
-		pr_err("cpsw: error allocating net_device\n");
+		dev_err(&pdev->dev, "cpsw: error allocating net_device\n");
 		return -ENOMEM;
 	}
 
@@ -1936,10 +1969,10 @@ static int cpsw_probe_dual_emac(struct platform_device *pdev,
 	if (is_valid_ether_addr(data->slave_data[1].mac_addr)) {
 		memcpy(priv_sl2->mac_addr, data->slave_data[1].mac_addr,
 			ETH_ALEN);
-		pr_info("cpsw: Detected MACID = %pM\n", priv_sl2->mac_addr);
+		dev_info(&pdev->dev, "cpsw: Detected MACID = %pM\n", priv_sl2->mac_addr);
 	} else {
 		random_ether_addr(priv_sl2->mac_addr);
-		pr_info("cpsw: Random MACID = %pM\n", priv_sl2->mac_addr);
+		dev_info(&pdev->dev, "cpsw: Random MACID = %pM\n", priv_sl2->mac_addr);
 	}
 	memcpy(ndev->dev_addr, priv_sl2->mac_addr, ETH_ALEN);
 
@@ -1970,14 +2003,14 @@ static int cpsw_probe_dual_emac(struct platform_device *pdev,
 	ndev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
 
 	ndev->netdev_ops = &cpsw_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &cpsw_ethtool_ops);
+	ndev->ethtool_ops = &cpsw_ethtool_ops;
 	netif_napi_add(ndev, &priv_sl2->napi, cpsw_poll, CPSW_POLL_WEIGHT);
 
 	/* register the network device */
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 	ret = register_netdev(ndev);
 	if (ret) {
-		pr_err("cpsw: error registering net device\n");
+		dev_err(&pdev->dev, "cpsw: error registering net device\n");
 		free_netdev(ndev);
 		ret = -ENODEV;
 	}
@@ -1999,7 +2032,7 @@ static int cpsw_probe(struct platform_device *pdev)
 
 	ndev = alloc_etherdev(sizeof(struct cpsw_priv));
 	if (!ndev) {
-		pr_err("error allocating net_device\n");
+		dev_err(&pdev->dev, "error allocating net_device\n");
 		return -ENOMEM;
 	}
 
@@ -2014,7 +2047,7 @@ static int cpsw_probe(struct platform_device *pdev)
 	priv->cpts = devm_kzalloc(&pdev->dev, sizeof(struct cpts), GFP_KERNEL);
 	priv->irq_enabled = true;
 	if (!priv->cpts) {
-		pr_err("error allocating cpts\n");
+		dev_err(&pdev->dev, "error allocating cpts\n");
 		goto clean_ndev_ret;
 	}
 
@@ -2027,7 +2060,7 @@ static int cpsw_probe(struct platform_device *pdev)
 	pinctrl_pm_select_default_state(&pdev->dev);
 
 	if (cpsw_probe_dt(&priv->data, pdev)) {
-		pr_err("cpsw: platform data missing\n");
+		dev_err(&pdev->dev, "cpsw: platform data missing\n");
 		ret = -ENODEV;
 		goto clean_runtime_disable_ret;
 	}
@@ -2035,10 +2068,10 @@ static int cpsw_probe(struct platform_device *pdev)
 
 	if (is_valid_ether_addr(data->slave_data[0].mac_addr)) {
 		memcpy(priv->mac_addr, data->slave_data[0].mac_addr, ETH_ALEN);
-		pr_info("Detected MACID = %pM\n", priv->mac_addr);
+		dev_info(&pdev->dev, "Detected MACID = %pM\n", priv->mac_addr);
 	} else {
 		eth_random_addr(priv->mac_addr);
-		pr_info("Random MACID = %pM\n", priv->mac_addr);
+		dev_info(&pdev->dev, "Random MACID = %pM\n", priv->mac_addr);
 	}
 
 	memcpy(ndev->dev_addr, priv->mac_addr, ETH_ALEN);
@@ -2199,7 +2232,7 @@ static int cpsw_probe(struct platform_device *pdev)
 	ndev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
 
 	ndev->netdev_ops = &cpsw_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &cpsw_ethtool_ops);
+	ndev->ethtool_ops = &cpsw_ethtool_ops;
 	netif_napi_add(ndev, &priv->napi, cpsw_poll, CPSW_POLL_WEIGHT);
 
 	/* register the network device */
diff --git a/drivers/net/ethernet/ti/cpts.c b/drivers/net/ethernet/ti/cpts.c
index 243513980b51..6b56f85951e5 100644
--- a/drivers/net/ethernet/ti/cpts.c
+++ b/drivers/net/ethernet/ti/cpts.c
@@ -236,13 +236,11 @@ static void cpts_overflow_check(struct work_struct *work)
 	schedule_delayed_work(&cpts->overflow_work, CPTS_OVERFLOW_PERIOD);
 }
 
-#define CPTS_REF_CLOCK_NAME "cpsw_cpts_rft_clk"
-
-static void cpts_clk_init(struct cpts *cpts)
+static void cpts_clk_init(struct device *dev, struct cpts *cpts)
 {
-	cpts->refclk = clk_get(NULL, CPTS_REF_CLOCK_NAME);
+	cpts->refclk = devm_clk_get(dev, "cpts");
 	if (IS_ERR(cpts->refclk)) {
-		pr_err("Failed to clk_get %s\n", CPTS_REF_CLOCK_NAME);
+		dev_err(dev, "Failed to get cpts refclk\n");
 		cpts->refclk = NULL;
 		return;
 	}
@@ -252,7 +250,6 @@ static void cpts_clk_init(struct cpts *cpts)
 static void cpts_clk_release(struct cpts *cpts)
 {
 	clk_disable(cpts->refclk);
-	clk_put(cpts->refclk);
 }
 
 static int cpts_match(struct sk_buff *skb, unsigned int ptp_class,
@@ -390,7 +387,7 @@ int cpts_register(struct device *dev, struct cpts *cpts,
 	for (i = 0; i < CPTS_MAX_EVENTS; i++)
 		list_add(&cpts->pool_data[i].list, &cpts->pool);
 
-	cpts_clk_init(cpts);
+	cpts_clk_init(dev, cpts);
 	cpts_write32(cpts, CPTS_EN, control);
 	cpts_write32(cpts, TS_PEND_EN, int_enable);
 
diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c
index 88ef27067bf2..4a000f6dd6fc 100644
--- a/drivers/net/ethernet/ti/davinci_cpdma.c
+++ b/drivers/net/ethernet/ti/davinci_cpdma.c
@@ -158,9 +158,9 @@ cpdma_desc_pool_create(struct device *dev, u32 phys, u32 hw_addr,
 	int bitmap_size;
 	struct cpdma_desc_pool *pool;
 
-	pool = kzalloc(sizeof(*pool), GFP_KERNEL);
+	pool = devm_kzalloc(dev, sizeof(*pool), GFP_KERNEL);
 	if (!pool)
-		return NULL;
+		goto fail;
 
 	spin_lock_init(&pool->lock);
 
@@ -170,7 +170,7 @@ cpdma_desc_pool_create(struct device *dev, u32 phys, u32 hw_addr,
 	pool->num_desc	= size / pool->desc_size;
 
 	bitmap_size  = (pool->num_desc / BITS_PER_LONG) * sizeof(long);
-	pool->bitmap = kzalloc(bitmap_size, GFP_KERNEL);
+	pool->bitmap = devm_kzalloc(dev, bitmap_size, GFP_KERNEL);
 	if (!pool->bitmap)
 		goto fail;
 
@@ -187,10 +187,7 @@ cpdma_desc_pool_create(struct device *dev, u32 phys, u32 hw_addr,
 
 	if (pool->iomap)
 		return pool;
-
 fail:
-	kfree(pool->bitmap);
-	kfree(pool);
 	return NULL;
 }
 
@@ -203,7 +200,6 @@ static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool)
 
 	spin_lock_irqsave(&pool->lock, flags);
 	WARN_ON(pool->used_desc);
-	kfree(pool->bitmap);
 	if (pool->cpumap) {
 		dma_free_coherent(pool->dev, pool->mem_size, pool->cpumap,
 				  pool->phys);
@@ -211,7 +207,6 @@ static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool)
 		iounmap(pool->iomap);
 	}
 	spin_unlock_irqrestore(&pool->lock, flags);
-	kfree(pool);
 }
 
 static inline dma_addr_t desc_phys(struct cpdma_desc_pool *pool,
@@ -276,7 +271,7 @@ struct cpdma_ctlr *cpdma_ctlr_create(struct cpdma_params *params)
 {
 	struct cpdma_ctlr *ctlr;
 
-	ctlr = kzalloc(sizeof(*ctlr), GFP_KERNEL);
+	ctlr = devm_kzalloc(params->dev, sizeof(*ctlr), GFP_KERNEL);
 	if (!ctlr)
 		return NULL;
 
@@ -290,10 +285,8 @@ struct cpdma_ctlr *cpdma_ctlr_create(struct cpdma_params *params)
 					    ctlr->params.desc_hw_addr,
 					    ctlr->params.desc_mem_size,
 					    ctlr->params.desc_align);
-	if (!ctlr->pool) {
-		kfree(ctlr);
+	if (!ctlr->pool)
 		return NULL;
-	}
 
 	if (WARN_ON(ctlr->num_chan > CPDMA_MAX_CHANNELS))
 		ctlr->num_chan = CPDMA_MAX_CHANNELS;
@@ -468,7 +461,6 @@ int cpdma_ctlr_destroy(struct cpdma_ctlr *ctlr)
 
 	cpdma_desc_pool_destroy(ctlr->pool);
 	spin_unlock_irqrestore(&ctlr->lock, flags);
-	kfree(ctlr);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(cpdma_ctlr_destroy);
@@ -507,21 +499,22 @@ struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num,
 				     cpdma_handler_fn handler)
 {
 	struct cpdma_chan *chan;
-	int ret, offset = (chan_num % CPDMA_MAX_CHANNELS) * 4;
+	int offset = (chan_num % CPDMA_MAX_CHANNELS) * 4;
 	unsigned long flags;
 
 	if (__chan_linear(chan_num) >= ctlr->num_chan)
 		return NULL;
 
-	ret = -ENOMEM;
-	chan = kzalloc(sizeof(*chan), GFP_KERNEL);
+	chan = devm_kzalloc(ctlr->dev, sizeof(*chan), GFP_KERNEL);
 	if (!chan)
-		goto err_chan_alloc;
+		return ERR_PTR(-ENOMEM);
 
 	spin_lock_irqsave(&ctlr->lock, flags);
-	ret = -EBUSY;
-	if (ctlr->channels[chan_num])
-		goto err_chan_busy;
+	if (ctlr->channels[chan_num]) {
+		spin_unlock_irqrestore(&ctlr->lock, flags);
+		devm_kfree(ctlr->dev, chan);
+		return ERR_PTR(-EBUSY);
+	}
 
 	chan->ctlr	= ctlr;
 	chan->state	= CPDMA_STATE_IDLE;
@@ -551,12 +544,6 @@ struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num,
 	ctlr->channels[chan_num] = chan;
 	spin_unlock_irqrestore(&ctlr->lock, flags);
 	return chan;
-
-err_chan_busy:
-	spin_unlock_irqrestore(&ctlr->lock, flags);
-	kfree(chan);
-err_chan_alloc:
-	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL_GPL(cpdma_chan_create);
 
diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c
index 8f0e69ce07ca..35a139e9a833 100644
--- a/drivers/net/ethernet/ti/davinci_emac.c
+++ b/drivers/net/ethernet/ti/davinci_emac.c
@@ -1567,7 +1567,6 @@ static int emac_dev_open(struct net_device *ndev)
 	while ((res = platform_get_resource(priv->pdev, IORESOURCE_IRQ,
 					    res_num))) {
 		for (irq_num = res->start; irq_num <= res->end; irq_num++) {
-			dev_err(emac_dev, "Request IRQ %d\n", irq_num);
 			if (request_irq(irq_num, emac_irq, 0, ndev->name,
 					ndev)) {
 				dev_err(emac_dev,
@@ -1865,7 +1864,6 @@ static int davinci_emac_probe(struct platform_device *pdev)
 	struct emac_priv *priv;
 	unsigned long hw_ram_addr;
 	struct emac_platform_data *pdata;
-	struct device *emac_dev;
 	struct cpdma_params dma_params;
 	struct clk *emac_clk;
 	unsigned long emac_bus_frequency;
@@ -1911,7 +1909,6 @@ static int davinci_emac_probe(struct platform_device *pdev)
 	priv->coal_intvl = 0;
 	priv->bus_freq_mhz = (u32)(emac_bus_frequency / 1000000);
 
-	emac_dev = &ndev->dev;
 	/* Get EMAC platform data */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	priv->emac_base_phys = res->start + pdata->ctrl_reg_offset;
@@ -1930,7 +1927,7 @@ static int davinci_emac_probe(struct platform_device *pdev)
 		hw_ram_addr = (u32 __force)res->start + pdata->ctrl_ram_offset;
 
 	memset(&dma_params, 0, sizeof(dma_params));
-	dma_params.dev			= emac_dev;
+	dma_params.dev			= &pdev->dev;
 	dma_params.dmaregs		= priv->emac_base;
 	dma_params.rxthresh		= priv->emac_base + 0x120;
 	dma_params.rxfree		= priv->emac_base + 0x140;
@@ -1980,7 +1977,7 @@ static int davinci_emac_probe(struct platform_device *pdev)
 	}
 
 	ndev->netdev_ops = &emac_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &ethtool_ops);
+	ndev->ethtool_ops = &ethtool_ops;
 	netif_napi_add(ndev, &priv->napi, emac_poll, EMAC_POLL_WEIGHT);
 
 	/* register the network device */
@@ -1994,7 +1991,7 @@ static int davinci_emac_probe(struct platform_device *pdev)
 
 
 	if (netif_msg_probe(priv)) {
-		dev_notice(emac_dev, "DaVinci EMAC Probe found device "\
+		dev_notice(&pdev->dev, "DaVinci EMAC Probe found device "
 			   "(regs: %p, irq: %d)\n",
 			   (void *)priv->emac_base_phys, ndev->irq);
 	}
diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c
index 0cca9dec5d82..735dc53d4b01 100644
--- a/drivers/net/ethernet/ti/davinci_mdio.c
+++ b/drivers/net/ethernet/ti/davinci_mdio.c
@@ -303,7 +303,7 @@ static int davinci_mdio_probe_dt(struct mdio_platform_data *data,
 		return -EINVAL;
 
 	if (of_property_read_u32(node, "bus_freq", &prop)) {
-		pr_err("Missing bus_freq property in the DT.\n");
+		dev_err(&pdev->dev, "Missing bus_freq property in the DT.\n");
 		return -EINVAL;
 	}
 	data->bus_freq = prop;
@@ -321,15 +321,14 @@ static int davinci_mdio_probe(struct platform_device *pdev)
 	struct phy_device *phy;
 	int ret, addr;
 
-	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 
-	data->bus = mdiobus_alloc();
+	data->bus = devm_mdiobus_alloc(dev);
 	if (!data->bus) {
 		dev_err(dev, "failed to alloc mii bus\n");
-		ret = -ENOMEM;
-		goto bail_out;
+		return -ENOMEM;
 	}
 
 	if (dev->of_node) {
@@ -349,12 +348,9 @@ static int davinci_mdio_probe(struct platform_device *pdev)
 	data->bus->parent	= dev;
 	data->bus->priv		= data;
 
-	/* Select default pin state */
-	pinctrl_pm_select_default_state(&pdev->dev);
-
 	pm_runtime_enable(&pdev->dev);
 	pm_runtime_get_sync(&pdev->dev);
-	data->clk = clk_get(&pdev->dev, "fck");
+	data->clk = devm_clk_get(dev, "fck");
 	if (IS_ERR(data->clk)) {
 		dev_err(dev, "failed to get device clock\n");
 		ret = PTR_ERR(data->clk);
@@ -367,24 +363,9 @@ static int davinci_mdio_probe(struct platform_device *pdev)
 	spin_lock_init(&data->lock);
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(dev, "could not find register map resource\n");
-		ret = -ENOENT;
-		goto bail_out;
-	}
-
-	res = devm_request_mem_region(dev, res->start, resource_size(res),
-					    dev_name(dev));
-	if (!res) {
-		dev_err(dev, "could not allocate register map resource\n");
-		ret = -ENXIO;
-		goto bail_out;
-	}
-
-	data->regs = devm_ioremap_nocache(dev, res->start, resource_size(res));
-	if (!data->regs) {
-		dev_err(dev, "could not map mdio registers\n");
-		ret = -ENOMEM;
+	data->regs = devm_ioremap_resource(dev, res);
+	if (IS_ERR(data->regs)) {
+		ret = PTR_ERR(data->regs);
 		goto bail_out;
 	}
 
@@ -406,16 +387,9 @@ static int davinci_mdio_probe(struct platform_device *pdev)
 	return 0;
 
 bail_out:
-	if (data->bus)
-		mdiobus_free(data->bus);
-
-	if (data->clk)
-		clk_put(data->clk);
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 
-	kfree(data);
-
 	return ret;
 }
 
@@ -423,18 +397,12 @@ static int davinci_mdio_remove(struct platform_device *pdev)
 {
 	struct davinci_mdio_data *data = platform_get_drvdata(pdev);
 
-	if (data->bus) {
+	if (data->bus)
 		mdiobus_unregister(data->bus);
-		mdiobus_free(data->bus);
-	}
 
-	if (data->clk)
-		clk_put(data->clk);
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 
-	kfree(data);
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/tile/tilegx.c b/drivers/net/ethernet/tile/tilegx.c
index 449011b0e007..4c70360967c2 100644
--- a/drivers/net/ethernet/tile/tilegx.c
+++ b/drivers/net/ethernet/tile/tilegx.c
@@ -2191,8 +2191,6 @@ static void tile_net_setup(struct net_device *dev)
 static void tile_net_dev_init(const char *name, const uint8_t *mac)
 {
 	int ret;
-	int i;
-	int nz_addr = 0;
 	struct net_device *dev;
 	struct tile_net_priv *priv;
 
@@ -2212,7 +2210,6 @@ static void tile_net_dev_init(const char *name, const uint8_t *mac)
 
 	/* Initialize "priv". */
 	priv = netdev_priv(dev);
-	memset(priv, 0, sizeof(*priv));
 	priv->dev = dev;
 	priv->channel = -1;
 	priv->loopify_channel = -1;
@@ -2223,15 +2220,10 @@ static void tile_net_dev_init(const char *name, const uint8_t *mac)
 	 * be done before the device is opened.  If the MAC is all zeroes,
 	 * we use a random address, since we're probably on the simulator.
 	 */
-	for (i = 0; i < 6; i++)
-		nz_addr |= mac[i];
-
-	if (nz_addr) {
-		memcpy(dev->dev_addr, mac, ETH_ALEN);
-		dev->addr_len = 6;
-	} else {
+	if (!is_zero_ether_addr(mac))
+		ether_addr_copy(dev->dev_addr, mac);
+	else
 		eth_hw_addr_random(dev);
-	}
 
 	/* Register the network device. */
 	ret = register_netdev(dev);
diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.c b/drivers/net/ethernet/toshiba/ps3_gelic_net.c
index d899d0072ae0..bb7992804664 100644
--- a/drivers/net/ethernet/toshiba/ps3_gelic_net.c
+++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.c
@@ -1561,7 +1561,7 @@ static struct gelic_card *gelic_alloc_card_net(struct net_device **netdev)
 	 * alloc netdev
 	 */
 	*netdev = alloc_etherdev(sizeof(struct gelic_port));
-	if (!netdev) {
+	if (!*netdev) {
 		kfree(card->unalign);
 		return NULL;
 	}
diff --git a/drivers/net/ethernet/via/Kconfig b/drivers/net/ethernet/via/Kconfig
index 8a049a2b4474..f66ddaee0c87 100644
--- a/drivers/net/ethernet/via/Kconfig
+++ b/drivers/net/ethernet/via/Kconfig
@@ -19,7 +19,7 @@ if NET_VENDOR_VIA
 
 config VIA_RHINE
 	tristate "VIA Rhine support"
-	depends on PCI
+	depends on (PCI || USE_OF)
 	select CRC32
 	select MII
 	---help---
diff --git a/drivers/net/ethernet/via/via-rhine.c b/drivers/net/ethernet/via/via-rhine.c
index f61dc2b72bb2..2d72f96a9e2c 100644
--- a/drivers/net/ethernet/via/via-rhine.c
+++ b/drivers/net/ethernet/via/via-rhine.c
@@ -94,6 +94,10 @@ static const int multicast_filter_limit = 32;
 #include <linux/ioport.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/of_irq.h>
+#include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
@@ -116,13 +120,6 @@ static const int multicast_filter_limit = 32;
 static const char version[] =
 	"v1.10-LK" DRV_VERSION " " DRV_RELDATE " Written by Donald Becker";
 
-/* This driver was written to use PCI memory space. Some early versions
-   of the Rhine may only work correctly with I/O space accesses. */
-#ifdef CONFIG_VIA_RHINE_MMIO
-#define USE_MMIO
-#else
-#endif
-
 MODULE_AUTHOR("Donald Becker <becker@scyld.com>");
 MODULE_DESCRIPTION("VIA Rhine PCI Fast Ethernet driver");
 MODULE_LICENSE("GPL");
@@ -260,6 +257,12 @@ enum rhine_quirks {
 	rq6patterns	= 0x0040,	/* 6 instead of 4 patterns for WOL */
 	rqStatusWBRace	= 0x0080,	/* Tx Status Writeback Error possible */
 	rqRhineI	= 0x0100,	/* See comment below */
+	rqIntPHY	= 0x0200,	/* Integrated PHY */
+	rqMgmt		= 0x0400,	/* Management adapter */
+	rqNeedEnMMIO	= 0x0800,	/* Whether the core needs to be
+					 * switched from PIO mode to MMIO
+					 * (only applies to PCI)
+					 */
 };
 /*
  * rqRhineI: VT86C100A (aka Rhine-I) uses different bits to enable
@@ -279,6 +282,15 @@ static DEFINE_PCI_DEVICE_TABLE(rhine_pci_tbl) = {
 };
 MODULE_DEVICE_TABLE(pci, rhine_pci_tbl);
 
+/* OpenFirmware identifiers for platform-bus devices
+ * The .data field is currently only used to store quirks
+ */
+static u32 vt8500_quirks = rqWOL | rqForceReset | rq6patterns;
+static struct of_device_id rhine_of_tbl[] = {
+	{ .compatible = "via,vt8500-rhine", .data = &vt8500_quirks },
+	{ }	/* terminate list */
+};
+MODULE_DEVICE_TABLE(of, rhine_of_tbl);
 
 /* Offsets to the device registers. */
 enum register_offsets {
@@ -338,13 +350,11 @@ enum bcr1_bits {
 	BCR1_MED1=0x80,		/* for VT6102 */
 };
 
-#ifdef USE_MMIO
 /* Registers we check that mmio and reg are the same. */
 static const int mmio_verify_registers[] = {
 	RxConfig, TxConfig, IntrEnable, ConfigA, ConfigB, ConfigC, ConfigD,
 	0
 };
-#endif
 
 /* Bits in the interrupt status/mask registers. */
 enum intr_status_bits {
@@ -446,7 +456,7 @@ struct rhine_private {
 	unsigned char *tx_bufs;
 	dma_addr_t tx_bufs_dma;
 
-	struct pci_dev *pdev;
+	int irq;
 	long pioaddr;
 	struct net_device *dev;
 	struct napi_struct napi;
@@ -649,20 +659,46 @@ static void rhine_chip_reset(struct net_device *dev)
 		   "failed" : "succeeded");
 }
 
-#ifdef USE_MMIO
 static void enable_mmio(long pioaddr, u32 quirks)
 {
 	int n;
-	if (quirks & rqRhineI) {
-		/* More recent docs say that this bit is reserved ... */
-		n = inb(pioaddr + ConfigA) | 0x20;
-		outb(n, pioaddr + ConfigA);
-	} else {
-		n = inb(pioaddr + ConfigD) | 0x80;
-		outb(n, pioaddr + ConfigD);
+
+	if (quirks & rqNeedEnMMIO) {
+		if (quirks & rqRhineI) {
+			/* More recent docs say that this bit is reserved */
+			n = inb(pioaddr + ConfigA) | 0x20;
+			outb(n, pioaddr + ConfigA);
+		} else {
+			n = inb(pioaddr + ConfigD) | 0x80;
+			outb(n, pioaddr + ConfigD);
+		}
 	}
 }
-#endif
+
+static inline int verify_mmio(struct device *hwdev,
+			      long pioaddr,
+			      void __iomem *ioaddr,
+			      u32 quirks)
+{
+	if (quirks & rqNeedEnMMIO) {
+		int i = 0;
+
+		/* Check that selected MMIO registers match the PIO ones */
+		while (mmio_verify_registers[i]) {
+			int reg = mmio_verify_registers[i++];
+			unsigned char a = inb(pioaddr+reg);
+			unsigned char b = readb(ioaddr+reg);
+
+			if (a != b) {
+				dev_err(hwdev,
+					"MMIO do not match PIO [%02x] (%02x != %02x)\n",
+					reg, a, b);
+				return -EIO;
+			}
+		}
+	}
+	return 0;
+}
 
 /*
  * Loads bytes 0x00-0x05, 0x6E-0x6F, 0x78-0x7B from EEPROM
@@ -682,14 +718,12 @@ static void rhine_reload_eeprom(long pioaddr, struct net_device *dev)
 	if (i > 512)
 		pr_info("%4d cycles used @ %s:%d\n", i, __func__, __LINE__);
 
-#ifdef USE_MMIO
 	/*
 	 * Reloading from EEPROM overwrites ConfigA-D, so we must re-enable
 	 * MMIO. If reloading EEPROM was done first this could be avoided, but
 	 * it is not known if that still works with the "win98-reboot" problem.
 	 */
 	enable_mmio(pioaddr, rp->quirks);
-#endif
 
 	/* Turn off EEPROM-controlled wake-up (magic packet) */
 	if (rp->quirks & rqWOL)
@@ -701,7 +735,7 @@ static void rhine_reload_eeprom(long pioaddr, struct net_device *dev)
 static void rhine_poll(struct net_device *dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
-	const int irq = rp->pdev->irq;
+	const int irq = rp->irq;
 
 	disable_irq(irq);
 	rhine_interrupt(irq, dev);
@@ -846,7 +880,8 @@ static void rhine_hw_init(struct net_device *dev, long pioaddr)
 		msleep(5);
 
 	/* Reload EEPROM controlled bytes cleared by soft reset */
-	rhine_reload_eeprom(pioaddr, dev);
+	if (dev_is_pci(dev->dev.parent))
+		rhine_reload_eeprom(pioaddr, dev);
 }
 
 static const struct net_device_ops rhine_netdev_ops = {
@@ -867,125 +902,37 @@ static const struct net_device_ops rhine_netdev_ops = {
 #endif
 };
 
-static int rhine_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
+static int rhine_init_one_common(struct device *hwdev, u32 quirks,
+				 long pioaddr, void __iomem *ioaddr, int irq)
 {
 	struct net_device *dev;
 	struct rhine_private *rp;
-	int i, rc;
-	u32 quirks;
-	long pioaddr;
-	long memaddr;
-	void __iomem *ioaddr;
-	int io_size, phy_id;
+	int i, rc, phy_id;
 	const char *name;
-#ifdef USE_MMIO
-	int bar = 1;
-#else
-	int bar = 0;
-#endif
-
-/* when built into the kernel, we only print version if device is found */
-#ifndef MODULE
-	pr_info_once("%s\n", version);
-#endif
-
-	io_size = 256;
-	phy_id = 0;
-	quirks = 0;
-	name = "Rhine";
-	if (pdev->revision < VTunknown0) {
-		quirks = rqRhineI;
-		io_size = 128;
-	}
-	else if (pdev->revision >= VT6102) {
-		quirks = rqWOL | rqForceReset;
-		if (pdev->revision < VT6105) {
-			name = "Rhine II";
-			quirks |= rqStatusWBRace;	/* Rhine-II exclusive */
-		}
-		else {
-			phy_id = 1;	/* Integrated PHY, phy_id fixed to 1 */
-			if (pdev->revision >= VT6105_B0)
-				quirks |= rq6patterns;
-			if (pdev->revision < VT6105M)
-				name = "Rhine III";
-			else
-				name = "Rhine III (Management Adapter)";
-		}
-	}
-
-	rc = pci_enable_device(pdev);
-	if (rc)
-		goto err_out;
 
 	/* this should always be supported */
-	rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+	rc = dma_set_mask(hwdev, DMA_BIT_MASK(32));
 	if (rc) {
-		dev_err(&pdev->dev,
-			"32-bit PCI DMA addresses not supported by the card!?\n");
-		goto err_out_pci_disable;
-	}
-
-	/* sanity check */
-	if ((pci_resource_len(pdev, 0) < io_size) ||
-	    (pci_resource_len(pdev, 1) < io_size)) {
-		rc = -EIO;
-		dev_err(&pdev->dev, "Insufficient PCI resources, aborting\n");
-		goto err_out_pci_disable;
+		dev_err(hwdev, "32-bit DMA addresses not supported by the card!?\n");
+		goto err_out;
 	}
 
-	pioaddr = pci_resource_start(pdev, 0);
-	memaddr = pci_resource_start(pdev, 1);
-
-	pci_set_master(pdev);
-
 	dev = alloc_etherdev(sizeof(struct rhine_private));
 	if (!dev) {
 		rc = -ENOMEM;
-		goto err_out_pci_disable;
+		goto err_out;
 	}
-	SET_NETDEV_DEV(dev, &pdev->dev);
+	SET_NETDEV_DEV(dev, hwdev);
 
 	rp = netdev_priv(dev);
 	rp->dev = dev;
 	rp->quirks = quirks;
 	rp->pioaddr = pioaddr;
-	rp->pdev = pdev;
+	rp->base = ioaddr;
+	rp->irq = irq;
 	rp->msg_enable = netif_msg_init(debug, RHINE_MSG_DEFAULT);
 
-	rc = pci_request_regions(pdev, DRV_NAME);
-	if (rc)
-		goto err_out_free_netdev;
-
-	ioaddr = pci_iomap(pdev, bar, io_size);
-	if (!ioaddr) {
-		rc = -EIO;
-		dev_err(&pdev->dev,
-			"ioremap failed for device %s, region 0x%X @ 0x%lX\n",
-			pci_name(pdev), io_size, memaddr);
-		goto err_out_free_res;
-	}
-
-#ifdef USE_MMIO
-	enable_mmio(pioaddr, quirks);
-
-	/* Check that selected MMIO registers match the PIO ones */
-	i = 0;
-	while (mmio_verify_registers[i]) {
-		int reg = mmio_verify_registers[i++];
-		unsigned char a = inb(pioaddr+reg);
-		unsigned char b = readb(ioaddr+reg);
-		if (a != b) {
-			rc = -EIO;
-			dev_err(&pdev->dev,
-				"MMIO do not match PIO [%02x] (%02x != %02x)\n",
-				reg, a, b);
-			goto err_out_unmap;
-		}
-	}
-#endif /* USE_MMIO */
-
-	rp->base = ioaddr;
+	phy_id = rp->quirks & rqIntPHY ? 1 : 0;
 
 	u64_stats_init(&rp->tx_stats.syncp);
 	u64_stats_init(&rp->rx_stats.syncp);
@@ -1030,7 +977,7 @@ static int rhine_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rp->quirks & rqRhineI)
 		dev->features |= NETIF_F_SG|NETIF_F_HW_CSUM;
 
-	if (pdev->revision >= VT6105M)
+	if (rp->quirks & rqMgmt)
 		dev->features |= NETIF_F_HW_VLAN_CTAG_TX |
 				 NETIF_F_HW_VLAN_CTAG_RX |
 				 NETIF_F_HW_VLAN_CTAG_FILTER;
@@ -1038,18 +985,21 @@ static int rhine_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* dev->name not defined before register_netdev()! */
 	rc = register_netdev(dev);
 	if (rc)
-		goto err_out_unmap;
+		goto err_out_free_netdev;
+
+	if (rp->quirks & rqRhineI)
+		name = "Rhine";
+	else if (rp->quirks & rqStatusWBRace)
+		name = "Rhine II";
+	else if (rp->quirks & rqMgmt)
+		name = "Rhine III (Management Adapter)";
+	else
+		name = "Rhine III";
 
 	netdev_info(dev, "VIA %s at 0x%lx, %pM, IRQ %d\n",
-		    name,
-#ifdef USE_MMIO
-		    memaddr,
-#else
-		    (long)ioaddr,
-#endif
-		    dev->dev_addr, pdev->irq);
+		    name, (long)ioaddr, dev->dev_addr, rp->irq);
 
-	pci_set_drvdata(pdev, dev);
+	dev_set_drvdata(hwdev, dev);
 
 	{
 		u16 mii_cmd;
@@ -1078,41 +1028,158 @@ static int rhine_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	return 0;
 
+err_out_free_netdev:
+	free_netdev(dev);
+err_out:
+	return rc;
+}
+
+static int rhine_init_one_pci(struct pci_dev *pdev,
+			      const struct pci_device_id *ent)
+{
+	struct device *hwdev = &pdev->dev;
+	int rc;
+	long pioaddr, memaddr;
+	void __iomem *ioaddr;
+	int io_size = pdev->revision < VTunknown0 ? 128 : 256;
+
+/* This driver was written to use PCI memory space. Some early versions
+ * of the Rhine may only work correctly with I/O space accesses.
+ * TODO: determine for which revisions this is true and assign the flag
+ *	 in code as opposed to this Kconfig option (???)
+ */
+#ifdef CONFIG_VIA_RHINE_MMIO
+	u32 quirks = rqNeedEnMMIO;
+#else
+	u32 quirks = 0;
+#endif
+
+/* when built into the kernel, we only print version if device is found */
+#ifndef MODULE
+	pr_info_once("%s\n", version);
+#endif
+
+	rc = pci_enable_device(pdev);
+	if (rc)
+		goto err_out;
+
+	if (pdev->revision < VTunknown0) {
+		quirks |= rqRhineI;
+	} else if (pdev->revision >= VT6102) {
+		quirks |= rqWOL | rqForceReset;
+		if (pdev->revision < VT6105) {
+			quirks |= rqStatusWBRace;
+		} else {
+			quirks |= rqIntPHY;
+			if (pdev->revision >= VT6105_B0)
+				quirks |= rq6patterns;
+			if (pdev->revision >= VT6105M)
+				quirks |= rqMgmt;
+		}
+	}
+
+	/* sanity check */
+	if ((pci_resource_len(pdev, 0) < io_size) ||
+	    (pci_resource_len(pdev, 1) < io_size)) {
+		rc = -EIO;
+		dev_err(hwdev, "Insufficient PCI resources, aborting\n");
+		goto err_out_pci_disable;
+	}
+
+	pioaddr = pci_resource_start(pdev, 0);
+	memaddr = pci_resource_start(pdev, 1);
+
+	pci_set_master(pdev);
+
+	rc = pci_request_regions(pdev, DRV_NAME);
+	if (rc)
+		goto err_out_pci_disable;
+
+	ioaddr = pci_iomap(pdev, (quirks & rqNeedEnMMIO ? 1 : 0), io_size);
+	if (!ioaddr) {
+		rc = -EIO;
+		dev_err(hwdev,
+			"ioremap failed for device %s, region 0x%X @ 0x%lX\n",
+			dev_name(hwdev), io_size, memaddr);
+		goto err_out_free_res;
+	}
+
+	enable_mmio(pioaddr, quirks);
+
+	rc = verify_mmio(hwdev, pioaddr, ioaddr, quirks);
+	if (rc)
+		goto err_out_unmap;
+
+	rc = rhine_init_one_common(&pdev->dev, quirks,
+				   pioaddr, ioaddr, pdev->irq);
+	if (!rc)
+		return 0;
+
 err_out_unmap:
 	pci_iounmap(pdev, ioaddr);
 err_out_free_res:
 	pci_release_regions(pdev);
-err_out_free_netdev:
-	free_netdev(dev);
 err_out_pci_disable:
 	pci_disable_device(pdev);
 err_out:
 	return rc;
 }
 
+static int rhine_init_one_platform(struct platform_device *pdev)
+{
+	const struct of_device_id *match;
+	const u32 *quirks;
+	int irq;
+	struct resource *res;
+	void __iomem *ioaddr;
+
+	match = of_match_device(rhine_of_tbl, &pdev->dev);
+	if (!match)
+		return -EINVAL;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	ioaddr = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(ioaddr))
+		return PTR_ERR(ioaddr);
+
+	irq = irq_of_parse_and_map(pdev->dev.of_node, 0);
+	if (!irq)
+		return -EINVAL;
+
+	quirks = match->data;
+	if (!quirks)
+		return -EINVAL;
+
+	return rhine_init_one_common(&pdev->dev, *quirks,
+				     (long)ioaddr, ioaddr, irq);
+}
+
 static int alloc_ring(struct net_device* dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	void *ring;
 	dma_addr_t ring_dma;
 
-	ring = pci_alloc_consistent(rp->pdev,
-				    RX_RING_SIZE * sizeof(struct rx_desc) +
-				    TX_RING_SIZE * sizeof(struct tx_desc),
-				    &ring_dma);
+	ring = dma_alloc_coherent(hwdev,
+				  RX_RING_SIZE * sizeof(struct rx_desc) +
+				  TX_RING_SIZE * sizeof(struct tx_desc),
+				  &ring_dma,
+				  GFP_ATOMIC);
 	if (!ring) {
 		netdev_err(dev, "Could not allocate DMA memory\n");
 		return -ENOMEM;
 	}
 	if (rp->quirks & rqRhineI) {
-		rp->tx_bufs = pci_alloc_consistent(rp->pdev,
-						   PKT_BUF_SZ * TX_RING_SIZE,
-						   &rp->tx_bufs_dma);
+		rp->tx_bufs = dma_alloc_coherent(hwdev,
+						 PKT_BUF_SZ * TX_RING_SIZE,
+						 &rp->tx_bufs_dma,
+						 GFP_ATOMIC);
 		if (rp->tx_bufs == NULL) {
-			pci_free_consistent(rp->pdev,
-				    RX_RING_SIZE * sizeof(struct rx_desc) +
-				    TX_RING_SIZE * sizeof(struct tx_desc),
-				    ring, ring_dma);
+			dma_free_coherent(hwdev,
+					  RX_RING_SIZE * sizeof(struct rx_desc) +
+					  TX_RING_SIZE * sizeof(struct tx_desc),
+					  ring, ring_dma);
 			return -ENOMEM;
 		}
 	}
@@ -1128,16 +1195,17 @@ static int alloc_ring(struct net_device* dev)
 static void free_ring(struct net_device* dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 
-	pci_free_consistent(rp->pdev,
-			    RX_RING_SIZE * sizeof(struct rx_desc) +
-			    TX_RING_SIZE * sizeof(struct tx_desc),
-			    rp->rx_ring, rp->rx_ring_dma);
+	dma_free_coherent(hwdev,
+			  RX_RING_SIZE * sizeof(struct rx_desc) +
+			  TX_RING_SIZE * sizeof(struct tx_desc),
+			  rp->rx_ring, rp->rx_ring_dma);
 	rp->tx_ring = NULL;
 
 	if (rp->tx_bufs)
-		pci_free_consistent(rp->pdev, PKT_BUF_SZ * TX_RING_SIZE,
-				    rp->tx_bufs, rp->tx_bufs_dma);
+		dma_free_coherent(hwdev, PKT_BUF_SZ * TX_RING_SIZE,
+				  rp->tx_bufs, rp->tx_bufs_dma);
 
 	rp->tx_bufs = NULL;
 
@@ -1146,6 +1214,7 @@ static void free_ring(struct net_device* dev)
 static void alloc_rbufs(struct net_device *dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	dma_addr_t next;
 	int i;
 
@@ -1174,9 +1243,9 @@ static void alloc_rbufs(struct net_device *dev)
 			break;
 
 		rp->rx_skbuff_dma[i] =
-			pci_map_single(rp->pdev, skb->data, rp->rx_buf_sz,
-				       PCI_DMA_FROMDEVICE);
-		if (dma_mapping_error(&rp->pdev->dev, rp->rx_skbuff_dma[i])) {
+			dma_map_single(hwdev, skb->data, rp->rx_buf_sz,
+				       DMA_FROM_DEVICE);
+		if (dma_mapping_error(hwdev, rp->rx_skbuff_dma[i])) {
 			rp->rx_skbuff_dma[i] = 0;
 			dev_kfree_skb(skb);
 			break;
@@ -1190,6 +1259,7 @@ static void alloc_rbufs(struct net_device *dev)
 static void free_rbufs(struct net_device* dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	int i;
 
 	/* Free all the skbuffs in the Rx queue. */
@@ -1197,9 +1267,9 @@ static void free_rbufs(struct net_device* dev)
 		rp->rx_ring[i].rx_status = 0;
 		rp->rx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */
 		if (rp->rx_skbuff[i]) {
-			pci_unmap_single(rp->pdev,
+			dma_unmap_single(hwdev,
 					 rp->rx_skbuff_dma[i],
-					 rp->rx_buf_sz, PCI_DMA_FROMDEVICE);
+					 rp->rx_buf_sz, DMA_FROM_DEVICE);
 			dev_kfree_skb(rp->rx_skbuff[i]);
 		}
 		rp->rx_skbuff[i] = NULL;
@@ -1230,6 +1300,7 @@ static void alloc_tbufs(struct net_device* dev)
 static void free_tbufs(struct net_device* dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	int i;
 
 	for (i = 0; i < TX_RING_SIZE; i++) {
@@ -1238,10 +1309,10 @@ static void free_tbufs(struct net_device* dev)
 		rp->tx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */
 		if (rp->tx_skbuff[i]) {
 			if (rp->tx_skbuff_dma[i]) {
-				pci_unmap_single(rp->pdev,
+				dma_unmap_single(hwdev,
 						 rp->tx_skbuff_dma[i],
 						 rp->tx_skbuff[i]->len,
-						 PCI_DMA_TODEVICE);
+						 DMA_TO_DEVICE);
 			}
 			dev_kfree_skb(rp->tx_skbuff[i]);
 		}
@@ -1278,8 +1349,9 @@ static void rhine_set_carrier(struct mii_if_info *mii)
 		/* autoneg is off: Link is always assumed to be up */
 		if (!netif_carrier_ok(dev))
 			netif_carrier_on(dev);
-	} else	/* Let MMI library update carrier status */
-		rhine_check_media(dev, 0);
+	}
+
+	rhine_check_media(dev, 0);
 
 	netif_info(rp, link, dev, "force_media %d, carrier %d\n",
 		   mii->force_media, netif_carrier_ok(dev));
@@ -1469,7 +1541,7 @@ static void init_registers(struct net_device *dev)
 
 	rhine_set_rx_mode(dev);
 
-	if (rp->pdev->revision >= VT6105M)
+	if (rp->quirks & rqMgmt)
 		rhine_init_cam_filter(dev);
 
 	napi_enable(&rp->napi);
@@ -1581,16 +1653,15 @@ static int rhine_open(struct net_device *dev)
 	void __iomem *ioaddr = rp->base;
 	int rc;
 
-	rc = request_irq(rp->pdev->irq, rhine_interrupt, IRQF_SHARED, dev->name,
-			dev);
+	rc = request_irq(rp->irq, rhine_interrupt, IRQF_SHARED, dev->name, dev);
 	if (rc)
 		return rc;
 
-	netif_dbg(rp, ifup, dev, "%s() irq %d\n", __func__, rp->pdev->irq);
+	netif_dbg(rp, ifup, dev, "%s() irq %d\n", __func__, rp->irq);
 
 	rc = alloc_ring(dev);
 	if (rc) {
-		free_irq(rp->pdev->irq, dev);
+		free_irq(rp->irq, dev);
 		return rc;
 	}
 	alloc_rbufs(dev);
@@ -1659,6 +1730,7 @@ static netdev_tx_t rhine_start_tx(struct sk_buff *skb,
 				  struct net_device *dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	void __iomem *ioaddr = rp->base;
 	unsigned entry;
 
@@ -1695,9 +1767,9 @@ static netdev_tx_t rhine_start_tx(struct sk_buff *skb,
 						       rp->tx_bufs));
 	} else {
 		rp->tx_skbuff_dma[entry] =
-			pci_map_single(rp->pdev, skb->data, skb->len,
-				       PCI_DMA_TODEVICE);
-		if (dma_mapping_error(&rp->pdev->dev, rp->tx_skbuff_dma[entry])) {
+			dma_map_single(hwdev, skb->data, skb->len,
+				       DMA_TO_DEVICE);
+		if (dma_mapping_error(hwdev, rp->tx_skbuff_dma[entry])) {
 			dev_kfree_skb_any(skb);
 			rp->tx_skbuff_dma[entry] = 0;
 			dev->stats.tx_dropped++;
@@ -1788,6 +1860,7 @@ static irqreturn_t rhine_interrupt(int irq, void *dev_instance)
 static void rhine_tx(struct net_device *dev)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	int txstatus = 0, entry = rp->dirty_tx % TX_RING_SIZE;
 
 	/* find and cleanup dirty tx descriptors */
@@ -1831,10 +1904,10 @@ static void rhine_tx(struct net_device *dev)
 		}
 		/* Free the original skb. */
 		if (rp->tx_skbuff_dma[entry]) {
-			pci_unmap_single(rp->pdev,
+			dma_unmap_single(hwdev,
 					 rp->tx_skbuff_dma[entry],
 					 rp->tx_skbuff[entry]->len,
-					 PCI_DMA_TODEVICE);
+					 DMA_TO_DEVICE);
 		}
 		dev_consume_skb_any(rp->tx_skbuff[entry]);
 		rp->tx_skbuff[entry] = NULL;
@@ -1863,6 +1936,7 @@ static inline u16 rhine_get_vlan_tci(struct sk_buff *skb, int data_size)
 static int rhine_rx(struct net_device *dev, int limit)
 {
 	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 	int count;
 	int entry = rp->cur_rx % RX_RING_SIZE;
 
@@ -1924,19 +1998,19 @@ static int rhine_rx(struct net_device *dev, int limit)
 			if (pkt_len < rx_copybreak)
 				skb = netdev_alloc_skb_ip_align(dev, pkt_len);
 			if (skb) {
-				pci_dma_sync_single_for_cpu(rp->pdev,
-							    rp->rx_skbuff_dma[entry],
-							    rp->rx_buf_sz,
-							    PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_cpu(hwdev,
+							rp->rx_skbuff_dma[entry],
+							rp->rx_buf_sz,
+							DMA_FROM_DEVICE);
 
 				skb_copy_to_linear_data(skb,
 						 rp->rx_skbuff[entry]->data,
 						 pkt_len);
 				skb_put(skb, pkt_len);
-				pci_dma_sync_single_for_device(rp->pdev,
-							       rp->rx_skbuff_dma[entry],
-							       rp->rx_buf_sz,
-							       PCI_DMA_FROMDEVICE);
+				dma_sync_single_for_device(hwdev,
+							   rp->rx_skbuff_dma[entry],
+							   rp->rx_buf_sz,
+							   DMA_FROM_DEVICE);
 			} else {
 				skb = rp->rx_skbuff[entry];
 				if (skb == NULL) {
@@ -1945,10 +2019,10 @@ static int rhine_rx(struct net_device *dev, int limit)
 				}
 				rp->rx_skbuff[entry] = NULL;
 				skb_put(skb, pkt_len);
-				pci_unmap_single(rp->pdev,
+				dma_unmap_single(hwdev,
 						 rp->rx_skbuff_dma[entry],
 						 rp->rx_buf_sz,
-						 PCI_DMA_FROMDEVICE);
+						 DMA_FROM_DEVICE);
 			}
 
 			if (unlikely(desc_length & DescTag))
@@ -1979,10 +2053,11 @@ static int rhine_rx(struct net_device *dev, int limit)
 			if (skb == NULL)
 				break;	/* Better luck next round. */
 			rp->rx_skbuff_dma[entry] =
-				pci_map_single(rp->pdev, skb->data,
+				dma_map_single(hwdev, skb->data,
 					       rp->rx_buf_sz,
-					       PCI_DMA_FROMDEVICE);
-			if (dma_mapping_error(&rp->pdev->dev, rp->rx_skbuff_dma[entry])) {
+					       DMA_FROM_DEVICE);
+			if (dma_mapping_error(hwdev,
+					      rp->rx_skbuff_dma[entry])) {
 				dev_kfree_skb(skb);
 				rp->rx_skbuff_dma[entry] = 0;
 				break;
@@ -2103,7 +2178,7 @@ static void rhine_set_rx_mode(struct net_device *dev)
 		/* Too many to match, or accept all multicasts. */
 		iowrite32(0xffffffff, ioaddr + MulticastFilter0);
 		iowrite32(0xffffffff, ioaddr + MulticastFilter1);
-	} else if (rp->pdev->revision >= VT6105M) {
+	} else if (rp->quirks & rqMgmt) {
 		int i = 0;
 		u32 mCAMmask = 0;	/* 32 mCAMs (6105M and better) */
 		netdev_for_each_mc_addr(ha, dev) {
@@ -2125,7 +2200,7 @@ static void rhine_set_rx_mode(struct net_device *dev)
 		iowrite32(mc_filter[1], ioaddr + MulticastFilter1);
 	}
 	/* enable/disable VLAN receive filtering */
-	if (rp->pdev->revision >= VT6105M) {
+	if (rp->quirks & rqMgmt) {
 		if (dev->flags & IFF_PROMISC)
 			BYTE_REG_BITS_OFF(BCR1_VIDFR, ioaddr + PCIBusConfig1);
 		else
@@ -2136,11 +2211,11 @@ static void rhine_set_rx_mode(struct net_device *dev)
 
 static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	struct rhine_private *rp = netdev_priv(dev);
+	struct device *hwdev = dev->dev.parent;
 
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
 	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->bus_info, pci_name(rp->pdev), sizeof(info->bus_info));
+	strlcpy(info->bus_info, dev_name(hwdev), sizeof(info->bus_info));
 }
 
 static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
@@ -2277,7 +2352,7 @@ static int rhine_close(struct net_device *dev)
 	/* Stop the chip's Tx and Rx processes. */
 	iowrite16(CmdStop, ioaddr + ChipCmd);
 
-	free_irq(rp->pdev->irq, dev);
+	free_irq(rp->irq, dev);
 	free_rbufs(dev);
 	free_tbufs(dev);
 	free_ring(dev);
@@ -2286,7 +2361,7 @@ static int rhine_close(struct net_device *dev)
 }
 
 
-static void rhine_remove_one(struct pci_dev *pdev)
+static void rhine_remove_one_pci(struct pci_dev *pdev)
 {
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct rhine_private *rp = netdev_priv(dev);
@@ -2300,7 +2375,21 @@ static void rhine_remove_one(struct pci_dev *pdev)
 	pci_disable_device(pdev);
 }
 
-static void rhine_shutdown (struct pci_dev *pdev)
+static int rhine_remove_one_platform(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct rhine_private *rp = netdev_priv(dev);
+
+	unregister_netdev(dev);
+
+	iounmap(rp->base);
+
+	free_netdev(dev);
+
+	return 0;
+}
+
+static void rhine_shutdown_pci(struct pci_dev *pdev)
 {
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct rhine_private *rp = netdev_priv(dev);
@@ -2354,8 +2443,7 @@ static void rhine_shutdown (struct pci_dev *pdev)
 #ifdef CONFIG_PM_SLEEP
 static int rhine_suspend(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct rhine_private *rp = netdev_priv(dev);
 
 	if (!netif_running(dev))
@@ -2367,23 +2455,21 @@ static int rhine_suspend(struct device *device)
 
 	netif_device_detach(dev);
 
-	rhine_shutdown(pdev);
+	if (dev_is_pci(device))
+		rhine_shutdown_pci(to_pci_dev(device));
 
 	return 0;
 }
 
 static int rhine_resume(struct device *device)
 {
-	struct pci_dev *pdev = to_pci_dev(device);
-	struct net_device *dev = pci_get_drvdata(pdev);
+	struct net_device *dev = dev_get_drvdata(device);
 	struct rhine_private *rp = netdev_priv(dev);
 
 	if (!netif_running(dev))
 		return 0;
 
-#ifdef USE_MMIO
 	enable_mmio(rp->pioaddr, rp->quirks);
-#endif
 	rhine_power_init(dev);
 	free_tbufs(dev);
 	free_rbufs(dev);
@@ -2408,15 +2494,26 @@ static SIMPLE_DEV_PM_OPS(rhine_pm_ops, rhine_suspend, rhine_resume);
 
 #endif /* !CONFIG_PM_SLEEP */
 
-static struct pci_driver rhine_driver = {
+static struct pci_driver rhine_driver_pci = {
 	.name		= DRV_NAME,
 	.id_table	= rhine_pci_tbl,
-	.probe		= rhine_init_one,
-	.remove		= rhine_remove_one,
-	.shutdown	= rhine_shutdown,
+	.probe		= rhine_init_one_pci,
+	.remove		= rhine_remove_one_pci,
+	.shutdown	= rhine_shutdown_pci,
 	.driver.pm	= RHINE_PM_OPS,
 };
 
+static struct platform_driver rhine_driver_platform = {
+	.probe		= rhine_init_one_platform,
+	.remove		= rhine_remove_one_platform,
+	.driver = {
+		.name	= DRV_NAME,
+		.owner	= THIS_MODULE,
+		.of_match_table	= rhine_of_tbl,
+		.pm		= RHINE_PM_OPS,
+	}
+};
+
 static struct dmi_system_id rhine_dmi_table[] __initdata = {
 	{
 		.ident = "EPIA-M",
@@ -2437,6 +2534,8 @@ static struct dmi_system_id rhine_dmi_table[] __initdata = {
 
 static int __init rhine_init(void)
 {
+	int ret_pci, ret_platform;
+
 /* when a module, this is printed whether or not devices are found in probe */
 #ifdef MODULE
 	pr_info("%s\n", version);
@@ -2449,13 +2548,19 @@ static int __init rhine_init(void)
 	else if (avoid_D3)
 		pr_info("avoid_D3 set\n");
 
-	return pci_register_driver(&rhine_driver);
+	ret_pci = pci_register_driver(&rhine_driver_pci);
+	ret_platform = platform_driver_register(&rhine_driver_platform);
+	if ((ret_pci < 0) && (ret_platform < 0))
+		return ret_pci;
+
+	return 0;
 }
 
 
 static void __exit rhine_cleanup(void)
 {
-	pci_unregister_driver(&rhine_driver);
+	platform_driver_unregister(&rhine_driver_platform);
+	pci_unregister_driver(&rhine_driver_pci);
 }
 
 
diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c
index fa193c4688da..4ef818a7a6c6 100644
--- a/drivers/net/ethernet/xilinx/ll_temac_main.c
+++ b/drivers/net/ethernet/xilinx/ll_temac_main.c
@@ -75,7 +75,7 @@ int temac_indirect_busywait(struct temac_local *lp)
 	long end = jiffies + 2;
 
 	while (!(temac_ior(lp, XTE_RDY0_OFFSET) & XTE_RDY0_HARD_ACS_RDY_MASK)) {
-		if (end - jiffies <= 0) {
+		if (time_before_eq(end, jiffies)) {
 			WARN_ON(1);
 			return -ETIMEDOUT;
 		}
diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c
index 64b4639f43b6..d4abf478e2bb 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c
@@ -22,7 +22,7 @@ int axienet_mdio_wait_until_ready(struct axienet_local *lp)
 	long end = jiffies + 2;
 	while (!(axienet_ior(lp, XAE_MDIO_MCR_OFFSET) &
 		 XAE_MDIO_MCR_READY_MASK)) {
-		if (end - jiffies <= 0) {
+		if (time_before_eq(end, jiffies)) {
 			WARN_ON(1);
 			return -ETIMEDOUT;
 		}
diff --git a/drivers/net/ethernet/xilinx/xilinx_emaclite.c b/drivers/net/ethernet/xilinx/xilinx_emaclite.c
index 0d87c67a5ff7..8c4aed3053eb 100644
--- a/drivers/net/ethernet/xilinx/xilinx_emaclite.c
+++ b/drivers/net/ethernet/xilinx/xilinx_emaclite.c
@@ -702,7 +702,7 @@ static int xemaclite_mdio_wait(struct net_local *lp)
 	*/
 	while (__raw_readl(lp->base_addr + XEL_MDIOCTRL_OFFSET) &
 			XEL_MDIOCTRL_MDIOSTS_MASK) {
-		if (end - jiffies <= 0) {
+		if (time_before_eq(end, jiffies)) {
 			WARN_ON(1);
 			return -ETIMEDOUT;
 		}
diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index d18f711d0b0c..6cc37c15e0bf 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -28,50 +28,119 @@
 #include <linux/hyperv.h>
 #include <linux/rndis.h>
 
-/* Fwd declaration */
-struct hv_netvsc_packet;
-struct ndis_tcp_ip_checksum_info;
+/* RSS related */
+#define OID_GEN_RECEIVE_SCALE_CAPABILITIES 0x00010203  /* query only */
+#define OID_GEN_RECEIVE_SCALE_PARAMETERS 0x00010204  /* query and set */
 
-/* Represent the xfer page packet which contains 1 or more netvsc packet */
-struct xferpage_packet {
-	struct list_head list_ent;
-	u32 status;
+#define NDIS_OBJECT_TYPE_RSS_CAPABILITIES 0x88
+#define NDIS_OBJECT_TYPE_RSS_PARAMETERS 0x89
 
-	/* # of netvsc packets this xfer packet contains */
-	u32 count;
+#define NDIS_RECEIVE_SCALE_CAPABILITIES_REVISION_2 2
+#define NDIS_RECEIVE_SCALE_PARAMETERS_REVISION_2 2
+
+struct ndis_obj_header {
+	u8 type;
+	u8 rev;
+	u16 size;
+} __packed;
+
+/* ndis_recv_scale_cap/cap_flag */
+#define NDIS_RSS_CAPS_MESSAGE_SIGNALED_INTERRUPTS 0x01000000
+#define NDIS_RSS_CAPS_CLASSIFICATION_AT_ISR       0x02000000
+#define NDIS_RSS_CAPS_CLASSIFICATION_AT_DPC       0x04000000
+#define NDIS_RSS_CAPS_USING_MSI_X                 0x08000000
+#define NDIS_RSS_CAPS_RSS_AVAILABLE_ON_PORTS      0x10000000
+#define NDIS_RSS_CAPS_SUPPORTS_MSI_X              0x20000000
+#define NDIS_RSS_CAPS_HASH_TYPE_TCP_IPV4          0x00000100
+#define NDIS_RSS_CAPS_HASH_TYPE_TCP_IPV6          0x00000200
+#define NDIS_RSS_CAPS_HASH_TYPE_TCP_IPV6_EX       0x00000400
+
+struct ndis_recv_scale_cap { /* NDIS_RECEIVE_SCALE_CAPABILITIES */
+	struct ndis_obj_header hdr;
+	u32 cap_flag;
+	u32 num_int_msg;
+	u32 num_recv_que;
+	u16 num_indirect_tabent;
+} __packed;
+
+
+/* ndis_recv_scale_param flags */
+#define NDIS_RSS_PARAM_FLAG_BASE_CPU_UNCHANGED     0x0001
+#define NDIS_RSS_PARAM_FLAG_HASH_INFO_UNCHANGED    0x0002
+#define NDIS_RSS_PARAM_FLAG_ITABLE_UNCHANGED       0x0004
+#define NDIS_RSS_PARAM_FLAG_HASH_KEY_UNCHANGED     0x0008
+#define NDIS_RSS_PARAM_FLAG_DISABLE_RSS            0x0010
+
+/* Hash info bits */
+#define NDIS_HASH_FUNC_TOEPLITZ 0x00000001
+#define NDIS_HASH_IPV4          0x00000100
+#define NDIS_HASH_TCP_IPV4      0x00000200
+#define NDIS_HASH_IPV6          0x00000400
+#define NDIS_HASH_IPV6_EX       0x00000800
+#define NDIS_HASH_TCP_IPV6      0x00001000
+#define NDIS_HASH_TCP_IPV6_EX   0x00002000
+
+#define NDIS_RSS_INDIRECTION_TABLE_MAX_SIZE_REVISION_2 (128 * 4)
+#define NDIS_RSS_HASH_SECRET_KEY_MAX_SIZE_REVISION_2   40
+
+#define ITAB_NUM 128
+#define HASH_KEYLEN NDIS_RSS_HASH_SECRET_KEY_MAX_SIZE_REVISION_2
+extern u8 netvsc_hash_key[];
+
+struct ndis_recv_scale_param { /* NDIS_RECEIVE_SCALE_PARAMETERS */
+	struct ndis_obj_header hdr;
+
+	/* Qualifies the rest of the information */
+	u16 flag;
+
+	/* The base CPU number to do receive processing. not used */
+	u16 base_cpu_number;
+
+	/* This describes the hash function and type being enabled */
+	u32 hashinfo;
+
+	/* The size of indirection table array */
+	u16 indirect_tabsize;
+
+	/* The offset of the indirection table from the beginning of this
+	 * structure
+	 */
+	u32 indirect_taboffset;
+
+	/* The size of the hash secret key */
+	u16 hashkey_size;
+
+	/* The offset of the secret key from the beginning of this structure */
+	u32 kashkey_offset;
+
+	u32 processor_masks_offset;
+	u32 num_processor_masks;
+	u32 processor_masks_entry_size;
 };
 
+/* Fwd declaration */
+struct ndis_tcp_ip_checksum_info;
+
 /*
  * Represent netvsc packet which contains 1 RNDIS and 1 ethernet frame
  * within the RNDIS
  */
 struct hv_netvsc_packet {
 	/* Bookkeeping stuff */
-	struct list_head list_ent;
 	u32 status;
 
 	struct hv_device *device;
 	bool is_data_pkt;
 	u16 vlan_tci;
 
-	/*
-	 * Valid only for receives when we break a xfer page packet
-	 * into multiple netvsc packets
-	 */
-	struct xferpage_packet *xfer_page_pkt;
+	u16 q_idx;
+	struct vmbus_channel *channel;
 
-	union {
-		struct {
-			u64 recv_completion_tid;
-			void *recv_completion_ctx;
-			void (*recv_completion)(void *context);
-		} recv;
-		struct {
-			u64 send_completion_tid;
-			void *send_completion_ctx;
-			void (*send_completion)(void *context);
-		} send;
-	} completion;
+	u64 send_completion_tid;
+	void *send_completion_ctx;
+	void (*send_completion)(void *context);
+
+	u32 send_buf_index;
 
 	/* This points to the memory after page_buf */
 	struct rndis_message *rndis_msg;
@@ -120,6 +189,7 @@ void netvsc_linkstatus_callback(struct hv_device *device_obj,
 int netvsc_recv_callback(struct hv_device *device_obj,
 			struct hv_netvsc_packet *packet,
 			struct ndis_tcp_ip_checksum_info *csum_info);
+void netvsc_channel_cb(void *context);
 int rndis_filter_open(struct hv_device *dev);
 int rndis_filter_close(struct hv_device *dev);
 int rndis_filter_device_add(struct hv_device *dev,
@@ -514,14 +584,16 @@ struct nvsp_message {
 
 #define NETVSC_RECEIVE_BUFFER_SIZE		(1024*1024*16)	/* 16MB */
 #define NETVSC_RECEIVE_BUFFER_SIZE_LEGACY	(1024*1024*15)  /* 15MB */
+#define NETVSC_SEND_BUFFER_SIZE			(1024 * 1024)   /* 1MB */
+#define NETVSC_INVALID_INDEX			-1
 
-#define NETVSC_RECEIVE_BUFFER_ID		0xcafe
 
-/* Preallocated receive packets */
-#define NETVSC_RECEIVE_PACKETLIST_COUNT		256
+#define NETVSC_RECEIVE_BUFFER_ID		0xcafe
 
 #define NETVSC_PACKET_SIZE                      2048
 
+#define VRSS_SEND_TAB_SIZE 16
+
 /* Per netvsc channel-specific */
 struct netvsc_device {
 	struct hv_device *dev;
@@ -532,12 +604,6 @@ struct netvsc_device {
 	wait_queue_head_t wait_drain;
 	bool start_remove;
 	bool destroy;
-	/*
-	 * List of free preallocated hv_netvsc_packet to represent receive
-	 * packet
-	 */
-	struct list_head recv_pkt_list;
-	spinlock_t recv_pkt_list_lock;
 
 	/* Receive buffer allocated by us but manages by NetVSP */
 	void *recv_buf;
@@ -546,6 +612,15 @@ struct netvsc_device {
 	u32 recv_section_cnt;
 	struct nvsp_1_receive_buffer_section *recv_section;
 
+	/* Send buffer allocated by us */
+	void *send_buf;
+	u32 send_buf_size;
+	u32 send_buf_gpadl_handle;
+	u32 send_section_cnt;
+	u32 send_section_size;
+	unsigned long *send_section_map;
+	int map_words;
+
 	/* Used for NetVSP initialization protocol */
 	struct completion channel_init_wait;
 	struct nvsp_message channel_init_pkt;
@@ -555,10 +630,20 @@ struct netvsc_device {
 
 	struct net_device *ndev;
 
+	struct vmbus_channel *chn_table[NR_CPUS];
+	u32 send_table[VRSS_SEND_TAB_SIZE];
+	u32 num_chn;
+	atomic_t queue_sends[NR_CPUS];
+
 	/* Holds rndis device info */
 	void *extension;
-	/* The recive buffer for this device */
+
+	int ring_size;
+
+	/* The primary channel callback buffer */
 	unsigned char cb_buffer[NETVSC_PACKET_SIZE];
+	/* The sub channel callback buffer */
+	unsigned char *sub_cb_buf;
 };
 
 /* NdisInitialize message */
@@ -706,6 +791,7 @@ enum ndis_per_pkt_info_type {
 	IEEE_8021Q_INFO,
 	ORIGINAL_PKTINFO,
 	PACKET_CANCEL_ID,
+	NBL_HASH_VALUE = PACKET_CANCEL_ID,
 	ORIGINAL_NET_BUFLIST,
 	CACHED_NET_BUFLIST,
 	SHORT_PKT_PADINFO,
@@ -852,6 +938,9 @@ struct ndis_tcp_lso_info {
 #define NDIS_LSO_PPI_SIZE (sizeof(struct rndis_per_packet_info) + \
 		sizeof(struct ndis_tcp_lso_info))
 
+#define NDIS_HASH_PPI_SIZE (sizeof(struct rndis_per_packet_info) + \
+		sizeof(u32))
+
 /* Format of Information buffer passed in a SetRequest for the OID */
 /* OID_GEN_RNDIS_CONFIG_PARAMETER. */
 struct rndis_config_parameter_info {
diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c
index f7629ecefa84..4ed38eaecea8 100644
--- a/drivers/net/hyperv/netvsc.c
+++ b/drivers/net/hyperv/netvsc.c
@@ -28,6 +28,7 @@
 #include <linux/slab.h>
 #include <linux/netdevice.h>
 #include <linux/if_ether.h>
+#include <asm/sync_bitops.h>
 
 #include "hyperv_net.h"
 
@@ -80,7 +81,7 @@ get_in_err:
 }
 
 
-static int netvsc_destroy_recv_buf(struct netvsc_device *net_device)
+static int netvsc_destroy_buf(struct netvsc_device *net_device)
 {
 	struct nvsp_message *revoke_packet;
 	int ret = 0;
@@ -146,10 +147,62 @@ static int netvsc_destroy_recv_buf(struct netvsc_device *net_device)
 		net_device->recv_section = NULL;
 	}
 
+	/* Deal with the send buffer we may have setup.
+	 * If we got a  send section size, it means we received a
+	 * SendsendBufferComplete msg (ie sent
+	 * NvspMessage1TypeSendReceiveBuffer msg) therefore, we need
+	 * to send a revoke msg here
+	 */
+	if (net_device->send_section_size) {
+		/* Send the revoke receive buffer */
+		revoke_packet = &net_device->revoke_packet;
+		memset(revoke_packet, 0, sizeof(struct nvsp_message));
+
+		revoke_packet->hdr.msg_type =
+			NVSP_MSG1_TYPE_REVOKE_SEND_BUF;
+		revoke_packet->msg.v1_msg.revoke_recv_buf.id = 0;
+
+		ret = vmbus_sendpacket(net_device->dev->channel,
+				       revoke_packet,
+				       sizeof(struct nvsp_message),
+				       (unsigned long)revoke_packet,
+				       VM_PKT_DATA_INBAND, 0);
+		/* If we failed here, we might as well return and
+		 * have a leak rather than continue and a bugchk
+		 */
+		if (ret != 0) {
+			netdev_err(ndev, "unable to send "
+				   "revoke send buffer to netvsp\n");
+			return ret;
+		}
+	}
+	/* Teardown the gpadl on the vsp end */
+	if (net_device->send_buf_gpadl_handle) {
+		ret = vmbus_teardown_gpadl(net_device->dev->channel,
+					   net_device->send_buf_gpadl_handle);
+
+		/* If we failed here, we might as well return and have a leak
+		 * rather than continue and a bugchk
+		 */
+		if (ret != 0) {
+			netdev_err(ndev,
+				   "unable to teardown send buffer's gpadl\n");
+			return ret;
+		}
+		net_device->send_buf_gpadl_handle = 0;
+	}
+	if (net_device->send_buf) {
+		/* Free up the receive buffer */
+		free_pages((unsigned long)net_device->send_buf,
+			   get_order(net_device->send_buf_size));
+		net_device->send_buf = NULL;
+	}
+	kfree(net_device->send_section_map);
+
 	return ret;
 }
 
-static int netvsc_init_recv_buf(struct hv_device *device)
+static int netvsc_init_buf(struct hv_device *device)
 {
 	int ret = 0;
 	int t;
@@ -248,10 +301,90 @@ static int netvsc_init_recv_buf(struct hv_device *device)
 		goto cleanup;
 	}
 
+	/* Now setup the send buffer.
+	 */
+	net_device->send_buf =
+		(void *)__get_free_pages(GFP_KERNEL|__GFP_ZERO,
+					 get_order(net_device->send_buf_size));
+	if (!net_device->send_buf) {
+		netdev_err(ndev, "unable to allocate send "
+			   "buffer of size %d\n", net_device->send_buf_size);
+		ret = -ENOMEM;
+		goto cleanup;
+	}
+
+	/* Establish the gpadl handle for this buffer on this
+	 * channel.  Note: This call uses the vmbus connection rather
+	 * than the channel to establish the gpadl handle.
+	 */
+	ret = vmbus_establish_gpadl(device->channel, net_device->send_buf,
+				    net_device->send_buf_size,
+				    &net_device->send_buf_gpadl_handle);
+	if (ret != 0) {
+		netdev_err(ndev,
+			   "unable to establish send buffer's gpadl\n");
+		goto cleanup;
+	}
+
+	/* Notify the NetVsp of the gpadl handle */
+	init_packet = &net_device->channel_init_pkt;
+	memset(init_packet, 0, sizeof(struct nvsp_message));
+	init_packet->hdr.msg_type = NVSP_MSG1_TYPE_SEND_SEND_BUF;
+	init_packet->msg.v1_msg.send_recv_buf.gpadl_handle =
+		net_device->send_buf_gpadl_handle;
+	init_packet->msg.v1_msg.send_recv_buf.id = 0;
+
+	/* Send the gpadl notification request */
+	ret = vmbus_sendpacket(device->channel, init_packet,
+			       sizeof(struct nvsp_message),
+			       (unsigned long)init_packet,
+			       VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (ret != 0) {
+		netdev_err(ndev,
+			   "unable to send send buffer's gpadl to netvsp\n");
+		goto cleanup;
+	}
+
+	t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ);
+	BUG_ON(t == 0);
+
+	/* Check the response */
+	if (init_packet->msg.v1_msg.
+	    send_send_buf_complete.status != NVSP_STAT_SUCCESS) {
+		netdev_err(ndev, "Unable to complete send buffer "
+			   "initialization with NetVsp - status %d\n",
+			   init_packet->msg.v1_msg.
+			   send_recv_buf_complete.status);
+		ret = -EINVAL;
+		goto cleanup;
+	}
+
+	/* Parse the response */
+	net_device->send_section_size = init_packet->msg.
+				v1_msg.send_send_buf_complete.section_size;
+
+	/* Section count is simply the size divided by the section size.
+	 */
+	net_device->send_section_cnt =
+		net_device->send_buf_size/net_device->send_section_size;
+
+	dev_info(&device->device, "Send section size: %d, Section count:%d\n",
+		 net_device->send_section_size, net_device->send_section_cnt);
+
+	/* Setup state for managing the send buffer. */
+	net_device->map_words = DIV_ROUND_UP(net_device->send_section_cnt,
+					     BITS_PER_LONG);
+
+	net_device->send_section_map =
+		kzalloc(net_device->map_words * sizeof(ulong), GFP_KERNEL);
+	if (net_device->send_section_map == NULL)
+		goto cleanup;
+
 	goto exit;
 
 cleanup:
-	netvsc_destroy_recv_buf(net_device);
+	netvsc_destroy_buf(net_device);
 
 exit:
 	return ret;
@@ -369,8 +502,9 @@ static int netvsc_connect_vsp(struct hv_device *device)
 		net_device->recv_buf_size = NETVSC_RECEIVE_BUFFER_SIZE_LEGACY;
 	else
 		net_device->recv_buf_size = NETVSC_RECEIVE_BUFFER_SIZE;
+	net_device->send_buf_size = NETVSC_SEND_BUFFER_SIZE;
 
-	ret = netvsc_init_recv_buf(device);
+	ret = netvsc_init_buf(device);
 
 cleanup:
 	return ret;
@@ -378,7 +512,7 @@ cleanup:
 
 static void netvsc_disconnect_vsp(struct netvsc_device *net_device)
 {
-	netvsc_destroy_recv_buf(net_device);
+	netvsc_destroy_buf(net_device);
 }
 
 /*
@@ -387,7 +521,6 @@ static void netvsc_disconnect_vsp(struct netvsc_device *net_device)
 int netvsc_device_remove(struct hv_device *device)
 {
 	struct netvsc_device *net_device;
-	struct hv_netvsc_packet *netvsc_packet, *pos;
 	unsigned long flags;
 
 	net_device = hv_get_drvdata(device);
@@ -416,11 +549,8 @@ int netvsc_device_remove(struct hv_device *device)
 	vmbus_close(device->channel);
 
 	/* Release all resources */
-	list_for_each_entry_safe(netvsc_packet, pos,
-				 &net_device->recv_pkt_list, list_ent) {
-		list_del(&netvsc_packet->list_ent);
-		kfree(netvsc_packet);
-	}
+	if (net_device->sub_cb_buf)
+		vfree(net_device->sub_cb_buf);
 
 	kfree(net_device);
 	return 0;
@@ -444,6 +574,12 @@ static inline u32 hv_ringbuf_avail_percent(
 	return avail_write * 100 / ring_info->ring_datasize;
 }
 
+static inline void netvsc_free_send_slot(struct netvsc_device *net_device,
+					 u32 index)
+{
+	sync_change_bit(index, net_device->send_section_map);
+}
+
 static void netvsc_send_completion(struct netvsc_device *net_device,
 				   struct hv_device *device,
 				   struct vmpacket_descriptor *packet)
@@ -451,6 +587,7 @@ static void netvsc_send_completion(struct netvsc_device *net_device,
 	struct nvsp_message *nvsp_packet;
 	struct hv_netvsc_packet *nvsc_packet;
 	struct net_device *ndev;
+	u32 send_index;
 
 	ndev = net_device->ndev;
 
@@ -461,7 +598,9 @@ static void netvsc_send_completion(struct netvsc_device *net_device,
 	    (nvsp_packet->hdr.msg_type ==
 	     NVSP_MSG1_TYPE_SEND_RECV_BUF_COMPLETE) ||
 	    (nvsp_packet->hdr.msg_type ==
-	     NVSP_MSG1_TYPE_SEND_SEND_BUF_COMPLETE)) {
+	     NVSP_MSG1_TYPE_SEND_SEND_BUF_COMPLETE) ||
+	    (nvsp_packet->hdr.msg_type ==
+	     NVSP_MSG5_TYPE_SUBCHANNEL)) {
 		/* Copy the response back */
 		memcpy(&net_device->channel_init_pkt, nvsp_packet,
 		       sizeof(struct nvsp_message));
@@ -469,28 +608,39 @@ static void netvsc_send_completion(struct netvsc_device *net_device,
 	} else if (nvsp_packet->hdr.msg_type ==
 		   NVSP_MSG1_TYPE_SEND_RNDIS_PKT_COMPLETE) {
 		int num_outstanding_sends;
+		u16 q_idx = 0;
+		struct vmbus_channel *channel = device->channel;
+		int queue_sends;
 
 		/* Get the send context */
 		nvsc_packet = (struct hv_netvsc_packet *)(unsigned long)
 			packet->trans_id;
 
 		/* Notify the layer above us */
-		if (nvsc_packet)
-			nvsc_packet->completion.send.send_completion(
-				nvsc_packet->completion.send.
-				send_completion_ctx);
+		if (nvsc_packet) {
+			send_index = nvsc_packet->send_buf_index;
+			if (send_index != NETVSC_INVALID_INDEX)
+				netvsc_free_send_slot(net_device, send_index);
+			q_idx = nvsc_packet->q_idx;
+			channel = nvsc_packet->channel;
+			nvsc_packet->send_completion(nvsc_packet->
+						     send_completion_ctx);
+		}
 
 		num_outstanding_sends =
 			atomic_dec_return(&net_device->num_outstanding_sends);
+		queue_sends = atomic_dec_return(&net_device->
+						queue_sends[q_idx]);
 
 		if (net_device->destroy && num_outstanding_sends == 0)
 			wake_up(&net_device->wait_drain);
 
-		if (netif_queue_stopped(ndev) && !net_device->start_remove &&
-			(hv_ringbuf_avail_percent(&device->channel->outbound)
-			> RING_AVAIL_PERCENT_HIWATER ||
-			num_outstanding_sends < 1))
-				netif_wake_queue(ndev);
+		if (netif_tx_queue_stopped(netdev_get_tx_queue(ndev, q_idx)) &&
+		    !net_device->start_remove &&
+		    (hv_ringbuf_avail_percent(&channel->outbound) >
+		     RING_AVAIL_PERCENT_HIWATER || queue_sends < 1))
+				netif_tx_wake_queue(netdev_get_tx_queue(
+						    ndev, q_idx));
 	} else {
 		netdev_err(ndev, "Unknown send completion packet type- "
 			   "%d received!!\n", nvsp_packet->hdr.msg_type);
@@ -498,6 +648,52 @@ static void netvsc_send_completion(struct netvsc_device *net_device,
 
 }
 
+static u32 netvsc_get_next_send_section(struct netvsc_device *net_device)
+{
+	unsigned long index;
+	u32 max_words = net_device->map_words;
+	unsigned long *map_addr = (unsigned long *)net_device->send_section_map;
+	u32 section_cnt = net_device->send_section_cnt;
+	int ret_val = NETVSC_INVALID_INDEX;
+	int i;
+	int prev_val;
+
+	for (i = 0; i < max_words; i++) {
+		if (!~(map_addr[i]))
+			continue;
+		index = ffz(map_addr[i]);
+		prev_val = sync_test_and_set_bit(index, &map_addr[i]);
+		if (prev_val)
+			continue;
+		if ((index + (i * BITS_PER_LONG)) >= section_cnt)
+			break;
+		ret_val = (index + (i * BITS_PER_LONG));
+		break;
+	}
+	return ret_val;
+}
+
+u32 netvsc_copy_to_send_buf(struct netvsc_device *net_device,
+			    unsigned int section_index,
+			    struct hv_netvsc_packet *packet)
+{
+	char *start = net_device->send_buf;
+	char *dest = (start + (section_index * net_device->send_section_size));
+	int i;
+	u32 msg_size = 0;
+
+	for (i = 0; i < packet->page_buf_cnt; i++) {
+		char *src = phys_to_virt(packet->page_buf[i].pfn << PAGE_SHIFT);
+		u32 offset = packet->page_buf[i].offset;
+		u32 len = packet->page_buf[i].len;
+
+		memcpy(dest, (src + offset), len);
+		msg_size += len;
+		dest += len;
+	}
+	return msg_size;
+}
+
 int netvsc_send(struct hv_device *device,
 			struct hv_netvsc_packet *packet)
 {
@@ -505,7 +701,12 @@ int netvsc_send(struct hv_device *device,
 	int ret = 0;
 	struct nvsp_message sendMessage;
 	struct net_device *ndev;
+	struct vmbus_channel *out_channel = NULL;
 	u64 req_id;
+	unsigned int section_index = NETVSC_INVALID_INDEX;
+	u32 msg_size = 0;
+	struct sk_buff *skb;
+
 
 	net_device = get_outbound_net_device(device);
 	if (!net_device)
@@ -521,25 +722,46 @@ int netvsc_send(struct hv_device *device,
 		sendMessage.msg.v1_msg.send_rndis_pkt.channel_type = 1;
 	}
 
-	/* Not using send buffer section */
+	/* Attempt to send via sendbuf */
+	if (packet->total_data_buflen < net_device->send_section_size) {
+		section_index = netvsc_get_next_send_section(net_device);
+		if (section_index != NETVSC_INVALID_INDEX) {
+			msg_size = netvsc_copy_to_send_buf(net_device,
+							   section_index,
+							   packet);
+			skb = (struct sk_buff *)
+			      (unsigned long)packet->send_completion_tid;
+			if (skb)
+				dev_kfree_skb_any(skb);
+			packet->page_buf_cnt = 0;
+		}
+	}
+	packet->send_buf_index = section_index;
+
+
 	sendMessage.msg.v1_msg.send_rndis_pkt.send_buf_section_index =
-		0xFFFFFFFF;
-	sendMessage.msg.v1_msg.send_rndis_pkt.send_buf_section_size = 0;
+		section_index;
+	sendMessage.msg.v1_msg.send_rndis_pkt.send_buf_section_size = msg_size;
 
-	if (packet->completion.send.send_completion)
+	if (packet->send_completion)
 		req_id = (ulong)packet;
 	else
 		req_id = 0;
 
+	out_channel = net_device->chn_table[packet->q_idx];
+	if (out_channel == NULL)
+		out_channel = device->channel;
+	packet->channel = out_channel;
+
 	if (packet->page_buf_cnt) {
-		ret = vmbus_sendpacket_pagebuffer(device->channel,
+		ret = vmbus_sendpacket_pagebuffer(out_channel,
 						  packet->page_buf,
 						  packet->page_buf_cnt,
 						  &sendMessage,
 						  sizeof(struct nvsp_message),
 						  req_id);
 	} else {
-		ret = vmbus_sendpacket(device->channel, &sendMessage,
+		ret = vmbus_sendpacket(out_channel, &sendMessage,
 				sizeof(struct nvsp_message),
 				req_id,
 				VM_PKT_DATA_INBAND,
@@ -548,17 +770,24 @@ int netvsc_send(struct hv_device *device,
 
 	if (ret == 0) {
 		atomic_inc(&net_device->num_outstanding_sends);
-		if (hv_ringbuf_avail_percent(&device->channel->outbound) <
+		atomic_inc(&net_device->queue_sends[packet->q_idx]);
+
+		if (hv_ringbuf_avail_percent(&out_channel->outbound) <
 			RING_AVAIL_PERCENT_LOWATER) {
-			netif_stop_queue(ndev);
+			netif_tx_stop_queue(netdev_get_tx_queue(
+					    ndev, packet->q_idx));
+
 			if (atomic_read(&net_device->
-				num_outstanding_sends) < 1)
-				netif_wake_queue(ndev);
+				queue_sends[packet->q_idx]) < 1)
+				netif_tx_wake_queue(netdev_get_tx_queue(
+						    ndev, packet->q_idx));
 		}
 	} else if (ret == -EAGAIN) {
-		netif_stop_queue(ndev);
-		if (atomic_read(&net_device->num_outstanding_sends) < 1) {
-			netif_wake_queue(ndev);
+		netif_tx_stop_queue(netdev_get_tx_queue(
+				    ndev, packet->q_idx));
+		if (atomic_read(&net_device->queue_sends[packet->q_idx]) < 1) {
+			netif_tx_wake_queue(netdev_get_tx_queue(
+					    ndev, packet->q_idx));
 			ret = -ENOSPC;
 		}
 	} else {
@@ -570,6 +799,7 @@ int netvsc_send(struct hv_device *device,
 }
 
 static void netvsc_send_recv_completion(struct hv_device *device,
+					struct vmbus_channel *channel,
 					struct netvsc_device *net_device,
 					u64 transaction_id, u32 status)
 {
@@ -587,7 +817,7 @@ static void netvsc_send_recv_completion(struct hv_device *device,
 
 retry_send_cmplt:
 	/* Send the completion */
-	ret = vmbus_sendpacket(device->channel, &recvcompMessage,
+	ret = vmbus_sendpacket(channel, &recvcompMessage,
 			       sizeof(struct nvsp_message), transaction_id,
 			       VM_PKT_COMP, 0);
 	if (ret == 0) {
@@ -613,76 +843,20 @@ retry_send_cmplt:
 	}
 }
 
-/* Send a receive completion packet to RNDIS device (ie NetVsp) */
-static void netvsc_receive_completion(void *context)
-{
-	struct hv_netvsc_packet *packet = context;
-	struct hv_device *device = packet->device;
-	struct netvsc_device *net_device;
-	u64 transaction_id = 0;
-	bool fsend_receive_comp = false;
-	unsigned long flags;
-	struct net_device *ndev;
-	u32 status = NVSP_STAT_NONE;
-
-	/*
-	 * Even though it seems logical to do a GetOutboundNetDevice() here to
-	 * send out receive completion, we are using GetInboundNetDevice()
-	 * since we may have disable outbound traffic already.
-	 */
-	net_device = get_inbound_net_device(device);
-	if (!net_device)
-		return;
-	ndev = net_device->ndev;
-
-	/* Overloading use of the lock. */
-	spin_lock_irqsave(&net_device->recv_pkt_list_lock, flags);
-
-	if (packet->status != NVSP_STAT_SUCCESS)
-		packet->xfer_page_pkt->status = NVSP_STAT_FAIL;
-
-	packet->xfer_page_pkt->count--;
-
-	/*
-	 * Last one in the line that represent 1 xfer page packet.
-	 * Return the xfer page packet itself to the freelist
-	 */
-	if (packet->xfer_page_pkt->count == 0) {
-		fsend_receive_comp = true;
-		transaction_id = packet->completion.recv.recv_completion_tid;
-		status = packet->xfer_page_pkt->status;
-		list_add_tail(&packet->xfer_page_pkt->list_ent,
-			      &net_device->recv_pkt_list);
-
-	}
-
-	/* Put the packet back */
-	list_add_tail(&packet->list_ent, &net_device->recv_pkt_list);
-	spin_unlock_irqrestore(&net_device->recv_pkt_list_lock, flags);
-
-	/* Send a receive completion for the xfer page packet */
-	if (fsend_receive_comp)
-		netvsc_send_recv_completion(device, net_device, transaction_id,
-					status);
-
-}
-
 static void netvsc_receive(struct netvsc_device *net_device,
+			struct vmbus_channel *channel,
 			struct hv_device *device,
 			struct vmpacket_descriptor *packet)
 {
 	struct vmtransfer_page_packet_header *vmxferpage_packet;
 	struct nvsp_message *nvsp_packet;
-	struct hv_netvsc_packet *netvsc_packet = NULL;
-	/* struct netvsc_driver *netvscDriver; */
-	struct xferpage_packet *xferpage_packet = NULL;
+	struct hv_netvsc_packet nv_pkt;
+	struct hv_netvsc_packet *netvsc_packet = &nv_pkt;
+	u32 status = NVSP_STAT_SUCCESS;
 	int i;
 	int count = 0;
-	unsigned long flags;
 	struct net_device *ndev;
 
-	LIST_HEAD(listHead);
-
 	ndev = net_device->ndev;
 
 	/*
@@ -715,77 +889,14 @@ static void netvsc_receive(struct netvsc_device *net_device,
 		return;
 	}
 
-	/*
-	 * Grab free packets (range count + 1) to represent this xfer
-	 * page packet. +1 to represent the xfer page packet itself.
-	 * We grab it here so that we know exactly how many we can
-	 * fulfil
-	 */
-	spin_lock_irqsave(&net_device->recv_pkt_list_lock, flags);
-	while (!list_empty(&net_device->recv_pkt_list)) {
-		list_move_tail(net_device->recv_pkt_list.next, &listHead);
-		if (++count == vmxferpage_packet->range_cnt + 1)
-			break;
-	}
-	spin_unlock_irqrestore(&net_device->recv_pkt_list_lock, flags);
-
-	/*
-	 * We need at least 2 netvsc pkts (1 to represent the xfer
-	 * page and at least 1 for the range) i.e. we can handled
-	 * some of the xfer page packet ranges...
-	 */
-	if (count < 2) {
-		netdev_err(ndev, "Got only %d netvsc pkt...needed "
-			"%d pkts. Dropping this xfer page packet completely!\n",
-			count, vmxferpage_packet->range_cnt + 1);
-
-		/* Return it to the freelist */
-		spin_lock_irqsave(&net_device->recv_pkt_list_lock, flags);
-		for (i = count; i != 0; i--) {
-			list_move_tail(listHead.next,
-				       &net_device->recv_pkt_list);
-		}
-		spin_unlock_irqrestore(&net_device->recv_pkt_list_lock,
-				       flags);
-
-		netvsc_send_recv_completion(device, net_device,
-					    vmxferpage_packet->d.trans_id,
-					    NVSP_STAT_FAIL);
-
-		return;
-	}
-
-	/* Remove the 1st packet to represent the xfer page packet itself */
-	xferpage_packet = (struct xferpage_packet *)listHead.next;
-	list_del(&xferpage_packet->list_ent);
-	xferpage_packet->status = NVSP_STAT_SUCCESS;
-
-	/* This is how much we can satisfy */
-	xferpage_packet->count = count - 1;
-
-	if (xferpage_packet->count != vmxferpage_packet->range_cnt) {
-		netdev_err(ndev, "Needed %d netvsc pkts to satisfy "
-			"this xfer page...got %d\n",
-			vmxferpage_packet->range_cnt, xferpage_packet->count);
-	}
+	count = vmxferpage_packet->range_cnt;
+	netvsc_packet->device = device;
+	netvsc_packet->channel = channel;
 
 	/* Each range represents 1 RNDIS pkt that contains 1 ethernet frame */
-	for (i = 0; i < (count - 1); i++) {
-		netvsc_packet = (struct hv_netvsc_packet *)listHead.next;
-		list_del(&netvsc_packet->list_ent);
-
+	for (i = 0; i < count; i++) {
 		/* Initialize the netvsc packet */
 		netvsc_packet->status = NVSP_STAT_SUCCESS;
-		netvsc_packet->xfer_page_pkt = xferpage_packet;
-		netvsc_packet->completion.recv.recv_completion =
-					netvsc_receive_completion;
-		netvsc_packet->completion.recv.recv_completion_ctx =
-					netvsc_packet;
-		netvsc_packet->device = device;
-		/* Save this so that we can send it back */
-		netvsc_packet->completion.recv.recv_completion_tid =
-					vmxferpage_packet->d.trans_id;
-
 		netvsc_packet->data = (void *)((unsigned long)net_device->
 			recv_buf + vmxferpage_packet->ranges[i].byte_offset);
 		netvsc_packet->total_data_buflen =
@@ -794,16 +905,53 @@ static void netvsc_receive(struct netvsc_device *net_device,
 		/* Pass it to the upper layer */
 		rndis_filter_receive(device, netvsc_packet);
 
-		netvsc_receive_completion(netvsc_packet->
-				completion.recv.recv_completion_ctx);
+		if (netvsc_packet->status != NVSP_STAT_SUCCESS)
+			status = NVSP_STAT_FAIL;
+	}
+
+	netvsc_send_recv_completion(device, channel, net_device,
+				    vmxferpage_packet->d.trans_id, status);
+}
+
+
+static void netvsc_send_table(struct hv_device *hdev,
+			      struct vmpacket_descriptor *vmpkt)
+{
+	struct netvsc_device *nvscdev;
+	struct net_device *ndev;
+	struct nvsp_message *nvmsg;
+	int i;
+	u32 count, *tab;
+
+	nvscdev = get_outbound_net_device(hdev);
+	if (!nvscdev)
+		return;
+	ndev = nvscdev->ndev;
+
+	nvmsg = (struct nvsp_message *)((unsigned long)vmpkt +
+					(vmpkt->offset8 << 3));
+
+	if (nvmsg->hdr.msg_type != NVSP_MSG5_TYPE_SEND_INDIRECTION_TABLE)
+		return;
+
+	count = nvmsg->msg.v5_msg.send_table.count;
+	if (count != VRSS_SEND_TAB_SIZE) {
+		netdev_err(ndev, "Received wrong send-table size:%u\n", count);
+		return;
 	}
 
+	tab = (u32 *)((unsigned long)&nvmsg->msg.v5_msg.send_table +
+		      nvmsg->msg.v5_msg.send_table.offset);
+
+	for (i = 0; i < count; i++)
+		nvscdev->send_table[i] = tab[i];
 }
 
-static void netvsc_channel_cb(void *context)
+void netvsc_channel_cb(void *context)
 {
 	int ret;
-	struct hv_device *device = context;
+	struct vmbus_channel *channel = (struct vmbus_channel *)context;
+	struct hv_device *device;
 	struct netvsc_device *net_device;
 	u32 bytes_recvd;
 	u64 request_id;
@@ -812,14 +960,19 @@ static void netvsc_channel_cb(void *context)
 	int bufferlen = NETVSC_PACKET_SIZE;
 	struct net_device *ndev;
 
+	if (channel->primary_channel != NULL)
+		device = channel->primary_channel->device_obj;
+	else
+		device = channel->device_obj;
+
 	net_device = get_inbound_net_device(device);
 	if (!net_device)
 		return;
 	ndev = net_device->ndev;
-	buffer = net_device->cb_buffer;
+	buffer = get_per_channel_state(channel);
 
 	do {
-		ret = vmbus_recvpacket_raw(device->channel, buffer, bufferlen,
+		ret = vmbus_recvpacket_raw(channel, buffer, bufferlen,
 					   &bytes_recvd, &request_id);
 		if (ret == 0) {
 			if (bytes_recvd > 0) {
@@ -831,8 +984,12 @@ static void netvsc_channel_cb(void *context)
 					break;
 
 				case VM_PKT_DATA_USING_XFER_PAGES:
-					netvsc_receive(net_device,
-							device, desc);
+					netvsc_receive(net_device, channel,
+						       device, desc);
+					break;
+
+				case VM_PKT_DATA_INBAND:
+					netvsc_send_table(device, desc);
 					break;
 
 				default:
@@ -880,11 +1037,9 @@ static void netvsc_channel_cb(void *context)
 int netvsc_device_add(struct hv_device *device, void *additional_info)
 {
 	int ret = 0;
-	int i;
 	int ring_size =
 	((struct netvsc_device_info *)additional_info)->ring_size;
 	struct netvsc_device *net_device;
-	struct hv_netvsc_packet *packet, *pos;
 	struct net_device *ndev;
 
 	net_device = alloc_net_device(device);
@@ -893,6 +1048,8 @@ int netvsc_device_add(struct hv_device *device, void *additional_info)
 		goto cleanup;
 	}
 
+	net_device->ring_size = ring_size;
+
 	/*
 	 * Coming into this function, struct net_device * is
 	 * registered as the driver private data.
@@ -903,24 +1060,14 @@ int netvsc_device_add(struct hv_device *device, void *additional_info)
 	ndev = net_device->ndev;
 
 	/* Initialize the NetVSC channel extension */
-	spin_lock_init(&net_device->recv_pkt_list_lock);
-
-	INIT_LIST_HEAD(&net_device->recv_pkt_list);
-
-	for (i = 0; i < NETVSC_RECEIVE_PACKETLIST_COUNT; i++) {
-		packet = kzalloc(sizeof(struct hv_netvsc_packet), GFP_KERNEL);
-		if (!packet)
-			break;
-
-		list_add_tail(&packet->list_ent,
-			      &net_device->recv_pkt_list);
-	}
 	init_completion(&net_device->channel_init_wait);
 
+	set_per_channel_state(device->channel, net_device->cb_buffer);
+
 	/* Open the channel */
 	ret = vmbus_open(device->channel, ring_size * PAGE_SIZE,
 			 ring_size * PAGE_SIZE, NULL, 0,
-			 netvsc_channel_cb, device);
+			 netvsc_channel_cb, device->channel);
 
 	if (ret != 0) {
 		netdev_err(ndev, "unable to open channel: %d\n", ret);
@@ -930,6 +1077,8 @@ int netvsc_device_add(struct hv_device *device, void *additional_info)
 	/* Channel is opened */
 	pr_info("hv_netvsc channel opened successfully\n");
 
+	net_device->chn_table[0] = device->channel;
+
 	/* Connect with the NetVsp */
 	ret = netvsc_connect_vsp(device);
 	if (ret != 0) {
@@ -946,16 +1095,8 @@ close:
 
 cleanup:
 
-	if (net_device) {
-		list_for_each_entry_safe(packet, pos,
-					 &net_device->recv_pkt_list,
-					 list_ent) {
-			list_del(&packet->list_ent);
-			kfree(packet);
-		}
-
+	if (net_device)
 		kfree(net_device);
-	}
 
 	return ret;
 }
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 7918d5132c1f..4fd71b75e666 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -101,7 +101,7 @@ static int netvsc_open(struct net_device *net)
 		return ret;
 	}
 
-	netif_start_queue(net);
+	netif_tx_start_all_queues(net);
 
 	nvdev = hv_get_drvdata(device_obj);
 	rdev = nvdev->extension;
@@ -149,15 +149,100 @@ static void *init_ppi_data(struct rndis_message *msg, u32 ppi_size,
 	return ppi;
 }
 
+union sub_key {
+	u64 k;
+	struct {
+		u8 pad[3];
+		u8 kb;
+		u32 ka;
+	};
+};
+
+/* Toeplitz hash function
+ * data: network byte order
+ * return: host byte order
+ */
+static u32 comp_hash(u8 *key, int klen, u8 *data, int dlen)
+{
+	union sub_key subk;
+	int k_next = 4;
+	u8 dt;
+	int i, j;
+	u32 ret = 0;
+
+	subk.k = 0;
+	subk.ka = ntohl(*(u32 *)key);
+
+	for (i = 0; i < dlen; i++) {
+		subk.kb = key[k_next];
+		k_next = (k_next + 1) % klen;
+		dt = data[i];
+		for (j = 0; j < 8; j++) {
+			if (dt & 0x80)
+				ret ^= subk.ka;
+			dt <<= 1;
+			subk.k <<= 1;
+		}
+	}
+
+	return ret;
+}
+
+static bool netvsc_set_hash(u32 *hash, struct sk_buff *skb)
+{
+	struct iphdr *iphdr;
+	int data_len;
+	bool ret = false;
+
+	if (eth_hdr(skb)->h_proto != htons(ETH_P_IP))
+		return false;
+
+	iphdr = ip_hdr(skb);
+
+	if (iphdr->version == 4) {
+		if (iphdr->protocol == IPPROTO_TCP)
+			data_len = 12;
+		else
+			data_len = 8;
+		*hash = comp_hash(netvsc_hash_key, HASH_KEYLEN,
+				  (u8 *)&iphdr->saddr, data_len);
+		ret = true;
+	}
+
+	return ret;
+}
+
+static u16 netvsc_select_queue(struct net_device *ndev, struct sk_buff *skb,
+			void *accel_priv, select_queue_fallback_t fallback)
+{
+	struct net_device_context *net_device_ctx = netdev_priv(ndev);
+	struct hv_device *hdev =  net_device_ctx->device_ctx;
+	struct netvsc_device *nvsc_dev = hv_get_drvdata(hdev);
+	u32 hash;
+	u16 q_idx = 0;
+
+	if (nvsc_dev == NULL || ndev->real_num_tx_queues <= 1)
+		return 0;
+
+	if (netvsc_set_hash(&hash, skb)) {
+		q_idx = nvsc_dev->send_table[hash % VRSS_SEND_TAB_SIZE] %
+			ndev->real_num_tx_queues;
+		skb_set_hash(skb, hash, PKT_HASH_TYPE_L3);
+	}
+
+	return q_idx;
+}
+
 static void netvsc_xmit_completion(void *context)
 {
 	struct hv_netvsc_packet *packet = (struct hv_netvsc_packet *)context;
 	struct sk_buff *skb = (struct sk_buff *)
-		(unsigned long)packet->completion.send.send_completion_tid;
+		(unsigned long)packet->send_completion_tid;
+	u32 index = packet->send_buf_index;
 
 	kfree(packet);
 
-	if (skb)
+	if (skb && (index == NETVSC_INVALID_INDEX))
 		dev_kfree_skb_any(skb);
 }
 
@@ -301,6 +386,7 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 	struct ndis_tcp_lso_info *lso_info;
 	int  hdr_offset;
 	u32 net_trans_info;
+	u32 hash;
 
 
 	/* We will atmost need two pages to describe the rndis
@@ -319,9 +405,8 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 	packet = kzalloc(sizeof(struct hv_netvsc_packet) +
 			 (num_data_pgs * sizeof(struct hv_page_buffer)) +
 			 sizeof(struct rndis_message) +
-			 NDIS_VLAN_PPI_SIZE +
-			 NDIS_CSUM_PPI_SIZE +
-			 NDIS_LSO_PPI_SIZE, GFP_ATOMIC);
+			 NDIS_VLAN_PPI_SIZE + NDIS_CSUM_PPI_SIZE +
+			 NDIS_LSO_PPI_SIZE + NDIS_HASH_PPI_SIZE, GFP_ATOMIC);
 	if (!packet) {
 		/* out of memory, drop packet */
 		netdev_err(net, "unable to allocate hv_netvsc_packet\n");
@@ -333,6 +418,8 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 
 	packet->vlan_tci = skb->vlan_tci;
 
+	packet->q_idx = skb_get_queue_mapping(skb);
+
 	packet->is_data_pkt = true;
 	packet->total_data_buflen = skb->len;
 
@@ -341,9 +428,9 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 				(num_data_pgs * sizeof(struct hv_page_buffer)));
 
 	/* Set the completion routine */
-	packet->completion.send.send_completion = netvsc_xmit_completion;
-	packet->completion.send.send_completion_ctx = packet;
-	packet->completion.send.send_completion_tid = (unsigned long)skb;
+	packet->send_completion = netvsc_xmit_completion;
+	packet->send_completion_ctx = packet;
+	packet->send_completion_tid = (unsigned long)skb;
 
 	isvlan = packet->vlan_tci & VLAN_TAG_PRESENT;
 
@@ -358,6 +445,14 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 
 	rndis_msg_size = RNDIS_MESSAGE_SIZE(struct rndis_packet);
 
+	hash = skb_get_hash_raw(skb);
+	if (hash != 0 && net->real_num_tx_queues > 1) {
+		rndis_msg_size += NDIS_HASH_PPI_SIZE;
+		ppi = init_ppi_data(rndis_msg, NDIS_HASH_PPI_SIZE,
+				    NBL_HASH_VALUE);
+		*(u32 *)((void *)ppi + ppi->ppi_offset) = hash;
+	}
+
 	if (isvlan) {
 		struct ndis_pkt_8021q_info *vlan;
 
@@ -558,6 +653,9 @@ int netvsc_recv_callback(struct hv_device *device_obj,
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
 				       packet->vlan_tci);
 
+	skb_record_rx_queue(skb, packet->channel->
+			    offermsg.offer.sub_channel_index);
+
 	net->stats.rx_packets++;
 	net->stats.rx_bytes += packet->total_data_buflen;
 
@@ -606,7 +704,7 @@ static int netvsc_change_mtu(struct net_device *ndev, int mtu)
 	hv_set_drvdata(hdev, ndev);
 	device_info.ring_size = ring_size;
 	rndis_filter_device_add(hdev, &device_info);
-	netif_wake_queue(ndev);
+	netif_tx_wake_all_queues(ndev);
 
 	return 0;
 }
@@ -652,6 +750,7 @@ static const struct net_device_ops device_ops = {
 	.ndo_change_mtu =		netvsc_change_mtu,
 	.ndo_validate_addr =		eth_validate_addr,
 	.ndo_set_mac_address =		netvsc_set_mac_addr,
+	.ndo_select_queue =		netvsc_select_queue,
 };
 
 /*
@@ -698,9 +797,11 @@ static int netvsc_probe(struct hv_device *dev,
 	struct net_device *net = NULL;
 	struct net_device_context *net_device_ctx;
 	struct netvsc_device_info device_info;
+	struct netvsc_device *nvdev;
 	int ret;
 
-	net = alloc_etherdev(sizeof(struct net_device_context));
+	net = alloc_etherdev_mq(sizeof(struct net_device_context),
+				num_online_cpus());
 	if (!net)
 		return -ENOMEM;
 
@@ -719,7 +820,7 @@ static int netvsc_probe(struct hv_device *dev,
 	net->features = NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_SG | NETIF_F_RXCSUM |
 			NETIF_F_IP_CSUM | NETIF_F_TSO;
 
-	SET_ETHTOOL_OPS(net, &ethtool_ops);
+	net->ethtool_ops = &ethtool_ops;
 	SET_NETDEV_DEV(net, &dev->device);
 
 	/* Notify the netvsc driver of the new device */
@@ -733,6 +834,10 @@ static int netvsc_probe(struct hv_device *dev,
 	}
 	memcpy(net->dev_addr, device_info.mac_adr, ETH_ALEN);
 
+	nvdev = hv_get_drvdata(dev);
+	netif_set_real_num_tx_queues(net, nvdev->num_chn);
+	netif_set_real_num_rx_queues(net, nvdev->num_chn);
+
 	ret = register_netdev(net);
 	if (ret != 0) {
 		pr_err("Unable to register netdev.\n");
diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index 143a98caf618..99c527adae5b 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -31,7 +31,7 @@
 #include "hyperv_net.h"
 
 
-#define RNDIS_EXT_LEN 100
+#define RNDIS_EXT_LEN PAGE_SIZE
 struct rndis_request {
 	struct list_head list_ent;
 	struct completion  wait_event;
@@ -94,6 +94,8 @@ static struct rndis_request *get_rndis_request(struct rndis_device *dev,
 	rndis_msg->ndis_msg_type = msg_type;
 	rndis_msg->msg_len = msg_len;
 
+	request->pkt.q_idx = 0;
+
 	/*
 	 * Set the request id. This field is always after the rndis header for
 	 * request/response packet types so we just used the SetRequest as a
@@ -234,7 +236,7 @@ static int rndis_filter_send_request(struct rndis_device *dev,
 			packet->page_buf[0].len;
 	}
 
-	packet->completion.send.send_completion = NULL;
+	packet->send_completion = NULL;
 
 	ret = netvsc_send(dev->net_dev->dev, packet);
 	return ret;
@@ -399,8 +401,6 @@ static void rndis_filter_receive_data(struct rndis_device *dev,
 	pkt->total_data_buflen = rndis_pkt->data_len;
 	pkt->data = (void *)((unsigned long)pkt->data + data_offset);
 
-	pkt->is_data_pkt = true;
-
 	vlan = rndis_get_ppi(rndis_pkt, IEEE_8021Q_INFO);
 	if (vlan) {
 		pkt->vlan_tci = VLAN_TAG_PRESENT | vlan->vlanid |
@@ -509,6 +509,19 @@ static int rndis_filter_query_device(struct rndis_device *dev, u32 oid,
 	query->info_buflen = 0;
 	query->dev_vc_handle = 0;
 
+	if (oid == OID_GEN_RECEIVE_SCALE_CAPABILITIES) {
+		struct ndis_recv_scale_cap *cap;
+
+		request->request_msg.msg_len +=
+			sizeof(struct ndis_recv_scale_cap);
+		query->info_buflen = sizeof(struct ndis_recv_scale_cap);
+		cap = (struct ndis_recv_scale_cap *)((unsigned long)query +
+						     query->info_buf_offset);
+		cap->hdr.type = NDIS_OBJECT_TYPE_RSS_CAPABILITIES;
+		cap->hdr.rev = NDIS_RECEIVE_SCALE_CAPABILITIES_REVISION_2;
+		cap->hdr.size = sizeof(struct ndis_recv_scale_cap);
+	}
+
 	ret = rndis_filter_send_request(dev, request);
 	if (ret != 0)
 		goto cleanup;
@@ -695,6 +708,89 @@ cleanup:
 	return ret;
 }
 
+u8 netvsc_hash_key[HASH_KEYLEN] = {
+	0x6d, 0x5a, 0x56, 0xda, 0x25, 0x5b, 0x0e, 0xc2,
+	0x41, 0x67, 0x25, 0x3d, 0x43, 0xa3, 0x8f, 0xb0,
+	0xd0, 0xca, 0x2b, 0xcb, 0xae, 0x7b, 0x30, 0xb4,
+	0x77, 0xcb, 0x2d, 0xa3, 0x80, 0x30, 0xf2, 0x0c,
+	0x6a, 0x42, 0xb7, 0x3b, 0xbe, 0xac, 0x01, 0xfa
+};
+
+int rndis_filter_set_rss_param(struct rndis_device *rdev, int num_queue)
+{
+	struct net_device *ndev = rdev->net_dev->ndev;
+	struct rndis_request *request;
+	struct rndis_set_request *set;
+	struct rndis_set_complete *set_complete;
+	u32 extlen = sizeof(struct ndis_recv_scale_param) +
+		     4*ITAB_NUM + HASH_KEYLEN;
+	struct ndis_recv_scale_param *rssp;
+	u32 *itab;
+	u8 *keyp;
+	int i, t, ret;
+
+	request = get_rndis_request(
+			rdev, RNDIS_MSG_SET,
+			RNDIS_MESSAGE_SIZE(struct rndis_set_request) + extlen);
+	if (!request)
+		return -ENOMEM;
+
+	set = &request->request_msg.msg.set_req;
+	set->oid = OID_GEN_RECEIVE_SCALE_PARAMETERS;
+	set->info_buflen = extlen;
+	set->info_buf_offset = sizeof(struct rndis_set_request);
+	set->dev_vc_handle = 0;
+
+	rssp = (struct ndis_recv_scale_param *)(set + 1);
+	rssp->hdr.type = NDIS_OBJECT_TYPE_RSS_PARAMETERS;
+	rssp->hdr.rev = NDIS_RECEIVE_SCALE_PARAMETERS_REVISION_2;
+	rssp->hdr.size = sizeof(struct ndis_recv_scale_param);
+	rssp->flag = 0;
+	rssp->hashinfo = NDIS_HASH_FUNC_TOEPLITZ | NDIS_HASH_IPV4 |
+			 NDIS_HASH_TCP_IPV4;
+	rssp->indirect_tabsize = 4*ITAB_NUM;
+	rssp->indirect_taboffset = sizeof(struct ndis_recv_scale_param);
+	rssp->hashkey_size = HASH_KEYLEN;
+	rssp->kashkey_offset = rssp->indirect_taboffset +
+			       rssp->indirect_tabsize;
+
+	/* Set indirection table entries */
+	itab = (u32 *)(rssp + 1);
+	for (i = 0; i < ITAB_NUM; i++)
+		itab[i] = i % num_queue;
+
+	/* Set hask key values */
+	keyp = (u8 *)((unsigned long)rssp + rssp->kashkey_offset);
+	for (i = 0; i < HASH_KEYLEN; i++)
+		keyp[i] = netvsc_hash_key[i];
+
+
+	ret = rndis_filter_send_request(rdev, request);
+	if (ret != 0)
+		goto cleanup;
+
+	t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
+	if (t == 0) {
+		netdev_err(ndev, "timeout before we got a set response...\n");
+		/* can't put_rndis_request, since we may still receive a
+		 * send-completion.
+		 */
+		return -ETIMEDOUT;
+	} else {
+		set_complete = &request->response_msg.msg.set_complete;
+		if (set_complete->status != RNDIS_STATUS_SUCCESS) {
+			netdev_err(ndev, "Fail to set RSS parameters:0x%x\n",
+				   set_complete->status);
+			ret = -EINVAL;
+		}
+	}
+
+cleanup:
+	put_rndis_request(rdev, request);
+	return ret;
+}
+
+
 static int rndis_filter_query_device_link_status(struct rndis_device *dev)
 {
 	u32 size = sizeof(u32);
@@ -886,6 +982,28 @@ static int rndis_filter_close_device(struct rndis_device *dev)
 	return ret;
 }
 
+static void netvsc_sc_open(struct vmbus_channel *new_sc)
+{
+	struct netvsc_device *nvscdev;
+	u16 chn_index = new_sc->offermsg.offer.sub_channel_index;
+	int ret;
+
+	nvscdev = hv_get_drvdata(new_sc->primary_channel->device_obj);
+
+	if (chn_index >= nvscdev->num_chn)
+		return;
+
+	set_per_channel_state(new_sc, nvscdev->sub_cb_buf + (chn_index - 1) *
+			      NETVSC_PACKET_SIZE);
+
+	ret = vmbus_open(new_sc, nvscdev->ring_size * PAGE_SIZE,
+			 nvscdev->ring_size * PAGE_SIZE, NULL, 0,
+			 netvsc_channel_cb, new_sc);
+
+	if (ret == 0)
+		nvscdev->chn_table[chn_index] = new_sc;
+}
+
 int rndis_filter_device_add(struct hv_device *dev,
 				  void *additional_info)
 {
@@ -894,6 +1012,10 @@ int rndis_filter_device_add(struct hv_device *dev,
 	struct rndis_device *rndis_device;
 	struct netvsc_device_info *device_info = additional_info;
 	struct ndis_offload_params offloads;
+	struct nvsp_message *init_packet;
+	int t;
+	struct ndis_recv_scale_cap rsscap;
+	u32 rsscap_size = sizeof(struct ndis_recv_scale_cap);
 
 	rndis_device = get_rndis_device();
 	if (!rndis_device)
@@ -913,6 +1035,7 @@ int rndis_filter_device_add(struct hv_device *dev,
 
 	/* Initialize the rndis device */
 	net_device = hv_get_drvdata(dev);
+	net_device->num_chn = 1;
 
 	net_device->extension = rndis_device;
 	rndis_device->net_dev = net_device;
@@ -952,7 +1075,6 @@ int rndis_filter_device_add(struct hv_device *dev,
 	if (ret)
 		goto err_dev_remv;
 
-
 	rndis_filter_query_device_link_status(rndis_device);
 
 	device_info->link_state = rndis_device->link_state;
@@ -961,7 +1083,66 @@ int rndis_filter_device_add(struct hv_device *dev,
 		 rndis_device->hw_mac_adr,
 		 device_info->link_state ? "down" : "up");
 
-	return ret;
+	if (net_device->nvsp_version < NVSP_PROTOCOL_VERSION_5)
+		return 0;
+
+	/* vRSS setup */
+	memset(&rsscap, 0, rsscap_size);
+	ret = rndis_filter_query_device(rndis_device,
+					OID_GEN_RECEIVE_SCALE_CAPABILITIES,
+					&rsscap, &rsscap_size);
+	if (ret || rsscap.num_recv_que < 2)
+		goto out;
+
+	net_device->num_chn = (num_online_cpus() < rsscap.num_recv_que) ?
+			       num_online_cpus() : rsscap.num_recv_que;
+	if (net_device->num_chn == 1)
+		goto out;
+
+	net_device->sub_cb_buf = vzalloc((net_device->num_chn - 1) *
+					 NETVSC_PACKET_SIZE);
+	if (!net_device->sub_cb_buf) {
+		net_device->num_chn = 1;
+		dev_info(&dev->device, "No memory for subchannels.\n");
+		goto out;
+	}
+
+	vmbus_set_sc_create_callback(dev->channel, netvsc_sc_open);
+
+	init_packet = &net_device->channel_init_pkt;
+	memset(init_packet, 0, sizeof(struct nvsp_message));
+	init_packet->hdr.msg_type = NVSP_MSG5_TYPE_SUBCHANNEL;
+	init_packet->msg.v5_msg.subchn_req.op = NVSP_SUBCHANNEL_ALLOCATE;
+	init_packet->msg.v5_msg.subchn_req.num_subchannels =
+						net_device->num_chn - 1;
+	ret = vmbus_sendpacket(dev->channel, init_packet,
+			       sizeof(struct nvsp_message),
+			       (unsigned long)init_packet,
+			       VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (ret)
+		goto out;
+	t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ);
+	if (t == 0) {
+		ret = -ETIMEDOUT;
+		goto out;
+	}
+	if (init_packet->msg.v5_msg.subchn_comp.status !=
+	    NVSP_STAT_SUCCESS) {
+		ret = -ENODEV;
+		goto out;
+	}
+	net_device->num_chn = 1 +
+		init_packet->msg.v5_msg.subchn_comp.num_subchannels;
+
+	vmbus_are_subchannels_present(dev->channel);
+
+	ret = rndis_filter_set_rss_param(rndis_device, net_device->num_chn);
+
+out:
+	if (ret)
+		net_device->num_chn = 1;
+	return 0; /* return 0 because primary channel can be used alone */
 
 err_dev_remv:
 	rndis_filter_device_remove(dev);
diff --git a/drivers/net/ieee802154/at86rf230.c b/drivers/net/ieee802154/at86rf230.c
index e36f194673a4..50899416f668 100644
--- a/drivers/net/ieee802154/at86rf230.c
+++ b/drivers/net/ieee802154/at86rf230.c
@@ -23,6 +23,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
+#include <linux/irq.h>
 #include <linux/gpio.h>
 #include <linux/delay.h>
 #include <linux/mutex.h>
@@ -692,10 +693,7 @@ at86rf230_xmit(struct ieee802154_dev *dev, struct sk_buff *skb)
 	if (rc < 0)
 		goto err_rx;
 
-	rc = at86rf230_start(dev);
-
-	return rc;
-
+	return at86rf230_start(dev);
 err_rx:
 	at86rf230_start(dev);
 err:
@@ -963,33 +961,24 @@ static irqreturn_t at86rf230_isr_level(int irq, void *data)
 	return at86rf230_isr(irq, data);
 }
 
-static int at86rf230_irq_polarity(struct at86rf230_local *lp, int pol)
-{
-	return at86rf230_write_subreg(lp, SR_IRQ_POLARITY, pol);
-}
-
 static int at86rf230_hw_init(struct at86rf230_local *lp)
 {
-	struct at86rf230_platform_data *pdata = lp->spi->dev.platform_data;
-	int rc, irq_pol;
-	u8 status;
+	int rc, irq_pol, irq_type;
+	u8 dvdd;
 	u8 csma_seed[2];
 
-	rc = at86rf230_read_subreg(lp, SR_TRX_STATUS, &status);
-	if (rc)
-		return rc;
-
 	rc = at86rf230_write_subreg(lp, SR_TRX_CMD, STATE_FORCE_TRX_OFF);
 	if (rc)
 		return rc;
 
+	irq_type = irq_get_trigger_type(lp->spi->irq);
 	/* configure irq polarity, defaults to high active */
-	if (pdata->irq_type & (IRQF_TRIGGER_FALLING | IRQF_TRIGGER_LOW))
+	if (irq_type & (IRQF_TRIGGER_FALLING | IRQF_TRIGGER_LOW))
 		irq_pol = IRQ_ACTIVE_LOW;
 	else
 		irq_pol = IRQ_ACTIVE_HIGH;
 
-	rc = at86rf230_irq_polarity(lp, irq_pol);
+	rc = at86rf230_write_subreg(lp, SR_IRQ_POLARITY, irq_pol);
 	if (rc)
 		return rc;
 
@@ -1017,10 +1006,10 @@ static int at86rf230_hw_init(struct at86rf230_local *lp)
 	/* Wait the next SLEEP cycle */
 	msleep(100);
 
-	rc = at86rf230_read_subreg(lp, SR_DVDD_OK, &status);
+	rc = at86rf230_read_subreg(lp, SR_DVDD_OK, &dvdd);
 	if (rc)
 		return rc;
-	if (!status) {
+	if (!dvdd) {
 		dev_err(&lp->spi->dev, "DVDD error\n");
 		return -EINVAL;
 	}
@@ -1032,7 +1021,6 @@ static struct at86rf230_platform_data *
 at86rf230_get_pdata(struct spi_device *spi)
 {
 	struct at86rf230_platform_data *pdata;
-	const char *irq_type;
 
 	if (!IS_ENABLED(CONFIG_OF) || !spi->dev.of_node)
 		return spi->dev.platform_data;
@@ -1044,19 +1032,6 @@ at86rf230_get_pdata(struct spi_device *spi)
 	pdata->rstn = of_get_named_gpio(spi->dev.of_node, "reset-gpio", 0);
 	pdata->slp_tr = of_get_named_gpio(spi->dev.of_node, "sleep-gpio", 0);
 
-	pdata->irq_type = IRQF_TRIGGER_RISING;
-	of_property_read_string(spi->dev.of_node, "irq-type", &irq_type);
-	if (!strcmp(irq_type, "level-high"))
-		pdata->irq_type = IRQF_TRIGGER_HIGH;
-	else if (!strcmp(irq_type, "level-low"))
-		pdata->irq_type = IRQF_TRIGGER_LOW;
-	else if (!strcmp(irq_type, "edge-rising"))
-		pdata->irq_type = IRQF_TRIGGER_RISING;
-	else if (!strcmp(irq_type, "edge-falling"))
-		pdata->irq_type = IRQF_TRIGGER_FALLING;
-	else
-		dev_warn(&spi->dev, "wrong irq-type specified using edge-rising\n");
-
 	spi->dev.platform_data = pdata;
 done:
 	return pdata;
@@ -1071,7 +1046,7 @@ static int at86rf230_probe(struct spi_device *spi)
 	u8 part = 0, version = 0, status;
 	irq_handler_t irq_handler;
 	work_func_t irq_worker;
-	int rc;
+	int rc, irq_type;
 	const char *chip;
 	struct ieee802154_ops *ops = NULL;
 
@@ -1087,27 +1062,17 @@ static int at86rf230_probe(struct spi_device *spi)
 	}
 
 	if (gpio_is_valid(pdata->rstn)) {
-		rc = gpio_request(pdata->rstn, "rstn");
+		rc = devm_gpio_request_one(&spi->dev, pdata->rstn,
+					   GPIOF_OUT_INIT_HIGH, "rstn");
 		if (rc)
 			return rc;
 	}
 
 	if (gpio_is_valid(pdata->slp_tr)) {
-		rc = gpio_request(pdata->slp_tr, "slp_tr");
-		if (rc)
-			goto err_slp_tr;
-	}
-
-	if (gpio_is_valid(pdata->rstn)) {
-		rc = gpio_direction_output(pdata->rstn, 1);
-		if (rc)
-			goto err_gpio_dir;
-	}
-
-	if (gpio_is_valid(pdata->slp_tr)) {
-		rc = gpio_direction_output(pdata->slp_tr, 0);
+		rc = devm_gpio_request_one(&spi->dev, pdata->slp_tr,
+					   GPIOF_OUT_INIT_LOW, "slp_tr");
 		if (rc)
-			goto err_gpio_dir;
+			return rc;
 	}
 
 	/* Reset */
@@ -1121,13 +1086,12 @@ static int at86rf230_probe(struct spi_device *spi)
 
 	rc = __at86rf230_detect_device(spi, &man_id, &part, &version);
 	if (rc < 0)
-		goto err_gpio_dir;
+		return rc;
 
 	if (man_id != 0x001f) {
 		dev_err(&spi->dev, "Non-Atmel dev found (MAN_ID %02x %02x)\n",
 			man_id >> 8, man_id & 0xFF);
-		rc = -EINVAL;
-		goto err_gpio_dir;
+		return -EINVAL;
 	}
 
 	switch (part) {
@@ -1154,16 +1118,12 @@ static int at86rf230_probe(struct spi_device *spi)
 	}
 
 	dev_info(&spi->dev, "Detected %s chip version %d\n", chip, version);
-	if (!ops) {
-		rc = -ENOTSUPP;
-		goto err_gpio_dir;
-	}
+	if (!ops)
+		return -ENOTSUPP;
 
 	dev = ieee802154_alloc_device(sizeof(*lp), ops);
-	if (!dev) {
-		rc = -ENOMEM;
-		goto err_gpio_dir;
-	}
+	if (!dev)
+		return -ENOMEM;
 
 	lp = dev->priv;
 	lp->dev = dev;
@@ -1176,7 +1136,10 @@ static int at86rf230_probe(struct spi_device *spi)
 	dev->extra_tx_headroom = 0;
 	dev->flags = IEEE802154_HW_OMIT_CKSUM | IEEE802154_HW_AACK;
 
-	if (pdata->irq_type & (IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING)) {
+	irq_type = irq_get_trigger_type(spi->irq);
+	if (!irq_type)
+		irq_type = IRQF_TRIGGER_RISING;
+	if (irq_type & (IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING)) {
 		irq_worker = at86rf230_irqwork;
 		irq_handler = at86rf230_isr;
 	} else {
@@ -1202,75 +1165,66 @@ static int at86rf230_probe(struct spi_device *spi)
 	if (rc)
 		goto err_hw_init;
 
-	rc = request_irq(spi->irq, irq_handler,
-			 IRQF_SHARED | pdata->irq_type,
-			 dev_name(&spi->dev), lp);
+	/* Read irq status register to reset irq line */
+	rc = at86rf230_read_subreg(lp, RG_IRQ_STATUS, 0xff, 0, &status);
 	if (rc)
 		goto err_hw_init;
 
-	/* Read irq status register to reset irq line */
-	rc = at86rf230_read_subreg(lp, RG_IRQ_STATUS, 0xff, 0, &status);
+	rc = devm_request_irq(&spi->dev, spi->irq, irq_handler,
+			      IRQF_SHARED | irq_type,
+			      dev_name(&spi->dev), lp);
 	if (rc)
-		goto err_irq;
+		goto err_hw_init;
 
 	rc = ieee802154_register_device(lp->dev);
 	if (rc)
-		goto err_irq;
+		goto err_hw_init;
 
 	return rc;
 
-err_irq:
-	free_irq(spi->irq, lp);
 err_hw_init:
 	flush_work(&lp->irqwork);
-	spi_set_drvdata(spi, NULL);
 	mutex_destroy(&lp->bmux);
 	ieee802154_free_device(lp->dev);
 
-err_gpio_dir:
-	if (gpio_is_valid(pdata->slp_tr))
-		gpio_free(pdata->slp_tr);
-err_slp_tr:
-	if (gpio_is_valid(pdata->rstn))
-		gpio_free(pdata->rstn);
 	return rc;
 }
 
 static int at86rf230_remove(struct spi_device *spi)
 {
 	struct at86rf230_local *lp = spi_get_drvdata(spi);
-	struct at86rf230_platform_data *pdata = spi->dev.platform_data;
 
 	/* mask all at86rf230 irq's */
 	at86rf230_write_subreg(lp, SR_IRQ_MASK, 0);
 	ieee802154_unregister_device(lp->dev);
-
-	free_irq(spi->irq, lp);
 	flush_work(&lp->irqwork);
-
-	if (gpio_is_valid(pdata->slp_tr))
-		gpio_free(pdata->slp_tr);
-	if (gpio_is_valid(pdata->rstn))
-		gpio_free(pdata->rstn);
-
 	mutex_destroy(&lp->bmux);
 	ieee802154_free_device(lp->dev);
-
 	dev_dbg(&spi->dev, "unregistered at86rf230\n");
+
 	return 0;
 }
 
-#if IS_ENABLED(CONFIG_OF)
-static struct of_device_id at86rf230_of_match[] = {
+static const struct of_device_id at86rf230_of_match[] = {
 	{ .compatible = "atmel,at86rf230", },
 	{ .compatible = "atmel,at86rf231", },
 	{ .compatible = "atmel,at86rf233", },
 	{ .compatible = "atmel,at86rf212", },
 	{ },
 };
-#endif
+MODULE_DEVICE_TABLE(of, at86rf230_of_match);
+
+static const struct spi_device_id at86rf230_device_id[] = {
+	{ .name = "at86rf230", },
+	{ .name = "at86rf231", },
+	{ .name = "at86rf233", },
+	{ .name = "at86rf212", },
+	{ },
+};
+MODULE_DEVICE_TABLE(spi, at86rf230_device_id);
 
 static struct spi_driver at86rf230_driver = {
+	.id_table = at86rf230_device_id,
 	.driver = {
 		.of_match_table = of_match_ptr(at86rf230_of_match),
 		.name	= "at86rf230",
diff --git a/drivers/net/ieee802154/fakelb.c b/drivers/net/ieee802154/fakelb.c
index b8d22173925d..27d83207d24c 100644
--- a/drivers/net/ieee802154/fakelb.c
+++ b/drivers/net/ieee802154/fakelb.c
@@ -26,6 +26,7 @@
 #include <linux/timer.h>
 #include <linux/platform_device.h>
 #include <linux/netdevice.h>
+#include <linux/device.h>
 #include <linux/spinlock.h>
 #include <net/mac802154.h>
 #include <net/wpan-phy.h>
@@ -228,7 +229,8 @@ static int fakelb_probe(struct platform_device *pdev)
 	int err = -ENOMEM;
 	int i;
 
-	priv = kzalloc(sizeof(struct fakelb_priv), GFP_KERNEL);
+	priv = devm_kzalloc(&pdev->dev, sizeof(struct fakelb_priv),
+			    GFP_KERNEL);
 	if (!priv)
 		goto err_alloc;
 
@@ -248,7 +250,6 @@ static int fakelb_probe(struct platform_device *pdev)
 err_slave:
 	list_for_each_entry(dp, &priv->list, list)
 		fakelb_del(dp);
-	kfree(priv);
 err_alloc:
 	return err;
 }
@@ -260,7 +261,6 @@ static int fakelb_remove(struct platform_device *pdev)
 
 	list_for_each_entry_safe(dp, temp, &priv->list, list)
 		fakelb_del(dp);
-	kfree(priv);
 
 	return 0;
 }
diff --git a/drivers/net/ieee802154/mrf24j40.c b/drivers/net/ieee802154/mrf24j40.c
index 78a6552ed707..4048062011ba 100644
--- a/drivers/net/ieee802154/mrf24j40.c
+++ b/drivers/net/ieee802154/mrf24j40.c
@@ -618,12 +618,12 @@ static int mrf24j40_probe(struct spi_device *spi)
 
 	printk(KERN_INFO "mrf24j40: probe(). IRQ: %d\n", spi->irq);
 
-	devrec = kzalloc(sizeof(struct mrf24j40), GFP_KERNEL);
+	devrec = devm_kzalloc(&spi->dev, sizeof(struct mrf24j40), GFP_KERNEL);
 	if (!devrec)
-		goto err_devrec;
-	devrec->buf = kzalloc(3, GFP_KERNEL);
+		goto err_ret;
+	devrec->buf = devm_kzalloc(&spi->dev, 3, GFP_KERNEL);
 	if (!devrec->buf)
-		goto err_buf;
+		goto err_ret;
 
 	spi->mode = SPI_MODE_0; /* TODO: Is this appropriate for right here? */
 	if (spi->max_speed_hz > MAX_SPI_SPEED_HZ)
@@ -638,7 +638,7 @@ static int mrf24j40_probe(struct spi_device *spi)
 
 	devrec->dev = ieee802154_alloc_device(0, &mrf24j40_ops);
 	if (!devrec->dev)
-		goto err_alloc_dev;
+		goto err_ret;
 
 	devrec->dev->priv = devrec;
 	devrec->dev->parent = &devrec->spi->dev;
@@ -676,12 +676,13 @@ static int mrf24j40_probe(struct spi_device *spi)
 	val &= ~0x3; /* Clear RX mode (normal) */
 	write_short_reg(devrec, REG_RXMCR, val);
 
-	ret = request_threaded_irq(spi->irq,
-				   NULL,
-				   mrf24j40_isr,
-				   IRQF_TRIGGER_LOW|IRQF_ONESHOT,
-				   dev_name(&spi->dev),
-				   devrec);
+	ret = devm_request_threaded_irq(&spi->dev,
+					spi->irq,
+					NULL,
+					mrf24j40_isr,
+					IRQF_TRIGGER_LOW|IRQF_ONESHOT,
+					dev_name(&spi->dev),
+					devrec);
 
 	if (ret) {
 		dev_err(printdev(devrec), "Unable to get IRQ");
@@ -695,11 +696,7 @@ err_read_reg:
 	ieee802154_unregister_device(devrec->dev);
 err_register_device:
 	ieee802154_free_device(devrec->dev);
-err_alloc_dev:
-	kfree(devrec->buf);
-err_buf:
-	kfree(devrec);
-err_devrec:
+err_ret:
 	return ret;
 }
 
@@ -709,15 +706,11 @@ static int mrf24j40_remove(struct spi_device *spi)
 
 	dev_dbg(printdev(devrec), "remove\n");
 
-	free_irq(spi->irq, devrec);
 	ieee802154_unregister_device(devrec->dev);
 	ieee802154_free_device(devrec->dev);
 	/* TODO: Will ieee802154_free_device() wait until ->xmit() is
 	 * complete? */
 
-	/* Clean up the SPI stuff. */
-	kfree(devrec->buf);
-	kfree(devrec);
 	return 0;
 }
 
diff --git a/drivers/net/irda/Kconfig b/drivers/net/irda/Kconfig
index 3da44d5d9149..8d101d63abca 100644
--- a/drivers/net/irda/Kconfig
+++ b/drivers/net/irda/Kconfig
@@ -396,7 +396,8 @@ config MCS_FIR
 
 config SH_IRDA
 	tristate "SuperH IrDA driver"
-	depends on IRDA && ARCH_SHMOBILE
+	depends on IRDA
+	depends on ARCH_SHMOBILE || COMPILE_TEST
 	help
 	  Say Y here if your want to enable SuperH IrDA devices.
 
diff --git a/drivers/net/irda/via-ircc.c b/drivers/net/irda/via-ircc.c
index 2900af091c2d..998bb89ede71 100644
--- a/drivers/net/irda/via-ircc.c
+++ b/drivers/net/irda/via-ircc.c
@@ -510,10 +510,8 @@ static void via_hw_init(struct via_ircc_cb *self)
  */
 static int via_ircc_read_dongle_id(int iobase)
 {
-	int dongle_id = 9;	/* Default to IBM */
-
 	IRDA_ERROR("via-ircc: dongle probing not supported, please specify dongle_id module parameter.\n");
-	return dongle_id;
+	return 9;	/* Default to IBM */
 }
 
 /*
@@ -926,7 +924,6 @@ static int via_ircc_dma_xmit(struct via_ircc_cb *self, u16 iobase)
 static int via_ircc_dma_xmit_complete(struct via_ircc_cb *self)
 {
 	int iobase;
-	int ret = TRUE;
 	u8 Tx_status;
 
 	IRDA_DEBUG(3, "%s()\n", __func__);
@@ -983,7 +980,7 @@ F01_E*/
 	// Tell the network layer, that we can accept more frames 
 	netif_wake_queue(self->netdev);
 //F01   }
-	return ret;
+	return TRUE;
 }
 
 /*
diff --git a/drivers/net/irda/w83977af_ir.c b/drivers/net/irda/w83977af_ir.c
index e641bb240362..11dbdf36d9c1 100644
--- a/drivers/net/irda/w83977af_ir.c
+++ b/drivers/net/irda/w83977af_ir.c
@@ -62,10 +62,6 @@
 #include "w83977af.h"
 #include "w83977af_ir.h"
 
-#ifdef  CONFIG_ARCH_NETWINDER            /* Adjust to NetWinder differences */
-#undef  CONFIG_NETWINDER_TX_DMA_PROBLEMS /* Not needed */
-#define CONFIG_NETWINDER_RX_DMA_PROBLEMS /* Must have this one! */
-#endif
 #define CONFIG_USE_W977_PNP        /* Currently needed */
 #define PIO_MAX_SPEED       115200 
 
@@ -332,7 +328,7 @@ static int w83977af_probe(int iobase, int irq, int dma)
  		w977_write_reg(0x74, dma+1, efbase[i]);
 #else
  		w977_write_reg(0x74, dma, efbase[i]);   
-#endif /*CONFIG_ARCH_NETWINDER */
+#endif /* CONFIG_ARCH_NETWINDER */
  		w977_write_reg(0x75, 0x04, efbase[i]);  /* Disable Tx DMA */
   	
  		/* Set append hardware CRC, enable IR bank selection */	
@@ -563,10 +559,6 @@ static netdev_tx_t w83977af_hard_xmit(struct sk_buff *skb,
 static void w83977af_dma_write(struct w83977af_ir *self, int iobase)
 {
 	__u8 set;
-#ifdef CONFIG_NETWINDER_TX_DMA_PROBLEMS
-	unsigned long flags;
-	__u8 hcr;
-#endif
         IRDA_DEBUG(4, "%s(), len=%d\n", __func__ , self->tx_buff.len);
 
 	/* Save current set */
@@ -579,30 +571,13 @@ static void w83977af_dma_write(struct w83977af_ir *self, int iobase)
 	/* Choose transmit DMA channel  */ 
 	switch_bank(iobase, SET2);
 	outb(ADCR1_D_CHSW|/*ADCR1_DMA_F|*/ADCR1_ADV_SL, iobase+ADCR1);
-#ifdef CONFIG_NETWINDER_TX_DMA_PROBLEMS
-	spin_lock_irqsave(&self->lock, flags);
-
-	disable_dma(self->io.dma);
-	clear_dma_ff(self->io.dma);
-	set_dma_mode(self->io.dma, DMA_MODE_READ);
-	set_dma_addr(self->io.dma, self->tx_buff_dma);
-	set_dma_count(self->io.dma, self->tx_buff.len);
-#else
 	irda_setup_dma(self->io.dma, self->tx_buff_dma, self->tx_buff.len,
 		       DMA_MODE_WRITE);	
-#endif
 	self->io.direction = IO_XMIT;
 	
 	/* Enable DMA */
  	switch_bank(iobase, SET0);
-#ifdef CONFIG_NETWINDER_TX_DMA_PROBLEMS
-	hcr = inb(iobase+HCR);
-	outb(hcr | HCR_EN_DMA, iobase+HCR);
-	enable_dma(self->io.dma);
-	spin_unlock_irqrestore(&self->lock, flags);
-#else	
 	outb(inb(iobase+HCR) | HCR_EN_DMA | HCR_TX_WT, iobase+HCR);
-#endif
 
 	/* Restore set register */
 	outb(set, iobase+SSR);
@@ -711,7 +686,7 @@ static int w83977af_dma_receive(struct w83977af_ir *self)
 {
 	int iobase;
 	__u8 set;
-#ifdef CONFIG_NETWINDER_RX_DMA_PROBLEMS
+#ifdef CONFIG_ARCH_NETWINDER
 	unsigned long flags;
 	__u8 hcr;
 #endif
@@ -736,7 +711,7 @@ static int w83977af_dma_receive(struct w83977af_ir *self)
 	self->io.direction = IO_RECV;
 	self->rx_buff.data = self->rx_buff.head;
 
-#ifdef CONFIG_NETWINDER_RX_DMA_PROBLEMS
+#ifdef CONFIG_ARCH_NETWINDER
 	spin_lock_irqsave(&self->lock, flags);
 
 	disable_dma(self->io.dma);
@@ -759,7 +734,7 @@ static int w83977af_dma_receive(struct w83977af_ir *self)
 	
 	/* Enable DMA */
 	switch_bank(iobase, SET0);
-#ifdef CONFIG_NETWINDER_RX_DMA_PROBLEMS
+#ifdef CONFIG_ARCH_NETWINDER
 	hcr = inb(iobase+HCR);
 	outb(hcr | HCR_EN_DMA, iobase+HCR);
 	enable_dma(self->io.dma);
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index d53e299ae1d9..958df383068a 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -30,8 +30,10 @@
 #include <linux/if_link.h>
 #include <linux/if_macvlan.h>
 #include <linux/hash.h>
+#include <linux/workqueue.h>
 #include <net/rtnetlink.h>
 #include <net/xfrm.h>
+#include <linux/netpoll.h>
 
 #define MACVLAN_HASH_SIZE	(1 << BITS_PER_BYTE)
 
@@ -40,10 +42,19 @@ struct macvlan_port {
 	struct hlist_head	vlan_hash[MACVLAN_HASH_SIZE];
 	struct list_head	vlans;
 	struct rcu_head		rcu;
+	struct sk_buff_head	bc_queue;
+	struct work_struct	bc_work;
 	bool 			passthru;
-	int			count;
 };
 
+#define MACVLAN_PORT_IS_EMPTY(port)    list_empty(&port->vlans)
+
+struct macvlan_skb_cb {
+	const struct macvlan_dev *src;
+};
+
+#define MACVLAN_SKB_CB(__skb) ((struct macvlan_skb_cb *)&((__skb)->cb[0]))
+
 static void macvlan_port_destroy(struct net_device *dev);
 
 static struct macvlan_port *macvlan_port_get_rcu(const struct net_device *dev)
@@ -120,7 +131,7 @@ static int macvlan_broadcast_one(struct sk_buff *skb,
 	struct net_device *dev = vlan->dev;
 
 	if (local)
-		return dev_forward_skb(dev, skb);
+		return __dev_forward_skb(dev, skb);
 
 	skb->dev = dev;
 	if (ether_addr_equal_64bits(eth->h_dest, dev->broadcast))
@@ -128,7 +139,7 @@ static int macvlan_broadcast_one(struct sk_buff *skb,
 	else
 		skb->pkt_type = PACKET_MULTICAST;
 
-	return netif_rx(skb);
+	return 0;
 }
 
 static u32 macvlan_hash_mix(const struct macvlan_dev *vlan)
@@ -175,32 +186,32 @@ static void macvlan_broadcast(struct sk_buff *skb,
 			if (likely(nskb))
 				err = macvlan_broadcast_one(
 					nskb, vlan, eth,
-					mode == MACVLAN_MODE_BRIDGE);
+					mode == MACVLAN_MODE_BRIDGE) ?:
+				      netif_rx_ni(nskb);
 			macvlan_count_rx(vlan, skb->len + ETH_HLEN,
 					 err == NET_RX_SUCCESS, 1);
 		}
 	}
 }
 
-/* called under rcu_read_lock() from netif_receive_skb */
-static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb)
+static void macvlan_process_broadcast(struct work_struct *w)
 {
-	struct macvlan_port *port;
-	struct sk_buff *skb = *pskb;
-	const struct ethhdr *eth = eth_hdr(skb);
-	const struct macvlan_dev *vlan;
-	const struct macvlan_dev *src;
-	struct net_device *dev;
-	unsigned int len = 0;
-	int ret = NET_RX_DROP;
+	struct macvlan_port *port = container_of(w, struct macvlan_port,
+						 bc_work);
+	struct sk_buff *skb;
+	struct sk_buff_head list;
+
+	skb_queue_head_init(&list);
+
+	spin_lock_bh(&port->bc_queue.lock);
+	skb_queue_splice_tail_init(&port->bc_queue, &list);
+	spin_unlock_bh(&port->bc_queue.lock);
+
+	while ((skb = __skb_dequeue(&list))) {
+		const struct macvlan_dev *src = MACVLAN_SKB_CB(skb)->src;
+
+		rcu_read_lock();
 
-	port = macvlan_port_get_rcu(skb->dev);
-	if (is_multicast_ether_addr(eth->h_dest)) {
-		skb = ip_check_defrag(skb, IP_DEFRAG_MACVLAN);
-		if (!skb)
-			return RX_HANDLER_CONSUMED;
-		eth = eth_hdr(skb);
-		src = macvlan_hash_lookup(port, eth->h_source);
 		if (!src)
 			/* frame comes from an external address */
 			macvlan_broadcast(skb, port, NULL,
@@ -213,20 +224,80 @@ static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb)
 			macvlan_broadcast(skb, port, src->dev,
 					  MACVLAN_MODE_VEPA |
 					  MACVLAN_MODE_BRIDGE);
-		else if (src->mode == MACVLAN_MODE_BRIDGE)
+		else
 			/*
 			 * flood only to VEPA ports, bridge ports
 			 * already saw the frame on the way out.
 			 */
 			macvlan_broadcast(skb, port, src->dev,
 					  MACVLAN_MODE_VEPA);
-		else {
+
+		rcu_read_unlock();
+
+		kfree_skb(skb);
+	}
+}
+
+static void macvlan_broadcast_enqueue(struct macvlan_port *port,
+				      struct sk_buff *skb)
+{
+	struct sk_buff *nskb;
+	int err = -ENOMEM;
+
+	nskb = skb_clone(skb, GFP_ATOMIC);
+	if (!nskb)
+		goto err;
+
+	spin_lock(&port->bc_queue.lock);
+	if (skb_queue_len(&port->bc_queue) < skb->dev->tx_queue_len) {
+		__skb_queue_tail(&port->bc_queue, nskb);
+		err = 0;
+	}
+	spin_unlock(&port->bc_queue.lock);
+
+	if (err)
+		goto free_nskb;
+
+	schedule_work(&port->bc_work);
+	return;
+
+free_nskb:
+	kfree_skb(nskb);
+err:
+	atomic_long_inc(&skb->dev->rx_dropped);
+}
+
+/* called under rcu_read_lock() from netif_receive_skb */
+static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb)
+{
+	struct macvlan_port *port;
+	struct sk_buff *skb = *pskb;
+	const struct ethhdr *eth = eth_hdr(skb);
+	const struct macvlan_dev *vlan;
+	const struct macvlan_dev *src;
+	struct net_device *dev;
+	unsigned int len = 0;
+	int ret = NET_RX_DROP;
+
+	port = macvlan_port_get_rcu(skb->dev);
+	if (is_multicast_ether_addr(eth->h_dest)) {
+		skb = ip_check_defrag(skb, IP_DEFRAG_MACVLAN);
+		if (!skb)
+			return RX_HANDLER_CONSUMED;
+		eth = eth_hdr(skb);
+		src = macvlan_hash_lookup(port, eth->h_source);
+		if (src && src->mode != MACVLAN_MODE_VEPA &&
+		    src->mode != MACVLAN_MODE_BRIDGE) {
 			/* forward to original port. */
 			vlan = src;
-			ret = macvlan_broadcast_one(skb, vlan, eth, 0);
+			ret = macvlan_broadcast_one(skb, vlan, eth, 0) ?:
+			      netif_rx(skb);
 			goto out;
 		}
 
+		MACVLAN_SKB_CB(skb)->src = src;
+		macvlan_broadcast_enqueue(port, skb);
+
 		return RX_HANDLER_PASS;
 	}
 
@@ -287,12 +358,26 @@ xmit_world:
 	return dev_queue_xmit(skb);
 }
 
+static inline netdev_tx_t macvlan_netpoll_send_skb(struct macvlan_dev *vlan, struct sk_buff *skb)
+{
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	if (vlan->netpoll)
+		netpoll_send_skb(vlan->netpoll, skb);
+#else
+	BUG();
+#endif
+	return NETDEV_TX_OK;
+}
+
 static netdev_tx_t macvlan_start_xmit(struct sk_buff *skb,
 				      struct net_device *dev)
 {
 	unsigned int len = skb->len;
 	int ret;
-	const struct macvlan_dev *vlan = netdev_priv(dev);
+	struct macvlan_dev *vlan = netdev_priv(dev);
+
+	if (unlikely(netpoll_tx_running(dev)))
+		return macvlan_netpoll_send_skb(vlan, skb);
 
 	if (vlan->fwd_priv) {
 		skb->dev = vlan->lowerdev;
@@ -424,35 +509,49 @@ hash_del:
 	return 0;
 }
 
-static int macvlan_set_mac_address(struct net_device *dev, void *p)
+static int macvlan_sync_address(struct net_device *dev, unsigned char *addr)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
 	struct net_device *lowerdev = vlan->lowerdev;
-	struct sockaddr *addr = p;
 	int err;
 
-	if (!is_valid_ether_addr(addr->sa_data))
-		return -EADDRNOTAVAIL;
-
 	if (!(dev->flags & IFF_UP)) {
 		/* Just copy in the new address */
-		memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
+		ether_addr_copy(dev->dev_addr, addr);
 	} else {
 		/* Rehash and update the device filters */
-		if (macvlan_addr_busy(vlan->port, addr->sa_data))
+		if (macvlan_addr_busy(vlan->port, addr))
 			return -EBUSY;
 
-		err = dev_uc_add(lowerdev, addr->sa_data);
-		if (err)
-			return err;
+		if (!vlan->port->passthru) {
+			err = dev_uc_add(lowerdev, addr);
+			if (err)
+				return err;
 
-		dev_uc_del(lowerdev, dev->dev_addr);
+			dev_uc_del(lowerdev, dev->dev_addr);
+		}
 
-		macvlan_hash_change_addr(vlan, addr->sa_data);
+		macvlan_hash_change_addr(vlan, addr);
 	}
 	return 0;
 }
 
+static int macvlan_set_mac_address(struct net_device *dev, void *p)
+{
+	struct macvlan_dev *vlan = netdev_priv(dev);
+	struct sockaddr *addr = p;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (vlan->mode == MACVLAN_MODE_PASSTHRU) {
+		dev_set_mac_address(vlan->lowerdev, addr);
+		return 0;
+	}
+
+	return macvlan_sync_address(dev, addr->sa_data);
+}
+
 static void macvlan_change_rx_flags(struct net_device *dev, int change)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
@@ -567,8 +666,7 @@ static void macvlan_uninit(struct net_device *dev)
 
 	free_percpu(vlan->pcpu_stats);
 
-	port->count -= 1;
-	if (!port->count)
+	if (MACVLAN_PORT_IS_EMPTY(port))
 		macvlan_port_destroy(port->dev);
 }
 
@@ -705,6 +803,50 @@ static netdev_features_t macvlan_fix_features(struct net_device *dev,
 	return features;
 }
 
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void macvlan_dev_poll_controller(struct net_device *dev)
+{
+	return;
+}
+
+static int macvlan_dev_netpoll_setup(struct net_device *dev, struct netpoll_info *npinfo)
+{
+	struct macvlan_dev *vlan = netdev_priv(dev);
+	struct net_device *real_dev = vlan->lowerdev;
+	struct netpoll *netpoll;
+	int err = 0;
+
+	netpoll = kzalloc(sizeof(*netpoll), GFP_KERNEL);
+	err = -ENOMEM;
+	if (!netpoll)
+		goto out;
+
+	err = __netpoll_setup(netpoll, real_dev);
+	if (err) {
+		kfree(netpoll);
+		goto out;
+	}
+
+	vlan->netpoll = netpoll;
+
+out:
+	return err;
+}
+
+static void macvlan_dev_netpoll_cleanup(struct net_device *dev)
+{
+	struct macvlan_dev *vlan = netdev_priv(dev);
+	struct netpoll *netpoll = vlan->netpoll;
+
+	if (!netpoll)
+		return;
+
+	vlan->netpoll = NULL;
+
+	__netpoll_free_async(netpoll);
+}
+#endif	/* CONFIG_NET_POLL_CONTROLLER */
+
 static const struct ethtool_ops macvlan_ethtool_ops = {
 	.get_link		= ethtool_op_get_link,
 	.get_settings		= macvlan_ethtool_get_settings,
@@ -730,6 +872,11 @@ static const struct net_device_ops macvlan_netdev_ops = {
 	.ndo_fdb_del		= macvlan_fdb_del,
 	.ndo_fdb_dump		= ndo_dflt_fdb_dump,
 	.ndo_get_lock_subclass  = macvlan_get_nest_level,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= macvlan_dev_poll_controller,
+	.ndo_netpoll_setup	= macvlan_dev_netpoll_setup,
+	.ndo_netpoll_cleanup	= macvlan_dev_netpoll_cleanup,
+#endif
 };
 
 void macvlan_common_setup(struct net_device *dev)
@@ -770,6 +917,9 @@ static int macvlan_port_create(struct net_device *dev)
 	for (i = 0; i < MACVLAN_HASH_SIZE; i++)
 		INIT_HLIST_HEAD(&port->vlan_hash[i]);
 
+	skb_queue_head_init(&port->bc_queue);
+	INIT_WORK(&port->bc_work, macvlan_process_broadcast);
+
 	err = netdev_rx_handler_register(dev, macvlan_handle_frame, port);
 	if (err)
 		kfree(port);
@@ -782,6 +932,7 @@ static void macvlan_port_destroy(struct net_device *dev)
 {
 	struct macvlan_port *port = macvlan_port_get_rtnl(dev);
 
+	cancel_work_sync(&port->bc_work);
 	dev->priv_flags &= ~IFF_MACVLAN_PORT;
 	netdev_rx_handler_unregister(dev);
 	kfree_rcu(port, rcu);
@@ -868,13 +1019,12 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev,
 		vlan->flags = nla_get_u16(data[IFLA_MACVLAN_FLAGS]);
 
 	if (vlan->mode == MACVLAN_MODE_PASSTHRU) {
-		if (port->count)
+		if (!MACVLAN_PORT_IS_EMPTY(port))
 			return -EINVAL;
 		port->passthru = true;
 		eth_hw_addr_inherit(dev, lowerdev);
 	}
 
-	port->count += 1;
 	err = register_netdevice(dev);
 	if (err < 0)
 		goto destroy_port;
@@ -892,8 +1042,7 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev,
 unregister_netdev:
 	unregister_netdevice(dev);
 destroy_port:
-	port->count -= 1;
-	if (!port->count)
+	if (MACVLAN_PORT_IS_EMPTY(port))
 		macvlan_port_destroy(lowerdev);
 
 	return err;
@@ -1028,6 +1177,25 @@ static int macvlan_device_event(struct notifier_block *unused,
 			netdev_update_features(vlan->dev);
 		}
 		break;
+	case NETDEV_CHANGEMTU:
+		list_for_each_entry(vlan, &port->vlans, list) {
+			if (vlan->dev->mtu <= dev->mtu)
+				continue;
+			dev_set_mtu(vlan->dev, dev->mtu);
+		}
+		break;
+	case NETDEV_CHANGEADDR:
+		if (!port->passthru)
+			return NOTIFY_DONE;
+
+		vlan = list_first_entry_or_null(&port->vlans,
+						struct macvlan_dev,
+						list);
+
+		if (macvlan_sync_address(vlan->dev, dev->dev_addr))
+			return NOTIFY_BAD;
+
+		break;
 	case NETDEV_UNREGISTER:
 		/* twiddle thumbs on netns device moves */
 		if (dev->reg_state != NETREG_UNREGISTERING)
@@ -1036,11 +1204,17 @@ static int macvlan_device_event(struct notifier_block *unused,
 		list_for_each_entry_safe(vlan, next, &port->vlans, list)
 			vlan->dev->rtnl_link_ops->dellink(vlan->dev, &list_kill);
 		unregister_netdevice_many(&list_kill);
-		list_del(&list_kill);
 		break;
 	case NETDEV_PRE_TYPE_CHANGE:
 		/* Forbid underlaying device to change its type. */
 		return NOTIFY_BAD;
+
+	case NETDEV_NOTIFY_PEERS:
+	case NETDEV_BONDING_FAILOVER:
+	case NETDEV_RESEND_IGMP:
+		/* Propagate to all vlans */
+		list_for_each_entry(vlan, &port->vlans, list)
+			call_netdevice_notifiers(event, vlan->dev);
 	}
 	return NOTIFY_DONE;
 }
diff --git a/drivers/net/ntb_netdev.c b/drivers/net/ntb_netdev.c
index 63aa9d9e34c5..5a7e6397440a 100644
--- a/drivers/net/ntb_netdev.c
+++ b/drivers/net/ntb_netdev.c
@@ -298,7 +298,6 @@ static int ntb_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 {
 	cmd->supported = SUPPORTED_Backplane;
 	cmd->advertising = ADVERTISED_Backplane;
-	cmd->speed = SPEED_UNKNOWN;
 	ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
 	cmd->duplex = DUPLEX_FULL;
 	cmd->port = PORT_OTHER;
@@ -348,7 +347,7 @@ static int ntb_netdev_probe(struct pci_dev *pdev)
 	memcpy(ndev->dev_addr, ndev->perm_addr, ndev->addr_len);
 
 	ndev->netdev_ops = &ntb_netdev_ops;
-	SET_ETHTOOL_OPS(ndev, &ntb_ethtool_ops);
+	ndev->ethtool_ops = &ntb_ethtool_ops;
 
 	dev->qp = ntb_transport_create_queue(ndev, pdev, &ntb_netdev_handlers);
 	if (!dev->qp) {
diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 6a17f92153b3..65de0cab8d07 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -24,6 +24,12 @@ config AMD_PHY
 	---help---
 	  Currently supports the am79c874
 
+config AMD_XGBE_PHY
+	tristate "Driver for the AMD 10GbE (amd-xgbe) PHYs"
+	depends on OF
+	---help---
+	  Currently supports the AMD 10GbE PHY
+
 config MARVELL_PHY
 	tristate "Drivers for Marvell PHYs"
 	---help---
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index 07d24024863e..7dc3d5b304cf 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -33,3 +33,4 @@ obj-$(CONFIG_MDIO_BUS_MUX_GPIO)	+= mdio-mux-gpio.o
 obj-$(CONFIG_MDIO_BUS_MUX_MMIOREG) += mdio-mux-mmioreg.o
 obj-$(CONFIG_MDIO_SUN4I)	+= mdio-sun4i.o
 obj-$(CONFIG_MDIO_MOXART)	+= mdio-moxart.o
+obj-$(CONFIG_AMD_XGBE_PHY)	+= amd-xgbe-phy.o
diff --git a/drivers/net/phy/amd-xgbe-phy.c b/drivers/net/phy/amd-xgbe-phy.c
new file mode 100644
index 000000000000..b57c22442867
--- /dev/null
+++ b/drivers/net/phy/amd-xgbe-phy.c
@@ -0,0 +1,1357 @@
+/*
+ * AMD 10Gb Ethernet PHY driver
+ *
+ * This file is available to you under your choice of the following two
+ * licenses:
+ *
+ * License 1: GPLv2
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ *
+ * This file is free software; you may copy, redistribute and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ *
+ * License 2: Modified BSD
+ *
+ * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Advanced Micro Devices, Inc. nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/unistd.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/skbuff.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/mii.h>
+#include <linux/ethtool.h>
+#include <linux/phy.h>
+#include <linux/mdio.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/of_device.h>
+#include <linux/uaccess.h>
+#include <asm/irq.h>
+
+
+MODULE_AUTHOR("Tom Lendacky <thomas.lendacky@amd.com>");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_VERSION("1.0.0-a");
+MODULE_DESCRIPTION("AMD 10GbE (amd-xgbe) PHY driver");
+
+#define XGBE_PHY_ID	0x000162d0
+#define XGBE_PHY_MASK	0xfffffff0
+
+#define XGBE_AN_INT_CMPLT		0x01
+#define XGBE_AN_INC_LINK		0x02
+#define XGBE_AN_PG_RCV			0x04
+
+#define XNP_MCF_NULL_MESSAGE		0x001
+#define XNP_ACK_PROCESSED		(1 << 12)
+#define XNP_MP_FORMATTED		(1 << 13)
+#define XNP_NP_EXCHANGE			(1 << 15)
+
+#ifndef MDIO_PMA_10GBR_PMD_CTRL
+#define MDIO_PMA_10GBR_PMD_CTRL		0x0096
+#endif
+#ifndef MDIO_PMA_10GBR_FEC_CTRL
+#define MDIO_PMA_10GBR_FEC_CTRL		0x00ab
+#endif
+#ifndef MDIO_AN_XNP
+#define MDIO_AN_XNP			0x0016
+#endif
+
+#ifndef MDIO_AN_INTMASK
+#define MDIO_AN_INTMASK			0x8001
+#endif
+#ifndef MDIO_AN_INT
+#define MDIO_AN_INT			0x8002
+#endif
+
+#ifndef MDIO_CTRL1_SPEED1G
+#define MDIO_CTRL1_SPEED1G		(MDIO_CTRL1_SPEED10G & ~BMCR_SPEED100)
+#endif
+
+/* SerDes integration register offsets */
+#define SIR0_STATUS			0x0040
+#define SIR1_SPEED			0x0000
+
+/* SerDes integration register entry bit positions and sizes */
+#define SIR0_STATUS_RX_READY_INDEX	0
+#define SIR0_STATUS_RX_READY_WIDTH	1
+#define SIR0_STATUS_TX_READY_INDEX	8
+#define SIR0_STATUS_TX_READY_WIDTH	1
+#define SIR1_SPEED_DATARATE_INDEX	4
+#define SIR1_SPEED_DATARATE_WIDTH	2
+#define SIR1_SPEED_PI_SPD_SEL_INDEX	12
+#define SIR1_SPEED_PI_SPD_SEL_WIDTH	4
+#define SIR1_SPEED_PLLSEL_INDEX		3
+#define SIR1_SPEED_PLLSEL_WIDTH		1
+#define SIR1_SPEED_RATECHANGE_INDEX	6
+#define SIR1_SPEED_RATECHANGE_WIDTH	1
+#define SIR1_SPEED_TXAMP_INDEX		8
+#define SIR1_SPEED_TXAMP_WIDTH		4
+#define SIR1_SPEED_WORDMODE_INDEX	0
+#define SIR1_SPEED_WORDMODE_WIDTH	3
+
+#define SPEED_10000_CDR			0x7
+#define SPEED_10000_PLL			0x1
+#define SPEED_10000_RATE		0x0
+#define SPEED_10000_TXAMP		0xa
+#define SPEED_10000_WORD		0x7
+
+#define SPEED_2500_CDR			0x2
+#define SPEED_2500_PLL			0x0
+#define SPEED_2500_RATE			0x2
+#define SPEED_2500_TXAMP		0xf
+#define SPEED_2500_WORD			0x1
+
+#define SPEED_1000_CDR			0x2
+#define SPEED_1000_PLL			0x0
+#define SPEED_1000_RATE			0x3
+#define SPEED_1000_TXAMP		0xf
+#define SPEED_1000_WORD			0x1
+
+
+/* SerDes RxTx register offsets */
+#define RXTX_REG20			0x0050
+#define RXTX_REG114			0x01c8
+
+/* SerDes RxTx register entry bit positions and sizes */
+#define RXTX_REG20_BLWC_ENA_INDEX	2
+#define RXTX_REG20_BLWC_ENA_WIDTH	1
+#define RXTX_REG114_PQ_REG_INDEX	9
+#define RXTX_REG114_PQ_REG_WIDTH	7
+
+#define RXTX_10000_BLWC			0
+#define RXTX_10000_PQ			0x1e
+
+#define RXTX_2500_BLWC			1
+#define RXTX_2500_PQ			0xa
+
+#define RXTX_1000_BLWC			1
+#define RXTX_1000_PQ			0xa
+
+/* Bit setting and getting macros
+ *  The get macro will extract the current bit field value from within
+ *  the variable
+ *
+ *  The set macro will clear the current bit field value within the
+ *  variable and then set the bit field of the variable to the
+ *  specified value
+ */
+#define GET_BITS(_var, _index, _width)					\
+	(((_var) >> (_index)) & ((0x1 << (_width)) - 1))
+
+#define SET_BITS(_var, _index, _width, _val)				\
+do {									\
+	(_var) &= ~(((0x1 << (_width)) - 1) << (_index));		\
+	(_var) |= (((_val) & ((0x1 << (_width)) - 1)) << (_index));	\
+} while (0)
+
+/* Macros for reading or writing SerDes integration registers
+ *  The ioread macros will get bit fields or full values using the
+ *  register definitions formed using the input names
+ *
+ *  The iowrite macros will set bit fields or full values using the
+ *  register definitions formed using the input names
+ */
+#define XSIR0_IOREAD(_priv, _reg)					\
+	ioread16((_priv)->sir0_regs + _reg)
+
+#define XSIR0_IOREAD_BITS(_priv, _reg, _field)				\
+	GET_BITS(XSIR0_IOREAD((_priv), _reg),				\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH)
+
+#define XSIR0_IOWRITE(_priv, _reg, _val)				\
+	iowrite16((_val), (_priv)->sir0_regs + _reg)
+
+#define XSIR0_IOWRITE_BITS(_priv, _reg, _field, _val)			\
+do {									\
+	u16 reg_val = XSIR0_IOREAD((_priv), _reg);			\
+	SET_BITS(reg_val,						\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH, (_val));			\
+	XSIR0_IOWRITE((_priv), _reg, reg_val);				\
+} while (0)
+
+#define XSIR1_IOREAD(_priv, _reg)					\
+	ioread16((_priv)->sir1_regs + _reg)
+
+#define XSIR1_IOREAD_BITS(_priv, _reg, _field)				\
+	GET_BITS(XSIR1_IOREAD((_priv), _reg),				\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH)
+
+#define XSIR1_IOWRITE(_priv, _reg, _val)				\
+	iowrite16((_val), (_priv)->sir1_regs + _reg)
+
+#define XSIR1_IOWRITE_BITS(_priv, _reg, _field, _val)			\
+do {									\
+	u16 reg_val = XSIR1_IOREAD((_priv), _reg);			\
+	SET_BITS(reg_val,						\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH, (_val));			\
+	XSIR1_IOWRITE((_priv), _reg, reg_val);				\
+} while (0)
+
+
+/* Macros for reading or writing SerDes RxTx registers
+ *  The ioread macros will get bit fields or full values using the
+ *  register definitions formed using the input names
+ *
+ *  The iowrite macros will set bit fields or full values using the
+ *  register definitions formed using the input names
+ */
+#define XRXTX_IOREAD(_priv, _reg)					\
+	ioread16((_priv)->rxtx_regs + _reg)
+
+#define XRXTX_IOREAD_BITS(_priv, _reg, _field)				\
+	GET_BITS(XRXTX_IOREAD((_priv), _reg),				\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH)
+
+#define XRXTX_IOWRITE(_priv, _reg, _val)				\
+	iowrite16((_val), (_priv)->rxtx_regs + _reg)
+
+#define XRXTX_IOWRITE_BITS(_priv, _reg, _field, _val)			\
+do {									\
+	u16 reg_val = XRXTX_IOREAD((_priv), _reg);			\
+	SET_BITS(reg_val,						\
+		 _reg##_##_field##_INDEX,				\
+		 _reg##_##_field##_WIDTH, (_val));			\
+	XRXTX_IOWRITE((_priv), _reg, reg_val);				\
+} while (0)
+
+
+enum amd_xgbe_phy_an {
+	AMD_XGBE_AN_READY = 0,
+	AMD_XGBE_AN_START,
+	AMD_XGBE_AN_EVENT,
+	AMD_XGBE_AN_PAGE_RECEIVED,
+	AMD_XGBE_AN_INCOMPAT_LINK,
+	AMD_XGBE_AN_COMPLETE,
+	AMD_XGBE_AN_NO_LINK,
+	AMD_XGBE_AN_EXIT,
+	AMD_XGBE_AN_ERROR,
+};
+
+enum amd_xgbe_phy_rx {
+	AMD_XGBE_RX_READY = 0,
+	AMD_XGBE_RX_BPA,
+	AMD_XGBE_RX_XNP,
+	AMD_XGBE_RX_COMPLETE,
+};
+
+enum amd_xgbe_phy_mode {
+	AMD_XGBE_MODE_KR,
+	AMD_XGBE_MODE_KX,
+};
+
+struct amd_xgbe_phy_priv {
+	struct platform_device *pdev;
+	struct device *dev;
+
+	struct phy_device *phydev;
+
+	/* SerDes related mmio resources */
+	struct resource *rxtx_res;
+	struct resource *sir0_res;
+	struct resource *sir1_res;
+
+	/* SerDes related mmio registers */
+	void __iomem *rxtx_regs;	/* SerDes Rx/Tx CSRs */
+	void __iomem *sir0_regs;	/* SerDes integration registers (1/2) */
+	void __iomem *sir1_regs;	/* SerDes integration registers (2/2) */
+
+	/* Maintain link status for re-starting auto-negotiation */
+	unsigned int link;
+	enum amd_xgbe_phy_mode mode;
+
+	/* Auto-negotiation state machine support */
+	struct mutex an_mutex;
+	enum amd_xgbe_phy_an an_result;
+	enum amd_xgbe_phy_an an_state;
+	enum amd_xgbe_phy_rx kr_state;
+	enum amd_xgbe_phy_rx kx_state;
+	struct work_struct an_work;
+	struct workqueue_struct *an_workqueue;
+};
+
+static int amd_xgbe_an_enable_kr_training(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
+	if (ret < 0)
+		return ret;
+
+	ret |= 0x02;
+	phy_write_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, ret);
+
+	return 0;
+}
+
+static int amd_xgbe_an_disable_kr_training(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~0x02;
+	phy_write_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, ret);
+
+	return 0;
+}
+
+static int amd_xgbe_phy_pcs_power_cycle(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		return ret;
+
+	ret |= MDIO_CTRL1_LPOWER;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	usleep_range(75, 100);
+
+	ret &= ~MDIO_CTRL1_LPOWER;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	return 0;
+}
+
+static void amd_xgbe_phy_serdes_start_ratechange(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+
+	/* Assert Rx and Tx ratechange */
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, RATECHANGE, 1);
+}
+
+static void amd_xgbe_phy_serdes_complete_ratechange(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+
+	/* Release Rx and Tx ratechange */
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, RATECHANGE, 0);
+
+	/* Wait for Rx and Tx ready */
+	while (!XSIR0_IOREAD_BITS(priv, SIR0_STATUS, RX_READY) &&
+	       !XSIR0_IOREAD_BITS(priv, SIR0_STATUS, TX_READY))
+		usleep_range(10, 20);
+}
+
+static int amd_xgbe_phy_xgmii_mode(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* Enable KR training */
+	ret = amd_xgbe_an_enable_kr_training(phydev);
+	if (ret < 0)
+		return ret;
+
+	/* Set PCS to KR/10G speed */
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_PCS_CTRL2_TYPE;
+	ret |= MDIO_PCS_CTRL2_10GBR;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2, ret);
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_CTRL1_SPEEDSEL;
+	ret |= MDIO_CTRL1_SPEED10G;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	ret = amd_xgbe_phy_pcs_power_cycle(phydev);
+	if (ret < 0)
+		return ret;
+
+	/* Set SerDes to 10G speed */
+	amd_xgbe_phy_serdes_start_ratechange(phydev);
+
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, DATARATE, SPEED_10000_RATE);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, WORDMODE, SPEED_10000_WORD);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, TXAMP, SPEED_10000_TXAMP);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, PLLSEL, SPEED_10000_PLL);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, PI_SPD_SEL, SPEED_10000_CDR);
+
+	XRXTX_IOWRITE_BITS(priv, RXTX_REG20, BLWC_ENA, RXTX_10000_BLWC);
+	XRXTX_IOWRITE_BITS(priv, RXTX_REG114, PQ_REG, RXTX_10000_PQ);
+
+	amd_xgbe_phy_serdes_complete_ratechange(phydev);
+
+	priv->mode = AMD_XGBE_MODE_KR;
+
+	return 0;
+}
+
+static int amd_xgbe_phy_gmii_2500_mode(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* Disable KR training */
+	ret = amd_xgbe_an_disable_kr_training(phydev);
+	if (ret < 0)
+		return ret;
+
+	/* Set PCS to KX/1G speed */
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_PCS_CTRL2_TYPE;
+	ret |= MDIO_PCS_CTRL2_10GBX;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2, ret);
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_CTRL1_SPEEDSEL;
+	ret |= MDIO_CTRL1_SPEED1G;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	ret = amd_xgbe_phy_pcs_power_cycle(phydev);
+	if (ret < 0)
+		return ret;
+
+	/* Set SerDes to 2.5G speed */
+	amd_xgbe_phy_serdes_start_ratechange(phydev);
+
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, DATARATE, SPEED_2500_RATE);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, WORDMODE, SPEED_2500_WORD);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, TXAMP, SPEED_2500_TXAMP);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, PLLSEL, SPEED_2500_PLL);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, PI_SPD_SEL, SPEED_2500_CDR);
+
+	XRXTX_IOWRITE_BITS(priv, RXTX_REG20, BLWC_ENA, RXTX_2500_BLWC);
+	XRXTX_IOWRITE_BITS(priv, RXTX_REG114, PQ_REG, RXTX_2500_PQ);
+
+	amd_xgbe_phy_serdes_complete_ratechange(phydev);
+
+	priv->mode = AMD_XGBE_MODE_KX;
+
+	return 0;
+}
+
+static int amd_xgbe_phy_gmii_mode(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* Disable KR training */
+	ret = amd_xgbe_an_disable_kr_training(phydev);
+	if (ret < 0)
+		return ret;
+
+	/* Set PCS to KX/1G speed */
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_PCS_CTRL2_TYPE;
+	ret |= MDIO_PCS_CTRL2_10GBX;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2, ret);
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_CTRL1_SPEEDSEL;
+	ret |= MDIO_CTRL1_SPEED1G;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	ret = amd_xgbe_phy_pcs_power_cycle(phydev);
+	if (ret < 0)
+		return ret;
+
+	/* Set SerDes to 1G speed */
+	amd_xgbe_phy_serdes_start_ratechange(phydev);
+
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, DATARATE, SPEED_1000_RATE);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, WORDMODE, SPEED_1000_WORD);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, TXAMP, SPEED_1000_TXAMP);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, PLLSEL, SPEED_1000_PLL);
+	XSIR1_IOWRITE_BITS(priv, SIR1_SPEED, PI_SPD_SEL, SPEED_1000_CDR);
+
+	XRXTX_IOWRITE_BITS(priv, RXTX_REG20, BLWC_ENA, RXTX_1000_BLWC);
+	XRXTX_IOWRITE_BITS(priv, RXTX_REG114, PQ_REG, RXTX_1000_PQ);
+
+	amd_xgbe_phy_serdes_complete_ratechange(phydev);
+
+	priv->mode = AMD_XGBE_MODE_KX;
+
+	return 0;
+}
+
+static int amd_xgbe_phy_switch_mode(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* If we are in KR switch to KX, and vice-versa */
+	if (priv->mode == AMD_XGBE_MODE_KR)
+		ret = amd_xgbe_phy_gmii_mode(phydev);
+	else
+		ret = amd_xgbe_phy_xgmii_mode(phydev);
+
+	return ret;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_switch_mode(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = amd_xgbe_phy_switch_mode(phydev);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	return AMD_XGBE_AN_START;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_tx_training(struct phy_device *phydev,
+						    enum amd_xgbe_phy_rx *state)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	int ad_reg, lp_reg, ret;
+
+	*state = AMD_XGBE_RX_COMPLETE;
+
+	/* If we're in KX mode then we're done */
+	if (priv->mode == AMD_XGBE_MODE_KX)
+		return AMD_XGBE_AN_EVENT;
+
+	/* Enable/Disable FEC */
+	ad_reg = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2);
+	if (ad_reg < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	lp_reg = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_LPA + 2);
+	if (lp_reg < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_FEC_CTRL);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	if ((ad_reg & 0xc000) && (lp_reg & 0xc000))
+		ret |= 0x01;
+	else
+		ret &= ~0x01;
+
+	phy_write_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_FEC_CTRL, ret);
+
+	/* Start KR training */
+	ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	ret |= 0x01;
+	phy_write_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, ret);
+
+	return AMD_XGBE_AN_EVENT;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_tx_xnp(struct phy_device *phydev,
+					       enum amd_xgbe_phy_rx *state)
+{
+	u16 msg;
+
+	*state = AMD_XGBE_RX_XNP;
+
+	msg = XNP_MCF_NULL_MESSAGE;
+	msg |= XNP_MP_FORMATTED;
+
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_XNP + 2, 0);
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_XNP + 1, 0);
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_XNP, msg);
+
+	return AMD_XGBE_AN_EVENT;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_rx_bpa(struct phy_device *phydev,
+					       enum amd_xgbe_phy_rx *state)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	unsigned int link_support;
+	int ret, ad_reg, lp_reg;
+
+	/* Read Base Ability register 2 first */
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_LPA + 1);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	/* Check for a supported mode, otherwise restart in a different one */
+	link_support = (priv->mode == AMD_XGBE_MODE_KR) ? 0x80 : 0x20;
+	if (!(ret & link_support))
+		return amd_xgbe_an_switch_mode(phydev);
+
+	/* Check Extended Next Page support */
+	ad_reg = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
+	if (ad_reg < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	lp_reg = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_LPA);
+	if (lp_reg < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	return ((ad_reg & XNP_NP_EXCHANGE) || (lp_reg & XNP_NP_EXCHANGE)) ?
+	       amd_xgbe_an_tx_xnp(phydev, state) :
+	       amd_xgbe_an_tx_training(phydev, state);
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_rx_xnp(struct phy_device *phydev,
+					       enum amd_xgbe_phy_rx *state)
+{
+	int ad_reg, lp_reg;
+
+	/* Check Extended Next Page support */
+	ad_reg = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
+	if (ad_reg < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	lp_reg = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_LPA);
+	if (lp_reg < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	return ((ad_reg & XNP_NP_EXCHANGE) || (lp_reg & XNP_NP_EXCHANGE)) ?
+	       amd_xgbe_an_tx_xnp(phydev, state) :
+	       amd_xgbe_an_tx_training(phydev, state);
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_start(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* Be sure we aren't looping trying to negotiate */
+	if (priv->mode == AMD_XGBE_MODE_KR) {
+		if (priv->kr_state != AMD_XGBE_RX_READY)
+			return AMD_XGBE_AN_NO_LINK;
+		priv->kr_state = AMD_XGBE_RX_BPA;
+	} else {
+		if (priv->kx_state != AMD_XGBE_RX_READY)
+			return AMD_XGBE_AN_NO_LINK;
+		priv->kx_state = AMD_XGBE_RX_BPA;
+	}
+
+	/* Set up Advertisement register 3 first */
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	if (phydev->supported & SUPPORTED_10000baseR_FEC)
+		ret |= 0xc000;
+	else
+		ret &= ~0xc000;
+
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2, ret);
+
+	/* Set up Advertisement register 2 next */
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	if (phydev->supported & SUPPORTED_10000baseKR_Full)
+		ret |= 0x80;
+	else
+		ret &= ~0x80;
+
+	if (phydev->supported & SUPPORTED_1000baseKX_Full)
+		ret |= 0x20;
+	else
+		ret &= ~0x20;
+
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1, ret);
+
+	/* Set up Advertisement register 1 last */
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	if (phydev->supported & SUPPORTED_Pause)
+		ret |= 0x400;
+	else
+		ret &= ~0x400;
+
+	if (phydev->supported & SUPPORTED_Asym_Pause)
+		ret |= 0x800;
+	else
+		ret &= ~0x800;
+
+	/* We don't intend to perform XNP */
+	ret &= ~XNP_NP_EXCHANGE;
+
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE, ret);
+
+	/* Enable and start auto-negotiation */
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_INT, 0);
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_CTRL1);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	ret |= MDIO_AN_CTRL1_ENABLE;
+	ret |= MDIO_AN_CTRL1_RESTART;
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_CTRL1, ret);
+
+	return AMD_XGBE_AN_EVENT;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_event(struct phy_device *phydev)
+{
+	enum amd_xgbe_phy_an new_state;
+	int ret;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_INT);
+	if (ret < 0)
+		return AMD_XGBE_AN_ERROR;
+
+	new_state = AMD_XGBE_AN_EVENT;
+	if (ret & XGBE_AN_PG_RCV)
+		new_state = AMD_XGBE_AN_PAGE_RECEIVED;
+	else if (ret & XGBE_AN_INC_LINK)
+		new_state = AMD_XGBE_AN_INCOMPAT_LINK;
+	else if (ret & XGBE_AN_INT_CMPLT)
+		new_state = AMD_XGBE_AN_COMPLETE;
+
+	if (new_state != AMD_XGBE_AN_EVENT)
+		phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_INT, 0);
+
+	return new_state;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_page_received(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	enum amd_xgbe_phy_rx *state;
+	int ret;
+
+	state = (priv->mode == AMD_XGBE_MODE_KR) ? &priv->kr_state
+						 : &priv->kx_state;
+
+	switch (*state) {
+	case AMD_XGBE_RX_BPA:
+		ret = amd_xgbe_an_rx_bpa(phydev, state);
+		break;
+
+	case AMD_XGBE_RX_XNP:
+		ret = amd_xgbe_an_rx_xnp(phydev, state);
+		break;
+
+	default:
+		ret = AMD_XGBE_AN_ERROR;
+	}
+
+	return ret;
+}
+
+static enum amd_xgbe_phy_an amd_xgbe_an_incompat_link(struct phy_device *phydev)
+{
+	return amd_xgbe_an_switch_mode(phydev);
+}
+
+static void amd_xgbe_an_state_machine(struct work_struct *work)
+{
+	struct amd_xgbe_phy_priv *priv = container_of(work,
+						      struct amd_xgbe_phy_priv,
+						      an_work);
+	struct phy_device *phydev = priv->phydev;
+	enum amd_xgbe_phy_an cur_state;
+	int sleep;
+
+	while (1) {
+		mutex_lock(&priv->an_mutex);
+
+		cur_state = priv->an_state;
+
+		switch (priv->an_state) {
+		case AMD_XGBE_AN_START:
+			priv->an_state = amd_xgbe_an_start(phydev);
+			break;
+
+		case AMD_XGBE_AN_EVENT:
+			priv->an_state = amd_xgbe_an_event(phydev);
+			break;
+
+		case AMD_XGBE_AN_PAGE_RECEIVED:
+			priv->an_state = amd_xgbe_an_page_received(phydev);
+			break;
+
+		case AMD_XGBE_AN_INCOMPAT_LINK:
+			priv->an_state = amd_xgbe_an_incompat_link(phydev);
+			break;
+
+		case AMD_XGBE_AN_COMPLETE:
+		case AMD_XGBE_AN_NO_LINK:
+		case AMD_XGBE_AN_EXIT:
+			goto exit_unlock;
+
+		default:
+			priv->an_state = AMD_XGBE_AN_ERROR;
+		}
+
+		if (priv->an_state == AMD_XGBE_AN_ERROR) {
+			netdev_err(phydev->attached_dev,
+				   "error during auto-negotiation, state=%u\n",
+				   cur_state);
+			goto exit_unlock;
+		}
+
+		sleep = (priv->an_state == AMD_XGBE_AN_EVENT) ? 1 : 0;
+
+		mutex_unlock(&priv->an_mutex);
+
+		if (sleep)
+			usleep_range(20, 50);
+	}
+
+exit_unlock:
+	priv->an_result = priv->an_state;
+	priv->an_state = AMD_XGBE_AN_READY;
+
+	mutex_unlock(&priv->an_mutex);
+}
+
+static int amd_xgbe_phy_soft_reset(struct phy_device *phydev)
+{
+	int count, ret;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		return ret;
+
+	ret |= MDIO_CTRL1_RESET;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	count = 50;
+	do {
+		msleep(20);
+		ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+		if (ret < 0)
+			return ret;
+	} while ((ret & MDIO_CTRL1_RESET) && --count);
+
+	if (ret & MDIO_CTRL1_RESET)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+static int amd_xgbe_phy_config_init(struct phy_device *phydev)
+{
+	/* Initialize supported features */
+	phydev->supported = SUPPORTED_Autoneg;
+	phydev->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
+	phydev->supported |= SUPPORTED_Backplane;
+	phydev->supported |= SUPPORTED_1000baseKX_Full |
+			     SUPPORTED_2500baseX_Full;
+	phydev->supported |= SUPPORTED_10000baseKR_Full |
+			     SUPPORTED_10000baseR_FEC;
+	phydev->advertising = phydev->supported;
+
+	/* Turn off and clear interrupts */
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_INTMASK, 0);
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_INT, 0);
+
+	return 0;
+}
+
+static int amd_xgbe_phy_setup_forced(struct phy_device *phydev)
+{
+	int ret;
+
+	/* Disable auto-negotiation */
+	ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_CTRL1);
+	if (ret < 0)
+		return ret;
+
+	ret &= ~MDIO_AN_CTRL1_ENABLE;
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_CTRL1, ret);
+
+	/* Validate/Set specified speed */
+	switch (phydev->speed) {
+	case SPEED_10000:
+		ret = amd_xgbe_phy_xgmii_mode(phydev);
+		break;
+
+	case SPEED_2500:
+		ret = amd_xgbe_phy_gmii_2500_mode(phydev);
+		break;
+
+	case SPEED_1000:
+		ret = amd_xgbe_phy_gmii_mode(phydev);
+		break;
+
+	default:
+		ret = -EINVAL;
+	}
+
+	if (ret < 0)
+		return ret;
+
+	/* Validate duplex mode */
+	if (phydev->duplex != DUPLEX_FULL)
+		return -EINVAL;
+
+	phydev->pause = 0;
+	phydev->asym_pause = 0;
+
+	return 0;
+}
+
+static int amd_xgbe_phy_config_aneg(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	u32 mmd_mask = phydev->c45_ids.devices_in_package;
+	int ret;
+
+	if (phydev->autoneg != AUTONEG_ENABLE)
+		return amd_xgbe_phy_setup_forced(phydev);
+
+	/* Make sure we have the AN MMD present */
+	if (!(mmd_mask & MDIO_DEVS_AN))
+		return -EINVAL;
+
+	/* Get the current speed mode */
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2);
+	if (ret < 0)
+		return ret;
+
+	/* Start/Restart the auto-negotiation state machine */
+	mutex_lock(&priv->an_mutex);
+	priv->an_result = AMD_XGBE_AN_READY;
+	priv->an_state = AMD_XGBE_AN_START;
+	priv->kr_state = AMD_XGBE_RX_READY;
+	priv->kx_state = AMD_XGBE_RX_READY;
+	mutex_unlock(&priv->an_mutex);
+
+	queue_work(priv->an_workqueue, &priv->an_work);
+
+	return 0;
+}
+
+static int amd_xgbe_phy_aneg_done(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	enum amd_xgbe_phy_an state;
+
+	mutex_lock(&priv->an_mutex);
+	state = priv->an_result;
+	mutex_unlock(&priv->an_mutex);
+
+	return (state == AMD_XGBE_AN_COMPLETE);
+}
+
+static int amd_xgbe_phy_update_link(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	enum amd_xgbe_phy_an state;
+	unsigned int check_again, autoneg;
+	int ret;
+
+	/* If we're doing auto-negotiation don't report link down */
+	mutex_lock(&priv->an_mutex);
+	state = priv->an_state;
+	mutex_unlock(&priv->an_mutex);
+
+	if (state != AMD_XGBE_AN_READY) {
+		phydev->link = 1;
+		return 0;
+	}
+
+	/* Since the device can be in the wrong mode when a link is
+	 * (re-)established (cable connected after the interface is
+	 * up, etc.), the link status may report no link. If there
+	 * is no link, try switching modes and checking the status
+	 * again.
+	 */
+	check_again = 1;
+again:
+	/* Link status is latched low, so read once to clear
+	 * and then read again to get current state
+	 */
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_STAT1);
+	if (ret < 0)
+		return ret;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_STAT1);
+	if (ret < 0)
+		return ret;
+
+	phydev->link = (ret & MDIO_STAT1_LSTATUS) ? 1 : 0;
+
+	if (!phydev->link) {
+		ret = amd_xgbe_phy_switch_mode(phydev);
+		if (check_again) {
+			check_again = 0;
+			goto again;
+		}
+	}
+
+	autoneg = (phydev->link && !priv->link) ? 1 : 0;
+	priv->link = phydev->link;
+	if (autoneg) {
+		/* Link is (back) up, re-start auto-negotiation */
+		ret = amd_xgbe_phy_config_aneg(phydev);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int amd_xgbe_phy_read_status(struct phy_device *phydev)
+{
+	u32 mmd_mask = phydev->c45_ids.devices_in_package;
+	int ret, mode, ad_ret, lp_ret;
+
+	ret = amd_xgbe_phy_update_link(phydev);
+	if (ret)
+		return ret;
+
+	mode = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2);
+	if (mode < 0)
+		return mode;
+	mode &= MDIO_PCS_CTRL2_TYPE;
+
+	if (phydev->autoneg == AUTONEG_ENABLE) {
+		if (!(mmd_mask & MDIO_DEVS_AN))
+			return -EINVAL;
+
+		if (!amd_xgbe_phy_aneg_done(phydev))
+			return 0;
+
+		/* Compare Advertisement and Link Partner register 1 */
+		ad_ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_ADVERTISE);
+		if (ad_ret < 0)
+			return ad_ret;
+		lp_ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_LPA);
+		if (lp_ret < 0)
+			return lp_ret;
+
+		ad_ret &= lp_ret;
+		phydev->pause = (ad_ret & 0x400) ? 1 : 0;
+		phydev->asym_pause = (ad_ret & 0x800) ? 1 : 0;
+
+		/* Compare Advertisement and Link Partner register 2 */
+		ad_ret = phy_read_mmd(phydev, MDIO_MMD_AN,
+				      MDIO_AN_ADVERTISE + 1);
+		if (ad_ret < 0)
+			return ad_ret;
+		lp_ret = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_LPA + 1);
+		if (lp_ret < 0)
+			return lp_ret;
+
+		ad_ret &= lp_ret;
+		if (ad_ret & 0x80) {
+			phydev->speed = SPEED_10000;
+			if (mode != MDIO_PCS_CTRL2_10GBR) {
+				ret = amd_xgbe_phy_xgmii_mode(phydev);
+				if (ret < 0)
+					return ret;
+			}
+		} else {
+			phydev->speed = SPEED_1000;
+			if (mode == MDIO_PCS_CTRL2_10GBR) {
+				ret = amd_xgbe_phy_gmii_mode(phydev);
+				if (ret < 0)
+					return ret;
+			}
+		}
+
+		phydev->duplex = DUPLEX_FULL;
+	} else {
+		phydev->speed = (mode == MDIO_PCS_CTRL2_10GBR) ? SPEED_10000
+							       : SPEED_1000;
+		phydev->duplex = DUPLEX_FULL;
+		phydev->pause = 0;
+		phydev->asym_pause = 0;
+	}
+
+	return 0;
+}
+
+static int amd_xgbe_phy_suspend(struct phy_device *phydev)
+{
+	int ret;
+
+	mutex_lock(&phydev->lock);
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		goto unlock;
+
+	ret |= MDIO_CTRL1_LPOWER;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	ret = 0;
+
+unlock:
+	mutex_unlock(&phydev->lock);
+
+	return ret;
+}
+
+static int amd_xgbe_phy_resume(struct phy_device *phydev)
+{
+	int ret;
+
+	mutex_lock(&phydev->lock);
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1);
+	if (ret < 0)
+		goto unlock;
+
+	ret &= ~MDIO_CTRL1_LPOWER;
+	phy_write_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL1, ret);
+
+	ret = 0;
+
+unlock:
+	mutex_unlock(&phydev->lock);
+
+	return ret;
+}
+
+static int amd_xgbe_phy_probe(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv;
+	struct platform_device *pdev;
+	struct device *dev;
+	char *wq_name;
+	int ret;
+
+	if (!phydev->dev.of_node)
+		return -EINVAL;
+
+	pdev = of_find_device_by_node(phydev->dev.of_node);
+	if (!pdev)
+		return -EINVAL;
+	dev = &pdev->dev;
+
+	wq_name = kasprintf(GFP_KERNEL, "%s-amd-xgbe-phy", phydev->bus->name);
+	if (!wq_name) {
+		ret = -ENOMEM;
+		goto err_pdev;
+	}
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv) {
+		ret = -ENOMEM;
+		goto err_name;
+	}
+
+	priv->pdev = pdev;
+	priv->dev = dev;
+	priv->phydev = phydev;
+
+	/* Get the device mmio areas */
+	priv->rxtx_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	priv->rxtx_regs = devm_ioremap_resource(dev, priv->rxtx_res);
+	if (IS_ERR(priv->rxtx_regs)) {
+		dev_err(dev, "rxtx ioremap failed\n");
+		ret = PTR_ERR(priv->rxtx_regs);
+		goto err_priv;
+	}
+
+	priv->sir0_res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	priv->sir0_regs = devm_ioremap_resource(dev, priv->sir0_res);
+	if (IS_ERR(priv->sir0_regs)) {
+		dev_err(dev, "sir0 ioremap failed\n");
+		ret = PTR_ERR(priv->sir0_regs);
+		goto err_rxtx;
+	}
+
+	priv->sir1_res = platform_get_resource(pdev, IORESOURCE_MEM, 2);
+	priv->sir1_regs = devm_ioremap_resource(dev, priv->sir1_res);
+	if (IS_ERR(priv->sir1_regs)) {
+		dev_err(dev, "sir1 ioremap failed\n");
+		ret = PTR_ERR(priv->sir1_regs);
+		goto err_sir0;
+	}
+
+	priv->link = 1;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_PCS, MDIO_CTRL2);
+	if (ret < 0)
+		goto err_sir1;
+	if ((ret & MDIO_PCS_CTRL2_TYPE) == MDIO_PCS_CTRL2_10GBR)
+		priv->mode = AMD_XGBE_MODE_KR;
+	else
+		priv->mode = AMD_XGBE_MODE_KX;
+
+	mutex_init(&priv->an_mutex);
+	INIT_WORK(&priv->an_work, amd_xgbe_an_state_machine);
+	priv->an_workqueue = create_singlethread_workqueue(wq_name);
+	if (!priv->an_workqueue) {
+		ret = -ENOMEM;
+		goto err_sir1;
+	}
+
+	phydev->priv = priv;
+
+	kfree(wq_name);
+	of_dev_put(pdev);
+
+	return 0;
+
+err_sir1:
+	devm_iounmap(dev, priv->sir1_regs);
+	devm_release_mem_region(dev, priv->sir1_res->start,
+				resource_size(priv->sir1_res));
+
+err_sir0:
+	devm_iounmap(dev, priv->sir0_regs);
+	devm_release_mem_region(dev, priv->sir0_res->start,
+				resource_size(priv->sir0_res));
+
+err_rxtx:
+	devm_iounmap(dev, priv->rxtx_regs);
+	devm_release_mem_region(dev, priv->rxtx_res->start,
+				resource_size(priv->rxtx_res));
+
+err_priv:
+	devm_kfree(dev, priv);
+
+err_name:
+	kfree(wq_name);
+
+err_pdev:
+	of_dev_put(pdev);
+
+	return ret;
+}
+
+static void amd_xgbe_phy_remove(struct phy_device *phydev)
+{
+	struct amd_xgbe_phy_priv *priv = phydev->priv;
+	struct device *dev = priv->dev;
+
+	/* Stop any in process auto-negotiation */
+	mutex_lock(&priv->an_mutex);
+	priv->an_state = AMD_XGBE_AN_EXIT;
+	mutex_unlock(&priv->an_mutex);
+
+	flush_workqueue(priv->an_workqueue);
+	destroy_workqueue(priv->an_workqueue);
+
+	/* Release resources */
+	devm_iounmap(dev, priv->sir1_regs);
+	devm_release_mem_region(dev, priv->sir1_res->start,
+				resource_size(priv->sir1_res));
+
+	devm_iounmap(dev, priv->sir0_regs);
+	devm_release_mem_region(dev, priv->sir0_res->start,
+				resource_size(priv->sir0_res));
+
+	devm_iounmap(dev, priv->rxtx_regs);
+	devm_release_mem_region(dev, priv->rxtx_res->start,
+				resource_size(priv->rxtx_res));
+
+	devm_kfree(dev, priv);
+}
+
+static int amd_xgbe_match_phy_device(struct phy_device *phydev)
+{
+	return phydev->c45_ids.device_ids[MDIO_MMD_PCS] == XGBE_PHY_ID;
+}
+
+static struct phy_driver amd_xgbe_phy_driver[] = {
+	{
+		.phy_id			= XGBE_PHY_ID,
+		.phy_id_mask		= XGBE_PHY_MASK,
+		.name			= "AMD XGBE PHY",
+		.features		= 0,
+		.probe			= amd_xgbe_phy_probe,
+		.remove			= amd_xgbe_phy_remove,
+		.soft_reset		= amd_xgbe_phy_soft_reset,
+		.config_init		= amd_xgbe_phy_config_init,
+		.suspend		= amd_xgbe_phy_suspend,
+		.resume			= amd_xgbe_phy_resume,
+		.config_aneg		= amd_xgbe_phy_config_aneg,
+		.aneg_done		= amd_xgbe_phy_aneg_done,
+		.read_status		= amd_xgbe_phy_read_status,
+		.match_phy_device	= amd_xgbe_match_phy_device,
+		.driver			= {
+			.owner = THIS_MODULE,
+		},
+	},
+};
+
+static int __init amd_xgbe_phy_init(void)
+{
+	return phy_drivers_register(amd_xgbe_phy_driver,
+				    ARRAY_SIZE(amd_xgbe_phy_driver));
+}
+
+static void __exit amd_xgbe_phy_exit(void)
+{
+	phy_drivers_unregister(amd_xgbe_phy_driver,
+			       ARRAY_SIZE(amd_xgbe_phy_driver));
+}
+
+module_init(amd_xgbe_phy_init);
+module_exit(amd_xgbe_phy_exit);
+
+static struct mdio_device_id __maybe_unused amd_xgbe_phy_ids[] = {
+	{ XGBE_PHY_ID, XGBE_PHY_MASK },
+	{ }
+};
+MODULE_DEVICE_TABLE(mdio, amd_xgbe_phy_ids);
diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c
index 643464d5a727..fdc1b418fa6a 100644
--- a/drivers/net/phy/at803x.c
+++ b/drivers/net/phy/at803x.c
@@ -16,9 +16,13 @@
 #include <linux/string.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <linux/of_gpio.h>
+#include <linux/gpio/consumer.h>
 
 #define AT803X_INTR_ENABLE			0x12
 #define AT803X_INTR_STATUS			0x13
+#define AT803X_SMART_SPEED			0x14
+#define AT803X_LED_CONTROL			0x18
 #define AT803X_WOL_ENABLE			0x01
 #define AT803X_DEVICE_ADDR			0x03
 #define AT803X_LOC_MAC_ADDR_0_15_OFFSET		0x804C
@@ -35,10 +39,52 @@
 #define AT803X_DEBUG_SYSTEM_MODE_CTRL		0x05
 #define AT803X_DEBUG_RGMII_TX_CLK_DLY		BIT(8)
 
+#define ATH8030_PHY_ID 0x004dd076
+#define ATH8031_PHY_ID 0x004dd074
+#define ATH8035_PHY_ID 0x004dd072
+
 MODULE_DESCRIPTION("Atheros 803x PHY driver");
 MODULE_AUTHOR("Matus Ujhelyi");
 MODULE_LICENSE("GPL");
 
+struct at803x_priv {
+	bool phy_reset:1;
+	struct gpio_desc *gpiod_reset;
+};
+
+struct at803x_context {
+	u16 bmcr;
+	u16 advertise;
+	u16 control1000;
+	u16 int_enable;
+	u16 smart_speed;
+	u16 led_control;
+};
+
+/* save relevant PHY registers to private copy */
+static void at803x_context_save(struct phy_device *phydev,
+				struct at803x_context *context)
+{
+	context->bmcr = phy_read(phydev, MII_BMCR);
+	context->advertise = phy_read(phydev, MII_ADVERTISE);
+	context->control1000 = phy_read(phydev, MII_CTRL1000);
+	context->int_enable = phy_read(phydev, AT803X_INTR_ENABLE);
+	context->smart_speed = phy_read(phydev, AT803X_SMART_SPEED);
+	context->led_control = phy_read(phydev, AT803X_LED_CONTROL);
+}
+
+/* restore relevant PHY registers from private copy */
+static void at803x_context_restore(struct phy_device *phydev,
+				   const struct at803x_context *context)
+{
+	phy_write(phydev, MII_BMCR, context->bmcr);
+	phy_write(phydev, MII_ADVERTISE, context->advertise);
+	phy_write(phydev, MII_CTRL1000, context->control1000);
+	phy_write(phydev, AT803X_INTR_ENABLE, context->int_enable);
+	phy_write(phydev, AT803X_SMART_SPEED, context->smart_speed);
+	phy_write(phydev, AT803X_LED_CONTROL, context->led_control);
+}
+
 static int at803x_set_wol(struct phy_device *phydev,
 			  struct ethtool_wolinfo *wol)
 {
@@ -142,43 +188,33 @@ static int at803x_resume(struct phy_device *phydev)
 	return 0;
 }
 
+static int at803x_probe(struct phy_device *phydev)
+{
+	struct device *dev = &phydev->dev;
+	struct at803x_priv *priv;
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->gpiod_reset = devm_gpiod_get(dev, "reset");
+	if (IS_ERR(priv->gpiod_reset))
+		priv->gpiod_reset = NULL;
+	else
+		gpiod_direction_output(priv->gpiod_reset, 1);
+
+	phydev->priv = priv;
+
+	return 0;
+}
+
 static int at803x_config_init(struct phy_device *phydev)
 {
-	int val;
 	int ret;
-	u32 features;
-
-	features = SUPPORTED_TP | SUPPORTED_MII | SUPPORTED_AUI |
-		   SUPPORTED_FIBRE | SUPPORTED_BNC;
-
-	val = phy_read(phydev, MII_BMSR);
-	if (val < 0)
-		return val;
-
-	if (val & BMSR_ANEGCAPABLE)
-		features |= SUPPORTED_Autoneg;
-	if (val & BMSR_100FULL)
-		features |= SUPPORTED_100baseT_Full;
-	if (val & BMSR_100HALF)
-		features |= SUPPORTED_100baseT_Half;
-	if (val & BMSR_10FULL)
-		features |= SUPPORTED_10baseT_Full;
-	if (val & BMSR_10HALF)
-		features |= SUPPORTED_10baseT_Half;
-
-	if (val & BMSR_ESTATEN) {
-		val = phy_read(phydev, MII_ESTATUS);
-		if (val < 0)
-			return val;
-
-		if (val & ESTATUS_1000_TFULL)
-			features |= SUPPORTED_1000baseT_Full;
-		if (val & ESTATUS_1000_THALF)
-			features |= SUPPORTED_1000baseT_Half;
-	}
 
-	phydev->supported = features;
-	phydev->advertising = features;
+	ret = genphy_config_init(phydev);
+	if (ret < 0)
+		return ret;
 
 	if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) {
 		ret = phy_write(phydev, AT803X_DEBUG_ADDR,
@@ -219,58 +255,99 @@ static int at803x_config_intr(struct phy_device *phydev)
 	return err;
 }
 
+static void at803x_link_change_notify(struct phy_device *phydev)
+{
+	struct at803x_priv *priv = phydev->priv;
+
+	/*
+	 * Conduct a hardware reset for AT8030 every time a link loss is
+	 * signalled. This is necessary to circumvent a hardware bug that
+	 * occurs when the cable is unplugged while TX packets are pending
+	 * in the FIFO. In such cases, the FIFO enters an error mode it
+	 * cannot recover from by software.
+	 */
+	if (phydev->drv->phy_id == ATH8030_PHY_ID) {
+		if (phydev->state == PHY_NOLINK) {
+			if (priv->gpiod_reset && !priv->phy_reset) {
+				struct at803x_context context;
+
+				at803x_context_save(phydev, &context);
+
+				gpiod_set_value(priv->gpiod_reset, 0);
+				msleep(1);
+				gpiod_set_value(priv->gpiod_reset, 1);
+				msleep(1);
+
+				at803x_context_restore(phydev, &context);
+
+				dev_dbg(&phydev->dev, "%s(): phy was reset\n",
+					__func__);
+				priv->phy_reset = true;
+			}
+		} else {
+			priv->phy_reset = false;
+		}
+	}
+}
+
 static struct phy_driver at803x_driver[] = {
 {
 	/* ATHEROS 8035 */
-	.phy_id		= 0x004dd072,
-	.name		= "Atheros 8035 ethernet",
-	.phy_id_mask	= 0xffffffef,
-	.config_init	= at803x_config_init,
-	.set_wol	= at803x_set_wol,
-	.get_wol	= at803x_get_wol,
-	.suspend	= at803x_suspend,
-	.resume		= at803x_resume,
-	.features	= PHY_GBIT_FEATURES,
-	.flags		= PHY_HAS_INTERRUPT,
-	.config_aneg	= genphy_config_aneg,
-	.read_status	= genphy_read_status,
-	.driver		= {
+	.phy_id			= ATH8035_PHY_ID,
+	.name			= "Atheros 8035 ethernet",
+	.phy_id_mask		= 0xffffffef,
+	.probe			= at803x_probe,
+	.config_init		= at803x_config_init,
+	.link_change_notify	= at803x_link_change_notify,
+	.set_wol		= at803x_set_wol,
+	.get_wol		= at803x_get_wol,
+	.suspend		= at803x_suspend,
+	.resume			= at803x_resume,
+	.features		= PHY_GBIT_FEATURES,
+	.flags			= PHY_HAS_INTERRUPT,
+	.config_aneg		= genphy_config_aneg,
+	.read_status		= genphy_read_status,
+	.driver			= {
 		.owner = THIS_MODULE,
 	},
 }, {
 	/* ATHEROS 8030 */
-	.phy_id		= 0x004dd076,
-	.name		= "Atheros 8030 ethernet",
-	.phy_id_mask	= 0xffffffef,
-	.config_init	= at803x_config_init,
-	.set_wol	= at803x_set_wol,
-	.get_wol	= at803x_get_wol,
-	.suspend	= at803x_suspend,
-	.resume		= at803x_resume,
-	.features	= PHY_GBIT_FEATURES,
-	.flags		= PHY_HAS_INTERRUPT,
-	.config_aneg	= genphy_config_aneg,
-	.read_status	= genphy_read_status,
-	.driver		= {
+	.phy_id			= ATH8030_PHY_ID,
+	.name			= "Atheros 8030 ethernet",
+	.phy_id_mask		= 0xffffffef,
+	.probe			= at803x_probe,
+	.config_init		= at803x_config_init,
+	.link_change_notify	= at803x_link_change_notify,
+	.set_wol		= at803x_set_wol,
+	.get_wol		= at803x_get_wol,
+	.suspend		= at803x_suspend,
+	.resume			= at803x_resume,
+	.features		= PHY_GBIT_FEATURES,
+	.flags			= PHY_HAS_INTERRUPT,
+	.config_aneg		= genphy_config_aneg,
+	.read_status		= genphy_read_status,
+	.driver			= {
 		.owner = THIS_MODULE,
 	},
 }, {
 	/* ATHEROS 8031 */
-	.phy_id		= 0x004dd074,
-	.name		= "Atheros 8031 ethernet",
-	.phy_id_mask	= 0xffffffef,
-	.config_init	= at803x_config_init,
-	.set_wol	= at803x_set_wol,
-	.get_wol	= at803x_get_wol,
-	.suspend	= at803x_suspend,
-	.resume		= at803x_resume,
-	.features	= PHY_GBIT_FEATURES,
-	.flags		= PHY_HAS_INTERRUPT,
-	.config_aneg	= genphy_config_aneg,
-	.read_status	= genphy_read_status,
-	.ack_interrupt  = &at803x_ack_interrupt,
-	.config_intr    = &at803x_config_intr,
-	.driver		= {
+	.phy_id			= ATH8031_PHY_ID,
+	.name			= "Atheros 8031 ethernet",
+	.phy_id_mask		= 0xffffffef,
+	.probe			= at803x_probe,
+	.config_init		= at803x_config_init,
+	.link_change_notify	= at803x_link_change_notify,
+	.set_wol		= at803x_set_wol,
+	.get_wol		= at803x_get_wol,
+	.suspend		= at803x_suspend,
+	.resume			= at803x_resume,
+	.features		= PHY_GBIT_FEATURES,
+	.flags			= PHY_HAS_INTERRUPT,
+	.config_aneg		= genphy_config_aneg,
+	.read_status		= genphy_read_status,
+	.ack_interrupt		= &at803x_ack_interrupt,
+	.config_intr		= &at803x_config_intr,
+	.driver			= {
 		.owner = THIS_MODULE,
 	},
 } };
@@ -283,17 +360,16 @@ static int __init atheros_init(void)
 
 static void __exit atheros_exit(void)
 {
-	return phy_drivers_unregister(at803x_driver,
-				      ARRAY_SIZE(at803x_driver));
+	phy_drivers_unregister(at803x_driver, ARRAY_SIZE(at803x_driver));
 }
 
 module_init(atheros_init);
 module_exit(atheros_exit);
 
 static struct mdio_device_id __maybe_unused atheros_tbl[] = {
-	{ 0x004dd076, 0xffffffef },
-	{ 0x004dd074, 0xffffffef },
-	{ 0x004dd072, 0xffffffef },
+	{ ATH8030_PHY_ID, 0xffffffef },
+	{ ATH8031_PHY_ID, 0xffffffef },
+	{ ATH8035_PHY_ID, 0xffffffef },
 	{ }
 };
 
diff --git a/drivers/net/phy/fixed.c b/drivers/net/phy/fixed.c
index ba55adfc7aae..d60d875cb445 100644
--- a/drivers/net/phy/fixed.c
+++ b/drivers/net/phy/fixed.c
@@ -21,6 +21,7 @@
 #include <linux/phy_fixed.h>
 #include <linux/err.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 
 #define MII_REGS_NUM 29
 
@@ -31,7 +32,7 @@ struct fixed_mdio_bus {
 };
 
 struct fixed_phy {
-	int id;
+	int addr;
 	u16 regs[MII_REGS_NUM];
 	struct phy_device *phydev;
 	struct fixed_phy_status status;
@@ -104,8 +105,8 @@ static int fixed_phy_update_regs(struct fixed_phy *fp)
 	if (fp->status.asym_pause)
 		lpa |= LPA_PAUSE_ASYM;
 
-	fp->regs[MII_PHYSID1] = fp->id >> 16;
-	fp->regs[MII_PHYSID2] = fp->id;
+	fp->regs[MII_PHYSID1] = 0;
+	fp->regs[MII_PHYSID2] = 0;
 
 	fp->regs[MII_BMSR] = bmsr;
 	fp->regs[MII_BMCR] = bmcr;
@@ -115,7 +116,7 @@ static int fixed_phy_update_regs(struct fixed_phy *fp)
 	return 0;
 }
 
-static int fixed_mdio_read(struct mii_bus *bus, int phy_id, int reg_num)
+static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num)
 {
 	struct fixed_mdio_bus *fmb = bus->priv;
 	struct fixed_phy *fp;
@@ -124,7 +125,7 @@ static int fixed_mdio_read(struct mii_bus *bus, int phy_id, int reg_num)
 		return -1;
 
 	list_for_each_entry(fp, &fmb->phys, node) {
-		if (fp->id == phy_id) {
+		if (fp->addr == phy_addr) {
 			/* Issue callback if user registered it. */
 			if (fp->link_update) {
 				fp->link_update(fp->phydev->attached_dev,
@@ -138,7 +139,7 @@ static int fixed_mdio_read(struct mii_bus *bus, int phy_id, int reg_num)
 	return 0xFFFF;
 }
 
-static int fixed_mdio_write(struct mii_bus *bus, int phy_id, int reg_num,
+static int fixed_mdio_write(struct mii_bus *bus, int phy_addr, int reg_num,
 			    u16 val)
 {
 	return 0;
@@ -160,7 +161,7 @@ int fixed_phy_set_link_update(struct phy_device *phydev,
 		return -EINVAL;
 
 	list_for_each_entry(fp, &fmb->phys, node) {
-		if (fp->id == phydev->phy_id) {
+		if (fp->addr == phydev->addr) {
 			fp->link_update = link_update;
 			fp->phydev = phydev;
 			return 0;
@@ -171,7 +172,7 @@ int fixed_phy_set_link_update(struct phy_device *phydev,
 }
 EXPORT_SYMBOL_GPL(fixed_phy_set_link_update);
 
-int fixed_phy_add(unsigned int irq, int phy_id,
+int fixed_phy_add(unsigned int irq, int phy_addr,
 		  struct fixed_phy_status *status)
 {
 	int ret;
@@ -184,9 +185,9 @@ int fixed_phy_add(unsigned int irq, int phy_id,
 
 	memset(fp->regs, 0xFF,  sizeof(fp->regs[0]) * MII_REGS_NUM);
 
-	fmb->irqs[phy_id] = irq;
+	fmb->irqs[phy_addr] = irq;
 
-	fp->id = phy_id;
+	fp->addr = phy_addr;
 	fp->status = *status;
 
 	ret = fixed_phy_update_regs(fp);
@@ -203,6 +204,66 @@ err_regs:
 }
 EXPORT_SYMBOL_GPL(fixed_phy_add);
 
+void fixed_phy_del(int phy_addr)
+{
+	struct fixed_mdio_bus *fmb = &platform_fmb;
+	struct fixed_phy *fp, *tmp;
+
+	list_for_each_entry_safe(fp, tmp, &fmb->phys, node) {
+		if (fp->addr == phy_addr) {
+			list_del(&fp->node);
+			kfree(fp);
+			return;
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(fixed_phy_del);
+
+static int phy_fixed_addr;
+static DEFINE_SPINLOCK(phy_fixed_addr_lock);
+
+int fixed_phy_register(unsigned int irq,
+		       struct fixed_phy_status *status,
+		       struct device_node *np)
+{
+	struct fixed_mdio_bus *fmb = &platform_fmb;
+	struct phy_device *phy;
+	int phy_addr;
+	int ret;
+
+	/* Get the next available PHY address, up to PHY_MAX_ADDR */
+	spin_lock(&phy_fixed_addr_lock);
+	if (phy_fixed_addr == PHY_MAX_ADDR) {
+		spin_unlock(&phy_fixed_addr_lock);
+		return -ENOSPC;
+	}
+	phy_addr = phy_fixed_addr++;
+	spin_unlock(&phy_fixed_addr_lock);
+
+	ret = fixed_phy_add(PHY_POLL, phy_addr, status);
+	if (ret < 0)
+		return ret;
+
+	phy = get_phy_device(fmb->mii_bus, phy_addr, false);
+	if (!phy || IS_ERR(phy)) {
+		fixed_phy_del(phy_addr);
+		return -EINVAL;
+	}
+
+	of_node_get(np);
+	phy->dev.of_node = np;
+
+	ret = phy_device_register(phy);
+	if (ret) {
+		phy_device_free(phy);
+		of_node_put(np);
+		fixed_phy_del(phy_addr);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int __init fixed_mdio_bus_init(void)
 {
 	struct fixed_mdio_bus *fmb = &platform_fmb;
diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 76f54b32a120..2e58aa54484c 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -69,6 +69,73 @@ struct mii_bus *mdiobus_alloc_size(size_t size)
 }
 EXPORT_SYMBOL(mdiobus_alloc_size);
 
+static void _devm_mdiobus_free(struct device *dev, void *res)
+{
+	mdiobus_free(*(struct mii_bus **)res);
+}
+
+static int devm_mdiobus_match(struct device *dev, void *res, void *data)
+{
+	struct mii_bus **r = res;
+
+	if (WARN_ON(!r || !*r))
+		return 0;
+
+	return *r == data;
+}
+
+/**
+ * devm_mdiobus_alloc_size - Resource-managed mdiobus_alloc_size()
+ * @dev:		Device to allocate mii_bus for
+ * @sizeof_priv:	Space to allocate for private structure.
+ *
+ * Managed mdiobus_alloc_size. mii_bus allocated with this function is
+ * automatically freed on driver detach.
+ *
+ * If an mii_bus allocated with this function needs to be freed separately,
+ * devm_mdiobus_free() must be used.
+ *
+ * RETURNS:
+ * Pointer to allocated mii_bus on success, NULL on failure.
+ */
+struct mii_bus *devm_mdiobus_alloc_size(struct device *dev, int sizeof_priv)
+{
+	struct mii_bus **ptr, *bus;
+
+	ptr = devres_alloc(_devm_mdiobus_free, sizeof(*ptr), GFP_KERNEL);
+	if (!ptr)
+		return NULL;
+
+	/* use raw alloc_dr for kmalloc caller tracing */
+	bus = mdiobus_alloc_size(sizeof_priv);
+	if (bus) {
+		*ptr = bus;
+		devres_add(dev, ptr);
+	} else {
+		devres_free(ptr);
+	}
+
+	return bus;
+}
+EXPORT_SYMBOL_GPL(devm_mdiobus_alloc_size);
+
+/**
+ * devm_mdiobus_free - Resource-managed mdiobus_free()
+ * @dev:		Device this mii_bus belongs to
+ * @bus:		the mii_bus associated with the device
+ *
+ * Free mii_bus allocated with devm_mdiobus_alloc_size().
+ */
+void devm_mdiobus_free(struct device *dev, struct mii_bus *bus)
+{
+	int rc;
+
+	rc = devres_release(dev, _devm_mdiobus_free,
+			    devm_mdiobus_match, bus);
+	WARN_ON(rc);
+}
+EXPORT_SYMBOL_GPL(devm_mdiobus_free);
+
 /**
  * mdiobus_release - mii_bus device release callback
  * @d: the target struct device that contains the mii_bus
@@ -233,6 +300,12 @@ struct phy_device *mdiobus_scan(struct mii_bus *bus, int addr)
 	if (IS_ERR(phydev) || phydev == NULL)
 		return phydev;
 
+	/*
+	 * For DT, see if the auto-probed phy has a correspoding child
+	 * in the bus node, and set the of_node pointer in this case.
+	 */
+	of_mdiobus_link_phydev(bus, phydev);
+
 	err = phy_device_register(phydev);
 	if (err) {
 		phy_device_free(phydev);
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index d849684231c1..bc7c7d2f75f2 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -283,6 +283,110 @@ static int ksz9021_config_init(struct phy_device *phydev)
 	return 0;
 }
 
+#define MII_KSZ9031RN_MMD_CTRL_REG	0x0d
+#define MII_KSZ9031RN_MMD_REGDATA_REG	0x0e
+#define OP_DATA				1
+#define KSZ9031_PS_TO_REG		60
+
+/* Extended registers */
+#define MII_KSZ9031RN_CONTROL_PAD_SKEW	4
+#define MII_KSZ9031RN_RX_DATA_PAD_SKEW	5
+#define MII_KSZ9031RN_TX_DATA_PAD_SKEW	6
+#define MII_KSZ9031RN_CLK_PAD_SKEW	8
+
+static int ksz9031_extended_write(struct phy_device *phydev,
+				  u8 mode, u32 dev_addr, u32 regnum, u16 val)
+{
+	phy_write(phydev, MII_KSZ9031RN_MMD_CTRL_REG, dev_addr);
+	phy_write(phydev, MII_KSZ9031RN_MMD_REGDATA_REG, regnum);
+	phy_write(phydev, MII_KSZ9031RN_MMD_CTRL_REG, (mode << 14) | dev_addr);
+	return phy_write(phydev, MII_KSZ9031RN_MMD_REGDATA_REG, val);
+}
+
+static int ksz9031_extended_read(struct phy_device *phydev,
+				 u8 mode, u32 dev_addr, u32 regnum)
+{
+	phy_write(phydev, MII_KSZ9031RN_MMD_CTRL_REG, dev_addr);
+	phy_write(phydev, MII_KSZ9031RN_MMD_REGDATA_REG, regnum);
+	phy_write(phydev, MII_KSZ9031RN_MMD_CTRL_REG, (mode << 14) | dev_addr);
+	return phy_read(phydev, MII_KSZ9031RN_MMD_REGDATA_REG);
+}
+
+static int ksz9031_of_load_skew_values(struct phy_device *phydev,
+				       struct device_node *of_node,
+				       u16 reg, size_t field_sz,
+				       char *field[], u8 numfields)
+{
+	int val[4] = {-1, -2, -3, -4};
+	int matches = 0;
+	u16 mask;
+	u16 maxval;
+	u16 newval;
+	int i;
+
+	for (i = 0; i < numfields; i++)
+		if (!of_property_read_u32(of_node, field[i], val + i))
+			matches++;
+
+	if (!matches)
+		return 0;
+
+	if (matches < numfields)
+		newval = ksz9031_extended_read(phydev, OP_DATA, 2, reg);
+	else
+		newval = 0;
+
+	maxval = (field_sz == 4) ? 0xf : 0x1f;
+	for (i = 0; i < numfields; i++)
+		if (val[i] != -(i + 1)) {
+			mask = 0xffff;
+			mask ^= maxval << (field_sz * i);
+			newval = (newval & mask) |
+				(((val[i] / KSZ9031_PS_TO_REG) & maxval)
+					<< (field_sz * i));
+		}
+
+	return ksz9031_extended_write(phydev, OP_DATA, 2, reg, newval);
+}
+
+static int ksz9031_config_init(struct phy_device *phydev)
+{
+	struct device *dev = &phydev->dev;
+	struct device_node *of_node = dev->of_node;
+	char *clk_skews[2] = {"rxc-skew-ps", "txc-skew-ps"};
+	char *rx_data_skews[4] = {
+		"rxd0-skew-ps", "rxd1-skew-ps",
+		"rxd2-skew-ps", "rxd3-skew-ps"
+	};
+	char *tx_data_skews[4] = {
+		"txd0-skew-ps", "txd1-skew-ps",
+		"txd2-skew-ps", "txd3-skew-ps"
+	};
+	char *control_skews[2] = {"txen-skew-ps", "rxdv-skew-ps"};
+
+	if (!of_node && dev->parent->of_node)
+		of_node = dev->parent->of_node;
+
+	if (of_node) {
+		ksz9031_of_load_skew_values(phydev, of_node,
+				MII_KSZ9031RN_CLK_PAD_SKEW, 5,
+				clk_skews, 2);
+
+		ksz9031_of_load_skew_values(phydev, of_node,
+				MII_KSZ9031RN_CONTROL_PAD_SKEW, 4,
+				control_skews, 2);
+
+		ksz9031_of_load_skew_values(phydev, of_node,
+				MII_KSZ9031RN_RX_DATA_PAD_SKEW, 4,
+				rx_data_skews, 4);
+
+		ksz9031_of_load_skew_values(phydev, of_node,
+				MII_KSZ9031RN_TX_DATA_PAD_SKEW, 4,
+				tx_data_skews, 4);
+	}
+	return 0;
+}
+
 #define KSZ8873MLL_GLOBAL_CONTROL_4	0x06
 #define KSZ8873MLL_GLOBAL_CONTROL_4_DUPLEX	(1 << 6)
 #define KSZ8873MLL_GLOBAL_CONTROL_4_SPEED	(1 << 4)
@@ -469,7 +573,7 @@ static struct phy_driver ksphy_driver[] = {
 	.features	= (PHY_GBIT_FEATURES | SUPPORTED_Pause
 				| SUPPORTED_Asym_Pause),
 	.flags		= PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT,
-	.config_init	= kszphy_config_init,
+	.config_init	= ksz9031_config_init,
 	.config_aneg	= genphy_config_aneg,
 	.read_status	= genphy_read_status,
 	.ack_interrupt	= kszphy_ack_interrupt,
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 3bc079a67a3d..f7c61812ea4a 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -720,6 +720,9 @@ void phy_state_machine(struct work_struct *work)
 
 	mutex_lock(&phydev->lock);
 
+	if (phydev->drv->link_change_notify)
+		phydev->drv->link_change_notify(phydev);
+
 	switch (phydev->state) {
 	case PHY_DOWN:
 	case PHY_STARTING:
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 4987a1c6dc52..35d753d22f78 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -33,6 +33,7 @@
 #include <linux/mdio.h>
 #include <linux/io.h>
 #include <linux/uaccess.h>
+#include <linux/of.h>
 
 #include <asm/irq.h>
 
@@ -1067,14 +1068,11 @@ int genphy_soft_reset(struct phy_device *phydev)
 }
 EXPORT_SYMBOL(genphy_soft_reset);
 
-static int genphy_config_init(struct phy_device *phydev)
+int genphy_config_init(struct phy_device *phydev)
 {
 	int val;
 	u32 features;
 
-	/* For now, I'll claim that the generic driver supports
-	 * all possible port types
-	 */
 	features = (SUPPORTED_TP | SUPPORTED_MII
 			| SUPPORTED_AUI | SUPPORTED_FIBRE |
 			SUPPORTED_BNC);
@@ -1107,8 +1105,8 @@ static int genphy_config_init(struct phy_device *phydev)
 			features |= SUPPORTED_1000baseT_Half;
 	}
 
-	phydev->supported = features;
-	phydev->advertising = features;
+	phydev->supported &= features;
+	phydev->advertising &= features;
 
 	return 0;
 }
@@ -1118,6 +1116,7 @@ static int gen10g_soft_reset(struct phy_device *phydev)
 	/* Do nothing for now */
 	return 0;
 }
+EXPORT_SYMBOL(genphy_config_init);
 
 static int gen10g_config_init(struct phy_device *phydev)
 {
@@ -1168,6 +1167,38 @@ static int gen10g_resume(struct phy_device *phydev)
 	return 0;
 }
 
+static void of_set_phy_supported(struct phy_device *phydev)
+{
+	struct device_node *node = phydev->dev.of_node;
+	u32 max_speed;
+
+	if (!IS_ENABLED(CONFIG_OF_MDIO))
+		return;
+
+	if (!node)
+		return;
+
+	if (!of_property_read_u32(node, "max-speed", &max_speed)) {
+		/* The default values for phydev->supported are provided by the PHY
+		 * driver "features" member, we want to reset to sane defaults fist
+		 * before supporting higher speeds.
+		 */
+		phydev->supported &= PHY_DEFAULT_FEATURES;
+
+		switch (max_speed) {
+		default:
+			return;
+
+		case SPEED_1000:
+			phydev->supported |= PHY_1000BT_FEATURES;
+		case SPEED_100:
+			phydev->supported |= PHY_100BT_FEATURES;
+		case SPEED_10:
+			phydev->supported |= PHY_10BT_FEATURES;
+		}
+	}
+}
+
 /**
  * phy_probe - probe and init a PHY device
  * @dev: device to probe and init
@@ -1202,7 +1233,8 @@ static int phy_probe(struct device *dev)
 	 * or both of these values
 	 */
 	phydev->supported = phydrv->features;
-	phydev->advertising = phydrv->features;
+	of_set_phy_supported(phydev);
+	phydev->advertising = phydev->supported;
 
 	/* Set the state to READY by default */
 	phydev->state = PHY_READY;
@@ -1295,7 +1327,9 @@ static struct phy_driver genphy_driver[] = {
 	.name		= "Generic PHY",
 	.soft_reset	= genphy_soft_reset,
 	.config_init	= genphy_config_init,
-	.features	= 0,
+	.features	= PHY_GBIT_FEATURES | SUPPORTED_MII |
+			  SUPPORTED_AUI | SUPPORTED_FIBRE |
+			  SUPPORTED_BNC,
 	.config_aneg	= genphy_config_aneg,
 	.aneg_done	= genphy_aneg_done,
 	.read_status	= genphy_read_status,
diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index fa1d69a38ccf..45483fdfbe06 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -64,65 +64,51 @@ static int rtl8211e_config_intr(struct phy_device *phydev)
 	return err;
 }
 
-/* RTL8201CP */
-static struct phy_driver rtl8201cp_driver = {
-	.phy_id         = 0x00008201,
-	.name           = "RTL8201CP Ethernet",
-	.phy_id_mask    = 0x0000ffff,
-	.features       = PHY_BASIC_FEATURES,
-	.flags          = PHY_HAS_INTERRUPT,
-	.config_aneg    = &genphy_config_aneg,
-	.read_status    = &genphy_read_status,
-	.driver         = { .owner = THIS_MODULE,},
-};
-
-/* RTL8211B */
-static struct phy_driver rtl8211b_driver = {
-	.phy_id		= 0x001cc912,
-	.name		= "RTL8211B Gigabit Ethernet",
-	.phy_id_mask	= 0x001fffff,
-	.features	= PHY_GBIT_FEATURES,
-	.flags		= PHY_HAS_INTERRUPT,
-	.config_aneg	= &genphy_config_aneg,
-	.read_status	= &genphy_read_status,
-	.ack_interrupt	= &rtl821x_ack_interrupt,
-	.config_intr	= &rtl8211b_config_intr,
-	.driver		= { .owner = THIS_MODULE,},
-};
-
-/* RTL8211E */
-static struct phy_driver rtl8211e_driver = {
-	.phy_id		= 0x001cc915,
-	.name		= "RTL8211E Gigabit Ethernet",
-	.phy_id_mask	= 0x001fffff,
-	.features	= PHY_GBIT_FEATURES,
-	.flags		= PHY_HAS_INTERRUPT,
-	.config_aneg	= &genphy_config_aneg,
-	.read_status	= &genphy_read_status,
-	.ack_interrupt	= &rtl821x_ack_interrupt,
-	.config_intr	= &rtl8211e_config_intr,
-	.suspend	= genphy_suspend,
-	.resume		= genphy_resume,
-	.driver		= { .owner = THIS_MODULE,},
+static struct phy_driver realtek_drvs[] = {
+	{
+		.phy_id         = 0x00008201,
+		.name           = "RTL8201CP Ethernet",
+		.phy_id_mask    = 0x0000ffff,
+		.features       = PHY_BASIC_FEATURES,
+		.flags          = PHY_HAS_INTERRUPT,
+		.config_aneg    = &genphy_config_aneg,
+		.read_status    = &genphy_read_status,
+		.driver         = { .owner = THIS_MODULE,},
+	}, {
+		.phy_id		= 0x001cc912,
+		.name		= "RTL8211B Gigabit Ethernet",
+		.phy_id_mask	= 0x001fffff,
+		.features	= PHY_GBIT_FEATURES,
+		.flags		= PHY_HAS_INTERRUPT,
+		.config_aneg	= &genphy_config_aneg,
+		.read_status	= &genphy_read_status,
+		.ack_interrupt	= &rtl821x_ack_interrupt,
+		.config_intr	= &rtl8211b_config_intr,
+		.driver		= { .owner = THIS_MODULE,},
+	}, {
+		.phy_id		= 0x001cc915,
+		.name		= "RTL8211E Gigabit Ethernet",
+		.phy_id_mask	= 0x001fffff,
+		.features	= PHY_GBIT_FEATURES,
+		.flags		= PHY_HAS_INTERRUPT,
+		.config_aneg	= &genphy_config_aneg,
+		.read_status	= &genphy_read_status,
+		.ack_interrupt	= &rtl821x_ack_interrupt,
+		.config_intr	= &rtl8211e_config_intr,
+		.suspend	= genphy_suspend,
+		.resume		= genphy_resume,
+		.driver		= { .owner = THIS_MODULE,},
+	},
 };
 
 static int __init realtek_init(void)
 {
-	int ret;
-
-	ret = phy_driver_register(&rtl8201cp_driver);
-	if (ret < 0)
-		return -ENODEV;
-	ret = phy_driver_register(&rtl8211b_driver);
-	if (ret < 0)
-		return -ENODEV;
-	return phy_driver_register(&rtl8211e_driver);
+	return phy_drivers_register(realtek_drvs, ARRAY_SIZE(realtek_drvs));
 }
 
 static void __exit realtek_exit(void)
 {
-	phy_driver_unregister(&rtl8211b_driver);
-	phy_driver_unregister(&rtl8211e_driver);
+	phy_drivers_unregister(realtek_drvs, ARRAY_SIZE(realtek_drvs));
 }
 
 module_init(realtek_init);
diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 11f34813e23f..180c49479c42 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -249,8 +249,7 @@ static int __init smsc_init(void)
 
 static void __exit smsc_exit(void)
 {
-	return phy_drivers_unregister(smsc_phy_driver,
-		ARRAY_SIZE(smsc_phy_driver));
+	phy_drivers_unregister(smsc_phy_driver, ARRAY_SIZE(smsc_phy_driver));
 }
 
 MODULE_DESCRIPTION("SMSC PHY driver");
diff --git a/drivers/net/phy/vitesse.c b/drivers/net/phy/vitesse.c
index 14372c65a7e8..5dc0935da99c 100644
--- a/drivers/net/phy/vitesse.c
+++ b/drivers/net/phy/vitesse.c
@@ -319,8 +319,7 @@ static int __init vsc82xx_init(void)
 
 static void __exit vsc82xx_exit(void)
 {
-	return phy_drivers_unregister(vsc82xx_driver,
-		ARRAY_SIZE(vsc82xx_driver));
+	phy_drivers_unregister(vsc82xx_driver, ARRAY_SIZE(vsc82xx_driver));
 }
 
 module_init(vsc82xx_init);
diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index e3923ebb693f..91d6c1272fcf 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -757,7 +757,7 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
 		err = get_filter(argp, &code);
 		if (err >= 0) {
-			struct sock_fprog fprog = {
+			struct sock_fprog_kern fprog = {
 				.len = err,
 				.filter = code,
 			};
@@ -778,7 +778,7 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
 		err = get_filter(argp, &code);
 		if (err >= 0) {
-			struct sock_fprog fprog = {
+			struct sock_fprog_kern fprog = {
 				.len = err,
 				.filter = code,
 			};
diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c
index 01805319e1e0..1aff970be33e 100644
--- a/drivers/net/ppp/pptp.c
+++ b/drivers/net/ppp/pptp.c
@@ -281,7 +281,7 @@ static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb)
 	nf_reset(skb);
 
 	skb->ip_summed = CHECKSUM_NONE;
-	ip_select_ident(skb, &rt->dst, NULL);
+	ip_select_ident(skb, NULL);
 	ip_send_check(iph);
 
 	ip_local_out(skb);
diff --git a/drivers/net/rionet.c b/drivers/net/rionet.c
index a8497183ff8b..dac7a0d9bb46 100644
--- a/drivers/net/rionet.c
+++ b/drivers/net/rionet.c
@@ -494,7 +494,7 @@ static int rionet_setup_netdev(struct rio_mport *mport, struct net_device *ndev)
 	ndev->mtu = RIO_MAX_MSG_SIZE - 14;
 	ndev->features = NETIF_F_LLTX;
 	SET_NETDEV_DEV(ndev, &mport->dev);
-	SET_ETHTOOL_OPS(ndev, &rionet_ethtool_ops);
+	ndev->ethtool_ops = &rionet_ethtool_ops;
 
 	spin_lock_init(&rnet->lock);
 	spin_lock_init(&rnet->tx_lock);
diff --git a/drivers/net/slip/slip.c b/drivers/net/slip/slip.c
index ad4a94e9ff57..87526443841f 100644
--- a/drivers/net/slip/slip.c
+++ b/drivers/net/slip/slip.c
@@ -83,6 +83,7 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/slab.h>
+#include <linux/workqueue.h>
 #include "slip.h"
 #ifdef CONFIG_INET
 #include <linux/ip.h>
@@ -416,36 +417,46 @@ static void sl_encaps(struct slip *sl, unsigned char *icp, int len)
 #endif
 }
 
-/*
- * Called by the driver when there's room for more data.  If we have
- * more packets to send, we send them here.
- */
-static void slip_write_wakeup(struct tty_struct *tty)
+/* Write out any remaining transmit buffer. Scheduled when tty is writable */
+static void slip_transmit(struct work_struct *work)
 {
+	struct slip *sl = container_of(work, struct slip, tx_work);
 	int actual;
-	struct slip *sl = tty->disc_data;
 
+	spin_lock_bh(&sl->lock);
 	/* First make sure we're connected. */
-	if (!sl || sl->magic != SLIP_MAGIC || !netif_running(sl->dev))
+	if (!sl->tty || sl->magic != SLIP_MAGIC || !netif_running(sl->dev)) {
+		spin_unlock_bh(&sl->lock);
 		return;
+	}
 
-	spin_lock_bh(&sl->lock);
 	if (sl->xleft <= 0)  {
 		/* Now serial buffer is almost free & we can start
 		 * transmission of another packet */
 		sl->dev->stats.tx_packets++;
-		clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
+		clear_bit(TTY_DO_WRITE_WAKEUP, &sl->tty->flags);
 		spin_unlock_bh(&sl->lock);
 		sl_unlock(sl);
 		return;
 	}
 
-	actual = tty->ops->write(tty, sl->xhead, sl->xleft);
+	actual = sl->tty->ops->write(sl->tty, sl->xhead, sl->xleft);
 	sl->xleft -= actual;
 	sl->xhead += actual;
 	spin_unlock_bh(&sl->lock);
 }
 
+/*
+ * Called by the driver when there's room for more data.
+ * Schedule the transmit.
+ */
+static void slip_write_wakeup(struct tty_struct *tty)
+{
+	struct slip *sl = tty->disc_data;
+
+	schedule_work(&sl->tx_work);
+}
+
 static void sl_tx_timeout(struct net_device *dev)
 {
 	struct slip *sl = netdev_priv(dev);
@@ -749,6 +760,7 @@ static struct slip *sl_alloc(dev_t line)
 	sl->magic       = SLIP_MAGIC;
 	sl->dev	      	= dev;
 	spin_lock_init(&sl->lock);
+	INIT_WORK(&sl->tx_work, slip_transmit);
 	sl->mode        = SL_MODE_DEFAULT;
 #ifdef CONFIG_SLIP_SMART
 	/* initialize timer_list struct */
@@ -872,8 +884,12 @@ static void slip_close(struct tty_struct *tty)
 	if (!sl || sl->magic != SLIP_MAGIC || sl->tty != tty)
 		return;
 
+	spin_lock_bh(&sl->lock);
 	tty->disc_data = NULL;
 	sl->tty = NULL;
+	spin_unlock_bh(&sl->lock);
+
+	flush_work(&sl->tx_work);
 
 	/* VSV = very important to remove timers */
 #ifdef CONFIG_SLIP_SMART
diff --git a/drivers/net/slip/slip.h b/drivers/net/slip/slip.h
index 67673cf1266b..cf32aadf508f 100644
--- a/drivers/net/slip/slip.h
+++ b/drivers/net/slip/slip.h
@@ -53,6 +53,7 @@ struct slip {
   struct tty_struct	*tty;		/* ptr to TTY structure		*/
   struct net_device	*dev;		/* easy for intr handling	*/
   spinlock_t		lock;
+  struct work_struct	tx_work;	/* Flushes transmit buffer	*/
 
 #ifdef SL_INCLUDE_CSLIP
   struct slcompress	*slcomp;	/* for header compression 	*/
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index ce4989be86d9..b4958c7ffa84 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -968,7 +968,7 @@ static void team_port_disable(struct team *team,
 static void __team_compute_features(struct team *team)
 {
 	struct team_port *port;
-	u32 vlan_features = TEAM_VLAN_FEATURES;
+	u32 vlan_features = TEAM_VLAN_FEATURES & NETIF_F_ALL_FOR_ALL;
 	unsigned short max_hard_header_len = ETH_HLEN;
 	unsigned int flags, dst_release_flag = IFF_XMIT_DST_RELEASE;
 
diff --git a/drivers/net/team/team_mode_loadbalance.c b/drivers/net/team/team_mode_loadbalance.c
index dbde3412ee5e..a58dfebb5512 100644
--- a/drivers/net/team/team_mode_loadbalance.c
+++ b/drivers/net/team/team_mode_loadbalance.c
@@ -49,7 +49,7 @@ struct lb_port_mapping {
 struct lb_priv_ex {
 	struct team *team;
 	struct lb_port_mapping tx_hash_to_port_mapping[LB_TX_HASHTABLE_SIZE];
-	struct sock_fprog *orig_fprog;
+	struct sock_fprog_kern *orig_fprog;
 	struct {
 		unsigned int refresh_interval; /* in tenths of second */
 		struct delayed_work refresh_dw;
@@ -241,15 +241,15 @@ static int lb_bpf_func_get(struct team *team, struct team_gsetter_ctx *ctx)
 	return 0;
 }
 
-static int __fprog_create(struct sock_fprog **pfprog, u32 data_len,
+static int __fprog_create(struct sock_fprog_kern **pfprog, u32 data_len,
 			  const void *data)
 {
-	struct sock_fprog *fprog;
+	struct sock_fprog_kern *fprog;
 	struct sock_filter *filter = (struct sock_filter *) data;
 
 	if (data_len % sizeof(struct sock_filter))
 		return -EINVAL;
-	fprog = kmalloc(sizeof(struct sock_fprog), GFP_KERNEL);
+	fprog = kmalloc(sizeof(*fprog), GFP_KERNEL);
 	if (!fprog)
 		return -ENOMEM;
 	fprog->filter = kmemdup(filter, data_len, GFP_KERNEL);
@@ -262,7 +262,7 @@ static int __fprog_create(struct sock_fprog **pfprog, u32 data_len,
 	return 0;
 }
 
-static void __fprog_destroy(struct sock_fprog *fprog)
+static void __fprog_destroy(struct sock_fprog_kern *fprog)
 {
 	kfree(fprog->filter);
 	kfree(fprog);
@@ -273,7 +273,7 @@ static int lb_bpf_func_set(struct team *team, struct team_gsetter_ctx *ctx)
 	struct lb_priv *lb_priv = get_lb_priv(team);
 	struct sk_filter *fp = NULL;
 	struct sk_filter *orig_fp;
-	struct sock_fprog *fprog = NULL;
+	struct sock_fprog_kern *fprog = NULL;
 	int err;
 
 	if (ctx->data.bin_val.len) {
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index ee328ba101e7..98bad1fb1bfb 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -498,12 +498,12 @@ static void tun_detach_all(struct net_device *dev)
 	for (i = 0; i < n; i++) {
 		tfile = rtnl_dereference(tun->tfiles[i]);
 		BUG_ON(!tfile);
-		wake_up_all(&tfile->wq.wait);
+		tfile->socket.sk->sk_data_ready(tfile->socket.sk);
 		RCU_INIT_POINTER(tfile->tun, NULL);
 		--tun->numqueues;
 	}
 	list_for_each_entry(tfile, &tun->disabled, next) {
-		wake_up_all(&tfile->wq.wait);
+		tfile->socket.sk->sk_data_ready(tfile->socket.sk);
 		RCU_INIT_POINTER(tfile->tun, NULL);
 	}
 	BUG_ON(tun->numqueues != 0);
@@ -807,8 +807,7 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Notify and wake up reader process */
 	if (tfile->flags & TUN_FASYNC)
 		kill_fasync(&tfile->fasync, SIGIO, POLL_IN);
-	wake_up_interruptible_poll(&tfile->wq.wait, POLLIN |
-				   POLLRDNORM | POLLRDBAND);
+	tfile->socket.sk->sk_data_ready(tfile->socket.sk);
 
 	rcu_read_unlock();
 	return NETDEV_TX_OK;
@@ -965,7 +964,7 @@ static unsigned int tun_chr_poll(struct file *file, poll_table *wait)
 
 	tun_debug(KERN_INFO, tun, "tun_chr_poll\n");
 
-	poll_wait(file, &tfile->wq.wait, wait);
+	poll_wait(file, sk_sleep(sk), wait);
 
 	if (!skb_queue_empty(&sk->sk_receive_queue))
 		mask |= POLLIN | POLLRDNORM;
@@ -1330,47 +1329,26 @@ done:
 static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
 			   const struct iovec *iv, ssize_t len, int noblock)
 {
-	DECLARE_WAITQUEUE(wait, current);
 	struct sk_buff *skb;
 	ssize_t ret = 0;
+	int peeked, err, off = 0;
 
 	tun_debug(KERN_INFO, tun, "tun_do_read\n");
 
-	if (unlikely(!noblock))
-		add_wait_queue(&tfile->wq.wait, &wait);
-	while (len) {
-		if (unlikely(!noblock))
-			current->state = TASK_INTERRUPTIBLE;
+	if (!len)
+		return ret;
 
-		/* Read frames from the queue */
-		if (!(skb = skb_dequeue(&tfile->socket.sk->sk_receive_queue))) {
-			if (noblock) {
-				ret = -EAGAIN;
-				break;
-			}
-			if (signal_pending(current)) {
-				ret = -ERESTARTSYS;
-				break;
-			}
-			if (tun->dev->reg_state != NETREG_REGISTERED) {
-				ret = -EIO;
-				break;
-			}
-
-			/* Nothing to read, let's sleep */
-			schedule();
-			continue;
-		}
+	if (tun->dev->reg_state != NETREG_REGISTERED)
+		return -EIO;
 
+	/* Read frames from queue */
+	skb = __skb_recv_datagram(tfile->socket.sk, noblock ? MSG_DONTWAIT : 0,
+				  &peeked, &off, &err);
+	if (skb) {
 		ret = tun_put_user(tun, tfile, skb, iv, len);
 		kfree_skb(skb);
-		break;
-	}
-
-	if (unlikely(!noblock)) {
-		current->state = TASK_RUNNING;
-		remove_wait_queue(&tfile->wq.wait, &wait);
-	}
+	} else
+		ret = err;
 
 	return ret;
 }
@@ -2199,8 +2177,8 @@ static int tun_chr_open(struct inode *inode, struct file * file)
 	tfile->flags = 0;
 	tfile->ifindex = 0;
 
-	rcu_assign_pointer(tfile->socket.wq, &tfile->wq);
 	init_waitqueue_head(&tfile->wq.wait);
+	RCU_INIT_POINTER(tfile->socket.wq, &tfile->wq);
 
 	tfile->socket.file = file;
 	tfile->socket.ops = &tun_socket_ops;
diff --git a/drivers/net/usb/catc.c b/drivers/net/usb/catc.c
index 630caf48f63a..8cfc3bb0c6a6 100644
--- a/drivers/net/usb/catc.c
+++ b/drivers/net/usb/catc.c
@@ -793,7 +793,7 @@ static int catc_probe(struct usb_interface *intf, const struct usb_device_id *id
 
 	netdev->netdev_ops = &catc_netdev_ops;
 	netdev->watchdog_timeo = TX_TIMEOUT;
-	SET_ETHTOOL_OPS(netdev, &ops);
+	netdev->ethtool_ops = &ops;
 
 	catc->usbdev = usbdev;
 	catc->netdev = netdev;
diff --git a/drivers/net/usb/cdc_mbim.c b/drivers/net/usb/cdc_mbim.c
index 2e025ddcef21..5ee7a1dbc023 100644
--- a/drivers/net/usb/cdc_mbim.c
+++ b/drivers/net/usb/cdc_mbim.c
@@ -24,13 +24,21 @@
 #include <net/ipv6.h>
 #include <net/addrconf.h>
 
+/* alternative VLAN for IP session 0 if not untagged */
+#define MBIM_IPS0_VID	4094
+
 /* driver specific data - must match cdc_ncm usage */
 struct cdc_mbim_state {
 	struct cdc_ncm_ctx *ctx;
 	atomic_t pmcount;
 	struct usb_driver *subdriver;
-	struct usb_interface *control;
-	struct usb_interface *data;
+	unsigned long _unused;
+	unsigned long flags;
+};
+
+/* flags for the cdc_mbim_state.flags field */
+enum cdc_mbim_flags {
+	FLAG_IPS0_VLAN = 1 << 0,	/* IP session 0 is tagged  */
 };
 
 /* using a counter to merge subdriver requests with our own into a combined state */
@@ -62,16 +70,91 @@ static int cdc_mbim_wdm_manage_power(struct usb_interface *intf, int status)
 	return cdc_mbim_manage_power(dev, status);
 }
 
+static int cdc_mbim_rx_add_vid(struct net_device *netdev, __be16 proto, u16 vid)
+{
+	struct usbnet *dev = netdev_priv(netdev);
+	struct cdc_mbim_state *info = (void *)&dev->data;
+
+	/* creation of this VLAN is a request to tag IP session 0 */
+	if (vid == MBIM_IPS0_VID)
+		info->flags |= FLAG_IPS0_VLAN;
+	else
+		if (vid >= 512)	/* we don't map these to MBIM session */
+			return -EINVAL;
+	return 0;
+}
+
+static int cdc_mbim_rx_kill_vid(struct net_device *netdev, __be16 proto, u16 vid)
+{
+	struct usbnet *dev = netdev_priv(netdev);
+	struct cdc_mbim_state *info = (void *)&dev->data;
+
+	/* this is a request for an untagged IP session 0 */
+	if (vid == MBIM_IPS0_VID)
+		info->flags &= ~FLAG_IPS0_VLAN;
+	return 0;
+}
+
+static const struct net_device_ops cdc_mbim_netdev_ops = {
+	.ndo_open             = usbnet_open,
+	.ndo_stop             = usbnet_stop,
+	.ndo_start_xmit       = usbnet_start_xmit,
+	.ndo_tx_timeout       = usbnet_tx_timeout,
+	.ndo_change_mtu       = usbnet_change_mtu,
+	.ndo_set_mac_address  = eth_mac_addr,
+	.ndo_validate_addr    = eth_validate_addr,
+	.ndo_vlan_rx_add_vid  = cdc_mbim_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = cdc_mbim_rx_kill_vid,
+};
+
+/* Change the control interface altsetting and update the .driver_info
+ * pointer if the matching entry after changing class codes points to
+ * a different struct
+ */
+static int cdc_mbim_set_ctrlalt(struct usbnet *dev, struct usb_interface *intf, u8 alt)
+{
+	struct usb_driver *driver = to_usb_driver(intf->dev.driver);
+	const struct usb_device_id *id;
+	struct driver_info *info;
+	int ret;
+
+	ret = usb_set_interface(dev->udev,
+				intf->cur_altsetting->desc.bInterfaceNumber,
+				alt);
+	if (ret)
+		return ret;
+
+	id = usb_match_id(intf, driver->id_table);
+	if (!id)
+		return -ENODEV;
+
+	info = (struct driver_info *)id->driver_info;
+	if (info != dev->driver_info) {
+		dev_dbg(&intf->dev, "driver_info updated to '%s'\n",
+			info->description);
+		dev->driver_info = info;
+	}
+	return 0;
+}
 
 static int cdc_mbim_bind(struct usbnet *dev, struct usb_interface *intf)
 {
 	struct cdc_ncm_ctx *ctx;
 	struct usb_driver *subdriver = ERR_PTR(-ENODEV);
 	int ret = -ENODEV;
-	u8 data_altsetting = cdc_ncm_select_altsetting(dev, intf);
+	u8 data_altsetting = 1;
 	struct cdc_mbim_state *info = (void *)&dev->data;
 
-	/* Probably NCM, defer for cdc_ncm_bind */
+	/* should we change control altsetting on a NCM/MBIM function? */
+	if (cdc_ncm_select_altsetting(intf) == CDC_NCM_COMM_ALTSETTING_MBIM) {
+		data_altsetting = CDC_NCM_DATA_ALTSETTING_MBIM;
+		ret = cdc_mbim_set_ctrlalt(dev, intf, CDC_NCM_COMM_ALTSETTING_MBIM);
+		if (ret)
+			goto err;
+		ret = -ENODEV;
+	}
+
+	/* we will hit this for NCM/MBIM functions if prefer_mbim is false */
 	if (!cdc_ncm_comm_intf_is_mbim(intf->cur_altsetting))
 		goto err;
 
@@ -101,7 +184,10 @@ static int cdc_mbim_bind(struct usbnet *dev, struct usb_interface *intf)
 	dev->net->flags |= IFF_NOARP;
 
 	/* no need to put the VLAN tci in the packet headers */
-	dev->net->features |= NETIF_F_HW_VLAN_CTAG_TX;
+	dev->net->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_FILTER;
+
+	/* monitor VLAN additions and removals */
+	dev->net->netdev_ops = &cdc_mbim_netdev_ops;
 err:
 	return ret;
 }
@@ -164,12 +250,24 @@ static struct sk_buff *cdc_mbim_tx_fixup(struct usbnet *dev, struct sk_buff *skb
 			skb_pull(skb, ETH_HLEN);
 		}
 
+		/* Is IP session <0> tagged too? */
+		if (info->flags & FLAG_IPS0_VLAN) {
+			/* drop all untagged packets */
+			if (!tci)
+				goto error;
+			/* map MBIM_IPS0_VID to IPS<0> */
+			if (tci == MBIM_IPS0_VID)
+				tci = 0;
+		}
+
 		/* mapping VLANs to MBIM sessions:
-		 *   no tag     => IPS session <0>
+		 *   no tag     => IPS session <0> if !FLAG_IPS0_VLAN
 		 *   1 - 255    => IPS session <vlanid>
 		 *   256 - 511  => DSS session <vlanid - 256>
-		 *   512 - 4095 => unsupported, drop
+		 *   512 - 4093 => unsupported, drop
+		 *   4094       => IPS session <0> if FLAG_IPS0_VLAN
 		 */
+
 		switch (tci & 0x0f00) {
 		case 0x0000: /* VLAN ID 0 - 255 */
 			if (!is_ip)
@@ -178,6 +276,8 @@ static struct sk_buff *cdc_mbim_tx_fixup(struct usbnet *dev, struct sk_buff *skb
 			c[3] = tci;
 			break;
 		case 0x0100: /* VLAN ID 256 - 511 */
+			if (is_ip)
+				goto error;
 			sign = cpu_to_le32(USB_CDC_MBIM_NDP16_DSS_SIGN);
 			c = (u8 *)&sign;
 			c[3] = tci;
@@ -223,8 +323,8 @@ static void do_neigh_solicit(struct usbnet *dev, u8 *buf, u16 tci)
 	/* need to send the NA on the VLAN dev, if any */
 	rcu_read_lock();
 	if (tci) {
-		netdev = __vlan_find_dev_deep(dev->net, htons(ETH_P_8021Q),
-					      tci);
+		netdev = __vlan_find_dev_deep_rcu(dev->net, htons(ETH_P_8021Q),
+						  tci);
 		if (!netdev) {
 			rcu_read_unlock();
 			return;
@@ -268,7 +368,7 @@ static struct sk_buff *cdc_mbim_process_dgram(struct usbnet *dev, u8 *buf, size_
 	__be16 proto = htons(ETH_P_802_3);
 	struct sk_buff *skb = NULL;
 
-	if (tci < 256) { /* IPS session? */
+	if (tci < 256 || tci == MBIM_IPS0_VID) { /* IPS session? */
 		if (len < sizeof(struct iphdr))
 			goto err;
 
@@ -320,6 +420,7 @@ static int cdc_mbim_rx_fixup(struct usbnet *dev, struct sk_buff *skb_in)
 	struct usb_cdc_ncm_dpe16 *dpe16;
 	int ndpoffset;
 	int loopcount = 50; /* arbitrary max preventing infinite loop */
+	u32 payload = 0;
 	u8 *c;
 	u16 tci;
 
@@ -338,6 +439,9 @@ next_ndp:
 	case cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN):
 		c = (u8 *)&ndp16->dwSignature;
 		tci = c[3];
+		/* tag IPS<0> packets too if MBIM_IPS0_VID exists */
+		if (!tci && info->flags & FLAG_IPS0_VLAN)
+			tci = MBIM_IPS0_VID;
 		break;
 	case cpu_to_le32(USB_CDC_MBIM_NDP16_DSS_SIGN):
 		c = (u8 *)&ndp16->dwSignature;
@@ -379,6 +483,7 @@ next_ndp:
 			if (!skb)
 				goto error;
 			usbnet_skb_return(dev, skb);
+			payload += len;	/* count payload bytes in this NTB */
 		}
 	}
 err_ndp:
@@ -387,6 +492,10 @@ err_ndp:
 	if (ndpoffset && loopcount--)
 		goto next_ndp;
 
+	/* update stats */
+	ctx->rx_overhead += skb_in->len - payload;
+	ctx->rx_ntbs++;
+
 	return 1;
 error:
 	return 0;
diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c
index 9a2bd11943eb..80a844e0ae03 100644
--- a/drivers/net/usb/cdc_ncm.c
+++ b/drivers/net/usb/cdc_ncm.c
@@ -65,19 +65,384 @@ static void cdc_ncm_tx_timeout_start(struct cdc_ncm_ctx *ctx);
 static enum hrtimer_restart cdc_ncm_tx_timer_cb(struct hrtimer *hr_timer);
 static struct usb_driver cdc_ncm_driver;
 
-static int cdc_ncm_setup(struct usbnet *dev)
+struct cdc_ncm_stats {
+	char stat_string[ETH_GSTRING_LEN];
+	int sizeof_stat;
+	int stat_offset;
+};
+
+#define CDC_NCM_STAT(str, m) { \
+		.stat_string = str, \
+		.sizeof_stat = sizeof(((struct cdc_ncm_ctx *)0)->m), \
+		.stat_offset = offsetof(struct cdc_ncm_ctx, m) }
+#define CDC_NCM_SIMPLE_STAT(m)	CDC_NCM_STAT(__stringify(m), m)
+
+static const struct cdc_ncm_stats cdc_ncm_gstrings_stats[] = {
+	CDC_NCM_SIMPLE_STAT(tx_reason_ntb_full),
+	CDC_NCM_SIMPLE_STAT(tx_reason_ndp_full),
+	CDC_NCM_SIMPLE_STAT(tx_reason_timeout),
+	CDC_NCM_SIMPLE_STAT(tx_reason_max_datagram),
+	CDC_NCM_SIMPLE_STAT(tx_overhead),
+	CDC_NCM_SIMPLE_STAT(tx_ntbs),
+	CDC_NCM_SIMPLE_STAT(rx_overhead),
+	CDC_NCM_SIMPLE_STAT(rx_ntbs),
+};
+
+static int cdc_ncm_get_sset_count(struct net_device __always_unused *netdev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return ARRAY_SIZE(cdc_ncm_gstrings_stats);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void cdc_ncm_get_ethtool_stats(struct net_device *netdev,
+				    struct ethtool_stats __always_unused *stats,
+				    u64 *data)
+{
+	struct usbnet *dev = netdev_priv(netdev);
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	int i;
+	char *p = NULL;
+
+	for (i = 0; i < ARRAY_SIZE(cdc_ncm_gstrings_stats); i++) {
+		p = (char *)ctx + cdc_ncm_gstrings_stats[i].stat_offset;
+		data[i] = (cdc_ncm_gstrings_stats[i].sizeof_stat == sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+	}
+}
+
+static void cdc_ncm_get_strings(struct net_device __always_unused *netdev, u32 stringset, u8 *data)
+{
+	u8 *p = data;
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(cdc_ncm_gstrings_stats); i++) {
+			memcpy(p, cdc_ncm_gstrings_stats[i].stat_string, ETH_GSTRING_LEN);
+			p += ETH_GSTRING_LEN;
+		}
+	}
+}
+
+static void cdc_ncm_update_rxtx_max(struct usbnet *dev, u32 new_rx, u32 new_tx);
+
+static const struct ethtool_ops cdc_ncm_ethtool_ops = {
+	.get_settings      = usbnet_get_settings,
+	.set_settings      = usbnet_set_settings,
+	.get_link          = usbnet_get_link,
+	.nway_reset        = usbnet_nway_reset,
+	.get_drvinfo       = usbnet_get_drvinfo,
+	.get_msglevel      = usbnet_get_msglevel,
+	.set_msglevel      = usbnet_set_msglevel,
+	.get_ts_info       = ethtool_op_get_ts_info,
+	.get_sset_count    = cdc_ncm_get_sset_count,
+	.get_strings       = cdc_ncm_get_strings,
+	.get_ethtool_stats = cdc_ncm_get_ethtool_stats,
+};
+
+static u32 cdc_ncm_check_rx_max(struct usbnet *dev, u32 new_rx)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u32 val, max, min;
+
+	/* clamp new_rx to sane values */
+	min = USB_CDC_NCM_NTB_MIN_IN_SIZE;
+	max = min_t(u32, CDC_NCM_NTB_MAX_SIZE_RX, le32_to_cpu(ctx->ncm_parm.dwNtbInMaxSize));
+
+	/* dwNtbInMaxSize spec violation? Use MIN size for both limits */
+	if (max < min) {
+		dev_warn(&dev->intf->dev, "dwNtbInMaxSize=%u is too small. Using %u\n",
+			 le32_to_cpu(ctx->ncm_parm.dwNtbInMaxSize), min);
+		max = min;
+	}
+
+	val = clamp_t(u32, new_rx, min, max);
+	if (val != new_rx)
+		dev_dbg(&dev->intf->dev, "rx_max must be in the [%u, %u] range\n", min, max);
+
+	return val;
+}
+
+static u32 cdc_ncm_check_tx_max(struct usbnet *dev, u32 new_tx)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u32 val, max, min;
+
+	/* clamp new_tx to sane values */
+	min = ctx->max_datagram_size + ctx->max_ndp_size + sizeof(struct usb_cdc_ncm_nth16);
+	max = min_t(u32, CDC_NCM_NTB_MAX_SIZE_TX, le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize));
+
+	/* some devices set dwNtbOutMaxSize too low for the above default */
+	min = min(min, max);
+
+	val = clamp_t(u32, new_tx, min, max);
+	if (val != new_tx)
+		dev_dbg(&dev->intf->dev, "tx_max must be in the [%u, %u] range\n", min, max);
+
+	return val;
+}
+
+static ssize_t cdc_ncm_show_min_tx_pkt(struct device *d, struct device_attribute *attr, char *buf)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+
+	return sprintf(buf, "%u\n", ctx->min_tx_pkt);
+}
+
+static ssize_t cdc_ncm_show_rx_max(struct device *d, struct device_attribute *attr, char *buf)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+
+	return sprintf(buf, "%u\n", ctx->rx_max);
+}
+
+static ssize_t cdc_ncm_show_tx_max(struct device *d, struct device_attribute *attr, char *buf)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+
+	return sprintf(buf, "%u\n", ctx->tx_max);
+}
+
+static ssize_t cdc_ncm_show_tx_timer_usecs(struct device *d, struct device_attribute *attr, char *buf)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+
+	return sprintf(buf, "%u\n", ctx->timer_interval / (u32)NSEC_PER_USEC);
+}
+
+static ssize_t cdc_ncm_store_min_tx_pkt(struct device *d,  struct device_attribute *attr, const char *buf, size_t len)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	unsigned long val;
+
+	/* no need to restrict values - anything from 0 to infinity is OK */
+	if (kstrtoul(buf, 0, &val))
+		return -EINVAL;
+
+	ctx->min_tx_pkt = val;
+	return len;
+}
+
+static ssize_t cdc_ncm_store_rx_max(struct device *d,  struct device_attribute *attr, const char *buf, size_t len)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	unsigned long val;
+
+	if (kstrtoul(buf, 0, &val) || cdc_ncm_check_rx_max(dev, val) != val)
+		return -EINVAL;
+
+	cdc_ncm_update_rxtx_max(dev, val, ctx->tx_max);
+	return len;
+}
+
+static ssize_t cdc_ncm_store_tx_max(struct device *d,  struct device_attribute *attr, const char *buf, size_t len)
 {
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
 	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	unsigned long val;
+
+	if (kstrtoul(buf, 0, &val) || cdc_ncm_check_tx_max(dev, val) != val)
+		return -EINVAL;
+
+	cdc_ncm_update_rxtx_max(dev, ctx->rx_max, val);
+	return len;
+}
+
+static ssize_t cdc_ncm_store_tx_timer_usecs(struct device *d,  struct device_attribute *attr, const char *buf, size_t len)
+{
+	struct usbnet *dev = netdev_priv(to_net_dev(d));
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	ssize_t ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret)
+		return ret;
+	if (val && (val < CDC_NCM_TIMER_INTERVAL_MIN || val > CDC_NCM_TIMER_INTERVAL_MAX))
+		return -EINVAL;
+
+	spin_lock_bh(&ctx->mtx);
+	ctx->timer_interval = val * NSEC_PER_USEC;
+	if (!ctx->timer_interval)
+		ctx->tx_timer_pending = 0;
+	spin_unlock_bh(&ctx->mtx);
+	return len;
+}
+
+static DEVICE_ATTR(min_tx_pkt, S_IRUGO | S_IWUSR, cdc_ncm_show_min_tx_pkt, cdc_ncm_store_min_tx_pkt);
+static DEVICE_ATTR(rx_max, S_IRUGO | S_IWUSR, cdc_ncm_show_rx_max, cdc_ncm_store_rx_max);
+static DEVICE_ATTR(tx_max, S_IRUGO | S_IWUSR, cdc_ncm_show_tx_max, cdc_ncm_store_tx_max);
+static DEVICE_ATTR(tx_timer_usecs, S_IRUGO | S_IWUSR, cdc_ncm_show_tx_timer_usecs, cdc_ncm_store_tx_timer_usecs);
+
+#define NCM_PARM_ATTR(name, format, tocpu)				\
+static ssize_t cdc_ncm_show_##name(struct device *d, struct device_attribute *attr, char *buf) \
+{ \
+	struct usbnet *dev = netdev_priv(to_net_dev(d)); \
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0]; \
+	return sprintf(buf, format "\n", tocpu(ctx->ncm_parm.name));	\
+} \
+static DEVICE_ATTR(name, S_IRUGO, cdc_ncm_show_##name, NULL)
+
+NCM_PARM_ATTR(bmNtbFormatsSupported, "0x%04x", le16_to_cpu);
+NCM_PARM_ATTR(dwNtbInMaxSize, "%u", le32_to_cpu);
+NCM_PARM_ATTR(wNdpInDivisor, "%u", le16_to_cpu);
+NCM_PARM_ATTR(wNdpInPayloadRemainder, "%u", le16_to_cpu);
+NCM_PARM_ATTR(wNdpInAlignment, "%u", le16_to_cpu);
+NCM_PARM_ATTR(dwNtbOutMaxSize, "%u", le32_to_cpu);
+NCM_PARM_ATTR(wNdpOutDivisor, "%u", le16_to_cpu);
+NCM_PARM_ATTR(wNdpOutPayloadRemainder, "%u", le16_to_cpu);
+NCM_PARM_ATTR(wNdpOutAlignment, "%u", le16_to_cpu);
+NCM_PARM_ATTR(wNtbOutMaxDatagrams, "%u", le16_to_cpu);
+
+static struct attribute *cdc_ncm_sysfs_attrs[] = {
+	&dev_attr_min_tx_pkt.attr,
+	&dev_attr_rx_max.attr,
+	&dev_attr_tx_max.attr,
+	&dev_attr_tx_timer_usecs.attr,
+	&dev_attr_bmNtbFormatsSupported.attr,
+	&dev_attr_dwNtbInMaxSize.attr,
+	&dev_attr_wNdpInDivisor.attr,
+	&dev_attr_wNdpInPayloadRemainder.attr,
+	&dev_attr_wNdpInAlignment.attr,
+	&dev_attr_dwNtbOutMaxSize.attr,
+	&dev_attr_wNdpOutDivisor.attr,
+	&dev_attr_wNdpOutPayloadRemainder.attr,
+	&dev_attr_wNdpOutAlignment.attr,
+	&dev_attr_wNtbOutMaxDatagrams.attr,
+	NULL,
+};
+
+static struct attribute_group cdc_ncm_sysfs_attr_group = {
+	.name = "cdc_ncm",
+	.attrs = cdc_ncm_sysfs_attrs,
+};
+
+/* handle rx_max and tx_max changes */
+static void cdc_ncm_update_rxtx_max(struct usbnet *dev, u32 new_rx, u32 new_tx)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u8 iface_no = ctx->control->cur_altsetting->desc.bInterfaceNumber;
 	u32 val;
-	u8 flags;
-	u8 iface_no;
-	int err;
-	int eth_hlen;
-	u16 mbim_mtu;
-	u16 ntb_fmt_supported;
-	__le16 max_datagram_size;
 
-	iface_no = ctx->control->cur_altsetting->desc.bInterfaceNumber;
+	val = cdc_ncm_check_rx_max(dev, new_rx);
+
+	/* inform device about NTB input size changes */
+	if (val != ctx->rx_max) {
+		__le32 dwNtbInMaxSize = cpu_to_le32(val);
+
+		dev_info(&dev->intf->dev, "setting rx_max = %u\n", val);
+
+		/* tell device to use new size */
+		if (usbnet_write_cmd(dev, USB_CDC_SET_NTB_INPUT_SIZE,
+				     USB_TYPE_CLASS | USB_DIR_OUT
+				     | USB_RECIP_INTERFACE,
+				     0, iface_no, &dwNtbInMaxSize, 4) < 0)
+			dev_dbg(&dev->intf->dev, "Setting NTB Input Size failed\n");
+		else
+			ctx->rx_max = val;
+	}
+
+	/* usbnet use these values for sizing rx queues */
+	if (dev->rx_urb_size != ctx->rx_max) {
+		dev->rx_urb_size = ctx->rx_max;
+		if (netif_running(dev->net))
+			usbnet_unlink_rx_urbs(dev);
+	}
+
+	val = cdc_ncm_check_tx_max(dev, new_tx);
+	if (val != ctx->tx_max)
+		dev_info(&dev->intf->dev, "setting tx_max = %u\n", val);
+
+	/* Adding a pad byte here if necessary simplifies the handling
+	 * in cdc_ncm_fill_tx_frame, making tx_max always represent
+	 * the real skb max size.
+	 *
+	 * We cannot use dev->maxpacket here because this is called from
+	 * .bind which is called before usbnet sets up dev->maxpacket
+	 */
+	if (val != le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize) &&
+	    val % usb_maxpacket(dev->udev, dev->out, 1) == 0)
+		val++;
+
+	/* we might need to flush any pending tx buffers if running */
+	if (netif_running(dev->net) && val > ctx->tx_max) {
+		netif_tx_lock_bh(dev->net);
+		usbnet_start_xmit(NULL, dev->net);
+		/* make sure tx_curr_skb is reallocated if it was empty */
+		if (ctx->tx_curr_skb) {
+			dev_kfree_skb_any(ctx->tx_curr_skb);
+			ctx->tx_curr_skb = NULL;
+		}
+		ctx->tx_max = val;
+		netif_tx_unlock_bh(dev->net);
+	} else {
+		ctx->tx_max = val;
+	}
+
+	dev->hard_mtu = ctx->tx_max;
+
+	/* max qlen depend on hard_mtu and rx_urb_size */
+	usbnet_update_max_qlen(dev);
+
+	/* never pad more than 3 full USB packets per transfer */
+	ctx->min_tx_pkt = clamp_t(u16, ctx->tx_max - 3 * usb_maxpacket(dev->udev, dev->out, 1),
+				  CDC_NCM_MIN_TX_PKT, ctx->tx_max);
+}
+
+/* helpers for NCM and MBIM differences */
+static u8 cdc_ncm_flags(struct usbnet *dev)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+
+	if (cdc_ncm_comm_intf_is_mbim(dev->intf->cur_altsetting) && ctx->mbim_desc)
+		return ctx->mbim_desc->bmNetworkCapabilities;
+	if (ctx->func_desc)
+		return ctx->func_desc->bmNetworkCapabilities;
+	return 0;
+}
+
+static int cdc_ncm_eth_hlen(struct usbnet *dev)
+{
+	if (cdc_ncm_comm_intf_is_mbim(dev->intf->cur_altsetting))
+		return 0;
+	return ETH_HLEN;
+}
+
+static u32 cdc_ncm_min_dgram_size(struct usbnet *dev)
+{
+	if (cdc_ncm_comm_intf_is_mbim(dev->intf->cur_altsetting))
+		return CDC_MBIM_MIN_DATAGRAM_SIZE;
+	return CDC_NCM_MIN_DATAGRAM_SIZE;
+}
+
+static u32 cdc_ncm_max_dgram_size(struct usbnet *dev)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+
+	if (cdc_ncm_comm_intf_is_mbim(dev->intf->cur_altsetting) && ctx->mbim_desc)
+		return le16_to_cpu(ctx->mbim_desc->wMaxSegmentSize);
+	if (ctx->ether_desc)
+		return le16_to_cpu(ctx->ether_desc->wMaxSegmentSize);
+	return CDC_NCM_MAX_DATAGRAM_SIZE;
+}
+
+/* initial one-time device setup.  MUST be called with the data interface
+ * in altsetting 0
+ */
+static int cdc_ncm_init(struct usbnet *dev)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u8 iface_no = ctx->control->cur_altsetting->desc.bInterfaceNumber;
+	int err;
 
 	err = usbnet_read_cmd(dev, USB_CDC_GET_NTB_PARAMETERS,
 			      USB_TYPE_CLASS | USB_DIR_IN
@@ -89,7 +454,36 @@ static int cdc_ncm_setup(struct usbnet *dev)
 		return err; /* GET_NTB_PARAMETERS is required */
 	}
 
-	/* read correct set of parameters according to device mode */
+	/* set CRC Mode */
+	if (cdc_ncm_flags(dev) & USB_CDC_NCM_NCAP_CRC_MODE) {
+		dev_dbg(&dev->intf->dev, "Setting CRC mode off\n");
+		err = usbnet_write_cmd(dev, USB_CDC_SET_CRC_MODE,
+				       USB_TYPE_CLASS | USB_DIR_OUT
+				       | USB_RECIP_INTERFACE,
+				       USB_CDC_NCM_CRC_NOT_APPENDED,
+				       iface_no, NULL, 0);
+		if (err < 0)
+			dev_err(&dev->intf->dev, "SET_CRC_MODE failed\n");
+	}
+
+	/* set NTB format, if both formats are supported.
+	 *
+	 * "The host shall only send this command while the NCM Data
+	 *  Interface is in alternate setting 0."
+	 */
+	if (le16_to_cpu(ctx->ncm_parm.bmNtbFormatsSupported) &
+						USB_CDC_NCM_NTB32_SUPPORTED) {
+		dev_dbg(&dev->intf->dev, "Setting NTB format to 16-bit\n");
+		err = usbnet_write_cmd(dev, USB_CDC_SET_NTB_FORMAT,
+				       USB_TYPE_CLASS | USB_DIR_OUT
+				       | USB_RECIP_INTERFACE,
+				       USB_CDC_NCM_NTB16_FORMAT,
+				       iface_no, NULL, 0);
+		if (err < 0)
+			dev_err(&dev->intf->dev, "SET_NTB_FORMAT failed\n");
+	}
+
+	/* set initial device values */
 	ctx->rx_max = le32_to_cpu(ctx->ncm_parm.dwNtbInMaxSize);
 	ctx->tx_max = le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize);
 	ctx->tx_remainder = le16_to_cpu(ctx->ncm_parm.wNdpOutPayloadRemainder);
@@ -97,72 +491,79 @@ static int cdc_ncm_setup(struct usbnet *dev)
 	ctx->tx_ndp_modulus = le16_to_cpu(ctx->ncm_parm.wNdpOutAlignment);
 	/* devices prior to NCM Errata shall set this field to zero */
 	ctx->tx_max_datagrams = le16_to_cpu(ctx->ncm_parm.wNtbOutMaxDatagrams);
-	ntb_fmt_supported = le16_to_cpu(ctx->ncm_parm.bmNtbFormatsSupported);
-
-	/* there are some minor differences in NCM and MBIM defaults */
-	if (cdc_ncm_comm_intf_is_mbim(ctx->control->cur_altsetting)) {
-		if (!ctx->mbim_desc)
-			return -EINVAL;
-		eth_hlen = 0;
-		flags = ctx->mbim_desc->bmNetworkCapabilities;
-		ctx->max_datagram_size = le16_to_cpu(ctx->mbim_desc->wMaxSegmentSize);
-		if (ctx->max_datagram_size < CDC_MBIM_MIN_DATAGRAM_SIZE)
-			ctx->max_datagram_size = CDC_MBIM_MIN_DATAGRAM_SIZE;
-	} else {
-		if (!ctx->func_desc)
-			return -EINVAL;
-		eth_hlen = ETH_HLEN;
-		flags = ctx->func_desc->bmNetworkCapabilities;
-		ctx->max_datagram_size = le16_to_cpu(ctx->ether_desc->wMaxSegmentSize);
-		if (ctx->max_datagram_size < CDC_NCM_MIN_DATAGRAM_SIZE)
-			ctx->max_datagram_size = CDC_NCM_MIN_DATAGRAM_SIZE;
-	}
-
-	/* common absolute max for NCM and MBIM */
-	if (ctx->max_datagram_size > CDC_NCM_MAX_DATAGRAM_SIZE)
-		ctx->max_datagram_size = CDC_NCM_MAX_DATAGRAM_SIZE;
 
 	dev_dbg(&dev->intf->dev,
 		"dwNtbInMaxSize=%u dwNtbOutMaxSize=%u wNdpOutPayloadRemainder=%u wNdpOutDivisor=%u wNdpOutAlignment=%u wNtbOutMaxDatagrams=%u flags=0x%x\n",
 		ctx->rx_max, ctx->tx_max, ctx->tx_remainder, ctx->tx_modulus,
-		ctx->tx_ndp_modulus, ctx->tx_max_datagrams, flags);
+		ctx->tx_ndp_modulus, ctx->tx_max_datagrams, cdc_ncm_flags(dev));
 
 	/* max count of tx datagrams */
 	if ((ctx->tx_max_datagrams == 0) ||
 			(ctx->tx_max_datagrams > CDC_NCM_DPT_DATAGRAMS_MAX))
 		ctx->tx_max_datagrams = CDC_NCM_DPT_DATAGRAMS_MAX;
 
-	/* verify maximum size of received NTB in bytes */
-	if (ctx->rx_max < USB_CDC_NCM_NTB_MIN_IN_SIZE) {
-		dev_dbg(&dev->intf->dev, "Using min receive length=%d\n",
-			USB_CDC_NCM_NTB_MIN_IN_SIZE);
-		ctx->rx_max = USB_CDC_NCM_NTB_MIN_IN_SIZE;
-	}
+	/* set up maximum NDP size */
+	ctx->max_ndp_size = sizeof(struct usb_cdc_ncm_ndp16) + (ctx->tx_max_datagrams + 1) * sizeof(struct usb_cdc_ncm_dpe16);
 
-	if (ctx->rx_max > CDC_NCM_NTB_MAX_SIZE_RX) {
-		dev_dbg(&dev->intf->dev, "Using default maximum receive length=%d\n",
-			CDC_NCM_NTB_MAX_SIZE_RX);
-		ctx->rx_max = CDC_NCM_NTB_MAX_SIZE_RX;
-	}
+	/* initial coalescing timer interval */
+	ctx->timer_interval = CDC_NCM_TIMER_INTERVAL_USEC * NSEC_PER_USEC;
 
-	/* inform device about NTB input size changes */
-	if (ctx->rx_max != le32_to_cpu(ctx->ncm_parm.dwNtbInMaxSize)) {
-		__le32 dwNtbInMaxSize = cpu_to_le32(ctx->rx_max);
+	return 0;
+}
 
-		err = usbnet_write_cmd(dev, USB_CDC_SET_NTB_INPUT_SIZE,
-				       USB_TYPE_CLASS | USB_DIR_OUT
-				       | USB_RECIP_INTERFACE,
-				       0, iface_no, &dwNtbInMaxSize, 4);
-		if (err < 0)
-			dev_dbg(&dev->intf->dev, "Setting NTB Input Size failed\n");
+/* set a new max datagram size */
+static void cdc_ncm_set_dgram_size(struct usbnet *dev, int new_size)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u8 iface_no = ctx->control->cur_altsetting->desc.bInterfaceNumber;
+	__le16 max_datagram_size;
+	u16 mbim_mtu;
+	int err;
+
+	/* set default based on descriptors */
+	ctx->max_datagram_size = clamp_t(u32, new_size,
+					 cdc_ncm_min_dgram_size(dev),
+					 CDC_NCM_MAX_DATAGRAM_SIZE);
+
+	/* inform the device about the selected Max Datagram Size? */
+	if (!(cdc_ncm_flags(dev) & USB_CDC_NCM_NCAP_MAX_DATAGRAM_SIZE))
+		goto out;
+
+	/* read current mtu value from device */
+	err = usbnet_read_cmd(dev, USB_CDC_GET_MAX_DATAGRAM_SIZE,
+			      USB_TYPE_CLASS | USB_DIR_IN | USB_RECIP_INTERFACE,
+			      0, iface_no, &max_datagram_size, 2);
+	if (err < 0) {
+		dev_dbg(&dev->intf->dev, "GET_MAX_DATAGRAM_SIZE failed\n");
+		goto out;
 	}
 
-	/* verify maximum size of transmitted NTB in bytes */
-	if (ctx->tx_max > CDC_NCM_NTB_MAX_SIZE_TX) {
-		dev_dbg(&dev->intf->dev, "Using default maximum transmit length=%d\n",
-			CDC_NCM_NTB_MAX_SIZE_TX);
-		ctx->tx_max = CDC_NCM_NTB_MAX_SIZE_TX;
+	if (le16_to_cpu(max_datagram_size) == ctx->max_datagram_size)
+		goto out;
+
+	max_datagram_size = cpu_to_le16(ctx->max_datagram_size);
+	err = usbnet_write_cmd(dev, USB_CDC_SET_MAX_DATAGRAM_SIZE,
+			       USB_TYPE_CLASS | USB_DIR_OUT | USB_RECIP_INTERFACE,
+			       0, iface_no, &max_datagram_size, 2);
+	if (err < 0)
+		dev_dbg(&dev->intf->dev, "SET_MAX_DATAGRAM_SIZE failed\n");
+
+out:
+	/* set MTU to max supported by the device if necessary */
+	dev->net->mtu = min_t(int, dev->net->mtu, ctx->max_datagram_size - cdc_ncm_eth_hlen(dev));
+
+	/* do not exceed operater preferred MTU */
+	if (ctx->mbim_extended_desc) {
+		mbim_mtu = le16_to_cpu(ctx->mbim_extended_desc->wMTU);
+		if (mbim_mtu != 0 && mbim_mtu < dev->net->mtu)
+			dev->net->mtu = mbim_mtu;
 	}
+}
+
+static void cdc_ncm_fix_modulus(struct usbnet *dev)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u32 val;
 
 	/*
 	 * verify that the structure alignment is:
@@ -199,68 +600,31 @@ static int cdc_ncm_setup(struct usbnet *dev)
 	}
 
 	/* adjust TX-remainder according to NCM specification. */
-	ctx->tx_remainder = ((ctx->tx_remainder - eth_hlen) &
+	ctx->tx_remainder = ((ctx->tx_remainder - cdc_ncm_eth_hlen(dev)) &
 			     (ctx->tx_modulus - 1));
+}
 
-	/* additional configuration */
-
-	/* set CRC Mode */
-	if (flags & USB_CDC_NCM_NCAP_CRC_MODE) {
-		err = usbnet_write_cmd(dev, USB_CDC_SET_CRC_MODE,
-				       USB_TYPE_CLASS | USB_DIR_OUT
-				       | USB_RECIP_INTERFACE,
-				       USB_CDC_NCM_CRC_NOT_APPENDED,
-				       iface_no, NULL, 0);
-		if (err < 0)
-			dev_dbg(&dev->intf->dev, "Setting CRC mode off failed\n");
-	}
-
-	/* set NTB format, if both formats are supported */
-	if (ntb_fmt_supported & USB_CDC_NCM_NTH32_SIGN) {
-		err = usbnet_write_cmd(dev, USB_CDC_SET_NTB_FORMAT,
-				       USB_TYPE_CLASS | USB_DIR_OUT
-				       | USB_RECIP_INTERFACE,
-				       USB_CDC_NCM_NTB16_FORMAT,
-				       iface_no, NULL, 0);
-		if (err < 0)
-			dev_dbg(&dev->intf->dev, "Setting NTB format to 16-bit failed\n");
-	}
-
-	/* inform the device about the selected Max Datagram Size */
-	if (!(flags & USB_CDC_NCM_NCAP_MAX_DATAGRAM_SIZE))
-		goto out;
-
-	/* read current mtu value from device */
-	err = usbnet_read_cmd(dev, USB_CDC_GET_MAX_DATAGRAM_SIZE,
-			      USB_TYPE_CLASS | USB_DIR_IN | USB_RECIP_INTERFACE,
-			      0, iface_no, &max_datagram_size, 2);
-	if (err < 0) {
-		dev_dbg(&dev->intf->dev, "GET_MAX_DATAGRAM_SIZE failed\n");
-		goto out;
-	}
-
-	if (le16_to_cpu(max_datagram_size) == ctx->max_datagram_size)
-		goto out;
+static int cdc_ncm_setup(struct usbnet *dev)
+{
+	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
+	u32 def_rx, def_tx;
 
-	max_datagram_size = cpu_to_le16(ctx->max_datagram_size);
-	err = usbnet_write_cmd(dev, USB_CDC_SET_MAX_DATAGRAM_SIZE,
-			       USB_TYPE_CLASS | USB_DIR_OUT | USB_RECIP_INTERFACE,
-			       0, iface_no, &max_datagram_size, 2);
-	if (err < 0)
-		dev_dbg(&dev->intf->dev, "SET_MAX_DATAGRAM_SIZE failed\n");
+	/* be conservative when selecting intial buffer size to
+	 * increase the number of hosts this will work for
+	 */
+	def_rx = min_t(u32, CDC_NCM_NTB_DEF_SIZE_RX,
+		       le32_to_cpu(ctx->ncm_parm.dwNtbInMaxSize));
+	def_tx = min_t(u32, CDC_NCM_NTB_DEF_SIZE_TX,
+		       le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize));
 
-out:
-	/* set MTU to max supported by the device if necessary */
-	if (dev->net->mtu > ctx->max_datagram_size - eth_hlen)
-		dev->net->mtu = ctx->max_datagram_size - eth_hlen;
+	/* clamp rx_max and tx_max and inform device */
+	cdc_ncm_update_rxtx_max(dev, def_rx, def_tx);
 
-	/* do not exceed operater preferred MTU */
-	if (ctx->mbim_extended_desc) {
-		mbim_mtu = le16_to_cpu(ctx->mbim_extended_desc->wMTU);
-		if (mbim_mtu != 0 && mbim_mtu < dev->net->mtu)
-			dev->net->mtu = mbim_mtu;
-	}
+	/* sanitize the modulus and remainder values */
+	cdc_ncm_fix_modulus(dev);
 
+	/* set max datagram size */
+	cdc_ncm_set_dgram_size(dev, cdc_ncm_max_dgram_size(dev));
 	return 0;
 }
 
@@ -424,10 +788,21 @@ advance:
 	}
 
 	/* check if we got everything */
-	if (!ctx->data || (!ctx->mbim_desc && !ctx->ether_desc)) {
-		dev_dbg(&intf->dev, "CDC descriptors missing\n");
+	if (!ctx->data) {
+		dev_dbg(&intf->dev, "CDC Union missing and no IAD found\n");
 		goto error;
 	}
+	if (cdc_ncm_comm_intf_is_mbim(intf->cur_altsetting)) {
+		if (!ctx->mbim_desc) {
+			dev_dbg(&intf->dev, "MBIM functional descriptor missing\n");
+			goto error;
+		}
+	} else {
+		if (!ctx->ether_desc || !ctx->func_desc) {
+			dev_dbg(&intf->dev, "NCM or ECM functional descriptors missing\n");
+			goto error;
+		}
+	}
 
 	/* claim data interface, if different from control */
 	if (ctx->data != ctx->control) {
@@ -447,8 +822,8 @@ advance:
 		goto error2;
 	}
 
-	/* initialize data interface */
-	if (cdc_ncm_setup(dev))
+	/* initialize basic device settings */
+	if (cdc_ncm_init(dev))
 		goto error2;
 
 	/* configure data interface */
@@ -477,18 +852,14 @@ advance:
 		dev_info(&intf->dev, "MAC-Address: %pM\n", dev->net->dev_addr);
 	}
 
-	/* usbnet use these values for sizing tx/rx queues */
-	dev->hard_mtu = ctx->tx_max;
-	dev->rx_urb_size = ctx->rx_max;
+	/* finish setting up the device specific data */
+	cdc_ncm_setup(dev);
 
-	/* cdc_ncm_setup will override dwNtbOutMaxSize if it is
-	 * outside the sane range. Adding a pad byte here if necessary
-	 * simplifies the handling in cdc_ncm_fill_tx_frame, making
-	 * tx_max always represent the real skb max size.
-	 */
-	if (ctx->tx_max != le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize) &&
-	    ctx->tx_max % usb_maxpacket(dev->udev, dev->out, 1) == 0)
-		ctx->tx_max++;
+	/* override ethtool_ops */
+	dev->net->ethtool_ops = &cdc_ncm_ethtool_ops;
+
+	/* add our sysfs attrs */
+	dev->net->sysfs_groups[0] = &cdc_ncm_sysfs_attr_group;
 
 	return 0;
 
@@ -541,10 +912,10 @@ void cdc_ncm_unbind(struct usbnet *dev, struct usb_interface *intf)
 }
 EXPORT_SYMBOL_GPL(cdc_ncm_unbind);
 
-/* Select the MBIM altsetting iff it is preferred and available,
- * returning the number of the corresponding data interface altsetting
+/* Return the number of the MBIM control interface altsetting iff it
+ * is preferred and available,
  */
-u8 cdc_ncm_select_altsetting(struct usbnet *dev, struct usb_interface *intf)
+u8 cdc_ncm_select_altsetting(struct usb_interface *intf)
 {
 	struct usb_host_interface *alt;
 
@@ -563,15 +934,15 @@ u8 cdc_ncm_select_altsetting(struct usbnet *dev, struct usb_interface *intf)
 	 *   the rules given in section 6 (USB Device Model) of this
 	 *   specification."
 	 */
-	if (prefer_mbim && intf->num_altsetting == 2) {
+	if (intf->num_altsetting < 2)
+		return intf->cur_altsetting->desc.bAlternateSetting;
+
+	if (prefer_mbim) {
 		alt = usb_altnum_to_altsetting(intf, CDC_NCM_COMM_ALTSETTING_MBIM);
-		if (alt && cdc_ncm_comm_intf_is_mbim(alt) &&
-		    !usb_set_interface(dev->udev,
-				       intf->cur_altsetting->desc.bInterfaceNumber,
-				       CDC_NCM_COMM_ALTSETTING_MBIM))
-			return CDC_NCM_DATA_ALTSETTING_MBIM;
+		if (alt && cdc_ncm_comm_intf_is_mbim(alt))
+			return CDC_NCM_COMM_ALTSETTING_MBIM;
 	}
-	return CDC_NCM_DATA_ALTSETTING_NCM;
+	return CDC_NCM_COMM_ALTSETTING_NCM;
 }
 EXPORT_SYMBOL_GPL(cdc_ncm_select_altsetting);
 
@@ -580,12 +951,11 @@ static int cdc_ncm_bind(struct usbnet *dev, struct usb_interface *intf)
 	int ret;
 
 	/* MBIM backwards compatible function? */
-	cdc_ncm_select_altsetting(dev, intf);
-	if (cdc_ncm_comm_intf_is_mbim(intf->cur_altsetting))
+	if (cdc_ncm_select_altsetting(intf) != CDC_NCM_COMM_ALTSETTING_NCM)
 		return -ENODEV;
 
-	/* NCM data altsetting is always 1 */
-	ret = cdc_ncm_bind_common(dev, intf, 1);
+	/* The NCM data altsetting is fixed */
+	ret = cdc_ncm_bind_common(dev, intf, CDC_NCM_DATA_ALTSETTING_NCM);
 
 	/*
 	 * We should get an event when network connection is "connected" or
@@ -628,7 +998,7 @@ static struct usb_cdc_ncm_ndp16 *cdc_ncm_ndp(struct cdc_ncm_ctx *ctx, struct sk_
 	cdc_ncm_align_tail(skb, ctx->tx_ndp_modulus, 0, ctx->tx_max);
 
 	/* verify that there is room for the NDP and the datagram (reserve) */
-	if ((ctx->tx_max - skb->len - reserve) < CDC_NCM_NDP_SIZE)
+	if ((ctx->tx_max - skb->len - reserve) < ctx->max_ndp_size)
 		return NULL;
 
 	/* link to it */
@@ -638,7 +1008,7 @@ static struct usb_cdc_ncm_ndp16 *cdc_ncm_ndp(struct cdc_ncm_ctx *ctx, struct sk_
 		nth16->wNdpIndex = cpu_to_le16(skb->len);
 
 	/* push a new empty NDP */
-	ndp16 = (struct usb_cdc_ncm_ndp16 *)memset(skb_put(skb, CDC_NCM_NDP_SIZE), 0, CDC_NCM_NDP_SIZE);
+	ndp16 = (struct usb_cdc_ncm_ndp16 *)memset(skb_put(skb, ctx->max_ndp_size), 0, ctx->max_ndp_size);
 	ndp16->dwSignature = sign;
 	ndp16->wLength = cpu_to_le16(sizeof(struct usb_cdc_ncm_ndp16) + sizeof(struct usb_cdc_ncm_dpe16));
 	return ndp16;
@@ -683,6 +1053,9 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 
 		/* count total number of frames in this NTB */
 		ctx->tx_curr_frame_num = 0;
+
+		/* recent payload counter for this skb_out */
+		ctx->tx_curr_frame_payload = 0;
 	}
 
 	for (n = ctx->tx_curr_frame_num; n < ctx->tx_max_datagrams; n++) {
@@ -720,6 +1093,7 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 				ctx->tx_rem_sign = sign;
 				skb = NULL;
 				ready2send = 1;
+				ctx->tx_reason_ntb_full++;	/* count reason for transmitting */
 			}
 			break;
 		}
@@ -733,12 +1107,14 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 		ndp16->dpe16[index].wDatagramIndex = cpu_to_le16(skb_out->len);
 		ndp16->wLength = cpu_to_le16(ndplen + sizeof(struct usb_cdc_ncm_dpe16));
 		memcpy(skb_put(skb_out, skb->len), skb->data, skb->len);
+		ctx->tx_curr_frame_payload += skb->len;	/* count real tx payload data */
 		dev_kfree_skb_any(skb);
 		skb = NULL;
 
 		/* send now if this NDP is full */
 		if (index >= CDC_NCM_DPT_DATAGRAMS_MAX) {
 			ready2send = 1;
+			ctx->tx_reason_ndp_full++;	/* count reason for transmitting */
 			break;
 		}
 	}
@@ -758,7 +1134,7 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 		ctx->tx_curr_skb = skb_out;
 		goto exit_no_skb;
 
-	} else if ((n < ctx->tx_max_datagrams) && (ready2send == 0)) {
+	} else if ((n < ctx->tx_max_datagrams) && (ready2send == 0) && (ctx->timer_interval > 0)) {
 		/* wait for more frames */
 		/* push variables */
 		ctx->tx_curr_skb = skb_out;
@@ -768,11 +1144,13 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 		goto exit_no_skb;
 
 	} else {
+		if (n == ctx->tx_max_datagrams)
+			ctx->tx_reason_max_datagram++;	/* count reason for transmitting */
 		/* frame goes out */
 		/* variables will be reset at next call */
 	}
 
-	/* If collected data size is less or equal CDC_NCM_MIN_TX_PKT
+	/* If collected data size is less or equal ctx->min_tx_pkt
 	 * bytes, we send buffers as it is. If we get more data, it
 	 * would be more efficient for USB HS mobile device with DMA
 	 * engine to receive a full size NTB, than canceling DMA
@@ -782,7 +1160,7 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 	 * a ZLP after full sized NTBs.
 	 */
 	if (!(dev->driver_info->flags & FLAG_SEND_ZLP) &&
-	    skb_out->len > CDC_NCM_MIN_TX_PKT)
+	    skb_out->len > ctx->min_tx_pkt)
 		memset(skb_put(skb_out, ctx->tx_max - skb_out->len), 0,
 		       ctx->tx_max - skb_out->len);
 	else if (skb_out->len < ctx->tx_max && (skb_out->len % dev->maxpacket) == 0)
@@ -795,11 +1173,22 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign)
 	/* return skb */
 	ctx->tx_curr_skb = NULL;
 	dev->net->stats.tx_packets += ctx->tx_curr_frame_num;
+
+	/* keep private stats: framing overhead and number of NTBs */
+	ctx->tx_overhead += skb_out->len - ctx->tx_curr_frame_payload;
+	ctx->tx_ntbs++;
+
+	/* usbnet has already counted all the framing overhead.
+	 * Adjust the stats so that the tx_bytes counter show real
+	 * payload data instead.
+	 */
+	dev->net->stats.tx_bytes -= skb_out->len - ctx->tx_curr_frame_payload;
+
 	return skb_out;
 
 exit_no_skb:
-	/* Start timer, if there is a remaining skb */
-	if (ctx->tx_curr_skb != NULL)
+	/* Start timer, if there is a remaining non-empty skb */
+	if (ctx->tx_curr_skb != NULL && n > 0)
 		cdc_ncm_tx_timeout_start(ctx);
 	return NULL;
 }
@@ -810,7 +1199,7 @@ static void cdc_ncm_tx_timeout_start(struct cdc_ncm_ctx *ctx)
 	/* start timer, if not already started */
 	if (!(hrtimer_active(&ctx->tx_timer) || atomic_read(&ctx->stop)))
 		hrtimer_start(&ctx->tx_timer,
-				ktime_set(0, CDC_NCM_TIMER_INTERVAL),
+				ktime_set(0, ctx->timer_interval),
 				HRTIMER_MODE_REL);
 }
 
@@ -835,6 +1224,7 @@ static void cdc_ncm_txpath_bh(unsigned long param)
 		cdc_ncm_tx_timeout_start(ctx);
 		spin_unlock_bh(&ctx->mtx);
 	} else if (dev->net != NULL) {
+		ctx->tx_reason_timeout++;	/* count reason for transmitting */
 		spin_unlock_bh(&ctx->mtx);
 		netif_tx_lock_bh(dev->net);
 		usbnet_start_xmit(NULL, dev->net);
@@ -970,6 +1360,7 @@ int cdc_ncm_rx_fixup(struct usbnet *dev, struct sk_buff *skb_in)
 	struct usb_cdc_ncm_dpe16 *dpe16;
 	int ndpoffset;
 	int loopcount = 50; /* arbitrary max preventing infinite loop */
+	u32 payload = 0;
 
 	ndpoffset = cdc_ncm_rx_verify_nth16(ctx, skb_in);
 	if (ndpoffset < 0)
@@ -1015,13 +1406,13 @@ next_ndp:
 			break;
 
 		} else {
-			skb = skb_clone(skb_in, GFP_ATOMIC);
+			/* create a fresh copy to reduce truesize */
+			skb = netdev_alloc_skb_ip_align(dev->net,  len);
 			if (!skb)
 				goto error;
-			skb->len = len;
-			skb->data = ((u8 *)skb_in->data) + offset;
-			skb_set_tail_pointer(skb, len);
+			memcpy(skb_put(skb, len), skb_in->data + offset, len);
 			usbnet_skb_return(dev, skb);
+			payload += len;	/* count payload bytes in this NTB */
 		}
 	}
 err_ndp:
@@ -1030,6 +1421,10 @@ err_ndp:
 	if (ndpoffset && loopcount--)
 		goto next_ndp;
 
+	/* update stats */
+	ctx->rx_overhead += skb_in->len - payload;
+	ctx->rx_ntbs++;
+
 	return 1;
 error:
 	return 0;
@@ -1049,14 +1444,14 @@ cdc_ncm_speed_change(struct usbnet *dev,
 	 */
 	if ((tx_speed > 1000000) && (rx_speed > 1000000)) {
 		netif_info(dev, link, dev->net,
-		       "%u mbit/s downlink %u mbit/s uplink\n",
-		       (unsigned int)(rx_speed / 1000000U),
-		       (unsigned int)(tx_speed / 1000000U));
+			   "%u mbit/s downlink %u mbit/s uplink\n",
+			   (unsigned int)(rx_speed / 1000000U),
+			   (unsigned int)(tx_speed / 1000000U));
 	} else {
 		netif_info(dev, link, dev->net,
-		       "%u kbit/s downlink %u kbit/s uplink\n",
-		       (unsigned int)(rx_speed / 1000U),
-		       (unsigned int)(tx_speed / 1000U));
+			   "%u kbit/s downlink %u kbit/s uplink\n",
+			   (unsigned int)(rx_speed / 1000U),
+			   (unsigned int)(tx_speed / 1000U));
 	}
 }
 
@@ -1086,11 +1481,10 @@ static void cdc_ncm_status(struct usbnet *dev, struct urb *urb)
 		 * USB_CDC_NOTIFY_NETWORK_CONNECTION notification shall be
 		 * sent by device after USB_CDC_NOTIFY_SPEED_CHANGE.
 		 */
-		ctx->connected = le16_to_cpu(event->wValue);
 		netif_info(dev, link, dev->net,
 			   "network connection: %sconnected\n",
-			   ctx->connected ? "" : "dis");
-		usbnet_link_change(dev, ctx->connected, 0);
+			   !!event->wValue ? "" : "dis");
+		usbnet_link_change(dev, !!event->wValue, 0);
 		break;
 
 	case USB_CDC_NOTIFY_SPEED_CHANGE:
@@ -1110,23 +1504,11 @@ static void cdc_ncm_status(struct usbnet *dev, struct urb *urb)
 	}
 }
 
-static int cdc_ncm_check_connect(struct usbnet *dev)
-{
-	struct cdc_ncm_ctx *ctx;
-
-	ctx = (struct cdc_ncm_ctx *)dev->data[0];
-	if (ctx == NULL)
-		return 1;	/* disconnected */
-
-	return !ctx->connected;
-}
-
 static const struct driver_info cdc_ncm_info = {
 	.description = "CDC NCM",
 	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET,
 	.bind = cdc_ncm_bind,
 	.unbind = cdc_ncm_unbind,
-	.check_connect = cdc_ncm_check_connect,
 	.manage_power = usbnet_manage_power,
 	.status = cdc_ncm_status,
 	.rx_fixup = cdc_ncm_rx_fixup,
@@ -1140,7 +1522,6 @@ static const struct driver_info wwan_info = {
 			| FLAG_WWAN,
 	.bind = cdc_ncm_bind,
 	.unbind = cdc_ncm_unbind,
-	.check_connect = cdc_ncm_check_connect,
 	.manage_power = usbnet_manage_power,
 	.status = cdc_ncm_status,
 	.rx_fixup = cdc_ncm_rx_fixup,
@@ -1154,7 +1535,6 @@ static const struct driver_info wwan_noarp_info = {
 			| FLAG_WWAN | FLAG_NOARP,
 	.bind = cdc_ncm_bind,
 	.unbind = cdc_ncm_unbind,
-	.check_connect = cdc_ncm_check_connect,
 	.manage_power = usbnet_manage_power,
 	.status = cdc_ncm_status,
 	.rx_fixup = cdc_ncm_rx_fixup,
diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c
index 660bd5ea9fc0..a3a05869309d 100644
--- a/drivers/net/usb/hso.c
+++ b/drivers/net/usb/hso.c
@@ -2425,7 +2425,7 @@ static void hso_net_init(struct net_device *net)
 	net->type = ARPHRD_NONE;
 	net->mtu = DEFAULT_MTU - 14;
 	net->tx_queue_len = 10;
-	SET_ETHTOOL_OPS(net, &ops);
+	net->ethtool_ops = &ops;
 
 	/* and initialize the semaphore */
 	spin_lock_init(&hso_net->net_lock);
diff --git a/drivers/net/usb/huawei_cdc_ncm.c b/drivers/net/usb/huawei_cdc_ncm.c
index 312178d7b698..5d95a13dbe2a 100644
--- a/drivers/net/usb/huawei_cdc_ncm.c
+++ b/drivers/net/usb/huawei_cdc_ncm.c
@@ -84,12 +84,13 @@ static int huawei_cdc_ncm_bind(struct usbnet *usbnet_dev,
 	ctx = drvstate->ctx;
 
 	if (usbnet_dev->status)
-		/* CDC-WMC r1.1 requires wMaxCommand to be "at least 256
-		 * decimal (0x100)"
+		/* The wMaxCommand buffer must be big enough to hold
+		 * any message from the modem. Experience has shown
+		 * that some replies are more than 256 bytes long
 		 */
 		subdriver = usb_cdc_wdm_register(ctx->control,
 						 &usbnet_dev->status->desc,
-						 256, /* wMaxCommand */
+						 1024, /* wMaxCommand */
 						 huawei_cdc_ncm_wdm_manage_power);
 	if (IS_ERR(subdriver)) {
 		ret = PTR_ERR(subdriver);
@@ -172,24 +173,11 @@ err:
 	return ret;
 }
 
-static int huawei_cdc_ncm_check_connect(struct usbnet *usbnet_dev)
-{
-	struct cdc_ncm_ctx *ctx;
-
-	ctx = (struct cdc_ncm_ctx *)usbnet_dev->data[0];
-
-	if (ctx == NULL)
-		return 1; /* disconnected */
-
-	return !ctx->connected;
-}
-
 static const struct driver_info huawei_cdc_ncm_info = {
 	.description = "Huawei CDC NCM device",
 	.flags = FLAG_NO_SETINT | FLAG_MULTI_PACKET | FLAG_WWAN,
 	.bind = huawei_cdc_ncm_bind,
 	.unbind = huawei_cdc_ncm_unbind,
-	.check_connect = huawei_cdc_ncm_check_connect,
 	.manage_power = huawei_cdc_ncm_manage_power,
 	.rx_fixup = cdc_ncm_rx_fixup,
 	.tx_fixup = cdc_ncm_tx_fixup,
diff --git a/drivers/net/usb/ipheth.c b/drivers/net/usb/ipheth.c
index 973275fef250..76465b117b72 100644
--- a/drivers/net/usb/ipheth.c
+++ b/drivers/net/usb/ipheth.c
@@ -534,7 +534,7 @@ static int ipheth_probe(struct usb_interface *intf,
 	usb_set_intfdata(intf, dev);
 
 	SET_NETDEV_DEV(netdev, &intf->dev);
-	SET_ETHTOOL_OPS(netdev, &ops);
+	netdev->ethtool_ops = &ops;
 
 	retval = register_netdev(netdev);
 	if (retval) {
diff --git a/drivers/net/usb/kaweth.c b/drivers/net/usb/kaweth.c
index a359d3bb7c5b..dcb6d33141e0 100644
--- a/drivers/net/usb/kaweth.c
+++ b/drivers/net/usb/kaweth.c
@@ -1171,7 +1171,7 @@ err_fw:
 	netdev->netdev_ops = &kaweth_netdev_ops;
 	netdev->watchdog_timeo = KAWETH_TX_TIMEOUT;
 	netdev->mtu = le16_to_cpu(kaweth->configuration.segment_size);
-	SET_ETHTOOL_OPS(netdev, &ops);
+	netdev->ethtool_ops = &ops;
 
 	/* kaweth is zeroed as part of alloc_netdev */
 	INIT_DELAYED_WORK(&kaweth->lowmem_work, kaweth_resubmit_tl);
diff --git a/drivers/net/usb/pegasus.c b/drivers/net/usb/pegasus.c
index 03e8a15d7deb..f84080215915 100644
--- a/drivers/net/usb/pegasus.c
+++ b/drivers/net/usb/pegasus.c
@@ -1159,7 +1159,7 @@ static int pegasus_probe(struct usb_interface *intf,
 
 	net->watchdog_timeo = PEGASUS_TX_TIMEOUT;
 	net->netdev_ops = &pegasus_netdev_ops;
-	SET_ETHTOOL_OPS(net, &ops);
+	net->ethtool_ops = &ops;
 	pegasus->mii.dev = net;
 	pegasus->mii.mdio_read = mdio_read;
 	pegasus->mii.mdio_write = mdio_write;
diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c
index dc4bf06948c7..cf62d7e8329f 100644
--- a/drivers/net/usb/qmi_wwan.c
+++ b/drivers/net/usb/qmi_wwan.c
@@ -763,7 +763,12 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x2357, 0x9000, 4)},	/* TP-LINK MA260 */
 	{QMI_FIXED_INTF(0x1bc7, 0x1200, 5)},	/* Telit LE920 */
 	{QMI_FIXED_INTF(0x1bc7, 0x1201, 2)},	/* Telit LE920 */
-	{QMI_FIXED_INTF(0x0b3c, 0xc005, 6)},    /* Olivetti Olicard 200 */
+	{QMI_FIXED_INTF(0x0b3c, 0xc000, 4)},	/* Olivetti Olicard 100 */
+	{QMI_FIXED_INTF(0x0b3c, 0xc001, 4)},	/* Olivetti Olicard 120 */
+	{QMI_FIXED_INTF(0x0b3c, 0xc002, 4)},	/* Olivetti Olicard 140 */
+	{QMI_FIXED_INTF(0x0b3c, 0xc004, 6)},	/* Olivetti Olicard 155 */
+	{QMI_FIXED_INTF(0x0b3c, 0xc005, 6)},	/* Olivetti Olicard 200 */
+	{QMI_FIXED_INTF(0x0b3c, 0xc00a, 6)},	/* Olivetti Olicard 160 */
 	{QMI_FIXED_INTF(0x0b3c, 0xc00b, 4)},	/* Olivetti Olicard 500 */
 	{QMI_FIXED_INTF(0x1e2d, 0x0060, 4)},	/* Cinterion PLxx */
 	{QMI_FIXED_INTF(0x1e2d, 0x0053, 4)},	/* Cinterion PHxx,PXxx */
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 3fbfb0869030..25431965a625 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -630,12 +630,10 @@ int set_registers(struct r8152 *tp, u16 value, u16 index, u16 size, void *data)
 	int ret;
 	void *tmp;
 
-	tmp = kmalloc(size, GFP_KERNEL);
+	tmp = kmemdup(data, size, GFP_KERNEL);
 	if (!tmp)
 		return -ENOMEM;
 
-	memcpy(tmp, data, size);
-
 	ret = usb_control_msg(tp->udev, usb_sndctrlpipe(tp->udev, 0),
 			       RTL8152_REQ_SET_REGS, RTL8152_REQT_WRITE,
 			       value, index, tmp, size, 500);
@@ -3452,7 +3450,7 @@ static int rtl8152_probe(struct usb_interface *intf,
 			      NETIF_F_TSO | NETIF_F_FRAGLIST |
 			      NETIF_F_IPV6_CSUM | NETIF_F_TSO6;
 
-	SET_ETHTOOL_OPS(netdev, &ops);
+	netdev->ethtool_ops = &ops;
 	netif_set_gso_max_size(netdev, RTL_LIMITED_TSO_SIZE);
 
 	tp->mii.dev = netdev;
diff --git a/drivers/net/usb/rtl8150.c b/drivers/net/usb/rtl8150.c
index da2c4583bd2d..6e87e5710048 100644
--- a/drivers/net/usb/rtl8150.c
+++ b/drivers/net/usb/rtl8150.c
@@ -878,7 +878,7 @@ static int rtl8150_probe(struct usb_interface *intf,
 	dev->netdev = netdev;
 	netdev->netdev_ops = &rtl8150_netdev_ops;
 	netdev->watchdog_timeo = RTL8150_TX_TIMEOUT;
-	SET_ETHTOOL_OPS(netdev, &ops);
+	netdev->ethtool_ops = &ops;
 	dev->intr_interval = 100;	/* 100ms */
 
 	if (!alloc_all_urbs(dev)) {
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 8a852b5f215f..7d9f84a91f37 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1646,7 +1646,7 @@ static int virtnet_probe(struct virtio_device *vdev)
 	dev->netdev_ops = &virtnet_netdev;
 	dev->features = NETIF_F_HIGHDMA;
 
-	SET_ETHTOOL_OPS(dev, &virtnet_ethtool_ops);
+	dev->ethtool_ops = &virtnet_ethtool_ops;
 	SET_NETDEV_DEV(dev, &vdev->dev);
 
 	/* Do we support "hardware" checksums? */
@@ -1724,6 +1724,13 @@ static int virtnet_probe(struct virtio_device *vdev)
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ))
 		vi->has_cvq = true;
 
+	if (vi->any_header_sg) {
+		if (vi->mergeable_rx_bufs)
+			dev->needed_headroom = sizeof(struct virtio_net_hdr_mrg_rxbuf);
+		else
+			dev->needed_headroom = sizeof(struct virtio_net_hdr);
+	}
+
 	/* Use single tx/rx queue pair as default */
 	vi->curr_queue_pairs = 1;
 	vi->max_queue_pairs = max_queue_pairs;
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 97394345e5dd..b76f7dcde0db 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -2589,8 +2589,8 @@ vmxnet3_open(struct net_device *netdev)
 	for (i = 0; i < adapter->num_tx_queues; i++)
 		spin_lock_init(&adapter->tx_queue[i].tx_lock);
 
-	err = vmxnet3_create_queues(adapter, VMXNET3_DEF_TX_RING_SIZE,
-				    VMXNET3_DEF_RX_RING_SIZE,
+	err = vmxnet3_create_queues(adapter, adapter->tx_ring_size,
+				    adapter->rx_ring_size,
 				    VMXNET3_DEF_RX_RING_SIZE);
 	if (err)
 		goto queue_err;
@@ -2968,6 +2968,9 @@ vmxnet3_probe_device(struct pci_dev *pdev,
 	adapter->netdev = netdev;
 	adapter->pdev = pdev;
 
+	adapter->tx_ring_size = VMXNET3_DEF_TX_RING_SIZE;
+	adapter->rx_ring_size = VMXNET3_DEF_RX_RING_SIZE;
+
 	spin_lock_init(&adapter->cmd_lock);
 	adapter->adapter_pa = dma_map_single(&adapter->pdev->dev, adapter,
 					     sizeof(struct vmxnet3_adapter),
diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index 600ab56c0008..b725fd9e7803 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -431,8 +431,8 @@ vmxnet3_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
 		ethtool_cmd_speed_set(ecmd, adapter->link_speed);
 		ecmd->duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(ecmd, -1);
-		ecmd->duplex = -1;
+		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
+		ecmd->duplex = DUPLEX_UNKNOWN;
 	}
 	return 0;
 }
@@ -449,8 +449,8 @@ vmxnet3_get_ringparam(struct net_device *netdev,
 	param->rx_mini_max_pending = 0;
 	param->rx_jumbo_max_pending = 0;
 
-	param->rx_pending = adapter->rx_queue[0].rx_ring[0].size;
-	param->tx_pending = adapter->tx_queue[0].tx_ring.size;
+	param->rx_pending = adapter->rx_ring_size;
+	param->tx_pending = adapter->tx_ring_size;
 	param->rx_mini_pending = 0;
 	param->rx_jumbo_pending = 0;
 }
@@ -529,9 +529,11 @@ vmxnet3_set_ringparam(struct net_device *netdev,
 			 * size */
 			netdev_err(netdev, "failed to apply new sizes, "
 				   "try the default ones\n");
+			new_rx_ring_size = VMXNET3_DEF_RX_RING_SIZE;
+			new_tx_ring_size = VMXNET3_DEF_TX_RING_SIZE;
 			err = vmxnet3_create_queues(adapter,
-						    VMXNET3_DEF_TX_RING_SIZE,
-						    VMXNET3_DEF_RX_RING_SIZE,
+						    new_tx_ring_size,
+						    new_rx_ring_size,
 						    VMXNET3_DEF_RX_RING_SIZE);
 			if (err) {
 				netdev_err(netdev, "failed to create queues "
@@ -545,6 +547,8 @@ vmxnet3_set_ringparam(struct net_device *netdev,
 			netdev_err(netdev, "failed to re-activate, error %d."
 				   " Closing it\n", err);
 	}
+	adapter->tx_ring_size = new_tx_ring_size;
+	adapter->rx_ring_size = new_rx_ring_size;
 
 out:
 	clear_bit(VMXNET3_STATE_BIT_RESETTING, &adapter->state);
@@ -579,7 +583,7 @@ vmxnet3_get_rss_indir_size(struct net_device *netdev)
 }
 
 static int
-vmxnet3_get_rss_indir(struct net_device *netdev, u32 *p)
+vmxnet3_get_rss(struct net_device *netdev, u32 *p, u8 *key)
 {
 	struct vmxnet3_adapter *adapter = netdev_priv(netdev);
 	struct UPT1_RSSConf *rssConf = adapter->rss_conf;
@@ -592,7 +596,7 @@ vmxnet3_get_rss_indir(struct net_device *netdev, u32 *p)
 }
 
 static int
-vmxnet3_set_rss_indir(struct net_device *netdev, const u32 *p)
+vmxnet3_set_rss(struct net_device *netdev, const u32 *p, const u8 *key)
 {
 	unsigned int i;
 	unsigned long flags;
@@ -628,12 +632,12 @@ static const struct ethtool_ops vmxnet3_ethtool_ops = {
 	.get_rxnfc         = vmxnet3_get_rxnfc,
 #ifdef VMXNET3_RSS
 	.get_rxfh_indir_size = vmxnet3_get_rss_indir_size,
-	.get_rxfh_indir    = vmxnet3_get_rss_indir,
-	.set_rxfh_indir    = vmxnet3_set_rss_indir,
+	.get_rxfh          = vmxnet3_get_rss,
+	.set_rxfh          = vmxnet3_set_rss,
 #endif
 };
 
 void vmxnet3_set_ethtool_ops(struct net_device *netdev)
 {
-	SET_ETHTOOL_OPS(netdev, &vmxnet3_ethtool_ops);
+	netdev->ethtool_ops = &vmxnet3_ethtool_ops;
 }
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index 190569d02450..29ee77f2c97f 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -349,6 +349,11 @@ struct vmxnet3_adapter {
 	u32     link_speed; /* in mbps */
 
 	u64     tx_timeout_count;
+
+	/* Ring sizes */
+	u32 tx_ring_size;
+	u32 rx_ring_size;
+
 	struct work_struct work;
 
 	unsigned long  state;    /* VMXNET3_STATE_BIT_xxx */
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 4dbb2ed85b97..ade33ef82823 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -127,6 +127,7 @@ struct vxlan_dev {
 	struct list_head  next;		/* vxlan's per namespace list */
 	struct vxlan_sock *vn_sock;	/* listening socket */
 	struct net_device *dev;
+	struct net	  *net;		/* netns for packet i/o */
 	struct vxlan_rdst default_dst;	/* default destination */
 	union vxlan_addr  saddr;	/* source address */
 	__be16		  dst_port;
@@ -134,7 +135,7 @@ struct vxlan_dev {
 	__u16		  port_max;
 	__u8		  tos;		/* TOS override */
 	__u8		  ttl;
-	u32		  flags;	/* VXLAN_F_* below */
+	u32		  flags;	/* VXLAN_F_* in vxlan.h */
 
 	struct work_struct sock_work;
 	struct work_struct igmp_join;
@@ -149,13 +150,6 @@ struct vxlan_dev {
 	struct hlist_head fdb_head[FDB_HASH_SIZE];
 };
 
-#define VXLAN_F_LEARN	0x01
-#define VXLAN_F_PROXY	0x02
-#define VXLAN_F_RSC	0x04
-#define VXLAN_F_L2MISS	0x08
-#define VXLAN_F_L3MISS	0x10
-#define VXLAN_F_IPV6	0x20 /* internal flag */
-
 /* salt for hash table */
 static u32 vxlan_salt __read_mostly;
 static struct workqueue_struct *vxlan_wq;
@@ -571,6 +565,7 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head, struct sk_buff
 			goto out;
 	}
 	skb_gro_pull(skb, sizeof(struct vxlanhdr)); /* pull vxlan header */
+	skb_gro_postpull_rcsum(skb, vh, sizeof(struct vxlanhdr));
 
 	off_eth = skb_gro_offset(skb);
 	hlen = off_eth + sizeof(*eh);
@@ -605,6 +600,7 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head, struct sk_buff
 	}
 
 	skb_gro_pull(skb, sizeof(*eh)); /* pull inner eth header */
+	skb_gro_postpull_rcsum(skb, eh, sizeof(*eh));
 	pp = ptype->callbacks.gro_receive(head, skb);
 
 out_unlock:
@@ -1160,15 +1156,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!vs)
 		goto drop;
 
-	/* If the NIC driver gave us an encapsulated packet
-	 * with the encapsulation mark, the device checksummed it
-	 * for us. Otherwise force the upper layers to verify it.
-	 */
-	if ((skb->ip_summed != CHECKSUM_UNNECESSARY && skb->ip_summed != CHECKSUM_PARTIAL) ||
-	    !skb->encapsulation)
-		skb->ip_summed = CHECKSUM_NONE;
-
-	skb->encapsulation = 0;
+	skb_pop_rcv_encapsulation(skb);
 
 	vs->rcv(vs, skb, vxh->vx_vni);
 	return 0;
@@ -1203,7 +1191,9 @@ static void vxlan_rcv(struct vxlan_sock *vs,
 
 	remote_ip = &vxlan->default_dst.remote_ip;
 	skb_reset_mac_header(skb);
+	skb_scrub_packet(skb, !net_eq(vxlan->net, dev_net(vxlan->dev)));
 	skb->protocol = eth_type_trans(skb, vxlan->dev);
+	skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
 
 	/* Ignore packet loops (and multicast echo) */
 	if (ether_addr_equal(eth_hdr(skb)->h_source, vxlan->dev->dev_addr))
@@ -1599,18 +1589,11 @@ __be16 vxlan_src_port(__u16 port_min, __u16 port_max, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(vxlan_src_port);
 
-static int handle_offloads(struct sk_buff *skb)
+static inline struct sk_buff *vxlan_handle_offloads(struct sk_buff *skb,
+						    bool udp_csum)
 {
-	if (skb_is_gso(skb)) {
-		int err = skb_unclone(skb, GFP_ATOMIC);
-		if (unlikely(err))
-			return err;
-
-		skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL;
-	} else if (skb->ip_summed != CHECKSUM_PARTIAL)
-		skb->ip_summed = CHECKSUM_NONE;
-
-	return 0;
+	int type = udp_csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
+	return iptunnel_handle_offloads(skb, udp_csum, type);
 }
 
 #if IS_ENABLED(CONFIG_IPV6)
@@ -1618,7 +1601,8 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs,
 			   struct dst_entry *dst, struct sk_buff *skb,
 			   struct net_device *dev, struct in6_addr *saddr,
 			   struct in6_addr *daddr, __u8 prio, __u8 ttl,
-			   __be16 src_port, __be16 dst_port, __be32 vni)
+			   __be16 src_port, __be16 dst_port, __be32 vni,
+			   bool xnet)
 {
 	struct ipv6hdr *ip6h;
 	struct vxlanhdr *vxh;
@@ -1626,12 +1610,11 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs,
 	int min_headroom;
 	int err;
 
-	if (!skb->encapsulation) {
-		skb_reset_inner_headers(skb);
-		skb->encapsulation = 1;
-	}
+	skb = vxlan_handle_offloads(skb, !udp_get_no_check6_tx(vs->sock->sk));
+	if (IS_ERR(skb))
+		return -EINVAL;
 
-	skb_scrub_packet(skb, false);
+	skb_scrub_packet(skb, xnet);
 
 	min_headroom = LL_RESERVED_SPACE(dst->dev) + dst->header_len
 			+ VXLAN_HLEN + sizeof(struct ipv6hdr)
@@ -1663,27 +1646,14 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs,
 	uh->source = src_port;
 
 	uh->len = htons(skb->len);
-	uh->check = 0;
 
 	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
 	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
 			      IPSKB_REROUTED);
 	skb_dst_set(skb, dst);
 
-	if (!skb_is_gso(skb) && !(dst->dev->features & NETIF_F_IPV6_CSUM)) {
-		__wsum csum = skb_checksum(skb, 0, skb->len, 0);
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-		uh->check = csum_ipv6_magic(saddr, daddr, skb->len,
-					    IPPROTO_UDP, csum);
-		if (uh->check == 0)
-			uh->check = CSUM_MANGLED_0;
-	} else {
-		skb->ip_summed = CHECKSUM_PARTIAL;
-		skb->csum_start = skb_transport_header(skb) - skb->head;
-		skb->csum_offset = offsetof(struct udphdr, check);
-		uh->check = ~csum_ipv6_magic(saddr, daddr,
-					     skb->len, IPPROTO_UDP, 0);
-	}
+	udp6_set_csum(udp_get_no_check6_tx(vs->sock->sk), skb,
+		      saddr, daddr, skb->len);
 
 	__skb_push(skb, sizeof(*ip6h));
 	skb_reset_network_header(skb);
@@ -1699,10 +1669,6 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs,
 	ip6h->daddr	  = *daddr;
 	ip6h->saddr	  = *saddr;
 
-	err = handle_offloads(skb);
-	if (err)
-		return err;
-
 	ip6tunnel_xmit(skb, dev);
 	return 0;
 }
@@ -1711,17 +1677,16 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs,
 int vxlan_xmit_skb(struct vxlan_sock *vs,
 		   struct rtable *rt, struct sk_buff *skb,
 		   __be32 src, __be32 dst, __u8 tos, __u8 ttl, __be16 df,
-		   __be16 src_port, __be16 dst_port, __be32 vni)
+		   __be16 src_port, __be16 dst_port, __be32 vni, bool xnet)
 {
 	struct vxlanhdr *vxh;
 	struct udphdr *uh;
 	int min_headroom;
 	int err;
 
-	if (!skb->encapsulation) {
-		skb_reset_inner_headers(skb);
-		skb->encapsulation = 1;
-	}
+	skb = vxlan_handle_offloads(skb, !vs->sock->sk->sk_no_check_tx);
+	if (IS_ERR(skb))
+		return -EINVAL;
 
 	min_headroom = LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len
 			+ VXLAN_HLEN + sizeof(struct iphdr)
@@ -1753,14 +1718,12 @@ int vxlan_xmit_skb(struct vxlan_sock *vs,
 	uh->source = src_port;
 
 	uh->len = htons(skb->len);
-	uh->check = 0;
 
-	err = handle_offloads(skb);
-	if (err)
-		return err;
+	udp_set_csum(vs->sock->sk->sk_no_check_tx, skb,
+		     src, dst, skb->len);
 
 	return iptunnel_xmit(vs->sock->sk, rt, skb, src, dst, IPPROTO_UDP,
-			     tos, ttl, df, false);
+			     tos, ttl, df, xnet);
 }
 EXPORT_SYMBOL_GPL(vxlan_xmit_skb);
 
@@ -1853,7 +1816,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		fl4.daddr = dst->sin.sin_addr.s_addr;
 		fl4.saddr = vxlan->saddr.sin.sin_addr.s_addr;
 
-		rt = ip_route_output_key(dev_net(dev), &fl4);
+		rt = ip_route_output_key(vxlan->net, &fl4);
 		if (IS_ERR(rt)) {
 			netdev_dbg(dev, "no route to %pI4\n",
 				   &dst->sin.sin_addr.s_addr);
@@ -1874,7 +1837,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			struct vxlan_dev *dst_vxlan;
 
 			ip_rt_put(rt);
-			dst_vxlan = vxlan_find_vni(dev_net(dev), vni, dst_port);
+			dst_vxlan = vxlan_find_vni(vxlan->net, vni, dst_port);
 			if (!dst_vxlan)
 				goto tx_error;
 			vxlan_encap_bypass(skb, vxlan, dst_vxlan);
@@ -1887,7 +1850,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		err = vxlan_xmit_skb(vxlan->vn_sock, rt, skb,
 				     fl4.saddr, dst->sin.sin_addr.s_addr,
 				     tos, ttl, df, src_port, dst_port,
-				     htonl(vni << 8));
+				     htonl(vni << 8),
+				     !net_eq(vxlan->net, dev_net(vxlan->dev)));
 
 		if (err < 0)
 			goto rt_tx_error;
@@ -1927,7 +1891,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			struct vxlan_dev *dst_vxlan;
 
 			dst_release(ndst);
-			dst_vxlan = vxlan_find_vni(dev_net(dev), vni, dst_port);
+			dst_vxlan = vxlan_find_vni(vxlan->net, vni, dst_port);
 			if (!dst_vxlan)
 				goto tx_error;
 			vxlan_encap_bypass(skb, vxlan, dst_vxlan);
@@ -1938,7 +1902,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 
 		err = vxlan6_xmit_skb(vxlan->vn_sock, ndst, skb,
 				      dev, &fl6.saddr, &fl6.daddr, 0, ttl,
-				      src_port, dst_port, htonl(vni << 8));
+				      src_port, dst_port, htonl(vni << 8),
+				      !net_eq(vxlan->net, dev_net(vxlan->dev)));
 #endif
 	}
 
@@ -2082,7 +2047,7 @@ static void vxlan_vs_add_dev(struct vxlan_sock *vs, struct vxlan_dev *vxlan)
 static int vxlan_init(struct net_device *dev)
 {
 	struct vxlan_dev *vxlan = netdev_priv(dev);
-	struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
+	struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
 	struct vxlan_sock *vs;
 
 	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
@@ -2090,7 +2055,7 @@ static int vxlan_init(struct net_device *dev)
 		return -ENOMEM;
 
 	spin_lock(&vn->sock_lock);
-	vs = vxlan_find_sock(dev_net(dev), vxlan->dst_port);
+	vs = vxlan_find_sock(vxlan->net, vxlan->dst_port);
 	if (vs) {
 		/* If we have a socket with same port already, reuse it */
 		atomic_inc(&vs->refcnt);
@@ -2172,8 +2137,8 @@ static void vxlan_flush(struct vxlan_dev *vxlan)
 /* Cleanup timer and forwarding table on shutdown */
 static int vxlan_stop(struct net_device *dev)
 {
-	struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
 	struct vxlan_dev *vxlan = netdev_priv(dev);
+	struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
 	struct vxlan_sock *vs = vxlan->vn_sock;
 
 	if (vs && vxlan_addr_multicast(&vxlan->default_dst.remote_ip) &&
@@ -2202,7 +2167,7 @@ static int vxlan_change_mtu(struct net_device *dev, int new_mtu)
 	struct net_device *lowerdev;
 	int max_mtu;
 
-	lowerdev = __dev_get_by_index(dev_net(dev), dst->remote_ifindex);
+	lowerdev = __dev_get_by_index(vxlan->net, dst->remote_ifindex);
 	if (lowerdev == NULL)
 		return eth_change_mtu(dev, new_mtu);
 
@@ -2275,9 +2240,9 @@ static void vxlan_setup(struct net_device *dev)
 	eth_hw_addr_random(dev);
 	ether_setup(dev);
 	if (vxlan->default_dst.remote_ip.sa.sa_family == AF_INET6)
-		dev->hard_header_len = ETH_HLEN + VXLAN6_HEADROOM;
+		dev->needed_headroom = ETH_HLEN + VXLAN6_HEADROOM;
 	else
-		dev->hard_header_len = ETH_HLEN + VXLAN_HEADROOM;
+		dev->needed_headroom = ETH_HLEN + VXLAN_HEADROOM;
 
 	dev->netdev_ops = &vxlan_netdev_ops;
 	dev->destructor = free_netdev;
@@ -2285,7 +2250,6 @@ static void vxlan_setup(struct net_device *dev)
 
 	dev->tx_queue_len = 0;
 	dev->features	|= NETIF_F_LLTX;
-	dev->features	|= NETIF_F_NETNS_LOCAL;
 	dev->features	|= NETIF_F_SG | NETIF_F_HW_CSUM;
 	dev->features   |= NETIF_F_RXCSUM;
 	dev->features   |= NETIF_F_GSO_SOFTWARE;
@@ -2401,7 +2365,7 @@ static void vxlan_del_work(struct work_struct *work)
  * could be used for both IPv4 and IPv6 communications, but
  * users may set bindv6only=1.
  */
-static struct socket *create_v6_sock(struct net *net, __be16 port)
+static struct socket *create_v6_sock(struct net *net, __be16 port, u32 flags)
 {
 	struct sock *sk;
 	struct socket *sock;
@@ -2438,18 +2402,25 @@ static struct socket *create_v6_sock(struct net *net, __be16 port)
 
 	/* Disable multicast loopback */
 	inet_sk(sk)->mc_loop = 0;
+
+	if (flags & VXLAN_F_UDP_ZERO_CSUM6_TX)
+		udp_set_no_check6_tx(sk, true);
+
+	if (flags & VXLAN_F_UDP_ZERO_CSUM6_RX)
+		udp_set_no_check6_rx(sk, true);
+
 	return sock;
 }
 
 #else
 
-static struct socket *create_v6_sock(struct net *net, __be16 port)
+static struct socket *create_v6_sock(struct net *net, __be16 port, u32 flags)
 {
 		return ERR_PTR(-EPFNOSUPPORT);
 }
 #endif
 
-static struct socket *create_v4_sock(struct net *net, __be16 port)
+static struct socket *create_v4_sock(struct net *net, __be16 port, u32 flags)
 {
 	struct sock *sk;
 	struct socket *sock;
@@ -2482,18 +2453,24 @@ static struct socket *create_v4_sock(struct net *net, __be16 port)
 
 	/* Disable multicast loopback */
 	inet_sk(sk)->mc_loop = 0;
+
+	if (!(flags & VXLAN_F_UDP_CSUM))
+		sock->sk->sk_no_check_tx = 1;
+
 	return sock;
 }
 
 /* Create new listen socket if needed */
 static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port,
-					      vxlan_rcv_t *rcv, void *data, bool ipv6)
+					      vxlan_rcv_t *rcv, void *data,
+					      u32 flags)
 {
 	struct vxlan_net *vn = net_generic(net, vxlan_net_id);
 	struct vxlan_sock *vs;
 	struct socket *sock;
 	struct sock *sk;
 	unsigned int h;
+	bool ipv6 = !!(flags & VXLAN_F_IPV6);
 
 	vs = kzalloc(sizeof(*vs), GFP_KERNEL);
 	if (!vs)
@@ -2505,9 +2482,9 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port,
 	INIT_WORK(&vs->del_work, vxlan_del_work);
 
 	if (ipv6)
-		sock = create_v6_sock(net, port);
+		sock = create_v6_sock(net, port, flags);
 	else
-		sock = create_v4_sock(net, port);
+		sock = create_v4_sock(net, port, flags);
 	if (IS_ERR(sock)) {
 		kfree(vs);
 		return ERR_CAST(sock);
@@ -2545,12 +2522,12 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port,
 
 struct vxlan_sock *vxlan_sock_add(struct net *net, __be16 port,
 				  vxlan_rcv_t *rcv, void *data,
-				  bool no_share, bool ipv6)
+				  bool no_share, u32 flags)
 {
 	struct vxlan_net *vn = net_generic(net, vxlan_net_id);
 	struct vxlan_sock *vs;
 
-	vs = vxlan_socket_create(net, port, rcv, data, ipv6);
+	vs = vxlan_socket_create(net, port, rcv, data, flags);
 	if (!IS_ERR(vs))
 		return vs;
 
@@ -2578,12 +2555,12 @@ EXPORT_SYMBOL_GPL(vxlan_sock_add);
 static void vxlan_sock_work(struct work_struct *work)
 {
 	struct vxlan_dev *vxlan = container_of(work, struct vxlan_dev, sock_work);
-	struct net *net = dev_net(vxlan->dev);
+	struct net *net = vxlan->net;
 	struct vxlan_net *vn = net_generic(net, vxlan_net_id);
 	__be16 port = vxlan->dst_port;
 	struct vxlan_sock *nvs;
 
-	nvs = vxlan_sock_add(net, port, vxlan_rcv, NULL, false, vxlan->flags & VXLAN_F_IPV6);
+	nvs = vxlan_sock_add(net, port, vxlan_rcv, NULL, false, vxlan->flags);
 	spin_lock(&vn->sock_lock);
 	if (!IS_ERR(nvs))
 		vxlan_vs_add_dev(nvs, vxlan);
@@ -2605,6 +2582,8 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
 	if (!data[IFLA_VXLAN_ID])
 		return -EINVAL;
 
+	vxlan->net = dev_net(dev);
+
 	vni = nla_get_u32(data[IFLA_VXLAN_ID]);
 	dst->remote_vni = vni;
 
@@ -2660,8 +2639,7 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
 		if (!tb[IFLA_MTU])
 			dev->mtu = lowerdev->mtu - (use_ipv6 ? VXLAN6_HEADROOM : VXLAN_HEADROOM);
 
-		/* update header length based on lower device */
-		dev->hard_header_len = lowerdev->hard_header_len +
+		dev->needed_headroom = lowerdev->hard_header_len +
 				       (use_ipv6 ? VXLAN6_HEADROOM : VXLAN_HEADROOM);
 	} else if (use_ipv6)
 		vxlan->flags |= VXLAN_F_IPV6;
@@ -2705,12 +2683,23 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
 	if (data[IFLA_VXLAN_PORT])
 		vxlan->dst_port = nla_get_be16(data[IFLA_VXLAN_PORT]);
 
+	if (data[IFLA_VXLAN_UDP_CSUM] && nla_get_u8(data[IFLA_VXLAN_UDP_CSUM]))
+		vxlan->flags |= VXLAN_F_UDP_CSUM;
+
+	if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX] &&
+	    nla_get_u8(data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX]))
+		vxlan->flags |= VXLAN_F_UDP_ZERO_CSUM6_TX;
+
+	if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX] &&
+	    nla_get_u8(data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX]))
+		vxlan->flags |= VXLAN_F_UDP_ZERO_CSUM6_RX;
+
 	if (vxlan_find_vni(net, vni, vxlan->dst_port)) {
 		pr_info("duplicate VNI %u\n", vni);
 		return -EEXIST;
 	}
 
-	SET_ETHTOOL_OPS(dev, &vxlan_ethtool_ops);
+	dev->ethtool_ops = &vxlan_ethtool_ops;
 
 	/* create an fdb entry for a valid default destination */
 	if (!vxlan_addr_any(&vxlan->default_dst.remote_ip)) {
@@ -2739,8 +2728,8 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
 
 static void vxlan_dellink(struct net_device *dev, struct list_head *head)
 {
-	struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
 	struct vxlan_dev *vxlan = netdev_priv(dev);
+	struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
 
 	spin_lock(&vn->sock_lock);
 	if (!hlist_unhashed(&vxlan->hlist))
@@ -2768,7 +2757,10 @@ static size_t vxlan_get_size(const struct net_device *dev)
 		nla_total_size(sizeof(__u32)) +	/* IFLA_VXLAN_AGEING */
 		nla_total_size(sizeof(__u32)) +	/* IFLA_VXLAN_LIMIT */
 		nla_total_size(sizeof(struct ifla_vxlan_port_range)) +
-		nla_total_size(sizeof(__be16))+ /* IFLA_VXLAN_PORT */
+		nla_total_size(sizeof(__be16)) + /* IFLA_VXLAN_PORT */
+		nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_UDP_CSUM */
+		nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_UDP_ZERO_CSUM6_TX */
+		nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_UDP_ZERO_CSUM6_RX */
 		0;
 }
 
@@ -2828,7 +2820,13 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
 			!!(vxlan->flags & VXLAN_F_L3MISS)) ||
 	    nla_put_u32(skb, IFLA_VXLAN_AGEING, vxlan->age_interval) ||
 	    nla_put_u32(skb, IFLA_VXLAN_LIMIT, vxlan->addrmax) ||
-	    nla_put_be16(skb, IFLA_VXLAN_PORT, vxlan->dst_port))
+	    nla_put_be16(skb, IFLA_VXLAN_PORT, vxlan->dst_port) ||
+	    nla_put_u8(skb, IFLA_VXLAN_UDP_CSUM,
+			!!(vxlan->flags & VXLAN_F_UDP_CSUM)) ||
+	    nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_TX,
+			!!(vxlan->flags & VXLAN_F_UDP_ZERO_CSUM6_TX)) ||
+	    nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_RX,
+			!!(vxlan->flags & VXLAN_F_UDP_ZERO_CSUM6_RX)))
 		goto nla_put_failure;
 
 	if (nla_put(skb, IFLA_VXLAN_PORT_RANGE, sizeof(ports), &ports))
@@ -2905,8 +2903,33 @@ static __net_init int vxlan_init_net(struct net *net)
 	return 0;
 }
 
+static void __net_exit vxlan_exit_net(struct net *net)
+{
+	struct vxlan_net *vn = net_generic(net, vxlan_net_id);
+	struct vxlan_dev *vxlan, *next;
+	struct net_device *dev, *aux;
+	LIST_HEAD(list);
+
+	rtnl_lock();
+	for_each_netdev_safe(net, dev, aux)
+		if (dev->rtnl_link_ops == &vxlan_link_ops)
+			unregister_netdevice_queue(dev, &list);
+
+	list_for_each_entry_safe(vxlan, next, &vn->vxlan_list, next) {
+		/* If vxlan->dev is in the same netns, it has already been added
+		 * to the list by the previous loop.
+		 */
+		if (!net_eq(dev_net(vxlan->dev), net))
+			unregister_netdevice_queue(dev, &list);
+	}
+
+	unregister_netdevice_many(&list);
+	rtnl_unlock();
+}
+
 static struct pernet_operations vxlan_net_ops = {
 	.init = vxlan_init_net,
+	.exit = vxlan_exit_net,
 	.id   = &vxlan_net_id,
 	.size = sizeof(struct vxlan_net),
 };
diff --git a/drivers/net/wan/farsync.c b/drivers/net/wan/farsync.c
index bcfff0d62de4..93ace042d0aa 100644
--- a/drivers/net/wan/farsync.c
+++ b/drivers/net/wan/farsync.c
@@ -26,6 +26,7 @@
 #include <linux/ioport.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
+#include <linux/delay.h>
 #include <linux/if.h>
 #include <linux/hdlc.h>
 #include <asm/io.h>
@@ -678,7 +679,6 @@ static inline void
 fst_cpureset(struct fst_card_info *card)
 {
 	unsigned char interrupt_line_register;
-	unsigned long j = jiffies + 1;
 	unsigned int regval;
 
 	if (card->family == FST_FAMILY_TXU) {
@@ -696,16 +696,12 @@ fst_cpureset(struct fst_card_info *card)
 		/*
 		 * We are delaying here to allow the 9054 to reset itself
 		 */
-		j = jiffies + 1;
-		while (jiffies < j)
-			/* Do nothing */ ;
+		usleep_range(10, 20);
 		outw(0x240f, card->pci_conf + CNTRL_9054 + 2);
 		/*
 		 * We are delaying here to allow the 9054 to reload its eeprom
 		 */
-		j = jiffies + 1;
-		while (jiffies < j)
-			/* Do nothing */ ;
+		usleep_range(10, 20);
 		outw(0x040f, card->pci_conf + CNTRL_9054 + 2);
 
 		if (pci_write_config_byte
@@ -886,20 +882,18 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
  *      Receive a frame through the DMA
  */
 static inline void
-fst_rx_dma(struct fst_card_info *card, dma_addr_t skb,
-	   dma_addr_t mem, int len)
+fst_rx_dma(struct fst_card_info *card, dma_addr_t dma, u32 mem, int len)
 {
 	/*
 	 * This routine will setup the DMA and start it
 	 */
 
-	dbg(DBG_RX, "In fst_rx_dma %lx %lx %d\n",
-	    (unsigned long) skb, (unsigned long) mem, len);
+	dbg(DBG_RX, "In fst_rx_dma %x %x %d\n", (u32)dma, mem, len);
 	if (card->dmarx_in_progress) {
 		dbg(DBG_ASS, "In fst_rx_dma while dma in progress\n");
 	}
 
-	outl(skb, card->pci_conf + DMAPADR0);	/* Copy to here */
+	outl(dma, card->pci_conf + DMAPADR0);	/* Copy to here */
 	outl(mem, card->pci_conf + DMALADR0);	/* from here */
 	outl(len, card->pci_conf + DMASIZ0);	/* for this length */
 	outl(0x00000000c, card->pci_conf + DMADPR0);	/* In this direction */
@@ -915,20 +909,19 @@ fst_rx_dma(struct fst_card_info *card, dma_addr_t skb,
  *      Send a frame through the DMA
  */
 static inline void
-fst_tx_dma(struct fst_card_info *card, unsigned char *skb,
-	   unsigned char *mem, int len)
+fst_tx_dma(struct fst_card_info *card, dma_addr_t dma, u32 mem, int len)
 {
 	/*
 	 * This routine will setup the DMA and start it.
 	 */
 
-	dbg(DBG_TX, "In fst_tx_dma %p %p %d\n", skb, mem, len);
+	dbg(DBG_TX, "In fst_tx_dma %x %x %d\n", (u32)dma, mem, len);
 	if (card->dmatx_in_progress) {
 		dbg(DBG_ASS, "In fst_tx_dma while dma in progress\n");
 	}
 
-	outl((unsigned long) skb, card->pci_conf + DMAPADR1);	/* Copy from here */
-	outl((unsigned long) mem, card->pci_conf + DMALADR1);	/* to here */
+	outl(dma, card->pci_conf + DMAPADR1);	/* Copy from here */
+	outl(mem, card->pci_conf + DMALADR1);	/* to here */
 	outl(len, card->pci_conf + DMASIZ1);	/* for this length */
 	outl(0x000000004, card->pci_conf + DMADPR1);	/* In this direction */
 
@@ -1405,9 +1398,7 @@ do_bottom_half_tx(struct fst_card_info *card)
 					card->dma_len_tx = skb->len;
 					card->dma_txpos = port->txpos;
 					fst_tx_dma(card,
-						   (char *) card->
-						   tx_dma_handle_card,
-						   (char *)
+						   card->tx_dma_handle_card,
 						   BUF_OFFSET(txBuffer[pi]
 							      [port->txpos][0]),
 						   skb->len);
diff --git a/drivers/net/wan/sdla.c b/drivers/net/wan/sdla.c
index de3bbf43fc5a..cdd45fb8a1f6 100644
--- a/drivers/net/wan/sdla.c
+++ b/drivers/net/wan/sdla.c
@@ -1322,10 +1322,6 @@ NOTE:  This is rather a useless action right now, as the
 
 static int sdla_change_mtu(struct net_device *dev, int new_mtu)
 {
-	struct frad_local *flp;
-
-	flp = netdev_priv(dev);
-
 	if (netif_running(dev))
 		return -EBUSY;
 
diff --git a/drivers/net/wimax/i2400m/control.c b/drivers/net/wimax/i2400m/control.c
index 4a01e5c7fe09..4c417903e9be 100644
--- a/drivers/net/wimax/i2400m/control.c
+++ b/drivers/net/wimax/i2400m/control.c
@@ -1061,7 +1061,7 @@ int i2400m_firmware_check(struct i2400m *i2400m)
 		goto error_bad_major;
 	}
 	result = 0;
-	if (minor < I2400M_HDIv_MINOR_2 && minor > I2400M_HDIv_MINOR)
+	if (minor > I2400M_HDIv_MINOR_2 || minor < I2400M_HDIv_MINOR)
 		dev_warn(dev, "untested minor fw version %u.%u.%u\n",
 			 major, minor, branch);
 	/* Yes, we ignore the branch -- we don't have to track it */
diff --git a/drivers/net/wimax/i2400m/driver.c b/drivers/net/wimax/i2400m/driver.c
index 9c34d2fccfac..9c78090e72f8 100644
--- a/drivers/net/wimax/i2400m/driver.c
+++ b/drivers/net/wimax/i2400m/driver.c
@@ -500,26 +500,23 @@ int i2400m_pm_notifier(struct notifier_block *notifier,
  */
 int i2400m_pre_reset(struct i2400m *i2400m)
 {
-	int result;
 	struct device *dev = i2400m_dev(i2400m);
 
 	d_fnstart(3, dev, "(i2400m %p)\n", i2400m);
 	d_printf(1, dev, "pre-reset shut down\n");
 
-	result = 0;
 	mutex_lock(&i2400m->init_mutex);
 	if (i2400m->updown) {
 		netif_tx_disable(i2400m->wimax_dev.net_dev);
 		__i2400m_dev_stop(i2400m);
-		result = 0;
 		/* down't set updown to zero -- this way
 		 * post_reset can restore properly */
 	}
 	mutex_unlock(&i2400m->init_mutex);
 	if (i2400m->bus_release)
 		i2400m->bus_release(i2400m);
-	d_fnend(3, dev, "(i2400m %p) = %d\n", i2400m, result);
-	return result;
+	d_fnend(3, dev, "(i2400m %p) = 0\n", i2400m);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(i2400m_pre_reset);
 
diff --git a/drivers/net/wireless/at76c50x-usb.c b/drivers/net/wireless/at76c50x-usb.c
index 99b3bfa717d5..d48776e4f343 100644
--- a/drivers/net/wireless/at76c50x-usb.c
+++ b/drivers/net/wireless/at76c50x-usb.c
@@ -365,15 +365,15 @@ static inline unsigned long at76_get_timeout(struct dfu_status *s)
 static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 				int manifest_sync_timeout)
 {
-	u8 *block;
-	struct dfu_status dfu_stat_buf;
 	int ret = 0;
 	int need_dfu_state = 1;
 	int is_done = 0;
-	u8 dfu_state = 0;
 	u32 dfu_timeout = 0;
 	int bsize = 0;
 	int blockno = 0;
+	struct dfu_status *dfu_stat_buf = NULL;
+	u8 *dfu_state = NULL;
+	u8 *block = NULL;
 
 	at76_dbg(DBG_DFU, "%s( %p, %u, %d)", __func__, buf, size,
 		 manifest_sync_timeout);
@@ -383,13 +383,28 @@ static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 		return -EINVAL;
 	}
 
+	dfu_stat_buf = kmalloc(sizeof(struct dfu_status), GFP_KERNEL);
+	if (!dfu_stat_buf) {
+		ret = -ENOMEM;
+		goto exit;
+	}
+
 	block = kmalloc(FW_BLOCK_SIZE, GFP_KERNEL);
-	if (!block)
-		return -ENOMEM;
+	if (!block) {
+		ret = -ENOMEM;
+		goto exit;
+	}
+
+	dfu_state = kmalloc(sizeof(u8), GFP_KERNEL);
+	if (!dfu_state) {
+		ret = -ENOMEM;
+		goto exit;
+	}
+	*dfu_state = 0;
 
 	do {
 		if (need_dfu_state) {
-			ret = at76_dfu_get_state(udev, &dfu_state);
+			ret = at76_dfu_get_state(udev, dfu_state);
 			if (ret < 0) {
 				dev_err(&udev->dev,
 					"cannot get DFU state: %d\n", ret);
@@ -398,13 +413,13 @@ static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 			need_dfu_state = 0;
 		}
 
-		switch (dfu_state) {
+		switch (*dfu_state) {
 		case STATE_DFU_DOWNLOAD_SYNC:
 			at76_dbg(DBG_DFU, "STATE_DFU_DOWNLOAD_SYNC");
-			ret = at76_dfu_get_status(udev, &dfu_stat_buf);
+			ret = at76_dfu_get_status(udev, dfu_stat_buf);
 			if (ret >= 0) {
-				dfu_state = dfu_stat_buf.state;
-				dfu_timeout = at76_get_timeout(&dfu_stat_buf);
+				*dfu_state = dfu_stat_buf->state;
+				dfu_timeout = at76_get_timeout(dfu_stat_buf);
 				need_dfu_state = 0;
 			} else
 				dev_err(&udev->dev,
@@ -447,12 +462,12 @@ static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 		case STATE_DFU_MANIFEST_SYNC:
 			at76_dbg(DBG_DFU, "STATE_DFU_MANIFEST_SYNC");
 
-			ret = at76_dfu_get_status(udev, &dfu_stat_buf);
+			ret = at76_dfu_get_status(udev, dfu_stat_buf);
 			if (ret < 0)
 				break;
 
-			dfu_state = dfu_stat_buf.state;
-			dfu_timeout = at76_get_timeout(&dfu_stat_buf);
+			*dfu_state = dfu_stat_buf->state;
+			dfu_timeout = at76_get_timeout(dfu_stat_buf);
 			need_dfu_state = 0;
 
 			/* override the timeout from the status response,
@@ -484,14 +499,17 @@ static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 			break;
 
 		default:
-			at76_dbg(DBG_DFU, "DFU UNKNOWN STATE (%d)", dfu_state);
+			at76_dbg(DBG_DFU, "DFU UNKNOWN STATE (%d)", *dfu_state);
 			ret = -EINVAL;
 			break;
 		}
 	} while (!is_done && (ret >= 0));
 
 exit:
+	kfree(dfu_state);
 	kfree(block);
+	kfree(dfu_stat_buf);
+
 	if (ret >= 0)
 		ret = 0;
 
@@ -1277,6 +1295,7 @@ static int at76_load_external_fw(struct usb_device *udev, struct fwentry *fwe)
 			dev_err(&udev->dev,
 				"loading %dth firmware block failed: %d\n",
 				blockno, ret);
+			ret = -EIO;
 			goto exit;
 		}
 		buf += bsize;
@@ -1410,6 +1429,8 @@ static int at76_startup_device(struct at76_priv *priv)
 	/* remove BSSID from previous run */
 	memset(priv->bssid, 0, ETH_ALEN);
 
+	priv->scanning = false;
+
 	if (at76_set_radio(priv, 1) == 1)
 		at76_wait_completion(priv, CMD_RADIO_ON);
 
@@ -1483,6 +1504,52 @@ static void at76_work_submit_rx(struct work_struct *work)
 	mutex_unlock(&priv->mtx);
 }
 
+/* This is a workaround to make scan working:
+ * currently mac80211 does not process frames with no frequency
+ * information.
+ * However during scan the HW performs a sweep by itself, and we
+ * are unable to know where the radio is actually tuned.
+ * This function tries to do its best to guess this information..
+ * During scan, If the current frame is a beacon or a probe response,
+ * the channel information is extracted from it.
+ * When not scanning, for other frames, or if it happens that for
+ * whatever reason we fail to parse beacons and probe responses, this
+ * function returns the priv->channel information, that should be correct
+ * at least when we are not scanning.
+ */
+static inline int at76_guess_freq(struct at76_priv *priv)
+{
+	size_t el_off;
+	const u8 *el;
+	int channel = priv->channel;
+	int len = priv->rx_skb->len;
+	struct ieee80211_hdr *hdr = (void *)priv->rx_skb->data;
+
+	if (!priv->scanning)
+		goto exit;
+
+	if (len < 24)
+		goto exit;
+
+	if (ieee80211_is_probe_resp(hdr->frame_control)) {
+		el_off = offsetof(struct ieee80211_mgmt, u.probe_resp.variable);
+		el = ((struct ieee80211_mgmt *)hdr)->u.probe_resp.variable;
+	} else if (ieee80211_is_beacon(hdr->frame_control)) {
+		el_off = offsetof(struct ieee80211_mgmt, u.beacon.variable);
+		el = ((struct ieee80211_mgmt *)hdr)->u.beacon.variable;
+	} else {
+		goto exit;
+	}
+	len -= el_off;
+
+	el = cfg80211_find_ie(WLAN_EID_DS_PARAMS, el, len);
+	if (el && el[1] > 0)
+		channel = el[2];
+
+exit:
+	return ieee80211_channel_to_frequency(channel, IEEE80211_BAND_2GHZ);
+}
+
 static void at76_rx_tasklet(unsigned long param)
 {
 	struct urb *urb = (struct urb *)param;
@@ -1523,6 +1590,8 @@ static void at76_rx_tasklet(unsigned long param)
 	rx_status.signal = buf->rssi;
 	rx_status.flag |= RX_FLAG_DECRYPTED;
 	rx_status.flag |= RX_FLAG_IV_STRIPPED;
+	rx_status.band = IEEE80211_BAND_2GHZ;
+	rx_status.freq = at76_guess_freq(priv);
 
 	at76_dbg(DBG_MAC80211, "calling ieee80211_rx_irqsafe(): %d/%d",
 		 priv->rx_skb->len, priv->rx_skb->data_len);
@@ -1875,6 +1944,8 @@ static void at76_dwork_hw_scan(struct work_struct *work)
 	if (is_valid_ether_addr(priv->bssid))
 		at76_join(priv);
 
+	priv->scanning = false;
+
 	mutex_unlock(&priv->mtx);
 
 	ieee80211_scan_completed(priv->hw, false);
@@ -1929,6 +2000,7 @@ static int at76_hw_scan(struct ieee80211_hw *hw,
 		goto exit;
 	}
 
+	priv->scanning = true;
 	ieee80211_queue_delayed_work(priv->hw, &priv->dwork_hw_scan,
 				     SCAN_POLL_INTERVAL);
 
@@ -2020,6 +2092,44 @@ static void at76_configure_filter(struct ieee80211_hw *hw,
 	ieee80211_queue_work(hw, &priv->work_set_promisc);
 }
 
+static int at76_set_wep(struct at76_priv *priv)
+{
+	int ret = 0;
+	struct mib_mac_wep *mib_data = &priv->mib_buf.data.wep_mib;
+
+	priv->mib_buf.type = MIB_MAC_WEP;
+	priv->mib_buf.size = sizeof(struct mib_mac_wep);
+	priv->mib_buf.index = 0;
+
+	memset(mib_data, 0, sizeof(*mib_data));
+
+	if (priv->wep_enabled) {
+		if (priv->wep_keys_len[priv->wep_key_id] > WEP_SMALL_KEY_LEN)
+			mib_data->encryption_level = 2;
+		else
+			mib_data->encryption_level = 1;
+
+		/* always exclude unencrypted if WEP is active */
+		mib_data->exclude_unencrypted = 1;
+	} else {
+		mib_data->exclude_unencrypted = 0;
+		mib_data->encryption_level = 0;
+	}
+
+	mib_data->privacy_invoked = priv->wep_enabled;
+	mib_data->wep_default_key_id = priv->wep_key_id;
+	memcpy(mib_data->wep_default_keyvalue, priv->wep_keys,
+	       sizeof(priv->wep_keys));
+
+	ret = at76_set_mib(priv, &priv->mib_buf);
+
+	if (ret < 0)
+		wiphy_err(priv->hw->wiphy,
+			  "set_mib (wep) failed: %d\n", ret);
+
+	return ret;
+}
+
 static int at76_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 			struct ieee80211_vif *vif, struct ieee80211_sta *sta,
 			struct ieee80211_key_conf *key)
@@ -2062,7 +2172,7 @@ static int at76_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 			priv->wep_enabled = 1;
 	}
 
-	at76_startup_device(priv);
+	at76_set_wep(priv);
 
 	mutex_unlock(&priv->mtx);
 
@@ -2330,16 +2440,22 @@ static int at76_probe(struct usb_interface *interface,
 	struct usb_device *udev;
 	int op_mode;
 	int need_ext_fw = 0;
-	struct mib_fw_version fwv;
+	struct mib_fw_version *fwv = NULL;
 	int board_type = (int)id->driver_info;
 
 	udev = usb_get_dev(interface_to_usbdev(interface));
 
+	fwv = kmalloc(sizeof(*fwv), GFP_KERNEL);
+	if (!fwv) {
+		ret = -ENOMEM;
+		goto exit;
+	}
+
 	/* Load firmware into kernel memory */
 	fwe = at76_load_firmware(udev, board_type);
 	if (!fwe) {
 		ret = -ENOENT;
-		goto error;
+		goto exit;
 	}
 
 	op_mode = at76_get_op_mode(udev);
@@ -2353,7 +2469,7 @@ static int at76_probe(struct usb_interface *interface,
 		dev_err(&interface->dev,
 			"cannot handle a device in HW_CONFIG_MODE\n");
 		ret = -EBUSY;
-		goto error;
+		goto exit;
 	}
 
 	if (op_mode != OPMODE_NORMAL_NIC_WITH_FLASH
@@ -2366,10 +2482,10 @@ static int at76_probe(struct usb_interface *interface,
 			dev_err(&interface->dev,
 				"error %d downloading internal firmware\n",
 				ret);
-			goto error;
+			goto exit;
 		}
 		usb_put_dev(udev);
-		return ret;
+		goto exit;
 	}
 
 	/* Internal firmware already inside the device.  Get firmware
@@ -2382,8 +2498,8 @@ static int at76_probe(struct usb_interface *interface,
 	 * query the device for the fw version */
 	if ((fwe->fw_version.major > 0 || fwe->fw_version.minor >= 100)
 	    || (op_mode == OPMODE_NORMAL_NIC_WITH_FLASH)) {
-		ret = at76_get_mib(udev, MIB_FW_VERSION, &fwv, sizeof(fwv));
-		if (ret < 0 || (fwv.major | fwv.minor) == 0)
+		ret = at76_get_mib(udev, MIB_FW_VERSION, fwv, sizeof(*fwv));
+		if (ret < 0 || (fwv->major | fwv->minor) == 0)
 			need_ext_fw = 1;
 	} else
 		/* No way to check firmware version, reload to be sure */
@@ -2394,37 +2510,37 @@ static int at76_probe(struct usb_interface *interface,
 			   "downloading external firmware\n");
 
 		ret = at76_load_external_fw(udev, fwe);
-		if (ret)
-			goto error;
+		if (ret < 0)
+			goto exit;
 
 		/* Re-check firmware version */
-		ret = at76_get_mib(udev, MIB_FW_VERSION, &fwv, sizeof(fwv));
+		ret = at76_get_mib(udev, MIB_FW_VERSION, fwv, sizeof(*fwv));
 		if (ret < 0) {
 			dev_err(&interface->dev,
 				"error %d getting firmware version\n", ret);
-			goto error;
+			goto exit;
 		}
 	}
 
 	priv = at76_alloc_new_device(udev);
 	if (!priv) {
 		ret = -ENOMEM;
-		goto error;
+		goto exit;
 	}
 
 	usb_set_intfdata(interface, priv);
 
-	memcpy(&priv->fw_version, &fwv, sizeof(struct mib_fw_version));
+	memcpy(&priv->fw_version, fwv, sizeof(struct mib_fw_version));
 	priv->board_type = board_type;
 
 	ret = at76_init_new_device(priv, interface);
 	if (ret < 0)
 		at76_delete_device(priv);
 
-	return ret;
-
-error:
-	usb_put_dev(udev);
+exit:
+	kfree(fwv);
+	if (ret < 0)
+		usb_put_dev(udev);
 	return ret;
 }
 
diff --git a/drivers/net/wireless/at76c50x-usb.h b/drivers/net/wireless/at76c50x-usb.h
index f14a65473fe8..55090a38ac95 100644
--- a/drivers/net/wireless/at76c50x-usb.h
+++ b/drivers/net/wireless/at76c50x-usb.h
@@ -219,18 +219,6 @@ struct at76_req_join {
 	u8 reserved;
 } __packed;
 
-struct set_mib_buffer {
-	u8 type;
-	u8 size;
-	u8 index;
-	u8 reserved;
-	union {
-		u8 byte;
-		__le16 word;
-		u8 addr[ETH_ALEN];
-	} data;
-} __packed;
-
 struct mib_local {
 	u16 reserved0;
 	u8 beacon_enable;
@@ -334,6 +322,19 @@ struct mib_mdomain {
 	u8 channel_list[14];	/* 0 for invalid channels */
 } __packed;
 
+struct set_mib_buffer {
+	u8 type;
+	u8 size;
+	u8 index;
+	u8 reserved;
+	union {
+		u8 byte;
+		__le16 word;
+		u8 addr[ETH_ALEN];
+		struct mib_mac_wep wep_mib;
+	} data;
+} __packed;
+
 struct at76_fw_header {
 	__le32 crc;		/* CRC32 of the whole image */
 	__le32 board_type;	/* firmware compatibility code */
@@ -417,6 +418,7 @@ struct at76_priv {
 	int scan_max_time;	/* scan max channel time */
 	int scan_mode;		/* SCAN_TYPE_ACTIVE, SCAN_TYPE_PASSIVE */
 	int scan_need_any;	/* if set, need to scan for any ESSID */
+	bool scanning;		/* if set, the scan is running */
 
 	u16 assoc_id;		/* current association ID, if associated */
 
diff --git a/drivers/net/wireless/ath/ar5523/ar5523.c b/drivers/net/wireless/ath/ar5523/ar5523.c
index 507d9a9ee69a..f92050617ae6 100644
--- a/drivers/net/wireless/ath/ar5523/ar5523.c
+++ b/drivers/net/wireless/ath/ar5523/ar5523.c
@@ -1090,7 +1090,8 @@ static int ar5523_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 	return ret;
 }
 
-static void ar5523_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void ar5523_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			 u32 queues, bool drop)
 {
 	struct ar5523 *ar = hw->priv;
 
diff --git a/drivers/net/wireless/ath/ath10k/bmi.c b/drivers/net/wireless/ath/ath10k/bmi.c
index a1f099628850..17d221abd58c 100644
--- a/drivers/net/wireless/ath/ath10k/bmi.c
+++ b/drivers/net/wireless/ath/ath10k/bmi.c
@@ -175,7 +175,7 @@ int ath10k_bmi_write_memory(struct ath10k *ar,
 	return 0;
 }
 
-int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 *param)
+int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 param, u32 *result)
 {
 	struct bmi_cmd cmd;
 	union bmi_resp resp;
@@ -184,7 +184,7 @@ int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 *param)
 	int ret;
 
 	ath10k_dbg(ATH10K_DBG_BMI, "bmi execute address 0x%x param 0x%x\n",
-		   address, *param);
+		   address, param);
 
 	if (ar->bmi.done_sent) {
 		ath10k_warn("command disallowed\n");
@@ -193,7 +193,7 @@ int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 *param)
 
 	cmd.id            = __cpu_to_le32(BMI_EXECUTE);
 	cmd.execute.addr  = __cpu_to_le32(address);
-	cmd.execute.param = __cpu_to_le32(*param);
+	cmd.execute.param = __cpu_to_le32(param);
 
 	ret = ath10k_hif_exchange_bmi_msg(ar, &cmd, cmdlen, &resp, &resplen);
 	if (ret) {
@@ -204,10 +204,13 @@ int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 *param)
 	if (resplen < sizeof(resp.execute)) {
 		ath10k_warn("invalid execute response length (%d)\n",
 			    resplen);
-		return ret;
+		return -EIO;
 	}
 
-	*param = __le32_to_cpu(resp.execute.result);
+	*result = __le32_to_cpu(resp.execute.result);
+
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi execute result 0x%x\n", *result);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ath/ath10k/bmi.h b/drivers/net/wireless/ath/ath10k/bmi.h
index 8d81ce1cec21..111ab701465c 100644
--- a/drivers/net/wireless/ath/ath10k/bmi.h
+++ b/drivers/net/wireless/ath/ath10k/bmi.h
@@ -201,7 +201,8 @@ int ath10k_bmi_write_memory(struct ath10k *ar, u32 address,
 									\
 		addr = host_interest_item_address(HI_ITEM(item));	\
 		ret = ath10k_bmi_read_memory(ar, addr, (u8 *)&tmp, 4); \
-		*val = __le32_to_cpu(tmp);				\
+		if (!ret)						\
+			*val = __le32_to_cpu(tmp);			\
 		ret;							\
 	 })
 
@@ -217,7 +218,7 @@ int ath10k_bmi_write_memory(struct ath10k *ar, u32 address,
 		ret;							\
 	})
 
-int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 *param);
+int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 param, u32 *result);
 int ath10k_bmi_lz_stream_start(struct ath10k *ar, u32 address);
 int ath10k_bmi_lz_data(struct ath10k *ar, const void *buffer, u32 length);
 int ath10k_bmi_fast_download(struct ath10k *ar, u32 address,
diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index a79499c82350..d185dc0cd12b 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -329,6 +329,33 @@ exit:
 	return ret;
 }
 
+void __ath10k_ce_send_revert(struct ath10k_ce_pipe *pipe)
+{
+	struct ath10k *ar = pipe->ar;
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+	struct ath10k_ce_ring *src_ring = pipe->src_ring;
+	u32 ctrl_addr = pipe->ctrl_addr;
+
+	lockdep_assert_held(&ar_pci->ce_lock);
+
+	/*
+	 * This function must be called only if there is an incomplete
+	 * scatter-gather transfer (before index register is updated)
+	 * that needs to be cleaned up.
+	 */
+	if (WARN_ON_ONCE(src_ring->write_index == src_ring->sw_index))
+		return;
+
+	if (WARN_ON_ONCE(src_ring->write_index ==
+			 ath10k_ce_src_ring_write_index_get(ar, ctrl_addr)))
+		return;
+
+	src_ring->write_index--;
+	src_ring->write_index &= src_ring->nentries_mask;
+
+	src_ring->per_transfer_context[src_ring->write_index] = NULL;
+}
+
 int ath10k_ce_send(struct ath10k_ce_pipe *ce_state,
 		   void *per_transfer_context,
 		   u32 buffer,
@@ -840,35 +867,17 @@ void ath10k_ce_recv_cb_register(struct ath10k_ce_pipe *ce_state,
 
 static int ath10k_ce_init_src_ring(struct ath10k *ar,
 				   unsigned int ce_id,
-				   struct ath10k_ce_pipe *ce_state,
 				   const struct ce_attr *attr)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
-	struct ath10k_ce_ring *src_ring;
-	unsigned int nentries = attr->src_nentries;
-	unsigned int ce_nbytes;
-	u32 ctrl_addr = ath10k_ce_base_address(ce_id);
-	dma_addr_t base_addr;
-	char *ptr;
-
-	nentries = roundup_pow_of_two(nentries);
-
-	if (ce_state->src_ring) {
-		WARN_ON(ce_state->src_ring->nentries != nentries);
-		return 0;
-	}
-
-	ce_nbytes = sizeof(struct ath10k_ce_ring) + (nentries * sizeof(void *));
-	ptr = kzalloc(ce_nbytes, GFP_KERNEL);
-	if (ptr == NULL)
-		return -ENOMEM;
+	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
+	struct ath10k_ce_ring *src_ring = ce_state->src_ring;
+	u32 nentries, ctrl_addr = ath10k_ce_base_address(ce_id);
 
-	ce_state->src_ring = (struct ath10k_ce_ring *)ptr;
-	src_ring = ce_state->src_ring;
+	nentries = roundup_pow_of_two(attr->src_nentries);
 
-	ptr += sizeof(struct ath10k_ce_ring);
-	src_ring->nentries = nentries;
-	src_ring->nentries_mask = nentries - 1;
+	memset(src_ring->per_transfer_context, 0,
+	       nentries * sizeof(*src_ring->per_transfer_context));
 
 	src_ring->sw_index = ath10k_ce_src_ring_read_index_get(ar, ctrl_addr);
 	src_ring->sw_index &= src_ring->nentries_mask;
@@ -878,21 +887,87 @@ static int ath10k_ce_init_src_ring(struct ath10k *ar,
 		ath10k_ce_src_ring_write_index_get(ar, ctrl_addr);
 	src_ring->write_index &= src_ring->nentries_mask;
 
-	src_ring->per_transfer_context = (void **)ptr;
+	ath10k_ce_src_ring_base_addr_set(ar, ctrl_addr,
+					 src_ring->base_addr_ce_space);
+	ath10k_ce_src_ring_size_set(ar, ctrl_addr, nentries);
+	ath10k_ce_src_ring_dmax_set(ar, ctrl_addr, attr->src_sz_max);
+	ath10k_ce_src_ring_byte_swap_set(ar, ctrl_addr, 0);
+	ath10k_ce_src_ring_lowmark_set(ar, ctrl_addr, 0);
+	ath10k_ce_src_ring_highmark_set(ar, ctrl_addr, nentries);
+
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot init ce src ring id %d entries %d base_addr %p\n",
+		   ce_id, nentries, src_ring->base_addr_owner_space);
+
+	return 0;
+}
+
+static int ath10k_ce_init_dest_ring(struct ath10k *ar,
+				    unsigned int ce_id,
+				    const struct ce_attr *attr)
+{
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
+	struct ath10k_ce_ring *dest_ring = ce_state->dest_ring;
+	u32 nentries, ctrl_addr = ath10k_ce_base_address(ce_id);
+
+	nentries = roundup_pow_of_two(attr->dest_nentries);
+
+	memset(dest_ring->per_transfer_context, 0,
+	       nentries * sizeof(*dest_ring->per_transfer_context));
+
+	dest_ring->sw_index = ath10k_ce_dest_ring_read_index_get(ar, ctrl_addr);
+	dest_ring->sw_index &= dest_ring->nentries_mask;
+	dest_ring->write_index =
+		ath10k_ce_dest_ring_write_index_get(ar, ctrl_addr);
+	dest_ring->write_index &= dest_ring->nentries_mask;
+
+	ath10k_ce_dest_ring_base_addr_set(ar, ctrl_addr,
+					  dest_ring->base_addr_ce_space);
+	ath10k_ce_dest_ring_size_set(ar, ctrl_addr, nentries);
+	ath10k_ce_dest_ring_byte_swap_set(ar, ctrl_addr, 0);
+	ath10k_ce_dest_ring_lowmark_set(ar, ctrl_addr, 0);
+	ath10k_ce_dest_ring_highmark_set(ar, ctrl_addr, nentries);
+
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot ce dest ring id %d entries %d base_addr %p\n",
+		   ce_id, nentries, dest_ring->base_addr_owner_space);
+
+	return 0;
+}
+
+static struct ath10k_ce_ring *
+ath10k_ce_alloc_src_ring(struct ath10k *ar, unsigned int ce_id,
+			 const struct ce_attr *attr)
+{
+	struct ath10k_ce_ring *src_ring;
+	u32 nentries = attr->src_nentries;
+	dma_addr_t base_addr;
+
+	nentries = roundup_pow_of_two(nentries);
+
+	src_ring = kzalloc(sizeof(*src_ring) +
+			   (nentries *
+			    sizeof(*src_ring->per_transfer_context)),
+			   GFP_KERNEL);
+	if (src_ring == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	src_ring->nentries = nentries;
+	src_ring->nentries_mask = nentries - 1;
 
 	/*
 	 * Legacy platforms that do not support cache
 	 * coherent DMA are unsupported
 	 */
 	src_ring->base_addr_owner_space_unaligned =
-		pci_alloc_consistent(ar_pci->pdev,
-				     (nentries * sizeof(struct ce_desc) +
-				      CE_DESC_RING_ALIGN),
-				     &base_addr);
+		dma_alloc_coherent(ar->dev,
+				   (nentries * sizeof(struct ce_desc) +
+				    CE_DESC_RING_ALIGN),
+				   &base_addr, GFP_KERNEL);
 	if (!src_ring->base_addr_owner_space_unaligned) {
-		kfree(ce_state->src_ring);
-		ce_state->src_ring = NULL;
-		return -ENOMEM;
+		kfree(src_ring);
+		return ERR_PTR(-ENOMEM);
 	}
 
 	src_ring->base_addr_ce_space_unaligned = base_addr;
@@ -912,88 +987,54 @@ static int ath10k_ce_init_src_ring(struct ath10k *ar,
 		kmalloc((nentries * sizeof(struct ce_desc) +
 			 CE_DESC_RING_ALIGN), GFP_KERNEL);
 	if (!src_ring->shadow_base_unaligned) {
-		pci_free_consistent(ar_pci->pdev,
-				    (nentries * sizeof(struct ce_desc) +
-				     CE_DESC_RING_ALIGN),
-				    src_ring->base_addr_owner_space,
-				    src_ring->base_addr_ce_space);
-		kfree(ce_state->src_ring);
-		ce_state->src_ring = NULL;
-		return -ENOMEM;
+		dma_free_coherent(ar->dev,
+				  (nentries * sizeof(struct ce_desc) +
+				   CE_DESC_RING_ALIGN),
+				  src_ring->base_addr_owner_space,
+				  src_ring->base_addr_ce_space);
+		kfree(src_ring);
+		return ERR_PTR(-ENOMEM);
 	}
 
 	src_ring->shadow_base = PTR_ALIGN(
 			src_ring->shadow_base_unaligned,
 			CE_DESC_RING_ALIGN);
 
-	ath10k_ce_src_ring_base_addr_set(ar, ctrl_addr,
-					 src_ring->base_addr_ce_space);
-	ath10k_ce_src_ring_size_set(ar, ctrl_addr, nentries);
-	ath10k_ce_src_ring_dmax_set(ar, ctrl_addr, attr->src_sz_max);
-	ath10k_ce_src_ring_byte_swap_set(ar, ctrl_addr, 0);
-	ath10k_ce_src_ring_lowmark_set(ar, ctrl_addr, 0);
-	ath10k_ce_src_ring_highmark_set(ar, ctrl_addr, nentries);
-
-	ath10k_dbg(ATH10K_DBG_BOOT,
-		   "boot ce src ring id %d entries %d base_addr %p\n",
-		   ce_id, nentries, src_ring->base_addr_owner_space);
-
-	return 0;
+	return src_ring;
 }
 
-static int ath10k_ce_init_dest_ring(struct ath10k *ar,
-				    unsigned int ce_id,
-				    struct ath10k_ce_pipe *ce_state,
-				    const struct ce_attr *attr)
+static struct ath10k_ce_ring *
+ath10k_ce_alloc_dest_ring(struct ath10k *ar, unsigned int ce_id,
+			  const struct ce_attr *attr)
 {
-	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_ce_ring *dest_ring;
-	unsigned int nentries = attr->dest_nentries;
-	unsigned int ce_nbytes;
-	u32 ctrl_addr = ath10k_ce_base_address(ce_id);
+	u32 nentries;
 	dma_addr_t base_addr;
-	char *ptr;
 
-	nentries = roundup_pow_of_two(nentries);
+	nentries = roundup_pow_of_two(attr->dest_nentries);
 
-	if (ce_state->dest_ring) {
-		WARN_ON(ce_state->dest_ring->nentries != nentries);
-		return 0;
-	}
-
-	ce_nbytes = sizeof(struct ath10k_ce_ring) + (nentries * sizeof(void *));
-	ptr = kzalloc(ce_nbytes, GFP_KERNEL);
-	if (ptr == NULL)
-		return -ENOMEM;
+	dest_ring = kzalloc(sizeof(*dest_ring) +
+			    (nentries *
+			     sizeof(*dest_ring->per_transfer_context)),
+			    GFP_KERNEL);
+	if (dest_ring == NULL)
+		return ERR_PTR(-ENOMEM);
 
-	ce_state->dest_ring = (struct ath10k_ce_ring *)ptr;
-	dest_ring = ce_state->dest_ring;
-
-	ptr += sizeof(struct ath10k_ce_ring);
 	dest_ring->nentries = nentries;
 	dest_ring->nentries_mask = nentries - 1;
 
-	dest_ring->sw_index = ath10k_ce_dest_ring_read_index_get(ar, ctrl_addr);
-	dest_ring->sw_index &= dest_ring->nentries_mask;
-	dest_ring->write_index =
-		ath10k_ce_dest_ring_write_index_get(ar, ctrl_addr);
-	dest_ring->write_index &= dest_ring->nentries_mask;
-
-	dest_ring->per_transfer_context = (void **)ptr;
-
 	/*
 	 * Legacy platforms that do not support cache
 	 * coherent DMA are unsupported
 	 */
 	dest_ring->base_addr_owner_space_unaligned =
-		pci_alloc_consistent(ar_pci->pdev,
-				     (nentries * sizeof(struct ce_desc) +
-				      CE_DESC_RING_ALIGN),
-				     &base_addr);
+		dma_alloc_coherent(ar->dev,
+				   (nentries * sizeof(struct ce_desc) +
+				    CE_DESC_RING_ALIGN),
+				   &base_addr, GFP_KERNEL);
 	if (!dest_ring->base_addr_owner_space_unaligned) {
-		kfree(ce_state->dest_ring);
-		ce_state->dest_ring = NULL;
-		return -ENOMEM;
+		kfree(dest_ring);
+		return ERR_PTR(-ENOMEM);
 	}
 
 	dest_ring->base_addr_ce_space_unaligned = base_addr;
@@ -1012,39 +1053,7 @@ static int ath10k_ce_init_dest_ring(struct ath10k *ar,
 			dest_ring->base_addr_ce_space_unaligned,
 			CE_DESC_RING_ALIGN);
 
-	ath10k_ce_dest_ring_base_addr_set(ar, ctrl_addr,
-					  dest_ring->base_addr_ce_space);
-	ath10k_ce_dest_ring_size_set(ar, ctrl_addr, nentries);
-	ath10k_ce_dest_ring_byte_swap_set(ar, ctrl_addr, 0);
-	ath10k_ce_dest_ring_lowmark_set(ar, ctrl_addr, 0);
-	ath10k_ce_dest_ring_highmark_set(ar, ctrl_addr, nentries);
-
-	ath10k_dbg(ATH10K_DBG_BOOT,
-		   "boot ce dest ring id %d entries %d base_addr %p\n",
-		   ce_id, nentries, dest_ring->base_addr_owner_space);
-
-	return 0;
-}
-
-static struct ath10k_ce_pipe *ath10k_ce_init_state(struct ath10k *ar,
-					     unsigned int ce_id,
-					     const struct ce_attr *attr)
-{
-	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
-	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
-	u32 ctrl_addr = ath10k_ce_base_address(ce_id);
-
-	spin_lock_bh(&ar_pci->ce_lock);
-
-	ce_state->ar = ar;
-	ce_state->id = ce_id;
-	ce_state->ctrl_addr = ctrl_addr;
-	ce_state->attr_flags = attr->flags;
-	ce_state->src_sz_max = attr->src_sz_max;
-
-	spin_unlock_bh(&ar_pci->ce_lock);
-
-	return ce_state;
+	return dest_ring;
 }
 
 /*
@@ -1054,11 +1063,11 @@ static struct ath10k_ce_pipe *ath10k_ce_init_state(struct ath10k *ar,
  * initialization. It may be that only one side or the other is
  * initialized by software/firmware.
  */
-struct ath10k_ce_pipe *ath10k_ce_init(struct ath10k *ar,
-				unsigned int ce_id,
-				const struct ce_attr *attr)
+int ath10k_ce_init_pipe(struct ath10k *ar, unsigned int ce_id,
+			const struct ce_attr *attr)
 {
-	struct ath10k_ce_pipe *ce_state;
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
 	int ret;
 
 	/*
@@ -1074,64 +1083,128 @@ struct ath10k_ce_pipe *ath10k_ce_init(struct ath10k *ar,
 
 	ret = ath10k_pci_wake(ar);
 	if (ret)
-		return NULL;
+		return ret;
 
-	ce_state = ath10k_ce_init_state(ar, ce_id, attr);
-	if (!ce_state) {
-		ath10k_err("Failed to initialize CE state for ID: %d\n", ce_id);
-		goto out;
-	}
+	spin_lock_bh(&ar_pci->ce_lock);
+	ce_state->ar = ar;
+	ce_state->id = ce_id;
+	ce_state->ctrl_addr = ath10k_ce_base_address(ce_id);
+	ce_state->attr_flags = attr->flags;
+	ce_state->src_sz_max = attr->src_sz_max;
+	spin_unlock_bh(&ar_pci->ce_lock);
 
 	if (attr->src_nentries) {
-		ret = ath10k_ce_init_src_ring(ar, ce_id, ce_state, attr);
+		ret = ath10k_ce_init_src_ring(ar, ce_id, attr);
 		if (ret) {
 			ath10k_err("Failed to initialize CE src ring for ID: %d (%d)\n",
 				   ce_id, ret);
-			ath10k_ce_deinit(ce_state);
-			ce_state = NULL;
 			goto out;
 		}
 	}
 
 	if (attr->dest_nentries) {
-		ret = ath10k_ce_init_dest_ring(ar, ce_id, ce_state, attr);
+		ret = ath10k_ce_init_dest_ring(ar, ce_id, attr);
 		if (ret) {
 			ath10k_err("Failed to initialize CE dest ring for ID: %d (%d)\n",
 				   ce_id, ret);
-			ath10k_ce_deinit(ce_state);
-			ce_state = NULL;
 			goto out;
 		}
 	}
 
 out:
 	ath10k_pci_sleep(ar);
-	return ce_state;
+	return ret;
 }
 
-void ath10k_ce_deinit(struct ath10k_ce_pipe *ce_state)
+static void ath10k_ce_deinit_src_ring(struct ath10k *ar, unsigned int ce_id)
+{
+	u32 ctrl_addr = ath10k_ce_base_address(ce_id);
+
+	ath10k_ce_src_ring_base_addr_set(ar, ctrl_addr, 0);
+	ath10k_ce_src_ring_size_set(ar, ctrl_addr, 0);
+	ath10k_ce_src_ring_dmax_set(ar, ctrl_addr, 0);
+	ath10k_ce_src_ring_highmark_set(ar, ctrl_addr, 0);
+}
+
+static void ath10k_ce_deinit_dest_ring(struct ath10k *ar, unsigned int ce_id)
+{
+	u32 ctrl_addr = ath10k_ce_base_address(ce_id);
+
+	ath10k_ce_dest_ring_base_addr_set(ar, ctrl_addr, 0);
+	ath10k_ce_dest_ring_size_set(ar, ctrl_addr, 0);
+	ath10k_ce_dest_ring_highmark_set(ar, ctrl_addr, 0);
+}
+
+void ath10k_ce_deinit_pipe(struct ath10k *ar, unsigned int ce_id)
+{
+	int ret;
+
+	ret = ath10k_pci_wake(ar);
+	if (ret)
+		return;
+
+	ath10k_ce_deinit_src_ring(ar, ce_id);
+	ath10k_ce_deinit_dest_ring(ar, ce_id);
+
+	ath10k_pci_sleep(ar);
+}
+
+int ath10k_ce_alloc_pipe(struct ath10k *ar, int ce_id,
+			 const struct ce_attr *attr)
+{
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
+	int ret;
+
+	if (attr->src_nentries) {
+		ce_state->src_ring = ath10k_ce_alloc_src_ring(ar, ce_id, attr);
+		if (IS_ERR(ce_state->src_ring)) {
+			ret = PTR_ERR(ce_state->src_ring);
+			ath10k_err("failed to allocate copy engine source ring %d: %d\n",
+				   ce_id, ret);
+			ce_state->src_ring = NULL;
+			return ret;
+		}
+	}
+
+	if (attr->dest_nentries) {
+		ce_state->dest_ring = ath10k_ce_alloc_dest_ring(ar, ce_id,
+								attr);
+		if (IS_ERR(ce_state->dest_ring)) {
+			ret = PTR_ERR(ce_state->dest_ring);
+			ath10k_err("failed to allocate copy engine destination ring %d: %d\n",
+				   ce_id, ret);
+			ce_state->dest_ring = NULL;
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+void ath10k_ce_free_pipe(struct ath10k *ar, int ce_id)
 {
-	struct ath10k *ar = ce_state->ar;
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
 
 	if (ce_state->src_ring) {
 		kfree(ce_state->src_ring->shadow_base_unaligned);
-		pci_free_consistent(ar_pci->pdev,
-				    (ce_state->src_ring->nentries *
-				     sizeof(struct ce_desc) +
-				     CE_DESC_RING_ALIGN),
-				    ce_state->src_ring->base_addr_owner_space,
-				    ce_state->src_ring->base_addr_ce_space);
+		dma_free_coherent(ar->dev,
+				  (ce_state->src_ring->nentries *
+				   sizeof(struct ce_desc) +
+				   CE_DESC_RING_ALIGN),
+				  ce_state->src_ring->base_addr_owner_space,
+				  ce_state->src_ring->base_addr_ce_space);
 		kfree(ce_state->src_ring);
 	}
 
 	if (ce_state->dest_ring) {
-		pci_free_consistent(ar_pci->pdev,
-				    (ce_state->dest_ring->nentries *
-				     sizeof(struct ce_desc) +
-				     CE_DESC_RING_ALIGN),
-				    ce_state->dest_ring->base_addr_owner_space,
-				    ce_state->dest_ring->base_addr_ce_space);
+		dma_free_coherent(ar->dev,
+				  (ce_state->dest_ring->nentries *
+				   sizeof(struct ce_desc) +
+				   CE_DESC_RING_ALIGN),
+				  ce_state->dest_ring->base_addr_owner_space,
+				  ce_state->dest_ring->base_addr_ce_space);
 		kfree(ce_state->dest_ring);
 	}
 
diff --git a/drivers/net/wireless/ath/ath10k/ce.h b/drivers/net/wireless/ath/ath10k/ce.h
index 8eb7f99ed992..7a5a36fc59c1 100644
--- a/drivers/net/wireless/ath/ath10k/ce.h
+++ b/drivers/net/wireless/ath/ath10k/ce.h
@@ -104,7 +104,8 @@ struct ath10k_ce_ring {
 	void *shadow_base_unaligned;
 	struct ce_desc *shadow_base;
 
-	void **per_transfer_context;
+	/* keep last */
+	void *per_transfer_context[0];
 };
 
 struct ath10k_ce_pipe {
@@ -159,6 +160,8 @@ int ath10k_ce_send_nolock(struct ath10k_ce_pipe *ce_state,
 			  unsigned int transfer_id,
 			  unsigned int flags);
 
+void __ath10k_ce_send_revert(struct ath10k_ce_pipe *pipe);
+
 void ath10k_ce_send_cb_register(struct ath10k_ce_pipe *ce_state,
 				void (*send_cb)(struct ath10k_ce_pipe *),
 				int disable_interrupts);
@@ -210,10 +213,12 @@ int ath10k_ce_completed_send_next(struct ath10k_ce_pipe *ce_state,
 
 /*==================CE Engine Initialization=======================*/
 
-/* Initialize an instance of a CE */
-struct ath10k_ce_pipe *ath10k_ce_init(struct ath10k *ar,
-				unsigned int ce_id,
-				const struct ce_attr *attr);
+int ath10k_ce_init_pipe(struct ath10k *ar, unsigned int ce_id,
+			const struct ce_attr *attr);
+void ath10k_ce_deinit_pipe(struct ath10k *ar, unsigned int ce_id);
+int ath10k_ce_alloc_pipe(struct ath10k *ar, int ce_id,
+			  const struct ce_attr *attr);
+void ath10k_ce_free_pipe(struct ath10k *ar, int ce_id);
 
 /*==================CE Engine Shutdown=======================*/
 /*
@@ -236,8 +241,6 @@ int ath10k_ce_cancel_send_next(struct ath10k_ce_pipe *ce_state,
 			       unsigned int *nbytesp,
 			       unsigned int *transfer_idp);
 
-void ath10k_ce_deinit(struct ath10k_ce_pipe *ce_state);
-
 /*==================CE Interrupt Handlers====================*/
 void ath10k_ce_per_engine_service_any(struct ath10k *ar);
 void ath10k_ce_per_engine_service(struct ath10k *ar, unsigned int ce_id);
diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index ebc5fc2ede75..82017f56e661 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -58,36 +58,6 @@ static void ath10k_send_suspend_complete(struct ath10k *ar)
 	complete(&ar->target_suspend);
 }
 
-static int ath10k_init_connect_htc(struct ath10k *ar)
-{
-	int status;
-
-	status = ath10k_wmi_connect_htc_service(ar);
-	if (status)
-		goto conn_fail;
-
-	/* Start HTC */
-	status = ath10k_htc_start(&ar->htc);
-	if (status)
-		goto conn_fail;
-
-	/* Wait for WMI event to be ready */
-	status = ath10k_wmi_wait_for_service_ready(ar);
-	if (status <= 0) {
-		ath10k_warn("wmi service ready event not received");
-		status = -ETIMEDOUT;
-		goto timeout;
-	}
-
-	ath10k_dbg(ATH10K_DBG_BOOT, "boot wmi ready\n");
-	return 0;
-
-timeout:
-	ath10k_htc_stop(&ar->htc);
-conn_fail:
-	return status;
-}
-
 static int ath10k_init_configure_target(struct ath10k *ar)
 {
 	u32 param_host;
@@ -249,30 +219,40 @@ exit:
 
 static int ath10k_download_and_run_otp(struct ath10k *ar)
 {
-	u32 address = ar->hw_params.patch_load_addr;
-	u32 exec_param;
+	u32 result, address = ar->hw_params.patch_load_addr;
 	int ret;
 
 	/* OTP is optional */
 
-	if (!ar->otp_data || !ar->otp_len)
+	if (!ar->otp_data || !ar->otp_len) {
+		ath10k_warn("Not running otp, calibration will be incorrect (otp-data %p otp_len %zd)!\n",
+			    ar->otp_data, ar->otp_len);
 		return 0;
+	}
+
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot upload otp to 0x%x len %zd\n",
+		   address, ar->otp_len);
 
 	ret = ath10k_bmi_fast_download(ar, address, ar->otp_data, ar->otp_len);
 	if (ret) {
 		ath10k_err("could not write otp (%d)\n", ret);
-		goto exit;
+		return ret;
 	}
 
-	exec_param = 0;
-	ret = ath10k_bmi_execute(ar, address, &exec_param);
+	ret = ath10k_bmi_execute(ar, address, 0, &result);
 	if (ret) {
 		ath10k_err("could not execute otp (%d)\n", ret);
-		goto exit;
+		return ret;
 	}
 
-exit:
-	return ret;
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot otp execute result %d\n", result);
+
+	if (result != 0) {
+		ath10k_err("otp calibration failed: %d", result);
+		return -EINVAL;
+	}
+
+	return 0;
 }
 
 static int ath10k_download_fw(struct ath10k *ar)
@@ -389,8 +369,8 @@ static int ath10k_core_fetch_firmware_api_n(struct ath10k *ar, const char *name)
 	/* first fetch the firmware file (firmware-*.bin) */
 	ar->firmware = ath10k_fetch_fw_file(ar, ar->hw_params.fw.dir, name);
 	if (IS_ERR(ar->firmware)) {
-		ath10k_err("Could not fetch firmware file '%s': %ld\n",
-			   name, PTR_ERR(ar->firmware));
+		ath10k_err("could not fetch firmware file '%s/%s': %ld\n",
+			   ar->hw_params.fw.dir, name, PTR_ERR(ar->firmware));
 		return PTR_ERR(ar->firmware);
 	}
 
@@ -401,14 +381,14 @@ static int ath10k_core_fetch_firmware_api_n(struct ath10k *ar, const char *name)
 	magic_len = strlen(ATH10K_FIRMWARE_MAGIC) + 1;
 
 	if (len < magic_len) {
-		ath10k_err("firmware image too small to contain magic: %zu\n",
-			   len);
+		ath10k_err("firmware file '%s/%s' too small to contain magic: %zu\n",
+			   ar->hw_params.fw.dir, name, len);
 		ret = -EINVAL;
 		goto err;
 	}
 
 	if (memcmp(data, ATH10K_FIRMWARE_MAGIC, magic_len) != 0) {
-		ath10k_err("Invalid firmware magic\n");
+		ath10k_err("invalid firmware magic\n");
 		ret = -EINVAL;
 		goto err;
 	}
@@ -430,7 +410,7 @@ static int ath10k_core_fetch_firmware_api_n(struct ath10k *ar, const char *name)
 		data += sizeof(*hdr);
 
 		if (len < ie_len) {
-			ath10k_err("Invalid length for FW IE %d (%zu < %zu)\n",
+			ath10k_err("invalid length for FW IE %d (%zu < %zu)\n",
 				   ie_id, len, ie_len);
 			ret = -EINVAL;
 			goto err;
@@ -513,8 +493,8 @@ static int ath10k_core_fetch_firmware_api_n(struct ath10k *ar, const char *name)
 	}
 
 	if (!ar->firmware_data || !ar->firmware_len) {
-		ath10k_warn("No ATH10K_FW_IE_FW_IMAGE found from %s, skipping\n",
-			    name);
+		ath10k_warn("No ATH10K_FW_IE_FW_IMAGE found from '%s/%s', skipping\n",
+			    ar->hw_params.fw.dir, name);
 		ret = -ENOMEDIUM;
 		goto err;
 	}
@@ -531,7 +511,9 @@ static int ath10k_core_fetch_firmware_api_n(struct ath10k *ar, const char *name)
 					 ar->hw_params.fw.board);
 	if (IS_ERR(ar->board)) {
 		ret = PTR_ERR(ar->board);
-		ath10k_err("could not fetch board data (%d)\n", ret);
+		ath10k_err("could not fetch board data '%s/%s' (%d)\n",
+			   ar->hw_params.fw.dir, ar->hw_params.fw.board,
+			   ret);
 		goto err;
 	}
 
@@ -549,19 +531,21 @@ static int ath10k_core_fetch_firmware_files(struct ath10k *ar)
 {
 	int ret;
 
+	ar->fw_api = 2;
+	ath10k_dbg(ATH10K_DBG_BOOT, "trying fw api %d\n", ar->fw_api);
+
 	ret = ath10k_core_fetch_firmware_api_n(ar, ATH10K_FW_API2_FILE);
-	if (ret == 0) {
-		ar->fw_api = 2;
-		goto out;
-	}
+	if (ret == 0)
+		goto success;
+
+	ar->fw_api = 1;
+	ath10k_dbg(ATH10K_DBG_BOOT, "trying fw api %d\n", ar->fw_api);
 
 	ret = ath10k_core_fetch_firmware_api_1(ar);
 	if (ret)
 		return ret;
 
-	ar->fw_api = 1;
-
-out:
+success:
 	ath10k_dbg(ATH10K_DBG_BOOT, "using fw api %d\n", ar->fw_api);
 
 	return 0;
@@ -572,16 +556,22 @@ static int ath10k_init_download_firmware(struct ath10k *ar)
 	int ret;
 
 	ret = ath10k_download_board_data(ar);
-	if (ret)
+	if (ret) {
+		ath10k_err("failed to download board data: %d\n", ret);
 		return ret;
+	}
 
 	ret = ath10k_download_and_run_otp(ar);
-	if (ret)
+	if (ret) {
+		ath10k_err("failed to run otp: %d\n", ret);
 		return ret;
+	}
 
 	ret = ath10k_download_fw(ar);
-	if (ret)
+	if (ret) {
+		ath10k_err("failed to download firmware: %d\n", ret);
 		return ret;
+	}
 
 	return ret;
 }
@@ -660,8 +650,9 @@ static void ath10k_core_restart(struct work_struct *work)
 
 	switch (ar->state) {
 	case ATH10K_STATE_ON:
-		ath10k_halt(ar);
 		ar->state = ATH10K_STATE_RESTARTING;
+		del_timer_sync(&ar->scan.timeout);
+		ath10k_reset_scan((unsigned long)ar);
 		ieee80211_restart_hw(ar->hw);
 		break;
 	case ATH10K_STATE_OFF:
@@ -670,6 +661,8 @@ static void ath10k_core_restart(struct work_struct *work)
 		ath10k_warn("cannot restart a device that hasn't been started\n");
 		break;
 	case ATH10K_STATE_RESTARTING:
+		/* hw restart might be requested from multiple places */
+		break;
 	case ATH10K_STATE_RESTARTED:
 		ar->state = ATH10K_STATE_WEDGED;
 		/* fall through */
@@ -681,70 +674,6 @@ static void ath10k_core_restart(struct work_struct *work)
 	mutex_unlock(&ar->conf_mutex);
 }
 
-struct ath10k *ath10k_core_create(void *hif_priv, struct device *dev,
-				  const struct ath10k_hif_ops *hif_ops)
-{
-	struct ath10k *ar;
-
-	ar = ath10k_mac_create();
-	if (!ar)
-		return NULL;
-
-	ar->ath_common.priv = ar;
-	ar->ath_common.hw = ar->hw;
-
-	ar->p2p = !!ath10k_p2p;
-	ar->dev = dev;
-
-	ar->hif.priv = hif_priv;
-	ar->hif.ops = hif_ops;
-
-	init_completion(&ar->scan.started);
-	init_completion(&ar->scan.completed);
-	init_completion(&ar->scan.on_channel);
-	init_completion(&ar->target_suspend);
-
-	init_completion(&ar->install_key_done);
-	init_completion(&ar->vdev_setup_done);
-
-	setup_timer(&ar->scan.timeout, ath10k_reset_scan, (unsigned long)ar);
-
-	ar->workqueue = create_singlethread_workqueue("ath10k_wq");
-	if (!ar->workqueue)
-		goto err_wq;
-
-	mutex_init(&ar->conf_mutex);
-	spin_lock_init(&ar->data_lock);
-
-	INIT_LIST_HEAD(&ar->peers);
-	init_waitqueue_head(&ar->peer_mapping_wq);
-
-	init_completion(&ar->offchan_tx_completed);
-	INIT_WORK(&ar->offchan_tx_work, ath10k_offchan_tx_work);
-	skb_queue_head_init(&ar->offchan_tx_queue);
-
-	INIT_WORK(&ar->wmi_mgmt_tx_work, ath10k_mgmt_over_wmi_tx_work);
-	skb_queue_head_init(&ar->wmi_mgmt_tx_queue);
-
-	INIT_WORK(&ar->restart_work, ath10k_core_restart);
-
-	return ar;
-
-err_wq:
-	ath10k_mac_destroy(ar);
-	return NULL;
-}
-EXPORT_SYMBOL(ath10k_core_create);
-
-void ath10k_core_destroy(struct ath10k *ar)
-{
-	flush_workqueue(ar->workqueue);
-	destroy_workqueue(ar->workqueue);
-
-	ath10k_mac_destroy(ar);
-}
-EXPORT_SYMBOL(ath10k_core_destroy);
-
 int ath10k_core_start(struct ath10k *ar)
 {
 	int status;
@@ -785,10 +714,28 @@ int ath10k_core_start(struct ath10k *ar)
 		goto err;
 	}
 
+	status = ath10k_htt_init(ar);
+	if (status) {
+		ath10k_err("failed to init htt: %d\n", status);
+		goto err_wmi_detach;
+	}
+
+	status = ath10k_htt_tx_alloc(&ar->htt);
+	if (status) {
+		ath10k_err("failed to alloc htt tx: %d\n", status);
+		goto err_wmi_detach;
+	}
+
+	status = ath10k_htt_rx_alloc(&ar->htt);
+	if (status) {
+		ath10k_err("failed to alloc htt rx: %d\n", status);
+		goto err_htt_tx_detach;
+	}
+
 	status = ath10k_hif_start(ar);
 	if (status) {
 		ath10k_err("could not start HIF: %d\n", status);
-		goto err_wmi_detach;
+		goto err_htt_rx_detach;
 	}
 
 	status = ath10k_htc_wait_target(&ar->htc);
@@ -797,15 +744,30 @@ int ath10k_core_start(struct ath10k *ar)
 		goto err_hif_stop;
 	}
 
-	status = ath10k_htt_attach(ar);
+	status = ath10k_htt_connect(&ar->htt);
 	if (status) {
-		ath10k_err("could not attach htt (%d)\n", status);
+		ath10k_err("failed to connect htt (%d)\n", status);
 		goto err_hif_stop;
 	}
 
-	status = ath10k_init_connect_htc(ar);
-	if (status)
-		goto err_htt_detach;
+	status = ath10k_wmi_connect(ar);
+	if (status) {
+		ath10k_err("could not connect wmi: %d\n", status);
+		goto err_hif_stop;
+	}
+
+	status = ath10k_htc_start(&ar->htc);
+	if (status) {
+		ath10k_err("failed to start htc: %d\n", status);
+		goto err_hif_stop;
+	}
+
+	status = ath10k_wmi_wait_for_service_ready(ar);
+	if (status <= 0) {
+		ath10k_warn("wmi service ready event not received");
+		status = -ETIMEDOUT;
+		goto err_htc_stop;
+	}
 
 	ath10k_dbg(ATH10K_DBG_BOOT, "firmware %s booted\n",
 		   ar->hw->wiphy->fw_version);
@@ -813,31 +775,36 @@ int ath10k_core_start(struct ath10k *ar)
 	status = ath10k_wmi_cmd_init(ar);
 	if (status) {
 		ath10k_err("could not send WMI init command (%d)\n", status);
-		goto err_disconnect_htc;
+		goto err_htc_stop;
 	}
 
 	status = ath10k_wmi_wait_for_unified_ready(ar);
 	if (status <= 0) {
 		ath10k_err("wmi unified ready event not received\n");
 		status = -ETIMEDOUT;
-		goto err_disconnect_htc;
+		goto err_htc_stop;
 	}
 
-	status = ath10k_htt_attach_target(&ar->htt);
-	if (status)
-		goto err_disconnect_htc;
+	status = ath10k_htt_setup(&ar->htt);
+	if (status) {
+		ath10k_err("failed to setup htt: %d\n", status);
+		goto err_htc_stop;
+	}
 
 	status = ath10k_debug_start(ar);
 	if (status)
-		goto err_disconnect_htc;
+		goto err_htc_stop;
 
 	ar->free_vdev_map = (1 << TARGET_NUM_VDEVS) - 1;
 	INIT_LIST_HEAD(&ar->arvifs);
 
 	if (!test_bit(ATH10K_FLAG_FIRST_BOOT_DONE, &ar->dev_flags))
-		ath10k_info("%s (0x%x) fw %s api %d htt %d.%d\n",
-			    ar->hw_params.name, ar->target_version,
-			    ar->hw->wiphy->fw_version, ar->fw_api,
+		ath10k_info("%s (0x%08x, 0x%08x) fw %s api %d htt %d.%d\n",
+			    ar->hw_params.name,
+			    ar->target_version,
+			    ar->chip_id,
+			    ar->hw->wiphy->fw_version,
+			    ar->fw_api,
 			    ar->htt.target_version_major,
 			    ar->htt.target_version_minor);
 
@@ -845,12 +812,14 @@ int ath10k_core_start(struct ath10k *ar)
 
 	return 0;
 
-err_disconnect_htc:
+err_htc_stop:
 	ath10k_htc_stop(&ar->htc);
-err_htt_detach:
-	ath10k_htt_detach(&ar->htt);
 err_hif_stop:
 	ath10k_hif_stop(ar);
+err_htt_rx_detach:
+	ath10k_htt_rx_free(&ar->htt);
+err_htt_tx_detach:
+	ath10k_htt_tx_free(&ar->htt);
 err_wmi_detach:
 	ath10k_wmi_detach(ar);
 err:
@@ -885,10 +854,14 @@ void ath10k_core_stop(struct ath10k *ar)
 	lockdep_assert_held(&ar->conf_mutex);
 
 	/* try to suspend target */
-	ath10k_wait_for_suspend(ar, WMI_PDEV_SUSPEND_AND_DISABLE_INTR);
+	if (ar->state != ATH10K_STATE_RESTARTING)
+		ath10k_wait_for_suspend(ar, WMI_PDEV_SUSPEND_AND_DISABLE_INTR);
+
 	ath10k_debug_stop(ar);
 	ath10k_htc_stop(&ar->htc);
-	ath10k_htt_detach(&ar->htt);
+	ath10k_hif_stop(ar);
+	ath10k_htt_tx_free(&ar->htt);
+	ath10k_htt_rx_free(&ar->htt);
 	ath10k_wmi_detach(ar);
 }
 EXPORT_SYMBOL(ath10k_core_stop);
@@ -980,22 +953,15 @@ static int ath10k_core_check_chip_id(struct ath10k *ar)
 	return 0;
 }
 
-int ath10k_core_register(struct ath10k *ar, u32 chip_id)
+static void ath10k_core_register_work(struct work_struct *work)
 {
+	struct ath10k *ar = container_of(work, struct ath10k, register_work);
 	int status;
 
-	ar->chip_id = chip_id;
-
-	status = ath10k_core_check_chip_id(ar);
-	if (status) {
-		ath10k_err("Unsupported chip id 0x%08x\n", ar->chip_id);
-		return status;
-	}
-
 	status = ath10k_core_probe_fw(ar);
 	if (status) {
 		ath10k_err("could not probe fw (%d)\n", status);
-		return status;
+		goto err;
 	}
 
 	status = ath10k_mac_register(ar);
@@ -1010,18 +976,43 @@ int ath10k_core_register(struct ath10k *ar, u32 chip_id)
 		goto err_unregister_mac;
 	}
 
-	return 0;
+	set_bit(ATH10K_FLAG_CORE_REGISTERED, &ar->dev_flags);
+	return;
 
 err_unregister_mac:
 	ath10k_mac_unregister(ar);
 err_release_fw:
 	ath10k_core_free_firmware_files(ar);
-	return status;
+err:
+	device_release_driver(ar->dev);
+	return;
+}
+
+int ath10k_core_register(struct ath10k *ar, u32 chip_id)
+{
+	int status;
+
+	ar->chip_id = chip_id;
+
+	status = ath10k_core_check_chip_id(ar);
+	if (status) {
+		ath10k_err("Unsupported chip id 0x%08x\n", ar->chip_id);
+		return status;
+	}
+
+	queue_work(ar->workqueue, &ar->register_work);
+
+	return 0;
 }
 EXPORT_SYMBOL(ath10k_core_register);
 
 void ath10k_core_unregister(struct ath10k *ar)
 {
+	cancel_work_sync(&ar->register_work);
+
+	if (!test_bit(ATH10K_FLAG_CORE_REGISTERED, &ar->dev_flags))
+		return;
+
 	/* We must unregister from mac80211 before we stop HTC and HIF.
 	 * Otherwise we will fail to submit commands to FW and mac80211 will be
 	 * unhappy about callback failures. */
@@ -1033,6 +1024,71 @@ void ath10k_core_unregister(struct ath10k *ar)
 }
 EXPORT_SYMBOL(ath10k_core_unregister);
 
+struct ath10k *ath10k_core_create(void *hif_priv, struct device *dev,
+				  const struct ath10k_hif_ops *hif_ops)
+{
+	struct ath10k *ar;
+
+	ar = ath10k_mac_create();
+	if (!ar)
+		return NULL;
+
+	ar->ath_common.priv = ar;
+	ar->ath_common.hw = ar->hw;
+
+	ar->p2p = !!ath10k_p2p;
+	ar->dev = dev;
+
+	ar->hif.priv = hif_priv;
+	ar->hif.ops = hif_ops;
+
+	init_completion(&ar->scan.started);
+	init_completion(&ar->scan.completed);
+	init_completion(&ar->scan.on_channel);
+	init_completion(&ar->target_suspend);
+
+	init_completion(&ar->install_key_done);
+	init_completion(&ar->vdev_setup_done);
+
+	setup_timer(&ar->scan.timeout, ath10k_reset_scan, (unsigned long)ar);
+
+	ar->workqueue = create_singlethread_workqueue("ath10k_wq");
+	if (!ar->workqueue)
+		goto err_wq;
+
+	mutex_init(&ar->conf_mutex);
+	spin_lock_init(&ar->data_lock);
+
+	INIT_LIST_HEAD(&ar->peers);
+	init_waitqueue_head(&ar->peer_mapping_wq);
+
+	init_completion(&ar->offchan_tx_completed);
+	INIT_WORK(&ar->offchan_tx_work, ath10k_offchan_tx_work);
+	skb_queue_head_init(&ar->offchan_tx_queue);
+
+	INIT_WORK(&ar->wmi_mgmt_tx_work, ath10k_mgmt_over_wmi_tx_work);
+	skb_queue_head_init(&ar->wmi_mgmt_tx_queue);
+
+	INIT_WORK(&ar->register_work, ath10k_core_register_work);
+	INIT_WORK(&ar->restart_work, ath10k_core_restart);
+
+	return ar;
+
+err_wq:
+	ath10k_mac_destroy(ar);
+	return NULL;
+}
+EXPORT_SYMBOL(ath10k_core_create);
+
+void ath10k_core_destroy(struct ath10k *ar)
+{
+	flush_workqueue(ar->workqueue);
+	destroy_workqueue(ar->workqueue);
+
+	ath10k_mac_destroy(ar);
+}
+EXPORT_SYMBOL(ath10k_core_destroy);
+
 MODULE_AUTHOR("Qualcomm Atheros");
 MODULE_DESCRIPTION("Core module for QCA988X PCIe devices.");
 MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 0e71979d837c..68ceef61933d 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -119,6 +119,7 @@ struct ath10k_peer_stat {
 	u8 peer_macaddr[ETH_ALEN];
 	u32 peer_rssi;
 	u32 peer_tx_rate;
+	u32 peer_rx_rate; /* 10x only */
 };
 
 struct ath10k_target_stats {
@@ -130,6 +131,12 @@ struct ath10k_target_stats {
 	u32 cycle_count;
 	u32 phy_err_count;
 	u32 chan_tx_power;
+	u32 ack_rx_bad;
+	u32 rts_bad;
+	u32 rts_good;
+	u32 fcs_bad;
+	u32 no_beacons;
+	u32 mib_int_count;
 
 	/* PDEV TX stats */
 	s32 comp_queued;
@@ -260,6 +267,8 @@ struct ath10k_vif {
 	u8 fixed_rate;
 	u8 fixed_nss;
 	u8 force_sgi;
+	bool use_cts_prot;
+	int num_legacy_stations;
 };
 
 struct ath10k_vif_iter {
@@ -326,6 +335,7 @@ enum ath10k_dev_flags {
 	/* Indicates that ath10k device is during CAC phase of DFS */
 	ATH10K_CAC_RUNNING,
 	ATH10K_FLAG_FIRST_BOOT_DONE,
+	ATH10K_FLAG_CORE_REGISTERED,
 };
 
 struct ath10k {
@@ -419,13 +429,24 @@ struct ath10k {
 	struct cfg80211_chan_def chandef;
 
 	int free_vdev_map;
+	bool promisc;
+	bool monitor;
 	int monitor_vdev_id;
-	bool monitor_enabled;
-	bool monitor_present;
+	bool monitor_started;
 	unsigned int filter_flags;
 	unsigned long dev_flags;
 	u32 dfs_block_radar_events;
 
+	/* protected by conf_mutex */
+	bool radar_enabled;
+	int num_started_vdevs;
+
+	/* Protected by conf-mutex */
+	u8 supp_tx_chainmask;
+	u8 supp_rx_chainmask;
+	u8 cfg_tx_chainmask;
+	u8 cfg_rx_chainmask;
+
 	struct wmi_pdev_set_wmm_params_arg wmm_params;
 	struct completion install_key_done;
 
@@ -456,6 +477,7 @@ struct ath10k {
 
 	enum ath10k_state state;
 
+	struct work_struct register_work;
 	struct work_struct restart_work;
 
 	/* cycle count is reported twice for each visited channel during scan.
diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 6debd281350a..1b7ff4ba122c 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -161,7 +161,7 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 	u8 *tmp = ev->data;
 	struct ath10k_target_stats *stats;
 	int num_pdev_stats, num_vdev_stats, num_peer_stats;
-	struct wmi_pdev_stats *ps;
+	struct wmi_pdev_stats_10x *ps;
 	int i;
 
 	spin_lock_bh(&ar->data_lock);
@@ -173,7 +173,7 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 	num_peer_stats = __le32_to_cpu(ev->num_peer_stats); /* 0 or max peers */
 
 	if (num_pdev_stats) {
-		ps = (struct wmi_pdev_stats *)tmp;
+		ps = (struct wmi_pdev_stats_10x *)tmp;
 
 		stats->ch_noise_floor = __le32_to_cpu(ps->chan_nf);
 		stats->tx_frame_count = __le32_to_cpu(ps->tx_frame_count);
@@ -228,7 +228,18 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 		stats->phy_err_drop = __le32_to_cpu(ps->wal.rx.phy_err_drop);
 		stats->mpdu_errs = __le32_to_cpu(ps->wal.rx.mpdu_errs);
 
-		tmp += sizeof(struct wmi_pdev_stats);
+		if (test_bit(ATH10K_FW_FEATURE_WMI_10X,
+			     ar->fw_features)) {
+			stats->ack_rx_bad = __le32_to_cpu(ps->ack_rx_bad);
+			stats->rts_bad = __le32_to_cpu(ps->rts_bad);
+			stats->rts_good = __le32_to_cpu(ps->rts_good);
+			stats->fcs_bad = __le32_to_cpu(ps->fcs_bad);
+			stats->no_beacons = __le32_to_cpu(ps->no_beacons);
+			stats->mib_int_count = __le32_to_cpu(ps->mib_int_count);
+			tmp += sizeof(struct wmi_pdev_stats_10x);
+		} else {
+			tmp += sizeof(struct wmi_pdev_stats_old);
+		}
 	}
 
 	/* 0 or max vdevs */
@@ -243,22 +254,29 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 	}
 
 	if (num_peer_stats) {
-		struct wmi_peer_stats *peer_stats;
+		struct wmi_peer_stats_10x *peer_stats;
 		struct ath10k_peer_stat *s;
 
 		stats->peers = num_peer_stats;
 
 		for (i = 0; i < num_peer_stats; i++) {
-			peer_stats = (struct wmi_peer_stats *)tmp;
+			peer_stats = (struct wmi_peer_stats_10x *)tmp;
 			s = &stats->peer_stat[i];
 
-			WMI_MAC_ADDR_TO_CHAR_ARRAY(&peer_stats->peer_macaddr,
-						   s->peer_macaddr);
+			memcpy(s->peer_macaddr, &peer_stats->peer_macaddr.addr,
+			       ETH_ALEN);
 			s->peer_rssi = __le32_to_cpu(peer_stats->peer_rssi);
 			s->peer_tx_rate =
 				__le32_to_cpu(peer_stats->peer_tx_rate);
-
-			tmp += sizeof(struct wmi_peer_stats);
+			if (test_bit(ATH10K_FW_FEATURE_WMI_10X,
+				     ar->fw_features)) {
+				s->peer_rx_rate =
+					__le32_to_cpu(peer_stats->peer_rx_rate);
+				tmp += sizeof(struct wmi_peer_stats_10x);
+
+			} else {
+				tmp += sizeof(struct wmi_peer_stats_old);
+			}
 		}
 	}
 
@@ -272,7 +290,7 @@ static ssize_t ath10k_read_fw_stats(struct file *file, char __user *user_buf,
 	struct ath10k *ar = file->private_data;
 	struct ath10k_target_stats *fw_stats;
 	char *buf = NULL;
-	unsigned int len = 0, buf_len = 2500;
+	unsigned int len = 0, buf_len = 8000;
 	ssize_t ret_cnt = 0;
 	long left;
 	int i;
@@ -320,6 +338,16 @@ static ssize_t ath10k_read_fw_stats(struct file *file, char __user *user_buf,
 			 "Cycle count", fw_stats->cycle_count);
 	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
 			 "PHY error count", fw_stats->phy_err_count);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "RTS bad count", fw_stats->rts_bad);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "RTS good count", fw_stats->rts_good);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "FCS bad count", fw_stats->fcs_bad);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "No beacon count", fw_stats->no_beacons);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "MIB int count", fw_stats->mib_int_count);
 
 	len += scnprintf(buf + len, buf_len - len, "\n");
 	len += scnprintf(buf + len, buf_len - len, "%30s\n",
@@ -411,8 +439,8 @@ static ssize_t ath10k_read_fw_stats(struct file *file, char __user *user_buf,
 			 "MPDU errors (FCS, MIC, ENC)", fw_stats->mpdu_errs);
 
 	len += scnprintf(buf + len, buf_len - len, "\n");
-	len += scnprintf(buf + len, buf_len - len, "%30s\n",
-			 "ath10k PEER stats");
+	len += scnprintf(buf + len, buf_len - len, "%30s (%d)\n",
+			 "ath10k PEER stats", fw_stats->peers);
 	len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
 				 "=================");
 
@@ -425,6 +453,9 @@ static ssize_t ath10k_read_fw_stats(struct file *file, char __user *user_buf,
 		len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
 				 "Peer TX rate",
 				 fw_stats->peer_stat[i].peer_tx_rate);
+		len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+				 "Peer RX rate",
+				 fw_stats->peer_stat[i].peer_rx_rate);
 		len += scnprintf(buf + len, buf_len - len, "\n");
 	}
 	spin_unlock_bh(&ar->data_lock);
@@ -451,27 +482,37 @@ static ssize_t ath10k_read_simulate_fw_crash(struct file *file,
 					     char __user *user_buf,
 					     size_t count, loff_t *ppos)
 {
-	const char buf[] = "To simulate firmware crash write the keyword"
-			   " `crash` to this file.\nThis will force firmware"
-			   " to report a crash to the host system.\n";
+	const char buf[] = "To simulate firmware crash write one of the"
+			   " keywords to this file:\n `soft` - this will send"
+			   " WMI_FORCE_FW_HANG_ASSERT to firmware if FW"
+			   " supports that command.\n `hard` - this will send"
+			   " to firmware command with illegal parameters"
+			   " causing firmware crash.\n";
+
 	return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));
 }
 
+/* Simulate firmware crash:
+ * 'soft': Call wmi command causing firmware hang. This firmware hang is
+ * recoverable by warm firmware reset.
+ * 'hard': Force firmware crash by setting any vdev parameter for not allowed
+ * vdev id. This is hard firmware crash because it is recoverable only by cold
+ * firmware reset.
+ */
 static ssize_t ath10k_write_simulate_fw_crash(struct file *file,
 					      const char __user *user_buf,
 					      size_t count, loff_t *ppos)
 {
 	struct ath10k *ar = file->private_data;
-	char buf[32] = {};
+	char buf[32];
 	int ret;
 
 	mutex_lock(&ar->conf_mutex);
 
 	simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, user_buf, count);
-	if (strcmp(buf, "crash") && strcmp(buf, "crash\n")) {
-		ret = -EINVAL;
-		goto exit;
-	}
+
+	/* make sure that buf is null terminated */
+	buf[sizeof(buf) - 1] = 0;
 
 	if (ar->state != ATH10K_STATE_ON &&
 	    ar->state != ATH10K_STATE_RESTARTED) {
@@ -479,14 +520,30 @@ static ssize_t ath10k_write_simulate_fw_crash(struct file *file,
 		goto exit;
 	}
 
-	ath10k_info("simulating firmware crash\n");
+	/* drop the possible '\n' from the end */
+	if (buf[count - 1] == '\n') {
+		buf[count - 1] = 0;
+		count--;
+	}
 
-	ret = ath10k_wmi_force_fw_hang(ar, WMI_FORCE_FW_HANG_ASSERT, 0);
-	if (ret)
-		ath10k_warn("failed to force fw hang (%d)\n", ret);
+	if (!strcmp(buf, "soft")) {
+		ath10k_info("simulating soft firmware crash\n");
+		ret = ath10k_wmi_force_fw_hang(ar, WMI_FORCE_FW_HANG_ASSERT, 0);
+	} else if (!strcmp(buf, "hard")) {
+		ath10k_info("simulating hard firmware crash\n");
+		ret = ath10k_wmi_vdev_set_param(ar, TARGET_NUM_VDEVS + 1,
+					ar->wmi.vdev_param->rts_threshold, 0);
+	} else {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	if (ret) {
+		ath10k_warn("failed to simulate firmware crash: %d\n", ret);
+		goto exit;
+	}
 
-	if (ret == 0)
-		ret = count;
+	ret = count;
 
 exit:
 	mutex_unlock(&ar->conf_mutex);
diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 7f1bccd3597f..e493db4b4a41 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -157,6 +157,9 @@ int ath10k_htc_send(struct ath10k_htc *htc,
 			goto err_pull;
 		}
 		ep->tx_credits -= credits;
+		ath10k_dbg(ATH10K_DBG_HTC,
+			   "htc ep %d consumed %d credits (total %d)\n",
+			   eid, credits, ep->tx_credits);
 		spin_unlock_bh(&htc->tx_lock);
 	}
 
@@ -185,6 +188,9 @@ err_credits:
 	if (ep->tx_credit_flow_enabled) {
 		spin_lock_bh(&htc->tx_lock);
 		ep->tx_credits += credits;
+		ath10k_dbg(ATH10K_DBG_HTC,
+			   "htc ep %d reverted %d credits back (total %d)\n",
+			   eid, credits, ep->tx_credits);
 		spin_unlock_bh(&htc->tx_lock);
 
 		if (ep->ep_ops.ep_tx_credits)
@@ -234,12 +240,12 @@ ath10k_htc_process_credit_report(struct ath10k_htc *htc,
 		if (report->eid >= ATH10K_HTC_EP_COUNT)
 			break;
 
-		ath10k_dbg(ATH10K_DBG_HTC, "ep %d got %d credits\n",
-			   report->eid, report->credits);
-
 		ep = &htc->endpoint[report->eid];
 		ep->tx_credits += report->credits;
 
+		ath10k_dbg(ATH10K_DBG_HTC, "htc ep %d got %d credits (total %d)\n",
+			   report->eid, report->credits, ep->tx_credits);
+
 		if (ep->ep_ops.ep_tx_credits) {
 			spin_unlock_bh(&htc->tx_lock);
 			ep->ep_ops.ep_tx_credits(htc->ar);
@@ -824,17 +830,11 @@ int ath10k_htc_start(struct ath10k_htc *htc)
 	return 0;
 }
 
-/*
- * stop HTC communications, i.e. stop interrupt reception, and flush all
- * queued buffers
- */
 void ath10k_htc_stop(struct ath10k_htc *htc)
 {
 	spin_lock_bh(&htc->tx_lock);
 	htc->stopped = true;
 	spin_unlock_bh(&htc->tx_lock);
-
-	ath10k_hif_stop(htc->ar);
 }
 
 /* registered target arrival callback from the HIF layer */
diff --git a/drivers/net/wireless/ath/ath10k/htt.c b/drivers/net/wireless/ath/ath10k/htt.c
index 69697af59ce0..19c12cc8d663 100644
--- a/drivers/net/wireless/ath/ath10k/htt.c
+++ b/drivers/net/wireless/ath/ath10k/htt.c
@@ -22,7 +22,7 @@
 #include "core.h"
 #include "debug.h"
 
-static int ath10k_htt_htc_attach(struct ath10k_htt *htt)
+int ath10k_htt_connect(struct ath10k_htt *htt)
 {
 	struct ath10k_htc_svc_conn_req conn_req;
 	struct ath10k_htc_svc_conn_resp conn_resp;
@@ -48,39 +48,14 @@ static int ath10k_htt_htc_attach(struct ath10k_htt *htt)
 	return 0;
 }
 
-int ath10k_htt_attach(struct ath10k *ar)
+int ath10k_htt_init(struct ath10k *ar)
 {
 	struct ath10k_htt *htt = &ar->htt;
-	int ret;
 
 	htt->ar = ar;
 	htt->max_throughput_mbps = 800;
 
 	/*
-	 * Connect to HTC service.
-	 * This has to be done before calling ath10k_htt_rx_attach,
-	 * since ath10k_htt_rx_attach involves sending a rx ring configure
-	 * message to the target.
-	 */
-	ret = ath10k_htt_htc_attach(htt);
-	if (ret) {
-		ath10k_err("could not attach htt htc (%d)\n", ret);
-		goto err_htc_attach;
-	}
-
-	ret = ath10k_htt_tx_attach(htt);
-	if (ret) {
-		ath10k_err("could not attach htt tx (%d)\n", ret);
-		goto err_htc_attach;
-	}
-
-	ret = ath10k_htt_rx_attach(htt);
-	if (ret) {
-		ath10k_err("could not attach htt rx (%d)\n", ret);
-		goto err_rx_attach;
-	}
-
-	/*
 	 * Prefetch enough data to satisfy target
 	 * classification engine.
 	 * This is for LL chips. HL chips will probably
@@ -93,11 +68,6 @@ int ath10k_htt_attach(struct ath10k *ar)
 		2; /* ip4 dscp or ip6 priority */
 
 	return 0;
-
-err_rx_attach:
-	ath10k_htt_tx_detach(htt);
-err_htc_attach:
-	return ret;
 }
 
 #define HTT_TARGET_VERSION_TIMEOUT_HZ (3*HZ)
@@ -117,7 +87,7 @@ static int ath10k_htt_verify_version(struct ath10k_htt *htt)
 	return 0;
 }
 
-int ath10k_htt_attach_target(struct ath10k_htt *htt)
+int ath10k_htt_setup(struct ath10k_htt *htt)
 {
 	int status;
 
@@ -140,9 +110,3 @@ int ath10k_htt_attach_target(struct ath10k_htt *htt)
 
 	return ath10k_htt_send_rx_ring_cfg_ll(htt);
 }
-
-void ath10k_htt_detach(struct ath10k_htt *htt)
-{
-	ath10k_htt_rx_detach(htt);
-	ath10k_htt_tx_detach(htt);
-}
diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index 654867fc1ae7..9a263462c793 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -21,6 +21,7 @@
 #include <linux/bug.h>
 #include <linux/interrupt.h>
 #include <linux/dmapool.h>
+#include <net/mac80211.h>
 
 #include "htc.h"
 #include "rx_desc.h"
@@ -1172,23 +1173,6 @@ struct htt_peer_unmap_event {
 	u16 peer_id;
 };
 
-struct htt_rx_info {
-	struct sk_buff *skb;
-	enum htt_rx_mpdu_status status;
-	enum htt_rx_mpdu_encrypt_type encrypt_type;
-	s8 signal;
-	struct {
-		u8 info0;
-		u32 info1;
-		u32 info2;
-	} rate;
-
-	u32 tsf;
-	bool fcs_err;
-	bool amsdu_more;
-	bool mic_err;
-};
-
 struct ath10k_htt_txbuf {
 	struct htt_data_tx_desc_frag frags[2];
 	struct ath10k_htc_hdr htc_hdr;
@@ -1289,6 +1273,9 @@ struct ath10k_htt {
 	struct tasklet_struct txrx_compl_task;
 	struct sk_buff_head tx_compl_q;
 	struct sk_buff_head rx_compl_q;
+
+	/* rx_status template */
+	struct ieee80211_rx_status rx_status;
 };
 
 #define RX_HTT_HDR_STATUS_LEN 64
@@ -1341,14 +1328,16 @@ struct htt_rx_desc {
 #define HTT_LOG2_MAX_CACHE_LINE_SIZE 7	/* 2^7 = 128 */
 #define HTT_MAX_CACHE_LINE_SIZE_MASK ((1 << HTT_LOG2_MAX_CACHE_LINE_SIZE) - 1)
 
-int ath10k_htt_attach(struct ath10k *ar);
-int ath10k_htt_attach_target(struct ath10k_htt *htt);
-void ath10k_htt_detach(struct ath10k_htt *htt);
+int ath10k_htt_connect(struct ath10k_htt *htt);
+int ath10k_htt_init(struct ath10k *ar);
+int ath10k_htt_setup(struct ath10k_htt *htt);
+
+int ath10k_htt_tx_alloc(struct ath10k_htt *htt);
+void ath10k_htt_tx_free(struct ath10k_htt *htt);
+
+int ath10k_htt_rx_alloc(struct ath10k_htt *htt);
+void ath10k_htt_rx_free(struct ath10k_htt *htt);
 
-int ath10k_htt_tx_attach(struct ath10k_htt *htt);
-void ath10k_htt_tx_detach(struct ath10k_htt *htt);
-int ath10k_htt_rx_attach(struct ath10k_htt *htt);
-void ath10k_htt_rx_detach(struct ath10k_htt *htt);
 void ath10k_htt_htc_tx_complete(struct ath10k *ar, struct sk_buff *skb);
 void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb);
 int ath10k_htt_h2t_ver_req_msg(struct ath10k_htt *htt);
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index cdcbe2de95f9..6c102b1312ff 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -225,10 +225,26 @@ static void ath10k_htt_rx_ring_refill_retry(unsigned long arg)
 	ath10k_htt_rx_msdu_buff_replenish(htt);
 }
 
-void ath10k_htt_rx_detach(struct ath10k_htt *htt)
+static void ath10k_htt_rx_ring_clean_up(struct ath10k_htt *htt)
 {
-	int sw_rd_idx = htt->rx_ring.sw_rd_idx.msdu_payld;
+	struct sk_buff *skb;
+	int i;
+
+	for (i = 0; i < htt->rx_ring.size; i++) {
+		skb = htt->rx_ring.netbufs_ring[i];
+		if (!skb)
+			continue;
 
+		dma_unmap_single(htt->ar->dev, ATH10K_SKB_CB(skb)->paddr,
+				 skb->len + skb_tailroom(skb),
+				 DMA_FROM_DEVICE);
+		dev_kfree_skb_any(skb);
+		htt->rx_ring.netbufs_ring[i] = NULL;
+	}
+}
+
+void ath10k_htt_rx_free(struct ath10k_htt *htt)
+{
 	del_timer_sync(&htt->rx_ring.refill_retry_timer);
 	tasklet_kill(&htt->rx_replenish_task);
 	tasklet_kill(&htt->txrx_compl_task);
@@ -236,18 +252,7 @@ void ath10k_htt_rx_detach(struct ath10k_htt *htt)
 	skb_queue_purge(&htt->tx_compl_q);
 	skb_queue_purge(&htt->rx_compl_q);
 
-	while (sw_rd_idx != __le32_to_cpu(*(htt->rx_ring.alloc_idx.vaddr))) {
-		struct sk_buff *skb =
-				htt->rx_ring.netbufs_ring[sw_rd_idx];
-		struct ath10k_skb_cb *cb = ATH10K_SKB_CB(skb);
-
-		dma_unmap_single(htt->ar->dev, cb->paddr,
-				 skb->len + skb_tailroom(skb),
-				 DMA_FROM_DEVICE);
-		dev_kfree_skb_any(htt->rx_ring.netbufs_ring[sw_rd_idx]);
-		sw_rd_idx++;
-		sw_rd_idx &= htt->rx_ring.size_mask;
-	}
+	ath10k_htt_rx_ring_clean_up(htt);
 
 	dma_free_coherent(htt->ar->dev,
 			  (htt->rx_ring.size *
@@ -277,6 +282,7 @@ static inline struct sk_buff *ath10k_htt_rx_netbuf_pop(struct ath10k_htt *htt)
 
 	idx = htt->rx_ring.sw_rd_idx.msdu_payld;
 	msdu = htt->rx_ring.netbufs_ring[idx];
+	htt->rx_ring.netbufs_ring[idx] = NULL;
 
 	idx++;
 	idx &= htt->rx_ring.size_mask;
@@ -297,6 +303,7 @@ static void ath10k_htt_rx_free_msdu_chain(struct sk_buff *skb)
 	}
 }
 
+/* return: < 0 fatal error, 0 - non chained msdu, 1 chained msdu */
 static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt,
 				   u8 **fw_desc, int *fw_desc_len,
 				   struct sk_buff **head_msdu,
@@ -305,12 +312,13 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt,
 	int msdu_len, msdu_chaining = 0;
 	struct sk_buff *msdu;
 	struct htt_rx_desc *rx_desc;
+	bool corrupted = false;
 
 	lockdep_assert_held(&htt->rx_ring.lock);
 
 	if (htt->rx_confused) {
 		ath10k_warn("htt is confused. refusing rx\n");
-		return 0;
+		return -1;
 	}
 
 	msdu = *head_msdu = ath10k_htt_rx_netbuf_pop(htt);
@@ -398,7 +406,6 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt,
 		msdu_len = MS(__le32_to_cpu(rx_desc->msdu_start.info0),
 			      RX_MSDU_START_INFO0_MSDU_LENGTH);
 		msdu_chained = rx_desc->frag_info.ring2_more_count;
-		msdu_chaining = msdu_chained;
 
 		if (msdu_len_invalid)
 			msdu_len = 0;
@@ -426,11 +433,15 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt,
 
 			msdu->next = next;
 			msdu = next;
+			msdu_chaining = 1;
 		}
 
 		last_msdu = __le32_to_cpu(rx_desc->msdu_end.info0) &
 				RX_MSDU_END_INFO0_LAST_MSDU;
 
+		if (msdu_chaining && !last_msdu)
+			corrupted = true;
+
 		if (last_msdu) {
 			msdu->next = NULL;
 			break;
@@ -442,6 +453,23 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt,
 	}
 	*tail_msdu = msdu;
 
+	if (*head_msdu == NULL)
+		msdu_chaining = -1;
+
+	/*
+	 * Apparently FW sometimes reports weird chained MSDU sequences with
+	 * more than one rx descriptor. This seems like a bug but needs more
+	 * analyzing. For the time being fix it by dropping such sequences to
+	 * avoid blowing up the host system.
+	 */
+	if (corrupted) {
+		ath10k_warn("failed to pop chained msdus, dropping\n");
+		ath10k_htt_rx_free_msdu_chain(*head_msdu);
+		*head_msdu = NULL;
+		*tail_msdu = NULL;
+		msdu_chaining = -EINVAL;
+	}
+
 	/*
 	 * Don't refill the ring yet.
 	 *
@@ -464,7 +492,7 @@ static void ath10k_htt_rx_replenish_task(unsigned long ptr)
 	ath10k_htt_rx_msdu_buff_replenish(htt);
 }
 
-int ath10k_htt_rx_attach(struct ath10k_htt *htt)
+int ath10k_htt_rx_alloc(struct ath10k_htt *htt)
 {
 	dma_addr_t paddr;
 	void *vaddr;
@@ -490,7 +518,7 @@ int ath10k_htt_rx_attach(struct ath10k_htt *htt)
 	htt->rx_ring.fill_level = ath10k_htt_rx_ring_fill_level(htt);
 
 	htt->rx_ring.netbufs_ring =
-		kmalloc(htt->rx_ring.size * sizeof(struct sk_buff *),
+		kzalloc(htt->rx_ring.size * sizeof(struct sk_buff *),
 			GFP_KERNEL);
 	if (!htt->rx_ring.netbufs_ring)
 		goto err_netbuf;
@@ -636,6 +664,203 @@ struct amsdu_subframe_hdr {
 	__be16 len;
 } __packed;
 
+static const u8 rx_legacy_rate_idx[] = {
+	3,	/* 0x00  - 11Mbps  */
+	2,	/* 0x01  - 5.5Mbps */
+	1,	/* 0x02  - 2Mbps   */
+	0,	/* 0x03  - 1Mbps   */
+	3,	/* 0x04  - 11Mbps  */
+	2,	/* 0x05  - 5.5Mbps */
+	1,	/* 0x06  - 2Mbps   */
+	0,	/* 0x07  - 1Mbps   */
+	10,	/* 0x08  - 48Mbps  */
+	8,	/* 0x09  - 24Mbps  */
+	6,	/* 0x0A  - 12Mbps  */
+	4,	/* 0x0B  - 6Mbps   */
+	11,	/* 0x0C  - 54Mbps  */
+	9,	/* 0x0D  - 36Mbps  */
+	7,	/* 0x0E  - 18Mbps  */
+	5,	/* 0x0F  - 9Mbps   */
+};
+
+static void ath10k_htt_rx_h_rates(struct ath10k *ar,
+				  enum ieee80211_band band,
+				  u8 info0, u32 info1, u32 info2,
+				  struct ieee80211_rx_status *status)
+{
+	u8 cck, rate, rate_idx, bw, sgi, mcs, nss;
+	u8 preamble = 0;
+
+	/* Check if valid fields */
+	if (!(info0 & HTT_RX_INDICATION_INFO0_START_VALID))
+		return;
+
+	preamble = MS(info1, HTT_RX_INDICATION_INFO1_PREAMBLE_TYPE);
+
+	switch (preamble) {
+	case HTT_RX_LEGACY:
+		cck = info0 & HTT_RX_INDICATION_INFO0_LEGACY_RATE_CCK;
+		rate = MS(info0, HTT_RX_INDICATION_INFO0_LEGACY_RATE);
+		rate_idx = 0;
+
+		if (rate < 0x08 || rate > 0x0F)
+			break;
+
+		switch (band) {
+		case IEEE80211_BAND_2GHZ:
+			if (cck)
+				rate &= ~BIT(3);
+			rate_idx = rx_legacy_rate_idx[rate];
+			break;
+		case IEEE80211_BAND_5GHZ:
+			rate_idx = rx_legacy_rate_idx[rate];
+			/* We are using same rate table registering
+			   HW - ath10k_rates[]. In case of 5GHz skip
+			   CCK rates, so -4 here */
+			rate_idx -= 4;
+			break;
+		default:
+			break;
+		}
+
+		status->rate_idx = rate_idx;
+		break;
+	case HTT_RX_HT:
+	case HTT_RX_HT_WITH_TXBF:
+		/* HT-SIG - Table 20-11 in info1 and info2 */
+		mcs = info1 & 0x1F;
+		nss = mcs >> 3;
+		bw = (info1 >> 7) & 1;
+		sgi = (info2 >> 7) & 1;
+
+		status->rate_idx = mcs;
+		status->flag |= RX_FLAG_HT;
+		if (sgi)
+			status->flag |= RX_FLAG_SHORT_GI;
+		if (bw)
+			status->flag |= RX_FLAG_40MHZ;
+		break;
+	case HTT_RX_VHT:
+	case HTT_RX_VHT_WITH_TXBF:
+		/* VHT-SIG-A1 in info 1, VHT-SIG-A2 in info2
+		   TODO check this */
+		mcs = (info2 >> 4) & 0x0F;
+		nss = ((info1 >> 10) & 0x07) + 1;
+		bw = info1 & 3;
+		sgi = info2 & 1;
+
+		status->rate_idx = mcs;
+		status->vht_nss = nss;
+
+		if (sgi)
+			status->flag |= RX_FLAG_SHORT_GI;
+
+		switch (bw) {
+		/* 20MHZ */
+		case 0:
+			break;
+		/* 40MHZ */
+		case 1:
+			status->flag |= RX_FLAG_40MHZ;
+			break;
+		/* 80MHZ */
+		case 2:
+			status->vht_flag |= RX_VHT_FLAG_80MHZ;
+		}
+
+		status->flag |= RX_FLAG_VHT;
+		break;
+	default:
+		break;
+	}
+}
+
+static void ath10k_htt_rx_h_protected(struct ath10k_htt *htt,
+				      struct ieee80211_rx_status *rx_status,
+				      struct sk_buff *skb,
+				      enum htt_rx_mpdu_encrypt_type enctype,
+				      enum rx_msdu_decap_format fmt,
+				      bool dot11frag)
+{
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+
+	rx_status->flag &= ~(RX_FLAG_DECRYPTED |
+			     RX_FLAG_IV_STRIPPED |
+			     RX_FLAG_MMIC_STRIPPED);
+
+	if (enctype == HTT_RX_MPDU_ENCRYPT_NONE)
+		return;
+
+	/*
+	 * There's no explicit rx descriptor flag to indicate whether a given
+	 * frame has been decrypted or not. We're forced to use the decap
+	 * format as an implicit indication. However fragmentation rx is always
+	 * raw and it probably never reports undecrypted raws.
+	 *
+	 * This makes sure sniffed frames are reported as-is without stripping
+	 * the protected flag.
+	 */
+	if (fmt == RX_MSDU_DECAP_RAW && !dot11frag)
+		return;
+
+	rx_status->flag |= RX_FLAG_DECRYPTED |
+			   RX_FLAG_IV_STRIPPED |
+			   RX_FLAG_MMIC_STRIPPED;
+	hdr->frame_control = __cpu_to_le16(__le16_to_cpu(hdr->frame_control) &
+					   ~IEEE80211_FCTL_PROTECTED);
+}
+
+static bool ath10k_htt_rx_h_channel(struct ath10k *ar,
+				    struct ieee80211_rx_status *status)
+{
+	struct ieee80211_channel *ch;
+
+	spin_lock_bh(&ar->data_lock);
+	ch = ar->scan_channel;
+	if (!ch)
+		ch = ar->rx_channel;
+	spin_unlock_bh(&ar->data_lock);
+
+	if (!ch)
+		return false;
+
+	status->band = ch->band;
+	status->freq = ch->center_freq;
+
+	return true;
+}
+
+static void ath10k_process_rx(struct ath10k *ar,
+			      struct ieee80211_rx_status *rx_status,
+			      struct sk_buff *skb)
+{
+	struct ieee80211_rx_status *status;
+
+	status = IEEE80211_SKB_RXCB(skb);
+	*status = *rx_status;
+
+	ath10k_dbg(ATH10K_DBG_DATA,
+		   "rx skb %p len %u %s%s%s%s%s %srate_idx %u vht_nss %u freq %u band %u flag 0x%x fcs-err %imic-err %i\n",
+		   skb,
+		   skb->len,
+		   status->flag == 0 ? "legacy" : "",
+		   status->flag & RX_FLAG_HT ? "ht" : "",
+		   status->flag & RX_FLAG_VHT ? "vht" : "",
+		   status->flag & RX_FLAG_40MHZ ? "40" : "",
+		   status->vht_flag & RX_VHT_FLAG_80MHZ ? "80" : "",
+		   status->flag & RX_FLAG_SHORT_GI ? "sgi " : "",
+		   status->rate_idx,
+		   status->vht_nss,
+		   status->freq,
+		   status->band, status->flag,
+		   !!(status->flag & RX_FLAG_FAILED_FCS_CRC),
+		   !!(status->flag & RX_FLAG_MMIC_ERROR));
+	ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "rx skb: ",
+			skb->data, skb->len);
+
+	ieee80211_rx(ar->hw, skb);
+}
+
 static int ath10k_htt_rx_nwifi_hdrlen(struct ieee80211_hdr *hdr)
 {
 	/* nwifi header is padded to 4 bytes. this fixes 4addr rx */
@@ -643,11 +868,12 @@ static int ath10k_htt_rx_nwifi_hdrlen(struct ieee80211_hdr *hdr)
 }
 
 static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
-				struct htt_rx_info *info)
+				struct ieee80211_rx_status *rx_status,
+				struct sk_buff *skb_in)
 {
 	struct htt_rx_desc *rxd;
+	struct sk_buff *skb = skb_in;
 	struct sk_buff *first;
-	struct sk_buff *skb = info->skb;
 	enum rx_msdu_decap_format fmt;
 	enum htt_rx_mpdu_encrypt_type enctype;
 	struct ieee80211_hdr *hdr;
@@ -728,24 +954,28 @@ static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
 			break;
 		}
 
-		info->skb = skb;
-		info->encrypt_type = enctype;
+		skb_in = skb;
+		ath10k_htt_rx_h_protected(htt, rx_status, skb_in, enctype, fmt,
+					  false);
 		skb = skb->next;
-		info->skb->next = NULL;
+		skb_in->next = NULL;
 
 		if (skb)
-			info->amsdu_more = true;
+			rx_status->flag |= RX_FLAG_AMSDU_MORE;
+		else
+			rx_status->flag &= ~RX_FLAG_AMSDU_MORE;
 
-		ath10k_process_rx(htt->ar, info);
+		ath10k_process_rx(htt->ar, rx_status, skb_in);
 	}
 
 	/* FIXME: It might be nice to re-assemble the A-MSDU when there's a
 	 * monitor interface active for sniffing purposes. */
 }
 
-static void ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
+static void ath10k_htt_rx_msdu(struct ath10k_htt *htt,
+			       struct ieee80211_rx_status *rx_status,
+			       struct sk_buff *skb)
 {
-	struct sk_buff *skb = info->skb;
 	struct htt_rx_desc *rxd;
 	struct ieee80211_hdr *hdr;
 	enum rx_msdu_decap_format fmt;
@@ -808,66 +1038,9 @@ static void ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 		break;
 	}
 
-	info->skb = skb;
-	info->encrypt_type = enctype;
+	ath10k_htt_rx_h_protected(htt, rx_status, skb, enctype, fmt, false);
 
-	ath10k_process_rx(htt->ar, info);
-}
-
-static bool ath10k_htt_rx_has_decrypt_err(struct sk_buff *skb)
-{
-	struct htt_rx_desc *rxd;
-	u32 flags;
-
-	rxd = (void *)skb->data - sizeof(*rxd);
-	flags = __le32_to_cpu(rxd->attention.flags);
-
-	if (flags & RX_ATTENTION_FLAGS_DECRYPT_ERR)
-		return true;
-
-	return false;
-}
-
-static bool ath10k_htt_rx_has_fcs_err(struct sk_buff *skb)
-{
-	struct htt_rx_desc *rxd;
-	u32 flags;
-
-	rxd = (void *)skb->data - sizeof(*rxd);
-	flags = __le32_to_cpu(rxd->attention.flags);
-
-	if (flags & RX_ATTENTION_FLAGS_FCS_ERR)
-		return true;
-
-	return false;
-}
-
-static bool ath10k_htt_rx_has_mic_err(struct sk_buff *skb)
-{
-	struct htt_rx_desc *rxd;
-	u32 flags;
-
-	rxd = (void *)skb->data - sizeof(*rxd);
-	flags = __le32_to_cpu(rxd->attention.flags);
-
-	if (flags & RX_ATTENTION_FLAGS_TKIP_MIC_ERR)
-		return true;
-
-	return false;
-}
-
-static bool ath10k_htt_rx_is_mgmt(struct sk_buff *skb)
-{
-	struct htt_rx_desc *rxd;
-	u32 flags;
-
-	rxd = (void *)skb->data - sizeof(*rxd);
-	flags = __le32_to_cpu(rxd->attention.flags);
-
-	if (flags & RX_ATTENTION_FLAGS_MGMT_TYPE)
-		return true;
-
-	return false;
+	ath10k_process_rx(htt->ar, rx_status, skb);
 }
 
 static int ath10k_htt_rx_get_csum_state(struct sk_buff *skb)
@@ -952,21 +1125,73 @@ static int ath10k_unchain_msdu(struct sk_buff *msdu_head)
 	return 0;
 }
 
+static bool ath10k_htt_rx_amsdu_allowed(struct ath10k_htt *htt,
+					struct sk_buff *head,
+					enum htt_rx_mpdu_status status,
+					bool channel_set,
+					u32 attention)
+{
+	if (head->len == 0) {
+		ath10k_dbg(ATH10K_DBG_HTT,
+			   "htt rx dropping due to zero-len\n");
+		return false;
+	}
+
+	if (attention & RX_ATTENTION_FLAGS_DECRYPT_ERR) {
+		ath10k_dbg(ATH10K_DBG_HTT,
+			   "htt rx dropping due to decrypt-err\n");
+		return false;
+	}
+
+	if (!channel_set) {
+		ath10k_warn("no channel configured; ignoring frame!\n");
+		return false;
+	}
+
+	/* Skip mgmt frames while we handle this in WMI */
+	if (status == HTT_RX_IND_MPDU_STATUS_MGMT_CTRL ||
+	    attention & RX_ATTENTION_FLAGS_MGMT_TYPE) {
+		ath10k_dbg(ATH10K_DBG_HTT, "htt rx mgmt ctrl\n");
+		return false;
+	}
+
+	if (status != HTT_RX_IND_MPDU_STATUS_OK &&
+	    status != HTT_RX_IND_MPDU_STATUS_TKIP_MIC_ERR &&
+	    status != HTT_RX_IND_MPDU_STATUS_ERR_INV_PEER &&
+	    !htt->ar->monitor_started) {
+		ath10k_dbg(ATH10K_DBG_HTT,
+			   "htt rx ignoring frame w/ status %d\n",
+			   status);
+		return false;
+	}
+
+	if (test_bit(ATH10K_CAC_RUNNING, &htt->ar->dev_flags)) {
+		ath10k_dbg(ATH10K_DBG_HTT,
+			   "htt rx CAC running\n");
+		return false;
+	}
+
+	return true;
+}
+
 static void ath10k_htt_rx_handler(struct ath10k_htt *htt,
 				  struct htt_rx_indication *rx)
 {
-	struct htt_rx_info info;
+	struct ieee80211_rx_status *rx_status = &htt->rx_status;
 	struct htt_rx_indication_mpdu_range *mpdu_ranges;
+	struct htt_rx_desc *rxd;
+	enum htt_rx_mpdu_status status;
 	struct ieee80211_hdr *hdr;
 	int num_mpdu_ranges;
+	u32 attention;
 	int fw_desc_len;
 	u8 *fw_desc;
+	bool channel_set;
 	int i, j;
+	int ret;
 
 	lockdep_assert_held(&htt->rx_ring.lock);
 
-	memset(&info, 0, sizeof(info));
-
 	fw_desc_len = __le16_to_cpu(rx->prefix.fw_rx_desc_bytes);
 	fw_desc = (u8 *)&rx->fw_desc;
 
@@ -974,106 +1199,90 @@ static void ath10k_htt_rx_handler(struct ath10k_htt *htt,
 			     HTT_RX_INDICATION_INFO1_NUM_MPDU_RANGES);
 	mpdu_ranges = htt_rx_ind_get_mpdu_ranges(rx);
 
+	/* Fill this once, while this is per-ppdu */
+	if (rx->ppdu.info0 & HTT_RX_INDICATION_INFO0_START_VALID) {
+		memset(rx_status, 0, sizeof(*rx_status));
+		rx_status->signal  = ATH10K_DEFAULT_NOISE_FLOOR +
+				     rx->ppdu.combined_rssi;
+	}
+
+	if (rx->ppdu.info0 & HTT_RX_INDICATION_INFO0_END_VALID) {
+		/* TSF available only in 32-bit */
+		rx_status->mactime = __le32_to_cpu(rx->ppdu.tsf) & 0xffffffff;
+		rx_status->flag |= RX_FLAG_MACTIME_END;
+	}
+
+	channel_set = ath10k_htt_rx_h_channel(htt->ar, rx_status);
+
+	if (channel_set) {
+		ath10k_htt_rx_h_rates(htt->ar, rx_status->band,
+				      rx->ppdu.info0,
+				      __le32_to_cpu(rx->ppdu.info1),
+				      __le32_to_cpu(rx->ppdu.info2),
+				      rx_status);
+	}
+
 	ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "htt rx ind: ",
 			rx, sizeof(*rx) +
 			(sizeof(struct htt_rx_indication_mpdu_range) *
 				num_mpdu_ranges));
 
 	for (i = 0; i < num_mpdu_ranges; i++) {
-		info.status = mpdu_ranges[i].mpdu_range_status;
+		status = mpdu_ranges[i].mpdu_range_status;
 
 		for (j = 0; j < mpdu_ranges[i].mpdu_count; j++) {
 			struct sk_buff *msdu_head, *msdu_tail;
-			enum htt_rx_mpdu_status status;
-			int msdu_chaining;
 
 			msdu_head = NULL;
 			msdu_tail = NULL;
-			msdu_chaining = ath10k_htt_rx_amsdu_pop(htt,
-							 &fw_desc,
-							 &fw_desc_len,
-							 &msdu_head,
-							 &msdu_tail);
-
-			if (!msdu_head) {
-				ath10k_warn("htt rx no data!\n");
-				continue;
-			}
-
-			if (msdu_head->len == 0) {
-				ath10k_dbg(ATH10K_DBG_HTT,
-					   "htt rx dropping due to zero-len\n");
+			ret = ath10k_htt_rx_amsdu_pop(htt,
+						      &fw_desc,
+						      &fw_desc_len,
+						      &msdu_head,
+						      &msdu_tail);
+
+			if (ret < 0) {
+				ath10k_warn("failed to pop amsdu from htt rx ring %d\n",
+					    ret);
 				ath10k_htt_rx_free_msdu_chain(msdu_head);
 				continue;
 			}
 
-			if (ath10k_htt_rx_has_decrypt_err(msdu_head)) {
-				ath10k_dbg(ATH10K_DBG_HTT,
-					   "htt rx dropping due to decrypt-err\n");
-				ath10k_htt_rx_free_msdu_chain(msdu_head);
-				continue;
-			}
+			rxd = container_of((void *)msdu_head->data,
+					   struct htt_rx_desc,
+					   msdu_payload);
+			attention = __le32_to_cpu(rxd->attention.flags);
 
-			status = info.status;
-
-			/* Skip mgmt frames while we handle this in WMI */
-			if (status == HTT_RX_IND_MPDU_STATUS_MGMT_CTRL ||
-			    ath10k_htt_rx_is_mgmt(msdu_head)) {
-				ath10k_dbg(ATH10K_DBG_HTT, "htt rx mgmt ctrl\n");
+			if (!ath10k_htt_rx_amsdu_allowed(htt, msdu_head,
+							 status,
+							 channel_set,
+							 attention)) {
 				ath10k_htt_rx_free_msdu_chain(msdu_head);
 				continue;
 			}
 
-			if (status != HTT_RX_IND_MPDU_STATUS_OK &&
-			    status != HTT_RX_IND_MPDU_STATUS_TKIP_MIC_ERR &&
-			    status != HTT_RX_IND_MPDU_STATUS_ERR_INV_PEER &&
-			    !htt->ar->monitor_enabled) {
-				ath10k_dbg(ATH10K_DBG_HTT,
-					   "htt rx ignoring frame w/ status %d\n",
-					   status);
+			if (ret > 0 &&
+			    ath10k_unchain_msdu(msdu_head) < 0) {
 				ath10k_htt_rx_free_msdu_chain(msdu_head);
 				continue;
 			}
 
-			if (test_bit(ATH10K_CAC_RUNNING, &htt->ar->dev_flags)) {
-				ath10k_dbg(ATH10K_DBG_HTT,
-					   "htt rx CAC running\n");
-				ath10k_htt_rx_free_msdu_chain(msdu_head);
-				continue;
-			}
-
-			if (msdu_chaining &&
-			    (ath10k_unchain_msdu(msdu_head) < 0)) {
-				ath10k_htt_rx_free_msdu_chain(msdu_head);
-				continue;
-			}
-
-			info.skb     = msdu_head;
-			info.fcs_err = ath10k_htt_rx_has_fcs_err(msdu_head);
-			info.mic_err = ath10k_htt_rx_has_mic_err(msdu_head);
-
-			if (info.fcs_err)
-				ath10k_dbg(ATH10K_DBG_HTT,
-					   "htt rx has FCS err\n");
-
-			if (info.mic_err)
-				ath10k_dbg(ATH10K_DBG_HTT,
-					   "htt rx has MIC err\n");
-
-			info.signal  = ATH10K_DEFAULT_NOISE_FLOOR;
-			info.signal += rx->ppdu.combined_rssi;
+			if (attention & RX_ATTENTION_FLAGS_FCS_ERR)
+				rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
+			else
+				rx_status->flag &= ~RX_FLAG_FAILED_FCS_CRC;
 
-			info.rate.info0 = rx->ppdu.info0;
-			info.rate.info1 = __le32_to_cpu(rx->ppdu.info1);
-			info.rate.info2 = __le32_to_cpu(rx->ppdu.info2);
-			info.tsf = __le32_to_cpu(rx->ppdu.tsf);
+			if (attention & RX_ATTENTION_FLAGS_TKIP_MIC_ERR)
+				rx_status->flag |= RX_FLAG_MMIC_ERROR;
+			else
+				rx_status->flag &= ~RX_FLAG_MMIC_ERROR;
 
 			hdr = ath10k_htt_rx_skb_get_hdr(msdu_head);
 
 			if (ath10k_htt_rx_hdr_is_amsdu(hdr))
-				ath10k_htt_rx_amsdu(htt, &info);
+				ath10k_htt_rx_amsdu(htt, rx_status, msdu_head);
 			else
-				ath10k_htt_rx_msdu(htt, &info);
+				ath10k_htt_rx_msdu(htt, rx_status, msdu_head);
 		}
 	}
 
@@ -1084,11 +1293,12 @@ static void ath10k_htt_rx_frag_handler(struct ath10k_htt *htt,
 				struct htt_rx_fragment_indication *frag)
 {
 	struct sk_buff *msdu_head, *msdu_tail;
+	enum htt_rx_mpdu_encrypt_type enctype;
 	struct htt_rx_desc *rxd;
 	enum rx_msdu_decap_format fmt;
-	struct htt_rx_info info = {};
+	struct ieee80211_rx_status *rx_status = &htt->rx_status;
 	struct ieee80211_hdr *hdr;
-	int msdu_chaining;
+	int ret;
 	bool tkip_mic_err;
 	bool decrypt_err;
 	u8 *fw_desc;
@@ -1102,24 +1312,21 @@ static void ath10k_htt_rx_frag_handler(struct ath10k_htt *htt,
 	msdu_tail = NULL;
 
 	spin_lock_bh(&htt->rx_ring.lock);
-	msdu_chaining = ath10k_htt_rx_amsdu_pop(htt, &fw_desc, &fw_desc_len,
-						&msdu_head, &msdu_tail);
+	ret = ath10k_htt_rx_amsdu_pop(htt, &fw_desc, &fw_desc_len,
+				      &msdu_head, &msdu_tail);
 	spin_unlock_bh(&htt->rx_ring.lock);
 
 	ath10k_dbg(ATH10K_DBG_HTT_DUMP, "htt rx frag ahead\n");
 
-	if (!msdu_head) {
-		ath10k_warn("htt rx frag no data\n");
-		return;
-	}
-
-	if (msdu_chaining || msdu_head != msdu_tail) {
-		ath10k_warn("aggregation with fragmentation?!\n");
+	if (ret) {
+		ath10k_warn("failed to pop amsdu from httr rx ring for fragmented rx %d\n",
+			    ret);
 		ath10k_htt_rx_free_msdu_chain(msdu_head);
 		return;
 	}
 
 	/* FIXME: implement signal strength */
+	rx_status->flag |= RX_FLAG_NO_SIGNAL_VAL;
 
 	hdr = (struct ieee80211_hdr *)msdu_head->data;
 	rxd = (void *)msdu_head->data - sizeof(*rxd);
@@ -1136,57 +1343,55 @@ static void ath10k_htt_rx_frag_handler(struct ath10k_htt *htt,
 		goto end;
 	}
 
-	info.skb = msdu_head;
-	info.status = HTT_RX_IND_MPDU_STATUS_OK;
-	info.encrypt_type = MS(__le32_to_cpu(rxd->mpdu_start.info0),
-				RX_MPDU_START_INFO0_ENCRYPT_TYPE);
-	info.skb->ip_summed = ath10k_htt_rx_get_csum_state(info.skb);
+	enctype = MS(__le32_to_cpu(rxd->mpdu_start.info0),
+		     RX_MPDU_START_INFO0_ENCRYPT_TYPE);
+	ath10k_htt_rx_h_protected(htt, rx_status, msdu_head, enctype, fmt,
+				  true);
+	msdu_head->ip_summed = ath10k_htt_rx_get_csum_state(msdu_head);
 
-	if (tkip_mic_err) {
+	if (tkip_mic_err)
 		ath10k_warn("tkip mic error\n");
-		info.status = HTT_RX_IND_MPDU_STATUS_TKIP_MIC_ERR;
-	}
 
 	if (decrypt_err) {
 		ath10k_warn("decryption err in fragmented rx\n");
-		dev_kfree_skb_any(info.skb);
+		dev_kfree_skb_any(msdu_head);
 		goto end;
 	}
 
-	if (info.encrypt_type != HTT_RX_MPDU_ENCRYPT_NONE) {
+	if (enctype != HTT_RX_MPDU_ENCRYPT_NONE) {
 		hdrlen = ieee80211_hdrlen(hdr->frame_control);
-		paramlen = ath10k_htt_rx_crypto_param_len(info.encrypt_type);
+		paramlen = ath10k_htt_rx_crypto_param_len(enctype);
 
 		/* It is more efficient to move the header than the payload */
-		memmove((void *)info.skb->data + paramlen,
-			(void *)info.skb->data,
+		memmove((void *)msdu_head->data + paramlen,
+			(void *)msdu_head->data,
 			hdrlen);
-		skb_pull(info.skb, paramlen);
-		hdr = (struct ieee80211_hdr *)info.skb->data;
+		skb_pull(msdu_head, paramlen);
+		hdr = (struct ieee80211_hdr *)msdu_head->data;
 	}
 
 	/* remove trailing FCS */
 	trim  = 4;
 
 	/* remove crypto trailer */
-	trim += ath10k_htt_rx_crypto_tail_len(info.encrypt_type);
+	trim += ath10k_htt_rx_crypto_tail_len(enctype);
 
 	/* last fragment of TKIP frags has MIC */
 	if (!ieee80211_has_morefrags(hdr->frame_control) &&
-	    info.encrypt_type == HTT_RX_MPDU_ENCRYPT_TKIP_WPA)
+	    enctype == HTT_RX_MPDU_ENCRYPT_TKIP_WPA)
 		trim += 8;
 
-	if (trim > info.skb->len) {
+	if (trim > msdu_head->len) {
 		ath10k_warn("htt rx fragment: trailer longer than the frame itself? drop\n");
-		dev_kfree_skb_any(info.skb);
+		dev_kfree_skb_any(msdu_head);
 		goto end;
 	}
 
-	skb_trim(info.skb, info.skb->len - trim);
+	skb_trim(msdu_head, msdu_head->len - trim);
 
 	ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "htt rx frag mpdu: ",
-			info.skb->data, info.skb->len);
-	ath10k_process_rx(htt->ar, &info);
+			msdu_head->data, msdu_head->len);
+	ath10k_process_rx(htt->ar, rx_status, msdu_head);
 
 end:
 	if (fw_desc_len > 0) {
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index 7a3e2e40dd5c..7064354d1f4f 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -83,7 +83,7 @@ void ath10k_htt_tx_free_msdu_id(struct ath10k_htt *htt, u16 msdu_id)
 	__clear_bit(msdu_id, htt->used_msdu_ids);
 }
 
-int ath10k_htt_tx_attach(struct ath10k_htt *htt)
+int ath10k_htt_tx_alloc(struct ath10k_htt *htt)
 {
 	spin_lock_init(&htt->tx_lock);
 	init_waitqueue_head(&htt->empty_tx_wq);
@@ -120,7 +120,7 @@ int ath10k_htt_tx_attach(struct ath10k_htt *htt)
 	return 0;
 }
 
-static void ath10k_htt_tx_cleanup_pending(struct ath10k_htt *htt)
+static void ath10k_htt_tx_free_pending(struct ath10k_htt *htt)
 {
 	struct htt_tx_done tx_done = {0};
 	int msdu_id;
@@ -141,9 +141,9 @@ static void ath10k_htt_tx_cleanup_pending(struct ath10k_htt *htt)
 	spin_unlock_bh(&htt->tx_lock);
 }
 
-void ath10k_htt_tx_detach(struct ath10k_htt *htt)
+void ath10k_htt_tx_free(struct ath10k_htt *htt)
 {
-	ath10k_htt_tx_cleanup_pending(htt);
+	ath10k_htt_tx_free_pending(htt);
 	kfree(htt->pending_tx);
 	kfree(htt->used_msdu_ids);
 	dma_pool_destroy(htt->tx_pool);
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 35fc44e281f5..007e855f4ba9 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -28,6 +28,7 @@
 #define QCA988X_HW_2_0_CHIP_ID_REV	0x2
 #define QCA988X_HW_2_0_FW_DIR		"ath10k/QCA988X/hw2.0"
 #define QCA988X_HW_2_0_FW_FILE		"firmware.bin"
+#define QCA988X_HW_2_0_FW_2_FILE	"firmware-2.bin"
 #define QCA988X_HW_2_0_OTP_FILE		"otp.bin"
 #define QCA988X_HW_2_0_BOARD_DATA_FILE	"board.bin"
 #define QCA988X_HW_2_0_PATCH_LOAD_ADDR	0x1234
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 511a2f81e7af..a21080028c54 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -54,7 +54,10 @@ static int ath10k_send_key(struct ath10k_vif *arvif,
 	switch (key->cipher) {
 	case WLAN_CIPHER_SUITE_CCMP:
 		arg.key_cipher = WMI_CIPHER_AES_CCM;
-		key->flags |= IEEE80211_KEY_FLAG_SW_MGMT_TX;
+		if (arvif->vdev_type == WMI_VDEV_TYPE_AP)
+			key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV_MGMT;
+		else
+			key->flags |= IEEE80211_KEY_FLAG_SW_MGMT_TX;
 		break;
 	case WLAN_CIPHER_SUITE_TKIP:
 		arg.key_cipher = WMI_CIPHER_TKIP;
@@ -165,7 +168,7 @@ static int ath10k_clear_peer_keys(struct ath10k_vif *arvif,
 			first_errno = ret;
 
 		if (ret)
-			ath10k_warn("could not remove peer wep key %d (%d)\n",
+			ath10k_warn("failed to remove peer wep key %d: %d\n",
 				    i, ret);
 
 		peer->keys[i] = NULL;
@@ -213,7 +216,8 @@ static int ath10k_clear_vdev_key(struct ath10k_vif *arvif,
 			first_errno = ret;
 
 		if (ret)
-			ath10k_warn("could not remove key for %pM\n", addr);
+			ath10k_warn("failed to remove key for %pM: %d\n",
+				    addr, ret);
 	}
 
 	return first_errno;
@@ -323,14 +327,14 @@ static int ath10k_peer_create(struct ath10k *ar, u32 vdev_id, const u8 *addr)
 
 	ret = ath10k_wmi_peer_create(ar, vdev_id, addr);
 	if (ret) {
-		ath10k_warn("Failed to create wmi peer %pM on vdev %i: %i\n",
+		ath10k_warn("failed to create wmi peer %pM on vdev %i: %i\n",
 			    addr, vdev_id, ret);
 		return ret;
 	}
 
 	ret = ath10k_wait_for_peer_created(ar, vdev_id, addr);
 	if (ret) {
-		ath10k_warn("Failed to wait for created wmi peer %pM on vdev %i: %i\n",
+		ath10k_warn("failed to wait for created wmi peer %pM on vdev %i: %i\n",
 			    addr, vdev_id, ret);
 		return ret;
 	}
@@ -351,7 +355,7 @@ static int ath10k_mac_set_kickout(struct ath10k_vif *arvif)
 	ret = ath10k_wmi_pdev_set_param(ar, param,
 					ATH10K_KICKOUT_THRESHOLD);
 	if (ret) {
-		ath10k_warn("Failed to set kickout threshold on vdev %i: %d\n",
+		ath10k_warn("failed to set kickout threshold on vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		return ret;
 	}
@@ -360,7 +364,7 @@ static int ath10k_mac_set_kickout(struct ath10k_vif *arvif)
 	ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, param,
 					ATH10K_KEEPALIVE_MIN_IDLE);
 	if (ret) {
-		ath10k_warn("Failed to set keepalive minimum idle time on vdev %i : %d\n",
+		ath10k_warn("failed to set keepalive minimum idle time on vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		return ret;
 	}
@@ -369,7 +373,7 @@ static int ath10k_mac_set_kickout(struct ath10k_vif *arvif)
 	ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, param,
 					ATH10K_KEEPALIVE_MAX_IDLE);
 	if (ret) {
-		ath10k_warn("Failed to set keepalive maximum idle time on vdev %i: %d\n",
+		ath10k_warn("failed to set keepalive maximum idle time on vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		return ret;
 	}
@@ -378,7 +382,7 @@ static int ath10k_mac_set_kickout(struct ath10k_vif *arvif)
 	ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, param,
 					ATH10K_KEEPALIVE_MAX_UNRESPONSIVE);
 	if (ret) {
-		ath10k_warn("Failed to set keepalive maximum unresponsive time on vdev %i: %d\n",
+		ath10k_warn("failed to set keepalive maximum unresponsive time on vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		return ret;
 	}
@@ -488,92 +492,20 @@ static inline int ath10k_vdev_setup_sync(struct ath10k *ar)
 	return 0;
 }
 
-static int ath10k_vdev_start(struct ath10k_vif *arvif)
+static bool ath10k_monitor_is_enabled(struct ath10k *ar)
 {
-	struct ath10k *ar = arvif->ar;
-	struct cfg80211_chan_def *chandef = &ar->chandef;
-	struct wmi_vdev_start_request_arg arg = {};
-	int ret = 0;
-
 	lockdep_assert_held(&ar->conf_mutex);
 
-	reinit_completion(&ar->vdev_setup_done);
-
-	arg.vdev_id = arvif->vdev_id;
-	arg.dtim_period = arvif->dtim_period;
-	arg.bcn_intval = arvif->beacon_interval;
-
-	arg.channel.freq = chandef->chan->center_freq;
-	arg.channel.band_center_freq1 = chandef->center_freq1;
-	arg.channel.mode = chan_to_phymode(chandef);
-
-	arg.channel.min_power = 0;
-	arg.channel.max_power = chandef->chan->max_power * 2;
-	arg.channel.max_reg_power = chandef->chan->max_reg_power * 2;
-	arg.channel.max_antenna_gain = chandef->chan->max_antenna_gain * 2;
-
-	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
-		arg.ssid = arvif->u.ap.ssid;
-		arg.ssid_len = arvif->u.ap.ssid_len;
-		arg.hidden_ssid = arvif->u.ap.hidden_ssid;
-
-		/* For now allow DFS for AP mode */
-		arg.channel.chan_radar =
-			!!(chandef->chan->flags & IEEE80211_CHAN_RADAR);
-	} else if (arvif->vdev_type == WMI_VDEV_TYPE_IBSS) {
-		arg.ssid = arvif->vif->bss_conf.ssid;
-		arg.ssid_len = arvif->vif->bss_conf.ssid_len;
-	}
-
 	ath10k_dbg(ATH10K_DBG_MAC,
-		   "mac vdev %d start center_freq %d phymode %s\n",
-		   arg.vdev_id, arg.channel.freq,
-		   ath10k_wmi_phymode_str(arg.channel.mode));
+		   "mac monitor refs: promisc %d monitor %d cac %d\n",
+		   ar->promisc, ar->monitor,
+		   test_bit(ATH10K_CAC_RUNNING, &ar->dev_flags));
 
-	ret = ath10k_wmi_vdev_start(ar, &arg);
-	if (ret) {
-		ath10k_warn("WMI vdev %i start failed: ret %d\n",
-			    arg.vdev_id, ret);
-		return ret;
-	}
-
-	ret = ath10k_vdev_setup_sync(ar);
-	if (ret) {
-		ath10k_warn("vdev %i setup failed %d\n",
-			    arg.vdev_id, ret);
-		return ret;
-	}
-
-	return ret;
+	return ar->promisc || ar->monitor ||
+	       test_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
 }
 
-static int ath10k_vdev_stop(struct ath10k_vif *arvif)
-{
-	struct ath10k *ar = arvif->ar;
-	int ret;
-
-	lockdep_assert_held(&ar->conf_mutex);
-
-	reinit_completion(&ar->vdev_setup_done);
-
-	ret = ath10k_wmi_vdev_stop(ar, arvif->vdev_id);
-	if (ret) {
-		ath10k_warn("WMI vdev %i stop failed: ret %d\n",
-			    arvif->vdev_id, ret);
-		return ret;
-	}
-
-	ret = ath10k_vdev_setup_sync(ar);
-	if (ret) {
-		ath10k_warn("vdev %i setup sync failed %d\n",
-			    arvif->vdev_id, ret);
-		return ret;
-	}
-
-	return ret;
-}
-
-static int ath10k_monitor_start(struct ath10k *ar, int vdev_id)
+static int ath10k_monitor_vdev_start(struct ath10k *ar, int vdev_id)
 {
 	struct cfg80211_chan_def *chandef = &ar->chandef;
 	struct ieee80211_channel *channel = chandef->chan;
@@ -582,11 +514,6 @@ static int ath10k_monitor_start(struct ath10k *ar, int vdev_id)
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	if (!ar->monitor_present) {
-		ath10k_warn("mac montor stop -- monitor is not present\n");
-		return -EINVAL;
-	}
-
 	arg.vdev_id = vdev_id;
 	arg.channel.freq = channel->center_freq;
 	arg.channel.band_center_freq1 = chandef->center_freq1;
@@ -604,88 +531,75 @@ static int ath10k_monitor_start(struct ath10k *ar, int vdev_id)
 
 	ret = ath10k_wmi_vdev_start(ar, &arg);
 	if (ret) {
-		ath10k_warn("Monitor vdev %i start failed: ret %d\n",
+		ath10k_warn("failed to request monitor vdev %i start: %d\n",
 			    vdev_id, ret);
 		return ret;
 	}
 
 	ret = ath10k_vdev_setup_sync(ar);
 	if (ret) {
-		ath10k_warn("Monitor vdev %i setup failed %d\n",
+		ath10k_warn("failed to synchronize setup for monitor vdev %i: %d\n",
 			    vdev_id, ret);
 		return ret;
 	}
 
 	ret = ath10k_wmi_vdev_up(ar, vdev_id, 0, ar->mac_addr);
 	if (ret) {
-		ath10k_warn("Monitor vdev %i up failed: %d\n",
+		ath10k_warn("failed to put up monitor vdev %i: %d\n",
 			    vdev_id, ret);
 		goto vdev_stop;
 	}
 
 	ar->monitor_vdev_id = vdev_id;
-	ar->monitor_enabled = true;
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor vdev %i started\n",
+		   ar->monitor_vdev_id);
 	return 0;
 
 vdev_stop:
 	ret = ath10k_wmi_vdev_stop(ar, ar->monitor_vdev_id);
 	if (ret)
-		ath10k_warn("Monitor vdev %i stop failed: %d\n",
+		ath10k_warn("failed to stop monitor vdev %i after start failure: %d\n",
 			    ar->monitor_vdev_id, ret);
 
 	return ret;
 }
 
-static int ath10k_monitor_stop(struct ath10k *ar)
+static int ath10k_monitor_vdev_stop(struct ath10k *ar)
 {
 	int ret = 0;
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	if (!ar->monitor_present) {
-		ath10k_warn("mac montor stop -- monitor is not present\n");
-		return -EINVAL;
-	}
-
-	if (!ar->monitor_enabled) {
-		ath10k_warn("mac montor stop -- monitor is not enabled\n");
-		return -EINVAL;
-	}
-
 	ret = ath10k_wmi_vdev_down(ar, ar->monitor_vdev_id);
 	if (ret)
-		ath10k_warn("Monitor vdev %i down failed: %d\n",
+		ath10k_warn("failed to put down monitor vdev %i: %d\n",
 			    ar->monitor_vdev_id, ret);
 
 	ret = ath10k_wmi_vdev_stop(ar, ar->monitor_vdev_id);
 	if (ret)
-		ath10k_warn("Monitor vdev %i stop failed: %d\n",
+		ath10k_warn("failed to to request monitor vdev %i stop: %d\n",
 			    ar->monitor_vdev_id, ret);
 
 	ret = ath10k_vdev_setup_sync(ar);
 	if (ret)
-		ath10k_warn("Monitor_down sync failed, vdev %i: %d\n",
+		ath10k_warn("failed to synchronise monitor vdev %i: %d\n",
 			    ar->monitor_vdev_id, ret);
 
-	ar->monitor_enabled = false;
+	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor vdev %i stopped\n",
+		   ar->monitor_vdev_id);
 	return ret;
 }
 
-static int ath10k_monitor_create(struct ath10k *ar)
+static int ath10k_monitor_vdev_create(struct ath10k *ar)
 {
 	int bit, ret = 0;
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	if (ar->monitor_present) {
-		ath10k_warn("Monitor mode already enabled\n");
-		return 0;
-	}
-
 	bit = ffs(ar->free_vdev_map);
 	if (bit == 0) {
-		ath10k_warn("No free VDEV slots\n");
+		ath10k_warn("failed to find free vdev id for monitor vdev\n");
 		return -ENOMEM;
 	}
 
@@ -696,7 +610,7 @@ static int ath10k_monitor_create(struct ath10k *ar)
 				     WMI_VDEV_TYPE_MONITOR,
 				     0, ar->mac_addr);
 	if (ret) {
-		ath10k_warn("WMI vdev %i monitor create failed: ret %d\n",
+		ath10k_warn("failed to request monitor vdev %i creation: %d\n",
 			    ar->monitor_vdev_id, ret);
 		goto vdev_fail;
 	}
@@ -704,7 +618,6 @@ static int ath10k_monitor_create(struct ath10k *ar)
 	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor vdev %d created\n",
 		   ar->monitor_vdev_id);
 
-	ar->monitor_present = true;
 	return 0;
 
 vdev_fail:
@@ -715,48 +628,123 @@ vdev_fail:
 	return ret;
 }
 
-static int ath10k_monitor_destroy(struct ath10k *ar)
+static int ath10k_monitor_vdev_delete(struct ath10k *ar)
 {
 	int ret = 0;
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	if (!ar->monitor_present)
-		return 0;
-
 	ret = ath10k_wmi_vdev_delete(ar, ar->monitor_vdev_id);
 	if (ret) {
-		ath10k_warn("WMI vdev %i monitor delete failed: %d\n",
+		ath10k_warn("failed to request wmi monitor vdev %i removal: %d\n",
 			    ar->monitor_vdev_id, ret);
 		return ret;
 	}
 
 	ar->free_vdev_map |= 1 << (ar->monitor_vdev_id);
-	ar->monitor_present = false;
 
 	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor vdev %d deleted\n",
 		   ar->monitor_vdev_id);
 	return ret;
 }
 
-static int ath10k_start_cac(struct ath10k *ar)
+static int ath10k_monitor_start(struct ath10k *ar)
 {
 	int ret;
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	set_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
+	if (!ath10k_monitor_is_enabled(ar)) {
+		ath10k_warn("trying to start monitor with no references\n");
+		return 0;
+	}
+
+	if (ar->monitor_started) {
+		ath10k_dbg(ATH10K_DBG_MAC, "mac monitor already started\n");
+		return 0;
+	}
 
-	ret = ath10k_monitor_create(ar);
+	ret = ath10k_monitor_vdev_create(ar);
 	if (ret) {
-		clear_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
+		ath10k_warn("failed to create monitor vdev: %d\n", ret);
 		return ret;
 	}
 
-	ret = ath10k_monitor_start(ar, ar->monitor_vdev_id);
+	ret = ath10k_monitor_vdev_start(ar, ar->monitor_vdev_id);
+	if (ret) {
+		ath10k_warn("failed to start monitor vdev: %d\n", ret);
+		ath10k_monitor_vdev_delete(ar);
+		return ret;
+	}
+
+	ar->monitor_started = true;
+	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor started\n");
+
+	return 0;
+}
+
+static void ath10k_monitor_stop(struct ath10k *ar)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (ath10k_monitor_is_enabled(ar)) {
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac monitor will be stopped later\n");
+		return;
+	}
+
+	if (!ar->monitor_started) {
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac monitor probably failed to start earlier\n");
+		return;
+	}
+
+	ret = ath10k_monitor_vdev_stop(ar);
+	if (ret)
+		ath10k_warn("failed to stop monitor vdev: %d\n", ret);
+
+	ret = ath10k_monitor_vdev_delete(ar);
+	if (ret)
+		ath10k_warn("failed to delete monitor vdev: %d\n", ret);
+
+	ar->monitor_started = false;
+	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor stopped\n");
+}
+
+static int ath10k_recalc_rtscts_prot(struct ath10k_vif *arvif)
+{
+	struct ath10k *ar = arvif->ar;
+	u32 vdev_param, rts_cts = 0;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	vdev_param = ar->wmi.vdev_param->enable_rtscts;
+
+	if (arvif->use_cts_prot || arvif->num_legacy_stations > 0)
+		rts_cts |= SM(WMI_RTSCTS_ENABLED, WMI_RTSCTS_SET);
+
+	if (arvif->num_legacy_stations > 0)
+		rts_cts |= SM(WMI_RTSCTS_ACROSS_SW_RETRIES,
+			      WMI_RTSCTS_PROFILE);
+
+	return ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, vdev_param,
+					 rts_cts);
+}
+
+static int ath10k_start_cac(struct ath10k *ar)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	set_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
+
+	ret = ath10k_monitor_start(ar);
 	if (ret) {
+		ath10k_warn("failed to start monitor (cac): %d\n", ret);
 		clear_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
-		ath10k_monitor_destroy(ar);
 		return ret;
 	}
 
@@ -774,58 +762,26 @@ static int ath10k_stop_cac(struct ath10k *ar)
 	if (!test_bit(ATH10K_CAC_RUNNING, &ar->dev_flags))
 		return 0;
 
-	ath10k_monitor_stop(ar);
-	ath10k_monitor_destroy(ar);
 	clear_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
+	ath10k_monitor_stop(ar);
 
 	ath10k_dbg(ATH10K_DBG_MAC, "mac cac finished\n");
 
 	return 0;
 }
 
-static const char *ath10k_dfs_state(enum nl80211_dfs_state dfs_state)
+static void ath10k_recalc_radar_detection(struct ath10k *ar)
 {
-	switch (dfs_state) {
-	case NL80211_DFS_USABLE:
-		return "USABLE";
-	case NL80211_DFS_UNAVAILABLE:
-		return "UNAVAILABLE";
-	case NL80211_DFS_AVAILABLE:
-		return "AVAILABLE";
-	default:
-		WARN_ON(1);
-		return "bug";
-	}
-}
-
-static void ath10k_config_radar_detection(struct ath10k *ar)
-{
-	struct ieee80211_channel *chan = ar->hw->conf.chandef.chan;
-	bool radar = ar->hw->conf.radar_enabled;
-	bool chan_radar = !!(chan->flags & IEEE80211_CHAN_RADAR);
-	enum nl80211_dfs_state dfs_state = chan->dfs_state;
 	int ret;
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	ath10k_dbg(ATH10K_DBG_MAC,
-		   "mac radar config update: chan %dMHz radar %d chan radar %d chan state %s\n",
-		   chan->center_freq, radar, chan_radar,
-		   ath10k_dfs_state(dfs_state));
-
-	/*
-	 * It's safe to call it even if CAC is not started.
-	 * This call here guarantees changing channel, etc. will stop CAC.
-	 */
 	ath10k_stop_cac(ar);
 
-	if (!radar)
-		return;
-
-	if (!chan_radar)
+	if (!ar->radar_enabled)
 		return;
 
-	if (dfs_state != NL80211_DFS_USABLE)
+	if (ar->num_started_vdevs > 0)
 		return;
 
 	ret = ath10k_start_cac(ar);
@@ -835,11 +791,106 @@ static void ath10k_config_radar_detection(struct ath10k *ar)
 		 * radiation is not allowed, make this channel DFS_UNAVAILABLE
 		 * by indicating that radar was detected.
 		 */
-		ath10k_warn("failed to start CAC (%d)\n", ret);
+		ath10k_warn("failed to start CAC: %d\n", ret);
 		ieee80211_radar_detected(ar->hw);
 	}
 }
 
+static int ath10k_vdev_start(struct ath10k_vif *arvif)
+{
+	struct ath10k *ar = arvif->ar;
+	struct cfg80211_chan_def *chandef = &ar->chandef;
+	struct wmi_vdev_start_request_arg arg = {};
+	int ret = 0;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	reinit_completion(&ar->vdev_setup_done);
+
+	arg.vdev_id = arvif->vdev_id;
+	arg.dtim_period = arvif->dtim_period;
+	arg.bcn_intval = arvif->beacon_interval;
+
+	arg.channel.freq = chandef->chan->center_freq;
+	arg.channel.band_center_freq1 = chandef->center_freq1;
+	arg.channel.mode = chan_to_phymode(chandef);
+
+	arg.channel.min_power = 0;
+	arg.channel.max_power = chandef->chan->max_power * 2;
+	arg.channel.max_reg_power = chandef->chan->max_reg_power * 2;
+	arg.channel.max_antenna_gain = chandef->chan->max_antenna_gain * 2;
+
+	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
+		arg.ssid = arvif->u.ap.ssid;
+		arg.ssid_len = arvif->u.ap.ssid_len;
+		arg.hidden_ssid = arvif->u.ap.hidden_ssid;
+
+		/* For now allow DFS for AP mode */
+		arg.channel.chan_radar =
+			!!(chandef->chan->flags & IEEE80211_CHAN_RADAR);
+	} else if (arvif->vdev_type == WMI_VDEV_TYPE_IBSS) {
+		arg.ssid = arvif->vif->bss_conf.ssid;
+		arg.ssid_len = arvif->vif->bss_conf.ssid_len;
+	}
+
+	ath10k_dbg(ATH10K_DBG_MAC,
+		   "mac vdev %d start center_freq %d phymode %s\n",
+		   arg.vdev_id, arg.channel.freq,
+		   ath10k_wmi_phymode_str(arg.channel.mode));
+
+	ret = ath10k_wmi_vdev_start(ar, &arg);
+	if (ret) {
+		ath10k_warn("failed to start WMI vdev %i: %d\n",
+			    arg.vdev_id, ret);
+		return ret;
+	}
+
+	ret = ath10k_vdev_setup_sync(ar);
+	if (ret) {
+		ath10k_warn("failed to synchronise setup for vdev %i: %d\n",
+			    arg.vdev_id, ret);
+		return ret;
+	}
+
+	ar->num_started_vdevs++;
+	ath10k_recalc_radar_detection(ar);
+
+	return ret;
+}
+
+static int ath10k_vdev_stop(struct ath10k_vif *arvif)
+{
+	struct ath10k *ar = arvif->ar;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	reinit_completion(&ar->vdev_setup_done);
+
+	ret = ath10k_wmi_vdev_stop(ar, arvif->vdev_id);
+	if (ret) {
+		ath10k_warn("failed to stop WMI vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	ret = ath10k_vdev_setup_sync(ar);
+	if (ret) {
+		ath10k_warn("failed to syncronise setup for vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	WARN_ON(ar->num_started_vdevs == 0);
+
+	if (ar->num_started_vdevs != 0) {
+		ar->num_started_vdevs--;
+		ath10k_recalc_radar_detection(ar);
+	}
+
+	return ret;
+}
+
 static void ath10k_control_beaconing(struct ath10k_vif *arvif,
 				struct ieee80211_bss_conf *info)
 {
@@ -880,7 +931,7 @@ static void ath10k_control_beaconing(struct ath10k_vif *arvif,
 	ret = ath10k_wmi_vdev_up(arvif->ar, arvif->vdev_id, arvif->aid,
 				 arvif->bssid);
 	if (ret) {
-		ath10k_warn("Failed to bring up vdev %d: %i\n",
+		ath10k_warn("failed to bring up vdev %d: %i\n",
 			    arvif->vdev_id, ret);
 		ath10k_vdev_stop(arvif);
 		return;
@@ -904,7 +955,7 @@ static void ath10k_control_ibss(struct ath10k_vif *arvif,
 	if (!info->ibss_joined) {
 		ret = ath10k_peer_delete(arvif->ar, arvif->vdev_id, self_peer);
 		if (ret)
-			ath10k_warn("Failed to delete IBSS self peer:%pM for VDEV:%d ret:%d\n",
+			ath10k_warn("failed to delete IBSS self peer %pM for vdev %d: %d\n",
 				    self_peer, arvif->vdev_id, ret);
 
 		if (is_zero_ether_addr(arvif->bssid))
@@ -913,7 +964,7 @@ static void ath10k_control_ibss(struct ath10k_vif *arvif,
 		ret = ath10k_peer_delete(arvif->ar, arvif->vdev_id,
 					 arvif->bssid);
 		if (ret) {
-			ath10k_warn("Failed to delete IBSS BSSID peer:%pM for VDEV:%d ret:%d\n",
+			ath10k_warn("failed to delete IBSS BSSID peer %pM for vdev %d: %d\n",
 				    arvif->bssid, arvif->vdev_id, ret);
 			return;
 		}
@@ -925,7 +976,7 @@ static void ath10k_control_ibss(struct ath10k_vif *arvif,
 
 	ret = ath10k_peer_create(arvif->ar, arvif->vdev_id, self_peer);
 	if (ret) {
-		ath10k_warn("Failed to create IBSS self peer:%pM for VDEV:%d ret:%d\n",
+		ath10k_warn("failed to create IBSS self peer %pM for vdev %d: %d\n",
 			    self_peer, arvif->vdev_id, ret);
 		return;
 	}
@@ -934,7 +985,7 @@ static void ath10k_control_ibss(struct ath10k_vif *arvif,
 	ret = ath10k_wmi_vdev_set_param(arvif->ar, arvif->vdev_id, vdev_param,
 					ATH10K_DEFAULT_ATIM);
 	if (ret)
-		ath10k_warn("Failed to set IBSS ATIM for VDEV:%d ret:%d\n",
+		ath10k_warn("failed to set IBSS ATIM for vdev %d: %d\n",
 			    arvif->vdev_id, ret);
 }
 
@@ -961,7 +1012,7 @@ static int ath10k_mac_vif_setup_ps(struct ath10k_vif *arvif)
 		ret = ath10k_wmi_set_sta_ps_param(ar, arvif->vdev_id, param,
 						  conf->dynamic_ps_timeout);
 		if (ret) {
-			ath10k_warn("Failed to set inactivity time for vdev %d: %i\n",
+			ath10k_warn("failed to set inactivity time for vdev %d: %i\n",
 				    arvif->vdev_id, ret);
 			return ret;
 		}
@@ -974,8 +1025,8 @@ static int ath10k_mac_vif_setup_ps(struct ath10k_vif *arvif)
 
 	ret = ath10k_wmi_set_psmode(ar, arvif->vdev_id, psmode);
 	if (ret) {
-		ath10k_warn("Failed to set PS Mode: %d for VDEV: %d\n",
-			    psmode, arvif->vdev_id);
+		ath10k_warn("failed to set PS Mode %d for vdev %d: %d\n",
+			    psmode, arvif->vdev_id, ret);
 		return ret;
 	}
 
@@ -1429,7 +1480,7 @@ static void ath10k_bss_assoc(struct ieee80211_hw *hw,
 
 	ap_sta = ieee80211_find_sta(vif, bss_conf->bssid);
 	if (!ap_sta) {
-		ath10k_warn("Failed to find station entry for %pM, vdev %i\n",
+		ath10k_warn("failed to find station entry for bss %pM vdev %i\n",
 			    bss_conf->bssid, arvif->vdev_id);
 		rcu_read_unlock();
 		return;
@@ -1442,7 +1493,7 @@ static void ath10k_bss_assoc(struct ieee80211_hw *hw,
 	ret = ath10k_peer_assoc_prepare(ar, arvif, ap_sta,
 					bss_conf, &peer_arg);
 	if (ret) {
-		ath10k_warn("Peer assoc prepare failed for %pM vdev %i\n: %d",
+		ath10k_warn("failed to prepare peer assoc for %pM vdev %i: %d\n",
 			    bss_conf->bssid, arvif->vdev_id, ret);
 		rcu_read_unlock();
 		return;
@@ -1452,7 +1503,7 @@ static void ath10k_bss_assoc(struct ieee80211_hw *hw,
 
 	ret = ath10k_wmi_peer_assoc(ar, &peer_arg);
 	if (ret) {
-		ath10k_warn("Peer assoc failed for %pM vdev %i\n: %d",
+		ath10k_warn("failed to run peer assoc for %pM vdev %i: %d\n",
 			    bss_conf->bssid, arvif->vdev_id, ret);
 		return;
 	}
@@ -1473,7 +1524,7 @@ static void ath10k_bss_assoc(struct ieee80211_hw *hw,
 
 	ret = ath10k_wmi_vdev_up(ar, arvif->vdev_id, arvif->aid, arvif->bssid);
 	if (ret) {
-		ath10k_warn("VDEV: %d up failed: ret %d\n",
+		ath10k_warn("failed to set vdev %d up: %d\n",
 			    arvif->vdev_id, ret);
 		return;
 	}
@@ -1524,7 +1575,7 @@ static void ath10k_bss_disassoc(struct ieee80211_hw *hw,
 }
 
 static int ath10k_station_assoc(struct ath10k *ar, struct ath10k_vif *arvif,
-				struct ieee80211_sta *sta)
+				struct ieee80211_sta *sta, bool reassoc)
 {
 	struct wmi_peer_assoc_complete_arg peer_arg;
 	int ret = 0;
@@ -1533,34 +1584,46 @@ static int ath10k_station_assoc(struct ath10k *ar, struct ath10k_vif *arvif,
 
 	ret = ath10k_peer_assoc_prepare(ar, arvif, sta, NULL, &peer_arg);
 	if (ret) {
-		ath10k_warn("WMI peer assoc prepare failed for %pM vdev %i: %i\n",
+		ath10k_warn("failed to prepare WMI peer assoc for %pM vdev %i: %i\n",
 			    sta->addr, arvif->vdev_id, ret);
 		return ret;
 	}
 
+	peer_arg.peer_reassoc = reassoc;
 	ret = ath10k_wmi_peer_assoc(ar, &peer_arg);
 	if (ret) {
-		ath10k_warn("Peer assoc failed for STA %pM vdev %i: %d\n",
+		ath10k_warn("failed to run peer assoc for STA %pM vdev %i: %d\n",
 			    sta->addr, arvif->vdev_id, ret);
 		return ret;
 	}
 
 	ret = ath10k_setup_peer_smps(ar, arvif, sta->addr, &sta->ht_cap);
 	if (ret) {
-		ath10k_warn("failed to setup peer SMPS for vdev: %d\n", ret);
+		ath10k_warn("failed to setup peer SMPS for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
 		return ret;
 	}
 
+	if (!sta->wme) {
+		arvif->num_legacy_stations++;
+		ret  = ath10k_recalc_rtscts_prot(arvif);
+		if (ret) {
+			ath10k_warn("failed to recalculate rts/cts prot for vdev %d: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
 	ret = ath10k_install_peer_wep_keys(arvif, sta->addr);
 	if (ret) {
-		ath10k_warn("could not install peer wep keys for vdev %i: %d\n",
+		ath10k_warn("failed to install peer wep keys for vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		return ret;
 	}
 
 	ret = ath10k_peer_assoc_qos_ap(ar, arvif, sta);
 	if (ret) {
-		ath10k_warn("could not set qos params for STA %pM for vdev %i: %d\n",
+		ath10k_warn("failed to set qos params for STA %pM for vdev %i: %d\n",
 			    sta->addr, arvif->vdev_id, ret);
 		return ret;
 	}
@@ -1575,9 +1638,19 @@ static int ath10k_station_disassoc(struct ath10k *ar, struct ath10k_vif *arvif,
 
 	lockdep_assert_held(&ar->conf_mutex);
 
+	if (!sta->wme) {
+		arvif->num_legacy_stations--;
+		ret = ath10k_recalc_rtscts_prot(arvif);
+		if (ret) {
+			ath10k_warn("failed to recalculate rts/cts prot for vdev %d: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
 	ret = ath10k_clear_peer_keys(arvif, sta->addr);
 	if (ret) {
-		ath10k_warn("could not clear all peer wep keys for vdev %i: %d\n",
+		ath10k_warn("failed to clear all peer wep keys for vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		return ret;
 	}
@@ -1685,19 +1758,44 @@ static int ath10k_update_channel_list(struct ath10k *ar)
 	return ret;
 }
 
+static enum wmi_dfs_region
+ath10k_mac_get_dfs_region(enum nl80211_dfs_regions dfs_region)
+{
+	switch (dfs_region) {
+	case NL80211_DFS_UNSET:
+		return WMI_UNINIT_DFS_DOMAIN;
+	case NL80211_DFS_FCC:
+		return WMI_FCC_DFS_DOMAIN;
+	case NL80211_DFS_ETSI:
+		return WMI_ETSI_DFS_DOMAIN;
+	case NL80211_DFS_JP:
+		return WMI_MKK4_DFS_DOMAIN;
+	}
+	return WMI_UNINIT_DFS_DOMAIN;
+}
+
 static void ath10k_regd_update(struct ath10k *ar)
 {
 	struct reg_dmn_pair_mapping *regpair;
 	int ret;
+	enum wmi_dfs_region wmi_dfs_reg;
+	enum nl80211_dfs_regions nl_dfs_reg;
 
 	lockdep_assert_held(&ar->conf_mutex);
 
 	ret = ath10k_update_channel_list(ar);
 	if (ret)
-		ath10k_warn("could not update channel list (%d)\n", ret);
+		ath10k_warn("failed to update channel list: %d\n", ret);
 
 	regpair = ar->ath_common.regulatory.regpair;
 
+	if (config_enabled(CONFIG_ATH10K_DFS_CERTIFIED) && ar->dfs_detector) {
+		nl_dfs_reg = ar->dfs_detector->region;
+		wmi_dfs_reg = ath10k_mac_get_dfs_region(nl_dfs_reg);
+	} else {
+		wmi_dfs_reg = WMI_UNINIT_DFS_DOMAIN;
+	}
+
 	/* Target allows setting up per-band regdomain but ath_common provides
 	 * a combined one only */
 	ret = ath10k_wmi_pdev_set_regdomain(ar,
@@ -1705,9 +1803,10 @@ static void ath10k_regd_update(struct ath10k *ar)
 					    regpair->reg_domain, /* 2ghz */
 					    regpair->reg_domain, /* 5ghz */
 					    regpair->reg_2ghz_ctl,
-					    regpair->reg_5ghz_ctl);
+					    regpair->reg_5ghz_ctl,
+					    wmi_dfs_reg);
 	if (ret)
-		ath10k_warn("could not set pdev regdomain (%d)\n", ret);
+		ath10k_warn("failed to set pdev regdomain: %d\n", ret);
 }
 
 static void ath10k_reg_notifier(struct wiphy *wiphy,
@@ -1725,7 +1824,7 @@ static void ath10k_reg_notifier(struct wiphy *wiphy,
 		result = ar->dfs_detector->set_dfs_domain(ar->dfs_detector,
 							  request->dfs_region);
 		if (!result)
-			ath10k_warn("dfs region 0x%X not supported, will trigger radar for every pulse\n",
+			ath10k_warn("DFS region 0x%X not supported, will trigger radar for every pulse\n",
 				    request->dfs_region);
 	}
 
@@ -1759,10 +1858,10 @@ static u8 ath10k_tx_h_get_vdev_id(struct ath10k *ar,
 	if (info->control.vif)
 		return ath10k_vif_to_arvif(info->control.vif)->vdev_id;
 
-	if (ar->monitor_enabled)
+	if (ar->monitor_started)
 		return ar->monitor_vdev_id;
 
-	ath10k_warn("could not resolve vdev id\n");
+	ath10k_warn("failed to resolve vdev id\n");
 	return 0;
 }
 
@@ -1792,8 +1891,13 @@ static void ath10k_tx_wep_key_work(struct work_struct *work)
 						wep_key_work);
 	int ret, keyidx = arvif->def_wep_key_newidx;
 
+	mutex_lock(&arvif->ar->conf_mutex);
+
+	if (arvif->ar->state != ATH10K_STATE_ON)
+		goto unlock;
+
 	if (arvif->def_wep_key_idx == keyidx)
-		return;
+		goto unlock;
 
 	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d set keyidx %d\n",
 		   arvif->vdev_id, keyidx);
@@ -1803,11 +1907,16 @@ static void ath10k_tx_wep_key_work(struct work_struct *work)
 					arvif->ar->wmi.vdev_param->def_keyid,
 					keyidx);
 	if (ret) {
-		ath10k_warn("could not update wep keyidx (%d)\n", ret);
-		return;
+		ath10k_warn("failed to update wep key index for vdev %d: %d\n",
+			    arvif->vdev_id,
+			    ret);
+		goto unlock;
 	}
 
 	arvif->def_wep_key_idx = keyidx;
+
+unlock:
+	mutex_unlock(&arvif->ar->conf_mutex);
 }
 
 static void ath10k_tx_h_update_wep_key(struct sk_buff *skb)
@@ -1879,7 +1988,7 @@ static void ath10k_tx_htt(struct ath10k *ar, struct sk_buff *skb)
 			     ar->fw_features)) {
 			if (skb_queue_len(&ar->wmi_mgmt_tx_queue) >=
 			    ATH10K_MAX_NUM_MGMT_PENDING) {
-				ath10k_warn("wmi mgmt_tx queue limit reached\n");
+				ath10k_warn("reached WMI management tranmist queue limit\n");
 				ret = -EBUSY;
 				goto exit;
 			}
@@ -1903,7 +2012,7 @@ static void ath10k_tx_htt(struct ath10k *ar, struct sk_buff *skb)
 
 exit:
 	if (ret) {
-		ath10k_warn("tx failed (%d). dropping packet.\n", ret);
+		ath10k_warn("failed to transmit packet, dropping: %d\n", ret);
 		ieee80211_free_txskb(ar->hw, skb);
 	}
 }
@@ -1964,7 +2073,7 @@ void ath10k_offchan_tx_work(struct work_struct *work)
 		if (!peer) {
 			ret = ath10k_peer_create(ar, vdev_id, peer_addr);
 			if (ret)
-				ath10k_warn("peer %pM on vdev %d not created (%d)\n",
+				ath10k_warn("failed to create peer %pM on vdev %d: %d\n",
 					    peer_addr, vdev_id, ret);
 		}
 
@@ -1984,7 +2093,7 @@ void ath10k_offchan_tx_work(struct work_struct *work)
 		if (!peer) {
 			ret = ath10k_peer_delete(ar, vdev_id, peer_addr);
 			if (ret)
-				ath10k_warn("peer %pM on vdev %d not deleted (%d)\n",
+				ath10k_warn("failed to delete peer %pM on vdev %d: %d\n",
 					    peer_addr, vdev_id, ret);
 		}
 
@@ -2018,7 +2127,8 @@ void ath10k_mgmt_over_wmi_tx_work(struct work_struct *work)
 
 		ret = ath10k_wmi_mgmt_tx(ar, skb);
 		if (ret) {
-			ath10k_warn("wmi mgmt_tx failed (%d)\n", ret);
+			ath10k_warn("failed to transmit management frame via WMI: %d\n",
+				    ret);
 			ieee80211_free_txskb(ar->hw, skb);
 		}
 	}
@@ -2043,7 +2153,7 @@ void ath10k_reset_scan(unsigned long ptr)
 		return;
 	}
 
-	ath10k_warn("scan timeout. resetting. fw issue?\n");
+	ath10k_warn("scan timed out, firmware problem?\n");
 
 	if (ar->scan.is_roc)
 		ieee80211_remain_on_channel_expired(ar->hw);
@@ -2079,7 +2189,7 @@ static int ath10k_abort_scan(struct ath10k *ar)
 
 	ret = ath10k_wmi_stop_scan(ar, &arg);
 	if (ret) {
-		ath10k_warn("could not submit wmi stop scan (%d)\n", ret);
+		ath10k_warn("failed to stop wmi scan: %d\n", ret);
 		spin_lock_bh(&ar->data_lock);
 		ar->scan.in_progress = false;
 		ath10k_offchan_tx_purge(ar);
@@ -2099,7 +2209,7 @@ static int ath10k_abort_scan(struct ath10k *ar)
 
 	spin_lock_bh(&ar->data_lock);
 	if (ar->scan.in_progress) {
-		ath10k_warn("could not stop scan. its still in progress\n");
+		ath10k_warn("failed to stop scan, it's still in progress\n");
 		ar->scan.in_progress = false;
 		ath10k_offchan_tx_purge(ar);
 		ret = -ETIMEDOUT;
@@ -2187,72 +2297,171 @@ static void ath10k_tx(struct ieee80211_hw *hw,
 	ath10k_tx_htt(ar, skb);
 }
 
-/*
- * Initialize various parameters with default vaules.
- */
+/* Must not be called with conf_mutex held as workers can use that also. */
+static void ath10k_drain_tx(struct ath10k *ar)
+{
+	/* make sure rcu-protected mac80211 tx path itself is drained */
+	synchronize_net();
+
+	ath10k_offchan_tx_purge(ar);
+	ath10k_mgmt_over_wmi_tx_purge(ar);
+
+	cancel_work_sync(&ar->offchan_tx_work);
+	cancel_work_sync(&ar->wmi_mgmt_tx_work);
+}
+
 void ath10k_halt(struct ath10k *ar)
 {
+	struct ath10k_vif *arvif;
+
 	lockdep_assert_held(&ar->conf_mutex);
 
-	ath10k_stop_cac(ar);
+	if (ath10k_monitor_is_enabled(ar)) {
+		clear_bit(ATH10K_CAC_RUNNING, &ar->dev_flags);
+		ar->promisc = false;
+		ar->monitor = false;
+		ath10k_monitor_stop(ar);
+	}
+
 	del_timer_sync(&ar->scan.timeout);
-	ath10k_offchan_tx_purge(ar);
-	ath10k_mgmt_over_wmi_tx_purge(ar);
+	ath10k_reset_scan((unsigned long)ar);
 	ath10k_peer_cleanup_all(ar);
 	ath10k_core_stop(ar);
 	ath10k_hif_power_down(ar);
 
 	spin_lock_bh(&ar->data_lock);
-	if (ar->scan.in_progress) {
-		del_timer(&ar->scan.timeout);
-		ar->scan.in_progress = false;
-		ieee80211_scan_completed(ar->hw, true);
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (!arvif->beacon)
+			continue;
+
+		dma_unmap_single(arvif->ar->dev,
+				 ATH10K_SKB_CB(arvif->beacon)->paddr,
+				 arvif->beacon->len, DMA_TO_DEVICE);
+		dev_kfree_skb_any(arvif->beacon);
+		arvif->beacon = NULL;
 	}
 	spin_unlock_bh(&ar->data_lock);
 }
 
+static int ath10k_get_antenna(struct ieee80211_hw *hw, u32 *tx_ant, u32 *rx_ant)
+{
+	struct ath10k *ar = hw->priv;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->cfg_tx_chainmask) {
+		*tx_ant = ar->cfg_tx_chainmask;
+		*rx_ant = ar->cfg_rx_chainmask;
+	} else {
+		*tx_ant = ar->supp_tx_chainmask;
+		*rx_ant = ar->supp_rx_chainmask;
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+}
+
+static int __ath10k_set_antenna(struct ath10k *ar, u32 tx_ant, u32 rx_ant)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ar->cfg_tx_chainmask = tx_ant;
+	ar->cfg_rx_chainmask = rx_ant;
+
+	if ((ar->state != ATH10K_STATE_ON) &&
+	    (ar->state != ATH10K_STATE_RESTARTED))
+		return 0;
+
+	ret = ath10k_wmi_pdev_set_param(ar, ar->wmi.pdev_param->tx_chain_mask,
+					tx_ant);
+	if (ret) {
+		ath10k_warn("failed to set tx-chainmask: %d, req 0x%x\n",
+			    ret, tx_ant);
+		return ret;
+	}
+
+	ret = ath10k_wmi_pdev_set_param(ar, ar->wmi.pdev_param->rx_chain_mask,
+					rx_ant);
+	if (ret) {
+		ath10k_warn("failed to set rx-chainmask: %d, req 0x%x\n",
+			    ret, rx_ant);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int ath10k_set_antenna(struct ieee80211_hw *hw, u32 tx_ant, u32 rx_ant)
+{
+	struct ath10k *ar = hw->priv;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+	ret = __ath10k_set_antenna(ar, tx_ant, rx_ant);
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
 static int ath10k_start(struct ieee80211_hw *hw)
 {
 	struct ath10k *ar = hw->priv;
 	int ret = 0;
 
+	/*
+	 * This makes sense only when restarting hw. It is harmless to call
+	 * uncoditionally. This is necessary to make sure no HTT/WMI tx
+	 * commands will be submitted while restarting.
+	 */
+	ath10k_drain_tx(ar);
+
 	mutex_lock(&ar->conf_mutex);
 
-	if (ar->state != ATH10K_STATE_OFF &&
-	    ar->state != ATH10K_STATE_RESTARTING) {
+	switch (ar->state) {
+	case ATH10K_STATE_OFF:
+		ar->state = ATH10K_STATE_ON;
+		break;
+	case ATH10K_STATE_RESTARTING:
+		ath10k_halt(ar);
+		ar->state = ATH10K_STATE_RESTARTED;
+		break;
+	case ATH10K_STATE_ON:
+	case ATH10K_STATE_RESTARTED:
+	case ATH10K_STATE_WEDGED:
+		WARN_ON(1);
 		ret = -EINVAL;
-		goto exit;
+		goto err;
 	}
 
 	ret = ath10k_hif_power_up(ar);
 	if (ret) {
-		ath10k_err("could not init hif (%d)\n", ret);
-		ar->state = ATH10K_STATE_OFF;
-		goto exit;
+		ath10k_err("Could not init hif: %d\n", ret);
+		goto err_off;
 	}
 
 	ret = ath10k_core_start(ar);
 	if (ret) {
-		ath10k_err("could not init core (%d)\n", ret);
-		ath10k_hif_power_down(ar);
-		ar->state = ATH10K_STATE_OFF;
-		goto exit;
+		ath10k_err("Could not init core: %d\n", ret);
+		goto err_power_down;
 	}
 
-	if (ar->state == ATH10K_STATE_OFF)
-		ar->state = ATH10K_STATE_ON;
-	else if (ar->state == ATH10K_STATE_RESTARTING)
-		ar->state = ATH10K_STATE_RESTARTED;
-
 	ret = ath10k_wmi_pdev_set_param(ar, ar->wmi.pdev_param->pmf_qos, 1);
-	if (ret)
-		ath10k_warn("could not enable WMI_PDEV_PARAM_PMF_QOS (%d)\n",
-			    ret);
+	if (ret) {
+		ath10k_warn("failed to enable PMF QOS: %d\n", ret);
+		goto err_core_stop;
+	}
 
 	ret = ath10k_wmi_pdev_set_param(ar, ar->wmi.pdev_param->dynamic_bw, 1);
-	if (ret)
-		ath10k_warn("could not init WMI_PDEV_PARAM_DYNAMIC_BW (%d)\n",
-			    ret);
+	if (ret) {
+		ath10k_warn("failed to enable dynamic BW: %d\n", ret);
+		goto err_core_stop;
+	}
+
+	if (ar->cfg_tx_chainmask)
+		__ath10k_set_antenna(ar, ar->cfg_tx_chainmask,
+				     ar->cfg_rx_chainmask);
 
 	/*
 	 * By default FW set ARP frames ac to voice (6). In that case ARP
@@ -2266,15 +2475,27 @@ static int ath10k_start(struct ieee80211_hw *hw)
 	ret = ath10k_wmi_pdev_set_param(ar,
 					ar->wmi.pdev_param->arp_ac_override, 0);
 	if (ret) {
-		ath10k_warn("could not set arp ac override parameter: %d\n",
+		ath10k_warn("failed to set arp ac override parameter: %d\n",
 			    ret);
-		goto exit;
+		goto err_core_stop;
 	}
 
+	ar->num_started_vdevs = 0;
 	ath10k_regd_update(ar);
-	ret = 0;
 
-exit:
+	mutex_unlock(&ar->conf_mutex);
+	return 0;
+
+err_core_stop:
+	ath10k_core_stop(ar);
+
+err_power_down:
+	ath10k_hif_power_down(ar);
+
+err_off:
+	ar->state = ATH10K_STATE_OFF;
+
+err:
 	mutex_unlock(&ar->conf_mutex);
 	return ret;
 }
@@ -2283,19 +2504,15 @@ static void ath10k_stop(struct ieee80211_hw *hw)
 {
 	struct ath10k *ar = hw->priv;
 
+	ath10k_drain_tx(ar);
+
 	mutex_lock(&ar->conf_mutex);
-	if (ar->state == ATH10K_STATE_ON ||
-	    ar->state == ATH10K_STATE_RESTARTED ||
-	    ar->state == ATH10K_STATE_WEDGED)
+	if (ar->state != ATH10K_STATE_OFF) {
 		ath10k_halt(ar);
-
-	ar->state = ATH10K_STATE_OFF;
+		ar->state = ATH10K_STATE_OFF;
+	}
 	mutex_unlock(&ar->conf_mutex);
 
-	ath10k_mgmt_over_wmi_tx_purge(ar);
-
-	cancel_work_sync(&ar->offchan_tx_work);
-	cancel_work_sync(&ar->wmi_mgmt_tx_work);
 	cancel_work_sync(&ar->restart_work);
 }
 
@@ -2309,7 +2526,7 @@ static int ath10k_config_ps(struct ath10k *ar)
 	list_for_each_entry(arvif, &ar->arvifs, list) {
 		ret = ath10k_mac_vif_setup_ps(arvif);
 		if (ret) {
-			ath10k_warn("could not setup powersave (%d)\n", ret);
+			ath10k_warn("failed to setup powersave: %d\n", ret);
 			break;
 		}
 	}
@@ -2343,7 +2560,6 @@ static const char *chandef_get_width(enum nl80211_chan_width width)
 static void ath10k_config_chan(struct ath10k *ar)
 {
 	struct ath10k_vif *arvif;
-	bool monitor_was_enabled;
 	int ret;
 
 	lockdep_assert_held(&ar->conf_mutex);
@@ -2357,10 +2573,8 @@ static void ath10k_config_chan(struct ath10k *ar)
 
 	/* First stop monitor interface. Some FW versions crash if there's a
 	 * lone monitor interface. */
-	monitor_was_enabled = ar->monitor_enabled;
-
-	if (ar->monitor_enabled)
-		ath10k_monitor_stop(ar);
+	if (ar->monitor_started)
+		ath10k_monitor_vdev_stop(ar);
 
 	list_for_each_entry(arvif, &ar->arvifs, list) {
 		if (!arvif->is_started)
@@ -2371,7 +2585,7 @@ static void ath10k_config_chan(struct ath10k *ar)
 
 		ret = ath10k_vdev_stop(arvif);
 		if (ret) {
-			ath10k_warn("could not stop vdev %d (%d)\n",
+			ath10k_warn("failed to stop vdev %d: %d\n",
 				    arvif->vdev_id, ret);
 			continue;
 		}
@@ -2388,7 +2602,7 @@ static void ath10k_config_chan(struct ath10k *ar)
 
 		ret = ath10k_vdev_start(arvif);
 		if (ret) {
-			ath10k_warn("could not start vdev %d (%d)\n",
+			ath10k_warn("failed to start vdev %d: %d\n",
 				    arvif->vdev_id, ret);
 			continue;
 		}
@@ -2399,14 +2613,14 @@ static void ath10k_config_chan(struct ath10k *ar)
 		ret = ath10k_wmi_vdev_up(arvif->ar, arvif->vdev_id, arvif->aid,
 					 arvif->bssid);
 		if (ret) {
-			ath10k_warn("could not bring vdev up %d (%d)\n",
+			ath10k_warn("failed to bring vdev up %d: %d\n",
 				    arvif->vdev_id, ret);
 			continue;
 		}
 	}
 
-	if (monitor_was_enabled)
-		ath10k_monitor_start(ar, ar->monitor_vdev_id);
+	if (ath10k_monitor_is_enabled(ar))
+		ath10k_monitor_vdev_start(ar, ar->monitor_vdev_id);
 }
 
 static int ath10k_config(struct ieee80211_hw *hw, u32 changed)
@@ -2420,15 +2634,17 @@ static int ath10k_config(struct ieee80211_hw *hw, u32 changed)
 
 	if (changed & IEEE80211_CONF_CHANGE_CHANNEL) {
 		ath10k_dbg(ATH10K_DBG_MAC,
-			   "mac config channel %d mhz flags 0x%x\n",
+			   "mac config channel %dMHz flags 0x%x radar %d\n",
 			   conf->chandef.chan->center_freq,
-			   conf->chandef.chan->flags);
+			   conf->chandef.chan->flags,
+			   conf->radar_enabled);
 
 		spin_lock_bh(&ar->data_lock);
 		ar->rx_channel = conf->chandef.chan;
 		spin_unlock_bh(&ar->data_lock);
 
-		ath10k_config_radar_detection(ar);
+		ar->radar_enabled = conf->radar_enabled;
+		ath10k_recalc_radar_detection(ar);
 
 		if (!cfg80211_chandef_identical(&ar->chandef, &conf->chandef)) {
 			ar->chandef = conf->chandef;
@@ -2444,14 +2660,14 @@ static int ath10k_config(struct ieee80211_hw *hw, u32 changed)
 		ret = ath10k_wmi_pdev_set_param(ar, param,
 						hw->conf.power_level * 2);
 		if (ret)
-			ath10k_warn("mac failed to set 2g txpower %d (%d)\n",
+			ath10k_warn("failed to set 2g txpower %d: %d\n",
 				    hw->conf.power_level, ret);
 
 		param = ar->wmi.pdev_param->txpower_limit5g;
 		ret = ath10k_wmi_pdev_set_param(ar, param,
 						hw->conf.power_level * 2);
 		if (ret)
-			ath10k_warn("mac failed to set 5g txpower %d (%d)\n",
+			ath10k_warn("failed to set 5g txpower %d: %d\n",
 				    hw->conf.power_level, ret);
 	}
 
@@ -2459,10 +2675,19 @@ static int ath10k_config(struct ieee80211_hw *hw, u32 changed)
 		ath10k_config_ps(ar);
 
 	if (changed & IEEE80211_CONF_CHANGE_MONITOR) {
-		if (conf->flags & IEEE80211_CONF_MONITOR)
-			ret = ath10k_monitor_create(ar);
-		else
-			ret = ath10k_monitor_destroy(ar);
+		if (conf->flags & IEEE80211_CONF_MONITOR && !ar->monitor) {
+			ar->monitor = true;
+			ret = ath10k_monitor_start(ar);
+			if (ret) {
+				ath10k_warn("failed to start monitor (config): %d\n",
+					    ret);
+				ar->monitor = false;
+			}
+		} else if (!(conf->flags & IEEE80211_CONF_MONITOR) &&
+			   ar->monitor) {
+			ar->monitor = false;
+			ath10k_monitor_stop(ar);
+		}
 	}
 
 	mutex_unlock(&ar->conf_mutex);
@@ -2497,12 +2722,6 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 	INIT_WORK(&arvif->wep_key_work, ath10k_tx_wep_key_work);
 	INIT_LIST_HEAD(&arvif->list);
 
-	if ((vif->type == NL80211_IFTYPE_MONITOR) && ar->monitor_present) {
-		ath10k_warn("Only one monitor interface allowed\n");
-		ret = -EBUSY;
-		goto err;
-	}
-
 	bit = ffs(ar->free_vdev_map);
 	if (bit == 0) {
 		ret = -EBUSY;
@@ -2545,7 +2764,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 	ret = ath10k_wmi_vdev_create(ar, arvif->vdev_id, arvif->vdev_type,
 				     arvif->vdev_subtype, vif->addr);
 	if (ret) {
-		ath10k_warn("WMI vdev %i create failed: ret %d\n",
+		ath10k_warn("failed to create WMI vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 		goto err;
 	}
@@ -2557,7 +2776,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 	ret = ath10k_wmi_vdev_set_param(ar, 0, vdev_param,
 					arvif->def_wep_key_idx);
 	if (ret) {
-		ath10k_warn("Failed to set vdev %i default keyid: %d\n",
+		ath10k_warn("failed to set vdev %i default key id: %d\n",
 			    arvif->vdev_id, ret);
 		goto err_vdev_delete;
 	}
@@ -2567,7 +2786,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 					ATH10K_HW_TXRX_NATIVE_WIFI);
 	/* 10.X firmware does not support this VDEV parameter. Do not warn */
 	if (ret && ret != -EOPNOTSUPP) {
-		ath10k_warn("Failed to set vdev %i TX encap: %d\n",
+		ath10k_warn("failed to set vdev %i TX encapsulation: %d\n",
 			    arvif->vdev_id, ret);
 		goto err_vdev_delete;
 	}
@@ -2575,14 +2794,14 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
 		ret = ath10k_peer_create(ar, arvif->vdev_id, vif->addr);
 		if (ret) {
-			ath10k_warn("Failed to create vdev %i peer for AP: %d\n",
+			ath10k_warn("failed to create vdev %i peer for AP: %d\n",
 				    arvif->vdev_id, ret);
 			goto err_vdev_delete;
 		}
 
 		ret = ath10k_mac_set_kickout(arvif);
 		if (ret) {
-			ath10k_warn("Failed to set vdev %i kickout parameters: %d\n",
+			ath10k_warn("failed to set vdev %i kickout parameters: %d\n",
 				    arvif->vdev_id, ret);
 			goto err_peer_delete;
 		}
@@ -2594,7 +2813,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
 						  param, value);
 		if (ret) {
-			ath10k_warn("Failed to set vdev %i RX wake policy: %d\n",
+			ath10k_warn("failed to set vdev %i RX wake policy: %d\n",
 				    arvif->vdev_id, ret);
 			goto err_peer_delete;
 		}
@@ -2604,7 +2823,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
 						  param, value);
 		if (ret) {
-			ath10k_warn("Failed to set vdev %i TX wake thresh: %d\n",
+			ath10k_warn("failed to set vdev %i TX wake thresh: %d\n",
 				    arvif->vdev_id, ret);
 			goto err_peer_delete;
 		}
@@ -2614,7 +2833,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
 						  param, value);
 		if (ret) {
-			ath10k_warn("Failed to set vdev %i PSPOLL count: %d\n",
+			ath10k_warn("failed to set vdev %i PSPOLL count: %d\n",
 				    arvif->vdev_id, ret);
 			goto err_peer_delete;
 		}
@@ -2622,21 +2841,18 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 
 	ret = ath10k_mac_set_rts(arvif, ar->hw->wiphy->rts_threshold);
 	if (ret) {
-		ath10k_warn("failed to set rts threshold for vdev %d (%d)\n",
+		ath10k_warn("failed to set rts threshold for vdev %d: %d\n",
 			    arvif->vdev_id, ret);
 		goto err_peer_delete;
 	}
 
 	ret = ath10k_mac_set_frag(arvif, ar->hw->wiphy->frag_threshold);
 	if (ret) {
-		ath10k_warn("failed to set frag threshold for vdev %d (%d)\n",
+		ath10k_warn("failed to set frag threshold for vdev %d: %d\n",
 			    arvif->vdev_id, ret);
 		goto err_peer_delete;
 	}
 
-	if (arvif->vdev_type == WMI_VDEV_TYPE_MONITOR)
-		ar->monitor_present = true;
-
 	mutex_unlock(&ar->conf_mutex);
 	return 0;
 
@@ -2668,6 +2884,9 @@ static void ath10k_remove_interface(struct ieee80211_hw *hw,
 
 	spin_lock_bh(&ar->data_lock);
 	if (arvif->beacon) {
+		dma_unmap_single(arvif->ar->dev,
+				 ATH10K_SKB_CB(arvif->beacon)->paddr,
+				 arvif->beacon->len, DMA_TO_DEVICE);
 		dev_kfree_skb_any(arvif->beacon);
 		arvif->beacon = NULL;
 	}
@@ -2679,7 +2898,7 @@ static void ath10k_remove_interface(struct ieee80211_hw *hw,
 	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
 		ret = ath10k_peer_delete(arvif->ar, arvif->vdev_id, vif->addr);
 		if (ret)
-			ath10k_warn("Failed to remove peer for AP vdev %i: %d\n",
+			ath10k_warn("failed to remove peer for AP vdev %i: %d\n",
 				    arvif->vdev_id, ret);
 
 		kfree(arvif->u.ap.noa_data);
@@ -2690,12 +2909,9 @@ static void ath10k_remove_interface(struct ieee80211_hw *hw,
 
 	ret = ath10k_wmi_vdev_delete(ar, arvif->vdev_id);
 	if (ret)
-		ath10k_warn("WMI vdev %i delete failed: %d\n",
+		ath10k_warn("failed to delete WMI vdev %i: %d\n",
 			    arvif->vdev_id, ret);
 
-	if (arvif->vdev_type == WMI_VDEV_TYPE_MONITOR)
-		ar->monitor_present = false;
-
 	ath10k_peer_cleanup(ar, arvif->vdev_id);
 
 	mutex_unlock(&ar->conf_mutex);
@@ -2728,28 +2944,17 @@ static void ath10k_configure_filter(struct ieee80211_hw *hw,
 	*total_flags &= SUPPORTED_FILTERS;
 	ar->filter_flags = *total_flags;
 
-	/* Monitor must not be started if it wasn't created first.
-	 * Promiscuous mode may be started on a non-monitor interface - in
-	 * such case the monitor vdev is not created so starting the
-	 * monitor makes no sense. Since ath10k uses no special RX filters
-	 * (only BSS filter in STA mode) there's no need for any special
-	 * action here. */
-	if ((ar->filter_flags & FIF_PROMISC_IN_BSS) &&
-	    !ar->monitor_enabled && ar->monitor_present) {
-		ath10k_dbg(ATH10K_DBG_MAC, "mac monitor %d start\n",
-			   ar->monitor_vdev_id);
-
-		ret = ath10k_monitor_start(ar, ar->monitor_vdev_id);
-		if (ret)
-			ath10k_warn("Unable to start monitor mode\n");
-	} else if (!(ar->filter_flags & FIF_PROMISC_IN_BSS) &&
-		   ar->monitor_enabled && ar->monitor_present) {
-		ath10k_dbg(ATH10K_DBG_MAC, "mac monitor %d stop\n",
-			   ar->monitor_vdev_id);
-
-		ret = ath10k_monitor_stop(ar);
-		if (ret)
-			ath10k_warn("Unable to stop monitor mode\n");
+	if (ar->filter_flags & FIF_PROMISC_IN_BSS && !ar->promisc) {
+		ar->promisc = true;
+		ret = ath10k_monitor_start(ar);
+		if (ret) {
+			ath10k_warn("failed to start monitor (promisc): %d\n",
+				    ret);
+			ar->promisc = false;
+		}
+	} else if (!(ar->filter_flags & FIF_PROMISC_IN_BSS) && ar->promisc) {
+		ar->promisc = false;
+		ath10k_monitor_stop(ar);
 	}
 
 	mutex_unlock(&ar->conf_mutex);
@@ -2780,7 +2985,7 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 			   arvif->vdev_id, arvif->beacon_interval);
 
 		if (ret)
-			ath10k_warn("Failed to set beacon interval for vdev %d: %i\n",
+			ath10k_warn("failed to set beacon interval for vdev %d: %i\n",
 				    arvif->vdev_id, ret);
 	}
 
@@ -2793,7 +2998,7 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_pdev_set_param(ar, pdev_param,
 						WMI_BEACON_STAGGERED_MODE);
 		if (ret)
-			ath10k_warn("Failed to set beacon mode for vdev %d: %i\n",
+			ath10k_warn("failed to set beacon mode for vdev %d: %i\n",
 				    arvif->vdev_id, ret);
 	}
 
@@ -2808,7 +3013,7 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, vdev_param,
 						arvif->dtim_period);
 		if (ret)
-			ath10k_warn("Failed to set dtim period for vdev %d: %i\n",
+			ath10k_warn("failed to set dtim period for vdev %d: %i\n",
 				    arvif->vdev_id, ret);
 	}
 
@@ -2820,7 +3025,12 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		arvif->u.ap.hidden_ssid = info->hidden_ssid;
 	}
 
-	if (changed & BSS_CHANGED_BSSID) {
+	/*
+	 * Firmware manages AP self-peer internally so make sure to not create
+	 * it in driver. Otherwise AP self-peer deletion may timeout later.
+	 */
+	if (changed & BSS_CHANGED_BSSID &&
+	    vif->type != NL80211_IFTYPE_AP) {
 		if (!is_zero_ether_addr(info->bssid)) {
 			ath10k_dbg(ATH10K_DBG_MAC,
 				   "mac vdev %d create peer %pM\n",
@@ -2829,7 +3039,7 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 			ret = ath10k_peer_create(ar, arvif->vdev_id,
 						 info->bssid);
 			if (ret)
-				ath10k_warn("Failed to add peer %pM for vdev %d when changing bssid: %i\n",
+				ath10k_warn("failed to add peer %pM for vdev %d when changing bssid: %i\n",
 					    info->bssid, arvif->vdev_id, ret);
 
 			if (vif->type == NL80211_IFTYPE_STATION) {
@@ -2868,20 +3078,13 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		ath10k_control_beaconing(arvif, info);
 
 	if (changed & BSS_CHANGED_ERP_CTS_PROT) {
-		u32 cts_prot;
-		if (info->use_cts_prot)
-			cts_prot = 1;
-		else
-			cts_prot = 0;
-
+		arvif->use_cts_prot = info->use_cts_prot;
 		ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d cts_prot %d\n",
-			   arvif->vdev_id, cts_prot);
+			   arvif->vdev_id, info->use_cts_prot);
 
-		vdev_param = ar->wmi.vdev_param->enable_rtscts;
-		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, vdev_param,
-						cts_prot);
+		ret = ath10k_recalc_rtscts_prot(arvif);
 		if (ret)
-			ath10k_warn("Failed to set CTS prot for vdev %d: %d\n",
+			ath10k_warn("failed to recalculate rts/cts prot for vdev %d: %d\n",
 				    arvif->vdev_id, ret);
 	}
 
@@ -2900,7 +3103,7 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, vdev_param,
 						slottime);
 		if (ret)
-			ath10k_warn("Failed to set erp slot for vdev %d: %i\n",
+			ath10k_warn("failed to set erp slot for vdev %d: %i\n",
 				    arvif->vdev_id, ret);
 	}
 
@@ -2919,7 +3122,7 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id, vdev_param,
 						preamble);
 		if (ret)
-			ath10k_warn("Failed to set preamble for vdev %d: %i\n",
+			ath10k_warn("failed to set preamble for vdev %d: %i\n",
 				    arvif->vdev_id, ret);
 	}
 
@@ -2990,7 +3193,7 @@ static int ath10k_hw_scan(struct ieee80211_hw *hw,
 
 	ret = ath10k_start_scan(ar, &arg);
 	if (ret) {
-		ath10k_warn("could not start hw scan (%d)\n", ret);
+		ath10k_warn("failed to start hw scan: %d\n", ret);
 		spin_lock_bh(&ar->data_lock);
 		ar->scan.in_progress = false;
 		spin_unlock_bh(&ar->data_lock);
@@ -3010,8 +3213,7 @@ static void ath10k_cancel_hw_scan(struct ieee80211_hw *hw,
 	mutex_lock(&ar->conf_mutex);
 	ret = ath10k_abort_scan(ar);
 	if (ret) {
-		ath10k_warn("couldn't abort scan (%d). forcefully sending scan completion to mac80211\n",
-			    ret);
+		ath10k_warn("failed to abort scan: %d\n", ret);
 		ieee80211_scan_completed(hw, 1 /* aborted */);
 	}
 	mutex_unlock(&ar->conf_mutex);
@@ -3089,7 +3291,7 @@ static int ath10k_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 	if (!peer) {
 		if (cmd == SET_KEY) {
-			ath10k_warn("cannot install key for non-existent peer %pM\n",
+			ath10k_warn("failed to install key for non-existent peer %pM\n",
 				    peer_addr);
 			ret = -EOPNOTSUPP;
 			goto exit;
@@ -3112,7 +3314,7 @@ static int ath10k_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 	ret = ath10k_install_key(arvif, key, cmd, peer_addr);
 	if (ret) {
-		ath10k_warn("key installation failed for vdev %i peer %pM: %d\n",
+		ath10k_warn("failed to install key for vdev %i peer %pM: %d\n",
 			    arvif->vdev_id, peer_addr, ret);
 		goto exit;
 	}
@@ -3127,7 +3329,7 @@ static int ath10k_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 		peer->keys[key->keyidx] = NULL;
 	else if (peer == NULL)
 		/* impossible unless FW goes crazy */
-		ath10k_warn("peer %pM disappeared!\n", peer_addr);
+		ath10k_warn("Peer %pM disappeared!\n", peer_addr);
 	spin_unlock_bh(&ar->data_lock);
 
 exit:
@@ -3195,6 +3397,16 @@ static void ath10k_sta_rc_update_wk(struct work_struct *wk)
 				    sta->addr, smps, err);
 	}
 
+	if (changed & IEEE80211_RC_SUPP_RATES_CHANGED) {
+		ath10k_dbg(ATH10K_DBG_MAC, "mac update sta %pM supp rates\n",
+			   sta->addr);
+
+		err = ath10k_station_assoc(ar, arvif, sta, true);
+		if (err)
+			ath10k_warn("failed to reassociate station: %pM\n",
+				    sta->addr);
+	}
+
 	mutex_unlock(&ar->conf_mutex);
 }
 
@@ -3236,7 +3448,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 			max_num_peers = TARGET_NUM_PEERS;
 
 		if (ar->num_peers >= max_num_peers) {
-			ath10k_warn("Number of peers exceeded: peers number %d (max peers %d)\n",
+			ath10k_warn("number of peers exceeded: peers number %d (max peers %d)\n",
 				    ar->num_peers, max_num_peers);
 			ret = -ENOBUFS;
 			goto exit;
@@ -3248,7 +3460,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 
 		ret = ath10k_peer_create(ar, arvif->vdev_id, sta->addr);
 		if (ret)
-			ath10k_warn("Failed to add peer %pM for vdev %d when adding a new sta: %i\n",
+			ath10k_warn("failed to add peer %pM for vdev %d when adding a new sta: %i\n",
 				    sta->addr, arvif->vdev_id, ret);
 	} else if ((old_state == IEEE80211_STA_NONE &&
 		    new_state == IEEE80211_STA_NOTEXIST)) {
@@ -3260,7 +3472,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 			   arvif->vdev_id, sta->addr);
 		ret = ath10k_peer_delete(ar, arvif->vdev_id, sta->addr);
 		if (ret)
-			ath10k_warn("Failed to delete peer %pM for vdev %d: %i\n",
+			ath10k_warn("failed to delete peer %pM for vdev %d: %i\n",
 				    sta->addr, arvif->vdev_id, ret);
 
 		if (vif->type == NL80211_IFTYPE_STATION)
@@ -3275,9 +3487,9 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 		ath10k_dbg(ATH10K_DBG_MAC, "mac sta %pM associated\n",
 			   sta->addr);
 
-		ret = ath10k_station_assoc(ar, arvif, sta);
+		ret = ath10k_station_assoc(ar, arvif, sta, false);
 		if (ret)
-			ath10k_warn("Failed to associate station %pM for vdev %i: %i\n",
+			ath10k_warn("failed to associate station %pM for vdev %i: %i\n",
 				    sta->addr, arvif->vdev_id, ret);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTH &&
@@ -3291,7 +3503,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 
 		ret = ath10k_station_disassoc(ar, arvif, sta);
 		if (ret)
-			ath10k_warn("Failed to disassociate station: %pM vdev %i ret %i\n",
+			ath10k_warn("failed to disassociate station: %pM vdev %i: %i\n",
 				    sta->addr, arvif->vdev_id, ret);
 	}
 exit:
@@ -3339,7 +3551,7 @@ static int ath10k_conf_tx_uapsd(struct ath10k *ar, struct ieee80211_vif *vif,
 					  WMI_STA_PS_PARAM_UAPSD,
 					  arvif->u.sta.uapsd);
 	if (ret) {
-		ath10k_warn("could not set uapsd params %d\n", ret);
+		ath10k_warn("failed to set uapsd params: %d\n", ret);
 		goto exit;
 	}
 
@@ -3352,7 +3564,7 @@ static int ath10k_conf_tx_uapsd(struct ath10k *ar, struct ieee80211_vif *vif,
 					  WMI_STA_PS_PARAM_RX_WAKE_POLICY,
 					  value);
 	if (ret)
-		ath10k_warn("could not set rx wake param %d\n", ret);
+		ath10k_warn("failed to set rx wake param: %d\n", ret);
 
 exit:
 	return ret;
@@ -3402,13 +3614,13 @@ static int ath10k_conf_tx(struct ieee80211_hw *hw,
 	/* FIXME: FW accepts wmm params per hw, not per vif */
 	ret = ath10k_wmi_pdev_set_wmm_params(ar, &ar->wmm_params);
 	if (ret) {
-		ath10k_warn("could not set wmm params %d\n", ret);
+		ath10k_warn("failed to set wmm params: %d\n", ret);
 		goto exit;
 	}
 
 	ret = ath10k_conf_tx_uapsd(ar, vif, ac, params->uapsd);
 	if (ret)
-		ath10k_warn("could not set sta uapsd %d\n", ret);
+		ath10k_warn("failed to set sta uapsd: %d\n", ret);
 
 exit:
 	mutex_unlock(&ar->conf_mutex);
@@ -3461,7 +3673,7 @@ static int ath10k_remain_on_channel(struct ieee80211_hw *hw,
 
 	ret = ath10k_start_scan(ar, &arg);
 	if (ret) {
-		ath10k_warn("could not start roc scan (%d)\n", ret);
+		ath10k_warn("failed to start roc scan: %d\n", ret);
 		spin_lock_bh(&ar->data_lock);
 		ar->scan.in_progress = false;
 		spin_unlock_bh(&ar->data_lock);
@@ -3470,7 +3682,7 @@ static int ath10k_remain_on_channel(struct ieee80211_hw *hw,
 
 	ret = wait_for_completion_timeout(&ar->scan.on_channel, 3*HZ);
 	if (ret == 0) {
-		ath10k_warn("could not switch to channel for roc scan\n");
+		ath10k_warn("failed to switch to channel for roc scan\n");
 		ath10k_abort_scan(ar);
 		ret = -ETIMEDOUT;
 		goto exit;
@@ -3511,7 +3723,7 @@ static int ath10k_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 
 		ret = ath10k_mac_set_rts(arvif, value);
 		if (ret) {
-			ath10k_warn("could not set rts threshold for vdev %d (%d)\n",
+			ath10k_warn("failed to set rts threshold for vdev %d: %d\n",
 				    arvif->vdev_id, ret);
 			break;
 		}
@@ -3534,7 +3746,7 @@ static int ath10k_set_frag_threshold(struct ieee80211_hw *hw, u32 value)
 
 		ret = ath10k_mac_set_rts(arvif, value);
 		if (ret) {
-			ath10k_warn("could not set fragmentation threshold for vdev %d (%d)\n",
+			ath10k_warn("failed to set fragmentation threshold for vdev %d: %d\n",
 				    arvif->vdev_id, ret);
 			break;
 		}
@@ -3544,7 +3756,8 @@ static int ath10k_set_frag_threshold(struct ieee80211_hw *hw, u32 value)
 	return ret;
 }
 
-static void ath10k_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void ath10k_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			 u32 queues, bool drop)
 {
 	struct ath10k *ar = hw->priv;
 	bool skip;
@@ -3573,7 +3786,7 @@ static void ath10k_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
 		}), ATH10K_FLUSH_TIMEOUT_HZ);
 
 	if (ret <= 0 || skip)
-		ath10k_warn("tx not flushed (skip %i ar-state %i): %i\n",
+		ath10k_warn("failed to flush transmit queue (skip %i ar-state %i): %i\n",
 			    skip, ar->state, ret);
 
 skip:
@@ -3608,7 +3821,7 @@ static int ath10k_suspend(struct ieee80211_hw *hw,
 
 	ret = ath10k_hif_suspend(ar);
 	if (ret) {
-		ath10k_warn("could not suspend hif (%d)\n", ret);
+		ath10k_warn("failed to suspend hif: %d\n", ret);
 		goto resume;
 	}
 
@@ -3617,7 +3830,7 @@ static int ath10k_suspend(struct ieee80211_hw *hw,
 resume:
 	ret = ath10k_wmi_pdev_resume_target(ar);
 	if (ret)
-		ath10k_warn("could not resume target (%d)\n", ret);
+		ath10k_warn("failed to resume target: %d\n", ret);
 
 	ret = 1;
 exit:
@@ -3634,14 +3847,14 @@ static int ath10k_resume(struct ieee80211_hw *hw)
 
 	ret = ath10k_hif_resume(ar);
 	if (ret) {
-		ath10k_warn("could not resume hif (%d)\n", ret);
+		ath10k_warn("failed to resume hif: %d\n", ret);
 		ret = 1;
 		goto exit;
 	}
 
 	ret = ath10k_wmi_pdev_resume_target(ar);
 	if (ret) {
-		ath10k_warn("could not resume target (%d)\n", ret);
+		ath10k_warn("failed to resume target: %d\n", ret);
 		ret = 1;
 		goto exit;
 	}
@@ -3964,7 +4177,7 @@ static int ath10k_set_fixed_rate_param(struct ath10k_vif *arvif,
 	ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 					vdev_param, fixed_rate);
 	if (ret) {
-		ath10k_warn("Could not set fixed_rate param 0x%02x: %d\n",
+		ath10k_warn("failed to set fixed rate param 0x%02x: %d\n",
 			    fixed_rate, ret);
 		ret = -EINVAL;
 		goto exit;
@@ -3977,7 +4190,7 @@ static int ath10k_set_fixed_rate_param(struct ath10k_vif *arvif,
 					vdev_param, fixed_nss);
 
 	if (ret) {
-		ath10k_warn("Could not set fixed_nss param %d: %d\n",
+		ath10k_warn("failed to set fixed nss param %d: %d\n",
 			    fixed_nss, ret);
 		ret = -EINVAL;
 		goto exit;
@@ -3990,7 +4203,7 @@ static int ath10k_set_fixed_rate_param(struct ath10k_vif *arvif,
 					force_sgi);
 
 	if (ret) {
-		ath10k_warn("Could not set sgi param %d: %d\n",
+		ath10k_warn("failed to set sgi param %d: %d\n",
 			    force_sgi, ret);
 		ret = -EINVAL;
 		goto exit;
@@ -4026,7 +4239,7 @@ static int ath10k_set_bitrate_mask(struct ieee80211_hw *hw,
 	}
 
 	if (fixed_rate == WMI_FIXED_RATE_NONE && force_sgi) {
-		ath10k_warn("Could not force SGI usage for default rate settings\n");
+		ath10k_warn("failed to force SGI usage for default rate settings\n");
 		return -EINVAL;
 	}
 
@@ -4034,14 +4247,6 @@ static int ath10k_set_bitrate_mask(struct ieee80211_hw *hw,
 					   fixed_nss, force_sgi);
 }
 
-static void ath10k_channel_switch_beacon(struct ieee80211_hw *hw,
-					 struct ieee80211_vif *vif,
-					 struct cfg80211_chan_def *chandef)
-{
-	/* there's no need to do anything here. vif->csa_active is enough */
-	return;
-}
-
 static void ath10k_sta_rc_update(struct ieee80211_hw *hw,
 				 struct ieee80211_vif *vif,
 				 struct ieee80211_sta *sta,
@@ -4072,8 +4277,8 @@ static void ath10k_sta_rc_update(struct ieee80211_hw *hw,
 			bw = WMI_PEER_CHWIDTH_80MHZ;
 			break;
 		case IEEE80211_STA_RX_BW_160:
-			ath10k_warn("mac sta rc update for %pM: invalid bw %d\n",
-				    sta->addr, sta->bandwidth);
+			ath10k_warn("Invalid bandwith %d in rc update for %pM\n",
+				    sta->bandwidth, sta->addr);
 			bw = WMI_PEER_CHWIDTH_20MHZ;
 			break;
 		}
@@ -4099,8 +4304,8 @@ static void ath10k_sta_rc_update(struct ieee80211_hw *hw,
 			smps = WMI_PEER_SMPS_DYNAMIC;
 			break;
 		case IEEE80211_SMPS_NUM_MODES:
-			ath10k_warn("mac sta rc update for %pM: invalid smps: %d\n",
-				    sta->addr, sta->smps_mode);
+			ath10k_warn("Invalid smps %d in sta rc update for %pM\n",
+				    sta->smps_mode, sta->addr);
 			smps = WMI_PEER_SMPS_PS_NONE;
 			break;
 		}
@@ -4108,15 +4313,6 @@ static void ath10k_sta_rc_update(struct ieee80211_hw *hw,
 		arsta->smps = smps;
 	}
 
-	if (changed & IEEE80211_RC_SUPP_RATES_CHANGED) {
-		/* FIXME: Not implemented. Probably the only way to do it would
-		 * be to re-assoc the peer. */
-		changed &= ~IEEE80211_RC_SUPP_RATES_CHANGED;
-		ath10k_dbg(ATH10K_DBG_MAC,
-			   "mac sta rc update for %pM: changing supported rates not implemented\n",
-			   sta->addr);
-	}
-
 	arsta->changed |= changed;
 
 	spin_unlock_bh(&ar->data_lock);
@@ -4154,10 +4350,11 @@ static const struct ieee80211_ops ath10k_ops = {
 	.set_frag_threshold		= ath10k_set_frag_threshold,
 	.flush				= ath10k_flush,
 	.tx_last_beacon			= ath10k_tx_last_beacon,
+	.set_antenna			= ath10k_set_antenna,
+	.get_antenna			= ath10k_get_antenna,
 	.restart_complete		= ath10k_restart_complete,
 	.get_survey			= ath10k_get_survey,
 	.set_bitrate_mask		= ath10k_set_bitrate_mask,
-	.channel_switch_beacon		= ath10k_channel_switch_beacon,
 	.sta_rc_update			= ath10k_sta_rc_update,
 	.get_tsf			= ath10k_get_tsf,
 #ifdef CONFIG_PM
@@ -4503,6 +4700,18 @@ int ath10k_mac_register(struct ath10k *ar)
 		BIT(NL80211_IFTYPE_ADHOC) |
 		BIT(NL80211_IFTYPE_AP);
 
+	if (test_bit(ATH10K_FW_FEATURE_WMI_10X, ar->fw_features)) {
+		/* TODO:  Have to deal with 2x2 chips if/when the come out. */
+		ar->supp_tx_chainmask = TARGET_10X_TX_CHAIN_MASK;
+		ar->supp_rx_chainmask = TARGET_10X_RX_CHAIN_MASK;
+	} else {
+		ar->supp_tx_chainmask = TARGET_TX_CHAIN_MASK;
+		ar->supp_rx_chainmask = TARGET_RX_CHAIN_MASK;
+	}
+
+	ar->hw->wiphy->available_antennas_rx = ar->supp_rx_chainmask;
+	ar->hw->wiphy->available_antennas_tx = ar->supp_tx_chainmask;
+
 	if (!test_bit(ATH10K_FW_FEATURE_NO_P2P, ar->fw_features))
 		ar->hw->wiphy->interface_modes |=
 			BIT(NL80211_IFTYPE_P2P_CLIENT) |
@@ -4516,7 +4725,6 @@ int ath10k_mac_register(struct ath10k *ar)
 			IEEE80211_HW_REPORTS_TX_ACK_STATUS |
 			IEEE80211_HW_HAS_RATE_CONTROL |
 			IEEE80211_HW_SUPPORTS_STATIC_SMPS |
-			IEEE80211_HW_WANT_MONITOR_VIF |
 			IEEE80211_HW_AP_LINK_PS |
 			IEEE80211_HW_SPECTRUM_MGMT;
 
@@ -4570,19 +4778,19 @@ int ath10k_mac_register(struct ath10k *ar)
 							     NL80211_DFS_UNSET);
 
 		if (!ar->dfs_detector)
-			ath10k_warn("dfs pattern detector init failed\n");
+			ath10k_warn("failed to initialise DFS pattern detector\n");
 	}
 
 	ret = ath_regd_init(&ar->ath_common.regulatory, ar->hw->wiphy,
 			    ath10k_reg_notifier);
 	if (ret) {
-		ath10k_err("Regulatory initialization failed: %i\n", ret);
+		ath10k_err("failed to initialise regulatory: %i\n", ret);
 		goto err_free;
 	}
 
 	ret = ieee80211_register_hw(ar->hw);
 	if (ret) {
-		ath10k_err("ieee80211 registration failed: %d\n", ret);
+		ath10k_err("failed to register ieee80211: %d\n", ret);
 		goto err_free;
 	}
 
diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 9d242d801d9d..d0004d59c97e 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -39,15 +39,28 @@ enum ath10k_pci_irq_mode {
 	ATH10K_PCI_IRQ_MSI = 2,
 };
 
-static unsigned int ath10k_target_ps;
+enum ath10k_pci_reset_mode {
+	ATH10K_PCI_RESET_AUTO = 0,
+	ATH10K_PCI_RESET_WARM_ONLY = 1,
+};
+
+static unsigned int ath10k_pci_target_ps;
 static unsigned int ath10k_pci_irq_mode = ATH10K_PCI_IRQ_AUTO;
+static unsigned int ath10k_pci_reset_mode = ATH10K_PCI_RESET_AUTO;
 
-module_param(ath10k_target_ps, uint, 0644);
-MODULE_PARM_DESC(ath10k_target_ps, "Enable ath10k Target (SoC) PS option");
+module_param_named(target_ps, ath10k_pci_target_ps, uint, 0644);
+MODULE_PARM_DESC(target_ps, "Enable ath10k Target (SoC) PS option");
 
 module_param_named(irq_mode, ath10k_pci_irq_mode, uint, 0644);
 MODULE_PARM_DESC(irq_mode, "0: auto, 1: legacy, 2: msi (default: 0)");
 
+module_param_named(reset_mode, ath10k_pci_reset_mode, uint, 0644);
+MODULE_PARM_DESC(reset_mode, "0: auto, 1: warm only (default: 0)");
+
+/* how long wait to wait for target to initialise, in ms */
+#define ATH10K_PCI_TARGET_WAIT 3000
+#define ATH10K_PCI_NUM_WARM_RESET_ATTEMPTS 3
+
 #define QCA988X_2_0_DEVICE_ID	(0x003c)
 
 static DEFINE_PCI_DEVICE_TABLE(ath10k_pci_id_table) = {
@@ -346,9 +359,10 @@ static int ath10k_pci_diag_read_mem(struct ath10k *ar, u32 address, void *data,
 	 *   2) Buffer in DMA-able space
 	 */
 	orig_nbytes = nbytes;
-	data_buf = (unsigned char *)pci_alloc_consistent(ar_pci->pdev,
-							 orig_nbytes,
-							 &ce_data_base);
+	data_buf = (unsigned char *)dma_alloc_coherent(ar->dev,
+						       orig_nbytes,
+						       &ce_data_base,
+						       GFP_ATOMIC);
 
 	if (!data_buf) {
 		ret = -ENOMEM;
@@ -442,12 +456,12 @@ done:
 				__le32_to_cpu(((__le32 *)data_buf)[i]);
 		}
 	} else
-		ath10k_dbg(ATH10K_DBG_PCI, "%s failure (0x%x)\n",
-			   __func__, address);
+		ath10k_warn("failed to read diag value at 0x%x: %d\n",
+			    address, ret);
 
 	if (data_buf)
-		pci_free_consistent(ar_pci->pdev, orig_nbytes,
-				    data_buf, ce_data_base);
+		dma_free_coherent(ar->dev, orig_nbytes, data_buf,
+				  ce_data_base);
 
 	return ret;
 }
@@ -490,9 +504,10 @@ static int ath10k_pci_diag_write_mem(struct ath10k *ar, u32 address,
 	 *   2) Buffer in DMA-able space
 	 */
 	orig_nbytes = nbytes;
-	data_buf = (unsigned char *)pci_alloc_consistent(ar_pci->pdev,
-							 orig_nbytes,
-							 &ce_data_base);
+	data_buf = (unsigned char *)dma_alloc_coherent(ar->dev,
+						       orig_nbytes,
+						       &ce_data_base,
+						       GFP_ATOMIC);
 	if (!data_buf) {
 		ret = -ENOMEM;
 		goto done;
@@ -588,13 +603,13 @@ static int ath10k_pci_diag_write_mem(struct ath10k *ar, u32 address,
 
 done:
 	if (data_buf) {
-		pci_free_consistent(ar_pci->pdev, orig_nbytes, data_buf,
-				    ce_data_base);
+		dma_free_coherent(ar->dev, orig_nbytes, data_buf,
+				  ce_data_base);
 	}
 
 	if (ret != 0)
-		ath10k_dbg(ATH10K_DBG_PCI, "%s failure (0x%x)\n", __func__,
-			   address);
+		ath10k_warn("failed to write diag value at 0x%x: %d\n",
+			    address, ret);
 
 	return ret;
 }
@@ -747,17 +762,21 @@ static int ath10k_pci_hif_tx_sg(struct ath10k *ar, u8 pipe_id,
 	struct ath10k_pci_pipe *pci_pipe = &ar_pci->pipe_info[pipe_id];
 	struct ath10k_ce_pipe *ce_pipe = pci_pipe->ce_hdl;
 	struct ath10k_ce_ring *src_ring = ce_pipe->src_ring;
-	unsigned int nentries_mask = src_ring->nentries_mask;
-	unsigned int sw_index = src_ring->sw_index;
-	unsigned int write_index = src_ring->write_index;
-	int err, i;
+	unsigned int nentries_mask;
+	unsigned int sw_index;
+	unsigned int write_index;
+	int err, i = 0;
 
 	spin_lock_bh(&ar_pci->ce_lock);
 
+	nentries_mask = src_ring->nentries_mask;
+	sw_index = src_ring->sw_index;
+	write_index = src_ring->write_index;
+
 	if (unlikely(CE_RING_DELTA(nentries_mask,
 				   write_index, sw_index - 1) < n_items)) {
 		err = -ENOBUFS;
-		goto unlock;
+		goto err;
 	}
 
 	for (i = 0; i < n_items - 1; i++) {
@@ -774,7 +793,7 @@ static int ath10k_pci_hif_tx_sg(struct ath10k *ar, u8 pipe_id,
 					    items[i].transfer_id,
 					    CE_SEND_FLAG_GATHER);
 		if (err)
-			goto unlock;
+			goto err;
 	}
 
 	/* `i` is equal to `n_items -1` after for() */
@@ -792,10 +811,15 @@ static int ath10k_pci_hif_tx_sg(struct ath10k *ar, u8 pipe_id,
 				    items[i].transfer_id,
 				    0);
 	if (err)
-		goto unlock;
+		goto err;
+
+	spin_unlock_bh(&ar_pci->ce_lock);
+	return 0;
+
+err:
+	for (; i > 0; i--)
+		__ath10k_ce_send_revert(ce_pipe);
 
-	err = 0;
-unlock:
 	spin_unlock_bh(&ar_pci->ce_lock);
 	return err;
 }
@@ -803,6 +827,9 @@ unlock:
 static u16 ath10k_pci_hif_get_free_queue_number(struct ath10k *ar, u8 pipe)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+
+	ath10k_dbg(ATH10K_DBG_PCI, "pci hif get free queue number\n");
+
 	return ath10k_ce_num_free_src_entries(ar_pci->pipe_info[pipe].ce_hdl);
 }
 
@@ -854,6 +881,8 @@ static void ath10k_pci_hif_dump_area(struct ath10k *ar)
 static void ath10k_pci_hif_send_complete_check(struct ath10k *ar, u8 pipe,
 					       int force)
 {
+	ath10k_dbg(ATH10K_DBG_PCI, "pci hif send complete check\n");
+
 	if (!force) {
 		int resources;
 		/*
@@ -880,7 +909,7 @@ static void ath10k_pci_hif_set_callbacks(struct ath10k *ar,
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 
-	ath10k_dbg(ATH10K_DBG_PCI, "%s\n", __func__);
+	ath10k_dbg(ATH10K_DBG_PCI, "pci hif set callbacks\n");
 
 	memcpy(&ar_pci->msg_callbacks_current, callbacks,
 	       sizeof(ar_pci->msg_callbacks_current));
@@ -938,6 +967,8 @@ static int ath10k_pci_hif_map_service_to_pipe(struct ath10k *ar,
 {
 	int ret = 0;
 
+	ath10k_dbg(ATH10K_DBG_PCI, "pci hif map service\n");
+
 	/* polling for received messages not supported */
 	*dl_is_polled = 0;
 
@@ -997,6 +1028,8 @@ static void ath10k_pci_hif_get_default_pipe(struct ath10k *ar,
 {
 	int ul_is_polled, dl_is_polled;
 
+	ath10k_dbg(ATH10K_DBG_PCI, "pci hif get default pipe\n");
+
 	(void)ath10k_pci_hif_map_service_to_pipe(ar,
 						 ATH10K_HTC_SVC_ID_RSVD_CTRL,
 						 ul_pipe,
@@ -1098,6 +1131,8 @@ static int ath10k_pci_hif_start(struct ath10k *ar)
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int ret, ret_early;
 
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot hif start\n");
+
 	ath10k_pci_free_early_irq(ar);
 	ath10k_pci_kill_tasklet(ar);
 
@@ -1233,18 +1268,10 @@ static void ath10k_pci_buffer_cleanup(struct ath10k *ar)
 
 static void ath10k_pci_ce_deinit(struct ath10k *ar)
 {
-	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
-	struct ath10k_pci_pipe *pipe_info;
-	int pipe_num;
+	int i;
 
-	for (pipe_num = 0; pipe_num < CE_COUNT; pipe_num++) {
-		pipe_info = &ar_pci->pipe_info[pipe_num];
-		if (pipe_info->ce_hdl) {
-			ath10k_ce_deinit(pipe_info->ce_hdl);
-			pipe_info->ce_hdl = NULL;
-			pipe_info->buf_sz = 0;
-		}
-	}
+	for (i = 0; i < CE_COUNT; i++)
+		ath10k_ce_deinit_pipe(ar, i);
 }
 
 static void ath10k_pci_hif_stop(struct ath10k *ar)
@@ -1252,7 +1279,10 @@ static void ath10k_pci_hif_stop(struct ath10k *ar)
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int ret;
 
-	ath10k_dbg(ATH10K_DBG_PCI, "%s\n", __func__);
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot hif stop\n");
+
+	if (WARN_ON(!ar_pci->started))
+		return;
 
 	ret = ath10k_ce_disable_interrupts(ar);
 	if (ret)
@@ -1697,30 +1727,49 @@ static int ath10k_pci_init_config(struct ath10k *ar)
 	return 0;
 }
 
+static int ath10k_pci_alloc_ce(struct ath10k *ar)
+{
+	int i, ret;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		ret = ath10k_ce_alloc_pipe(ar, i, &host_ce_config_wlan[i]);
+		if (ret) {
+			ath10k_err("failed to allocate copy engine pipe %d: %d\n",
+				   i, ret);
+			return ret;
+		}
+	}
 
+	return 0;
+}
+
+static void ath10k_pci_free_ce(struct ath10k *ar)
+{
+	int i;
+
+	for (i = 0; i < CE_COUNT; i++)
+		ath10k_ce_free_pipe(ar, i);
+}
 
 static int ath10k_pci_ce_init(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pipe_info;
 	const struct ce_attr *attr;
-	int pipe_num;
+	int pipe_num, ret;
 
 	for (pipe_num = 0; pipe_num < CE_COUNT; pipe_num++) {
 		pipe_info = &ar_pci->pipe_info[pipe_num];
+		pipe_info->ce_hdl = &ar_pci->ce_states[pipe_num];
 		pipe_info->pipe_num = pipe_num;
 		pipe_info->hif_ce_state = ar;
 		attr = &host_ce_config_wlan[pipe_num];
 
-		pipe_info->ce_hdl = ath10k_ce_init(ar, pipe_num, attr);
-		if (pipe_info->ce_hdl == NULL) {
-			ath10k_err("failed to initialize CE for pipe: %d\n",
-				   pipe_num);
-
-			/* It is safe to call it here. It checks if ce_hdl is
-			 * valid for each pipe */
-			ath10k_pci_ce_deinit(ar);
-			return -1;
+		ret = ath10k_ce_init_pipe(ar, pipe_num, attr);
+		if (ret) {
+			ath10k_err("failed to initialize copy engine pipe %d: %d\n",
+				   pipe_num, ret);
+			return ret;
 		}
 
 		if (pipe_num == CE_COUNT - 1) {
@@ -1741,16 +1790,15 @@ static int ath10k_pci_ce_init(struct ath10k *ar)
 static void ath10k_pci_fw_interrupt_handler(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
-	u32 fw_indicator_address, fw_indicator;
+	u32 fw_indicator;
 
 	ath10k_pci_wake(ar);
 
-	fw_indicator_address = ar_pci->fw_indicator_address;
-	fw_indicator = ath10k_pci_read32(ar, fw_indicator_address);
+	fw_indicator = ath10k_pci_read32(ar, FW_INDICATOR_ADDRESS);
 
 	if (fw_indicator & FW_IND_EVENT_PENDING) {
 		/* ACK: clear Target-side pending event */
-		ath10k_pci_write32(ar, fw_indicator_address,
+		ath10k_pci_write32(ar, FW_INDICATOR_ADDRESS,
 				   fw_indicator & ~FW_IND_EVENT_PENDING);
 
 		if (ar_pci->started) {
@@ -1767,13 +1815,32 @@ static void ath10k_pci_fw_interrupt_handler(struct ath10k *ar)
 	ath10k_pci_sleep(ar);
 }
 
+/* this function effectively clears target memory controller assert line */
+static void ath10k_pci_warm_reset_si0(struct ath10k *ar)
+{
+	u32 val;
+
+	val = ath10k_pci_soc_read32(ar, SOC_RESET_CONTROL_ADDRESS);
+	ath10k_pci_soc_write32(ar, SOC_RESET_CONTROL_ADDRESS,
+			       val | SOC_RESET_CONTROL_SI0_RST_MASK);
+	val = ath10k_pci_soc_read32(ar, SOC_RESET_CONTROL_ADDRESS);
+
+	msleep(10);
+
+	val = ath10k_pci_soc_read32(ar, SOC_RESET_CONTROL_ADDRESS);
+	ath10k_pci_soc_write32(ar, SOC_RESET_CONTROL_ADDRESS,
+			       val & ~SOC_RESET_CONTROL_SI0_RST_MASK);
+	val = ath10k_pci_soc_read32(ar, SOC_RESET_CONTROL_ADDRESS);
+
+	msleep(10);
+}
+
 static int ath10k_pci_warm_reset(struct ath10k *ar)
 {
-	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int ret = 0;
 	u32 val;
 
-	ath10k_dbg(ATH10K_DBG_BOOT, "boot performing warm chip reset\n");
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot warm reset\n");
 
 	ret = ath10k_do_pci_wake(ar);
 	if (ret) {
@@ -1801,7 +1868,7 @@ static int ath10k_pci_warm_reset(struct ath10k *ar)
 	msleep(100);
 
 	/* clear fw indicator */
-	ath10k_pci_write32(ar, ar_pci->fw_indicator_address, 0);
+	ath10k_pci_write32(ar, FW_INDICATOR_ADDRESS, 0);
 
 	/* clear target LF timer interrupts */
 	val = ath10k_pci_read32(ar, RTC_SOC_BASE_ADDRESS +
@@ -1826,6 +1893,8 @@ static int ath10k_pci_warm_reset(struct ath10k *ar)
 				SOC_RESET_CONTROL_ADDRESS);
 	msleep(10);
 
+	ath10k_pci_warm_reset_si0(ar);
+
 	/* debug */
 	val = ath10k_pci_read32(ar, SOC_CORE_BASE_ADDRESS +
 				PCIE_INTR_CAUSE_ADDRESS);
@@ -1934,7 +2003,9 @@ static int __ath10k_pci_hif_power_up(struct ath10k *ar, bool cold_reset)
 		irq_mode = "legacy";
 
 	if (!test_bit(ATH10K_FLAG_FIRST_BOOT_DONE, &ar->dev_flags))
-		ath10k_info("pci irq %s\n", irq_mode);
+		ath10k_info("pci irq %s irq_mode %d reset_mode %d\n",
+			    irq_mode, ath10k_pci_irq_mode,
+			    ath10k_pci_reset_mode);
 
 	return 0;
 
@@ -1952,23 +2023,52 @@ err:
 	return ret;
 }
 
+static int ath10k_pci_hif_power_up_warm(struct ath10k *ar)
+{
+	int i, ret;
+
+	/*
+	 * Sometime warm reset succeeds after retries.
+	 *
+	 * FIXME: It might be possible to tune ath10k_pci_warm_reset() to work
+	 * at first try.
+	 */
+	for (i = 0; i < ATH10K_PCI_NUM_WARM_RESET_ATTEMPTS; i++) {
+		ret = __ath10k_pci_hif_power_up(ar, false);
+		if (ret == 0)
+			break;
+
+		ath10k_warn("failed to warm reset (attempt %d out of %d): %d\n",
+			    i + 1, ATH10K_PCI_NUM_WARM_RESET_ATTEMPTS, ret);
+	}
+
+	return ret;
+}
+
 static int ath10k_pci_hif_power_up(struct ath10k *ar)
 {
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot hif power up\n");
+
 	/*
 	 * Hardware CUS232 version 2 has some issues with cold reset and the
 	 * preferred (and safer) way to perform a device reset is through a
 	 * warm reset.
 	 *
-	 * Warm reset doesn't always work though (notably after a firmware
-	 * crash) so fall back to cold reset if necessary.
+	 * Warm reset doesn't always work though so fall back to cold reset may
+	 * be necessary.
 	 */
-	ret = __ath10k_pci_hif_power_up(ar, false);
+	ret = ath10k_pci_hif_power_up_warm(ar);
 	if (ret) {
-		ath10k_warn("failed to power up target using warm reset (%d), trying cold reset\n",
+		ath10k_warn("failed to power up target using warm reset: %d\n",
 			    ret);
 
+		if (ath10k_pci_reset_mode == ATH10K_PCI_RESET_WARM_ONLY)
+			return ret;
+
+		ath10k_warn("trying cold reset\n");
+
 		ret = __ath10k_pci_hif_power_up(ar, true);
 		if (ret) {
 			ath10k_err("failed to power up target using cold reset too (%d)\n",
@@ -1984,12 +2084,14 @@ static void ath10k_pci_hif_power_down(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot hif power down\n");
+
 	ath10k_pci_free_early_irq(ar);
 	ath10k_pci_kill_tasklet(ar);
 	ath10k_pci_deinit_irq(ar);
+	ath10k_pci_ce_deinit(ar);
 	ath10k_pci_warm_reset(ar);
 
-	ath10k_pci_ce_deinit(ar);
 	if (!test_bit(ATH10K_PCI_FEATURE_SOC_POWER_SAVE, ar_pci->features))
 		ath10k_do_pci_sleep(ar);
 }
@@ -2137,7 +2239,6 @@ static irqreturn_t ath10k_pci_interrupt_handler(int irq, void *arg)
 static void ath10k_pci_early_irq_tasklet(unsigned long data)
 {
 	struct ath10k *ar = (struct ath10k *)data;
-	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	u32 fw_ind;
 	int ret;
 
@@ -2148,14 +2249,11 @@ static void ath10k_pci_early_irq_tasklet(unsigned long data)
 		return;
 	}
 
-	fw_ind = ath10k_pci_read32(ar, ar_pci->fw_indicator_address);
+	fw_ind = ath10k_pci_read32(ar, FW_INDICATOR_ADDRESS);
 	if (fw_ind & FW_IND_EVENT_PENDING) {
-		ath10k_pci_write32(ar, ar_pci->fw_indicator_address,
+		ath10k_pci_write32(ar, FW_INDICATOR_ADDRESS,
 				   fw_ind & ~FW_IND_EVENT_PENDING);
-
-		/* Some structures are unavailable during early boot or at
-		 * driver teardown so just print that the device has crashed. */
-		ath10k_warn("device crashed - no diagnostics available\n");
+		ath10k_pci_hif_dump_area(ar);
 	}
 
 	ath10k_pci_sleep(ar);
@@ -2385,33 +2483,69 @@ static int ath10k_pci_deinit_irq(struct ath10k *ar)
 static int ath10k_pci_wait_for_target_init(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
-	int wait_limit = 300; /* 3 sec */
+	unsigned long timeout;
 	int ret;
+	u32 val;
+
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot waiting target to initialise\n");
 
 	ret = ath10k_pci_wake(ar);
 	if (ret) {
-		ath10k_err("failed to wake up target: %d\n", ret);
+		ath10k_err("failed to wake up target for init: %d\n", ret);
 		return ret;
 	}
 
-	while (wait_limit-- &&
-	       !(ioread32(ar_pci->mem + FW_INDICATOR_ADDRESS) &
-		 FW_IND_INITIALIZED)) {
+	timeout = jiffies + msecs_to_jiffies(ATH10K_PCI_TARGET_WAIT);
+
+	do {
+		val = ath10k_pci_read32(ar, FW_INDICATOR_ADDRESS);
+
+		ath10k_dbg(ATH10K_DBG_BOOT, "boot target indicator %x\n", val);
+
+		/* target should never return this */
+		if (val == 0xffffffff)
+			continue;
+
+		/* the device has crashed so don't bother trying anymore */
+		if (val & FW_IND_EVENT_PENDING)
+			break;
+
+		if (val & FW_IND_INITIALIZED)
+			break;
+
 		if (ar_pci->num_msi_intrs == 0)
 			/* Fix potential race by repeating CORE_BASE writes */
-			iowrite32(PCIE_INTR_FIRMWARE_MASK |
-				  PCIE_INTR_CE_MASK_ALL,
-				  ar_pci->mem + (SOC_CORE_BASE_ADDRESS |
-						 PCIE_INTR_ENABLE_ADDRESS));
+			ath10k_pci_soc_write32(ar, PCIE_INTR_ENABLE_ADDRESS,
+					       PCIE_INTR_FIRMWARE_MASK |
+					       PCIE_INTR_CE_MASK_ALL);
+
 		mdelay(10);
-	}
+	} while (time_before(jiffies, timeout));
 
-	if (wait_limit < 0) {
-		ath10k_err("target stalled\n");
+	if (val == 0xffffffff) {
+		ath10k_err("failed to read device register, device is gone\n");
 		ret = -EIO;
 		goto out;
 	}
 
+	if (val & FW_IND_EVENT_PENDING) {
+		ath10k_warn("device has crashed during init\n");
+		ath10k_pci_write32(ar, FW_INDICATOR_ADDRESS,
+				   val & ~FW_IND_EVENT_PENDING);
+		ath10k_pci_hif_dump_area(ar);
+		ret = -ECOMM;
+		goto out;
+	}
+
+	if (!(val & FW_IND_INITIALIZED)) {
+		ath10k_err("failed to receive initialized event from target: %08x\n",
+			   val);
+		ret = -ETIMEDOUT;
+		goto out;
+	}
+
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot target initialised\n");
+
 out:
 	ath10k_pci_sleep(ar);
 	return ret;
@@ -2422,6 +2556,8 @@ static int ath10k_pci_cold_reset(struct ath10k *ar)
 	int i, ret;
 	u32 val;
 
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot cold reset\n");
+
 	ret = ath10k_do_pci_wake(ar);
 	if (ret) {
 		ath10k_err("failed to wake up target: %d\n",
@@ -2453,6 +2589,9 @@ static int ath10k_pci_cold_reset(struct ath10k *ar)
 	}
 
 	ath10k_do_pci_sleep(ar);
+
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot cold reset complete\n");
+
 	return 0;
 }
 
@@ -2484,7 +2623,7 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 	struct ath10k_pci *ar_pci;
 	u32 lcr_val, chip_id;
 
-	ath10k_dbg(ATH10K_DBG_PCI, "%s\n", __func__);
+	ath10k_dbg(ATH10K_DBG_PCI, "pci probe\n");
 
 	ar_pci = kzalloc(sizeof(*ar_pci), GFP_KERNEL);
 	if (ar_pci == NULL)
@@ -2503,7 +2642,7 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 		goto err_ar_pci;
 	}
 
-	if (ath10k_target_ps)
+	if (ath10k_pci_target_ps)
 		set_bit(ATH10K_PCI_FEATURE_SOC_POWER_SAVE, ar_pci->features);
 
 	ath10k_pci_dump_features(ar_pci);
@@ -2516,23 +2655,10 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 	}
 
 	ar_pci->ar = ar;
-	ar_pci->fw_indicator_address = FW_INDICATOR_ADDRESS;
 	atomic_set(&ar_pci->keep_awake_count, 0);
 
 	pci_set_drvdata(pdev, ar);
 
-	/*
-	 * Without any knowledge of the Host, the Target may have been reset or
-	 * power cycled and its Config Space may no longer reflect the PCI
-	 * address space that was assigned earlier by the PCI infrastructure.
-	 * Refresh it now.
-	 */
-	ret = pci_assign_resource(pdev, BAR_NUM);
-	if (ret) {
-		ath10k_err("failed to assign PCI space: %d\n", ret);
-		goto err_ar;
-	}
-
 	ret = pci_enable_device(pdev);
 	if (ret) {
 		ath10k_err("failed to enable PCI device: %d\n", ret);
@@ -2594,16 +2720,24 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 
 	ath10k_do_pci_sleep(ar);
 
+	ret = ath10k_pci_alloc_ce(ar);
+	if (ret) {
+		ath10k_err("failed to allocate copy engine pipes: %d\n", ret);
+		goto err_iomap;
+	}
+
 	ath10k_dbg(ATH10K_DBG_BOOT, "boot pci_mem 0x%p\n", ar_pci->mem);
 
 	ret = ath10k_core_register(ar, chip_id);
 	if (ret) {
 		ath10k_err("failed to register driver core: %d\n", ret);
-		goto err_iomap;
+		goto err_free_ce;
 	}
 
 	return 0;
 
+err_free_ce:
+	ath10k_pci_free_ce(ar);
 err_iomap:
 	pci_iounmap(pdev, mem);
 err_master:
@@ -2626,7 +2760,7 @@ static void ath10k_pci_remove(struct pci_dev *pdev)
 	struct ath10k *ar = pci_get_drvdata(pdev);
 	struct ath10k_pci *ar_pci;
 
-	ath10k_dbg(ATH10K_DBG_PCI, "%s\n", __func__);
+	ath10k_dbg(ATH10K_DBG_PCI, "pci remove\n");
 
 	if (!ar)
 		return;
@@ -2636,9 +2770,8 @@ static void ath10k_pci_remove(struct pci_dev *pdev)
 	if (!ar_pci)
 		return;
 
-	tasklet_kill(&ar_pci->msi_fw_err);
-
 	ath10k_core_unregister(ar);
+	ath10k_pci_free_ce(ar);
 
 	pci_iounmap(pdev, ar_pci->mem);
 	pci_release_region(pdev, BAR_NUM);
@@ -2680,6 +2813,5 @@ module_exit(ath10k_pci_exit);
 MODULE_AUTHOR("Qualcomm Atheros");
 MODULE_DESCRIPTION("Driver support for Atheros QCA988X PCIe devices");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" QCA988X_HW_2_0_FW_FILE);
-MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" QCA988X_HW_2_0_OTP_FILE);
+MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" QCA988X_HW_2_0_FW_2_FILE);
 MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" QCA988X_HW_2_0_BOARD_DATA_FILE);
diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h
index b43fdb4f7319..dfdebb4157aa 100644
--- a/drivers/net/wireless/ath/ath10k/pci.h
+++ b/drivers/net/wireless/ath/ath10k/pci.h
@@ -189,9 +189,6 @@ struct ath10k_pci {
 
 	struct ath10k_hif_cb msg_callbacks_current;
 
-	/* Target address used to signal a pending firmware event */
-	u32 fw_indicator_address;
-
 	/* Copy Engine used for Diagnostic Accesses */
 	struct ath10k_ce_pipe *ce_diag;
 
diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index 0541dd939ce9..82669a77e553 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -100,189 +100,6 @@ exit:
 		wake_up(&htt->empty_tx_wq);
 }
 
-static const u8 rx_legacy_rate_idx[] = {
-	3,	/* 0x00  - 11Mbps  */
-	2,	/* 0x01  - 5.5Mbps */
-	1,	/* 0x02  - 2Mbps   */
-	0,	/* 0x03  - 1Mbps   */
-	3,	/* 0x04  - 11Mbps  */
-	2,	/* 0x05  - 5.5Mbps */
-	1,	/* 0x06  - 2Mbps   */
-	0,	/* 0x07  - 1Mbps   */
-	10,	/* 0x08  - 48Mbps  */
-	8,	/* 0x09  - 24Mbps  */
-	6,	/* 0x0A  - 12Mbps  */
-	4,	/* 0x0B  - 6Mbps   */
-	11,	/* 0x0C  - 54Mbps  */
-	9,	/* 0x0D  - 36Mbps  */
-	7,	/* 0x0E  - 18Mbps  */
-	5,	/* 0x0F  - 9Mbps   */
-};
-
-static void process_rx_rates(struct ath10k *ar, struct htt_rx_info *info,
-			     enum ieee80211_band band,
-			     struct ieee80211_rx_status *status)
-{
-	u8 cck, rate, rate_idx, bw, sgi, mcs, nss;
-	u8 info0 = info->rate.info0;
-	u32 info1 = info->rate.info1;
-	u32 info2 = info->rate.info2;
-	u8 preamble = 0;
-
-	/* Check if valid fields */
-	if (!(info0 & HTT_RX_INDICATION_INFO0_START_VALID))
-		return;
-
-	preamble = MS(info1, HTT_RX_INDICATION_INFO1_PREAMBLE_TYPE);
-
-	switch (preamble) {
-	case HTT_RX_LEGACY:
-		cck = info0 & HTT_RX_INDICATION_INFO0_LEGACY_RATE_CCK;
-		rate = MS(info0, HTT_RX_INDICATION_INFO0_LEGACY_RATE);
-		rate_idx = 0;
-
-		if (rate < 0x08 || rate > 0x0F)
-			break;
-
-		switch (band) {
-		case IEEE80211_BAND_2GHZ:
-			if (cck)
-				rate &= ~BIT(3);
-			rate_idx = rx_legacy_rate_idx[rate];
-			break;
-		case IEEE80211_BAND_5GHZ:
-			rate_idx = rx_legacy_rate_idx[rate];
-			/* We are using same rate table registering
-			   HW - ath10k_rates[]. In case of 5GHz skip
-			   CCK rates, so -4 here */
-			rate_idx -= 4;
-			break;
-		default:
-			break;
-		}
-
-		status->rate_idx = rate_idx;
-		break;
-	case HTT_RX_HT:
-	case HTT_RX_HT_WITH_TXBF:
-		/* HT-SIG - Table 20-11 in info1 and info2 */
-		mcs = info1 & 0x1F;
-		nss = mcs >> 3;
-		bw = (info1 >> 7) & 1;
-		sgi = (info2 >> 7) & 1;
-
-		status->rate_idx = mcs;
-		status->flag |= RX_FLAG_HT;
-		if (sgi)
-			status->flag |= RX_FLAG_SHORT_GI;
-		if (bw)
-			status->flag |= RX_FLAG_40MHZ;
-		break;
-	case HTT_RX_VHT:
-	case HTT_RX_VHT_WITH_TXBF:
-		/* VHT-SIG-A1 in info 1, VHT-SIG-A2 in info2
-		   TODO check this */
-		mcs = (info2 >> 4) & 0x0F;
-		nss = ((info1 >> 10) & 0x07) + 1;
-		bw = info1 & 3;
-		sgi = info2 & 1;
-
-		status->rate_idx = mcs;
-		status->vht_nss = nss;
-
-		if (sgi)
-			status->flag |= RX_FLAG_SHORT_GI;
-
-		switch (bw) {
-		/* 20MHZ */
-		case 0:
-			break;
-		/* 40MHZ */
-		case 1:
-			status->flag |= RX_FLAG_40MHZ;
-			break;
-		/* 80MHZ */
-		case 2:
-			status->vht_flag |= RX_VHT_FLAG_80MHZ;
-		}
-
-		status->flag |= RX_FLAG_VHT;
-		break;
-	default:
-		break;
-	}
-}
-
-void ath10k_process_rx(struct ath10k *ar, struct htt_rx_info *info)
-{
-	struct ieee80211_rx_status *status;
-	struct ieee80211_channel *ch;
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)info->skb->data;
-
-	status = IEEE80211_SKB_RXCB(info->skb);
-	memset(status, 0, sizeof(*status));
-
-	if (info->encrypt_type != HTT_RX_MPDU_ENCRYPT_NONE) {
-		status->flag |= RX_FLAG_DECRYPTED | RX_FLAG_IV_STRIPPED |
-				RX_FLAG_MMIC_STRIPPED;
-		hdr->frame_control = __cpu_to_le16(
-				__le16_to_cpu(hdr->frame_control) &
-				~IEEE80211_FCTL_PROTECTED);
-	}
-
-	if (info->mic_err)
-		status->flag |= RX_FLAG_MMIC_ERROR;
-
-	if (info->fcs_err)
-		status->flag |= RX_FLAG_FAILED_FCS_CRC;
-
-	if (info->amsdu_more)
-		status->flag |= RX_FLAG_AMSDU_MORE;
-
-	status->signal = info->signal;
-
-	spin_lock_bh(&ar->data_lock);
-	ch = ar->scan_channel;
-	if (!ch)
-		ch = ar->rx_channel;
-	spin_unlock_bh(&ar->data_lock);
-
-	if (!ch) {
-		ath10k_warn("no channel configured; ignoring frame!\n");
-		dev_kfree_skb_any(info->skb);
-		return;
-	}
-
-	process_rx_rates(ar, info, ch->band, status);
-	status->band = ch->band;
-	status->freq = ch->center_freq;
-
-	if (info->rate.info0 & HTT_RX_INDICATION_INFO0_END_VALID) {
-		/* TSF available only in 32-bit */
-		status->mactime = info->tsf & 0xffffffff;
-		status->flag |= RX_FLAG_MACTIME_END;
-	}
-
-	ath10k_dbg(ATH10K_DBG_DATA,
-		   "rx skb %p len %u %s%s%s%s%s %srate_idx %u vht_nss %u freq %u band %u flag 0x%x fcs-err %i\n",
-		   info->skb,
-		   info->skb->len,
-		   status->flag == 0 ? "legacy" : "",
-		   status->flag & RX_FLAG_HT ? "ht" : "",
-		   status->flag & RX_FLAG_VHT ? "vht" : "",
-		   status->flag & RX_FLAG_40MHZ ? "40" : "",
-		   status->vht_flag & RX_VHT_FLAG_80MHZ ? "80" : "",
-		   status->flag & RX_FLAG_SHORT_GI ? "sgi " : "",
-		   status->rate_idx,
-		   status->vht_nss,
-		   status->freq,
-		   status->band, status->flag, info->fcs_err);
-	ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "rx skb: ",
-			info->skb->data, info->skb->len);
-
-	ieee80211_rx(ar->hw, info->skb);
-}
-
 struct ath10k_peer *ath10k_peer_find(struct ath10k *ar, int vdev_id,
 				     const u8 *addr)
 {
diff --git a/drivers/net/wireless/ath/ath10k/txrx.h b/drivers/net/wireless/ath/ath10k/txrx.h
index 356dc9c04c9e..aee3e20058f8 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.h
+++ b/drivers/net/wireless/ath/ath10k/txrx.h
@@ -21,7 +21,6 @@
 
 void ath10k_txrx_tx_unref(struct ath10k_htt *htt,
 			  const struct htt_tx_done *tx_done);
-void ath10k_process_rx(struct ath10k *ar, struct htt_rx_info *info);
 
 struct ath10k_peer *ath10k_peer_find(struct ath10k *ar, int vdev_id,
 				     const u8 *addr);
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index cb1f7b5bcf4c..4b7782a529ac 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -639,6 +639,7 @@ int ath10k_wmi_mgmt_tx(struct ath10k *ar, struct sk_buff *skb)
 	struct sk_buff *wmi_skb;
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int len;
+	u32 buf_len = skb->len;
 	u16 fc;
 
 	hdr = (struct ieee80211_hdr *)skb->data;
@@ -648,6 +649,15 @@ int ath10k_wmi_mgmt_tx(struct ath10k *ar, struct sk_buff *skb)
 		return -EINVAL;
 
 	len = sizeof(cmd->hdr) + skb->len;
+
+	if ((ieee80211_is_action(hdr->frame_control) ||
+	     ieee80211_is_deauth(hdr->frame_control) ||
+	     ieee80211_is_disassoc(hdr->frame_control)) &&
+	     ieee80211_has_protected(hdr->frame_control)) {
+		len += IEEE80211_CCMP_MIC_LEN;
+		buf_len += IEEE80211_CCMP_MIC_LEN;
+	}
+
 	len = round_up(len, 4);
 
 	wmi_skb = ath10k_wmi_alloc_skb(len);
@@ -659,7 +669,7 @@ int ath10k_wmi_mgmt_tx(struct ath10k *ar, struct sk_buff *skb)
 	cmd->hdr.vdev_id = __cpu_to_le32(ATH10K_SKB_CB(skb)->vdev_id);
 	cmd->hdr.tx_rate = 0;
 	cmd->hdr.tx_power = 0;
-	cmd->hdr.buf_len = __cpu_to_le32((u32)(skb->len));
+	cmd->hdr.buf_len = __cpu_to_le32(buf_len);
 
 	memcpy(cmd->hdr.peer_macaddr.addr, ieee80211_get_DA(hdr), ETH_ALEN);
 	memcpy(cmd->buf, skb->data, skb->len);
@@ -957,10 +967,16 @@ static int ath10k_wmi_event_mgmt_rx(struct ath10k *ar, struct sk_buff *skb)
 	 * frames with Protected Bit set. */
 	if (ieee80211_has_protected(hdr->frame_control) &&
 	    !ieee80211_is_auth(hdr->frame_control)) {
-		status->flag |= RX_FLAG_DECRYPTED | RX_FLAG_IV_STRIPPED |
-				RX_FLAG_MMIC_STRIPPED;
-		hdr->frame_control = __cpu_to_le16(fc &
+		status->flag |= RX_FLAG_DECRYPTED;
+
+		if (!ieee80211_is_action(hdr->frame_control) &&
+		    !ieee80211_is_deauth(hdr->frame_control) &&
+		    !ieee80211_is_disassoc(hdr->frame_control)) {
+			status->flag |= RX_FLAG_IV_STRIPPED |
+					RX_FLAG_MMIC_STRIPPED;
+			hdr->frame_control = __cpu_to_le16(fc &
 					~IEEE80211_FCTL_PROTECTED);
+		}
 	}
 
 	ath10k_dbg(ATH10K_DBG_MGMT,
@@ -1362,13 +1378,10 @@ static void ath10k_wmi_event_host_swba(struct ath10k *ar, struct sk_buff *skb)
 	struct sk_buff *bcn;
 	int ret, vdev_id = 0;
 
-	ath10k_dbg(ATH10K_DBG_MGMT, "WMI_HOST_SWBA_EVENTID\n");
-
 	ev = (struct wmi_host_swba_event *)skb->data;
 	map = __le32_to_cpu(ev->vdev_map);
 
-	ath10k_dbg(ATH10K_DBG_MGMT, "host swba:\n"
-		   "-vdev map 0x%x\n",
+	ath10k_dbg(ATH10K_DBG_MGMT, "mgmt swba vdev_map 0x%x\n",
 		   ev->vdev_map);
 
 	for (; map; map >>= 1, vdev_id++) {
@@ -1385,12 +1398,7 @@ static void ath10k_wmi_event_host_swba(struct ath10k *ar, struct sk_buff *skb)
 		bcn_info = &ev->bcn_info[i];
 
 		ath10k_dbg(ATH10K_DBG_MGMT,
-			   "-bcn_info[%d]:\n"
-			   "--tim_len %d\n"
-			   "--tim_mcast %d\n"
-			   "--tim_changed %d\n"
-			   "--tim_num_ps_pending %d\n"
-			   "--tim_bitmap 0x%08x%08x%08x%08x\n",
+			   "mgmt event bcn_info %d tim_len %d mcast %d changed %d num_ps_pending %d bitmap 0x%08x%08x%08x%08x\n",
 			   i,
 			   __le32_to_cpu(bcn_info->tim_info.tim_len),
 			   __le32_to_cpu(bcn_info->tim_info.tim_mcast),
@@ -1439,6 +1447,7 @@ static void ath10k_wmi_event_host_swba(struct ath10k *ar, struct sk_buff *skb)
 					 ATH10K_SKB_CB(arvif->beacon)->paddr,
 					 arvif->beacon->len, DMA_TO_DEVICE);
 			dev_kfree_skb_any(arvif->beacon);
+			arvif->beacon = NULL;
 		}
 
 		ATH10K_SKB_CB(bcn)->paddr = dma_map_single(arvif->ar->dev,
@@ -1448,6 +1457,7 @@ static void ath10k_wmi_event_host_swba(struct ath10k *ar, struct sk_buff *skb)
 					ATH10K_SKB_CB(bcn)->paddr);
 		if (ret) {
 			ath10k_warn("failed to map beacon: %d\n", ret);
+			dev_kfree_skb_any(bcn);
 			goto skip;
 		}
 
@@ -2365,7 +2375,7 @@ void ath10k_wmi_detach(struct ath10k *ar)
 	ar->wmi.num_mem_chunks = 0;
 }
 
-int ath10k_wmi_connect_htc_service(struct ath10k *ar)
+int ath10k_wmi_connect(struct ath10k *ar)
 {
 	int status;
 	struct ath10k_htc_svc_conn_req conn_req;
@@ -2393,8 +2403,9 @@ int ath10k_wmi_connect_htc_service(struct ath10k *ar)
 	return 0;
 }
 
-int ath10k_wmi_pdev_set_regdomain(struct ath10k *ar, u16 rd, u16 rd2g,
-				  u16 rd5g, u16 ctl2g, u16 ctl5g)
+static int ath10k_wmi_main_pdev_set_regdomain(struct ath10k *ar, u16 rd,
+					      u16 rd2g, u16 rd5g, u16 ctl2g,
+					      u16 ctl5g)
 {
 	struct wmi_pdev_set_regdomain_cmd *cmd;
 	struct sk_buff *skb;
@@ -2418,6 +2429,46 @@ int ath10k_wmi_pdev_set_regdomain(struct ath10k *ar, u16 rd, u16 rd2g,
 				   ar->wmi.cmd->pdev_set_regdomain_cmdid);
 }
 
+static int ath10k_wmi_10x_pdev_set_regdomain(struct ath10k *ar, u16 rd,
+					     u16 rd2g, u16 rd5g,
+					     u16 ctl2g, u16 ctl5g,
+					     enum wmi_dfs_region dfs_reg)
+{
+	struct wmi_pdev_set_regdomain_cmd_10x *cmd;
+	struct sk_buff *skb;
+
+	skb = ath10k_wmi_alloc_skb(sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_set_regdomain_cmd_10x *)skb->data;
+	cmd->reg_domain = __cpu_to_le32(rd);
+	cmd->reg_domain_2G = __cpu_to_le32(rd2g);
+	cmd->reg_domain_5G = __cpu_to_le32(rd5g);
+	cmd->conformance_test_limit_2G = __cpu_to_le32(ctl2g);
+	cmd->conformance_test_limit_5G = __cpu_to_le32(ctl5g);
+	cmd->dfs_domain = __cpu_to_le32(dfs_reg);
+
+	ath10k_dbg(ATH10K_DBG_WMI,
+		   "wmi pdev regdomain rd %x rd2g %x rd5g %x ctl2g %x ctl5g %x dfs_region %x\n",
+		   rd, rd2g, rd5g, ctl2g, ctl5g, dfs_reg);
+
+	return ath10k_wmi_cmd_send(ar, skb,
+				   ar->wmi.cmd->pdev_set_regdomain_cmdid);
+}
+
+int ath10k_wmi_pdev_set_regdomain(struct ath10k *ar, u16 rd, u16 rd2g,
+				  u16 rd5g, u16 ctl2g, u16 ctl5g,
+				  enum wmi_dfs_region dfs_reg)
+{
+	if (test_bit(ATH10K_FW_FEATURE_WMI_10X, ar->fw_features))
+		return ath10k_wmi_10x_pdev_set_regdomain(ar, rd, rd2g, rd5g,
+							ctl2g, ctl5g, dfs_reg);
+	else
+		return ath10k_wmi_main_pdev_set_regdomain(ar, rd, rd2g, rd5g,
+							 ctl2g, ctl5g);
+}
+
 int ath10k_wmi_pdev_set_channel(struct ath10k *ar,
 				const struct wmi_channel_arg *arg)
 {
@@ -3456,8 +3507,9 @@ int ath10k_wmi_peer_assoc(struct ath10k *ar,
 		__cpu_to_le32(arg->peer_vht_rates.tx_mcs_set);
 
 	ath10k_dbg(ATH10K_DBG_WMI,
-		   "wmi peer assoc vdev %d addr %pM\n",
-		   arg->vdev_id, arg->addr);
+		   "wmi peer assoc vdev %d addr %pM (%s)\n",
+		   arg->vdev_id, arg->addr,
+		   arg->peer_reassoc ? "reassociate" : "new");
 	return ath10k_wmi_cmd_send(ar, skb, ar->wmi.cmd->peer_assoc_cmdid);
 }
 
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index f51d5ca0141f..e93df2c10413 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -198,16 +198,6 @@ struct wmi_mac_addr {
 	} __packed;
 } __packed;
 
-/* macro to convert MAC address from WMI word format to char array */
-#define WMI_MAC_ADDR_TO_CHAR_ARRAY(pwmi_mac_addr, c_macaddr) do { \
-	(c_macaddr)[0] =  ((pwmi_mac_addr)->word0) & 0xff; \
-	(c_macaddr)[1] = (((pwmi_mac_addr)->word0) >> 8) & 0xff; \
-	(c_macaddr)[2] = (((pwmi_mac_addr)->word0) >> 16) & 0xff; \
-	(c_macaddr)[3] = (((pwmi_mac_addr)->word0) >> 24) & 0xff; \
-	(c_macaddr)[4] =  ((pwmi_mac_addr)->word1) & 0xff; \
-	(c_macaddr)[5] = (((pwmi_mac_addr)->word1) >> 8) & 0xff; \
-	} while (0)
-
 struct wmi_cmd_map {
 	u32 init_cmdid;
 	u32 start_scan_cmdid;
@@ -2185,6 +2175,31 @@ struct wmi_pdev_set_regdomain_cmd {
 	__le32 conformance_test_limit_5G;
 } __packed;
 
+enum wmi_dfs_region {
+	/* Uninitialized dfs domain */
+	WMI_UNINIT_DFS_DOMAIN = 0,
+
+	/* FCC3 dfs domain */
+	WMI_FCC_DFS_DOMAIN = 1,
+
+	/* ETSI dfs domain */
+	WMI_ETSI_DFS_DOMAIN = 2,
+
+	/*Japan dfs domain */
+	WMI_MKK4_DFS_DOMAIN = 3,
+};
+
+struct wmi_pdev_set_regdomain_cmd_10x {
+	__le32 reg_domain;
+	__le32 reg_domain_2G;
+	__le32 reg_domain_5G;
+	__le32 conformance_test_limit_2G;
+	__le32 conformance_test_limit_5G;
+
+	/* dfs domain from wmi_dfs_region */
+	__le32 dfs_domain;
+} __packed;
+
 /* Command to set/unset chip in quiet mode */
 struct wmi_pdev_set_quiet_cmd {
 	/* period in TUs */
@@ -2210,6 +2225,19 @@ enum ath10k_protmode {
 	ATH10K_PROT_RTSCTS   = 2,    /* RTS-CTS */
 };
 
+enum wmi_rtscts_profile {
+	WMI_RTSCTS_FOR_NO_RATESERIES = 0,
+	WMI_RTSCTS_FOR_SECOND_RATESERIES,
+	WMI_RTSCTS_ACROSS_SW_RETRIES
+};
+
+#define WMI_RTSCTS_ENABLED		1
+#define WMI_RTSCTS_SET_MASK		0x0f
+#define WMI_RTSCTS_SET_LSB		0
+
+#define WMI_RTSCTS_PROFILE_MASK		0xf0
+#define WMI_RTSCTS_PROFILE_LSB		4
+
 enum wmi_beacon_gen_mode {
 	WMI_BEACON_STAGGERED_MODE = 0,
 	WMI_BEACON_BURST_MODE = 1
@@ -2295,9 +2323,9 @@ struct wmi_pdev_param_map {
 #define WMI_PDEV_PARAM_UNSUPPORTED 0
 
 enum wmi_pdev_param {
-	/* TX chian mask */
+	/* TX chain mask */
 	WMI_PDEV_PARAM_TX_CHAIN_MASK = 0x1,
-	/* RX chian mask */
+	/* RX chain mask */
 	WMI_PDEV_PARAM_RX_CHAIN_MASK,
 	/* TX power limit for 2G Radio */
 	WMI_PDEV_PARAM_TXPOWER_LIMIT2G,
@@ -2682,6 +2710,9 @@ struct wal_dbg_tx_stats {
 	/* wal pdev resets  */
 	__le32 pdev_resets;
 
+	/* frames dropped due to non-availability of stateless TIDs */
+	__le32 stateless_tid_alloc_failure;
+
 	__le32 phy_underrun;
 
 	/* MPDU is more than txop limit */
@@ -2738,13 +2769,21 @@ enum wmi_stats_id {
 	WMI_REQUEST_AP_STAT	= 0x02
 };
 
+struct wlan_inst_rssi_args {
+	__le16 cfg_retry_count;
+	__le16 retry_count;
+};
+
 struct wmi_request_stats_cmd {
 	__le32 stats_id;
 
-	/*
-	 * Space to add parameters like
-	 * peer mac addr
-	 */
+	__le32 vdev_id;
+
+	/* peer MAC address */
+	struct wmi_mac_addr peer_macaddr;
+
+	/* Instantaneous RSSI arguments */
+	struct wlan_inst_rssi_args inst_rssi_args;
 } __packed;
 
 /* Suspend option */
@@ -2795,7 +2834,18 @@ struct wmi_stats_event {
  * PDEV statistics
  * TODO: add all PDEV stats here
  */
-struct wmi_pdev_stats {
+struct wmi_pdev_stats_old {
+	__le32 chan_nf;        /* Channel noise floor */
+	__le32 tx_frame_count; /* TX frame count */
+	__le32 rx_frame_count; /* RX frame count */
+	__le32 rx_clear_count; /* rx clear count */
+	__le32 cycle_count;    /* cycle count */
+	__le32 phy_err_count;  /* Phy error count */
+	__le32 chan_tx_pwr;    /* channel tx power */
+	struct wal_dbg_stats wal; /* WAL dbg stats */
+} __packed;
+
+struct wmi_pdev_stats_10x {
 	__le32 chan_nf;        /* Channel noise floor */
 	__le32 tx_frame_count; /* TX frame count */
 	__le32 rx_frame_count; /* RX frame count */
@@ -2804,6 +2854,12 @@ struct wmi_pdev_stats {
 	__le32 phy_err_count;  /* Phy error count */
 	__le32 chan_tx_pwr;    /* channel tx power */
 	struct wal_dbg_stats wal; /* WAL dbg stats */
+	__le32 ack_rx_bad;
+	__le32 rts_bad;
+	__le32 rts_good;
+	__le32 fcs_bad;
+	__le32 no_beacons;
+	__le32 mib_int_count;
 } __packed;
 
 /*
@@ -2818,10 +2874,17 @@ struct wmi_vdev_stats {
  * peer statistics.
  * TODO: add more stats
  */
-struct wmi_peer_stats {
+struct wmi_peer_stats_old {
+	struct wmi_mac_addr peer_macaddr;
+	__le32 peer_rssi;
+	__le32 peer_tx_rate;
+} __packed;
+
+struct wmi_peer_stats_10x {
 	struct wmi_mac_addr peer_macaddr;
 	__le32 peer_rssi;
 	__le32 peer_tx_rate;
+	__le32 peer_rx_rate;
 } __packed;
 
 struct wmi_vdev_create_cmd {
@@ -4196,13 +4259,14 @@ void ath10k_wmi_detach(struct ath10k *ar);
 int ath10k_wmi_wait_for_service_ready(struct ath10k *ar);
 int ath10k_wmi_wait_for_unified_ready(struct ath10k *ar);
 
-int ath10k_wmi_connect_htc_service(struct ath10k *ar);
+int ath10k_wmi_connect(struct ath10k *ar);
 int ath10k_wmi_pdev_set_channel(struct ath10k *ar,
 				const struct wmi_channel_arg *);
 int ath10k_wmi_pdev_suspend_target(struct ath10k *ar, u32 suspend_opt);
 int ath10k_wmi_pdev_resume_target(struct ath10k *ar);
 int ath10k_wmi_pdev_set_regdomain(struct ath10k *ar, u16 rd, u16 rd2g,
-				  u16 rd5g, u16 ctl2g, u16 ctl5g);
+				  u16 rd5g, u16 ctl2g, u16 ctl5g,
+				  enum wmi_dfs_region dfs_reg);
 int ath10k_wmi_pdev_set_param(struct ath10k *ar, u32 id, u32 value);
 int ath10k_wmi_cmd_init(struct ath10k *ar);
 int ath10k_wmi_start_scan(struct ath10k *ar, const struct wmi_start_scan_arg *);
diff --git a/drivers/net/wireless/ath/ath5k/phy.c b/drivers/net/wireless/ath/ath5k/phy.c
index 1a2973b7acf2..0fce1c76638e 100644
--- a/drivers/net/wireless/ath/ath5k/phy.c
+++ b/drivers/net/wireless/ath/ath5k/phy.c
@@ -3709,8 +3709,8 @@ ath5k_hw_txpower(struct ath5k_hw *ah, struct ieee80211_channel *channel,
 			AR5K_REG_MS(AR5K_TUNE_MAX_TXPOWER, AR5K_TPC_CHIRP),
 			AR5K_TPC);
 	} else {
-		ath5k_hw_reg_write(ah, AR5K_PHY_TXPOWER_RATE_MAX |
-			AR5K_TUNE_MAX_TXPOWER, AR5K_PHY_TXPOWER_RATE_MAX);
+		ath5k_hw_reg_write(ah, AR5K_TUNE_MAX_TXPOWER,
+			AR5K_PHY_TXPOWER_RATE_MAX);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/ath/ath6kl/Kconfig b/drivers/net/wireless/ath/ath6kl/Kconfig
index e39e5860a2e9..9c125ff083f7 100644
--- a/drivers/net/wireless/ath/ath6kl/Kconfig
+++ b/drivers/net/wireless/ath/ath6kl/Kconfig
@@ -1,11 +1,19 @@
 config ATH6KL
 	tristate "Atheros mobile chipsets support"
+	depends on CFG80211
+        ---help---
+	  This module adds core support for wireless adapters based on
+	  Atheros AR6003 and AR6004 chipsets. You still need separate
+	  bus drivers for USB and SDIO to be able to use real devices.
+
+	  If you choose to build it as a module, it will be called
+	  ath6kl_core. Please note that AR6002 and AR6001 are not
+	  supported by this driver.
 
 config ATH6KL_SDIO
 	tristate "Atheros ath6kl SDIO support"
 	depends on ATH6KL
 	depends on MMC
-	depends on CFG80211
 	---help---
 	  This module adds support for wireless adapters based on
 	  Atheros AR6003 and AR6004 chipsets running over SDIO. If you
@@ -17,25 +25,31 @@ config ATH6KL_USB
 	tristate "Atheros ath6kl USB support"
 	depends on ATH6KL
 	depends on USB
-	depends on CFG80211
 	---help---
 	  This module adds support for wireless adapters based on
-	  Atheros AR6004 chipset running over USB. This is still under
-	  implementation and it isn't functional. If you choose to
-	  build it as a module, it will be called ath6kl_usb.
+	  Atheros AR6004 chipset and chipsets based on it running over
+	  USB. If you choose to build it as a module, it will be
+	  called ath6kl_usb.
 
 config ATH6KL_DEBUG
 	bool "Atheros ath6kl debugging"
 	depends on ATH6KL
 	---help---
-	  Enables debug support
+	  Enables ath6kl debug support, including debug messages
+	  enabled with debug_mask module parameter and debugfs
+	  interface.
+
+	  If unsure, say Y to make it easier to debug problems.
 
 config ATH6KL_TRACING
 	bool "Atheros ath6kl tracing support"
 	depends on ATH6KL
 	depends on EVENT_TRACING
 	---help---
-	  Select this to ath6kl use tracing infrastructure.
+	  Select this to ath6kl use tracing infrastructure which, for
+	  example, can be enabled with help of trace-cmd. All debug
+	  messages and commands are delivered to using individually
+	  enablable trace points.
 
 	  If unsure, say Y to make it easier to debug problems.
 
@@ -47,3 +61,5 @@ config ATH6KL_REGDOMAIN
 	  Enabling this makes it possible to change the regdomain in
 	  the firmware. This can be only enabled if regulatory requirements
 	  are taken into account.
+
+	  If unsure, say N.
diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c
index c2c6f4604958..0e26f4a34fda 100644
--- a/drivers/net/wireless/ath/ath6kl/cfg80211.c
+++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c
@@ -724,8 +724,9 @@ ath6kl_add_bss_if_needed(struct ath6kl_vif *vif,
 			ath6kl_dbg(ATH6KL_DBG_WLAN_CFG,
 				   "added bss %pM to cfg80211\n", bssid);
 		kfree(ie);
-	} else
+	} else {
 		ath6kl_dbg(ATH6KL_DBG_WLAN_CFG, "cfg80211 already has a bss\n");
+	}
 
 	return bss;
 }
@@ -970,7 +971,6 @@ static int ath6kl_set_probed_ssids(struct ath6kl *ar,
 					  ssid_list[i].flag,
 					  ssid_list[i].ssid.ssid_len,
 					  ssid_list[i].ssid.ssid);
-
 	}
 
 	/* Make sure no old entries are left behind */
@@ -1759,7 +1759,7 @@ static bool is_rate_ht40(s32 rate, u8 *mcs, bool *sgi)
 }
 
 static int ath6kl_get_station(struct wiphy *wiphy, struct net_device *dev,
-			      u8 *mac, struct station_info *sinfo)
+			      const u8 *mac, struct station_info *sinfo)
 {
 	struct ath6kl *ar = ath6kl_priv(dev);
 	struct ath6kl_vif *vif = netdev_priv(dev);
@@ -1897,7 +1897,6 @@ static int ath6kl_wow_usr(struct ath6kl *ar, struct ath6kl_vif *vif,
 
 	/* Configure the patterns that we received from the user. */
 	for (i = 0; i < wow->n_patterns; i++) {
-
 		/*
 		 * Convert given nl80211 specific mask value to equivalent
 		 * driver specific mask value and send it to the chip along
@@ -2850,8 +2849,9 @@ static int ath6kl_start_ap(struct wiphy *wiphy, struct net_device *dev,
 	if (p.prwise_crypto_type == 0) {
 		p.prwise_crypto_type = NONE_CRYPT;
 		ath6kl_set_cipher(vif, 0, true);
-	} else if (info->crypto.n_ciphers_pairwise == 1)
+	} else if (info->crypto.n_ciphers_pairwise == 1) {
 		ath6kl_set_cipher(vif, info->crypto.ciphers_pairwise[0], true);
+	}
 
 	switch (info->crypto.cipher_group) {
 	case WLAN_CIPHER_SUITE_WEP40:
@@ -2897,7 +2897,6 @@ static int ath6kl_start_ap(struct wiphy *wiphy, struct net_device *dev,
 	}
 
 	if (info->inactivity_timeout) {
-
 		inactivity_timeout = info->inactivity_timeout;
 
 		if (ar->hw.flags & ATH6KL_HW_AP_INACTIVITY_MINS)
@@ -2975,7 +2974,7 @@ static int ath6kl_stop_ap(struct wiphy *wiphy, struct net_device *dev)
 static const u8 bcast_addr[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
 
 static int ath6kl_del_station(struct wiphy *wiphy, struct net_device *dev,
-			      u8 *mac)
+			      const u8 *mac)
 {
 	struct ath6kl *ar = ath6kl_priv(dev);
 	struct ath6kl_vif *vif = netdev_priv(dev);
@@ -2986,7 +2985,8 @@ static int ath6kl_del_station(struct wiphy *wiphy, struct net_device *dev,
 }
 
 static int ath6kl_change_station(struct wiphy *wiphy, struct net_device *dev,
-				 u8 *mac, struct station_parameters *params)
+				 const u8 *mac,
+				 struct station_parameters *params)
 {
 	struct ath6kl *ar = ath6kl_priv(dev);
 	struct ath6kl_vif *vif = netdev_priv(dev);
diff --git a/drivers/net/wireless/ath/ath6kl/core.c b/drivers/net/wireless/ath/ath6kl/core.c
index 4b46adbe8c92..b0b652042760 100644
--- a/drivers/net/wireless/ath/ath6kl/core.c
+++ b/drivers/net/wireless/ath/ath6kl/core.c
@@ -45,9 +45,9 @@ module_param(testmode, uint, 0644);
 module_param(recovery_enable, uint, 0644);
 module_param(heart_beat_poll, uint, 0644);
 MODULE_PARM_DESC(recovery_enable, "Enable recovery from firmware error");
-MODULE_PARM_DESC(heart_beat_poll, "Enable fw error detection periodic"   \
-		 "polling. This also specifies the polling interval in"  \
-		 "msecs. Set reocvery_enable for this to be effective");
+MODULE_PARM_DESC(heart_beat_poll,
+		 "Enable fw error detection periodic polling in msecs - Also set recovery_enable for this to be effective");
+
 
 void ath6kl_core_tx_complete(struct ath6kl *ar, struct sk_buff *skb)
 {
diff --git a/drivers/net/wireless/ath/ath6kl/debug.c b/drivers/net/wireless/ath/ath6kl/debug.c
index dbfd17d0a5fa..55c4064dd506 100644
--- a/drivers/net/wireless/ath/ath6kl/debug.c
+++ b/drivers/net/wireless/ath/ath6kl/debug.c
@@ -172,7 +172,6 @@ void ath6kl_dump_registers(struct ath6kl_device *dev,
 			   struct ath6kl_irq_proc_registers *irq_proc_reg,
 			   struct ath6kl_irq_enable_reg *irq_enable_reg)
 {
-
 	ath6kl_dbg(ATH6KL_DBG_IRQ, ("<------- Register Table -------->\n"));
 
 	if (irq_proc_reg != NULL) {
@@ -219,7 +218,6 @@ void ath6kl_dump_registers(struct ath6kl_device *dev,
 				   "GMBOX lookahead alias 1:   0x%x\n",
 				   irq_proc_reg->rx_gmbox_lkahd_alias[1]);
 		}
-
 	}
 
 	if (irq_enable_reg != NULL) {
@@ -1396,7 +1394,6 @@ static ssize_t ath6kl_create_qos_write(struct file *file,
 						const char __user *user_buf,
 						size_t count, loff_t *ppos)
 {
-
 	struct ath6kl *ar = file->private_data;
 	struct ath6kl_vif *vif;
 	char buf[200];
@@ -1575,7 +1572,6 @@ static ssize_t ath6kl_delete_qos_write(struct file *file,
 				const char __user *user_buf,
 				size_t count, loff_t *ppos)
 {
-
 	struct ath6kl *ar = file->private_data;
 	struct ath6kl_vif *vif;
 	char buf[100];
diff --git a/drivers/net/wireless/ath/ath6kl/debug.h b/drivers/net/wireless/ath/ath6kl/debug.h
index ca9ba005f287..e194c10d9f00 100644
--- a/drivers/net/wireless/ath/ath6kl/debug.h
+++ b/drivers/net/wireless/ath/ath6kl/debug.h
@@ -97,8 +97,8 @@ static inline void ath6kl_dump_registers(struct ath6kl_device *dev,
 		struct ath6kl_irq_proc_registers *irq_proc_reg,
 		struct ath6kl_irq_enable_reg *irq_en_reg)
 {
-
 }
+
 static inline void dump_cred_dist_stats(struct htc_target *target)
 {
 }
diff --git a/drivers/net/wireless/ath/ath6kl/hif.c b/drivers/net/wireless/ath/ath6kl/hif.c
index fea7709b5dda..18c070850a09 100644
--- a/drivers/net/wireless/ath/ath6kl/hif.c
+++ b/drivers/net/wireless/ath/ath6kl/hif.c
@@ -37,7 +37,6 @@ static int ath6kl_hif_cp_scat_dma_buf(struct hif_scatter_req *req,
 	buf = req->virt_dma_buf;
 
 	for (i = 0; i < req->scat_entries; i++) {
-
 		if (from_dma)
 			memcpy(req->scat_list[i].buf, buf,
 			       req->scat_list[i].len);
@@ -116,7 +115,6 @@ static void ath6kl_hif_dump_fw_crash(struct ath6kl *ar)
 			    le32_to_cpu(regdump_val[i + 2]),
 			    le32_to_cpu(regdump_val[i + 3]));
 	}
-
 }
 
 static int ath6kl_hif_proc_dbg_intr(struct ath6kl_device *dev)
@@ -701,5 +699,4 @@ int ath6kl_hif_setup(struct ath6kl_device *dev)
 
 fail_setup:
 	return status;
-
 }
diff --git a/drivers/net/wireless/ath/ath6kl/hif.h b/drivers/net/wireless/ath/ath6kl/hif.h
index 61f6b21fb0ae..dc6bd8cd9b83 100644
--- a/drivers/net/wireless/ath/ath6kl/hif.h
+++ b/drivers/net/wireless/ath/ath6kl/hif.h
@@ -197,9 +197,9 @@ struct hif_scatter_req {
 	/* bounce buffer for upper layers to copy to/from */
 	u8 *virt_dma_buf;
 
-	struct hif_scatter_item scat_list[1];
-
 	u32 scat_q_depth;
+
+	struct hif_scatter_item scat_list[0];
 };
 
 struct ath6kl_irq_proc_registers {
diff --git a/drivers/net/wireless/ath/ath6kl/htc_mbox.c b/drivers/net/wireless/ath/ath6kl/htc_mbox.c
index 65e5b719093d..e481f14b9878 100644
--- a/drivers/net/wireless/ath/ath6kl/htc_mbox.c
+++ b/drivers/net/wireless/ath/ath6kl/htc_mbox.c
@@ -112,9 +112,9 @@ static void ath6kl_credit_init(struct ath6kl_htc_credit_info *cred_info,
 		if (cur_ep_dist->endpoint == ENDPOINT_0)
 			continue;
 
-		if (cur_ep_dist->svc_id == WMI_CONTROL_SVC)
+		if (cur_ep_dist->svc_id == WMI_CONTROL_SVC) {
 			cur_ep_dist->cred_norm = cur_ep_dist->cred_per_msg;
-		else {
+		} else {
 			/*
 			 * For the remaining data endpoints, we assume that
 			 * each cred_per_msg are the same. We use a simple
@@ -129,7 +129,6 @@ static void ath6kl_credit_init(struct ath6kl_htc_credit_info *cred_info,
 			count = (count * 3) >> 2;
 			count = max(count, cur_ep_dist->cred_per_msg);
 			cur_ep_dist->cred_norm = count;
-
 		}
 
 		ath6kl_dbg(ATH6KL_DBG_CREDIT,
@@ -549,7 +548,6 @@ static int htc_check_credits(struct htc_target *target,
 			     enum htc_endpoint_id eid, unsigned int len,
 			     int *req_cred)
 {
-
 	*req_cred = (len > target->tgt_cred_sz) ?
 		     DIV_ROUND_UP(len, target->tgt_cred_sz) : 1;
 
@@ -608,7 +606,6 @@ static void ath6kl_htc_tx_pkts_get(struct htc_target *target,
 	unsigned int len;
 
 	while (true) {
-
 		flags = 0;
 
 		if (list_empty(&endpoint->txq))
@@ -889,7 +886,6 @@ static void ath6kl_htc_tx_from_queue(struct htc_target *target,
 		ac = target->dev->ar->ep2ac_map[endpoint->eid];
 
 	while (true) {
-
 		if (list_empty(&endpoint->txq))
 			break;
 
@@ -1190,7 +1186,6 @@ static void ath6kl_htc_mbox_flush_txep(struct htc_target *target,
 		list_add_tail(&packet->list, &container);
 		htc_tx_complete(endpoint, &container);
 	}
-
 }
 
 static void ath6kl_htc_flush_txep_all(struct htc_target *target)
@@ -1394,7 +1389,6 @@ static int ath6kl_htc_rx_setup(struct htc_target *target,
 
 	ep_cb = ep->ep_cb;
 	for (j = 0; j < n_msg; j++) {
-
 		/*
 		 * Reset flag, any packets allocated using the
 		 * rx_alloc() API cannot be recycled on
@@ -1424,9 +1418,9 @@ static int ath6kl_htc_rx_setup(struct htc_target *target,
 				}
 			}
 
-			if (list_empty(&ep->rx_bufq))
+			if (list_empty(&ep->rx_bufq)) {
 				packet = NULL;
-			else {
+			} else {
 				packet = list_first_entry(&ep->rx_bufq,
 						struct htc_packet, list);
 				list_del(&packet->list);
@@ -1487,7 +1481,6 @@ static int ath6kl_htc_rx_alloc(struct htc_target *target,
 	spin_lock_bh(&target->rx_lock);
 
 	for (i = 0; i < msg; i++) {
-
 		htc_hdr = (struct htc_frame_hdr *)&lk_ahds[i];
 
 		if (htc_hdr->eid >= ENDPOINT_MAX) {
@@ -1708,7 +1701,6 @@ static int htc_parse_trailer(struct htc_target *target,
 		lk_ahd = (struct htc_lookahead_report *) record_buf;
 		if ((lk_ahd->pre_valid == ((~lk_ahd->post_valid) & 0xFF)) &&
 		    next_lk_ahds) {
-
 			ath6kl_dbg(ATH6KL_DBG_HTC,
 				   "htc rx lk_ahd found pre_valid 0x%x post_valid 0x%x\n",
 				   lk_ahd->pre_valid, lk_ahd->post_valid);
@@ -1755,7 +1747,6 @@ static int htc_parse_trailer(struct htc_target *target,
 	}
 
 	return 0;
-
 }
 
 static int htc_proc_trailer(struct htc_target *target,
@@ -1776,7 +1767,6 @@ static int htc_proc_trailer(struct htc_target *target,
 	status = 0;
 
 	while (len > 0) {
-
 		if (len < sizeof(struct htc_record_hdr)) {
 			status = -ENOMEM;
 			break;
@@ -2098,7 +2088,6 @@ static int ath6kl_htc_rx_fetch(struct htc_target *target,
 		}
 
 		if (!fetched_pkts) {
-
 			packet = list_first_entry(rx_pktq, struct htc_packet,
 						   list);
 
@@ -2173,7 +2162,6 @@ int ath6kl_htc_rxmsg_pending_handler(struct htc_target *target,
 	look_aheads[0] = msg_look_ahead;
 
 	while (true) {
-
 		/*
 		 * First lookahead sets the expected endpoint IDs for all
 		 * packets in a bundle.
@@ -2825,8 +2813,9 @@ static int ath6kl_htc_reset(struct htc_target *target)
 			packet->buf = packet->buf_start;
 			packet->endpoint = ENDPOINT_0;
 			list_add_tail(&packet->list, &target->free_ctrl_rxbuf);
-		} else
+		} else {
 			list_add_tail(&packet->list, &target->free_ctrl_txbuf);
+		}
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/ath/ath6kl/htc_pipe.c b/drivers/net/wireless/ath/ath6kl/htc_pipe.c
index 67aa924ed8b3..756fe52a12c8 100644
--- a/drivers/net/wireless/ath/ath6kl/htc_pipe.c
+++ b/drivers/net/wireless/ath/ath6kl/htc_pipe.c
@@ -137,7 +137,6 @@ static void get_htc_packet_credit_based(struct htc_target *target,
 			credits_required = 0;
 
 		} else {
-
 			if (ep->cred_dist.credits < credits_required)
 				break;
 
@@ -169,7 +168,6 @@ static void get_htc_packet_credit_based(struct htc_target *target,
 		/* queue this packet into the caller's queue */
 		list_add_tail(&packet->list, queue);
 	}
-
 }
 
 static void get_htc_packet(struct htc_target *target,
@@ -279,7 +277,6 @@ static int htc_issue_packets(struct htc_target *target,
 			list_add(&packet->list, pkt_queue);
 			break;
 		}
-
 	}
 
 	if (status != 0) {
@@ -385,7 +382,6 @@ static enum htc_send_queue_result htc_try_send(struct htc_target *target,
 			 */
 			list_for_each_entry_safe(packet, tmp_pkt,
 						 txq, list) {
-
 				ath6kl_dbg(ATH6KL_DBG_HTC,
 					   "%s: Indicat overflowed TX pkts: %p\n",
 					   __func__, packet);
@@ -403,7 +399,6 @@ static enum htc_send_queue_result htc_try_send(struct htc_target *target,
 					list_move_tail(&packet->list,
 						       &send_queue);
 				}
-
 			}
 
 			if (list_empty(&send_queue)) {
@@ -454,7 +449,6 @@ static enum htc_send_queue_result htc_try_send(struct htc_target *target,
 	 * enough transmit resources.
 	 */
 	while (true) {
-
 		if (get_queue_depth(&ep->txq) == 0)
 			break;
 
@@ -495,8 +489,8 @@ static enum htc_send_queue_result htc_try_send(struct htc_target *target,
 		}
 
 		spin_lock_bh(&target->tx_lock);
-
 	}
+
 	/* done with this endpoint, we can clear the count */
 	ep->tx_proc_cnt = 0;
 	spin_unlock_bh(&target->tx_lock);
@@ -1106,7 +1100,6 @@ free_skb:
 	dev_kfree_skb(skb);
 
 	return status;
-
 }
 
 static void htc_flush_rx_queue(struct htc_target *target,
@@ -1258,7 +1251,6 @@ static int ath6kl_htc_pipe_conn_service(struct htc_target *target,
 		tx_alloc = 0;
 
 	} else {
-
 		tx_alloc = htc_get_credit_alloc(target, conn_req->svc_id);
 		if (tx_alloc == 0) {
 			status = -ENOMEM;
diff --git a/drivers/net/wireless/ath/ath6kl/init.c b/drivers/net/wireless/ath/ath6kl/init.c
index 4f316bdcbab5..d5ef211f261c 100644
--- a/drivers/net/wireless/ath/ath6kl/init.c
+++ b/drivers/net/wireless/ath/ath6kl/init.c
@@ -1192,7 +1192,6 @@ static int ath6kl_upload_board_file(struct ath6kl *ar)
 
 	if (board_ext_address &&
 	    ar->fw_board_len == (board_data_size + board_ext_data_size)) {
-
 		/* write extended board data */
 		ath6kl_dbg(ATH6KL_DBG_BOOT,
 			   "writing extended board data to 0x%x (%d B)\n",
diff --git a/drivers/net/wireless/ath/ath6kl/main.c b/drivers/net/wireless/ath/ath6kl/main.c
index 5839fc23bdc7..d56554674da4 100644
--- a/drivers/net/wireless/ath/ath6kl/main.c
+++ b/drivers/net/wireless/ath/ath6kl/main.c
@@ -571,7 +571,6 @@ void ath6kl_scan_complete_evt(struct ath6kl_vif *vif, int status)
 
 static int ath6kl_commit_ch_switch(struct ath6kl_vif *vif, u16 channel)
 {
-
 	struct ath6kl *ar = vif->ar;
 
 	vif->profile.ch = cpu_to_le16(channel);
@@ -600,7 +599,6 @@ static int ath6kl_commit_ch_switch(struct ath6kl_vif *vif, u16 channel)
 
 static void ath6kl_check_ch_switch(struct ath6kl *ar, u16 channel)
 {
-
 	struct ath6kl_vif *vif;
 	int res = 0;
 
@@ -692,9 +690,9 @@ void ath6kl_tkip_micerr_event(struct ath6kl_vif *vif, u8 keyid, bool ismcast)
 		cfg80211_michael_mic_failure(vif->ndev, sta->mac,
 					     NL80211_KEYTYPE_PAIRWISE, keyid,
 					     tsc, GFP_KERNEL);
-	} else
+	} else {
 		ath6kl_cfg80211_tkip_micerr_event(vif, keyid, ismcast);
-
+	}
 }
 
 static void ath6kl_update_target_stats(struct ath6kl_vif *vif, u8 *ptr, u32 len)
@@ -1093,8 +1091,9 @@ static int ath6kl_open(struct net_device *dev)
 	if (test_bit(CONNECTED, &vif->flags)) {
 		netif_carrier_on(dev);
 		netif_wake_queue(dev);
-	} else
+	} else {
 		netif_carrier_off(dev);
+	}
 
 	return 0;
 }
@@ -1146,7 +1145,6 @@ static int ath6kl_set_features(struct net_device *dev,
 			dev->features = features | NETIF_F_RXCSUM;
 			return err;
 		}
-
 	}
 
 	return err;
diff --git a/drivers/net/wireless/ath/ath6kl/sdio.c b/drivers/net/wireless/ath/ath6kl/sdio.c
index 7126bdd4236c..339d89f14d32 100644
--- a/drivers/net/wireless/ath/ath6kl/sdio.c
+++ b/drivers/net/wireless/ath/ath6kl/sdio.c
@@ -348,7 +348,7 @@ static int ath6kl_sdio_alloc_prep_scat_req(struct ath6kl_sdio *ar_sdio,
 	int i, scat_req_sz, scat_list_sz, size;
 	u8 *virt_buf;
 
-	scat_list_sz = (n_scat_entry - 1) * sizeof(struct hif_scatter_item);
+	scat_list_sz = n_scat_entry * sizeof(struct hif_scatter_item);
 	scat_req_sz = sizeof(*s_req) + scat_list_sz;
 
 	if (!virt_scat)
@@ -425,8 +425,9 @@ static int ath6kl_sdio_read_write_sync(struct ath6kl *ar, u32 addr, u8 *buf,
 			memcpy(tbuf, buf, len);
 
 		bounced = true;
-	} else
+	} else {
 		tbuf = buf;
+	}
 
 	ret = ath6kl_sdio_io(ar_sdio->func, request, addr, tbuf, len);
 	if ((request & HIF_READ) && bounced)
@@ -441,9 +442,9 @@ static int ath6kl_sdio_read_write_sync(struct ath6kl *ar, u32 addr, u8 *buf,
 static void __ath6kl_sdio_write_async(struct ath6kl_sdio *ar_sdio,
 				      struct bus_request *req)
 {
-	if (req->scat_req)
+	if (req->scat_req) {
 		ath6kl_sdio_scat_rw(ar_sdio, req);
-	else {
+	} else {
 		void *context;
 		int status;
 
@@ -656,7 +657,6 @@ static void ath6kl_sdio_scatter_req_add(struct ath6kl *ar,
 	list_add_tail(&s_req->list, &ar_sdio->scat_req);
 
 	spin_unlock_bh(&ar_sdio->scat_lock);
-
 }
 
 /* scatter gather read write request */
@@ -674,9 +674,9 @@ static int ath6kl_sdio_async_rw_scatter(struct ath6kl *ar,
 		   "hif-scatter: total len: %d scatter entries: %d\n",
 		   scat_req->len, scat_req->scat_entries);
 
-	if (request & HIF_SYNCHRONOUS)
+	if (request & HIF_SYNCHRONOUS) {
 		status = ath6kl_sdio_scat_rw(ar_sdio, scat_req->busrequest);
-	else {
+	} else {
 		spin_lock_bh(&ar_sdio->wr_async_lock);
 		list_add_tail(&scat_req->busrequest->list, &ar_sdio->wr_asyncq);
 		spin_unlock_bh(&ar_sdio->wr_async_lock);
@@ -856,7 +856,6 @@ static int ath6kl_sdio_suspend(struct ath6kl *ar, struct cfg80211_wowlan *wow)
 
 	if (ar->suspend_mode == WLAN_POWER_STATE_WOW ||
 	    (!ar->suspend_mode && wow)) {
-
 		ret = ath6kl_set_sdio_pm_caps(ar);
 		if (ret)
 			goto cut_pwr;
@@ -878,7 +877,6 @@ static int ath6kl_sdio_suspend(struct ath6kl *ar, struct cfg80211_wowlan *wow)
 
 	if (ar->suspend_mode == WLAN_POWER_STATE_DEEP_SLEEP ||
 	    !ar->suspend_mode || try_deepsleep) {
-
 		flags = sdio_get_host_pm_caps(func);
 		if (!(flags & MMC_PM_KEEP_POWER))
 			goto cut_pwr;
@@ -1061,7 +1059,6 @@ static int ath6kl_sdio_bmi_credits(struct ath6kl *ar)
 
 	timeout = jiffies + msecs_to_jiffies(BMI_COMMUNICATION_TIMEOUT);
 	while (time_before(jiffies, timeout) && !ar->bmi.cmd_credits) {
-
 		/*
 		 * Hit the credit counter with a 4-byte access, the first byte
 		 * read will hit the counter and cause a decrement, while the
diff --git a/drivers/net/wireless/ath/ath6kl/target.h b/drivers/net/wireless/ath/ath6kl/target.h
index a580a629a0da..d5eeeae7711b 100644
--- a/drivers/net/wireless/ath/ath6kl/target.h
+++ b/drivers/net/wireless/ath/ath6kl/target.h
@@ -289,7 +289,7 @@ struct host_interest {
 	u32 hi_hp_rx_traffic_ratio;                    /* 0xd8 */
 
 	/* test applications flags */
-	u32 hi_test_apps_related    ;                  /* 0xdc */
+	u32 hi_test_apps_related;                      /* 0xdc */
 	/* location of test script */
 	u32 hi_ota_testscript;                         /* 0xe0 */
 	/* location of CAL data */
diff --git a/drivers/net/wireless/ath/ath6kl/txrx.c b/drivers/net/wireless/ath/ath6kl/txrx.c
index ebb24045a8ae..40432fe7a5d2 100644
--- a/drivers/net/wireless/ath/ath6kl/txrx.c
+++ b/drivers/net/wireless/ath/ath6kl/txrx.c
@@ -125,8 +125,9 @@ static bool ath6kl_process_uapsdq(struct ath6kl_sta *conn,
 		*flags |= WMI_DATA_HDR_FLAGS_UAPSD;
 		spin_unlock_bh(&conn->psq_lock);
 		return false;
-	} else if (!conn->apsd_info)
+	} else if (!conn->apsd_info) {
 		return false;
+	}
 
 	if (test_bit(WMM_ENABLED, &vif->flags)) {
 		ether_type = be16_to_cpu(datap->h_proto);
@@ -316,8 +317,9 @@ int ath6kl_control_tx(void *devt, struct sk_buff *skb,
 		cookie = NULL;
 		ath6kl_err("wmi ctrl ep full, dropping pkt : 0x%p, len:%d\n",
 			   skb, skb->len);
-	} else
+	} else {
 		cookie = ath6kl_alloc_cookie(ar);
+	}
 
 	if (cookie == NULL) {
 		spin_unlock_bh(&ar->lock);
@@ -359,7 +361,7 @@ int ath6kl_data_tx(struct sk_buff *skb, struct net_device *dev)
 	struct ath6kl_vif *vif = netdev_priv(dev);
 	u32 map_no = 0;
 	u16 htc_tag = ATH6KL_DATA_PKT_TAG;
-	u8 ac = 99 ; /* initialize to unmapped ac */
+	u8 ac = 99; /* initialize to unmapped ac */
 	bool chk_adhoc_ps_mapping = false;
 	int ret;
 	struct wmi_tx_meta_v2 meta_v2;
@@ -449,8 +451,9 @@ int ath6kl_data_tx(struct sk_buff *skb, struct net_device *dev)
 			if (ret)
 				goto fail_tx;
 		}
-	} else
+	} else {
 		goto fail_tx;
+	}
 
 	spin_lock_bh(&ar->lock);
 
@@ -702,7 +705,6 @@ void ath6kl_tx_complete(struct htc_target *target,
 
 	/* reap completed packets */
 	while (!list_empty(packet_queue)) {
-
 		packet = list_first_entry(packet_queue, struct htc_packet,
 					  list);
 		list_del(&packet->list);
@@ -1089,8 +1091,9 @@ static void aggr_deque_frms(struct aggr_info_conn *agg_conn, u8 tid,
 			else
 				skb_queue_tail(&rxtid->q, node->skb);
 			node->skb = NULL;
-		} else
+		} else {
 			stats->num_hole++;
+		}
 
 		rxtid->seq_next = ATH6KL_NEXT_SEQ_NO(rxtid->seq_next);
 		idx = AGGR_WIN_IDX(rxtid->seq_next, rxtid->hold_q_sz);
@@ -1211,7 +1214,7 @@ static bool aggr_process_recv_frm(struct aggr_info_conn *agg_conn, u8 tid,
 		return is_queued;
 
 	spin_lock_bh(&rxtid->lock);
-	for (idx = 0 ; idx < rxtid->hold_q_sz; idx++) {
+	for (idx = 0; idx < rxtid->hold_q_sz; idx++) {
 		if (rxtid->hold_q[idx].skb) {
 			/*
 			 * There is a frame in the queue and no
@@ -1265,7 +1268,6 @@ static void ath6kl_uapsd_trigger_frame_rx(struct ath6kl_vif *vif,
 	is_apsdq_empty_at_start = is_apsdq_empty;
 
 	while ((!is_apsdq_empty) && (num_frames_to_deliver)) {
-
 		spin_lock_bh(&conn->psq_lock);
 		skb = skb_dequeue(&conn->apsdq);
 		is_apsdq_empty = skb_queue_empty(&conn->apsdq);
@@ -1606,16 +1608,18 @@ void ath6kl_rx(struct htc_target *target, struct htc_packet *packet)
 			if (!conn)
 				return;
 			aggr_conn = conn->aggr_conn;
-		} else
+		} else {
 			aggr_conn = vif->aggr_cntxt->aggr_conn;
+		}
 
 		if (aggr_process_recv_frm(aggr_conn, tid, seq_no,
 					  is_amsdu, skb)) {
 			/* aggregation code will handle the skb */
 			return;
 		}
-	} else if (!is_broadcast_ether_addr(datap->h_dest))
+	} else if (!is_broadcast_ether_addr(datap->h_dest)) {
 		vif->net_stats.multicast++;
+	}
 
 	ath6kl_deliver_frames_to_nw_stack(vif->ndev, skb);
 }
@@ -1710,8 +1714,9 @@ void aggr_recv_addba_req_evt(struct ath6kl_vif *vif, u8 tid_mux, u16 seq_no,
 		sta = ath6kl_find_sta_by_aid(vif->ar, aid);
 		if (sta)
 			aggr_conn = sta->aggr_conn;
-	} else
+	} else {
 		aggr_conn = vif->aggr_cntxt->aggr_conn;
+	}
 
 	if (!aggr_conn)
 		return;
@@ -1766,7 +1771,6 @@ void aggr_conn_init(struct ath6kl_vif *vif, struct aggr_info *aggr_info,
 		skb_queue_head_init(&rxtid->q);
 		spin_lock_init(&rxtid->lock);
 	}
-
 }
 
 struct aggr_info *aggr_init(struct ath6kl_vif *vif)
@@ -1806,8 +1810,9 @@ void aggr_recv_delba_req_evt(struct ath6kl_vif *vif, u8 tid_mux)
 		sta = ath6kl_find_sta_by_aid(vif->ar, aid);
 		if (sta)
 			aggr_conn = sta->aggr_conn;
-	} else
+	} else {
 		aggr_conn = vif->aggr_cntxt->aggr_conn;
+	}
 
 	if (!aggr_conn)
 		return;
diff --git a/drivers/net/wireless/ath/ath6kl/usb.c b/drivers/net/wireless/ath/ath6kl/usb.c
index 56c3fd5cef65..3afc5a463d06 100644
--- a/drivers/net/wireless/ath/ath6kl/usb.c
+++ b/drivers/net/wireless/ath/ath6kl/usb.c
@@ -236,7 +236,6 @@ static void ath6kl_usb_free_pipe_resources(struct ath6kl_usb_pipe *pipe)
 			break;
 		kfree(urb_context);
 	}
-
 }
 
 static void ath6kl_usb_cleanup_pipe_resources(struct ath6kl_usb *ar_usb)
@@ -245,7 +244,6 @@ static void ath6kl_usb_cleanup_pipe_resources(struct ath6kl_usb *ar_usb)
 
 	for (i = 0; i < ATH6KL_USB_PIPE_MAX; i++)
 		ath6kl_usb_free_pipe_resources(&ar_usb->pipes[i]);
-
 }
 
 static u8 ath6kl_usb_get_logical_pipe_num(struct ath6kl_usb *ar_usb,
diff --git a/drivers/net/wireless/ath/ath6kl/wmi.c b/drivers/net/wireless/ath/ath6kl/wmi.c
index 8b4ce28e3ce8..4d7f9e4712e9 100644
--- a/drivers/net/wireless/ath/ath6kl/wmi.c
+++ b/drivers/net/wireless/ath/ath6kl/wmi.c
@@ -289,8 +289,9 @@ int ath6kl_wmi_implicit_create_pstream(struct wmi *wmi, u8 if_idx,
 			   ath6kl_wmi_determine_user_priority(((u8 *) llc_hdr) +
 					sizeof(struct ath6kl_llc_snap_hdr),
 					layer2_priority);
-		} else
+		} else {
 			usr_pri = layer2_priority & 0x7;
+		}
 
 		/*
 		 * Queue the EAPOL frames in the same WMM_AC_VO queue
@@ -359,8 +360,9 @@ int ath6kl_wmi_dot11_hdr_remove(struct wmi *wmi, struct sk_buff *skb)
 		hdr_size = roundup(sizeof(struct ieee80211_qos_hdr),
 				   sizeof(u32));
 		skb_pull(skb, hdr_size);
-	} else if (sub_type == cpu_to_le16(IEEE80211_STYPE_DATA))
+	} else if (sub_type == cpu_to_le16(IEEE80211_STYPE_DATA)) {
 		skb_pull(skb, sizeof(struct ieee80211_hdr_3addr));
+	}
 
 	datap = skb->data;
 	llc_hdr = (struct ath6kl_llc_snap_hdr *)(datap);
@@ -936,7 +938,6 @@ ath6kl_regd_find_country_by_rd(u16 regdmn)
 
 static void ath6kl_wmi_regdomain_event(struct wmi *wmi, u8 *datap, int len)
 {
-
 	struct ath6kl_wmi_regdomain *ev;
 	struct country_code_to_enum_rd *country = NULL;
 	struct reg_dmn_pair_mapping *regpair = NULL;
@@ -946,10 +947,9 @@ static void ath6kl_wmi_regdomain_event(struct wmi *wmi, u8 *datap, int len)
 	ev = (struct ath6kl_wmi_regdomain *) datap;
 	reg_code = le32_to_cpu(ev->reg_code);
 
-	if ((reg_code >> ATH6KL_COUNTRY_RD_SHIFT) & COUNTRY_ERD_FLAG)
+	if ((reg_code >> ATH6KL_COUNTRY_RD_SHIFT) & COUNTRY_ERD_FLAG) {
 		country = ath6kl_regd_find_country((u16) reg_code);
-	else if (!(((u16) reg_code & WORLD_SKU_MASK) == WORLD_SKU_PREFIX)) {
-
+	} else if (!(((u16) reg_code & WORLD_SKU_MASK) == WORLD_SKU_PREFIX)) {
 		regpair = ath6kl_get_regpair((u16) reg_code);
 		country = ath6kl_regd_find_country_by_rd((u16) reg_code);
 		if (regpair)
@@ -1499,7 +1499,6 @@ static int ath6kl_wmi_cac_event_rx(struct wmi *wmi, u8 *datap, int len,
 
 	if ((reply->cac_indication == CAC_INDICATION_ADMISSION_RESP) &&
 	    (reply->status_code != IEEE80211_TSPEC_STATUS_ADMISS_ACCEPTED)) {
-
 		ts = (struct ieee80211_tspec_ie *) &(reply->tspec_suggestion);
 		tsinfo = le16_to_cpu(ts->tsinfo);
 		tsid = (tsinfo >> IEEE80211_WMM_IE_TSPEC_TID_SHIFT) &
@@ -1530,7 +1529,6 @@ static int ath6kl_wmi_cac_event_rx(struct wmi *wmi, u8 *datap, int len,
 	 * for delete qos stream from AP
 	 */
 	else if (reply->cac_indication == CAC_INDICATION_DELETE) {
-
 		ts = (struct ieee80211_tspec_ie *) &(reply->tspec_suggestion);
 		tsinfo = le16_to_cpu(ts->tsinfo);
 		ts_id = ((tsinfo >> IEEE80211_WMM_IE_TSPEC_TID_SHIFT) &
@@ -2322,7 +2320,7 @@ int ath6kl_wmi_addkey_cmd(struct wmi *wmi, u8 if_idx, u8 key_index,
 	return ret;
 }
 
-int ath6kl_wmi_add_krk_cmd(struct wmi *wmi, u8 if_idx, u8 *krk)
+int ath6kl_wmi_add_krk_cmd(struct wmi *wmi, u8 if_idx, const u8 *krk)
 {
 	struct sk_buff *skb;
 	struct wmi_add_krk_cmd *cmd;
@@ -2479,7 +2477,6 @@ static int ath6kl_wmi_sync_point(struct wmi *wmi, u8 if_idx)
 		goto free_data_skb;
 
 	for (index = 0; index < num_pri_streams; index++) {
-
 		if (WARN_ON(!data_sync_bufs[index].skb))
 			goto free_data_skb;
 
@@ -2704,7 +2701,6 @@ static void ath6kl_wmi_relinquish_implicit_pstream_credits(struct wmi *wmi)
 
 	for (i = 0; i < WMM_NUM_AC; i++) {
 		if (stream_exist & (1 << i)) {
-
 			/*
 			 * FIXME: Is this lock & unlock inside
 			 * for loop correct? may need rework.
@@ -2870,8 +2866,9 @@ int ath6kl_wmi_set_host_sleep_mode_cmd(struct wmi *wmi, u8 if_idx,
 	if (host_mode == ATH6KL_HOST_MODE_ASLEEP) {
 		ath6kl_wmi_relinquish_implicit_pstream_credits(wmi);
 		cmd->asleep = cpu_to_le32(1);
-	} else
+	} else {
 		cmd->awake = cpu_to_le32(1);
+	}
 
 	ret = ath6kl_wmi_cmd_send(wmi, if_idx, skb,
 				  WMI_SET_HOST_SLEEP_MODE_CMDID,
diff --git a/drivers/net/wireless/ath/ath6kl/wmi.h b/drivers/net/wireless/ath/ath6kl/wmi.h
index 5c702ae4d9f8..bb23fc00111d 100644
--- a/drivers/net/wireless/ath/ath6kl/wmi.h
+++ b/drivers/net/wireless/ath/ath6kl/wmi.h
@@ -898,7 +898,6 @@ struct wmi_start_scan_cmd {
  *  flags here
  */
 enum wmi_scan_ctrl_flags_bits {
-
 	/* set if can scan in the connect cmd */
 	CONNECT_SCAN_CTRL_FLAGS = 0x01,
 
@@ -2617,7 +2616,7 @@ int ath6kl_wmi_addkey_cmd(struct wmi *wmi, u8 if_idx, u8 key_index,
 			  u8 *key_material,
 			  u8 key_op_ctrl, u8 *mac_addr,
 			  enum wmi_sync_flag sync_flag);
-int ath6kl_wmi_add_krk_cmd(struct wmi *wmi, u8 if_idx, u8 *krk);
+int ath6kl_wmi_add_krk_cmd(struct wmi *wmi, u8 if_idx, const u8 *krk);
 int ath6kl_wmi_deletekey_cmd(struct wmi *wmi, u8 if_idx, u8 key_index);
 int ath6kl_wmi_setpmkid_cmd(struct wmi *wmi, u8 if_idx, const u8 *bssid,
 			    const u8 *pmkid, bool set);
diff --git a/drivers/net/wireless/ath/ath9k/Makefile b/drivers/net/wireless/ath/ath9k/Makefile
index 8e1c7b0fe76c..8fcd586d1c39 100644
--- a/drivers/net/wireless/ath/ath9k/Makefile
+++ b/drivers/net/wireless/ath/ath9k/Makefile
@@ -53,7 +53,8 @@ obj-$(CONFIG_ATH9K_HW) += ath9k_hw.o
 obj-$(CONFIG_ATH9K_COMMON) += ath9k_common.o
 ath9k_common-y:=	common.o \
 			common-init.o \
-			common-beacon.o
+			common-beacon.o \
+			common-debug.o
 
 ath9k_htc-y +=	htc_hst.o \
 		hif_usb.o \
diff --git a/drivers/net/wireless/ath/ath9k/ar9003_2p2_initvals.h b/drivers/net/wireless/ath/ath9k/ar9003_2p2_initvals.h
index 0a6163e9248c..c38399bc9aa9 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_2p2_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar9003_2p2_initvals.h
@@ -410,7 +410,7 @@ static const u32 ar9300_2p2_baseband_core[][2] = {
 	{0x00009e30, 0x06336f77},
 	{0x00009e34, 0x6af6532f},
 	{0x00009e38, 0x0cc80c00},
-	{0x00009e40, 0x0d261820},
+	{0x00009e40, 0x0d261800},
 	{0x00009e4c, 0x00001004},
 	{0x00009e50, 0x00ff03f1},
 	{0x00009e54, 0x00000000},
diff --git a/drivers/net/wireless/ath/ath9k/ar9330_1p1_initvals.h b/drivers/net/wireless/ath/ath9k/ar9330_1p1_initvals.h
index f76139bbb74f..2c42ff05efa3 100644
--- a/drivers/net/wireless/ath/ath9k/ar9330_1p1_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar9330_1p1_initvals.h
@@ -592,7 +592,7 @@ static const u32 ar9331_1p1_baseband_core[][2] = {
 	{0x00009e30, 0x06336f77},
 	{0x00009e34, 0x6af6532f},
 	{0x00009e38, 0x0cc80c00},
-	{0x00009e40, 0x0d261820},
+	{0x00009e40, 0x0d261800},
 	{0x00009e4c, 0x00001004},
 	{0x00009e50, 0x00ff03f1},
 	{0x00009fc0, 0x803e4788},
diff --git a/drivers/net/wireless/ath/ath9k/ar9330_1p2_initvals.h b/drivers/net/wireless/ath/ath9k/ar9330_1p2_initvals.h
index 0ac8be96097f..2154efcd3900 100644
--- a/drivers/net/wireless/ath/ath9k/ar9330_1p2_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar9330_1p2_initvals.h
@@ -231,7 +231,7 @@ static const u32 ar9331_1p2_baseband_core[][2] = {
 	{0x00009e30, 0x06336f77},
 	{0x00009e34, 0x6af6532f},
 	{0x00009e38, 0x0cc80c00},
-	{0x00009e40, 0x0d261820},
+	{0x00009e40, 0x0d261800},
 	{0x00009e4c, 0x00001004},
 	{0x00009e50, 0x00ff03f1},
 	{0x00009fc0, 0x803e4788},
diff --git a/drivers/net/wireless/ath/ath9k/ar9340_initvals.h b/drivers/net/wireless/ath/ath9k/ar9340_initvals.h
index a01f0edb6518..b995ffe88b33 100644
--- a/drivers/net/wireless/ath/ath9k/ar9340_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar9340_initvals.h
@@ -318,7 +318,7 @@ static const u32 ar9340_1p0_baseband_core[][2] = {
 	{0x00009e30, 0x06336f77},
 	{0x00009e34, 0x6af6532f},
 	{0x00009e38, 0x0cc80c00},
-	{0x00009e40, 0x0d261820},
+	{0x00009e40, 0x0d261800},
 	{0x00009e4c, 0x00001004},
 	{0x00009e50, 0x00ff03f1},
 	{0x00009e54, 0x00000000},
@@ -348,9 +348,9 @@ static const u32 ar9340_1p0_baseband_core[][2] = {
 	{0x0000a370, 0x00000000},
 	{0x0000a390, 0x00000001},
 	{0x0000a394, 0x00000444},
-	{0x0000a398, 0x00000000},
-	{0x0000a39c, 0x210d0401},
-	{0x0000a3a0, 0xab9a7144},
+	{0x0000a398, 0x001f0e0f},
+	{0x0000a39c, 0x0075393f},
+	{0x0000a3a0, 0xb79f6427},
 	{0x0000a3a4, 0x00000000},
 	{0x0000a3a8, 0xaaaaaaaa},
 	{0x0000a3ac, 0x3c466478},
diff --git a/drivers/net/wireless/ath/ath9k/ar953x_initvals.h b/drivers/net/wireless/ath/ath9k/ar953x_initvals.h
index 3c9113d9b1bc..8e5c3b9786e3 100644
--- a/drivers/net/wireless/ath/ath9k/ar953x_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar953x_initvals.h
@@ -257,9 +257,9 @@ static const u32 qca953x_1p0_baseband_core[][2] = {
 	{0x0000a370, 0x00000000},
 	{0x0000a390, 0x00000001},
 	{0x0000a394, 0x00000444},
-	{0x0000a398, 0x1f020503},
-	{0x0000a39c, 0x29180c03},
-	{0x0000a3a0, 0x9a8b6844},
+	{0x0000a398, 0x001f0e0f},
+	{0x0000a39c, 0x0075393f},
+	{0x0000a3a0, 0xb79f6427},
 	{0x0000a3a4, 0x000000ff},
 	{0x0000a3a8, 0x6a6a6a6a},
 	{0x0000a3ac, 0x6a6a6a6a},
diff --git a/drivers/net/wireless/ath/ath9k/ar9580_1p0_initvals.h b/drivers/net/wireless/ath/ath9k/ar9580_1p0_initvals.h
index e6aec2c0207f..a5ca65240af3 100644
--- a/drivers/net/wireless/ath/ath9k/ar9580_1p0_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar9580_1p0_initvals.h
@@ -90,7 +90,7 @@ static const u32 ar9580_1p0_baseband_core[][2] = {
 	{0x00009e30, 0x06336f77},
 	{0x00009e34, 0x6af6532f},
 	{0x00009e38, 0x0cc80c00},
-	{0x00009e40, 0x0d261820},
+	{0x00009e40, 0x0d261800},
 	{0x00009e4c, 0x00001004},
 	{0x00009e50, 0x00ff03f1},
 	{0x00009e54, 0x00000000},
diff --git a/drivers/net/wireless/ath/ath9k/ath9k.h b/drivers/net/wireless/ath/ath9k/ath9k.h
index 3ba03dde4215..2ca8f7e06174 100644
--- a/drivers/net/wireless/ath/ath9k/ath9k.h
+++ b/drivers/net/wireless/ath/ath9k/ath9k.h
@@ -23,8 +23,8 @@
 #include <linux/leds.h>
 #include <linux/completion.h>
 
-#include "debug.h"
 #include "common.h"
+#include "debug.h"
 #include "mci.h"
 #include "dfs.h"
 #include "spectral.h"
@@ -114,6 +114,9 @@ int ath_descdma_setup(struct ath_softc *sc, struct ath_descdma *dd,
 #define ATH_TXFIFO_DEPTH           8
 #define ATH_TX_ERROR               0x01
 
+/* Stop tx traffic 1ms before the GO goes away */
+#define ATH_P2P_PS_STOP_TIME       1000
+
 #define IEEE80211_SEQ_SEQ_SHIFT    4
 #define IEEE80211_SEQ_MAX          4096
 #define IEEE80211_WEP_IVLEN        3
@@ -271,6 +274,7 @@ struct ath_node {
 #ifdef CONFIG_ATH9K_STATION_STATISTICS
 	struct ath_rx_rate_stats rx_rate_stats;
 #endif
+	u8 key_idx[4];
 };
 
 struct ath_tx_control {
@@ -366,11 +370,15 @@ void ath9k_release_buffered_frames(struct ieee80211_hw *hw,
 /********/
 
 struct ath_vif {
+	struct ieee80211_vif *vif;
 	struct ath_node mcast_node;
 	int av_bslot;
 	bool primary_sta_vif;
 	__le64 tsf_adjust; /* TSF adjustment for staggered beacons */
 	struct ath_buf *av_bcbuf;
+
+	/* P2P Client */
+	struct ieee80211_noa_data noa;
 };
 
 struct ath9k_vif_iter_data {
@@ -463,6 +471,8 @@ int ath_update_survey_stats(struct ath_softc *sc);
 void ath_update_survey_nf(struct ath_softc *sc, int channel);
 void ath9k_queue_reset(struct ath_softc *sc, enum ath_reset_type type);
 void ath_ps_full_sleep(unsigned long data);
+void ath9k_p2p_ps_timer(void *priv);
+void ath9k_update_p2p_ps(struct ath_softc *sc, struct ieee80211_vif *vif);
 
 /**********/
 /* BTCOEX */
@@ -713,6 +723,9 @@ struct ath_softc {
 	struct completion paprd_complete;
 	wait_queue_head_t tx_wait;
 
+	struct ath_gen_timer *p2p_ps_timer;
+	struct ath_vif *p2p_ps_vif;
+
 	unsigned long driver_data;
 
 	u8 gtt_cnt;
@@ -757,6 +770,7 @@ struct ath_softc {
 	struct ath_ant_comb ant_comb;
 	u8 ant_tx, ant_rx;
 	struct dfs_pattern_detector *dfs_detector;
+	u64 dfs_prev_pulse_ts;
 	u32 wow_enabled;
 	/* relay(fs) channel for spectral scan */
 	struct rchan *rfs_chan_spec_scan;
diff --git a/drivers/net/wireless/ath/ath9k/beacon.c b/drivers/net/wireless/ath/ath9k/beacon.c
index bd9e634879e6..e387f0b2954a 100644
--- a/drivers/net/wireless/ath/ath9k/beacon.c
+++ b/drivers/net/wireless/ath/ath9k/beacon.c
@@ -537,8 +537,6 @@ static void ath9k_cache_beacon_config(struct ath_softc *sc,
 	cur_conf->dtim_period = bss_conf->dtim_period;
 	cur_conf->dtim_count = 1;
 	cur_conf->ibss_creator = bss_conf->ibss_creator;
-	cur_conf->bmiss_timeout =
-		ATH_DEFAULT_BMISS_LIMIT * cur_conf->beacon_interval;
 
 	/*
 	 * It looks like mac80211 may end up using beacon interval of zero in
@@ -549,6 +547,9 @@ static void ath9k_cache_beacon_config(struct ath_softc *sc,
 	if (cur_conf->beacon_interval == 0)
 		cur_conf->beacon_interval = 100;
 
+	cur_conf->bmiss_timeout =
+		ATH_DEFAULT_BMISS_LIMIT * cur_conf->beacon_interval;
+
 	/*
 	 * We don't parse dtim period from mac80211 during the driver
 	 * initialization as it breaks association with hidden-ssid
diff --git a/drivers/net/wireless/ath/ath9k/common-debug.c b/drivers/net/wireless/ath/ath9k/common-debug.c
new file mode 100644
index 000000000000..3b289f933405
--- /dev/null
+++ b/drivers/net/wireless/ath/ath9k/common-debug.c
@@ -0,0 +1,253 @@
+/*
+ * Copyright (c) 2008-2011 Atheros Communications Inc.
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+
+#include "common.h"
+
+static ssize_t read_file_modal_eeprom(struct file *file, char __user *user_buf,
+				      size_t count, loff_t *ppos)
+{
+	struct ath_hw *ah = file->private_data;
+	u32 len = 0, size = 6000;
+	char *buf;
+	size_t retval;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (buf == NULL)
+		return -ENOMEM;
+
+	len = ah->eep_ops->dump_eeprom(ah, false, buf, len, size);
+
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	return retval;
+}
+
+static const struct file_operations fops_modal_eeprom = {
+	.read = read_file_modal_eeprom,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+
+void ath9k_cmn_debug_modal_eeprom(struct dentry *debugfs_phy,
+				  struct ath_hw *ah)
+{
+	debugfs_create_file("modal_eeprom", S_IRUSR, debugfs_phy, ah,
+			    &fops_modal_eeprom);
+}
+EXPORT_SYMBOL(ath9k_cmn_debug_modal_eeprom);
+
+static ssize_t read_file_base_eeprom(struct file *file, char __user *user_buf,
+				     size_t count, loff_t *ppos)
+{
+	struct ath_hw *ah = file->private_data;
+	u32 len = 0, size = 1500;
+	ssize_t retval = 0;
+	char *buf;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	len = ah->eep_ops->dump_eeprom(ah, true, buf, len, size);
+
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	return retval;
+}
+
+static const struct file_operations fops_base_eeprom = {
+	.read = read_file_base_eeprom,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+void ath9k_cmn_debug_base_eeprom(struct dentry *debugfs_phy,
+				 struct ath_hw *ah)
+{
+	debugfs_create_file("base_eeprom", S_IRUSR, debugfs_phy, ah,
+			    &fops_base_eeprom);
+}
+EXPORT_SYMBOL(ath9k_cmn_debug_base_eeprom);
+
+void ath9k_cmn_debug_stat_rx(struct ath_rx_stats *rxstats,
+			     struct ath_rx_status *rs)
+{
+#define RX_PHY_ERR_INC(c) rxstats->phy_err_stats[c]++
+#define RX_CMN_STAT_INC(c) (rxstats->c++)
+
+	RX_CMN_STAT_INC(rx_pkts_all);
+	rxstats->rx_bytes_all += rs->rs_datalen;
+
+	if (rs->rs_status & ATH9K_RXERR_CRC)
+		RX_CMN_STAT_INC(crc_err);
+	if (rs->rs_status & ATH9K_RXERR_DECRYPT)
+		RX_CMN_STAT_INC(decrypt_crc_err);
+	if (rs->rs_status & ATH9K_RXERR_MIC)
+		RX_CMN_STAT_INC(mic_err);
+	if (rs->rs_status & ATH9K_RX_DELIM_CRC_PRE)
+		RX_CMN_STAT_INC(pre_delim_crc_err);
+	if (rs->rs_status & ATH9K_RX_DELIM_CRC_POST)
+		RX_CMN_STAT_INC(post_delim_crc_err);
+	if (rs->rs_status & ATH9K_RX_DECRYPT_BUSY)
+		RX_CMN_STAT_INC(decrypt_busy_err);
+
+	if (rs->rs_status & ATH9K_RXERR_PHY) {
+		RX_CMN_STAT_INC(phy_err);
+		if (rs->rs_phyerr < ATH9K_PHYERR_MAX)
+			RX_PHY_ERR_INC(rs->rs_phyerr);
+	}
+
+#undef RX_CMN_STAT_INC
+#undef RX_PHY_ERR_INC
+}
+EXPORT_SYMBOL(ath9k_cmn_debug_stat_rx);
+
+static ssize_t read_file_recv(struct file *file, char __user *user_buf,
+			      size_t count, loff_t *ppos)
+{
+#define RXS_ERR(s, e)					\
+	do {						\
+		len += scnprintf(buf + len, size - len,	\
+				 "%18s : %10u\n", s,	\
+				 rxstats->e);		\
+	} while (0)
+
+	struct ath_rx_stats *rxstats = file->private_data;
+	char *buf;
+	unsigned int len = 0, size = 1600;
+	ssize_t retval = 0;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (buf == NULL)
+		return -ENOMEM;
+
+	RXS_ERR("PKTS-ALL", rx_pkts_all);
+	RXS_ERR("BYTES-ALL", rx_bytes_all);
+	RXS_ERR("BEACONS", rx_beacons);
+	RXS_ERR("FRAGS", rx_frags);
+	RXS_ERR("SPECTRAL", rx_spectral);
+
+	RXS_ERR("CRC ERR", crc_err);
+	RXS_ERR("DECRYPT CRC ERR", decrypt_crc_err);
+	RXS_ERR("PHY ERR", phy_err);
+	RXS_ERR("MIC ERR", mic_err);
+	RXS_ERR("PRE-DELIM CRC ERR", pre_delim_crc_err);
+	RXS_ERR("POST-DELIM CRC ERR", post_delim_crc_err);
+	RXS_ERR("DECRYPT BUSY ERR", decrypt_busy_err);
+	RXS_ERR("LENGTH-ERR", rx_len_err);
+	RXS_ERR("OOM-ERR", rx_oom_err);
+	RXS_ERR("RATE-ERR", rx_rate_err);
+	RXS_ERR("TOO-MANY-FRAGS", rx_too_many_frags_err);
+
+	if (len > size)
+		len = size;
+
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	return retval;
+
+#undef RXS_ERR
+}
+
+static const struct file_operations fops_recv = {
+	.read = read_file_recv,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+void ath9k_cmn_debug_recv(struct dentry *debugfs_phy,
+			  struct ath_rx_stats *rxstats)
+{
+	debugfs_create_file("recv", S_IRUSR, debugfs_phy, rxstats,
+			    &fops_recv);
+}
+EXPORT_SYMBOL(ath9k_cmn_debug_recv);
+
+static ssize_t read_file_phy_err(struct file *file, char __user *user_buf,
+				 size_t count, loff_t *ppos)
+{
+#define PHY_ERR(s, p) \
+	len += scnprintf(buf + len, size - len, "%22s : %10u\n", s, \
+			 rxstats->phy_err_stats[p]);
+
+	struct ath_rx_stats *rxstats = file->private_data;
+	char *buf;
+	unsigned int len = 0, size = 1600;
+	ssize_t retval = 0;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (buf == NULL)
+		return -ENOMEM;
+
+	PHY_ERR("UNDERRUN ERR", ATH9K_PHYERR_UNDERRUN);
+	PHY_ERR("TIMING ERR", ATH9K_PHYERR_TIMING);
+	PHY_ERR("PARITY ERR", ATH9K_PHYERR_PARITY);
+	PHY_ERR("RATE ERR", ATH9K_PHYERR_RATE);
+	PHY_ERR("LENGTH ERR", ATH9K_PHYERR_LENGTH);
+	PHY_ERR("RADAR ERR", ATH9K_PHYERR_RADAR);
+	PHY_ERR("SERVICE ERR", ATH9K_PHYERR_SERVICE);
+	PHY_ERR("TOR ERR", ATH9K_PHYERR_TOR);
+	PHY_ERR("OFDM-TIMING ERR", ATH9K_PHYERR_OFDM_TIMING);
+	PHY_ERR("OFDM-SIGNAL-PARITY ERR", ATH9K_PHYERR_OFDM_SIGNAL_PARITY);
+	PHY_ERR("OFDM-RATE ERR", ATH9K_PHYERR_OFDM_RATE_ILLEGAL);
+	PHY_ERR("OFDM-LENGTH ERR", ATH9K_PHYERR_OFDM_LENGTH_ILLEGAL);
+	PHY_ERR("OFDM-POWER-DROP ERR", ATH9K_PHYERR_OFDM_POWER_DROP);
+	PHY_ERR("OFDM-SERVICE ERR", ATH9K_PHYERR_OFDM_SERVICE);
+	PHY_ERR("OFDM-RESTART ERR", ATH9K_PHYERR_OFDM_RESTART);
+	PHY_ERR("FALSE-RADAR-EXT ERR", ATH9K_PHYERR_FALSE_RADAR_EXT);
+	PHY_ERR("CCK-TIMING ERR", ATH9K_PHYERR_CCK_TIMING);
+	PHY_ERR("CCK-HEADER-CRC ERR", ATH9K_PHYERR_CCK_HEADER_CRC);
+	PHY_ERR("CCK-RATE ERR", ATH9K_PHYERR_CCK_RATE_ILLEGAL);
+	PHY_ERR("CCK-SERVICE ERR", ATH9K_PHYERR_CCK_SERVICE);
+	PHY_ERR("CCK-RESTART ERR", ATH9K_PHYERR_CCK_RESTART);
+	PHY_ERR("CCK-LENGTH ERR", ATH9K_PHYERR_CCK_LENGTH_ILLEGAL);
+	PHY_ERR("CCK-POWER-DROP ERR", ATH9K_PHYERR_CCK_POWER_DROP);
+	PHY_ERR("HT-CRC ERR", ATH9K_PHYERR_HT_CRC_ERROR);
+	PHY_ERR("HT-LENGTH ERR", ATH9K_PHYERR_HT_LENGTH_ILLEGAL);
+	PHY_ERR("HT-RATE ERR", ATH9K_PHYERR_HT_RATE_ILLEGAL);
+
+	if (len > size)
+		len = size;
+
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	return retval;
+
+#undef PHY_ERR
+}
+
+static const struct file_operations fops_phy_err = {
+	.read = read_file_phy_err,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+void ath9k_cmn_debug_phy_err(struct dentry *debugfs_phy,
+			     struct ath_rx_stats *rxstats)
+{
+	debugfs_create_file("phy_err", S_IRUSR, debugfs_phy, rxstats,
+			    &fops_phy_err);
+}
+EXPORT_SYMBOL(ath9k_cmn_debug_phy_err);
diff --git a/drivers/net/wireless/ath/ath9k/common-debug.h b/drivers/net/wireless/ath/ath9k/common-debug.h
new file mode 100644
index 000000000000..7c9788490f7f
--- /dev/null
+++ b/drivers/net/wireless/ath/ath9k/common-debug.h
@@ -0,0 +1,72 @@
+/*
+ * Copyright (c) 2008-2011 Atheros Communications Inc.
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+
+
+
+/**
+ * struct ath_rx_stats - RX Statistics
+ * @rx_pkts_all:  No. of total frames received, including ones that
+	may have had errors.
+ * @rx_bytes_all:  No. of total bytes received, including ones that
+	may have had errors.
+ * @crc_err: No. of frames with incorrect CRC value
+ * @decrypt_crc_err: No. of frames whose CRC check failed after
+	decryption process completed
+ * @phy_err: No. of frames whose reception failed because the PHY
+	encountered an error
+ * @mic_err: No. of frames with incorrect TKIP MIC verification failure
+ * @pre_delim_crc_err: Pre-Frame delimiter CRC error detections
+ * @post_delim_crc_err: Post-Frame delimiter CRC error detections
+ * @decrypt_busy_err: Decryption interruptions counter
+ * @phy_err_stats: Individual PHY error statistics
+ * @rx_len_err:  No. of frames discarded due to bad length.
+ * @rx_oom_err:  No. of frames dropped due to OOM issues.
+ * @rx_rate_err:  No. of frames dropped due to rate errors.
+ * @rx_too_many_frags_err:  Frames dropped due to too-many-frags received.
+ * @rx_beacons:  No. of beacons received.
+ * @rx_frags:  No. of rx-fragements received.
+ * @rx_spectral: No of spectral packets received.
+ */
+struct ath_rx_stats {
+	u32 rx_pkts_all;
+	u32 rx_bytes_all;
+	u32 crc_err;
+	u32 decrypt_crc_err;
+	u32 phy_err;
+	u32 mic_err;
+	u32 pre_delim_crc_err;
+	u32 post_delim_crc_err;
+	u32 decrypt_busy_err;
+	u32 phy_err_stats[ATH9K_PHYERR_MAX];
+	u32 rx_len_err;
+	u32 rx_oom_err;
+	u32 rx_rate_err;
+	u32 rx_too_many_frags_err;
+	u32 rx_beacons;
+	u32 rx_frags;
+	u32 rx_spectral;
+};
+
+void ath9k_cmn_debug_modal_eeprom(struct dentry *debugfs_phy,
+				  struct ath_hw *ah);
+void ath9k_cmn_debug_base_eeprom(struct dentry *debugfs_phy,
+				 struct ath_hw *ah);
+void ath9k_cmn_debug_stat_rx(struct ath_rx_stats *rxstats,
+			     struct ath_rx_status *rs);
+void ath9k_cmn_debug_recv(struct dentry *debugfs_phy,
+			  struct ath_rx_stats *rxstats);
+void ath9k_cmn_debug_phy_err(struct dentry *debugfs_phy,
+			     struct ath_rx_stats *rxstats);
diff --git a/drivers/net/wireless/ath/ath9k/common.h b/drivers/net/wireless/ath/ath9k/common.h
index ca38116838f0..ffc454b18637 100644
--- a/drivers/net/wireless/ath/ath9k/common.h
+++ b/drivers/net/wireless/ath/ath9k/common.h
@@ -23,6 +23,7 @@
 
 #include "common-init.h"
 #include "common-beacon.h"
+#include "common-debug.h"
 
 /* Common header for Atheros 802.11n base driver cores */
 
diff --git a/drivers/net/wireless/ath/ath9k/debug.c b/drivers/net/wireless/ath/ath9k/debug.c
index 780ff1bee6f6..6cc42be48d4e 100644
--- a/drivers/net/wireless/ath/ath9k/debug.c
+++ b/drivers/net/wireless/ath/ath9k/debug.c
@@ -948,151 +948,11 @@ static const struct file_operations fops_reset = {
 	.llseek = default_llseek,
 };
 
-static ssize_t read_file_recv(struct file *file, char __user *user_buf,
-			      size_t count, loff_t *ppos)
-{
-#define RXS_ERR(s, e)					    \
-	do {						    \
-		len += scnprintf(buf + len, size - len,	    \
-				 "%18s : %10u\n", s,	    \
-				 sc->debug.stats.rxstats.e);\
-	} while (0)
-
-	struct ath_softc *sc = file->private_data;
-	char *buf;
-	unsigned int len = 0, size = 1600;
-	ssize_t retval = 0;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	RXS_ERR("PKTS-ALL", rx_pkts_all);
-	RXS_ERR("BYTES-ALL", rx_bytes_all);
-	RXS_ERR("BEACONS", rx_beacons);
-	RXS_ERR("FRAGS", rx_frags);
-	RXS_ERR("SPECTRAL", rx_spectral);
-
-	RXS_ERR("CRC ERR", crc_err);
-	RXS_ERR("DECRYPT CRC ERR", decrypt_crc_err);
-	RXS_ERR("PHY ERR", phy_err);
-	RXS_ERR("MIC ERR", mic_err);
-	RXS_ERR("PRE-DELIM CRC ERR", pre_delim_crc_err);
-	RXS_ERR("POST-DELIM CRC ERR", post_delim_crc_err);
-	RXS_ERR("DECRYPT BUSY ERR", decrypt_busy_err);
-	RXS_ERR("LENGTH-ERR", rx_len_err);
-	RXS_ERR("OOM-ERR", rx_oom_err);
-	RXS_ERR("RATE-ERR", rx_rate_err);
-	RXS_ERR("TOO-MANY-FRAGS", rx_too_many_frags_err);
-
-	if (len > size)
-		len = size;
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-
-#undef RXS_ERR
-}
-
 void ath_debug_stat_rx(struct ath_softc *sc, struct ath_rx_status *rs)
 {
-#define RX_PHY_ERR_INC(c) sc->debug.stats.rxstats.phy_err_stats[c]++
-
-	RX_STAT_INC(rx_pkts_all);
-	sc->debug.stats.rxstats.rx_bytes_all += rs->rs_datalen;
-
-	if (rs->rs_status & ATH9K_RXERR_CRC)
-		RX_STAT_INC(crc_err);
-	if (rs->rs_status & ATH9K_RXERR_DECRYPT)
-		RX_STAT_INC(decrypt_crc_err);
-	if (rs->rs_status & ATH9K_RXERR_MIC)
-		RX_STAT_INC(mic_err);
-	if (rs->rs_status & ATH9K_RX_DELIM_CRC_PRE)
-		RX_STAT_INC(pre_delim_crc_err);
-	if (rs->rs_status & ATH9K_RX_DELIM_CRC_POST)
-		RX_STAT_INC(post_delim_crc_err);
-	if (rs->rs_status & ATH9K_RX_DECRYPT_BUSY)
-		RX_STAT_INC(decrypt_busy_err);
-
-	if (rs->rs_status & ATH9K_RXERR_PHY) {
-		RX_STAT_INC(phy_err);
-		if (rs->rs_phyerr < ATH9K_PHYERR_MAX)
-			RX_PHY_ERR_INC(rs->rs_phyerr);
-	}
-
-#undef RX_PHY_ERR_INC
+	ath9k_cmn_debug_stat_rx(&sc->debug.stats.rxstats, rs);
 }
 
-static const struct file_operations fops_recv = {
-	.read = read_file_recv,
-	.open = simple_open,
-	.owner = THIS_MODULE,
-	.llseek = default_llseek,
-};
-
-static ssize_t read_file_phy_err(struct file *file, char __user *user_buf,
-				 size_t count, loff_t *ppos)
-{
-#define PHY_ERR(s, p) \
-	len += scnprintf(buf + len, size - len, "%22s : %10u\n", s, \
-			 sc->debug.stats.rxstats.phy_err_stats[p]);
-
-	struct ath_softc *sc = file->private_data;
-	char *buf;
-	unsigned int len = 0, size = 1600;
-	ssize_t retval = 0;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	PHY_ERR("UNDERRUN ERR", ATH9K_PHYERR_UNDERRUN);
-	PHY_ERR("TIMING ERR", ATH9K_PHYERR_TIMING);
-	PHY_ERR("PARITY ERR", ATH9K_PHYERR_PARITY);
-	PHY_ERR("RATE ERR", ATH9K_PHYERR_RATE);
-	PHY_ERR("LENGTH ERR", ATH9K_PHYERR_LENGTH);
-	PHY_ERR("RADAR ERR", ATH9K_PHYERR_RADAR);
-	PHY_ERR("SERVICE ERR", ATH9K_PHYERR_SERVICE);
-	PHY_ERR("TOR ERR", ATH9K_PHYERR_TOR);
-	PHY_ERR("OFDM-TIMING ERR", ATH9K_PHYERR_OFDM_TIMING);
-	PHY_ERR("OFDM-SIGNAL-PARITY ERR", ATH9K_PHYERR_OFDM_SIGNAL_PARITY);
-	PHY_ERR("OFDM-RATE ERR", ATH9K_PHYERR_OFDM_RATE_ILLEGAL);
-	PHY_ERR("OFDM-LENGTH ERR", ATH9K_PHYERR_OFDM_LENGTH_ILLEGAL);
-	PHY_ERR("OFDM-POWER-DROP ERR", ATH9K_PHYERR_OFDM_POWER_DROP);
-	PHY_ERR("OFDM-SERVICE ERR", ATH9K_PHYERR_OFDM_SERVICE);
-	PHY_ERR("OFDM-RESTART ERR", ATH9K_PHYERR_OFDM_RESTART);
-	PHY_ERR("FALSE-RADAR-EXT ERR", ATH9K_PHYERR_FALSE_RADAR_EXT);
-	PHY_ERR("CCK-TIMING ERR", ATH9K_PHYERR_CCK_TIMING);
-	PHY_ERR("CCK-HEADER-CRC ERR", ATH9K_PHYERR_CCK_HEADER_CRC);
-	PHY_ERR("CCK-RATE ERR", ATH9K_PHYERR_CCK_RATE_ILLEGAL);
-	PHY_ERR("CCK-SERVICE ERR", ATH9K_PHYERR_CCK_SERVICE);
-	PHY_ERR("CCK-RESTART ERR", ATH9K_PHYERR_CCK_RESTART);
-	PHY_ERR("CCK-LENGTH ERR", ATH9K_PHYERR_CCK_LENGTH_ILLEGAL);
-	PHY_ERR("CCK-POWER-DROP ERR", ATH9K_PHYERR_CCK_POWER_DROP);
-	PHY_ERR("HT-CRC ERR", ATH9K_PHYERR_HT_CRC_ERROR);
-	PHY_ERR("HT-LENGTH ERR", ATH9K_PHYERR_HT_LENGTH_ILLEGAL);
-	PHY_ERR("HT-RATE ERR", ATH9K_PHYERR_HT_RATE_ILLEGAL);
-
-	if (len > size)
-		len = size;
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-
-#undef PHY_ERR
-}
-
-static const struct file_operations fops_phy_err = {
-	.read = read_file_phy_err,
-	.open = simple_open,
-	.owner = THIS_MODULE,
-	.llseek = default_llseek,
-};
-
 static ssize_t read_file_regidx(struct file *file, char __user *user_buf,
                                 size_t count, loff_t *ppos)
 {
@@ -1268,62 +1128,6 @@ static const struct file_operations fops_dump_nfcal = {
 	.llseek = default_llseek,
 };
 
-static ssize_t read_file_base_eeprom(struct file *file, char __user *user_buf,
-				     size_t count, loff_t *ppos)
-{
-	struct ath_softc *sc = file->private_data;
-	struct ath_hw *ah = sc->sc_ah;
-	u32 len = 0, size = 1500;
-	ssize_t retval = 0;
-	char *buf;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
-
-	len = ah->eep_ops->dump_eeprom(ah, true, buf, len, size);
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-}
-
-static const struct file_operations fops_base_eeprom = {
-	.read = read_file_base_eeprom,
-	.open = simple_open,
-	.owner = THIS_MODULE,
-	.llseek = default_llseek,
-};
-
-static ssize_t read_file_modal_eeprom(struct file *file, char __user *user_buf,
-				      size_t count, loff_t *ppos)
-{
-	struct ath_softc *sc = file->private_data;
-	struct ath_hw *ah = sc->sc_ah;
-	u32 len = 0, size = 6000;
-	char *buf;
-	size_t retval;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	len = ah->eep_ops->dump_eeprom(ah, false, buf, len, size);
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-}
-
-static const struct file_operations fops_modal_eeprom = {
-	.read = read_file_modal_eeprom,
-	.open = simple_open,
-	.owner = THIS_MODULE,
-	.llseek = default_llseek,
-};
-
 #ifdef CONFIG_ATH9K_BTCOEX_SUPPORT
 static ssize_t read_file_btcoex(struct file *file, char __user *user_buf,
 				size_t count, loff_t *ppos)
@@ -1524,10 +1328,10 @@ int ath9k_init_debug(struct ath_hw *ah)
 			    &fops_misc);
 	debugfs_create_file("reset", S_IRUSR, sc->debug.debugfs_phy, sc,
 			    &fops_reset);
-	debugfs_create_file("recv", S_IRUSR, sc->debug.debugfs_phy, sc,
-			    &fops_recv);
-	debugfs_create_file("phy_err", S_IRUSR, sc->debug.debugfs_phy, sc,
-			    &fops_phy_err);
+
+	ath9k_cmn_debug_recv(sc->debug.debugfs_phy, &sc->debug.stats.rxstats);
+	ath9k_cmn_debug_phy_err(sc->debug.debugfs_phy, &sc->debug.stats.rxstats);
+
 	debugfs_create_u8("rx_chainmask", S_IRUSR, sc->debug.debugfs_phy,
 			  &ah->rxchainmask);
 	debugfs_create_u8("tx_chainmask", S_IRUSR, sc->debug.debugfs_phy,
@@ -1547,10 +1351,10 @@ int ath9k_init_debug(struct ath_hw *ah)
 			    &fops_regdump);
 	debugfs_create_file("dump_nfcal", S_IRUSR, sc->debug.debugfs_phy, sc,
 			    &fops_dump_nfcal);
-	debugfs_create_file("base_eeprom", S_IRUSR, sc->debug.debugfs_phy, sc,
-			    &fops_base_eeprom);
-	debugfs_create_file("modal_eeprom", S_IRUSR, sc->debug.debugfs_phy, sc,
-			    &fops_modal_eeprom);
+
+	ath9k_cmn_debug_base_eeprom(sc->debug.debugfs_phy, sc->sc_ah);
+	ath9k_cmn_debug_modal_eeprom(sc->debug.debugfs_phy, sc->sc_ah);
+
 	debugfs_create_u32("gpio_mask", S_IRUSR | S_IWUSR,
 			   sc->debug.debugfs_phy, &sc->sc_ah->gpio_mask);
 	debugfs_create_u32("gpio_val", S_IRUSR | S_IWUSR,
diff --git a/drivers/net/wireless/ath/ath9k/debug.h b/drivers/net/wireless/ath/ath9k/debug.h
index 559a68c2709c..53ae15bd0c9d 100644
--- a/drivers/net/wireless/ath/ath9k/debug.h
+++ b/drivers/net/wireless/ath/ath9k/debug.h
@@ -221,50 +221,6 @@ struct ath_rx_rate_stats {
 	} cck_stats[4];
 };
 
-/**
- * struct ath_rx_stats - RX Statistics
- * @rx_pkts_all:  No. of total frames received, including ones that
-	may have had errors.
- * @rx_bytes_all:  No. of total bytes received, including ones that
-	may have had errors.
- * @crc_err: No. of frames with incorrect CRC value
- * @decrypt_crc_err: No. of frames whose CRC check failed after
-	decryption process completed
- * @phy_err: No. of frames whose reception failed because the PHY
-	encountered an error
- * @mic_err: No. of frames with incorrect TKIP MIC verification failure
- * @pre_delim_crc_err: Pre-Frame delimiter CRC error detections
- * @post_delim_crc_err: Post-Frame delimiter CRC error detections
- * @decrypt_busy_err: Decryption interruptions counter
- * @phy_err_stats: Individual PHY error statistics
- * @rx_len_err:  No. of frames discarded due to bad length.
- * @rx_oom_err:  No. of frames dropped due to OOM issues.
- * @rx_rate_err:  No. of frames dropped due to rate errors.
- * @rx_too_many_frags_err:  Frames dropped due to too-many-frags received.
- * @rx_beacons:  No. of beacons received.
- * @rx_frags:  No. of rx-fragements received.
- * @rx_spectral: No of spectral packets received.
- */
-struct ath_rx_stats {
-	u32 rx_pkts_all;
-	u32 rx_bytes_all;
-	u32 crc_err;
-	u32 decrypt_crc_err;
-	u32 phy_err;
-	u32 mic_err;
-	u32 pre_delim_crc_err;
-	u32 post_delim_crc_err;
-	u32 decrypt_busy_err;
-	u32 phy_err_stats[ATH9K_PHYERR_MAX];
-	u32 rx_len_err;
-	u32 rx_oom_err;
-	u32 rx_rate_err;
-	u32 rx_too_many_frags_err;
-	u32 rx_beacons;
-	u32 rx_frags;
-	u32 rx_spectral;
-};
-
 #define ANT_MAIN 0
 #define ANT_ALT  1
 
diff --git a/drivers/net/wireless/ath/ath9k/dfs.c b/drivers/net/wireless/ath/ath9k/dfs.c
index 857bb28b3894..726271c7c330 100644
--- a/drivers/net/wireless/ath/ath9k/dfs.c
+++ b/drivers/net/wireless/ath/ath9k/dfs.c
@@ -178,12 +178,12 @@ void ath9k_dfs_process_phyerr(struct ath_softc *sc, void *data,
 	pe.ts = mactime;
 	if (ath9k_postprocess_radar_event(sc, &ard, &pe)) {
 		struct dfs_pattern_detector *pd = sc->dfs_detector;
-		static u64 last_ts;
 		ath_dbg(common, DFS,
 			"ath9k_dfs_process_phyerr: channel=%d, ts=%llu, "
 			"width=%d, rssi=%d, delta_ts=%llu\n",
-			pe.freq, pe.ts, pe.width, pe.rssi, pe.ts-last_ts);
-		last_ts = pe.ts;
+			pe.freq, pe.ts, pe.width, pe.rssi,
+			pe.ts - sc->dfs_prev_pulse_ts);
+		sc->dfs_prev_pulse_ts = pe.ts;
 		DFS_STAT_INC(sc, pulses_processed);
 		if (pd != NULL && pd->add_pulse(pd, &pe)) {
 			DFS_STAT_INC(sc, radar_detected);
diff --git a/drivers/net/wireless/ath/ath9k/htc.h b/drivers/net/wireless/ath/ath9k/htc.h
index dab1f0cab993..09a5d72f3ff5 100644
--- a/drivers/net/wireless/ath/ath9k/htc.h
+++ b/drivers/net/wireless/ath/ath9k/htc.h
@@ -325,14 +325,14 @@ static inline struct ath9k_htc_tx_ctl *HTC_SKB_CB(struct sk_buff *skb)
 
 #define TX_STAT_INC(c) (hif_dev->htc_handle->drv_priv->debug.tx_stats.c++)
 #define TX_STAT_ADD(c, a) (hif_dev->htc_handle->drv_priv->debug.tx_stats.c += a)
-#define RX_STAT_INC(c) (hif_dev->htc_handle->drv_priv->debug.rx_stats.c++)
-#define RX_STAT_ADD(c, a) (hif_dev->htc_handle->drv_priv->debug.rx_stats.c += a)
+#define RX_STAT_INC(c) (hif_dev->htc_handle->drv_priv->debug.skbrx_stats.c++)
+#define RX_STAT_ADD(c, a) (hif_dev->htc_handle->drv_priv->debug.skbrx_stats.c += a)
 #define CAB_STAT_INC   priv->debug.tx_stats.cab_queued++
 
 #define TX_QSTAT_INC(q) (priv->debug.tx_stats.queue_stats[q]++)
 
 void ath9k_htc_err_stat_rx(struct ath9k_htc_priv *priv,
-			   struct ath_htc_rx_status *rxs);
+			   struct ath_rx_status *rs);
 
 struct ath_tx_stats {
 	u32 buf_queued;
@@ -345,25 +345,18 @@ struct ath_tx_stats {
 	u32 queue_stats[IEEE80211_NUM_ACS];
 };
 
-struct ath_rx_stats {
+struct ath_skbrx_stats {
 	u32 skb_allocated;
 	u32 skb_completed;
 	u32 skb_completed_bytes;
 	u32 skb_dropped;
-	u32 err_crc;
-	u32 err_decrypt_crc;
-	u32 err_mic;
-	u32 err_pre_delim;
-	u32 err_post_delim;
-	u32 err_decrypt_busy;
-	u32 err_phy;
-	u32 err_phy_stats[ATH9K_PHYERR_MAX];
 };
 
 struct ath9k_debug {
 	struct dentry *debugfs_phy;
 	struct ath_tx_stats tx_stats;
 	struct ath_rx_stats rx_stats;
+	struct ath_skbrx_stats skbrx_stats;
 };
 
 void ath9k_htc_get_et_strings(struct ieee80211_hw *hw,
@@ -385,7 +378,7 @@ void ath9k_htc_get_et_stats(struct ieee80211_hw *hw,
 #define TX_QSTAT_INC(c) do { } while (0)
 
 static inline void ath9k_htc_err_stat_rx(struct ath9k_htc_priv *priv,
-					 struct ath_htc_rx_status *rxs)
+					 struct ath_rx_status *rs)
 {
 }
 
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_debug.c b/drivers/net/wireless/ath/ath9k/htc_drv_debug.c
index fb071ee4fcfb..8b529e4b8ac4 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_debug.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_debug.c
@@ -243,39 +243,14 @@ static const struct file_operations fops_xmit = {
 };
 
 void ath9k_htc_err_stat_rx(struct ath9k_htc_priv *priv,
-			   struct ath_htc_rx_status *rxs)
+			     struct ath_rx_status *rs)
 {
-#define RX_PHY_ERR_INC(c) priv->debug.rx_stats.err_phy_stats[c]++
-
-	if (rxs->rs_status & ATH9K_RXERR_CRC)
-		priv->debug.rx_stats.err_crc++;
-	if (rxs->rs_status & ATH9K_RXERR_DECRYPT)
-		priv->debug.rx_stats.err_decrypt_crc++;
-	if (rxs->rs_status & ATH9K_RXERR_MIC)
-		priv->debug.rx_stats.err_mic++;
-	if (rxs->rs_status & ATH9K_RX_DELIM_CRC_PRE)
-		priv->debug.rx_stats.err_pre_delim++;
-	if (rxs->rs_status & ATH9K_RX_DELIM_CRC_POST)
-		priv->debug.rx_stats.err_post_delim++;
-	if (rxs->rs_status & ATH9K_RX_DECRYPT_BUSY)
-		priv->debug.rx_stats.err_decrypt_busy++;
-
-	if (rxs->rs_status & ATH9K_RXERR_PHY) {
-		priv->debug.rx_stats.err_phy++;
-		if (rxs->rs_phyerr < ATH9K_PHYERR_MAX)
-			RX_PHY_ERR_INC(rxs->rs_phyerr);
-	}
-
-#undef RX_PHY_ERR_INC
+	ath9k_cmn_debug_stat_rx(&priv->debug.rx_stats, rs);
 }
 
-static ssize_t read_file_recv(struct file *file, char __user *user_buf,
+static ssize_t read_file_skb_rx(struct file *file, char __user *user_buf,
 			      size_t count, loff_t *ppos)
 {
-#define PHY_ERR(s, p)							\
-	len += scnprintf(buf + len, size - len, "%20s : %10u\n", s,	\
-			 priv->debug.rx_stats.err_phy_stats[p]);
-
 	struct ath9k_htc_priv *priv = file->private_data;
 	char *buf;
 	unsigned int len = 0, size = 1500;
@@ -287,63 +262,13 @@ static ssize_t read_file_recv(struct file *file, char __user *user_buf,
 
 	len += scnprintf(buf + len, size - len,
 			 "%20s : %10u\n", "SKBs allocated",
-			 priv->debug.rx_stats.skb_allocated);
+			 priv->debug.skbrx_stats.skb_allocated);
 	len += scnprintf(buf + len, size - len,
 			 "%20s : %10u\n", "SKBs completed",
-			 priv->debug.rx_stats.skb_completed);
+			 priv->debug.skbrx_stats.skb_completed);
 	len += scnprintf(buf + len, size - len,
 			 "%20s : %10u\n", "SKBs Dropped",
-			 priv->debug.rx_stats.skb_dropped);
-
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "CRC ERR",
-			 priv->debug.rx_stats.err_crc);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "DECRYPT CRC ERR",
-			 priv->debug.rx_stats.err_decrypt_crc);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "MIC ERR",
-			 priv->debug.rx_stats.err_mic);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "PRE-DELIM CRC ERR",
-			 priv->debug.rx_stats.err_pre_delim);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "POST-DELIM CRC ERR",
-			 priv->debug.rx_stats.err_post_delim);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "DECRYPT BUSY ERR",
-			 priv->debug.rx_stats.err_decrypt_busy);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10u\n", "TOTAL PHY ERR",
-			 priv->debug.rx_stats.err_phy);
-
-
-	PHY_ERR("UNDERRUN", ATH9K_PHYERR_UNDERRUN);
-	PHY_ERR("TIMING", ATH9K_PHYERR_TIMING);
-	PHY_ERR("PARITY", ATH9K_PHYERR_PARITY);
-	PHY_ERR("RATE", ATH9K_PHYERR_RATE);
-	PHY_ERR("LENGTH", ATH9K_PHYERR_LENGTH);
-	PHY_ERR("RADAR", ATH9K_PHYERR_RADAR);
-	PHY_ERR("SERVICE", ATH9K_PHYERR_SERVICE);
-	PHY_ERR("TOR", ATH9K_PHYERR_TOR);
-	PHY_ERR("OFDM-TIMING", ATH9K_PHYERR_OFDM_TIMING);
-	PHY_ERR("OFDM-SIGNAL-PARITY", ATH9K_PHYERR_OFDM_SIGNAL_PARITY);
-	PHY_ERR("OFDM-RATE", ATH9K_PHYERR_OFDM_RATE_ILLEGAL);
-	PHY_ERR("OFDM-LENGTH", ATH9K_PHYERR_OFDM_LENGTH_ILLEGAL);
-	PHY_ERR("OFDM-POWER-DROP", ATH9K_PHYERR_OFDM_POWER_DROP);
-	PHY_ERR("OFDM-SERVICE", ATH9K_PHYERR_OFDM_SERVICE);
-	PHY_ERR("OFDM-RESTART", ATH9K_PHYERR_OFDM_RESTART);
-	PHY_ERR("FALSE-RADAR-EXT", ATH9K_PHYERR_FALSE_RADAR_EXT);
-	PHY_ERR("CCK-TIMING", ATH9K_PHYERR_CCK_TIMING);
-	PHY_ERR("CCK-HEADER-CRC", ATH9K_PHYERR_CCK_HEADER_CRC);
-	PHY_ERR("CCK-RATE", ATH9K_PHYERR_CCK_RATE_ILLEGAL);
-	PHY_ERR("CCK-SERVICE", ATH9K_PHYERR_CCK_SERVICE);
-	PHY_ERR("CCK-RESTART", ATH9K_PHYERR_CCK_RESTART);
-	PHY_ERR("CCK-LENGTH", ATH9K_PHYERR_CCK_LENGTH_ILLEGAL);
-	PHY_ERR("CCK-POWER-DROP", ATH9K_PHYERR_CCK_POWER_DROP);
-	PHY_ERR("HT-CRC", ATH9K_PHYERR_HT_CRC_ERROR);
-	PHY_ERR("HT-LENGTH", ATH9K_PHYERR_HT_LENGTH_ILLEGAL);
-	PHY_ERR("HT-RATE", ATH9K_PHYERR_HT_RATE_ILLEGAL);
+			 priv->debug.skbrx_stats.skb_dropped);
 
 	if (len > size)
 		len = size;
@@ -352,12 +277,10 @@ static ssize_t read_file_recv(struct file *file, char __user *user_buf,
 	kfree(buf);
 
 	return retval;
-
-#undef PHY_ERR
 }
 
-static const struct file_operations fops_recv = {
-	.read = read_file_recv,
+static const struct file_operations fops_skb_rx = {
+	.read = read_file_skb_rx,
 	.open = simple_open,
 	.owner = THIS_MODULE,
 	.llseek = default_llseek,
@@ -486,423 +409,6 @@ static const struct file_operations fops_debug = {
 	.llseek = default_llseek,
 };
 
-static ssize_t read_file_base_eeprom(struct file *file, char __user *user_buf,
-				     size_t count, loff_t *ppos)
-{
-	struct ath9k_htc_priv *priv = file->private_data;
-	struct ath_common *common = ath9k_hw_common(priv->ah);
-	struct base_eep_header *pBase = NULL;
-	unsigned int len = 0, size = 1500;
-	ssize_t retval = 0;
-	char *buf;
-
-	pBase = ath9k_htc_get_eeprom_base(priv);
-
-	if (pBase == NULL) {
-		ath_err(common, "Unknown EEPROM type\n");
-		return 0;
-	}
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n", "Major Version",
-			 pBase->version >> 12);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n", "Minor Version",
-			 pBase->version & 0xFFF);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n", "Checksum",
-			 pBase->checksum);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n", "Length",
-			 pBase->length);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n", "RegDomain1",
-			 pBase->regDmn[0]);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n", "RegDomain2",
-			 pBase->regDmn[1]);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "TX Mask", pBase->txMask);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "RX Mask", pBase->rxMask);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Allow 5GHz",
-			 !!(pBase->opCapFlags & AR5416_OPFLAGS_11A));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Allow 2GHz",
-			 !!(pBase->opCapFlags & AR5416_OPFLAGS_11G));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Disable 2GHz HT20",
-			 !!(pBase->opCapFlags & AR5416_OPFLAGS_N_2G_HT20));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Disable 2GHz HT40",
-			 !!(pBase->opCapFlags & AR5416_OPFLAGS_N_2G_HT40));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Disable 5Ghz HT20",
-			 !!(pBase->opCapFlags & AR5416_OPFLAGS_N_5G_HT20));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Disable 5Ghz HT40",
-			 !!(pBase->opCapFlags & AR5416_OPFLAGS_N_5G_HT40));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Big Endian",
-			 !!(pBase->eepMisc & 0x01));
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Cal Bin Major Ver",
-			 (pBase->binBuildNumber >> 24) & 0xFF);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Cal Bin Minor Ver",
-			 (pBase->binBuildNumber >> 16) & 0xFF);
-	len += scnprintf(buf + len, size - len,
-			 "%20s : %10d\n",
-			 "Cal Bin Build",
-			 (pBase->binBuildNumber >> 8) & 0xFF);
-
-	/*
-	 * UB91 specific data.
-	 */
-	if (AR_SREV_9271(priv->ah)) {
-		struct base_eep_header_4k *pBase4k =
-			&priv->ah->eeprom.map4k.baseEepHeader;
-
-		len += scnprintf(buf + len, size - len,
-				 "%20s : %10d\n",
-				 "TX Gain type",
-				 pBase4k->txGainType);
-	}
-
-	/*
-	 * UB95 specific data.
-	 */
-	if (priv->ah->hw_version.usbdev == AR9287_USB) {
-		struct base_eep_ar9287_header *pBase9287 =
-			&priv->ah->eeprom.map9287.baseEepHeader;
-
-		len += scnprintf(buf + len, size - len,
-				 "%20s : %10ddB\n",
-				 "Power Table Offset",
-				 pBase9287->pwrTableOffset);
-
-		len += scnprintf(buf + len, size - len,
-				 "%20s : %10d\n",
-				 "OpenLoop Power Ctrl",
-				 pBase9287->openLoopPwrCntl);
-	}
-
-	len += scnprintf(buf + len, size - len, "%20s : %pM\n", "MacAddress",
-			 pBase->macAddr);
-	if (len > size)
-		len = size;
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-}
-
-static const struct file_operations fops_base_eeprom = {
-	.read = read_file_base_eeprom,
-	.open = simple_open,
-	.owner = THIS_MODULE,
-	.llseek = default_llseek,
-};
-
-static ssize_t read_4k_modal_eeprom(struct file *file,
-				    char __user *user_buf,
-				    size_t count, loff_t *ppos)
-{
-#define PR_EEP(_s, _val)						\
-	do {								\
-		len += scnprintf(buf + len, size - len, "%20s : %10d\n",\
-				 _s, (_val));				\
-	} while (0)
-
-	struct ath9k_htc_priv *priv = file->private_data;
-	struct modal_eep_4k_header *pModal = &priv->ah->eeprom.map4k.modalHeader;
-	unsigned int len = 0, size = 2048;
-	ssize_t retval = 0;
-	char *buf;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	PR_EEP("Chain0 Ant. Control", pModal->antCtrlChain[0]);
-	PR_EEP("Ant. Common Control", pModal->antCtrlCommon);
-	PR_EEP("Chain0 Ant. Gain", pModal->antennaGainCh[0]);
-	PR_EEP("Switch Settle", pModal->switchSettling);
-	PR_EEP("Chain0 TxRxAtten", pModal->txRxAttenCh[0]);
-	PR_EEP("Chain0 RxTxMargin", pModal->rxTxMarginCh[0]);
-	PR_EEP("ADC Desired size", pModal->adcDesiredSize);
-	PR_EEP("PGA Desired size", pModal->pgaDesiredSize);
-	PR_EEP("Chain0 xlna Gain", pModal->xlnaGainCh[0]);
-	PR_EEP("txEndToXpaOff", pModal->txEndToXpaOff);
-	PR_EEP("txEndToRxOn", pModal->txEndToRxOn);
-	PR_EEP("txFrameToXpaOn", pModal->txFrameToXpaOn);
-	PR_EEP("CCA Threshold)", pModal->thresh62);
-	PR_EEP("Chain0 NF Threshold", pModal->noiseFloorThreshCh[0]);
-	PR_EEP("xpdGain", pModal->xpdGain);
-	PR_EEP("External PD", pModal->xpd);
-	PR_EEP("Chain0 I Coefficient", pModal->iqCalICh[0]);
-	PR_EEP("Chain0 Q Coefficient", pModal->iqCalQCh[0]);
-	PR_EEP("pdGainOverlap", pModal->pdGainOverlap);
-	PR_EEP("O/D Bias Version", pModal->version);
-	PR_EEP("CCK OutputBias", pModal->ob_0);
-	PR_EEP("BPSK OutputBias", pModal->ob_1);
-	PR_EEP("QPSK OutputBias", pModal->ob_2);
-	PR_EEP("16QAM OutputBias", pModal->ob_3);
-	PR_EEP("64QAM OutputBias", pModal->ob_4);
-	PR_EEP("CCK Driver1_Bias", pModal->db1_0);
-	PR_EEP("BPSK Driver1_Bias", pModal->db1_1);
-	PR_EEP("QPSK Driver1_Bias", pModal->db1_2);
-	PR_EEP("16QAM Driver1_Bias", pModal->db1_3);
-	PR_EEP("64QAM Driver1_Bias", pModal->db1_4);
-	PR_EEP("CCK Driver2_Bias", pModal->db2_0);
-	PR_EEP("BPSK Driver2_Bias", pModal->db2_1);
-	PR_EEP("QPSK Driver2_Bias", pModal->db2_2);
-	PR_EEP("16QAM Driver2_Bias", pModal->db2_3);
-	PR_EEP("64QAM Driver2_Bias", pModal->db2_4);
-	PR_EEP("xPA Bias Level", pModal->xpaBiasLvl);
-	PR_EEP("txFrameToDataStart", pModal->txFrameToDataStart);
-	PR_EEP("txFrameToPaOn", pModal->txFrameToPaOn);
-	PR_EEP("HT40 Power Inc.", pModal->ht40PowerIncForPdadc);
-	PR_EEP("Chain0 bswAtten", pModal->bswAtten[0]);
-	PR_EEP("Chain0 bswMargin", pModal->bswMargin[0]);
-	PR_EEP("HT40 Switch Settle", pModal->swSettleHt40);
-	PR_EEP("Chain0 xatten2Db", pModal->xatten2Db[0]);
-	PR_EEP("Chain0 xatten2Margin", pModal->xatten2Margin[0]);
-	PR_EEP("Ant. Diversity ctl1", pModal->antdiv_ctl1);
-	PR_EEP("Ant. Diversity ctl2", pModal->antdiv_ctl2);
-	PR_EEP("TX Diversity", pModal->tx_diversity);
-
-	if (len > size)
-		len = size;
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-
-#undef PR_EEP
-}
-
-static ssize_t read_def_modal_eeprom(struct file *file,
-				     char __user *user_buf,
-				     size_t count, loff_t *ppos)
-{
-#define PR_EEP(_s, _val)						\
-	do {								\
-		if (pBase->opCapFlags & AR5416_OPFLAGS_11G) {		\
-			pModal = &priv->ah->eeprom.def.modalHeader[1];	\
-			len += scnprintf(buf + len, size - len, "%20s : %8d%7s", \
-					 _s, (_val), "|");		\
-		}							\
-		if (pBase->opCapFlags & AR5416_OPFLAGS_11A) {		\
-			pModal = &priv->ah->eeprom.def.modalHeader[0];	\
-			len += scnprintf(buf + len, size - len, "%9d\n",\
-					(_val));			\
-		}							\
-	} while (0)
-
-	struct ath9k_htc_priv *priv = file->private_data;
-	struct base_eep_header *pBase = &priv->ah->eeprom.def.baseEepHeader;
-	struct modal_eep_header *pModal = NULL;
-	unsigned int len = 0, size = 3500;
-	ssize_t retval = 0;
-	char *buf;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	len += scnprintf(buf + len, size - len,
-			 "%31s %15s\n", "2G", "5G");
-	len += scnprintf(buf + len, size - len,
-			 "%32s %16s\n", "====", "====\n");
-
-	PR_EEP("Chain0 Ant. Control", pModal->antCtrlChain[0]);
-	PR_EEP("Chain1 Ant. Control", pModal->antCtrlChain[1]);
-	PR_EEP("Chain2 Ant. Control", pModal->antCtrlChain[2]);
-	PR_EEP("Ant. Common Control", pModal->antCtrlCommon);
-	PR_EEP("Chain0 Ant. Gain", pModal->antennaGainCh[0]);
-	PR_EEP("Chain1 Ant. Gain", pModal->antennaGainCh[1]);
-	PR_EEP("Chain2 Ant. Gain", pModal->antennaGainCh[2]);
-	PR_EEP("Switch Settle", pModal->switchSettling);
-	PR_EEP("Chain0 TxRxAtten", pModal->txRxAttenCh[0]);
-	PR_EEP("Chain1 TxRxAtten", pModal->txRxAttenCh[1]);
-	PR_EEP("Chain2 TxRxAtten", pModal->txRxAttenCh[2]);
-	PR_EEP("Chain0 RxTxMargin", pModal->rxTxMarginCh[0]);
-	PR_EEP("Chain1 RxTxMargin", pModal->rxTxMarginCh[1]);
-	PR_EEP("Chain2 RxTxMargin", pModal->rxTxMarginCh[2]);
-	PR_EEP("ADC Desired size", pModal->adcDesiredSize);
-	PR_EEP("PGA Desired size", pModal->pgaDesiredSize);
-	PR_EEP("Chain0 xlna Gain", pModal->xlnaGainCh[0]);
-	PR_EEP("Chain1 xlna Gain", pModal->xlnaGainCh[1]);
-	PR_EEP("Chain2 xlna Gain", pModal->xlnaGainCh[2]);
-	PR_EEP("txEndToXpaOff", pModal->txEndToXpaOff);
-	PR_EEP("txEndToRxOn", pModal->txEndToRxOn);
-	PR_EEP("txFrameToXpaOn", pModal->txFrameToXpaOn);
-	PR_EEP("CCA Threshold)", pModal->thresh62);
-	PR_EEP("Chain0 NF Threshold", pModal->noiseFloorThreshCh[0]);
-	PR_EEP("Chain1 NF Threshold", pModal->noiseFloorThreshCh[1]);
-	PR_EEP("Chain2 NF Threshold", pModal->noiseFloorThreshCh[2]);
-	PR_EEP("xpdGain", pModal->xpdGain);
-	PR_EEP("External PD", pModal->xpd);
-	PR_EEP("Chain0 I Coefficient", pModal->iqCalICh[0]);
-	PR_EEP("Chain1 I Coefficient", pModal->iqCalICh[1]);
-	PR_EEP("Chain2 I Coefficient", pModal->iqCalICh[2]);
-	PR_EEP("Chain0 Q Coefficient", pModal->iqCalQCh[0]);
-	PR_EEP("Chain1 Q Coefficient", pModal->iqCalQCh[1]);
-	PR_EEP("Chain2 Q Coefficient", pModal->iqCalQCh[2]);
-	PR_EEP("pdGainOverlap", pModal->pdGainOverlap);
-	PR_EEP("Chain0 OutputBias", pModal->ob);
-	PR_EEP("Chain0 DriverBias", pModal->db);
-	PR_EEP("xPA Bias Level", pModal->xpaBiasLvl);
-	PR_EEP("2chain pwr decrease", pModal->pwrDecreaseFor2Chain);
-	PR_EEP("3chain pwr decrease", pModal->pwrDecreaseFor3Chain);
-	PR_EEP("txFrameToDataStart", pModal->txFrameToDataStart);
-	PR_EEP("txFrameToPaOn", pModal->txFrameToPaOn);
-	PR_EEP("HT40 Power Inc.", pModal->ht40PowerIncForPdadc);
-	PR_EEP("Chain0 bswAtten", pModal->bswAtten[0]);
-	PR_EEP("Chain1 bswAtten", pModal->bswAtten[1]);
-	PR_EEP("Chain2 bswAtten", pModal->bswAtten[2]);
-	PR_EEP("Chain0 bswMargin", pModal->bswMargin[0]);
-	PR_EEP("Chain1 bswMargin", pModal->bswMargin[1]);
-	PR_EEP("Chain2 bswMargin", pModal->bswMargin[2]);
-	PR_EEP("HT40 Switch Settle", pModal->swSettleHt40);
-	PR_EEP("Chain0 xatten2Db", pModal->xatten2Db[0]);
-	PR_EEP("Chain1 xatten2Db", pModal->xatten2Db[1]);
-	PR_EEP("Chain2 xatten2Db", pModal->xatten2Db[2]);
-	PR_EEP("Chain0 xatten2Margin", pModal->xatten2Margin[0]);
-	PR_EEP("Chain1 xatten2Margin", pModal->xatten2Margin[1]);
-	PR_EEP("Chain2 xatten2Margin", pModal->xatten2Margin[2]);
-	PR_EEP("Chain1 OutputBias", pModal->ob_ch1);
-	PR_EEP("Chain1 DriverBias", pModal->db_ch1);
-	PR_EEP("LNA Control", pModal->lna_ctl);
-	PR_EEP("XPA Bias Freq0", pModal->xpaBiasLvlFreq[0]);
-	PR_EEP("XPA Bias Freq1", pModal->xpaBiasLvlFreq[1]);
-	PR_EEP("XPA Bias Freq2", pModal->xpaBiasLvlFreq[2]);
-
-	if (len > size)
-		len = size;
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-
-#undef PR_EEP
-}
-
-static ssize_t read_9287_modal_eeprom(struct file *file,
-				      char __user *user_buf,
-				      size_t count, loff_t *ppos)
-{
-#define PR_EEP(_s, _val)						\
-	do {								\
-		len += scnprintf(buf + len, size - len, "%20s : %10d\n",\
-				 _s, (_val));				\
-	} while (0)
-
-	struct ath9k_htc_priv *priv = file->private_data;
-	struct modal_eep_ar9287_header *pModal = &priv->ah->eeprom.map9287.modalHeader;
-	unsigned int len = 0, size = 3000;
-	ssize_t retval = 0;
-	char *buf;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (buf == NULL)
-		return -ENOMEM;
-
-	PR_EEP("Chain0 Ant. Control", pModal->antCtrlChain[0]);
-	PR_EEP("Chain1 Ant. Control", pModal->antCtrlChain[1]);
-	PR_EEP("Ant. Common Control", pModal->antCtrlCommon);
-	PR_EEP("Chain0 Ant. Gain", pModal->antennaGainCh[0]);
-	PR_EEP("Chain1 Ant. Gain", pModal->antennaGainCh[1]);
-	PR_EEP("Switch Settle", pModal->switchSettling);
-	PR_EEP("Chain0 TxRxAtten", pModal->txRxAttenCh[0]);
-	PR_EEP("Chain1 TxRxAtten", pModal->txRxAttenCh[1]);
-	PR_EEP("Chain0 RxTxMargin", pModal->rxTxMarginCh[0]);
-	PR_EEP("Chain1 RxTxMargin", pModal->rxTxMarginCh[1]);
-	PR_EEP("ADC Desired size", pModal->adcDesiredSize);
-	PR_EEP("txEndToXpaOff", pModal->txEndToXpaOff);
-	PR_EEP("txEndToRxOn", pModal->txEndToRxOn);
-	PR_EEP("txFrameToXpaOn", pModal->txFrameToXpaOn);
-	PR_EEP("CCA Threshold)", pModal->thresh62);
-	PR_EEP("Chain0 NF Threshold", pModal->noiseFloorThreshCh[0]);
-	PR_EEP("Chain1 NF Threshold", pModal->noiseFloorThreshCh[1]);
-	PR_EEP("xpdGain", pModal->xpdGain);
-	PR_EEP("External PD", pModal->xpd);
-	PR_EEP("Chain0 I Coefficient", pModal->iqCalICh[0]);
-	PR_EEP("Chain1 I Coefficient", pModal->iqCalICh[1]);
-	PR_EEP("Chain0 Q Coefficient", pModal->iqCalQCh[0]);
-	PR_EEP("Chain1 Q Coefficient", pModal->iqCalQCh[1]);
-	PR_EEP("pdGainOverlap", pModal->pdGainOverlap);
-	PR_EEP("xPA Bias Level", pModal->xpaBiasLvl);
-	PR_EEP("txFrameToDataStart", pModal->txFrameToDataStart);
-	PR_EEP("txFrameToPaOn", pModal->txFrameToPaOn);
-	PR_EEP("HT40 Power Inc.", pModal->ht40PowerIncForPdadc);
-	PR_EEP("Chain0 bswAtten", pModal->bswAtten[0]);
-	PR_EEP("Chain1 bswAtten", pModal->bswAtten[1]);
-	PR_EEP("Chain0 bswMargin", pModal->bswMargin[0]);
-	PR_EEP("Chain1 bswMargin", pModal->bswMargin[1]);
-	PR_EEP("HT40 Switch Settle", pModal->swSettleHt40);
-	PR_EEP("AR92x7 Version", pModal->version);
-	PR_EEP("DriverBias1", pModal->db1);
-	PR_EEP("DriverBias2", pModal->db1);
-	PR_EEP("CCK OutputBias", pModal->ob_cck);
-	PR_EEP("PSK OutputBias", pModal->ob_psk);
-	PR_EEP("QAM OutputBias", pModal->ob_qam);
-	PR_EEP("PAL_OFF OutputBias", pModal->ob_pal_off);
-
-	if (len > size)
-		len = size;
-
-	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
-	kfree(buf);
-
-	return retval;
-
-#undef PR_EEP
-}
-
-static ssize_t read_file_modal_eeprom(struct file *file, char __user *user_buf,
-				      size_t count, loff_t *ppos)
-{
-	struct ath9k_htc_priv *priv = file->private_data;
-
-	if (AR_SREV_9271(priv->ah))
-		return read_4k_modal_eeprom(file, user_buf, count, ppos);
-	else if (priv->ah->hw_version.usbdev == AR9280_USB)
-		return read_def_modal_eeprom(file, user_buf, count, ppos);
-	else if (priv->ah->hw_version.usbdev == AR9287_USB)
-		return read_9287_modal_eeprom(file, user_buf, count, ppos);
-
-	return 0;
-}
-
-static const struct file_operations fops_modal_eeprom = {
-	.read = read_file_modal_eeprom,
-	.open = simple_open,
-	.owner = THIS_MODULE,
-	.llseek = default_llseek,
-};
-
-
 /* Ethtool support for get-stats */
 #define AMKSTR(nm) #nm "_BE", #nm "_BK", #nm "_VI", #nm "_VO"
 static const char ath9k_htc_gstrings_stats[][ETH_GSTRING_LEN] = {
@@ -947,6 +453,8 @@ int ath9k_htc_get_et_sset_count(struct ieee80211_hw *hw,
 
 #define STXBASE priv->debug.tx_stats
 #define SRXBASE priv->debug.rx_stats
+#define SKBTXBASE priv->debug.tx_stats
+#define SKBRXBASE priv->debug.skbrx_stats
 #define ASTXQ(a)					\
 	data[i++] = STXBASE.a[IEEE80211_AC_BE];		\
 	data[i++] = STXBASE.a[IEEE80211_AC_BK];		\
@@ -960,24 +468,24 @@ void ath9k_htc_get_et_stats(struct ieee80211_hw *hw,
 	struct ath9k_htc_priv *priv = hw->priv;
 	int i = 0;
 
-	data[i++] = STXBASE.skb_success;
-	data[i++] = STXBASE.skb_success_bytes;
-	data[i++] = SRXBASE.skb_completed;
-	data[i++] = SRXBASE.skb_completed_bytes;
+	data[i++] = SKBTXBASE.skb_success;
+	data[i++] = SKBTXBASE.skb_success_bytes;
+	data[i++] = SKBRXBASE.skb_completed;
+	data[i++] = SKBRXBASE.skb_completed_bytes;
 
 	ASTXQ(queue_stats);
 
-	data[i++] = SRXBASE.err_crc;
-	data[i++] = SRXBASE.err_decrypt_crc;
-	data[i++] = SRXBASE.err_phy;
-	data[i++] = SRXBASE.err_mic;
-	data[i++] = SRXBASE.err_pre_delim;
-	data[i++] = SRXBASE.err_post_delim;
-	data[i++] = SRXBASE.err_decrypt_busy;
+	data[i++] = SRXBASE.crc_err;
+	data[i++] = SRXBASE.decrypt_crc_err;
+	data[i++] = SRXBASE.phy_err;
+	data[i++] = SRXBASE.mic_err;
+	data[i++] = SRXBASE.pre_delim_crc_err;
+	data[i++] = SRXBASE.post_delim_crc_err;
+	data[i++] = SRXBASE.decrypt_busy_err;
 
-	data[i++] = SRXBASE.err_phy_stats[ATH9K_PHYERR_RADAR];
-	data[i++] = SRXBASE.err_phy_stats[ATH9K_PHYERR_OFDM_TIMING];
-	data[i++] = SRXBASE.err_phy_stats[ATH9K_PHYERR_CCK_TIMING];
+	data[i++] = SRXBASE.phy_err_stats[ATH9K_PHYERR_RADAR];
+	data[i++] = SRXBASE.phy_err_stats[ATH9K_PHYERR_OFDM_TIMING];
+	data[i++] = SRXBASE.phy_err_stats[ATH9K_PHYERR_CCK_TIMING];
 
 	WARN_ON(i != ATH9K_HTC_SSTATS_LEN);
 }
@@ -1001,18 +509,21 @@ int ath9k_htc_init_debug(struct ath_hw *ah)
 			    priv, &fops_tgt_rx_stats);
 	debugfs_create_file("xmit", S_IRUSR, priv->debug.debugfs_phy,
 			    priv, &fops_xmit);
-	debugfs_create_file("recv", S_IRUSR, priv->debug.debugfs_phy,
-			    priv, &fops_recv);
+	debugfs_create_file("skb_rx", S_IRUSR, priv->debug.debugfs_phy,
+			    priv, &fops_skb_rx);
+
+	ath9k_cmn_debug_recv(priv->debug.debugfs_phy, &priv->debug.rx_stats);
+	ath9k_cmn_debug_phy_err(priv->debug.debugfs_phy, &priv->debug.rx_stats);
+
 	debugfs_create_file("slot", S_IRUSR, priv->debug.debugfs_phy,
 			    priv, &fops_slot);
 	debugfs_create_file("queue", S_IRUSR, priv->debug.debugfs_phy,
 			    priv, &fops_queue);
 	debugfs_create_file("debug", S_IRUSR | S_IWUSR, priv->debug.debugfs_phy,
 			    priv, &fops_debug);
-	debugfs_create_file("base_eeprom", S_IRUSR, priv->debug.debugfs_phy,
-			    priv, &fops_base_eeprom);
-	debugfs_create_file("modal_eeprom", S_IRUSR, priv->debug.debugfs_phy,
-			    priv, &fops_modal_eeprom);
+
+	ath9k_cmn_debug_base_eeprom(priv->debug.debugfs_phy, priv->ah);
+	ath9k_cmn_debug_modal_eeprom(priv->debug.debugfs_phy, priv->ah);
 
 	return 0;
 }
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c b/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
index 289f3d8924b5..bb86eb2ffc95 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
@@ -996,8 +996,6 @@ static bool ath9k_rx_prepare(struct ath9k_htc_priv *priv,
 		goto rx_next;
 	}
 
-	ath9k_htc_err_stat_rx(priv, rxstatus);
-
 	/* Get the RX status information */
 
 	memset(rx_status, 0, sizeof(struct ieee80211_rx_status));
@@ -1005,6 +1003,7 @@ static bool ath9k_rx_prepare(struct ath9k_htc_priv *priv,
 	/* Copy everything from ath_htc_rx_status (HTC_RX_FRAME_HEADER).
 	 * After this, we can drop this part of skb. */
 	rx_status_htc_to_ath(&rx_stats, rxstatus);
+	ath9k_htc_err_stat_rx(priv, &rx_stats);
 	rx_status->mactime = be64_to_cpu(rxstatus->rs_tstamp);
 	skb_pull(skb, HTC_RX_FRAME_HEADER_SIZE);
 
diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c
index c8a9dfab1fee..2a8ed8375ec0 100644
--- a/drivers/net/wireless/ath/ath9k/hw.c
+++ b/drivers/net/wireless/ath/ath9k/hw.c
@@ -26,7 +26,6 @@
 #include "ar9003_mac.h"
 #include "ar9003_mci.h"
 #include "ar9003_phy.h"
-#include "debug.h"
 #include "ath9k.h"
 
 static bool ath9k_hw_set_reset_reg(struct ath_hw *ah, u32 type);
@@ -246,6 +245,8 @@ static void ath9k_hw_read_revisions(struct ath_hw *ah)
 		return;
 	case AR9300_DEVID_AR953X:
 		ah->hw_version.macVersion = AR_SREV_VERSION_9531;
+		if (ah->get_mac_revision)
+			ah->hw_version.macRev = ah->get_mac_revision();
 		return;
 	}
 
diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c
index 36ae6490e554..0246b990fe87 100644
--- a/drivers/net/wireless/ath/ath9k/init.c
+++ b/drivers/net/wireless/ath/ath9k/init.c
@@ -61,6 +61,10 @@ static int ath9k_ps_enable;
 module_param_named(ps_enable, ath9k_ps_enable, int, 0444);
 MODULE_PARM_DESC(ps_enable, "Enable WLAN PowerSave");
 
+static int ath9k_use_chanctx;
+module_param_named(use_chanctx, ath9k_use_chanctx, int, 0444);
+MODULE_PARM_DESC(use_chanctx, "Enable channel context for concurrency");
+
 bool is_ath9k_unloaded;
 
 #ifdef CONFIG_MAC80211_LEDS
@@ -508,7 +512,7 @@ static int ath9k_init_softc(u16 devid, struct ath_softc *sc,
 	sc->tx99_power = MAX_RATE_POWER + 1;
 	init_waitqueue_head(&sc->tx_wait);
 
-	if (!pdata) {
+	if (!pdata || pdata->use_eeprom) {
 		ah->ah_flags |= AH_USE_EEPROM;
 		sc->sc_ah->led_pin = -1;
 	} else {
@@ -589,6 +593,9 @@ static int ath9k_init_softc(u16 devid, struct ath_softc *sc,
 	if (ret)
 		goto err_btcoex;
 
+	sc->p2p_ps_timer = ath_gen_timer_alloc(sc->sc_ah, ath9k_p2p_ps_timer,
+		NULL, sc, AR_FIRST_NDP_TIMER);
+
 	ath9k_cmn_init_crypto(sc->sc_ah);
 	ath9k_init_misc(sc);
 	ath_fill_led_pin(sc);
@@ -643,17 +650,20 @@ static void ath9k_init_txpower_limits(struct ath_softc *sc)
 }
 
 static const struct ieee80211_iface_limit if_limits[] = {
-	{ .max = 2048,	.types = BIT(NL80211_IFTYPE_STATION) |
-				 BIT(NL80211_IFTYPE_P2P_CLIENT) |
-				 BIT(NL80211_IFTYPE_WDS) },
+	{ .max = 2048,	.types = BIT(NL80211_IFTYPE_STATION) },
 	{ .max = 8,	.types =
 #ifdef CONFIG_MAC80211_MESH
 				 BIT(NL80211_IFTYPE_MESH_POINT) |
 #endif
-				 BIT(NL80211_IFTYPE_AP) |
+				 BIT(NL80211_IFTYPE_AP) },
+	{ .max = 1,	.types = BIT(NL80211_IFTYPE_P2P_CLIENT) |
 				 BIT(NL80211_IFTYPE_P2P_GO) },
 };
 
+static const struct ieee80211_iface_limit wds_limits[] = {
+	{ .max = 2048,	.types = BIT(NL80211_IFTYPE_WDS) },
+};
+
 static const struct ieee80211_iface_limit if_dfs_limits[] = {
 	{ .max = 1,	.types = BIT(NL80211_IFTYPE_AP) |
 #ifdef CONFIG_MAC80211_MESH
@@ -670,6 +680,13 @@ static const struct ieee80211_iface_combination if_comb[] = {
 		.num_different_channels = 1,
 		.beacon_int_infra_match = true,
 	},
+	{
+		.limits = wds_limits,
+		.n_limits = ARRAY_SIZE(wds_limits),
+		.max_interfaces = 2048,
+		.num_different_channels = 1,
+		.beacon_int_infra_match = true,
+	},
 #ifdef CONFIG_ATH9K_DFS_CERTIFIED
 	{
 		.limits = if_dfs_limits,
@@ -711,19 +728,23 @@ static void ath9k_set_hw_capab(struct ath_softc *sc, struct ieee80211_hw *hw)
 	if (AR_SREV_9160_10_OR_LATER(sc->sc_ah) || ath9k_modparam_nohwcrypt)
 		hw->flags |= IEEE80211_HW_MFP_CAPABLE;
 
-	hw->wiphy->features |= NL80211_FEATURE_ACTIVE_MONITOR;
+	hw->wiphy->features |= (NL80211_FEATURE_ACTIVE_MONITOR |
+				NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE);
 
 	if (!config_enabled(CONFIG_ATH9K_TX99)) {
 		hw->wiphy->interface_modes =
 			BIT(NL80211_IFTYPE_P2P_GO) |
 			BIT(NL80211_IFTYPE_P2P_CLIENT) |
 			BIT(NL80211_IFTYPE_AP) |
-			BIT(NL80211_IFTYPE_WDS) |
 			BIT(NL80211_IFTYPE_STATION) |
 			BIT(NL80211_IFTYPE_ADHOC) |
 			BIT(NL80211_IFTYPE_MESH_POINT);
 		hw->wiphy->iface_combinations = if_comb;
-		hw->wiphy->n_iface_combinations = ARRAY_SIZE(if_comb);
+		if (!ath9k_use_chanctx) {
+			hw->wiphy->n_iface_combinations = ARRAY_SIZE(if_comb);
+			hw->wiphy->interface_modes |= BIT(NL80211_IFTYPE_WDS);
+		} else
+			hw->wiphy->n_iface_combinations = 1;
 	}
 
 	hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
@@ -855,6 +876,9 @@ static void ath9k_deinit_softc(struct ath_softc *sc)
 {
 	int i = 0;
 
+	if (sc->p2p_ps_timer)
+		ath_gen_timer_free(sc->sc_ah, sc->p2p_ps_timer);
+
 	ath9k_deinit_btcoex(sc);
 
 	for (i = 0; i < ATH9K_NUM_TX_QUEUES; i++)
diff --git a/drivers/net/wireless/ath/ath9k/mac.c b/drivers/net/wireless/ath/ath9k/mac.c
index 51ce36f108f9..275205ab5f15 100644
--- a/drivers/net/wireless/ath/ath9k/mac.c
+++ b/drivers/net/wireless/ath/ath9k/mac.c
@@ -958,3 +958,25 @@ void ath9k_hw_set_interrupts(struct ath_hw *ah)
 	return;
 }
 EXPORT_SYMBOL(ath9k_hw_set_interrupts);
+
+#define ATH9K_HW_MAX_DCU       10
+#define ATH9K_HW_SLICE_PER_DCU 16
+#define ATH9K_HW_BIT_IN_SLICE  16
+void ath9k_hw_set_tx_filter(struct ath_hw *ah, u8 destidx, bool set)
+{
+	int dcu_idx;
+	u32 filter;
+
+	for (dcu_idx = 0; dcu_idx < 10; dcu_idx++) {
+		filter = SM(set, AR_D_TXBLK_WRITE_COMMAND);
+		filter |= SM(dcu_idx, AR_D_TXBLK_WRITE_DCU);
+		filter |= SM((destidx / ATH9K_HW_SLICE_PER_DCU),
+			     AR_D_TXBLK_WRITE_SLICE);
+		filter |= BIT(destidx % ATH9K_HW_BIT_IN_SLICE);
+		ath_dbg(ath9k_hw_common(ah), PS,
+			"DCU%d staid %d set %d txfilter %08x\n",
+			dcu_idx, destidx, set, filter);
+		REG_WRITE(ah, AR_D_TXBLK_BASE, filter);
+	}
+}
+EXPORT_SYMBOL(ath9k_hw_set_tx_filter);
diff --git a/drivers/net/wireless/ath/ath9k/mac.h b/drivers/net/wireless/ath/ath9k/mac.h
index 89df634e81f9..da7686757535 100644
--- a/drivers/net/wireless/ath/ath9k/mac.h
+++ b/drivers/net/wireless/ath/ath9k/mac.h
@@ -729,6 +729,7 @@ void ath9k_hw_startpcureceive(struct ath_hw *ah, bool is_scanning);
 void ath9k_hw_abortpcurecv(struct ath_hw *ah);
 bool ath9k_hw_stopdmarecv(struct ath_hw *ah, bool *reset);
 int ath9k_hw_beaconq_setup(struct ath_hw *ah);
+void ath9k_hw_set_tx_filter(struct ath_hw *ah, u8 destidx, bool set);
 
 /* Interrupt Handling */
 bool ath9k_hw_intrpend(struct ath_hw *ah);
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index d69853b848ce..62ac95d6bb9d 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -261,6 +261,8 @@ static bool ath_complete_reset(struct ath_softc *sc, bool start)
 	sc->gtt_cnt = 0;
 	ieee80211_wake_queues(sc->hw);
 
+	ath9k_p2p_ps_timer(sc);
+
 	return true;
 }
 
@@ -419,6 +421,7 @@ static void ath_node_attach(struct ath_softc *sc, struct ieee80211_sta *sta,
 	an->sc = sc;
 	an->sta = sta;
 	an->vif = vif;
+	memset(&an->key_idx, 0, sizeof(an->key_idx));
 
 	ath_tx_node_init(sc, an);
 }
@@ -1119,6 +1122,8 @@ static int ath9k_add_interface(struct ieee80211_hw *hw,
 	if (ath9k_uses_beacons(vif->type))
 		ath9k_beacon_assign_slot(sc, vif);
 
+	avp->vif = vif;
+
 	an->sc = sc;
 	an->sta = NULL;
 	an->vif = vif;
@@ -1163,6 +1168,29 @@ static int ath9k_change_interface(struct ieee80211_hw *hw,
 	return 0;
 }
 
+static void
+ath9k_update_p2p_ps_timer(struct ath_softc *sc, struct ath_vif *avp)
+{
+	struct ath_hw *ah = sc->sc_ah;
+	s32 tsf, target_tsf;
+
+	if (!avp || !avp->noa.has_next_tsf)
+		return;
+
+	ath9k_hw_gen_timer_stop(ah, sc->p2p_ps_timer);
+
+	tsf = ath9k_hw_gettsf32(sc->sc_ah);
+
+	target_tsf = avp->noa.next_tsf;
+	if (!avp->noa.absent)
+		target_tsf -= ATH_P2P_PS_STOP_TIME;
+
+	if (target_tsf - tsf < ATH_P2P_PS_STOP_TIME)
+		target_tsf = tsf + ATH_P2P_PS_STOP_TIME;
+
+	ath9k_hw_gen_timer_start(ah, sc->p2p_ps_timer, (u32) target_tsf, 1000000);
+}
+
 static void ath9k_remove_interface(struct ieee80211_hw *hw,
 				   struct ieee80211_vif *vif)
 {
@@ -1174,6 +1202,13 @@ static void ath9k_remove_interface(struct ieee80211_hw *hw,
 
 	mutex_lock(&sc->mutex);
 
+	spin_lock_bh(&sc->sc_pcu_lock);
+	if (avp == sc->p2p_ps_vif) {
+		sc->p2p_ps_vif = NULL;
+		ath9k_update_p2p_ps_timer(sc, NULL);
+	}
+	spin_unlock_bh(&sc->sc_pcu_lock);
+
 	sc->nvifs--;
 	sc->tx99_vif = NULL;
 
@@ -1427,8 +1462,10 @@ static int ath9k_sta_add(struct ieee80211_hw *hw,
 		return 0;
 
 	key = ath_key_config(common, vif, sta, &ps_key);
-	if (key > 0)
+	if (key > 0) {
 		an->ps_key = key;
+		an->key_idx[0] = key;
+	}
 
 	return 0;
 }
@@ -1446,6 +1483,7 @@ static void ath9k_del_ps_key(struct ath_softc *sc,
 
 	ath_key_delete(common, &ps_key);
 	an->ps_key = 0;
+	an->key_idx[0] = 0;
 }
 
 static int ath9k_sta_remove(struct ieee80211_hw *hw,
@@ -1460,6 +1498,19 @@ static int ath9k_sta_remove(struct ieee80211_hw *hw,
 	return 0;
 }
 
+static void ath9k_sta_set_tx_filter(struct ath_hw *ah,
+				    struct ath_node *an,
+				    bool set)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(an->key_idx); i++) {
+		if (!an->key_idx[i])
+			continue;
+		ath9k_hw_set_tx_filter(ah, an->key_idx[i], set);
+	}
+}
+
 static void ath9k_sta_notify(struct ieee80211_hw *hw,
 			 struct ieee80211_vif *vif,
 			 enum sta_notify_cmd cmd,
@@ -1472,8 +1523,10 @@ static void ath9k_sta_notify(struct ieee80211_hw *hw,
 	case STA_NOTIFY_SLEEP:
 		an->sleeping = true;
 		ath_tx_aggr_sleep(sta, sc, an);
+		ath9k_sta_set_tx_filter(sc->sc_ah, an, true);
 		break;
 	case STA_NOTIFY_AWAKE:
+		ath9k_sta_set_tx_filter(sc->sc_ah, an, false);
 		an->sleeping = false;
 		ath_tx_aggr_wakeup(sc, an);
 		break;
@@ -1529,7 +1582,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw,
 {
 	struct ath_softc *sc = hw->priv;
 	struct ath_common *common = ath9k_hw_common(sc->sc_ah);
-	int ret = 0;
+	struct ath_node *an = NULL;
+	int ret = 0, i;
 
 	if (ath9k_modparam_nohwcrypt)
 		return -ENOSPC;
@@ -1551,13 +1605,16 @@ static int ath9k_set_key(struct ieee80211_hw *hw,
 
 	mutex_lock(&sc->mutex);
 	ath9k_ps_wakeup(sc);
-	ath_dbg(common, CONFIG, "Set HW Key\n");
+	ath_dbg(common, CONFIG, "Set HW Key %d\n", cmd);
+	if (sta)
+		an = (struct ath_node *)sta->drv_priv;
 
 	switch (cmd) {
 	case SET_KEY:
 		if (sta)
 			ath9k_del_ps_key(sc, vif, sta);
 
+		key->hw_key_idx = 0;
 		ret = ath_key_config(common, vif, sta, key);
 		if (ret >= 0) {
 			key->hw_key_idx = ret;
@@ -1570,9 +1627,27 @@ static int ath9k_set_key(struct ieee80211_hw *hw,
 				key->flags |= IEEE80211_KEY_FLAG_SW_MGMT_TX;
 			ret = 0;
 		}
+		if (an && key->hw_key_idx) {
+			for (i = 0; i < ARRAY_SIZE(an->key_idx); i++) {
+				if (an->key_idx[i])
+					continue;
+				an->key_idx[i] = key->hw_key_idx;
+				break;
+			}
+			WARN_ON(i == ARRAY_SIZE(an->key_idx));
+		}
 		break;
 	case DISABLE_KEY:
 		ath_key_delete(common, key);
+		if (an) {
+			for (i = 0; i < ARRAY_SIZE(an->key_idx); i++) {
+				if (an->key_idx[i] != key->hw_key_idx)
+					continue;
+				an->key_idx[i] = 0;
+				break;
+			}
+		}
+		key->hw_key_idx = 0;
 		break;
 	default:
 		ret = -EINVAL;
@@ -1636,6 +1711,66 @@ static void ath9k_bss_assoc_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
 		ath9k_set_assoc_state(sc, vif);
 }
 
+void ath9k_p2p_ps_timer(void *priv)
+{
+	struct ath_softc *sc = priv;
+	struct ath_vif *avp = sc->p2p_ps_vif;
+	struct ieee80211_vif *vif;
+	struct ieee80211_sta *sta;
+	struct ath_node *an;
+	u32 tsf;
+
+	if (!avp)
+		return;
+
+	tsf = ath9k_hw_gettsf32(sc->sc_ah);
+	if (!avp->noa.absent)
+		tsf += ATH_P2P_PS_STOP_TIME;
+
+	if (!avp->noa.has_next_tsf ||
+	    avp->noa.next_tsf - tsf > BIT(31))
+		ieee80211_update_p2p_noa(&avp->noa, tsf);
+
+	ath9k_update_p2p_ps_timer(sc, avp);
+
+	rcu_read_lock();
+
+	vif = avp->vif;
+	sta = ieee80211_find_sta(vif, vif->bss_conf.bssid);
+	if (!sta)
+		goto out;
+
+	an = (void *) sta->drv_priv;
+	if (an->sleeping == !!avp->noa.absent)
+		goto out;
+
+	an->sleeping = avp->noa.absent;
+	if (an->sleeping)
+		ath_tx_aggr_sleep(sta, sc, an);
+	else
+		ath_tx_aggr_wakeup(sc, an);
+
+out:
+	rcu_read_unlock();
+}
+
+void ath9k_update_p2p_ps(struct ath_softc *sc, struct ieee80211_vif *vif)
+{
+	struct ath_vif *avp = (void *)vif->drv_priv;
+	u32 tsf;
+
+	if (!sc->p2p_ps_timer)
+		return;
+
+	if (vif->type != NL80211_IFTYPE_STATION || !vif->p2p)
+		return;
+
+	sc->p2p_ps_vif = avp;
+	tsf = ath9k_hw_gettsf32(sc->sc_ah);
+	ieee80211_parse_p2p_noa(&vif->bss_conf.p2p_noa_attr, &avp->noa, tsf);
+	ath9k_update_p2p_ps_timer(sc, avp);
+}
+
 static void ath9k_bss_info_changed(struct ieee80211_hw *hw,
 				   struct ieee80211_vif *vif,
 				   struct ieee80211_bss_conf *bss_conf,
@@ -1650,6 +1785,7 @@ static void ath9k_bss_info_changed(struct ieee80211_hw *hw,
 	struct ath_hw *ah = sc->sc_ah;
 	struct ath_common *common = ath9k_hw_common(ah);
 	struct ath_vif *avp = (void *)vif->drv_priv;
+	unsigned long flags;
 	int slottime;
 
 	ath9k_ps_wakeup(sc);
@@ -1710,6 +1846,15 @@ static void ath9k_bss_info_changed(struct ieee80211_hw *hw,
 		}
 	}
 
+	if (changed & BSS_CHANGED_P2P_PS) {
+		spin_lock_bh(&sc->sc_pcu_lock);
+		spin_lock_irqsave(&sc->sc_pm_lock, flags);
+		if (!(sc->ps_flags & PS_BEACON_SYNC))
+			ath9k_update_p2p_ps(sc, vif);
+		spin_unlock_irqrestore(&sc->sc_pm_lock, flags);
+		spin_unlock_bh(&sc->sc_pcu_lock);
+	}
+
 	if (changed & CHECK_ANI)
 		ath_check_ani(sc);
 
@@ -1883,7 +2028,8 @@ static bool ath9k_has_tx_pending(struct ath_softc *sc)
 	return !!npend;
 }
 
-static void ath9k_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void ath9k_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			u32 queues, bool drop)
 {
 	struct ath_softc *sc = hw->priv;
 	struct ath_hw *ah = sc->sc_ah;
@@ -2084,14 +2230,6 @@ static void ath9k_sw_scan_complete(struct ieee80211_hw *hw)
 	clear_bit(ATH_OP_SCANNING, &common->op_flags);
 }
 
-static void ath9k_channel_switch_beacon(struct ieee80211_hw *hw,
-					struct ieee80211_vif *vif,
-					struct cfg80211_chan_def *chandef)
-{
-	/* depend on vif->csa_active only */
-	return;
-}
-
 struct ieee80211_ops ath9k_ops = {
 	.tx 		    = ath9k_tx,
 	.start 		    = ath9k_start,
@@ -2139,5 +2277,4 @@ struct ieee80211_ops ath9k_ops = {
 #endif
 	.sw_scan_start	    = ath9k_sw_scan_start,
 	.sw_scan_complete   = ath9k_sw_scan_complete,
-	.channel_switch_beacon     = ath9k_channel_switch_beacon,
 };
diff --git a/drivers/net/wireless/ath/ath9k/pci.c b/drivers/net/wireless/ath/ath9k/pci.c
index 914dbc6b1720..4dec09e565ed 100644
--- a/drivers/net/wireless/ath/ath9k/pci.c
+++ b/drivers/net/wireless/ath/ath9k/pci.c
@@ -686,7 +686,7 @@ static bool ath_pci_eeprom_read(struct ath_common *common, u32 off, u16 *data)
 	struct ath_softc *sc = (struct ath_softc *) common->priv;
 	struct ath9k_platform_data *pdata = sc->dev->platform_data;
 
-	if (pdata) {
+	if (pdata && !pdata->use_eeprom) {
 		if (off >= (ARRAY_SIZE(pdata->eeprom_data))) {
 			ath_err(common,
 				"%s: eeprom read failed, offset %08x is out of range\n",
@@ -914,6 +914,7 @@ static int ath_pci_suspend(struct device *device)
 	 */
 	ath9k_stop_btcoex(sc);
 	ath9k_hw_disable(sc->sc_ah);
+	del_timer_sync(&sc->sleep_timer);
 	ath9k_hw_setpower(sc->sc_ah, ATH9K_PM_FULL_SLEEP);
 
 	return 0;
diff --git a/drivers/net/wireless/ath/ath9k/recv.c b/drivers/net/wireless/ath/ath9k/recv.c
index 19df969ec909..9105a92364f7 100644
--- a/drivers/net/wireless/ath/ath9k/recv.c
+++ b/drivers/net/wireless/ath/ath9k/recv.c
@@ -34,7 +34,8 @@ static inline bool ath9k_check_auto_sleep(struct ath_softc *sc)
  * buffer (or rx fifo). This can incorrectly acknowledge packets
  * to a sender if last desc is self-linked.
  */
-static void ath_rx_buf_link(struct ath_softc *sc, struct ath_rxbuf *bf)
+static void ath_rx_buf_link(struct ath_softc *sc, struct ath_rxbuf *bf,
+			    bool flush)
 {
 	struct ath_hw *ah = sc->sc_ah;
 	struct ath_common *common = ath9k_hw_common(ah);
@@ -59,18 +60,19 @@ static void ath_rx_buf_link(struct ath_softc *sc, struct ath_rxbuf *bf)
 			     common->rx_bufsize,
 			     0);
 
-	if (sc->rx.rxlink == NULL)
-		ath9k_hw_putrxbuf(ah, bf->bf_daddr);
-	else
+	if (sc->rx.rxlink)
 		*sc->rx.rxlink = bf->bf_daddr;
+	else if (!flush)
+		ath9k_hw_putrxbuf(ah, bf->bf_daddr);
 
 	sc->rx.rxlink = &ds->ds_link;
 }
 
-static void ath_rx_buf_relink(struct ath_softc *sc, struct ath_rxbuf *bf)
+static void ath_rx_buf_relink(struct ath_softc *sc, struct ath_rxbuf *bf,
+			      bool flush)
 {
 	if (sc->rx.buf_hold)
-		ath_rx_buf_link(sc, sc->rx.buf_hold);
+		ath_rx_buf_link(sc, sc->rx.buf_hold, flush);
 
 	sc->rx.buf_hold = bf;
 }
@@ -442,7 +444,7 @@ int ath_startrecv(struct ath_softc *sc)
 	sc->rx.buf_hold = NULL;
 	sc->rx.rxlink = NULL;
 	list_for_each_entry_safe(bf, tbf, &sc->rx.rxbuf, list) {
-		ath_rx_buf_link(sc, bf);
+		ath_rx_buf_link(sc, bf, false);
 	}
 
 	/* We could have deleted elements so the list may be empty now */
@@ -538,7 +540,10 @@ static void ath_rx_ps_beacon(struct ath_softc *sc, struct sk_buff *skb)
 		sc->ps_flags &= ~PS_BEACON_SYNC;
 		ath_dbg(common, PS,
 			"Reconfigure beacon timers based on synchronized timestamp\n");
-		ath9k_set_beacon(sc);
+		if (!(WARN_ON_ONCE(sc->cur_beacon_conf.beacon_interval == 0)))
+			ath9k_set_beacon(sc);
+		if (sc->p2p_ps_vif)
+			ath9k_update_p2p_ps(sc, sc->p2p_ps_vif->vif);
 	}
 
 	if (ath_beacon_dtim_pending_cab(skb)) {
@@ -1115,12 +1120,12 @@ requeue_drop_frag:
 requeue:
 		list_add_tail(&bf->list, &sc->rx.rxbuf);
 
-		if (edma) {
-			ath_rx_edma_buf_link(sc, qtype);
-		} else {
-			ath_rx_buf_relink(sc, bf);
+		if (!edma) {
+			ath_rx_buf_relink(sc, bf, flush);
 			if (!flush)
 				ath9k_hw_rxena(ah);
+		} else if (!flush) {
+			ath_rx_edma_buf_link(sc, qtype);
 		}
 
 		if (!budget--)
diff --git a/drivers/net/wireless/ath/ath9k/reg.h b/drivers/net/wireless/ath/ath9k/reg.h
index b1fd3fa84983..f1bbce3f7774 100644
--- a/drivers/net/wireless/ath/ath9k/reg.h
+++ b/drivers/net/wireless/ath/ath9k/reg.h
@@ -505,9 +505,6 @@
 #define AR_D_QCUMASK         0x000003FF
 #define AR_D_QCUMASK_RESV0   0xFFFFFC00
 
-#define AR_D_TXBLK_CMD  0x1038
-#define AR_D_TXBLK_DATA(i) (AR_D_TXBLK_CMD+(i))
-
 #define AR_D0_LCL_IFS     0x1040
 #define AR_D1_LCL_IFS     0x1044
 #define AR_D2_LCL_IFS     0x1048
diff --git a/drivers/net/wireless/ath/carl9170/main.c b/drivers/net/wireless/ath/carl9170/main.c
index 4c8cdb097b65..f8ded84b7be8 100644
--- a/drivers/net/wireless/ath/carl9170/main.c
+++ b/drivers/net/wireless/ath/carl9170/main.c
@@ -1707,7 +1707,9 @@ found:
 	return 0;
 }
 
-static void carl9170_op_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void carl9170_op_flush(struct ieee80211_hw *hw,
+			      struct ieee80211_vif *vif,
+			      u32 queues, bool drop)
 {
 	struct ar9170 *ar = hw->priv;
 	unsigned int vid;
diff --git a/drivers/net/wireless/ath/carl9170/usb.c b/drivers/net/wireless/ath/carl9170/usb.c
index ca115f33746f..f35c7f30f9a6 100644
--- a/drivers/net/wireless/ath/carl9170/usb.c
+++ b/drivers/net/wireless/ath/carl9170/usb.c
@@ -1076,8 +1076,14 @@ static int carl9170_usb_probe(struct usb_interface *intf,
 
 	carl9170_set_state(ar, CARL9170_STOPPED);
 
-	return request_firmware_nowait(THIS_MODULE, 1, CARL9170FW_NAME,
+	err = request_firmware_nowait(THIS_MODULE, 1, CARL9170FW_NAME,
 		&ar->udev->dev, GFP_KERNEL, ar, carl9170_usb_firmware_step2);
+	if (err) {
+		usb_put_dev(udev);
+		usb_put_dev(udev);
+		carl9170_free(ar);
+	}
+	return err;
 }
 
 static void carl9170_usb_disconnect(struct usb_interface *intf)
diff --git a/drivers/net/wireless/ath/dfs_pattern_detector.c b/drivers/net/wireless/ath/dfs_pattern_detector.c
index a1a69c5db409..650be79c7ac9 100644
--- a/drivers/net/wireless/ath/dfs_pattern_detector.c
+++ b/drivers/net/wireless/ath/dfs_pattern_detector.c
@@ -73,9 +73,52 @@ static const struct radar_types etsi_radar_types_v15 = {
 	.radar_types		= etsi_radar_ref_types_v15,
 };
 
-/* for now, we support ETSI radar types, FCC and JP are TODO */
+#define FCC_PATTERN(ID, WMIN, WMAX, PMIN, PMAX, PRF, PPB)	\
+{								\
+	ID, WIDTH_LOWER(WMIN), WIDTH_UPPER(WMAX),		\
+	PMIN - PRI_TOLERANCE,					\
+	PMAX * PRF + PRI_TOLERANCE, PRF, PPB * PRF,		\
+	PPB_THRESH(PPB), PRI_TOLERANCE,				\
+}
+
+static const struct radar_detector_specs fcc_radar_ref_types[] = {
+	FCC_PATTERN(0, 0, 1, 1428, 1428, 1, 18),
+	FCC_PATTERN(1, 0, 5, 150, 230, 1, 23),
+	FCC_PATTERN(2, 6, 10, 200, 500, 1, 16),
+	FCC_PATTERN(3, 11, 20, 200, 500, 1, 12),
+	FCC_PATTERN(4, 50, 100, 1000, 2000, 20, 1),
+	FCC_PATTERN(5, 0, 1, 333, 333, 1, 9),
+};
+
+static const struct radar_types fcc_radar_types = {
+	.region			= NL80211_DFS_FCC,
+	.num_radar_types	= ARRAY_SIZE(fcc_radar_ref_types),
+	.radar_types		= fcc_radar_ref_types,
+};
+
+#define JP_PATTERN FCC_PATTERN
+static const struct radar_detector_specs jp_radar_ref_types[] = {
+	JP_PATTERN(0, 0, 1, 1428, 1428, 1, 18),
+	JP_PATTERN(1, 2, 3, 3846, 3846, 1, 18),
+	JP_PATTERN(2, 0, 1, 1388, 1388, 1, 18),
+	JP_PATTERN(3, 1, 2, 4000, 4000, 1, 18),
+	JP_PATTERN(4, 0, 5, 150, 230, 1, 23),
+	JP_PATTERN(5, 6, 10, 200, 500, 1, 16),
+	JP_PATTERN(6, 11, 20, 200, 500, 1, 12),
+	JP_PATTERN(7, 50, 100, 1000, 2000, 20, 1),
+	JP_PATTERN(5, 0, 1, 333, 333, 1, 9),
+};
+
+static const struct radar_types jp_radar_types = {
+	.region			= NL80211_DFS_JP,
+	.num_radar_types	= ARRAY_SIZE(jp_radar_ref_types),
+	.radar_types		= jp_radar_ref_types,
+};
+
 static const struct radar_types *dfs_domains[] = {
 	&etsi_radar_types_v15,
+	&fcc_radar_types,
+	&jp_radar_types,
 };
 
 /**
diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c
index 7bf0ef8a1f56..63986931829e 100644
--- a/drivers/net/wireless/ath/wcn36xx/smd.c
+++ b/drivers/net/wireless/ath/wcn36xx/smd.c
@@ -2068,7 +2068,7 @@ static void wcn36xx_smd_rsp_process(struct wcn36xx *wcn, void *buf, size_t len)
 		if (!msg_ind)
 			goto nomem;
 		msg_ind->msg_len = len;
-		msg_ind->msg = kmalloc(len, GFP_KERNEL);
+		msg_ind->msg = kmemdup(buf, len, GFP_KERNEL);
 		if (!msg_ind->msg) {
 			kfree(msg_ind);
 nomem:
@@ -2080,7 +2080,6 @@ nomem:
 				    msg_header->msg_type);
 			break;
 		}
-		memcpy(msg_ind->msg, buf, len);
 		mutex_lock(&wcn->hal_ind_mutex);
 		list_add_tail(&msg_ind->list, &wcn->hal_ind_queue);
 		queue_work(wcn->hal_ind_wq, &wcn->hal_ind_work);
diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 4806a49cb61b..820d4ebd9322 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -172,7 +172,7 @@ static int wil_cid_fill_sinfo(struct wil6210_priv *wil, int cid,
 
 static int wil_cfg80211_get_station(struct wiphy *wiphy,
 				    struct net_device *ndev,
-				    u8 *mac, struct station_info *sinfo)
+				    const u8 *mac, struct station_info *sinfo)
 {
 	struct wil6210_priv *wil = wiphy_to_wil(wiphy);
 	int rc;
@@ -288,6 +288,7 @@ static int wil_cfg80211_scan(struct wiphy *wiphy,
 	}
 
 	wil->scan_request = request;
+	mod_timer(&wil->scan_timer, jiffies + WIL6210_SCAN_TO);
 
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.cmd.num_channels = 0;
@@ -671,7 +672,7 @@ static int wil_cfg80211_stop_ap(struct wiphy *wiphy,
 }
 
 static int wil_cfg80211_del_station(struct wiphy *wiphy,
-				    struct net_device *dev, u8 *mac)
+				    struct net_device *dev, const u8 *mac)
 {
 	struct wil6210_priv *wil = wiphy_to_wil(wiphy);
 
diff --git a/drivers/net/wireless/ath/wil6210/debugfs.c b/drivers/net/wireless/ath/wil6210/debugfs.c
index ecdabe4adec3..8d4bc4bfb664 100644
--- a/drivers/net/wireless/ath/wil6210/debugfs.c
+++ b/drivers/net/wireless/ath/wil6210/debugfs.c
@@ -35,7 +35,7 @@ static void wil_print_vring(struct seq_file *s, struct wil6210_priv *wil,
 	void __iomem *x = wmi_addr(wil, vring->hwtail);
 
 	seq_printf(s, "VRING %s = {\n", name);
-	seq_printf(s, "  pa     = 0x%016llx\n", (unsigned long long)vring->pa);
+	seq_printf(s, "  pa     = %pad\n", &vring->pa);
 	seq_printf(s, "  va     = 0x%p\n", vring->va);
 	seq_printf(s, "  size   = %d\n", vring->size);
 	seq_printf(s, "  swtail = %d\n", vring->swtail);
@@ -473,7 +473,7 @@ static int wil_txdesc_debugfs_show(struct seq_file *s, void *data)
 			   u[0], u[1], u[2], u[3]);
 		seq_printf(s, "  DMA = 0x%08x 0x%08x 0x%08x 0x%08x\n",
 			   u[4], u[5], u[6], u[7]);
-		seq_printf(s, "  SKB = %p\n", skb);
+		seq_printf(s, "  SKB = 0x%p\n", skb);
 
 		if (skb) {
 			skb_get(skb);
diff --git a/drivers/net/wireless/ath/wil6210/interrupt.c b/drivers/net/wireless/ath/wil6210/interrupt.c
index 5824cd41e4ba..73593aa3cd98 100644
--- a/drivers/net/wireless/ath/wil6210/interrupt.c
+++ b/drivers/net/wireless/ath/wil6210/interrupt.c
@@ -338,7 +338,7 @@ static irqreturn_t wil6210_irq_misc_thread(int irq, void *cookie)
 	}
 
 	if (isr)
-		wil_err(wil, "un-handled MISC ISR bits 0x%08x\n", isr);
+		wil_dbg_irq(wil, "un-handled MISC ISR bits 0x%08x\n", isr);
 
 	wil->isr_misc = 0;
 
diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 95f4efe9ef37..11e6d9d22eae 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -81,7 +81,7 @@ static void wil_disconnect_cid(struct wil6210_priv *wil, int cid)
 	memset(&sta->stats, 0, sizeof(sta->stats));
 }
 
-static void _wil6210_disconnect(struct wil6210_priv *wil, void *bssid)
+static void _wil6210_disconnect(struct wil6210_priv *wil, const u8 *bssid)
 {
 	int cid = -ENOENT;
 	struct net_device *ndev = wil_to_ndev(wil);
@@ -150,6 +150,15 @@ static void wil_connect_timer_fn(ulong x)
 	schedule_work(&wil->disconnect_worker);
 }
 
+static void wil_scan_timer_fn(ulong x)
+{
+	struct wil6210_priv *wil = (void *)x;
+
+	clear_bit(wil_status_fwready, &wil->status);
+	wil_err(wil, "Scan timeout detected, start fw error recovery\n");
+	schedule_work(&wil->fw_error_worker);
+}
+
 static void wil_fw_error_worker(struct work_struct *work)
 {
 	struct wil6210_priv *wil = container_of(work,
@@ -161,12 +170,30 @@ static void wil_fw_error_worker(struct work_struct *work)
 	if (no_fw_recovery)
 		return;
 
+	/* increment @recovery_count if less then WIL6210_FW_RECOVERY_TO
+	 * passed since last recovery attempt
+	 */
+	if (time_is_after_jiffies(wil->last_fw_recovery +
+				  WIL6210_FW_RECOVERY_TO))
+		wil->recovery_count++;
+	else
+		wil->recovery_count = 1; /* fw was alive for a long time */
+
+	if (wil->recovery_count > WIL6210_FW_RECOVERY_RETRIES) {
+		wil_err(wil, "too many recovery attempts (%d), giving up\n",
+			wil->recovery_count);
+		return;
+	}
+
+	wil->last_fw_recovery = jiffies;
+
 	mutex_lock(&wil->mutex);
 	switch (wdev->iftype) {
 	case NL80211_IFTYPE_STATION:
 	case NL80211_IFTYPE_P2P_CLIENT:
 	case NL80211_IFTYPE_MONITOR:
-		wil_info(wil, "fw error recovery started...\n");
+		wil_info(wil, "fw error recovery started (try %d)...\n",
+			 wil->recovery_count);
 		wil_reset(wil);
 
 		/* need to re-allocate Rx ring after reset */
@@ -230,6 +257,7 @@ int wil_priv_init(struct wil6210_priv *wil)
 
 	wil->pending_connect_cid = -1;
 	setup_timer(&wil->connect_timer, wil_connect_timer_fn, (ulong)wil);
+	setup_timer(&wil->scan_timer, wil_scan_timer_fn, (ulong)wil);
 
 	INIT_WORK(&wil->connect_worker, wil_connect_worker);
 	INIT_WORK(&wil->disconnect_worker, wil_disconnect_worker);
@@ -249,10 +277,12 @@ int wil_priv_init(struct wil6210_priv *wil)
 		return -EAGAIN;
 	}
 
+	wil->last_fw_recovery = jiffies;
+
 	return 0;
 }
 
-void wil6210_disconnect(struct wil6210_priv *wil, void *bssid)
+void wil6210_disconnect(struct wil6210_priv *wil, const u8 *bssid)
 {
 	del_timer_sync(&wil->connect_timer);
 	_wil6210_disconnect(wil, bssid);
@@ -260,6 +290,7 @@ void wil6210_disconnect(struct wil6210_priv *wil, void *bssid)
 
 void wil_priv_deinit(struct wil6210_priv *wil)
 {
+	del_timer_sync(&wil->scan_timer);
 	cancel_work_sync(&wil->disconnect_worker);
 	cancel_work_sync(&wil->fw_error_worker);
 	mutex_lock(&wil->mutex);
@@ -363,8 +394,8 @@ static int wil_wait_for_fw_ready(struct wil6210_priv *wil)
 		wil_err(wil, "Firmware not ready\n");
 		return -ETIME;
 	} else {
-		wil_dbg_misc(wil, "FW ready after %d ms\n",
-			     jiffies_to_msecs(to-left));
+		wil_info(wil, "FW ready after %d ms. HW version 0x%08x\n",
+			 jiffies_to_msecs(to-left), wil->hw_version);
 	}
 	return 0;
 }
@@ -391,6 +422,7 @@ int wil_reset(struct wil6210_priv *wil)
 	if (wil->scan_request) {
 		wil_dbg_misc(wil, "Abort scan_request 0x%p\n",
 			     wil->scan_request);
+		del_timer_sync(&wil->scan_timer);
 		cfg80211_scan_done(wil->scan_request, true);
 		wil->scan_request = NULL;
 	}
@@ -520,6 +552,7 @@ static int __wil_down(struct wil6210_priv *wil)
 	napi_disable(&wil->napi_tx);
 
 	if (wil->scan_request) {
+		del_timer_sync(&wil->scan_timer);
 		cfg80211_scan_done(wil->scan_request, true);
 		wil->scan_request = NULL;
 	}
diff --git a/drivers/net/wireless/ath/wil6210/netdev.c b/drivers/net/wireless/ath/wil6210/netdev.c
index fdcaeb820e75..106b6dcb773a 100644
--- a/drivers/net/wireless/ath/wil6210/netdev.c
+++ b/drivers/net/wireless/ath/wil6210/netdev.c
@@ -32,12 +32,26 @@ static int wil_stop(struct net_device *ndev)
 	return wil_down(wil);
 }
 
+static int wil_change_mtu(struct net_device *ndev, int new_mtu)
+{
+	struct wil6210_priv *wil = ndev_to_wil(ndev);
+
+	if (new_mtu < 68 || new_mtu > IEEE80211_MAX_DATA_LEN_DMG)
+		return -EINVAL;
+
+	wil_dbg_misc(wil, "change MTU %d -> %d\n", ndev->mtu, new_mtu);
+	ndev->mtu = new_mtu;
+
+	return 0;
+}
+
 static const struct net_device_ops wil_netdev_ops = {
 	.ndo_open		= wil_open,
 	.ndo_stop		= wil_stop,
 	.ndo_start_xmit		= wil_start_xmit,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_change_mtu		= wil_change_mtu,
 };
 
 static int wil6210_netdev_poll_rx(struct napi_struct *napi, int budget)
diff --git a/drivers/net/wireless/ath/wil6210/pcie_bus.c b/drivers/net/wireless/ath/wil6210/pcie_bus.c
index f1e1bb338d68..1e2e07b9d13d 100644
--- a/drivers/net/wireless/ath/wil6210/pcie_bus.c
+++ b/drivers/net/wireless/ath/wil6210/pcie_bus.c
@@ -74,8 +74,6 @@ static int wil_if_pcie_enable(struct wil6210_priv *wil)
 	if (rc)
 		goto release_irq;
 
-	wil_info(wil, "HW version: 0x%08x\n", wil->hw_version);
-
 	return 0;
 
  release_irq:
@@ -140,7 +138,7 @@ static int wil_pcie_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err_release_reg;
 	}
 	/* rollback to err_iounmap */
-	dev_info(&pdev->dev, "CSR at %pR -> %p\n", &pdev->resource[0], csr);
+	dev_info(&pdev->dev, "CSR at %pR -> 0x%p\n", &pdev->resource[0], csr);
 
 	wil = wil_if_alloc(dev, csr);
 	if (IS_ERR(wil)) {
diff --git a/drivers/net/wireless/ath/wil6210/rx_reorder.c b/drivers/net/wireless/ath/wil6210/rx_reorder.c
index d04629fe053f..747ae1275877 100644
--- a/drivers/net/wireless/ath/wil6210/rx_reorder.c
+++ b/drivers/net/wireless/ath/wil6210/rx_reorder.c
@@ -49,10 +49,17 @@ static void wil_release_reorder_frames(struct wil6210_priv *wil,
 {
 	int index;
 
-	while (seq_less(r->head_seq_num, hseq)) {
+	/* note: this function is never called with
+	 * hseq preceding r->head_seq_num, i.e it is always true
+	 * !seq_less(hseq, r->head_seq_num)
+	 * and thus on loop exit it should be
+	 * r->head_seq_num == hseq
+	 */
+	while (seq_less(r->head_seq_num, hseq) && r->stored_mpdu_num) {
 		index = reorder_index(r, r->head_seq_num);
 		wil_release_reorder_frame(wil, r, index);
 	}
+	r->head_seq_num = hseq;
 }
 
 static void wil_reorder_release(struct wil6210_priv *wil,
@@ -91,6 +98,22 @@ void wil_rx_reorder(struct wil6210_priv *wil, struct sk_buff *skb)
 
 	spin_lock(&r->reorder_lock);
 
+	/** Due to the race between WMI events, where BACK establishment
+	 * reported, and data Rx, few packets may be pass up before reorder
+	 * buffer get allocated. Catch up by pretending SSN is what we
+	 * see in the 1-st Rx packet
+	 */
+	if (r->first_time) {
+		r->first_time = false;
+		if (seq != r->head_seq_num) {
+			wil_err(wil, "Error: 1-st frame with wrong sequence"
+				" %d, should be %d. Fixing...\n", seq,
+				r->head_seq_num);
+			r->head_seq_num = seq;
+			r->ssn = seq;
+		}
+	}
+
 	/* frame with out of date sequence number */
 	if (seq_less(seq, r->head_seq_num)) {
 		dev_kfree_skb(skb);
@@ -162,6 +185,7 @@ struct wil_tid_ampdu_rx *wil_tid_ampdu_rx_alloc(struct wil6210_priv *wil,
 	r->head_seq_num = ssn;
 	r->buf_size = size;
 	r->stored_mpdu_num = 0;
+	r->first_time = true;
 	return r;
 }
 
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index c8c547457eb4..0784ef3d4ce2 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -64,6 +64,22 @@ static inline int wil_vring_avail_tx(struct vring *vring)
 	return vring->size - used - 1;
 }
 
+/**
+ * wil_vring_wmark_low - low watermark for available descriptor space
+ */
+static inline int wil_vring_wmark_low(struct vring *vring)
+{
+	return vring->size/8;
+}
+
+/**
+ * wil_vring_wmark_high - high watermark for available descriptor space
+ */
+static inline int wil_vring_wmark_high(struct vring *vring)
+{
+	return vring->size/4;
+}
+
 static int wil_vring_alloc(struct wil6210_priv *wil, struct vring *vring)
 {
 	struct device *dev = wil_to_dev(wil);
@@ -98,8 +114,8 @@ static int wil_vring_alloc(struct wil6210_priv *wil, struct vring *vring)
 		_d->dma.status = TX_DMA_STATUS_DU;
 	}
 
-	wil_dbg_misc(wil, "vring[%d] 0x%p:0x%016llx 0x%p\n", vring->size,
-		     vring->va, (unsigned long long)vring->pa, vring->ctx);
+	wil_dbg_misc(wil, "vring[%d] 0x%p:%pad 0x%p\n", vring->size,
+		     vring->va, &vring->pa, vring->ctx);
 
 	return 0;
 }
@@ -880,8 +896,8 @@ static int wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
 	pa = dma_map_single(dev, skb->data,
 			skb_headlen(skb), DMA_TO_DEVICE);
 
-	wil_dbg_txrx(wil, "Tx skb %d bytes %p -> %#08llx\n", skb_headlen(skb),
-		     skb->data, (unsigned long long)pa);
+	wil_dbg_txrx(wil, "Tx skb %d bytes 0x%p -> %pad\n", skb_headlen(skb),
+		     skb->data, &pa);
 	wil_hex_dump_txrx("Tx ", DUMP_PREFIX_OFFSET, 16, 1,
 			  skb->data, skb_headlen(skb), false);
 
@@ -1007,7 +1023,7 @@ netdev_tx_t wil_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	rc = wil_tx_vring(wil, vring, skb);
 
 	/* do we still have enough room in the vring? */
-	if (wil_vring_avail_tx(vring) < vring->size/8)
+	if (wil_vring_avail_tx(vring) < wil_vring_wmark_low(vring))
 		netif_tx_stop_all_queues(wil_to_ndev(wil));
 
 	switch (rc) {
@@ -1116,7 +1132,7 @@ int wil_tx_complete(struct wil6210_priv *wil, int ringid)
 			done++;
 		}
 	}
-	if (wil_vring_avail_tx(vring) > vring->size/4)
+	if (wil_vring_avail_tx(vring) > wil_vring_wmark_high(vring))
 		netif_tx_wake_all_queues(wil_to_ndev(wil));
 
 	return done;
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 2a2dec75f026..e25edc52398f 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -35,11 +35,14 @@ static inline u32 WIL_GET_BITS(u32 x, int b0, int b1)
 #define WIL6210_MEM_SIZE (2*1024*1024UL)
 
 #define WIL6210_RX_RING_SIZE	(128)
-#define WIL6210_TX_RING_SIZE	(128)
+#define WIL6210_TX_RING_SIZE	(512)
 #define WIL6210_MAX_TX_RINGS	(24) /* HW limit */
 #define WIL6210_MAX_CID		(8) /* HW limit */
 #define WIL6210_NAPI_BUDGET	(16) /* arbitrary */
 #define WIL6210_ITR_TRSH	(10000) /* arbitrary - about 15 IRQs/msec */
+#define WIL6210_FW_RECOVERY_RETRIES	(5) /* try to recover this many times */
+#define WIL6210_FW_RECOVERY_TO	msecs_to_jiffies(5000)
+#define WIL6210_SCAN_TO		msecs_to_jiffies(10000)
 
 /* Hardware definitions begin */
 
@@ -301,6 +304,7 @@ struct wil_tid_ampdu_rx {
 	u16 buf_size;
 	u16 timeout;
 	u8 dialog_token;
+	bool first_time; /* is it 1-st time this buffer used? */
 };
 
 struct wil6210_stats {
@@ -360,6 +364,8 @@ struct wil6210_priv {
 	u32 fw_version;
 	u32 hw_version;
 	u8 n_mids; /* number of additional MIDs as reported by FW */
+	int recovery_count; /* num of FW recovery attempts in a short time */
+	unsigned long last_fw_recovery; /* jiffies of last fw recovery */
 	/* profile */
 	u32 monitor_flags;
 	u32 secure_pcp; /* create secure PCP? */
@@ -381,6 +387,7 @@ struct wil6210_priv {
 	struct work_struct disconnect_worker;
 	struct work_struct fw_error_worker;	/* for FW error recovery */
 	struct timer_list connect_timer;
+	struct timer_list scan_timer; /* detect scan timeout */
 	int pending_connect_cid;
 	struct list_head pending_wmi_ev;
 	/*
@@ -507,7 +514,7 @@ void wil_wdev_free(struct wil6210_priv *wil);
 int wmi_set_mac_address(struct wil6210_priv *wil, void *addr);
 int wmi_pcp_start(struct wil6210_priv *wil, int bi, u8 wmi_nettype, u8 chan);
 int wmi_pcp_stop(struct wil6210_priv *wil);
-void wil6210_disconnect(struct wil6210_priv *wil, void *bssid);
+void wil6210_disconnect(struct wil6210_priv *wil, const u8 *bssid);
 
 int wil_rx_init(struct wil6210_priv *wil);
 void wil_rx_fini(struct wil6210_priv *wil);
diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c
index 2ba56eef0c45..6cc0e182cc70 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.c
+++ b/drivers/net/wireless/ath/wil6210/wmi.c
@@ -192,7 +192,7 @@ static int __wmi_send(struct wil6210_priv *wil, u16 cmdid, void *buf, u16 len)
 	might_sleep();
 
 	if (!test_bit(wil_status_fwready, &wil->status)) {
-		wil_err(wil, "FW not ready\n");
+		wil_err(wil, "WMI: cannot send command while FW not ready\n");
 		return -EAGAIN;
 	}
 
@@ -276,8 +276,8 @@ static void wmi_evt_ready(struct wil6210_priv *wil, int id, void *d, int len)
 	wil->fw_version = le32_to_cpu(evt->sw_version);
 	wil->n_mids = evt->numof_additional_mids;
 
-	wil_dbg_wmi(wil, "FW ver. %d; MAC %pM; %d MID's\n", wil->fw_version,
-		    evt->mac, wil->n_mids);
+	wil_info(wil, "FW ver. %d; MAC %pM; %d MID's\n", wil->fw_version,
+		 evt->mac, wil->n_mids);
 
 	if (!is_valid_ether_addr(ndev->dev_addr)) {
 		memcpy(ndev->dev_addr, evt->mac, ETH_ALEN);
@@ -290,7 +290,7 @@ static void wmi_evt_ready(struct wil6210_priv *wil, int id, void *d, int len)
 static void wmi_evt_fw_ready(struct wil6210_priv *wil, int id, void *d,
 			     int len)
 {
-	wil_dbg_wmi(wil, "WMI: FW ready\n");
+	wil_dbg_wmi(wil, "WMI: got FW ready event\n");
 
 	set_bit(wil_status_fwready, &wil->status);
 	/* reuse wmi_ready for the firmware ready indication */
@@ -348,9 +348,10 @@ static void wmi_evt_scan_complete(struct wil6210_priv *wil, int id,
 {
 	if (wil->scan_request) {
 		struct wmi_scan_complete_event *data = d;
-		bool aborted = (data->status != 0);
+		bool aborted = (data->status != WMI_SCAN_SUCCESS);
 
 		wil_dbg_wmi(wil, "SCAN_COMPLETE(0x%08x)\n", data->status);
+		del_timer_sync(&wil->scan_timer);
 		cfg80211_scan_done(wil->scan_request, aborted);
 		wil->scan_request = NULL;
 	} else {
@@ -658,21 +659,27 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 	u8 *cmd;
 	void __iomem *src;
 	ulong flags;
+	unsigned n;
 
 	if (!test_bit(wil_status_reset_done, &wil->status)) {
 		wil_err(wil, "Reset not completed\n");
 		return;
 	}
 
-	for (;;) {
+	for (n = 0;; n++) {
 		u16 len;
 
 		r->head = ioread32(wil->csr + HOST_MBOX +
 				   offsetof(struct wil6210_mbox_ctl, rx.head));
-		if (r->tail == r->head)
+		if (r->tail == r->head) {
+			if (n == 0)
+				wil_dbg_wmi(wil, "No events?\n");
 			return;
+		}
 
-		/* read cmd from tail */
+		wil_dbg_wmi(wil, "Mbox head %08x tail %08x\n",
+			    r->head, r->tail);
+		/* read cmd descriptor from tail */
 		wil_memcpy_fromio_32(&d_tail, wil->csr + HOSTADDR(r->tail),
 				     sizeof(struct wil6210_mbox_ring_desc));
 		if (d_tail.sync == 0) {
@@ -680,13 +687,18 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 			return;
 		}
 
+		/* read cmd header from descriptor */
 		if (0 != wmi_read_hdr(wil, d_tail.addr, &hdr)) {
 			wil_err(wil, "Mbox evt at 0x%08x?\n",
 				le32_to_cpu(d_tail.addr));
 			return;
 		}
-
 		len = le16_to_cpu(hdr.len);
+		wil_dbg_wmi(wil, "Mbox evt %04x %04x %04x %02x\n",
+			    le16_to_cpu(hdr.seq), len, le16_to_cpu(hdr.type),
+			    hdr.flags);
+
+		/* read cmd buffer from descriptor */
 		src = wmi_buffer(wil, d_tail.addr) +
 		      sizeof(struct wil6210_mbox_hdr);
 		evt = kmalloc(ALIGN(offsetof(struct pending_wmi_event,
@@ -702,9 +714,6 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 		iowrite32(0, wil->csr + HOSTADDR(r->tail) +
 			  offsetof(struct wil6210_mbox_ring_desc, sync));
 		/* indicate */
-		wil_dbg_wmi(wil, "Mbox evt %04x %04x %04x %02x\n",
-			    le16_to_cpu(hdr.seq), len, le16_to_cpu(hdr.type),
-			    hdr.flags);
 		if ((hdr.type == WIL_MBOX_HDR_TYPE_WMI) &&
 		    (len >= sizeof(struct wil6210_mbox_hdr_wmi))) {
 			struct wil6210_mbox_hdr_wmi *wmi = &evt->event.wmi;
@@ -734,6 +743,8 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 			wil_dbg_wmi(wil, "queue_work -> %d\n", q);
 		}
 	}
+	if (n > 1)
+		wil_dbg_wmi(wil, "%s -> %d events processed\n", __func__, n);
 }
 
 int wmi_call(struct wil6210_priv *wil, u16 cmdid, void *buf, u16 len,
@@ -802,6 +813,7 @@ int wmi_pcp_start(struct wil6210_priv *wil, int bi, u8 wmi_nettype, u8 chan)
 		.network_type = wmi_nettype,
 		.disable_sec_offload = 1,
 		.channel = chan - 1,
+		.pcp_max_assoc_sta = WIL6210_MAX_CID,
 	};
 	struct {
 		struct wil6210_mbox_hdr_wmi wmi;
diff --git a/drivers/net/wireless/ath/wil6210/wmi.h b/drivers/net/wireless/ath/wil6210/wmi.h
index 50b8528394f4..17334c852866 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.h
+++ b/drivers/net/wireless/ath/wil6210/wmi.h
@@ -28,7 +28,7 @@
 #define __WILOCITY_WMI_H__
 
 /* General */
-
+#define WILOCITY_MAX_ASSOC_STA (8)
 #define WMI_MAC_LEN		(6)
 #define WMI_PROX_RANGE_NUM	(3)
 
@@ -219,15 +219,6 @@ struct wmi_disconnect_sta_cmd {
 	__le16 disconnect_reason;
 } __packed;
 
-/*
- * WMI_RECONNECT_CMDID
- */
-struct wmi_reconnect_cmd {
-	u8 channel;			/* hint */
-	u8 reserved;
-	u8 bssid[WMI_MAC_LEN];		/* mandatory if set */
-} __packed;
-
 
 /*
  * WMI_SET_PMK_CMDID
@@ -296,11 +287,13 @@ enum wmi_scan_type {
 	WMI_LONG_SCAN		= 0,
 	WMI_SHORT_SCAN		= 1,
 	WMI_PBC_SCAN		= 2,
+	WMI_ACTIVE_SCAN		= 3,
+	WMI_DIRECT_SCAN		= 4,
 };
 
 struct wmi_start_scan_cmd {
-	u8 reserved[8];
-
+	u8 direct_scan_mac_addr[6];
+	u8 reserved[2];
 	__le32 home_dwell_time;	/* Max duration in the home channel(ms) */
 	__le32 force_scan_interval;	/* Time interval between scans (ms)*/
 	u8 scan_type;		/* wmi_scan_type */
@@ -332,6 +325,7 @@ struct wmi_probed_ssid_cmd {
 	u8 ssid[WMI_MAX_SSID_LEN];
 } __packed;
 
+
 /*
  * WMI_SET_APPIE_CMDID
  * Add Application specified IE to a management frame
@@ -427,7 +421,7 @@ struct wmi_bcon_ctrl_cmd {
 	__le16 frag_num;
 	__le64 ss_mask;
 	u8 network_type;
-	u8 reserved;
+	u8 pcp_max_assoc_sta;
 	u8 disable_sec_offload;
 	u8 disable_sec;
 } __packed;
@@ -450,7 +444,7 @@ enum wmi_port_role {
 struct wmi_port_allocate_cmd {
 	u8 mac[WMI_MAC_LEN];
 	u8 port_role;
-	u8 midid;
+	u8 mid;
 } __packed;
 
 /*
@@ -467,6 +461,7 @@ struct wmi_delete_port_cmd {
 enum wmi_discovery_mode {
 	WMI_DISCOVERY_MODE_NON_OFFLOAD	= 0,
 	WMI_DISCOVERY_MODE_OFFLOAD	= 1,
+	WMI_DISCOVERY_MODE_PEER2PEER	= 2,
 };
 
 struct wmi_p2p_cfg_cmd {
@@ -493,7 +488,8 @@ struct wmi_power_mgmt_cfg_cmd {
  */
 struct wmi_pcp_start_cmd {
 	__le16 bcon_interval;
-	u8 reserved0[10];
+	u8 pcp_max_assoc_sta;
+	u8 reserved0[9];
 	u8 network_type;
 	u8 channel;
 	u8 disable_sec_offload;
@@ -857,6 +853,7 @@ enum wmi_event_id {
 	WMI_RF_MGMT_STATUS_EVENTID		= 0x1853,
 	WMI_BF_SM_MGMT_DONE_EVENTID		= 0x1838,
 	WMI_RX_MGMT_PACKET_EVENTID		= 0x1840,
+	WMI_TX_MGMT_PACKET_EVENTID		= 0x1841,
 
 	/* Performance monitoring events */
 	WMI_DATA_PORT_OPEN_EVENTID		= 0x1860,
@@ -1040,16 +1037,23 @@ enum wmi_disconnect_reason {
 struct wmi_disconnect_event {
 	__le16 protocol_reason_status;	/* reason code, see 802.11 spec. */
 	u8 bssid[WMI_MAC_LEN];		/* set if known */
-	u8 disconnect_reason;		/* see wmi_disconnect_reason_e */
-	u8 assoc_resp_len;
-	u8 assoc_info[0];
+	u8 disconnect_reason;		/* see wmi_disconnect_reason */
+	u8 assoc_resp_len;		/* not in use */
+	u8 assoc_info[0];		/* not in use */
 } __packed;
 
 /*
  * WMI_SCAN_COMPLETE_EVENTID
  */
+enum scan_status {
+	WMI_SCAN_SUCCESS	= 0,
+	WMI_SCAN_FAILED		= 1,
+	WMI_SCAN_ABORTED	= 2,
+	WMI_SCAN_REJECTED	= 3,
+};
+
 struct wmi_scan_complete_event {
-	__le32 status;
+	__le32 status;	/* scan_status */
 } __packed;
 
 /*
@@ -1256,6 +1260,14 @@ struct wmi_rx_mgmt_info {
 	u8 channel;	/* From Radio MNGR */
 } __packed;
 
+
+/*
+ * WMI_TX_MGMT_PACKET_EVENTID
+ */
+struct wmi_tx_mgmt_packet_event {
+	u8 payload[0];
+} __packed;
+
 struct wmi_rx_mgmt_packet_event {
 	struct wmi_rx_mgmt_info info;
 	u8 payload[0];
diff --git a/drivers/net/wireless/b43/Kconfig b/drivers/net/wireless/b43/Kconfig
index 088d544ec63f..40fd9b7b1426 100644
--- a/drivers/net/wireless/b43/Kconfig
+++ b/drivers/net/wireless/b43/Kconfig
@@ -1,7 +1,8 @@
 config B43
 	tristate "Broadcom 43xx wireless support (mac80211 stack)"
-	depends on SSB_POSSIBLE && MAC80211 && HAS_DMA
-	select SSB
+	depends on (BCMA_POSSIBLE || SSB_POSSIBLE) && MAC80211 && HAS_DMA
+	select BCMA if B43_BCMA
+	select SSB if B43_SSB
 	select FW_LOADER
 	---help---
 	  b43 is a driver for the Broadcom 43xx series wireless devices.
@@ -27,14 +28,33 @@ config B43
 	  If unsure, say M.
 
 config B43_BCMA
-	bool "Support for BCMA bus"
-	depends on B43 && (BCMA = y || BCMA = B43)
-	default y
+	bool
 
 config B43_SSB
 	bool
-	depends on B43 && (SSB = y || SSB = B43)
-	default y
+
+choice
+	prompt "Supported bus types"
+	depends on B43
+	default B43_BUSES_BCMA_AND_SSB
+
+config B43_BUSES_BCMA_AND_SSB
+	bool "BCMA and SSB"
+	depends on BCMA_POSSIBLE && SSB_POSSIBLE
+	select B43_BCMA
+	select B43_SSB
+
+config B43_BUSES_BCMA
+	bool "BCMA only"
+	depends on BCMA_POSSIBLE
+	select B43_BCMA
+
+config B43_BUSES_SSB
+	bool "SSB only"
+	depends on SSB_POSSIBLE
+	select B43_SSB
+
+endchoice
 
 # Auto-select SSB PCI-HOST support, if possible
 config B43_PCI_AUTOSELECT
@@ -53,7 +73,7 @@ config B43_PCICORE_AUTOSELECT
 
 config B43_PCMCIA
 	bool "Broadcom 43xx PCMCIA device support"
-	depends on B43 && SSB_PCMCIAHOST_POSSIBLE
+	depends on B43 && B43_SSB && SSB_PCMCIAHOST_POSSIBLE
 	select SSB_PCMCIAHOST
 	---help---
 	  Broadcom 43xx PCMCIA device support.
@@ -73,7 +93,7 @@ config B43_PCMCIA
 
 config B43_SDIO
 	bool "Broadcom 43xx SDIO device support"
-	depends on B43 && SSB_SDIOHOST_POSSIBLE
+	depends on B43 && B43_SSB && SSB_SDIOHOST_POSSIBLE
 	select SSB_SDIOHOST
 	---help---
 	  Broadcom 43xx device support for Soft-MAC SDIO devices.
@@ -98,7 +118,7 @@ config B43_BCMA_PIO
 
 config B43_PIO
 	bool
-	depends on B43
+	depends on B43 && B43_SSB
 	select SSB_BLOCKIO
 	default y
 
@@ -116,7 +136,7 @@ config B43_PHY_N
 
 config B43_PHY_LP
 	bool "Support for low-power (LP-PHY) devices"
-	depends on B43
+	depends on B43 && B43_SSB
 	default y
 	---help---
 	  Support for the LP-PHY.
diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h
index 54376fddfaf9..4113b6934764 100644
--- a/drivers/net/wireless/b43/b43.h
+++ b/drivers/net/wireless/b43/b43.h
@@ -915,10 +915,6 @@ struct b43_wl {
 	char rng_name[30 + 1];
 #endif /* CONFIG_B43_HWRNG */
 
-	/* List of all wireless devices on this chip */
-	struct list_head devlist;
-	u8 nr_devs;
-
 	bool radiotap_enabled;
 	bool radio_enabled;
 
diff --git a/drivers/net/wireless/b43/bus.h b/drivers/net/wireless/b43/bus.h
index 184c95659279..f3205c6988bc 100644
--- a/drivers/net/wireless/b43/bus.h
+++ b/drivers/net/wireless/b43/bus.h
@@ -5,7 +5,9 @@ enum b43_bus_type {
 #ifdef CONFIG_B43_BCMA
 	B43_BUS_BCMA,
 #endif
+#ifdef CONFIG_B43_SSB
 	B43_BUS_SSB,
+#endif
 };
 
 struct b43_bus_dev {
@@ -52,13 +54,21 @@ struct b43_bus_dev {
 
 static inline bool b43_bus_host_is_pcmcia(struct b43_bus_dev *dev)
 {
+#ifdef CONFIG_B43_SSB
 	return (dev->bus_type == B43_BUS_SSB &&
 		dev->sdev->bus->bustype == SSB_BUSTYPE_PCMCIA);
+#else
+	return false;
+#endif
 }
 static inline bool b43_bus_host_is_sdio(struct b43_bus_dev *dev)
 {
+#ifdef CONFIG_B43_SSB
 	return (dev->bus_type == B43_BUS_SSB &&
 		dev->sdev->bus->bustype == SSB_BUSTYPE_SDIO);
+#else
+	return false;
+#endif
 }
 
 struct b43_bus_dev *b43_bus_dev_bcma_init(struct bcma_device *core);
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 69fc3d65531a..0d6a0bb1f876 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -182,7 +182,7 @@ static struct ieee80211_rate __b43_ratetable[] = {
 #define b43_g_ratetable		(__b43_ratetable + 0)
 #define b43_g_ratetable_size	12
 
-#define CHAN4G(_channel, _freq, _flags) {			\
+#define CHAN2G(_channel, _freq, _flags) {			\
 	.band			= IEEE80211_BAND_2GHZ,		\
 	.center_freq		= (_freq),			\
 	.hw_value		= (_channel),			\
@@ -191,23 +191,31 @@ static struct ieee80211_rate __b43_ratetable[] = {
 	.max_power		= 30,				\
 }
 static struct ieee80211_channel b43_2ghz_chantable[] = {
-	CHAN4G(1, 2412, 0),
-	CHAN4G(2, 2417, 0),
-	CHAN4G(3, 2422, 0),
-	CHAN4G(4, 2427, 0),
-	CHAN4G(5, 2432, 0),
-	CHAN4G(6, 2437, 0),
-	CHAN4G(7, 2442, 0),
-	CHAN4G(8, 2447, 0),
-	CHAN4G(9, 2452, 0),
-	CHAN4G(10, 2457, 0),
-	CHAN4G(11, 2462, 0),
-	CHAN4G(12, 2467, 0),
-	CHAN4G(13, 2472, 0),
-	CHAN4G(14, 2484, 0),
+	CHAN2G(1, 2412, 0),
+	CHAN2G(2, 2417, 0),
+	CHAN2G(3, 2422, 0),
+	CHAN2G(4, 2427, 0),
+	CHAN2G(5, 2432, 0),
+	CHAN2G(6, 2437, 0),
+	CHAN2G(7, 2442, 0),
+	CHAN2G(8, 2447, 0),
+	CHAN2G(9, 2452, 0),
+	CHAN2G(10, 2457, 0),
+	CHAN2G(11, 2462, 0),
+	CHAN2G(12, 2467, 0),
+	CHAN2G(13, 2472, 0),
+	CHAN2G(14, 2484, 0),
 };
-#undef CHAN4G
+#undef CHAN2G
 
+#define CHAN4G(_channel, _flags) {				\
+	.band			= IEEE80211_BAND_5GHZ,		\
+	.center_freq		= 4000 + (5 * (_channel)),	\
+	.hw_value		= (_channel),			\
+	.flags			= (_flags),			\
+	.max_antenna_gain	= 0,				\
+	.max_power		= 30,				\
+}
 #define CHAN5G(_channel, _flags) {				\
 	.band			= IEEE80211_BAND_5GHZ,		\
 	.center_freq		= 5000 + (5 * (_channel)),	\
@@ -217,6 +225,18 @@ static struct ieee80211_channel b43_2ghz_chantable[] = {
 	.max_power		= 30,				\
 }
 static struct ieee80211_channel b43_5ghz_nphy_chantable[] = {
+	CHAN4G(184, 0),		CHAN4G(186, 0),
+	CHAN4G(188, 0),		CHAN4G(190, 0),
+	CHAN4G(192, 0),		CHAN4G(194, 0),
+	CHAN4G(196, 0),		CHAN4G(198, 0),
+	CHAN4G(200, 0),		CHAN4G(202, 0),
+	CHAN4G(204, 0),		CHAN4G(206, 0),
+	CHAN4G(208, 0),		CHAN4G(210, 0),
+	CHAN4G(212, 0),		CHAN4G(214, 0),
+	CHAN4G(216, 0),		CHAN4G(218, 0),
+	CHAN4G(220, 0),		CHAN4G(222, 0),
+	CHAN4G(224, 0),		CHAN4G(226, 0),
+	CHAN4G(228, 0),
 	CHAN5G(32, 0),		CHAN5G(34, 0),
 	CHAN5G(36, 0),		CHAN5G(38, 0),
 	CHAN5G(40, 0),		CHAN5G(42, 0),
@@ -260,18 +280,7 @@ static struct ieee80211_channel b43_5ghz_nphy_chantable[] = {
 	CHAN5G(170, 0),		CHAN5G(172, 0),
 	CHAN5G(174, 0),		CHAN5G(176, 0),
 	CHAN5G(178, 0),		CHAN5G(180, 0),
-	CHAN5G(182, 0),		CHAN5G(184, 0),
-	CHAN5G(186, 0),		CHAN5G(188, 0),
-	CHAN5G(190, 0),		CHAN5G(192, 0),
-	CHAN5G(194, 0),		CHAN5G(196, 0),
-	CHAN5G(198, 0),		CHAN5G(200, 0),
-	CHAN5G(202, 0),		CHAN5G(204, 0),
-	CHAN5G(206, 0),		CHAN5G(208, 0),
-	CHAN5G(210, 0),		CHAN5G(212, 0),
-	CHAN5G(214, 0),		CHAN5G(216, 0),
-	CHAN5G(218, 0),		CHAN5G(220, 0),
-	CHAN5G(222, 0),		CHAN5G(224, 0),
-	CHAN5G(226, 0),		CHAN5G(228, 0),
+	CHAN5G(182, 0),
 };
 
 static struct ieee80211_channel b43_5ghz_aphy_chantable[] = {
@@ -295,6 +304,7 @@ static struct ieee80211_channel b43_5ghz_aphy_chantable[] = {
 	CHAN5G(208, 0),		CHAN5G(212, 0),
 	CHAN5G(216, 0),
 };
+#undef CHAN4G
 #undef CHAN5G
 
 static struct ieee80211_supported_band b43_band_5GHz_nphy = {
@@ -1175,18 +1185,7 @@ static void b43_bcma_phy_reset(struct b43_wldev *dev)
 	bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, flags);
 	udelay(2);
 
-	/* Take PHY out of reset */
-	flags = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
-	flags &= ~B43_BCMA_IOCTL_PHY_RESET;
-	flags |= BCMA_IOCTL_FGC;
-	bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, flags);
-	udelay(1);
-
-	/* Do not force clock anymore */
-	flags = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
-	flags &= ~BCMA_IOCTL_FGC;
-	bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, flags);
-	udelay(1);
+	b43_phy_take_out_of_reset(dev);
 }
 
 static void b43_bcma_wireless_core_reset(struct b43_wldev *dev, bool gmode)
@@ -1195,18 +1194,22 @@ static void b43_bcma_wireless_core_reset(struct b43_wldev *dev, bool gmode)
 		  B43_BCMA_CLKCTLST_PHY_PLL_REQ;
 	u32 status = B43_BCMA_CLKCTLST_80211_PLL_ST |
 		     B43_BCMA_CLKCTLST_PHY_PLL_ST;
+	u32 flags;
+
+	flags = B43_BCMA_IOCTL_PHY_CLKEN;
+	if (gmode)
+		flags |= B43_BCMA_IOCTL_GMODE;
+	b43_device_enable(dev, flags);
 
-	b43_device_enable(dev, B43_BCMA_IOCTL_PHY_CLKEN);
 	bcma_core_set_clockmode(dev->dev->bdev, BCMA_CLKMODE_FAST);
 	b43_bcma_phy_reset(dev);
 	bcma_core_pll_ctl(dev->dev->bdev, req, status, true);
 }
 #endif
 
+#ifdef CONFIG_B43_SSB
 static void b43_ssb_wireless_core_reset(struct b43_wldev *dev, bool gmode)
 {
-	struct ssb_device *sdev = dev->dev->sdev;
-	u32 tmslow;
 	u32 flags = 0;
 
 	if (gmode)
@@ -1218,18 +1221,9 @@ static void b43_ssb_wireless_core_reset(struct b43_wldev *dev, bool gmode)
 	b43_device_enable(dev, flags);
 	msleep(2);		/* Wait for the PLL to turn on. */
 
-	/* Now take the PHY out of Reset again */
-	tmslow = ssb_read32(sdev, SSB_TMSLOW);
-	tmslow |= SSB_TMSLOW_FGC;
-	tmslow &= ~B43_TMSLOW_PHYRESET;
-	ssb_write32(sdev, SSB_TMSLOW, tmslow);
-	ssb_read32(sdev, SSB_TMSLOW);	/* flush */
-	msleep(1);
-	tmslow &= ~SSB_TMSLOW_FGC;
-	ssb_write32(sdev, SSB_TMSLOW, tmslow);
-	ssb_read32(sdev, SSB_TMSLOW);	/* flush */
-	msleep(1);
+	b43_phy_take_out_of_reset(dev);
 }
+#endif
 
 void b43_wireless_core_reset(struct b43_wldev *dev, bool gmode)
 {
@@ -2704,32 +2698,37 @@ static int b43_upload_initvals(struct b43_wldev *dev)
 	struct b43_firmware *fw = &dev->fw;
 	const struct b43_iv *ivals;
 	size_t count;
-	int err;
 
 	hdr = (const struct b43_fw_header *)(fw->initvals.data->data);
 	ivals = (const struct b43_iv *)(fw->initvals.data->data + hdr_len);
 	count = be32_to_cpu(hdr->size);
-	err = b43_write_initvals(dev, ivals, count,
+	return b43_write_initvals(dev, ivals, count,
 				 fw->initvals.data->size - hdr_len);
-	if (err)
-		goto out;
-	if (fw->initvals_band.data) {
-		hdr = (const struct b43_fw_header *)(fw->initvals_band.data->data);
-		ivals = (const struct b43_iv *)(fw->initvals_band.data->data + hdr_len);
-		count = be32_to_cpu(hdr->size);
-		err = b43_write_initvals(dev, ivals, count,
-					 fw->initvals_band.data->size - hdr_len);
-		if (err)
-			goto out;
-	}
-out:
+}
 
-	return err;
+static int b43_upload_initvals_band(struct b43_wldev *dev)
+{
+	const size_t hdr_len = sizeof(struct b43_fw_header);
+	const struct b43_fw_header *hdr;
+	struct b43_firmware *fw = &dev->fw;
+	const struct b43_iv *ivals;
+	size_t count;
+
+	if (!fw->initvals_band.data)
+		return 0;
+
+	hdr = (const struct b43_fw_header *)(fw->initvals_band.data->data);
+	ivals = (const struct b43_iv *)(fw->initvals_band.data->data + hdr_len);
+	count = be32_to_cpu(hdr->size);
+	return b43_write_initvals(dev, ivals, count,
+				  fw->initvals_band.data->size - hdr_len);
 }
 
 /* Initialize the GPIOs
  * http://bcm-specs.sipsolutions.net/GPIO
  */
+
+#ifdef CONFIG_B43_SSB
 static struct ssb_device *b43_ssb_gpio_dev(struct b43_wldev *dev)
 {
 	struct ssb_bus *bus = dev->dev->sdev->bus;
@@ -2740,10 +2739,13 @@ static struct ssb_device *b43_ssb_gpio_dev(struct b43_wldev *dev)
 	return bus->chipco.dev;
 #endif
 }
+#endif
 
 static int b43_gpio_init(struct b43_wldev *dev)
 {
+#ifdef CONFIG_B43_SSB
 	struct ssb_device *gpiodev;
+#endif
 	u32 mask, set;
 
 	b43_maskset32(dev, B43_MMIO_MACCTL, ~B43_MACCTL_GPOUTSMSK, 0);
@@ -2802,7 +2804,9 @@ static int b43_gpio_init(struct b43_wldev *dev)
 /* Turn off all GPIO stuff. Call this on module unload, for example. */
 static void b43_gpio_cleanup(struct b43_wldev *dev)
 {
+#ifdef CONFIG_B43_SSB
 	struct ssb_device *gpiodev;
+#endif
 
 	switch (dev->dev->bus_type) {
 #ifdef CONFIG_B43_BCMA
@@ -3086,6 +3090,10 @@ static int b43_chip_init(struct b43_wldev *dev)
 	if (err)
 		goto err_gpio_clean;
 
+	err = b43_upload_initvals_band(dev);
+	if (err)
+		goto err_gpio_clean;
+
 	/* Turn the Analog on and initialize the PHY. */
 	phy->ops->switch_analog(dev, 1);
 	err = b43_phy_init(dev);
@@ -3685,37 +3693,6 @@ static void b43_op_set_tsf(struct ieee80211_hw *hw,
 	mutex_unlock(&wl->mutex);
 }
 
-static void b43_put_phy_into_reset(struct b43_wldev *dev)
-{
-	u32 tmp;
-
-	switch (dev->dev->bus_type) {
-#ifdef CONFIG_B43_BCMA
-	case B43_BUS_BCMA:
-		b43err(dev->wl,
-		       "Putting PHY into reset not supported on BCMA\n");
-		break;
-#endif
-#ifdef CONFIG_B43_SSB
-	case B43_BUS_SSB:
-		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
-		tmp &= ~B43_TMSLOW_GMODE;
-		tmp |= B43_TMSLOW_PHYRESET;
-		tmp |= SSB_TMSLOW_FGC;
-		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
-		msleep(1);
-
-		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
-		tmp &= ~SSB_TMSLOW_FGC;
-		tmp |= B43_TMSLOW_PHYRESET;
-		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
-		msleep(1);
-
-		break;
-#endif
-	}
-}
-
 static const char *band_to_string(enum ieee80211_band band)
 {
 	switch (band) {
@@ -3731,94 +3708,75 @@ static const char *band_to_string(enum ieee80211_band band)
 }
 
 /* Expects wl->mutex locked */
-static int b43_switch_band(struct b43_wl *wl, struct ieee80211_channel *chan)
+static int b43_switch_band(struct b43_wldev *dev,
+			   struct ieee80211_channel *chan)
 {
-	struct b43_wldev *up_dev = NULL;
-	struct b43_wldev *down_dev;
-	struct b43_wldev *d;
-	int err;
-	bool uninitialized_var(gmode);
-	int prev_status;
+	struct b43_phy *phy = &dev->phy;
+	bool gmode;
+	u32 tmp;
 
-	/* Find a device and PHY which supports the band. */
-	list_for_each_entry(d, &wl->devlist, list) {
-		switch (chan->band) {
-		case IEEE80211_BAND_5GHZ:
-			if (d->phy.supports_5ghz) {
-				up_dev = d;
-				gmode = false;
-			}
-			break;
-		case IEEE80211_BAND_2GHZ:
-			if (d->phy.supports_2ghz) {
-				up_dev = d;
-				gmode = true;
-			}
-			break;
-		default:
-			B43_WARN_ON(1);
-			return -EINVAL;
-		}
-		if (up_dev)
-			break;
+	switch (chan->band) {
+	case IEEE80211_BAND_5GHZ:
+		gmode = false;
+		break;
+	case IEEE80211_BAND_2GHZ:
+		gmode = true;
+		break;
+	default:
+		B43_WARN_ON(1);
+		return -EINVAL;
 	}
-	if (!up_dev) {
-		b43err(wl, "Could not find a device for %s-GHz band operation\n",
+
+	if (!((gmode && phy->supports_2ghz) ||
+	      (!gmode && phy->supports_5ghz))) {
+		b43err(dev->wl, "This device doesn't support %s-GHz band\n",
 		       band_to_string(chan->band));
 		return -ENODEV;
 	}
-	if ((up_dev == wl->current_dev) &&
-	    (!!wl->current_dev->phy.gmode == !!gmode)) {
+
+	if (!!phy->gmode == !!gmode) {
 		/* This device is already running. */
 		return 0;
 	}
-	b43dbg(wl, "Switching to %s-GHz band\n",
+
+	b43dbg(dev->wl, "Switching to %s GHz band\n",
 	       band_to_string(chan->band));
-	down_dev = wl->current_dev;
 
-	prev_status = b43_status(down_dev);
-	/* Shutdown the currently running core. */
-	if (prev_status >= B43_STAT_STARTED)
-		down_dev = b43_wireless_core_stop(down_dev);
-	if (prev_status >= B43_STAT_INITIALIZED)
-		b43_wireless_core_exit(down_dev);
+	/* Some new devices don't need disabling radio for band switching */
+	if (!(phy->type == B43_PHYTYPE_N && phy->rev >= 3))
+		b43_software_rfkill(dev, true);
 
-	if (down_dev != up_dev) {
-		/* We switch to a different core, so we put PHY into
-		 * RESET on the old core. */
-		b43_put_phy_into_reset(down_dev);
+	phy->gmode = gmode;
+	b43_phy_put_into_reset(dev);
+	switch (dev->dev->bus_type) {
+#ifdef CONFIG_B43_BCMA
+	case B43_BUS_BCMA:
+		tmp = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
+		if (gmode)
+			tmp |= B43_BCMA_IOCTL_GMODE;
+		else
+			tmp &= ~B43_BCMA_IOCTL_GMODE;
+		bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, tmp);
+		break;
+#endif
+#ifdef CONFIG_B43_SSB
+	case B43_BUS_SSB:
+		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
+		if (gmode)
+			tmp |= B43_TMSLOW_GMODE;
+		else
+			tmp &= ~B43_TMSLOW_GMODE;
+		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
+		break;
+#endif
 	}
+	b43_phy_take_out_of_reset(dev);
 
-	/* Now start the new core. */
-	up_dev->phy.gmode = gmode;
-	if (prev_status >= B43_STAT_INITIALIZED) {
-		err = b43_wireless_core_init(up_dev);
-		if (err) {
-			b43err(wl, "Fatal: Could not initialize device for "
-			       "selected %s-GHz band\n",
-			       band_to_string(chan->band));
-			goto init_failure;
-		}
-	}
-	if (prev_status >= B43_STAT_STARTED) {
-		err = b43_wireless_core_start(up_dev);
-		if (err) {
-			b43err(wl, "Fatal: Could not start device for "
-			       "selected %s-GHz band\n",
-			       band_to_string(chan->band));
-			b43_wireless_core_exit(up_dev);
-			goto init_failure;
-		}
-	}
-	B43_WARN_ON(b43_status(up_dev) != prev_status);
+	b43_upload_initvals_band(dev);
 
-	wl->current_dev = up_dev;
+	b43_phy_init(dev);
 
 	return 0;
-init_failure:
-	/* Whoops, failed to init the new core. No core is operating now. */
-	wl->current_dev = NULL;
-	return err;
 }
 
 /* Write the short and long frame retry limit values. */
@@ -3851,8 +3809,10 @@ static int b43_op_config(struct ieee80211_hw *hw, u32 changed)
 
 	dev = wl->current_dev;
 
+	b43_mac_suspend(dev);
+
 	/* Switch the band (if necessary). This might change the active core. */
-	err = b43_switch_band(wl, conf->chandef.chan);
+	err = b43_switch_band(dev, conf->chandef.chan);
 	if (err)
 		goto out_unlock_mutex;
 
@@ -3871,8 +3831,6 @@ static int b43_op_config(struct ieee80211_hw *hw, u32 changed)
 	else
 		phy->is_40mhz = false;
 
-	b43_mac_suspend(dev);
-
 	if (changed & IEEE80211_CONF_CHANGE_RETRY_LIMITS)
 		b43_set_retry_limits(dev, conf->short_frame_max_tx_count,
 					  conf->long_frame_max_tx_count);
@@ -4582,8 +4540,12 @@ static void b43_imcfglo_timeouts_workaround(struct b43_wldev *dev)
 	struct ssb_bus *bus;
 	u32 tmp;
 
+#ifdef CONFIG_B43_SSB
 	if (dev->dev->bus_type != B43_BUS_SSB)
 		return;
+#else
+	return;
+#endif
 
 	bus = dev->dev->sdev->bus;
 
@@ -4738,7 +4700,7 @@ static int b43_wireless_core_init(struct b43_wldev *dev)
 	}
 	if (sprom->boardflags_lo & B43_BFL_XTAL_NOSLOW)
 		hf |= B43_HF_DSCRQ; /* Disable slowclock requests from ucode. */
-#ifdef CONFIG_SSB_DRIVER_PCICORE
+#if defined(CONFIG_B43_SSB) && defined(CONFIG_SSB_DRIVER_PCICORE)
 	if (dev->dev->bus_type == B43_BUS_SSB &&
 	    dev->dev->sdev->bus->bustype == SSB_BUSTYPE_PCI &&
 	    dev->dev->sdev->bus->pcicore.dev->id.revision <= 10)
@@ -5129,10 +5091,82 @@ static void b43_wireless_core_detach(struct b43_wldev *dev)
 	b43_phy_free(dev);
 }
 
+static void b43_supported_bands(struct b43_wldev *dev, bool *have_2ghz_phy,
+				bool *have_5ghz_phy)
+{
+	u16 dev_id = 0;
+
+#ifdef CONFIG_B43_BCMA
+	if (dev->dev->bus_type == B43_BUS_BCMA &&
+	    dev->dev->bdev->bus->hosttype == BCMA_HOSTTYPE_PCI)
+		dev_id = dev->dev->bdev->bus->host_pci->device;
+#endif
+#ifdef CONFIG_B43_SSB
+	if (dev->dev->bus_type == B43_BUS_SSB &&
+	    dev->dev->sdev->bus->bustype == SSB_BUSTYPE_PCI)
+		dev_id = dev->dev->sdev->bus->host_pci->device;
+#endif
+	/* Override with SPROM value if available */
+	if (dev->dev->bus_sprom->dev_id)
+		dev_id = dev->dev->bus_sprom->dev_id;
+
+	/* Note: below IDs can be "virtual" (not maching e.g. real PCI ID) */
+	switch (dev_id) {
+	case 0x4324: /* BCM4306 */
+	case 0x4312: /* BCM4311 */
+	case 0x4319: /* BCM4318 */
+	case 0x4328: /* BCM4321 */
+	case 0x432b: /* BCM4322 */
+	case 0x4350: /* BCM43222 */
+	case 0x4353: /* BCM43224 */
+	case 0x0576: /* BCM43224 */
+	case 0x435f: /* BCM6362 */
+	case 0x4331: /* BCM4331 */
+	case 0x4359: /* BCM43228 */
+	case 0x43a0: /* BCM4360 */
+	case 0x43b1: /* BCM4352 */
+		/* Dual band devices */
+		*have_2ghz_phy = true;
+		*have_5ghz_phy = true;
+		return;
+	case 0x4321: /* BCM4306 */
+	case 0x4313: /* BCM4311 */
+	case 0x431a: /* BCM4318 */
+	case 0x432a: /* BCM4321 */
+	case 0x432d: /* BCM4322 */
+	case 0x4352: /* BCM43222 */
+	case 0x4333: /* BCM4331 */
+	case 0x43a2: /* BCM4360 */
+	case 0x43b3: /* BCM4352 */
+		/* 5 GHz only devices */
+		*have_2ghz_phy = false;
+		*have_5ghz_phy = true;
+		return;
+	}
+
+	/* As a fallback, try to guess using PHY type */
+	switch (dev->phy.type) {
+	case B43_PHYTYPE_A:
+		*have_2ghz_phy = false;
+		*have_5ghz_phy = true;
+		return;
+	case B43_PHYTYPE_G:
+	case B43_PHYTYPE_N:
+	case B43_PHYTYPE_LP:
+	case B43_PHYTYPE_HT:
+	case B43_PHYTYPE_LCN:
+		*have_2ghz_phy = true;
+		*have_5ghz_phy = false;
+		return;
+	}
+
+	B43_WARN_ON(1);
+}
+
 static int b43_wireless_core_attach(struct b43_wldev *dev)
 {
 	struct b43_wl *wl = dev->wl;
-	struct pci_dev *pdev = NULL;
+	struct b43_phy *phy = &dev->phy;
 	int err;
 	u32 tmp;
 	bool have_2ghz_phy = false, have_5ghz_phy = false;
@@ -5144,19 +5178,15 @@ static int b43_wireless_core_attach(struct b43_wldev *dev)
 	 * that in core_init(), too.
 	 */
 
-#ifdef CONFIG_B43_SSB
-	if (dev->dev->bus_type == B43_BUS_SSB &&
-	    dev->dev->sdev->bus->bustype == SSB_BUSTYPE_PCI)
-		pdev = dev->dev->sdev->bus->host_pci;
-#endif
-
 	err = b43_bus_powerup(dev, 0);
 	if (err) {
 		b43err(wl, "Bus powerup failed\n");
 		goto out;
 	}
 
-	/* Get the PHY type. */
+	phy->do_full_init = true;
+
+	/* Try to guess supported bands for the first init needs */
 	switch (dev->dev->bus_type) {
 #ifdef CONFIG_B43_BCMA
 	case B43_BUS_BCMA:
@@ -5178,51 +5208,32 @@ static int b43_wireless_core_attach(struct b43_wldev *dev)
 	}
 
 	dev->phy.gmode = have_2ghz_phy;
-	dev->phy.radio_on = true;
 	b43_wireless_core_reset(dev, dev->phy.gmode);
 
+	/* Get the PHY type. */
 	err = b43_phy_versioning(dev);
 	if (err)
 		goto err_powerdown;
-	/* Check if this device supports multiband. */
-	if (!pdev ||
-	    (pdev->device != 0x4312 &&
-	     pdev->device != 0x4319 && pdev->device != 0x4324)) {
-		/* No multiband support. */
-		have_2ghz_phy = false;
+
+	/* Get real info about supported bands */
+	b43_supported_bands(dev, &have_2ghz_phy, &have_5ghz_phy);
+
+	/* We don't support 5 GHz on some PHYs yet */
+	switch (dev->phy.type) {
+	case B43_PHYTYPE_A:
+	case B43_PHYTYPE_G:
+	case B43_PHYTYPE_N:
+	case B43_PHYTYPE_LP:
+	case B43_PHYTYPE_HT:
+		b43warn(wl, "5 GHz band is unsupported on this PHY\n");
 		have_5ghz_phy = false;
-		switch (dev->phy.type) {
-		case B43_PHYTYPE_A:
-			have_5ghz_phy = true;
-			break;
-		case B43_PHYTYPE_LP: //FIXME not always!
-#if 0 //FIXME enabling 5GHz causes a NULL pointer dereference
-			have_5ghz_phy = 1;
-#endif
-		case B43_PHYTYPE_G:
-		case B43_PHYTYPE_N:
-		case B43_PHYTYPE_HT:
-		case B43_PHYTYPE_LCN:
-			have_2ghz_phy = true;
-			break;
-		default:
-			B43_WARN_ON(1);
-		}
 	}
-	if (dev->phy.type == B43_PHYTYPE_A) {
-		/* FIXME */
-		b43err(wl, "IEEE 802.11a devices are unsupported\n");
+
+	if (!have_2ghz_phy && !have_5ghz_phy) {
+		b43err(wl, "b43 can't support any band on this device\n");
 		err = -EOPNOTSUPP;
 		goto err_powerdown;
 	}
-	if (1 /* disable A-PHY */) {
-		/* FIXME: For now we disable the A-PHY on multi-PHY devices. */
-		if (dev->phy.type != B43_PHYTYPE_N &&
-		    dev->phy.type != B43_PHYTYPE_LP) {
-			have_2ghz_phy = true;
-			have_5ghz_phy = false;
-		}
-	}
 
 	err = b43_phy_allocate(dev);
 	if (err)
@@ -5270,7 +5281,6 @@ static void b43_one_core_detach(struct b43_bus_dev *dev)
 	b43_debugfs_remove_device(wldev);
 	b43_wireless_core_detach(wldev);
 	list_del(&wldev->list);
-	wl->nr_devs--;
 	b43_bus_set_wldev(dev, NULL);
 	kfree(wldev);
 }
@@ -5295,8 +5305,6 @@ static int b43_one_core_attach(struct b43_bus_dev *dev, struct b43_wl *wl)
 	if (err)
 		goto err_kfree_wldev;
 
-	list_add(&wldev->list, &wl->devlist);
-	wl->nr_devs++;
 	b43_bus_set_wldev(dev, wldev);
 	b43_debugfs_add_device(wldev);
 
@@ -5314,6 +5322,7 @@ static int b43_one_core_attach(struct b43_bus_dev *dev, struct b43_wl *wl)
 	(pdev->subsystem_vendor == PCI_VENDOR_ID_##_subvendor) &&	\
 	(pdev->subsystem_device == _subdevice)				)
 
+#ifdef CONFIG_B43_SSB
 static void b43_sprom_fixup(struct ssb_bus *bus)
 {
 	struct pci_dev *pdev;
@@ -5345,6 +5354,7 @@ static void b43_wireless_exit(struct b43_bus_dev *dev, struct b43_wl *wl)
 	ssb_set_devtypedata(dev->sdev, NULL);
 	ieee80211_free_hw(hw);
 }
+#endif
 
 static struct b43_wl *b43_wireless_init(struct b43_bus_dev *dev)
 {
@@ -5386,7 +5396,6 @@ static struct b43_wl *b43_wireless_init(struct b43_bus_dev *dev)
 	wl->hw = hw;
 	mutex_init(&wl->mutex);
 	spin_lock_init(&wl->hardirq_lock);
-	INIT_LIST_HEAD(&wl->devlist);
 	INIT_WORK(&wl->beacon_update_trigger, b43_beacon_update_trigger_work);
 	INIT_WORK(&wl->txpower_adjust_work, b43_phy_txpower_adjust_work);
 	INIT_WORK(&wl->tx_work, b43_tx_work);
@@ -5486,39 +5495,42 @@ int b43_ssb_probe(struct ssb_device *sdev, const struct ssb_device_id *id)
 	struct b43_bus_dev *dev;
 	struct b43_wl *wl;
 	int err;
-	int first = 0;
 
 	dev = b43_bus_dev_ssb_init(sdev);
 	if (!dev)
 		return -ENOMEM;
 
 	wl = ssb_get_devtypedata(sdev);
-	if (!wl) {
-		/* Probing the first core. Must setup common struct b43_wl */
-		first = 1;
-		b43_sprom_fixup(sdev->bus);
-		wl = b43_wireless_init(dev);
-		if (IS_ERR(wl)) {
-			err = PTR_ERR(wl);
-			goto out;
-		}
-		ssb_set_devtypedata(sdev, wl);
-		B43_WARN_ON(ssb_get_devtypedata(sdev) != wl);
+	if (wl) {
+		b43err(NULL, "Dual-core devices are not supported\n");
+		err = -ENOTSUPP;
+		goto err_ssb_kfree_dev;
+	}
+
+	b43_sprom_fixup(sdev->bus);
+
+	wl = b43_wireless_init(dev);
+	if (IS_ERR(wl)) {
+		err = PTR_ERR(wl);
+		goto err_ssb_kfree_dev;
 	}
+	ssb_set_devtypedata(sdev, wl);
+	B43_WARN_ON(ssb_get_devtypedata(sdev) != wl);
+
 	err = b43_one_core_attach(dev, wl);
 	if (err)
-		goto err_wireless_exit;
+		goto err_ssb_wireless_exit;
 
 	/* setup and start work to load firmware */
 	INIT_WORK(&wl->firmware_load, b43_request_firmware);
 	schedule_work(&wl->firmware_load);
 
-      out:
 	return err;
 
-      err_wireless_exit:
-	if (first)
-		b43_wireless_exit(dev, wl);
+err_ssb_wireless_exit:
+	b43_wireless_exit(dev, wl);
+err_ssb_kfree_dev:
+	kfree(dev);
 	return err;
 }
 
@@ -5546,13 +5558,8 @@ static void b43_ssb_remove(struct ssb_device *sdev)
 	/* Unregister HW RNG driver */
 	b43_rng_exit(wl);
 
-	if (list_empty(&wl->devlist)) {
-		b43_leds_unregister(wl);
-		/* Last core on the chip unregistered.
-		 * We can destroy common struct b43_wl.
-		 */
-		b43_wireless_exit(dev, wl);
-	}
+	b43_leds_unregister(wl);
+	b43_wireless_exit(dev, wl);
 }
 
 static struct ssb_driver b43_ssb_driver = {
diff --git a/drivers/net/wireless/b43/phy_common.c b/drivers/net/wireless/b43/phy_common.c
index dbaa51890198..08244b3b327e 100644
--- a/drivers/net/wireless/b43/phy_common.c
+++ b/drivers/net/wireless/b43/phy_common.c
@@ -96,12 +96,16 @@ int b43_phy_init(struct b43_wldev *dev)
 
 	phy->channel = ops->get_default_chan(dev);
 
-	ops->software_rfkill(dev, false);
+	phy->ops->switch_analog(dev, true);
+	b43_software_rfkill(dev, false);
+
 	err = ops->init(dev);
 	if (err) {
 		b43err(dev->wl, "PHY init failed\n");
 		goto err_block_rf;
 	}
+	phy->do_full_init = false;
+
 	/* Make sure to switch hardware and firmware (SHM) to
 	 * the default channel. */
 	err = b43_switch_channel(dev, ops->get_default_chan(dev));
@@ -113,10 +117,11 @@ int b43_phy_init(struct b43_wldev *dev)
 	return 0;
 
 err_phy_exit:
+	phy->do_full_init = true;
 	if (ops->exit)
 		ops->exit(dev);
 err_block_rf:
-	ops->software_rfkill(dev, true);
+	b43_software_rfkill(dev, true);
 
 	return err;
 }
@@ -125,7 +130,8 @@ void b43_phy_exit(struct b43_wldev *dev)
 {
 	const struct b43_phy_operations *ops = dev->phy.ops;
 
-	ops->software_rfkill(dev, true);
+	b43_software_rfkill(dev, true);
+	dev->phy.do_full_init = true;
 	if (ops->exit)
 		ops->exit(dev);
 }
@@ -312,6 +318,90 @@ void b43_phy_maskset(struct b43_wldev *dev, u16 offset, u16 mask, u16 set)
 	}
 }
 
+void b43_phy_put_into_reset(struct b43_wldev *dev)
+{
+	u32 tmp;
+
+	switch (dev->dev->bus_type) {
+#ifdef CONFIG_B43_BCMA
+	case B43_BUS_BCMA:
+		tmp = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
+		tmp &= ~B43_BCMA_IOCTL_GMODE;
+		tmp |= B43_BCMA_IOCTL_PHY_RESET;
+		tmp |= BCMA_IOCTL_FGC;
+		bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, tmp);
+		udelay(1);
+
+		tmp = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
+		tmp &= ~BCMA_IOCTL_FGC;
+		bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, tmp);
+		udelay(1);
+		break;
+#endif
+#ifdef CONFIG_B43_SSB
+	case B43_BUS_SSB:
+		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
+		tmp &= ~B43_TMSLOW_GMODE;
+		tmp |= B43_TMSLOW_PHYRESET;
+		tmp |= SSB_TMSLOW_FGC;
+		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
+		usleep_range(1000, 2000);
+
+		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
+		tmp &= ~SSB_TMSLOW_FGC;
+		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
+		usleep_range(1000, 2000);
+
+		break;
+#endif
+	}
+}
+
+void b43_phy_take_out_of_reset(struct b43_wldev *dev)
+{
+	u32 tmp;
+
+	switch (dev->dev->bus_type) {
+#ifdef CONFIG_B43_BCMA
+	case B43_BUS_BCMA:
+		/* Unset reset bit (with forcing clock) */
+		tmp = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
+		tmp &= ~B43_BCMA_IOCTL_PHY_RESET;
+		tmp &= ~B43_BCMA_IOCTL_PHY_CLKEN;
+		tmp |= BCMA_IOCTL_FGC;
+		bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, tmp);
+		udelay(1);
+
+		/* Do not force clock anymore */
+		tmp = bcma_aread32(dev->dev->bdev, BCMA_IOCTL);
+		tmp &= ~BCMA_IOCTL_FGC;
+		tmp |= B43_BCMA_IOCTL_PHY_CLKEN;
+		bcma_awrite32(dev->dev->bdev, BCMA_IOCTL, tmp);
+		udelay(1);
+		break;
+#endif
+#ifdef CONFIG_B43_SSB
+	case B43_BUS_SSB:
+		/* Unset reset bit (with forcing clock) */
+		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
+		tmp &= ~B43_TMSLOW_PHYRESET;
+		tmp &= ~B43_TMSLOW_PHYCLKEN;
+		tmp |= SSB_TMSLOW_FGC;
+		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
+		ssb_read32(dev->dev->sdev, SSB_TMSLOW); /* flush */
+		usleep_range(1000, 2000);
+
+		tmp = ssb_read32(dev->dev->sdev, SSB_TMSLOW);
+		tmp &= ~SSB_TMSLOW_FGC;
+		tmp |= B43_TMSLOW_PHYCLKEN;
+		ssb_write32(dev->dev->sdev, SSB_TMSLOW, tmp);
+		ssb_read32(dev->dev->sdev, SSB_TMSLOW); /* flush */
+		usleep_range(1000, 2000);
+		break;
+#endif
+	}
+}
+
 int b43_switch_channel(struct b43_wldev *dev, unsigned int new_channel)
 {
 	struct b43_phy *phy = &(dev->phy);
diff --git a/drivers/net/wireless/b43/phy_common.h b/drivers/net/wireless/b43/phy_common.h
index f1b999349876..4ad6240d9ff4 100644
--- a/drivers/net/wireless/b43/phy_common.h
+++ b/drivers/net/wireless/b43/phy_common.h
@@ -231,9 +231,12 @@ struct b43_phy {
 	/* HT info */
 	bool is_40mhz;
 
-	/* GMODE bit enabled? */
+	/* Is GMODE (2 GHz mode) bit enabled? */
 	bool gmode;
 
+	/* After power reset full init has to be performed */
+	bool do_full_init;
+
 	/* Analog Type */
 	u8 analog;
 	/* B43_PHYTYPE_ */
@@ -390,6 +393,9 @@ void b43_phy_lock(struct b43_wldev *dev);
  */
 void b43_phy_unlock(struct b43_wldev *dev);
 
+void b43_phy_put_into_reset(struct b43_wldev *dev);
+void b43_phy_take_out_of_reset(struct b43_wldev *dev);
+
 /**
  * b43_switch_channel - Switch to another channel
  */
diff --git a/drivers/net/wireless/b43/phy_g.c b/drivers/net/wireless/b43/phy_g.c
index 12f467b8d564..8f5c14bc10e6 100644
--- a/drivers/net/wireless/b43/phy_g.c
+++ b/drivers/net/wireless/b43/phy_g.c
@@ -1587,6 +1587,7 @@ static void b43_phy_initb5(struct b43_wldev *dev)
 	b43_write16(dev, 0x03E4, (b43_read16(dev, 0x03E4) & 0xFFC0) | 0x0004);
 }
 
+/* http://bcm-v4.sipsolutions.net/802.11/PHY/Init/B6 */
 static void b43_phy_initb6(struct b43_wldev *dev)
 {
 	struct b43_phy *phy = &dev->phy;
@@ -1670,7 +1671,7 @@ static void b43_phy_initb6(struct b43_wldev *dev)
 		b43_radio_write16(dev, 0x50, 0x20);
 	}
 	if (phy->radio_rev <= 2) {
-		b43_radio_write16(dev, 0x7C, 0x20);
+		b43_radio_write16(dev, 0x50, 0x20);
 		b43_radio_write16(dev, 0x5A, 0x70);
 		b43_radio_write16(dev, 0x5B, 0x7B);
 		b43_radio_write16(dev, 0x5C, 0xB0);
@@ -1686,9 +1687,8 @@ static void b43_phy_initb6(struct b43_wldev *dev)
 		b43_phy_write(dev, 0x2A, 0x8AC0);
 	b43_phy_write(dev, 0x0038, 0x0668);
 	b43_set_txpower_g(dev, &gphy->bbatt, &gphy->rfatt, gphy->tx_control);
-	if (phy->radio_rev <= 5) {
+	if (phy->radio_rev == 4 || phy->radio_rev == 5)
 		b43_phy_maskset(dev, 0x5D, 0xFF80, 0x0003);
-	}
 	if (phy->radio_rev <= 2)
 		b43_radio_write16(dev, 0x005D, 0x000D);
 
diff --git a/drivers/net/wireless/b43/phy_n.c b/drivers/net/wireless/b43/phy_n.c
index 24ccbe96e0c8..86569f6a8705 100644
--- a/drivers/net/wireless/b43/phy_n.c
+++ b/drivers/net/wireless/b43/phy_n.c
@@ -257,6 +257,72 @@ static void b43_nphy_rf_ctl_override(struct b43_wldev *dev, u16 field,
 	}
 }
 
+static void b43_nphy_rf_ctl_intc_override_rev7(struct b43_wldev *dev,
+					       enum n_intc_override intc_override,
+					       u16 value, u8 core_sel)
+{
+	u16 reg, tmp, tmp2, val;
+	int core;
+
+	for (core = 0; core < 2; core++) {
+		if ((core_sel == 1 && core != 0) ||
+		    (core_sel == 2 && core != 1))
+			continue;
+
+		reg = (core == 0) ? B43_NPHY_RFCTL_INTC1 : B43_NPHY_RFCTL_INTC2;
+
+		switch (intc_override) {
+		case N_INTC_OVERRIDE_OFF:
+			b43_phy_write(dev, reg, 0);
+			b43_nphy_force_rf_sequence(dev, B43_RFSEQ_RESET2RX);
+			break;
+		case N_INTC_OVERRIDE_TRSW:
+			b43_phy_maskset(dev, reg, ~0xC0, value << 6);
+			b43_phy_set(dev, reg, 0x400);
+
+			b43_phy_mask(dev, 0x2ff, ~0xC000 & 0xFFFF);
+			b43_phy_set(dev, 0x2ff, 0x2000);
+			b43_phy_set(dev, 0x2ff, 0x0001);
+			break;
+		case N_INTC_OVERRIDE_PA:
+			tmp = 0x0030;
+			if (b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ)
+				val = value << 5;
+			else
+				val = value << 4;
+			b43_phy_maskset(dev, reg, ~tmp, val);
+			b43_phy_set(dev, reg, 0x1000);
+			break;
+		case N_INTC_OVERRIDE_EXT_LNA_PU:
+			if (b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ) {
+				tmp = 0x0001;
+				tmp2 = 0x0004;
+				val = value;
+			} else {
+				tmp = 0x0004;
+				tmp2 = 0x0001;
+				val = value << 2;
+			}
+			b43_phy_maskset(dev, reg, ~tmp, val);
+			b43_phy_mask(dev, reg, ~tmp2);
+			break;
+		case N_INTC_OVERRIDE_EXT_LNA_GAIN:
+			if (b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ) {
+				tmp = 0x0002;
+				tmp2 = 0x0008;
+				val = value << 1;
+			} else {
+				tmp = 0x0008;
+				tmp2 = 0x0002;
+				val = value << 3;
+			}
+			b43_phy_maskset(dev, reg, ~tmp, val);
+			b43_phy_mask(dev, reg, ~tmp2);
+			break;
+		}
+	}
+}
+
 /* http://bcm-v4.sipsolutions.net/802.11/PHY/N/RFCtrlIntcOverride */
 static void b43_nphy_rf_ctl_intc_override(struct b43_wldev *dev,
 					  enum n_intc_override intc_override,
@@ -265,6 +331,12 @@ static void b43_nphy_rf_ctl_intc_override(struct b43_wldev *dev,
 	u8 i, j;
 	u16 reg, tmp, val;
 
+	if (dev->phy.rev >= 7) {
+		b43_nphy_rf_ctl_intc_override_rev7(dev, intc_override, value,
+						   core);
+		return;
+	}
+
 	B43_WARN_ON(dev->phy.rev < 3);
 
 	for (i = 0; i < 2; i++) {
@@ -419,7 +491,8 @@ static void b43_nphy_stay_in_carrier_search(struct b43_wldev *dev, bool enable)
 		static const u16 clip[] = { 0xFFFF, 0xFFFF };
 		if (nphy->deaf_count++ == 0) {
 			nphy->classifier_state = b43_nphy_classifier(dev, 0, 0);
-			b43_nphy_classifier(dev, 0x7, 0);
+			b43_nphy_classifier(dev, 0x7,
+					    B43_NPHY_CLASSCTL_WAITEDEN);
 			b43_nphy_read_clip_detection(dev, nphy->clip_state);
 			b43_nphy_write_clip_detection(dev, clip);
 		}
@@ -627,13 +700,11 @@ static void b43_radio_2057_init_post(struct b43_wldev *dev)
 	b43_radio_mask(dev, R2057_RFPLL_MISC_CAL_RESETN, ~0x78);
 	b43_radio_mask(dev, R2057_XTAL_CONFIG2, ~0x80);
 
-	if (dev->phy.n->init_por) {
+	if (dev->phy.do_full_init) {
 		b43_radio_2057_rcal(dev);
 		b43_radio_2057_rccal(dev);
 	}
 	b43_radio_mask(dev, R2057_RFPLL_MASTER, ~0x8);
-
-	dev->phy.n->init_por = false;
 }
 
 /* http://bcm-v4.sipsolutions.net/802.11/Radio/2057/Init */
@@ -734,9 +805,16 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 	u16 bias, cbias;
 	u16 pag_boost, padg_boost, pgag_boost, mixg_boost;
 	u16 paa_boost, pada_boost, pgaa_boost, mixa_boost;
+	bool is_pkg_fab_smic;
 
 	B43_WARN_ON(dev->phy.rev < 3);
 
+	is_pkg_fab_smic =
+		((dev->dev->chip_id == BCMA_CHIP_ID_BCM43224 ||
+		  dev->dev->chip_id == BCMA_CHIP_ID_BCM43225 ||
+		  dev->dev->chip_id == BCMA_CHIP_ID_BCM43421) &&
+		 dev->dev->chip_pkg == BCMA_PKG_ID_BCM43224_FAB_SMIC);
+
 	b43_chantab_radio_2056_upload(dev, e);
 	b2056_upload_syn_pll_cp2(dev, band == IEEE80211_BAND_5GHZ);
 
@@ -744,7 +822,8 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 	    b43_current_band(dev->wl) == IEEE80211_BAND_2GHZ) {
 		b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER1, 0x1F);
 		b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER2, 0x1F);
-		if (dev->dev->chip_id == 0x4716) {
+		if (dev->dev->chip_id == BCMA_CHIP_ID_BCM4716 ||
+		    dev->dev->chip_id == BCMA_CHIP_ID_BCM47162) {
 			b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER4, 0x14);
 			b43_radio_write(dev, B2056_SYN_PLL_CP2, 0);
 		} else {
@@ -752,6 +831,13 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 			b43_radio_write(dev, B2056_SYN_PLL_CP2, 0x14);
 		}
 	}
+	if (sprom->boardflags2_hi & B43_BFH2_GPLL_WAR2 &&
+	    b43_current_band(dev->wl) == IEEE80211_BAND_2GHZ) {
+		b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER1, 0x1f);
+		b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER2, 0x1f);
+		b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER4, 0x0b);
+		b43_radio_write(dev, B2056_SYN_PLL_CP2, 0x20);
+	}
 	if (sprom->boardflags2_lo & B43_BFL2_APLL_WAR &&
 	    b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ) {
 		b43_radio_write(dev, B2056_SYN_PLL_LOOPFILTER1, 0x1F);
@@ -767,7 +853,8 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 				b43_radio_write(dev,
 					offset | B2056_TX_PADG_IDAC, 0xcc);
 
-				if (dev->dev->chip_id == 0x4716) {
+				if (dev->dev->chip_id == BCMA_CHIP_ID_BCM4716 ||
+				    dev->dev->chip_id == BCMA_CHIP_ID_BCM47162) {
 					bias = 0x40;
 					cbias = 0x45;
 					pag_boost = 0x5;
@@ -776,6 +863,10 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 				} else {
 					bias = 0x25;
 					cbias = 0x20;
+					if (is_pkg_fab_smic) {
+						bias = 0x2a;
+						cbias = 0x38;
+					}
 					pag_boost = 0x4;
 					pgag_boost = 0x03;
 					mixg_boost = 0x65;
@@ -844,6 +935,8 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 			mixa_boost = 0xF;
 		}
 
+		cbias = is_pkg_fab_smic ? 0x35 : 0x30;
+
 		for (i = 0; i < 2; i++) {
 			offset = i ? B2056_TX1 : B2056_TX0;
 
@@ -862,11 +955,11 @@ static void b43_radio_2056_setup(struct b43_wldev *dev,
 			b43_radio_write(dev,
 				offset | B2056_TX_PADA_CASCBIAS, 0x03);
 			b43_radio_write(dev,
-				offset | B2056_TX_INTPAA_IAUX_STAT, 0x50);
+				offset | B2056_TX_INTPAA_IAUX_STAT, 0x30);
 			b43_radio_write(dev,
-				offset | B2056_TX_INTPAA_IMAIN_STAT, 0x50);
+				offset | B2056_TX_INTPAA_IMAIN_STAT, 0x30);
 			b43_radio_write(dev,
-				offset | B2056_TX_INTPAA_CASCBIAS, 0x30);
+				offset | B2056_TX_INTPAA_CASCBIAS, cbias);
 		}
 	}
 
@@ -933,7 +1026,7 @@ static void b43_radio_init2056_post(struct b43_wldev *dev)
 	b43_radio_mask(dev, B2056_SYN_COM_RESET, ~0x2);
 	b43_radio_mask(dev, B2056_SYN_PLL_MAST2, ~0xFC);
 	b43_radio_mask(dev, B2056_SYN_RCCAL_CTRL0, ~0x1);
-	if (dev->phy.n->init_por)
+	if (dev->phy.do_full_init)
 		b43_radio_2056_rcal(dev);
 }
 
@@ -946,8 +1039,6 @@ static void b43_radio_init2056(struct b43_wldev *dev)
 	b43_radio_init2056_pre(dev);
 	b2056_upload_inittabs(dev, 0, 0);
 	b43_radio_init2056_post(dev);
-
-	dev->phy.n->init_por = false;
 }
 
 /**************************************************
@@ -1164,23 +1255,20 @@ static void b43_nphy_run_samples(struct b43_wldev *dev, u16 samps, u16 loops,
 	u16 seq_mode;
 	u32 tmp;
 
-	if (nphy->hang_avoid)
-		b43_nphy_stay_in_carrier_search(dev, true);
+	b43_nphy_stay_in_carrier_search(dev, true);
 
 	if ((nphy->bb_mult_save & 0x80000000) == 0) {
 		tmp = b43_ntab_read(dev, B43_NTAB16(15, 87));
 		nphy->bb_mult_save = (tmp & 0xFFFF) | 0x80000000;
 	}
 
+	/* TODO: add modify_bbmult argument */
 	if (!dev->phy.is_40mhz)
 		tmp = 0x6464;
 	else
 		tmp = 0x4747;
 	b43_ntab_write(dev, B43_NTAB16(15, 87), tmp);
 
-	if (nphy->hang_avoid)
-		b43_nphy_stay_in_carrier_search(dev, false);
-
 	b43_phy_write(dev, B43_NPHY_SAMP_DEPCNT, (samps - 1));
 
 	if (loops != 0xFFFF)
@@ -1213,6 +1301,8 @@ static void b43_nphy_run_samples(struct b43_wldev *dev, u16 samps, u16 loops,
 		b43err(dev->wl, "run samples timeout\n");
 
 	b43_phy_write(dev, B43_NPHY_RFSEQMODE, seq_mode);
+
+	b43_nphy_stay_in_carrier_search(dev, false);
 }
 
 /**************************************************
@@ -1588,8 +1678,8 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 	struct b43_phy_n *nphy = dev->phy.n;
 
 	u16 saved_regs_phy_rfctl[2];
-	u16 saved_regs_phy[13];
-	u16 regs_to_store[] = {
+	u16 saved_regs_phy[22];
+	u16 regs_to_store_rev3[] = {
 		B43_NPHY_AFECTL_OVER1, B43_NPHY_AFECTL_OVER,
 		B43_NPHY_AFECTL_C1, B43_NPHY_AFECTL_C2,
 		B43_NPHY_TXF_40CO_B1S1, B43_NPHY_RFCTL_OVER,
@@ -1598,6 +1688,20 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 		B43_NPHY_RFCTL_LUT_TRSW_UP1, B43_NPHY_RFCTL_LUT_TRSW_UP2,
 		B43_NPHY_RFCTL_RSSIO1, B43_NPHY_RFCTL_RSSIO2
 	};
+	u16 regs_to_store_rev7[] = {
+		B43_NPHY_AFECTL_OVER1, B43_NPHY_AFECTL_OVER,
+		B43_NPHY_AFECTL_C1, B43_NPHY_AFECTL_C2,
+		B43_NPHY_TXF_40CO_B1S1, B43_NPHY_RFCTL_OVER,
+		0x342, 0x343, 0x346, 0x347,
+		0x2ff,
+		B43_NPHY_TXF_40CO_B1S0, B43_NPHY_TXF_40CO_B32S1,
+		B43_NPHY_RFCTL_CMD,
+		B43_NPHY_RFCTL_LUT_TRSW_UP1, B43_NPHY_RFCTL_LUT_TRSW_UP2,
+		0x340, 0x341, 0x344, 0x345,
+		B43_NPHY_RFCTL_RSSIO1, B43_NPHY_RFCTL_RSSIO2
+	};
+	u16 *regs_to_store;
+	int regs_amount;
 
 	u16 class;
 
@@ -1617,6 +1721,15 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 	u8 rx_core_state;
 	int core, i, j, vcm;
 
+	if (dev->phy.rev >= 7) {
+		regs_to_store = regs_to_store_rev7;
+		regs_amount = ARRAY_SIZE(regs_to_store_rev7);
+	} else {
+		regs_to_store = regs_to_store_rev3;
+		regs_amount = ARRAY_SIZE(regs_to_store_rev3);
+	}
+	BUG_ON(regs_amount > ARRAY_SIZE(saved_regs_phy));
+
 	class = b43_nphy_classifier(dev, 0, 0);
 	b43_nphy_classifier(dev, 7, 4);
 	b43_nphy_read_clip_detection(dev, clip_state);
@@ -1624,22 +1737,29 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 
 	saved_regs_phy_rfctl[0] = b43_phy_read(dev, B43_NPHY_RFCTL_INTC1);
 	saved_regs_phy_rfctl[1] = b43_phy_read(dev, B43_NPHY_RFCTL_INTC2);
-	for (i = 0; i < ARRAY_SIZE(regs_to_store); i++)
+	for (i = 0; i < regs_amount; i++)
 		saved_regs_phy[i] = b43_phy_read(dev, regs_to_store[i]);
 
 	b43_nphy_rf_ctl_intc_override(dev, N_INTC_OVERRIDE_OFF, 0, 7);
 	b43_nphy_rf_ctl_intc_override(dev, N_INTC_OVERRIDE_TRSW, 1, 7);
-	b43_nphy_rf_ctl_override(dev, 0x1, 0, 0, false);
-	b43_nphy_rf_ctl_override(dev, 0x2, 1, 0, false);
-	b43_nphy_rf_ctl_override(dev, 0x80, 1, 0, false);
-	b43_nphy_rf_ctl_override(dev, 0x40, 1, 0, false);
-
-	if (b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ) {
-		b43_nphy_rf_ctl_override(dev, 0x20, 0, 0, false);
-		b43_nphy_rf_ctl_override(dev, 0x10, 1, 0, false);
+
+	if (dev->phy.rev >= 7) {
+		/* TODO */
+		if (b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ) {
+		} else {
+		}
 	} else {
-		b43_nphy_rf_ctl_override(dev, 0x10, 0, 0, false);
-		b43_nphy_rf_ctl_override(dev, 0x20, 1, 0, false);
+		b43_nphy_rf_ctl_override(dev, 0x1, 0, 0, false);
+		b43_nphy_rf_ctl_override(dev, 0x2, 1, 0, false);
+		b43_nphy_rf_ctl_override(dev, 0x80, 1, 0, false);
+		b43_nphy_rf_ctl_override(dev, 0x40, 1, 0, false);
+		if (b43_current_band(dev->wl) == IEEE80211_BAND_5GHZ) {
+			b43_nphy_rf_ctl_override(dev, 0x20, 0, 0, false);
+			b43_nphy_rf_ctl_override(dev, 0x10, 1, 0, false);
+		} else {
+			b43_nphy_rf_ctl_override(dev, 0x10, 0, 0, false);
+			b43_nphy_rf_ctl_override(dev, 0x20, 1, 0, false);
+		}
 	}
 
 	rx_core_state = b43_nphy_get_rx_core_state(dev);
@@ -1654,8 +1774,11 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 
 		/* Grab RSSI results for every possible VCM */
 		for (vcm = 0; vcm < 8; vcm++) {
-			b43_radio_maskset(dev, r | B2056_RX_RSSI_MISC, 0xE3,
-					vcm << 2);
+			if (dev->phy.rev >= 7)
+				;
+			else
+				b43_radio_maskset(dev, r | B2056_RX_RSSI_MISC,
+						  0xE3, vcm << 2);
 			b43_nphy_poll_rssi(dev, N_RSSI_NB, results[vcm], 8);
 		}
 
@@ -1682,8 +1805,11 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 		}
 
 		/* Select the best VCM */
-		b43_radio_maskset(dev, r | B2056_RX_RSSI_MISC, 0xE3,
-				  vcm_final << 2);
+		if (dev->phy.rev >= 7)
+			;
+		else
+			b43_radio_maskset(dev, r | B2056_RX_RSSI_MISC,
+					  0xE3, vcm_final << 2);
 
 		for (i = 0; i < 4; i++) {
 			if (core != i / 2)
@@ -1736,9 +1862,9 @@ static void b43_nphy_rev3_rssi_cal(struct b43_wldev *dev)
 
 	b43_phy_set(dev, B43_NPHY_RFCTL_OVER, 0x1);
 	b43_phy_set(dev, B43_NPHY_RFCTL_CMD, B43_NPHY_RFCTL_CMD_RXTX);
-	b43_phy_mask(dev, B43_NPHY_TXF_40CO_B1S1, ~0x1);
+	b43_phy_mask(dev, B43_NPHY_RFCTL_OVER, ~0x1);
 
-	for (i = 0; i < ARRAY_SIZE(regs_to_store); i++)
+	for (i = 0; i < regs_amount; i++)
 		b43_phy_write(dev, regs_to_store[i], saved_regs_phy[i]);
 
 	/* Store for future configuration */
@@ -2494,8 +2620,8 @@ static void b43_nphy_workarounds_rev3plus(struct b43_wldev *dev)
 	struct ssb_sprom *sprom = dev->dev->bus_sprom;
 
 	/* TX to RX */
-	u8 tx2rx_events[8] = { 0x4, 0x3, 0x6, 0x5, 0x2, 0x1, 0x8, 0x1F };
-	u8 tx2rx_delays[8] = { 8, 4, 2, 2, 4, 4, 6, 1 };
+	u8 tx2rx_events[7] = { 0x4, 0x3, 0x5, 0x2, 0x1, 0x8, 0x1F };
+	u8 tx2rx_delays[7] = { 8, 4, 4, 4, 4, 6, 1 };
 	/* RX to TX */
 	u8 rx2tx_events_ipa[9] = { 0x0, 0x1, 0x2, 0x8, 0x5, 0x6, 0xF, 0x3,
 					0x1F };
@@ -2503,6 +2629,23 @@ static void b43_nphy_workarounds_rev3plus(struct b43_wldev *dev)
 	u8 rx2tx_events[9] = { 0x0, 0x1, 0x2, 0x8, 0x5, 0x6, 0x3, 0x4, 0x1F };
 	u8 rx2tx_delays[9] = { 8, 6, 6, 4, 4, 18, 42, 1, 1 };
 
+	u16 vmids[5][4] = {
+		{ 0xa2, 0xb4, 0xb4, 0x89, }, /* 0 */
+		{ 0xb4, 0xb4, 0xb4, 0x24, }, /* 1 */
+		{ 0xa2, 0xb4, 0xb4, 0x74, }, /* 2 */
+		{ 0xa2, 0xb4, 0xb4, 0x270, }, /* 3 */
+		{ 0xa2, 0xb4, 0xb4, 0x00, }, /* 4 and 5 */
+	};
+	u16 gains[5][4] = {
+		{ 0x02, 0x02, 0x02, 0x00, }, /* 0 */
+		{ 0x02, 0x02, 0x02, 0x02, }, /* 1 */
+		{ 0x02, 0x02, 0x02, 0x04, }, /* 2 */
+		{ 0x02, 0x02, 0x02, 0x00, }, /* 3 */
+		{ 0x02, 0x02, 0x02, 0x00, }, /* 4 and 5 */
+	};
+	u16 *vmid, *gain;
+
+	u8 pdet_range;
 	u16 tmp16;
 	u32 tmp32;
 
@@ -2561,7 +2704,71 @@ static void b43_nphy_workarounds_rev3plus(struct b43_wldev *dev)
 	b43_ntab_write(dev, B43_NTAB16(8, 0), 2);
 	b43_ntab_write(dev, B43_NTAB16(8, 16), 2);
 
-	/* TODO */
+	if (b43_current_band(dev->wl) == IEEE80211_BAND_2GHZ)
+		pdet_range = sprom->fem.ghz2.pdet_range;
+	else
+		pdet_range = sprom->fem.ghz5.pdet_range;
+	vmid = vmids[min_t(u16, pdet_range, 4)];
+	gain = gains[min_t(u16, pdet_range, 4)];
+	switch (pdet_range) {
+	case 3:
+		if (!(dev->phy.rev >= 4 &&
+		      b43_current_band(dev->wl) == IEEE80211_BAND_2GHZ))
+			break;
+		/* FALL THROUGH */
+	case 0:
+	case 1:
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x08), 4, vmid);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x18), 4, vmid);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x0c), 4, gain);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x1c), 4, gain);
+		break;
+	case 2:
+		if (dev->phy.rev >= 6) {
+			if (b43_current_band(dev->wl) == IEEE80211_BAND_2GHZ)
+				vmid[3] = 0x94;
+			else
+				vmid[3] = 0x8e;
+			gain[3] = 3;
+		} else if (dev->phy.rev == 5) {
+			vmid[3] = 0x84;
+			gain[3] = 2;
+		}
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x08), 4, vmid);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x18), 4, vmid);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x0c), 4, gain);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x1c), 4, gain);
+		break;
+	case 4:
+	case 5:
+		if (b43_current_band(dev->wl) != IEEE80211_BAND_2GHZ) {
+			if (pdet_range == 4) {
+				vmid[3] = 0x8e;
+				tmp16 = 0x96;
+				gain[3] = 0x2;
+			} else {
+				vmid[3] = 0x89;
+				tmp16 = 0x89;
+				gain[3] = 0;
+			}
+		} else {
+			if (pdet_range == 4) {
+				vmid[3] = 0x89;
+				tmp16 = 0x8b;
+				gain[3] = 0x2;
+			} else {
+				vmid[3] = 0x74;
+				tmp16 = 0x70;
+				gain[3] = 0;
+			}
+		}
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x08), 4, vmid);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x0c), 4, gain);
+		vmid[3] = tmp16;
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x18), 4, vmid);
+		b43_ntab_write_bulk(dev, B43_NTAB16(8, 0x1c), 4, gain);
+		break;
+	}
 
 	b43_radio_write(dev, B2056_RX0 | B2056_RX_MIXA_MAST_BIAS, 0x00);
 	b43_radio_write(dev, B2056_RX1 | B2056_RX_MIXA_MAST_BIAS, 0x00);
@@ -2600,7 +2807,7 @@ static void b43_nphy_workarounds_rev3plus(struct b43_wldev *dev)
 	/* Dropped probably-always-true condition */
 	b43_phy_write(dev, B43_NPHY_ED_CRS40ASSERTTHRESH0, 0x03eb);
 	b43_phy_write(dev, B43_NPHY_ED_CRS40ASSERTTHRESH1, 0x03eb);
-	b43_phy_write(dev, B43_NPHY_ED_CRS40DEASSERTTHRESH1, 0x0341);
+	b43_phy_write(dev, B43_NPHY_ED_CRS40DEASSERTTHRESH0, 0x0341);
 	b43_phy_write(dev, B43_NPHY_ED_CRS40DEASSERTTHRESH1, 0x0341);
 	b43_phy_write(dev, B43_NPHY_ED_CRS20LASSERTTHRESH0, 0x042b);
 	b43_phy_write(dev, B43_NPHY_ED_CRS20LASSERTTHRESH1, 0x042b);
@@ -3211,6 +3418,20 @@ static void b43_nphy_tx_prepare_adjusted_power_table(struct b43_wldev *dev)
 	u8 idx, delta;
 	u8 i, stf_mode;
 
+	/* Array adj_pwr_tbl corresponds to the hardware table. It consists of
+	 * 21 groups, each containing 4 entries.
+	 *
+	 * First group has entries for CCK modulation.
+	 * The rest of groups has 1 entry per modulation (SISO, CDD, STBC, SDM).
+	 *
+	 * Group 0 is for CCK
+	 * Groups 1..4 use BPSK (group per coding rate)
+	 * Groups 5..8 use QPSK (group per coding rate)
+	 * Groups 9..12 use 16-QAM (group per coding rate)
+	 * Groups 13..16 use 64-QAM (group per coding rate)
+	 * Groups 17..20 are unknown
+	 */
+
 	for (i = 0; i < 4; i++)
 		nphy->adj_pwr_tbl[i] = nphy->tx_power_offset[i];
 
@@ -3409,10 +3630,8 @@ static void b43_nphy_tx_power_ctl_setup(struct b43_wldev *dev)
 	}
 
 	b43_nphy_tx_prepare_adjusted_power_table(dev);
-	/*
 	b43_ntab_write_bulk(dev, B43_NTAB16(26, 64), 84, nphy->adj_pwr_tbl);
 	b43_ntab_write_bulk(dev, B43_NTAB16(27, 64), 84, nphy->adj_pwr_tbl);
-	*/
 
 	if (nphy->hang_avoid)
 		b43_nphy_stay_in_carrier_search(dev, false);
@@ -5124,7 +5343,7 @@ static int b43_phy_initn(struct b43_wldev *dev)
 	b43_phy_write(dev, B43_NPHY_TXMACIF_HOLDOFF, 0x0015);
 	b43_phy_write(dev, B43_NPHY_TXMACDELAY, 0x0320);
 	if (phy->rev >= 3 && phy->rev <= 6)
-		b43_phy_write(dev, B43_NPHY_PLOAD_CSENSE_EXTLEN, 0x0014);
+		b43_phy_write(dev, B43_NPHY_PLOAD_CSENSE_EXTLEN, 0x0032);
 	b43_nphy_tx_lp_fbw(dev);
 	if (phy->rev >= 3)
 		b43_nphy_spur_workaround(dev);
@@ -5338,7 +5557,6 @@ static void b43_nphy_op_prepare_structs(struct b43_wldev *dev)
 	nphy->hang_avoid = (phy->rev == 3 || phy->rev == 4);
 	nphy->spur_avoid = (phy->rev >= 3) ?
 				B43_SPUR_AVOID_AUTO : B43_SPUR_AVOID_DISABLE;
-	nphy->init_por = true;
 	nphy->gain_boost = true; /* this way we follow wl, assume it is true */
 	nphy->txrx_chain = 2; /* sth different than 0 and 1 for now */
 	nphy->phyrxchain = 3; /* to avoid b43_nphy_set_rx_core_state like wl */
@@ -5379,8 +5597,6 @@ static void b43_nphy_op_prepare_structs(struct b43_wldev *dev)
 		nphy->ipa2g_on = sprom->fem.ghz2.extpa_gain == 2;
 		nphy->ipa5g_on = sprom->fem.ghz5.extpa_gain == 2;
 	}
-
-	nphy->init_por = true;
 }
 
 static void b43_nphy_op_free(struct b43_wldev *dev)
@@ -5441,8 +5657,11 @@ static u16 b43_nphy_op_radio_read(struct b43_wldev *dev, u16 reg)
 {
 	/* Register 1 is a 32-bit register. */
 	B43_WARN_ON(reg == 1);
-	/* N-PHY needs 0x100 for read access */
-	reg |= 0x100;
+
+	if (dev->phy.rev >= 7)
+		reg |= 0x200; /* Radio 0x2057 */
+	else
+		reg |= 0x100;
 
 	b43_write16(dev, B43_MMIO_RADIO_CONTROL, reg);
 	return b43_read16(dev, B43_MMIO_RADIO_DATA_LOW);
@@ -5488,10 +5707,12 @@ static void b43_nphy_op_software_rfkill(struct b43_wldev *dev,
 		}
 	} else {
 		if (dev->phy.rev >= 7) {
-			b43_radio_2057_init(dev);
+			if (!dev->phy.radio_on)
+				b43_radio_2057_init(dev);
 			b43_switch_channel(dev, dev->phy.channel);
 		} else if (dev->phy.rev >= 3) {
-			b43_radio_init2056(dev);
+			if (!dev->phy.radio_on)
+				b43_radio_init2056(dev);
 			b43_switch_channel(dev, dev->phy.channel);
 		} else {
 			b43_radio_init2055(dev);
diff --git a/drivers/net/wireless/b43/phy_n.h b/drivers/net/wireless/b43/phy_n.h
index 9a5b6bc27d24..ecfbf66dbc3b 100644
--- a/drivers/net/wireless/b43/phy_n.h
+++ b/drivers/net/wireless/b43/phy_n.h
@@ -931,7 +931,6 @@ struct b43_phy_n {
 	u16 papd_epsilon_offset[2];
 	s32 preamble_override;
 	u32 bb_mult_save;
-	bool init_por;
 
 	bool gain_boost;
 	bool elna_gain_config;
diff --git a/drivers/net/wireless/b43/radio_2056.c b/drivers/net/wireless/b43/radio_2056.c
index b4fd9345d673..2ce25607c60d 100644
--- a/drivers/net/wireless/b43/radio_2056.c
+++ b/drivers/net/wireless/b43/radio_2056.c
@@ -48,7 +48,7 @@ struct b2056_inittabs_pts {
 	unsigned int rx_length;
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev3_syn[] = {
+static const struct b2056_inittab_entry b2056_inittab_phy_rev3_syn[] = {
 	[B2056_SYN_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -232,7 +232,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev3_syn[] = {
 	[B2056_SYN_LOGEN_TX_CMOS_VALID]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev3_tx[] = {
+static const struct b2056_inittab_entry b2056_inittab_phy_rev3_tx[] = {
 	[B2056_TX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -380,7 +380,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev3_tx[] = {
 	[B2056_TX_STATUS_TXLPF_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev3_rx[] = {
+static const struct b2056_inittab_entry b2056_inittab_phy_rev3_rx[] = {
 	[B2056_RX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -530,7 +530,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev3_rx[] = {
 	[B2056_RX_STATUS_HPC_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev4_syn[] = {
+static const struct b2056_inittab_entry b2056_inittab_phy_rev4_syn[] = {
 	[B2056_SYN_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -714,7 +714,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev4_syn[] = {
 	[B2056_SYN_LOGEN_TX_CMOS_VALID]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev4_tx[] = {
+static const struct b2056_inittab_entry b2056_inittab_phy_rev4_tx[] = {
 	[B2056_TX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -862,7 +862,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev4_tx[] = {
 	[B2056_TX_STATUS_TXLPF_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev4_rx[] = {
+static const struct b2056_inittab_entry b2056_inittab_phy_rev4_rx[] = {
 	[B2056_RX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1012,7 +1012,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev4_rx[] = {
 	[B2056_RX_STATUS_HPC_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev5_syn[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev5_syn[] = {
 	[B2056_SYN_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1196,7 +1196,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev5_syn[] = {
 	[B2056_SYN_LOGEN_TX_CMOS_VALID]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev5_tx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev5_tx[] = {
 	[B2056_TX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1352,7 +1352,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev5_tx[] = {
 	[B2056_TX_GMBB_IDAC7]		= { .ghz5 = 0x0075, .ghz2 = 0x0075, UPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev5_rx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev5_rx[] = {
 	[B2056_RX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1502,7 +1502,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev5_rx[] = {
 	[B2056_RX_STATUS_HPC_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev6_syn[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev6_syn[] = {
 	[B2056_SYN_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1686,7 +1686,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev6_syn[] = {
 	[B2056_SYN_LOGEN_TX_CMOS_VALID]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev6_tx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev6_tx[] = {
 	[B2056_TX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1842,7 +1842,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev6_tx[] = {
 	[B2056_TX_GMBB_IDAC7]		= { .ghz5 = 0x0070, .ghz2 = 0x0070, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev6_rx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev6_rx[] = {
 	[B2056_RX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -1992,7 +1992,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev6_rx[] = {
 	[B2056_RX_STATUS_HPC_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev7_syn[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev7_9_syn[] = {
 	[B2056_SYN_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -2176,7 +2176,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev7_syn[] = {
 	[B2056_SYN_LOGEN_TX_CMOS_VALID]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev7_tx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev7_9_tx[] = {
 	[B2056_TX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -2332,7 +2332,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev7_tx[] = {
 	[B2056_TX_GMBB_IDAC7]		= { .ghz5 = 0x0075, .ghz2 = 0x0075, UPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev7_rx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev7_9_rx[] = {
 	[B2056_RX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -2482,7 +2482,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev7_rx[] = {
 	[B2056_RX_STATUS_HPC_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev8_syn[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev8_syn[] = {
 	[B2056_SYN_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_SYN_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -2666,7 +2666,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev8_syn[] = {
 	[B2056_SYN_LOGEN_TX_CMOS_VALID]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev8_tx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev8_tx[] = {
 	[B2056_TX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_TX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -2822,7 +2822,7 @@ static const struct b2056_inittab_entry b2056_inittab_rev8_tx[] = {
 	[B2056_TX_GMBB_IDAC7]		= { .ghz5 = 0x0070, .ghz2 = 0x0070, NOUPLOAD, },
 };
 
-static const struct b2056_inittab_entry b2056_inittab_rev8_rx[] = {
+static const struct b2056_inittab_entry b2056_inittab_radio_rev8_rx[] = {
 	[B2056_RX_RESERVED_ADDR2]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR3]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 	[B2056_RX_RESERVED_ADDR4]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
@@ -2972,24 +2972,69 @@ static const struct b2056_inittab_entry b2056_inittab_rev8_rx[] = {
 	[B2056_RX_STATUS_HPC_RC]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, NOUPLOAD, },
 };
 
-#define INITTABSPTS(prefix) \
-	.syn		= prefix##_syn,			\
-	.syn_length	= ARRAY_SIZE(prefix##_syn),	\
-	.tx		= prefix##_tx,			\
-	.tx_length	= ARRAY_SIZE(prefix##_tx),	\
-	.rx		= prefix##_rx,			\
-	.rx_length	= ARRAY_SIZE(prefix##_rx)
+static const struct b2056_inittab_entry b2056_inittab_radio_rev11_syn[] = {
+	[B2056_SYN_PLL_PFD]		= { .ghz5 = 0x0006, .ghz2 = 0x0006, UPLOAD, },
+	[B2056_SYN_PLL_CP2]		= { .ghz5 = 0x003f, .ghz2 = 0x003f, UPLOAD, },
+	[B2056_SYN_PLL_LOOPFILTER1]	= { .ghz5 = 0x0006, .ghz2 = 0x0006, UPLOAD, },
+	[B2056_SYN_PLL_LOOPFILTER2]	= { .ghz5 = 0x0006, .ghz2 = 0x0006, UPLOAD, },
+	[B2056_SYN_PLL_LOOPFILTER4]	= { .ghz5 = 0x002b, .ghz2 = 0x002b, UPLOAD, },
+	[B2056_SYN_PLL_VCO2]		= { .ghz5 = 0x00f7, .ghz2 = 0x00f7, UPLOAD, },
+	[B2056_SYN_PLL_VCOCAL12]	= { .ghz5 = 0x0007, .ghz2 = 0x0007, UPLOAD, },
+	[B2056_SYN_LOGENBUF2]		= { .ghz5 = 0x008f, .ghz2 = 0x008f, UPLOAD, },
+};
 
-static const struct b2056_inittabs_pts b2056_inittabs[] = {
-	[3] = { INITTABSPTS(b2056_inittab_rev3) },
-	[4] = { INITTABSPTS(b2056_inittab_rev4) },
-	[5] = { INITTABSPTS(b2056_inittab_rev5) },
-	[6] = { INITTABSPTS(b2056_inittab_rev6) },
-	[7] = { INITTABSPTS(b2056_inittab_rev7) },
-	[8] = { INITTABSPTS(b2056_inittab_rev8) },
-	[9] = { INITTABSPTS(b2056_inittab_rev7) },
+static const struct b2056_inittab_entry b2056_inittab_radio_rev11_tx[] = {
+	[B2056_TX_PA_SPARE2]		= { .ghz5 = 0x00ee, .ghz2 = 0x00ee, UPLOAD, },
+	[B2056_TX_INTPAA_IAUX_STAT]	= { .ghz5 = 0x0050, .ghz2 = 0x0050, UPLOAD, },
+	[B2056_TX_INTPAA_IMAIN_STAT]	= { .ghz5 = 0x0050, .ghz2 = 0x0050, UPLOAD, },
+	[B2056_TX_INTPAA_PASLOPE]	= { .ghz5 = 0x00f0, .ghz2 = 0x00f0, UPLOAD, },
+	[B2056_TX_INTPAG_PASLOPE]	= { .ghz5 = 0x00f0, .ghz2 = 0x00f0, UPLOAD, },
+	[B2056_TX_PADA_IDAC]		= { .ghz5 = 0x00ff, .ghz2 = 0x00ff, UPLOAD, },
+	[B2056_TX_PADA_SLOPE]		= { .ghz5 = 0x0070, .ghz2 = 0x0070, UPLOAD, },
+	[B2056_TX_PADG_SLOPE]		= { .ghz5 = 0x0070, .ghz2 = 0x0070, UPLOAD, },
+	[B2056_TX_PGAA_IDAC]		= { .ghz5 = 0x00ff, .ghz2 = 0x00ff, UPLOAD, },
+	[B2056_TX_PGAA_SLOPE]		= { .ghz5 = 0x0077, .ghz2 = 0x0077, UPLOAD, },
+	[B2056_TX_PGAG_SLOPE]		= { .ghz5 = 0x0077, .ghz2 = 0x0077, UPLOAD, },
+	[B2056_TX_GMBB_IDAC]		= { .ghz5 = 0x0000, .ghz2 = 0x0000, UPLOAD, },
+	[B2056_TX_TXSPARE1]		= { .ghz5 = 0x0030, .ghz2 = 0x0030, UPLOAD, },
+};
+
+static const struct b2056_inittab_entry b2056_inittab_radio_rev11_rx[] = {
+	[B2056_RX_BIASPOLE_LNAA1_IDAC]	= { .ghz5 = 0x0017, .ghz2 = 0x0017, UPLOAD, },
+	[B2056_RX_LNAA2_IDAC]		= { .ghz5 = 0x00ff, .ghz2 = 0x00ff, UPLOAD, },
+	[B2056_RX_BIASPOLE_LNAG1_IDAC]	= { .ghz5 = 0x0017, .ghz2 = 0x0017, UPLOAD, },
+	[B2056_RX_LNAG2_IDAC]		= { .ghz5 = 0x00f0, .ghz2 = 0x00f0, UPLOAD, },
+	[B2056_RX_MIXA_VCM]		= { .ghz5 = 0x0055, .ghz2 = 0x0055, UPLOAD, },
+	[B2056_RX_MIXA_LOB_BIAS]	= { .ghz5 = 0x0088, .ghz2 = 0x0088, UPLOAD, },
+	[B2056_RX_MIXA_BIAS_AUX]	= { .ghz5 = 0x0007, .ghz2 = 0x0007, UPLOAD, },
+	[B2056_RX_MIXG_VCM]		= { .ghz5 = 0x0055, .ghz2 = 0x0055, UPLOAD, },
+	[B2056_RX_TIA_IOPAMP]		= { .ghz5 = 0x0026, .ghz2 = 0x0026, UPLOAD, },
+	[B2056_RX_TIA_QOPAMP]		= { .ghz5 = 0x0026, .ghz2 = 0x0026, UPLOAD, },
+	[B2056_RX_TIA_IMISC]		= { .ghz5 = 0x000f, .ghz2 = 0x000f, UPLOAD, },
+	[B2056_RX_TIA_QMISC]		= { .ghz5 = 0x000f, .ghz2 = 0x000f, UPLOAD, },
+	[B2056_RX_RXLPF_OUTVCM]		= { .ghz5 = 0x0004, .ghz2 = 0x0004, UPLOAD, },
+	[B2056_RX_VGA_BIAS_DCCANCEL]	= { .ghz5 = 0x0000, .ghz2 = 0x0000, UPLOAD, },
+	[B2056_RX_RXSPARE3]		= { .ghz5 = 0x0005, .ghz2 = 0x0005, UPLOAD, },
 };
 
+#define INITTABSPTS(prefix) \
+	static const struct b2056_inittabs_pts prefix = {	\
+		.syn		= prefix##_syn,			\
+		.syn_length	= ARRAY_SIZE(prefix##_syn),	\
+		.tx		= prefix##_tx,			\
+		.tx_length	= ARRAY_SIZE(prefix##_tx),	\
+		.rx		= prefix##_rx,			\
+		.rx_length	= ARRAY_SIZE(prefix##_rx),	\
+	}
+
+INITTABSPTS(b2056_inittab_phy_rev3);
+INITTABSPTS(b2056_inittab_phy_rev4);
+INITTABSPTS(b2056_inittab_radio_rev5);
+INITTABSPTS(b2056_inittab_radio_rev6);
+INITTABSPTS(b2056_inittab_radio_rev7_9);
+INITTABSPTS(b2056_inittab_radio_rev8);
+INITTABSPTS(b2056_inittab_radio_rev11);
+
 #define RADIOREGS3(r00, r01, r02, r03, r04, r05, r06, r07, r08, r09, \
 		   r10, r11, r12, r13, r14, r15, r16, r17, r18, r19, \
 		   r20, r21, r22, r23, r24, r25, r26, r27, r28, r29, \
@@ -3041,7 +3086,7 @@ static const struct b2056_inittabs_pts b2056_inittabs[] = {
 	.phy_regs.phy_bw6	= r5
 
 /* http://bcm-v4.sipsolutions.net/802.11/Radio/2056/ChannelTable */
-static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev3[] = {
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_phy_rev3[] = {
   {	.freq			= 4920,
 	RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x04,
 		   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
@@ -4036,7 +4081,7 @@ static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev3[] =
   },
 };
 
-static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev4[] = {
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_phy_rev4[] = {
   {	.freq			= 4920,
 	RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x04,
 		   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
@@ -5031,7 +5076,7 @@ static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev4[] =
   },
 };
 
-static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev5[] = {
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_radio_rev5[] = {
   {	.freq			= 4920,
 	RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x04,
 		   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
@@ -6026,7 +6071,7 @@ static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev5[] =
   },
 };
 
-static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev6[] = {
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_radio_rev6[] = {
   {	.freq			= 4920,
 	RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x04,
 		   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
@@ -7021,7 +7066,7 @@ static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev6[] =
   },
 };
 
-static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev7_9[] = {
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_radio_rev7_9[] = {
   {	.freq			= 4920,
 	RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x04,
 		   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
@@ -8016,7 +8061,7 @@ static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev7_9[]
   },
 };
 
-static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev8[] = {
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_radio_rev8[] = {
   {	.freq			= 4920,
 	RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x04,
 		   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
@@ -9011,6 +9056,1154 @@ static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_rev8[] =
   },
 };
 
+static const struct b43_nphy_channeltab_entry_rev3 b43_nphy_channeltab_radio_rev11[] = {
+	{
+		.freq			= 4920,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xec, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07b4, 0x07b0, 0x07ac, 0x0214, 0x0215, 0x0216),
+	},
+	{
+		.freq			= 4930,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xed, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07b8, 0x07b4, 0x07b0, 0x0213, 0x0214, 0x0215),
+	},
+	{
+		.freq			= 4940,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xee, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07bc, 0x07b8, 0x07b4, 0x0212, 0x0213, 0x0214),
+	},
+	{
+		.freq			= 4950,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xef, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x00, 0x00, 0x00, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07c0, 0x07bc, 0x07b8, 0x0211, 0x0212, 0x0213),
+	},
+	{
+		.freq			= 4960,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xf0, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07c4, 0x07c0, 0x07bc, 0x020f, 0x0211, 0x0212),
+	},
+	{
+		.freq			= 4970,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xf1, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07c8, 0x07c4, 0x07c0, 0x020e, 0x020f, 0x0211),
+	},
+	{
+		.freq			= 4980,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xf2, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07cc, 0x07c8, 0x07c4, 0x020d, 0x020e, 0x020f),
+	},
+	{
+		.freq			= 4990,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xf3, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07d0, 0x07cc, 0x07c8, 0x020c, 0x020d, 0x020e),
+	},
+	{
+		.freq			= 5000,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xf4, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07d4, 0x07d0, 0x07cc, 0x020b, 0x020c, 0x020d),
+	},
+	{
+		.freq			= 5010,
+		RADIOREGS3(0xff, 0x01, 0x01, 0x01, 0xf5, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07d8, 0x07d4, 0x07d0, 0x020a, 0x020b, 0x020c),
+	},
+	{
+		.freq			= 5020,
+		RADIOREGS3(0xf7, 0x01, 0x01, 0x01, 0xf6, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07dc, 0x07d8, 0x07d4, 0x0209, 0x020a, 0x020b),
+	},
+	{
+		.freq			= 5030,
+		RADIOREGS3(0xf7, 0x01, 0x01, 0x01, 0xf7, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07e0, 0x07dc, 0x07d8, 0x0208, 0x0209, 0x020a),
+	},
+	{
+		.freq			= 5040,
+		RADIOREGS3(0xef, 0x01, 0x01, 0x01, 0xf8, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07e4, 0x07e0, 0x07dc, 0x0207, 0x0208, 0x0209),
+	},
+	{
+		.freq			= 5050,
+		RADIOREGS3(0xef, 0x01, 0x01, 0x01, 0xf9, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07e8, 0x07e4, 0x07e0, 0x0206, 0x0207, 0x0208),
+	},
+	{
+		.freq			= 5060,
+		RADIOREGS3(0xe6, 0x01, 0x01, 0x01, 0xfa, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfe, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfe, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07ec, 0x07e8, 0x07e4, 0x0205, 0x0206, 0x0207),
+	},
+	{
+		.freq			= 5070,
+		RADIOREGS3(0xe6, 0x01, 0x01, 0x01, 0xfb, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfd, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfd, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07f0, 0x07ec, 0x07e8, 0x0204, 0x0205, 0x0206),
+	},
+	{
+		.freq			= 5080,
+		RADIOREGS3(0xde, 0x01, 0x01, 0x01, 0xfc, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfd, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfd, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07f4, 0x07f0, 0x07ec, 0x0203, 0x0204, 0x0205),
+	},
+	{
+		.freq			= 5090,
+		RADIOREGS3(0xde, 0x01, 0x01, 0x01, 0xfd, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x01, 0x01, 0x01, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfd, 0x00, 0x09, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfd, 0x00, 0x09, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07f8, 0x07f4, 0x07f0, 0x0202, 0x0203, 0x0204),
+	},
+	{
+		.freq			= 5100,
+		RADIOREGS3(0xd6, 0x01, 0x01, 0x01, 0xfe, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfd, 0x00, 0x08, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfd, 0x00, 0x08, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x07fc, 0x07f8, 0x07f4, 0x0201, 0x0202, 0x0203),
+	},
+	{
+		.freq			= 5110,
+		RADIOREGS3(0xd6, 0x01, 0x01, 0x01, 0xff, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfc, 0x00, 0x08, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfc, 0x00, 0x08, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0800, 0x07fc, 0x07f8, 0x0200, 0x0201, 0x0202),
+	},
+	{
+		.freq			= 5120,
+		RADIOREGS3(0xce, 0x01, 0x01, 0x02, 0x00, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfc, 0x00, 0x08, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfc, 0x00, 0x08, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0804, 0x0800, 0x07fc, 0x01ff, 0x0200, 0x0201),
+	},
+	{
+		.freq			= 5130,
+		RADIOREGS3(0xce, 0x01, 0x01, 0x02, 0x01, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfc, 0x00, 0x08, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfc, 0x00, 0x08, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0808, 0x0804, 0x0800, 0x01fe, 0x01ff, 0x0200),
+	},
+	{
+		.freq			= 5140,
+		RADIOREGS3(0xc6, 0x01, 0x01, 0x02, 0x02, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfb, 0x00, 0x08, 0x00, 0x77, 0x00, 0x0f,
+			   0x00, 0x6f, 0x00, 0xfb, 0x00, 0x08, 0x00, 0x77,
+			   0x00, 0x0f, 0x00, 0x6f, 0x00),
+		PHYREGS(0x080c, 0x0808, 0x0804, 0x01fd, 0x01fe, 0x01ff),
+	},
+	{
+		.freq			= 5160,
+		RADIOREGS3(0xbe, 0x01, 0x01, 0x02, 0x04, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfa, 0x00, 0x07, 0x00, 0x77, 0x00, 0x0e,
+			   0x00, 0x6f, 0x00, 0xfa, 0x00, 0x07, 0x00, 0x77,
+			   0x00, 0x0e, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0814, 0x0810, 0x080c, 0x01fb, 0x01fc, 0x01fd),
+	},
+	{
+		.freq			= 5170,
+		RADIOREGS3(0xbe, 0x01, 0x01, 0x02, 0x05, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xfa, 0x00, 0x07, 0x00, 0x77, 0x00, 0x0e,
+			   0x00, 0x6f, 0x00, 0xfa, 0x00, 0x07, 0x00, 0x77,
+			   0x00, 0x0e, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0818, 0x0814, 0x0810, 0x01fa, 0x01fb, 0x01fc),
+	},
+	{
+		.freq			= 5180,
+		RADIOREGS3(0xb6, 0x01, 0x01, 0x02, 0x06, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xf9, 0x00, 0x06, 0x00, 0x77, 0x00, 0x0e,
+			   0x00, 0x6f, 0x00, 0xf9, 0x00, 0x06, 0x00, 0x77,
+			   0x00, 0x0e, 0x00, 0x6f, 0x00),
+		PHYREGS(0x081c, 0x0818, 0x0814, 0x01f9, 0x01fa, 0x01fb),
+	},
+	{
+		.freq			= 5190,
+		RADIOREGS3(0xb6, 0x01, 0x01, 0x02, 0x07, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xf9, 0x00, 0x06, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xf9, 0x00, 0x06, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0820, 0x081c, 0x0818, 0x01f8, 0x01f9, 0x01fa),
+	},
+	{
+		.freq			= 5200,
+		RADIOREGS3(0xaf, 0x01, 0x01, 0x02, 0x08, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xf9, 0x00, 0x05, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xf9, 0x00, 0x05, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0824, 0x0820, 0x081c, 0x01f7, 0x01f8, 0x01f9),
+	},
+	{
+		.freq			= 5210,
+		RADIOREGS3(0xaf, 0x01, 0x01, 0x02, 0x09, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8f, 0x0f, 0x00,
+			   0xff, 0xf9, 0x00, 0x05, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xf9, 0x00, 0x05, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0828, 0x0824, 0x0820, 0x01f6, 0x01f7, 0x01f8),
+	},
+	{
+		.freq			= 5220,
+		RADIOREGS3(0xa7, 0x01, 0x01, 0x02, 0x0a, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8e, 0x0f, 0x00,
+			   0xfe, 0xd8, 0x00, 0x05, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xd8, 0x00, 0x05, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x082c, 0x0828, 0x0824, 0x01f5, 0x01f6, 0x01f7),
+	},
+	{
+		.freq			= 5230,
+		RADIOREGS3(0xa7, 0x01, 0x01, 0x02, 0x0b, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8e, 0x0f, 0x00,
+			   0xee, 0xd8, 0x00, 0x05, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xd8, 0x00, 0x05, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0830, 0x082c, 0x0828, 0x01f4, 0x01f5, 0x01f6),
+	},
+	{
+		.freq			= 5240,
+		RADIOREGS3(0xa0, 0x01, 0x01, 0x02, 0x0c, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8e, 0x0f, 0x00,
+			   0xee, 0xc8, 0x00, 0x05, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xc8, 0x00, 0x05, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0834, 0x0830, 0x082c, 0x01f3, 0x01f4, 0x01f5),
+	},
+	{
+		.freq			= 5250,
+		RADIOREGS3(0xa0, 0x01, 0x01, 0x02, 0x0d, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8e, 0x0f, 0x00,
+			   0xed, 0xc7, 0x00, 0x05, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xc7, 0x00, 0x05, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0838, 0x0834, 0x0830, 0x01f2, 0x01f3, 0x01f4),
+	},
+	{
+		.freq			= 5260,
+		RADIOREGS3(0x98, 0x01, 0x01, 0x02, 0x0e, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x02, 0x02, 0x02, 0x8e, 0x0e, 0x00,
+			   0xed, 0xc7, 0x00, 0x04, 0x00, 0x77, 0x00, 0x0d,
+			   0x00, 0x6f, 0x00, 0xc7, 0x00, 0x04, 0x00, 0x77,
+			   0x00, 0x0d, 0x00, 0x6f, 0x00),
+		PHYREGS(0x083c, 0x0838, 0x0834, 0x01f1, 0x01f2, 0x01f3),
+	},
+	{
+		.freq			= 5270,
+		RADIOREGS3(0x98, 0x01, 0x01, 0x02, 0x0f, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8e, 0x0e, 0x00,
+			   0xed, 0xc7, 0x00, 0x04, 0x00, 0x77, 0x00, 0x0c,
+			   0x00, 0x6f, 0x00, 0xc7, 0x00, 0x04, 0x00, 0x77,
+			   0x00, 0x0c, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0840, 0x083c, 0x0838, 0x01f0, 0x01f1, 0x01f2),
+	},
+	{
+		.freq			= 5280,
+		RADIOREGS3(0x91, 0x01, 0x01, 0x02, 0x10, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0e, 0x00,
+			   0xdc, 0xb7, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0c,
+			   0x00, 0x6f, 0x00, 0xb7, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0c, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0844, 0x0840, 0x083c, 0x01f0, 0x01f0, 0x01f1),
+	},
+	{
+		.freq			= 5290,
+		RADIOREGS3(0x91, 0x01, 0x01, 0x02, 0x11, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0e, 0x00,
+			   0xdc, 0xb7, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0c,
+			   0x00, 0x6f, 0x00, 0xb7, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0c, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0848, 0x0844, 0x0840, 0x01ef, 0x01f0, 0x01f0),
+	},
+	{
+		.freq			= 5300,
+		RADIOREGS3(0x8a, 0x01, 0x01, 0x02, 0x12, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0e, 0x00,
+			   0xdc, 0xb7, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0c,
+			   0x00, 0x6f, 0x00, 0xb7, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0c, 0x00, 0x6f, 0x00),
+		PHYREGS(0x084c, 0x0848, 0x0844, 0x01ee, 0x01ef, 0x01f0),
+	},
+	{
+		.freq			= 5310,
+		RADIOREGS3(0x8a, 0x01, 0x01, 0x02, 0x13, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0e, 0x00,
+			   0xdc, 0xb7, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0c,
+			   0x00, 0x6f, 0x00, 0xb7, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0c, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0850, 0x084c, 0x0848, 0x01ed, 0x01ee, 0x01ef),
+	},
+	{
+		.freq			= 5320,
+		RADIOREGS3(0x83, 0x01, 0x01, 0x02, 0x14, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0e, 0x00,
+			   0xdb, 0xb7, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0c,
+			   0x00, 0x6f, 0x00, 0xb7, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0c, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0854, 0x0850, 0x084c, 0x01ec, 0x01ed, 0x01ee),
+	},
+	{
+		.freq			= 5330,
+		RADIOREGS3(0x83, 0x01, 0x01, 0x02, 0x15, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0d, 0x00,
+			   0xcb, 0xa6, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0b,
+			   0x00, 0x6f, 0x00, 0xa6, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0b, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0858, 0x0854, 0x0850, 0x01eb, 0x01ec, 0x01ed),
+	},
+	{
+		.freq			= 5340,
+		RADIOREGS3(0x7c, 0x01, 0x01, 0x02, 0x16, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8d, 0x0d, 0x00,
+			   0xca, 0xa6, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0b,
+			   0x00, 0x6f, 0x00, 0xa6, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0b, 0x00, 0x6f, 0x00),
+		PHYREGS(0x085c, 0x0858, 0x0854, 0x01ea, 0x01eb, 0x01ec),
+	},
+	{
+		.freq			= 5350,
+		RADIOREGS3(0x7c, 0x01, 0x01, 0x02, 0x17, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0d, 0x00,
+			   0xca, 0xa6, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0b,
+			   0x00, 0x6f, 0x00, 0xa6, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0b, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0860, 0x085c, 0x0858, 0x01e9, 0x01ea, 0x01eb),
+	},
+	{
+		.freq			= 5360,
+		RADIOREGS3(0x75, 0x01, 0x01, 0x02, 0x18, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0d, 0x00,
+			   0xc9, 0x95, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x95, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0864, 0x0860, 0x085c, 0x01e8, 0x01e9, 0x01ea),
+	},
+	{
+		.freq			= 5370,
+		RADIOREGS3(0x75, 0x01, 0x01, 0x02, 0x19, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0d, 0x00,
+			   0xc9, 0x95, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x95, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0868, 0x0864, 0x0860, 0x01e7, 0x01e8, 0x01e9),
+	},
+	{
+		.freq			= 5380,
+		RADIOREGS3(0x6e, 0x01, 0x01, 0x02, 0x1a, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0c, 0x00,
+			   0xb8, 0x95, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x95, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x086c, 0x0868, 0x0864, 0x01e6, 0x01e7, 0x01e8),
+	},
+	{
+		.freq			= 5390,
+		RADIOREGS3(0x6e, 0x01, 0x01, 0x02, 0x1b, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0c, 0x00,
+			   0xb8, 0x84, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0870, 0x086c, 0x0868, 0x01e5, 0x01e6, 0x01e7),
+	},
+	{
+		.freq			= 5400,
+		RADIOREGS3(0x67, 0x01, 0x01, 0x02, 0x1c, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0c, 0x00,
+			   0xb8, 0x84, 0x00, 0x03, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x03, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0874, 0x0870, 0x086c, 0x01e5, 0x01e5, 0x01e6),
+	},
+	{
+		.freq			= 5410,
+		RADIOREGS3(0x67, 0x01, 0x01, 0x02, 0x1d, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0c, 0x00,
+			   0xb7, 0x84, 0x00, 0x02, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x02, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0878, 0x0874, 0x0870, 0x01e4, 0x01e5, 0x01e5),
+	},
+	{
+		.freq			= 5420,
+		RADIOREGS3(0x61, 0x01, 0x01, 0x02, 0x1e, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0c, 0x00,
+			   0xa7, 0x84, 0x00, 0x02, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x02, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x087c, 0x0878, 0x0874, 0x01e3, 0x01e4, 0x01e5),
+	},
+	{
+		.freq			= 5430,
+		RADIOREGS3(0x61, 0x01, 0x01, 0x02, 0x1f, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x03, 0x03, 0x03, 0x8c, 0x0b, 0x00,
+			   0xa6, 0x84, 0x00, 0x02, 0x00, 0x77, 0x00, 0x0a,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x02, 0x00, 0x77,
+			   0x00, 0x0a, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0880, 0x087c, 0x0878, 0x01e2, 0x01e3, 0x01e4),
+	},
+	{
+		.freq			= 5440,
+		RADIOREGS3(0x5a, 0x01, 0x01, 0x02, 0x20, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8b, 0x0b, 0x00,
+			   0xa6, 0x84, 0x00, 0x02, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x02, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0884, 0x0880, 0x087c, 0x01e1, 0x01e2, 0x01e3),
+	},
+	{
+		.freq			= 5450,
+		RADIOREGS3(0x5a, 0x01, 0x01, 0x02, 0x21, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8b, 0x0b, 0x00,
+			   0x95, 0x84, 0x00, 0x01, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x01, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0888, 0x0884, 0x0880, 0x01e0, 0x01e1, 0x01e2),
+	},
+	{
+		.freq			= 5460,
+		RADIOREGS3(0x53, 0x01, 0x01, 0x02, 0x22, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8b, 0x0b, 0x00,
+			   0x95, 0x84, 0x00, 0x01, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x84, 0x00, 0x01, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x088c, 0x0888, 0x0884, 0x01df, 0x01e0, 0x01e1),
+	},
+	{
+		.freq			= 5470,
+		RADIOREGS3(0x53, 0x01, 0x01, 0x02, 0x23, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8b, 0x0b, 0x00,
+			   0x94, 0x73, 0x00, 0x01, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x01, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0890, 0x088c, 0x0888, 0x01de, 0x01df, 0x01e0),
+	},
+	{
+		.freq			= 5480,
+		RADIOREGS3(0x4d, 0x01, 0x01, 0x02, 0x24, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x0a, 0x00,
+			   0x84, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0894, 0x0890, 0x088c, 0x01dd, 0x01de, 0x01df),
+	},
+	{
+		.freq			= 5490,
+		RADIOREGS3(0x4d, 0x01, 0x01, 0x02, 0x25, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x0a, 0x00,
+			   0x83, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x0898, 0x0894, 0x0890, 0x01dd, 0x01dd, 0x01de),
+	},
+	{
+		.freq			= 5500,
+		RADIOREGS3(0x47, 0x01, 0x01, 0x02, 0x26, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x0a, 0x00,
+			   0x82, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x089c, 0x0898, 0x0894, 0x01dc, 0x01dd, 0x01dd),
+	},
+	{
+		.freq			= 5510,
+		RADIOREGS3(0x47, 0x01, 0x01, 0x02, 0x27, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x0a, 0x00,
+			   0x82, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08a0, 0x089c, 0x0898, 0x01db, 0x01dc, 0x01dd),
+	},
+	{
+		.freq			= 5520,
+		RADIOREGS3(0x40, 0x01, 0x01, 0x02, 0x28, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x0a, 0x00,
+			   0x72, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08a4, 0x08a0, 0x089c, 0x01da, 0x01db, 0x01dc),
+	},
+	{
+		.freq			= 5530,
+		RADIOREGS3(0x40, 0x01, 0x01, 0x02, 0x29, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x09, 0x00,
+			   0x72, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08a8, 0x08a4, 0x08a0, 0x01d9, 0x01da, 0x01db),
+	},
+	{
+		.freq			= 5540,
+		RADIOREGS3(0x3a, 0x01, 0x01, 0x02, 0x2a, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x8a, 0x09, 0x00,
+			   0x71, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08ac, 0x08a8, 0x08a4, 0x01d8, 0x01d9, 0x01da),
+	},
+	{
+		.freq			= 5550,
+		RADIOREGS3(0x3a, 0x01, 0x01, 0x02, 0x2b, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x89, 0x09, 0x00,
+			   0x61, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08b0, 0x08ac, 0x08a8, 0x01d7, 0x01d8, 0x01d9),
+	},
+	{
+		.freq			= 5560,
+		RADIOREGS3(0x34, 0x01, 0x01, 0x02, 0x2c, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x89, 0x09, 0x00,
+			   0x61, 0x73, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x73, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08b4, 0x08b0, 0x08ac, 0x01d7, 0x01d7, 0x01d8),
+	},
+	{
+		.freq			= 5570,
+		RADIOREGS3(0x34, 0x01, 0x01, 0x02, 0x2d, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x89, 0x09, 0x00,
+			   0x61, 0x62, 0x00, 0x00, 0x00, 0x77, 0x00, 0x09,
+			   0x00, 0x6f, 0x00, 0x62, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x09, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08b8, 0x08b4, 0x08b0, 0x01d6, 0x01d7, 0x01d7),
+	},
+	{
+		.freq			= 5580,
+		RADIOREGS3(0x2e, 0x01, 0x01, 0x02, 0x2e, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x89, 0x08, 0x00,
+			   0x60, 0x62, 0x00, 0x00, 0x00, 0x77, 0x00, 0x08,
+			   0x00, 0x6f, 0x00, 0x62, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x08, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08bc, 0x08b8, 0x08b4, 0x01d5, 0x01d6, 0x01d7),
+	},
+	{
+		.freq			= 5590,
+		RADIOREGS3(0x2e, 0x01, 0x01, 0x02, 0x2f, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x04, 0x04, 0x04, 0x89, 0x08, 0x00,
+			   0x50, 0x61, 0x00, 0x00, 0x00, 0x77, 0x00, 0x08,
+			   0x00, 0x6f, 0x00, 0x61, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x08, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08c0, 0x08bc, 0x08b8, 0x01d4, 0x01d5, 0x01d6),
+	},
+	{
+		.freq			= 5600,
+		RADIOREGS3(0x28, 0x01, 0x01, 0x02, 0x30, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x89, 0x08, 0x00,
+			   0x50, 0x51, 0x00, 0x00, 0x00, 0x77, 0x00, 0x08,
+			   0x00, 0x6f, 0x00, 0x51, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x08, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08c4, 0x08c0, 0x08bc, 0x01d3, 0x01d4, 0x01d5),
+	},
+	{
+		.freq			= 5610,
+		RADIOREGS3(0x28, 0x01, 0x01, 0x02, 0x31, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x89, 0x08, 0x00,
+			   0x50, 0x51, 0x00, 0x00, 0x00, 0x77, 0x00, 0x08,
+			   0x00, 0x6f, 0x00, 0x51, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x08, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08c8, 0x08c4, 0x08c0, 0x01d2, 0x01d3, 0x01d4),
+	},
+	{
+		.freq			= 5620,
+		RADIOREGS3(0x21, 0x01, 0x01, 0x02, 0x32, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x89, 0x08, 0x00,
+			   0x50, 0x50, 0x00, 0x00, 0x00, 0x77, 0x00, 0x07,
+			   0x00, 0x6f, 0x00, 0x50, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x07, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08cc, 0x08c8, 0x08c4, 0x01d2, 0x01d2, 0x01d3),
+	},
+	{
+		.freq			= 5630,
+		RADIOREGS3(0x21, 0x01, 0x01, 0x02, 0x33, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x88, 0x07, 0x00,
+			   0x50, 0x50, 0x00, 0x00, 0x00, 0x77, 0x00, 0x07,
+			   0x00, 0x6f, 0x00, 0x50, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x07, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08d0, 0x08cc, 0x08c8, 0x01d1, 0x01d2, 0x01d2),
+	},
+	{
+		.freq			= 5640,
+		RADIOREGS3(0x1c, 0x01, 0x01, 0x02, 0x34, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x88, 0x07, 0x00,
+			   0x40, 0x50, 0x00, 0x00, 0x00, 0x77, 0x00, 0x07,
+			   0x00, 0x6f, 0x00, 0x50, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x07, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08d4, 0x08d0, 0x08cc, 0x01d0, 0x01d1, 0x01d2),
+	},
+	{
+		.freq			= 5650,
+		RADIOREGS3(0x1c, 0x01, 0x01, 0x02, 0x35, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x88, 0x07, 0x00,
+			   0x40, 0x40, 0x00, 0x00, 0x00, 0x77, 0x00, 0x07,
+			   0x00, 0x6f, 0x00, 0x40, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x07, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08d8, 0x08d4, 0x08d0, 0x01cf, 0x01d0, 0x01d1),
+	},
+	{
+		.freq			= 5660,
+		RADIOREGS3(0x16, 0x01, 0x01, 0x02, 0x36, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x88, 0x07, 0x00,
+			   0x40, 0x40, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6f, 0x00, 0x40, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08dc, 0x08d8, 0x08d4, 0x01ce, 0x01cf, 0x01d0),
+	},
+	{
+		.freq			= 5670,
+		RADIOREGS3(0x16, 0x01, 0x01, 0x02, 0x37, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x88, 0x07, 0x00,
+			   0x40, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6f, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08e0, 0x08dc, 0x08d8, 0x01ce, 0x01ce, 0x01cf),
+	},
+	{
+		.freq			= 5680,
+		RADIOREGS3(0x10, 0x01, 0x01, 0x02, 0x38, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x06, 0x00,
+			   0x30, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6f, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08e4, 0x08e0, 0x08dc, 0x01cd, 0x01ce, 0x01ce),
+	},
+	{
+		.freq			= 5690,
+		RADIOREGS3(0x10, 0x01, 0x01, 0x02, 0x39, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x06, 0x00,
+			   0x30, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6f, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6f, 0x00),
+		PHYREGS(0x08e8, 0x08e4, 0x08e0, 0x01cc, 0x01cd, 0x01ce),
+	},
+	{
+		.freq			= 5700,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x3a, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x06, 0x00,
+			   0x30, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6e, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6e, 0x00),
+		PHYREGS(0x08ec, 0x08e8, 0x08e4, 0x01cb, 0x01cc, 0x01cd),
+	},
+	{
+		.freq			= 5710,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x3b, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x06, 0x00,
+			   0x30, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6e, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6e, 0x00),
+		PHYREGS(0x08f0, 0x08ec, 0x08e8, 0x01ca, 0x01cb, 0x01cc),
+	},
+	{
+		.freq			= 5720,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x3c, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x06, 0x00,
+			   0x30, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6e, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6e, 0x00),
+		PHYREGS(0x08f4, 0x08f0, 0x08ec, 0x01c9, 0x01ca, 0x01cb),
+	},
+	{
+		.freq			= 5725,
+		RADIOREGS3(0x03, 0x01, 0x02, 0x04, 0x79, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x05, 0x05, 0x05, 0x87, 0x06, 0x00,
+			   0x30, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6e, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6e, 0x00),
+		PHYREGS(0x08f6, 0x08f2, 0x08ee, 0x01c9, 0x01ca, 0x01cb),
+	},
+	{
+		.freq			= 5730,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x3d, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x05, 0x00,
+			   0x20, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6e, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6e, 0x00),
+		PHYREGS(0x08f8, 0x08f4, 0x08f0, 0x01c9, 0x01c9, 0x01ca),
+	},
+	{
+		.freq			= 5735,
+		RADIOREGS3(0x03, 0x01, 0x02, 0x04, 0x7b, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x05, 0x05, 0x05, 0x87, 0x05, 0x00,
+			   0x20, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6d, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6d, 0x00),
+		PHYREGS(0x08fa, 0x08f6, 0x08f2, 0x01c8, 0x01c9, 0x01ca),
+	},
+	{
+		.freq			= 5740,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x3e, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x05, 0x00,
+			   0x20, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6d, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6d, 0x00),
+		PHYREGS(0x08fc, 0x08f8, 0x08f4, 0x01c8, 0x01c9, 0x01c9),
+	},
+	{
+		.freq			= 5745,
+		RADIOREGS3(0xfe, 0x00, 0x02, 0x04, 0x7d, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x05, 0x05, 0x05, 0x87, 0x05, 0x00,
+			   0x20, 0x30, 0x00, 0x00, 0x00, 0x77, 0x00, 0x06,
+			   0x00, 0x6d, 0x00, 0x30, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x06, 0x00, 0x6d, 0x00),
+		PHYREGS(0x08fe, 0x08fa, 0x08f6, 0x01c8, 0x01c8, 0x01c9),
+	},
+	{
+		.freq			= 5750,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x3f, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x87, 0x05, 0x00,
+			   0x20, 0x20, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6d, 0x00, 0x20, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6d, 0x00),
+		PHYREGS(0x0900, 0x08fc, 0x08f8, 0x01c7, 0x01c8, 0x01c9),
+	},
+	{
+		.freq			= 5755,
+		RADIOREGS3(0xfe, 0x00, 0x02, 0x04, 0x7f, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x05, 0x05, 0x05, 0x87, 0x05, 0x00,
+			   0x10, 0x20, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6c, 0x00, 0x20, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6c, 0x00),
+		PHYREGS(0x0902, 0x08fe, 0x08fa, 0x01c7, 0x01c8, 0x01c8),
+	},
+	{
+		.freq			= 5760,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x40, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x86, 0x05, 0x00,
+			   0x10, 0x20, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6c, 0x00, 0x20, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6c, 0x00),
+		PHYREGS(0x0904, 0x0900, 0x08fc, 0x01c6, 0x01c7, 0x01c8),
+	},
+	{
+		.freq			= 5765,
+		RADIOREGS3(0xf8, 0x00, 0x02, 0x04, 0x81, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x05, 0x05, 0x05, 0x86, 0x05, 0x00,
+			   0x10, 0x10, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6c, 0x00, 0x10, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6c, 0x00),
+		PHYREGS(0x0906, 0x0902, 0x08fe, 0x01c6, 0x01c7, 0x01c8),
+	},
+	{
+		.freq			= 5770,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x41, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x86, 0x04, 0x00,
+			   0x10, 0x10, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x10, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x0908, 0x0904, 0x0900, 0x01c6, 0x01c6, 0x01c7),
+	},
+	{
+		.freq			= 5775,
+		RADIOREGS3(0xf8, 0x00, 0x02, 0x04, 0x83, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x05, 0x05, 0x05, 0x86, 0x04, 0x00,
+			   0x10, 0x10, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x10, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x090a, 0x0906, 0x0902, 0x01c5, 0x01c6, 0x01c7),
+	},
+	{
+		.freq			= 5780,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x42, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x05, 0x05, 0x05, 0x86, 0x04, 0x00,
+			   0x10, 0x10, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x10, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x090c, 0x0908, 0x0904, 0x01c5, 0x01c6, 0x01c6),
+	},
+	{
+		.freq			= 5785,
+		RADIOREGS3(0xf2, 0x00, 0x02, 0x04, 0x85, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x10, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x10, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x090e, 0x090a, 0x0906, 0x01c4, 0x01c5, 0x01c6),
+	},
+	{
+		.freq			= 5790,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x43, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x10, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x10, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x0910, 0x090c, 0x0908, 0x01c4, 0x01c5, 0x01c6),
+	},
+	{
+		.freq			= 5795,
+		RADIOREGS3(0xf2, 0x00, 0x02, 0x04, 0x87, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x0912, 0x090e, 0x090a, 0x01c4, 0x01c4, 0x01c5),
+	},
+	{
+		.freq			= 5800,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x44, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6b, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6b, 0x00),
+		PHYREGS(0x0914, 0x0910, 0x090c, 0x01c3, 0x01c4, 0x01c5),
+	},
+	{
+		.freq			= 5805,
+		RADIOREGS3(0xed, 0x00, 0x02, 0x04, 0x89, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6a, 0x00),
+		PHYREGS(0x0916, 0x0912, 0x090e, 0x01c3, 0x01c4, 0x01c4),
+	},
+	{
+		.freq			= 5810,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x45, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6a, 0x00),
+		PHYREGS(0x0918, 0x0914, 0x0910, 0x01c2, 0x01c3, 0x01c4),
+	},
+	{
+		.freq			= 5815,
+		RADIOREGS3(0xed, 0x00, 0x02, 0x04, 0x8b, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6a, 0x00),
+		PHYREGS(0x091a, 0x0916, 0x0912, 0x01c2, 0x01c3, 0x01c4),
+	},
+	{
+		.freq			= 5820,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x46, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x6a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x6a, 0x00),
+		PHYREGS(0x091c, 0x0918, 0x0914, 0x01c2, 0x01c2, 0x01c3),
+	},
+	{
+		.freq			= 5825,
+		RADIOREGS3(0xed, 0x00, 0x02, 0x04, 0x8d, 0x05, 0x05, 0x02,
+			   0x15, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x69, 0x00),
+		PHYREGS(0x091e, 0x091a, 0x0916, 0x01c1, 0x01c2, 0x01c3),
+	},
+	{
+		.freq			= 5830,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x47, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x05,
+			   0x00, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x05, 0x00, 0x69, 0x00),
+		PHYREGS(0x0920, 0x091c, 0x0918, 0x01c1, 0x01c2, 0x01c2),
+	},
+	{
+		.freq			= 5840,
+		RADIOREGS3(0x0a, 0x01, 0x01, 0x02, 0x48, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x86, 0x04, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x69, 0x00),
+		PHYREGS(0x0924, 0x0920, 0x091c, 0x01c0, 0x01c1, 0x01c2),
+	},
+	{
+		.freq			= 5850,
+		RADIOREGS3(0xe0, 0x00, 0x01, 0x02, 0x49, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x69, 0x00),
+		PHYREGS(0x0928, 0x0924, 0x0920, 0x01bf, 0x01c0, 0x01c1),
+	},
+	{
+		.freq			= 5860,
+		RADIOREGS3(0xde, 0x00, 0x01, 0x02, 0x4a, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x69, 0x00),
+		PHYREGS(0x092c, 0x0928, 0x0924, 0x01bf, 0x01bf, 0x01c0),
+	},
+	{
+		.freq			= 5870,
+		RADIOREGS3(0xdb, 0x00, 0x01, 0x02, 0x4b, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x68, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x68, 0x00),
+		PHYREGS(0x0930, 0x092c, 0x0928, 0x01be, 0x01bf, 0x01bf),
+	},
+	{
+		.freq			= 5880,
+		RADIOREGS3(0xd8, 0x00, 0x01, 0x02, 0x4c, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x68, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x68, 0x00),
+		PHYREGS(0x0934, 0x0930, 0x092c, 0x01bd, 0x01be, 0x01bf),
+	},
+	{
+		.freq			= 5890,
+		RADIOREGS3(0xd6, 0x00, 0x01, 0x02, 0x4d, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x68, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x68, 0x00),
+		PHYREGS(0x0938, 0x0934, 0x0930, 0x01bc, 0x01bd, 0x01be),
+	},
+	{
+		.freq			= 5900,
+		RADIOREGS3(0xd3, 0x00, 0x01, 0x02, 0x4e, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x68, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x68, 0x00),
+		PHYREGS(0x093c, 0x0938, 0x0934, 0x01bc, 0x01bc, 0x01bd),
+	},
+	{
+		.freq			= 5910,
+		RADIOREGS3(0xd6, 0x00, 0x01, 0x02, 0x4f, 0x05, 0x05, 0x02,
+			   0x0c, 0x01, 0x06, 0x06, 0x06, 0x85, 0x03, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x00, 0x77, 0x00, 0x04,
+			   0x00, 0x68, 0x00, 0x00, 0x00, 0x00, 0x00, 0x77,
+			   0x00, 0x04, 0x00, 0x68, 0x00),
+		PHYREGS(0x0940, 0x093c, 0x0938, 0x01bb, 0x01bc, 0x01bc),
+	},
+	{
+		.freq			= 2412,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x6c, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x04, 0x04, 0x04, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x78, 0x00, 0x03, 0x00, 0x70, 0x00,
+			   0x0b, 0x00, 0x0a, 0x00, 0x89, 0x00, 0x03, 0x00,
+			   0x70, 0x00, 0x0b, 0x00, 0x0a),
+		PHYREGS(0x03c9, 0x03c5, 0x03c1, 0x043a, 0x043f, 0x0443),
+	},
+	{
+		.freq			= 2417,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x71, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x05, 0x05, 0x05, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x78, 0x00, 0x03, 0x00, 0x70, 0x00,
+			   0x0b, 0x00, 0x0a, 0x00, 0x89, 0x00, 0x03, 0x00,
+			   0x70, 0x00, 0x0b, 0x00, 0x0a),
+		PHYREGS(0x03cb, 0x03c7, 0x03c3, 0x0438, 0x043d, 0x0441),
+	},
+	{
+		.freq			= 2422,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x76, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x05, 0x05, 0x05, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x67, 0x00, 0x03, 0x00, 0x70, 0x00,
+			   0x0b, 0x00, 0x0a, 0x00, 0x89, 0x00, 0x03, 0x00,
+			   0x70, 0x00, 0x0b, 0x00, 0x0a),
+		PHYREGS(0x03cd, 0x03c9, 0x03c5, 0x0436, 0x043a, 0x043f),
+	},
+	{
+		.freq			= 2427,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x7b, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x05, 0x05, 0x05, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x57, 0x00, 0x03, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x0a, 0x00, 0x78, 0x00, 0x03, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x0a),
+		PHYREGS(0x03cf, 0x03cb, 0x03c7, 0x0434, 0x0438, 0x043d),
+	},
+	{
+		.freq			= 2432,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x80, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x05, 0x05, 0x05, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x56, 0x00, 0x03, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x0a, 0x00, 0x77, 0x00, 0x03, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x0a),
+		PHYREGS(0x03d1, 0x03cd, 0x03c9, 0x0431, 0x0436, 0x043a),
+	},
+	{
+		.freq			= 2437,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x85, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x05, 0x05, 0x05, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x46, 0x00, 0x03, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x0a, 0x00, 0x76, 0x00, 0x03, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x0a),
+		PHYREGS(0x03d3, 0x03cf, 0x03cb, 0x042f, 0x0434, 0x0438),
+	},
+	{
+		.freq			= 2442,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x8a, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x05, 0x05, 0x05, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x45, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x0a, 0x00, 0x66, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x0a),
+		PHYREGS(0x03d5, 0x03d1, 0x03cd, 0x042d, 0x0431, 0x0436),
+	},
+	{
+		.freq			= 2447,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x8f, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x06, 0x06, 0x06, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x34, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x09, 0x00, 0x55, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x09),
+		PHYREGS(0x03d7, 0x03d3, 0x03cf, 0x042b, 0x042f, 0x0434),
+	},
+	{
+		.freq			= 2452,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x94, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x06, 0x06, 0x06, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x23, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x09, 0x00, 0x45, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x09),
+		PHYREGS(0x03d9, 0x03d5, 0x03d1, 0x0429, 0x042d, 0x0431),
+	},
+	{
+		.freq			= 2457,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x99, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x06, 0x06, 0x06, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x12, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x0a, 0x00, 0x09, 0x00, 0x34, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x0a, 0x00, 0x09),
+		PHYREGS(0x03db, 0x03d7, 0x03d3, 0x0427, 0x042b, 0x042f),
+	},
+	{
+		.freq			= 2462,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0x9e, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x06, 0x06, 0x06, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x02, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x09, 0x00, 0x09, 0x00, 0x33, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x09, 0x00, 0x09),
+		PHYREGS(0x03dd, 0x03d9, 0x03d5, 0x0424, 0x0429, 0x042d),
+	},
+	{
+		.freq			= 2467,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0xa3, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x06, 0x06, 0x06, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x01, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x09, 0x00, 0x09, 0x00, 0x22, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x09, 0x00, 0x09),
+		PHYREGS(0x03df, 0x03db, 0x03d7, 0x0422, 0x0427, 0x042b),
+	},
+	{
+		.freq			= 2472,
+		RADIOREGS3(0x00, 0x01, 0x03, 0x09, 0xa8, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x07, 0x07, 0x07, 0x8f, 0x30, 0x00,
+			   0x00, 0x00, 0x01, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x09, 0x00, 0x09, 0x00, 0x11, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x09, 0x00, 0x09),
+		PHYREGS(0x03e1, 0x03dd, 0x03d9, 0x0420, 0x0424, 0x0429),
+	},
+	{
+		.freq			= 2484,
+		RADIOREGS3(0xff, 0x01, 0x03, 0x09, 0xb4, 0x06, 0x06, 0x04,
+			   0x2b, 0x01, 0x07, 0x07, 0x07, 0x8f, 0x20, 0x00,
+			   0x00, 0x00, 0x00, 0x00, 0x02, 0x00, 0x70, 0x00,
+			   0x09, 0x00, 0x09, 0x00, 0x00, 0x00, 0x02, 0x00,
+			   0x70, 0x00, 0x09, 0x00, 0x09),
+		PHYREGS(0x03e6, 0x03e2, 0x03de, 0x041b, 0x041f, 0x0424),
+	},
+};
+
+static const struct b2056_inittabs_pts
+*b43_nphy_get_inittabs_rev3(struct b43_wldev *dev)
+{
+	struct b43_phy *phy = &dev->phy;
+
+	switch (dev->phy.rev) {
+	case 3:
+		return &b2056_inittab_phy_rev3;
+	case 4:
+		return &b2056_inittab_phy_rev4;
+	default:
+		switch (phy->radio_rev) {
+		case 5:
+			return &b2056_inittab_radio_rev5;
+		case 6:
+			return &b2056_inittab_radio_rev6;
+		case 7:
+		case 9:
+			return &b2056_inittab_radio_rev7_9;
+		case 8:
+			return &b2056_inittab_radio_rev8;
+		case 11:
+			return &b2056_inittab_radio_rev11;
+		}
+	}
+
+	return NULL;
+}
+
 static void b2056_upload_inittab(struct b43_wldev *dev, bool ghz5,
 				 bool ignore_uploadflag, u16 routing,
 				 const struct b2056_inittab_entry *e,
@@ -9037,11 +10230,11 @@ void b2056_upload_inittabs(struct b43_wldev *dev,
 {
 	const struct b2056_inittabs_pts *pts;
 
-	if (dev->phy.rev >= ARRAY_SIZE(b2056_inittabs)) {
+	pts = b43_nphy_get_inittabs_rev3(dev);
+	if (!pts) {
 		B43_WARN_ON(1);
 		return;
 	}
-	pts = &b2056_inittabs[dev->phy.rev];
 
 	b2056_upload_inittab(dev, ghz5, ignore_uploadflag,
 				B2056_SYN, pts->syn, pts->syn_length);
@@ -9060,11 +10253,12 @@ void b2056_upload_syn_pll_cp2(struct b43_wldev *dev, bool ghz5)
 	const struct b2056_inittabs_pts *pts;
 	const struct b2056_inittab_entry *e;
 
-	if (dev->phy.rev >= ARRAY_SIZE(b2056_inittabs)) {
+	pts = b43_nphy_get_inittabs_rev3(dev);
+	if (!pts) {
 		B43_WARN_ON(1);
 		return;
 	}
-	pts = &b2056_inittabs[dev->phy.rev];
+
 	e = &pts->syn[B2056_SYN_PLL_CP2];
 
 	b43_radio_write(dev, B2056_SYN_PLL_CP2, ghz5 ? e->ghz5 : e->ghz2);
@@ -9073,38 +10267,46 @@ void b2056_upload_syn_pll_cp2(struct b43_wldev *dev, bool ghz5)
 const struct b43_nphy_channeltab_entry_rev3 *
 b43_nphy_get_chantabent_rev3(struct b43_wldev *dev, u16 freq)
 {
+	struct b43_phy *phy = &dev->phy;
 	const struct b43_nphy_channeltab_entry_rev3 *e;
 	unsigned int length, i;
 
-	switch (dev->phy.rev) {
+	switch (phy->rev) {
 	case 3:
-		e = b43_nphy_channeltab_rev3;
-		length = ARRAY_SIZE(b43_nphy_channeltab_rev3);
+		e = b43_nphy_channeltab_phy_rev3;
+		length = ARRAY_SIZE(b43_nphy_channeltab_phy_rev3);
 		break;
 	case 4:
-		e = b43_nphy_channeltab_rev4;
-		length = ARRAY_SIZE(b43_nphy_channeltab_rev4);
-		break;
-	case 5:
-		e = b43_nphy_channeltab_rev5;
-		length = ARRAY_SIZE(b43_nphy_channeltab_rev5);
-		break;
-	case 6:
-		e = b43_nphy_channeltab_rev6;
-		length = ARRAY_SIZE(b43_nphy_channeltab_rev6);
-		break;
-	case 7:
-	case 9:
-		e = b43_nphy_channeltab_rev7_9;
-		length = ARRAY_SIZE(b43_nphy_channeltab_rev7_9);
-		break;
-	case 8:
-		e = b43_nphy_channeltab_rev8;
-		length = ARRAY_SIZE(b43_nphy_channeltab_rev8);
+		e = b43_nphy_channeltab_phy_rev4;
+		length = ARRAY_SIZE(b43_nphy_channeltab_phy_rev4);
 		break;
 	default:
-		B43_WARN_ON(1);
-		return NULL;
+		switch (phy->radio_rev) {
+		case 5:
+			e = b43_nphy_channeltab_radio_rev5;
+			length = ARRAY_SIZE(b43_nphy_channeltab_radio_rev5);
+			break;
+		case 6:
+			e = b43_nphy_channeltab_radio_rev6;
+			length = ARRAY_SIZE(b43_nphy_channeltab_radio_rev6);
+			break;
+		case 7:
+		case 9:
+			e = b43_nphy_channeltab_radio_rev7_9;
+			length = ARRAY_SIZE(b43_nphy_channeltab_radio_rev7_9);
+			break;
+		case 8:
+			e = b43_nphy_channeltab_radio_rev8;
+			length = ARRAY_SIZE(b43_nphy_channeltab_radio_rev8);
+			break;
+		case 11:
+			e = b43_nphy_channeltab_radio_rev11;
+			length = ARRAY_SIZE(b43_nphy_channeltab_radio_rev11);
+			break;
+		default:
+			B43_WARN_ON(1);
+			return NULL;
+		}
 	}
 
 	for (i = 0; i < length; i++, e++) {
diff --git a/drivers/net/wireless/b43/tables_nphy.c b/drivers/net/wireless/b43/tables_nphy.c
index 94c755fdda14..4047c05e3807 100644
--- a/drivers/net/wireless/b43/tables_nphy.c
+++ b/drivers/net/wireless/b43/tables_nphy.c
@@ -1627,74 +1627,7 @@ static const u32 b43_ntab_tdtrn_r3[] = {
 	0xfa58fc00, 0x0b64fc7e, 0x0800f7b6, 0x00f006be,
 };
 
-static const u32 b43_ntab_noisevar0_r3[] = {
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
-};
-
-static const u32 b43_ntab_noisevar1_r3[] = {
+static const u32 b43_ntab_noisevar_r3[] = {
 	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
 	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
 	0x02110211, 0x0000014d, 0x02110211, 0x0000014d,
@@ -3109,31 +3042,32 @@ static void b43_nphy_tables_init_rev3(struct b43_wldev *dev)
 		antswlut = sprom->fem.ghz2.antswlut;
 
 	/* Static tables */
-	ntab_upload(dev, B43_NTAB_FRAMESTRUCT_R3, b43_ntab_framestruct_r3);
-	ntab_upload(dev, B43_NTAB_PILOT_R3, b43_ntab_pilot_r3);
-	ntab_upload(dev, B43_NTAB_TMAP_R3, b43_ntab_tmap_r3);
-	ntab_upload(dev, B43_NTAB_INTLEVEL_R3, b43_ntab_intlevel_r3);
-	ntab_upload(dev, B43_NTAB_TDTRN_R3, b43_ntab_tdtrn_r3);
-	ntab_upload(dev, B43_NTAB_NOISEVAR0_R3, b43_ntab_noisevar0_r3);
-	ntab_upload(dev, B43_NTAB_NOISEVAR1_R3, b43_ntab_noisevar1_r3);
-	ntab_upload(dev, B43_NTAB_MCS_R3, b43_ntab_mcs_r3);
-	ntab_upload(dev, B43_NTAB_TDI20A0_R3, b43_ntab_tdi20a0_r3);
-	ntab_upload(dev, B43_NTAB_TDI20A1_R3, b43_ntab_tdi20a1_r3);
-	ntab_upload(dev, B43_NTAB_TDI40A0_R3, b43_ntab_tdi40a0_r3);
-	ntab_upload(dev, B43_NTAB_TDI40A1_R3, b43_ntab_tdi40a1_r3);
-	ntab_upload(dev, B43_NTAB_PILOTLT_R3, b43_ntab_pilotlt_r3);
-	ntab_upload(dev, B43_NTAB_CHANEST_R3, b43_ntab_channelest_r3);
-	ntab_upload(dev, B43_NTAB_FRAMELT_R3, b43_ntab_framelookup_r3);
-	ntab_upload(dev, B43_NTAB_C0_ESTPLT_R3, b43_ntab_estimatepowerlt0_r3);
-	ntab_upload(dev, B43_NTAB_C1_ESTPLT_R3, b43_ntab_estimatepowerlt1_r3);
-	ntab_upload(dev, B43_NTAB_C0_ADJPLT_R3, b43_ntab_adjustpower0_r3);
-	ntab_upload(dev, B43_NTAB_C1_ADJPLT_R3, b43_ntab_adjustpower1_r3);
-	ntab_upload(dev, B43_NTAB_C0_GAINCTL_R3, b43_ntab_gainctl0_r3);
-	ntab_upload(dev, B43_NTAB_C1_GAINCTL_R3, b43_ntab_gainctl1_r3);
-	ntab_upload(dev, B43_NTAB_C0_IQLT_R3, b43_ntab_iqlt0_r3);
-	ntab_upload(dev, B43_NTAB_C1_IQLT_R3, b43_ntab_iqlt1_r3);
-	ntab_upload(dev, B43_NTAB_C0_LOFEEDTH_R3, b43_ntab_loftlt0_r3);
-	ntab_upload(dev, B43_NTAB_C1_LOFEEDTH_R3, b43_ntab_loftlt1_r3);
+	if (dev->phy.do_full_init) {
+		ntab_upload(dev, B43_NTAB_FRAMESTRUCT_R3, b43_ntab_framestruct_r3);
+		ntab_upload(dev, B43_NTAB_PILOT_R3, b43_ntab_pilot_r3);
+		ntab_upload(dev, B43_NTAB_TMAP_R3, b43_ntab_tmap_r3);
+		ntab_upload(dev, B43_NTAB_INTLEVEL_R3, b43_ntab_intlevel_r3);
+		ntab_upload(dev, B43_NTAB_TDTRN_R3, b43_ntab_tdtrn_r3);
+		ntab_upload(dev, B43_NTAB_NOISEVAR_R3, b43_ntab_noisevar_r3);
+		ntab_upload(dev, B43_NTAB_MCS_R3, b43_ntab_mcs_r3);
+		ntab_upload(dev, B43_NTAB_TDI20A0_R3, b43_ntab_tdi20a0_r3);
+		ntab_upload(dev, B43_NTAB_TDI20A1_R3, b43_ntab_tdi20a1_r3);
+		ntab_upload(dev, B43_NTAB_TDI40A0_R3, b43_ntab_tdi40a0_r3);
+		ntab_upload(dev, B43_NTAB_TDI40A1_R3, b43_ntab_tdi40a1_r3);
+		ntab_upload(dev, B43_NTAB_PILOTLT_R3, b43_ntab_pilotlt_r3);
+		ntab_upload(dev, B43_NTAB_CHANEST_R3, b43_ntab_channelest_r3);
+		ntab_upload(dev, B43_NTAB_FRAMELT_R3, b43_ntab_framelookup_r3);
+		ntab_upload(dev, B43_NTAB_C0_ESTPLT_R3, b43_ntab_estimatepowerlt0_r3);
+		ntab_upload(dev, B43_NTAB_C1_ESTPLT_R3, b43_ntab_estimatepowerlt1_r3);
+		ntab_upload(dev, B43_NTAB_C0_ADJPLT_R3, b43_ntab_adjustpower0_r3);
+		ntab_upload(dev, B43_NTAB_C1_ADJPLT_R3, b43_ntab_adjustpower1_r3);
+		ntab_upload(dev, B43_NTAB_C0_GAINCTL_R3, b43_ntab_gainctl0_r3);
+		ntab_upload(dev, B43_NTAB_C1_GAINCTL_R3, b43_ntab_gainctl1_r3);
+		ntab_upload(dev, B43_NTAB_C0_IQLT_R3, b43_ntab_iqlt0_r3);
+		ntab_upload(dev, B43_NTAB_C1_IQLT_R3, b43_ntab_iqlt1_r3);
+		ntab_upload(dev, B43_NTAB_C0_LOFEEDTH_R3, b43_ntab_loftlt0_r3);
+		ntab_upload(dev, B43_NTAB_C1_LOFEEDTH_R3, b43_ntab_loftlt1_r3);
+	}
 
 	/* Volatile tables */
 	if (antswlut < ARRAY_SIZE(b43_ntab_antswctl_r3))
@@ -3146,20 +3080,22 @@ static void b43_nphy_tables_init_rev3(struct b43_wldev *dev)
 static void b43_nphy_tables_init_rev0(struct b43_wldev *dev)
 {
 	/* Static tables */
-	ntab_upload(dev, B43_NTAB_FRAMESTRUCT, b43_ntab_framestruct);
-	ntab_upload(dev, B43_NTAB_FRAMELT, b43_ntab_framelookup);
-	ntab_upload(dev, B43_NTAB_TMAP, b43_ntab_tmap);
-	ntab_upload(dev, B43_NTAB_TDTRN, b43_ntab_tdtrn);
-	ntab_upload(dev, B43_NTAB_INTLEVEL, b43_ntab_intlevel);
-	ntab_upload(dev, B43_NTAB_PILOT, b43_ntab_pilot);
-	ntab_upload(dev, B43_NTAB_TDI20A0, b43_ntab_tdi20a0);
-	ntab_upload(dev, B43_NTAB_TDI20A1, b43_ntab_tdi20a1);
-	ntab_upload(dev, B43_NTAB_TDI40A0, b43_ntab_tdi40a0);
-	ntab_upload(dev, B43_NTAB_TDI40A1, b43_ntab_tdi40a1);
-	ntab_upload(dev, B43_NTAB_CHANEST, b43_ntab_channelest);
-	ntab_upload(dev, B43_NTAB_MCS, b43_ntab_mcs);
-	ntab_upload(dev, B43_NTAB_NOISEVAR10, b43_ntab_noisevar10);
-	ntab_upload(dev, B43_NTAB_NOISEVAR11, b43_ntab_noisevar11);
+	if (dev->phy.do_full_init) {
+		ntab_upload(dev, B43_NTAB_FRAMESTRUCT, b43_ntab_framestruct);
+		ntab_upload(dev, B43_NTAB_FRAMELT, b43_ntab_framelookup);
+		ntab_upload(dev, B43_NTAB_TMAP, b43_ntab_tmap);
+		ntab_upload(dev, B43_NTAB_TDTRN, b43_ntab_tdtrn);
+		ntab_upload(dev, B43_NTAB_INTLEVEL, b43_ntab_intlevel);
+		ntab_upload(dev, B43_NTAB_PILOT, b43_ntab_pilot);
+		ntab_upload(dev, B43_NTAB_TDI20A0, b43_ntab_tdi20a0);
+		ntab_upload(dev, B43_NTAB_TDI20A1, b43_ntab_tdi20a1);
+		ntab_upload(dev, B43_NTAB_TDI40A0, b43_ntab_tdi40a0);
+		ntab_upload(dev, B43_NTAB_TDI40A1, b43_ntab_tdi40a1);
+		ntab_upload(dev, B43_NTAB_CHANEST, b43_ntab_channelest);
+		ntab_upload(dev, B43_NTAB_MCS, b43_ntab_mcs);
+		ntab_upload(dev, B43_NTAB_NOISEVAR10, b43_ntab_noisevar10);
+		ntab_upload(dev, B43_NTAB_NOISEVAR11, b43_ntab_noisevar11);
+	}
 
 	/* Volatile tables */
 	ntab_upload(dev, B43_NTAB_BDI, b43_ntab_bdi);
diff --git a/drivers/net/wireless/b43/tables_nphy.h b/drivers/net/wireless/b43/tables_nphy.h
index 9ff33adcff89..3a58aee4c4cf 100644
--- a/drivers/net/wireless/b43/tables_nphy.h
+++ b/drivers/net/wireless/b43/tables_nphy.h
@@ -143,8 +143,7 @@ struct nphy_gain_ctl_workaround_entry *b43_nphy_get_gain_ctl_workaround_ent(
 #define B43_NTAB_TMAP_R3		B43_NTAB32(12,   0) /* TM AP  */
 #define B43_NTAB_INTLEVEL_R3		B43_NTAB32(13,   0) /* INT LV  */
 #define B43_NTAB_TDTRN_R3		B43_NTAB32(14,   0) /* TD TRN  */
-#define B43_NTAB_NOISEVAR0_R3		B43_NTAB32(16,   0) /* noise variance 0  */
-#define B43_NTAB_NOISEVAR1_R3		B43_NTAB32(16, 128) /* noise variance 1  */
+#define B43_NTAB_NOISEVAR_R3		B43_NTAB32(16,   0) /* noise variance */
 #define B43_NTAB_MCS_R3			B43_NTAB16(18,   0) /* MCS  */
 #define B43_NTAB_TDI20A0_R3		B43_NTAB32(19, 128) /* TDI 20/0  */
 #define B43_NTAB_TDI20A1_R3		B43_NTAB32(19, 256) /* TDI 20/1  */
diff --git a/drivers/net/wireless/b43/wa.c b/drivers/net/wireless/b43/wa.c
index 9b1a038be08b..c218c08fb2f5 100644
--- a/drivers/net/wireless/b43/wa.c
+++ b/drivers/net/wireless/b43/wa.c
@@ -441,7 +441,7 @@ static void b43_wa_altagc(struct b43_wldev *dev)
 
 static void b43_wa_tr_ltov(struct b43_wldev *dev) /* TR Lookup Table Original Values */
 {
-	b43_gtab_write(dev, B43_GTAB_ORIGTR, 0, 0xC480);
+	b43_gtab_write(dev, B43_GTAB_ORIGTR, 0, 0x7654);
 }
 
 static void b43_wa_cpll_nonpilot(struct b43_wldev *dev)
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index 31adb8cf0291..6e6ef3fc2247 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -408,7 +408,7 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		mac_ctl |= B43_TXH_MAC_HWSEQ;
 	if (info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
 		mac_ctl |= B43_TXH_MAC_STMSDU;
-	if (phy->type == B43_PHYTYPE_A)
+	if (!phy->gmode)
 		mac_ctl |= B43_TXH_MAC_5GHZ;
 
 	/* Overwrite rates[0].count to make the retry calculation
@@ -811,9 +811,13 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr)
 		break;
 	case B43_PHYTYPE_G:
 		status.band = IEEE80211_BAND_2GHZ;
-		/* chanid is the radio channel cookie value as used
-		 * to tune the radio. */
-		status.freq = chanid + 2400;
+		/* Somewhere between 478.104 and 508.1084 firmware for G-PHY
+		 * has been modified to be compatible with N-PHY and others.
+		 */
+		if (dev->fw.rev >= 508)
+			status.freq = ieee80211_channel_to_frequency(chanid, status.band);
+		else
+			status.freq = chanid + 2400;
 		break;
 	case B43_PHYTYPE_N:
 	case B43_PHYTYPE_LP:
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/Makefile b/drivers/net/wireless/brcm80211/brcmfmac/Makefile
index 1d2ceac3a221..98e67c18f276 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/Makefile
+++ b/drivers/net/wireless/brcm80211/brcmfmac/Makefile
@@ -33,7 +33,7 @@ brcmfmac-objs += \
 		bcdc.o \
 		dhd_common.o \
 		dhd_linux.o \
-		nvram.o \
+		firmware.o \
 		btcoex.o
 brcmfmac-$(CONFIG_BRCMFMAC_SDIO) += \
 		dhd_sdio.o \
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/dhd.h b/drivers/net/wireless/brcm80211/brcmfmac/dhd.h
index 939d6b132922..16f9ab2568a8 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/dhd.h
+++ b/drivers/net/wireless/brcm80211/brcmfmac/dhd.h
@@ -186,7 +186,7 @@ void brcmf_del_if(struct brcmf_pub *drvr, s32 bssidx);
 void brcmf_txflowblock_if(struct brcmf_if *ifp,
 			  enum brcmf_netif_stop_reason reason, bool state);
 u32 brcmf_get_chip_info(struct brcmf_if *ifp);
-void brcmf_txfinalize(struct brcmf_pub *drvr, struct sk_buff *txp,
+void brcmf_txfinalize(struct brcmf_pub *drvr, struct sk_buff *txp, u8 ifidx,
 		      bool success);
 
 /* Sets dongle media info (drv_version, mac address). */
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/dhd_bus.h b/drivers/net/wireless/brcm80211/brcmfmac/dhd_bus.h
index c4535616064e..7735328fff21 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/dhd_bus.h
+++ b/drivers/net/wireless/brcm80211/brcmfmac/dhd_bus.h
@@ -63,7 +63,6 @@ struct brcmf_bus_dcmd {
  */
 struct brcmf_bus_ops {
 	int (*preinit)(struct device *dev);
-	int (*init)(struct device *dev);
 	void (*stop)(struct device *dev);
 	int (*txdata)(struct device *dev, struct sk_buff *skb);
 	int (*txctl)(struct device *dev, unsigned char *msg, uint len);
@@ -99,6 +98,7 @@ struct brcmf_bus {
 	unsigned long tx_realloc;
 	u32 chip;
 	u32 chiprev;
+	bool always_use_fws_queue;
 
 	struct brcmf_bus_ops *ops;
 };
@@ -113,11 +113,6 @@ static inline int brcmf_bus_preinit(struct brcmf_bus *bus)
 	return bus->ops->preinit(bus->dev);
 }
 
-static inline int brcmf_bus_init(struct brcmf_bus *bus)
-{
-	return bus->ops->init(bus->dev);
-}
-
 static inline void brcmf_bus_stop(struct brcmf_bus *bus)
 {
 	bus->ops->stop(bus->dev);
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/dhd_common.c b/drivers/net/wireless/brcm80211/brcmfmac/dhd_common.c
index 6a8983a1fb9c..ed3e32ce8c23 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/dhd_common.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/dhd_common.c
@@ -32,6 +32,9 @@
 #define BRCMF_DEFAULT_SCAN_UNASSOC_TIME	40
 #define BRCMF_DEFAULT_PACKET_FILTER	"100 0 0 0 0x01 0x00"
 
+/* boost value for RSSI_DELTA in preferred join selection */
+#define BRCMF_JOIN_PREF_RSSI_BOOST	8
+
 
 bool brcmf_c_prec_enq(struct device *dev, struct pktq *q,
 		      struct sk_buff *pkt, int prec)
@@ -246,6 +249,7 @@ int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
 {
 	s8 eventmask[BRCMF_EVENTING_MASK_LEN];
 	u8 buf[BRCMF_DCMD_SMLEN];
+	struct brcmf_join_pref_params join_pref_params[2];
 	char *ptr;
 	s32 err;
 
@@ -298,6 +302,20 @@ int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
 		goto done;
 	}
 
+	/* Setup join_pref to select target by RSSI(with boost on 5GHz) */
+	join_pref_params[0].type = BRCMF_JOIN_PREF_RSSI_DELTA;
+	join_pref_params[0].len = 2;
+	join_pref_params[0].rssi_gain = BRCMF_JOIN_PREF_RSSI_BOOST;
+	join_pref_params[0].band = WLC_BAND_5G;
+	join_pref_params[1].type = BRCMF_JOIN_PREF_RSSI;
+	join_pref_params[1].len = 2;
+	join_pref_params[1].rssi_gain = 0;
+	join_pref_params[1].band = 0;
+	err = brcmf_fil_iovar_data_set(ifp, "join_pref", join_pref_params,
+				       sizeof(join_pref_params));
+	if (err)
+		brcmf_err("Set join_pref error (%d)\n", err);
+
 	/* Setup event_msgs, enable E_IF */
 	err = brcmf_fil_iovar_data_get(ifp, "event_msgs", eventmask,
 				       BRCMF_EVENTING_MASK_LEN);
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/dhd_linux.c b/drivers/net/wireless/brcm80211/brcmfmac/dhd_linux.c
index 7d28cd385092..09dd8c13d844 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/dhd_linux.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/dhd_linux.c
@@ -190,7 +190,7 @@ static netdev_tx_t brcmf_netdev_start_xmit(struct sk_buff *skb,
 	int ret;
 	struct brcmf_if *ifp = netdev_priv(ndev);
 	struct brcmf_pub *drvr = ifp->drvr;
-	struct ethhdr *eh;
+	struct ethhdr *eh = (struct ethhdr *)(skb->data);
 
 	brcmf_dbg(DATA, "Enter, idx=%d\n", ifp->bssidx);
 
@@ -236,6 +236,9 @@ static netdev_tx_t brcmf_netdev_start_xmit(struct sk_buff *skb,
 		goto done;
 	}
 
+	if (eh->h_proto == htons(ETH_P_PAE))
+		atomic_inc(&ifp->pend_8021x_cnt);
+
 	ret = brcmf_fws_process_skb(ifp, skb);
 
 done:
@@ -538,31 +541,26 @@ void brcmf_rx_frame(struct device *dev, struct sk_buff *skb)
 		brcmf_netif_rx(ifp, skb);
 }
 
-void brcmf_txfinalize(struct brcmf_pub *drvr, struct sk_buff *txp,
+void brcmf_txfinalize(struct brcmf_pub *drvr, struct sk_buff *txp, u8 ifidx,
 		      bool success)
 {
 	struct brcmf_if *ifp;
 	struct ethhdr *eh;
-	u8 ifidx;
 	u16 type;
-	int res;
-
-	res = brcmf_proto_hdrpull(drvr, false, &ifidx, txp);
 
 	ifp = drvr->iflist[ifidx];
 	if (!ifp)
 		goto done;
 
-	if (res == 0) {
-		eh = (struct ethhdr *)(txp->data);
-		type = ntohs(eh->h_proto);
+	eh = (struct ethhdr *)(txp->data);
+	type = ntohs(eh->h_proto);
 
-		if (type == ETH_P_PAE) {
-			atomic_dec(&ifp->pend_8021x_cnt);
-			if (waitqueue_active(&ifp->pend_8021x_wait))
-				wake_up(&ifp->pend_8021x_wait);
-		}
+	if (type == ETH_P_PAE) {
+		atomic_dec(&ifp->pend_8021x_cnt);
+		if (waitqueue_active(&ifp->pend_8021x_wait))
+			wake_up(&ifp->pend_8021x_wait);
 	}
+
 	if (!success)
 		ifp->stats.tx_errors++;
 done:
@@ -573,13 +571,17 @@ void brcmf_txcomplete(struct device *dev, struct sk_buff *txp, bool success)
 {
 	struct brcmf_bus *bus_if = dev_get_drvdata(dev);
 	struct brcmf_pub *drvr = bus_if->drvr;
+	u8 ifidx;
 
 	/* await txstatus signal for firmware if active */
 	if (brcmf_fws_fc_active(drvr->fws)) {
 		if (!success)
 			brcmf_fws_bustxfail(drvr->fws, txp);
 	} else {
-		brcmf_txfinalize(drvr, txp, success);
+		if (brcmf_proto_hdrpull(drvr, false, &ifidx, txp))
+			brcmu_pkt_buf_free_skb(txp);
+		else
+			brcmf_txfinalize(drvr, txp, ifidx, success);
 	}
 }
 
@@ -914,13 +916,6 @@ int brcmf_bus_start(struct device *dev)
 
 	brcmf_dbg(TRACE, "\n");
 
-	/* Bring up the bus */
-	ret = brcmf_bus_init(bus_if);
-	if (ret != 0) {
-		brcmf_err("brcmf_sdbrcm_bus_init failed %d\n", ret);
-		return ret;
-	}
-
 	/* add primary networking interface */
 	ifp = brcmf_add_if(drvr, 0, 0, "wlan%d", NULL);
 	if (IS_ERR(ifp))
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c b/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c
index 13c89a0c4ba7..8fa0dbbbda72 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c
@@ -42,7 +42,7 @@
 #include <soc.h>
 #include "sdio_host.h"
 #include "chip.h"
-#include "nvram.h"
+#include "firmware.h"
 
 #define DCMD_RESP_TIMEOUT  2000	/* In milli second */
 
@@ -632,43 +632,28 @@ static const struct brcmf_firmware_names brcmf_fwname_data[] = {
 	{ BCM4354_CHIP_ID, 0xFFFFFFFF, BRCMF_FIRMWARE_NVRAM(BCM4354) }
 };
 
-
-static const struct firmware *brcmf_sdio_get_fw(struct brcmf_sdio *bus,
-						  enum brcmf_firmware_type type)
+static const char *brcmf_sdio_get_fwname(struct brcmf_chip *ci,
+					 enum brcmf_firmware_type type)
 {
-	const struct firmware *fw;
-	const char *name;
-	int err, i;
+	int i;
 
 	for (i = 0; i < ARRAY_SIZE(brcmf_fwname_data); i++) {
-		if (brcmf_fwname_data[i].chipid == bus->ci->chip &&
-		    brcmf_fwname_data[i].revmsk & BIT(bus->ci->chiprev)) {
+		if (brcmf_fwname_data[i].chipid == ci->chip &&
+		    brcmf_fwname_data[i].revmsk & BIT(ci->chiprev)) {
 			switch (type) {
 			case BRCMF_FIRMWARE_BIN:
-				name = brcmf_fwname_data[i].bin;
-				break;
+				return brcmf_fwname_data[i].bin;
 			case BRCMF_FIRMWARE_NVRAM:
-				name = brcmf_fwname_data[i].nv;
-				break;
+				return brcmf_fwname_data[i].nv;
 			default:
 				brcmf_err("invalid firmware type (%d)\n", type);
 				return NULL;
 			}
-			goto found;
 		}
 	}
 	brcmf_err("Unknown chipid %d [%d]\n",
-		  bus->ci->chip, bus->ci->chiprev);
+		  ci->chip, ci->chiprev);
 	return NULL;
-
-found:
-	err = request_firmware(&fw, name, &bus->sdiodev->func[2]->dev);
-	if ((err) || (!fw)) {
-		brcmf_err("fail to request firmware %s (%d)\n", name, err);
-		return NULL;
-	}
-
-	return fw;
 }
 
 static void pkt_align(struct sk_buff *p, int len, int align)
@@ -3278,20 +3263,13 @@ static int brcmf_sdio_download_code_file(struct brcmf_sdio *bus,
 }
 
 static int brcmf_sdio_download_nvram(struct brcmf_sdio *bus,
-				     const struct firmware *nv)
+				     void *vars, u32 varsz)
 {
-	void *vars;
-	u32 varsz;
 	int address;
 	int err;
 
 	brcmf_dbg(TRACE, "Enter\n");
 
-	vars = brcmf_nvram_strip(nv, &varsz);
-
-	if (vars == NULL)
-		return -EINVAL;
-
 	address = bus->ci->ramsize - varsz + bus->ci->rambase;
 	err = brcmf_sdiod_ramrw(bus->sdiodev, true, address, vars, varsz);
 	if (err)
@@ -3300,15 +3278,14 @@ static int brcmf_sdio_download_nvram(struct brcmf_sdio *bus,
 	else if (!brcmf_sdio_verifymemory(bus->sdiodev, address, vars, varsz))
 		err = -EIO;
 
-	brcmf_nvram_free(vars);
-
 	return err;
 }
 
-static int brcmf_sdio_download_firmware(struct brcmf_sdio *bus)
+static int brcmf_sdio_download_firmware(struct brcmf_sdio *bus,
+					const struct firmware *fw,
+					void *nvram, u32 nvlen)
 {
 	int bcmerror = -EFAULT;
-	const struct firmware *fw;
 	u32 rstvec;
 
 	sdio_claim_host(bus->sdiodev->func[1]);
@@ -3317,12 +3294,6 @@ static int brcmf_sdio_download_firmware(struct brcmf_sdio *bus)
 	/* Keep arm in reset */
 	brcmf_chip_enter_download(bus->ci);
 
-	fw = brcmf_sdio_get_fw(bus, BRCMF_FIRMWARE_BIN);
-	if (fw == NULL) {
-		bcmerror = -ENOENT;
-		goto err;
-	}
-
 	rstvec = get_unaligned_le32(fw->data);
 	brcmf_dbg(SDIO, "firmware rstvec: %x\n", rstvec);
 
@@ -3330,17 +3301,12 @@ static int brcmf_sdio_download_firmware(struct brcmf_sdio *bus)
 	release_firmware(fw);
 	if (bcmerror) {
 		brcmf_err("dongle image file download failed\n");
+		brcmf_fw_nvram_free(nvram);
 		goto err;
 	}
 
-	fw = brcmf_sdio_get_fw(bus, BRCMF_FIRMWARE_NVRAM);
-	if (fw == NULL) {
-		bcmerror = -ENOENT;
-		goto err;
-	}
-
-	bcmerror = brcmf_sdio_download_nvram(bus, fw);
-	release_firmware(fw);
+	bcmerror = brcmf_sdio_download_nvram(bus, nvram, nvlen);
+	brcmf_fw_nvram_free(nvram);
 	if (bcmerror) {
 		brcmf_err("dongle nvram file download failed\n");
 		goto err;
@@ -3490,97 +3456,6 @@ done:
 	return err;
 }
 
-static int brcmf_sdio_bus_init(struct device *dev)
-{
-	struct brcmf_bus *bus_if = dev_get_drvdata(dev);
-	struct brcmf_sdio_dev *sdiodev = bus_if->bus_priv.sdio;
-	struct brcmf_sdio *bus = sdiodev->bus;
-	int err, ret = 0;
-	u8 saveclk;
-
-	brcmf_dbg(TRACE, "Enter\n");
-
-	/* try to download image and nvram to the dongle */
-	if (bus_if->state == BRCMF_BUS_DOWN) {
-		bus->alp_only = true;
-		err = brcmf_sdio_download_firmware(bus);
-		if (err)
-			return err;
-		bus->alp_only = false;
-	}
-
-	if (!bus->sdiodev->bus_if->drvr)
-		return 0;
-
-	/* Start the watchdog timer */
-	bus->sdcnt.tickcnt = 0;
-	brcmf_sdio_wd_timer(bus, BRCMF_WD_POLL_MS);
-
-	sdio_claim_host(bus->sdiodev->func[1]);
-
-	/* Make sure backplane clock is on, needed to generate F2 interrupt */
-	brcmf_sdio_clkctl(bus, CLK_AVAIL, false);
-	if (bus->clkstate != CLK_AVAIL)
-		goto exit;
-
-	/* Force clocks on backplane to be sure F2 interrupt propagates */
-	saveclk = brcmf_sdiod_regrb(bus->sdiodev,
-				    SBSDIO_FUNC1_CHIPCLKCSR, &err);
-	if (!err) {
-		brcmf_sdiod_regwb(bus->sdiodev, SBSDIO_FUNC1_CHIPCLKCSR,
-				  (saveclk | SBSDIO_FORCE_HT), &err);
-	}
-	if (err) {
-		brcmf_err("Failed to force clock for F2: err %d\n", err);
-		goto exit;
-	}
-
-	/* Enable function 2 (frame transfers) */
-	w_sdreg32(bus, SDPCM_PROT_VERSION << SMB_DATA_VERSION_SHIFT,
-		  offsetof(struct sdpcmd_regs, tosbmailboxdata));
-	err = sdio_enable_func(bus->sdiodev->func[SDIO_FUNC_2]);
-
-
-	brcmf_dbg(INFO, "enable F2: err=%d\n", err);
-
-	/* If F2 successfully enabled, set core and enable interrupts */
-	if (!err) {
-		/* Set up the interrupt mask and enable interrupts */
-		bus->hostintmask = HOSTINTMASK;
-		w_sdreg32(bus, bus->hostintmask,
-			  offsetof(struct sdpcmd_regs, hostintmask));
-
-		brcmf_sdiod_regwb(bus->sdiodev, SBSDIO_WATERMARK, 8, &err);
-	} else {
-		/* Disable F2 again */
-		sdio_disable_func(bus->sdiodev->func[SDIO_FUNC_2]);
-		ret = -ENODEV;
-	}
-
-	if (brcmf_chip_sr_capable(bus->ci)) {
-		brcmf_sdio_sr_init(bus);
-	} else {
-		/* Restore previous clock setting */
-		brcmf_sdiod_regwb(bus->sdiodev, SBSDIO_FUNC1_CHIPCLKCSR,
-				  saveclk, &err);
-	}
-
-	if (ret == 0) {
-		ret = brcmf_sdiod_intr_register(bus->sdiodev);
-		if (ret != 0)
-			brcmf_err("intr register failed:%d\n", ret);
-	}
-
-	/* If we didn't come up, turn off backplane clock */
-	if (ret != 0)
-		brcmf_sdio_clkctl(bus, CLK_NONE, false);
-
-exit:
-	sdio_release_host(bus->sdiodev->func[1]);
-
-	return ret;
-}
-
 void brcmf_sdio_isr(struct brcmf_sdio *bus)
 {
 	brcmf_dbg(TRACE, "Enter\n");
@@ -4020,13 +3895,114 @@ brcmf_sdio_watchdog(unsigned long data)
 static struct brcmf_bus_ops brcmf_sdio_bus_ops = {
 	.stop = brcmf_sdio_bus_stop,
 	.preinit = brcmf_sdio_bus_preinit,
-	.init = brcmf_sdio_bus_init,
 	.txdata = brcmf_sdio_bus_txdata,
 	.txctl = brcmf_sdio_bus_txctl,
 	.rxctl = brcmf_sdio_bus_rxctl,
 	.gettxq = brcmf_sdio_bus_gettxq,
 };
 
+static void brcmf_sdio_firmware_callback(struct device *dev,
+					 const struct firmware *code,
+					 void *nvram, u32 nvram_len)
+{
+	struct brcmf_bus *bus_if = dev_get_drvdata(dev);
+	struct brcmf_sdio_dev *sdiodev = bus_if->bus_priv.sdio;
+	struct brcmf_sdio *bus = sdiodev->bus;
+	int err = 0;
+	u8 saveclk;
+
+	brcmf_dbg(TRACE, "Enter: dev=%s\n", dev_name(dev));
+
+	/* try to download image and nvram to the dongle */
+	if (bus_if->state == BRCMF_BUS_DOWN) {
+		bus->alp_only = true;
+		err = brcmf_sdio_download_firmware(bus, code, nvram, nvram_len);
+		if (err)
+			goto fail;
+		bus->alp_only = false;
+	}
+
+	if (!bus_if->drvr)
+		return;
+
+	/* Start the watchdog timer */
+	bus->sdcnt.tickcnt = 0;
+	brcmf_sdio_wd_timer(bus, BRCMF_WD_POLL_MS);
+
+	sdio_claim_host(sdiodev->func[1]);
+
+	/* Make sure backplane clock is on, needed to generate F2 interrupt */
+	brcmf_sdio_clkctl(bus, CLK_AVAIL, false);
+	if (bus->clkstate != CLK_AVAIL)
+		goto release;
+
+	/* Force clocks on backplane to be sure F2 interrupt propagates */
+	saveclk = brcmf_sdiod_regrb(sdiodev, SBSDIO_FUNC1_CHIPCLKCSR, &err);
+	if (!err) {
+		brcmf_sdiod_regwb(sdiodev, SBSDIO_FUNC1_CHIPCLKCSR,
+				  (saveclk | SBSDIO_FORCE_HT), &err);
+	}
+	if (err) {
+		brcmf_err("Failed to force clock for F2: err %d\n", err);
+		goto release;
+	}
+
+	/* Enable function 2 (frame transfers) */
+	w_sdreg32(bus, SDPCM_PROT_VERSION << SMB_DATA_VERSION_SHIFT,
+		  offsetof(struct sdpcmd_regs, tosbmailboxdata));
+	err = sdio_enable_func(sdiodev->func[SDIO_FUNC_2]);
+
+
+	brcmf_dbg(INFO, "enable F2: err=%d\n", err);
+
+	/* If F2 successfully enabled, set core and enable interrupts */
+	if (!err) {
+		/* Set up the interrupt mask and enable interrupts */
+		bus->hostintmask = HOSTINTMASK;
+		w_sdreg32(bus, bus->hostintmask,
+			  offsetof(struct sdpcmd_regs, hostintmask));
+
+		brcmf_sdiod_regwb(sdiodev, SBSDIO_WATERMARK, 8, &err);
+	} else {
+		/* Disable F2 again */
+		sdio_disable_func(sdiodev->func[SDIO_FUNC_2]);
+		goto release;
+	}
+
+	if (brcmf_chip_sr_capable(bus->ci)) {
+		brcmf_sdio_sr_init(bus);
+	} else {
+		/* Restore previous clock setting */
+		brcmf_sdiod_regwb(sdiodev, SBSDIO_FUNC1_CHIPCLKCSR,
+				  saveclk, &err);
+	}
+
+	if (err == 0) {
+		err = brcmf_sdiod_intr_register(sdiodev);
+		if (err != 0)
+			brcmf_err("intr register failed:%d\n", err);
+	}
+
+	/* If we didn't come up, turn off backplane clock */
+	if (err != 0)
+		brcmf_sdio_clkctl(bus, CLK_NONE, false);
+
+	sdio_release_host(sdiodev->func[1]);
+
+	err = brcmf_bus_start(dev);
+	if (err != 0) {
+		brcmf_err("dongle is not responding\n");
+		goto fail;
+	}
+	return;
+
+release:
+	sdio_release_host(sdiodev->func[1]);
+fail:
+	brcmf_dbg(TRACE, "failed: dev=%s, err=%d\n", dev_name(dev), err);
+	device_release_driver(dev);
+}
+
 struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 {
 	int ret;
@@ -4110,8 +4086,13 @@ struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 		goto fail;
 	}
 
+	/* Query the F2 block size, set roundup accordingly */
+	bus->blocksize = bus->sdiodev->func[2]->cur_blksize;
+	bus->roundup = min(max_roundup, bus->blocksize);
+
 	/* Allocate buffers */
 	if (bus->sdiodev->bus_if->maxctl) {
+		bus->sdiodev->bus_if->maxctl += bus->roundup;
 		bus->rxblen =
 		    roundup((bus->sdiodev->bus_if->maxctl + SDPCM_HDRLEN),
 			    ALIGNMENT) + bus->head_align;
@@ -4139,10 +4120,6 @@ struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 	bus->idletime = BRCMF_IDLE_INTERVAL;
 	bus->idleclock = BRCMF_IDLE_ACTIVE;
 
-	/* Query the F2 block size, set roundup accordingly */
-	bus->blocksize = bus->sdiodev->func[2]->cur_blksize;
-	bus->roundup = min(max_roundup, bus->blocksize);
-
 	/* SR state */
 	bus->sleeping = false;
 	bus->sr_enabled = false;
@@ -4150,10 +4127,14 @@ struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 	brcmf_sdio_debugfs_create(bus);
 	brcmf_dbg(INFO, "completed!!\n");
 
-	/* if firmware path present try to download and bring up bus */
-	ret = brcmf_bus_start(bus->sdiodev->dev);
+	ret = brcmf_fw_get_firmwares(sdiodev->dev, BRCMF_FW_REQUEST_NVRAM,
+				     brcmf_sdio_get_fwname(bus->ci,
+							   BRCMF_FIRMWARE_BIN),
+				     brcmf_sdio_get_fwname(bus->ci,
+							   BRCMF_FIRMWARE_NVRAM),
+				     brcmf_sdio_firmware_callback);
 	if (ret != 0) {
-		brcmf_err("dongle is not responding\n");
+		brcmf_err("async firmware request failed: %d\n", ret);
 		goto fail;
 	}
 
@@ -4173,9 +4154,7 @@ void brcmf_sdio_remove(struct brcmf_sdio *bus)
 		/* De-register interrupt handler */
 		brcmf_sdiod_intr_unregister(bus->sdiodev);
 
-		if (bus->sdiodev->bus_if->drvr) {
-			brcmf_detach(bus->sdiodev->dev);
-		}
+		brcmf_detach(bus->sdiodev->dev);
 
 		cancel_work_sync(&bus->datawork);
 		if (bus->brcmf_wq)
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/firmware.c b/drivers/net/wireless/brcm80211/brcmfmac/firmware.c
new file mode 100644
index 000000000000..7b7d237c1ddb
--- /dev/null
+++ b/drivers/net/wireless/brcm80211/brcmfmac/firmware.c
@@ -0,0 +1,332 @@
+/*
+ * Copyright (c) 2013 Broadcom Corporation
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
+ * SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
+ * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
+ * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/device.h>
+#include <linux/firmware.h>
+
+#include "dhd_dbg.h"
+#include "firmware.h"
+
+enum nvram_parser_state {
+	IDLE,
+	KEY,
+	VALUE,
+	COMMENT,
+	END
+};
+
+/**
+ * struct nvram_parser - internal info for parser.
+ *
+ * @state: current parser state.
+ * @fwnv: input buffer being parsed.
+ * @nvram: output buffer with parse result.
+ * @nvram_len: lenght of parse result.
+ * @line: current line.
+ * @column: current column in line.
+ * @pos: byte offset in input buffer.
+ * @entry: start position of key,value entry.
+ */
+struct nvram_parser {
+	enum nvram_parser_state state;
+	const struct firmware *fwnv;
+	u8 *nvram;
+	u32 nvram_len;
+	u32 line;
+	u32 column;
+	u32 pos;
+	u32 entry;
+};
+
+static bool is_nvram_char(char c)
+{
+	/* comment marker excluded */
+	if (c == '#')
+		return false;
+
+	/* key and value may have any other readable character */
+	return (c > 0x20 && c < 0x7f);
+}
+
+static bool is_whitespace(char c)
+{
+	return (c == ' ' || c == '\r' || c == '\n' || c == '\t');
+}
+
+static enum nvram_parser_state brcmf_nvram_handle_idle(struct nvram_parser *nvp)
+{
+	char c;
+
+	c = nvp->fwnv->data[nvp->pos];
+	if (c == '\n')
+		return COMMENT;
+	if (is_whitespace(c))
+		goto proceed;
+	if (c == '#')
+		return COMMENT;
+	if (is_nvram_char(c)) {
+		nvp->entry = nvp->pos;
+		return KEY;
+	}
+	brcmf_dbg(INFO, "warning: ln=%d:col=%d: ignoring invalid character\n",
+		  nvp->line, nvp->column);
+proceed:
+	nvp->column++;
+	nvp->pos++;
+	return IDLE;
+}
+
+static enum nvram_parser_state brcmf_nvram_handle_key(struct nvram_parser *nvp)
+{
+	enum nvram_parser_state st = nvp->state;
+	char c;
+
+	c = nvp->fwnv->data[nvp->pos];
+	if (c == '=') {
+		st = VALUE;
+	} else if (!is_nvram_char(c)) {
+		brcmf_dbg(INFO, "warning: ln=%d:col=%d: '=' expected, skip invalid key entry\n",
+			  nvp->line, nvp->column);
+		return COMMENT;
+	}
+
+	nvp->column++;
+	nvp->pos++;
+	return st;
+}
+
+static enum nvram_parser_state
+brcmf_nvram_handle_value(struct nvram_parser *nvp)
+{
+	char c;
+	char *skv;
+	char *ekv;
+	u32 cplen;
+
+	c = nvp->fwnv->data[nvp->pos];
+	if (!is_nvram_char(c)) {
+		/* key,value pair complete */
+		ekv = (u8 *)&nvp->fwnv->data[nvp->pos];
+		skv = (u8 *)&nvp->fwnv->data[nvp->entry];
+		cplen = ekv - skv;
+		/* copy to output buffer */
+		memcpy(&nvp->nvram[nvp->nvram_len], skv, cplen);
+		nvp->nvram_len += cplen;
+		nvp->nvram[nvp->nvram_len] = '\0';
+		nvp->nvram_len++;
+		return IDLE;
+	}
+	nvp->pos++;
+	nvp->column++;
+	return VALUE;
+}
+
+static enum nvram_parser_state
+brcmf_nvram_handle_comment(struct nvram_parser *nvp)
+{
+	char *eol, *sol;
+
+	sol = (char *)&nvp->fwnv->data[nvp->pos];
+	eol = strchr(sol, '\n');
+	if (eol == NULL)
+		return END;
+
+	/* eat all moving to next line */
+	nvp->line++;
+	nvp->column = 1;
+	nvp->pos += (eol - sol) + 1;
+	return IDLE;
+}
+
+static enum nvram_parser_state brcmf_nvram_handle_end(struct nvram_parser *nvp)
+{
+	/* final state */
+	return END;
+}
+
+static enum nvram_parser_state
+(*nv_parser_states[])(struct nvram_parser *nvp) = {
+	brcmf_nvram_handle_idle,
+	brcmf_nvram_handle_key,
+	brcmf_nvram_handle_value,
+	brcmf_nvram_handle_comment,
+	brcmf_nvram_handle_end
+};
+
+static int brcmf_init_nvram_parser(struct nvram_parser *nvp,
+				   const struct firmware *nv)
+{
+	memset(nvp, 0, sizeof(*nvp));
+	nvp->fwnv = nv;
+	/* Alloc for extra 0 byte + roundup by 4 + length field */
+	nvp->nvram = kzalloc(nv->size + 1 + 3 + sizeof(u32), GFP_KERNEL);
+	if (!nvp->nvram)
+		return -ENOMEM;
+
+	nvp->line = 1;
+	nvp->column = 1;
+	return 0;
+}
+
+/* brcmf_nvram_strip :Takes a buffer of "<var>=<value>\n" lines read from a fil
+ * and ending in a NUL. Removes carriage returns, empty lines, comment lines,
+ * and converts newlines to NULs. Shortens buffer as needed and pads with NULs.
+ * End of buffer is completed with token identifying length of buffer.
+ */
+static void *brcmf_fw_nvram_strip(const struct firmware *nv, u32 *new_length)
+{
+	struct nvram_parser nvp;
+	u32 pad;
+	u32 token;
+	__le32 token_le;
+
+	if (brcmf_init_nvram_parser(&nvp, nv) < 0)
+		return NULL;
+
+	while (nvp.pos < nv->size) {
+		nvp.state = nv_parser_states[nvp.state](&nvp);
+		if (nvp.state == END)
+			break;
+	}
+	pad = nvp.nvram_len;
+	*new_length = roundup(nvp.nvram_len + 1, 4);
+	while (pad != *new_length) {
+		nvp.nvram[pad] = 0;
+		pad++;
+	}
+
+	token = *new_length / 4;
+	token = (~token << 16) | (token & 0x0000FFFF);
+	token_le = cpu_to_le32(token);
+
+	memcpy(&nvp.nvram[*new_length], &token_le, sizeof(token_le));
+	*new_length += sizeof(token_le);
+
+	return nvp.nvram;
+}
+
+void brcmf_fw_nvram_free(void *nvram)
+{
+	kfree(nvram);
+}
+
+struct brcmf_fw {
+	struct device *dev;
+	u16 flags;
+	const struct firmware *code;
+	const char *nvram_name;
+	void (*done)(struct device *dev, const struct firmware *fw,
+		     void *nvram_image, u32 nvram_len);
+};
+
+static void brcmf_fw_request_nvram_done(const struct firmware *fw, void *ctx)
+{
+	struct brcmf_fw *fwctx = ctx;
+	u32 nvram_length = 0;
+	void *nvram = NULL;
+
+	brcmf_dbg(TRACE, "enter: dev=%s\n", dev_name(fwctx->dev));
+	if (!fw && !(fwctx->flags & BRCMF_FW_REQ_NV_OPTIONAL))
+		goto fail;
+
+	if (fw) {
+		nvram = brcmf_fw_nvram_strip(fw, &nvram_length);
+		release_firmware(fw);
+		if (!nvram && !(fwctx->flags & BRCMF_FW_REQ_NV_OPTIONAL))
+			goto fail;
+	}
+
+	fwctx->done(fwctx->dev, fwctx->code, nvram, nvram_length);
+	kfree(fwctx);
+	return;
+
+fail:
+	brcmf_dbg(TRACE, "failed: dev=%s\n", dev_name(fwctx->dev));
+	if (fwctx->code)
+		release_firmware(fwctx->code);
+	device_release_driver(fwctx->dev);
+	kfree(fwctx);
+}
+
+static void brcmf_fw_request_code_done(const struct firmware *fw, void *ctx)
+{
+	struct brcmf_fw *fwctx = ctx;
+	int ret;
+
+	brcmf_dbg(TRACE, "enter: dev=%s\n", dev_name(fwctx->dev));
+	if (!fw)
+		goto fail;
+
+	/* only requested code so done here */
+	if (!(fwctx->flags & BRCMF_FW_REQUEST_NVRAM)) {
+		fwctx->done(fwctx->dev, fw, NULL, 0);
+		kfree(fwctx);
+		return;
+	}
+	fwctx->code = fw;
+	ret = request_firmware_nowait(THIS_MODULE, true, fwctx->nvram_name,
+				      fwctx->dev, GFP_KERNEL, fwctx,
+				      brcmf_fw_request_nvram_done);
+
+	if (!ret)
+		return;
+
+	/* when nvram is optional call .done() callback here */
+	if (fwctx->flags & BRCMF_FW_REQ_NV_OPTIONAL) {
+		fwctx->done(fwctx->dev, fw, NULL, 0);
+		kfree(fwctx);
+		return;
+	}
+
+	/* failed nvram request */
+	release_firmware(fw);
+fail:
+	brcmf_dbg(TRACE, "failed: dev=%s\n", dev_name(fwctx->dev));
+	device_release_driver(fwctx->dev);
+	kfree(fwctx);
+}
+
+int brcmf_fw_get_firmwares(struct device *dev, u16 flags,
+			   const char *code, const char *nvram,
+			   void (*fw_cb)(struct device *dev,
+					 const struct firmware *fw,
+					 void *nvram_image, u32 nvram_len))
+{
+	struct brcmf_fw *fwctx;
+
+	brcmf_dbg(TRACE, "enter: dev=%s\n", dev_name(dev));
+	if (!fw_cb || !code)
+		return -EINVAL;
+
+	if ((flags & BRCMF_FW_REQUEST_NVRAM) && !nvram)
+		return -EINVAL;
+
+	fwctx = kzalloc(sizeof(*fwctx), GFP_KERNEL);
+	if (!fwctx)
+		return -ENOMEM;
+
+	fwctx->dev = dev;
+	fwctx->flags = flags;
+	fwctx->done = fw_cb;
+	if (flags & BRCMF_FW_REQUEST_NVRAM)
+		fwctx->nvram_name = nvram;
+
+	return request_firmware_nowait(THIS_MODULE, true, code, dev,
+				       GFP_KERNEL, fwctx,
+				       brcmf_fw_request_code_done);
+}
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/nvram.h b/drivers/net/wireless/brcm80211/brcmfmac/firmware.h
index d454580928c9..6431bfd7afff 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/nvram.h
+++ b/drivers/net/wireless/brcm80211/brcmfmac/firmware.h
@@ -13,12 +13,24 @@
  * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
  * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
-#ifndef BRCMFMAC_NVRAM_H
-#define BRCMFMAC_NVRAM_H
+#ifndef BRCMFMAC_FIRMWARE_H
+#define BRCMFMAC_FIRMWARE_H
 
+#define BRCMF_FW_REQUEST		0x000F
+#define  BRCMF_FW_REQUEST_NVRAM		0x0001
+#define BRCMF_FW_REQ_FLAGS		0x00F0
+#define  BRCMF_FW_REQ_NV_OPTIONAL	0x0010
 
-void *brcmf_nvram_strip(const struct firmware *nv, u32 *new_length);
-void brcmf_nvram_free(void *nvram);
-
+void brcmf_fw_nvram_free(void *nvram);
+/*
+ * Request firmware(s) asynchronously. When the asynchronous request
+ * fails it will not use the callback, but call device_release_driver()
+ * instead which will call the driver .remove() callback.
+ */
+int brcmf_fw_get_firmwares(struct device *dev, u16 flags,
+			   const char *code, const char *nvram,
+			   void (*fw_cb)(struct device *dev,
+					 const struct firmware *fw,
+					 void *nvram_image, u32 nvram_len));
 
-#endif /* BRCMFMAC_NVRAM_H */
+#endif /* BRCMFMAC_FIRMWARE_H */
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/brcm80211/brcmfmac/fwil_types.h
index 614e4888504f..2bc68a2137fc 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/brcm80211/brcmfmac/fwil_types.h
@@ -53,6 +53,14 @@
 #define BRCMF_OBSS_COEX_OFF		0
 #define BRCMF_OBSS_COEX_ON		1
 
+/* join preference types for join_pref iovar */
+enum brcmf_join_pref_types {
+	BRCMF_JOIN_PREF_RSSI = 1,
+	BRCMF_JOIN_PREF_WPA,
+	BRCMF_JOIN_PREF_BAND,
+	BRCMF_JOIN_PREF_RSSI_DELTA,
+};
+
 enum brcmf_fil_p2p_if_types {
 	BRCMF_FIL_P2P_IF_CLIENT,
 	BRCMF_FIL_P2P_IF_GO,
@@ -282,6 +290,22 @@ struct brcmf_assoc_params_le {
 	__le16 chanspec_list[1];
 };
 
+/**
+ * struct join_pref params - parameters for preferred join selection.
+ *
+ * @type: preference type (see enum brcmf_join_pref_types).
+ * @len: length of bytes following (currently always 2).
+ * @rssi_gain: signal gain for selection (only when @type is RSSI_DELTA).
+ * @band: band to which selection preference applies.
+ *	This is used if @type is BAND or RSSI_DELTA.
+ */
+struct brcmf_join_pref_params {
+	u8 type;
+	u8 len;
+	u8 rssi_gain;
+	u8 band;
+};
+
 /* used for join with or without a specific bssid and channel list */
 struct brcmf_join_params {
 	struct brcmf_ssid_le ssid_le;
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/fwsignal.c b/drivers/net/wireless/brcm80211/brcmfmac/fwsignal.c
index c3e7d76dbf35..699908de314a 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/fwsignal.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/fwsignal.c
@@ -476,6 +476,7 @@ struct brcmf_fws_info {
 	bool bus_flow_blocked;
 	bool creditmap_received;
 	u8 mode;
+	bool avoid_queueing;
 };
 
 /*
@@ -1369,13 +1370,12 @@ done:
 }
 
 static int brcmf_fws_txstatus_suppressed(struct brcmf_fws_info *fws, int fifo,
-					 struct sk_buff *skb, u32 genbit,
-					 u16 seq)
+					 struct sk_buff *skb, u8 ifidx,
+					 u32 genbit, u16 seq)
 {
 	struct brcmf_fws_mac_descriptor *entry = brcmf_skbcb(skb)->mac;
 	u32 hslot;
 	int ret;
-	u8 ifidx;
 
 	hslot = brcmf_skb_htod_tag_get_field(skb, HSLOT);
 
@@ -1389,29 +1389,21 @@ static int brcmf_fws_txstatus_suppressed(struct brcmf_fws_info *fws, int fifo,
 
 	entry->generation = genbit;
 
-	ret = brcmf_proto_hdrpull(fws->drvr, false, &ifidx, skb);
-	if (ret == 0) {
-		brcmf_skb_htod_tag_set_field(skb, GENERATION, genbit);
-		brcmf_skbcb(skb)->htod_seq = seq;
-		if (brcmf_skb_htod_seq_get_field(skb, FROMFW)) {
-			brcmf_skb_htod_seq_set_field(skb, FROMDRV, 1);
-			brcmf_skb_htod_seq_set_field(skb, FROMFW, 0);
-		} else {
-			brcmf_skb_htod_seq_set_field(skb, FROMDRV, 0);
-		}
-		ret = brcmf_fws_enq(fws, BRCMF_FWS_SKBSTATE_SUPPRESSED, fifo,
-				    skb);
+	brcmf_skb_htod_tag_set_field(skb, GENERATION, genbit);
+	brcmf_skbcb(skb)->htod_seq = seq;
+	if (brcmf_skb_htod_seq_get_field(skb, FROMFW)) {
+		brcmf_skb_htod_seq_set_field(skb, FROMDRV, 1);
+		brcmf_skb_htod_seq_set_field(skb, FROMFW, 0);
+	} else {
+		brcmf_skb_htod_seq_set_field(skb, FROMDRV, 0);
 	}
+	ret = brcmf_fws_enq(fws, BRCMF_FWS_SKBSTATE_SUPPRESSED, fifo, skb);
 
 	if (ret != 0) {
-		/* suppress q is full or hdrpull failed, drop this packet */
-		brcmf_fws_hanger_poppkt(&fws->hanger, hslot, &skb,
-					true);
+		/* suppress q is full drop this packet */
+		brcmf_fws_hanger_poppkt(&fws->hanger, hslot, &skb, true);
 	} else {
-		/*
-		 * Mark suppressed to avoid a double free during
-		 * wlfc cleanup
-		 */
+		/* Mark suppressed to avoid a double free during wlfc cleanup */
 		brcmf_fws_hanger_mark_suppressed(&fws->hanger, hslot);
 	}
 
@@ -1428,6 +1420,7 @@ brcmf_fws_txs_process(struct brcmf_fws_info *fws, u8 flags, u32 hslot,
 	struct sk_buff *skb;
 	struct brcmf_skbuff_cb *skcb;
 	struct brcmf_fws_mac_descriptor *entry = NULL;
+	u8 ifidx;
 
 	brcmf_dbg(DATA, "flags %d\n", flags);
 
@@ -1476,12 +1469,15 @@ brcmf_fws_txs_process(struct brcmf_fws_info *fws, u8 flags, u32 hslot,
 	}
 	brcmf_fws_macdesc_return_req_credit(skb);
 
+	if (brcmf_proto_hdrpull(fws->drvr, false, &ifidx, skb)) {
+		brcmu_pkt_buf_free_skb(skb);
+		return -EINVAL;
+	}
 	if (!remove_from_hanger)
-		ret = brcmf_fws_txstatus_suppressed(fws, fifo, skb, genbit,
-						    seq);
-
+		ret = brcmf_fws_txstatus_suppressed(fws, fifo, skb, ifidx,
+						    genbit, seq);
 	if (remove_from_hanger || ret)
-		brcmf_txfinalize(fws->drvr, skb, true);
+		brcmf_txfinalize(fws->drvr, skb, ifidx, true);
 
 	return 0;
 }
@@ -1868,7 +1864,7 @@ int brcmf_fws_process_skb(struct brcmf_if *ifp, struct sk_buff *skb)
 	struct ethhdr *eh = (struct ethhdr *)(skb->data);
 	int fifo = BRCMF_FWS_FIFO_BCMC;
 	bool multicast = is_multicast_ether_addr(eh->h_dest);
-	bool pae = eh->h_proto == htons(ETH_P_PAE);
+	int rc = 0;
 
 	brcmf_dbg(DATA, "tx proto=0x%X\n", ntohs(eh->h_proto));
 	/* determine the priority */
@@ -1876,8 +1872,13 @@ int brcmf_fws_process_skb(struct brcmf_if *ifp, struct sk_buff *skb)
 		skb->priority = cfg80211_classify8021d(skb, NULL);
 
 	drvr->tx_multicast += !!multicast;
-	if (pae)
-		atomic_inc(&ifp->pend_8021x_cnt);
+
+	if (fws->avoid_queueing) {
+		rc = brcmf_proto_txdata(drvr, ifp->ifidx, 0, skb);
+		if (rc < 0)
+			brcmf_txfinalize(drvr, skb, ifp->ifidx, false);
+		return rc;
+	}
 
 	/* set control buffer information */
 	skcb->if_flags = 0;
@@ -1899,15 +1900,12 @@ int brcmf_fws_process_skb(struct brcmf_if *ifp, struct sk_buff *skb)
 		brcmf_fws_schedule_deq(fws);
 	} else {
 		brcmf_err("drop skb: no hanger slot\n");
-		if (pae) {
-			atomic_dec(&ifp->pend_8021x_cnt);
-			if (waitqueue_active(&ifp->pend_8021x_wait))
-				wake_up(&ifp->pend_8021x_wait);
-		}
-		brcmu_pkt_buf_free_skb(skb);
+		brcmf_txfinalize(drvr, skb, ifp->ifidx, false);
+		rc = -ENOMEM;
 	}
 	brcmf_fws_unlock(fws);
-	return 0;
+
+	return rc;
 }
 
 void brcmf_fws_reset_interface(struct brcmf_if *ifp)
@@ -1982,7 +1980,8 @@ static void brcmf_fws_dequeue_worker(struct work_struct *worker)
 				ret = brcmf_proto_txdata(drvr, ifidx, 0, skb);
 				brcmf_fws_lock(fws);
 				if (ret < 0)
-					brcmf_txfinalize(drvr, skb, false);
+					brcmf_txfinalize(drvr, skb, ifidx,
+							 false);
 				if (fws->bus_flow_blocked)
 					break;
 			}
@@ -2039,6 +2038,13 @@ int brcmf_fws_init(struct brcmf_pub *drvr)
 	fws->drvr = drvr;
 	fws->fcmode = fcmode;
 
+	if ((drvr->bus_if->always_use_fws_queue == false) &&
+	    (fcmode == BRCMF_FWS_FCMODE_NONE)) {
+		fws->avoid_queueing = true;
+		brcmf_dbg(INFO, "FWS queueing will be avoided\n");
+		return 0;
+	}
+
 	fws->fws_wq = create_singlethread_workqueue("brcmf_fws_wq");
 	if (fws->fws_wq == NULL) {
 		brcmf_err("workqueue creation failed\n");
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/nvram.c b/drivers/net/wireless/brcm80211/brcmfmac/nvram.c
deleted file mode 100644
index d5ef86db631b..000000000000
--- a/drivers/net/wireless/brcm80211/brcmfmac/nvram.c
+++ /dev/null
@@ -1,94 +0,0 @@
-/*
- * Copyright (c) 2013 Broadcom Corporation
- *
- * Permission to use, copy, modify, and/or distribute this software for any
- * purpose with or without fee is hereby granted, provided that the above
- * copyright notice and this permission notice appear in all copies.
- *
- * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
- * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
- * SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
- * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
- * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
- */
-
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/firmware.h>
-
-#include "nvram.h"
-
-/* brcmf_nvram_strip :Takes a buffer of "<var>=<value>\n" lines read from a file
- * and ending in a NUL. Removes carriage returns, empty lines, comment lines,
- * and converts newlines to NULs. Shortens buffer as needed and pads with NULs.
- * End of buffer is completed with token identifying length of buffer.
- */
-void *brcmf_nvram_strip(const struct firmware *nv, u32 *new_length)
-{
-	u8 *nvram;
-	u32 i;
-	u32 len;
-	u32 column;
-	u8 val;
-	bool comment;
-	u32 token;
-	__le32 token_le;
-
-	/* Alloc for extra 0 byte + roundup by 4 + length field */
-	nvram = kmalloc(nv->size + 1 + 3 + sizeof(token_le), GFP_KERNEL);
-	if (!nvram)
-		return NULL;
-
-	len = 0;
-	column = 0;
-	comment = false;
-	for (i = 0; i < nv->size; i++) {
-		val = nv->data[i];
-		if (val == 0)
-			break;
-		if (val == '\r')
-			continue;
-		if (comment && (val != '\n'))
-			continue;
-		comment = false;
-		if (val == '#') {
-			comment = true;
-			continue;
-		}
-		if (val == '\n') {
-			if (column == 0)
-				continue;
-			nvram[len] = 0;
-			len++;
-			column = 0;
-			continue;
-		}
-		nvram[len] = val;
-		len++;
-		column++;
-	}
-	column = len;
-	*new_length = roundup(len + 1, 4);
-	while (column != *new_length) {
-		nvram[column] = 0;
-		column++;
-	}
-
-	token = *new_length / 4;
-	token = (~token << 16) | (token & 0x0000FFFF);
-	token_le = cpu_to_le32(token);
-
-	memcpy(&nvram[*new_length], &token_le, sizeof(token_le));
-	*new_length += sizeof(token_le);
-
-	return nvram;
-}
-
-void brcmf_nvram_free(void *nvram)
-{
-	kfree(nvram);
-}
-
-
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/brcm80211/brcmfmac/usb.c
index 24f65cd53859..6db51a666f61 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/usb.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/usb.c
@@ -25,6 +25,7 @@
 #include <dhd_bus.h>
 #include <dhd_dbg.h>
 
+#include "firmware.h"
 #include "usb_rdl.h"
 #include "usb.h"
 
@@ -61,12 +62,6 @@ struct brcmf_usb_image {
 	u8 *image;
 	int image_len;
 };
-static struct list_head fw_image_list;
-
-struct intr_transfer_buf {
-	u32 notification;
-	u32 reserved;
-};
 
 struct brcmf_usbdev_info {
 	struct brcmf_usbdev bus_pub; /* MUST BE FIRST */
@@ -75,7 +70,7 @@ struct brcmf_usbdev_info {
 	struct list_head rx_postq;
 	struct list_head tx_freeq;
 	struct list_head tx_postq;
-	uint rx_pipe, tx_pipe, intr_pipe, rx_pipe2;
+	uint rx_pipe, tx_pipe, rx_pipe2;
 
 	int rx_low_watermark;
 	int tx_low_watermark;
@@ -87,7 +82,7 @@ struct brcmf_usbdev_info {
 	struct brcmf_usbreq *tx_reqs;
 	struct brcmf_usbreq *rx_reqs;
 
-	u8 *image;	/* buffer for combine fw and nvram */
+	const u8 *image;	/* buffer for combine fw and nvram */
 	int image_len;
 
 	struct usb_device *usbdev;
@@ -104,10 +99,6 @@ struct brcmf_usbdev_info {
 	ulong ctl_op;
 
 	struct urb *bulk_urb; /* used for FW download */
-	struct urb *intr_urb; /* URB for interrupt endpoint */
-	int intr_size;          /* Size of interrupt message */
-	int interval;           /* Interrupt polling interval */
-	struct intr_transfer_buf intr; /* Data buffer for interrupt endpoint */
 };
 
 static void brcmf_usb_rx_refill(struct brcmf_usbdev_info *devinfo,
@@ -531,39 +522,6 @@ brcmf_usb_state_change(struct brcmf_usbdev_info *devinfo, int state)
 	}
 }
 
-static void
-brcmf_usb_intr_complete(struct urb *urb)
-{
-	struct brcmf_usbdev_info *devinfo =
-			(struct brcmf_usbdev_info *)urb->context;
-	int err;
-
-	brcmf_dbg(USB, "Enter, urb->status=%d\n", urb->status);
-
-	if (devinfo == NULL)
-		return;
-
-	if (unlikely(urb->status)) {
-		if (urb->status == -ENOENT ||
-		    urb->status == -ESHUTDOWN ||
-		    urb->status == -ENODEV) {
-			brcmf_usb_state_change(devinfo,
-					       BRCMFMAC_USB_STATE_DOWN);
-		}
-	}
-
-	if (devinfo->bus_pub.state == BRCMFMAC_USB_STATE_DOWN) {
-		brcmf_err("intr cb when DBUS down, ignoring\n");
-		return;
-	}
-
-	if (devinfo->bus_pub.state == BRCMFMAC_USB_STATE_UP) {
-		err = usb_submit_urb(devinfo->intr_urb, GFP_ATOMIC);
-		if (err)
-			brcmf_err("usb_submit_urb, err=%d\n", err);
-	}
-}
-
 static int brcmf_usb_tx(struct device *dev, struct sk_buff *skb)
 {
 	struct brcmf_usbdev_info *devinfo = brcmf_usb_get_businfo(dev);
@@ -619,7 +577,6 @@ static int brcmf_usb_up(struct device *dev)
 {
 	struct brcmf_usbdev_info *devinfo = brcmf_usb_get_businfo(dev);
 	u16 ifnum;
-	int ret;
 
 	brcmf_dbg(USB, "Enter\n");
 	if (devinfo->bus_pub.state == BRCMFMAC_USB_STATE_UP)
@@ -628,23 +585,6 @@ static int brcmf_usb_up(struct device *dev)
 	/* Success, indicate devinfo is fully up */
 	brcmf_usb_state_change(devinfo, BRCMFMAC_USB_STATE_UP);
 
-	if (devinfo->intr_urb) {
-		usb_fill_int_urb(devinfo->intr_urb, devinfo->usbdev,
-			devinfo->intr_pipe,
-			&devinfo->intr,
-			devinfo->intr_size,
-			(usb_complete_t)brcmf_usb_intr_complete,
-			devinfo,
-			devinfo->interval);
-
-		ret = usb_submit_urb(devinfo->intr_urb, GFP_ATOMIC);
-		if (ret) {
-			brcmf_err("USB_SUBMIT_URB failed with status %d\n",
-				  ret);
-			return -EINVAL;
-		}
-	}
-
 	if (devinfo->ctl_urb) {
 		devinfo->ctl_in_pipe = usb_rcvctrlpipe(devinfo->usbdev, 0);
 		devinfo->ctl_out_pipe = usb_sndctrlpipe(devinfo->usbdev, 0);
@@ -681,8 +621,6 @@ static void brcmf_usb_down(struct device *dev)
 		return;
 
 	brcmf_usb_state_change(devinfo, BRCMFMAC_USB_STATE_DOWN);
-	if (devinfo->intr_urb)
-		usb_kill_urb(devinfo->intr_urb);
 
 	if (devinfo->ctl_urb)
 		usb_kill_urb(devinfo->ctl_urb);
@@ -1021,7 +959,7 @@ brcmf_usb_fw_download(struct brcmf_usbdev_info *devinfo)
 	}
 
 	err = brcmf_usb_dlstart(devinfo,
-		devinfo->image, devinfo->image_len);
+		(u8 *)devinfo->image, devinfo->image_len);
 	if (err == 0)
 		err = brcmf_usb_dlrun(devinfo);
 	return err;
@@ -1036,7 +974,6 @@ static void brcmf_usb_detach(struct brcmf_usbdev_info *devinfo)
 	brcmf_usb_free_q(&devinfo->rx_freeq, false);
 	brcmf_usb_free_q(&devinfo->tx_freeq, false);
 
-	usb_free_urb(devinfo->intr_urb);
 	usb_free_urb(devinfo->ctl_urb);
 	usb_free_urb(devinfo->bulk_urb);
 
@@ -1080,68 +1017,20 @@ static int check_file(const u8 *headers)
 	return -1;
 }
 
-static int brcmf_usb_get_fw(struct brcmf_usbdev_info *devinfo)
+static const char *brcmf_usb_get_fwname(struct brcmf_usbdev_info *devinfo)
 {
-	s8 *fwname;
-	const struct firmware *fw;
-	struct brcmf_usb_image *fw_image;
-	int err;
-
-	brcmf_dbg(USB, "Enter\n");
 	switch (devinfo->bus_pub.devid) {
 	case 43143:
-		fwname = BRCMF_USB_43143_FW_NAME;
-		break;
+		return BRCMF_USB_43143_FW_NAME;
 	case 43235:
 	case 43236:
 	case 43238:
-		fwname = BRCMF_USB_43236_FW_NAME;
-		break;
+		return BRCMF_USB_43236_FW_NAME;
 	case 43242:
-		fwname = BRCMF_USB_43242_FW_NAME;
-		break;
+		return BRCMF_USB_43242_FW_NAME;
 	default:
-		return -EINVAL;
-		break;
-	}
-	brcmf_dbg(USB, "Loading FW %s\n", fwname);
-	list_for_each_entry(fw_image, &fw_image_list, list) {
-		if (fw_image->fwname == fwname) {
-			devinfo->image = fw_image->image;
-			devinfo->image_len = fw_image->image_len;
-			return 0;
-		}
-	}
-	/* fw image not yet loaded. Load it now and add to list */
-	err = request_firmware(&fw, fwname, devinfo->dev);
-	if (!fw) {
-		brcmf_err("fail to request firmware %s\n", fwname);
-		return err;
-	}
-	if (check_file(fw->data) < 0) {
-		brcmf_err("invalid firmware %s\n", fwname);
-		return -EINVAL;
+		return NULL;
 	}
-
-	fw_image = kzalloc(sizeof(*fw_image), GFP_ATOMIC);
-	if (!fw_image)
-		return -ENOMEM;
-	INIT_LIST_HEAD(&fw_image->list);
-	list_add_tail(&fw_image->list, &fw_image_list);
-	fw_image->fwname = fwname;
-	fw_image->image = vmalloc(fw->size);
-	if (!fw_image->image)
-		return -ENOMEM;
-
-	memcpy(fw_image->image, fw->data, fw->size);
-	fw_image->image_len = fw->size;
-
-	release_firmware(fw);
-
-	devinfo->image = fw_image->image;
-	devinfo->image_len = fw_image->image_len;
-
-	return 0;
 }
 
 
@@ -1186,11 +1075,6 @@ struct brcmf_usbdev *brcmf_usb_attach(struct brcmf_usbdev_info *devinfo,
 		goto error;
 	devinfo->tx_freecount = ntxq;
 
-	devinfo->intr_urb = usb_alloc_urb(0, GFP_ATOMIC);
-	if (!devinfo->intr_urb) {
-		brcmf_err("usb_alloc_urb (intr) failed\n");
-		goto error;
-	}
 	devinfo->ctl_urb = usb_alloc_urb(0, GFP_ATOMIC);
 	if (!devinfo->ctl_urb) {
 		brcmf_err("usb_alloc_urb (ctl) failed\n");
@@ -1202,16 +1086,6 @@ struct brcmf_usbdev *brcmf_usb_attach(struct brcmf_usbdev_info *devinfo,
 		goto error;
 	}
 
-	if (!brcmf_usb_dlneeded(devinfo))
-		return &devinfo->bus_pub;
-
-	brcmf_dbg(USB, "Start fw downloading\n");
-	if (brcmf_usb_get_fw(devinfo))
-		goto error;
-
-	if (brcmf_usb_fw_download(devinfo))
-		goto error;
-
 	return &devinfo->bus_pub;
 
 error:
@@ -1222,18 +1096,77 @@ error:
 
 static struct brcmf_bus_ops brcmf_usb_bus_ops = {
 	.txdata = brcmf_usb_tx,
-	.init = brcmf_usb_up,
 	.stop = brcmf_usb_down,
 	.txctl = brcmf_usb_tx_ctlpkt,
 	.rxctl = brcmf_usb_rx_ctlpkt,
 };
 
+static int brcmf_usb_bus_setup(struct brcmf_usbdev_info *devinfo)
+{
+	int ret;
+
+	/* Attach to the common driver interface */
+	ret = brcmf_attach(devinfo->dev);
+	if (ret) {
+		brcmf_err("brcmf_attach failed\n");
+		return ret;
+	}
+
+	ret = brcmf_usb_up(devinfo->dev);
+	if (ret)
+		goto fail;
+
+	ret = brcmf_bus_start(devinfo->dev);
+	if (ret)
+		goto fail;
+
+	return 0;
+fail:
+	brcmf_detach(devinfo->dev);
+	return ret;
+}
+
+static void brcmf_usb_probe_phase2(struct device *dev,
+				   const struct firmware *fw,
+				   void *nvram, u32 nvlen)
+{
+	struct brcmf_bus *bus = dev_get_drvdata(dev);
+	struct brcmf_usbdev_info *devinfo;
+	int ret;
+
+	brcmf_dbg(USB, "Start fw downloading\n");
+	ret = check_file(fw->data);
+	if (ret < 0) {
+		brcmf_err("invalid firmware\n");
+		release_firmware(fw);
+		goto error;
+	}
+
+	devinfo = bus->bus_priv.usb->devinfo;
+	devinfo->image = fw->data;
+	devinfo->image_len = fw->size;
+
+	ret = brcmf_usb_fw_download(devinfo);
+	release_firmware(fw);
+	if (ret)
+		goto error;
+
+	ret = brcmf_usb_bus_setup(devinfo);
+	if (ret)
+		goto error;
+
+	return;
+error:
+	brcmf_dbg(TRACE, "failed: dev=%s, err=%d\n", dev_name(dev), ret);
+	device_release_driver(dev);
+}
+
 static int brcmf_usb_probe_cb(struct brcmf_usbdev_info *devinfo)
 {
 	struct brcmf_bus *bus = NULL;
 	struct brcmf_usbdev *bus_pub = NULL;
-	int ret;
 	struct device *dev = devinfo->dev;
+	int ret;
 
 	brcmf_dbg(USB, "Enter\n");
 	bus_pub = brcmf_usb_attach(devinfo, BRCMF_USB_NRXQ, BRCMF_USB_NTXQ);
@@ -1254,22 +1187,18 @@ static int brcmf_usb_probe_cb(struct brcmf_usbdev_info *devinfo)
 	bus->chip = bus_pub->devid;
 	bus->chiprev = bus_pub->chiprev;
 	bus->proto_type = BRCMF_PROTO_BCDC;
+	bus->always_use_fws_queue = true;
 
-	/* Attach to the common driver interface */
-	ret = brcmf_attach(dev);
-	if (ret) {
-		brcmf_err("brcmf_attach failed\n");
-		goto fail;
-	}
-
-	ret = brcmf_bus_start(dev);
-	if (ret) {
-		brcmf_err("dongle is not responding\n");
-		brcmf_detach(dev);
-		goto fail;
+	if (!brcmf_usb_dlneeded(devinfo)) {
+		ret = brcmf_usb_bus_setup(devinfo);
+		if (ret)
+			goto fail;
 	}
-
+	/* request firmware here */
+	brcmf_fw_get_firmwares(dev, 0, brcmf_usb_get_fwname(devinfo), NULL,
+			       brcmf_usb_probe_phase2);
 	return 0;
+
 fail:
 	/* Release resources in reverse order */
 	kfree(bus);
@@ -1357,9 +1286,6 @@ brcmf_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 		goto fail;
 	}
 
-	endpoint_num = endpoint->bEndpointAddress & USB_ENDPOINT_NUMBER_MASK;
-	devinfo->intr_pipe = usb_rcvintpipe(usb, endpoint_num);
-
 	devinfo->rx_pipe = 0;
 	devinfo->rx_pipe2 = 0;
 	devinfo->tx_pipe = 0;
@@ -1391,16 +1317,9 @@ brcmf_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 		}
 	}
 
-	/* Allocate interrupt URB and data buffer */
-	/* RNDIS says 8-byte intr, our old drivers used 4-byte */
-	if (IFEPDESC(usb, CONTROL_IF, 0).wMaxPacketSize == cpu_to_le16(16))
-		devinfo->intr_size = 8;
-	else
-		devinfo->intr_size = 4;
-
-	devinfo->interval = IFEPDESC(usb, CONTROL_IF, 0).bInterval;
-
-	if (usb->speed == USB_SPEED_HIGH)
+	if (usb->speed == USB_SPEED_SUPER)
+		brcmf_dbg(USB, "Broadcom super speed USB wireless device detected\n");
+	else if (usb->speed == USB_SPEED_HIGH)
 		brcmf_dbg(USB, "Broadcom high speed USB wireless device detected\n");
 	else
 		brcmf_dbg(USB, "Broadcom full speed USB wireless device detected\n");
@@ -1455,23 +1374,18 @@ static int brcmf_usb_resume(struct usb_interface *intf)
 	struct brcmf_usbdev_info *devinfo = brcmf_usb_get_businfo(&usb->dev);
 
 	brcmf_dbg(USB, "Enter\n");
-	if (!brcmf_attach(devinfo->dev))
-		return brcmf_bus_start(&usb->dev);
-
-	return 0;
+	return brcmf_usb_bus_setup(devinfo);
 }
 
 static int brcmf_usb_reset_resume(struct usb_interface *intf)
 {
 	struct usb_device *usb = interface_to_usbdev(intf);
 	struct brcmf_usbdev_info *devinfo = brcmf_usb_get_businfo(&usb->dev);
-
 	brcmf_dbg(USB, "Enter\n");
 
-	if (!brcmf_usb_fw_download(devinfo))
-		return brcmf_usb_resume(intf);
-
-	return -EIO;
+	return brcmf_fw_get_firmwares(&usb->dev, 0,
+				      brcmf_usb_get_fwname(devinfo), NULL,
+				      brcmf_usb_probe_phase2);
 }
 
 #define BRCMF_USB_VENDOR_ID_BROADCOM	0x0a5c
@@ -1506,16 +1420,6 @@ static struct usb_driver brcmf_usbdrvr = {
 	.disable_hub_initiated_lpm = 1,
 };
 
-static void brcmf_release_fw(struct list_head *q)
-{
-	struct brcmf_usb_image *fw_image, *next;
-
-	list_for_each_entry_safe(fw_image, next, q, list) {
-		vfree(fw_image->image);
-		list_del_init(&fw_image->list);
-	}
-}
-
 static int brcmf_usb_reset_device(struct device *dev, void *notused)
 {
 	/* device past is the usb interface so we
@@ -1534,12 +1438,10 @@ void brcmf_usb_exit(void)
 	ret = driver_for_each_device(drv, NULL, NULL,
 				     brcmf_usb_reset_device);
 	usb_deregister(&brcmf_usbdrvr);
-	brcmf_release_fw(&fw_image_list);
 }
 
 void brcmf_usb_register(void)
 {
 	brcmf_dbg(USB, "Enter\n");
-	INIT_LIST_HEAD(&fw_image_list);
 	usb_register(&brcmf_usbdrvr);
 }
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c b/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
index be1985296bdc..d8fa276e368b 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
@@ -221,9 +221,9 @@ static const struct ieee80211_regdomain brcmf_regdom = {
 		 */
 		REG_RULE(2484-10, 2484+10, 20, 6, 20, 0),
 		/* IEEE 802.11a, channel 36..64 */
-		REG_RULE(5150-10, 5350+10, 40, 6, 20, 0),
+		REG_RULE(5150-10, 5350+10, 80, 6, 20, 0),
 		/* IEEE 802.11a, channel 100..165 */
-		REG_RULE(5470-10, 5850+10, 40, 6, 20, 0), }
+		REG_RULE(5470-10, 5850+10, 80, 6, 20, 0), }
 };
 
 static const u32 __wl_cipher_suites[] = {
@@ -341,6 +341,60 @@ static u8 brcmf_mw_to_qdbm(u16 mw)
 	return qdbm;
 }
 
+static u16 chandef_to_chanspec(struct brcmu_d11inf *d11inf,
+			       struct cfg80211_chan_def *ch)
+{
+	struct brcmu_chan ch_inf;
+	s32 primary_offset;
+
+	brcmf_dbg(TRACE, "chandef: control %d center %d width %d\n",
+		  ch->chan->center_freq, ch->center_freq1, ch->width);
+	ch_inf.chnum = ieee80211_frequency_to_channel(ch->center_freq1);
+	primary_offset = ch->center_freq1 - ch->chan->center_freq;
+	switch (ch->width) {
+	case NL80211_CHAN_WIDTH_20:
+		ch_inf.bw = BRCMU_CHAN_BW_20;
+		WARN_ON(primary_offset != 0);
+		break;
+	case NL80211_CHAN_WIDTH_40:
+		ch_inf.bw = BRCMU_CHAN_BW_40;
+		if (primary_offset < 0)
+			ch_inf.sb = BRCMU_CHAN_SB_U;
+		else
+			ch_inf.sb = BRCMU_CHAN_SB_L;
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		ch_inf.bw = BRCMU_CHAN_BW_80;
+		if (primary_offset < 0) {
+			if (primary_offset < -CH_10MHZ_APART)
+				ch_inf.sb = BRCMU_CHAN_SB_UU;
+			else
+				ch_inf.sb = BRCMU_CHAN_SB_UL;
+		} else {
+			if (primary_offset > CH_10MHZ_APART)
+				ch_inf.sb = BRCMU_CHAN_SB_LL;
+			else
+				ch_inf.sb = BRCMU_CHAN_SB_LU;
+		}
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+	switch (ch->chan->band) {
+	case IEEE80211_BAND_2GHZ:
+		ch_inf.band = BRCMU_CHAN_BAND_2G;
+		break;
+	case IEEE80211_BAND_5GHZ:
+		ch_inf.band = BRCMU_CHAN_BAND_5G;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+	d11inf->encchspec(&ch_inf);
+
+	return ch_inf.chspec;
+}
+
 u16 channel_to_chanspec(struct brcmu_d11inf *d11inf,
 			struct ieee80211_channel *ch)
 {
@@ -586,6 +640,9 @@ s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg,
 		if (err)
 			brcmf_err("Scan abort  failed\n");
 	}
+
+	brcmf_set_mpc(ifp, 1);
+
 	/*
 	 * e-scan can be initiated by scheduled scan
 	 * which takes precedence.
@@ -595,12 +652,10 @@ s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg,
 		cfg->sched_escan = false;
 		if (!aborted)
 			cfg80211_sched_scan_results(cfg_to_wiphy(cfg));
-		brcmf_set_mpc(ifp, 1);
 	} else if (scan_request) {
 		brcmf_dbg(SCAN, "ESCAN Completed scan: %s\n",
 			  aborted ? "Aborted" : "Done");
 		cfg80211_scan_done(scan_request, aborted);
-		brcmf_set_mpc(ifp, 1);
 	}
 	if (!test_and_clear_bit(BRCMF_SCAN_STATUS_BUSY, &cfg->scan_status))
 		brcmf_dbg(SCAN, "Scan complete, probably P2P scan\n");
@@ -1236,8 +1291,8 @@ brcmf_cfg80211_join_ibss(struct wiphy *wiphy, struct net_device *ndev,
 				params->chandef.chan->center_freq);
 		if (params->channel_fixed) {
 			/* adding chanspec */
-			chanspec = channel_to_chanspec(&cfg->d11inf,
-						       params->chandef.chan);
+			chanspec = chandef_to_chanspec(&cfg->d11inf,
+						       &params->chandef);
 			join_params.params_le.chanspec_list[0] =
 				cpu_to_le16(chanspec);
 			join_params.params_le.chanspec_num = cpu_to_le32(1);
@@ -2182,7 +2237,7 @@ brcmf_cfg80211_config_default_mgmt_key(struct wiphy *wiphy,
 
 static s32
 brcmf_cfg80211_get_station(struct wiphy *wiphy, struct net_device *ndev,
-			   u8 *mac, struct station_info *sinfo)
+			   const u8 *mac, struct station_info *sinfo)
 {
 	struct brcmf_if *ifp = netdev_priv(ndev);
 	struct brcmf_cfg80211_profile *profile = &ifp->vif->profile;
@@ -3124,7 +3179,7 @@ brcmf_cfg80211_sched_scan_start(struct wiphy *wiphy,
 	}
 
 	if (!request->n_ssids || !request->n_match_sets) {
-		brcmf_err("Invalid sched scan req!! n_ssids:%d\n",
+		brcmf_dbg(SCAN, "Invalid sched scan req!! n_ssids:%d\n",
 			  request->n_ssids);
 		return -EINVAL;
 	}
@@ -3734,23 +3789,6 @@ brcmf_config_ap_mgmt_ie(struct brcmf_cfg80211_vif *vif,
 }
 
 static s32
-brcmf_cfg80211_set_channel(struct brcmf_cfg80211_info *cfg,
-			   struct brcmf_if *ifp,
-			   struct ieee80211_channel *channel)
-{
-	u16 chanspec;
-	s32 err;
-
-	brcmf_dbg(TRACE, "band=%d, center_freq=%d\n", channel->band,
-		  channel->center_freq);
-
-	chanspec = channel_to_chanspec(&cfg->d11inf, channel);
-	err = brcmf_fil_iovar_int_set(ifp, "chanspec", chanspec);
-
-	return err;
-}
-
-static s32
 brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev,
 			struct cfg80211_ap_settings *settings)
 {
@@ -3765,11 +3803,12 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev,
 	struct brcmf_join_params join_params;
 	enum nl80211_iftype dev_role;
 	struct brcmf_fil_bss_enable_le bss_enable;
+	u16 chanspec;
 
-	brcmf_dbg(TRACE, "channel_type=%d, beacon_interval=%d, dtim_period=%d,\n",
-		  cfg80211_get_chandef_type(&settings->chandef),
-		  settings->beacon_interval,
-		  settings->dtim_period);
+	brcmf_dbg(TRACE, "ctrlchn=%d, center=%d, bw=%d, beacon_interval=%d, dtim_period=%d,\n",
+		  settings->chandef.chan->hw_value,
+		  settings->chandef.center_freq1, settings->chandef.width,
+		  settings->beacon_interval, settings->dtim_period);
 	brcmf_dbg(TRACE, "ssid=%s(%zu), auth_type=%d, inactivity_timeout=%d\n",
 		  settings->ssid, settings->ssid_len, settings->auth_type,
 		  settings->inactivity_timeout);
@@ -3826,9 +3865,10 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev,
 
 	brcmf_config_ap_mgmt_ie(ifp->vif, &settings->beacon);
 
-	err = brcmf_cfg80211_set_channel(cfg, ifp, settings->chandef.chan);
+	chanspec = chandef_to_chanspec(&cfg->d11inf, &settings->chandef);
+	err = brcmf_fil_iovar_int_set(ifp, "chanspec", chanspec);
 	if (err < 0) {
-		brcmf_err("Set Channel failed, %d\n", err);
+		brcmf_err("Set Channel failed: chspec=%d, %d\n", chanspec, err);
 		goto exit;
 	}
 
@@ -3975,7 +4015,7 @@ brcmf_cfg80211_change_beacon(struct wiphy *wiphy, struct net_device *ndev,
 
 static int
 brcmf_cfg80211_del_station(struct wiphy *wiphy, struct net_device *ndev,
-			   u8 *mac)
+			   const u8 *mac)
 {
 	struct brcmf_cfg80211_info *cfg = wiphy_to_cfg(wiphy);
 	struct brcmf_scb_val_le scbval;
@@ -4203,7 +4243,7 @@ static int brcmf_convert_nl80211_tdls_oper(enum nl80211_tdls_operation oper)
 }
 
 static int brcmf_cfg80211_tdls_oper(struct wiphy *wiphy,
-				    struct net_device *ndev, u8 *peer,
+				    struct net_device *ndev, const u8 *peer,
 				    enum nl80211_tdls_operation oper)
 {
 	struct brcmf_if *ifp;
@@ -4364,6 +4404,8 @@ static struct wiphy *brcmf_setup_wiphy(struct device *phydev)
 			WIPHY_FLAG_OFFCHAN_TX |
 			WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL |
 			WIPHY_FLAG_SUPPORTS_TDLS;
+	if (!brcmf_roamoff)
+		wiphy->flags |= WIPHY_FLAG_SUPPORTS_FW_ROAM;
 	wiphy->mgmt_stypes = brcmf_txrx_stypes;
 	wiphy->max_remain_on_channel_duration = 5000;
 	brcmf_wiphy_pno_params(wiphy);
@@ -4685,7 +4727,6 @@ brcmf_notify_connect_status(struct brcmf_if *ifp,
 	struct brcmf_cfg80211_profile *profile = &ifp->vif->profile;
 	struct ieee80211_channel *chan;
 	s32 err = 0;
-	u16 reason;
 
 	if (brcmf_is_apmode(ifp->vif)) {
 		err = brcmf_notify_connect_status_ap(cfg, ndev, e, data);
@@ -4706,16 +4747,6 @@ brcmf_notify_connect_status(struct brcmf_if *ifp,
 		brcmf_dbg(CONN, "Linkdown\n");
 		if (!brcmf_is_ibssmode(ifp->vif)) {
 			brcmf_bss_connect_done(cfg, ndev, e, false);
-			if (test_and_clear_bit(BRCMF_VIF_STATUS_CONNECTED,
-					       &ifp->vif->sme_state)) {
-				reason = 0;
-				if (((e->event_code == BRCMF_E_DEAUTH_IND) ||
-				     (e->event_code == BRCMF_E_DISASSOC_IND)) &&
-				    (e->reason != WLAN_REASON_UNSPECIFIED))
-					reason = e->reason;
-				cfg80211_disconnected(ndev, reason, NULL, 0,
-						      GFP_KERNEL);
-			}
 		}
 		brcmf_link_down(ifp->vif);
 		brcmf_init_prof(ndev_to_prof(ndev));
@@ -5215,6 +5246,9 @@ static s32 brcmf_construct_reginfo(struct brcmf_cfg80211_info *cfg,
 		if (!(bw_cap[band] & WLC_BW_40MHZ_BIT) &&
 		    ch.bw == BRCMU_CHAN_BW_40)
 			continue;
+		if (!(bw_cap[band] & WLC_BW_80MHZ_BIT) &&
+		    ch.bw == BRCMU_CHAN_BW_80)
+			continue;
 		update = false;
 		for (j = 0; (j < *n_cnt && (*n_cnt < array_size)); j++) {
 			if (band_chan_arr[j].hw_value == ch.chnum) {
@@ -5231,10 +5265,13 @@ static s32 brcmf_construct_reginfo(struct brcmf_cfg80211_info *cfg,
 				ieee80211_channel_to_frequency(ch.chnum, band);
 			band_chan_arr[index].hw_value = ch.chnum;
 
-			if (ch.bw == BRCMU_CHAN_BW_40) {
-				/* assuming the order is HT20, HT40 Upper,
-				 * HT40 lower from chanspecs
-				 */
+			/* assuming the chanspecs order is HT20,
+			 * HT40 upper, HT40 lower, and VHT80.
+			 */
+			if (ch.bw == BRCMU_CHAN_BW_80) {
+				band_chan_arr[index].flags &=
+					~IEEE80211_CHAN_NO_80MHZ;
+			} else if (ch.bw == BRCMU_CHAN_BW_40) {
 				ht40_flag = band_chan_arr[index].flags &
 					    IEEE80211_CHAN_NO_HT40;
 				if (ch.sb == BRCMU_CHAN_SB_U) {
@@ -5255,8 +5292,13 @@ static s32 brcmf_construct_reginfo(struct brcmf_cfg80211_info *cfg,
 						    IEEE80211_CHAN_NO_HT40MINUS;
 				}
 			} else {
+				/* disable other bandwidths for now as mentioned
+				 * order assure they are enabled for subsequent
+				 * chanspecs.
+				 */
 				band_chan_arr[index].flags =
-							IEEE80211_CHAN_NO_HT40;
+						IEEE80211_CHAN_NO_HT40 |
+						IEEE80211_CHAN_NO_80MHZ;
 				ch.bw = BRCMU_CHAN_BW_20;
 				cfg->d11inf.encchspec(&ch);
 				channel = ch.chspec;
@@ -5323,13 +5365,63 @@ static void brcmf_get_bwcap(struct brcmf_if *ifp, u32 bw_cap[])
 	}
 }
 
+static void brcmf_update_ht_cap(struct ieee80211_supported_band *band,
+				u32 bw_cap[2], u32 nchain)
+{
+	band->ht_cap.ht_supported = true;
+	if (bw_cap[band->band] & WLC_BW_40MHZ_BIT) {
+		band->ht_cap.cap |= IEEE80211_HT_CAP_SGI_40;
+		band->ht_cap.cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40;
+	}
+	band->ht_cap.cap |= IEEE80211_HT_CAP_SGI_20;
+	band->ht_cap.cap |= IEEE80211_HT_CAP_DSSSCCK40;
+	band->ht_cap.ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K;
+	band->ht_cap.ampdu_density = IEEE80211_HT_MPDU_DENSITY_16;
+	memset(band->ht_cap.mcs.rx_mask, 0xff, nchain);
+	band->ht_cap.mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED;
+}
+
+static __le16 brcmf_get_mcs_map(u32 nchain, enum ieee80211_vht_mcs_support supp)
+{
+	u16 mcs_map;
+	int i;
+
+	for (i = 0, mcs_map = 0xFFFF; i < nchain; i++)
+		mcs_map = (mcs_map << 2) | supp;
+
+	return cpu_to_le16(mcs_map);
+}
+
+static void brcmf_update_vht_cap(struct ieee80211_supported_band *band,
+				 u32 bw_cap[2], u32 nchain)
+{
+	__le16 mcs_map;
+
+	/* not allowed in 2.4G band */
+	if (band->band == IEEE80211_BAND_2GHZ)
+		return;
+
+	band->vht_cap.vht_supported = true;
+	/* 80MHz is mandatory */
+	band->vht_cap.cap |= IEEE80211_VHT_CAP_SHORT_GI_80;
+	if (bw_cap[band->band] & WLC_BW_160MHZ_BIT) {
+		band->vht_cap.cap |= IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ;
+		band->vht_cap.cap |= IEEE80211_VHT_CAP_SHORT_GI_160;
+	}
+	/* all support 256-QAM */
+	mcs_map = brcmf_get_mcs_map(nchain, IEEE80211_VHT_MCS_SUPPORT_0_9);
+	band->vht_cap.vht_mcs.rx_mcs_map = mcs_map;
+	band->vht_cap.vht_mcs.tx_mcs_map = mcs_map;
+}
+
 static s32 brcmf_update_wiphybands(struct brcmf_cfg80211_info *cfg)
 {
 	struct brcmf_if *ifp = netdev_priv(cfg_to_ndev(cfg));
 	struct wiphy *wiphy;
 	s32 phy_list;
 	u32 band_list[3];
-	u32 nmode;
+	u32 nmode = 0;
+	u32 vhtmode = 0;
 	u32 bw_cap[2] = { 0, 0 };
 	u32 rxchain;
 	u32 nchain;
@@ -5360,14 +5452,16 @@ static s32 brcmf_update_wiphybands(struct brcmf_cfg80211_info *cfg)
 	brcmf_dbg(INFO, "BRCMF_C_GET_BANDLIST reported: 0x%08x 0x%08x 0x%08x phy\n",
 		  band_list[0], band_list[1], band_list[2]);
 
+	(void)brcmf_fil_iovar_int_get(ifp, "vhtmode", &vhtmode);
 	err = brcmf_fil_iovar_int_get(ifp, "nmode", &nmode);
 	if (err) {
 		brcmf_err("nmode error (%d)\n", err);
 	} else {
 		brcmf_get_bwcap(ifp, bw_cap);
 	}
-	brcmf_dbg(INFO, "nmode=%d, bw_cap=(%d, %d)\n", nmode,
-		  bw_cap[IEEE80211_BAND_2GHZ], bw_cap[IEEE80211_BAND_5GHZ]);
+	brcmf_dbg(INFO, "nmode=%d, vhtmode=%d, bw_cap=(%d, %d)\n",
+		  nmode, vhtmode, bw_cap[IEEE80211_BAND_2GHZ],
+		  bw_cap[IEEE80211_BAND_5GHZ]);
 
 	err = brcmf_fil_iovar_int_get(ifp, "rxchain", &rxchain);
 	if (err) {
@@ -5398,17 +5492,10 @@ static s32 brcmf_update_wiphybands(struct brcmf_cfg80211_info *cfg)
 		else
 			continue;
 
-		if (bw_cap[band->band] & WLC_BW_40MHZ_BIT) {
-			band->ht_cap.cap |= IEEE80211_HT_CAP_SGI_40;
-			band->ht_cap.cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40;
-		}
-		band->ht_cap.cap |= IEEE80211_HT_CAP_SGI_20;
-		band->ht_cap.cap |= IEEE80211_HT_CAP_DSSSCCK40;
-		band->ht_cap.ht_supported = true;
-		band->ht_cap.ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K;
-		band->ht_cap.ampdu_density = IEEE80211_HT_MPDU_DENSITY_16;
-		memset(band->ht_cap.mcs.rx_mask, 0xff, nchain);
-		band->ht_cap.mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED;
+		if (nmode)
+			brcmf_update_ht_cap(band, bw_cap, nchain);
+		if (vhtmode)
+			brcmf_update_vht_cap(band, bw_cap, nchain);
 		bands[band->band] = band;
 	}
 
diff --git a/drivers/net/wireless/brcm80211/brcmsmac/mac80211_if.c b/drivers/net/wireless/brcm80211/brcmsmac/mac80211_if.c
index 8c5fa4e58139..43c71bfaa474 100644
--- a/drivers/net/wireless/brcm80211/brcmsmac/mac80211_if.c
+++ b/drivers/net/wireless/brcm80211/brcmsmac/mac80211_if.c
@@ -897,7 +897,8 @@ static bool brcms_tx_flush_completed(struct brcms_info *wl)
 	return result;
 }
 
-static void brcms_ops_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void brcms_ops_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    u32 queues, bool drop)
 {
 	struct brcms_info *wl = hw->priv;
 	int ret;
diff --git a/drivers/net/wireless/brcm80211/brcmsmac/main.c b/drivers/net/wireless/brcm80211/brcmsmac/main.c
index 9417cb5a2553..af8ba64ace39 100644
--- a/drivers/net/wireless/brcm80211/brcmsmac/main.c
+++ b/drivers/net/wireless/brcm80211/brcmsmac/main.c
@@ -4870,14 +4870,11 @@ static void brcms_c_detach_module(struct brcms_c_info *wlc)
 /*
  * low level detach
  */
-static int brcms_b_detach(struct brcms_c_info *wlc)
+static void brcms_b_detach(struct brcms_c_info *wlc)
 {
 	uint i;
 	struct brcms_hw_band *band;
 	struct brcms_hardware *wlc_hw = wlc->hw;
-	int callbacks;
-
-	callbacks = 0;
 
 	brcms_b_detach_dmapio(wlc_hw);
 
@@ -4900,9 +4897,6 @@ static int brcms_b_detach(struct brcms_c_info *wlc)
 		ai_detach(wlc_hw->sih);
 		wlc_hw->sih = NULL;
 	}
-
-	return callbacks;
-
 }
 
 /*
@@ -4917,14 +4911,15 @@ static int brcms_b_detach(struct brcms_c_info *wlc)
  */
 uint brcms_c_detach(struct brcms_c_info *wlc)
 {
-	uint callbacks = 0;
+	uint callbacks;
 
 	if (wlc == NULL)
 		return 0;
 
-	callbacks += brcms_b_detach(wlc);
+	brcms_b_detach(wlc);
 
 	/* delete software timers */
+	callbacks = 0;
 	if (!brcms_c_radio_monitor_stop(wlc))
 		callbacks++;
 
diff --git a/drivers/net/wireless/brcm80211/brcmutil/d11.c b/drivers/net/wireless/brcm80211/brcmutil/d11.c
index 30e54e2c6c9b..2b2522bdd8eb 100644
--- a/drivers/net/wireless/brcm80211/brcmutil/d11.c
+++ b/drivers/net/wireless/brcm80211/brcmutil/d11.c
@@ -21,19 +21,46 @@
 #include <brcmu_wifi.h>
 #include <brcmu_d11.h>
 
-static void brcmu_d11n_encchspec(struct brcmu_chan *ch)
+static u16 d11n_sb(enum brcmu_chan_sb sb)
 {
-	ch->chspec = ch->chnum & BRCMU_CHSPEC_CH_MASK;
+	switch (sb) {
+	case BRCMU_CHAN_SB_NONE:
+		return BRCMU_CHSPEC_D11N_SB_N;
+	case BRCMU_CHAN_SB_L:
+		return BRCMU_CHSPEC_D11N_SB_L;
+	case BRCMU_CHAN_SB_U:
+		return BRCMU_CHSPEC_D11N_SB_U;
+	default:
+		WARN_ON(1);
+	}
+	return 0;
+}
 
-	switch (ch->bw) {
+static u16 d11n_bw(enum brcmu_chan_bw bw)
+{
+	switch (bw) {
 	case BRCMU_CHAN_BW_20:
-		ch->chspec |= BRCMU_CHSPEC_D11N_BW_20 | BRCMU_CHSPEC_D11N_SB_N;
-		break;
+		return BRCMU_CHSPEC_D11N_BW_20;
 	case BRCMU_CHAN_BW_40:
+		return BRCMU_CHSPEC_D11N_BW_40;
 	default:
-		WARN_ON_ONCE(1);
-		break;
+		WARN_ON(1);
 	}
+	return 0;
+}
+
+static void brcmu_d11n_encchspec(struct brcmu_chan *ch)
+{
+	if (ch->bw == BRCMU_CHAN_BW_20)
+		ch->sb = BRCMU_CHAN_SB_NONE;
+
+	ch->chspec = 0;
+	brcmu_maskset16(&ch->chspec, BRCMU_CHSPEC_CH_MASK,
+			BRCMU_CHSPEC_CH_SHIFT, ch->chnum);
+	brcmu_maskset16(&ch->chspec, BRCMU_CHSPEC_D11N_SB_MASK,
+			0, d11n_sb(ch->sb));
+	brcmu_maskset16(&ch->chspec, BRCMU_CHSPEC_D11N_BW_MASK,
+			0, d11n_bw(ch->bw));
 
 	if (ch->chnum <= CH_MAX_2G_CHANNEL)
 		ch->chspec |= BRCMU_CHSPEC_D11N_BND_2G;
@@ -41,23 +68,34 @@ static void brcmu_d11n_encchspec(struct brcmu_chan *ch)
 		ch->chspec |= BRCMU_CHSPEC_D11N_BND_5G;
 }
 
-static void brcmu_d11ac_encchspec(struct brcmu_chan *ch)
+static u16 d11ac_bw(enum brcmu_chan_bw bw)
 {
-	ch->chspec = ch->chnum & BRCMU_CHSPEC_CH_MASK;
-
-	switch (ch->bw) {
+	switch (bw) {
 	case BRCMU_CHAN_BW_20:
-		ch->chspec |= BRCMU_CHSPEC_D11AC_BW_20;
-		break;
+		return BRCMU_CHSPEC_D11AC_BW_20;
 	case BRCMU_CHAN_BW_40:
+		return BRCMU_CHSPEC_D11AC_BW_40;
 	case BRCMU_CHAN_BW_80:
-	case BRCMU_CHAN_BW_80P80:
-	case BRCMU_CHAN_BW_160:
+		return BRCMU_CHSPEC_D11AC_BW_80;
 	default:
-		WARN_ON_ONCE(1);
-		break;
+		WARN_ON(1);
 	}
+	return 0;
+}
 
+static void brcmu_d11ac_encchspec(struct brcmu_chan *ch)
+{
+	if (ch->bw == BRCMU_CHAN_BW_20 || ch->sb == BRCMU_CHAN_SB_NONE)
+		ch->sb = BRCMU_CHAN_SB_L;
+
+	brcmu_maskset16(&ch->chspec, BRCMU_CHSPEC_CH_MASK,
+			BRCMU_CHSPEC_CH_SHIFT, ch->chnum);
+	brcmu_maskset16(&ch->chspec, BRCMU_CHSPEC_D11AC_SB_MASK,
+			BRCMU_CHSPEC_D11AC_SB_SHIFT, ch->sb);
+	brcmu_maskset16(&ch->chspec, BRCMU_CHSPEC_D11AC_BW_MASK,
+			0, d11ac_bw(ch->bw));
+
+	ch->chspec &= ~BRCMU_CHSPEC_D11AC_BND_MASK;
 	if (ch->chnum <= CH_MAX_2G_CHANNEL)
 		ch->chspec |= BRCMU_CHSPEC_D11AC_BND_2G;
 	else
@@ -73,6 +111,7 @@ static void brcmu_d11n_decchspec(struct brcmu_chan *ch)
 	switch (ch->chspec & BRCMU_CHSPEC_D11N_BW_MASK) {
 	case BRCMU_CHSPEC_D11N_BW_20:
 		ch->bw = BRCMU_CHAN_BW_20;
+		ch->sb = BRCMU_CHAN_SB_NONE;
 		break;
 	case BRCMU_CHSPEC_D11N_BW_40:
 		ch->bw = BRCMU_CHAN_BW_40;
@@ -112,6 +151,7 @@ static void brcmu_d11ac_decchspec(struct brcmu_chan *ch)
 	switch (ch->chspec & BRCMU_CHSPEC_D11AC_BW_MASK) {
 	case BRCMU_CHSPEC_D11AC_BW_20:
 		ch->bw = BRCMU_CHAN_BW_20;
+		ch->sb = BRCMU_CHAN_SB_NONE;
 		break;
 	case BRCMU_CHSPEC_D11AC_BW_40:
 		ch->bw = BRCMU_CHAN_BW_40;
@@ -128,6 +168,25 @@ static void brcmu_d11ac_decchspec(struct brcmu_chan *ch)
 		break;
 	case BRCMU_CHSPEC_D11AC_BW_80:
 		ch->bw = BRCMU_CHAN_BW_80;
+		ch->sb = brcmu_maskget16(ch->chspec, BRCMU_CHSPEC_D11AC_SB_MASK,
+					 BRCMU_CHSPEC_D11AC_SB_SHIFT);
+		switch (ch->sb) {
+		case BRCMU_CHAN_SB_LL:
+			ch->chnum -= CH_30MHZ_APART;
+			break;
+		case BRCMU_CHAN_SB_LU:
+			ch->chnum -= CH_10MHZ_APART;
+			break;
+		case BRCMU_CHAN_SB_UL:
+			ch->chnum += CH_10MHZ_APART;
+			break;
+		case BRCMU_CHAN_SB_UU:
+			ch->chnum += CH_30MHZ_APART;
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			break;
+		}
 		break;
 	case BRCMU_CHSPEC_D11AC_BW_8080:
 	case BRCMU_CHSPEC_D11AC_BW_160:
diff --git a/drivers/net/wireless/brcm80211/include/brcmu_d11.h b/drivers/net/wireless/brcm80211/include/brcmu_d11.h
index 8660a2cba098..f9745ea8b3e0 100644
--- a/drivers/net/wireless/brcm80211/include/brcmu_d11.h
+++ b/drivers/net/wireless/brcm80211/include/brcmu_d11.h
@@ -108,13 +108,7 @@ enum brcmu_chan_bw {
 };
 
 enum brcmu_chan_sb {
-	BRCMU_CHAN_SB_NONE = 0,
-	BRCMU_CHAN_SB_L,
-	BRCMU_CHAN_SB_U,
-	BRCMU_CHAN_SB_LL,
-	BRCMU_CHAN_SB_LU,
-	BRCMU_CHAN_SB_UL,
-	BRCMU_CHAN_SB_UU,
+	BRCMU_CHAN_SB_NONE = -1,
 	BRCMU_CHAN_SB_LLL,
 	BRCMU_CHAN_SB_LLU,
 	BRCMU_CHAN_SB_LUL,
@@ -123,6 +117,12 @@ enum brcmu_chan_sb {
 	BRCMU_CHAN_SB_ULU,
 	BRCMU_CHAN_SB_UUL,
 	BRCMU_CHAN_SB_UUU,
+	BRCMU_CHAN_SB_L = BRCMU_CHAN_SB_LLL,
+	BRCMU_CHAN_SB_U = BRCMU_CHAN_SB_LLU,
+	BRCMU_CHAN_SB_LL = BRCMU_CHAN_SB_LLL,
+	BRCMU_CHAN_SB_LU = BRCMU_CHAN_SB_LLU,
+	BRCMU_CHAN_SB_UL = BRCMU_CHAN_SB_LUL,
+	BRCMU_CHAN_SB_UU = BRCMU_CHAN_SB_LUU,
 };
 
 struct brcmu_chan {
diff --git a/drivers/net/wireless/brcm80211/include/brcmu_wifi.h b/drivers/net/wireless/brcm80211/include/brcmu_wifi.h
index 74419d4bd123..76b5d3a86294 100644
--- a/drivers/net/wireless/brcm80211/include/brcmu_wifi.h
+++ b/drivers/net/wireless/brcm80211/include/brcmu_wifi.h
@@ -29,6 +29,7 @@
 #define CH_UPPER_SB			0x01
 #define CH_LOWER_SB			0x02
 #define CH_EWA_VALID			0x04
+#define CH_30MHZ_APART			6
 #define CH_20MHZ_APART			4
 #define CH_10MHZ_APART			2
 #define CH_5MHZ_APART			1 /* 2G band channels are 5 Mhz apart */
diff --git a/drivers/net/wireless/cw1200/sta.c b/drivers/net/wireless/cw1200/sta.c
index 103f7bce8932..cd0cad7f7759 100644
--- a/drivers/net/wireless/cw1200/sta.c
+++ b/drivers/net/wireless/cw1200/sta.c
@@ -936,7 +936,8 @@ static int __cw1200_flush(struct cw1200_common *priv, bool drop)
 	return ret;
 }
 
-void cw1200_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+void cw1200_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		  u32 queues, bool drop)
 {
 	struct cw1200_common *priv = hw->priv;
 
diff --git a/drivers/net/wireless/cw1200/sta.h b/drivers/net/wireless/cw1200/sta.h
index 35babb62cc6a..b7e386b7662b 100644
--- a/drivers/net/wireless/cw1200/sta.h
+++ b/drivers/net/wireless/cw1200/sta.h
@@ -40,7 +40,8 @@ int cw1200_set_key(struct ieee80211_hw *dev, enum set_key_cmd cmd,
 
 int cw1200_set_rts_threshold(struct ieee80211_hw *hw, u32 value);
 
-void cw1200_flush(struct ieee80211_hw *hw, u32 queues, bool drop);
+void cw1200_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		  u32 queues, bool drop);
 
 u64 cw1200_prepare_multicast(struct ieee80211_hw *hw,
 			     struct netdev_hw_addr_list *mc_list);
diff --git a/drivers/net/wireless/hostap/hostap_main.c b/drivers/net/wireless/hostap/hostap_main.c
index 67db34e56d7e..52919ad42726 100644
--- a/drivers/net/wireless/hostap/hostap_main.c
+++ b/drivers/net/wireless/hostap/hostap_main.c
@@ -882,7 +882,7 @@ void hostap_setup_dev(struct net_device *dev, local_info_t *local,
 	dev->mtu = local->mtu;
 
 
-	SET_ETHTOOL_OPS(dev, &prism2_ethtool_ops);
+	dev->ethtool_ops = &prism2_ethtool_ops;
 
 }
 
diff --git a/drivers/net/wireless/iwlegacy/3945.c b/drivers/net/wireless/iwlegacy/3945.c
index d37a6fd90d40..b598e2803500 100644
--- a/drivers/net/wireless/iwlegacy/3945.c
+++ b/drivers/net/wireless/iwlegacy/3945.c
@@ -573,7 +573,7 @@ il3945_hdl_rx(struct il_priv *il, struct il_rx_buf *rxb)
 		rx_status.flag |= RX_FLAG_SHORTPRE;
 
 	if ((unlikely(rx_stats->phy_count > 20))) {
-		D_DROP("dsp size out of range [0,20]: %d/n",
+		D_DROP("dsp size out of range [0,20]: %d\n",
 		       rx_stats->phy_count);
 		return;
 	}
diff --git a/drivers/net/wireless/iwlegacy/4965-mac.c b/drivers/net/wireless/iwlegacy/4965-mac.c
index 888ad5c74639..c159c05db6ef 100644
--- a/drivers/net/wireless/iwlegacy/4965-mac.c
+++ b/drivers/net/wireless/iwlegacy/4965-mac.c
@@ -670,7 +670,7 @@ il4965_hdl_rx(struct il_priv *il, struct il_rx_buf *rxb)
 	}
 
 	if ((unlikely(phy_res->cfg_phy_cnt > 20))) {
-		D_DROP("dsp size out of range [0,20]: %d/n",
+		D_DROP("dsp size out of range [0,20]: %d\n",
 		       phy_res->cfg_phy_cnt);
 		return;
 	}
diff --git a/drivers/net/wireless/iwlegacy/common.c b/drivers/net/wireless/iwlegacy/common.c
index 4f42174d9994..ecc674627e6e 100644
--- a/drivers/net/wireless/iwlegacy/common.c
+++ b/drivers/net/wireless/iwlegacy/common.c
@@ -4755,7 +4755,8 @@ out:
 }
 EXPORT_SYMBOL(il_mac_change_interface);
 
-void il_mac_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+void il_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		  u32 queues, bool drop)
 {
 	struct il_priv *il = hw->priv;
 	unsigned long timeout = jiffies + msecs_to_jiffies(500);
diff --git a/drivers/net/wireless/iwlegacy/common.h b/drivers/net/wireless/iwlegacy/common.h
index dfb13c70efe8..ea5c0f863c4e 100644
--- a/drivers/net/wireless/iwlegacy/common.h
+++ b/drivers/net/wireless/iwlegacy/common.h
@@ -1723,7 +1723,8 @@ void il_mac_remove_interface(struct ieee80211_hw *hw,
 			     struct ieee80211_vif *vif);
 int il_mac_change_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			    enum nl80211_iftype newtype, bool newp2p);
-void il_mac_flush(struct ieee80211_hw *hw, u32 queues, bool drop);
+void il_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		  u32 queues, bool drop);
 int il_alloc_txq_mem(struct il_priv *il);
 void il_free_txq_mem(struct il_priv *il);
 
diff --git a/drivers/net/wireless/iwlwifi/Kconfig b/drivers/net/wireless/iwlwifi/Kconfig
index 74b3b4de7bb7..7fd50428b934 100644
--- a/drivers/net/wireless/iwlwifi/Kconfig
+++ b/drivers/net/wireless/iwlwifi/Kconfig
@@ -2,10 +2,6 @@ config IWLWIFI
 	tristate "Intel Wireless WiFi Next Gen AGN - Wireless-N/Advanced-N/Ultimate-N (iwlwifi) "
 	depends on PCI && MAC80211 && HAS_IOMEM
 	select FW_LOADER
-	select NEW_LEDS
-	select LEDS_CLASS
-	select LEDS_TRIGGERS
-	select MAC80211_LEDS
 	---help---
 	  Select to build the driver supporting the:
 
@@ -43,6 +39,14 @@ config IWLWIFI
 	  say M here and read <file:Documentation/kbuild/modules.txt>.  The
 	  module will be called iwlwifi.
 
+config IWLWIFI_LEDS
+	bool
+	depends on IWLWIFI
+	depends on LEDS_CLASS=y || LEDS_CLASS=IWLWIFI
+	select LEDS_TRIGGERS
+	select MAC80211_LEDS
+	default y
+
 config IWLDVM
 	tristate "Intel Wireless WiFi DVM Firmware support"
 	depends on IWLWIFI
@@ -124,7 +128,6 @@ config IWLWIFI_DEBUG_EXPERIMENTAL_UCODE
 	  Enable use of experimental ucode for testing and debugging.
 
 config IWLWIFI_DEVICE_TRACING
-
 	bool "iwlwifi device access tracing"
 	depends on IWLWIFI
 	depends on EVENT_TRACING
diff --git a/drivers/net/wireless/iwlwifi/dvm/Makefile b/drivers/net/wireless/iwlwifi/dvm/Makefile
index dce7ab2e0c4b..4d19685f31c3 100644
--- a/drivers/net/wireless/iwlwifi/dvm/Makefile
+++ b/drivers/net/wireless/iwlwifi/dvm/Makefile
@@ -4,9 +4,10 @@ iwldvm-objs		+= main.o rs.o mac80211.o ucode.o tx.o
 iwldvm-objs		+= lib.o calib.o tt.o sta.o rx.o
 
 iwldvm-objs		+= power.o
-iwldvm-objs		+= scan.o led.o
+iwldvm-objs		+= scan.o
 iwldvm-objs		+= rxon.o devices.o
 
+iwldvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwldvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o
 
 ccflags-y += -D__CHECK_ENDIAN__ -I$(src)/../
diff --git a/drivers/net/wireless/iwlwifi/dvm/calib.c b/drivers/net/wireless/iwlwifi/dvm/calib.c
index be1086c87157..20e6aa910700 100644
--- a/drivers/net/wireless/iwlwifi/dvm/calib.c
+++ b/drivers/net/wireless/iwlwifi/dvm/calib.c
@@ -94,7 +94,6 @@ int iwl_send_calib_results(struct iwl_priv *priv)
 {
 	struct iwl_host_cmd hcmd = {
 		.id = REPLY_PHY_CALIBRATION_CMD,
-		.flags = CMD_SYNC,
 	};
 	struct iwl_calib_result *res;
 
diff --git a/drivers/net/wireless/iwlwifi/dvm/debugfs.c b/drivers/net/wireless/iwlwifi/dvm/debugfs.c
index d2fe2596d54e..0ffb6ff1a255 100644
--- a/drivers/net/wireless/iwlwifi/dvm/debugfs.c
+++ b/drivers/net/wireless/iwlwifi/dvm/debugfs.c
@@ -1481,7 +1481,7 @@ static ssize_t iwl_dbgfs_ucode_bt_stats_read(struct file *file,
 
 	/* make request to uCode to retrieve statistics information */
 	mutex_lock(&priv->mutex);
-	ret = iwl_send_statistics_request(priv, CMD_SYNC, false);
+	ret = iwl_send_statistics_request(priv, 0, false);
 	mutex_unlock(&priv->mutex);
 
 	if (ret)
@@ -1868,7 +1868,7 @@ static ssize_t iwl_dbgfs_clear_ucode_statistics_write(struct file *file,
 
 	/* make request to uCode to retrieve statistics information */
 	mutex_lock(&priv->mutex);
-	iwl_send_statistics_request(priv, CMD_SYNC, true);
+	iwl_send_statistics_request(priv, 0, true);
 	mutex_unlock(&priv->mutex);
 
 	return count;
@@ -2188,7 +2188,6 @@ static int iwl_cmd_echo_test(struct iwl_priv *priv)
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_ECHO,
 		.len = { 0 },
-		.flags = CMD_SYNC,
 	};
 
 	ret = iwl_dvm_send_cmd(priv, &cmd);
@@ -2320,7 +2319,7 @@ static ssize_t iwl_dbgfs_fw_restart_write(struct file *file,
 	mutex_lock(&priv->mutex);
 
 	/* take the return value to make compiler happy - it will fail anyway */
-	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_ERROR, CMD_SYNC, 0, NULL);
+	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_ERROR, 0, 0, NULL);
 
 	mutex_unlock(&priv->mutex);
 
diff --git a/drivers/net/wireless/iwlwifi/dvm/dev.h b/drivers/net/wireless/iwlwifi/dvm/dev.h
index 3441f70d0ff9..a6f22c32a279 100644
--- a/drivers/net/wireless/iwlwifi/dvm/dev.h
+++ b/drivers/net/wireless/iwlwifi/dvm/dev.h
@@ -888,9 +888,11 @@ struct iwl_priv {
 
 	struct iwl_event_log event_log;
 
+#ifdef CONFIG_IWLWIFI_LEDS
 	struct led_classdev led;
 	unsigned long blink_on, blink_off;
 	bool led_registered;
+#endif
 
 	/* WoWLAN GTK rekey data */
 	u8 kck[NL80211_KCK_LEN], kek[NL80211_KEK_LEN];
diff --git a/drivers/net/wireless/iwlwifi/dvm/devices.c b/drivers/net/wireless/iwlwifi/dvm/devices.c
index 758c54eeb206..34b41e5f7cfc 100644
--- a/drivers/net/wireless/iwlwifi/dvm/devices.c
+++ b/drivers/net/wireless/iwlwifi/dvm/devices.c
@@ -417,7 +417,6 @@ static int iwl5000_hw_channel_switch(struct iwl_priv *priv,
 	struct iwl_host_cmd hcmd = {
 		.id = REPLY_CHANNEL_SWITCH,
 		.len = { sizeof(cmd), },
-		.flags = CMD_SYNC,
 		.data = { &cmd, },
 	};
 
@@ -579,7 +578,6 @@ static int iwl6000_hw_channel_switch(struct iwl_priv *priv,
 	struct iwl_host_cmd hcmd = {
 		.id = REPLY_CHANNEL_SWITCH,
 		.len = { sizeof(*cmd), },
-		.flags = CMD_SYNC,
 		.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
 	};
 	int err;
diff --git a/drivers/net/wireless/iwlwifi/dvm/led.h b/drivers/net/wireless/iwlwifi/dvm/led.h
index 6a0817d9c4fa..1c6b2252d0f2 100644
--- a/drivers/net/wireless/iwlwifi/dvm/led.h
+++ b/drivers/net/wireless/iwlwifi/dvm/led.h
@@ -36,8 +36,20 @@ struct iwl_priv;
 #define IWL_LED_ACTIVITY       (0<<1)
 #define IWL_LED_LINK           (1<<1)
 
+#ifdef CONFIG_IWLWIFI_LEDS
 void iwlagn_led_enable(struct iwl_priv *priv);
 void iwl_leds_init(struct iwl_priv *priv);
 void iwl_leds_exit(struct iwl_priv *priv);
+#else
+static inline void iwlagn_led_enable(struct iwl_priv *priv)
+{
+}
+static inline void iwl_leds_init(struct iwl_priv *priv)
+{
+}
+static inline void iwl_leds_exit(struct iwl_priv *priv)
+{
+}
+#endif
 
 #endif /* __iwl_leds_h__ */
diff --git a/drivers/net/wireless/iwlwifi/dvm/lib.c b/drivers/net/wireless/iwlwifi/dvm/lib.c
index 576f7ee38ca5..2191621d69c1 100644
--- a/drivers/net/wireless/iwlwifi/dvm/lib.c
+++ b/drivers/net/wireless/iwlwifi/dvm/lib.c
@@ -81,7 +81,7 @@ int iwlagn_send_tx_power(struct iwl_priv *priv)
 	else
 		tx_ant_cfg_cmd = REPLY_TX_POWER_DBM_CMD;
 
-	return iwl_dvm_send_cmd_pdu(priv, tx_ant_cfg_cmd, CMD_SYNC,
+	return iwl_dvm_send_cmd_pdu(priv, tx_ant_cfg_cmd, 0,
 			sizeof(tx_power_cmd), &tx_power_cmd);
 }
 
@@ -141,7 +141,6 @@ int iwlagn_txfifo_flush(struct iwl_priv *priv, u32 scd_q_msk)
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_TXFIFO_FLUSH,
 		.len = { sizeof(struct iwl_txfifo_flush_cmd), },
-		.flags = CMD_SYNC,
 		.data = { &flush_cmd, },
 	};
 
@@ -180,7 +179,7 @@ void iwlagn_dev_txfifo_flush(struct iwl_priv *priv)
 		goto done;
 	}
 	IWL_DEBUG_INFO(priv, "wait transmit/flush all frames\n");
-	iwl_trans_wait_tx_queue_empty(priv->trans);
+	iwl_trans_wait_tx_queue_empty(priv->trans, 0xffffffff);
 done:
 	ieee80211_wake_queues(priv->hw);
 	mutex_unlock(&priv->mutex);
@@ -333,12 +332,12 @@ void iwlagn_send_advance_bt_config(struct iwl_priv *priv)
 		memcpy(&bt_cmd_v2.basic, &basic,
 			sizeof(basic));
 		ret = iwl_dvm_send_cmd_pdu(priv, REPLY_BT_CONFIG,
-			CMD_SYNC, sizeof(bt_cmd_v2), &bt_cmd_v2);
+			0, sizeof(bt_cmd_v2), &bt_cmd_v2);
 	} else {
 		memcpy(&bt_cmd_v1.basic, &basic,
 			sizeof(basic));
 		ret = iwl_dvm_send_cmd_pdu(priv, REPLY_BT_CONFIG,
-			CMD_SYNC, sizeof(bt_cmd_v1), &bt_cmd_v1);
+			0, sizeof(bt_cmd_v1), &bt_cmd_v1);
 	}
 	if (ret)
 		IWL_ERR(priv, "failed to send BT Coex Config\n");
@@ -1044,7 +1043,6 @@ int iwlagn_send_patterns(struct iwl_priv *priv,
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_WOWLAN_PATTERNS,
 		.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
-		.flags = CMD_SYNC,
 	};
 	int i, err;
 
@@ -1201,7 +1199,6 @@ int iwlagn_suspend(struct iwl_priv *priv, struct cfg80211_wowlan *wowlan)
 		if (key_data.use_rsc_tsc) {
 			struct iwl_host_cmd rsc_tsc_cmd = {
 				.id = REPLY_WOWLAN_TSC_RSC_PARAMS,
-				.flags = CMD_SYNC,
 				.data[0] = key_data.rsc_tsc,
 				.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
 				.len[0] = sizeof(*key_data.rsc_tsc),
@@ -1215,7 +1212,7 @@ int iwlagn_suspend(struct iwl_priv *priv, struct cfg80211_wowlan *wowlan)
 		if (key_data.use_tkip) {
 			ret = iwl_dvm_send_cmd_pdu(priv,
 						 REPLY_WOWLAN_TKIP_PARAMS,
-						 CMD_SYNC, sizeof(tkip_cmd),
+						 0, sizeof(tkip_cmd),
 						 &tkip_cmd);
 			if (ret)
 				goto out;
@@ -1231,20 +1228,20 @@ int iwlagn_suspend(struct iwl_priv *priv, struct cfg80211_wowlan *wowlan)
 
 			ret = iwl_dvm_send_cmd_pdu(priv,
 						 REPLY_WOWLAN_KEK_KCK_MATERIAL,
-						 CMD_SYNC, sizeof(kek_kck_cmd),
+						 0, sizeof(kek_kck_cmd),
 						 &kek_kck_cmd);
 			if (ret)
 				goto out;
 		}
 	}
 
-	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_D3_CONFIG, CMD_SYNC,
+	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_D3_CONFIG, 0,
 				     sizeof(d3_cfg_cmd), &d3_cfg_cmd);
 	if (ret)
 		goto out;
 
 	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_WOWLAN_WAKEUP_FILTER,
-				 CMD_SYNC, sizeof(wakeup_filter_cmd),
+				 0, sizeof(wakeup_filter_cmd),
 				 &wakeup_filter_cmd);
 	if (ret)
 		goto out;
diff --git a/drivers/net/wireless/iwlwifi/dvm/mac80211.c b/drivers/net/wireless/iwlwifi/dvm/mac80211.c
index dd55c9cf7ba8..29af7b51e370 100644
--- a/drivers/net/wireless/iwlwifi/dvm/mac80211.c
+++ b/drivers/net/wireless/iwlwifi/dvm/mac80211.c
@@ -1091,7 +1091,8 @@ static void iwlagn_configure_filter(struct ieee80211_hw *hw,
 			FIF_BCN_PRBRESP_PROMISC | FIF_CONTROL;
 }
 
-static void iwlagn_mac_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void iwlagn_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			     u32 queues, bool drop)
 {
 	struct iwl_priv *priv = IWL_MAC80211_GET_DVM(hw);
 
@@ -1119,7 +1120,7 @@ static void iwlagn_mac_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
 		}
 	}
 	IWL_DEBUG_MAC80211(priv, "wait transmit/flush all frames\n");
-	iwl_trans_wait_tx_queue_empty(priv->trans);
+	iwl_trans_wait_tx_queue_empty(priv->trans, 0xffffffff);
 done:
 	mutex_unlock(&priv->mutex);
 	IWL_DEBUG_MAC80211(priv, "leave\n");
diff --git a/drivers/net/wireless/iwlwifi/dvm/main.c b/drivers/net/wireless/iwlwifi/dvm/main.c
index 6a6df71af1d7..0b7f46f0b079 100644
--- a/drivers/net/wireless/iwlwifi/dvm/main.c
+++ b/drivers/net/wireless/iwlwifi/dvm/main.c
@@ -128,7 +128,6 @@ int iwlagn_send_beacon_cmd(struct iwl_priv *priv)
 	struct iwl_tx_beacon_cmd *tx_beacon_cmd;
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_TX_BEACON,
-		.flags = CMD_SYNC,
 	};
 	struct ieee80211_tx_info *info;
 	u32 frame_size;
@@ -311,8 +310,7 @@ int iwl_send_statistics_request(struct iwl_priv *priv, u8 flags, bool clear)
 					sizeof(struct iwl_statistics_cmd),
 					&statistics_cmd);
 	else
-		return iwl_dvm_send_cmd_pdu(priv, REPLY_STATISTICS_CMD,
-					CMD_SYNC,
+		return iwl_dvm_send_cmd_pdu(priv, REPLY_STATISTICS_CMD, 0,
 					sizeof(struct iwl_statistics_cmd),
 					&statistics_cmd);
 }
@@ -622,7 +620,7 @@ static void iwl_rf_kill_ct_config(struct iwl_priv *priv)
 
 		ret = iwl_dvm_send_cmd_pdu(priv,
 				       REPLY_CT_KILL_CONFIG_CMD,
-				       CMD_SYNC, sizeof(adv_cmd), &adv_cmd);
+				       0, sizeof(adv_cmd), &adv_cmd);
 		if (ret)
 			IWL_ERR(priv, "REPLY_CT_KILL_CONFIG_CMD failed\n");
 		else
@@ -637,7 +635,7 @@ static void iwl_rf_kill_ct_config(struct iwl_priv *priv)
 
 		ret = iwl_dvm_send_cmd_pdu(priv,
 				       REPLY_CT_KILL_CONFIG_CMD,
-				       CMD_SYNC, sizeof(cmd), &cmd);
+				       0, sizeof(cmd), &cmd);
 		if (ret)
 			IWL_ERR(priv, "REPLY_CT_KILL_CONFIG_CMD failed\n");
 		else
@@ -673,9 +671,7 @@ static int iwlagn_send_tx_ant_config(struct iwl_priv *priv, u8 valid_tx_ant)
 
 	if (IWL_UCODE_API(priv->fw->ucode_ver) > 1) {
 		IWL_DEBUG_HC(priv, "select valid tx ant: %u\n", valid_tx_ant);
-		return iwl_dvm_send_cmd_pdu(priv,
-					TX_ANT_CONFIGURATION_CMD,
-					CMD_SYNC,
+		return iwl_dvm_send_cmd_pdu(priv, TX_ANT_CONFIGURATION_CMD, 0,
 					sizeof(struct iwl_tx_ant_config_cmd),
 					&tx_ant_cmd);
 	} else {
@@ -703,7 +699,7 @@ static void iwl_send_bt_config(struct iwl_priv *priv)
 		(bt_cmd.flags == BT_COEX_DISABLE) ? "disable" : "active");
 
 	if (iwl_dvm_send_cmd_pdu(priv, REPLY_BT_CONFIG,
-			     CMD_SYNC, sizeof(struct iwl_bt_cmd), &bt_cmd))
+			     0, sizeof(struct iwl_bt_cmd), &bt_cmd))
 		IWL_ERR(priv, "failed to send BT Coex Config\n");
 }
 
@@ -987,7 +983,7 @@ static void iwl_bg_restart(struct work_struct *data)
 			ieee80211_restart_hw(priv->hw);
 		else
 			IWL_ERR(priv,
-				"Cannot request restart before registrating with mac80211");
+				"Cannot request restart before registrating with mac80211\n");
 	} else {
 		WARN_ON(1);
 	}
@@ -1127,7 +1123,6 @@ static void iwl_option_config(struct iwl_priv *priv)
 static int iwl_eeprom_init_hw_params(struct iwl_priv *priv)
 {
 	struct iwl_nvm_data *data = priv->nvm_data;
-	char *debug_msg;
 
 	if (data->sku_cap_11n_enable &&
 	    !priv->cfg->ht_params) {
@@ -1141,8 +1136,8 @@ static int iwl_eeprom_init_hw_params(struct iwl_priv *priv)
 		return -EINVAL;
 	}
 
-	debug_msg = "Device SKU: 24GHz %s %s, 52GHz %s %s, 11.n %s %s\n";
-	IWL_DEBUG_INFO(priv, debug_msg,
+	IWL_DEBUG_INFO(priv,
+		       "Device SKU: 24GHz %s %s, 52GHz %s %s, 11.n %s %s\n",
 		       data->sku_cap_band_24GHz_enable ? "" : "NOT", "enabled",
 		       data->sku_cap_band_52GHz_enable ? "" : "NOT", "enabled",
 		       data->sku_cap_11n_enable ? "" : "NOT", "enabled");
@@ -1350,7 +1345,7 @@ static struct iwl_op_mode *iwl_op_mode_dvm_start(struct iwl_trans *trans,
 	iwl_set_hw_params(priv);
 
 	if (!(priv->nvm_data->sku_cap_ipan_enable)) {
-		IWL_DEBUG_INFO(priv, "Your EEPROM disabled PAN");
+		IWL_DEBUG_INFO(priv, "Your EEPROM disabled PAN\n");
 		ucode_flags &= ~IWL_UCODE_TLV_FLAGS_PAN;
 		/*
 		 * if not PAN, then don't support P2P -- might be a uCode
@@ -2019,10 +2014,10 @@ void iwlagn_lift_passive_no_rx(struct iwl_priv *priv)
 
 	for (mq = 0; mq < IWLAGN_FIRST_AMPDU_QUEUE; mq++) {
 		if (!test_bit(mq, &priv->transport_queue_stop)) {
-			IWL_DEBUG_TX_QUEUES(priv, "Wake queue %d", mq);
+			IWL_DEBUG_TX_QUEUES(priv, "Wake queue %d\n", mq);
 			ieee80211_wake_queue(priv->hw, mq);
 		} else {
-			IWL_DEBUG_TX_QUEUES(priv, "Don't wake queue %d", mq);
+			IWL_DEBUG_TX_QUEUES(priv, "Don't wake queue %d\n", mq);
 		}
 	}
 
@@ -2053,6 +2048,17 @@ static bool iwl_set_hw_rfkill_state(struct iwl_op_mode *op_mode, bool state)
 	return false;
 }
 
+static void iwl_napi_add(struct iwl_op_mode *op_mode,
+			 struct napi_struct *napi,
+			 struct net_device *napi_dev,
+			 int (*poll)(struct napi_struct *, int),
+			 int weight)
+{
+	struct iwl_priv *priv = IWL_OP_MODE_GET_DVM(op_mode);
+
+	ieee80211_napi_add(priv->hw, napi, napi_dev, poll, weight);
+}
+
 static const struct iwl_op_mode_ops iwl_dvm_ops = {
 	.start = iwl_op_mode_dvm_start,
 	.stop = iwl_op_mode_dvm_stop,
@@ -2065,6 +2071,7 @@ static const struct iwl_op_mode_ops iwl_dvm_ops = {
 	.cmd_queue_full = iwl_cmd_queue_full,
 	.nic_config = iwl_nic_config,
 	.wimax_active = iwl_wimax_active,
+	.napi_add = iwl_napi_add,
 };
 
 /*****************************************************************************
diff --git a/drivers/net/wireless/iwlwifi/dvm/power.c b/drivers/net/wireless/iwlwifi/dvm/power.c
index b4e61417013a..f2c1439566b5 100644
--- a/drivers/net/wireless/iwlwifi/dvm/power.c
+++ b/drivers/net/wireless/iwlwifi/dvm/power.c
@@ -278,7 +278,7 @@ static int iwl_set_power(struct iwl_priv *priv, struct iwl_powertable_cmd *cmd)
 			le32_to_cpu(cmd->sleep_interval[3]),
 			le32_to_cpu(cmd->sleep_interval[4]));
 
-	return iwl_dvm_send_cmd_pdu(priv, POWER_TABLE_CMD, CMD_SYNC,
+	return iwl_dvm_send_cmd_pdu(priv, POWER_TABLE_CMD, 0,
 				sizeof(struct iwl_powertable_cmd), cmd);
 }
 
@@ -361,7 +361,7 @@ int iwl_power_set_mode(struct iwl_priv *priv, struct iwl_powertable_cmd *cmd,
 
 		memcpy(&priv->power_data.sleep_cmd, cmd, sizeof(*cmd));
 	} else
-		IWL_ERR(priv, "set power fail, ret = %d", ret);
+		IWL_ERR(priv, "set power fail, ret = %d\n", ret);
 
 	return ret;
 }
diff --git a/drivers/net/wireless/iwlwifi/dvm/rs.c b/drivers/net/wireless/iwlwifi/dvm/rs.c
index aa773a2da4ab..32b78a66536d 100644
--- a/drivers/net/wireless/iwlwifi/dvm/rs.c
+++ b/drivers/net/wireless/iwlwifi/dvm/rs.c
@@ -1453,7 +1453,7 @@ static int rs_move_legacy_other(struct iwl_priv *priv,
 			tbl->action = IWL_LEGACY_SWITCH_SISO;
 		break;
 	default:
-		IWL_ERR(priv, "Invalid BT load %d", priv->bt_traffic_load);
+		IWL_ERR(priv, "Invalid BT load %d\n", priv->bt_traffic_load);
 		break;
 	}
 
@@ -1628,7 +1628,7 @@ static int rs_move_siso_to_other(struct iwl_priv *priv,
 			tbl->action = IWL_SISO_SWITCH_ANTENNA1;
 		break;
 	default:
-		IWL_ERR(priv, "Invalid BT load %d", priv->bt_traffic_load);
+		IWL_ERR(priv, "Invalid BT load %d\n", priv->bt_traffic_load);
 		break;
 	}
 
@@ -1799,7 +1799,7 @@ static int rs_move_mimo2_to_other(struct iwl_priv *priv,
 			tbl->action = IWL_MIMO2_SWITCH_SISO_A;
 		break;
 	default:
-		IWL_ERR(priv, "Invalid BT load %d", priv->bt_traffic_load);
+		IWL_ERR(priv, "Invalid BT load %d\n", priv->bt_traffic_load);
 		break;
 	}
 
@@ -1969,7 +1969,7 @@ static int rs_move_mimo3_to_other(struct iwl_priv *priv,
 			tbl->action = IWL_MIMO3_SWITCH_SISO_A;
 		break;
 	default:
-		IWL_ERR(priv, "Invalid BT load %d", priv->bt_traffic_load);
+		IWL_ERR(priv, "Invalid BT load %d\n", priv->bt_traffic_load);
 		break;
 	}
 
@@ -2709,7 +2709,7 @@ static void rs_initialize_lq(struct iwl_priv *priv,
 	rs_set_expected_tpt_table(lq_sta, tbl);
 	rs_fill_link_cmd(NULL, lq_sta, rate);
 	priv->stations[lq_sta->lq.sta_id].lq = &lq_sta->lq;
-	iwl_send_lq_cmd(priv, ctx, &lq_sta->lq, CMD_SYNC, true);
+	iwl_send_lq_cmd(priv, ctx, &lq_sta->lq, 0, true);
 }
 
 static void rs_get_rate(void *priv_r, struct ieee80211_sta *sta, void *priv_sta,
diff --git a/drivers/net/wireless/iwlwifi/dvm/rx.c b/drivers/net/wireless/iwlwifi/dvm/rx.c
index cd8377346aff..debec963c610 100644
--- a/drivers/net/wireless/iwlwifi/dvm/rx.c
+++ b/drivers/net/wireless/iwlwifi/dvm/rx.c
@@ -786,7 +786,7 @@ static void iwlagn_pass_packet_to_mac80211(struct iwl_priv *priv,
 
 	memcpy(IEEE80211_SKB_RXCB(skb), stats, sizeof(*stats));
 
-	ieee80211_rx_ni(priv->hw, skb);
+	ieee80211_rx(priv->hw, skb);
 }
 
 static u32 iwlagn_translate_rx_status(struct iwl_priv *priv, u32 decrypt_in)
diff --git a/drivers/net/wireless/iwlwifi/dvm/rxon.c b/drivers/net/wireless/iwlwifi/dvm/rxon.c
index 503a81e58185..ed50de6362ed 100644
--- a/drivers/net/wireless/iwlwifi/dvm/rxon.c
+++ b/drivers/net/wireless/iwlwifi/dvm/rxon.c
@@ -104,7 +104,7 @@ static int iwlagn_disable_bss(struct iwl_priv *priv,
 
 	send->filter_flags &= ~RXON_FILTER_ASSOC_MSK;
 	ret = iwl_dvm_send_cmd_pdu(priv, ctx->rxon_cmd,
-				CMD_SYNC, sizeof(*send), send);
+				0, sizeof(*send), send);
 
 	send->filter_flags = old_filter;
 
@@ -134,7 +134,7 @@ static int iwlagn_disable_pan(struct iwl_priv *priv,
 	send->filter_flags &= ~RXON_FILTER_ASSOC_MSK;
 	send->dev_type = RXON_DEV_TYPE_P2P;
 	ret = iwl_dvm_send_cmd_pdu(priv, ctx->rxon_cmd,
-				CMD_SYNC, sizeof(*send), send);
+				0, sizeof(*send), send);
 
 	send->filter_flags = old_filter;
 	send->dev_type = old_dev_type;
@@ -160,7 +160,7 @@ static int iwlagn_disconn_pan(struct iwl_priv *priv,
 	int ret;
 
 	send->filter_flags &= ~RXON_FILTER_ASSOC_MSK;
-	ret = iwl_dvm_send_cmd_pdu(priv, ctx->rxon_cmd, CMD_SYNC,
+	ret = iwl_dvm_send_cmd_pdu(priv, ctx->rxon_cmd, 0,
 				sizeof(*send), send);
 
 	send->filter_flags = old_filter;
@@ -189,7 +189,7 @@ static void iwlagn_update_qos(struct iwl_priv *priv,
 		      ctx->qos_data.qos_active,
 		      ctx->qos_data.def_qos_parm.qos_flags);
 
-	ret = iwl_dvm_send_cmd_pdu(priv, ctx->qos_cmd, CMD_SYNC,
+	ret = iwl_dvm_send_cmd_pdu(priv, ctx->qos_cmd, 0,
 			       sizeof(struct iwl_qosparam_cmd),
 			       &ctx->qos_data.def_qos_parm);
 	if (ret)
@@ -353,7 +353,7 @@ static int iwl_send_rxon_timing(struct iwl_priv *priv,
 			le16_to_cpu(ctx->timing.atim_window));
 
 	return iwl_dvm_send_cmd_pdu(priv, ctx->rxon_timing_cmd,
-				CMD_SYNC, sizeof(ctx->timing), &ctx->timing);
+				0, sizeof(ctx->timing), &ctx->timing);
 }
 
 static int iwlagn_rxon_disconn(struct iwl_priv *priv,
@@ -495,7 +495,7 @@ static int iwlagn_rxon_connect(struct iwl_priv *priv,
 	 * Associated RXON doesn't clear the station table in uCode,
 	 * so we don't need to restore stations etc. after this.
 	 */
-	ret = iwl_dvm_send_cmd_pdu(priv, ctx->rxon_cmd, CMD_SYNC,
+	ret = iwl_dvm_send_cmd_pdu(priv, ctx->rxon_cmd, 0,
 		      sizeof(struct iwl_rxon_cmd), &ctx->staging);
 	if (ret) {
 		IWL_ERR(priv, "Error setting new RXON (%d)\n", ret);
@@ -610,7 +610,7 @@ int iwlagn_set_pan_params(struct iwl_priv *priv)
 	cmd.slots[0].width = cpu_to_le16(slot0);
 	cmd.slots[1].width = cpu_to_le16(slot1);
 
-	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_WIPAN_PARAMS, CMD_SYNC,
+	ret = iwl_dvm_send_cmd_pdu(priv, REPLY_WIPAN_PARAMS, 0,
 			sizeof(cmd), &cmd);
 	if (ret)
 		IWL_ERR(priv, "Error setting PAN parameters (%d)\n", ret);
@@ -823,7 +823,7 @@ static int iwl_check_rxon_cmd(struct iwl_priv *priv,
 
 	if ((rxon->flags & (RXON_FLG_CCK_MSK | RXON_FLG_AUTO_DETECT_MSK))
 			== (RXON_FLG_CCK_MSK | RXON_FLG_AUTO_DETECT_MSK)) {
-		IWL_WARN(priv, "CCK and auto detect");
+		IWL_WARN(priv, "CCK and auto detect\n");
 		errors |= BIT(8);
 	}
 
@@ -1395,7 +1395,7 @@ static void iwlagn_chain_noise_reset(struct iwl_priv *priv)
 			priv->phy_calib_chain_noise_reset_cmd);
 		ret = iwl_dvm_send_cmd_pdu(priv,
 					REPLY_PHY_CALIBRATION_CMD,
-					CMD_SYNC, sizeof(cmd), &cmd);
+					0, sizeof(cmd), &cmd);
 		if (ret)
 			IWL_ERR(priv,
 				"Could not send REPLY_PHY_CALIBRATION_CMD\n");
diff --git a/drivers/net/wireless/iwlwifi/dvm/scan.c b/drivers/net/wireless/iwlwifi/dvm/scan.c
index be98b913ed58..43bef901e8f9 100644
--- a/drivers/net/wireless/iwlwifi/dvm/scan.c
+++ b/drivers/net/wireless/iwlwifi/dvm/scan.c
@@ -59,7 +59,7 @@ static int iwl_send_scan_abort(struct iwl_priv *priv)
 	int ret;
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_SCAN_ABORT_CMD,
-		.flags = CMD_SYNC | CMD_WANT_SKB,
+		.flags = CMD_WANT_SKB,
 	};
 	__le32 *status;
 
@@ -639,7 +639,6 @@ static int iwlagn_request_scan(struct iwl_priv *priv, struct ieee80211_vif *vif)
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_SCAN_CMD,
 		.len = { sizeof(struct iwl_scan_cmd), },
-		.flags = CMD_SYNC,
 	};
 	struct iwl_scan_cmd *scan;
 	struct iwl_rxon_context *ctx = &priv->contexts[IWL_RXON_CTX_BSS];
diff --git a/drivers/net/wireless/iwlwifi/dvm/sta.c b/drivers/net/wireless/iwlwifi/dvm/sta.c
index 9cdd91cdf661..6ec86adbe4a1 100644
--- a/drivers/net/wireless/iwlwifi/dvm/sta.c
+++ b/drivers/net/wireless/iwlwifi/dvm/sta.c
@@ -39,7 +39,7 @@ static int iwl_sta_ucode_activate(struct iwl_priv *priv, u8 sta_id)
 	lockdep_assert_held(&priv->sta_lock);
 
 	if (sta_id >= IWLAGN_STATION_COUNT) {
-		IWL_ERR(priv, "invalid sta_id %u", sta_id);
+		IWL_ERR(priv, "invalid sta_id %u\n", sta_id);
 		return -EINVAL;
 	}
 	if (!(priv->stations[sta_id].used & IWL_STA_DRIVER_ACTIVE))
@@ -165,7 +165,7 @@ int iwl_send_add_sta(struct iwl_priv *priv,
 	iwl_free_resp(&cmd);
 
 	if (cmd.handler_status)
-		IWL_ERR(priv, "%s - error in the CMD response %d", __func__,
+		IWL_ERR(priv, "%s - error in the CMD response %d\n", __func__,
 			cmd.handler_status);
 
 	return cmd.handler_status;
@@ -261,7 +261,7 @@ int iwl_sta_update_ht(struct iwl_priv *priv, struct iwl_rxon_context *ctx,
 	cmd.station_flags = flags;
 	cmd.sta.sta_id = sta_id;
 
-	return iwl_send_add_sta(priv, &cmd, CMD_SYNC);
+	return iwl_send_add_sta(priv, &cmd, 0);
 }
 
 static void iwl_set_ht_add_station(struct iwl_priv *priv, u8 index,
@@ -413,7 +413,7 @@ int iwl_add_station_common(struct iwl_priv *priv, struct iwl_rxon_context *ctx,
 	spin_unlock_bh(&priv->sta_lock);
 
 	/* Add station to device's station table */
-	ret = iwl_send_add_sta(priv, &sta_cmd, CMD_SYNC);
+	ret = iwl_send_add_sta(priv, &sta_cmd, 0);
 	if (ret) {
 		spin_lock_bh(&priv->sta_lock);
 		IWL_ERR(priv, "Adding station %pM failed.\n",
@@ -456,7 +456,6 @@ static int iwl_send_remove_station(struct iwl_priv *priv,
 	struct iwl_host_cmd cmd = {
 		.id = REPLY_REMOVE_STA,
 		.len = { sizeof(struct iwl_rem_sta_cmd), },
-		.flags = CMD_SYNC,
 		.data = { &rm_sta_cmd, },
 	};
 
@@ -740,7 +739,7 @@ void iwl_restore_stations(struct iwl_priv *priv, struct iwl_rxon_context *ctx)
 					send_lq = true;
 			}
 			spin_unlock_bh(&priv->sta_lock);
-			ret = iwl_send_add_sta(priv, &sta_cmd, CMD_SYNC);
+			ret = iwl_send_add_sta(priv, &sta_cmd, 0);
 			if (ret) {
 				spin_lock_bh(&priv->sta_lock);
 				IWL_ERR(priv, "Adding station %pM failed.\n",
@@ -756,8 +755,7 @@ void iwl_restore_stations(struct iwl_priv *priv, struct iwl_rxon_context *ctx)
 			 * current LQ command
 			 */
 			if (send_lq)
-				iwl_send_lq_cmd(priv, ctx, &lq,
-						CMD_SYNC, true);
+				iwl_send_lq_cmd(priv, ctx, &lq, 0, true);
 			spin_lock_bh(&priv->sta_lock);
 			priv->stations[i].used &= ~IWL_STA_UCODE_INPROGRESS;
 		}
@@ -968,7 +966,7 @@ int iwlagn_add_bssid_station(struct iwl_priv *priv,
 		return -ENOMEM;
 	}
 
-	ret = iwl_send_lq_cmd(priv, ctx, link_cmd, CMD_SYNC, true);
+	ret = iwl_send_lq_cmd(priv, ctx, link_cmd, 0, true);
 	if (ret)
 		IWL_ERR(priv, "Link quality command failed (%d)\n", ret);
 
@@ -999,7 +997,6 @@ static int iwl_send_static_wepkey_cmd(struct iwl_priv *priv,
 	struct iwl_host_cmd cmd = {
 		.id = ctx->wep_key_cmd,
 		.data = { wep_cmd, },
-		.flags = CMD_SYNC,
 	};
 
 	might_sleep();
@@ -1248,7 +1245,7 @@ int iwl_remove_dynamic_key(struct iwl_priv *priv,
 	sta_cmd.sta.modify_mask = STA_MODIFY_KEY_MASK;
 	sta_cmd.mode = STA_CONTROL_MODIFY_MSK;
 
-	return iwl_send_add_sta(priv, &sta_cmd, CMD_SYNC);
+	return iwl_send_add_sta(priv, &sta_cmd, 0);
 }
 
 int iwl_set_dynamic_key(struct iwl_priv *priv,
@@ -1284,13 +1281,13 @@ int iwl_set_dynamic_key(struct iwl_priv *priv,
 		ieee80211_get_key_rx_seq(keyconf, 0, &seq);
 		ieee80211_get_tkip_rx_p1k(keyconf, addr, seq.tkip.iv32, p1k);
 		ret = iwlagn_send_sta_key(priv, keyconf, sta_id,
-					  seq.tkip.iv32, p1k, CMD_SYNC);
+					  seq.tkip.iv32, p1k, 0);
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 	case WLAN_CIPHER_SUITE_WEP40:
 	case WLAN_CIPHER_SUITE_WEP104:
 		ret = iwlagn_send_sta_key(priv, keyconf, sta_id,
-					  0, NULL, CMD_SYNC);
+					  0, NULL, 0);
 		break;
 	default:
 		IWL_ERR(priv, "Unknown cipher %x\n", keyconf->cipher);
@@ -1409,7 +1406,7 @@ int iwl_sta_tx_modify_enable_tid(struct iwl_priv *priv, int sta_id, int tid)
 	memcpy(&sta_cmd, &priv->stations[sta_id].sta, sizeof(struct iwl_addsta_cmd));
 	spin_unlock_bh(&priv->sta_lock);
 
-	return iwl_send_add_sta(priv, &sta_cmd, CMD_SYNC);
+	return iwl_send_add_sta(priv, &sta_cmd, 0);
 }
 
 int iwl_sta_rx_agg_start(struct iwl_priv *priv, struct ieee80211_sta *sta,
@@ -1433,7 +1430,7 @@ int iwl_sta_rx_agg_start(struct iwl_priv *priv, struct ieee80211_sta *sta,
 	memcpy(&sta_cmd, &priv->stations[sta_id].sta, sizeof(struct iwl_addsta_cmd));
 	spin_unlock_bh(&priv->sta_lock);
 
-	return iwl_send_add_sta(priv, &sta_cmd, CMD_SYNC);
+	return iwl_send_add_sta(priv, &sta_cmd, 0);
 }
 
 int iwl_sta_rx_agg_stop(struct iwl_priv *priv, struct ieee80211_sta *sta,
@@ -1458,7 +1455,7 @@ int iwl_sta_rx_agg_stop(struct iwl_priv *priv, struct ieee80211_sta *sta,
 	memcpy(&sta_cmd, &priv->stations[sta_id].sta, sizeof(struct iwl_addsta_cmd));
 	spin_unlock_bh(&priv->sta_lock);
 
-	return iwl_send_add_sta(priv, &sta_cmd, CMD_SYNC);
+	return iwl_send_add_sta(priv, &sta_cmd, 0);
 }
 
 
diff --git a/drivers/net/wireless/iwlwifi/dvm/tt.c b/drivers/net/wireless/iwlwifi/dvm/tt.c
index 058c5892c427..acb981a0a0aa 100644
--- a/drivers/net/wireless/iwlwifi/dvm/tt.c
+++ b/drivers/net/wireless/iwlwifi/dvm/tt.c
@@ -236,7 +236,7 @@ static void iwl_prepare_ct_kill_task(struct iwl_priv *priv)
 {
 	IWL_DEBUG_TEMP(priv, "Prepare to enter IWL_TI_CT_KILL\n");
 	/* make request to retrieve statistics information */
-	iwl_send_statistics_request(priv, CMD_SYNC, false);
+	iwl_send_statistics_request(priv, 0, false);
 	/* Reschedule the ct_kill wait timer */
 	mod_timer(&priv->thermal_throttle.ct_kill_waiting_tm,
 		 jiffies + msecs_to_jiffies(CT_KILL_WAITING_DURATION));
diff --git a/drivers/net/wireless/iwlwifi/dvm/tx.c b/drivers/net/wireless/iwlwifi/dvm/tx.c
index 398dd096674c..3255a1723d17 100644
--- a/drivers/net/wireless/iwlwifi/dvm/tx.c
+++ b/drivers/net/wireless/iwlwifi/dvm/tx.c
@@ -402,10 +402,10 @@ int iwlagn_tx_skb(struct iwl_priv *priv,
 		/* aggregation is on for this <sta,tid> */
 		if (info->flags & IEEE80211_TX_CTL_AMPDU &&
 		    tid_data->agg.state != IWL_AGG_ON) {
-			IWL_ERR(priv, "TX_CTL_AMPDU while not in AGG:"
-				" Tx flags = 0x%08x, agg.state = %d",
+			IWL_ERR(priv,
+				"TX_CTL_AMPDU while not in AGG: Tx flags = 0x%08x, agg.state = %d\n",
 				info->flags, tid_data->agg.state);
-			IWL_ERR(priv, "sta_id = %d, tid = %d seq_num = %d",
+			IWL_ERR(priv, "sta_id = %d, tid = %d seq_num = %d\n",
 				sta_id, tid,
 				IEEE80211_SEQ_TO_SN(tid_data->seq_number));
 			goto drop_unlock_sta;
@@ -416,7 +416,7 @@ int iwlagn_tx_skb(struct iwl_priv *priv,
 		 */
 		if (WARN_ONCE(tid_data->agg.state != IWL_AGG_ON &&
 			      tid_data->agg.state != IWL_AGG_OFF,
-		    "Tx while agg.state = %d", tid_data->agg.state))
+			      "Tx while agg.state = %d\n", tid_data->agg.state))
 			goto drop_unlock_sta;
 
 		seq_number = tid_data->seq_number;
@@ -778,8 +778,8 @@ static void iwlagn_check_ratid_empty(struct iwl_priv *priv, int sta_id, u8 tid)
 		/* There are no packets for this RA / TID in the HW any more */
 		if (tid_data->agg.ssn == tid_data->next_reclaimed) {
 			IWL_DEBUG_TX_QUEUES(priv,
-				"Can continue DELBA flow ssn = next_recl ="
-				" %d", tid_data->next_reclaimed);
+				"Can continue DELBA flow ssn = next_recl = %d\n",
+				tid_data->next_reclaimed);
 			iwl_trans_txq_disable(priv->trans,
 					      tid_data->agg.txq_id);
 			iwlagn_dealloc_agg_txq(priv, tid_data->agg.txq_id);
@@ -791,8 +791,8 @@ static void iwlagn_check_ratid_empty(struct iwl_priv *priv, int sta_id, u8 tid)
 		/* There are no packets for this RA / TID in the HW any more */
 		if (tid_data->agg.ssn == tid_data->next_reclaimed) {
 			IWL_DEBUG_TX_QUEUES(priv,
-				"Can continue ADDBA flow ssn = next_recl ="
-				" %d", tid_data->next_reclaimed);
+				"Can continue ADDBA flow ssn = next_recl = %d\n",
+				tid_data->next_reclaimed);
 			tid_data->agg.state = IWL_AGG_STARTING;
 			ieee80211_start_tx_ba_cb_irqsafe(vif, addr, tid);
 		}
@@ -1216,8 +1216,8 @@ int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_cmd_buffer *rxb,
 			    ctx->vif->type == NL80211_IFTYPE_STATION) {
 				/* block and stop all queues */
 				priv->passive_no_rx = true;
-				IWL_DEBUG_TX_QUEUES(priv, "stop all queues: "
-						    "passive channel");
+				IWL_DEBUG_TX_QUEUES(priv,
+					"stop all queues: passive channel\n");
 				ieee80211_stop_queues(priv->hw);
 
 				IWL_DEBUG_TX_REPLY(priv,
@@ -1271,7 +1271,7 @@ int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_cmd_buffer *rxb,
 
 	while (!skb_queue_empty(&skbs)) {
 		skb = __skb_dequeue(&skbs);
-		ieee80211_tx_status_ni(priv->hw, skb);
+		ieee80211_tx_status(priv->hw, skb);
 	}
 
 	return 0;
@@ -1411,7 +1411,7 @@ int iwlagn_rx_reply_compressed_ba(struct iwl_priv *priv,
 
 	while (!skb_queue_empty(&reclaimed_skbs)) {
 		skb = __skb_dequeue(&reclaimed_skbs);
-		ieee80211_tx_status_ni(priv->hw, skb);
+		ieee80211_tx_status(priv->hw, skb);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/iwlwifi/dvm/ucode.c b/drivers/net/wireless/iwlwifi/dvm/ucode.c
index cf03ef5619d9..d5cee1530597 100644
--- a/drivers/net/wireless/iwlwifi/dvm/ucode.c
+++ b/drivers/net/wireless/iwlwifi/dvm/ucode.c
@@ -172,7 +172,7 @@ static int iwl_send_wimax_coex(struct iwl_priv *priv)
 	memset(&coex_cmd, 0, sizeof(coex_cmd));
 
 	return iwl_dvm_send_cmd_pdu(priv,
-				COEX_PRIORITY_TABLE_CMD, CMD_SYNC,
+				COEX_PRIORITY_TABLE_CMD, 0,
 				sizeof(coex_cmd), &coex_cmd);
 }
 
@@ -205,7 +205,7 @@ void iwl_send_prio_tbl(struct iwl_priv *priv)
 	memcpy(prio_tbl_cmd.prio_tbl, iwl_bt_prio_tbl,
 		sizeof(iwl_bt_prio_tbl));
 	if (iwl_dvm_send_cmd_pdu(priv,
-				REPLY_BT_COEX_PRIO_TABLE, CMD_SYNC,
+				REPLY_BT_COEX_PRIO_TABLE, 0,
 				sizeof(prio_tbl_cmd), &prio_tbl_cmd))
 		IWL_ERR(priv, "failed to send BT prio tbl command\n");
 }
@@ -218,7 +218,7 @@ int iwl_send_bt_env(struct iwl_priv *priv, u8 action, u8 type)
 	env_cmd.action = action;
 	env_cmd.type = type;
 	ret = iwl_dvm_send_cmd_pdu(priv,
-			       REPLY_BT_COEX_PROT_ENV, CMD_SYNC,
+			       REPLY_BT_COEX_PROT_ENV, 0,
 			       sizeof(env_cmd), &env_cmd);
 	if (ret)
 		IWL_ERR(priv, "failed to send BT env command\n");
diff --git a/drivers/net/wireless/iwlwifi/iwl-1000.c b/drivers/net/wireless/iwlwifi/iwl-1000.c
index 854ba84ccb73..c3817fae16c0 100644
--- a/drivers/net/wireless/iwlwifi/iwl-1000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-1000.c
@@ -62,6 +62,7 @@ static const struct iwl_base_params iwl1000_base_params = {
 	.led_compensation = 51,
 	.wd_timeout = IWL_WATCHDOG_DISABLED,
 	.max_event_log_size = 128,
+	.scd_chain_ext_wa = true,
 };
 
 static const struct iwl_ht_params iwl1000_ht_params = {
diff --git a/drivers/net/wireless/iwlwifi/iwl-2000.c b/drivers/net/wireless/iwlwifi/iwl-2000.c
index 3e63323637f3..21e5d0843a62 100644
--- a/drivers/net/wireless/iwlwifi/iwl-2000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-2000.c
@@ -75,6 +75,7 @@ static const struct iwl_base_params iwl2000_base_params = {
 	.wd_timeout = IWL_DEF_WD_TIMEOUT,
 	.max_event_log_size = 512,
 	.shadow_reg_enable = false, /* TODO: fix bugs using this feature */
+	.scd_chain_ext_wa = true,
 };
 
 
@@ -88,6 +89,7 @@ static const struct iwl_base_params iwl2030_base_params = {
 	.wd_timeout = IWL_LONG_WD_TIMEOUT,
 	.max_event_log_size = 512,
 	.shadow_reg_enable = false, /* TODO: fix bugs using this feature */
+	.scd_chain_ext_wa = true,
 };
 
 static const struct iwl_ht_params iwl2000_ht_params = {
diff --git a/drivers/net/wireless/iwlwifi/iwl-5000.c b/drivers/net/wireless/iwlwifi/iwl-5000.c
index 6674f2c4541c..332bbede39e5 100644
--- a/drivers/net/wireless/iwlwifi/iwl-5000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-5000.c
@@ -61,6 +61,7 @@ static const struct iwl_base_params iwl5000_base_params = {
 	.led_compensation = 51,
 	.wd_timeout = IWL_WATCHDOG_DISABLED,
 	.max_event_log_size = 512,
+	.scd_chain_ext_wa = true,
 };
 
 static const struct iwl_ht_params iwl5000_ht_params = {
diff --git a/drivers/net/wireless/iwlwifi/iwl-6000.c b/drivers/net/wireless/iwlwifi/iwl-6000.c
index 8048de90233f..8f2c3c8c6b84 100644
--- a/drivers/net/wireless/iwlwifi/iwl-6000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-6000.c
@@ -85,6 +85,7 @@ static const struct iwl_base_params iwl6000_base_params = {
 	.wd_timeout = IWL_DEF_WD_TIMEOUT,
 	.max_event_log_size = 512,
 	.shadow_reg_enable = false, /* TODO: fix bugs using this feature */
+	.scd_chain_ext_wa = true,
 };
 
 static const struct iwl_base_params iwl6050_base_params = {
@@ -97,6 +98,7 @@ static const struct iwl_base_params iwl6050_base_params = {
 	.wd_timeout = IWL_DEF_WD_TIMEOUT,
 	.max_event_log_size = 1024,
 	.shadow_reg_enable = false, /* TODO: fix bugs using this feature */
+	.scd_chain_ext_wa = true,
 };
 
 static const struct iwl_base_params iwl6000_g2_base_params = {
@@ -109,6 +111,7 @@ static const struct iwl_base_params iwl6000_g2_base_params = {
 	.wd_timeout = IWL_LONG_WD_TIMEOUT,
 	.max_event_log_size = 512,
 	.shadow_reg_enable = false, /* TODO: fix bugs using this feature */
+	.scd_chain_ext_wa = true,
 };
 
 static const struct iwl_ht_params iwl6000_ht_params = {
diff --git a/drivers/net/wireless/iwlwifi/iwl-7000.c b/drivers/net/wireless/iwlwifi/iwl-7000.c
index 4c2d4ef28b22..48730064da73 100644
--- a/drivers/net/wireless/iwlwifi/iwl-7000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-7000.c
@@ -71,12 +71,12 @@
 #define IWL3160_UCODE_API_MAX	9
 
 /* Oldest version we won't warn about */
-#define IWL7260_UCODE_API_OK	8
-#define IWL3160_UCODE_API_OK	8
+#define IWL7260_UCODE_API_OK	9
+#define IWL3160_UCODE_API_OK	9
 
 /* Lowest firmware API version supported */
-#define IWL7260_UCODE_API_MIN	7
-#define IWL3160_UCODE_API_MIN	7
+#define IWL7260_UCODE_API_MIN	8
+#define IWL3160_UCODE_API_MIN	8
 
 /* NVM versions */
 #define IWL7260_NVM_VERSION		0x0a1d
@@ -98,7 +98,7 @@
 #define NVM_HW_SECTION_NUM_FAMILY_7000		0
 
 static const struct iwl_base_params iwl7000_base_params = {
-	.eeprom_size = OTP_LOW_IMAGE_SIZE,
+	.eeprom_size = OTP_LOW_IMAGE_SIZE_FAMILY_7000,
 	.num_of_queues = IWLAGN_NUM_QUEUES,
 	.pll_cfg_val = 0,
 	.shadow_ram_support = true,
@@ -107,6 +107,7 @@ static const struct iwl_base_params iwl7000_base_params = {
 	.max_event_log_size = 512,
 	.shadow_reg_enable = true,
 	.pcie_l1_allowed = true,
+	.apmg_wake_up_wa = true,
 };
 
 static const struct iwl_ht_params iwl7000_ht_params = {
diff --git a/drivers/net/wireless/iwlwifi/iwl-8000.c b/drivers/net/wireless/iwlwifi/iwl-8000.c
index f5bd82b88592..51c41531d81d 100644
--- a/drivers/net/wireless/iwlwifi/iwl-8000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-8000.c
@@ -83,9 +83,10 @@
 #define IWL8000_MODULE_FIRMWARE(api) IWL8000_FW_PRE __stringify(api) ".ucode"
 
 #define NVM_HW_SECTION_NUM_FAMILY_8000		10
+#define DEFAULT_NVM_FILE_FAMILY_8000		"iwl_nvm_8000.bin"
 
 static const struct iwl_base_params iwl8000_base_params = {
-	.eeprom_size = OTP_LOW_IMAGE_SIZE,
+	.eeprom_size = OTP_LOW_IMAGE_SIZE_FAMILY_8000,
 	.num_of_queues = IWLAGN_NUM_QUEUES,
 	.pll_cfg_val = 0,
 	.shadow_ram_support = true,
@@ -118,6 +119,7 @@ const struct iwl_cfg iwl8260_2ac_cfg = {
 	.ht_params = &iwl8000_ht_params,
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
+	.default_nvm_file = DEFAULT_NVM_FILE_FAMILY_8000,
 };
 
 const struct iwl_cfg iwl8260_n_cfg = {
@@ -127,6 +129,7 @@ const struct iwl_cfg iwl8260_n_cfg = {
 	.ht_params = &iwl8000_ht_params,
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
+	.default_nvm_file = DEFAULT_NVM_FILE_FAMILY_8000,
 };
 
 MODULE_FIRMWARE(IWL8000_MODULE_FIRMWARE(IWL8000_UCODE_API_OK));
diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-hw.h b/drivers/net/wireless/iwlwifi/iwl-agn-hw.h
index 7f37fb86837b..04a483d38659 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn-hw.h
+++ b/drivers/net/wireless/iwlwifi/iwl-agn-hw.h
@@ -102,9 +102,7 @@
 
 /* EEPROM */
 #define IWLAGN_EEPROM_IMG_SIZE		2048
-/* OTP */
-/* lower blocks contain EEPROM image and calibration data */
-#define OTP_LOW_IMAGE_SIZE		(2 * 512 * sizeof(u16)) /* 2 KB */
+
 /* high blocks contain PAPD data */
 #define OTP_HIGH_IMAGE_SIZE_6x00        (6 * 512 * sizeof(u16)) /* 6 KB */
 #define OTP_HIGH_IMAGE_SIZE_1000        (0x200 * sizeof(u16)) /* 1024 bytes */
diff --git a/drivers/net/wireless/iwlwifi/iwl-config.h b/drivers/net/wireless/iwlwifi/iwl-config.h
index 3f17dc3f2c8a..b7047905f41a 100644
--- a/drivers/net/wireless/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/iwlwifi/iwl-config.h
@@ -146,6 +146,9 @@ static inline u8 num_of_ant(u8 mask)
  * @wd_timeout: TX queues watchdog timeout
  * @max_event_log_size: size of event log buffer size for ucode event logging
  * @shadow_reg_enable: HW shadow register support
+ * @apmg_wake_up_wa: should the MAC access REQ be asserted when a command
+ *	is in flight. This is due to a HW bug in 7260, 3160 and 7265.
+ * @scd_chain_ext_wa: should the chain extension feature in SCD be disabled.
  */
 struct iwl_base_params {
 	int eeprom_size;
@@ -160,6 +163,8 @@ struct iwl_base_params {
 	u32 max_event_log_size;
 	const bool shadow_reg_enable;
 	const bool pcie_l1_allowed;
+	const bool apmg_wake_up_wa;
+	const bool scd_chain_ext_wa;
 };
 
 /*
@@ -188,6 +193,11 @@ struct iwl_ht_params {
 #define EEPROM_6000_REG_BAND_24_HT40_CHANNELS	0x80
 #define EEPROM_REGULATORY_BAND_NO_HT40		0
 
+/* lower blocks contain EEPROM image and calibration data */
+#define OTP_LOW_IMAGE_SIZE		(2 * 512 * sizeof(u16)) /* 2 KB */
+#define OTP_LOW_IMAGE_SIZE_FAMILY_7000	(16 * 512 * sizeof(u16)) /* 16 KB */
+#define OTP_LOW_IMAGE_SIZE_FAMILY_8000	(32 * 512 * sizeof(u16)) /* 32 KB */
+
 struct iwl_eeprom_params {
 	const u8 regulatory_bands[7];
 	bool enhanced_txpower;
@@ -264,6 +274,8 @@ struct iwl_cfg {
 	u8   nvm_hw_section_num;
 	bool lp_xtal_workaround;
 	const struct iwl_pwr_tx_backoff *pwr_tx_backoffs;
+	bool no_power_up_nic_in_init;
+	const char *default_nvm_file;
 };
 
 /*
diff --git a/drivers/net/wireless/iwlwifi/iwl-debug.c b/drivers/net/wireless/iwlwifi/iwl-debug.c
index 8a44f594528d..09feff4fa226 100644
--- a/drivers/net/wireless/iwlwifi/iwl-debug.c
+++ b/drivers/net/wireless/iwlwifi/iwl-debug.c
@@ -61,8 +61,6 @@
  *
  *****************************************************************************/
 
-#define DEBUG
-
 #include <linux/device.h>
 #include <linux/interrupt.h>
 #include <linux/export.h>
@@ -128,8 +126,8 @@ void __iwl_dbg(struct device *dev,
 #ifdef CONFIG_IWLWIFI_DEBUG
 	if (iwl_have_debug_level(level) &&
 	    (!limit || net_ratelimit()))
-		dev_dbg(dev, "%c %s %pV", in_interrupt() ? 'I' : 'U',
-			function, &vaf);
+		dev_printk(KERN_DEBUG, dev, "%c %s %pV",
+			   in_interrupt() ? 'I' : 'U', function, &vaf);
 #endif
 	trace_iwlwifi_dbg(level, in_interrupt(), function, &vaf);
 	va_end(args);
diff --git a/drivers/net/wireless/iwlwifi/iwl-debug.h b/drivers/net/wireless/iwlwifi/iwl-debug.h
index c8cbdbe15924..295083510e72 100644
--- a/drivers/net/wireless/iwlwifi/iwl-debug.h
+++ b/drivers/net/wireless/iwlwifi/iwl-debug.h
@@ -47,12 +47,32 @@ void __iwl_warn(struct device *dev, const char *fmt, ...) __printf(2, 3);
 void __iwl_info(struct device *dev, const char *fmt, ...) __printf(2, 3);
 void __iwl_crit(struct device *dev, const char *fmt, ...) __printf(2, 3);
 
+/* not all compilers can evaluate strlen() at compile time, so use sizeof() */
+#define CHECK_FOR_NEWLINE(f) BUILD_BUG_ON(f[sizeof(f) - 2] != '\n')
+
 /* No matter what is m (priv, bus, trans), this will work */
-#define IWL_ERR(m, f, a...) __iwl_err((m)->dev, false, false, f, ## a)
-#define IWL_ERR_DEV(d, f, a...) __iwl_err((d), false, false, f, ## a)
-#define IWL_WARN(m, f, a...) __iwl_warn((m)->dev, f, ## a)
-#define IWL_INFO(m, f, a...) __iwl_info((m)->dev, f, ## a)
-#define IWL_CRIT(m, f, a...) __iwl_crit((m)->dev, f, ## a)
+#define IWL_ERR_DEV(d, f, a...)						\
+	do {								\
+		CHECK_FOR_NEWLINE(f);					\
+		__iwl_err((d), false, false, f, ## a);			\
+	} while (0)
+#define IWL_ERR(m, f, a...)						\
+	IWL_ERR_DEV((m)->dev, f, ## a)
+#define IWL_WARN(m, f, a...)						\
+	do {								\
+		CHECK_FOR_NEWLINE(f);					\
+		__iwl_warn((m)->dev, f, ## a);				\
+	} while (0)
+#define IWL_INFO(m, f, a...)						\
+	do {								\
+		CHECK_FOR_NEWLINE(f);					\
+		__iwl_info((m)->dev, f, ## a);				\
+	} while (0)
+#define IWL_CRIT(m, f, a...)						\
+	do {								\
+		CHECK_FOR_NEWLINE(f);					\
+		__iwl_crit((m)->dev, f, ## a);				\
+	} while (0)
 
 #if defined(CONFIG_IWLWIFI_DEBUG) || defined(CONFIG_IWLWIFI_DEVICE_TRACING)
 void __iwl_dbg(struct device *dev,
@@ -72,12 +92,17 @@ do {									\
 		       DUMP_PREFIX_OFFSET, 16, 1, p, len, 1);		\
 } while (0)
 
+#define __IWL_DEBUG_DEV(dev, level, limit, fmt, args...)		\
+	do {								\
+		CHECK_FOR_NEWLINE(fmt);					\
+		__iwl_dbg(dev, level, limit, __func__, fmt, ##args);	\
+	} while (0)
 #define IWL_DEBUG(m, level, fmt, args...)				\
-	__iwl_dbg((m)->dev, level, false, __func__, fmt, ##args)
+	__IWL_DEBUG_DEV((m)->dev, level, false, fmt, ##args)
 #define IWL_DEBUG_DEV(dev, level, fmt, args...)				\
-	__iwl_dbg((dev), level, false, __func__, fmt, ##args)
+	__IWL_DEBUG_DEV(dev, level, false, fmt, ##args)
 #define IWL_DEBUG_LIMIT(m, level, fmt, args...)				\
-	__iwl_dbg((m)->dev, level, true, __func__, fmt, ##args)
+	__IWL_DEBUG_DEV((m)->dev, level, true, fmt, ##args)
 
 #ifdef CONFIG_IWLWIFI_DEBUG
 #define iwl_print_hex_dump(m, level, p, len)				\
diff --git a/drivers/net/wireless/iwlwifi/iwl-drv.c b/drivers/net/wireless/iwlwifi/iwl-drv.c
index 0a3e841b44a9..f2a5c12269a3 100644
--- a/drivers/net/wireless/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/iwlwifi/iwl-drv.c
@@ -1243,6 +1243,7 @@ struct iwl_mod_params iwlwifi_mod_params = {
 	.bt_coex_active = true,
 	.power_level = IWL_POWER_INDEX_1,
 	.wd_disable = true,
+	.uapsd_disable = false,
 	/* the rest are 0 by default */
 };
 IWL_EXPORT_SYMBOL(iwlwifi_mod_params);
@@ -1356,6 +1357,10 @@ MODULE_PARM_DESC(wd_disable,
 module_param_named(nvm_file, iwlwifi_mod_params.nvm_file, charp, S_IRUGO);
 MODULE_PARM_DESC(nvm_file, "NVM file name");
 
+module_param_named(uapsd_disable, iwlwifi_mod_params.uapsd_disable,
+		   bool, S_IRUGO);
+MODULE_PARM_DESC(uapsd_disable, "disable U-APSD functionality (default: N)");
+
 /*
  * set bt_coex_active to true, uCode will do kill/defer
  * every time the priority line is asserted (BT is sending signals on the
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-error-dump.h b/drivers/net/wireless/iwlwifi/iwl-fw-error-dump.h
index 58c8941c0d95..2953ffceda38 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-error-dump.h
+++ b/drivers/net/wireless/iwlwifi/iwl-fw-error-dump.h
@@ -71,10 +71,15 @@
  * enum iwl_fw_error_dump_type - types of data in the dump file
  * @IWL_FW_ERROR_DUMP_SRAM:
  * @IWL_FW_ERROR_DUMP_REG:
+ * @IWL_FW_ERROR_DUMP_RXF:
+ * @IWL_FW_ERROR_DUMP_TXCMD: last TX command data, structured as
+ *	&struct iwl_fw_error_dump_txcmd packets
  */
 enum iwl_fw_error_dump_type {
 	IWL_FW_ERROR_DUMP_SRAM = 0,
 	IWL_FW_ERROR_DUMP_REG = 1,
+	IWL_FW_ERROR_DUMP_RXF = 2,
+	IWL_FW_ERROR_DUMP_TXCMD = 3,
 
 	IWL_FW_ERROR_DUMP_MAX,
 };
@@ -89,7 +94,7 @@ struct iwl_fw_error_dump_data {
 	__le32 type;
 	__le32 len;
 	__u8 data[];
-} __packed __aligned(4);
+} __packed;
 
 /**
  * struct iwl_fw_error_dump_file - the layout of the header of the file
@@ -101,6 +106,29 @@ struct iwl_fw_error_dump_file {
 	__le32 barker;
 	__le32 file_len;
 	u8 data[0];
-} __packed __aligned(4);
+} __packed;
+
+/**
+ * struct iwl_fw_error_dump_txcmd - TX command data
+ * @cmdlen: original length of command
+ * @caplen: captured length of command (may be less)
+ * @data: captured command data, @caplen bytes
+ */
+struct iwl_fw_error_dump_txcmd {
+	__le32 cmdlen;
+	__le32 caplen;
+	u8 data[];
+} __packed;
+
+/**
+ * iwl_mvm_fw_error_next_data - advance fw error dump data pointer
+ * @data: previous data block
+ * Returns: next data block
+ */
+static inline struct iwl_fw_error_dump_data *
+iwl_mvm_fw_error_next_data(struct iwl_fw_error_dump_data *data)
+{
+	return (void *)(data->data + le32_to_cpu(data->len));
+}
 
 #endif /* __fw_error_dump_h__ */
diff --git a/drivers/net/wireless/iwlwifi/iwl-fw.h b/drivers/net/wireless/iwlwifi/iwl-fw.h
index d14f19339d61..0aa7c0085c9f 100644
--- a/drivers/net/wireless/iwlwifi/iwl-fw.h
+++ b/drivers/net/wireless/iwlwifi/iwl-fw.h
@@ -74,29 +74,24 @@
  * @IWL_UCODE_TLV_FLAGS_MFP: This uCode image supports MFP (802.11w).
  * @IWL_UCODE_TLV_FLAGS_P2P: This uCode image supports P2P.
  * @IWL_UCODE_TLV_FLAGS_DW_BC_TABLE: The SCD byte count table is in DWORDS
- * @IWL_UCODE_TLV_FLAGS_UAPSD: This uCode image supports uAPSD
+ * @IWL_UCODE_TLV_FLAGS_UAPSD_SUPPORT: This uCode image supports uAPSD
  * @IWL_UCODE_TLV_FLAGS_SHORT_BL: 16 entries of black list instead of 64 in scan
  *	offload profile config command.
- * @IWL_UCODE_TLV_FLAGS_RX_ENERGY_API: supports rx signal strength api
- * @IWL_UCODE_TLV_FLAGS_TIME_EVENT_API_V2: using the new time event API.
  * @IWL_UCODE_TLV_FLAGS_D3_6_IPV6_ADDRS: D3 image supports up to six
  *	(rather than two) IPv6 addresses
- * @IWL_UCODE_TLV_FLAGS_BF_UPDATED: new beacon filtering API
  * @IWL_UCODE_TLV_FLAGS_NO_BASIC_SSID: not sending a probe with the SSID element
  *	from the probe request template.
- * @IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API: modified D3 API to allow keeping
- *	connection when going back to D0
  * @IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_SMALL: new NS offload (small version)
  * @IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_LARGE: new NS offload (large version)
- * @IWL_UCODE_TLV_FLAGS_SCHED_SCAN: this uCode image supports scheduled scan.
- * @IWL_UCODE_TLV_FLAGS_STA_KEY_CMD: new ADD_STA and ADD_STA_KEY command API
- * @IWL_UCODE_TLV_FLAGS_DEVICE_PS_CMD: support device wide power command
- *	containing CAM (Continuous Active Mode) indication.
+ * @IWL_UCODE_TLV_FLAGS_P2P_PM: P2P client supports PM as a stand alone MAC
  * @IWL_UCODE_TLV_FLAGS_P2P_BSS_PS_DCM: support power save on BSS station and
  *	P2P client interfaces simultaneously if they are in different bindings.
+ * @IWL_UCODE_TLV_FLAGS_P2P_BSS_PS_SCM: support power save on BSS station and
+ *	P2P client interfaces simultaneously if they are in same bindings.
  * @IWL_UCODE_TLV_FLAGS_P2P_PS_UAPSD: P2P client supports uAPSD power save
  * @IWL_UCODE_TLV_FLAGS_BCAST_FILTERING: uCode supports broadcast filtering.
  * @IWL_UCODE_TLV_FLAGS_GO_UAPSD: AP/GO interfaces support uAPSD clients
+ * @IWL_UCODE_TLV_FLAGS_EBS_SUPPORT: this uCode image supports EBS.
  */
 enum iwl_ucode_tlv_flag {
 	IWL_UCODE_TLV_FLAGS_PAN			= BIT(0),
@@ -104,22 +99,16 @@ enum iwl_ucode_tlv_flag {
 	IWL_UCODE_TLV_FLAGS_MFP			= BIT(2),
 	IWL_UCODE_TLV_FLAGS_P2P			= BIT(3),
 	IWL_UCODE_TLV_FLAGS_DW_BC_TABLE		= BIT(4),
-	IWL_UCODE_TLV_FLAGS_NEWBT_COEX		= BIT(5),
-	IWL_UCODE_TLV_FLAGS_PM_CMD_SUPPORT	= BIT(6),
 	IWL_UCODE_TLV_FLAGS_SHORT_BL		= BIT(7),
-	IWL_UCODE_TLV_FLAGS_RX_ENERGY_API	= BIT(8),
-	IWL_UCODE_TLV_FLAGS_TIME_EVENT_API_V2	= BIT(9),
 	IWL_UCODE_TLV_FLAGS_D3_6_IPV6_ADDRS	= BIT(10),
-	IWL_UCODE_TLV_FLAGS_BF_UPDATED		= BIT(11),
 	IWL_UCODE_TLV_FLAGS_NO_BASIC_SSID	= BIT(12),
-	IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API	= BIT(14),
 	IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_SMALL	= BIT(15),
 	IWL_UCODE_TLV_FLAGS_NEW_NSOFFL_LARGE	= BIT(16),
-	IWL_UCODE_TLV_FLAGS_SCHED_SCAN		= BIT(17),
-	IWL_UCODE_TLV_FLAGS_STA_KEY_CMD		= BIT(19),
-	IWL_UCODE_TLV_FLAGS_DEVICE_PS_CMD	= BIT(20),
+	IWL_UCODE_TLV_FLAGS_P2P_PM		= BIT(21),
 	IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_DCM	= BIT(22),
+	IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_SCM	= BIT(23),
 	IWL_UCODE_TLV_FLAGS_UAPSD_SUPPORT	= BIT(24),
+	IWL_UCODE_TLV_FLAGS_EBS_SUPPORT		= BIT(25),
 	IWL_UCODE_TLV_FLAGS_P2P_PS_UAPSD	= BIT(26),
 	IWL_UCODE_TLV_FLAGS_BCAST_FILTERING	= BIT(29),
 	IWL_UCODE_TLV_FLAGS_GO_UAPSD		= BIT(30),
@@ -128,9 +117,11 @@ enum iwl_ucode_tlv_flag {
 /**
  * enum iwl_ucode_tlv_api - ucode api
  * @IWL_UCODE_TLV_API_WOWLAN_CONFIG_TID: wowlan config includes tid field.
+ * @IWL_UCODE_TLV_API_CSA_FLOW: ucode can do unbind-bind flow for CSA.
  */
 enum iwl_ucode_tlv_api {
 	IWL_UCODE_TLV_API_WOWLAN_CONFIG_TID	= BIT(0),
+	IWL_UCODE_TLV_API_CSA_FLOW		= BIT(4),
 };
 
 /**
@@ -183,6 +174,7 @@ enum iwl_ucode_sec {
 #define IWL_UCODE_SECTION_MAX 12
 #define IWL_API_ARRAY_SIZE	1
 #define IWL_CAPABILITIES_ARRAY_SIZE	1
+#define CPU1_CPU2_SEPARATOR_SECTION	0xFFFFCCCC
 
 struct iwl_ucode_capabilities {
 	u32 max_probe_length;
@@ -205,6 +197,11 @@ struct fw_img {
 	bool is_dual_cpus;
 };
 
+struct iwl_sf_region {
+	u32 addr;
+	u32 size;
+};
+
 /* uCode version contains 4 values: Major/Minor/API/Serial */
 #define IWL_UCODE_MAJOR(ver)	(((ver) & 0xFF000000) >> 24)
 #define IWL_UCODE_MINOR(ver)	(((ver) & 0x00FF0000) >> 16)
diff --git a/drivers/net/wireless/iwlwifi/iwl-io.c b/drivers/net/wireless/iwlwifi/iwl-io.c
index 44cc3cf45762..5eef4ae7333b 100644
--- a/drivers/net/wireless/iwlwifi/iwl-io.c
+++ b/drivers/net/wireless/iwlwifi/iwl-io.c
@@ -33,6 +33,7 @@
 #include "iwl-io.h"
 #include "iwl-csr.h"
 #include "iwl-debug.h"
+#include "iwl-prph.h"
 #include "iwl-fh.h"
 
 #define IWL_POLL_INTERVAL 10	/* microseconds */
@@ -183,6 +184,23 @@ void iwl_clear_bits_prph(struct iwl_trans *trans, u32 ofs, u32 mask)
 }
 IWL_EXPORT_SYMBOL(iwl_clear_bits_prph);
 
+void iwl_force_nmi(struct iwl_trans *trans)
+{
+	/*
+	 * In HW previous to the 8000 HW family, and in the 8000 HW family
+	 * itself when the revision step==0, the DEVICE_SET_NMI_REG is used
+	 * to force an NMI. Otherwise, a different register -
+	 * DEVICE_SET_NMI_8000B_REG - is used.
+	 */
+	if ((trans->cfg->device_family != IWL_DEVICE_FAMILY_8000) ||
+	    ((trans->hw_rev & 0xc) == 0x0))
+		iwl_write_prph(trans, DEVICE_SET_NMI_REG, DEVICE_SET_NMI_VAL);
+	else
+		iwl_write_prph(trans, DEVICE_SET_NMI_8000B_REG,
+			       DEVICE_SET_NMI_8000B_VAL);
+}
+IWL_EXPORT_SYMBOL(iwl_force_nmi);
+
 static const char *get_fh_string(int cmd)
 {
 #define IWL_CMD(x) case x: return #x
diff --git a/drivers/net/wireless/iwlwifi/iwl-io.h b/drivers/net/wireless/iwlwifi/iwl-io.h
index 665ddd9dbbc4..705d12c079e8 100644
--- a/drivers/net/wireless/iwlwifi/iwl-io.h
+++ b/drivers/net/wireless/iwlwifi/iwl-io.h
@@ -80,6 +80,7 @@ void iwl_set_bits_prph(struct iwl_trans *trans, u32 ofs, u32 mask);
 void iwl_set_bits_mask_prph(struct iwl_trans *trans, u32 ofs,
 			    u32 bits, u32 mask);
 void iwl_clear_bits_prph(struct iwl_trans *trans, u32 ofs, u32 mask);
+void iwl_force_nmi(struct iwl_trans *trans);
 
 /* Error handling */
 int iwl_dump_fh(struct iwl_trans *trans, char **buf);
diff --git a/drivers/net/wireless/iwlwifi/iwl-modparams.h b/drivers/net/wireless/iwlwifi/iwl-modparams.h
index d994317db85b..d051857729ab 100644
--- a/drivers/net/wireless/iwlwifi/iwl-modparams.h
+++ b/drivers/net/wireless/iwlwifi/iwl-modparams.h
@@ -119,6 +119,7 @@ struct iwl_mod_params {
 #endif
 	int ant_coupling;
 	char *nvm_file;
+	bool uapsd_disable;
 };
 
 #endif /* #__iwl_modparams_h__ */
diff --git a/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c
index 6be30c698506..85eee79c495c 100644
--- a/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c
@@ -62,6 +62,7 @@
 #include <linux/types.h>
 #include <linux/slab.h>
 #include <linux/export.h>
+#include <linux/etherdevice.h>
 #include "iwl-drv.h"
 #include "iwl-modparams.h"
 #include "iwl-nvm-parse.h"
@@ -127,19 +128,20 @@ static const u8 iwl_nvm_channels[] = {
 
 static const u8 iwl_nvm_channels_family_8000[] = {
 	/* 2.4 GHz */
-	1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,
+	1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
 	/* 5 GHz */
 	36, 40, 44, 48, 52, 56, 60, 64, 68, 72, 76, 80, 84, 88, 92,
 	96, 100, 104, 108, 112, 116, 120, 124, 128, 132, 136, 140, 144,
 	149, 153, 157, 161, 165, 169, 173, 177, 181
 };
 
-#define IWL_NUM_CHANNELS	ARRAY_SIZE(iwl_nvm_channels)
+#define IWL_NUM_CHANNELS		ARRAY_SIZE(iwl_nvm_channels)
 #define IWL_NUM_CHANNELS_FAMILY_8000	ARRAY_SIZE(iwl_nvm_channels_family_8000)
-#define NUM_2GHZ_CHANNELS	14
-#define FIRST_2GHZ_HT_MINUS	5
-#define LAST_2GHZ_HT_PLUS	9
-#define LAST_5GHZ_HT		161
+#define NUM_2GHZ_CHANNELS		14
+#define NUM_2GHZ_CHANNELS_FAMILY_8000	14
+#define FIRST_2GHZ_HT_MINUS		5
+#define LAST_2GHZ_HT_PLUS		9
+#define LAST_5GHZ_HT			161
 
 #define DEFAULT_MAX_TX_POWER 16
 
@@ -202,21 +204,23 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
 	struct ieee80211_channel *channel;
 	u16 ch_flags;
 	bool is_5ghz;
-	int num_of_ch;
+	int num_of_ch, num_2ghz_channels;
 	const u8 *nvm_chan;
 
 	if (cfg->device_family != IWL_DEVICE_FAMILY_8000) {
 		num_of_ch = IWL_NUM_CHANNELS;
 		nvm_chan = &iwl_nvm_channels[0];
+		num_2ghz_channels = NUM_2GHZ_CHANNELS;
 	} else {
 		num_of_ch = IWL_NUM_CHANNELS_FAMILY_8000;
 		nvm_chan = &iwl_nvm_channels_family_8000[0];
+		num_2ghz_channels = NUM_2GHZ_CHANNELS_FAMILY_8000;
 	}
 
 	for (ch_idx = 0; ch_idx < num_of_ch; ch_idx++) {
 		ch_flags = __le16_to_cpup(nvm_ch_flags + ch_idx);
 
-		if (ch_idx >= NUM_2GHZ_CHANNELS &&
+		if (ch_idx >= num_2ghz_channels &&
 		    !data->sku_cap_band_52GHz_enable)
 			ch_flags &= ~NVM_CHANNEL_VALID;
 
@@ -225,7 +229,7 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
 					 "Ch. %d Flags %x [%sGHz] - No traffic\n",
 					 nvm_chan[ch_idx],
 					 ch_flags,
-					 (ch_idx >= NUM_2GHZ_CHANNELS) ?
+					 (ch_idx >= num_2ghz_channels) ?
 					 "5.2" : "2.4");
 			continue;
 		}
@@ -234,7 +238,7 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
 		n_channels++;
 
 		channel->hw_value = nvm_chan[ch_idx];
-		channel->band = (ch_idx < NUM_2GHZ_CHANNELS) ?
+		channel->band = (ch_idx < num_2ghz_channels) ?
 				IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ;
 		channel->center_freq =
 			ieee80211_channel_to_frequency(
@@ -242,7 +246,7 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
 
 		/* TODO: Need to be dependent to the NVM */
 		channel->flags = IEEE80211_CHAN_NO_HT40;
-		if (ch_idx < NUM_2GHZ_CHANNELS &&
+		if (ch_idx < num_2ghz_channels &&
 		    (ch_flags & NVM_CHANNEL_40MHZ)) {
 			if (nvm_chan[ch_idx] <= LAST_2GHZ_HT_PLUS)
 				channel->flags &= ~IEEE80211_CHAN_NO_HT40PLUS;
@@ -250,7 +254,7 @@ static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
 				channel->flags &= ~IEEE80211_CHAN_NO_HT40MINUS;
 		} else if (nvm_chan[ch_idx] <= LAST_5GHZ_HT &&
 			   (ch_flags & NVM_CHANNEL_40MHZ)) {
-			if ((ch_idx - NUM_2GHZ_CHANNELS) % 2 == 0)
+			if ((ch_idx - num_2ghz_channels) % 2 == 0)
 				channel->flags &= ~IEEE80211_CHAN_NO_HT40PLUS;
 			else
 				channel->flags &= ~IEEE80211_CHAN_NO_HT40MINUS;
@@ -447,13 +451,7 @@ static void iwl_set_hw_address(const struct iwl_cfg *cfg,
 			       struct iwl_nvm_data *data,
 			       const __le16 *nvm_sec)
 {
-	u8 hw_addr[ETH_ALEN];
-
-	if (cfg->device_family != IWL_DEVICE_FAMILY_8000)
-		memcpy(hw_addr, nvm_sec + HW_ADDR, ETH_ALEN);
-	else
-		memcpy(hw_addr, nvm_sec + MAC_ADDRESS_OVERRIDE_FAMILY_8000,
-		       ETH_ALEN);
+	const u8 *hw_addr = (const u8 *)(nvm_sec + HW_ADDR);
 
 	/* The byte order is little endian 16 bit, meaning 214365 */
 	data->hw_addr[0] = hw_addr[1];
@@ -464,6 +462,41 @@ static void iwl_set_hw_address(const struct iwl_cfg *cfg,
 	data->hw_addr[5] = hw_addr[4];
 }
 
+static void iwl_set_hw_address_family_8000(const struct iwl_cfg *cfg,
+					   struct iwl_nvm_data *data,
+					   const __le16 *mac_override,
+					   const __le16 *nvm_hw)
+{
+	const u8 *hw_addr;
+
+	if (mac_override) {
+		hw_addr = (const u8 *)(mac_override +
+				 MAC_ADDRESS_OVERRIDE_FAMILY_8000);
+
+		/* The byte order is little endian 16 bit, meaning 214365 */
+		data->hw_addr[0] = hw_addr[1];
+		data->hw_addr[1] = hw_addr[0];
+		data->hw_addr[2] = hw_addr[3];
+		data->hw_addr[3] = hw_addr[2];
+		data->hw_addr[4] = hw_addr[5];
+		data->hw_addr[5] = hw_addr[4];
+
+		if (is_valid_ether_addr(hw_addr))
+			return;
+	}
+
+	/* take the MAC address from the OTP */
+	hw_addr = (const u8 *)(nvm_hw + HW_ADDR0_FAMILY_8000);
+	data->hw_addr[0] = hw_addr[3];
+	data->hw_addr[1] = hw_addr[2];
+	data->hw_addr[2] = hw_addr[1];
+	data->hw_addr[3] = hw_addr[0];
+
+	hw_addr = (const u8 *)(nvm_hw + HW_ADDR1_FAMILY_8000);
+	data->hw_addr[4] = hw_addr[1];
+	data->hw_addr[5] = hw_addr[0];
+}
+
 struct iwl_nvm_data *
 iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
 		   const __le16 *nvm_hw, const __le16 *nvm_sw,
@@ -523,7 +556,7 @@ iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
 				rx_chains);
 	} else {
 		/* MAC address in family 8000 */
-		iwl_set_hw_address(cfg, data, mac_override);
+		iwl_set_hw_address_family_8000(cfg, data, mac_override, nvm_hw);
 
 		iwl_init_sbands(dev, cfg, data, regulatory,
 				sku & NVM_SKU_CAP_11AC_ENABLE, tx_chains,
diff --git a/drivers/net/wireless/iwlwifi/iwl-op-mode.h b/drivers/net/wireless/iwlwifi/iwl-op-mode.h
index ea29504ac617..99785c892f96 100644
--- a/drivers/net/wireless/iwlwifi/iwl-op-mode.h
+++ b/drivers/net/wireless/iwlwifi/iwl-op-mode.h
@@ -63,6 +63,7 @@
 #ifndef __iwl_op_mode_h__
 #define __iwl_op_mode_h__
 
+#include <linux/netdevice.h>
 #include <linux/debugfs.h>
 
 struct iwl_op_mode;
@@ -112,8 +113,11 @@ struct iwl_cfg;
  * @stop: stop the op_mode. Must free all the memory allocated.
  *	May sleep
  * @rx: Rx notification to the op_mode. rxb is the Rx buffer itself. Cmd is the
- *	HCMD this Rx responds to.
- *	This callback may sleep, it is called from a threaded IRQ handler.
+ *	HCMD this Rx responds to. Can't sleep.
+ * @napi_add: NAPI initialisation. The transport is fully responsible for NAPI,
+ *	but the higher layers need to know about it (in particular mac80211 to
+ *	to able to call the right NAPI RX functions); this function is needed
+ *	to eventually call netif_napi_add() with higher layer involvement.
  * @queue_full: notifies that a HW queue is full.
  *	Must be atomic and called with BH disabled.
  * @queue_not_full: notifies that a HW queue is not full any more.
@@ -143,6 +147,11 @@ struct iwl_op_mode_ops {
 	void (*stop)(struct iwl_op_mode *op_mode);
 	int (*rx)(struct iwl_op_mode *op_mode, struct iwl_rx_cmd_buffer *rxb,
 		  struct iwl_device_cmd *cmd);
+	void (*napi_add)(struct iwl_op_mode *op_mode,
+			 struct napi_struct *napi,
+			 struct net_device *napi_dev,
+			 int (*poll)(struct napi_struct *, int),
+			 int weight);
 	void (*queue_full)(struct iwl_op_mode *op_mode, int queue);
 	void (*queue_not_full)(struct iwl_op_mode *op_mode, int queue);
 	bool (*hw_rf_kill)(struct iwl_op_mode *op_mode, bool state);
@@ -180,7 +189,6 @@ static inline int iwl_op_mode_rx(struct iwl_op_mode *op_mode,
 				  struct iwl_rx_cmd_buffer *rxb,
 				  struct iwl_device_cmd *cmd)
 {
-	might_sleep();
 	return op_mode->ops->rx(op_mode, rxb, cmd);
 }
 
@@ -249,4 +257,15 @@ static inline int iwl_op_mode_exit_d0i3(struct iwl_op_mode *op_mode)
 	return op_mode->ops->exit_d0i3(op_mode);
 }
 
+static inline void iwl_op_mode_napi_add(struct iwl_op_mode *op_mode,
+					struct napi_struct *napi,
+					struct net_device *napi_dev,
+					int (*poll)(struct napi_struct *, int),
+					int weight)
+{
+	if (!op_mode->ops->napi_add)
+		return;
+	op_mode->ops->napi_add(op_mode, napi, napi_dev, poll, weight);
+}
+
 #endif /* __iwl_op_mode_h__ */
diff --git a/drivers/net/wireless/iwlwifi/iwl-phy-db.c b/drivers/net/wireless/iwlwifi/iwl-phy-db.c
index b761ac4822a3..d4fb5cad07ea 100644
--- a/drivers/net/wireless/iwlwifi/iwl-phy-db.c
+++ b/drivers/net/wireless/iwlwifi/iwl-phy-db.c
@@ -345,7 +345,6 @@ static int iwl_send_phy_db_cmd(struct iwl_phy_db *phy_db, u16 type,
 	struct iwl_phy_db_cmd phy_db_cmd;
 	struct iwl_host_cmd cmd = {
 		.id = PHY_DB_CMD,
-		.flags = CMD_SYNC,
 	};
 
 	IWL_DEBUG_INFO(phy_db->trans,
@@ -393,13 +392,13 @@ static int iwl_phy_db_send_all_channel_groups(
 					  entry->data);
 		if (err) {
 			IWL_ERR(phy_db->trans,
-				"Can't SEND phy_db section %d (%d), err %d",
+				"Can't SEND phy_db section %d (%d), err %d\n",
 				type, i, err);
 			return err;
 		}
 
 		IWL_DEBUG_INFO(phy_db->trans,
-			       "Sent PHY_DB HCMD, type = %d num = %d",
+			       "Sent PHY_DB HCMD, type = %d num = %d\n",
 			       type, i);
 	}
 
@@ -451,7 +450,7 @@ int iwl_send_phy_db_data(struct iwl_phy_db *phy_db)
 						 IWL_NUM_PAPD_CH_GROUPS);
 	if (err) {
 		IWL_ERR(phy_db->trans,
-			"Cannot send channel specific PAPD groups");
+			"Cannot send channel specific PAPD groups\n");
 		return err;
 	}
 
@@ -461,7 +460,7 @@ int iwl_send_phy_db_data(struct iwl_phy_db *phy_db)
 						 IWL_NUM_TXP_CH_GROUPS);
 	if (err) {
 		IWL_ERR(phy_db->trans,
-			"Cannot send channel specific TX power groups");
+			"Cannot send channel specific TX power groups\n");
 		return err;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-prph.h b/drivers/net/wireless/iwlwifi/iwl-prph.h
index 5f657c501406..4997e27672b3 100644
--- a/drivers/net/wireless/iwlwifi/iwl-prph.h
+++ b/drivers/net/wireless/iwlwifi/iwl-prph.h
@@ -105,6 +105,9 @@
 
 /* Device NMI register */
 #define DEVICE_SET_NMI_REG 0x00a01c30
+#define DEVICE_SET_NMI_VAL 0x1
+#define DEVICE_SET_NMI_8000B_REG 0x00a01c24
+#define DEVICE_SET_NMI_8000B_VAL 0x1000000
 
 /* Shared registers (0x0..0x3ff, via target indirect or periphery */
 #define SHR_BASE	0x00a10000
@@ -348,4 +351,12 @@ enum secure_load_status_reg {
 
 #define LMPM_SECURE_TIME_OUT	(100)
 
+/* Rx FIFO */
+#define RXF_SIZE_ADDR			(0xa00c88)
+#define RXF_SIZE_BYTE_CND_POS		(7)
+#define RXF_SIZE_BYTE_CNT_MSK		(0x3ff << RXF_SIZE_BYTE_CND_POS)
+
+#define RXF_LD_FENCE_OFFSET_ADDR	(0xa00c10)
+#define RXF_FIFO_RD_FENCE_ADDR		(0xa00c0c)
+
 #endif				/* __iwl_prph_h__ */
diff --git a/drivers/net/wireless/iwlwifi/iwl-trans.h b/drivers/net/wireless/iwlwifi/iwl-trans.h
index 8cdb0dd618a6..34d49e171fb4 100644
--- a/drivers/net/wireless/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/iwlwifi/iwl-trans.h
@@ -189,10 +189,9 @@ static inline u32 iwl_rx_packet_payload_len(const struct iwl_rx_packet *pkt)
 /**
  * enum CMD_MODE - how to send the host commands ?
  *
- * @CMD_SYNC: The caller will be stalled until the fw responds to the command
  * @CMD_ASYNC: Return right away and don't wait for the response
- * @CMD_WANT_SKB: valid only with CMD_SYNC. The caller needs the buffer of the
- *	response. The caller needs to call iwl_free_resp when done.
+ * @CMD_WANT_SKB: Not valid with CMD_ASYNC. The caller needs the buffer of
+ *	the response. The caller needs to call iwl_free_resp when done.
  * @CMD_HIGH_PRIO: The command is high priority - it goes to the front of the
  *	command queue, but after other high priority commands. valid only
  *	with CMD_ASYNC.
@@ -202,7 +201,6 @@ static inline u32 iwl_rx_packet_payload_len(const struct iwl_rx_packet *pkt)
  *	(i.e. mark it as non-idle).
  */
 enum CMD_MODE {
-	CMD_SYNC		= 0,
 	CMD_ASYNC		= BIT(0),
 	CMD_WANT_SKB		= BIT(1),
 	CMD_SEND_IN_RFKILL	= BIT(2),
@@ -427,7 +425,7 @@ struct iwl_trans;
  * @send_cmd:send a host command. Must return -ERFKILL if RFkill is asserted.
  *	If RFkill is asserted in the middle of a SYNC host command, it must
  *	return -ERFKILL straight away.
- *	May sleep only if CMD_SYNC is set
+ *	May sleep only if CMD_ASYNC is not set
  * @tx: send an skb
  *	Must be atomic
  * @reclaim: free packet until ssn. Returns a list of freed packets.
@@ -437,8 +435,7 @@ struct iwl_trans;
  *	this one. The op_mode must not configure the HCMD queue. May sleep.
  * @txq_disable: de-configure a Tx queue to send AMPDUs
  *	Must be atomic
- * @wait_tx_queue_empty: wait until all tx queues are empty
- *	May sleep
+ * @wait_tx_queue_empty: wait until tx queues are empty. May sleep.
  * @dbgfs_register: add the dbgfs files under this directory. Files will be
  *	automatically deleted.
  * @write8: write a u8 to a register at offset ofs from the BAR
@@ -464,6 +461,11 @@ struct iwl_trans;
  * @unref: release a reference previously taken with @ref. Note that
  *	initially the reference count is 1, making an initial @unref
  *	necessary to allow low power states.
+ * @dump_data: fill a data dump with debug data, maybe containing last
+ *	TX'ed commands and similar. When called with a NULL buffer and
+ *	zero buffer length, provide only the (estimated) required buffer
+ *	length. Return the used buffer length.
+ *	Note that the transport must fill in the proper file headers.
  */
 struct iwl_trans_ops {
 
@@ -471,6 +473,8 @@ struct iwl_trans_ops {
 	void (*op_mode_leave)(struct iwl_trans *iwl_trans);
 	int (*start_fw)(struct iwl_trans *trans, const struct fw_img *fw,
 			bool run_in_rfkill);
+	int (*update_sf)(struct iwl_trans *trans,
+			 struct iwl_sf_region *st_fwrd_space);
 	void (*fw_alive)(struct iwl_trans *trans, u32 scd_addr);
 	void (*stop_device)(struct iwl_trans *trans);
 
@@ -490,7 +494,7 @@ struct iwl_trans_ops {
 	void (*txq_disable)(struct iwl_trans *trans, int queue);
 
 	int (*dbgfs_register)(struct iwl_trans *trans, struct dentry* dir);
-	int (*wait_tx_queue_empty)(struct iwl_trans *trans);
+	int (*wait_tx_queue_empty)(struct iwl_trans *trans, u32 txq_bm);
 
 	void (*write8)(struct iwl_trans *trans, u32 ofs, u8 val);
 	void (*write32)(struct iwl_trans *trans, u32 ofs, u32 val);
@@ -512,6 +516,10 @@ struct iwl_trans_ops {
 			      u32 value);
 	void (*ref)(struct iwl_trans *trans);
 	void (*unref)(struct iwl_trans *trans);
+
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+	u32 (*dump_data)(struct iwl_trans *trans, void *buf, u32 buflen);
+#endif
 };
 
 /**
@@ -630,6 +638,17 @@ static inline int iwl_trans_start_fw(struct iwl_trans *trans,
 	return trans->ops->start_fw(trans, fw, run_in_rfkill);
 }
 
+static inline int iwl_trans_update_sf(struct iwl_trans *trans,
+				      struct iwl_sf_region *st_fwrd_space)
+{
+	might_sleep();
+
+	if (trans->ops->update_sf)
+		return trans->ops->update_sf(trans, st_fwrd_space);
+
+	return 0;
+}
+
 static inline void iwl_trans_stop_device(struct iwl_trans *trans)
 {
 	might_sleep();
@@ -665,6 +684,16 @@ static inline void iwl_trans_unref(struct iwl_trans *trans)
 		trans->ops->unref(trans);
 }
 
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+static inline u32 iwl_trans_dump_data(struct iwl_trans *trans,
+				      void *buf, u32 buflen)
+{
+	if (!trans->ops->dump_data)
+		return 0;
+	return trans->ops->dump_data(trans, buf, buflen);
+}
+#endif
+
 static inline int iwl_trans_send_cmd(struct iwl_trans *trans,
 				     struct iwl_host_cmd *cmd)
 {
@@ -678,7 +707,7 @@ static inline int iwl_trans_send_cmd(struct iwl_trans *trans,
 		return -EIO;
 
 	if (unlikely(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d", __func__, trans->state);
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
 		return -EIO;
 	}
 
@@ -720,7 +749,7 @@ static inline int iwl_trans_tx(struct iwl_trans *trans, struct sk_buff *skb,
 		return -EIO;
 
 	if (unlikely(trans->state != IWL_TRANS_FW_ALIVE))
-		IWL_ERR(trans, "%s bad state = %d", __func__, trans->state);
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
 
 	return trans->ops->tx(trans, skb, dev_cmd, queue);
 }
@@ -729,7 +758,7 @@ static inline void iwl_trans_reclaim(struct iwl_trans *trans, int queue,
 				     int ssn, struct sk_buff_head *skbs)
 {
 	if (unlikely(trans->state != IWL_TRANS_FW_ALIVE))
-		IWL_ERR(trans, "%s bad state = %d", __func__, trans->state);
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
 
 	trans->ops->reclaim(trans, queue, ssn, skbs);
 }
@@ -746,7 +775,7 @@ static inline void iwl_trans_txq_enable(struct iwl_trans *trans, int queue,
 	might_sleep();
 
 	if (unlikely((trans->state != IWL_TRANS_FW_ALIVE)))
-		IWL_ERR(trans, "%s bad state = %d", __func__, trans->state);
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
 
 	trans->ops->txq_enable(trans, queue, fifo, sta_id, tid,
 				 frame_limit, ssn);
@@ -759,12 +788,13 @@ static inline void iwl_trans_ac_txq_enable(struct iwl_trans *trans, int queue,
 			     IWL_MAX_TID_COUNT, IWL_FRAME_LIMIT, 0);
 }
 
-static inline int iwl_trans_wait_tx_queue_empty(struct iwl_trans *trans)
+static inline int iwl_trans_wait_tx_queue_empty(struct iwl_trans *trans,
+						u32 txq_bm)
 {
 	if (unlikely(trans->state != IWL_TRANS_FW_ALIVE))
-		IWL_ERR(trans, "%s bad state = %d", __func__, trans->state);
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
 
-	return trans->ops->wait_tx_queue_empty(trans);
+	return trans->ops->wait_tx_queue_empty(trans, txq_bm);
 }
 
 static inline int iwl_trans_dbgfs_register(struct iwl_trans *trans,
diff --git a/drivers/net/wireless/iwlwifi/mvm/Makefile b/drivers/net/wireless/iwlwifi/mvm/Makefile
index ccdd3b7c4cce..c30d7f64ec1e 100644
--- a/drivers/net/wireless/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/iwlwifi/mvm/Makefile
@@ -3,8 +3,9 @@ iwlmvm-y += fw.o mac80211.o nvm.o ops.o phy-ctxt.o mac-ctxt.o
 iwlmvm-y += utils.o rx.o tx.o binding.o quota.o sta.o sf.o
 iwlmvm-y += scan.o time-event.o rs.o
 iwlmvm-y += power.o coex.o
-iwlmvm-y += led.o tt.o offloading.o
+iwlmvm-y += tt.o offloading.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
+iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM_SLEEP) += d3.o
 
 ccflags-y += -D__CHECK_ENDIAN__ -I$(src)/../
diff --git a/drivers/net/wireless/iwlwifi/mvm/coex.c b/drivers/net/wireless/iwlwifi/mvm/coex.c
index 0489314425cb..c8c3b38228f0 100644
--- a/drivers/net/wireless/iwlwifi/mvm/coex.c
+++ b/drivers/net/wireless/iwlwifi/mvm/coex.c
@@ -104,12 +104,9 @@ static const u8 iwl_bt_prio_tbl[BT_COEX_PRIO_TBL_EVT_MAX] = {
 #define BT_DISABLE_REDUCED_TXPOWER_THRESHOLD	(-65)
 #define BT_ANTENNA_COUPLING_THRESHOLD		(30)
 
-int iwl_send_bt_prio_tbl(struct iwl_mvm *mvm)
+static int iwl_send_bt_prio_tbl(struct iwl_mvm *mvm)
 {
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_NEWBT_COEX))
-		return 0;
-
-	return iwl_mvm_send_cmd_pdu(mvm, BT_COEX_PRIO_TABLE, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, BT_COEX_PRIO_TABLE, 0,
 				    sizeof(struct iwl_bt_coex_prio_tbl_cmd),
 				    &iwl_bt_prio_tbl);
 }
@@ -127,10 +124,10 @@ const u32 iwl_bt_cts_kill_msk[BT_KILL_MSK_MAX] = {
 };
 
 static const __le32 iwl_bt_prio_boost[BT_COEX_BOOST_SIZE] = {
-	cpu_to_le32(0xf0f0f0f0),
-	cpu_to_le32(0xc0c0c0c0),
-	cpu_to_le32(0xfcfcfcfc),
-	cpu_to_le32(0xff00ff00),
+	cpu_to_le32(0xf0f0f0f0), /* 50% */
+	cpu_to_le32(0xc0c0c0c0), /* 25% */
+	cpu_to_le32(0xfcfcfcfc), /* 75% */
+	cpu_to_le32(0xfefefefe), /* 87.5% */
 };
 
 static const __le32 iwl_single_shared_ant[BT_COEX_MAX_LUT][BT_COEX_LUT_SIZE] = {
@@ -303,8 +300,8 @@ static const __le64 iwl_ci_mask[][3] = {
 };
 
 static const __le32 iwl_bt_mprio_lut[BT_COEX_MULTI_PRIO_LUT_SIZE] = {
-	cpu_to_le32(0x22002200),
-	cpu_to_le32(0x33113311),
+	cpu_to_le32(0x28412201),
+	cpu_to_le32(0x11118451),
 };
 
 struct corunning_block_luts {
@@ -568,13 +565,13 @@ int iwl_send_bt_init_conf(struct iwl_mvm *mvm)
 		.id = BT_CONFIG,
 		.len = { sizeof(*bt_cmd), },
 		.dataflags = { IWL_HCMD_DFL_NOCOPY, },
-		.flags = CMD_SYNC,
 	};
 	int ret;
 	u32 flags;
 
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_NEWBT_COEX))
-		return 0;
+	ret = iwl_send_bt_prio_tbl(mvm);
+	if (ret)
+		return ret;
 
 	bt_cmd = kzalloc(sizeof(*bt_cmd), GFP_KERNEL);
 	if (!bt_cmd)
@@ -582,10 +579,12 @@ int iwl_send_bt_init_conf(struct iwl_mvm *mvm)
 	cmd.data[0] = bt_cmd;
 
 	bt_cmd->max_kill = 5;
-	bt_cmd->bt4_antenna_isolation_thr = BT_ANTENNA_COUPLING_THRESHOLD,
-	bt_cmd->bt4_antenna_isolation = iwlwifi_mod_params.ant_coupling,
-	bt_cmd->bt4_tx_tx_delta_freq_thr = 15,
-	bt_cmd->bt4_tx_rx_max_freq0 = 15,
+	bt_cmd->bt4_antenna_isolation_thr = BT_ANTENNA_COUPLING_THRESHOLD;
+	bt_cmd->bt4_antenna_isolation = iwlwifi_mod_params.ant_coupling;
+	bt_cmd->bt4_tx_tx_delta_freq_thr = 15;
+	bt_cmd->bt4_tx_rx_max_freq0 = 15;
+	bt_cmd->override_primary_lut = BT_COEX_INVALID_LUT;
+	bt_cmd->override_secondary_lut = BT_COEX_INVALID_LUT;
 
 	flags = iwlwifi_mod_params.bt_coex_active ?
 			BT_COEX_NW : BT_COEX_DISABLE;
@@ -663,7 +662,6 @@ static int iwl_mvm_bt_udpate_ctrl_kill_msk(struct iwl_mvm *mvm,
 		.data[0] = &bt_cmd,
 		.len = { sizeof(*bt_cmd), },
 		.dataflags = { IWL_HCMD_DFL_NOCOPY, },
-		.flags = CMD_SYNC,
 	};
 	int ret = 0;
 
@@ -717,7 +715,8 @@ static int iwl_mvm_bt_udpate_ctrl_kill_msk(struct iwl_mvm *mvm,
 	return ret;
 }
 
-int iwl_mvm_bt_coex_reduced_txp(struct iwl_mvm *mvm, u8 sta_id, bool enable)
+static int iwl_mvm_bt_coex_reduced_txp(struct iwl_mvm *mvm, u8 sta_id,
+				       bool enable)
 {
 	struct iwl_bt_coex_cmd *bt_cmd;
 	/* Send ASYNC since this can be sent from an atomic context */
@@ -735,8 +734,7 @@ int iwl_mvm_bt_coex_reduced_txp(struct iwl_mvm *mvm, u8 sta_id, bool enable)
 		return 0;
 
 	/* nothing to do */
-	if (mvmsta->bt_reduced_txpower_dbg ||
-	    mvmsta->bt_reduced_txpower == enable)
+	if (mvmsta->bt_reduced_txpower == enable)
 		return 0;
 
 	bt_cmd = kzalloc(sizeof(*bt_cmd), GFP_ATOMIC);
@@ -803,23 +801,10 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
+		/* Count BSSes vifs */
+		data->num_bss_ifaces++;
 		/* default smps_mode for BSS / P2P client is AUTOMATIC */
 		smps_mode = IEEE80211_SMPS_AUTOMATIC;
-		data->num_bss_ifaces++;
-
-		/*
-		 * Count unassoc BSSes, relax SMSP constraints
-		 * and disable reduced Tx Power
-		 */
-		if (!vif->bss_conf.assoc) {
-			iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
-					    smps_mode);
-			if (iwl_mvm_bt_coex_reduced_txp(mvm,
-							mvmvif->ap_sta_id,
-							false))
-				IWL_ERR(mvm, "Couldn't send BT_CONFIG cmd\n");
-			return;
-		}
 		break;
 	case NL80211_IFTYPE_AP:
 		/* default smps_mode for AP / GO is OFF */
@@ -845,8 +830,12 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 		/* ... relax constraints and disable rssi events */
 		iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
 				    smps_mode);
-		if (vif->type == NL80211_IFTYPE_STATION)
+		data->reduced_tx_power = false;
+		if (vif->type == NL80211_IFTYPE_STATION) {
+			iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id,
+						    false);
 			iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
+		}
 		return;
 	}
 
@@ -857,6 +846,11 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 		smps_mode = vif->type == NL80211_IFTYPE_AP ?
 				IEEE80211_SMPS_OFF :
 				IEEE80211_SMPS_DYNAMIC;
+
+	/* relax SMPS contraints for next association */
+	if (!vif->bss_conf.assoc)
+		smps_mode = IEEE80211_SMPS_AUTOMATIC;
+
 	IWL_DEBUG_COEX(data->mvm,
 		       "mac %d: bt_status %d bt_activity_grading %d smps_req %d\n",
 		       mvmvif->id, data->notif->bt_status, bt_activity_grading,
@@ -903,22 +897,18 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 		/* if secondary is not NULL, it might be a GO */
 		data->secondary = chanctx_conf;
 
-	/* don't reduce the Tx power if in loose scheme */
+	/*
+	 * don't reduce the Tx power if one of these is true:
+	 *  we are in LOOSE
+	 *  single share antenna product
+	 *  BT is active
+	 *  we are associated
+	 */
 	if (iwl_get_coex_type(mvm, vif) == BT_COEX_LOOSE_LUT ||
-	    mvm->cfg->bt_shared_single_ant) {
-		data->reduced_tx_power = false;
-		iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
-		return;
-	}
-
-	/* reduced Txpower only if BT is on, so ...*/
-	if (!data->notif->bt_status) {
-		/* ... cancel reduced Tx power ... */
-		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id, false))
-			IWL_ERR(mvm, "Couldn't send BT_CONFIG cmd\n");
+	    mvm->cfg->bt_shared_single_ant || !vif->bss_conf.assoc ||
+	    !data->notif->bt_status) {
 		data->reduced_tx_power = false;
-
-		/* ... and there is no need to get reports on RSSI any more. */
+		iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id, false);
 		iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
 		return;
 	}
@@ -1022,9 +1012,9 @@ static void iwl_mvm_bt_coex_notif_handle(struct iwl_mvm *mvm)
 
 	/* Don't spam the fw with the same command over and over */
 	if (memcmp(&cmd, &mvm->last_bt_ci_cmd, sizeof(cmd))) {
-		if (iwl_mvm_send_cmd_pdu(mvm, BT_COEX_CI, CMD_SYNC,
+		if (iwl_mvm_send_cmd_pdu(mvm, BT_COEX_CI, 0,
 					 sizeof(cmd), &cmd))
-			IWL_ERR(mvm, "Failed to send BT_CI cmd");
+			IWL_ERR(mvm, "Failed to send BT_CI cmd\n");
 		memcpy(&mvm->last_bt_ci_cmd, &cmd, sizeof(cmd));
 	}
 
@@ -1039,7 +1029,6 @@ static void iwl_mvm_bt_coex_notif_handle(struct iwl_mvm *mvm)
 		IWL_ERR(mvm, "Failed to update the ctrl_kill_msk\n");
 }
 
-/* upon association, the fw will send in BT Coex notification */
 int iwl_mvm_rx_bt_coex_notif(struct iwl_mvm *mvm,
 			     struct iwl_rx_cmd_buffer *rxb,
 			     struct iwl_device_cmd *dev_cmd)
@@ -1215,6 +1204,17 @@ bool iwl_mvm_bt_coex_is_mimo_allowed(struct iwl_mvm *mvm,
 	return iwl_get_coex_type(mvm, mvmsta->vif) == BT_COEX_TIGHT_LUT;
 }
 
+bool iwl_mvm_bt_coex_is_tpc_allowed(struct iwl_mvm *mvm,
+				    enum ieee80211_band band)
+{
+	u32 bt_activity = le32_to_cpu(mvm->last_bt_notif.bt_activity_grading);
+
+	if (band != IEEE80211_BAND_2GHZ)
+		return false;
+
+	return bt_activity >= BT_LOW_TRAFFIC;
+}
+
 u8 iwl_mvm_bt_coex_tx_prio(struct iwl_mvm *mvm, struct ieee80211_hdr *hdr,
 			   struct ieee80211_tx_info *info, u8 ac)
 {
@@ -1249,9 +1249,6 @@ u8 iwl_mvm_bt_coex_tx_prio(struct iwl_mvm *mvm, struct ieee80211_hdr *hdr,
 
 void iwl_mvm_bt_coex_vif_change(struct iwl_mvm *mvm)
 {
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_NEWBT_COEX))
-		return;
-
 	iwl_mvm_bt_coex_notif_handle(mvm);
 }
 
@@ -1270,7 +1267,6 @@ int iwl_mvm_rx_ant_coupling_notif(struct iwl_mvm *mvm,
 		.id = BT_CONFIG,
 		.len = { sizeof(*bt_cmd), },
 		.dataflags = { IWL_HCMD_DFL_NOCOPY, },
-		.flags = CMD_SYNC,
 	};
 
 	if (!IWL_MVM_BT_COEX_CORUNNING)
diff --git a/drivers/net/wireless/iwlwifi/mvm/d3.c b/drivers/net/wireless/iwlwifi/mvm/d3.c
index e56f5a0edf85..645b3cfc29a5 100644
--- a/drivers/net/wireless/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/iwlwifi/mvm/d3.c
@@ -193,8 +193,7 @@ static void iwl_mvm_wowlan_program_keys(struct ieee80211_hw *hw,
 			wkc.wep_key.key_offset = data->wep_key_idx;
 		}
 
-		ret = iwl_mvm_send_cmd_pdu(mvm, WEP_KEY, CMD_SYNC,
-					   sizeof(wkc), &wkc);
+		ret = iwl_mvm_send_cmd_pdu(mvm, WEP_KEY, 0, sizeof(wkc), &wkc);
 		data->error = ret != 0;
 
 		mvm->ptk_ivlen = key->iv_len;
@@ -341,7 +340,6 @@ static int iwl_mvm_send_patterns(struct iwl_mvm *mvm,
 	struct iwl_host_cmd cmd = {
 		.id = WOWLAN_PATTERNS,
 		.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
-		.flags = CMD_SYNC,
 	};
 	int i, err;
 
@@ -518,7 +516,6 @@ static int iwl_mvm_send_remote_wake_cfg(struct iwl_mvm *mvm,
 		.id = REMOTE_WAKE_CONFIG_CMD,
 		.len = { sizeof(*cfg), },
 		.dataflags = { IWL_HCMD_DFL_NOCOPY, },
-		.flags = CMD_SYNC,
 	};
 	int ret;
 
@@ -666,10 +663,8 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	if (WARN_ON(!vif->bss_conf.assoc))
 		return -EINVAL;
-	/* hack */
-	vif->bss_conf.assoc = false;
+
 	ret = iwl_mvm_mac_ctxt_add(mvm, vif);
-	vif->bss_conf.assoc = true;
 	if (ret)
 		return ret;
 
@@ -705,7 +700,7 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		return ret;
 	rcu_assign_pointer(mvm->fw_id_to_mac_id[mvmvif->ap_sta_id], ap_sta);
 
-	ret = iwl_mvm_mac_ctxt_changed(mvm, vif);
+	ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false);
 	if (ret)
 		return ret;
 
@@ -719,7 +714,7 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	for (i = 1; i < MAX_BINDINGS; i++)
 		quota_cmd.quotas[i].id_and_color = cpu_to_le32(FW_CTXT_INVALID);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, TIME_QUOTA_CMD, CMD_SYNC,
+	ret = iwl_mvm_send_cmd_pdu(mvm, TIME_QUOTA_CMD, 0,
 				   sizeof(quota_cmd), &quota_cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send quota: %d\n", ret);
@@ -739,15 +734,13 @@ static int iwl_mvm_get_last_nonqos_seq(struct iwl_mvm *mvm,
 	};
 	struct iwl_host_cmd cmd = {
 		.id = NON_QOS_TX_COUNTER_CMD,
-		.flags = CMD_SYNC | CMD_WANT_SKB,
+		.flags = CMD_WANT_SKB,
 	};
 	int err;
 	u32 size;
 
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API) {
-		cmd.data[0] = &query_cmd;
-		cmd.len[0] = sizeof(query_cmd);
-	}
+	cmd.data[0] = &query_cmd;
+	cmd.len[0] = sizeof(query_cmd);
 
 	err = iwl_mvm_send_cmd(mvm, &cmd);
 	if (err)
@@ -758,10 +751,8 @@ static int iwl_mvm_get_last_nonqos_seq(struct iwl_mvm *mvm,
 		err = -EINVAL;
 	} else {
 		err = le16_to_cpup((__le16 *)cmd.resp_pkt->data);
-		/* new API returns next, not last-used seqno */
-		if (mvm->fw->ucode_capa.flags &
-				IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API)
-			err = (u16) (err - 0x10);
+		/* firmware returns next, not last-used seqno */
+		err = (u16) (err - 0x10);
 	}
 
 	iwl_free_resp(&cmd);
@@ -785,11 +776,7 @@ void iwl_mvm_set_last_nonqos_seq(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	mvmvif->seqno_valid = false;
 
-	if (!(mvm->fw->ucode_capa.flags &
-			IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API))
-		return;
-
-	if (iwl_mvm_send_cmd_pdu(mvm, NON_QOS_TX_COUNTER_CMD, CMD_SYNC,
+	if (iwl_mvm_send_cmd_pdu(mvm, NON_QOS_TX_COUNTER_CMD, 0,
 				 sizeof(query_cmd), &query_cmd))
 		IWL_ERR(mvm, "failed to set non-QoS seqno\n");
 }
@@ -804,7 +791,7 @@ iwl_mvm_send_wowlan_config_cmd(struct iwl_mvm *mvm,
 	if (mvm->fw->ucode_capa.api[0] & IWL_UCODE_TLV_API_WOWLAN_CONFIG_TID)
 		cmd_len = sizeof(*cmd);
 
-	return iwl_mvm_send_cmd_pdu(mvm, WOWLAN_CONFIGURATION, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, WOWLAN_CONFIGURATION, 0,
 				    cmd_len, cmd);
 }
 
@@ -833,7 +820,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 	};
 	struct iwl_host_cmd d3_cfg_cmd = {
 		.id = D3_CONFIG_CMD,
-		.flags = CMD_SYNC | CMD_WANT_SKB,
+		.flags = CMD_WANT_SKB,
 		.data[0] = &d3_cfg_cmd_data,
 		.len[0] = sizeof(d3_cfg_cmd_data),
 	};
@@ -983,7 +970,6 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 		if (key_data.use_rsc_tsc) {
 			struct iwl_host_cmd rsc_tsc_cmd = {
 				.id = WOWLAN_TSC_RSC_PARAM,
-				.flags = CMD_SYNC,
 				.data[0] = key_data.rsc_tsc,
 				.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
 				.len[0] = sizeof(*key_data.rsc_tsc),
@@ -997,7 +983,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 		if (key_data.use_tkip) {
 			ret = iwl_mvm_send_cmd_pdu(mvm,
 						   WOWLAN_TKIP_PARAM,
-						   CMD_SYNC, sizeof(tkip_cmd),
+						   0, sizeof(tkip_cmd),
 						   &tkip_cmd);
 			if (ret)
 				goto out;
@@ -1014,8 +1000,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 			kek_kck_cmd.replay_ctr = mvmvif->rekey_data.replay_ctr;
 
 			ret = iwl_mvm_send_cmd_pdu(mvm,
-						   WOWLAN_KEK_KCK_MATERIAL,
-						   CMD_SYNC,
+						   WOWLAN_KEK_KCK_MATERIAL, 0,
 						   sizeof(kek_kck_cmd),
 						   &kek_kck_cmd);
 			if (ret)
@@ -1031,7 +1016,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 	if (ret)
 		goto out;
 
-	ret = iwl_mvm_send_proto_offload(mvm, vif, false, CMD_SYNC);
+	ret = iwl_mvm_send_proto_offload(mvm, vif, false, 0);
 	if (ret)
 		goto out;
 
@@ -1043,7 +1028,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 	if (ret)
 		goto out;
 
-	ret = iwl_mvm_power_update_mac(mvm, vif);
+	ret = iwl_mvm_power_update_mac(mvm);
 	if (ret)
 		goto out;
 
@@ -1082,6 +1067,15 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 
 int iwl_mvm_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan)
 {
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	if (iwl_mvm_is_d0i3_supported(mvm)) {
+		mutex_lock(&mvm->d0i3_suspend_mutex);
+		__set_bit(D0I3_DEFER_WAKEUP, &mvm->d0i3_suspend_flags);
+		mutex_unlock(&mvm->d0i3_suspend_mutex);
+		return 0;
+	}
+
 	return __iwl_mvm_suspend(hw, wowlan, false);
 }
 
@@ -1277,7 +1271,7 @@ static void iwl_mvm_set_tkip_rx_seq(struct tkip_sc *scs,
 }
 
 static void iwl_mvm_set_key_rx_seq(struct ieee80211_key_conf *key,
-				   struct iwl_wowlan_status_v6 *status)
+				   struct iwl_wowlan_status *status)
 {
 	union iwl_all_tsc_rsc *rsc = &status->gtk.rsc.all_tsc_rsc;
 
@@ -1294,7 +1288,7 @@ static void iwl_mvm_set_key_rx_seq(struct ieee80211_key_conf *key,
 }
 
 struct iwl_mvm_d3_gtk_iter_data {
-	struct iwl_wowlan_status_v6 *status;
+	struct iwl_wowlan_status *status;
 	void *last_gtk;
 	u32 cipher;
 	bool find_phase, unhandled_cipher;
@@ -1370,7 +1364,7 @@ static void iwl_mvm_d3_update_gtks(struct ieee80211_hw *hw,
 
 static bool iwl_mvm_setup_connection_keep(struct iwl_mvm *mvm,
 					  struct ieee80211_vif *vif,
-					  struct iwl_wowlan_status_v6 *status)
+					  struct iwl_wowlan_status *status)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_d3_gtk_iter_data gtkdata = {
@@ -1465,10 +1459,10 @@ static bool iwl_mvm_query_wakeup_reasons(struct iwl_mvm *mvm,
 	} err_info;
 	struct iwl_host_cmd cmd = {
 		.id = WOWLAN_GET_STATUSES,
-		.flags = CMD_SYNC | CMD_WANT_SKB,
+		.flags = CMD_WANT_SKB,
 	};
 	struct iwl_wowlan_status_data status;
-	struct iwl_wowlan_status_v6 *status_v6;
+	struct iwl_wowlan_status *fw_status;
 	int ret, len, status_size, i;
 	bool keep;
 	struct ieee80211_sta *ap_sta;
@@ -1491,7 +1485,7 @@ static bool iwl_mvm_query_wakeup_reasons(struct iwl_mvm *mvm,
 	}
 
 	/* only for tracing for now */
-	ret = iwl_mvm_send_cmd_pdu(mvm, OFFLOADS_QUERY_CMD, CMD_SYNC, 0, NULL);
+	ret = iwl_mvm_send_cmd_pdu(mvm, OFFLOADS_QUERY_CMD, 0, 0, NULL);
 	if (ret)
 		IWL_ERR(mvm, "failed to query offload statistics (%d)\n", ret);
 
@@ -1505,10 +1499,7 @@ static bool iwl_mvm_query_wakeup_reasons(struct iwl_mvm *mvm,
 	if (!cmd.resp_pkt)
 		goto out_unlock;
 
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API)
-		status_size = sizeof(struct iwl_wowlan_status_v6);
-	else
-		status_size = sizeof(struct iwl_wowlan_status_v4);
+	status_size = sizeof(*fw_status);
 
 	len = iwl_rx_packet_payload_len(cmd.resp_pkt);
 	if (len < status_size) {
@@ -1516,35 +1507,18 @@ static bool iwl_mvm_query_wakeup_reasons(struct iwl_mvm *mvm,
 		goto out_free_resp;
 	}
 
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_D3_CONTINUITY_API) {
-		status_v6 = (void *)cmd.resp_pkt->data;
-
-		status.pattern_number = le16_to_cpu(status_v6->pattern_number);
-		for (i = 0; i < 8; i++)
-			status.qos_seq_ctr[i] =
-				le16_to_cpu(status_v6->qos_seq_ctr[i]);
-		status.wakeup_reasons = le32_to_cpu(status_v6->wakeup_reasons);
-		status.wake_packet_length =
-			le32_to_cpu(status_v6->wake_packet_length);
-		status.wake_packet_bufsize =
-			le32_to_cpu(status_v6->wake_packet_bufsize);
-		status.wake_packet = status_v6->wake_packet;
-	} else {
-		struct iwl_wowlan_status_v4 *status_v4;
-		status_v6 = NULL;
-		status_v4 = (void *)cmd.resp_pkt->data;
-
-		status.pattern_number = le16_to_cpu(status_v4->pattern_number);
-		for (i = 0; i < 8; i++)
-			status.qos_seq_ctr[i] =
-				le16_to_cpu(status_v4->qos_seq_ctr[i]);
-		status.wakeup_reasons = le32_to_cpu(status_v4->wakeup_reasons);
-		status.wake_packet_length =
-			le32_to_cpu(status_v4->wake_packet_length);
-		status.wake_packet_bufsize =
-			le32_to_cpu(status_v4->wake_packet_bufsize);
-		status.wake_packet = status_v4->wake_packet;
-	}
+	fw_status = (void *)cmd.resp_pkt->data;
+
+	status.pattern_number = le16_to_cpu(fw_status->pattern_number);
+	for (i = 0; i < 8; i++)
+		status.qos_seq_ctr[i] =
+			le16_to_cpu(fw_status->qos_seq_ctr[i]);
+	status.wakeup_reasons = le32_to_cpu(fw_status->wakeup_reasons);
+	status.wake_packet_length =
+		le32_to_cpu(fw_status->wake_packet_length);
+	status.wake_packet_bufsize =
+		le32_to_cpu(fw_status->wake_packet_bufsize);
+	status.wake_packet = fw_status->wake_packet;
 
 	if (len != status_size + ALIGN(status.wake_packet_bufsize, 4)) {
 		IWL_ERR(mvm, "Invalid WoWLAN status response!\n");
@@ -1571,7 +1545,7 @@ static bool iwl_mvm_query_wakeup_reasons(struct iwl_mvm *mvm,
 
 	iwl_mvm_report_wakeup_reasons(mvm, vif, &status);
 
-	keep = iwl_mvm_setup_connection_keep(mvm, vif, status_v6);
+	keep = iwl_mvm_setup_connection_keep(mvm, vif, fw_status);
 
 	iwl_free_resp(&cmd);
 	return keep;
@@ -1674,6 +1648,19 @@ int iwl_mvm_resume(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
+	if (iwl_mvm_is_d0i3_supported(mvm)) {
+		bool exit_now;
+
+		mutex_lock(&mvm->d0i3_suspend_mutex);
+		__clear_bit(D0I3_DEFER_WAKEUP, &mvm->d0i3_suspend_flags);
+		exit_now = __test_and_clear_bit(D0I3_PENDING_WAKEUP,
+						&mvm->d0i3_suspend_flags);
+		mutex_unlock(&mvm->d0i3_suspend_mutex);
+		if (exit_now)
+			_iwl_mvm_exit_d0i3(mvm);
+		return 0;
+	}
+
 	return __iwl_mvm_resume(mvm, false);
 }
 
diff --git a/drivers/net/wireless/iwlwifi/mvm/debugfs-vif.c b/drivers/net/wireless/iwlwifi/mvm/debugfs-vif.c
index 9b59e1d7ae71..2e90ff795c13 100644
--- a/drivers/net/wireless/iwlwifi/mvm/debugfs-vif.c
+++ b/drivers/net/wireless/iwlwifi/mvm/debugfs-vif.c
@@ -103,10 +103,6 @@ static void iwl_dbgfs_update_pm(struct iwl_mvm *mvm,
 		IWL_DEBUG_POWER(mvm, "tx_data_timeout=%d\n", val);
 		dbgfs_pm->tx_data_timeout = val;
 		break;
-	case MVM_DEBUGFS_PM_DISABLE_POWER_OFF:
-		IWL_DEBUG_POWER(mvm, "disable_power_off=%d\n", val);
-		dbgfs_pm->disable_power_off = val;
-		break;
 	case MVM_DEBUGFS_PM_LPRX_ENA:
 		IWL_DEBUG_POWER(mvm, "lprx %s\n", val ? "enabled" : "disabled");
 		dbgfs_pm->lprx_ena = val;
@@ -154,12 +150,6 @@ static ssize_t iwl_dbgfs_pm_params_write(struct ieee80211_vif *vif, char *buf,
 		if (sscanf(buf + 16, "%d", &val) != 1)
 			return -EINVAL;
 		param = MVM_DEBUGFS_PM_TX_DATA_TIMEOUT;
-	} else if (!strncmp("disable_power_off=", buf, 18) &&
-		   !(mvm->fw->ucode_capa.flags &
-		     IWL_UCODE_TLV_FLAGS_DEVICE_PS_CMD)) {
-		if (sscanf(buf + 18, "%d", &val) != 1)
-			return -EINVAL;
-		param = MVM_DEBUGFS_PM_DISABLE_POWER_OFF;
 	} else if (!strncmp("lprx=", buf, 5)) {
 		if (sscanf(buf + 5, "%d", &val) != 1)
 			return -EINVAL;
@@ -185,7 +175,7 @@ static ssize_t iwl_dbgfs_pm_params_write(struct ieee80211_vif *vif, char *buf,
 
 	mutex_lock(&mvm->mutex);
 	iwl_dbgfs_update_pm(mvm, vif, param, val);
-	ret = iwl_mvm_power_update_mac(mvm, vif);
+	ret = iwl_mvm_power_update_mac(mvm);
 	mutex_unlock(&mvm->mutex);
 
 	return ret ?: count;
@@ -272,10 +262,9 @@ static ssize_t iwl_dbgfs_mac_params_read(struct file *file,
 			struct iwl_mvm_sta *mvm_sta = (void *)sta->drv_priv;
 
 			pos += scnprintf(buf+pos, bufsz-pos,
-					 "ap_sta_id %d - reduced Tx power %d force %d\n",
+					 "ap_sta_id %d - reduced Tx power %d\n",
 					 ap_sta_id,
-					 mvm_sta->bt_reduced_txpower,
-					 mvm_sta->bt_reduced_txpower_dbg);
+					 mvm_sta->bt_reduced_txpower);
 		}
 	}
 
@@ -293,41 +282,6 @@ static ssize_t iwl_dbgfs_mac_params_read(struct file *file,
 	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
 }
 
-static ssize_t iwl_dbgfs_reduced_txp_write(struct ieee80211_vif *vif,
-					   char *buf, size_t count,
-					   loff_t *ppos)
-{
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm *mvm = mvmvif->mvm;
-	struct iwl_mvm_sta *mvmsta;
-	bool reduced_tx_power;
-	int ret;
-
-	if (mvmvif->ap_sta_id >= ARRAY_SIZE(mvm->fw_id_to_mac_id))
-		return -ENOTCONN;
-
-	if (strtobool(buf, &reduced_tx_power) != 0)
-		return -EINVAL;
-
-	mutex_lock(&mvm->mutex);
-
-	mvmsta = iwl_mvm_sta_from_staid_protected(mvm, mvmvif->ap_sta_id);
-	if (IS_ERR_OR_NULL(mvmsta)) {
-		mutex_unlock(&mvm->mutex);
-		return -ENOTCONN;
-	}
-
-	mvmsta->bt_reduced_txpower_dbg = false;
-	ret = iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id,
-					  reduced_tx_power);
-	if (!ret)
-		mvmsta->bt_reduced_txpower_dbg = true;
-
-	mutex_unlock(&mvm->mutex);
-
-	return ret ? : count;
-}
-
 static void iwl_dbgfs_update_bf(struct ieee80211_vif *vif,
 				enum iwl_dbgfs_bf_mask param, int value)
 {
@@ -462,9 +416,9 @@ static ssize_t iwl_dbgfs_bf_params_write(struct ieee80211_vif *vif, char *buf,
 	mutex_lock(&mvm->mutex);
 	iwl_dbgfs_update_bf(vif, param, value);
 	if (param == MVM_DEBUGFS_BF_ENABLE_BEACON_FILTER && !value)
-		ret = iwl_mvm_disable_beacon_filter(mvm, vif, CMD_SYNC);
+		ret = iwl_mvm_disable_beacon_filter(mvm, vif, 0);
 	else
-		ret = iwl_mvm_enable_beacon_filter(mvm, vif, CMD_SYNC);
+		ret = iwl_mvm_enable_beacon_filter(mvm, vif, 0);
 	mutex_unlock(&mvm->mutex);
 
 	return ret ?: count;
@@ -568,7 +522,6 @@ MVM_DEBUGFS_READ_FILE_OPS(mac_params);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(pm_params, 32);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(bf_params, 256);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(low_latency, 10);
-MVM_DEBUGFS_WRITE_FILE_OPS(reduced_txp, 10);
 
 void iwl_mvm_vif_dbgfs_register(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
@@ -592,8 +545,7 @@ void iwl_mvm_vif_dbgfs_register(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		return;
 	}
 
-	if ((mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_PM_CMD_SUPPORT) &&
-	    iwlmvm_mod_params.power_scheme != IWL_POWER_SCHEME_CAM &&
+	if (iwlmvm_mod_params.power_scheme != IWL_POWER_SCHEME_CAM &&
 	    ((vif->type == NL80211_IFTYPE_STATION && !vif->p2p) ||
 	     (vif->type == NL80211_IFTYPE_STATION && vif->p2p &&
 	      mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_DCM)))
@@ -601,7 +553,6 @@ void iwl_mvm_vif_dbgfs_register(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 					 S_IRUSR);
 
 	MVM_DEBUGFS_ADD_FILE_VIF(mac_params, mvmvif->dbgfs_dir, S_IRUSR);
-	MVM_DEBUGFS_ADD_FILE_VIF(reduced_txp, mvmvif->dbgfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE_VIF(low_latency, mvmvif->dbgfs_dir,
 				 S_IRUSR | S_IWUSR);
 
diff --git a/drivers/net/wireless/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/iwlwifi/mvm/debugfs.c
index 1b52deea6081..29ca72695eaa 100644
--- a/drivers/net/wireless/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/iwlwifi/mvm/debugfs.c
@@ -65,9 +65,8 @@
 #include "mvm.h"
 #include "sta.h"
 #include "iwl-io.h"
-#include "iwl-prph.h"
 #include "debugfs.h"
-#include "fw-error-dump.h"
+#include "iwl-fw-error-dump.h"
 
 static ssize_t iwl_dbgfs_tx_flush_write(struct iwl_mvm *mvm, char *buf,
 					size_t count, loff_t *ppos)
@@ -136,9 +135,6 @@ static int iwl_dbgfs_fw_error_dump_open(struct inode *inode, struct file *file)
 
 	file->private_data = mvm->fw_error_dump;
 	mvm->fw_error_dump = NULL;
-	kfree(mvm->fw_error_sram);
-	mvm->fw_error_sram = NULL;
-	mvm->fw_error_sram_len = 0;
 	ret = 0;
 
 out:
@@ -684,7 +680,7 @@ static ssize_t iwl_dbgfs_fw_restart_write(struct iwl_mvm *mvm, char *buf,
 		mvm->restart_fw++;
 
 	/* take the return value to make compiler happy - it will fail anyway */
-	ret = iwl_mvm_send_cmd_pdu(mvm, REPLY_ERROR, CMD_SYNC, 0, NULL);
+	ret = iwl_mvm_send_cmd_pdu(mvm, REPLY_ERROR, 0, 0, NULL);
 
 	mutex_unlock(&mvm->mutex);
 
@@ -694,7 +690,7 @@ static ssize_t iwl_dbgfs_fw_restart_write(struct iwl_mvm *mvm, char *buf,
 static ssize_t iwl_dbgfs_fw_nmi_write(struct iwl_mvm *mvm, char *buf,
 				      size_t count, loff_t *ppos)
 {
-	iwl_write_prph(mvm->trans, DEVICE_SET_NMI_REG, 1);
+	iwl_force_nmi(mvm->trans);
 
 	return count;
 }
@@ -841,7 +837,7 @@ static ssize_t iwl_dbgfs_bcast_filters_write(struct iwl_mvm *mvm, char *buf,
 	/* send updated bcast filtering configuration */
 	if (mvm->dbgfs_bcast_filtering.override &&
 	    iwl_mvm_bcast_filter_build_cmd(mvm, &cmd))
-		err = iwl_mvm_send_cmd_pdu(mvm, BCAST_FILTER_CMD, CMD_SYNC,
+		err = iwl_mvm_send_cmd_pdu(mvm, BCAST_FILTER_CMD, 0,
 					   sizeof(cmd), &cmd);
 	mutex_unlock(&mvm->mutex);
 
@@ -913,7 +909,7 @@ static ssize_t iwl_dbgfs_bcast_filters_macs_write(struct iwl_mvm *mvm,
 	/* send updated bcast filtering configuration */
 	if (mvm->dbgfs_bcast_filtering.override &&
 	    iwl_mvm_bcast_filter_build_cmd(mvm, &cmd))
-		err = iwl_mvm_send_cmd_pdu(mvm, BCAST_FILTER_CMD, CMD_SYNC,
+		err = iwl_mvm_send_cmd_pdu(mvm, BCAST_FILTER_CMD, 0,
 					   sizeof(cmd), &cmd);
 	mutex_unlock(&mvm->mutex);
 
@@ -1004,6 +1000,7 @@ static ssize_t iwl_dbgfs_d0i3_refs_read(struct file *file,
 	PRINT_MVM_REF(IWL_MVM_REF_P2P_CLIENT);
 	PRINT_MVM_REF(IWL_MVM_REF_AP_IBSS);
 	PRINT_MVM_REF(IWL_MVM_REF_USER);
+	PRINT_MVM_REF(IWL_MVM_REF_EXIT_WORK);
 
 	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
 }
@@ -1108,9 +1105,9 @@ MVM_DEBUGFS_READ_WRITE_FILE_OPS(scan_ant_rxchain, 8);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(d0i3_refs, 8);
 
 static const struct file_operations iwl_dbgfs_fw_error_dump_ops = {
-        .open = iwl_dbgfs_fw_error_dump_open,
-        .read = iwl_dbgfs_fw_error_dump_read,
-        .release = iwl_dbgfs_fw_error_dump_release,
+	.open = iwl_dbgfs_fw_error_dump_open,
+	.read = iwl_dbgfs_fw_error_dump_read,
+	.release = iwl_dbgfs_fw_error_dump_release,
 };
 
 #ifdef CONFIG_IWLWIFI_BCAST_FILTERING
@@ -1138,9 +1135,8 @@ int iwl_mvm_dbgfs_register(struct iwl_mvm *mvm, struct dentry *dbgfs_dir)
 	MVM_DEBUGFS_ADD_FILE(fw_error_dump, dbgfs_dir, S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(bt_notif, dbgfs_dir, S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(bt_cmd, dbgfs_dir, S_IRUSR);
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_DEVICE_PS_CMD)
-		MVM_DEBUGFS_ADD_FILE(disable_power_off, mvm->debugfs_dir,
-				     S_IRUSR | S_IWUSR);
+	MVM_DEBUGFS_ADD_FILE(disable_power_off, mvm->debugfs_dir,
+			     S_IRUSR | S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(fw_rx_stats, mvm->debugfs_dir, S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(drv_rx_stats, mvm->debugfs_dir, S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(fw_restart, mvm->debugfs_dir, S_IWUSR);
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-coex.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-coex.h
index 21877e5966a8..5fe82c29c8ad 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-coex.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-coex.h
@@ -141,7 +141,8 @@ enum iwl_bt_coex_lut_type {
 	BT_COEX_TX_DIS_LUT,
 
 	BT_COEX_MAX_LUT,
-};
+	BT_COEX_INVALID_LUT = 0xff,
+}; /* BT_COEX_DECISION_LUT_INDEX_API_E_VER_1 */
 
 #define BT_COEX_LUT_SIZE (12)
 #define BT_COEX_CORUN_LUT_SIZE (32)
@@ -154,19 +155,23 @@ enum iwl_bt_coex_lut_type {
  * @flags:&enum iwl_bt_coex_flags
  * @max_kill:
  * @bt_reduced_tx_power: enum %iwl_bt_reduced_tx_power
- * @bt4_antenna_isolation:
- * @bt4_antenna_isolation_thr:
- * @bt4_tx_tx_delta_freq_thr:
- * @bt4_tx_rx_max_freq0:
- * @bt_prio_boost:
+ * @override_primary_lut: enum %iwl_bt_coex_lut_type: BT_COEX_INVALID_LUT
+ *	should be set by default
+ * @override_secondary_lut: enum %iwl_bt_coex_lut_type: BT_COEX_INVALID_LUT
+ *	should be set by default
+ * @bt4_antenna_isolation: antenna isolation
+ * @bt4_antenna_isolation_thr: antenna threshold value
+ * @bt4_tx_tx_delta_freq_thr: TxTx delta frequency
+ * @bt4_tx_rx_max_freq0: TxRx max frequency
+ * @bt_prio_boost: BT priority boost registers
  * @wifi_tx_prio_boost: SW boost of wifi tx priority
  * @wifi_rx_prio_boost: SW boost of wifi rx priority
- * @kill_ack_msk:
- * @kill_cts_msk:
- * @decision_lut:
- * @bt4_multiprio_lut:
- * @bt4_corun_lut20:
- * @bt4_corun_lut40:
+ * @kill_ack_msk: kill ACK mask. 1 - Tx ACK, 0 - kill Tx of ACK.
+ * @kill_cts_msk: kill CTS mask. 1 - Tx CTS, 0 - kill Tx of CTS.
+ * @decision_lut: PTA decision LUT, per Prio-Ch
+ * @bt4_multiprio_lut: multi priority LUT configuration
+ * @bt4_corun_lut20: co-running 20 MHz LUT configuration
+ * @bt4_corun_lut40: co-running 40 MHz LUT configuration
  * @valid_bit_msk: enum %iwl_bt_coex_valid_bit_msk
  *
  * The structure is used for the BT_COEX command.
@@ -175,7 +180,8 @@ struct iwl_bt_coex_cmd {
 	__le32 flags;
 	u8 max_kill;
 	u8 bt_reduced_tx_power;
-	u8 reserved[2];
+	u8 override_primary_lut;
+	u8 override_secondary_lut;
 
 	u8 bt4_antenna_isolation;
 	u8 bt4_antenna_isolation_thr;
@@ -194,7 +200,7 @@ struct iwl_bt_coex_cmd {
 	__le32 bt4_corun_lut40[BT_COEX_CORUN_LUT_SIZE];
 
 	__le32 valid_bit_msk;
-} __packed; /* BT_COEX_CMD_API_S_VER_3 */
+} __packed; /* BT_COEX_CMD_API_S_VER_5 */
 
 /**
  * struct iwl_bt_coex_ci_cmd - bt coex channel inhibition command
@@ -282,7 +288,7 @@ enum iwl_bt_activity_grading {
 	BT_ON_NO_CONNECTION	= 1,
 	BT_LOW_TRAFFIC		= 2,
 	BT_HIGH_TRAFFIC		= 3,
-};
+}; /* BT_COEX_BT_ACTIVITY_GRADING_API_E_VER_1 */
 
 /**
  * struct iwl_bt_coex_profile_notif - notification about BT coex
@@ -310,7 +316,7 @@ struct iwl_bt_coex_profile_notif {
 	__le32 primary_ch_lut;
 	__le32 secondary_ch_lut;
 	__le32 bt_activity_grading;
-} __packed; /* BT_COEX_PROFILE_NTFY_API_S_VER_2 */
+} __packed; /* BT_COEX_PROFILE_NTFY_API_S_VER_3 */
 
 enum iwl_bt_coex_prio_table_event {
 	BT_COEX_PRIO_TBL_EVT_INIT_CALIB1		= 0,
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-d3.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-d3.h
index 10fcc1a79ebd..13696fe419b7 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-d3.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-d3.h
@@ -345,21 +345,6 @@ enum iwl_wowlan_wakeup_reason {
 	IWL_WOWLAN_WAKEUP_BY_REM_WAKE_WAKEUP_PACKET		= BIT(12),
 }; /* WOWLAN_WAKE_UP_REASON_API_E_VER_2 */
 
-struct iwl_wowlan_status_v4 {
-	__le64 replay_ctr;
-	__le16 pattern_number;
-	__le16 non_qos_seq_ctr;
-	__le16 qos_seq_ctr[8];
-	__le32 wakeup_reasons;
-	__le32 rekey_status;
-	__le32 num_of_gtk_rekeys;
-	__le32 transmitted_ndps;
-	__le32 received_beacons;
-	__le32 wake_packet_length;
-	__le32 wake_packet_bufsize;
-	u8 wake_packet[]; /* can be truncated from _length to _bufsize */
-} __packed; /* WOWLAN_STATUSES_API_S_VER_4 */
-
 struct iwl_wowlan_gtk_status {
 	u8 key_index;
 	u8 reserved[3];
@@ -368,7 +353,7 @@ struct iwl_wowlan_gtk_status {
 	struct iwl_wowlan_rsc_tsc_params_cmd rsc;
 } __packed;
 
-struct iwl_wowlan_status_v6 {
+struct iwl_wowlan_status {
 	struct iwl_wowlan_gtk_status gtk;
 	__le64 replay_ctr;
 	__le16 pattern_number;
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-rs.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-rs.h
index 39148b5bb332..8bb5b94bf963 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-rs.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-rs.h
@@ -334,7 +334,7 @@ enum {
  */
 struct iwl_lq_cmd {
 	u8 sta_id;
-	u8 reserved1;
+	u8 reduced_tpc;
 	u16 control;
 	/* LINK_QUAL_GENERAL_PARAMS_API_S_VER_1 */
 	u8 flags;
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h
index d73a89ecd78a..6959fda3fe09 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h
@@ -169,8 +169,12 @@ enum iwl_scan_type {
 	SCAN_TYPE_DISCOVERY_FORCED	= 6,
 }; /* SCAN_ACTIVITY_TYPE_E_VER_1 */
 
-/* Maximal number of channels to scan */
-#define MAX_NUM_SCAN_CHANNELS 0x24
+/**
+ * Maximal number of channels to scan
+ * it should be equal to:
+ * max(IWL_NUM_CHANNELS, IWL_NUM_CHANNELS_FAMILY_8000)
+ */
+#define MAX_NUM_SCAN_CHANNELS 50
 
 /**
  * struct iwl_scan_cmd - scan request command
@@ -534,13 +538,16 @@ struct iwl_scan_offload_schedule {
  *
  * IWL_SCAN_OFFLOAD_FLAG_PASS_ALL: pass all results - no filtering.
  * IWL_SCAN_OFFLOAD_FLAG_CACHED_CHANNEL: add cached channels to partial scan.
- * IWL_SCAN_OFFLOAD_FLAG_ENERGY_SCAN: use energy based scan before partial scan
- *	on A band.
+ * IWL_SCAN_OFFLOAD_FLAG_EBS_QUICK_MODE: EBS duration is 100mSec - typical
+ *	beacon period. Finding channel activity in this mode is not guaranteed.
+ * IWL_SCAN_OFFLOAD_FLAG_EBS_ACCURATE_MODE: EBS duration is 200mSec.
+ *	Assuming beacon period is 100ms finding channel activity is guaranteed.
  */
 enum iwl_scan_offload_flags {
 	IWL_SCAN_OFFLOAD_FLAG_PASS_ALL		= BIT(0),
 	IWL_SCAN_OFFLOAD_FLAG_CACHED_CHANNEL	= BIT(2),
-	IWL_SCAN_OFFLOAD_FLAG_ENERGY_SCAN	= BIT(3),
+	IWL_SCAN_OFFLOAD_FLAG_EBS_QUICK_MODE	= BIT(5),
+	IWL_SCAN_OFFLOAD_FLAG_EBS_ACCURATE_MODE	= BIT(6),
 };
 
 /**
@@ -563,17 +570,24 @@ enum iwl_scan_offload_compleate_status {
 	IWL_SCAN_OFFLOAD_ABORTED	= 2,
 };
 
+enum iwl_scan_ebs_status {
+	IWL_SCAN_EBS_SUCCESS,
+	IWL_SCAN_EBS_FAILED,
+	IWL_SCAN_EBS_CHAN_NOT_FOUND,
+};
+
 /**
  * iwl_scan_offload_complete - SCAN_OFFLOAD_COMPLETE_NTF_API_S_VER_1
  * @last_schedule_line:		last schedule line executed (fast or regular)
  * @last_schedule_iteration:	last scan iteration executed before scan abort
  * @status:			enum iwl_scan_offload_compleate_status
+ * @ebs_status: last EBS status, see IWL_SCAN_EBS_*
  */
 struct iwl_scan_offload_complete {
 	u8 last_schedule_line;
 	u8 last_schedule_iteration;
 	u8 status;
-	u8 reserved;
+	u8 ebs_status;
 } __packed;
 
 /**
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-sta.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-sta.h
index d63647867262..39cebee8016f 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-sta.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-sta.h
@@ -255,22 +255,19 @@ struct iwl_mvm_keyinfo {
 } __packed;
 
 /**
- * struct iwl_mvm_add_sta_cmd_v5 - Add/modify a station in the fw's sta table.
+ * struct iwl_mvm_add_sta_cmd - Add/modify a station in the fw's sta table.
  * ( REPLY_ADD_STA = 0x18 )
  * @add_modify: 1: modify existing, 0: add new station
- * @unicast_tx_key_id: unicast tx key id. Relevant only when unicast key sent
- * @multicast_tx_key_id: multicast tx key id. Relevant only when multicast key
- *	sent
+ * @awake_acs:
+ * @tid_disable_tx: is tid BIT(tid) enabled for Tx. Clear BIT(x) to enable
+ *	AMPDU for tid x. Set %STA_MODIFY_TID_DISABLE_TX to change this field.
  * @mac_id_n_color: the Mac context this station belongs to
  * @addr[ETH_ALEN]: station's MAC address
  * @sta_id: index of station in uCode's station table
  * @modify_mask: STA_MODIFY_*, selects which parameters to modify vs. leave
  *	alone. 1 - modify, 0 - don't change.
- * @key: look at %iwl_mvm_keyinfo
  * @station_flags: look at %iwl_sta_flags
  * @station_flags_msk: what of %station_flags have changed
- * @tid_disable_tx: is tid BIT(tid) enabled for Tx. Clear BIT(x) to enable
- *	AMPDU for tid x. Set %STA_MODIFY_TID_DISABLE_TX to change this field.
  * @add_immediate_ba_tid: tid for which to add block-ack support (Rx)
  *	Set %STA_MODIFY_ADD_BA_TID to use this field, and also set
  *	add_immediate_ba_ssn.
@@ -294,40 +291,7 @@ struct iwl_mvm_keyinfo {
  * ADD_STA sets up the table entry for one station, either creating a new
  * entry, or modifying a pre-existing one.
  */
-struct iwl_mvm_add_sta_cmd_v5 {
-	u8 add_modify;
-	u8 unicast_tx_key_id;
-	u8 multicast_tx_key_id;
-	u8 reserved1;
-	__le32 mac_id_n_color;
-	u8 addr[ETH_ALEN];
-	__le16 reserved2;
-	u8 sta_id;
-	u8 modify_mask;
-	__le16 reserved3;
-	struct iwl_mvm_keyinfo key;
-	__le32 station_flags;
-	__le32 station_flags_msk;
-	__le16 tid_disable_tx;
-	__le16 reserved4;
-	u8 add_immediate_ba_tid;
-	u8 remove_immediate_ba_tid;
-	__le16 add_immediate_ba_ssn;
-	__le16 sleep_tx_count;
-	__le16 sleep_state_flags;
-	__le16 assoc_id;
-	__le16 beamform_flags;
-	__le32 tfd_queue_msk;
-} __packed; /* ADD_STA_CMD_API_S_VER_5 */
-
-/**
- * struct iwl_mvm_add_sta_cmd_v7 - Add / modify a station
- * VER_7 of this command is quite similar to VER_5 except
- * exclusion of all fields related to the security key installation.
- * It only differs from VER_6 by the "awake_acs" field that is
- * reserved and ignored in VER_6.
- */
-struct iwl_mvm_add_sta_cmd_v7 {
+struct iwl_mvm_add_sta_cmd {
 	u8 add_modify;
 	u8 awake_acs;
 	__le16 tid_disable_tx;
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-tx.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-tx.h
index 8e122f3a7a74..6cc5f52b807f 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-tx.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-tx.h
@@ -482,7 +482,8 @@ struct iwl_mvm_tx_resp {
 	u8 pa_integ_res_b[3];
 	u8 pa_integ_res_c[3];
 	__le16 measurement_req_id;
-	__le16 reserved;
+	u8 reduced_tpc;
+	u8 reserved;
 
 	__le32 tfd_info;
 	__le16 seq_ctl;
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/iwlwifi/mvm/fw-api.h
index 6e75b52588de..309a9b9a94fe 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api.h
@@ -71,6 +71,7 @@
 #include "fw-api-power.h"
 #include "fw-api-d3.h"
 #include "fw-api-coex.h"
+#include "fw-api-scan.h"
 
 /* maximal number of Tx queues in any platform */
 #define IWL_MVM_MAX_QUEUES	20
@@ -604,52 +605,7 @@ enum {
 	TE_V1_NOTIF_INTERNAL_FRAG_END = BIT(7),
 }; /* MAC_EVENT_ACTION_API_E_VER_2 */
 
-
-/**
- * struct iwl_time_event_cmd_api_v1 - configuring Time Events
- * with struct MAC_TIME_EVENT_DATA_API_S_VER_1 (see also
- * with version 2. determined by IWL_UCODE_TLV_FLAGS)
- * ( TIME_EVENT_CMD = 0x29 )
- * @id_and_color: ID and color of the relevant MAC
- * @action: action to perform, one of FW_CTXT_ACTION_*
- * @id: this field has two meanings, depending on the action:
- *	If the action is ADD, then it means the type of event to add.
- *	For all other actions it is the unique event ID assigned when the
- *	event was added by the FW.
- * @apply_time: When to start the Time Event (in GP2)
- * @max_delay: maximum delay to event's start (apply time), in TU
- * @depends_on: the unique ID of the event we depend on (if any)
- * @interval: interval between repetitions, in TU
- * @interval_reciprocal: 2^32 / interval
- * @duration: duration of event in TU
- * @repeat: how many repetitions to do, can be TE_REPEAT_ENDLESS
- * @dep_policy: one of TE_V1_INDEPENDENT, TE_V1_DEP_OTHER, TE_V1_DEP_TSF
- *	and TE_V1_EVENT_SOCIOPATHIC
- * @is_present: 0 or 1, are we present or absent during the Time Event
- * @max_frags: maximal number of fragments the Time Event can be divided to
- * @notify: notifications using TE_V1_NOTIF_* (whom to notify when)
- */
-struct iwl_time_event_cmd_v1 {
-	/* COMMON_INDEX_HDR_API_S_VER_1 */
-	__le32 id_and_color;
-	__le32 action;
-	__le32 id;
-	/* MAC_TIME_EVENT_DATA_API_S_VER_1 */
-	__le32 apply_time;
-	__le32 max_delay;
-	__le32 dep_policy;
-	__le32 depends_on;
-	__le32 is_present;
-	__le32 max_frags;
-	__le32 interval;
-	__le32 interval_reciprocal;
-	__le32 duration;
-	__le32 repeat;
-	__le32 notify;
-} __packed; /* MAC_TIME_EVENT_CMD_API_S_VER_1 */
-
-
-/* Time event - defines for command API v2 */
+/* Time event - defines for command API */
 
 /*
  * @TE_V2_FRAG_NONE: fragmentation of the time event is NOT allowed.
@@ -680,7 +636,7 @@ enum {
 #define TE_V2_PLACEMENT_POS	12
 #define TE_V2_ABSENCE_POS	15
 
-/* Time event policy values (for time event cmd api v2)
+/* Time event policy values
  * A notification (both event and fragment) includes a status indicating weather
  * the FW was able to schedule the event or not. For fragment start/end
  * notification the status is always success. There is no start/end fragment
@@ -727,7 +683,7 @@ enum {
 };
 
 /**
- * struct iwl_time_event_cmd_api_v2 - configuring Time Events
+ * struct iwl_time_event_cmd_api - configuring Time Events
  * with struct MAC_TIME_EVENT_DATA_API_S_VER_2 (see also
  * with version 1. determined by IWL_UCODE_TLV_FLAGS)
  * ( TIME_EVENT_CMD = 0x29 )
@@ -750,7 +706,7 @@ enum {
  *	TE_EVENT_SOCIOPATHIC
  *	using TE_ABSENCE and using TE_NOTIF_*
  */
-struct iwl_time_event_cmd_v2 {
+struct iwl_time_event_cmd {
 	/* COMMON_INDEX_HDR_API_S_VER_1 */
 	__le32 id_and_color;
 	__le32 action;
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw.c b/drivers/net/wireless/iwlwifi/mvm/fw.c
index 7ce20062f32d..883e702152d5 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/iwlwifi/mvm/fw.c
@@ -99,7 +99,7 @@ static int iwl_send_tx_ant_cfg(struct iwl_mvm *mvm, u8 valid_tx_ant)
 	};
 
 	IWL_DEBUG_FW(mvm, "select valid tx ant: %u\n", valid_tx_ant);
-	return iwl_mvm_send_cmd_pdu(mvm, TX_ANT_CONFIGURATION_CMD, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, TX_ANT_CONFIGURATION_CMD, 0,
 				    sizeof(tx_ant_cmd), &tx_ant_cmd);
 }
 
@@ -137,6 +137,8 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait,
 		alive_data->scd_base_addr = le32_to_cpu(palive2->scd_base_ptr);
 		mvm->umac_error_event_table =
 			le32_to_cpu(palive2->error_info_addr);
+		mvm->sf_space.addr = le32_to_cpu(palive2->st_fwrd_addr);
+		mvm->sf_space.size = le32_to_cpu(palive2->st_fwrd_size);
 
 		alive_data->valid = le16_to_cpu(palive2->status) ==
 				    IWL_ALIVE_STATUS_OK;
@@ -180,6 +182,7 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 	int ret, i;
 	enum iwl_ucode_type old_type = mvm->cur_ucode;
 	static const u8 alive_cmd[] = { MVM_ALIVE };
+	struct iwl_sf_region st_fwrd_space;
 
 	fw = iwl_get_ucode_image(mvm, ucode_type);
 	if (WARN_ON(!fw))
@@ -215,6 +218,14 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 		return -EIO;
 	}
 
+	/*
+	 * update the sdio allocation according to the pointer we get in the
+	 * alive notification.
+	 */
+	st_fwrd_space.addr = mvm->sf_space.addr;
+	st_fwrd_space.size = mvm->sf_space.size;
+	ret = iwl_trans_update_sf(mvm->trans, &st_fwrd_space);
+
 	iwl_trans_fw_alive(mvm->trans, alive_data.scd_base_addr);
 
 	/*
@@ -256,7 +267,7 @@ static int iwl_send_phy_cfg_cmd(struct iwl_mvm *mvm)
 	IWL_DEBUG_INFO(mvm, "Sending Phy CFG command: 0x%x\n",
 		       phy_cfg_cmd.phy_cfg);
 
-	return iwl_mvm_send_cmd_pdu(mvm, PHY_CONFIGURATION_CMD, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, PHY_CONFIGURATION_CMD, 0,
 				    sizeof(phy_cfg_cmd), &phy_cfg_cmd);
 }
 
@@ -288,14 +299,14 @@ int iwl_run_init_mvm_ucode(struct iwl_mvm *mvm, bool read_nvm)
 		goto error;
 	}
 
-	ret = iwl_send_bt_prio_tbl(mvm);
+	ret = iwl_send_bt_init_conf(mvm);
 	if (ret)
 		goto error;
 
 	/* Read the NVM only at driver load time, no need to do this twice */
 	if (read_nvm) {
 		/* Read nvm */
-		ret = iwl_nvm_init(mvm);
+		ret = iwl_nvm_init(mvm, true);
 		if (ret) {
 			IWL_ERR(mvm, "Failed to read NVM: %d\n", ret);
 			goto error;
@@ -303,7 +314,7 @@ int iwl_run_init_mvm_ucode(struct iwl_mvm *mvm, bool read_nvm)
 	}
 
 	/* In case we read the NVM from external file, load it to the NIC */
-	if (iwlwifi_mod_params.nvm_file)
+	if (mvm->nvm_file_name)
 		iwl_mvm_load_nvm_to_nic(mvm);
 
 	ret = iwl_nvm_check_version(mvm->nvm_data, mvm->trans);
@@ -424,10 +435,6 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	if (ret)
 		goto error;
 
-	ret = iwl_send_bt_prio_tbl(mvm);
-	if (ret)
-		goto error;
-
 	ret = iwl_send_bt_init_conf(mvm);
 	if (ret)
 		goto error;
@@ -468,12 +475,6 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	/* Initialize tx backoffs to the minimal possible */
 	iwl_mvm_tt_tx_backoff(mvm, 0);
 
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_PM_CMD_SUPPORT)) {
-		ret = iwl_power_legacy_set_cam_mode(mvm);
-		if (ret)
-			goto error;
-	}
-
 	ret = iwl_mvm_power_update_device(mvm);
 	if (ret)
 		goto error;
diff --git a/drivers/net/wireless/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/iwlwifi/mvm/mac-ctxt.c
index 9ccec10bba16..8b5302777632 100644
--- a/drivers/net/wireless/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/iwlwifi/mvm/mac-ctxt.c
@@ -667,12 +667,9 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 	if (vif->bss_conf.qos)
 		cmd->qos_flags |= cpu_to_le32(MAC_QOS_FLG_UPDATE_EDCA);
 
-	/* Don't use cts to self as the fw doesn't support it currently. */
 	if (vif->bss_conf.use_cts_prot) {
 		cmd->protection_flags |= cpu_to_le32(MAC_PROT_FLG_TGG_PROTECT);
-		if (IWL_UCODE_API(mvm->fw->ucode_ver) >= 8)
-			cmd->protection_flags |=
-				cpu_to_le32(MAC_PROT_FLG_SELF_CTS_EN);
+		cmd->protection_flags |= cpu_to_le32(MAC_PROT_FLG_SELF_CTS_EN);
 	}
 	IWL_DEBUG_RATE(mvm, "use_cts_prot %d, ht_operation_mode %d\n",
 		       vif->bss_conf.use_cts_prot,
@@ -688,7 +685,7 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 static int iwl_mvm_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
 				     struct iwl_mac_ctx_cmd *cmd)
 {
-	int ret = iwl_mvm_send_cmd_pdu(mvm, MAC_CONTEXT_CMD, CMD_SYNC,
+	int ret = iwl_mvm_send_cmd_pdu(mvm, MAC_CONTEXT_CMD, 0,
 				       sizeof(*cmd), cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send MAC context (action:%d): %d\n",
@@ -696,19 +693,39 @@ static int iwl_mvm_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
 	return ret;
 }
 
-/*
- * Fill the specific data for mac context of type station or p2p client
- */
-static void iwl_mvm_mac_ctxt_cmd_fill_sta(struct iwl_mvm *mvm,
-					  struct ieee80211_vif *vif,
-					  struct iwl_mac_data_sta *ctxt_sta,
-					  bool force_assoc_off)
+static int iwl_mvm_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
+				    struct ieee80211_vif *vif,
+				    u32 action, bool force_assoc_off)
 {
+	struct iwl_mac_ctx_cmd cmd = {};
+	struct iwl_mac_data_sta *ctxt_sta;
+
+	WARN_ON(vif->type != NL80211_IFTYPE_STATION);
+
+	/* Fill the common data for all mac context types */
+	iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
+
+	if (vif->p2p) {
+		struct ieee80211_p2p_noa_attr *noa =
+			&vif->bss_conf.p2p_noa_attr;
+
+		cmd.p2p_sta.ctwin = cpu_to_le32(noa->oppps_ctwindow &
+					IEEE80211_P2P_OPPPS_CTWINDOW_MASK);
+		ctxt_sta = &cmd.p2p_sta.sta;
+	} else {
+		ctxt_sta = &cmd.sta;
+	}
+
 	/* We need the dtim_period to set the MAC as associated */
 	if (vif->bss_conf.assoc && vif->bss_conf.dtim_period &&
 	    !force_assoc_off) {
 		u32 dtim_offs;
 
+		/* Allow beacons to pass through as long as we are not
+		 * associated, or we do not have dtim period information.
+		 */
+		cmd.filter_flags |= cpu_to_le32(MAC_FILTER_IN_BEACON);
+
 		/*
 		 * The DTIM count counts down, so when it is N that means N
 		 * more beacon intervals happen until the DTIM TBTT. Therefore
@@ -755,51 +772,6 @@ static void iwl_mvm_mac_ctxt_cmd_fill_sta(struct iwl_mvm *mvm,
 
 	ctxt_sta->listen_interval = cpu_to_le32(mvm->hw->conf.listen_interval);
 	ctxt_sta->assoc_id = cpu_to_le32(vif->bss_conf.aid);
-}
-
-static int iwl_mvm_mac_ctxt_cmd_station(struct iwl_mvm *mvm,
-					struct ieee80211_vif *vif,
-					u32 action)
-{
-	struct iwl_mac_ctx_cmd cmd = {};
-
-	WARN_ON(vif->type != NL80211_IFTYPE_STATION || vif->p2p);
-
-	/* Fill the common data for all mac context types */
-	iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
-
-	/* Allow beacons to pass through as long as we are not associated,or we
-	 * do not have dtim period information */
-	if (!vif->bss_conf.assoc || !vif->bss_conf.dtim_period)
-		cmd.filter_flags |= cpu_to_le32(MAC_FILTER_IN_BEACON);
-	else
-		cmd.filter_flags &= ~cpu_to_le32(MAC_FILTER_IN_BEACON);
-
-	/* Fill the data specific for station mode */
-	iwl_mvm_mac_ctxt_cmd_fill_sta(mvm, vif, &cmd.sta,
-				      action == FW_CTXT_ACTION_ADD);
-
-	return iwl_mvm_mac_ctxt_send_cmd(mvm, &cmd);
-}
-
-static int iwl_mvm_mac_ctxt_cmd_p2p_client(struct iwl_mvm *mvm,
-					   struct ieee80211_vif *vif,
-					   u32 action)
-{
-	struct iwl_mac_ctx_cmd cmd = {};
-	struct ieee80211_p2p_noa_attr *noa = &vif->bss_conf.p2p_noa_attr;
-
-	WARN_ON(vif->type != NL80211_IFTYPE_STATION || !vif->p2p);
-
-	/* Fill the common data for all mac context types */
-	iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
-
-	/* Fill the data specific for station mode */
-	iwl_mvm_mac_ctxt_cmd_fill_sta(mvm, vif, &cmd.p2p_sta.sta,
-				      action == FW_CTXT_ACTION_ADD);
-
-	cmd.p2p_sta.ctwin = cpu_to_le32(noa->oppps_ctwindow &
-					IEEE80211_P2P_OPPPS_CTWINDOW_MASK);
 
 	return iwl_mvm_mac_ctxt_send_cmd(mvm, &cmd);
 }
@@ -1137,16 +1109,12 @@ static int iwl_mvm_mac_ctxt_cmd_go(struct iwl_mvm *mvm,
 }
 
 static int iwl_mvm_mac_ctx_send(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-				u32 action)
+				u32 action, bool force_assoc_off)
 {
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
-		if (!vif->p2p)
-			return iwl_mvm_mac_ctxt_cmd_station(mvm, vif,
-							    action);
-		else
-			return iwl_mvm_mac_ctxt_cmd_p2p_client(mvm, vif,
-							       action);
+		return iwl_mvm_mac_ctxt_cmd_sta(mvm, vif, action,
+						force_assoc_off);
 		break;
 	case NL80211_IFTYPE_AP:
 		if (!vif->p2p)
@@ -1176,7 +1144,8 @@ int iwl_mvm_mac_ctxt_add(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		      vif->addr, ieee80211_vif_type_p2p(vif)))
 		return -EIO;
 
-	ret = iwl_mvm_mac_ctx_send(mvm, vif, FW_CTXT_ACTION_ADD);
+	ret = iwl_mvm_mac_ctx_send(mvm, vif, FW_CTXT_ACTION_ADD,
+				   true);
 	if (ret)
 		return ret;
 
@@ -1187,7 +1156,8 @@ int iwl_mvm_mac_ctxt_add(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	return 0;
 }
 
-int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     bool force_assoc_off)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
@@ -1195,7 +1165,8 @@ int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		      vif->addr, ieee80211_vif_type_p2p(vif)))
 		return -EIO;
 
-	return iwl_mvm_mac_ctx_send(mvm, vif, FW_CTXT_ACTION_MODIFY);
+	return iwl_mvm_mac_ctx_send(mvm, vif, FW_CTXT_ACTION_MODIFY,
+				    force_assoc_off);
 }
 
 int iwl_mvm_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
@@ -1214,7 +1185,7 @@ int iwl_mvm_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 							   mvmvif->color));
 	cmd.action = cpu_to_le32(FW_CTXT_ACTION_REMOVE);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, MAC_CONTEXT_CMD, CMD_SYNC,
+	ret = iwl_mvm_send_cmd_pdu(mvm, MAC_CONTEXT_CMD, 0,
 				   sizeof(cmd), &cmd);
 	if (ret) {
 		IWL_ERR(mvm, "Failed to remove MAC context: %d\n", ret);
@@ -1240,11 +1211,23 @@ int iwl_mvm_rx_beacon_notif(struct iwl_mvm *mvm,
 	u32 rate __maybe_unused =
 		le32_to_cpu(beacon->beacon_notify_hdr.initial_rate);
 
+	lockdep_assert_held(&mvm->mutex);
+
 	IWL_DEBUG_RX(mvm, "beacon status %#x retries:%d tsf:0x%16llX rate:%d\n",
 		     status & TX_STATUS_MSK,
 		     beacon->beacon_notify_hdr.failure_frame,
 		     le64_to_cpu(beacon->tsf),
 		     rate);
+
+	if (unlikely(mvm->csa_vif && mvm->csa_vif->csa_active)) {
+		if (!ieee80211_csa_is_complete(mvm->csa_vif)) {
+			iwl_mvm_mac_ctxt_beacon_changed(mvm, mvm->csa_vif);
+		} else {
+			ieee80211_csa_finish(mvm->csa_vif);
+			mvm->csa_vif = NULL;
+		}
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/iwlwifi/mvm/mac80211.c
index 8735ef1f44ae..7215f5980186 100644
--- a/drivers/net/wireless/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/iwlwifi/mvm/mac80211.c
@@ -295,7 +295,9 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	    !iwlwifi_mod_params.sw_crypto)
 		hw->flags |= IEEE80211_HW_MFP_CAPABLE;
 
-	if (0 && mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_UAPSD_SUPPORT) {
+	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_UAPSD_SUPPORT &&
+	    IWL_UCODE_API(mvm->fw->ucode_ver) >= 9 &&
+	    !iwlwifi_mod_params.uapsd_disable) {
 		hw->flags |= IEEE80211_HW_SUPPORTS_UAPSD;
 		hw->uapsd_queues = IWL_UAPSD_AC_INFO;
 		hw->uapsd_max_sp_len = IWL_UAPSD_MAX_SP;
@@ -309,11 +311,8 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 		BIT(NL80211_IFTYPE_P2P_CLIENT) |
 		BIT(NL80211_IFTYPE_AP) |
 		BIT(NL80211_IFTYPE_P2P_GO) |
-		BIT(NL80211_IFTYPE_P2P_DEVICE);
-
-	/* IBSS has bugs in older versions */
-	if (IWL_UCODE_API(mvm->fw->ucode_ver) >= 8)
-		hw->wiphy->interface_modes |= BIT(NL80211_IFTYPE_ADHOC);
+		BIT(NL80211_IFTYPE_P2P_DEVICE) |
+		BIT(NL80211_IFTYPE_ADHOC);
 
 	hw->wiphy->flags |= WIPHY_FLAG_IBSS_RSN;
 	hw->wiphy->regulatory_flags |= REGULATORY_CUSTOM_REG |
@@ -322,6 +321,9 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_GO_UAPSD)
 		hw->wiphy->flags |= WIPHY_FLAG_AP_UAPSD;
 
+	if (mvm->fw->ucode_capa.api[0] & IWL_UCODE_TLV_API_CSA_FLOW)
+		hw->wiphy->flags |= WIPHY_FLAG_HAS_CHANNEL_SWITCH;
+
 	hw->wiphy->iface_combinations = iwl_mvm_iface_combinations;
 	hw->wiphy->n_iface_combinations =
 		ARRAY_SIZE(iwl_mvm_iface_combinations);
@@ -365,14 +367,11 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	else
 		hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
 
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_SCHED_SCAN) {
-		hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_SCHED_SCAN;
-		hw->wiphy->max_sched_scan_ssids = PROBE_OPTION_MAX;
-		hw->wiphy->max_match_sets = IWL_SCAN_MAX_PROFILES;
-		/* we create the 802.11 header and zero length SSID IE. */
-		hw->wiphy->max_sched_scan_ie_len =
-					SCAN_OFFLOAD_PROBE_REQ_SIZE - 24 - 2;
-	}
+	hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_SCHED_SCAN;
+	hw->wiphy->max_sched_scan_ssids = PROBE_OPTION_MAX;
+	hw->wiphy->max_match_sets = IWL_SCAN_MAX_PROFILES;
+	/* we create the 802.11 header and zero length SSID IE. */
+	hw->wiphy->max_sched_scan_ie_len = SCAN_OFFLOAD_PROBE_REQ_SIZE - 24 - 2;
 
 	hw->wiphy->features |= NL80211_FEATURE_P2P_GO_CTWIN |
 			       NL80211_FEATURE_P2P_GO_OPPPS;
@@ -386,7 +385,11 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	}
 
 #ifdef CONFIG_PM_SLEEP
-	if (mvm->fw->img[IWL_UCODE_WOWLAN].sec[0].len &&
+	if (iwl_mvm_is_d0i3_supported(mvm) &&
+	    device_can_wakeup(mvm->trans->dev)) {
+		mvm->wowlan.flags = WIPHY_WOWLAN_ANY;
+		hw->wiphy->wowlan = &mvm->wowlan;
+	} else if (mvm->fw->img[IWL_UCODE_WOWLAN].sec[0].len &&
 	    mvm->trans->ops->d3_suspend &&
 	    mvm->trans->ops->d3_resume &&
 	    device_can_wakeup(mvm->trans->dev)) {
@@ -540,13 +543,22 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 		return -EACCES;
 
 	/* return from D0i3 before starting a new Tx aggregation */
-	if (action == IEEE80211_AMPDU_TX_START) {
+	switch (action) {
+	case IEEE80211_AMPDU_TX_START:
+	case IEEE80211_AMPDU_TX_STOP_CONT:
+	case IEEE80211_AMPDU_TX_STOP_FLUSH:
+	case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
+	case IEEE80211_AMPDU_TX_OPERATIONAL:
 		iwl_mvm_ref(mvm, IWL_MVM_REF_TX_AGG);
 		tx_agg_ref = true;
 
 		/*
-		 * wait synchronously until D0i3 exit to get the correct
-		 * sequence number for the tid
+		 * for tx start, wait synchronously until D0i3 exit to
+		 * get the correct sequence number for the tid.
+		 * additionally, some other ampdu actions use direct
+		 * target access, which is not handled automatically
+		 * by the trans layer (unlike commands), so wait for
+		 * d0i3 exit in these cases as well.
 		 */
 		if (!wait_event_timeout(mvm->d0i3_exit_waitq,
 			  !test_bit(IWL_MVM_STATUS_IN_D0I3, &mvm->status), HZ)) {
@@ -554,6 +566,9 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 			iwl_mvm_unref(mvm, IWL_MVM_REF_TX_AGG);
 			return -EIO;
 		}
+		break;
+	default:
+		break;
 	}
 
 	mutex_lock(&mvm->mutex);
@@ -758,7 +773,7 @@ static int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		.pwr_restriction = cpu_to_le16(tx_power),
 	};
 
-	return iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, 0,
 				    sizeof(reduce_txpwr_cmd),
 				    &reduce_txpwr_cmd);
 }
@@ -817,18 +832,17 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 	if (ret)
 		goto out_release;
 
-	ret = iwl_mvm_power_update_mac(mvm, vif);
+	ret = iwl_mvm_power_update_mac(mvm);
 	if (ret)
 		goto out_release;
 
 	/* beacon filtering */
-	ret = iwl_mvm_disable_beacon_filter(mvm, vif, CMD_SYNC);
+	ret = iwl_mvm_disable_beacon_filter(mvm, vif, 0);
 	if (ret)
 		goto out_remove_mac;
 
-	if (!mvm->bf_allowed_vif && false &&
-	    vif->type == NL80211_IFTYPE_STATION && !vif->p2p &&
-	    mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BF_UPDATED){
+	if (!mvm->bf_allowed_vif &&
+	    vif->type == NL80211_IFTYPE_STATION && !vif->p2p) {
 		mvm->bf_allowed_vif = mvmvif;
 		vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER |
 				     IEEE80211_VIF_SUPPORTS_CQM_RSSI;
@@ -969,7 +983,7 @@ static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
 	if (mvm->vif_count && vif->type != NL80211_IFTYPE_P2P_DEVICE)
 		mvm->vif_count--;
 
-	iwl_mvm_power_update_mac(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
 	iwl_mvm_mac_ctxt_remove(mvm, vif);
 
 out_release:
@@ -1223,10 +1237,14 @@ static int iwl_mvm_configure_bcast_filter(struct iwl_mvm *mvm,
 	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BCAST_FILTERING))
 		return 0;
 
+	/* bcast filtering isn't supported for P2P client */
+	if (vif->p2p)
+		return 0;
+
 	if (!iwl_mvm_bcast_filter_build_cmd(mvm, &cmd))
 		return 0;
 
-	return iwl_mvm_send_cmd_pdu(mvm, BCAST_FILTER_CMD, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, BCAST_FILTER_CMD, 0,
 				    sizeof(cmd), &cmd);
 }
 #else
@@ -1253,7 +1271,7 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 	if (changes & BSS_CHANGED_ASSOC && bss_conf->assoc)
 		iwl_mvm_mac_ctxt_recalc_tsf_id(mvm, vif);
 
-	ret = iwl_mvm_mac_ctxt_changed(mvm, vif);
+	ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false);
 	if (ret)
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
 
@@ -1333,10 +1351,10 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 		iwl_mvm_remove_time_event(mvm, mvmvif,
 					  &mvmvif->time_event_data);
 		iwl_mvm_sf_update(mvm, vif, false);
-		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, CMD_SYNC));
+		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
 	} else if (changes & (BSS_CHANGED_PS | BSS_CHANGED_P2P_PS |
 			      BSS_CHANGED_QOS)) {
-		ret = iwl_mvm_power_update_mac(mvm, vif);
+		ret = iwl_mvm_power_update_mac(mvm);
 		if (ret)
 			IWL_ERR(mvm, "failed to update power mode\n");
 	}
@@ -1347,16 +1365,19 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 	}
 
 	if (changes & BSS_CHANGED_CQM) {
-		IWL_DEBUG_MAC80211(mvm, "cqm info_changed");
+		IWL_DEBUG_MAC80211(mvm, "cqm info_changed\n");
 		/* reset cqm events tracking */
 		mvmvif->bf_data.last_cqm_event = 0;
-		ret = iwl_mvm_update_beacon_filter(mvm, vif, false, CMD_SYNC);
-		if (ret)
-			IWL_ERR(mvm, "failed to update CQM thresholds\n");
+		if (mvmvif->bf_data.bf_enabled) {
+			ret = iwl_mvm_enable_beacon_filter(mvm, vif, 0);
+			if (ret)
+				IWL_ERR(mvm,
+					"failed to update CQM thresholds\n");
+		}
 	}
 
 	if (changes & BSS_CHANGED_ARP_FILTER) {
-		IWL_DEBUG_MAC80211(mvm, "arp filter changed");
+		IWL_DEBUG_MAC80211(mvm, "arp filter changed\n");
 		iwl_mvm_configure_bcast_filter(mvm, vif);
 	}
 }
@@ -1402,7 +1423,7 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 	mvmvif->ap_ibss_active = true;
 
 	/* power updated needs to be done before quotas */
-	iwl_mvm_power_update_mac(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
 
 	ret = iwl_mvm_update_quotas(mvm, vif);
 	if (ret)
@@ -1410,7 +1431,7 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 
 	/* Need to update the P2P Device MAC (only GO, IBSS is single vif) */
 	if (vif->p2p && mvm->p2p_device_vif)
-		iwl_mvm_mac_ctxt_changed(mvm, mvm->p2p_device_vif);
+		iwl_mvm_mac_ctxt_changed(mvm, mvm->p2p_device_vif, false);
 
 	iwl_mvm_ref(mvm, IWL_MVM_REF_AP_IBSS);
 
@@ -1420,7 +1441,7 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 	return 0;
 
 out_quota_failed:
-	iwl_mvm_power_update_mac(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
 	mvmvif->ap_ibss_active = false;
 	iwl_mvm_send_rm_bcast_sta(mvm, &mvmvif->bcast_sta);
 out_unbind:
@@ -1450,13 +1471,13 @@ static void iwl_mvm_stop_ap_ibss(struct ieee80211_hw *hw,
 
 	/* Need to update the P2P Device MAC (only GO, IBSS is single vif) */
 	if (vif->p2p && mvm->p2p_device_vif)
-		iwl_mvm_mac_ctxt_changed(mvm, mvm->p2p_device_vif);
+		iwl_mvm_mac_ctxt_changed(mvm, mvm->p2p_device_vif, false);
 
 	iwl_mvm_update_quotas(mvm, NULL);
 	iwl_mvm_send_rm_bcast_sta(mvm, &mvmvif->bcast_sta);
 	iwl_mvm_binding_remove_vif(mvm, vif);
 
-	iwl_mvm_power_update_mac(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
 
 	iwl_mvm_mac_ctxt_remove(mvm, vif);
 
@@ -1477,7 +1498,7 @@ iwl_mvm_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 
 	if (changes & (BSS_CHANGED_ERP_CTS_PROT | BSS_CHANGED_HT |
 		       BSS_CHANGED_BANDWIDTH) &&
-	    iwl_mvm_mac_ctxt_changed(mvm, vif))
+	    iwl_mvm_mac_ctxt_changed(mvm, vif, false))
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
 
 	/* Need to send a new beacon template to the FW */
@@ -1495,6 +1516,9 @@ static void iwl_mvm_bss_info_changed(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
+	if (changes & BSS_CHANGED_IDLE && !bss_conf->idle)
+		iwl_mvm_sched_scan_stop(mvm, true);
+
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
 		iwl_mvm_bss_info_changed_station(mvm, vif, bss_conf, changes);
@@ -1525,7 +1549,7 @@ static int iwl_mvm_mac_hw_scan(struct ieee80211_hw *hw,
 
 	switch (mvm->scan_status) {
 	case IWL_MVM_SCAN_SCHED:
-		ret = iwl_mvm_sched_scan_stop(mvm);
+		ret = iwl_mvm_sched_scan_stop(mvm, true);
 		if (ret) {
 			ret = -EBUSY;
 			goto out;
@@ -1697,6 +1721,11 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 		ret = iwl_mvm_add_sta(mvm, vif, sta);
 	} else if (old_state == IEEE80211_STA_NONE &&
 		   new_state == IEEE80211_STA_AUTH) {
+		/*
+		 * EBS may be disabled due to previous failures reported by FW.
+		 * Reset EBS status here assuming environment has been changed.
+		 */
+		mvm->last_ebs_successful = true;
 		ret = 0;
 	} else if (old_state == IEEE80211_STA_AUTH &&
 		   new_state == IEEE80211_STA_ASSOC) {
@@ -1708,14 +1737,12 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTHORIZED) {
 		/* enable beacon filtering */
-		if (vif->bss_conf.dtim_period)
-			WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif,
-							     CMD_SYNC));
+		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
 		ret = 0;
 	} else if (old_state == IEEE80211_STA_AUTHORIZED &&
 		   new_state == IEEE80211_STA_ASSOC) {
 		/* disable beacon filtering */
-		WARN_ON(iwl_mvm_disable_beacon_filter(mvm, vif, CMD_SYNC));
+		WARN_ON(iwl_mvm_disable_beacon_filter(mvm, vif, 0));
 		ret = 0;
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTH) {
@@ -1772,7 +1799,7 @@ static int iwl_mvm_mac_conf_tx(struct ieee80211_hw *hw,
 		int ret;
 
 		mutex_lock(&mvm->mutex);
-		ret = iwl_mvm_mac_ctxt_changed(mvm, vif);
+		ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false);
 		mutex_unlock(&mvm->mutex);
 		return ret;
 	}
@@ -1865,7 +1892,7 @@ static int iwl_mvm_mac_sched_scan_stop(struct ieee80211_hw *hw,
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_sched_scan_stop(mvm);
+	ret = iwl_mvm_sched_scan_stop(mvm, false);
 	mutex_unlock(&mvm->mutex);
 	iwl_mvm_wait_for_async_handlers(mvm);
 
@@ -2161,10 +2188,10 @@ static void iwl_mvm_change_chanctx(struct ieee80211_hw *hw,
 		return;
 
 	mutex_lock(&mvm->mutex);
+	iwl_mvm_bt_coex_vif_change(mvm);
 	iwl_mvm_phy_ctxt_changed(mvm, phy_ctxt, &ctx->min_def,
 				 ctx->rx_chains_static,
 				 ctx->rx_chains_dynamic);
-	iwl_mvm_bt_coex_vif_change(mvm);
 	mutex_unlock(&mvm->mutex);
 }
 
@@ -2184,6 +2211,11 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_AP:
+		/* Unless it's a CSA flow we have nothing to do here */
+		if (vif->csa_active) {
+			mvmvif->ap_ibss_active = true;
+			break;
+		}
 	case NL80211_IFTYPE_ADHOC:
 		/*
 		 * The AP binding flow is handled as part of the start_ap flow
@@ -2207,7 +2239,7 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 	 * Power state must be updated before quotas,
 	 * otherwise fw will complain.
 	 */
-	iwl_mvm_power_update_mac(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
 
 	/* Setting the quota at this stage is only required for monitor
 	 * interfaces. For the other types, the bss_info changed flow
@@ -2220,11 +2252,17 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 			goto out_remove_binding;
 	}
 
+	/* Handle binding during CSA */
+	if (vif->type == NL80211_IFTYPE_AP) {
+		iwl_mvm_update_quotas(mvm, vif);
+		iwl_mvm_mac_ctxt_changed(mvm, vif, false);
+	}
+
 	goto out_unlock;
 
  out_remove_binding:
 	iwl_mvm_binding_remove_vif(mvm, vif);
-	iwl_mvm_power_update_mac(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
  out_unlock:
 	mutex_unlock(&mvm->mutex);
 	if (ret)
@@ -2244,22 +2282,29 @@ static void iwl_mvm_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	iwl_mvm_remove_time_event(mvm, mvmvif, &mvmvif->time_event_data);
 
 	switch (vif->type) {
-	case NL80211_IFTYPE_AP:
 	case NL80211_IFTYPE_ADHOC:
 		goto out_unlock;
 	case NL80211_IFTYPE_MONITOR:
 		mvmvif->monitor_active = false;
 		iwl_mvm_update_quotas(mvm, NULL);
 		break;
+	case NL80211_IFTYPE_AP:
+		/* This part is triggered only during CSA */
+		if (!vif->csa_active || !mvmvif->ap_ibss_active)
+			goto out_unlock;
+
+		mvmvif->ap_ibss_active = false;
+		iwl_mvm_update_quotas(mvm, NULL);
+		/*TODO: bt_coex notification here? */
 	default:
 		break;
 	}
 
 	iwl_mvm_binding_remove_vif(mvm, vif);
-	iwl_mvm_power_update_mac(mvm, vif);
 
 out_unlock:
 	mvmvif->phy_ctxt = NULL;
+	iwl_mvm_power_update_mac(mvm);
 	mutex_unlock(&mvm->mutex);
 }
 
@@ -2323,9 +2368,8 @@ static int __iwl_mvm_mac_testmode_cmd(struct iwl_mvm *mvm,
 			return -EINVAL;
 
 		if (nla_get_u32(tb[IWL_MVM_TM_ATTR_BEACON_FILTER_STATE]))
-			return iwl_mvm_enable_beacon_filter(mvm, vif,
-							    CMD_SYNC);
-		return iwl_mvm_disable_beacon_filter(mvm, vif, CMD_SYNC);
+			return iwl_mvm_enable_beacon_filter(mvm, vif, 0);
+		return iwl_mvm_disable_beacon_filter(mvm, vif, 0);
 	}
 
 	return -EOPNOTSUPP;
@@ -2346,6 +2390,53 @@ static int iwl_mvm_mac_testmode_cmd(struct ieee80211_hw *hw,
 }
 #endif
 
+static void iwl_mvm_channel_switch_beacon(struct ieee80211_hw *hw,
+					  struct ieee80211_vif *vif,
+					  struct cfg80211_chan_def *chandef)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	mutex_lock(&mvm->mutex);
+	if (WARN(mvm->csa_vif && mvm->csa_vif->csa_active,
+		 "Another CSA is already in progress"))
+		goto out_unlock;
+
+	IWL_DEBUG_MAC80211(mvm, "CSA started to freq %d\n",
+			   chandef->center_freq1);
+	mvm->csa_vif = vif;
+
+out_unlock:
+	mutex_unlock(&mvm->mutex);
+}
+
+static void iwl_mvm_mac_flush(struct ieee80211_hw *hw,
+			      struct ieee80211_vif *vif, u32 queues, bool drop)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif;
+	struct iwl_mvm_sta *mvmsta;
+
+	if (!vif || vif->type != NL80211_IFTYPE_STATION)
+		return;
+
+	mutex_lock(&mvm->mutex);
+	mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	mvmsta = iwl_mvm_sta_from_staid_protected(mvm, mvmvif->ap_sta_id);
+
+	if (WARN_ON_ONCE(!mvmsta))
+		goto done;
+
+	if (drop) {
+		if (iwl_mvm_flush_tx_path(mvm, mvmsta->tfd_queue_msk, true))
+			IWL_ERR(mvm, "flush request fail\n");
+	} else {
+		iwl_trans_wait_tx_queue_empty(mvm->trans,
+					      mvmsta->tfd_queue_msk);
+	}
+done:
+	mutex_unlock(&mvm->mutex);
+}
+
 const struct ieee80211_ops iwl_mvm_hw_ops = {
 	.tx = iwl_mvm_mac_tx,
 	.ampdu_action = iwl_mvm_mac_ampdu_action,
@@ -2369,6 +2460,7 @@ const struct ieee80211_ops iwl_mvm_hw_ops = {
 	.sta_rc_update = iwl_mvm_sta_rc_update,
 	.conf_tx = iwl_mvm_mac_conf_tx,
 	.mgd_prepare_tx = iwl_mvm_mac_mgd_prepare_tx,
+	.flush = iwl_mvm_mac_flush,
 	.sched_scan_start = iwl_mvm_mac_sched_scan_start,
 	.sched_scan_stop = iwl_mvm_mac_sched_scan_stop,
 	.set_key = iwl_mvm_mac_set_key,
@@ -2388,6 +2480,8 @@ const struct ieee80211_ops iwl_mvm_hw_ops = {
 
 	.set_tim = iwl_mvm_set_tim,
 
+	.channel_switch_beacon = iwl_mvm_channel_switch_beacon,
+
 	CFG80211_TESTMODE_CMD(iwl_mvm_mac_testmode_cmd)
 
 #ifdef CONFIG_PM_SLEEP
diff --git a/drivers/net/wireless/iwlwifi/mvm/mvm.h b/drivers/net/wireless/iwlwifi/mvm/mvm.h
index f1ec0986c3c9..fcc6c29482d0 100644
--- a/drivers/net/wireless/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/iwlwifi/mvm/mvm.h
@@ -164,7 +164,6 @@ enum iwl_dbgfs_pm_mask {
 	MVM_DEBUGFS_PM_SKIP_DTIM_PERIODS = BIT(2),
 	MVM_DEBUGFS_PM_RX_DATA_TIMEOUT = BIT(3),
 	MVM_DEBUGFS_PM_TX_DATA_TIMEOUT = BIT(4),
-	MVM_DEBUGFS_PM_DISABLE_POWER_OFF = BIT(5),
 	MVM_DEBUGFS_PM_LPRX_ENA = BIT(6),
 	MVM_DEBUGFS_PM_LPRX_RSSI_THRESHOLD = BIT(7),
 	MVM_DEBUGFS_PM_SNOOZE_ENABLE = BIT(8),
@@ -177,7 +176,6 @@ struct iwl_dbgfs_pm {
 	u32 tx_data_timeout;
 	bool skip_over_dtim;
 	u8 skip_dtim_periods;
-	bool disable_power_off;
 	bool lprx_ena;
 	u32 lprx_rssi_threshold;
 	bool snooze_ena;
@@ -232,6 +230,7 @@ enum iwl_mvm_ref_type {
 	IWL_MVM_REF_USER,
 	IWL_MVM_REF_TX,
 	IWL_MVM_REF_TX_AGG,
+	IWL_MVM_REF_EXIT_WORK,
 
 	IWL_MVM_REF_COUNT,
 };
@@ -265,6 +264,7 @@ struct iwl_mvm_vif_bf_data {
  * @uploaded: indicates the MAC context has been added to the device
  * @ap_ibss_active: indicates that AP/IBSS is configured and that the interface
  *	should get quota etc.
+ * @pm_enabled - Indicate if MAC power management is allowed
  * @monitor_active: indicates that monitor context is configured, and that the
  *	interface should get quota etc.
  * @low_latency: indicates that this interface is in low-latency mode
@@ -283,6 +283,7 @@ struct iwl_mvm_vif {
 
 	bool uploaded;
 	bool ap_ibss_active;
+	bool pm_enabled;
 	bool monitor_active;
 	bool low_latency;
 	struct iwl_mvm_vif_bf_data bf_data;
@@ -451,6 +452,11 @@ struct iwl_mvm_frame_stats {
 	int last_frame_idx;
 };
 
+enum {
+	D0I3_DEFER_WAKEUP,
+	D0I3_PENDING_WAKEUP,
+};
+
 struct iwl_mvm {
 	/* for logger access */
 	struct device *dev;
@@ -484,6 +490,7 @@ struct iwl_mvm {
 	u32 log_event_table;
 	u32 umac_error_event_table;
 	bool support_umac_log;
+	struct iwl_sf_region sf_space;
 
 	u32 ampdu_ref;
 
@@ -495,6 +502,7 @@ struct iwl_mvm {
 	u8 queue_to_mac80211[IWL_MAX_HW_QUEUES];
 	atomic_t queue_stop_count[IWL_MAX_HW_QUEUES];
 
+	const char *nvm_file_name;
 	struct iwl_nvm_data *nvm_data;
 	/* NVM sections */
 	struct iwl_nvm_section nvm_sections[NVM_MAX_NUM_SECTIONS];
@@ -535,6 +543,8 @@ struct iwl_mvm {
 	/* Internal station */
 	struct iwl_mvm_int_sta aux_sta;
 
+	bool last_ebs_successful;
+
 	u8 scan_last_antenna_idx; /* to toggle TX between antennas */
 	u8 mgmt_last_antenna_idx;
 
@@ -578,8 +588,12 @@ struct iwl_mvm {
 	void *fw_error_dump;
 	void *fw_error_sram;
 	u32 fw_error_sram_len;
+	u32 *fw_error_rxf;
+	u32 fw_error_rxf_len;
 
+#ifdef CONFIG_IWLWIFI_LEDS
 	struct led_classdev led;
+#endif
 
 	struct ieee80211_vif *p2p_device_vif;
 
@@ -601,6 +615,9 @@ struct iwl_mvm {
 	bool d0i3_offloading;
 	struct work_struct d0i3_exit_work;
 	struct sk_buff_head d0i3_tx;
+	/* protect d0i3_suspend_flags */
+	struct mutex d0i3_suspend_mutex;
+	unsigned long d0i3_suspend_flags;
 	/* sync d0i3_tx queue and IWL_MVM_STATUS_IN_D0I3 status flag */
 	spinlock_t d0i3_tx_lock;
 	wait_queue_head_t d0i3_exit_waitq;
@@ -629,8 +646,8 @@ struct iwl_mvm {
 
 	/* Indicate if device power save is allowed */
 	bool ps_disabled;
-	/* Indicate if device power management is allowed */
-	bool pm_disabled;
+
+	struct ieee80211_vif *csa_vif;
 };
 
 /* Extract MVM priv from op_mode and _hw */
@@ -705,6 +722,7 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm);
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm);
 void iwl_mvm_fw_error_sram_dump(struct iwl_mvm *mvm);
+void iwl_mvm_fw_error_rxf_dump(struct iwl_mvm *mvm);
 #endif
 u8 first_antenna(u8 mask);
 u8 iwl_mvm_next_antenna(struct iwl_mvm *mvm, u8 valid, u8 last_idx);
@@ -745,7 +763,7 @@ int iwl_mvm_rx_statistics(struct iwl_mvm *mvm,
 			  struct iwl_device_cmd *cmd);
 
 /* NVM */
-int iwl_nvm_init(struct iwl_mvm *mvm);
+int iwl_nvm_init(struct iwl_mvm *mvm, bool read_nvm_from_nic);
 int iwl_mvm_load_nvm_to_nic(struct iwl_mvm *mvm);
 
 int iwl_mvm_up(struct iwl_mvm *mvm);
@@ -796,7 +814,8 @@ void iwl_mvm_phy_ctxt_unref(struct iwl_mvm *mvm,
 int iwl_mvm_mac_ctxt_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 void iwl_mvm_mac_ctxt_release(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mac_ctxt_add(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
-int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     bool force_assoc_off);
 int iwl_mvm_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 u32 iwl_mvm_mac_get_queues_mask(struct iwl_mvm *mvm,
 				struct ieee80211_vif *vif);
@@ -840,7 +859,7 @@ int iwl_mvm_config_sched_scan_profiles(struct iwl_mvm *mvm,
 				       struct cfg80211_sched_scan_request *req);
 int iwl_mvm_sched_scan_start(struct iwl_mvm *mvm,
 			     struct cfg80211_sched_scan_request *req);
-int iwl_mvm_sched_scan_stop(struct iwl_mvm *mvm);
+int iwl_mvm_sched_scan_stop(struct iwl_mvm *mvm, bool notify);
 int iwl_mvm_rx_sched_scan_results(struct iwl_mvm *mvm,
 				  struct iwl_rx_cmd_buffer *rxb,
 				  struct iwl_device_cmd *cmd);
@@ -874,10 +893,8 @@ void iwl_mvm_update_frame_stats(struct iwl_mvm *mvm,
 int rs_pretty_print_rate(char *buf, const u32 rate);
 
 /* power management */
-int iwl_power_legacy_set_cam_mode(struct iwl_mvm *mvm);
-
 int iwl_mvm_power_update_device(struct iwl_mvm *mvm);
-int iwl_mvm_power_update_mac(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_power_update_mac(struct iwl_mvm *mvm);
 int iwl_mvm_power_mac_dbgfs_read(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				 char *buf, int bufsz);
 
@@ -886,8 +903,18 @@ int iwl_mvm_power_uapsd_misbehaving_ap_notif(struct iwl_mvm *mvm,
 					     struct iwl_rx_cmd_buffer *rxb,
 					     struct iwl_device_cmd *cmd);
 
+#ifdef CONFIG_IWLWIFI_LEDS
 int iwl_mvm_leds_init(struct iwl_mvm *mvm);
 void iwl_mvm_leds_exit(struct iwl_mvm *mvm);
+#else
+static inline int iwl_mvm_leds_init(struct iwl_mvm *mvm)
+{
+	return 0;
+}
+static inline void iwl_mvm_leds_exit(struct iwl_mvm *mvm)
+{
+}
+#endif
 
 /* D3 (WoWLAN, NetDetect) */
 int iwl_mvm_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan);
@@ -922,9 +949,9 @@ int iwl_mvm_send_proto_offload(struct iwl_mvm *mvm,
 void iwl_mvm_ref(struct iwl_mvm *mvm, enum iwl_mvm_ref_type ref_type);
 void iwl_mvm_unref(struct iwl_mvm *mvm, enum iwl_mvm_ref_type ref_type);
 void iwl_mvm_d0i3_enable_tx(struct iwl_mvm *mvm, __le16 *qos_seq);
+int _iwl_mvm_exit_d0i3(struct iwl_mvm *mvm);
 
 /* BT Coex */
-int iwl_send_bt_prio_tbl(struct iwl_mvm *mvm);
 int iwl_send_bt_init_conf(struct iwl_mvm *mvm);
 int iwl_mvm_rx_bt_coex_notif(struct iwl_mvm *mvm,
 			     struct iwl_rx_cmd_buffer *rxb,
@@ -936,9 +963,10 @@ u16 iwl_mvm_coex_agg_time_limit(struct iwl_mvm *mvm,
 				struct ieee80211_sta *sta);
 bool iwl_mvm_bt_coex_is_mimo_allowed(struct iwl_mvm *mvm,
 				     struct ieee80211_sta *sta);
+bool iwl_mvm_bt_coex_is_tpc_allowed(struct iwl_mvm *mvm,
+				    enum ieee80211_band band);
 u8 iwl_mvm_bt_coex_tx_prio(struct iwl_mvm *mvm, struct ieee80211_hdr *hdr,
 			   struct ieee80211_tx_info *info, u8 ac);
-int iwl_mvm_bt_coex_reduced_txp(struct iwl_mvm *mvm, u8 sta_id, bool enable);
 
 enum iwl_bt_kill_msk {
 	BT_KILL_MSK_DEFAULT,
@@ -969,17 +997,11 @@ int iwl_mvm_enable_beacon_filter(struct iwl_mvm *mvm,
 int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 				  struct ieee80211_vif *vif,
 				  u32 flags);
-int iwl_mvm_update_beacon_abort(struct iwl_mvm *mvm,
-				struct ieee80211_vif *vif, bool enable);
-int iwl_mvm_update_beacon_filter(struct iwl_mvm *mvm,
-				 struct ieee80211_vif *vif,
-				 bool force,
-				 u32 flags);
-
 /* SMPS */
 void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				enum iwl_mvm_smps_type_request req_type,
 				enum ieee80211_smps_mode smps_request);
+bool iwl_mvm_rx_diversity_allowed(struct iwl_mvm *mvm);
 
 /* Low latency */
 int iwl_mvm_update_low_latency(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/iwlwifi/mvm/nvm.c b/drivers/net/wireless/iwlwifi/mvm/nvm.c
index cf2d09f53782..808f78f6fbf9 100644
--- a/drivers/net/wireless/iwlwifi/mvm/nvm.c
+++ b/drivers/net/wireless/iwlwifi/mvm/nvm.c
@@ -74,6 +74,12 @@
 #define NVM_WRITE_OPCODE 1
 #define NVM_READ_OPCODE 0
 
+/* load nvm chunk response */
+enum {
+	READ_NVM_CHUNK_SUCCEED = 0,
+	READ_NVM_CHUNK_NOT_VALID_ADDRESS = 1
+};
+
 /*
  * prepare the NVM host command w/ the pointers to the nvm buffer
  * and send it to fw
@@ -90,7 +96,7 @@ static int iwl_nvm_write_chunk(struct iwl_mvm *mvm, u16 section,
 	struct iwl_host_cmd cmd = {
 		.id = NVM_ACCESS_CMD,
 		.len = { sizeof(struct iwl_nvm_access_cmd), length },
-		.flags = CMD_SYNC | CMD_SEND_IN_RFKILL,
+		.flags = CMD_SEND_IN_RFKILL,
 		.data = { &nvm_access_cmd, data },
 		/* data may come from vmalloc, so use _DUP */
 		.dataflags = { 0, IWL_HCMD_DFL_DUP },
@@ -112,7 +118,7 @@ static int iwl_nvm_read_chunk(struct iwl_mvm *mvm, u16 section,
 	struct iwl_rx_packet *pkt;
 	struct iwl_host_cmd cmd = {
 		.id = NVM_ACCESS_CMD,
-		.flags = CMD_SYNC | CMD_WANT_SKB | CMD_SEND_IN_RFKILL,
+		.flags = CMD_WANT_SKB | CMD_SEND_IN_RFKILL,
 		.data = { &nvm_access_cmd, },
 	};
 	int ret, bytes_read, offset_read;
@@ -139,10 +145,26 @@ static int iwl_nvm_read_chunk(struct iwl_mvm *mvm, u16 section,
 	offset_read = le16_to_cpu(nvm_resp->offset);
 	resp_data = nvm_resp->data;
 	if (ret) {
-		IWL_ERR(mvm,
-			"NVM access command failed with status %d (device: %s)\n",
-			ret, mvm->cfg->name);
-		ret = -EINVAL;
+		if ((offset != 0) &&
+		    (ret == READ_NVM_CHUNK_NOT_VALID_ADDRESS)) {
+			/*
+			 * meaning of NOT_VALID_ADDRESS:
+			 * driver try to read chunk from address that is
+			 * multiple of 2K and got an error since addr is empty.
+			 * meaning of (offset != 0): driver already
+			 * read valid data from another chunk so this case
+			 * is not an error.
+			 */
+			IWL_DEBUG_EEPROM(mvm->trans->dev,
+					 "NVM access command failed on offset 0x%x since that section size is multiple 2K\n",
+					 offset);
+			ret = 0;
+		} else {
+			IWL_DEBUG_EEPROM(mvm->trans->dev,
+					 "NVM access command failed with status %d (device: %s)\n",
+					 ret, mvm->cfg->name);
+			ret = -EIO;
+		}
 		goto exit;
 	}
 
@@ -211,9 +233,9 @@ static int iwl_nvm_read_section(struct iwl_mvm *mvm, u16 section,
 	while (ret == length) {
 		ret = iwl_nvm_read_chunk(mvm, section, offset, length, data);
 		if (ret < 0) {
-			IWL_ERR(mvm,
-				"Cannot read NVM from section %d offset %d, length %d\n",
-				section, offset, length);
+			IWL_DEBUG_EEPROM(mvm->trans->dev,
+					 "Cannot read NVM from section %d offset %d, length %d\n",
+					 section, offset, length);
 			return ret;
 		}
 		offset += ret;
@@ -238,13 +260,20 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm)
 			return NULL;
 		}
 	} else {
+		/* SW and REGULATORY sections are mandatory */
 		if (!mvm->nvm_sections[NVM_SECTION_TYPE_SW].data ||
-		    !mvm->nvm_sections[NVM_SECTION_TYPE_MAC_OVERRIDE].data ||
 		    !mvm->nvm_sections[NVM_SECTION_TYPE_REGULATORY].data) {
 			IWL_ERR(mvm,
 				"Can't parse empty family 8000 NVM sections\n");
 			return NULL;
 		}
+		/* MAC_OVERRIDE or at least HW section must exist */
+		if (!mvm->nvm_sections[mvm->cfg->nvm_hw_section_num].data &&
+		    !mvm->nvm_sections[NVM_SECTION_TYPE_MAC_OVERRIDE].data) {
+			IWL_ERR(mvm,
+				"Can't parse mac_address, empty sections\n");
+			return NULL;
+		}
 	}
 
 	if (WARN_ON(!mvm->cfg))
@@ -311,16 +340,16 @@ static int iwl_mvm_read_external_nvm(struct iwl_mvm *mvm)
 	 * get here after that we assume the NVM request can be satisfied
 	 * synchronously.
 	 */
-	ret = request_firmware(&fw_entry, iwlwifi_mod_params.nvm_file,
+	ret = request_firmware(&fw_entry, mvm->nvm_file_name,
 			       mvm->trans->dev);
 	if (ret) {
 		IWL_ERR(mvm, "ERROR: %s isn't available %d\n",
-			iwlwifi_mod_params.nvm_file, ret);
+			mvm->nvm_file_name, ret);
 		return ret;
 	}
 
 	IWL_INFO(mvm, "Loaded NVM file %s (%zu bytes)\n",
-		 iwlwifi_mod_params.nvm_file, fw_entry->size);
+		 mvm->nvm_file_name, fw_entry->size);
 
 	if (fw_entry->size < sizeof(*file_sec)) {
 		IWL_ERR(mvm, "NVM file too small\n");
@@ -427,53 +456,28 @@ int iwl_mvm_load_nvm_to_nic(struct iwl_mvm *mvm)
 	return ret;
 }
 
-int iwl_nvm_init(struct iwl_mvm *mvm)
+int iwl_nvm_init(struct iwl_mvm *mvm, bool read_nvm_from_nic)
 {
-	int ret, i, section;
+	int ret, section;
 	u8 *nvm_buffer, *temp;
-	int nvm_to_read[NVM_MAX_NUM_SECTIONS];
-	int num_of_sections_to_read;
 
 	if (WARN_ON_ONCE(mvm->cfg->nvm_hw_section_num >= NVM_MAX_NUM_SECTIONS))
 		return -EINVAL;
 
-	/* load external NVM if configured */
-	if (iwlwifi_mod_params.nvm_file) {
-		/* move to External NVM flow */
-		ret = iwl_mvm_read_external_nvm(mvm);
-		if (ret)
-			return ret;
-	} else {
-		/* list of NVM sections we are allowed/need to read */
-		if (mvm->trans->cfg->device_family != IWL_DEVICE_FAMILY_8000) {
-			nvm_to_read[0] = mvm->cfg->nvm_hw_section_num;
-			nvm_to_read[1] = NVM_SECTION_TYPE_SW;
-			nvm_to_read[2] = NVM_SECTION_TYPE_CALIBRATION;
-			nvm_to_read[3] = NVM_SECTION_TYPE_PRODUCTION;
-			num_of_sections_to_read = 4;
-		} else {
-			nvm_to_read[0] = NVM_SECTION_TYPE_SW;
-			nvm_to_read[1] = NVM_SECTION_TYPE_CALIBRATION;
-			nvm_to_read[2] = NVM_SECTION_TYPE_PRODUCTION;
-			nvm_to_read[3] = NVM_SECTION_TYPE_REGULATORY;
-			nvm_to_read[4] = NVM_SECTION_TYPE_MAC_OVERRIDE;
-			num_of_sections_to_read = 5;
-		}
-
+	/* load NVM values from nic */
+	if (read_nvm_from_nic) {
 		/* Read From FW NVM */
 		IWL_DEBUG_EEPROM(mvm->trans->dev, "Read from NVM\n");
 
-		/* TODO: find correct NVM max size for a section */
 		nvm_buffer = kmalloc(mvm->cfg->base_params->eeprom_size,
 				     GFP_KERNEL);
 		if (!nvm_buffer)
 			return -ENOMEM;
-		for (i = 0; i < num_of_sections_to_read; i++) {
-			section = nvm_to_read[i];
+		for (section = 0; section < NVM_MAX_NUM_SECTIONS; section++) {
 			/* we override the constness for initial read */
 			ret = iwl_nvm_read_section(mvm, section, nvm_buffer);
 			if (ret < 0)
-				break;
+				continue;
 			temp = kmemdup(nvm_buffer, ret, GFP_KERNEL);
 			if (!temp) {
 				ret = -ENOMEM;
@@ -502,15 +506,21 @@ int iwl_nvm_init(struct iwl_mvm *mvm)
 					mvm->nvm_hw_blob.size = ret;
 					break;
 				}
-				WARN(1, "section: %d", section);
 			}
 #endif
 		}
 		kfree(nvm_buffer);
-		if (ret < 0)
+	}
+
+	/* load external NVM if configured */
+	if (mvm->nvm_file_name) {
+		/* move to External NVM flow */
+		ret = iwl_mvm_read_external_nvm(mvm);
+		if (ret)
 			return ret;
 	}
 
+	/* parse the relevant nvm sections */
 	mvm->nvm_data = iwl_parse_nvm_sections(mvm);
 	if (!mvm->nvm_data)
 		return -ENODATA;
diff --git a/drivers/net/wireless/iwlwifi/mvm/ops.c b/drivers/net/wireless/iwlwifi/mvm/ops.c
index 9545d7fdd4bf..cc2f7de396de 100644
--- a/drivers/net/wireless/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/iwlwifi/mvm/ops.c
@@ -79,8 +79,8 @@
 #include "iwl-prph.h"
 #include "rs.h"
 #include "fw-api-scan.h"
-#include "fw-error-dump.h"
 #include "time-event.h"
+#include "iwl-fw-error-dump.h"
 
 /*
  * module name, copyright, version, etc.
@@ -220,7 +220,7 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 	RX_HANDLER(BA_NOTIF, iwl_mvm_rx_ba_notif, false),
 
 	RX_HANDLER(BT_PROFILE_NOTIFICATION, iwl_mvm_rx_bt_coex_notif, true),
-	RX_HANDLER(BEACON_NOTIFICATION, iwl_mvm_rx_beacon_notif, false),
+	RX_HANDLER(BEACON_NOTIFICATION, iwl_mvm_rx_beacon_notif, true),
 	RX_HANDLER(STATISTICS_NOTIFICATION, iwl_mvm_rx_statistics, true),
 	RX_HANDLER(ANTENNA_COUPLING_NOTIFICATION,
 		   iwl_mvm_rx_ant_coupling_notif, true),
@@ -402,6 +402,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	mvm->sf_state = SF_UNINIT;
 
 	mutex_init(&mvm->mutex);
+	mutex_init(&mvm->d0i3_suspend_mutex);
 	spin_lock_init(&mvm->async_handlers_lock);
 	INIT_LIST_HEAD(&mvm->time_event_list);
 	INIT_LIST_HEAD(&mvm->async_handlers_list);
@@ -465,13 +466,24 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 
 	min_backoff = calc_min_backoff(trans, cfg);
 	iwl_mvm_tt_initialize(mvm, min_backoff);
+	/* set the nvm_file_name according to priority */
+	if (iwlwifi_mod_params.nvm_file)
+		mvm->nvm_file_name = iwlwifi_mod_params.nvm_file;
+	else
+		mvm->nvm_file_name = mvm->cfg->default_nvm_file;
+
+	if (WARN(cfg->no_power_up_nic_in_init && !mvm->nvm_file_name,
+		 "not allowing power-up and not having nvm_file\n"))
+		goto out_free;
 
 	/*
-	 * If the NVM exists in an external file,
-	 * there is no need to unnecessarily power up the NIC at driver load
+	 * Even if nvm exists in the nvm_file driver should read agin the nvm
+	 * from the nic because there might be entries that exist in the OTP
+	 * and not in the file.
+	 * for nics with no_power_up_nic_in_init: rely completley on nvm_file
 	 */
-	if (iwlwifi_mod_params.nvm_file) {
-		err = iwl_nvm_init(mvm);
+	if (cfg->no_power_up_nic_in_init && mvm->nvm_file_name) {
+		err = iwl_nvm_init(mvm, false);
 		if (err)
 			goto out_free;
 	} else {
@@ -518,7 +530,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
  out_free:
 	iwl_phy_db_free(mvm->phy_db);
 	kfree(mvm->scan_cmd);
-	if (!iwlwifi_mod_params.nvm_file)
+	if (!cfg->no_power_up_nic_in_init || !mvm->nvm_file_name)
 		iwl_trans_op_mode_leave(trans);
 	ieee80211_free_hw(mvm->hw);
 	return NULL;
@@ -538,6 +550,7 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode)
 	kfree(mvm->scan_cmd);
 	vfree(mvm->fw_error_dump);
 	kfree(mvm->fw_error_sram);
+	kfree(mvm->fw_error_rxf);
 	kfree(mvm->mcast_filter_cmd);
 	mvm->mcast_filter_cmd = NULL;
 
@@ -814,6 +827,7 @@ void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm)
 	struct iwl_fw_error_dump_file *dump_file;
 	struct iwl_fw_error_dump_data *dump_data;
 	u32 file_len;
+	u32 trans_len;
 
 	lockdep_assert_held(&mvm->mutex);
 
@@ -821,8 +835,13 @@ void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm)
 		return;
 
 	file_len = mvm->fw_error_sram_len +
+		   mvm->fw_error_rxf_len +
 		   sizeof(*dump_file) +
-		   sizeof(*dump_data);
+		   sizeof(*dump_data) * 2;
+
+	trans_len = iwl_trans_dump_data(mvm->trans, NULL, 0);
+	if (trans_len)
+		file_len += trans_len;
 
 	dump_file = vmalloc(file_len);
 	if (!dump_file)
@@ -833,7 +852,12 @@ void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm)
 	dump_file->barker = cpu_to_le32(IWL_FW_ERROR_DUMP_BARKER);
 	dump_file->file_len = cpu_to_le32(file_len);
 	dump_data = (void *)dump_file->data;
-	dump_data->type = IWL_FW_ERROR_DUMP_SRAM;
+	dump_data->type = cpu_to_le32(IWL_FW_ERROR_DUMP_RXF);
+	dump_data->len = cpu_to_le32(mvm->fw_error_rxf_len);
+	memcpy(dump_data->data, mvm->fw_error_rxf, mvm->fw_error_rxf_len);
+
+	dump_data = iwl_mvm_fw_error_next_data(dump_data);
+	dump_data->type = cpu_to_le32(IWL_FW_ERROR_DUMP_SRAM);
 	dump_data->len = cpu_to_le32(mvm->fw_error_sram_len);
 
 	/*
@@ -842,6 +866,23 @@ void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm)
 	 * mvm->fw_error_sram right now.
 	 */
 	memcpy(dump_data->data, mvm->fw_error_sram, mvm->fw_error_sram_len);
+
+	kfree(mvm->fw_error_rxf);
+	mvm->fw_error_rxf = NULL;
+	mvm->fw_error_rxf_len = 0;
+
+	kfree(mvm->fw_error_sram);
+	mvm->fw_error_sram = NULL;
+	mvm->fw_error_sram_len = 0;
+
+	if (trans_len) {
+		void *buf = iwl_mvm_fw_error_next_data(dump_data);
+		u32 real_trans_len = iwl_trans_dump_data(mvm->trans, buf,
+							 trans_len);
+		dump_data = (void *)((u8 *)buf + real_trans_len);
+		dump_file->file_len =
+			cpu_to_le32(file_len - trans_len + real_trans_len);
+	}
 }
 #endif
 
@@ -853,6 +894,7 @@ static void iwl_mvm_nic_error(struct iwl_op_mode *op_mode)
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	iwl_mvm_fw_error_sram_dump(mvm);
+	iwl_mvm_fw_error_rxf_dump(mvm);
 #endif
 
 	iwl_mvm_nic_restart(mvm);
@@ -1126,9 +1168,9 @@ static void iwl_mvm_d0i3_exit_work(struct work_struct *wk)
 	struct iwl_mvm *mvm = container_of(wk, struct iwl_mvm, d0i3_exit_work);
 	struct iwl_host_cmd get_status_cmd = {
 		.id = WOWLAN_GET_STATUSES,
-		.flags = CMD_SYNC | CMD_HIGH_PRIO | CMD_WANT_SKB,
+		.flags = CMD_HIGH_PRIO | CMD_WANT_SKB,
 	};
-	struct iwl_wowlan_status_v6 *status;
+	struct iwl_wowlan_status *status;
 	int ret;
 	u32 disconnection_reasons, wakeup_reasons;
 	__le16 *qos_seq = NULL;
@@ -1158,18 +1200,27 @@ static void iwl_mvm_d0i3_exit_work(struct work_struct *wk)
 	iwl_free_resp(&get_status_cmd);
 out:
 	iwl_mvm_d0i3_enable_tx(mvm, qos_seq);
+	iwl_mvm_unref(mvm, IWL_MVM_REF_EXIT_WORK);
 	mutex_unlock(&mvm->mutex);
 }
 
-static int iwl_mvm_exit_d0i3(struct iwl_op_mode *op_mode)
+int _iwl_mvm_exit_d0i3(struct iwl_mvm *mvm)
 {
-	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
 	u32 flags = CMD_ASYNC | CMD_HIGH_PRIO | CMD_SEND_IN_IDLE |
 		    CMD_WAKE_UP_TRANS;
 	int ret;
 
 	IWL_DEBUG_RPM(mvm, "MVM exiting D0i3\n");
 
+	mutex_lock(&mvm->d0i3_suspend_mutex);
+	if (test_bit(D0I3_DEFER_WAKEUP, &mvm->d0i3_suspend_flags)) {
+		IWL_DEBUG_RPM(mvm, "Deferring d0i3 exit until resume\n");
+		__set_bit(D0I3_PENDING_WAKEUP, &mvm->d0i3_suspend_flags);
+		mutex_unlock(&mvm->d0i3_suspend_mutex);
+		return 0;
+	}
+	mutex_unlock(&mvm->d0i3_suspend_mutex);
+
 	ret = iwl_mvm_send_cmd_pdu(mvm, D0I3_END_CMD, flags, 0, NULL);
 	if (ret)
 		goto out;
@@ -1183,6 +1234,25 @@ out:
 	return ret;
 }
 
+static int iwl_mvm_exit_d0i3(struct iwl_op_mode *op_mode)
+{
+	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
+
+	iwl_mvm_ref(mvm, IWL_MVM_REF_EXIT_WORK);
+	return _iwl_mvm_exit_d0i3(mvm);
+}
+
+static void iwl_mvm_napi_add(struct iwl_op_mode *op_mode,
+			     struct napi_struct *napi,
+			     struct net_device *napi_dev,
+			     int (*poll)(struct napi_struct *, int),
+			     int weight)
+{
+	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
+
+	ieee80211_napi_add(mvm->hw, napi, napi_dev, poll, weight);
+}
+
 static const struct iwl_op_mode_ops iwl_mvm_ops = {
 	.start = iwl_op_mode_mvm_start,
 	.stop = iwl_op_mode_mvm_stop,
@@ -1196,4 +1266,5 @@ static const struct iwl_op_mode_ops iwl_mvm_ops = {
 	.nic_config = iwl_mvm_nic_config,
 	.enter_d0i3 = iwl_mvm_enter_d0i3,
 	.exit_d0i3 = iwl_mvm_exit_d0i3,
+	.napi_add = iwl_mvm_napi_add,
 };
diff --git a/drivers/net/wireless/iwlwifi/mvm/phy-ctxt.c b/drivers/net/wireless/iwlwifi/mvm/phy-ctxt.c
index 237efe0ac1c4..539f3a942d43 100644
--- a/drivers/net/wireless/iwlwifi/mvm/phy-ctxt.c
+++ b/drivers/net/wireless/iwlwifi/mvm/phy-ctxt.c
@@ -156,6 +156,18 @@ static void iwl_mvm_phy_ctxt_cmd_data(struct iwl_mvm *mvm,
 	idle_cnt = chains_static;
 	active_cnt = chains_dynamic;
 
+	/* In scenarios where we only ever use a single-stream rates,
+	 * i.e. legacy 11b/g/a associations, single-stream APs or even
+	 * static SMPS, enable both chains to get diversity, improving
+	 * the case where we're far enough from the AP that attenuation
+	 * between the two antennas is sufficiently different to impact
+	 * performance.
+	 */
+	if (active_cnt == 1 && iwl_mvm_rx_diversity_allowed(mvm)) {
+		idle_cnt = 2;
+		active_cnt = 2;
+	}
+
 	cmd->rxchain_info = cpu_to_le32(mvm->fw->valid_rx_ant <<
 					PHY_RX_CHAIN_VALID_POS);
 	cmd->rxchain_info |= cpu_to_le32(idle_cnt << PHY_RX_CHAIN_CNT_POS);
@@ -187,7 +199,7 @@ static int iwl_mvm_phy_ctxt_apply(struct iwl_mvm *mvm,
 	iwl_mvm_phy_ctxt_cmd_data(mvm, &cmd, chandef,
 				  chains_static, chains_dynamic);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, PHY_CONTEXT_CMD, CMD_SYNC,
+	ret = iwl_mvm_send_cmd_pdu(mvm, PHY_CONTEXT_CMD, 0,
 				   sizeof(struct iwl_phy_context_cmd),
 				   &cmd);
 	if (ret)
@@ -202,18 +214,15 @@ int iwl_mvm_phy_ctxt_add(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt,
 			 struct cfg80211_chan_def *chandef,
 			 u8 chains_static, u8 chains_dynamic)
 {
-	int ret;
-
 	WARN_ON(!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
 		ctxt->ref);
 	lockdep_assert_held(&mvm->mutex);
 
 	ctxt->channel = chandef->chan;
-	ret = iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef,
-				     chains_static, chains_dynamic,
-				     FW_CTXT_ACTION_ADD, 0);
 
-	return ret;
+	return iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef,
+				      chains_static, chains_dynamic,
+				      FW_CTXT_ACTION_ADD, 0);
 }
 
 /*
diff --git a/drivers/net/wireless/iwlwifi/mvm/power.c b/drivers/net/wireless/iwlwifi/mvm/power.c
index 6b636eab3339..c182a8baf685 100644
--- a/drivers/net/wireless/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/iwlwifi/mvm/power.c
@@ -123,28 +123,6 @@ void iwl_mvm_beacon_filter_set_cqm_params(struct iwl_mvm *mvm,
 	cmd->ba_enable_beacon_abort = cpu_to_le32(mvmvif->bf_data.ba_enabled);
 }
 
-int iwl_mvm_update_beacon_abort(struct iwl_mvm *mvm,
-				struct ieee80211_vif *vif, bool enable)
-{
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_beacon_filter_cmd cmd = {
-		IWL_BF_CMD_CONFIG_DEFAULTS,
-		.bf_enable_beacon_filter = cpu_to_le32(1),
-		.ba_enable_beacon_abort = cpu_to_le32(enable),
-	};
-
-	if (!mvmvif->bf_data.bf_enabled)
-		return 0;
-
-	if (mvm->cur_ucode == IWL_UCODE_WOWLAN)
-		cmd.ba_escape_timer = cpu_to_le32(IWL_BA_ESCAPE_TIMER_D3);
-
-	mvmvif->bf_data.ba_enabled = enable;
-	iwl_mvm_beacon_filter_set_cqm_params(mvm, vif, &cmd);
-	iwl_mvm_beacon_filter_debugfs_parameters(vif, &cmd);
-	return iwl_mvm_beacon_filter_send_cmd(mvm, &cmd, CMD_SYNC);
-}
-
 static void iwl_mvm_power_log(struct iwl_mvm *mvm,
 			      struct iwl_mac_power_cmd *cmd)
 {
@@ -268,6 +246,57 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 		IWL_MVM_PS_HEAVY_RX_THLD_PERCENT;
 }
 
+static void iwl_mvm_binding_iterator(void *_data, u8 *mac,
+				      struct ieee80211_vif *vif)
+{
+	unsigned long *data = _data;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	if (!mvmvif->phy_ctxt)
+		return;
+
+	if (vif->type == NL80211_IFTYPE_STATION ||
+	    vif->type == NL80211_IFTYPE_AP)
+		__set_bit(mvmvif->phy_ctxt->id, data);
+}
+
+static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned long phy_ctxt_counter = 0;
+
+	ieee80211_iterate_active_interfaces_atomic(mvm->hw,
+						   IEEE80211_IFACE_ITER_NORMAL,
+						   iwl_mvm_binding_iterator,
+						   &phy_ctxt_counter);
+
+	if (!memcmp(mvmvif->uapsd_misbehaving_bssid, vif->bss_conf.bssid,
+		    ETH_ALEN))
+		return false;
+
+	if (vif->p2p &&
+	    !(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_P2P_PS_UAPSD))
+		return false;
+	/*
+	 * Avoid using uAPSD if P2P client is associated to GO that uses
+	 * opportunistic power save. This is due to current FW limitation.
+	 */
+	if (vif->p2p &&
+	    (vif->bss_conf.p2p_noa_attr.oppps_ctwindow &
+	    IEEE80211_P2P_OPPPS_ENABLE_BIT))
+		return false;
+
+	/*
+	 * Avoid using uAPSD if client is in DCM -
+	 * low latency issue in Miracast
+	 */
+	if (hweight8(phy_ctxt_counter) >= 2)
+		return false;
+
+	return true;
+}
+
 static void iwl_mvm_power_build_cmd(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif,
 				    struct iwl_mac_power_cmd *cmd)
@@ -280,7 +309,6 @@ static void iwl_mvm_power_build_cmd(struct iwl_mvm *mvm,
 	bool radar_detect = false;
 	struct iwl_mvm_vif *mvmvif __maybe_unused =
 		iwl_mvm_vif_from_mac80211(vif);
-	bool allow_uapsd = true;
 
 	cmd->id_and_color = cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->id,
 							    mvmvif->color));
@@ -303,13 +331,8 @@ static void iwl_mvm_power_build_cmd(struct iwl_mvm *mvm,
 
 	cmd->flags |= cpu_to_le16(POWER_FLAGS_POWER_SAVE_ENA_MSK);
 
-#ifdef CONFIG_IWLWIFI_DEBUGFS
-	if (mvmvif->dbgfs_pm.mask & MVM_DEBUGFS_PM_DISABLE_POWER_OFF &&
-	    mvmvif->dbgfs_pm.disable_power_off)
-		cmd->flags &= cpu_to_le16(~POWER_FLAGS_POWER_SAVE_ENA_MSK);
-#endif
 	if (!vif->bss_conf.ps || iwl_mvm_vif_low_latency(mvmvif) ||
-	    mvm->pm_disabled)
+	    !mvmvif->pm_enabled)
 		return;
 
 	cmd->flags |= cpu_to_le16(POWER_FLAGS_POWER_MANAGEMENT_ENA_MSK);
@@ -351,23 +374,7 @@ static void iwl_mvm_power_build_cmd(struct iwl_mvm *mvm,
 			cpu_to_le32(IWL_MVM_WOWLAN_PS_TX_DATA_TIMEOUT);
 	}
 
-	if (!memcmp(mvmvif->uapsd_misbehaving_bssid, vif->bss_conf.bssid,
-		    ETH_ALEN))
-		allow_uapsd = false;
-
-	if (vif->p2p &&
-	    !(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_P2P_PS_UAPSD))
-		allow_uapsd = false;
-	/*
-	 * Avoid using uAPSD if P2P client is associated to GO that uses
-	 * opportunistic power save. This is due to current FW limitation.
-	 */
-	if (vif->p2p &&
-	    vif->bss_conf.p2p_noa_attr.oppps_ctwindow &
-	    IEEE80211_P2P_OPPPS_ENABLE_BIT)
-		allow_uapsd = false;
-
-	if (allow_uapsd)
+	if (iwl_mvm_power_allow_uapsd(mvm, vif))
 		iwl_mvm_power_configure_uapsd(mvm, vif, cmd);
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
@@ -421,20 +428,13 @@ static int iwl_mvm_power_send_cmd(struct iwl_mvm *mvm,
 {
 	struct iwl_mac_power_cmd cmd = {};
 
-	if (vif->type != NL80211_IFTYPE_STATION)
-		return 0;
-
-	if (vif->p2p &&
-	    !(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_DCM))
-		return 0;
-
 	iwl_mvm_power_build_cmd(mvm, vif, &cmd);
 	iwl_mvm_power_log(mvm, &cmd);
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	memcpy(&iwl_mvm_vif_from_mac80211(vif)->mac_pwr_cmd, &cmd, sizeof(cmd));
 #endif
 
-	return iwl_mvm_send_cmd_pdu(mvm, MAC_PM_POWER_TABLE, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, MAC_PM_POWER_TABLE, 0,
 				    sizeof(cmd), &cmd);
 }
 
@@ -444,12 +444,6 @@ int iwl_mvm_power_update_device(struct iwl_mvm *mvm)
 		.flags = cpu_to_le16(DEVICE_POWER_FLAGS_POWER_SAVE_ENA_MSK),
 	};
 
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_PM_CMD_SUPPORT))
-		return 0;
-
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_DEVICE_PS_CMD))
-		return 0;
-
 	if (iwlmvm_mod_params.power_scheme == IWL_POWER_SCHEME_CAM)
 		mvm->ps_disabled = true;
 
@@ -466,7 +460,7 @@ int iwl_mvm_power_update_device(struct iwl_mvm *mvm)
 			"Sending device power command with flags = 0x%X\n",
 			cmd.flags);
 
-	return iwl_mvm_send_cmd_pdu(mvm, POWER_TABLE_CMD, CMD_SYNC, sizeof(cmd),
+	return iwl_mvm_send_cmd_pdu(mvm, POWER_TABLE_CMD, 0, sizeof(cmd),
 				    &cmd);
 }
 
@@ -508,86 +502,69 @@ int iwl_mvm_power_uapsd_misbehaving_ap_notif(struct iwl_mvm *mvm,
 	return 0;
 }
 
-struct iwl_power_constraint {
+struct iwl_power_vifs {
 	struct ieee80211_vif *bf_vif;
 	struct ieee80211_vif *bss_vif;
 	struct ieee80211_vif *p2p_vif;
-	u16 bss_phyctx_id;
-	u16 p2p_phyctx_id;
-	bool pm_disabled;
-	bool ps_disabled;
-	struct iwl_mvm *mvm;
+	struct ieee80211_vif *ap_vif;
+	struct ieee80211_vif *monitor_vif;
+	bool p2p_active;
+	bool bss_active;
+	bool ap_active;
+	bool monitor_active;
 };
 
 static void iwl_mvm_power_iterator(void *_data, u8 *mac,
 				   struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_power_constraint *power_iterator = _data;
-	struct iwl_mvm *mvm = power_iterator->mvm;
+	struct iwl_power_vifs *power_iterator = _data;
 
+	mvmvif->pm_enabled = false;
 	switch (ieee80211_vif_type_p2p(vif)) {
 	case NL80211_IFTYPE_P2P_DEVICE:
 		break;
 
 	case NL80211_IFTYPE_P2P_GO:
 	case NL80211_IFTYPE_AP:
-		/* no BSS power mgmt if we have an active AP */
-		if (mvmvif->ap_ibss_active)
-			power_iterator->pm_disabled = true;
+		/* only a single MAC of the same type */
+		WARN_ON(power_iterator->ap_vif);
+		power_iterator->ap_vif = vif;
+		if (mvmvif->phy_ctxt)
+			if (mvmvif->phy_ctxt->id < MAX_PHYS)
+				power_iterator->ap_active = true;
 		break;
 
 	case NL80211_IFTYPE_MONITOR:
-		/* no BSS power mgmt and no device power save */
-		power_iterator->pm_disabled = true;
-		power_iterator->ps_disabled = true;
+		/* only a single MAC of the same type */
+		WARN_ON(power_iterator->monitor_vif);
+		power_iterator->monitor_vif = vif;
+		if (mvmvif->phy_ctxt)
+			if (mvmvif->phy_ctxt->id < MAX_PHYS)
+				power_iterator->monitor_active = true;
 		break;
 
 	case NL80211_IFTYPE_P2P_CLIENT:
-		if (mvmvif->phy_ctxt)
-			power_iterator->p2p_phyctx_id = mvmvif->phy_ctxt->id;
-
-		/* we should have only one P2P vif */
+		/* only a single MAC of the same type */
 		WARN_ON(power_iterator->p2p_vif);
 		power_iterator->p2p_vif = vif;
-
-		IWL_DEBUG_POWER(mvm, "p2p: p2p_id=%d, bss_id=%d\n",
-				power_iterator->p2p_phyctx_id,
-				power_iterator->bss_phyctx_id);
-		if (!(mvm->fw->ucode_capa.flags &
-		      IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_DCM)) {
-			/* no BSS power mgmt if we have a P2P client*/
-			power_iterator->pm_disabled = true;
-		} else if (power_iterator->p2p_phyctx_id < MAX_PHYS &&
-			   power_iterator->bss_phyctx_id < MAX_PHYS &&
-			   power_iterator->p2p_phyctx_id ==
-			   power_iterator->bss_phyctx_id) {
-			power_iterator->pm_disabled = true;
-		}
+		if (mvmvif->phy_ctxt)
+			if (mvmvif->phy_ctxt->id < MAX_PHYS)
+				power_iterator->p2p_active = true;
 		break;
 
 	case NL80211_IFTYPE_STATION:
-		if (mvmvif->phy_ctxt)
-			power_iterator->bss_phyctx_id = mvmvif->phy_ctxt->id;
-
-		/* we should have only one BSS vif */
+		/* only a single MAC of the same type */
 		WARN_ON(power_iterator->bss_vif);
 		power_iterator->bss_vif = vif;
+		if (mvmvif->phy_ctxt)
+			if (mvmvif->phy_ctxt->id < MAX_PHYS)
+				power_iterator->bss_active = true;
 
 		if (mvmvif->bf_data.bf_enabled &&
 		    !WARN_ON(power_iterator->bf_vif))
 			power_iterator->bf_vif = vif;
 
-		IWL_DEBUG_POWER(mvm, "bss: p2p_id=%d, bss_id=%d\n",
-				power_iterator->p2p_phyctx_id,
-				power_iterator->bss_phyctx_id);
-		if (mvm->fw->ucode_capa.flags &
-		    IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_DCM &&
-			(power_iterator->p2p_phyctx_id < MAX_PHYS &&
-			 power_iterator->bss_phyctx_id < MAX_PHYS &&
-			 power_iterator->p2p_phyctx_id ==
-			 power_iterator->bss_phyctx_id))
-			power_iterator->pm_disabled = true;
 		break;
 
 	default:
@@ -596,70 +573,73 @@ static void iwl_mvm_power_iterator(void *_data, u8 *mac,
 }
 
 static void
-iwl_mvm_power_get_global_constraint(struct iwl_mvm *mvm,
-				    struct iwl_power_constraint *constraint)
+iwl_mvm_power_set_pm(struct iwl_mvm *mvm,
+				    struct iwl_power_vifs *vifs)
 {
-	lockdep_assert_held(&mvm->mutex);
+	struct iwl_mvm_vif *bss_mvmvif = NULL;
+	struct iwl_mvm_vif *p2p_mvmvif = NULL;
+	struct iwl_mvm_vif *ap_mvmvif = NULL;
+	bool client_same_channel = false;
+	bool ap_same_channel = false;
 
-	if (iwlmvm_mod_params.power_scheme == IWL_POWER_SCHEME_CAM) {
-		constraint->pm_disabled = true;
-		constraint->ps_disabled = true;
-	}
+	lockdep_assert_held(&mvm->mutex);
 
+	/* get vifs info + set pm_enable to false */
 	ieee80211_iterate_active_interfaces_atomic(mvm->hw,
 					    IEEE80211_IFACE_ITER_NORMAL,
-					    iwl_mvm_power_iterator, constraint);
-}
-
-int iwl_mvm_power_update_mac(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
-{
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_power_constraint constraint = {
-		    .p2p_phyctx_id = MAX_PHYS,
-		    .bss_phyctx_id = MAX_PHYS,
-		    .mvm = mvm,
-	};
-	bool ba_enable;
-	int ret;
+					    iwl_mvm_power_iterator, vifs);
 
-	lockdep_assert_held(&mvm->mutex);
+	if (vifs->bss_vif)
+		bss_mvmvif = iwl_mvm_vif_from_mac80211(vifs->bss_vif);
 
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_PM_CMD_SUPPORT))
-		return 0;
+	if (vifs->p2p_vif)
+		p2p_mvmvif = iwl_mvm_vif_from_mac80211(vifs->p2p_vif);
 
-	iwl_mvm_power_get_global_constraint(mvm, &constraint);
-	mvm->ps_disabled = constraint.ps_disabled;
-	mvm->pm_disabled = constraint.pm_disabled;
+	if (vifs->ap_vif)
+		ap_mvmvif = iwl_mvm_vif_from_mac80211(vifs->ap_vif);
 
-	/* don't update device power state unless we add / remove monitor */
-	if (vif->type == NL80211_IFTYPE_MONITOR) {
-		ret = iwl_mvm_power_update_device(mvm);
-		if (ret)
-			return ret;
+	/* enable PM on bss if bss stand alone */
+	if (vifs->bss_active && !vifs->p2p_active && !vifs->ap_active) {
+		bss_mvmvif->pm_enabled = true;
+		return;
 	}
 
-	if (constraint.bss_vif) {
-		ret = iwl_mvm_power_send_cmd(mvm, constraint.bss_vif);
-		if (ret)
-			return ret;
+	/* enable PM on p2p if p2p stand alone */
+	if (vifs->p2p_active && !vifs->bss_active && !vifs->ap_active) {
+		if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_P2P_PM)
+			p2p_mvmvif->pm_enabled = true;
+		return;
 	}
 
-	if (constraint.p2p_vif) {
-		ret = iwl_mvm_power_send_cmd(mvm, constraint.p2p_vif);
-		if (ret)
-			return ret;
+	if (vifs->bss_active && vifs->p2p_active)
+		client_same_channel = (bss_mvmvif->phy_ctxt->id ==
+				       p2p_mvmvif->phy_ctxt->id);
+	if (vifs->bss_active && vifs->ap_active)
+		ap_same_channel = (bss_mvmvif->phy_ctxt->id ==
+				   ap_mvmvif->phy_ctxt->id);
+
+	/* clients are not stand alone: enable PM if DCM */
+	if (!(client_same_channel || ap_same_channel) &&
+	    (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_DCM)) {
+		if (vifs->bss_active)
+			bss_mvmvif->pm_enabled = true;
+		if (vifs->p2p_active &&
+		    (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_P2P_PM))
+			p2p_mvmvif->pm_enabled = true;
+		return;
 	}
 
-	if (!constraint.bf_vif)
-		return 0;
-
-	vif = constraint.bf_vif;
-	mvmvif = iwl_mvm_vif_from_mac80211(vif);
-
-	ba_enable = !(constraint.pm_disabled || constraint.ps_disabled ||
-		      !vif->bss_conf.ps || iwl_mvm_vif_low_latency(mvmvif));
-
-	return iwl_mvm_update_beacon_abort(mvm, constraint.bf_vif, ba_enable);
+	/*
+	 * There is only one channel in the system and there are only
+	 * bss and p2p clients that share it
+	 */
+	if (client_same_channel && !vifs->ap_active &&
+	    (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BSS_P2P_PS_SCM)) {
+		/* share same channel*/
+		bss_mvmvif->pm_enabled = true;
+		if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_P2P_PM)
+			p2p_mvmvif->pm_enabled = true;
+	}
 }
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
@@ -671,19 +651,10 @@ int iwl_mvm_power_mac_dbgfs_read(struct iwl_mvm *mvm,
 	struct iwl_mac_power_cmd cmd = {};
 	int pos = 0;
 
-	if (WARN_ON(!(mvm->fw->ucode_capa.flags &
-		      IWL_UCODE_TLV_FLAGS_PM_CMD_SUPPORT)))
-		return 0;
-
 	mutex_lock(&mvm->mutex);
 	memcpy(&cmd, &mvmvif->mac_pwr_cmd, sizeof(cmd));
 	mutex_unlock(&mvm->mutex);
 
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_DEVICE_PS_CMD))
-		pos += scnprintf(buf+pos, bufsz-pos, "disable_power_off = %d\n",
-				 (cmd.flags &
-				 cpu_to_le16(POWER_FLAGS_POWER_SAVE_ENA_MSK)) ?
-				 0 : 1);
 	pos += scnprintf(buf+pos, bufsz-pos, "power_scheme = %d\n",
 			 iwlmvm_mod_params.power_scheme);
 	pos += scnprintf(buf+pos, bufsz-pos, "flags = 0x%x\n",
@@ -790,7 +761,7 @@ static int _iwl_mvm_enable_beacon_filter(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
 
-	if (mvmvif != mvm->bf_allowed_vif ||
+	if (mvmvif != mvm->bf_allowed_vif || !vif->bss_conf.dtim_period ||
 	    vif->type != NL80211_IFTYPE_STATION || vif->p2p)
 		return 0;
 
@@ -818,6 +789,26 @@ int iwl_mvm_enable_beacon_filter(struct iwl_mvm *mvm,
 	return _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd, flags, false);
 }
 
+static int iwl_mvm_update_beacon_abort(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif,
+				       bool enable)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_beacon_filter_cmd cmd = {
+		IWL_BF_CMD_CONFIG_DEFAULTS,
+		.bf_enable_beacon_filter = cpu_to_le32(1),
+	};
+
+	if (!mvmvif->bf_data.bf_enabled)
+		return 0;
+
+	if (mvm->cur_ucode == IWL_UCODE_WOWLAN)
+		cmd.ba_escape_timer = cpu_to_le32(IWL_BA_ESCAPE_TIMER_D3);
+
+	mvmvif->bf_data.ba_enabled = enable;
+	return _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd, 0, false);
+}
+
 int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 				  struct ieee80211_vif *vif,
 				  u32 flags)
@@ -826,8 +817,7 @@ int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
 
-	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BF_UPDATED) ||
-	    vif->type != NL80211_IFTYPE_STATION || vif->p2p)
+	if (vif->type != NL80211_IFTYPE_STATION || vif->p2p)
 		return 0;
 
 	ret = iwl_mvm_beacon_filter_send_cmd(mvm, &cmd, flags);
@@ -838,6 +828,55 @@ int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 	return ret;
 }
 
+int iwl_mvm_power_update_mac(struct iwl_mvm *mvm)
+{
+	struct iwl_mvm_vif *mvmvif;
+	struct iwl_power_vifs vifs = {};
+	bool ba_enable;
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	iwl_mvm_power_set_pm(mvm, &vifs);
+
+	/* disable PS if CAM */
+	if (iwlmvm_mod_params.power_scheme == IWL_POWER_SCHEME_CAM) {
+		mvm->ps_disabled = true;
+	} else {
+	/* don't update device power state unless we add / remove monitor */
+		if (vifs.monitor_vif) {
+			if (vifs.monitor_active)
+				mvm->ps_disabled = true;
+			ret = iwl_mvm_power_update_device(mvm);
+			if (ret)
+				return ret;
+		}
+	}
+
+	if (vifs.bss_vif) {
+		ret = iwl_mvm_power_send_cmd(mvm, vifs.bss_vif);
+		if (ret)
+			return ret;
+	}
+
+	if (vifs.p2p_vif) {
+		ret = iwl_mvm_power_send_cmd(mvm, vifs.p2p_vif);
+		if (ret)
+			return ret;
+	}
+
+	if (!vifs.bf_vif)
+		return 0;
+
+	mvmvif = iwl_mvm_vif_from_mac80211(vifs.bf_vif);
+
+	ba_enable = !(!mvmvif->pm_enabled || mvm->ps_disabled ||
+		      !vifs.bf_vif->bss_conf.ps ||
+		      iwl_mvm_vif_low_latency(mvmvif));
+
+	return iwl_mvm_update_beacon_abort(mvm, vifs.bf_vif, ba_enable);
+}
+
 int iwl_mvm_update_d0i3_power_mode(struct iwl_mvm *mvm,
 				   struct ieee80211_vif *vif,
 				   bool enable, u32 flags)
@@ -861,9 +900,10 @@ int iwl_mvm_update_d0i3_power_mode(struct iwl_mvm *mvm,
 		if (WARN_ON(!dtimper_msec))
 			return 0;
 
-		cmd.flags |=
-			cpu_to_le16(POWER_FLAGS_SKIP_OVER_DTIM_MSK);
 		cmd.skip_dtim_periods = 300 / dtimper_msec;
+		if (cmd.skip_dtim_periods)
+			cmd.flags |=
+				cpu_to_le16(POWER_FLAGS_SKIP_OVER_DTIM_MSK);
 	}
 	iwl_mvm_power_log(mvm, &cmd);
 #ifdef CONFIG_IWLWIFI_DEBUGFS
@@ -894,33 +934,3 @@ int iwl_mvm_update_d0i3_power_mode(struct iwl_mvm *mvm,
 
 	return ret;
 }
-
-int iwl_mvm_update_beacon_filter(struct iwl_mvm *mvm,
-				 struct ieee80211_vif *vif,
-				 bool force,
-				 u32 flags)
-{
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-
-	if (mvmvif != mvm->bf_allowed_vif)
-		return 0;
-
-	if (!mvmvif->bf_data.bf_enabled) {
-		/* disable beacon filtering explicitly if force is true */
-		if (force)
-			return iwl_mvm_disable_beacon_filter(mvm, vif, flags);
-		return 0;
-	}
-
-	return iwl_mvm_enable_beacon_filter(mvm, vif, flags);
-}
-
-int iwl_power_legacy_set_cam_mode(struct iwl_mvm *mvm)
-{
-	struct iwl_powertable_cmd cmd = {
-		.keep_alive_seconds = POWER_KEEP_ALIVE_PERIOD_SEC,
-	};
-
-	return iwl_mvm_send_cmd_pdu(mvm, POWER_TABLE_CMD, CMD_SYNC,
-				    sizeof(cmd), &cmd);
-}
diff --git a/drivers/net/wireless/iwlwifi/mvm/quota.c b/drivers/net/wireless/iwlwifi/mvm/quota.c
index 35e86e06dffd..ba68d7b84505 100644
--- a/drivers/net/wireless/iwlwifi/mvm/quota.c
+++ b/drivers/net/wireless/iwlwifi/mvm/quota.c
@@ -285,7 +285,7 @@ int iwl_mvm_update_quotas(struct iwl_mvm *mvm, struct ieee80211_vif *newvif)
 
 	iwl_mvm_adjust_quota_for_noa(mvm, &cmd);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, TIME_QUOTA_CMD, CMD_SYNC,
+	ret = iwl_mvm_send_cmd_pdu(mvm, TIME_QUOTA_CMD, 0,
 				   sizeof(cmd), &cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send quota: %d\n", ret);
diff --git a/drivers/net/wireless/iwlwifi/mvm/rs.c b/drivers/net/wireless/iwlwifi/mvm/rs.c
index e1c838899363..306a6caa4868 100644
--- a/drivers/net/wireless/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/iwlwifi/mvm/rs.c
@@ -211,7 +211,7 @@ static const struct rs_tx_column rs_tx_columns[] = {
 		.next_columns = {
 			RS_COLUMN_LEGACY_ANT_B,
 			RS_COLUMN_SISO_ANT_A,
-			RS_COLUMN_SISO_ANT_B,
+			RS_COLUMN_MIMO2,
 			RS_COLUMN_INVALID,
 			RS_COLUMN_INVALID,
 			RS_COLUMN_INVALID,
@@ -223,8 +223,8 @@ static const struct rs_tx_column rs_tx_columns[] = {
 		.ant = ANT_B,
 		.next_columns = {
 			RS_COLUMN_LEGACY_ANT_A,
-			RS_COLUMN_SISO_ANT_A,
 			RS_COLUMN_SISO_ANT_B,
+			RS_COLUMN_MIMO2,
 			RS_COLUMN_INVALID,
 			RS_COLUMN_INVALID,
 			RS_COLUMN_INVALID,
@@ -238,10 +238,10 @@ static const struct rs_tx_column rs_tx_columns[] = {
 			RS_COLUMN_SISO_ANT_B,
 			RS_COLUMN_MIMO2,
 			RS_COLUMN_SISO_ANT_A_SGI,
-			RS_COLUMN_SISO_ANT_B_SGI,
 			RS_COLUMN_LEGACY_ANT_A,
 			RS_COLUMN_LEGACY_ANT_B,
 			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
 		},
 		.checks = {
 			rs_siso_allow,
@@ -254,10 +254,10 @@ static const struct rs_tx_column rs_tx_columns[] = {
 			RS_COLUMN_SISO_ANT_A,
 			RS_COLUMN_MIMO2,
 			RS_COLUMN_SISO_ANT_B_SGI,
-			RS_COLUMN_SISO_ANT_A_SGI,
 			RS_COLUMN_LEGACY_ANT_A,
 			RS_COLUMN_LEGACY_ANT_B,
 			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
 		},
 		.checks = {
 			rs_siso_allow,
@@ -271,10 +271,10 @@ static const struct rs_tx_column rs_tx_columns[] = {
 			RS_COLUMN_SISO_ANT_B_SGI,
 			RS_COLUMN_MIMO2_SGI,
 			RS_COLUMN_SISO_ANT_A,
-			RS_COLUMN_SISO_ANT_B,
-			RS_COLUMN_MIMO2,
 			RS_COLUMN_LEGACY_ANT_A,
 			RS_COLUMN_LEGACY_ANT_B,
+			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
 		},
 		.checks = {
 			rs_siso_allow,
@@ -289,10 +289,10 @@ static const struct rs_tx_column rs_tx_columns[] = {
 			RS_COLUMN_SISO_ANT_A_SGI,
 			RS_COLUMN_MIMO2_SGI,
 			RS_COLUMN_SISO_ANT_B,
-			RS_COLUMN_SISO_ANT_A,
-			RS_COLUMN_MIMO2,
 			RS_COLUMN_LEGACY_ANT_A,
 			RS_COLUMN_LEGACY_ANT_B,
+			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
 		},
 		.checks = {
 			rs_siso_allow,
@@ -304,12 +304,12 @@ static const struct rs_tx_column rs_tx_columns[] = {
 		.ant = ANT_AB,
 		.next_columns = {
 			RS_COLUMN_SISO_ANT_A,
-			RS_COLUMN_SISO_ANT_B,
-			RS_COLUMN_SISO_ANT_A_SGI,
-			RS_COLUMN_SISO_ANT_B_SGI,
 			RS_COLUMN_MIMO2_SGI,
 			RS_COLUMN_LEGACY_ANT_A,
 			RS_COLUMN_LEGACY_ANT_B,
+			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
 		},
 		.checks = {
 			rs_mimo_allow,
@@ -321,12 +321,12 @@ static const struct rs_tx_column rs_tx_columns[] = {
 		.sgi = true,
 		.next_columns = {
 			RS_COLUMN_SISO_ANT_A_SGI,
-			RS_COLUMN_SISO_ANT_B_SGI,
-			RS_COLUMN_SISO_ANT_A,
-			RS_COLUMN_SISO_ANT_B,
 			RS_COLUMN_MIMO2,
 			RS_COLUMN_LEGACY_ANT_A,
 			RS_COLUMN_LEGACY_ANT_B,
+			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
+			RS_COLUMN_INVALID,
 		},
 		.checks = {
 			rs_mimo_allow,
@@ -527,6 +527,9 @@ static void rs_rate_scale_clear_tbl_windows(struct iwl_mvm *mvm,
 	IWL_DEBUG_RATE(mvm, "Clearing up window stats\n");
 	for (i = 0; i < IWL_RATE_COUNT; i++)
 		rs_rate_scale_clear_window(&tbl->win[i]);
+
+	for (i = 0; i < ARRAY_SIZE(tbl->tpc_win); i++)
+		rs_rate_scale_clear_window(&tbl->tpc_win[i]);
 }
 
 static inline u8 rs_is_valid_ant(u8 valid_antenna, u8 ant_type)
@@ -656,17 +659,34 @@ static int _rs_collect_tx_data(struct iwl_scale_tbl_info *tbl,
 	return 0;
 }
 
-static int rs_collect_tx_data(struct iwl_scale_tbl_info *tbl,
-			      int scale_index, int attempts, int successes)
+static int rs_collect_tx_data(struct iwl_lq_sta *lq_sta,
+			      struct iwl_scale_tbl_info *tbl,
+			      int scale_index, int attempts, int successes,
+			      u8 reduced_txp)
 {
 	struct iwl_rate_scale_data *window = NULL;
+	int ret;
 
 	if (scale_index < 0 || scale_index >= IWL_RATE_COUNT)
 		return -EINVAL;
 
+	if (tbl->column != RS_COLUMN_INVALID) {
+		lq_sta->tx_stats[tbl->column][scale_index].total += attempts;
+		lq_sta->tx_stats[tbl->column][scale_index].success += successes;
+	}
+
 	/* Select window for current tx bit rate */
 	window = &(tbl->win[scale_index]);
 
+	ret = _rs_collect_tx_data(tbl, scale_index, attempts, successes,
+				  window);
+	if (ret)
+		return ret;
+
+	if (WARN_ON_ONCE(reduced_txp > TPC_MAX_REDUCTION))
+		return -EINVAL;
+
+	window = &tbl->tpc_win[reduced_txp];
 	return _rs_collect_tx_data(tbl, scale_index, attempts, successes,
 				   window);
 }
@@ -1000,6 +1020,7 @@ static void rs_tx_status(void *mvm_r, struct ieee80211_supported_band *sband,
 	u32 ucode_rate;
 	struct rs_rate rate;
 	struct iwl_scale_tbl_info *curr_tbl, *other_tbl, *tmp_tbl;
+	u8 reduced_txp = (uintptr_t)info->status.status_driver_data[0];
 
 	/* Treat uninitialized rate scaling data same as non-existing. */
 	if (!lq_sta) {
@@ -1141,9 +1162,10 @@ static void rs_tx_status(void *mvm_r, struct ieee80211_supported_band *sband,
 	if (info->flags & IEEE80211_TX_STAT_AMPDU) {
 		ucode_rate = le32_to_cpu(table->rs_table[0]);
 		rs_rate_from_ucode_rate(ucode_rate, info->band, &rate);
-		rs_collect_tx_data(curr_tbl, rate.index,
+		rs_collect_tx_data(lq_sta, curr_tbl, rate.index,
 				   info->status.ampdu_len,
-				   info->status.ampdu_ack_len);
+				   info->status.ampdu_ack_len,
+				   reduced_txp);
 
 		/* Update success/fail counts if not searching for new mode */
 		if (lq_sta->rs_state == RS_STATE_STAY_IN_COLUMN) {
@@ -1176,8 +1198,9 @@ static void rs_tx_status(void *mvm_r, struct ieee80211_supported_band *sband,
 			else
 				continue;
 
-			rs_collect_tx_data(tmp_tbl, rate.index, 1,
-					   i < retries ? 0 : legacy_success);
+			rs_collect_tx_data(lq_sta, tmp_tbl, rate.index, 1,
+					   i < retries ? 0 : legacy_success,
+					   reduced_txp);
 		}
 
 		/* Update success/fail counts if not searching for new mode */
@@ -1188,6 +1211,7 @@ static void rs_tx_status(void *mvm_r, struct ieee80211_supported_band *sband,
 	}
 	/* The last TX rate is cached in lq_sta; it's set in if/else above */
 	lq_sta->last_rate_n_flags = ucode_rate;
+	IWL_DEBUG_RATE(mvm, "reduced txpower: %d\n", reduced_txp);
 done:
 	/* See if there's a better rate or modulation mode to try. */
 	if (sta && sta->supp_rates[sband->band])
@@ -1311,105 +1335,50 @@ static void rs_set_expected_tpt_table(struct iwl_lq_sta *lq_sta,
 	tbl->expected_tpt = rs_get_expected_tpt_table(lq_sta, column, rate->bw);
 }
 
-/*
- * Find starting rate for new "search" high-throughput mode of modulation.
- * Goal is to find lowest expected rate (under perfect conditions) that is
- * above the current measured throughput of "active" mode, to give new mode
- * a fair chance to prove itself without too many challenges.
- *
- * This gets called when transitioning to more aggressive modulation
- * (i.e. legacy to SISO or MIMO, or SISO to MIMO), as well as less aggressive
- * (i.e. MIMO to SISO).  When moving to MIMO, bit rate will typically need
- * to decrease to match "active" throughput.  When moving from MIMO to SISO,
- * bit rate will typically need to increase, but not if performance was bad.
- */
 static s32 rs_get_best_rate(struct iwl_mvm *mvm,
 			    struct iwl_lq_sta *lq_sta,
 			    struct iwl_scale_tbl_info *tbl,	/* "search" */
-			    u16 rate_mask, s8 index)
+			    unsigned long rate_mask, s8 index)
 {
-	/* "active" values */
 	struct iwl_scale_tbl_info *active_tbl =
 	    &(lq_sta->lq_info[lq_sta->active_tbl]);
-	s32 active_sr = active_tbl->win[index].success_ratio;
-	s32 active_tpt = active_tbl->expected_tpt[index];
-	/* expected "search" throughput */
+	s32 success_ratio = active_tbl->win[index].success_ratio;
+	u16 expected_current_tpt = active_tbl->expected_tpt[index];
 	const u16 *tpt_tbl = tbl->expected_tpt;
-
-	s32 new_rate, high, low, start_hi;
 	u16 high_low;
-	s8 rate = index;
-
-	new_rate = high = low = start_hi = IWL_RATE_INVALID;
-
-	while (1) {
-		high_low = rs_get_adjacent_rate(mvm, rate, rate_mask,
-						tbl->rate.type);
-
-		low = high_low & 0xff;
-		high = (high_low >> 8) & 0xff;
+	u32 target_tpt;
+	int rate_idx;
 
-		/*
-		 * Lower the "search" bit rate, to give new "search" mode
-		 * approximately the same throughput as "active" if:
-		 *
-		 * 1) "Active" mode has been working modestly well (but not
-		 *    great), and expected "search" throughput (under perfect
-		 *    conditions) at candidate rate is above the actual
-		 *    measured "active" throughput (but less than expected
-		 *    "active" throughput under perfect conditions).
-		 * OR
-		 * 2) "Active" mode has been working perfectly or very well
-		 *    and expected "search" throughput (under perfect
-		 *    conditions) at candidate rate is above expected
-		 *    "active" throughput (under perfect conditions).
-		 */
-		if ((((100 * tpt_tbl[rate]) > lq_sta->last_tpt) &&
-		     ((active_sr > RS_SR_FORCE_DECREASE) &&
-		      (active_sr <= IWL_RATE_HIGH_TH) &&
-		      (tpt_tbl[rate] <= active_tpt))) ||
-		    ((active_sr >= IWL_RATE_SCALE_SWITCH) &&
-		     (tpt_tbl[rate] > active_tpt))) {
-			/* (2nd or later pass)
-			 * If we've already tried to raise the rate, and are
-			 * now trying to lower it, use the higher rate. */
-			if (start_hi != IWL_RATE_INVALID) {
-				new_rate = start_hi;
-				break;
-			}
-
-			new_rate = rate;
+	if (success_ratio > RS_SR_NO_DECREASE) {
+		target_tpt = 100 * expected_current_tpt;
+		IWL_DEBUG_RATE(mvm,
+			       "SR %d high. Find rate exceeding EXPECTED_CURRENT %d\n",
+			       success_ratio, target_tpt);
+	} else {
+		target_tpt = lq_sta->last_tpt;
+		IWL_DEBUG_RATE(mvm,
+			       "SR %d not thag good. Find rate exceeding ACTUAL_TPT %d\n",
+			       success_ratio, target_tpt);
+	}
 
-			/* Loop again with lower rate */
-			if (low != IWL_RATE_INVALID)
-				rate = low;
+	rate_idx = find_first_bit(&rate_mask, BITS_PER_LONG);
 
-			/* Lower rate not available, use the original */
-			else
-				break;
-
-		/* Else try to raise the "search" rate to match "active" */
-		} else {
-			/* (2nd or later pass)
-			 * If we've already tried to lower the rate, and are
-			 * now trying to raise it, use the lower rate. */
-			if (new_rate != IWL_RATE_INVALID)
-				break;
+	while (rate_idx != IWL_RATE_INVALID) {
+		if (target_tpt < (100 * tpt_tbl[rate_idx]))
+			break;
 
-			/* Loop again with higher rate */
-			else if (high != IWL_RATE_INVALID) {
-				start_hi = high;
-				rate = high;
+		high_low = rs_get_adjacent_rate(mvm, rate_idx, rate_mask,
+						tbl->rate.type);
 
-			/* Higher rate not available, use the original */
-			} else {
-				new_rate = rate;
-				break;
-			}
-		}
+		rate_idx = (high_low >> 8) & 0xff;
 	}
 
-	return new_rate;
+	IWL_DEBUG_RATE(mvm, "Best rate found %d target_tp %d expected_new %d\n",
+		       rate_idx, target_tpt,
+		       rate_idx != IWL_RATE_INVALID ?
+		       100 * tpt_tbl[rate_idx] : IWL_INVALID_VALUE);
+
+	return rate_idx;
 }
 
 static u32 rs_bw_from_sta_bw(struct ieee80211_sta *sta)
@@ -1584,7 +1553,7 @@ static enum rs_column rs_get_next_column(struct iwl_mvm *mvm,
 
 		tpt = lq_sta->last_tpt / 100;
 		expected_tpt_tbl = rs_get_expected_tpt_table(lq_sta, next_col,
-							     tbl->rate.bw);
+						     rs_bw_from_sta_bw(sta));
 		if (WARN_ON_ONCE(!expected_tpt_tbl))
 			continue;
 
@@ -1625,7 +1594,7 @@ static int rs_switch_to_column(struct iwl_mvm *mvm,
 	const struct rs_tx_column *curr_column = &rs_tx_columns[tbl->column];
 	u32 sz = (sizeof(struct iwl_scale_tbl_info) -
 		  (sizeof(struct iwl_rate_scale_data) * IWL_RATE_COUNT));
-	u16 rate_mask = 0;
+	unsigned long rate_mask = 0;
 	u32 rate_idx = 0;
 
 	memcpy(search_tbl, tbl, sz);
@@ -1667,7 +1636,7 @@ static int rs_switch_to_column(struct iwl_mvm *mvm,
 		    !(BIT(rate_idx) & rate_mask)) {
 			IWL_DEBUG_RATE(mvm,
 				       "can not switch with index %d"
-				       " rate mask %x\n",
+				       " rate mask %lx\n",
 				       rate_idx, rate_mask);
 
 			goto err;
@@ -1769,6 +1738,203 @@ out:
 	return action;
 }
 
+static void rs_get_adjacent_txp(struct iwl_mvm *mvm, int index,
+				int *weaker, int *stronger)
+{
+	*weaker = index + TPC_TX_POWER_STEP;
+	if (*weaker > TPC_MAX_REDUCTION)
+		*weaker = TPC_INVALID;
+
+	*stronger = index - TPC_TX_POWER_STEP;
+	if (*stronger < 0)
+		*stronger = TPC_INVALID;
+}
+
+static bool rs_tpc_allowed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			   struct rs_rate *rate, enum ieee80211_band band)
+{
+	int index = rate->index;
+	bool cam = (iwlmvm_mod_params.power_scheme == IWL_POWER_SCHEME_CAM);
+	bool sta_ps_disabled = (vif->type == NL80211_IFTYPE_STATION &&
+				!vif->bss_conf.ps);
+
+	IWL_DEBUG_RATE(mvm, "cam: %d sta_ps_disabled %d\n",
+		       cam, sta_ps_disabled);
+	/*
+	 * allow tpc only if power management is enabled, or bt coex
+	 * activity grade allows it and we are on 2.4Ghz.
+	 */
+	if ((cam || sta_ps_disabled) &&
+	    !iwl_mvm_bt_coex_is_tpc_allowed(mvm, band))
+		return false;
+
+	IWL_DEBUG_RATE(mvm, "check rate, table type: %d\n", rate->type);
+	if (is_legacy(rate))
+		return index == IWL_RATE_54M_INDEX;
+	if (is_ht(rate))
+		return index == IWL_RATE_MCS_7_INDEX;
+	if (is_vht(rate))
+		return index == IWL_RATE_MCS_7_INDEX ||
+		       index == IWL_RATE_MCS_8_INDEX ||
+		       index == IWL_RATE_MCS_9_INDEX;
+
+	WARN_ON_ONCE(1);
+	return false;
+}
+
+enum tpc_action {
+	TPC_ACTION_STAY,
+	TPC_ACTION_DECREASE,
+	TPC_ACTION_INCREASE,
+	TPC_ACTION_NO_RESTIRCTION,
+};
+
+static enum tpc_action rs_get_tpc_action(struct iwl_mvm *mvm,
+					 s32 sr, int weak, int strong,
+					 int current_tpt,
+					 int weak_tpt, int strong_tpt)
+{
+	/* stay until we have valid tpt */
+	if (current_tpt == IWL_INVALID_VALUE) {
+		IWL_DEBUG_RATE(mvm, "no current tpt. stay.\n");
+		return TPC_ACTION_STAY;
+	}
+
+	/* Too many failures, increase txp */
+	if (sr <= TPC_SR_FORCE_INCREASE || current_tpt == 0) {
+		IWL_DEBUG_RATE(mvm, "increase txp because of weak SR\n");
+		return TPC_ACTION_NO_RESTIRCTION;
+	}
+
+	/* try decreasing first if applicable */
+	if (weak != TPC_INVALID) {
+		if (weak_tpt == IWL_INVALID_VALUE &&
+		    (strong_tpt == IWL_INVALID_VALUE ||
+		     current_tpt >= strong_tpt)) {
+			IWL_DEBUG_RATE(mvm,
+				       "no weak txp measurement. decrease txp\n");
+			return TPC_ACTION_DECREASE;
+		}
+
+		if (weak_tpt > current_tpt) {
+			IWL_DEBUG_RATE(mvm,
+				       "lower txp has better tpt. decrease txp\n");
+			return TPC_ACTION_DECREASE;
+		}
+	}
+
+	/* next, increase if needed */
+	if (sr < TPC_SR_NO_INCREASE && strong != TPC_INVALID) {
+		if (weak_tpt == IWL_INVALID_VALUE &&
+		    strong_tpt != IWL_INVALID_VALUE &&
+		    current_tpt < strong_tpt) {
+			IWL_DEBUG_RATE(mvm,
+				       "higher txp has better tpt. increase txp\n");
+			return TPC_ACTION_INCREASE;
+		}
+
+		if (weak_tpt < current_tpt &&
+		    (strong_tpt == IWL_INVALID_VALUE ||
+		     strong_tpt > current_tpt)) {
+			IWL_DEBUG_RATE(mvm,
+				       "lower txp has worse tpt. increase txp\n");
+			return TPC_ACTION_INCREASE;
+		}
+	}
+
+	IWL_DEBUG_RATE(mvm, "no need to increase or decrease txp - stay\n");
+	return TPC_ACTION_STAY;
+}
+
+static bool rs_tpc_perform(struct iwl_mvm *mvm,
+			   struct ieee80211_sta *sta,
+			   struct iwl_lq_sta *lq_sta,
+			   struct iwl_scale_tbl_info *tbl)
+{
+	struct iwl_mvm_sta *mvm_sta = (void *)sta->drv_priv;
+	struct ieee80211_vif *vif = mvm_sta->vif;
+	struct ieee80211_chanctx_conf *chanctx_conf;
+	enum ieee80211_band band;
+	struct iwl_rate_scale_data *window;
+	struct rs_rate *rate = &tbl->rate;
+	enum tpc_action action;
+	s32 sr;
+	u8 cur = lq_sta->lq.reduced_tpc;
+	int current_tpt;
+	int weak, strong;
+	int weak_tpt = IWL_INVALID_VALUE, strong_tpt = IWL_INVALID_VALUE;
+
+#ifdef CONFIG_MAC80211_DEBUGFS
+	if (lq_sta->dbg_fixed_txp_reduction <= TPC_MAX_REDUCTION) {
+		IWL_DEBUG_RATE(mvm, "fixed tpc: %d\n",
+			       lq_sta->dbg_fixed_txp_reduction);
+		lq_sta->lq.reduced_tpc = lq_sta->dbg_fixed_txp_reduction;
+		return cur != lq_sta->dbg_fixed_txp_reduction;
+	}
+#endif
+
+	rcu_read_lock();
+	chanctx_conf = rcu_dereference(vif->chanctx_conf);
+	if (WARN_ON(!chanctx_conf))
+		band = IEEE80211_NUM_BANDS;
+	else
+		band = chanctx_conf->def.chan->band;
+	rcu_read_unlock();
+
+	if (!rs_tpc_allowed(mvm, vif, rate, band)) {
+		IWL_DEBUG_RATE(mvm,
+			       "tpc is not allowed. remove txp restrictions\n");
+		lq_sta->lq.reduced_tpc = TPC_NO_REDUCTION;
+		return cur != TPC_NO_REDUCTION;
+	}
+
+	rs_get_adjacent_txp(mvm, cur, &weak, &strong);
+
+	/* Collect measured throughputs for current and adjacent rates */
+	window = tbl->tpc_win;
+	sr = window[cur].success_ratio;
+	current_tpt = window[cur].average_tpt;
+	if (weak != TPC_INVALID)
+		weak_tpt = window[weak].average_tpt;
+	if (strong != TPC_INVALID)
+		strong_tpt = window[strong].average_tpt;
+
+	IWL_DEBUG_RATE(mvm,
+		       "(TPC: %d): cur_tpt %d SR %d weak %d strong %d weak_tpt %d strong_tpt %d\n",
+		       cur, current_tpt, sr, weak, strong,
+		       weak_tpt, strong_tpt);
+
+	action = rs_get_tpc_action(mvm, sr, weak, strong,
+				   current_tpt, weak_tpt, strong_tpt);
+
+	/* override actions if we are on the edge */
+	if (weak == TPC_INVALID && action == TPC_ACTION_DECREASE) {
+		IWL_DEBUG_RATE(mvm, "already in lowest txp, stay\n");
+		action = TPC_ACTION_STAY;
+	} else if (strong == TPC_INVALID &&
+		   (action == TPC_ACTION_INCREASE ||
+		    action == TPC_ACTION_NO_RESTIRCTION)) {
+		IWL_DEBUG_RATE(mvm, "already in highest txp, stay\n");
+		action = TPC_ACTION_STAY;
+	}
+
+	switch (action) {
+	case TPC_ACTION_DECREASE:
+		lq_sta->lq.reduced_tpc = weak;
+		return true;
+	case TPC_ACTION_INCREASE:
+		lq_sta->lq.reduced_tpc = strong;
+		return true;
+	case TPC_ACTION_NO_RESTIRCTION:
+		lq_sta->lq.reduced_tpc = TPC_NO_REDUCTION;
+		return true;
+	case TPC_ACTION_STAY:
+		/* do nothing */
+		break;
+	}
+	return false;
+}
+
 /*
  * Do rate scaling and search for new modulation mode.
  */
@@ -2019,6 +2185,9 @@ static void rs_rate_scale_perform(struct iwl_mvm *mvm,
 		break;
 	case RS_ACTION_STAY:
 		/* No change */
+		if (lq_sta->rs_state == RS_STATE_STAY_IN_COLUMN)
+			update_lq = rs_tpc_perform(mvm, sta, lq_sta, tbl);
+		break;
 	default:
 		break;
 	}
@@ -2271,10 +2440,6 @@ static void rs_vht_set_enabled_rates(struct ieee80211_sta *sta,
 			if (i == IWL_RATE_9M_INDEX)
 				continue;
 
-			/* Disable MCS9 as a workaround */
-			if (i == IWL_RATE_MCS_9_INDEX)
-				continue;
-
 			/* VHT MCS9 isn't valid for 20Mhz for NSS=1,2 */
 			if (i == IWL_RATE_MCS_9_INDEX &&
 			    sta->bandwidth == IEEE80211_STA_RX_BW_20)
@@ -2293,10 +2458,6 @@ static void rs_vht_set_enabled_rates(struct ieee80211_sta *sta,
 			if (i == IWL_RATE_9M_INDEX)
 				continue;
 
-			/* Disable MCS9 as a workaround */
-			if (i == IWL_RATE_MCS_9_INDEX)
-				continue;
-
 			/* VHT MCS9 isn't valid for 20Mhz for NSS=1,2 */
 			if (i == IWL_RATE_MCS_9_INDEX &&
 			    sta->bandwidth == IEEE80211_STA_RX_BW_20)
@@ -2478,6 +2639,7 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	lq_sta->is_agg = 0;
 #ifdef CONFIG_MAC80211_DEBUGFS
 	lq_sta->dbg_fixed_rate = 0;
+	lq_sta->dbg_fixed_txp_reduction = TPC_INVALID;
 #endif
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	iwl_mvm_reset_frame_stats(mvm, &mvm->drv_rx_stats);
@@ -2653,6 +2815,7 @@ static void rs_fill_lq_cmd(struct iwl_mvm *mvm,
 		rs_build_rates_table_from_fixed(mvm, lq_cmd,
 						lq_sta->band,
 						lq_sta->dbg_fixed_rate);
+		lq_cmd->reduced_tpc = 0;
 		ant = (lq_sta->dbg_fixed_rate & RATE_MCS_ANT_ABC_MSK) >>
 			RATE_MCS_ANT_POS;
 	} else
@@ -2783,7 +2946,6 @@ static ssize_t rs_sta_dbgfs_scale_table_write(struct file *file,
 	size_t buf_size;
 	u32 parsed_rate;
 
-
 	mvm = lq_sta->drv;
 	memset(buf, 0, sizeof(buf));
 	buf_size = min(count, sizeof(buf) -  1);
@@ -2856,6 +3018,7 @@ static ssize_t rs_sta_dbgfs_scale_table_read(struct file *file,
 			lq_sta->lq.agg_disable_start_th,
 			lq_sta->lq.agg_frame_cnt_limit);
 
+	desc += sprintf(buff+desc, "reduced tpc=%d\n", lq_sta->lq.reduced_tpc);
 	desc += sprintf(buff+desc,
 			"Start idx [0]=0x%x [1]=0x%x [2]=0x%x [3]=0x%x\n",
 			lq_sta->lq.initial_rate_index[0],
@@ -2928,6 +3091,94 @@ static const struct file_operations rs_sta_dbgfs_stats_table_ops = {
 	.llseek = default_llseek,
 };
 
+static ssize_t rs_sta_dbgfs_drv_tx_stats_read(struct file *file,
+					      char __user *user_buf,
+					      size_t count, loff_t *ppos)
+{
+	static const char * const column_name[] = {
+		[RS_COLUMN_LEGACY_ANT_A] = "LEGACY_ANT_A",
+		[RS_COLUMN_LEGACY_ANT_B] = "LEGACY_ANT_B",
+		[RS_COLUMN_SISO_ANT_A] = "SISO_ANT_A",
+		[RS_COLUMN_SISO_ANT_B] = "SISO_ANT_B",
+		[RS_COLUMN_SISO_ANT_A_SGI] = "SISO_ANT_A_SGI",
+		[RS_COLUMN_SISO_ANT_B_SGI] = "SISO_ANT_B_SGI",
+		[RS_COLUMN_MIMO2] = "MIMO2",
+		[RS_COLUMN_MIMO2_SGI] = "MIMO2_SGI",
+	};
+
+	static const char * const rate_name[] = {
+		[IWL_RATE_1M_INDEX] = "1M",
+		[IWL_RATE_2M_INDEX] = "2M",
+		[IWL_RATE_5M_INDEX] = "5.5M",
+		[IWL_RATE_11M_INDEX] = "11M",
+		[IWL_RATE_6M_INDEX] = "6M|MCS0",
+		[IWL_RATE_9M_INDEX] = "9M",
+		[IWL_RATE_12M_INDEX] = "12M|MCS1",
+		[IWL_RATE_18M_INDEX] = "18M|MCS2",
+		[IWL_RATE_24M_INDEX] = "24M|MCS3",
+		[IWL_RATE_36M_INDEX] = "36M|MCS4",
+		[IWL_RATE_48M_INDEX] = "48M|MCS5",
+		[IWL_RATE_54M_INDEX] = "54M|MCS6",
+		[IWL_RATE_MCS_7_INDEX] = "MCS7",
+		[IWL_RATE_MCS_8_INDEX] = "MCS8",
+		[IWL_RATE_MCS_9_INDEX] = "MCS9",
+	};
+
+	char *buff, *pos, *endpos;
+	int col, rate;
+	ssize_t ret;
+	struct iwl_lq_sta *lq_sta = file->private_data;
+	struct rs_rate_stats *stats;
+	static const size_t bufsz = 1024;
+
+	buff = kmalloc(bufsz, GFP_KERNEL);
+	if (!buff)
+		return -ENOMEM;
+
+	pos = buff;
+	endpos = pos + bufsz;
+
+	pos += scnprintf(pos, endpos - pos, "COLUMN,");
+	for (rate = 0; rate < IWL_RATE_COUNT; rate++)
+		pos += scnprintf(pos, endpos - pos, "%s,", rate_name[rate]);
+	pos += scnprintf(pos, endpos - pos, "\n");
+
+	for (col = 0; col < RS_COLUMN_COUNT; col++) {
+		pos += scnprintf(pos, endpos - pos,
+				 "%s,", column_name[col]);
+
+		for (rate = 0; rate < IWL_RATE_COUNT; rate++) {
+			stats = &(lq_sta->tx_stats[col][rate]);
+			pos += scnprintf(pos, endpos - pos,
+					 "%llu/%llu,",
+					 stats->success,
+					 stats->total);
+		}
+		pos += scnprintf(pos, endpos - pos, "\n");
+	}
+
+	ret = simple_read_from_buffer(user_buf, count, ppos, buff, pos - buff);
+	kfree(buff);
+	return ret;
+}
+
+static ssize_t rs_sta_dbgfs_drv_tx_stats_write(struct file *file,
+					       const char __user *user_buf,
+					       size_t count, loff_t *ppos)
+{
+	struct iwl_lq_sta *lq_sta = file->private_data;
+	memset(lq_sta->tx_stats, 0, sizeof(lq_sta->tx_stats));
+
+	return count;
+}
+
+static const struct file_operations rs_sta_dbgfs_drv_tx_stats_ops = {
+	.read = rs_sta_dbgfs_drv_tx_stats_read,
+	.write = rs_sta_dbgfs_drv_tx_stats_write,
+	.open = simple_open,
+	.llseek = default_llseek,
+};
+
 static void rs_add_debugfs(void *mvm, void *mvm_sta, struct dentry *dir)
 {
 	struct iwl_lq_sta *lq_sta = mvm_sta;
@@ -2937,9 +3188,15 @@ static void rs_add_debugfs(void *mvm, void *mvm_sta, struct dentry *dir)
 	lq_sta->rs_sta_dbgfs_stats_table_file =
 		debugfs_create_file("rate_stats_table", S_IRUSR, dir,
 				    lq_sta, &rs_sta_dbgfs_stats_table_ops);
+	lq_sta->rs_sta_dbgfs_drv_tx_stats_file =
+		debugfs_create_file("drv_tx_stats", S_IRUSR | S_IWUSR, dir,
+				    lq_sta, &rs_sta_dbgfs_drv_tx_stats_ops);
 	lq_sta->rs_sta_dbgfs_tx_agg_tid_en_file =
 		debugfs_create_u8("tx_agg_tid_enable", S_IRUSR | S_IWUSR, dir,
 				  &lq_sta->tx_agg_tid_en);
+	lq_sta->rs_sta_dbgfs_reduced_txp_file =
+		debugfs_create_u8("reduced_tpc", S_IRUSR | S_IWUSR, dir,
+				  &lq_sta->dbg_fixed_txp_reduction);
 }
 
 static void rs_remove_debugfs(void *mvm, void *mvm_sta)
@@ -2947,7 +3204,9 @@ static void rs_remove_debugfs(void *mvm, void *mvm_sta)
 	struct iwl_lq_sta *lq_sta = mvm_sta;
 	debugfs_remove(lq_sta->rs_sta_dbgfs_scale_table_file);
 	debugfs_remove(lq_sta->rs_sta_dbgfs_stats_table_file);
+	debugfs_remove(lq_sta->rs_sta_dbgfs_drv_tx_stats_file);
 	debugfs_remove(lq_sta->rs_sta_dbgfs_tx_agg_tid_en_file);
+	debugfs_remove(lq_sta->rs_sta_dbgfs_reduced_txp_file);
 }
 #endif
 
diff --git a/drivers/net/wireless/iwlwifi/mvm/rs.h b/drivers/net/wireless/iwlwifi/mvm/rs.h
index 0acfac96a56c..374a83d7db25 100644
--- a/drivers/net/wireless/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/iwlwifi/mvm/rs.h
@@ -158,6 +158,13 @@ enum {
 #define RS_SR_FORCE_DECREASE		1920	/*  15% */
 #define RS_SR_NO_DECREASE		10880	/*  85% */
 
+#define TPC_SR_FORCE_INCREASE		9600	/* 75% */
+#define TPC_SR_NO_INCREASE		10880	/* 85% */
+#define TPC_TX_POWER_STEP		3
+#define TPC_MAX_REDUCTION		15
+#define TPC_NO_REDUCTION		0
+#define TPC_INVALID			0xff
+
 #define LINK_QUAL_AGG_TIME_LIMIT_DEF	(4000) /* 4 milliseconds */
 #define LINK_QUAL_AGG_TIME_LIMIT_MAX	(8000)
 #define LINK_QUAL_AGG_TIME_LIMIT_MIN	(100)
@@ -266,9 +273,16 @@ enum rs_column {
 	RS_COLUMN_MIMO2_SGI,
 
 	RS_COLUMN_LAST = RS_COLUMN_MIMO2_SGI,
+	RS_COLUMN_COUNT = RS_COLUMN_LAST + 1,
 	RS_COLUMN_INVALID,
 };
 
+/* Packet stats per rate */
+struct rs_rate_stats {
+	u64 success;
+	u64 total;
+};
+
 /**
  * struct iwl_scale_tbl_info -- tx params and success history for all rates
  *
@@ -280,6 +294,8 @@ struct iwl_scale_tbl_info {
 	enum rs_column column;
 	const u16 *expected_tpt;	/* throughput metrics; expected_tpt_G, etc. */
 	struct iwl_rate_scale_data win[IWL_RATE_COUNT]; /* rate histories */
+	/* per txpower-reduction history */
+	struct iwl_rate_scale_data tpc_win[TPC_MAX_REDUCTION + 1];
 };
 
 enum {
@@ -315,6 +331,8 @@ struct iwl_lq_sta {
 	bool is_vht;
 	enum ieee80211_band band;
 
+	struct rs_rate_stats tx_stats[RS_COLUMN_COUNT][IWL_RATE_COUNT];
+
 	/* The following are bitmaps of rates; IWL_RATE_6M_MASK, etc. */
 	unsigned long active_legacy_rate;
 	unsigned long active_siso_rate;
@@ -334,8 +352,11 @@ struct iwl_lq_sta {
 #ifdef CONFIG_MAC80211_DEBUGFS
 	struct dentry *rs_sta_dbgfs_scale_table_file;
 	struct dentry *rs_sta_dbgfs_stats_table_file;
+	struct dentry *rs_sta_dbgfs_drv_tx_stats_file;
 	struct dentry *rs_sta_dbgfs_tx_agg_tid_en_file;
+	struct dentry *rs_sta_dbgfs_reduced_txp_file;
 	u32 dbg_fixed_rate;
+	u8 dbg_fixed_txp_reduction;
 #endif
 	struct iwl_mvm *drv;
 
@@ -345,6 +366,9 @@ struct iwl_lq_sta {
 	u32 last_rate_n_flags;
 	/* packets destined for this STA are aggregated */
 	u8 is_agg;
+
+	/* tx power reduce for this sta */
+	int tpc_reduce;
 };
 
 /* Initialize station's rate scaling information after adding station */
diff --git a/drivers/net/wireless/iwlwifi/mvm/rx.c b/drivers/net/wireless/iwlwifi/mvm/rx.c
index 6061553a5e44..cf7276967acd 100644
--- a/drivers/net/wireless/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/iwlwifi/mvm/rx.c
@@ -60,7 +60,6 @@
  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  *****************************************************************************/
 #include "iwl-trans.h"
-
 #include "mvm.h"
 #include "fw-api.h"
 
@@ -130,42 +129,7 @@ static void iwl_mvm_pass_packet_to_mac80211(struct iwl_mvm *mvm,
 
 	memcpy(IEEE80211_SKB_RXCB(skb), stats, sizeof(*stats));
 
-	ieee80211_rx_ni(mvm->hw, skb);
-}
-
-static void iwl_mvm_calc_rssi(struct iwl_mvm *mvm,
-			      struct iwl_rx_phy_info *phy_info,
-			      struct ieee80211_rx_status *rx_status)
-{
-	int rssi_a, rssi_b, rssi_a_dbm, rssi_b_dbm, max_rssi_dbm;
-	u32 agc_a, agc_b;
-	u32 val;
-
-	val = le32_to_cpu(phy_info->non_cfg_phy[IWL_RX_INFO_AGC_IDX]);
-	agc_a = (val & IWL_OFDM_AGC_A_MSK) >> IWL_OFDM_AGC_A_POS;
-	agc_b = (val & IWL_OFDM_AGC_B_MSK) >> IWL_OFDM_AGC_B_POS;
-
-	val = le32_to_cpu(phy_info->non_cfg_phy[IWL_RX_INFO_RSSI_AB_IDX]);
-	rssi_a = (val & IWL_OFDM_RSSI_INBAND_A_MSK) >> IWL_OFDM_RSSI_A_POS;
-	rssi_b = (val & IWL_OFDM_RSSI_INBAND_B_MSK) >> IWL_OFDM_RSSI_B_POS;
-
-	/*
-	 * dBm = rssi dB - agc dB - constant.
-	 * Higher AGC (higher radio gain) means lower signal.
-	 */
-	rssi_a_dbm = rssi_a - IWL_RSSI_OFFSET - agc_a;
-	rssi_b_dbm = rssi_b - IWL_RSSI_OFFSET - agc_b;
-	max_rssi_dbm = max_t(int, rssi_a_dbm, rssi_b_dbm);
-
-	IWL_DEBUG_STATS(mvm, "Rssi In A %d B %d Max %d AGCA %d AGCB %d\n",
-			rssi_a_dbm, rssi_b_dbm, max_rssi_dbm, agc_a, agc_b);
-
-	rx_status->signal = max_rssi_dbm;
-	rx_status->chains = (le16_to_cpu(phy_info->phy_flags) &
-				RX_RES_PHY_FLAGS_ANTENNA)
-					>> RX_RES_PHY_FLAGS_ANTENNA_POS;
-	rx_status->chain_signal[0] = rssi_a_dbm;
-	rx_status->chain_signal[1] = rssi_b_dbm;
+	ieee80211_rx(mvm->hw, skb);
 }
 
 /*
@@ -337,10 +301,7 @@ int iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb,
 	 */
 	/*rx_status.flag |= RX_FLAG_MACTIME_MPDU;*/
 
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_RX_ENERGY_API)
-		iwl_mvm_get_signal_strength(mvm, phy_info, &rx_status);
-	else
-		iwl_mvm_calc_rssi(mvm, phy_info, &rx_status);
+	iwl_mvm_get_signal_strength(mvm, phy_info, &rx_status);
 
 	IWL_DEBUG_STATS_LIMIT(mvm, "Rssi %d, TSF %llu\n", rx_status.signal,
 			      (unsigned long long)rx_status.mactime);
@@ -394,6 +355,8 @@ int iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb,
 		rx_status.rate_idx = rate_n_flags & RATE_VHT_MCS_RATE_CODE_MSK;
 		rx_status.flag |= RX_FLAG_VHT;
 		rx_status.flag |= stbc << RX_FLAG_STBC_SHIFT;
+		if (rate_n_flags & RATE_MCS_BF_MSK)
+			rx_status.vht_flag |= RX_VHT_FLAG_BF;
 	} else {
 		rx_status.rate_idx =
 			iwl_mvm_legacy_rate_to_mac80211_idx(rate_n_flags,
diff --git a/drivers/net/wireless/iwlwifi/mvm/scan.c b/drivers/net/wireless/iwlwifi/mvm/scan.c
index c28de54c75d4..4b6c7d4bd199 100644
--- a/drivers/net/wireless/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/iwlwifi/mvm/scan.c
@@ -306,7 +306,6 @@ int iwl_mvm_scan_request(struct iwl_mvm *mvm,
 		.id = SCAN_REQUEST_CMD,
 		.len = { 0, },
 		.data = { mvm->scan_cmd, },
-		.flags = CMD_SYNC,
 		.dataflags = { IWL_HCMD_DFL_NOCOPY, },
 	};
 	struct iwl_scan_cmd *cmd = mvm->scan_cmd;
@@ -319,7 +318,10 @@ int iwl_mvm_scan_request(struct iwl_mvm *mvm,
 	struct iwl_mvm_scan_params params = {};
 
 	lockdep_assert_held(&mvm->mutex);
-	BUG_ON(mvm->scan_cmd == NULL);
+
+	/* we should have failed registration if scan_cmd was NULL */
+	if (WARN_ON(mvm->scan_cmd == NULL))
+		return -ENOMEM;
 
 	IWL_DEBUG_SCAN(mvm, "Handling mac80211 scan request\n");
 	mvm->scan_status = IWL_MVM_SCAN_OS;
@@ -514,7 +516,7 @@ int iwl_mvm_cancel_scan(struct iwl_mvm *mvm)
 				   ARRAY_SIZE(scan_abort_notif),
 				   iwl_mvm_scan_abort_notif, NULL);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, SCAN_ABORT_CMD, CMD_SYNC, 0, NULL);
+	ret = iwl_mvm_send_cmd_pdu(mvm, SCAN_ABORT_CMD, 0, 0, NULL);
 	if (ret) {
 		IWL_ERR(mvm, "Couldn't send SCAN_ABORT_CMD: %d\n", ret);
 		/* mac80211's state will be cleaned in the nic_restart flow */
@@ -538,9 +540,13 @@ int iwl_mvm_rx_scan_offload_complete_notif(struct iwl_mvm *mvm,
 	/* scan status must be locked for proper checking */
 	lockdep_assert_held(&mvm->mutex);
 
-	IWL_DEBUG_SCAN(mvm, "Scheduled scan completed, status %s\n",
+	IWL_DEBUG_SCAN(mvm,
+		       "Scheduled scan completed, status %s EBS status %s:%d\n",
 		       scan_notif->status == IWL_SCAN_OFFLOAD_COMPLETED ?
-		       "completed" : "aborted");
+		       "completed" : "aborted", scan_notif->ebs_status ==
+		       IWL_SCAN_EBS_SUCCESS ? "success" : "failed",
+		       scan_notif->ebs_status);
+
 
 	/* only call mac80211 completion if the stop was initiated by FW */
 	if (mvm->scan_status == IWL_MVM_SCAN_SCHED) {
@@ -548,6 +554,8 @@ int iwl_mvm_rx_scan_offload_complete_notif(struct iwl_mvm *mvm,
 		ieee80211_sched_scan_stopped(mvm->hw);
 	}
 
+	mvm->last_ebs_successful = !scan_notif->ebs_status;
+
 	return 0;
 }
 
@@ -740,7 +748,6 @@ int iwl_mvm_config_sched_scan(struct iwl_mvm *mvm,
 	struct iwl_scan_offload_cfg *scan_cfg;
 	struct iwl_host_cmd cmd = {
 		.id = SCAN_OFFLOAD_CONFIG_CMD,
-		.flags = CMD_SYNC,
 	};
 	struct iwl_mvm_scan_params params = {};
 
@@ -798,7 +805,6 @@ int iwl_mvm_config_sched_scan_profiles(struct iwl_mvm *mvm,
 	struct iwl_scan_offload_blacklist *blacklist;
 	struct iwl_host_cmd cmd = {
 		.id = SCAN_OFFLOAD_UPDATE_PROFILES_CMD,
-		.flags = CMD_SYNC,
 		.len[1] = sizeof(*profile_cfg),
 		.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
 		.dataflags[1] = IWL_HCMD_DFL_NOCOPY,
@@ -884,7 +890,12 @@ int iwl_mvm_sched_scan_start(struct iwl_mvm *mvm,
 		scan_req.flags |= cpu_to_le16(IWL_SCAN_OFFLOAD_FLAG_PASS_ALL);
 	}
 
-	return iwl_mvm_send_cmd_pdu(mvm, SCAN_OFFLOAD_REQUEST_CMD, CMD_SYNC,
+	if (mvm->last_ebs_successful &&
+	    mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_EBS_SUPPORT)
+		scan_req.flags |=
+			cpu_to_le16(IWL_SCAN_OFFLOAD_FLAG_EBS_ACCURATE_MODE);
+
+	return iwl_mvm_send_cmd_pdu(mvm, SCAN_OFFLOAD_REQUEST_CMD, 0,
 				    sizeof(scan_req), &scan_req);
 }
 
@@ -893,7 +904,6 @@ static int iwl_mvm_send_sched_scan_abort(struct iwl_mvm *mvm)
 	int ret;
 	struct iwl_host_cmd cmd = {
 		.id = SCAN_OFFLOAD_ABORT_CMD,
-		.flags = CMD_SYNC,
 	};
 	u32 status;
 
@@ -922,7 +932,7 @@ static int iwl_mvm_send_sched_scan_abort(struct iwl_mvm *mvm)
 	return ret;
 }
 
-int iwl_mvm_sched_scan_stop(struct iwl_mvm *mvm)
+int iwl_mvm_sched_scan_stop(struct iwl_mvm *mvm, bool notify)
 {
 	int ret;
 	struct iwl_notification_wait wait_scan_done;
@@ -960,5 +970,8 @@ int iwl_mvm_sched_scan_stop(struct iwl_mvm *mvm)
 	 */
 	mvm->scan_status = IWL_MVM_SCAN_NONE;
 
+	if (notify)
+		ieee80211_sched_scan_stopped(mvm->hw);
+
 	return 0;
 }
diff --git a/drivers/net/wireless/iwlwifi/mvm/sf.c b/drivers/net/wireless/iwlwifi/mvm/sf.c
index 88809b2d1654..7edfd15efc9d 100644
--- a/drivers/net/wireless/iwlwifi/mvm/sf.c
+++ b/drivers/net/wireless/iwlwifi/mvm/sf.c
@@ -237,9 +237,6 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 		.sta_vif_ap_sta_id = IWL_MVM_STATION_COUNT,
 	};
 
-	if (IWL_UCODE_API(mvm->fw->ucode_ver) < 8)
-		return 0;
-
 	/*
 	 * Ignore the call if we are in HW Restart flow, or if the handled
 	 * vif is a p2p device.
diff --git a/drivers/net/wireless/iwlwifi/mvm/sta.c b/drivers/net/wireless/iwlwifi/mvm/sta.c
index f339ef884250..1fb01ea2e704 100644
--- a/drivers/net/wireless/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/iwlwifi/mvm/sta.c
@@ -66,115 +66,6 @@
 #include "sta.h"
 #include "rs.h"
 
-static void iwl_mvm_add_sta_cmd_v7_to_v5(struct iwl_mvm_add_sta_cmd_v7 *cmd_v7,
-					 struct iwl_mvm_add_sta_cmd_v5 *cmd_v5)
-{
-	memset(cmd_v5, 0, sizeof(*cmd_v5));
-
-	cmd_v5->add_modify = cmd_v7->add_modify;
-	cmd_v5->tid_disable_tx = cmd_v7->tid_disable_tx;
-	cmd_v5->mac_id_n_color = cmd_v7->mac_id_n_color;
-	memcpy(cmd_v5->addr, cmd_v7->addr, ETH_ALEN);
-	cmd_v5->sta_id = cmd_v7->sta_id;
-	cmd_v5->modify_mask = cmd_v7->modify_mask;
-	cmd_v5->station_flags = cmd_v7->station_flags;
-	cmd_v5->station_flags_msk = cmd_v7->station_flags_msk;
-	cmd_v5->add_immediate_ba_tid = cmd_v7->add_immediate_ba_tid;
-	cmd_v5->remove_immediate_ba_tid = cmd_v7->remove_immediate_ba_tid;
-	cmd_v5->add_immediate_ba_ssn = cmd_v7->add_immediate_ba_ssn;
-	cmd_v5->sleep_tx_count = cmd_v7->sleep_tx_count;
-	cmd_v5->sleep_state_flags = cmd_v7->sleep_state_flags;
-	cmd_v5->assoc_id = cmd_v7->assoc_id;
-	cmd_v5->beamform_flags = cmd_v7->beamform_flags;
-	cmd_v5->tfd_queue_msk = cmd_v7->tfd_queue_msk;
-}
-
-static void
-iwl_mvm_add_sta_key_to_add_sta_cmd_v5(struct iwl_mvm_add_sta_key_cmd *key_cmd,
-				      struct iwl_mvm_add_sta_cmd_v5 *sta_cmd,
-				      u32 mac_id_n_color)
-{
-	memset(sta_cmd, 0, sizeof(*sta_cmd));
-
-	sta_cmd->sta_id = key_cmd->sta_id;
-	sta_cmd->add_modify = STA_MODE_MODIFY;
-	sta_cmd->modify_mask = STA_MODIFY_KEY;
-	sta_cmd->mac_id_n_color = cpu_to_le32(mac_id_n_color);
-
-	sta_cmd->key.key_offset = key_cmd->key_offset;
-	sta_cmd->key.key_flags = key_cmd->key_flags;
-	memcpy(sta_cmd->key.key, key_cmd->key, sizeof(sta_cmd->key.key));
-	sta_cmd->key.tkip_rx_tsc_byte2 = key_cmd->tkip_rx_tsc_byte2;
-	memcpy(sta_cmd->key.tkip_rx_ttak, key_cmd->tkip_rx_ttak,
-	       sizeof(sta_cmd->key.tkip_rx_ttak));
-}
-
-static int iwl_mvm_send_add_sta_cmd_status(struct iwl_mvm *mvm,
-					   struct iwl_mvm_add_sta_cmd_v7 *cmd,
-					   int *status)
-{
-	struct iwl_mvm_add_sta_cmd_v5 cmd_v5;
-
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_STA_KEY_CMD)
-		return iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(*cmd),
-						   cmd, status);
-
-	iwl_mvm_add_sta_cmd_v7_to_v5(cmd, &cmd_v5);
-
-	return iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd_v5),
-					   &cmd_v5, status);
-}
-
-static int iwl_mvm_send_add_sta_cmd(struct iwl_mvm *mvm, u32 flags,
-				    struct iwl_mvm_add_sta_cmd_v7 *cmd)
-{
-	struct iwl_mvm_add_sta_cmd_v5 cmd_v5;
-
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_STA_KEY_CMD)
-		return iwl_mvm_send_cmd_pdu(mvm, ADD_STA, flags,
-					    sizeof(*cmd), cmd);
-
-	iwl_mvm_add_sta_cmd_v7_to_v5(cmd, &cmd_v5);
-
-	return iwl_mvm_send_cmd_pdu(mvm, ADD_STA, flags, sizeof(cmd_v5),
-				    &cmd_v5);
-}
-
-static int
-iwl_mvm_send_add_sta_key_cmd_status(struct iwl_mvm *mvm,
-				    struct iwl_mvm_add_sta_key_cmd *cmd,
-				    u32 mac_id_n_color,
-				    int *status)
-{
-	struct iwl_mvm_add_sta_cmd_v5 sta_cmd;
-
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_STA_KEY_CMD)
-		return iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA_KEY,
-						   sizeof(*cmd), cmd, status);
-
-	iwl_mvm_add_sta_key_to_add_sta_cmd_v5(cmd, &sta_cmd, mac_id_n_color);
-
-	return iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(sta_cmd),
-					   &sta_cmd, status);
-}
-
-static int iwl_mvm_send_add_sta_key_cmd(struct iwl_mvm *mvm,
-					u32 flags,
-					struct iwl_mvm_add_sta_key_cmd *cmd,
-					u32 mac_id_n_color)
-{
-	struct iwl_mvm_add_sta_cmd_v5 sta_cmd;
-
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_STA_KEY_CMD)
-		return iwl_mvm_send_cmd_pdu(mvm, ADD_STA_KEY, flags,
-					    sizeof(*cmd), cmd);
-
-	iwl_mvm_add_sta_key_to_add_sta_cmd_v5(cmd, &sta_cmd, mac_id_n_color);
-
-	return iwl_mvm_send_cmd_pdu(mvm, ADD_STA, flags, sizeof(sta_cmd),
-				    &sta_cmd);
-}
-
 static int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm,
 				    enum nl80211_iftype iftype)
 {
@@ -207,7 +98,7 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			   bool update)
 {
 	struct iwl_mvm_sta *mvm_sta = (void *)sta->drv_priv;
-	struct iwl_mvm_add_sta_cmd_v7 add_sta_cmd;
+	struct iwl_mvm_add_sta_cmd add_sta_cmd;
 	int ret;
 	u32 status;
 	u32 agg_size = 0, mpdu_dens = 0;
@@ -295,7 +186,8 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		cpu_to_le32(mpdu_dens << STA_FLG_AGG_MPDU_DENS_SHIFT);
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_add_sta_cmd_status(mvm, &add_sta_cmd, &status);
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(add_sta_cmd),
+					  &add_sta_cmd, &status);
 	if (ret)
 		return ret;
 
@@ -380,7 +272,7 @@ int iwl_mvm_update_sta(struct iwl_mvm *mvm,
 int iwl_mvm_drain_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 		      bool drain)
 {
-	struct iwl_mvm_add_sta_cmd_v7 cmd = {};
+	struct iwl_mvm_add_sta_cmd cmd = {};
 	int ret;
 	u32 status;
 
@@ -393,7 +285,8 @@ int iwl_mvm_drain_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 	cmd.station_flags_msk = cpu_to_le32(STA_FLG_DRAIN_FLOW);
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_add_sta_cmd_status(mvm, &cmd, &status);
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+					  &cmd, &status);
 	if (ret)
 		return ret;
 
@@ -434,7 +327,7 @@ static int iwl_mvm_rm_sta_common(struct iwl_mvm *mvm, u8 sta_id)
 		return -EINVAL;
 	}
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, REMOVE_STA, CMD_SYNC,
+	ret = iwl_mvm_send_cmd_pdu(mvm, REMOVE_STA, 0,
 				   sizeof(rm_sta_cmd), &rm_sta_cmd);
 	if (ret) {
 		IWL_ERR(mvm, "Failed to remove station. Id=%d\n", sta_id);
@@ -498,7 +391,7 @@ void iwl_mvm_sta_drained_wk(struct work_struct *wk)
 				sta_id);
 			continue;
 		}
-		rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], NULL);
+		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta_id], NULL);
 		clear_bit(sta_id, mvm->sta_drained);
 	}
 
@@ -520,14 +413,6 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 		/* flush its queues here since we are freeing mvm_sta */
 		ret = iwl_mvm_flush_tx_path(mvm, mvm_sta->tfd_queue_msk, true);
 
-		/*
-		 * Put a non-NULL since the fw station isn't removed.
-		 * It will be removed after the MAC will be set as
-		 * unassoc.
-		 */
-		rcu_assign_pointer(mvm->fw_id_to_mac_id[mvm_sta->sta_id],
-				   ERR_PTR(-EINVAL));
-
 		/* if we are associated - we can't remove the AP STA now */
 		if (vif->bss_conf.assoc)
 			return ret;
@@ -557,7 +442,7 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 	} else {
 		spin_unlock_bh(&mvm_sta->lock);
 		ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->sta_id);
-		rcu_assign_pointer(mvm->fw_id_to_mac_id[mvm_sta->sta_id], NULL);
+		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->sta_id], NULL);
 	}
 
 	return ret;
@@ -571,7 +456,7 @@ int iwl_mvm_rm_sta_id(struct iwl_mvm *mvm,
 
 	lockdep_assert_held(&mvm->mutex);
 
-	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], NULL);
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta_id], NULL);
 	return ret;
 }
 
@@ -593,7 +478,7 @@ int iwl_mvm_allocate_int_sta(struct iwl_mvm *mvm, struct iwl_mvm_int_sta *sta,
 
 void iwl_mvm_dealloc_int_sta(struct iwl_mvm *mvm, struct iwl_mvm_int_sta *sta)
 {
-	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta->sta_id], NULL);
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta->sta_id], NULL);
 	memset(sta, 0, sizeof(struct iwl_mvm_int_sta));
 	sta->sta_id = IWL_MVM_STATION_COUNT;
 }
@@ -603,13 +488,13 @@ static int iwl_mvm_add_int_sta_common(struct iwl_mvm *mvm,
 				      const u8 *addr,
 				      u16 mac_id, u16 color)
 {
-	struct iwl_mvm_add_sta_cmd_v7 cmd;
+	struct iwl_mvm_add_sta_cmd cmd;
 	int ret;
 	u32 status;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	memset(&cmd, 0, sizeof(struct iwl_mvm_add_sta_cmd_v7));
+	memset(&cmd, 0, sizeof(cmd));
 	cmd.sta_id = sta->sta_id;
 	cmd.mac_id_n_color = cpu_to_le32(FW_CMD_ID_AND_COLOR(mac_id,
 							     color));
@@ -619,7 +504,8 @@ static int iwl_mvm_add_int_sta_common(struct iwl_mvm *mvm,
 	if (addr)
 		memcpy(cmd.addr, addr, ETH_ALEN);
 
-	ret = iwl_mvm_send_add_sta_cmd_status(mvm, &cmd, &status);
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+					  &cmd, &status);
 	if (ret)
 		return ret;
 
@@ -753,7 +639,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		       int tid, u16 ssn, bool start)
 {
 	struct iwl_mvm_sta *mvm_sta = (void *)sta->drv_priv;
-	struct iwl_mvm_add_sta_cmd_v7 cmd = {};
+	struct iwl_mvm_add_sta_cmd cmd = {};
 	int ret;
 	u32 status;
 
@@ -777,7 +663,8 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 				  STA_MODIFY_REMOVE_BA_TID;
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_add_sta_cmd_status(mvm, &cmd, &status);
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+					  &cmd, &status);
 	if (ret)
 		return ret;
 
@@ -812,7 +699,7 @@ static int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			      int tid, u8 queue, bool start)
 {
 	struct iwl_mvm_sta *mvm_sta = (void *)sta->drv_priv;
-	struct iwl_mvm_add_sta_cmd_v7 cmd = {};
+	struct iwl_mvm_add_sta_cmd cmd = {};
 	int ret;
 	u32 status;
 
@@ -834,7 +721,8 @@ static int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	cmd.tid_disable_tx = cpu_to_le16(mvm_sta->tid_disable_agg);
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_add_sta_cmd_status(mvm, &cmd, &status);
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+					  &cmd, &status);
 	if (ret)
 		return ret;
 
@@ -1129,12 +1017,11 @@ static int iwl_mvm_send_sta_key(struct iwl_mvm *mvm,
 				u8 sta_id, u32 tkip_iv32, u16 *tkip_p1k,
 				u32 cmd_flags)
 {
-	__le16 key_flags;
 	struct iwl_mvm_add_sta_key_cmd cmd = {};
+	__le16 key_flags;
 	int ret, status;
 	u16 keyidx;
 	int i;
-	u32 mac_id_n_color = mvm_sta->mac_id_n_color;
 
 	keyidx = (keyconf->keyidx << STA_KEY_FLG_KEYID_POS) &
 		 STA_KEY_FLG_KEYID_MSK;
@@ -1166,13 +1053,12 @@ static int iwl_mvm_send_sta_key(struct iwl_mvm *mvm,
 	cmd.sta_id = sta_id;
 
 	status = ADD_STA_SUCCESS;
-	if (cmd_flags == CMD_SYNC)
-		ret = iwl_mvm_send_add_sta_key_cmd_status(mvm, &cmd,
-							  mac_id_n_color,
-							  &status);
+	if (cmd_flags & CMD_ASYNC)
+		ret =  iwl_mvm_send_cmd_pdu(mvm, ADD_STA_KEY, CMD_ASYNC,
+					    sizeof(cmd), &cmd);
 	else
-		ret = iwl_mvm_send_add_sta_key_cmd(mvm, CMD_ASYNC, &cmd,
-						   mac_id_n_color);
+		ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA_KEY, sizeof(cmd),
+						  &cmd, &status);
 
 	switch (status) {
 	case ADD_STA_SUCCESS:
@@ -1225,7 +1111,7 @@ static int iwl_mvm_send_sta_igtk(struct iwl_mvm *mvm,
 		       remove_key ? "removing" : "installing",
 		       igtk_cmd.sta_id);
 
-	return iwl_mvm_send_cmd_pdu(mvm, MGMT_MCAST_KEY, CMD_SYNC,
+	return iwl_mvm_send_cmd_pdu(mvm, MGMT_MCAST_KEY, 0,
 				    sizeof(igtk_cmd), &igtk_cmd);
 }
 
@@ -1312,15 +1198,15 @@ int iwl_mvm_set_sta_key(struct iwl_mvm *mvm,
 		ieee80211_get_key_rx_seq(keyconf, 0, &seq);
 		ieee80211_get_tkip_rx_p1k(keyconf, addr, seq.tkip.iv32, p1k);
 		ret = iwl_mvm_send_sta_key(mvm, mvm_sta, keyconf, sta_id,
-					   seq.tkip.iv32, p1k, CMD_SYNC);
+					   seq.tkip.iv32, p1k, 0);
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 		ret = iwl_mvm_send_sta_key(mvm, mvm_sta, keyconf, sta_id,
-					   0, NULL, CMD_SYNC);
+					   0, NULL, 0);
 		break;
 	default:
 		ret = iwl_mvm_send_sta_key(mvm, mvm_sta, keyconf,
-					   sta_id, 0, NULL, CMD_SYNC);
+					   sta_id, 0, NULL, 0);
 	}
 
 	if (ret)
@@ -1399,9 +1285,8 @@ int iwl_mvm_remove_sta_key(struct iwl_mvm *mvm,
 	cmd.sta_id = sta_id;
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_add_sta_key_cmd_status(mvm, &cmd,
-						  mvm_sta->mac_id_n_color,
-						  &status);
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA_KEY, sizeof(cmd),
+					  &cmd, &status);
 
 	switch (status) {
 	case ADD_STA_SUCCESS:
@@ -1448,7 +1333,7 @@ void iwl_mvm_sta_modify_ps_wake(struct iwl_mvm *mvm,
 				struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_mvm_add_sta_cmd_v7 cmd = {
+	struct iwl_mvm_add_sta_cmd cmd = {
 		.add_modify = STA_MODE_MODIFY,
 		.sta_id = mvmsta->sta_id,
 		.station_flags_msk = cpu_to_le32(STA_FLG_PS),
@@ -1456,7 +1341,7 @@ void iwl_mvm_sta_modify_ps_wake(struct iwl_mvm *mvm,
 	};
 	int ret;
 
-	ret = iwl_mvm_send_add_sta_cmd(mvm, CMD_ASYNC, &cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC, sizeof(cmd), &cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send ADD_STA command (%d)\n", ret);
 }
@@ -1468,7 +1353,7 @@ void iwl_mvm_sta_modify_sleep_tx_count(struct iwl_mvm *mvm,
 				       bool agg)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_mvm_add_sta_cmd_v7 cmd = {
+	struct iwl_mvm_add_sta_cmd cmd = {
 		.add_modify = STA_MODE_MODIFY,
 		.sta_id = mvmsta->sta_id,
 		.modify_mask = STA_MODIFY_SLEEPING_STA_TX_COUNT,
@@ -1538,7 +1423,7 @@ void iwl_mvm_sta_modify_sleep_tx_count(struct iwl_mvm *mvm,
 		cmd.sleep_state_flags |= cpu_to_le16(STA_SLEEP_STATE_UAPSD);
 	}
 
-	ret = iwl_mvm_send_add_sta_cmd(mvm, CMD_ASYNC, &cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC, sizeof(cmd), &cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send ADD_STA command (%d)\n", ret);
 }
diff --git a/drivers/net/wireless/iwlwifi/mvm/sta.h b/drivers/net/wireless/iwlwifi/mvm/sta.h
index 2ed84c421481..d98e8a2142b8 100644
--- a/drivers/net/wireless/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/iwlwifi/mvm/sta.h
@@ -253,6 +253,8 @@ enum iwl_mvm_agg_state {
  *	This is basically (last acked packet++).
  * @rate_n_flags: Rate at which Tx was attempted. Holds the data between the
  *	Tx response (TX_CMD), and the block ack notification (COMPRESSED_BA).
+ * @reduced_tpc: Reduced tx power. Holds the data between the
+ *	Tx response (TX_CMD), and the block ack notification (COMPRESSED_BA).
  * @state: state of the BA agreement establishment / tear down.
  * @txq_id: Tx queue used by the BA session
  * @ssn: the first packet to be sent in AGG HW queue in Tx AGG start flow, or
@@ -265,6 +267,7 @@ struct iwl_mvm_tid_data {
 	u16 next_reclaimed;
 	/* The rest is Tx AGG related */
 	u32 rate_n_flags;
+	u8 reduced_tpc;
 	enum iwl_mvm_agg_state state;
 	u16 txq_id;
 	u16 ssn;
@@ -284,8 +287,6 @@ static inline u16 iwl_mvm_tid_queued(struct iwl_mvm_tid_data *tid_data)
  * @tid_disable_agg: bitmap: if bit(tid) is set, the fw won't send ampdus for
  *	tid.
  * @max_agg_bufsize: the maximal size of the AGG buffer for this station
- * @bt_reduced_txpower_dbg: debug mode in which %bt_reduced_txpower is forced
- *	by debugfs.
  * @bt_reduced_txpower: is reduced tx power enabled for this station
  * @next_status_eosp: the next reclaimed packet is a PS-Poll response and
  *	we need to signal the EOSP
@@ -306,7 +307,6 @@ struct iwl_mvm_sta {
 	u32 mac_id_n_color;
 	u16 tid_disable_agg;
 	u8 max_agg_bufsize;
-	bool bt_reduced_txpower_dbg;
 	bool bt_reduced_txpower;
 	bool next_status_eosp;
 	spinlock_t lock;
diff --git a/drivers/net/wireless/iwlwifi/mvm/time-event.c b/drivers/net/wireless/iwlwifi/mvm/time-event.c
index 61331245ad93..80100f6cc12a 100644
--- a/drivers/net/wireless/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/iwlwifi/mvm/time-event.c
@@ -273,67 +273,10 @@ static bool iwl_mvm_time_event_response(struct iwl_notif_wait_data *notif_wait,
 	return true;
 }
 
-/* used to convert from time event API v2 to v1 */
-#define TE_V2_DEP_POLICY_MSK (TE_V2_DEP_OTHER | TE_V2_DEP_TSF |\
-			     TE_V2_EVENT_SOCIOPATHIC)
-static inline u16 te_v2_get_notify(__le16 policy)
-{
-	return le16_to_cpu(policy) & TE_V2_NOTIF_MSK;
-}
-
-static inline u16 te_v2_get_dep_policy(__le16 policy)
-{
-	return (le16_to_cpu(policy) & TE_V2_DEP_POLICY_MSK) >>
-		TE_V2_PLACEMENT_POS;
-}
-
-static inline u16 te_v2_get_absence(__le16 policy)
-{
-	return (le16_to_cpu(policy) & TE_V2_ABSENCE) >> TE_V2_ABSENCE_POS;
-}
-
-static void iwl_mvm_te_v2_to_v1(const struct iwl_time_event_cmd_v2 *cmd_v2,
-				struct iwl_time_event_cmd_v1 *cmd_v1)
-{
-	cmd_v1->id_and_color = cmd_v2->id_and_color;
-	cmd_v1->action = cmd_v2->action;
-	cmd_v1->id = cmd_v2->id;
-	cmd_v1->apply_time = cmd_v2->apply_time;
-	cmd_v1->max_delay = cmd_v2->max_delay;
-	cmd_v1->depends_on = cmd_v2->depends_on;
-	cmd_v1->interval = cmd_v2->interval;
-	cmd_v1->duration = cmd_v2->duration;
-	if (cmd_v2->repeat == TE_V2_REPEAT_ENDLESS)
-		cmd_v1->repeat = cpu_to_le32(TE_V1_REPEAT_ENDLESS);
-	else
-		cmd_v1->repeat = cpu_to_le32(cmd_v2->repeat);
-	cmd_v1->max_frags = cpu_to_le32(cmd_v2->max_frags);
-	cmd_v1->interval_reciprocal = 0; /* unused */
-
-	cmd_v1->dep_policy = cpu_to_le32(te_v2_get_dep_policy(cmd_v2->policy));
-	cmd_v1->is_present = cpu_to_le32(!te_v2_get_absence(cmd_v2->policy));
-	cmd_v1->notify = cpu_to_le32(te_v2_get_notify(cmd_v2->policy));
-}
-
-static int iwl_mvm_send_time_event_cmd(struct iwl_mvm *mvm,
-				       const struct iwl_time_event_cmd_v2 *cmd)
-{
-	struct iwl_time_event_cmd_v1 cmd_v1;
-
-	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_TIME_EVENT_API_V2)
-		return iwl_mvm_send_cmd_pdu(mvm, TIME_EVENT_CMD, CMD_SYNC,
-					    sizeof(*cmd), cmd);
-
-	iwl_mvm_te_v2_to_v1(cmd, &cmd_v1);
-	return iwl_mvm_send_cmd_pdu(mvm, TIME_EVENT_CMD, CMD_SYNC,
-				    sizeof(cmd_v1), &cmd_v1);
-}
-
-
 static int iwl_mvm_time_event_send_add(struct iwl_mvm *mvm,
 				       struct ieee80211_vif *vif,
 				       struct iwl_mvm_time_event_data *te_data,
-				       struct iwl_time_event_cmd_v2 *te_cmd)
+				       struct iwl_time_event_cmd *te_cmd)
 {
 	static const u8 time_event_response[] = { TIME_EVENT_CMD };
 	struct iwl_notification_wait wait_time_event;
@@ -369,7 +312,8 @@ static int iwl_mvm_time_event_send_add(struct iwl_mvm *mvm,
 				   ARRAY_SIZE(time_event_response),
 				   iwl_mvm_time_event_response, te_data);
 
-	ret = iwl_mvm_send_time_event_cmd(mvm, te_cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, TIME_EVENT_CMD, 0,
+					    sizeof(*te_cmd), te_cmd);
 	if (ret) {
 		IWL_ERR(mvm, "Couldn't send TIME_EVENT_CMD: %d\n", ret);
 		iwl_remove_notification(&mvm->notif_wait, &wait_time_event);
@@ -397,7 +341,7 @@ void iwl_mvm_protect_session(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_time_event_data *te_data = &mvmvif->time_event_data;
-	struct iwl_time_event_cmd_v2 time_cmd = {};
+	struct iwl_time_event_cmd time_cmd = {};
 
 	lockdep_assert_held(&mvm->mutex);
 
@@ -453,7 +397,7 @@ void iwl_mvm_remove_time_event(struct iwl_mvm *mvm,
 			       struct iwl_mvm_vif *mvmvif,
 			       struct iwl_mvm_time_event_data *te_data)
 {
-	struct iwl_time_event_cmd_v2 time_cmd = {};
+	struct iwl_time_event_cmd time_cmd = {};
 	u32 id, uid;
 	int ret;
 
@@ -490,7 +434,8 @@ void iwl_mvm_remove_time_event(struct iwl_mvm *mvm,
 		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->id, mvmvif->color));
 
 	IWL_DEBUG_TE(mvm, "Removing TE 0x%x\n", le32_to_cpu(time_cmd.id));
-	ret = iwl_mvm_send_time_event_cmd(mvm, &time_cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, TIME_EVENT_CMD, 0,
+				   sizeof(time_cmd), &time_cmd);
 	if (WARN_ON(ret))
 		return;
 }
@@ -510,7 +455,7 @@ int iwl_mvm_start_p2p_roc(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_time_event_data *te_data = &mvmvif->time_event_data;
-	struct iwl_time_event_cmd_v2 time_cmd = {};
+	struct iwl_time_event_cmd time_cmd = {};
 
 	lockdep_assert_held(&mvm->mutex);
 	if (te_data->running) {
diff --git a/drivers/net/wireless/iwlwifi/mvm/tt.c b/drivers/net/wireless/iwlwifi/mvm/tt.c
index 7a99fa361954..868561512783 100644
--- a/drivers/net/wireless/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/iwlwifi/mvm/tt.c
@@ -409,7 +409,6 @@ void iwl_mvm_tt_tx_backoff(struct iwl_mvm *mvm, u32 backoff)
 		.id = REPLY_THERMAL_MNG_BACKOFF,
 		.len = { sizeof(u32), },
 		.data = { &backoff, },
-		.flags = CMD_SYNC,
 	};
 
 	backoff = max(backoff, mvm->thermal_throttle.min_backoff);
@@ -468,13 +467,14 @@ void iwl_mvm_tt_handler(struct iwl_mvm *mvm)
 	}
 
 	if (params->support_tx_backoff) {
-		tx_backoff = 0;
+		tx_backoff = tt->min_backoff;
 		for (i = 0; i < TT_TX_BACKOFF_SIZE; i++) {
 			if (temperature < params->tx_backoff[i].temperature)
 				break;
-			tx_backoff = params->tx_backoff[i].backoff;
+			tx_backoff = max(tt->min_backoff,
+					 params->tx_backoff[i].backoff);
 		}
-		if (tx_backoff != 0)
+		if (tx_backoff != tt->min_backoff)
 			throttle_enable = true;
 		if (tt->tx_backoff != tx_backoff)
 			iwl_mvm_tt_tx_backoff(mvm, tx_backoff);
@@ -484,7 +484,8 @@ void iwl_mvm_tt_handler(struct iwl_mvm *mvm)
 		IWL_WARN(mvm,
 			 "Due to high temperature thermal throttling initiated\n");
 		tt->throttle = true;
-	} else if (tt->throttle && !tt->dynamic_smps && tt->tx_backoff == 0 &&
+	} else if (tt->throttle && !tt->dynamic_smps &&
+		   tt->tx_backoff == tt->min_backoff &&
 		   temperature <= params->tx_protection_exit) {
 		IWL_WARN(mvm,
 			 "Temperature is back to normal thermal throttling stopped\n");
diff --git a/drivers/net/wireless/iwlwifi/mvm/tx.c b/drivers/net/wireless/iwlwifi/mvm/tx.c
index 879aeac46cc1..3846a6c41eb1 100644
--- a/drivers/net/wireless/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/iwlwifi/mvm/tx.c
@@ -636,7 +636,11 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 			seq_ctl = le16_to_cpu(hdr->seq_ctrl);
 		}
 
-		ieee80211_tx_status_ni(mvm->hw, skb);
+		BUILD_BUG_ON(ARRAY_SIZE(info->status.status_driver_data) < 1);
+		info->status.status_driver_data[0] =
+				(void *)(uintptr_t)tx_resp->reduced_tpc;
+
+		ieee80211_tx_status(mvm->hw, skb);
 	}
 
 	if (txq_id >= mvm->first_agg_queue) {
@@ -815,6 +819,7 @@ static void iwl_mvm_rx_tx_cmd_agg(struct iwl_mvm *mvm,
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 		mvmsta->tid_data[tid].rate_n_flags =
 			le32_to_cpu(tx_resp->initial_rate);
+		mvmsta->tid_data[tid].reduced_tpc = tx_resp->reduced_tpc;
 	}
 
 	rcu_read_unlock();
@@ -928,6 +933,8 @@ int iwl_mvm_rx_ba_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb,
 			info->status.ampdu_len = ba_notif->txed;
 			iwl_mvm_hwrate_to_tx_status(tid_data->rate_n_flags,
 						    info);
+			info->status.status_driver_data[0] =
+				(void *)(uintptr_t)tid_data->reduced_tpc;
 		}
 	}
 
@@ -937,7 +944,7 @@ int iwl_mvm_rx_ba_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb,
 
 	while (!skb_queue_empty(&reclaimed_skbs)) {
 		skb = __skb_dequeue(&reclaimed_skbs);
-		ieee80211_tx_status_ni(mvm->hw, skb);
+		ieee80211_tx_status(mvm->hw, skb);
 	}
 
 	return 0;
@@ -951,7 +958,7 @@ int iwl_mvm_flush_tx_path(struct iwl_mvm *mvm, u32 tfd_msk, bool sync)
 		.flush_ctl = cpu_to_le16(DUMP_TX_FIFO_FLUSH),
 	};
 
-	u32 flags = sync ? CMD_SYNC : CMD_ASYNC;
+	u32 flags = sync ? 0 : CMD_ASYNC;
 
 	ret = iwl_mvm_send_cmd_pdu(mvm, TXPATH_FLUSH, flags,
 				   sizeof(flush_cmd), &flush_cmd);
diff --git a/drivers/net/wireless/iwlwifi/mvm/utils.c b/drivers/net/wireless/iwlwifi/mvm/utils.c
index 2180902266ae..aa9fc77e8413 100644
--- a/drivers/net/wireless/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/iwlwifi/mvm/utils.c
@@ -64,6 +64,7 @@
 
 #include "iwl-debug.h"
 #include "iwl-io.h"
+#include "iwl-prph.h"
 
 #include "mvm.h"
 #include "fw-api-rs.h"
@@ -143,7 +144,7 @@ int iwl_mvm_send_cmd_status(struct iwl_mvm *mvm, struct iwl_host_cmd *cmd,
 		      "cmd flags %x", cmd->flags))
 		return -EINVAL;
 
-	cmd->flags |= CMD_SYNC | CMD_WANT_SKB;
+	cmd->flags |= CMD_WANT_SKB;
 
 	ret = iwl_trans_send_cmd(mvm->trans, cmd);
 	if (ret == -ERFKILL) {
@@ -469,6 +470,8 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
 			mvm->status, table.valid);
 	}
 
+	/* Do not change this output - scripts rely on it */
+
 	IWL_ERR(mvm, "Loaded firmware version: %s\n", mvm->fw->fw_version);
 
 	trace_iwlwifi_dev_ucode_error(trans->dev, table.error_id, table.tsf_low,
@@ -516,13 +519,14 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
 		iwl_mvm_dump_umac_error_log(mvm);
 }
 
+#ifdef CONFIG_IWLWIFI_DEBUGFS
 void iwl_mvm_fw_error_sram_dump(struct iwl_mvm *mvm)
 {
 	const struct fw_img *img;
 	u32 ofs, sram_len;
 	void *sram;
 
-	if (!mvm->ucode_loaded || mvm->fw_error_sram)
+	if (!mvm->ucode_loaded || mvm->fw_error_sram || mvm->fw_error_dump)
 		return;
 
 	img = &mvm->fw->img[mvm->cur_ucode];
@@ -538,6 +542,48 @@ void iwl_mvm_fw_error_sram_dump(struct iwl_mvm *mvm)
 	mvm->fw_error_sram_len = sram_len;
 }
 
+void iwl_mvm_fw_error_rxf_dump(struct iwl_mvm *mvm)
+{
+	int i, reg_val;
+	unsigned long flags;
+
+	if (!mvm->ucode_loaded || mvm->fw_error_rxf || mvm->fw_error_dump)
+		return;
+
+	/* reading buffer size */
+	reg_val = iwl_trans_read_prph(mvm->trans, RXF_SIZE_ADDR);
+	mvm->fw_error_rxf_len =
+		(reg_val & RXF_SIZE_BYTE_CNT_MSK) >> RXF_SIZE_BYTE_CND_POS;
+
+	/* the register holds the value divided by 128 */
+	mvm->fw_error_rxf_len = mvm->fw_error_rxf_len << 7;
+
+	if (!mvm->fw_error_rxf_len)
+		return;
+
+	mvm->fw_error_rxf =  kzalloc(mvm->fw_error_rxf_len, GFP_ATOMIC);
+	if (!mvm->fw_error_rxf) {
+		mvm->fw_error_rxf_len = 0;
+		return;
+	}
+
+	if (!iwl_trans_grab_nic_access(mvm->trans, false, &flags)) {
+		kfree(mvm->fw_error_rxf);
+		mvm->fw_error_rxf = NULL;
+		mvm->fw_error_rxf_len = 0;
+		return;
+	}
+
+	for (i = 0; i < (mvm->fw_error_rxf_len / sizeof(u32)); i++) {
+		iwl_trans_write_prph(mvm->trans, RXF_LD_FENCE_OFFSET_ADDR,
+				     i * sizeof(u32));
+		mvm->fw_error_rxf[i] =
+			iwl_trans_read_prph(mvm->trans, RXF_FIFO_RD_FENCE_ADDR);
+	}
+	iwl_trans_release_nic_access(mvm->trans, &flags);
+}
+#endif
+
 /**
  * iwl_mvm_send_lq_cmd() - Send link quality command
  * @init: This command is sent as part of station initialization right
@@ -553,7 +599,7 @@ int iwl_mvm_send_lq_cmd(struct iwl_mvm *mvm, struct iwl_lq_cmd *lq, bool init)
 	struct iwl_host_cmd cmd = {
 		.id = LQ_CMD,
 		.len = { sizeof(struct iwl_lq_cmd), },
-		.flags = init ? CMD_SYNC : CMD_ASYNC,
+		.flags = init ? 0 : CMD_ASYNC,
 		.data = { lq, },
 	};
 
@@ -604,6 +650,39 @@ void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	ieee80211_request_smps(vif, smps_mode);
 }
 
+static void iwl_mvm_diversity_iter(void *_data, u8 *mac,
+				   struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	bool *result = _data;
+	int i;
+
+	for (i = 0; i < NUM_IWL_MVM_SMPS_REQ; i++) {
+		if (mvmvif->smps_requests[i] == IEEE80211_SMPS_STATIC ||
+		    mvmvif->smps_requests[i] == IEEE80211_SMPS_DYNAMIC)
+			*result = false;
+	}
+}
+
+bool iwl_mvm_rx_diversity_allowed(struct iwl_mvm *mvm)
+{
+	bool result = true;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (num_of_ant(mvm->fw->valid_rx_ant) == 1)
+		return false;
+
+	if (!mvm->cfg->rx_with_siso_diversity)
+		return false;
+
+	ieee80211_iterate_active_interfaces_atomic(
+			mvm->hw, IEEE80211_IFACE_ITER_NORMAL,
+			iwl_mvm_diversity_iter, &result);
+
+	return result;
+}
+
 int iwl_mvm_update_low_latency(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			       bool value)
 {
@@ -623,7 +702,7 @@ int iwl_mvm_update_low_latency(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	iwl_mvm_bt_coex_vif_change(mvm);
 
-	return iwl_mvm_power_update_mac(mvm, vif);
+	return iwl_mvm_power_update_mac(mvm);
 }
 
 static void iwl_mvm_ll_iter(void *_data, u8 *mac, struct ieee80211_vif *vif)
diff --git a/drivers/net/wireless/iwlwifi/pcie/drv.c b/drivers/net/wireless/iwlwifi/pcie/drv.c
index 3d1d57f9f5bc..7091a18d5a72 100644
--- a/drivers/net/wireless/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/iwlwifi/pcie/drv.c
@@ -417,7 +417,7 @@ static u64 splx_get_pwr_limit(struct iwl_trans *trans, union acpi_object *splx)
 	    splx->package.count != 2 ||
 	    splx->package.elements[0].type != ACPI_TYPE_INTEGER ||
 	    splx->package.elements[0].integer.value != 0) {
-		IWL_ERR(trans, "Unsupported splx structure");
+		IWL_ERR(trans, "Unsupported splx structure\n");
 		return 0;
 	}
 
@@ -426,14 +426,14 @@ static u64 splx_get_pwr_limit(struct iwl_trans *trans, union acpi_object *splx)
 	    limits->package.count < 2 ||
 	    limits->package.elements[0].type != ACPI_TYPE_INTEGER ||
 	    limits->package.elements[1].type != ACPI_TYPE_INTEGER) {
-		IWL_ERR(trans, "Invalid limits element");
+		IWL_ERR(trans, "Invalid limits element\n");
 		return 0;
 	}
 
 	domain_type = &limits->package.elements[0];
 	power_limit = &limits->package.elements[1];
 	if (!(domain_type->integer.value & SPL_DOMAINTYPE_WIFI)) {
-		IWL_DEBUG_INFO(trans, "WiFi power is not limited");
+		IWL_DEBUG_INFO(trans, "WiFi power is not limited\n");
 		return 0;
 	}
 
@@ -450,26 +450,26 @@ static void set_dflt_pwr_limit(struct iwl_trans *trans, struct pci_dev *pdev)
 	pxsx_handle = ACPI_HANDLE(&pdev->dev);
 	if (!pxsx_handle) {
 		IWL_DEBUG_INFO(trans,
-			       "Could not retrieve root port ACPI handle");
+			       "Could not retrieve root port ACPI handle\n");
 		return;
 	}
 
 	/* Get the method's handle */
 	status = acpi_get_handle(pxsx_handle, (acpi_string)SPL_METHOD, &handle);
 	if (ACPI_FAILURE(status)) {
-		IWL_DEBUG_INFO(trans, "SPL method not found");
+		IWL_DEBUG_INFO(trans, "SPL method not found\n");
 		return;
 	}
 
 	/* Call SPLC with no arguments */
 	status = acpi_evaluate_object(handle, NULL, NULL, &splx);
 	if (ACPI_FAILURE(status)) {
-		IWL_ERR(trans, "SPLC invocation failed (0x%x)", status);
+		IWL_ERR(trans, "SPLC invocation failed (0x%x)\n", status);
 		return;
 	}
 
 	trans->dflt_pwr_limit = splx_get_pwr_limit(trans, splx.pointer);
-	IWL_DEBUG_INFO(trans, "Default power limit set to %lld",
+	IWL_DEBUG_INFO(trans, "Default power limit set to %lld\n",
 		       trans->dflt_pwr_limit);
 	kfree(splx.pointer);
 }
diff --git a/drivers/net/wireless/iwlwifi/pcie/internal.h b/drivers/net/wireless/iwlwifi/pcie/internal.h
index 9091513ea738..6c22b23a2845 100644
--- a/drivers/net/wireless/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/iwlwifi/pcie/internal.h
@@ -102,7 +102,7 @@ struct iwl_rxq {
 	u32 write_actual;
 	struct list_head rx_free;
 	struct list_head rx_used;
-	int need_update;
+	bool need_update;
 	struct iwl_rb_status *rb_stts;
 	dma_addr_t rb_stts_dma;
 	spinlock_t lock;
@@ -117,21 +117,19 @@ struct iwl_dma_ptr {
 /**
  * iwl_queue_inc_wrap - increment queue index, wrap back to beginning
  * @index -- current index
- * @n_bd -- total number of entries in queue (must be power of 2)
  */
-static inline int iwl_queue_inc_wrap(int index, int n_bd)
+static inline int iwl_queue_inc_wrap(int index)
 {
-	return ++index & (n_bd - 1);
+	return ++index & (TFD_QUEUE_SIZE_MAX - 1);
 }
 
 /**
  * iwl_queue_dec_wrap - decrement queue index, wrap back to end
  * @index -- current index
- * @n_bd -- total number of entries in queue (must be power of 2)
  */
-static inline int iwl_queue_dec_wrap(int index, int n_bd)
+static inline int iwl_queue_dec_wrap(int index)
 {
-	return --index & (n_bd - 1);
+	return --index & (TFD_QUEUE_SIZE_MAX - 1);
 }
 
 struct iwl_cmd_meta {
@@ -145,13 +143,13 @@ struct iwl_cmd_meta {
  *
  * Contains common data for Rx and Tx queues.
  *
- * Note the difference between n_bd and n_window: the hardware
- * always assumes 256 descriptors, so n_bd is always 256 (unless
+ * Note the difference between TFD_QUEUE_SIZE_MAX and n_window: the hardware
+ * always assumes 256 descriptors, so TFD_QUEUE_SIZE_MAX is always 256 (unless
  * there might be HW changes in the future). For the normal TX
  * queues, n_window, which is the size of the software queue data
  * is also 256; however, for the command queue, n_window is only
  * 32 since we don't need so many commands pending. Since the HW
- * still uses 256 BDs for DMA though, n_bd stays 256. As a result,
+ * still uses 256 BDs for DMA though, TFD_QUEUE_SIZE_MAX stays 256. As a result,
  * the software buffers (in the variables @meta, @txb in struct
  * iwl_txq) only have 32 entries, while the HW buffers (@tfds in
  * the same struct) have 256.
@@ -162,7 +160,6 @@ struct iwl_cmd_meta {
  * data is a window overlayed over the HW queue.
  */
 struct iwl_queue {
-	int n_bd;              /* number of BDs in this queue */
 	int write_ptr;       /* 1-st empty entry (index) host_w*/
 	int read_ptr;         /* last used entry (index) host_r*/
 	/* use for monitoring and recovering the stuck queue */
@@ -231,7 +228,7 @@ struct iwl_txq {
 	spinlock_t lock;
 	struct timer_list stuck_timer;
 	struct iwl_trans_pcie *trans_pcie;
-	u8 need_update;
+	bool need_update;
 	u8 active;
 	bool ampdu;
 };
@@ -270,6 +267,9 @@ struct iwl_trans_pcie {
 	struct iwl_trans *trans;
 	struct iwl_drv *drv;
 
+	struct net_device napi_dev;
+	struct napi_struct napi;
+
 	/* INT ICT Table */
 	__le32 *ict_tbl;
 	dma_addr_t ict_tbl_dma;
@@ -362,7 +362,7 @@ void iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, int fifo,
 void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int queue);
 int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 		      struct iwl_device_cmd *dev_cmd, int txq_id);
-void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq);
+void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans);
 int iwl_trans_pcie_send_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd);
 void iwl_pcie_hcmd_complete(struct iwl_trans *trans,
 			    struct iwl_rx_cmd_buffer *rxb, int handler_status);
@@ -370,6 +370,13 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 			    struct sk_buff_head *skbs);
 void iwl_trans_pcie_tx_reset(struct iwl_trans *trans);
 
+static inline u16 iwl_pcie_tfd_tb_get_len(struct iwl_tfd *tfd, u8 idx)
+{
+	struct iwl_tfd_tb *tb = &tfd->tbs[idx];
+
+	return le16_to_cpu(tb->hi_n_len) >> 4;
+}
+
 /*****************************************************
 * Error handling
 ******************************************************/
diff --git a/drivers/net/wireless/iwlwifi/pcie/rx.c b/drivers/net/wireless/iwlwifi/pcie/rx.c
index fdfa3969cac9..a2698e5e062c 100644
--- a/drivers/net/wireless/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/iwlwifi/pcie/rx.c
@@ -145,15 +145,13 @@ int iwl_pcie_rx_stop(struct iwl_trans *trans)
 /*
  * iwl_pcie_rxq_inc_wr_ptr - Update the write pointer for the RX queue
  */
-static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans,
-				    struct iwl_rxq *rxq)
+static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	u32 reg;
 
-	spin_lock(&rxq->lock);
-
-	if (rxq->need_update == 0)
-		goto exit_unlock;
+	lockdep_assert_held(&rxq->lock);
 
 	/*
 	 * explicitly wake up the NIC if:
@@ -169,13 +167,27 @@ static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans,
 				       reg);
 			iwl_set_bit(trans, CSR_GP_CNTRL,
 				    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
-			goto exit_unlock;
+			rxq->need_update = true;
+			return;
 		}
 	}
 
 	rxq->write_actual = round_down(rxq->write, 8);
 	iwl_write32(trans, FH_RSCSR_CHNL0_WPTR, rxq->write_actual);
-	rxq->need_update = 0;
+}
+
+static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_rxq *rxq = &trans_pcie->rxq;
+
+	spin_lock(&rxq->lock);
+
+	if (!rxq->need_update)
+		goto exit_unlock;
+
+	iwl_pcie_rxq_inc_wr_ptr(trans);
+	rxq->need_update = false;
 
  exit_unlock:
 	spin_unlock(&rxq->lock);
@@ -236,9 +248,8 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans)
 	 * Increment device's write pointer in multiples of 8. */
 	if (rxq->write_actual != (rxq->write & ~0x7)) {
 		spin_lock(&rxq->lock);
-		rxq->need_update = 1;
+		iwl_pcie_rxq_inc_wr_ptr(trans);
 		spin_unlock(&rxq->lock);
-		iwl_pcie_rxq_inc_wr_ptr(trans, rxq);
 	}
 }
 
@@ -362,20 +373,9 @@ static void iwl_pcie_rxq_free_rbs(struct iwl_trans *trans)
  * Also restock the Rx queue via iwl_pcie_rxq_restock.
  * This is called as a scheduled work item (except for during initialization)
  */
-static void iwl_pcie_rx_replenish(struct iwl_trans *trans)
-{
-	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-
-	iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL);
-
-	spin_lock(&trans_pcie->irq_lock);
-	iwl_pcie_rxq_restock(trans);
-	spin_unlock(&trans_pcie->irq_lock);
-}
-
-static void iwl_pcie_rx_replenish_now(struct iwl_trans *trans)
+static void iwl_pcie_rx_replenish(struct iwl_trans *trans, gfp_t gfp)
 {
-	iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC);
+	iwl_pcie_rxq_alloc_rbs(trans, gfp);
 
 	iwl_pcie_rxq_restock(trans);
 }
@@ -385,7 +385,7 @@ static void iwl_pcie_rx_replenish_work(struct work_struct *data)
 	struct iwl_trans_pcie *trans_pcie =
 	    container_of(data, struct iwl_trans_pcie, rx_replenish);
 
-	iwl_pcie_rx_replenish(trans_pcie->trans);
+	iwl_pcie_rx_replenish(trans_pcie->trans, GFP_KERNEL);
 }
 
 static int iwl_pcie_rx_alloc(struct iwl_trans *trans)
@@ -521,14 +521,13 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	memset(rxq->rb_stts, 0, sizeof(*rxq->rb_stts));
 	spin_unlock(&rxq->lock);
 
-	iwl_pcie_rx_replenish(trans);
+	iwl_pcie_rx_replenish(trans, GFP_KERNEL);
 
 	iwl_pcie_rx_hw_init(trans, rxq);
 
-	spin_lock(&trans_pcie->irq_lock);
-	rxq->need_update = 1;
-	iwl_pcie_rxq_inc_wr_ptr(trans, rxq);
-	spin_unlock(&trans_pcie->irq_lock);
+	spin_lock(&rxq->lock);
+	iwl_pcie_rxq_inc_wr_ptr(trans);
+	spin_unlock(&rxq->lock);
 
 	return 0;
 }
@@ -673,7 +672,6 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 	/* Reuse the page if possible. For notification packets and
 	 * SKBs that fail to Rx correctly, add them back into the
 	 * rx_free list for reuse later. */
-	spin_lock(&rxq->lock);
 	if (rxb->page != NULL) {
 		rxb->page_dma =
 			dma_map_page(trans->dev, rxb->page, 0,
@@ -694,7 +692,6 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 		}
 	} else
 		list_add_tail(&rxb->list, &rxq->rx_used);
-	spin_unlock(&rxq->lock);
 }
 
 /*
@@ -709,6 +706,8 @@ static void iwl_pcie_rx_handle(struct iwl_trans *trans)
 	u32 count = 8;
 	int total_empty;
 
+restart:
+	spin_lock(&rxq->lock);
 	/* uCode's read index (stored in shared DRAM) indicates the last Rx
 	 * buffer that the driver may process (last buffer filled by ucode). */
 	r = le16_to_cpu(ACCESS_ONCE(rxq->rb_stts->closed_rb_num)) & 0x0FFF;
@@ -743,18 +742,25 @@ static void iwl_pcie_rx_handle(struct iwl_trans *trans)
 			count++;
 			if (count >= 8) {
 				rxq->read = i;
-				iwl_pcie_rx_replenish_now(trans);
+				spin_unlock(&rxq->lock);
+				iwl_pcie_rx_replenish(trans, GFP_ATOMIC);
 				count = 0;
+				goto restart;
 			}
 		}
 	}
 
 	/* Backtrack one entry */
 	rxq->read = i;
+	spin_unlock(&rxq->lock);
+
 	if (fill_rx)
-		iwl_pcie_rx_replenish_now(trans);
+		iwl_pcie_rx_replenish(trans, GFP_ATOMIC);
 	else
 		iwl_pcie_rxq_restock(trans);
+
+	if (trans_pcie->napi.poll)
+		napi_gro_flush(&trans_pcie->napi, false);
 }
 
 /*
@@ -844,7 +850,7 @@ static u32 iwl_pcie_int_cause_ict(struct iwl_trans *trans)
 				trans_pcie->ict_index, read);
 		trans_pcie->ict_tbl[trans_pcie->ict_index] = 0;
 		trans_pcie->ict_index =
-			iwl_queue_inc_wrap(trans_pcie->ict_index, ICT_COUNT);
+			((trans_pcie->ict_index + 1) & (ICT_COUNT - 1));
 
 		read = le32_to_cpu(trans_pcie->ict_tbl[trans_pcie->ict_index]);
 		trace_iwlwifi_dev_ict_read(trans->dev, trans_pcie->ict_index,
@@ -876,7 +882,6 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id)
 	struct isr_statistics *isr_stats = &trans_pcie->isr_stats;
 	u32 inta = 0;
 	u32 handled = 0;
-	u32 i;
 
 	lock_map_acquire(&trans->sync_cmd_lockdep_map);
 
@@ -1028,9 +1033,8 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id)
 	/* uCode wakes up after power-down sleep */
 	if (inta & CSR_INT_BIT_WAKEUP) {
 		IWL_DEBUG_ISR(trans, "Wakeup interrupt\n");
-		iwl_pcie_rxq_inc_wr_ptr(trans, &trans_pcie->rxq);
-		for (i = 0; i < trans->cfg->base_params->num_of_queues; i++)
-			iwl_pcie_txq_inc_wr_ptr(trans, &trans_pcie->txq[i]);
+		iwl_pcie_rxq_check_wrptr(trans);
+		iwl_pcie_txq_check_wrptrs(trans);
 
 		isr_stats->wakeup++;
 
@@ -1068,8 +1072,6 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id)
 		iwl_write8(trans, CSR_INT_PERIODIC_REG,
 			    CSR_INT_PERIODIC_DIS);
 
-		iwl_pcie_rx_handle(trans);
-
 		/*
 		 * Enable periodic interrupt in 8 msec only if we received
 		 * real RX interrupt (instead of just periodic int), to catch
@@ -1082,6 +1084,10 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id)
 				   CSR_INT_PERIODIC_ENA);
 
 		isr_stats->rx++;
+
+		local_bh_disable();
+		iwl_pcie_rx_handle(trans);
+		local_bh_enable();
 	}
 
 	/* This "Tx" DMA channel is used only for loading uCode */
diff --git a/drivers/net/wireless/iwlwifi/pcie/trans.c b/drivers/net/wireless/iwlwifi/pcie/trans.c
index 2365553f1ef7..788085bc65d7 100644
--- a/drivers/net/wireless/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/iwlwifi/pcie/trans.c
@@ -73,6 +73,7 @@
 #include "iwl-csr.h"
 #include "iwl-prph.h"
 #include "iwl-agn-hw.h"
+#include "iwl-fw-error-dump.h"
 #include "internal.h"
 
 static u32 iwl_trans_pcie_read_shr(struct iwl_trans *trans, u32 reg)
@@ -103,7 +104,6 @@ static void iwl_pcie_set_pwr(struct iwl_trans *trans, bool vaux)
 
 /* PCI registers */
 #define PCI_CFG_RETRY_TIMEOUT	0x041
-#define CPU1_CPU2_SEPARATOR_SECTION	0xFFFFCCCC
 
 static void iwl_pcie_apm_config(struct iwl_trans *trans)
 {
@@ -454,6 +454,7 @@ static int iwl_pcie_prepare_card_hw(struct iwl_trans *trans)
 {
 	int ret;
 	int t = 0;
+	int iter;
 
 	IWL_DEBUG_INFO(trans, "iwl_trans_prepare_card_hw enter\n");
 
@@ -462,18 +463,23 @@ static int iwl_pcie_prepare_card_hw(struct iwl_trans *trans)
 	if (ret >= 0)
 		return 0;
 
-	/* If HW is not ready, prepare the conditions to check again */
-	iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
-		    CSR_HW_IF_CONFIG_REG_PREPARE);
+	for (iter = 0; iter < 10; iter++) {
+		/* If HW is not ready, prepare the conditions to check again */
+		iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
+			    CSR_HW_IF_CONFIG_REG_PREPARE);
+
+		do {
+			ret = iwl_pcie_set_hw_ready(trans);
+			if (ret >= 0)
+				return 0;
 
-	do {
-		ret = iwl_pcie_set_hw_ready(trans);
-		if (ret >= 0)
-			return 0;
+			usleep_range(200, 1000);
+			t += 200;
+		} while (t < 150000);
+		msleep(25);
+	}
 
-		usleep_range(200, 1000);
-		t += 200;
-	} while (t < 150000);
+	IWL_DEBUG_INFO(trans, "got NIC after %d iterations\n", iter);
 
 	return ret;
 }
@@ -1053,6 +1059,12 @@ static void iwl_trans_pcie_write_prph(struct iwl_trans *trans, u32 addr,
 	iwl_trans_pcie_write32(trans, HBUS_TARG_PRPH_WDAT, val);
 }
 
+static int iwl_pcie_dummy_napi_poll(struct napi_struct *napi, int budget)
+{
+	WARN_ON(1);
+	return 0;
+}
+
 static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 				     const struct iwl_trans_config *trans_cfg)
 {
@@ -1079,6 +1091,18 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 
 	trans_pcie->command_names = trans_cfg->command_names;
 	trans_pcie->bc_table_dword = trans_cfg->bc_table_dword;
+
+	/* Initialize NAPI here - it should be before registering to mac80211
+	 * in the opmode but after the HW struct is allocated.
+	 * As this function may be called again in some corner cases don't
+	 * do anything if NAPI was already initialized.
+	 */
+	if (!trans_pcie->napi.poll && trans->op_mode->ops->napi_add) {
+		init_dummy_netdev(&trans_pcie->napi_dev);
+		iwl_op_mode_napi_add(trans->op_mode, &trans_pcie->napi,
+				     &trans_pcie->napi_dev,
+				     iwl_pcie_dummy_napi_poll, 64);
+	}
 }
 
 void iwl_trans_pcie_free(struct iwl_trans *trans)
@@ -1099,6 +1123,9 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	pci_disable_device(trans_pcie->pci_dev);
 	kmem_cache_destroy(trans->dev_cmd_pool);
 
+	if (trans_pcie->napi.poll)
+		netif_napi_del(&trans_pcie->napi);
+
 	kfree(trans);
 }
 
@@ -1237,7 +1264,7 @@ static int iwl_trans_pcie_write_mem(struct iwl_trans *trans, u32 addr,
 
 #define IWL_FLUSH_WAIT_MS	2000
 
-static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans)
+static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, u32 txq_bm)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq;
@@ -1250,13 +1277,31 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans)
 
 	/* waiting for all the tx frames complete might take a while */
 	for (cnt = 0; cnt < trans->cfg->base_params->num_of_queues; cnt++) {
+		u8 wr_ptr;
+
 		if (cnt == trans_pcie->cmd_queue)
 			continue;
+		if (!test_bit(cnt, trans_pcie->queue_used))
+			continue;
+		if (!(BIT(cnt) & txq_bm))
+			continue;
+
+		IWL_DEBUG_TX_QUEUES(trans, "Emptying queue %d...\n", cnt);
 		txq = &trans_pcie->txq[cnt];
 		q = &txq->q;
-		while (q->read_ptr != q->write_ptr && !time_after(jiffies,
-		       now + msecs_to_jiffies(IWL_FLUSH_WAIT_MS)))
+		wr_ptr = ACCESS_ONCE(q->write_ptr);
+
+		while (q->read_ptr != ACCESS_ONCE(q->write_ptr) &&
+		       !time_after(jiffies,
+				   now + msecs_to_jiffies(IWL_FLUSH_WAIT_MS))) {
+			u8 write_ptr = ACCESS_ONCE(q->write_ptr);
+
+			if (WARN_ONCE(wr_ptr != write_ptr,
+				      "WR pointer moved while flushing %d -> %d\n",
+				      wr_ptr, write_ptr))
+				return -ETIMEDOUT;
 			msleep(1);
+		}
 
 		if (q->read_ptr != q->write_ptr) {
 			IWL_ERR(trans,
@@ -1264,6 +1309,7 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans)
 			ret = -ETIMEDOUT;
 			break;
 		}
+		IWL_DEBUG_TX_QUEUES(trans, "Queue %d is now empty.\n", cnt);
 	}
 
 	if (!ret)
@@ -1298,8 +1344,8 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans)
 		IWL_ERR(trans,
 			"Q %d is %sactive and mapped to fifo %d ra_tid 0x%04x [%d,%d]\n",
 			cnt, active ? "" : "in", fifo, tbl_dw,
-			iwl_read_prph(trans,
-				      SCD_QUEUE_RDPTR(cnt)) & (txq->q.n_bd - 1),
+			iwl_read_prph(trans, SCD_QUEUE_RDPTR(cnt)) &
+				(TFD_QUEUE_SIZE_MAX - 1),
 			iwl_read_prph(trans, SCD_QUEUE_WRPTR(cnt)));
 	}
 
@@ -1630,6 +1676,61 @@ err:
 	IWL_ERR(trans, "failed to create the trans debugfs entry\n");
 	return -ENOMEM;
 }
+
+static u32 iwl_trans_pcie_get_cmdlen(struct iwl_tfd *tfd)
+{
+	u32 cmdlen = 0;
+	int i;
+
+	for (i = 0; i < IWL_NUM_OF_TBS; i++)
+		cmdlen += iwl_pcie_tfd_tb_get_len(tfd, i);
+
+	return cmdlen;
+}
+
+static u32 iwl_trans_pcie_dump_data(struct iwl_trans *trans,
+				    void *buf, u32 buflen)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_fw_error_dump_data *data;
+	struct iwl_txq *cmdq = &trans_pcie->txq[trans_pcie->cmd_queue];
+	struct iwl_fw_error_dump_txcmd *txcmd;
+	u32 len;
+	int i, ptr;
+
+	if (!buf)
+		return sizeof(*data) +
+		       cmdq->q.n_window * (sizeof(*txcmd) +
+					   TFD_MAX_PAYLOAD_SIZE);
+
+	len = 0;
+	data = buf;
+	data->type = cpu_to_le32(IWL_FW_ERROR_DUMP_TXCMD);
+	txcmd = (void *)data->data;
+	spin_lock_bh(&cmdq->lock);
+	ptr = cmdq->q.write_ptr;
+	for (i = 0; i < cmdq->q.n_window; i++) {
+		u8 idx = get_cmd_index(&cmdq->q, ptr);
+		u32 caplen, cmdlen;
+
+		cmdlen = iwl_trans_pcie_get_cmdlen(&cmdq->tfds[ptr]);
+		caplen = min_t(u32, TFD_MAX_PAYLOAD_SIZE, cmdlen);
+
+		if (cmdlen) {
+			len += sizeof(*txcmd) + caplen;
+			txcmd->cmdlen = cpu_to_le32(cmdlen);
+			txcmd->caplen = cpu_to_le32(caplen);
+			memcpy(txcmd->data, cmdq->entries[idx].cmd, caplen);
+			txcmd = (void *)((u8 *)txcmd->data + caplen);
+		}
+
+		ptr = iwl_queue_dec_wrap(ptr);
+	}
+	spin_unlock_bh(&cmdq->lock);
+
+	data->len = cpu_to_le32(len);
+	return sizeof(*data) + len;
+}
 #else
 static int iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans,
 					 struct dentry *dir)
@@ -1672,6 +1773,10 @@ static const struct iwl_trans_ops trans_ops_pcie = {
 	.grab_nic_access = iwl_trans_pcie_grab_nic_access,
 	.release_nic_access = iwl_trans_pcie_release_nic_access,
 	.set_bits_mask = iwl_trans_pcie_set_bits_mask,
+
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+	.dump_data = iwl_trans_pcie_dump_data,
+#endif
 };
 
 struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
diff --git a/drivers/net/wireless/iwlwifi/pcie/tx.c b/drivers/net/wireless/iwlwifi/pcie/tx.c
index 3b0c72c10054..038940afbdc5 100644
--- a/drivers/net/wireless/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/iwlwifi/pcie/tx.c
@@ -70,20 +70,20 @@ static int iwl_queue_space(const struct iwl_queue *q)
 
 	/*
 	 * To avoid ambiguity between empty and completely full queues, there
-	 * should always be less than q->n_bd elements in the queue.
-	 * If q->n_window is smaller than q->n_bd, there is no need to reserve
-	 * any queue entries for this purpose.
+	 * should always be less than TFD_QUEUE_SIZE_MAX elements in the queue.
+	 * If q->n_window is smaller than TFD_QUEUE_SIZE_MAX, there is no need
+	 * to reserve any queue entries for this purpose.
 	 */
-	if (q->n_window < q->n_bd)
+	if (q->n_window < TFD_QUEUE_SIZE_MAX)
 		max = q->n_window;
 	else
-		max = q->n_bd - 1;
+		max = TFD_QUEUE_SIZE_MAX - 1;
 
 	/*
-	 * q->n_bd is a power of 2, so the following is equivalent to modulo by
-	 * q->n_bd and is well defined for negative dividends.
+	 * TFD_QUEUE_SIZE_MAX is a power of 2, so the following is equivalent to
+	 * modulo by TFD_QUEUE_SIZE_MAX and is well defined.
 	 */
-	used = (q->write_ptr - q->read_ptr) & (q->n_bd - 1);
+	used = (q->write_ptr - q->read_ptr) & (TFD_QUEUE_SIZE_MAX - 1);
 
 	if (WARN_ON(used > max))
 		return 0;
@@ -94,17 +94,11 @@ static int iwl_queue_space(const struct iwl_queue *q)
 /*
  * iwl_queue_init - Initialize queue's high/low-water and read/write indexes
  */
-static int iwl_queue_init(struct iwl_queue *q, int count, int slots_num, u32 id)
+static int iwl_queue_init(struct iwl_queue *q, int slots_num, u32 id)
 {
-	q->n_bd = count;
 	q->n_window = slots_num;
 	q->id = id;
 
-	/* count must be power-of-two size, otherwise iwl_queue_inc_wrap
-	 * and iwl_queue_dec_wrap are broken. */
-	if (WARN_ON(!is_power_of_2(count)))
-		return -EINVAL;
-
 	/* slots_num must be power-of-two size, otherwise
 	 * get_cmd_index is broken. */
 	if (WARN_ON(!is_power_of_2(slots_num)))
@@ -197,17 +191,17 @@ static void iwl_pcie_txq_stuck_timer(unsigned long data)
 		IWL_ERR(trans,
 			"Q %d is %sactive and mapped to fifo %d ra_tid 0x%04x [%d,%d]\n",
 			i, active ? "" : "in", fifo, tbl_dw,
-			iwl_read_prph(trans,
-				      SCD_QUEUE_RDPTR(i)) & (txq->q.n_bd - 1),
+			iwl_read_prph(trans, SCD_QUEUE_RDPTR(i)) &
+				(TFD_QUEUE_SIZE_MAX - 1),
 			iwl_read_prph(trans, SCD_QUEUE_WRPTR(i)));
 	}
 
 	for (i = q->read_ptr; i != q->write_ptr;
-	     i = iwl_queue_inc_wrap(i, q->n_bd))
+	     i = iwl_queue_inc_wrap(i))
 		IWL_ERR(trans, "scratch %d = 0x%08x\n", i,
 			le32_to_cpu(txq->scratchbufs[i].scratch));
 
-	iwl_write_prph(trans, DEVICE_SET_NMI_REG, 1);
+	iwl_force_nmi(trans);
 }
 
 /*
@@ -287,14 +281,14 @@ static void iwl_pcie_txq_inval_byte_cnt_tbl(struct iwl_trans *trans,
 /*
  * iwl_pcie_txq_inc_wr_ptr - Send new write index to hardware
  */
-void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq)
+static void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans,
+				    struct iwl_txq *txq)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	u32 reg = 0;
 	int txq_id = txq->q.id;
 
-	if (txq->need_update == 0)
-		return;
+	lockdep_assert_held(&txq->lock);
 
 	/*
 	 * explicitly wake up the NIC if:
@@ -317,6 +311,7 @@ void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq)
 				       txq_id, reg);
 			iwl_set_bit(trans, CSR_GP_CNTRL,
 				    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
+			txq->need_update = true;
 			return;
 		}
 	}
@@ -327,8 +322,23 @@ void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq)
 	 */
 	IWL_DEBUG_TX(trans, "Q:%d WR: 0x%x\n", txq_id, txq->q.write_ptr);
 	iwl_write32(trans, HBUS_TARG_WRPTR, txq->q.write_ptr | (txq_id << 8));
+}
+
+void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int i;
+
+	for (i = 0; i < trans->cfg->base_params->num_of_queues; i++) {
+		struct iwl_txq *txq = &trans_pcie->txq[i];
 
-	txq->need_update = 0;
+		spin_lock_bh(&txq->lock);
+		if (trans_pcie->txq[i].need_update) {
+			iwl_pcie_txq_inc_wr_ptr(trans, txq);
+			trans_pcie->txq[i].need_update = false;
+		}
+		spin_unlock_bh(&txq->lock);
+	}
 }
 
 static inline dma_addr_t iwl_pcie_tfd_tb_get_addr(struct iwl_tfd *tfd, u8 idx)
@@ -343,13 +353,6 @@ static inline dma_addr_t iwl_pcie_tfd_tb_get_addr(struct iwl_tfd *tfd, u8 idx)
 	return addr;
 }
 
-static inline u16 iwl_pcie_tfd_tb_get_len(struct iwl_tfd *tfd, u8 idx)
-{
-	struct iwl_tfd_tb *tb = &tfd->tbs[idx];
-
-	return le16_to_cpu(tb->hi_n_len) >> 4;
-}
-
 static inline void iwl_pcie_tfd_set_tb(struct iwl_tfd *tfd, u8 idx,
 				       dma_addr_t addr, u16 len)
 {
@@ -409,13 +412,17 @@ static void iwl_pcie_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
 {
 	struct iwl_tfd *tfd_tmp = txq->tfds;
 
-	/* rd_ptr is bounded by n_bd and idx is bounded by n_window */
+	/* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and
+	 * idx is bounded by n_window
+	 */
 	int rd_ptr = txq->q.read_ptr;
 	int idx = get_cmd_index(&txq->q, rd_ptr);
 
 	lockdep_assert_held(&txq->lock);
 
-	/* We have only q->n_window txq->entries, but we use q->n_bd tfds */
+	/* We have only q->n_window txq->entries, but we use
+	 * TFD_QUEUE_SIZE_MAX tfds
+	 */
 	iwl_pcie_tfd_unmap(trans, &txq->entries[idx].meta, &tfd_tmp[rd_ptr]);
 
 	/* free SKB */
@@ -436,7 +443,7 @@ static void iwl_pcie_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
 }
 
 static int iwl_pcie_txq_build_tfd(struct iwl_trans *trans, struct iwl_txq *txq,
-				  dma_addr_t addr, u16 len, u8 reset)
+				  dma_addr_t addr, u16 len, bool reset)
 {
 	struct iwl_queue *q;
 	struct iwl_tfd *tfd, *tfd_tmp;
@@ -542,15 +549,14 @@ static int iwl_pcie_txq_init(struct iwl_trans *trans, struct iwl_txq *txq,
 {
 	int ret;
 
-	txq->need_update = 0;
+	txq->need_update = false;
 
 	/* TFD_QUEUE_SIZE_MAX must be power-of-two size, otherwise
 	 * iwl_queue_inc_wrap and iwl_queue_dec_wrap are broken. */
 	BUILD_BUG_ON(TFD_QUEUE_SIZE_MAX & (TFD_QUEUE_SIZE_MAX - 1));
 
 	/* Initialize queue's high/low-water marks, and head/tail indexes */
-	ret = iwl_queue_init(&txq->q, TFD_QUEUE_SIZE_MAX, slots_num,
-			txq_id);
+	ret = iwl_queue_init(&txq->q, slots_num, txq_id);
 	if (ret)
 		return ret;
 
@@ -575,15 +581,12 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 	struct iwl_txq *txq = &trans_pcie->txq[txq_id];
 	struct iwl_queue *q = &txq->q;
 
-	if (!q->n_bd)
-		return;
-
 	spin_lock_bh(&txq->lock);
 	while (q->write_ptr != q->read_ptr) {
 		IWL_DEBUG_TX_REPLY(trans, "Q %d Free %d\n",
 				   txq_id, q->read_ptr);
 		iwl_pcie_txq_free_tfd(trans, txq);
-		q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd);
+		q->read_ptr = iwl_queue_inc_wrap(q->read_ptr);
 	}
 	txq->active = false;
 	spin_unlock_bh(&txq->lock);
@@ -620,10 +623,12 @@ static void iwl_pcie_txq_free(struct iwl_trans *trans, int txq_id)
 		}
 
 	/* De-alloc circular buffer of TFDs */
-	if (txq->q.n_bd) {
-		dma_free_coherent(dev, sizeof(struct iwl_tfd) *
-				  txq->q.n_bd, txq->tfds, txq->q.dma_addr);
+	if (txq->tfds) {
+		dma_free_coherent(dev,
+				  sizeof(struct iwl_tfd) * TFD_QUEUE_SIZE_MAX,
+				  txq->tfds, txq->q.dma_addr);
 		txq->q.dma_addr = 0;
+		txq->tfds = NULL;
 
 		dma_free_coherent(dev,
 				  sizeof(*txq->scratchbufs) * txq->q.n_window,
@@ -680,7 +685,8 @@ void iwl_pcie_tx_start(struct iwl_trans *trans, u32 scd_base_addr)
 	/* The chain extension of the SCD doesn't work well. This feature is
 	 * enabled by default by the HW, so we need to disable it manually.
 	 */
-	iwl_write_prph(trans, SCD_CHAINEXT_EN, 0);
+	if (trans->cfg->base_params->scd_chain_ext_wa)
+		iwl_write_prph(trans, SCD_CHAINEXT_EN, 0);
 
 	iwl_trans_ac_txq_enable(trans, trans_pcie->cmd_queue,
 				trans_pcie->cmd_fifo);
@@ -931,8 +937,7 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq = &trans_pcie->txq[txq_id];
-	/* n_bd is usually 256 => n_bd - 1 = 0xff */
-	int tfd_num = ssn & (txq->q.n_bd - 1);
+	int tfd_num = ssn & (TFD_QUEUE_SIZE_MAX - 1);
 	struct iwl_queue *q = &txq->q;
 	int last_to_free;
 
@@ -956,12 +961,12 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 
 	/*Since we free until index _not_ inclusive, the one before index is
 	 * the last we will free. This one must be used */
-	last_to_free = iwl_queue_dec_wrap(tfd_num, q->n_bd);
+	last_to_free = iwl_queue_dec_wrap(tfd_num);
 
 	if (!iwl_queue_used(q, last_to_free)) {
 		IWL_ERR(trans,
 			"%s: Read index for DMA queue txq id (%d), last_to_free %d is out of range [0-%d] %d %d.\n",
-			__func__, txq_id, last_to_free, q->n_bd,
+			__func__, txq_id, last_to_free, TFD_QUEUE_SIZE_MAX,
 			q->write_ptr, q->read_ptr);
 		goto out;
 	}
@@ -971,7 +976,7 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 
 	for (;
 	     q->read_ptr != tfd_num;
-	     q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd)) {
+	     q->read_ptr = iwl_queue_inc_wrap(q->read_ptr)) {
 
 		if (WARN_ON_ONCE(txq->entries[txq->q.read_ptr].skb == NULL))
 			continue;
@@ -1010,25 +1015,26 @@ static void iwl_pcie_cmdq_reclaim(struct iwl_trans *trans, int txq_id, int idx)
 
 	lockdep_assert_held(&txq->lock);
 
-	if ((idx >= q->n_bd) || (!iwl_queue_used(q, idx))) {
+	if ((idx >= TFD_QUEUE_SIZE_MAX) || (!iwl_queue_used(q, idx))) {
 		IWL_ERR(trans,
 			"%s: Read index for DMA queue txq id (%d), index %d is out of range [0-%d] %d %d.\n",
-			__func__, txq_id, idx, q->n_bd,
+			__func__, txq_id, idx, TFD_QUEUE_SIZE_MAX,
 			q->write_ptr, q->read_ptr);
 		return;
 	}
 
-	for (idx = iwl_queue_inc_wrap(idx, q->n_bd); q->read_ptr != idx;
-	     q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd)) {
+	for (idx = iwl_queue_inc_wrap(idx); q->read_ptr != idx;
+	     q->read_ptr = iwl_queue_inc_wrap(q->read_ptr)) {
 
 		if (nfreed++ > 0) {
 			IWL_ERR(trans, "HCMD skipped: index (%d) %d %d\n",
 				idx, q->write_ptr, q->read_ptr);
-			iwl_write_prph(trans, DEVICE_SET_NMI_REG, 1);
+			iwl_force_nmi(trans);
 		}
 	}
 
-	if (q->read_ptr == q->write_ptr) {
+	if (trans->cfg->base_params->apmg_wake_up_wa &&
+	    q->read_ptr == q->write_ptr) {
 		spin_lock_irqsave(&trans_pcie->reg_lock, flags);
 		WARN_ON(!trans_pcie->cmd_in_flight);
 		trans_pcie->cmd_in_flight = false;
@@ -1309,28 +1315,39 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 	cmd_pos = offsetof(struct iwl_device_cmd, payload);
 	copy_size = sizeof(out_cmd->hdr);
 	for (i = 0; i < IWL_MAX_CMD_TBS_PER_TFD; i++) {
-		int copy = 0;
+		int copy;
 
 		if (!cmd->len[i])
 			continue;
 
-		/* need at least IWL_HCMD_SCRATCHBUF_SIZE copied */
-		if (copy_size < IWL_HCMD_SCRATCHBUF_SIZE) {
-			copy = IWL_HCMD_SCRATCHBUF_SIZE - copy_size;
-
-			if (copy > cmd->len[i])
-				copy = cmd->len[i];
-		}
-
 		/* copy everything if not nocopy/dup */
 		if (!(cmd->dataflags[i] & (IWL_HCMD_DFL_NOCOPY |
-					   IWL_HCMD_DFL_DUP)))
+					   IWL_HCMD_DFL_DUP))) {
 			copy = cmd->len[i];
 
-		if (copy) {
 			memcpy((u8 *)out_cmd + cmd_pos, cmd->data[i], copy);
 			cmd_pos += copy;
 			copy_size += copy;
+			continue;
+		}
+
+		/*
+		 * Otherwise we need at least IWL_HCMD_SCRATCHBUF_SIZE copied
+		 * in total (for the scratchbuf handling), but copy up to what
+		 * we can fit into the payload for debug dump purposes.
+		 */
+		copy = min_t(int, TFD_MAX_PAYLOAD_SIZE - cmd_pos, cmd->len[i]);
+
+		memcpy((u8 *)out_cmd + cmd_pos, cmd->data[i], copy);
+		cmd_pos += copy;
+
+		/* However, treat copy_size the proper way, we need it below */
+		if (copy_size < IWL_HCMD_SCRATCHBUF_SIZE) {
+			copy = IWL_HCMD_SCRATCHBUF_SIZE - copy_size;
+
+			if (copy > cmd->len[i])
+				copy = cmd->len[i];
+			copy_size += copy;
 		}
 	}
 
@@ -1345,7 +1362,7 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 	memcpy(&txq->scratchbufs[q->write_ptr], &out_cmd->hdr, scratch_size);
 	iwl_pcie_txq_build_tfd(trans, txq,
 			       iwl_pcie_get_scratchbuf_dma(txq, q->write_ptr),
-			       scratch_size, 1);
+			       scratch_size, true);
 
 	/* map first command fragment, if any remains */
 	if (copy_size > scratch_size) {
@@ -1361,7 +1378,7 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 		}
 
 		iwl_pcie_txq_build_tfd(trans, txq, phys_addr,
-				       copy_size - scratch_size, 0);
+				       copy_size - scratch_size, false);
 	}
 
 	/* map the remaining (adjusted) nocopy/dup fragments */
@@ -1384,7 +1401,7 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 			goto out;
 		}
 
-		iwl_pcie_txq_build_tfd(trans, txq, phys_addr, cmdlen[i], 0);
+		iwl_pcie_txq_build_tfd(trans, txq, phys_addr, cmdlen[i], false);
 	}
 
 	out_meta->flags = cmd->flags;
@@ -1392,8 +1409,6 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 		kfree(txq->entries[idx].free_buf);
 	txq->entries[idx].free_buf = dup_buf;
 
-	txq->need_update = 1;
-
 	trace_iwlwifi_dev_hcmd(trans->dev, cmd, cmd_size, &out_cmd->hdr);
 
 	/* start timer if queue currently empty */
@@ -1405,9 +1420,11 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 	/*
 	 * wake up the NIC to make sure that the firmware will see the host
 	 * command - we will let the NIC sleep once all the host commands
-	 * returned.
+	 * returned. This needs to be done only on NICs that have
+	 * apmg_wake_up_wa set.
 	 */
-	if (!trans_pcie->cmd_in_flight) {
+	if (trans->cfg->base_params->apmg_wake_up_wa &&
+	    !trans_pcie->cmd_in_flight) {
 		trans_pcie->cmd_in_flight = true;
 		__iwl_trans_pcie_set_bit(trans, CSR_GP_CNTRL,
 					 CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
@@ -1427,7 +1444,7 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 	}
 
 	/* Increment and update queue's write index */
-	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
+	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr);
 	iwl_pcie_txq_inc_wr_ptr(trans, txq);
 
 	spin_unlock_irqrestore(&trans_pcie->reg_lock, flags);
@@ -1583,7 +1600,7 @@ static int iwl_pcie_send_hcmd_sync(struct iwl_trans *trans,
 			       get_cmd_string(trans_pcie, cmd->id));
 		ret = -ETIMEDOUT;
 
-		iwl_write_prph(trans, DEVICE_SET_NMI_REG, 1);
+		iwl_force_nmi(trans);
 		iwl_trans_fw_error(trans);
 
 		goto cancel;
@@ -1661,7 +1678,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	dma_addr_t tb0_phys, tb1_phys, scratch_phys;
 	void *tb1_addr;
 	u16 len, tb1_len, tb2_len;
-	u8 wait_write_ptr = 0;
+	bool wait_write_ptr;
 	__le16 fc = hdr->frame_control;
 	u8 hdr_len = ieee80211_hdrlen(fc);
 	u16 wifi_seq;
@@ -1722,7 +1739,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	memcpy(&txq->scratchbufs[q->write_ptr], &dev_cmd->hdr,
 	       IWL_HCMD_SCRATCHBUF_SIZE);
 	iwl_pcie_txq_build_tfd(trans, txq, tb0_phys,
-			       IWL_HCMD_SCRATCHBUF_SIZE, 1);
+			       IWL_HCMD_SCRATCHBUF_SIZE, true);
 
 	/* there must be data left over for TB1 or this code must be changed */
 	BUILD_BUG_ON(sizeof(struct iwl_tx_cmd) < IWL_HCMD_SCRATCHBUF_SIZE);
@@ -1732,7 +1749,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	tb1_phys = dma_map_single(trans->dev, tb1_addr, tb1_len, DMA_TO_DEVICE);
 	if (unlikely(dma_mapping_error(trans->dev, tb1_phys)))
 		goto out_err;
-	iwl_pcie_txq_build_tfd(trans, txq, tb1_phys, tb1_len, 0);
+	iwl_pcie_txq_build_tfd(trans, txq, tb1_phys, tb1_len, false);
 
 	/*
 	 * Set up TFD's third entry to point directly to remainder
@@ -1748,7 +1765,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 					   &txq->tfds[q->write_ptr]);
 			goto out_err;
 		}
-		iwl_pcie_txq_build_tfd(trans, txq, tb2_phys, tb2_len, 0);
+		iwl_pcie_txq_build_tfd(trans, txq, tb2_phys, tb2_len, false);
 	}
 
 	/* Set up entry for this TFD in Tx byte-count array */
@@ -1762,12 +1779,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	trace_iwlwifi_dev_tx_data(trans->dev, skb,
 				  skb->data + hdr_len, tb2_len);
 
-	if (!ieee80211_has_morefrags(fc)) {
-		txq->need_update = 1;
-	} else {
-		wait_write_ptr = 1;
-		txq->need_update = 0;
-	}
+	wait_write_ptr = ieee80211_has_morefrags(fc);
 
 	/* start timer if queue currently empty */
 	if (txq->need_update && q->read_ptr == q->write_ptr &&
@@ -1775,22 +1787,19 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 		mod_timer(&txq->stuck_timer, jiffies + trans_pcie->wd_timeout);
 
 	/* Tell device the write index *just past* this latest filled TFD */
-	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
-	iwl_pcie_txq_inc_wr_ptr(trans, txq);
+	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr);
+	if (!wait_write_ptr)
+		iwl_pcie_txq_inc_wr_ptr(trans, txq);
 
 	/*
 	 * At this point the frame is "transmitted" successfully
-	 * and we will get a TX status notification eventually,
-	 * regardless of the value of ret. "ret" only indicates
-	 * whether or not we should update the write pointer.
+	 * and we will get a TX status notification eventually.
 	 */
 	if (iwl_queue_space(q) < q->high_mark) {
-		if (wait_write_ptr) {
-			txq->need_update = 1;
+		if (wait_write_ptr)
 			iwl_pcie_txq_inc_wr_ptr(trans, txq);
-		} else {
+		else
 			iwl_stop_queue(trans, txq);
-		}
 	}
 	spin_unlock(&txq->lock);
 	return 0;
diff --git a/drivers/net/wireless/libertas/cfg.c b/drivers/net/wireless/libertas/cfg.c
index 54e344aed6e0..47a998d8f99e 100644
--- a/drivers/net/wireless/libertas/cfg.c
+++ b/drivers/net/wireless/libertas/cfg.c
@@ -1006,9 +1006,8 @@ struct cmd_key_material {
 } __packed;
 
 static int lbs_set_key_material(struct lbs_private *priv,
-				int key_type,
-				int key_info,
-				u8 *key, u16 key_len)
+				int key_type, int key_info,
+				const u8 *key, u16 key_len)
 {
 	struct cmd_key_material cmd;
 	int ret;
@@ -1610,7 +1609,7 @@ static int lbs_cfg_del_key(struct wiphy *wiphy, struct net_device *netdev,
  */
 
 static int lbs_cfg_get_station(struct wiphy *wiphy, struct net_device *dev,
-			      u8 *mac, struct station_info *sinfo)
+			       const u8 *mac, struct station_info *sinfo)
 {
 	struct lbs_private *priv = wiphy_priv(wiphy);
 	s8 signal, noise;
diff --git a/drivers/net/wireless/libertas/defs.h b/drivers/net/wireless/libertas/defs.h
index ab966f08024a..407784aca627 100644
--- a/drivers/net/wireless/libertas/defs.h
+++ b/drivers/net/wireless/libertas/defs.h
@@ -90,7 +90,8 @@ do { if ((lbs_debug & (grp)) == (grp)) \
 #define lbs_deb_cfg80211(fmt, args...)  LBS_DEB_LL(LBS_DEB_CFG80211, " cfg80211", fmt, ##args)
 
 #ifdef DEBUG
-static inline void lbs_deb_hex(unsigned int grp, const char *prompt, u8 *buf, int len)
+static inline void lbs_deb_hex(unsigned int grp, const char *prompt,
+			       const u8 *buf, int len)
 {
 	int i = 0;
 
diff --git a/drivers/net/wireless/libertas/rx.c b/drivers/net/wireless/libertas/rx.c
index c7366b07b568..e446fed7b345 100644
--- a/drivers/net/wireless/libertas/rx.c
+++ b/drivers/net/wireless/libertas/rx.c
@@ -71,8 +71,10 @@ int lbs_process_rxed_packet(struct lbs_private *priv, struct sk_buff *skb)
 
 	skb->ip_summed = CHECKSUM_NONE;
 
-	if (priv->wdev->iftype == NL80211_IFTYPE_MONITOR)
-		return process_rxed_802_11_packet(priv, skb);
+	if (priv->wdev->iftype == NL80211_IFTYPE_MONITOR) {
+		ret = process_rxed_802_11_packet(priv, skb);
+		goto done;
+	}
 
 	p_rx_pd = (struct rxpd *) skb->data;
 	p_rx_pkt = (struct rxpackethdr *) ((u8 *)p_rx_pd +
@@ -86,7 +88,7 @@ int lbs_process_rxed_packet(struct lbs_private *priv, struct sk_buff *skb)
 	if (skb->len < (ETH_HLEN + 8 + sizeof(struct rxpd))) {
 		lbs_deb_rx("rx err: frame received with bad length\n");
 		dev->stats.rx_length_errors++;
-		ret = 0;
+		ret = -EINVAL;
 		dev_kfree_skb(skb);
 		goto done;
 	}
diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index 9d7a52f5a410..a312c653d116 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -1676,7 +1676,9 @@ static int mac80211_hwsim_ampdu_action(struct ieee80211_hw *hw,
 	return 0;
 }
 
-static void mac80211_hwsim_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void mac80211_hwsim_flush(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 u32 queues, bool drop)
 {
 	/* Not implemented, queues only on kernel side */
 }
@@ -2056,6 +2058,7 @@ static int mac80211_hwsim_create_radio(int channels, const char *reg_alpha2,
 			    WIPHY_FLAG_AP_UAPSD |
 			    WIPHY_FLAG_HAS_CHANNEL_SWITCH;
 	hw->wiphy->features |= NL80211_FEATURE_ACTIVE_MONITOR;
+	hw->wiphy->features |= NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE;
 
 	/* ask mac80211 to reserve space for magic */
 	hw->vif_data_size = sizeof(struct hwsim_vif_priv);
diff --git a/drivers/net/wireless/mwifiex/11ac.c b/drivers/net/wireless/mwifiex/11ac.c
index c92f27aa71ed..706831df1fa2 100644
--- a/drivers/net/wireless/mwifiex/11ac.c
+++ b/drivers/net/wireless/mwifiex/11ac.c
@@ -212,8 +212,7 @@ int mwifiex_cmd_append_11ac_tlv(struct mwifiex_private *priv,
 				      sizeof(struct mwifiex_ie_types_header));
 			memcpy((u8 *)vht_op +
 				sizeof(struct mwifiex_ie_types_header),
-			       (u8 *)bss_desc->bcn_vht_oper +
-			       sizeof(struct ieee_types_header),
+			       (u8 *)bss_desc->bcn_vht_oper,
 			       le16_to_cpu(vht_op->header.len));
 
 			/* negotiate the channel width and central freq
diff --git a/drivers/net/wireless/mwifiex/11n.c b/drivers/net/wireless/mwifiex/11n.c
index d14ead8beca8..e1c2f67ae85e 100644
--- a/drivers/net/wireless/mwifiex/11n.c
+++ b/drivers/net/wireless/mwifiex/11n.c
@@ -345,8 +345,7 @@ mwifiex_cmd_append_11n_tlv(struct mwifiex_private *priv,
 
 			memcpy((u8 *) ht_info +
 			       sizeof(struct mwifiex_ie_types_header),
-			       (u8 *) bss_desc->bcn_ht_oper +
-			       sizeof(struct ieee_types_header),
+			       (u8 *)bss_desc->bcn_ht_oper,
 			       le16_to_cpu(ht_info->header.len));
 
 			if (!(sband->ht_cap.cap &
@@ -750,3 +749,45 @@ void mwifiex_set_ba_params(struct mwifiex_private *priv)
 
 	return;
 }
+
+u8 mwifiex_get_sec_chan_offset(int chan)
+{
+	u8 sec_offset;
+
+	switch (chan) {
+	case 36:
+	case 44:
+	case 52:
+	case 60:
+	case 100:
+	case 108:
+	case 116:
+	case 124:
+	case 132:
+	case 140:
+	case 149:
+	case 157:
+		sec_offset = IEEE80211_HT_PARAM_CHA_SEC_ABOVE;
+		break;
+	case 40:
+	case 48:
+	case 56:
+	case 64:
+	case 104:
+	case 112:
+	case 120:
+	case 128:
+	case 136:
+	case 144:
+	case 153:
+	case 161:
+		sec_offset = IEEE80211_HT_PARAM_CHA_SEC_BELOW;
+		break;
+	case 165:
+	default:
+		sec_offset = IEEE80211_HT_PARAM_CHA_SEC_NONE;
+		break;
+	}
+
+	return sec_offset;
+}
diff --git a/drivers/net/wireless/mwifiex/11n.h b/drivers/net/wireless/mwifiex/11n.h
index 40b007a00f4b..0b73fa08f5d4 100644
--- a/drivers/net/wireless/mwifiex/11n.h
+++ b/drivers/net/wireless/mwifiex/11n.h
@@ -63,6 +63,7 @@ int mwifiex_cmd_amsdu_aggr_ctrl(struct host_cmd_ds_command *cmd,
 				int cmd_action,
 				struct mwifiex_ds_11n_amsdu_aggr_ctrl *aa_ctrl);
 void mwifiex_del_tx_ba_stream_tbl_by_ra(struct mwifiex_private *priv, u8 *ra);
+u8 mwifiex_get_sec_chan_offset(int chan);
 
 static inline u8
 mwifiex_is_station_ampdu_allowed(struct mwifiex_private *priv,
@@ -199,7 +200,7 @@ static inline int mwifiex_is_sta_11n_enabled(struct mwifiex_private *priv,
 }
 
 static inline u8
-mwifiex_tdls_peer_11n_enabled(struct mwifiex_private *priv, u8 *ra)
+mwifiex_tdls_peer_11n_enabled(struct mwifiex_private *priv, const u8 *ra)
 {
 	struct mwifiex_sta_node *node = mwifiex_get_sta_entry(priv, ra);
 	if (node)
diff --git a/drivers/net/wireless/mwifiex/11n_aggr.c b/drivers/net/wireless/mwifiex/11n_aggr.c
index 63211707f939..5b32106182f8 100644
--- a/drivers/net/wireless/mwifiex/11n_aggr.c
+++ b/drivers/net/wireless/mwifiex/11n_aggr.c
@@ -100,6 +100,7 @@ mwifiex_11n_form_amsdu_txpd(struct mwifiex_private *priv,
 			    struct sk_buff *skb)
 {
 	struct txpd *local_tx_pd;
+	struct mwifiex_txinfo *tx_info = MWIFIEX_SKB_TXCB(skb);
 
 	skb_push(skb, sizeof(*local_tx_pd));
 
@@ -118,6 +119,9 @@ mwifiex_11n_form_amsdu_txpd(struct mwifiex_private *priv,
 	local_tx_pd->tx_pkt_length = cpu_to_le16(skb->len -
 						 sizeof(*local_tx_pd));
 
+	if (tx_info->flags & MWIFIEX_BUF_FLAG_TDLS_PKT)
+		local_tx_pd->flags |= MWIFIEX_TXPD_FLAGS_TDLS_PACKET;
+
 	if (local_tx_pd->tx_control == 0)
 		/* TxCtrl set by user or default */
 		local_tx_pd->tx_control = cpu_to_le32(priv->pkt_tx_ctrl);
@@ -160,6 +164,7 @@ mwifiex_11n_aggregate_pkt(struct mwifiex_private *priv,
 	int pad = 0, ret;
 	struct mwifiex_tx_param tx_param;
 	struct txpd *ptx_pd = NULL;
+	struct timeval tv;
 	int headroom = adapter->iface_type == MWIFIEX_USB ? 0 : INTF_HEADER_LEN;
 
 	skb_src = skb_peek(&pra_list->skb_head);
@@ -182,8 +187,14 @@ mwifiex_11n_aggregate_pkt(struct mwifiex_private *priv,
 
 	tx_info_aggr->bss_type = tx_info_src->bss_type;
 	tx_info_aggr->bss_num = tx_info_src->bss_num;
+
+	if (tx_info_src->flags & MWIFIEX_BUF_FLAG_TDLS_PKT)
+		tx_info_aggr->flags |= MWIFIEX_BUF_FLAG_TDLS_PKT;
 	skb_aggr->priority = skb_src->priority;
 
+	do_gettimeofday(&tv);
+	skb_aggr->tstamp = timeval_to_ktime(tv);
+
 	do {
 		/* Check if AMSDU can accommodate this MSDU */
 		if (skb_tailroom(skb_aggr) < (skb_src->len + LLC_SNAP_LEN))
@@ -236,18 +247,11 @@ mwifiex_11n_aggregate_pkt(struct mwifiex_private *priv,
 		ret = adapter->if_ops.host_to_card(adapter, MWIFIEX_USB_EP_DATA,
 						   skb_aggr, NULL);
 	} else {
-		/*
-		 * Padding per MSDU will affect the length of next
-		 * packet and hence the exact length of next packet
-		 * is uncertain here.
-		 *
-		 * Also, aggregation of transmission buffer, while
-		 * downloading the data to the card, wont gain much
-		 * on the AMSDU packets as the AMSDU packets utilizes
-		 * the transmission buffer space to the maximum
-		 * (adapter->tx_buf_size).
-		 */
-		tx_param.next_pkt_len = 0;
+		if (skb_src)
+			tx_param.next_pkt_len =
+					skb_src->len + sizeof(struct txpd);
+		else
+			tx_param.next_pkt_len = 0;
 
 		ret = adapter->if_ops.host_to_card(adapter, MWIFIEX_TYPE_DATA,
 						   skb_aggr, &tx_param);
diff --git a/drivers/net/wireless/mwifiex/README b/drivers/net/wireless/mwifiex/README
index b9242c3dca43..3b55ce5690a5 100644
--- a/drivers/net/wireless/mwifiex/README
+++ b/drivers/net/wireless/mwifiex/README
@@ -200,4 +200,11 @@ getlog
 
 	cat getlog
 
+fw_dump
+	This command is used to dump firmware memory into files.
+	Separate file will be created for each memory segment.
+	Usage:
+
+	cat fw_dump
+
 ===============================================================================
diff --git a/drivers/net/wireless/mwifiex/cfg80211.c b/drivers/net/wireless/mwifiex/cfg80211.c
index 21ee27ab7b74..e95dec91a561 100644
--- a/drivers/net/wireless/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/mwifiex/cfg80211.c
@@ -994,7 +994,7 @@ mwifiex_dump_station_info(struct mwifiex_private *priv,
  */
 static int
 mwifiex_cfg80211_get_station(struct wiphy *wiphy, struct net_device *dev,
-			     u8 *mac, struct station_info *sinfo)
+			     const u8 *mac, struct station_info *sinfo)
 {
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
 
@@ -1270,7 +1270,7 @@ static int mwifiex_cfg80211_change_beacon(struct wiphy *wiphy,
  */
 static int
 mwifiex_cfg80211_del_station(struct wiphy *wiphy, struct net_device *dev,
-			     u8 *mac)
+			     const u8 *mac)
 {
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
 	struct mwifiex_sta_node *sta_node;
@@ -2629,7 +2629,7 @@ static int mwifiex_cfg80211_set_coalesce(struct wiphy *wiphy,
  */
 static int
 mwifiex_cfg80211_tdls_mgmt(struct wiphy *wiphy, struct net_device *dev,
-			   u8 *peer, u8 action_code, u8 dialog_token,
+			   const u8 *peer, u8 action_code, u8 dialog_token,
 			   u16 status_code, u32 peer_capability,
 			   const u8 *extra_ies, size_t extra_ies_len)
 {
@@ -2701,7 +2701,7 @@ mwifiex_cfg80211_tdls_mgmt(struct wiphy *wiphy, struct net_device *dev,
 
 static int
 mwifiex_cfg80211_tdls_oper(struct wiphy *wiphy, struct net_device *dev,
-			   u8 *peer, enum nl80211_tdls_operation action)
+			   const u8 *peer, enum nl80211_tdls_operation action)
 {
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
 
@@ -2748,9 +2748,8 @@ mwifiex_cfg80211_tdls_oper(struct wiphy *wiphy, struct net_device *dev,
 }
 
 static int
-mwifiex_cfg80211_add_station(struct wiphy *wiphy,
-			     struct net_device *dev,
-			     u8 *mac, struct station_parameters *params)
+mwifiex_cfg80211_add_station(struct wiphy *wiphy, struct net_device *dev,
+			     const u8 *mac, struct station_parameters *params)
 {
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
 
@@ -2765,9 +2764,9 @@ mwifiex_cfg80211_add_station(struct wiphy *wiphy,
 }
 
 static int
-mwifiex_cfg80211_change_station(struct wiphy *wiphy,
-				struct net_device *dev,
-				u8 *mac, struct station_parameters *params)
+mwifiex_cfg80211_change_station(struct wiphy *wiphy, struct net_device *dev,
+				const u8 *mac,
+				struct station_parameters *params)
 {
 	int ret;
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
diff --git a/drivers/net/wireless/mwifiex/cmdevt.c b/drivers/net/wireless/mwifiex/cmdevt.c
index 1062c918a7bf..8dee6c86f4f1 100644
--- a/drivers/net/wireless/mwifiex/cmdevt.c
+++ b/drivers/net/wireless/mwifiex/cmdevt.c
@@ -955,8 +955,6 @@ mwifiex_cmd_timeout_func(unsigned long function_context)
 			adapter->cmd_wait_q.status = -ETIMEDOUT;
 			wake_up_interruptible(&adapter->cmd_wait_q.wait);
 			mwifiex_cancel_pending_ioctl(adapter);
-			/* reset cmd_sent flag to unblock new commands */
-			adapter->cmd_sent = false;
 		}
 	}
 	if (adapter->hw_status == MWIFIEX_HW_STATUS_INITIALIZING)
diff --git a/drivers/net/wireless/mwifiex/debugfs.c b/drivers/net/wireless/mwifiex/debugfs.c
index b8a49aad12fd..7b419bbcd544 100644
--- a/drivers/net/wireless/mwifiex/debugfs.c
+++ b/drivers/net/wireless/mwifiex/debugfs.c
@@ -257,6 +257,29 @@ free_and_exit:
 }
 
 /*
+ * Proc firmware dump read handler.
+ *
+ * This function is called when the 'fw_dump' file is opened for
+ * reading.
+ * This function dumps firmware memory in different files
+ * (ex. DTCM, ITCM, SQRAM etc.) based on the the segments for
+ * debugging.
+ */
+static ssize_t
+mwifiex_fw_dump_read(struct file *file, char __user *ubuf,
+		     size_t count, loff_t *ppos)
+{
+	struct mwifiex_private *priv = file->private_data;
+
+	if (!priv->adapter->if_ops.fw_dump)
+		return -EIO;
+
+	priv->adapter->if_ops.fw_dump(priv->adapter);
+
+	return 0;
+}
+
+/*
  * Proc getlog file read handler.
  *
  * This function is called when the 'getlog' file is opened for reading
@@ -699,6 +722,7 @@ static const struct file_operations mwifiex_dfs_##name##_fops = {       \
 MWIFIEX_DFS_FILE_READ_OPS(info);
 MWIFIEX_DFS_FILE_READ_OPS(debug);
 MWIFIEX_DFS_FILE_READ_OPS(getlog);
+MWIFIEX_DFS_FILE_READ_OPS(fw_dump);
 MWIFIEX_DFS_FILE_OPS(regrdwr);
 MWIFIEX_DFS_FILE_OPS(rdeeprom);
 
@@ -722,6 +746,7 @@ mwifiex_dev_debugfs_init(struct mwifiex_private *priv)
 	MWIFIEX_DFS_ADD_FILE(getlog);
 	MWIFIEX_DFS_ADD_FILE(regrdwr);
 	MWIFIEX_DFS_ADD_FILE(rdeeprom);
+	MWIFIEX_DFS_ADD_FILE(fw_dump);
 }
 
 /*
diff --git a/drivers/net/wireless/mwifiex/decl.h b/drivers/net/wireless/mwifiex/decl.h
index e7b3e16e5d34..38da6ff6f416 100644
--- a/drivers/net/wireless/mwifiex/decl.h
+++ b/drivers/net/wireless/mwifiex/decl.h
@@ -42,12 +42,12 @@
 #define MWIFIEX_MAX_TX_BASTREAM_SUPPORTED	2
 #define MWIFIEX_MAX_RX_BASTREAM_SUPPORTED	16
 
-#define MWIFIEX_STA_AMPDU_DEF_TXWINSIZE        16
-#define MWIFIEX_STA_AMPDU_DEF_RXWINSIZE        32
+#define MWIFIEX_STA_AMPDU_DEF_TXWINSIZE        64
+#define MWIFIEX_STA_AMPDU_DEF_RXWINSIZE        64
 #define MWIFIEX_UAP_AMPDU_DEF_TXWINSIZE        32
 #define MWIFIEX_UAP_AMPDU_DEF_RXWINSIZE        16
-#define MWIFIEX_11AC_STA_AMPDU_DEF_TXWINSIZE   32
-#define MWIFIEX_11AC_STA_AMPDU_DEF_RXWINSIZE   48
+#define MWIFIEX_11AC_STA_AMPDU_DEF_TXWINSIZE   64
+#define MWIFIEX_11AC_STA_AMPDU_DEF_RXWINSIZE   64
 #define MWIFIEX_11AC_UAP_AMPDU_DEF_TXWINSIZE   48
 #define MWIFIEX_11AC_UAP_AMPDU_DEF_RXWINSIZE   32
 
diff --git a/drivers/net/wireless/mwifiex/fw.h b/drivers/net/wireless/mwifiex/fw.h
index b485dc1ae5eb..3175dd04834b 100644
--- a/drivers/net/wireless/mwifiex/fw.h
+++ b/drivers/net/wireless/mwifiex/fw.h
@@ -169,6 +169,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 #define TLV_TYPE_GWK_CIPHER         (PROPRIETARY_TLV_BASE_ID + 146)
 #define TLV_TYPE_COALESCE_RULE      (PROPRIETARY_TLV_BASE_ID + 154)
 #define TLV_TYPE_KEY_PARAM_V2       (PROPRIETARY_TLV_BASE_ID + 156)
+#define TLV_TYPE_TDLS_IDLE_TIMEOUT  (PROPRIETARY_TLV_BASE_ID + 194)
 #define TLV_TYPE_FW_API_REV         (PROPRIETARY_TLV_BASE_ID + 199)
 
 #define MWIFIEX_TX_DATA_BUF_SIZE_2K        2048
@@ -229,6 +230,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 #define ISENABLED_40MHZ_INTOLERANT(Dot11nDevCap) (Dot11nDevCap & BIT(8))
 #define ISSUPP_RXLDPC(Dot11nDevCap) (Dot11nDevCap & BIT(22))
 #define ISSUPP_BEAMFORMING(Dot11nDevCap) (Dot11nDevCap & BIT(30))
+#define ISALLOWED_CHANWIDTH40(ht_param) (ht_param & BIT(2))
 
 /* httxcfg bitmap
  * 0		reserved
@@ -403,7 +405,7 @@ enum P2P_MODES {
 #define HS_CFG_CANCEL			0xffffffff
 #define HS_CFG_COND_DEF			0x00000000
 #define HS_CFG_GPIO_DEF			0xff
-#define HS_CFG_GAP_DEF			0
+#define HS_CFG_GAP_DEF			0xff
 #define HS_CFG_COND_BROADCAST_DATA	0x00000001
 #define HS_CFG_COND_UNICAST_DATA	0x00000002
 #define HS_CFG_COND_MAC_EVENT		0x00000004
@@ -487,6 +489,7 @@ enum P2P_MODES {
 #define EVENT_UAP_MIC_COUNTERMEASURES   0x0000004c
 #define EVENT_HOSTWAKE_STAIE		0x0000004d
 #define EVENT_CHANNEL_SWITCH_ANN        0x00000050
+#define EVENT_TDLS_GENERIC_EVENT        0x00000052
 #define EVENT_EXT_SCAN_REPORT           0x00000058
 #define EVENT_REMAIN_ON_CHAN_EXPIRED    0x0000005f
 
@@ -519,6 +522,7 @@ enum P2P_MODES {
 #define ACT_TDLS_DELETE            0x00
 #define ACT_TDLS_CREATE            0x01
 #define ACT_TDLS_CONFIG            0x02
+#define TDLS_EVENT_LINK_TEAR_DOWN  3
 
 #define MWIFIEX_FW_V15		   15
 
@@ -535,6 +539,7 @@ struct mwifiex_ie_types_data {
 #define MWIFIEX_TxPD_POWER_MGMT_NULL_PACKET 0x01
 #define MWIFIEX_TxPD_POWER_MGMT_LAST_PACKET 0x08
 #define MWIFIEX_TXPD_FLAGS_TDLS_PACKET      0x10
+#define MWIFIEX_RXPD_FLAGS_TDLS_PACKET      0x01
 
 struct txpd {
 	u8 bss_type;
@@ -577,7 +582,7 @@ struct rxpd {
 	 * [Bit 7] Reserved
 	 */
 	u8 ht_info;
-	u8 reserved;
+	u8 flags;
 } __packed;
 
 struct uap_txpd {
@@ -708,6 +713,13 @@ struct mwifiex_ie_types_vendor_param_set {
 	u8 ie[MWIFIEX_MAX_VSIE_LEN];
 };
 
+#define MWIFIEX_TDLS_IDLE_TIMEOUT	60
+
+struct mwifiex_ie_types_tdls_idle_timeout {
+	struct mwifiex_ie_types_header header;
+	__le16 value;
+} __packed;
+
 struct mwifiex_ie_types_rsn_param_set {
 	struct mwifiex_ie_types_header header;
 	u8 rsn_ie[1];
@@ -1745,6 +1757,15 @@ struct host_cmd_ds_802_11_subsc_evt {
 	__le16 events;
 } __packed;
 
+struct mwifiex_tdls_generic_event {
+	__le16 type;
+	u8 peer_mac[ETH_ALEN];
+	union {
+		__le16 reason_code;
+		__le16 reserved;
+	} u;
+} __packed;
+
 struct mwifiex_ie {
 	__le16 ie_index;
 	__le16 mgmt_subtype_mask;
diff --git a/drivers/net/wireless/mwifiex/ioctl.h b/drivers/net/wireless/mwifiex/ioctl.h
index ee494db54060..1b576722671d 100644
--- a/drivers/net/wireless/mwifiex/ioctl.h
+++ b/drivers/net/wireless/mwifiex/ioctl.h
@@ -303,7 +303,7 @@ struct mwifiex_ds_ant_cfg {
 	u32 rx_ant;
 };
 
-#define MWIFIEX_NUM_OF_CMD_BUFFER	20
+#define MWIFIEX_NUM_OF_CMD_BUFFER	50
 #define MWIFIEX_SIZE_OF_CMD_BUFFER	2048
 
 enum {
diff --git a/drivers/net/wireless/mwifiex/main.c b/drivers/net/wireless/mwifiex/main.c
index 9c771b3e9918..cbabc12fbda3 100644
--- a/drivers/net/wireless/mwifiex/main.c
+++ b/drivers/net/wireless/mwifiex/main.c
@@ -521,7 +521,6 @@ done:
 		release_firmware(adapter->firmware);
 		adapter->firmware = NULL;
 	}
-	complete(&adapter->fw_load);
 	if (init_failed)
 		mwifiex_free_adapter(adapter);
 	up(sem);
@@ -535,7 +534,6 @@ static int mwifiex_init_hw_fw(struct mwifiex_adapter *adapter)
 {
 	int ret;
 
-	init_completion(&adapter->fw_load);
 	ret = request_firmware_nowait(THIS_MODULE, 1, adapter->fw_name,
 				      adapter->dev, GFP_KERNEL, adapter,
 				      mwifiex_fw_dpc);
diff --git a/drivers/net/wireless/mwifiex/main.h b/drivers/net/wireless/mwifiex/main.h
index d53e1e8c9467..1398afa84064 100644
--- a/drivers/net/wireless/mwifiex/main.h
+++ b/drivers/net/wireless/mwifiex/main.h
@@ -672,6 +672,7 @@ struct mwifiex_if_ops {
 	int (*init_fw_port) (struct mwifiex_adapter *);
 	int (*dnld_fw) (struct mwifiex_adapter *, struct mwifiex_fw_image *);
 	void (*card_reset) (struct mwifiex_adapter *);
+	void (*fw_dump)(struct mwifiex_adapter *);
 	int (*clean_pcie_ring) (struct mwifiex_adapter *adapter);
 };
 
@@ -787,7 +788,6 @@ struct mwifiex_adapter {
 	struct mwifiex_wait_queue cmd_wait_q;
 	u8 scan_wait_q_woken;
 	spinlock_t queue_lock;		/* lock for tx queues */
-	struct completion fw_load;
 	u8 country_code[IEEE80211_COUNTRY_STRING_LEN];
 	u16 max_mgmt_ie_index;
 	u8 scan_delay_cnt;
@@ -910,8 +910,6 @@ int mwifiex_handle_uap_rx_forward(struct mwifiex_private *priv,
 				  struct sk_buff *skb);
 int mwifiex_process_sta_event(struct mwifiex_private *);
 int mwifiex_process_uap_event(struct mwifiex_private *);
-struct mwifiex_sta_node *
-mwifiex_get_sta_entry(struct mwifiex_private *priv, u8 *mac);
 void mwifiex_delete_all_station_list(struct mwifiex_private *priv);
 void *mwifiex_process_sta_txpd(struct mwifiex_private *, struct sk_buff *skb);
 void *mwifiex_process_uap_txpd(struct mwifiex_private *, struct sk_buff *skb);
@@ -1101,7 +1099,7 @@ mwifiex_11h_get_csa_closed_channel(struct mwifiex_private *priv)
 		return 0;
 
 	/* Clear csa channel, if DFS channel move time has passed */
-	if (jiffies > priv->csa_expire_time) {
+	if (time_after(jiffies, priv->csa_expire_time)) {
 		priv->csa_chan = 0;
 		priv->csa_expire_time = 0;
 	}
@@ -1220,26 +1218,26 @@ void mwifiex_dnld_txpwr_table(struct mwifiex_private *priv);
 extern const struct ethtool_ops mwifiex_ethtool_ops;
 
 void mwifiex_del_all_sta_list(struct mwifiex_private *priv);
-void mwifiex_del_sta_entry(struct mwifiex_private *priv, u8 *mac);
+void mwifiex_del_sta_entry(struct mwifiex_private *priv, const u8 *mac);
 void
 mwifiex_set_sta_ht_cap(struct mwifiex_private *priv, const u8 *ies,
 		       int ies_len, struct mwifiex_sta_node *node);
 struct mwifiex_sta_node *
-mwifiex_add_sta_entry(struct mwifiex_private *priv, u8 *mac);
+mwifiex_add_sta_entry(struct mwifiex_private *priv, const u8 *mac);
 struct mwifiex_sta_node *
-mwifiex_get_sta_entry(struct mwifiex_private *priv, u8 *mac);
-int mwifiex_send_tdls_data_frame(struct mwifiex_private *priv, u8 *peer,
+mwifiex_get_sta_entry(struct mwifiex_private *priv, const u8 *mac);
+int mwifiex_send_tdls_data_frame(struct mwifiex_private *priv, const u8 *peer,
 				 u8 action_code, u8 dialog_token,
 				 u16 status_code, const u8 *extra_ies,
 				 size_t extra_ies_len);
-int mwifiex_send_tdls_action_frame(struct mwifiex_private *priv,
-				 u8 *peer, u8 action_code, u8 dialog_token,
-				 u16 status_code, const u8 *extra_ies,
-				 size_t extra_ies_len);
+int mwifiex_send_tdls_action_frame(struct mwifiex_private *priv, const u8 *peer,
+				   u8 action_code, u8 dialog_token,
+				   u16 status_code, const u8 *extra_ies,
+				   size_t extra_ies_len);
 void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 				       u8 *buf, int len);
-int mwifiex_tdls_oper(struct mwifiex_private *priv, u8 *peer, u8 action);
-int mwifiex_get_tdls_link_status(struct mwifiex_private *priv, u8 *mac);
+int mwifiex_tdls_oper(struct mwifiex_private *priv, const u8 *peer, u8 action);
+int mwifiex_get_tdls_link_status(struct mwifiex_private *priv, const u8 *mac);
 void mwifiex_disable_all_tdls_links(struct mwifiex_private *priv);
 bool mwifiex_is_bss_in_11ac_mode(struct mwifiex_private *priv);
 u8 mwifiex_get_center_freq_index(struct mwifiex_private *priv, u8 band,
diff --git a/drivers/net/wireless/mwifiex/pcie.c b/drivers/net/wireless/mwifiex/pcie.c
index a7e8b96b2d90..2cc9b6fca490 100644
--- a/drivers/net/wireless/mwifiex/pcie.c
+++ b/drivers/net/wireless/mwifiex/pcie.c
@@ -50,7 +50,7 @@ mwifiex_map_pci_memory(struct mwifiex_adapter *adapter, struct sk_buff *skb,
 		return -1;
 	}
 	mapping.len = size;
-	memcpy(skb->cb, &mapping, sizeof(mapping));
+	mwifiex_store_mapping(skb, &mapping);
 	return 0;
 }
 
@@ -60,7 +60,7 @@ static void mwifiex_unmap_pci_memory(struct mwifiex_adapter *adapter,
 	struct pcie_service_card *card = adapter->card;
 	struct mwifiex_dma_mapping mapping;
 
-	MWIFIEX_SKB_PACB(skb, &mapping);
+	mwifiex_get_mapping(skb, &mapping);
 	pci_unmap_single(card->dev, mapping.addr, mapping.len, flags);
 }
 
@@ -221,9 +221,6 @@ static void mwifiex_pcie_remove(struct pci_dev *pdev)
 	if (!adapter || !adapter->priv_num)
 		return;
 
-	/* In case driver is removed when asynchronous FW load is in progress */
-	wait_for_completion(&adapter->fw_load);
-
 	if (user_rmmod) {
 #ifdef CONFIG_PM_SLEEP
 		if (adapter->is_suspended)
@@ -1074,6 +1071,7 @@ static int mwifiex_pcie_send_data_complete(struct mwifiex_adapter *adapter)
  * is mapped to PCI device memory. Tx ring pointers are advanced accordingly.
  * Download ready interrupt to FW is deffered if Tx ring is not full and
  * additional payload can be accomodated.
+ * Caller must ensure tx_param parameter to this function is not NULL.
  */
 static int
 mwifiex_pcie_send_data(struct mwifiex_adapter *adapter, struct sk_buff *skb,
diff --git a/drivers/net/wireless/mwifiex/scan.c b/drivers/net/wireless/mwifiex/scan.c
index 7b3af3d29ded..45c5b3450cf5 100644
--- a/drivers/net/wireless/mwifiex/scan.c
+++ b/drivers/net/wireless/mwifiex/scan.c
@@ -29,9 +29,6 @@
 #define MWIFIEX_MAX_CHANNELS_PER_SPECIFIC_SCAN   14
 
 #define MWIFIEX_DEF_CHANNELS_PER_SCAN_CMD	4
-#define MWIFIEX_LIMIT_1_CHANNEL_PER_SCAN_CMD	15
-#define MWIFIEX_LIMIT_2_CHANNELS_PER_SCAN_CMD	27
-#define MWIFIEX_LIMIT_3_CHANNELS_PER_SCAN_CMD	35
 
 /* Memory needed to store a max sized Channel List TLV for a firmware scan */
 #define CHAN_TLV_MAX_SIZE  (sizeof(struct mwifiex_ie_types_header)         \
@@ -1055,20 +1052,10 @@ mwifiex_config_scan(struct mwifiex_private *priv,
 
 	/*
 	 * In associated state we will reduce the number of channels scanned per
-	 * scan command to avoid any traffic delay/loss. This number is decided
-	 * based on total number of channels to be scanned due to constraints
-	 * of command buffers.
+	 * scan command to 1 to avoid any traffic delay/loss.
 	 */
-	if (priv->media_connected) {
-		if (chan_num < MWIFIEX_LIMIT_1_CHANNEL_PER_SCAN_CMD)
+	if (priv->media_connected)
 			*max_chan_per_scan = 1;
-		else if (chan_num < MWIFIEX_LIMIT_2_CHANNELS_PER_SCAN_CMD)
-			*max_chan_per_scan = 2;
-		else if (chan_num < MWIFIEX_LIMIT_3_CHANNELS_PER_SCAN_CMD)
-			*max_chan_per_scan = 3;
-		else
-			*max_chan_per_scan = 4;
-	}
 }
 
 /*
@@ -1353,23 +1340,17 @@ int mwifiex_update_bss_desc_with_ie(struct mwifiex_adapter *adapter,
 					      bss_entry->beacon_buf);
 			break;
 		case WLAN_EID_BSS_COEX_2040:
-			bss_entry->bcn_bss_co_2040 = current_ptr +
-				sizeof(struct ieee_types_header);
-			bss_entry->bss_co_2040_offset = (u16) (current_ptr +
-					sizeof(struct ieee_types_header) -
-						bss_entry->beacon_buf);
+			bss_entry->bcn_bss_co_2040 = current_ptr;
+			bss_entry->bss_co_2040_offset =
+				(u16) (current_ptr - bss_entry->beacon_buf);
 			break;
 		case WLAN_EID_EXT_CAPABILITY:
-			bss_entry->bcn_ext_cap = current_ptr +
-				sizeof(struct ieee_types_header);
-			bss_entry->ext_cap_offset = (u16) (current_ptr +
-					sizeof(struct ieee_types_header) -
-					bss_entry->beacon_buf);
+			bss_entry->bcn_ext_cap = current_ptr;
+			bss_entry->ext_cap_offset =
+				(u16) (current_ptr - bss_entry->beacon_buf);
 			break;
 		case WLAN_EID_OPMODE_NOTIF:
-			bss_entry->oper_mode =
-				(void *)(current_ptr +
-					 sizeof(struct ieee_types_header));
+			bss_entry->oper_mode = (void *)current_ptr;
 			bss_entry->oper_mode_offset =
 					(u16)((u8 *)bss_entry->oper_mode -
 					      bss_entry->beacon_buf);
@@ -1757,6 +1738,19 @@ mwifiex_parse_single_response_buf(struct mwifiex_private *priv, u8 **bss_info,
 	return 0;
 }
 
+static void mwifiex_complete_scan(struct mwifiex_private *priv)
+{
+	struct mwifiex_adapter *adapter = priv->adapter;
+
+	if (adapter->curr_cmd->wait_q_enabled) {
+		adapter->cmd_wait_q.status = 0;
+		if (!priv->scan_request) {
+			dev_dbg(adapter->dev, "complete internal scan\n");
+			mwifiex_complete_cmd(adapter, adapter->curr_cmd);
+		}
+	}
+}
+
 static void mwifiex_check_next_scan_command(struct mwifiex_private *priv)
 {
 	struct mwifiex_adapter *adapter = priv->adapter;
@@ -1770,16 +1764,9 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv)
 		adapter->scan_processing = false;
 		spin_unlock_irqrestore(&adapter->mwifiex_cmd_lock, flags);
 
-		/* Need to indicate IOCTL complete */
-		if (adapter->curr_cmd->wait_q_enabled) {
-			adapter->cmd_wait_q.status = 0;
-			if (!priv->scan_request) {
-				dev_dbg(adapter->dev,
-					"complete internal scan\n");
-				mwifiex_complete_cmd(adapter,
-						     adapter->curr_cmd);
-			}
-		}
+		if (!adapter->ext_scan)
+			mwifiex_complete_scan(priv);
+
 		if (priv->report_scan_result)
 			priv->report_scan_result = false;
 
@@ -1984,6 +1971,9 @@ int mwifiex_cmd_802_11_scan_ext(struct mwifiex_private *priv,
 int mwifiex_ret_802_11_scan_ext(struct mwifiex_private *priv)
 {
 	dev_dbg(priv->adapter->dev, "info: EXT scan returns successfully\n");
+
+	mwifiex_complete_scan(priv);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/mwifiex/sdio.c b/drivers/net/wireless/mwifiex/sdio.c
index d206f04d4994..4ce3d7b33991 100644
--- a/drivers/net/wireless/mwifiex/sdio.c
+++ b/drivers/net/wireless/mwifiex/sdio.c
@@ -85,6 +85,8 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id)
 		card->supports_sdio_new_mode = data->supports_sdio_new_mode;
 		card->has_control_mask = data->has_control_mask;
 		card->tx_buf_size = data->tx_buf_size;
+		card->mp_tx_agg_buf_size = data->mp_tx_agg_buf_size;
+		card->mp_rx_agg_buf_size = data->mp_rx_agg_buf_size;
 	}
 
 	sdio_claim_host(func);
@@ -177,9 +179,6 @@ mwifiex_sdio_remove(struct sdio_func *func)
 	if (!adapter || !adapter->priv_num)
 		return;
 
-	/* In case driver is removed when asynchronous FW load is in progress */
-	wait_for_completion(&adapter->fw_load);
-
 	if (user_rmmod) {
 		if (adapter->is_suspended)
 			mwifiex_sdio_resume(adapter->dev);
@@ -1679,8 +1678,12 @@ static int mwifiex_sdio_host_to_card(struct mwifiex_adapter *adapter,
 	if (ret) {
 		if (type == MWIFIEX_TYPE_CMD)
 			adapter->cmd_sent = false;
-		if (type == MWIFIEX_TYPE_DATA)
+		if (type == MWIFIEX_TYPE_DATA) {
 			adapter->data_sent = false;
+			/* restore curr_wr_port in error cases */
+			card->curr_wr_port = port;
+			card->mp_wr_bitmap |= (u32)(1 << card->curr_wr_port);
+		}
 	} else {
 		if (type == MWIFIEX_TYPE_DATA) {
 			if (!(card->mp_wr_bitmap & (1 << card->curr_wr_port)))
@@ -1842,8 +1845,8 @@ static int mwifiex_init_sdio(struct mwifiex_adapter *adapter)
 	card->mpa_rx.len_arr = kzalloc(sizeof(*card->mpa_rx.len_arr) *
 				       card->mp_agg_pkt_limit, GFP_KERNEL);
 	ret = mwifiex_alloc_sdio_mpa_buffers(adapter,
-					     SDIO_MP_TX_AGGR_DEF_BUF_SIZE,
-					     SDIO_MP_RX_AGGR_DEF_BUF_SIZE);
+					     card->mp_tx_agg_buf_size,
+					     card->mp_rx_agg_buf_size);
 	if (ret) {
 		dev_err(adapter->dev, "failed to alloc sdio mp-a buffers\n");
 		kfree(card->mp_regs);
diff --git a/drivers/net/wireless/mwifiex/sdio.h b/drivers/net/wireless/mwifiex/sdio.h
index c71201b2e2a3..6eea30b43ed7 100644
--- a/drivers/net/wireless/mwifiex/sdio.h
+++ b/drivers/net/wireless/mwifiex/sdio.h
@@ -64,10 +64,8 @@
 #define UP_LD_CMD_PORT_HOST_INT_STATUS	(0x40U)
 #define DN_LD_CMD_PORT_HOST_INT_STATUS	(0x80U)
 
-#define SDIO_MP_TX_AGGR_DEF_BUF_SIZE        (8192)	/* 8K */
-
-/* Multi port RX aggregation buffer size */
-#define SDIO_MP_RX_AGGR_DEF_BUF_SIZE        (16384)	/* 16K */
+#define MWIFIEX_MP_AGGR_BUF_SIZE_16K	(16384)
+#define MWIFIEX_MP_AGGR_BUF_SIZE_32K	(32768)
 
 /* Misc. Config Register : Auto Re-enable interrupts */
 #define AUTO_RE_ENABLE_INT              BIT(4)
@@ -234,6 +232,8 @@ struct sdio_mmc_card {
 	bool supports_sdio_new_mode;
 	bool has_control_mask;
 	u16 tx_buf_size;
+	u32 mp_tx_agg_buf_size;
+	u32 mp_rx_agg_buf_size;
 
 	u32 mp_rd_bitmap;
 	u32 mp_wr_bitmap;
@@ -258,6 +258,8 @@ struct mwifiex_sdio_device {
 	bool supports_sdio_new_mode;
 	bool has_control_mask;
 	u16 tx_buf_size;
+	u32 mp_tx_agg_buf_size;
+	u32 mp_rx_agg_buf_size;
 };
 
 static const struct mwifiex_sdio_card_reg mwifiex_reg_sd87xx = {
@@ -315,6 +317,8 @@ static const struct mwifiex_sdio_device mwifiex_sdio_sd8786 = {
 	.supports_sdio_new_mode = false,
 	.has_control_mask = true,
 	.tx_buf_size = MWIFIEX_TX_DATA_BUF_SIZE_2K,
+	.mp_tx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_16K,
+	.mp_rx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_16K,
 };
 
 static const struct mwifiex_sdio_device mwifiex_sdio_sd8787 = {
@@ -325,6 +329,8 @@ static const struct mwifiex_sdio_device mwifiex_sdio_sd8787 = {
 	.supports_sdio_new_mode = false,
 	.has_control_mask = true,
 	.tx_buf_size = MWIFIEX_TX_DATA_BUF_SIZE_2K,
+	.mp_tx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_16K,
+	.mp_rx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_16K,
 };
 
 static const struct mwifiex_sdio_device mwifiex_sdio_sd8797 = {
@@ -335,6 +341,8 @@ static const struct mwifiex_sdio_device mwifiex_sdio_sd8797 = {
 	.supports_sdio_new_mode = false,
 	.has_control_mask = true,
 	.tx_buf_size = MWIFIEX_TX_DATA_BUF_SIZE_2K,
+	.mp_tx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_16K,
+	.mp_rx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_16K,
 };
 
 static const struct mwifiex_sdio_device mwifiex_sdio_sd8897 = {
@@ -345,6 +353,8 @@ static const struct mwifiex_sdio_device mwifiex_sdio_sd8897 = {
 	.supports_sdio_new_mode = true,
 	.has_control_mask = false,
 	.tx_buf_size = MWIFIEX_TX_DATA_BUF_SIZE_4K,
+	.mp_tx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_32K,
+	.mp_rx_agg_buf_size = MWIFIEX_MP_AGGR_BUF_SIZE_32K,
 };
 
 /*
diff --git a/drivers/net/wireless/mwifiex/sta_cmd.c b/drivers/net/wireless/mwifiex/sta_cmd.c
index e3cac1495cc7..88202ce0c139 100644
--- a/drivers/net/wireless/mwifiex/sta_cmd.c
+++ b/drivers/net/wireless/mwifiex/sta_cmd.c
@@ -1546,6 +1546,7 @@ mwifiex_cmd_tdls_oper(struct mwifiex_private *priv,
 	struct mwifiex_ie_types_extcap *extcap;
 	struct mwifiex_ie_types_vhtcap *vht_capab;
 	struct mwifiex_ie_types_aid *aid;
+	struct mwifiex_ie_types_tdls_idle_timeout *timeout;
 	u8 *pos, qos_info;
 	u16 config_len = 0;
 	struct station_parameters *params = priv->sta_params;
@@ -1643,6 +1644,12 @@ mwifiex_cmd_tdls_oper(struct mwifiex_private *priv,
 			config_len += sizeof(struct mwifiex_ie_types_aid);
 		}
 
+		timeout = (void *)(pos + config_len);
+		timeout->header.type = cpu_to_le16(TLV_TYPE_TDLS_IDLE_TIMEOUT);
+		timeout->header.len = cpu_to_le16(sizeof(timeout->value));
+		timeout->value = cpu_to_le16(MWIFIEX_TDLS_IDLE_TIMEOUT);
+		config_len += sizeof(struct mwifiex_ie_types_tdls_idle_timeout);
+
 		break;
 	default:
 		dev_err(priv->adapter->dev, "Unknown TDLS operation\n");
diff --git a/drivers/net/wireless/mwifiex/sta_cmdresp.c b/drivers/net/wireless/mwifiex/sta_cmdresp.c
index bfebb0144df5..577f2979ed8f 100644
--- a/drivers/net/wireless/mwifiex/sta_cmdresp.c
+++ b/drivers/net/wireless/mwifiex/sta_cmdresp.c
@@ -865,14 +865,20 @@ static int mwifiex_ret_tdls_oper(struct mwifiex_private *priv,
 
 	switch (action) {
 	case ACT_TDLS_DELETE:
-		if (reason)
-			dev_err(priv->adapter->dev,
-				"TDLS link delete for %pM failed: reason %d\n",
-				cmd_tdls_oper->peer_mac, reason);
-		else
+		if (reason) {
+			if (!node || reason == TDLS_ERR_LINK_NONEXISTENT)
+				dev_dbg(priv->adapter->dev,
+					"TDLS link delete for %pM failed: reason %d\n",
+					cmd_tdls_oper->peer_mac, reason);
+			else
+				dev_err(priv->adapter->dev,
+					"TDLS link delete for %pM failed: reason %d\n",
+					cmd_tdls_oper->peer_mac, reason);
+		} else {
 			dev_dbg(priv->adapter->dev,
-				"TDLS link config for %pM successful\n",
+				"TDLS link delete for %pM successful\n",
 				cmd_tdls_oper->peer_mac);
+		}
 		break;
 	case ACT_TDLS_CREATE:
 		if (reason) {
diff --git a/drivers/net/wireless/mwifiex/sta_event.c b/drivers/net/wireless/mwifiex/sta_event.c
index 368450cc56c7..f6395ef11a72 100644
--- a/drivers/net/wireless/mwifiex/sta_event.c
+++ b/drivers/net/wireless/mwifiex/sta_event.c
@@ -134,6 +134,46 @@ mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason_code)
 		netif_carrier_off(priv->netdev);
 }
 
+static int mwifiex_parse_tdls_event(struct mwifiex_private *priv,
+				    struct sk_buff *event_skb)
+{
+	int ret = 0;
+	struct mwifiex_adapter *adapter = priv->adapter;
+	struct mwifiex_sta_node *sta_ptr;
+	struct mwifiex_tdls_generic_event *tdls_evt =
+			(void *)event_skb->data + sizeof(adapter->event_cause);
+
+	/* reserved 2 bytes are not mandatory in tdls event */
+	if (event_skb->len < (sizeof(struct mwifiex_tdls_generic_event) -
+			      sizeof(u16) - sizeof(adapter->event_cause))) {
+		dev_err(adapter->dev, "Invalid event length!\n");
+		return -1;
+	}
+
+	sta_ptr = mwifiex_get_sta_entry(priv, tdls_evt->peer_mac);
+	if (!sta_ptr) {
+		dev_err(adapter->dev, "cannot get sta entry!\n");
+		return -1;
+	}
+
+	switch (le16_to_cpu(tdls_evt->type)) {
+	case TDLS_EVENT_LINK_TEAR_DOWN:
+		cfg80211_tdls_oper_request(priv->netdev,
+					   tdls_evt->peer_mac,
+					   NL80211_TDLS_TEARDOWN,
+					   le16_to_cpu(tdls_evt->u.reason_code),
+					   GFP_KERNEL);
+		ret = mwifiex_tdls_oper(priv, tdls_evt->peer_mac,
+					MWIFIEX_TDLS_DISABLE_LINK);
+		queue_work(adapter->workqueue, &adapter->main_work);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
 /*
  * This function handles events generated by firmware.
  *
@@ -459,6 +499,10 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv)
 			false);
 		break;
 
+	case EVENT_TDLS_GENERIC_EVENT:
+		ret = mwifiex_parse_tdls_event(priv, adapter->event_skb);
+		break;
+
 	default:
 		dev_dbg(adapter->dev, "event: unknown event id: %#x\n",
 			eventcause);
diff --git a/drivers/net/wireless/mwifiex/sta_rx.c b/drivers/net/wireless/mwifiex/sta_rx.c
index ed26387eccf5..8b639d7fe6df 100644
--- a/drivers/net/wireless/mwifiex/sta_rx.c
+++ b/drivers/net/wireless/mwifiex/sta_rx.c
@@ -183,6 +183,7 @@ int mwifiex_process_sta_rx_packet(struct mwifiex_private *priv,
 	struct rx_packet_hdr *rx_pkt_hdr;
 	u8 ta[ETH_ALEN];
 	u16 rx_pkt_type, rx_pkt_offset, rx_pkt_length, seq_num;
+	struct mwifiex_sta_node *sta_ptr;
 
 	local_rx_pd = (struct rxpd *) (skb->data);
 	rx_pkt_type = le16_to_cpu(local_rx_pd->rx_pkt_type);
@@ -213,14 +214,25 @@ int mwifiex_process_sta_rx_packet(struct mwifiex_private *priv,
 	 * If the packet is not an unicast packet then send the packet
 	 * directly to os. Don't pass thru rx reordering
 	 */
-	if (!IS_11N_ENABLED(priv) ||
+	if ((!IS_11N_ENABLED(priv) &&
+	     !(ISSUPP_TDLS_ENABLED(priv->adapter->fw_cap_info) &&
+	       !(local_rx_pd->flags & MWIFIEX_RXPD_FLAGS_TDLS_PACKET))) ||
 	    !ether_addr_equal_unaligned(priv->curr_addr, rx_pkt_hdr->eth803_hdr.h_dest)) {
 		mwifiex_process_rx_packet(priv, skb);
 		return ret;
 	}
 
-	if (mwifiex_queuing_ra_based(priv)) {
+	if (mwifiex_queuing_ra_based(priv) ||
+	    (ISSUPP_TDLS_ENABLED(priv->adapter->fw_cap_info) &&
+	     local_rx_pd->flags & MWIFIEX_RXPD_FLAGS_TDLS_PACKET)) {
 		memcpy(ta, rx_pkt_hdr->eth803_hdr.h_source, ETH_ALEN);
+		if (local_rx_pd->flags & MWIFIEX_RXPD_FLAGS_TDLS_PACKET &&
+		    local_rx_pd->priority < MAX_NUM_TID) {
+			sta_ptr = mwifiex_get_sta_entry(priv, ta);
+			if (sta_ptr)
+				sta_ptr->rx_seq[local_rx_pd->priority] =
+					      le16_to_cpu(local_rx_pd->seq_num);
+		}
 	} else {
 		if (rx_pkt_type != PKT_TYPE_BAR)
 			priv->rx_seq[local_rx_pd->priority] = seq_num;
diff --git a/drivers/net/wireless/mwifiex/sta_tx.c b/drivers/net/wireless/mwifiex/sta_tx.c
index 1236a5de7bca..5fce7e78a36e 100644
--- a/drivers/net/wireless/mwifiex/sta_tx.c
+++ b/drivers/net/wireless/mwifiex/sta_tx.c
@@ -128,6 +128,7 @@ int mwifiex_send_null_packet(struct mwifiex_private *priv, u8 flags)
 {
 	struct mwifiex_adapter *adapter = priv->adapter;
 	struct txpd *local_tx_pd;
+	struct mwifiex_tx_param tx_param;
 /* sizeof(struct txpd) + Interface specific header */
 #define NULL_PACKET_HDR 64
 	u32 data_len = NULL_PACKET_HDR;
@@ -168,8 +169,9 @@ int mwifiex_send_null_packet(struct mwifiex_private *priv, u8 flags)
 						   skb, NULL);
 	} else {
 		skb_push(skb, INTF_HEADER_LEN);
+		tx_param.next_pkt_len = 0;
 		ret = adapter->if_ops.host_to_card(adapter, MWIFIEX_TYPE_DATA,
-						   skb, NULL);
+						   skb, &tx_param);
 	}
 	switch (ret) {
 	case -EBUSY:
diff --git a/drivers/net/wireless/mwifiex/tdls.c b/drivers/net/wireless/mwifiex/tdls.c
index 97662a1ba58c..e73034fbbde9 100644
--- a/drivers/net/wireless/mwifiex/tdls.c
+++ b/drivers/net/wireless/mwifiex/tdls.c
@@ -25,8 +25,8 @@
 #define TDLS_RESP_FIX_LEN     8
 #define TDLS_CONFIRM_FIX_LEN  6
 
-static void
-mwifiex_restore_tdls_packets(struct mwifiex_private *priv, u8 *mac, u8 status)
+static void mwifiex_restore_tdls_packets(struct mwifiex_private *priv,
+					 const u8 *mac, u8 status)
 {
 	struct mwifiex_ra_list_tbl *ra_list;
 	struct list_head *tid_list;
@@ -84,7 +84,8 @@ mwifiex_restore_tdls_packets(struct mwifiex_private *priv, u8 *mac, u8 status)
 	return;
 }
 
-static void mwifiex_hold_tdls_packets(struct mwifiex_private *priv, u8 *mac)
+static void mwifiex_hold_tdls_packets(struct mwifiex_private *priv,
+				      const u8 *mac)
 {
 	struct mwifiex_ra_list_tbl *ra_list;
 	struct list_head *ra_list_head;
@@ -185,8 +186,50 @@ static int mwifiex_tdls_add_vht_capab(struct mwifiex_private *priv,
 	return 0;
 }
 
+static int
+mwifiex_tdls_add_ht_oper(struct mwifiex_private *priv, const u8 *mac,
+			 u8 vht_enabled, struct sk_buff *skb)
+{
+	struct ieee80211_ht_operation *ht_oper;
+	struct mwifiex_sta_node *sta_ptr;
+	struct mwifiex_bssdescriptor *bss_desc =
+					&priv->curr_bss_params.bss_descriptor;
+	u8 *pos;
+
+	sta_ptr = mwifiex_get_sta_entry(priv, mac);
+	if (unlikely(!sta_ptr)) {
+		dev_warn(priv->adapter->dev,
+			 "TDLS peer station not found in list\n");
+		return -1;
+	}
+
+	pos = (void *)skb_put(skb, sizeof(struct ieee80211_ht_operation) + 2);
+	*pos++ = WLAN_EID_HT_OPERATION;
+	*pos++ = sizeof(struct ieee80211_ht_operation);
+	ht_oper = (void *)pos;
+
+	ht_oper->primary_chan = bss_desc->channel;
+
+	/* follow AP's channel bandwidth */
+	if (ISSUPP_CHANWIDTH40(priv->adapter->hw_dot_11n_dev_cap) &&
+	    bss_desc->bcn_ht_cap &&
+	    ISALLOWED_CHANWIDTH40(bss_desc->bcn_ht_oper->ht_param))
+		ht_oper->ht_param = bss_desc->bcn_ht_oper->ht_param;
+
+	if (vht_enabled) {
+		ht_oper->ht_param =
+			  mwifiex_get_sec_chan_offset(bss_desc->channel);
+		ht_oper->ht_param |= BIT(2);
+	}
+
+	memcpy(&sta_ptr->tdls_cap.ht_oper, ht_oper,
+	       sizeof(struct ieee80211_ht_operation));
+
+	return 0;
+}
+
 static int mwifiex_tdls_add_vht_oper(struct mwifiex_private *priv,
-				     u8 *mac, struct sk_buff *skb)
+				     const u8 *mac, struct sk_buff *skb)
 {
 	struct mwifiex_bssdescriptor *bss_desc;
 	struct ieee80211_vht_operation *vht_oper;
@@ -325,8 +368,9 @@ static void mwifiex_tdls_add_qos_capab(struct sk_buff *skb)
 }
 
 static int mwifiex_prep_tdls_encap_data(struct mwifiex_private *priv,
-			     u8 *peer, u8 action_code, u8 dialog_token,
-			     u16 status_code, struct sk_buff *skb)
+					const u8 *peer, u8 action_code,
+					u8 dialog_token,
+					u16 status_code, struct sk_buff *skb)
 {
 	struct ieee80211_tdls_data *tf;
 	int ret;
@@ -428,6 +472,17 @@ static int mwifiex_prep_tdls_encap_data(struct mwifiex_private *priv,
 				dev_kfree_skb_any(skb);
 				return ret;
 			}
+			ret = mwifiex_tdls_add_ht_oper(priv, peer, 1, skb);
+			if (ret) {
+				dev_kfree_skb_any(skb);
+				return ret;
+			}
+		} else {
+			ret = mwifiex_tdls_add_ht_oper(priv, peer, 0, skb);
+			if (ret) {
+				dev_kfree_skb_any(skb);
+				return ret;
+			}
 		}
 		break;
 
@@ -453,7 +508,8 @@ static int mwifiex_prep_tdls_encap_data(struct mwifiex_private *priv,
 }
 
 static void
-mwifiex_tdls_add_link_ie(struct sk_buff *skb, u8 *src_addr, u8 *peer, u8 *bssid)
+mwifiex_tdls_add_link_ie(struct sk_buff *skb, const u8 *src_addr,
+			 const u8 *peer, const u8 *bssid)
 {
 	struct ieee80211_tdls_lnkie *lnkid;
 
@@ -467,8 +523,8 @@ mwifiex_tdls_add_link_ie(struct sk_buff *skb, u8 *src_addr, u8 *peer, u8 *bssid)
 	memcpy(lnkid->resp_sta, peer, ETH_ALEN);
 }
 
-int mwifiex_send_tdls_data_frame(struct mwifiex_private *priv,
-				 u8 *peer, u8 action_code, u8 dialog_token,
+int mwifiex_send_tdls_data_frame(struct mwifiex_private *priv, const u8 *peer,
+				 u8 action_code, u8 dialog_token,
 				 u16 status_code, const u8 *extra_ies,
 				 size_t extra_ies_len)
 {
@@ -560,7 +616,8 @@ int mwifiex_send_tdls_data_frame(struct mwifiex_private *priv,
 }
 
 static int
-mwifiex_construct_tdls_action_frame(struct mwifiex_private *priv, u8 *peer,
+mwifiex_construct_tdls_action_frame(struct mwifiex_private *priv,
+				    const u8 *peer,
 				    u8 action_code, u8 dialog_token,
 				    u16 status_code, struct sk_buff *skb)
 {
@@ -638,10 +695,10 @@ mwifiex_construct_tdls_action_frame(struct mwifiex_private *priv, u8 *peer,
 	return 0;
 }
 
-int mwifiex_send_tdls_action_frame(struct mwifiex_private *priv,
-				 u8 *peer, u8 action_code, u8 dialog_token,
-				 u16 status_code, const u8 *extra_ies,
-				 size_t extra_ies_len)
+int mwifiex_send_tdls_action_frame(struct mwifiex_private *priv, const u8 *peer,
+				   u8 action_code, u8 dialog_token,
+				   u16 status_code, const u8 *extra_ies,
+				   size_t extra_ies_len)
 {
 	struct sk_buff *skb;
 	struct mwifiex_txinfo *tx_info;
@@ -848,7 +905,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 }
 
 static int
-mwifiex_tdls_process_config_link(struct mwifiex_private *priv, u8 *peer)
+mwifiex_tdls_process_config_link(struct mwifiex_private *priv, const u8 *peer)
 {
 	struct mwifiex_sta_node *sta_ptr;
 	struct mwifiex_ds_tdls_oper tdls_oper;
@@ -869,7 +926,7 @@ mwifiex_tdls_process_config_link(struct mwifiex_private *priv, u8 *peer)
 }
 
 static int
-mwifiex_tdls_process_create_link(struct mwifiex_private *priv, u8 *peer)
+mwifiex_tdls_process_create_link(struct mwifiex_private *priv, const u8 *peer)
 {
 	struct mwifiex_sta_node *sta_ptr;
 	struct mwifiex_ds_tdls_oper tdls_oper;
@@ -896,7 +953,7 @@ mwifiex_tdls_process_create_link(struct mwifiex_private *priv, u8 *peer)
 }
 
 static int
-mwifiex_tdls_process_disable_link(struct mwifiex_private *priv, u8 *peer)
+mwifiex_tdls_process_disable_link(struct mwifiex_private *priv, const u8 *peer)
 {
 	struct mwifiex_sta_node *sta_ptr;
 	struct mwifiex_ds_tdls_oper tdls_oper;
@@ -925,7 +982,7 @@ mwifiex_tdls_process_disable_link(struct mwifiex_private *priv, u8 *peer)
 }
 
 static int
-mwifiex_tdls_process_enable_link(struct mwifiex_private *priv, u8 *peer)
+mwifiex_tdls_process_enable_link(struct mwifiex_private *priv, const u8 *peer)
 {
 	struct mwifiex_sta_node *sta_ptr;
 	struct ieee80211_mcs_info mcs;
@@ -982,7 +1039,7 @@ mwifiex_tdls_process_enable_link(struct mwifiex_private *priv, u8 *peer)
 	return 0;
 }
 
-int mwifiex_tdls_oper(struct mwifiex_private *priv, u8 *peer, u8 action)
+int mwifiex_tdls_oper(struct mwifiex_private *priv, const u8 *peer, u8 action)
 {
 	switch (action) {
 	case MWIFIEX_TDLS_ENABLE_LINK:
@@ -997,7 +1054,7 @@ int mwifiex_tdls_oper(struct mwifiex_private *priv, u8 *peer, u8 action)
 	return 0;
 }
 
-int mwifiex_get_tdls_link_status(struct mwifiex_private *priv, u8 *mac)
+int mwifiex_get_tdls_link_status(struct mwifiex_private *priv, const u8 *mac)
 {
 	struct mwifiex_sta_node *sta_ptr;
 
diff --git a/drivers/net/wireless/mwifiex/uap_cmd.c b/drivers/net/wireless/mwifiex/uap_cmd.c
index 9be6544bdded..32643555dd2a 100644
--- a/drivers/net/wireless/mwifiex/uap_cmd.c
+++ b/drivers/net/wireless/mwifiex/uap_cmd.c
@@ -175,17 +175,19 @@ mwifiex_set_ht_params(struct mwifiex_private *priv,
 		switch (GET_RXSTBC(cap_info)) {
 		case MWIFIEX_RX_STBC1:
 			/* HT_CAP 1X1 mode */
-			memset(&bss_cfg->ht_cap.mcs, 0xff, 1);
+			bss_cfg->ht_cap.mcs.rx_mask[0] = 0xff;
 			break;
 		case MWIFIEX_RX_STBC12:	/* fall through */
 		case MWIFIEX_RX_STBC123:
 			/* HT_CAP 2X2 mode */
-			memset(&bss_cfg->ht_cap.mcs, 0xff, 2);
+			bss_cfg->ht_cap.mcs.rx_mask[0] = 0xff;
+			bss_cfg->ht_cap.mcs.rx_mask[1] = 0xff;
 			break;
 		default:
 			dev_warn(priv->adapter->dev,
 				 "Unsupported RX-STBC, default to 2x2\n");
-			memset(&bss_cfg->ht_cap.mcs, 0xff, 2);
+			bss_cfg->ht_cap.mcs.rx_mask[0] = 0xff;
+			bss_cfg->ht_cap.mcs.rx_mask[1] = 0xff;
 			break;
 		}
 		priv->ap_11n_enabled = 1;
diff --git a/drivers/net/wireless/mwifiex/usb.c b/drivers/net/wireless/mwifiex/usb.c
index edbe4aff00d8..a8ce8130cfae 100644
--- a/drivers/net/wireless/mwifiex/usb.c
+++ b/drivers/net/wireless/mwifiex/usb.c
@@ -22,9 +22,9 @@
 
 #define USB_VERSION	"1.0"
 
+static u8 user_rmmod;
 static struct mwifiex_if_ops usb_ops;
 static struct semaphore add_remove_card_sem;
-static struct usb_card_rec *usb_card;
 
 static struct usb_device_id mwifiex_usb_table[] = {
 	/* 8797 */
@@ -532,28 +532,38 @@ static int mwifiex_usb_resume(struct usb_interface *intf)
 static void mwifiex_usb_disconnect(struct usb_interface *intf)
 {
 	struct usb_card_rec *card = usb_get_intfdata(intf);
+	struct mwifiex_adapter *adapter;
 
-	if (!card) {
-		pr_err("%s: card is NULL\n", __func__);
+	if (!card || !card->adapter) {
+		pr_err("%s: card or card->adapter is NULL\n", __func__);
 		return;
 	}
 
-	mwifiex_usb_free(card);
+	adapter = card->adapter;
+	if (!adapter->priv_num)
+		return;
 
-	if (card->adapter) {
-		struct mwifiex_adapter *adapter = card->adapter;
+	if (user_rmmod) {
+#ifdef CONFIG_PM
+		if (adapter->is_suspended)
+			mwifiex_usb_resume(intf);
+#endif
 
-		if (!adapter->priv_num)
-			return;
+		mwifiex_deauthenticate_all(adapter);
 
-		dev_dbg(adapter->dev, "%s: removing card\n", __func__);
-		mwifiex_remove_card(adapter, &add_remove_card_sem);
+		mwifiex_init_shutdown_fw(mwifiex_get_priv(adapter,
+							  MWIFIEX_BSS_ROLE_ANY),
+					 MWIFIEX_FUNC_SHUTDOWN);
 	}
 
+	mwifiex_usb_free(card);
+
+	dev_dbg(adapter->dev, "%s: removing card\n", __func__);
+	mwifiex_remove_card(adapter, &add_remove_card_sem);
+
 	usb_set_intfdata(intf, NULL);
 	usb_put_dev(interface_to_usbdev(intf));
 	kfree(card);
-	usb_card = NULL;
 
 	return;
 }
@@ -565,6 +575,7 @@ static struct usb_driver mwifiex_usb_driver = {
 	.id_table = mwifiex_usb_table,
 	.suspend = mwifiex_usb_suspend,
 	.resume = mwifiex_usb_resume,
+	.soft_unbind = 1,
 };
 
 static int mwifiex_usb_tx_init(struct mwifiex_adapter *adapter)
@@ -762,7 +773,6 @@ static int mwifiex_register_dev(struct mwifiex_adapter *adapter)
 
 	card->adapter = adapter;
 	adapter->dev = &card->udev->dev;
-	usb_card = card;
 
 	switch (le16_to_cpu(card->udev->descriptor.idProduct)) {
 	case USB8897_PID_1:
@@ -1025,25 +1035,8 @@ static void mwifiex_usb_cleanup_module(void)
 	if (!down_interruptible(&add_remove_card_sem))
 		up(&add_remove_card_sem);
 
-	if (usb_card && usb_card->adapter) {
-		struct mwifiex_adapter *adapter = usb_card->adapter;
-
-		/* In case driver is removed when asynchronous FW downloading is
-		 * in progress
-		 */
-		wait_for_completion(&adapter->fw_load);
-
-#ifdef CONFIG_PM
-		if (adapter->is_suspended)
-			mwifiex_usb_resume(usb_card->intf);
-#endif
-
-		mwifiex_deauthenticate_all(adapter);
-
-		mwifiex_init_shutdown_fw(mwifiex_get_priv(adapter,
-							  MWIFIEX_BSS_ROLE_ANY),
-					 MWIFIEX_FUNC_SHUTDOWN);
-	}
+	/* set the flag as user is removing this module */
+	user_rmmod = 1;
 
 	usb_deregister(&mwifiex_usb_driver);
 }
diff --git a/drivers/net/wireless/mwifiex/util.c b/drivers/net/wireless/mwifiex/util.c
index c3824e37f3f2..6da5abf52e61 100644
--- a/drivers/net/wireless/mwifiex/util.c
+++ b/drivers/net/wireless/mwifiex/util.c
@@ -259,7 +259,7 @@ int mwifiex_complete_cmd(struct mwifiex_adapter *adapter,
  * NULL is returned if station entry is not found in associated STA list.
  */
 struct mwifiex_sta_node *
-mwifiex_get_sta_entry(struct mwifiex_private *priv, u8 *mac)
+mwifiex_get_sta_entry(struct mwifiex_private *priv, const u8 *mac)
 {
 	struct mwifiex_sta_node *node;
 
@@ -280,7 +280,7 @@ mwifiex_get_sta_entry(struct mwifiex_private *priv, u8 *mac)
  * If received mac address is NULL, NULL is returned.
  */
 struct mwifiex_sta_node *
-mwifiex_add_sta_entry(struct mwifiex_private *priv, u8 *mac)
+mwifiex_add_sta_entry(struct mwifiex_private *priv, const u8 *mac)
 {
 	struct mwifiex_sta_node *node;
 	unsigned long flags;
@@ -332,7 +332,7 @@ mwifiex_set_sta_ht_cap(struct mwifiex_private *priv, const u8 *ies,
 }
 
 /* This function will delete a station entry from station list */
-void mwifiex_del_sta_entry(struct mwifiex_private *priv, u8 *mac)
+void mwifiex_del_sta_entry(struct mwifiex_private *priv, const u8 *mac)
 {
 	struct mwifiex_sta_node *node;
 	unsigned long flags;
diff --git a/drivers/net/wireless/mwifiex/util.h b/drivers/net/wireless/mwifiex/util.h
index ddae57021397..caadb3737b9e 100644
--- a/drivers/net/wireless/mwifiex/util.h
+++ b/drivers/net/wireless/mwifiex/util.h
@@ -20,32 +20,55 @@
 #ifndef _MWIFIEX_UTIL_H_
 #define _MWIFIEX_UTIL_H_
 
+struct mwifiex_dma_mapping {
+	dma_addr_t addr;
+	size_t len;
+};
+
+struct mwifiex_cb {
+	struct mwifiex_dma_mapping dma_mapping;
+	union {
+		struct mwifiex_rxinfo rx_info;
+		struct mwifiex_txinfo tx_info;
+	};
+};
+
 static inline struct mwifiex_rxinfo *MWIFIEX_SKB_RXCB(struct sk_buff *skb)
 {
-	return (struct mwifiex_rxinfo *)(skb->cb + sizeof(dma_addr_t));
+	struct mwifiex_cb *cb = (struct mwifiex_cb *)skb->cb;
+
+	BUILD_BUG_ON(sizeof(struct mwifiex_cb) > sizeof(skb->cb));
+	return &cb->rx_info;
 }
 
 static inline struct mwifiex_txinfo *MWIFIEX_SKB_TXCB(struct sk_buff *skb)
 {
-	return (struct mwifiex_txinfo *)(skb->cb + sizeof(dma_addr_t));
+	struct mwifiex_cb *cb = (struct mwifiex_cb *)skb->cb;
+
+	return &cb->tx_info;
 }
 
-struct mwifiex_dma_mapping {
-	dma_addr_t addr;
-	size_t len;
-};
+static inline void mwifiex_store_mapping(struct sk_buff *skb,
+					 struct mwifiex_dma_mapping *mapping)
+{
+	struct mwifiex_cb *cb = (struct mwifiex_cb *)skb->cb;
+
+	memcpy(&cb->dma_mapping, mapping, sizeof(*mapping));
+}
 
-static inline void MWIFIEX_SKB_PACB(struct sk_buff *skb,
-					struct mwifiex_dma_mapping *mapping)
+static inline void mwifiex_get_mapping(struct sk_buff *skb,
+				       struct mwifiex_dma_mapping *mapping)
 {
-	memcpy(mapping, skb->cb, sizeof(*mapping));
+	struct mwifiex_cb *cb = (struct mwifiex_cb *)skb->cb;
+
+	memcpy(mapping, &cb->dma_mapping, sizeof(*mapping));
 }
 
 static inline dma_addr_t MWIFIEX_SKB_DMA_ADDR(struct sk_buff *skb)
 {
 	struct mwifiex_dma_mapping mapping;
 
-	MWIFIEX_SKB_PACB(skb, &mapping);
+	mwifiex_get_mapping(skb, &mapping);
 
 	return mapping.addr;
 }
diff --git a/drivers/net/wireless/mwifiex/wmm.c b/drivers/net/wireless/mwifiex/wmm.c
index 0a7cc742aed7..d3671d009f6c 100644
--- a/drivers/net/wireless/mwifiex/wmm.c
+++ b/drivers/net/wireless/mwifiex/wmm.c
@@ -92,7 +92,7 @@ mwifiex_wmm_ac_debug_print(const struct ieee_types_wmm_ac_parameters *ac_param)
  * The function also initializes the list with the provided RA.
  */
 static struct mwifiex_ra_list_tbl *
-mwifiex_wmm_allocate_ralist_node(struct mwifiex_adapter *adapter, u8 *ra)
+mwifiex_wmm_allocate_ralist_node(struct mwifiex_adapter *adapter, const u8 *ra)
 {
 	struct mwifiex_ra_list_tbl *ra_list;
 
@@ -139,8 +139,7 @@ static u8 mwifiex_get_random_ba_threshold(void)
  * This function allocates and adds a RA list for all TIDs
  * with the given RA.
  */
-void
-mwifiex_ralist_add(struct mwifiex_private *priv, u8 *ra)
+void mwifiex_ralist_add(struct mwifiex_private *priv, const u8 *ra)
 {
 	int i;
 	struct mwifiex_ra_list_tbl *ra_list;
@@ -164,6 +163,7 @@ mwifiex_ralist_add(struct mwifiex_private *priv, u8 *ra)
 		if (!mwifiex_queuing_ra_based(priv)) {
 			if (mwifiex_get_tdls_link_status(priv, ra) ==
 			    TDLS_SETUP_COMPLETE) {
+				ra_list->tdls_link = true;
 				ra_list->is_11n_enabled =
 					mwifiex_tdls_peer_11n_enabled(priv, ra);
 			} else {
@@ -426,15 +426,6 @@ mwifiex_wmm_init(struct mwifiex_adapter *adapter)
 							priv->tos_to_tid_inv[i];
 		}
 
-		priv->aggr_prio_tbl[6].amsdu
-					= priv->aggr_prio_tbl[6].ampdu_ap
-					= priv->aggr_prio_tbl[6].ampdu_user
-					= BA_STREAM_NOT_ALLOWED;
-
-		priv->aggr_prio_tbl[7].amsdu = priv->aggr_prio_tbl[7].ampdu_ap
-					= priv->aggr_prio_tbl[7].ampdu_user
-					= BA_STREAM_NOT_ALLOWED;
-
 		mwifiex_set_ba_params(priv);
 		mwifiex_reset_11n_rx_seq_num(priv);
 
@@ -575,7 +566,7 @@ mwifiex_clean_txrx(struct mwifiex_private *priv)
  */
 static struct mwifiex_ra_list_tbl *
 mwifiex_wmm_get_ralist_node(struct mwifiex_private *priv, u8 tid,
-			    u8 *ra_addr)
+			    const u8 *ra_addr)
 {
 	struct mwifiex_ra_list_tbl *ra_list;
 
@@ -596,7 +587,8 @@ mwifiex_wmm_get_ralist_node(struct mwifiex_private *priv, u8 tid,
  * retrieved.
  */
 struct mwifiex_ra_list_tbl *
-mwifiex_wmm_get_queue_raptr(struct mwifiex_private *priv, u8 tid, u8 *ra_addr)
+mwifiex_wmm_get_queue_raptr(struct mwifiex_private *priv, u8 tid,
+			    const u8 *ra_addr)
 {
 	struct mwifiex_ra_list_tbl *ra_list;
 
@@ -657,7 +649,7 @@ mwifiex_wmm_add_buf_txqueue(struct mwifiex_private *priv,
 		if (ntohs(eth_hdr->h_proto) == ETH_P_TDLS)
 			dev_dbg(adapter->dev,
 				"TDLS setup packet for %pM. Don't block\n", ra);
-		else
+		else if (memcmp(priv->cfg_bssid, ra, ETH_ALEN))
 			tdls_status = mwifiex_get_tdls_link_status(priv, ra);
 	}
 
diff --git a/drivers/net/wireless/mwifiex/wmm.h b/drivers/net/wireless/mwifiex/wmm.h
index 83e42083ebff..eca56e371a57 100644
--- a/drivers/net/wireless/mwifiex/wmm.h
+++ b/drivers/net/wireless/mwifiex/wmm.h
@@ -99,7 +99,7 @@ mwifiex_wmm_is_ra_list_empty(struct list_head *ra_list_hhead)
 
 void mwifiex_wmm_add_buf_txqueue(struct mwifiex_private *priv,
 				 struct sk_buff *skb);
-void mwifiex_ralist_add(struct mwifiex_private *priv, u8 *ra);
+void mwifiex_ralist_add(struct mwifiex_private *priv, const u8 *ra);
 void mwifiex_rotate_priolists(struct mwifiex_private *priv,
 			      struct mwifiex_ra_list_tbl *ra, int tid);
 
@@ -123,7 +123,8 @@ void mwifiex_wmm_setup_ac_downgrade(struct mwifiex_private *priv);
 int mwifiex_ret_wmm_get_status(struct mwifiex_private *priv,
 			       const struct host_cmd_ds_command *resp);
 struct mwifiex_ra_list_tbl *
-mwifiex_wmm_get_queue_raptr(struct mwifiex_private *priv, u8 tid, u8 *ra_addr);
+mwifiex_wmm_get_queue_raptr(struct mwifiex_private *priv, u8 tid,
+			    const u8 *ra_addr);
 u8 mwifiex_wmm_downgrade_tid(struct mwifiex_private *priv, u32 tid);
 
 #endif /* !_MWIFIEX_WMM_H_ */
diff --git a/drivers/net/wireless/orinoco/hw.c b/drivers/net/wireless/orinoco/hw.c
index 49300d04efdf..e27e32851f1e 100644
--- a/drivers/net/wireless/orinoco/hw.c
+++ b/drivers/net/wireless/orinoco/hw.c
@@ -988,8 +988,8 @@ int __orinoco_hw_setup_enc(struct orinoco_private *priv)
  * tsc must be NULL or up to 8 bytes
  */
 int __orinoco_hw_set_tkip_key(struct orinoco_private *priv, int key_idx,
-			      int set_tx, u8 *key, u8 *rsc, size_t rsc_len,
-			      u8 *tsc, size_t tsc_len)
+			      int set_tx, const u8 *key, const u8 *rsc,
+			      size_t rsc_len, const u8 *tsc, size_t tsc_len)
 {
 	struct {
 		__le16 idx;
diff --git a/drivers/net/wireless/orinoco/hw.h b/drivers/net/wireless/orinoco/hw.h
index 8f6831f4e328..466d1ede76f1 100644
--- a/drivers/net/wireless/orinoco/hw.h
+++ b/drivers/net/wireless/orinoco/hw.h
@@ -38,8 +38,8 @@ int __orinoco_hw_set_wap(struct orinoco_private *priv);
 int __orinoco_hw_setup_wepkeys(struct orinoco_private *priv);
 int __orinoco_hw_setup_enc(struct orinoco_private *priv);
 int __orinoco_hw_set_tkip_key(struct orinoco_private *priv, int key_idx,
-			      int set_tx, u8 *key, u8 *rsc, size_t rsc_len,
-			      u8 *tsc, size_t tsc_len);
+			      int set_tx, const u8 *key, const u8 *rsc,
+			      size_t rsc_len, const u8 *tsc, size_t tsc_len);
 int orinoco_clear_tkip_key(struct orinoco_private *priv, int key_idx);
 int __orinoco_hw_set_multicast_list(struct orinoco_private *priv,
 				    struct net_device *dev,
diff --git a/drivers/net/wireless/orinoco/orinoco_usb.c b/drivers/net/wireless/orinoco/orinoco_usb.c
index 3ac71339d040..c90939ced0e4 100644
--- a/drivers/net/wireless/orinoco/orinoco_usb.c
+++ b/drivers/net/wireless/orinoco/orinoco_usb.c
@@ -1673,7 +1673,7 @@ static int ezusb_probe(struct usb_interface *interface,
 		firmware.code = fw_entry->data;
 	}
 	if (firmware.size && firmware.code) {
-		if (ezusb_firmware_download(upriv, &firmware))
+		if (ezusb_firmware_download(upriv, &firmware) < 0)
 			goto error;
 	} else {
 		err("No firmware to download");
diff --git a/drivers/net/wireless/orinoco/wext.c b/drivers/net/wireless/orinoco/wext.c
index b7a867b50b94..6abdaf0aa052 100644
--- a/drivers/net/wireless/orinoco/wext.c
+++ b/drivers/net/wireless/orinoco/wext.c
@@ -52,9 +52,9 @@ static int orinoco_set_key(struct orinoco_private *priv, int index,
 	priv->keys[index].seq_len = seq_len;
 
 	if (key_len)
-		memcpy(priv->keys[index].key, key, key_len);
+		memcpy((void *)priv->keys[index].key, key, key_len);
 	if (seq_len)
-		memcpy(priv->keys[index].seq, seq, seq_len);
+		memcpy((void *)priv->keys[index].seq, seq, seq_len);
 
 	switch (alg) {
 	case ORINOCO_ALG_TKIP:
diff --git a/drivers/net/wireless/p54/main.c b/drivers/net/wireless/p54/main.c
index eede90b63f84..7be3a4839640 100644
--- a/drivers/net/wireless/p54/main.c
+++ b/drivers/net/wireless/p54/main.c
@@ -669,7 +669,8 @@ static unsigned int p54_flush_count(struct p54_common *priv)
 	return total;
 }
 
-static void p54_flush(struct ieee80211_hw *dev, u32 queues, bool drop)
+static void p54_flush(struct ieee80211_hw *dev, struct ieee80211_vif *vif,
+		      u32 queues, bool drop)
 {
 	struct p54_common *priv = dev->priv;
 	unsigned int total, i;
diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c
index cbf0a589d32a..8330fa33e50b 100644
--- a/drivers/net/wireless/ray_cs.c
+++ b/drivers/net/wireless/ray_cs.c
@@ -343,7 +343,7 @@ static void ray_detach(struct pcmcia_device *link)
 	ray_release(link);
 
 	local = netdev_priv(dev);
-	del_timer(&local->timer);
+	del_timer_sync(&local->timer);
 
 	if (link->priv) {
 		unregister_netdev(dev);
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index 39d22a154341..d2a9a08210be 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -517,7 +517,7 @@ static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev,
 				 u8 key_index, bool unicast, bool multicast);
 
 static int rndis_get_station(struct wiphy *wiphy, struct net_device *dev,
-					u8 *mac, struct station_info *sinfo);
+			     const u8 *mac, struct station_info *sinfo);
 
 static int rndis_dump_station(struct wiphy *wiphy, struct net_device *dev,
 			       int idx, u8 *mac, struct station_info *sinfo);
@@ -2490,7 +2490,7 @@ static void rndis_fill_station_info(struct usbnet *usbdev,
 }
 
 static int rndis_get_station(struct wiphy *wiphy, struct net_device *dev,
-					u8 *mac, struct station_info *sinfo)
+			     const u8 *mac, struct station_info *sinfo)
 {
 	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
 	struct usbnet *usbdev = priv->usbdev;
diff --git a/drivers/net/wireless/rsi/rsi_91x_mac80211.c b/drivers/net/wireless/rsi/rsi_91x_mac80211.c
index 84164747ace0..54aaeb09debf 100644
--- a/drivers/net/wireless/rsi/rsi_91x_mac80211.c
+++ b/drivers/net/wireless/rsi/rsi_91x_mac80211.c
@@ -656,6 +656,7 @@ static int rsi_mac80211_ampdu_action(struct ieee80211_hw *hw,
 	case IEEE80211_AMPDU_TX_START:
 		common->vif_info[ii].seq_start = seq_no;
 		ieee80211_start_tx_ba_cb_irqsafe(vif, sta->addr, tid);
+		status = 0;
 		break;
 
 	case IEEE80211_AMPDU_TX_STOP_CONT:
diff --git a/drivers/net/wireless/rsi/rsi_91x_mgmt.c b/drivers/net/wireless/rsi/rsi_91x_mgmt.c
index 1b28cda6ca88..2eefbf159bc0 100644
--- a/drivers/net/wireless/rsi/rsi_91x_mgmt.c
+++ b/drivers/net/wireless/rsi/rsi_91x_mgmt.c
@@ -1083,7 +1083,7 @@ void rsi_inform_bss_status(struct rsi_common *common,
 {
 	if (status) {
 		rsi_hal_send_sta_notify_frame(common,
-					      NL80211_IFTYPE_STATION,
+					      RSI_IFTYPE_STATION,
 					      STA_CONNECTED,
 					      bssid,
 					      qos_enable,
@@ -1092,7 +1092,7 @@ void rsi_inform_bss_status(struct rsi_common *common,
 			rsi_send_auto_rate_request(common);
 	} else {
 		rsi_hal_send_sta_notify_frame(common,
-					      NL80211_IFTYPE_STATION,
+					      RSI_IFTYPE_STATION,
 					      STA_DISCONNECTED,
 					      bssid,
 					      qos_enable,
diff --git a/drivers/net/wireless/rsi/rsi_91x_sdio.c b/drivers/net/wireless/rsi/rsi_91x_sdio.c
index 2e39d38d6a9e..46e7af446f01 100644
--- a/drivers/net/wireless/rsi/rsi_91x_sdio.c
+++ b/drivers/net/wireless/rsi/rsi_91x_sdio.c
@@ -285,7 +285,6 @@ static void rsi_reset_card(struct sdio_func *pfunction)
 		if (err) {
 			rsi_dbg(ERR_ZONE, "%s: CCCR speed reg read failed: %d\n",
 				__func__, err);
-			card->state &= ~MMC_STATE_HIGHSPEED;
 		} else {
 			err = rsi_cmd52writebyte(card,
 						 SDIO_CCCR_SPEED,
@@ -296,14 +295,13 @@ static void rsi_reset_card(struct sdio_func *pfunction)
 					__func__, err);
 				return;
 			}
-			mmc_card_set_highspeed(card);
 			host->ios.timing = MMC_TIMING_SD_HS;
 			host->ops->set_ios(host, &host->ios);
 		}
 	}
 
 	/* Set clock */
-	if (mmc_card_highspeed(card))
+	if (mmc_card_hs(card))
 		clock = 50000000;
 	else
 		clock = card->cis.max_dtr;
diff --git a/drivers/net/wireless/rsi/rsi_common.h b/drivers/net/wireless/rsi/rsi_common.h
index f2f70784d4ad..d3fbe33d2324 100644
--- a/drivers/net/wireless/rsi/rsi_common.h
+++ b/drivers/net/wireless/rsi/rsi_common.h
@@ -63,7 +63,7 @@ static inline int rsi_create_kthread(struct rsi_common *common,
 				     u8 *name)
 {
 	init_completion(&thread->completion);
-	thread->task = kthread_run(func_ptr, common, name);
+	thread->task = kthread_run(func_ptr, common, "%s", name);
 	if (IS_ERR(thread->task))
 		return (int)PTR_ERR(thread->task);
 
diff --git a/drivers/net/wireless/rsi/rsi_mgmt.h b/drivers/net/wireless/rsi/rsi_mgmt.h
index ac67c4ad63c2..225215a3b8bb 100644
--- a/drivers/net/wireless/rsi/rsi_mgmt.h
+++ b/drivers/net/wireless/rsi/rsi_mgmt.h
@@ -73,6 +73,7 @@
 #define RX_BA_INDICATION                1
 #define RSI_TBL_SZ                      40
 #define MAX_RETRIES                     8
+#define RSI_IFTYPE_STATION		 0
 
 #define STD_RATE_MCS7                   0x07
 #define STD_RATE_MCS6                   0x06
diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c
index 2f1cd929c6f6..a511cccc9f01 100644
--- a/drivers/net/wireless/rt2x00/rt2500pci.c
+++ b/drivers/net/wireless/rt2x00/rt2500pci.c
@@ -1681,8 +1681,13 @@ static int rt2500pci_init_eeprom(struct rt2x00_dev *rt2x00dev)
 	/*
 	 * Detect if this device has an hardware controlled radio.
 	 */
-	if (rt2x00_get_field16(eeprom, EEPROM_ANTENNA_HARDWARE_RADIO))
+	if (rt2x00_get_field16(eeprom, EEPROM_ANTENNA_HARDWARE_RADIO)) {
 		__set_bit(CAPABILITY_HW_BUTTON, &rt2x00dev->cap_flags);
+		/*
+		 * On this device RFKILL initialized during probe does not work.
+		 */
+		__set_bit(REQUIRE_DELAYED_RFKILL, &rt2x00dev->cap_flags);
+	}
 
 	/*
 	 * Check if the BBP tuning should be enabled.
diff --git a/drivers/net/wireless/rt2x00/rt2800lib.c b/drivers/net/wireless/rt2x00/rt2800lib.c
index 41d4a8167dc3..c17fcf272728 100644
--- a/drivers/net/wireless/rt2x00/rt2800lib.c
+++ b/drivers/net/wireless/rt2x00/rt2800lib.c
@@ -1005,10 +1005,9 @@ void rt2800_write_beacon(struct queue_entry *entry, struct txentry_desc *txdesc)
 				   entry->skb->len + padding_len);
 
 	/*
-	 * Enable beaconing again.
+	 * Restore beaconing state.
 	 */
-	rt2x00_set_field32(&reg, BCN_TIME_CFG_BEACON_GEN, 1);
-	rt2800_register_write(rt2x00dev, BCN_TIME_CFG, reg);
+	rt2800_register_write(rt2x00dev, BCN_TIME_CFG, orig_reg);
 
 	/*
 	 * Clean up beacon skb.
@@ -1039,13 +1038,14 @@ static inline void rt2800_clear_beacon_register(struct rt2x00_dev *rt2x00dev,
 void rt2800_clear_beacon(struct queue_entry *entry)
 {
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
-	u32 reg;
+	u32 orig_reg, reg;
 
 	/*
 	 * Disable beaconing while we are reloading the beacon data,
 	 * otherwise we might be sending out invalid data.
 	 */
-	rt2800_register_read(rt2x00dev, BCN_TIME_CFG, &reg);
+	rt2800_register_read(rt2x00dev, BCN_TIME_CFG, &orig_reg);
+	reg = orig_reg;
 	rt2x00_set_field32(&reg, BCN_TIME_CFG_BEACON_GEN, 0);
 	rt2800_register_write(rt2x00dev, BCN_TIME_CFG, reg);
 
@@ -1055,10 +1055,9 @@ void rt2800_clear_beacon(struct queue_entry *entry)
 	rt2800_clear_beacon_register(rt2x00dev, entry->entry_idx);
 
 	/*
-	 * Enabled beaconing again.
+	 * Restore beaconing state.
 	 */
-	rt2x00_set_field32(&reg, BCN_TIME_CFG_BEACON_GEN, 1);
-	rt2800_register_write(rt2x00dev, BCN_TIME_CFG, reg);
+	rt2800_register_write(rt2x00dev, BCN_TIME_CFG, orig_reg);
 }
 EXPORT_SYMBOL_GPL(rt2800_clear_beacon);
 
diff --git a/drivers/net/wireless/rt2x00/rt2800usb.c b/drivers/net/wireless/rt2x00/rt2800usb.c
index a49c3d73ea2c..e11dab2216c6 100644
--- a/drivers/net/wireless/rt2x00/rt2800usb.c
+++ b/drivers/net/wireless/rt2x00/rt2800usb.c
@@ -229,6 +229,27 @@ static enum hrtimer_restart rt2800usb_tx_sta_fifo_timeout(struct hrtimer *timer)
 /*
  * Firmware functions
  */
+static int rt2800usb_autorun_detect(struct rt2x00_dev *rt2x00dev)
+{
+	__le32 reg;
+	u32 fw_mode;
+
+	/* cannot use rt2x00usb_register_read here as it uses different
+	 * mode (MULTI_READ vs. DEVICE_MODE) and does not pass the
+	 * magic value USB_MODE_AUTORUN (0x11) to the device, thus the
+	 * returned value would be invalid.
+	 */
+	rt2x00usb_vendor_request(rt2x00dev, USB_DEVICE_MODE,
+				 USB_VENDOR_REQUEST_IN, 0, USB_MODE_AUTORUN,
+				 &reg, sizeof(reg), REGISTER_TIMEOUT_FIRMWARE);
+	fw_mode = le32_to_cpu(reg);
+
+	if ((fw_mode & 0x00000003) == 2)
+		return 1;
+
+	return 0;
+}
+
 static char *rt2800usb_get_firmware_name(struct rt2x00_dev *rt2x00dev)
 {
 	return FIRMWARE_RT2870;
@@ -257,8 +278,13 @@ static int rt2800usb_write_firmware(struct rt2x00_dev *rt2x00dev,
 	/*
 	 * Write firmware to device.
 	 */
-	rt2x00usb_register_multiwrite(rt2x00dev, FIRMWARE_IMAGE_BASE,
-				      data + offset, length);
+	if (rt2800usb_autorun_detect(rt2x00dev)) {
+		rt2x00_info(rt2x00dev,
+			    "Firmware loading not required - NIC in AutoRun mode\n");
+	} else {
+		rt2x00usb_register_multiwrite(rt2x00dev, FIRMWARE_IMAGE_BASE,
+					      data + offset, length);
+	}
 
 	rt2x00usb_register_write(rt2x00dev, H2M_MAILBOX_CID, ~0);
 	rt2x00usb_register_write(rt2x00dev, H2M_MAILBOX_STATUS, ~0);
@@ -735,11 +761,18 @@ static void rt2800usb_fill_rxdone(struct queue_entry *entry,
 /*
  * Device probe functions.
  */
+static int rt2800usb_efuse_detect(struct rt2x00_dev *rt2x00dev)
+{
+	if (rt2800usb_autorun_detect(rt2x00dev))
+		return 1;
+	return rt2800_efuse_detect(rt2x00dev);
+}
+
 static int rt2800usb_read_eeprom(struct rt2x00_dev *rt2x00dev)
 {
 	int retval;
 
-	if (rt2800_efuse_detect(rt2x00dev))
+	if (rt2800usb_efuse_detect(rt2x00dev))
 		retval = rt2800_read_eeprom_efuse(rt2x00dev);
 	else
 		retval = rt2x00usb_eeprom_read(rt2x00dev, rt2x00dev->eeprom,
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index e3b885d8f7db..d13f25cd70d5 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -693,6 +693,7 @@ enum rt2x00_capability_flags {
 	REQUIRE_SW_SEQNO,
 	REQUIRE_HT_TX_DESC,
 	REQUIRE_PS_AUTOWAKE,
+	REQUIRE_DELAYED_RFKILL,
 
 	/*
 	 * Capabilities
@@ -1448,7 +1449,8 @@ int rt2x00mac_conf_tx(struct ieee80211_hw *hw,
 		      struct ieee80211_vif *vif, u16 queue,
 		      const struct ieee80211_tx_queue_params *params);
 void rt2x00mac_rfkill_poll(struct ieee80211_hw *hw);
-void rt2x00mac_flush(struct ieee80211_hw *hw, u32 queues, bool drop);
+void rt2x00mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		     u32 queues, bool drop);
 int rt2x00mac_set_antenna(struct ieee80211_hw *hw, u32 tx_ant, u32 rx_ant);
 int rt2x00mac_get_antenna(struct ieee80211_hw *hw, u32 *tx_ant, u32 *rx_ant);
 void rt2x00mac_get_ringparam(struct ieee80211_hw *hw,
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index 2bde6729f5e6..4fa43a2eeb73 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -1126,9 +1126,10 @@ static void rt2x00lib_uninitialize(struct rt2x00_dev *rt2x00dev)
 		return;
 
 	/*
-	 * Unregister extra components.
+	 * Stop rfkill polling.
 	 */
-	rt2x00rfkill_unregister(rt2x00dev);
+	if (test_bit(REQUIRE_DELAYED_RFKILL, &rt2x00dev->cap_flags))
+		rt2x00rfkill_unregister(rt2x00dev);
 
 	/*
 	 * Allow the HW to uninitialize.
@@ -1166,6 +1167,12 @@ static int rt2x00lib_initialize(struct rt2x00_dev *rt2x00dev)
 
 	set_bit(DEVICE_STATE_INITIALIZED, &rt2x00dev->flags);
 
+	/*
+	 * Start rfkill polling.
+	 */
+	if (test_bit(REQUIRE_DELAYED_RFKILL, &rt2x00dev->cap_flags))
+		rt2x00rfkill_register(rt2x00dev);
+
 	return 0;
 }
 
@@ -1375,7 +1382,12 @@ int rt2x00lib_probe_dev(struct rt2x00_dev *rt2x00dev)
 	rt2x00link_register(rt2x00dev);
 	rt2x00leds_register(rt2x00dev);
 	rt2x00debug_register(rt2x00dev);
-	rt2x00rfkill_register(rt2x00dev);
+
+	/*
+	 * Start rfkill polling.
+	 */
+	if (!test_bit(REQUIRE_DELAYED_RFKILL, &rt2x00dev->cap_flags))
+		rt2x00rfkill_register(rt2x00dev);
 
 	return 0;
 
@@ -1391,6 +1403,12 @@ void rt2x00lib_remove_dev(struct rt2x00_dev *rt2x00dev)
 	clear_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags);
 
 	/*
+	 * Stop rfkill polling.
+	 */
+	if (!test_bit(REQUIRE_DELAYED_RFKILL, &rt2x00dev->cap_flags))
+		rt2x00rfkill_unregister(rt2x00dev);
+
+	/*
 	 * Disable radio.
 	 */
 	rt2x00lib_disable_radio(rt2x00dev);
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index a87ee9b6585a..004dff9b962d 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -487,6 +487,8 @@ int rt2x00mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	crypto.cipher = rt2x00crypto_key_to_cipher(key);
 	if (crypto.cipher == CIPHER_NONE)
 		return -EOPNOTSUPP;
+	if (crypto.cipher == CIPHER_TKIP && rt2x00_is_usb(rt2x00dev))
+		return -EOPNOTSUPP;
 
 	crypto.cmd = cmd;
 
@@ -749,7 +751,8 @@ void rt2x00mac_rfkill_poll(struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_rfkill_poll);
 
-void rt2x00mac_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+void rt2x00mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		     u32 queues, bool drop)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
 	struct data_queue *queue;
diff --git a/drivers/net/wireless/rt2x00/rt2x00usb.c b/drivers/net/wireless/rt2x00/rt2x00usb.c
index 10572452cc21..86c43d112a4b 100644
--- a/drivers/net/wireless/rt2x00/rt2x00usb.c
+++ b/drivers/net/wireless/rt2x00/rt2x00usb.c
@@ -68,6 +68,12 @@ int rt2x00usb_vendor_request(struct rt2x00_dev *rt2x00dev,
 		}
 	}
 
+	/* If the port is powered down, we get a -EPROTO error, and this
+	 * leads to a endless loop. So just say that the device is gone.
+	 */
+	if (status == -EPROTO)
+		clear_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags);
+
 	rt2x00_err(rt2x00dev,
 		   "Vendor Request 0x%02x failed for offset 0x%04x with error %d\n",
 		   request, offset, status);
diff --git a/drivers/net/wireless/rt2x00/rt2x00usb.h b/drivers/net/wireless/rt2x00/rt2x00usb.h
index e7bcf62347d5..831b65f93feb 100644
--- a/drivers/net/wireless/rt2x00/rt2x00usb.h
+++ b/drivers/net/wireless/rt2x00/rt2x00usb.h
@@ -93,6 +93,7 @@ enum rt2x00usb_mode_offset {
 	USB_MODE_SLEEP = 7,	/* RT73USB */
 	USB_MODE_FIRMWARE = 8,	/* RT73USB */
 	USB_MODE_WAKEUP = 9,	/* RT73USB */
+	USB_MODE_AUTORUN = 17, /* RT2800USB */
 };
 
 /**
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index 24402984ee57..9048a9cbe52c 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -2031,13 +2031,14 @@ static void rt61pci_write_beacon(struct queue_entry *entry,
 static void rt61pci_clear_beacon(struct queue_entry *entry)
 {
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
-	u32 reg;
+	u32 orig_reg, reg;
 
 	/*
 	 * Disable beaconing while we are reloading the beacon data,
 	 * otherwise we might be sending out invalid data.
 	 */
-	rt2x00mmio_register_read(rt2x00dev, TXRX_CSR9, &reg);
+	rt2x00mmio_register_read(rt2x00dev, TXRX_CSR9, &orig_reg);
+	reg = orig_reg;
 	rt2x00_set_field32(&reg, TXRX_CSR9_BEACON_GEN, 0);
 	rt2x00mmio_register_write(rt2x00dev, TXRX_CSR9, reg);
 
@@ -2048,10 +2049,9 @@ static void rt61pci_clear_beacon(struct queue_entry *entry)
 				  HW_BEACON_OFFSET(entry->entry_idx), 0);
 
 	/*
-	 * Enable beaconing again.
+	 * Restore global beaconing state.
 	 */
-	rt2x00_set_field32(&reg, TXRX_CSR9_BEACON_GEN, 1);
-	rt2x00mmio_register_write(rt2x00dev, TXRX_CSR9, reg);
+	rt2x00mmio_register_write(rt2x00dev, TXRX_CSR9, orig_reg);
 }
 
 /*
diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c
index a140170b1eb3..95724ff9c726 100644
--- a/drivers/net/wireless/rt2x00/rt73usb.c
+++ b/drivers/net/wireless/rt2x00/rt73usb.c
@@ -1597,13 +1597,14 @@ static void rt73usb_clear_beacon(struct queue_entry *entry)
 {
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
 	unsigned int beacon_base;
-	u32 reg;
+	u32 orig_reg, reg;
 
 	/*
 	 * Disable beaconing while we are reloading the beacon data,
 	 * otherwise we might be sending out invalid data.
 	 */
-	rt2x00usb_register_read(rt2x00dev, TXRX_CSR9, &reg);
+	rt2x00usb_register_read(rt2x00dev, TXRX_CSR9, &orig_reg);
+	reg = orig_reg;
 	rt2x00_set_field32(&reg, TXRX_CSR9_BEACON_GEN, 0);
 	rt2x00usb_register_write(rt2x00dev, TXRX_CSR9, reg);
 
@@ -1614,10 +1615,9 @@ static void rt73usb_clear_beacon(struct queue_entry *entry)
 	rt2x00usb_register_write(rt2x00dev, beacon_base, 0);
 
 	/*
-	 * Enable beaconing again.
+	 * Restore beaconing state.
 	 */
-	rt2x00_set_field32(&reg, TXRX_CSR9_BEACON_GEN, 1);
-	rt2x00usb_register_write(rt2x00dev, TXRX_CSR9, reg);
+	rt2x00usb_register_write(rt2x00dev, TXRX_CSR9, orig_reg);
 }
 
 static int rt73usb_get_tx_data_len(struct queue_entry *entry)
diff --git a/drivers/net/wireless/rtl818x/rtl8180/Makefile b/drivers/net/wireless/rtl818x/rtl8180/Makefile
index 08b056db4a3b..21005bd8b43c 100644
--- a/drivers/net/wireless/rtl818x/rtl8180/Makefile
+++ b/drivers/net/wireless/rtl818x/rtl8180/Makefile
@@ -1,5 +1,5 @@
-rtl8180-objs		:= dev.o rtl8225.o sa2400.o max2820.o grf5101.o rtl8225se.o
+rtl818x_pci-objs	:= dev.o rtl8225.o sa2400.o max2820.o grf5101.o rtl8225se.o
 
-obj-$(CONFIG_RTL8180)	+= rtl8180.o
+obj-$(CONFIG_RTL8180)	+= rtl818x_pci.o
 
 ccflags-y += -Idrivers/net/wireless/rtl818x
diff --git a/drivers/net/wireless/rtl818x/rtl8180/dev.c b/drivers/net/wireless/rtl818x/rtl8180/dev.c
index 98d8256f0377..2c1c02bafa10 100644
--- a/drivers/net/wireless/rtl818x/rtl8180/dev.c
+++ b/drivers/net/wireless/rtl818x/rtl8180/dev.c
@@ -284,6 +284,8 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
 			rx_status.band = dev->conf.chandef.chan->band;
 			rx_status.mactime = tsft;
 			rx_status.flag |= RX_FLAG_MACTIME_START;
+			if (flags & RTL818X_RX_DESC_FLAG_SPLCP)
+				rx_status.flag |= RX_FLAG_SHORTPRE;
 			if (flags & RTL818X_RX_DESC_FLAG_CRC32_ERR)
 				rx_status.flag |= RX_FLAG_FAILED_FCS_CRC;
 
@@ -461,18 +463,23 @@ static void rtl8180_tx(struct ieee80211_hw *dev,
 			    RTL818X_TX_DESC_FLAG_NO_ENC;
 
 	rc_flags = info->control.rates[0].flags;
+
+	/* HW will perform RTS-CTS when only RTS flags is set.
+	 * HW will perform CTS-to-self when both RTS and CTS flags are set.
+	 * RTS rate and RTS duration will be used also for CTS-to-self.
+	 */
 	if (rc_flags & IEEE80211_TX_RC_USE_RTS_CTS) {
 		tx_flags |= RTL818X_TX_DESC_FLAG_RTS;
 		tx_flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
+		rts_duration = ieee80211_rts_duration(dev, priv->vif,
+						skb->len, info);
 	} else if (rc_flags & IEEE80211_TX_RC_USE_CTS_PROTECT) {
-		tx_flags |= RTL818X_TX_DESC_FLAG_CTS;
+		tx_flags |= RTL818X_TX_DESC_FLAG_RTS | RTL818X_TX_DESC_FLAG_CTS;
 		tx_flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
+		rts_duration = ieee80211_ctstoself_duration(dev, priv->vif,
+						skb->len, info);
 	}
 
-	if (rc_flags & IEEE80211_TX_RC_USE_RTS_CTS)
-		rts_duration = ieee80211_rts_duration(dev, priv->vif, skb->len,
-						      info);
-
 	if (priv->chip_family == RTL818X_CHIP_FAMILY_RTL8180) {
 		unsigned int remainder;
 
@@ -683,9 +690,8 @@ static void rtl8180_int_enable(struct ieee80211_hw *dev)
 	struct rtl8180_priv *priv = dev->priv;
 
 	if (priv->chip_family == RTL818X_CHIP_FAMILY_RTL8187SE) {
-		rtl818x_iowrite32(priv, &priv->map->IMR, IMR_TMGDOK |
-			  IMR_TBDER | IMR_THPDER |
-			  IMR_THPDER | IMR_THPDOK |
+		rtl818x_iowrite32(priv, &priv->map->IMR,
+			  IMR_TBDER | IMR_TBDOK |
 			  IMR_TVODER | IMR_TVODOK |
 			  IMR_TVIDER | IMR_TVIDOK |
 			  IMR_TBEDER | IMR_TBEDOK |
@@ -911,7 +917,10 @@ static int rtl8180_init_hw(struct ieee80211_hw *dev)
 		reg32 &= 0x00ffff00;
 		reg32 |= 0xb8000054;
 		rtl818x_iowrite32(priv, &priv->map->RF_PARA, reg32);
-	}
+	} else
+		/* stop unused queus (no dma alloc) */
+		rtl818x_iowrite8(priv, &priv->map->TX_DMA_POLLING,
+			    (1<<1) | (1<<2));
 
 	priv->rf->init(dev);
 
diff --git a/drivers/net/wireless/rtl818x/rtl8187/dev.c b/drivers/net/wireless/rtl818x/rtl8187/dev.c
index 0ca17cda48fa..629ad8cfa17b 100644
--- a/drivers/net/wireless/rtl818x/rtl8187/dev.c
+++ b/drivers/net/wireless/rtl818x/rtl8187/dev.c
@@ -253,14 +253,21 @@ static void rtl8187_tx(struct ieee80211_hw *dev,
 	flags |= ieee80211_get_tx_rate(dev, info)->hw_value << 24;
 	if (ieee80211_has_morefrags(tx_hdr->frame_control))
 		flags |= RTL818X_TX_DESC_FLAG_MOREFRAG;
+
+	/* HW will perform RTS-CTS when only RTS flags is set.
+	 * HW will perform CTS-to-self when both RTS and CTS flags are set.
+	 * RTS rate and RTS duration will be used also for CTS-to-self.
+	 */
 	if (info->control.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS) {
 		flags |= RTL818X_TX_DESC_FLAG_RTS;
 		flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
 		rts_dur = ieee80211_rts_duration(dev, priv->vif,
 						 skb->len, info);
 	} else if (info->control.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT) {
-		flags |= RTL818X_TX_DESC_FLAG_CTS;
+		flags |= RTL818X_TX_DESC_FLAG_RTS | RTL818X_TX_DESC_FLAG_CTS;
 		flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
+		rts_dur = ieee80211_ctstoself_duration(dev, priv->vif,
+						 skb->len, info);
 	}
 
 	if (info->flags & IEEE80211_TX_CTL_ASSIGN_SEQ) {
@@ -381,6 +388,8 @@ static void rtl8187_rx_cb(struct urb *urb)
 	rx_status.freq = dev->conf.chandef.chan->center_freq;
 	rx_status.band = dev->conf.chandef.chan->band;
 	rx_status.flag |= RX_FLAG_MACTIME_START;
+	if (flags & RTL818X_RX_DESC_FLAG_SPLCP)
+		rx_status.flag |= RX_FLAG_SHORTPRE;
 	if (flags & RTL818X_RX_DESC_FLAG_CRC32_ERR)
 		rx_status.flag |= RX_FLAG_FAILED_FCS_CRC;
 	memcpy(IEEE80211_SKB_RXCB(skb), &rx_status, sizeof(rx_status));
diff --git a/drivers/net/wireless/rtl818x/rtl818x.h b/drivers/net/wireless/rtl818x/rtl818x.h
index 45ea4e1c4abe..7abef95d278b 100644
--- a/drivers/net/wireless/rtl818x/rtl818x.h
+++ b/drivers/net/wireless/rtl818x/rtl818x.h
@@ -334,9 +334,9 @@ struct rtl818x_csr {
  * I don't like to introduce a ton of "reserved"..
  * They are for RTL8187SE
  */
-#define REG_ADDR1(addr)	((u8 __iomem *)priv->map + addr)
-#define REG_ADDR2(addr)	((__le16 __iomem *)priv->map + (addr >> 1))
-#define REG_ADDR4(addr)	((__le32 __iomem *)priv->map + (addr >> 2))
+#define REG_ADDR1(addr)	((u8 __iomem *)priv->map + (addr))
+#define REG_ADDR2(addr)	((__le16 __iomem *)priv->map + ((addr) >> 1))
+#define REG_ADDR4(addr)	((__le32 __iomem *)priv->map + ((addr) >> 2))
 
 #define FEMR_SE		REG_ADDR2(0x1D4)
 #define ARFR		REG_ADDR2(0x1E0)
diff --git a/drivers/net/wireless/rtlwifi/core.c b/drivers/net/wireless/rtlwifi/core.c
index 4ec424f26672..b1ed6d0796f6 100644
--- a/drivers/net/wireless/rtlwifi/core.c
+++ b/drivers/net/wireless/rtlwifi/core.c
@@ -1387,7 +1387,8 @@ static void rtl_op_rfkill_poll(struct ieee80211_hw *hw)
  * before switch channel or power save, or tx buffer packet
  * maybe send after offchannel or rf sleep, this may cause
  * dis-association by AP */
-static void rtl_op_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void rtl_op_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			 u32 queues, bool drop)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
diff --git a/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c b/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
index 94cd9df98381..b14cf5a10f44 100644
--- a/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
@@ -2515,23 +2515,3 @@ void rtl88ee_suspend(struct ieee80211_hw *hw)
 void rtl88ee_resume(struct ieee80211_hw *hw)
 {
 }
-
-/* Turn on AAP (RCR:bit 0) for promicuous mode. */
-void rtl88ee_allow_all_destaddr(struct ieee80211_hw *hw,
-				bool allow_all_da, bool write_into_reg)
-{
-	struct rtl_priv *rtlpriv = rtl_priv(hw);
-	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
-
-	if (allow_all_da) /* Set BIT0 */
-		rtlpci->receive_config |= RCR_AAP;
-	 else /* Clear BIT0 */
-		rtlpci->receive_config &= ~RCR_AAP;
-
-	if (write_into_reg)
-		rtl_write_dword(rtlpriv, REG_RCR, rtlpci->receive_config);
-
-	RT_TRACE(rtlpriv, COMP_TURBO | COMP_INIT, DBG_LOUD,
-		 "receive_config = 0x%08X, write_into_reg =%d\n",
-		 rtlpci->receive_config, write_into_reg);
-}
diff --git a/drivers/net/wireless/rtlwifi/rtl8188ee/hw.h b/drivers/net/wireless/rtlwifi/rtl8188ee/hw.h
index b4460a41bd01..1850fde881b5 100644
--- a/drivers/net/wireless/rtlwifi/rtl8188ee/hw.h
+++ b/drivers/net/wireless/rtlwifi/rtl8188ee/hw.h
@@ -61,8 +61,6 @@ void rtl8188ee_bt_reg_init(struct ieee80211_hw *hw);
 void rtl8188ee_bt_hw_init(struct ieee80211_hw *hw);
 void rtl88ee_suspend(struct ieee80211_hw *hw);
 void rtl88ee_resume(struct ieee80211_hw *hw);
-void rtl88ee_allow_all_destaddr(struct ieee80211_hw *hw,
-				bool allow_all_da, bool write_into_reg);
 void rtl88ee_fw_clk_off_timer_callback(unsigned long data);
 
 #endif
diff --git a/drivers/net/wireless/rtlwifi/rtl8188ee/sw.c b/drivers/net/wireless/rtlwifi/rtl8188ee/sw.c
index 1b4101bf9974..842d69349a37 100644
--- a/drivers/net/wireless/rtlwifi/rtl8188ee/sw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8188ee/sw.c
@@ -93,7 +93,7 @@ int rtl88e_init_sw_vars(struct ieee80211_hw *hw)
 	u8 tid;
 
 	rtl8188ee_bt_reg_init(hw);
-	rtlpci->msi_support = true;
+	rtlpci->msi_support = rtlpriv->cfg->mod_params->msi_support;
 
 	rtlpriv->dm.dm_initialgain_enable = 1;
 	rtlpriv->dm.dm_flag = 0;
@@ -255,7 +255,6 @@ static struct rtl_hal_ops rtl8188ee_hal_ops = {
 	.enable_hw_sec = rtl88ee_enable_hw_security_config,
 	.set_key = rtl88ee_set_key,
 	.init_sw_leds = rtl88ee_init_sw_leds,
-	.allow_all_destaddr = rtl88ee_allow_all_destaddr,
 	.get_bbreg = rtl88e_phy_query_bb_reg,
 	.set_bbreg = rtl88e_phy_set_bb_reg,
 	.get_rfreg = rtl88e_phy_query_rf_reg,
@@ -267,6 +266,7 @@ static struct rtl_mod_params rtl88ee_mod_params = {
 	.inactiveps = true,
 	.swctrl_lps = false,
 	.fwctrl_lps = true,
+	.msi_support = false,
 	.debug = DBG_EMERG,
 };
 
@@ -383,10 +383,12 @@ module_param_named(debug, rtl88ee_mod_params.debug, int, 0444);
 module_param_named(ips, rtl88ee_mod_params.inactiveps, bool, 0444);
 module_param_named(swlps, rtl88ee_mod_params.swctrl_lps, bool, 0444);
 module_param_named(fwlps, rtl88ee_mod_params.fwctrl_lps, bool, 0444);
+module_param_named(msi, rtl88ee_mod_params.msi_support, bool, 0444);
 MODULE_PARM_DESC(swenc, "Set to 1 for software crypto (default 0)\n");
 MODULE_PARM_DESC(ips, "Set to 0 to not use link power save (default 1)\n");
 MODULE_PARM_DESC(swlps, "Set to 1 to use SW control power save (default 0)\n");
 MODULE_PARM_DESC(fwlps, "Set to 1 to use FW control power save (default 1)\n");
+MODULE_PARM_DESC(msi, "Set to 1 to use MSI interrupts mode (default 0)\n");
 MODULE_PARM_DESC(debug, "Set debug level (0-5) (default 0)");
 
 static SIMPLE_DEV_PM_OPS(rtlwifi_pm_ops, rtl_pci_suspend, rtl_pci_resume);
diff --git a/drivers/net/wireless/rtlwifi/rtl8192ce/hw.c b/drivers/net/wireless/rtlwifi/rtl8192ce/hw.c
index 55adf043aef7..cdecb0fd4d8e 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192ce/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192ce/hw.c
@@ -2423,24 +2423,3 @@ void rtl92ce_suspend(struct ieee80211_hw *hw)
 void rtl92ce_resume(struct ieee80211_hw *hw)
 {
 }
-
-/* Turn on AAP (RCR:bit 0) for promicuous mode. */
-void rtl92ce_allow_all_destaddr(struct ieee80211_hw *hw,
-	bool allow_all_da, bool write_into_reg)
-{
-	struct rtl_priv *rtlpriv = rtl_priv(hw);
-	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
-
-	if (allow_all_da) {/* Set BIT0 */
-		rtlpci->receive_config |= RCR_AAP;
-	} else {/* Clear BIT0 */
-		rtlpci->receive_config &= ~RCR_AAP;
-	}
-
-	if (write_into_reg)
-		rtl_write_dword(rtlpriv, REG_RCR, rtlpci->receive_config);
-
-	RT_TRACE(rtlpriv, COMP_TURBO | COMP_INIT, DBG_LOUD,
-		 "receive_config=0x%08X, write_into_reg=%d\n",
-		 rtlpci->receive_config, write_into_reg);
-}
diff --git a/drivers/net/wireless/rtlwifi/rtl8192ce/hw.h b/drivers/net/wireless/rtlwifi/rtl8192ce/hw.h
index 2d063b0c7760..5533070f266c 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192ce/hw.h
+++ b/drivers/net/wireless/rtlwifi/rtl8192ce/hw.h
@@ -76,7 +76,5 @@ void rtl8192ce_bt_reg_init(struct ieee80211_hw *hw);
 void rtl8192ce_bt_hw_init(struct ieee80211_hw *hw);
 void rtl92ce_suspend(struct ieee80211_hw *hw);
 void rtl92ce_resume(struct ieee80211_hw *hw);
-void rtl92ce_allow_all_destaddr(struct ieee80211_hw *hw,
-				bool allow_all_da, bool write_into_reg);
 
 #endif
diff --git a/drivers/net/wireless/rtlwifi/rtl8192ce/sw.c b/drivers/net/wireless/rtlwifi/rtl8192ce/sw.c
index b790320d2030..12f21f4073e8 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192ce/sw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192ce/sw.c
@@ -229,7 +229,6 @@ static struct rtl_hal_ops rtl8192ce_hal_ops = {
 	.enable_hw_sec = rtl92ce_enable_hw_security_config,
 	.set_key = rtl92ce_set_key,
 	.init_sw_leds = rtl92ce_init_sw_leds,
-	.allow_all_destaddr = rtl92ce_allow_all_destaddr,
 	.get_bbreg = rtl92c_phy_query_bb_reg,
 	.set_bbreg = rtl92c_phy_set_bb_reg,
 	.set_rfreg = rtl92ce_phy_set_rf_reg,
diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/hw.c b/drivers/net/wireless/rtlwifi/rtl8192cu/hw.c
index 07cb06da6729..a903c2671b4d 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192cu/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192cu/hw.c
@@ -511,7 +511,7 @@ static int _rtl92cu_init_power_on(struct ieee80211_hw *hw)
 			pr_info("MAC auto ON okay!\n");
 			break;
 		}
-		if (pollingCount++ > 100) {
+		if (pollingCount++ > 1000) {
 			RT_TRACE(rtlpriv, COMP_INIT, DBG_EMERG,
 				 "Failed to polling REG_APS_FSMCO[APFM_ONMAC] done!\n");
 			return -ENODEV;
diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c b/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c
index c61311084d7e..361435f8608a 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c
@@ -395,9 +395,6 @@ static struct usb_driver rtl8192cu_driver = {
 	/* .resume = rtl_usb_resume, */
 	/* .reset_resume = rtl8192c_resume, */
 #endif /* CONFIG_PM */
-#ifdef CONFIG_AUTOSUSPEND
-	.supports_autosuspend = 1,
-#endif
 	.disable_hub_initiated_lpm = 1,
 };
 
diff --git a/drivers/net/wireless/rtlwifi/rtl8192se/hw.c b/drivers/net/wireless/rtlwifi/rtl8192se/hw.c
index 9098558d916d..1c7101bcd790 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192se/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192se/hw.c
@@ -2544,23 +2544,3 @@ void rtl92se_resume(struct ieee80211_hw *hw)
 		pci_write_config_dword(rtlpci->pdev, 0x40,
 			val & 0xffff00ff);
 }
-
-/* Turn on AAP (RCR:bit 0) for promicuous mode. */
-void rtl92se_allow_all_destaddr(struct ieee80211_hw *hw,
-				bool allow_all_da, bool write_into_reg)
-{
-	struct rtl_priv *rtlpriv = rtl_priv(hw);
-	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
-
-	if (allow_all_da) /* Set BIT0 */
-		rtlpci->receive_config |= RCR_AAP;
-	else /* Clear BIT0 */
-		rtlpci->receive_config &= ~RCR_AAP;
-
-	if (write_into_reg)
-		rtl_write_dword(rtlpriv, RCR, rtlpci->receive_config);
-
-	RT_TRACE(rtlpriv, COMP_TURBO | COMP_INIT, DBG_LOUD,
-		 "receive_config=0x%08X, write_into_reg=%d\n",
-		 rtlpci->receive_config, write_into_reg);
-}
diff --git a/drivers/net/wireless/rtlwifi/rtl8192se/hw.h b/drivers/net/wireless/rtlwifi/rtl8192se/hw.h
index da48aa8cbe6f..4cacee10f31e 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192se/hw.h
+++ b/drivers/net/wireless/rtlwifi/rtl8192se/hw.h
@@ -74,7 +74,5 @@ void rtl92se_set_key(struct ieee80211_hw *hw,
 		     u8 enc_algo, bool is_wepkey, bool clear_all);
 void rtl92se_suspend(struct ieee80211_hw *hw);
 void rtl92se_resume(struct ieee80211_hw *hw);
-void rtl92se_allow_all_destaddr(struct ieee80211_hw *hw,
-				bool allow_all_da, bool write_into_reg);
 
 #endif
diff --git a/drivers/net/wireless/rtlwifi/rtl8192se/sw.c b/drivers/net/wireless/rtlwifi/rtl8192se/sw.c
index 2e8e6f8d2d51..1bff2a0f7600 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192se/sw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192se/sw.c
@@ -290,7 +290,6 @@ static struct rtl_hal_ops rtl8192se_hal_ops = {
 	.enable_hw_sec = rtl92se_enable_hw_security_config,
 	.set_key = rtl92se_set_key,
 	.init_sw_leds = rtl92se_init_sw_leds,
-	.allow_all_destaddr = rtl92se_allow_all_destaddr,
 	.get_bbreg = rtl92s_phy_query_bb_reg,
 	.set_bbreg = rtl92s_phy_set_bb_reg,
 	.get_rfreg = rtl92s_phy_query_rf_reg,
diff --git a/drivers/net/wireless/rtlwifi/rtl8723ae/hal_bt_coexist.c b/drivers/net/wireless/rtlwifi/rtl8723ae/hal_bt_coexist.c
index 48fee1be78c2..5b4a714f3c8c 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723ae/hal_bt_coexist.c
+++ b/drivers/net/wireless/rtlwifi/rtl8723ae/hal_bt_coexist.c
@@ -32,7 +32,6 @@
 #include "dm.h"
 #include "fw.h"
 #include "../rtl8723com/fw_common.h"
-#include "../rtl8723com/fw_common.h"
 #include "phy.h"
 #include "reg.h"
 #include "hal_btc.h"
diff --git a/drivers/net/wireless/rtlwifi/rtl8723ae/hw.c b/drivers/net/wireless/rtlwifi/rtl8723ae/hw.c
index 65c9e80e1f78..87f69166a7ed 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723ae/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8723ae/hw.c
@@ -2383,24 +2383,3 @@ void rtl8723ae_suspend(struct ieee80211_hw *hw)
 void rtl8723ae_resume(struct ieee80211_hw *hw)
 {
 }
-
-/* Turn on AAP (RCR:bit 0) for promicuous mode. */
-void rtl8723ae_allow_all_destaddr(struct ieee80211_hw *hw,
-	bool allow_all_da, bool write_into_reg)
-{
-	struct rtl_priv *rtlpriv = rtl_priv(hw);
-	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
-
-	if (allow_all_da) /* Set BIT0 */
-		rtlpci->receive_config |= RCR_AAP;
-	else /* Clear BIT0 */
-		rtlpci->receive_config &= ~RCR_AAP;
-
-	if (write_into_reg)
-		rtl_write_dword(rtlpriv, REG_RCR, rtlpci->receive_config);
-
-
-	RT_TRACE(rtlpriv, COMP_TURBO | COMP_INIT, DBG_LOUD,
-		 "receive_config=0x%08X, write_into_reg=%d\n",
-		 rtlpci->receive_config, write_into_reg);
-}
diff --git a/drivers/net/wireless/rtlwifi/rtl8723ae/hw.h b/drivers/net/wireless/rtlwifi/rtl8723ae/hw.h
index 6fa24f79b1d7..d3bc39fb27a5 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723ae/hw.h
+++ b/drivers/net/wireless/rtlwifi/rtl8723ae/hw.h
@@ -67,7 +67,5 @@ void rtl8723ae_bt_reg_init(struct ieee80211_hw *hw);
 void rtl8723ae_bt_hw_init(struct ieee80211_hw *hw);
 void rtl8723ae_suspend(struct ieee80211_hw *hw);
 void rtl8723ae_resume(struct ieee80211_hw *hw);
-void rtl8723ae_allow_all_destaddr(struct ieee80211_hw *hw,
-				  bool allow_all_da, bool write_into_reg);
 
 #endif
diff --git a/drivers/net/wireless/rtlwifi/rtl8723ae/sw.c b/drivers/net/wireless/rtlwifi/rtl8723ae/sw.c
index 1087a3bd07fa..73cba1eec8cf 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723ae/sw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8723ae/sw.c
@@ -238,7 +238,6 @@ static struct rtl_hal_ops rtl8723ae_hal_ops = {
 	.enable_hw_sec = rtl8723ae_enable_hw_security_config,
 	.set_key = rtl8723ae_set_key,
 	.init_sw_leds = rtl8723ae_init_sw_leds,
-	.allow_all_destaddr = rtl8723ae_allow_all_destaddr,
 	.get_bbreg = rtl8723_phy_query_bb_reg,
 	.set_bbreg = rtl8723_phy_set_bb_reg,
 	.get_rfreg = rtl8723ae_phy_query_rf_reg,
diff --git a/drivers/net/wireless/rtlwifi/rtl8723be/hw.c b/drivers/net/wireless/rtlwifi/rtl8723be/hw.c
index 0fdf0909321f..3d555495b453 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723be/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8723be/hw.c
@@ -2501,23 +2501,3 @@ void rtl8723be_suspend(struct ieee80211_hw *hw)
 void rtl8723be_resume(struct ieee80211_hw *hw)
 {
 }
-
-/* Turn on AAP (RCR:bit 0) for promicuous mode. */
-void rtl8723be_allow_all_destaddr(struct ieee80211_hw *hw, bool allow_all_da,
-				  bool write_into_reg)
-{
-	struct rtl_priv *rtlpriv = rtl_priv(hw);
-	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
-
-	if (allow_all_da) /* Set BIT0 */
-		rtlpci->receive_config |= RCR_AAP;
-	else /* Clear BIT0 */
-		rtlpci->receive_config &= ~RCR_AAP;
-
-	if (write_into_reg)
-		rtl_write_dword(rtlpriv, REG_RCR, rtlpci->receive_config);
-
-	RT_TRACE(rtlpriv, COMP_TURBO | COMP_INIT, DBG_LOUD,
-		 "receive_config = 0x%08X, write_into_reg =%d\n",
-		 rtlpci->receive_config, write_into_reg);
-}
diff --git a/drivers/net/wireless/rtlwifi/rtl8723be/hw.h b/drivers/net/wireless/rtlwifi/rtl8723be/hw.h
index b7449a9b57e4..64c7551af6b7 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723be/hw.h
+++ b/drivers/net/wireless/rtlwifi/rtl8723be/hw.h
@@ -59,6 +59,4 @@ void rtl8723be_bt_reg_init(struct ieee80211_hw *hw);
 void rtl8723be_bt_hw_init(struct ieee80211_hw *hw);
 void rtl8723be_suspend(struct ieee80211_hw *hw);
 void rtl8723be_resume(struct ieee80211_hw *hw);
-void rtl8723be_allow_all_destaddr(struct ieee80211_hw *hw, bool allow_all_da,
-				  bool write_into_reg);
 #endif
diff --git a/drivers/net/wireless/rtlwifi/rtl8723be/sw.c b/drivers/net/wireless/rtlwifi/rtl8723be/sw.c
index b4577ebc4bb0..ff12bf41644b 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723be/sw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8723be/sw.c
@@ -92,7 +92,7 @@ int rtl8723be_init_sw_vars(struct ieee80211_hw *hw)
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
 
 	rtl8723be_bt_reg_init(hw);
-	rtlpci->msi_support = true;
+	rtlpci->msi_support = rtlpriv->cfg->mod_params->msi_support;
 	rtlpriv->btcoexist.btc_ops = rtl_btc_get_ops_pointer();
 
 	rtlpriv->dm.dm_initialgain_enable = 1;
@@ -253,6 +253,7 @@ static struct rtl_mod_params rtl8723be_mod_params = {
 	.inactiveps = true,
 	.swctrl_lps = false,
 	.fwctrl_lps = true,
+	.msi_support = false,
 	.debug = DBG_EMERG,
 };
 
@@ -365,9 +366,11 @@ module_param_named(debug, rtl8723be_mod_params.debug, int, 0444);
 module_param_named(ips, rtl8723be_mod_params.inactiveps, bool, 0444);
 module_param_named(swlps, rtl8723be_mod_params.swctrl_lps, bool, 0444);
 module_param_named(fwlps, rtl8723be_mod_params.fwctrl_lps, bool, 0444);
+module_param_named(msi, rtl8723be_mod_params.msi_support, bool, 0444);
 MODULE_PARM_DESC(swenc, "using hardware crypto (default 0 [hardware])\n");
 MODULE_PARM_DESC(ips, "using no link power save (default 1 is open)\n");
 MODULE_PARM_DESC(fwlps, "using linked fw control power save (default 1 is open)\n");
+MODULE_PARM_DESC(msi, "Set to 1 to use MSI interrupts mode (default 0)\n");
 MODULE_PARM_DESC(debug, "Set debug level (0-5) (default 0)");
 
 static SIMPLE_DEV_PM_OPS(rtlwifi_pm_ops, rtl_pci_suspend, rtl_pci_resume);
diff --git a/drivers/net/wireless/rtlwifi/rtl8723be/trx.c b/drivers/net/wireless/rtlwifi/rtl8723be/trx.c
index e0a0d8c8fed5..969eaea5eddd 100644
--- a/drivers/net/wireless/rtlwifi/rtl8723be/trx.c
+++ b/drivers/net/wireless/rtlwifi/rtl8723be/trx.c
@@ -33,7 +33,6 @@
 #include "trx.h"
 #include "led.h"
 #include "dm.h"
-#include "phy.h"
 
 static u8 _rtl8723be_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue)
 {
diff --git a/drivers/net/wireless/rtlwifi/wifi.h b/drivers/net/wireless/rtlwifi/wifi.h
index 6965afdf572a..407a7936d364 100644
--- a/drivers/net/wireless/rtlwifi/wifi.h
+++ b/drivers/net/wireless/rtlwifi/wifi.h
@@ -1960,8 +1960,6 @@ struct rtl_hal_ops {
 			  u32 regaddr, u32 bitmask);
 	void (*set_rfreg) (struct ieee80211_hw *hw, enum radio_path rfpath,
 			   u32 regaddr, u32 bitmask, u32 data);
-	void (*allow_all_destaddr)(struct ieee80211_hw *hw,
-		bool allow_all_da, bool write_into_reg);
 	void (*linked_set_reg) (struct ieee80211_hw *hw);
 	void (*chk_switch_dmdp) (struct ieee80211_hw *hw);
 	void (*dualmac_easy_concurrent) (struct ieee80211_hw *hw);
@@ -2030,6 +2028,10 @@ struct rtl_mod_params {
 
 	/* default: 1 = using linked fw power save */
 	bool fwctrl_lps;
+
+	/* default: 0 = not using MSI interrupts mode */
+	/* submodules should set their own defalut value */
+	bool msi_support;
 };
 
 struct rtl_hal_usbint_cfg {
diff --git a/drivers/net/wireless/ti/wl1251/acx.c b/drivers/net/wireless/ti/wl1251/acx.c
index 5a4ec56c83d0..5695628757ee 100644
--- a/drivers/net/wireless/ti/wl1251/acx.c
+++ b/drivers/net/wireless/ti/wl1251/acx.c
@@ -2,7 +2,6 @@
 
 #include <linux/module.h>
 #include <linux/slab.h>
-#include <linux/crc7.h>
 
 #include "wl1251.h"
 #include "reg.h"
diff --git a/drivers/net/wireless/ti/wl1251/cmd.c b/drivers/net/wireless/ti/wl1251/cmd.c
index bf1fa18b9786..ede31f048ef9 100644
--- a/drivers/net/wireless/ti/wl1251/cmd.c
+++ b/drivers/net/wireless/ti/wl1251/cmd.c
@@ -2,7 +2,6 @@
 
 #include <linux/module.h>
 #include <linux/slab.h>
-#include <linux/crc7.h>
 #include <linux/etherdevice.h>
 
 #include "wl1251.h"
diff --git a/drivers/net/wireless/ti/wl1251/event.c b/drivers/net/wireless/ti/wl1251/event.c
index db0105313745..c98630394a1a 100644
--- a/drivers/net/wireless/ti/wl1251/event.c
+++ b/drivers/net/wireless/ti/wl1251/event.c
@@ -124,11 +124,12 @@ static int wl1251_event_process(struct wl1251 *wl, struct event_mailbox *mbox)
 			return ret;
 	}
 
-	if (wl->vif && vector & SYNCHRONIZATION_TIMEOUT_EVENT_ID) {
+	if (vector & SYNCHRONIZATION_TIMEOUT_EVENT_ID) {
 		wl1251_debug(DEBUG_EVENT, "SYNCHRONIZATION_TIMEOUT_EVENT");
 
 		/* indicate to the stack, that beacons have been lost */
-		ieee80211_beacon_loss(wl->vif);
+		if (wl->vif && wl->vif->type == NL80211_IFTYPE_STATION)
+			ieee80211_beacon_loss(wl->vif);
 	}
 
 	if (vector & REGAINED_BSS_EVENT_ID) {
diff --git a/drivers/net/wireless/ti/wl1251/main.c b/drivers/net/wireless/ti/wl1251/main.c
index 757e25784a8a..4e782f18ae34 100644
--- a/drivers/net/wireless/ti/wl1251/main.c
+++ b/drivers/net/wireless/ti/wl1251/main.c
@@ -550,6 +550,34 @@ static void wl1251_op_remove_interface(struct ieee80211_hw *hw,
 	mutex_unlock(&wl->mutex);
 }
 
+static int wl1251_build_null_data(struct wl1251 *wl)
+{
+	struct sk_buff *skb = NULL;
+	int size;
+	void *ptr;
+	int ret = -ENOMEM;
+
+	if (wl->bss_type == BSS_TYPE_IBSS) {
+		size = sizeof(struct wl12xx_null_data_template);
+		ptr = NULL;
+	} else {
+		skb = ieee80211_nullfunc_get(wl->hw, wl->vif);
+		if (!skb)
+			goto out;
+		size = skb->len;
+		ptr = skb->data;
+	}
+
+	ret = wl1251_cmd_template_set(wl, CMD_NULL_DATA, ptr, size);
+
+out:
+	dev_kfree_skb(skb);
+	if (ret)
+		wl1251_warning("cmd buld null data failed: %d", ret);
+
+	return ret;
+}
+
 static int wl1251_build_qos_null_data(struct wl1251 *wl)
 {
 	struct ieee80211_qos_hdr template;
@@ -687,16 +715,6 @@ static int wl1251_op_config(struct ieee80211_hw *hw, u32 changed)
 		wl->power_level = conf->power_level;
 	}
 
-	/*
-	 * Tell stack that connection is lost because hw encryption isn't
-	 * supported in monitor mode.
-	 * This requires temporary enabling of the hw connection monitor flag
-	 */
-	if ((changed & IEEE80211_CONF_CHANGE_MONITOR) && wl->vif) {
-		wl->hw->flags |= IEEE80211_HW_CONNECTION_MONITOR;
-		ieee80211_connection_loss(wl->vif);
-	}
-
 out_sleep:
 	wl1251_ps_elp_sleep(wl);
 
@@ -1103,24 +1121,19 @@ static void wl1251_op_bss_info_changed(struct ieee80211_hw *hw,
 		wl->rssi_thold = bss_conf->cqm_rssi_thold;
 	}
 
-	if (changed & BSS_CHANGED_BSSID) {
+	if ((changed & BSS_CHANGED_BSSID) &&
+	    memcmp(wl->bssid, bss_conf->bssid, ETH_ALEN)) {
 		memcpy(wl->bssid, bss_conf->bssid, ETH_ALEN);
 
-		skb = ieee80211_nullfunc_get(wl->hw, wl->vif);
-		if (!skb)
-			goto out_sleep;
-
-		ret = wl1251_cmd_template_set(wl, CMD_NULL_DATA,
-					      skb->data, skb->len);
-		dev_kfree_skb(skb);
-		if (ret < 0)
-			goto out_sleep;
+		if (!is_zero_ether_addr(wl->bssid)) {
+			ret = wl1251_build_null_data(wl);
+			if (ret < 0)
+				goto out_sleep;
 
-		ret = wl1251_build_qos_null_data(wl);
-		if (ret < 0)
-			goto out;
+			ret = wl1251_build_qos_null_data(wl);
+			if (ret < 0)
+				goto out_sleep;
 
-		if (wl->bss_type != BSS_TYPE_IBSS) {
 			ret = wl1251_join(wl, wl->bss_type, wl->channel,
 					  wl->beacon_int, wl->dtim_period);
 			if (ret < 0)
@@ -1129,9 +1142,6 @@ static void wl1251_op_bss_info_changed(struct ieee80211_hw *hw,
 	}
 
 	if (changed & BSS_CHANGED_ASSOC) {
-		/* Disable temporary enabled hw connection monitor flag */
-		wl->hw->flags &= ~IEEE80211_HW_CONNECTION_MONITOR;
-
 		if (bss_conf->assoc) {
 			wl->beacon_int = bss_conf->beacon_int;
 
@@ -1216,8 +1226,8 @@ static void wl1251_op_bss_info_changed(struct ieee80211_hw *hw,
 		if (ret < 0)
 			goto out_sleep;
 
-		ret = wl1251_join(wl, wl->bss_type, wl->beacon_int,
-				  wl->channel, wl->dtim_period);
+		ret = wl1251_join(wl, wl->bss_type, wl->channel,
+				  wl->beacon_int, wl->dtim_period);
 
 		if (ret < 0)
 			goto out_sleep;
diff --git a/drivers/net/wireless/ti/wl1251/spi.c b/drivers/net/wireless/ti/wl1251/spi.c
index b06d36d99362..a0aa8fa72392 100644
--- a/drivers/net/wireless/ti/wl1251/spi.c
+++ b/drivers/net/wireless/ti/wl1251/spi.c
@@ -23,6 +23,7 @@
 #include <linux/irq.h>
 #include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/swab.h>
 #include <linux/crc7.h>
 #include <linux/spi/spi.h>
 #include <linux/wl12xx.h>
@@ -83,47 +84,44 @@ static void wl1251_spi_reset(struct wl1251 *wl)
 
 static void wl1251_spi_wake(struct wl1251 *wl)
 {
-	u8 crc[WSPI_INIT_CMD_CRC_LEN], *cmd;
 	struct spi_transfer t;
 	struct spi_message m;
+	u8 *cmd = kzalloc(WSPI_INIT_CMD_LEN, GFP_KERNEL);
 
-	cmd = kzalloc(WSPI_INIT_CMD_LEN, GFP_KERNEL);
 	if (!cmd) {
 		wl1251_error("could not allocate cmd for spi init");
 		return;
 	}
 
-	memset(crc, 0, sizeof(crc));
 	memset(&t, 0, sizeof(t));
 	spi_message_init(&m);
 
 	/* Set WSPI_INIT_COMMAND
 	 * the data is being send from the MSB to LSB
 	 */
-	cmd[2] = 0xff;
-	cmd[3] = 0xff;
-	cmd[1] = WSPI_INIT_CMD_START | WSPI_INIT_CMD_TX;
-	cmd[0] = 0;
-	cmd[7] = 0;
-	cmd[6] |= HW_ACCESS_WSPI_INIT_CMD_MASK << 3;
-	cmd[6] |= HW_ACCESS_WSPI_FIXED_BUSY_LEN & WSPI_INIT_CMD_FIXEDBUSY_LEN;
+	cmd[0] = 0xff;
+	cmd[1] = 0xff;
+	cmd[2] = WSPI_INIT_CMD_START | WSPI_INIT_CMD_TX;
+	cmd[3] = 0;
+	cmd[4] = 0;
+	cmd[5] = HW_ACCESS_WSPI_INIT_CMD_MASK << 3;
+	cmd[5] |= HW_ACCESS_WSPI_FIXED_BUSY_LEN & WSPI_INIT_CMD_FIXEDBUSY_LEN;
+
+	cmd[6] = WSPI_INIT_CMD_IOD | WSPI_INIT_CMD_IP | WSPI_INIT_CMD_CS
+		| WSPI_INIT_CMD_WSPI | WSPI_INIT_CMD_WS;
 
 	if (HW_ACCESS_WSPI_FIXED_BUSY_LEN == 0)
-		cmd[5] |=  WSPI_INIT_CMD_DIS_FIXEDBUSY;
+		cmd[6] |= WSPI_INIT_CMD_DIS_FIXEDBUSY;
 	else
-		cmd[5] |= WSPI_INIT_CMD_EN_FIXEDBUSY;
-
-	cmd[5] |= WSPI_INIT_CMD_IOD | WSPI_INIT_CMD_IP | WSPI_INIT_CMD_CS
-		| WSPI_INIT_CMD_WSPI | WSPI_INIT_CMD_WS;
-
-	crc[0] = cmd[1];
-	crc[1] = cmd[0];
-	crc[2] = cmd[7];
-	crc[3] = cmd[6];
-	crc[4] = cmd[5];
+		cmd[6] |= WSPI_INIT_CMD_EN_FIXEDBUSY;
 
-	cmd[4] |= crc7(0, crc, WSPI_INIT_CMD_CRC_LEN) << 1;
-	cmd[4] |= WSPI_INIT_CMD_END;
+	cmd[7] = crc7_be(0, cmd+2, WSPI_INIT_CMD_CRC_LEN) | WSPI_INIT_CMD_END;
+	/*
+	 * The above is the logical order; it must actually be stored
+	 * in the buffer byte-swapped.
+	 */
+	__swab32s((u32 *)cmd);
+	__swab32s((u32 *)cmd+1);
 
 	t.tx_buf = cmd;
 	t.len = WSPI_INIT_CMD_LEN;
diff --git a/drivers/net/wireless/ti/wlcore/debugfs.h b/drivers/net/wireless/ti/wlcore/debugfs.h
index f7381dd69009..0f2cfb0d2a9e 100644
--- a/drivers/net/wireless/ti/wlcore/debugfs.h
+++ b/drivers/net/wireless/ti/wlcore/debugfs.h
@@ -57,7 +57,7 @@ static const struct file_operations name## _ops = {			\
 					    wl, &name## _ops);		\
 		if (!entry || IS_ERR(entry))				\
 			goto err;					\
-	} while (0);
+	} while (0)
 
 
 #define DEBUGFS_ADD_PREFIX(prefix, name, parent)			\
@@ -66,7 +66,7 @@ static const struct file_operations name## _ops = {			\
 				    wl, &prefix## _## name## _ops);	\
 		if (!entry || IS_ERR(entry))				\
 			goto err;					\
-	} while (0);
+	} while (0)
 
 #define DEBUGFS_FWSTATS_FILE(sub, name, fmt, struct_type)		\
 static ssize_t sub## _ ##name## _read(struct file *file,		\
diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index e71eae353368..3d6028e62750 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -1416,7 +1416,7 @@ void wl1271_rx_filter_free(struct wl12xx_rx_filter *filter)
 
 int wl1271_rx_filter_alloc_field(struct wl12xx_rx_filter *filter,
 				 u16 offset, u8 flags,
-				 u8 *pattern, u8 len)
+				 const u8 *pattern, u8 len)
 {
 	struct wl12xx_rx_filter_field *field;
 
@@ -5184,7 +5184,8 @@ out:
 	mutex_unlock(&wl->mutex);
 }
 
-static void wlcore_op_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void wlcore_op_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    u32 queues, bool drop)
 {
 	struct wl1271 *wl = hw->priv;
 
diff --git a/drivers/net/wireless/ti/wlcore/sdio.c b/drivers/net/wireless/ti/wlcore/sdio.c
index 29ef2492951f..d3dd7bfdf3f1 100644
--- a/drivers/net/wireless/ti/wlcore/sdio.c
+++ b/drivers/net/wireless/ti/wlcore/sdio.c
@@ -217,7 +217,7 @@ static struct wl1271_if_operations sdio_ops = {
 static int wl1271_probe(struct sdio_func *func,
 				  const struct sdio_device_id *id)
 {
-	struct wlcore_platdev_data *pdev_data;
+	struct wlcore_platdev_data pdev_data;
 	struct wl12xx_sdio_glue *glue;
 	struct resource res[1];
 	mmc_pm_flag_t mmcflags;
@@ -228,16 +228,13 @@ static int wl1271_probe(struct sdio_func *func,
 	if (func->num != 0x02)
 		return -ENODEV;
 
-	pdev_data = kzalloc(sizeof(*pdev_data), GFP_KERNEL);
-	if (!pdev_data)
-		goto out;
-
-	pdev_data->if_ops = &sdio_ops;
+	memset(&pdev_data, 0x00, sizeof(pdev_data));
+	pdev_data.if_ops = &sdio_ops;
 
 	glue = kzalloc(sizeof(*glue), GFP_KERNEL);
 	if (!glue) {
 		dev_err(&func->dev, "can't allocate glue\n");
-		goto out_free_pdev_data;
+		goto out;
 	}
 
 	glue->dev = &func->dev;
@@ -248,9 +245,9 @@ static int wl1271_probe(struct sdio_func *func,
 	/* Use block mode for transferring over one block size of data */
 	func->card->quirks |= MMC_QUIRK_BLKSZ_FOR_BYTE_MODE;
 
-	pdev_data->pdata = wl12xx_get_platform_data();
-	if (IS_ERR(pdev_data->pdata)) {
-		ret = PTR_ERR(pdev_data->pdata);
+	pdev_data.pdata = wl12xx_get_platform_data();
+	if (IS_ERR(pdev_data.pdata)) {
+		ret = PTR_ERR(pdev_data.pdata);
 		dev_err(glue->dev, "missing wlan platform data: %d\n", ret);
 		goto out_free_glue;
 	}
@@ -260,7 +257,7 @@ static int wl1271_probe(struct sdio_func *func,
 	dev_dbg(glue->dev, "sdio PM caps = 0x%x\n", mmcflags);
 
 	if (mmcflags & MMC_PM_KEEP_POWER)
-		pdev_data->pdata->pwr_in_suspend = true;
+		pdev_data.pdata->pwr_in_suspend = true;
 
 	sdio_set_drvdata(func, glue);
 
@@ -289,7 +286,7 @@ static int wl1271_probe(struct sdio_func *func,
 
 	memset(res, 0x00, sizeof(res));
 
-	res[0].start = pdev_data->pdata->irq;
+	res[0].start = pdev_data.pdata->irq;
 	res[0].flags = IORESOURCE_IRQ;
 	res[0].name = "irq";
 
@@ -299,8 +296,8 @@ static int wl1271_probe(struct sdio_func *func,
 		goto out_dev_put;
 	}
 
-	ret = platform_device_add_data(glue->core, pdev_data,
-				       sizeof(*pdev_data));
+	ret = platform_device_add_data(glue->core, &pdev_data,
+				       sizeof(pdev_data));
 	if (ret) {
 		dev_err(glue->dev, "can't add platform data\n");
 		goto out_dev_put;
@@ -319,9 +316,6 @@ out_dev_put:
 out_free_glue:
 	kfree(glue);
 
-out_free_pdev_data:
-	kfree(pdev_data);
-
 out:
 	return ret;
 }
diff --git a/drivers/net/wireless/ti/wlcore/spi.c b/drivers/net/wireless/ti/wlcore/spi.c
index dbe826dd7c23..392c882b28f0 100644
--- a/drivers/net/wireless/ti/wlcore/spi.c
+++ b/drivers/net/wireless/ti/wlcore/spi.c
@@ -24,11 +24,12 @@
 #include <linux/interrupt.h>
 #include <linux/irq.h>
 #include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/swab.h>
 #include <linux/crc7.h>
 #include <linux/spi/spi.h>
 #include <linux/wl12xx.h>
 #include <linux/platform_device.h>
-#include <linux/slab.h>
 
 #include "wlcore.h"
 #include "wl12xx_80211.h"
@@ -110,18 +111,16 @@ static void wl12xx_spi_reset(struct device *child)
 static void wl12xx_spi_init(struct device *child)
 {
 	struct wl12xx_spi_glue *glue = dev_get_drvdata(child->parent);
-	u8 crc[WSPI_INIT_CMD_CRC_LEN], *cmd;
 	struct spi_transfer t;
 	struct spi_message m;
+	u8 *cmd = kzalloc(WSPI_INIT_CMD_LEN, GFP_KERNEL);
 
-	cmd = kzalloc(WSPI_INIT_CMD_LEN, GFP_KERNEL);
 	if (!cmd) {
 		dev_err(child->parent,
 			"could not allocate cmd for spi init\n");
 		return;
 	}
 
-	memset(crc, 0, sizeof(crc));
 	memset(&t, 0, sizeof(t));
 	spi_message_init(&m);
 
@@ -129,30 +128,29 @@ static void wl12xx_spi_init(struct device *child)
 	 * Set WSPI_INIT_COMMAND
 	 * the data is being send from the MSB to LSB
 	 */
-	cmd[2] = 0xff;
-	cmd[3] = 0xff;
-	cmd[1] = WSPI_INIT_CMD_START | WSPI_INIT_CMD_TX;
-	cmd[0] = 0;
-	cmd[7] = 0;
-	cmd[6] |= HW_ACCESS_WSPI_INIT_CMD_MASK << 3;
-	cmd[6] |= HW_ACCESS_WSPI_FIXED_BUSY_LEN & WSPI_INIT_CMD_FIXEDBUSY_LEN;
+	cmd[0] = 0xff;
+	cmd[1] = 0xff;
+	cmd[2] = WSPI_INIT_CMD_START | WSPI_INIT_CMD_TX;
+	cmd[3] = 0;
+	cmd[4] = 0;
+	cmd[5] = HW_ACCESS_WSPI_INIT_CMD_MASK << 3;
+	cmd[5] |= HW_ACCESS_WSPI_FIXED_BUSY_LEN & WSPI_INIT_CMD_FIXEDBUSY_LEN;
+
+	cmd[6] = WSPI_INIT_CMD_IOD | WSPI_INIT_CMD_IP | WSPI_INIT_CMD_CS
+		| WSPI_INIT_CMD_WSPI | WSPI_INIT_CMD_WS;
 
 	if (HW_ACCESS_WSPI_FIXED_BUSY_LEN == 0)
-		cmd[5] |=  WSPI_INIT_CMD_DIS_FIXEDBUSY;
+		cmd[6] |= WSPI_INIT_CMD_DIS_FIXEDBUSY;
 	else
-		cmd[5] |= WSPI_INIT_CMD_EN_FIXEDBUSY;
-
-	cmd[5] |= WSPI_INIT_CMD_IOD | WSPI_INIT_CMD_IP | WSPI_INIT_CMD_CS
-		| WSPI_INIT_CMD_WSPI | WSPI_INIT_CMD_WS;
-
-	crc[0] = cmd[1];
-	crc[1] = cmd[0];
-	crc[2] = cmd[7];
-	crc[3] = cmd[6];
-	crc[4] = cmd[5];
+		cmd[6] |= WSPI_INIT_CMD_EN_FIXEDBUSY;
 
-	cmd[4] |= crc7(0, crc, WSPI_INIT_CMD_CRC_LEN) << 1;
-	cmd[4] |= WSPI_INIT_CMD_END;
+	cmd[7] = crc7_be(0, cmd+2, WSPI_INIT_CMD_CRC_LEN) | WSPI_INIT_CMD_END;
+	/*
+	 * The above is the logical order; it must actually be stored
+	 * in the buffer byte-swapped.
+	 */
+	__swab32s((u32 *)cmd);
+	__swab32s((u32 *)cmd+1);
 
 	t.tx_buf = cmd;
 	t.len = WSPI_INIT_CMD_LEN;
@@ -327,27 +325,25 @@ static struct wl1271_if_operations spi_ops = {
 static int wl1271_probe(struct spi_device *spi)
 {
 	struct wl12xx_spi_glue *glue;
-	struct wlcore_platdev_data *pdev_data;
+	struct wlcore_platdev_data pdev_data;
 	struct resource res[1];
 	int ret = -ENOMEM;
 
-	pdev_data = kzalloc(sizeof(*pdev_data), GFP_KERNEL);
-	if (!pdev_data)
-		goto out;
+	memset(&pdev_data, 0x00, sizeof(pdev_data));
 
-	pdev_data->pdata = dev_get_platdata(&spi->dev);
-	if (!pdev_data->pdata) {
+	pdev_data.pdata = dev_get_platdata(&spi->dev);
+	if (!pdev_data.pdata) {
 		dev_err(&spi->dev, "no platform data\n");
 		ret = -ENODEV;
-		goto out_free_pdev_data;
+		goto out;
 	}
 
-	pdev_data->if_ops = &spi_ops;
+	pdev_data.if_ops = &spi_ops;
 
 	glue = kzalloc(sizeof(*glue), GFP_KERNEL);
 	if (!glue) {
 		dev_err(&spi->dev, "can't allocate glue\n");
-		goto out_free_pdev_data;
+		goto out;
 	}
 
 	glue->dev = &spi->dev;
@@ -385,8 +381,8 @@ static int wl1271_probe(struct spi_device *spi)
 		goto out_dev_put;
 	}
 
-	ret = platform_device_add_data(glue->core, pdev_data,
-				       sizeof(*pdev_data));
+	ret = platform_device_add_data(glue->core, &pdev_data,
+				       sizeof(pdev_data));
 	if (ret) {
 		dev_err(glue->dev, "can't add platform data\n");
 		goto out_dev_put;
@@ -406,9 +402,6 @@ out_dev_put:
 out_free_glue:
 	kfree(glue);
 
-out_free_pdev_data:
-	kfree(pdev_data);
-
 out:
 	return ret;
 }
diff --git a/drivers/net/wireless/ti/wlcore/wlcore_i.h b/drivers/net/wireless/ti/wlcore/wlcore_i.h
index 756e890bc5ee..c2c34a84ff3d 100644
--- a/drivers/net/wireless/ti/wlcore/wlcore_i.h
+++ b/drivers/net/wireless/ti/wlcore/wlcore_i.h
@@ -512,8 +512,8 @@ int wl1271_recalc_rx_streaming(struct wl1271 *wl, struct wl12xx_vif *wlvif);
 void wl12xx_queue_recovery_work(struct wl1271 *wl);
 size_t wl12xx_copy_fwlog(struct wl1271 *wl, u8 *memblock, size_t maxlen);
 int wl1271_rx_filter_alloc_field(struct wl12xx_rx_filter *filter,
-					u16 offset, u8 flags,
-					u8 *pattern, u8 len);
+				 u16 offset, u8 flags,
+				 const u8 *pattern, u8 len);
 void wl1271_rx_filter_free(struct wl12xx_rx_filter *filter);
 struct wl12xx_rx_filter *wl1271_rx_filter_alloc(void);
 int wl1271_rx_filter_get_fields_size(struct wl12xx_rx_filter *filter);
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 0d4a285cbd7e..2532ce85d718 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -99,22 +99,43 @@ struct xenvif_rx_meta {
  */
 #define XEN_NETBK_LEGACY_SLOTS_MAX XEN_NETIF_NR_SLOTS_MIN
 
-struct xenvif {
-	/* Unique identifier for this interface. */
-	domid_t          domid;
-	unsigned int     handle;
+/* Queue name is interface name with "-qNNN" appended */
+#define QUEUE_NAME_SIZE (IFNAMSIZ + 5)
 
-	/* Is this interface disabled? True when backend discovers
-	 * frontend is rogue.
+/* IRQ name is queue name with "-tx" or "-rx" appended */
+#define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3)
+
+struct xenvif;
+
+struct xenvif_stats {
+	/* Stats fields to be updated per-queue.
+	 * A subset of struct net_device_stats that contains only the
+	 * fields that are updated in netback.c for each queue.
 	 */
-	bool disabled;
+	unsigned int rx_bytes;
+	unsigned int rx_packets;
+	unsigned int tx_bytes;
+	unsigned int tx_packets;
+
+	/* Additional stats used by xenvif */
+	unsigned long rx_gso_checksum_fixup;
+	unsigned long tx_zerocopy_sent;
+	unsigned long tx_zerocopy_success;
+	unsigned long tx_zerocopy_fail;
+	unsigned long tx_frag_overflow;
+};
+
+struct xenvif_queue { /* Per-queue data for xenvif */
+	unsigned int id; /* Queue ID, 0-based */
+	char name[QUEUE_NAME_SIZE]; /* DEVNAME-qN */
+	struct xenvif *vif; /* Parent VIF */
 
 	/* Use NAPI for guest TX */
 	struct napi_struct napi;
 	/* When feature-split-event-channels = 0, tx_irq = rx_irq. */
 	unsigned int tx_irq;
 	/* Only used when feature-split-event-channels = 1 */
-	char tx_irq_name[IFNAMSIZ+4]; /* DEVNAME-tx */
+	char tx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-tx */
 	struct xen_netif_tx_back_ring tx;
 	struct sk_buff_head tx_queue;
 	struct page *mmap_pages[MAX_PENDING_REQS];
@@ -150,7 +171,7 @@ struct xenvif {
 	/* When feature-split-event-channels = 0, tx_irq = rx_irq. */
 	unsigned int rx_irq;
 	/* Only used when feature-split-event-channels = 1 */
-	char rx_irq_name[IFNAMSIZ+4]; /* DEVNAME-rx */
+	char rx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-rx */
 	struct xen_netif_rx_back_ring rx;
 	struct sk_buff_head rx_queue;
 	RING_IDX rx_last_skb_slots;
@@ -158,14 +179,29 @@ struct xenvif {
 
 	struct timer_list wake_queue;
 
-	/* This array is allocated seperately as it is large */
-	struct gnttab_copy *grant_copy_op;
+	struct gnttab_copy grant_copy_op[MAX_GRANT_COPY_OPS];
 
 	/* We create one meta structure per ring request we consume, so
 	 * the maximum number is the same as the ring size.
 	 */
 	struct xenvif_rx_meta meta[XEN_NETIF_RX_RING_SIZE];
 
+	/* Transmit shaping: allow 'credit_bytes' every 'credit_usec'. */
+	unsigned long   credit_bytes;
+	unsigned long   credit_usec;
+	unsigned long   remaining_credit;
+	struct timer_list credit_timeout;
+	u64 credit_window_start;
+
+	/* Statistics */
+	struct xenvif_stats stats;
+};
+
+struct xenvif {
+	/* Unique identifier for this interface. */
+	domid_t          domid;
+	unsigned int     handle;
+
 	u8               fe_dev_addr[6];
 
 	/* Frontend feature information. */
@@ -179,19 +215,14 @@ struct xenvif {
 	/* Internal feature information. */
 	u8 can_queue:1;	    /* can queue packets for receiver? */
 
-	/* Transmit shaping: allow 'credit_bytes' every 'credit_usec'. */
-	unsigned long   credit_bytes;
-	unsigned long   credit_usec;
-	unsigned long   remaining_credit;
-	struct timer_list credit_timeout;
-	u64 credit_window_start;
+	/* Is this interface disabled? True when backend discovers
+	 * frontend is rogue.
+	 */
+	bool disabled;
 
-	/* Statistics */
-	unsigned long rx_gso_checksum_fixup;
-	unsigned long tx_zerocopy_sent;
-	unsigned long tx_zerocopy_success;
-	unsigned long tx_zerocopy_fail;
-	unsigned long tx_frag_overflow;
+	/* Queues */
+	struct xenvif_queue *queues;
+	unsigned int num_queues; /* active queues, resource allocated */
 
 	/* Miscellaneous private stuff. */
 	struct net_device *dev;
@@ -206,7 +237,10 @@ struct xenvif *xenvif_alloc(struct device *parent,
 			    domid_t domid,
 			    unsigned int handle);
 
-int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
+int xenvif_init_queue(struct xenvif_queue *queue);
+void xenvif_deinit_queue(struct xenvif_queue *queue);
+
+int xenvif_connect(struct xenvif_queue *queue, unsigned long tx_ring_ref,
 		   unsigned long rx_ring_ref, unsigned int tx_evtchn,
 		   unsigned int rx_evtchn);
 void xenvif_disconnect(struct xenvif *vif);
@@ -217,44 +251,47 @@ void xenvif_xenbus_fini(void);
 
 int xenvif_schedulable(struct xenvif *vif);
 
-int xenvif_must_stop_queue(struct xenvif *vif);
+int xenvif_must_stop_queue(struct xenvif_queue *queue);
+
+int xenvif_queue_stopped(struct xenvif_queue *queue);
+void xenvif_wake_queue(struct xenvif_queue *queue);
 
 /* (Un)Map communication rings. */
-void xenvif_unmap_frontend_rings(struct xenvif *vif);
-int xenvif_map_frontend_rings(struct xenvif *vif,
+void xenvif_unmap_frontend_rings(struct xenvif_queue *queue);
+int xenvif_map_frontend_rings(struct xenvif_queue *queue,
 			      grant_ref_t tx_ring_ref,
 			      grant_ref_t rx_ring_ref);
 
 /* Check for SKBs from frontend and schedule backend processing */
-void xenvif_napi_schedule_or_enable_events(struct xenvif *vif);
+void xenvif_napi_schedule_or_enable_events(struct xenvif_queue *queue);
 
 /* Prevent the device from generating any further traffic. */
 void xenvif_carrier_off(struct xenvif *vif);
 
-int xenvif_tx_action(struct xenvif *vif, int budget);
+int xenvif_tx_action(struct xenvif_queue *queue, int budget);
 
 int xenvif_kthread_guest_rx(void *data);
-void xenvif_kick_thread(struct xenvif *vif);
+void xenvif_kick_thread(struct xenvif_queue *queue);
 
 int xenvif_dealloc_kthread(void *data);
 
 /* Determine whether the needed number of slots (req) are available,
  * and set req_event if not.
  */
-bool xenvif_rx_ring_slots_available(struct xenvif *vif, int needed);
+bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue, int needed);
 
-void xenvif_stop_queue(struct xenvif *vif);
+void xenvif_carrier_on(struct xenvif *vif);
 
 /* Callback from stack when TX packet can be released */
 void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success);
 
 /* Unmap a pending page and release it back to the guest */
-void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx);
+void xenvif_idx_unmap(struct xenvif_queue *queue, u16 pending_idx);
 
-static inline pending_ring_idx_t nr_pending_reqs(struct xenvif *vif)
+static inline pending_ring_idx_t nr_pending_reqs(struct xenvif_queue *queue)
 {
 	return MAX_PENDING_REQS -
-		vif->pending_prod + vif->pending_cons;
+		queue->pending_prod + queue->pending_cons;
 }
 
 /* Callback from stack when TX packet can be released */
@@ -264,5 +301,6 @@ extern bool separate_tx_rx_irq;
 
 extern unsigned int rx_drain_timeout_msecs;
 extern unsigned int rx_drain_timeout_jiffies;
+extern unsigned int xenvif_max_queues;
 
 #endif /* __XEN_NETBACK__COMMON_H__ */
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index 20e9defa1060..9e97c7ca0ddd 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -43,6 +43,16 @@
 #define XENVIF_QUEUE_LENGTH 32
 #define XENVIF_NAPI_WEIGHT  64
 
+static inline void xenvif_stop_queue(struct xenvif_queue *queue)
+{
+	struct net_device *dev = queue->vif->dev;
+
+	if (!queue->vif->can_queue)
+		return;
+
+	netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id));
+}
+
 int xenvif_schedulable(struct xenvif *vif)
 {
 	return netif_running(vif->dev) && netif_carrier_ok(vif->dev);
@@ -50,33 +60,34 @@ int xenvif_schedulable(struct xenvif *vif)
 
 static irqreturn_t xenvif_tx_interrupt(int irq, void *dev_id)
 {
-	struct xenvif *vif = dev_id;
+	struct xenvif_queue *queue = dev_id;
 
-	if (RING_HAS_UNCONSUMED_REQUESTS(&vif->tx))
-		napi_schedule(&vif->napi);
+	if (RING_HAS_UNCONSUMED_REQUESTS(&queue->tx))
+		napi_schedule(&queue->napi);
 
 	return IRQ_HANDLED;
 }
 
-static int xenvif_poll(struct napi_struct *napi, int budget)
+int xenvif_poll(struct napi_struct *napi, int budget)
 {
-	struct xenvif *vif = container_of(napi, struct xenvif, napi);
+	struct xenvif_queue *queue =
+		container_of(napi, struct xenvif_queue, napi);
 	int work_done;
 
 	/* This vif is rogue, we pretend we've there is nothing to do
 	 * for this vif to deschedule it from NAPI. But this interface
 	 * will be turned off in thread context later.
 	 */
-	if (unlikely(vif->disabled)) {
+	if (unlikely(queue->vif->disabled)) {
 		napi_complete(napi);
 		return 0;
 	}
 
-	work_done = xenvif_tx_action(vif, budget);
+	work_done = xenvif_tx_action(queue, budget);
 
 	if (work_done < budget) {
 		napi_complete(napi);
-		xenvif_napi_schedule_or_enable_events(vif);
+		xenvif_napi_schedule_or_enable_events(queue);
 	}
 
 	return work_done;
@@ -84,9 +95,9 @@ static int xenvif_poll(struct napi_struct *napi, int budget)
 
 static irqreturn_t xenvif_rx_interrupt(int irq, void *dev_id)
 {
-	struct xenvif *vif = dev_id;
+	struct xenvif_queue *queue = dev_id;
 
-	xenvif_kick_thread(vif);
+	xenvif_kick_thread(queue);
 
 	return IRQ_HANDLED;
 }
@@ -99,28 +110,59 @@ static irqreturn_t xenvif_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void xenvif_wake_queue(unsigned long data)
+int xenvif_queue_stopped(struct xenvif_queue *queue)
+{
+	struct net_device *dev = queue->vif->dev;
+	unsigned int id = queue->id;
+	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, id));
+}
+
+void xenvif_wake_queue(struct xenvif_queue *queue)
+{
+	struct net_device *dev = queue->vif->dev;
+	unsigned int id = queue->id;
+	netif_tx_wake_queue(netdev_get_tx_queue(dev, id));
+}
+
+/* Callback to wake the queue and drain it on timeout */
+static void xenvif_wake_queue_callback(unsigned long data)
 {
-	struct xenvif *vif = (struct xenvif *)data;
+	struct xenvif_queue *queue = (struct xenvif_queue *)data;
 
-	if (netif_queue_stopped(vif->dev)) {
-		netdev_err(vif->dev, "draining TX queue\n");
-		vif->rx_queue_purge = true;
-		xenvif_kick_thread(vif);
-		netif_wake_queue(vif->dev);
+	if (xenvif_queue_stopped(queue)) {
+		netdev_err(queue->vif->dev, "draining TX queue\n");
+		queue->rx_queue_purge = true;
+		xenvif_kick_thread(queue);
+		xenvif_wake_queue(queue);
 	}
 }
 
 static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct xenvif *vif = netdev_priv(dev);
+	struct xenvif_queue *queue = NULL;
+	unsigned int num_queues = vif->num_queues;
+	u16 index;
 	int min_slots_needed;
 
 	BUG_ON(skb->dev != dev);
 
-	/* Drop the packet if vif is not ready */
-	if (vif->task == NULL ||
-	    vif->dealloc_task == NULL ||
+	/* Drop the packet if queues are not set up */
+	if (num_queues < 1)
+		goto drop;
+
+	/* Obtain the queue to be used to transmit this packet */
+	index = skb_get_queue_mapping(skb);
+	if (index >= num_queues) {
+		pr_warn_ratelimited("Invalid queue %hu for packet on interface %s\n.",
+				    index, vif->dev->name);
+		index %= num_queues;
+	}
+	queue = &vif->queues[index];
+
+	/* Drop the packet if queue is not ready */
+	if (queue->task == NULL ||
+	    queue->dealloc_task == NULL ||
 	    !xenvif_schedulable(vif))
 		goto drop;
 
@@ -139,16 +181,16 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 * then turn off the queue to give the ring a chance to
 	 * drain.
 	 */
-	if (!xenvif_rx_ring_slots_available(vif, min_slots_needed)) {
-		vif->wake_queue.function = xenvif_wake_queue;
-		vif->wake_queue.data = (unsigned long)vif;
-		xenvif_stop_queue(vif);
-		mod_timer(&vif->wake_queue,
+	if (!xenvif_rx_ring_slots_available(queue, min_slots_needed)) {
+		queue->wake_queue.function = xenvif_wake_queue_callback;
+		queue->wake_queue.data = (unsigned long)queue;
+		xenvif_stop_queue(queue);
+		mod_timer(&queue->wake_queue,
 			jiffies + rx_drain_timeout_jiffies);
 	}
 
-	skb_queue_tail(&vif->rx_queue, skb);
-	xenvif_kick_thread(vif);
+	skb_queue_tail(&queue->rx_queue, skb);
+	xenvif_kick_thread(queue);
 
 	return NETDEV_TX_OK;
 
@@ -161,25 +203,65 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
 static struct net_device_stats *xenvif_get_stats(struct net_device *dev)
 {
 	struct xenvif *vif = netdev_priv(dev);
+	struct xenvif_queue *queue = NULL;
+	unsigned int num_queues = vif->num_queues;
+	unsigned long rx_bytes = 0;
+	unsigned long rx_packets = 0;
+	unsigned long tx_bytes = 0;
+	unsigned long tx_packets = 0;
+	unsigned int index;
+
+	if (vif->queues == NULL)
+		goto out;
+
+	/* Aggregate tx and rx stats from each queue */
+	for (index = 0; index < num_queues; ++index) {
+		queue = &vif->queues[index];
+		rx_bytes += queue->stats.rx_bytes;
+		rx_packets += queue->stats.rx_packets;
+		tx_bytes += queue->stats.tx_bytes;
+		tx_packets += queue->stats.tx_packets;
+	}
+
+out:
+	vif->dev->stats.rx_bytes = rx_bytes;
+	vif->dev->stats.rx_packets = rx_packets;
+	vif->dev->stats.tx_bytes = tx_bytes;
+	vif->dev->stats.tx_packets = tx_packets;
+
 	return &vif->dev->stats;
 }
 
 static void xenvif_up(struct xenvif *vif)
 {
-	napi_enable(&vif->napi);
-	enable_irq(vif->tx_irq);
-	if (vif->tx_irq != vif->rx_irq)
-		enable_irq(vif->rx_irq);
-	xenvif_napi_schedule_or_enable_events(vif);
+	struct xenvif_queue *queue = NULL;
+	unsigned int num_queues = vif->num_queues;
+	unsigned int queue_index;
+
+	for (queue_index = 0; queue_index < num_queues; ++queue_index) {
+		queue = &vif->queues[queue_index];
+		napi_enable(&queue->napi);
+		enable_irq(queue->tx_irq);
+		if (queue->tx_irq != queue->rx_irq)
+			enable_irq(queue->rx_irq);
+		xenvif_napi_schedule_or_enable_events(queue);
+	}
 }
 
 static void xenvif_down(struct xenvif *vif)
 {
-	napi_disable(&vif->napi);
-	disable_irq(vif->tx_irq);
-	if (vif->tx_irq != vif->rx_irq)
-		disable_irq(vif->rx_irq);
-	del_timer_sync(&vif->credit_timeout);
+	struct xenvif_queue *queue = NULL;
+	unsigned int num_queues = vif->num_queues;
+	unsigned int queue_index;
+
+	for (queue_index = 0; queue_index < num_queues; ++queue_index) {
+		queue = &vif->queues[queue_index];
+		napi_disable(&queue->napi);
+		disable_irq(queue->tx_irq);
+		if (queue->tx_irq != queue->rx_irq)
+			disable_irq(queue->rx_irq);
+		del_timer_sync(&queue->credit_timeout);
+	}
 }
 
 static int xenvif_open(struct net_device *dev)
@@ -187,7 +269,7 @@ static int xenvif_open(struct net_device *dev)
 	struct xenvif *vif = netdev_priv(dev);
 	if (netif_carrier_ok(dev))
 		xenvif_up(vif);
-	netif_start_queue(dev);
+	netif_tx_start_all_queues(dev);
 	return 0;
 }
 
@@ -196,7 +278,7 @@ static int xenvif_close(struct net_device *dev)
 	struct xenvif *vif = netdev_priv(dev);
 	if (netif_carrier_ok(dev))
 		xenvif_down(vif);
-	netif_stop_queue(dev);
+	netif_tx_stop_all_queues(dev);
 	return 0;
 }
 
@@ -236,29 +318,29 @@ static const struct xenvif_stat {
 } xenvif_stats[] = {
 	{
 		"rx_gso_checksum_fixup",
-		offsetof(struct xenvif, rx_gso_checksum_fixup)
+		offsetof(struct xenvif_stats, rx_gso_checksum_fixup)
 	},
 	/* If (sent != success + fail), there are probably packets never
 	 * freed up properly!
 	 */
 	{
 		"tx_zerocopy_sent",
-		offsetof(struct xenvif, tx_zerocopy_sent),
+		offsetof(struct xenvif_stats, tx_zerocopy_sent),
 	},
 	{
 		"tx_zerocopy_success",
-		offsetof(struct xenvif, tx_zerocopy_success),
+		offsetof(struct xenvif_stats, tx_zerocopy_success),
 	},
 	{
 		"tx_zerocopy_fail",
-		offsetof(struct xenvif, tx_zerocopy_fail)
+		offsetof(struct xenvif_stats, tx_zerocopy_fail)
 	},
 	/* Number of packets exceeding MAX_SKB_FRAG slots. You should use
 	 * a guest with the same MAX_SKB_FRAG
 	 */
 	{
 		"tx_frag_overflow",
-		offsetof(struct xenvif, tx_frag_overflow)
+		offsetof(struct xenvif_stats, tx_frag_overflow)
 	},
 };
 
@@ -275,11 +357,20 @@ static int xenvif_get_sset_count(struct net_device *dev, int string_set)
 static void xenvif_get_ethtool_stats(struct net_device *dev,
 				     struct ethtool_stats *stats, u64 * data)
 {
-	void *vif = netdev_priv(dev);
+	struct xenvif *vif = netdev_priv(dev);
+	unsigned int num_queues = vif->num_queues;
 	int i;
-
-	for (i = 0; i < ARRAY_SIZE(xenvif_stats); i++)
-		data[i] = *(unsigned long *)(vif + xenvif_stats[i].offset);
+	unsigned int queue_index;
+	struct xenvif_stats *vif_stats;
+
+	for (i = 0; i < ARRAY_SIZE(xenvif_stats); i++) {
+		unsigned long accum = 0;
+		for (queue_index = 0; queue_index < num_queues; ++queue_index) {
+			vif_stats = &vif->queues[queue_index].stats;
+			accum += *(unsigned long *)(vif_stats + xenvif_stats[i].offset);
+		}
+		data[i] = accum;
+	}
 }
 
 static void xenvif_get_strings(struct net_device *dev, u32 stringset, u8 * data)
@@ -321,10 +412,14 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
 	struct net_device *dev;
 	struct xenvif *vif;
 	char name[IFNAMSIZ] = {};
-	int i;
 
 	snprintf(name, IFNAMSIZ - 1, "vif%u.%u", domid, handle);
-	dev = alloc_netdev(sizeof(struct xenvif), name, ether_setup);
+	/* Allocate a netdev with the max. supported number of queues.
+	 * When the guest selects the desired number, it will be updated
+	 * via netif_set_real_num_*_queues().
+	 */
+	dev = alloc_netdev_mq(sizeof(struct xenvif), name, ether_setup,
+			      xenvif_max_queues);
 	if (dev == NULL) {
 		pr_warn("Could not allocate netdev for %s\n", name);
 		return ERR_PTR(-ENOMEM);
@@ -334,66 +429,26 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
 
 	vif = netdev_priv(dev);
 
-	vif->grant_copy_op = vmalloc(sizeof(struct gnttab_copy) *
-				     MAX_GRANT_COPY_OPS);
-	if (vif->grant_copy_op == NULL) {
-		pr_warn("Could not allocate grant copy space for %s\n", name);
-		free_netdev(dev);
-		return ERR_PTR(-ENOMEM);
-	}
-
 	vif->domid  = domid;
 	vif->handle = handle;
 	vif->can_sg = 1;
 	vif->ip_csum = 1;
 	vif->dev = dev;
-
 	vif->disabled = false;
 
-	vif->credit_bytes = vif->remaining_credit = ~0UL;
-	vif->credit_usec  = 0UL;
-	init_timer(&vif->credit_timeout);
-	vif->credit_window_start = get_jiffies_64();
-
-	init_timer(&vif->wake_queue);
+	/* Start out with no queues. */
+	vif->queues = NULL;
+	vif->num_queues = 0;
 
 	dev->netdev_ops	= &xenvif_netdev_ops;
 	dev->hw_features = NETIF_F_SG |
 		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 		NETIF_F_TSO | NETIF_F_TSO6;
 	dev->features = dev->hw_features | NETIF_F_RXCSUM;
-	SET_ETHTOOL_OPS(dev, &xenvif_ethtool_ops);
+	dev->ethtool_ops = &xenvif_ethtool_ops;
 
 	dev->tx_queue_len = XENVIF_QUEUE_LENGTH;
 
-	skb_queue_head_init(&vif->rx_queue);
-	skb_queue_head_init(&vif->tx_queue);
-
-	vif->pending_cons = 0;
-	vif->pending_prod = MAX_PENDING_REQS;
-	for (i = 0; i < MAX_PENDING_REQS; i++)
-		vif->pending_ring[i] = i;
-	spin_lock_init(&vif->callback_lock);
-	spin_lock_init(&vif->response_lock);
-	/* If ballooning is disabled, this will consume real memory, so you
-	 * better enable it. The long term solution would be to use just a
-	 * bunch of valid page descriptors, without dependency on ballooning
-	 */
-	err = alloc_xenballooned_pages(MAX_PENDING_REQS,
-				       vif->mmap_pages,
-				       false);
-	if (err) {
-		netdev_err(dev, "Could not reserve mmap_pages\n");
-		return ERR_PTR(-ENOMEM);
-	}
-	for (i = 0; i < MAX_PENDING_REQS; i++) {
-		vif->pending_tx_info[i].callback_struct = (struct ubuf_info)
-			{ .callback = xenvif_zerocopy_callback,
-			  .ctx = NULL,
-			  .desc = i };
-		vif->grant_tx_handle[i] = NETBACK_INVALID_HANDLE;
-	}
-
 	/*
 	 * Initialise a dummy MAC address. We choose the numerically
 	 * largest non-broadcast address to prevent the address getting
@@ -403,8 +458,6 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
 	memset(dev->dev_addr, 0xFF, ETH_ALEN);
 	dev->dev_addr[0] &= ~0x01;
 
-	netif_napi_add(dev, &vif->napi, xenvif_poll, XENVIF_NAPI_WEIGHT);
-
 	netif_carrier_off(dev);
 
 	err = register_netdev(dev);
@@ -421,98 +474,147 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
 	return vif;
 }
 
-int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
+int xenvif_init_queue(struct xenvif_queue *queue)
+{
+	int err, i;
+
+	queue->credit_bytes = queue->remaining_credit = ~0UL;
+	queue->credit_usec  = 0UL;
+	init_timer(&queue->credit_timeout);
+	queue->credit_window_start = get_jiffies_64();
+
+	skb_queue_head_init(&queue->rx_queue);
+	skb_queue_head_init(&queue->tx_queue);
+
+	queue->pending_cons = 0;
+	queue->pending_prod = MAX_PENDING_REQS;
+	for (i = 0; i < MAX_PENDING_REQS; ++i)
+		queue->pending_ring[i] = i;
+
+	spin_lock_init(&queue->callback_lock);
+	spin_lock_init(&queue->response_lock);
+
+	/* If ballooning is disabled, this will consume real memory, so you
+	 * better enable it. The long term solution would be to use just a
+	 * bunch of valid page descriptors, without dependency on ballooning
+	 */
+	err = alloc_xenballooned_pages(MAX_PENDING_REQS,
+				       queue->mmap_pages,
+				       false);
+	if (err) {
+		netdev_err(queue->vif->dev, "Could not reserve mmap_pages\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < MAX_PENDING_REQS; i++) {
+		queue->pending_tx_info[i].callback_struct = (struct ubuf_info)
+			{ .callback = xenvif_zerocopy_callback,
+			  .ctx = NULL,
+			  .desc = i };
+		queue->grant_tx_handle[i] = NETBACK_INVALID_HANDLE;
+	}
+
+	init_timer(&queue->wake_queue);
+
+	netif_napi_add(queue->vif->dev, &queue->napi, xenvif_poll,
+			XENVIF_NAPI_WEIGHT);
+
+	return 0;
+}
+
+void xenvif_carrier_on(struct xenvif *vif)
+{
+	rtnl_lock();
+	if (!vif->can_sg && vif->dev->mtu > ETH_DATA_LEN)
+		dev_set_mtu(vif->dev, ETH_DATA_LEN);
+	netdev_update_features(vif->dev);
+	netif_carrier_on(vif->dev);
+	if (netif_running(vif->dev))
+		xenvif_up(vif);
+	rtnl_unlock();
+}
+
+int xenvif_connect(struct xenvif_queue *queue, unsigned long tx_ring_ref,
 		   unsigned long rx_ring_ref, unsigned int tx_evtchn,
 		   unsigned int rx_evtchn)
 {
 	struct task_struct *task;
 	int err = -ENOMEM;
 
-	BUG_ON(vif->tx_irq);
-	BUG_ON(vif->task);
-	BUG_ON(vif->dealloc_task);
+	BUG_ON(queue->tx_irq);
+	BUG_ON(queue->task);
+	BUG_ON(queue->dealloc_task);
 
-	err = xenvif_map_frontend_rings(vif, tx_ring_ref, rx_ring_ref);
+	err = xenvif_map_frontend_rings(queue, tx_ring_ref, rx_ring_ref);
 	if (err < 0)
 		goto err;
 
-	init_waitqueue_head(&vif->wq);
-	init_waitqueue_head(&vif->dealloc_wq);
+	init_waitqueue_head(&queue->wq);
+	init_waitqueue_head(&queue->dealloc_wq);
 
 	if (tx_evtchn == rx_evtchn) {
 		/* feature-split-event-channels == 0 */
 		err = bind_interdomain_evtchn_to_irqhandler(
-			vif->domid, tx_evtchn, xenvif_interrupt, 0,
-			vif->dev->name, vif);
+			queue->vif->domid, tx_evtchn, xenvif_interrupt, 0,
+			queue->name, queue);
 		if (err < 0)
 			goto err_unmap;
-		vif->tx_irq = vif->rx_irq = err;
-		disable_irq(vif->tx_irq);
+		queue->tx_irq = queue->rx_irq = err;
+		disable_irq(queue->tx_irq);
 	} else {
 		/* feature-split-event-channels == 1 */
-		snprintf(vif->tx_irq_name, sizeof(vif->tx_irq_name),
-			 "%s-tx", vif->dev->name);
+		snprintf(queue->tx_irq_name, sizeof(queue->tx_irq_name),
+			 "%s-tx", queue->name);
 		err = bind_interdomain_evtchn_to_irqhandler(
-			vif->domid, tx_evtchn, xenvif_tx_interrupt, 0,
-			vif->tx_irq_name, vif);
+			queue->vif->domid, tx_evtchn, xenvif_tx_interrupt, 0,
+			queue->tx_irq_name, queue);
 		if (err < 0)
 			goto err_unmap;
-		vif->tx_irq = err;
-		disable_irq(vif->tx_irq);
+		queue->tx_irq = err;
+		disable_irq(queue->tx_irq);
 
-		snprintf(vif->rx_irq_name, sizeof(vif->rx_irq_name),
-			 "%s-rx", vif->dev->name);
+		snprintf(queue->rx_irq_name, sizeof(queue->rx_irq_name),
+			 "%s-rx", queue->name);
 		err = bind_interdomain_evtchn_to_irqhandler(
-			vif->domid, rx_evtchn, xenvif_rx_interrupt, 0,
-			vif->rx_irq_name, vif);
+			queue->vif->domid, rx_evtchn, xenvif_rx_interrupt, 0,
+			queue->rx_irq_name, queue);
 		if (err < 0)
 			goto err_tx_unbind;
-		vif->rx_irq = err;
-		disable_irq(vif->rx_irq);
+		queue->rx_irq = err;
+		disable_irq(queue->rx_irq);
 	}
 
 	task = kthread_create(xenvif_kthread_guest_rx,
-			      (void *)vif, "%s-guest-rx", vif->dev->name);
+			      (void *)queue, "%s-guest-rx", queue->name);
 	if (IS_ERR(task)) {
-		pr_warn("Could not allocate kthread for %s\n", vif->dev->name);
+		pr_warn("Could not allocate kthread for %s\n", queue->name);
 		err = PTR_ERR(task);
 		goto err_rx_unbind;
 	}
-
-	vif->task = task;
+	queue->task = task;
 
 	task = kthread_create(xenvif_dealloc_kthread,
-			      (void *)vif, "%s-dealloc", vif->dev->name);
+			      (void *)queue, "%s-dealloc", queue->name);
 	if (IS_ERR(task)) {
-		pr_warn("Could not allocate kthread for %s\n", vif->dev->name);
+		pr_warn("Could not allocate kthread for %s\n", queue->name);
 		err = PTR_ERR(task);
 		goto err_rx_unbind;
 	}
+	queue->dealloc_task = task;
 
-	vif->dealloc_task = task;
-
-	rtnl_lock();
-	if (!vif->can_sg && vif->dev->mtu > ETH_DATA_LEN)
-		dev_set_mtu(vif->dev, ETH_DATA_LEN);
-	netdev_update_features(vif->dev);
-	netif_carrier_on(vif->dev);
-	if (netif_running(vif->dev))
-		xenvif_up(vif);
-	rtnl_unlock();
-
-	wake_up_process(vif->task);
-	wake_up_process(vif->dealloc_task);
+	wake_up_process(queue->task);
+	wake_up_process(queue->dealloc_task);
 
 	return 0;
 
 err_rx_unbind:
-	unbind_from_irqhandler(vif->rx_irq, vif);
-	vif->rx_irq = 0;
+	unbind_from_irqhandler(queue->rx_irq, queue);
+	queue->rx_irq = 0;
 err_tx_unbind:
-	unbind_from_irqhandler(vif->tx_irq, vif);
-	vif->tx_irq = 0;
+	unbind_from_irqhandler(queue->tx_irq, queue);
+	queue->tx_irq = 0;
 err_unmap:
-	xenvif_unmap_frontend_rings(vif);
+	xenvif_unmap_frontend_rings(queue);
 err:
 	module_put(THIS_MODULE);
 	return err;
@@ -529,38 +631,77 @@ void xenvif_carrier_off(struct xenvif *vif)
 	rtnl_unlock();
 }
 
+static void xenvif_wait_unmap_timeout(struct xenvif_queue *queue,
+				      unsigned int worst_case_skb_lifetime)
+{
+	int i, unmap_timeout = 0;
+
+	for (i = 0; i < MAX_PENDING_REQS; ++i) {
+		if (queue->grant_tx_handle[i] != NETBACK_INVALID_HANDLE) {
+			unmap_timeout++;
+			schedule_timeout(msecs_to_jiffies(1000));
+			if (unmap_timeout > worst_case_skb_lifetime &&
+			    net_ratelimit())
+				netdev_err(queue->vif->dev,
+					   "Page still granted! Index: %x\n",
+					   i);
+			i = -1;
+		}
+	}
+}
+
 void xenvif_disconnect(struct xenvif *vif)
 {
+	struct xenvif_queue *queue = NULL;
+	unsigned int num_queues = vif->num_queues;
+	unsigned int queue_index;
+
 	if (netif_carrier_ok(vif->dev))
 		xenvif_carrier_off(vif);
 
-	if (vif->task) {
-		del_timer_sync(&vif->wake_queue);
-		kthread_stop(vif->task);
-		vif->task = NULL;
-	}
+	for (queue_index = 0; queue_index < num_queues; ++queue_index) {
+		queue = &vif->queues[queue_index];
 
-	if (vif->dealloc_task) {
-		kthread_stop(vif->dealloc_task);
-		vif->dealloc_task = NULL;
-	}
+		if (queue->task) {
+			del_timer_sync(&queue->wake_queue);
+			kthread_stop(queue->task);
+			queue->task = NULL;
+		}
 
-	if (vif->tx_irq) {
-		if (vif->tx_irq == vif->rx_irq)
-			unbind_from_irqhandler(vif->tx_irq, vif);
-		else {
-			unbind_from_irqhandler(vif->tx_irq, vif);
-			unbind_from_irqhandler(vif->rx_irq, vif);
+		if (queue->dealloc_task) {
+			kthread_stop(queue->dealloc_task);
+			queue->dealloc_task = NULL;
 		}
-		vif->tx_irq = 0;
+
+		if (queue->tx_irq) {
+			if (queue->tx_irq == queue->rx_irq)
+				unbind_from_irqhandler(queue->tx_irq, queue);
+			else {
+				unbind_from_irqhandler(queue->tx_irq, queue);
+				unbind_from_irqhandler(queue->rx_irq, queue);
+			}
+			queue->tx_irq = 0;
+		}
+
+		xenvif_unmap_frontend_rings(queue);
 	}
+}
 
-	xenvif_unmap_frontend_rings(vif);
+/* Reverse the relevant parts of xenvif_init_queue().
+ * Used for queue teardown from xenvif_free(), and on the
+ * error handling paths in xenbus.c:connect().
+ */
+void xenvif_deinit_queue(struct xenvif_queue *queue)
+{
+	free_xenballooned_pages(MAX_PENDING_REQS, queue->mmap_pages);
+	netif_napi_del(&queue->napi);
 }
 
 void xenvif_free(struct xenvif *vif)
 {
-	int i, unmap_timeout = 0;
+	struct xenvif_queue *queue = NULL;
+	unsigned int num_queues = vif->num_queues;
+	unsigned int queue_index;
 	/* Here we want to avoid timeout messages if an skb can be legitimately
 	 * stuck somewhere else. Realistically this could be an another vif's
 	 * internal or QDisc queue. That another vif also has this
@@ -575,33 +716,18 @@ void xenvif_free(struct xenvif *vif)
 	unsigned int worst_case_skb_lifetime = (rx_drain_timeout_msecs/1000) *
 		DIV_ROUND_UP(XENVIF_QUEUE_LENGTH, (XEN_NETIF_RX_RING_SIZE / MAX_SKB_FRAGS));
 
-	for (i = 0; i < MAX_PENDING_REQS; ++i) {
-		if (vif->grant_tx_handle[i] != NETBACK_INVALID_HANDLE) {
-			unmap_timeout++;
-			schedule_timeout(msecs_to_jiffies(1000));
-			if (unmap_timeout > worst_case_skb_lifetime &&
-			    net_ratelimit())
-				netdev_err(vif->dev,
-					   "Page still granted! Index: %x\n",
-					   i);
-			/* If there are still unmapped pages, reset the loop to
-			 * start checking again. We shouldn't exit here until
-			 * dealloc thread and NAPI instance release all the
-			 * pages. If a kernel bug causes the skbs to stall
-			 * somewhere, the interface cannot be brought down
-			 * properly.
-			 */
-			i = -1;
-		}
-	}
-
-	free_xenballooned_pages(MAX_PENDING_REQS, vif->mmap_pages);
+	unregister_netdev(vif->dev);
 
-	netif_napi_del(&vif->napi);
+	for (queue_index = 0; queue_index < num_queues; ++queue_index) {
+		queue = &vif->queues[queue_index];
+		xenvif_wait_unmap_timeout(queue, worst_case_skb_lifetime);
+		xenvif_deinit_queue(queue);
+	}
 
-	unregister_netdev(vif->dev);
+	vfree(vif->queues);
+	vif->queues = NULL;
+	vif->num_queues = 0;
 
-	vfree(vif->grant_copy_op);
 	free_netdev(vif->dev);
 
 	module_put(THIS_MODULE);
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 7367208ee8cd..1844a47636b6 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -62,6 +62,11 @@ unsigned int rx_drain_timeout_msecs = 10000;
 module_param(rx_drain_timeout_msecs, uint, 0444);
 unsigned int rx_drain_timeout_jiffies;
 
+unsigned int xenvif_max_queues;
+module_param_named(max_queues, xenvif_max_queues, uint, 0644);
+MODULE_PARM_DESC(max_queues,
+		 "Maximum number of queues per virtual interface");
+
 /*
  * This is the maximum slots a skb can have. If a guest sends a skb
  * which exceeds this limit it is considered malicious.
@@ -70,33 +75,33 @@ unsigned int rx_drain_timeout_jiffies;
 static unsigned int fatal_skb_slots = FATAL_SKB_SLOTS_DEFAULT;
 module_param(fatal_skb_slots, uint, 0444);
 
-static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
+static void xenvif_idx_release(struct xenvif_queue *queue, u16 pending_idx,
 			       u8 status);
 
-static void make_tx_response(struct xenvif *vif,
+static void make_tx_response(struct xenvif_queue *queue,
 			     struct xen_netif_tx_request *txp,
 			     s8       st);
 
-static inline int tx_work_todo(struct xenvif *vif);
-static inline int rx_work_todo(struct xenvif *vif);
+static inline int tx_work_todo(struct xenvif_queue *queue);
+static inline int rx_work_todo(struct xenvif_queue *queue);
 
-static struct xen_netif_rx_response *make_rx_response(struct xenvif *vif,
+static struct xen_netif_rx_response *make_rx_response(struct xenvif_queue *queue,
 					     u16      id,
 					     s8       st,
 					     u16      offset,
 					     u16      size,
 					     u16      flags);
 
-static inline unsigned long idx_to_pfn(struct xenvif *vif,
+static inline unsigned long idx_to_pfn(struct xenvif_queue *queue,
 				       u16 idx)
 {
-	return page_to_pfn(vif->mmap_pages[idx]);
+	return page_to_pfn(queue->mmap_pages[idx]);
 }
 
-static inline unsigned long idx_to_kaddr(struct xenvif *vif,
+static inline unsigned long idx_to_kaddr(struct xenvif_queue *queue,
 					 u16 idx)
 {
-	return (unsigned long)pfn_to_kaddr(idx_to_pfn(vif, idx));
+	return (unsigned long)pfn_to_kaddr(idx_to_pfn(queue, idx));
 }
 
 #define callback_param(vif, pending_idx) \
@@ -104,13 +109,13 @@ static inline unsigned long idx_to_kaddr(struct xenvif *vif,
 
 /* Find the containing VIF's structure from a pointer in pending_tx_info array
  */
-static inline struct xenvif *ubuf_to_vif(const struct ubuf_info *ubuf)
+static inline struct xenvif_queue *ubuf_to_queue(const struct ubuf_info *ubuf)
 {
 	u16 pending_idx = ubuf->desc;
 	struct pending_tx_info *temp =
 		container_of(ubuf, struct pending_tx_info, callback_struct);
 	return container_of(temp - pending_idx,
-			    struct xenvif,
+			    struct xenvif_queue,
 			    pending_tx_info[0]);
 }
 
@@ -136,24 +141,24 @@ static inline pending_ring_idx_t pending_index(unsigned i)
 	return i & (MAX_PENDING_REQS-1);
 }
 
-bool xenvif_rx_ring_slots_available(struct xenvif *vif, int needed)
+bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue, int needed)
 {
 	RING_IDX prod, cons;
 
 	do {
-		prod = vif->rx.sring->req_prod;
-		cons = vif->rx.req_cons;
+		prod = queue->rx.sring->req_prod;
+		cons = queue->rx.req_cons;
 
 		if (prod - cons >= needed)
 			return true;
 
-		vif->rx.sring->req_event = prod + 1;
+		queue->rx.sring->req_event = prod + 1;
 
 		/* Make sure event is visible before we check prod
 		 * again.
 		 */
 		mb();
-	} while (vif->rx.sring->req_prod != prod);
+	} while (queue->rx.sring->req_prod != prod);
 
 	return false;
 }
@@ -163,7 +168,8 @@ bool xenvif_rx_ring_slots_available(struct xenvif *vif, int needed)
  * adding 'size' bytes to a buffer which currently contains 'offset'
  * bytes.
  */
-static bool start_new_rx_buffer(int offset, unsigned long size, int head)
+static bool start_new_rx_buffer(int offset, unsigned long size, int head,
+				bool full_coalesce)
 {
 	/* simple case: we have completely filled the current buffer. */
 	if (offset == MAX_BUFFER_OFFSET)
@@ -175,6 +181,7 @@ static bool start_new_rx_buffer(int offset, unsigned long size, int head)
 	 *     (i)   this frag would fit completely in the next buffer
 	 * and (ii)  there is already some data in the current buffer
 	 * and (iii) this is not the head buffer.
+	 * and (iv)  there is no need to fully utilize the buffers
 	 *
 	 * Where:
 	 * - (i) stops us splitting a frag into two copies
@@ -185,6 +192,8 @@ static bool start_new_rx_buffer(int offset, unsigned long size, int head)
 	 *   by (ii) but is explicitly checked because
 	 *   netfront relies on the first buffer being
 	 *   non-empty and can crash otherwise.
+	 * - (iv) is needed for skbs which can use up more than MAX_SKB_FRAGS
+	 *   slot
 	 *
 	 * This means we will effectively linearise small
 	 * frags but do not needlessly split large buffers
@@ -192,7 +201,8 @@ static bool start_new_rx_buffer(int offset, unsigned long size, int head)
 	 * own buffers as before.
 	 */
 	BUG_ON(size > MAX_BUFFER_OFFSET);
-	if ((offset + size > MAX_BUFFER_OFFSET) && offset && !head)
+	if ((offset + size > MAX_BUFFER_OFFSET) && offset && !head &&
+	    !full_coalesce)
 		return true;
 
 	return false;
@@ -207,13 +217,13 @@ struct netrx_pending_operations {
 	grant_ref_t copy_gref;
 };
 
-static struct xenvif_rx_meta *get_next_rx_buffer(struct xenvif *vif,
+static struct xenvif_rx_meta *get_next_rx_buffer(struct xenvif_queue *queue,
 						 struct netrx_pending_operations *npo)
 {
 	struct xenvif_rx_meta *meta;
 	struct xen_netif_rx_request *req;
 
-	req = RING_GET_REQUEST(&vif->rx, vif->rx.req_cons++);
+	req = RING_GET_REQUEST(&queue->rx, queue->rx.req_cons++);
 
 	meta = npo->meta + npo->meta_prod++;
 	meta->gso_type = XEN_NETIF_GSO_TYPE_NONE;
@@ -227,15 +237,22 @@ static struct xenvif_rx_meta *get_next_rx_buffer(struct xenvif *vif,
 	return meta;
 }
 
+struct xenvif_rx_cb {
+	int meta_slots_used;
+	bool full_coalesce;
+};
+
+#define XENVIF_RX_CB(skb) ((struct xenvif_rx_cb *)(skb)->cb)
+
 /*
  * Set up the grant operations for this fragment. If it's a flipping
  * interface, we also set up the unmap request from here.
  */
-static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
+static void xenvif_gop_frag_copy(struct xenvif_queue *queue, struct sk_buff *skb,
 				 struct netrx_pending_operations *npo,
 				 struct page *page, unsigned long size,
 				 unsigned long offset, int *head,
-				 struct xenvif *foreign_vif,
+				 struct xenvif_queue *foreign_queue,
 				 grant_ref_t foreign_gref)
 {
 	struct gnttab_copy *copy_gop;
@@ -261,14 +278,17 @@ static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
 		if (bytes > size)
 			bytes = size;
 
-		if (start_new_rx_buffer(npo->copy_off, bytes, *head)) {
+		if (start_new_rx_buffer(npo->copy_off,
+					bytes,
+					*head,
+					XENVIF_RX_CB(skb)->full_coalesce)) {
 			/*
 			 * Netfront requires there to be some data in the head
 			 * buffer.
 			 */
 			BUG_ON(*head);
 
-			meta = get_next_rx_buffer(vif, npo);
+			meta = get_next_rx_buffer(queue, npo);
 		}
 
 		if (npo->copy_off + bytes > MAX_BUFFER_OFFSET)
@@ -278,8 +298,8 @@ static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
 		copy_gop->flags = GNTCOPY_dest_gref;
 		copy_gop->len = bytes;
 
-		if (foreign_vif) {
-			copy_gop->source.domid = foreign_vif->domid;
+		if (foreign_queue) {
+			copy_gop->source.domid = foreign_queue->vif->domid;
 			copy_gop->source.u.ref = foreign_gref;
 			copy_gop->flags |= GNTCOPY_source_gref;
 		} else {
@@ -289,7 +309,7 @@ static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
 		}
 		copy_gop->source.offset = offset;
 
-		copy_gop->dest.domid = vif->domid;
+		copy_gop->dest.domid = queue->vif->domid;
 		copy_gop->dest.offset = npo->copy_off;
 		copy_gop->dest.u.ref = npo->copy_gref;
 
@@ -314,8 +334,8 @@ static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
 				gso_type = XEN_NETIF_GSO_TYPE_TCPV6;
 		}
 
-		if (*head && ((1 << gso_type) & vif->gso_mask))
-			vif->rx.req_cons++;
+		if (*head && ((1 << gso_type) & queue->vif->gso_mask))
+			queue->rx.req_cons++;
 
 		*head = 0; /* There must be something in this buffer now. */
 
@@ -337,13 +357,13 @@ static const struct ubuf_info *xenvif_find_gref(const struct sk_buff *const skb,
 						const int i,
 						const struct ubuf_info *ubuf)
 {
-	struct xenvif *foreign_vif = ubuf_to_vif(ubuf);
+	struct xenvif_queue *foreign_queue = ubuf_to_queue(ubuf);
 
 	do {
 		u16 pending_idx = ubuf->desc;
 
 		if (skb_shinfo(skb)->frags[i].page.p ==
-		    foreign_vif->mmap_pages[pending_idx])
+		    foreign_queue->mmap_pages[pending_idx])
 			break;
 		ubuf = (struct ubuf_info *) ubuf->ctx;
 	} while (ubuf);
@@ -364,7 +384,8 @@ static const struct ubuf_info *xenvif_find_gref(const struct sk_buff *const skb,
  * frontend-side LRO).
  */
 static int xenvif_gop_skb(struct sk_buff *skb,
-			  struct netrx_pending_operations *npo)
+			  struct netrx_pending_operations *npo,
+			  struct xenvif_queue *queue)
 {
 	struct xenvif *vif = netdev_priv(skb->dev);
 	int nr_frags = skb_shinfo(skb)->nr_frags;
@@ -390,7 +411,7 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 
 	/* Set up a GSO prefix descriptor, if necessary */
 	if ((1 << gso_type) & vif->gso_prefix_mask) {
-		req = RING_GET_REQUEST(&vif->rx, vif->rx.req_cons++);
+		req = RING_GET_REQUEST(&queue->rx, queue->rx.req_cons++);
 		meta = npo->meta + npo->meta_prod++;
 		meta->gso_type = gso_type;
 		meta->gso_size = skb_shinfo(skb)->gso_size;
@@ -398,7 +419,7 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 		meta->id = req->id;
 	}
 
-	req = RING_GET_REQUEST(&vif->rx, vif->rx.req_cons++);
+	req = RING_GET_REQUEST(&queue->rx, queue->rx.req_cons++);
 	meta = npo->meta + npo->meta_prod++;
 
 	if ((1 << gso_type) & vif->gso_mask) {
@@ -422,7 +443,7 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 		if (data + len > skb_tail_pointer(skb))
 			len = skb_tail_pointer(skb) - data;
 
-		xenvif_gop_frag_copy(vif, skb, npo,
+		xenvif_gop_frag_copy(queue, skb, npo,
 				     virt_to_page(data), len, offset, &head,
 				     NULL,
 				     0);
@@ -433,7 +454,7 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 		/* This variable also signals whether foreign_gref has a real
 		 * value or not.
 		 */
-		struct xenvif *foreign_vif = NULL;
+		struct xenvif_queue *foreign_queue = NULL;
 		grant_ref_t foreign_gref;
 
 		if ((skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) &&
@@ -458,8 +479,9 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 			if (likely(ubuf)) {
 				u16 pending_idx = ubuf->desc;
 
-				foreign_vif = ubuf_to_vif(ubuf);
-				foreign_gref = foreign_vif->pending_tx_info[pending_idx].req.gref;
+				foreign_queue = ubuf_to_queue(ubuf);
+				foreign_gref =
+					foreign_queue->pending_tx_info[pending_idx].req.gref;
 				/* Just a safety measure. If this was the last
 				 * element on the list, the for loop will
 				 * iterate again if a local page were added to
@@ -477,13 +499,13 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 				 */
 				ubuf = head_ubuf;
 		}
-		xenvif_gop_frag_copy(vif, skb, npo,
+		xenvif_gop_frag_copy(queue, skb, npo,
 				     skb_frag_page(&skb_shinfo(skb)->frags[i]),
 				     skb_frag_size(&skb_shinfo(skb)->frags[i]),
 				     skb_shinfo(skb)->frags[i].page_offset,
 				     &head,
-				     foreign_vif,
-				     foreign_vif ? foreign_gref : UINT_MAX);
+				     foreign_queue,
+				     foreign_queue ? foreign_gref : UINT_MAX);
 	}
 
 	return npo->meta_prod - old_meta_prod;
@@ -515,7 +537,7 @@ static int xenvif_check_gop(struct xenvif *vif, int nr_meta_slots,
 	return status;
 }
 
-static void xenvif_add_frag_responses(struct xenvif *vif, int status,
+static void xenvif_add_frag_responses(struct xenvif_queue *queue, int status,
 				      struct xenvif_rx_meta *meta,
 				      int nr_meta_slots)
 {
@@ -536,23 +558,17 @@ static void xenvif_add_frag_responses(struct xenvif *vif, int status,
 			flags = XEN_NETRXF_more_data;
 
 		offset = 0;
-		make_rx_response(vif, meta[i].id, status, offset,
+		make_rx_response(queue, meta[i].id, status, offset,
 				 meta[i].size, flags);
 	}
 }
 
-struct xenvif_rx_cb {
-	int meta_slots_used;
-};
-
-#define XENVIF_RX_CB(skb) ((struct xenvif_rx_cb *)(skb)->cb)
-
-void xenvif_kick_thread(struct xenvif *vif)
+void xenvif_kick_thread(struct xenvif_queue *queue)
 {
-	wake_up(&vif->wq);
+	wake_up(&queue->wq);
 }
 
-static void xenvif_rx_action(struct xenvif *vif)
+static void xenvif_rx_action(struct xenvif_queue *queue)
 {
 	s8 status;
 	u16 flags;
@@ -565,13 +581,13 @@ static void xenvif_rx_action(struct xenvif *vif)
 	bool need_to_notify = false;
 
 	struct netrx_pending_operations npo = {
-		.copy  = vif->grant_copy_op,
-		.meta  = vif->meta,
+		.copy  = queue->grant_copy_op,
+		.meta  = queue->meta,
 	};
 
 	skb_queue_head_init(&rxq);
 
-	while ((skb = skb_dequeue(&vif->rx_queue)) != NULL) {
+	while ((skb = skb_dequeue(&queue->rx_queue)) != NULL) {
 		RING_IDX max_slots_needed;
 		RING_IDX old_req_cons;
 		RING_IDX ring_slots_used;
@@ -602,10 +618,15 @@ static void xenvif_rx_action(struct xenvif *vif)
 
 		/* To avoid the estimate becoming too pessimal for some
 		 * frontends that limit posted rx requests, cap the estimate
-		 * at MAX_SKB_FRAGS.
+		 * at MAX_SKB_FRAGS. In this case netback will fully coalesce
+		 * the skb into the provided slots.
 		 */
-		if (max_slots_needed > MAX_SKB_FRAGS)
+		if (max_slots_needed > MAX_SKB_FRAGS) {
 			max_slots_needed = MAX_SKB_FRAGS;
+			XENVIF_RX_CB(skb)->full_coalesce = true;
+		} else {
+			XENVIF_RX_CB(skb)->full_coalesce = false;
+		}
 
 		/* We may need one more slot for GSO metadata */
 		if (skb_is_gso(skb) &&
@@ -614,42 +635,42 @@ static void xenvif_rx_action(struct xenvif *vif)
 			max_slots_needed++;
 
 		/* If the skb may not fit then bail out now */
-		if (!xenvif_rx_ring_slots_available(vif, max_slots_needed)) {
-			skb_queue_head(&vif->rx_queue, skb);
+		if (!xenvif_rx_ring_slots_available(queue, max_slots_needed)) {
+			skb_queue_head(&queue->rx_queue, skb);
 			need_to_notify = true;
-			vif->rx_last_skb_slots = max_slots_needed;
+			queue->rx_last_skb_slots = max_slots_needed;
 			break;
 		} else
-			vif->rx_last_skb_slots = 0;
+			queue->rx_last_skb_slots = 0;
 
-		old_req_cons = vif->rx.req_cons;
-		XENVIF_RX_CB(skb)->meta_slots_used = xenvif_gop_skb(skb, &npo);
-		ring_slots_used = vif->rx.req_cons - old_req_cons;
+		old_req_cons = queue->rx.req_cons;
+		XENVIF_RX_CB(skb)->meta_slots_used = xenvif_gop_skb(skb, &npo, queue);
+		ring_slots_used = queue->rx.req_cons - old_req_cons;
 
 		BUG_ON(ring_slots_used > max_slots_needed);
 
 		__skb_queue_tail(&rxq, skb);
 	}
 
-	BUG_ON(npo.meta_prod > ARRAY_SIZE(vif->meta));
+	BUG_ON(npo.meta_prod > ARRAY_SIZE(queue->meta));
 
 	if (!npo.copy_prod)
 		goto done;
 
 	BUG_ON(npo.copy_prod > MAX_GRANT_COPY_OPS);
-	gnttab_batch_copy(vif->grant_copy_op, npo.copy_prod);
+	gnttab_batch_copy(queue->grant_copy_op, npo.copy_prod);
 
 	while ((skb = __skb_dequeue(&rxq)) != NULL) {
 
-		if ((1 << vif->meta[npo.meta_cons].gso_type) &
-		    vif->gso_prefix_mask) {
-			resp = RING_GET_RESPONSE(&vif->rx,
-						 vif->rx.rsp_prod_pvt++);
+		if ((1 << queue->meta[npo.meta_cons].gso_type) &
+		    queue->vif->gso_prefix_mask) {
+			resp = RING_GET_RESPONSE(&queue->rx,
+						 queue->rx.rsp_prod_pvt++);
 
 			resp->flags = XEN_NETRXF_gso_prefix | XEN_NETRXF_more_data;
 
-			resp->offset = vif->meta[npo.meta_cons].gso_size;
-			resp->id = vif->meta[npo.meta_cons].id;
+			resp->offset = queue->meta[npo.meta_cons].gso_size;
+			resp->id = queue->meta[npo.meta_cons].id;
 			resp->status = XENVIF_RX_CB(skb)->meta_slots_used;
 
 			npo.meta_cons++;
@@ -657,10 +678,10 @@ static void xenvif_rx_action(struct xenvif *vif)
 		}
 
 
-		vif->dev->stats.tx_bytes += skb->len;
-		vif->dev->stats.tx_packets++;
+		queue->stats.tx_bytes += skb->len;
+		queue->stats.tx_packets++;
 
-		status = xenvif_check_gop(vif,
+		status = xenvif_check_gop(queue->vif,
 					  XENVIF_RX_CB(skb)->meta_slots_used,
 					  &npo);
 
@@ -676,22 +697,22 @@ static void xenvif_rx_action(struct xenvif *vif)
 			flags |= XEN_NETRXF_data_validated;
 
 		offset = 0;
-		resp = make_rx_response(vif, vif->meta[npo.meta_cons].id,
+		resp = make_rx_response(queue, queue->meta[npo.meta_cons].id,
 					status, offset,
-					vif->meta[npo.meta_cons].size,
+					queue->meta[npo.meta_cons].size,
 					flags);
 
-		if ((1 << vif->meta[npo.meta_cons].gso_type) &
-		    vif->gso_mask) {
+		if ((1 << queue->meta[npo.meta_cons].gso_type) &
+		    queue->vif->gso_mask) {
 			struct xen_netif_extra_info *gso =
 				(struct xen_netif_extra_info *)
-				RING_GET_RESPONSE(&vif->rx,
-						  vif->rx.rsp_prod_pvt++);
+				RING_GET_RESPONSE(&queue->rx,
+						  queue->rx.rsp_prod_pvt++);
 
 			resp->flags |= XEN_NETRXF_extra_info;
 
-			gso->u.gso.type = vif->meta[npo.meta_cons].gso_type;
-			gso->u.gso.size = vif->meta[npo.meta_cons].gso_size;
+			gso->u.gso.type = queue->meta[npo.meta_cons].gso_type;
+			gso->u.gso.size = queue->meta[npo.meta_cons].gso_size;
 			gso->u.gso.pad = 0;
 			gso->u.gso.features = 0;
 
@@ -699,11 +720,11 @@ static void xenvif_rx_action(struct xenvif *vif)
 			gso->flags = 0;
 		}
 
-		xenvif_add_frag_responses(vif, status,
-					  vif->meta + npo.meta_cons + 1,
+		xenvif_add_frag_responses(queue, status,
+					  queue->meta + npo.meta_cons + 1,
 					  XENVIF_RX_CB(skb)->meta_slots_used);
 
-		RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&vif->rx, ret);
+		RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&queue->rx, ret);
 
 		need_to_notify |= !!ret;
 
@@ -713,20 +734,20 @@ static void xenvif_rx_action(struct xenvif *vif)
 
 done:
 	if (need_to_notify)
-		notify_remote_via_irq(vif->rx_irq);
+		notify_remote_via_irq(queue->rx_irq);
 }
 
-void xenvif_napi_schedule_or_enable_events(struct xenvif *vif)
+void xenvif_napi_schedule_or_enable_events(struct xenvif_queue *queue)
 {
 	int more_to_do;
 
-	RING_FINAL_CHECK_FOR_REQUESTS(&vif->tx, more_to_do);
+	RING_FINAL_CHECK_FOR_REQUESTS(&queue->tx, more_to_do);
 
 	if (more_to_do)
-		napi_schedule(&vif->napi);
+		napi_schedule(&queue->napi);
 }
 
-static void tx_add_credit(struct xenvif *vif)
+static void tx_add_credit(struct xenvif_queue *queue)
 {
 	unsigned long max_burst, max_credit;
 
@@ -734,55 +755,57 @@ static void tx_add_credit(struct xenvif *vif)
 	 * Allow a burst big enough to transmit a jumbo packet of up to 128kB.
 	 * Otherwise the interface can seize up due to insufficient credit.
 	 */
-	max_burst = RING_GET_REQUEST(&vif->tx, vif->tx.req_cons)->size;
+	max_burst = RING_GET_REQUEST(&queue->tx, queue->tx.req_cons)->size;
 	max_burst = min(max_burst, 131072UL);
-	max_burst = max(max_burst, vif->credit_bytes);
+	max_burst = max(max_burst, queue->credit_bytes);
 
 	/* Take care that adding a new chunk of credit doesn't wrap to zero. */
-	max_credit = vif->remaining_credit + vif->credit_bytes;
-	if (max_credit < vif->remaining_credit)
+	max_credit = queue->remaining_credit + queue->credit_bytes;
+	if (max_credit < queue->remaining_credit)
 		max_credit = ULONG_MAX; /* wrapped: clamp to ULONG_MAX */
 
-	vif->remaining_credit = min(max_credit, max_burst);
+	queue->remaining_credit = min(max_credit, max_burst);
 }
 
 static void tx_credit_callback(unsigned long data)
 {
-	struct xenvif *vif = (struct xenvif *)data;
-	tx_add_credit(vif);
-	xenvif_napi_schedule_or_enable_events(vif);
+	struct xenvif_queue *queue = (struct xenvif_queue *)data;
+	tx_add_credit(queue);
+	xenvif_napi_schedule_or_enable_events(queue);
 }
 
-static void xenvif_tx_err(struct xenvif *vif,
+static void xenvif_tx_err(struct xenvif_queue *queue,
 			  struct xen_netif_tx_request *txp, RING_IDX end)
 {
-	RING_IDX cons = vif->tx.req_cons;
+	RING_IDX cons = queue->tx.req_cons;
 	unsigned long flags;
 
 	do {
-		spin_lock_irqsave(&vif->response_lock, flags);
-		make_tx_response(vif, txp, XEN_NETIF_RSP_ERROR);
-		spin_unlock_irqrestore(&vif->response_lock, flags);
+		spin_lock_irqsave(&queue->response_lock, flags);
+		make_tx_response(queue, txp, XEN_NETIF_RSP_ERROR);
+		spin_unlock_irqrestore(&queue->response_lock, flags);
 		if (cons == end)
 			break;
-		txp = RING_GET_REQUEST(&vif->tx, cons++);
+		txp = RING_GET_REQUEST(&queue->tx, cons++);
 	} while (1);
-	vif->tx.req_cons = cons;
+	queue->tx.req_cons = cons;
 }
 
 static void xenvif_fatal_tx_err(struct xenvif *vif)
 {
 	netdev_err(vif->dev, "fatal error; disabling device\n");
 	vif->disabled = true;
-	xenvif_kick_thread(vif);
+	/* Disable the vif from queue 0's kthread */
+	if (vif->queues)
+		xenvif_kick_thread(&vif->queues[0]);
 }
 
-static int xenvif_count_requests(struct xenvif *vif,
+static int xenvif_count_requests(struct xenvif_queue *queue,
 				 struct xen_netif_tx_request *first,
 				 struct xen_netif_tx_request *txp,
 				 int work_to_do)
 {
-	RING_IDX cons = vif->tx.req_cons;
+	RING_IDX cons = queue->tx.req_cons;
 	int slots = 0;
 	int drop_err = 0;
 	int more_data;
@@ -794,10 +817,10 @@ static int xenvif_count_requests(struct xenvif *vif,
 		struct xen_netif_tx_request dropped_tx = { 0 };
 
 		if (slots >= work_to_do) {
-			netdev_err(vif->dev,
+			netdev_err(queue->vif->dev,
 				   "Asked for %d slots but exceeds this limit\n",
 				   work_to_do);
-			xenvif_fatal_tx_err(vif);
+			xenvif_fatal_tx_err(queue->vif);
 			return -ENODATA;
 		}
 
@@ -805,10 +828,10 @@ static int xenvif_count_requests(struct xenvif *vif,
 		 * considered malicious.
 		 */
 		if (unlikely(slots >= fatal_skb_slots)) {
-			netdev_err(vif->dev,
+			netdev_err(queue->vif->dev,
 				   "Malicious frontend using %d slots, threshold %u\n",
 				   slots, fatal_skb_slots);
-			xenvif_fatal_tx_err(vif);
+			xenvif_fatal_tx_err(queue->vif);
 			return -E2BIG;
 		}
 
@@ -821,7 +844,7 @@ static int xenvif_count_requests(struct xenvif *vif,
 		 */
 		if (!drop_err && slots >= XEN_NETBK_LEGACY_SLOTS_MAX) {
 			if (net_ratelimit())
-				netdev_dbg(vif->dev,
+				netdev_dbg(queue->vif->dev,
 					   "Too many slots (%d) exceeding limit (%d), dropping packet\n",
 					   slots, XEN_NETBK_LEGACY_SLOTS_MAX);
 			drop_err = -E2BIG;
@@ -830,7 +853,7 @@ static int xenvif_count_requests(struct xenvif *vif,
 		if (drop_err)
 			txp = &dropped_tx;
 
-		memcpy(txp, RING_GET_REQUEST(&vif->tx, cons + slots),
+		memcpy(txp, RING_GET_REQUEST(&queue->tx, cons + slots),
 		       sizeof(*txp));
 
 		/* If the guest submitted a frame >= 64 KiB then
@@ -844,7 +867,7 @@ static int xenvif_count_requests(struct xenvif *vif,
 		 */
 		if (!drop_err && txp->size > first->size) {
 			if (net_ratelimit())
-				netdev_dbg(vif->dev,
+				netdev_dbg(queue->vif->dev,
 					   "Invalid tx request, slot size %u > remaining size %u\n",
 					   txp->size, first->size);
 			drop_err = -EIO;
@@ -854,9 +877,9 @@ static int xenvif_count_requests(struct xenvif *vif,
 		slots++;
 
 		if (unlikely((txp->offset + txp->size) > PAGE_SIZE)) {
-			netdev_err(vif->dev, "Cross page boundary, txp->offset: %x, size: %u\n",
+			netdev_err(queue->vif->dev, "Cross page boundary, txp->offset: %x, size: %u\n",
 				 txp->offset, txp->size);
-			xenvif_fatal_tx_err(vif);
+			xenvif_fatal_tx_err(queue->vif);
 			return -EINVAL;
 		}
 
@@ -868,7 +891,7 @@ static int xenvif_count_requests(struct xenvif *vif,
 	} while (more_data);
 
 	if (drop_err) {
-		xenvif_tx_err(vif, first, cons + slots);
+		xenvif_tx_err(queue, first, cons + slots);
 		return drop_err;
 	}
 
@@ -882,17 +905,17 @@ struct xenvif_tx_cb {
 
 #define XENVIF_TX_CB(skb) ((struct xenvif_tx_cb *)(skb)->cb)
 
-static inline void xenvif_tx_create_map_op(struct xenvif *vif,
+static inline void xenvif_tx_create_map_op(struct xenvif_queue *queue,
 					  u16 pending_idx,
 					  struct xen_netif_tx_request *txp,
 					  struct gnttab_map_grant_ref *mop)
 {
-	vif->pages_to_map[mop-vif->tx_map_ops] = vif->mmap_pages[pending_idx];
-	gnttab_set_map_op(mop, idx_to_kaddr(vif, pending_idx),
+	queue->pages_to_map[mop-queue->tx_map_ops] = queue->mmap_pages[pending_idx];
+	gnttab_set_map_op(mop, idx_to_kaddr(queue, pending_idx),
 			  GNTMAP_host_map | GNTMAP_readonly,
-			  txp->gref, vif->domid);
+			  txp->gref, queue->vif->domid);
 
-	memcpy(&vif->pending_tx_info[pending_idx].req, txp,
+	memcpy(&queue->pending_tx_info[pending_idx].req, txp,
 	       sizeof(*txp));
 }
 
@@ -913,7 +936,7 @@ static inline struct sk_buff *xenvif_alloc_skb(unsigned int size)
 	return skb;
 }
 
-static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
+static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif_queue *queue,
 							struct sk_buff *skb,
 							struct xen_netif_tx_request *txp,
 							struct gnttab_map_grant_ref *gop)
@@ -940,9 +963,9 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
 
 	for (shinfo->nr_frags = start; shinfo->nr_frags < nr_slots;
 	     shinfo->nr_frags++, txp++, gop++) {
-		index = pending_index(vif->pending_cons++);
-		pending_idx = vif->pending_ring[index];
-		xenvif_tx_create_map_op(vif, pending_idx, txp, gop);
+		index = pending_index(queue->pending_cons++);
+		pending_idx = queue->pending_ring[index];
+		xenvif_tx_create_map_op(queue, pending_idx, txp, gop);
 		frag_set_pending_idx(&frags[shinfo->nr_frags], pending_idx);
 	}
 
@@ -950,7 +973,7 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
 		struct sk_buff *nskb = xenvif_alloc_skb(0);
 		if (unlikely(nskb == NULL)) {
 			if (net_ratelimit())
-				netdev_err(vif->dev,
+				netdev_err(queue->vif->dev,
 					   "Can't allocate the frag_list skb.\n");
 			return NULL;
 		}
@@ -960,9 +983,9 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
 
 		for (shinfo->nr_frags = 0; shinfo->nr_frags < frag_overflow;
 		     shinfo->nr_frags++, txp++, gop++) {
-			index = pending_index(vif->pending_cons++);
-			pending_idx = vif->pending_ring[index];
-			xenvif_tx_create_map_op(vif, pending_idx, txp, gop);
+			index = pending_index(queue->pending_cons++);
+			pending_idx = queue->pending_ring[index];
+			xenvif_tx_create_map_op(queue, pending_idx, txp, gop);
 			frag_set_pending_idx(&frags[shinfo->nr_frags],
 					     pending_idx);
 		}
@@ -973,34 +996,34 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
 	return gop;
 }
 
-static inline void xenvif_grant_handle_set(struct xenvif *vif,
+static inline void xenvif_grant_handle_set(struct xenvif_queue *queue,
 					   u16 pending_idx,
 					   grant_handle_t handle)
 {
-	if (unlikely(vif->grant_tx_handle[pending_idx] !=
+	if (unlikely(queue->grant_tx_handle[pending_idx] !=
 		     NETBACK_INVALID_HANDLE)) {
-		netdev_err(vif->dev,
+		netdev_err(queue->vif->dev,
 			   "Trying to overwrite active handle! pending_idx: %x\n",
 			   pending_idx);
 		BUG();
 	}
-	vif->grant_tx_handle[pending_idx] = handle;
+	queue->grant_tx_handle[pending_idx] = handle;
 }
 
-static inline void xenvif_grant_handle_reset(struct xenvif *vif,
+static inline void xenvif_grant_handle_reset(struct xenvif_queue *queue,
 					     u16 pending_idx)
 {
-	if (unlikely(vif->grant_tx_handle[pending_idx] ==
+	if (unlikely(queue->grant_tx_handle[pending_idx] ==
 		     NETBACK_INVALID_HANDLE)) {
-		netdev_err(vif->dev,
+		netdev_err(queue->vif->dev,
 			   "Trying to unmap invalid handle! pending_idx: %x\n",
 			   pending_idx);
 		BUG();
 	}
-	vif->grant_tx_handle[pending_idx] = NETBACK_INVALID_HANDLE;
+	queue->grant_tx_handle[pending_idx] = NETBACK_INVALID_HANDLE;
 }
 
-static int xenvif_tx_check_gop(struct xenvif *vif,
+static int xenvif_tx_check_gop(struct xenvif_queue *queue,
 			       struct sk_buff *skb,
 			       struct gnttab_map_grant_ref **gopp_map,
 			       struct gnttab_copy **gopp_copy)
@@ -1017,12 +1040,12 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
 	(*gopp_copy)++;
 	if (unlikely(err)) {
 		if (net_ratelimit())
-			netdev_dbg(vif->dev,
+			netdev_dbg(queue->vif->dev,
 				   "Grant copy of header failed! status: %d pending_idx: %u ref: %u\n",
 				   (*gopp_copy)->status,
 				   pending_idx,
 				   (*gopp_copy)->source.u.ref);
-		xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
+		xenvif_idx_release(queue, pending_idx, XEN_NETIF_RSP_ERROR);
 	}
 
 check_frags:
@@ -1035,24 +1058,24 @@ check_frags:
 		newerr = gop_map->status;
 
 		if (likely(!newerr)) {
-			xenvif_grant_handle_set(vif,
+			xenvif_grant_handle_set(queue,
 						pending_idx,
 						gop_map->handle);
 			/* Had a previous error? Invalidate this fragment. */
 			if (unlikely(err))
-				xenvif_idx_unmap(vif, pending_idx);
+				xenvif_idx_unmap(queue, pending_idx);
 			continue;
 		}
 
 		/* Error on this fragment: respond to client with an error. */
 		if (net_ratelimit())
-			netdev_dbg(vif->dev,
+			netdev_dbg(queue->vif->dev,
 				   "Grant map of %d. frag failed! status: %d pending_idx: %u ref: %u\n",
 				   i,
 				   gop_map->status,
 				   pending_idx,
 				   gop_map->ref);
-		xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
+		xenvif_idx_release(queue, pending_idx, XEN_NETIF_RSP_ERROR);
 
 		/* Not the first error? Preceding frags already invalidated. */
 		if (err)
@@ -1060,7 +1083,7 @@ check_frags:
 		/* First error: invalidate preceding fragments. */
 		for (j = 0; j < i; j++) {
 			pending_idx = frag_get_pending_idx(&shinfo->frags[j]);
-			xenvif_idx_unmap(vif, pending_idx);
+			xenvif_idx_unmap(queue, pending_idx);
 		}
 
 		/* Remember the error: invalidate all subsequent fragments. */
@@ -1084,7 +1107,7 @@ check_frags:
 		shinfo = skb_shinfo(first_skb);
 		for (j = 0; j < shinfo->nr_frags; j++) {
 			pending_idx = frag_get_pending_idx(&shinfo->frags[j]);
-			xenvif_idx_unmap(vif, pending_idx);
+			xenvif_idx_unmap(queue, pending_idx);
 		}
 	}
 
@@ -1092,7 +1115,7 @@ check_frags:
 	return err;
 }
 
-static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb)
+static void xenvif_fill_frags(struct xenvif_queue *queue, struct sk_buff *skb)
 {
 	struct skb_shared_info *shinfo = skb_shinfo(skb);
 	int nr_frags = shinfo->nr_frags;
@@ -1110,23 +1133,23 @@ static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb)
 		/* If this is not the first frag, chain it to the previous*/
 		if (prev_pending_idx == INVALID_PENDING_IDX)
 			skb_shinfo(skb)->destructor_arg =
-				&callback_param(vif, pending_idx);
+				&callback_param(queue, pending_idx);
 		else
-			callback_param(vif, prev_pending_idx).ctx =
-				&callback_param(vif, pending_idx);
+			callback_param(queue, prev_pending_idx).ctx =
+				&callback_param(queue, pending_idx);
 
-		callback_param(vif, pending_idx).ctx = NULL;
+		callback_param(queue, pending_idx).ctx = NULL;
 		prev_pending_idx = pending_idx;
 
-		txp = &vif->pending_tx_info[pending_idx].req;
-		page = virt_to_page(idx_to_kaddr(vif, pending_idx));
+		txp = &queue->pending_tx_info[pending_idx].req;
+		page = virt_to_page(idx_to_kaddr(queue, pending_idx));
 		__skb_fill_page_desc(skb, i, page, txp->offset, txp->size);
 		skb->len += txp->size;
 		skb->data_len += txp->size;
 		skb->truesize += txp->size;
 
 		/* Take an extra reference to offset network stack's put_page */
-		get_page(vif->mmap_pages[pending_idx]);
+		get_page(queue->mmap_pages[pending_idx]);
 	}
 	/* FIXME: __skb_fill_page_desc set this to true because page->pfmemalloc
 	 * overlaps with "index", and "mapping" is not set. I think mapping
@@ -1136,33 +1159,33 @@ static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb)
 	skb->pfmemalloc	= false;
 }
 
-static int xenvif_get_extras(struct xenvif *vif,
+static int xenvif_get_extras(struct xenvif_queue *queue,
 				struct xen_netif_extra_info *extras,
 				int work_to_do)
 {
 	struct xen_netif_extra_info extra;
-	RING_IDX cons = vif->tx.req_cons;
+	RING_IDX cons = queue->tx.req_cons;
 
 	do {
 		if (unlikely(work_to_do-- <= 0)) {
-			netdev_err(vif->dev, "Missing extra info\n");
-			xenvif_fatal_tx_err(vif);
+			netdev_err(queue->vif->dev, "Missing extra info\n");
+			xenvif_fatal_tx_err(queue->vif);
 			return -EBADR;
 		}
 
-		memcpy(&extra, RING_GET_REQUEST(&vif->tx, cons),
+		memcpy(&extra, RING_GET_REQUEST(&queue->tx, cons),
 		       sizeof(extra));
 		if (unlikely(!extra.type ||
 			     extra.type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
-			vif->tx.req_cons = ++cons;
-			netdev_err(vif->dev,
+			queue->tx.req_cons = ++cons;
+			netdev_err(queue->vif->dev,
 				   "Invalid extra type: %d\n", extra.type);
-			xenvif_fatal_tx_err(vif);
+			xenvif_fatal_tx_err(queue->vif);
 			return -EINVAL;
 		}
 
 		memcpy(&extras[extra.type - 1], &extra, sizeof(extra));
-		vif->tx.req_cons = ++cons;
+		queue->tx.req_cons = ++cons;
 	} while (extra.flags & XEN_NETIF_EXTRA_FLAG_MORE);
 
 	return work_to_do;
@@ -1197,7 +1220,7 @@ static int xenvif_set_skb_gso(struct xenvif *vif,
 	return 0;
 }
 
-static int checksum_setup(struct xenvif *vif, struct sk_buff *skb)
+static int checksum_setup(struct xenvif_queue *queue, struct sk_buff *skb)
 {
 	bool recalculate_partial_csum = false;
 
@@ -1207,7 +1230,7 @@ static int checksum_setup(struct xenvif *vif, struct sk_buff *skb)
 	 * recalculate the partial checksum.
 	 */
 	if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) {
-		vif->rx_gso_checksum_fixup++;
+		queue->stats.rx_gso_checksum_fixup++;
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		recalculate_partial_csum = true;
 	}
@@ -1219,31 +1242,31 @@ static int checksum_setup(struct xenvif *vif, struct sk_buff *skb)
 	return skb_checksum_setup(skb, recalculate_partial_csum);
 }
 
-static bool tx_credit_exceeded(struct xenvif *vif, unsigned size)
+static bool tx_credit_exceeded(struct xenvif_queue *queue, unsigned size)
 {
 	u64 now = get_jiffies_64();
-	u64 next_credit = vif->credit_window_start +
-		msecs_to_jiffies(vif->credit_usec / 1000);
+	u64 next_credit = queue->credit_window_start +
+		msecs_to_jiffies(queue->credit_usec / 1000);
 
 	/* Timer could already be pending in rare cases. */
-	if (timer_pending(&vif->credit_timeout))
+	if (timer_pending(&queue->credit_timeout))
 		return true;
 
 	/* Passed the point where we can replenish credit? */
 	if (time_after_eq64(now, next_credit)) {
-		vif->credit_window_start = now;
-		tx_add_credit(vif);
+		queue->credit_window_start = now;
+		tx_add_credit(queue);
 	}
 
 	/* Still too big to send right now? Set a callback. */
-	if (size > vif->remaining_credit) {
-		vif->credit_timeout.data     =
-			(unsigned long)vif;
-		vif->credit_timeout.function =
+	if (size > queue->remaining_credit) {
+		queue->credit_timeout.data     =
+			(unsigned long)queue;
+		queue->credit_timeout.function =
 			tx_credit_callback;
-		mod_timer(&vif->credit_timeout,
+		mod_timer(&queue->credit_timeout,
 			  next_credit);
-		vif->credit_window_start = next_credit;
+		queue->credit_window_start = next_credit;
 
 		return true;
 	}
@@ -1251,16 +1274,16 @@ static bool tx_credit_exceeded(struct xenvif *vif, unsigned size)
 	return false;
 }
 
-static void xenvif_tx_build_gops(struct xenvif *vif,
+static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 				     int budget,
 				     unsigned *copy_ops,
 				     unsigned *map_ops)
 {
-	struct gnttab_map_grant_ref *gop = vif->tx_map_ops, *request_gop;
+	struct gnttab_map_grant_ref *gop = queue->tx_map_ops, *request_gop;
 	struct sk_buff *skb;
 	int ret;
 
-	while (skb_queue_len(&vif->tx_queue) < budget) {
+	while (skb_queue_len(&queue->tx_queue) < budget) {
 		struct xen_netif_tx_request txreq;
 		struct xen_netif_tx_request txfrags[XEN_NETBK_LEGACY_SLOTS_MAX];
 		struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX-1];
@@ -1270,69 +1293,69 @@ static void xenvif_tx_build_gops(struct xenvif *vif,
 		unsigned int data_len;
 		pending_ring_idx_t index;
 
-		if (vif->tx.sring->req_prod - vif->tx.req_cons >
+		if (queue->tx.sring->req_prod - queue->tx.req_cons >
 		    XEN_NETIF_TX_RING_SIZE) {
-			netdev_err(vif->dev,
+			netdev_err(queue->vif->dev,
 				   "Impossible number of requests. "
 				   "req_prod %d, req_cons %d, size %ld\n",
-				   vif->tx.sring->req_prod, vif->tx.req_cons,
+				   queue->tx.sring->req_prod, queue->tx.req_cons,
 				   XEN_NETIF_TX_RING_SIZE);
-			xenvif_fatal_tx_err(vif);
+			xenvif_fatal_tx_err(queue->vif);
 			break;
 		}
 
-		work_to_do = RING_HAS_UNCONSUMED_REQUESTS(&vif->tx);
+		work_to_do = RING_HAS_UNCONSUMED_REQUESTS(&queue->tx);
 		if (!work_to_do)
 			break;
 
-		idx = vif->tx.req_cons;
+		idx = queue->tx.req_cons;
 		rmb(); /* Ensure that we see the request before we copy it. */
-		memcpy(&txreq, RING_GET_REQUEST(&vif->tx, idx), sizeof(txreq));
+		memcpy(&txreq, RING_GET_REQUEST(&queue->tx, idx), sizeof(txreq));
 
 		/* Credit-based scheduling. */
-		if (txreq.size > vif->remaining_credit &&
-		    tx_credit_exceeded(vif, txreq.size))
+		if (txreq.size > queue->remaining_credit &&
+		    tx_credit_exceeded(queue, txreq.size))
 			break;
 
-		vif->remaining_credit -= txreq.size;
+		queue->remaining_credit -= txreq.size;
 
 		work_to_do--;
-		vif->tx.req_cons = ++idx;
+		queue->tx.req_cons = ++idx;
 
 		memset(extras, 0, sizeof(extras));
 		if (txreq.flags & XEN_NETTXF_extra_info) {
-			work_to_do = xenvif_get_extras(vif, extras,
+			work_to_do = xenvif_get_extras(queue, extras,
 						       work_to_do);
-			idx = vif->tx.req_cons;
+			idx = queue->tx.req_cons;
 			if (unlikely(work_to_do < 0))
 				break;
 		}
 
-		ret = xenvif_count_requests(vif, &txreq, txfrags, work_to_do);
+		ret = xenvif_count_requests(queue, &txreq, txfrags, work_to_do);
 		if (unlikely(ret < 0))
 			break;
 
 		idx += ret;
 
 		if (unlikely(txreq.size < ETH_HLEN)) {
-			netdev_dbg(vif->dev,
+			netdev_dbg(queue->vif->dev,
 				   "Bad packet size: %d\n", txreq.size);
-			xenvif_tx_err(vif, &txreq, idx);
+			xenvif_tx_err(queue, &txreq, idx);
 			break;
 		}
 
 		/* No crossing a page as the payload mustn't fragment. */
 		if (unlikely((txreq.offset + txreq.size) > PAGE_SIZE)) {
-			netdev_err(vif->dev,
+			netdev_err(queue->vif->dev,
 				   "txreq.offset: %x, size: %u, end: %lu\n",
 				   txreq.offset, txreq.size,
 				   (txreq.offset&~PAGE_MASK) + txreq.size);
-			xenvif_fatal_tx_err(vif);
+			xenvif_fatal_tx_err(queue->vif);
 			break;
 		}
 
-		index = pending_index(vif->pending_cons);
-		pending_idx = vif->pending_ring[index];
+		index = pending_index(queue->pending_cons);
+		pending_idx = queue->pending_ring[index];
 
 		data_len = (txreq.size > PKT_PROT_LEN &&
 			    ret < XEN_NETBK_LEGACY_SLOTS_MAX) ?
@@ -1340,9 +1363,9 @@ static void xenvif_tx_build_gops(struct xenvif *vif,
 
 		skb = xenvif_alloc_skb(data_len);
 		if (unlikely(skb == NULL)) {
-			netdev_dbg(vif->dev,
+			netdev_dbg(queue->vif->dev,
 				   "Can't allocate a skb in start_xmit.\n");
-			xenvif_tx_err(vif, &txreq, idx);
+			xenvif_tx_err(queue, &txreq, idx);
 			break;
 		}
 
@@ -1350,7 +1373,7 @@ static void xenvif_tx_build_gops(struct xenvif *vif,
 			struct xen_netif_extra_info *gso;
 			gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1];
 
-			if (xenvif_set_skb_gso(vif, skb, gso)) {
+			if (xenvif_set_skb_gso(queue->vif, skb, gso)) {
 				/* Failure in xenvif_set_skb_gso is fatal. */
 				kfree_skb(skb);
 				break;
@@ -1360,18 +1383,18 @@ static void xenvif_tx_build_gops(struct xenvif *vif,
 		XENVIF_TX_CB(skb)->pending_idx = pending_idx;
 
 		__skb_put(skb, data_len);
-		vif->tx_copy_ops[*copy_ops].source.u.ref = txreq.gref;
-		vif->tx_copy_ops[*copy_ops].source.domid = vif->domid;
-		vif->tx_copy_ops[*copy_ops].source.offset = txreq.offset;
+		queue->tx_copy_ops[*copy_ops].source.u.ref = txreq.gref;
+		queue->tx_copy_ops[*copy_ops].source.domid = queue->vif->domid;
+		queue->tx_copy_ops[*copy_ops].source.offset = txreq.offset;
 
-		vif->tx_copy_ops[*copy_ops].dest.u.gmfn =
+		queue->tx_copy_ops[*copy_ops].dest.u.gmfn =
 			virt_to_mfn(skb->data);
-		vif->tx_copy_ops[*copy_ops].dest.domid = DOMID_SELF;
-		vif->tx_copy_ops[*copy_ops].dest.offset =
+		queue->tx_copy_ops[*copy_ops].dest.domid = DOMID_SELF;
+		queue->tx_copy_ops[*copy_ops].dest.offset =
 			offset_in_page(skb->data);
 
-		vif->tx_copy_ops[*copy_ops].len = data_len;
-		vif->tx_copy_ops[*copy_ops].flags = GNTCOPY_source_gref;
+		queue->tx_copy_ops[*copy_ops].len = data_len;
+		queue->tx_copy_ops[*copy_ops].flags = GNTCOPY_source_gref;
 
 		(*copy_ops)++;
 
@@ -1380,42 +1403,42 @@ static void xenvif_tx_build_gops(struct xenvif *vif,
 			skb_shinfo(skb)->nr_frags++;
 			frag_set_pending_idx(&skb_shinfo(skb)->frags[0],
 					     pending_idx);
-			xenvif_tx_create_map_op(vif, pending_idx, &txreq, gop);
+			xenvif_tx_create_map_op(queue, pending_idx, &txreq, gop);
 			gop++;
 		} else {
 			frag_set_pending_idx(&skb_shinfo(skb)->frags[0],
 					     INVALID_PENDING_IDX);
-			memcpy(&vif->pending_tx_info[pending_idx].req, &txreq,
+			memcpy(&queue->pending_tx_info[pending_idx].req, &txreq,
 			       sizeof(txreq));
 		}
 
-		vif->pending_cons++;
+		queue->pending_cons++;
 
-		request_gop = xenvif_get_requests(vif, skb, txfrags, gop);
+		request_gop = xenvif_get_requests(queue, skb, txfrags, gop);
 		if (request_gop == NULL) {
 			kfree_skb(skb);
-			xenvif_tx_err(vif, &txreq, idx);
+			xenvif_tx_err(queue, &txreq, idx);
 			break;
 		}
 		gop = request_gop;
 
-		__skb_queue_tail(&vif->tx_queue, skb);
+		__skb_queue_tail(&queue->tx_queue, skb);
 
-		vif->tx.req_cons = idx;
+		queue->tx.req_cons = idx;
 
-		if (((gop-vif->tx_map_ops) >= ARRAY_SIZE(vif->tx_map_ops)) ||
-		    (*copy_ops >= ARRAY_SIZE(vif->tx_copy_ops)))
+		if (((gop-queue->tx_map_ops) >= ARRAY_SIZE(queue->tx_map_ops)) ||
+		    (*copy_ops >= ARRAY_SIZE(queue->tx_copy_ops)))
 			break;
 	}
 
-	(*map_ops) = gop - vif->tx_map_ops;
+	(*map_ops) = gop - queue->tx_map_ops;
 	return;
 }
 
 /* Consolidate skb with a frag_list into a brand new one with local pages on
  * frags. Returns 0 or -ENOMEM if can't allocate new pages.
  */
-static int xenvif_handle_frag_list(struct xenvif *vif, struct sk_buff *skb)
+static int xenvif_handle_frag_list(struct xenvif_queue *queue, struct sk_buff *skb)
 {
 	unsigned int offset = skb_headlen(skb);
 	skb_frag_t frags[MAX_SKB_FRAGS];
@@ -1423,10 +1446,10 @@ static int xenvif_handle_frag_list(struct xenvif *vif, struct sk_buff *skb)
 	struct ubuf_info *uarg;
 	struct sk_buff *nskb = skb_shinfo(skb)->frag_list;
 
-	vif->tx_zerocopy_sent += 2;
-	vif->tx_frag_overflow++;
+	queue->stats.tx_zerocopy_sent += 2;
+	queue->stats.tx_frag_overflow++;
 
-	xenvif_fill_frags(vif, nskb);
+	xenvif_fill_frags(queue, nskb);
 	/* Subtract frags size, we will correct it later */
 	skb->truesize -= skb->data_len;
 	skb->len += nskb->len;
@@ -1478,37 +1501,37 @@ static int xenvif_handle_frag_list(struct xenvif *vif, struct sk_buff *skb)
 	return 0;
 }
 
-static int xenvif_tx_submit(struct xenvif *vif)
+static int xenvif_tx_submit(struct xenvif_queue *queue)
 {
-	struct gnttab_map_grant_ref *gop_map = vif->tx_map_ops;
-	struct gnttab_copy *gop_copy = vif->tx_copy_ops;
+	struct gnttab_map_grant_ref *gop_map = queue->tx_map_ops;
+	struct gnttab_copy *gop_copy = queue->tx_copy_ops;
 	struct sk_buff *skb;
 	int work_done = 0;
 
-	while ((skb = __skb_dequeue(&vif->tx_queue)) != NULL) {
+	while ((skb = __skb_dequeue(&queue->tx_queue)) != NULL) {
 		struct xen_netif_tx_request *txp;
 		u16 pending_idx;
 		unsigned data_len;
 
 		pending_idx = XENVIF_TX_CB(skb)->pending_idx;
-		txp = &vif->pending_tx_info[pending_idx].req;
+		txp = &queue->pending_tx_info[pending_idx].req;
 
 		/* Check the remap error code. */
-		if (unlikely(xenvif_tx_check_gop(vif, skb, &gop_map, &gop_copy))) {
+		if (unlikely(xenvif_tx_check_gop(queue, skb, &gop_map, &gop_copy))) {
 			skb_shinfo(skb)->nr_frags = 0;
 			kfree_skb(skb);
 			continue;
 		}
 
 		data_len = skb->len;
-		callback_param(vif, pending_idx).ctx = NULL;
+		callback_param(queue, pending_idx).ctx = NULL;
 		if (data_len < txp->size) {
 			/* Append the packet payload as a fragment. */
 			txp->offset += data_len;
 			txp->size -= data_len;
 		} else {
 			/* Schedule a response immediately. */
-			xenvif_idx_release(vif, pending_idx,
+			xenvif_idx_release(queue, pending_idx,
 					   XEN_NETIF_RSP_OKAY);
 		}
 
@@ -1517,12 +1540,12 @@ static int xenvif_tx_submit(struct xenvif *vif)
 		else if (txp->flags & XEN_NETTXF_data_validated)
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 
-		xenvif_fill_frags(vif, skb);
+		xenvif_fill_frags(queue, skb);
 
 		if (unlikely(skb_has_frag_list(skb))) {
-			if (xenvif_handle_frag_list(vif, skb)) {
+			if (xenvif_handle_frag_list(queue, skb)) {
 				if (net_ratelimit())
-					netdev_err(vif->dev,
+					netdev_err(queue->vif->dev,
 						   "Not enough memory to consolidate frag_list!\n");
 				skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
 				kfree_skb(skb);
@@ -1535,12 +1558,12 @@ static int xenvif_tx_submit(struct xenvif *vif)
 			__pskb_pull_tail(skb, target - skb_headlen(skb));
 		}
 
-		skb->dev      = vif->dev;
+		skb->dev      = queue->vif->dev;
 		skb->protocol = eth_type_trans(skb, skb->dev);
 		skb_reset_network_header(skb);
 
-		if (checksum_setup(vif, skb)) {
-			netdev_dbg(vif->dev,
+		if (checksum_setup(queue, skb)) {
+			netdev_dbg(queue->vif->dev,
 				   "Can't setup checksum in net_tx_action\n");
 			/* We have to set this flag to trigger the callback */
 			if (skb_shinfo(skb)->destructor_arg)
@@ -1565,8 +1588,8 @@ static int xenvif_tx_submit(struct xenvif *vif)
 				DIV_ROUND_UP(skb->len - hdrlen, mss);
 		}
 
-		vif->dev->stats.rx_bytes += skb->len;
-		vif->dev->stats.rx_packets++;
+		queue->stats.rx_bytes += skb->len;
+		queue->stats.rx_packets++;
 
 		work_done++;
 
@@ -1577,7 +1600,7 @@ static int xenvif_tx_submit(struct xenvif *vif)
 		 */
 		if (skb_shinfo(skb)->destructor_arg) {
 			skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
-			vif->tx_zerocopy_sent++;
+			queue->stats.tx_zerocopy_sent++;
 		}
 
 		netif_receive_skb(skb);
@@ -1590,47 +1613,47 @@ void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success)
 {
 	unsigned long flags;
 	pending_ring_idx_t index;
-	struct xenvif *vif = ubuf_to_vif(ubuf);
+	struct xenvif_queue *queue = ubuf_to_queue(ubuf);
 
 	/* This is the only place where we grab this lock, to protect callbacks
 	 * from each other.
 	 */
-	spin_lock_irqsave(&vif->callback_lock, flags);
+	spin_lock_irqsave(&queue->callback_lock, flags);
 	do {
 		u16 pending_idx = ubuf->desc;
 		ubuf = (struct ubuf_info *) ubuf->ctx;
-		BUG_ON(vif->dealloc_prod - vif->dealloc_cons >=
+		BUG_ON(queue->dealloc_prod - queue->dealloc_cons >=
 			MAX_PENDING_REQS);
-		index = pending_index(vif->dealloc_prod);
-		vif->dealloc_ring[index] = pending_idx;
+		index = pending_index(queue->dealloc_prod);
+		queue->dealloc_ring[index] = pending_idx;
 		/* Sync with xenvif_tx_dealloc_action:
 		 * insert idx then incr producer.
 		 */
 		smp_wmb();
-		vif->dealloc_prod++;
+		queue->dealloc_prod++;
 	} while (ubuf);
-	wake_up(&vif->dealloc_wq);
-	spin_unlock_irqrestore(&vif->callback_lock, flags);
+	wake_up(&queue->dealloc_wq);
+	spin_unlock_irqrestore(&queue->callback_lock, flags);
 
 	if (likely(zerocopy_success))
-		vif->tx_zerocopy_success++;
+		queue->stats.tx_zerocopy_success++;
 	else
-		vif->tx_zerocopy_fail++;
+		queue->stats.tx_zerocopy_fail++;
 }
 
-static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
+static inline void xenvif_tx_dealloc_action(struct xenvif_queue *queue)
 {
 	struct gnttab_unmap_grant_ref *gop;
 	pending_ring_idx_t dc, dp;
 	u16 pending_idx, pending_idx_release[MAX_PENDING_REQS];
 	unsigned int i = 0;
 
-	dc = vif->dealloc_cons;
-	gop = vif->tx_unmap_ops;
+	dc = queue->dealloc_cons;
+	gop = queue->tx_unmap_ops;
 
 	/* Free up any grants we have finished using */
 	do {
-		dp = vif->dealloc_prod;
+		dp = queue->dealloc_prod;
 
 		/* Ensure we see all indices enqueued by all
 		 * xenvif_zerocopy_callback().
@@ -1638,38 +1661,38 @@ static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
 		smp_rmb();
 
 		while (dc != dp) {
-			BUG_ON(gop - vif->tx_unmap_ops > MAX_PENDING_REQS);
+			BUG_ON(gop - queue->tx_unmap_ops > MAX_PENDING_REQS);
 			pending_idx =
-				vif->dealloc_ring[pending_index(dc++)];
+				queue->dealloc_ring[pending_index(dc++)];
 
-			pending_idx_release[gop-vif->tx_unmap_ops] =
+			pending_idx_release[gop-queue->tx_unmap_ops] =
 				pending_idx;
-			vif->pages_to_unmap[gop-vif->tx_unmap_ops] =
-				vif->mmap_pages[pending_idx];
+			queue->pages_to_unmap[gop-queue->tx_unmap_ops] =
+				queue->mmap_pages[pending_idx];
 			gnttab_set_unmap_op(gop,
-					    idx_to_kaddr(vif, pending_idx),
+					    idx_to_kaddr(queue, pending_idx),
 					    GNTMAP_host_map,
-					    vif->grant_tx_handle[pending_idx]);
-			xenvif_grant_handle_reset(vif, pending_idx);
+					    queue->grant_tx_handle[pending_idx]);
+			xenvif_grant_handle_reset(queue, pending_idx);
 			++gop;
 		}
 
-	} while (dp != vif->dealloc_prod);
+	} while (dp != queue->dealloc_prod);
 
-	vif->dealloc_cons = dc;
+	queue->dealloc_cons = dc;
 
-	if (gop - vif->tx_unmap_ops > 0) {
+	if (gop - queue->tx_unmap_ops > 0) {
 		int ret;
-		ret = gnttab_unmap_refs(vif->tx_unmap_ops,
+		ret = gnttab_unmap_refs(queue->tx_unmap_ops,
 					NULL,
-					vif->pages_to_unmap,
-					gop - vif->tx_unmap_ops);
+					queue->pages_to_unmap,
+					gop - queue->tx_unmap_ops);
 		if (ret) {
-			netdev_err(vif->dev, "Unmap fail: nr_ops %tx ret %d\n",
-				   gop - vif->tx_unmap_ops, ret);
-			for (i = 0; i < gop - vif->tx_unmap_ops; ++i) {
+			netdev_err(queue->vif->dev, "Unmap fail: nr_ops %tx ret %d\n",
+				   gop - queue->tx_unmap_ops, ret);
+			for (i = 0; i < gop - queue->tx_unmap_ops; ++i) {
 				if (gop[i].status != GNTST_okay)
-					netdev_err(vif->dev,
+					netdev_err(queue->vif->dev,
 						   " host_addr: %llx handle: %x status: %d\n",
 						   gop[i].host_addr,
 						   gop[i].handle,
@@ -1679,91 +1702,91 @@ static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
 		}
 	}
 
-	for (i = 0; i < gop - vif->tx_unmap_ops; ++i)
-		xenvif_idx_release(vif, pending_idx_release[i],
+	for (i = 0; i < gop - queue->tx_unmap_ops; ++i)
+		xenvif_idx_release(queue, pending_idx_release[i],
 				   XEN_NETIF_RSP_OKAY);
 }
 
 
 /* Called after netfront has transmitted */
-int xenvif_tx_action(struct xenvif *vif, int budget)
+int xenvif_tx_action(struct xenvif_queue *queue, int budget)
 {
 	unsigned nr_mops, nr_cops = 0;
 	int work_done, ret;
 
-	if (unlikely(!tx_work_todo(vif)))
+	if (unlikely(!tx_work_todo(queue)))
 		return 0;
 
-	xenvif_tx_build_gops(vif, budget, &nr_cops, &nr_mops);
+	xenvif_tx_build_gops(queue, budget, &nr_cops, &nr_mops);
 
 	if (nr_cops == 0)
 		return 0;
 
-	gnttab_batch_copy(vif->tx_copy_ops, nr_cops);
+	gnttab_batch_copy(queue->tx_copy_ops, nr_cops);
 	if (nr_mops != 0) {
-		ret = gnttab_map_refs(vif->tx_map_ops,
+		ret = gnttab_map_refs(queue->tx_map_ops,
 				      NULL,
-				      vif->pages_to_map,
+				      queue->pages_to_map,
 				      nr_mops);
 		BUG_ON(ret);
 	}
 
-	work_done = xenvif_tx_submit(vif);
+	work_done = xenvif_tx_submit(queue);
 
 	return work_done;
 }
 
-static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
+static void xenvif_idx_release(struct xenvif_queue *queue, u16 pending_idx,
 			       u8 status)
 {
 	struct pending_tx_info *pending_tx_info;
 	pending_ring_idx_t index;
 	unsigned long flags;
 
-	pending_tx_info = &vif->pending_tx_info[pending_idx];
-	spin_lock_irqsave(&vif->response_lock, flags);
-	make_tx_response(vif, &pending_tx_info->req, status);
-	index = pending_index(vif->pending_prod);
-	vif->pending_ring[index] = pending_idx;
+	pending_tx_info = &queue->pending_tx_info[pending_idx];
+	spin_lock_irqsave(&queue->response_lock, flags);
+	make_tx_response(queue, &pending_tx_info->req, status);
+	index = pending_index(queue->pending_prod);
+	queue->pending_ring[index] = pending_idx;
 	/* TX shouldn't use the index before we give it back here */
 	mb();
-	vif->pending_prod++;
-	spin_unlock_irqrestore(&vif->response_lock, flags);
+	queue->pending_prod++;
+	spin_unlock_irqrestore(&queue->response_lock, flags);
 }
 
 
-static void make_tx_response(struct xenvif *vif,
+static void make_tx_response(struct xenvif_queue *queue,
 			     struct xen_netif_tx_request *txp,
 			     s8       st)
 {
-	RING_IDX i = vif->tx.rsp_prod_pvt;
+	RING_IDX i = queue->tx.rsp_prod_pvt;
 	struct xen_netif_tx_response *resp;
 	int notify;
 
-	resp = RING_GET_RESPONSE(&vif->tx, i);
+	resp = RING_GET_RESPONSE(&queue->tx, i);
 	resp->id     = txp->id;
 	resp->status = st;
 
 	if (txp->flags & XEN_NETTXF_extra_info)
-		RING_GET_RESPONSE(&vif->tx, ++i)->status = XEN_NETIF_RSP_NULL;
+		RING_GET_RESPONSE(&queue->tx, ++i)->status = XEN_NETIF_RSP_NULL;
 
-	vif->tx.rsp_prod_pvt = ++i;
-	RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&vif->tx, notify);
+	queue->tx.rsp_prod_pvt = ++i;
+	RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&queue->tx, notify);
 	if (notify)
-		notify_remote_via_irq(vif->tx_irq);
+		notify_remote_via_irq(queue->tx_irq);
 }
 
-static struct xen_netif_rx_response *make_rx_response(struct xenvif *vif,
+static struct xen_netif_rx_response *make_rx_response(struct xenvif_queue *queue,
 					     u16      id,
 					     s8       st,
 					     u16      offset,
 					     u16      size,
 					     u16      flags)
 {
-	RING_IDX i = vif->rx.rsp_prod_pvt;
+	RING_IDX i = queue->rx.rsp_prod_pvt;
 	struct xen_netif_rx_response *resp;
 
-	resp = RING_GET_RESPONSE(&vif->rx, i);
+	resp = RING_GET_RESPONSE(&queue->rx, i);
 	resp->offset     = offset;
 	resp->flags      = flags;
 	resp->id         = id;
@@ -1771,26 +1794,26 @@ static struct xen_netif_rx_response *make_rx_response(struct xenvif *vif,
 	if (st < 0)
 		resp->status = (s16)st;
 
-	vif->rx.rsp_prod_pvt = ++i;
+	queue->rx.rsp_prod_pvt = ++i;
 
 	return resp;
 }
 
-void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx)
+void xenvif_idx_unmap(struct xenvif_queue *queue, u16 pending_idx)
 {
 	int ret;
 	struct gnttab_unmap_grant_ref tx_unmap_op;
 
 	gnttab_set_unmap_op(&tx_unmap_op,
-			    idx_to_kaddr(vif, pending_idx),
+			    idx_to_kaddr(queue, pending_idx),
 			    GNTMAP_host_map,
-			    vif->grant_tx_handle[pending_idx]);
-	xenvif_grant_handle_reset(vif, pending_idx);
+			    queue->grant_tx_handle[pending_idx]);
+	xenvif_grant_handle_reset(queue, pending_idx);
 
 	ret = gnttab_unmap_refs(&tx_unmap_op, NULL,
-				&vif->mmap_pages[pending_idx], 1);
+				&queue->mmap_pages[pending_idx], 1);
 	if (ret) {
-		netdev_err(vif->dev,
+		netdev_err(queue->vif->dev,
 			   "Unmap fail: ret: %d pending_idx: %d host_addr: %llx handle: %x status: %d\n",
 			   ret,
 			   pending_idx,
@@ -1800,41 +1823,40 @@ void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx)
 		BUG();
 	}
 
-	xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_OKAY);
+	xenvif_idx_release(queue, pending_idx, XEN_NETIF_RSP_OKAY);
 }
 
-static inline int rx_work_todo(struct xenvif *vif)
+static inline int rx_work_todo(struct xenvif_queue *queue)
 {
-	return (!skb_queue_empty(&vif->rx_queue) &&
-	       xenvif_rx_ring_slots_available(vif, vif->rx_last_skb_slots)) ||
-	       vif->rx_queue_purge;
+	return (!skb_queue_empty(&queue->rx_queue) &&
+	       xenvif_rx_ring_slots_available(queue, queue->rx_last_skb_slots)) ||
+	       queue->rx_queue_purge;
 }
 
-static inline int tx_work_todo(struct xenvif *vif)
+static inline int tx_work_todo(struct xenvif_queue *queue)
 {
-
-	if (likely(RING_HAS_UNCONSUMED_REQUESTS(&vif->tx)))
+	if (likely(RING_HAS_UNCONSUMED_REQUESTS(&queue->tx)))
 		return 1;
 
 	return 0;
 }
 
-static inline bool tx_dealloc_work_todo(struct xenvif *vif)
+static inline bool tx_dealloc_work_todo(struct xenvif_queue *queue)
 {
-	return vif->dealloc_cons != vif->dealloc_prod;
+	return queue->dealloc_cons != queue->dealloc_prod;
 }
 
-void xenvif_unmap_frontend_rings(struct xenvif *vif)
+void xenvif_unmap_frontend_rings(struct xenvif_queue *queue)
 {
-	if (vif->tx.sring)
-		xenbus_unmap_ring_vfree(xenvif_to_xenbus_device(vif),
-					vif->tx.sring);
-	if (vif->rx.sring)
-		xenbus_unmap_ring_vfree(xenvif_to_xenbus_device(vif),
-					vif->rx.sring);
+	if (queue->tx.sring)
+		xenbus_unmap_ring_vfree(xenvif_to_xenbus_device(queue->vif),
+					queue->tx.sring);
+	if (queue->rx.sring)
+		xenbus_unmap_ring_vfree(xenvif_to_xenbus_device(queue->vif),
+					queue->rx.sring);
 }
 
-int xenvif_map_frontend_rings(struct xenvif *vif,
+int xenvif_map_frontend_rings(struct xenvif_queue *queue,
 			      grant_ref_t tx_ring_ref,
 			      grant_ref_t rx_ring_ref)
 {
@@ -1844,85 +1866,78 @@ int xenvif_map_frontend_rings(struct xenvif *vif,
 
 	int err = -ENOMEM;
 
-	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(vif),
+	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(queue->vif),
 				     tx_ring_ref, &addr);
 	if (err)
 		goto err;
 
 	txs = (struct xen_netif_tx_sring *)addr;
-	BACK_RING_INIT(&vif->tx, txs, PAGE_SIZE);
+	BACK_RING_INIT(&queue->tx, txs, PAGE_SIZE);
 
-	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(vif),
+	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(queue->vif),
 				     rx_ring_ref, &addr);
 	if (err)
 		goto err;
 
 	rxs = (struct xen_netif_rx_sring *)addr;
-	BACK_RING_INIT(&vif->rx, rxs, PAGE_SIZE);
+	BACK_RING_INIT(&queue->rx, rxs, PAGE_SIZE);
 
 	return 0;
 
 err:
-	xenvif_unmap_frontend_rings(vif);
+	xenvif_unmap_frontend_rings(queue);
 	return err;
 }
 
-void xenvif_stop_queue(struct xenvif *vif)
+static void xenvif_start_queue(struct xenvif_queue *queue)
 {
-	if (!vif->can_queue)
-		return;
-
-	netif_stop_queue(vif->dev);
-}
-
-static void xenvif_start_queue(struct xenvif *vif)
-{
-	if (xenvif_schedulable(vif))
-		netif_wake_queue(vif->dev);
+	if (xenvif_schedulable(queue->vif))
+		xenvif_wake_queue(queue);
 }
 
 int xenvif_kthread_guest_rx(void *data)
 {
-	struct xenvif *vif = data;
+	struct xenvif_queue *queue = data;
 	struct sk_buff *skb;
 
 	while (!kthread_should_stop()) {
-		wait_event_interruptible(vif->wq,
-					 rx_work_todo(vif) ||
-					 vif->disabled ||
+		wait_event_interruptible(queue->wq,
+					 rx_work_todo(queue) ||
+					 queue->vif->disabled ||
 					 kthread_should_stop());
 
 		/* This frontend is found to be rogue, disable it in
 		 * kthread context. Currently this is only set when
 		 * netback finds out frontend sends malformed packet,
 		 * but we cannot disable the interface in softirq
-		 * context so we defer it here.
+		 * context so we defer it here, if this thread is
+		 * associated with queue 0.
 		 */
-		if (unlikely(vif->disabled && netif_carrier_ok(vif->dev)))
-			xenvif_carrier_off(vif);
+		if (unlikely(queue->vif->disabled && netif_carrier_ok(queue->vif->dev) && queue->id == 0))
+			xenvif_carrier_off(queue->vif);
 
 		if (kthread_should_stop())
 			break;
 
-		if (vif->rx_queue_purge) {
-			skb_queue_purge(&vif->rx_queue);
-			vif->rx_queue_purge = false;
+		if (queue->rx_queue_purge) {
+			skb_queue_purge(&queue->rx_queue);
+			queue->rx_queue_purge = false;
 		}
 
-		if (!skb_queue_empty(&vif->rx_queue))
-			xenvif_rx_action(vif);
+		if (!skb_queue_empty(&queue->rx_queue))
+			xenvif_rx_action(queue);
 
-		if (skb_queue_empty(&vif->rx_queue) &&
-		    netif_queue_stopped(vif->dev)) {
-			del_timer_sync(&vif->wake_queue);
-			xenvif_start_queue(vif);
+		if (skb_queue_empty(&queue->rx_queue) &&
+		    xenvif_queue_stopped(queue)) {
+			del_timer_sync(&queue->wake_queue);
+			xenvif_start_queue(queue);
 		}
 
 		cond_resched();
 	}
 
 	/* Bin any remaining skbs */
-	while ((skb = skb_dequeue(&vif->rx_queue)) != NULL)
+	while ((skb = skb_dequeue(&queue->rx_queue)) != NULL)
 		dev_kfree_skb(skb);
 
 	return 0;
@@ -1930,22 +1945,22 @@ int xenvif_kthread_guest_rx(void *data)
 
 int xenvif_dealloc_kthread(void *data)
 {
-	struct xenvif *vif = data;
+	struct xenvif_queue *queue = data;
 
 	while (!kthread_should_stop()) {
-		wait_event_interruptible(vif->dealloc_wq,
-					 tx_dealloc_work_todo(vif) ||
+		wait_event_interruptible(queue->dealloc_wq,
+					 tx_dealloc_work_todo(queue) ||
 					 kthread_should_stop());
 		if (kthread_should_stop())
 			break;
 
-		xenvif_tx_dealloc_action(vif);
+		xenvif_tx_dealloc_action(queue);
 		cond_resched();
 	}
 
 	/* Unmap anything remaining*/
-	if (tx_dealloc_work_todo(vif))
-		xenvif_tx_dealloc_action(vif);
+	if (tx_dealloc_work_todo(queue))
+		xenvif_tx_dealloc_action(queue);
 
 	return 0;
 }
@@ -1957,6 +1972,9 @@ static int __init netback_init(void)
 	if (!xen_domain())
 		return -ENODEV;
 
+	/* Allow as many queues as there are CPUs, by default */
+	xenvif_max_queues = num_online_cpus();
+
 	if (fatal_skb_slots < XEN_NETBK_LEGACY_SLOTS_MAX) {
 		pr_info("fatal_skb_slots too small (%d), bump it to XEN_NETBK_LEGACY_SLOTS_MAX (%d)\n",
 			fatal_skb_slots, XEN_NETBK_LEGACY_SLOTS_MAX);
diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index 7a206cffb062..3d85acd84bad 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -19,6 +19,8 @@
 */
 
 #include "common.h"
+#include <linux/vmalloc.h>
+#include <linux/rtnetlink.h>
 
 struct backend_info {
 	struct xenbus_device *dev;
@@ -34,8 +36,9 @@ struct backend_info {
 	u8 have_hotplug_status_watch:1;
 };
 
-static int connect_rings(struct backend_info *);
-static void connect(struct backend_info *);
+static int connect_rings(struct backend_info *be, struct xenvif_queue *queue);
+static void connect(struct backend_info *be);
+static int read_xenbus_vif_flags(struct backend_info *be);
 static void backend_create_xenvif(struct backend_info *be);
 static void unregister_hotplug_status_watch(struct backend_info *be);
 static void set_backend_state(struct backend_info *be,
@@ -157,6 +160,12 @@ static int netback_probe(struct xenbus_device *dev,
 	if (err)
 		pr_debug("Error writing feature-split-event-channels\n");
 
+	/* Multi-queue support: This is an optional feature. */
+	err = xenbus_printf(XBT_NIL, dev->nodename,
+			    "multi-queue-max-queues", "%u", xenvif_max_queues);
+	if (err)
+		pr_debug("Error writing multi-queue-max-queues\n");
+
 	err = xenbus_switch_state(dev, XenbusStateInitWait);
 	if (err)
 		goto fail;
@@ -485,10 +494,26 @@ static void connect(struct backend_info *be)
 {
 	int err;
 	struct xenbus_device *dev = be->dev;
+	unsigned long credit_bytes, credit_usec;
+	unsigned int queue_index;
+	unsigned int requested_num_queues;
+	struct xenvif_queue *queue;
 
-	err = connect_rings(be);
-	if (err)
+	/* Check whether the frontend requested multiple queues
+	 * and read the number requested.
+	 */
+	err = xenbus_scanf(XBT_NIL, dev->otherend,
+			   "multi-queue-num-queues",
+			   "%u", &requested_num_queues);
+	if (err < 0) {
+		requested_num_queues = 1; /* Fall back to single queue */
+	} else if (requested_num_queues > xenvif_max_queues) {
+		/* buggy or malicious guest */
+		xenbus_dev_fatal(dev, err,
+				 "guest requested %u queues, exceeding the maximum of %u.",
+				 requested_num_queues, xenvif_max_queues);
 		return;
+	}
 
 	err = xen_net_read_mac(dev, be->vif->fe_dev_addr);
 	if (err) {
@@ -496,9 +521,56 @@ static void connect(struct backend_info *be)
 		return;
 	}
 
-	xen_net_read_rate(dev, &be->vif->credit_bytes,
-			  &be->vif->credit_usec);
-	be->vif->remaining_credit = be->vif->credit_bytes;
+	xen_net_read_rate(dev, &credit_bytes, &credit_usec);
+	read_xenbus_vif_flags(be);
+
+	/* Use the number of queues requested by the frontend */
+	be->vif->queues = vzalloc(requested_num_queues *
+				  sizeof(struct xenvif_queue));
+	be->vif->num_queues = requested_num_queues;
+
+	for (queue_index = 0; queue_index < requested_num_queues; ++queue_index) {
+		queue = &be->vif->queues[queue_index];
+		queue->vif = be->vif;
+		queue->id = queue_index;
+		snprintf(queue->name, sizeof(queue->name), "%s-q%u",
+				be->vif->dev->name, queue->id);
+
+		err = xenvif_init_queue(queue);
+		if (err) {
+			/* xenvif_init_queue() cleans up after itself on
+			 * failure, but we need to clean up any previously
+			 * initialised queues. Set num_queues to i so that
+			 * earlier queues can be destroyed using the regular
+			 * disconnect logic.
+			 */
+			be->vif->num_queues = queue_index;
+			goto err;
+		}
+
+		queue->remaining_credit = credit_bytes;
+
+		err = connect_rings(be, queue);
+		if (err) {
+			/* connect_rings() cleans up after itself on failure,
+			 * but we need to clean up after xenvif_init_queue() here,
+			 * and also clean up any previously initialised queues.
+			 */
+			xenvif_deinit_queue(queue);
+			be->vif->num_queues = queue_index;
+			goto err;
+		}
+	}
+
+	/* Initialisation completed, tell core driver the number of
+	 * active queues.
+	 */
+	rtnl_lock();
+	netif_set_real_num_tx_queues(be->vif->dev, requested_num_queues);
+	netif_set_real_num_rx_queues(be->vif->dev, requested_num_queues);
+	rtnl_unlock();
+
+	xenvif_carrier_on(be->vif);
 
 	unregister_hotplug_status_watch(be);
 	err = xenbus_watch_pathfmt(dev, &be->hotplug_status_watch,
@@ -507,45 +579,107 @@ static void connect(struct backend_info *be)
 	if (!err)
 		be->have_hotplug_status_watch = 1;
 
-	netif_wake_queue(be->vif->dev);
+	netif_tx_wake_all_queues(be->vif->dev);
+
+	return;
+
+err:
+	if (be->vif->num_queues > 0)
+		xenvif_disconnect(be->vif); /* Clean up existing queues */
+	vfree(be->vif->queues);
+	be->vif->queues = NULL;
+	be->vif->num_queues = 0;
+	return;
 }
 
 
-static int connect_rings(struct backend_info *be)
+static int connect_rings(struct backend_info *be, struct xenvif_queue *queue)
 {
-	struct xenvif *vif = be->vif;
 	struct xenbus_device *dev = be->dev;
+	unsigned int num_queues = queue->vif->num_queues;
 	unsigned long tx_ring_ref, rx_ring_ref;
-	unsigned int tx_evtchn, rx_evtchn, rx_copy;
+	unsigned int tx_evtchn, rx_evtchn;
 	int err;
-	int val;
+	char *xspath;
+	size_t xspathsize;
+	const size_t xenstore_path_ext_size = 11; /* sufficient for "/queue-NNN" */
+
+	/* If the frontend requested 1 queue, or we have fallen back
+	 * to single queue due to lack of frontend support for multi-
+	 * queue, expect the remaining XenStore keys in the toplevel
+	 * directory. Otherwise, expect them in a subdirectory called
+	 * queue-N.
+	 */
+	if (num_queues == 1) {
+		xspath = kzalloc(strlen(dev->otherend) + 1, GFP_KERNEL);
+		if (!xspath) {
+			xenbus_dev_fatal(dev, -ENOMEM,
+					 "reading ring references");
+			return -ENOMEM;
+		}
+		strcpy(xspath, dev->otherend);
+	} else {
+		xspathsize = strlen(dev->otherend) + xenstore_path_ext_size;
+		xspath = kzalloc(xspathsize, GFP_KERNEL);
+		if (!xspath) {
+			xenbus_dev_fatal(dev, -ENOMEM,
+					 "reading ring references");
+			return -ENOMEM;
+		}
+		snprintf(xspath, xspathsize, "%s/queue-%u", dev->otherend,
+			 queue->id);
+	}
 
-	err = xenbus_gather(XBT_NIL, dev->otherend,
+	err = xenbus_gather(XBT_NIL, xspath,
 			    "tx-ring-ref", "%lu", &tx_ring_ref,
 			    "rx-ring-ref", "%lu", &rx_ring_ref, NULL);
 	if (err) {
 		xenbus_dev_fatal(dev, err,
 				 "reading %s/ring-ref",
-				 dev->otherend);
-		return err;
+				 xspath);
+		goto err;
 	}
 
 	/* Try split event channels first, then single event channel. */
-	err = xenbus_gather(XBT_NIL, dev->otherend,
+	err = xenbus_gather(XBT_NIL, xspath,
 			    "event-channel-tx", "%u", &tx_evtchn,
 			    "event-channel-rx", "%u", &rx_evtchn, NULL);
 	if (err < 0) {
-		err = xenbus_scanf(XBT_NIL, dev->otherend,
+		err = xenbus_scanf(XBT_NIL, xspath,
 				   "event-channel", "%u", &tx_evtchn);
 		if (err < 0) {
 			xenbus_dev_fatal(dev, err,
 					 "reading %s/event-channel(-tx/rx)",
-					 dev->otherend);
-			return err;
+					 xspath);
+			goto err;
 		}
 		rx_evtchn = tx_evtchn;
 	}
 
+	/* Map the shared frame, irq etc. */
+	err = xenvif_connect(queue, tx_ring_ref, rx_ring_ref,
+			     tx_evtchn, rx_evtchn);
+	if (err) {
+		xenbus_dev_fatal(dev, err,
+				 "mapping shared-frames %lu/%lu port tx %u rx %u",
+				 tx_ring_ref, rx_ring_ref,
+				 tx_evtchn, rx_evtchn);
+		goto err;
+	}
+
+	err = 0;
+err: /* Regular return falls through with err == 0 */
+	kfree(xspath);
+	return err;
+}
+
+static int read_xenbus_vif_flags(struct backend_info *be)
+{
+	struct xenvif *vif = be->vif;
+	struct xenbus_device *dev = be->dev;
+	unsigned int rx_copy;
+	int err, val;
+
 	err = xenbus_scanf(XBT_NIL, dev->otherend, "request-rx-copy", "%u",
 			   &rx_copy);
 	if (err == -ENOENT) {
@@ -621,16 +755,6 @@ static int connect_rings(struct backend_info *be)
 		val = 0;
 	vif->ipv6_csum = !!val;
 
-	/* Map the shared frame, irq etc. */
-	err = xenvif_connect(vif, tx_ring_ref, rx_ring_ref,
-			     tx_evtchn, rx_evtchn);
-	if (err) {
-		xenbus_dev_fatal(dev, err,
-				 "mapping shared-frames %lu/%lu port tx %u rx %u",
-				 tx_ring_ref, rx_ring_ref,
-				 tx_evtchn, rx_evtchn);
-		return err;
-	}
 	return 0;
 }
 
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index 158b5e639fc7..2ccb4a02368b 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -57,6 +57,12 @@
 #include <xen/interface/memory.h>
 #include <xen/interface/grant_table.h>
 
+/* Module parameters */
+static unsigned int xennet_max_queues;
+module_param_named(max_queues, xennet_max_queues, uint, 0644);
+MODULE_PARM_DESC(max_queues,
+		 "Maximum number of queues per virtual interface");
+
 static const struct ethtool_ops xennet_ethtool_ops;
 
 struct netfront_cb {
@@ -73,6 +79,12 @@ struct netfront_cb {
 #define NET_RX_RING_SIZE __CONST_RING_SIZE(xen_netif_rx, PAGE_SIZE)
 #define TX_MAX_TARGET min_t(int, NET_TX_RING_SIZE, 256)
 
+/* Queue name is interface name with "-qNNN" appended */
+#define QUEUE_NAME_SIZE (IFNAMSIZ + 6)
+
+/* IRQ name is queue name with "-tx" or "-rx" appended */
+#define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3)
+
 struct netfront_stats {
 	u64			rx_packets;
 	u64			tx_packets;
@@ -81,9 +93,12 @@ struct netfront_stats {
 	struct u64_stats_sync	syncp;
 };
 
-struct netfront_info {
-	struct list_head list;
-	struct net_device *netdev;
+struct netfront_info;
+
+struct netfront_queue {
+	unsigned int id; /* Queue ID, 0-based */
+	char name[QUEUE_NAME_SIZE]; /* DEVNAME-qN */
+	struct netfront_info *info;
 
 	struct napi_struct napi;
 
@@ -93,10 +108,8 @@ struct netfront_info {
 	unsigned int tx_evtchn, rx_evtchn;
 	unsigned int tx_irq, rx_irq;
 	/* Only used when split event channels support is enabled */
-	char tx_irq_name[IFNAMSIZ+4]; /* DEVNAME-tx */
-	char rx_irq_name[IFNAMSIZ+4]; /* DEVNAME-rx */
-
-	struct xenbus_device *xbdev;
+	char tx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-tx */
+	char rx_irq_name[IRQ_NAME_SIZE]; /* DEVNAME-qN-rx */
 
 	spinlock_t   tx_lock;
 	struct xen_netif_tx_front_ring tx;
@@ -140,11 +153,21 @@ struct netfront_info {
 	unsigned long rx_pfn_array[NET_RX_RING_SIZE];
 	struct multicall_entry rx_mcl[NET_RX_RING_SIZE+1];
 	struct mmu_update rx_mmu[NET_RX_RING_SIZE];
+};
+
+struct netfront_info {
+	struct list_head list;
+	struct net_device *netdev;
+
+	struct xenbus_device *xbdev;
+
+	/* Multi-queue support */
+	struct netfront_queue *queues;
 
 	/* Statistics */
 	struct netfront_stats __percpu *stats;
 
-	unsigned long rx_gso_checksum_fixup;
+	atomic_t rx_gso_checksum_fixup;
 };
 
 struct netfront_rx_info {
@@ -187,21 +210,21 @@ static int xennet_rxidx(RING_IDX idx)
 	return idx & (NET_RX_RING_SIZE - 1);
 }
 
-static struct sk_buff *xennet_get_rx_skb(struct netfront_info *np,
+static struct sk_buff *xennet_get_rx_skb(struct netfront_queue *queue,
 					 RING_IDX ri)
 {
 	int i = xennet_rxidx(ri);
-	struct sk_buff *skb = np->rx_skbs[i];
-	np->rx_skbs[i] = NULL;
+	struct sk_buff *skb = queue->rx_skbs[i];
+	queue->rx_skbs[i] = NULL;
 	return skb;
 }
 
-static grant_ref_t xennet_get_rx_ref(struct netfront_info *np,
+static grant_ref_t xennet_get_rx_ref(struct netfront_queue *queue,
 					    RING_IDX ri)
 {
 	int i = xennet_rxidx(ri);
-	grant_ref_t ref = np->grant_rx_ref[i];
-	np->grant_rx_ref[i] = GRANT_INVALID_REF;
+	grant_ref_t ref = queue->grant_rx_ref[i];
+	queue->grant_rx_ref[i] = GRANT_INVALID_REF;
 	return ref;
 }
 
@@ -221,41 +244,40 @@ static bool xennet_can_sg(struct net_device *dev)
 
 static void rx_refill_timeout(unsigned long data)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct netfront_info *np = netdev_priv(dev);
-	napi_schedule(&np->napi);
+	struct netfront_queue *queue = (struct netfront_queue *)data;
+	napi_schedule(&queue->napi);
 }
 
-static int netfront_tx_slot_available(struct netfront_info *np)
+static int netfront_tx_slot_available(struct netfront_queue *queue)
 {
-	return (np->tx.req_prod_pvt - np->tx.rsp_cons) <
+	return (queue->tx.req_prod_pvt - queue->tx.rsp_cons) <
 		(TX_MAX_TARGET - MAX_SKB_FRAGS - 2);
 }
 
-static void xennet_maybe_wake_tx(struct net_device *dev)
+static void xennet_maybe_wake_tx(struct netfront_queue *queue)
 {
-	struct netfront_info *np = netdev_priv(dev);
+	struct net_device *dev = queue->info->netdev;
+	struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, queue->id);
 
-	if (unlikely(netif_queue_stopped(dev)) &&
-	    netfront_tx_slot_available(np) &&
+	if (unlikely(netif_tx_queue_stopped(dev_queue)) &&
+	    netfront_tx_slot_available(queue) &&
 	    likely(netif_running(dev)))
-		netif_wake_queue(dev);
+		netif_tx_wake_queue(netdev_get_tx_queue(dev, queue->id));
 }
 
-static void xennet_alloc_rx_buffers(struct net_device *dev)
+static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
 {
 	unsigned short id;
-	struct netfront_info *np = netdev_priv(dev);
 	struct sk_buff *skb;
 	struct page *page;
 	int i, batch_target, notify;
-	RING_IDX req_prod = np->rx.req_prod_pvt;
+	RING_IDX req_prod = queue->rx.req_prod_pvt;
 	grant_ref_t ref;
 	unsigned long pfn;
 	void *vaddr;
 	struct xen_netif_rx_request *req;
 
-	if (unlikely(!netif_carrier_ok(dev)))
+	if (unlikely(!netif_carrier_ok(queue->info->netdev)))
 		return;
 
 	/*
@@ -264,9 +286,10 @@ static void xennet_alloc_rx_buffers(struct net_device *dev)
 	 * allocator, so should reduce the chance of failed allocation requests
 	 * both for ourself and for other kernel subsystems.
 	 */
-	batch_target = np->rx_target - (req_prod - np->rx.rsp_cons);
-	for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) {
-		skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN,
+	batch_target = queue->rx_target - (req_prod - queue->rx.rsp_cons);
+	for (i = skb_queue_len(&queue->rx_batch); i < batch_target; i++) {
+		skb = __netdev_alloc_skb(queue->info->netdev,
+					 RX_COPY_THRESHOLD + NET_IP_ALIGN,
 					 GFP_ATOMIC | __GFP_NOWARN);
 		if (unlikely(!skb))
 			goto no_skb;
@@ -279,7 +302,7 @@ static void xennet_alloc_rx_buffers(struct net_device *dev)
 			kfree_skb(skb);
 no_skb:
 			/* Could not allocate any skbuffs. Try again later. */
-			mod_timer(&np->rx_refill_timer,
+			mod_timer(&queue->rx_refill_timer,
 				  jiffies + (HZ/10));
 
 			/* Any skbuffs queued for refill? Force them out. */
@@ -289,44 +312,44 @@ no_skb:
 		}
 
 		skb_add_rx_frag(skb, 0, page, 0, 0, PAGE_SIZE);
-		__skb_queue_tail(&np->rx_batch, skb);
+		__skb_queue_tail(&queue->rx_batch, skb);
 	}
 
 	/* Is the batch large enough to be worthwhile? */
-	if (i < (np->rx_target/2)) {
-		if (req_prod > np->rx.sring->req_prod)
+	if (i < (queue->rx_target/2)) {
+		if (req_prod > queue->rx.sring->req_prod)
 			goto push;
 		return;
 	}
 
 	/* Adjust our fill target if we risked running out of buffers. */
-	if (((req_prod - np->rx.sring->rsp_prod) < (np->rx_target / 4)) &&
-	    ((np->rx_target *= 2) > np->rx_max_target))
-		np->rx_target = np->rx_max_target;
+	if (((req_prod - queue->rx.sring->rsp_prod) < (queue->rx_target / 4)) &&
+	    ((queue->rx_target *= 2) > queue->rx_max_target))
+		queue->rx_target = queue->rx_max_target;
 
  refill:
 	for (i = 0; ; i++) {
-		skb = __skb_dequeue(&np->rx_batch);
+		skb = __skb_dequeue(&queue->rx_batch);
 		if (skb == NULL)
 			break;
 
-		skb->dev = dev;
+		skb->dev = queue->info->netdev;
 
 		id = xennet_rxidx(req_prod + i);
 
-		BUG_ON(np->rx_skbs[id]);
-		np->rx_skbs[id] = skb;
+		BUG_ON(queue->rx_skbs[id]);
+		queue->rx_skbs[id] = skb;
 
-		ref = gnttab_claim_grant_reference(&np->gref_rx_head);
+		ref = gnttab_claim_grant_reference(&queue->gref_rx_head);
 		BUG_ON((signed short)ref < 0);
-		np->grant_rx_ref[id] = ref;
+		queue->grant_rx_ref[id] = ref;
 
 		pfn = page_to_pfn(skb_frag_page(&skb_shinfo(skb)->frags[0]));
 		vaddr = page_address(skb_frag_page(&skb_shinfo(skb)->frags[0]));
 
-		req = RING_GET_REQUEST(&np->rx, req_prod + i);
+		req = RING_GET_REQUEST(&queue->rx, req_prod + i);
 		gnttab_grant_foreign_access_ref(ref,
-						np->xbdev->otherend_id,
+						queue->info->xbdev->otherend_id,
 						pfn_to_mfn(pfn),
 						0);
 
@@ -337,72 +360,77 @@ no_skb:
 	wmb();		/* barrier so backend seens requests */
 
 	/* Above is a suitable barrier to ensure backend will see requests. */
-	np->rx.req_prod_pvt = req_prod + i;
+	queue->rx.req_prod_pvt = req_prod + i;
  push:
-	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->rx, notify);
+	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->rx, notify);
 	if (notify)
-		notify_remote_via_irq(np->rx_irq);
+		notify_remote_via_irq(queue->rx_irq);
 }
 
 static int xennet_open(struct net_device *dev)
 {
 	struct netfront_info *np = netdev_priv(dev);
-
-	napi_enable(&np->napi);
-
-	spin_lock_bh(&np->rx_lock);
-	if (netif_carrier_ok(dev)) {
-		xennet_alloc_rx_buffers(dev);
-		np->rx.sring->rsp_event = np->rx.rsp_cons + 1;
-		if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx))
-			napi_schedule(&np->napi);
+	unsigned int num_queues = dev->real_num_tx_queues;
+	unsigned int i = 0;
+	struct netfront_queue *queue = NULL;
+
+	for (i = 0; i < num_queues; ++i) {
+		queue = &np->queues[i];
+		napi_enable(&queue->napi);
+
+		spin_lock_bh(&queue->rx_lock);
+		if (netif_carrier_ok(dev)) {
+			xennet_alloc_rx_buffers(queue);
+			queue->rx.sring->rsp_event = queue->rx.rsp_cons + 1;
+			if (RING_HAS_UNCONSUMED_RESPONSES(&queue->rx))
+				napi_schedule(&queue->napi);
+		}
+		spin_unlock_bh(&queue->rx_lock);
 	}
-	spin_unlock_bh(&np->rx_lock);
 
-	netif_start_queue(dev);
+	netif_tx_start_all_queues(dev);
 
 	return 0;
 }
 
-static void xennet_tx_buf_gc(struct net_device *dev)
+static void xennet_tx_buf_gc(struct netfront_queue *queue)
 {
 	RING_IDX cons, prod;
 	unsigned short id;
-	struct netfront_info *np = netdev_priv(dev);
 	struct sk_buff *skb;
 
-	BUG_ON(!netif_carrier_ok(dev));
+	BUG_ON(!netif_carrier_ok(queue->info->netdev));
 
 	do {
-		prod = np->tx.sring->rsp_prod;
+		prod = queue->tx.sring->rsp_prod;
 		rmb(); /* Ensure we see responses up to 'rp'. */
 
-		for (cons = np->tx.rsp_cons; cons != prod; cons++) {
+		for (cons = queue->tx.rsp_cons; cons != prod; cons++) {
 			struct xen_netif_tx_response *txrsp;
 
-			txrsp = RING_GET_RESPONSE(&np->tx, cons);
+			txrsp = RING_GET_RESPONSE(&queue->tx, cons);
 			if (txrsp->status == XEN_NETIF_RSP_NULL)
 				continue;
 
 			id  = txrsp->id;
-			skb = np->tx_skbs[id].skb;
+			skb = queue->tx_skbs[id].skb;
 			if (unlikely(gnttab_query_foreign_access(
-				np->grant_tx_ref[id]) != 0)) {
+				queue->grant_tx_ref[id]) != 0)) {
 				pr_alert("%s: warning -- grant still in use by backend domain\n",
 					 __func__);
 				BUG();
 			}
 			gnttab_end_foreign_access_ref(
-				np->grant_tx_ref[id], GNTMAP_readonly);
+				queue->grant_tx_ref[id], GNTMAP_readonly);
 			gnttab_release_grant_reference(
-				&np->gref_tx_head, np->grant_tx_ref[id]);
-			np->grant_tx_ref[id] = GRANT_INVALID_REF;
-			np->grant_tx_page[id] = NULL;
-			add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, id);
+				&queue->gref_tx_head, queue->grant_tx_ref[id]);
+			queue->grant_tx_ref[id] = GRANT_INVALID_REF;
+			queue->grant_tx_page[id] = NULL;
+			add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, id);
 			dev_kfree_skb_irq(skb);
 		}
 
-		np->tx.rsp_cons = prod;
+		queue->tx.rsp_cons = prod;
 
 		/*
 		 * Set a new event, then check for race with update of tx_cons.
@@ -412,21 +440,20 @@ static void xennet_tx_buf_gc(struct net_device *dev)
 		 * data is outstanding: in such cases notification from Xen is
 		 * likely to be the only kick that we'll get.
 		 */
-		np->tx.sring->rsp_event =
-			prod + ((np->tx.sring->req_prod - prod) >> 1) + 1;
+		queue->tx.sring->rsp_event =
+			prod + ((queue->tx.sring->req_prod - prod) >> 1) + 1;
 		mb();		/* update shared area */
-	} while ((cons == prod) && (prod != np->tx.sring->rsp_prod));
+	} while ((cons == prod) && (prod != queue->tx.sring->rsp_prod));
 
-	xennet_maybe_wake_tx(dev);
+	xennet_maybe_wake_tx(queue);
 }
 
-static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
+static void xennet_make_frags(struct sk_buff *skb, struct netfront_queue *queue,
 			      struct xen_netif_tx_request *tx)
 {
-	struct netfront_info *np = netdev_priv(dev);
 	char *data = skb->data;
 	unsigned long mfn;
-	RING_IDX prod = np->tx.req_prod_pvt;
+	RING_IDX prod = queue->tx.req_prod_pvt;
 	int frags = skb_shinfo(skb)->nr_frags;
 	unsigned int offset = offset_in_page(data);
 	unsigned int len = skb_headlen(skb);
@@ -443,19 +470,19 @@ static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
 		data += tx->size;
 		offset = 0;
 
-		id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs);
-		np->tx_skbs[id].skb = skb_get(skb);
-		tx = RING_GET_REQUEST(&np->tx, prod++);
+		id = get_id_from_freelist(&queue->tx_skb_freelist, queue->tx_skbs);
+		queue->tx_skbs[id].skb = skb_get(skb);
+		tx = RING_GET_REQUEST(&queue->tx, prod++);
 		tx->id = id;
-		ref = gnttab_claim_grant_reference(&np->gref_tx_head);
+		ref = gnttab_claim_grant_reference(&queue->gref_tx_head);
 		BUG_ON((signed short)ref < 0);
 
 		mfn = virt_to_mfn(data);
-		gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id,
+		gnttab_grant_foreign_access_ref(ref, queue->info->xbdev->otherend_id,
 						mfn, GNTMAP_readonly);
 
-		np->grant_tx_page[id] = virt_to_page(data);
-		tx->gref = np->grant_tx_ref[id] = ref;
+		queue->grant_tx_page[id] = virt_to_page(data);
+		tx->gref = queue->grant_tx_ref[id] = ref;
 		tx->offset = offset;
 		tx->size = len;
 		tx->flags = 0;
@@ -487,21 +514,21 @@ static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
 
 			tx->flags |= XEN_NETTXF_more_data;
 
-			id = get_id_from_freelist(&np->tx_skb_freelist,
-						  np->tx_skbs);
-			np->tx_skbs[id].skb = skb_get(skb);
-			tx = RING_GET_REQUEST(&np->tx, prod++);
+			id = get_id_from_freelist(&queue->tx_skb_freelist,
+						  queue->tx_skbs);
+			queue->tx_skbs[id].skb = skb_get(skb);
+			tx = RING_GET_REQUEST(&queue->tx, prod++);
 			tx->id = id;
-			ref = gnttab_claim_grant_reference(&np->gref_tx_head);
+			ref = gnttab_claim_grant_reference(&queue->gref_tx_head);
 			BUG_ON((signed short)ref < 0);
 
 			mfn = pfn_to_mfn(page_to_pfn(page));
 			gnttab_grant_foreign_access_ref(ref,
-							np->xbdev->otherend_id,
+							queue->info->xbdev->otherend_id,
 							mfn, GNTMAP_readonly);
 
-			np->grant_tx_page[id] = page;
-			tx->gref = np->grant_tx_ref[id] = ref;
+			queue->grant_tx_page[id] = page;
+			tx->gref = queue->grant_tx_ref[id] = ref;
 			tx->offset = offset;
 			tx->size = bytes;
 			tx->flags = 0;
@@ -518,7 +545,7 @@ static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
 		}
 	}
 
-	np->tx.req_prod_pvt = prod;
+	queue->tx.req_prod_pvt = prod;
 }
 
 /*
@@ -544,6 +571,24 @@ static int xennet_count_skb_frag_slots(struct sk_buff *skb)
 	return pages;
 }
 
+static u16 xennet_select_queue(struct net_device *dev, struct sk_buff *skb,
+			       void *accel_priv, select_queue_fallback_t fallback)
+{
+	unsigned int num_queues = dev->real_num_tx_queues;
+	u32 hash;
+	u16 queue_idx;
+
+	/* First, check if there is only one queue */
+	if (num_queues == 1) {
+		queue_idx = 0;
+	} else {
+		hash = skb_get_hash(skb);
+		queue_idx = hash % num_queues;
+	}
+
+	return queue_idx;
+}
+
 static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	unsigned short id;
@@ -559,6 +604,16 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	unsigned int offset = offset_in_page(data);
 	unsigned int len = skb_headlen(skb);
 	unsigned long flags;
+	struct netfront_queue *queue = NULL;
+	unsigned int num_queues = dev->real_num_tx_queues;
+	u16 queue_index;
+
+	/* Drop the packet if no queues are set up */
+	if (num_queues < 1)
+		goto drop;
+	/* Determine which queue to transmit this SKB on */
+	queue_index = skb_get_queue_mapping(skb);
+	queue = &np->queues[queue_index];
 
 	/* If skb->len is too big for wire format, drop skb and alert
 	 * user about misconfiguration.
@@ -578,30 +633,30 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto drop;
 	}
 
-	spin_lock_irqsave(&np->tx_lock, flags);
+	spin_lock_irqsave(&queue->tx_lock, flags);
 
 	if (unlikely(!netif_carrier_ok(dev) ||
 		     (slots > 1 && !xennet_can_sg(dev)) ||
 		     netif_needs_gso(skb, netif_skb_features(skb)))) {
-		spin_unlock_irqrestore(&np->tx_lock, flags);
+		spin_unlock_irqrestore(&queue->tx_lock, flags);
 		goto drop;
 	}
 
-	i = np->tx.req_prod_pvt;
+	i = queue->tx.req_prod_pvt;
 
-	id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs);
-	np->tx_skbs[id].skb = skb;
+	id = get_id_from_freelist(&queue->tx_skb_freelist, queue->tx_skbs);
+	queue->tx_skbs[id].skb = skb;
 
-	tx = RING_GET_REQUEST(&np->tx, i);
+	tx = RING_GET_REQUEST(&queue->tx, i);
 
 	tx->id   = id;
-	ref = gnttab_claim_grant_reference(&np->gref_tx_head);
+	ref = gnttab_claim_grant_reference(&queue->gref_tx_head);
 	BUG_ON((signed short)ref < 0);
 	mfn = virt_to_mfn(data);
 	gnttab_grant_foreign_access_ref(
-		ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly);
-	np->grant_tx_page[id] = virt_to_page(data);
-	tx->gref = np->grant_tx_ref[id] = ref;
+		ref, queue->info->xbdev->otherend_id, mfn, GNTMAP_readonly);
+	queue->grant_tx_page[id] = virt_to_page(data);
+	tx->gref = queue->grant_tx_ref[id] = ref;
 	tx->offset = offset;
 	tx->size = len;
 
@@ -617,7 +672,7 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		struct xen_netif_extra_info *gso;
 
 		gso = (struct xen_netif_extra_info *)
-			RING_GET_REQUEST(&np->tx, ++i);
+			RING_GET_REQUEST(&queue->tx, ++i);
 
 		tx->flags |= XEN_NETTXF_extra_info;
 
@@ -632,14 +687,14 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		gso->flags = 0;
 	}
 
-	np->tx.req_prod_pvt = i + 1;
+	queue->tx.req_prod_pvt = i + 1;
 
-	xennet_make_frags(skb, dev, tx);
+	xennet_make_frags(skb, queue, tx);
 	tx->size = skb->len;
 
-	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->tx, notify);
+	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->tx, notify);
 	if (notify)
-		notify_remote_via_irq(np->tx_irq);
+		notify_remote_via_irq(queue->tx_irq);
 
 	u64_stats_update_begin(&stats->syncp);
 	stats->tx_bytes += skb->len;
@@ -647,12 +702,12 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	u64_stats_update_end(&stats->syncp);
 
 	/* Note: It is not safe to access skb after xennet_tx_buf_gc()! */
-	xennet_tx_buf_gc(dev);
+	xennet_tx_buf_gc(queue);
 
-	if (!netfront_tx_slot_available(np))
-		netif_stop_queue(dev);
+	if (!netfront_tx_slot_available(queue))
+		netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id));
 
-	spin_unlock_irqrestore(&np->tx_lock, flags);
+	spin_unlock_irqrestore(&queue->tx_lock, flags);
 
 	return NETDEV_TX_OK;
 
@@ -665,32 +720,38 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 static int xennet_close(struct net_device *dev)
 {
 	struct netfront_info *np = netdev_priv(dev);
-	netif_stop_queue(np->netdev);
-	napi_disable(&np->napi);
+	unsigned int num_queues = dev->real_num_tx_queues;
+	unsigned int i;
+	struct netfront_queue *queue;
+	netif_tx_stop_all_queues(np->netdev);
+	for (i = 0; i < num_queues; ++i) {
+		queue = &np->queues[i];
+		napi_disable(&queue->napi);
+	}
 	return 0;
 }
 
-static void xennet_move_rx_slot(struct netfront_info *np, struct sk_buff *skb,
+static void xennet_move_rx_slot(struct netfront_queue *queue, struct sk_buff *skb,
 				grant_ref_t ref)
 {
-	int new = xennet_rxidx(np->rx.req_prod_pvt);
-
-	BUG_ON(np->rx_skbs[new]);
-	np->rx_skbs[new] = skb;
-	np->grant_rx_ref[new] = ref;
-	RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->id = new;
-	RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->gref = ref;
-	np->rx.req_prod_pvt++;
+	int new = xennet_rxidx(queue->rx.req_prod_pvt);
+
+	BUG_ON(queue->rx_skbs[new]);
+	queue->rx_skbs[new] = skb;
+	queue->grant_rx_ref[new] = ref;
+	RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->id = new;
+	RING_GET_REQUEST(&queue->rx, queue->rx.req_prod_pvt)->gref = ref;
+	queue->rx.req_prod_pvt++;
 }
 
-static int xennet_get_extras(struct netfront_info *np,
+static int xennet_get_extras(struct netfront_queue *queue,
 			     struct xen_netif_extra_info *extras,
 			     RING_IDX rp)
 
 {
 	struct xen_netif_extra_info *extra;
-	struct device *dev = &np->netdev->dev;
-	RING_IDX cons = np->rx.rsp_cons;
+	struct device *dev = &queue->info->netdev->dev;
+	RING_IDX cons = queue->rx.rsp_cons;
 	int err = 0;
 
 	do {
@@ -705,7 +766,7 @@ static int xennet_get_extras(struct netfront_info *np,
 		}
 
 		extra = (struct xen_netif_extra_info *)
-			RING_GET_RESPONSE(&np->rx, ++cons);
+			RING_GET_RESPONSE(&queue->rx, ++cons);
 
 		if (unlikely(!extra->type ||
 			     extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
@@ -718,33 +779,33 @@ static int xennet_get_extras(struct netfront_info *np,
 			       sizeof(*extra));
 		}
 
-		skb = xennet_get_rx_skb(np, cons);
-		ref = xennet_get_rx_ref(np, cons);
-		xennet_move_rx_slot(np, skb, ref);
+		skb = xennet_get_rx_skb(queue, cons);
+		ref = xennet_get_rx_ref(queue, cons);
+		xennet_move_rx_slot(queue, skb, ref);
 	} while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE);
 
-	np->rx.rsp_cons = cons;
+	queue->rx.rsp_cons = cons;
 	return err;
 }
 
-static int xennet_get_responses(struct netfront_info *np,
+static int xennet_get_responses(struct netfront_queue *queue,
 				struct netfront_rx_info *rinfo, RING_IDX rp,
 				struct sk_buff_head *list)
 {
 	struct xen_netif_rx_response *rx = &rinfo->rx;
 	struct xen_netif_extra_info *extras = rinfo->extras;
-	struct device *dev = &np->netdev->dev;
-	RING_IDX cons = np->rx.rsp_cons;
-	struct sk_buff *skb = xennet_get_rx_skb(np, cons);
-	grant_ref_t ref = xennet_get_rx_ref(np, cons);
+	struct device *dev = &queue->info->netdev->dev;
+	RING_IDX cons = queue->rx.rsp_cons;
+	struct sk_buff *skb = xennet_get_rx_skb(queue, cons);
+	grant_ref_t ref = xennet_get_rx_ref(queue, cons);
 	int max = MAX_SKB_FRAGS + (rx->status <= RX_COPY_THRESHOLD);
 	int slots = 1;
 	int err = 0;
 	unsigned long ret;
 
 	if (rx->flags & XEN_NETRXF_extra_info) {
-		err = xennet_get_extras(np, extras, rp);
-		cons = np->rx.rsp_cons;
+		err = xennet_get_extras(queue, extras, rp);
+		cons = queue->rx.rsp_cons;
 	}
 
 	for (;;) {
@@ -753,7 +814,7 @@ static int xennet_get_responses(struct netfront_info *np,
 			if (net_ratelimit())
 				dev_warn(dev, "rx->offset: %x, size: %u\n",
 					 rx->offset, rx->status);
-			xennet_move_rx_slot(np, skb, ref);
+			xennet_move_rx_slot(queue, skb, ref);
 			err = -EINVAL;
 			goto next;
 		}
@@ -774,7 +835,7 @@ static int xennet_get_responses(struct netfront_info *np,
 		ret = gnttab_end_foreign_access_ref(ref, 0);
 		BUG_ON(!ret);
 
-		gnttab_release_grant_reference(&np->gref_rx_head, ref);
+		gnttab_release_grant_reference(&queue->gref_rx_head, ref);
 
 		__skb_queue_tail(list, skb);
 
@@ -789,9 +850,9 @@ next:
 			break;
 		}
 
-		rx = RING_GET_RESPONSE(&np->rx, cons + slots);
-		skb = xennet_get_rx_skb(np, cons + slots);
-		ref = xennet_get_rx_ref(np, cons + slots);
+		rx = RING_GET_RESPONSE(&queue->rx, cons + slots);
+		skb = xennet_get_rx_skb(queue, cons + slots);
+		ref = xennet_get_rx_ref(queue, cons + slots);
 		slots++;
 	}
 
@@ -802,7 +863,7 @@ next:
 	}
 
 	if (unlikely(err))
-		np->rx.rsp_cons = cons + slots;
+		queue->rx.rsp_cons = cons + slots;
 
 	return err;
 }
@@ -836,17 +897,17 @@ static int xennet_set_skb_gso(struct sk_buff *skb,
 	return 0;
 }
 
-static RING_IDX xennet_fill_frags(struct netfront_info *np,
+static RING_IDX xennet_fill_frags(struct netfront_queue *queue,
 				  struct sk_buff *skb,
 				  struct sk_buff_head *list)
 {
 	struct skb_shared_info *shinfo = skb_shinfo(skb);
-	RING_IDX cons = np->rx.rsp_cons;
+	RING_IDX cons = queue->rx.rsp_cons;
 	struct sk_buff *nskb;
 
 	while ((nskb = __skb_dequeue(list))) {
 		struct xen_netif_rx_response *rx =
-			RING_GET_RESPONSE(&np->rx, ++cons);
+			RING_GET_RESPONSE(&queue->rx, ++cons);
 		skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0];
 
 		if (shinfo->nr_frags == MAX_SKB_FRAGS) {
@@ -879,7 +940,7 @@ static int checksum_setup(struct net_device *dev, struct sk_buff *skb)
 	 */
 	if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) {
 		struct netfront_info *np = netdev_priv(dev);
-		np->rx_gso_checksum_fixup++;
+		atomic_inc(&np->rx_gso_checksum_fixup);
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		recalculate_partial_csum = true;
 	}
@@ -891,11 +952,10 @@ static int checksum_setup(struct net_device *dev, struct sk_buff *skb)
 	return skb_checksum_setup(skb, recalculate_partial_csum);
 }
 
-static int handle_incoming_queue(struct net_device *dev,
+static int handle_incoming_queue(struct netfront_queue *queue,
 				 struct sk_buff_head *rxq)
 {
-	struct netfront_info *np = netdev_priv(dev);
-	struct netfront_stats *stats = this_cpu_ptr(np->stats);
+	struct netfront_stats *stats = this_cpu_ptr(queue->info->stats);
 	int packets_dropped = 0;
 	struct sk_buff *skb;
 
@@ -906,13 +966,13 @@ static int handle_incoming_queue(struct net_device *dev,
 			__pskb_pull_tail(skb, pull_to - skb_headlen(skb));
 
 		/* Ethernet work: Delayed to here as it peeks the header. */
-		skb->protocol = eth_type_trans(skb, dev);
+		skb->protocol = eth_type_trans(skb, queue->info->netdev);
 		skb_reset_network_header(skb);
 
-		if (checksum_setup(dev, skb)) {
+		if (checksum_setup(queue->info->netdev, skb)) {
 			kfree_skb(skb);
 			packets_dropped++;
-			dev->stats.rx_errors++;
+			queue->info->netdev->stats.rx_errors++;
 			continue;
 		}
 
@@ -922,7 +982,7 @@ static int handle_incoming_queue(struct net_device *dev,
 		u64_stats_update_end(&stats->syncp);
 
 		/* Pass it up. */
-		napi_gro_receive(&np->napi, skb);
+		napi_gro_receive(&queue->napi, skb);
 	}
 
 	return packets_dropped;
@@ -930,8 +990,8 @@ static int handle_incoming_queue(struct net_device *dev,
 
 static int xennet_poll(struct napi_struct *napi, int budget)
 {
-	struct netfront_info *np = container_of(napi, struct netfront_info, napi);
-	struct net_device *dev = np->netdev;
+	struct netfront_queue *queue = container_of(napi, struct netfront_queue, napi);
+	struct net_device *dev = queue->info->netdev;
 	struct sk_buff *skb;
 	struct netfront_rx_info rinfo;
 	struct xen_netif_rx_response *rx = &rinfo.rx;
@@ -944,29 +1004,29 @@ static int xennet_poll(struct napi_struct *napi, int budget)
 	unsigned long flags;
 	int err;
 
-	spin_lock(&np->rx_lock);
+	spin_lock(&queue->rx_lock);
 
 	skb_queue_head_init(&rxq);
 	skb_queue_head_init(&errq);
 	skb_queue_head_init(&tmpq);
 
-	rp = np->rx.sring->rsp_prod;
+	rp = queue->rx.sring->rsp_prod;
 	rmb(); /* Ensure we see queued responses up to 'rp'. */
 
-	i = np->rx.rsp_cons;
+	i = queue->rx.rsp_cons;
 	work_done = 0;
 	while ((i != rp) && (work_done < budget)) {
-		memcpy(rx, RING_GET_RESPONSE(&np->rx, i), sizeof(*rx));
+		memcpy(rx, RING_GET_RESPONSE(&queue->rx, i), sizeof(*rx));
 		memset(extras, 0, sizeof(rinfo.extras));
 
-		err = xennet_get_responses(np, &rinfo, rp, &tmpq);
+		err = xennet_get_responses(queue, &rinfo, rp, &tmpq);
 
 		if (unlikely(err)) {
 err:
 			while ((skb = __skb_dequeue(&tmpq)))
 				__skb_queue_tail(&errq, skb);
 			dev->stats.rx_errors++;
-			i = np->rx.rsp_cons;
+			i = queue->rx.rsp_cons;
 			continue;
 		}
 
@@ -978,7 +1038,7 @@ err:
 
 			if (unlikely(xennet_set_skb_gso(skb, gso))) {
 				__skb_queue_head(&tmpq, skb);
-				np->rx.rsp_cons += skb_queue_len(&tmpq);
+				queue->rx.rsp_cons += skb_queue_len(&tmpq);
 				goto err;
 			}
 		}
@@ -992,7 +1052,7 @@ err:
 		skb->data_len = rx->status;
 		skb->len += rx->status;
 
-		i = xennet_fill_frags(np, skb, &tmpq);
+		i = xennet_fill_frags(queue, skb, &tmpq);
 
 		if (rx->flags & XEN_NETRXF_csum_blank)
 			skb->ip_summed = CHECKSUM_PARTIAL;
@@ -1001,22 +1061,22 @@ err:
 
 		__skb_queue_tail(&rxq, skb);
 
-		np->rx.rsp_cons = ++i;
+		queue->rx.rsp_cons = ++i;
 		work_done++;
 	}
 
 	__skb_queue_purge(&errq);
 
-	work_done -= handle_incoming_queue(dev, &rxq);
+	work_done -= handle_incoming_queue(queue, &rxq);
 
 	/* If we get a callback with very few responses, reduce fill target. */
 	/* NB. Note exponential increase, linear decrease. */
-	if (((np->rx.req_prod_pvt - np->rx.sring->rsp_prod) >
-	     ((3*np->rx_target) / 4)) &&
-	    (--np->rx_target < np->rx_min_target))
-		np->rx_target = np->rx_min_target;
+	if (((queue->rx.req_prod_pvt - queue->rx.sring->rsp_prod) >
+	     ((3*queue->rx_target) / 4)) &&
+	    (--queue->rx_target < queue->rx_min_target))
+		queue->rx_target = queue->rx_min_target;
 
-	xennet_alloc_rx_buffers(dev);
+	xennet_alloc_rx_buffers(queue);
 
 	if (work_done < budget) {
 		int more_to_do = 0;
@@ -1025,14 +1085,14 @@ err:
 
 		local_irq_save(flags);
 
-		RING_FINAL_CHECK_FOR_RESPONSES(&np->rx, more_to_do);
+		RING_FINAL_CHECK_FOR_RESPONSES(&queue->rx, more_to_do);
 		if (!more_to_do)
 			__napi_complete(napi);
 
 		local_irq_restore(flags);
 	}
 
-	spin_unlock(&np->rx_lock);
+	spin_unlock(&queue->rx_lock);
 
 	return work_done;
 }
@@ -1080,43 +1140,43 @@ static struct rtnl_link_stats64 *xennet_get_stats64(struct net_device *dev,
 	return tot;
 }
 
-static void xennet_release_tx_bufs(struct netfront_info *np)
+static void xennet_release_tx_bufs(struct netfront_queue *queue)
 {
 	struct sk_buff *skb;
 	int i;
 
 	for (i = 0; i < NET_TX_RING_SIZE; i++) {
 		/* Skip over entries which are actually freelist references */
-		if (skb_entry_is_link(&np->tx_skbs[i]))
+		if (skb_entry_is_link(&queue->tx_skbs[i]))
 			continue;
 
-		skb = np->tx_skbs[i].skb;
-		get_page(np->grant_tx_page[i]);
-		gnttab_end_foreign_access(np->grant_tx_ref[i],
+		skb = queue->tx_skbs[i].skb;
+		get_page(queue->grant_tx_page[i]);
+		gnttab_end_foreign_access(queue->grant_tx_ref[i],
 					  GNTMAP_readonly,
-					  (unsigned long)page_address(np->grant_tx_page[i]));
-		np->grant_tx_page[i] = NULL;
-		np->grant_tx_ref[i] = GRANT_INVALID_REF;
-		add_id_to_freelist(&np->tx_skb_freelist, np->tx_skbs, i);
+					  (unsigned long)page_address(queue->grant_tx_page[i]));
+		queue->grant_tx_page[i] = NULL;
+		queue->grant_tx_ref[i] = GRANT_INVALID_REF;
+		add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, i);
 		dev_kfree_skb_irq(skb);
 	}
 }
 
-static void xennet_release_rx_bufs(struct netfront_info *np)
+static void xennet_release_rx_bufs(struct netfront_queue *queue)
 {
 	int id, ref;
 
-	spin_lock_bh(&np->rx_lock);
+	spin_lock_bh(&queue->rx_lock);
 
 	for (id = 0; id < NET_RX_RING_SIZE; id++) {
 		struct sk_buff *skb;
 		struct page *page;
 
-		skb = np->rx_skbs[id];
+		skb = queue->rx_skbs[id];
 		if (!skb)
 			continue;
 
-		ref = np->grant_rx_ref[id];
+		ref = queue->grant_rx_ref[id];
 		if (ref == GRANT_INVALID_REF)
 			continue;
 
@@ -1128,21 +1188,28 @@ static void xennet_release_rx_bufs(struct netfront_info *np)
 		get_page(page);
 		gnttab_end_foreign_access(ref, 0,
 					  (unsigned long)page_address(page));
-		np->grant_rx_ref[id] = GRANT_INVALID_REF;
+		queue->grant_rx_ref[id] = GRANT_INVALID_REF;
 
 		kfree_skb(skb);
 	}
 
-	spin_unlock_bh(&np->rx_lock);
+	spin_unlock_bh(&queue->rx_lock);
 }
 
 static void xennet_uninit(struct net_device *dev)
 {
 	struct netfront_info *np = netdev_priv(dev);
-	xennet_release_tx_bufs(np);
-	xennet_release_rx_bufs(np);
-	gnttab_free_grant_references(np->gref_tx_head);
-	gnttab_free_grant_references(np->gref_rx_head);
+	unsigned int num_queues = dev->real_num_tx_queues;
+	struct netfront_queue *queue;
+	unsigned int i;
+
+	for (i = 0; i < num_queues; ++i) {
+		queue = &np->queues[i];
+		xennet_release_tx_bufs(queue);
+		xennet_release_rx_bufs(queue);
+		gnttab_free_grant_references(queue->gref_tx_head);
+		gnttab_free_grant_references(queue->gref_rx_head);
+	}
 }
 
 static netdev_features_t xennet_fix_features(struct net_device *dev,
@@ -1203,25 +1270,24 @@ static int xennet_set_features(struct net_device *dev,
 
 static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id)
 {
-	struct netfront_info *np = dev_id;
-	struct net_device *dev = np->netdev;
+	struct netfront_queue *queue = dev_id;
 	unsigned long flags;
 
-	spin_lock_irqsave(&np->tx_lock, flags);
-	xennet_tx_buf_gc(dev);
-	spin_unlock_irqrestore(&np->tx_lock, flags);
+	spin_lock_irqsave(&queue->tx_lock, flags);
+	xennet_tx_buf_gc(queue);
+	spin_unlock_irqrestore(&queue->tx_lock, flags);
 
 	return IRQ_HANDLED;
 }
 
 static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id)
 {
-	struct netfront_info *np = dev_id;
-	struct net_device *dev = np->netdev;
+	struct netfront_queue *queue = dev_id;
+	struct net_device *dev = queue->info->netdev;
 
 	if (likely(netif_carrier_ok(dev) &&
-		   RING_HAS_UNCONSUMED_RESPONSES(&np->rx)))
-			napi_schedule(&np->napi);
+		   RING_HAS_UNCONSUMED_RESPONSES(&queue->rx)))
+		napi_schedule(&queue->napi);
 
 	return IRQ_HANDLED;
 }
@@ -1236,7 +1302,12 @@ static irqreturn_t xennet_interrupt(int irq, void *dev_id)
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void xennet_poll_controller(struct net_device *dev)
 {
-	xennet_interrupt(0, dev);
+	/* Poll each queue */
+	struct netfront_info *info = netdev_priv(dev);
+	unsigned int num_queues = dev->real_num_tx_queues;
+	unsigned int i;
+	for (i = 0; i < num_queues; ++i)
+		xennet_interrupt(0, &info->queues[i]);
 }
 #endif
 
@@ -1251,6 +1322,7 @@ static const struct net_device_ops xennet_netdev_ops = {
 	.ndo_validate_addr   = eth_validate_addr,
 	.ndo_fix_features    = xennet_fix_features,
 	.ndo_set_features    = xennet_set_features,
+	.ndo_select_queue    = xennet_select_queue,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller = xennet_poll_controller,
 #endif
@@ -1258,66 +1330,30 @@ static const struct net_device_ops xennet_netdev_ops = {
 
 static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 {
-	int i, err;
+	int err;
 	struct net_device *netdev;
 	struct netfront_info *np;
 
-	netdev = alloc_etherdev(sizeof(struct netfront_info));
+	netdev = alloc_etherdev_mq(sizeof(struct netfront_info), xennet_max_queues);
 	if (!netdev)
 		return ERR_PTR(-ENOMEM);
 
 	np                   = netdev_priv(netdev);
 	np->xbdev            = dev;
 
-	spin_lock_init(&np->tx_lock);
-	spin_lock_init(&np->rx_lock);
-
-	skb_queue_head_init(&np->rx_batch);
-	np->rx_target     = RX_DFL_MIN_TARGET;
-	np->rx_min_target = RX_DFL_MIN_TARGET;
-	np->rx_max_target = RX_MAX_TARGET;
-
-	init_timer(&np->rx_refill_timer);
-	np->rx_refill_timer.data = (unsigned long)netdev;
-	np->rx_refill_timer.function = rx_refill_timeout;
+	/* No need to use rtnl_lock() before the call below as it
+	 * happens before register_netdev().
+	 */
+	netif_set_real_num_tx_queues(netdev, 0);
+	np->queues = NULL;
 
 	err = -ENOMEM;
 	np->stats = netdev_alloc_pcpu_stats(struct netfront_stats);
 	if (np->stats == NULL)
 		goto exit;
 
-	/* Initialise tx_skbs as a free chain containing every entry. */
-	np->tx_skb_freelist = 0;
-	for (i = 0; i < NET_TX_RING_SIZE; i++) {
-		skb_entry_set_link(&np->tx_skbs[i], i+1);
-		np->grant_tx_ref[i] = GRANT_INVALID_REF;
-		np->grant_tx_page[i] = NULL;
-	}
-
-	/* Clear out rx_skbs */
-	for (i = 0; i < NET_RX_RING_SIZE; i++) {
-		np->rx_skbs[i] = NULL;
-		np->grant_rx_ref[i] = GRANT_INVALID_REF;
-	}
-
-	/* A grant for every tx ring slot */
-	if (gnttab_alloc_grant_references(TX_MAX_TARGET,
-					  &np->gref_tx_head) < 0) {
-		pr_alert("can't alloc tx grant refs\n");
-		err = -ENOMEM;
-		goto exit_free_stats;
-	}
-	/* A grant for every rx ring slot */
-	if (gnttab_alloc_grant_references(RX_MAX_TARGET,
-					  &np->gref_rx_head) < 0) {
-		pr_alert("can't alloc rx grant refs\n");
-		err = -ENOMEM;
-		goto exit_free_tx;
-	}
-
 	netdev->netdev_ops	= &xennet_netdev_ops;
 
-	netif_napi_add(netdev, &np->napi, xennet_poll, 64);
 	netdev->features        = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
 				  NETIF_F_GSO_ROBUST;
 	netdev->hw_features	= NETIF_F_SG |
@@ -1332,7 +1368,7 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev)
          */
 	netdev->features |= netdev->hw_features;
 
-	SET_ETHTOOL_OPS(netdev, &xennet_ethtool_ops);
+	netdev->ethtool_ops = &xennet_ethtool_ops;
 	SET_NETDEV_DEV(netdev, &dev->dev);
 
 	netif_set_gso_max_size(netdev, XEN_NETIF_MAX_TX_SIZE - MAX_TCP_HEADER);
@@ -1343,10 +1379,6 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 
 	return netdev;
 
- exit_free_tx:
-	gnttab_free_grant_references(np->gref_tx_head);
- exit_free_stats:
-	free_percpu(np->stats);
  exit:
 	free_netdev(netdev);
 	return ERR_PTR(err);
@@ -1404,30 +1436,37 @@ static void xennet_end_access(int ref, void *page)
 
 static void xennet_disconnect_backend(struct netfront_info *info)
 {
-	/* Stop old i/f to prevent errors whilst we rebuild the state. */
-	spin_lock_bh(&info->rx_lock);
-	spin_lock_irq(&info->tx_lock);
-	netif_carrier_off(info->netdev);
-	spin_unlock_irq(&info->tx_lock);
-	spin_unlock_bh(&info->rx_lock);
-
-	if (info->tx_irq && (info->tx_irq == info->rx_irq))
-		unbind_from_irqhandler(info->tx_irq, info);
-	if (info->tx_irq && (info->tx_irq != info->rx_irq)) {
-		unbind_from_irqhandler(info->tx_irq, info);
-		unbind_from_irqhandler(info->rx_irq, info);
-	}
-	info->tx_evtchn = info->rx_evtchn = 0;
-	info->tx_irq = info->rx_irq = 0;
+	unsigned int i = 0;
+	unsigned int num_queues = info->netdev->real_num_tx_queues;
+
+	for (i = 0; i < num_queues; ++i) {
+		struct netfront_queue *queue = &info->queues[i];
+
+		/* Stop old i/f to prevent errors whilst we rebuild the state. */
+		spin_lock_bh(&queue->rx_lock);
+		spin_lock_irq(&queue->tx_lock);
+		netif_carrier_off(queue->info->netdev);
+		spin_unlock_irq(&queue->tx_lock);
+		spin_unlock_bh(&queue->rx_lock);
+
+		if (queue->tx_irq && (queue->tx_irq == queue->rx_irq))
+			unbind_from_irqhandler(queue->tx_irq, queue);
+		if (queue->tx_irq && (queue->tx_irq != queue->rx_irq)) {
+			unbind_from_irqhandler(queue->tx_irq, queue);
+			unbind_from_irqhandler(queue->rx_irq, queue);
+		}
+		queue->tx_evtchn = queue->rx_evtchn = 0;
+		queue->tx_irq = queue->rx_irq = 0;
 
-	/* End access and free the pages */
-	xennet_end_access(info->tx_ring_ref, info->tx.sring);
-	xennet_end_access(info->rx_ring_ref, info->rx.sring);
+		/* End access and free the pages */
+		xennet_end_access(queue->tx_ring_ref, queue->tx.sring);
+		xennet_end_access(queue->rx_ring_ref, queue->rx.sring);
 
-	info->tx_ring_ref = GRANT_INVALID_REF;
-	info->rx_ring_ref = GRANT_INVALID_REF;
-	info->tx.sring = NULL;
-	info->rx.sring = NULL;
+		queue->tx_ring_ref = GRANT_INVALID_REF;
+		queue->rx_ring_ref = GRANT_INVALID_REF;
+		queue->tx.sring = NULL;
+		queue->rx.sring = NULL;
+	}
 }
 
 /**
@@ -1468,100 +1507,86 @@ static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[])
 	return 0;
 }
 
-static int setup_netfront_single(struct netfront_info *info)
+static int setup_netfront_single(struct netfront_queue *queue)
 {
 	int err;
 
-	err = xenbus_alloc_evtchn(info->xbdev, &info->tx_evtchn);
+	err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn);
 	if (err < 0)
 		goto fail;
 
-	err = bind_evtchn_to_irqhandler(info->tx_evtchn,
+	err = bind_evtchn_to_irqhandler(queue->tx_evtchn,
 					xennet_interrupt,
-					0, info->netdev->name, info);
+					0, queue->info->netdev->name, queue);
 	if (err < 0)
 		goto bind_fail;
-	info->rx_evtchn = info->tx_evtchn;
-	info->rx_irq = info->tx_irq = err;
+	queue->rx_evtchn = queue->tx_evtchn;
+	queue->rx_irq = queue->tx_irq = err;
 
 	return 0;
 
 bind_fail:
-	xenbus_free_evtchn(info->xbdev, info->tx_evtchn);
-	info->tx_evtchn = 0;
+	xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn);
+	queue->tx_evtchn = 0;
 fail:
 	return err;
 }
 
-static int setup_netfront_split(struct netfront_info *info)
+static int setup_netfront_split(struct netfront_queue *queue)
 {
 	int err;
 
-	err = xenbus_alloc_evtchn(info->xbdev, &info->tx_evtchn);
+	err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->tx_evtchn);
 	if (err < 0)
 		goto fail;
-	err = xenbus_alloc_evtchn(info->xbdev, &info->rx_evtchn);
+	err = xenbus_alloc_evtchn(queue->info->xbdev, &queue->rx_evtchn);
 	if (err < 0)
 		goto alloc_rx_evtchn_fail;
 
-	snprintf(info->tx_irq_name, sizeof(info->tx_irq_name),
-		 "%s-tx", info->netdev->name);
-	err = bind_evtchn_to_irqhandler(info->tx_evtchn,
+	snprintf(queue->tx_irq_name, sizeof(queue->tx_irq_name),
+		 "%s-tx", queue->name);
+	err = bind_evtchn_to_irqhandler(queue->tx_evtchn,
 					xennet_tx_interrupt,
-					0, info->tx_irq_name, info);
+					0, queue->tx_irq_name, queue);
 	if (err < 0)
 		goto bind_tx_fail;
-	info->tx_irq = err;
+	queue->tx_irq = err;
 
-	snprintf(info->rx_irq_name, sizeof(info->rx_irq_name),
-		 "%s-rx", info->netdev->name);
-	err = bind_evtchn_to_irqhandler(info->rx_evtchn,
+	snprintf(queue->rx_irq_name, sizeof(queue->rx_irq_name),
+		 "%s-rx", queue->name);
+	err = bind_evtchn_to_irqhandler(queue->rx_evtchn,
 					xennet_rx_interrupt,
-					0, info->rx_irq_name, info);
+					0, queue->rx_irq_name, queue);
 	if (err < 0)
 		goto bind_rx_fail;
-	info->rx_irq = err;
+	queue->rx_irq = err;
 
 	return 0;
 
 bind_rx_fail:
-	unbind_from_irqhandler(info->tx_irq, info);
-	info->tx_irq = 0;
+	unbind_from_irqhandler(queue->tx_irq, queue);
+	queue->tx_irq = 0;
 bind_tx_fail:
-	xenbus_free_evtchn(info->xbdev, info->rx_evtchn);
-	info->rx_evtchn = 0;
+	xenbus_free_evtchn(queue->info->xbdev, queue->rx_evtchn);
+	queue->rx_evtchn = 0;
 alloc_rx_evtchn_fail:
-	xenbus_free_evtchn(info->xbdev, info->tx_evtchn);
-	info->tx_evtchn = 0;
+	xenbus_free_evtchn(queue->info->xbdev, queue->tx_evtchn);
+	queue->tx_evtchn = 0;
 fail:
 	return err;
 }
 
-static int setup_netfront(struct xenbus_device *dev, struct netfront_info *info)
+static int setup_netfront(struct xenbus_device *dev,
+			struct netfront_queue *queue, unsigned int feature_split_evtchn)
 {
 	struct xen_netif_tx_sring *txs;
 	struct xen_netif_rx_sring *rxs;
 	int err;
-	struct net_device *netdev = info->netdev;
-	unsigned int feature_split_evtchn;
-
-	info->tx_ring_ref = GRANT_INVALID_REF;
-	info->rx_ring_ref = GRANT_INVALID_REF;
-	info->rx.sring = NULL;
-	info->tx.sring = NULL;
-	netdev->irq = 0;
-
-	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
-			   "feature-split-event-channels", "%u",
-			   &feature_split_evtchn);
-	if (err < 0)
-		feature_split_evtchn = 0;
 
-	err = xen_net_read_mac(dev, netdev->dev_addr);
-	if (err) {
-		xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
-		goto fail;
-	}
+	queue->tx_ring_ref = GRANT_INVALID_REF;
+	queue->rx_ring_ref = GRANT_INVALID_REF;
+	queue->rx.sring = NULL;
+	queue->tx.sring = NULL;
 
 	txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
 	if (!txs) {
@@ -1570,13 +1595,13 @@ static int setup_netfront(struct xenbus_device *dev, struct netfront_info *info)
 		goto fail;
 	}
 	SHARED_RING_INIT(txs);
-	FRONT_RING_INIT(&info->tx, txs, PAGE_SIZE);
+	FRONT_RING_INIT(&queue->tx, txs, PAGE_SIZE);
 
 	err = xenbus_grant_ring(dev, virt_to_mfn(txs));
 	if (err < 0)
 		goto grant_tx_ring_fail;
+	queue->tx_ring_ref = err;
 
-	info->tx_ring_ref = err;
 	rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
 	if (!rxs) {
 		err = -ENOMEM;
@@ -1584,21 +1609,21 @@ static int setup_netfront(struct xenbus_device *dev, struct netfront_info *info)
 		goto alloc_rx_ring_fail;
 	}
 	SHARED_RING_INIT(rxs);
-	FRONT_RING_INIT(&info->rx, rxs, PAGE_SIZE);
+	FRONT_RING_INIT(&queue->rx, rxs, PAGE_SIZE);
 
 	err = xenbus_grant_ring(dev, virt_to_mfn(rxs));
 	if (err < 0)
 		goto grant_rx_ring_fail;
-	info->rx_ring_ref = err;
+	queue->rx_ring_ref = err;
 
 	if (feature_split_evtchn)
-		err = setup_netfront_split(info);
+		err = setup_netfront_split(queue);
 	/* setup single event channel if
 	 *  a) feature-split-event-channels == 0
 	 *  b) feature-split-event-channels == 1 but failed to setup
 	 */
 	if (!feature_split_evtchn || (feature_split_evtchn && err))
-		err = setup_netfront_single(info);
+		err = setup_netfront_single(queue);
 
 	if (err)
 		goto alloc_evtchn_fail;
@@ -1609,17 +1634,225 @@ static int setup_netfront(struct xenbus_device *dev, struct netfront_info *info)
 	 * granted pages because backend is not accessing it at this point.
 	 */
 alloc_evtchn_fail:
-	gnttab_end_foreign_access_ref(info->rx_ring_ref, 0);
+	gnttab_end_foreign_access_ref(queue->rx_ring_ref, 0);
 grant_rx_ring_fail:
 	free_page((unsigned long)rxs);
 alloc_rx_ring_fail:
-	gnttab_end_foreign_access_ref(info->tx_ring_ref, 0);
+	gnttab_end_foreign_access_ref(queue->tx_ring_ref, 0);
 grant_tx_ring_fail:
 	free_page((unsigned long)txs);
 fail:
 	return err;
 }
 
+/* Queue-specific initialisation
+ * This used to be done in xennet_create_dev() but must now
+ * be run per-queue.
+ */
+static int xennet_init_queue(struct netfront_queue *queue)
+{
+	unsigned short i;
+	int err = 0;
+
+	spin_lock_init(&queue->tx_lock);
+	spin_lock_init(&queue->rx_lock);
+
+	skb_queue_head_init(&queue->rx_batch);
+	queue->rx_target     = RX_DFL_MIN_TARGET;
+	queue->rx_min_target = RX_DFL_MIN_TARGET;
+	queue->rx_max_target = RX_MAX_TARGET;
+
+	init_timer(&queue->rx_refill_timer);
+	queue->rx_refill_timer.data = (unsigned long)queue;
+	queue->rx_refill_timer.function = rx_refill_timeout;
+
+	snprintf(queue->name, sizeof(queue->name), "%s-q%u",
+		 queue->info->netdev->name, queue->id);
+
+	/* Initialise tx_skbs as a free chain containing every entry. */
+	queue->tx_skb_freelist = 0;
+	for (i = 0; i < NET_TX_RING_SIZE; i++) {
+		skb_entry_set_link(&queue->tx_skbs[i], i+1);
+		queue->grant_tx_ref[i] = GRANT_INVALID_REF;
+		queue->grant_tx_page[i] = NULL;
+	}
+
+	/* Clear out rx_skbs */
+	for (i = 0; i < NET_RX_RING_SIZE; i++) {
+		queue->rx_skbs[i] = NULL;
+		queue->grant_rx_ref[i] = GRANT_INVALID_REF;
+	}
+
+	/* A grant for every tx ring slot */
+	if (gnttab_alloc_grant_references(TX_MAX_TARGET,
+					  &queue->gref_tx_head) < 0) {
+		pr_alert("can't alloc tx grant refs\n");
+		err = -ENOMEM;
+		goto exit;
+	}
+
+	/* A grant for every rx ring slot */
+	if (gnttab_alloc_grant_references(RX_MAX_TARGET,
+					  &queue->gref_rx_head) < 0) {
+		pr_alert("can't alloc rx grant refs\n");
+		err = -ENOMEM;
+		goto exit_free_tx;
+	}
+
+	return 0;
+
+ exit_free_tx:
+	gnttab_free_grant_references(queue->gref_tx_head);
+ exit:
+	return err;
+}
+
+static int write_queue_xenstore_keys(struct netfront_queue *queue,
+			   struct xenbus_transaction *xbt, int write_hierarchical)
+{
+	/* Write the queue-specific keys into XenStore in the traditional
+	 * way for a single queue, or in a queue subkeys for multiple
+	 * queues.
+	 */
+	struct xenbus_device *dev = queue->info->xbdev;
+	int err;
+	const char *message;
+	char *path;
+	size_t pathsize;
+
+	/* Choose the correct place to write the keys */
+	if (write_hierarchical) {
+		pathsize = strlen(dev->nodename) + 10;
+		path = kzalloc(pathsize, GFP_KERNEL);
+		if (!path) {
+			err = -ENOMEM;
+			message = "out of memory while writing ring references";
+			goto error;
+		}
+		snprintf(path, pathsize, "%s/queue-%u",
+				dev->nodename, queue->id);
+	} else {
+		path = (char *)dev->nodename;
+	}
+
+	/* Write ring references */
+	err = xenbus_printf(*xbt, path, "tx-ring-ref", "%u",
+			queue->tx_ring_ref);
+	if (err) {
+		message = "writing tx-ring-ref";
+		goto error;
+	}
+
+	err = xenbus_printf(*xbt, path, "rx-ring-ref", "%u",
+			queue->rx_ring_ref);
+	if (err) {
+		message = "writing rx-ring-ref";
+		goto error;
+	}
+
+	/* Write event channels; taking into account both shared
+	 * and split event channel scenarios.
+	 */
+	if (queue->tx_evtchn == queue->rx_evtchn) {
+		/* Shared event channel */
+		err = xenbus_printf(*xbt, path,
+				"event-channel", "%u", queue->tx_evtchn);
+		if (err) {
+			message = "writing event-channel";
+			goto error;
+		}
+	} else {
+		/* Split event channels */
+		err = xenbus_printf(*xbt, path,
+				"event-channel-tx", "%u", queue->tx_evtchn);
+		if (err) {
+			message = "writing event-channel-tx";
+			goto error;
+		}
+
+		err = xenbus_printf(*xbt, path,
+				"event-channel-rx", "%u", queue->rx_evtchn);
+		if (err) {
+			message = "writing event-channel-rx";
+			goto error;
+		}
+	}
+
+	if (write_hierarchical)
+		kfree(path);
+	return 0;
+
+error:
+	if (write_hierarchical)
+		kfree(path);
+	xenbus_dev_fatal(dev, err, "%s", message);
+	return err;
+}
+
+static void xennet_destroy_queues(struct netfront_info *info)
+{
+	unsigned int i;
+
+	rtnl_lock();
+
+	for (i = 0; i < info->netdev->real_num_tx_queues; i++) {
+		struct netfront_queue *queue = &info->queues[i];
+
+		if (netif_running(info->netdev))
+			napi_disable(&queue->napi);
+		netif_napi_del(&queue->napi);
+	}
+
+	rtnl_unlock();
+
+	kfree(info->queues);
+	info->queues = NULL;
+}
+
+static int xennet_create_queues(struct netfront_info *info,
+				unsigned int num_queues)
+{
+	unsigned int i;
+	int ret;
+
+	info->queues = kcalloc(num_queues, sizeof(struct netfront_queue),
+			       GFP_KERNEL);
+	if (!info->queues)
+		return -ENOMEM;
+
+	rtnl_lock();
+
+	for (i = 0; i < num_queues; i++) {
+		struct netfront_queue *queue = &info->queues[i];
+
+		queue->id = i;
+		queue->info = info;
+
+		ret = xennet_init_queue(queue);
+		if (ret < 0) {
+			dev_warn(&info->netdev->dev, "only created %d queues\n",
+				 num_queues);
+			num_queues = i;
+			break;
+		}
+
+		netif_napi_add(queue->info->netdev, &queue->napi,
+			       xennet_poll, 64);
+		if (netif_running(info->netdev))
+			napi_enable(&queue->napi);
+	}
+
+	netif_set_real_num_tx_queues(info->netdev, num_queues);
+
+	rtnl_unlock();
+
+	if (num_queues == 0) {
+		dev_err(&info->netdev->dev, "no queues\n");
+		return -EINVAL;
+	}
+	return 0;
+}
+
 /* Common code used when first setting up, and when resuming. */
 static int talk_to_netback(struct xenbus_device *dev,
 			   struct netfront_info *info)
@@ -1627,11 +1860,61 @@ static int talk_to_netback(struct xenbus_device *dev,
 	const char *message;
 	struct xenbus_transaction xbt;
 	int err;
+	unsigned int feature_split_evtchn;
+	unsigned int i = 0;
+	unsigned int max_queues = 0;
+	struct netfront_queue *queue = NULL;
+	unsigned int num_queues = 1;
 
-	/* Create shared ring, alloc event channel. */
-	err = setup_netfront(dev, info);
-	if (err)
+	info->netdev->irq = 0;
+
+	/* Check if backend supports multiple queues */
+	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
+			   "multi-queue-max-queues", "%u", &max_queues);
+	if (err < 0)
+		max_queues = 1;
+	num_queues = min(max_queues, xennet_max_queues);
+
+	/* Check feature-split-event-channels */
+	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
+			   "feature-split-event-channels", "%u",
+			   &feature_split_evtchn);
+	if (err < 0)
+		feature_split_evtchn = 0;
+
+	/* Read mac addr. */
+	err = xen_net_read_mac(dev, info->netdev->dev_addr);
+	if (err) {
+		xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
 		goto out;
+	}
+
+	if (info->queues)
+		xennet_destroy_queues(info);
+
+	err = xennet_create_queues(info, num_queues);
+	if (err < 0)
+		goto destroy_ring;
+
+	/* Create shared ring, alloc event channel -- for each queue */
+	for (i = 0; i < num_queues; ++i) {
+		queue = &info->queues[i];
+		err = setup_netfront(dev, queue, feature_split_evtchn);
+		if (err) {
+			/* setup_netfront() will tidy up the current
+			 * queue on error, but we need to clean up
+			 * those already allocated.
+			 */
+			if (i > 0) {
+				rtnl_lock();
+				netif_set_real_num_tx_queues(info->netdev, i);
+				rtnl_unlock();
+				goto destroy_ring;
+			} else {
+				goto out;
+			}
+		}
+	}
 
 again:
 	err = xenbus_transaction_start(&xbt);
@@ -1640,41 +1923,29 @@ again:
 		goto destroy_ring;
 	}
 
-	err = xenbus_printf(xbt, dev->nodename, "tx-ring-ref", "%u",
-			    info->tx_ring_ref);
-	if (err) {
-		message = "writing tx ring-ref";
-		goto abort_transaction;
-	}
-	err = xenbus_printf(xbt, dev->nodename, "rx-ring-ref", "%u",
-			    info->rx_ring_ref);
-	if (err) {
-		message = "writing rx ring-ref";
-		goto abort_transaction;
-	}
-
-	if (info->tx_evtchn == info->rx_evtchn) {
-		err = xenbus_printf(xbt, dev->nodename,
-				    "event-channel", "%u", info->tx_evtchn);
-		if (err) {
-			message = "writing event-channel";
-			goto abort_transaction;
-		}
+	if (num_queues == 1) {
+		err = write_queue_xenstore_keys(&info->queues[0], &xbt, 0); /* flat */
+		if (err)
+			goto abort_transaction_no_dev_fatal;
 	} else {
-		err = xenbus_printf(xbt, dev->nodename,
-				    "event-channel-tx", "%u", info->tx_evtchn);
+		/* Write the number of queues */
+		err = xenbus_printf(xbt, dev->nodename, "multi-queue-num-queues",
+				    "%u", num_queues);
 		if (err) {
-			message = "writing event-channel-tx";
-			goto abort_transaction;
+			message = "writing multi-queue-num-queues";
+			goto abort_transaction_no_dev_fatal;
 		}
-		err = xenbus_printf(xbt, dev->nodename,
-				    "event-channel-rx", "%u", info->rx_evtchn);
-		if (err) {
-			message = "writing event-channel-rx";
-			goto abort_transaction;
+
+		/* Write the keys for each queue */
+		for (i = 0; i < num_queues; ++i) {
+			queue = &info->queues[i];
+			err = write_queue_xenstore_keys(queue, &xbt, 1); /* hierarchical */
+			if (err)
+				goto abort_transaction_no_dev_fatal;
 		}
 	}
 
+	/* The remaining keys are not queue-specific */
 	err = xenbus_printf(xbt, dev->nodename, "request-rx-copy", "%u",
 			    1);
 	if (err) {
@@ -1724,10 +1995,16 @@ again:
 	return 0;
 
  abort_transaction:
-	xenbus_transaction_end(xbt, 1);
 	xenbus_dev_fatal(dev, err, "%s", message);
+abort_transaction_no_dev_fatal:
+	xenbus_transaction_end(xbt, 1);
  destroy_ring:
 	xennet_disconnect_backend(info);
+	kfree(info->queues);
+	info->queues = NULL;
+	rtnl_lock();
+	netif_set_real_num_tx_queues(info->netdev, 0);
+	rtnl_lock();
  out:
 	return err;
 }
@@ -1735,11 +2012,14 @@ again:
 static int xennet_connect(struct net_device *dev)
 {
 	struct netfront_info *np = netdev_priv(dev);
+	unsigned int num_queues = 0;
 	int i, requeue_idx, err;
 	struct sk_buff *skb;
 	grant_ref_t ref;
 	struct xen_netif_rx_request *req;
 	unsigned int feature_rx_copy;
+	unsigned int j = 0;
+	struct netfront_queue *queue = NULL;
 
 	err = xenbus_scanf(XBT_NIL, np->xbdev->otherend,
 			   "feature-rx-copy", "%u", &feature_rx_copy);
@@ -1756,40 +2036,47 @@ static int xennet_connect(struct net_device *dev)
 	if (err)
 		return err;
 
+	/* talk_to_netback() sets the correct number of queues */
+	num_queues = dev->real_num_tx_queues;
+
 	rtnl_lock();
 	netdev_update_features(dev);
 	rtnl_unlock();
 
-	spin_lock_bh(&np->rx_lock);
-	spin_lock_irq(&np->tx_lock);
+	/* By now, the queue structures have been set up */
+	for (j = 0; j < num_queues; ++j) {
+		queue = &np->queues[j];
+		spin_lock_bh(&queue->rx_lock);
+		spin_lock_irq(&queue->tx_lock);
 
-	/* Step 1: Discard all pending TX packet fragments. */
-	xennet_release_tx_bufs(np);
+		/* Step 1: Discard all pending TX packet fragments. */
+		xennet_release_tx_bufs(queue);
 
-	/* Step 2: Rebuild the RX buffer freelist and the RX ring itself. */
-	for (requeue_idx = 0, i = 0; i < NET_RX_RING_SIZE; i++) {
-		skb_frag_t *frag;
-		const struct page *page;
-		if (!np->rx_skbs[i])
-			continue;
+		/* Step 2: Rebuild the RX buffer freelist and the RX ring itself. */
+		for (requeue_idx = 0, i = 0; i < NET_RX_RING_SIZE; i++) {
+			skb_frag_t *frag;
+			const struct page *page;
+			if (!queue->rx_skbs[i])
+				continue;
 
-		skb = np->rx_skbs[requeue_idx] = xennet_get_rx_skb(np, i);
-		ref = np->grant_rx_ref[requeue_idx] = xennet_get_rx_ref(np, i);
-		req = RING_GET_REQUEST(&np->rx, requeue_idx);
+			skb = queue->rx_skbs[requeue_idx] = xennet_get_rx_skb(queue, i);
+			ref = queue->grant_rx_ref[requeue_idx] = xennet_get_rx_ref(queue, i);
+			req = RING_GET_REQUEST(&queue->rx, requeue_idx);
 
-		frag = &skb_shinfo(skb)->frags[0];
-		page = skb_frag_page(frag);
-		gnttab_grant_foreign_access_ref(
-			ref, np->xbdev->otherend_id,
-			pfn_to_mfn(page_to_pfn(page)),
-			0);
-		req->gref = ref;
-		req->id   = requeue_idx;
+			frag = &skb_shinfo(skb)->frags[0];
+			page = skb_frag_page(frag);
+			gnttab_grant_foreign_access_ref(
+				ref, queue->info->xbdev->otherend_id,
+				pfn_to_mfn(page_to_pfn(page)),
+				0);
+			req->gref = ref;
+			req->id   = requeue_idx;
 
-		requeue_idx++;
-	}
+			requeue_idx++;
+		}
 
-	np->rx.req_prod_pvt = requeue_idx;
+		queue->rx.req_prod_pvt = requeue_idx;
+	}
 
 	/*
 	 * Step 3: All public and private state should now be sane.  Get
@@ -1798,14 +2085,17 @@ static int xennet_connect(struct net_device *dev)
 	 * packets.
 	 */
 	netif_carrier_on(np->netdev);
-	notify_remote_via_irq(np->tx_irq);
-	if (np->tx_irq != np->rx_irq)
-		notify_remote_via_irq(np->rx_irq);
-	xennet_tx_buf_gc(dev);
-	xennet_alloc_rx_buffers(dev);
-
-	spin_unlock_irq(&np->tx_lock);
-	spin_unlock_bh(&np->rx_lock);
+	for (j = 0; j < num_queues; ++j) {
+		queue = &np->queues[j];
+		notify_remote_via_irq(queue->tx_irq);
+		if (queue->tx_irq != queue->rx_irq)
+			notify_remote_via_irq(queue->rx_irq);
+		xennet_tx_buf_gc(queue);
+		xennet_alloc_rx_buffers(queue);
+
+		spin_unlock_irq(&queue->tx_lock);
+		spin_unlock_bh(&queue->rx_lock);
+	}
 
 	return 0;
 }
@@ -1878,7 +2168,7 @@ static void xennet_get_ethtool_stats(struct net_device *dev,
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(xennet_stats); i++)
-		data[i] = *(unsigned long *)(np + xennet_stats[i].offset);
+		data[i] = atomic_read((atomic_t *)(np + xennet_stats[i].offset));
 }
 
 static void xennet_get_strings(struct net_device *dev, u32 stringset, u8 * data)
@@ -1909,8 +2199,12 @@ static ssize_t show_rxbuf_min(struct device *dev,
 {
 	struct net_device *netdev = to_net_dev(dev);
 	struct netfront_info *info = netdev_priv(netdev);
+	unsigned int num_queues = netdev->real_num_tx_queues;
 
-	return sprintf(buf, "%u\n", info->rx_min_target);
+	if (num_queues)
+		return sprintf(buf, "%u\n", info->queues[0].rx_min_target);
+	else
+		return sprintf(buf, "%u\n", RX_MIN_TARGET);
 }
 
 static ssize_t store_rxbuf_min(struct device *dev,
@@ -1919,8 +2213,11 @@ static ssize_t store_rxbuf_min(struct device *dev,
 {
 	struct net_device *netdev = to_net_dev(dev);
 	struct netfront_info *np = netdev_priv(netdev);
+	unsigned int num_queues = netdev->real_num_tx_queues;
 	char *endp;
 	unsigned long target;
+	unsigned int i;
+	struct netfront_queue *queue;
 
 	if (!capable(CAP_NET_ADMIN))
 		return -EPERM;
@@ -1934,16 +2231,19 @@ static ssize_t store_rxbuf_min(struct device *dev,
 	if (target > RX_MAX_TARGET)
 		target = RX_MAX_TARGET;
 
-	spin_lock_bh(&np->rx_lock);
-	if (target > np->rx_max_target)
-		np->rx_max_target = target;
-	np->rx_min_target = target;
-	if (target > np->rx_target)
-		np->rx_target = target;
+	for (i = 0; i < num_queues; ++i) {
+		queue = &np->queues[i];
+		spin_lock_bh(&queue->rx_lock);
+		if (target > queue->rx_max_target)
+			queue->rx_max_target = target;
+		queue->rx_min_target = target;
+		if (target > queue->rx_target)
+			queue->rx_target = target;
 
-	xennet_alloc_rx_buffers(netdev);
+		xennet_alloc_rx_buffers(queue);
 
-	spin_unlock_bh(&np->rx_lock);
+		spin_unlock_bh(&queue->rx_lock);
+	}
 	return len;
 }
 
@@ -1952,8 +2252,12 @@ static ssize_t show_rxbuf_max(struct device *dev,
 {
 	struct net_device *netdev = to_net_dev(dev);
 	struct netfront_info *info = netdev_priv(netdev);
+	unsigned int num_queues = netdev->real_num_tx_queues;
 
-	return sprintf(buf, "%u\n", info->rx_max_target);
+	if (num_queues)
+		return sprintf(buf, "%u\n", info->queues[0].rx_max_target);
+	else
+		return sprintf(buf, "%u\n", RX_MAX_TARGET);
 }
 
 static ssize_t store_rxbuf_max(struct device *dev,
@@ -1962,8 +2266,11 @@ static ssize_t store_rxbuf_max(struct device *dev,
 {
 	struct net_device *netdev = to_net_dev(dev);
 	struct netfront_info *np = netdev_priv(netdev);
+	unsigned int num_queues = netdev->real_num_tx_queues;
 	char *endp;
 	unsigned long target;
+	unsigned int i = 0;
+	struct netfront_queue *queue = NULL;
 
 	if (!capable(CAP_NET_ADMIN))
 		return -EPERM;
@@ -1977,16 +2284,19 @@ static ssize_t store_rxbuf_max(struct device *dev,
 	if (target > RX_MAX_TARGET)
 		target = RX_MAX_TARGET;
 
-	spin_lock_bh(&np->rx_lock);
-	if (target < np->rx_min_target)
-		np->rx_min_target = target;
-	np->rx_max_target = target;
-	if (target < np->rx_target)
-		np->rx_target = target;
+	for (i = 0; i < num_queues; ++i) {
+		queue = &np->queues[i];
+		spin_lock_bh(&queue->rx_lock);
+		if (target < queue->rx_min_target)
+			queue->rx_min_target = target;
+		queue->rx_max_target = target;
+		if (target < queue->rx_target)
+			queue->rx_target = target;
 
-	xennet_alloc_rx_buffers(netdev);
+		xennet_alloc_rx_buffers(queue);
 
-	spin_unlock_bh(&np->rx_lock);
+		spin_unlock_bh(&queue->rx_lock);
+	}
 	return len;
 }
 
@@ -1995,8 +2305,12 @@ static ssize_t show_rxbuf_cur(struct device *dev,
 {
 	struct net_device *netdev = to_net_dev(dev);
 	struct netfront_info *info = netdev_priv(netdev);
+	unsigned int num_queues = netdev->real_num_tx_queues;
 
-	return sprintf(buf, "%u\n", info->rx_target);
+	if (num_queues)
+		return sprintf(buf, "%u\n", info->queues[0].rx_target);
+	else
+		return sprintf(buf, "0\n");
 }
 
 static struct device_attribute xennet_attrs[] = {
@@ -2043,6 +2357,9 @@ static const struct xenbus_device_id netfront_ids[] = {
 static int xennet_remove(struct xenbus_device *dev)
 {
 	struct netfront_info *info = dev_get_drvdata(&dev->dev);
+	unsigned int num_queues = info->netdev->real_num_tx_queues;
+	struct netfront_queue *queue = NULL;
+	unsigned int i = 0;
 
 	dev_dbg(&dev->dev, "%s\n", dev->nodename);
 
@@ -2052,7 +2369,15 @@ static int xennet_remove(struct xenbus_device *dev)
 
 	unregister_netdev(info->netdev);
 
-	del_timer_sync(&info->rx_refill_timer);
+	for (i = 0; i < num_queues; ++i) {
+		queue = &info->queues[i];
+		del_timer_sync(&queue->rx_refill_timer);
+	}
+
+	if (num_queues) {
+		kfree(info->queues);
+		info->queues = NULL;
+	}
 
 	free_percpu(info->stats);
 
@@ -2078,6 +2403,9 @@ static int __init netif_init(void)
 
 	pr_info("Initialising Xen virtual ethernet driver\n");
 
+	/* Allow as many queues as there are CPUs, by default */
+	xennet_max_queues = num_online_cpus();
+
 	return xenbus_register_frontend(&netfront_driver);
 }
 module_init(netif_init);
diff --git a/drivers/nfc/Kconfig b/drivers/nfc/Kconfig
index 65d4ca19d132..26c66a126551 100644
--- a/drivers/nfc/Kconfig
+++ b/drivers/nfc/Kconfig
@@ -71,5 +71,6 @@ config NFC_PORT100
 source "drivers/nfc/pn544/Kconfig"
 source "drivers/nfc/microread/Kconfig"
 source "drivers/nfc/nfcmrvl/Kconfig"
+source "drivers/nfc/st21nfca/Kconfig"
 
 endmenu
diff --git a/drivers/nfc/Makefile b/drivers/nfc/Makefile
index ae42a3fa60c9..23225b0287fd 100644
--- a/drivers/nfc/Makefile
+++ b/drivers/nfc/Makefile
@@ -11,5 +11,6 @@ obj-$(CONFIG_NFC_SIM)		+= nfcsim.o
 obj-$(CONFIG_NFC_PORT100)	+= port100.o
 obj-$(CONFIG_NFC_MRVL)		+= nfcmrvl/
 obj-$(CONFIG_NFC_TRF7970A)	+= trf7970a.o
+obj-$(CONFIG_NFC_ST21NFCA)  += st21nfca/
 
 ccflags-$(CONFIG_NFC_DEBUG) := -DDEBUG
diff --git a/drivers/nfc/pn544/i2c.c b/drivers/nfc/pn544/i2c.c
index f2acd85be86e..440291ab7263 100644
--- a/drivers/nfc/pn544/i2c.c
+++ b/drivers/nfc/pn544/i2c.c
@@ -22,6 +22,8 @@
 #include <linux/module.h>
 #include <linux/i2c.h>
 #include <linux/gpio.h>
+#include <linux/of_gpio.h>
+#include <linux/of_irq.h>
 #include <linux/miscdevice.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
@@ -857,6 +859,92 @@ exit_state_wait_secure_write_answer:
 	}
 }
 
+#ifdef CONFIG_OF
+
+static int pn544_hci_i2c_of_request_resources(struct i2c_client *client)
+{
+	struct pn544_i2c_phy *phy = i2c_get_clientdata(client);
+	struct device_node *pp;
+	int ret;
+
+	pp = client->dev.of_node;
+	if (!pp) {
+		ret = -ENODEV;
+		goto err_dt;
+	}
+
+	/* Obtention of EN GPIO from device tree */
+	ret = of_get_named_gpio(pp, "enable-gpios", 0);
+	if (ret < 0) {
+		if (ret != -EPROBE_DEFER)
+			nfc_err(&client->dev,
+				"Failed to get EN gpio, error: %d\n", ret);
+		goto err_dt;
+	}
+	phy->gpio_en = ret;
+
+	/* Configuration of EN GPIO */
+	ret = gpio_request(phy->gpio_en, "pn544_en");
+	if (ret) {
+		nfc_err(&client->dev, "Fail EN pin\n");
+		goto err_dt;
+	}
+	ret = gpio_direction_output(phy->gpio_en, 0);
+	if (ret) {
+		nfc_err(&client->dev, "Fail EN pin direction\n");
+		goto err_gpio_en;
+	}
+
+	/* Obtention of FW GPIO from device tree */
+	ret = of_get_named_gpio(pp, "firmware-gpios", 0);
+	if (ret < 0) {
+		if (ret != -EPROBE_DEFER)
+			nfc_err(&client->dev,
+				"Failed to get FW gpio, error: %d\n", ret);
+		goto err_gpio_en;
+	}
+	phy->gpio_fw = ret;
+
+	/* Configuration of FW GPIO */
+	ret = gpio_request(phy->gpio_fw, "pn544_fw");
+	if (ret) {
+		nfc_err(&client->dev, "Fail FW pin\n");
+		goto err_gpio_en;
+	}
+	ret = gpio_direction_output(phy->gpio_fw, 0);
+	if (ret) {
+		nfc_err(&client->dev, "Fail FW pin direction\n");
+		goto err_gpio_fw;
+	}
+
+	/* IRQ */
+	ret = irq_of_parse_and_map(pp, 0);
+	if (ret < 0) {
+		nfc_err(&client->dev,
+			"Unable to get irq, error: %d\n", ret);
+		goto err_gpio_fw;
+	}
+	client->irq = ret;
+
+	return 0;
+
+err_gpio_fw:
+	gpio_free(phy->gpio_fw);
+err_gpio_en:
+	gpio_free(phy->gpio_en);
+err_dt:
+	return ret;
+}
+
+#else
+
+static int pn544_hci_i2c_of_request_resources(struct i2c_client *client)
+{
+	return -ENODEV;
+}
+
+#endif
+
 static int pn544_hci_i2c_probe(struct i2c_client *client,
 			       const struct i2c_device_id *id)
 {
@@ -887,25 +975,36 @@ static int pn544_hci_i2c_probe(struct i2c_client *client,
 	i2c_set_clientdata(client, phy);
 
 	pdata = client->dev.platform_data;
-	if (pdata == NULL) {
-		nfc_err(&client->dev, "No platform data\n");
-		return -EINVAL;
-	}
 
-	if (pdata->request_resources == NULL) {
-		nfc_err(&client->dev, "request_resources() missing\n");
-		return -EINVAL;
-	}
+	/* No platform data, using device tree. */
+	if (!pdata && client->dev.of_node) {
+		r = pn544_hci_i2c_of_request_resources(client);
+		if (r) {
+			nfc_err(&client->dev, "No DT data\n");
+			return r;
+		}
+	/* Using platform data. */
+	} else if (pdata) {
 
-	r = pdata->request_resources(client);
-	if (r) {
-		nfc_err(&client->dev, "Cannot get platform resources\n");
-		return r;
-	}
+		if (pdata->request_resources == NULL) {
+			nfc_err(&client->dev, "request_resources() missing\n");
+			return -EINVAL;
+		}
 
-	phy->gpio_en = pdata->get_gpio(NFC_GPIO_ENABLE);
-	phy->gpio_fw = pdata->get_gpio(NFC_GPIO_FW_RESET);
-	phy->gpio_irq = pdata->get_gpio(NFC_GPIO_IRQ);
+		r = pdata->request_resources(client);
+		if (r) {
+			nfc_err(&client->dev,
+				"Cannot get platform resources\n");
+			return r;
+		}
+
+		phy->gpio_en = pdata->get_gpio(NFC_GPIO_ENABLE);
+		phy->gpio_fw = pdata->get_gpio(NFC_GPIO_FW_RESET);
+		phy->gpio_irq = pdata->get_gpio(NFC_GPIO_IRQ);
+	} else {
+		nfc_err(&client->dev, "No platform data\n");
+		return -EINVAL;
+	}
 
 	pn544_hci_i2c_platform_init(phy);
 
@@ -930,8 +1029,12 @@ err_hci:
 	free_irq(client->irq, phy);
 
 err_rti:
-	if (pdata->free_resources != NULL)
+	if (!pdata) {
+		gpio_free(phy->gpio_en);
+		gpio_free(phy->gpio_fw);
+	} else if (pdata->free_resources) {
 		pdata->free_resources();
+	}
 
 	return r;
 }
@@ -953,15 +1056,30 @@ static int pn544_hci_i2c_remove(struct i2c_client *client)
 		pn544_hci_i2c_disable(phy);
 
 	free_irq(client->irq, phy);
-	if (pdata->free_resources)
+
+	/* No platform data, GPIOs have been requested by this driver */
+	if (!pdata) {
+		gpio_free(phy->gpio_en);
+		gpio_free(phy->gpio_fw);
+	/* Using platform data */
+	} else if (pdata->free_resources) {
 		pdata->free_resources();
+	}
 
 	return 0;
 }
 
+static const struct of_device_id of_pn544_i2c_match[] = {
+	{ .compatible = "nxp,pn544-i2c", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, of_pn544_i2c_match);
+
 static struct i2c_driver pn544_hci_i2c_driver = {
 	.driver = {
 		   .name = PN544_HCI_I2C_DRIVER_NAME,
+		   .owner  = THIS_MODULE,
+		   .of_match_table = of_match_ptr(of_pn544_i2c_match),
 		  },
 	.probe = pn544_hci_i2c_probe,
 	.id_table = pn544_hci_i2c_id_table,
diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c
index b7a372af5eb7..4ac4d31f6c59 100644
--- a/drivers/nfc/port100.c
+++ b/drivers/nfc/port100.c
@@ -28,7 +28,8 @@
 			   NFC_PROTO_MIFARE_MASK   | \
 			   NFC_PROTO_FELICA_MASK   | \
 			   NFC_PROTO_NFC_DEP_MASK  | \
-			   NFC_PROTO_ISO14443_MASK)
+			   NFC_PROTO_ISO14443_MASK | \
+			   NFC_PROTO_ISO14443_B_MASK)
 
 #define PORT100_CAPABILITIES (NFC_DIGITAL_DRV_CAPS_IN_CRC | \
 			      NFC_DIGITAL_DRV_CAPS_TG_CRC)
@@ -120,6 +121,7 @@ struct port100_in_rf_setting {
 #define PORT100_COMM_TYPE_IN_212F 0x01
 #define PORT100_COMM_TYPE_IN_424F 0x02
 #define PORT100_COMM_TYPE_IN_106A 0x03
+#define PORT100_COMM_TYPE_IN_106B 0x07
 
 static const struct port100_in_rf_setting in_rf_settings[] = {
 	[NFC_DIGITAL_RF_TECH_212F] = {
@@ -140,6 +142,12 @@ static const struct port100_in_rf_setting in_rf_settings[] = {
 		.in_recv_set_number = 15,
 		.in_recv_comm_type  = PORT100_COMM_TYPE_IN_106A,
 	},
+	[NFC_DIGITAL_RF_TECH_106B] = {
+		.in_send_set_number = 3,
+		.in_send_comm_type  = PORT100_COMM_TYPE_IN_106B,
+		.in_recv_set_number = 15,
+		.in_recv_comm_type  = PORT100_COMM_TYPE_IN_106B,
+	},
 	/* Ensures the array has NFC_DIGITAL_RF_TECH_LAST elements */
 	[NFC_DIGITAL_RF_TECH_LAST] = { 0 },
 };
@@ -340,6 +348,32 @@ in_protocols[][PORT100_IN_MAX_NUM_PROTOCOLS + 1] = {
 	[NFC_DIGITAL_FRAMING_NFC_DEP_ACTIVATED] = {
 		{ PORT100_IN_PROT_END, 0 },
 	},
+	[NFC_DIGITAL_FRAMING_NFCB] = {
+		{ PORT100_IN_PROT_INITIAL_GUARD_TIME,     20 },
+		{ PORT100_IN_PROT_ADD_CRC,                 1 },
+		{ PORT100_IN_PROT_CHECK_CRC,               1 },
+		{ PORT100_IN_PROT_MULTI_CARD,              0 },
+		{ PORT100_IN_PROT_ADD_PARITY,              0 },
+		{ PORT100_IN_PROT_CHECK_PARITY,            0 },
+		{ PORT100_IN_PROT_BITWISE_AC_RECV_MODE,    0 },
+		{ PORT100_IN_PROT_VALID_BIT_NUMBER,        8 },
+		{ PORT100_IN_PROT_CRYPTO1,                 0 },
+		{ PORT100_IN_PROT_ADD_SOF,                 1 },
+		{ PORT100_IN_PROT_CHECK_SOF,               1 },
+		{ PORT100_IN_PROT_ADD_EOF,                 1 },
+		{ PORT100_IN_PROT_CHECK_EOF,               1 },
+		{ PORT100_IN_PROT_DEAF_TIME,               4 },
+		{ PORT100_IN_PROT_CRM,                     0 },
+		{ PORT100_IN_PROT_CRM_MIN_LEN,             0 },
+		{ PORT100_IN_PROT_T1_TAG_FRAME,            0 },
+		{ PORT100_IN_PROT_RFCA,                    0 },
+		{ PORT100_IN_PROT_GUARD_TIME_AT_INITIATOR, 6 },
+		{ PORT100_IN_PROT_END,                     0 },
+	},
+	[NFC_DIGITAL_FRAMING_NFCB_T4T] = {
+		/* nfc_digital_framing_nfcb */
+		{ PORT100_IN_PROT_END,                     0 },
+	},
 	/* Ensures the array has NFC_DIGITAL_FRAMING_LAST elements */
 	[NFC_DIGITAL_FRAMING_LAST] = {
 		{ PORT100_IN_PROT_END, 0 },
diff --git a/drivers/nfc/st21nfca/Kconfig b/drivers/nfc/st21nfca/Kconfig
new file mode 100644
index 000000000000..ee459f066ade
--- /dev/null
+++ b/drivers/nfc/st21nfca/Kconfig
@@ -0,0 +1,23 @@
+config NFC_ST21NFCA
+	tristate "STMicroelectronics ST21NFCA NFC driver"
+	depends on NFC_HCI
+	select CRC_CCITT
+	default n
+	---help---
+	  STMicroelectronics ST21NFCA core driver. It implements the chipset
+	  HCI logic and hooks into the NFC kernel APIs. Physical layers will
+	  register against it.
+
+	  To compile this driver as a module, choose m here. The module will
+	  be called st21nfca.
+	  Say N if unsure.
+
+config NFC_ST21NFCA_I2C
+	tristate "NFC ST21NFCA i2c support"
+	depends on NFC_ST21NFCA && I2C && NFC_SHDLC
+	---help---
+	  This module adds support for the STMicroelectronics st21nfca i2c interface.
+	  Select this if your platform is using the i2c bus.
+
+	  If you choose to build a module, it'll be called st21nfca_i2c.
+	  Say N if unsure.
diff --git a/drivers/nfc/st21nfca/Makefile b/drivers/nfc/st21nfca/Makefile
new file mode 100644
index 000000000000..038ed093a119
--- /dev/null
+++ b/drivers/nfc/st21nfca/Makefile
@@ -0,0 +1,8 @@
+#
+# Makefile for ST21NFCA HCI based NFC driver
+#
+
+st21nfca_i2c-objs  = i2c.o
+
+obj-$(CONFIG_NFC_ST21NFCA)     += st21nfca.o
+obj-$(CONFIG_NFC_ST21NFCA_I2C) += st21nfca_i2c.o
diff --git a/drivers/nfc/st21nfca/i2c.c b/drivers/nfc/st21nfca/i2c.c
new file mode 100644
index 000000000000..3f954ed86d98
--- /dev/null
+++ b/drivers/nfc/st21nfca/i2c.c
@@ -0,0 +1,724 @@
+/*
+ * I2C Link Layer for ST21NFCA HCI based Driver
+ * Copyright (C) 2014  STMicroelectronics SAS. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/crc-ccitt.h>
+#include <linux/module.h>
+#include <linux/i2c.h>
+#include <linux/gpio.h>
+#include <linux/of_irq.h>
+#include <linux/of_gpio.h>
+#include <linux/miscdevice.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/nfc.h>
+#include <linux/firmware.h>
+#include <linux/unaligned/access_ok.h>
+#include <linux/platform_data/st21nfca.h>
+
+#include <net/nfc/hci.h>
+#include <net/nfc/llc.h>
+#include <net/nfc/nfc.h>
+
+#include "st21nfca.h"
+
+/*
+ * Every frame starts with ST21NFCA_SOF_EOF and ends with ST21NFCA_SOF_EOF.
+ * Because ST21NFCA_SOF_EOF is a possible data value, there is a mecanism
+ * called byte stuffing has been introduced.
+ *
+ * if byte == ST21NFCA_SOF_EOF or ST21NFCA_ESCAPE_BYTE_STUFFING
+ * - insert ST21NFCA_ESCAPE_BYTE_STUFFING (escape byte)
+ * - xor byte with ST21NFCA_BYTE_STUFFING_MASK
+ */
+#define ST21NFCA_SOF_EOF		0x7e
+#define ST21NFCA_BYTE_STUFFING_MASK	0x20
+#define ST21NFCA_ESCAPE_BYTE_STUFFING	0x7d
+
+/* SOF + 00 */
+#define ST21NFCA_FRAME_HEADROOM			2
+
+/* 2 bytes crc + EOF */
+#define ST21NFCA_FRAME_TAILROOM 3
+#define IS_START_OF_FRAME(buf) (buf[0] == ST21NFCA_SOF_EOF && \
+				buf[1] == 0)
+
+#define ST21NFCA_HCI_I2C_DRIVER_NAME "st21nfca_hci_i2c"
+
+static struct i2c_device_id st21nfca_hci_i2c_id_table[] = {
+	{ST21NFCA_HCI_DRIVER_NAME, 0},
+	{}
+};
+
+MODULE_DEVICE_TABLE(i2c, st21nfca_hci_i2c_id_table);
+
+struct st21nfca_i2c_phy {
+	struct i2c_client *i2c_dev;
+	struct nfc_hci_dev *hdev;
+
+	unsigned int gpio_ena;
+	unsigned int gpio_irq;
+	unsigned int irq_polarity;
+
+	struct sk_buff *pending_skb;
+	int current_read_len;
+	/*
+	 * crc might have fail because i2c macro
+	 * is disable due to other interface activity
+	 */
+	int crc_trials;
+
+	int powered;
+	int run_mode;
+
+	/*
+	 * < 0 if hardware error occured (e.g. i2c err)
+	 * and prevents normal operation.
+	 */
+	int hard_fault;
+	struct mutex phy_lock;
+};
+static u8 len_seq[] = { 13, 24, 15, 29 };
+static u16 wait_tab[] = { 2, 3, 5, 15, 20, 40};
+
+#define I2C_DUMP_SKB(info, skb)					\
+do {								\
+	pr_debug("%s:\n", info);				\
+	print_hex_dump(KERN_DEBUG, "i2c: ", DUMP_PREFIX_OFFSET,	\
+		       16, 1, (skb)->data, (skb)->len, 0);	\
+} while (0)
+
+/*
+ * In order to get the CLF in a known state we generate an internal reboot
+ * using a proprietary command.
+ * Once the reboot is completed, we expect to receive a ST21NFCA_SOF_EOF
+ * fill buffer.
+ */
+static int st21nfca_hci_platform_init(struct st21nfca_i2c_phy *phy)
+{
+	u16 wait_reboot[] = { 50, 300, 1000 };
+	char reboot_cmd[] = { 0x7E, 0x66, 0x48, 0xF6, 0x7E };
+	u8 tmp[ST21NFCA_HCI_LLC_MAX_SIZE];
+	int i, r = -1;
+
+	for (i = 0; i < ARRAY_SIZE(wait_reboot) && r < 0; i++) {
+		r = i2c_master_send(phy->i2c_dev, reboot_cmd,
+				    sizeof(reboot_cmd));
+		if (r < 0)
+			msleep(wait_reboot[i]);
+	}
+	if (r < 0)
+		return r;
+
+	/* CLF is spending about 20ms to do an internal reboot */
+	msleep(20);
+	r = -1;
+	for (i = 0; i < ARRAY_SIZE(wait_reboot) && r < 0; i++) {
+		r = i2c_master_recv(phy->i2c_dev, tmp,
+				    ST21NFCA_HCI_LLC_MAX_SIZE);
+		if (r < 0)
+			msleep(wait_reboot[i]);
+	}
+	if (r < 0)
+		return r;
+
+	for (i = 0; i < ST21NFCA_HCI_LLC_MAX_SIZE &&
+		tmp[i] == ST21NFCA_SOF_EOF; i++)
+		;
+
+	if (r != ST21NFCA_HCI_LLC_MAX_SIZE)
+		return -ENODEV;
+
+	usleep_range(1000, 1500);
+	return 0;
+}
+
+static int st21nfca_hci_i2c_enable(void *phy_id)
+{
+	struct st21nfca_i2c_phy *phy = phy_id;
+
+	gpio_set_value(phy->gpio_ena, 1);
+	phy->powered = 1;
+	phy->run_mode = ST21NFCA_HCI_MODE;
+
+	usleep_range(10000, 15000);
+
+	return 0;
+}
+
+static void st21nfca_hci_i2c_disable(void *phy_id)
+{
+	struct st21nfca_i2c_phy *phy = phy_id;
+
+	pr_info("\n");
+	gpio_set_value(phy->gpio_ena, 0);
+
+	phy->powered = 0;
+}
+
+static void st21nfca_hci_add_len_crc(struct sk_buff *skb)
+{
+	u16 crc;
+	u8 tmp;
+
+	*skb_push(skb, 1) = 0;
+
+	crc = crc_ccitt(0xffff, skb->data, skb->len);
+	crc = ~crc;
+
+	tmp = crc & 0x00ff;
+	*skb_put(skb, 1) = tmp;
+
+	tmp = (crc >> 8) & 0x00ff;
+	*skb_put(skb, 1) = tmp;
+}
+
+static void st21nfca_hci_remove_len_crc(struct sk_buff *skb)
+{
+	skb_pull(skb, ST21NFCA_FRAME_HEADROOM);
+	skb_trim(skb, skb->len - ST21NFCA_FRAME_TAILROOM);
+}
+
+/*
+ * Writing a frame must not return the number of written bytes.
+ * It must return either zero for success, or <0 for error.
+ * In addition, it must not alter the skb
+ */
+static int st21nfca_hci_i2c_write(void *phy_id, struct sk_buff *skb)
+{
+	int r = -1, i, j;
+	struct st21nfca_i2c_phy *phy = phy_id;
+	struct i2c_client *client = phy->i2c_dev;
+	u8 tmp[ST21NFCA_HCI_LLC_MAX_SIZE * 2];
+
+	I2C_DUMP_SKB("st21nfca_hci_i2c_write", skb);
+
+
+	if (phy->hard_fault != 0)
+		return phy->hard_fault;
+
+	/*
+	 * Compute CRC before byte stuffing computation on frame
+	 * Note st21nfca_hci_add_len_crc is doing a byte stuffing
+	 * on its own value
+	 */
+	st21nfca_hci_add_len_crc(skb);
+
+	/* add ST21NFCA_SOF_EOF on tail */
+	*skb_put(skb, 1) = ST21NFCA_SOF_EOF;
+	/* add ST21NFCA_SOF_EOF on head */
+	*skb_push(skb, 1) = ST21NFCA_SOF_EOF;
+
+	/*
+	 * Compute byte stuffing
+	 * if byte == ST21NFCA_SOF_EOF or ST21NFCA_ESCAPE_BYTE_STUFFING
+	 * insert ST21NFCA_ESCAPE_BYTE_STUFFING (escape byte)
+	 * xor byte with ST21NFCA_BYTE_STUFFING_MASK
+	 */
+	tmp[0] = skb->data[0];
+	for (i = 1, j = 1; i < skb->len - 1; i++, j++) {
+		if (skb->data[i] == ST21NFCA_SOF_EOF
+		    || skb->data[i] == ST21NFCA_ESCAPE_BYTE_STUFFING) {
+			tmp[j] = ST21NFCA_ESCAPE_BYTE_STUFFING;
+			j++;
+			tmp[j] = skb->data[i] ^ ST21NFCA_BYTE_STUFFING_MASK;
+		} else {
+			tmp[j] = skb->data[i];
+		}
+	}
+	tmp[j] = skb->data[i];
+	j++;
+
+	/*
+	 * Manage sleep mode
+	 * Try 3 times to send data with delay between each
+	 */
+	mutex_lock(&phy->phy_lock);
+	for (i = 0; i < ARRAY_SIZE(wait_tab) && r < 0; i++) {
+		r = i2c_master_send(client, tmp, j);
+		if (r < 0)
+			msleep(wait_tab[i]);
+	}
+	mutex_unlock(&phy->phy_lock);
+
+	if (r >= 0) {
+		if (r != j)
+			r = -EREMOTEIO;
+		else
+			r = 0;
+	}
+
+	st21nfca_hci_remove_len_crc(skb);
+
+	return r;
+}
+
+static int get_frame_size(u8 *buf, int buflen)
+{
+	int len = 0;
+	if (buf[len + 1] == ST21NFCA_SOF_EOF)
+		return 0;
+
+	for (len = 1; len < buflen && buf[len] != ST21NFCA_SOF_EOF; len++)
+		;
+
+	return len;
+}
+
+static int check_crc(u8 *buf, int buflen)
+{
+	u16 crc;
+
+	crc = crc_ccitt(0xffff, buf, buflen - 2);
+	crc = ~crc;
+
+	if (buf[buflen - 2] != (crc & 0xff) || buf[buflen - 1] != (crc >> 8)) {
+		pr_err(ST21NFCA_HCI_DRIVER_NAME
+		       ": CRC error 0x%x != 0x%x 0x%x\n", crc, buf[buflen - 1],
+		       buf[buflen - 2]);
+
+		pr_info(DRIVER_DESC ": %s : BAD CRC\n", __func__);
+		print_hex_dump(KERN_DEBUG, "crc: ", DUMP_PREFIX_NONE,
+			       16, 2, buf, buflen, false);
+		return -EPERM;
+	}
+	return 0;
+}
+
+/*
+ * Prepare received data for upper layer.
+ * Received data include byte stuffing, crc and sof/eof
+ * which is not usable by hci part.
+ * returns:
+ * frame size without sof/eof, header and byte stuffing
+ * -EBADMSG : frame was incorrect and discarded
+ */
+static int st21nfca_hci_i2c_repack(struct sk_buff *skb)
+{
+	int i, j, r, size;
+	if (skb->len < 1 || (skb->len > 1 && skb->data[1] != 0))
+		return -EBADMSG;
+
+	size = get_frame_size(skb->data, skb->len);
+	if (size > 0) {
+		skb_trim(skb, size);
+		/* remove ST21NFCA byte stuffing for upper layer */
+		for (i = 1, j = 0; i < skb->len; i++) {
+			if (skb->data[i + j] ==
+					(u8) ST21NFCA_ESCAPE_BYTE_STUFFING) {
+				skb->data[i] = skb->data[i + j + 1]
+						| ST21NFCA_BYTE_STUFFING_MASK;
+				i++;
+				j++;
+			}
+			skb->data[i] = skb->data[i + j];
+		}
+		/* remove byte stuffing useless byte */
+		skb_trim(skb, i - j);
+		/* remove ST21NFCA_SOF_EOF from head */
+		skb_pull(skb, 1);
+
+		r = check_crc(skb->data, skb->len);
+		if (r != 0) {
+			i = 0;
+			return -EBADMSG;
+		}
+
+		/* remove headbyte */
+		skb_pull(skb, 1);
+		/* remove crc. Byte Stuffing is already removed here */
+		skb_trim(skb, skb->len - 2);
+		return skb->len;
+	}
+	return 0;
+}
+
+/*
+ * Reads an shdlc frame and returns it in a newly allocated sk_buff. Guarantees
+ * that i2c bus will be flushed and that next read will start on a new frame.
+ * returned skb contains only LLC header and payload.
+ * returns:
+ * frame size : if received frame is complete (find ST21NFCA_SOF_EOF at
+ * end of read)
+ * -EAGAIN : if received frame is incomplete (not find ST21NFCA_SOF_EOF
+ * at end of read)
+ * -EREMOTEIO : i2c read error (fatal)
+ * -EBADMSG : frame was incorrect and discarded
+ * (value returned from st21nfca_hci_i2c_repack)
+ * -EIO : if no ST21NFCA_SOF_EOF is found after reaching
+ * the read length end sequence
+ */
+static int st21nfca_hci_i2c_read(struct st21nfca_i2c_phy *phy,
+				 struct sk_buff *skb)
+{
+	int r, i;
+	u8 len;
+	u8 buf[ST21NFCA_HCI_LLC_MAX_PAYLOAD];
+	struct i2c_client *client = phy->i2c_dev;
+
+	if (phy->current_read_len < ARRAY_SIZE(len_seq)) {
+		len = len_seq[phy->current_read_len];
+
+		/*
+		 * Add retry mecanism
+		 * Operation on I2C interface may fail in case of operation on
+		 * RF or SWP interface
+		 */
+		r = 0;
+		mutex_lock(&phy->phy_lock);
+		for (i = 0; i < ARRAY_SIZE(wait_tab) && r <= 0; i++) {
+			r = i2c_master_recv(client, buf, len);
+			if (r < 0)
+				msleep(wait_tab[i]);
+		}
+		mutex_unlock(&phy->phy_lock);
+
+		if (r != len) {
+			phy->current_read_len = 0;
+			return -EREMOTEIO;
+		}
+
+		/*
+		 * The first read sequence does not start with SOF.
+		 * Data is corrupeted so we drop it.
+		 */
+		if (!phy->current_read_len && buf[0] != ST21NFCA_SOF_EOF) {
+			skb_trim(skb, 0);
+			phy->current_read_len = 0;
+			return -EIO;
+		} else if (phy->current_read_len &&
+			IS_START_OF_FRAME(buf)) {
+			/*
+			 * Previous frame transmission was interrupted and
+			 * the frame got repeated.
+			 * Received frame start with ST21NFCA_SOF_EOF + 00.
+			 */
+			skb_trim(skb, 0);
+			phy->current_read_len = 0;
+		}
+
+		memcpy(skb_put(skb, len), buf, len);
+
+		if (skb->data[skb->len - 1] == ST21NFCA_SOF_EOF) {
+			phy->current_read_len = 0;
+			return st21nfca_hci_i2c_repack(skb);
+		}
+		phy->current_read_len++;
+		return -EAGAIN;
+	}
+	return -EIO;
+}
+
+/*
+ * Reads an shdlc frame from the chip. This is not as straightforward as it
+ * seems. The frame format is data-crc, and corruption can occur anywhere
+ * while transiting on i2c bus, such that we could read an invalid data.
+ * The tricky case is when we read a corrupted data or crc. We must detect
+ * this here in order to determine that data can be transmitted to the hci
+ * core. This is the reason why we check the crc here.
+ * The CLF will repeat a frame until we send a RR on that frame.
+ *
+ * On ST21NFCA, IRQ goes in idle when read starts. As no size information are
+ * available in the incoming data, other IRQ might come. Every IRQ will trigger
+ * a read sequence with different length and will fill the current frame.
+ * The reception is complete once we reach a ST21NFCA_SOF_EOF.
+ */
+static irqreturn_t st21nfca_hci_irq_thread_fn(int irq, void *phy_id)
+{
+	struct st21nfca_i2c_phy *phy = phy_id;
+	struct i2c_client *client;
+
+	int r;
+
+	if (!phy || irq != phy->i2c_dev->irq) {
+		WARN_ON_ONCE(1);
+		return IRQ_NONE;
+	}
+
+	client = phy->i2c_dev;
+	dev_dbg(&client->dev, "IRQ\n");
+
+	if (phy->hard_fault != 0)
+		return IRQ_HANDLED;
+
+	r = st21nfca_hci_i2c_read(phy, phy->pending_skb);
+	if (r == -EREMOTEIO) {
+		phy->hard_fault = r;
+
+		nfc_hci_recv_frame(phy->hdev, NULL);
+
+		return IRQ_HANDLED;
+	} else if (r == -EAGAIN || r == -EIO) {
+		return IRQ_HANDLED;
+	} else if (r == -EBADMSG && phy->crc_trials < ARRAY_SIZE(wait_tab)) {
+		/*
+		 * With ST21NFCA, only one interface (I2C, RF or SWP)
+		 * may be active at a time.
+		 * Having incorrect crc is usually due to i2c macrocell
+		 * deactivation in the middle of a transmission.
+		 * It may generate corrupted data on i2c.
+		 * We give sometime to get i2c back.
+		 * The complete frame will be repeated.
+		 */
+		msleep(wait_tab[phy->crc_trials]);
+		phy->crc_trials++;
+		phy->current_read_len = 0;
+		kfree_skb(phy->pending_skb);
+	} else if (r > 0) {
+		/*
+		 * We succeeded to read data from the CLF and
+		 * data is valid.
+		 * Reset counter.
+		 */
+		nfc_hci_recv_frame(phy->hdev, phy->pending_skb);
+		phy->crc_trials = 0;
+	}
+
+	phy->pending_skb = alloc_skb(ST21NFCA_HCI_LLC_MAX_SIZE * 2, GFP_KERNEL);
+	if (phy->pending_skb == NULL) {
+		phy->hard_fault = -ENOMEM;
+		nfc_hci_recv_frame(phy->hdev, NULL);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static struct nfc_phy_ops i2c_phy_ops = {
+	.write = st21nfca_hci_i2c_write,
+	.enable = st21nfca_hci_i2c_enable,
+	.disable = st21nfca_hci_i2c_disable,
+};
+
+#ifdef CONFIG_OF
+static int st21nfca_hci_i2c_of_request_resources(struct i2c_client *client)
+{
+	struct st21nfca_i2c_phy *phy = i2c_get_clientdata(client);
+	struct device_node *pp;
+	int gpio;
+	int r;
+
+	pp = client->dev.of_node;
+	if (!pp)
+		return -ENODEV;
+
+	/* Get GPIO from device tree */
+	gpio = of_get_named_gpio(pp, "enable-gpios", 0);
+	if (gpio < 0) {
+		nfc_err(&client->dev, "Failed to retrieve enable-gpios from device tree\n");
+		return gpio;
+	}
+
+	/* GPIO request and configuration */
+	r = devm_gpio_request(&client->dev, gpio, "clf_enable");
+	if (r) {
+		nfc_err(&client->dev, "Failed to request enable pin\n");
+		return -ENODEV;
+	}
+
+	r = gpio_direction_output(gpio, 1);
+	if (r) {
+		nfc_err(&client->dev, "Failed to set enable pin direction as output\n");
+		return -ENODEV;
+	}
+	phy->gpio_ena = gpio;
+
+	/* IRQ */
+	r = irq_of_parse_and_map(pp, 0);
+	if (r < 0) {
+		nfc_err(&client->dev,
+				"Unable to get irq, error: %d\n", r);
+		return r;
+	}
+
+	phy->irq_polarity = irq_get_trigger_type(r);
+	client->irq = r;
+
+	return 0;
+}
+#else
+static int st21nfca_hci_i2c_of_request_resources(struct i2c_client *client)
+{
+	return -ENODEV;
+}
+#endif
+
+static int st21nfca_hci_i2c_request_resources(struct i2c_client *client)
+{
+	struct st21nfca_nfc_platform_data *pdata;
+	struct st21nfca_i2c_phy *phy = i2c_get_clientdata(client);
+	int r;
+	int irq;
+
+	pdata = client->dev.platform_data;
+	if (pdata == NULL) {
+		nfc_err(&client->dev, "No platform data\n");
+		return -EINVAL;
+	}
+
+	/* store for later use */
+	phy->gpio_irq = pdata->gpio_irq;
+	phy->gpio_ena = pdata->gpio_ena;
+	phy->irq_polarity = pdata->irq_polarity;
+
+	r = devm_gpio_request(&client->dev, phy->gpio_irq, "wake_up");
+	if (r) {
+		pr_err("%s : gpio_request failed\n", __FILE__);
+		return -ENODEV;
+	}
+
+	r = gpio_direction_input(phy->gpio_irq);
+	if (r) {
+		pr_err("%s : gpio_direction_input failed\n", __FILE__);
+		return -ENODEV;
+	}
+
+	if (phy->gpio_ena > 0) {
+		r = devm_gpio_request(&client->dev,
+					phy->gpio_ena, "clf_enable");
+		if (r) {
+			pr_err("%s : ena gpio_request failed\n", __FILE__);
+			return -ENODEV;
+		}
+		r = gpio_direction_output(phy->gpio_ena, 1);
+
+		if (r) {
+			pr_err("%s : ena gpio_direction_output failed\n",
+			       __FILE__);
+			return -ENODEV;
+		}
+	}
+
+	/* IRQ */
+	irq = gpio_to_irq(phy->gpio_irq);
+	if (irq < 0) {
+		nfc_err(&client->dev,
+				"Unable to get irq number for GPIO %d error %d\n",
+				phy->gpio_irq, r);
+		return -ENODEV;
+	}
+	client->irq = irq;
+
+	return 0;
+}
+
+static int st21nfca_hci_i2c_probe(struct i2c_client *client,
+				  const struct i2c_device_id *id)
+{
+	struct st21nfca_i2c_phy *phy;
+	struct st21nfca_nfc_platform_data *pdata;
+	int r;
+
+	dev_dbg(&client->dev, "%s\n", __func__);
+	dev_dbg(&client->dev, "IRQ: %d\n", client->irq);
+
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) {
+		nfc_err(&client->dev, "Need I2C_FUNC_I2C\n");
+		return -ENODEV;
+	}
+
+	phy = devm_kzalloc(&client->dev, sizeof(struct st21nfca_i2c_phy),
+			   GFP_KERNEL);
+	if (!phy) {
+		nfc_err(&client->dev,
+			"Cannot allocate memory for st21nfca i2c phy.\n");
+		return -ENOMEM;
+	}
+
+	phy->i2c_dev = client;
+	phy->pending_skb = alloc_skb(ST21NFCA_HCI_LLC_MAX_SIZE * 2, GFP_KERNEL);
+	if (phy->pending_skb == NULL)
+		return -ENOMEM;
+
+	phy->current_read_len = 0;
+	phy->crc_trials = 0;
+	mutex_init(&phy->phy_lock);
+	i2c_set_clientdata(client, phy);
+
+	pdata = client->dev.platform_data;
+	if (!pdata && client->dev.of_node) {
+		r = st21nfca_hci_i2c_of_request_resources(client);
+		if (r) {
+			nfc_err(&client->dev, "No platform data\n");
+			return r;
+		}
+	} else if (pdata) {
+		r = st21nfca_hci_i2c_request_resources(client);
+		if (r) {
+			nfc_err(&client->dev, "Cannot get platform resources\n");
+			return r;
+		}
+	} else {
+		nfc_err(&client->dev, "st21nfca platform resources not available\n");
+		return -ENODEV;
+	}
+
+	r = st21nfca_hci_platform_init(phy);
+	if (r < 0) {
+		nfc_err(&client->dev, "Unable to reboot st21nfca\n");
+		return -ENODEV;
+	}
+
+	r = devm_request_threaded_irq(&client->dev, client->irq, NULL,
+				st21nfca_hci_irq_thread_fn,
+				phy->irq_polarity | IRQF_ONESHOT,
+				ST21NFCA_HCI_DRIVER_NAME, phy);
+	if (r < 0) {
+		nfc_err(&client->dev, "Unable to register IRQ handler\n");
+		return r;
+	}
+
+	return st21nfca_hci_probe(phy, &i2c_phy_ops, LLC_SHDLC_NAME,
+			       ST21NFCA_FRAME_HEADROOM, ST21NFCA_FRAME_TAILROOM,
+			       ST21NFCA_HCI_LLC_MAX_PAYLOAD, &phy->hdev);
+}
+
+static int st21nfca_hci_i2c_remove(struct i2c_client *client)
+{
+	struct st21nfca_i2c_phy *phy = i2c_get_clientdata(client);
+
+	dev_dbg(&client->dev, "%s\n", __func__);
+
+	st21nfca_hci_remove(phy->hdev);
+
+	if (phy->powered)
+		st21nfca_hci_i2c_disable(phy);
+
+	return 0;
+}
+
+static const struct of_device_id of_st21nfca_i2c_match[] = {
+	{ .compatible = "st,st21nfca_i2c", },
+	{}
+};
+
+static struct i2c_driver st21nfca_hci_i2c_driver = {
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = ST21NFCA_HCI_I2C_DRIVER_NAME,
+		.owner = THIS_MODULE,
+		.of_match_table = of_match_ptr(of_st21nfca_i2c_match),
+	},
+	.probe = st21nfca_hci_i2c_probe,
+	.id_table = st21nfca_hci_i2c_id_table,
+	.remove = st21nfca_hci_i2c_remove,
+};
+
+module_i2c_driver(st21nfca_hci_i2c_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION(DRIVER_DESC);
diff --git a/drivers/nfc/st21nfca/st21nfca.c b/drivers/nfc/st21nfca/st21nfca.c
new file mode 100644
index 000000000000..51e0f00b3a4f
--- /dev/null
+++ b/drivers/nfc/st21nfca/st21nfca.c
@@ -0,0 +1,698 @@
+/*
+ * HCI based Driver for STMicroelectronics NFC Chip
+ *
+ * Copyright (C) 2014  STMicroelectronics SAS. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/module.h>
+#include <linux/nfc.h>
+#include <net/nfc/hci.h>
+#include <net/nfc/llc.h>
+
+#include "st21nfca.h"
+
+#define DRIVER_DESC "HCI NFC driver for ST21NFCA"
+
+#define FULL_VERSION_LEN 3
+
+/* Proprietary gates, events, commands and registers */
+
+/* Commands that apply to all RF readers */
+#define ST21NFCA_RF_READER_CMD_PRESENCE_CHECK	0x30
+
+#define ST21NFCA_RF_READER_ISO15693_GATE	0x12
+#define ST21NFCA_RF_READER_ISO15693_INVENTORY 0x01
+
+/*
+ * Reader gate for communication with contact-less cards using Type A
+ * protocol ISO14443-3 but not compliant with ISO14443-4
+ */
+#define ST21NFCA_RF_READER_14443_3_A_GATE	0x15
+#define ST21NFCA_RF_READER_14443_3_A_UID	0x02
+#define ST21NFCA_RF_READER_14443_3_A_ATQA	0x03
+#define ST21NFCA_RF_READER_14443_3_A_SAK	0x04
+
+#define ST21NFCA_DEVICE_MGNT_GATE		0x01
+#define ST21NFCA_DEVICE_MGNT_PIPE		0x02
+
+#define ST21NFCA_DM_GETINFO         0x13
+#define ST21NFCA_DM_GETINFO_PIPE_LIST       0x02
+#define ST21NFCA_DM_GETINFO_PIPE_INFO       0x01
+#define ST21NFCA_DM_PIPE_CREATED        0x02
+#define ST21NFCA_DM_PIPE_OPEN           0x04
+#define ST21NFCA_DM_RF_ACTIVE           0x80
+
+#define ST21NFCA_DM_IS_PIPE_OPEN(p) \
+	((p & 0x0f) == (ST21NFCA_DM_PIPE_CREATED | ST21NFCA_DM_PIPE_OPEN))
+
+#define ST21NFCA_NFC_MODE	0x03	/* NFC_MODE parameter*/
+
+static DECLARE_BITMAP(dev_mask, ST21NFCA_NUM_DEVICES);
+
+static struct nfc_hci_gate st21nfca_gates[] = {
+	{NFC_HCI_ADMIN_GATE, NFC_HCI_ADMIN_PIPE},
+	{NFC_HCI_LOOPBACK_GATE, NFC_HCI_INVALID_PIPE},
+	{NFC_HCI_ID_MGMT_GATE, NFC_HCI_INVALID_PIPE},
+	{NFC_HCI_LINK_MGMT_GATE, NFC_HCI_LINK_MGMT_PIPE},
+	{NFC_HCI_RF_READER_B_GATE, NFC_HCI_INVALID_PIPE},
+	{NFC_HCI_RF_READER_A_GATE, NFC_HCI_INVALID_PIPE},
+	{ST21NFCA_DEVICE_MGNT_GATE, ST21NFCA_DEVICE_MGNT_PIPE},
+	{ST21NFCA_RF_READER_F_GATE, NFC_HCI_INVALID_PIPE},
+	{ST21NFCA_RF_READER_14443_3_A_GATE, NFC_HCI_INVALID_PIPE},
+	{ST21NFCA_RF_READER_ISO15693_GATE, NFC_HCI_INVALID_PIPE},
+};
+
+struct st21nfca_pipe_info {
+	u8 pipe_state;
+	u8 src_host_id;
+	u8 src_gate_id;
+	u8 dst_host_id;
+	u8 dst_gate_id;
+} __packed;
+
+/* Largest headroom needed for outgoing custom commands */
+#define ST21NFCA_CMDS_HEADROOM  7
+
+static int st21nfca_hci_load_session(struct nfc_hci_dev *hdev)
+{
+	int i, j, r;
+	struct sk_buff *skb_pipe_list, *skb_pipe_info;
+	struct st21nfca_pipe_info *info;
+
+	u8 pipe_list[] = { ST21NFCA_DM_GETINFO_PIPE_LIST,
+		NFC_HCI_TERMINAL_HOST_ID
+	};
+	u8 pipe_info[] = { ST21NFCA_DM_GETINFO_PIPE_INFO,
+		NFC_HCI_TERMINAL_HOST_ID, 0
+	};
+
+	skb_pipe_list = alloc_skb(ST21NFCA_HCI_LLC_MAX_SIZE, GFP_KERNEL);
+	if (!skb_pipe_list) {
+		r = -ENOMEM;
+		goto free_list;
+	}
+
+	skb_pipe_info = alloc_skb(ST21NFCA_HCI_LLC_MAX_SIZE, GFP_KERNEL);
+	if (!skb_pipe_info) {
+		r = -ENOMEM;
+		goto free_info;
+	}
+
+	/* On ST21NFCA device pipes number are dynamics
+	 * A maximum of 16 pipes can be created at the same time
+	 * If pipes are already created, hci_dev_up will fail.
+	 * Doing a clear all pipe is a bad idea because:
+	 * - It does useless EEPROM cycling
+	 * - It might cause issue for secure elements support
+	 * (such as removing connectivity or APDU reader pipe)
+	 * A better approach on ST21NFCA is to:
+	 * - get a pipe list for each host.
+	 * (eg: NFC_HCI_HOST_CONTROLLER_ID for now).
+	 * (TODO Later on UICC HOST and eSE HOST)
+	 * - get pipe information
+	 * - match retrieved pipe list in st21nfca_gates
+	 * ST21NFCA_DEVICE_MGNT_GATE is a proprietary gate
+	 * with ST21NFCA_DEVICE_MGNT_PIPE.
+	 * Pipe can be closed and need to be open.
+	 */
+	r = nfc_hci_connect_gate(hdev, NFC_HCI_HOST_CONTROLLER_ID,
+		ST21NFCA_DEVICE_MGNT_GATE, ST21NFCA_DEVICE_MGNT_PIPE);
+	if (r < 0)
+		goto free_info;
+
+	/* Get pipe list */
+	r = nfc_hci_send_cmd(hdev, ST21NFCA_DEVICE_MGNT_GATE,
+			ST21NFCA_DM_GETINFO, pipe_list, sizeof(pipe_list),
+			&skb_pipe_list);
+	if (r < 0)
+		goto free_info;
+
+	/* Complete the existing gate_pipe table */
+	for (i = 0; i < skb_pipe_list->len; i++) {
+		pipe_info[2] = skb_pipe_list->data[i];
+		r = nfc_hci_send_cmd(hdev, ST21NFCA_DEVICE_MGNT_GATE,
+					ST21NFCA_DM_GETINFO, pipe_info,
+					sizeof(pipe_info), &skb_pipe_info);
+
+		if (r)
+			continue;
+
+		/*
+		 * Match pipe ID and gate ID
+		 * Output format from ST21NFC_DM_GETINFO is:
+		 * - pipe state (1byte)
+		 * - source hid (1byte)
+		 * - source gid (1byte)
+		 * - destination hid (1byte)
+		 * - destination gid (1byte)
+		 */
+		info = (struct st21nfca_pipe_info *) skb_pipe_info->data;
+		for (j = 0; (j < ARRAY_SIZE(st21nfca_gates)) &&
+			(st21nfca_gates[j].gate != info->dst_gate_id);
+			j++)
+			;
+
+		if (j < ARRAY_SIZE(st21nfca_gates) &&
+			st21nfca_gates[j].gate == info->dst_gate_id &&
+			ST21NFCA_DM_IS_PIPE_OPEN(info->pipe_state)) {
+			st21nfca_gates[j].pipe = pipe_info[2];
+			hdev->gate2pipe[st21nfca_gates[j].gate] =
+				st21nfca_gates[j].pipe;
+		}
+	}
+
+	/*
+	 * 3 gates have a well known pipe ID.
+	 * They will never appear in the pipe list
+	 */
+	if (skb_pipe_list->len + 3 < ARRAY_SIZE(st21nfca_gates)) {
+		for (i = skb_pipe_list->len + 3;
+				i < ARRAY_SIZE(st21nfca_gates); i++) {
+			r = nfc_hci_connect_gate(hdev,
+					NFC_HCI_HOST_CONTROLLER_ID,
+					st21nfca_gates[i].gate,
+					st21nfca_gates[i].pipe);
+			if (r < 0)
+				goto free_info;
+		}
+	}
+
+	memcpy(hdev->init_data.gates, st21nfca_gates, sizeof(st21nfca_gates));
+free_info:
+	kfree_skb(skb_pipe_info);
+free_list:
+	kfree_skb(skb_pipe_list);
+	return r;
+}
+
+static int st21nfca_hci_open(struct nfc_hci_dev *hdev)
+{
+	struct st21nfca_hci_info *info = nfc_hci_get_clientdata(hdev);
+	int r;
+
+	mutex_lock(&info->info_lock);
+
+	if (info->state != ST21NFCA_ST_COLD) {
+		r = -EBUSY;
+		goto out;
+	}
+
+	r = info->phy_ops->enable(info->phy_id);
+
+	if (r == 0)
+		info->state = ST21NFCA_ST_READY;
+
+out:
+	mutex_unlock(&info->info_lock);
+	return r;
+}
+
+static void st21nfca_hci_close(struct nfc_hci_dev *hdev)
+{
+	struct st21nfca_hci_info *info = nfc_hci_get_clientdata(hdev);
+
+	mutex_lock(&info->info_lock);
+
+	if (info->state == ST21NFCA_ST_COLD)
+		goto out;
+
+	info->phy_ops->disable(info->phy_id);
+	info->state = ST21NFCA_ST_COLD;
+
+out:
+	mutex_unlock(&info->info_lock);
+}
+
+static int st21nfca_hci_ready(struct nfc_hci_dev *hdev)
+{
+	struct sk_buff *skb;
+
+	u8 param;
+	int r;
+
+	param = NFC_HCI_UICC_HOST_ID;
+	r = nfc_hci_set_param(hdev, NFC_HCI_ADMIN_GATE,
+			      NFC_HCI_ADMIN_WHITELIST, &param, 1);
+	if (r < 0)
+		return r;
+
+	/* Set NFC_MODE in device management gate to enable */
+	r = nfc_hci_get_param(hdev, ST21NFCA_DEVICE_MGNT_GATE,
+			      ST21NFCA_NFC_MODE, &skb);
+	if (r < 0)
+		return r;
+
+	if (skb->data[0] == 0) {
+		kfree_skb(skb);
+		param = 1;
+
+		r = nfc_hci_set_param(hdev, ST21NFCA_DEVICE_MGNT_GATE,
+					ST21NFCA_NFC_MODE, &param, 1);
+		if (r < 0)
+			return r;
+	}
+
+	r = nfc_hci_send_event(hdev, NFC_HCI_RF_READER_A_GATE,
+			       NFC_HCI_EVT_END_OPERATION, NULL, 0);
+	if (r < 0)
+		return r;
+
+	r = nfc_hci_get_param(hdev, NFC_HCI_ID_MGMT_GATE,
+			      NFC_HCI_ID_MGMT_VERSION_SW, &skb);
+	if (r < 0)
+		return r;
+
+	if (skb->len != FULL_VERSION_LEN) {
+		kfree_skb(skb);
+		return -EINVAL;
+	}
+
+	print_hex_dump(KERN_DEBUG, "FULL VERSION SOFTWARE INFO: ",
+		       DUMP_PREFIX_NONE, 16, 1,
+		       skb->data, FULL_VERSION_LEN, false);
+
+	kfree_skb(skb);
+
+	return 0;
+}
+
+static int st21nfca_hci_xmit(struct nfc_hci_dev *hdev, struct sk_buff *skb)
+{
+	struct st21nfca_hci_info *info = nfc_hci_get_clientdata(hdev);
+
+	return info->phy_ops->write(info->phy_id, skb);
+}
+
+static int st21nfca_hci_start_poll(struct nfc_hci_dev *hdev,
+				   u32 im_protocols, u32 tm_protocols)
+{
+	int r;
+
+	pr_info(DRIVER_DESC ": %s protocols 0x%x 0x%x\n",
+		__func__, im_protocols, tm_protocols);
+
+	r = nfc_hci_send_event(hdev, NFC_HCI_RF_READER_A_GATE,
+			       NFC_HCI_EVT_END_OPERATION, NULL, 0);
+	if (r < 0)
+		return r;
+	if (im_protocols) {
+		/*
+		 * enable polling according to im_protocols & tm_protocols
+		 * - CLOSE pipe according to im_protocols & tm_protocols
+		 */
+		if ((NFC_HCI_RF_READER_B_GATE & im_protocols) == 0) {
+			r = nfc_hci_disconnect_gate(hdev,
+					NFC_HCI_RF_READER_B_GATE);
+			if (r < 0)
+				return r;
+		}
+
+		if ((NFC_HCI_RF_READER_A_GATE & im_protocols) == 0) {
+			r = nfc_hci_disconnect_gate(hdev,
+					NFC_HCI_RF_READER_A_GATE);
+			if (r < 0)
+				return r;
+		}
+
+		if ((ST21NFCA_RF_READER_F_GATE & im_protocols) == 0) {
+			r = nfc_hci_disconnect_gate(hdev,
+					ST21NFCA_RF_READER_F_GATE);
+			if (r < 0)
+				return r;
+		}
+
+		if ((ST21NFCA_RF_READER_14443_3_A_GATE & im_protocols) == 0) {
+			r = nfc_hci_disconnect_gate(hdev,
+					ST21NFCA_RF_READER_14443_3_A_GATE);
+			if (r < 0)
+				return r;
+		}
+
+		if ((ST21NFCA_RF_READER_ISO15693_GATE & im_protocols) == 0) {
+			r = nfc_hci_disconnect_gate(hdev,
+					ST21NFCA_RF_READER_ISO15693_GATE);
+			if (r < 0)
+				return r;
+		}
+
+		r = nfc_hci_send_event(hdev, NFC_HCI_RF_READER_A_GATE,
+				       NFC_HCI_EVT_READER_REQUESTED, NULL, 0);
+		if (r < 0)
+			nfc_hci_send_event(hdev, NFC_HCI_RF_READER_A_GATE,
+					   NFC_HCI_EVT_END_OPERATION, NULL, 0);
+	}
+	return r;
+}
+
+static int st21nfca_get_iso14443_3_atqa(struct nfc_hci_dev *hdev, u16 *atqa)
+{
+	int r;
+	struct sk_buff *atqa_skb = NULL;
+
+	r = nfc_hci_get_param(hdev, ST21NFCA_RF_READER_14443_3_A_GATE,
+			      ST21NFCA_RF_READER_14443_3_A_ATQA, &atqa_skb);
+	if (r < 0)
+		goto exit;
+
+	if (atqa_skb->len != 2) {
+		r = -EPROTO;
+		goto exit;
+	}
+
+	*atqa = be16_to_cpu(*(__be16 *) atqa_skb->data);
+
+exit:
+	kfree_skb(atqa_skb);
+	return r;
+}
+
+static int st21nfca_get_iso14443_3_sak(struct nfc_hci_dev *hdev, u8 *sak)
+{
+	int r;
+	struct sk_buff *sak_skb = NULL;
+
+	r = nfc_hci_get_param(hdev, ST21NFCA_RF_READER_14443_3_A_GATE,
+			      ST21NFCA_RF_READER_14443_3_A_SAK, &sak_skb);
+	if (r < 0)
+		goto exit;
+
+	if (sak_skb->len != 1) {
+		r = -EPROTO;
+		goto exit;
+	}
+
+	*sak = sak_skb->data[0];
+
+exit:
+	kfree_skb(sak_skb);
+	return r;
+}
+
+static int st21nfca_get_iso14443_3_uid(struct nfc_hci_dev *hdev, u8 *gate,
+				       int *len)
+{
+	int r;
+	struct sk_buff *uid_skb = NULL;
+
+	r = nfc_hci_get_param(hdev, ST21NFCA_RF_READER_14443_3_A_GATE,
+			      ST21NFCA_RF_READER_14443_3_A_UID, &uid_skb);
+	if (r < 0)
+		goto exit;
+
+	if (uid_skb->len == 0 || uid_skb->len > NFC_NFCID1_MAXSIZE) {
+		r = -EPROTO;
+		goto exit;
+	}
+
+	gate = uid_skb->data;
+	*len = uid_skb->len;
+exit:
+	kfree_skb(uid_skb);
+	return r;
+}
+
+static int st21nfca_get_iso15693_inventory(struct nfc_hci_dev *hdev,
+					   struct nfc_target *target)
+{
+	int r;
+	struct sk_buff *inventory_skb = NULL;
+
+	r = nfc_hci_get_param(hdev, ST21NFCA_RF_READER_ISO15693_GATE,
+			      ST21NFCA_RF_READER_ISO15693_INVENTORY,
+			      &inventory_skb);
+	if (r < 0)
+		goto exit;
+
+	skb_pull(inventory_skb, 2);
+
+	if (inventory_skb->len == 0 ||
+	    inventory_skb->len > NFC_ISO15693_UID_MAXSIZE) {
+		r = -EPROTO;
+		goto exit;
+	}
+
+	memcpy(target->iso15693_uid, inventory_skb->data, inventory_skb->len);
+	target->iso15693_dsfid	= inventory_skb->data[1];
+	target->is_iso15693 = 1;
+exit:
+	kfree_skb(inventory_skb);
+	return r;
+}
+
+static int st21nfca_hci_target_from_gate(struct nfc_hci_dev *hdev, u8 gate,
+					 struct nfc_target *target)
+{
+	int r, len;
+	u16 atqa;
+	u8 sak;
+	u8 uid[NFC_NFCID1_MAXSIZE];
+
+	switch (gate) {
+	case ST21NFCA_RF_READER_F_GATE:
+		target->supported_protocols = NFC_PROTO_FELICA_MASK;
+		break;
+	case ST21NFCA_RF_READER_14443_3_A_GATE:
+		/* ISO14443-3 type 1 or 2 tags */
+		r = st21nfca_get_iso14443_3_atqa(hdev, &atqa);
+		if (r < 0)
+			return r;
+		if (atqa == 0x000c) {
+			target->supported_protocols = NFC_PROTO_JEWEL_MASK;
+			target->sens_res = 0x0c00;
+		} else {
+			r = st21nfca_get_iso14443_3_sak(hdev, &sak);
+			if (r < 0)
+				return r;
+
+			r = st21nfca_get_iso14443_3_uid(hdev, uid, &len);
+			if (r < 0)
+				return r;
+
+			target->supported_protocols =
+			    nfc_hci_sak_to_protocol(sak);
+			if (target->supported_protocols == 0xffffffff)
+				return -EPROTO;
+
+			target->sens_res = atqa;
+			target->sel_res = sak;
+			memcpy(target->nfcid1, uid, len);
+			target->nfcid1_len = len;
+		}
+
+		break;
+	case ST21NFCA_RF_READER_ISO15693_GATE:
+		target->supported_protocols = NFC_PROTO_ISO15693_MASK;
+		r = st21nfca_get_iso15693_inventory(hdev, target);
+		if (r < 0)
+			return r;
+		break;
+	default:
+		return -EPROTO;
+	}
+
+	return 0;
+}
+
+#define ST21NFCA_CB_TYPE_READER_ISO15693 1
+static void st21nfca_hci_data_exchange_cb(void *context, struct sk_buff *skb,
+					  int err)
+{
+	struct st21nfca_hci_info *info = context;
+
+	switch (info->async_cb_type) {
+	case ST21NFCA_CB_TYPE_READER_ISO15693:
+		if (err == 0)
+			skb_trim(skb, skb->len - 1);
+		info->async_cb(info->async_cb_context, skb, err);
+		break;
+	default:
+		if (err == 0)
+			kfree_skb(skb);
+		break;
+	}
+}
+
+/*
+ * Returns:
+ * <= 0: driver handled the data exchange
+ *    1: driver doesn't especially handle, please do standard processing
+ */
+static int st21nfca_hci_im_transceive(struct nfc_hci_dev *hdev,
+				      struct nfc_target *target,
+				      struct sk_buff *skb,
+				      data_exchange_cb_t cb, void *cb_context)
+{
+	struct st21nfca_hci_info *info = nfc_hci_get_clientdata(hdev);
+
+	pr_info(DRIVER_DESC ": %s for gate=%d len=%d\n", __func__,
+		target->hci_reader_gate, skb->len);
+
+	switch (target->hci_reader_gate) {
+	case ST21NFCA_RF_READER_F_GATE:
+		*skb_push(skb, 1) = 0x1a;
+		return nfc_hci_send_cmd_async(hdev, target->hci_reader_gate,
+					      ST21NFCA_WR_XCHG_DATA, skb->data,
+					      skb->len, cb, cb_context);
+	case ST21NFCA_RF_READER_14443_3_A_GATE:
+		*skb_push(skb, 1) = 0x1a;	/* CTR, see spec:10.2.2.1 */
+
+		return nfc_hci_send_cmd_async(hdev, target->hci_reader_gate,
+					      ST21NFCA_WR_XCHG_DATA, skb->data,
+					      skb->len, cb, cb_context);
+	case ST21NFCA_RF_READER_ISO15693_GATE:
+		info->async_cb_type = ST21NFCA_CB_TYPE_READER_ISO15693;
+		info->async_cb = cb;
+		info->async_cb_context = cb_context;
+
+		*skb_push(skb, 1) = 0x17;
+
+		return nfc_hci_send_cmd_async(hdev, target->hci_reader_gate,
+					      ST21NFCA_WR_XCHG_DATA, skb->data,
+					      skb->len,
+					      st21nfca_hci_data_exchange_cb,
+					      info);
+		break;
+	default:
+		return 1;
+	}
+}
+
+static int st21nfca_hci_check_presence(struct nfc_hci_dev *hdev,
+				       struct nfc_target *target)
+{
+	u8 fwi = 0x11;
+	switch (target->hci_reader_gate) {
+	case NFC_HCI_RF_READER_A_GATE:
+	case NFC_HCI_RF_READER_B_GATE:
+		/*
+		 * PRESENCE_CHECK on those gates is available
+		 * However, the answer to this command is taking 3 * fwi
+		 * if the card is no present.
+		 * Instead, we send an empty I-Frame with a very short
+		 * configurable fwi ~604µs.
+		 */
+		return nfc_hci_send_cmd(hdev, target->hci_reader_gate,
+					ST21NFCA_WR_XCHG_DATA, &fwi, 1, NULL);
+	case ST21NFCA_RF_READER_14443_3_A_GATE:
+		return nfc_hci_send_cmd(hdev, target->hci_reader_gate,
+					ST21NFCA_RF_READER_CMD_PRESENCE_CHECK,
+					NULL, 0, NULL);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static struct nfc_hci_ops st21nfca_hci_ops = {
+	.open = st21nfca_hci_open,
+	.close = st21nfca_hci_close,
+	.load_session = st21nfca_hci_load_session,
+	.hci_ready = st21nfca_hci_ready,
+	.xmit = st21nfca_hci_xmit,
+	.start_poll = st21nfca_hci_start_poll,
+	.target_from_gate = st21nfca_hci_target_from_gate,
+	.im_transceive = st21nfca_hci_im_transceive,
+	.check_presence = st21nfca_hci_check_presence,
+};
+
+int st21nfca_hci_probe(void *phy_id, struct nfc_phy_ops *phy_ops,
+		       char *llc_name, int phy_headroom, int phy_tailroom,
+		       int phy_payload, struct nfc_hci_dev **hdev)
+{
+	struct st21nfca_hci_info *info;
+	int r = 0;
+	int dev_num;
+	u32 protocols;
+	struct nfc_hci_init_data init_data;
+	unsigned long quirks = 0;
+
+	info = kzalloc(sizeof(struct st21nfca_hci_info), GFP_KERNEL);
+	if (!info) {
+		r = -ENOMEM;
+		goto err_alloc_hdev;
+	}
+
+	info->phy_ops = phy_ops;
+	info->phy_id = phy_id;
+	info->state = ST21NFCA_ST_COLD;
+	mutex_init(&info->info_lock);
+
+	init_data.gate_count = ARRAY_SIZE(st21nfca_gates);
+
+	memcpy(init_data.gates, st21nfca_gates, sizeof(st21nfca_gates));
+
+	/*
+	 * Session id must include the driver name + i2c bus addr
+	 * persistent info to discriminate 2 identical chips
+	 */
+	dev_num = find_first_zero_bit(dev_mask, ST21NFCA_NUM_DEVICES);
+	if (dev_num >= ST21NFCA_NUM_DEVICES)
+		goto err_alloc_hdev;
+
+	scnprintf(init_data.session_id, sizeof(init_data.session_id), "%s%2x",
+		  "ST21AH", dev_num);
+
+	protocols = NFC_PROTO_JEWEL_MASK |
+	    NFC_PROTO_MIFARE_MASK |
+	    NFC_PROTO_FELICA_MASK |
+	    NFC_PROTO_ISO14443_MASK |
+	    NFC_PROTO_ISO14443_B_MASK |
+	    NFC_PROTO_ISO15693_MASK;
+
+	set_bit(NFC_HCI_QUIRK_SHORT_CLEAR, &quirks);
+
+	info->hdev =
+	    nfc_hci_allocate_device(&st21nfca_hci_ops, &init_data, quirks,
+				    protocols, llc_name,
+				    phy_headroom + ST21NFCA_CMDS_HEADROOM,
+				    phy_tailroom, phy_payload);
+
+	if (!info->hdev) {
+		pr_err("Cannot allocate nfc hdev.\n");
+		r = -ENOMEM;
+		goto err_alloc_hdev;
+	}
+
+	nfc_hci_set_clientdata(info->hdev, info);
+
+	r = nfc_hci_register_device(info->hdev);
+	if (r)
+		goto err_regdev;
+
+	*hdev = info->hdev;
+
+	return 0;
+
+err_regdev:
+	nfc_hci_free_device(info->hdev);
+
+err_alloc_hdev:
+	kfree(info);
+
+	return r;
+}
+EXPORT_SYMBOL(st21nfca_hci_probe);
+
+void st21nfca_hci_remove(struct nfc_hci_dev *hdev)
+{
+	struct st21nfca_hci_info *info = nfc_hci_get_clientdata(hdev);
+
+	nfc_hci_unregister_device(hdev);
+	nfc_hci_free_device(hdev);
+	kfree(info);
+}
+EXPORT_SYMBOL(st21nfca_hci_remove);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION(DRIVER_DESC);
diff --git a/drivers/nfc/st21nfca/st21nfca.h b/drivers/nfc/st21nfca/st21nfca.h
new file mode 100644
index 000000000000..334cd90bcc8c
--- /dev/null
+++ b/drivers/nfc/st21nfca/st21nfca.h
@@ -0,0 +1,87 @@
+/*
+ * Copyright (C) 2014  STMicroelectronics SAS. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __LOCAL_ST21NFCA_H_
+#define __LOCAL_ST21NFCA_H_
+
+#include <net/nfc/hci.h>
+
+#define HCI_MODE 0
+
+/* framing in HCI mode */
+#define ST21NFCA_SOF_EOF_LEN    2
+
+/* Almost every time value is 0 */
+#define ST21NFCA_HCI_LLC_LEN    1
+
+/* Size in worst case :
+ * In normal case CRC len = 2 but byte stuffing
+ * may appear in case one CRC byte = ST21NFCA_SOF_EOF
+ */
+#define ST21NFCA_HCI_LLC_CRC    4
+
+#define ST21NFCA_HCI_LLC_LEN_CRC        (ST21NFCA_SOF_EOF_LEN + \
+						ST21NFCA_HCI_LLC_LEN + \
+						ST21NFCA_HCI_LLC_CRC)
+#define ST21NFCA_HCI_LLC_MIN_SIZE       (1 + ST21NFCA_HCI_LLC_LEN_CRC)
+
+/* Worst case when adding byte stuffing between each byte */
+#define ST21NFCA_HCI_LLC_MAX_PAYLOAD    29
+#define ST21NFCA_HCI_LLC_MAX_SIZE       (ST21NFCA_HCI_LLC_LEN_CRC + 1 + \
+					ST21NFCA_HCI_LLC_MAX_PAYLOAD)
+
+#define DRIVER_DESC "HCI NFC driver for ST21NFCA"
+
+#define ST21NFCA_HCI_MODE 0
+
+#define ST21NFCA_NUM_DEVICES 256
+
+int st21nfca_hci_probe(void *phy_id, struct nfc_phy_ops *phy_ops,
+		       char *llc_name, int phy_headroom, int phy_tailroom,
+		       int phy_payload, struct nfc_hci_dev **hdev);
+void st21nfca_hci_remove(struct nfc_hci_dev *hdev);
+
+enum st21nfca_state {
+	ST21NFCA_ST_COLD,
+	ST21NFCA_ST_READY,
+};
+
+struct st21nfca_hci_info {
+	struct nfc_phy_ops *phy_ops;
+	void *phy_id;
+
+	struct nfc_hci_dev *hdev;
+
+	enum st21nfca_state state;
+
+	struct mutex info_lock;
+
+	int async_cb_type;
+	data_exchange_cb_t async_cb;
+	void *async_cb_context;
+
+} __packed;
+
+/* Reader RF commands */
+#define ST21NFCA_WR_XCHG_DATA            0x10
+
+#define ST21NFCA_RF_READER_F_GATE               0x14
+#define ST21NFCA_RF_READER_F_DATARATE 0x01
+#define ST21NFCA_RF_READER_F_DATARATE_106 0x01
+#define ST21NFCA_RF_READER_F_DATARATE_212 0x02
+#define ST21NFCA_RF_READER_F_DATARATE_424 0x04
+
+#endif /* __LOCAL_ST21NFCA_H_ */
diff --git a/drivers/nfc/trf7970a.c b/drivers/nfc/trf7970a.c
index d9babe986473..3b78b031e617 100644
--- a/drivers/nfc/trf7970a.c
+++ b/drivers/nfc/trf7970a.c
@@ -16,6 +16,7 @@
 #include <linux/device.h>
 #include <linux/netdevice.h>
 #include <linux/interrupt.h>
+#include <linux/pm_runtime.h>
 #include <linux/nfc.h>
 #include <linux/skbuff.h>
 #include <linux/delay.h>
@@ -67,14 +68,14 @@
  * only the SRX bit set, it means that all of the data has been received
  * (once what's in the fifo has been read).  However, depending on timing
  * an interrupt status with only the SRX bit set may not be recived.  In
- * those cases, the timeout mechanism is used to wait 5 ms in case more
- * data arrives.  After 5 ms, it is assumed that all of the data has been
+ * those cases, the timeout mechanism is used to wait 20 ms in case more
+ * data arrives.  After 20 ms, it is assumed that all of the data has been
  * received and the accumulated rx data is sent upstream.  The
  * 'TRF7970A_ST_WAIT_FOR_RX_DATA_CONT' state is used for this purpose
  * (i.e., it indicates that some data has been received but we're not sure
  * if there is more coming so a timeout in this state means all data has
- * been received and there isn't an error).  The delay is 5 ms since delays
- * over 2 ms have been observed during testing (a little extra just in case).
+ * been received and there isn't an error).  The delay is 20 ms since delays
+ * of ~16 ms have been observed during testing.
  *
  * Type 2 write and sector select commands respond with a 4-bit ACK or NACK.
  * Having only 4 bits in the FIFO won't normally generate an interrupt so
@@ -104,8 +105,11 @@
 
 #define TRF7970A_SUPPORTED_PROTOCOLS \
 		(NFC_PROTO_MIFARE_MASK | NFC_PROTO_ISO14443_MASK |	\
+		 NFC_PROTO_ISO14443_B_MASK | NFC_PROTO_FELICA_MASK | \
 		 NFC_PROTO_ISO15693_MASK)
 
+#define TRF7970A_AUTOSUSPEND_DELAY		30000 /* 30 seconds */
+
 /* TX data must be prefixed with a FIFO reset cmd, a cmd that depends
  * on what the current framing is, the address of the TX length byte 1
  * register (0x1d), and the 2 byte length of the data to be transmitted.
@@ -120,7 +124,7 @@
 /* TX length is 3 nibbles long ==> 4KB - 1 bytes max */
 #define TRF7970A_TX_MAX				(4096 - 1)
 
-#define TRF7970A_WAIT_FOR_RX_DATA_TIMEOUT	5
+#define TRF7970A_WAIT_FOR_RX_DATA_TIMEOUT	20
 #define TRF7970A_WAIT_FOR_FIFO_DRAIN_TIMEOUT	3
 #define TRF7970A_WAIT_TO_ISSUE_ISO15693_EOF	20
 
@@ -330,13 +334,15 @@ struct trf7970a {
 	struct regulator		*regulator;
 	struct nfc_digital_dev		*ddev;
 	u32				quirks;
-	bool				powering_up;
 	bool				aborting;
 	struct sk_buff			*tx_skb;
 	struct sk_buff			*rx_skb;
 	nfc_digital_cmd_complete_t	cb;
 	void				*cb_arg;
+	u8				chip_status_ctrl;
 	u8				iso_ctrl;
+	u8				iso_ctrl_tech;
+	u8				modulator_sys_clk_ctrl;
 	u8				special_fcn_reg1;
 	int				technology;
 	int				framing;
@@ -681,7 +687,9 @@ static irqreturn_t trf7970a_irq(int irq, void *dev_id)
 			trf->ignore_timeout =
 				!cancel_delayed_work(&trf->timeout_work);
 			trf7970a_drain_fifo(trf, status);
-		} else if (!(status & TRF7970A_IRQ_STATUS_TX)) {
+		} else if (status == TRF7970A_IRQ_STATUS_TX) {
+			trf7970a_cmd(trf, TRF7970A_CMD_FIFO_RESET);
+		} else {
 			trf7970a_send_err_upstream(trf, -EIO);
 		}
 		break;
@@ -757,8 +765,8 @@ static int trf7970a_init(struct trf7970a *trf)
 	if (ret)
 		goto err_out;
 
-	ret = trf7970a_write(trf, TRF7970A_MODULATOR_SYS_CLK_CTRL,
-			TRF7970A_MODULATOR_DEPTH_OOK);
+	/* Must clear NFC Target Detection Level reg due to erratum */
+	ret = trf7970a_write(trf, TRF7970A_NFC_TARGET_LEVEL, 0);
 	if (ret)
 		goto err_out;
 
@@ -774,12 +782,7 @@ static int trf7970a_init(struct trf7970a *trf)
 
 	trf->special_fcn_reg1 = 0;
 
-	ret = trf7970a_write(trf, TRF7970A_CHIP_STATUS_CTRL,
-			TRF7970A_CHIP_STATUS_RF_ON |
-				TRF7970A_CHIP_STATUS_VRS5_3);
-	if (ret)
-		goto err_out;
-
+	trf->iso_ctrl = 0xff;
 	return 0;
 
 err_out:
@@ -791,53 +794,29 @@ static void trf7970a_switch_rf_off(struct trf7970a *trf)
 {
 	dev_dbg(trf->dev, "Switching rf off\n");
 
-	gpio_set_value(trf->en_gpio, 0);
-	gpio_set_value(trf->en2_gpio, 0);
+	trf->chip_status_ctrl &= ~TRF7970A_CHIP_STATUS_RF_ON;
+
+	trf7970a_write(trf, TRF7970A_CHIP_STATUS_CTRL, trf->chip_status_ctrl);
 
 	trf->aborting = false;
 	trf->state = TRF7970A_ST_OFF;
+
+	pm_runtime_mark_last_busy(trf->dev);
+	pm_runtime_put_autosuspend(trf->dev);
 }
 
-static int trf7970a_switch_rf_on(struct trf7970a *trf)
+static void trf7970a_switch_rf_on(struct trf7970a *trf)
 {
-	unsigned long delay;
-	int ret;
-
 	dev_dbg(trf->dev, "Switching rf on\n");
 
-	if (trf->powering_up)
-		usleep_range(5000, 6000);
-
-	gpio_set_value(trf->en2_gpio, 1);
-	usleep_range(1000, 2000);
-	gpio_set_value(trf->en_gpio, 1);
+	pm_runtime_get_sync(trf->dev);
 
-	/* The delay between enabling the trf7970a and issuing the first
-	 * command is significantly longer the very first time after powering
-	 * up.  Make sure the longer delay is only done the first time.
-	 */
-	if (trf->powering_up) {
-		delay = 20000;
-		trf->powering_up = false;
-	} else {
-		delay = 5000;
-	}
-
-	usleep_range(delay, delay + 1000);
-
-	ret = trf7970a_init(trf);
-	if (ret)
-		trf7970a_switch_rf_off(trf);
-	else
-		trf->state = TRF7970A_ST_IDLE;
-
-	return ret;
+	trf->state = TRF7970A_ST_IDLE;
 }
 
 static int trf7970a_switch_rf(struct nfc_digital_dev *ddev, bool on)
 {
 	struct trf7970a *trf = nfc_digital_get_drvdata(ddev);
-	int ret = 0;
 
 	dev_dbg(trf->dev, "Switching RF - state: %d, on: %d\n", trf->state, on);
 
@@ -846,7 +825,7 @@ static int trf7970a_switch_rf(struct nfc_digital_dev *ddev, bool on)
 	if (on) {
 		switch (trf->state) {
 		case TRF7970A_ST_OFF:
-			ret = trf7970a_switch_rf_on(trf);
+			trf7970a_switch_rf_on(trf);
 			break;
 		case TRF7970A_ST_IDLE:
 		case TRF7970A_ST_IDLE_RX_BLOCKED:
@@ -871,7 +850,7 @@ static int trf7970a_switch_rf(struct nfc_digital_dev *ddev, bool on)
 	}
 
 	mutex_unlock(&trf->lock);
-	return ret;
+	return 0;
 }
 
 static int trf7970a_config_rf_tech(struct trf7970a *trf, int tech)
@@ -882,10 +861,24 @@ static int trf7970a_config_rf_tech(struct trf7970a *trf, int tech)
 
 	switch (tech) {
 	case NFC_DIGITAL_RF_TECH_106A:
-		trf->iso_ctrl = TRF7970A_ISO_CTRL_14443A_106;
+		trf->iso_ctrl_tech = TRF7970A_ISO_CTRL_14443A_106;
+		trf->modulator_sys_clk_ctrl = TRF7970A_MODULATOR_DEPTH_OOK;
+		break;
+	case NFC_DIGITAL_RF_TECH_106B:
+		trf->iso_ctrl_tech = TRF7970A_ISO_CTRL_14443B_106;
+		trf->modulator_sys_clk_ctrl = TRF7970A_MODULATOR_DEPTH_ASK10;
+		break;
+	case NFC_DIGITAL_RF_TECH_212F:
+		trf->iso_ctrl_tech = TRF7970A_ISO_CTRL_FELICA_212;
+		trf->modulator_sys_clk_ctrl = TRF7970A_MODULATOR_DEPTH_ASK10;
+		break;
+	case NFC_DIGITAL_RF_TECH_424F:
+		trf->iso_ctrl_tech = TRF7970A_ISO_CTRL_FELICA_424;
+		trf->modulator_sys_clk_ctrl = TRF7970A_MODULATOR_DEPTH_ASK10;
 		break;
 	case NFC_DIGITAL_RF_TECH_ISO15693:
-		trf->iso_ctrl = TRF7970A_ISO_CTRL_15693_SGL_1OF4_2648;
+		trf->iso_ctrl_tech = TRF7970A_ISO_CTRL_15693_SGL_1OF4_2648;
+		trf->modulator_sys_clk_ctrl = TRF7970A_MODULATOR_DEPTH_OOK;
 		break;
 	default:
 		dev_dbg(trf->dev, "Unsupported rf technology: %d\n", tech);
@@ -899,24 +892,31 @@ static int trf7970a_config_rf_tech(struct trf7970a *trf, int tech)
 
 static int trf7970a_config_framing(struct trf7970a *trf, int framing)
 {
+	u8 iso_ctrl = trf->iso_ctrl_tech;
+	int ret;
+
 	dev_dbg(trf->dev, "framing: %d\n", framing);
 
 	switch (framing) {
 	case NFC_DIGITAL_FRAMING_NFCA_SHORT:
 	case NFC_DIGITAL_FRAMING_NFCA_STANDARD:
 		trf->tx_cmd = TRF7970A_CMD_TRANSMIT_NO_CRC;
-		trf->iso_ctrl |= TRF7970A_ISO_CTRL_RX_CRC_N;
+		iso_ctrl |= TRF7970A_ISO_CTRL_RX_CRC_N;
 		break;
 	case NFC_DIGITAL_FRAMING_NFCA_STANDARD_WITH_CRC_A:
 	case NFC_DIGITAL_FRAMING_NFCA_T4T:
+	case NFC_DIGITAL_FRAMING_NFCB:
+	case NFC_DIGITAL_FRAMING_NFCB_T4T:
+	case NFC_DIGITAL_FRAMING_NFCF:
+	case NFC_DIGITAL_FRAMING_NFCF_T3T:
 	case NFC_DIGITAL_FRAMING_ISO15693_INVENTORY:
 	case NFC_DIGITAL_FRAMING_ISO15693_T5T:
 		trf->tx_cmd = TRF7970A_CMD_TRANSMIT;
-		trf->iso_ctrl &= ~TRF7970A_ISO_CTRL_RX_CRC_N;
+		iso_ctrl &= ~TRF7970A_ISO_CTRL_RX_CRC_N;
 		break;
 	case NFC_DIGITAL_FRAMING_NFCA_T2T:
 		trf->tx_cmd = TRF7970A_CMD_TRANSMIT;
-		trf->iso_ctrl |= TRF7970A_ISO_CTRL_RX_CRC_N;
+		iso_ctrl |= TRF7970A_ISO_CTRL_RX_CRC_N;
 		break;
 	default:
 		dev_dbg(trf->dev, "Unsupported Framing: %d\n", framing);
@@ -925,24 +925,46 @@ static int trf7970a_config_framing(struct trf7970a *trf, int framing)
 
 	trf->framing = framing;
 
-	return trf7970a_write(trf, TRF7970A_ISO_CTRL, trf->iso_ctrl);
+	if (iso_ctrl != trf->iso_ctrl) {
+		ret = trf7970a_write(trf, TRF7970A_ISO_CTRL, iso_ctrl);
+		if (ret)
+			return ret;
+
+		trf->iso_ctrl = iso_ctrl;
+
+		ret = trf7970a_write(trf, TRF7970A_MODULATOR_SYS_CLK_CTRL,
+				trf->modulator_sys_clk_ctrl);
+		if (ret)
+			return ret;
+	}
+
+	if (!(trf->chip_status_ctrl & TRF7970A_CHIP_STATUS_RF_ON)) {
+		ret = trf7970a_write(trf, TRF7970A_CHIP_STATUS_CTRL,
+				trf->chip_status_ctrl |
+					TRF7970A_CHIP_STATUS_RF_ON);
+		if (ret)
+			return ret;
+
+		trf->chip_status_ctrl |= TRF7970A_CHIP_STATUS_RF_ON;
+
+		usleep_range(5000, 6000);
+	}
+
+	return 0;
 }
 
 static int trf7970a_in_configure_hw(struct nfc_digital_dev *ddev, int type,
 		int param)
 {
 	struct trf7970a *trf = nfc_digital_get_drvdata(ddev);
-	int ret = 0;
+	int ret;
 
 	dev_dbg(trf->dev, "Configure hw - type: %d, param: %d\n", type, param);
 
 	mutex_lock(&trf->lock);
 
-	if (trf->state == TRF7970A_ST_OFF) {
-		ret = trf7970a_switch_rf_on(trf);
-		if (ret)
-			goto err_out;
-	}
+	if (trf->state == TRF7970A_ST_OFF)
+		trf7970a_switch_rf_on(trf);
 
 	switch (type) {
 	case NFC_DIGITAL_CONFIG_RF_TECH:
@@ -956,7 +978,6 @@ static int trf7970a_in_configure_hw(struct nfc_digital_dev *ddev, int type,
 		ret = -EINVAL;
 	}
 
-err_out:
 	mutex_unlock(&trf->lock);
 	return ret;
 }
@@ -1191,7 +1212,18 @@ static void trf7970a_abort_cmd(struct nfc_digital_dev *ddev)
 	dev_dbg(trf->dev, "Abort process initiated\n");
 
 	mutex_lock(&trf->lock);
-	trf->aborting = true;
+
+	switch (trf->state) {
+	case TRF7970A_ST_WAIT_FOR_TX_FIFO:
+	case TRF7970A_ST_WAIT_FOR_RX_DATA:
+	case TRF7970A_ST_WAIT_FOR_RX_DATA_CONT:
+	case TRF7970A_ST_WAIT_TO_ISSUE_EOF:
+		trf->aborting = true;
+		break;
+	default:
+		break;
+	}
+
 	mutex_unlock(&trf->lock);
 }
 
@@ -1206,12 +1238,25 @@ static struct nfc_digital_ops trf7970a_nfc_ops = {
 	.abort_cmd		= trf7970a_abort_cmd,
 };
 
+static int trf7970a_get_autosuspend_delay(struct device_node *np)
+{
+	int autosuspend_delay, ret;
+
+	ret = of_property_read_u32(np, "autosuspend-delay", &autosuspend_delay);
+	if (ret)
+		autosuspend_delay = TRF7970A_AUTOSUSPEND_DELAY;
+
+	of_node_put(np);
+
+	return autosuspend_delay;
+}
+
 static int trf7970a_probe(struct spi_device *spi)
 {
 	struct device_node *np = spi->dev.of_node;
 	const struct spi_device_id *id = spi_get_device_id(spi);
 	struct trf7970a *trf;
-	int ret;
+	int uvolts, autosuspend_delay, ret;
 
 	if (!np) {
 		dev_err(&spi->dev, "No Device Tree entry\n");
@@ -1281,7 +1326,10 @@ static int trf7970a_probe(struct spi_device *spi)
 		goto err_destroy_lock;
 	}
 
-	trf->powering_up = true;
+	uvolts = regulator_get_voltage(trf->regulator);
+
+	if (uvolts > 4000000)
+		trf->chip_status_ctrl = TRF7970A_CHIP_STATUS_VRS5_3;
 
 	trf->ddev = nfc_digital_allocate_device(&trf7970a_nfc_ops,
 			TRF7970A_SUPPORTED_PROTOCOLS,
@@ -1297,6 +1345,12 @@ static int trf7970a_probe(struct spi_device *spi)
 	nfc_digital_set_drvdata(trf->ddev, trf);
 	spi_set_drvdata(spi, trf);
 
+	autosuspend_delay = trf7970a_get_autosuspend_delay(np);
+
+	pm_runtime_set_autosuspend_delay(trf->dev, autosuspend_delay);
+	pm_runtime_use_autosuspend(trf->dev);
+	pm_runtime_enable(trf->dev);
+
 	ret = nfc_digital_register_device(trf->ddev);
 	if (ret) {
 		dev_err(trf->dev, "Can't register NFC digital device: %d\n",
@@ -1307,6 +1361,7 @@ static int trf7970a_probe(struct spi_device *spi)
 	return 0;
 
 err_free_ddev:
+	pm_runtime_disable(trf->dev);
 	nfc_digital_free_device(trf->ddev);
 err_disable_regulator:
 	regulator_disable(trf->regulator);
@@ -1321,15 +1376,16 @@ static int trf7970a_remove(struct spi_device *spi)
 
 	mutex_lock(&trf->lock);
 
-	trf7970a_switch_rf_off(trf);
-	trf7970a_init(trf);
-
 	switch (trf->state) {
 	case TRF7970A_ST_WAIT_FOR_TX_FIFO:
 	case TRF7970A_ST_WAIT_FOR_RX_DATA:
 	case TRF7970A_ST_WAIT_FOR_RX_DATA_CONT:
 	case TRF7970A_ST_WAIT_TO_ISSUE_EOF:
 		trf7970a_send_err_upstream(trf, -ECANCELED);
+		/* FALLTHROUGH */
+	case TRF7970A_ST_IDLE:
+	case TRF7970A_ST_IDLE_RX_BLOCKED:
+		pm_runtime_put_sync(trf->dev);
 		break;
 	default:
 		break;
@@ -1337,6 +1393,8 @@ static int trf7970a_remove(struct spi_device *spi)
 
 	mutex_unlock(&trf->lock);
 
+	pm_runtime_disable(trf->dev);
+
 	nfc_digital_unregister_device(trf->ddev);
 	nfc_digital_free_device(trf->ddev);
 
@@ -1347,6 +1405,70 @@ static int trf7970a_remove(struct spi_device *spi)
 	return 0;
 }
 
+#ifdef CONFIG_PM_RUNTIME
+static int trf7970a_pm_runtime_suspend(struct device *dev)
+{
+	struct spi_device *spi = container_of(dev, struct spi_device, dev);
+	struct trf7970a *trf = spi_get_drvdata(spi);
+	int ret;
+
+	dev_dbg(dev, "Runtime suspend\n");
+
+	if (trf->state != TRF7970A_ST_OFF) {
+		dev_dbg(dev, "Can't suspend - not in OFF state (%d)\n",
+				trf->state);
+		return -EBUSY;
+	}
+
+	gpio_set_value(trf->en_gpio, 0);
+	gpio_set_value(trf->en2_gpio, 0);
+
+	ret = regulator_disable(trf->regulator);
+	if (ret)
+		dev_err(dev, "%s - Can't disable VIN: %d\n", __func__, ret);
+
+	return ret;
+}
+
+static int trf7970a_pm_runtime_resume(struct device *dev)
+{
+	struct spi_device *spi = container_of(dev, struct spi_device, dev);
+	struct trf7970a *trf = spi_get_drvdata(spi);
+	int ret;
+
+	dev_dbg(dev, "Runtime resume\n");
+
+	ret = regulator_enable(trf->regulator);
+	if (ret) {
+		dev_err(dev, "%s - Can't enable VIN: %d\n", __func__, ret);
+		return ret;
+	}
+
+	usleep_range(5000, 6000);
+
+	gpio_set_value(trf->en2_gpio, 1);
+	usleep_range(1000, 2000);
+	gpio_set_value(trf->en_gpio, 1);
+
+	usleep_range(20000, 21000);
+
+	ret = trf7970a_init(trf);
+	if (ret) {
+		dev_err(dev, "%s - Can't initialize: %d\n", __func__, ret);
+		return ret;
+	}
+
+	pm_runtime_mark_last_busy(dev);
+
+	return 0;
+}
+#endif
+
+static const struct dev_pm_ops trf7970a_pm_ops = {
+	SET_RUNTIME_PM_OPS(trf7970a_pm_runtime_suspend,
+			trf7970a_pm_runtime_resume, NULL)
+};
+
 static const struct spi_device_id trf7970a_id_table[] = {
 	{ "trf7970a", TRF7970A_QUIRK_IRQ_STATUS_READ_ERRATA },
 	{ }
@@ -1360,6 +1482,7 @@ static struct spi_driver trf7970a_spi_driver = {
 	.driver		= {
 		.name	= "trf7970a",
 		.owner	= THIS_MODULE,
+		.pm	= &trf7970a_pm_ops,
 	},
 };
 
diff --git a/drivers/of/base.c b/drivers/of/base.c
index 8368d96ae7b4..b9864806e9b8 100644
--- a/drivers/of/base.c
+++ b/drivers/of/base.c
@@ -227,7 +227,8 @@ static int __of_node_add(struct device_node *np)
 	np->kobj.kset = of_kset;
 	if (!np->parent) {
 		/* Nodes without parents are new top level trees */
-		rc = kobject_add(&np->kobj, NULL, safe_name(&of_kset->kobj, "base"));
+		rc = kobject_add(&np->kobj, NULL, "%s",
+				 safe_name(&of_kset->kobj, "base"));
 	} else {
 		name = safe_name(&np->parent->kobj, kbasename(np->full_name));
 		if (!name || !name[0])
@@ -1960,9 +1961,9 @@ int of_attach_node(struct device_node *np)
 
 	raw_spin_lock_irqsave(&devtree_lock, flags);
 	np->sibling = np->parent->child;
-	np->allnext = of_allnodes;
+	np->allnext = np->parent->allnext;
+	np->parent->allnext = np;
 	np->parent->child = np;
-	of_allnodes = np;
 	of_node_clear_flag(np, OF_DETACHED);
 	raw_spin_unlock_irqrestore(&devtree_lock, flags);
 
diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
index c4cddf0cd96d..b777d8f46bd5 100644
--- a/drivers/of/fdt.c
+++ b/drivers/of/fdt.c
@@ -880,6 +880,21 @@ void __init __weak early_init_dt_add_memory_arch(u64 base, u64 size)
 	const u64 phys_offset = __pa(PAGE_OFFSET);
 	base &= PAGE_MASK;
 	size &= PAGE_MASK;
+
+	if (sizeof(phys_addr_t) < sizeof(u64)) {
+		if (base > ULONG_MAX) {
+			pr_warning("Ignoring memory block 0x%llx - 0x%llx\n",
+					base, base + size);
+			return;
+		}
+
+		if (base + size > ULONG_MAX) {
+			pr_warning("Ignoring memory range 0x%lx - 0x%llx\n",
+					ULONG_MAX, base + size);
+			size = ULONG_MAX - base;
+		}
+	}
+
 	if (base + size < phys_offset) {
 		pr_warning("Ignoring memory block 0x%llx - 0x%llx\n",
 			   base, base + size);
diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 9a95831bd065..a3bf2122a8d5 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -14,6 +14,7 @@
 #include <linux/netdevice.h>
 #include <linux/err.h>
 #include <linux/phy.h>
+#include <linux/phy_fixed.h>
 #include <linux/of.h>
 #include <linux/of_irq.h>
 #include <linux/of_mdio.h>
@@ -22,27 +23,6 @@
 MODULE_AUTHOR("Grant Likely <grant.likely@secretlab.ca>");
 MODULE_LICENSE("GPL");
 
-static void of_set_phy_supported(struct phy_device *phydev, u32 max_speed)
-{
-	/* The default values for phydev->supported are provided by the PHY
-	 * driver "features" member, we want to reset to sane defaults fist
-	 * before supporting higher speeds.
-	 */
-	phydev->supported &= PHY_DEFAULT_FEATURES;
-
-	switch (max_speed) {
-	default:
-		return;
-
-	case SPEED_1000:
-		phydev->supported |= PHY_1000BT_FEATURES;
-	case SPEED_100:
-		phydev->supported |= PHY_100BT_FEATURES;
-	case SPEED_10:
-		phydev->supported |= PHY_10BT_FEATURES;
-	}
-}
-
 /* Extract the clause 22 phy ID from the compatible string of the form
  * ethernet-phy-idAAAA.BBBB */
 static int of_get_phy_id(struct device_node *device, u32 *phy_id)
@@ -66,7 +46,6 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi
 	struct phy_device *phy;
 	bool is_c45;
 	int rc;
-	u32 max_speed = 0;
 	u32 phy_id;
 
 	is_c45 = of_device_is_compatible(child,
@@ -103,17 +82,33 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi
 		return 1;
 	}
 
-	/* Set phydev->supported based on the "max-speed" property
-	 * if present */
-	if (!of_property_read_u32(child, "max-speed", &max_speed))
-		of_set_phy_supported(phy, max_speed);
-
 	dev_dbg(&mdio->dev, "registered phy %s at address %i\n",
 		child->name, addr);
 
 	return 0;
 }
 
+static int of_mdio_parse_addr(struct device *dev, const struct device_node *np)
+{
+	u32 addr;
+	int ret;
+
+	ret = of_property_read_u32(np, "reg", &addr);
+	if (ret < 0) {
+		dev_err(dev, "%s has invalid PHY address\n", np->full_name);
+		return ret;
+	}
+
+	/* A PHY must have a reg property in the range [0-31] */
+	if (addr >= PHY_MAX_ADDR) {
+		dev_err(dev, "%s PHY address %i is too large\n",
+			np->full_name, addr);
+		return -EINVAL;
+	}
+
+	return addr;
+}
+
 /**
  * of_mdiobus_register - Register mii_bus and create PHYs from the device tree
  * @mdio: pointer to mii_bus structure
@@ -126,9 +121,8 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 {
 	struct device_node *child;
 	const __be32 *paddr;
-	u32 addr;
 	bool scanphys = false;
-	int rc, i, len;
+	int addr, rc, i;
 
 	/* Mask out all PHYs from auto probing.  Instead the PHYs listed in
 	 * the device tree are populated after the bus has been registered */
@@ -148,19 +142,9 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 
 	/* Loop over the child nodes and register a phy_device for each one */
 	for_each_available_child_of_node(np, child) {
-		/* A PHY must have a reg property in the range [0-31] */
-		paddr = of_get_property(child, "reg", &len);
-		if (!paddr || len < sizeof(*paddr)) {
+		addr = of_mdio_parse_addr(&mdio->dev, child);
+		if (addr < 0) {
 			scanphys = true;
-			dev_err(&mdio->dev, "%s has invalid PHY address\n",
-				child->full_name);
-			continue;
-		}
-
-		addr = be32_to_cpup(paddr);
-		if (addr >= PHY_MAX_ADDR) {
-			dev_err(&mdio->dev, "%s PHY address %i is too large\n",
-				child->full_name, addr);
 			continue;
 		}
 
@@ -175,7 +159,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 	/* auto scan for PHYs with empty reg property */
 	for_each_available_child_of_node(np, child) {
 		/* Skip PHYs with reg property set */
-		paddr = of_get_property(child, "reg", &len);
+		paddr = of_get_property(child, "reg", NULL);
 		if (paddr)
 			continue;
 
@@ -198,6 +182,40 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 }
 EXPORT_SYMBOL(of_mdiobus_register);
 
+/**
+ * of_mdiobus_link_phydev - Find a device node for a phy
+ * @mdio: pointer to mii_bus structure
+ * @phydev: phydev for which the of_node pointer should be set
+ *
+ * Walk the list of subnodes of a mdio bus and look for a node that matches the
+ * phy's address with its 'reg' property. If found, set the of_node pointer for
+ * the phy. This allows auto-probed pyh devices to be supplied with information
+ * passed in via DT.
+ */
+void of_mdiobus_link_phydev(struct mii_bus *mdio,
+			    struct phy_device *phydev)
+{
+	struct device *dev = &phydev->dev;
+	struct device_node *child;
+
+	if (dev->of_node || !mdio->dev.of_node)
+		return;
+
+	for_each_available_child_of_node(mdio->dev.of_node, child) {
+		int addr;
+
+		addr = of_mdio_parse_addr(&mdio->dev, child);
+		if (addr < 0)
+			continue;
+
+		if (addr == phydev->addr) {
+			dev->of_node = child;
+			return;
+		}
+	}
+}
+EXPORT_SYMBOL(of_mdiobus_link_phydev);
+
 /* Helper function for of_phy_find_device */
 static int of_phy_match(struct device *dev, void *phy_np)
 {
@@ -245,44 +263,6 @@ struct phy_device *of_phy_connect(struct net_device *dev,
 EXPORT_SYMBOL(of_phy_connect);
 
 /**
- * of_phy_connect_fixed_link - Parse fixed-link property and return a dummy phy
- * @dev: pointer to net_device claiming the phy
- * @hndlr: Link state callback for the network device
- * @iface: PHY data interface type
- *
- * This function is a temporary stop-gap and will be removed soon.  It is
- * only to support the fs_enet, ucc_geth and gianfar Ethernet drivers.  Do
- * not call this function from new drivers.
- */
-struct phy_device *of_phy_connect_fixed_link(struct net_device *dev,
-					     void (*hndlr)(struct net_device *),
-					     phy_interface_t iface)
-{
-	struct device_node *net_np;
-	char bus_id[MII_BUS_ID_SIZE + 3];
-	struct phy_device *phy;
-	const __be32 *phy_id;
-	int sz;
-
-	if (!dev->dev.parent)
-		return NULL;
-
-	net_np = dev->dev.parent->of_node;
-	if (!net_np)
-		return NULL;
-
-	phy_id = of_get_property(net_np, "fixed-link", &sz);
-	if (!phy_id || sz < sizeof(*phy_id))
-		return NULL;
-
-	sprintf(bus_id, PHY_ID_FMT, "fixed-0", be32_to_cpu(phy_id[0]));
-
-	phy = phy_connect(dev, bus_id, hndlr, iface);
-	return IS_ERR(phy) ? NULL : phy;
-}
-EXPORT_SYMBOL(of_phy_connect_fixed_link);
-
-/**
  * of_phy_attach - Attach to a PHY without starting the state machine
  * @dev: pointer to net_device claiming the phy
  * @phy_np: Node pointer for the PHY
@@ -301,3 +281,71 @@ struct phy_device *of_phy_attach(struct net_device *dev,
 	return phy_attach_direct(dev, phy, flags, iface) ? NULL : phy;
 }
 EXPORT_SYMBOL(of_phy_attach);
+
+#if defined(CONFIG_FIXED_PHY)
+/*
+ * of_phy_is_fixed_link() and of_phy_register_fixed_link() must
+ * support two DT bindings:
+ * - the old DT binding, where 'fixed-link' was a property with 5
+ *   cells encoding various informations about the fixed PHY
+ * - the new DT binding, where 'fixed-link' is a sub-node of the
+ *   Ethernet device.
+ */
+bool of_phy_is_fixed_link(struct device_node *np)
+{
+	struct device_node *dn;
+	int len;
+
+	/* New binding */
+	dn = of_get_child_by_name(np, "fixed-link");
+	if (dn) {
+		of_node_put(dn);
+		return true;
+	}
+
+	/* Old binding */
+	if (of_get_property(np, "fixed-link", &len) &&
+	    len == (5 * sizeof(__be32)))
+		return true;
+
+	return false;
+}
+EXPORT_SYMBOL(of_phy_is_fixed_link);
+
+int of_phy_register_fixed_link(struct device_node *np)
+{
+	struct fixed_phy_status status = {};
+	struct device_node *fixed_link_node;
+	const __be32 *fixed_link_prop;
+	int len;
+
+	/* New binding */
+	fixed_link_node = of_get_child_by_name(np, "fixed-link");
+	if (fixed_link_node) {
+		status.link = 1;
+		status.duplex = of_property_read_bool(fixed_link_node,
+						      "full-duplex");
+		if (of_property_read_u32(fixed_link_node, "speed", &status.speed))
+			return -EINVAL;
+		status.pause = of_property_read_bool(fixed_link_node, "pause");
+		status.asym_pause = of_property_read_bool(fixed_link_node,
+							  "asym-pause");
+		of_node_put(fixed_link_node);
+		return fixed_phy_register(PHY_POLL, &status, np);
+	}
+
+	/* Old binding */
+	fixed_link_prop = of_get_property(np, "fixed-link", &len);
+	if (fixed_link_prop && len == (5 * sizeof(__be32))) {
+		status.link = 1;
+		status.duplex = be32_to_cpu(fixed_link_prop[1]);
+		status.speed = be32_to_cpu(fixed_link_prop[2]);
+		status.pause = be32_to_cpu(fixed_link_prop[3]);
+		status.asym_pause = be32_to_cpu(fixed_link_prop[4]);
+		return fixed_phy_register(PHY_POLL, &status, np);
+	}
+
+	return -ENODEV;
+}
+EXPORT_SYMBOL(of_phy_register_fixed_link);
+#endif
diff --git a/drivers/of/platform.c b/drivers/of/platform.c
index 6c48d73a7fd7..500436f9be7f 100644
--- a/drivers/of/platform.c
+++ b/drivers/of/platform.c
@@ -166,10 +166,6 @@ static void of_dma_configure(struct platform_device *pdev)
 	int ret;
 	struct device *dev = &pdev->dev;
 
-#if defined(CONFIG_MICROBLAZE)
-	pdev->archdata.dma_mask = 0xffffffffUL;
-#endif
-
 	/*
 	 * Set default dma-mask to 32 bit. Drivers are expected to setup
 	 * the correct supported dma_mask.
diff --git a/drivers/pci/access.c b/drivers/pci/access.c
index 8c148f39e8d7..d292d7cb3417 100644
--- a/drivers/pci/access.c
+++ b/drivers/pci/access.c
@@ -231,10 +231,7 @@ static int pci_vpd_pci22_wait(struct pci_dev *dev)
 		}
 
 		if (time_after(jiffies, timeout)) {
-			dev_printk(KERN_DEBUG, &dev->dev,
-				   "vpd r/w failed.  This is likely a firmware "
-				   "bug on this device.  Contact the card "
-				   "vendor for a firmware update.");
+			dev_printk(KERN_DEBUG, &dev->dev, "vpd r/w failed.  This is likely a firmware bug on this device.  Contact the card vendor for a firmware update\n");
 			return -ETIMEDOUT;
 		}
 		if (fatal_signal_pending(current))
diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 447d393725e1..73aef51a28f0 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -226,6 +226,7 @@ int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
 					 type_mask, alignf, alignf_data,
 					 &pci_32_bit);
 }
+EXPORT_SYMBOL(pci_bus_alloc_resource);
 
 void __weak pcibios_resource_survey_bus(struct pci_bus *bus) { }
 
@@ -253,6 +254,7 @@ void pci_bus_add_device(struct pci_dev *dev)
 
 	dev->is_added = 1;
 }
+EXPORT_SYMBOL_GPL(pci_bus_add_device);
 
 /**
  * pci_bus_add_devices - start driver for PCI devices
@@ -279,6 +281,7 @@ void pci_bus_add_devices(const struct pci_bus *bus)
 			pci_bus_add_devices(child);
 	}
 }
+EXPORT_SYMBOL(pci_bus_add_devices);
 
 /** pci_walk_bus - walk devices on/under bus, calling callback.
  *  @top      bus whose devices should be walked
@@ -344,6 +347,3 @@ void pci_bus_put(struct pci_bus *bus)
 }
 EXPORT_SYMBOL(pci_bus_put);
 
-EXPORT_SYMBOL(pci_bus_alloc_resource);
-EXPORT_SYMBOL_GPL(pci_bus_add_device);
-EXPORT_SYMBOL(pci_bus_add_devices);
diff --git a/drivers/pci/host/pci-exynos.c b/drivers/pci/host/pci-exynos.c
index 1632661c5b7f..c5d0ca384502 100644
--- a/drivers/pci/host/pci-exynos.c
+++ b/drivers/pci/host/pci-exynos.c
@@ -545,7 +545,6 @@ static int __init add_pcie_port(struct pcie_port *pp,
 	pp->root_bus_nr = -1;
 	pp->ops = &exynos_pcie_host_ops;
 
-	spin_lock_init(&pp->conf_lock);
 	ret = dw_pcie_host_init(pp);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to initialize host\n");
diff --git a/drivers/pci/host/pci-imx6.c b/drivers/pci/host/pci-imx6.c
index a5645ae4aef0..a568efaa331c 100644
--- a/drivers/pci/host/pci-imx6.c
+++ b/drivers/pci/host/pci-imx6.c
@@ -507,7 +507,6 @@ static int __init imx6_add_pcie_port(struct pcie_port *pp,
 	pp->root_bus_nr = -1;
 	pp->ops = &imx6_pcie_host_ops;
 
-	spin_lock_init(&pp->conf_lock);
 	ret = dw_pcie_host_init(pp);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to initialize host\n");
diff --git a/drivers/pci/host/pci-mvebu.c b/drivers/pci/host/pci-mvebu.c
index e384e2534594..ce23e0f076b6 100644
--- a/drivers/pci/host/pci-mvebu.c
+++ b/drivers/pci/host/pci-mvebu.c
@@ -113,7 +113,6 @@ struct mvebu_pcie {
 struct mvebu_pcie_port {
 	char *name;
 	void __iomem *base;
-	spinlock_t conf_lock;
 	u32 port;
 	u32 lane;
 	int devfn;
@@ -329,9 +328,11 @@ static void mvebu_pcie_add_windows(struct mvebu_pcie_port *port,
 		ret = mvebu_mbus_add_window_remap_by_id(target, attribute, base,
 							sz, remap);
 		if (ret) {
+			phys_addr_t end = base + sz - 1;
+
 			dev_err(&port->pcie->pdev->dev,
-				"Could not create MBus window at 0x%x, size 0x%x: %d\n",
-				base, sz, ret);
+				"Could not create MBus window at [mem %pa-%pa]: %d\n",
+				&base, &end, ret);
 			mvebu_pcie_del_windows(port, base - size_mapped,
 					       size_mapped);
 			return;
@@ -613,9 +614,9 @@ static inline struct mvebu_pcie *sys_to_pcie(struct pci_sys_data *sys)
 	return sys->private_data;
 }
 
-static struct mvebu_pcie_port *
-mvebu_pcie_find_port(struct mvebu_pcie *pcie, struct pci_bus *bus,
-		     int devfn)
+static struct mvebu_pcie_port *mvebu_pcie_find_port(struct mvebu_pcie *pcie,
+						    struct pci_bus *bus,
+						    int devfn)
 {
 	int i;
 
@@ -638,7 +639,6 @@ static int mvebu_pcie_wr_conf(struct pci_bus *bus, u32 devfn,
 {
 	struct mvebu_pcie *pcie = sys_to_pcie(bus->sysdata);
 	struct mvebu_pcie_port *port;
-	unsigned long flags;
 	int ret;
 
 	port = mvebu_pcie_find_port(pcie, bus, devfn);
@@ -664,10 +664,8 @@ static int mvebu_pcie_wr_conf(struct pci_bus *bus, u32 devfn,
 		return PCIBIOS_DEVICE_NOT_FOUND;
 
 	/* Access the real PCIe interface */
-	spin_lock_irqsave(&port->conf_lock, flags);
 	ret = mvebu_pcie_hw_wr_conf(port, bus, devfn,
 				    where, size, val);
-	spin_unlock_irqrestore(&port->conf_lock, flags);
 
 	return ret;
 }
@@ -678,7 +676,6 @@ static int mvebu_pcie_rd_conf(struct pci_bus *bus, u32 devfn, int where,
 {
 	struct mvebu_pcie *pcie = sys_to_pcie(bus->sysdata);
 	struct mvebu_pcie_port *port;
-	unsigned long flags;
 	int ret;
 
 	port = mvebu_pcie_find_port(pcie, bus, devfn);
@@ -710,10 +707,8 @@ static int mvebu_pcie_rd_conf(struct pci_bus *bus, u32 devfn, int where,
 	}
 
 	/* Access the real PCIe interface */
-	spin_lock_irqsave(&port->conf_lock, flags);
 	ret = mvebu_pcie_hw_rd_conf(port, bus, devfn,
 				    where, size, val);
-	spin_unlock_irqrestore(&port->conf_lock, flags);
 
 	return ret;
 }
@@ -786,10 +781,10 @@ static void mvebu_pcie_add_bus(struct pci_bus *bus)
 }
 
 static resource_size_t mvebu_pcie_align_resource(struct pci_dev *dev,
-						const struct resource *res,
-						resource_size_t start,
-						resource_size_t size,
-						resource_size_t align)
+						 const struct resource *res,
+						 resource_size_t start,
+						 resource_size_t size,
+						 resource_size_t align)
 {
 	if (dev->bus->number != 0)
 		return start;
@@ -839,7 +834,8 @@ static void mvebu_pcie_enable(struct mvebu_pcie *pcie)
  * found, maps it.
  */
 static void __iomem *mvebu_pcie_map_registers(struct platform_device *pdev,
-		      struct device_node *np, struct mvebu_pcie_port *port)
+					      struct device_node *np,
+					      struct mvebu_pcie_port *port)
 {
 	struct resource regs;
 	int ret = 0;
@@ -1060,7 +1056,6 @@ static int mvebu_pcie_probe(struct platform_device *pdev)
 		mvebu_pcie_set_local_dev_nr(port, 1);
 
 		port->dn = child;
-		spin_lock_init(&port->conf_lock);
 		mvebu_sw_pci_bridge_init(port);
 		i++;
 	}
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index e3bf9e6d5d9a..1eaf4df3618a 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -643,7 +643,6 @@ static int dw_pcie_rd_conf(struct pci_bus *bus, u32 devfn, int where,
 			int size, u32 *val)
 {
 	struct pcie_port *pp = sys_to_pcie(bus->sysdata);
-	unsigned long flags;
 	int ret;
 
 	if (!pp) {
@@ -656,13 +655,11 @@ static int dw_pcie_rd_conf(struct pci_bus *bus, u32 devfn, int where,
 		return PCIBIOS_DEVICE_NOT_FOUND;
 	}
 
-	spin_lock_irqsave(&pp->conf_lock, flags);
 	if (bus->number != pp->root_bus_nr)
 		ret = dw_pcie_rd_other_conf(pp, bus, devfn,
 						where, size, val);
 	else
 		ret = dw_pcie_rd_own_conf(pp, where, size, val);
-	spin_unlock_irqrestore(&pp->conf_lock, flags);
 
 	return ret;
 }
@@ -671,7 +668,6 @@ static int dw_pcie_wr_conf(struct pci_bus *bus, u32 devfn,
 			int where, int size, u32 val)
 {
 	struct pcie_port *pp = sys_to_pcie(bus->sysdata);
-	unsigned long flags;
 	int ret;
 
 	if (!pp) {
@@ -682,13 +678,11 @@ static int dw_pcie_wr_conf(struct pci_bus *bus, u32 devfn,
 	if (dw_pcie_valid_config(pp, bus, PCI_SLOT(devfn)) == 0)
 		return PCIBIOS_DEVICE_NOT_FOUND;
 
-	spin_lock_irqsave(&pp->conf_lock, flags);
 	if (bus->number != pp->root_bus_nr)
 		ret = dw_pcie_wr_other_conf(pp, bus, devfn,
 						where, size, val);
 	else
 		ret = dw_pcie_wr_own_conf(pp, where, size, val);
-	spin_unlock_irqrestore(&pp->conf_lock, flags);
 
 	return ret;
 }
diff --git a/drivers/pci/host/pcie-designware.h b/drivers/pci/host/pcie-designware.h
index a169d22d517e..77f592faa7bf 100644
--- a/drivers/pci/host/pcie-designware.h
+++ b/drivers/pci/host/pcie-designware.h
@@ -41,7 +41,6 @@ struct pcie_port {
 	void __iomem		*va_cfg1_base;
 	u64			io_base;
 	u64			mem_base;
-	spinlock_t		conf_lock;
 	struct resource		cfg;
 	struct resource		io;
 	struct resource		mem;
diff --git a/drivers/pci/host/pcie-rcar.c b/drivers/pci/host/pcie-rcar.c
index 8e06124aa80f..f7d3de32c9a0 100644
--- a/drivers/pci/host/pcie-rcar.c
+++ b/drivers/pci/host/pcie-rcar.c
@@ -277,9 +277,8 @@ static int rcar_pcie_read_conf(struct pci_bus *bus, unsigned int devfn,
 	else if (size == 2)
 		*val = (*val >> (8 * (where & 2))) & 0xffff;
 
-	dev_dbg(&bus->dev, "pcie-config-read: bus=%3d devfn=0x%04x "
-		"where=0x%04x size=%d val=0x%08lx\n", bus->number,
-		devfn, where, size, (unsigned long)*val);
+	dev_dbg(&bus->dev, "pcie-config-read: bus=%3d devfn=0x%04x where=0x%04x size=%d val=0x%08lx\n",
+		bus->number, devfn, where, size, (unsigned long)*val);
 
 	return ret;
 }
@@ -302,9 +301,8 @@ static int rcar_pcie_write_conf(struct pci_bus *bus, unsigned int devfn,
 	if (ret != PCIBIOS_SUCCESSFUL)
 		return ret;
 
-	dev_dbg(&bus->dev, "pcie-config-write: bus=%3d devfn=0x%04x "
-		"where=0x%04x size=%d val=0x%08lx\n", bus->number,
-		devfn, where, size, (unsigned long)val);
+	dev_dbg(&bus->dev, "pcie-config-write: bus=%3d devfn=0x%04x where=0x%04x size=%d val=0x%08lx\n",
+		bus->number, devfn, where, size, (unsigned long)val);
 
 	if (size == 1) {
 		shift = 8 * (where & 3);
diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h
index 2b859249303b..b0e61bf261a7 100644
--- a/drivers/pci/hotplug/acpiphp.h
+++ b/drivers/pci/hotplug/acpiphp.h
@@ -142,6 +142,16 @@ static inline acpi_handle func_to_handle(struct acpiphp_func *func)
 	return func_to_acpi_device(func)->handle;
 }
 
+struct acpiphp_root_context {
+	struct acpi_hotplug_context hp;
+	struct acpiphp_bridge *root_bridge;
+};
+
+static inline struct acpiphp_root_context *to_acpiphp_root_context(struct acpi_hotplug_context *hp)
+{
+	return container_of(hp, struct acpiphp_root_context, hp);
+}
+
 /*
  * struct acpiphp_attention_info - device specific attention registration
  *
diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c
index 728c31f4c2c5..e291efcd02a2 100644
--- a/drivers/pci/hotplug/acpiphp_core.c
+++ b/drivers/pci/hotplug/acpiphp_core.c
@@ -63,10 +63,6 @@ MODULE_LICENSE("GPL");
 MODULE_PARM_DESC(disable, "disable acpiphp driver");
 module_param_named(disable, acpiphp_disabled, bool, 0444);
 
-/* export the attention callback registration methods */
-EXPORT_SYMBOL_GPL(acpiphp_register_attention);
-EXPORT_SYMBOL_GPL(acpiphp_unregister_attention);
-
 static int enable_slot		(struct hotplug_slot *slot);
 static int disable_slot		(struct hotplug_slot *slot);
 static int set_attention_status (struct hotplug_slot *slot, u8 value);
@@ -104,6 +100,7 @@ int acpiphp_register_attention(struct acpiphp_attention_info *info)
 	}
 	return retval;
 }
+EXPORT_SYMBOL_GPL(acpiphp_register_attention);
 
 
 /**
@@ -124,6 +121,7 @@ int acpiphp_unregister_attention(struct acpiphp_attention_info *info)
 	}
 	return retval;
 }
+EXPORT_SYMBOL_GPL(acpiphp_unregister_attention);
 
 
 /**
diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 75e178330215..602d153c7055 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -351,11 +351,9 @@ static acpi_status acpiphp_add_context(acpi_handle handle, u32 lvl, void *data,
 			slot->slot = NULL;
 			bridge->nr_slots--;
 			if (retval == -EBUSY)
-				pr_warn("Slot %llu already registered by another "
-					"hotplug driver\n", sun);
+				pr_warn("Slot %llu already registered by another hotplug driver\n", sun);
 			else
-				pr_warn("acpiphp_register_hotplug_slot failed "
-					"(err code = 0x%x)\n", retval);
+				pr_warn("acpiphp_register_hotplug_slot failed (err code = 0x%x)\n", retval);
 		}
 		/* Even if the slot registration fails, we can still use it. */
 	}
@@ -373,17 +371,13 @@ static acpi_status acpiphp_add_context(acpi_handle handle, u32 lvl, void *data,
 
 static struct acpiphp_bridge *acpiphp_dev_to_bridge(struct acpi_device *adev)
 {
-	struct acpiphp_context *context;
 	struct acpiphp_bridge *bridge = NULL;
 
 	acpi_lock_hp_context();
-	context = acpiphp_get_context(adev);
-	if (context) {
-		bridge = context->bridge;
+	if (adev->hp) {
+		bridge = to_acpiphp_root_context(adev->hp)->root_bridge;
 		if (bridge)
 			get_bridge(bridge);
-
-		acpiphp_put_context(context);
 	}
 	acpi_unlock_hp_context();
 	return bridge;
@@ -881,7 +875,17 @@ void acpiphp_enumerate_slots(struct pci_bus *bus)
 	 */
 	get_device(&bus->dev);
 
-	if (!pci_is_root_bus(bridge->pci_bus)) {
+	acpi_lock_hp_context();
+	if (pci_is_root_bus(bridge->pci_bus)) {
+		struct acpiphp_root_context *root_context;
+
+		root_context = kzalloc(sizeof(*root_context), GFP_KERNEL);
+		if (!root_context)
+			goto err;
+
+		root_context->root_bridge = bridge;
+		acpi_set_hp_context(adev, &root_context->hp, NULL, NULL, NULL);
+	} else {
 		struct acpiphp_context *context;
 
 		/*
@@ -890,21 +894,16 @@ void acpiphp_enumerate_slots(struct pci_bus *bus)
 		 * parent is going to be handled by pciehp, in which case this
 		 * bridge is not interesting to us either.
 		 */
-		acpi_lock_hp_context();
 		context = acpiphp_get_context(adev);
-		if (!context) {
-			acpi_unlock_hp_context();
-			put_device(&bus->dev);
-			pci_dev_put(bridge->pci_dev);
-			kfree(bridge);
-			return;
-		}
+		if (!context)
+			goto err;
+
 		bridge->context = context;
 		context->bridge = bridge;
 		/* Get a reference to the parent bridge. */
 		get_bridge(context->func.parent);
-		acpi_unlock_hp_context();
 	}
+	acpi_unlock_hp_context();
 
 	/* Must be added to the list prior to calling acpiphp_add_context(). */
 	mutex_lock(&bridge_mutex);
@@ -919,6 +918,30 @@ void acpiphp_enumerate_slots(struct pci_bus *bus)
 		cleanup_bridge(bridge);
 		put_bridge(bridge);
 	}
+	return;
+
+ err:
+	acpi_unlock_hp_context();
+	put_device(&bus->dev);
+	pci_dev_put(bridge->pci_dev);
+	kfree(bridge);
+}
+
+void acpiphp_drop_bridge(struct acpiphp_bridge *bridge)
+{
+	if (pci_is_root_bus(bridge->pci_bus)) {
+		struct acpiphp_root_context *root_context;
+		struct acpi_device *adev;
+
+		acpi_lock_hp_context();
+		adev = ACPI_COMPANION(bridge->pci_bus->bridge);
+		root_context = to_acpiphp_root_context(adev->hp);
+		adev->hp = NULL;
+		acpi_unlock_hp_context();
+		kfree(root_context);
+	}
+	cleanup_bridge(bridge);
+	put_bridge(bridge);
 }
 
 /**
@@ -936,8 +959,7 @@ void acpiphp_remove_slots(struct pci_bus *bus)
 	list_for_each_entry(bridge, &bridge_list, list)
 		if (bridge->pci_bus == bus) {
 			mutex_unlock(&bridge_mutex);
-			cleanup_bridge(bridge);
-			put_bridge(bridge);
+			acpiphp_drop_bridge(bridge);
 			return;
 		}
 
diff --git a/drivers/pci/hotplug/cpci_hotplug.h b/drivers/pci/hotplug/cpci_hotplug.h
index 1356211431d0..6a0ddf757349 100644
--- a/drivers/pci/hotplug/cpci_hotplug.h
+++ b/drivers/pci/hotplug/cpci_hotplug.h
@@ -56,9 +56,9 @@ struct cpci_hp_controller_ops {
 	int (*enable_irq) (void);
 	int (*disable_irq) (void);
 	int (*check_irq) (void *dev_id);
-	int (*hardware_test) (struct slot* slot, u32 value);
-	u8  (*get_power) (struct slot* slot);
-	int (*set_power) (struct slot* slot, int value);
+	int (*hardware_test) (struct slot *slot, u32 value);
+	u8  (*get_power) (struct slot *slot);
+	int (*set_power) (struct slot *slot, int value);
 };
 
 struct cpci_hp_controller {
@@ -89,13 +89,13 @@ int cpci_hp_stop(void);
 u8 cpci_get_attention_status(struct slot *slot);
 u8 cpci_get_latch_status(struct slot *slot);
 u8 cpci_get_adapter_status(struct slot *slot);
-u16 cpci_get_hs_csr(struct slot * slot);
+u16 cpci_get_hs_csr(struct slot *slot);
 int cpci_set_attention_status(struct slot *slot, int status);
-int cpci_check_and_clear_ins(struct slot * slot);
-int cpci_check_ext(struct slot * slot);
-int cpci_clear_ext(struct slot * slot);
-int cpci_led_on(struct slot * slot);
-int cpci_led_off(struct slot * slot);
+int cpci_check_and_clear_ins(struct slot *slot);
+int cpci_check_ext(struct slot *slot);
+int cpci_clear_ext(struct slot *slot);
+int cpci_led_on(struct slot *slot);
+int cpci_led_off(struct slot *slot);
 int cpci_configure_slot(struct slot *slot);
 int cpci_unconfigure_slot(struct slot *slot);
 
diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c
index 00c81a3cefc9..e09cf7827d68 100644
--- a/drivers/pci/hotplug/cpci_hotplug_core.c
+++ b/drivers/pci/hotplug/cpci_hotplug_core.c
@@ -65,10 +65,10 @@ static int thread_finished;
 static int enable_slot(struct hotplug_slot *slot);
 static int disable_slot(struct hotplug_slot *slot);
 static int set_attention_status(struct hotplug_slot *slot, u8 value);
-static int get_power_status(struct hotplug_slot *slot, u8 * value);
-static int get_attention_status(struct hotplug_slot *slot, u8 * value);
-static int get_adapter_status(struct hotplug_slot *slot, u8 * value);
-static int get_latch_status(struct hotplug_slot *slot, u8 * value);
+static int get_power_status(struct hotplug_slot *slot, u8 *value);
+static int get_attention_status(struct hotplug_slot *slot, u8 *value);
+static int get_adapter_status(struct hotplug_slot *slot, u8 *value);
+static int get_latch_status(struct hotplug_slot *slot, u8 *value);
 
 static struct hotplug_slot_ops cpci_hotplug_slot_ops = {
 	.enable_slot = enable_slot,
@@ -168,7 +168,7 @@ cpci_get_power_status(struct slot *slot)
 }
 
 static int
-get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
+get_power_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	struct slot *slot = hotplug_slot->private;
 
@@ -177,7 +177,7 @@ get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
 }
 
 static int
-get_attention_status(struct hotplug_slot *hotplug_slot, u8 * value)
+get_attention_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	struct slot *slot = hotplug_slot->private;
 
@@ -192,14 +192,14 @@ set_attention_status(struct hotplug_slot *hotplug_slot, u8 status)
 }
 
 static int
-get_adapter_status(struct hotplug_slot *hotplug_slot, u8 * value)
+get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	*value = hotplug_slot->info->adapter_status;
 	return 0;
 }
 
 static int
-get_latch_status(struct hotplug_slot *hotplug_slot, u8 * value)
+get_latch_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	*value = hotplug_slot->info->latch_status;
 	return 0;
@@ -299,6 +299,7 @@ error_slot:
 error:
 	return status;
 }
+EXPORT_SYMBOL_GPL(cpci_hp_register_bus);
 
 int
 cpci_hp_unregister_bus(struct pci_bus *bus)
@@ -329,6 +330,7 @@ cpci_hp_unregister_bus(struct pci_bus *bus)
 	up_write(&list_rwsem);
 	return status;
 }
+EXPORT_SYMBOL_GPL(cpci_hp_unregister_bus);
 
 /* This is the interrupt mode interrupt handler */
 static irqreturn_t
@@ -360,7 +362,7 @@ static int
 init_slots(int clear_ins)
 {
 	struct slot *slot;
-	struct pci_dev* dev;
+	struct pci_dev *dev;
 
 	dbg("%s - enter", __func__);
 	down_read(&list_rwsem);
@@ -614,6 +616,7 @@ cpci_hp_register_controller(struct cpci_hp_controller *new_controller)
 		controller = new_controller;
 	return status;
 }
+EXPORT_SYMBOL_GPL(cpci_hp_register_controller);
 
 static void
 cleanup_slots(void)
@@ -653,6 +656,7 @@ cpci_hp_unregister_controller(struct cpci_hp_controller *old_controller)
 		status = -ENODEV;
 	return status;
 }
+EXPORT_SYMBOL_GPL(cpci_hp_unregister_controller);
 
 int
 cpci_hp_start(void)
@@ -690,6 +694,7 @@ cpci_hp_start(void)
 	dbg("%s - exit", __func__);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(cpci_hp_start);
 
 int
 cpci_hp_stop(void)
@@ -704,6 +709,7 @@ cpci_hp_stop(void)
 	cpci_stop_thread();
 	return 0;
 }
+EXPORT_SYMBOL_GPL(cpci_hp_stop);
 
 int __init
 cpci_hotplug_init(int debug)
@@ -721,10 +727,3 @@ cpci_hotplug_exit(void)
 	cpci_hp_stop();
 	cpci_hp_unregister_controller(controller);
 }
-
-EXPORT_SYMBOL_GPL(cpci_hp_register_controller);
-EXPORT_SYMBOL_GPL(cpci_hp_unregister_controller);
-EXPORT_SYMBOL_GPL(cpci_hp_register_bus);
-EXPORT_SYMBOL_GPL(cpci_hp_unregister_bus);
-EXPORT_SYMBOL_GPL(cpci_hp_start);
-EXPORT_SYMBOL_GPL(cpci_hp_stop);
diff --git a/drivers/pci/hotplug/cpci_hotplug_pci.c b/drivers/pci/hotplug/cpci_hotplug_pci.c
index f6ef64c2ccb5..7d48ecae6695 100644
--- a/drivers/pci/hotplug/cpci_hotplug_pci.c
+++ b/drivers/pci/hotplug/cpci_hotplug_pci.c
@@ -46,7 +46,7 @@ extern int cpci_debug;
 #define warn(format, arg...) printk(KERN_WARNING "%s: " format "\n", MY_NAME , ## arg)
 
 
-u8 cpci_get_attention_status(struct slot* slot)
+u8 cpci_get_attention_status(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -66,7 +66,7 @@ u8 cpci_get_attention_status(struct slot* slot)
 	return hs_csr & 0x0008 ? 1 : 0;
 }
 
-int cpci_set_attention_status(struct slot* slot, int status)
+int cpci_set_attention_status(struct slot *slot, int status)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -93,7 +93,7 @@ int cpci_set_attention_status(struct slot* slot, int status)
 	return 1;
 }
 
-u16 cpci_get_hs_csr(struct slot* slot)
+u16 cpci_get_hs_csr(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -111,7 +111,7 @@ u16 cpci_get_hs_csr(struct slot* slot)
 	return hs_csr;
 }
 
-int cpci_check_and_clear_ins(struct slot* slot)
+int cpci_check_and_clear_ins(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -140,7 +140,7 @@ int cpci_check_and_clear_ins(struct slot* slot)
 	return ins;
 }
 
-int cpci_check_ext(struct slot* slot)
+int cpci_check_ext(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -161,7 +161,7 @@ int cpci_check_ext(struct slot* slot)
 	return ext;
 }
 
-int cpci_clear_ext(struct slot* slot)
+int cpci_clear_ext(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -187,7 +187,7 @@ int cpci_clear_ext(struct slot* slot)
 	return 0;
 }
 
-int cpci_led_on(struct slot* slot)
+int cpci_led_on(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -216,7 +216,7 @@ int cpci_led_on(struct slot* slot)
 	return 0;
 }
 
-int cpci_led_off(struct slot* slot)
+int cpci_led_off(struct slot *slot)
 {
 	int hs_cap;
 	u16 hs_csr;
@@ -303,7 +303,7 @@ int cpci_configure_slot(struct slot *slot)
 	return ret;
 }
 
-int cpci_unconfigure_slot(struct slot* slot)
+int cpci_unconfigure_slot(struct slot *slot)
 {
 	struct pci_dev *dev, *temp;
 
diff --git a/drivers/pci/hotplug/cpcihp_generic.c b/drivers/pci/hotplug/cpcihp_generic.c
index 7536eef620b0..04fcd7811400 100644
--- a/drivers/pci/hotplug/cpcihp_generic.c
+++ b/drivers/pci/hotplug/cpcihp_generic.c
@@ -78,8 +78,8 @@ static struct cpci_hp_controller generic_hpc;
 
 static int __init validate_parameters(void)
 {
-	char* str;
-	char* p;
+	char *str;
+	char *p;
 	unsigned long tmp;
 
 	if(!bridge) {
@@ -142,8 +142,8 @@ static int query_enum(void)
 static int __init cpcihp_generic_init(void)
 {
 	int status;
-	struct resource* r;
-	struct pci_dev* dev;
+	struct resource *r;
+	struct pci_dev *dev;
 
 	info(DRIVER_DESC " version: " DRIVER_VERSION);
 	status = validate_parameters();
diff --git a/drivers/pci/hotplug/cpcihp_zt5550.c b/drivers/pci/hotplug/cpcihp_zt5550.c
index e8c4a7ccf578..6757b3ef7e10 100644
--- a/drivers/pci/hotplug/cpcihp_zt5550.c
+++ b/drivers/pci/hotplug/cpcihp_zt5550.c
@@ -295,7 +295,7 @@ static struct pci_driver zt5550_hc_driver = {
 
 static int __init zt5550_init(void)
 {
-	struct resource* r;
+	struct resource *r;
 	int rc;
 
 	info(DRIVER_DESC " version: " DRIVER_VERSION);
diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h
index 516b87738b6e..0450f405807d 100644
--- a/drivers/pci/hotplug/cpqphp.h
+++ b/drivers/pci/hotplug/cpqphp.h
@@ -255,7 +255,7 @@ struct pci_func {
 	struct pci_resource *io_head;
 	struct pci_resource *bus_head;
 	struct timer_list *p_task_event;
-	struct pci_dev* pci_dev;
+	struct pci_dev *pci_dev;
 };
 
 struct slot {
@@ -278,7 +278,7 @@ struct slot {
 };
 
 struct pci_resource {
-	struct pci_resource * next;
+	struct pci_resource *next;
 	u32 base;
 	u32 length;
 };
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 037e2612c5bd..4aaee746df88 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -94,7 +94,7 @@ static inline int is_slot66mhz(struct slot *slot)
  *
  * Returns pointer to the head of the SMBIOS tables (or %NULL).
  */
-static void __iomem * detect_SMBIOS_pointer(void __iomem *begin, void __iomem *end)
+static void __iomem *detect_SMBIOS_pointer(void __iomem *begin, void __iomem *end)
 {
 	void __iomem *fp;
 	void __iomem *endp;
@@ -131,7 +131,7 @@ static void __iomem * detect_SMBIOS_pointer(void __iomem *begin, void __iomem *e
  *
  * For unexpected switch opens
  */
-static int init_SERR(struct controller * ctrl)
+static int init_SERR(struct controller *ctrl)
 {
 	u32 tempdword;
 	u32 number_of_slots;
@@ -291,7 +291,7 @@ static void release_slot(struct hotplug_slot *hotplug_slot)
 	kfree(slot);
 }
 
-static int ctrl_slot_cleanup (struct controller * ctrl)
+static int ctrl_slot_cleanup (struct controller *ctrl)
 {
 	struct slot *old_slot, *next_slot;
 
@@ -706,8 +706,7 @@ static int ctrl_slot_setup(struct controller *ctrl,
 		hotplug_slot_info->adapter_status =
 			get_presence_status(ctrl, slot);
 
-		dbg("registering bus %d, dev %d, number %d, "
-				"ctrl->slot_device_offset %d, slot %d\n",
+		dbg("registering bus %d, dev %d, number %d, ctrl->slot_device_offset %d, slot %d\n",
 				slot->bus, slot->device,
 				slot->number, ctrl->slot_device_offset,
 				slot_number);
@@ -837,8 +836,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	bus = pdev->subordinate;
 	if (!bus) {
-		dev_notice(&pdev->dev, "the device is not a bridge, "
-				"skipping\n");
+		dev_notice(&pdev->dev, "the device is not a bridge, skipping\n");
 		rc = -ENODEV;
 		goto err_disable_device;
 	}
diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c
index f593585f2784..bde47fce3248 100644
--- a/drivers/pci/hotplug/cpqphp_ctrl.c
+++ b/drivers/pci/hotplug/cpqphp_ctrl.c
@@ -39,9 +39,9 @@
 #include <linux/kthread.h>
 #include "cpqphp.h"
 
-static u32 configure_new_device(struct controller* ctrl, struct pci_func *func,
+static u32 configure_new_device(struct controller *ctrl, struct pci_func *func,
 			u8 behind_bridge, struct resource_lists *resources);
-static int configure_new_function(struct controller* ctrl, struct pci_func *func,
+static int configure_new_function(struct controller *ctrl, struct pci_func *func,
 			u8 behind_bridge, struct resource_lists *resources);
 static void interrupt_event_handler(struct controller *ctrl);
 
@@ -64,7 +64,7 @@ static void long_delay(int delay)
 
 /* FIXME: The following line needs to be somewhere else... */
 #define WRONG_BUS_FREQUENCY 0x07
-static u8 handle_switch_change(u8 change, struct controller * ctrl)
+static u8 handle_switch_change(u8 change, struct controller *ctrl)
 {
 	int hp_slot;
 	u8 rc = 0;
@@ -138,7 +138,7 @@ static struct slot *cpqhp_find_slot(struct controller *ctrl, u8 device)
 }
 
 
-static u8 handle_presence_change(u16 change, struct controller * ctrl)
+static u8 handle_presence_change(u16 change, struct controller *ctrl)
 {
 	int hp_slot;
 	u8 rc = 0;
@@ -232,7 +232,7 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl)
 }
 
 
-static u8 handle_power_fault(u8 change, struct controller * ctrl)
+static u8 handle_power_fault(u8 change, struct controller *ctrl)
 {
 	int hp_slot;
 	u8 rc = 0;
@@ -997,7 +997,7 @@ struct pci_func *cpqhp_slot_create(u8 busnumber)
  *
  * Returns %0 if successful, !0 otherwise.
  */
-static int slot_remove(struct pci_func * old_slot)
+static int slot_remove(struct pci_func *old_slot)
 {
 	struct pci_func *next;
 
@@ -1109,7 +1109,7 @@ struct pci_func *cpqhp_slot_find(u8 bus, u8 device, u8 index)
 
 /* DJZ: I don't think is_bridge will work as is.
  * FIXME */
-static int is_bridge(struct pci_func * func)
+static int is_bridge(struct pci_func *func)
 {
 	/* Check the header type */
 	if (((func->config_space[0x03] >> 16) & 0xFF) == 0x01)
@@ -1625,7 +1625,7 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl)
  * @replace_flag: whether replacing or adding a new device
  * @ctrl: target controller
  */
-static u32 remove_board(struct pci_func * func, u32 replace_flag, struct controller * ctrl)
+static u32 remove_board(struct pci_func *func, u32 replace_flag, struct controller *ctrl)
 {
 	int index;
 	u8 skip = 0;
@@ -1742,7 +1742,7 @@ static void pushbutton_helper_thread(unsigned long data)
 
 
 /* this is the main worker thread */
-static int event_thread(void* data)
+static int event_thread(void *data)
 {
 	struct controller *ctrl;
 
@@ -1992,7 +1992,7 @@ int cpqhp_process_SI(struct controller *ctrl, struct pci_func *func)
 	u16 temp_word;
 	u32 tempdword;
 	int rc;
-	struct slot* p_slot;
+	struct slot *p_slot;
 	int physical_slot = 0;
 
 	tempdword = 0;
@@ -2088,7 +2088,7 @@ int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func)
 	u8 replace_flag;
 	u32 rc = 0;
 	unsigned int devfn;
-	struct slot* p_slot;
+	struct slot *p_slot;
 	struct pci_bus *pci_bus = ctrl->pci_bus;
 	int physical_slot=0;
 
@@ -2270,8 +2270,8 @@ int cpqhp_hardware_test(struct controller *ctrl, int test_num)
  *
  * Returns 0 if success.
  */
-static u32 configure_new_device(struct controller * ctrl, struct pci_func * func,
-				 u8 behind_bridge, struct resource_lists * resources)
+static u32 configure_new_device(struct controller  *ctrl, struct pci_func  *func,
+				 u8 behind_bridge, struct resource_lists  *resources)
 {
 	u8 temp_byte, function, max_functions, stop_it;
 	int rc;
diff --git a/drivers/pci/hotplug/cpqphp_nvram.c b/drivers/pci/hotplug/cpqphp_nvram.c
index 9600a392eaae..0968a9bcb345 100644
--- a/drivers/pci/hotplug/cpqphp_nvram.c
+++ b/drivers/pci/hotplug/cpqphp_nvram.c
@@ -107,7 +107,7 @@ static spinlock_t int15_lock;
  */
 
 
-static u32 add_byte( u32 **p_buffer, u8 value, u32 *used, u32 *avail)
+static u32 add_byte(u32 **p_buffer, u8 value, u32 *used, u32 *avail)
 {
 	u8 **tByte;
 
@@ -122,7 +122,7 @@ static u32 add_byte( u32 **p_buffer, u8 value, u32 *used, u32 *avail)
 }
 
 
-static u32 add_dword( u32 **p_buffer, u32 value, u32 *used, u32 *avail)
+static u32 add_dword(u32 **p_buffer, u32 value, u32 *used, u32 *avail)
 {
 	if ((*used + 4) > *avail)
 		return(1);
@@ -267,12 +267,12 @@ static u32 store_HRT (void __iomem *rom_start)
 	ctrl = cpqhp_ctrl_list;
 
 	/* The revision of this structure */
-	rc = add_byte( &pFill, 1 + ctrl->push_flag, &usedbytes, &available);
+	rc = add_byte(&pFill, 1 + ctrl->push_flag, &usedbytes, &available);
 	if (rc)
 		return(rc);
 
 	/* The number of controllers */
-	rc = add_byte( &pFill, 1, &usedbytes, &available);
+	rc = add_byte(&pFill, 1, &usedbytes, &available);
 	if (rc)
 		return(rc);
 
@@ -282,22 +282,22 @@ static u32 store_HRT (void __iomem *rom_start)
 		numCtrl++;
 
 		/* The bus number */
-		rc = add_byte( &pFill, ctrl->bus, &usedbytes, &available);
+		rc = add_byte(&pFill, ctrl->bus, &usedbytes, &available);
 		if (rc)
 			return(rc);
 
 		/* The device Number */
-		rc = add_byte( &pFill, PCI_SLOT(ctrl->pci_dev->devfn), &usedbytes, &available);
+		rc = add_byte(&pFill, PCI_SLOT(ctrl->pci_dev->devfn), &usedbytes, &available);
 		if (rc)
 			return(rc);
 
 		/* The function Number */
-		rc = add_byte( &pFill, PCI_FUNC(ctrl->pci_dev->devfn), &usedbytes, &available);
+		rc = add_byte(&pFill, PCI_FUNC(ctrl->pci_dev->devfn), &usedbytes, &available);
 		if (rc)
 			return(rc);
 
 		/* Skip the number of available entries */
-		rc = add_dword( &pFill, 0, &usedbytes, &available);
+		rc = add_dword(&pFill, 0, &usedbytes, &available);
 		if (rc)
 			return(rc);
 
@@ -311,12 +311,12 @@ static u32 store_HRT (void __iomem *rom_start)
 			loop ++;
 
 			/* base */
-			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
 			/* length */
-			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
@@ -336,12 +336,12 @@ static u32 store_HRT (void __iomem *rom_start)
 			loop ++;
 
 			/* base */
-			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
 			/* length */
-			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
@@ -361,12 +361,12 @@ static u32 store_HRT (void __iomem *rom_start)
 			loop ++;
 
 			/* base */
-			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
 			/* length */
-			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
@@ -386,12 +386,12 @@ static u32 store_HRT (void __iomem *rom_start)
 			loop ++;
 
 			/* base */
-			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
 			/* length */
-			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
+			rc = add_dword(&pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index a3e3c2002b58..1c8c2f130d31 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -81,7 +81,7 @@ static void __iomem *detect_HRT_floating_pointer(void __iomem *begin, void __iom
 }
 
 
-int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
+int cpqhp_configure_device (struct controller *ctrl, struct pci_func *func)
 {
 	struct pci_bus *child;
 	int num;
@@ -121,7 +121,7 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
 }
 
 
-int cpqhp_unconfigure_device(struct pci_func* func)
+int cpqhp_unconfigure_device(struct pci_func *func)
 {
 	int j;
 
@@ -129,7 +129,7 @@ int cpqhp_unconfigure_device(struct pci_func* func)
 
 	pci_lock_rescan_remove();
 	for (j=0; j<8 ; j++) {
-		struct pci_dev* temp = pci_get_bus_and_slot(func->bus, PCI_DEVFN(func->device, j));
+		struct pci_dev *temp = pci_get_bus_and_slot(func->bus, PCI_DEVFN(func->device, j));
 		if (temp) {
 			pci_dev_put(temp);
 			pci_stop_and_remove_bus_device(temp);
@@ -203,7 +203,7 @@ int cpqhp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num)
 }
 
 
-static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev_num)
+static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 *dev_num)
 {
 	u16 tdevice;
 	u32 work;
@@ -280,7 +280,7 @@ static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num
 }
 
 
-int cpqhp_get_bus_dev (struct controller *ctrl, u8 * bus_num, u8 * dev_num, u8 slot)
+int cpqhp_get_bus_dev (struct controller *ctrl, u8 *bus_num, u8 *dev_num, u8 slot)
 {
 	/* plain (bridges allowed) */
 	return PCI_GetBusDevHelper(ctrl, bus_num, dev_num, slot, 0);
@@ -465,7 +465,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
  *
  * returns 0 if success
  */
-int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
+int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func *new_slot)
 {
 	long rc;
 	u8 class_code;
@@ -549,7 +549,7 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
  *
  * returns 0 if success
  */
-int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func)
+int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func *func)
 {
 	u8 cloop;
 	u8 header_type;
@@ -686,7 +686,7 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func)
  *
  * returns 0 if success
  */
-int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
+int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func *func)
 {
 	u8 cloop;
 	u8 header_type;
@@ -949,7 +949,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
  *
  * returns 0 if success
  */
-int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func)
+int cpqhp_configure_board(struct controller *ctrl, struct pci_func *func)
 {
 	int cloop;
 	u8 header_type;
@@ -1027,7 +1027,7 @@ int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func)
  *
  * returns 0 if the board is the same nonzero otherwise
  */
-int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func)
+int cpqhp_valid_replace(struct controller *ctrl, struct pci_func *func)
 {
 	u8 cloop;
 	u8 header_type;
@@ -1419,7 +1419,7 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
  *
  * returns 0 if success
  */
-int cpqhp_return_board_resources(struct pci_func * func, struct resource_lists * resources)
+int cpqhp_return_board_resources(struct pci_func *func, struct resource_lists *resources)
 {
 	int rc = 0;
 	struct pci_resource *node;
@@ -1475,7 +1475,7 @@ int cpqhp_return_board_resources(struct pci_func * func, struct resource_lists *
  *
  * Puts node back in the resource list pointed to by head
  */
-void cpqhp_destroy_resource_list (struct resource_lists * resources)
+void cpqhp_destroy_resource_list (struct resource_lists *resources)
 {
 	struct pci_resource *res, *tres;
 
@@ -1522,7 +1522,7 @@ void cpqhp_destroy_resource_list (struct resource_lists * resources)
  *
  * Puts node back in the resource list pointed to by head
  */
-void cpqhp_destroy_board_resources (struct pci_func * func)
+void cpqhp_destroy_board_resources (struct pci_func *func)
 {
 	struct pci_resource *res, *tres;
 
diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c
index 17c1f36315d1..4a392c44e3d3 100644
--- a/drivers/pci/hotplug/cpqphp_sysfs.c
+++ b/drivers/pci/hotplug/cpqphp_sysfs.c
@@ -79,7 +79,7 @@ static int show_ctrl (struct controller *ctrl, char *buf)
 
 static int show_dev (struct controller *ctrl, char *buf)
 {
-	char * out = buf;
+	char *out = buf;
 	int index;
 	struct pci_resource *res;
 	struct pci_func *new_slot;
diff --git a/drivers/pci/hotplug/ibmphp_core.c b/drivers/pci/hotplug/ibmphp_core.c
index cf3ac1e4b099..f7b8684a7739 100644
--- a/drivers/pci/hotplug/ibmphp_core.c
+++ b/drivers/pci/hotplug/ibmphp_core.c
@@ -74,7 +74,7 @@ static inline int get_max_adapter_speed (struct hotplug_slot *hs, u8 *value)
 static inline int get_cur_bus_info(struct slot **sl)
 {
 	int rc = 1;
-	struct slot * slot_cur = *sl;
+	struct slot *slot_cur = *sl;
 
 	debug("options = %x\n", slot_cur->ctrl->options);
 	debug("revision = %x\n", slot_cur->ctrl->revision);
@@ -114,8 +114,8 @@ static inline int slot_update(struct slot **sl)
 
 static int __init get_max_slots (void)
 {
-	struct slot * slot_cur;
-	struct list_head * tmp;
+	struct slot *slot_cur;
+	struct list_head *tmp;
 	u8 slot_count = 0;
 
 	list_for_each(tmp, &ibmphp_slot_head) {
@@ -280,7 +280,7 @@ static int set_attention_status(struct hotplug_slot *hotplug_slot, u8 value)
 	return rc;
 }
 
-static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	int rc = -ENODEV;
 	struct slot *pslot;
@@ -311,7 +311,7 @@ static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 * value)
 	return rc;
 }
 
-static int get_latch_status(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_latch_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	int rc = -ENODEV;
 	struct slot *pslot;
@@ -338,7 +338,7 @@ static int get_latch_status(struct hotplug_slot *hotplug_slot, u8 * value)
 }
 
 
-static int get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_power_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	int rc = -ENODEV;
 	struct slot *pslot;
@@ -364,7 +364,7 @@ static int get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
 	return rc;
 }
 
-static int get_adapter_present(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_adapter_present(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	int rc = -ENODEV;
 	struct slot *pslot;
@@ -433,7 +433,7 @@ static int get_max_bus_speed(struct slot *slot)
 }
 
 /*
-static int get_max_adapter_speed_1(struct hotplug_slot *hotplug_slot, u8 * value, u8 flag)
+static int get_max_adapter_speed_1(struct hotplug_slot *hotplug_slot, u8 *value, u8 flag)
 {
 	int rc = -ENODEV;
 	struct slot *pslot;
@@ -471,7 +471,7 @@ static int get_max_adapter_speed_1(struct hotplug_slot *hotplug_slot, u8 * value
 	return rc;
 }
 
-static int get_bus_name(struct hotplug_slot *hotplug_slot, char * value)
+static int get_bus_name(struct hotplug_slot *hotplug_slot, char *value)
 {
 	int rc = -ENODEV;
 	struct slot *pslot = NULL;
@@ -671,7 +671,7 @@ static struct pci_func *ibm_slot_find(u8 busno, u8 device, u8 function)
 {
 	struct pci_func *func_cur;
 	struct slot *slot_cur;
-	struct list_head * tmp;
+	struct list_head *tmp;
 	list_for_each(tmp, &ibmphp_slot_head) {
 		slot_cur = list_entry(tmp, struct slot, ibm_slot_list);
 		if (slot_cur->func) {
@@ -696,8 +696,8 @@ static struct pci_func *ibm_slot_find(u8 busno, u8 device, u8 function)
 static void free_slots(void)
 {
 	struct slot *slot_cur;
-	struct list_head * tmp;
-	struct list_head * next;
+	struct list_head *tmp;
+	struct list_head *next;
 
 	debug("%s -- enter\n", __func__);
 
@@ -825,10 +825,10 @@ static int ibm_configure_device(struct pci_func *func)
 /*******************************************************
  * Returns whether the bus is empty or not
  *******************************************************/
-static int is_bus_empty(struct slot * slot_cur)
+static int is_bus_empty(struct slot *slot_cur)
 {
 	int rc;
-	struct slot * tmp_slot;
+	struct slot *tmp_slot;
 	u8 i = slot_cur->bus_on->slot_min;
 
 	while (i <= slot_cur->bus_on->slot_max) {
@@ -856,7 +856,7 @@ static int is_bus_empty(struct slot * slot_cur)
  * Parameters: slot
  * Returns: bus is set (0) or error code
  ***********************************************************/
-static int set_bus(struct slot * slot_cur)
+static int set_bus(struct slot *slot_cur)
 {
 	int rc;
 	u8 speed;
@@ -956,7 +956,7 @@ static int set_bus(struct slot * slot_cur)
 static int check_limitations(struct slot *slot_cur)
 {
 	u8 i;
-	struct slot * tmp_slot;
+	struct slot *tmp_slot;
 	u8 count = 0;
 	u8 limitation = 0;
 
@@ -1045,8 +1045,7 @@ static int enable_slot(struct hotplug_slot *hs)
 	rc = check_limitations(slot_cur);
 	if (rc) {
 		err("Adding this card exceeds the limitations of this bus.\n");
-		err("(i.e., >1 133MHz cards running on same bus, or "
-		     ">2 66 PCI cards running on same bus.\n");
+		err("(i.e., >1 133MHz cards running on same bus, or >2 66 PCI cards running on same bus.\n");
 		err("Try hot-adding into another bus\n");
 		rc = -EINVAL;
 		goto error_nopower;
@@ -1070,12 +1069,10 @@ static int enable_slot(struct hotplug_slot *hs)
 					!(SLOT_PWRGD(slot_cur->status)))
 			err("power fault occurred trying to power up\n");
 		else if (SLOT_BUS_SPEED(slot_cur->status)) {
-			err("bus speed mismatch occurred.  please check "
-				"current bus speed and card capability\n");
+			err("bus speed mismatch occurred.  please check current bus speed and card capability\n");
 			print_card_capability(slot_cur);
 		} else if (SLOT_BUS_MODE(slot_cur->ext_status)) {
-			err("bus mode mismatch occurred.  please check "
-				"current bus mode and card capability\n");
+			err("bus mode mismatch occurred.  please check current bus mode and card capability\n");
 			print_card_capability(slot_cur);
 		}
 		ibmphp_update_slot_info(slot_cur);
@@ -1098,8 +1095,7 @@ static int enable_slot(struct hotplug_slot *hs)
 		goto error_power;
 	}
 	if (SLOT_POWER(slot_cur->status) && (SLOT_BUS_SPEED(slot_cur->status))) {
-		err("bus speed mismatch occurred.  please check current bus "
-					"speed and card capability\n");
+		err("bus speed mismatch occurred.  please check current bus speed and card capability\n");
 		print_card_capability(slot_cur);
 		goto error_power;
 	}
diff --git a/drivers/pci/hotplug/ibmphp_ebda.c b/drivers/pci/hotplug/ibmphp_ebda.c
index bd044158b36c..0f65ac555434 100644
--- a/drivers/pci/hotplug/ibmphp_ebda.c
+++ b/drivers/pci/hotplug/ibmphp_ebda.c
@@ -563,7 +563,7 @@ static int first_slot_num (u8 slot_num, u8 first_slot, u8 var)
 	return rc;
 }
 
-static struct opt_rio_lo * find_rxe_num (u8 slot_num)
+static struct opt_rio_lo *find_rxe_num (u8 slot_num)
 {
 	struct opt_rio_lo *opt_lo_ptr;
 
@@ -575,7 +575,7 @@ static struct opt_rio_lo * find_rxe_num (u8 slot_num)
 	return NULL;
 }
 
-static struct opt_rio * find_chassis_num (u8 slot_num)
+static struct opt_rio *find_chassis_num (u8 slot_num)
 {
 	struct opt_rio *opt_vg_ptr;
 
@@ -593,7 +593,7 @@ static struct opt_rio * find_chassis_num (u8 slot_num)
 static u8 calculate_first_slot (u8 slot_num)
 {
 	u8 first_slot = 1;
-	struct slot * slot_cur;
+	struct slot *slot_cur;
 
 	list_for_each_entry(slot_cur, &ibmphp_slot_head, ibm_slot_list) {
 		if (slot_cur->ctrl) {
@@ -607,7 +607,7 @@ static u8 calculate_first_slot (u8 slot_num)
 
 #define SLOT_NAME_SIZE 30
 
-static char *create_file_name (struct slot * slot_cur)
+static char *create_file_name (struct slot *slot_cur)
 {
 	struct opt_rio *opt_vg_ptr = NULL;
 	struct opt_rio_lo *opt_lo_ptr = NULL;
@@ -1192,7 +1192,7 @@ int ibmphp_register_pci (void)
 	}
 	return rc;
 }
-static int ibmphp_probe (struct pci_dev * dev, const struct pci_device_id *ids)
+static int ibmphp_probe (struct pci_dev *dev, const struct pci_device_id *ids)
 {
 	struct controller *ctrl;
 
diff --git a/drivers/pci/hotplug/ibmphp_hpc.c b/drivers/pci/hotplug/ibmphp_hpc.c
index 5fc7a089f532..a936022956e6 100644
--- a/drivers/pci/hotplug/ibmphp_hpc.c
+++ b/drivers/pci/hotplug/ibmphp_hpc.c
@@ -533,7 +533,7 @@ static u8 hpc_readcmdtoindex (u8 cmd, u8 index)
 *
 * Return   0 or error codes
 *---------------------------------------------------------------------*/
-int ibmphp_hpc_readslot (struct slot * pslot, u8 cmd, u8 * pstatus)
+int ibmphp_hpc_readslot (struct slot *pslot, u8 cmd, u8 *pstatus)
 {
 	void __iomem *wpg_bbar = NULL;
 	struct controller *ctlr_ptr;
@@ -672,7 +672,7 @@ int ibmphp_hpc_readslot (struct slot * pslot, u8 cmd, u8 * pstatus)
 *
 * Action: issue a WRITE command to HPC
 *---------------------------------------------------------------------*/
-int ibmphp_hpc_writeslot (struct slot * pslot, u8 cmd)
+int ibmphp_hpc_writeslot (struct slot *pslot, u8 cmd)
 {
 	void __iomem *wpg_bbar = NULL;
 	struct controller *ctlr_ptr;
@@ -1102,7 +1102,7 @@ void __exit ibmphp_hpc_stop_poll_thread (void)
 * Value:
 *---------------------------------------------------------------------*/
 static int hpc_wait_ctlr_notworking (int timeout, struct controller *ctlr_ptr, void __iomem *wpg_bbar,
-				    u8 * pstatus)
+				    u8 *pstatus)
 {
 	int rc = 0;
 	u8 done = 0;
diff --git a/drivers/pci/hotplug/ibmphp_pci.c b/drivers/pci/hotplug/ibmphp_pci.c
index 639ea3a75e14..2fd296706ce7 100644
--- a/drivers/pci/hotplug/ibmphp_pci.c
+++ b/drivers/pci/hotplug/ibmphp_pci.c
@@ -47,7 +47,7 @@ static u8 find_sec_number (u8 primary_busno, u8 slotno);
  * We also assign the same irq numbers for multi function devices.
  * These are PIC mode, so shouldn't matter n.e.ways (hopefully)
  */
-static void assign_alt_irq (struct pci_func * cur_func, u8 class_code)
+static void assign_alt_irq (struct pci_func *cur_func, u8 class_code)
 {
 	int j;
 	for (j = 0; j < 4; j++) {
@@ -137,8 +137,8 @@ int ibmphp_configure_card (struct pci_func *func, u8 slotno)
 				     "Please choose another device.\n", cur_func->device);
 				return -ENODEV;
 			} else if (class == PCI_CLASS_DISPLAY_VGA) {
-				err ("The device %x is not supported for hot plugging. "
-				     "Please choose another device.\n", cur_func->device);
+				err ("The device %x is not supported for hot plugging. Please choose another device.\n",
+				     cur_func->device);
 				return -ENODEV;
 			}
 			switch (hdr_type) {
@@ -179,8 +179,8 @@ int ibmphp_configure_card (struct pci_func *func, u8 slotno)
 				case PCI_HEADER_TYPE_MULTIBRIDGE:
 					class >>= 8;
 					if (class != PCI_CLASS_BRIDGE_PCI) {
-						err ("This %x is not PCI-to-PCI bridge, and as is not supported for hot-plugging. "
-						     "Please insert another card.\n", cur_func->device);
+						err ("This %x is not PCI-to-PCI bridge, and as is not supported for hot-plugging.  Please insert another card.\n",
+						     cur_func->device);
 						return -ENODEV;
 					}
 					assign_alt_irq (cur_func, class_code);
@@ -247,8 +247,8 @@ int ibmphp_configure_card (struct pci_func *func, u8 slotno)
 					class >>= 8;
 					debug ("class now is %x\n", class);
 					if (class != PCI_CLASS_BRIDGE_PCI) {
-						err ("This %x is not PCI-to-PCI bridge, and as is not supported for hot-plugging. "
-						     "Please insert another card.\n", cur_func->device);
+						err ("This %x is not PCI-to-PCI bridge, and as is not supported for hot-plugging.  Please insert another card.\n",
+						     cur_func->device);
 						return -ENODEV;
 					}
 
@@ -1073,7 +1073,7 @@ error:
  * Input: bridge function
  * Output: amount of resources needed
  *****************************************************************************/
-static struct res_needed *scan_behind_bridge (struct pci_func * func, u8 busno)
+static struct res_needed *scan_behind_bridge (struct pci_func *func, u8 busno)
 {
 	int count, len[6];
 	u16 vendor_id;
@@ -1125,13 +1125,11 @@ static struct res_needed *scan_behind_bridge (struct pci_func * func, u8 busno)
 
 				class >>= 8;	/* to take revision out, class = class.subclass.prog i/f */
 				if (class == PCI_CLASS_NOT_DEFINED_VGA) {
-					err ("The device %x is VGA compatible and as is not supported for hot plugging. "
-					     "Please choose another device.\n", device);
+					err ("The device %x is VGA compatible and as is not supported for hot plugging.  Please choose another device.\n", device);
 					amount->not_correct = 1;
 					return amount;
 				} else if (class == PCI_CLASS_DISPLAY_VGA) {
-					err ("The device %x is not supported for hot plugging. "
-					     "Please choose another device.\n", device);
+					err ("The device %x is not supported for hot plugging.  Please choose another device.\n", device);
 					amount->not_correct = 1;
 					return amount;
 				}
@@ -1483,12 +1481,10 @@ static int unconfigure_boot_card (struct slot *slot_cur)
 			debug ("hdr_type %x, class %x\n", hdr_type, class);
 			class >>= 8;	/* to take revision out, class = class.subclass.prog i/f */
 			if (class == PCI_CLASS_NOT_DEFINED_VGA) {
-				err ("The device %x function %x is VGA compatible and is not supported for hot removing. "
-				     "Please choose another device.\n", device, function);
+				err ("The device %x function %x is VGA compatible and is not supported for hot removing.  Please choose another device.\n", device, function);
 				return -ENODEV;
 			} else if (class == PCI_CLASS_DISPLAY_VGA) {
-				err ("The device %x function %x is not supported for hot removing. "
-				     "Please choose another device.\n", device, function);
+				err ("The device %x function %x is not supported for hot removing.  Please choose another device.\n", device, function);
 				return -ENODEV;
 			}
 
@@ -1513,9 +1509,7 @@ static int unconfigure_boot_card (struct slot *slot_cur)
 				case PCI_HEADER_TYPE_BRIDGE:
 					class >>= 8;
 					if (class != PCI_CLASS_BRIDGE_PCI) {
-						err ("This device %x function %x is not PCI-to-PCI bridge, "
-						     "and is not supported for hot-removing. "
-						     "Please try another card.\n", device, function);
+						err ("This device %x function %x is not PCI-to-PCI bridge, and is not supported for hot-removing.  Please try another card.\n", device, function);
 						return -ENODEV;
 					}
 					rc = unconfigure_boot_bridge (busno, device, function);
@@ -1529,9 +1523,7 @@ static int unconfigure_boot_card (struct slot *slot_cur)
 				case PCI_HEADER_TYPE_MULTIBRIDGE:
 					class >>= 8;
 					if (class != PCI_CLASS_BRIDGE_PCI) {
-						err ("This device %x function %x is not PCI-to-PCI bridge, "
-						     "and is not supported for hot-removing. "
-						     "Please try another card.\n", device, function);
+						err ("This device %x function %x is not PCI-to-PCI bridge,  and is not supported for hot-removing.  Please try another card.\n", device, function);
 						return -ENODEV;
 					}
 					rc = unconfigure_boot_bridge (busno, device, function);
diff --git a/drivers/pci/hotplug/ibmphp_res.c b/drivers/pci/hotplug/ibmphp_res.c
index a265acb2d518..f34745abd5b6 100644
--- a/drivers/pci/hotplug/ibmphp_res.c
+++ b/drivers/pci/hotplug/ibmphp_res.c
@@ -46,9 +46,9 @@ static struct bus_node *find_bus_wprev (u8, struct bus_node **, u8);
 
 static LIST_HEAD(gbuses);
 
-static struct bus_node * __init alloc_error_bus (struct ebda_pci_rsrc * curr, u8 busno, int flag)
+static struct bus_node * __init alloc_error_bus (struct ebda_pci_rsrc *curr, u8 busno, int flag)
 {
-	struct bus_node * newbus;
+	struct bus_node *newbus;
 
 	if (!(curr) && !(flag)) {
 		err ("NULL pointer passed\n");
@@ -69,7 +69,7 @@ static struct bus_node * __init alloc_error_bus (struct ebda_pci_rsrc * curr, u8
 	return newbus;
 }
 
-static struct resource_node * __init alloc_resources (struct ebda_pci_rsrc * curr)
+static struct resource_node * __init alloc_resources (struct ebda_pci_rsrc *curr)
 {
 	struct resource_node *rs;
 
@@ -93,7 +93,7 @@ static struct resource_node * __init alloc_resources (struct ebda_pci_rsrc * cur
 
 static int __init alloc_bus_range (struct bus_node **new_bus, struct range_node **new_range, struct ebda_pci_rsrc *curr, int flag, u8 first_bus)
 {
-	struct bus_node * newbus;
+	struct bus_node *newbus;
 	struct range_node *newrange;
 	u8 num_ranges = 0;
 
@@ -789,8 +789,7 @@ int ibmphp_remove_resource (struct resource_node *res)
 	bus_cur = find_bus_wprev (res->busno, NULL, 0);
 
 	if (!bus_cur) {
-		err ("cannot find corresponding bus of the io resource to remove  "
-			"bailing out...\n");
+		err ("cannot find corresponding bus of the io resource to remove  bailing out...\n");
 		return -ENODEV;
 	}
 
@@ -934,9 +933,9 @@ int ibmphp_remove_resource (struct resource_node *res)
 	return 0;
 }
 
-static struct range_node * find_range (struct bus_node *bus_cur, struct resource_node * res)
+static struct range_node *find_range (struct bus_node *bus_cur, struct resource_node *res)
 {
-	struct range_node * range = NULL;
+	struct range_node *range = NULL;
 
 	switch (res->type) {
 		case IO:
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index cfa92a984e62..56d8486dc167 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -59,14 +59,12 @@ static bool debug;
 #define DRIVER_DESC	"PCI Hot Plug PCI Core"
 
 
-//////////////////////////////////////////////////////////////////
-
 static LIST_HEAD(pci_hotplug_slot_list);
 static DEFINE_MUTEX(pci_hp_mutex);
 
 /* Weee, fun with macros... */
-#define GET_STATUS(name,type)	\
-static int get_##name (struct hotplug_slot *slot, type *value)		\
+#define GET_STATUS(name, type)	\
+static int get_##name(struct hotplug_slot *slot, type *value)		\
 {									\
 	struct hotplug_slot_ops *ops = slot->ops;			\
 	int retval = 0;							\
@@ -92,42 +90,41 @@ static ssize_t power_read_file(struct pci_slot *slot, char *buf)
 
 	retval = get_power_status(slot->hotplug, &value);
 	if (retval)
-		goto exit;
-	retval = sprintf (buf, "%d\n", value);
-exit:
-	return retval;
+		return retval;
+
+	return sprintf(buf, "%d\n", value);
 }
 
 static ssize_t power_write_file(struct pci_slot *pci_slot, const char *buf,
-		size_t count)
+				size_t count)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	unsigned long lpower;
 	u8 power;
 	int retval = 0;
 
-	lpower = simple_strtoul (buf, NULL, 10);
+	lpower = simple_strtoul(buf, NULL, 10);
 	power = (u8)(lpower & 0xff);
-	dbg ("power = %d\n", power);
+	dbg("power = %d\n", power);
 
 	if (!try_module_get(slot->ops->owner)) {
 		retval = -ENODEV;
 		goto exit;
 	}
 	switch (power) {
-		case 0:
-			if (slot->ops->disable_slot)
-				retval = slot->ops->disable_slot(slot);
-			break;
-
-		case 1:
-			if (slot->ops->enable_slot)
-				retval = slot->ops->enable_slot(slot);
-			break;
-
-		default:
-			err ("Illegal value specified for power\n");
-			retval = -EINVAL;
+	case 0:
+		if (slot->ops->disable_slot)
+			retval = slot->ops->disable_slot(slot);
+		break;
+
+	case 1:
+		if (slot->ops->enable_slot)
+			retval = slot->ops->enable_slot(slot);
+		break;
+
+	default:
+		err("Illegal value specified for power\n");
+		retval = -EINVAL;
 	}
 	module_put(slot->ops->owner);
 
@@ -150,24 +147,22 @@ static ssize_t attention_read_file(struct pci_slot *slot, char *buf)
 
 	retval = get_attention_status(slot->hotplug, &value);
 	if (retval)
-		goto exit;
-	retval = sprintf(buf, "%d\n", value);
+		return retval;
 
-exit:
-	return retval;
+	return sprintf(buf, "%d\n", value);
 }
 
 static ssize_t attention_write_file(struct pci_slot *slot, const char *buf,
-		size_t count)
+				    size_t count)
 {
 	struct hotplug_slot_ops *ops = slot->hotplug->ops;
 	unsigned long lattention;
 	u8 attention;
 	int retval = 0;
 
-	lattention = simple_strtoul (buf, NULL, 10);
+	lattention = simple_strtoul(buf, NULL, 10);
 	attention = (u8)(lattention & 0xff);
-	dbg (" - attention = %d\n", attention);
+	dbg(" - attention = %d\n", attention);
 
 	if (!try_module_get(ops->owner)) {
 		retval = -ENODEV;
@@ -196,11 +191,9 @@ static ssize_t latch_read_file(struct pci_slot *slot, char *buf)
 
 	retval = get_latch_status(slot->hotplug, &value);
 	if (retval)
-		goto exit;
-	retval = sprintf (buf, "%d\n", value);
+		return retval;
 
-exit:
-	return retval;
+	return sprintf(buf, "%d\n", value);
 }
 
 static struct pci_slot_attribute hotplug_slot_attr_latch = {
@@ -215,11 +208,9 @@ static ssize_t presence_read_file(struct pci_slot *slot, char *buf)
 
 	retval = get_adapter_status(slot->hotplug, &value);
 	if (retval)
-		goto exit;
-	retval = sprintf (buf, "%d\n", value);
+		return retval;
 
-exit:
-	return retval;
+	return sprintf(buf, "%d\n", value);
 }
 
 static struct pci_slot_attribute hotplug_slot_attr_presence = {
@@ -228,7 +219,7 @@ static struct pci_slot_attribute hotplug_slot_attr_presence = {
 };
 
 static ssize_t test_write_file(struct pci_slot *pci_slot, const char *buf,
-		size_t count)
+			       size_t count)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	unsigned long ltest;
@@ -237,7 +228,7 @@ static ssize_t test_write_file(struct pci_slot *pci_slot, const char *buf,
 
 	ltest = simple_strtoul (buf, NULL, 10);
 	test = (u32)(ltest & 0xffffffff);
-	dbg ("test = %d\n", test);
+	dbg("test = %d\n", test);
 
 	if (!try_module_get(slot->ops->owner)) {
 		retval = -ENODEV;
@@ -261,6 +252,7 @@ static struct pci_slot_attribute hotplug_slot_attr_test = {
 static bool has_power_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
+
 	if ((!slot) || (!slot->ops))
 		return false;
 	if ((slot->ops->enable_slot) ||
@@ -273,6 +265,7 @@ static bool has_power_file(struct pci_slot *pci_slot)
 static bool has_attention_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
+
 	if ((!slot) || (!slot->ops))
 		return false;
 	if ((slot->ops->set_attention_status) ||
@@ -284,6 +277,7 @@ static bool has_attention_file(struct pci_slot *pci_slot)
 static bool has_latch_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
+
 	if ((!slot) || (!slot->ops))
 		return false;
 	if (slot->ops->get_latch_status)
@@ -294,6 +288,7 @@ static bool has_latch_file(struct pci_slot *pci_slot)
 static bool has_adapter_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
+
 	if ((!slot) || (!slot->ops))
 		return false;
 	if (slot->ops->get_adapter_status)
@@ -304,6 +299,7 @@ static bool has_adapter_file(struct pci_slot *pci_slot)
 static bool has_test_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
+
 	if ((!slot) || (!slot->ops))
 		return false;
 	if (slot->ops->hardware_test)
@@ -397,13 +393,13 @@ static void fs_remove_slot(struct pci_slot *slot)
 	pci_hp_remove_module_link(slot);
 }
 
-static struct hotplug_slot *get_slot_from_name (const char *name)
+static struct hotplug_slot *get_slot_from_name(const char *name)
 {
 	struct hotplug_slot *slot;
 	struct list_head *tmp;
 
-	list_for_each (tmp, &pci_hotplug_slot_list) {
-		slot = list_entry (tmp, struct hotplug_slot, slot_list);
+	list_for_each(tmp, &pci_hotplug_slot_list) {
+		slot = list_entry(tmp, struct hotplug_slot, slot_list);
 		if (strcmp(hotplug_slot_name(slot), name) == 0)
 			return slot;
 	}
@@ -436,8 +432,7 @@ int __pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus,
 	if ((slot->info == NULL) || (slot->ops == NULL))
 		return -EINVAL;
 	if (slot->release == NULL) {
-		dbg("Why are you trying to register a hotplug slot "
-		    "without a proper release function?\n");
+		dbg("Why are you trying to register a hotplug slot without a proper release function?\n");
 		return -EINVAL;
 	}
 
@@ -468,6 +463,7 @@ out:
 	mutex_unlock(&pci_hp_mutex);
 	return result;
 }
+EXPORT_SYMBOL_GPL(__pci_hp_register);
 
 /**
  * pci_hp_deregister - deregister a hotplug_slot with the PCI hotplug subsystem
@@ -506,6 +502,7 @@ int pci_hp_deregister(struct hotplug_slot *hotplug)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(pci_hp_deregister);
 
 /**
  * pci_hp_change_slot_info - changes the slot's information structure in the core
@@ -527,24 +524,23 @@ int pci_hp_change_slot_info(struct hotplug_slot *hotplug,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(pci_hp_change_slot_info);
 
-static int __init pci_hotplug_init (void)
+static int __init pci_hotplug_init(void)
 {
 	int result;
 
 	result = cpci_hotplug_init(debug);
 	if (result) {
-		err ("cpci_hotplug_init with error %d\n", result);
-		goto err_cpci;
+		err("cpci_hotplug_init with error %d\n", result);
+		return result;
 	}
 
-	info (DRIVER_DESC " version: " DRIVER_VERSION "\n");
-
-err_cpci:
+	info(DRIVER_DESC " version: " DRIVER_VERSION "\n");
 	return result;
 }
 
-static void __exit pci_hotplug_exit (void)
+static void __exit pci_hotplug_exit(void)
 {
 	cpci_hotplug_exit();
 }
@@ -557,7 +553,3 @@ MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_LICENSE("GPL");
 module_param(debug, bool, 0644);
 MODULE_PARM_DESC(debug, "Debugging mode enabled or not");
-
-EXPORT_SYMBOL_GPL(__pci_hp_register);
-EXPORT_SYMBOL_GPL(pci_hp_deregister);
-EXPORT_SYMBOL_GPL(pci_hp_change_slot_info);
diff --git a/drivers/pci/hotplug/pciehp_acpi.c b/drivers/pci/hotplug/pciehp_acpi.c
index 20fea57d2149..93cc9266e8cb 100644
--- a/drivers/pci/hotplug/pciehp_acpi.c
+++ b/drivers/pci/hotplug/pciehp_acpi.c
@@ -103,10 +103,10 @@ static int __init dummy_probe(struct pcie_device *dev)
 }
 
 static struct pcie_port_service_driver __initdata dummy_driver = {
-        .name           = "pciehp_dummy",
+	.name		= "pciehp_dummy",
 	.port_type	= PCIE_ANY_PORT,
 	.service	= PCIE_PORT_SERVICE_HP,
-        .probe          = dummy_probe,
+	.probe		= dummy_probe,
 };
 
 static int __init select_detection_mode(void)
diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c
index 0e0a2fff20a3..a2297db80813 100644
--- a/drivers/pci/hotplug/pciehp_core.c
+++ b/drivers/pci/hotplug/pciehp_core.c
@@ -266,8 +266,7 @@ static int pciehp_probe(struct pcie_device *dev)
 	rc = init_slot(ctrl);
 	if (rc) {
 		if (rc == -EBUSY)
-			ctrl_warn(ctrl, "Slot already registered by another "
-				  "hotplug driver\n");
+			ctrl_warn(ctrl, "Slot already registered by another hotplug driver\n");
 		else
 			ctrl_err(ctrl, "Slot initialization failed\n");
 		goto err_out_release_ctlr;
@@ -312,12 +311,12 @@ static void pciehp_remove(struct pcie_device *dev)
 }
 
 #ifdef CONFIG_PM
-static int pciehp_suspend (struct pcie_device *dev)
+static int pciehp_suspend(struct pcie_device *dev)
 {
 	return 0;
 }
 
-static int pciehp_resume (struct pcie_device *dev)
+static int pciehp_resume(struct pcie_device *dev)
 {
 	struct controller *ctrl;
 	struct slot *slot;
diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c
index c75e6a678dcc..ff32e85e1de6 100644
--- a/drivers/pci/hotplug/pciehp_ctrl.c
+++ b/drivers/pci/hotplug/pciehp_ctrl.c
@@ -175,7 +175,7 @@ void pciehp_handle_linkstate_change(struct slot *p_slot)
    hotplug controller logic
  */
 
-static void set_slot_off(struct controller *ctrl, struct slot * pslot)
+static void set_slot_off(struct controller *ctrl, struct slot *pslot)
 {
 	/* turn off slot, turn on Amber LED, turn off Green LED if supported*/
 	if (POWER_CTRL(ctrl)) {
@@ -376,14 +376,12 @@ static void handle_button_press_event(struct slot *p_slot)
 		pciehp_get_power_status(p_slot, &getstatus);
 		if (getstatus) {
 			p_slot->state = BLINKINGOFF_STATE;
-			ctrl_info(ctrl,
-				  "PCI slot #%s - powering off due to button "
-				  "press.\n", slot_name(p_slot));
+			ctrl_info(ctrl, "PCI slot #%s - powering off due to button press\n",
+				  slot_name(p_slot));
 		} else {
 			p_slot->state = BLINKINGON_STATE;
-			ctrl_info(ctrl,
-				  "PCI slot #%s - powering on due to button "
-				  "press.\n", slot_name(p_slot));
+			ctrl_info(ctrl, "PCI slot #%s - powering on due to button press\n",
+				  slot_name(p_slot));
 		}
 		/* blink green LED and turn off amber */
 		pciehp_green_led_blink(p_slot);
@@ -404,8 +402,8 @@ static void handle_button_press_event(struct slot *p_slot)
 		else
 			pciehp_green_led_off(p_slot);
 		pciehp_set_attention_status(p_slot, 0);
-		ctrl_info(ctrl, "PCI slot #%s - action canceled "
-			  "due to button press\n", slot_name(p_slot));
+		ctrl_info(ctrl, "PCI slot #%s - action canceled due to button press\n",
+			  slot_name(p_slot));
 		p_slot->state = STATIC_STATE;
 		break;
 	case POWEROFF_STATE:
diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c
index 1463412cf7f8..42914e04d110 100644
--- a/drivers/pci/hotplug/pciehp_hpc.c
+++ b/drivers/pci/hotplug/pciehp_hpc.c
@@ -174,12 +174,10 @@ static void pcie_write_cmd(struct controller *ctrl, u16 cmd, u16 mask)
 			 * event even though it supports none of power
 			 * controller, attention led, power led and EMI.
 			 */
-			ctrl_dbg(ctrl, "Unexpected CMD_COMPLETED. Need to "
-				 "wait for command completed event.\n");
+			ctrl_dbg(ctrl, "Unexpected CMD_COMPLETED. Need to wait for command completed event\n");
 			ctrl->no_cmd_complete = 0;
 		} else {
-			ctrl_dbg(ctrl, "Unexpected CMD_COMPLETED. Maybe "
-				 "the controller is broken.\n");
+			ctrl_dbg(ctrl, "Unexpected CMD_COMPLETED. Maybe the controller is broken\n");
 		}
 	}
 
@@ -203,7 +201,7 @@ static void pcie_write_cmd(struct controller *ctrl, u16 cmd, u16 mask)
 		if (!(slot_ctrl & PCI_EXP_SLTCTL_HPIE) ||
 		    !(slot_ctrl & PCI_EXP_SLTCTL_CCIE))
 			poll = 1;
-                pcie_wait_cmd(ctrl, poll);
+		pcie_wait_cmd(ctrl, poll);
 	}
 	mutex_unlock(&ctrl->ctrl_lock);
 }
@@ -276,15 +274,15 @@ int pciehp_check_link_status(struct controller *ctrl)
 	bool found;
 	u16 lnk_status;
 
-        /*
-         * Data Link Layer Link Active Reporting must be capable for
-         * hot-plug capable downstream port. But old controller might
-         * not implement it. In this case, we wait for 1000 ms.
-         */
-        if (ctrl->link_active_reporting)
-                pcie_wait_link_active(ctrl);
-        else
-                msleep(1000);
+	/*
+	 * Data Link Layer Link Active Reporting must be capable for
+	 * hot-plug capable downstream port. But old controller might
+	 * not implement it. In this case, we wait for 1000 ms.
+	*/
+	if (ctrl->link_active_reporting)
+		pcie_wait_link_active(ctrl);
+	else
+		msleep(1000);
 
 	/* wait 100ms before read pci conf, and try in 1s */
 	msleep(100);
@@ -295,7 +293,7 @@ int pciehp_check_link_status(struct controller *ctrl)
 	ctrl_dbg(ctrl, "%s: lnk_status = %x\n", __func__, lnk_status);
 	if ((lnk_status & PCI_EXP_LNKSTA_LT) ||
 	    !(lnk_status & PCI_EXP_LNKSTA_NLW)) {
-		ctrl_err(ctrl, "Link Training Error occurs \n");
+		ctrl_err(ctrl, "Link Training Error occurs\n");
 		return -1;
 	}
 
@@ -414,7 +412,7 @@ void pciehp_set_attention_status(struct slot *slot, u8 value)
 		return;
 
 	switch (value) {
-	case 0 :	/* turn off */
+	case 0:		/* turn off */
 		slot_cmd = PCI_EXP_SLTCTL_ATTN_IND_OFF;
 		break;
 	case 1:		/* turn on */
@@ -470,7 +468,7 @@ void pciehp_green_led_blink(struct slot *slot)
 		 PCI_EXP_SLTCTL_PWR_IND_BLINK);
 }
 
-int pciehp_power_on_slot(struct slot * slot)
+int pciehp_power_on_slot(struct slot *slot)
 {
 	struct controller *ctrl = slot->ctrl;
 	struct pci_dev *pdev = ctrl_dev(ctrl);
@@ -496,7 +494,7 @@ int pciehp_power_on_slot(struct slot * slot)
 	return retval;
 }
 
-void pciehp_power_off_slot(struct slot * slot)
+void pciehp_power_off_slot(struct slot *slot)
 {
 	struct controller *ctrl = slot->ctrl;
 
@@ -756,7 +754,7 @@ static inline void dbg_ctrl(struct controller *ctrl)
 	ctrl_info(ctrl, "Slot Control           : 0x%04x\n", reg16);
 }
 
-#define FLAG(x,y)	(((x) & (y)) ? '+' : '-')
+#define FLAG(x, y)	(((x) & (y)) ? '+' : '-')
 
 struct controller *pcie_init(struct pcie_device *dev)
 {
@@ -783,14 +781,14 @@ struct controller *pcie_init(struct pcie_device *dev)
 	 */
 	if (NO_CMD_CMPL(ctrl) ||
 	    !(POWER_CTRL(ctrl) | ATTN_LED(ctrl) | PWR_LED(ctrl) | EMI(ctrl)))
-	    ctrl->no_cmd_complete = 1;
-
-        /* Check if Data Link Layer Link Active Reporting is implemented */
-        pcie_capability_read_dword(pdev, PCI_EXP_LNKCAP, &link_cap);
-        if (link_cap & PCI_EXP_LNKCAP_DLLLARC) {
-                ctrl_dbg(ctrl, "Link Active Reporting supported\n");
-                ctrl->link_active_reporting = 1;
-        }
+		ctrl->no_cmd_complete = 1;
+
+	/* Check if Data Link Layer Link Active Reporting is implemented */
+	pcie_capability_read_dword(pdev, PCI_EXP_LNKCAP, &link_cap);
+	if (link_cap & PCI_EXP_LNKCAP_DLLLARC) {
+		ctrl_dbg(ctrl, "Link Active Reporting supported\n");
+		ctrl->link_active_reporting = 1;
+	}
 
 	/* Clear all remaining event bits in Slot Status register */
 	pcie_capability_write_word(pdev, PCI_EXP_SLTSTA,
diff --git a/drivers/pci/hotplug/pciehp_pci.c b/drivers/pci/hotplug/pciehp_pci.c
index b6cb1df67097..5f871f4c4af1 100644
--- a/drivers/pci/hotplug/pciehp_pci.c
+++ b/drivers/pci/hotplug/pciehp_pci.c
@@ -46,9 +46,8 @@ int pciehp_configure_device(struct slot *p_slot)
 
 	dev = pci_get_slot(parent, PCI_DEVFN(0, 0));
 	if (dev) {
-		ctrl_err(ctrl, "Device %s already exists "
-			 "at %04x:%02x:00, cannot hot-add\n", pci_name(dev),
-			 pci_domain_nr(parent), parent->number);
+		ctrl_err(ctrl, "Device %s already exists at %04x:%02x:00, cannot hot-add\n",
+			 pci_name(dev), pci_domain_nr(parent), parent->number);
 		pci_dev_put(dev);
 		ret = -EEXIST;
 		goto out;
diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c
index ac69094e4b20..d062c008fc95 100644
--- a/drivers/pci/hotplug/pcihp_skeleton.c
+++ b/drivers/pci/hotplug/pcihp_skeleton.c
@@ -51,7 +51,7 @@ static LIST_HEAD(slot_list);
 #define dbg(format, arg...)					\
 	do {							\
 		if (debug)					\
-			printk (KERN_DEBUG "%s: " format "\n",	\
+			printk(KERN_DEBUG "%s: " format "\n",	\
 				MY_NAME , ## arg);		\
 	} while (0)
 #define err(format, arg...) printk(KERN_ERR "%s: " format "\n", MY_NAME , ## arg)
@@ -128,18 +128,18 @@ static int set_attention_status(struct hotplug_slot *hotplug_slot, u8 status)
 	dbg("%s - physical_slot = %s\n", __func__, hotplug_slot->name);
 
 	switch (status) {
-		case 0:
-			/*
-			 * Fill in code here to turn light off
-			 */
-			break;
-
-		case 1:
-		default:
-			/*
-			 * Fill in code here to turn light on
-			 */
-			break;
+	case 0:
+		/*
+		 * Fill in code here to turn light off
+		 */
+		break;
+
+	case 1:
+	default:
+		/*
+		 * Fill in code here to turn light on
+		 */
+		break;
 	}
 
 	return retval;
@@ -153,12 +153,12 @@ static int hardware_test(struct hotplug_slot *hotplug_slot, u32 value)
 	dbg("%s - physical_slot = %s\n", __func__, hotplug_slot->name);
 
 	switch (value) {
-		case 0:
-			/* Specify a test here */
-			break;
-		case 1:
-			/* Specify another test here */
-			break;
+	case 0:
+		/* Specify a test here */
+		break;
+	case 1:
+		/* Specify another test here */
+		break;
 	}
 
 	return retval;
diff --git a/drivers/pci/hotplug/rpaphp_core.c b/drivers/pci/hotplug/rpaphp_core.c
index 984d708552f6..93aa29f6d39c 100644
--- a/drivers/pci/hotplug/rpaphp_core.c
+++ b/drivers/pci/hotplug/rpaphp_core.c
@@ -39,6 +39,7 @@
 
 bool rpaphp_debug;
 LIST_HEAD(rpaphp_slot_head);
+EXPORT_SYMBOL_GPL(rpaphp_slot_head);
 
 #define DRIVER_VERSION	"0.1"
 #define DRIVER_AUTHOR	"Linda Xie <lxie@us.ibm.com>"
@@ -88,7 +89,7 @@ static int set_attention_status(struct hotplug_slot *hotplug_slot, u8 value)
  * @hotplug_slot: slot to get status
  * @value: pointer to store status
  */
-static int get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_power_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	int retval, level;
 	struct slot *slot = (struct slot *)hotplug_slot->private;
@@ -104,14 +105,14 @@ static int get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
  * @hotplug_slot: slot to get status
  * @value: pointer to store status
  */
-static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	struct slot *slot = (struct slot *)hotplug_slot->private;
 	*value = slot->hotplug_slot->info->attention_status;
 	return 0;
 }
 
-static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 * value)
+static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	struct slot *slot = (struct slot *)hotplug_slot->private;
 	int rc, state;
@@ -241,6 +242,7 @@ int rpaphp_get_drc_props(struct device_node *dn, int *drc_index,
 
 	return -EINVAL;
 }
+EXPORT_SYMBOL_GPL(rpaphp_get_drc_props);
 
 static int is_php_type(char *drc_type)
 {
@@ -350,6 +352,7 @@ int rpaphp_add_slot(struct device_node *dn)
 	/* XXX FIXME: reports a failure only if last entry in loop failed */
 	return retval;
 }
+EXPORT_SYMBOL_GPL(rpaphp_add_slot);
 
 static void __exit cleanup_slots(void)
 {
@@ -443,7 +446,3 @@ struct hotplug_slot_ops rpaphp_hotplug_slot_ops = {
 
 module_init(rpaphp_init);
 module_exit(rpaphp_exit);
-
-EXPORT_SYMBOL_GPL(rpaphp_add_slot);
-EXPORT_SYMBOL_GPL(rpaphp_slot_head);
-EXPORT_SYMBOL_GPL(rpaphp_get_drc_props);
diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c
index 613043f7576f..bada20999870 100644
--- a/drivers/pci/hotplug/sgi_hotplug.c
+++ b/drivers/pci/hotplug/sgi_hotplug.c
@@ -188,7 +188,7 @@ static int sn_hp_slot_private_alloc(struct hotplug_slot *bss_hotplug_slot,
 	return 0;
 }
 
-static struct hotplug_slot * sn_hp_destroy(void)
+static struct hotplug_slot *sn_hp_destroy(void)
 {
 	struct slot *slot;
 	struct pci_slot *pci_slot;
@@ -250,15 +250,13 @@ static int sn_slot_enable(struct hotplug_slot *bss_hotplug_slot,
 	}
 
 	if (rc == PCI_L1_ERR) {
-		dev_dbg(&slot->pci_bus->self->dev,
-			"L1 failure %d with message: %s",
+		dev_dbg(&slot->pci_bus->self->dev, "L1 failure %d with message: %s",
 			resp.resp_sub_errno, resp.resp_l1_msg);
 		return -EPERM;
 	}
 
 	if (rc) {
-		dev_dbg(&slot->pci_bus->self->dev,
-			"insert failed with error %d sub-error %d\n",
+		dev_dbg(&slot->pci_bus->self->dev, "insert failed with error %d sub-error %d\n",
 			rc, resp.resp_sub_errno);
 		return -EIO;
 	}
@@ -288,21 +286,18 @@ static int sn_slot_disable(struct hotplug_slot *bss_hotplug_slot,
 	}
 
 	if ((action == PCI_REQ_SLOT_ELIGIBLE) && (rc == PCI_EMPTY_33MHZ)) {
-		dev_dbg(&slot->pci_bus->self->dev,
-			"Cannot remove last 33MHz card\n");
+		dev_dbg(&slot->pci_bus->self->dev, "Cannot remove last 33MHz card\n");
 		return -EPERM;
 	}
 
 	if ((action == PCI_REQ_SLOT_ELIGIBLE) && (rc == PCI_L1_ERR)) {
-		dev_dbg(&slot->pci_bus->self->dev,
-			"L1 failure %d with message \n%s\n",
+		dev_dbg(&slot->pci_bus->self->dev, "L1 failure %d with message \n%s\n",
 			resp.resp_sub_errno, resp.resp_l1_msg);
 		return -EPERM;
 	}
 
 	if ((action == PCI_REQ_SLOT_ELIGIBLE) && rc) {
-		dev_dbg(&slot->pci_bus->self->dev,
-			"remove failed with error %d sub-error %d\n",
+		dev_dbg(&slot->pci_bus->self->dev, "remove failed with error %d sub-error %d\n",
 			rc, resp.resp_sub_errno);
 		return -EIO;
 	}
@@ -417,8 +412,7 @@ static int enable_slot(struct hotplug_slot *bss_hotplug_slot)
 		phandle = acpi_device_handle(PCI_CONTROLLER(slot->pci_bus)->companion);
 
 		if (acpi_bus_get_device(phandle, &pdevice)) {
-			dev_dbg(&slot->pci_bus->self->dev,
-				"no parent device, assuming NULL\n");
+			dev_dbg(&slot->pci_bus->self->dev, "no parent device, assuming NULL\n");
 			pdevice = NULL;
 		}
 
@@ -447,10 +441,8 @@ static int enable_slot(struct hotplug_slot *bss_hotplug_slot)
 
 				ret = acpi_bus_scan(chandle);
 				if (ACPI_FAILURE(ret)) {
-					printk(KERN_ERR "%s: acpi_bus_scan "
-					       "failed (0x%x) for slot %d "
-					       "func %d\n", __func__,
-					       ret, (int)(adr>>16),
+					printk(KERN_ERR "%s: acpi_bus_scan failed (0x%x) for slot %d func %d\n",
+					       __func__, ret, (int)(adr>>16),
 					       (int)(adr&0xffff));
 					/* try to continue on */
 				}
@@ -471,11 +463,9 @@ static int enable_slot(struct hotplug_slot *bss_hotplug_slot)
 	mutex_unlock(&sn_hotplug_mutex);
 
 	if (rc == 0)
-		dev_dbg(&slot->pci_bus->self->dev,
-			"insert operation successful\n");
+		dev_dbg(&slot->pci_bus->self->dev, "insert operation successful\n");
 	else
-		dev_dbg(&slot->pci_bus->self->dev,
-			"insert operation failed rc = %d\n", rc);
+		dev_dbg(&slot->pci_bus->self->dev, "insert operation failed rc = %d\n", rc);
 
 	return rc;
 }
@@ -561,8 +551,7 @@ static int disable_slot(struct hotplug_slot *bss_hotplug_slot)
 		acpi_status ret;
 		ret = acpi_unload_table_id(ssdt_id);
 		if (ACPI_FAILURE(ret)) {
-			printk(KERN_ERR "%s: acpi_unload_table_id "
-			       "failed (0x%x) for id %d\n",
+			printk(KERN_ERR "%s: acpi_unload_table_id failed (0x%x) for id %d\n",
 			       __func__, ret, ssdt_id);
 			/* try to continue on */
 		}
diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h
index 61529097464d..5897d516427b 100644
--- a/drivers/pci/hotplug/shpchp.h
+++ b/drivers/pci/hotplug/shpchp.h
@@ -180,7 +180,7 @@ int shpchp_configure_device(struct slot *p_slot);
 int shpchp_unconfigure_device(struct slot *p_slot);
 void cleanup_slots(struct controller *ctrl);
 void shpchp_queue_pushbutton_work(struct work_struct *work);
-int shpc_init( struct controller *ctrl, struct pci_dev *pdev);
+int shpc_init(struct controller *ctrl, struct pci_dev *pdev);
 
 static inline const char *slot_name(struct slot *slot)
 {
@@ -295,7 +295,7 @@ static inline void amd_pogo_errata_restore_misc_reg(struct slot *p_slot)
 		pci_write_config_dword(p_slot->ctrl->pci_dev, PCIX_MEM_BASE_LIMIT_OFFSET, rse_set);
 	}
 	/* restore MiscII register */
-	pci_read_config_dword( p_slot->ctrl->pci_dev, PCIX_MISCII_OFFSET, &pcix_misc2_temp );
+	pci_read_config_dword(p_slot->ctrl->pci_dev, PCIX_MISCII_OFFSET, &pcix_misc2_temp );
 
 	if (p_slot->ctrl->pcix_misc2_reg & SERRFATALENABLE_MASK)
 		pcix_misc2_temp |= SERRFATALENABLE_MASK;
diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c
index faf13abd5b99..294ef4b10cf1 100644
--- a/drivers/pci/hotplug/shpchp_core.c
+++ b/drivers/pci/hotplug/shpchp_core.c
@@ -143,8 +143,7 @@ static int init_slots(struct controller *ctrl)
 		snprintf(name, SLOT_NAME_SIZE, "%d", slot->number);
 		hotplug_slot->ops = &shpchp_hotplug_slot_ops;
 
-		ctrl_dbg(ctrl, "Registering domain:bus:dev=%04x:%02x:%02x "
-			 "hp_slot=%x sun=%x slot_device_offset=%x\n",
+		ctrl_dbg(ctrl, "Registering domain:bus:dev=%04x:%02x:%02x hp_slot=%x sun=%x slot_device_offset=%x\n",
 			 pci_domain_nr(ctrl->pci_dev->subordinate),
 			 slot->bus, slot->device, slot->hp_slot, slot->number,
 			 ctrl->slot_device_offset);
diff --git a/drivers/pci/hotplug/shpchp_ctrl.c b/drivers/pci/hotplug/shpchp_ctrl.c
index 6efc2ec5e4db..a81fb67ea9a1 100644
--- a/drivers/pci/hotplug/shpchp_ctrl.c
+++ b/drivers/pci/hotplug/shpchp_ctrl.c
@@ -162,7 +162,7 @@ u8 shpchp_handle_power_fault(u8 hp_slot, struct controller *ctrl)
 
 	p_slot = shpchp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset);
 
-	if ( !(p_slot->hpc_ops->query_power_fault(p_slot))) {
+	if (!(p_slot->hpc_ops->query_power_fault(p_slot))) {
 		/*
 		 * Power fault Cleared
 		 */
@@ -196,8 +196,8 @@ static int change_bus_speed(struct controller *ctrl, struct slot *p_slot,
 
 	ctrl_dbg(ctrl, "Change speed to %d\n", speed);
 	if ((rc = p_slot->hpc_ops->set_bus_speed_mode(p_slot, speed))) {
-		ctrl_err(ctrl, "%s: Issue of set bus speed mode command "
-			 "failed\n", __func__);
+		ctrl_err(ctrl, "%s: Issue of set bus speed mode command failed\n",
+			 __func__);
 		return WRONG_BUS_FREQUENCY;
 	}
 	return rc;
@@ -215,8 +215,8 @@ static int fix_bus_speed(struct controller *ctrl, struct slot *pslot,
 	 */
 	if (flag) {
 		if (asp < bsp) {
-			ctrl_err(ctrl, "Speed of bus %x and adapter %x "
-				 "mismatch\n", bsp, asp);
+			ctrl_err(ctrl, "Speed of bus %x and adapter %x mismatch\n",
+				 bsp, asp);
 			rc = WRONG_BUS_FREQUENCY;
 		}
 		return rc;
@@ -250,8 +250,7 @@ static int board_added(struct slot *p_slot)
 
 	hp_slot = p_slot->device - ctrl->slot_device_offset;
 
-	ctrl_dbg(ctrl,
-		 "%s: p_slot->device, slot_offset, hp_slot = %d, %d ,%d\n",
+	ctrl_dbg(ctrl, "%s: p_slot->device, slot_offset, hp_slot = %d, %d ,%d\n",
 		 __func__, p_slot->device, ctrl->slot_device_offset, hp_slot);
 
 	/* Power on slot without connecting to bus */
@@ -263,8 +262,8 @@ static int board_added(struct slot *p_slot)
 
 	if ((ctrl->pci_dev->vendor == 0x8086) && (ctrl->pci_dev->device == 0x0332)) {
 		if ((rc = p_slot->hpc_ops->set_bus_speed_mode(p_slot, PCI_SPEED_33MHz))) {
-			ctrl_err(ctrl, "%s: Issue of set bus speed mode command"
-				 " failed\n", __func__);
+			ctrl_err(ctrl, "%s: Issue of set bus speed mode command failed\n",
+				 __func__);
 			return WRONG_BUS_FREQUENCY;
 		}
 
@@ -277,8 +276,7 @@ static int board_added(struct slot *p_slot)
 
 	rc = p_slot->hpc_ops->get_adapter_speed(p_slot, &asp);
 	if (rc) {
-		ctrl_err(ctrl, "Can't get adapter speed or "
-			 "bus mode mismatch\n");
+		ctrl_err(ctrl, "Can't get adapter speed or bus mode mismatch\n");
 		return WRONG_BUS_FREQUENCY;
 	}
 
@@ -289,8 +287,8 @@ static int board_added(struct slot *p_slot)
 	if (!list_empty(&ctrl->pci_dev->subordinate->devices))
 		slots_not_empty = 1;
 
-	ctrl_dbg(ctrl, "%s: slots_not_empty %d, adapter_speed %d, bus_speed %d,"
-		 " max_bus_speed %d\n", __func__, slots_not_empty, asp,
+	ctrl_dbg(ctrl, "%s: slots_not_empty %d, adapter_speed %d, bus_speed %d, max_bus_speed %d\n",
+		 __func__, slots_not_empty, asp,
 		 bsp, msp);
 
 	rc = fix_bus_speed(ctrl, p_slot, slots_not_empty, asp, bsp, msp);
@@ -490,12 +488,12 @@ static void handle_button_press_event(struct slot *p_slot)
 		p_slot->hpc_ops->get_power_status(p_slot, &getstatus);
 		if (getstatus) {
 			p_slot->state = BLINKINGOFF_STATE;
-			ctrl_info(ctrl, "PCI slot #%s - powering off due to "
-				  "button press.\n", slot_name(p_slot));
+			ctrl_info(ctrl, "PCI slot #%s - powering off due to button press\n",
+				  slot_name(p_slot));
 		} else {
 			p_slot->state = BLINKINGON_STATE;
-			ctrl_info(ctrl, "PCI slot #%s - powering on due to "
-				  "button press.\n", slot_name(p_slot));
+			ctrl_info(ctrl, "PCI slot #%s - powering on due to button press\n",
+				  slot_name(p_slot));
 		}
 		/* blink green LED and turn off amber */
 		p_slot->hpc_ops->green_led_blink(p_slot);
@@ -518,8 +516,8 @@ static void handle_button_press_event(struct slot *p_slot)
 		else
 			p_slot->hpc_ops->green_led_off(p_slot);
 		p_slot->hpc_ops->set_attention_status(p_slot, 0);
-		ctrl_info(ctrl, "PCI slot #%s - action canceled due to "
-			  "button press\n", slot_name(p_slot));
+		ctrl_info(ctrl, "PCI slot #%s - action canceled due to button press\n",
+			  slot_name(p_slot));
 		p_slot->state = STATIC_STATE;
 		break;
 	case POWEROFF_STATE:
diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c
index 2d7f474ca0ec..29e22352822c 100644
--- a/drivers/pci/hotplug/shpchp_hpc.c
+++ b/drivers/pci/hotplug/shpchp_hpc.c
@@ -341,8 +341,7 @@ static int shpc_write_cmd(struct slot *slot, u8 t_slot, u8 cmd)
 
 	cmd_status = hpc_check_cmd_status(slot->ctrl);
 	if (cmd_status) {
-		ctrl_err(ctrl,
-			 "Failed to issued command 0x%x (error code = %d)\n",
+		ctrl_err(ctrl, "Failed to issued command 0x%x (error code = %d)\n",
 			 cmd, cmd_status);
 		retval = -EIO;
 	}
@@ -404,7 +403,7 @@ static int hpc_get_attention_status(struct slot *slot, u8 *status)
 	return 0;
 }
 
-static int hpc_get_power_status(struct slot * slot, u8 *status)
+static int hpc_get_power_status(struct slot *slot, u8 *status)
 {
 	struct controller *ctrl = slot->ctrl;
 	u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot));
@@ -528,7 +527,7 @@ static int hpc_get_mode1_ECC_cap(struct slot *slot, u8 *mode)
 	return retval;
 }
 
-static int hpc_query_power_fault(struct slot * slot)
+static int hpc_query_power_fault(struct slot *slot)
 {
 	struct controller *ctrl = slot->ctrl;
 	u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot));
@@ -614,7 +613,7 @@ static void hpc_release_ctlr(struct controller *ctrl)
 	release_mem_region(ctrl->mmio_base, ctrl->mmio_size);
 }
 
-static int hpc_power_on_slot(struct slot * slot)
+static int hpc_power_on_slot(struct slot *slot)
 {
 	int retval;
 
@@ -625,7 +624,7 @@ static int hpc_power_on_slot(struct slot * slot)
 	return retval;
 }
 
-static int hpc_slot_enable(struct slot * slot)
+static int hpc_slot_enable(struct slot *slot)
 {
 	int retval;
 
@@ -638,7 +637,7 @@ static int hpc_slot_enable(struct slot * slot)
 	return retval;
 }
 
-static int hpc_slot_disable(struct slot * slot)
+static int hpc_slot_disable(struct slot *slot)
 {
 	int retval;
 
@@ -720,7 +719,7 @@ static int shpc_get_cur_bus_speed(struct controller *ctrl)
 }
 
 
-static int hpc_set_bus_speed_mode(struct slot * slot, enum pci_bus_speed value)
+static int hpc_set_bus_speed_mode(struct slot *slot, enum pci_bus_speed value)
 {
 	int retval;
 	struct controller *ctrl = slot->ctrl;
@@ -974,8 +973,8 @@ int shpc_init(struct controller *ctrl, struct pci_dev *pdev)
 		for (i = 0; i < 9 + num_slots; i++) {
 			rc = shpc_indirect_read(ctrl, i, &tempdword);
 			if (rc) {
-				ctrl_err(ctrl,
-					 "Cannot read creg (index = %d)\n", i);
+				ctrl_err(ctrl, "Cannot read creg (index = %d)\n",
+					 i);
 				goto abort;
 			}
 			ctrl_dbg(ctrl, " offset %d: value %x\n", i, tempdword);
@@ -1060,10 +1059,8 @@ int shpc_init(struct controller *ctrl, struct pci_dev *pdev)
 		/* Installs the interrupt handler */
 		rc = pci_enable_msi(pdev);
 		if (rc) {
-			ctrl_info(ctrl,
-				  "Can't get msi for the hotplug controller\n");
-			ctrl_info(ctrl,
-				  "Use INTx for the hotplug controller\n");
+			ctrl_info(ctrl, "Can't get msi for the hotplug controller\n");
+			ctrl_info(ctrl, "Use INTx for the hotplug controller\n");
 		}
 
 		rc = request_irq(ctrl->pci_dev->irq, shpc_isr, IRQF_SHARED,
@@ -1071,8 +1068,8 @@ int shpc_init(struct controller *ctrl, struct pci_dev *pdev)
 		ctrl_dbg(ctrl, "request_irq %d (returns %d)\n",
 			 ctrl->pci_dev->irq, rc);
 		if (rc) {
-			ctrl_err(ctrl, "Can't get irq %d for the hotplug "
-				 "controller\n", ctrl->pci_dev->irq);
+			ctrl_err(ctrl, "Can't get irq %d for the hotplug controller\n",
+				 ctrl->pci_dev->irq);
 			goto abort_iounmap;
 		}
 	}
diff --git a/drivers/pci/hotplug/shpchp_pci.c b/drivers/pci/hotplug/shpchp_pci.c
index 9202d133485c..469454e0cc48 100644
--- a/drivers/pci/hotplug/shpchp_pci.c
+++ b/drivers/pci/hotplug/shpchp_pci.c
@@ -46,9 +46,9 @@ int shpchp_configure_device(struct slot *p_slot)
 
 	dev = pci_get_slot(parent, PCI_DEVFN(p_slot->device, 0));
 	if (dev) {
-		ctrl_err(ctrl, "Device %s already exists "
-			 "at %04x:%02x:%02x, cannot hot-add\n", pci_name(dev),
-			 pci_domain_nr(parent), p_slot->bus, p_slot->device);
+		ctrl_err(ctrl, "Device %s already exists at %04x:%02x:%02x, cannot hot-add\n",
+			 pci_name(dev), pci_domain_nr(parent),
+			 p_slot->bus, p_slot->device);
 		pci_dev_put(dev);
 		ret = -EINVAL;
 		goto out;
diff --git a/drivers/pci/hotplug/shpchp_sysfs.c b/drivers/pci/hotplug/shpchp_sysfs.c
index e8c31fe20566..52875b360463 100644
--- a/drivers/pci/hotplug/shpchp_sysfs.c
+++ b/drivers/pci/hotplug/shpchp_sysfs.c
@@ -38,7 +38,7 @@
 static ssize_t show_ctrl (struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev;
-	char * out = buf;
+	char *out = buf;
 	int index, busnr;
 	struct resource *res;
 	struct pci_bus *bus;
diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c
index d68b030ab533..a94dd2c4183a 100644
--- a/drivers/pci/htirq.c
+++ b/drivers/pci/htirq.c
@@ -102,7 +102,7 @@ int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update)
 	spin_unlock_irqrestore(&ht_irq_lock, flags);
 
 	max_irq = (data >> 16) & 0xff;
-	if ( idx > max_irq)
+	if (idx > max_irq)
 		return -EINVAL;
 
 	cfg = kmalloc(sizeof(*cfg), GFP_KERNEL);
@@ -131,6 +131,7 @@ int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update)
 
 	return irq;
 }
+EXPORT_SYMBOL(__ht_create_irq);
 
 /**
  * ht_create_irq - create an irq and attach it to a device.
@@ -146,6 +147,7 @@ int ht_create_irq(struct pci_dev *dev, int idx)
 {
 	return __ht_create_irq(dev, idx, NULL);
 }
+EXPORT_SYMBOL(ht_create_irq);
 
 /**
  * ht_destroy_irq - destroy an irq created with ht_create_irq
@@ -165,7 +167,4 @@ void ht_destroy_irq(unsigned int irq)
 
 	kfree(cfg);
 }
-
-EXPORT_SYMBOL(__ht_create_irq);
-EXPORT_SYMBOL(ht_create_irq);
 EXPORT_SYMBOL(ht_destroy_irq);
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 27a7e67ddfe4..13f3d3037272 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -413,7 +413,7 @@ static void free_msi_irqs(struct pci_dev *dev)
 	if (dev->msi_irq_groups) {
 		sysfs_remove_groups(&dev->dev.kobj, dev->msi_irq_groups);
 		msi_attrs = dev->msi_irq_groups[0]->attrs;
-		list_for_each_entry(entry, &dev->msi_list, list) {
+		while (msi_attrs[count]) {
 			dev_attr = container_of(msi_attrs[count],
 						struct device_attribute, attr);
 			kfree(dev_attr->attr.name);
@@ -980,8 +980,7 @@ int pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries, int nvec)
 
 	/* Check whether driver already requested for MSI irq */
 	if (dev->msi_enabled) {
-		dev_info(&dev->dev, "can't enable MSI-X "
-		       "(MSI IRQ already assigned)\n");
+		dev_info(&dev->dev, "can't enable MSI-X (MSI IRQ already assigned)\n");
 		return -EINVAL;
 	}
 	status = msix_capability_init(dev, entries, nvec);
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 837d71f5390b..3f8e3dbcaa7c 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -77,6 +77,7 @@ int pci_add_dynid(struct pci_driver *drv,
 
 	return retval;
 }
+EXPORT_SYMBOL_GPL(pci_add_dynid);
 
 static void pci_free_dynids(struct pci_driver *drv)
 {
@@ -98,15 +99,15 @@ static void pci_free_dynids(struct pci_driver *drv)
  *
  * Allow PCI IDs to be added to an existing driver via sysfs.
  */
-static ssize_t
-store_new_id(struct device_driver *driver, const char *buf, size_t count)
+static ssize_t store_new_id(struct device_driver *driver, const char *buf,
+			    size_t count)
 {
 	struct pci_driver *pdrv = to_pci_driver(driver);
 	const struct pci_device_id *ids = pdrv->id_table;
-	__u32 vendor, device, subvendor=PCI_ANY_ID,
-		subdevice=PCI_ANY_ID, class=0, class_mask=0;
-	unsigned long driver_data=0;
-	int fields=0;
+	__u32 vendor, device, subvendor = PCI_ANY_ID,
+		subdevice = PCI_ANY_ID, class = 0, class_mask = 0;
+	unsigned long driver_data = 0;
+	int fields = 0;
 	int retval = 0;
 
 	fields = sscanf(buf, "%x %x %x %x %x %x %lx",
@@ -166,8 +167,8 @@ static DRIVER_ATTR(new_id, S_IWUSR, NULL, store_new_id);
  *
  * Removes a dynamic pci device ID to this driver.
  */
-static ssize_t
-store_remove_id(struct device_driver *driver, const char *buf, size_t count)
+static ssize_t store_remove_id(struct device_driver *driver, const char *buf,
+			       size_t count)
 {
 	struct pci_dynid *dynid, *n;
 	struct pci_driver *pdrv = to_pci_driver(driver);
@@ -235,6 +236,7 @@ const struct pci_device_id *pci_match_id(const struct pci_device_id *ids,
 	}
 	return NULL;
 }
+EXPORT_SYMBOL(pci_match_id);
 
 static const struct pci_device_id pci_device_id_any = {
 	.vendor = PCI_ANY_ID,
@@ -372,8 +374,7 @@ static int pci_call_probe(struct pci_driver *drv, struct pci_dev *dev,
  * returns 0 on success, else error.
  * side-effect: pci_dev->driver is set to drv when drv claims pci_dev.
  */
-static int
-__pci_device_probe(struct pci_driver *drv, struct pci_dev *pci_dev)
+static int __pci_device_probe(struct pci_driver *drv, struct pci_dev *pci_dev)
 {
 	const struct pci_device_id *id;
 	int error = 0;
@@ -390,7 +391,7 @@ __pci_device_probe(struct pci_driver *drv, struct pci_dev *pci_dev)
 	return error;
 }
 
-static int pci_device_probe(struct device * dev)
+static int pci_device_probe(struct device *dev)
 {
 	int error = 0;
 	struct pci_driver *drv;
@@ -406,10 +407,10 @@ static int pci_device_probe(struct device * dev)
 	return error;
 }
 
-static int pci_device_remove(struct device * dev)
+static int pci_device_remove(struct device *dev)
 {
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
 
 	if (drv) {
 		if (drv->remove) {
@@ -537,8 +538,8 @@ static int pci_pm_reenable_device(struct pci_dev *pci_dev)
 
 static int pci_legacy_suspend(struct device *dev, pm_message_t state)
 {
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
 
 	if (drv && drv->suspend) {
 		pci_power_t prev = pci_dev->current_state;
@@ -564,8 +565,8 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state)
 
 static int pci_legacy_suspend_late(struct device *dev, pm_message_t state)
 {
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
 
 	if (drv && drv->suspend_late) {
 		pci_power_t prev = pci_dev->current_state;
@@ -595,8 +596,8 @@ static int pci_legacy_suspend_late(struct device *dev, pm_message_t state)
 
 static int pci_legacy_resume_early(struct device *dev)
 {
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
 
 	return drv && drv->resume_early ?
 			drv->resume_early(pci_dev) : 0;
@@ -604,8 +605,8 @@ static int pci_legacy_resume_early(struct device *dev)
 
 static int pci_legacy_resume(struct device *dev)
 {
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
 
 	pci_fixup_device(pci_fixup_resume, pci_dev);
 
@@ -1255,6 +1256,7 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner,
 	/* register with core */
 	return driver_register(&drv->driver);
 }
+EXPORT_SYMBOL(__pci_register_driver);
 
 /**
  * pci_unregister_driver - unregister a pci driver
@@ -1266,12 +1268,12 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner,
  * driverless.
  */
 
-void
-pci_unregister_driver(struct pci_driver *drv)
+void pci_unregister_driver(struct pci_driver *drv)
 {
 	driver_unregister(&drv->driver);
 	pci_free_dynids(drv);
 }
+EXPORT_SYMBOL(pci_unregister_driver);
 
 static struct pci_driver pci_compat_driver = {
 	.name = "compat"
@@ -1284,19 +1286,19 @@ static struct pci_driver pci_compat_driver = {
  * Returns the appropriate pci_driver structure or %NULL if there is no
  * registered driver for the device.
  */
-struct pci_driver *
-pci_dev_driver(const struct pci_dev *dev)
+struct pci_driver *pci_dev_driver(const struct pci_dev *dev)
 {
 	if (dev->driver)
 		return dev->driver;
 	else {
 		int i;
-		for(i=0; i<=PCI_ROM_RESOURCE; i++)
+		for (i = 0; i <= PCI_ROM_RESOURCE; i++)
 			if (dev->resource[i].flags & IORESOURCE_BUSY)
 				return &pci_compat_driver;
 	}
 	return NULL;
 }
+EXPORT_SYMBOL(pci_dev_driver);
 
 /**
  * pci_bus_match - Tell if a PCI device structure has a matching PCI device id structure
@@ -1342,6 +1344,7 @@ struct pci_dev *pci_dev_get(struct pci_dev *dev)
 		get_device(&dev->dev);
 	return dev;
 }
+EXPORT_SYMBOL(pci_dev_get);
 
 /**
  * pci_dev_put - release a use of the pci device structure
@@ -1355,6 +1358,7 @@ void pci_dev_put(struct pci_dev *dev)
 	if (dev)
 		put_device(&dev->dev);
 }
+EXPORT_SYMBOL(pci_dev_put);
 
 static int pci_uevent(struct device *dev, struct kobj_uevent_env *env)
 {
@@ -1400,19 +1404,10 @@ struct bus_type pci_bus_type = {
 	.drv_groups	= pci_drv_groups,
 	.pm		= PCI_PM_OPS_PTR,
 };
+EXPORT_SYMBOL(pci_bus_type);
 
 static int __init pci_driver_init(void)
 {
 	return bus_register(&pci_bus_type);
 }
-
 postcore_initcall(pci_driver_init);
-
-EXPORT_SYMBOL_GPL(pci_add_dynid);
-EXPORT_SYMBOL(pci_match_id);
-EXPORT_SYMBOL(__pci_register_driver);
-EXPORT_SYMBOL(pci_unregister_driver);
-EXPORT_SYMBOL(pci_dev_driver);
-EXPORT_SYMBOL(pci_bus_type);
-EXPORT_SYMBOL(pci_dev_get);
-EXPORT_SYMBOL(pci_dev_put);
diff --git a/drivers/pci/pci-label.c b/drivers/pci/pci-label.c
index 45113daaa778..a3fbe2012ea3 100644
--- a/drivers/pci/pci-label.c
+++ b/drivers/pci/pci-label.c
@@ -40,9 +40,8 @@ enum smbios_attr_enum {
 	SMBIOS_ATTR_INSTANCE_SHOW,
 };
 
-static size_t
-find_smbios_instance_string(struct pci_dev *pdev, char *buf,
-			    enum smbios_attr_enum attribute)
+static size_t find_smbios_instance_string(struct pci_dev *pdev, char *buf,
+					  enum smbios_attr_enum attribute)
 {
 	const struct dmi_device *dmi;
 	struct dmi_dev_onboard *donboard;
@@ -74,9 +73,8 @@ find_smbios_instance_string(struct pci_dev *pdev, char *buf,
 	return 0;
 }
 
-static umode_t
-smbios_instance_string_exist(struct kobject *kobj, struct attribute *attr,
-			     int n)
+static umode_t smbios_instance_string_exist(struct kobject *kobj,
+					    struct attribute *attr, int n)
 {
 	struct device *dev;
 	struct pci_dev *pdev;
@@ -88,8 +86,8 @@ smbios_instance_string_exist(struct kobject *kobj, struct attribute *attr,
 					   S_IRUGO : 0;
 }
 
-static ssize_t
-smbioslabel_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t smbioslabel_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev;
 	pdev = to_pci_dev(dev);
@@ -98,9 +96,8 @@ smbioslabel_show(struct device *dev, struct device_attribute *attr, char *buf)
 					   SMBIOS_ATTR_LABEL_SHOW);
 }
 
-static ssize_t
-smbiosinstance_show(struct device *dev,
-		    struct device_attribute *attr, char *buf)
+static ssize_t smbiosinstance_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev;
 	pdev = to_pci_dev(dev);
@@ -130,26 +127,22 @@ static struct attribute_group smbios_attr_group = {
 	.is_visible = smbios_instance_string_exist,
 };
 
-static int
-pci_create_smbiosname_file(struct pci_dev *pdev)
+static int pci_create_smbiosname_file(struct pci_dev *pdev)
 {
 	return sysfs_create_group(&pdev->dev.kobj, &smbios_attr_group);
 }
 
-static void
-pci_remove_smbiosname_file(struct pci_dev *pdev)
+static void pci_remove_smbiosname_file(struct pci_dev *pdev)
 {
 	sysfs_remove_group(&pdev->dev.kobj, &smbios_attr_group);
 }
 #else
-static inline int
-pci_create_smbiosname_file(struct pci_dev *pdev)
+static inline int pci_create_smbiosname_file(struct pci_dev *pdev)
 {
 	return -1;
 }
 
-static inline void
-pci_remove_smbiosname_file(struct pci_dev *pdev)
+static inline void pci_remove_smbiosname_file(struct pci_dev *pdev)
 {
 }
 #endif
@@ -175,8 +168,8 @@ static void dsm_label_utf16s_to_utf8s(union acpi_object *obj, char *buf)
 	buf[len] = '\n';
 }
 
-static int
-dsm_get_label(struct device *dev, char *buf, enum acpi_attr_enum attr)
+static int dsm_get_label(struct device *dev, char *buf,
+			 enum acpi_attr_enum attr)
 {
 	acpi_handle handle;
 	union acpi_object *obj, *tmp;
@@ -212,8 +205,7 @@ dsm_get_label(struct device *dev, char *buf, enum acpi_attr_enum attr)
 	return len;
 }
 
-static bool
-device_has_dsm(struct device *dev)
+static bool device_has_dsm(struct device *dev)
 {
 	acpi_handle handle;
 
@@ -225,8 +217,8 @@ device_has_dsm(struct device *dev)
 				1 << DEVICE_LABEL_DSM);
 }
 
-static umode_t
-acpi_index_string_exist(struct kobject *kobj, struct attribute *attr, int n)
+static umode_t acpi_index_string_exist(struct kobject *kobj,
+				       struct attribute *attr, int n)
 {
 	struct device *dev;
 
@@ -238,14 +230,14 @@ acpi_index_string_exist(struct kobject *kobj, struct attribute *attr, int n)
 	return 0;
 }
 
-static ssize_t
-acpilabel_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t acpilabel_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
 {
 	return dsm_get_label(dev, buf, ACPI_ATTR_LABEL_SHOW);
 }
 
-static ssize_t
-acpiindex_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t acpiindex_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
 {
 	return dsm_get_label(dev, buf, ACPI_ATTR_INDEX_SHOW);
 }
@@ -271,33 +263,28 @@ static struct attribute_group acpi_attr_group = {
 	.is_visible = acpi_index_string_exist,
 };
 
-static int
-pci_create_acpi_index_label_files(struct pci_dev *pdev)
+static int pci_create_acpi_index_label_files(struct pci_dev *pdev)
 {
 	return sysfs_create_group(&pdev->dev.kobj, &acpi_attr_group);
 }
 
-static int
-pci_remove_acpi_index_label_files(struct pci_dev *pdev)
+static int pci_remove_acpi_index_label_files(struct pci_dev *pdev)
 {
 	sysfs_remove_group(&pdev->dev.kobj, &acpi_attr_group);
 	return 0;
 }
 #else
-static inline int
-pci_create_acpi_index_label_files(struct pci_dev *pdev)
+static inline int pci_create_acpi_index_label_files(struct pci_dev *pdev)
 {
 	return -1;
 }
 
-static inline int
-pci_remove_acpi_index_label_files(struct pci_dev *pdev)
+static inline int pci_remove_acpi_index_label_files(struct pci_dev *pdev)
 {
 	return -1;
 }
 
-static inline bool
-device_has_dsm(struct device *dev)
+static inline bool device_has_dsm(struct device *dev)
 {
 	return false;
 }
diff --git a/drivers/pci/pci-stub.c b/drivers/pci/pci-stub.c
index 2ff77509d8e5..886fb3570278 100644
--- a/drivers/pci/pci-stub.c
+++ b/drivers/pci/pci-stub.c
@@ -55,7 +55,7 @@ static int __init pci_stub_init(void)
 	p = ids;
 	while ((id = strsep(&p, ","))) {
 		unsigned int vendor, device, subvendor = PCI_ANY_ID,
-			subdevice = PCI_ANY_ID, class=0, class_mask=0;
+			subdevice = PCI_ANY_ID, class = 0, class_mask = 0;
 		int fields;
 
 		if (!strlen(id))
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 84c350994b06..9ff0a901ecf7 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -41,8 +41,8 @@ field##_show(struct device *dev, struct device_attribute *attr, char *buf)				\
 {									\
 	struct pci_dev *pdev;						\
 									\
-	pdev = to_pci_dev (dev);					\
-	return sprintf (buf, format_string, pdev->field);		\
+	pdev = to_pci_dev(dev);						\
+	return sprintf(buf, format_string, pdev->field);		\
 }									\
 static DEVICE_ATTR_RO(field)
 
@@ -58,7 +58,7 @@ static ssize_t broken_parity_status_show(struct device *dev,
 					 char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
-	return sprintf (buf, "%u\n", pdev->broken_parity_status);
+	return sprintf(buf, "%u\n", pdev->broken_parity_status);
 }
 
 static ssize_t broken_parity_status_store(struct device *dev,
@@ -77,10 +77,8 @@ static ssize_t broken_parity_status_store(struct device *dev,
 }
 static DEVICE_ATTR_RW(broken_parity_status);
 
-static ssize_t pci_dev_show_local_cpu(struct device *dev,
-		int type,
-		struct device_attribute *attr,
-		char *buf)
+static ssize_t pci_dev_show_local_cpu(struct device *dev, int type,
+				      struct device_attribute *attr, char *buf)
 {
 	const struct cpumask *mask;
 	int len;
@@ -101,14 +99,14 @@ static ssize_t pci_dev_show_local_cpu(struct device *dev,
 }
 
 static ssize_t local_cpus_show(struct device *dev,
-			struct device_attribute *attr, char *buf)
+			       struct device_attribute *attr, char *buf)
 {
 	return pci_dev_show_local_cpu(dev, 1, attr, buf);
 }
 static DEVICE_ATTR_RO(local_cpus);
 
 static ssize_t local_cpulist_show(struct device *dev,
-			struct device_attribute *attr, char *buf)
+				  struct device_attribute *attr, char *buf)
 {
 	return pci_dev_show_local_cpu(dev, 0, attr, buf);
 }
@@ -117,8 +115,7 @@ static DEVICE_ATTR_RO(local_cpulist);
 /*
  * PCI Bus Class Devices
  */
-static ssize_t pci_bus_show_cpuaffinity(struct device *dev,
-					int type,
+static ssize_t pci_bus_show_cpuaffinity(struct device *dev, int type,
 					struct device_attribute *attr,
 					char *buf)
 {
@@ -149,11 +146,11 @@ static ssize_t cpulistaffinity_show(struct device *dev,
 static DEVICE_ATTR_RO(cpulistaffinity);
 
 /* show resources */
-static ssize_t
-resource_show(struct device * dev, struct device_attribute *attr, char * buf)
+static ssize_t resource_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
 {
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	char * str = buf;
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	char *str = buf;
 	int i;
 	int max;
 	resource_size_t start, end;
@@ -166,7 +163,7 @@ resource_show(struct device * dev, struct device_attribute *attr, char * buf)
 	for (i = 0; i < max; i++) {
 		struct resource *res =  &pci_dev->resource[i];
 		pci_resource_to_user(pci_dev, i, res, &start, &end);
-		str += sprintf(str,"0x%016llx 0x%016llx 0x%016llx\n",
+		str += sprintf(str, "0x%016llx 0x%016llx 0x%016llx\n",
 			       (unsigned long long)start,
 			       (unsigned long long)end,
 			       (unsigned long long)res->flags);
@@ -175,7 +172,8 @@ resource_show(struct device * dev, struct device_attribute *attr, char * buf)
 }
 static DEVICE_ATTR_RO(resource);
 
-static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
 
@@ -187,9 +185,8 @@ static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
 }
 static DEVICE_ATTR_RO(modalias);
 
-static ssize_t enabled_store(struct device *dev,
-			     struct device_attribute *attr, const char *buf,
-			     size_t count)
+static ssize_t enabled_store(struct device *dev, struct device_attribute *attr,
+			     const char *buf, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	unsigned long val;
@@ -213,57 +210,56 @@ static ssize_t enabled_store(struct device *dev,
 	return result < 0 ? result : count;
 }
 
-static ssize_t enabled_show(struct device *dev,
-			    struct device_attribute *attr, char *buf)
+static ssize_t enabled_show(struct device *dev, struct device_attribute *attr,
+			    char *buf)
 {
 	struct pci_dev *pdev;
 
-	pdev = to_pci_dev (dev);
-	return sprintf (buf, "%u\n", atomic_read(&pdev->enable_cnt));
+	pdev = to_pci_dev(dev);
+	return sprintf(buf, "%u\n", atomic_read(&pdev->enable_cnt));
 }
 static DEVICE_ATTR_RW(enabled);
 
 #ifdef CONFIG_NUMA
-static ssize_t
-numa_node_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t numa_node_show(struct device *dev, struct device_attribute *attr,
+			      char *buf)
 {
-	return sprintf (buf, "%d\n", dev->numa_node);
+	return sprintf(buf, "%d\n", dev->numa_node);
 }
 static DEVICE_ATTR_RO(numa_node);
 #endif
 
-static ssize_t
-dma_mask_bits_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t dma_mask_bits_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 
-	return sprintf (buf, "%d\n", fls64(pdev->dma_mask));
+	return sprintf(buf, "%d\n", fls64(pdev->dma_mask));
 }
 static DEVICE_ATTR_RO(dma_mask_bits);
 
-static ssize_t
-consistent_dma_mask_bits_show(struct device *dev, struct device_attribute *attr,
-				 char *buf)
+static ssize_t consistent_dma_mask_bits_show(struct device *dev,
+					     struct device_attribute *attr,
+					     char *buf)
 {
-	return sprintf (buf, "%d\n", fls64(dev->coherent_dma_mask));
+	return sprintf(buf, "%d\n", fls64(dev->coherent_dma_mask));
 }
 static DEVICE_ATTR_RO(consistent_dma_mask_bits);
 
-static ssize_t
-msi_bus_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t msi_bus_show(struct device *dev, struct device_attribute *attr,
+			    char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 
 	if (!pdev->subordinate)
 		return 0;
 
-	return sprintf (buf, "%u\n",
-			!(pdev->subordinate->bus_flags & PCI_BUS_FLAGS_NO_MSI));
+	return sprintf(buf, "%u\n",
+		       !(pdev->subordinate->bus_flags & PCI_BUS_FLAGS_NO_MSI));
 }
 
-static ssize_t
-msi_bus_store(struct device *dev, struct device_attribute *attr,
-	      const char *buf, size_t count)
+static ssize_t msi_bus_store(struct device *dev, struct device_attribute *attr,
+			     const char *buf, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	unsigned long val;
@@ -290,8 +286,8 @@ msi_bus_store(struct device *dev, struct device_attribute *attr,
 	    !!val) {
 		pdev->subordinate->bus_flags ^= PCI_BUS_FLAGS_NO_MSI;
 
-		dev_warn(&pdev->dev, "forced subordinate bus to%s support MSI,"
-			 " bad things could happen\n", val ? "" : " not");
+		dev_warn(&pdev->dev, "forced subordinate bus to%s support MSI, bad things could happen\n",
+			 val ? "" : " not");
 	}
 
 	return count;
@@ -331,9 +327,9 @@ const struct attribute_group *pci_bus_groups[] = {
 	NULL,
 };
 
-static ssize_t
-dev_rescan_store(struct device *dev, struct device_attribute *attr,
-		 const char *buf, size_t count)
+static ssize_t dev_rescan_store(struct device *dev,
+				struct device_attribute *attr, const char *buf,
+				size_t count)
 {
 	unsigned long val;
 	struct pci_dev *pdev = to_pci_dev(dev);
@@ -352,9 +348,8 @@ static struct device_attribute dev_rescan_attr = __ATTR(rescan,
 							(S_IWUSR|S_IWGRP),
 							NULL, dev_rescan_store);
 
-static ssize_t
-remove_store(struct device *dev, struct device_attribute *attr,
-	     const char *buf, size_t count)
+static ssize_t remove_store(struct device *dev, struct device_attribute *attr,
+			    const char *buf, size_t count)
 {
 	unsigned long val;
 
@@ -369,9 +364,9 @@ static struct device_attribute dev_remove_attr = __ATTR(remove,
 							(S_IWUSR|S_IWGRP),
 							NULL, remove_store);
 
-static ssize_t
-dev_bus_rescan_store(struct device *dev, struct device_attribute *attr,
-		 const char *buf, size_t count)
+static ssize_t dev_bus_rescan_store(struct device *dev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t count)
 {
 	unsigned long val;
 	struct pci_bus *bus = to_pci_bus(dev);
@@ -412,7 +407,7 @@ static ssize_t d3cold_allowed_show(struct device *dev,
 				   struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
-	return sprintf (buf, "%u\n", pdev->d3cold_allowed);
+	return sprintf(buf, "%u\n", pdev->d3cold_allowed);
 }
 static DEVICE_ATTR_RW(d3cold_allowed);
 #endif
@@ -607,8 +602,8 @@ const struct attribute_group *pcibus_groups[] = {
 	NULL,
 };
 
-static ssize_t
-boot_vga_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t boot_vga_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct pci_dev *vga_dev = vga_default_device();
@@ -622,22 +617,21 @@ boot_vga_show(struct device *dev, struct device_attribute *attr, char *buf)
 }
 static struct device_attribute vga_attr = __ATTR_RO(boot_vga);
 
-static ssize_t
-pci_read_config(struct file *filp, struct kobject *kobj,
-		struct bin_attribute *bin_attr,
-		char *buf, loff_t off, size_t count)
+static ssize_t pci_read_config(struct file *filp, struct kobject *kobj,
+			       struct bin_attribute *bin_attr, char *buf,
+			       loff_t off, size_t count)
 {
-	struct pci_dev *dev = to_pci_dev(container_of(kobj,struct device,kobj));
+	struct pci_dev *dev = to_pci_dev(container_of(kobj, struct device,
+						      kobj));
 	unsigned int size = 64;
 	loff_t init_off = off;
-	u8 *data = (u8*) buf;
+	u8 *data = (u8 *) buf;
 
 	/* Several chips lock up trying to read undefined config space */
-	if (security_capable(filp->f_cred, &init_user_ns, CAP_SYS_ADMIN) == 0) {
+	if (security_capable(filp->f_cred, &init_user_ns, CAP_SYS_ADMIN) == 0)
 		size = dev->cfg_size;
-	} else if (dev->hdr_type == PCI_HEADER_TYPE_CARDBUS) {
+	else if (dev->hdr_type == PCI_HEADER_TYPE_CARDBUS)
 		size = 128;
-	}
 
 	if (off > size)
 		return 0;
@@ -700,15 +694,15 @@ pci_read_config(struct file *filp, struct kobject *kobj,
 	return count;
 }
 
-static ssize_t
-pci_write_config(struct file* filp, struct kobject *kobj,
-		 struct bin_attribute *bin_attr,
-		 char *buf, loff_t off, size_t count)
+static ssize_t pci_write_config(struct file *filp, struct kobject *kobj,
+				struct bin_attribute *bin_attr, char *buf,
+				loff_t off, size_t count)
 {
-	struct pci_dev *dev = to_pci_dev(container_of(kobj,struct device,kobj));
+	struct pci_dev *dev = to_pci_dev(container_of(kobj, struct device,
+						      kobj));
 	unsigned int size = count;
 	loff_t init_off = off;
-	u8 *data = (u8*) buf;
+	u8 *data = (u8 *) buf;
 
 	if (off > dev->cfg_size)
 		return 0;
@@ -728,10 +722,10 @@ pci_write_config(struct file* filp, struct kobject *kobj,
 	if ((off & 3) && size > 2) {
 		u16 val = data[off - init_off];
 		val |= (u16) data[off - init_off + 1] << 8;
-                pci_user_write_config_word(dev, off, val);
-                off += 2;
-                size -= 2;
-        }
+		pci_user_write_config_word(dev, off, val);
+		off += 2;
+		size -= 2;
+	}
 
 	while (size > 3) {
 		u32 val = data[off - init_off];
@@ -762,10 +756,9 @@ pci_write_config(struct file* filp, struct kobject *kobj,
 	return count;
 }
 
-static ssize_t
-read_vpd_attr(struct file *filp, struct kobject *kobj,
-	      struct bin_attribute *bin_attr,
-	      char *buf, loff_t off, size_t count)
+static ssize_t read_vpd_attr(struct file *filp, struct kobject *kobj,
+			     struct bin_attribute *bin_attr, char *buf,
+			     loff_t off, size_t count)
 {
 	struct pci_dev *dev =
 		to_pci_dev(container_of(kobj, struct device, kobj));
@@ -778,10 +771,9 @@ read_vpd_attr(struct file *filp, struct kobject *kobj,
 	return pci_read_vpd(dev, off, count, buf);
 }
 
-static ssize_t
-write_vpd_attr(struct file *filp, struct kobject *kobj,
-	       struct bin_attribute *bin_attr,
-	       char *buf, loff_t off, size_t count)
+static ssize_t write_vpd_attr(struct file *filp, struct kobject *kobj,
+			      struct bin_attribute *bin_attr, char *buf,
+			      loff_t off, size_t count)
 {
 	struct pci_dev *dev =
 		to_pci_dev(container_of(kobj, struct device, kobj));
@@ -807,20 +799,18 @@ write_vpd_attr(struct file *filp, struct kobject *kobj,
  * Reads 1, 2, or 4 bytes from legacy I/O port space using an arch specific
  * callback routine (pci_legacy_read).
  */
-static ssize_t
-pci_read_legacy_io(struct file *filp, struct kobject *kobj,
-		   struct bin_attribute *bin_attr,
-		   char *buf, loff_t off, size_t count)
+static ssize_t pci_read_legacy_io(struct file *filp, struct kobject *kobj,
+				  struct bin_attribute *bin_attr, char *buf,
+				  loff_t off, size_t count)
 {
-        struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+	struct pci_bus *bus = to_pci_bus(container_of(kobj, struct device,
 						      kobj));
 
-        /* Only support 1, 2 or 4 byte accesses */
-        if (count != 1 && count != 2 && count != 4)
-                return -EINVAL;
+	/* Only support 1, 2 or 4 byte accesses */
+	if (count != 1 && count != 2 && count != 4)
+		return -EINVAL;
 
-        return pci_legacy_read(bus, off, (u32 *)buf, count);
+	return pci_legacy_read(bus, off, (u32 *)buf, count);
 }
 
 /**
@@ -835,19 +825,18 @@ pci_read_legacy_io(struct file *filp, struct kobject *kobj,
  * Writes 1, 2, or 4 bytes from legacy I/O port space using an arch specific
  * callback routine (pci_legacy_write).
  */
-static ssize_t
-pci_write_legacy_io(struct file *filp, struct kobject *kobj,
-		    struct bin_attribute *bin_attr,
-		    char *buf, loff_t off, size_t count)
+static ssize_t pci_write_legacy_io(struct file *filp, struct kobject *kobj,
+				   struct bin_attribute *bin_attr, char *buf,
+				   loff_t off, size_t count)
 {
-        struct pci_bus *bus = to_pci_bus(container_of(kobj,
-						      struct device,
+	struct pci_bus *bus = to_pci_bus(container_of(kobj, struct device,
 						      kobj));
-        /* Only support 1, 2 or 4 byte accesses */
-        if (count != 1 && count != 2 && count != 4)
-                return -EINVAL;
 
-        return pci_legacy_write(bus, off, *(u32 *)buf, count);
+	/* Only support 1, 2 or 4 byte accesses */
+	if (count != 1 && count != 2 && count != 4)
+		return -EINVAL;
+
+	return pci_legacy_write(bus, off, *(u32 *)buf, count);
 }
 
 /**
@@ -861,16 +850,14 @@ pci_write_legacy_io(struct file *filp, struct kobject *kobj,
  * legacy memory space (first meg of bus space) into application virtual
  * memory space.
  */
-static int
-pci_mmap_legacy_mem(struct file *filp, struct kobject *kobj,
-		    struct bin_attribute *attr,
-                    struct vm_area_struct *vma)
+static int pci_mmap_legacy_mem(struct file *filp, struct kobject *kobj,
+			       struct bin_attribute *attr,
+			       struct vm_area_struct *vma)
 {
-        struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+	struct pci_bus *bus = to_pci_bus(container_of(kobj, struct device,
 						      kobj));
 
-        return pci_mmap_legacy_page_range(bus, vma, pci_mmap_mem);
+	return pci_mmap_legacy_page_range(bus, vma, pci_mmap_mem);
 }
 
 /**
@@ -884,16 +871,14 @@ pci_mmap_legacy_mem(struct file *filp, struct kobject *kobj,
  * legacy IO space (first meg of bus space) into application virtual
  * memory space. Returns -ENOSYS if the operation isn't supported
  */
-static int
-pci_mmap_legacy_io(struct file *filp, struct kobject *kobj,
-		   struct bin_attribute *attr,
-		   struct vm_area_struct *vma)
+static int pci_mmap_legacy_io(struct file *filp, struct kobject *kobj,
+			      struct bin_attribute *attr,
+			      struct vm_area_struct *vma)
 {
-        struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+	struct pci_bus *bus = to_pci_bus(container_of(kobj, struct device,
 						      kobj));
 
-        return pci_mmap_legacy_page_range(bus, vma, pci_mmap_io);
+	return pci_mmap_legacy_page_range(bus, vma, pci_mmap_io);
 }
 
 /**
@@ -903,10 +888,9 @@ pci_mmap_legacy_io(struct file *filp, struct kobject *kobj,
  *
  * Stub implementation. Can be overridden by arch if necessary.
  */
-void __weak
-pci_adjust_legacy_attr(struct pci_bus *b, enum pci_mmap_state mmap_type)
+void __weak pci_adjust_legacy_attr(struct pci_bus *b,
+				   enum pci_mmap_state mmap_type)
 {
-	return;
 }
 
 /**
@@ -961,8 +945,7 @@ legacy_io_err:
 	kfree(b->legacy_io);
 	b->legacy_io = NULL;
 kzalloc_err:
-	printk(KERN_WARNING "pci: warning: could not create legacy I/O port "
-	       "and ISA memory resources to sysfs\n");
+	printk(KERN_WARNING "pci: warning: could not create legacy I/O port and ISA memory resources to sysfs\n");
 	return;
 }
 
@@ -1005,9 +988,8 @@ int pci_mmap_fits(struct pci_dev *pdev, int resno, struct vm_area_struct *vma,
  *
  * Use the regular PCI mapping routines to map a PCI resource into userspace.
  */
-static int
-pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
-		  struct vm_area_struct *vma, int write_combine)
+static int pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
+			     struct vm_area_struct *vma, int write_combine)
 {
 	struct pci_dev *pdev = to_pci_dev(container_of(kobj,
 						       struct device, kobj));
@@ -1023,8 +1005,7 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 		return -ENODEV;
 
 	if (!pci_mmap_fits(pdev, i, vma, PCI_MMAP_SYSFS)) {
-		WARN(1, "process \"%s\" tried to map 0x%08lx bytes "
-			"at page 0x%08lx on %s BAR %d (start 0x%16Lx, size 0x%16Lx)\n",
+		WARN(1, "process \"%s\" tried to map 0x%08lx bytes at page 0x%08lx on %s BAR %d (start 0x%16Lx, size 0x%16Lx)\n",
 			current->comm, vma->vm_end-vma->vm_start, vma->vm_pgoff,
 			pci_name(pdev), i,
 			(u64)pci_resource_start(pdev, i),
@@ -1046,26 +1027,23 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 	return pci_mmap_page_range(pdev, vma, mmap_type, write_combine);
 }
 
-static int
-pci_mmap_resource_uc(struct file *filp, struct kobject *kobj,
-		     struct bin_attribute *attr,
-		     struct vm_area_struct *vma)
+static int pci_mmap_resource_uc(struct file *filp, struct kobject *kobj,
+				struct bin_attribute *attr,
+				struct vm_area_struct *vma)
 {
 	return pci_mmap_resource(kobj, attr, vma, 0);
 }
 
-static int
-pci_mmap_resource_wc(struct file *filp, struct kobject *kobj,
-		     struct bin_attribute *attr,
-		     struct vm_area_struct *vma)
+static int pci_mmap_resource_wc(struct file *filp, struct kobject *kobj,
+				struct bin_attribute *attr,
+				struct vm_area_struct *vma)
 {
 	return pci_mmap_resource(kobj, attr, vma, 1);
 }
 
-static ssize_t
-pci_resource_io(struct file *filp, struct kobject *kobj,
-		struct bin_attribute *attr, char *buf,
-		loff_t off, size_t count, bool write)
+static ssize_t pci_resource_io(struct file *filp, struct kobject *kobj,
+			       struct bin_attribute *attr, char *buf,
+			       loff_t off, size_t count, bool write)
 {
 	struct pci_dev *pdev = to_pci_dev(container_of(kobj,
 						       struct device, kobj));
@@ -1110,18 +1088,16 @@ pci_resource_io(struct file *filp, struct kobject *kobj,
 	return -EINVAL;
 }
 
-static ssize_t
-pci_read_resource_io(struct file *filp, struct kobject *kobj,
-		     struct bin_attribute *attr, char *buf,
-		     loff_t off, size_t count)
+static ssize_t pci_read_resource_io(struct file *filp, struct kobject *kobj,
+				    struct bin_attribute *attr, char *buf,
+				    loff_t off, size_t count)
 {
 	return pci_resource_io(filp, kobj, attr, buf, off, count, false);
 }
 
-static ssize_t
-pci_write_resource_io(struct file *filp, struct kobject *kobj,
-		      struct bin_attribute *attr, char *buf,
-		      loff_t off, size_t count)
+static ssize_t pci_write_resource_io(struct file *filp, struct kobject *kobj,
+				     struct bin_attribute *attr, char *buf,
+				     loff_t off, size_t count)
 {
 	return pci_resource_io(filp, kobj, attr, buf, off, count, true);
 }
@@ -1133,8 +1109,7 @@ pci_write_resource_io(struct file *filp, struct kobject *kobj,
  * If we created resource files for @pdev, remove them from sysfs and
  * free their resources.
  */
-static void
-pci_remove_resource_files(struct pci_dev *pdev)
+static void pci_remove_resource_files(struct pci_dev *pdev)
 {
 	int i;
 
@@ -1237,10 +1212,9 @@ void __weak pci_remove_resource_files(struct pci_dev *dev) { return; }
  *
  * writing anything except 0 enables it
  */
-static ssize_t
-pci_write_rom(struct file *filp, struct kobject *kobj,
-	      struct bin_attribute *bin_attr,
-	      char *buf, loff_t off, size_t count)
+static ssize_t pci_write_rom(struct file *filp, struct kobject *kobj,
+			     struct bin_attribute *bin_attr, char *buf,
+			     loff_t off, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(container_of(kobj, struct device, kobj));
 
@@ -1264,10 +1238,9 @@ pci_write_rom(struct file *filp, struct kobject *kobj,
  * Put @count bytes starting at @off into @buf from the ROM in the PCI
  * device corresponding to @kobj.
  */
-static ssize_t
-pci_read_rom(struct file *filp, struct kobject *kobj,
-	     struct bin_attribute *bin_attr,
-	     char *buf, loff_t off, size_t count)
+static ssize_t pci_read_rom(struct file *filp, struct kobject *kobj,
+			    struct bin_attribute *bin_attr, char *buf,
+			    loff_t off, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(container_of(kobj, struct device, kobj));
 	void __iomem *rom;
@@ -1313,9 +1286,8 @@ static struct bin_attribute pcie_config_attr = {
 	.write = pci_write_config,
 };
 
-static ssize_t reset_store(struct device *dev,
-			   struct device_attribute *attr, const char *buf,
-			   size_t count)
+static ssize_t reset_store(struct device *dev, struct device_attribute *attr,
+			   const char *buf, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	unsigned long val;
@@ -1382,7 +1354,7 @@ error:
 	return retval;
 }
 
-int __must_check pci_create_sysfs_dev_files (struct pci_dev *pdev)
+int __must_check pci_create_sysfs_dev_files(struct pci_dev *pdev)
 {
 	int retval;
 	int rom_size = 0;
@@ -1520,7 +1492,6 @@ static int __init pci_sysfs_init(void)
 
 	return 0;
 }
-
 late_initcall(pci_sysfs_init);
 
 static struct attribute *pci_dev_dev_attrs[] = {
@@ -1529,7 +1500,7 @@ static struct attribute *pci_dev_dev_attrs[] = {
 };
 
 static umode_t pci_dev_attrs_are_visible(struct kobject *kobj,
-						struct attribute *a, int n)
+					 struct attribute *a, int n)
 {
 	struct device *dev = container_of(kobj, struct device, kobj);
 	struct pci_dev *pdev = to_pci_dev(dev);
@@ -1548,7 +1519,7 @@ static struct attribute *pci_dev_hp_attrs[] = {
 };
 
 static umode_t pci_dev_hp_attrs_are_visible(struct kobject *kobj,
-						struct attribute *a, int n)
+					    struct attribute *a, int n)
 {
 	struct device *dev = container_of(kobj, struct device, kobj);
 	struct pci_dev *pdev = to_pci_dev(dev);
@@ -1572,7 +1543,7 @@ static struct attribute *sriov_dev_attrs[] = {
 };
 
 static umode_t sriov_attrs_are_visible(struct kobject *kobj,
-					 struct attribute *a, int n)
+				       struct attribute *a, int n)
 {
 	struct device *dev = container_of(kobj, struct device, kobj);
 
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 436a76ab4bb1..1c8592b0e146 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -114,7 +114,7 @@ unsigned char pci_bus_max_busnr(struct pci_bus *bus)
 	max = bus->busn_res.end;
 	list_for_each_entry(tmp, &bus->children, node) {
 		n = pci_bus_max_busnr(tmp);
-		if(n > max)
+		if (n > max)
 			max = n;
 	}
 	return max;
@@ -226,6 +226,7 @@ int pci_find_capability(struct pci_dev *dev, int cap)
 
 	return pos;
 }
+EXPORT_SYMBOL(pci_find_capability);
 
 /**
  * pci_bus_find_capability - query for devices' capabilities
@@ -253,6 +254,7 @@ int pci_bus_find_capability(struct pci_bus *bus, unsigned int devfn, int cap)
 
 	return pos;
 }
+EXPORT_SYMBOL(pci_bus_find_capability);
 
 /**
  * pci_find_next_ext_capability - Find an extended capability
@@ -403,8 +405,8 @@ EXPORT_SYMBOL_GPL(pci_find_ht_capability);
  *  For given resource region of given device, return the resource
  *  region of parent bus the given region is contained in.
  */
-struct resource *
-pci_find_parent_resource(const struct pci_dev *dev, struct resource *res)
+struct resource *pci_find_parent_resource(const struct pci_dev *dev,
+					  struct resource *res)
 {
 	const struct pci_bus *bus = dev->bus;
 	struct resource *r;
@@ -436,6 +438,7 @@ pci_find_parent_resource(const struct pci_dev *dev, struct resource *res)
 	}
 	return NULL;
 }
+EXPORT_SYMBOL(pci_find_parent_resource);
 
 /**
  * pci_wait_for_pending - wait for @mask bit(s) to clear in status word @pos
@@ -470,8 +473,7 @@ int pci_wait_for_pending(struct pci_dev *dev, int pos, u16 mask)
  * Restore the BAR values for a given device, so as to make it
  * accessible by its driver.
  */
-static void
-pci_restore_bars(struct pci_dev *dev)
+static void pci_restore_bars(struct pci_dev *dev)
 {
 	int i;
 
@@ -496,7 +498,7 @@ static inline bool platform_pci_power_manageable(struct pci_dev *dev)
 }
 
 static inline int platform_pci_set_power_state(struct pci_dev *dev,
-                                                pci_power_t t)
+					       pci_power_t t)
 {
 	return pci_platform_pm ? pci_platform_pm->set_state(dev, t) : -ENOSYS;
 }
@@ -553,8 +555,8 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 	 */
 	if (state != PCI_D0 && dev->current_state <= PCI_D3cold
 	    && dev->current_state > state) {
-		dev_err(&dev->dev, "invalid power transition "
-			"(from state %d to %d)\n", dev->current_state, state);
+		dev_err(&dev->dev, "invalid power transition (from state %d to %d)\n",
+			dev->current_state, state);
 		return -EINVAL;
 	}
 
@@ -601,8 +603,8 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
 	dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
 	if (dev->current_state != state && printk_ratelimit())
-		dev_info(&dev->dev, "Refused to change power state, "
-			"currently in D%d\n", dev->current_state);
+		dev_info(&dev->dev, "Refused to change power state, currently in D%d\n",
+			 dev->current_state);
 
 	/*
 	 * According to section 5.4.1 of the "PCI BUS POWER MANAGEMENT
@@ -846,6 +848,7 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 
 	return error;
 }
+EXPORT_SYMBOL(pci_set_power_state);
 
 /**
  * pci_choose_state - Choose the power state of a PCI device
@@ -884,12 +887,10 @@ pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state)
 	}
 	return PCI_D0;
 }
-
 EXPORT_SYMBOL(pci_choose_state);
 
 #define PCI_EXP_SAVE_REGS	7
 
-
 static struct pci_cap_saved_state *_pci_find_saved_cap(struct pci_dev *pci_dev,
 						       u16 cap, bool extended)
 {
@@ -1001,8 +1002,7 @@ static void pci_restore_pcix_state(struct pci_dev *dev)
  * pci_save_state - save the PCI configuration space of a device before suspending
  * @dev: - PCI device that we're dealing with
  */
-int
-pci_save_state(struct pci_dev *dev)
+int pci_save_state(struct pci_dev *dev)
 {
 	int i;
 	/* XXX: 100% dword access ok here? */
@@ -1017,6 +1017,7 @@ pci_save_state(struct pci_dev *dev)
 		return i;
 	return 0;
 }
+EXPORT_SYMBOL(pci_save_state);
 
 static void pci_restore_config_dword(struct pci_dev *pdev, int offset,
 				     u32 saved_val, int retry)
@@ -1028,8 +1029,8 @@ static void pci_restore_config_dword(struct pci_dev *pdev, int offset,
 		return;
 
 	for (;;) {
-		dev_dbg(&pdev->dev, "restoring config space at offset "
-			"%#x (was %#x, writing %#x)\n", offset, val, saved_val);
+		dev_dbg(&pdev->dev, "restoring config space at offset %#x (was %#x, writing %#x)\n",
+			offset, val, saved_val);
 		pci_write_config_dword(pdev, offset, saved_val);
 		if (retry-- <= 0)
 			return;
@@ -1087,6 +1088,7 @@ void pci_restore_state(struct pci_dev *dev)
 
 	dev->state_saved = false;
 }
+EXPORT_SYMBOL(pci_restore_state);
 
 struct pci_saved_state {
 	u32 config_space[16];
@@ -1231,6 +1233,7 @@ int pci_reenable_device(struct pci_dev *dev)
 		return do_pci_enable_device(dev, (1 << PCI_NUM_RESOURCES) - 1);
 	return 0;
 }
+EXPORT_SYMBOL(pci_reenable_device);
 
 static void pci_enable_bridge(struct pci_dev *dev)
 {
@@ -1305,6 +1308,7 @@ int pci_enable_device_io(struct pci_dev *dev)
 {
 	return pci_enable_device_flags(dev, IORESOURCE_IO);
 }
+EXPORT_SYMBOL(pci_enable_device_io);
 
 /**
  * pci_enable_device_mem - Initialize a device for use with Memory space
@@ -1318,6 +1322,7 @@ int pci_enable_device_mem(struct pci_dev *dev)
 {
 	return pci_enable_device_flags(dev, IORESOURCE_MEM);
 }
+EXPORT_SYMBOL(pci_enable_device_mem);
 
 /**
  * pci_enable_device - Initialize device before it's used by a driver.
@@ -1334,6 +1339,7 @@ int pci_enable_device(struct pci_dev *dev)
 {
 	return pci_enable_device_flags(dev, IORESOURCE_MEM | IORESOURCE_IO);
 }
+EXPORT_SYMBOL(pci_enable_device);
 
 /*
  * Managed PCI resources.  This manages device on/off, intx/msi/msix
@@ -1416,6 +1422,7 @@ int pcim_enable_device(struct pci_dev *pdev)
 	}
 	return rc;
 }
+EXPORT_SYMBOL(pcim_enable_device);
 
 /**
  * pcim_pin_device - Pin managed PCI device
@@ -1434,6 +1441,7 @@ void pcim_pin_device(struct pci_dev *pdev)
 	if (dr)
 		dr->pinned = 1;
 }
+EXPORT_SYMBOL(pcim_pin_device);
 
 /*
  * pcibios_add_device - provide arch specific hooks when adding device dev
@@ -1443,7 +1451,7 @@ void pcim_pin_device(struct pci_dev *pdev)
  * devices are added. This is the default implementation. Architecture
  * implementations can override this.
  */
-int __weak pcibios_add_device (struct pci_dev *dev)
+int __weak pcibios_add_device(struct pci_dev *dev)
 {
 	return 0;
 }
@@ -1515,8 +1523,7 @@ void pci_disable_enabled_device(struct pci_dev *dev)
  * Note we don't actually disable the device until all callers of
  * pci_enable_device() have called pci_disable_device().
  */
-void
-pci_disable_device(struct pci_dev *dev)
+void pci_disable_device(struct pci_dev *dev)
 {
 	struct pci_devres *dr;
 
@@ -1534,6 +1541,7 @@ pci_disable_device(struct pci_dev *dev)
 
 	dev->is_busmaster = 0;
 }
+EXPORT_SYMBOL(pci_disable_device);
 
 /**
  * pcibios_set_pcie_reset_state - set reset state for device dev
@@ -1562,6 +1570,7 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state)
 {
 	return pcibios_set_pcie_reset_state(dev, state);
 }
+EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);
 
 /**
  * pci_check_pme_status - Check if given device has generated PME.
@@ -1641,6 +1650,7 @@ bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
 
 	return !!(dev->pme_support & (1 << state));
 }
+EXPORT_SYMBOL(pci_pme_capable);
 
 static void pci_pme_list_scan(struct work_struct *work)
 {
@@ -1745,6 +1755,7 @@ void pci_pme_active(struct pci_dev *dev, bool enable)
 
 	dev_dbg(&dev->dev, "PME# %s\n", enable ? "enabled" : "disabled");
 }
+EXPORT_SYMBOL(pci_pme_active);
 
 /**
  * __pci_enable_wake - enable PCI device as wakeup event source
@@ -1830,6 +1841,7 @@ int pci_wake_from_d3(struct pci_dev *dev, bool enable)
 			pci_enable_wake(dev, PCI_D3cold, enable) :
 			pci_enable_wake(dev, PCI_D3hot, enable);
 }
+EXPORT_SYMBOL(pci_wake_from_d3);
 
 /**
  * pci_target_state - find an appropriate low power state for a given PCI dev
@@ -1908,6 +1920,7 @@ int pci_prepare_to_sleep(struct pci_dev *dev)
 
 	return error;
 }
+EXPORT_SYMBOL(pci_prepare_to_sleep);
 
 /**
  * pci_back_from_sleep - turn PCI device on during system-wide transition into working state
@@ -1920,6 +1933,7 @@ int pci_back_from_sleep(struct pci_dev *dev)
 	pci_enable_wake(dev, PCI_D0, false);
 	return pci_set_power_state(dev, PCI_D0);
 }
+EXPORT_SYMBOL(pci_back_from_sleep);
 
 /**
  * pci_finish_runtime_suspend - Carry out PCI-specific part of runtime suspend.
@@ -2415,8 +2429,7 @@ u8 pci_swizzle_interrupt_pin(const struct pci_dev *dev, u8 pin)
 	return (((pin - 1) + slot) % 4) + 1;
 }
 
-int
-pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
+int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
 {
 	u8 pin;
 
@@ -2478,6 +2491,7 @@ void pci_release_region(struct pci_dev *pdev, int bar)
 	if (dr)
 		dr->region_mask &= ~(1 << bar);
 }
+EXPORT_SYMBOL(pci_release_region);
 
 /**
  *	__pci_request_region - Reserved PCI I/O and memory resource
@@ -2498,8 +2512,8 @@ void pci_release_region(struct pci_dev *pdev, int bar)
  *	Returns 0 on success, or %EBUSY on error.  A warning
  *	message is also printed on failure.
  */
-static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name,
-									int exclusive)
+static int __pci_request_region(struct pci_dev *pdev, int bar,
+				const char *res_name, int exclusive)
 {
 	struct pci_devres *dr;
 
@@ -2510,8 +2524,7 @@ static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_n
 		if (!request_region(pci_resource_start(pdev, bar),
 			    pci_resource_len(pdev, bar), res_name))
 			goto err_out;
-	}
-	else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
+	} else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
 		if (!__request_mem_region(pci_resource_start(pdev, bar),
 					pci_resource_len(pdev, bar), res_name,
 					exclusive))
@@ -2548,6 +2561,7 @@ int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
 {
 	return __pci_request_region(pdev, bar, res_name, 0);
 }
+EXPORT_SYMBOL(pci_request_region);
 
 /**
  *	pci_request_region_exclusive - Reserved PCI I/O and memory resource
@@ -2567,10 +2581,13 @@ int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
  *	explicitly not allowed to map the resource via /dev/mem or
  *	sysfs.
  */
-int pci_request_region_exclusive(struct pci_dev *pdev, int bar, const char *res_name)
+int pci_request_region_exclusive(struct pci_dev *pdev, int bar,
+				 const char *res_name)
 {
 	return __pci_request_region(pdev, bar, res_name, IORESOURCE_EXCLUSIVE);
 }
+EXPORT_SYMBOL(pci_request_region_exclusive);
+
 /**
  * pci_release_selected_regions - Release selected PCI I/O and memory resources
  * @pdev: PCI device whose resources were previously reserved
@@ -2587,9 +2604,10 @@ void pci_release_selected_regions(struct pci_dev *pdev, int bars)
 		if (bars & (1 << i))
 			pci_release_region(pdev, i);
 }
+EXPORT_SYMBOL(pci_release_selected_regions);
 
 static int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
-				 const char *res_name, int excl)
+					  const char *res_name, int excl)
 {
 	int i;
 
@@ -2600,7 +2618,7 @@ static int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
 	return 0;
 
 err_out:
-	while(--i >= 0)
+	while (--i >= 0)
 		if (bars & (1 << i))
 			pci_release_region(pdev, i);
 
@@ -2619,13 +2637,15 @@ int pci_request_selected_regions(struct pci_dev *pdev, int bars,
 {
 	return __pci_request_selected_regions(pdev, bars, res_name, 0);
 }
+EXPORT_SYMBOL(pci_request_selected_regions);
 
-int pci_request_selected_regions_exclusive(struct pci_dev *pdev,
-				 int bars, const char *res_name)
+int pci_request_selected_regions_exclusive(struct pci_dev *pdev, int bars,
+					   const char *res_name)
 {
 	return __pci_request_selected_regions(pdev, bars, res_name,
 			IORESOURCE_EXCLUSIVE);
 }
+EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
 
 /**
  *	pci_release_regions - Release reserved PCI I/O and memory resources
@@ -2640,6 +2660,7 @@ void pci_release_regions(struct pci_dev *pdev)
 {
 	pci_release_selected_regions(pdev, (1 << 6) - 1);
 }
+EXPORT_SYMBOL(pci_release_regions);
 
 /**
  *	pci_request_regions - Reserved PCI I/O and memory resources
@@ -2658,6 +2679,7 @@ int pci_request_regions(struct pci_dev *pdev, const char *res_name)
 {
 	return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name);
 }
+EXPORT_SYMBOL(pci_request_regions);
 
 /**
  *	pci_request_regions_exclusive - Reserved PCI I/O and memory resources
@@ -2680,6 +2702,7 @@ int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name)
 	return pci_request_selected_regions_exclusive(pdev,
 					((1 << 6) - 1), res_name);
 }
+EXPORT_SYMBOL(pci_request_regions_exclusive);
 
 static void __pci_set_master(struct pci_dev *dev, bool enable)
 {
@@ -2749,6 +2772,7 @@ void pci_set_master(struct pci_dev *dev)
 	__pci_set_master(dev, true);
 	pcibios_set_master(dev);
 }
+EXPORT_SYMBOL(pci_set_master);
 
 /**
  * pci_clear_master - disables bus-mastering for device dev
@@ -2758,6 +2782,7 @@ void pci_clear_master(struct pci_dev *dev)
 {
 	__pci_set_master(dev, false);
 }
+EXPORT_SYMBOL(pci_clear_master);
 
 /**
  * pci_set_cacheline_size - ensure the CACHE_LINE_SIZE register is programmed
@@ -2790,30 +2815,13 @@ int pci_set_cacheline_size(struct pci_dev *dev)
 	if (cacheline_size == pci_cache_line_size)
 		return 0;
 
-	dev_printk(KERN_DEBUG, &dev->dev, "cache line size of %d is not "
-		   "supported\n", pci_cache_line_size << 2);
+	dev_printk(KERN_DEBUG, &dev->dev, "cache line size of %d is not supported\n",
+		   pci_cache_line_size << 2);
 
 	return -EINVAL;
 }
 EXPORT_SYMBOL_GPL(pci_set_cacheline_size);
 
-#ifdef PCI_DISABLE_MWI
-int pci_set_mwi(struct pci_dev *dev)
-{
-	return 0;
-}
-
-int pci_try_set_mwi(struct pci_dev *dev)
-{
-	return 0;
-}
-
-void pci_clear_mwi(struct pci_dev *dev)
-{
-}
-
-#else
-
 /**
  * pci_set_mwi - enables memory-write-invalidate PCI transaction
  * @dev: the PCI device for which MWI is enabled
@@ -2822,9 +2830,11 @@ void pci_clear_mwi(struct pci_dev *dev)
  *
  * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
  */
-int
-pci_set_mwi(struct pci_dev *dev)
+int pci_set_mwi(struct pci_dev *dev)
 {
+#ifdef PCI_DISABLE_MWI
+	return 0;
+#else
 	int rc;
 	u16 cmd;
 
@@ -2833,14 +2843,15 @@ pci_set_mwi(struct pci_dev *dev)
 		return rc;
 
 	pci_read_config_word(dev, PCI_COMMAND, &cmd);
-	if (! (cmd & PCI_COMMAND_INVALIDATE)) {
+	if (!(cmd & PCI_COMMAND_INVALIDATE)) {
 		dev_dbg(&dev->dev, "enabling Mem-Wr-Inval\n");
 		cmd |= PCI_COMMAND_INVALIDATE;
 		pci_write_config_word(dev, PCI_COMMAND, cmd);
 	}
-
 	return 0;
+#endif
 }
+EXPORT_SYMBOL(pci_set_mwi);
 
 /**
  * pci_try_set_mwi - enables memory-write-invalidate PCI transaction
@@ -2853,9 +2864,13 @@ pci_set_mwi(struct pci_dev *dev)
  */
 int pci_try_set_mwi(struct pci_dev *dev)
 {
-	int rc = pci_set_mwi(dev);
-	return rc;
+#ifdef PCI_DISABLE_MWI
+	return 0;
+#else
+	return pci_set_mwi(dev);
+#endif
 }
+EXPORT_SYMBOL(pci_try_set_mwi);
 
 /**
  * pci_clear_mwi - disables Memory-Write-Invalidate for device dev
@@ -2863,9 +2878,9 @@ int pci_try_set_mwi(struct pci_dev *dev)
  *
  * Disables PCI Memory-Write-Invalidate transaction on the device
  */
-void
-pci_clear_mwi(struct pci_dev *dev)
+void pci_clear_mwi(struct pci_dev *dev)
 {
+#ifndef PCI_DISABLE_MWI
 	u16 cmd;
 
 	pci_read_config_word(dev, PCI_COMMAND, &cmd);
@@ -2873,8 +2888,9 @@ pci_clear_mwi(struct pci_dev *dev)
 		cmd &= ~PCI_COMMAND_INVALIDATE;
 		pci_write_config_word(dev, PCI_COMMAND, cmd);
 	}
+#endif
 }
-#endif /* ! PCI_DISABLE_MWI */
+EXPORT_SYMBOL(pci_clear_mwi);
 
 /**
  * pci_intx - enables/disables PCI INTx for device dev
@@ -2883,18 +2899,16 @@ pci_clear_mwi(struct pci_dev *dev)
  *
  * Enables/disables PCI INTx for device dev
  */
-void
-pci_intx(struct pci_dev *pdev, int enable)
+void pci_intx(struct pci_dev *pdev, int enable)
 {
 	u16 pci_command, new;
 
 	pci_read_config_word(pdev, PCI_COMMAND, &pci_command);
 
-	if (enable) {
+	if (enable)
 		new = pci_command & ~PCI_COMMAND_INTX_DISABLE;
-	} else {
+	else
 		new = pci_command | PCI_COMMAND_INTX_DISABLE;
-	}
 
 	if (new != pci_command) {
 		struct pci_devres *dr;
@@ -2908,6 +2922,7 @@ pci_intx(struct pci_dev *pdev, int enable)
 		}
 	}
 }
+EXPORT_SYMBOL_GPL(pci_intx);
 
 /**
  * pci_intx_mask_supported - probe for INTx masking support
@@ -2937,8 +2952,8 @@ bool pci_intx_mask_supported(struct pci_dev *dev)
 	 * go ahead and check it.
 	 */
 	if ((new ^ orig) & ~PCI_COMMAND_INTX_DISABLE) {
-		dev_err(&dev->dev, "Command register changed from "
-			"0x%x to 0x%x: driver or hardware bug?\n", orig, new);
+		dev_err(&dev->dev, "Command register changed from 0x%x to 0x%x: driver or hardware bug?\n",
+			orig, new);
 	} else if ((new ^ orig) & PCI_COMMAND_INTX_DISABLE) {
 		mask_supported = true;
 		pci_write_config_word(dev, PCI_COMMAND, orig);
@@ -3120,12 +3135,16 @@ static int pci_af_flr(struct pci_dev *dev, int probe)
 	if (probe)
 		return 0;
 
-	/* Wait for Transaction Pending bit clean */
-	if (pci_wait_for_pending(dev, pos + PCI_AF_STATUS, PCI_AF_STATUS_TP))
+	/*
+	 * Wait for Transaction Pending bit to clear.  A word-aligned test
+	 * is used, so we use the conrol offset rather than status and shift
+	 * the test bit to match.
+	 */
+	if (pci_wait_for_pending(dev, pos + PCI_AF_CTRL,
+				 PCI_AF_STATUS_TP << 8))
 		goto clear;
 
-	dev_err(&dev->dev, "transaction is not cleared; "
-			"proceeding with reset anyway\n");
+	dev_err(&dev->dev, "transaction is not cleared; proceeding with reset anyway\n");
 
 clear:
 	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
@@ -3179,14 +3198,7 @@ static int pci_pm_reset(struct pci_dev *dev, int probe)
 	return 0;
 }
 
-/**
- * pci_reset_bridge_secondary_bus - Reset the secondary bus on a PCI bridge.
- * @dev: Bridge device
- *
- * Use the bridge control register to assert reset on the secondary bus.
- * Devices on the secondary bus are left in power-on state.
- */
-void pci_reset_bridge_secondary_bus(struct pci_dev *dev)
+void __weak pcibios_reset_secondary_bus(struct pci_dev *dev)
 {
 	u16 ctrl;
 
@@ -3211,6 +3223,18 @@ void pci_reset_bridge_secondary_bus(struct pci_dev *dev)
 	 */
 	ssleep(1);
 }
+
+/**
+ * pci_reset_bridge_secondary_bus - Reset the secondary bus on a PCI bridge.
+ * @dev: Bridge device
+ *
+ * Use the bridge control register to assert reset on the secondary bus.
+ * Devices on the secondary bus are left in power-on state.
+ */
+void pci_reset_bridge_secondary_bus(struct pci_dev *dev)
+{
+	pcibios_reset_secondary_bus(dev);
+}
 EXPORT_SYMBOL_GPL(pci_reset_bridge_secondary_bus);
 
 static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
@@ -4095,6 +4119,7 @@ int pci_select_bars(struct pci_dev *dev, unsigned long flags)
 			bars |= (1 << i);
 	return bars;
 }
+EXPORT_SYMBOL(pci_select_bars);
 
 /**
  * pci_resource_bar - get position of the BAR associated with a resource
@@ -4134,7 +4159,7 @@ void __init pci_register_set_vga_state(arch_set_vga_state_t func)
 }
 
 static int pci_set_vga_state_arch(struct pci_dev *dev, bool decode,
-		      unsigned int command_bits, u32 flags)
+				  unsigned int command_bits, u32 flags)
 {
 	if (arch_set_vga_state)
 		return arch_set_vga_state(dev, decode, command_bits,
@@ -4246,11 +4271,10 @@ static resource_size_t pci_specified_resource_alignment(struct pci_dev *dev)
 			bus == dev->bus->number &&
 			slot == PCI_SLOT(dev->devfn) &&
 			func == PCI_FUNC(dev->devfn)) {
-			if (align_order == -1) {
+			if (align_order == -1)
 				align = PAGE_SIZE;
-			} else {
+			else
 				align = 1 << align_order;
-			}
 			/* Found */
 			break;
 		}
@@ -4368,7 +4392,6 @@ static int __init pci_resource_alignment_sysfs_init(void)
 	return bus_create_file(&pci_bus_type,
 					&bus_attr_resource_alignment);
 }
-
 late_initcall(pci_resource_alignment_sysfs_init);
 
 static void pci_no_domains(void)
@@ -4447,41 +4470,3 @@ static int __init pci_setup(char *str)
 	return 0;
 }
 early_param("pci", pci_setup);
-
-EXPORT_SYMBOL(pci_reenable_device);
-EXPORT_SYMBOL(pci_enable_device_io);
-EXPORT_SYMBOL(pci_enable_device_mem);
-EXPORT_SYMBOL(pci_enable_device);
-EXPORT_SYMBOL(pcim_enable_device);
-EXPORT_SYMBOL(pcim_pin_device);
-EXPORT_SYMBOL(pci_disable_device);
-EXPORT_SYMBOL(pci_find_capability);
-EXPORT_SYMBOL(pci_bus_find_capability);
-EXPORT_SYMBOL(pci_release_regions);
-EXPORT_SYMBOL(pci_request_regions);
-EXPORT_SYMBOL(pci_request_regions_exclusive);
-EXPORT_SYMBOL(pci_release_region);
-EXPORT_SYMBOL(pci_request_region);
-EXPORT_SYMBOL(pci_request_region_exclusive);
-EXPORT_SYMBOL(pci_release_selected_regions);
-EXPORT_SYMBOL(pci_request_selected_regions);
-EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
-EXPORT_SYMBOL(pci_set_master);
-EXPORT_SYMBOL(pci_clear_master);
-EXPORT_SYMBOL(pci_set_mwi);
-EXPORT_SYMBOL(pci_try_set_mwi);
-EXPORT_SYMBOL(pci_clear_mwi);
-EXPORT_SYMBOL_GPL(pci_intx);
-EXPORT_SYMBOL(pci_assign_resource);
-EXPORT_SYMBOL(pci_find_parent_resource);
-EXPORT_SYMBOL(pci_select_bars);
-
-EXPORT_SYMBOL(pci_set_power_state);
-EXPORT_SYMBOL(pci_save_state);
-EXPORT_SYMBOL(pci_restore_state);
-EXPORT_SYMBOL(pci_pme_capable);
-EXPORT_SYMBOL(pci_pme_active);
-EXPORT_SYMBOL(pci_wake_from_d3);
-EXPORT_SYMBOL(pci_prepare_to_sleep);
-EXPORT_SYMBOL(pci_back_from_sleep);
-EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);
diff --git a/drivers/pci/pcie/aer/aer_inject.c b/drivers/pci/pcie/aer/aer_inject.c
index 587e7e853107..182224acedbe 100644
--- a/drivers/pci/pcie/aer/aer_inject.c
+++ b/drivers/pci/pcie/aer/aer_inject.c
@@ -397,16 +397,14 @@ static int aer_inject(struct aer_error_inj *einj)
 	if (!aer_mask_override && einj->cor_status &&
 	    !(einj->cor_status & ~cor_mask)) {
 		ret = -EINVAL;
-		printk(KERN_WARNING "The correctable error(s) is masked "
-				"by device\n");
+		printk(KERN_WARNING "The correctable error(s) is masked by device\n");
 		spin_unlock_irqrestore(&inject_lock, flags);
 		goto out_put;
 	}
 	if (!aer_mask_override && einj->uncor_status &&
 	    !(einj->uncor_status & ~uncor_mask)) {
 		ret = -EINVAL;
-		printk(KERN_WARNING "The uncorrectable error(s) is masked "
-				"by device\n");
+		printk(KERN_WARNING "The uncorrectable error(s) is masked by device\n");
 		spin_unlock_irqrestore(&inject_lock, flags);
 		goto out_put;
 	}
@@ -464,8 +462,7 @@ static int aer_inject(struct aer_error_inj *einj)
 			goto out_put;
 		}
 		aer_irq(-1, edev);
-	}
-	else
+	} else
 		ret = -EINVAL;
 out_put:
 	kfree(err_alloc);
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index b2c8881da764..5653ea94547f 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -542,8 +542,7 @@ static void aer_recover_work_func(struct work_struct *work);
 #define AER_RECOVER_RING_ORDER		4
 #define AER_RECOVER_RING_SIZE		(1 << AER_RECOVER_RING_ORDER)
 
-struct aer_recover_entry
-{
+struct aer_recover_entry {
 	u8	bus;
 	u8	devfn;
 	u16	domain;
diff --git a/drivers/pci/pcie/aer/aerdrv_errprint.c b/drivers/pci/pcie/aer/aerdrv_errprint.c
index 34ff7026440c..36ed31b52198 100644
--- a/drivers/pci/pcie/aer/aerdrv_errprint.c
+++ b/drivers/pci/pcie/aer/aerdrv_errprint.c
@@ -172,9 +172,7 @@ void aer_print_error(struct pci_dev *dev, struct aer_err_info *info)
 	int id = ((dev->bus->number << 8) | dev->devfn);
 
 	if (!info->status) {
-		dev_err(&dev->dev,
-			"PCIe Bus Error: severity=%s, type=Unaccessible, "
-			"id=%04x(Unregistered Agent ID)\n",
+		dev_err(&dev->dev, "PCIe Bus Error: severity=%s, type=Unaccessible, id=%04x(Unregistered Agent ID)\n",
 			aer_error_severity_string[info->severity], id);
 		goto out;
 	}
@@ -182,13 +180,11 @@ void aer_print_error(struct pci_dev *dev, struct aer_err_info *info)
 	layer = AER_GET_LAYER_ERROR(info->severity, info->status);
 	agent = AER_GET_AGENT(info->severity, info->status);
 
-	dev_err(&dev->dev,
-		"PCIe Bus Error: severity=%s, type=%s, id=%04x(%s)\n",
+	dev_err(&dev->dev, "PCIe Bus Error: severity=%s, type=%s, id=%04x(%s)\n",
 		aer_error_severity_string[info->severity],
 		aer_error_layer[layer], id, aer_agent_string[agent]);
 
-	dev_err(&dev->dev,
-		"  device [%04x:%04x] error status/mask=%08x/%08x\n",
+	dev_err(&dev->dev, "  device [%04x:%04x] error status/mask=%08x/%08x\n",
 		dev->vendor, dev->device,
 		info->status, info->mask);
 
diff --git a/drivers/pci/pcie/pme.c b/drivers/pci/pcie/pme.c
index bbc3bdd2b189..82e06a86cd77 100644
--- a/drivers/pci/pcie/pme.c
+++ b/drivers/pci/pcie/pme.c
@@ -199,8 +199,7 @@ static void pcie_pme_handle_request(struct pci_dev *port, u16 req_id)
 		 * assuming that the PME was reported by a PCIe-PCI bridge that
 		 * used devfn different from zero.
 		 */
-		dev_dbg(&port->dev, "PME interrupt generated for "
-			"non-existent device %02x:%02x.%d\n",
+		dev_dbg(&port->dev, "PME interrupt generated for non-existent device %02x:%02x.%d\n",
 			busnr, PCI_SLOT(devfn), PCI_FUNC(devfn));
 		found = pcie_pme_from_pci_bridge(bus, 0);
 	}
diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c
index 0d8fdc48e642..80887eaa0668 100644
--- a/drivers/pci/pcie/portdrv_pci.c
+++ b/drivers/pci/pcie/portdrv_pci.c
@@ -204,8 +204,8 @@ static int pcie_portdrv_probe(struct pci_dev *dev,
 		return -ENODEV;
 
 	if (!dev->irq && dev->pin) {
-		dev_warn(&dev->dev, "device [%04x:%04x] has invalid IRQ; "
-			 "check vendor BIOS\n", dev->vendor, dev->device);
+		dev_warn(&dev->dev, "device [%04x:%04x] has invalid IRQ; check vendor BIOS\n",
+			 dev->vendor, dev->device);
 	}
 	status = pcie_port_device_register(dev);
 	if (status)
@@ -397,7 +397,7 @@ static struct pci_driver pcie_portdriver = {
 static int __init dmi_pcie_pme_disable_msi(const struct dmi_system_id *d)
 {
 	pr_notice("%s detected: will not use MSI for PCIe PME signaling\n",
-			d->ident);
+		  d->ident);
 	pcie_pme_disable_msi();
 	return 0;
 }
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 2bbf5221afb3..e3cf8a2e6292 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -168,7 +168,7 @@ static inline unsigned long decode_bar(struct pci_dev *dev, u32 bar)
  * Returns 1 if the BAR is 64-bit, or 0 if 32-bit.
  */
 int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,
-			struct resource *res, unsigned int pos)
+		    struct resource *res, unsigned int pos)
 {
 	u32 l, sz, mask;
 	u64 l64, sz64, mask64;
@@ -433,8 +433,7 @@ static void pci_read_bridge_mmio_pref(struct pci_bus *child)
 			limit |= ((unsigned long) mem_limit_hi) << 32;
 #else
 			if (mem_base_hi || mem_limit_hi) {
-				dev_err(&dev->dev, "can't handle 64-bit "
-					"address space for bridge\n");
+				dev_err(&dev->dev, "can't handle 64-bit address space for bridge\n");
 				return;
 			}
 #endif
@@ -604,7 +603,6 @@ static enum pci_bus_speed agp_speed(int agp3, int agpstat)
 	return agp_speeds[index];
 }
 
-
 static void pci_set_bus_speed(struct pci_bus *bus)
 {
 	struct pci_dev *bridge = bus->self;
@@ -636,11 +634,10 @@ static void pci_set_bus_speed(struct pci_bus *bus)
 		} else if (status & PCI_X_SSTATUS_266MHZ) {
 			max = PCI_SPEED_133MHz_PCIX_266;
 		} else if (status & PCI_X_SSTATUS_133MHZ) {
-			if ((status & PCI_X_SSTATUS_VERS) == PCI_X_SSTATUS_V2) {
+			if ((status & PCI_X_SSTATUS_VERS) == PCI_X_SSTATUS_V2)
 				max = PCI_SPEED_133MHz_PCIX_ECC;
-			} else {
+			else
 				max = PCI_SPEED_133MHz_PCIX;
-			}
 		} else {
 			max = PCI_SPEED_66MHz_PCIX;
 		}
@@ -664,7 +661,6 @@ static void pci_set_bus_speed(struct pci_bus *bus)
 	}
 }
 
-
 static struct pci_bus *pci_alloc_child_bus(struct pci_bus *parent,
 					   struct pci_dev *bridge, int busnr)
 {
@@ -729,7 +725,8 @@ add_dev:
 	return child;
 }
 
-struct pci_bus *pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev, int busnr)
+struct pci_bus *pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev,
+				int busnr)
 {
 	struct pci_bus *child;
 
@@ -741,6 +738,7 @@ struct pci_bus *pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev, int
 	}
 	return child;
 }
+EXPORT_SYMBOL(pci_add_new_bus);
 
 /*
  * If it's a bridge, configure it and scan the bus behind it.
@@ -887,7 +885,7 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass)
 			 * as cards with a PCI-to-PCI bridge can be
 			 * inserted later.
 			 */
-			for (i=0; i<CARDBUS_RESERVE_BUSNR; i++) {
+			for (i = 0; i < CARDBUS_RESERVE_BUSNR; i++) {
 				struct pci_bus *parent = bus;
 				if (pci_find_bus(pci_domain_nr(bus),
 							max+i+1))
@@ -934,8 +932,7 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass)
 		    (child->number > bus->busn_res.end) ||
 		    (child->number < bus->number) ||
 		    (child->busn_res.end < bus->number)) {
-			dev_info(&child->dev, "%pR %s "
-				"hidden behind%s bridge %s %pR\n",
+			dev_info(&child->dev, "%pR %s hidden behind%s bridge %s %pR\n",
 				&child->busn_res,
 				(bus->number > child->busn_res.end &&
 				 bus->busn_res.end < child->number) ?
@@ -952,6 +949,7 @@ out:
 
 	return max;
 }
+EXPORT_SYMBOL(pci_scan_bridge);
 
 /*
  * Read interrupt line and base address registers.
@@ -992,7 +990,6 @@ void set_pcie_hotplug_bridge(struct pci_dev *pdev)
 		pdev->is_hotplug_bridge = 1;
 }
 
-
 /**
  * pci_ext_cfg_is_aliased - is ext config space just an alias of std config?
  * @dev: PCI device
@@ -1225,13 +1222,13 @@ int pci_setup_device(struct pci_dev *dev)
 		break;
 
 	default:				    /* unknown header */
-		dev_err(&dev->dev, "unknown header type %02x, "
-			"ignoring device\n", dev->hdr_type);
+		dev_err(&dev->dev, "unknown header type %02x, ignoring device\n",
+			dev->hdr_type);
 		return -EIO;
 
 	bad:
-		dev_err(&dev->dev, "ignoring class %#08x (doesn't match header "
-			"type %02x)\n", dev->class, dev->hdr_type);
+		dev_err(&dev->dev, "ignoring class %#08x (doesn't match header type %02x)\n",
+			dev->class, dev->hdr_type);
 		dev->class = PCI_CLASS_NOT_DEFINED;
 	}
 
@@ -1283,7 +1280,7 @@ struct pci_dev *pci_alloc_dev(struct pci_bus *bus)
 EXPORT_SYMBOL(pci_alloc_dev);
 
 bool pci_bus_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *l,
-				 int crs_timeout)
+				int crs_timeout)
 {
 	int delay = 1;
 
@@ -1306,10 +1303,9 @@ bool pci_bus_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *l,
 			return false;
 		/* Card hasn't responded in 60 seconds?  Must be stuck. */
 		if (delay > crs_timeout) {
-			printk(KERN_WARNING "pci %04x:%02x:%02x.%d: not "
-					"responding\n", pci_domain_nr(bus),
-					bus->number, PCI_SLOT(devfn),
-					PCI_FUNC(devfn));
+			printk(KERN_WARNING "pci %04x:%02x:%02x.%d: not responding\n",
+			       pci_domain_nr(bus), bus->number, PCI_SLOT(devfn),
+			       PCI_FUNC(devfn));
 			return false;
 		}
 	}
@@ -1519,6 +1515,7 @@ int pci_scan_slot(struct pci_bus *bus, int devfn)
 
 	return nr;
 }
+EXPORT_SYMBOL(pci_scan_slot);
 
 static int pcie_find_smpss(struct pci_dev *dev, void *data)
 {
@@ -1613,9 +1610,7 @@ static void pcie_write_mrrs(struct pci_dev *dev)
 	}
 
 	if (mrrs < 128)
-		dev_err(&dev->dev, "MRRS was unable to be configured with a "
-			"safe value.  If problems are experienced, try running "
-			"with pci=pcie_bus_safe.\n");
+		dev_err(&dev->dev, "MRRS was unable to be configured with a safe value.  If problems are experienced, try running with pci=pcie_bus_safe\n");
 }
 
 static void pcie_bus_detect_mps(struct pci_dev *dev)
@@ -1652,8 +1647,8 @@ static int pcie_bus_configure_set(struct pci_dev *dev, void *data)
 	pcie_write_mps(dev, mps);
 	pcie_write_mrrs(dev);
 
-	dev_info(&dev->dev, "Max Payload Size set to %4d/%4d (was %4d), "
-		 "Max Read Rq %4d\n", pcie_get_mps(dev), 128 << dev->pcie_mpss,
+	dev_info(&dev->dev, "Max Payload Size set to %4d/%4d (was %4d), Max Read Rq %4d\n",
+		 pcie_get_mps(dev), 128 << dev->pcie_mpss,
 		 orig_mps, pcie_get_readrq(dev));
 
 	return 0;
@@ -1716,7 +1711,7 @@ unsigned int pci_scan_child_bus(struct pci_bus *bus)
 		bus->is_added = 1;
 	}
 
-	for (pass=0; pass < 2; pass++)
+	for (pass = 0; pass < 2; pass++)
 		list_for_each_entry(dev, &bus->devices, bus_list) {
 			if (pci_is_bridge(dev))
 				max = pci_scan_bridge(bus, dev, max, pass);
@@ -1732,6 +1727,7 @@ unsigned int pci_scan_child_bus(struct pci_bus *bus)
 	dev_dbg(&bus->dev, "bus scan returning with max=%02x\n", max);
 	return max;
 }
+EXPORT_SYMBOL_GPL(pci_scan_child_bus);
 
 /**
  * pcibios_root_bridge_prepare - Platform-specific host bridge setup.
@@ -2040,11 +2036,6 @@ unsigned int pci_rescan_bus(struct pci_bus *bus)
 }
 EXPORT_SYMBOL_GPL(pci_rescan_bus);
 
-EXPORT_SYMBOL(pci_add_new_bus);
-EXPORT_SYMBOL(pci_scan_slot);
-EXPORT_SYMBOL(pci_scan_bridge);
-EXPORT_SYMBOL_GPL(pci_scan_child_bus);
-
 /*
  * pci_rescan_bus(), pci_rescan_bus_bridge_resize() and PCI device removal
  * routines should always be executed under this mutex.
@@ -2063,7 +2054,8 @@ void pci_unlock_rescan_remove(void)
 }
 EXPORT_SYMBOL_GPL(pci_unlock_rescan_remove);
 
-static int __init pci_sort_bf_cmp(const struct device *d_a, const struct device *d_b)
+static int __init pci_sort_bf_cmp(const struct device *d_a,
+				  const struct device *d_b)
 {
 	const struct pci_dev *a = to_pci_dev(d_a);
 	const struct pci_dev *b = to_pci_dev(d_b);
diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c
index 46d1378f2e9e..3f155e78513f 100644
--- a/drivers/pci/proc.c
+++ b/drivers/pci/proc.c
@@ -17,15 +17,14 @@
 
 static int proc_initialized;	/* = 0 */
 
-static loff_t
-proc_bus_pci_lseek(struct file *file, loff_t off, int whence)
+static loff_t proc_bus_pci_lseek(struct file *file, loff_t off, int whence)
 {
 	struct pci_dev *dev = PDE_DATA(file_inode(file));
 	return fixed_size_llseek(file, off, whence, dev->cfg_size);
 }
 
-static ssize_t
-proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
+static ssize_t proc_bus_pci_read(struct file *file, char __user *buf,
+				 size_t nbytes, loff_t *ppos)
 {
 	struct pci_dev *dev = PDE_DATA(file_inode(file));
 	unsigned int pos = *ppos;
@@ -108,8 +107,8 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp
 	return nbytes;
 }
 
-static ssize_t
-proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, loff_t *ppos)
+static ssize_t proc_bus_pci_write(struct file *file, const char __user *buf,
+				  size_t nbytes, loff_t *ppos)
 {
 	struct inode *ino = file_inode(file);
 	struct pci_dev *dev = PDE_DATA(ino);
@@ -413,7 +412,7 @@ int pci_proc_detach_device(struct pci_dev *dev)
 	return 0;
 }
 
-int pci_proc_detach_bus(struct pci_bus* bus)
+int pci_proc_detach_bus(struct pci_bus *bus)
 {
 	proc_remove(bus->procdir);
 	return 0;
@@ -423,6 +422,7 @@ static int proc_bus_pci_dev_open(struct inode *inode, struct file *file)
 {
 	return seq_open(file, &proc_bus_pci_devices_op);
 }
+
 static const struct file_operations proc_bus_pci_dev_operations = {
 	.owner		= THIS_MODULE,
 	.open		= proc_bus_pci_dev_open,
@@ -443,6 +443,4 @@ static int __init pci_proc_init(void)
 
 	return 0;
 }
-
 device_initcall(pci_proc_init);
-
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 92e68c7747f7..d0f69269eb6c 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -48,8 +48,8 @@ static void quirk_mellanox_tavor(struct pci_dev *dev)
 {
 	dev->broken_parity_status = 1;	/* This device gives false positives */
 }
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_MELLANOX,PCI_DEVICE_ID_MELLANOX_TAVOR,quirk_mellanox_tavor);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_MELLANOX,PCI_DEVICE_ID_MELLANOX_TAVOR_BRIDGE,quirk_mellanox_tavor);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_MELLANOX, PCI_DEVICE_ID_MELLANOX_TAVOR, quirk_mellanox_tavor);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_MELLANOX, PCI_DEVICE_ID_MELLANOX_TAVOR_BRIDGE, quirk_mellanox_tavor);
 
 /* Deal with broken BIOSes that neglect to enable passive release,
    which can cause problems in combination with the 82441FX/PPro MTRRs */
@@ -82,7 +82,7 @@ DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82441,	quirk_p
 static void quirk_isa_dma_hangs(struct pci_dev *dev)
 {
 	if (!isa_dma_bridge_buggy) {
-		isa_dma_bridge_buggy=1;
+		isa_dma_bridge_buggy = 1;
 		dev_info(&dev->dev, "Activating ISA DMA hang workarounds\n");
 	}
 }
@@ -123,7 +123,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_TGP_LPC, quirk
  */
 static void quirk_nopcipci(struct pci_dev *dev)
 {
-	if ((pci_pci_problems & PCIPCI_FAIL)==0) {
+	if ((pci_pci_problems & PCIPCI_FAIL) == 0) {
 		dev_info(&dev->dev, "Disabling direct PCI/PCI transfers\n");
 		pci_pci_problems |= PCIPCI_FAIL;
 	}
@@ -148,7 +148,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_8151_0,	quirk_nopci
  */
 static void quirk_triton(struct pci_dev *dev)
 {
-	if ((pci_pci_problems&PCIPCI_TRITON)==0) {
+	if ((pci_pci_problems&PCIPCI_TRITON) == 0) {
 		dev_info(&dev->dev, "Limiting direct PCI/PCI transfers\n");
 		pci_pci_problems |= PCIPCI_TRITON;
 	}
@@ -163,8 +163,8 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82439TX,	quirk_
  *	Made according to a windows driver based patch by George E. Breese
  *	see PCI Latency Adjust on http://www.viahardware.com/download/viatweak.shtm
  *	and http://www.georgebreese.com/net/software/#PCI
- *      Also see http://www.au-ja.org/review-kt133a-1-en.phtml for
- *      the info on which Mr Breese based his work.
+ *	Also see http://www.au-ja.org/review-kt133a-1-en.phtml for
+ *	the info on which Mr Breese based his work.
  *
  *	Updated based on further information from the site and also on
  *	information provided by VIA
@@ -177,14 +177,14 @@ static void quirk_vialatency(struct pci_dev *dev)
 	   a buggy southbridge */
 
 	p = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686, NULL);
-	if (p!=NULL) {
+	if (p != NULL) {
 		/* 0x40 - 0x4f == 686B, 0x10 - 0x2f == 686A; thanks Dan Hollis */
 		/* Check for buggy part revisions */
 		if (p->revision < 0x40 || p->revision > 0x42)
 			goto exit;
 	} else {
 		p = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231, NULL);
-		if (p==NULL)	/* No problem parts */
+		if (p == NULL)	/* No problem parts */
 			goto exit;
 		/* Check for buggy part revisions */
 		if (p->revision < 0x10 || p->revision > 0x12)
@@ -227,7 +227,7 @@ DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8361,		quirk_viala
  */
 static void quirk_viaetbf(struct pci_dev *dev)
 {
-	if ((pci_pci_problems&PCIPCI_VIAETBF)==0) {
+	if ((pci_pci_problems&PCIPCI_VIAETBF) == 0) {
 		dev_info(&dev->dev, "Limiting direct PCI/PCI transfers\n");
 		pci_pci_problems |= PCIPCI_VIAETBF;
 	}
@@ -236,7 +236,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C597_0,	quirk_via
 
 static void quirk_vsfx(struct pci_dev *dev)
 {
-	if ((pci_pci_problems&PCIPCI_VSFX)==0) {
+	if ((pci_pci_problems&PCIPCI_VSFX) == 0) {
 		dev_info(&dev->dev, "Limiting direct PCI/PCI transfers\n");
 		pci_pci_problems |= PCIPCI_VSFX;
 	}
@@ -251,7 +251,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C576,	quirk_vsfx)
  */
 static void quirk_alimagik(struct pci_dev *dev)
 {
-	if ((pci_pci_problems&PCIPCI_ALIMAGIK)==0) {
+	if ((pci_pci_problems&PCIPCI_ALIMAGIK) == 0) {
 		dev_info(&dev->dev, "Limiting direct PCI/PCI transfers\n");
 		pci_pci_problems |= PCIPCI_ALIMAGIK|PCIPCI_TRITON;
 	}
@@ -265,7 +265,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AL,	PCI_DEVICE_ID_AL_M1651,		quirk_alimagi
  */
 static void quirk_natoma(struct pci_dev *dev)
 {
-	if ((pci_pci_problems&PCIPCI_NATOMA)==0) {
+	if ((pci_pci_problems&PCIPCI_NATOMA) == 0) {
 		dev_info(&dev->dev, "Limiting direct PCI/PCI transfers\n");
 		pci_pci_problems |= PCIPCI_NATOMA;
 	}
@@ -315,8 +315,7 @@ static void quirk_cs5536_vsa(struct pci_dev *dev)
 	if (pci_resource_len(dev, 0) != 8) {
 		struct resource *res = &dev->resource[0];
 		res->end = res->start + 8 - 1;
-		dev_info(&dev->dev, "CS5536 ISA bridge bug detected "
-				"(incorrect header); workaround applied.\n");
+		dev_info(&dev->dev, "CS5536 ISA bridge bug detected (incorrect header); workaround applied\n");
 	}
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA, quirk_cs5536_vsa);
@@ -400,7 +399,8 @@ static void piix4_io_quirk(struct pci_dev *dev, const char *name, unsigned int p
 	 * let's get enough confirmation reports first.
 	 */
 	base &= -size;
-	dev_info(&dev->dev, "%s PIO at %04x-%04x\n", name, base, base + size - 1);
+	dev_info(&dev->dev, "%s PIO at %04x-%04x\n", name, base,
+		 base + size - 1);
 }
 
 static void piix4_mem_quirk(struct pci_dev *dev, const char *name, unsigned int port, unsigned int enable)
@@ -425,7 +425,8 @@ static void piix4_mem_quirk(struct pci_dev *dev, const char *name, unsigned int
 	 * reserve it, but let's get enough confirmation reports first.
 	 */
 	base &= -size;
-	dev_info(&dev->dev, "%s MMIO at %04x-%04x\n", name, base, base + size - 1);
+	dev_info(&dev->dev, "%s MMIO at %04x-%04x\n", name, base,
+		 base + size - 1);
 }
 
 /*
@@ -668,8 +669,7 @@ static void quirk_xio2000a(struct pci_dev *dev)
 	struct pci_dev *pdev;
 	u16 command;
 
-	dev_warn(&dev->dev, "TI XIO2000a quirk detected; "
-		"secondary bus fast back-to-back transfers disabled\n");
+	dev_warn(&dev->dev, "TI XIO2000a quirk detected; secondary bus fast back-to-back transfers disabled\n");
 	list_for_each_entry(pdev, &dev->subordinate->devices, bus_list) {
 		pci_read_config_word(pdev, PCI_COMMAND, &command);
 		if (command & PCI_COMMAND_FAST_BACK)
@@ -703,7 +703,7 @@ static void quirk_via_ioapic(struct pci_dev *dev)
 	       tmp == 0 ? "Disa" : "Ena");
 
 	/* Offset 0x58: External APIC IRQ output control */
-	pci_write_config_byte (dev, 0x58, tmp);
+	pci_write_config_byte(dev, 0x58, tmp);
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C686,	quirk_via_ioapic);
 DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C686,	quirk_via_ioapic);
@@ -761,8 +761,8 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SI,	PCI_ANY_ID,			quirk_ioapic_rmw);
 static void quirk_amd_8131_mmrbc(struct pci_dev *dev)
 {
 	if (dev->subordinate && dev->revision <= 0x12) {
-		dev_info(&dev->dev, "AMD8131 rev %x detected; "
-			"disabling PCI-X MMRBC\n", dev->revision);
+		dev_info(&dev->dev, "AMD8131 rev %x detected; disabling PCI-X MMRBC\n",
+			 dev->revision);
 		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MMRBC;
 	}
 }
@@ -916,12 +916,12 @@ static void quirk_amd_ordering(struct pci_dev *dev)
 {
 	u32 pcic;
 	pci_read_config_dword(dev, 0x4C, &pcic);
-	if ((pcic&6)!=6) {
+	if ((pcic & 6) != 6) {
 		pcic |= 6;
 		dev_warn(&dev->dev, "BIOS failed to enable PCI standards compliance; fixing this error\n");
 		pci_write_config_dword(dev, 0x4C, pcic);
 		pci_read_config_dword(dev, 0x84, &pcic);
-		pcic |= (1<<23);	/* Required in this mode */
+		pcic |= (1 << 23);	/* Required in this mode */
 		pci_write_config_dword(dev, 0x84, pcic);
 	}
 }
@@ -937,7 +937,7 @@ DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_FE_GATE_700C
  */
 static void quirk_dunord(struct pci_dev *dev)
 {
-	struct resource *r = &dev->resource [1];
+	struct resource *r = &dev->resource[1];
 
 	r->flags |= IORESOURCE_UNSET;
 	r->start = 0;
@@ -967,11 +967,13 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_TOSHIBA,	0x605,	quirk_transparent_bridge)
 static void quirk_mediagx_master(struct pci_dev *dev)
 {
 	u8 reg;
+
 	pci_read_config_byte(dev, 0x41, &reg);
 	if (reg & 2) {
 		reg &= ~2;
-		dev_info(&dev->dev, "Fixup for MediaGX/Geode Slave Disconnect Boundary (0x41=0x%02x)\n", reg);
-                pci_write_config_byte(dev, 0x41, reg);
+		dev_info(&dev->dev, "Fixup for MediaGX/Geode Slave Disconnect Boundary (0x41=0x%02x)\n",
+			 reg);
+		pci_write_config_byte(dev, 0x41, reg);
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_CYRIX,	PCI_DEVICE_ID_CYRIX_PCI_MASTER, quirk_mediagx_master);
@@ -1120,7 +1122,7 @@ static void asus_hides_smbus_hostbridge(struct pci_dev *dev)
 {
 	if (unlikely(dev->subsystem_vendor == PCI_VENDOR_ID_ASUSTEK)) {
 		if (dev->device == PCI_DEVICE_ID_INTEL_82845_HB)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0x8025: /* P4B-LX */
 			case 0x8070: /* P4B */
 			case 0x8088: /* P4B533 */
@@ -1128,14 +1130,14 @@ static void asus_hides_smbus_hostbridge(struct pci_dev *dev)
 				asus_hides_smbus = 1;
 			}
 		else if (dev->device == PCI_DEVICE_ID_INTEL_82845G_HB)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0x80b1: /* P4GE-V */
 			case 0x80b2: /* P4PE */
 			case 0x8093: /* P4B533-V */
 				asus_hides_smbus = 1;
 			}
 		else if (dev->device == PCI_DEVICE_ID_INTEL_82850_HB)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0x8030: /* P4T533 */
 				asus_hides_smbus = 1;
 			}
@@ -1175,7 +1177,7 @@ static void asus_hides_smbus_hostbridge(struct pci_dev *dev)
 			}
 	} else if (unlikely(dev->subsystem_vendor == PCI_VENDOR_ID_HP)) {
 		if (dev->device ==  PCI_DEVICE_ID_INTEL_82855PM_HB)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0x088C: /* HP Compaq nc8000 */
 			case 0x0890: /* HP Compaq nc6000 */
 				asus_hides_smbus = 1;
@@ -1192,20 +1194,20 @@ static void asus_hides_smbus_hostbridge(struct pci_dev *dev)
 			case 0x12bf: /* HP xw4100 */
 				asus_hides_smbus = 1;
 			}
-       } else if (unlikely(dev->subsystem_vendor == PCI_VENDOR_ID_SAMSUNG)) {
-               if (dev->device ==  PCI_DEVICE_ID_INTEL_82855PM_HB)
-                       switch(dev->subsystem_device) {
-                       case 0xC00C: /* Samsung P35 notebook */
-                               asus_hides_smbus = 1;
-                       }
+	} else if (unlikely(dev->subsystem_vendor == PCI_VENDOR_ID_SAMSUNG)) {
+		if (dev->device ==  PCI_DEVICE_ID_INTEL_82855PM_HB)
+			switch (dev->subsystem_device) {
+			case 0xC00C: /* Samsung P35 notebook */
+				asus_hides_smbus = 1;
+		}
 	} else if (unlikely(dev->subsystem_vendor == PCI_VENDOR_ID_COMPAQ)) {
 		if (dev->device == PCI_DEVICE_ID_INTEL_82855PM_HB)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0x0058: /* Compaq Evo N620c */
 				asus_hides_smbus = 1;
 			}
 		else if (dev->device == PCI_DEVICE_ID_INTEL_82810_IG3)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0xB16C: /* Compaq Deskpro EP 401963-001 (PCA# 010174) */
 				/* Motherboard doesn't have Host bridge
 				 * subvendor/subdevice IDs, therefore checking
@@ -1213,7 +1215,7 @@ static void asus_hides_smbus_hostbridge(struct pci_dev *dev)
 				asus_hides_smbus = 1;
 			}
 		else if (dev->device == PCI_DEVICE_ID_INTEL_82801DB_2)
-			switch(dev->subsystem_device) {
+			switch (dev->subsystem_device) {
 			case 0x00b8: /* Compaq Evo D510 CMT */
 			case 0x00b9: /* Compaq Evo D510 SFF */
 			case 0x00ba: /* Compaq Evo D510 USDT */
@@ -1261,7 +1263,8 @@ static void asus_hides_smbus_lpc(struct pci_dev *dev)
 		pci_write_config_word(dev, 0xF2, val & (~0x8));
 		pci_read_config_word(dev, 0xF2, &val);
 		if (val & 0x8)
-			dev_info(&dev->dev, "i801 SMBus device continues to play 'hide and seek'! 0x%x\n", val);
+			dev_info(&dev->dev, "i801 SMBus device continues to play 'hide and seek'! 0x%x\n",
+				 val);
 		else
 			dev_info(&dev->dev, "Enabled i801 SMBus device\n");
 	}
@@ -1409,7 +1412,8 @@ static void asus_hides_ac97_lpc(struct pci_dev *dev)
 		pci_write_config_byte(dev, 0x50, val & (~0xc0));
 		pci_read_config_byte(dev, 0x50, &val);
 		if (val & 0xc0)
-			dev_info(&dev->dev, "Onboard AC97/MC97 devices continue to play 'hide and seek'! 0x%x\n", val);
+			dev_info(&dev->dev, "Onboard AC97/MC97 devices continue to play 'hide and seek'! 0x%x\n",
+				 val);
 		else
 			dev_info(&dev->dev, "Enabled onboard AC97/MC97 devices\n");
 	}
@@ -1514,10 +1518,8 @@ static void quirk_alder_ioapic(struct pci_dev *pdev)
 
 	/* The next five BARs all seem to be rubbish, so just clean
 	 * them out */
-	for (i=1; i < 6; i++) {
+	for (i = 1; i < 6; i++)
 		memset(&pdev->resource[i], 0, sizeof(pdev->resource[i]));
-	}
-
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_EESSC,	quirk_alder_ioapic);
 #endif
@@ -1552,7 +1554,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_PXHV,	quirk_pci
  * Some Intel PCI Express chipsets have trouble with downstream
  * device power management.
  */
-static void quirk_intel_pcie_pm(struct pci_dev * dev)
+static void quirk_intel_pcie_pm(struct pci_dev *dev)
 {
 	pci_pm_d3_delay = 120;
 	dev->no_d1d2 = 1;
@@ -1721,8 +1723,8 @@ static void quirk_disable_amd_8111_boot_interrupt(struct pci_dev *dev)
 
 	pci_read_config_word(dev, AMD_8111_PCI_IRQ_ROUTING, &pci_config_word);
 	if (!pci_config_word) {
-		dev_info(&dev->dev, "boot interrupts on device [%04x:%04x] "
-			 "already disabled\n", dev->vendor, dev->device);
+		dev_info(&dev->dev, "boot interrupts on device [%04x:%04x] already disabled\n",
+			 dev->vendor, dev->device);
 		return;
 	}
 	pci_write_config_word(dev, AMD_8111_PCI_IRQ_ROUTING, 0);
@@ -1770,8 +1772,7 @@ static void quirk_plx_pci9050(struct pci_dev *dev)
 		if (pci_resource_len(dev, bar) == 0x80 &&
 		    (pci_resource_start(dev, bar) & 0x80)) {
 			struct resource *r = &dev->resource[bar];
-			dev_info(&dev->dev,
-				 "Re-allocating PLX PCI 9050 BAR %u to length 256 to avoid bit 7 bug\n",
+			dev_info(&dev->dev, "Re-allocating PLX PCI 9050 BAR %u to length 256 to avoid bit 7 bug\n",
 				 bar);
 			r->flags |= IORESOURCE_UNSET;
 			r->start = 0;
@@ -1818,9 +1819,7 @@ static void quirk_netmos(struct pci_dev *dev)
 	case PCI_DEVICE_ID_NETMOS_9845:
 	case PCI_DEVICE_ID_NETMOS_9855:
 		if (num_parallel) {
-			dev_info(&dev->dev, "Netmos %04x (%u parallel, "
-				"%u serial); changing class SERIAL to OTHER "
-				"(use parport_serial)\n",
+			dev_info(&dev->dev, "Netmos %04x (%u parallel, %u serial); changing class SERIAL to OTHER (use parport_serial)\n",
 				dev->device, num_parallel, num_serial);
 			dev->class = (PCI_CLASS_COMMUNICATION_OTHER << 8) |
 			    (dev->class & 0xff);
@@ -1887,8 +1886,7 @@ static void quirk_e100_interrupt(struct pci_dev *dev)
 
 	cmd_hi = readb(csr + 3);
 	if (cmd_hi == 0) {
-		dev_warn(&dev->dev, "Firmware left e100 interrupts enabled; "
-			"disabling\n");
+		dev_warn(&dev->dev, "Firmware left e100 interrupts enabled; disabling\n");
 		writeb(1, csr + 3);
 	}
 
@@ -1958,8 +1956,7 @@ static void quirk_nvidia_ck804_pcie_aer_ext_cap(struct pci_dev *dev)
 	if (pci_read_config_byte(dev, 0xf41, &b) == 0) {
 		if (!(b & 0x20)) {
 			pci_write_config_byte(dev, 0xf41, b | 0x20);
-			dev_info(&dev->dev,
-			       "Linking AER extended capability\n");
+			dev_info(&dev->dev, "Linking AER extended capability\n");
 		}
 	}
 }
@@ -1997,8 +1994,7 @@ static void quirk_via_cx700_pci_parking_caching(struct pci_dev *dev)
 			/* Turn off PCI Bus Parking */
 			pci_write_config_byte(dev, 0x76, b ^ 0x40);
 
-			dev_info(&dev->dev,
-				"Disabling VIA CX700 PCI parking\n");
+			dev_info(&dev->dev, "Disabling VIA CX700 PCI parking\n");
 		}
 	}
 
@@ -2013,8 +2009,7 @@ static void quirk_via_cx700_pci_parking_caching(struct pci_dev *dev)
 			/* Disable "Read FIFO Timer" */
 			pci_write_config_byte(dev, 0x77, 0x0);
 
-			dev_info(&dev->dev,
-				"Disabling VIA CX700 PCI caching\n");
+			dev_info(&dev->dev, "Disabling VIA CX700 PCI caching\n");
 		}
 	}
 }
@@ -2149,8 +2144,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8380_0, quirk_disab
 static void quirk_disable_msi(struct pci_dev *dev)
 {
 	if (dev->subordinate) {
-		dev_warn(&dev->dev, "MSI quirk detected; "
-			"subordinate MSI disabled\n");
+		dev_warn(&dev->dev, "MSI quirk detected; subordinate MSI disabled\n");
 		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
 	}
 }
@@ -2189,8 +2183,7 @@ static int msi_ht_cap_enabled(struct pci_dev *dev)
 		u8 flags;
 
 		if (pci_read_config_byte(dev, pos + HT_MSI_FLAGS,
-					 &flags) == 0)
-		{
+					 &flags) == 0) {
 			dev_info(&dev->dev, "Found %s HT MSI Mapping\n",
 				flags & HT_MSI_FLAGS_ENABLE ?
 				"enabled" : "disabled");
@@ -2207,8 +2200,7 @@ static int msi_ht_cap_enabled(struct pci_dev *dev)
 static void quirk_msi_ht_cap(struct pci_dev *dev)
 {
 	if (dev->subordinate && !msi_ht_cap_enabled(dev)) {
-		dev_warn(&dev->dev, "MSI quirk detected; "
-			"subordinate MSI disabled\n");
+		dev_warn(&dev->dev, "MSI quirk detected; subordinate MSI disabled\n");
 		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
 	}
 }
@@ -2232,8 +2224,7 @@ static void quirk_nvidia_ck804_msi_ht_cap(struct pci_dev *dev)
 	if (!pdev)
 		return;
 	if (!msi_ht_cap_enabled(dev) && !msi_ht_cap_enabled(pdev)) {
-		dev_warn(&dev->dev, "MSI quirk detected; "
-			"subordinate MSI disabled\n");
+		dev_warn(&dev->dev, "MSI quirk detected; subordinate MSI disabled\n");
 		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
 	}
 	pci_dev_put(pdev);
@@ -2279,8 +2270,7 @@ static void nvenet_msi_disable(struct pci_dev *dev)
 	if (board_name &&
 	    (strstr(board_name, "P5N32-SLI PREMIUM") ||
 	     strstr(board_name, "P5N32-E SLI"))) {
-		dev_info(&dev->dev,
-			 "Disabling msi for MCP55 NIC on P5N32-SLI\n");
+		dev_info(&dev->dev, "Disabling msi for MCP55 NIC on P5N32-SLI\n");
 		dev->no_msi = 1;
 	}
 }
@@ -2489,8 +2479,7 @@ static void __nv_msi_ht_cap_quirk(struct pci_dev *dev, int all)
 	 */
 	host_bridge = pci_get_bus_and_slot(0, PCI_DEVFN(0, 0));
 	if (host_bridge == NULL) {
-		dev_warn(&dev->dev,
-			 "nv_msi_ht_cap_quirk didn't locate host bridge\n");
+		dev_warn(&dev->dev, "nv_msi_ht_cap_quirk didn't locate host bridge\n");
 		return;
 	}
 
@@ -2817,8 +2806,7 @@ static void quirk_intel_mc_errata(struct pci_dev *dev)
 	 */
 	err = pci_read_config_word(dev, 0x48, &rcc);
 	if (err) {
-		dev_err(&dev->dev, "Error attempting to read the read "
-			"completion coalescing register.\n");
+		dev_err(&dev->dev, "Error attempting to read the read completion coalescing register\n");
 		return;
 	}
 
@@ -2829,13 +2817,11 @@ static void quirk_intel_mc_errata(struct pci_dev *dev)
 
 	err = pci_write_config_word(dev, 0x48, rcc);
 	if (err) {
-		dev_err(&dev->dev, "Error attempting to write the read "
-			"completion coalescing register.\n");
+		dev_err(&dev->dev, "Error attempting to write the read completion coalescing register\n");
 		return;
 	}
 
-	pr_info_once("Read completion coalescing disabled due to hardware "
-		     "errata relating to 256B MPS.\n");
+	pr_info_once("Read completion coalescing disabled due to hardware errata relating to 256B MPS\n");
 }
 /* Intel 5000 series memory controllers and ports 2-7 */
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x25c0, quirk_intel_mc_errata);
@@ -2944,8 +2930,7 @@ static void disable_igfx_irq(struct pci_dev *dev)
 
 	/* Check if any interrupt line is still enabled */
 	if (readl(regs + I915_DEIER_REG) != 0) {
-		dev_warn(&dev->dev, "BIOS left Intel GPU interrupts enabled; "
-			"disabling\n");
+		dev_warn(&dev->dev, "BIOS left Intel GPU interrupts enabled; disabling\n");
 
 		writel(0, regs + I915_DEIER_REG);
 	}
@@ -3040,7 +3025,7 @@ void pci_fixup_device(enum pci_fixup_pass pass, struct pci_dev *dev)
 {
 	struct pci_fixup *start, *end;
 
-	switch(pass) {
+	switch (pass) {
 	case pci_fixup_early:
 		start = __start_pci_fixups_early;
 		end = __end_pci_fixups_early;
@@ -3112,8 +3097,8 @@ static int __init pci_apply_final_quirks(void)
 			if (!tmp || cls == tmp)
 				continue;
 
-			printk(KERN_DEBUG "PCI: CLS mismatch (%u != %u), "
-			       "using %u bytes\n", cls << 2, tmp << 2,
+			printk(KERN_DEBUG "PCI: CLS mismatch (%u != %u), using %u bytes\n",
+			       cls << 2, tmp << 2,
 			       pci_dfl_cache_line_size << 2);
 			pci_cache_line_size = pci_dfl_cache_line_size;
 		}
@@ -3342,6 +3327,85 @@ int pci_dev_specific_reset(struct pci_dev *dev, int probe)
 	return -ENOTTY;
 }
 
+static void quirk_dma_func0_alias(struct pci_dev *dev)
+{
+	if (PCI_FUNC(dev->devfn) != 0) {
+		dev->dma_alias_devfn = PCI_DEVFN(PCI_SLOT(dev->devfn), 0);
+		dev->dev_flags |= PCI_DEV_FLAGS_DMA_ALIAS_DEVFN;
+	}
+}
+
+/*
+ * https://bugzilla.redhat.com/show_bug.cgi?id=605888
+ *
+ * Some Ricoh devices use function 0 as the PCIe requester ID for DMA.
+ */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_RICOH, 0xe832, quirk_dma_func0_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_RICOH, 0xe476, quirk_dma_func0_alias);
+
+static void quirk_dma_func1_alias(struct pci_dev *dev)
+{
+	if (PCI_FUNC(dev->devfn) != 1) {
+		dev->dma_alias_devfn = PCI_DEVFN(PCI_SLOT(dev->devfn), 1);
+		dev->dev_flags |= PCI_DEV_FLAGS_DMA_ALIAS_DEVFN;
+	}
+}
+
+/*
+ * Marvell 88SE9123 uses function 1 as the requester ID for DMA.  In some
+ * SKUs function 1 is present and is a legacy IDE controller, in other
+ * SKUs this function is not present, making this a ghost requester.
+ * https://bugzilla.kernel.org/show_bug.cgi?id=42679
+ */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9123,
+			 quirk_dma_func1_alias);
+/* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c14 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9130,
+			 quirk_dma_func1_alias);
+/* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c47 + c57 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9172,
+			 quirk_dma_func1_alias);
+/* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c59 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x917a,
+			 quirk_dma_func1_alias);
+/* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c46 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x91a0,
+			 quirk_dma_func1_alias);
+/* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c49 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9230,
+			 quirk_dma_func1_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_TTI, 0x0642,
+			 quirk_dma_func1_alias);
+/* https://bugs.gentoo.org/show_bug.cgi?id=497630 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_JMICRON,
+			 PCI_DEVICE_ID_JMICRON_JMB388_ESD,
+			 quirk_dma_func1_alias);
+
+/*
+ * A few PCIe-to-PCI bridges fail to expose a PCIe capability, resulting in
+ * using the wrong DMA alias for the device.  Some of these devices can be
+ * used as either forward or reverse bridges, so we need to test whether the
+ * device is operating in the correct mode.  We could probably apply this
+ * quirk to PCI_ANY_ID, but for now we'll just use known offenders.  The test
+ * is for a non-root, non-PCIe bridge where the upstream device is PCIe and
+ * is not a PCIe-to-PCI bridge, then @pdev is actually a PCIe-to-PCI bridge.
+ */
+static void quirk_use_pcie_bridge_dma_alias(struct pci_dev *pdev)
+{
+	if (!pci_is_root_bus(pdev->bus) &&
+	    pdev->hdr_type == PCI_HEADER_TYPE_BRIDGE &&
+	    !pci_is_pcie(pdev) && pci_is_pcie(pdev->bus->self) &&
+	    pci_pcie_type(pdev->bus->self) != PCI_EXP_TYPE_PCI_BRIDGE)
+		pdev->dev_flags |= PCI_DEV_FLAG_PCIE_BRIDGE_ALIAS;
+}
+/* ASM1083/1085, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c46 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
+			 quirk_use_pcie_bridge_dma_alias);
+/* Tundra 8113, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c43 */
+DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
+/* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
+DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
+
 static struct pci_dev *pci_func_0_dma_source(struct pci_dev *dev)
 {
 	if (!PCI_FUNC(dev->devfn))
diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c
index c1839450d4d6..f955edb9bea7 100644
--- a/drivers/pci/rom.c
+++ b/drivers/pci/rom.c
@@ -38,6 +38,7 @@ int pci_enable_rom(struct pci_dev *pdev)
 	pci_write_config_dword(pdev, pdev->rom_base_reg, rom_addr);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(pci_enable_rom);
 
 /**
  * pci_disable_rom - disable ROM decoding for a PCI device
@@ -53,6 +54,7 @@ void pci_disable_rom(struct pci_dev *pdev)
 	rom_addr &= ~PCI_ROM_ADDRESS_ENABLE;
 	pci_write_config_dword(pdev, pdev->rom_base_reg, rom_addr);
 }
+EXPORT_SYMBOL_GPL(pci_disable_rom);
 
 /**
  * pci_get_rom_size - obtain the actual size of the ROM image
@@ -135,7 +137,7 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size)
 		} else {
 			/* assign the ROM an address if it doesn't have one */
 			if (res->parent == NULL &&
-			    pci_assign_resource(pdev,PCI_ROM_RESOURCE))
+			    pci_assign_resource(pdev, PCI_ROM_RESOURCE))
 				return NULL;
 			start = pci_resource_start(pdev, PCI_ROM_RESOURCE);
 			*size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
@@ -166,6 +168,7 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size)
 	*size = pci_get_rom_size(pdev, rom, *size);
 	return rom;
 }
+EXPORT_SYMBOL(pci_map_rom);
 
 /**
  * pci_unmap_rom - unmap the ROM from kernel space
@@ -187,6 +190,7 @@ void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom)
 	if (!(res->flags & (IORESOURCE_ROM_ENABLE | IORESOURCE_ROM_SHADOW)))
 		pci_disable_rom(pdev);
 }
+EXPORT_SYMBOL(pci_unmap_rom);
 
 /**
  * pci_cleanup_rom - free the ROM copy created by pci_map_rom_copy
@@ -199,7 +203,7 @@ void pci_cleanup_rom(struct pci_dev *pdev)
 	struct resource *res = &pdev->resource[PCI_ROM_RESOURCE];
 
 	if (res->flags & IORESOURCE_ROM_COPY) {
-		kfree((void*)(unsigned long)res->start);
+		kfree((void *)(unsigned long)res->start);
 		res->flags |= IORESOURCE_UNSET;
 		res->flags &= ~IORESOURCE_ROM_COPY;
 		res->start = 0;
@@ -222,9 +226,4 @@ void __iomem *pci_platform_rom(struct pci_dev *pdev, size_t *size)
 
 	return NULL;
 }
-
-EXPORT_SYMBOL(pci_map_rom);
-EXPORT_SYMBOL(pci_unmap_rom);
-EXPORT_SYMBOL_GPL(pci_enable_rom);
-EXPORT_SYMBOL_GPL(pci_disable_rom);
 EXPORT_SYMBOL(pci_platform_rom);
diff --git a/drivers/pci/search.c b/drivers/pci/search.c
index 8e495bda678f..827ad831f1dd 100644
--- a/drivers/pci/search.c
+++ b/drivers/pci/search.c
@@ -17,14 +17,100 @@ DECLARE_RWSEM(pci_bus_sem);
 EXPORT_SYMBOL_GPL(pci_bus_sem);
 
 /*
+ * pci_for_each_dma_alias - Iterate over DMA aliases for a device
+ * @pdev: starting downstream device
+ * @fn: function to call for each alias
+ * @data: opaque data to pass to @fn
+ *
+ * Starting @pdev, walk up the bus calling @fn for each possible alias
+ * of @pdev at the root bus.
+ */
+int pci_for_each_dma_alias(struct pci_dev *pdev,
+			   int (*fn)(struct pci_dev *pdev,
+				     u16 alias, void *data), void *data)
+{
+	struct pci_bus *bus;
+	int ret;
+
+	ret = fn(pdev, PCI_DEVID(pdev->bus->number, pdev->devfn), data);
+	if (ret)
+		return ret;
+
+	/*
+	 * If the device is broken and uses an alias requester ID for
+	 * DMA, iterate over that too.
+	 */
+	if (unlikely(pdev->dev_flags & PCI_DEV_FLAGS_DMA_ALIAS_DEVFN)) {
+		ret = fn(pdev, PCI_DEVID(pdev->bus->number,
+					 pdev->dma_alias_devfn), data);
+		if (ret)
+			return ret;
+	}
+
+	for (bus = pdev->bus; !pci_is_root_bus(bus); bus = bus->parent) {
+		struct pci_dev *tmp;
+
+		/* Skip virtual buses */
+		if (!bus->self)
+			continue;
+
+		tmp = bus->self;
+
+		/*
+		 * PCIe-to-PCI/X bridges alias transactions from downstream
+		 * devices using the subordinate bus number (PCI Express to
+		 * PCI/PCI-X Bridge Spec, rev 1.0, sec 2.3).  For all cases
+		 * where the upstream bus is PCI/X we alias to the bridge
+		 * (there are various conditions in the previous reference
+		 * where the bridge may take ownership of transactions, even
+		 * when the secondary interface is PCI-X).
+		 */
+		if (pci_is_pcie(tmp)) {
+			switch (pci_pcie_type(tmp)) {
+			case PCI_EXP_TYPE_ROOT_PORT:
+			case PCI_EXP_TYPE_UPSTREAM:
+			case PCI_EXP_TYPE_DOWNSTREAM:
+				continue;
+			case PCI_EXP_TYPE_PCI_BRIDGE:
+				ret = fn(tmp,
+					 PCI_DEVID(tmp->subordinate->number,
+						   PCI_DEVFN(0, 0)), data);
+				if (ret)
+					return ret;
+				continue;
+			case PCI_EXP_TYPE_PCIE_BRIDGE:
+				ret = fn(tmp,
+					 PCI_DEVID(tmp->bus->number,
+						   tmp->devfn), data);
+				if (ret)
+					return ret;
+				continue;
+			}
+		} else {
+			if (tmp->dev_flags & PCI_DEV_FLAG_PCIE_BRIDGE_ALIAS)
+				ret = fn(tmp,
+					 PCI_DEVID(tmp->subordinate->number,
+						   PCI_DEVFN(0, 0)), data);
+			else
+				ret = fn(tmp,
+					 PCI_DEVID(tmp->bus->number,
+						   tmp->devfn), data);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+/*
  * find the upstream PCIe-to-PCI bridge of a PCI device
  * if the device is PCIE, return NULL
  * if the device isn't connected to a PCIe bridge (that is its parent is a
  * legacy PCI bridge and the bridge is directly connected to bus 0), return its
  * parent
  */
-struct pci_dev *
-pci_find_upstream_pcie_bridge(struct pci_dev *pdev)
+struct pci_dev *pci_find_upstream_pcie_bridge(struct pci_dev *pdev)
 {
 	struct pci_dev *tmp = NULL;
 
@@ -56,12 +142,12 @@ static struct pci_bus *pci_do_find_bus(struct pci_bus *bus, unsigned char busnr)
 	struct pci_bus *child;
 	struct pci_bus *tmp;
 
-	if(bus->number == busnr)
+	if (bus->number == busnr)
 		return bus;
 
 	list_for_each_entry(tmp, &bus->children, node) {
 		child = pci_do_find_bus(tmp, busnr);
-		if(child)
+		if (child)
 			return child;
 	}
 	return NULL;
@@ -76,7 +162,7 @@ static struct pci_bus *pci_do_find_bus(struct pci_bus *bus, unsigned char busnr)
  * in the global list of PCI buses.  If the bus is found, a pointer to its
  * data structure is returned.  If no bus is found, %NULL is returned.
  */
-struct pci_bus * pci_find_bus(int domain, int busnr)
+struct pci_bus *pci_find_bus(int domain, int busnr)
 {
 	struct pci_bus *bus = NULL;
 	struct pci_bus *tmp_bus;
@@ -90,6 +176,7 @@ struct pci_bus * pci_find_bus(int domain, int busnr)
 	}
 	return NULL;
 }
+EXPORT_SYMBOL(pci_find_bus);
 
 /**
  * pci_find_next_bus - begin or continue searching for a PCI bus
@@ -100,8 +187,7 @@ struct pci_bus * pci_find_bus(int domain, int busnr)
  * @from is not %NULL, searches continue from next device on the
  * global list.
  */
-struct pci_bus *
-pci_find_next_bus(const struct pci_bus *from)
+struct pci_bus *pci_find_next_bus(const struct pci_bus *from)
 {
 	struct list_head *n;
 	struct pci_bus *b = NULL;
@@ -114,6 +200,7 @@ pci_find_next_bus(const struct pci_bus *from)
 	up_read(&pci_bus_sem);
 	return b;
 }
+EXPORT_SYMBOL(pci_find_next_bus);
 
 /**
  * pci_get_slot - locate PCI device for a given PCI slot
@@ -147,6 +234,7 @@ struct pci_dev *pci_get_slot(struct pci_bus *bus, unsigned int devfn)
 	up_read(&pci_bus_sem);
 	return dev;
 }
+EXPORT_SYMBOL(pci_get_slot);
 
 /**
  * pci_get_domain_bus_and_slot - locate PCI device for a given PCI domain (segment), bus, and slot
@@ -251,6 +339,7 @@ struct pci_dev *pci_get_subsys(unsigned int vendor, unsigned int device,
 
 	return pci_get_dev_by_id(&id, from);
 }
+EXPORT_SYMBOL(pci_get_subsys);
 
 /**
  * pci_get_device - begin or continue searching for a PCI device by vendor/device id
@@ -266,11 +355,12 @@ struct pci_dev *pci_get_subsys(unsigned int vendor, unsigned int device,
  * from next device on the global list.  The reference count for @from is
  * always decremented if it is not %NULL.
  */
-struct pci_dev *
-pci_get_device(unsigned int vendor, unsigned int device, struct pci_dev *from)
+struct pci_dev *pci_get_device(unsigned int vendor, unsigned int device,
+			       struct pci_dev *from)
 {
 	return pci_get_subsys(vendor, device, PCI_ANY_ID, PCI_ANY_ID, from);
 }
+EXPORT_SYMBOL(pci_get_device);
 
 /**
  * pci_get_class - begin or continue searching for a PCI device by class
@@ -299,6 +389,7 @@ struct pci_dev *pci_get_class(unsigned int class, struct pci_dev *from)
 
 	return pci_get_dev_by_id(&id, from);
 }
+EXPORT_SYMBOL(pci_get_class);
 
 /**
  * pci_dev_present - Returns 1 if device matching the device list is present, 0 if not.
@@ -328,12 +419,3 @@ int pci_dev_present(const struct pci_device_id *ids)
 	return 0;
 }
 EXPORT_SYMBOL(pci_dev_present);
-
-/* For boot time work */
-EXPORT_SYMBOL(pci_find_bus);
-EXPORT_SYMBOL(pci_find_next_bus);
-/* For everyone */
-EXPORT_SYMBOL(pci_get_device);
-EXPORT_SYMBOL(pci_get_subsys);
-EXPORT_SYMBOL(pci_get_slot);
-EXPORT_SYMBOL(pci_get_class);
diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index fd9b545c3cf5..a5a63ecfb628 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -68,7 +68,7 @@ static int add_to_list(struct list_head *head,
 
 	tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
 	if (!tmp) {
-		pr_warning("add_to_list: kmalloc() failed!\n");
+		pr_warn("add_to_list: kmalloc() failed!\n");
 		return -ENOMEM;
 	}
 
@@ -148,8 +148,7 @@ static void pdev_sort_resources(struct pci_dev *dev, struct list_head *head)
 
 		tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
 		if (!tmp)
-			panic("pdev_sort_resources(): "
-			      "kmalloc() failed!\n");
+			panic("pdev_sort_resources(): kmalloc() failed!\n");
 		tmp->res = r;
 		tmp->dev = dev;
 
@@ -736,7 +735,7 @@ static resource_size_t calculate_iosize(resource_size_t size,
 {
 	if (size < min_size)
 		size = min_size;
-	if (old_size == 1 )
+	if (old_size == 1)
 		old_size = 0;
 	/* To be fixed in 2.5: we should have sort of HAVE_ISA
 	   flag in the struct pci_bus. */
@@ -757,7 +756,7 @@ static resource_size_t calculate_memsize(resource_size_t size,
 {
 	if (size < min_size)
 		size = min_size;
-	if (old_size == 1 )
+	if (old_size == 1)
 		old_size = 0;
 	if (size < old_size)
 		size = old_size;
@@ -859,9 +858,8 @@ static void pbus_size_io(struct pci_bus *bus, resource_size_t min_size,
 			resource_size(b_res), min_align);
 	if (!size0 && !size1) {
 		if (b_res->start || b_res->end)
-			dev_info(&bus->self->dev, "disabling bridge window "
-				 "%pR to %pR (unused)\n", b_res,
-				 &bus->busn_res);
+			dev_info(&bus->self->dev, "disabling bridge window %pR to %pR (unused)\n",
+				 b_res, &bus->busn_res);
 		b_res->flags = 0;
 		return;
 	}
@@ -872,10 +870,9 @@ static void pbus_size_io(struct pci_bus *bus, resource_size_t min_size,
 	if (size1 > size0 && realloc_head) {
 		add_to_list(realloc_head, bus->self, b_res, size1-size0,
 			    min_align);
-		dev_printk(KERN_DEBUG, &bus->self->dev, "bridge window "
-				 "%pR to %pR add_size %llx\n", b_res,
-				 &bus->busn_res,
-				 (unsigned long long)size1-size0);
+		dev_printk(KERN_DEBUG, &bus->self->dev, "bridge window %pR to %pR add_size %llx\n",
+			   b_res, &bus->busn_res,
+			   (unsigned long long)size1-size0);
 	}
 }
 
@@ -974,9 +971,8 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask,
 			if (order < 0)
 				order = 0;
 			if (order >= ARRAY_SIZE(aligns)) {
-				dev_warn(&dev->dev, "disabling BAR %d: %pR "
-					 "(bad alignment %#llx)\n", i, r,
-					 (unsigned long long) align);
+				dev_warn(&dev->dev, "disabling BAR %d: %pR (bad alignment %#llx)\n",
+					 i, r, (unsigned long long) align);
 				r->flags = 0;
 				continue;
 			}
@@ -1003,9 +999,8 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask,
 				resource_size(b_res), min_align);
 	if (!size0 && !size1) {
 		if (b_res->start || b_res->end)
-			dev_info(&bus->self->dev, "disabling bridge window "
-				 "%pR to %pR (unused)\n", b_res,
-				 &bus->busn_res);
+			dev_info(&bus->self->dev, "disabling bridge window %pR to %pR (unused)\n",
+				 b_res, &bus->busn_res);
 		b_res->flags = 0;
 		return 0;
 	}
@@ -1014,9 +1009,9 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask,
 	b_res->flags |= IORESOURCE_STARTALIGN;
 	if (size1 > size0 && realloc_head) {
 		add_to_list(realloc_head, bus->self, b_res, size1-size0, min_align);
-		dev_printk(KERN_DEBUG, &bus->self->dev, "bridge window "
-				 "%pR to %pR add_size %llx\n", b_res,
-				 &bus->busn_res, (unsigned long long)size1-size0);
+		dev_printk(KERN_DEBUG, &bus->self->dev, "bridge window %pR to %pR add_size %llx\n",
+			   b_res, &bus->busn_res,
+			   (unsigned long long)size1-size0);
 	}
 	return 0;
 }
@@ -1274,8 +1269,8 @@ void __pci_bus_assign_resources(const struct pci_bus *bus,
 			break;
 
 		default:
-			dev_info(&dev->dev, "not setting up bridge for bus "
-				 "%04x:%02x\n", pci_domain_nr(b), b->number);
+			dev_info(&dev->dev, "not setting up bridge for bus %04x:%02x\n",
+				 pci_domain_nr(b), b->number);
 			break;
 		}
 	}
@@ -1312,8 +1307,8 @@ static void __pci_bridge_assign_resources(const struct pci_dev *bridge,
 		break;
 
 	default:
-		dev_info(&bridge->dev, "not setting up bridge for bus "
-			 "%04x:%02x\n", pci_domain_nr(b), b->number);
+		dev_info(&bridge->dev, "not setting up bridge for bus %04x:%02x\n",
+			 pci_domain_nr(b), b->number);
 		break;
 	}
 }
@@ -1430,10 +1425,10 @@ static void pci_bus_dump_res(struct pci_bus *bus)
 
 	pci_bus_for_each_resource(bus, res, i) {
 		if (!res || !res->end || !res->flags)
-                        continue;
+			continue;
 
 		dev_printk(KERN_DEBUG, &bus->dev, "resource %d %pR\n", i, res);
-        }
+	}
 }
 
 static void pci_bus_dump_resources(struct pci_bus *bus)
@@ -1458,7 +1453,7 @@ static int pci_bus_get_depth(struct pci_bus *bus)
 	int depth = 0;
 	struct pci_bus *child_bus;
 
-	list_for_each_entry(child_bus, &bus->children, node){
+	list_for_each_entry(child_bus, &bus->children, node) {
 		int ret;
 
 		ret = pci_bus_get_depth(child_bus);
diff --git a/drivers/pci/setup-irq.c b/drivers/pci/setup-irq.c
index dbc4ffcf42de..4e2d595d50ca 100644
--- a/drivers/pci/setup-irq.c
+++ b/drivers/pci/setup-irq.c
@@ -22,10 +22,9 @@ void __weak pcibios_update_irq(struct pci_dev *dev, int irq)
 	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
 }
 
-static void
-pdev_fixup_irq(struct pci_dev *dev,
-	       u8 (*swizzle)(struct pci_dev *, u8 *),
-	       int (*map_irq)(const struct pci_dev *, u8, u8))
+static void pdev_fixup_irq(struct pci_dev *dev,
+			   u8 (*swizzle)(struct pci_dev *, u8 *),
+			   int (*map_irq)(const struct pci_dev *, u8, u8))
 {
 	u8 pin, slot;
 	int irq = 0;
@@ -58,11 +57,11 @@ pdev_fixup_irq(struct pci_dev *dev,
 	pcibios_update_irq(dev, irq);
 }
 
-void
-pci_fixup_irqs(u8 (*swizzle)(struct pci_dev *, u8 *),
-	       int (*map_irq)(const struct pci_dev *, u8, u8))
+void pci_fixup_irqs(u8 (*swizzle)(struct pci_dev *, u8 *),
+		    int (*map_irq)(const struct pci_dev *, u8, u8))
 {
 	struct pci_dev *dev = NULL;
+
 	for_each_pci_dev(dev)
 		pdev_fixup_irq(dev, swizzle, map_irq);
 }
diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c
index 33f9e32d94d0..caed1ce6facd 100644
--- a/drivers/pci/setup-res.c
+++ b/drivers/pci/setup-res.c
@@ -96,8 +96,8 @@ void pci_update_resource(struct pci_dev *dev, int resno)
 		pci_write_config_dword(dev, reg + 4, new);
 		pci_read_config_dword(dev, reg + 4, &check);
 		if (check != new) {
-			dev_err(&dev->dev, "BAR %d: error updating "
-			       "(high %#08x != %#08x)\n", resno, new, check);
+			dev_err(&dev->dev, "BAR %d: error updating (high %#08x != %#08x)\n",
+				resno, new, check);
 		}
 	}
 
@@ -289,8 +289,8 @@ int pci_assign_resource(struct pci_dev *dev, int resno)
 	res->flags |= IORESOURCE_UNSET;
 	align = pci_resource_alignment(dev, res);
 	if (!align) {
-		dev_info(&dev->dev, "BAR %d: can't assign %pR "
-			 "(bogus alignment)\n", resno, res);
+		dev_info(&dev->dev, "BAR %d: can't assign %pR (bogus alignment)\n",
+			 resno, res);
 		return -EINVAL;
 	}
 
@@ -314,6 +314,7 @@ int pci_assign_resource(struct pci_dev *dev, int resno)
 	}
 	return ret;
 }
+EXPORT_SYMBOL(pci_assign_resource);
 
 int pci_reassign_resource(struct pci_dev *dev, int resno, resource_size_t addsize,
 			resource_size_t min_align)
@@ -324,8 +325,8 @@ int pci_reassign_resource(struct pci_dev *dev, int resno, resource_size_t addsiz
 
 	res->flags |= IORESOURCE_UNSET;
 	if (!res->parent) {
-		dev_info(&dev->dev, "BAR %d: can't reassign an unassigned resource %pR "
-			 "\n", resno, res);
+		dev_info(&dev->dev, "BAR %d: can't reassign an unassigned resource %pR\n",
+			 resno, res);
 		return -EINVAL;
 	}
 
diff --git a/drivers/pci/syscall.c b/drivers/pci/syscall.c
index 24750a1b39b6..b91c4da68365 100644
--- a/drivers/pci/syscall.c
+++ b/drivers/pci/syscall.c
@@ -99,7 +99,7 @@ SYSCALL_DEFINE5(pciconfig_write, unsigned long, bus, unsigned long, dfn,
 	if (!dev)
 		return -ENODEV;
 
-	switch(len) {
+	switch (len) {
 	case 1:
 		err = get_user(byte, (u8 __user *)buf);
 		if (err)
diff --git a/drivers/phy/Kconfig b/drivers/phy/Kconfig
index 16a2f067c242..64b98d242ea6 100644
--- a/drivers/phy/Kconfig
+++ b/drivers/phy/Kconfig
@@ -112,6 +112,7 @@ config PHY_EXYNOS5250_SATA
 config PHY_SUN4I_USB
 	tristate "Allwinner sunxi SoC USB PHY driver"
 	depends on ARCH_SUNXI && HAS_IOMEM && OF
+	depends on RESET_CONTROLLER
 	select GENERIC_PHY
 	help
 	  Enable this to support the transceiver that is part of Allwinner
@@ -122,6 +123,7 @@ config PHY_SUN4I_USB
 
 config PHY_SAMSUNG_USB2
 	tristate "Samsung USB 2.0 PHY driver"
+	depends on HAS_IOMEM
 	select GENERIC_PHY
 	select MFD_SYSCON
 	help
diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c
index c64a2f3b2d62..49c446530101 100644
--- a/drivers/phy/phy-core.c
+++ b/drivers/phy/phy-core.c
@@ -614,8 +614,9 @@ struct phy *phy_create(struct device *dev, const struct phy_ops *ops,
 	return phy;
 
 put_dev:
-	put_device(&phy->dev);
-	ida_remove(&phy_ida, phy->id);
+	put_device(&phy->dev);  /* calls phy_release() which frees resources */
+	return ERR_PTR(ret);
+
 free_phy:
 	kfree(phy);
 	return ERR_PTR(ret);
@@ -799,7 +800,7 @@ static void phy_release(struct device *dev)
 
 	phy = to_phy(dev);
 	dev_vdbg(dev, "releasing '%s'\n", dev_name(dev));
-	ida_remove(&phy_ida, phy->id);
+	ida_simple_remove(&phy_ida, phy->id);
 	kfree(phy);
 }
 
diff --git a/drivers/phy/phy-omap-usb2.c b/drivers/phy/phy-omap-usb2.c
index 7007c11fe07d..34b396146c8a 100644
--- a/drivers/phy/phy-omap-usb2.c
+++ b/drivers/phy/phy-omap-usb2.c
@@ -233,8 +233,8 @@ static int omap_usb2_probe(struct platform_device *pdev)
 	if (phy_data->flags & OMAP_USB2_CALIBRATE_FALSE_DISCONNECT) {
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 		phy->phy_base = devm_ioremap_resource(&pdev->dev, res);
-		if (!phy->phy_base)
-			return -ENOMEM;
+		if (IS_ERR(phy->phy_base))
+			return PTR_ERR(phy->phy_base);
 		phy->flags |= OMAP_USB2_CALIBRATE_FALSE_DISCONNECT;
 	}
 
@@ -262,7 +262,6 @@ static int omap_usb2_probe(struct platform_device *pdev)
 	otg->phy		= &phy->phy;
 
 	platform_set_drvdata(pdev, phy);
-	pm_runtime_enable(phy->dev);
 
 	generic_phy = devm_phy_create(phy->dev, &ops, NULL);
 	if (IS_ERR(generic_phy))
@@ -270,10 +269,13 @@ static int omap_usb2_probe(struct platform_device *pdev)
 
 	phy_set_drvdata(generic_phy, phy);
 
+	pm_runtime_enable(phy->dev);
 	phy_provider = devm_of_phy_provider_register(phy->dev,
 			of_phy_simple_xlate);
-	if (IS_ERR(phy_provider))
+	if (IS_ERR(phy_provider)) {
+		pm_runtime_disable(phy->dev);
 		return PTR_ERR(phy_provider);
+	}
 
 	phy->wkupclk = devm_clk_get(phy->dev, "wkupclk");
 	if (IS_ERR(phy->wkupclk)) {
@@ -317,6 +319,7 @@ static int omap_usb2_remove(struct platform_device *pdev)
 	if (!IS_ERR(phy->optclk))
 		clk_unprepare(phy->optclk);
 	usb_remove_phy(&phy->phy);
+	pm_runtime_disable(phy->dev);
 
 	return 0;
 }
diff --git a/drivers/phy/phy-samsung-usb2.c b/drivers/phy/phy-samsung-usb2.c
index 8a8c6bc8709a..1e69a32c221d 100644
--- a/drivers/phy/phy-samsung-usb2.c
+++ b/drivers/phy/phy-samsung-usb2.c
@@ -107,6 +107,7 @@ static const struct of_device_id samsung_usb2_phy_of_match[] = {
 #endif
 	{ },
 };
+MODULE_DEVICE_TABLE(of, samsung_usb2_phy_of_match);
 
 static int samsung_usb2_phy_probe(struct platform_device *pdev)
 {
diff --git a/drivers/pinctrl/berlin/berlin.c b/drivers/pinctrl/berlin/berlin.c
index edf5d2fd2b22..86db2235ab00 100644
--- a/drivers/pinctrl/berlin/berlin.c
+++ b/drivers/pinctrl/berlin/berlin.c
@@ -320,7 +320,7 @@ int berlin_pinctrl_probe(struct platform_device *pdev,
 
 	regmap = dev_get_regmap(&pdev->dev, NULL);
 	if (!regmap)
-		return PTR_ERR(regmap);
+		return -ENODEV;
 
 	pctrl = devm_kzalloc(dev, sizeof(*pctrl), GFP_KERNEL);
 	if (!pctrl)
diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.c b/drivers/pinctrl/sunxi/pinctrl-sunxi.c
index f1ca75e6d7b1..5f38c7f67834 100644
--- a/drivers/pinctrl/sunxi/pinctrl-sunxi.c
+++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.c
@@ -211,6 +211,10 @@ static int sunxi_pctrl_dt_node_to_map(struct pinctrl_dev *pctldev,
 			configlen++;
 
 		pinconfig = kzalloc(configlen * sizeof(*pinconfig), GFP_KERNEL);
+		if (!pinconfig) {
+			kfree(*map);
+			return -ENOMEM;
+		}
 
 		if (!of_property_read_u32(node, "allwinner,drive", &val)) {
 			u16 strength = (val + 1) * 10;
diff --git a/drivers/platform/chrome/chromeos_laptop.c b/drivers/platform/chrome/chromeos_laptop.c
index 7f3aad0e115c..7f1a2e2711bd 100644
--- a/drivers/platform/chrome/chromeos_laptop.c
+++ b/drivers/platform/chrome/chromeos_laptop.c
@@ -84,21 +84,19 @@ static struct i2c_board_info tsl2563_als_device = {
 	I2C_BOARD_INFO("tsl2563", TAOS_ALS_I2C_ADDR),
 };
 
+static int mxt_t19_keys[] = {
+	KEY_RESERVED,
+	KEY_RESERVED,
+	KEY_RESERVED,
+	KEY_RESERVED,
+	KEY_RESERVED,
+	BTN_LEFT
+};
+
 static struct mxt_platform_data atmel_224s_tp_platform_data = {
-	.x_line			= 18,
-	.y_line			= 12,
-	.x_size			= 102*20,
-	.y_size			= 68*20,
-	.blen			= 0x80,	/* Gain setting is in upper 4 bits */
-	.threshold		= 0x32,
-	.voltage		= 0,	/* 3.3V */
-	.orient			= MXT_VERTICAL_FLIP,
 	.irqflags		= IRQF_TRIGGER_FALLING,
-	.is_tp			= true,
-	.key_map		= { KEY_RESERVED,
-				    KEY_RESERVED,
-				    KEY_RESERVED,
-				    BTN_LEFT },
+	.t19_num_keys		= ARRAY_SIZE(mxt_t19_keys),
+	.t19_keymap		= mxt_t19_keys,
 	.config			= NULL,
 	.config_length		= 0,
 };
@@ -110,16 +108,7 @@ static struct i2c_board_info atmel_224s_tp_device = {
 };
 
 static struct mxt_platform_data atmel_1664s_platform_data = {
-	.x_line			= 32,
-	.y_line			= 50,
-	.x_size			= 1700,
-	.y_size			= 2560,
-	.blen			= 0x89,	/* Gain setting is in upper 4 bits */
-	.threshold		= 0x28,
-	.voltage		= 0,	/* 3.3V */
-	.orient			= MXT_ROTATED_90_COUNTER,
 	.irqflags		= IRQF_TRIGGER_FALLING,
-	.is_tp			= false,
 	.config			= NULL,
 	.config_length		= 0,
 };
diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 27df2c533b09..172f26ce59ac 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -102,7 +102,7 @@ config DELL_LAPTOP
 	default n
 	---help---
 	This driver adds support for rfkill and backlight control to Dell
-	laptops.
+	laptops (except for some models covered by the Compal driver).
 
 config DELL_WMI
 	tristate "Dell WMI extras"
@@ -127,6 +127,16 @@ config DELL_WMI_AIO
 	  To compile this driver as a module, choose M here: the module will
 	  be called dell-wmi-aio.
 
+config DELL_SMO8800
+	tristate "Dell Latitude freefall driver (ACPI SMO8800/SMO8810)"
+	depends on ACPI
+	---help---
+	  Say Y here if you want to support SMO8800/SMO8810 freefall device
+	  on Dell Latitude laptops.
+
+	  To compile this driver as a module, choose M here: the module will
+	  be called dell-smo8800.
+
 
 config FUJITSU_LAPTOP
 	tristate "Fujitsu Laptop Extras"
@@ -265,23 +275,21 @@ config PANASONIC_LAPTOP
 	  R2, R3, R5, T2, W2 and Y2 series), say Y.
 
 config COMPAL_LAPTOP
-	tristate "Compal Laptop Extras"
+	tristate "Compal (and others) Laptop Extras"
 	depends on ACPI
 	depends on BACKLIGHT_CLASS_DEVICE
 	depends on RFKILL
 	depends on HWMON
 	depends on POWER_SUPPLY
 	---help---
-	  This is a driver for laptops built by Compal:
-
-	  Compal FL90/IFL90
-	  Compal FL91/IFL91
-	  Compal FL92/JFL92
-	  Compal FT00/IFT00
+	  This is a driver for laptops built by Compal, and some models by
+	  other brands (e.g. Dell, Toshiba).
 
-	  It adds support for Bluetooth, WLAN and LCD brightness control.
+	  It adds support for rfkill, Bluetooth, WLAN and LCD brightness
+	  control.
 
-	  If you have an Compal FL9x/IFL9x/FT00 laptop, say Y or M here.
+	  For a (possibly incomplete) list of supported laptops, please refer
+	  to: Documentation/platform/x86-laptop-drivers.txt
 
 config SONY_LAPTOP
 	tristate "Sony Laptop Extras"
@@ -724,7 +732,7 @@ config IBM_RTL
 
 config XO1_RFKILL
 	tristate "OLPC XO-1 software RF kill switch"
-	depends on OLPC
+	depends on OLPC || COMPILE_TEST
 	depends on RFKILL
 	---help---
 	  Support for enabling/disabling the WLAN interface on the OLPC XO-1
@@ -732,6 +740,7 @@ config XO1_RFKILL
 
 config XO15_EBOOK
 	tristate "OLPC XO-1.5 ebook switch"
+	depends on OLPC || COMPILE_TEST
 	depends on ACPI && INPUT
 	---help---
 	  Support for the ebook switch on the OLPC XO-1.5 laptop.
diff --git a/drivers/platform/x86/Makefile b/drivers/platform/x86/Makefile
index 1a2eafc9d48e..c4ca428fd3db 100644
--- a/drivers/platform/x86/Makefile
+++ b/drivers/platform/x86/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_COMPAL_LAPTOP)	+= compal-laptop.o
 obj-$(CONFIG_DELL_LAPTOP)	+= dell-laptop.o
 obj-$(CONFIG_DELL_WMI)		+= dell-wmi.o
 obj-$(CONFIG_DELL_WMI_AIO)	+= dell-wmi-aio.o
+obj-$(CONFIG_DELL_SMO8800)	+= dell-smo8800.o
 obj-$(CONFIG_ACER_WMI)		+= acer-wmi.o
 obj-$(CONFIG_ACERHDF)		+= acerhdf.o
 obj-$(CONFIG_HP_ACCEL)		+= hp_accel.o
diff --git a/drivers/platform/x86/alienware-wmi.c b/drivers/platform/x86/alienware-wmi.c
index 541f9514f76f..297b6640213f 100644
--- a/drivers/platform/x86/alienware-wmi.c
+++ b/drivers/platform/x86/alienware-wmi.c
@@ -32,6 +32,7 @@
 #define WMAX_METHOD_HDMI_STATUS		0x2
 #define WMAX_METHOD_BRIGHTNESS		0x3
 #define WMAX_METHOD_ZONE_CONTROL	0x4
+#define WMAX_METHOD_HDMI_CABLE		0x5
 
 MODULE_AUTHOR("Mario Limonciello <mario_limonciello@dell.com>");
 MODULE_DESCRIPTION("Alienware special feature control");
@@ -350,12 +351,11 @@ static int alienware_zone_init(struct platform_device *dev)
 	char *name;
 
 	if (interface == WMAX) {
-		global_led.max_brightness = 100;
 		lighting_control_state = WMAX_RUNNING;
 	} else if (interface == LEGACY) {
-		global_led.max_brightness = 0x0F;
 		lighting_control_state = LEGACY_RUNNING;
 	}
+	global_led.max_brightness = 0x0F;
 	global_brightness = global_led.max_brightness;
 
 	/*
@@ -423,41 +423,85 @@ static void alienware_zone_exit(struct platform_device *dev)
 	The HDMI mux sysfs node indicates the status of the HDMI input mux.
 	It can toggle between standard system GPU output and HDMI input.
 */
-static ssize_t show_hdmi(struct device *dev, struct device_attribute *attr,
-			 char *buf)
+static acpi_status alienware_hdmi_command(struct hdmi_args *in_args,
+					  u32 command, int *out_data)
 {
 	acpi_status status;
-	struct acpi_buffer input;
 	union acpi_object *obj;
-	u32 tmp = 0;
-	struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
+	struct acpi_buffer input;
+	struct acpi_buffer output;
+
+	input.length = (acpi_size) sizeof(*in_args);
+	input.pointer = in_args;
+	if (out_data != NULL) {
+		output.length = ACPI_ALLOCATE_BUFFER;
+		output.pointer = NULL;
+		status = wmi_evaluate_method(WMAX_CONTROL_GUID, 1,
+					     command, &input, &output);
+	} else
+		status = wmi_evaluate_method(WMAX_CONTROL_GUID, 1,
+					     command, &input, NULL);
+
+	if (ACPI_SUCCESS(status) && out_data != NULL) {
+		obj = (union acpi_object *)output.pointer;
+		if (obj && obj->type == ACPI_TYPE_INTEGER)
+			*out_data = (u32) obj->integer.value;
+	}
+	return status;
+
+}
+
+static ssize_t show_hdmi_cable(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	acpi_status status;
+	u32 out_data;
 	struct hdmi_args in_args = {
 		.arg = 0,
 	};
-	input.length = (acpi_size) sizeof(in_args);
-	input.pointer = &in_args;
-	status = wmi_evaluate_method(WMAX_CONTROL_GUID, 1,
-				     WMAX_METHOD_HDMI_STATUS, &input, &output);
+	status =
+	    alienware_hdmi_command(&in_args, WMAX_METHOD_HDMI_CABLE,
+				   (u32 *) &out_data);
+	if (ACPI_SUCCESS(status)) {
+		if (out_data == 0)
+			return scnprintf(buf, PAGE_SIZE,
+					 "[unconnected] connected unknown\n");
+		else if (out_data == 1)
+			return scnprintf(buf, PAGE_SIZE,
+					 "unconnected [connected] unknown\n");
+	}
+	pr_err("alienware-wmi: unknown HDMI cable status: %d\n", status);
+	return scnprintf(buf, PAGE_SIZE, "unconnected connected [unknown]\n");
+}
+
+static ssize_t show_hdmi_source(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	acpi_status status;
+	u32 out_data;
+	struct hdmi_args in_args = {
+		.arg = 0,
+	};
+	status =
+	    alienware_hdmi_command(&in_args, WMAX_METHOD_HDMI_STATUS,
+				   (u32 *) &out_data);
 
 	if (ACPI_SUCCESS(status)) {
-		obj = (union acpi_object *)output.pointer;
-		if (obj && obj->type == ACPI_TYPE_INTEGER)
-			tmp = (u32) obj->integer.value;
-		if (tmp == 1)
+		if (out_data == 1)
 			return scnprintf(buf, PAGE_SIZE,
 					 "[input] gpu unknown\n");
-		else if (tmp == 2)
+		else if (out_data == 2)
 			return scnprintf(buf, PAGE_SIZE,
 					 "input [gpu] unknown\n");
 	}
-	pr_err("alienware-wmi: unknown HDMI status: %d\n", status);
+	pr_err("alienware-wmi: unknown HDMI source status: %d\n", out_data);
 	return scnprintf(buf, PAGE_SIZE, "input gpu [unknown]\n");
 }
 
-static ssize_t toggle_hdmi(struct device *dev, struct device_attribute *attr,
-			   const char *buf, size_t count)
+static ssize_t toggle_hdmi_source(struct device *dev,
+				  struct device_attribute *attr,
+				  const char *buf, size_t count)
 {
-	struct acpi_buffer input;
 	acpi_status status;
 	struct hdmi_args args;
 	if (strcmp(buf, "gpu\n") == 0)
@@ -467,33 +511,46 @@ static ssize_t toggle_hdmi(struct device *dev, struct device_attribute *attr,
 	else
 		args.arg = 3;
 	pr_debug("alienware-wmi: setting hdmi to %d : %s", args.arg, buf);
-	input.length = (acpi_size) sizeof(args);
-	input.pointer = &args;
-	status = wmi_evaluate_method(WMAX_CONTROL_GUID, 1,
-				     WMAX_METHOD_HDMI_SOURCE, &input, NULL);
+
+	status = alienware_hdmi_command(&args, WMAX_METHOD_HDMI_SOURCE, NULL);
+
 	if (ACPI_FAILURE(status))
 		pr_err("alienware-wmi: HDMI toggle failed: results: %u\n",
 		       status);
 	return count;
 }
 
-static DEVICE_ATTR(hdmi, S_IRUGO | S_IWUSR, show_hdmi, toggle_hdmi);
+static DEVICE_ATTR(cable, S_IRUGO, show_hdmi_cable, NULL);
+static DEVICE_ATTR(source, S_IRUGO | S_IWUSR, show_hdmi_source,
+		   toggle_hdmi_source);
+
+static struct attribute *hdmi_attrs[] = {
+	&dev_attr_cable.attr,
+	&dev_attr_source.attr,
+	NULL,
+};
 
-static void remove_hdmi(struct platform_device *device)
+static struct attribute_group hdmi_attribute_group = {
+	.name = "hdmi",
+	.attrs = hdmi_attrs,
+};
+
+static void remove_hdmi(struct platform_device *dev)
 {
-	device_remove_file(&device->dev, &dev_attr_hdmi);
+	sysfs_remove_group(&dev->dev.kobj, &hdmi_attribute_group);
 }
 
-static int create_hdmi(void)
+static int create_hdmi(struct platform_device *dev)
 {
-	int ret = -ENOMEM;
-	ret = device_create_file(&platform_device->dev, &dev_attr_hdmi);
+	int ret;
+
+	ret = sysfs_create_group(&dev->dev.kobj, &hdmi_attribute_group);
 	if (ret)
 		goto error_create_hdmi;
 	return 0;
 
 error_create_hdmi:
-	remove_hdmi(platform_device);
+	remove_hdmi(dev);
 	return ret;
 }
 
@@ -527,7 +584,7 @@ static int __init alienware_wmi_init(void)
 		goto fail_platform_device2;
 
 	if (interface == WMAX) {
-		ret = create_hdmi();
+		ret = create_hdmi(platform_device);
 		if (ret)
 			goto fail_prep_hdmi;
 	}
diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c
index 563f59efa669..ddf0eefd862c 100644
--- a/drivers/platform/x86/asus-nb-wmi.c
+++ b/drivers/platform/x86/asus-nb-wmi.c
@@ -137,6 +137,15 @@ static struct dmi_system_id asus_quirks[] = {
 	},
 	{
 		.callback = dmi_matched,
+		.ident = "ASUSTeK COMPUTER INC. X550CA",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "X550CA"),
+		},
+		.driver_data = &quirk_asus_x401u,
+	},
+	{
+		.callback = dmi_matched,
 		.ident = "ASUSTeK COMPUTER INC. X55A",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c
index 91ef69a52263..3c6ccedc82b6 100644
--- a/drivers/platform/x86/asus-wmi.c
+++ b/drivers/platform/x86/asus-wmi.c
@@ -266,7 +266,7 @@ static int asus_wmi_evaluate_method(u32 method_id, u32 arg0, u32 arg1,
 	struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
 	acpi_status status;
 	union acpi_object *obj;
-	u32 tmp;
+	u32 tmp = 0;
 
 	status = wmi_evaluate_method(ASUS_WMI_MGMT_GUID, 1, method_id,
 				     &input, &output);
@@ -277,8 +277,6 @@ static int asus_wmi_evaluate_method(u32 method_id, u32 arg0, u32 arg1,
 	obj = (union acpi_object *)output.pointer;
 	if (obj && obj->type == ACPI_TYPE_INTEGER)
 		tmp = (u32) obj->integer.value;
-	else
-		tmp = 0;
 
 	if (retval)
 		*retval = tmp;
diff --git a/drivers/platform/x86/dell-smo8800.c b/drivers/platform/x86/dell-smo8800.c
new file mode 100644
index 000000000000..a653716055d1
--- /dev/null
+++ b/drivers/platform/x86/dell-smo8800.c
@@ -0,0 +1,233 @@
+/*
+ *  dell-smo8800.c - Dell Latitude ACPI SMO8800/SMO8810 freefall sensor driver
+ *
+ *  Copyright (C) 2012 Sonal Santan <sonal.santan@gmail.com>
+ *  Copyright (C) 2014 Pali Rohár <pali.rohar@gmail.com>
+ *
+ *  This is loosely based on lis3lv02d driver.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ */
+
+#define DRIVER_NAME "smo8800"
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/acpi.h>
+#include <linux/interrupt.h>
+#include <linux/miscdevice.h>
+
+struct smo8800_device {
+	u32 irq;                     /* acpi device irq */
+	atomic_t counter;            /* count after last read */
+	struct miscdevice miscdev;   /* for /dev/freefall */
+	unsigned long misc_opened;   /* whether the device is open */
+	wait_queue_head_t misc_wait; /* Wait queue for the misc dev */
+	struct device *dev;          /* acpi device */
+};
+
+static irqreturn_t smo8800_interrupt_quick(int irq, void *data)
+{
+	struct smo8800_device *smo8800 = data;
+
+	atomic_inc(&smo8800->counter);
+	wake_up_interruptible(&smo8800->misc_wait);
+	return IRQ_WAKE_THREAD;
+}
+
+static irqreturn_t smo8800_interrupt_thread(int irq, void *data)
+{
+	struct smo8800_device *smo8800 = data;
+
+	dev_info(smo8800->dev, "detected free fall\n");
+	return IRQ_HANDLED;
+}
+
+static acpi_status smo8800_get_resource(struct acpi_resource *resource,
+					void *context)
+{
+	struct acpi_resource_extended_irq *irq;
+
+	if (resource->type != ACPI_RESOURCE_TYPE_EXTENDED_IRQ)
+		return AE_OK;
+
+	irq = &resource->data.extended_irq;
+	if (!irq || !irq->interrupt_count)
+		return AE_OK;
+
+	*((u32 *)context) = irq->interrupts[0];
+	return AE_CTRL_TERMINATE;
+}
+
+static u32 smo8800_get_irq(struct acpi_device *device)
+{
+	u32 irq = 0;
+	acpi_status status;
+
+	status = acpi_walk_resources(device->handle, METHOD_NAME__CRS,
+				     smo8800_get_resource, &irq);
+	if (ACPI_FAILURE(status)) {
+		dev_err(&device->dev, "acpi_walk_resources failed\n");
+		return 0;
+	}
+
+	return irq;
+}
+
+static ssize_t smo8800_misc_read(struct file *file, char __user *buf,
+				 size_t count, loff_t *pos)
+{
+	struct smo8800_device *smo8800 = container_of(file->private_data,
+					 struct smo8800_device, miscdev);
+
+	u32 data = 0;
+	unsigned char byte_data = 0;
+	ssize_t retval = 1;
+
+	if (count < 1)
+		return -EINVAL;
+
+	atomic_set(&smo8800->counter, 0);
+	retval = wait_event_interruptible(smo8800->misc_wait,
+				(data = atomic_xchg(&smo8800->counter, 0)));
+
+	if (retval)
+		return retval;
+
+	byte_data = 1;
+	retval = 1;
+
+	if (data < 255)
+		byte_data = data;
+	else
+		byte_data = 255;
+
+	if (put_user(byte_data, buf))
+		retval = -EFAULT;
+
+	return retval;
+}
+
+static int smo8800_misc_open(struct inode *inode, struct file *file)
+{
+	struct smo8800_device *smo8800 = container_of(file->private_data,
+					 struct smo8800_device, miscdev);
+
+	if (test_and_set_bit(0, &smo8800->misc_opened))
+		return -EBUSY; /* already open */
+
+	atomic_set(&smo8800->counter, 0);
+	return 0;
+}
+
+static int smo8800_misc_release(struct inode *inode, struct file *file)
+{
+	struct smo8800_device *smo8800 = container_of(file->private_data,
+					 struct smo8800_device, miscdev);
+
+	clear_bit(0, &smo8800->misc_opened); /* release the device */
+	return 0;
+}
+
+static const struct file_operations smo8800_misc_fops = {
+	.owner = THIS_MODULE,
+	.read = smo8800_misc_read,
+	.open = smo8800_misc_open,
+	.release = smo8800_misc_release,
+};
+
+static int smo8800_add(struct acpi_device *device)
+{
+	int err;
+	struct smo8800_device *smo8800;
+
+	smo8800 = devm_kzalloc(&device->dev, sizeof(*smo8800), GFP_KERNEL);
+	if (!smo8800) {
+		dev_err(&device->dev, "failed to allocate device data\n");
+		return -ENOMEM;
+	}
+
+	smo8800->dev = &device->dev;
+	smo8800->miscdev.minor = MISC_DYNAMIC_MINOR;
+	smo8800->miscdev.name = "freefall";
+	smo8800->miscdev.fops = &smo8800_misc_fops;
+
+	init_waitqueue_head(&smo8800->misc_wait);
+
+	err = misc_register(&smo8800->miscdev);
+	if (err) {
+		dev_err(&device->dev, "failed to register misc dev: %d\n", err);
+		return err;
+	}
+
+	device->driver_data = smo8800;
+
+	smo8800->irq = smo8800_get_irq(device);
+	if (!smo8800->irq) {
+		dev_err(&device->dev, "failed to obtain IRQ\n");
+		err = -EINVAL;
+		goto error;
+	}
+
+	err = request_threaded_irq(smo8800->irq, smo8800_interrupt_quick,
+				   smo8800_interrupt_thread,
+				   IRQF_TRIGGER_RISING | IRQF_ONESHOT,
+				   DRIVER_NAME, smo8800);
+	if (err) {
+		dev_err(&device->dev,
+			"failed to request thread for IRQ %d: %d\n",
+			smo8800->irq, err);
+		goto error;
+	}
+
+	dev_dbg(&device->dev, "device /dev/freefall registered with IRQ %d\n",
+		 smo8800->irq);
+	return 0;
+
+error:
+	misc_deregister(&smo8800->miscdev);
+	return err;
+}
+
+static int smo8800_remove(struct acpi_device *device)
+{
+	struct smo8800_device *smo8800 = device->driver_data;
+
+	free_irq(smo8800->irq, smo8800);
+	misc_deregister(&smo8800->miscdev);
+	dev_dbg(&device->dev, "device /dev/freefall unregistered\n");
+	return 0;
+}
+
+static const struct acpi_device_id smo8800_ids[] = {
+	{ "SMO8800", 0 },
+	{ "SMO8810", 0 },
+	{ "", 0 },
+};
+
+MODULE_DEVICE_TABLE(acpi, smo8800_ids);
+
+static struct acpi_driver smo8800_driver = {
+	.name = DRIVER_NAME,
+	.class = "Latitude",
+	.ids = smo8800_ids,
+	.ops = {
+		.add = smo8800_add,
+		.remove = smo8800_remove,
+	},
+	.owner = THIS_MODULE,
+};
+
+module_acpi_driver(smo8800_driver);
+
+MODULE_DESCRIPTION("Dell Latitude freefall driver (ACPI SMO8800/SMO8810)");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Sonal Santan, Pali Rohár");
diff --git a/drivers/platform/x86/hp-wmi.c b/drivers/platform/x86/hp-wmi.c
index 8ba8956b5a48..484a8673b835 100644
--- a/drivers/platform/x86/hp-wmi.c
+++ b/drivers/platform/x86/hp-wmi.c
@@ -53,6 +53,7 @@ MODULE_ALIAS("wmi:5FB7F034-2C63-45e9-BE91-3D44E2C707E4");
 #define HPWMI_ALS_QUERY 0x3
 #define HPWMI_HARDWARE_QUERY 0x4
 #define HPWMI_WIRELESS_QUERY 0x5
+#define HPWMI_BIOS_QUERY 0x9
 #define HPWMI_HOTKEY_QUERY 0xc
 #define HPWMI_FEATURE_QUERY 0xd
 #define HPWMI_WIRELESS2_QUERY 0x1b
@@ -144,6 +145,7 @@ static const struct key_entry hp_wmi_keymap[] = {
 	{ KE_KEY, 0x2142, { KEY_MEDIA } },
 	{ KE_KEY, 0x213b, { KEY_INFO } },
 	{ KE_KEY, 0x2169, { KEY_DIRECTION } },
+	{ KE_KEY, 0x216a, { KEY_SETUP } },
 	{ KE_KEY, 0x231b, { KEY_HELP } },
 	{ KE_END, 0 }
 };
@@ -304,6 +306,19 @@ static int hp_wmi_bios_2009_later(void)
 	return (state & 0x10) ? 1 : 0;
 }
 
+static int hp_wmi_enable_hotkeys(void)
+{
+	int ret;
+	int query = 0x6e;
+
+	ret = hp_wmi_perform_query(HPWMI_BIOS_QUERY, 1, &query, sizeof(query),
+				   0);
+
+	if (ret)
+		return -EINVAL;
+	return 0;
+}
+
 static int hp_wmi_set_block(void *data, bool blocked)
 {
 	enum hp_wmi_radio r = (enum hp_wmi_radio) data;
@@ -648,6 +663,9 @@ static int __init hp_wmi_input_setup(void)
 			    hp_wmi_tablet_state());
 	input_sync(hp_wmi_input_dev);
 
+	if (hp_wmi_bios_2009_later() == 4)
+		hp_wmi_enable_hotkeys();
+
 	status = wmi_install_notify_handler(HPWMI_EVENT_GUID, hp_wmi_notify, NULL);
 	if (ACPI_FAILURE(status)) {
 		err = -EIO;
diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c
index 6dd060a0bb65..b4c495a62eec 100644
--- a/drivers/platform/x86/ideapad-laptop.c
+++ b/drivers/platform/x86/ideapad-laptop.c
@@ -36,6 +36,8 @@
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 #include <linux/i8042.h>
+#include <linux/dmi.h>
+#include <linux/device.h>
 
 #define IDEAPAD_RFKILL_DEV_NUM	(3)
 
@@ -819,6 +821,19 @@ static void ideapad_acpi_notify(acpi_handle handle, u32 event, void *data)
 	}
 }
 
+/* Blacklist for devices where the ideapad rfkill interface does not work */
+static struct dmi_system_id rfkill_blacklist[] = {
+	/* The Lenovo Yoga 2 11 always reports everything as blocked */
+	{
+		.ident = "Lenovo Yoga 2 11",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo Yoga 2 11"),
+		},
+	},
+	{}
+};
+
 static int ideapad_acpi_add(struct platform_device *pdev)
 {
 	int ret, i;
@@ -833,7 +848,7 @@ static int ideapad_acpi_add(struct platform_device *pdev)
 	if (read_method_int(adev->handle, "_CFG", &cfg))
 		return -ENODEV;
 
-	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;
 
@@ -844,7 +859,7 @@ static int ideapad_acpi_add(struct platform_device *pdev)
 
 	ret = ideapad_sysfs_init(priv);
 	if (ret)
-		goto sysfs_failed;
+		return ret;
 
 	ret = ideapad_debugfs_init(priv);
 	if (ret)
@@ -854,11 +869,10 @@ static int ideapad_acpi_add(struct platform_device *pdev)
 	if (ret)
 		goto input_failed;
 
-	for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) {
-		if (test_bit(ideapad_rfk_data[i].cfgbit, &priv->cfg))
-			ideapad_register_rfkill(priv, i);
-		else
-			priv->rfk[i] = NULL;
+	if (!dmi_check_system(rfkill_blacklist)) {
+		for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++)
+			if (test_bit(ideapad_rfk_data[i].cfgbit, &priv->cfg))
+				ideapad_register_rfkill(priv, i);
 	}
 	ideapad_sync_rfk_state(priv);
 	ideapad_sync_touchpad_state(priv);
@@ -884,8 +898,6 @@ input_failed:
 	ideapad_debugfs_exit(priv);
 debugfs_failed:
 	ideapad_sysfs_exit(priv);
-sysfs_failed:
-	kfree(priv);
 	return ret;
 }
 
@@ -903,7 +915,6 @@ static int ideapad_acpi_remove(struct platform_device *pdev)
 	ideapad_debugfs_exit(priv);
 	ideapad_sysfs_exit(priv);
 	dev_set_drvdata(&pdev->dev, NULL);
-	kfree(priv);
 
 	return 0;
 }
diff --git a/drivers/platform/x86/intel_mid_thermal.c b/drivers/platform/x86/intel_mid_thermal.c
index 93fab8b70ce1..ab7860a21a22 100644
--- a/drivers/platform/x86/intel_mid_thermal.c
+++ b/drivers/platform/x86/intel_mid_thermal.c
@@ -481,7 +481,8 @@ static int mid_thermal_probe(struct platform_device *pdev)
 	int i;
 	struct platform_info *pinfo;
 
-	pinfo = kzalloc(sizeof(struct platform_info), GFP_KERNEL);
+	pinfo = devm_kzalloc(&pdev->dev, sizeof(struct platform_info),
+			     GFP_KERNEL);
 	if (!pinfo)
 		return -ENOMEM;
 
@@ -489,7 +490,6 @@ static int mid_thermal_probe(struct platform_device *pdev)
 	ret = mid_initialize_adc(&pdev->dev);
 	if (ret) {
 		dev_err(&pdev->dev, "ADC init failed");
-		kfree(pinfo);
 		return ret;
 	}
 
@@ -520,7 +520,6 @@ err:
 		thermal_zone_device_unregister(pinfo->tzd[i]);
 	}
 	configure_adc(0);
-	kfree(pinfo);
 	return ret;
 }
 
@@ -541,8 +540,6 @@ static int mid_thermal_remove(struct platform_device *pdev)
 		thermal_zone_device_unregister(pinfo->tzd[i]);
 	}
 
-	kfree(pinfo);
-
 	/* Stop the ADC */
 	return configure_adc(0);
 }
diff --git a/drivers/platform/x86/intel_pmic_gpio.c b/drivers/platform/x86/intel_pmic_gpio.c
index 2805988485f6..40929e4f7ad7 100644
--- a/drivers/platform/x86/intel_pmic_gpio.c
+++ b/drivers/platform/x86/intel_pmic_gpio.c
@@ -91,7 +91,7 @@ static void pmic_program_irqtype(int gpio, int type)
 
 static int pmic_gpio_direction_input(struct gpio_chip *chip, unsigned offset)
 {
-	if (offset > 8) {
+	if (offset >= 8) {
 		pr_err("only pin 0-7 support input\n");
 		return -1;/* we only have 8 GPIO can use as input */
 	}
@@ -130,7 +130,7 @@ static int pmic_gpio_get(struct gpio_chip *chip, unsigned offset)
 	int ret;
 
 	/* we only have 8 GPIO pins we can use as input */
-	if (offset > 8)
+	if (offset >= 8)
 		return -EOPNOTSUPP;
 	ret = intel_scu_ipc_ioread8(GPIO0 + offset, &r);
 	if (ret < 0)
diff --git a/drivers/platform/x86/pvpanic.c b/drivers/platform/x86/pvpanic.c
index c9f6e511daa6..073a90a63dbc 100644
--- a/drivers/platform/x86/pvpanic.c
+++ b/drivers/platform/x86/pvpanic.c
@@ -70,6 +70,7 @@ pvpanic_panic_notify(struct notifier_block *nb, unsigned long code,
 
 static struct notifier_block pvpanic_panic_nb = {
 	.notifier_call = pvpanic_panic_notify,
+	.priority = 1, /* let this called before broken drm_fb_helper */
 };
 
 
diff --git a/drivers/platform/x86/samsung-laptop.c b/drivers/platform/x86/samsung-laptop.c
index d1f030053176..5a5966512277 100644
--- a/drivers/platform/x86/samsung-laptop.c
+++ b/drivers/platform/x86/samsung-laptop.c
@@ -27,6 +27,7 @@
 #include <linux/debugfs.h>
 #include <linux/ctype.h>
 #include <linux/efi.h>
+#include <linux/suspend.h>
 #include <acpi/video.h>
 
 /*
@@ -340,6 +341,8 @@ struct samsung_laptop {
 	struct samsung_laptop_debug debug;
 	struct samsung_quirks *quirks;
 
+	struct notifier_block pm_nb;
+
 	bool handle_backlight;
 	bool has_stepping_quirk;
 
@@ -348,6 +351,8 @@ struct samsung_laptop {
 
 struct samsung_quirks {
 	bool broken_acpi_video;
+	bool four_kbd_backlight_levels;
+	bool enable_kbd_backlight;
 };
 
 static struct samsung_quirks samsung_unknown = {};
@@ -356,6 +361,11 @@ static struct samsung_quirks samsung_broken_acpi_video = {
 	.broken_acpi_video = true,
 };
 
+static struct samsung_quirks samsung_np740u3e = {
+	.four_kbd_backlight_levels = true,
+	.enable_kbd_backlight = true,
+};
+
 static bool force;
 module_param(force, bool, 0);
 MODULE_PARM_DESC(force,
@@ -1051,6 +1061,8 @@ static int __init samsung_leds_init(struct samsung_laptop *samsung)
 		samsung->kbd_led.brightness_set = kbd_led_set;
 		samsung->kbd_led.brightness_get = kbd_led_get;
 		samsung->kbd_led.max_brightness = 8;
+		if (samsung->quirks->four_kbd_backlight_levels)
+			samsung->kbd_led.max_brightness = 4;
 
 		ret = led_classdev_register(&samsung->platform_device->dev,
 					   &samsung->kbd_led);
@@ -1414,6 +1426,19 @@ static void samsung_platform_exit(struct samsung_laptop *samsung)
 	}
 }
 
+static int samsung_pm_notification(struct notifier_block *nb,
+				   unsigned long val, void *ptr)
+{
+	struct samsung_laptop *samsung;
+
+	samsung = container_of(nb, struct samsung_laptop, pm_nb);
+	if (val == PM_POST_HIBERNATION &&
+	    samsung->quirks->enable_kbd_backlight)
+		kbd_backlight_enable(samsung);
+
+	return 0;
+}
+
 static int __init samsung_platform_init(struct samsung_laptop *samsung)
 {
 	struct platform_device *pdev;
@@ -1534,6 +1559,15 @@ static struct dmi_system_id __initdata samsung_dmi_table[] = {
 		},
 	 .driver_data = &samsung_broken_acpi_video,
 	},
+	{
+	 .callback = samsung_dmi_matched,
+	 .ident = "730U3E/740U3E",
+	 .matches = {
+		DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."),
+		DMI_MATCH(DMI_PRODUCT_NAME, "730U3E/740U3E"),
+		},
+	 .driver_data = &samsung_np740u3e,
+	},
 	{ },
 };
 MODULE_DEVICE_TABLE(dmi, samsung_dmi_table);
@@ -1608,6 +1642,9 @@ static int __init samsung_init(void)
 	if (ret)
 		goto error_debugfs;
 
+	samsung->pm_nb.notifier_call = samsung_pm_notification;
+	register_pm_notifier(&samsung->pm_nb);
+
 	samsung_platform_device = samsung->platform_device;
 	return ret;
 
@@ -1633,6 +1670,7 @@ static void __exit samsung_exit(void)
 	struct samsung_laptop *samsung;
 
 	samsung = platform_get_drvdata(samsung_platform_device);
+	unregister_pm_notifier(&samsung->pm_nb);
 
 	samsung_debugfs_exit(samsung);
 	samsung_leds_exit(samsung);
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 15e61c16736e..d82f196e3cfe 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -3171,8 +3171,10 @@ static int __init hotkey_init(struct ibm_init_struct *iibm)
 		KEY_MICMUTE,	/* 0x1a: Mic mute (since ?400 or so) */
 
 		/* (assignments unknown, please report if found) */
-		KEY_UNKNOWN, KEY_UNKNOWN, KEY_UNKNOWN, KEY_UNKNOWN,
 		KEY_UNKNOWN,
+
+		/* Extra keys in use since the X240 / T440 / T540 */
+		KEY_CONFIG, KEY_SEARCH, KEY_SCALE, KEY_COMPUTER,
 		},
 	};
 
diff --git a/drivers/platform/x86/toshiba_acpi.c b/drivers/platform/x86/toshiba_acpi.c
index 46473ca7566b..76441dcbe5ff 100644
--- a/drivers/platform/x86/toshiba_acpi.c
+++ b/drivers/platform/x86/toshiba_acpi.c
@@ -56,6 +56,7 @@
 #include <linux/workqueue.h>
 #include <linux/i8042.h>
 #include <linux/acpi.h>
+#include <linux/dmi.h>
 #include <asm/uaccess.h>
 
 MODULE_AUTHOR("John Belmonte");
@@ -213,6 +214,30 @@ static const struct key_entry toshiba_acpi_keymap[] = {
 	{ KE_END, 0 },
 };
 
+/* alternative keymap */
+static const struct dmi_system_id toshiba_alt_keymap_dmi[] = {
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "TOSHIBA"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Satellite M840"),
+		},
+	},
+	{}
+};
+
+static const struct key_entry toshiba_acpi_alt_keymap[] = {
+	{ KE_KEY, 0x157, { KEY_MUTE } },
+	{ KE_KEY, 0x102, { KEY_ZOOMOUT } },
+	{ KE_KEY, 0x103, { KEY_ZOOMIN } },
+	{ KE_KEY, 0x139, { KEY_ZOOMRESET } },
+	{ KE_KEY, 0x13e, { KEY_SWITCHVIDEOMODE } },
+	{ KE_KEY, 0x13c, { KEY_BRIGHTNESSDOWN } },
+	{ KE_KEY, 0x13d, { KEY_BRIGHTNESSUP } },
+	{ KE_KEY, 0x158, { KEY_WLAN } },
+	{ KE_KEY, 0x13f, { KEY_TOUCHPAD_TOGGLE } },
+	{ KE_END, 0 },
+};
+
 /* utility
  */
 
@@ -1440,6 +1465,7 @@ static int toshiba_acpi_setup_keyboard(struct toshiba_acpi_dev *dev)
 	acpi_handle ec_handle;
 	int error;
 	u32 hci_result;
+	const struct key_entry *keymap = toshiba_acpi_keymap;
 
 	dev->hotkey_dev = input_allocate_device();
 	if (!dev->hotkey_dev)
@@ -1449,7 +1475,9 @@ static int toshiba_acpi_setup_keyboard(struct toshiba_acpi_dev *dev)
 	dev->hotkey_dev->phys = "toshiba_acpi/input0";
 	dev->hotkey_dev->id.bustype = BUS_HOST;
 
-	error = sparse_keymap_setup(dev->hotkey_dev, toshiba_acpi_keymap, NULL);
+	if (dmi_check_system(toshiba_alt_keymap_dmi))
+		keymap = toshiba_acpi_alt_keymap;
+	error = sparse_keymap_setup(dev->hotkey_dev, keymap, NULL);
 	if (error)
 		goto err_free_dev;
 
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index 6aea373547f6..ee3de3421f2d 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -74,7 +74,7 @@ config DP83640_PHY
 
 config PTP_1588_CLOCK_PCH
 	tristate "Intel PCH EG20T as PTP clock"
-	depends on X86 || COMPILE_TEST
+	depends on X86_32 || COMPILE_TEST
 	depends on HAS_IOMEM && NET
 	select PTP_1588_CLOCK
 	help
diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
index e25d2bc898e5..296b0ec8744d 100644
--- a/drivers/ptp/ptp_clock.c
+++ b/drivers/ptp/ptp_clock.c
@@ -142,7 +142,10 @@ static int ptp_clock_adjtime(struct posix_clock *pc, struct timex *tx)
 		delta = ktime_to_ns(kt);
 		err = ops->adjtime(ops, delta);
 	} else if (tx->modes & ADJ_FREQUENCY) {
-		err = ops->adjfreq(ops, scaled_ppm_to_ppb(tx->freq));
+		s32 ppb = scaled_ppm_to_ppb(tx->freq);
+		if (ppb > ops->max_adj || ppb < -ops->max_adj)
+			return -ERANGE;
+		err = ops->adjfreq(ops, ppb);
 		ptp->dialed_frequency = tx->freq;
 	} else if (tx->modes == 0) {
 		tx->freq = ptp->dialed_frequency;
diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig
index 5b34ff29ea38..4ad7b89a4cb4 100644
--- a/drivers/pwm/Kconfig
+++ b/drivers/pwm/Kconfig
@@ -62,6 +62,15 @@ config PWM_ATMEL_TCB
 	  To compile this driver as a module, choose M here: the module
 	  will be called pwm-atmel-tcb.
 
+config PWM_BCM_KONA
+	tristate "Kona PWM support"
+	depends on ARCH_BCM_MOBILE
+	help
+	  Generic PWM framework driver for Broadcom Kona PWM block.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called pwm-bcm-kona.
+
 config PWM_BFIN
 	tristate "Blackfin PWM support"
 	depends on BFIN_GPTIMERS
diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile
index e57d2c38a794..5c86a19d5d39 100644
--- a/drivers/pwm/Makefile
+++ b/drivers/pwm/Makefile
@@ -3,6 +3,7 @@ obj-$(CONFIG_PWM_SYSFS)		+= sysfs.o
 obj-$(CONFIG_PWM_AB8500)	+= pwm-ab8500.o
 obj-$(CONFIG_PWM_ATMEL)		+= pwm-atmel.o
 obj-$(CONFIG_PWM_ATMEL_TCB)	+= pwm-atmel-tcb.o
+obj-$(CONFIG_PWM_BCM_KONA)	+= pwm-bcm-kona.o
 obj-$(CONFIG_PWM_BFIN)		+= pwm-bfin.o
 obj-$(CONFIG_PWM_CLPS711X)	+= pwm-clps711x.o
 obj-$(CONFIG_PWM_EP93XX)	+= pwm-ep93xx.o
diff --git a/drivers/pwm/core.c b/drivers/pwm/core.c
index a80471399c20..4b66bf09ee55 100644
--- a/drivers/pwm/core.c
+++ b/drivers/pwm/core.c
@@ -661,10 +661,16 @@ struct pwm_device *pwm_get(struct device *dev, const char *con_id)
 		}
 	}
 
+	mutex_unlock(&pwm_lookup_lock);
+
 	if (chip)
 		pwm = pwm_request_from_chip(chip, index, con_id ?: dev_id);
+	if (IS_ERR(pwm))
+		return pwm;
+
+	pwm_set_period(pwm, p->period);
+	pwm_set_polarity(pwm, p->polarity);
 
-	mutex_unlock(&pwm_lookup_lock);
 
 	return pwm;
 }
diff --git a/drivers/pwm/pwm-ab8500.c b/drivers/pwm/pwm-ab8500.c
index 1d07a6f99375..4c07a8420b37 100644
--- a/drivers/pwm/pwm-ab8500.c
+++ b/drivers/pwm/pwm-ab8500.c
@@ -20,10 +20,6 @@
 #define AB8500_PWM_OUT_CTRL2_REG	0x61
 #define AB8500_PWM_OUT_CTRL7_REG	0x66
 
-/* backlight driver constants */
-#define ENABLE_PWM			1
-#define DISABLE_PWM			0
-
 struct ab8500_pwm_chip {
 	struct pwm_chip chip;
 };
@@ -64,7 +60,7 @@ static int ab8500_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm)
 
 	ret = abx500_mask_and_set_register_interruptible(chip->dev,
 				AB8500_MISC, AB8500_PWM_OUT_CTRL7_REG,
-				1 << (chip->base - 1), ENABLE_PWM);
+				1 << (chip->base - 1), 1 << (chip->base - 1));
 	if (ret < 0)
 		dev_err(chip->dev, "%s: Failed to enable PWM, Error %d\n",
 							pwm->label, ret);
@@ -77,11 +73,10 @@ static void ab8500_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm)
 
 	ret = abx500_mask_and_set_register_interruptible(chip->dev,
 				AB8500_MISC, AB8500_PWM_OUT_CTRL7_REG,
-				1 << (chip->base - 1), DISABLE_PWM);
+				1 << (chip->base - 1), 0);
 	if (ret < 0)
 		dev_err(chip->dev, "%s: Failed to disable PWM, Error %d\n",
 							pwm->label, ret);
-	return;
 }
 
 static const struct pwm_ops ab8500_pwm_ops = {
@@ -101,10 +96,8 @@ static int ab8500_pwm_probe(struct platform_device *pdev)
 	 * device which is required for ab8500 read and write
 	 */
 	ab8500 = devm_kzalloc(&pdev->dev, sizeof(*ab8500), GFP_KERNEL);
-	if (ab8500 == NULL) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (ab8500 == NULL)
 		return -ENOMEM;
-	}
 
 	ab8500->chip.dev = &pdev->dev;
 	ab8500->chip.ops = &ab8500_pwm_ops;
diff --git a/drivers/pwm/pwm-atmel.c b/drivers/pwm/pwm-atmel.c
index 0adc952cc4ef..6e700a541ca3 100644
--- a/drivers/pwm/pwm-atmel.c
+++ b/drivers/pwm/pwm-atmel.c
@@ -357,6 +357,7 @@ static int atmel_pwm_probe(struct platform_device *pdev)
 
 	atmel_pwm->chip.base = -1;
 	atmel_pwm->chip.npwm = 4;
+	atmel_pwm->chip.can_sleep = true;
 	atmel_pwm->config = data->config;
 
 	ret = pwmchip_add(&atmel_pwm->chip);
diff --git a/drivers/pwm/pwm-bcm-kona.c b/drivers/pwm/pwm-bcm-kona.c
new file mode 100644
index 000000000000..02bc048892a9
--- /dev/null
+++ b/drivers/pwm/pwm-bcm-kona.c
@@ -0,0 +1,318 @@
+/*
+ * Copyright (C) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/ioport.h>
+#include <linux/math64.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/pwm.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+/*
+ * The Kona PWM has some unusual characteristics.  Here are the main points.
+ *
+ * 1) There is no disable bit and the hardware docs advise programming a zero
+ *    duty to achieve output equivalent to that of a normal disable operation.
+ *
+ * 2) Changes to prescale, duty, period, and polarity do not take effect until
+ *    a subsequent rising edge of the trigger bit.
+ *
+ * 3) If the smooth bit and trigger bit are both low, the output is a constant
+ *    high signal.  Otherwise, the earlier waveform continues to be output.
+ *
+ * 4) If the smooth bit is set on the rising edge of the trigger bit, output
+ *    will transition to the new settings on a period boundary (which could be
+ *    seconds away).  If the smooth bit is clear, new settings will be applied
+ *    as soon as possible (the hardware always has a 400ns delay).
+ *
+ * 5) When the external clock that feeds the PWM is disabled, output is pegged
+ *    high or low depending on its state at that exact instant.
+ */
+
+#define PWM_CONTROL_OFFSET			(0x00000000)
+#define PWM_CONTROL_SMOOTH_SHIFT(chan)		(24 + (chan))
+#define PWM_CONTROL_TYPE_SHIFT(chan)		(16 + (chan))
+#define PWM_CONTROL_POLARITY_SHIFT(chan)	(8 + (chan))
+#define PWM_CONTROL_TRIGGER_SHIFT(chan)		(chan)
+
+#define PRESCALE_OFFSET				(0x00000004)
+#define PRESCALE_SHIFT(chan)			((chan) << 2)
+#define PRESCALE_MASK(chan)			(0x7 << PRESCALE_SHIFT(chan))
+#define PRESCALE_MIN				(0x00000000)
+#define PRESCALE_MAX				(0x00000007)
+
+#define PERIOD_COUNT_OFFSET(chan)		(0x00000008 + ((chan) << 3))
+#define PERIOD_COUNT_MIN			(0x00000002)
+#define PERIOD_COUNT_MAX			(0x00ffffff)
+
+#define DUTY_CYCLE_HIGH_OFFSET(chan)		(0x0000000c + ((chan) << 3))
+#define DUTY_CYCLE_HIGH_MIN			(0x00000000)
+#define DUTY_CYCLE_HIGH_MAX			(0x00ffffff)
+
+struct kona_pwmc {
+	struct pwm_chip chip;
+	void __iomem *base;
+	struct clk *clk;
+};
+
+static inline struct kona_pwmc *to_kona_pwmc(struct pwm_chip *_chip)
+{
+	return container_of(_chip, struct kona_pwmc, chip);
+}
+
+static void kona_pwmc_apply_settings(struct kona_pwmc *kp, unsigned int chan)
+{
+	unsigned int value = readl(kp->base + PWM_CONTROL_OFFSET);
+
+	/* Clear trigger bit but set smooth bit to maintain old output */
+	value |= 1 << PWM_CONTROL_SMOOTH_SHIFT(chan);
+	value &= ~(1 << PWM_CONTROL_TRIGGER_SHIFT(chan));
+	writel(value, kp->base + PWM_CONTROL_OFFSET);
+
+	/* Set trigger bit and clear smooth bit to apply new settings */
+	value &= ~(1 << PWM_CONTROL_SMOOTH_SHIFT(chan));
+	value |= 1 << PWM_CONTROL_TRIGGER_SHIFT(chan);
+	writel(value, kp->base + PWM_CONTROL_OFFSET);
+}
+
+static int kona_pwmc_config(struct pwm_chip *chip, struct pwm_device *pwm,
+			    int duty_ns, int period_ns)
+{
+	struct kona_pwmc *kp = to_kona_pwmc(chip);
+	u64 val, div, rate;
+	unsigned long prescale = PRESCALE_MIN, pc, dc;
+	unsigned int value, chan = pwm->hwpwm;
+
+	/*
+	 * Find period count, duty count and prescale to suit duty_ns and
+	 * period_ns. This is done according to formulas described below:
+	 *
+	 * period_ns = 10^9 * (PRESCALE + 1) * PC / PWM_CLK_RATE
+	 * duty_ns = 10^9 * (PRESCALE + 1) * DC / PWM_CLK_RATE
+	 *
+	 * PC = (PWM_CLK_RATE * period_ns) / (10^9 * (PRESCALE + 1))
+	 * DC = (PWM_CLK_RATE * duty_ns) / (10^9 * (PRESCALE + 1))
+	 */
+
+	rate = clk_get_rate(kp->clk);
+
+	while (1) {
+		div = 1000000000;
+		div *= 1 + prescale;
+		val = rate * period_ns;
+		pc = div64_u64(val, div);
+		val = rate * duty_ns;
+		dc = div64_u64(val, div);
+
+		/* If duty_ns or period_ns are not achievable then return */
+		if (pc < PERIOD_COUNT_MIN || dc < DUTY_CYCLE_HIGH_MIN)
+			return -EINVAL;
+
+		/* If pc and dc are in bounds, the calculation is done */
+		if (pc <= PERIOD_COUNT_MAX && dc <= DUTY_CYCLE_HIGH_MAX)
+			break;
+
+		/* Otherwise, increase prescale and recalculate pc and dc */
+		if (++prescale > PRESCALE_MAX)
+			return -EINVAL;
+	}
+
+	/* If the PWM channel is enabled, write the settings to the HW */
+	if (test_bit(PWMF_ENABLED, &pwm->flags)) {
+		value = readl(kp->base + PRESCALE_OFFSET);
+		value &= ~PRESCALE_MASK(chan);
+		value |= prescale << PRESCALE_SHIFT(chan);
+		writel(value, kp->base + PRESCALE_OFFSET);
+
+		writel(pc, kp->base + PERIOD_COUNT_OFFSET(chan));
+
+		writel(dc, kp->base + DUTY_CYCLE_HIGH_OFFSET(chan));
+
+		kona_pwmc_apply_settings(kp, chan);
+	}
+
+	return 0;
+}
+
+static int kona_pwmc_set_polarity(struct pwm_chip *chip, struct pwm_device *pwm,
+				  enum pwm_polarity polarity)
+{
+	struct kona_pwmc *kp = to_kona_pwmc(chip);
+	unsigned int chan = pwm->hwpwm;
+	unsigned int value;
+	int ret;
+
+	ret = clk_prepare_enable(kp->clk);
+	if (ret < 0) {
+		dev_err(chip->dev, "failed to enable clock: %d\n", ret);
+		return ret;
+	}
+
+	value = readl(kp->base + PWM_CONTROL_OFFSET);
+
+	if (polarity == PWM_POLARITY_NORMAL)
+		value |= 1 << PWM_CONTROL_POLARITY_SHIFT(chan);
+	else
+		value &= ~(1 << PWM_CONTROL_POLARITY_SHIFT(chan));
+
+	writel(value, kp->base + PWM_CONTROL_OFFSET);
+
+	kona_pwmc_apply_settings(kp, chan);
+
+	/* Wait for waveform to settle before gating off the clock */
+	ndelay(400);
+
+	clk_disable_unprepare(kp->clk);
+
+	return 0;
+}
+
+static int kona_pwmc_enable(struct pwm_chip *chip, struct pwm_device *pwm)
+{
+	struct kona_pwmc *kp = to_kona_pwmc(chip);
+	int ret;
+
+	ret = clk_prepare_enable(kp->clk);
+	if (ret < 0) {
+		dev_err(chip->dev, "failed to enable clock: %d\n", ret);
+		return ret;
+	}
+
+	ret = kona_pwmc_config(chip, pwm, pwm->duty_cycle, pwm->period);
+	if (ret < 0) {
+		clk_disable_unprepare(kp->clk);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void kona_pwmc_disable(struct pwm_chip *chip, struct pwm_device *pwm)
+{
+	struct kona_pwmc *kp = to_kona_pwmc(chip);
+	unsigned int chan = pwm->hwpwm;
+
+	/* Simulate a disable by configuring for zero duty */
+	writel(0, kp->base + DUTY_CYCLE_HIGH_OFFSET(chan));
+	kona_pwmc_apply_settings(kp, chan);
+
+	/* Wait for waveform to settle before gating off the clock */
+	ndelay(400);
+
+	clk_disable_unprepare(kp->clk);
+}
+
+static const struct pwm_ops kona_pwm_ops = {
+	.config = kona_pwmc_config,
+	.set_polarity = kona_pwmc_set_polarity,
+	.enable = kona_pwmc_enable,
+	.disable = kona_pwmc_disable,
+	.owner = THIS_MODULE,
+};
+
+static int kona_pwmc_probe(struct platform_device *pdev)
+{
+	struct kona_pwmc *kp;
+	struct resource *res;
+	unsigned int chan;
+	unsigned int value = 0;
+	int ret = 0;
+
+	kp = devm_kzalloc(&pdev->dev, sizeof(*kp), GFP_KERNEL);
+	if (kp == NULL)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, kp);
+
+	kp->chip.dev = &pdev->dev;
+	kp->chip.ops = &kona_pwm_ops;
+	kp->chip.base = -1;
+	kp->chip.npwm = 6;
+	kp->chip.of_xlate = of_pwm_xlate_with_flags;
+	kp->chip.of_pwm_n_cells = 3;
+	kp->chip.can_sleep = true;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	kp->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(kp->base))
+		return PTR_ERR(kp->base);
+
+	kp->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(kp->clk)) {
+		dev_err(&pdev->dev, "failed to get clock: %ld\n",
+			PTR_ERR(kp->clk));
+		return PTR_ERR(kp->clk);
+	}
+
+	ret = clk_prepare_enable(kp->clk);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to enable clock: %d\n", ret);
+		return ret;
+	}
+
+	/* Set smooth mode, push/pull, and normal polarity for all channels */
+	for (chan = 0; chan < kp->chip.npwm; chan++) {
+		value |= (1 << PWM_CONTROL_SMOOTH_SHIFT(chan));
+		value |= (1 << PWM_CONTROL_TYPE_SHIFT(chan));
+		value |= (1 << PWM_CONTROL_POLARITY_SHIFT(chan));
+	}
+
+	writel(value, kp->base + PWM_CONTROL_OFFSET);
+
+	clk_disable_unprepare(kp->clk);
+
+	ret = pwmchip_add(&kp->chip);
+	if (ret < 0)
+		dev_err(&pdev->dev, "failed to add PWM chip: %d\n", ret);
+
+	return ret;
+}
+
+static int kona_pwmc_remove(struct platform_device *pdev)
+{
+	struct kona_pwmc *kp = platform_get_drvdata(pdev);
+	unsigned int chan;
+
+	for (chan = 0; chan < kp->chip.npwm; chan++)
+		if (test_bit(PWMF_ENABLED, &kp->chip.pwms[chan].flags))
+			clk_disable_unprepare(kp->clk);
+
+	return pwmchip_remove(&kp->chip);
+}
+
+static const struct of_device_id bcm_kona_pwmc_dt[] = {
+	{ .compatible = "brcm,kona-pwm" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, bcm_kona_pwmc_dt);
+
+static struct platform_driver kona_pwmc_driver = {
+	.driver = {
+		.name = "bcm-kona-pwm",
+		.of_match_table = bcm_kona_pwmc_dt,
+	},
+	.probe = kona_pwmc_probe,
+	.remove = kona_pwmc_remove,
+};
+module_platform_driver(kona_pwmc_driver);
+
+MODULE_AUTHOR("Broadcom Corporation <bcm-kernel-feedback-list@broadcom.com>");
+MODULE_AUTHOR("Tim Kryger <tkryger@broadcom.com>");
+MODULE_DESCRIPTION("Broadcom Kona PWM driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/pwm/pwm-fsl-ftm.c b/drivers/pwm/pwm-fsl-ftm.c
index 420169e96b5f..a18bc8fea385 100644
--- a/drivers/pwm/pwm-fsl-ftm.c
+++ b/drivers/pwm/pwm-fsl-ftm.c
@@ -454,6 +454,7 @@ static int fsl_pwm_probe(struct platform_device *pdev)
 	fpc->chip.of_pwm_n_cells = 3;
 	fpc->chip.base = -1;
 	fpc->chip.npwm = 8;
+	fpc->chip.can_sleep = true;
 
 	ret = pwmchip_add(&fpc->chip);
 	if (ret < 0) {
diff --git a/drivers/pwm/pwm-imx.c b/drivers/pwm/pwm-imx.c
index cc4773344874..d797c7b84c3f 100644
--- a/drivers/pwm/pwm-imx.c
+++ b/drivers/pwm/pwm-imx.c
@@ -241,10 +241,8 @@ static int imx_pwm_probe(struct platform_device *pdev)
 		return -ENODEV;
 
 	imx = devm_kzalloc(&pdev->dev, sizeof(*imx), GFP_KERNEL);
-	if (imx == NULL) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (imx == NULL)
 		return -ENOMEM;
-	}
 
 	imx->clk_per = devm_clk_get(&pdev->dev, "per");
 	if (IS_ERR(imx->clk_per)) {
diff --git a/drivers/pwm/pwm-lp3943.c b/drivers/pwm/pwm-lp3943.c
index a40b9c34e9ff..2c39b0e50fa4 100644
--- a/drivers/pwm/pwm-lp3943.c
+++ b/drivers/pwm/pwm-lp3943.c
@@ -278,6 +278,7 @@ static int lp3943_pwm_probe(struct platform_device *pdev)
 	lp3943_pwm->chip.dev = &pdev->dev;
 	lp3943_pwm->chip.ops = &lp3943_pwm_ops;
 	lp3943_pwm->chip.npwm = LP3943_NUM_PWMS;
+	lp3943_pwm->chip.can_sleep = true;
 
 	platform_set_drvdata(pdev, lp3943_pwm);
 
diff --git a/drivers/pwm/pwm-lpss.c b/drivers/pwm/pwm-lpss.c
index 449e372050a0..44ce6c6103ae 100644
--- a/drivers/pwm/pwm-lpss.c
+++ b/drivers/pwm/pwm-lpss.c
@@ -6,6 +6,7 @@
  * Author: Chew Kean Ho <kean.ho.chew@intel.com>
  * Author: Chang Rebecca Swee Fun <rebecca.swee.fun.chang@intel.com>
  * Author: Chew Chiau Ee <chiau.ee.chew@intel.com>
+ * Author: Alan Cox <alan@linux.intel.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -19,6 +20,9 @@
 #include <linux/module.h>
 #include <linux/pwm.h>
 #include <linux/platform_device.h>
+#include <linux/pci.h>
+
+static int pci_drv, plat_drv;	/* So we know which drivers registered */
 
 #define PWM				0x00000000
 #define PWM_ENABLE			BIT(31)
@@ -34,6 +38,16 @@ struct pwm_lpss_chip {
 	struct pwm_chip chip;
 	void __iomem *regs;
 	struct clk *clk;
+	unsigned long clk_rate;
+};
+
+struct pwm_lpss_boardinfo {
+	unsigned long clk_rate;
+};
+
+/* BayTrail */
+static const struct pwm_lpss_boardinfo byt_info = {
+	25000000
 };
 
 static inline struct pwm_lpss_chip *to_lpwm(struct pwm_chip *chip)
@@ -55,7 +69,7 @@ static int pwm_lpss_config(struct pwm_chip *chip, struct pwm_device *pwm,
 	/* The equation is: base_unit = ((freq / c) * 65536) + correction */
 	base_unit = freq * 65536;
 
-	c = clk_get_rate(lpwm->clk);
+	c = lpwm->clk_rate;
 	if (!c)
 		return -EINVAL;
 
@@ -113,52 +127,48 @@ static const struct pwm_ops pwm_lpss_ops = {
 	.owner = THIS_MODULE,
 };
 
-static const struct acpi_device_id pwm_lpss_acpi_match[] = {
-	{ "80860F09", 0 },
-	{ },
-};
-MODULE_DEVICE_TABLE(acpi, pwm_lpss_acpi_match);
-
-static int pwm_lpss_probe(struct platform_device *pdev)
+static struct pwm_lpss_chip *pwm_lpss_probe(struct device *dev,
+					    struct resource *r,
+					    const struct pwm_lpss_boardinfo *info)
 {
 	struct pwm_lpss_chip *lpwm;
-	struct resource *r;
 	int ret;
 
-	lpwm = devm_kzalloc(&pdev->dev, sizeof(*lpwm), GFP_KERNEL);
+	lpwm = devm_kzalloc(dev, sizeof(*lpwm), GFP_KERNEL);
 	if (!lpwm)
-		return -ENOMEM;
-
-	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		return ERR_PTR(-ENOMEM);
 
-	lpwm->regs = devm_ioremap_resource(&pdev->dev, r);
+	lpwm->regs = devm_ioremap_resource(dev, r);
 	if (IS_ERR(lpwm->regs))
-		return PTR_ERR(lpwm->regs);
-
-	lpwm->clk = devm_clk_get(&pdev->dev, NULL);
-	if (IS_ERR(lpwm->clk)) {
-		dev_err(&pdev->dev, "failed to get PWM clock\n");
-		return PTR_ERR(lpwm->clk);
+		return ERR_CAST(lpwm->regs);
+
+	if (info) {
+		lpwm->clk_rate = info->clk_rate;
+	} else {
+		lpwm->clk = devm_clk_get(dev, NULL);
+		if (IS_ERR(lpwm->clk)) {
+			dev_err(dev, "failed to get PWM clock\n");
+			return ERR_CAST(lpwm->clk);
+		}
+		lpwm->clk_rate = clk_get_rate(lpwm->clk);
 	}
 
-	lpwm->chip.dev = &pdev->dev;
+	lpwm->chip.dev = dev;
 	lpwm->chip.ops = &pwm_lpss_ops;
 	lpwm->chip.base = -1;
 	lpwm->chip.npwm = 1;
 
 	ret = pwmchip_add(&lpwm->chip);
 	if (ret) {
-		dev_err(&pdev->dev, "failed to add PWM chip: %d\n", ret);
-		return ret;
+		dev_err(dev, "failed to add PWM chip: %d\n", ret);
+		return ERR_PTR(ret);
 	}
 
-	platform_set_drvdata(pdev, lpwm);
-	return 0;
+	return lpwm;
 }
 
-static int pwm_lpss_remove(struct platform_device *pdev)
+static int pwm_lpss_remove(struct pwm_lpss_chip *lpwm)
 {
-	struct pwm_lpss_chip *lpwm = platform_get_drvdata(pdev);
 	u32 ctrl;
 
 	ctrl = readl(lpwm->regs + PWM);
@@ -167,15 +177,104 @@ static int pwm_lpss_remove(struct platform_device *pdev)
 	return pwmchip_remove(&lpwm->chip);
 }
 
-static struct platform_driver pwm_lpss_driver = {
+static int pwm_lpss_probe_pci(struct pci_dev *pdev,
+			      const struct pci_device_id *id)
+{
+	const struct pwm_lpss_boardinfo *info;
+	struct pwm_lpss_chip *lpwm;
+	int err;
+
+	err = pci_enable_device(pdev);
+	if (err < 0)
+		return err;
+
+	info = (struct pwm_lpss_boardinfo *)id->driver_data;
+	lpwm = pwm_lpss_probe(&pdev->dev, &pdev->resource[0], info);
+	if (IS_ERR(lpwm))
+		return PTR_ERR(lpwm);
+
+	pci_set_drvdata(pdev, lpwm);
+	return 0;
+}
+
+static void pwm_lpss_remove_pci(struct pci_dev *pdev)
+{
+	struct pwm_lpss_chip *lpwm = pci_get_drvdata(pdev);
+
+	pwm_lpss_remove(lpwm);
+	pci_disable_device(pdev);
+}
+
+static struct pci_device_id pwm_lpss_pci_ids[] = {
+	{ PCI_VDEVICE(INTEL, 0x0f08), (unsigned long)&byt_info},
+	{ PCI_VDEVICE(INTEL, 0x0f09), (unsigned long)&byt_info},
+	{ },
+};
+MODULE_DEVICE_TABLE(pci, pwm_lpss_pci_ids);
+
+static struct pci_driver pwm_lpss_driver_pci = {
+	.name = "pwm-lpss",
+	.id_table = pwm_lpss_pci_ids,
+	.probe = pwm_lpss_probe_pci,
+	.remove = pwm_lpss_remove_pci,
+};
+
+static int pwm_lpss_probe_platform(struct platform_device *pdev)
+{
+	struct pwm_lpss_chip *lpwm;
+	struct resource *r;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+	lpwm = pwm_lpss_probe(&pdev->dev, r, NULL);
+	if (IS_ERR(lpwm))
+		return PTR_ERR(lpwm);
+
+	platform_set_drvdata(pdev, lpwm);
+	return 0;
+}
+
+static int pwm_lpss_remove_platform(struct platform_device *pdev)
+{
+	struct pwm_lpss_chip *lpwm = platform_get_drvdata(pdev);
+
+	return pwm_lpss_remove(lpwm);
+}
+
+static const struct acpi_device_id pwm_lpss_acpi_match[] = {
+	{ "80860F09", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, pwm_lpss_acpi_match);
+
+static struct platform_driver pwm_lpss_driver_platform = {
 	.driver = {
 		.name = "pwm-lpss",
 		.acpi_match_table = pwm_lpss_acpi_match,
 	},
-	.probe = pwm_lpss_probe,
-	.remove = pwm_lpss_remove,
+	.probe = pwm_lpss_probe_platform,
+	.remove = pwm_lpss_remove_platform,
 };
-module_platform_driver(pwm_lpss_driver);
+
+static int __init pwm_init(void)
+{
+	pci_drv = pci_register_driver(&pwm_lpss_driver_pci);
+	plat_drv = platform_driver_register(&pwm_lpss_driver_platform);
+	if (pci_drv && plat_drv)
+		return pci_drv;
+
+	return 0;
+}
+module_init(pwm_init);
+
+static void __exit pwm_exit(void)
+{
+	if (!pci_drv)
+		pci_unregister_driver(&pwm_lpss_driver_pci);
+	if (!plat_drv)
+		platform_driver_unregister(&pwm_lpss_driver_platform);
+}
+module_exit(pwm_exit);
 
 MODULE_DESCRIPTION("PWM driver for Intel LPSS");
 MODULE_AUTHOR("Mika Westerberg <mika.westerberg@linux.intel.com>");
diff --git a/drivers/pwm/pwm-mxs.c b/drivers/pwm/pwm-mxs.c
index 9475bc7a6f97..4f1bb4e0a426 100644
--- a/drivers/pwm/pwm-mxs.c
+++ b/drivers/pwm/pwm-mxs.c
@@ -147,6 +147,7 @@ static int mxs_pwm_probe(struct platform_device *pdev)
 	mxs->chip.dev = &pdev->dev;
 	mxs->chip.ops = &mxs_pwm_ops;
 	mxs->chip.base = -1;
+	mxs->chip.can_sleep = true;
 	ret = of_property_read_u32(np, "fsl,pwm-number", &mxs->chip.npwm);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "failed to get pwm number: %d\n", ret);
diff --git a/drivers/pwm/pwm-pxa.c b/drivers/pwm/pwm-pxa.c
index cd356d870244..0b312ec420b6 100644
--- a/drivers/pwm/pwm-pxa.c
+++ b/drivers/pwm/pwm-pxa.c
@@ -179,10 +179,8 @@ static int pwm_probe(struct platform_device *pdev)
 		return -EINVAL;
 
 	pwm = devm_kzalloc(&pdev->dev, sizeof(*pwm), GFP_KERNEL);
-	if (pwm == NULL) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (pwm == NULL)
 		return -ENOMEM;
-	}
 
 	pwm->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(pwm->clk))
diff --git a/drivers/pwm/pwm-renesas-tpu.c b/drivers/pwm/pwm-renesas-tpu.c
index aff6ba9b49e7..3b71b42e89d5 100644
--- a/drivers/pwm/pwm-renesas-tpu.c
+++ b/drivers/pwm/pwm-renesas-tpu.c
@@ -21,13 +21,14 @@
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/of.h>
-#include <linux/platform_data/pwm-renesas-tpu.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/pwm.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
+#define TPU_CHANNEL_MAX		4
+
 #define TPU_TSTR		0x00	/* Timer start register (shared) */
 
 #define TPU_TCRn		0x00	/* Timer control register */
@@ -87,7 +88,6 @@ struct tpu_pwm_device {
 
 struct tpu_device {
 	struct platform_device *pdev;
-	enum pwm_polarity polarities[TPU_CHANNEL_MAX];
 	struct pwm_chip chip;
 	spinlock_t lock;
 
@@ -229,7 +229,7 @@ static int tpu_pwm_request(struct pwm_chip *chip, struct pwm_device *_pwm)
 
 	pwm->tpu = tpu;
 	pwm->channel = _pwm->hwpwm;
-	pwm->polarity = tpu->polarities[pwm->channel];
+	pwm->polarity = PWM_POLARITY_NORMAL;
 	pwm->prescaler = 0;
 	pwm->period = 0;
 	pwm->duty = 0;
@@ -388,16 +388,6 @@ static const struct pwm_ops tpu_pwm_ops = {
  * Probe and remove
  */
 
-static void tpu_parse_pdata(struct tpu_device *tpu)
-{
-	struct tpu_pwm_platform_data *pdata = tpu->pdev->dev.platform_data;
-	unsigned int i;
-
-	for (i = 0; i < ARRAY_SIZE(tpu->polarities); ++i)
-		tpu->polarities[i] = pdata ? pdata->channels[i].polarity
-				   : PWM_POLARITY_NORMAL;
-}
-
 static int tpu_probe(struct platform_device *pdev)
 {
 	struct tpu_device *tpu;
@@ -405,17 +395,12 @@ static int tpu_probe(struct platform_device *pdev)
 	int ret;
 
 	tpu = devm_kzalloc(&pdev->dev, sizeof(*tpu), GFP_KERNEL);
-	if (tpu == NULL) {
-		dev_err(&pdev->dev, "failed to allocate driver data\n");
+	if (tpu == NULL)
 		return -ENOMEM;
-	}
 
 	spin_lock_init(&tpu->lock);
 	tpu->pdev = pdev;
 
-	/* Initialize device configuration from platform data. */
-	tpu_parse_pdata(tpu);
-
 	/* Map memory, get clock and pin control. */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	tpu->base = devm_ioremap_resource(&pdev->dev, res);
diff --git a/drivers/pwm/pwm-samsung.c b/drivers/pwm/pwm-samsung.c
index d66529a995a1..ba6b650cf8dc 100644
--- a/drivers/pwm/pwm-samsung.c
+++ b/drivers/pwm/pwm-samsung.c
@@ -335,9 +335,6 @@ static int pwm_samsung_config(struct pwm_chip *chip, struct pwm_device *pwm,
 	writel(tcnt, our_chip->base + REG_TCNTB(pwm->hwpwm));
 	writel(tcmp, our_chip->base + REG_TCMPB(pwm->hwpwm));
 
-	if (test_bit(PWMF_ENABLED, &pwm->flags))
-		pwm_samsung_enable(chip, pwm);
-
 	chan->period_ns = period_ns;
 	chan->tin_ns = tin_ns;
 	chan->duty_ns = duty_ns;
diff --git a/drivers/pwm/pwm-spear.c b/drivers/pwm/pwm-spear.c
index cb2d4f0f9711..6fd93e6a4122 100644
--- a/drivers/pwm/pwm-spear.c
+++ b/drivers/pwm/pwm-spear.c
@@ -179,10 +179,8 @@ static int spear_pwm_probe(struct platform_device *pdev)
 	u32 val;
 
 	pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
-	if (!pc) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (!pc)
 		return -ENOMEM;
-	}
 
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	pc->mmio_base = devm_ioremap_resource(&pdev->dev, r);
@@ -222,7 +220,7 @@ static int spear_pwm_probe(struct platform_device *pdev)
 	}
 
 	ret = pwmchip_add(&pc->chip);
-	if (!ret) {
+	if (ret < 0) {
 		clk_unprepare(pc->clk);
 		dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret);
 	}
diff --git a/drivers/pwm/pwm-tegra.c b/drivers/pwm/pwm-tegra.c
index 74298c561c4e..61d86b9498ca 100644
--- a/drivers/pwm/pwm-tegra.c
+++ b/drivers/pwm/pwm-tegra.c
@@ -173,10 +173,8 @@ static int tegra_pwm_probe(struct platform_device *pdev)
 	int ret;
 
 	pwm = devm_kzalloc(&pdev->dev, sizeof(*pwm), GFP_KERNEL);
-	if (!pwm) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (!pwm)
 		return -ENOMEM;
-	}
 
 	pwm->dev = &pdev->dev;
 
diff --git a/drivers/pwm/pwm-tiecap.c b/drivers/pwm/pwm-tiecap.c
index 032092c7a6ae..74efbe7f20c3 100644
--- a/drivers/pwm/pwm-tiecap.c
+++ b/drivers/pwm/pwm-tiecap.c
@@ -209,10 +209,8 @@ static int ecap_pwm_probe(struct platform_device *pdev)
 	u16 status;
 
 	pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
-	if (!pc) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (!pc)
 		return -ENOMEM;
-	}
 
 	clk = devm_clk_get(&pdev->dev, "fck");
 	if (IS_ERR(clk)) {
diff --git a/drivers/pwm/pwm-tiehrpwm.c b/drivers/pwm/pwm-tiehrpwm.c
index aee4471424d1..cb75133085a8 100644
--- a/drivers/pwm/pwm-tiehrpwm.c
+++ b/drivers/pwm/pwm-tiehrpwm.c
@@ -138,12 +138,12 @@ static inline struct ehrpwm_pwm_chip *to_ehrpwm_pwm_chip(struct pwm_chip *chip)
 	return container_of(chip, struct ehrpwm_pwm_chip, chip);
 }
 
-static u16 ehrpwm_read(void __iomem *base, int offset)
+static inline u16 ehrpwm_read(void __iomem *base, int offset)
 {
 	return readw(base + offset);
 }
 
-static void ehrpwm_write(void __iomem *base, int offset, unsigned int val)
+static inline void ehrpwm_write(void __iomem *base, int offset, unsigned int val)
 {
 	writew(val & 0xFFFF, base + offset);
 }
@@ -440,10 +440,8 @@ static int ehrpwm_pwm_probe(struct platform_device *pdev)
 	u16 status;
 
 	pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
-	if (!pc) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (!pc)
 		return -ENOMEM;
-	}
 
 	clk = devm_clk_get(&pdev->dev, "fck");
 	if (IS_ERR(clk)) {
@@ -531,6 +529,7 @@ static int ehrpwm_pwm_remove(struct platform_device *pdev)
 	return pwmchip_remove(&pc->chip);
 }
 
+#ifdef CONFIG_PM_SLEEP
 static void ehrpwm_pwm_save_context(struct ehrpwm_pwm_chip *pc)
 {
 	pm_runtime_get_sync(pc->chip.dev);
@@ -557,7 +556,6 @@ static void ehrpwm_pwm_restore_context(struct ehrpwm_pwm_chip *pc)
 	ehrpwm_write(pc->mmio_base, TBCTL, pc->ctx.tbctl);
 }
 
-#ifdef CONFIG_PM_SLEEP
 static int ehrpwm_pwm_suspend(struct device *dev)
 {
 	struct ehrpwm_pwm_chip *pc = dev_get_drvdata(dev);
diff --git a/drivers/pwm/pwm-twl.c b/drivers/pwm/pwm-twl.c
index b99a50e626a6..04f76725d591 100644
--- a/drivers/pwm/pwm-twl.c
+++ b/drivers/pwm/pwm-twl.c
@@ -265,14 +265,6 @@ static void twl6030_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm)
 
 	ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_TOGGLE3_REG);
 	if (ret < 0) {
-		dev_err(chip->dev, "%s: Failed to read TOGGLE3\n", pwm->label);
-		goto out;
-	}
-
-	val |= TWL6030_PWM_TOGGLE(pwm->hwpwm, TWL6030_PWMXS | TWL6030_PWMXEN);
-
-	ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_TOGGLE3_REG);
-	if (ret < 0) {
 		dev_err(chip->dev, "%s: Failed to disable PWM\n", pwm->label);
 		goto out;
 	}
diff --git a/drivers/pwm/pwm-vt8500.c b/drivers/pwm/pwm-vt8500.c
index 323125abf3f4..652e6b5b859b 100644
--- a/drivers/pwm/pwm-vt8500.c
+++ b/drivers/pwm/pwm-vt8500.c
@@ -211,10 +211,8 @@ static int vt8500_pwm_probe(struct platform_device *pdev)
 	}
 
 	chip = devm_kzalloc(&pdev->dev, sizeof(*chip), GFP_KERNEL);
-	if (chip == NULL) {
-		dev_err(&pdev->dev, "failed to allocate memory\n");
+	if (chip == NULL)
 		return -ENOMEM;
-	}
 
 	chip->chip.dev = &pdev->dev;
 	chip->chip.ops = &vt8500_pwm_ops;
diff --git a/drivers/regulator/as3722-regulator.c b/drivers/regulator/as3722-regulator.c
index 85585219ce82..ad9e0c9b7daf 100644
--- a/drivers/regulator/as3722-regulator.c
+++ b/drivers/regulator/as3722-regulator.c
@@ -433,6 +433,7 @@ static struct regulator_ops as3722_ldo3_extcntrl_ops = {
 };
 
 static const struct regulator_linear_range as3722_ldo_ranges[] = {
+	REGULATOR_LINEAR_RANGE(0, 0x00, 0x00, 0),
 	REGULATOR_LINEAR_RANGE(825000, 0x01, 0x24, 25000),
 	REGULATOR_LINEAR_RANGE(1725000, 0x40, 0x7F, 25000),
 };
@@ -609,6 +610,7 @@ static bool as3722_sd0_is_low_voltage(struct as3722_regulators *as3722_regs)
 }
 
 static const struct regulator_linear_range as3722_sd2345_ranges[] = {
+	REGULATOR_LINEAR_RANGE(0, 0x00, 0x00, 0),
 	REGULATOR_LINEAR_RANGE(612500, 0x01, 0x40, 12500),
 	REGULATOR_LINEAR_RANGE(1425000, 0x41, 0x70, 25000),
 	REGULATOR_LINEAR_RANGE(2650000, 0x71, 0x7F, 50000),
diff --git a/drivers/regulator/bcm590xx-regulator.c b/drivers/regulator/bcm590xx-regulator.c
index 57544e254a78..58ece59367ae 100644
--- a/drivers/regulator/bcm590xx-regulator.c
+++ b/drivers/regulator/bcm590xx-regulator.c
@@ -119,6 +119,10 @@ static const unsigned int ldo_c_table[] = {
 	2900000, 3000000, 3300000,
 };
 
+static const unsigned int ldo_vbus[] = {
+	5000000,
+};
+
 /* DCDC group CSR: supported voltages in microvolts */
 static const struct regulator_linear_range dcdc_csr_ranges[] = {
 	REGULATOR_LINEAR_RANGE(860000, 2, 50, 10000),
@@ -192,6 +196,7 @@ static struct bcm590xx_info bcm590xx_regs[] = {
 	BCM590XX_REG_TABLE(gpldo4, ldo_a_table),
 	BCM590XX_REG_TABLE(gpldo5, ldo_a_table),
 	BCM590XX_REG_TABLE(gpldo6, ldo_a_table),
+	BCM590XX_REG_TABLE(vbus, ldo_vbus),
 };
 
 struct bcm590xx_reg {
diff --git a/drivers/regulator/ltc3589.c b/drivers/regulator/ltc3589.c
index 110a99ee1162..c8105182b8b8 100644
--- a/drivers/regulator/ltc3589.c
+++ b/drivers/regulator/ltc3589.c
@@ -255,7 +255,7 @@ static int ltc3589_parse_regulators_dt(struct ltc3589 *ltc3589)
 	struct device_node *node;
 	int i, ret;
 
-	node = of_find_node_by_name(dev->of_node, "regulators");
+	node = of_get_child_by_name(dev->of_node, "regulators");
 	if (!node) {
 		dev_err(dev, "regulators node not found\n");
 		return -EINVAL;
diff --git a/drivers/regulator/palmas-regulator.c b/drivers/regulator/palmas-regulator.c
index 864ed02ce4b7..93b4ad842901 100644
--- a/drivers/regulator/palmas-regulator.c
+++ b/drivers/regulator/palmas-regulator.c
@@ -37,12 +37,14 @@ struct regs_info {
 };
 
 static const struct regulator_linear_range smps_low_ranges[] = {
+	REGULATOR_LINEAR_RANGE(0, 0x0, 0x0, 0),
 	REGULATOR_LINEAR_RANGE(500000, 0x1, 0x6, 0),
 	REGULATOR_LINEAR_RANGE(510000, 0x7, 0x79, 10000),
 	REGULATOR_LINEAR_RANGE(1650000, 0x7A, 0x7f, 0),
 };
 
 static const struct regulator_linear_range smps_high_ranges[] = {
+	REGULATOR_LINEAR_RANGE(0, 0x0, 0x0, 0),
 	REGULATOR_LINEAR_RANGE(1000000, 0x1, 0x6, 0),
 	REGULATOR_LINEAR_RANGE(1020000, 0x7, 0x79, 20000),
 	REGULATOR_LINEAR_RANGE(3300000, 0x7A, 0x7f, 0),
@@ -323,6 +325,10 @@ static int palmas_set_mode_smps(struct regulator_dev *dev, unsigned int mode)
 	if (rail_enable)
 		palmas_smps_write(pmic->palmas,
 			palmas_regs_info[id].ctrl_addr, reg);
+
+	/* Switch the enable value to ensure this is used for enable */
+	pmic->desc[id].enable_val = pmic->current_reg_mode[id];
+
 	return 0;
 }
 
@@ -962,6 +968,14 @@ static int palmas_regulators_probe(struct platform_device *pdev)
 				return ret;
 			pmic->current_reg_mode[id] = reg &
 					PALMAS_SMPS12_CTRL_MODE_ACTIVE_MASK;
+
+			pmic->desc[id].enable_reg =
+					PALMAS_BASE_TO_REG(PALMAS_SMPS_BASE,
+						palmas_regs_info[id].ctrl_addr);
+			pmic->desc[id].enable_mask =
+					PALMAS_SMPS12_CTRL_MODE_ACTIVE_MASK;
+			/* set_mode overrides this value */
+			pmic->desc[id].enable_val = SMPS_CTRL_MODE_ON;
 		}
 
 		pmic->desc[id].type = REGULATOR_VOLTAGE;
diff --git a/drivers/regulator/tps65218-regulator.c b/drivers/regulator/tps65218-regulator.c
index 69b4b7750410..9effe48c605e 100644
--- a/drivers/regulator/tps65218-regulator.c
+++ b/drivers/regulator/tps65218-regulator.c
@@ -209,7 +209,7 @@ static const struct regulator_desc regulators[] = {
 			   1, -1, -1, TPS65218_REG_ENABLE1,
 			   TPS65218_ENABLE1_DC6_EN, NULL, NULL, 0, 0),
 	TPS65218_REGULATOR("LDO1", TPS65218_LDO_1, tps65218_ldo1_dcdc34_ops, 64,
-			   TPS65218_REG_CONTROL_DCDC4,
+			   TPS65218_REG_CONTROL_LDO1,
 			   TPS65218_CONTROL_LDO1_MASK, TPS65218_REG_ENABLE2,
 			   TPS65218_ENABLE2_LDO1_EN, NULL, ldo1_dcdc3_ranges,
 			   2, 0),
@@ -240,6 +240,7 @@ static int tps65218_regulator_probe(struct platform_device *pdev)
 	config.init_data = init_data;
 	config.driver_data = tps;
 	config.regmap = tps->regmap;
+	config.of_node = pdev->dev.of_node;
 
 	rdev = devm_regulator_register(&pdev->dev, &regulators[id], &config);
 	if (IS_ERR(rdev)) {
diff --git a/drivers/regulator/virtual.c b/drivers/regulator/virtual.c
index f53e78b9a84e..6ff95b045984 100644
--- a/drivers/regulator/virtual.c
+++ b/drivers/regulator/virtual.c
@@ -266,11 +266,11 @@ static ssize_t set_mode(struct device *dev, struct device_attribute *attr,
 	return count;
 }
 
-static DEVICE_ATTR(min_microvolts, 0666, show_min_uV, set_min_uV);
-static DEVICE_ATTR(max_microvolts, 0666, show_max_uV, set_max_uV);
-static DEVICE_ATTR(min_microamps, 0666, show_min_uA, set_min_uA);
-static DEVICE_ATTR(max_microamps, 0666, show_max_uA, set_max_uA);
-static DEVICE_ATTR(mode, 0666, show_mode, set_mode);
+static DEVICE_ATTR(min_microvolts, 0664, show_min_uV, set_min_uV);
+static DEVICE_ATTR(max_microvolts, 0664, show_max_uV, set_max_uV);
+static DEVICE_ATTR(min_microamps, 0664, show_min_uA, set_min_uA);
+static DEVICE_ATTR(max_microamps, 0664, show_max_uA, set_max_uA);
+static DEVICE_ATTR(mode, 0664, show_mode, set_mode);
 
 static struct attribute *regulator_virtual_attributes[] = {
 	&dev_attr_min_microvolts.attr,
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
index ce1743d0b679..5e343bab9458 100644
--- a/drivers/remoteproc/Kconfig
+++ b/drivers/remoteproc/Kconfig
@@ -44,7 +44,7 @@ config STE_MODEM_RPROC
 config DA8XX_REMOTEPROC
 	tristate "DA8xx/OMAP-L13x remoteproc support"
 	depends on ARCH_DAVINCI_DA8XX
-	select CMA
+	select CMA if MMU
 	select REMOTEPROC
 	select RPMSG
 	help
diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index 71988b69eca6..0754f5c7cb3b 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -530,11 +530,11 @@ config RTC_DRV_RV3029C2
 	  will be called rtc-rv3029c2.
 
 config RTC_DRV_S5M
-	tristate "Samsung S5M series"
+	tristate "Samsung S2M/S5M series"
 	depends on MFD_SEC_CORE
 	help
 	  If you say yes here you will get support for the
-	  RTC of Samsung S5M PMIC series.
+	  RTC of Samsung S2MPS14 and S5M PMIC series.
 
 	  This driver can also be built as a module. If so, the module
 	  will be called rtc-s5m.
diff --git a/drivers/rtc/rtc-puv3.c b/drivers/rtc/rtc-puv3.c
index 1ecfe3bd92ac..1cff2a21db67 100644
--- a/drivers/rtc/rtc-puv3.c
+++ b/drivers/rtc/rtc-puv3.c
@@ -71,7 +71,7 @@ static int puv3_rtc_setpie(struct device *dev, int enabled)
 {
 	unsigned int tmp;
 
-	dev_debug(dev, "%s: pie=%d\n", __func__, enabled);
+	dev_dbg(dev, "%s: pie=%d\n", __func__, enabled);
 
 	spin_lock_irq(&puv3_rtc_pie_lock);
 	tmp = readl(RTC_RTSR) & ~RTC_RTSR_HZE;
@@ -140,7 +140,7 @@ static int puv3_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm)
 	rtc_tm_to_time(tm, &rtcalarm_count);
 	writel(rtcalarm_count, RTC_RTAR);
 
-	puv3_rtc_setaie(&dev->dev, alrm->enabled);
+	puv3_rtc_setaie(dev, alrm->enabled);
 
 	if (alrm->enabled)
 		enable_irq_wake(puv3_rtc_alarmno);
diff --git a/drivers/rtc/rtc-s5m.c b/drivers/rtc/rtc-s5m.c
index 8ec2d6a1dbe1..8f06250a0389 100644
--- a/drivers/rtc/rtc-s5m.c
+++ b/drivers/rtc/rtc-s5m.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2013 Samsung Electronics Co., Ltd
+ * Copyright (c) 2013-2014 Samsung Electronics Co., Ltd
  *	http://www.samsung.com
  *
  *  Copyright (C) 2013 Google, Inc
@@ -17,27 +17,76 @@
 
 #include <linux/module.h>
 #include <linux/i2c.h>
-#include <linux/slab.h>
 #include <linux/bcd.h>
-#include <linux/bitops.h>
 #include <linux/regmap.h>
 #include <linux/rtc.h>
-#include <linux/delay.h>
 #include <linux/platform_device.h>
 #include <linux/mfd/samsung/core.h>
 #include <linux/mfd/samsung/irq.h>
 #include <linux/mfd/samsung/rtc.h>
+#include <linux/mfd/samsung/s2mps14.h>
 
 /*
  * Maximum number of retries for checking changes in UDR field
- * of SEC_RTC_UDR_CON register (to limit possible endless loop).
+ * of S5M_RTC_UDR_CON register (to limit possible endless loop).
  *
  * After writing to RTC registers (setting time or alarm) read the UDR field
- * in SEC_RTC_UDR_CON register. UDR is auto-cleared when data have
+ * in S5M_RTC_UDR_CON register. UDR is auto-cleared when data have
  * been transferred.
  */
 #define UDR_READ_RETRY_CNT	5
 
+/* Registers used by the driver which are different between chipsets. */
+struct s5m_rtc_reg_config {
+	/* Number of registers used for setting time/alarm0/alarm1 */
+	unsigned int regs_count;
+	/* First register for time, seconds */
+	unsigned int time;
+	/* RTC control register */
+	unsigned int ctrl;
+	/* First register for alarm 0, seconds */
+	unsigned int alarm0;
+	/* First register for alarm 1, seconds */
+	unsigned int alarm1;
+	/* SMPL/WTSR register */
+	unsigned int smpl_wtsr;
+	/*
+	 * Register for update flag (UDR). Typically setting UDR field to 1
+	 * will enable update of time or alarm register. Then it will be
+	 * auto-cleared after successful update.
+	 */
+	unsigned int rtc_udr_update;
+	/* Mask for UDR field in 'rtc_udr_update' register */
+	unsigned int rtc_udr_mask;
+};
+
+/* Register map for S5M8763 and S5M8767 */
+static const struct s5m_rtc_reg_config s5m_rtc_regs = {
+	.regs_count		= 8,
+	.time			= S5M_RTC_SEC,
+	.ctrl			= S5M_ALARM1_CONF,
+	.alarm0			= S5M_ALARM0_SEC,
+	.alarm1			= S5M_ALARM1_SEC,
+	.smpl_wtsr		= S5M_WTSR_SMPL_CNTL,
+	.rtc_udr_update		= S5M_RTC_UDR_CON,
+	.rtc_udr_mask		= S5M_RTC_UDR_MASK,
+};
+
+/*
+ * Register map for S2MPS14.
+ * It may be also suitable for S2MPS11 but this was not tested.
+ */
+static const struct s5m_rtc_reg_config s2mps_rtc_regs = {
+	.regs_count		= 7,
+	.time			= S2MPS_RTC_SEC,
+	.ctrl			= S2MPS_RTC_CTRL,
+	.alarm0			= S2MPS_ALARM0_SEC,
+	.alarm1			= S2MPS_ALARM1_SEC,
+	.smpl_wtsr		= S2MPS_WTSR_SMPL_CNTL,
+	.rtc_udr_update		= S2MPS_RTC_UDR_CON,
+	.rtc_udr_mask		= S2MPS_RTC_WUDR_MASK,
+};
+
 struct s5m_rtc_info {
 	struct device *dev;
 	struct i2c_client *i2c;
@@ -48,13 +97,14 @@ struct s5m_rtc_info {
 	int device_type;
 	int rtc_24hr_mode;
 	bool wtsr_smpl;
+	const struct s5m_rtc_reg_config	*regs;
 };
 
 static const struct regmap_config s5m_rtc_regmap_config = {
 	.reg_bits = 8,
 	.val_bits = 8,
 
-	.max_register = SEC_RTC_REG_MAX,
+	.max_register = S5M_RTC_REG_MAX,
 };
 
 static const struct regmap_config s2mps14_rtc_regmap_config = {
@@ -119,8 +169,9 @@ static inline int s5m8767_wait_for_udr_update(struct s5m_rtc_info *info)
 	unsigned int data;
 
 	do {
-		ret = regmap_read(info->regmap, SEC_RTC_UDR_CON, &data);
-	} while (--retry && (data & RTC_UDR_MASK) && !ret);
+		ret = regmap_read(info->regmap, info->regs->rtc_udr_update,
+				&data);
+	} while (--retry && (data & info->regs->rtc_udr_mask) && !ret);
 
 	if (!retry)
 		dev_err(info->dev, "waiting for UDR update, reached max number of retries\n");
@@ -128,21 +179,53 @@ static inline int s5m8767_wait_for_udr_update(struct s5m_rtc_info *info)
 	return ret;
 }
 
+static inline int s5m_check_peding_alarm_interrupt(struct s5m_rtc_info *info,
+		struct rtc_wkalrm *alarm)
+{
+	int ret;
+	unsigned int val;
+
+	switch (info->device_type) {
+	case S5M8767X:
+	case S5M8763X:
+		ret = regmap_read(info->regmap, S5M_RTC_STATUS, &val);
+		val &= S5M_ALARM0_STATUS;
+		break;
+	case S2MPS14X:
+		ret = regmap_read(info->s5m87xx->regmap_pmic, S2MPS14_REG_ST2,
+				&val);
+		val &= S2MPS_ALARM0_STATUS;
+		break;
+	default:
+		return -EINVAL;
+	}
+	if (ret < 0)
+		return ret;
+
+	if (val)
+		alarm->pending = 1;
+	else
+		alarm->pending = 0;
+
+	return 0;
+}
+
 static inline int s5m8767_rtc_set_time_reg(struct s5m_rtc_info *info)
 {
 	int ret;
 	unsigned int data;
 
-	ret = regmap_read(info->regmap, SEC_RTC_UDR_CON, &data);
+	ret = regmap_read(info->regmap, info->regs->rtc_udr_update, &data);
 	if (ret < 0) {
 		dev_err(info->dev, "failed to read update reg(%d)\n", ret);
 		return ret;
 	}
 
-	data |= RTC_TIME_EN_MASK;
-	data |= RTC_UDR_MASK;
+	data |= info->regs->rtc_udr_mask;
+	if (info->device_type == S5M8763X || info->device_type == S5M8767X)
+		data |= S5M_RTC_TIME_EN_MASK;
 
-	ret = regmap_write(info->regmap, SEC_RTC_UDR_CON, data);
+	ret = regmap_write(info->regmap, info->regs->rtc_udr_update, data);
 	if (ret < 0) {
 		dev_err(info->dev, "failed to write update reg(%d)\n", ret);
 		return ret;
@@ -158,17 +241,27 @@ static inline int s5m8767_rtc_set_alarm_reg(struct s5m_rtc_info *info)
 	int ret;
 	unsigned int data;
 
-	ret = regmap_read(info->regmap, SEC_RTC_UDR_CON, &data);
+	ret = regmap_read(info->regmap, info->regs->rtc_udr_update, &data);
 	if (ret < 0) {
 		dev_err(info->dev, "%s: fail to read update reg(%d)\n",
 			__func__, ret);
 		return ret;
 	}
 
-	data &= ~RTC_TIME_EN_MASK;
-	data |= RTC_UDR_MASK;
+	data |= info->regs->rtc_udr_mask;
+	switch (info->device_type) {
+	case S5M8763X:
+	case S5M8767X:
+		data &= ~S5M_RTC_TIME_EN_MASK;
+		break;
+	case S2MPS14X:
+		data |= S2MPS_RTC_RUDR_MASK;
+		break;
+	default:
+		return -EINVAL;
+	}
 
-	ret = regmap_write(info->regmap, SEC_RTC_UDR_CON, data);
+	ret = regmap_write(info->regmap, info->regs->rtc_udr_update, data);
 	if (ret < 0) {
 		dev_err(info->dev, "%s: fail to write update reg(%d)\n",
 			__func__, ret);
@@ -215,10 +308,22 @@ static void s5m8763_tm_to_data(struct rtc_time *tm, u8 *data)
 static int s5m_rtc_read_time(struct device *dev, struct rtc_time *tm)
 {
 	struct s5m_rtc_info *info = dev_get_drvdata(dev);
-	u8 data[8];
+	u8 data[info->regs->regs_count];
 	int ret;
 
-	ret = regmap_bulk_read(info->regmap, SEC_RTC_SEC, data, 8);
+	if (info->device_type == S2MPS14X) {
+		ret = regmap_update_bits(info->regmap,
+				info->regs->rtc_udr_update,
+				S2MPS_RTC_RUDR_MASK, S2MPS_RTC_RUDR_MASK);
+		if (ret) {
+			dev_err(dev,
+				"Failed to prepare registers for time reading: %d\n",
+				ret);
+			return ret;
+		}
+	}
+	ret = regmap_bulk_read(info->regmap, info->regs->time, data,
+			info->regs->regs_count);
 	if (ret < 0)
 		return ret;
 
@@ -228,6 +333,7 @@ static int s5m_rtc_read_time(struct device *dev, struct rtc_time *tm)
 		break;
 
 	case S5M8767X:
+	case S2MPS14X:
 		s5m8767_data_to_tm(data, tm, info->rtc_24hr_mode);
 		break;
 
@@ -245,7 +351,7 @@ static int s5m_rtc_read_time(struct device *dev, struct rtc_time *tm)
 static int s5m_rtc_set_time(struct device *dev, struct rtc_time *tm)
 {
 	struct s5m_rtc_info *info = dev_get_drvdata(dev);
-	u8 data[8];
+	u8 data[info->regs->regs_count];
 	int ret = 0;
 
 	switch (info->device_type) {
@@ -253,6 +359,7 @@ static int s5m_rtc_set_time(struct device *dev, struct rtc_time *tm)
 		s5m8763_tm_to_data(tm, data);
 		break;
 	case S5M8767X:
+	case S2MPS14X:
 		ret = s5m8767_tm_to_data(tm, data);
 		break;
 	default:
@@ -266,7 +373,8 @@ static int s5m_rtc_set_time(struct device *dev, struct rtc_time *tm)
 		1900 + tm->tm_year, 1 + tm->tm_mon, tm->tm_mday,
 		tm->tm_hour, tm->tm_min, tm->tm_sec, tm->tm_wday);
 
-	ret = regmap_raw_write(info->regmap, SEC_RTC_SEC, data, 8);
+	ret = regmap_raw_write(info->regmap, info->regs->time, data,
+			info->regs->regs_count);
 	if (ret < 0)
 		return ret;
 
@@ -278,70 +386,60 @@ static int s5m_rtc_set_time(struct device *dev, struct rtc_time *tm)
 static int s5m_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 {
 	struct s5m_rtc_info *info = dev_get_drvdata(dev);
-	u8 data[8];
+	u8 data[info->regs->regs_count];
 	unsigned int val;
 	int ret, i;
 
-	ret = regmap_bulk_read(info->regmap, SEC_ALARM0_SEC, data, 8);
+	ret = regmap_bulk_read(info->regmap, info->regs->alarm0, data,
+			info->regs->regs_count);
 	if (ret < 0)
 		return ret;
 
 	switch (info->device_type) {
 	case S5M8763X:
 		s5m8763_data_to_tm(data, &alrm->time);
-		ret = regmap_read(info->regmap, SEC_ALARM0_CONF, &val);
+		ret = regmap_read(info->regmap, S5M_ALARM0_CONF, &val);
 		if (ret < 0)
 			return ret;
 
 		alrm->enabled = !!val;
-
-		ret = regmap_read(info->regmap, SEC_RTC_STATUS, &val);
-		if (ret < 0)
-			return ret;
-
 		break;
 
 	case S5M8767X:
+	case S2MPS14X:
 		s5m8767_data_to_tm(data, &alrm->time, info->rtc_24hr_mode);
-		dev_dbg(dev, "%s: %d/%d/%d %d:%d:%d(%d)\n", __func__,
-			1900 + alrm->time.tm_year, 1 + alrm->time.tm_mon,
-			alrm->time.tm_mday, alrm->time.tm_hour,
-			alrm->time.tm_min, alrm->time.tm_sec,
-			alrm->time.tm_wday);
-
 		alrm->enabled = 0;
-		for (i = 0; i < 7; i++) {
+		for (i = 0; i < info->regs->regs_count; i++) {
 			if (data[i] & ALARM_ENABLE_MASK) {
 				alrm->enabled = 1;
 				break;
 			}
 		}
-
-		alrm->pending = 0;
-		ret = regmap_read(info->regmap, SEC_RTC_STATUS, &val);
-		if (ret < 0)
-			return ret;
 		break;
 
 	default:
 		return -EINVAL;
 	}
 
-	if (val & ALARM0_STATUS)
-		alrm->pending = 1;
-	else
-		alrm->pending = 0;
+	dev_dbg(dev, "%s: %d/%d/%d %d:%d:%d(%d)\n", __func__,
+		1900 + alrm->time.tm_year, 1 + alrm->time.tm_mon,
+		alrm->time.tm_mday, alrm->time.tm_hour,
+		alrm->time.tm_min, alrm->time.tm_sec,
+		alrm->time.tm_wday);
+
+	ret = s5m_check_peding_alarm_interrupt(info, alrm);
 
 	return 0;
 }
 
 static int s5m_rtc_stop_alarm(struct s5m_rtc_info *info)
 {
-	u8 data[8];
+	u8 data[info->regs->regs_count];
 	int ret, i;
 	struct rtc_time tm;
 
-	ret = regmap_bulk_read(info->regmap, SEC_ALARM0_SEC, data, 8);
+	ret = regmap_bulk_read(info->regmap, info->regs->alarm0, data,
+			info->regs->regs_count);
 	if (ret < 0)
 		return ret;
 
@@ -352,14 +450,16 @@ static int s5m_rtc_stop_alarm(struct s5m_rtc_info *info)
 
 	switch (info->device_type) {
 	case S5M8763X:
-		ret = regmap_write(info->regmap, SEC_ALARM0_CONF, 0);
+		ret = regmap_write(info->regmap, S5M_ALARM0_CONF, 0);
 		break;
 
 	case S5M8767X:
-		for (i = 0; i < 7; i++)
+	case S2MPS14X:
+		for (i = 0; i < info->regs->regs_count; i++)
 			data[i] &= ~ALARM_ENABLE_MASK;
 
-		ret = regmap_raw_write(info->regmap, SEC_ALARM0_SEC, data, 8);
+		ret = regmap_raw_write(info->regmap, info->regs->alarm0, data,
+				info->regs->regs_count);
 		if (ret < 0)
 			return ret;
 
@@ -377,11 +477,12 @@ static int s5m_rtc_stop_alarm(struct s5m_rtc_info *info)
 static int s5m_rtc_start_alarm(struct s5m_rtc_info *info)
 {
 	int ret;
-	u8 data[8];
+	u8 data[info->regs->regs_count];
 	u8 alarm0_conf;
 	struct rtc_time tm;
 
-	ret = regmap_bulk_read(info->regmap, SEC_ALARM0_SEC, data, 8);
+	ret = regmap_bulk_read(info->regmap, info->regs->alarm0, data,
+			info->regs->regs_count);
 	if (ret < 0)
 		return ret;
 
@@ -393,10 +494,11 @@ static int s5m_rtc_start_alarm(struct s5m_rtc_info *info)
 	switch (info->device_type) {
 	case S5M8763X:
 		alarm0_conf = 0x77;
-		ret = regmap_write(info->regmap, SEC_ALARM0_CONF, alarm0_conf);
+		ret = regmap_write(info->regmap, S5M_ALARM0_CONF, alarm0_conf);
 		break;
 
 	case S5M8767X:
+	case S2MPS14X:
 		data[RTC_SEC] |= ALARM_ENABLE_MASK;
 		data[RTC_MIN] |= ALARM_ENABLE_MASK;
 		data[RTC_HOUR] |= ALARM_ENABLE_MASK;
@@ -408,7 +510,8 @@ static int s5m_rtc_start_alarm(struct s5m_rtc_info *info)
 		if (data[RTC_YEAR1] & 0x7f)
 			data[RTC_YEAR1] |= ALARM_ENABLE_MASK;
 
-		ret = regmap_raw_write(info->regmap, SEC_ALARM0_SEC, data, 8);
+		ret = regmap_raw_write(info->regmap, info->regs->alarm0, data,
+				info->regs->regs_count);
 		if (ret < 0)
 			return ret;
 		ret = s5m8767_rtc_set_alarm_reg(info);
@@ -425,7 +528,7 @@ static int s5m_rtc_start_alarm(struct s5m_rtc_info *info)
 static int s5m_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 {
 	struct s5m_rtc_info *info = dev_get_drvdata(dev);
-	u8 data[8];
+	u8 data[info->regs->regs_count];
 	int ret;
 
 	switch (info->device_type) {
@@ -434,6 +537,7 @@ static int s5m_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 		break;
 
 	case S5M8767X:
+	case S2MPS14X:
 		s5m8767_tm_to_data(&alrm->time, data);
 		break;
 
@@ -450,7 +554,8 @@ static int s5m_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 	if (ret < 0)
 		return ret;
 
-	ret = regmap_raw_write(info->regmap, SEC_ALARM0_SEC, data, 8);
+	ret = regmap_raw_write(info->regmap, info->regs->alarm0, data,
+			info->regs->regs_count);
 	if (ret < 0)
 		return ret;
 
@@ -495,7 +600,7 @@ static const struct rtc_class_ops s5m_rtc_ops = {
 static void s5m_rtc_enable_wtsr(struct s5m_rtc_info *info, bool enable)
 {
 	int ret;
-	ret = regmap_update_bits(info->regmap, SEC_WTSR_SMPL_CNTL,
+	ret = regmap_update_bits(info->regmap, info->regs->smpl_wtsr,
 				 WTSR_ENABLE_MASK,
 				 enable ? WTSR_ENABLE_MASK : 0);
 	if (ret < 0)
@@ -506,7 +611,7 @@ static void s5m_rtc_enable_wtsr(struct s5m_rtc_info *info, bool enable)
 static void s5m_rtc_enable_smpl(struct s5m_rtc_info *info, bool enable)
 {
 	int ret;
-	ret = regmap_update_bits(info->regmap, SEC_WTSR_SMPL_CNTL,
+	ret = regmap_update_bits(info->regmap, info->regs->smpl_wtsr,
 				 SMPL_ENABLE_MASK,
 				 enable ? SMPL_ENABLE_MASK : 0);
 	if (ret < 0)
@@ -517,50 +622,41 @@ static void s5m_rtc_enable_smpl(struct s5m_rtc_info *info, bool enable)
 static int s5m8767_rtc_init_reg(struct s5m_rtc_info *info)
 {
 	u8 data[2];
-	unsigned int tp_read;
 	int ret;
-	struct rtc_time tm;
 
-	ret = regmap_read(info->regmap, SEC_RTC_UDR_CON, &tp_read);
-	if (ret < 0) {
-		dev_err(info->dev, "%s: fail to read control reg(%d)\n",
-			__func__, ret);
-		return ret;
-	}
+	switch (info->device_type) {
+	case S5M8763X:
+	case S5M8767X:
+		/* UDR update time. Default of 7.32 ms is too long. */
+		ret = regmap_update_bits(info->regmap, S5M_RTC_UDR_CON,
+				S5M_RTC_UDR_T_MASK, S5M_RTC_UDR_T_450_US);
+		if (ret < 0)
+			dev_err(info->dev, "%s: fail to change UDR time: %d\n",
+					__func__, ret);
 
-	/* Set RTC control register : Binary mode, 24hour mode */
-	data[0] = (1 << BCD_EN_SHIFT) | (1 << MODEL24_SHIFT);
-	data[1] = (0 << BCD_EN_SHIFT) | (1 << MODEL24_SHIFT);
+		/* Set RTC control register : Binary mode, 24hour mode */
+		data[0] = (1 << BCD_EN_SHIFT) | (1 << MODEL24_SHIFT);
+		data[1] = (0 << BCD_EN_SHIFT) | (1 << MODEL24_SHIFT);
+
+		ret = regmap_raw_write(info->regmap, S5M_ALARM0_CONF, data, 2);
+		break;
+
+	case S2MPS14X:
+		data[0] = (0 << BCD_EN_SHIFT) | (1 << MODEL24_SHIFT);
+		ret = regmap_write(info->regmap, info->regs->ctrl, data[0]);
+		break;
+
+	default:
+		return -EINVAL;
+	}
 
 	info->rtc_24hr_mode = 1;
-	ret = regmap_raw_write(info->regmap, SEC_ALARM0_CONF, data, 2);
 	if (ret < 0) {
 		dev_err(info->dev, "%s: fail to write controlm reg(%d)\n",
 			__func__, ret);
 		return ret;
 	}
 
-	/* In first boot time, Set rtc time to 1/1/2012 00:00:00(SUN) */
-	if ((tp_read & RTC_TCON_MASK) == 0) {
-		dev_dbg(info->dev, "rtc init\n");
-		tm.tm_sec = 0;
-		tm.tm_min = 0;
-		tm.tm_hour = 0;
-		tm.tm_wday = 0;
-		tm.tm_mday = 1;
-		tm.tm_mon = 0;
-		tm.tm_year = 112;
-		tm.tm_yday = 0;
-		tm.tm_isdst = 0;
-		ret = s5m_rtc_set_time(info->dev, &tm);
-	}
-
-	ret = regmap_update_bits(info->regmap, SEC_RTC_UDR_CON,
-				 RTC_TCON_MASK, tp_read | RTC_TCON_MASK);
-	if (ret < 0)
-		dev_err(info->dev, "%s: fail to update TCON reg(%d)\n",
-			__func__, ret);
-
 	return ret;
 }
 
@@ -570,7 +666,7 @@ static int s5m_rtc_probe(struct platform_device *pdev)
 	struct sec_platform_data *pdata = s5m87xx->pdata;
 	struct s5m_rtc_info *info;
 	const struct regmap_config *regmap_cfg;
-	int ret;
+	int ret, alarm_irq;
 
 	if (!pdata) {
 		dev_err(pdev->dev.parent, "Platform data not supplied\n");
@@ -584,12 +680,18 @@ static int s5m_rtc_probe(struct platform_device *pdev)
 	switch (pdata->device_type) {
 	case S2MPS14X:
 		regmap_cfg = &s2mps14_rtc_regmap_config;
+		info->regs = &s2mps_rtc_regs;
+		alarm_irq = S2MPS14_IRQ_RTCA0;
 		break;
 	case S5M8763X:
 		regmap_cfg = &s5m_rtc_regmap_config;
+		info->regs = &s5m_rtc_regs;
+		alarm_irq = S5M8763_IRQ_ALARM0;
 		break;
 	case S5M8767X:
 		regmap_cfg = &s5m_rtc_regmap_config;
+		info->regs = &s5m_rtc_regs;
+		alarm_irq = S5M8767_IRQ_RTCA1;
 		break;
 	default:
 		dev_err(&pdev->dev, "Device type is not supported by RTC driver\n");
@@ -615,20 +717,11 @@ static int s5m_rtc_probe(struct platform_device *pdev)
 	info->device_type = s5m87xx->device_type;
 	info->wtsr_smpl = s5m87xx->wtsr_smpl;
 
-	switch (pdata->device_type) {
-	case S5M8763X:
-		info->irq = regmap_irq_get_virq(s5m87xx->irq_data,
-				S5M8763_IRQ_ALARM0);
-		break;
-
-	case S5M8767X:
-		info->irq = regmap_irq_get_virq(s5m87xx->irq_data,
-				S5M8767_IRQ_RTCA1);
-		break;
-
-	default:
+	info->irq = regmap_irq_get_virq(s5m87xx->irq_data, alarm_irq);
+	if (info->irq <= 0) {
 		ret = -EINVAL;
-		dev_err(&pdev->dev, "Unsupported device type: %d\n", ret);
+		dev_err(&pdev->dev, "Failed to get virtual IRQ %d\n",
+				alarm_irq);
 		goto err;
 	}
 
@@ -676,7 +769,7 @@ static void s5m_rtc_shutdown(struct platform_device *pdev)
 	if (info->wtsr_smpl) {
 		for (i = 0; i < 3; i++) {
 			s5m_rtc_enable_wtsr(info, false);
-			regmap_read(info->regmap, SEC_WTSR_SMPL_CNTL, &val);
+			regmap_read(info->regmap, info->regs->smpl_wtsr, &val);
 			pr_debug("%s: WTSR_SMPL reg(0x%02x)\n", __func__, val);
 			if (val & WTSR_ENABLE_MASK)
 				pr_emerg("%s: fail to disable WTSR\n",
@@ -730,7 +823,8 @@ static int s5m_rtc_suspend(struct device *dev)
 static SIMPLE_DEV_PM_OPS(s5m_rtc_pm_ops, s5m_rtc_suspend, s5m_rtc_resume);
 
 static const struct platform_device_id s5m_rtc_id[] = {
-	{ "s5m-rtc", 0 },
+	{ "s5m-rtc",		S5M8767X },
+	{ "s2mps14-rtc",	S2MPS14X },
 };
 
 static struct platform_driver s5m_rtc_driver = {
@@ -749,6 +843,6 @@ module_platform_driver(s5m_rtc_driver);
 
 /* Module information */
 MODULE_AUTHOR("Sangbeom Kim <sbkim73@samsung.com>");
-MODULE_DESCRIPTION("Samsung S5M RTC driver");
+MODULE_DESCRIPTION("Samsung S5M/S2MPS14 RTC driver");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:s5m-rtc");
diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c
index ee0e85abe1fd..0f471750327e 100644
--- a/drivers/s390/block/dcssblk.c
+++ b/drivers/s390/block/dcssblk.c
@@ -593,7 +593,7 @@ dcssblk_add_store(struct device *dev, struct device_attribute *attr, const char
 	dev_info->start = dcssblk_find_lowest_addr(dev_info);
 	dev_info->end = dcssblk_find_highest_addr(dev_info);
 
-	dev_set_name(&dev_info->dev, dev_info->segment_name);
+	dev_set_name(&dev_info->dev, "%s", dev_info->segment_name);
 	dev_info->dev.release = dcssblk_release_segment;
 	dev_info->dev.groups = dcssblk_dev_attr_groups;
 	INIT_LIST_HEAD(&dev_info->lh);
diff --git a/drivers/s390/char/Makefile b/drivers/s390/char/Makefile
index 629fcc275e92..78b6ace7edcb 100644
--- a/drivers/s390/char/Makefile
+++ b/drivers/s390/char/Makefile
@@ -19,7 +19,6 @@ obj-$(CONFIG_SCLP_VT220_TTY) += sclp_vt220.o
 obj-$(CONFIG_SCLP_CPI) += sclp_cpi.o
 obj-$(CONFIG_SCLP_ASYNC) += sclp_async.o
 
-obj-$(CONFIG_ZVM_WATCHDOG) += vmwatchdog.o
 obj-$(CONFIG_VMLOGRDR) += vmlogrdr.o
 obj-$(CONFIG_VMCP) += vmcp.o
 
diff --git a/drivers/s390/char/sclp_vt220.c b/drivers/s390/char/sclp_vt220.c
index cd9c91909596..b9a9f721716d 100644
--- a/drivers/s390/char/sclp_vt220.c
+++ b/drivers/s390/char/sclp_vt220.c
@@ -838,8 +838,6 @@ sclp_vt220_con_init(void)
 {
 	int rc;
 
-	if (!CONSOLE_IS_SCLP)
-		return 0;
 	rc = __sclp_vt220_init(sclp_console_pages);
 	if (rc)
 		return rc;
diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c
index cf31d3321dab..a8848db7b09d 100644
--- a/drivers/s390/char/vmlogrdr.c
+++ b/drivers/s390/char/vmlogrdr.c
@@ -761,7 +761,7 @@ static int vmlogrdr_register_device(struct vmlogrdr_priv_t *priv)
 
 	dev = kzalloc(sizeof(struct device), GFP_KERNEL);
 	if (dev) {
-		dev_set_name(dev, priv->internal_name);
+		dev_set_name(dev, "%s", priv->internal_name);
 		dev->bus = &iucv_bus;
 		dev->parent = iucv_root;
 		dev->driver = &vmlogrdr_driver;
diff --git a/drivers/s390/char/vmwatchdog.c b/drivers/s390/char/vmwatchdog.c
deleted file mode 100644
index d5eac985976b..000000000000
--- a/drivers/s390/char/vmwatchdog.c
+++ /dev/null
@@ -1,338 +0,0 @@
-/*
- * Watchdog implementation based on z/VM Watchdog Timer API
- *
- * Copyright IBM Corp. 2004, 2009
- *
- * The user space watchdog daemon can use this driver as
- * /dev/vmwatchdog to have z/VM execute the specified CP
- * command when the timeout expires. The default command is
- * "IPL", which which cause an immediate reboot.
- */
-#define KMSG_COMPONENT "vmwatchdog"
-#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
-
-#include <linux/init.h>
-#include <linux/fs.h>
-#include <linux/kernel.h>
-#include <linux/miscdevice.h>
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/slab.h>
-#include <linux/suspend.h>
-#include <linux/watchdog.h>
-
-#include <asm/ebcdic.h>
-#include <asm/io.h>
-#include <asm/uaccess.h>
-
-#define MAX_CMDLEN 240
-#define MIN_INTERVAL 15
-static char vmwdt_cmd[MAX_CMDLEN] = "IPL";
-static bool vmwdt_conceal;
-
-static bool vmwdt_nowayout = WATCHDOG_NOWAYOUT;
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Arnd Bergmann <arndb@de.ibm.com>");
-MODULE_DESCRIPTION("z/VM Watchdog Timer");
-module_param_string(cmd, vmwdt_cmd, MAX_CMDLEN, 0644);
-MODULE_PARM_DESC(cmd, "CP command that is run when the watchdog triggers");
-module_param_named(conceal, vmwdt_conceal, bool, 0644);
-MODULE_PARM_DESC(conceal, "Enable the CONCEAL CP option while the watchdog "
-		" is active");
-module_param_named(nowayout, vmwdt_nowayout, bool, 0);
-MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started"
-		" (default=CONFIG_WATCHDOG_NOWAYOUT)");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
-static unsigned int vmwdt_interval = 60;
-static unsigned long vmwdt_is_open;
-static int vmwdt_expect_close;
-
-static DEFINE_MUTEX(vmwdt_mutex);
-
-#define VMWDT_OPEN	0	/* devnode is open or suspend in progress */
-#define VMWDT_RUNNING	1	/* The watchdog is armed */
-
-enum vmwdt_func {
-	/* function codes */
-	wdt_init   = 0,
-	wdt_change = 1,
-	wdt_cancel = 2,
-	/* flags */
-	wdt_conceal = 0x80000000,
-};
-
-static int __diag288(enum vmwdt_func func, unsigned int timeout,
-			    char *cmd, size_t len)
-{
-	register unsigned long __func asm("2") = func;
-	register unsigned long __timeout asm("3") = timeout;
-	register unsigned long __cmdp asm("4") = virt_to_phys(cmd);
-	register unsigned long __cmdl asm("5") = len;
-	int err;
-
-	err = -EINVAL;
-	asm volatile(
-		"	diag	%1,%3,0x288\n"
-		"0:	la	%0,0\n"
-		"1:\n"
-		EX_TABLE(0b,1b)
-		: "+d" (err) : "d"(__func), "d"(__timeout),
-		  "d"(__cmdp), "d"(__cmdl) : "1", "cc");
-	return err;
-}
-
-static int vmwdt_keepalive(void)
-{
-	/* we allocate new memory every time to avoid having
-	 * to track the state. static allocation is not an
-	 * option since that might not be contiguous in real
-	 * storage in case of a modular build */
-	static char *ebc_cmd;
-	size_t len;
-	int ret;
-	unsigned int func;
-
-	ebc_cmd = kmalloc(MAX_CMDLEN, GFP_KERNEL);
-	if (!ebc_cmd)
-		return -ENOMEM;
-
-	len = strlcpy(ebc_cmd, vmwdt_cmd, MAX_CMDLEN);
-	ASCEBC(ebc_cmd, MAX_CMDLEN);
-	EBC_TOUPPER(ebc_cmd, MAX_CMDLEN);
-
-	func = vmwdt_conceal ? (wdt_init | wdt_conceal) : wdt_init;
-	set_bit(VMWDT_RUNNING, &vmwdt_is_open);
-	ret = __diag288(func, vmwdt_interval, ebc_cmd, len);
-	WARN_ON(ret != 0);
-	kfree(ebc_cmd);
-	return ret;
-}
-
-static int vmwdt_disable(void)
-{
-	char cmd[] = {'\0'};
-	int ret = __diag288(wdt_cancel, 0, cmd, 0);
-	WARN_ON(ret != 0);
-	clear_bit(VMWDT_RUNNING, &vmwdt_is_open);
-	return ret;
-}
-
-static int __init vmwdt_probe(void)
-{
-	/* there is no real way to see if the watchdog is supported,
-	 * so we try initializing it with a NOP command ("BEGIN")
-	 * that won't cause any harm even if the following disable
-	 * fails for some reason */
-	char ebc_begin[] = {
-		194, 197, 199, 201, 213
-	};
-	if (__diag288(wdt_init, 15, ebc_begin, sizeof(ebc_begin)) != 0)
-		return -EINVAL;
-	return vmwdt_disable();
-}
-
-static int vmwdt_open(struct inode *i, struct file *f)
-{
-	int ret;
-	if (test_and_set_bit(VMWDT_OPEN, &vmwdt_is_open))
-		return -EBUSY;
-	ret = vmwdt_keepalive();
-	if (ret)
-		clear_bit(VMWDT_OPEN, &vmwdt_is_open);
-	return ret ? ret : nonseekable_open(i, f);
-}
-
-static int vmwdt_close(struct inode *i, struct file *f)
-{
-	if (vmwdt_expect_close == 42)
-		vmwdt_disable();
-	vmwdt_expect_close = 0;
-	clear_bit(VMWDT_OPEN, &vmwdt_is_open);
-	return 0;
-}
-
-static struct watchdog_info vmwdt_info = {
-	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
-	.firmware_version = 0,
-	.identity = "z/VM Watchdog Timer",
-};
-
-static int __vmwdt_ioctl(unsigned int cmd, unsigned long arg)
-{
-	switch (cmd) {
-	case WDIOC_GETSUPPORT:
-		if (copy_to_user((void __user *)arg, &vmwdt_info,
-					sizeof(vmwdt_info)))
-			return -EFAULT;
-		return 0;
-	case WDIOC_GETSTATUS:
-	case WDIOC_GETBOOTSTATUS:
-		return put_user(0, (int __user *)arg);
-	case WDIOC_GETTEMP:
-		return -EINVAL;
-	case WDIOC_SETOPTIONS:
-		{
-			int options, ret;
-			if (get_user(options, (int __user *)arg))
-				return -EFAULT;
-			ret = -EINVAL;
-			if (options & WDIOS_DISABLECARD) {
-				ret = vmwdt_disable();
-				if (ret)
-					return ret;
-			}
-			if (options & WDIOS_ENABLECARD) {
-				ret = vmwdt_keepalive();
-			}
-			return ret;
-		}
-	case WDIOC_GETTIMEOUT:
-		return put_user(vmwdt_interval, (int __user *)arg);
-	case WDIOC_SETTIMEOUT:
-		{
-			int interval;
-			if (get_user(interval, (int __user *)arg))
-				return -EFAULT;
-			if (interval < MIN_INTERVAL)
-				return -EINVAL;
-			vmwdt_interval = interval;
-		}
-		return vmwdt_keepalive();
-	case WDIOC_KEEPALIVE:
-		return vmwdt_keepalive();
-	}
-	return -EINVAL;
-}
-
-static long vmwdt_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
-{
-	int rc;
-
-	mutex_lock(&vmwdt_mutex);
-	rc = __vmwdt_ioctl(cmd, arg);
-	mutex_unlock(&vmwdt_mutex);
-	return (long) rc;
-}
-
-static ssize_t vmwdt_write(struct file *f, const char __user *buf,
-				size_t count, loff_t *ppos)
-{
-	if(count) {
-		if (!vmwdt_nowayout) {
-			size_t i;
-
-			/* note: just in case someone wrote the magic character
-			 * five months ago... */
-			vmwdt_expect_close = 0;
-
-			for (i = 0; i != count; i++) {
-				char c;
-				if (get_user(c, buf+i))
-					return -EFAULT;
-				if (c == 'V')
-					vmwdt_expect_close = 42;
-			}
-		}
-		/* someone wrote to us, we should restart timer */
-		vmwdt_keepalive();
-	}
-	return count;
-}
-
-static int vmwdt_resume(void)
-{
-	clear_bit(VMWDT_OPEN, &vmwdt_is_open);
-	return NOTIFY_DONE;
-}
-
-/*
- * It makes no sense to go into suspend while the watchdog is running.
- * Depending on the memory size, the watchdog might trigger, while we
- * are still saving the memory.
- * We reuse the open flag to ensure that suspend and watchdog open are
- * exclusive operations
- */
-static int vmwdt_suspend(void)
-{
-	if (test_and_set_bit(VMWDT_OPEN, &vmwdt_is_open)) {
-		pr_err("The system cannot be suspended while the watchdog"
-			" is in use\n");
-		return notifier_from_errno(-EBUSY);
-	}
-	if (test_bit(VMWDT_RUNNING, &vmwdt_is_open)) {
-		clear_bit(VMWDT_OPEN, &vmwdt_is_open);
-		pr_err("The system cannot be suspended while the watchdog"
-			" is running\n");
-		return notifier_from_errno(-EBUSY);
-	}
-	return NOTIFY_DONE;
-}
-
-/*
- * This function is called for suspend and resume.
- */
-static int vmwdt_power_event(struct notifier_block *this, unsigned long event,
-			     void *ptr)
-{
-	switch (event) {
-	case PM_POST_HIBERNATION:
-	case PM_POST_SUSPEND:
-		return vmwdt_resume();
-	case PM_HIBERNATION_PREPARE:
-	case PM_SUSPEND_PREPARE:
-		return vmwdt_suspend();
-	default:
-		return NOTIFY_DONE;
-	}
-}
-
-static struct notifier_block vmwdt_power_notifier = {
-	.notifier_call = vmwdt_power_event,
-};
-
-static const struct file_operations vmwdt_fops = {
-	.open    = &vmwdt_open,
-	.release = &vmwdt_close,
-	.unlocked_ioctl = &vmwdt_ioctl,
-	.write   = &vmwdt_write,
-	.owner   = THIS_MODULE,
-	.llseek  = noop_llseek,
-};
-
-static struct miscdevice vmwdt_dev = {
-	.minor      = WATCHDOG_MINOR,
-	.name       = "watchdog",
-	.fops       = &vmwdt_fops,
-};
-
-static int __init vmwdt_init(void)
-{
-	int ret;
-
-	ret = vmwdt_probe();
-	if (ret)
-		return ret;
-	ret = register_pm_notifier(&vmwdt_power_notifier);
-	if (ret)
-		return ret;
-	/*
-	 * misc_register() has to be the last action in module_init(), because
-	 * file operations will be available right after this.
-	 */
-	ret = misc_register(&vmwdt_dev);
-	if (ret) {
-		unregister_pm_notifier(&vmwdt_power_notifier);
-		return ret;
-	}
-	return 0;
-}
-module_init(vmwdt_init);
-
-static void __exit vmwdt_exit(void)
-{
-	unregister_pm_notifier(&vmwdt_power_notifier);
-	misc_deregister(&vmwdt_dev);
-}
-module_exit(vmwdt_exit);
diff --git a/drivers/s390/cio/airq.c b/drivers/s390/cio/airq.c
index 445564c790f6..00bfbee0af9e 100644
--- a/drivers/s390/cio/airq.c
+++ b/drivers/s390/cio/airq.c
@@ -196,11 +196,11 @@ EXPORT_SYMBOL(airq_iv_release);
  */
 unsigned long airq_iv_alloc(struct airq_iv *iv, unsigned long num)
 {
-	unsigned long bit, i;
+	unsigned long bit, i, flags;
 
 	if (!iv->avail || num == 0)
 		return -1UL;
-	spin_lock(&iv->lock);
+	spin_lock_irqsave(&iv->lock, flags);
 	bit = find_first_bit_inv(iv->avail, iv->bits);
 	while (bit + num <= iv->bits) {
 		for (i = 1; i < num; i++)
@@ -218,9 +218,8 @@ unsigned long airq_iv_alloc(struct airq_iv *iv, unsigned long num)
 	}
 	if (bit + num > iv->bits)
 		bit = -1UL;
-	spin_unlock(&iv->lock);
+	spin_unlock_irqrestore(&iv->lock, flags);
 	return bit;
-
 }
 EXPORT_SYMBOL(airq_iv_alloc);
 
@@ -232,11 +231,11 @@ EXPORT_SYMBOL(airq_iv_alloc);
  */
 void airq_iv_free(struct airq_iv *iv, unsigned long bit, unsigned long num)
 {
-	unsigned long i;
+	unsigned long i, flags;
 
 	if (!iv->avail || num == 0)
 		return;
-	spin_lock(&iv->lock);
+	spin_lock_irqsave(&iv->lock, flags);
 	for (i = 0; i < num; i++) {
 		/* Clear (possibly left over) interrupt bit */
 		clear_bit_inv(bit + i, iv->vector);
@@ -248,7 +247,7 @@ void airq_iv_free(struct airq_iv *iv, unsigned long bit, unsigned long num)
 		while (iv->end > 0 && !test_bit_inv(iv->end - 1, iv->avail))
 			iv->end--;
 	}
-	spin_unlock(&iv->lock);
+	spin_unlock_irqrestore(&iv->lock, flags);
 }
 EXPORT_SYMBOL(airq_iv_free);
 
diff --git a/drivers/s390/cio/ccwgroup.c b/drivers/s390/cio/ccwgroup.c
index dfd7bc681c25..e443b0d0b236 100644
--- a/drivers/s390/cio/ccwgroup.c
+++ b/drivers/s390/cio/ccwgroup.c
@@ -184,7 +184,7 @@ static ssize_t ccwgroup_ungroup_store(struct device *dev,
 				      const char *buf, size_t count)
 {
 	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
-	int rc;
+	int rc = 0;
 
 	/* Prevent concurrent online/offline processing and ungrouping. */
 	if (atomic_cmpxchg(&gdev->onoff, 0, 1) != 0)
@@ -196,11 +196,12 @@ static ssize_t ccwgroup_ungroup_store(struct device *dev,
 
 	if (device_remove_file_self(dev, attr))
 		ccwgroup_ungroup(gdev);
+	else
+		rc = -ENODEV;
 out:
 	if (rc) {
-		if (rc != -EAGAIN)
-			/* Release onoff "lock" when ungrouping failed. */
-			atomic_set(&gdev->onoff, 0);
+		/* Release onoff "lock" when ungrouping failed. */
+		atomic_set(&gdev->onoff, 0);
 		return rc;
 	}
 	return count;
@@ -227,6 +228,7 @@ static void ccwgroup_ungroup_workfn(struct work_struct *work)
 		container_of(work, struct ccwgroup_device, ungroup_work);
 
 	ccwgroup_ungroup(gdev);
+	put_device(&gdev->dev);
 }
 
 static void ccwgroup_release(struct device *dev)
@@ -412,8 +414,10 @@ static int ccwgroup_notifier(struct notifier_block *nb, unsigned long action,
 {
 	struct ccwgroup_device *gdev = to_ccwgroupdev(data);
 
-	if (action == BUS_NOTIFY_UNBIND_DRIVER)
+	if (action == BUS_NOTIFY_UNBIND_DRIVER) {
+		get_device(&gdev->dev);
 		schedule_work(&gdev->ungroup_work);
+	}
 
 	return NOTIFY_OK;
 }
@@ -582,11 +586,7 @@ void ccwgroup_driver_unregister(struct ccwgroup_driver *cdriver)
 					 __ccwgroup_match_all))) {
 		struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
 
-		mutex_lock(&gdev->reg_mutex);
-		__ccwgroup_remove_symlinks(gdev);
-		device_unregister(dev);
-		__ccwgroup_remove_cdev_refs(gdev);
-		mutex_unlock(&gdev->reg_mutex);
+		ccwgroup_ungroup(gdev);
 		put_device(dev);
 	}
 	driver_unregister(&cdriver->driver);
@@ -633,13 +633,7 @@ void ccwgroup_remove_ccwdev(struct ccw_device *cdev)
 	get_device(&gdev->dev);
 	spin_unlock_irq(cdev->ccwlock);
 	/* Unregister group device. */
-	mutex_lock(&gdev->reg_mutex);
-	if (device_is_registered(&gdev->dev)) {
-		__ccwgroup_remove_symlinks(gdev);
-		device_unregister(&gdev->dev);
-		__ccwgroup_remove_cdev_refs(gdev);
-	}
-	mutex_unlock(&gdev->reg_mutex);
+	ccwgroup_ungroup(gdev);
 	/* Release ccwgroup device reference for local processing. */
 	put_device(&gdev->dev);
 }
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 77f9c92df4b9..2905d8b0ec95 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -602,6 +602,7 @@ void __init init_cio_interrupts(void)
 
 #ifdef CONFIG_CCW_CONSOLE
 static struct subchannel *console_sch;
+static struct lock_class_key console_sch_key;
 
 /*
  * Use cio_tsch to update the subchannel status and call the interrupt handler
@@ -686,6 +687,7 @@ struct subchannel *cio_probe_console(void)
 	if (IS_ERR(sch))
 		return sch;
 
+	lockdep_set_class(sch->lock, &console_sch_key);
 	isc_register(CONSOLE_ISC);
 	sch->config.isc = CONSOLE_ISC;
 	sch->config.intparm = (u32)(addr_t)sch;
diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index d8d9b5b5cc56..dfef5e63cb7b 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -678,18 +678,11 @@ static const struct attribute_group *ccwdev_attr_groups[] = {
 	NULL,
 };
 
-/* this is a simple abstraction for device_register that sets the
- * correct bus type and adds the bus specific files */
-static int ccw_device_register(struct ccw_device *cdev)
+static int ccw_device_add(struct ccw_device *cdev)
 {
 	struct device *dev = &cdev->dev;
-	int ret;
 
 	dev->bus = &ccw_bus_type;
-	ret = dev_set_name(&cdev->dev, "0.%x.%04x", cdev->private->dev_id.ssid,
-			   cdev->private->dev_id.devno);
-	if (ret)
-		return ret;
 	return device_add(dev);
 }
 
@@ -764,22 +757,46 @@ static void ccw_device_todo(struct work_struct *work);
 static int io_subchannel_initialize_dev(struct subchannel *sch,
 					struct ccw_device *cdev)
 {
-	cdev->private->cdev = cdev;
-	cdev->private->int_class = IRQIO_CIO;
-	atomic_set(&cdev->private->onoff, 0);
+	struct ccw_device_private *priv = cdev->private;
+	int ret;
+
+	priv->cdev = cdev;
+	priv->int_class = IRQIO_CIO;
+	priv->state = DEV_STATE_NOT_OPER;
+	priv->dev_id.devno = sch->schib.pmcw.dev;
+	priv->dev_id.ssid = sch->schid.ssid;
+	priv->schid = sch->schid;
+
+	INIT_WORK(&priv->todo_work, ccw_device_todo);
+	INIT_LIST_HEAD(&priv->cmb_list);
+	init_waitqueue_head(&priv->wait_q);
+	init_timer(&priv->timer);
+
+	atomic_set(&priv->onoff, 0);
+	cdev->ccwlock = sch->lock;
 	cdev->dev.parent = &sch->dev;
 	cdev->dev.release = ccw_device_release;
-	INIT_WORK(&cdev->private->todo_work, ccw_device_todo);
 	cdev->dev.groups = ccwdev_attr_groups;
 	/* Do first half of device_register. */
 	device_initialize(&cdev->dev);
+	ret = dev_set_name(&cdev->dev, "0.%x.%04x", cdev->private->dev_id.ssid,
+			   cdev->private->dev_id.devno);
+	if (ret)
+		goto out_put;
 	if (!get_device(&sch->dev)) {
-		/* Release reference from device_initialize(). */
-		put_device(&cdev->dev);
-		return -ENODEV;
+		ret = -ENODEV;
+		goto out_put;
 	}
-	cdev->private->flags.initialized = 1;
+	priv->flags.initialized = 1;
+	spin_lock_irq(sch->lock);
+	sch_set_cdev(sch, cdev);
+	spin_unlock_irq(sch->lock);
 	return 0;
+
+out_put:
+	/* Release reference from device_initialize(). */
+	put_device(&cdev->dev);
+	return ret;
 }
 
 static struct ccw_device * io_subchannel_create_ccwdev(struct subchannel *sch)
@@ -858,7 +875,7 @@ static void io_subchannel_register(struct ccw_device *cdev)
 	dev_set_uevent_suppress(&sch->dev, 0);
 	kobject_uevent(&sch->dev.kobj, KOBJ_ADD);
 	/* make it known to the system */
-	ret = ccw_device_register(cdev);
+	ret = ccw_device_add(cdev);
 	if (ret) {
 		CIO_MSG_EVENT(0, "Could not register ccw dev 0.%x.%04x: %d\n",
 			      cdev->private->dev_id.ssid,
@@ -923,26 +940,11 @@ io_subchannel_recog_done(struct ccw_device *cdev)
 
 static void io_subchannel_recog(struct ccw_device *cdev, struct subchannel *sch)
 {
-	struct ccw_device_private *priv;
-
-	cdev->ccwlock = sch->lock;
-
-	/* Init private data. */
-	priv = cdev->private;
-	priv->dev_id.devno = sch->schib.pmcw.dev;
-	priv->dev_id.ssid = sch->schid.ssid;
-	priv->schid = sch->schid;
-	priv->state = DEV_STATE_NOT_OPER;
-	INIT_LIST_HEAD(&priv->cmb_list);
-	init_waitqueue_head(&priv->wait_q);
-	init_timer(&priv->timer);
-
 	/* Increase counter of devices currently in recognition. */
 	atomic_inc(&ccw_device_init_count);
 
 	/* Start async. device sensing. */
 	spin_lock_irq(sch->lock);
-	sch_set_cdev(sch, cdev);
 	ccw_device_recognition(cdev);
 	spin_unlock_irq(sch->lock);
 }
@@ -1083,7 +1085,7 @@ static int io_subchannel_probe(struct subchannel *sch)
 		dev_set_uevent_suppress(&sch->dev, 0);
 		kobject_uevent(&sch->dev.kobj, KOBJ_ADD);
 		cdev = sch_get_cdev(sch);
-		rc = ccw_device_register(cdev);
+		rc = ccw_device_add(cdev);
 		if (rc) {
 			/* Release online reference. */
 			put_device(&cdev->dev);
@@ -1597,7 +1599,6 @@ int __init ccw_device_enable_console(struct ccw_device *cdev)
 	if (rc)
 		return rc;
 	sch->driver = &io_subchannel_driver;
-	sch_set_cdev(sch, cdev);
 	io_subchannel_recog(cdev, sch);
 	/* Now wait for the async. recognition to come to an end. */
 	spin_lock_irq(cdev->ccwlock);
@@ -1639,6 +1640,7 @@ struct ccw_device * __init ccw_device_create_console(struct ccw_driver *drv)
 		put_device(&sch->dev);
 		return ERR_PTR(-ENOMEM);
 	}
+	set_io_private(sch, io_priv);
 	cdev = io_subchannel_create_ccwdev(sch);
 	if (IS_ERR(cdev)) {
 		put_device(&sch->dev);
@@ -1646,7 +1648,6 @@ struct ccw_device * __init ccw_device_create_console(struct ccw_driver *drv)
 		return cdev;
 	}
 	cdev->drv = drv;
-	set_io_private(sch, io_priv);
 	ccw_device_set_int_class(cdev);
 	return cdev;
 }
diff --git a/drivers/s390/cio/qdio_debug.c b/drivers/s390/cio/qdio_debug.c
index 4221b02085ad..f1f3baa8e6e4 100644
--- a/drivers/s390/cio/qdio_debug.c
+++ b/drivers/s390/cio/qdio_debug.c
@@ -7,6 +7,7 @@
 #include <linux/debugfs.h>
 #include <linux/uaccess.h>
 #include <linux/export.h>
+#include <linux/slab.h>
 #include <asm/debug.h>
 #include "qdio_debug.h"
 #include "qdio.h"
@@ -16,11 +17,51 @@ debug_info_t *qdio_dbf_error;
 
 static struct dentry *debugfs_root;
 #define QDIO_DEBUGFS_NAME_LEN	10
+#define QDIO_DBF_NAME_LEN	20
 
-void qdio_allocate_dbf(struct qdio_initialize *init_data,
+struct qdio_dbf_entry {
+	char dbf_name[QDIO_DBF_NAME_LEN];
+	debug_info_t *dbf_info;
+	struct list_head dbf_list;
+};
+
+static LIST_HEAD(qdio_dbf_list);
+static DEFINE_MUTEX(qdio_dbf_list_mutex);
+
+static debug_info_t *qdio_get_dbf_entry(char *name)
+{
+	struct qdio_dbf_entry *entry;
+	debug_info_t *rc = NULL;
+
+	mutex_lock(&qdio_dbf_list_mutex);
+	list_for_each_entry(entry, &qdio_dbf_list, dbf_list) {
+		if (strcmp(entry->dbf_name, name) == 0) {
+			rc = entry->dbf_info;
+			break;
+		}
+	}
+	mutex_unlock(&qdio_dbf_list_mutex);
+	return rc;
+}
+
+static void qdio_clear_dbf_list(void)
+{
+	struct qdio_dbf_entry *entry, *tmp;
+
+	mutex_lock(&qdio_dbf_list_mutex);
+	list_for_each_entry_safe(entry, tmp, &qdio_dbf_list, dbf_list) {
+		list_del(&entry->dbf_list);
+		debug_unregister(entry->dbf_info);
+		kfree(entry);
+	}
+	mutex_unlock(&qdio_dbf_list_mutex);
+}
+
+int qdio_allocate_dbf(struct qdio_initialize *init_data,
 		       struct qdio_irq *irq_ptr)
 {
-	char text[20];
+	char text[QDIO_DBF_NAME_LEN];
+	struct qdio_dbf_entry *new_entry;
 
 	DBF_EVENT("qfmt:%1d", init_data->q_format);
 	DBF_HEX(init_data->adapter_name, 8);
@@ -38,11 +79,34 @@ void qdio_allocate_dbf(struct qdio_initialize *init_data,
 	DBF_EVENT("irq:%8lx", (unsigned long)irq_ptr);
 
 	/* allocate trace view for the interface */
-	snprintf(text, 20, "qdio_%s", dev_name(&init_data->cdev->dev));
-	irq_ptr->debug_area = debug_register(text, 2, 1, 16);
-	debug_register_view(irq_ptr->debug_area, &debug_hex_ascii_view);
-	debug_set_level(irq_ptr->debug_area, DBF_WARN);
-	DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf created");
+	snprintf(text, QDIO_DBF_NAME_LEN, "qdio_%s",
+					dev_name(&init_data->cdev->dev));
+	irq_ptr->debug_area = qdio_get_dbf_entry(text);
+	if (irq_ptr->debug_area)
+		DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf reused");
+	else {
+		irq_ptr->debug_area = debug_register(text, 2, 1, 16);
+		if (!irq_ptr->debug_area)
+			return -ENOMEM;
+		if (debug_register_view(irq_ptr->debug_area,
+						&debug_hex_ascii_view)) {
+			debug_unregister(irq_ptr->debug_area);
+			return -ENOMEM;
+		}
+		debug_set_level(irq_ptr->debug_area, DBF_WARN);
+		DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf created");
+		new_entry = kzalloc(sizeof(struct qdio_dbf_entry), GFP_KERNEL);
+		if (!new_entry) {
+			debug_unregister(irq_ptr->debug_area);
+			return -ENOMEM;
+		}
+		strlcpy(new_entry->dbf_name, text, QDIO_DBF_NAME_LEN);
+		new_entry->dbf_info = irq_ptr->debug_area;
+		mutex_lock(&qdio_dbf_list_mutex);
+		list_add(&new_entry->dbf_list, &qdio_dbf_list);
+		mutex_unlock(&qdio_dbf_list_mutex);
+	}
+	return 0;
 }
 
 static int qstat_show(struct seq_file *m, void *v)
@@ -300,6 +364,7 @@ int __init qdio_debug_init(void)
 
 void qdio_debug_exit(void)
 {
+	qdio_clear_dbf_list();
 	debugfs_remove(debugfs_root);
 	if (qdio_dbf_setup)
 		debug_unregister(qdio_dbf_setup);
diff --git a/drivers/s390/cio/qdio_debug.h b/drivers/s390/cio/qdio_debug.h
index dfac9bfefea3..f33ce8577619 100644
--- a/drivers/s390/cio/qdio_debug.h
+++ b/drivers/s390/cio/qdio_debug.h
@@ -75,7 +75,7 @@ static inline void DBF_DEV_HEX(struct qdio_irq *dev, void *addr,
 	}
 }
 
-void qdio_allocate_dbf(struct qdio_initialize *init_data,
+int qdio_allocate_dbf(struct qdio_initialize *init_data,
 		       struct qdio_irq *irq_ptr);
 void qdio_setup_debug_entries(struct qdio_irq *irq_ptr,
 			      struct ccw_device *cdev);
diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index 77466c4faabb..848e3b64ea6e 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -409,17 +409,16 @@ static inline void qdio_stop_polling(struct qdio_q *q)
 		set_buf_state(q, q->u.in.ack_start, SLSB_P_INPUT_NOT_INIT);
 }
 
-static inline void account_sbals(struct qdio_q *q, int count)
+static inline void account_sbals(struct qdio_q *q, unsigned int count)
 {
-	int pos = 0;
+	int pos;
 
 	q->q_stats.nr_sbal_total += count;
 	if (count == QDIO_MAX_BUFFERS_MASK) {
 		q->q_stats.nr_sbals[7]++;
 		return;
 	}
-	while (count >>= 1)
-		pos++;
+	pos = ilog2(count);
 	q->q_stats.nr_sbals[pos]++;
 }
 
@@ -1234,12 +1233,10 @@ int qdio_free(struct ccw_device *cdev)
 		return -ENODEV;
 
 	DBF_EVENT("qfree:%4x", cdev->private->schid.sch_no);
+	DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf abandoned");
 	mutex_lock(&irq_ptr->setup_mutex);
 
-	if (irq_ptr->debug_area != NULL) {
-		debug_unregister(irq_ptr->debug_area);
-		irq_ptr->debug_area = NULL;
-	}
+	irq_ptr->debug_area = NULL;
 	cdev->private->qdio_data = NULL;
 	mutex_unlock(&irq_ptr->setup_mutex);
 
@@ -1276,7 +1273,8 @@ int qdio_allocate(struct qdio_initialize *init_data)
 		goto out_err;
 
 	mutex_init(&irq_ptr->setup_mutex);
-	qdio_allocate_dbf(init_data, irq_ptr);
+	if (qdio_allocate_dbf(init_data, irq_ptr))
+		goto out_rel;
 
 	/*
 	 * Allocate a page for the chsc calls in qdio_establish.
diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c
index 8eec1653c9cc..69ef4f8cfac8 100644
--- a/drivers/s390/crypto/ap_bus.c
+++ b/drivers/s390/crypto/ap_bus.c
@@ -77,12 +77,12 @@ MODULE_ALIAS("z90crypt");
  * Module parameter
  */
 int ap_domain_index = -1;	/* Adjunct Processor Domain Index */
-module_param_named(domain, ap_domain_index, int, 0000);
+module_param_named(domain, ap_domain_index, int, S_IRUSR|S_IRGRP);
 MODULE_PARM_DESC(domain, "domain index for ap devices");
 EXPORT_SYMBOL(ap_domain_index);
 
 static int ap_thread_flag = 0;
-module_param_named(poll_thread, ap_thread_flag, int, 0000);
+module_param_named(poll_thread, ap_thread_flag, int, S_IRUSR|S_IRGRP);
 MODULE_PARM_DESC(poll_thread, "Turn on/off poll thread, default is 0 (off).");
 
 static struct device *ap_root_device = NULL;
diff --git a/drivers/s390/crypto/zcrypt_api.c b/drivers/s390/crypto/zcrypt_api.c
index 5222ebe15705..0e18c5dcd91f 100644
--- a/drivers/s390/crypto/zcrypt_api.c
+++ b/drivers/s390/crypto/zcrypt_api.c
@@ -356,7 +356,7 @@ struct zcrypt_ops *zcrypt_msgtype_request(unsigned char *name, int variant)
 
 	zops = __ops_lookup(name, variant);
 	if (!zops) {
-		request_module(name);
+		request_module("%s", name);
 		zops = __ops_lookup(name, variant);
 	}
 	if ((!zops) || (!try_module_get(zops->owner)))
diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c
index 1e1fc671f89a..d2c0b442bce5 100644
--- a/drivers/s390/kvm/virtio_ccw.c
+++ b/drivers/s390/kvm/virtio_ccw.c
@@ -27,6 +27,7 @@
 #include <linux/module.h>
 #include <linux/io.h>
 #include <linux/kvm_para.h>
+#include <linux/notifier.h>
 #include <asm/setup.h>
 #include <asm/irq.h>
 #include <asm/cio.h>
@@ -62,6 +63,7 @@ struct virtio_ccw_device {
 	struct vq_config_block *config_block;
 	bool is_thinint;
 	bool going_away;
+	bool device_lost;
 	void *airq_info;
 };
 
@@ -1010,11 +1012,14 @@ static void virtio_ccw_remove(struct ccw_device *cdev)
 	unsigned long flags;
 	struct virtio_ccw_device *vcdev = virtio_grab_drvdata(cdev);
 
-	if (vcdev && cdev->online)
+	if (vcdev && cdev->online) {
+		if (vcdev->device_lost)
+			virtio_break_device(&vcdev->vdev);
 		unregister_virtio_device(&vcdev->vdev);
-	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
-	dev_set_drvdata(&cdev->dev, NULL);
-	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+		spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+		dev_set_drvdata(&cdev->dev, NULL);
+		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+	}
 	cdev->handler = NULL;
 }
 
@@ -1023,12 +1028,14 @@ static int virtio_ccw_offline(struct ccw_device *cdev)
 	unsigned long flags;
 	struct virtio_ccw_device *vcdev = virtio_grab_drvdata(cdev);
 
-	if (vcdev) {
-		unregister_virtio_device(&vcdev->vdev);
-		spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
-		dev_set_drvdata(&cdev->dev, NULL);
-		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
-	}
+	if (!vcdev)
+		return 0;
+	if (vcdev->device_lost)
+		virtio_break_device(&vcdev->vdev);
+	unregister_virtio_device(&vcdev->vdev);
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	dev_set_drvdata(&cdev->dev, NULL);
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
 	return 0;
 }
 
@@ -1096,8 +1103,26 @@ out_free:
 
 static int virtio_ccw_cio_notify(struct ccw_device *cdev, int event)
 {
-	/* TODO: Check whether we need special handling here. */
-	return 0;
+	int rc;
+	struct virtio_ccw_device *vcdev = dev_get_drvdata(&cdev->dev);
+
+	/*
+	 * Make sure vcdev is set
+	 * i.e. set_offline/remove callback not already running
+	 */
+	if (!vcdev)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case CIO_GONE:
+		vcdev->device_lost = true;
+		rc = NOTIFY_DONE;
+		break;
+	default:
+		rc = NOTIFY_DONE;
+		break;
+	}
+	return rc;
 }
 
 static struct ccw_device_id virtio_ids[] = {
diff --git a/drivers/s390/net/claw.c b/drivers/s390/net/claw.c
index fd7b3bd80789..d837c3c5330f 100644
--- a/drivers/s390/net/claw.c
+++ b/drivers/s390/net/claw.c
@@ -3348,7 +3348,7 @@ static int __init claw_init(void)
 	}
 	CLAW_DBF_TEXT(2, setup, "init_mod");
 	claw_root_dev = root_device_register("claw");
-	ret = PTR_RET(claw_root_dev);
+	ret = PTR_ERR_OR_ZERO(claw_root_dev);
 	if (ret)
 		goto register_err;
 	ret = ccw_driver_register(&claw_ccw_driver);
diff --git a/drivers/s390/net/ctcm_main.c b/drivers/s390/net/ctcm_main.c
index 70b3a023100e..03b6ad035577 100644
--- a/drivers/s390/net/ctcm_main.c
+++ b/drivers/s390/net/ctcm_main.c
@@ -1837,7 +1837,7 @@ static int __init ctcm_init(void)
 	if (ret)
 		goto out_err;
 	ctcm_root_dev = root_device_register("ctcm");
-	ret = PTR_RET(ctcm_root_dev);
+	ret = PTR_ERR_OR_ZERO(ctcm_root_dev);
 	if (ret)
 		goto register_err;
 	ret = ccw_driver_register(&ctcm_ccw_driver);
diff --git a/drivers/s390/net/ctcm_sysfs.c b/drivers/s390/net/ctcm_sysfs.c
index 985b5dcbdac8..6bcfbbb20f04 100644
--- a/drivers/s390/net/ctcm_sysfs.c
+++ b/drivers/s390/net/ctcm_sysfs.c
@@ -34,8 +34,9 @@ static ssize_t ctcm_buffer_write(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
 	struct net_device *ndev;
-	int bs1;
+	unsigned int bs1;
 	struct ctcm_priv *priv = dev_get_drvdata(dev);
+	int rc;
 
 	ndev = priv->channel[CTCM_READ]->netdev;
 	if (!(priv && priv->channel[CTCM_READ] && ndev)) {
@@ -43,7 +44,9 @@ static ssize_t ctcm_buffer_write(struct device *dev,
 		return -ENODEV;
 	}
 
-	sscanf(buf, "%u", &bs1);
+	rc = sscanf(buf, "%u", &bs1);
+	if (rc != 1)
+		goto einval;
 	if (bs1 > CTCM_BUFSIZE_LIMIT)
 					goto einval;
 	if (bs1 < (576 + LL_HEADER_LENGTH + 2))
@@ -143,13 +146,14 @@ static ssize_t ctcm_proto_show(struct device *dev,
 static ssize_t ctcm_proto_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	int value;
+	int value, rc;
 	struct ctcm_priv *priv = dev_get_drvdata(dev);
 
 	if (!priv)
 		return -ENODEV;
-	sscanf(buf, "%u", &value);
-	if (!((value == CTCM_PROTO_S390)  ||
+	rc = sscanf(buf, "%d", &value);
+	if ((rc != 1) ||
+	    !((value == CTCM_PROTO_S390)  ||
 	      (value == CTCM_PROTO_LINUX) ||
 	      (value == CTCM_PROTO_MPC) ||
 	      (value == CTCM_PROTO_OS390)))
diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c
index c461f2aac610..0a7d87c372b8 100644
--- a/drivers/s390/net/lcs.c
+++ b/drivers/s390/net/lcs.c
@@ -1943,14 +1943,16 @@ static ssize_t
 lcs_portno_store (struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
 {
         struct lcs_card *card;
-        int value;
+	int value, rc;
 
 	card = dev_get_drvdata(dev);
 
         if (!card)
                 return 0;
 
-        sscanf(buf, "%u", &value);
+	rc = sscanf(buf, "%d", &value);
+	if (rc != 1)
+		return -EINVAL;
         /* TODO: sanity checks */
         card->portno = value;
 
@@ -1997,14 +1999,17 @@ static ssize_t
 lcs_timeout_store (struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
 {
         struct lcs_card *card;
-        int value;
+	unsigned int value;
+	int rc;
 
 	card = dev_get_drvdata(dev);
 
         if (!card)
                 return 0;
 
-        sscanf(buf, "%u", &value);
+	rc = sscanf(buf, "%u", &value);
+	if (rc != 1)
+		return -EINVAL;
         /* TODO: sanity checks */
         card->lancmd_timeout = value;
 
@@ -2442,7 +2447,7 @@ __init lcs_init_module(void)
 	if (rc)
 		goto out_err;
 	lcs_root_dev = root_device_register("lcs");
-	rc = PTR_RET(lcs_root_dev);
+	rc = PTR_ERR_OR_ZERO(lcs_root_dev);
 	if (rc)
 		goto register_err;
 	rc = ccw_driver_register(&lcs_ccw_driver);
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 5333b2c018e7..a2088af51cc5 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -268,10 +268,8 @@ static inline int qeth_is_ipa_enabled(struct qeth_ipa_info *ipa,
 #define QETH_NO_PRIO_QUEUEING 0
 #define QETH_PRIO_Q_ING_PREC  1
 #define QETH_PRIO_Q_ING_TOS   2
-#define IP_TOS_LOWDELAY 0x10
-#define IP_TOS_HIGHTHROUGHPUT 0x08
-#define IP_TOS_HIGHRELIABILITY 0x04
-#define IP_TOS_NOTIMPORTANT 0x02
+#define QETH_PRIO_Q_ING_SKB   3
+#define QETH_PRIO_Q_ING_VLAN  4
 
 /* Packing */
 #define QETH_LOW_WATERMARK_PACK  2
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index e89f38c31176..f54bec54d677 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -20,6 +20,7 @@
 #include <linux/kthread.h>
 #include <linux/slab.h>
 #include <net/iucv/af_iucv.h>
+#include <net/dsfield.h>
 
 #include <asm/ebcdic.h>
 #include <asm/chpid.h>
@@ -1013,7 +1014,7 @@ static long __qeth_check_irb_error(struct ccw_device *cdev,
 
 	card = CARD_FROM_CDEV(cdev);
 
-	if (!IS_ERR(irb))
+	if (!card || !IS_ERR(irb))
 		return 0;
 
 	switch (PTR_ERR(irb)) {
@@ -1029,7 +1030,7 @@ static long __qeth_check_irb_error(struct ccw_device *cdev,
 		QETH_CARD_TEXT(card, 2, "ckirberr");
 		QETH_CARD_TEXT_(card, 2, "  rc%d", -ETIMEDOUT);
 		if (intparm == QETH_RCD_PARM) {
-			if (card && (card->data.ccwdev == cdev)) {
+			if (card->data.ccwdev == cdev) {
 				card->data.state = CH_STATE_DOWN;
 				wake_up(&card->wait_q);
 			}
@@ -3662,42 +3663,56 @@ void qeth_qdio_output_handler(struct ccw_device *ccwdev,
 }
 EXPORT_SYMBOL_GPL(qeth_qdio_output_handler);
 
+/**
+ * Note: Function assumes that we have 4 outbound queues.
+ */
 int qeth_get_priority_queue(struct qeth_card *card, struct sk_buff *skb,
 			int ipv, int cast_type)
 {
-	if (!ipv && (card->info.type == QETH_CARD_TYPE_OSD ||
-		     card->info.type == QETH_CARD_TYPE_OSX))
-		return card->qdio.default_out_queue;
-	switch (card->qdio.no_out_queues) {
-	case 4:
-		if (cast_type && card->info.is_multicast_different)
-			return card->info.is_multicast_different &
-				(card->qdio.no_out_queues - 1);
-		if (card->qdio.do_prio_queueing && (ipv == 4)) {
-			const u8 tos = ip_hdr(skb)->tos;
-
-			if (card->qdio.do_prio_queueing ==
-				QETH_PRIO_Q_ING_TOS) {
-				if (tos & IP_TOS_NOTIMPORTANT)
-					return 3;
-				if (tos & IP_TOS_HIGHRELIABILITY)
-					return 2;
-				if (tos & IP_TOS_HIGHTHROUGHPUT)
-					return 1;
-				if (tos & IP_TOS_LOWDELAY)
-					return 0;
-			}
-			if (card->qdio.do_prio_queueing ==
-				QETH_PRIO_Q_ING_PREC)
-				return 3 - (tos >> 6);
-		} else if (card->qdio.do_prio_queueing && (ipv == 6)) {
-			/* TODO: IPv6!!! */
+	__be16 *tci;
+	u8 tos;
+
+	if (cast_type && card->info.is_multicast_different)
+		return card->info.is_multicast_different &
+			(card->qdio.no_out_queues - 1);
+
+	switch (card->qdio.do_prio_queueing) {
+	case QETH_PRIO_Q_ING_TOS:
+	case QETH_PRIO_Q_ING_PREC:
+		switch (ipv) {
+		case 4:
+			tos = ipv4_get_dsfield(ip_hdr(skb));
+			break;
+		case 6:
+			tos = ipv6_get_dsfield(ipv6_hdr(skb));
+			break;
+		default:
+			return card->qdio.default_out_queue;
 		}
-		return card->qdio.default_out_queue;
-	case 1: /* fallthrough for single-out-queue 1920-device */
+		if (card->qdio.do_prio_queueing == QETH_PRIO_Q_ING_PREC)
+			return ~tos >> 6 & 3;
+		if (tos & IPTOS_MINCOST)
+			return 3;
+		if (tos & IPTOS_RELIABILITY)
+			return 2;
+		if (tos & IPTOS_THROUGHPUT)
+			return 1;
+		if (tos & IPTOS_LOWDELAY)
+			return 0;
+		break;
+	case QETH_PRIO_Q_ING_SKB:
+		if (skb->priority > 5)
+			return 0;
+		return ~skb->priority >> 1 & 3;
+	case QETH_PRIO_Q_ING_VLAN:
+		tci = &((struct ethhdr *)skb->data)->h_proto;
+		if (*tci == ETH_P_8021Q)
+			return ~*(tci + 1) >> (VLAN_PRIO_SHIFT + 1) & 3;
+		break;
 	default:
-		return card->qdio.default_out_queue;
+		break;
 	}
+	return card->qdio.default_out_queue;
 }
 EXPORT_SYMBOL_GPL(qeth_get_priority_queue);
 
@@ -5703,6 +5718,7 @@ int qeth_core_ethtool_get_settings(struct net_device *netdev,
 	struct qeth_card *card = netdev->ml_priv;
 	enum qeth_link_types link_type;
 	struct carrier_info carrier_info;
+	u32 speed;
 
 	if ((card->info.type == QETH_CARD_TYPE_IQD) || (card->info.guestlan))
 		link_type = QETH_LINK_TYPE_10GBIT_ETH;
@@ -5717,28 +5733,29 @@ int qeth_core_ethtool_get_settings(struct net_device *netdev,
 	case QETH_LINK_TYPE_FAST_ETH:
 	case QETH_LINK_TYPE_LANE_ETH100:
 		qeth_set_ecmd_adv_sup(ecmd, SPEED_100, PORT_TP);
-		ecmd->speed = SPEED_100;
+		speed = SPEED_100;
 		ecmd->port = PORT_TP;
 		break;
 
 	case QETH_LINK_TYPE_GBIT_ETH:
 	case QETH_LINK_TYPE_LANE_ETH1000:
 		qeth_set_ecmd_adv_sup(ecmd, SPEED_1000, PORT_FIBRE);
-		ecmd->speed = SPEED_1000;
+		speed = SPEED_1000;
 		ecmd->port = PORT_FIBRE;
 		break;
 
 	case QETH_LINK_TYPE_10GBIT_ETH:
 		qeth_set_ecmd_adv_sup(ecmd, SPEED_10000, PORT_FIBRE);
-		ecmd->speed = SPEED_10000;
+		speed = SPEED_10000;
 		ecmd->port = PORT_FIBRE;
 		break;
 
 	default:
 		qeth_set_ecmd_adv_sup(ecmd, SPEED_10, PORT_TP);
-		ecmd->speed = SPEED_10;
+		speed = SPEED_10;
 		ecmd->port = PORT_TP;
 	}
+	ethtool_cmd_speed_set(ecmd, speed);
 
 	/* Check if we can obtain more accurate information.	 */
 	/* If QUERY_CARD_INFO command is not supported or fails, */
@@ -5783,18 +5800,19 @@ int qeth_core_ethtool_get_settings(struct net_device *netdev,
 
 	switch (carrier_info.port_speed) {
 	case CARD_INFO_PORTS_10M:
-		ecmd->speed = SPEED_10;
+		speed = SPEED_10;
 		break;
 	case CARD_INFO_PORTS_100M:
-		ecmd->speed = SPEED_100;
+		speed = SPEED_100;
 		break;
 	case CARD_INFO_PORTS_1G:
-		ecmd->speed = SPEED_1000;
+		speed = SPEED_1000;
 		break;
 	case CARD_INFO_PORTS_10G:
-		ecmd->speed = SPEED_10000;
+		speed = SPEED_10000;
 		break;
 	}
+	ethtool_cmd_speed_set(ecmd, speed);
 
 	return 0;
 }
@@ -5816,7 +5834,7 @@ static int __init qeth_core_init(void)
 	if (rc)
 		goto out_err;
 	qeth_core_root_dev = root_device_register("qeth");
-	rc = PTR_RET(qeth_core_root_dev);
+	rc = PTR_ERR_OR_ZERO(qeth_core_root_dev);
 	if (rc)
 		goto register_err;
 	qeth_core_header_cache = kmem_cache_create("qeth_hdr",
diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c
index 425c0ecf1f3b..8a25a2be9890 100644
--- a/drivers/s390/net/qeth_core_sys.c
+++ b/drivers/s390/net/qeth_core_sys.c
@@ -217,6 +217,10 @@ static ssize_t qeth_dev_prioqing_show(struct device *dev,
 		return sprintf(buf, "%s\n", "by precedence");
 	case QETH_PRIO_Q_ING_TOS:
 		return sprintf(buf, "%s\n", "by type of service");
+	case QETH_PRIO_Q_ING_SKB:
+		return sprintf(buf, "%s\n", "by skb-priority");
+	case QETH_PRIO_Q_ING_VLAN:
+		return sprintf(buf, "%s\n", "by VLAN headers");
 	default:
 		return sprintf(buf, "always queue %i\n",
 			       card->qdio.default_out_queue);
@@ -250,11 +254,23 @@ static ssize_t qeth_dev_prioqing_store(struct device *dev,
 	}
 
 	tmp = strsep((char **) &buf, "\n");
-	if (!strcmp(tmp, "prio_queueing_prec"))
+	if (!strcmp(tmp, "prio_queueing_prec")) {
 		card->qdio.do_prio_queueing = QETH_PRIO_Q_ING_PREC;
-	else if (!strcmp(tmp, "prio_queueing_tos"))
+		card->qdio.default_out_queue = QETH_DEFAULT_QUEUE;
+	} else if (!strcmp(tmp, "prio_queueing_skb")) {
+		card->qdio.do_prio_queueing = QETH_PRIO_Q_ING_SKB;
+		card->qdio.default_out_queue = QETH_DEFAULT_QUEUE;
+	} else if (!strcmp(tmp, "prio_queueing_tos")) {
 		card->qdio.do_prio_queueing = QETH_PRIO_Q_ING_TOS;
-	else if (!strcmp(tmp, "no_prio_queueing:0")) {
+		card->qdio.default_out_queue = QETH_DEFAULT_QUEUE;
+	} else if (!strcmp(tmp, "prio_queueing_vlan")) {
+		if (!card->options.layer2) {
+			rc = -ENOTSUPP;
+			goto out;
+		}
+		card->qdio.do_prio_queueing = QETH_PRIO_Q_ING_VLAN;
+		card->qdio.default_out_queue = QETH_DEFAULT_QUEUE;
+	} else if (!strcmp(tmp, "no_prio_queueing:0")) {
 		card->qdio.do_prio_queueing = QETH_NO_PRIO_QUEUEING;
 		card->qdio.default_out_queue = 0;
 	} else if (!strcmp(tmp, "no_prio_queueing:1")) {
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 8dea3f12ccc1..5ef5b4f45758 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -725,15 +725,20 @@ static int qeth_l2_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	int elements = 0;
 	struct qeth_card *card = dev->ml_priv;
 	struct sk_buff *new_skb = skb;
-	int ipv = qeth_get_ip_version(skb);
 	int cast_type = qeth_l2_get_cast_type(card, skb);
-	struct qeth_qdio_out_q *queue = card->qdio.out_qs
-		[qeth_get_priority_queue(card, skb, ipv, cast_type)];
+	struct qeth_qdio_out_q *queue;
 	int tx_bytes = skb->len;
 	int data_offset = -1;
 	int elements_needed = 0;
 	int hd_len = 0;
 
+	if (card->qdio.do_prio_queueing || (cast_type &&
+					card->info.is_multicast_different))
+		queue = card->qdio.out_qs[qeth_get_priority_queue(card, skb,
+					qeth_get_ip_version(skb), cast_type)];
+	else
+		queue = card->qdio.out_qs[card->qdio.default_out_queue];
+
 	if ((card->state != CARD_STATE_UP) || !card->lan_online) {
 		card->stats.tx_carrier_errors++;
 		goto tx_drop;
@@ -964,10 +969,9 @@ static int qeth_l2_setup_netdev(struct qeth_card *card)
 	card->dev->watchdog_timeo = QETH_TX_TIMEOUT;
 	card->dev->mtu = card->info.initial_mtu;
 	card->dev->netdev_ops = &qeth_l2_netdev_ops;
-	if (card->info.type != QETH_CARD_TYPE_OSN)
-		SET_ETHTOOL_OPS(card->dev, &qeth_l2_ethtool_ops);
-	else
-		SET_ETHTOOL_OPS(card->dev, &qeth_l2_osn_ops);
+	card->dev->ethtool_ops =
+		(card->info.type != QETH_CARD_TYPE_OSN) ?
+		&qeth_l2_ethtool_ops : &qeth_l2_osn_ops;
 	card->dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
 	card->info.broadcast_capable = 1;
 	qeth_l2_request_initial_mac(card);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 3524d34ff694..14e0b5810e8c 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -63,7 +63,7 @@ void qeth_l3_ipaddr4_to_string(const __u8 *addr, char *buf)
 int qeth_l3_string_to_ipaddr4(const char *buf, __u8 *addr)
 {
 	int count = 0, rc = 0;
-	int in[4];
+	unsigned int in[4];
 	char c;
 
 	rc = sscanf(buf, "%u.%u.%u.%u%c",
@@ -1659,7 +1659,7 @@ static void qeth_l3_add_vlan_mc(struct qeth_card *card)
 	for_each_set_bit(vid, card->active_vlans, VLAN_N_VID) {
 		struct net_device *netdev;
 
-		netdev = __vlan_find_dev_deep(card->dev, htons(ETH_P_8021Q),
+		netdev = __vlan_find_dev_deep_rcu(card->dev, htons(ETH_P_8021Q),
 					      vid);
 		if (netdev == NULL ||
 		    !(netdev->flags & IFF_UP))
@@ -1721,7 +1721,7 @@ static void qeth_l3_add_vlan_mc6(struct qeth_card *card)
 	for_each_set_bit(vid, card->active_vlans, VLAN_N_VID) {
 		struct net_device *netdev;
 
-		netdev = __vlan_find_dev_deep(card->dev, htons(ETH_P_8021Q),
+		netdev = __vlan_find_dev_deep_rcu(card->dev, htons(ETH_P_8021Q),
 					      vid);
 		if (netdev == NULL ||
 		    !(netdev->flags & IFF_UP))
@@ -1766,7 +1766,7 @@ static void qeth_l3_free_vlan_addresses4(struct qeth_card *card,
 
 	QETH_CARD_TEXT(card, 4, "frvaddr4");
 
-	netdev = __vlan_find_dev_deep(card->dev, htons(ETH_P_8021Q), vid);
+	netdev = __vlan_find_dev_deep_rcu(card->dev, htons(ETH_P_8021Q), vid);
 	if (!netdev)
 		return;
 	in_dev = in_dev_get(netdev);
@@ -1796,7 +1796,7 @@ static void qeth_l3_free_vlan_addresses6(struct qeth_card *card,
 
 	QETH_CARD_TEXT(card, 4, "frvaddr6");
 
-	netdev = __vlan_find_dev_deep(card->dev, htons(ETH_P_8021Q), vid);
+	netdev = __vlan_find_dev_deep_rcu(card->dev, htons(ETH_P_8021Q), vid);
 	if (!netdev)
 		return;
 	in6_dev = in6_dev_get(netdev);
@@ -2089,7 +2089,7 @@ static int qeth_l3_verify_vlan_dev(struct net_device *dev,
 		struct net_device *netdev;
 
 		rcu_read_lock();
-		netdev = __vlan_find_dev_deep(card->dev, htons(ETH_P_8021Q),
+		netdev = __vlan_find_dev_deep_rcu(card->dev, htons(ETH_P_8021Q),
 					      vid);
 		rcu_read_unlock();
 		if (netdev == dev) {
@@ -2926,8 +2926,11 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct sk_buff *new_skb = NULL;
 	int ipv = qeth_get_ip_version(skb);
 	int cast_type = qeth_l3_get_cast_type(card, skb);
-	struct qeth_qdio_out_q *queue = card->qdio.out_qs
-		[qeth_get_priority_queue(card, skb, ipv, cast_type)];
+	struct qeth_qdio_out_q *queue =
+		card->qdio.out_qs[card->qdio.do_prio_queueing
+			|| (cast_type && card->info.is_multicast_different) ?
+			qeth_get_priority_queue(card, skb, ipv, cast_type) :
+			card->qdio.default_out_queue];
 	int tx_bytes = skb->len;
 	bool large_send;
 	int data_offset = -1;
@@ -3298,7 +3301,7 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 	card->dev->ml_priv = card;
 	card->dev->watchdog_timeo = QETH_TX_TIMEOUT;
 	card->dev->mtu = card->info.initial_mtu;
-	SET_ETHTOOL_OPS(card->dev, &qeth_l3_ethtool_ops);
+	card->dev->ethtool_ops = &qeth_l3_ethtool_ops;
 	card->dev->features |=	NETIF_F_HW_VLAN_CTAG_TX |
 				NETIF_F_HW_VLAN_CTAG_RX |
 				NETIF_F_HW_VLAN_CTAG_FILTER;
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index 02832d64d918..baca5897039f 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -1773,6 +1773,7 @@ config SCSI_BFA_FC
 config SCSI_VIRTIO
 	tristate "virtio-scsi support"
 	depends on VIRTIO
+	select BLK_DEV_INTEGRITY
 	help
           This is the virtual HBA driver for virtio.  If the kernel will
           be used in a virtual machine, say Y or M.
diff --git a/drivers/scsi/NCR5380.c b/drivers/scsi/NCR5380.c
index bcd223868227..93d13fc9a293 100644
--- a/drivers/scsi/NCR5380.c
+++ b/drivers/scsi/NCR5380.c
@@ -27,8 +27,6 @@
  */
 
 /*
- * $Log: NCR5380.c,v $
-
  * Revision 1.10 1998/9/2	Alan Cox
  *				(alan@lxorguk.ukuu.org.uk)
  * Fixed up the timer lockups reported so far. Things still suck. Looking 
@@ -89,13 +87,6 @@
 #include <scsi/scsi_dbg.h>
 #include <scsi/scsi_transport_spi.h>
 
-#ifndef NDEBUG
-#define NDEBUG 0
-#endif
-#ifndef NDEBUG_ABORT
-#define NDEBUG_ABORT 0
-#endif
-
 #if (NDEBUG & NDEBUG_LISTS)
 #define LIST(x,y) {printk("LINE:%d   Adding %p to %p\n", __LINE__, (void*)(x), (void*)(y)); if ((x)==(y)) udelay(5); }
 #define REMOVE(w,x,y,z) {printk("LINE:%d   Removing: %p->%p  %p->%p \n", __LINE__, (void*)(w), (void*)(x), (void*)(y), (void*)(z)); if ((x)==(y)) udelay(5); }
@@ -1005,7 +996,7 @@ static int NCR5380_queue_command_lck(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)
 		LIST(cmd, tmp);
 		tmp->host_scribble = (unsigned char *) cmd;
 	}
-	dprintk(NDEBUG_QUEUES, ("scsi%d : command added to %s of queue\n", instance->host_no, (cmd->cmnd[0] == REQUEST_SENSE) ? "head" : "tail"));
+	dprintk(NDEBUG_QUEUES, "scsi%d : command added to %s of queue\n", instance->host_no, (cmd->cmnd[0] == REQUEST_SENSE) ? "head" : "tail");
 
 	/* Run the coroutine if it isn't already running. */
 	/* Kick off command processing */
@@ -1040,7 +1031,7 @@ static void NCR5380_main(struct work_struct *work)
 		/* Lock held here */
 		done = 1;
 		if (!hostdata->connected && !hostdata->selecting) {
-			dprintk(NDEBUG_MAIN, ("scsi%d : not connected\n", instance->host_no));
+			dprintk(NDEBUG_MAIN, "scsi%d : not connected\n", instance->host_no);
 			/*
 			 * Search through the issue_queue for a command destined
 			 * for a target that's not busy.
@@ -1048,7 +1039,7 @@ static void NCR5380_main(struct work_struct *work)
 			for (tmp = (Scsi_Cmnd *) hostdata->issue_queue, prev = NULL; tmp; prev = tmp, tmp = (Scsi_Cmnd *) tmp->host_scribble) 
 			{
 				if (prev != tmp)
-					dprintk(NDEBUG_LISTS, ("MAIN tmp=%p   target=%d   busy=%d lun=%d\n", tmp, tmp->target, hostdata->busy[tmp->target], tmp->lun));
+					dprintk(NDEBUG_LISTS, "MAIN tmp=%p   target=%d   busy=%d lun=%d\n", tmp, tmp->device->id, hostdata->busy[tmp->device->id], tmp->device->lun);
 				/*  When we find one, remove it from the issue queue. */
 				if (!(hostdata->busy[tmp->device->id] & (1 << tmp->device->lun))) {
 					if (prev) {
@@ -1066,7 +1057,7 @@ static void NCR5380_main(struct work_struct *work)
 					 * On failure, we must add the command back to the
 					 *   issue queue so we can keep trying. 
 					 */
-					dprintk(NDEBUG_MAIN|NDEBUG_QUEUES, ("scsi%d : main() : command for target %d lun %d removed from issue_queue\n", instance->host_no, tmp->target, tmp->lun));
+					dprintk(NDEBUG_MAIN|NDEBUG_QUEUES, "scsi%d : main() : command for target %d lun %d removed from issue_queue\n", instance->host_no, tmp->device->id, tmp->device->lun);
 	
 					/*
 					 * A successful selection is defined as one that 
@@ -1095,7 +1086,7 @@ static void NCR5380_main(struct work_struct *work)
 						tmp->host_scribble = (unsigned char *) hostdata->issue_queue;
 						hostdata->issue_queue = tmp;
 						done = 0;
-						dprintk(NDEBUG_MAIN|NDEBUG_QUEUES, ("scsi%d : main(): select() failed, returned to issue_queue\n", instance->host_no));
+						dprintk(NDEBUG_MAIN|NDEBUG_QUEUES, "scsi%d : main(): select() failed, returned to issue_queue\n", instance->host_no);
 					}
 					/* lock held here still */
 				}	/* if target/lun is not busy */
@@ -1125,9 +1116,9 @@ static void NCR5380_main(struct work_struct *work)
 #endif
 		    && (!hostdata->time_expires || time_before_eq(hostdata->time_expires, jiffies))
 		    ) {
-			dprintk(NDEBUG_MAIN, ("scsi%d : main() : performing information transfer\n", instance->host_no));
+			dprintk(NDEBUG_MAIN, "scsi%d : main() : performing information transfer\n", instance->host_no);
 			NCR5380_information_transfer(instance);
-			dprintk(NDEBUG_MAIN, ("scsi%d : main() : done set false\n", instance->host_no));
+			dprintk(NDEBUG_MAIN, "scsi%d : main() : done set false\n", instance->host_no);
 			done = 0;
 		} else
 			break;
@@ -1159,8 +1150,8 @@ static irqreturn_t NCR5380_intr(int dummy, void *dev_id)
 	unsigned char basr;
 	unsigned long flags;
 
-	dprintk(NDEBUG_INTR, ("scsi : NCR5380 irq %d triggered\n",
-		instance->irq));
+	dprintk(NDEBUG_INTR, "scsi : NCR5380 irq %d triggered\n",
+		instance->irq);
 
 	do {
 		done = 1;
@@ -1173,14 +1164,14 @@ static irqreturn_t NCR5380_intr(int dummy, void *dev_id)
 			NCR5380_dprint(NDEBUG_INTR, instance);
 			if ((NCR5380_read(STATUS_REG) & (SR_SEL | SR_IO)) == (SR_SEL | SR_IO)) {
 				done = 0;
-				dprintk(NDEBUG_INTR, ("scsi%d : SEL interrupt\n", instance->host_no));
+				dprintk(NDEBUG_INTR, "scsi%d : SEL interrupt\n", instance->host_no);
 				NCR5380_reselect(instance);
 				(void) NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 			} else if (basr & BASR_PARITY_ERROR) {
-				dprintk(NDEBUG_INTR, ("scsi%d : PARITY interrupt\n", instance->host_no));
+				dprintk(NDEBUG_INTR, "scsi%d : PARITY interrupt\n", instance->host_no);
 				(void) NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 			} else if ((NCR5380_read(STATUS_REG) & SR_RST) == SR_RST) {
-				dprintk(NDEBUG_INTR, ("scsi%d : RESET interrupt\n", instance->host_no));
+				dprintk(NDEBUG_INTR, "scsi%d : RESET interrupt\n", instance->host_no);
 				(void) NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 			} else {
 #if defined(REAL_DMA)
@@ -1210,7 +1201,7 @@ static irqreturn_t NCR5380_intr(int dummy, void *dev_id)
 					NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 				}
 #else
-				dprintk(NDEBUG_INTR, ("scsi : unknown interrupt, BASR 0x%X, MR 0x%X, SR 0x%x\n", basr, NCR5380_read(MODE_REG), NCR5380_read(STATUS_REG)));
+				dprintk(NDEBUG_INTR, "scsi : unknown interrupt, BASR 0x%X, MR 0x%X, SR 0x%x\n", basr, NCR5380_read(MODE_REG), NCR5380_read(STATUS_REG));
 				(void) NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 #endif
 			}
@@ -1304,7 +1295,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd * cmd, int tag)
 	hostdata->restart_select = 0;
 
 	NCR5380_dprint(NDEBUG_ARBITRATION, instance);
-	dprintk(NDEBUG_ARBITRATION, ("scsi%d : starting arbitration, id = %d\n", instance->host_no, instance->this_id));
+	dprintk(NDEBUG_ARBITRATION, "scsi%d : starting arbitration, id = %d\n", instance->host_no, instance->this_id);
 
 	/* 
 	 * Set the phase bits to 0, otherwise the NCR5380 won't drive the 
@@ -1333,7 +1324,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd * cmd, int tag)
 		goto failed;
 	}
 
-	dprintk(NDEBUG_ARBITRATION, ("scsi%d : arbitration complete\n", instance->host_no));
+	dprintk(NDEBUG_ARBITRATION, "scsi%d : arbitration complete\n", instance->host_no);
 
 	/* 
 	 * The arbitration delay is 2.2us, but this is a minimum and there is 
@@ -1347,7 +1338,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd * cmd, int tag)
 	/* Check for lost arbitration */
 	if ((NCR5380_read(INITIATOR_COMMAND_REG) & ICR_ARBITRATION_LOST) || (NCR5380_read(CURRENT_SCSI_DATA_REG) & hostdata->id_higher_mask) || (NCR5380_read(INITIATOR_COMMAND_REG) & ICR_ARBITRATION_LOST)) {
 		NCR5380_write(MODE_REG, MR_BASE);
-		dprintk(NDEBUG_ARBITRATION, ("scsi%d : lost arbitration, deasserting MR_ARBITRATE\n", instance->host_no));
+		dprintk(NDEBUG_ARBITRATION, "scsi%d : lost arbitration, deasserting MR_ARBITRATE\n", instance->host_no);
 		goto failed;
 	}
 	NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | ICR_ASSERT_SEL);
@@ -1360,7 +1351,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd * cmd, int tag)
 	    (NCR5380_read(INITIATOR_COMMAND_REG) & ICR_ARBITRATION_LOST)) {
 		NCR5380_write(MODE_REG, MR_BASE);
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-		dprintk(NDEBUG_ARBITRATION, ("scsi%d : lost arbitration, deasserting ICR_ASSERT_SEL\n", instance->host_no));
+		dprintk(NDEBUG_ARBITRATION, "scsi%d : lost arbitration, deasserting ICR_ASSERT_SEL\n", instance->host_no);
 		goto failed;
 	}
 	/* 
@@ -1370,7 +1361,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd * cmd, int tag)
 
 	udelay(2);
 
-	dprintk(NDEBUG_ARBITRATION, ("scsi%d : won arbitration\n", instance->host_no));
+	dprintk(NDEBUG_ARBITRATION, "scsi%d : won arbitration\n", instance->host_no);
 
 	/* 
 	 * Now that we have won arbitration, start Selection process, asserting 
@@ -1422,7 +1413,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd * cmd, int tag)
 
 	udelay(1);
 
-	dprintk(NDEBUG_SELECTION, ("scsi%d : selecting target %d\n", instance->host_no, scmd_id(cmd)));
+	dprintk(NDEBUG_SELECTION, "scsi%d : selecting target %d\n", instance->host_no, scmd_id(cmd));
 
 	/* 
 	 * The SCSI specification calls for a 250 ms timeout for the actual 
@@ -1487,7 +1478,7 @@ part2:
 		collect_stats(hostdata, cmd);
 		cmd->scsi_done(cmd);
 		NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
-		dprintk(NDEBUG_SELECTION, ("scsi%d : target did not respond within 250ms\n", instance->host_no));
+		dprintk(NDEBUG_SELECTION, "scsi%d : target did not respond within 250ms\n", instance->host_no);
 		NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 		return 0;
 	}
@@ -1520,7 +1511,7 @@ part2:
 		goto failed;
 	}
 
-	dprintk(NDEBUG_SELECTION, ("scsi%d : target %d selected, going into MESSAGE OUT phase.\n", instance->host_no, cmd->device->id));
+	dprintk(NDEBUG_SELECTION, "scsi%d : target %d selected, going into MESSAGE OUT phase.\n", instance->host_no, cmd->device->id);
 	tmp[0] = IDENTIFY(((instance->irq == SCSI_IRQ_NONE) ? 0 : 1), cmd->device->lun);
 
 	len = 1;
@@ -1530,7 +1521,7 @@ part2:
 	data = tmp;
 	phase = PHASE_MSGOUT;
 	NCR5380_transfer_pio(instance, &phase, &len, &data);
-	dprintk(NDEBUG_SELECTION, ("scsi%d : nexus established.\n", instance->host_no));
+	dprintk(NDEBUG_SELECTION, "scsi%d : nexus established.\n", instance->host_no);
 	/* XXX need to handle errors here */
 	hostdata->connected = cmd;
 	hostdata->busy[cmd->device->id] |= (1 << cmd->device->lun);
@@ -1583,9 +1574,9 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance, unsigned char *phase
 	NCR5380_setup(instance);
 
 	if (!(p & SR_IO))
-		dprintk(NDEBUG_PIO, ("scsi%d : pio write %d bytes\n", instance->host_no, c));
+		dprintk(NDEBUG_PIO, "scsi%d : pio write %d bytes\n", instance->host_no, c);
 	else
-		dprintk(NDEBUG_PIO, ("scsi%d : pio read %d bytes\n", instance->host_no, c));
+		dprintk(NDEBUG_PIO, "scsi%d : pio read %d bytes\n", instance->host_no, c);
 
 	/* 
 	 * The NCR5380 chip will only drive the SCSI bus when the 
@@ -1620,11 +1611,11 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance, unsigned char *phase
 			break;
 		}
 
-		dprintk(NDEBUG_HANDSHAKE, ("scsi%d : REQ detected\n", instance->host_no));
+		dprintk(NDEBUG_HANDSHAKE, "scsi%d : REQ detected\n", instance->host_no);
 
 		/* Check for phase mismatch */
 		if ((tmp & PHASE_MASK) != p) {
-			dprintk(NDEBUG_HANDSHAKE, ("scsi%d : phase mismatch\n", instance->host_no));
+			dprintk(NDEBUG_HANDSHAKE, "scsi%d : phase mismatch\n", instance->host_no);
 			NCR5380_dprint_phase(NDEBUG_HANDSHAKE, instance);
 			break;
 		}
@@ -1660,7 +1651,7 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance, unsigned char *phase
 
 		/* FIXME - if this fails bus reset ?? */
 		NCR5380_poll_politely(instance, STATUS_REG, SR_REQ, 0, 5*HZ);
-		dprintk(NDEBUG_HANDSHAKE, ("scsi%d : req false, handshake complete\n", instance->host_no));
+		dprintk(NDEBUG_HANDSHAKE, "scsi%d : req false, handshake complete\n", instance->host_no);
 
 /*
  * We have several special cases to consider during REQ/ACK handshaking : 
@@ -1681,7 +1672,7 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance, unsigned char *phase
 		}
 	} while (--c);
 
-	dprintk(NDEBUG_PIO, ("scsi%d : residual %d\n", instance->host_no, c));
+	dprintk(NDEBUG_PIO, "scsi%d : residual %d\n", instance->host_no, c);
 
 	*count = c;
 	*data = d;
@@ -1828,7 +1819,7 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 		c -= 2;
 	}
 #endif
-	dprintk(NDEBUG_DMA, ("scsi%d : initializing DMA channel %d for %s, %d bytes %s %0x\n", instance->host_no, instance->dma_channel, (p & SR_IO) ? "reading" : "writing", c, (p & SR_IO) ? "to" : "from", (unsigned) d));
+	dprintk(NDEBUG_DMA, "scsi%d : initializing DMA channel %d for %s, %d bytes %s %0x\n", instance->host_no, instance->dma_channel, (p & SR_IO) ? "reading" : "writing", c, (p & SR_IO) ? "to" : "from", (unsigned) d);
 	hostdata->dma_len = (p & SR_IO) ? NCR5380_dma_read_setup(instance, d, c) : NCR5380_dma_write_setup(instance, d, c);
 #endif
 
@@ -1857,7 +1848,7 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 		NCR5380_write(MODE_REG, MR_BASE | MR_DMA_MODE);
 #endif				/* def REAL_DMA */
 
-	dprintk(NDEBUG_DMA, ("scsi%d : mode reg = 0x%X\n", instance->host_no, NCR5380_read(MODE_REG)));
+	dprintk(NDEBUG_DMA, "scsi%d : mode reg = 0x%X\n", instance->host_no, NCR5380_read(MODE_REG));
 
 	/* 
 	 *	On the PAS16 at least I/O recovery delays are not needed here.
@@ -1934,7 +1925,7 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 		}
 	}
 
-	dprintk(NDEBUG_DMA, ("scsi%d : polled DMA transfer complete, basr 0x%X, sr 0x%X\n", instance->host_no, tmp, NCR5380_read(STATUS_REG)));
+	dprintk(NDEBUG_DMA, "scsi%d : polled DMA transfer complete, basr 0x%X, sr 0x%X\n", instance->host_no, tmp, NCR5380_read(STATUS_REG));
 
 	NCR5380_write(MODE_REG, MR_BASE);
 	NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
@@ -1948,7 +1939,7 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 #ifdef READ_OVERRUNS
 	if (*phase == p && (p & SR_IO) && residue == 0) {
 		if (overrun) {
-			dprintk(NDEBUG_DMA, ("Got an input overrun, using saved byte\n"));
+			dprintk(NDEBUG_DMA, "Got an input overrun, using saved byte\n");
 			**data = saved_data;
 			*data += 1;
 			*count -= 1;
@@ -1957,13 +1948,13 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 			printk("No overrun??\n");
 			cnt = toPIO = 2;
 		}
-		dprintk(NDEBUG_DMA, ("Doing %d-byte PIO to 0x%X\n", cnt, *data));
+		dprintk(NDEBUG_DMA, "Doing %d-byte PIO to 0x%X\n", cnt, *data);
 		NCR5380_transfer_pio(instance, phase, &cnt, data);
 		*count -= toPIO - cnt;
 	}
 #endif
 
-	dprintk(NDEBUG_DMA, ("Return with data ptr = 0x%X, count %d, last 0x%X, next 0x%X\n", *data, *count, *(*data + *count - 1), *(*data + *count)));
+	dprintk(NDEBUG_DMA, "Return with data ptr = 0x%X, count %d, last 0x%X, next 0x%X\n", *data, *count, *(*data + *count - 1), *(*data + *count));
 	return 0;
 
 #elif defined(REAL_DMA)
@@ -2013,7 +2004,7 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 		foo = NCR5380_pwrite(instance, d, c);
 #else
 		int timeout;
-		dprintk(NDEBUG_C400_PWRITE, ("About to pwrite %d bytes\n", c));
+		dprintk(NDEBUG_C400_PWRITE, "About to pwrite %d bytes\n", c);
 		if (!(foo = NCR5380_pwrite(instance, d, c))) {
 			/*
 			 * Wait for the last byte to be sent.  If REQ is being asserted for 
@@ -2024,19 +2015,19 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 				while (!(NCR5380_read(BUS_AND_STATUS_REG) & BASR_DRQ) && (NCR5380_read(BUS_AND_STATUS_REG) & BASR_PHASE_MATCH));
 
 				if (!timeout)
-					dprintk(NDEBUG_LAST_BYTE_SENT, ("scsi%d : timed out on last byte\n", instance->host_no));
+					dprintk(NDEBUG_LAST_BYTE_SENT, "scsi%d : timed out on last byte\n", instance->host_no);
 
 				if (hostdata->flags & FLAG_CHECK_LAST_BYTE_SENT) {
 					hostdata->flags &= ~FLAG_CHECK_LAST_BYTE_SENT;
 					if (NCR5380_read(TARGET_COMMAND_REG) & TCR_LAST_BYTE_SENT) {
 						hostdata->flags |= FLAG_HAS_LAST_BYTE_SENT;
-						dprintk(NDEBUG_LAST_WRITE_SENT, ("scsi%d : last bit sent works\n", instance->host_no));
+						dprintk(NDEBUG_LAST_BYTE_SENT, "scsi%d : last byte sent works\n", instance->host_no);
 					}
 				}
 			} else {
-				dprintk(NDEBUG_C400_PWRITE, ("Waiting for LASTBYTE\n"));
+				dprintk(NDEBUG_C400_PWRITE, "Waiting for LASTBYTE\n");
 				while (!(NCR5380_read(TARGET_COMMAND_REG) & TCR_LAST_BYTE_SENT));
-				dprintk(NDEBUG_C400_PWRITE, ("Got LASTBYTE\n"));
+				dprintk(NDEBUG_C400_PWRITE, "Got LASTBYTE\n");
 			}
 		}
 #endif
@@ -2045,9 +2036,9 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance, unsigned char *phase
 	NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 
 	if ((!(p & SR_IO)) && (hostdata->flags & FLAG_NCR53C400)) {
-		dprintk(NDEBUG_C400_PWRITE, ("53C400w: Checking for IRQ\n"));
+		dprintk(NDEBUG_C400_PWRITE, "53C400w: Checking for IRQ\n");
 		if (NCR5380_read(BUS_AND_STATUS_REG) & BASR_IRQ) {
-			dprintk(NDEBUG_C400_PWRITE, ("53C400w:    got it, reading reset interrupt reg\n"));
+			dprintk(NDEBUG_C400_PWRITE, "53C400w:    got it, reading reset interrupt reg\n");
 			NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 		} else {
 			printk("53C400w:    IRQ NOT THERE!\n");
@@ -2139,7 +2130,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 					--cmd->SCp.buffers_residual;
 					cmd->SCp.this_residual = cmd->SCp.buffer->length;
 					cmd->SCp.ptr = sg_virt(cmd->SCp.buffer);
-					dprintk(NDEBUG_INFORMATION, ("scsi%d : %d bytes and %d buffers left\n", instance->host_no, cmd->SCp.this_residual, cmd->SCp.buffers_residual));
+					dprintk(NDEBUG_INFORMATION, "scsi%d : %d bytes and %d buffers left\n", instance->host_no, cmd->SCp.this_residual, cmd->SCp.buffers_residual);
 				}
 				/*
 				 * The preferred transfer method is going to be 
@@ -2219,7 +2210,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 				case LINKED_FLG_CMD_COMPLETE:
 					/* Accept message by clearing ACK */
 					NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-					dprintk(NDEBUG_LINKED, ("scsi%d : target %d lun %d linked command complete.\n", instance->host_no, cmd->device->id, cmd->device->lun));
+					dprintk(NDEBUG_LINKED, "scsi%d : target %d lun %d linked command complete.\n", instance->host_no, cmd->device->id, cmd->device->lun);
 					/* 
 					 * Sanity check : A linked command should only terminate with
 					 * one of these messages if there are more linked commands
@@ -2235,7 +2226,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 					/* The next command is still part of this process */
 					cmd->next_link->tag = cmd->tag;
 					cmd->result = cmd->SCp.Status | (cmd->SCp.Message << 8);
-					dprintk(NDEBUG_LINKED, ("scsi%d : target %d lun %d linked request done, calling scsi_done().\n", instance->host_no, cmd->device->id, cmd->device->lun));
+					dprintk(NDEBUG_LINKED, "scsi%d : target %d lun %d linked request done, calling scsi_done().\n", instance->host_no, cmd->device->id, cmd->device->lun);
 					collect_stats(hostdata, cmd);
 					cmd->scsi_done(cmd);
 					cmd = hostdata->connected;
@@ -2247,7 +2238,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 					sink = 1;
 					NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 					hostdata->connected = NULL;
-					dprintk(NDEBUG_QUEUES, ("scsi%d : command for target %d, lun %d completed\n", instance->host_no, cmd->device->id, cmd->device->lun));
+					dprintk(NDEBUG_QUEUES, "scsi%d : command for target %d, lun %d completed\n", instance->host_no, cmd->device->id, cmd->device->lun);
 					hostdata->busy[cmd->device->id] &= ~(1 << cmd->device->lun);
 
 					/* 
@@ -2281,13 +2272,13 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 					if ((cmd->cmnd[0] != REQUEST_SENSE) && (status_byte(cmd->SCp.Status) == CHECK_CONDITION)) {
 						scsi_eh_prep_cmnd(cmd, &hostdata->ses, NULL, 0, ~0);
 
-						dprintk(NDEBUG_AUTOSENSE, ("scsi%d : performing request sense\n", instance->host_no));
+						dprintk(NDEBUG_AUTOSENSE, "scsi%d : performing request sense\n", instance->host_no);
 
 						LIST(cmd, hostdata->issue_queue);
 						cmd->host_scribble = (unsigned char *)
 						    hostdata->issue_queue;
 						hostdata->issue_queue = (Scsi_Cmnd *) cmd;
-						dprintk(NDEBUG_QUEUES, ("scsi%d : REQUEST SENSE added to head of issue queue\n", instance->host_no));
+						dprintk(NDEBUG_QUEUES, "scsi%d : REQUEST SENSE added to head of issue queue\n", instance->host_no);
 					} else
 #endif				/* def AUTOSENSE */
 					{
@@ -2327,7 +2318,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 						    hostdata->disconnected_queue;
 						hostdata->connected = NULL;
 						hostdata->disconnected_queue = cmd;
-						dprintk(NDEBUG_QUEUES, ("scsi%d : command for target %d lun %d was moved from connected to" "  the disconnected_queue\n", instance->host_no, cmd->device->id, cmd->device->lun));
+						dprintk(NDEBUG_QUEUES, "scsi%d : command for target %d lun %d was moved from connected to" "  the disconnected_queue\n", instance->host_no, cmd->device->id, cmd->device->lun);
 						/* 
 						 * Restore phase bits to 0 so an interrupted selection, 
 						 * arbitration can resume.
@@ -2373,14 +2364,14 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 					extended_msg[0] = EXTENDED_MESSAGE;
 					/* Accept first byte by clearing ACK */
 					NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-					dprintk(NDEBUG_EXTENDED, ("scsi%d : receiving extended message\n", instance->host_no));
+					dprintk(NDEBUG_EXTENDED, "scsi%d : receiving extended message\n", instance->host_no);
 
 					len = 2;
 					data = extended_msg + 1;
 					phase = PHASE_MSGIN;
 					NCR5380_transfer_pio(instance, &phase, &len, &data);
 
-					dprintk(NDEBUG_EXTENDED, ("scsi%d : length=%d, code=0x%02x\n", instance->host_no, (int) extended_msg[1], (int) extended_msg[2]));
+					dprintk(NDEBUG_EXTENDED, "scsi%d : length=%d, code=0x%02x\n", instance->host_no, (int) extended_msg[1], (int) extended_msg[2]);
 
 					if (!len && extended_msg[1] <= (sizeof(extended_msg) - 1)) {
 						/* Accept third byte by clearing ACK */
@@ -2390,7 +2381,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 						phase = PHASE_MSGIN;
 
 						NCR5380_transfer_pio(instance, &phase, &len, &data);
-						dprintk(NDEBUG_EXTENDED, ("scsi%d : message received, residual %d\n", instance->host_no, len));
+						dprintk(NDEBUG_EXTENDED, "scsi%d : message received, residual %d\n", instance->host_no, len);
 
 						switch (extended_msg[2]) {
 						case EXTENDED_SDTR:
@@ -2456,7 +2447,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 				NCR5380_transfer_pio(instance, &phase, &len, &data);
 				if (!cmd->device->disconnect && should_disconnect(cmd->cmnd[0])) {
 					NCR5380_set_timer(hostdata, USLEEP_SLEEP);
-					dprintk(NDEBUG_USLEEP, ("scsi%d : issued command, sleeping until %ul\n", instance->host_no, hostdata->time_expires));
+					dprintk(NDEBUG_USLEEP, "scsi%d : issued command, sleeping until %lu\n", instance->host_no, hostdata->time_expires);
 					return;
 				}
 				break;
@@ -2468,7 +2459,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 				break;
 			default:
 				printk("scsi%d : unknown phase\n", instance->host_no);
-				NCR5380_dprint(NDEBUG_ALL, instance);
+				NCR5380_dprint(NDEBUG_ANY, instance);
 			}	/* switch(phase) */
 		}		/* if (tmp * SR_REQ) */
 		else {
@@ -2476,7 +2467,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance) {
 			 */
 			if (!cmd->device->disconnect && time_after_eq(jiffies, poll_time)) {
 				NCR5380_set_timer(hostdata, USLEEP_SLEEP);
-				dprintk(NDEBUG_USLEEP, ("scsi%d : poll timed out, sleeping until %ul\n", instance->host_no, hostdata->time_expires));
+				dprintk(NDEBUG_USLEEP, "scsi%d : poll timed out, sleeping until %lu\n", instance->host_no, hostdata->time_expires);
 				return;
 			}
 		}
@@ -2517,7 +2508,7 @@ static void NCR5380_reselect(struct Scsi_Host *instance) {
 	hostdata->restart_select = 1;
 
 	target_mask = NCR5380_read(CURRENT_SCSI_DATA_REG) & ~(hostdata->id_mask);
-	dprintk(NDEBUG_SELECTION, ("scsi%d : reselect\n", instance->host_no));
+	dprintk(NDEBUG_SELECTION, "scsi%d : reselect\n", instance->host_no);
 
 	/* 
 	 * At this point, we have detected that our SCSI ID is on the bus,
@@ -2597,7 +2588,7 @@ static void NCR5380_reselect(struct Scsi_Host *instance) {
 		do_abort(instance);
 	} else {
 		hostdata->connected = tmp;
-		dprintk(NDEBUG_RESELECTION, ("scsi%d : nexus established, target = %d, lun = %d, tag = %d\n", instance->host_no, tmp->target, tmp->lun, tmp->tag));
+		dprintk(NDEBUG_RESELECTION, "scsi%d : nexus established, target = %d, lun = %d, tag = %d\n", instance->host_no, tmp->device->id, tmp->device->lun, tmp->tag);
 	}
 }
 
@@ -2682,8 +2673,8 @@ static int NCR5380_abort(Scsi_Cmnd * cmd) {
 
 	NCR5380_setup(instance);
 
-	dprintk(NDEBUG_ABORT, ("scsi%d : abort called\n", instance->host_no));
-	dprintk(NDEBUG_ABORT, ("        basr 0x%X, sr 0x%X\n", NCR5380_read(BUS_AND_STATUS_REG), NCR5380_read(STATUS_REG)));
+	dprintk(NDEBUG_ABORT, "scsi%d : abort called\n", instance->host_no);
+	dprintk(NDEBUG_ABORT, "        basr 0x%X, sr 0x%X\n", NCR5380_read(BUS_AND_STATUS_REG), NCR5380_read(STATUS_REG));
 
 #if 0
 /*
@@ -2693,7 +2684,7 @@ static int NCR5380_abort(Scsi_Cmnd * cmd) {
  */
 
 	if (hostdata->connected == cmd) {
-		dprintk(NDEBUG_ABORT, ("scsi%d : aborting connected command\n", instance->host_no));
+		dprintk(NDEBUG_ABORT, "scsi%d : aborting connected command\n", instance->host_no);
 		hostdata->aborted = 1;
 /*
  * We should perform BSY checking, and make sure we haven't slipped
@@ -2721,14 +2712,14 @@ static int NCR5380_abort(Scsi_Cmnd * cmd) {
  *          from the issue queue.
  */
  
-	dprintk(NDEBUG_ABORT, ("scsi%d : abort going into loop.\n", instance->host_no));
+	dprintk(NDEBUG_ABORT, "scsi%d : abort going into loop.\n", instance->host_no);
 	for (prev = (Scsi_Cmnd **) & (hostdata->issue_queue), tmp = (Scsi_Cmnd *) hostdata->issue_queue; tmp; prev = (Scsi_Cmnd **) & (tmp->host_scribble), tmp = (Scsi_Cmnd *) tmp->host_scribble)
 		if (cmd == tmp) {
 			REMOVE(5, *prev, tmp, tmp->host_scribble);
 			(*prev) = (Scsi_Cmnd *) tmp->host_scribble;
 			tmp->host_scribble = NULL;
 			tmp->result = DID_ABORT << 16;
-			dprintk(NDEBUG_ABORT, ("scsi%d : abort removed command from issue queue.\n", instance->host_no));
+			dprintk(NDEBUG_ABORT, "scsi%d : abort removed command from issue queue.\n", instance->host_no);
 			tmp->scsi_done(tmp);
 			return SUCCESS;
 		}
@@ -2750,7 +2741,7 @@ static int NCR5380_abort(Scsi_Cmnd * cmd) {
  */
 
 	if (hostdata->connected) {
-		dprintk(NDEBUG_ABORT, ("scsi%d : abort failed, command connected.\n", instance->host_no));
+		dprintk(NDEBUG_ABORT, "scsi%d : abort failed, command connected.\n", instance->host_no);
 		return FAILED;
 	}
 /*
@@ -2780,11 +2771,11 @@ static int NCR5380_abort(Scsi_Cmnd * cmd) {
 
 	for (tmp = (Scsi_Cmnd *) hostdata->disconnected_queue; tmp; tmp = (Scsi_Cmnd *) tmp->host_scribble)
 		if (cmd == tmp) {
-			dprintk(NDEBUG_ABORT, ("scsi%d : aborting disconnected command.\n", instance->host_no));
+			dprintk(NDEBUG_ABORT, "scsi%d : aborting disconnected command.\n", instance->host_no);
 
 			if (NCR5380_select(instance, cmd, (int) cmd->tag))
 				return FAILED;
-			dprintk(NDEBUG_ABORT, ("scsi%d : nexus reestablished.\n", instance->host_no));
+			dprintk(NDEBUG_ABORT, "scsi%d : nexus reestablished.\n", instance->host_no);
 
 			do_abort(instance);
 
diff --git a/drivers/scsi/NCR5380.h b/drivers/scsi/NCR5380.h
index 14964d0a0e9d..c79ddfa6f53c 100644
--- a/drivers/scsi/NCR5380.h
+++ b/drivers/scsi/NCR5380.h
@@ -21,10 +21,6 @@
  * 1+ (800) 334-5454
  */
 
-/*
- * $Log: NCR5380.h,v $
- */
-
 #ifndef NCR5380_H
 #define NCR5380_H
 
@@ -60,6 +56,9 @@
 #define NDEBUG_C400_PREAD	0x100000
 #define NDEBUG_C400_PWRITE	0x200000
 #define NDEBUG_LISTS		0x400000
+#define NDEBUG_ABORT		0x800000
+#define NDEBUG_TAGS		0x1000000
+#define NDEBUG_MERGING		0x2000000
 
 #define NDEBUG_ANY		0xFFFFFFFFUL
 
@@ -292,9 +291,24 @@ struct NCR5380_hostdata {
 
 #ifdef __KERNEL__
 
-#define dprintk(a,b)			do {} while(0)
-#define NCR5380_dprint(a,b)		do {} while(0)
-#define NCR5380_dprint_phase(a,b)	do {} while(0)
+#ifndef NDEBUG
+#define NDEBUG (0)
+#endif
+
+#define dprintk(flg, fmt, ...) \
+	do { if ((NDEBUG) & (flg)) pr_debug(fmt, ## __VA_ARGS__); } while (0)
+
+#if NDEBUG
+#define NCR5380_dprint(flg, arg) \
+	do { if ((NDEBUG) & (flg)) NCR5380_print(arg); } while (0)
+#define NCR5380_dprint_phase(flg, arg) \
+	do { if ((NDEBUG) & (flg)) NCR5380_print_phase(arg); } while (0)
+static void NCR5380_print_phase(struct Scsi_Host *instance);
+static void NCR5380_print(struct Scsi_Host *instance);
+#else
+#define NCR5380_dprint(flg, arg)       do {} while (0)
+#define NCR5380_dprint_phase(flg, arg) do {} while (0)
+#endif
 
 #if defined(AUTOPROBE_IRQ)
 static int NCR5380_probe_irq(struct Scsi_Host *instance, int possible);
@@ -307,10 +321,6 @@ static irqreturn_t NCR5380_intr(int irq, void *dev_id);
 #endif
 static void NCR5380_main(struct work_struct *work);
 static void __maybe_unused NCR5380_print_options(struct Scsi_Host *instance);
-#ifdef NDEBUG
-static void NCR5380_print_phase(struct Scsi_Host *instance);
-static void NCR5380_print(struct Scsi_Host *instance);
-#endif
 static int NCR5380_abort(Scsi_Cmnd * cmd);
 static int NCR5380_bus_reset(Scsi_Cmnd * cmd);
 static int NCR5380_queue_command(struct Scsi_Host *, struct scsi_cmnd *);
diff --git a/drivers/scsi/aic7xxx/aic79xx_pci.c b/drivers/scsi/aic7xxx/aic79xx_pci.c
index 14b5f8d0e7f4..cc9bd26f5d1a 100644
--- a/drivers/scsi/aic7xxx/aic79xx_pci.c
+++ b/drivers/scsi/aic7xxx/aic79xx_pci.c
@@ -827,7 +827,7 @@ ahd_pci_intr(struct ahd_softc *ahd)
 		for (bit = 0; bit < 8; bit++) {
 
 			if ((pci_status[i] & (0x1 << bit)) != 0) {
-				static const char *s;
+				const char *s;
 
 				s = pci_status_strings[bit];
 				if (i == 7/*TARG*/ && bit == 3)
@@ -887,23 +887,15 @@ ahd_pci_split_intr(struct ahd_softc *ahd, u_int intstat)
 
 		for (bit = 0; bit < 8; bit++) {
 
-			if ((split_status[i] & (0x1 << bit)) != 0) {
-				static const char *s;
-
-				s = split_status_strings[bit];
-				printk(s, ahd_name(ahd),
+			if ((split_status[i] & (0x1 << bit)) != 0)
+				printk(split_status_strings[bit], ahd_name(ahd),
 				       split_status_source[i]);
-			}
 
 			if (i > 1)
 				continue;
 
-			if ((sg_split_status[i] & (0x1 << bit)) != 0) {
-				static const char *s;
-
-				s = split_status_strings[bit];
-				printk(s, ahd_name(ahd), "SG");
-			}
+			if ((sg_split_status[i] & (0x1 << bit)) != 0)
+				printk(split_status_strings[bit], ahd_name(ahd), "SG");
 		}
 	}
 	/*
diff --git a/drivers/scsi/arm/acornscsi.c b/drivers/scsi/arm/acornscsi.c
index 059ff477a398..2e797a367608 100644
--- a/drivers/scsi/arm/acornscsi.c
+++ b/drivers/scsi/arm/acornscsi.c
@@ -62,13 +62,6 @@
  */
 #undef CONFIG_SCSI_ACORNSCSI_TAGGED_QUEUE
 /*
- * SCSI-II Linked command support.
- *
- * The higher level code doesn't support linked commands yet, and so the option
- * is undef'd here.
- */
-#undef CONFIG_SCSI_ACORNSCSI_LINK
-/*
  * SCSI-II Synchronous transfer support.
  *
  * Tried and tested...
@@ -160,10 +153,6 @@
 #error "Yippee!  ABORT TAG is now defined!  Remove this error!"
 #endif
 
-#ifdef CONFIG_SCSI_ACORNSCSI_LINK
-#error SCSI2 LINKed commands not supported (yet)!
-#endif
-
 #ifdef USE_DMAC
 /*
  * DMAC setup parameters
@@ -1668,42 +1657,6 @@ void acornscsi_message(AS_Host *host)
 	}
 	break;
 
-#ifdef CONFIG_SCSI_ACORNSCSI_LINK
-    case LINKED_CMD_COMPLETE:
-    case LINKED_FLG_CMD_COMPLETE:
-	/*
-	 * We don't support linked commands yet
-	 */
-	if (0) {
-#if (DEBUG & DEBUG_LINK)
-	    printk("scsi%d.%c: lun %d tag %d linked command complete\n",
-		    host->host->host_no, acornscsi_target(host), host->SCpnt->tag);
-#endif
-	    /*
-	     * A linked command should only terminate with one of these messages
-	     * if there are more linked commands available.
-	     */
-	    if (!host->SCpnt->next_link) {
-		printk(KERN_WARNING "scsi%d.%c: lun %d tag %d linked command complete, but no next_link\n",
-			instance->host_no, acornscsi_target(host), host->SCpnt->tag);
-		acornscsi_sbic_issuecmd(host, CMND_ASSERTATN);
-		msgqueue_addmsg(&host->scsi.msgs, 1, ABORT);
-	    } else {
-		struct scsi_cmnd *SCpnt = host->SCpnt;
-
-		acornscsi_dma_cleanup(host);
-
-		host->SCpnt = host->SCpnt->next_link;
-		host->SCpnt->tag = SCpnt->tag;
-		SCpnt->result = DID_OK | host->scsi.SCp.Message << 8 | host->Scsi.SCp.Status;
-		SCpnt->done(SCpnt);
-
-		/* initialise host->SCpnt->SCp */
-	    }
-	    break;
-	}
-#endif
-
     default: /* reject message */
 	printk(KERN_ERR "scsi%d.%c: unrecognised message %02X, rejecting\n",
 		host->host->host_no, acornscsi_target(host),
@@ -2825,9 +2778,6 @@ char *acornscsi_info(struct Scsi_Host *host)
 #ifdef CONFIG_SCSI_ACORNSCSI_TAGGED_QUEUE
     " TAG"
 #endif
-#ifdef CONFIG_SCSI_ACORNSCSI_LINK
-    " LINK"
-#endif
 #if (DEBUG & DEBUG_NO_WRITE)
     " NOWRITE (" __stringify(NO_WRITE) ")"
 #endif
@@ -2851,9 +2801,6 @@ static int acornscsi_show_info(struct seq_file *m, struct Scsi_Host *instance)
 #ifdef CONFIG_SCSI_ACORNSCSI_TAGGED_QUEUE
     " TAG"
 #endif
-#ifdef CONFIG_SCSI_ACORNSCSI_LINK
-    " LINK"
-#endif
 #if (DEBUG & DEBUG_NO_WRITE)
     " NOWRITE (" __stringify(NO_WRITE) ")"
 #endif
diff --git a/drivers/scsi/arm/cumana_1.c b/drivers/scsi/arm/cumana_1.c
index f8e060900052..8ef810a4476e 100644
--- a/drivers/scsi/arm/cumana_1.c
+++ b/drivers/scsi/arm/cumana_1.c
@@ -36,9 +36,6 @@
 	void __iomem *base;		\
 	void __iomem *dma
 
-#define BOARD_NORMAL	0
-#define BOARD_NCR53C400	1
-
 #include "../NCR5380.h"
 
 void cumanascsi_setup(char *str, int *ints)
diff --git a/drivers/scsi/arm/oak.c b/drivers/scsi/arm/oak.c
index 4266eef8aca1..188e734c7ff0 100644
--- a/drivers/scsi/arm/oak.c
+++ b/drivers/scsi/arm/oak.c
@@ -37,9 +37,6 @@
 #define NCR5380_implementation_fields	\
 	void __iomem *base
 
-#define BOARD_NORMAL	0
-#define BOARD_NCR53C400	1
-
 #include "../NCR5380.h"
 
 #undef START_DMA_INITIATOR_RECEIVE_REG
diff --git a/drivers/scsi/atari_NCR5380.c b/drivers/scsi/atari_NCR5380.c
index 0f3cdbc80ba6..1814aa20b724 100644
--- a/drivers/scsi/atari_NCR5380.c
+++ b/drivers/scsi/atari_NCR5380.c
@@ -370,7 +370,7 @@ static int is_lun_busy(Scsi_Cmnd *cmd, int should_be_tagged)
 		return 0;
 	if (TagAlloc[cmd->device->id][cmd->device->lun].nr_allocated >=
 	    TagAlloc[cmd->device->id][cmd->device->lun].queue_size) {
-		TAG_PRINTK("scsi%d: target %d lun %d: no free tags\n",
+		dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d: no free tags\n",
 			   H_NO(cmd), cmd->device->id, cmd->device->lun);
 		return 1;
 	}
@@ -394,7 +394,7 @@ static void cmd_get_tag(Scsi_Cmnd *cmd, int should_be_tagged)
 	    !setup_use_tagged_queuing || !cmd->device->tagged_supported) {
 		cmd->tag = TAG_NONE;
 		hostdata->busy[cmd->device->id] |= (1 << cmd->device->lun);
-		TAG_PRINTK("scsi%d: target %d lun %d now allocated by untagged "
+		dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d now allocated by untagged "
 			   "command\n", H_NO(cmd), cmd->device->id, cmd->device->lun);
 	} else {
 		TAG_ALLOC *ta = &TagAlloc[cmd->device->id][cmd->device->lun];
@@ -402,7 +402,7 @@ static void cmd_get_tag(Scsi_Cmnd *cmd, int should_be_tagged)
 		cmd->tag = find_first_zero_bit(ta->allocated, MAX_TAGS);
 		set_bit(cmd->tag, ta->allocated);
 		ta->nr_allocated++;
-		TAG_PRINTK("scsi%d: using tag %d for target %d lun %d "
+		dprintk(NDEBUG_TAGS, "scsi%d: using tag %d for target %d lun %d "
 			   "(now %d tags in use)\n",
 			   H_NO(cmd), cmd->tag, cmd->device->id,
 			   cmd->device->lun, ta->nr_allocated);
@@ -420,7 +420,7 @@ static void cmd_free_tag(Scsi_Cmnd *cmd)
 
 	if (cmd->tag == TAG_NONE) {
 		hostdata->busy[cmd->device->id] &= ~(1 << cmd->device->lun);
-		TAG_PRINTK("scsi%d: target %d lun %d untagged cmd finished\n",
+		dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d untagged cmd finished\n",
 			   H_NO(cmd), cmd->device->id, cmd->device->lun);
 	} else if (cmd->tag >= MAX_TAGS) {
 		printk(KERN_NOTICE "scsi%d: trying to free bad tag %d!\n",
@@ -429,7 +429,7 @@ static void cmd_free_tag(Scsi_Cmnd *cmd)
 		TAG_ALLOC *ta = &TagAlloc[cmd->device->id][cmd->device->lun];
 		clear_bit(cmd->tag, ta->allocated);
 		ta->nr_allocated--;
-		TAG_PRINTK("scsi%d: freed tag %d for target %d lun %d\n",
+		dprintk(NDEBUG_TAGS, "scsi%d: freed tag %d for target %d lun %d\n",
 			   H_NO(cmd), cmd->tag, cmd->device->id, cmd->device->lun);
 	}
 }
@@ -478,7 +478,7 @@ static void merge_contiguous_buffers(Scsi_Cmnd *cmd)
 	for (endaddr = virt_to_phys(cmd->SCp.ptr + cmd->SCp.this_residual - 1) + 1;
 	     cmd->SCp.buffers_residual &&
 	     virt_to_phys(sg_virt(&cmd->SCp.buffer[1])) == endaddr;) {
-		MER_PRINTK("VTOP(%p) == %08lx -> merging\n",
+		dprintk(NDEBUG_MERGING, "VTOP(%p) == %08lx -> merging\n",
 			   page_address(sg_page(&cmd->SCp.buffer[1])), endaddr);
 #if (NDEBUG & NDEBUG_MERGING)
 		++cnt;
@@ -490,7 +490,7 @@ static void merge_contiguous_buffers(Scsi_Cmnd *cmd)
 	}
 #if (NDEBUG & NDEBUG_MERGING)
 	if (oldlen != cmd->SCp.this_residual)
-		MER_PRINTK("merged %d buffers from %p, new length %08x\n",
+		dprintk(NDEBUG_MERGING, "merged %d buffers from %p, new length %08x\n",
 			   cnt, cmd->SCp.ptr, cmd->SCp.this_residual);
 #endif
 }
@@ -626,16 +626,6 @@ static void NCR5380_print_phase(struct Scsi_Host *instance)
 	}
 }
 
-#else /* !NDEBUG */
-
-/* dummies... */
-static inline void NCR5380_print(struct Scsi_Host *instance)
-{
-};
-static inline void NCR5380_print_phase(struct Scsi_Host *instance)
-{
-};
-
 #endif
 
 /*
@@ -676,7 +666,7 @@ static inline void NCR5380_all_init(void)
 {
 	static int done = 0;
 	if (!done) {
-		INI_PRINTK("scsi : NCR5380_all_init()\n");
+		dprintk(NDEBUG_INIT, "scsi : NCR5380_all_init()\n");
 		done = 1;
 	}
 }
@@ -739,8 +729,8 @@ static void NCR5380_print_status(struct Scsi_Host *instance)
 	Scsi_Cmnd *ptr;
 	unsigned long flags;
 
-	NCR_PRINT(NDEBUG_ANY);
-	NCR_PRINT_PHASE(NDEBUG_ANY);
+	NCR5380_dprint(NDEBUG_ANY, instance);
+	NCR5380_dprint_phase(NDEBUG_ANY, instance);
 
 	hostdata = (struct NCR5380_hostdata *)instance->hostdata;
 
@@ -984,7 +974,7 @@ static int NCR5380_queue_command_lck(Scsi_Cmnd *cmd, void (*done)(Scsi_Cmnd *))
 	}
 	local_irq_restore(flags);
 
-	QU_PRINTK("scsi%d: command added to %s of queue\n", H_NO(cmd),
+	dprintk(NDEBUG_QUEUES, "scsi%d: command added to %s of queue\n", H_NO(cmd),
 		  (cmd->cmnd[0] == REQUEST_SENSE) ? "head" : "tail");
 
 	/* If queue_command() is called from an interrupt (real one or bottom
@@ -1054,7 +1044,7 @@ static void NCR5380_main(struct work_struct *work)
 		done = 1;
 
 		if (!hostdata->connected) {
-			MAIN_PRINTK("scsi%d: not connected\n", HOSTNO);
+			dprintk(NDEBUG_MAIN, "scsi%d: not connected\n", HOSTNO);
 			/*
 			 * Search through the issue_queue for a command destined
 			 * for a target that's not busy.
@@ -1107,7 +1097,7 @@ static void NCR5380_main(struct work_struct *work)
 					 * On failure, we must add the command back to the
 					 *   issue queue so we can keep trying.
 					 */
-					MAIN_PRINTK("scsi%d: main(): command for target %d "
+					dprintk(NDEBUG_MAIN, "scsi%d: main(): command for target %d "
 						    "lun %d removed from issue_queue\n",
 						    HOSTNO, tmp->device->id, tmp->device->lun);
 					/*
@@ -1140,7 +1130,7 @@ static void NCR5380_main(struct work_struct *work)
 #endif
 						falcon_dont_release--;
 						local_irq_restore(flags);
-						MAIN_PRINTK("scsi%d: main(): select() failed, "
+						dprintk(NDEBUG_MAIN, "scsi%d: main(): select() failed, "
 							    "returned to issue_queue\n", HOSTNO);
 						if (hostdata->connected)
 							break;
@@ -1155,10 +1145,10 @@ static void NCR5380_main(struct work_struct *work)
 #endif
 		    ) {
 			local_irq_restore(flags);
-			MAIN_PRINTK("scsi%d: main: performing information transfer\n",
+			dprintk(NDEBUG_MAIN, "scsi%d: main: performing information transfer\n",
 				    HOSTNO);
 			NCR5380_information_transfer(instance);
-			MAIN_PRINTK("scsi%d: main: done set false\n", HOSTNO);
+			dprintk(NDEBUG_MAIN, "scsi%d: main: done set false\n", HOSTNO);
 			done = 0;
 		}
 	} while (!done);
@@ -1204,12 +1194,12 @@ static void NCR5380_dma_complete(struct Scsi_Host *instance)
 			    (BASR_PHASE_MATCH|BASR_ACK)) {
 				saved_data = NCR5380_read(INPUT_DATA_REG);
 				overrun = 1;
-				DMA_PRINTK("scsi%d: read overrun handled\n", HOSTNO);
+				dprintk(NDEBUG_DMA, "scsi%d: read overrun handled\n", HOSTNO);
 			}
 		}
 	}
 
-	DMA_PRINTK("scsi%d: real DMA transfer complete, basr 0x%X, sr 0x%X\n",
+	dprintk(NDEBUG_DMA, "scsi%d: real DMA transfer complete, basr 0x%X, sr 0x%X\n",
 		   HOSTNO, NCR5380_read(BUS_AND_STATUS_REG),
 		   NCR5380_read(STATUS_REG));
 
@@ -1229,13 +1219,13 @@ static void NCR5380_dma_complete(struct Scsi_Host *instance)
 		if ((NCR5380_read(STATUS_REG) & PHASE_MASK) == p && (p & SR_IO)) {
 			cnt = toPIO = atari_read_overruns;
 			if (overrun) {
-				DMA_PRINTK("Got an input overrun, using saved byte\n");
+				dprintk(NDEBUG_DMA, "Got an input overrun, using saved byte\n");
 				*(*data)++ = saved_data;
 				(*count)--;
 				cnt--;
 				toPIO--;
 			}
-			DMA_PRINTK("Doing %d-byte PIO to 0x%08lx\n", cnt, (long)*data);
+			dprintk(NDEBUG_DMA, "Doing %d-byte PIO to 0x%08lx\n", cnt, (long)*data);
 			NCR5380_transfer_pio(instance, &p, &cnt, data);
 			*count -= toPIO - cnt;
 		}
@@ -1261,25 +1251,25 @@ static irqreturn_t NCR5380_intr(int irq, void *dev_id)
 	int done = 1, handled = 0;
 	unsigned char basr;
 
-	INT_PRINTK("scsi%d: NCR5380 irq triggered\n", HOSTNO);
+	dprintk(NDEBUG_INTR, "scsi%d: NCR5380 irq triggered\n", HOSTNO);
 
 	/* Look for pending interrupts */
 	basr = NCR5380_read(BUS_AND_STATUS_REG);
-	INT_PRINTK("scsi%d: BASR=%02x\n", HOSTNO, basr);
+	dprintk(NDEBUG_INTR, "scsi%d: BASR=%02x\n", HOSTNO, basr);
 	/* dispatch to appropriate routine if found and done=0 */
 	if (basr & BASR_IRQ) {
-		NCR_PRINT(NDEBUG_INTR);
+		NCR5380_dprint(NDEBUG_INTR, instance);
 		if ((NCR5380_read(STATUS_REG) & (SR_SEL|SR_IO)) == (SR_SEL|SR_IO)) {
 			done = 0;
 			ENABLE_IRQ();
-			INT_PRINTK("scsi%d: SEL interrupt\n", HOSTNO);
+			dprintk(NDEBUG_INTR, "scsi%d: SEL interrupt\n", HOSTNO);
 			NCR5380_reselect(instance);
 			(void)NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 		} else if (basr & BASR_PARITY_ERROR) {
-			INT_PRINTK("scsi%d: PARITY interrupt\n", HOSTNO);
+			dprintk(NDEBUG_INTR, "scsi%d: PARITY interrupt\n", HOSTNO);
 			(void)NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 		} else if ((NCR5380_read(STATUS_REG) & SR_RST) == SR_RST) {
-			INT_PRINTK("scsi%d: RESET interrupt\n", HOSTNO);
+			dprintk(NDEBUG_INTR, "scsi%d: RESET interrupt\n", HOSTNO);
 			(void)NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 		} else {
 			/*
@@ -1298,7 +1288,7 @@ static irqreturn_t NCR5380_intr(int irq, void *dev_id)
 			    ((basr & BASR_END_DMA_TRANSFER) ||
 			     !(basr & BASR_PHASE_MATCH))) {
 
-				INT_PRINTK("scsi%d: PHASE MISM or EOP interrupt\n", HOSTNO);
+				dprintk(NDEBUG_INTR, "scsi%d: PHASE MISM or EOP interrupt\n", HOSTNO);
 				NCR5380_dma_complete( instance );
 				done = 0;
 				ENABLE_IRQ();
@@ -1323,7 +1313,7 @@ static irqreturn_t NCR5380_intr(int irq, void *dev_id)
 	}
 
 	if (!done) {
-		INT_PRINTK("scsi%d: in int routine, calling main\n", HOSTNO);
+		dprintk(NDEBUG_INTR, "scsi%d: in int routine, calling main\n", HOSTNO);
 		/* Put a call to NCR5380_main() on the queue... */
 		queue_main();
 	}
@@ -1396,8 +1386,8 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 	unsigned long flags;
 
 	hostdata->restart_select = 0;
-	NCR_PRINT(NDEBUG_ARBITRATION);
-	ARB_PRINTK("scsi%d: starting arbitration, id = %d\n", HOSTNO,
+	NCR5380_dprint(NDEBUG_ARBITRATION, instance);
+	dprintk(NDEBUG_ARBITRATION, "scsi%d: starting arbitration, id = %d\n", HOSTNO,
 		   instance->this_id);
 
 	/*
@@ -1442,7 +1432,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 		;
 #endif
 
-	ARB_PRINTK("scsi%d: arbitration complete\n", HOSTNO);
+	dprintk(NDEBUG_ARBITRATION, "scsi%d: arbitration complete\n", HOSTNO);
 
 	if (hostdata->connected) {
 		NCR5380_write(MODE_REG, MR_BASE);
@@ -1463,7 +1453,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 	    (NCR5380_read(INITIATOR_COMMAND_REG) & ICR_ARBITRATION_LOST) ||
 	    hostdata->connected) {
 		NCR5380_write(MODE_REG, MR_BASE);
-		ARB_PRINTK("scsi%d: lost arbitration, deasserting MR_ARBITRATE\n",
+		dprintk(NDEBUG_ARBITRATION, "scsi%d: lost arbitration, deasserting MR_ARBITRATE\n",
 			   HOSTNO);
 		return -1;
 	}
@@ -1478,7 +1468,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 	    hostdata->connected) {
 		NCR5380_write(MODE_REG, MR_BASE);
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-		ARB_PRINTK("scsi%d: lost arbitration, deasserting ICR_ASSERT_SEL\n",
+		dprintk(NDEBUG_ARBITRATION, "scsi%d: lost arbitration, deasserting ICR_ASSERT_SEL\n",
 			   HOSTNO);
 		return -1;
 	}
@@ -1501,7 +1491,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 		return -1;
 	}
 
-	ARB_PRINTK("scsi%d: won arbitration\n", HOSTNO);
+	dprintk(NDEBUG_ARBITRATION, "scsi%d: won arbitration\n", HOSTNO);
 
 	/*
 	 * Now that we have won arbitration, start Selection process, asserting
@@ -1561,7 +1551,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 
 	udelay(1);
 
-	SEL_PRINTK("scsi%d: selecting target %d\n", HOSTNO, cmd->device->id);
+	dprintk(NDEBUG_SELECTION, "scsi%d: selecting target %d\n", HOSTNO, cmd->device->id);
 
 	/*
 	 * The SCSI specification calls for a 250 ms timeout for the actual
@@ -1617,7 +1607,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 			printk(KERN_ERR "scsi%d: weirdness\n", HOSTNO);
 			if (hostdata->restart_select)
 				printk(KERN_NOTICE "\trestart select\n");
-			NCR_PRINT(NDEBUG_ANY);
+			NCR5380_dprint(NDEBUG_ANY, instance);
 			NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 			return -1;
 		}
@@ -1630,7 +1620,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 #endif
 		cmd->scsi_done(cmd);
 		NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
-		SEL_PRINTK("scsi%d: target did not respond within 250ms\n", HOSTNO);
+		dprintk(NDEBUG_SELECTION, "scsi%d: target did not respond within 250ms\n", HOSTNO);
 		NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 		return 0;
 	}
@@ -1656,7 +1646,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 	while (!(NCR5380_read(STATUS_REG) & SR_REQ))
 		;
 
-	SEL_PRINTK("scsi%d: target %d selected, going into MESSAGE OUT phase.\n",
+	dprintk(NDEBUG_SELECTION, "scsi%d: target %d selected, going into MESSAGE OUT phase.\n",
 		   HOSTNO, cmd->device->id);
 	tmp[0] = IDENTIFY(1, cmd->device->lun);
 
@@ -1676,7 +1666,7 @@ static int NCR5380_select(struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag)
 	data = tmp;
 	phase = PHASE_MSGOUT;
 	NCR5380_transfer_pio(instance, &phase, &len, &data);
-	SEL_PRINTK("scsi%d: nexus established.\n", HOSTNO);
+	dprintk(NDEBUG_SELECTION, "scsi%d: nexus established.\n", HOSTNO);
 	/* XXX need to handle errors here */
 	hostdata->connected = cmd;
 #ifndef SUPPORT_TAGS
@@ -1737,12 +1727,12 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance,
 		while (!((tmp = NCR5380_read(STATUS_REG)) & SR_REQ))
 			;
 
-		HSH_PRINTK("scsi%d: REQ detected\n", HOSTNO);
+		dprintk(NDEBUG_HANDSHAKE, "scsi%d: REQ detected\n", HOSTNO);
 
 		/* Check for phase mismatch */
 		if ((tmp & PHASE_MASK) != p) {
-			PIO_PRINTK("scsi%d: phase mismatch\n", HOSTNO);
-			NCR_PRINT_PHASE(NDEBUG_PIO);
+			dprintk(NDEBUG_PIO, "scsi%d: phase mismatch\n", HOSTNO);
+			NCR5380_dprint_phase(NDEBUG_PIO, instance);
 			break;
 		}
 
@@ -1764,25 +1754,25 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance,
 		if (!(p & SR_IO)) {
 			if (!((p & SR_MSG) && c > 1)) {
 				NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | ICR_ASSERT_DATA);
-				NCR_PRINT(NDEBUG_PIO);
+				NCR5380_dprint(NDEBUG_PIO, instance);
 				NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE |
 					      ICR_ASSERT_DATA | ICR_ASSERT_ACK);
 			} else {
 				NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE |
 					      ICR_ASSERT_DATA | ICR_ASSERT_ATN);
-				NCR_PRINT(NDEBUG_PIO);
+				NCR5380_dprint(NDEBUG_PIO, instance);
 				NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE |
 					      ICR_ASSERT_DATA | ICR_ASSERT_ATN | ICR_ASSERT_ACK);
 			}
 		} else {
-			NCR_PRINT(NDEBUG_PIO);
+			NCR5380_dprint(NDEBUG_PIO, instance);
 			NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | ICR_ASSERT_ACK);
 		}
 
 		while (NCR5380_read(STATUS_REG) & SR_REQ)
 			;
 
-		HSH_PRINTK("scsi%d: req false, handshake complete\n", HOSTNO);
+		dprintk(NDEBUG_HANDSHAKE, "scsi%d: req false, handshake complete\n", HOSTNO);
 
 		/*
 		 * We have several special cases to consider during REQ/ACK handshaking :
@@ -1803,7 +1793,7 @@ static int NCR5380_transfer_pio(struct Scsi_Host *instance,
 		}
 	} while (--c);
 
-	PIO_PRINTK("scsi%d: residual %d\n", HOSTNO, c);
+	dprintk(NDEBUG_PIO, "scsi%d: residual %d\n", HOSTNO, c);
 
 	*count = c;
 	*data = d;
@@ -1917,7 +1907,7 @@ static int NCR5380_transfer_dma(struct Scsi_Host *instance,
 	if (atari_read_overruns && (p & SR_IO))
 		c -= atari_read_overruns;
 
-	DMA_PRINTK("scsi%d: initializing DMA for %s, %d bytes %s %p\n",
+	dprintk(NDEBUG_DMA, "scsi%d: initializing DMA for %s, %d bytes %s %p\n",
 		   HOSTNO, (p & SR_IO) ? "reading" : "writing",
 		   c, (p & SR_IO) ? "to" : "from", d);
 
@@ -1997,7 +1987,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 			phase = (tmp & PHASE_MASK);
 			if (phase != old_phase) {
 				old_phase = phase;
-				NCR_PRINT_PHASE(NDEBUG_INFORMATION);
+				NCR5380_dprint_phase(NDEBUG_INFORMATION, instance);
 			}
 
 			if (sink && (phase != PHASE_MSGOUT)) {
@@ -2039,7 +2029,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					 * they are at contiguous physical addresses.
 					 */
 					merge_contiguous_buffers(cmd);
-					INF_PRINTK("scsi%d: %d bytes and %d buffers left\n",
+					dprintk(NDEBUG_INFORMATION, "scsi%d: %d bytes and %d buffers left\n",
 						   HOSTNO, cmd->SCp.this_residual,
 						   cmd->SCp.buffers_residual);
 				}
@@ -2123,7 +2113,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					/* Accept message by clearing ACK */
 					NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 
-					LNK_PRINTK("scsi%d: target %d lun %d linked command "
+					dprintk(NDEBUG_LINKED, "scsi%d: target %d lun %d linked command "
 						   "complete.\n", HOSTNO, cmd->device->id, cmd->device->lun);
 
 					/* Enable reselect interrupts */
@@ -2148,7 +2138,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					 * and don't free it! */
 					cmd->next_link->tag = cmd->tag;
 					cmd->result = cmd->SCp.Status | (cmd->SCp.Message << 8);
-					LNK_PRINTK("scsi%d: target %d lun %d linked request "
+					dprintk(NDEBUG_LINKED, "scsi%d: target %d lun %d linked request "
 						   "done, calling scsi_done().\n",
 						   HOSTNO, cmd->device->id, cmd->device->lun);
 #ifdef NCR5380_STATS
@@ -2165,7 +2155,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					/* ++guenther: possible race with Falcon locking */
 					falcon_dont_release++;
 					hostdata->connected = NULL;
-					QU_PRINTK("scsi%d: command for target %d, lun %d "
+					dprintk(NDEBUG_QUEUES, "scsi%d: command for target %d, lun %d "
 						  "completed\n", HOSTNO, cmd->device->id, cmd->device->lun);
 #ifdef SUPPORT_TAGS
 					cmd_free_tag(cmd);
@@ -2179,7 +2169,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 						/* ++Andreas: the mid level code knows about
 						   QUEUE_FULL now. */
 						TAG_ALLOC *ta = &TagAlloc[cmd->device->id][cmd->device->lun];
-						TAG_PRINTK("scsi%d: target %d lun %d returned "
+						dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d returned "
 							   "QUEUE_FULL after %d commands\n",
 							   HOSTNO, cmd->device->id, cmd->device->lun,
 							   ta->nr_allocated);
@@ -2224,14 +2214,14 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					    (status_byte(cmd->SCp.Status) == CHECK_CONDITION)) {
 						scsi_eh_prep_cmnd(cmd, &hostdata->ses, NULL, 0, ~0);
 
-						ASEN_PRINTK("scsi%d: performing request sense\n", HOSTNO);
+						dprintk(NDEBUG_AUTOSENSE, "scsi%d: performing request sense\n", HOSTNO);
 
 						local_irq_save(flags);
 						LIST(cmd,hostdata->issue_queue);
 						SET_NEXT(cmd, hostdata->issue_queue);
 						hostdata->issue_queue = (Scsi_Cmnd *) cmd;
 						local_irq_restore(flags);
-						QU_PRINTK("scsi%d: REQUEST SENSE added to head of "
+						dprintk(NDEBUG_QUEUES, "scsi%d: REQUEST SENSE added to head of "
 							  "issue queue\n", H_NO(cmd));
 					} else
 #endif /* def AUTOSENSE */
@@ -2277,7 +2267,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 						cmd->device->tagged_supported = 0;
 						hostdata->busy[cmd->device->id] |= (1 << cmd->device->lun);
 						cmd->tag = TAG_NONE;
-						TAG_PRINTK("scsi%d: target %d lun %d rejected "
+						dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d rejected "
 							   "QUEUE_TAG message; tagged queuing "
 							   "disabled\n",
 							   HOSTNO, cmd->device->id, cmd->device->lun);
@@ -2294,7 +2284,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					hostdata->connected = NULL;
 					hostdata->disconnected_queue = cmd;
 					local_irq_restore(flags);
-					QU_PRINTK("scsi%d: command for target %d lun %d was "
+					dprintk(NDEBUG_QUEUES, "scsi%d: command for target %d lun %d was "
 						  "moved from connected to the "
 						  "disconnected_queue\n", HOSTNO,
 						  cmd->device->id, cmd->device->lun);
@@ -2344,13 +2334,13 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 					/* Accept first byte by clearing ACK */
 					NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 
-					EXT_PRINTK("scsi%d: receiving extended message\n", HOSTNO);
+					dprintk(NDEBUG_EXTENDED, "scsi%d: receiving extended message\n", HOSTNO);
 
 					len = 2;
 					data = extended_msg + 1;
 					phase = PHASE_MSGIN;
 					NCR5380_transfer_pio(instance, &phase, &len, &data);
-					EXT_PRINTK("scsi%d: length=%d, code=0x%02x\n", HOSTNO,
+					dprintk(NDEBUG_EXTENDED, "scsi%d: length=%d, code=0x%02x\n", HOSTNO,
 						   (int)extended_msg[1], (int)extended_msg[2]);
 
 					if (!len && extended_msg[1] <=
@@ -2362,7 +2352,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 						phase = PHASE_MSGIN;
 
 						NCR5380_transfer_pio(instance, &phase, &len, &data);
-						EXT_PRINTK("scsi%d: message received, residual %d\n",
+						dprintk(NDEBUG_EXTENDED, "scsi%d: message received, residual %d\n",
 							   HOSTNO, len);
 
 						switch (extended_msg[2]) {
@@ -2451,7 +2441,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance)
 				break;
 			default:
 				printk("scsi%d: unknown phase\n", HOSTNO);
-				NCR_PRINT(NDEBUG_ANY);
+				NCR5380_dprint(NDEBUG_ANY, instance);
 			} /* switch(phase) */
 		} /* if (tmp * SR_REQ) */
 	} /* while (1) */
@@ -2493,7 +2483,7 @@ static void NCR5380_reselect(struct Scsi_Host *instance)
 
 	target_mask = NCR5380_read(CURRENT_SCSI_DATA_REG) & ~(hostdata->id_mask);
 
-	RSL_PRINTK("scsi%d: reselect\n", HOSTNO);
+	dprintk(NDEBUG_RESELECTION, "scsi%d: reselect\n", HOSTNO);
 
 	/*
 	 * At this point, we have detected that our SCSI ID is on the bus,
@@ -2544,7 +2534,7 @@ static void NCR5380_reselect(struct Scsi_Host *instance)
 		if (!NCR5380_transfer_pio(instance, &phase, &len, &data) &&
 		    msg[1] == SIMPLE_QUEUE_TAG)
 			tag = msg[2];
-		TAG_PRINTK("scsi%d: target mask %02x, lun %d sent tag %d at "
+		dprintk(NDEBUG_TAGS, "scsi%d: target mask %02x, lun %d sent tag %d at "
 			   "reselection\n", HOSTNO, target_mask, lun, tag);
 	}
 #endif
@@ -2598,7 +2588,7 @@ static void NCR5380_reselect(struct Scsi_Host *instance)
 	NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 
 	hostdata->connected = tmp;
-	RSL_PRINTK("scsi%d: nexus established, target = %d, lun = %d, tag = %d\n",
+	dprintk(NDEBUG_RESELECTION, "scsi%d: nexus established, target = %d, lun = %d, tag = %d\n",
 		   HOSTNO, tmp->device->id, tmp->device->lun, tmp->tag);
 	falcon_dont_release--;
 }
@@ -2640,7 +2630,7 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 		printk(KERN_ERR "scsi%d: !!BINGO!! Falcon has no lock in NCR5380_abort\n",
 		       HOSTNO);
 
-	ABRT_PRINTK("scsi%d: abort called basr 0x%02x, sr 0x%02x\n", HOSTNO,
+	dprintk(NDEBUG_ABORT, "scsi%d: abort called basr 0x%02x, sr 0x%02x\n", HOSTNO,
 		    NCR5380_read(BUS_AND_STATUS_REG),
 		    NCR5380_read(STATUS_REG));
 
@@ -2653,7 +2643,7 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 
 	if (hostdata->connected == cmd) {
 
-		ABRT_PRINTK("scsi%d: aborting connected command\n", HOSTNO);
+		dprintk(NDEBUG_ABORT, "scsi%d: aborting connected command\n", HOSTNO);
 		/*
 		 * We should perform BSY checking, and make sure we haven't slipped
 		 * into BUS FREE.
@@ -2683,11 +2673,11 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 			local_irq_restore(flags);
 			cmd->scsi_done(cmd);
 			falcon_release_lock_if_possible(hostdata);
-			return SCSI_ABORT_SUCCESS;
+			return SUCCESS;
 		} else {
 /*			local_irq_restore(flags); */
 			printk("scsi%d: abort of connected command failed!\n", HOSTNO);
-			return SCSI_ABORT_ERROR;
+			return FAILED;
 		}
 	}
 #endif
@@ -2705,13 +2695,13 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 			SET_NEXT(tmp, NULL);
 			tmp->result = DID_ABORT << 16;
 			local_irq_restore(flags);
-			ABRT_PRINTK("scsi%d: abort removed command from issue queue.\n",
+			dprintk(NDEBUG_ABORT, "scsi%d: abort removed command from issue queue.\n",
 				    HOSTNO);
 			/* Tagged queuing note: no tag to free here, hasn't been assigned
 			 * yet... */
 			tmp->scsi_done(tmp);
 			falcon_release_lock_if_possible(hostdata);
-			return SCSI_ABORT_SUCCESS;
+			return SUCCESS;
 		}
 	}
 
@@ -2728,8 +2718,8 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 
 	if (hostdata->connected) {
 		local_irq_restore(flags);
-		ABRT_PRINTK("scsi%d: abort failed, command connected.\n", HOSTNO);
-		return SCSI_ABORT_SNOOZE;
+		dprintk(NDEBUG_ABORT, "scsi%d: abort failed, command connected.\n", HOSTNO);
+		return FAILED;
 	}
 
 	/*
@@ -2761,12 +2751,12 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 	     tmp = NEXT(tmp)) {
 		if (cmd == tmp) {
 			local_irq_restore(flags);
-			ABRT_PRINTK("scsi%d: aborting disconnected command.\n", HOSTNO);
+			dprintk(NDEBUG_ABORT, "scsi%d: aborting disconnected command.\n", HOSTNO);
 
 			if (NCR5380_select(instance, cmd, (int)cmd->tag))
-				return SCSI_ABORT_BUSY;
+				return FAILED;
 
-			ABRT_PRINTK("scsi%d: nexus reestablished.\n", HOSTNO);
+			dprintk(NDEBUG_ABORT, "scsi%d: nexus reestablished.\n", HOSTNO);
 
 			do_abort(instance);
 
@@ -2791,7 +2781,7 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 					local_irq_restore(flags);
 					tmp->scsi_done(tmp);
 					falcon_release_lock_if_possible(hostdata);
-					return SCSI_ABORT_SUCCESS;
+					return SUCCESS;
 				}
 			}
 		}
@@ -2816,7 +2806,7 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
 	 */
 	falcon_release_lock_if_possible(hostdata);
 
-	return SCSI_ABORT_NOT_RUNNING;
+	return FAILED;
 }
 
 
@@ -2825,7 +2815,7 @@ int NCR5380_abort(Scsi_Cmnd *cmd)
  *
  * Purpose : reset the SCSI bus.
  *
- * Returns : SCSI_RESET_WAKEUP
+ * Returns : SUCCESS or FAILURE
  *
  */
 
@@ -2834,7 +2824,7 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 	SETUP_HOSTDATA(cmd->device->host);
 	int i;
 	unsigned long flags;
-#if 1
+#if defined(RESET_RUN_DONE)
 	Scsi_Cmnd *connected, *disconnected_queue;
 #endif
 
@@ -2859,7 +2849,14 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 	 * through anymore ... */
 	(void)NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 
-#if 1	/* XXX Should now be done by midlevel code, but it's broken XXX */
+	/* MSch 20140115 - looking at the generic NCR5380 driver, all of this
+	 * should go.
+	 * Catch-22: if we don't clear all queues, the SCSI driver lock will
+	 * not be reset by atari_scsi_reset()!
+	 */
+
+#if defined(RESET_RUN_DONE)
+	/* XXX Should now be done by midlevel code, but it's broken XXX */
 	/* XXX see below                                            XXX */
 
 	/* MSch: old-style reset: actually abort all command processing here */
@@ -2890,7 +2887,7 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 	 */
 
 	if ((cmd = connected)) {
-		ABRT_PRINTK("scsi%d: reset aborted a connected command\n", H_NO(cmd));
+		dprintk(NDEBUG_ABORT, "scsi%d: reset aborted a connected command\n", H_NO(cmd));
 		cmd->result = (cmd->result & 0xffff) | (DID_RESET << 16);
 		cmd->scsi_done(cmd);
 	}
@@ -2902,7 +2899,7 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 		cmd->scsi_done(cmd);
 	}
 	if (i > 0)
-		ABRT_PRINTK("scsi: reset aborted %d disconnected command(s)\n", i);
+		dprintk(NDEBUG_ABORT, "scsi: reset aborted %d disconnected command(s)\n", i);
 
 	/* The Falcon lock should be released after a reset...
 	 */
@@ -2915,7 +2912,7 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 	 * the midlevel code that the reset was SUCCESSFUL, and there is no
 	 * need to 'wake up' the commands by a request_sense
 	 */
-	return SCSI_RESET_SUCCESS | SCSI_RESET_BUS_RESET;
+	return SUCCESS;
 #else /* 1 */
 
 	/* MSch: new-style reset handling: let the mid-level do what it can */
@@ -2942,11 +2939,11 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 	 */
 
 	if (hostdata->issue_queue)
-		ABRT_PRINTK("scsi%d: reset aborted issued command(s)\n", H_NO(cmd));
+		dprintk(NDEBUG_ABORT, "scsi%d: reset aborted issued command(s)\n", H_NO(cmd));
 	if (hostdata->connected)
-		ABRT_PRINTK("scsi%d: reset aborted a connected command\n", H_NO(cmd));
+		dprintk(NDEBUG_ABORT, "scsi%d: reset aborted a connected command\n", H_NO(cmd));
 	if (hostdata->disconnected_queue)
-		ABRT_PRINTK("scsi%d: reset aborted disconnected command(s)\n", H_NO(cmd));
+		dprintk(NDEBUG_ABORT, "scsi%d: reset aborted disconnected command(s)\n", H_NO(cmd));
 
 	local_irq_save(flags);
 	hostdata->issue_queue = NULL;
@@ -2963,6 +2960,6 @@ static int NCR5380_bus_reset(Scsi_Cmnd *cmd)
 	local_irq_restore(flags);
 
 	/* we did no complete reset of all commands, so a wakeup is required */
-	return SCSI_RESET_WAKEUP | SCSI_RESET_BUS_RESET;
+	return SUCCESS;
 #endif /* 1 */
 }
diff --git a/drivers/scsi/atari_scsi.c b/drivers/scsi/atari_scsi.c
index a8d721ff19eb..b522134528d6 100644
--- a/drivers/scsi/atari_scsi.c
+++ b/drivers/scsi/atari_scsi.c
@@ -67,12 +67,6 @@
 
 #include <linux/module.h>
 
-#define NDEBUG (0)
-
-#define NDEBUG_ABORT		0x00100000
-#define NDEBUG_TAGS		0x00200000
-#define NDEBUG_MERGING		0x00400000
-
 #define AUTOSENSE
 /* For the Atari version, use only polled IO or REAL_DMA */
 #define	REAL_DMA
@@ -314,7 +308,7 @@ static irqreturn_t scsi_tt_intr(int irq, void *dummy)
 
 	dma_stat = tt_scsi_dma.dma_ctrl;
 
-	INT_PRINTK("scsi%d: NCR5380 interrupt, DMA status = %02x\n",
+	dprintk(NDEBUG_INTR, "scsi%d: NCR5380 interrupt, DMA status = %02x\n",
 		   atari_scsi_host->host_no, dma_stat & 0xff);
 
 	/* Look if it was the DMA that has interrupted: First possibility
@@ -340,7 +334,7 @@ static irqreturn_t scsi_tt_intr(int irq, void *dummy)
 	if ((dma_stat & 0x02) && !(dma_stat & 0x40)) {
 		atari_dma_residual = HOSTDATA_DMALEN - (SCSI_DMA_READ_P(dma_addr) - atari_dma_startaddr);
 
-		DMA_PRINTK("SCSI DMA: There are %ld residual bytes.\n",
+		dprintk(NDEBUG_DMA, "SCSI DMA: There are %ld residual bytes.\n",
 			   atari_dma_residual);
 
 		if ((signed int)atari_dma_residual < 0)
@@ -371,7 +365,7 @@ static irqreturn_t scsi_tt_intr(int irq, void *dummy)
 			 * other command.  These shouldn't disconnect anyway.
 			 */
 			if (atari_dma_residual & 0x1ff) {
-				DMA_PRINTK("SCSI DMA: DMA bug corrected, "
+				dprintk(NDEBUG_DMA, "SCSI DMA: DMA bug corrected, "
 					   "difference %ld bytes\n",
 					   512 - (atari_dma_residual & 0x1ff));
 				atari_dma_residual = (atari_dma_residual + 511) & ~0x1ff;
@@ -438,7 +432,7 @@ static irqreturn_t scsi_falcon_intr(int irq, void *dummy)
 			       "ST-DMA fifo\n", transferred & 15);
 
 		atari_dma_residual = HOSTDATA_DMALEN - transferred;
-		DMA_PRINTK("SCSI DMA: There are %ld residual bytes.\n",
+		dprintk(NDEBUG_DMA, "SCSI DMA: There are %ld residual bytes.\n",
 			   atari_dma_residual);
 	} else
 		atari_dma_residual = 0;
@@ -474,11 +468,11 @@ static void atari_scsi_fetch_restbytes(void)
 		/* there are 'nr' bytes left for the last long address
 		   before the DMA pointer */
 		phys_dst ^= nr;
-		DMA_PRINTK("SCSI DMA: there are %d rest bytes for phys addr 0x%08lx",
+		dprintk(NDEBUG_DMA, "SCSI DMA: there are %d rest bytes for phys addr 0x%08lx",
 			   nr, phys_dst);
 		/* The content of the DMA pointer is a physical address!  */
 		dst = phys_to_virt(phys_dst);
-		DMA_PRINTK(" = virt addr %p\n", dst);
+		dprintk(NDEBUG_DMA, " = virt addr %p\n", dst);
 		for (src = (char *)&tt_scsi_dma.dma_restdata; nr != 0; --nr)
 			*dst++ = *src++;
 	}
@@ -827,7 +821,7 @@ static int atari_scsi_bus_reset(Scsi_Cmnd *cmd)
 	} else {
 		atari_turnon_irq(IRQ_MFP_FSCSI);
 	}
-	if ((rv & SCSI_RESET_ACTION) == SCSI_RESET_SUCCESS)
+	if (rv == SUCCESS)
 		falcon_release_lock_if_possible(hostdata);
 
 	return rv;
@@ -883,7 +877,7 @@ static unsigned long atari_scsi_dma_setup(struct Scsi_Host *instance,
 {
 	unsigned long addr = virt_to_phys(data);
 
-	DMA_PRINTK("scsi%d: setting up dma, data = %p, phys = %lx, count = %ld, "
+	dprintk(NDEBUG_DMA, "scsi%d: setting up dma, data = %p, phys = %lx, count = %ld, "
 		   "dir = %d\n", instance->host_no, data, addr, count, dir);
 
 	if (!IS_A_TT() && !STRAM_ADDR(addr)) {
@@ -1063,7 +1057,7 @@ static unsigned long atari_dma_xfer_len(unsigned long wanted_len,
 		possible_len = limit;
 
 	if (possible_len != wanted_len)
-		DMA_PRINTK("Sorry, must cut DMA transfer size to %ld bytes "
+		dprintk(NDEBUG_DMA, "Sorry, must cut DMA transfer size to %ld bytes "
 			   "instead of %ld\n", possible_len, wanted_len);
 
 	return possible_len;
diff --git a/drivers/scsi/atari_scsi.h b/drivers/scsi/atari_scsi.h
index 11c624bb122d..3299d91d7336 100644
--- a/drivers/scsi/atari_scsi.h
+++ b/drivers/scsi/atari_scsi.h
@@ -54,125 +54,6 @@
 #define	NCR5380_dma_xfer_len(i,cmd,phase) \
 	atari_dma_xfer_len(cmd->SCp.this_residual,cmd,((phase) & SR_IO) ? 0 : 1)
 
-/* former generic SCSI error handling stuff */
-
-#define SCSI_ABORT_SNOOZE 0
-#define SCSI_ABORT_SUCCESS 1
-#define SCSI_ABORT_PENDING 2
-#define SCSI_ABORT_BUSY 3
-#define SCSI_ABORT_NOT_RUNNING 4
-#define SCSI_ABORT_ERROR 5
-
-#define SCSI_RESET_SNOOZE 0
-#define SCSI_RESET_PUNT 1
-#define SCSI_RESET_SUCCESS 2
-#define SCSI_RESET_PENDING 3
-#define SCSI_RESET_WAKEUP 4
-#define SCSI_RESET_NOT_RUNNING 5
-#define SCSI_RESET_ERROR 6
-
-#define SCSI_RESET_SYNCHRONOUS		0x01
-#define SCSI_RESET_ASYNCHRONOUS		0x02
-#define SCSI_RESET_SUGGEST_BUS_RESET	0x04
-#define SCSI_RESET_SUGGEST_HOST_RESET	0x08
-
-#define SCSI_RESET_BUS_RESET 0x100
-#define SCSI_RESET_HOST_RESET 0x200
-#define SCSI_RESET_ACTION   0xff
-
-/* Debugging printk definitions:
- *
- *  ARB  -> arbitration
- *  ASEN -> auto-sense
- *  DMA  -> DMA
- *  HSH  -> PIO handshake
- *  INF  -> information transfer
- *  INI  -> initialization
- *  INT  -> interrupt
- *  LNK  -> linked commands
- *  MAIN -> NCR5380_main() control flow
- *  NDAT -> no data-out phase
- *  NWR  -> no write commands
- *  PIO  -> PIO transfers
- *  PDMA -> pseudo DMA (unused on Atari)
- *  QU   -> queues
- *  RSL  -> reselections
- *  SEL  -> selections
- *  USL  -> usleep cpde (unused on Atari)
- *  LBS  -> last byte sent (unused on Atari)
- *  RSS  -> restarting of selections
- *  EXT  -> extended messages
- *  ABRT -> aborting and resetting
- *  TAG  -> queue tag handling
- *  MER  -> merging of consec. buffers
- *
- */
-
-#define dprint(flg, format...)			\
-({						\
-	if (NDEBUG & (flg))			\
-		printk(KERN_DEBUG format);	\
-})
-
-#define ARB_PRINTK(format, args...) \
-	dprint(NDEBUG_ARBITRATION, format , ## args)
-#define ASEN_PRINTK(format, args...) \
-	dprint(NDEBUG_AUTOSENSE, format , ## args)
-#define DMA_PRINTK(format, args...) \
-	dprint(NDEBUG_DMA, format , ## args)
-#define HSH_PRINTK(format, args...) \
-	dprint(NDEBUG_HANDSHAKE, format , ## args)
-#define INF_PRINTK(format, args...) \
-	dprint(NDEBUG_INFORMATION, format , ## args)
-#define INI_PRINTK(format, args...) \
-	dprint(NDEBUG_INIT, format , ## args)
-#define INT_PRINTK(format, args...) \
-	dprint(NDEBUG_INTR, format , ## args)
-#define LNK_PRINTK(format, args...) \
-	dprint(NDEBUG_LINKED, format , ## args)
-#define MAIN_PRINTK(format, args...) \
-	dprint(NDEBUG_MAIN, format , ## args)
-#define NDAT_PRINTK(format, args...) \
-	dprint(NDEBUG_NO_DATAOUT, format , ## args)
-#define NWR_PRINTK(format, args...) \
-	dprint(NDEBUG_NO_WRITE, format , ## args)
-#define PIO_PRINTK(format, args...) \
-	dprint(NDEBUG_PIO, format , ## args)
-#define PDMA_PRINTK(format, args...) \
-	dprint(NDEBUG_PSEUDO_DMA, format , ## args)
-#define QU_PRINTK(format, args...) \
-	dprint(NDEBUG_QUEUES, format , ## args)
-#define RSL_PRINTK(format, args...) \
-	dprint(NDEBUG_RESELECTION, format , ## args)
-#define SEL_PRINTK(format, args...) \
-	dprint(NDEBUG_SELECTION, format , ## args)
-#define USL_PRINTK(format, args...) \
-	dprint(NDEBUG_USLEEP, format , ## args)
-#define LBS_PRINTK(format, args...) \
-	dprint(NDEBUG_LAST_BYTE_SENT, format , ## args)
-#define RSS_PRINTK(format, args...) \
-	dprint(NDEBUG_RESTART_SELECT, format , ## args)
-#define EXT_PRINTK(format, args...) \
-	dprint(NDEBUG_EXTENDED, format , ## args)
-#define ABRT_PRINTK(format, args...) \
-	dprint(NDEBUG_ABORT, format , ## args)
-#define TAG_PRINTK(format, args...) \
-	dprint(NDEBUG_TAGS, format , ## args)
-#define MER_PRINTK(format, args...) \
-	dprint(NDEBUG_MERGING, format , ## args)
-
-/* conditional macros for NCR5380_print_{,phase,status} */
-
-#define NCR_PRINT(mask)	\
-	((NDEBUG & (mask)) ? NCR5380_print(instance) : (void)0)
-
-#define NCR_PRINT_PHASE(mask) \
-	((NDEBUG & (mask)) ? NCR5380_print_phase(instance) : (void)0)
-
-#define NCR_PRINT_STATUS(mask) \
-	((NDEBUG & (mask)) ? NCR5380_print_status(instance) : (void)0)
-
-
 #endif /* ndef ASM */
 #endif /* ATARI_SCSI_H */
 
diff --git a/drivers/scsi/be2iscsi/be.h b/drivers/scsi/be2iscsi/be.h
index 1bfb0bd01198..860f527d8f26 100644
--- a/drivers/scsi/be2iscsi/be.h
+++ b/drivers/scsi/be2iscsi/be.h
@@ -83,9 +83,20 @@ static inline void queue_tail_inc(struct be_queue_info *q)
 
 /*ISCSI */
 
+struct be_aic_obj {		/* Adaptive interrupt coalescing (AIC) info */
+	bool enable;
+	u32 min_eqd;		/* in usecs */
+	u32 max_eqd;		/* in usecs */
+	u32 prev_eqd;		/* in usecs */
+	u32 et_eqd;		/* configured val when aic is off */
+	ulong jiffs;
+	u64 eq_prev;		/* Used to calculate eqe */
+};
+
 struct be_eq_obj {
 	bool todo_mcc_cq;
 	bool todo_cq;
+	u32 cq_count;
 	struct be_queue_info q;
 	struct beiscsi_hba *phba;
 	struct be_queue_info *cq;
diff --git a/drivers/scsi/be2iscsi/be_cmds.h b/drivers/scsi/be2iscsi/be_cmds.h
index 7cf7f99ee442..cc7405c0eca0 100644
--- a/drivers/scsi/be2iscsi/be_cmds.h
+++ b/drivers/scsi/be2iscsi/be_cmds.h
@@ -71,6 +71,7 @@ struct be_mcc_wrb {
 #define BEISCSI_FW_MBX_TIMEOUT	100
 
 /* MBOX Command VER */
+#define MBX_CMD_VER1	0x01
 #define MBX_CMD_VER2	0x02
 
 struct be_mcc_compl {
@@ -271,6 +272,12 @@ struct be_cmd_resp_eq_create {
 	u16 rsvd0;		/* sword */
 } __packed;
 
+struct be_set_eqd {
+	u32 eq_id;
+	u32 phase;
+	u32 delay_multiplier;
+} __packed;
+
 struct mgmt_chap_format {
 	u32 flags;
 	u8  intr_chap_name[256];
@@ -622,7 +629,7 @@ struct be_cmd_req_modify_eq_delay {
 		u32 eq_id;
 		u32 phase;
 		u32 delay_multiplier;
-	} delay[8];
+	} delay[MAX_CPUS];
 } __packed;
 
 /******************** Get MAC ADDR *******************/
@@ -708,6 +715,8 @@ unsigned int be_cmd_get_port_speed(struct beiscsi_hba *phba);
 
 void free_mcc_tag(struct be_ctrl_info *ctrl, unsigned int tag);
 
+int be_cmd_modify_eq_delay(struct beiscsi_hba *phba, struct be_set_eqd *,
+			    int num);
 int beiscsi_mccq_compl(struct beiscsi_hba *phba,
 			uint32_t tag, struct be_mcc_wrb **wrb,
 			struct be_dma_mem *mbx_cmd_mem);
@@ -1005,6 +1014,26 @@ struct tcp_connect_and_offload_in {
 	u8 rsvd0[3];
 } __packed;
 
+struct tcp_connect_and_offload_in_v1 {
+	struct be_cmd_req_hdr hdr;
+	struct ip_addr_format ip_address;
+	u16 tcp_port;
+	u16 cid;
+	u16 cq_id;
+	u16 defq_id;
+	struct phys_addr dataout_template_pa;
+	u16 hdr_ring_id;
+	u16 data_ring_id;
+	u8 do_offload;
+	u8 ifd_state;
+	u8 rsvd0[2];
+	u16 tcp_window_size;
+	u8 tcp_window_scale_count;
+	u8 rsvd1;
+	u32 tcp_mss:24;
+	u8 rsvd2;
+} __packed;
+
 struct tcp_connect_and_offload_out {
 	struct be_cmd_resp_hdr hdr;
 	u32 connection_handle;
diff --git a/drivers/scsi/be2iscsi/be_iscsi.c b/drivers/scsi/be2iscsi/be_iscsi.c
index a3df43324c98..fd284ff36ecf 100644
--- a/drivers/scsi/be2iscsi/be_iscsi.c
+++ b/drivers/scsi/be2iscsi/be_iscsi.c
@@ -1106,7 +1106,7 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep,
 	struct beiscsi_hba *phba = beiscsi_ep->phba;
 	struct tcp_connect_and_offload_out *ptcpcnct_out;
 	struct be_dma_mem nonemb_cmd;
-	unsigned int tag;
+	unsigned int tag, req_memsize;
 	int ret = -ENOMEM;
 
 	beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG,
@@ -1127,8 +1127,14 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep,
 		       (beiscsi_ep->ep_cid)] = ep;
 
 	beiscsi_ep->cid_vld = 0;
+
+	if (is_chip_be2_be3r(phba))
+		req_memsize = sizeof(struct tcp_connect_and_offload_in);
+	else
+		req_memsize = sizeof(struct tcp_connect_and_offload_in_v1);
+
 	nonemb_cmd.va = pci_alloc_consistent(phba->ctrl.pdev,
-				sizeof(struct tcp_connect_and_offload_in),
+				req_memsize,
 				&nonemb_cmd.dma);
 	if (nonemb_cmd.va == NULL) {
 
@@ -1139,7 +1145,7 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep,
 		beiscsi_free_ep(beiscsi_ep);
 		return -ENOMEM;
 	}
-	nonemb_cmd.size = sizeof(struct tcp_connect_and_offload_in);
+	nonemb_cmd.size = req_memsize;
 	memset(nonemb_cmd.va, 0, nonemb_cmd.size);
 	tag = mgmt_open_connection(phba, dst_addr, beiscsi_ep, &nonemb_cmd);
 	if (tag <= 0) {
diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c
index 0d822297aa80..56467df3d6de 100644
--- a/drivers/scsi/be2iscsi/be_main.c
+++ b/drivers/scsi/be2iscsi/be_main.c
@@ -599,15 +599,7 @@ static struct beiscsi_hba *beiscsi_hba_alloc(struct pci_dev *pcidev)
 	pci_set_drvdata(pcidev, phba);
 	phba->interface_handle = 0xFFFFFFFF;
 
-	if (iscsi_host_add(shost, &phba->pcidev->dev))
-		goto free_devices;
-
 	return phba;
-
-free_devices:
-	pci_dev_put(phba->pcidev);
-	iscsi_host_free(phba->shost);
-	return NULL;
 }
 
 static void beiscsi_unmap_pci_function(struct beiscsi_hba *phba)
@@ -2279,6 +2271,7 @@ static int be_iopoll(struct blk_iopoll *iop, int budget)
 
 	pbe_eq = container_of(iop, struct be_eq_obj, iopoll);
 	ret = beiscsi_process_cq(pbe_eq);
+	pbe_eq->cq_count += ret;
 	if (ret < budget) {
 		phba = pbe_eq->phba;
 		blk_iopoll_complete(iop);
@@ -3692,7 +3685,7 @@ static void hwi_cleanup(struct beiscsi_hba *phba)
 	struct hwi_controller *phwi_ctrlr;
 	struct hwi_context_memory *phwi_context;
 	struct hwi_async_pdu_context *pasync_ctx;
-	int i, eq_num, ulp_num;
+	int i, eq_for_mcc, ulp_num;
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	phwi_context = phwi_ctrlr->phwi_ctxt;
@@ -3729,16 +3722,17 @@ static void hwi_cleanup(struct beiscsi_hba *phba)
 		if (q->created)
 			beiscsi_cmd_q_destroy(ctrl, q, QTYPE_CQ);
 	}
+
+	be_mcc_queues_destroy(phba);
 	if (phba->msix_enabled)
-		eq_num = 1;
+		eq_for_mcc = 1;
 	else
-		eq_num = 0;
-	for (i = 0; i < (phba->num_cpus + eq_num); i++) {
+		eq_for_mcc = 0;
+	for (i = 0; i < (phba->num_cpus + eq_for_mcc); i++) {
 		q = &phwi_context->be_eq[i].q;
 		if (q->created)
 			beiscsi_cmd_q_destroy(ctrl, q, QTYPE_EQ);
 	}
-	be_mcc_queues_destroy(phba);
 	be_cmd_fw_uninit(ctrl);
 }
 
@@ -3833,9 +3827,9 @@ static int hwi_init_port(struct beiscsi_hba *phba)
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	phwi_context = phwi_ctrlr->phwi_ctxt;
-	phwi_context->max_eqd = 0;
+	phwi_context->max_eqd = 128;
 	phwi_context->min_eqd = 0;
-	phwi_context->cur_eqd = 64;
+	phwi_context->cur_eqd = 0;
 	be_cmd_fw_initialize(&phba->ctrl);
 
 	status = beiscsi_create_eqs(phba, phwi_context);
@@ -4204,6 +4198,8 @@ static int hba_setup_cid_tbls(struct beiscsi_hba *phba)
 		kfree(phba->ep_array);
 		phba->ep_array = NULL;
 		ret = -ENOMEM;
+
+		goto free_memory;
 	}
 
 	for (i = 0; i < phba->params.cxns_per_ctrl; i++) {
@@ -5290,6 +5286,57 @@ static void beiscsi_msix_enable(struct beiscsi_hba *phba)
 	return;
 }
 
+static void be_eqd_update(struct beiscsi_hba *phba)
+{
+	struct be_set_eqd set_eqd[MAX_CPUS];
+	struct be_aic_obj *aic;
+	struct be_eq_obj *pbe_eq;
+	struct hwi_controller *phwi_ctrlr;
+	struct hwi_context_memory *phwi_context;
+	int eqd, i, num = 0;
+	ulong now;
+	u32 pps, delta;
+	unsigned int tag;
+
+	phwi_ctrlr = phba->phwi_ctrlr;
+	phwi_context = phwi_ctrlr->phwi_ctxt;
+
+	for (i = 0; i <= phba->num_cpus; i++) {
+		aic = &phba->aic_obj[i];
+		pbe_eq = &phwi_context->be_eq[i];
+		now = jiffies;
+		if (!aic->jiffs || time_before(now, aic->jiffs) ||
+		    pbe_eq->cq_count < aic->eq_prev) {
+			aic->jiffs = now;
+			aic->eq_prev = pbe_eq->cq_count;
+			continue;
+		}
+		delta = jiffies_to_msecs(now - aic->jiffs);
+		pps = (((u32)(pbe_eq->cq_count - aic->eq_prev) * 1000) / delta);
+		eqd = (pps / 1500) << 2;
+
+		if (eqd < 8)
+			eqd = 0;
+		eqd = min_t(u32, eqd, phwi_context->max_eqd);
+		eqd = max_t(u32, eqd, phwi_context->min_eqd);
+
+		aic->jiffs = now;
+		aic->eq_prev = pbe_eq->cq_count;
+
+		if (eqd != aic->prev_eqd) {
+			set_eqd[num].delay_multiplier = (eqd * 65)/100;
+			set_eqd[num].eq_id = pbe_eq->q.id;
+			aic->prev_eqd = eqd;
+			num++;
+		}
+	}
+	if (num) {
+		tag = be_cmd_modify_eq_delay(phba, set_eqd, num);
+		if (tag)
+			beiscsi_mccq_compl(phba, tag, NULL, NULL);
+	}
+}
+
 /*
  * beiscsi_hw_health_check()- Check adapter health
  * @work: work item to check HW health
@@ -5303,6 +5350,8 @@ beiscsi_hw_health_check(struct work_struct *work)
 		container_of(work, struct beiscsi_hba,
 			     beiscsi_hw_check_task.work);
 
+	be_eqd_update(phba);
+
 	beiscsi_ue_detect(phba);
 
 	schedule_delayed_work(&phba->beiscsi_hw_check_task,
@@ -5579,7 +5628,7 @@ static int beiscsi_dev_probe(struct pci_dev *pcidev,
 		phba->ctrl.mcc_numtag[i + 1] = 0;
 		phba->ctrl.mcc_tag_available++;
 		memset(&phba->ctrl.ptag_state[i].tag_mem_state, 0,
-		       sizeof(struct beiscsi_mcc_tag_state));
+		       sizeof(struct be_dma_mem));
 	}
 
 	phba->ctrl.mcc_alloc_index = phba->ctrl.mcc_free_index = 0;
@@ -5621,6 +5670,9 @@ static int beiscsi_dev_probe(struct pci_dev *pcidev,
 	}
 	hwi_enable_intr(phba);
 
+	if (iscsi_host_add(phba->shost, &phba->pcidev->dev))
+		goto free_blkenbld;
+
 	if (beiscsi_setup_boot_info(phba))
 		/*
 		 * log error but continue, because we may not be using
diff --git a/drivers/scsi/be2iscsi/be_main.h b/drivers/scsi/be2iscsi/be_main.h
index 9380b55bdeaf..9ceab426eec9 100644
--- a/drivers/scsi/be2iscsi/be_main.h
+++ b/drivers/scsi/be2iscsi/be_main.h
@@ -36,7 +36,7 @@
 #include <scsi/scsi_transport_iscsi.h>
 
 #define DRV_NAME		"be2iscsi"
-#define BUILD_STR		"10.2.125.0"
+#define BUILD_STR		"10.2.273.0"
 #define BE_NAME			"Emulex OneConnect" \
 				"Open-iSCSI Driver version" BUILD_STR
 #define DRV_DESC		BE_NAME " " "Driver"
@@ -71,8 +71,8 @@
 
 #define BEISCSI_SGLIST_ELEMENTS	30
 
-#define BEISCSI_CMD_PER_LUN	128	/* scsi_host->cmd_per_lun */
-#define BEISCSI_MAX_SECTORS	2048	/* scsi_host->max_sectors */
+#define BEISCSI_CMD_PER_LUN	128 /* scsi_host->cmd_per_lun */
+#define BEISCSI_MAX_SECTORS	1024 /* scsi_host->max_sectors */
 #define BEISCSI_TEMPLATE_HDR_PER_CXN_SIZE 128 /* Template size per cxn */
 
 #define BEISCSI_MAX_CMD_LEN	16	/* scsi_host->max_cmd_len */
@@ -427,6 +427,7 @@ struct beiscsi_hba {
 	struct mgmt_session_info boot_sess;
 	struct invalidate_command_table inv_tbl[128];
 
+	struct be_aic_obj aic_obj[MAX_CPUS];
 	unsigned int attr_log_enable;
 	int (*iotask_fn)(struct iscsi_task *,
 			struct scatterlist *sg,
diff --git a/drivers/scsi/be2iscsi/be_mgmt.c b/drivers/scsi/be2iscsi/be_mgmt.c
index 088bdf752cfa..07934b0b9ee1 100644
--- a/drivers/scsi/be2iscsi/be_mgmt.c
+++ b/drivers/scsi/be2iscsi/be_mgmt.c
@@ -155,6 +155,43 @@ void beiscsi_ue_detect(struct beiscsi_hba *phba)
 	}
 }
 
+int be_cmd_modify_eq_delay(struct beiscsi_hba *phba,
+		 struct be_set_eqd *set_eqd, int num)
+{
+	struct be_ctrl_info *ctrl = &phba->ctrl;
+	struct be_mcc_wrb *wrb;
+	struct be_cmd_req_modify_eq_delay *req;
+	unsigned int tag = 0;
+	int i;
+
+	spin_lock(&ctrl->mbox_lock);
+	tag = alloc_mcc_tag(phba);
+	if (!tag) {
+		spin_unlock(&ctrl->mbox_lock);
+		return tag;
+	}
+
+	wrb = wrb_from_mccq(phba);
+	req = embedded_payload(wrb);
+
+	wrb->tag0 |= tag;
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_MODIFY_EQ_DELAY, sizeof(*req));
+
+	req->num_eq = cpu_to_le32(num);
+	for (i = 0; i < num; i++) {
+		req->delay[i].eq_id = cpu_to_le32(set_eqd[i].eq_id);
+		req->delay[i].phase = 0;
+		req->delay[i].delay_multiplier =
+				cpu_to_le32(set_eqd[i].delay_multiplier);
+	}
+
+	be_mcc_notify(phba);
+	spin_unlock(&ctrl->mbox_lock);
+	return tag;
+}
+
 /**
  * mgmt_reopen_session()- Reopen a session based on reopen_type
  * @phba: Device priv structure instance
@@ -447,8 +484,8 @@ unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl,
 					 struct be_dma_mem *nonemb_cmd)
 {
 	struct be_cmd_resp_hdr *resp;
-	struct be_mcc_wrb *wrb = wrb_from_mccq(phba);
-	struct be_sge *mcc_sge = nonembedded_sgl(wrb);
+	struct be_mcc_wrb *wrb;
+	struct be_sge *mcc_sge;
 	unsigned int tag = 0;
 	struct iscsi_bsg_request *bsg_req = job->request;
 	struct be_bsg_vendor_cmd *req = nonemb_cmd->va;
@@ -465,7 +502,6 @@ unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl,
 	req->sector = sector;
 	req->offset = offset;
 	spin_lock(&ctrl->mbox_lock);
-	memset(wrb, 0, sizeof(*wrb));
 
 	switch (bsg_req->rqst_data.h_vendor.vendor_cmd[0]) {
 	case BEISCSI_WRITE_FLASH:
@@ -495,6 +531,8 @@ unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl,
 		return tag;
 	}
 
+	wrb = wrb_from_mccq(phba);
+	mcc_sge = nonembedded_sgl(wrb);
 	be_wrb_hdr_prepare(wrb, nonemb_cmd->size, false,
 			   job->request_payload.sg_cnt);
 	mcc_sge->pa_hi = cpu_to_le32(upper_32_bits(nonemb_cmd->dma));
@@ -525,7 +563,6 @@ int mgmt_epfw_cleanup(struct beiscsi_hba *phba, unsigned short ulp_num)
 	int status = 0;
 
 	spin_lock(&ctrl->mbox_lock);
-	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
 	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI,
@@ -675,7 +712,7 @@ int mgmt_open_connection(struct beiscsi_hba *phba,
 	struct sockaddr_in6 *daddr_in6 = (struct sockaddr_in6 *)dst_addr;
 	struct be_ctrl_info *ctrl = &phba->ctrl;
 	struct be_mcc_wrb *wrb;
-	struct tcp_connect_and_offload_in *req;
+	struct tcp_connect_and_offload_in_v1 *req;
 	unsigned short def_hdr_id;
 	unsigned short def_data_id;
 	struct phys_addr template_address = { 0, 0 };
@@ -702,17 +739,16 @@ int mgmt_open_connection(struct beiscsi_hba *phba,
 		return tag;
 	}
 	wrb = wrb_from_mccq(phba);
-	memset(wrb, 0, sizeof(*wrb));
 	sge = nonembedded_sgl(wrb);
 
 	req = nonemb_cmd->va;
 	memset(req, 0, sizeof(*req));
 	wrb->tag0 |= tag;
 
-	be_wrb_hdr_prepare(wrb, sizeof(*req), false, 1);
+	be_wrb_hdr_prepare(wrb, nonemb_cmd->size, false, 1);
 	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI,
 			   OPCODE_COMMON_ISCSI_TCP_CONNECT_AND_OFFLOAD,
-			   sizeof(*req));
+			   nonemb_cmd->size);
 	if (dst_addr->sa_family == PF_INET) {
 		__be32 s_addr = daddr_in->sin_addr.s_addr;
 		req->ip_address.ip_type = BE2_IPV4;
@@ -758,6 +794,13 @@ int mgmt_open_connection(struct beiscsi_hba *phba,
 	sge->pa_hi = cpu_to_le32(upper_32_bits(nonemb_cmd->dma));
 	sge->pa_lo = cpu_to_le32(nonemb_cmd->dma & 0xFFFFFFFF);
 	sge->len = cpu_to_le32(nonemb_cmd->size);
+
+	if (!is_chip_be2_be3r(phba)) {
+		req->hdr.version = MBX_CMD_VER1;
+		req->tcp_window_size = 0;
+		req->tcp_window_scale_count = 2;
+	}
+
 	be_mcc_notify(phba);
 	spin_unlock(&ctrl->mbox_lock);
 	return tag;
@@ -804,7 +847,7 @@ static int mgmt_exec_nonemb_cmd(struct beiscsi_hba *phba,
 				int resp_buf_len)
 {
 	struct be_ctrl_info *ctrl = &phba->ctrl;
-	struct be_mcc_wrb *wrb = wrb_from_mccq(phba);
+	struct be_mcc_wrb *wrb;
 	struct be_sge *sge;
 	unsigned int tag;
 	int rc = 0;
@@ -816,7 +859,8 @@ static int mgmt_exec_nonemb_cmd(struct beiscsi_hba *phba,
 		rc = -ENOMEM;
 		goto free_cmd;
 	}
-	memset(wrb, 0, sizeof(*wrb));
+
+	wrb = wrb_from_mccq(phba);
 	wrb->tag0 |= tag;
 	sge = nonembedded_sgl(wrb);
 
@@ -964,10 +1008,8 @@ int mgmt_set_ip(struct beiscsi_hba *phba,
 		BE2_IPV6 : BE2_IPV4 ;
 
 	rc = mgmt_get_if_info(phba, ip_type, &if_info);
-	if (rc) {
-		kfree(if_info);
+	if (rc)
 		return rc;
-	}
 
 	if (boot_proto == ISCSI_BOOTPROTO_DHCP) {
 		if (if_info->dhcp_state) {
diff --git a/drivers/scsi/be2iscsi/be_mgmt.h b/drivers/scsi/be2iscsi/be_mgmt.h
index 01b8c97284c0..24a8fc577477 100644
--- a/drivers/scsi/be2iscsi/be_mgmt.h
+++ b/drivers/scsi/be2iscsi/be_mgmt.h
@@ -335,5 +335,7 @@ void beiscsi_offload_cxn_v0(struct beiscsi_offload_params *params,
 void beiscsi_offload_cxn_v2(struct beiscsi_offload_params *params,
 			     struct wrb_handle *pwrb_handle);
 void beiscsi_ue_detect(struct beiscsi_hba *phba);
+int be_cmd_modify_eq_delay(struct beiscsi_hba *phba,
+			 struct be_set_eqd *, int num);
 
 #endif
diff --git a/drivers/scsi/bfa/bfad.c b/drivers/scsi/bfa/bfad.c
index cc0fbcdc5192..7593b7c1d336 100644
--- a/drivers/scsi/bfa/bfad.c
+++ b/drivers/scsi/bfa/bfad.c
@@ -507,7 +507,7 @@ bfa_fcb_pbc_vport_create(struct bfad_s *bfad, struct bfi_pbc_vport_s pbc_vport)
 	struct bfad_vport_s   *vport;
 	int rc;
 
-	vport = kzalloc(sizeof(struct bfad_vport_s), GFP_KERNEL);
+	vport = kzalloc(sizeof(struct bfad_vport_s), GFP_ATOMIC);
 	if (!vport) {
 		bfa_trc(bfad, 0);
 		return;
diff --git a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
index f54843023466..785d0d71781e 100644
--- a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
+++ b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
@@ -516,23 +516,17 @@ static void bnx2fc_recv_frame(struct sk_buff *skb)
 	skb_pull(skb, sizeof(struct fcoe_hdr));
 	fr_len = skb->len - sizeof(struct fcoe_crc_eof);
 
-	stats = per_cpu_ptr(lport->stats, get_cpu());
-	stats->RxFrames++;
-	stats->RxWords += fr_len / FCOE_WORD_TO_BYTE;
-
 	fp = (struct fc_frame *)skb;
 	fc_frame_init(fp);
 	fr_dev(fp) = lport;
 	fr_sof(fp) = hp->fcoe_sof;
 	if (skb_copy_bits(skb, fr_len, &crc_eof, sizeof(crc_eof))) {
-		put_cpu();
 		kfree_skb(skb);
 		return;
 	}
 	fr_eof(fp) = crc_eof.fcoe_eof;
 	fr_crc(fp) = crc_eof.fcoe_crc32;
 	if (pskb_trim(skb, fr_len)) {
-		put_cpu();
 		kfree_skb(skb);
 		return;
 	}
@@ -544,7 +538,6 @@ static void bnx2fc_recv_frame(struct sk_buff *skb)
 		port = lport_priv(vn_port);
 		if (!ether_addr_equal(port->data_src_addr, dest_mac)) {
 			BNX2FC_HBA_DBG(lport, "fpma mismatch\n");
-			put_cpu();
 			kfree_skb(skb);
 			return;
 		}
@@ -552,7 +545,6 @@ static void bnx2fc_recv_frame(struct sk_buff *skb)
 	if (fh->fh_r_ctl == FC_RCTL_DD_SOL_DATA &&
 	    fh->fh_type == FC_TYPE_FCP) {
 		/* Drop FCP data. We dont this in L2 path */
-		put_cpu();
 		kfree_skb(skb);
 		return;
 	}
@@ -562,7 +554,6 @@ static void bnx2fc_recv_frame(struct sk_buff *skb)
 		case ELS_LOGO:
 			if (ntoh24(fh->fh_s_id) == FC_FID_FLOGI) {
 				/* drop non-FIP LOGO */
-				put_cpu();
 				kfree_skb(skb);
 				return;
 			}
@@ -572,22 +563,23 @@ static void bnx2fc_recv_frame(struct sk_buff *skb)
 
 	if (fh->fh_r_ctl == FC_RCTL_BA_ABTS) {
 		/* Drop incoming ABTS */
-		put_cpu();
 		kfree_skb(skb);
 		return;
 	}
 
+	stats = per_cpu_ptr(lport->stats, smp_processor_id());
+	stats->RxFrames++;
+	stats->RxWords += fr_len / FCOE_WORD_TO_BYTE;
+
 	if (le32_to_cpu(fr_crc(fp)) !=
 			~crc32(~0, skb->data, fr_len)) {
 		if (stats->InvalidCRCCount < 5)
 			printk(KERN_WARNING PFX "dropping frame with "
 			       "CRC error\n");
 		stats->InvalidCRCCount++;
-		put_cpu();
 		kfree_skb(skb);
 		return;
 	}
-	put_cpu();
 	fc_exch_recv(lport, fp);
 }
 
diff --git a/drivers/scsi/bnx2fc/bnx2fc_hwi.c b/drivers/scsi/bnx2fc/bnx2fc_hwi.c
index 46a37657307f..512aed3ae4f1 100644
--- a/drivers/scsi/bnx2fc/bnx2fc_hwi.c
+++ b/drivers/scsi/bnx2fc/bnx2fc_hwi.c
@@ -1966,26 +1966,29 @@ static void bnx2fc_free_hash_table(struct bnx2fc_hba *hba)
 {
 	int i;
 	int segment_count;
-	int hash_table_size;
 	u32 *pbl;
 
-	segment_count = hba->hash_tbl_segment_count;
-	hash_table_size = BNX2FC_NUM_MAX_SESS * BNX2FC_MAX_ROWS_IN_HASH_TBL *
-		sizeof(struct fcoe_hash_table_entry);
+	if (hba->hash_tbl_segments) {
 
-	pbl = hba->hash_tbl_pbl;
-	for (i = 0; i < segment_count; ++i) {
-		dma_addr_t dma_address;
+		pbl = hba->hash_tbl_pbl;
+		if (pbl) {
+			segment_count = hba->hash_tbl_segment_count;
+			for (i = 0; i < segment_count; ++i) {
+				dma_addr_t dma_address;
 
-		dma_address = le32_to_cpu(*pbl);
-		++pbl;
-		dma_address += ((u64)le32_to_cpu(*pbl)) << 32;
-		++pbl;
-		dma_free_coherent(&hba->pcidev->dev,
-				  BNX2FC_HASH_TBL_CHUNK_SIZE,
-				  hba->hash_tbl_segments[i],
-				  dma_address);
+				dma_address = le32_to_cpu(*pbl);
+				++pbl;
+				dma_address += ((u64)le32_to_cpu(*pbl)) << 32;
+				++pbl;
+				dma_free_coherent(&hba->pcidev->dev,
+						  BNX2FC_HASH_TBL_CHUNK_SIZE,
+						  hba->hash_tbl_segments[i],
+						  dma_address);
+			}
+		}
 
+		kfree(hba->hash_tbl_segments);
+		hba->hash_tbl_segments = NULL;
 	}
 
 	if (hba->hash_tbl_pbl) {
@@ -2023,7 +2026,7 @@ static int bnx2fc_allocate_hash_table(struct bnx2fc_hba *hba)
 	dma_segment_array = kzalloc(dma_segment_array_size, GFP_KERNEL);
 	if (!dma_segment_array) {
 		printk(KERN_ERR PFX "hash table pointers (dma) alloc failed\n");
-		return -ENOMEM;
+		goto cleanup_ht;
 	}
 
 	for (i = 0; i < segment_count; ++i) {
@@ -2034,15 +2037,7 @@ static int bnx2fc_allocate_hash_table(struct bnx2fc_hba *hba)
 					   GFP_KERNEL);
 		if (!hba->hash_tbl_segments[i]) {
 			printk(KERN_ERR PFX "hash segment alloc failed\n");
-			while (--i >= 0) {
-				dma_free_coherent(&hba->pcidev->dev,
-						    BNX2FC_HASH_TBL_CHUNK_SIZE,
-						    hba->hash_tbl_segments[i],
-						    dma_segment_array[i]);
-				hba->hash_tbl_segments[i] = NULL;
-			}
-			kfree(dma_segment_array);
-			return -ENOMEM;
+			goto cleanup_dma;
 		}
 		memset(hba->hash_tbl_segments[i], 0,
 		       BNX2FC_HASH_TBL_CHUNK_SIZE);
@@ -2054,8 +2049,7 @@ static int bnx2fc_allocate_hash_table(struct bnx2fc_hba *hba)
 					       GFP_KERNEL);
 	if (!hba->hash_tbl_pbl) {
 		printk(KERN_ERR PFX "hash table pbl alloc failed\n");
-		kfree(dma_segment_array);
-		return -ENOMEM;
+		goto cleanup_dma;
 	}
 	memset(hba->hash_tbl_pbl, 0, PAGE_SIZE);
 
@@ -2080,6 +2074,22 @@ static int bnx2fc_allocate_hash_table(struct bnx2fc_hba *hba)
 	}
 	kfree(dma_segment_array);
 	return 0;
+
+cleanup_dma:
+	for (i = 0; i < segment_count; ++i) {
+		if (hba->hash_tbl_segments[i])
+			dma_free_coherent(&hba->pcidev->dev,
+					    BNX2FC_HASH_TBL_CHUNK_SIZE,
+					    hba->hash_tbl_segments[i],
+					    dma_segment_array[i]);
+	}
+
+	kfree(dma_segment_array);
+
+cleanup_ht:
+	kfree(hba->hash_tbl_segments);
+	hba->hash_tbl_segments = NULL;
+	return -ENOMEM;
 }
 
 /**
diff --git a/drivers/scsi/bnx2fc/bnx2fc_io.c b/drivers/scsi/bnx2fc/bnx2fc_io.c
index 32a5e0a2a669..7bc47fc7c686 100644
--- a/drivers/scsi/bnx2fc/bnx2fc_io.c
+++ b/drivers/scsi/bnx2fc/bnx2fc_io.c
@@ -282,6 +282,8 @@ struct bnx2fc_cmd_mgr *bnx2fc_cmd_mgr_alloc(struct bnx2fc_hba *hba)
 				       arr_sz, GFP_KERNEL);
 	if (!cmgr->free_list_lock) {
 		printk(KERN_ERR PFX "failed to alloc free_list_lock\n");
+		kfree(cmgr->free_list);
+		cmgr->free_list = NULL;
 		goto mem_err;
 	}
 
diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c
index 5248c888552b..7bcf67eec921 100644
--- a/drivers/scsi/device_handler/scsi_dh_alua.c
+++ b/drivers/scsi/device_handler/scsi_dh_alua.c
@@ -120,6 +120,7 @@ static struct request *get_alua_req(struct scsi_device *sdev,
 			    "%s: blk_get_request failed\n", __func__);
 		return NULL;
 	}
+	blk_rq_set_block_pc(rq);
 
 	if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_NOIO)) {
 		blk_put_request(rq);
@@ -128,7 +129,6 @@ static struct request *get_alua_req(struct scsi_device *sdev,
 		return NULL;
 	}
 
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			 REQ_FAILFAST_DRIVER;
 	rq->retries = ALUA_FAILOVER_RETRIES;
diff --git a/drivers/scsi/device_handler/scsi_dh_emc.c b/drivers/scsi/device_handler/scsi_dh_emc.c
index e1c8be06de9d..6f07f7fe3aa1 100644
--- a/drivers/scsi/device_handler/scsi_dh_emc.c
+++ b/drivers/scsi/device_handler/scsi_dh_emc.c
@@ -280,6 +280,7 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
 		return NULL;
 	}
 
+	blk_rq_set_block_pc(rq);
 	rq->cmd_len = COMMAND_SIZE(cmd);
 	rq->cmd[0] = cmd;
 
@@ -304,7 +305,6 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
 		break;
 	}
 
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			 REQ_FAILFAST_DRIVER;
 	rq->timeout = CLARIION_TIMEOUT;
diff --git a/drivers/scsi/device_handler/scsi_dh_hp_sw.c b/drivers/scsi/device_handler/scsi_dh_hp_sw.c
index 084062bb8ee9..e9d9fea9e272 100644
--- a/drivers/scsi/device_handler/scsi_dh_hp_sw.c
+++ b/drivers/scsi/device_handler/scsi_dh_hp_sw.c
@@ -120,7 +120,7 @@ retry:
 	if (!req)
 		return SCSI_DH_RES_TEMP_UNAVAIL;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			  REQ_FAILFAST_DRIVER;
 	req->cmd_len = COMMAND_SIZE(TEST_UNIT_READY);
@@ -250,7 +250,7 @@ static int hp_sw_start_stop(struct hp_sw_dh_data *h)
 	if (!req)
 		return SCSI_DH_RES_TEMP_UNAVAIL;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			  REQ_FAILFAST_DRIVER;
 	req->cmd_len = COMMAND_SIZE(START_STOP);
diff --git a/drivers/scsi/device_handler/scsi_dh_rdac.c b/drivers/scsi/device_handler/scsi_dh_rdac.c
index 4b9cf93f3fb6..826069db9848 100644
--- a/drivers/scsi/device_handler/scsi_dh_rdac.c
+++ b/drivers/scsi/device_handler/scsi_dh_rdac.c
@@ -279,6 +279,7 @@ static struct request *get_rdac_req(struct scsi_device *sdev,
 				"get_rdac_req: blk_get_request failed.\n");
 		return NULL;
 	}
+	blk_rq_set_block_pc(rq);
 
 	if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_NOIO)) {
 		blk_put_request(rq);
@@ -287,7 +288,6 @@ static struct request *get_rdac_req(struct scsi_device *sdev,
 		return NULL;
 	}
 
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
 			 REQ_FAILFAST_DRIVER;
 	rq->retries = RDAC_RETRIES;
diff --git a/drivers/scsi/dtc.c b/drivers/scsi/dtc.c
index eb29fe7eaf49..0a667fe05006 100644
--- a/drivers/scsi/dtc.c
+++ b/drivers/scsi/dtc.c
@@ -3,8 +3,6 @@
 #define PSEUDO_DMA
 #define DONT_USE_INTR
 #define UNSAFE			/* Leave interrupts enabled during pseudo-dma I/O */
-#define xNDEBUG (NDEBUG_INTR+NDEBUG_RESELECTION+\
-		 NDEBUG_SELECTION+NDEBUG_ARBITRATION)
 #define DMA_WORKS_RIGHT
 
 
diff --git a/drivers/scsi/esas2r/esas2r_main.c b/drivers/scsi/esas2r/esas2r_main.c
index f37f3e3dd5d5..6504a195c874 100644
--- a/drivers/scsi/esas2r/esas2r_main.c
+++ b/drivers/scsi/esas2r/esas2r_main.c
@@ -390,7 +390,7 @@ static int esas2r_probe(struct pci_dev *pcid,
 	esas2r_log_dev(ESAS2R_LOG_INFO, &(pcid->dev),
 		       "pci_enable_device() OK");
 	esas2r_log_dev(ESAS2R_LOG_INFO, &(pcid->dev),
-		       "after pci_device_enable() enable_cnt: %d",
+		       "after pci_enable_device() enable_cnt: %d",
 		       pcid->enable_cnt.counter);
 
 	host = scsi_host_alloc(&driver_template, host_alloc_size);
diff --git a/drivers/scsi/fnic/fnic.h b/drivers/scsi/fnic/fnic.h
index 528d43b7b569..1d3521e13d77 100644
--- a/drivers/scsi/fnic/fnic.h
+++ b/drivers/scsi/fnic/fnic.h
@@ -39,14 +39,15 @@
 
 #define DRV_NAME		"fnic"
 #define DRV_DESCRIPTION		"Cisco FCoE HBA Driver"
-#define DRV_VERSION		"1.5.0.45"
+#define DRV_VERSION		"1.6.0.10"
 #define PFX			DRV_NAME ": "
 #define DFX                     DRV_NAME "%d: "
 
 #define DESC_CLEAN_LOW_WATERMARK 8
 #define FNIC_UCSM_DFLT_THROTTLE_CNT_BLD	16 /* UCSM default throttle count */
 #define FNIC_MIN_IO_REQ			256 /* Min IO throttle count */
-#define FNIC_MAX_IO_REQ		2048 /* scsi_cmnd tag map entries */
+#define FNIC_MAX_IO_REQ		1024 /* scsi_cmnd tag map entries */
+#define FNIC_DFLT_IO_REQ        256 /* Default scsi_cmnd tag map entries */
 #define	FNIC_IO_LOCKS		64 /* IO locks: power of 2 */
 #define FNIC_DFLT_QUEUE_DEPTH	32
 #define	FNIC_STATS_RATE_LIMIT	4 /* limit rate at which stats are pulled up */
diff --git a/drivers/scsi/fnic/fnic_debugfs.c b/drivers/scsi/fnic/fnic_debugfs.c
index b6073f875761..2c613bdea78f 100644
--- a/drivers/scsi/fnic/fnic_debugfs.c
+++ b/drivers/scsi/fnic/fnic_debugfs.c
@@ -25,6 +25,21 @@ static struct dentry *fnic_trace_debugfs_file;
 static struct dentry *fnic_trace_enable;
 static struct dentry *fnic_stats_debugfs_root;
 
+static struct dentry *fnic_fc_trace_debugfs_file;
+static struct dentry *fnic_fc_rdata_trace_debugfs_file;
+static struct dentry *fnic_fc_trace_enable;
+static struct dentry *fnic_fc_trace_clear;
+
+struct fc_trace_flag_type {
+	u8 fc_row_file;
+	u8 fc_normal_file;
+	u8 fnic_trace;
+	u8 fc_trace;
+	u8 fc_clear;
+};
+
+static struct fc_trace_flag_type *fc_trc_flag;
+
 /*
  * fnic_debugfs_init - Initialize debugfs for fnic debug logging
  *
@@ -56,6 +71,18 @@ int fnic_debugfs_init(void)
 		return rc;
 	}
 
+	/* Allocate memory to structure */
+	fc_trc_flag = (struct fc_trace_flag_type *)
+		vmalloc(sizeof(struct fc_trace_flag_type));
+
+	if (fc_trc_flag) {
+		fc_trc_flag->fc_row_file = 0;
+		fc_trc_flag->fc_normal_file = 1;
+		fc_trc_flag->fnic_trace = 2;
+		fc_trc_flag->fc_trace = 3;
+		fc_trc_flag->fc_clear = 4;
+	}
+
 	rc = 0;
 	return rc;
 }
@@ -74,15 +101,19 @@ void fnic_debugfs_terminate(void)
 
 	debugfs_remove(fnic_trace_debugfs_root);
 	fnic_trace_debugfs_root = NULL;
+
+	if (fc_trc_flag)
+		vfree(fc_trc_flag);
 }
 
 /*
- * fnic_trace_ctrl_open - Open the trace_enable file
+ * fnic_trace_ctrl_open - Open the trace_enable file for fnic_trace
+ *               Or Open fc_trace_enable file for fc_trace
  * @inode: The inode pointer.
  * @file: The file pointer to attach the trace enable/disable flag.
  *
  * Description:
- * This routine opens a debugsfs file trace_enable.
+ * This routine opens a debugsfs file trace_enable or fc_trace_enable.
  *
  * Returns:
  * This function returns zero if successful.
@@ -94,15 +125,19 @@ static int fnic_trace_ctrl_open(struct inode *inode, struct file *filp)
 }
 
 /*
- * fnic_trace_ctrl_read - Read a trace_enable debugfs file
+ * fnic_trace_ctrl_read -
+ *          Read  trace_enable ,fc_trace_enable
+ *              or fc_trace_clear debugfs file
  * @filp: The file pointer to read from.
  * @ubuf: The buffer to copy the data to.
  * @cnt: The number of bytes to read.
  * @ppos: The position in the file to start reading from.
  *
  * Description:
- * This routine reads value of variable fnic_tracing_enabled
- * and stores into local @buf. It will start reading file at @ppos and
+ * This routine reads value of variable fnic_tracing_enabled or
+ * fnic_fc_tracing_enabled or fnic_fc_trace_cleared
+ * and stores into local @buf.
+ * It will start reading file at @ppos and
  * copy up to @cnt of data to @ubuf from @buf.
  *
  * Returns:
@@ -114,13 +149,25 @@ static ssize_t fnic_trace_ctrl_read(struct file *filp,
 {
 	char buf[64];
 	int len;
-	len = sprintf(buf, "%u\n", fnic_tracing_enabled);
+	u8 *trace_type;
+	len = 0;
+	trace_type = (u8 *)filp->private_data;
+	if (*trace_type == fc_trc_flag->fnic_trace)
+		len = sprintf(buf, "%u\n", fnic_tracing_enabled);
+	else if (*trace_type == fc_trc_flag->fc_trace)
+		len = sprintf(buf, "%u\n", fnic_fc_tracing_enabled);
+	else if (*trace_type == fc_trc_flag->fc_clear)
+		len = sprintf(buf, "%u\n", fnic_fc_trace_cleared);
+	else
+		pr_err("fnic: Cannot read to any debugfs file\n");
 
 	return simple_read_from_buffer(ubuf, cnt, ppos, buf, len);
 }
 
 /*
- * fnic_trace_ctrl_write - Write to trace_enable debugfs file
+ * fnic_trace_ctrl_write -
+ * Write to trace_enable, fc_trace_enable or
+ *         fc_trace_clear debugfs file
  * @filp: The file pointer to write from.
  * @ubuf: The buffer to copy the data from.
  * @cnt: The number of bytes to write.
@@ -128,7 +175,8 @@ static ssize_t fnic_trace_ctrl_read(struct file *filp,
  *
  * Description:
  * This routine writes data from user buffer @ubuf to buffer @buf and
- * sets fnic_tracing_enabled value as per user input.
+ * sets fc_trace_enable ,tracing_enable or fnic_fc_trace_cleared
+ * value as per user input.
  *
  * Returns:
  * This function returns the amount of data that was written.
@@ -140,6 +188,8 @@ static ssize_t fnic_trace_ctrl_write(struct file *filp,
 	char buf[64];
 	unsigned long val;
 	int ret;
+	u8 *trace_type;
+	trace_type = (u8 *)filp->private_data;
 
 	if (cnt >= sizeof(buf))
 		return -EINVAL;
@@ -153,12 +203,27 @@ static ssize_t fnic_trace_ctrl_write(struct file *filp,
 	if (ret < 0)
 		return ret;
 
-	fnic_tracing_enabled = val;
+	if (*trace_type == fc_trc_flag->fnic_trace)
+		fnic_tracing_enabled = val;
+	else if (*trace_type == fc_trc_flag->fc_trace)
+		fnic_fc_tracing_enabled = val;
+	else if (*trace_type == fc_trc_flag->fc_clear)
+		fnic_fc_trace_cleared = val;
+	else
+		pr_err("fnic: cannot write to any debufs file\n");
+
 	(*ppos)++;
 
 	return cnt;
 }
 
+static const struct file_operations fnic_trace_ctrl_fops = {
+	.owner = THIS_MODULE,
+	.open = fnic_trace_ctrl_open,
+	.read = fnic_trace_ctrl_read,
+	.write = fnic_trace_ctrl_write,
+};
+
 /*
  * fnic_trace_debugfs_open - Open the fnic trace log
  * @inode: The inode pointer
@@ -178,19 +243,36 @@ static int fnic_trace_debugfs_open(struct inode *inode,
 				  struct file *file)
 {
 	fnic_dbgfs_t *fnic_dbg_prt;
+	u8 *rdata_ptr;
+	rdata_ptr = (u8 *)inode->i_private;
 	fnic_dbg_prt = kzalloc(sizeof(fnic_dbgfs_t), GFP_KERNEL);
 	if (!fnic_dbg_prt)
 		return -ENOMEM;
 
-	fnic_dbg_prt->buffer = vmalloc((3*(trace_max_pages * PAGE_SIZE)));
-	if (!fnic_dbg_prt->buffer) {
-		kfree(fnic_dbg_prt);
-		return -ENOMEM;
+	if (*rdata_ptr == fc_trc_flag->fnic_trace) {
+		fnic_dbg_prt->buffer = vmalloc(3 *
+					(trace_max_pages * PAGE_SIZE));
+		if (!fnic_dbg_prt->buffer) {
+			kfree(fnic_dbg_prt);
+			return -ENOMEM;
+		}
+		memset((void *)fnic_dbg_prt->buffer, 0,
+		3 * (trace_max_pages * PAGE_SIZE));
+		fnic_dbg_prt->buffer_len = fnic_get_trace_data(fnic_dbg_prt);
+	} else {
+		fnic_dbg_prt->buffer =
+			vmalloc(3 * (fnic_fc_trace_max_pages * PAGE_SIZE));
+		if (!fnic_dbg_prt->buffer) {
+			kfree(fnic_dbg_prt);
+			return -ENOMEM;
+		}
+		memset((void *)fnic_dbg_prt->buffer, 0,
+			3 * (fnic_fc_trace_max_pages * PAGE_SIZE));
+		fnic_dbg_prt->buffer_len =
+			fnic_fc_trace_get_data(fnic_dbg_prt, *rdata_ptr);
 	}
-	memset((void *)fnic_dbg_prt->buffer, 0,
-			  (3*(trace_max_pages * PAGE_SIZE)));
-	fnic_dbg_prt->buffer_len = fnic_get_trace_data(fnic_dbg_prt);
 	file->private_data = fnic_dbg_prt;
+
 	return 0;
 }
 
@@ -272,13 +354,6 @@ static int fnic_trace_debugfs_release(struct inode *inode,
 	return 0;
 }
 
-static const struct file_operations fnic_trace_ctrl_fops = {
-	.owner = THIS_MODULE,
-	.open = fnic_trace_ctrl_open,
-	.read = fnic_trace_ctrl_read,
-	.write = fnic_trace_ctrl_write,
-};
-
 static const struct file_operations fnic_trace_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.open = fnic_trace_debugfs_open,
@@ -306,9 +381,10 @@ int fnic_trace_debugfs_init(void)
 		return rc;
 	}
 	fnic_trace_enable = debugfs_create_file("tracing_enable",
-					  S_IFREG|S_IRUGO|S_IWUSR,
-					  fnic_trace_debugfs_root,
-					  NULL, &fnic_trace_ctrl_fops);
+					S_IFREG|S_IRUGO|S_IWUSR,
+					fnic_trace_debugfs_root,
+					&(fc_trc_flag->fnic_trace),
+					&fnic_trace_ctrl_fops);
 
 	if (!fnic_trace_enable) {
 		printk(KERN_DEBUG
@@ -317,10 +393,10 @@ int fnic_trace_debugfs_init(void)
 	}
 
 	fnic_trace_debugfs_file = debugfs_create_file("trace",
-						  S_IFREG|S_IRUGO|S_IWUSR,
-						  fnic_trace_debugfs_root,
-						  NULL,
-						  &fnic_trace_debugfs_fops);
+					S_IFREG|S_IRUGO|S_IWUSR,
+					fnic_trace_debugfs_root,
+					&(fc_trc_flag->fnic_trace),
+					&fnic_trace_debugfs_fops);
 
 	if (!fnic_trace_debugfs_file) {
 		printk(KERN_DEBUG
@@ -340,14 +416,104 @@ int fnic_trace_debugfs_init(void)
  */
 void fnic_trace_debugfs_terminate(void)
 {
-	if (fnic_trace_debugfs_file) {
-		debugfs_remove(fnic_trace_debugfs_file);
-		fnic_trace_debugfs_file = NULL;
+	debugfs_remove(fnic_trace_debugfs_file);
+	fnic_trace_debugfs_file = NULL;
+
+	debugfs_remove(fnic_trace_enable);
+	fnic_trace_enable = NULL;
+}
+
+/*
+ * fnic_fc_trace_debugfs_init -
+ * Initialize debugfs for fnic control frame trace logging
+ *
+ * Description:
+ * When Debugfs is configured this routine sets up the fnic_fc debugfs
+ * file system. If not already created, this routine will create the
+ * create file trace to log fnic fc trace buffer output into debugfs and
+ * it will also create file fc_trace_enable to control enable/disable of
+ * trace logging into trace buffer.
+ */
+
+int fnic_fc_trace_debugfs_init(void)
+{
+	int rc = -1;
+
+	if (!fnic_trace_debugfs_root) {
+		pr_err("fnic:Debugfs root directory doesn't exist\n");
+		return rc;
+	}
+
+	fnic_fc_trace_enable = debugfs_create_file("fc_trace_enable",
+					S_IFREG|S_IRUGO|S_IWUSR,
+					fnic_trace_debugfs_root,
+					&(fc_trc_flag->fc_trace),
+					&fnic_trace_ctrl_fops);
+
+	if (!fnic_fc_trace_enable) {
+		pr_err("fnic: Failed create fc_trace_enable file\n");
+		return rc;
+	}
+
+	fnic_fc_trace_clear = debugfs_create_file("fc_trace_clear",
+					S_IFREG|S_IRUGO|S_IWUSR,
+					fnic_trace_debugfs_root,
+					&(fc_trc_flag->fc_clear),
+					&fnic_trace_ctrl_fops);
+
+	if (!fnic_fc_trace_clear) {
+		pr_err("fnic: Failed to create fc_trace_enable file\n");
+		return rc;
+	}
+
+	fnic_fc_rdata_trace_debugfs_file =
+		debugfs_create_file("fc_trace_rdata",
+				    S_IFREG|S_IRUGO|S_IWUSR,
+				    fnic_trace_debugfs_root,
+				    &(fc_trc_flag->fc_normal_file),
+				    &fnic_trace_debugfs_fops);
+
+	if (!fnic_fc_rdata_trace_debugfs_file) {
+		pr_err("fnic: Failed create fc_rdata_trace file\n");
+		return rc;
 	}
-	if (fnic_trace_enable) {
-		debugfs_remove(fnic_trace_enable);
-		fnic_trace_enable = NULL;
+
+	fnic_fc_trace_debugfs_file =
+		debugfs_create_file("fc_trace",
+				    S_IFREG|S_IRUGO|S_IWUSR,
+				    fnic_trace_debugfs_root,
+				    &(fc_trc_flag->fc_row_file),
+				    &fnic_trace_debugfs_fops);
+
+	if (!fnic_fc_trace_debugfs_file) {
+		pr_err("fnic: Failed to create fc_trace file\n");
+		return rc;
 	}
+	rc = 0;
+	return rc;
+}
+
+/*
+ * fnic_fc_trace_debugfs_terminate - Tear down debugfs infrastructure
+ *
+ * Description:
+ * When Debugfs is configured this routine removes debugfs file system
+ * elements that are specific to fnic_fc trace logging.
+ */
+
+void fnic_fc_trace_debugfs_terminate(void)
+{
+	debugfs_remove(fnic_fc_trace_debugfs_file);
+	fnic_fc_trace_debugfs_file = NULL;
+
+	debugfs_remove(fnic_fc_rdata_trace_debugfs_file);
+	fnic_fc_rdata_trace_debugfs_file = NULL;
+
+	debugfs_remove(fnic_fc_trace_enable);
+	fnic_fc_trace_enable = NULL;
+
+	debugfs_remove(fnic_fc_trace_clear);
+	fnic_fc_trace_clear = NULL;
 }
 
 /*
diff --git a/drivers/scsi/fnic/fnic_fcs.c b/drivers/scsi/fnic/fnic_fcs.c
index 1671325aec7f..1b948f633fc5 100644
--- a/drivers/scsi/fnic/fnic_fcs.c
+++ b/drivers/scsi/fnic/fnic_fcs.c
@@ -66,19 +66,35 @@ void fnic_handle_link(struct work_struct *work)
 	fnic->link_down_cnt = vnic_dev_link_down_cnt(fnic->vdev);
 
 	if (old_link_status == fnic->link_status) {
-		if (!fnic->link_status)
+		if (!fnic->link_status) {
 			/* DOWN -> DOWN */
 			spin_unlock_irqrestore(&fnic->fnic_lock, flags);
-		else {
+			fnic_fc_trace_set_data(fnic->lport->host->host_no,
+				FNIC_FC_LE, "Link Status: DOWN->DOWN",
+				strlen("Link Status: DOWN->DOWN"));
+		} else {
 			if (old_link_down_cnt != fnic->link_down_cnt) {
 				/* UP -> DOWN -> UP */
 				fnic->lport->host_stats.link_failure_count++;
 				spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+				fnic_fc_trace_set_data(
+					fnic->lport->host->host_no,
+					FNIC_FC_LE,
+					"Link Status:UP_DOWN_UP",
+					strlen("Link_Status:UP_DOWN_UP")
+					);
 				FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
 					     "link down\n");
 				fcoe_ctlr_link_down(&fnic->ctlr);
 				if (fnic->config.flags & VFCF_FIP_CAPABLE) {
 					/* start FCoE VLAN discovery */
+					fnic_fc_trace_set_data(
+						fnic->lport->host->host_no,
+						FNIC_FC_LE,
+						"Link Status: UP_DOWN_UP_VLAN",
+						strlen(
+						"Link Status: UP_DOWN_UP_VLAN")
+						);
 					fnic_fcoe_send_vlan_req(fnic);
 					return;
 				}
@@ -88,22 +104,36 @@ void fnic_handle_link(struct work_struct *work)
 			} else
 				/* UP -> UP */
 				spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+				fnic_fc_trace_set_data(
+					fnic->lport->host->host_no, FNIC_FC_LE,
+					"Link Status: UP_UP",
+					strlen("Link Status: UP_UP"));
 		}
 	} else if (fnic->link_status) {
 		/* DOWN -> UP */
 		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
 		if (fnic->config.flags & VFCF_FIP_CAPABLE) {
 			/* start FCoE VLAN discovery */
+				fnic_fc_trace_set_data(
+				fnic->lport->host->host_no,
+				FNIC_FC_LE, "Link Status: DOWN_UP_VLAN",
+				strlen("Link Status: DOWN_UP_VLAN"));
 			fnic_fcoe_send_vlan_req(fnic);
 			return;
 		}
 		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, "link up\n");
+		fnic_fc_trace_set_data(fnic->lport->host->host_no, FNIC_FC_LE,
+			"Link Status: DOWN_UP", strlen("Link Status: DOWN_UP"));
 		fcoe_ctlr_link_up(&fnic->ctlr);
 	} else {
 		/* UP -> DOWN */
 		fnic->lport->host_stats.link_failure_count++;
 		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
 		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, "link down\n");
+		fnic_fc_trace_set_data(
+			fnic->lport->host->host_no, FNIC_FC_LE,
+			"Link Status: UP_DOWN",
+			strlen("Link Status: UP_DOWN"));
 		fcoe_ctlr_link_down(&fnic->ctlr);
 	}
 
@@ -267,11 +297,6 @@ static inline int is_fnic_fip_flogi_reject(struct fcoe_ctlr *fip,
 
 	if (desc->fip_dtype == FIP_DT_FLOGI) {
 
-		shost_printk(KERN_DEBUG, lport->host,
-			  " FIP TYPE FLOGI: fab name:%llx "
-			  "vfid:%d map:%x\n",
-			  fip->sel_fcf->fabric_name, fip->sel_fcf->vfid,
-			  fip->sel_fcf->fc_map);
 		if (dlen < sizeof(*els) + sizeof(*fh) + 1)
 			return 0;
 
@@ -616,6 +641,10 @@ static inline int fnic_import_rq_eth_pkt(struct fnic *fnic, struct sk_buff *skb)
 					"using UCSM\n");
 			goto drop;
 		}
+		if ((fnic_fc_trace_set_data(fnic->lport->host->host_no,
+			FNIC_FC_RECV|0x80, (char *)skb->data, skb->len)) != 0) {
+			printk(KERN_ERR "fnic ctlr frame trace error!!!");
+		}
 		skb_queue_tail(&fnic->fip_frame_queue, skb);
 		queue_work(fnic_fip_queue, &fnic->fip_frame_work);
 		return 1;		/* let caller know packet was used */
@@ -844,6 +873,10 @@ static void fnic_rq_cmpl_frame_recv(struct vnic_rq *rq, struct cq_desc
 	}
 	fr_dev(fp) = fnic->lport;
 	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+	if ((fnic_fc_trace_set_data(fnic->lport->host->host_no, FNIC_FC_RECV,
+					(char *)skb->data, skb->len)) != 0) {
+		printk(KERN_ERR "fnic ctlr frame trace error!!!");
+	}
 
 	skb_queue_tail(&fnic->frame_queue, skb);
 	queue_work(fnic_event_queue, &fnic->frame_work);
@@ -951,6 +984,15 @@ void fnic_eth_send(struct fcoe_ctlr *fip, struct sk_buff *skb)
 		vlan_hdr->h_vlan_proto = htons(ETH_P_8021Q);
 		vlan_hdr->h_vlan_encapsulated_proto = eth_hdr->h_proto;
 		vlan_hdr->h_vlan_TCI = htons(fnic->vlan_id);
+		if ((fnic_fc_trace_set_data(fnic->lport->host->host_no,
+			FNIC_FC_SEND|0x80, (char *)eth_hdr, skb->len)) != 0) {
+			printk(KERN_ERR "fnic ctlr frame trace error!!!");
+		}
+	} else {
+		if ((fnic_fc_trace_set_data(fnic->lport->host->host_no,
+			FNIC_FC_SEND|0x80, (char *)skb->data, skb->len)) != 0) {
+			printk(KERN_ERR "fnic ctlr frame trace error!!!");
+		}
 	}
 
 	pa = pci_map_single(fnic->pdev, skb->data, skb->len, PCI_DMA_TODEVICE);
@@ -1023,6 +1065,11 @@ static int fnic_send_frame(struct fnic *fnic, struct fc_frame *fp)
 
 	pa = pci_map_single(fnic->pdev, eth_hdr, tot_len, PCI_DMA_TODEVICE);
 
+	if ((fnic_fc_trace_set_data(fnic->lport->host->host_no, FNIC_FC_SEND,
+				(char *)eth_hdr, tot_len)) != 0) {
+		printk(KERN_ERR "fnic ctlr frame trace error!!!");
+	}
+
 	spin_lock_irqsave(&fnic->wq_lock[0], flags);
 
 	if (!vnic_wq_desc_avail(wq)) {
diff --git a/drivers/scsi/fnic/fnic_main.c b/drivers/scsi/fnic/fnic_main.c
index 33e4ec2bfe73..8c56fdc3a456 100644
--- a/drivers/scsi/fnic/fnic_main.c
+++ b/drivers/scsi/fnic/fnic_main.c
@@ -74,6 +74,11 @@ module_param(fnic_trace_max_pages, uint, S_IRUGO|S_IWUSR);
 MODULE_PARM_DESC(fnic_trace_max_pages, "Total allocated memory pages "
 					"for fnic trace buffer");
 
+unsigned int fnic_fc_trace_max_pages = 64;
+module_param(fnic_fc_trace_max_pages, uint, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(fnic_fc_trace_max_pages,
+		 "Total allocated memory pages for fc trace buffer");
+
 static unsigned int fnic_max_qdepth = FNIC_DFLT_QUEUE_DEPTH;
 module_param(fnic_max_qdepth, uint, S_IRUGO|S_IWUSR);
 MODULE_PARM_DESC(fnic_max_qdepth, "Queue depth to report for each LUN");
@@ -111,7 +116,7 @@ static struct scsi_host_template fnic_host_template = {
 	.change_queue_type = fc_change_queue_type,
 	.this_id = -1,
 	.cmd_per_lun = 3,
-	.can_queue = FNIC_MAX_IO_REQ,
+	.can_queue = FNIC_DFLT_IO_REQ,
 	.use_clustering = ENABLE_CLUSTERING,
 	.sg_tablesize = FNIC_MAX_SG_DESC_CNT,
 	.max_sectors = 0xffff,
@@ -773,6 +778,7 @@ static int fnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		shost_printk(KERN_INFO, fnic->lport->host,
 			     "firmware uses non-FIP mode\n");
 		fcoe_ctlr_init(&fnic->ctlr, FIP_MODE_NON_FIP);
+		fnic->ctlr.state = FIP_ST_NON_FIP;
 	}
 	fnic->state = FNIC_IN_FC_MODE;
 
@@ -1033,11 +1039,20 @@ static int __init fnic_init_module(void)
 	/* Allocate memory for trace buffer */
 	err = fnic_trace_buf_init();
 	if (err < 0) {
-		printk(KERN_ERR PFX "Trace buffer initialization Failed "
-				  "Fnic Tracing utility is disabled\n");
+		printk(KERN_ERR PFX
+		       "Trace buffer initialization Failed. "
+		       "Fnic Tracing utility is disabled\n");
 		fnic_trace_free();
 	}
 
+    /* Allocate memory for fc trace buffer */
+	err = fnic_fc_trace_init();
+	if (err < 0) {
+		printk(KERN_ERR PFX "FC trace buffer initialization Failed "
+		       "FC frame tracing utility is disabled\n");
+		fnic_fc_trace_free();
+	}
+
 	/* Create a cache for allocation of default size sgls */
 	len = sizeof(struct fnic_dflt_sgl_list);
 	fnic_sgl_cache[FNIC_SGL_CACHE_DFLT] = kmem_cache_create
@@ -1118,6 +1133,7 @@ err_create_fnic_sgl_slab_max:
 	kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]);
 err_create_fnic_sgl_slab_dflt:
 	fnic_trace_free();
+	fnic_fc_trace_free();
 	fnic_debugfs_terminate();
 	return err;
 }
@@ -1135,6 +1151,7 @@ static void __exit fnic_cleanup_module(void)
 	kmem_cache_destroy(fnic_io_req_cache);
 	fc_release_transport(fnic_fc_transport);
 	fnic_trace_free();
+	fnic_fc_trace_free();
 	fnic_debugfs_terminate();
 }
 
diff --git a/drivers/scsi/fnic/fnic_scsi.c b/drivers/scsi/fnic/fnic_scsi.c
index 0521436d05d6..ea28b5ca4c73 100644
--- a/drivers/scsi/fnic/fnic_scsi.c
+++ b/drivers/scsi/fnic/fnic_scsi.c
@@ -1312,8 +1312,9 @@ static void fnic_cleanup_io(struct fnic *fnic, int exclude_id)
 
 cleanup_scsi_cmd:
 		sc->result = DID_TRANSPORT_DISRUPTED << 16;
-		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "fnic_cleanup_io:"
-			      " DID_TRANSPORT_DISRUPTED\n");
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "%s: sc duration = %lu DID_TRANSPORT_DISRUPTED\n",
+			      __func__, (jiffies - start_time));
 
 		if (atomic64_read(&fnic->io_cmpl_skip))
 			atomic64_dec(&fnic->io_cmpl_skip);
@@ -1733,6 +1734,7 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 	struct fnic_stats *fnic_stats;
 	struct abort_stats *abts_stats;
 	struct terminate_stats *term_stats;
+	enum fnic_ioreq_state old_ioreq_state;
 	int tag;
 	DECLARE_COMPLETION_ONSTACK(tm_done);
 
@@ -1793,6 +1795,7 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 	 * the completion wont be done till mid-layer, since abort
 	 * has already started.
 	 */
+	old_ioreq_state = CMD_STATE(sc);
 	CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING;
 	CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE;
 
@@ -1816,6 +1819,8 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 	if (fnic_queue_abort_io_req(fnic, sc->request->tag, task_req,
 				    fc_lun.scsi_lun, io_req)) {
 		spin_lock_irqsave(io_lock, flags);
+		if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING)
+			CMD_STATE(sc) = old_ioreq_state;
 		io_req = (struct fnic_io_req *)CMD_SP(sc);
 		if (io_req)
 			io_req->abts_done = NULL;
@@ -1859,12 +1864,8 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 	if (CMD_ABTS_STATUS(sc) == FCPIO_INVALID_CODE) {
 		spin_unlock_irqrestore(io_lock, flags);
 		if (task_req == FCPIO_ITMF_ABT_TASK) {
-			FNIC_SCSI_DBG(KERN_INFO,
-				fnic->lport->host, "Abort Driver Timeout\n");
 			atomic64_inc(&abts_stats->abort_drv_timeouts);
 		} else {
-			FNIC_SCSI_DBG(KERN_INFO, fnic->lport->host,
-				"Terminate Driver Timeout\n");
 			atomic64_inc(&term_stats->terminate_drv_timeouts);
 		}
 		CMD_FLAGS(sc) |= FNIC_IO_ABT_TERM_TIMED_OUT;
diff --git a/drivers/scsi/fnic/fnic_trace.c b/drivers/scsi/fnic/fnic_trace.c
index e002e7187dc0..c77285926827 100644
--- a/drivers/scsi/fnic/fnic_trace.c
+++ b/drivers/scsi/fnic/fnic_trace.c
@@ -20,6 +20,7 @@
 #include <linux/errno.h>
 #include <linux/spinlock.h>
 #include <linux/kallsyms.h>
+#include <linux/time.h>
 #include "fnic_io.h"
 #include "fnic.h"
 
@@ -32,6 +33,16 @@ static DEFINE_SPINLOCK(fnic_trace_lock);
 static fnic_trace_dbg_t fnic_trace_entries;
 int fnic_tracing_enabled = 1;
 
+/* static char *fnic_fc_ctlr_trace_buf_p; */
+
+static int fc_trace_max_entries;
+static unsigned long fnic_fc_ctlr_trace_buf_p;
+static fnic_trace_dbg_t fc_trace_entries;
+int fnic_fc_tracing_enabled = 1;
+int fnic_fc_trace_cleared = 1;
+static DEFINE_SPINLOCK(fnic_fc_trace_lock);
+
+
 /*
  * fnic_trace_get_buf - Give buffer pointer to user to fill up trace information
  *
@@ -428,10 +439,10 @@ int fnic_trace_buf_init(void)
 	}
 	err = fnic_trace_debugfs_init();
 	if (err < 0) {
-		printk(KERN_ERR PFX "Failed to initialize debugfs for tracing\n");
+		pr_err("fnic: Failed to initialize debugfs for tracing\n");
 		goto err_fnic_trace_debugfs_init;
 	}
-	printk(KERN_INFO PFX "Successfully Initialized Trace Buffer\n");
+	pr_info("fnic: Successfully Initialized Trace Buffer\n");
 	return err;
 err_fnic_trace_debugfs_init:
 	fnic_trace_free();
@@ -456,3 +467,314 @@ void fnic_trace_free(void)
 	}
 	printk(KERN_INFO PFX "Successfully Freed Trace Buffer\n");
 }
+
+/*
+ * fnic_fc_ctlr_trace_buf_init -
+ * Initialize trace buffer to log fnic control frames
+ * Description:
+ * Initialize trace buffer data structure by allocating
+ * required memory for trace data as well as for Indexes.
+ * Frame size is 256 bytes and
+ * memory is allocated for 1024 entries of 256 bytes.
+ * Page_offset(Index) is set to the address of trace entry
+ * and page_offset is initialized by adding frame size
+ * to the previous page_offset entry.
+ */
+
+int fnic_fc_trace_init(void)
+{
+	unsigned long fc_trace_buf_head;
+	int err = 0;
+	int i;
+
+	fc_trace_max_entries = (fnic_fc_trace_max_pages * PAGE_SIZE)/
+				FC_TRC_SIZE_BYTES;
+	fnic_fc_ctlr_trace_buf_p = (unsigned long)vmalloc(
+					fnic_fc_trace_max_pages * PAGE_SIZE);
+	if (!fnic_fc_ctlr_trace_buf_p) {
+		pr_err("fnic: Failed to allocate memory for "
+		       "FC Control Trace Buf\n");
+		err = -ENOMEM;
+		goto err_fnic_fc_ctlr_trace_buf_init;
+	}
+
+	memset((void *)fnic_fc_ctlr_trace_buf_p, 0,
+			fnic_fc_trace_max_pages * PAGE_SIZE);
+
+	/* Allocate memory for page offset */
+	fc_trace_entries.page_offset = vmalloc(fc_trace_max_entries *
+						sizeof(unsigned long));
+	if (!fc_trace_entries.page_offset) {
+		pr_err("fnic:Failed to allocate memory for page_offset\n");
+		if (fnic_fc_ctlr_trace_buf_p) {
+			pr_err("fnic: Freeing FC Control Trace Buf\n");
+			vfree((void *)fnic_fc_ctlr_trace_buf_p);
+			fnic_fc_ctlr_trace_buf_p = 0;
+		}
+		err = -ENOMEM;
+		goto err_fnic_fc_ctlr_trace_buf_init;
+	}
+	memset((void *)fc_trace_entries.page_offset, 0,
+	       (fc_trace_max_entries * sizeof(unsigned long)));
+
+	fc_trace_entries.rd_idx = fc_trace_entries.wr_idx = 0;
+	fc_trace_buf_head = fnic_fc_ctlr_trace_buf_p;
+
+	/*
+	* Set up fc_trace_entries.page_offset field with memory location
+	* for every trace entry
+	*/
+	for (i = 0; i < fc_trace_max_entries; i++) {
+		fc_trace_entries.page_offset[i] = fc_trace_buf_head;
+		fc_trace_buf_head += FC_TRC_SIZE_BYTES;
+	}
+	err = fnic_fc_trace_debugfs_init();
+	if (err < 0) {
+		pr_err("fnic: Failed to initialize FC_CTLR tracing.\n");
+		goto err_fnic_fc_ctlr_trace_debugfs_init;
+	}
+	pr_info("fnic: Successfully Initialized FC_CTLR Trace Buffer\n");
+	return err;
+
+err_fnic_fc_ctlr_trace_debugfs_init:
+	fnic_fc_trace_free();
+err_fnic_fc_ctlr_trace_buf_init:
+	return err;
+}
+
+/*
+ * Fnic_fc_ctlr_trace_free - Free memory of fnic_fc_ctlr trace data structures.
+ */
+void fnic_fc_trace_free(void)
+{
+	fnic_fc_tracing_enabled = 0;
+	fnic_fc_trace_debugfs_terminate();
+	if (fc_trace_entries.page_offset) {
+		vfree((void *)fc_trace_entries.page_offset);
+		fc_trace_entries.page_offset = NULL;
+	}
+	if (fnic_fc_ctlr_trace_buf_p) {
+		vfree((void *)fnic_fc_ctlr_trace_buf_p);
+		fnic_fc_ctlr_trace_buf_p = 0;
+	}
+	pr_info("fnic:Successfully FC_CTLR Freed Trace Buffer\n");
+}
+
+/*
+ * fnic_fc_ctlr_set_trace_data:
+ *       Maintain rd & wr idx accordingly and set data
+ * Passed parameters:
+ *       host_no: host number accociated with fnic
+ *       frame_type: send_frame, rece_frame or link event
+ *       fc_frame: pointer to fc_frame
+ *       frame_len: Length of the fc_frame
+ * Description:
+ *   This routine will get next available wr_idx and
+ *   copy all passed trace data to the buffer pointed by wr_idx
+ *   and increment wr_idx. It will also make sure that we dont
+ *   overwrite the entry which we are reading and also
+ *   wrap around if we reach the maximum entries.
+ * Returned Value:
+ *   It will return 0 for success or -1 for failure
+ */
+int fnic_fc_trace_set_data(u32 host_no, u8 frame_type,
+				char *frame, u32 fc_trc_frame_len)
+{
+	unsigned long flags;
+	struct fc_trace_hdr *fc_buf;
+	unsigned long eth_fcoe_hdr_len;
+	char *fc_trace;
+
+	if (fnic_fc_tracing_enabled == 0)
+		return 0;
+
+	spin_lock_irqsave(&fnic_fc_trace_lock, flags);
+
+	if (fnic_fc_trace_cleared == 1) {
+		fc_trace_entries.rd_idx = fc_trace_entries.wr_idx = 0;
+		pr_info("fnic: Reseting the read idx\n");
+		memset((void *)fnic_fc_ctlr_trace_buf_p, 0,
+				fnic_fc_trace_max_pages * PAGE_SIZE);
+		fnic_fc_trace_cleared = 0;
+	}
+
+	fc_buf = (struct fc_trace_hdr *)
+		fc_trace_entries.page_offset[fc_trace_entries.wr_idx];
+
+	fc_trace_entries.wr_idx++;
+
+	if (fc_trace_entries.wr_idx >= fc_trace_max_entries)
+		fc_trace_entries.wr_idx = 0;
+
+	if (fc_trace_entries.wr_idx == fc_trace_entries.rd_idx) {
+		fc_trace_entries.rd_idx++;
+		if (fc_trace_entries.rd_idx >= fc_trace_max_entries)
+			fc_trace_entries.rd_idx = 0;
+	}
+
+	fc_buf->time_stamp = CURRENT_TIME;
+	fc_buf->host_no = host_no;
+	fc_buf->frame_type = frame_type;
+
+	fc_trace = (char *)FC_TRACE_ADDRESS(fc_buf);
+
+	/* During the receive path, we do not have eth hdr as well as fcoe hdr
+	 * at trace entry point so we will stuff 0xff just to make it generic.
+	 */
+	if (frame_type == FNIC_FC_RECV) {
+		eth_fcoe_hdr_len = sizeof(struct ethhdr) +
+					sizeof(struct fcoe_hdr);
+		fc_trc_frame_len = fc_trc_frame_len + eth_fcoe_hdr_len;
+		memset((char *)fc_trace, 0xff, eth_fcoe_hdr_len);
+		/* Copy the rest of data frame */
+		memcpy((char *)(fc_trace + eth_fcoe_hdr_len), (void *)frame,
+		min_t(u8, fc_trc_frame_len,
+			(u8)(FC_TRC_SIZE_BYTES - FC_TRC_HEADER_SIZE)));
+	} else {
+		memcpy((char *)fc_trace, (void *)frame,
+		min_t(u8, fc_trc_frame_len,
+			(u8)(FC_TRC_SIZE_BYTES - FC_TRC_HEADER_SIZE)));
+	}
+
+	/* Store the actual received length */
+	fc_buf->frame_len = fc_trc_frame_len;
+
+	spin_unlock_irqrestore(&fnic_fc_trace_lock, flags);
+	return 0;
+}
+
+/*
+ * fnic_fc_ctlr_get_trace_data: Copy trace buffer to a memory file
+ * Passed parameter:
+ *       @fnic_dbgfs_t: pointer to debugfs trace buffer
+ *       rdata_flag: 1 => Unformated file
+ *                   0 => formated file
+ * Description:
+ *       This routine will copy the trace data to memory file with
+ *       proper formatting and also copy to another memory
+ *       file without formatting for further procesing.
+ * Retrun Value:
+ *       Number of bytes that were dumped into fnic_dbgfs_t
+ */
+
+int fnic_fc_trace_get_data(fnic_dbgfs_t *fnic_dbgfs_prt, u8 rdata_flag)
+{
+	int rd_idx, wr_idx;
+	unsigned long flags;
+	int len = 0, j;
+	struct fc_trace_hdr *tdata;
+	char *fc_trace;
+
+	spin_lock_irqsave(&fnic_fc_trace_lock, flags);
+	if (fc_trace_entries.wr_idx == fc_trace_entries.rd_idx) {
+		spin_unlock_irqrestore(&fnic_fc_trace_lock, flags);
+		pr_info("fnic: Buffer is empty\n");
+		return 0;
+	}
+	rd_idx = fc_trace_entries.rd_idx;
+	wr_idx = fc_trace_entries.wr_idx;
+	if (rdata_flag == 0) {
+		len += snprintf(fnic_dbgfs_prt->buffer + len,
+			(fnic_fc_trace_max_pages * PAGE_SIZE * 3) - len,
+			"Time Stamp (UTC)\t\t"
+			"Host No:   F Type:  len:     FCoE_FRAME:\n");
+	}
+
+	while (rd_idx != wr_idx) {
+		tdata = (struct fc_trace_hdr *)
+			fc_trace_entries.page_offset[rd_idx];
+		if (!tdata) {
+			pr_info("fnic: Rd data is NULL\n");
+			spin_unlock_irqrestore(&fnic_fc_trace_lock, flags);
+			return 0;
+		}
+		if (rdata_flag == 0) {
+			copy_and_format_trace_data(tdata,
+				fnic_dbgfs_prt, &len, rdata_flag);
+		} else {
+			fc_trace = (char *)tdata;
+			for (j = 0; j < FC_TRC_SIZE_BYTES; j++) {
+				len += snprintf(fnic_dbgfs_prt->buffer + len,
+				(fnic_fc_trace_max_pages * PAGE_SIZE * 3)
+				- len, "%02x", fc_trace[j] & 0xff);
+			} /* for loop */
+			len += snprintf(fnic_dbgfs_prt->buffer + len,
+				(fnic_fc_trace_max_pages * PAGE_SIZE * 3) - len,
+				"\n");
+		}
+		rd_idx++;
+		if (rd_idx > (fc_trace_max_entries - 1))
+			rd_idx = 0;
+	}
+
+	spin_unlock_irqrestore(&fnic_fc_trace_lock, flags);
+	return len;
+}
+
+/*
+ * copy_and_format_trace_data: Copy formatted data to char * buffer
+ * Passed Parameter:
+ *      @fc_trace_hdr_t: pointer to trace data
+ *      @fnic_dbgfs_t: pointer to debugfs trace buffer
+ *      @orig_len: pointer to len
+ *      rdata_flag: 0 => Formated file, 1 => Unformated file
+ * Description:
+ *      This routine will format and copy the passed trace data
+ *      for formated file or unformated file accordingly.
+ */
+
+void copy_and_format_trace_data(struct fc_trace_hdr *tdata,
+				fnic_dbgfs_t *fnic_dbgfs_prt, int *orig_len,
+				u8 rdata_flag)
+{
+	struct tm tm;
+	int j, i = 1, len;
+	char *fc_trace, *fmt;
+	int ethhdr_len = sizeof(struct ethhdr) - 1;
+	int fcoehdr_len = sizeof(struct fcoe_hdr);
+	int fchdr_len = sizeof(struct fc_frame_header);
+	int max_size = fnic_fc_trace_max_pages * PAGE_SIZE * 3;
+
+	tdata->frame_type = tdata->frame_type & 0x7F;
+
+	len = *orig_len;
+
+	time_to_tm(tdata->time_stamp.tv_sec, 0, &tm);
+
+	fmt = "%02d:%02d:%04ld %02d:%02d:%02d.%09lu ns%8x       %c%8x\t";
+	len += snprintf(fnic_dbgfs_prt->buffer + len,
+		(fnic_fc_trace_max_pages * PAGE_SIZE * 3) - len,
+		fmt,
+		tm.tm_mon + 1, tm.tm_mday, tm.tm_year + 1900,
+		tm.tm_hour, tm.tm_min, tm.tm_sec,
+		tdata->time_stamp.tv_nsec, tdata->host_no,
+		tdata->frame_type, tdata->frame_len);
+
+	fc_trace = (char *)FC_TRACE_ADDRESS(tdata);
+
+	for (j = 0; j < min_t(u8, tdata->frame_len,
+		(u8)(FC_TRC_SIZE_BYTES - FC_TRC_HEADER_SIZE)); j++) {
+		if (tdata->frame_type == FNIC_FC_LE) {
+			len += snprintf(fnic_dbgfs_prt->buffer + len,
+				max_size - len, "%c", fc_trace[j]);
+		} else {
+			len += snprintf(fnic_dbgfs_prt->buffer + len,
+				max_size - len, "%02x", fc_trace[j] & 0xff);
+			len += snprintf(fnic_dbgfs_prt->buffer + len,
+				max_size - len, " ");
+			if (j == ethhdr_len ||
+				j == ethhdr_len + fcoehdr_len ||
+				j == ethhdr_len + fcoehdr_len + fchdr_len ||
+				(i > 3 && j%fchdr_len == 0)) {
+				len += snprintf(fnic_dbgfs_prt->buffer
+					+ len, (fnic_fc_trace_max_pages
+					* PAGE_SIZE * 3) - len,
+					"\n\t\t\t\t\t\t\t\t");
+				i++;
+			}
+		} /* end of else*/
+	} /* End of for loop*/
+	len += snprintf(fnic_dbgfs_prt->buffer + len,
+		max_size - len, "\n");
+	*orig_len = len;
+}
diff --git a/drivers/scsi/fnic/fnic_trace.h b/drivers/scsi/fnic/fnic_trace.h
index d412f2ee3c4f..a8aa0578fcb0 100644
--- a/drivers/scsi/fnic/fnic_trace.h
+++ b/drivers/scsi/fnic/fnic_trace.h
@@ -19,6 +19,17 @@
 #define __FNIC_TRACE_H__
 
 #define FNIC_ENTRY_SIZE_BYTES 64
+#define FC_TRC_SIZE_BYTES 256
+#define FC_TRC_HEADER_SIZE sizeof(struct fc_trace_hdr)
+
+/*
+ * Fisrt bit of FNIC_FC_RECV and FNIC_FC_SEND is used to represent the type
+ * of frame 1 => Eth frame, 0=> FC frame
+ */
+
+#define FNIC_FC_RECV 0x52 /* Character R */
+#define FNIC_FC_SEND 0x54 /* Character T */
+#define FNIC_FC_LE 0x4C /* Character L */
 
 extern ssize_t simple_read_from_buffer(void __user *to,
 					  size_t count,
@@ -30,6 +41,10 @@ extern unsigned int fnic_trace_max_pages;
 extern int fnic_tracing_enabled;
 extern unsigned int trace_max_pages;
 
+extern unsigned int fnic_fc_trace_max_pages;
+extern int fnic_fc_tracing_enabled;
+extern int fnic_fc_trace_cleared;
+
 typedef struct fnic_trace_dbg {
 	int wr_idx;
 	int rd_idx;
@@ -56,6 +71,16 @@ struct fnic_trace_data {
 
 typedef struct fnic_trace_data fnic_trace_data_t;
 
+struct fc_trace_hdr {
+	struct timespec time_stamp;
+	u32 host_no;
+	u8 frame_type;
+	u8 frame_len;
+} __attribute__((__packed__));
+
+#define FC_TRACE_ADDRESS(a) \
+	((unsigned long)(a) + sizeof(struct fc_trace_hdr))
+
 #define FNIC_TRACE_ENTRY_SIZE \
 		  (FNIC_ENTRY_SIZE_BYTES - sizeof(fnic_trace_data_t))
 
@@ -88,4 +113,17 @@ int fnic_debugfs_init(void);
 void fnic_debugfs_terminate(void);
 int fnic_trace_debugfs_init(void);
 void fnic_trace_debugfs_terminate(void);
+
+/* Fnic FC CTLR Trace releated function */
+int fnic_fc_trace_init(void);
+void fnic_fc_trace_free(void);
+int fnic_fc_trace_set_data(u32 host_no, u8 frame_type,
+				char *frame, u32 fc_frame_len);
+int fnic_fc_trace_get_data(fnic_dbgfs_t *fnic_dbgfs_prt, u8 rdata_flag);
+void copy_and_format_trace_data(struct fc_trace_hdr *tdata,
+				fnic_dbgfs_t *fnic_dbgfs_prt,
+				int *len, u8 rdata_flag);
+int fnic_fc_trace_debugfs_init(void);
+void fnic_fc_trace_debugfs_terminate(void);
+
 #endif
diff --git a/drivers/scsi/g_NCR5380.c b/drivers/scsi/g_NCR5380.c
index 7176365e916b..a1bc8ca958e1 100644
--- a/drivers/scsi/g_NCR5380.c
+++ b/drivers/scsi/g_NCR5380.c
@@ -78,10 +78,6 @@
  *     
  */
 
-/*
- * $Log: generic_NCR5380.c,v $
- */
-
 /* settings for DTC3181E card with only Mustek scanner attached */
 #define USLEEP
 #define USLEEP_POLL	1
diff --git a/drivers/scsi/g_NCR5380.h b/drivers/scsi/g_NCR5380.h
index 1bcdb7beb77b..703adf78e0b2 100644
--- a/drivers/scsi/g_NCR5380.h
+++ b/drivers/scsi/g_NCR5380.h
@@ -25,10 +25,6 @@
  * 1+ (800) 334-5454
  */
 
-/*
- * $Log: generic_NCR5380.h,v $
- */
-
 #ifndef GENERIC_NCR5380_H
 #define GENERIC_NCR5380_H
 
@@ -58,8 +54,6 @@ static const char* generic_NCR5380_info(struct Scsi_Host *);
 #define CAN_QUEUE 16
 #endif
 
-#ifndef HOSTS_C
-
 #define __STRVAL(x) #x
 #define STRVAL(x) __STRVAL(x)
 
@@ -131,7 +125,6 @@ static const char* generic_NCR5380_info(struct Scsi_Host *);
 #define BOARD_NCR53C400A 2
 #define BOARD_DTC3181E	3
 
-#endif /* else def HOSTS_C */
 #endif /* ndef ASM */
 #endif /* GENERIC_NCR5380_H */
 
diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 9a6e4a2cd072..31184b35370f 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -48,6 +48,7 @@
 #include <linux/bitmap.h>
 #include <linux/atomic.h>
 #include <linux/jiffies.h>
+#include <linux/percpu.h>
 #include <asm/div64.h>
 #include "hpsa_cmd.h"
 #include "hpsa.h"
@@ -115,9 +116,15 @@ static const struct pci_device_id hpsa_pci_device_id[] = {
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C3},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C4},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C5},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C6},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C7},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C8},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21C9},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21CA},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21CB},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21CC},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21CD},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSI,     0x103C, 0x21CE},
 	{PCI_VENDOR_ID_HP_3PAR, 0x0075, 0x1590, 0x0076},
 	{PCI_VENDOR_ID_HP_3PAR, 0x0075, 0x1590, 0x0087},
 	{PCI_VENDOR_ID_HP_3PAR, 0x0075, 0x1590, 0x007D},
@@ -165,9 +172,15 @@ static struct board_type products[] = {
 	{0x21C3103C, "Smart Array", &SA5_access},
 	{0x21C4103C, "Smart Array", &SA5_access},
 	{0x21C5103C, "Smart Array", &SA5_access},
+	{0x21C6103C, "Smart Array", &SA5_access},
 	{0x21C7103C, "Smart Array", &SA5_access},
 	{0x21C8103C, "Smart Array", &SA5_access},
 	{0x21C9103C, "Smart Array", &SA5_access},
+	{0x21CA103C, "Smart Array", &SA5_access},
+	{0x21CB103C, "Smart Array", &SA5_access},
+	{0x21CC103C, "Smart Array", &SA5_access},
+	{0x21CD103C, "Smart Array", &SA5_access},
+	{0x21CE103C, "Smart Array", &SA5_access},
 	{0x00761590, "HP Storage P1224 Array Controller", &SA5_access},
 	{0x00871590, "HP Storage P1224e Array Controller", &SA5_access},
 	{0x007D1590, "HP Storage P1228 Array Controller", &SA5_access},
@@ -181,7 +194,8 @@ static int number_of_controllers;
 static irqreturn_t do_hpsa_intr_intx(int irq, void *dev_id);
 static irqreturn_t do_hpsa_intr_msi(int irq, void *dev_id);
 static int hpsa_ioctl(struct scsi_device *dev, int cmd, void *arg);
-static void start_io(struct ctlr_info *h);
+static void lock_and_start_io(struct ctlr_info *h);
+static void start_io(struct ctlr_info *h, unsigned long *flags);
 
 #ifdef CONFIG_COMPAT
 static int hpsa_compat_ioctl(struct scsi_device *dev, int cmd, void *arg);
@@ -683,7 +697,7 @@ static inline void addQ(struct list_head *list, struct CommandList *c)
 static inline u32 next_command(struct ctlr_info *h, u8 q)
 {
 	u32 a;
-	struct reply_pool *rq = &h->reply_queue[q];
+	struct reply_queue_buffer *rq = &h->reply_queue[q];
 	unsigned long flags;
 
 	if (h->transMethod & CFGTBL_Trans_io_accel1)
@@ -832,8 +846,8 @@ static void enqueue_cmd_and_start_io(struct ctlr_info *h,
 	spin_lock_irqsave(&h->lock, flags);
 	addQ(&h->reqQ, c);
 	h->Qdepth++;
+	start_io(h, &flags);
 	spin_unlock_irqrestore(&h->lock, flags);
-	start_io(h);
 }
 
 static inline void removeQ(struct CommandList *c)
@@ -1542,9 +1556,13 @@ static int handle_ioaccel_mode2_error(struct ctlr_info *h,
 			dev_warn(&h->pdev->dev,
 				"%s: task complete with check condition.\n",
 				"HP SSD Smart Path");
+			cmd->result |= SAM_STAT_CHECK_CONDITION;
 			if (c2->error_data.data_present !=
-					IOACCEL2_SENSE_DATA_PRESENT)
+					IOACCEL2_SENSE_DATA_PRESENT) {
+				memset(cmd->sense_buffer, 0,
+					SCSI_SENSE_BUFFERSIZE);
 				break;
+			}
 			/* copy the sense data */
 			data_len = c2->error_data.sense_data_len;
 			if (data_len > SCSI_SENSE_BUFFERSIZE)
@@ -1554,7 +1572,6 @@ static int handle_ioaccel_mode2_error(struct ctlr_info *h,
 					sizeof(c2->error_data.sense_data_buff);
 			memcpy(cmd->sense_buffer,
 				c2->error_data.sense_data_buff, data_len);
-			cmd->result |= SAM_STAT_CHECK_CONDITION;
 			retry = 1;
 			break;
 		case IOACCEL2_STATUS_SR_TASK_COMP_BUSY:
@@ -1639,16 +1656,6 @@ static void process_ioaccel2_completion(struct ctlr_info *h,
 	if (is_logical_dev_addr_mode(dev->scsi3addr) &&
 		c2->error_data.serv_response ==
 			IOACCEL2_SERV_RESPONSE_FAILURE) {
-		if (c2->error_data.status ==
-			IOACCEL2_STATUS_SR_IOACCEL_DISABLED)
-			dev_warn(&h->pdev->dev,
-				"%s: Path is unavailable, retrying on standard path.\n",
-				"HP SSD Smart Path");
-		else
-			dev_warn(&h->pdev->dev,
-				"%s: Error 0x%02x, retrying on standard path.\n",
-				"HP SSD Smart Path", c2->error_data.status);
-
 		dev->offload_enabled = 0;
 		h->drv_req_rescan = 1;	/* schedule controller for a rescan */
 		cmd->result = DID_SOFT_ERROR << 16;
@@ -1979,20 +1986,26 @@ static inline void hpsa_scsi_do_simple_cmd_core(struct ctlr_info *h,
 	wait_for_completion(&wait);
 }
 
+static u32 lockup_detected(struct ctlr_info *h)
+{
+	int cpu;
+	u32 rc, *lockup_detected;
+
+	cpu = get_cpu();
+	lockup_detected = per_cpu_ptr(h->lockup_detected, cpu);
+	rc = *lockup_detected;
+	put_cpu();
+	return rc;
+}
+
 static void hpsa_scsi_do_simple_cmd_core_if_no_lockup(struct ctlr_info *h,
 	struct CommandList *c)
 {
-	unsigned long flags;
-
 	/* If controller lockup detected, fake a hardware error. */
-	spin_lock_irqsave(&h->lock, flags);
-	if (unlikely(h->lockup_detected)) {
-		spin_unlock_irqrestore(&h->lock, flags);
+	if (unlikely(lockup_detected(h)))
 		c->err_info->CommandStatus = CMD_HARDWARE_ERR;
-	} else {
-		spin_unlock_irqrestore(&h->lock, flags);
+	else
 		hpsa_scsi_do_simple_cmd_core(h, c);
-	}
 }
 
 #define MAX_DRIVER_CMD_RETRIES 25
@@ -2417,7 +2430,7 @@ static int hpsa_get_device_id(struct ctlr_info *h, unsigned char *scsi3addr,
 		buflen = 16;
 	buf = kzalloc(64, GFP_KERNEL);
 	if (!buf)
-		return -1;
+		return -ENOMEM;
 	rc = hpsa_scsi_do_inquiry(h, scsi3addr, VPD_PAGE | 0x83, buf, 64);
 	if (rc == 0)
 		memcpy(device_id, &buf[8], buflen);
@@ -2503,27 +2516,21 @@ static int hpsa_get_volume_status(struct ctlr_info *h,
 		return HPSA_VPD_LV_STATUS_UNSUPPORTED;
 
 	/* Does controller have VPD for logical volume status? */
-	if (!hpsa_vpd_page_supported(h, scsi3addr, HPSA_VPD_LV_STATUS)) {
-		dev_warn(&h->pdev->dev, "Logical volume status VPD page is unsupported.\n");
+	if (!hpsa_vpd_page_supported(h, scsi3addr, HPSA_VPD_LV_STATUS))
 		goto exit_failed;
-	}
 
 	/* Get the size of the VPD return buffer */
 	rc = hpsa_scsi_do_inquiry(h, scsi3addr, VPD_PAGE | HPSA_VPD_LV_STATUS,
 					buf, HPSA_VPD_HEADER_SZ);
-	if (rc != 0) {
-		dev_warn(&h->pdev->dev, "Logical volume status VPD inquiry failed.\n");
+	if (rc != 0)
 		goto exit_failed;
-	}
 	size = buf[3];
 
 	/* Now get the whole VPD buffer */
 	rc = hpsa_scsi_do_inquiry(h, scsi3addr, VPD_PAGE | HPSA_VPD_LV_STATUS,
 					buf, size + HPSA_VPD_HEADER_SZ);
-	if (rc != 0) {
-		dev_warn(&h->pdev->dev, "Logical volume status VPD inquiry failed.\n");
+	if (rc != 0)
 		goto exit_failed;
-	}
 	status = buf[4]; /* status byte */
 
 	kfree(buf);
@@ -2536,11 +2543,11 @@ exit_failed:
 /* Determine offline status of a volume.
  * Return either:
  *  0 (not offline)
- * -1 (offline for unknown reasons)
+ *  0xff (offline for unknown reasons)
  *  # (integer code indicating one of several NOT READY states
  *     describing why a volume is to be kept offline)
  */
-static unsigned char hpsa_volume_offline(struct ctlr_info *h,
+static int hpsa_volume_offline(struct ctlr_info *h,
 					unsigned char scsi3addr[])
 {
 	struct CommandList *c;
@@ -2639,11 +2646,15 @@ static int hpsa_update_device_info(struct ctlr_info *h,
 
 	if (this_device->devtype == TYPE_DISK &&
 		is_logical_dev_addr_mode(scsi3addr)) {
+		int volume_offline;
+
 		hpsa_get_raid_level(h, scsi3addr, &this_device->raid_level);
 		if (h->fw_support & MISC_FW_RAID_OFFLOAD_BASIC)
 			hpsa_get_ioaccel_status(h, scsi3addr, this_device);
-		this_device->volume_offline =
-			hpsa_volume_offline(h, scsi3addr);
+		volume_offline = hpsa_volume_offline(h, scsi3addr);
+		if (volume_offline < 0 || volume_offline > 0xff)
+			volume_offline = HPSA_VPD_LV_STATUS_UNSUPPORTED;
+		this_device->volume_offline = volume_offline & 0xff;
 	} else {
 		this_device->raid_level = RAID_UNKNOWN;
 		this_device->offload_config = 0;
@@ -2836,6 +2847,8 @@ static int hpsa_get_pdisk_of_ioaccel2(struct ctlr_info *h,
 
 	/* Get the list of physical devices */
 	physicals = kzalloc(reportsize, GFP_KERNEL);
+	if (physicals == NULL)
+		return 0;
 	if (hpsa_scsi_do_report_phys_luns(h, (struct ReportLUNdata *) physicals,
 		reportsize, extended)) {
 		dev_err(&h->pdev->dev,
@@ -2847,26 +2860,20 @@ static int hpsa_get_pdisk_of_ioaccel2(struct ctlr_info *h,
 	nphysicals = be32_to_cpu(*((__be32 *)physicals->LUNListLength)) /
 							responsesize;
 
-
 	/* find ioaccel2 handle in list of physicals: */
 	for (i = 0; i < nphysicals; i++) {
+		struct ext_report_lun_entry *entry = &physicals->LUN[i];
+
 		/* handle is in bytes 28-31 of each lun */
-		if (memcmp(&((struct ReportExtendedLUNdata *)
-				physicals)->LUN[i][20], &find, 4) != 0) {
+		if (entry->ioaccel_handle != find)
 			continue; /* didn't match */
-		}
 		found = 1;
-		memcpy(scsi3addr, &((struct ReportExtendedLUNdata *)
-					physicals)->LUN[i][0], 8);
+		memcpy(scsi3addr, entry->lunid, 8);
 		if (h->raid_offload_debug > 0)
 			dev_info(&h->pdev->dev,
-				"%s: Searched h=0x%08x, Found h=0x%08x, scsiaddr 0x%02x%02x%02x%02x%02x%02x%02x%02x\n",
+				"%s: Searched h=0x%08x, Found h=0x%08x, scsiaddr 0x%8phN\n",
 				__func__, find,
-				((struct ReportExtendedLUNdata *)
-					physicals)->LUN[i][20],
-				scsi3addr[0], scsi3addr[1], scsi3addr[2],
-				scsi3addr[3], scsi3addr[4], scsi3addr[5],
-				scsi3addr[6], scsi3addr[7]);
+				entry->ioaccel_handle, scsi3addr);
 		break; /* found it */
 	}
 
@@ -2951,7 +2958,8 @@ u8 *figure_lunaddrbytes(struct ctlr_info *h, int raid_ctlr_position, int i,
 		return RAID_CTLR_LUNID;
 
 	if (i < logicals_start)
-		return &physdev_list->LUN[i - (raid_ctlr_position == 0)][0];
+		return &physdev_list->LUN[i -
+				(raid_ctlr_position == 0)].lunid[0];
 
 	if (i < last_device)
 		return &logdev_list->LUN[i - nphysicals -
@@ -2963,19 +2971,24 @@ u8 *figure_lunaddrbytes(struct ctlr_info *h, int raid_ctlr_position, int i,
 static int hpsa_hba_mode_enabled(struct ctlr_info *h)
 {
 	int rc;
+	int hba_mode_enabled;
 	struct bmic_controller_parameters *ctlr_params;
 	ctlr_params = kzalloc(sizeof(struct bmic_controller_parameters),
 		GFP_KERNEL);
 
 	if (!ctlr_params)
-		return 0;
+		return -ENOMEM;
 	rc = hpsa_bmic_ctrl_mode_sense(h, RAID_CTLR_LUNID, 0, ctlr_params,
 		sizeof(struct bmic_controller_parameters));
-	if (rc != 0) {
+	if (rc) {
 		kfree(ctlr_params);
-		return 0;
+		return rc;
 	}
-	return ctlr_params->nvram_flags & (1 << 3) ? 1 : 0;
+
+	hba_mode_enabled =
+		((ctlr_params->nvram_flags & HBA_MODE_ENABLED_FLAG) != 0);
+	kfree(ctlr_params);
+	return hba_mode_enabled;
 }
 
 static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno)
@@ -3001,7 +3014,7 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno)
 	int reportlunsize = sizeof(*physdev_list) + HPSA_MAX_PHYS_LUN * 24;
 	int i, n_ext_target_devs, ndevs_to_allocate;
 	int raid_ctlr_position;
-	u8 rescan_hba_mode;
+	int rescan_hba_mode;
 	DECLARE_BITMAP(lunzerobits, MAX_EXT_TARGETS);
 
 	currentsd = kzalloc(sizeof(*currentsd) * HPSA_MAX_DEVICES, GFP_KERNEL);
@@ -3016,6 +3029,8 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno)
 	memset(lunzerobits, 0, sizeof(lunzerobits));
 
 	rescan_hba_mode = hpsa_hba_mode_enabled(h);
+	if (rescan_hba_mode < 0)
+		goto out;
 
 	if (!h->hba_mode_enabled && rescan_hba_mode)
 		dev_warn(&h->pdev->dev, "HBA mode enabled\n");
@@ -3053,7 +3068,7 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno)
 		ndev_allocated++;
 	}
 
-	if (unlikely(is_scsi_rev_5(h)))
+	if (is_scsi_rev_5(h))
 		raid_ctlr_position = 0;
 	else
 		raid_ctlr_position = nphysicals + nlogicals;
@@ -3950,7 +3965,6 @@ static int hpsa_scsi_queue_command_lck(struct scsi_cmnd *cmd,
 	struct hpsa_scsi_dev_t *dev;
 	unsigned char scsi3addr[8];
 	struct CommandList *c;
-	unsigned long flags;
 	int rc = 0;
 
 	/* Get the ptr to our adapter structure out of cmd->host. */
@@ -3963,14 +3977,11 @@ static int hpsa_scsi_queue_command_lck(struct scsi_cmnd *cmd,
 	}
 	memcpy(scsi3addr, dev->scsi3addr, sizeof(scsi3addr));
 
-	spin_lock_irqsave(&h->lock, flags);
-	if (unlikely(h->lockup_detected)) {
-		spin_unlock_irqrestore(&h->lock, flags);
+	if (unlikely(lockup_detected(h))) {
 		cmd->result = DID_ERROR << 16;
 		done(cmd);
 		return 0;
 	}
-	spin_unlock_irqrestore(&h->lock, flags);
 	c = cmd_alloc(h);
 	if (c == NULL) {			/* trouble... */
 		dev_err(&h->pdev->dev, "cmd_alloc returned NULL!\n");
@@ -4082,16 +4093,13 @@ static int do_not_scan_if_controller_locked_up(struct ctlr_info *h)
 	 * we can prevent new rescan threads from piling up on a
 	 * locked up controller.
 	 */
-	spin_lock_irqsave(&h->lock, flags);
-	if (unlikely(h->lockup_detected)) {
-		spin_unlock_irqrestore(&h->lock, flags);
+	if (unlikely(lockup_detected(h))) {
 		spin_lock_irqsave(&h->scan_lock, flags);
 		h->scan_finished = 1;
 		wake_up_all(&h->scan_wait_queue);
 		spin_unlock_irqrestore(&h->scan_lock, flags);
 		return 1;
 	}
-	spin_unlock_irqrestore(&h->lock, flags);
 	return 0;
 }
 
@@ -4942,7 +4950,7 @@ static int hpsa_passthru_ioctl(struct ctlr_info *h, void __user *argp)
 		buff = kmalloc(iocommand.buf_size, GFP_KERNEL);
 		if (buff == NULL)
 			return -EFAULT;
-		if (iocommand.Request.Type.Direction == XFER_WRITE) {
+		if (iocommand.Request.Type.Direction & XFER_WRITE) {
 			/* Copy the data into the buffer we created */
 			if (copy_from_user(buff, iocommand.buf,
 				iocommand.buf_size)) {
@@ -5005,7 +5013,7 @@ static int hpsa_passthru_ioctl(struct ctlr_info *h, void __user *argp)
 		rc = -EFAULT;
 		goto out;
 	}
-	if (iocommand.Request.Type.Direction == XFER_READ &&
+	if ((iocommand.Request.Type.Direction & XFER_READ) &&
 		iocommand.buf_size > 0) {
 		/* Copy the data out of the buffer we created */
 		if (copy_to_user(iocommand.buf, buff, iocommand.buf_size)) {
@@ -5082,7 +5090,7 @@ static int hpsa_big_passthru_ioctl(struct ctlr_info *h, void __user *argp)
 			status = -ENOMEM;
 			goto cleanup1;
 		}
-		if (ioc->Request.Type.Direction == XFER_WRITE) {
+		if (ioc->Request.Type.Direction & XFER_WRITE) {
 			if (copy_from_user(buff[sg_used], data_ptr, sz)) {
 				status = -ENOMEM;
 				goto cleanup1;
@@ -5134,7 +5142,7 @@ static int hpsa_big_passthru_ioctl(struct ctlr_info *h, void __user *argp)
 		status = -EFAULT;
 		goto cleanup0;
 	}
-	if (ioc->Request.Type.Direction == XFER_READ && ioc->buf_size > 0) {
+	if ((ioc->Request.Type.Direction & XFER_READ) && ioc->buf_size > 0) {
 		/* Copy the data out of the buffer we created */
 		BYTE __user *ptr = ioc->buf;
 		for (i = 0; i < sg_used; i++) {
@@ -5438,13 +5446,12 @@ static void __iomem *remap_pci_mem(ulong base, ulong size)
 
 /* Takes cmds off the submission queue and sends them to the hardware,
  * then puts them on the queue of cmds waiting for completion.
+ * Assumes h->lock is held
  */
-static void start_io(struct ctlr_info *h)
+static void start_io(struct ctlr_info *h, unsigned long *flags)
 {
 	struct CommandList *c;
-	unsigned long flags;
 
-	spin_lock_irqsave(&h->lock, flags);
 	while (!list_empty(&h->reqQ)) {
 		c = list_entry(h->reqQ.next, struct CommandList, list);
 		/* can't do anything if fifo is full */
@@ -5467,14 +5474,20 @@ static void start_io(struct ctlr_info *h)
 		 * condition.
 		 */
 		h->commands_outstanding++;
-		if (h->commands_outstanding > h->max_outstanding)
-			h->max_outstanding = h->commands_outstanding;
 
 		/* Tell the controller execute command */
-		spin_unlock_irqrestore(&h->lock, flags);
+		spin_unlock_irqrestore(&h->lock, *flags);
 		h->access.submit_command(h, c);
-		spin_lock_irqsave(&h->lock, flags);
+		spin_lock_irqsave(&h->lock, *flags);
 	}
+}
+
+static void lock_and_start_io(struct ctlr_info *h)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&h->lock, flags);
+	start_io(h, &flags);
 	spin_unlock_irqrestore(&h->lock, flags);
 }
 
@@ -5542,7 +5555,7 @@ static inline void finish_cmd(struct CommandList *c)
 	else if (c->cmd_type == CMD_IOCTL_PEND)
 		complete(c->waiting);
 	if (unlikely(io_may_be_stalled))
-		start_io(h);
+		lock_and_start_io(h);
 }
 
 static inline u32 hpsa_tag_contains_index(u32 tag)
@@ -5819,12 +5832,12 @@ static int hpsa_controller_hard_reset(struct pci_dev *pdev,
 		dev_info(&pdev->dev, "using doorbell to reset controller\n");
 		writel(use_doorbell, vaddr + SA5_DOORBELL);
 
-		/* PMC hardware guys tell us we need a 5 second delay after
+		/* PMC hardware guys tell us we need a 10 second delay after
 		 * doorbell reset and before any attempt to talk to the board
 		 * at all to ensure that this actually works and doesn't fall
 		 * over in some weird corner cases.
 		 */
-		msleep(5000);
+		msleep(10000);
 	} else { /* Try to do it the PCI power state way */
 
 		/* Quoting from the Open CISS Specification: "The Power
@@ -6145,6 +6158,8 @@ static void hpsa_interrupt_mode(struct ctlr_info *h)
 	if (pci_find_capability(h->pdev, PCI_CAP_ID_MSIX)) {
 		dev_info(&h->pdev->dev, "MSIX\n");
 		h->msix_vector = MAX_REPLY_QUEUES;
+		if (h->msix_vector > num_online_cpus())
+			h->msix_vector = num_online_cpus();
 		err = pci_enable_msix(h->pdev, hpsa_msix_entries,
 				      h->msix_vector);
 		if (err > 0) {
@@ -6594,6 +6609,17 @@ static void hpsa_free_cmd_pool(struct ctlr_info *h)
 			h->ioaccel_cmd_pool, h->ioaccel_cmd_pool_dhandle);
 }
 
+static void hpsa_irq_affinity_hints(struct ctlr_info *h)
+{
+	int i, cpu, rc;
+
+	cpu = cpumask_first(cpu_online_mask);
+	for (i = 0; i < h->msix_vector; i++) {
+		rc = irq_set_affinity_hint(h->intr[i], get_cpu_mask(cpu));
+		cpu = cpumask_next(cpu, cpu_online_mask);
+	}
+}
+
 static int hpsa_request_irq(struct ctlr_info *h,
 	irqreturn_t (*msixhandler)(int, void *),
 	irqreturn_t (*intxhandler)(int, void *))
@@ -6613,6 +6639,7 @@ static int hpsa_request_irq(struct ctlr_info *h,
 			rc = request_irq(h->intr[i], msixhandler,
 					0, h->devname,
 					&h->q[i]);
+		hpsa_irq_affinity_hints(h);
 	} else {
 		/* Use single reply pool */
 		if (h->msix_vector > 0 || h->msi_vector) {
@@ -6664,12 +6691,15 @@ static void free_irqs(struct ctlr_info *h)
 	if (!h->msix_vector || h->intr_mode != PERF_MODE_INT) {
 		/* Single reply queue, only one irq to free */
 		i = h->intr_mode;
+		irq_set_affinity_hint(h->intr[i], NULL);
 		free_irq(h->intr[i], &h->q[i]);
 		return;
 	}
 
-	for (i = 0; i < h->msix_vector; i++)
+	for (i = 0; i < h->msix_vector; i++) {
+		irq_set_affinity_hint(h->intr[i], NULL);
 		free_irq(h->intr[i], &h->q[i]);
+	}
 }
 
 static void hpsa_free_irqs_and_disable_msix(struct ctlr_info *h)
@@ -6686,6 +6716,20 @@ static void hpsa_free_irqs_and_disable_msix(struct ctlr_info *h)
 #endif /* CONFIG_PCI_MSI */
 }
 
+static void hpsa_free_reply_queues(struct ctlr_info *h)
+{
+	int i;
+
+	for (i = 0; i < h->nreply_queues; i++) {
+		if (!h->reply_queue[i].head)
+			continue;
+		pci_free_consistent(h->pdev, h->reply_queue_size,
+			h->reply_queue[i].head, h->reply_queue[i].busaddr);
+		h->reply_queue[i].head = NULL;
+		h->reply_queue[i].busaddr = 0;
+	}
+}
+
 static void hpsa_undo_allocations_after_kdump_soft_reset(struct ctlr_info *h)
 {
 	hpsa_free_irqs_and_disable_msix(h);
@@ -6693,8 +6737,7 @@ static void hpsa_undo_allocations_after_kdump_soft_reset(struct ctlr_info *h)
 	hpsa_free_cmd_pool(h);
 	kfree(h->ioaccel1_blockFetchTable);
 	kfree(h->blockFetchTable);
-	pci_free_consistent(h->pdev, h->reply_pool_size,
-		h->reply_pool, h->reply_pool_dhandle);
+	hpsa_free_reply_queues(h);
 	if (h->vaddr)
 		iounmap(h->vaddr);
 	if (h->transtable)
@@ -6719,16 +6762,38 @@ static void fail_all_cmds_on_list(struct ctlr_info *h, struct list_head *list)
 	}
 }
 
+static void set_lockup_detected_for_all_cpus(struct ctlr_info *h, u32 value)
+{
+	int i, cpu;
+
+	cpu = cpumask_first(cpu_online_mask);
+	for (i = 0; i < num_online_cpus(); i++) {
+		u32 *lockup_detected;
+		lockup_detected = per_cpu_ptr(h->lockup_detected, cpu);
+		*lockup_detected = value;
+		cpu = cpumask_next(cpu, cpu_online_mask);
+	}
+	wmb(); /* be sure the per-cpu variables are out to memory */
+}
+
 static void controller_lockup_detected(struct ctlr_info *h)
 {
 	unsigned long flags;
+	u32 lockup_detected;
 
 	h->access.set_intr_mask(h, HPSA_INTR_OFF);
 	spin_lock_irqsave(&h->lock, flags);
-	h->lockup_detected = readl(h->vaddr + SA5_SCRATCHPAD_OFFSET);
+	lockup_detected = readl(h->vaddr + SA5_SCRATCHPAD_OFFSET);
+	if (!lockup_detected) {
+		/* no heartbeat, but controller gave us a zero. */
+		dev_warn(&h->pdev->dev,
+			"lockup detected but scratchpad register is zero\n");
+		lockup_detected = 0xffffffff;
+	}
+	set_lockup_detected_for_all_cpus(h, lockup_detected);
 	spin_unlock_irqrestore(&h->lock, flags);
 	dev_warn(&h->pdev->dev, "Controller lockup detected: 0x%08x\n",
-			h->lockup_detected);
+			lockup_detected);
 	pci_disable_device(h->pdev);
 	spin_lock_irqsave(&h->lock, flags);
 	fail_all_cmds_on_list(h, &h->cmpQ);
@@ -6863,7 +6928,7 @@ static void hpsa_monitor_ctlr_worker(struct work_struct *work)
 	struct ctlr_info *h = container_of(to_delayed_work(work),
 					struct ctlr_info, monitor_ctlr_work);
 	detect_controller_lockup(h);
-	if (h->lockup_detected)
+	if (lockup_detected(h))
 		return;
 
 	if (hpsa_ctlr_needs_rescan(h) || hpsa_offline_devices_ready(h)) {
@@ -6913,7 +6978,6 @@ reinit_after_soft_reset:
 	 * the 5 lower bits of the address are used by the hardware. and by
 	 * the driver.  See comments in hpsa.h for more info.
 	 */
-#define COMMANDLIST_ALIGNMENT 128
 	BUILD_BUG_ON(sizeof(struct CommandList) % COMMANDLIST_ALIGNMENT);
 	h = kzalloc(sizeof(*h), GFP_KERNEL);
 	if (!h)
@@ -6928,6 +6992,13 @@ reinit_after_soft_reset:
 	spin_lock_init(&h->offline_device_lock);
 	spin_lock_init(&h->scan_lock);
 	spin_lock_init(&h->passthru_count_lock);
+
+	/* Allocate and clear per-cpu variable lockup_detected */
+	h->lockup_detected = alloc_percpu(u32);
+	if (!h->lockup_detected)
+		goto clean1;
+	set_lockup_detected_for_all_cpus(h, 0);
+
 	rc = hpsa_pci_init(h);
 	if (rc != 0)
 		goto clean1;
@@ -7051,6 +7122,8 @@ clean4:
 	free_irqs(h);
 clean2:
 clean1:
+	if (h->lockup_detected)
+		free_percpu(h->lockup_detected);
 	kfree(h);
 	return rc;
 }
@@ -7059,16 +7132,10 @@ static void hpsa_flush_cache(struct ctlr_info *h)
 {
 	char *flush_buf;
 	struct CommandList *c;
-	unsigned long flags;
 
 	/* Don't bother trying to flush the cache if locked up */
-	spin_lock_irqsave(&h->lock, flags);
-	if (unlikely(h->lockup_detected)) {
-		spin_unlock_irqrestore(&h->lock, flags);
+	if (unlikely(lockup_detected(h)))
 		return;
-	}
-	spin_unlock_irqrestore(&h->lock, flags);
-
 	flush_buf = kzalloc(4, GFP_KERNEL);
 	if (!flush_buf)
 		return;
@@ -7144,8 +7211,7 @@ static void hpsa_remove_one(struct pci_dev *pdev)
 	pci_free_consistent(h->pdev,
 		h->nr_cmds * sizeof(struct ErrorInfo),
 		h->errinfo_pool, h->errinfo_pool_dhandle);
-	pci_free_consistent(h->pdev, h->reply_pool_size,
-		h->reply_pool, h->reply_pool_dhandle);
+	hpsa_free_reply_queues(h);
 	kfree(h->cmd_pool_bits);
 	kfree(h->blockFetchTable);
 	kfree(h->ioaccel1_blockFetchTable);
@@ -7153,6 +7219,7 @@ static void hpsa_remove_one(struct pci_dev *pdev)
 	kfree(h->hba_inquiry_data);
 	pci_disable_device(pdev);
 	pci_release_regions(pdev);
+	free_percpu(h->lockup_detected);
 	kfree(h);
 }
 
@@ -7257,8 +7324,16 @@ static void hpsa_enter_performant_mode(struct ctlr_info *h, u32 trans_support)
 	 * 10 = 6 s/g entry or 24k
 	 */
 
+	/* If the controller supports either ioaccel method then
+	 * we can also use the RAID stack submit path that does not
+	 * perform the superfluous readl() after each command submission.
+	 */
+	if (trans_support & (CFGTBL_Trans_io_accel1 | CFGTBL_Trans_io_accel2))
+		access = SA5_performant_access_no_read;
+
 	/* Controller spec: zero out this buffer. */
-	memset(h->reply_pool, 0, h->reply_pool_size);
+	for (i = 0; i < h->nreply_queues; i++)
+		memset(h->reply_queue[i].head, 0, h->reply_queue_size);
 
 	bft[7] = SG_ENTRIES_IN_CMD + 4;
 	calc_bucket_map(bft, ARRAY_SIZE(bft),
@@ -7274,8 +7349,7 @@ static void hpsa_enter_performant_mode(struct ctlr_info *h, u32 trans_support)
 
 	for (i = 0; i < h->nreply_queues; i++) {
 		writel(0, &h->transtable->RepQAddr[i].upper);
-		writel(h->reply_pool_dhandle +
-			(h->max_commands * sizeof(u64) * i),
+		writel(h->reply_queue[i].busaddr,
 			&h->transtable->RepQAddr[i].lower);
 	}
 
@@ -7323,8 +7397,10 @@ static void hpsa_enter_performant_mode(struct ctlr_info *h, u32 trans_support)
 				h->ioaccel1_blockFetchTable);
 
 		/* initialize all reply queue entries to unused */
-		memset(h->reply_pool, (u8) IOACCEL_MODE1_REPLY_UNUSED,
-				h->reply_pool_size);
+		for (i = 0; i < h->nreply_queues; i++)
+			memset(h->reply_queue[i].head,
+				(u8) IOACCEL_MODE1_REPLY_UNUSED,
+				h->reply_queue_size);
 
 		/* set all the constant fields in the accelerator command
 		 * frames once at init time to save CPU cycles later.
@@ -7386,7 +7462,6 @@ static int hpsa_alloc_ioaccel_cmd_and_bft(struct ctlr_info *h)
 	 * because the 7 lower bits of the address are used by the
 	 * hardware.
 	 */
-#define IOACCEL1_COMMANDLIST_ALIGNMENT 128
 	BUILD_BUG_ON(sizeof(struct io_accel1_cmd) %
 			IOACCEL1_COMMANDLIST_ALIGNMENT);
 	h->ioaccel_cmd_pool =
@@ -7424,7 +7499,6 @@ static int ioaccel2_alloc_cmds_and_bft(struct ctlr_info *h)
 	if (h->ioaccel_maxsg > IOACCEL2_MAXSGENTRIES)
 		h->ioaccel_maxsg = IOACCEL2_MAXSGENTRIES;
 
-#define IOACCEL2_COMMANDLIST_ALIGNMENT 128
 	BUILD_BUG_ON(sizeof(struct io_accel2_cmd) %
 			IOACCEL2_COMMANDLIST_ALIGNMENT);
 	h->ioaccel2_cmd_pool =
@@ -7482,16 +7556,17 @@ static void hpsa_put_ctlr_into_performant_mode(struct ctlr_info *h)
 		}
 	}
 
-	/* TODO, check that this next line h->nreply_queues is correct */
 	h->nreply_queues = h->msix_vector > 0 ? h->msix_vector : 1;
 	hpsa_get_max_perf_mode_cmds(h);
 	/* Performant mode ring buffer and supporting data structures */
-	h->reply_pool_size = h->max_commands * sizeof(u64) * h->nreply_queues;
-	h->reply_pool = pci_alloc_consistent(h->pdev, h->reply_pool_size,
-				&(h->reply_pool_dhandle));
+	h->reply_queue_size = h->max_commands * sizeof(u64);
 
 	for (i = 0; i < h->nreply_queues; i++) {
-		h->reply_queue[i].head = &h->reply_pool[h->max_commands * i];
+		h->reply_queue[i].head = pci_alloc_consistent(h->pdev,
+						h->reply_queue_size,
+						&(h->reply_queue[i].busaddr));
+		if (!h->reply_queue[i].head)
+			goto clean_up;
 		h->reply_queue[i].size = h->max_commands;
 		h->reply_queue[i].wraparound = 1;  /* spec: init to 1 */
 		h->reply_queue[i].current_entry = 0;
@@ -7500,18 +7575,14 @@ static void hpsa_put_ctlr_into_performant_mode(struct ctlr_info *h)
 	/* Need a block fetch table for performant mode */
 	h->blockFetchTable = kmalloc(((SG_ENTRIES_IN_CMD + 1) *
 				sizeof(u32)), GFP_KERNEL);
-
-	if ((h->reply_pool == NULL)
-		|| (h->blockFetchTable == NULL))
+	if (!h->blockFetchTable)
 		goto clean_up;
 
 	hpsa_enter_performant_mode(h, trans_support);
 	return;
 
 clean_up:
-	if (h->reply_pool)
-		pci_free_consistent(h->pdev, h->reply_pool_size,
-			h->reply_pool, h->reply_pool_dhandle);
+	hpsa_free_reply_queues(h);
 	kfree(h->blockFetchTable);
 }
 
diff --git a/drivers/scsi/hpsa.h b/drivers/scsi/hpsa.h
index 44235a27e1b6..24472cec7de3 100644
--- a/drivers/scsi/hpsa.h
+++ b/drivers/scsi/hpsa.h
@@ -57,11 +57,12 @@ struct hpsa_scsi_dev_t {
 
 };
 
-struct reply_pool {
+struct reply_queue_buffer {
 	u64 *head;
 	size_t size;
 	u8 wraparound;
 	u32 current_entry;
+	dma_addr_t busaddr;
 };
 
 #pragma pack(1)
@@ -90,6 +91,7 @@ struct bmic_controller_parameters {
 	u8   automatic_drive_slamming;
 	u8   reserved1;
 	u8   nvram_flags;
+#define HBA_MODE_ENABLED_FLAG (1 << 3)
 	u8   cache_nvram_flags;
 	u8   drive_config_flags;
 	u16  reserved2;
@@ -115,11 +117,8 @@ struct ctlr_info {
 	int 	nr_cmds; /* Number of commands allowed on this controller */
 	struct CfgTable __iomem *cfgtable;
 	int	interrupts_enabled;
-	int	major;
 	int 	max_commands;
 	int	commands_outstanding;
-	int 	max_outstanding; /* Debug */
-	int	usage_count;  /* number of opens all all minor devices */
 #	define PERF_MODE_INT	0
 #	define DOORBELL_INT	1
 #	define SIMPLE_MODE_INT	2
@@ -176,11 +175,9 @@ struct ctlr_info {
 	/*
 	 * Performant mode completion buffers
 	 */
-	u64 *reply_pool;
-	size_t reply_pool_size;
-	struct reply_pool reply_queue[MAX_REPLY_QUEUES];
+	size_t reply_queue_size;
+	struct reply_queue_buffer reply_queue[MAX_REPLY_QUEUES];
 	u8 nreply_queues;
-	dma_addr_t reply_pool_dhandle;
 	u32 *blockFetchTable;
 	u32 *ioaccel1_blockFetchTable;
 	u32 *ioaccel2_blockFetchTable;
@@ -195,7 +192,7 @@ struct ctlr_info {
 	u64 last_heartbeat_timestamp;
 	u32 heartbeat_sample_interval;
 	atomic_t firmware_flash_in_progress;
-	u32 lockup_detected;
+	u32 *lockup_detected;
 	struct delayed_work monitor_ctlr_work;
 	int remove_in_progress;
 	u32 fifo_recently_full;
@@ -232,11 +229,9 @@ struct ctlr_info {
 #define CTLR_STATE_CHANGE_EVENT_AIO_CONFIG_CHANGE	(1 << 31)
 
 #define RESCAN_REQUIRED_EVENT_BITS \
-		(CTLR_STATE_CHANGE_EVENT | \
-		CTLR_ENCLOSURE_HOT_PLUG_EVENT | \
+		(CTLR_ENCLOSURE_HOT_PLUG_EVENT | \
 		CTLR_STATE_CHANGE_EVENT_PHYSICAL_DRV | \
 		CTLR_STATE_CHANGE_EVENT_LOGICAL_DRV | \
-		CTLR_STATE_CHANGE_EVENT_REDUNDANT_CNTRL | \
 		CTLR_STATE_CHANGE_EVENT_AIO_ENABLED_DISABLED | \
 		CTLR_STATE_CHANGE_EVENT_AIO_CONFIG_CHANGE)
 	spinlock_t offline_device_lock;
@@ -345,22 +340,23 @@ struct offline_device_entry {
 static void SA5_submit_command(struct ctlr_info *h,
 	struct CommandList *c)
 {
-	dev_dbg(&h->pdev->dev, "Sending %x, tag = %x\n", c->busaddr,
-		c->Header.Tag.lower);
 	writel(c->busaddr, h->vaddr + SA5_REQUEST_PORT_OFFSET);
 	(void) readl(h->vaddr + SA5_SCRATCHPAD_OFFSET);
 }
 
+static void SA5_submit_command_no_read(struct ctlr_info *h,
+	struct CommandList *c)
+{
+	writel(c->busaddr, h->vaddr + SA5_REQUEST_PORT_OFFSET);
+}
+
 static void SA5_submit_command_ioaccel2(struct ctlr_info *h,
 	struct CommandList *c)
 {
-	dev_dbg(&h->pdev->dev, "Sending %x, tag = %x\n", c->busaddr,
-		c->Header.Tag.lower);
 	if (c->cmd_type == CMD_IOACCEL2)
 		writel(c->busaddr, h->vaddr + IOACCEL2_INBOUND_POSTQ_32);
 	else
 		writel(c->busaddr, h->vaddr + SA5_REQUEST_PORT_OFFSET);
-	(void) readl(h->vaddr + SA5_SCRATCHPAD_OFFSET);
 }
 
 /*
@@ -398,7 +394,7 @@ static void SA5_performant_intr_mask(struct ctlr_info *h, unsigned long val)
 
 static unsigned long SA5_performant_completed(struct ctlr_info *h, u8 q)
 {
-	struct reply_pool *rq = &h->reply_queue[q];
+	struct reply_queue_buffer *rq = &h->reply_queue[q];
 	unsigned long flags, register_value = FIFO_EMPTY;
 
 	/* msi auto clears the interrupt pending bit. */
@@ -477,7 +473,6 @@ static bool SA5_intr_pending(struct ctlr_info *h)
 {
 	unsigned long register_value  =
 		readl(h->vaddr + SA5_INTR_STATUS);
-	dev_dbg(&h->pdev->dev, "intr_pending %lx\n", register_value);
 	return register_value & SA5_INTR_PENDING;
 }
 
@@ -514,7 +509,7 @@ static bool SA5_ioaccel_mode1_intr_pending(struct ctlr_info *h)
 static unsigned long SA5_ioaccel_mode1_completed(struct ctlr_info *h, u8 q)
 {
 	u64 register_value;
-	struct reply_pool *rq = &h->reply_queue[q];
+	struct reply_queue_buffer *rq = &h->reply_queue[q];
 	unsigned long flags;
 
 	BUG_ON(q >= h->nreply_queues);
@@ -572,6 +567,14 @@ static struct access_method SA5_performant_access = {
 	SA5_performant_completed,
 };
 
+static struct access_method SA5_performant_access_no_read = {
+	SA5_submit_command_no_read,
+	SA5_performant_intr_mask,
+	SA5_fifo_full,
+	SA5_performant_intr_pending,
+	SA5_performant_completed,
+};
+
 struct board_type {
 	u32	board_id;
 	char	*product_name;
diff --git a/drivers/scsi/hpsa_cmd.h b/drivers/scsi/hpsa_cmd.h
index b5cc7052339f..b5125dc31439 100644
--- a/drivers/scsi/hpsa_cmd.h
+++ b/drivers/scsi/hpsa_cmd.h
@@ -151,7 +151,7 @@
 #define HPSA_VPD_HEADER_SZ              4
 
 /* Logical volume states */
-#define HPSA_VPD_LV_STATUS_UNSUPPORTED			-1
+#define HPSA_VPD_LV_STATUS_UNSUPPORTED			0xff
 #define HPSA_LV_OK                                      0x0
 #define HPSA_LV_UNDERGOING_ERASE			0x0F
 #define HPSA_LV_UNDERGOING_RPI				0x12
@@ -238,11 +238,21 @@ struct ReportLUNdata {
 	u8 LUN[HPSA_MAX_LUN][8];
 };
 
+struct ext_report_lun_entry {
+	u8 lunid[8];
+	u8 wwid[8];
+	u8 device_type;
+	u8 device_flags;
+	u8 lun_count; /* multi-lun device, how many luns */
+	u8 redundant_paths;
+	u32 ioaccel_handle; /* ioaccel1 only uses lower 16 bits */
+};
+
 struct ReportExtendedLUNdata {
 	u8 LUNListLength[4];
 	u8 extended_response_flag;
 	u8 reserved[3];
-	u8 LUN[HPSA_MAX_LUN][24];
+	struct ext_report_lun_entry LUN[HPSA_MAX_LUN];
 };
 
 struct SenseSubsystem_info {
@@ -375,6 +385,7 @@ struct ctlr_info; /* defined in hpsa.h */
  *        or a bus address.
  */
 
+#define COMMANDLIST_ALIGNMENT 128
 struct CommandList {
 	struct CommandListHeader Header;
 	struct RequestBlock      Request;
@@ -389,21 +400,7 @@ struct CommandList {
 	struct list_head list;
 	struct completion *waiting;
 	void   *scsi_cmd;
-
-/* on 64 bit architectures, to get this to be 32-byte-aligned
- * it so happens we need PAD_64 bytes of padding, on 32 bit systems,
- * we need PAD_32 bytes of padding (see below).   This does that.
- * If it happens that 64 bit and 32 bit systems need different
- * padding, PAD_32 and PAD_64 can be set independently, and.
- * the code below will do the right thing.
- */
-#define IS_32_BIT ((8 - sizeof(long))/4)
-#define IS_64_BIT (!IS_32_BIT)
-#define PAD_32 (40)
-#define PAD_64 (12)
-#define COMMANDLIST_PAD (IS_32_BIT * PAD_32 + IS_64_BIT * PAD_64)
-	u8 pad[COMMANDLIST_PAD];
-};
+} __aligned(COMMANDLIST_ALIGNMENT);
 
 /* Max S/G elements in I/O accelerator command */
 #define IOACCEL1_MAXSGENTRIES           24
@@ -413,6 +410,7 @@ struct CommandList {
  * Structure for I/O accelerator (mode 1) commands.
  * Note that this structure must be 128-byte aligned in size.
  */
+#define IOACCEL1_COMMANDLIST_ALIGNMENT 128
 struct io_accel1_cmd {
 	u16 dev_handle;			/* 0x00 - 0x01 */
 	u8  reserved1;			/* 0x02 */
@@ -440,12 +438,7 @@ struct io_accel1_cmd {
 	struct vals32 host_addr;	/* 0x70 - 0x77 */
 	u8  CISS_LUN[8];		/* 0x78 - 0x7F */
 	struct SGDescriptor SG[IOACCEL1_MAXSGENTRIES];
-#define IOACCEL1_PAD_64 0
-#define IOACCEL1_PAD_32 0
-#define IOACCEL1_PAD (IS_32_BIT * IOACCEL1_PAD_32 + \
-			IS_64_BIT * IOACCEL1_PAD_64)
-	u8 pad[IOACCEL1_PAD];
-};
+} __aligned(IOACCEL1_COMMANDLIST_ALIGNMENT);
 
 #define IOACCEL1_FUNCTION_SCSIIO        0x00
 #define IOACCEL1_SGLOFFSET              32
@@ -510,14 +503,11 @@ struct io_accel2_scsi_response {
 	u8 sense_data_buff[32];		/* sense/response data buffer */
 };
 
-#define IOACCEL2_64_PAD 76
-#define IOACCEL2_32_PAD 76
-#define IOACCEL2_PAD (IS_32_BIT * IOACCEL2_32_PAD + \
-			IS_64_BIT * IOACCEL2_64_PAD)
 /*
  * Structure for I/O accelerator (mode 2 or m2) commands.
  * Note that this structure must be 128-byte aligned in size.
  */
+#define IOACCEL2_COMMANDLIST_ALIGNMENT 128
 struct io_accel2_cmd {
 	u8  IU_type;			/* IU Type */
 	u8  direction;			/* direction, memtype, and encryption */
@@ -544,8 +534,7 @@ struct io_accel2_cmd {
 	u32 tweak_upper;		/* Encryption tweak, upper 4 bytes */
 	struct ioaccel2_sg_element sg[IOACCEL2_MAXSGENTRIES];
 	struct io_accel2_scsi_response error_data;
-	u8 pad[IOACCEL2_PAD];
-};
+} __aligned(IOACCEL2_COMMANDLIST_ALIGNMENT);
 
 /*
  * defines for Mode 2 command struct
@@ -636,7 +625,7 @@ struct TransTable_struct {
 	u32            RepQCount;
 	u32            RepQCtrAddrLow32;
 	u32            RepQCtrAddrHigh32;
-#define MAX_REPLY_QUEUES 8
+#define MAX_REPLY_QUEUES 64
 	struct vals32  RepQAddr[MAX_REPLY_QUEUES];
 };
 
diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c
index 2ebfb2bb0f42..7b23f21f22f1 100644
--- a/drivers/scsi/ibmvscsi/ibmvscsi.c
+++ b/drivers/scsi/ibmvscsi/ibmvscsi.c
@@ -185,6 +185,11 @@ static struct viosrp_crq *crq_queue_next_crq(struct crq_queue *queue)
 	if (crq->valid & 0x80) {
 		if (++queue->cur == queue->size)
 			queue->cur = 0;
+
+		/* Ensure the read of the valid bit occurs before reading any
+		 * other bits of the CRQ entry
+		 */
+		rmb();
 	} else
 		crq = NULL;
 	spin_unlock_irqrestore(&queue->lock, flags);
@@ -203,6 +208,11 @@ static int ibmvscsi_send_crq(struct ibmvscsi_host_data *hostdata,
 {
 	struct vio_dev *vdev = to_vio_dev(hostdata->dev);
 
+	/*
+	 * Ensure the command buffer is flushed to memory before handing it
+	 * over to the VIOS to prevent it from fetching any stale data.
+	 */
+	mb();
 	return plpar_hcall_norets(H_SEND_CRQ, vdev->unit_address, word1, word2);
 }
 
@@ -797,7 +807,8 @@ static void purge_requests(struct ibmvscsi_host_data *hostdata, int error_code)
 				       evt->hostdata->dev);
 			if (evt->cmnd_done)
 				evt->cmnd_done(evt->cmnd);
-		} else if (evt->done)
+		} else if (evt->done && evt->crq.format != VIOSRP_MAD_FORMAT &&
+			   evt->iu.srp.login_req.opcode != SRP_LOGIN_REQ)
 			evt->done(evt);
 		free_event_struct(&evt->hostdata->pool, evt);
 		spin_lock_irqsave(hostdata->host->host_lock, flags);
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 11854845393b..a669f2d11c31 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -244,7 +244,7 @@ iscsi_sw_tcp_conn_restore_callbacks(struct iscsi_conn *conn)
 	sk->sk_data_ready   = tcp_sw_conn->old_data_ready;
 	sk->sk_state_change = tcp_sw_conn->old_state_change;
 	sk->sk_write_space  = tcp_sw_conn->old_write_space;
-	sk->sk_no_check	 = 0;
+	sk->sk_no_check_tx = 0;
 	write_unlock_bh(&sk->sk_callback_lock);
 }
 
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 26dc005bb0f0..3d1bc67bac9d 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -338,7 +338,7 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_task *task)
 	struct iscsi_session *session = conn->session;
 	struct scsi_cmnd *sc = task->sc;
 	struct iscsi_scsi_req *hdr;
-	unsigned hdrlength, cmd_len;
+	unsigned hdrlength, cmd_len, transfer_length;
 	itt_t itt;
 	int rc;
 
@@ -391,11 +391,11 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_task *task)
 	if (scsi_get_prot_op(sc) != SCSI_PROT_NORMAL)
 		task->protected = true;
 
+	transfer_length = scsi_transfer_length(sc);
+	hdr->data_length = cpu_to_be32(transfer_length);
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
-		unsigned out_len = scsi_out(sc)->length;
 		struct iscsi_r2t_info *r2t = &task->unsol_r2t;
 
-		hdr->data_length = cpu_to_be32(out_len);
 		hdr->flags |= ISCSI_FLAG_CMD_WRITE;
 		/*
 		 * Write counters:
@@ -414,18 +414,19 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_task *task)
 		memset(r2t, 0, sizeof(*r2t));
 
 		if (session->imm_data_en) {
-			if (out_len >= session->first_burst)
+			if (transfer_length >= session->first_burst)
 				task->imm_count = min(session->first_burst,
 							conn->max_xmit_dlength);
 			else
-				task->imm_count = min(out_len,
-							conn->max_xmit_dlength);
+				task->imm_count = min(transfer_length,
+						      conn->max_xmit_dlength);
 			hton24(hdr->dlength, task->imm_count);
 		} else
 			zero_data(hdr->dlength);
 
 		if (!session->initial_r2t_en) {
-			r2t->data_length = min(session->first_burst, out_len) -
+			r2t->data_length = min(session->first_burst,
+					       transfer_length) -
 					       task->imm_count;
 			r2t->data_offset = task->imm_count;
 			r2t->ttt = cpu_to_be32(ISCSI_RESERVED_TAG);
@@ -438,7 +439,6 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_task *task)
 	} else {
 		hdr->flags |= ISCSI_FLAG_CMD_FINAL;
 		zero_data(hdr->dlength);
-		hdr->data_length = cpu_to_be32(scsi_in(sc)->length);
 
 		if (sc->sc_data_direction == DMA_FROM_DEVICE)
 			hdr->flags |= ISCSI_FLAG_CMD_READ;
@@ -466,7 +466,7 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_task *task)
 			  scsi_bidi_cmnd(sc) ? "bidirectional" :
 			  sc->sc_data_direction == DMA_TO_DEVICE ?
 			  "write" : "read", conn->id, sc, sc->cmnd[0],
-			  task->itt, scsi_bufflen(sc),
+			  task->itt, transfer_length,
 			  scsi_bidi_cmnd(sc) ? scsi_in(sc)->length : 0,
 			  session->cmdsn,
 			  session->max_cmdsn - session->exp_cmdsn + 1);
@@ -1442,9 +1442,9 @@ static int iscsi_xmit_task(struct iscsi_conn *conn)
 		conn->task = NULL;
 	}
 	/* regular RX path uses back_lock */
-	spin_lock_bh(&conn->session->back_lock);
+	spin_lock(&conn->session->back_lock);
 	__iscsi_put_task(task);
-	spin_unlock_bh(&conn->session->back_lock);
+	spin_unlock(&conn->session->back_lock);
 	return rc;
 }
 
diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h
index 94a3cafe7197..434e9037908e 100644
--- a/drivers/scsi/lpfc/lpfc.h
+++ b/drivers/scsi/lpfc/lpfc.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
@@ -640,6 +640,7 @@ struct lpfc_hba {
 #define HBA_DEVLOSS_TMO         0x2000 /* HBA in devloss timeout */
 #define HBA_RRQ_ACTIVE		0x4000 /* process the rrq active list */
 #define HBA_FCP_IOQ_FLUSH	0x8000 /* FCP I/O queues being flushed */
+#define HBA_FW_DUMP_OP		0x10000 /* Skips fn reset before FW dump */
 	uint32_t fcp_ring_in_use; /* When polling test if intr-hndlr active*/
 	struct lpfc_dmabuf slim2p;
 
diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c
index 8d5b6ceec9c9..1d7a5c34ee8c 100644
--- a/drivers/scsi/lpfc/lpfc_attr.c
+++ b/drivers/scsi/lpfc/lpfc_attr.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
@@ -919,10 +919,15 @@ lpfc_sli4_pdev_reg_request(struct lpfc_hba *phba, uint32_t opcode)
 		phba->cfg_sriov_nr_virtfn = 0;
 	}
 
+	if (opcode == LPFC_FW_DUMP)
+		phba->hba_flag |= HBA_FW_DUMP_OP;
+
 	status = lpfc_do_offline(phba, LPFC_EVT_OFFLINE);
 
-	if (status != 0)
+	if (status != 0) {
+		phba->hba_flag &= ~HBA_FW_DUMP_OP;
 		return status;
+	}
 
 	/* wait for the device to be quiesced before firmware reset */
 	msleep(100);
@@ -2364,7 +2369,7 @@ lpfc_oas_tgt_store(struct device *dev, struct device_attribute *attr,
 	uint8_t wwpn[WWN_SZ];
 	int rc;
 
-	if (!phba->cfg_EnableXLane)
+	if (!phba->cfg_fof)
 		return -EPERM;
 
 	/* count may include a LF at end of string */
@@ -2432,7 +2437,7 @@ lpfc_oas_vpt_store(struct device *dev, struct device_attribute *attr,
 	uint8_t wwpn[WWN_SZ];
 	int rc;
 
-	if (!phba->cfg_EnableXLane)
+	if (!phba->cfg_fof)
 		return -EPERM;
 
 	/* count may include a LF at end of string */
@@ -2499,7 +2504,7 @@ lpfc_oas_lun_state_store(struct device *dev, struct device_attribute *attr,
 	struct lpfc_hba *phba = ((struct lpfc_vport *)shost->hostdata)->phba;
 	int val = 0;
 
-	if (!phba->cfg_EnableXLane)
+	if (!phba->cfg_fof)
 		return -EPERM;
 
 	if (!isdigit(buf[0]))
@@ -2565,7 +2570,7 @@ lpfc_oas_lun_state_set(struct lpfc_hba *phba, uint8_t vpt_wwpn[],
 
 	int rc = 0;
 
-	if (!phba->cfg_EnableXLane)
+	if (!phba->cfg_fof)
 		return -EPERM;
 
 	if (oas_state) {
@@ -2670,7 +2675,7 @@ lpfc_oas_lun_show(struct device *dev, struct device_attribute *attr,
 	uint64_t oas_lun;
 	int len = 0;
 
-	if (!phba->cfg_EnableXLane)
+	if (!phba->cfg_fof)
 		return -EPERM;
 
 	if (wwn_to_u64(phba->cfg_oas_vpt_wwpn) == 0)
@@ -2716,7 +2721,7 @@ lpfc_oas_lun_store(struct device *dev, struct device_attribute *attr,
 	uint64_t scsi_lun;
 	ssize_t rc;
 
-	if (!phba->cfg_EnableXLane)
+	if (!phba->cfg_fof)
 		return -EPERM;
 
 	if (wwn_to_u64(phba->cfg_oas_vpt_wwpn) == 0)
@@ -4655,7 +4660,7 @@ LPFC_ATTR_R(EnableXLane, 0, 0, 1, "Enable Express Lane Feature.");
 #       0x0 - 0x7f  = CS_CTL field in FC header (high 7 bits)
 # Value range is [0x0,0x7f]. Default value is 0
 */
-LPFC_ATTR_R(XLanePriority, 0, 0x0, 0x7f, "CS_CTL for Express Lane Feature.");
+LPFC_ATTR_RW(XLanePriority, 0, 0x0, 0x7f, "CS_CTL for Express Lane Feature.");
 
 /*
 # lpfc_enable_bg: Enable BlockGuard (Emulex's Implementation of T10-DIF)
diff --git a/drivers/scsi/lpfc/lpfc_bsg.c b/drivers/scsi/lpfc/lpfc_bsg.c
index ca2f4ea7cdef..5b5c825d9576 100644
--- a/drivers/scsi/lpfc/lpfc_bsg.c
+++ b/drivers/scsi/lpfc/lpfc_bsg.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2009-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2009-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_bsg.h b/drivers/scsi/lpfc/lpfc_bsg.h
index a94d4c9dfaa5..928ef609f363 100644
--- a/drivers/scsi/lpfc/lpfc_bsg.h
+++ b/drivers/scsi/lpfc/lpfc_bsg.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2010-2012 Emulex.  All rights reserved.                *
+ * Copyright (C) 2010-2014 Emulex.  All rights reserved.                *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_crtn.h b/drivers/scsi/lpfc/lpfc_crtn.h
index adda0bf7a244..db5604f01a1a 100644
--- a/drivers/scsi/lpfc/lpfc_crtn.h
+++ b/drivers/scsi/lpfc/lpfc_crtn.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
@@ -289,6 +289,7 @@ int lpfc_sli_issue_iocb(struct lpfc_hba *, uint32_t,
 void lpfc_sli_pcimem_bcopy(void *, void *, uint32_t);
 void lpfc_sli_bemem_bcopy(void *, void *, uint32_t);
 void lpfc_sli_abort_iocb_ring(struct lpfc_hba *, struct lpfc_sli_ring *);
+void lpfc_sli_abort_fcp_rings(struct lpfc_hba *phba);
 void lpfc_sli_hba_iocb_abort(struct lpfc_hba *);
 void lpfc_sli_flush_fcp_rings(struct lpfc_hba *);
 int lpfc_sli_ringpostbuf_put(struct lpfc_hba *, struct lpfc_sli_ring *,
@@ -310,6 +311,9 @@ int lpfc_sli_issue_abort_iotag(struct lpfc_hba *, struct lpfc_sli_ring *,
 int lpfc_sli_sum_iocb(struct lpfc_vport *, uint16_t, uint64_t, lpfc_ctx_cmd);
 int lpfc_sli_abort_iocb(struct lpfc_vport *, struct lpfc_sli_ring *, uint16_t,
 			uint64_t, lpfc_ctx_cmd);
+int
+lpfc_sli_abort_taskmgmt(struct lpfc_vport *, struct lpfc_sli_ring *,
+			uint16_t, uint64_t, lpfc_ctx_cmd);
 
 void lpfc_mbox_timeout(unsigned long);
 void lpfc_mbox_timeout_handler(struct lpfc_hba *);
diff --git a/drivers/scsi/lpfc/lpfc_debugfs.c b/drivers/scsi/lpfc/lpfc_debugfs.c
index 828c08e9389e..b0aedce3f54b 100644
--- a/drivers/scsi/lpfc/lpfc_debugfs.c
+++ b/drivers/scsi/lpfc/lpfc_debugfs.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2007-2012 Emulex.  All rights reserved.           *
+ * Copyright (C) 2007-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
@@ -2314,7 +2314,7 @@ proc_cq:
 			goto too_big;
 	}
 
-	if (phba->cfg_EnableXLane) {
+	if (phba->cfg_fof) {
 
 		/* OAS CQ */
 		qp = phba->sli4_hba.oas_cq;
diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c
index 624fe0b3cc0b..7a5d81a65be8 100644
--- a/drivers/scsi/lpfc/lpfc_els.c
+++ b/drivers/scsi/lpfc/lpfc_els.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index 294c072e9083..2a17e31265b8 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
@@ -5634,6 +5634,9 @@ lpfc_nlp_init(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 		ndlp->active_rrqs_xri_bitmap =
 				mempool_alloc(vport->phba->active_rrq_pool,
 					      GFP_KERNEL);
+		if (ndlp->active_rrqs_xri_bitmap)
+			memset(ndlp->active_rrqs_xri_bitmap, 0,
+			       ndlp->phba->cfg_rrq_xri_bitmap_sz);
 	}
 
 
diff --git a/drivers/scsi/lpfc/lpfc_hw.h b/drivers/scsi/lpfc/lpfc_hw.h
index 3d9438ce59ab..236259252379 100644
--- a/drivers/scsi/lpfc/lpfc_hw.h
+++ b/drivers/scsi/lpfc/lpfc_hw.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_hw4.h b/drivers/scsi/lpfc/lpfc_hw4.h
index fd79f7de7666..f432ec180cf8 100644
--- a/drivers/scsi/lpfc/lpfc_hw4.h
+++ b/drivers/scsi/lpfc/lpfc_hw4.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2009-2013 Emulex.  All rights reserved.                *
+ * Copyright (C) 2009-2014 Emulex.  All rights reserved.                *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index 635eeb3d6987..06f9a5b79e66 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
@@ -820,57 +820,153 @@ lpfc_hba_down_prep(struct lpfc_hba *phba)
 }
 
 /**
- * lpfc_hba_down_post_s3 - Perform lpfc uninitialization after HBA reset
+ * lpfc_sli4_free_sp_events - Cleanup sp_queue_events to free
+ * rspiocb which got deferred
+ *
  * @phba: pointer to lpfc HBA data structure.
  *
- * This routine will do uninitialization after the HBA is reset when bring
- * down the SLI Layer.
+ * This routine will cleanup completed slow path events after HBA is reset
+ * when bringing down the SLI Layer.
+ *
  *
  * Return codes
- *   0 - success.
- *   Any other value - error.
+ *   void.
  **/
-static int
-lpfc_hba_down_post_s3(struct lpfc_hba *phba)
+static void
+lpfc_sli4_free_sp_events(struct lpfc_hba *phba)
+{
+	struct lpfc_iocbq *rspiocbq;
+	struct hbq_dmabuf *dmabuf;
+	struct lpfc_cq_event *cq_event;
+
+	spin_lock_irq(&phba->hbalock);
+	phba->hba_flag &= ~HBA_SP_QUEUE_EVT;
+	spin_unlock_irq(&phba->hbalock);
+
+	while (!list_empty(&phba->sli4_hba.sp_queue_event)) {
+		/* Get the response iocb from the head of work queue */
+		spin_lock_irq(&phba->hbalock);
+		list_remove_head(&phba->sli4_hba.sp_queue_event,
+				 cq_event, struct lpfc_cq_event, list);
+		spin_unlock_irq(&phba->hbalock);
+
+		switch (bf_get(lpfc_wcqe_c_code, &cq_event->cqe.wcqe_cmpl)) {
+		case CQE_CODE_COMPL_WQE:
+			rspiocbq = container_of(cq_event, struct lpfc_iocbq,
+						 cq_event);
+			lpfc_sli_release_iocbq(phba, rspiocbq);
+			break;
+		case CQE_CODE_RECEIVE:
+		case CQE_CODE_RECEIVE_V1:
+			dmabuf = container_of(cq_event, struct hbq_dmabuf,
+					      cq_event);
+			lpfc_in_buf_free(phba, &dmabuf->dbuf);
+		}
+	}
+}
+
+/**
+ * lpfc_hba_free_post_buf - Perform lpfc uninitialization after HBA reset
+ * @phba: pointer to lpfc HBA data structure.
+ *
+ * This routine will cleanup posted ELS buffers after the HBA is reset
+ * when bringing down the SLI Layer.
+ *
+ *
+ * Return codes
+ *   void.
+ **/
+static void
+lpfc_hba_free_post_buf(struct lpfc_hba *phba)
 {
 	struct lpfc_sli *psli = &phba->sli;
 	struct lpfc_sli_ring *pring;
 	struct lpfc_dmabuf *mp, *next_mp;
-	LIST_HEAD(completions);
-	int i;
+	LIST_HEAD(buflist);
+	int count;
 
 	if (phba->sli3_options & LPFC_SLI3_HBQ_ENABLED)
 		lpfc_sli_hbqbuf_free_all(phba);
 	else {
 		/* Cleanup preposted buffers on the ELS ring */
 		pring = &psli->ring[LPFC_ELS_RING];
-		list_for_each_entry_safe(mp, next_mp, &pring->postbufq, list) {
+		spin_lock_irq(&phba->hbalock);
+		list_splice_init(&pring->postbufq, &buflist);
+		spin_unlock_irq(&phba->hbalock);
+
+		count = 0;
+		list_for_each_entry_safe(mp, next_mp, &buflist, list) {
 			list_del(&mp->list);
-			pring->postbufq_cnt--;
+			count++;
 			lpfc_mbuf_free(phba, mp->virt, mp->phys);
 			kfree(mp);
 		}
+
+		spin_lock_irq(&phba->hbalock);
+		pring->postbufq_cnt -= count;
+		spin_unlock_irq(&phba->hbalock);
 	}
+}
+
+/**
+ * lpfc_hba_clean_txcmplq - Perform lpfc uninitialization after HBA reset
+ * @phba: pointer to lpfc HBA data structure.
+ *
+ * This routine will cleanup the txcmplq after the HBA is reset when bringing
+ * down the SLI Layer.
+ *
+ * Return codes
+ *   void
+ **/
+static void
+lpfc_hba_clean_txcmplq(struct lpfc_hba *phba)
+{
+	struct lpfc_sli *psli = &phba->sli;
+	struct lpfc_sli_ring *pring;
+	LIST_HEAD(completions);
+	int i;
 
-	spin_lock_irq(&phba->hbalock);
 	for (i = 0; i < psli->num_rings; i++) {
 		pring = &psli->ring[i];
-
+		if (phba->sli_rev >= LPFC_SLI_REV4)
+			spin_lock_irq(&pring->ring_lock);
+		else
+			spin_lock_irq(&phba->hbalock);
 		/* At this point in time the HBA is either reset or DOA. Either
 		 * way, nothing should be on txcmplq as it will NEVER complete.
 		 */
 		list_splice_init(&pring->txcmplq, &completions);
-		spin_unlock_irq(&phba->hbalock);
+		pring->txcmplq_cnt = 0;
+
+		if (phba->sli_rev >= LPFC_SLI_REV4)
+			spin_unlock_irq(&pring->ring_lock);
+		else
+			spin_unlock_irq(&phba->hbalock);
 
 		/* Cancel all the IOCBs from the completions list */
 		lpfc_sli_cancel_iocbs(phba, &completions, IOSTAT_LOCAL_REJECT,
 				      IOERR_SLI_ABORTED);
-
 		lpfc_sli_abort_iocb_ring(phba, pring);
-		spin_lock_irq(&phba->hbalock);
 	}
-	spin_unlock_irq(&phba->hbalock);
+}
 
+/**
+ * lpfc_hba_down_post_s3 - Perform lpfc uninitialization after HBA reset
+	int i;
+ * @phba: pointer to lpfc HBA data structure.
+ *
+ * This routine will do uninitialization after the HBA is reset when bring
+ * down the SLI Layer.
+ *
+ * Return codes
+ *   0 - success.
+ *   Any other value - error.
+ **/
+static int
+lpfc_hba_down_post_s3(struct lpfc_hba *phba)
+{
+	lpfc_hba_free_post_buf(phba);
+	lpfc_hba_clean_txcmplq(phba);
 	return 0;
 }
 
@@ -890,13 +986,12 @@ lpfc_hba_down_post_s4(struct lpfc_hba *phba)
 {
 	struct lpfc_scsi_buf *psb, *psb_next;
 	LIST_HEAD(aborts);
-	int ret;
 	unsigned long iflag = 0;
 	struct lpfc_sglq *sglq_entry = NULL;
 
-	ret = lpfc_hba_down_post_s3(phba);
-	if (ret)
-		return ret;
+	lpfc_hba_free_post_buf(phba);
+	lpfc_hba_clean_txcmplq(phba);
+
 	/* At this point in time the HBA is either reset or DOA. Either
 	 * way, nothing should be on lpfc_abts_els_sgl_list, it needs to be
 	 * on the lpfc_sgl_list so that it can either be freed if the
@@ -932,6 +1027,8 @@ lpfc_hba_down_post_s4(struct lpfc_hba *phba)
 	spin_lock_irqsave(&phba->scsi_buf_list_put_lock, iflag);
 	list_splice(&aborts, &phba->lpfc_scsi_buf_list_put);
 	spin_unlock_irqrestore(&phba->scsi_buf_list_put_lock, iflag);
+
+	lpfc_sli4_free_sp_events(phba);
 	return 0;
 }
 
@@ -1250,7 +1347,6 @@ static void
 lpfc_handle_deferred_eratt(struct lpfc_hba *phba)
 {
 	uint32_t old_host_status = phba->work_hs;
-	struct lpfc_sli_ring  *pring;
 	struct lpfc_sli *psli = &phba->sli;
 
 	/* If the pci channel is offline, ignore possible errors,
@@ -1279,8 +1375,7 @@ lpfc_handle_deferred_eratt(struct lpfc_hba *phba)
 	 * dropped by the firmware. Error iocb (I/O) on txcmplq and let the
 	 * SCSI layer retry it after re-establishing link.
 	 */
-	pring = &psli->ring[psli->fcp_ring];
-	lpfc_sli_abort_iocb_ring(phba, pring);
+	lpfc_sli_abort_fcp_rings(phba);
 
 	/*
 	 * There was a firmware error. Take the hba offline and then
@@ -1348,7 +1443,6 @@ lpfc_handle_eratt_s3(struct lpfc_hba *phba)
 {
 	struct lpfc_vport *vport = phba->pport;
 	struct lpfc_sli   *psli = &phba->sli;
-	struct lpfc_sli_ring  *pring;
 	uint32_t event_data;
 	unsigned long temperature;
 	struct temp_event temp_event_data;
@@ -1400,8 +1494,7 @@ lpfc_handle_eratt_s3(struct lpfc_hba *phba)
 		* Error iocb (I/O) on txcmplq and let the SCSI layer
 		* retry it after re-establishing link.
 		*/
-		pring = &psli->ring[psli->fcp_ring];
-		lpfc_sli_abort_iocb_ring(phba, pring);
+		lpfc_sli_abort_fcp_rings(phba);
 
 		/*
 		 * There was a firmware error.  Take the hba offline and then
@@ -1940,78 +2033,81 @@ lpfc_get_hba_model_desc(struct lpfc_hba *phba, uint8_t *mdp, uint8_t *descp)
 
 	switch (dev_id) {
 	case PCI_DEVICE_ID_FIREFLY:
-		m = (typeof(m)){"LP6000", "PCI", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LP6000", "PCI",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_SUPERFLY:
 		if (vp->rev.biuRev >= 1 && vp->rev.biuRev <= 3)
-			m = (typeof(m)){"LP7000", "PCI",
-					"Fibre Channel Adapter"};
+			m = (typeof(m)){"LP7000", "PCI", ""};
 		else
-			m = (typeof(m)){"LP7000E", "PCI",
-					"Fibre Channel Adapter"};
+			m = (typeof(m)){"LP7000E", "PCI", ""};
+		m.function = "Obsolete, Unsupported Fibre Channel Adapter";
 		break;
 	case PCI_DEVICE_ID_DRAGONFLY:
 		m = (typeof(m)){"LP8000", "PCI",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_CENTAUR:
 		if (FC_JEDEC_ID(vp->rev.biuRev) == CENTAUR_2G_JEDEC_ID)
-			m = (typeof(m)){"LP9002", "PCI",
-					"Fibre Channel Adapter"};
+			m = (typeof(m)){"LP9002", "PCI", ""};
 		else
-			m = (typeof(m)){"LP9000", "PCI",
-					"Fibre Channel Adapter"};
+			m = (typeof(m)){"LP9000", "PCI", ""};
+		m.function = "Obsolete, Unsupported Fibre Channel Adapter";
 		break;
 	case PCI_DEVICE_ID_RFLY:
 		m = (typeof(m)){"LP952", "PCI",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_PEGASUS:
 		m = (typeof(m)){"LP9802", "PCI-X",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_THOR:
 		m = (typeof(m)){"LP10000", "PCI-X",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_VIPER:
 		m = (typeof(m)){"LPX1000",  "PCI-X",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_PFLY:
 		m = (typeof(m)){"LP982", "PCI-X",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_TFLY:
 		m = (typeof(m)){"LP1050", "PCI-X",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_HELIOS:
 		m = (typeof(m)){"LP11000", "PCI-X2",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_HELIOS_SCSP:
 		m = (typeof(m)){"LP11000-SP", "PCI-X2",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_HELIOS_DCSP:
 		m = (typeof(m)){"LP11002-SP",  "PCI-X2",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_NEPTUNE:
-		m = (typeof(m)){"LPe1000", "PCIe", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LPe1000", "PCIe",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_NEPTUNE_SCSP:
-		m = (typeof(m)){"LPe1000-SP", "PCIe", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LPe1000-SP", "PCIe",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_NEPTUNE_DCSP:
-		m = (typeof(m)){"LPe1002-SP", "PCIe", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LPe1002-SP", "PCIe",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_BMID:
 		m = (typeof(m)){"LP1150", "PCI-X2", "Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_BSMB:
-		m = (typeof(m)){"LP111", "PCI-X2", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LP111", "PCI-X2",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_ZEPHYR:
 		m = (typeof(m)){"LPe11000", "PCIe", "Fibre Channel Adapter"};
@@ -2030,16 +2126,20 @@ lpfc_get_hba_model_desc(struct lpfc_hba *phba, uint8_t *mdp, uint8_t *descp)
 		m = (typeof(m)){"LPe111", "PCIe", "Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_LP101:
-		m = (typeof(m)){"LP101", "PCI-X", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LP101", "PCI-X",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_LP10000S:
-		m = (typeof(m)){"LP10000-S", "PCI", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LP10000-S", "PCI",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_LP11000S:
-		m = (typeof(m)){"LP11000-S", "PCI-X2", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LP11000-S", "PCI-X2",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_LPE11000S:
-		m = (typeof(m)){"LPe11000-S", "PCIe", "Fibre Channel Adapter"};
+		m = (typeof(m)){"LPe11000-S", "PCIe",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_SAT:
 		m = (typeof(m)){"LPe12000", "PCIe", "Fibre Channel Adapter"};
@@ -2060,20 +2160,21 @@ lpfc_get_hba_model_desc(struct lpfc_hba *phba, uint8_t *mdp, uint8_t *descp)
 		m = (typeof(m)){"LPe12000-S", "PCIe", "Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_HORNET:
-		m = (typeof(m)){"LP21000", "PCIe", "FCoE Adapter"};
+		m = (typeof(m)){"LP21000", "PCIe",
+				"Obsolete, Unsupported FCoE Adapter"};
 		GE = 1;
 		break;
 	case PCI_DEVICE_ID_PROTEUS_VF:
 		m = (typeof(m)){"LPev12000", "PCIe IOV",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_PROTEUS_PF:
 		m = (typeof(m)){"LPev12000", "PCIe IOV",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_PROTEUS_S:
 		m = (typeof(m)){"LPemv12002-S", "PCIe IOV",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_TIGERSHARK:
 		oneConnect = 1;
@@ -2089,17 +2190,24 @@ lpfc_get_hba_model_desc(struct lpfc_hba *phba, uint8_t *mdp, uint8_t *descp)
 		break;
 	case PCI_DEVICE_ID_BALIUS:
 		m = (typeof(m)){"LPVe12002", "PCIe Shared I/O",
-				"Fibre Channel Adapter"};
+				"Obsolete, Unsupported Fibre Channel Adapter"};
 		break;
 	case PCI_DEVICE_ID_LANCER_FC:
-	case PCI_DEVICE_ID_LANCER_FC_VF:
 		m = (typeof(m)){"LPe16000", "PCIe", "Fibre Channel Adapter"};
 		break;
+	case PCI_DEVICE_ID_LANCER_FC_VF:
+		m = (typeof(m)){"LPe16000", "PCIe",
+				"Obsolete, Unsupported Fibre Channel Adapter"};
+		break;
 	case PCI_DEVICE_ID_LANCER_FCOE:
-	case PCI_DEVICE_ID_LANCER_FCOE_VF:
 		oneConnect = 1;
 		m = (typeof(m)){"OCe15100", "PCIe", "FCoE"};
 		break;
+	case PCI_DEVICE_ID_LANCER_FCOE_VF:
+		oneConnect = 1;
+		m = (typeof(m)){"OCe15100", "PCIe",
+				"Obsolete, Unsupported FCoE"};
+		break;
 	case PCI_DEVICE_ID_SKYHAWK:
 	case PCI_DEVICE_ID_SKYHAWK_VF:
 		oneConnect = 1;
@@ -4614,7 +4722,10 @@ lpfc_reset_hba(struct lpfc_hba *phba)
 		phba->link_state = LPFC_HBA_ERROR;
 		return;
 	}
-	lpfc_offline_prep(phba, LPFC_MBX_WAIT);
+	if (phba->sli.sli_flag & LPFC_SLI_ACTIVE)
+		lpfc_offline_prep(phba, LPFC_MBX_WAIT);
+	else
+		lpfc_offline_prep(phba, LPFC_MBX_NO_WAIT);
 	lpfc_offline(phba);
 	lpfc_sli_brdrestart(phba);
 	lpfc_online(phba);
@@ -9663,9 +9774,6 @@ lpfc_pci_resume_one_s3(struct pci_dev *pdev)
 static void
 lpfc_sli_prep_dev_for_recover(struct lpfc_hba *phba)
 {
-	struct lpfc_sli *psli = &phba->sli;
-	struct lpfc_sli_ring  *pring;
-
 	lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
 			"2723 PCI channel I/O abort preparing for recovery\n");
 
@@ -9673,8 +9781,7 @@ lpfc_sli_prep_dev_for_recover(struct lpfc_hba *phba)
 	 * There may be errored I/Os through HBA, abort all I/Os on txcmplq
 	 * and let the SCSI mid-layer to retry them to recover.
 	 */
-	pring = &psli->ring[psli->fcp_ring];
-	lpfc_sli_abort_iocb_ring(phba, pring);
+	lpfc_sli_abort_fcp_rings(phba);
 }
 
 /**
@@ -10417,17 +10524,13 @@ lpfc_pci_resume_one_s4(struct pci_dev *pdev)
 static void
 lpfc_sli4_prep_dev_for_recover(struct lpfc_hba *phba)
 {
-	struct lpfc_sli *psli = &phba->sli;
-	struct lpfc_sli_ring  *pring;
-
 	lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
 			"2828 PCI channel I/O abort preparing for recovery\n");
 	/*
 	 * There may be errored I/Os through HBA, abort all I/Os on txcmplq
 	 * and let the SCSI mid-layer to retry them to recover.
 	 */
-	pring = &psli->ring[psli->fcp_ring];
-	lpfc_sli_abort_iocb_ring(phba, pring);
+	lpfc_sli_abort_fcp_rings(phba);
 }
 
 /**
@@ -10898,7 +11001,7 @@ lpfc_sli4_oas_verify(struct lpfc_hba *phba)
 	if (phba->sli4_hba.pc_sli4_params.oas_supported) {
 		phba->cfg_fof = 1;
 	} else {
-		phba->cfg_EnableXLane = 0;
+		phba->cfg_fof = 0;
 		if (phba->device_data_mem_pool)
 			mempool_destroy(phba->device_data_mem_pool);
 		phba->device_data_mem_pool = NULL;
@@ -10928,7 +11031,7 @@ lpfc_fof_queue_setup(struct lpfc_hba *phba)
 	if (rc)
 		return -ENOMEM;
 
-	if (phba->cfg_EnableXLane) {
+	if (phba->cfg_fof) {
 
 		rc = lpfc_cq_create(phba, phba->sli4_hba.oas_cq,
 				    phba->sli4_hba.fof_eq, LPFC_WCQ, LPFC_FCP);
@@ -10947,8 +11050,7 @@ lpfc_fof_queue_setup(struct lpfc_hba *phba)
 	return 0;
 
 out_oas_wq:
-	if (phba->cfg_EnableXLane)
-		lpfc_cq_destroy(phba, phba->sli4_hba.oas_cq);
+	lpfc_cq_destroy(phba, phba->sli4_hba.oas_cq);
 out_oas_cq:
 	lpfc_eq_destroy(phba, phba->sli4_hba.fof_eq);
 	return rc;
@@ -10982,7 +11084,7 @@ lpfc_fof_queue_create(struct lpfc_hba *phba)
 
 	phba->sli4_hba.fof_eq = qdesc;
 
-	if (phba->cfg_EnableXLane) {
+	if (phba->cfg_fof) {
 
 		/* Create OAS CQ */
 		qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.cq_esize,
diff --git a/drivers/scsi/lpfc/lpfc_mem.c b/drivers/scsi/lpfc/lpfc_mem.c
index ed419aad2b1f..3fa65338d3f5 100644
--- a/drivers/scsi/lpfc/lpfc_mem.c
+++ b/drivers/scsi/lpfc/lpfc_mem.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2012 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index 462453ee0bda..2df11daad85b 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
@@ -73,7 +73,7 @@ lpfc_rport_data_from_scsi_device(struct scsi_device *sdev)
 {
 	struct lpfc_vport *vport = (struct lpfc_vport *)sdev->host->hostdata;
 
-	if (vport->phba->cfg_EnableXLane)
+	if (vport->phba->cfg_fof)
 		return ((struct lpfc_device_data *)sdev->hostdata)->rport_data;
 	else
 		return (struct lpfc_rport_data *)sdev->hostdata;
@@ -3462,7 +3462,7 @@ lpfc_scsi_prep_dma_buf_s4(struct lpfc_hba *phba, struct lpfc_scsi_buf *lpfc_cmd)
 	 * If the OAS driver feature is enabled and the lun is enabled for
 	 * OAS, set the oas iocb related flags.
 	 */
-	if ((phba->cfg_EnableXLane) && ((struct lpfc_device_data *)
+	if ((phba->cfg_fof) && ((struct lpfc_device_data *)
 		scsi_cmnd->device->hostdata)->oas_enabled)
 		lpfc_cmd->cur_iocbq.iocb_flag |= LPFC_IO_OAS;
 	return 0;
@@ -4314,6 +4314,7 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd,
 		fcp_cmnd->fcpCntl1 = SIMPLE_Q;
 
 	sli4 = (phba->sli_rev == LPFC_SLI_REV4);
+	piocbq->iocb.un.fcpi.fcpi_XRdy = 0;
 
 	/*
 	 * There are three possibilities here - use scatter-gather segment, use
@@ -4782,7 +4783,9 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 	struct lpfc_scsi_buf *lpfc_cmd;
 	IOCB_t *cmd, *icmd;
 	int ret = SUCCESS, status = 0;
-	unsigned long flags;
+	struct lpfc_sli_ring *pring_s4;
+	int ring_number, ret_val;
+	unsigned long flags, iflags;
 	DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waitq);
 
 	status = fc_block_scsi_eh(cmnd);
@@ -4833,6 +4836,14 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 
 	BUG_ON(iocb->context1 != lpfc_cmd);
 
+	/* abort issued in recovery is still in progress */
+	if (iocb->iocb_flag & LPFC_DRIVER_ABORTED) {
+		lpfc_printf_vlog(vport, KERN_WARNING, LOG_FCP,
+			 "3389 SCSI Layer I/O Abort Request is pending\n");
+		spin_unlock_irqrestore(&phba->hbalock, flags);
+		goto wait_for_cmpl;
+	}
+
 	abtsiocb = __lpfc_sli_get_iocbq(phba);
 	if (abtsiocb == NULL) {
 		ret = FAILED;
@@ -4871,11 +4882,23 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 
 	abtsiocb->iocb_cmpl = lpfc_sli_abort_fcp_cmpl;
 	abtsiocb->vport = vport;
+	if (phba->sli_rev == LPFC_SLI_REV4) {
+		ring_number = MAX_SLI3_CONFIGURED_RINGS + iocb->fcp_wqidx;
+		pring_s4 = &phba->sli.ring[ring_number];
+		/* Note: both hbalock and ring_lock must be set here */
+		spin_lock_irqsave(&pring_s4->ring_lock, iflags);
+		ret_val = __lpfc_sli_issue_iocb(phba, pring_s4->ringno,
+						abtsiocb, 0);
+		spin_unlock_irqrestore(&pring_s4->ring_lock, iflags);
+	} else {
+		ret_val = __lpfc_sli_issue_iocb(phba, LPFC_FCP_RING,
+						abtsiocb, 0);
+	}
 	/* no longer need the lock after this point */
 	spin_unlock_irqrestore(&phba->hbalock, flags);
 
-	if (lpfc_sli_issue_iocb(phba, LPFC_FCP_RING, abtsiocb, 0) ==
-	    IOCB_ERROR) {
+
+	if (ret_val == IOCB_ERROR) {
 		lpfc_sli_release_iocbq(phba, abtsiocb);
 		ret = FAILED;
 		goto out;
@@ -4885,12 +4908,16 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 		lpfc_sli_handle_fast_ring_event(phba,
 			&phba->sli.ring[LPFC_FCP_RING], HA_R0RE_REQ);
 
+wait_for_cmpl:
 	lpfc_cmd->waitq = &waitq;
 	/* Wait for abort to complete */
 	wait_event_timeout(waitq,
 			  (lpfc_cmd->pCmd != cmnd),
 			   msecs_to_jiffies(2*vport->cfg_devloss_tmo*1000));
+
+	spin_lock_irqsave(shost->host_lock, flags);
 	lpfc_cmd->waitq = NULL;
+	spin_unlock_irqrestore(shost->host_lock, flags);
 
 	if (lpfc_cmd->pCmd == cmnd) {
 		ret = FAILED;
@@ -5172,8 +5199,9 @@ lpfc_reset_flush_io_context(struct lpfc_vport *vport, uint16_t tgt_id,
 
 	cnt = lpfc_sli_sum_iocb(vport, tgt_id, lun_id, context);
 	if (cnt)
-		lpfc_sli_abort_iocb(vport, &phba->sli.ring[phba->sli.fcp_ring],
-				    tgt_id, lun_id, context);
+		lpfc_sli_abort_taskmgmt(vport,
+					&phba->sli.ring[phba->sli.fcp_ring],
+					tgt_id, lun_id, context);
 	later = msecs_to_jiffies(2 * vport->cfg_devloss_tmo * 1000) + jiffies;
 	while (time_after(later, jiffies) && cnt) {
 		schedule_timeout_uninterruptible(msecs_to_jiffies(20));
@@ -5491,7 +5519,7 @@ lpfc_slave_alloc(struct scsi_device *sdev)
 	if (!rport || fc_remote_port_chkready(rport))
 		return -ENXIO;
 
-	if (phba->cfg_EnableXLane) {
+	if (phba->cfg_fof) {
 
 		/*
 		 * Check to see if the device data structure for the lun
@@ -5616,7 +5644,7 @@ lpfc_slave_destroy(struct scsi_device *sdev)
 	struct lpfc_device_data *device_data = sdev->hostdata;
 
 	atomic_dec(&phba->sdev_cnt);
-	if ((phba->cfg_EnableXLane) && (device_data)) {
+	if ((phba->cfg_fof) && (device_data)) {
 		spin_lock_irqsave(&phba->devicelock, flags);
 		device_data->available = false;
 		if (!device_data->oas_enabled)
@@ -5655,7 +5683,7 @@ lpfc_create_device_data(struct lpfc_hba *phba, struct lpfc_name *vport_wwpn,
 	int memory_flags;
 
 	if (unlikely(!phba) || !vport_wwpn || !target_wwpn  ||
-	    !(phba->cfg_EnableXLane))
+	    !(phba->cfg_fof))
 		return NULL;
 
 	/* Attempt to create the device data to contain lun info */
@@ -5693,7 +5721,7 @@ lpfc_delete_device_data(struct lpfc_hba *phba,
 {
 
 	if (unlikely(!phba) || !lun_info  ||
-	    !(phba->cfg_EnableXLane))
+	    !(phba->cfg_fof))
 		return;
 
 	if (!list_empty(&lun_info->listentry))
@@ -5727,7 +5755,7 @@ __lpfc_get_device_data(struct lpfc_hba *phba, struct list_head *list,
 	struct lpfc_device_data *lun_info;
 
 	if (unlikely(!phba) || !list || !vport_wwpn || !target_wwpn ||
-	    !phba->cfg_EnableXLane)
+	    !phba->cfg_fof)
 		return NULL;
 
 	/* Check to see if the lun is already enabled for OAS. */
@@ -5789,7 +5817,7 @@ lpfc_find_next_oas_lun(struct lpfc_hba *phba, struct lpfc_name *vport_wwpn,
 	    !starting_lun || !found_vport_wwpn ||
 	    !found_target_wwpn || !found_lun || !found_lun_status ||
 	    (*starting_lun == NO_MORE_OAS_LUN) ||
-	    !phba->cfg_EnableXLane)
+	    !phba->cfg_fof)
 		return false;
 
 	lun = *starting_lun;
@@ -5873,7 +5901,7 @@ lpfc_enable_oas_lun(struct lpfc_hba *phba, struct lpfc_name *vport_wwpn,
 	unsigned long flags;
 
 	if (unlikely(!phba) || !vport_wwpn || !target_wwpn ||
-	    !phba->cfg_EnableXLane)
+	    !phba->cfg_fof)
 		return false;
 
 	spin_lock_irqsave(&phba->devicelock, flags);
@@ -5930,7 +5958,7 @@ lpfc_disable_oas_lun(struct lpfc_hba *phba, struct lpfc_name *vport_wwpn,
 	unsigned long flags;
 
 	if (unlikely(!phba) || !vport_wwpn || !target_wwpn ||
-	    !phba->cfg_EnableXLane)
+	    !phba->cfg_fof)
 		return false;
 
 	spin_lock_irqsave(&phba->devicelock, flags);
diff --git a/drivers/scsi/lpfc/lpfc_scsi.h b/drivers/scsi/lpfc/lpfc_scsi.h
index 0120bfccf50b..0389ac1e7b83 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.h
+++ b/drivers/scsi/lpfc/lpfc_scsi.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 6bb51f8e3c1b..32ada0505576 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
@@ -265,6 +265,16 @@ lpfc_sli4_eq_get(struct lpfc_queue *q)
 		return NULL;
 
 	q->hba_index = idx;
+
+	/*
+	 * insert barrier for instruction interlock : data from the hardware
+	 * must have the valid bit checked before it can be copied and acted
+	 * upon. Given what was seen in lpfc_sli4_cq_get() of speculative
+	 * instructions allowing action on content before valid bit checked,
+	 * add barrier here as well. May not be needed as "content" is a
+	 * single 32-bit entity here (vs multi word structure for cq's).
+	 */
+	mb();
 	return eqe;
 }
 
@@ -370,6 +380,17 @@ lpfc_sli4_cq_get(struct lpfc_queue *q)
 
 	cqe = q->qe[q->hba_index].cqe;
 	q->hba_index = idx;
+
+	/*
+	 * insert barrier for instruction interlock : data from the hardware
+	 * must have the valid bit checked before it can be copied and acted
+	 * upon. Speculative instructions were allowing a bcopy at the start
+	 * of lpfc_sli4_fp_handle_wcqe(), which is called immediately
+	 * after our return, to copy data before the valid bit check above
+	 * was done. As such, some of the copied data was stale. The barrier
+	 * ensures the check is before any data is copied.
+	 */
+	mb();
 	return cqe;
 }
 
@@ -3511,14 +3532,27 @@ lpfc_sli_abort_iocb_ring(struct lpfc_hba *phba, struct lpfc_sli_ring *pring)
 	/* Error everything on txq and txcmplq
 	 * First do the txq.
 	 */
-	spin_lock_irq(&phba->hbalock);
-	list_splice_init(&pring->txq, &completions);
+	if (phba->sli_rev >= LPFC_SLI_REV4) {
+		spin_lock_irq(&pring->ring_lock);
+		list_splice_init(&pring->txq, &completions);
+		pring->txq_cnt = 0;
+		spin_unlock_irq(&pring->ring_lock);
 
-	/* Next issue ABTS for everything on the txcmplq */
-	list_for_each_entry_safe(iocb, next_iocb, &pring->txcmplq, list)
-		lpfc_sli_issue_abort_iotag(phba, pring, iocb);
+		spin_lock_irq(&phba->hbalock);
+		/* Next issue ABTS for everything on the txcmplq */
+		list_for_each_entry_safe(iocb, next_iocb, &pring->txcmplq, list)
+			lpfc_sli_issue_abort_iotag(phba, pring, iocb);
+		spin_unlock_irq(&phba->hbalock);
+	} else {
+		spin_lock_irq(&phba->hbalock);
+		list_splice_init(&pring->txq, &completions);
+		pring->txq_cnt = 0;
 
-	spin_unlock_irq(&phba->hbalock);
+		/* Next issue ABTS for everything on the txcmplq */
+		list_for_each_entry_safe(iocb, next_iocb, &pring->txcmplq, list)
+			lpfc_sli_issue_abort_iotag(phba, pring, iocb);
+		spin_unlock_irq(&phba->hbalock);
+	}
 
 	/* Cancel all the IOCBs from the completions list */
 	lpfc_sli_cancel_iocbs(phba, &completions, IOSTAT_LOCAL_REJECT,
@@ -3526,6 +3560,36 @@ lpfc_sli_abort_iocb_ring(struct lpfc_hba *phba, struct lpfc_sli_ring *pring)
 }
 
 /**
+ * lpfc_sli_abort_fcp_rings - Abort all iocbs in all FCP rings
+ * @phba: Pointer to HBA context object.
+ * @pring: Pointer to driver SLI ring object.
+ *
+ * This function aborts all iocbs in FCP rings and frees all the iocb
+ * objects in txq. This function issues an abort iocb for all the iocb commands
+ * in txcmplq. The iocbs in the txcmplq is not guaranteed to complete before
+ * the return of this function. The caller is not required to hold any locks.
+ **/
+void
+lpfc_sli_abort_fcp_rings(struct lpfc_hba *phba)
+{
+	struct lpfc_sli *psli = &phba->sli;
+	struct lpfc_sli_ring  *pring;
+	uint32_t i;
+
+	/* Look on all the FCP Rings for the iotag */
+	if (phba->sli_rev >= LPFC_SLI_REV4) {
+		for (i = 0; i < phba->cfg_fcp_io_channel; i++) {
+			pring = &psli->ring[i + MAX_SLI3_CONFIGURED_RINGS];
+			lpfc_sli_abort_iocb_ring(phba, pring);
+		}
+	} else {
+		pring = &psli->ring[psli->fcp_ring];
+		lpfc_sli_abort_iocb_ring(phba, pring);
+	}
+}
+
+
+/**
  * lpfc_sli_flush_fcp_rings - flush all iocbs in the fcp ring
  * @phba: Pointer to HBA context object.
  *
@@ -3542,28 +3606,55 @@ lpfc_sli_flush_fcp_rings(struct lpfc_hba *phba)
 	LIST_HEAD(txcmplq);
 	struct lpfc_sli *psli = &phba->sli;
 	struct lpfc_sli_ring  *pring;
-
-	/* Currently, only one fcp ring */
-	pring = &psli->ring[psli->fcp_ring];
+	uint32_t i;
 
 	spin_lock_irq(&phba->hbalock);
-	/* Retrieve everything on txq */
-	list_splice_init(&pring->txq, &txq);
-
-	/* Retrieve everything on the txcmplq */
-	list_splice_init(&pring->txcmplq, &txcmplq);
-
 	/* Indicate the I/O queues are flushed */
 	phba->hba_flag |= HBA_FCP_IOQ_FLUSH;
 	spin_unlock_irq(&phba->hbalock);
 
-	/* Flush the txq */
-	lpfc_sli_cancel_iocbs(phba, &txq, IOSTAT_LOCAL_REJECT,
-			      IOERR_SLI_DOWN);
+	/* Look on all the FCP Rings for the iotag */
+	if (phba->sli_rev >= LPFC_SLI_REV4) {
+		for (i = 0; i < phba->cfg_fcp_io_channel; i++) {
+			pring = &psli->ring[i + MAX_SLI3_CONFIGURED_RINGS];
+
+			spin_lock_irq(&pring->ring_lock);
+			/* Retrieve everything on txq */
+			list_splice_init(&pring->txq, &txq);
+			/* Retrieve everything on the txcmplq */
+			list_splice_init(&pring->txcmplq, &txcmplq);
+			pring->txq_cnt = 0;
+			pring->txcmplq_cnt = 0;
+			spin_unlock_irq(&pring->ring_lock);
+
+			/* Flush the txq */
+			lpfc_sli_cancel_iocbs(phba, &txq,
+					      IOSTAT_LOCAL_REJECT,
+					      IOERR_SLI_DOWN);
+			/* Flush the txcmpq */
+			lpfc_sli_cancel_iocbs(phba, &txcmplq,
+					      IOSTAT_LOCAL_REJECT,
+					      IOERR_SLI_DOWN);
+		}
+	} else {
+		pring = &psli->ring[psli->fcp_ring];
 
-	/* Flush the txcmpq */
-	lpfc_sli_cancel_iocbs(phba, &txcmplq, IOSTAT_LOCAL_REJECT,
-			      IOERR_SLI_DOWN);
+		spin_lock_irq(&phba->hbalock);
+		/* Retrieve everything on txq */
+		list_splice_init(&pring->txq, &txq);
+		/* Retrieve everything on the txcmplq */
+		list_splice_init(&pring->txcmplq, &txcmplq);
+		pring->txq_cnt = 0;
+		pring->txcmplq_cnt = 0;
+		spin_unlock_irq(&phba->hbalock);
+
+		/* Flush the txq */
+		lpfc_sli_cancel_iocbs(phba, &txq, IOSTAT_LOCAL_REJECT,
+				      IOERR_SLI_DOWN);
+		/* Flush the txcmpq */
+		lpfc_sli_cancel_iocbs(phba, &txcmplq, IOSTAT_LOCAL_REJECT,
+				      IOERR_SLI_DOWN);
+	}
 }
 
 /**
@@ -3966,12 +4057,13 @@ lpfc_sli4_brdreset(struct lpfc_hba *phba)
 {
 	struct lpfc_sli *psli = &phba->sli;
 	uint16_t cfg_value;
-	int rc;
+	int rc = 0;
 
 	/* Reset HBA */
 	lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
-			"0295 Reset HBA Data: x%x x%x\n",
-			phba->pport->port_state, psli->sli_flag);
+			"0295 Reset HBA Data: x%x x%x x%x\n",
+			phba->pport->port_state, psli->sli_flag,
+			phba->hba_flag);
 
 	/* perform board reset */
 	phba->fc_eventTag = 0;
@@ -3984,6 +4076,12 @@ lpfc_sli4_brdreset(struct lpfc_hba *phba)
 	phba->fcf.fcf_flag = 0;
 	spin_unlock_irq(&phba->hbalock);
 
+	/* SLI4 INTF 2: if FW dump is being taken skip INIT_PORT */
+	if (phba->hba_flag & HBA_FW_DUMP_OP) {
+		phba->hba_flag &= ~HBA_FW_DUMP_OP;
+		return rc;
+	}
+
 	/* Now physically reset the device */
 	lpfc_printf_log(phba, KERN_INFO, LOG_INIT,
 			"0389 Performing PCI function reset!\n");
@@ -4981,7 +5079,7 @@ lpfc_sli4_arm_cqeq_intr(struct lpfc_hba *phba)
 		} while (++fcp_eqidx < phba->cfg_fcp_io_channel);
 	}
 
-	if (phba->cfg_EnableXLane)
+	if (phba->cfg_fof)
 		lpfc_sli4_cq_release(phba->sli4_hba.oas_cq, LPFC_QUEUE_REARM);
 
 	if (phba->sli4_hba.hba_eq) {
@@ -6701,7 +6799,6 @@ lpfc_mbox_timeout_handler(struct lpfc_hba *phba)
 	LPFC_MBOXQ_t *pmbox = phba->sli.mbox_active;
 	MAILBOX_t *mb = &pmbox->u.mb;
 	struct lpfc_sli *psli = &phba->sli;
-	struct lpfc_sli_ring *pring;
 
 	/* If the mailbox completed, process the completion and return */
 	if (lpfc_sli4_process_missed_mbox_completions(phba))
@@ -6743,8 +6840,7 @@ lpfc_mbox_timeout_handler(struct lpfc_hba *phba)
 	psli->sli_flag &= ~LPFC_SLI_ACTIVE;
 	spin_unlock_irq(&phba->hbalock);
 
-	pring = &psli->ring[psli->fcp_ring];
-	lpfc_sli_abort_iocb_ring(phba, pring);
+	lpfc_sli_abort_fcp_rings(phba);
 
 	lpfc_printf_log(phba, KERN_ERR, LOG_MBOX | LOG_SLI,
 			"0345 Resetting board due to mailbox timeout\n");
@@ -8112,6 +8208,7 @@ lpfc_sli4_iocb2wqe(struct lpfc_hba *phba, struct lpfc_iocbq *iocbq,
 	abort_tag = (uint32_t) iocbq->iotag;
 	xritag = iocbq->sli4_xritag;
 	wqe->generic.wqe_com.word7 = 0; /* The ct field has moved so reset */
+	wqe->generic.wqe_com.word10 = 0;
 	/* words0-2 bpl convert bde */
 	if (iocbq->iocb.un.genreq64.bdl.bdeFlags == BUFF_TYPE_BLP_64) {
 		numBdes = iocbq->iocb.un.genreq64.bdl.bdeSize /
@@ -8618,8 +8715,7 @@ __lpfc_sli_issue_iocb_s4(struct lpfc_hba *phba, uint32_t ring_number,
 
 	if ((piocb->iocb_flag & LPFC_IO_FCP) ||
 	    (piocb->iocb_flag & LPFC_USE_FCPWQIDX)) {
-		if (!phba->cfg_EnableXLane || (!(piocb->iocb_flag &
-			LPFC_IO_OAS))) {
+		if (!phba->cfg_fof || (!(piocb->iocb_flag & LPFC_IO_OAS))) {
 			wq = phba->sli4_hba.fcp_wq[piocb->fcp_wqidx];
 		} else {
 			wq = phba->sli4_hba.oas_wq;
@@ -8714,7 +8810,7 @@ lpfc_sli_issue_iocb(struct lpfc_hba *phba, uint32_t ring_number,
 
 	if (phba->sli_rev == LPFC_SLI_REV4) {
 		if (piocb->iocb_flag &  LPFC_IO_FCP) {
-			if (!phba->cfg_EnableXLane || (!(piocb->iocb_flag &
+			if (!phba->cfg_fof || (!(piocb->iocb_flag &
 				LPFC_IO_OAS))) {
 				if (unlikely(!phba->sli4_hba.fcp_wq))
 					return IOCB_ERROR;
@@ -9149,6 +9245,7 @@ lpfc_sli_queue_setup(struct lpfc_hba *phba)
 		pring->sli.sli3.next_cmdidx  = 0;
 		pring->sli.sli3.local_getidx = 0;
 		pring->sli.sli3.cmdidx = 0;
+		pring->flag = 0;
 		INIT_LIST_HEAD(&pring->txq);
 		INIT_LIST_HEAD(&pring->txcmplq);
 		INIT_LIST_HEAD(&pring->iocb_continueq);
@@ -9784,43 +9881,6 @@ abort_iotag_exit:
 }
 
 /**
- * lpfc_sli_iocb_ring_abort - Unconditionally abort all iocbs on an iocb ring
- * @phba: Pointer to HBA context object.
- * @pring: Pointer to driver SLI ring object.
- *
- * This function aborts all iocbs in the given ring and frees all the iocb
- * objects in txq. This function issues abort iocbs unconditionally for all
- * the iocb commands in txcmplq. The iocbs in the txcmplq is not guaranteed
- * to complete before the return of this function. The caller is not required
- * to hold any locks.
- **/
-static void
-lpfc_sli_iocb_ring_abort(struct lpfc_hba *phba, struct lpfc_sli_ring *pring)
-{
-	LIST_HEAD(completions);
-	struct lpfc_iocbq *iocb, *next_iocb;
-
-	if (pring->ringno == LPFC_ELS_RING)
-		lpfc_fabric_abort_hba(phba);
-
-	spin_lock_irq(&phba->hbalock);
-
-	/* Take off all the iocbs on txq for cancelling */
-	list_splice_init(&pring->txq, &completions);
-	pring->txq_cnt = 0;
-
-	/* Next issue ABTS for everything on the txcmplq */
-	list_for_each_entry_safe(iocb, next_iocb, &pring->txcmplq, list)
-		lpfc_sli_abort_iotag_issue(phba, pring, iocb);
-
-	spin_unlock_irq(&phba->hbalock);
-
-	/* Cancel all the IOCBs from the completions list */
-	lpfc_sli_cancel_iocbs(phba, &completions, IOSTAT_LOCAL_REJECT,
-			      IOERR_SLI_ABORTED);
-}
-
-/**
  * lpfc_sli_hba_iocb_abort - Abort all iocbs to an hba.
  * @phba: pointer to lpfc HBA data structure.
  *
@@ -9835,7 +9895,7 @@ lpfc_sli_hba_iocb_abort(struct lpfc_hba *phba)
 
 	for (i = 0; i < psli->num_rings; i++) {
 		pring = &psli->ring[i];
-		lpfc_sli_iocb_ring_abort(phba, pring);
+		lpfc_sli_abort_iocb_ring(phba, pring);
 	}
 }
 
@@ -10060,6 +10120,124 @@ lpfc_sli_abort_iocb(struct lpfc_vport *vport, struct lpfc_sli_ring *pring,
 }
 
 /**
+ * lpfc_sli_abort_taskmgmt - issue abort for all commands on a host/target/LUN
+ * @vport: Pointer to virtual port.
+ * @pring: Pointer to driver SLI ring object.
+ * @tgt_id: SCSI ID of the target.
+ * @lun_id: LUN ID of the scsi device.
+ * @taskmgmt_cmd: LPFC_CTX_LUN/LPFC_CTX_TGT/LPFC_CTX_HOST.
+ *
+ * This function sends an abort command for every SCSI command
+ * associated with the given virtual port pending on the ring
+ * filtered by lpfc_sli_validate_fcp_iocb function.
+ * When taskmgmt_cmd == LPFC_CTX_LUN, the function sends abort only to the
+ * FCP iocbs associated with lun specified by tgt_id and lun_id
+ * parameters
+ * When taskmgmt_cmd == LPFC_CTX_TGT, the function sends abort only to the
+ * FCP iocbs associated with SCSI target specified by tgt_id parameter.
+ * When taskmgmt_cmd == LPFC_CTX_HOST, the function sends abort to all
+ * FCP iocbs associated with virtual port.
+ * This function returns number of iocbs it aborted .
+ * This function is called with no locks held right after a taskmgmt
+ * command is sent.
+ **/
+int
+lpfc_sli_abort_taskmgmt(struct lpfc_vport *vport, struct lpfc_sli_ring *pring,
+			uint16_t tgt_id, uint64_t lun_id, lpfc_ctx_cmd cmd)
+{
+	struct lpfc_hba *phba = vport->phba;
+	struct lpfc_iocbq *abtsiocbq;
+	struct lpfc_iocbq *iocbq;
+	IOCB_t *icmd;
+	int sum, i, ret_val;
+	unsigned long iflags;
+	struct lpfc_sli_ring *pring_s4;
+	uint32_t ring_number;
+
+	spin_lock_irq(&phba->hbalock);
+
+	/* all I/Os are in process of being flushed */
+	if (phba->hba_flag & HBA_FCP_IOQ_FLUSH) {
+		spin_unlock_irq(&phba->hbalock);
+		return 0;
+	}
+	sum = 0;
+
+	for (i = 1; i <= phba->sli.last_iotag; i++) {
+		iocbq = phba->sli.iocbq_lookup[i];
+
+		if (lpfc_sli_validate_fcp_iocb(iocbq, vport, tgt_id, lun_id,
+					       cmd) != 0)
+			continue;
+
+		/*
+		 * If the iocbq is already being aborted, don't take a second
+		 * action, but do count it.
+		 */
+		if (iocbq->iocb_flag & LPFC_DRIVER_ABORTED)
+			continue;
+
+		/* issue ABTS for this IOCB based on iotag */
+		abtsiocbq = __lpfc_sli_get_iocbq(phba);
+		if (abtsiocbq == NULL)
+			continue;
+
+		icmd = &iocbq->iocb;
+		abtsiocbq->iocb.un.acxri.abortType = ABORT_TYPE_ABTS;
+		abtsiocbq->iocb.un.acxri.abortContextTag = icmd->ulpContext;
+		if (phba->sli_rev == LPFC_SLI_REV4)
+			abtsiocbq->iocb.un.acxri.abortIoTag =
+							 iocbq->sli4_xritag;
+		else
+			abtsiocbq->iocb.un.acxri.abortIoTag = icmd->ulpIoTag;
+		abtsiocbq->iocb.ulpLe = 1;
+		abtsiocbq->iocb.ulpClass = icmd->ulpClass;
+		abtsiocbq->vport = vport;
+
+		/* ABTS WQE must go to the same WQ as the WQE to be aborted */
+		abtsiocbq->fcp_wqidx = iocbq->fcp_wqidx;
+		if (iocbq->iocb_flag & LPFC_IO_FCP)
+			abtsiocbq->iocb_flag |= LPFC_USE_FCPWQIDX;
+
+		if (lpfc_is_link_up(phba))
+			abtsiocbq->iocb.ulpCommand = CMD_ABORT_XRI_CN;
+		else
+			abtsiocbq->iocb.ulpCommand = CMD_CLOSE_XRI_CN;
+
+		/* Setup callback routine and issue the command. */
+		abtsiocbq->iocb_cmpl = lpfc_sli_abort_fcp_cmpl;
+
+		/*
+		 * Indicate the IO is being aborted by the driver and set
+		 * the caller's flag into the aborted IO.
+		 */
+		iocbq->iocb_flag |= LPFC_DRIVER_ABORTED;
+
+		if (phba->sli_rev == LPFC_SLI_REV4) {
+			ring_number = MAX_SLI3_CONFIGURED_RINGS +
+					 iocbq->fcp_wqidx;
+			pring_s4 = &phba->sli.ring[ring_number];
+			/* Note: both hbalock and ring_lock must be set here */
+			spin_lock_irqsave(&pring_s4->ring_lock, iflags);
+			ret_val = __lpfc_sli_issue_iocb(phba, pring_s4->ringno,
+							abtsiocbq, 0);
+			spin_unlock_irqrestore(&pring_s4->ring_lock, iflags);
+		} else {
+			ret_val = __lpfc_sli_issue_iocb(phba, pring->ringno,
+							abtsiocbq, 0);
+		}
+
+
+		if (ret_val == IOCB_ERROR)
+			__lpfc_sli_release_iocbq(phba, abtsiocbq);
+		else
+			sum++;
+	}
+	spin_unlock_irq(&phba->hbalock);
+	return sum;
+}
+
+/**
  * lpfc_sli_wake_iocb_wait - lpfc_sli_issue_iocb_wait's completion handler
  * @phba: Pointer to HBA context object.
  * @cmdiocbq: Pointer to command iocb.
diff --git a/drivers/scsi/lpfc/lpfc_sli.h b/drivers/scsi/lpfc/lpfc_sli.h
index 6f04080f4ea8..edb48832c39b 100644
--- a/drivers/scsi/lpfc/lpfc_sli.h
+++ b/drivers/scsi/lpfc/lpfc_sli.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_sli4.h b/drivers/scsi/lpfc/lpfc_sli4.h
index 9b8cda866176..7f50aa04d66a 100644
--- a/drivers/scsi/lpfc/lpfc_sli4.h
+++ b/drivers/scsi/lpfc/lpfc_sli4.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2009-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2009-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
diff --git a/drivers/scsi/lpfc/lpfc_version.h b/drivers/scsi/lpfc/lpfc_version.h
index e32cbec70324..41675c1193e7 100644
--- a/drivers/scsi/lpfc/lpfc_version.h
+++ b/drivers/scsi/lpfc/lpfc_version.h
@@ -1,7 +1,7 @@
 /*******************************************************************
  * This file is part of the Emulex Linux Device Driver for         *
  * Fibre Channel Host Bus Adapters.                                *
- * Copyright (C) 2004-2013 Emulex.  All rights reserved.           *
+ * Copyright (C) 2004-2014 Emulex.  All rights reserved.           *
  * EMULEX and SLI are trademarks of Emulex.                        *
  * www.emulex.com                                                  *
  *                                                                 *
@@ -18,7 +18,7 @@
  * included with this package.                                     *
  *******************************************************************/
 
-#define LPFC_DRIVER_VERSION "8.3.45"
+#define LPFC_DRIVER_VERSION "10.2.8001.0."
 #define LPFC_DRIVER_NAME		"lpfc"
 
 /* Used for SLI 2/3 */
@@ -30,4 +30,4 @@
 
 #define LPFC_MODULE_DESC "Emulex LightPulse Fibre Channel SCSI driver " \
 		LPFC_DRIVER_VERSION
-#define LPFC_COPYRIGHT "Copyright(c) 2004-2013 Emulex.  All rights reserved."
+#define LPFC_COPYRIGHT "Copyright(c) 2004-2014 Emulex.  All rights reserved."
diff --git a/drivers/scsi/mac_scsi.c b/drivers/scsi/mac_scsi.c
index f5cdc68cd5b6..6a039eb1cbce 100644
--- a/drivers/scsi/mac_scsi.c
+++ b/drivers/scsi/mac_scsi.c
@@ -25,10 +25,6 @@
  * 1+ (800) 334-5454
  */
 
-/*
- * $Log: mac_NCR5380.c,v $
- */
-
 #include <linux/types.h>
 #include <linux/stddef.h>
 #include <linux/ctype.h>
@@ -58,12 +54,6 @@
 
 #include "NCR5380.h"
 
-#if 0
-#define NDEBUG (NDEBUG_INTR | NDEBUG_PSEUDO_DMA | NDEBUG_ARBITRATION | NDEBUG_SELECTION | NDEBUG_RESELECTION)
-#else
-#define NDEBUG (NDEBUG_ABORT)
-#endif
-
 #define RESET_BOOT
 #define DRIVER_SETUP
 
diff --git a/drivers/scsi/mac_scsi.h b/drivers/scsi/mac_scsi.h
index 7dc62fce1c4c..06969b06e54b 100644
--- a/drivers/scsi/mac_scsi.h
+++ b/drivers/scsi/mac_scsi.h
@@ -22,10 +22,6 @@
  * 1+ (800) 334-5454
  */
 
-/*
- * $Log: cumana_NCR5380.h,v $
- */
-
 #ifndef MAC_NCR5380_H
 #define MAC_NCR5380_H
 
@@ -51,8 +47,6 @@
 
 #include <scsi/scsicam.h>
 
-#ifndef HOSTS_C
-
 #define NCR5380_implementation_fields \
     int port, ctrl
 
@@ -75,10 +69,6 @@
 #define NCR5380_show_info macscsi_show_info
 #define NCR5380_write_info macscsi_write_info
 
-#define BOARD_NORMAL	0
-#define BOARD_NCR53C400	1
-
-#endif /* ndef HOSTS_C */
 #endif /* ndef ASM */
 #endif /* MAC_NCR5380_H */
 
diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c
index d84d02c2aad9..112799b131a9 100644
--- a/drivers/scsi/megaraid/megaraid_sas_base.c
+++ b/drivers/scsi/megaraid/megaraid_sas_base.c
@@ -3061,7 +3061,8 @@ megasas_transition_to_ready(struct megasas_instance *instance, int ocr)
 	u32 cur_state;
 	u32 abs_state, curr_abs_state;
 
-	fw_state = instance->instancet->read_fw_status_reg(instance->reg_set) & MFI_STATE_MASK;
+	abs_state = instance->instancet->read_fw_status_reg(instance->reg_set);
+	fw_state = abs_state & MFI_STATE_MASK;
 
 	if (fw_state != MFI_STATE_READY)
 		printk(KERN_INFO "megasas: Waiting for FW to come to ready"
@@ -3069,9 +3070,6 @@ megasas_transition_to_ready(struct megasas_instance *instance, int ocr)
 
 	while (fw_state != MFI_STATE_READY) {
 
-		abs_state =
-		instance->instancet->read_fw_status_reg(instance->reg_set);
-
 		switch (fw_state) {
 
 		case MFI_STATE_FAULT:
@@ -3223,10 +3221,8 @@ megasas_transition_to_ready(struct megasas_instance *instance, int ocr)
 		 * The cur_state should not last for more than max_wait secs
 		 */
 		for (i = 0; i < (max_wait * 1000); i++) {
-			fw_state = instance->instancet->read_fw_status_reg(instance->reg_set) &
-					MFI_STATE_MASK ;
-		curr_abs_state =
-		instance->instancet->read_fw_status_reg(instance->reg_set);
+			curr_abs_state = instance->instancet->
+				read_fw_status_reg(instance->reg_set);
 
 			if (abs_state == curr_abs_state) {
 				msleep(1);
@@ -3242,6 +3238,9 @@ megasas_transition_to_ready(struct megasas_instance *instance, int ocr)
 			       "in %d secs\n", fw_state, max_wait);
 			return -ENODEV;
 		}
+
+		abs_state = curr_abs_state;
+		fw_state = curr_abs_state & MFI_STATE_MASK;
 	}
 	printk(KERN_INFO "megasas: FW now in Ready state\n");
 
diff --git a/drivers/scsi/mpt2sas/mpt2sas_base.c b/drivers/scsi/mpt2sas/mpt2sas_base.c
index bde63f7452bd..8b88118e20e6 100644
--- a/drivers/scsi/mpt2sas/mpt2sas_base.c
+++ b/drivers/scsi/mpt2sas/mpt2sas_base.c
@@ -1739,14 +1739,14 @@ mpt2sas_base_free_smid(struct MPT2SAS_ADAPTER *ioc, u16 smid)
 			list_for_each_entry_safe(chain_req, next,
 			    &ioc->scsi_lookup[i].chain_list, tracker_list) {
 				list_del_init(&chain_req->tracker_list);
-				list_add_tail(&chain_req->tracker_list,
+				list_add(&chain_req->tracker_list,
 				    &ioc->free_chain_list);
 			}
 		}
 		ioc->scsi_lookup[i].cb_idx = 0xFF;
 		ioc->scsi_lookup[i].scmd = NULL;
 		ioc->scsi_lookup[i].direct_io = 0;
-		list_add_tail(&ioc->scsi_lookup[i].tracker_list,
+		list_add(&ioc->scsi_lookup[i].tracker_list,
 		    &ioc->free_list);
 		spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
 
@@ -1764,13 +1764,13 @@ mpt2sas_base_free_smid(struct MPT2SAS_ADAPTER *ioc, u16 smid)
 		/* hi-priority */
 		i = smid - ioc->hi_priority_smid;
 		ioc->hpr_lookup[i].cb_idx = 0xFF;
-		list_add_tail(&ioc->hpr_lookup[i].tracker_list,
+		list_add(&ioc->hpr_lookup[i].tracker_list,
 		    &ioc->hpr_free_list);
 	} else if (smid <= ioc->hba_queue_depth) {
 		/* internal queue */
 		i = smid - ioc->internal_smid;
 		ioc->internal_lookup[i].cb_idx = 0xFF;
-		list_add_tail(&ioc->internal_lookup[i].tracker_list,
+		list_add(&ioc->internal_lookup[i].tracker_list,
 		    &ioc->internal_free_list);
 	}
 	spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
diff --git a/drivers/scsi/mpt2sas/mpt2sas_base.h b/drivers/scsi/mpt2sas/mpt2sas_base.h
index 1f2ac3a28621..fd3b998c75b1 100644
--- a/drivers/scsi/mpt2sas/mpt2sas_base.h
+++ b/drivers/scsi/mpt2sas/mpt2sas_base.h
@@ -1065,7 +1065,7 @@ void mpt2sas_scsih_event_callback(struct MPT2SAS_ADAPTER *ioc, u8 msix_index,
     u32 reply);
 int mpt2sas_scsih_issue_tm(struct MPT2SAS_ADAPTER *ioc, u16 handle,
 	uint channel, uint id, uint lun, u8 type, u16 smid_task,
-	ulong timeout, unsigned long serial_number, enum mutex_type m_type);
+	ulong timeout, enum mutex_type m_type);
 void mpt2sas_scsih_set_tm_flag(struct MPT2SAS_ADAPTER *ioc, u16 handle);
 void mpt2sas_scsih_clear_tm_flag(struct MPT2SAS_ADAPTER *ioc, u16 handle);
 void mpt2sas_expander_remove(struct MPT2SAS_ADAPTER *ioc, u64 sas_address);
diff --git a/drivers/scsi/mpt2sas/mpt2sas_ctl.c b/drivers/scsi/mpt2sas/mpt2sas_ctl.c
index b7f887c9b0bf..62df8f9d4271 100644
--- a/drivers/scsi/mpt2sas/mpt2sas_ctl.c
+++ b/drivers/scsi/mpt2sas/mpt2sas_ctl.c
@@ -987,7 +987,7 @@ _ctl_do_mpt_command(struct MPT2SAS_ADAPTER *ioc, struct mpt2_ioctl_command karg,
 			mpt2sas_scsih_issue_tm(ioc,
 			    le16_to_cpu(mpi_request->FunctionDependent1), 0, 0,
 			    0, MPI2_SCSITASKMGMT_TASKTYPE_TARGET_RESET, 0, 10,
-			    0, TM_MUTEX_ON);
+			    TM_MUTEX_ON);
 			ioc->tm_cmds.status = MPT2_CMD_NOT_USED;
 		} else
 			mpt2sas_base_hard_reset_handler(ioc, CAN_SLEEP,
diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c
index 6fd7d40b2c4d..5055f925d2cd 100644
--- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c
+++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c
@@ -2368,7 +2368,6 @@ mpt2sas_scsih_clear_tm_flag(struct MPT2SAS_ADAPTER *ioc, u16 handle)
  * @type: MPI2_SCSITASKMGMT_TASKTYPE__XXX (defined in mpi2_init.h)
  * @smid_task: smid assigned to the task
  * @timeout: timeout in seconds
- * @serial_number: the serial_number from scmd
  * @m_type: TM_MUTEX_ON or TM_MUTEX_OFF
  * Context: user
  *
@@ -2381,7 +2380,7 @@ mpt2sas_scsih_clear_tm_flag(struct MPT2SAS_ADAPTER *ioc, u16 handle)
 int
 mpt2sas_scsih_issue_tm(struct MPT2SAS_ADAPTER *ioc, u16 handle, uint channel,
     uint id, uint lun, u8 type, u16 smid_task, ulong timeout,
-	unsigned long serial_number, enum mutex_type m_type)
+	enum mutex_type m_type)
 {
 	Mpi2SCSITaskManagementRequest_t *mpi_request;
 	Mpi2SCSITaskManagementReply_t *mpi_reply;
@@ -2634,8 +2633,7 @@ _scsih_abort(struct scsi_cmnd *scmd)
 	handle = sas_device_priv_data->sas_target->handle;
 	r = mpt2sas_scsih_issue_tm(ioc, handle, scmd->device->channel,
 	    scmd->device->id, scmd->device->lun,
-	    MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30,
-	    scmd->serial_number, TM_MUTEX_ON);
+	    MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30, TM_MUTEX_ON);
 
  out:
 	sdev_printk(KERN_INFO, scmd->device, "task abort: %s scmd(%p)\n",
@@ -2696,8 +2694,7 @@ _scsih_dev_reset(struct scsi_cmnd *scmd)
 
 	r = mpt2sas_scsih_issue_tm(ioc, handle, scmd->device->channel,
 	    scmd->device->id, scmd->device->lun,
-	    MPI2_SCSITASKMGMT_TASKTYPE_LOGICAL_UNIT_RESET, 0, 30, 0,
-	    TM_MUTEX_ON);
+	    MPI2_SCSITASKMGMT_TASKTYPE_LOGICAL_UNIT_RESET, 0, 30, TM_MUTEX_ON);
 
  out:
 	sdev_printk(KERN_INFO, scmd->device, "device reset: %s scmd(%p)\n",
@@ -2757,7 +2754,7 @@ _scsih_target_reset(struct scsi_cmnd *scmd)
 
 	r = mpt2sas_scsih_issue_tm(ioc, handle, scmd->device->channel,
 	    scmd->device->id, 0, MPI2_SCSITASKMGMT_TASKTYPE_TARGET_RESET, 0,
-	    30, 0, TM_MUTEX_ON);
+	    30, TM_MUTEX_ON);
 
  out:
 	starget_printk(KERN_INFO, starget, "target reset: %s scmd(%p)\n",
@@ -3953,9 +3950,9 @@ _scsih_setup_direct_io(struct MPT2SAS_ADAPTER *ioc, struct scsi_cmnd *scmd,
  * SCSI_MLQUEUE_HOST_BUSY if the entire host queue is full
  */
 static int
-_scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
+_scsih_qcmd(struct Scsi_Host *shost, struct scsi_cmnd *scmd)
 {
-	struct MPT2SAS_ADAPTER *ioc = shost_priv(scmd->device->host);
+	struct MPT2SAS_ADAPTER *ioc = shost_priv(shost);
 	struct MPT2SAS_DEVICE *sas_device_priv_data;
 	struct MPT2SAS_TARGET *sas_target_priv_data;
 	struct _raid_device *raid_device;
@@ -3963,7 +3960,6 @@ _scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
 	u32 mpi_control;
 	u16 smid;
 
-	scmd->scsi_done = done;
 	sas_device_priv_data = scmd->device->hostdata;
 	if (!sas_device_priv_data || !sas_device_priv_data->sas_target) {
 		scmd->result = DID_NO_CONNECT << 16;
@@ -4039,7 +4035,7 @@ _scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
 	    MPT_TARGET_FLAGS_RAID_COMPONENT)
 		mpi_request->Function = MPI2_FUNCTION_RAID_SCSI_IO_PASSTHROUGH;
 	else
-		mpi_request->Function = MPI2_FUNCTION_SCSI_IO_REQUEST;
+	mpi_request->Function = MPI2_FUNCTION_SCSI_IO_REQUEST;
 	mpi_request->DevHandle =
 	    cpu_to_le16(sas_device_priv_data->sas_target->handle);
 	mpi_request->DataLength = cpu_to_le32(scsi_bufflen(scmd));
@@ -4083,8 +4079,6 @@ _scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
 	return SCSI_MLQUEUE_HOST_BUSY;
 }
 
-static DEF_SCSI_QCMD(_scsih_qcmd)
-
 /**
  * _scsih_normalize_sense - normalize descriptor and fixed format sense data
  * @sense_buffer: sense data returned by target
@@ -5880,7 +5874,7 @@ broadcast_aen_retry:
 
 		spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
 		r = mpt2sas_scsih_issue_tm(ioc, handle, 0, 0, lun,
-		    MPI2_SCSITASKMGMT_TASKTYPE_QUERY_TASK, smid, 30, 0,
+		    MPI2_SCSITASKMGMT_TASKTYPE_QUERY_TASK, smid, 30,
 		    TM_MUTEX_OFF);
 		if (r == FAILED) {
 			sdev_printk(KERN_WARNING, sdev,
@@ -5922,7 +5916,7 @@ broadcast_aen_retry:
 
 		r = mpt2sas_scsih_issue_tm(ioc, handle, sdev->channel, sdev->id,
 		    sdev->lun, MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30,
-		    scmd->serial_number, TM_MUTEX_OFF);
+		    TM_MUTEX_OFF);
 		if (r == FAILED) {
 			sdev_printk(KERN_WARNING, sdev,
 			    "mpt2sas_scsih_issue_tm: ABORT_TASK: FAILED : "
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.h b/drivers/scsi/mpt3sas/mpt3sas_base.h
index 0ebf5d913c80..9b90a6fef706 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_base.h
+++ b/drivers/scsi/mpt3sas/mpt3sas_base.h
@@ -993,7 +993,7 @@ void mpt3sas_scsih_reset_handler(struct MPT3SAS_ADAPTER *ioc, int reset_phase);
 
 int mpt3sas_scsih_issue_tm(struct MPT3SAS_ADAPTER *ioc, u16 handle,
 	uint channel, uint id, uint lun, u8 type, u16 smid_task,
-	ulong timeout, unsigned long serial_number,  enum mutex_type m_type);
+	ulong timeout, enum mutex_type m_type);
 void mpt3sas_scsih_set_tm_flag(struct MPT3SAS_ADAPTER *ioc, u16 handle);
 void mpt3sas_scsih_clear_tm_flag(struct MPT3SAS_ADAPTER *ioc, u16 handle);
 void mpt3sas_expander_remove(struct MPT3SAS_ADAPTER *ioc, u64 sas_address);
diff --git a/drivers/scsi/mpt3sas/mpt3sas_ctl.c b/drivers/scsi/mpt3sas/mpt3sas_ctl.c
index 9b89de14a0a3..ba9cbe598a91 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_ctl.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_ctl.c
@@ -980,7 +980,7 @@ _ctl_do_mpt_command(struct MPT3SAS_ADAPTER *ioc, struct mpt3_ioctl_command karg,
 			mpt3sas_scsih_issue_tm(ioc,
 			    le16_to_cpu(mpi_request->FunctionDependent1), 0, 0,
 			    0, MPI2_SCSITASKMGMT_TASKTYPE_TARGET_RESET, 0, 30,
-			    0, TM_MUTEX_ON);
+			    TM_MUTEX_ON);
 		} else
 			mpt3sas_base_hard_reset_handler(ioc, CAN_SLEEP,
 			    FORCE_BIG_HAMMER);
diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c
index a961fe11b527..18e713db1d32 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c
@@ -2029,7 +2029,6 @@ mpt3sas_scsih_clear_tm_flag(struct MPT3SAS_ADAPTER *ioc, u16 handle)
  * @type: MPI2_SCSITASKMGMT_TASKTYPE__XXX (defined in mpi2_init.h)
  * @smid_task: smid assigned to the task
  * @timeout: timeout in seconds
- * @serial_number: the serial_number from scmd
  * @m_type: TM_MUTEX_ON or TM_MUTEX_OFF
  * Context: user
  *
@@ -2042,7 +2041,7 @@ mpt3sas_scsih_clear_tm_flag(struct MPT3SAS_ADAPTER *ioc, u16 handle)
 int
 mpt3sas_scsih_issue_tm(struct MPT3SAS_ADAPTER *ioc, u16 handle, uint channel,
 	uint id, uint lun, u8 type, u16 smid_task, ulong timeout,
-	unsigned long serial_number, enum mutex_type m_type)
+	enum mutex_type m_type)
 {
 	Mpi2SCSITaskManagementRequest_t *mpi_request;
 	Mpi2SCSITaskManagementReply_t *mpi_reply;
@@ -2293,8 +2292,7 @@ _scsih_abort(struct scsi_cmnd *scmd)
 	handle = sas_device_priv_data->sas_target->handle;
 	r = mpt3sas_scsih_issue_tm(ioc, handle, scmd->device->channel,
 	    scmd->device->id, scmd->device->lun,
-	    MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30,
-	    scmd->serial_number, TM_MUTEX_ON);
+	    MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30, TM_MUTEX_ON);
 
  out:
 	sdev_printk(KERN_INFO, scmd->device, "task abort: %s scmd(%p)\n",
@@ -2353,8 +2351,7 @@ _scsih_dev_reset(struct scsi_cmnd *scmd)
 
 	r = mpt3sas_scsih_issue_tm(ioc, handle, scmd->device->channel,
 	    scmd->device->id, scmd->device->lun,
-	    MPI2_SCSITASKMGMT_TASKTYPE_LOGICAL_UNIT_RESET, 0, 30, 0,
-	    TM_MUTEX_ON);
+	    MPI2_SCSITASKMGMT_TASKTYPE_LOGICAL_UNIT_RESET, 0, 30, TM_MUTEX_ON);
 
  out:
 	sdev_printk(KERN_INFO, scmd->device, "device reset: %s scmd(%p)\n",
@@ -2414,7 +2411,7 @@ _scsih_target_reset(struct scsi_cmnd *scmd)
 
 	r = mpt3sas_scsih_issue_tm(ioc, handle, scmd->device->channel,
 	    scmd->device->id, 0, MPI2_SCSITASKMGMT_TASKTYPE_TARGET_RESET, 0,
-	    30, 0, TM_MUTEX_ON);
+	    30, TM_MUTEX_ON);
 
  out:
 	starget_printk(KERN_INFO, starget, "target reset: %s scmd(%p)\n",
@@ -3518,7 +3515,7 @@ _scsih_eedp_error_handling(struct scsi_cmnd *scmd, u16 ioc_status)
 
 
 /**
- * _scsih_qcmd_lck - main scsi request entry point
+ * _scsih_qcmd - main scsi request entry point
  * @scmd: pointer to scsi command object
  * @done: function pointer to be invoked on completion
  *
@@ -3529,9 +3526,9 @@ _scsih_eedp_error_handling(struct scsi_cmnd *scmd, u16 ioc_status)
  * SCSI_MLQUEUE_HOST_BUSY if the entire host queue is full
  */
 static int
-_scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
+_scsih_qcmd(struct Scsi_Host *shost, struct scsi_cmnd *scmd)
 {
-	struct MPT3SAS_ADAPTER *ioc = shost_priv(scmd->device->host);
+	struct MPT3SAS_ADAPTER *ioc = shost_priv(shost);
 	struct MPT3SAS_DEVICE *sas_device_priv_data;
 	struct MPT3SAS_TARGET *sas_target_priv_data;
 	Mpi2SCSIIORequest_t *mpi_request;
@@ -3544,7 +3541,6 @@ _scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
 		scsi_print_command(scmd);
 #endif
 
-	scmd->scsi_done = done;
 	sas_device_priv_data = scmd->device->hostdata;
 	if (!sas_device_priv_data || !sas_device_priv_data->sas_target) {
 		scmd->result = DID_NO_CONNECT << 16;
@@ -3659,8 +3655,6 @@ _scsih_qcmd_lck(struct scsi_cmnd *scmd, void (*done)(struct scsi_cmnd *))
  out:
 	return SCSI_MLQUEUE_HOST_BUSY;
 }
-static DEF_SCSI_QCMD(_scsih_qcmd)
-
 
 /**
  * _scsih_normalize_sense - normalize descriptor and fixed format sense data
@@ -5425,7 +5419,7 @@ _scsih_sas_broadcast_primitive_event(struct MPT3SAS_ADAPTER *ioc,
 
 		spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
 		r = mpt3sas_scsih_issue_tm(ioc, handle, 0, 0, lun,
-		    MPI2_SCSITASKMGMT_TASKTYPE_QUERY_TASK, smid, 30, 0,
+		    MPI2_SCSITASKMGMT_TASKTYPE_QUERY_TASK, smid, 30,
 		    TM_MUTEX_OFF);
 		if (r == FAILED) {
 			sdev_printk(KERN_WARNING, sdev,
@@ -5467,7 +5461,7 @@ _scsih_sas_broadcast_primitive_event(struct MPT3SAS_ADAPTER *ioc,
 
 		r = mpt3sas_scsih_issue_tm(ioc, handle, sdev->channel, sdev->id,
 		    sdev->lun, MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30,
-		    scmd->serial_number, TM_MUTEX_OFF);
+		    TM_MUTEX_OFF);
 		if (r == FAILED) {
 			sdev_printk(KERN_WARNING, sdev,
 			    "mpt3sas_scsih_issue_tm: ABORT_TASK: FAILED : "
diff --git a/drivers/scsi/mvsas/mv_94xx.c b/drivers/scsi/mvsas/mv_94xx.c
index 1e4479f3331a..9270d15ff1a4 100644
--- a/drivers/scsi/mvsas/mv_94xx.c
+++ b/drivers/scsi/mvsas/mv_94xx.c
@@ -564,7 +564,7 @@ static void mvs_94xx_interrupt_enable(struct mvs_info *mvi)
 	u32 tmp;
 
 	tmp = mr32(MVS_GBL_CTL);
-	tmp |= (IRQ_SAS_A | IRQ_SAS_B);
+	tmp |= (MVS_IRQ_SAS_A | MVS_IRQ_SAS_B);
 	mw32(MVS_GBL_INT_STAT, tmp);
 	writel(tmp, regs + 0x0C);
 	writel(tmp, regs + 0x10);
@@ -580,7 +580,7 @@ static void mvs_94xx_interrupt_disable(struct mvs_info *mvi)
 
 	tmp = mr32(MVS_GBL_CTL);
 
-	tmp &= ~(IRQ_SAS_A | IRQ_SAS_B);
+	tmp &= ~(MVS_IRQ_SAS_A | MVS_IRQ_SAS_B);
 	mw32(MVS_GBL_INT_STAT, tmp);
 	writel(tmp, regs + 0x0C);
 	writel(tmp, regs + 0x10);
@@ -596,7 +596,7 @@ static u32 mvs_94xx_isr_status(struct mvs_info *mvi, int irq)
 	if (!(mvi->flags & MVF_FLAG_SOC)) {
 		stat = mr32(MVS_GBL_INT_STAT);
 
-		if (!(stat & (IRQ_SAS_A | IRQ_SAS_B)))
+		if (!(stat & (MVS_IRQ_SAS_A | MVS_IRQ_SAS_B)))
 			return 0;
 	}
 	return stat;
@@ -606,8 +606,8 @@ static irqreturn_t mvs_94xx_isr(struct mvs_info *mvi, int irq, u32 stat)
 {
 	void __iomem *regs = mvi->regs;
 
-	if (((stat & IRQ_SAS_A) && mvi->id == 0) ||
-			((stat & IRQ_SAS_B) && mvi->id == 1)) {
+	if (((stat & MVS_IRQ_SAS_A) && mvi->id == 0) ||
+			((stat & MVS_IRQ_SAS_B) && mvi->id == 1)) {
 		mw32_f(MVS_INT_STAT, CINT_DONE);
 
 		spin_lock(&mvi->lock);
diff --git a/drivers/scsi/mvsas/mv_94xx.h b/drivers/scsi/mvsas/mv_94xx.h
index 487aa6f97412..14e197497b46 100644
--- a/drivers/scsi/mvsas/mv_94xx.h
+++ b/drivers/scsi/mvsas/mv_94xx.h
@@ -150,35 +150,35 @@ enum chip_register_bits {
 
 enum pci_interrupt_cause {
 	/*  MAIN_IRQ_CAUSE (R10200) Bits*/
-	IRQ_COM_IN_I2O_IOP0            = (1 << 0),
-	IRQ_COM_IN_I2O_IOP1            = (1 << 1),
-	IRQ_COM_IN_I2O_IOP2            = (1 << 2),
-	IRQ_COM_IN_I2O_IOP3            = (1 << 3),
-	IRQ_COM_OUT_I2O_HOS0           = (1 << 4),
-	IRQ_COM_OUT_I2O_HOS1           = (1 << 5),
-	IRQ_COM_OUT_I2O_HOS2           = (1 << 6),
-	IRQ_COM_OUT_I2O_HOS3           = (1 << 7),
-	IRQ_PCIF_TO_CPU_DRBL0          = (1 << 8),
-	IRQ_PCIF_TO_CPU_DRBL1          = (1 << 9),
-	IRQ_PCIF_TO_CPU_DRBL2          = (1 << 10),
-	IRQ_PCIF_TO_CPU_DRBL3          = (1 << 11),
-	IRQ_PCIF_DRBL0                 = (1 << 12),
-	IRQ_PCIF_DRBL1                 = (1 << 13),
-	IRQ_PCIF_DRBL2                 = (1 << 14),
-	IRQ_PCIF_DRBL3                 = (1 << 15),
-	IRQ_XOR_A                      = (1 << 16),
-	IRQ_XOR_B                      = (1 << 17),
-	IRQ_SAS_A                      = (1 << 18),
-	IRQ_SAS_B                      = (1 << 19),
-	IRQ_CPU_CNTRL                  = (1 << 20),
-	IRQ_GPIO                       = (1 << 21),
-	IRQ_UART                       = (1 << 22),
-	IRQ_SPI                        = (1 << 23),
-	IRQ_I2C                        = (1 << 24),
-	IRQ_SGPIO                      = (1 << 25),
-	IRQ_COM_ERR                    = (1 << 29),
-	IRQ_I2O_ERR                    = (1 << 30),
-	IRQ_PCIE_ERR                   = (1 << 31),
+	MVS_IRQ_COM_IN_I2O_IOP0        = (1 << 0),
+	MVS_IRQ_COM_IN_I2O_IOP1        = (1 << 1),
+	MVS_IRQ_COM_IN_I2O_IOP2        = (1 << 2),
+	MVS_IRQ_COM_IN_I2O_IOP3        = (1 << 3),
+	MVS_IRQ_COM_OUT_I2O_HOS0       = (1 << 4),
+	MVS_IRQ_COM_OUT_I2O_HOS1       = (1 << 5),
+	MVS_IRQ_COM_OUT_I2O_HOS2       = (1 << 6),
+	MVS_IRQ_COM_OUT_I2O_HOS3       = (1 << 7),
+	MVS_IRQ_PCIF_TO_CPU_DRBL0      = (1 << 8),
+	MVS_IRQ_PCIF_TO_CPU_DRBL1      = (1 << 9),
+	MVS_IRQ_PCIF_TO_CPU_DRBL2      = (1 << 10),
+	MVS_IRQ_PCIF_TO_CPU_DRBL3      = (1 << 11),
+	MVS_IRQ_PCIF_DRBL0             = (1 << 12),
+	MVS_IRQ_PCIF_DRBL1             = (1 << 13),
+	MVS_IRQ_PCIF_DRBL2             = (1 << 14),
+	MVS_IRQ_PCIF_DRBL3             = (1 << 15),
+	MVS_IRQ_XOR_A                  = (1 << 16),
+	MVS_IRQ_XOR_B                  = (1 << 17),
+	MVS_IRQ_SAS_A                  = (1 << 18),
+	MVS_IRQ_SAS_B                  = (1 << 19),
+	MVS_IRQ_CPU_CNTRL              = (1 << 20),
+	MVS_IRQ_GPIO                   = (1 << 21),
+	MVS_IRQ_UART                   = (1 << 22),
+	MVS_IRQ_SPI                    = (1 << 23),
+	MVS_IRQ_I2C                    = (1 << 24),
+	MVS_IRQ_SGPIO                  = (1 << 25),
+	MVS_IRQ_COM_ERR                = (1 << 29),
+	MVS_IRQ_I2O_ERR                = (1 << 30),
+	MVS_IRQ_PCIE_ERR               = (1 << 31),
 };
 
 union reg_phy_cfg {
diff --git a/drivers/scsi/mvsas/mv_init.c b/drivers/scsi/mvsas/mv_init.c
index 5ff978be249d..eacee48a955c 100644
--- a/drivers/scsi/mvsas/mv_init.c
+++ b/drivers/scsi/mvsas/mv_init.c
@@ -728,6 +728,15 @@ static struct pci_device_id mvs_pci_table[] = {
 		.class_mask	= 0,
 		.driver_data	= chip_9485,
 	},
+	{
+		.vendor		= PCI_VENDOR_ID_MARVELL_EXT,
+		.device		= 0x9485,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= 0x9485,
+		.class		= 0,
+		.class_mask	= 0,
+		.driver_data	= chip_9485,
+	},
 	{ PCI_VDEVICE(OCZ, 0x1021), chip_9485}, /* OCZ RevoDrive3 */
 	{ PCI_VDEVICE(OCZ, 0x1022), chip_9485}, /* OCZ RevoDrive3/zDriveR4 (exact model unknown) */
 	{ PCI_VDEVICE(OCZ, 0x1040), chip_9485}, /* OCZ RevoDrive3/zDriveR4 (exact model unknown) */
diff --git a/drivers/scsi/osd/osd_initiator.c b/drivers/scsi/osd/osd_initiator.c
index bac04c2335aa..5f4cbf0c4759 100644
--- a/drivers/scsi/osd/osd_initiator.c
+++ b/drivers/scsi/osd/osd_initiator.c
@@ -1570,6 +1570,7 @@ static struct request *_make_request(struct request_queue *q, bool has_write,
 		if (unlikely(!req))
 			return ERR_PTR(-ENOMEM);
 
+		blk_rq_set_block_pc(req);
 		return req;
 	}
 }
@@ -1590,7 +1591,6 @@ static int _init_blk_request(struct osd_request *or,
 	}
 
 	or->request = req;
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= REQ_QUIET;
 
 	req->timeout = or->timeout;
@@ -1608,7 +1608,7 @@ static int _init_blk_request(struct osd_request *or,
 				ret = PTR_ERR(req);
 				goto out;
 			}
-			req->cmd_type = REQ_TYPE_BLOCK_PC;
+			blk_rq_set_block_pc(req);
 			or->in.req = or->request->next_rq = req;
 		}
 	} else if (has_in)
diff --git a/drivers/scsi/osst.c b/drivers/scsi/osst.c
index 21883a2d6324..0727ea7cc387 100644
--- a/drivers/scsi/osst.c
+++ b/drivers/scsi/osst.c
@@ -365,7 +365,7 @@ static int osst_execute(struct osst_request *SRpnt, const unsigned char *cmd,
 	if (!req)
 		return DRIVER_ERROR << 24;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_QUIET;
 
 	SRpnt->bio = NULL;
diff --git a/drivers/scsi/pas16.h b/drivers/scsi/pas16.h
index 3721342835e9..aa528f53c533 100644
--- a/drivers/scsi/pas16.h
+++ b/drivers/scsi/pas16.h
@@ -129,8 +129,6 @@ static int pas16_bus_reset(Scsi_Cmnd *);
 #define CAN_QUEUE 32 
 #endif
 
-#ifndef HOSTS_C
-
 #define NCR5380_implementation_fields \
     volatile unsigned short io_port
 
@@ -171,6 +169,5 @@ static int pas16_bus_reset(Scsi_Cmnd *);
    
 #define PAS16_IRQS 0xd4a8 
 
-#endif /* else def HOSTS_C */
 #endif /* ndef ASM */
 #endif /* PAS16_H */
diff --git a/drivers/scsi/pm8001/pm8001_ctl.c b/drivers/scsi/pm8001/pm8001_ctl.c
index 28b4e8139153..a368d77b8d41 100644
--- a/drivers/scsi/pm8001/pm8001_ctl.c
+++ b/drivers/scsi/pm8001/pm8001_ctl.c
@@ -395,6 +395,8 @@ static ssize_t pm8001_ctl_bios_version_show(struct device *cdev,
 	payload.offset = 0;
 	payload.length = 4096;
 	payload.func_specific = kzalloc(4096, GFP_KERNEL);
+	if (!payload.func_specific)
+		return -ENOMEM;
 	PM8001_CHIP_DISP->get_nvmd_req(pm8001_ha, &payload);
 	wait_for_completion(&completion);
 	virt_addr = pm8001_ha->memoryMap.region[NVMD].virt_ptr;
@@ -402,6 +404,7 @@ static ssize_t pm8001_ctl_bios_version_show(struct device *cdev,
 		bios_index++)
 		str += sprintf(str, "%c",
 			*((u8 *)((u8 *)virt_addr+bios_index)));
+	kfree(payload.func_specific);
 	return str - buf;
 }
 static DEVICE_ATTR(bios_version, S_IRUGO, pm8001_ctl_bios_version_show, NULL);
@@ -729,7 +732,7 @@ static ssize_t pm8001_show_update_fw(struct device *cdev,
 			flash_error_table[i].reason);
 }
 
-static DEVICE_ATTR(update_fw, S_IRUGO|S_IWUGO,
+static DEVICE_ATTR(update_fw, S_IRUGO|S_IWUSR|S_IWGRP,
 	pm8001_show_update_fw, pm8001_store_update_fw);
 struct device_attribute *pm8001_host_attrs[] = {
 	&dev_attr_interface_rev,
diff --git a/drivers/scsi/pm8001/pm8001_init.c b/drivers/scsi/pm8001/pm8001_init.c
index c4f31b21feb8..e90c89f1d480 100644
--- a/drivers/scsi/pm8001/pm8001_init.c
+++ b/drivers/scsi/pm8001/pm8001_init.c
@@ -677,7 +677,7 @@ static void pm8001_init_sas_add(struct pm8001_hba_info *pm8001_ha)
  * pm8001_get_phy_settings_info : Read phy setting values.
  * @pm8001_ha : our hba.
  */
-void pm8001_get_phy_settings_info(struct pm8001_hba_info *pm8001_ha)
+static int pm8001_get_phy_settings_info(struct pm8001_hba_info *pm8001_ha)
 {
 
 #ifdef PM8001_READ_VPD
@@ -691,11 +691,15 @@ void pm8001_get_phy_settings_info(struct pm8001_hba_info *pm8001_ha)
 	payload.offset = 0;
 	payload.length = 4096;
 	payload.func_specific = kzalloc(4096, GFP_KERNEL);
+	if (!payload.func_specific)
+		return -ENOMEM;
 	/* Read phy setting values from flash */
 	PM8001_CHIP_DISP->get_nvmd_req(pm8001_ha, &payload);
 	wait_for_completion(&completion);
 	pm8001_set_phy_profile(pm8001_ha, sizeof(u8), payload.func_specific);
+	kfree(payload.func_specific);
 #endif
+	return 0;
 }
 
 #ifdef PM8001_USE_MSIX
@@ -879,8 +883,11 @@ static int pm8001_pci_probe(struct pci_dev *pdev,
 	pm8001_init_sas_add(pm8001_ha);
 	/* phy setting support for motherboard controller */
 	if (pdev->subsystem_vendor != PCI_VENDOR_ID_ADAPTEC2 &&
-		pdev->subsystem_vendor != 0)
-		pm8001_get_phy_settings_info(pm8001_ha);
+		pdev->subsystem_vendor != 0) {
+		rc = pm8001_get_phy_settings_info(pm8001_ha);
+		if (rc)
+			goto err_out_shost;
+	}
 	pm8001_post_sas_ha_init(shost, chip);
 	rc = sas_register_ha(SHOST_TO_SAS_HA(shost));
 	if (rc)
diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index 07befcf365b8..16fe5196e6d9 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -664,7 +664,7 @@ do_read:
 		}
 
 		rval = qla2x00_read_sfp(vha, ha->sfp_data_dma, ha->sfp_data,
-		    addr, offset, SFP_BLOCK_SIZE, 0);
+		    addr, offset, SFP_BLOCK_SIZE, BIT_1);
 		if (rval != QLA_SUCCESS) {
 			ql_log(ql_log_warn, vha, 0x706d,
 			    "Unable to read SFP data (%x/%x/%x).\n", rval,
@@ -1495,7 +1495,7 @@ qla2x00_fw_dump_size_show(struct device *dev, struct device_attribute *attr,
 
 	if (!ha->fw_dumped)
 		size = 0;
-	else if (IS_QLA82XX(ha))
+	else if (IS_P3P_TYPE(ha))
 		size = ha->md_template_size + ha->md_dump_size;
 	else
 		size = ha->fw_dump_len;
diff --git a/drivers/scsi/qla2xxx/qla_bsg.c b/drivers/scsi/qla2xxx/qla_bsg.c
index 71ff340f6de4..524f9eb7fcd1 100644
--- a/drivers/scsi/qla2xxx/qla_bsg.c
+++ b/drivers/scsi/qla2xxx/qla_bsg.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2012 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -2054,9 +2054,49 @@ qla26xx_serdes_op(struct fc_bsg_job *bsg_job)
 		bsg_job->reply->reply_payload_rcv_len = sizeof(sr);
 		break;
 	default:
-		ql_log(ql_log_warn, vha, 0x708c,
+		ql_dbg(ql_dbg_user, vha, 0x708c,
 		    "Unknown serdes cmd %x.\n", sr.cmd);
-		rval = -EDOM;
+		rval = -EINVAL;
+		break;
+	}
+
+	bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] =
+	    rval ? EXT_STATUS_MAILBOX : 0;
+
+	bsg_job->reply_len = sizeof(struct fc_bsg_reply);
+	bsg_job->reply->result = DID_OK << 16;
+	bsg_job->job_done(bsg_job);
+	return 0;
+}
+
+static int
+qla8044_serdes_op(struct fc_bsg_job *bsg_job)
+{
+	struct Scsi_Host *host = bsg_job->shost;
+	scsi_qla_host_t *vha = shost_priv(host);
+	int rval = 0;
+	struct qla_serdes_reg_ex sr;
+
+	memset(&sr, 0, sizeof(sr));
+
+	sg_copy_to_buffer(bsg_job->request_payload.sg_list,
+	    bsg_job->request_payload.sg_cnt, &sr, sizeof(sr));
+
+	switch (sr.cmd) {
+	case INT_SC_SERDES_WRITE_REG:
+		rval = qla8044_write_serdes_word(vha, sr.addr, sr.val);
+		bsg_job->reply->reply_payload_rcv_len = 0;
+		break;
+	case INT_SC_SERDES_READ_REG:
+		rval = qla8044_read_serdes_word(vha, sr.addr, &sr.val);
+		sg_copy_from_buffer(bsg_job->reply_payload.sg_list,
+		    bsg_job->reply_payload.sg_cnt, &sr, sizeof(sr));
+		bsg_job->reply->reply_payload_rcv_len = sizeof(sr);
+		break;
+	default:
+		ql_dbg(ql_dbg_user, vha, 0x70cf,
+		    "Unknown serdes cmd %x.\n", sr.cmd);
+		rval = -EINVAL;
 		break;
 	}
 
@@ -2121,6 +2161,9 @@ qla2x00_process_vendor_specific(struct fc_bsg_job *bsg_job)
 	case QL_VND_SERDES_OP:
 		return qla26xx_serdes_op(bsg_job);
 
+	case QL_VND_SERDES_OP_EX:
+		return qla8044_serdes_op(bsg_job);
+
 	default:
 		return -ENOSYS;
 	}
diff --git a/drivers/scsi/qla2xxx/qla_bsg.h b/drivers/scsi/qla2xxx/qla_bsg.h
index e5c2126221e9..d38f9efa56fa 100644
--- a/drivers/scsi/qla2xxx/qla_bsg.h
+++ b/drivers/scsi/qla2xxx/qla_bsg.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -24,6 +24,7 @@
 #define QL_VND_READ_I2C		0x11
 #define QL_VND_FX00_MGMT_CMD	0x12
 #define QL_VND_SERDES_OP	0x13
+#define	QL_VND_SERDES_OP_EX	0x14
 
 /* BSG Vendor specific subcode returns */
 #define EXT_STATUS_OK			0
@@ -225,4 +226,10 @@ struct qla_serdes_reg {
 	uint16_t val;
 } __packed;
 
+struct qla_serdes_reg_ex {
+	uint16_t cmd;
+	uint32_t addr;
+	uint32_t val;
+} __packed;
+
 #endif
diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c
index 97255f7c3975..c72ee97bf3f7 100644
--- a/drivers/scsi/qla2xxx/qla_dbg.c
+++ b/drivers/scsi/qla2xxx/qla_dbg.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -15,7 +15,7 @@
  * |                              |                    | 0x0144,0x0146	|
  * |                              |                    | 0x015b-0x0160	|
  * |                              |                    | 0x016e-0x0170	|
- * | Mailbox commands             |       0x1187       | 0x1018-0x1019	|
+ * | Mailbox commands             |       0x118d       | 0x1018-0x1019	|
  * |                              |                    | 0x10ca         |
  * |                              |                    | 0x1115-0x1116  |
  * |                              |                    | 0x111a-0x111b	|
@@ -45,12 +45,16 @@
  * |                              |                    | 0x70ad-0x70ae  |
  * |                              |                    | 0x70d7-0x70db  |
  * |                              |                    | 0x70de-0x70df  |
- * | Task Management              |       0x803d       | 0x8025-0x8026  |
- * |                              |                    | 0x800b,0x8039  |
+ * | Task Management              |       0x803d       | 0x8000,0x800b  |
+ * |                              |                    | 0x8019         |
+ * |                              |                    | 0x8025,0x8026  |
+ * |                              |                    | 0x8031,0x8032  |
+ * |                              |                    | 0x8039,0x803c  |
  * | AER/EEH                      |       0x9011       |		|
  * | Virtual Port                 |       0xa007       |		|
- * | ISP82XX Specific             |       0xb14c       | 0xb002,0xb024  |
+ * | ISP82XX Specific             |       0xb157       | 0xb002,0xb024  |
  * |                              |                    | 0xb09e,0xb0ae  |
+ * |				  |		       | 0xb0c3,0xb0c6  |
  * |                              |                    | 0xb0e0-0xb0ef  |
  * |                              |                    | 0xb085,0xb0dc  |
  * |                              |                    | 0xb107,0xb108  |
@@ -60,12 +64,12 @@
  * |                              |                    | 0xb13c-0xb140  |
  * |                              |                    | 0xb149		|
  * | MultiQ                       |       0xc00c       |		|
- * | Misc                         |       0xd2ff       | 0xd017-0xd019	|
+ * | Misc                         |       0xd212       | 0xd017-0xd019	|
  * |                              |                    | 0xd020		|
- * |                              |                    | 0xd02e-0xd0ff	|
+ * |                              |                    | 0xd030-0xd0ff	|
  * |                              |                    | 0xd101-0xd1fe	|
- * |                              |                    | 0xd212-0xd2fe	|
- * | Target Mode		  |	  0xe070       | 0xe021		|
+ * |                              |                    | 0xd213-0xd2fe	|
+ * | Target Mode		  |	  0xe078       |		|
  * | Target Mode Management	  |	  0xf072       | 0xf002-0xf003	|
  * |                              |                    | 0xf046-0xf049  |
  * | Target Mode Task Management  |	  0x1000b      |		|
@@ -277,9 +281,15 @@ qla24xx_dump_memory(struct qla_hw_data *ha, uint32_t *code_ram,
 	if (rval != QLA_SUCCESS)
 		return rval;
 
+	set_bit(RISC_SRAM_DUMP_CMPL, &ha->fw_dump_cap_flags);
+
 	/* External Memory. */
-	return qla24xx_dump_ram(ha, 0x100000, *nxt,
+	rval = qla24xx_dump_ram(ha, 0x100000, *nxt,
 	    ha->fw_memory_size - 0x100000 + 1, nxt);
+	if (rval == QLA_SUCCESS)
+		set_bit(RISC_EXT_MEM_DUMP_CMPL, &ha->fw_dump_cap_flags);
+
+	return rval;
 }
 
 static uint32_t *
@@ -296,23 +306,15 @@ qla24xx_read_window(struct device_reg_24xx __iomem *reg, uint32_t iobase,
 	return buf;
 }
 
-int
-qla24xx_pause_risc(struct device_reg_24xx __iomem *reg)
+void
+qla24xx_pause_risc(struct device_reg_24xx __iomem *reg, struct qla_hw_data *ha)
 {
-	int rval = QLA_SUCCESS;
-	uint32_t cnt;
-
 	WRT_REG_DWORD(&reg->hccr, HCCRX_SET_RISC_PAUSE);
-	for (cnt = 30000;
-	    ((RD_REG_DWORD(&reg->host_status) & HSRX_RISC_PAUSED) == 0) &&
-	    rval == QLA_SUCCESS; cnt--) {
-		if (cnt)
-			udelay(100);
-		else
-			rval = QLA_FUNCTION_TIMEOUT;
-	}
 
-	return rval;
+	/* 100 usec delay is sufficient enough for hardware to pause RISC */
+	udelay(100);
+	if (RD_REG_DWORD(&reg->host_status) & HSRX_RISC_PAUSED)
+		set_bit(RISC_PAUSE_CMPL, &ha->fw_dump_cap_flags);
 }
 
 int
@@ -320,10 +322,14 @@ qla24xx_soft_reset(struct qla_hw_data *ha)
 {
 	int rval = QLA_SUCCESS;
 	uint32_t cnt;
-	uint16_t mb0, wd;
+	uint16_t wd;
 	struct device_reg_24xx __iomem *reg = &ha->iobase->isp24;
 
-	/* Reset RISC. */
+	/*
+	 * Reset RISC. The delay is dependent on system architecture.
+	 * Driver can proceed with the reset sequence after waiting
+	 * for a timeout period.
+	 */
 	WRT_REG_DWORD(&reg->ctrl_status, CSRX_DMA_SHUTDOWN|MWB_4096_BYTES);
 	for (cnt = 0; cnt < 30000; cnt++) {
 		if ((RD_REG_DWORD(&reg->ctrl_status) & CSRX_DMA_ACTIVE) == 0)
@@ -331,19 +337,14 @@ qla24xx_soft_reset(struct qla_hw_data *ha)
 
 		udelay(10);
 	}
+	if (!(RD_REG_DWORD(&reg->ctrl_status) & CSRX_DMA_ACTIVE))
+		set_bit(DMA_SHUTDOWN_CMPL, &ha->fw_dump_cap_flags);
 
 	WRT_REG_DWORD(&reg->ctrl_status,
 	    CSRX_ISP_SOFT_RESET|CSRX_DMA_SHUTDOWN|MWB_4096_BYTES);
 	pci_read_config_word(ha->pdev, PCI_COMMAND, &wd);
 
 	udelay(100);
-	/* Wait for firmware to complete NVRAM accesses. */
-	mb0 = (uint32_t) RD_REG_WORD(&reg->mailbox0);
-	for (cnt = 10000 ; cnt && mb0; cnt--) {
-		udelay(5);
-		mb0 = (uint32_t) RD_REG_WORD(&reg->mailbox0);
-		barrier();
-	}
 
 	/* Wait for soft-reset to complete. */
 	for (cnt = 0; cnt < 30000; cnt++) {
@@ -353,16 +354,21 @@ qla24xx_soft_reset(struct qla_hw_data *ha)
 
 		udelay(10);
 	}
+	if (!(RD_REG_DWORD(&reg->ctrl_status) & CSRX_ISP_SOFT_RESET))
+		set_bit(ISP_RESET_CMPL, &ha->fw_dump_cap_flags);
+
 	WRT_REG_DWORD(&reg->hccr, HCCRX_CLR_RISC_RESET);
 	RD_REG_DWORD(&reg->hccr);             /* PCI Posting. */
 
-	for (cnt = 30000; RD_REG_WORD(&reg->mailbox0) != 0 &&
+	for (cnt = 10000; RD_REG_WORD(&reg->mailbox0) != 0 &&
 	    rval == QLA_SUCCESS; cnt--) {
 		if (cnt)
-			udelay(100);
+			udelay(10);
 		else
 			rval = QLA_FUNCTION_TIMEOUT;
 	}
+	if (rval == QLA_SUCCESS)
+		set_bit(RISC_RDY_AFT_RESET, &ha->fw_dump_cap_flags);
 
 	return rval;
 }
@@ -659,12 +665,13 @@ qla2xxx_dump_post_process(scsi_qla_host_t *vha, int rval)
 
 	if (rval != QLA_SUCCESS) {
 		ql_log(ql_log_warn, vha, 0xd000,
-		    "Failed to dump firmware (%x).\n", rval);
+		    "Failed to dump firmware (%x), dump status flags (0x%lx).\n",
+		    rval, ha->fw_dump_cap_flags);
 		ha->fw_dumped = 0;
 	} else {
 		ql_log(ql_log_info, vha, 0xd001,
-		    "Firmware dump saved to temp buffer (%ld/%p).\n",
-		    vha->host_no, ha->fw_dump);
+		    "Firmware dump saved to temp buffer (%ld/%p), dump status flags (0x%lx).\n",
+		    vha->host_no, ha->fw_dump, ha->fw_dump_cap_flags);
 		ha->fw_dumped = 1;
 		qla2x00_post_uevent_work(vha, QLA_UEVENT_CODE_FW_DUMP);
 	}
@@ -1053,6 +1060,7 @@ qla24xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	risc_address = ext_mem_cnt = 0;
 	flags = 0;
+	ha->fw_dump_cap_flags = 0;
 
 	if (!hardware_locked)
 		spin_lock_irqsave(&ha->hardware_lock, flags);
@@ -1075,10 +1083,11 @@ qla24xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	fw->host_status = htonl(RD_REG_DWORD(&reg->host_status));
 
-	/* Pause RISC. */
-	rval = qla24xx_pause_risc(reg);
-	if (rval != QLA_SUCCESS)
-		goto qla24xx_fw_dump_failed_0;
+	/*
+	 * Pause RISC. No need to track timeout, as resetting the chip
+	 * is the right approach incase of pause timeout
+	 */
+	qla24xx_pause_risc(reg, ha);
 
 	/* Host interface registers. */
 	dmp_reg = &reg->flash_addr;
@@ -1302,6 +1311,7 @@ qla25xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	risc_address = ext_mem_cnt = 0;
 	flags = 0;
+	ha->fw_dump_cap_flags = 0;
 
 	if (!hardware_locked)
 		spin_lock_irqsave(&ha->hardware_lock, flags);
@@ -1325,10 +1335,11 @@ qla25xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	fw->host_status = htonl(RD_REG_DWORD(&reg->host_status));
 
-	/* Pause RISC. */
-	rval = qla24xx_pause_risc(reg);
-	if (rval != QLA_SUCCESS)
-		goto qla25xx_fw_dump_failed_0;
+	/*
+	 * Pause RISC. No need to track timeout, as resetting the chip
+	 * is the right approach incase of pause timeout
+	 */
+	qla24xx_pause_risc(reg, ha);
 
 	/* Host/Risc registers. */
 	iter_reg = fw->host_risc_reg;
@@ -1619,6 +1630,7 @@ qla81xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	risc_address = ext_mem_cnt = 0;
 	flags = 0;
+	ha->fw_dump_cap_flags = 0;
 
 	if (!hardware_locked)
 		spin_lock_irqsave(&ha->hardware_lock, flags);
@@ -1641,10 +1653,11 @@ qla81xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	fw->host_status = htonl(RD_REG_DWORD(&reg->host_status));
 
-	/* Pause RISC. */
-	rval = qla24xx_pause_risc(reg);
-	if (rval != QLA_SUCCESS)
-		goto qla81xx_fw_dump_failed_0;
+	/*
+	 * Pause RISC. No need to track timeout, as resetting the chip
+	 * is the right approach incase of pause timeout
+	 */
+	qla24xx_pause_risc(reg, ha);
 
 	/* Host/Risc registers. */
 	iter_reg = fw->host_risc_reg;
@@ -1938,6 +1951,7 @@ qla83xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	risc_address = ext_mem_cnt = 0;
 	flags = 0;
+	ha->fw_dump_cap_flags = 0;
 
 	if (!hardware_locked)
 		spin_lock_irqsave(&ha->hardware_lock, flags);
@@ -1959,10 +1973,11 @@ qla83xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 
 	fw->host_status = htonl(RD_REG_DWORD(&reg->host_status));
 
-	/* Pause RISC. */
-	rval = qla24xx_pause_risc(reg);
-	if (rval != QLA_SUCCESS)
-		goto qla83xx_fw_dump_failed_0;
+	/*
+	 * Pause RISC. No need to track timeout, as resetting the chip
+	 * is the right approach incase of pause timeout
+	 */
+	qla24xx_pause_risc(reg, ha);
 
 	WRT_REG_DWORD(&reg->iobase_addr, 0x6000);
 	dmp_reg = &reg->iobase_window;
@@ -2385,9 +2400,11 @@ qla83xx_fw_dump(scsi_qla_host_t *vha, int hardware_locked)
 			nxt += sizeof(fw->code_ram);
 			nxt += (ha->fw_memory_size - 0x100000 + 1);
 			goto copy_queue;
-		} else
+		} else {
+			set_bit(RISC_RDY_AFT_RESET, &ha->fw_dump_cap_flags);
 			ql_log(ql_log_warn, vha, 0xd010,
 			    "bigger hammer success?\n");
+		}
 	}
 
 	rval = qla24xx_dump_memory(ha, fw->code_ram, sizeof(fw->code_ram),
diff --git a/drivers/scsi/qla2xxx/qla_dbg.h b/drivers/scsi/qla2xxx/qla_dbg.h
index cc961040f8b1..e1fc4e66966a 100644
--- a/drivers/scsi/qla2xxx/qla_dbg.h
+++ b/drivers/scsi/qla2xxx/qla_dbg.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -353,5 +353,6 @@ extern int qla27xx_dump_mpi_ram(struct qla_hw_data *, uint32_t, uint32_t *,
 	uint32_t, void **);
 extern int qla24xx_dump_ram(struct qla_hw_data *, uint32_t, uint32_t *,
 	uint32_t, void **);
-extern int qla24xx_pause_risc(struct device_reg_24xx __iomem *);
+extern void qla24xx_pause_risc(struct device_reg_24xx __iomem *,
+	struct qla_hw_data *);
 extern int qla24xx_soft_reset(struct qla_hw_data *);
diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index 6a106136716c..de5d0ae19d83 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -965,6 +965,13 @@ struct mbx_cmd_32 {
  */
 #define MBC_WRITE_MPI_REGISTER		0x01    /* Write MPI Register. */
 
+/*
+ * ISP8044 mailbox commands
+ */
+#define MBC_SET_GET_ETH_SERDES_REG	0x150
+#define HCS_WRITE_SERDES		0x3
+#define HCS_READ_SERDES			0x4
+
 /* Firmware return data sizes */
 #define FCAL_MAP_SIZE	128
 
@@ -1622,25 +1629,35 @@ typedef struct {
 #define PO_MODE_DIF_PASS	2
 #define PO_MODE_DIF_REPLACE	3
 #define PO_MODE_DIF_TCP_CKSUM	6
-#define PO_ENABLE_DIF_BUNDLING	BIT_8
 #define PO_ENABLE_INCR_GUARD_SEED	BIT_3
-#define PO_DISABLE_INCR_REF_TAG	BIT_5
 #define PO_DISABLE_GUARD_CHECK	BIT_4
+#define PO_DISABLE_INCR_REF_TAG	BIT_5
+#define PO_DIS_HEADER_MODE	BIT_7
+#define PO_ENABLE_DIF_BUNDLING	BIT_8
+#define PO_DIS_FRAME_MODE	BIT_9
+#define PO_DIS_VALD_APP_ESC	BIT_10 /* Dis validation for escape tag/ffffh */
+#define PO_DIS_VALD_APP_REF_ESC BIT_11
+
+#define PO_DIS_APP_TAG_REPL	BIT_12 /* disable REG Tag replacement */
+#define PO_DIS_REF_TAG_REPL	BIT_13
+#define PO_DIS_APP_TAG_VALD	BIT_14 /* disable REF Tag validation */
+#define PO_DIS_REF_TAG_VALD	BIT_15
+
 /*
  * ISP queue - 64-Bit addressing, continuation crc entry structure definition.
  */
 struct crc_context {
 	uint32_t handle;		/* System handle. */
-	uint32_t ref_tag;
-	uint16_t app_tag;
+	__le32 ref_tag;
+	__le16 app_tag;
 	uint8_t ref_tag_mask[4];	/* Validation/Replacement Mask*/
 	uint8_t app_tag_mask[2];	/* Validation/Replacement Mask*/
-	uint16_t guard_seed;		/* Initial Guard Seed */
-	uint16_t prot_opts;		/* Requested Data Protection Mode */
-	uint16_t blk_size;		/* Data size in bytes */
+	__le16 guard_seed;		/* Initial Guard Seed */
+	__le16 prot_opts;		/* Requested Data Protection Mode */
+	__le16 blk_size;		/* Data size in bytes */
 	uint16_t runt_blk_guard;	/* Guard value for runt block (tape
 					 * only) */
-	uint32_t byte_count;		/* Total byte count/ total data
+	__le32 byte_count;		/* Total byte count/ total data
 					 * transfer count */
 	union {
 		struct {
@@ -1654,10 +1671,10 @@ struct crc_context {
 			uint32_t	reserved_6;
 		} nobundling;
 		struct {
-			uint32_t	dif_byte_count;	/* Total DIF byte
+			__le32	dif_byte_count;	/* Total DIF byte
 							 * count */
 			uint16_t	reserved_1;
-			uint16_t	dseg_count;	/* Data segment count */
+			__le16	dseg_count;	/* Data segment count */
 			uint32_t	reserved_2;
 			uint32_t	data_address[2];
 			uint32_t	data_length;
@@ -1748,6 +1765,8 @@ typedef struct {
 #define CS_PORT_CONFIG_CHG	0x2A	/* Port Configuration Changed */
 #define CS_PORT_BUSY		0x2B	/* Port Busy */
 #define CS_COMPLETE_CHKCOND	0x30	/* Error? */
+#define CS_IOCB_ERROR		0x31	/* Generic error for IOCB request
+					   failure */
 #define CS_BAD_PAYLOAD		0x80	/* Driver defined */
 #define CS_UNKNOWN		0x81	/* Driver defined */
 #define CS_RETRY		0x82	/* Driver defined */
@@ -2676,6 +2695,7 @@ struct rsp_que {
 	uint32_t __iomem *rsp_q_out;
 	uint16_t  ring_index;
 	uint16_t  out_ptr;
+	uint16_t  *in_ptr;		/* queue shadow in index */
 	uint16_t  length;
 	uint16_t  options;
 	uint16_t  rid;
@@ -2702,6 +2722,7 @@ struct req_que {
 	uint32_t __iomem *req_q_out;
 	uint16_t  ring_index;
 	uint16_t  in_ptr;
+	uint16_t  *out_ptr;		/* queue shadow out index */
 	uint16_t  cnt;
 	uint16_t  length;
 	uint16_t  options;
@@ -2907,6 +2928,8 @@ struct qla_hw_data {
 #define PCI_DEVICE_ID_QLOGIC_ISP8031	0x8031
 #define PCI_DEVICE_ID_QLOGIC_ISP2031	0x2031
 #define PCI_DEVICE_ID_QLOGIC_ISP2071	0x2071
+#define PCI_DEVICE_ID_QLOGIC_ISP2271	0x2271
+
 	uint32_t	device_type;
 #define DT_ISP2100                      BIT_0
 #define DT_ISP2200                      BIT_1
@@ -2928,7 +2951,8 @@ struct qla_hw_data {
 #define DT_ISPFX00			BIT_17
 #define DT_ISP8044			BIT_18
 #define DT_ISP2071			BIT_19
-#define DT_ISP_LAST			(DT_ISP2071 << 1)
+#define DT_ISP2271			BIT_20
+#define DT_ISP_LAST			(DT_ISP2271 << 1)
 
 #define DT_T10_PI                       BIT_25
 #define DT_IIDMA                        BIT_26
@@ -2959,6 +2983,7 @@ struct qla_hw_data {
 #define IS_QLA8031(ha)	(DT_MASK(ha) & DT_ISP8031)
 #define IS_QLAFX00(ha)	(DT_MASK(ha) & DT_ISPFX00)
 #define IS_QLA2071(ha)	(DT_MASK(ha) & DT_ISP2071)
+#define IS_QLA2271(ha)	(DT_MASK(ha) & DT_ISP2271)
 
 #define IS_QLA23XX(ha)  (IS_QLA2300(ha) || IS_QLA2312(ha) || IS_QLA2322(ha) || \
 			IS_QLA6312(ha) || IS_QLA6322(ha))
@@ -2967,7 +2992,7 @@ struct qla_hw_data {
 #define IS_QLA25XX(ha)  (IS_QLA2532(ha))
 #define IS_QLA83XX(ha)	(IS_QLA2031(ha) || IS_QLA8031(ha))
 #define IS_QLA84XX(ha)  (IS_QLA8432(ha))
-#define IS_QLA27XX(ha)  (IS_QLA2071(ha))
+#define IS_QLA27XX(ha)  (IS_QLA2071(ha) || IS_QLA2271(ha))
 #define IS_QLA24XX_TYPE(ha)     (IS_QLA24XX(ha) || IS_QLA54XX(ha) || \
 				IS_QLA84XX(ha))
 #define IS_CNA_CAPABLE(ha)	(IS_QLA81XX(ha) || IS_QLA82XX(ha) || \
@@ -3006,6 +3031,7 @@ struct qla_hw_data {
     (((ha)->fw_attributes_h << 16 | (ha)->fw_attributes) & BIT_22))
 #define IS_ATIO_MSIX_CAPABLE(ha) (IS_QLA83XX(ha))
 #define IS_TGT_MODE_CAPABLE(ha)	(ha->tgt.atio_q_length)
+#define IS_SHADOW_REG_CAPABLE(ha)  (IS_QLA27XX(ha))
 
 	/* HBA serial number */
 	uint8_t		serial0;
@@ -3136,7 +3162,15 @@ struct qla_hw_data {
 	struct qla2xxx_fw_dump *fw_dump;
 	uint32_t	fw_dump_len;
 	int		fw_dumped;
+	unsigned long	fw_dump_cap_flags;
+#define RISC_PAUSE_CMPL		0
+#define DMA_SHUTDOWN_CMPL	1
+#define ISP_RESET_CMPL		2
+#define RISC_RDY_AFT_RESET	3
+#define RISC_SRAM_DUMP_CMPL	4
+#define RISC_EXT_MEM_DUMP_CMPL	5
 	int		fw_dump_reading;
+	int		prev_minidump_failed;
 	dma_addr_t	eft_dma;
 	void		*eft;
 /* Current size of mctp dump is 0x086064 bytes */
diff --git a/drivers/scsi/qla2xxx/qla_dfs.c b/drivers/scsi/qla2xxx/qla_dfs.c
index 32ab80957688..2ca39b8e7166 100644
--- a/drivers/scsi/qla2xxx/qla_dfs.c
+++ b/drivers/scsi/qla2xxx/qla_dfs.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
diff --git a/drivers/scsi/qla2xxx/qla_fw.h b/drivers/scsi/qla2xxx/qla_fw.h
index 3a7353eaccbd..eb8f57249f1d 100644
--- a/drivers/scsi/qla2xxx/qla_fw.h
+++ b/drivers/scsi/qla2xxx/qla_fw.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -371,7 +371,10 @@ struct init_cb_24xx {
 	 * BIT 14 = Data Rate bit 1
 	 * BIT 15 = Data Rate bit 2
 	 * BIT 16 = Enable 75 ohm Termination Select
-	 * BIT 17-31 = Reserved
+	 * BIT 17-28 = Reserved
+	 * BIT 29 = Enable response queue 0 in index shadowing
+	 * BIT 30 = Enable request queue 0 out index shadowing
+	 * BIT 31 = Reserved
 	 */
 	uint32_t firmware_options_3;
 	uint16_t qos;
@@ -1134,13 +1137,6 @@ struct device_reg_24xx {
 #define MIN_MULTI_ID_FABRIC	64	/* Must be power-of-2. */
 #define MAX_MULTI_ID_FABRIC	256	/* ... */
 
-#define for_each_mapped_vp_idx(_ha, _idx)		\
-	for (_idx = find_next_bit((_ha)->vp_idx_map,	\
-		(_ha)->max_npiv_vports + 1, 1);		\
-	    _idx <= (_ha)->max_npiv_vports;		\
-	    _idx = find_next_bit((_ha)->vp_idx_map,	\
-		(_ha)->max_npiv_vports + 1, _idx + 1))	\
-
 struct mid_conf_entry_24xx {
 	uint16_t reserved_1;
 
diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h
index e665e8109933..d48dea8fab1b 100644
--- a/drivers/scsi/qla2xxx/qla_gbl.h
+++ b/drivers/scsi/qla2xxx/qla_gbl.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -220,6 +220,13 @@ extern unsigned long qla2x00_get_async_timeout(struct scsi_qla_host *);
 
 extern void *qla2x00_alloc_iocbs(scsi_qla_host_t *, srb_t *);
 extern int qla2x00_issue_marker(scsi_qla_host_t *, int);
+extern int qla24xx_walk_and_build_sglist_no_difb(struct qla_hw_data *, srb_t *,
+	uint32_t *, uint16_t, struct qla_tgt_cmd *);
+extern int qla24xx_walk_and_build_sglist(struct qla_hw_data *, srb_t *,
+	uint32_t *, uint16_t, struct qla_tgt_cmd *);
+extern int qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *, srb_t *,
+	uint32_t *, uint16_t, struct qla_tgt_cmd *);
+
 
 /*
  * Global Function Prototypes in qla_mbx.c source file.
@@ -347,6 +354,11 @@ extern int
 qla2x00_read_serdes_word(scsi_qla_host_t *, uint16_t, uint16_t *);
 
 extern int
+qla8044_write_serdes_word(scsi_qla_host_t *, uint32_t, uint32_t);
+extern int
+qla8044_read_serdes_word(scsi_qla_host_t *, uint32_t, uint32_t *);
+
+extern int
 qla2x00_set_serdes_params(scsi_qla_host_t *, uint16_t, uint16_t, uint16_t);
 
 extern int
diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
index e377f9d2f92a..a0df3b1b3823 100644
--- a/drivers/scsi/qla2xxx/qla_gs.c
+++ b/drivers/scsi/qla2xxx/qla_gs.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 38aeb54cd9d8..e2184412617d 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -1476,6 +1476,7 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *vha)
 	}
 
 	ha->fw_dumped = 0;
+	ha->fw_dump_cap_flags = 0;
 	dump_size = fixed_size = mem_size = eft_size = fce_size = mq_size = 0;
 	req_q_size = rsp_q_size = 0;
 
@@ -2061,6 +2062,10 @@ qla24xx_config_rings(struct scsi_qla_host *vha)
 	icb->atio_q_address[0] = cpu_to_le32(LSD(ha->tgt.atio_dma));
 	icb->atio_q_address[1] = cpu_to_le32(MSD(ha->tgt.atio_dma));
 
+	if (IS_SHADOW_REG_CAPABLE(ha))
+		icb->firmware_options_2 |=
+		    __constant_cpu_to_le32(BIT_30|BIT_29);
+
 	if (ha->mqenable || IS_QLA83XX(ha) || IS_QLA27XX(ha)) {
 		icb->qos = __constant_cpu_to_le16(QLA_DEFAULT_QUE_QOS);
 		icb->rid = __constant_cpu_to_le16(rid);
@@ -2138,6 +2143,8 @@ qla2x00_init_rings(scsi_qla_host_t *vha)
 		req = ha->req_q_map[que];
 		if (!req)
 			continue;
+		req->out_ptr = (void *)(req->ring + req->length);
+		*req->out_ptr = 0;
 		for (cnt = 1; cnt < req->num_outstanding_cmds; cnt++)
 			req->outstanding_cmds[cnt] = NULL;
 
@@ -2153,6 +2160,8 @@ qla2x00_init_rings(scsi_qla_host_t *vha)
 		rsp = ha->rsp_q_map[que];
 		if (!rsp)
 			continue;
+		rsp->in_ptr = (void *)(rsp->ring + rsp->length);
+		*rsp->in_ptr = 0;
 		/* Initialize response queue entries */
 		if (IS_QLAFX00(ha))
 			qlafx00_init_response_q_entries(rsp);
@@ -3406,7 +3415,7 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha)
 					    fcport->d_id.b.domain,
 					    fcport->d_id.b.area,
 					    fcport->d_id.b.al_pa);
-					fcport->loop_id = FC_NO_LOOP_ID;
+					qla2x00_clear_loop_id(fcport);
 				}
 			}
 		}
@@ -4727,7 +4736,6 @@ static int
 qla2x00_restart_isp(scsi_qla_host_t *vha)
 {
 	int status = 0;
-	uint32_t wait_time;
 	struct qla_hw_data *ha = vha->hw;
 	struct req_que *req = ha->req_q_map[0];
 	struct rsp_que *rsp = ha->rsp_q_map[0];
@@ -4744,14 +4752,12 @@ qla2x00_restart_isp(scsi_qla_host_t *vha)
 	if (!status && !(status = qla2x00_init_rings(vha))) {
 		clear_bit(RESET_MARKER_NEEDED, &vha->dpc_flags);
 		ha->flags.chip_reset_done = 1;
+
 		/* Initialize the queues in use */
 		qla25xx_init_queues(ha);
 
 		status = qla2x00_fw_ready(vha);
 		if (!status) {
-			ql_dbg(ql_dbg_taskm, vha, 0x8031,
-			    "Start configure loop status = %d.\n", status);
-
 			/* Issue a marker after FW becomes ready. */
 			qla2x00_marker(vha, req, rsp, 0, 0, MK_SYNC_ALL);
 
@@ -4766,24 +4772,12 @@ qla2x00_restart_isp(scsi_qla_host_t *vha)
 				qlt_24xx_process_atio_queue(vha);
 			spin_unlock_irqrestore(&ha->hardware_lock, flags);
 
-			/* Wait at most MAX_TARGET RSCNs for a stable link. */
-			wait_time = 256;
-			do {
-				clear_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags);
-				qla2x00_configure_loop(vha);
-				wait_time--;
-			} while (!atomic_read(&vha->loop_down_timer) &&
-				!(test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags))
-				&& wait_time && (test_bit(LOOP_RESYNC_NEEDED,
-				&vha->dpc_flags)));
+			set_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags);
 		}
 
 		/* if no cable then assume it's good */
 		if ((vha->device_flags & DFLG_NO_CABLE))
 			status = 0;
-
-		ql_dbg(ql_dbg_taskm, vha, 0x8032,
-		    "Configure loop done, status = 0x%x.\n", status);
 	}
 	return (status);
 }
@@ -6130,7 +6124,6 @@ int
 qla82xx_restart_isp(scsi_qla_host_t *vha)
 {
 	int status, rval;
-	uint32_t wait_time;
 	struct qla_hw_data *ha = vha->hw;
 	struct req_que *req = ha->req_q_map[0];
 	struct rsp_que *rsp = ha->rsp_q_map[0];
@@ -6144,31 +6137,15 @@ qla82xx_restart_isp(scsi_qla_host_t *vha)
 
 		status = qla2x00_fw_ready(vha);
 		if (!status) {
-			ql_log(ql_log_info, vha, 0x803c,
-			    "Start configure loop, status =%d.\n", status);
-
 			/* Issue a marker after FW becomes ready. */
 			qla2x00_marker(vha, req, rsp, 0, 0, MK_SYNC_ALL);
-
 			vha->flags.online = 1;
-			/* Wait at most MAX_TARGET RSCNs for a stable link. */
-			wait_time = 256;
-			do {
-				clear_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags);
-				qla2x00_configure_loop(vha);
-				wait_time--;
-			} while (!atomic_read(&vha->loop_down_timer) &&
-			    !(test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags)) &&
-			    wait_time &&
-			    (test_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags)));
+			set_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags);
 		}
 
 		/* if no cable then assume it's good */
 		if ((vha->device_flags & DFLG_NO_CABLE))
 			status = 0;
-
-		ql_log(ql_log_info, vha, 0x8000,
-		    "Configure loop done, status = 0x%x.\n", status);
 	}
 
 	if (!status) {
@@ -6182,8 +6159,6 @@ qla82xx_restart_isp(scsi_qla_host_t *vha)
 			vha->marker_needed = 1;
 		}
 
-		vha->flags.online = 1;
-
 		ha->isp_ops->enable_intrs(ha);
 
 		ha->isp_abort_cnt = 0;
diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h
index ce8b5fb0f347..b3b1d6fc2d6c 100644
--- a/drivers/scsi/qla2xxx/qla_inline.h
+++ b/drivers/scsi/qla2xxx/qla_inline.h
@@ -1,10 +1,11 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
 
+#include "qla_target.h"
 /**
  * qla24xx_calc_iocbs() - Determine number of Command Type 3 and
  * Continuation Type 1 IOCBs to allocate.
@@ -128,12 +129,20 @@ qla2x00_clear_loop_id(fc_port_t *fcport) {
 }
 
 static inline void
-qla2x00_clean_dsd_pool(struct qla_hw_data *ha, srb_t *sp)
+qla2x00_clean_dsd_pool(struct qla_hw_data *ha, srb_t *sp,
+	struct qla_tgt_cmd *tc)
 {
 	struct dsd_dma *dsd_ptr, *tdsd_ptr;
 	struct crc_context *ctx;
 
-	ctx = (struct crc_context *)GET_CMD_CTX_SP(sp);
+	if (sp)
+		ctx = (struct crc_context *)GET_CMD_CTX_SP(sp);
+	else if (tc)
+		ctx = (struct crc_context *)tc->ctx;
+	else {
+		BUG();
+		return;
+	}
 
 	/* clean up allocated prev pool */
 	list_for_each_entry_safe(dsd_ptr, tdsd_ptr,
diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c
index e607568bce49..760931529592 100644
--- a/drivers/scsi/qla2xxx/qla_iocb.c
+++ b/drivers/scsi/qla2xxx/qla_iocb.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -936,9 +936,9 @@ qla24xx_get_one_block_sg(uint32_t blk_sz, struct qla2_sgx *sgx,
 	return 1;
 }
 
-static int
+int
 qla24xx_walk_and_build_sglist_no_difb(struct qla_hw_data *ha, srb_t *sp,
-	uint32_t *dsd, uint16_t tot_dsds)
+	uint32_t *dsd, uint16_t tot_dsds, struct qla_tgt_cmd *tc)
 {
 	void *next_dsd;
 	uint8_t avail_dsds = 0;
@@ -948,21 +948,35 @@ qla24xx_walk_and_build_sglist_no_difb(struct qla_hw_data *ha, srb_t *sp,
 	uint32_t *cur_dsd = dsd;
 	uint16_t	used_dsds = tot_dsds;
 
-	uint32_t	prot_int;
+	uint32_t	prot_int; /* protection interval */
 	uint32_t	partial;
 	struct qla2_sgx sgx;
 	dma_addr_t	sle_dma;
 	uint32_t	sle_dma_len, tot_prot_dma_len = 0;
-	struct scsi_cmnd *cmd = GET_CMD_SP(sp);
-
-	prot_int = cmd->device->sector_size;
+	struct scsi_cmnd *cmd;
+	struct scsi_qla_host *vha;
 
 	memset(&sgx, 0, sizeof(struct qla2_sgx));
-	sgx.tot_bytes = scsi_bufflen(cmd);
-	sgx.cur_sg = scsi_sglist(cmd);
-	sgx.sp = sp;
-
-	sg_prot = scsi_prot_sglist(cmd);
+	if (sp) {
+		vha = sp->fcport->vha;
+		cmd = GET_CMD_SP(sp);
+		prot_int = cmd->device->sector_size;
+
+		sgx.tot_bytes = scsi_bufflen(cmd);
+		sgx.cur_sg = scsi_sglist(cmd);
+		sgx.sp = sp;
+
+		sg_prot = scsi_prot_sglist(cmd);
+	} else if (tc) {
+		vha = tc->vha;
+		prot_int      = tc->blk_sz;
+		sgx.tot_bytes = tc->bufflen;
+		sgx.cur_sg    = tc->sg;
+		sg_prot	      = tc->prot_sg;
+	} else {
+		BUG();
+		return 1;
+	}
 
 	while (qla24xx_get_one_block_sg(prot_int, &sgx, &partial)) {
 
@@ -995,10 +1009,18 @@ alloc_and_fill:
 				return 1;
 			}
 
-			list_add_tail(&dsd_ptr->list,
-			    &((struct crc_context *)sp->u.scmd.ctx)->dsd_list);
+			if (sp) {
+				list_add_tail(&dsd_ptr->list,
+				    &((struct crc_context *)
+					    sp->u.scmd.ctx)->dsd_list);
+
+				sp->flags |= SRB_CRC_CTX_DSD_VALID;
+			} else {
+				list_add_tail(&dsd_ptr->list,
+				    &(tc->ctx->dsd_list));
+				tc->ctx_dsd_alloced = 1;
+			}
 
-			sp->flags |= SRB_CRC_CTX_DSD_VALID;
 
 			/* add new list to cmd iocb or last list */
 			*cur_dsd++ = cpu_to_le32(LSD(dsd_ptr->dsd_list_dma));
@@ -1033,21 +1055,35 @@ alloc_and_fill:
 	return 0;
 }
 
-static int
+int
 qla24xx_walk_and_build_sglist(struct qla_hw_data *ha, srb_t *sp, uint32_t *dsd,
-	uint16_t tot_dsds)
+	uint16_t tot_dsds, struct qla_tgt_cmd *tc)
 {
 	void *next_dsd;
 	uint8_t avail_dsds = 0;
 	uint32_t dsd_list_len;
 	struct dsd_dma *dsd_ptr;
-	struct scatterlist *sg;
+	struct scatterlist *sg, *sgl;
 	uint32_t *cur_dsd = dsd;
 	int	i;
 	uint16_t	used_dsds = tot_dsds;
-	struct scsi_cmnd *cmd = GET_CMD_SP(sp);
+	struct scsi_cmnd *cmd;
+	struct scsi_qla_host *vha;
+
+	if (sp) {
+		cmd = GET_CMD_SP(sp);
+		sgl = scsi_sglist(cmd);
+		vha = sp->fcport->vha;
+	} else if (tc) {
+		sgl = tc->sg;
+		vha = tc->vha;
+	} else {
+		BUG();
+		return 1;
+	}
 
-	scsi_for_each_sg(cmd, sg, tot_dsds, i) {
+
+	for_each_sg(sgl, sg, tot_dsds, i) {
 		dma_addr_t	sle_dma;
 
 		/* Allocate additional continuation packets? */
@@ -1076,10 +1112,17 @@ qla24xx_walk_and_build_sglist(struct qla_hw_data *ha, srb_t *sp, uint32_t *dsd,
 				return 1;
 			}
 
-			list_add_tail(&dsd_ptr->list,
-			    &((struct crc_context *)sp->u.scmd.ctx)->dsd_list);
+			if (sp) {
+				list_add_tail(&dsd_ptr->list,
+				    &((struct crc_context *)
+					    sp->u.scmd.ctx)->dsd_list);
 
-			sp->flags |= SRB_CRC_CTX_DSD_VALID;
+				sp->flags |= SRB_CRC_CTX_DSD_VALID;
+			} else {
+				list_add_tail(&dsd_ptr->list,
+				    &(tc->ctx->dsd_list));
+				tc->ctx_dsd_alloced = 1;
+			}
 
 			/* add new list to cmd iocb or last list */
 			*cur_dsd++ = cpu_to_le32(LSD(dsd_ptr->dsd_list_dma));
@@ -1102,23 +1145,37 @@ qla24xx_walk_and_build_sglist(struct qla_hw_data *ha, srb_t *sp, uint32_t *dsd,
 	return 0;
 }
 
-static int
+int
 qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *ha, srb_t *sp,
-							uint32_t *dsd,
-	uint16_t tot_dsds)
+	uint32_t *dsd, uint16_t tot_dsds, struct qla_tgt_cmd *tc)
 {
 	void *next_dsd;
 	uint8_t avail_dsds = 0;
 	uint32_t dsd_list_len;
 	struct dsd_dma *dsd_ptr;
-	struct scatterlist *sg;
+	struct scatterlist *sg, *sgl;
 	int	i;
 	struct scsi_cmnd *cmd;
 	uint32_t *cur_dsd = dsd;
-	uint16_t	used_dsds = tot_dsds;
+	uint16_t used_dsds = tot_dsds;
+	struct scsi_qla_host *vha;
+
+	if (sp) {
+		cmd = GET_CMD_SP(sp);
+		sgl = scsi_prot_sglist(cmd);
+		vha = sp->fcport->vha;
+	} else if (tc) {
+		vha = tc->vha;
+		sgl = tc->prot_sg;
+	} else {
+		BUG();
+		return 1;
+	}
 
-	cmd = GET_CMD_SP(sp);
-	scsi_for_each_prot_sg(cmd, sg, tot_dsds, i) {
+	ql_dbg(ql_dbg_tgt, vha, 0xe021,
+		"%s: enter\n", __func__);
+
+	for_each_sg(sgl, sg, tot_dsds, i) {
 		dma_addr_t	sle_dma;
 
 		/* Allocate additional continuation packets? */
@@ -1147,10 +1204,17 @@ qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *ha, srb_t *sp,
 				return 1;
 			}
 
-			list_add_tail(&dsd_ptr->list,
-			    &((struct crc_context *)sp->u.scmd.ctx)->dsd_list);
+			if (sp) {
+				list_add_tail(&dsd_ptr->list,
+				    &((struct crc_context *)
+					    sp->u.scmd.ctx)->dsd_list);
 
-			sp->flags |= SRB_CRC_CTX_DSD_VALID;
+				sp->flags |= SRB_CRC_CTX_DSD_VALID;
+			} else {
+				list_add_tail(&dsd_ptr->list,
+				    &(tc->ctx->dsd_list));
+				tc->ctx_dsd_alloced = 1;
+			}
 
 			/* add new list to cmd iocb or last list */
 			*cur_dsd++ = cpu_to_le32(LSD(dsd_ptr->dsd_list_dma));
@@ -1386,10 +1450,10 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt,
 
 	if (!bundling && tot_prot_dsds) {
 		if (qla24xx_walk_and_build_sglist_no_difb(ha, sp,
-		    cur_dsd, tot_dsds))
+			cur_dsd, tot_dsds, NULL))
 			goto crc_queuing_error;
 	} else if (qla24xx_walk_and_build_sglist(ha, sp, cur_dsd,
-	    (tot_dsds - tot_prot_dsds)))
+			(tot_dsds - tot_prot_dsds), NULL))
 		goto crc_queuing_error;
 
 	if (bundling && tot_prot_dsds) {
@@ -1398,7 +1462,7 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt,
 			__constant_cpu_to_le16(CF_DIF_SEG_DESCR_ENABLE);
 		cur_dsd = (uint32_t *) &crc_ctx_pkt->u.bundling.dif_address;
 		if (qla24xx_walk_and_build_prot_sglist(ha, sp, cur_dsd,
-		    tot_prot_dsds))
+				tot_prot_dsds, NULL))
 			goto crc_queuing_error;
 	}
 	return QLA_SUCCESS;
@@ -1478,8 +1542,8 @@ qla24xx_start_scsi(srb_t *sp)
 	tot_dsds = nseg;
 	req_cnt = qla24xx_calc_iocbs(vha, tot_dsds);
 	if (req->cnt < (req_cnt + 2)) {
-		cnt = RD_REG_DWORD_RELAXED(req->req_q_out);
-
+		cnt = IS_SHADOW_REG_CAPABLE(ha) ? *req->out_ptr :
+		    RD_REG_DWORD_RELAXED(req->req_q_out);
 		if (req->ring_index < cnt)
 			req->cnt = cnt - req->ring_index;
 		else
@@ -1697,8 +1761,8 @@ qla24xx_dif_start_scsi(srb_t *sp)
 	tot_prot_dsds = nseg;
 	tot_dsds += nseg;
 	if (req->cnt < (req_cnt + 2)) {
-		cnt = RD_REG_DWORD_RELAXED(req->req_q_out);
-
+		cnt = IS_SHADOW_REG_CAPABLE(ha) ? *req->out_ptr :
+		    RD_REG_DWORD_RELAXED(req->req_q_out);
 		if (req->ring_index < cnt)
 			req->cnt = cnt - req->ring_index;
 		else
@@ -2825,8 +2889,8 @@ qla2x00_start_bidir(srb_t *sp, struct scsi_qla_host *vha, uint32_t tot_dsds)
 
 	/* Check for room on request queue. */
 	if (req->cnt < req_cnt + 2) {
-		cnt = RD_REG_DWORD_RELAXED(req->req_q_out);
-
+		cnt = IS_SHADOW_REG_CAPABLE(ha) ? *req->out_ptr :
+		    RD_REG_DWORD_RELAXED(req->req_q_out);
 		if  (req->ring_index < cnt)
 			req->cnt = cnt - req->ring_index;
 		else
diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index 95314ef2e505..a56825c73c31 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -2009,11 +2009,13 @@ qla2x00_status_entry(scsi_qla_host_t *vha, struct rsp_que *rsp, void *pkt)
 		ql_dbg(ql_dbg_io, vha, 0x3017,
 		    "Invalid status handle (0x%x).\n", sts->handle);
 
-		if (IS_P3P_TYPE(ha))
-			set_bit(FCOE_CTX_RESET_NEEDED, &vha->dpc_flags);
-		else
-			set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags);
-		qla2xxx_wake_dpc(vha);
+		if (!test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags)) {
+			if (IS_P3P_TYPE(ha))
+				set_bit(FCOE_CTX_RESET_NEEDED, &vha->dpc_flags);
+			else
+				set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags);
+			qla2xxx_wake_dpc(vha);
+		}
 		return;
 	}
 
@@ -2472,12 +2474,14 @@ void qla24xx_process_response_queue(struct scsi_qla_host *vha,
 		if (pkt->entry_status != 0) {
 			qla2x00_error_entry(vha, rsp, (sts_entry_t *) pkt);
 
-			(void)qlt_24xx_process_response_error(vha, pkt);
+			if (qlt_24xx_process_response_error(vha, pkt))
+				goto process_err;
 
 			((response_t *)pkt)->signature = RESPONSE_PROCESSED;
 			wmb();
 			continue;
 		}
+process_err:
 
 		switch (pkt->entry_type) {
 		case STATUS_TYPE:
@@ -2494,10 +2498,10 @@ void qla24xx_process_response_queue(struct scsi_qla_host *vha,
 			qla24xx_logio_entry(vha, rsp->req,
 			    (struct logio_entry_24xx *)pkt);
 			break;
-                case CT_IOCB_TYPE:
+		case CT_IOCB_TYPE:
 			qla24xx_els_ct_entry(vha, rsp->req, pkt, CT_IOCB_TYPE);
 			break;
-                case ELS_IOCB_TYPE:
+		case ELS_IOCB_TYPE:
 			qla24xx_els_ct_entry(vha, rsp->req, pkt, ELS_IOCB_TYPE);
 			break;
 		case ABTS_RECV_24XX:
@@ -2506,6 +2510,7 @@ void qla24xx_process_response_queue(struct scsi_qla_host *vha,
 		case ABTS_RESP_24XX:
 		case CTIO_TYPE7:
 		case NOTIFY_ACK_TYPE:
+		case CTIO_CRC2:
 			qlt_response_pkt_all_vps(vha, (response_t *)pkt);
 			break;
 		case MARKER_TYPE:
diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 2528709c4add..1c33a77db5c2 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -1319,7 +1319,7 @@ qla2x00_get_node_name_list(scsi_qla_host_t *vha, void **out_data, int *out_len)
 
 		left = 0;
 
-		list = kzalloc(dma_size, GFP_KERNEL);
+		list = kmemdup(pmap, dma_size, GFP_KERNEL);
 		if (!list) {
 			ql_log(ql_log_warn, vha, 0x1140,
 			    "%s(%ld): failed to allocate node names list "
@@ -1328,7 +1328,6 @@ qla2x00_get_node_name_list(scsi_qla_host_t *vha, void **out_data, int *out_len)
 			goto out_free;
 		}
 
-		memcpy(list, pmap, dma_size);
 restart:
 		dma_free_coherent(&ha->pdev->dev, dma_size, pmap, pmap_dma);
 	}
@@ -2644,7 +2643,10 @@ qla24xx_abort_command(srb_t *sp)
 		ql_dbg(ql_dbg_mbx, vha, 0x1090,
 		    "Failed to complete IOCB -- completion status (%x).\n",
 		    le16_to_cpu(abt->nport_handle));
-		rval = QLA_FUNCTION_FAILED;
+		if (abt->nport_handle == CS_IOCB_ERROR)
+			rval = QLA_FUNCTION_PARAMETER_ERROR;
+		else
+			rval = QLA_FUNCTION_FAILED;
 	} else {
 		ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x1091,
 		    "Done %s.\n", __func__);
@@ -2879,6 +2881,78 @@ qla2x00_read_serdes_word(scsi_qla_host_t *vha, uint16_t addr, uint16_t *data)
 	return rval;
 }
 
+int
+qla8044_write_serdes_word(scsi_qla_host_t *vha, uint32_t addr, uint32_t data)
+{
+	int rval;
+	mbx_cmd_t mc;
+	mbx_cmd_t *mcp = &mc;
+
+	if (!IS_QLA8044(vha->hw))
+		return QLA_FUNCTION_FAILED;
+
+	ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x1186,
+	    "Entered %s.\n", __func__);
+
+	mcp->mb[0] = MBC_SET_GET_ETH_SERDES_REG;
+	mcp->mb[1] = HCS_WRITE_SERDES;
+	mcp->mb[3] = LSW(addr);
+	mcp->mb[4] = MSW(addr);
+	mcp->mb[5] = LSW(data);
+	mcp->mb[6] = MSW(data);
+	mcp->out_mb = MBX_6|MBX_5|MBX_4|MBX_3|MBX_1|MBX_0;
+	mcp->in_mb = MBX_0;
+	mcp->tov = MBX_TOV_SECONDS;
+	mcp->flags = 0;
+	rval = qla2x00_mailbox_command(vha, mcp);
+
+	if (rval != QLA_SUCCESS) {
+		ql_dbg(ql_dbg_mbx, vha, 0x1187,
+		    "Failed=%x mb[0]=%x.\n", rval, mcp->mb[0]);
+	} else {
+		ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x1188,
+		    "Done %s.\n", __func__);
+	}
+
+	return rval;
+}
+
+int
+qla8044_read_serdes_word(scsi_qla_host_t *vha, uint32_t addr, uint32_t *data)
+{
+	int rval;
+	mbx_cmd_t mc;
+	mbx_cmd_t *mcp = &mc;
+
+	if (!IS_QLA8044(vha->hw))
+		return QLA_FUNCTION_FAILED;
+
+	ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x1189,
+	    "Entered %s.\n", __func__);
+
+	mcp->mb[0] = MBC_SET_GET_ETH_SERDES_REG;
+	mcp->mb[1] = HCS_READ_SERDES;
+	mcp->mb[3] = LSW(addr);
+	mcp->mb[4] = MSW(addr);
+	mcp->out_mb = MBX_4|MBX_3|MBX_1|MBX_0;
+	mcp->in_mb = MBX_2|MBX_1|MBX_0;
+	mcp->tov = MBX_TOV_SECONDS;
+	mcp->flags = 0;
+	rval = qla2x00_mailbox_command(vha, mcp);
+
+	*data = mcp->mb[2] << 16 | mcp->mb[1];
+
+	if (rval != QLA_SUCCESS) {
+		ql_dbg(ql_dbg_mbx, vha, 0x118a,
+		    "Failed=%x mb[0]=%x.\n", rval, mcp->mb[0]);
+	} else {
+		ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x118b,
+		    "Done %s.\n", __func__);
+	}
+
+	return rval;
+}
+
 /**
  * qla2x00_set_serdes_params() -
  * @ha: HA context
@@ -3660,6 +3734,9 @@ qla25xx_init_req_que(struct scsi_qla_host *vha, struct req_que *req)
 	ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x10d3,
 	    "Entered %s.\n", __func__);
 
+	if (IS_SHADOW_REG_CAPABLE(ha))
+		req->options |= BIT_13;
+
 	mcp->mb[0] = MBC_INITIALIZE_MULTIQ;
 	mcp->mb[1] = req->options;
 	mcp->mb[2] = MSW(LSD(req->dma));
@@ -3679,7 +3756,7 @@ qla25xx_init_req_que(struct scsi_qla_host *vha, struct req_que *req)
 	/* que in ptr index */
 	mcp->mb[8] = 0;
 	/* que out ptr index */
-	mcp->mb[9] = 0;
+	mcp->mb[9] = *req->out_ptr = 0;
 	mcp->out_mb = MBX_14|MBX_13|MBX_12|MBX_11|MBX_10|MBX_9|MBX_8|MBX_7|
 			MBX_6|MBX_5|MBX_4|MBX_3|MBX_2|MBX_1|MBX_0;
 	mcp->in_mb = MBX_0;
@@ -3688,7 +3765,7 @@ qla25xx_init_req_que(struct scsi_qla_host *vha, struct req_que *req)
 
 	if (IS_QLA81XX(ha) || IS_QLA83XX(ha) || IS_QLA27XX(ha))
 		mcp->in_mb |= MBX_1;
-	if (IS_QLA83XX(ha) || !IS_QLA27XX(ha)) {
+	if (IS_QLA83XX(ha) || IS_QLA27XX(ha)) {
 		mcp->out_mb |= MBX_15;
 		/* debug q create issue in SR-IOV */
 		mcp->in_mb |= MBX_9 | MBX_8 | MBX_7;
@@ -3697,7 +3774,7 @@ qla25xx_init_req_que(struct scsi_qla_host *vha, struct req_que *req)
 	spin_lock_irqsave(&ha->hardware_lock, flags);
 	if (!(req->options & BIT_0)) {
 		WRT_REG_DWORD(req->req_q_in, 0);
-		if (!IS_QLA83XX(ha) || !IS_QLA27XX(ha))
+		if (!IS_QLA83XX(ha) && !IS_QLA27XX(ha))
 			WRT_REG_DWORD(req->req_q_out, 0);
 	}
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
@@ -3726,6 +3803,9 @@ qla25xx_init_rsp_que(struct scsi_qla_host *vha, struct rsp_que *rsp)
 	ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x10d6,
 	    "Entered %s.\n", __func__);
 
+	if (IS_SHADOW_REG_CAPABLE(ha))
+		rsp->options |= BIT_13;
+
 	mcp->mb[0] = MBC_INITIALIZE_MULTIQ;
 	mcp->mb[1] = rsp->options;
 	mcp->mb[2] = MSW(LSD(rsp->dma));
@@ -3740,7 +3820,7 @@ qla25xx_init_rsp_que(struct scsi_qla_host *vha, struct rsp_que *rsp)
 
 	mcp->mb[4] = rsp->id;
 	/* que in ptr index */
-	mcp->mb[8] = 0;
+	mcp->mb[8] = *rsp->in_ptr = 0;
 	/* que out ptr index */
 	mcp->mb[9] = 0;
 	mcp->out_mb = MBX_14|MBX_13|MBX_9|MBX_8|MBX_7
diff --git a/drivers/scsi/qla2xxx/qla_mid.c b/drivers/scsi/qla2xxx/qla_mid.c
index f0a852257f99..89998244f48d 100644
--- a/drivers/scsi/qla2xxx/qla_mid.c
+++ b/drivers/scsi/qla2xxx/qla_mid.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
diff --git a/drivers/scsi/qla2xxx/qla_mr.c b/drivers/scsi/qla2xxx/qla_mr.c
index 0aaf6a9c87d3..abeb3901498b 100644
--- a/drivers/scsi/qla2xxx/qla_mr.c
+++ b/drivers/scsi/qla2xxx/qla_mr.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -527,21 +527,63 @@ qlafx00_soc_cpu_reset(scsi_qla_host_t *vha)
 	struct qla_hw_data *ha = vha->hw;
 	int i, core;
 	uint32_t cnt;
+	uint32_t reg_val;
+
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x80004, 0);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x82004, 0);
+
+	/* stop the XOR DMA engines */
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x60920, 0x02);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x60924, 0x02);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0xf0920, 0x02);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0xf0924, 0x02);
+
+	/* stop the IDMA engines */
+	reg_val = QLAFX00_GET_HBA_SOC_REG(ha, 0x60840);
+	reg_val &= ~(1<<12);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x60840, reg_val);
+
+	reg_val = QLAFX00_GET_HBA_SOC_REG(ha, 0x60844);
+	reg_val &= ~(1<<12);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x60844, reg_val);
+
+	reg_val = QLAFX00_GET_HBA_SOC_REG(ha, 0x60848);
+	reg_val &= ~(1<<12);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x60848, reg_val);
+
+	reg_val = QLAFX00_GET_HBA_SOC_REG(ha, 0x6084C);
+	reg_val &= ~(1<<12);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x6084C, reg_val);
+
+	for (i = 0; i < 100000; i++) {
+		if ((QLAFX00_GET_HBA_SOC_REG(ha, 0xd0000) & 0x10000000) == 0 &&
+		    (QLAFX00_GET_HBA_SOC_REG(ha, 0x10600) & 0x1) == 0)
+			break;
+		udelay(100);
+	}
 
 	/* Set all 4 cores in reset */
 	for (i = 0; i < 4; i++) {
 		QLAFX00_SET_HBA_SOC_REG(ha,
 		    (SOC_SW_RST_CONTROL_REG_CORE0 + 8*i), (0xF01));
-	}
-
-	/* Set all 4 core Clock gating control */
-	for (i = 0; i < 4; i++) {
 		QLAFX00_SET_HBA_SOC_REG(ha,
 		    (SOC_SW_RST_CONTROL_REG_CORE0 + 4 + 8*i), (0x01010101));
 	}
 
 	/* Reset all units in Fabric */
-	QLAFX00_SET_HBA_SOC_REG(ha, SOC_FABRIC_RST_CONTROL_REG, (0x11F0101));
+	QLAFX00_SET_HBA_SOC_REG(ha, SOC_FABRIC_RST_CONTROL_REG, (0x011f0101));
+
+	/* */
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x10610, 1);
+	QLAFX00_SET_HBA_SOC_REG(ha, 0x10600, 0);
+
+	/* Set all 4 core Memory Power Down Registers */
+	for (i = 0; i < 5; i++) {
+		QLAFX00_SET_HBA_SOC_REG(ha,
+		    (SOC_PWR_MANAGEMENT_PWR_DOWN_REG + 4*i), (0x0));
+	}
 
 	/* Reset all interrupt control registers */
 	for (i = 0; i < 115; i++) {
@@ -564,20 +606,19 @@ qlafx00_soc_cpu_reset(scsi_qla_host_t *vha)
 	QLAFX00_SET_HBA_SOC_REG(ha, SOC_FABRIC_CONTROL_REG, (0x2));
 	QLAFX00_SET_HBA_SOC_REG(ha, SOC_FABRIC_CONFIG_REG, (0x3));
 
-	spin_lock_irqsave(&ha->hardware_lock, flags);
-
 	/* Kick in Fabric units */
 	QLAFX00_SET_HBA_SOC_REG(ha, SOC_FABRIC_RST_CONTROL_REG, (0x0));
 
 	/* Kick in Core0 to start boot process */
 	QLAFX00_SET_HBA_SOC_REG(ha, SOC_SW_RST_CONTROL_REG_CORE0, (0xF00));
 
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+
 	/* Wait 10secs for soft-reset to complete. */
 	for (cnt = 10; cnt; cnt--) {
 		msleep(1000);
 		barrier();
 	}
-	spin_unlock_irqrestore(&ha->hardware_lock, flags);
 }
 
 /**
@@ -597,7 +638,6 @@ qlafx00_soft_reset(scsi_qla_host_t *vha)
 
 	ha->isp_ops->disable_intrs(ha);
 	qlafx00_soc_cpu_reset(vha);
-	ha->isp_ops->enable_intrs(ha);
 }
 
 /**
@@ -2675,7 +2715,7 @@ qlafx00_process_response_queue(struct scsi_qla_host *vha,
 	uint16_t lreq_q_out = 0;
 
 	lreq_q_in = RD_REG_DWORD(rsp->rsp_q_in);
-	lreq_q_out = RD_REG_DWORD(rsp->rsp_q_out);
+	lreq_q_out = rsp->ring_index;
 
 	while (lreq_q_in != lreq_q_out) {
 		lptr = rsp->ring_ptr;
@@ -3426,7 +3466,7 @@ qlafx00_fxdisc_iocb(srb_t *sp, struct fxdisc_entry_fx00 *pfxiocb)
 	    sp->fcport->vha, 0x3047,
 	    (uint8_t *)&fx_iocb, sizeof(struct fxdisc_entry_fx00));
 
-	memcpy((void *)pfxiocb, &fx_iocb,
+	memcpy_toio((void __iomem *)pfxiocb, &fx_iocb,
 	    sizeof(struct fxdisc_entry_fx00));
 	wmb();
 }
diff --git a/drivers/scsi/qla2xxx/qla_mr.h b/drivers/scsi/qla2xxx/qla_mr.h
index e529dfaeb854..aeaa1b40b1fc 100644
--- a/drivers/scsi/qla2xxx/qla_mr.h
+++ b/drivers/scsi/qla2xxx/qla_mr.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -351,6 +351,7 @@ struct config_info_data {
 #define SOC_FABRIC_RST_CONTROL_REG       0x0020840
 #define SOC_FABRIC_CONTROL_REG           0x0020200
 #define SOC_FABRIC_CONFIG_REG            0x0020204
+#define SOC_PWR_MANAGEMENT_PWR_DOWN_REG  0x001820C
 
 #define SOC_INTERRUPT_SOURCE_I_CONTROL_REG     0x0020B00
 #define SOC_CORE_TIMER_REG                     0x0021850
diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c
index 5511e24b1f11..58f3c912d96e 100644
--- a/drivers/scsi/qla2xxx/qla_nx.c
+++ b/drivers/scsi/qla2xxx/qla_nx.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -848,6 +848,7 @@ qla82xx_rom_lock(struct qla_hw_data *ha)
 {
 	int done = 0, timeout = 0;
 	uint32_t lock_owner = 0;
+	scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev);
 
 	while (!done) {
 		/* acquire semaphore2 from PCI HW block */
@@ -856,17 +857,21 @@ qla82xx_rom_lock(struct qla_hw_data *ha)
 			break;
 		if (timeout >= qla82xx_rom_lock_timeout) {
 			lock_owner = qla82xx_rd_32(ha, QLA82XX_ROM_LOCK_ID);
+			ql_log(ql_log_warn, vha, 0xb157,
+			    "%s: Simultaneous flash access by following ports, active port = %d: accessing port = %d",
+			    __func__, ha->portnum, lock_owner);
 			return -1;
 		}
 		timeout++;
 	}
-	qla82xx_wr_32(ha, QLA82XX_ROM_LOCK_ID, ROM_LOCK_DRIVER);
+	qla82xx_wr_32(ha, QLA82XX_ROM_LOCK_ID, ha->portnum);
 	return 0;
 }
 
 static void
 qla82xx_rom_unlock(struct qla_hw_data *ha)
 {
+	qla82xx_wr_32(ha, QLA82XX_ROM_LOCK_ID, 0xffffffff);
 	qla82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM2_UNLOCK));
 }
 
@@ -950,6 +955,7 @@ static int
 qla82xx_rom_fast_read(struct qla_hw_data *ha, int addr, int *valp)
 {
 	int ret, loops = 0;
+	uint32_t lock_owner = 0;
 	scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev);
 
 	while ((qla82xx_rom_lock(ha) != 0) && (loops < 50000)) {
@@ -958,8 +964,10 @@ qla82xx_rom_fast_read(struct qla_hw_data *ha, int addr, int *valp)
 		loops++;
 	}
 	if (loops >= 50000) {
+		lock_owner = qla82xx_rd_32(ha, QLA82XX_ROM_LOCK_ID);
 		ql_log(ql_log_fatal, vha, 0x00b9,
-		    "Failed to acquire SEM2 lock.\n");
+		    "Failed to acquire SEM2 lock, Lock Owner %u.\n",
+		    lock_owner);
 		return -1;
 	}
 	ret = qla82xx_do_rom_fast_read(ha, addr, valp);
@@ -1057,6 +1065,7 @@ static int
 ql82xx_rom_lock_d(struct qla_hw_data *ha)
 {
 	int loops = 0;
+	uint32_t lock_owner = 0;
 	scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev);
 
 	while ((qla82xx_rom_lock(ha) != 0) && (loops < 50000)) {
@@ -1065,8 +1074,9 @@ ql82xx_rom_lock_d(struct qla_hw_data *ha)
 		loops++;
 	}
 	if (loops >= 50000) {
+		lock_owner = qla82xx_rd_32(ha, QLA82XX_ROM_LOCK_ID);
 		ql_log(ql_log_warn, vha, 0xb010,
-		    "ROM lock failed.\n");
+		    "ROM lock failed, Lock Owner %u.\n", lock_owner);
 		return -1;
 	}
 	return 0;
@@ -2811,12 +2821,14 @@ static void
 qla82xx_rom_lock_recovery(struct qla_hw_data *ha)
 {
 	scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev);
+	uint32_t lock_owner = 0;
 
-	if (qla82xx_rom_lock(ha))
+	if (qla82xx_rom_lock(ha)) {
+		lock_owner = qla82xx_rd_32(ha, QLA82XX_ROM_LOCK_ID);
 		/* Someone else is holding the lock. */
 		ql_log(ql_log_info, vha, 0xb022,
-		    "Resetting rom_lock.\n");
-
+		    "Resetting rom_lock, Lock Owner %u.\n", lock_owner);
+	}
 	/*
 	 * Either we got the lock, or someone
 	 * else died while holding it.
@@ -2840,47 +2852,30 @@ static int
 qla82xx_device_bootstrap(scsi_qla_host_t *vha)
 {
 	int rval = QLA_SUCCESS;
-	int i, timeout;
+	int i;
 	uint32_t old_count, count;
 	struct qla_hw_data *ha = vha->hw;
-	int need_reset = 0, peg_stuck = 1;
+	int need_reset = 0;
 
 	need_reset = qla82xx_need_reset(ha);
 
-	old_count = qla82xx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER);
-
-	for (i = 0; i < 10; i++) {
-		timeout = msleep_interruptible(200);
-		if (timeout) {
-			qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE,
-				QLA8XXX_DEV_FAILED);
-			return QLA_FUNCTION_FAILED;
-		}
-
-		count = qla82xx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER);
-		if (count != old_count)
-			peg_stuck = 0;
-	}
-
 	if (need_reset) {
 		/* We are trying to perform a recovery here. */
-		if (peg_stuck)
+		if (ha->flags.isp82xx_fw_hung)
 			qla82xx_rom_lock_recovery(ha);
-		goto dev_initialize;
 	} else  {
-		/* Start of day for this ha context. */
-		if (peg_stuck) {
-			/* Either we are the first or recovery in progress. */
-			qla82xx_rom_lock_recovery(ha);
-			goto dev_initialize;
-		} else
-			/* Firmware already running. */
-			goto dev_ready;
+		old_count = qla82xx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER);
+		for (i = 0; i < 10; i++) {
+			msleep(200);
+			count = qla82xx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER);
+			if (count != old_count) {
+				rval = QLA_SUCCESS;
+				goto dev_ready;
+			}
+		}
+		qla82xx_rom_lock_recovery(ha);
 	}
 
-	return rval;
-
-dev_initialize:
 	/* set to DEV_INITIALIZING */
 	ql_log(ql_log_info, vha, 0x009e,
 	    "HW State: INITIALIZING.\n");
@@ -3142,18 +3137,18 @@ qla82xx_check_md_needed(scsi_qla_host_t *vha)
 
 	if (ql2xmdenable) {
 		if (!ha->fw_dumped) {
-			if (fw_major_version != ha->fw_major_version ||
+			if ((fw_major_version != ha->fw_major_version ||
 			    fw_minor_version != ha->fw_minor_version ||
-			    fw_subminor_version != ha->fw_subminor_version) {
+			    fw_subminor_version != ha->fw_subminor_version) ||
+			    (ha->prev_minidump_failed)) {
 				ql_dbg(ql_dbg_p3p, vha, 0xb02d,
-				    "Firmware version differs "
-				    "Previous version: %d:%d:%d - "
-				    "New version: %d:%d:%d\n",
+				    "Firmware version differs Previous version: %d:%d:%d - New version: %d:%d:%d, prev_minidump_failed: %d.\n",
 				    fw_major_version, fw_minor_version,
 				    fw_subminor_version,
 				    ha->fw_major_version,
 				    ha->fw_minor_version,
-				    ha->fw_subminor_version);
+				    ha->fw_subminor_version,
+				    ha->prev_minidump_failed);
 				/* Release MiniDump resources */
 				qla82xx_md_free(vha);
 				/* ALlocate MiniDump resources */
@@ -3682,8 +3677,10 @@ qla82xx_chip_reset_cleanup(scsi_qla_host_t *vha)
 			for (cnt = 1; cnt < req->num_outstanding_cmds; cnt++) {
 				sp = req->outstanding_cmds[cnt];
 				if (sp) {
-					if (!sp->u.scmd.ctx ||
-					    (sp->flags & SRB_FCP_CMND_DMA_VALID)) {
+					if ((!sp->u.scmd.ctx ||
+					    (sp->flags &
+						SRB_FCP_CMND_DMA_VALID)) &&
+						!ha->flags.isp82xx_fw_hung) {
 						spin_unlock_irqrestore(
 						    &ha->hardware_lock, flags);
 						if (ha->isp_ops->abort_command(sp)) {
diff --git a/drivers/scsi/qla2xxx/qla_nx.h b/drivers/scsi/qla2xxx/qla_nx.h
index 1bb93dbbccbb..59c477883a73 100644
--- a/drivers/scsi/qla2xxx/qla_nx.h
+++ b/drivers/scsi/qla2xxx/qla_nx.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -333,9 +333,6 @@
 #define QLA82XX_ROMUSB_ROM_INSTR_OPCODE		(ROMUSB_ROM + 0x0004)
 #define QLA82XX_ROMUSB_GLB_CAS_RST		(ROMUSB_GLB + 0x0038)
 
-/* Lock IDs for ROM lock */
-#define ROM_LOCK_DRIVER       0x0d417340
-
 #define QLA82XX_PCI_CRB_WINDOWSIZE 0x00100000	 /* all are 1MB windows */
 #define QLA82XX_PCI_CRB_WINDOW(A) \
 	(QLA82XX_PCI_CRBSPACE + (A)*QLA82XX_PCI_CRB_WINDOWSIZE)
@@ -1186,6 +1183,7 @@ static const int MD_MIU_TEST_AGT_RDDATA[] = { 0x410000A8, 0x410000AC,
 #define CRB_NIU_XG_PAUSE_CTL_P1        0x8
 
 #define qla82xx_get_temp_val(x)          ((x) >> 16)
+#define qla82xx_get_temp_val1(x)          ((x) && 0x0000FFFF)
 #define qla82xx_get_temp_state(x)        ((x) & 0xffff)
 #define qla82xx_encode_temp(val, state)  (((val) << 16) | (state))
 
diff --git a/drivers/scsi/qla2xxx/qla_nx2.c b/drivers/scsi/qla2xxx/qla_nx2.c
index 86cf10815db0..da9e3902f219 100644
--- a/drivers/scsi/qla2xxx/qla_nx2.c
+++ b/drivers/scsi/qla2xxx/qla_nx2.c
@@ -1,17 +1,20 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
 
 #include <linux/vmalloc.h>
+#include <linux/delay.h>
 
 #include "qla_def.h"
 #include "qla_gbl.h"
 
 #include <linux/delay.h>
 
+#define TIMEOUT_100_MS 100
+
 /* 8044 Flash Read/Write functions */
 uint32_t
 qla8044_rd_reg(struct qla_hw_data *ha, ulong addr)
@@ -117,6 +120,95 @@ qla8044_read_write_crb_reg(struct scsi_qla_host *vha,
 	qla8044_wr_reg_indirect(vha, waddr, value);
 }
 
+static int
+qla8044_poll_wait_for_ready(struct scsi_qla_host *vha, uint32_t addr1,
+	uint32_t mask)
+{
+	unsigned long timeout;
+	uint32_t temp;
+
+	/* jiffies after 100ms */
+	timeout = jiffies + msecs_to_jiffies(TIMEOUT_100_MS);
+	do {
+		qla8044_rd_reg_indirect(vha, addr1, &temp);
+		if ((temp & mask) != 0)
+			break;
+		if (time_after_eq(jiffies, timeout)) {
+			ql_log(ql_log_warn, vha, 0xb151,
+				"Error in processing rdmdio entry\n");
+			return -1;
+		}
+	} while (1);
+
+	return 0;
+}
+
+static uint32_t
+qla8044_ipmdio_rd_reg(struct scsi_qla_host *vha,
+	uint32_t addr1, uint32_t addr3, uint32_t mask, uint32_t addr)
+{
+	uint32_t temp;
+	int ret = 0;
+
+	ret = qla8044_poll_wait_for_ready(vha, addr1, mask);
+	if (ret == -1)
+		return -1;
+
+	temp = (0x40000000 | addr);
+	qla8044_wr_reg_indirect(vha, addr1, temp);
+
+	ret = qla8044_poll_wait_for_ready(vha, addr1, mask);
+	if (ret == -1)
+		return 0;
+
+	qla8044_rd_reg_indirect(vha, addr3, &ret);
+
+	return ret;
+}
+
+
+static int
+qla8044_poll_wait_ipmdio_bus_idle(struct scsi_qla_host *vha,
+	uint32_t addr1, uint32_t addr2, uint32_t addr3, uint32_t mask)
+{
+	unsigned long timeout;
+	uint32_t temp;
+
+	/* jiffies after 100 msecs */
+	timeout = jiffies + msecs_to_jiffies(TIMEOUT_100_MS);
+	do {
+		temp = qla8044_ipmdio_rd_reg(vha, addr1, addr3, mask, addr2);
+		if ((temp & 0x1) != 1)
+			break;
+		if (time_after_eq(jiffies, timeout)) {
+			ql_log(ql_log_warn, vha, 0xb152,
+			    "Error in processing mdiobus idle\n");
+			return -1;
+		}
+	} while (1);
+
+	return 0;
+}
+
+static int
+qla8044_ipmdio_wr_reg(struct scsi_qla_host *vha, uint32_t addr1,
+	uint32_t addr3, uint32_t mask, uint32_t addr, uint32_t value)
+{
+	int ret = 0;
+
+	ret = qla8044_poll_wait_for_ready(vha, addr1, mask);
+	if (ret == -1)
+		return -1;
+
+	qla8044_wr_reg_indirect(vha, addr3, value);
+	qla8044_wr_reg_indirect(vha, addr1, addr);
+
+	ret = qla8044_poll_wait_for_ready(vha, addr1, mask);
+	if (ret == -1)
+		return -1;
+
+	return 0;
+}
 /*
  * qla8044_rmw_crb_reg - Read value from raddr, AND with test_mask,
  * Shift Left,Right/OR/XOR with values RMW header and write value to waddr.
@@ -356,8 +448,8 @@ qla8044_flash_lock(scsi_qla_host_t *vha)
 			lock_owner = qla8044_rd_reg(ha,
 			    QLA8044_FLASH_LOCK_ID);
 			ql_log(ql_log_warn, vha, 0xb113,
-			    "%s: flash lock by %d failed, held by %d\n",
-				__func__, ha->portnum, lock_owner);
+			    "%s: Simultaneous flash access by following ports, active port = %d: accessing port = %d",
+			    __func__, ha->portnum, lock_owner);
 			ret_val = QLA_FUNCTION_FAILED;
 			break;
 		}
@@ -1541,7 +1633,7 @@ static void
 qla8044_need_reset_handler(struct scsi_qla_host *vha)
 {
 	uint32_t dev_state = 0, drv_state, drv_active;
-	unsigned long reset_timeout, dev_init_timeout;
+	unsigned long reset_timeout;
 	struct qla_hw_data *ha = vha->hw;
 
 	ql_log(ql_log_fatal, vha, 0xb0c2,
@@ -1555,84 +1647,78 @@ qla8044_need_reset_handler(struct scsi_qla_host *vha)
 		qla8044_idc_lock(ha);
 	}
 
+	dev_state = qla8044_rd_direct(vha,
+	    QLA8044_CRB_DEV_STATE_INDEX);
 	drv_state = qla8044_rd_direct(vha,
 	    QLA8044_CRB_DRV_STATE_INDEX);
 	drv_active = qla8044_rd_direct(vha,
 	    QLA8044_CRB_DRV_ACTIVE_INDEX);
 
 	ql_log(ql_log_info, vha, 0xb0c5,
-	    "%s(%ld): drv_state = 0x%x, drv_active = 0x%x\n",
-	    __func__, vha->host_no, drv_state, drv_active);
+	    "%s(%ld): drv_state = 0x%x, drv_active = 0x%x dev_state = 0x%x\n",
+	    __func__, vha->host_no, drv_state, drv_active, dev_state);
 
-	if (!ha->flags.nic_core_reset_owner) {
-		ql_dbg(ql_dbg_p3p, vha, 0xb0c3,
-		    "%s(%ld): reset acknowledged\n",
-		    __func__, vha->host_no);
-		qla8044_set_rst_ready(vha);
+	qla8044_set_rst_ready(vha);
 
-		/* Non-reset owners ACK Reset and wait for device INIT state
-		 * as part of Reset Recovery by Reset Owner
-		 */
-		dev_init_timeout = jiffies + (ha->fcoe_reset_timeout * HZ);
+	/* wait for 10 seconds for reset ack from all functions */
+	reset_timeout = jiffies + (ha->fcoe_reset_timeout * HZ);
 
-		do {
-			if (time_after_eq(jiffies, dev_init_timeout)) {
-				ql_log(ql_log_info, vha, 0xb0c4,
-				    "%s: Non Reset owner: Reset Ack Timeout!\n",
-				    __func__);
-				break;
-			}
+	do {
+		if (time_after_eq(jiffies, reset_timeout)) {
+			ql_log(ql_log_info, vha, 0xb0c4,
+			    "%s: Function %d: Reset Ack Timeout!, drv_state: 0x%08x, drv_active: 0x%08x\n",
+			    __func__, ha->portnum, drv_state, drv_active);
+			break;
+		}
 
-			qla8044_idc_unlock(ha);
-			msleep(1000);
-			qla8044_idc_lock(ha);
+		qla8044_idc_unlock(ha);
+		msleep(1000);
+		qla8044_idc_lock(ha);
 
-			dev_state = qla8044_rd_direct(vha,
-					QLA8044_CRB_DEV_STATE_INDEX);
-		} while (((drv_state & drv_active) != drv_active) &&
-		    (dev_state == QLA8XXX_DEV_NEED_RESET));
+		dev_state = qla8044_rd_direct(vha,
+		    QLA8044_CRB_DEV_STATE_INDEX);
+		drv_state = qla8044_rd_direct(vha,
+		    QLA8044_CRB_DRV_STATE_INDEX);
+		drv_active = qla8044_rd_direct(vha,
+		    QLA8044_CRB_DRV_ACTIVE_INDEX);
+	} while (((drv_state & drv_active) != drv_active) &&
+	    (dev_state == QLA8XXX_DEV_NEED_RESET));
+
+	/* Remove IDC participation of functions not acknowledging */
+	if (drv_state != drv_active) {
+		ql_log(ql_log_info, vha, 0xb0c7,
+		    "%s(%ld): Function %d turning off drv_active of non-acking function 0x%x\n",
+		    __func__, vha->host_no, ha->portnum,
+		    (drv_active ^ drv_state));
+		drv_active = drv_active & drv_state;
+		qla8044_wr_direct(vha, QLA8044_CRB_DRV_ACTIVE_INDEX,
+		    drv_active);
 	} else {
-		qla8044_set_rst_ready(vha);
-
-		/* wait for 10 seconds for reset ack from all functions */
-		reset_timeout = jiffies + (ha->fcoe_reset_timeout * HZ);
-
-		while ((drv_state & drv_active) != drv_active) {
-			if (time_after_eq(jiffies, reset_timeout)) {
-				ql_log(ql_log_info, vha, 0xb0c6,
-				    "%s: RESET TIMEOUT!"
-				    "drv_state: 0x%08x, drv_active: 0x%08x\n",
-				    QLA2XXX_DRIVER_NAME, drv_state, drv_active);
-				break;
-			}
-
-			qla8044_idc_unlock(ha);
-			msleep(1000);
-			qla8044_idc_lock(ha);
-
-			drv_state = qla8044_rd_direct(vha,
-			    QLA8044_CRB_DRV_STATE_INDEX);
-			drv_active = qla8044_rd_direct(vha,
-			    QLA8044_CRB_DRV_ACTIVE_INDEX);
-		}
-
-		if (drv_state != drv_active) {
-			ql_log(ql_log_info, vha, 0xb0c7,
-			    "%s(%ld): Reset_owner turning off drv_active "
-			    "of non-acking function 0x%x\n", __func__,
-			    vha->host_no, (drv_active ^ drv_state));
-			drv_active = drv_active & drv_state;
-			qla8044_wr_direct(vha, QLA8044_CRB_DRV_ACTIVE_INDEX,
-			    drv_active);
+		/*
+		 * Reset owner should execute reset recovery,
+		 * if all functions acknowledged
+		 */
+		if ((ha->flags.nic_core_reset_owner) &&
+		    (dev_state == QLA8XXX_DEV_NEED_RESET)) {
+			ha->flags.nic_core_reset_owner = 0;
+			qla8044_device_bootstrap(vha);
+			return;
 		}
+	}
 
-		/*
-		* Clear RESET OWNER, will be set at next reset
-		* by next RST_OWNER
-		*/
+	/* Exit if non active function */
+	if (!(drv_active & (1 << ha->portnum))) {
 		ha->flags.nic_core_reset_owner = 0;
+		return;
+	}
 
-		/* Start Reset Recovery */
+	/*
+	 * Execute Reset Recovery if Reset Owner or Function 7
+	 * is the only active function
+	 */
+	if (ha->flags.nic_core_reset_owner ||
+	    ((drv_state & drv_active) == QLA8044_FUN7_ACTIVE_INDEX)) {
+		ha->flags.nic_core_reset_owner = 0;
 		qla8044_device_bootstrap(vha);
 	}
 }
@@ -1655,6 +1741,19 @@ qla8044_set_drv_active(struct scsi_qla_host *vha)
 	qla8044_wr_direct(vha, QLA8044_CRB_DRV_ACTIVE_INDEX, drv_active);
 }
 
+static int
+qla8044_check_drv_active(struct scsi_qla_host *vha)
+{
+	uint32_t drv_active;
+	struct qla_hw_data *ha = vha->hw;
+
+	drv_active = qla8044_rd_direct(vha, QLA8044_CRB_DRV_ACTIVE_INDEX);
+	if (drv_active & (1 << ha->portnum))
+		return QLA_SUCCESS;
+	else
+		return QLA_TEST_FAILED;
+}
+
 static void
 qla8044_clear_idc_dontreset(struct scsi_qla_host *vha)
 {
@@ -1837,14 +1936,16 @@ qla8044_device_state_handler(struct scsi_qla_host *vha)
 
 	while (1) {
 		if (time_after_eq(jiffies, dev_init_timeout)) {
-			ql_log(ql_log_warn, vha, 0xb0cf,
-			    "%s: Device Init Failed 0x%x = %s\n",
-			    QLA2XXX_DRIVER_NAME, dev_state,
-			    dev_state < MAX_STATES ?
-			    qdev_state(dev_state) : "Unknown");
-
-			qla8044_wr_direct(vha, QLA8044_CRB_DEV_STATE_INDEX,
-			    QLA8XXX_DEV_FAILED);
+			if (qla8044_check_drv_active(vha) == QLA_SUCCESS) {
+				ql_log(ql_log_warn, vha, 0xb0cf,
+				    "%s: Device Init Failed 0x%x = %s\n",
+				    QLA2XXX_DRIVER_NAME, dev_state,
+				    dev_state < MAX_STATES ?
+				    qdev_state(dev_state) : "Unknown");
+				qla8044_wr_direct(vha,
+				    QLA8044_CRB_DEV_STATE_INDEX,
+				    QLA8XXX_DEV_FAILED);
+			}
 		}
 
 		dev_state = qla8044_rd_direct(vha, QLA8044_CRB_DEV_STATE_INDEX);
@@ -2017,6 +2118,13 @@ qla8044_watchdog(struct scsi_qla_host *vha)
 	    test_bit(FCOE_CTX_RESET_NEEDED, &vha->dpc_flags))) {
 		dev_state = qla8044_rd_direct(vha, QLA8044_CRB_DEV_STATE_INDEX);
 
+		if (qla8044_check_fw_alive(vha)) {
+			ha->flags.isp82xx_fw_hung = 1;
+			ql_log(ql_log_warn, vha, 0xb10a,
+			    "Firmware hung.\n");
+			qla82xx_clear_pending_mbx(vha);
+		}
+
 		if (qla8044_check_temp(vha)) {
 			set_bit(ISP_UNRECOVERABLE, &vha->dpc_flags);
 			ha->flags.isp82xx_fw_hung = 1;
@@ -2037,7 +2145,7 @@ qla8044_watchdog(struct scsi_qla_host *vha)
 			qla2xxx_wake_dpc(vha);
 		} else  {
 			/* Check firmware health */
-			if (qla8044_check_fw_alive(vha)) {
+			if (ha->flags.isp82xx_fw_hung) {
 				halt_status = qla8044_rd_direct(vha,
 					QLA8044_PEG_HALT_STATUS1_INDEX);
 				if (halt_status &
@@ -2073,12 +2181,8 @@ qla8044_watchdog(struct scsi_qla_host *vha)
 						    __func__);
 						set_bit(ISP_ABORT_NEEDED,
 						    &vha->dpc_flags);
-						qla82xx_clear_pending_mbx(vha);
 					}
 				}
-				ha->flags.isp82xx_fw_hung = 1;
-				ql_log(ql_log_warn, vha, 0xb10a,
-				    "Firmware hung.\n");
 				qla2xxx_wake_dpc(vha);
 			}
 		}
@@ -2286,8 +2390,6 @@ qla8044_minidump_process_rdmem(struct scsi_qla_host *vha,
 		}
 
 		if (j >= MAX_CTL_CHECK) {
-			printk_ratelimited(KERN_ERR
-			    "%s: failed to read through agent\n", __func__);
 			write_unlock_irqrestore(&ha->hw_lock, flags);
 			return QLA_SUCCESS;
 		}
@@ -2882,6 +2984,231 @@ error_exit:
 	return rval;
 }
 
+static uint32_t
+qla8044_minidump_process_rddfe(struct scsi_qla_host *vha,
+	struct qla8044_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr)
+{
+	int loop_cnt;
+	uint32_t addr1, addr2, value, data, temp, wrVal;
+	uint8_t stride, stride2;
+	uint16_t count;
+	uint32_t poll, mask, data_size, modify_mask;
+	uint32_t wait_count = 0;
+
+	uint32_t *data_ptr = *d_ptr;
+
+	struct qla8044_minidump_entry_rddfe *rddfe;
+	rddfe = (struct qla8044_minidump_entry_rddfe *) entry_hdr;
+
+	addr1 = rddfe->addr_1;
+	value = rddfe->value;
+	stride = rddfe->stride;
+	stride2 = rddfe->stride2;
+	count = rddfe->count;
+
+	poll = rddfe->poll;
+	mask = rddfe->mask;
+	modify_mask = rddfe->modify_mask;
+	data_size = rddfe->data_size;
+
+	addr2 = addr1 + stride;
+
+	for (loop_cnt = 0x0; loop_cnt < count; loop_cnt++) {
+		qla8044_wr_reg_indirect(vha, addr1, (0x40000000 | value));
+
+		wait_count = 0;
+		while (wait_count < poll) {
+			qla8044_rd_reg_indirect(vha, addr1, &temp);
+			if ((temp & mask) != 0)
+				break;
+			wait_count++;
+		}
+
+		if (wait_count == poll) {
+			ql_log(ql_log_warn, vha, 0xb153,
+			    "%s: TIMEOUT\n", __func__);
+			goto error;
+		} else {
+			qla8044_rd_reg_indirect(vha, addr2, &temp);
+			temp = temp & modify_mask;
+			temp = (temp | ((loop_cnt << 16) | loop_cnt));
+			wrVal = ((temp << 16) | temp);
+
+			qla8044_wr_reg_indirect(vha, addr2, wrVal);
+			qla8044_wr_reg_indirect(vha, addr1, value);
+
+			wait_count = 0;
+			while (wait_count < poll) {
+				qla8044_rd_reg_indirect(vha, addr1, &temp);
+				if ((temp & mask) != 0)
+					break;
+				wait_count++;
+			}
+			if (wait_count == poll) {
+				ql_log(ql_log_warn, vha, 0xb154,
+				    "%s: TIMEOUT\n", __func__);
+				goto error;
+			}
+
+			qla8044_wr_reg_indirect(vha, addr1,
+			    ((0x40000000 | value) + stride2));
+			wait_count = 0;
+			while (wait_count < poll) {
+				qla8044_rd_reg_indirect(vha, addr1, &temp);
+				if ((temp & mask) != 0)
+					break;
+				wait_count++;
+			}
+
+			if (wait_count == poll) {
+				ql_log(ql_log_warn, vha, 0xb155,
+				    "%s: TIMEOUT\n", __func__);
+				goto error;
+			}
+
+			qla8044_rd_reg_indirect(vha, addr2, &data);
+
+			*data_ptr++ = wrVal;
+			*data_ptr++ = data;
+		}
+
+	}
+
+	*d_ptr = data_ptr;
+	return QLA_SUCCESS;
+
+error:
+	return -1;
+
+}
+
+static uint32_t
+qla8044_minidump_process_rdmdio(struct scsi_qla_host *vha,
+	struct qla8044_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr)
+{
+	int ret = 0;
+	uint32_t addr1, addr2, value1, value2, data, selVal;
+	uint8_t stride1, stride2;
+	uint32_t addr3, addr4, addr5, addr6, addr7;
+	uint16_t count, loop_cnt;
+	uint32_t poll, mask;
+	uint32_t *data_ptr = *d_ptr;
+
+	struct qla8044_minidump_entry_rdmdio *rdmdio;
+
+	rdmdio = (struct qla8044_minidump_entry_rdmdio *) entry_hdr;
+
+	addr1 = rdmdio->addr_1;
+	addr2 = rdmdio->addr_2;
+	value1 = rdmdio->value_1;
+	stride1 = rdmdio->stride_1;
+	stride2 = rdmdio->stride_2;
+	count = rdmdio->count;
+
+	poll = rdmdio->poll;
+	mask = rdmdio->mask;
+	value2 = rdmdio->value_2;
+
+	addr3 = addr1 + stride1;
+
+	for (loop_cnt = 0; loop_cnt < count; loop_cnt++) {
+		ret = qla8044_poll_wait_ipmdio_bus_idle(vha, addr1, addr2,
+		    addr3, mask);
+		if (ret == -1)
+			goto error;
+
+		addr4 = addr2 - stride1;
+		ret = qla8044_ipmdio_wr_reg(vha, addr1, addr3, mask, addr4,
+		    value2);
+		if (ret == -1)
+			goto error;
+
+		addr5 = addr2 - (2 * stride1);
+		ret = qla8044_ipmdio_wr_reg(vha, addr1, addr3, mask, addr5,
+		    value1);
+		if (ret == -1)
+			goto error;
+
+		addr6 = addr2 - (3 * stride1);
+		ret = qla8044_ipmdio_wr_reg(vha, addr1, addr3, mask,
+		    addr6, 0x2);
+		if (ret == -1)
+			goto error;
+
+		ret = qla8044_poll_wait_ipmdio_bus_idle(vha, addr1, addr2,
+		    addr3, mask);
+		if (ret == -1)
+			goto error;
+
+		addr7 = addr2 - (4 * stride1);
+			data = qla8044_ipmdio_rd_reg(vha, addr1, addr3,
+			    mask, addr7);
+		if (data == -1)
+			goto error;
+
+		selVal = (value2 << 18) | (value1 << 2) | 2;
+
+		stride2 = rdmdio->stride_2;
+		*data_ptr++ = selVal;
+		*data_ptr++ = data;
+
+		value1 = value1 + stride2;
+		*d_ptr = data_ptr;
+	}
+
+	return 0;
+
+error:
+	return -1;
+}
+
+static uint32_t qla8044_minidump_process_pollwr(struct scsi_qla_host *vha,
+		struct qla8044_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr)
+{
+	uint32_t addr1, addr2, value1, value2, poll, mask, r_value;
+	uint32_t wait_count = 0;
+	struct qla8044_minidump_entry_pollwr *pollwr_hdr;
+
+	pollwr_hdr = (struct qla8044_minidump_entry_pollwr *)entry_hdr;
+	addr1 = pollwr_hdr->addr_1;
+	addr2 = pollwr_hdr->addr_2;
+	value1 = pollwr_hdr->value_1;
+	value2 = pollwr_hdr->value_2;
+
+	poll = pollwr_hdr->poll;
+	mask = pollwr_hdr->mask;
+
+	while (wait_count < poll) {
+		qla8044_rd_reg_indirect(vha, addr1, &r_value);
+
+		if ((r_value & poll) != 0)
+			break;
+		wait_count++;
+	}
+
+	if (wait_count == poll) {
+		ql_log(ql_log_warn, vha, 0xb156, "%s: TIMEOUT\n", __func__);
+		goto error;
+	}
+
+	qla8044_wr_reg_indirect(vha, addr2, value2);
+	qla8044_wr_reg_indirect(vha, addr1, value1);
+
+	wait_count = 0;
+	while (wait_count < poll) {
+		qla8044_rd_reg_indirect(vha, addr1, &r_value);
+
+		if ((r_value & poll) != 0)
+			break;
+		wait_count++;
+	}
+
+	return QLA_SUCCESS;
+
+error:
+	return -1;
+}
+
 /*
  *
  * qla8044_collect_md_data - Retrieve firmware minidump data.
@@ -3089,6 +3416,24 @@ qla8044_collect_md_data(struct scsi_qla_host *vha)
 			if (rval != QLA_SUCCESS)
 				qla8044_mark_entry_skipped(vha, entry_hdr, i);
 			break;
+		case QLA8044_RDDFE:
+			rval = qla8044_minidump_process_rddfe(vha, entry_hdr,
+			    &data_ptr);
+			if (rval != QLA_SUCCESS)
+				qla8044_mark_entry_skipped(vha, entry_hdr, i);
+			break;
+		case QLA8044_RDMDIO:
+			rval = qla8044_minidump_process_rdmdio(vha, entry_hdr,
+			    &data_ptr);
+			if (rval != QLA_SUCCESS)
+				qla8044_mark_entry_skipped(vha, entry_hdr, i);
+			break;
+		case QLA8044_POLLWR:
+			rval = qla8044_minidump_process_pollwr(vha, entry_hdr,
+			    &data_ptr);
+			if (rval != QLA_SUCCESS)
+				qla8044_mark_entry_skipped(vha, entry_hdr, i);
+			break;
 		case QLA82XX_RDNOP:
 		default:
 			qla8044_mark_entry_skipped(vha, entry_hdr, i);
@@ -3110,6 +3455,7 @@ skip_nxt_entry:
 		    "Dump data mismatch: Data collected: "
 		    "[0x%x], total_data_size:[0x%x]\n",
 		    data_collected, ha->md_dump_size);
+		rval = QLA_FUNCTION_FAILED;
 		goto md_failed;
 	}
 
@@ -3134,10 +3480,12 @@ qla8044_get_minidump(struct scsi_qla_host *vha)
 
 	if (!qla8044_collect_md_data(vha)) {
 		ha->fw_dumped = 1;
+		ha->prev_minidump_failed = 0;
 	} else {
 		ql_log(ql_log_fatal, vha, 0xb0db,
 		    "%s: Unable to collect minidump\n",
 		    __func__);
+		ha->prev_minidump_failed = 1;
 	}
 }
 
diff --git a/drivers/scsi/qla2xxx/qla_nx2.h b/drivers/scsi/qla2xxx/qla_nx2.h
index 2ab2eabab908..ada36057d7cd 100644
--- a/drivers/scsi/qla2xxx/qla_nx2.h
+++ b/drivers/scsi/qla2xxx/qla_nx2.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -133,6 +133,7 @@
 #define QLA8044_LINK_SPEED(f)		(0x36E0+(((f) >> 2) * 4))
 #define QLA8044_MAX_LINK_SPEED(f)       (0x36F0+(((f) / 4) * 4))
 #define QLA8044_LINK_SPEED_FACTOR	10
+#define QLA8044_FUN7_ACTIVE_INDEX	0x80
 
 /* FLASH API Defines */
 #define QLA8044_FLASH_MAX_WAIT_USEC	100
@@ -431,6 +432,50 @@ struct qla8044_minidump_entry_pollrd {
 	uint32_t rsvd_1;
 } __packed;
 
+struct qla8044_minidump_entry_rddfe {
+	struct qla8044_minidump_entry_hdr h;
+	uint32_t addr_1;
+	uint32_t value;
+	uint8_t stride;
+	uint8_t stride2;
+	uint16_t count;
+	uint32_t poll;
+	uint32_t mask;
+	uint32_t modify_mask;
+	uint32_t data_size;
+	uint32_t rsvd;
+
+} __packed;
+
+struct qla8044_minidump_entry_rdmdio {
+	struct qla8044_minidump_entry_hdr h;
+
+	uint32_t addr_1;
+	uint32_t addr_2;
+	uint32_t value_1;
+	uint8_t stride_1;
+	uint8_t stride_2;
+	uint16_t count;
+	uint32_t poll;
+	uint32_t mask;
+	uint32_t value_2;
+	uint32_t data_size;
+
+} __packed;
+
+struct qla8044_minidump_entry_pollwr {
+	struct qla8044_minidump_entry_hdr h;
+	uint32_t addr_1;
+	uint32_t addr_2;
+	uint32_t value_1;
+	uint32_t value_2;
+	uint32_t poll;
+	uint32_t mask;
+	uint32_t data_size;
+	uint32_t rsvd;
+
+}  __packed;
+
 /* RDMUX2 Entry */
 struct qla8044_minidump_entry_rdmux2 {
 	struct qla8044_minidump_entry_hdr h;
@@ -516,6 +561,9 @@ static const uint32_t qla8044_reg_tbl[] = {
 #define QLA8044_DBG_RSVD_ARRAY_LEN              8
 #define QLA8044_DBG_OCM_WNDREG_ARRAY_LEN        16
 #define QLA8044_SS_PCI_INDEX                    0
+#define QLA8044_RDDFE          38
+#define QLA8044_RDMDIO         39
+#define QLA8044_POLLWR         40
 
 struct qla8044_minidump_template_hdr {
 	uint32_t entry_type;
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index afc84814e9bb..d96bfb55e57b 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -616,7 +616,7 @@ qla2x00_sp_free_dma(void *vha, void *ptr)
 
 	if (sp->flags & SRB_CRC_CTX_DSD_VALID) {
 		/* List assured to be having elements */
-		qla2x00_clean_dsd_pool(ha, sp);
+		qla2x00_clean_dsd_pool(ha, sp, NULL);
 		sp->flags &= ~SRB_CRC_CTX_DSD_VALID;
 	}
 
@@ -781,7 +781,7 @@ static int
 qla2x00_eh_wait_on_command(struct scsi_cmnd *cmd)
 {
 #define ABORT_POLLING_PERIOD	1000
-#define ABORT_WAIT_ITER		((10 * 1000) / (ABORT_POLLING_PERIOD))
+#define ABORT_WAIT_ITER		((2 * 1000) / (ABORT_POLLING_PERIOD))
 	unsigned long wait_iter = ABORT_WAIT_ITER;
 	scsi_qla_host_t *vha = shost_priv(cmd->device->host);
 	struct qla_hw_data *ha = vha->hw;
@@ -844,11 +844,8 @@ qla2x00_wait_for_hba_online(scsi_qla_host_t *vha)
 }
 
 /*
- * qla2x00_wait_for_reset_ready
- *    Wait till the HBA is online after going through
- *    <= MAX_RETRIES_OF_ISP_ABORT  or
- *    finally HBA is disabled ie marked offline or flash
- *    operations are in progress.
+ * qla2x00_wait_for_hba_ready
+ * Wait till the HBA is ready before doing driver unload
  *
  * Input:
  *     ha - pointer to host adapter structure
@@ -857,35 +854,15 @@ qla2x00_wait_for_hba_online(scsi_qla_host_t *vha)
  *    Does context switching-Release SPIN_LOCK
  *    (if any) before calling this routine.
  *
- * Return:
- *    Success (Adapter is online/no flash ops) : 0
- *    Failed  (Adapter is offline/disabled/flash ops in progress) : 1
  */
-static int
-qla2x00_wait_for_reset_ready(scsi_qla_host_t *vha)
+static void
+qla2x00_wait_for_hba_ready(scsi_qla_host_t *vha)
 {
-	int		return_status;
-	unsigned long	wait_online;
 	struct qla_hw_data *ha = vha->hw;
-	scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev);
 
-	wait_online = jiffies + (MAX_LOOP_TIMEOUT * HZ);
-	while (((test_bit(ISP_ABORT_NEEDED, &base_vha->dpc_flags)) ||
-	    test_bit(ABORT_ISP_ACTIVE, &base_vha->dpc_flags) ||
-	    test_bit(ISP_ABORT_RETRY, &base_vha->dpc_flags) ||
-	    ha->optrom_state != QLA_SWAITING ||
-	    ha->dpc_active) && time_before(jiffies, wait_online))
+	while ((!(vha->flags.online) || ha->dpc_active ||
+	    ha->flags.mbox_busy))
 		msleep(1000);
-
-	if (base_vha->flags.online &&  ha->optrom_state == QLA_SWAITING)
-		return_status = QLA_SUCCESS;
-	else
-		return_status = QLA_FUNCTION_FAILED;
-
-	ql_dbg(ql_dbg_taskm, vha, 0x8019,
-	    "%s return status=%d.\n", __func__, return_status);
-
-	return return_status;
 }
 
 int
@@ -945,7 +922,7 @@ qla2xxx_eh_abort(struct scsi_cmnd *cmd)
 	int ret;
 	unsigned int id, lun;
 	unsigned long flags;
-	int wait = 0;
+	int rval, wait = 0;
 	struct qla_hw_data *ha = vha->hw;
 
 	if (!CMD_SP(cmd))
@@ -974,10 +951,20 @@ qla2xxx_eh_abort(struct scsi_cmnd *cmd)
 	sp_get(sp);
 
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
-	if (ha->isp_ops->abort_command(sp)) {
-		ret = FAILED;
+	rval = ha->isp_ops->abort_command(sp);
+	if (rval) {
+		if (rval == QLA_FUNCTION_PARAMETER_ERROR) {
+			/*
+			 * Decrement the ref_count since we can't find the
+			 * command
+			 */
+			atomic_dec(&sp->ref_count);
+			ret = SUCCESS;
+		} else
+			ret = FAILED;
+
 		ql_dbg(ql_dbg_taskm, vha, 0x8003,
-		    "Abort command mbx failed cmd=%p.\n", cmd);
+		    "Abort command mbx failed cmd=%p, rval=%x.\n", cmd, rval);
 	} else {
 		ql_dbg(ql_dbg_taskm, vha, 0x8004,
 		    "Abort command mbx success cmd=%p.\n", cmd);
@@ -985,6 +972,12 @@ qla2xxx_eh_abort(struct scsi_cmnd *cmd)
 	}
 
 	spin_lock_irqsave(&ha->hardware_lock, flags);
+	/*
+	 * Clear the slot in the oustanding_cmds array if we can't find the
+	 * command to reclaim the resources.
+	 */
+	if (rval == QLA_FUNCTION_PARAMETER_ERROR)
+		vha->req->outstanding_cmds[sp->handle] = NULL;
 	sp->done(ha, sp, 0);
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
 
@@ -1236,7 +1229,11 @@ qla2xxx_eh_host_reset(struct scsi_cmnd *cmd)
 	ql_log(ql_log_info, vha, 0x8018,
 	    "ADAPTER RESET ISSUED nexus=%ld:%d:%d.\n", vha->host_no, id, lun);
 
-	if (qla2x00_wait_for_reset_ready(vha) != QLA_SUCCESS)
+	/*
+	 * No point in issuing another reset if one is active.  Also do not
+	 * attempt a reset if we are updating flash.
+	 */
+	if (qla2x00_reset_active(vha) || ha->optrom_state != QLA_SWAITING)
 		goto eh_host_reset_lock;
 
 	if (vha != base_vha) {
@@ -2270,6 +2267,13 @@ qla2x00_set_isp_flags(struct qla_hw_data *ha)
 		ha->device_type |= DT_IIDMA;
 		ha->fw_srisc_address = RISC_START_ADDRESS_2400;
 		break;
+	case PCI_DEVICE_ID_QLOGIC_ISP2271:
+		ha->device_type |= DT_ISP2271;
+		ha->device_type |= DT_ZIO_SUPPORTED;
+		ha->device_type |= DT_FWI2;
+		ha->device_type |= DT_IIDMA;
+		ha->fw_srisc_address = RISC_START_ADDRESS_2400;
+		break;
 	}
 
 	if (IS_QLA82XX(ha))
@@ -2346,7 +2350,8 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	    pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8031 ||
 	    pdev->device == PCI_DEVICE_ID_QLOGIC_ISPF001 ||
 	    pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8044 ||
-	    pdev->device == PCI_DEVICE_ID_QLOGIC_ISP2071) {
+	    pdev->device == PCI_DEVICE_ID_QLOGIC_ISP2071 ||
+	    pdev->device == PCI_DEVICE_ID_QLOGIC_ISP2271) {
 		bars = pci_select_bars(pdev, IORESOURCE_MEM);
 		mem_only = 1;
 		ql_dbg_pci(ql_dbg_init, pdev, 0x0007,
@@ -2877,6 +2882,7 @@ skip_dpc:
 
 	base_vha->flags.init_done = 1;
 	base_vha->flags.online = 1;
+	ha->prev_minidump_failed = 0;
 
 	ql_dbg(ql_dbg_init, base_vha, 0x00f2,
 	    "Init done and hba is online.\n");
@@ -3136,6 +3142,8 @@ qla2x00_remove_one(struct pci_dev *pdev)
 	base_vha = pci_get_drvdata(pdev);
 	ha = base_vha->hw;
 
+	qla2x00_wait_for_hba_ready(base_vha);
+
 	set_bit(UNLOADING, &base_vha->dpc_flags);
 
 	if (IS_QLAFX00(ha))
@@ -3645,6 +3653,7 @@ qla2x00_free_fw_dump(struct qla_hw_data *ha)
 	ha->eft = NULL;
 	ha->eft_dma = 0;
 	ha->fw_dumped = 0;
+	ha->fw_dump_cap_flags = 0;
 	ha->fw_dump_reading = 0;
 	ha->fw_dump = NULL;
 	ha->fw_dump_len = 0;
@@ -4913,12 +4922,13 @@ qla2x00_do_dpc(void *data)
 				if (qlafx00_reset_initialize(base_vha)) {
 					/* Failed. Abort isp later. */
 					if (!test_bit(UNLOADING,
-					    &base_vha->dpc_flags))
+					    &base_vha->dpc_flags)) {
 						set_bit(ISP_UNRECOVERABLE,
 						    &base_vha->dpc_flags);
 						ql_dbg(ql_dbg_dpc, base_vha,
 						    0x4021,
 						    "Reset Recovery Failed\n");
+					}
 				}
 			}
 
@@ -5077,8 +5087,10 @@ intr_on_check:
 			ha->isp_ops->enable_intrs(ha);
 
 		if (test_and_clear_bit(BEACON_BLINK_NEEDED,
-					&base_vha->dpc_flags))
-			ha->isp_ops->beacon_blink(base_vha);
+					&base_vha->dpc_flags)) {
+			if (ha->beacon_blink_led == 1)
+				ha->isp_ops->beacon_blink(base_vha);
+		}
 
 		if (!IS_QLAFX00(ha))
 			qla2x00_do_dpc_all_vps(base_vha);
@@ -5325,7 +5337,7 @@ qla2x00_timer(scsi_qla_host_t *vha)
 #define FW_ISP82XX	7
 #define FW_ISP2031	8
 #define FW_ISP8031	9
-#define FW_ISP2071	10
+#define FW_ISP27XX	10
 
 #define FW_FILE_ISP21XX	"ql2100_fw.bin"
 #define FW_FILE_ISP22XX	"ql2200_fw.bin"
@@ -5337,7 +5349,7 @@ qla2x00_timer(scsi_qla_host_t *vha)
 #define FW_FILE_ISP82XX	"ql8200_fw.bin"
 #define FW_FILE_ISP2031	"ql2600_fw.bin"
 #define FW_FILE_ISP8031	"ql8300_fw.bin"
-#define FW_FILE_ISP2071	"ql2700_fw.bin"
+#define FW_FILE_ISP27XX	"ql2700_fw.bin"
 
 
 static DEFINE_MUTEX(qla_fw_lock);
@@ -5353,7 +5365,7 @@ static struct fw_blob qla_fw_blobs[FW_BLOBS] = {
 	{ .name = FW_FILE_ISP82XX, },
 	{ .name = FW_FILE_ISP2031, },
 	{ .name = FW_FILE_ISP8031, },
-	{ .name = FW_FILE_ISP2071, },
+	{ .name = FW_FILE_ISP27XX, },
 };
 
 struct fw_blob *
@@ -5382,8 +5394,8 @@ qla2x00_request_firmware(scsi_qla_host_t *vha)
 		blob = &qla_fw_blobs[FW_ISP2031];
 	} else if (IS_QLA8031(ha)) {
 		blob = &qla_fw_blobs[FW_ISP8031];
-	} else if (IS_QLA2071(ha)) {
-		blob = &qla_fw_blobs[FW_ISP2071];
+	} else if (IS_QLA27XX(ha)) {
+		blob = &qla_fw_blobs[FW_ISP27XX];
 	} else {
 		return NULL;
 	}
@@ -5714,6 +5726,7 @@ static struct pci_device_id qla2xxx_pci_tbl[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISPF001) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP8044) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP2071) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP2271) },
 	{ 0 },
 };
 MODULE_DEVICE_TABLE(pci, qla2xxx_pci_tbl);
diff --git a/drivers/scsi/qla2xxx/qla_settings.h b/drivers/scsi/qla2xxx/qla_settings.h
index 46ef0ac48f44..2fb7ebfbbc38 100644
--- a/drivers/scsi/qla2xxx/qla_settings.h
+++ b/drivers/scsi/qla2xxx/qla_settings.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c
index f28123e8ed65..bca173e56f16 100644
--- a/drivers/scsi/qla2xxx/qla_sup.c
+++ b/drivers/scsi/qla2xxx/qla_sup.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -1727,11 +1727,8 @@ qla83xx_beacon_blink(struct scsi_qla_host *vha)
 	if (IS_QLA2031(ha)) {
 		led_select_value = qla83xx_select_led_port(ha);
 
-		qla83xx_wr_reg(vha, led_select_value, 0x40002000);
-		qla83xx_wr_reg(vha, led_select_value + 4, 0x40002000);
-		msleep(1000);
-		qla83xx_wr_reg(vha, led_select_value, 0x40004000);
-		qla83xx_wr_reg(vha, led_select_value + 4, 0x40004000);
+		qla83xx_wr_reg(vha, led_select_value, 0x40000230);
+		qla83xx_wr_reg(vha, led_select_value + 4, 0x40000230);
 	} else if (IS_QLA8031(ha)) {
 		led_select_value = qla83xx_select_led_port(ha);
 
diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c
index 0cb73074c199..e632e14180cf 100644
--- a/drivers/scsi/qla2xxx/qla_target.c
+++ b/drivers/scsi/qla2xxx/qla_target.c
@@ -104,7 +104,6 @@ static void qlt_reject_free_srr_imm(struct scsi_qla_host *ha,
 /*
  * Global Variables
  */
-static struct kmem_cache *qla_tgt_cmd_cachep;
 static struct kmem_cache *qla_tgt_mgmt_cmd_cachep;
 static mempool_t *qla_tgt_mgmt_cmd_mempool;
 static struct workqueue_struct *qla_tgt_wq;
@@ -182,6 +181,11 @@ struct scsi_qla_host *qlt_find_host_by_vp_idx(struct scsi_qla_host *vha,
 void qlt_24xx_atio_pkt_all_vps(struct scsi_qla_host *vha,
 	struct atio_from_isp *atio)
 {
+	ql_dbg(ql_dbg_tgt, vha, 0xe072,
+		"%s: qla_target(%d): type %x ox_id %04x\n",
+		__func__, vha->vp_idx, atio->u.raw.entry_type,
+		be16_to_cpu(atio->u.isp24.fcp_hdr.ox_id));
+
 	switch (atio->u.raw.entry_type) {
 	case ATIO_TYPE7:
 	{
@@ -236,6 +240,10 @@ void qlt_24xx_atio_pkt_all_vps(struct scsi_qla_host *vha,
 void qlt_response_pkt_all_vps(struct scsi_qla_host *vha, response_t *pkt)
 {
 	switch (pkt->entry_type) {
+	case CTIO_CRC2:
+		ql_dbg(ql_dbg_tgt, vha, 0xe073,
+			"qla_target(%d):%s: CRC2 Response pkt\n",
+			vha->vp_idx, __func__);
 	case CTIO_TYPE7:
 	{
 		struct ctio7_from_24xx *entry = (struct ctio7_from_24xx *)pkt;
@@ -1120,7 +1128,7 @@ static void qlt_24xx_retry_term_exchange(struct scsi_qla_host *vha,
 	ctio->u.status1.flags =
 	    __constant_cpu_to_le16(CTIO7_FLAGS_STATUS_MODE_1 |
 		CTIO7_FLAGS_TERMINATE);
-	ctio->u.status1.ox_id = entry->fcp_hdr_le.ox_id;
+	ctio->u.status1.ox_id = cpu_to_le16(entry->fcp_hdr_le.ox_id);
 
 	qla2x00_start_iocbs(vha, vha->req);
 
@@ -1254,6 +1262,7 @@ static void qlt_24xx_send_task_mgmt_ctio(struct scsi_qla_host *ha,
 {
 	struct atio_from_isp *atio = &mcmd->orig_iocb.atio;
 	struct ctio7_to_24xx *ctio;
+	uint16_t temp;
 
 	ql_dbg(ql_dbg_tgt, ha, 0xe008,
 	    "Sending task mgmt CTIO7 (ha=%p, atio=%p, resp_code=%x\n",
@@ -1284,7 +1293,8 @@ static void qlt_24xx_send_task_mgmt_ctio(struct scsi_qla_host *ha,
 	ctio->u.status1.flags = (atio->u.isp24.attr << 9) |
 	    __constant_cpu_to_le16(CTIO7_FLAGS_STATUS_MODE_1 |
 		CTIO7_FLAGS_SEND_STATUS);
-	ctio->u.status1.ox_id = swab16(atio->u.isp24.fcp_hdr.ox_id);
+	temp = be16_to_cpu(atio->u.isp24.fcp_hdr.ox_id);
+	ctio->u.status1.ox_id = cpu_to_le16(temp);
 	ctio->u.status1.scsi_status =
 	    __constant_cpu_to_le16(SS_RESPONSE_INFO_LEN_VALID);
 	ctio->u.status1.response_len = __constant_cpu_to_le16(8);
@@ -1350,13 +1360,42 @@ static int qlt_pci_map_calc_cnt(struct qla_tgt_prm *prm)
 
 	prm->cmd->sg_mapped = 1;
 
-	/*
-	 * If greater than four sg entries then we need to allocate
-	 * the continuation entries
-	 */
-	if (prm->seg_cnt > prm->tgt->datasegs_per_cmd)
-		prm->req_cnt += DIV_ROUND_UP(prm->seg_cnt -
-		    prm->tgt->datasegs_per_cmd, prm->tgt->datasegs_per_cont);
+	if (cmd->se_cmd.prot_op == TARGET_PROT_NORMAL) {
+		/*
+		 * If greater than four sg entries then we need to allocate
+		 * the continuation entries
+		 */
+		if (prm->seg_cnt > prm->tgt->datasegs_per_cmd)
+			prm->req_cnt += DIV_ROUND_UP(prm->seg_cnt -
+			prm->tgt->datasegs_per_cmd,
+			prm->tgt->datasegs_per_cont);
+	} else {
+		/* DIF */
+		if ((cmd->se_cmd.prot_op == TARGET_PROT_DIN_INSERT) ||
+		    (cmd->se_cmd.prot_op == TARGET_PROT_DOUT_STRIP)) {
+			prm->seg_cnt = DIV_ROUND_UP(cmd->bufflen, cmd->blk_sz);
+			prm->tot_dsds = prm->seg_cnt;
+		} else
+			prm->tot_dsds = prm->seg_cnt;
+
+		if (cmd->prot_sg_cnt) {
+			prm->prot_sg      = cmd->prot_sg;
+			prm->prot_seg_cnt = pci_map_sg(prm->tgt->ha->pdev,
+				cmd->prot_sg, cmd->prot_sg_cnt,
+				cmd->dma_data_direction);
+			if (unlikely(prm->prot_seg_cnt == 0))
+				goto out_err;
+
+			if ((cmd->se_cmd.prot_op == TARGET_PROT_DIN_INSERT) ||
+			    (cmd->se_cmd.prot_op == TARGET_PROT_DOUT_STRIP)) {
+				/* Dif Bundling not support here */
+				prm->prot_seg_cnt = DIV_ROUND_UP(cmd->bufflen,
+								cmd->blk_sz);
+				prm->tot_dsds += prm->prot_seg_cnt;
+			} else
+				prm->tot_dsds += prm->prot_seg_cnt;
+		}
+	}
 
 	ql_dbg(ql_dbg_tgt, prm->cmd->vha, 0xe009, "seg_cnt=%d, req_cnt=%d\n",
 	    prm->seg_cnt, prm->req_cnt);
@@ -1377,6 +1416,16 @@ static inline void qlt_unmap_sg(struct scsi_qla_host *vha,
 	BUG_ON(!cmd->sg_mapped);
 	pci_unmap_sg(ha->pdev, cmd->sg, cmd->sg_cnt, cmd->dma_data_direction);
 	cmd->sg_mapped = 0;
+
+	if (cmd->prot_sg_cnt)
+		pci_unmap_sg(ha->pdev, cmd->prot_sg, cmd->prot_sg_cnt,
+			cmd->dma_data_direction);
+
+	if (cmd->ctx_dsd_alloced)
+		qla2x00_clean_dsd_pool(ha, NULL, cmd);
+
+	if (cmd->ctx)
+		dma_pool_free(ha->dl_dma_pool, cmd->ctx, cmd->ctx->crc_ctx_dma);
 }
 
 static int qlt_check_reserve_free_req(struct scsi_qla_host *vha,
@@ -1466,6 +1515,7 @@ static int qlt_24xx_build_ctio_pkt(struct qla_tgt_prm *prm,
 	struct ctio7_to_24xx *pkt;
 	struct qla_hw_data *ha = vha->hw;
 	struct atio_from_isp *atio = &prm->cmd->atio;
+	uint16_t temp;
 
 	pkt = (struct ctio7_to_24xx *)vha->req->ring_ptr;
 	prm->pkt = pkt;
@@ -1494,13 +1544,13 @@ static int qlt_24xx_build_ctio_pkt(struct qla_tgt_prm *prm,
 	pkt->initiator_id[2] = atio->u.isp24.fcp_hdr.s_id[0];
 	pkt->exchange_addr = atio->u.isp24.exchange_addr;
 	pkt->u.status0.flags |= (atio->u.isp24.attr << 9);
-	pkt->u.status0.ox_id = swab16(atio->u.isp24.fcp_hdr.ox_id);
+	temp = be16_to_cpu(atio->u.isp24.fcp_hdr.ox_id);
+	pkt->u.status0.ox_id = cpu_to_le16(temp);
 	pkt->u.status0.relative_offset = cpu_to_le32(prm->cmd->offset);
 
 	ql_dbg(ql_dbg_tgt, vha, 0xe00c,
 	    "qla_target(%d): handle(cmd) -> %08x, timeout %d, ox_id %#x\n",
-	    vha->vp_idx, pkt->handle, QLA_TGT_TIMEOUT,
-	    le16_to_cpu(pkt->u.status0.ox_id));
+	    vha->vp_idx, pkt->handle, QLA_TGT_TIMEOUT, temp);
 	return 0;
 }
 
@@ -1665,8 +1715,9 @@ static int qlt_pre_xmit_response(struct qla_tgt_cmd *cmd,
 		return QLA_TGT_PRE_XMIT_RESP_CMD_ABORTED;
 	}
 
-	ql_dbg(ql_dbg_tgt, vha, 0xe011, "qla_target(%d): tag=%u\n",
-	    vha->vp_idx, cmd->tag);
+	ql_dbg(ql_dbg_tgt, vha, 0xe011, "qla_target(%d): tag=%u ox_id %04x\n",
+		vha->vp_idx, cmd->tag,
+		be16_to_cpu(cmd->atio.u.isp24.fcp_hdr.ox_id));
 
 	prm->cmd = cmd;
 	prm->tgt = tgt;
@@ -1902,6 +1953,328 @@ skip_explict_conf:
 	/* Sense with len > 24, is it possible ??? */
 }
 
+
+
+/* diff  */
+static inline int
+qlt_hba_err_chk_enabled(struct se_cmd *se_cmd)
+{
+	/*
+	 * Uncomment when corresponding SCSI changes are done.
+	 *
+	 if (!sp->cmd->prot_chk)
+	 return 0;
+	 *
+	 */
+	switch (se_cmd->prot_op) {
+	case TARGET_PROT_DOUT_INSERT:
+	case TARGET_PROT_DIN_STRIP:
+		if (ql2xenablehba_err_chk >= 1)
+			return 1;
+		break;
+	case TARGET_PROT_DOUT_PASS:
+	case TARGET_PROT_DIN_PASS:
+		if (ql2xenablehba_err_chk >= 2)
+			return 1;
+		break;
+	case TARGET_PROT_DIN_INSERT:
+	case TARGET_PROT_DOUT_STRIP:
+		return 1;
+	default:
+		break;
+	}
+	return 0;
+}
+
+/*
+ * qla24xx_set_t10dif_tags_from_cmd - Extract Ref and App tags from SCSI command
+ *
+ */
+static inline void
+qlt_set_t10dif_tags(struct se_cmd *se_cmd, struct crc_context *ctx)
+{
+	uint32_t lba = 0xffffffff & se_cmd->t_task_lba;
+
+	/* wait til Mode Sense/Select cmd, modepage Ah, subpage 2
+	 * have been immplemented by TCM, before AppTag is avail.
+	 * Look for modesense_handlers[]
+	 */
+	ctx->app_tag = 0;
+	ctx->app_tag_mask[0] = 0x0;
+	ctx->app_tag_mask[1] = 0x0;
+
+	switch (se_cmd->prot_type) {
+	case TARGET_DIF_TYPE0_PROT:
+		/*
+		 * No check for ql2xenablehba_err_chk, as it would be an
+		 * I/O error if hba tag generation is not done.
+		 */
+		ctx->ref_tag = cpu_to_le32(lba);
+
+		if (!qlt_hba_err_chk_enabled(se_cmd))
+			break;
+
+		/* enable ALL bytes of the ref tag */
+		ctx->ref_tag_mask[0] = 0xff;
+		ctx->ref_tag_mask[1] = 0xff;
+		ctx->ref_tag_mask[2] = 0xff;
+		ctx->ref_tag_mask[3] = 0xff;
+		break;
+	/*
+	 * For TYpe 1 protection: 16 bit GUARD tag, 32 bit REF tag, and
+	 * 16 bit app tag.
+	 */
+	case TARGET_DIF_TYPE1_PROT:
+		ctx->ref_tag = cpu_to_le32(lba);
+
+		if (!qlt_hba_err_chk_enabled(se_cmd))
+			break;
+
+		/* enable ALL bytes of the ref tag */
+		ctx->ref_tag_mask[0] = 0xff;
+		ctx->ref_tag_mask[1] = 0xff;
+		ctx->ref_tag_mask[2] = 0xff;
+		ctx->ref_tag_mask[3] = 0xff;
+		break;
+	/*
+	 * For TYPE 2 protection: 16 bit GUARD + 32 bit REF tag has to
+	 * match LBA in CDB + N
+	 */
+	case TARGET_DIF_TYPE2_PROT:
+		ctx->ref_tag = cpu_to_le32(lba);
+
+		if (!qlt_hba_err_chk_enabled(se_cmd))
+			break;
+
+		/* enable ALL bytes of the ref tag */
+		ctx->ref_tag_mask[0] = 0xff;
+		ctx->ref_tag_mask[1] = 0xff;
+		ctx->ref_tag_mask[2] = 0xff;
+		ctx->ref_tag_mask[3] = 0xff;
+		break;
+
+	/* For Type 3 protection: 16 bit GUARD only */
+	case TARGET_DIF_TYPE3_PROT:
+		ctx->ref_tag_mask[0] = ctx->ref_tag_mask[1] =
+			ctx->ref_tag_mask[2] = ctx->ref_tag_mask[3] = 0x00;
+		break;
+	}
+}
+
+
+static inline int
+qlt_build_ctio_crc2_pkt(struct qla_tgt_prm *prm, scsi_qla_host_t *vha)
+{
+	uint32_t		*cur_dsd;
+	int			sgc;
+	uint32_t		transfer_length = 0;
+	uint32_t		data_bytes;
+	uint32_t		dif_bytes;
+	uint8_t			bundling = 1;
+	uint8_t			*clr_ptr;
+	struct crc_context	*crc_ctx_pkt = NULL;
+	struct qla_hw_data	*ha;
+	struct ctio_crc2_to_fw	*pkt;
+	dma_addr_t		crc_ctx_dma;
+	uint16_t		fw_prot_opts = 0;
+	struct qla_tgt_cmd	*cmd = prm->cmd;
+	struct se_cmd		*se_cmd = &cmd->se_cmd;
+	uint32_t h;
+	struct atio_from_isp *atio = &prm->cmd->atio;
+	uint16_t t16;
+
+	sgc = 0;
+	ha = vha->hw;
+
+	pkt = (struct ctio_crc2_to_fw *)vha->req->ring_ptr;
+	prm->pkt = pkt;
+	memset(pkt, 0, sizeof(*pkt));
+
+	ql_dbg(ql_dbg_tgt, vha, 0xe071,
+		"qla_target(%d):%s: se_cmd[%p] CRC2 prot_op[0x%x] cmd prot sg:cnt[%p:%x] lba[%llu]\n",
+		vha->vp_idx, __func__, se_cmd, se_cmd->prot_op,
+		prm->prot_sg, prm->prot_seg_cnt, se_cmd->t_task_lba);
+
+	if ((se_cmd->prot_op == TARGET_PROT_DIN_INSERT) ||
+	    (se_cmd->prot_op == TARGET_PROT_DOUT_STRIP))
+		bundling = 0;
+
+	/* Compute dif len and adjust data len to incude protection */
+	data_bytes = cmd->bufflen;
+	dif_bytes  = (data_bytes / cmd->blk_sz) * 8;
+
+	switch (se_cmd->prot_op) {
+	case TARGET_PROT_DIN_INSERT:
+	case TARGET_PROT_DOUT_STRIP:
+		transfer_length = data_bytes;
+		data_bytes += dif_bytes;
+		break;
+
+	case TARGET_PROT_DIN_STRIP:
+	case TARGET_PROT_DOUT_INSERT:
+	case TARGET_PROT_DIN_PASS:
+	case TARGET_PROT_DOUT_PASS:
+		transfer_length = data_bytes + dif_bytes;
+		break;
+
+	default:
+		BUG();
+		break;
+	}
+
+	if (!qlt_hba_err_chk_enabled(se_cmd))
+		fw_prot_opts |= 0x10; /* Disable Guard tag checking */
+	/* HBA error checking enabled */
+	else if (IS_PI_UNINIT_CAPABLE(ha)) {
+		if ((se_cmd->prot_type == TARGET_DIF_TYPE1_PROT) ||
+		    (se_cmd->prot_type == TARGET_DIF_TYPE2_PROT))
+			fw_prot_opts |= PO_DIS_VALD_APP_ESC;
+		else if (se_cmd->prot_type == TARGET_DIF_TYPE3_PROT)
+			fw_prot_opts |= PO_DIS_VALD_APP_REF_ESC;
+	}
+
+	switch (se_cmd->prot_op) {
+	case TARGET_PROT_DIN_INSERT:
+	case TARGET_PROT_DOUT_INSERT:
+		fw_prot_opts |= PO_MODE_DIF_INSERT;
+		break;
+	case TARGET_PROT_DIN_STRIP:
+	case TARGET_PROT_DOUT_STRIP:
+		fw_prot_opts |= PO_MODE_DIF_REMOVE;
+		break;
+	case TARGET_PROT_DIN_PASS:
+	case TARGET_PROT_DOUT_PASS:
+		fw_prot_opts |= PO_MODE_DIF_PASS;
+		/* FUTURE: does tcm require T10CRC<->IPCKSUM conversion? */
+		break;
+	default:/* Normal Request */
+		fw_prot_opts |= PO_MODE_DIF_PASS;
+		break;
+	}
+
+
+	/* ---- PKT ---- */
+	/* Update entry type to indicate Command Type CRC_2 IOCB */
+	pkt->entry_type  = CTIO_CRC2;
+	pkt->entry_count = 1;
+	pkt->vp_index = vha->vp_idx;
+
+	h = qlt_make_handle(vha);
+	if (unlikely(h == QLA_TGT_NULL_HANDLE)) {
+		/*
+		 * CTIO type 7 from the firmware doesn't provide a way to
+		 * know the initiator's LOOP ID, hence we can't find
+		 * the session and, so, the command.
+		 */
+		return -EAGAIN;
+	} else
+		ha->tgt.cmds[h-1] = prm->cmd;
+
+
+	pkt->handle  = h | CTIO_COMPLETION_HANDLE_MARK;
+	pkt->nport_handle = prm->cmd->loop_id;
+	pkt->timeout = __constant_cpu_to_le16(QLA_TGT_TIMEOUT);
+	pkt->initiator_id[0] = atio->u.isp24.fcp_hdr.s_id[2];
+	pkt->initiator_id[1] = atio->u.isp24.fcp_hdr.s_id[1];
+	pkt->initiator_id[2] = atio->u.isp24.fcp_hdr.s_id[0];
+	pkt->exchange_addr   = atio->u.isp24.exchange_addr;
+
+	/* silence compile warning */
+	t16 = be16_to_cpu(atio->u.isp24.fcp_hdr.ox_id);
+	pkt->ox_id  = cpu_to_le16(t16);
+
+	t16 = (atio->u.isp24.attr << 9);
+	pkt->flags |= cpu_to_le16(t16);
+	pkt->relative_offset = cpu_to_le32(prm->cmd->offset);
+
+	/* Set transfer direction */
+	if (cmd->dma_data_direction == DMA_TO_DEVICE)
+		pkt->flags = __constant_cpu_to_le16(CTIO7_FLAGS_DATA_IN);
+	else if (cmd->dma_data_direction == DMA_FROM_DEVICE)
+		pkt->flags = __constant_cpu_to_le16(CTIO7_FLAGS_DATA_OUT);
+
+
+	pkt->dseg_count = prm->tot_dsds;
+	/* Fibre channel byte count */
+	pkt->transfer_length = cpu_to_le32(transfer_length);
+
+
+	/* ----- CRC context -------- */
+
+	/* Allocate CRC context from global pool */
+	crc_ctx_pkt = cmd->ctx =
+	    dma_pool_alloc(ha->dl_dma_pool, GFP_ATOMIC, &crc_ctx_dma);
+
+	if (!crc_ctx_pkt)
+		goto crc_queuing_error;
+
+	/* Zero out CTX area. */
+	clr_ptr = (uint8_t *)crc_ctx_pkt;
+	memset(clr_ptr, 0, sizeof(*crc_ctx_pkt));
+
+	crc_ctx_pkt->crc_ctx_dma = crc_ctx_dma;
+	INIT_LIST_HEAD(&crc_ctx_pkt->dsd_list);
+
+	/* Set handle */
+	crc_ctx_pkt->handle = pkt->handle;
+
+	qlt_set_t10dif_tags(se_cmd, crc_ctx_pkt);
+
+	pkt->crc_context_address[0] = cpu_to_le32(LSD(crc_ctx_dma));
+	pkt->crc_context_address[1] = cpu_to_le32(MSD(crc_ctx_dma));
+	pkt->crc_context_len = CRC_CONTEXT_LEN_FW;
+
+
+	if (!bundling) {
+		cur_dsd = (uint32_t *) &crc_ctx_pkt->u.nobundling.data_address;
+	} else {
+		/*
+		 * Configure Bundling if we need to fetch interlaving
+		 * protection PCI accesses
+		 */
+		fw_prot_opts |= PO_ENABLE_DIF_BUNDLING;
+		crc_ctx_pkt->u.bundling.dif_byte_count = cpu_to_le32(dif_bytes);
+		crc_ctx_pkt->u.bundling.dseg_count =
+			cpu_to_le16(prm->tot_dsds - prm->prot_seg_cnt);
+		cur_dsd = (uint32_t *) &crc_ctx_pkt->u.bundling.data_address;
+	}
+
+	/* Finish the common fields of CRC pkt */
+	crc_ctx_pkt->blk_size   = cpu_to_le16(cmd->blk_sz);
+	crc_ctx_pkt->prot_opts  = cpu_to_le16(fw_prot_opts);
+	crc_ctx_pkt->byte_count = cpu_to_le32(data_bytes);
+	crc_ctx_pkt->guard_seed = __constant_cpu_to_le16(0);
+
+
+	/* Walks data segments */
+	pkt->flags |= __constant_cpu_to_le16(CTIO7_FLAGS_DSD_PTR);
+
+	if (!bundling && prm->prot_seg_cnt) {
+		if (qla24xx_walk_and_build_sglist_no_difb(ha, NULL, cur_dsd,
+			prm->tot_dsds, cmd))
+			goto crc_queuing_error;
+	} else if (qla24xx_walk_and_build_sglist(ha, NULL, cur_dsd,
+		(prm->tot_dsds - prm->prot_seg_cnt), cmd))
+		goto crc_queuing_error;
+
+	if (bundling && prm->prot_seg_cnt) {
+		/* Walks dif segments */
+		pkt->add_flags |= CTIO_CRC2_AF_DIF_DSD_ENA;
+
+		cur_dsd = (uint32_t *) &crc_ctx_pkt->u.bundling.dif_address;
+		if (qla24xx_walk_and_build_prot_sglist(ha, NULL, cur_dsd,
+			prm->prot_seg_cnt, cmd))
+			goto crc_queuing_error;
+	}
+	return QLA_SUCCESS;
+
+crc_queuing_error:
+	/* Cleanup will be performed by the caller */
+
+	return QLA_FUNCTION_FAILED;
+}
+
+
 /*
  * Callback to setup response of xmit_type of QLA_TGT_XMIT_DATA and *
  * QLA_TGT_XMIT_STATUS for >= 24xx silicon
@@ -1921,9 +2294,10 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
 	qlt_check_srr_debug(cmd, &xmit_type);
 
 	ql_dbg(ql_dbg_tgt, cmd->vha, 0xe018,
-	    "is_send_status=%d, cmd->bufflen=%d, cmd->sg_cnt=%d, "
-	    "cmd->dma_data_direction=%d\n", (xmit_type & QLA_TGT_XMIT_STATUS) ?
-	    1 : 0, cmd->bufflen, cmd->sg_cnt, cmd->dma_data_direction);
+	    "is_send_status=%d, cmd->bufflen=%d, cmd->sg_cnt=%d, cmd->dma_data_direction=%d se_cmd[%p]\n",
+	    (xmit_type & QLA_TGT_XMIT_STATUS) ?
+	    1 : 0, cmd->bufflen, cmd->sg_cnt, cmd->dma_data_direction,
+	    &cmd->se_cmd);
 
 	res = qlt_pre_xmit_response(cmd, &prm, xmit_type, scsi_status,
 	    &full_req_cnt);
@@ -1941,7 +2315,10 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
 	if (unlikely(res))
 		goto out_unmap_unlock;
 
-	res = qlt_24xx_build_ctio_pkt(&prm, vha);
+	if (cmd->se_cmd.prot_op && (xmit_type & QLA_TGT_XMIT_DATA))
+		res = qlt_build_ctio_crc2_pkt(&prm, vha);
+	else
+		res = qlt_24xx_build_ctio_pkt(&prm, vha);
 	if (unlikely(res != 0))
 		goto out_unmap_unlock;
 
@@ -1953,7 +2330,8 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
 		    __constant_cpu_to_le16(CTIO7_FLAGS_DATA_IN |
 			CTIO7_FLAGS_STATUS_MODE_0);
 
-		qlt_load_data_segments(&prm, vha);
+		if (cmd->se_cmd.prot_op == TARGET_PROT_NORMAL)
+			qlt_load_data_segments(&prm, vha);
 
 		if (prm.add_status_pkt == 0) {
 			if (xmit_type & QLA_TGT_XMIT_STATUS) {
@@ -1983,8 +2361,14 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
 			ql_dbg(ql_dbg_tgt, vha, 0xe019,
 			    "Building additional status packet\n");
 
+			/*
+			 * T10Dif: ctio_crc2_to_fw overlay ontop of
+			 * ctio7_to_24xx
+			 */
 			memcpy(ctio, pkt, sizeof(*ctio));
+			/* reset back to CTIO7 */
 			ctio->entry_count = 1;
+			ctio->entry_type = CTIO_TYPE7;
 			ctio->dseg_count = 0;
 			ctio->u.status1.flags &= ~__constant_cpu_to_le16(
 			    CTIO7_FLAGS_DATA_IN);
@@ -1993,6 +2377,11 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
 			pkt->handle |= CTIO_INTERMEDIATE_HANDLE_MARK;
 			pkt->u.status0.flags |= __constant_cpu_to_le16(
 			    CTIO7_FLAGS_DONT_RET_CTIO);
+
+			/* qlt_24xx_init_ctio_to_isp will correct
+			 * all neccessary fields that's part of CTIO7.
+			 * There should be no residual of CTIO-CRC2 data.
+			 */
 			qlt_24xx_init_ctio_to_isp((struct ctio7_to_24xx *)ctio,
 			    &prm);
 			pr_debug("Status CTIO7: %p\n", ctio);
@@ -2041,8 +2430,10 @@ int qlt_rdy_to_xfer(struct qla_tgt_cmd *cmd)
 	if (qlt_issue_marker(vha, 0) != QLA_SUCCESS)
 		return -EIO;
 
-	ql_dbg(ql_dbg_tgt, vha, 0xe01b, "CTIO_start: vha(%d)",
-	    (int)vha->vp_idx);
+	ql_dbg(ql_dbg_tgt, vha, 0xe01b,
+		"%s: CTIO_start: vha(%d) se_cmd %p ox_id %04x\n",
+		__func__, (int)vha->vp_idx, &cmd->se_cmd,
+		be16_to_cpu(cmd->atio.u.isp24.fcp_hdr.ox_id));
 
 	/* Calculate number of entries and segments required */
 	if (qlt_pci_map_calc_cnt(&prm) != 0)
@@ -2054,14 +2445,19 @@ int qlt_rdy_to_xfer(struct qla_tgt_cmd *cmd)
 	res = qlt_check_reserve_free_req(vha, prm.req_cnt);
 	if (res != 0)
 		goto out_unlock_free_unmap;
+	if (cmd->se_cmd.prot_op)
+		res = qlt_build_ctio_crc2_pkt(&prm, vha);
+	else
+		res = qlt_24xx_build_ctio_pkt(&prm, vha);
 
-	res = qlt_24xx_build_ctio_pkt(&prm, vha);
 	if (unlikely(res != 0))
 		goto out_unlock_free_unmap;
 	pkt = (struct ctio7_to_24xx *)prm.pkt;
 	pkt->u.status0.flags |= __constant_cpu_to_le16(CTIO7_FLAGS_DATA_OUT |
 	    CTIO7_FLAGS_STATUS_MODE_0);
-	qlt_load_data_segments(&prm, vha);
+
+	if (cmd->se_cmd.prot_op == TARGET_PROT_NORMAL)
+		qlt_load_data_segments(&prm, vha);
 
 	cmd->state = QLA_TGT_STATE_NEED_DATA;
 
@@ -2079,6 +2475,143 @@ out_unlock_free_unmap:
 }
 EXPORT_SYMBOL(qlt_rdy_to_xfer);
 
+
+/*
+ * Checks the guard or meta-data for the type of error
+ * detected by the HBA.
+ */
+static inline int
+qlt_handle_dif_error(struct scsi_qla_host *vha, struct qla_tgt_cmd *cmd,
+		struct ctio_crc_from_fw *sts)
+{
+	uint8_t		*ap = &sts->actual_dif[0];
+	uint8_t		*ep = &sts->expected_dif[0];
+	uint32_t	e_ref_tag, a_ref_tag;
+	uint16_t	e_app_tag, a_app_tag;
+	uint16_t	e_guard, a_guard;
+	uint64_t	lba = cmd->se_cmd.t_task_lba;
+
+	a_guard   = be16_to_cpu(*(uint16_t *)(ap + 0));
+	a_app_tag = be16_to_cpu(*(uint16_t *)(ap + 2));
+	a_ref_tag = be32_to_cpu(*(uint32_t *)(ap + 4));
+
+	e_guard   = be16_to_cpu(*(uint16_t *)(ep + 0));
+	e_app_tag = be16_to_cpu(*(uint16_t *)(ep + 2));
+	e_ref_tag = be32_to_cpu(*(uint32_t *)(ep + 4));
+
+	ql_dbg(ql_dbg_tgt, vha, 0xe075,
+	    "iocb(s) %p Returned STATUS.\n", sts);
+
+	ql_dbg(ql_dbg_tgt, vha, 0xf075,
+	    "dif check TGT cdb 0x%x lba 0x%llu: [Actual|Expected] Ref Tag[0x%x|0x%x], App Tag [0x%x|0x%x], Guard [0x%x|0x%x]\n",
+	    cmd->atio.u.isp24.fcp_cmnd.cdb[0], lba,
+	    a_ref_tag, e_ref_tag, a_app_tag, e_app_tag, a_guard, e_guard);
+
+	/*
+	 * Ignore sector if:
+	 * For type     3: ref & app tag is all 'f's
+	 * For type 0,1,2: app tag is all 'f's
+	 */
+	if ((a_app_tag == 0xffff) &&
+	    ((cmd->se_cmd.prot_type != TARGET_DIF_TYPE3_PROT) ||
+	     (a_ref_tag == 0xffffffff))) {
+		uint32_t blocks_done;
+
+		/* 2TB boundary case covered automatically with this */
+		blocks_done = e_ref_tag - (uint32_t)lba + 1;
+		cmd->se_cmd.bad_sector = e_ref_tag;
+		cmd->se_cmd.pi_err = 0;
+		ql_dbg(ql_dbg_tgt, vha, 0xf074,
+			"need to return scsi good\n");
+
+		/* Update protection tag */
+		if (cmd->prot_sg_cnt) {
+			uint32_t i, j = 0, k = 0, num_ent;
+			struct scatterlist *sg, *sgl;
+
+
+			sgl = cmd->prot_sg;
+
+			/* Patch the corresponding protection tags */
+			for_each_sg(sgl, sg, cmd->prot_sg_cnt, i) {
+				num_ent = sg_dma_len(sg) / 8;
+				if (k + num_ent < blocks_done) {
+					k += num_ent;
+					continue;
+				}
+				j = blocks_done - k - 1;
+				k = blocks_done;
+				break;
+			}
+
+			if (k != blocks_done) {
+				ql_log(ql_log_warn, vha, 0xf076,
+				    "unexpected tag values tag:lba=%u:%llu)\n",
+				    e_ref_tag, (unsigned long long)lba);
+				goto out;
+			}
+
+#if 0
+			struct sd_dif_tuple *spt;
+			/* TODO:
+			 * This section came from initiator. Is it valid here?
+			 * should ulp be override with actual val???
+			 */
+			spt = page_address(sg_page(sg)) + sg->offset;
+			spt += j;
+
+			spt->app_tag = 0xffff;
+			if (cmd->se_cmd.prot_type == SCSI_PROT_DIF_TYPE3)
+				spt->ref_tag = 0xffffffff;
+#endif
+		}
+
+		return 0;
+	}
+
+	/* check guard */
+	if (e_guard != a_guard) {
+		cmd->se_cmd.pi_err = TCM_LOGICAL_BLOCK_GUARD_CHECK_FAILED;
+		cmd->se_cmd.bad_sector = cmd->se_cmd.t_task_lba;
+
+		ql_log(ql_log_warn, vha, 0xe076,
+		    "Guard ERR: cdb 0x%x lba 0x%llx: [Actual|Expected] Ref Tag[0x%x|0x%x], App Tag [0x%x|0x%x], Guard [0x%x|0x%x] cmd=%p\n",
+		    cmd->atio.u.isp24.fcp_cmnd.cdb[0], lba,
+		    a_ref_tag, e_ref_tag, a_app_tag, e_app_tag,
+		    a_guard, e_guard, cmd);
+		goto out;
+	}
+
+	/* check ref tag */
+	if (e_ref_tag != a_ref_tag) {
+		cmd->se_cmd.pi_err = TCM_LOGICAL_BLOCK_REF_TAG_CHECK_FAILED;
+		cmd->se_cmd.bad_sector = e_ref_tag;
+
+		ql_log(ql_log_warn, vha, 0xe077,
+			"Ref Tag ERR: cdb 0x%x lba 0x%llx: [Actual|Expected] Ref Tag[0x%x|0x%x], App Tag [0x%x|0x%x], Guard [0x%x|0x%x] cmd=%p\n",
+			cmd->atio.u.isp24.fcp_cmnd.cdb[0], lba,
+			a_ref_tag, e_ref_tag, a_app_tag, e_app_tag,
+			a_guard, e_guard, cmd);
+		goto out;
+	}
+
+	/* check appl tag */
+	if (e_app_tag != a_app_tag) {
+		cmd->se_cmd.pi_err = TCM_LOGICAL_BLOCK_APP_TAG_CHECK_FAILED;
+		cmd->se_cmd.bad_sector = cmd->se_cmd.t_task_lba;
+
+		ql_log(ql_log_warn, vha, 0xe078,
+			"App Tag ERR: cdb 0x%x lba 0x%llx: [Actual|Expected] Ref Tag[0x%x|0x%x], App Tag [0x%x|0x%x], Guard [0x%x|0x%x] cmd=%p\n",
+			cmd->atio.u.isp24.fcp_cmnd.cdb[0], lba,
+			a_ref_tag, e_ref_tag, a_app_tag, e_app_tag,
+			a_guard, e_guard, cmd);
+		goto out;
+	}
+out:
+	return 1;
+}
+
+
 /* If hardware_lock held on entry, might drop it, then reaquire */
 /* This function sends the appropriate CTIO to ISP 2xxx or 24xx */
 static int __qlt_send_term_exchange(struct scsi_qla_host *vha,
@@ -2089,6 +2622,7 @@ static int __qlt_send_term_exchange(struct scsi_qla_host *vha,
 	struct qla_hw_data *ha = vha->hw;
 	request_t *pkt;
 	int ret = 0;
+	uint16_t temp;
 
 	ql_dbg(ql_dbg_tgt, vha, 0xe01c, "Sending TERM EXCH CTIO (ha=%p)\n", ha);
 
@@ -2125,7 +2659,8 @@ static int __qlt_send_term_exchange(struct scsi_qla_host *vha,
 	ctio24->u.status1.flags = (atio->u.isp24.attr << 9) |
 	    __constant_cpu_to_le16(CTIO7_FLAGS_STATUS_MODE_1 |
 		CTIO7_FLAGS_TERMINATE);
-	ctio24->u.status1.ox_id = swab16(atio->u.isp24.fcp_hdr.ox_id);
+	temp = be16_to_cpu(atio->u.isp24.fcp_hdr.ox_id);
+	ctio24->u.status1.ox_id = cpu_to_le16(temp);
 
 	/* Most likely, it isn't needed */
 	ctio24->u.status1.residual = get_unaligned((uint32_t *)
@@ -2155,21 +2690,46 @@ static void qlt_send_term_exchange(struct scsi_qla_host *vha,
 	rc = __qlt_send_term_exchange(vha, cmd, atio);
 	spin_unlock_irqrestore(&vha->hw->hardware_lock, flags);
 done:
-	if (rc == 1) {
+	/*
+	 * Terminate exchange will tell fw to release any active CTIO
+	 * that's in FW posession and cleanup the exchange.
+	 *
+	 * "cmd->state == QLA_TGT_STATE_ABORTED" means CTIO is still
+	 * down at FW.  Free the cmd later when CTIO comes back later
+	 * w/aborted(0x2) status.
+	 *
+	 * "cmd->state != QLA_TGT_STATE_ABORTED" means CTIO is already
+	 * back w/some err.  Free the cmd now.
+	 */
+	if ((rc == 1) && (cmd->state != QLA_TGT_STATE_ABORTED)) {
 		if (!ha_locked && !in_interrupt())
 			msleep(250); /* just in case */
 
+		if (cmd->sg_mapped)
+			qlt_unmap_sg(vha, cmd);
 		vha->hw->tgt.tgt_ops->free_cmd(cmd);
 	}
+	return;
 }
 
 void qlt_free_cmd(struct qla_tgt_cmd *cmd)
 {
-	BUG_ON(cmd->sg_mapped);
+	struct qla_tgt_sess *sess = cmd->sess;
 
+	ql_dbg(ql_dbg_tgt, cmd->vha, 0xe074,
+	    "%s: se_cmd[%p] ox_id %04x\n",
+	    __func__, &cmd->se_cmd,
+	    be16_to_cpu(cmd->atio.u.isp24.fcp_hdr.ox_id));
+
+	BUG_ON(cmd->sg_mapped);
 	if (unlikely(cmd->free_sg))
 		kfree(cmd->sg);
-	kmem_cache_free(qla_tgt_cmd_cachep, cmd);
+
+	if (!sess || !sess->se_sess) {
+		WARN_ON(1);
+		return;
+	}
+	percpu_ida_free(&sess->se_sess->sess_tag_pool, cmd->se_cmd.map_tag);
 }
 EXPORT_SYMBOL(qlt_free_cmd);
 
@@ -2374,6 +2934,7 @@ static void qlt_do_ctio_completion(struct scsi_qla_host *vha, uint32_t handle,
 		case CTIO_LIP_RESET:
 		case CTIO_TARGET_RESET:
 		case CTIO_ABORTED:
+			/* driver request abort via Terminate exchange */
 		case CTIO_TIMEOUT:
 		case CTIO_INVALID_RX_ID:
 			/* They are OK */
@@ -2404,18 +2965,58 @@ static void qlt_do_ctio_completion(struct scsi_qla_host *vha, uint32_t handle,
 			else
 				return;
 
+		case CTIO_DIF_ERROR: {
+			struct ctio_crc_from_fw *crc =
+				(struct ctio_crc_from_fw *)ctio;
+			ql_dbg(ql_dbg_tgt_mgt, vha, 0xf073,
+			    "qla_target(%d): CTIO with DIF_ERROR status %x received (state %x, se_cmd %p) actual_dif[0x%llx] expect_dif[0x%llx]\n",
+			    vha->vp_idx, status, cmd->state, se_cmd,
+			    *((u64 *)&crc->actual_dif[0]),
+			    *((u64 *)&crc->expected_dif[0]));
+
+			if (qlt_handle_dif_error(vha, cmd, ctio)) {
+				if (cmd->state == QLA_TGT_STATE_NEED_DATA) {
+					/* scsi Write/xfer rdy complete */
+					goto skip_term;
+				} else {
+					/* scsi read/xmit respond complete
+					 * call handle dif to send scsi status
+					 * rather than terminate exchange.
+					 */
+					cmd->state = QLA_TGT_STATE_PROCESSED;
+					ha->tgt.tgt_ops->handle_dif_err(cmd);
+					return;
+				}
+			} else {
+				/* Need to generate a SCSI good completion.
+				 * because FW did not send scsi status.
+				 */
+				status = 0;
+				goto skip_term;
+			}
+			break;
+		}
 		default:
 			ql_dbg(ql_dbg_tgt_mgt, vha, 0xf05b,
-			    "qla_target(%d): CTIO with error status "
-			    "0x%x received (state %x, se_cmd %p\n",
+			    "qla_target(%d): CTIO with error status 0x%x received (state %x, se_cmd %p\n",
 			    vha->vp_idx, status, cmd->state, se_cmd);
 			break;
 		}
 
-		if (cmd->state != QLA_TGT_STATE_NEED_DATA)
+
+		/* "cmd->state == QLA_TGT_STATE_ABORTED" means
+		 * cmd is already aborted/terminated, we don't
+		 * need to terminate again.  The exchange is already
+		 * cleaned up/freed at FW level.  Just cleanup at driver
+		 * level.
+		 */
+		if ((cmd->state != QLA_TGT_STATE_NEED_DATA) &&
+			(cmd->state != QLA_TGT_STATE_ABORTED)) {
 			if (qlt_term_ctio_exchange(vha, ctio, cmd, status))
 				return;
+		}
 	}
+skip_term:
 
 	if (cmd->state == QLA_TGT_STATE_PROCESSED) {
 		ql_dbg(ql_dbg_tgt, vha, 0xe01f, "Command %p finished\n", cmd);
@@ -2444,7 +3045,8 @@ static void qlt_do_ctio_completion(struct scsi_qla_host *vha, uint32_t handle,
 		    "not return a CTIO complete\n", vha->vp_idx, cmd->state);
 	}
 
-	if (unlikely(status != CTIO_SUCCESS)) {
+	if (unlikely(status != CTIO_SUCCESS) &&
+		(cmd->state != QLA_TGT_STATE_ABORTED)) {
 		ql_dbg(ql_dbg_tgt_mgt, vha, 0xf01f, "Finishing failed CTIO\n");
 		dump_stack();
 	}
@@ -2489,13 +3091,12 @@ static struct qla_tgt_sess *qlt_make_local_sess(struct scsi_qla_host *,
 /*
  * Process context for I/O path into tcm_qla2xxx code
  */
-static void qlt_do_work(struct work_struct *work)
+static void __qlt_do_work(struct qla_tgt_cmd *cmd)
 {
-	struct qla_tgt_cmd *cmd = container_of(work, struct qla_tgt_cmd, work);
 	scsi_qla_host_t *vha = cmd->vha;
 	struct qla_hw_data *ha = vha->hw;
 	struct qla_tgt *tgt = vha->vha_tgt.qla_tgt;
-	struct qla_tgt_sess *sess = NULL;
+	struct qla_tgt_sess *sess = cmd->sess;
 	struct atio_from_isp *atio = &cmd->atio;
 	unsigned char *cdb;
 	unsigned long flags;
@@ -2505,41 +3106,6 @@ static void qlt_do_work(struct work_struct *work)
 	if (tgt->tgt_stop)
 		goto out_term;
 
-	spin_lock_irqsave(&ha->hardware_lock, flags);
-	sess = ha->tgt.tgt_ops->find_sess_by_s_id(vha,
-	    atio->u.isp24.fcp_hdr.s_id);
-	/* Do kref_get() before dropping qla_hw_data->hardware_lock. */
-	if (sess)
-		kref_get(&sess->se_sess->sess_kref);
-	spin_unlock_irqrestore(&ha->hardware_lock, flags);
-
-	if (unlikely(!sess)) {
-		uint8_t *s_id =	atio->u.isp24.fcp_hdr.s_id;
-
-		ql_dbg(ql_dbg_tgt_mgt, vha, 0xf022,
-			"qla_target(%d): Unable to find wwn login"
-			" (s_id %x:%x:%x), trying to create it manually\n",
-			vha->vp_idx, s_id[0], s_id[1], s_id[2]);
-
-		if (atio->u.raw.entry_count > 1) {
-			ql_dbg(ql_dbg_tgt_mgt, vha, 0xf023,
-				"Dropping multy entry cmd %p\n", cmd);
-			goto out_term;
-		}
-
-		mutex_lock(&vha->vha_tgt.tgt_mutex);
-		sess = qlt_make_local_sess(vha, s_id);
-		/* sess has an extra creation ref. */
-		mutex_unlock(&vha->vha_tgt.tgt_mutex);
-
-		if (!sess)
-			goto out_term;
-	}
-
-	cmd->sess = sess;
-	cmd->loop_id = sess->loop_id;
-	cmd->conf_compl_supported = sess->conf_compl_supported;
-
 	cdb = &atio->u.isp24.fcp_cmnd.cdb[0];
 	cmd->tag = atio->u.isp24.exchange_addr;
 	cmd->unpacked_lun = scsilun_to_int(
@@ -2563,11 +3129,12 @@ static void qlt_do_work(struct work_struct *work)
 	    atio->u.isp24.fcp_cmnd.add_cdb_len]));
 
 	ql_dbg(ql_dbg_tgt, vha, 0xe022,
-	    "qla_target: START qla command: %p lun: 0x%04x (tag %d)\n",
-	    cmd, cmd->unpacked_lun, cmd->tag);
+		"qla_target: START qla cmd: %p se_cmd %p lun: 0x%04x (tag %d) len(%d) ox_id %x\n",
+		cmd, &cmd->se_cmd, cmd->unpacked_lun, cmd->tag, data_length,
+		cmd->atio.u.isp24.fcp_hdr.ox_id);
 
-	ret = vha->hw->tgt.tgt_ops->handle_cmd(vha, cmd, cdb, data_length,
-	    fcp_task_attr, data_dir, bidi);
+	ret = ha->tgt.tgt_ops->handle_cmd(vha, cmd, cdb, data_length,
+				          fcp_task_attr, data_dir, bidi);
 	if (ret != 0)
 		goto out_term;
 	/*
@@ -2586,17 +3153,114 @@ out_term:
 	 */
 	spin_lock_irqsave(&ha->hardware_lock, flags);
 	qlt_send_term_exchange(vha, NULL, &cmd->atio, 1);
-	kmem_cache_free(qla_tgt_cmd_cachep, cmd);
-	if (sess)
+	percpu_ida_free(&sess->se_sess->sess_tag_pool, cmd->se_cmd.map_tag);
+	ha->tgt.tgt_ops->put_sess(sess);
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+}
+
+static void qlt_do_work(struct work_struct *work)
+{
+	struct qla_tgt_cmd *cmd = container_of(work, struct qla_tgt_cmd, work);
+
+	__qlt_do_work(cmd);
+}
+
+static struct qla_tgt_cmd *qlt_get_tag(scsi_qla_host_t *vha,
+				       struct qla_tgt_sess *sess,
+				       struct atio_from_isp *atio)
+{
+	struct se_session *se_sess = sess->se_sess;
+	struct qla_tgt_cmd *cmd;
+	int tag;
+
+	tag = percpu_ida_alloc(&se_sess->sess_tag_pool, TASK_RUNNING);
+	if (tag < 0)
+		return NULL;
+
+	cmd = &((struct qla_tgt_cmd *)se_sess->sess_cmd_map)[tag];
+	memset(cmd, 0, sizeof(struct qla_tgt_cmd));
+
+	memcpy(&cmd->atio, atio, sizeof(*atio));
+	cmd->state = QLA_TGT_STATE_NEW;
+	cmd->tgt = vha->vha_tgt.qla_tgt;
+	cmd->vha = vha;
+	cmd->se_cmd.map_tag = tag;
+	cmd->sess = sess;
+	cmd->loop_id = sess->loop_id;
+	cmd->conf_compl_supported = sess->conf_compl_supported;
+
+	return cmd;
+}
+
+static void qlt_send_busy(struct scsi_qla_host *, struct atio_from_isp *,
+			  uint16_t);
+
+static void qlt_create_sess_from_atio(struct work_struct *work)
+{
+	struct qla_tgt_sess_op *op = container_of(work,
+					struct qla_tgt_sess_op, work);
+	scsi_qla_host_t *vha = op->vha;
+	struct qla_hw_data *ha = vha->hw;
+	struct qla_tgt_sess *sess;
+	struct qla_tgt_cmd *cmd;
+	unsigned long flags;
+	uint8_t *s_id = op->atio.u.isp24.fcp_hdr.s_id;
+
+	ql_dbg(ql_dbg_tgt_mgt, vha, 0xf022,
+		"qla_target(%d): Unable to find wwn login"
+		" (s_id %x:%x:%x), trying to create it manually\n",
+		vha->vp_idx, s_id[0], s_id[1], s_id[2]);
+
+	if (op->atio.u.raw.entry_count > 1) {
+		ql_dbg(ql_dbg_tgt_mgt, vha, 0xf023,
+		        "Dropping multy entry atio %p\n", &op->atio);
+		goto out_term;
+	}
+
+	mutex_lock(&vha->vha_tgt.tgt_mutex);
+	sess = qlt_make_local_sess(vha, s_id);
+	/* sess has an extra creation ref. */
+	mutex_unlock(&vha->vha_tgt.tgt_mutex);
+
+	if (!sess)
+		goto out_term;
+	/*
+	 * Now obtain a pre-allocated session tag using the original op->atio
+	 * packet header, and dispatch into __qlt_do_work() using the existing
+	 * process context.
+	 */
+	cmd = qlt_get_tag(vha, sess, &op->atio);
+	if (!cmd) {
+		spin_lock_irqsave(&ha->hardware_lock, flags);
+		qlt_send_busy(vha, &op->atio, SAM_STAT_BUSY);
 		ha->tgt.tgt_ops->put_sess(sess);
+		spin_unlock_irqrestore(&ha->hardware_lock, flags);
+		kfree(op);
+		return;
+	}
+	/*
+	 * __qlt_do_work() will call ha->tgt.tgt_ops->put_sess() to release
+	 * the extra reference taken above by qlt_make_local_sess()
+	 */
+	__qlt_do_work(cmd);
+	kfree(op);
+	return;
+
+out_term:
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	qlt_send_term_exchange(vha, NULL, &op->atio, 1);
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+	kfree(op);
+
 }
 
 /* ha->hardware_lock supposed to be held on entry */
 static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha,
 	struct atio_from_isp *atio)
 {
+	struct qla_hw_data *ha = vha->hw;
 	struct qla_tgt *tgt = vha->vha_tgt.qla_tgt;
+	struct qla_tgt_sess *sess;
 	struct qla_tgt_cmd *cmd;
 
 	if (unlikely(tgt->tgt_stop)) {
@@ -2605,18 +3269,31 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha,
 		return -EFAULT;
 	}
 
-	cmd = kmem_cache_zalloc(qla_tgt_cmd_cachep, GFP_ATOMIC);
+	sess = ha->tgt.tgt_ops->find_sess_by_s_id(vha, atio->u.isp24.fcp_hdr.s_id);
+	if (unlikely(!sess)) {
+		struct qla_tgt_sess_op *op = kzalloc(sizeof(struct qla_tgt_sess_op),
+						     GFP_ATOMIC);
+		if (!op)
+			return -ENOMEM;
+
+		memcpy(&op->atio, atio, sizeof(*atio));
+		INIT_WORK(&op->work, qlt_create_sess_from_atio);
+		queue_work(qla_tgt_wq, &op->work);
+		return 0;
+	}
+	/*
+	 * Do kref_get() before returning + dropping qla_hw_data->hardware_lock.
+	 */
+	kref_get(&sess->se_sess->sess_kref);
+
+	cmd = qlt_get_tag(vha, sess, atio);
 	if (!cmd) {
 		ql_dbg(ql_dbg_tgt_mgt, vha, 0xf05e,
 		    "qla_target(%d): Allocation of cmd failed\n", vha->vp_idx);
+		ha->tgt.tgt_ops->put_sess(sess);
 		return -ENOMEM;
 	}
 
-	memcpy(&cmd->atio, atio, sizeof(*atio));
-	cmd->state = QLA_TGT_STATE_NEW;
-	cmd->tgt = vha->vha_tgt.qla_tgt;
-	cmd->vha = vha;
-
 	INIT_WORK(&cmd->work, qlt_do_work);
 	queue_work(qla_tgt_wq, &cmd->work);
 	return 0;
@@ -3527,11 +4204,11 @@ static void qlt_24xx_atio_pkt(struct scsi_qla_host *vha,
 	switch (atio->u.raw.entry_type) {
 	case ATIO_TYPE7:
 		ql_dbg(ql_dbg_tgt, vha, 0xe02d,
-		    "ATIO_TYPE7 instance %d, lun %Lx, read/write %d/%d, "
-		    "add_cdb_len %d, data_length %04x, s_id %x:%x:%x\n",
+		    "ATIO_TYPE7 instance %d, lun %Lx, read/write %d/%d, cdb %x, add_cdb_len %x, data_length %04x, s_id %02x%02x%02x\n",
 		    vha->vp_idx, atio->u.isp24.fcp_cmnd.lun,
 		    atio->u.isp24.fcp_cmnd.rddata,
 		    atio->u.isp24.fcp_cmnd.wrdata,
+		    atio->u.isp24.fcp_cmnd.cdb[0],
 		    atio->u.isp24.fcp_cmnd.add_cdb_len,
 		    be32_to_cpu(get_unaligned((uint32_t *)
 			&atio->u.isp24.fcp_cmnd.add_cdb[
@@ -3629,11 +4306,13 @@ static void qlt_response_pkt(struct scsi_qla_host *vha, response_t *pkt)
 	tgt->irq_cmd_count++;
 
 	switch (pkt->entry_type) {
+	case CTIO_CRC2:
 	case CTIO_TYPE7:
 	{
 		struct ctio7_from_24xx *entry = (struct ctio7_from_24xx *)pkt;
-		ql_dbg(ql_dbg_tgt, vha, 0xe030, "CTIO_TYPE7: instance %d\n",
-		    vha->vp_idx);
+		ql_dbg(ql_dbg_tgt, vha, 0xe030,
+			"CTIO[0x%x] 12/CTIO7 7A/CRC2: instance %d\n",
+			entry->entry_type, vha->vp_idx);
 		qlt_do_ctio_completion(vha, entry->handle,
 		    le16_to_cpu(entry->status)|(pkt->entry_status << 16),
 		    entry);
@@ -4768,6 +5447,7 @@ qlt_24xx_process_response_error(struct scsi_qla_host *vha,
 	case ABTS_RESP_24XX:
 	case CTIO_TYPE7:
 	case NOTIFY_ACK_TYPE:
+	case CTIO_CRC2:
 		return 1;
 	default:
 		return 0;
@@ -4911,23 +5591,13 @@ int __init qlt_init(void)
 	if (!QLA_TGT_MODE_ENABLED())
 		return 0;
 
-	qla_tgt_cmd_cachep = kmem_cache_create("qla_tgt_cmd_cachep",
-	    sizeof(struct qla_tgt_cmd), __alignof__(struct qla_tgt_cmd), 0,
-	    NULL);
-	if (!qla_tgt_cmd_cachep) {
-		ql_log(ql_log_fatal, NULL, 0xe06c,
-		    "kmem_cache_create for qla_tgt_cmd_cachep failed\n");
-		return -ENOMEM;
-	}
-
 	qla_tgt_mgmt_cmd_cachep = kmem_cache_create("qla_tgt_mgmt_cmd_cachep",
 	    sizeof(struct qla_tgt_mgmt_cmd), __alignof__(struct
 	    qla_tgt_mgmt_cmd), 0, NULL);
 	if (!qla_tgt_mgmt_cmd_cachep) {
 		ql_log(ql_log_fatal, NULL, 0xe06d,
 		    "kmem_cache_create for qla_tgt_mgmt_cmd_cachep failed\n");
-		ret = -ENOMEM;
-		goto out;
+		return -ENOMEM;
 	}
 
 	qla_tgt_mgmt_cmd_mempool = mempool_create(25, mempool_alloc_slab,
@@ -4955,8 +5625,6 @@ out_cmd_mempool:
 	mempool_destroy(qla_tgt_mgmt_cmd_mempool);
 out_mgmt_cmd_cachep:
 	kmem_cache_destroy(qla_tgt_mgmt_cmd_cachep);
-out:
-	kmem_cache_destroy(qla_tgt_cmd_cachep);
 	return ret;
 }
 
@@ -4968,5 +5636,4 @@ void qlt_exit(void)
 	destroy_workqueue(qla_tgt_wq);
 	mempool_destroy(qla_tgt_mgmt_cmd_mempool);
 	kmem_cache_destroy(qla_tgt_mgmt_cmd_cachep);
-	kmem_cache_destroy(qla_tgt_cmd_cachep);
 }
diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h
index ce33d8c26406..d1d24fb0160a 100644
--- a/drivers/scsi/qla2xxx/qla_target.h
+++ b/drivers/scsi/qla2xxx/qla_target.h
@@ -293,6 +293,7 @@ struct ctio_to_2xxx {
 #define CTIO_ABORTED			0x02
 #define CTIO_INVALID_RX_ID		0x08
 #define CTIO_TIMEOUT			0x0B
+#define CTIO_DIF_ERROR			0x0C     /* DIF error detected  */
 #define CTIO_LIP_RESET			0x0E
 #define CTIO_TARGET_RESET		0x17
 #define CTIO_PORT_UNAVAILABLE		0x28
@@ -315,7 +316,7 @@ struct fcp_hdr {
 	uint8_t  seq_id;
 	uint8_t  df_ctl;
 	uint16_t seq_cnt;
-	uint16_t ox_id;
+	__be16   ox_id;
 	uint16_t rx_id;
 	uint32_t parameter;
 } __packed;
@@ -440,9 +441,9 @@ struct ctio7_to_24xx {
 	union {
 		struct {
 			uint16_t reserved1;
-			uint16_t flags;
+			__le16 flags;
 			uint32_t residual;
-			uint16_t ox_id;
+			__le16 ox_id;
 			uint16_t scsi_status;
 			uint32_t relative_offset;
 			uint32_t reserved2;
@@ -457,7 +458,7 @@ struct ctio7_to_24xx {
 			uint16_t sense_length;
 			uint16_t flags;
 			uint32_t residual;
-			uint16_t ox_id;
+			__le16 ox_id;
 			uint16_t scsi_status;
 			uint16_t response_len;
 			uint16_t reserved;
@@ -498,11 +499,12 @@ struct ctio7_from_24xx {
 #define CTIO7_FLAGS_DONT_RET_CTIO	BIT_8
 #define CTIO7_FLAGS_STATUS_MODE_0	0
 #define CTIO7_FLAGS_STATUS_MODE_1	BIT_6
+#define CTIO7_FLAGS_STATUS_MODE_2	BIT_7
 #define CTIO7_FLAGS_EXPLICIT_CONFORM	BIT_5
 #define CTIO7_FLAGS_CONFIRM_SATISF	BIT_4
 #define CTIO7_FLAGS_DSD_PTR		BIT_2
-#define CTIO7_FLAGS_DATA_IN		BIT_1
-#define CTIO7_FLAGS_DATA_OUT		BIT_0
+#define CTIO7_FLAGS_DATA_IN		BIT_1 /* data to initiator */
+#define CTIO7_FLAGS_DATA_OUT		BIT_0 /* data from initiator */
 
 #define ELS_PLOGI			0x3
 #define ELS_FLOGI			0x4
@@ -514,6 +516,68 @@ struct ctio7_from_24xx {
 #define ELS_ADISC			0x52
 
 /*
+ *CTIO Type CRC_2 IOCB
+ */
+struct ctio_crc2_to_fw {
+	uint8_t entry_type;		/* Entry type. */
+#define CTIO_CRC2 0x7A
+	uint8_t entry_count;		/* Entry count. */
+	uint8_t sys_define;		/* System defined. */
+	uint8_t entry_status;		/* Entry Status. */
+
+	uint32_t handle;		/* System handle. */
+	uint16_t nport_handle;		/* N_PORT handle. */
+	__le16 timeout;		/* Command timeout. */
+
+	uint16_t dseg_count;		/* Data segment count. */
+	uint8_t  vp_index;
+	uint8_t  add_flags;		/* additional flags */
+#define CTIO_CRC2_AF_DIF_DSD_ENA BIT_3
+
+	uint8_t  initiator_id[3];	/* initiator ID */
+	uint8_t  reserved1;
+	uint32_t exchange_addr;		/* rcv exchange address */
+	uint16_t reserved2;
+	__le16 flags;			/* refer to CTIO7 flags values */
+	uint32_t residual;
+	__le16 ox_id;
+	uint16_t scsi_status;
+	__le32 relative_offset;
+	uint32_t reserved5;
+	__le32 transfer_length;		/* total fc transfer length */
+	uint32_t reserved6;
+	__le32 crc_context_address[2];/* Data segment address. */
+	uint16_t crc_context_len;	/* Data segment length. */
+	uint16_t reserved_1;		/* MUST be set to 0. */
+} __packed;
+
+/* CTIO Type CRC_x Status IOCB */
+struct ctio_crc_from_fw {
+	uint8_t entry_type;		/* Entry type. */
+	uint8_t entry_count;		/* Entry count. */
+	uint8_t sys_define;		/* System defined. */
+	uint8_t entry_status;		/* Entry Status. */
+
+	uint32_t handle;		/* System handle. */
+	uint16_t status;
+	uint16_t timeout;		/* Command timeout. */
+	uint16_t dseg_count;		/* Data segment count. */
+	uint32_t reserved1;
+	uint16_t state_flags;
+#define CTIO_CRC_SF_DIF_CHOPPED BIT_4
+
+	uint32_t exchange_address;	/* rcv exchange address */
+	uint16_t reserved2;
+	uint16_t flags;
+	uint32_t resid_xfer_length;
+	uint16_t ox_id;
+	uint8_t  reserved3[12];
+	uint16_t runt_guard;		/* reported runt blk guard */
+	uint8_t  actual_dif[8];
+	uint8_t  expected_dif[8];
+} __packed;
+
+/*
  * ISP queue - ABTS received/response entries structure definition for 24xx.
  */
 #define ABTS_RECV_24XX		0x54 /* ABTS received (for 24xx) */
@@ -641,6 +705,7 @@ struct qla_tgt_func_tmpl {
 	int (*handle_cmd)(struct scsi_qla_host *, struct qla_tgt_cmd *,
 			unsigned char *, uint32_t, int, int, int);
 	void (*handle_data)(struct qla_tgt_cmd *);
+	void (*handle_dif_err)(struct qla_tgt_cmd *);
 	int (*handle_tmr)(struct qla_tgt_mgmt_cmd *, uint32_t, uint8_t,
 			uint32_t);
 	void (*free_cmd)(struct qla_tgt_cmd *);
@@ -805,6 +870,12 @@ struct qla_tgt {
 	struct list_head tgt_list_entry;
 };
 
+struct qla_tgt_sess_op {
+	struct scsi_qla_host *vha;
+	struct atio_from_isp atio;
+	struct work_struct work;
+};
+
 /*
  * Equivilant to IT Nexus (Initiator-Target)
  */
@@ -829,9 +900,9 @@ struct qla_tgt_sess {
 };
 
 struct qla_tgt_cmd {
+	struct se_cmd se_cmd;
 	struct qla_tgt_sess *sess;
 	int state;
-	struct se_cmd se_cmd;
 	struct work_struct free_work;
 	struct work_struct work;
 	/* Sense buffer that will be mapped into outgoing status */
@@ -843,6 +914,7 @@ struct qla_tgt_cmd {
 	unsigned int free_sg:1;
 	unsigned int aborted:1; /* Needed in case of SRR */
 	unsigned int write_data_transferred:1;
+	unsigned int ctx_dsd_alloced:1;
 
 	struct scatterlist *sg;	/* cmd data buffer SG vector */
 	int sg_cnt;		/* SG segments count */
@@ -857,6 +929,12 @@ struct qla_tgt_cmd {
 	struct scsi_qla_host *vha;
 
 	struct atio_from_isp atio;
+	/* t10dif */
+	struct scatterlist *prot_sg;
+	uint32_t prot_sg_cnt;
+	uint32_t blk_sz;
+	struct crc_context *ctx;
+
 };
 
 struct qla_tgt_sess_work_param {
@@ -901,6 +979,10 @@ struct qla_tgt_prm {
 	int sense_buffer_len;
 	int residual;
 	int add_status_pkt;
+	/* dif */
+	struct scatterlist *prot_sg;
+	uint16_t prot_seg_cnt;
+	uint16_t tot_dsds;
 };
 
 struct qla_tgt_srr_imm {
@@ -976,6 +1058,8 @@ extern void qlt_24xx_atio_pkt_all_vps(struct scsi_qla_host *,
 extern void qlt_response_pkt_all_vps(struct scsi_qla_host *, response_t *);
 extern int qlt_rdy_to_xfer(struct qla_tgt_cmd *);
 extern int qlt_xmit_response(struct qla_tgt_cmd *, int, uint8_t);
+extern int qlt_rdy_to_xfer_dif(struct qla_tgt_cmd *);
+extern int qlt_xmit_response_dif(struct qla_tgt_cmd *, int, uint8_t);
 extern void qlt_xmit_tm_rsp(struct qla_tgt_mgmt_cmd *);
 extern void qlt_free_mcmd(struct qla_tgt_mgmt_cmd *);
 extern void qlt_free_cmd(struct qla_tgt_cmd *cmd);
diff --git a/drivers/scsi/qla2xxx/qla_tmpl.c b/drivers/scsi/qla2xxx/qla_tmpl.c
index a804e9b744bb..cb9a0c4bc419 100644
--- a/drivers/scsi/qla2xxx/qla_tmpl.c
+++ b/drivers/scsi/qla2xxx/qla_tmpl.c
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -201,7 +201,6 @@ qla27xx_read_reg(__iomem struct device_reg_24xx *reg,
 		ql_dbg(ql_dbg_misc, NULL, 0xd014,
 		    "%s: @%x\n", __func__, offset);
 	}
-	qla27xx_insert32(offset, buf, len);
 	qla27xx_read32(window, buf, len);
 }
 
@@ -220,7 +219,7 @@ qla27xx_write_reg(__iomem struct device_reg_24xx *reg,
 
 static inline void
 qla27xx_read_window(__iomem struct device_reg_24xx *reg,
-	uint32_t base, uint offset, uint count, uint width, void *buf,
+	uint32_t addr, uint offset, uint count, uint width, void *buf,
 	ulong *len)
 {
 	void *window = (void *)reg + offset;
@@ -229,14 +228,14 @@ qla27xx_read_window(__iomem struct device_reg_24xx *reg,
 	if (buf) {
 		ql_dbg(ql_dbg_misc, NULL, 0xd016,
 		    "%s: base=%x offset=%x count=%x width=%x\n",
-		    __func__, base, offset, count, width);
+		    __func__, addr, offset, count, width);
 	}
-	qla27xx_write_reg(reg, IOBASE_ADDR, base, buf);
+	qla27xx_write_reg(reg, IOBASE_ADDR, addr, buf);
 	while (count--) {
-		qla27xx_insert32(base, buf, len);
+		qla27xx_insert32(addr, buf, len);
 		readn(window, buf, len);
 		window += width;
-		base += width;
+		addr++;
 	}
 }
 
@@ -336,7 +335,8 @@ qla27xx_fwdt_entry_t260(struct scsi_qla_host *vha,
 
 	ql_dbg(ql_dbg_misc, vha, 0xd204,
 	    "%s: rdpci [%lx]\n", __func__, *len);
-	qla27xx_read_reg(reg, ent->t260.pci_addr, buf, len);
+	qla27xx_insert32(ent->t260.pci_offset, buf, len);
+	qla27xx_read_reg(reg, ent->t260.pci_offset, buf, len);
 
 	return false;
 }
@@ -349,7 +349,7 @@ qla27xx_fwdt_entry_t261(struct scsi_qla_host *vha,
 
 	ql_dbg(ql_dbg_misc, vha, 0xd205,
 	    "%s: wrpci [%lx]\n", __func__, *len);
-	qla27xx_write_reg(reg, ent->t261.pci_addr, ent->t261.write_data, buf);
+	qla27xx_write_reg(reg, ent->t261.pci_offset, ent->t261.write_data, buf);
 
 	return false;
 }
@@ -392,9 +392,9 @@ qla27xx_fwdt_entry_t262(struct scsi_qla_host *vha,
 		goto done;
 	}
 
-	if (end < start) {
+	if (end < start || end == 0) {
 		ql_dbg(ql_dbg_misc, vha, 0xd023,
-		    "%s: bad range (start=%x end=%x)\n", __func__,
+		    "%s: unusable range (start=%x end=%x)\n", __func__,
 		    ent->t262.end_addr, ent->t262.start_addr);
 		qla27xx_skip_entry(ent, buf);
 		goto done;
@@ -452,17 +452,15 @@ qla27xx_fwdt_entry_t263(struct scsi_qla_host *vha,
 		ql_dbg(ql_dbg_misc, vha, 0xd025,
 		    "%s: unsupported atio queue\n", __func__);
 		qla27xx_skip_entry(ent, buf);
-		goto done;
 	} else {
 		ql_dbg(ql_dbg_misc, vha, 0xd026,
 		    "%s: unknown queue %u\n", __func__, ent->t263.queue_type);
 		qla27xx_skip_entry(ent, buf);
-		goto done;
 	}
 
 	if (buf)
 		ent->t263.num_queues = count;
-done:
+
 	return false;
 }
 
@@ -503,7 +501,7 @@ qla27xx_fwdt_entry_t265(struct scsi_qla_host *vha,
 	ql_dbg(ql_dbg_misc, vha, 0xd209,
 	    "%s: pause risc [%lx]\n", __func__, *len);
 	if (buf)
-		qla24xx_pause_risc(reg);
+		qla24xx_pause_risc(reg, vha->hw);
 
 	return false;
 }
@@ -590,7 +588,6 @@ qla27xx_fwdt_entry_t270(struct scsi_qla_host *vha,
 	struct qla27xx_fwdt_entry *ent, void *buf, ulong *len)
 {
 	struct device_reg_24xx __iomem *reg = qla27xx_isp_reg(vha);
-	void *window = (void *)reg + 0xc4;
 	ulong dwords = ent->t270.count;
 	ulong addr = ent->t270.addr;
 
@@ -599,10 +596,9 @@ qla27xx_fwdt_entry_t270(struct scsi_qla_host *vha,
 	qla27xx_write_reg(reg, IOBASE_ADDR, 0x40, buf);
 	while (dwords--) {
 		qla27xx_write_reg(reg, 0xc0, addr|0x80000000, buf);
-		qla27xx_read_reg(reg, 0xc4, buf, len);
 		qla27xx_insert32(addr, buf, len);
-		qla27xx_read32(window, buf, len);
-		addr++;
+		qla27xx_read_reg(reg, 0xc4, buf, len);
+		addr += sizeof(uint32_t);
 	}
 
 	return false;
@@ -614,12 +610,12 @@ qla27xx_fwdt_entry_t271(struct scsi_qla_host *vha,
 {
 	struct device_reg_24xx __iomem *reg = qla27xx_isp_reg(vha);
 	ulong addr = ent->t271.addr;
+	ulong data = ent->t271.data;
 
 	ql_dbg(ql_dbg_misc, vha, 0xd20f,
 	    "%s: wrremreg [%lx]\n", __func__, *len);
 	qla27xx_write_reg(reg, IOBASE_ADDR, 0x40, buf);
-	qla27xx_read_reg(reg, 0xc4, buf, len);
-	qla27xx_insert32(addr, buf, len);
+	qla27xx_write_reg(reg, 0xc4, data, buf);
 	qla27xx_write_reg(reg, 0xc0, addr, buf);
 
 	return false;
@@ -662,9 +658,59 @@ qla27xx_fwdt_entry_t273(struct scsi_qla_host *vha,
 			    "%s: failed pcicfg read at %lx\n", __func__, addr);
 		qla27xx_insert32(addr, buf, len);
 		qla27xx_insert32(value, buf, len);
-		addr += 4;
+		addr += sizeof(uint32_t);
+	}
+
+	return false;
+}
+
+static int
+qla27xx_fwdt_entry_t274(struct scsi_qla_host *vha,
+	struct qla27xx_fwdt_entry *ent, void *buf, ulong *len)
+{
+	uint count = 0;
+	uint i;
+
+	ql_dbg(ql_dbg_misc, vha, 0xd212,
+	    "%s: getqsh(%x) [%lx]\n", __func__, ent->t274.queue_type, *len);
+	if (ent->t274.queue_type == T274_QUEUE_TYPE_REQ_SHAD) {
+		for (i = 0; i < vha->hw->max_req_queues; i++) {
+			struct req_que *req = vha->hw->req_q_map[i];
+			if (req || !buf) {
+				qla27xx_insert16(i, buf, len);
+				qla27xx_insert16(1, buf, len);
+				qla27xx_insert32(req && req->out_ptr ?
+				    *req->out_ptr : 0, buf, len);
+				count++;
+			}
+		}
+	} else if (ent->t274.queue_type == T274_QUEUE_TYPE_RSP_SHAD) {
+		for (i = 0; i < vha->hw->max_rsp_queues; i++) {
+			struct rsp_que *rsp = vha->hw->rsp_q_map[i];
+			if (rsp || !buf) {
+				qla27xx_insert16(i, buf, len);
+				qla27xx_insert16(1, buf, len);
+				qla27xx_insert32(rsp && rsp->in_ptr ?
+				    *rsp->in_ptr : 0, buf, len);
+				count++;
+			}
+		}
+	} else if (ent->t274.queue_type == T274_QUEUE_TYPE_ATIO_SHAD) {
+		ql_dbg(ql_dbg_misc, vha, 0xd02e,
+		    "%s: unsupported atio queue\n", __func__);
+		qla27xx_skip_entry(ent, buf);
+	} else {
+		ql_dbg(ql_dbg_misc, vha, 0xd02f,
+		    "%s: unknown queue %u\n", __func__, ent->t274.queue_type);
+		qla27xx_skip_entry(ent, buf);
 	}
 
+	if (buf)
+		ent->t274.num_queues = count;
+
+	if (!count)
+		qla27xx_skip_entry(ent, buf);
+
 	return false;
 }
 
@@ -709,6 +755,7 @@ static struct qla27xx_fwdt_entry_call ql27xx_fwdt_entry_call_list[] = {
 	{ ENTRY_TYPE_WRREMREG		, qla27xx_fwdt_entry_t271  } ,
 	{ ENTRY_TYPE_RDREMRAM		, qla27xx_fwdt_entry_t272  } ,
 	{ ENTRY_TYPE_PCICFG		, qla27xx_fwdt_entry_t273  } ,
+	{ ENTRY_TYPE_GET_SHADOW		, qla27xx_fwdt_entry_t274  } ,
 	{ -1				, qla27xx_fwdt_entry_other }
 };
 
diff --git a/drivers/scsi/qla2xxx/qla_tmpl.h b/drivers/scsi/qla2xxx/qla_tmpl.h
index c9d2fff4d964..1967424c8e64 100644
--- a/drivers/scsi/qla2xxx/qla_tmpl.h
+++ b/drivers/scsi/qla2xxx/qla_tmpl.h
@@ -1,6 +1,6 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
@@ -52,6 +52,7 @@ struct __packed qla27xx_fwdt_template {
 #define ENTRY_TYPE_WRREMREG		271
 #define ENTRY_TYPE_RDREMRAM		272
 #define ENTRY_TYPE_PCICFG		273
+#define ENTRY_TYPE_GET_SHADOW		274
 
 #define CAPTURE_FLAG_PHYS_ONLY		BIT_0
 #define CAPTURE_FLAG_PHYS_VIRT		BIT_1
@@ -109,12 +110,12 @@ struct __packed qla27xx_fwdt_entry {
 		} t259;
 
 		struct __packed {
-			uint8_t pci_addr;
+			uint8_t pci_offset;
 			uint8_t reserved[3];
 		} t260;
 
 		struct __packed {
-			uint8_t pci_addr;
+			uint8_t pci_offset;
 			uint8_t reserved[3];
 			uint32_t write_data;
 		} t261;
@@ -186,6 +187,12 @@ struct __packed qla27xx_fwdt_entry {
 			uint32_t addr;
 			uint32_t count;
 		} t273;
+
+		struct __packed {
+			uint32_t num_queues;
+			uint8_t  queue_type;
+			uint8_t  reserved[3];
+		} t274;
 	};
 };
 
@@ -202,4 +209,8 @@ struct __packed qla27xx_fwdt_entry {
 #define T268_BUF_TYPE_EXCH_BUFOFF	2
 #define T268_BUF_TYPE_EXTD_LOGIN	3
 
+#define T274_QUEUE_TYPE_REQ_SHAD	1
+#define T274_QUEUE_TYPE_RSP_SHAD	2
+#define T274_QUEUE_TYPE_ATIO_SHAD	3
+
 #endif
diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h
index e36b94712544..4d2c98cbec4f 100644
--- a/drivers/scsi/qla2xxx/qla_version.h
+++ b/drivers/scsi/qla2xxx/qla_version.h
@@ -1,13 +1,13 @@
 /*
  * QLogic Fibre Channel HBA Driver
- * Copyright (c)  2003-2013 QLogic Corporation
+ * Copyright (c)  2003-2014 QLogic Corporation
  *
  * See LICENSE.qla2xxx for copyright and licensing details.
  */
 /*
  * Driver version
  */
-#define QLA2XXX_VERSION      "8.07.00.02-k"
+#define QLA2XXX_VERSION      "8.07.00.08-k"
 
 #define QLA_DRIVER_MAJOR_VER	8
 #define QLA_DRIVER_MINOR_VER	7
diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
index 68fb66fdb757..e2beab962096 100644
--- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c
+++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
@@ -472,6 +472,11 @@ static int tcm_qla2xxx_write_pending(struct se_cmd *se_cmd)
 	cmd->sg_cnt = se_cmd->t_data_nents;
 	cmd->sg = se_cmd->t_data_sg;
 
+	cmd->prot_sg_cnt = se_cmd->t_prot_nents;
+	cmd->prot_sg = se_cmd->t_prot_sg;
+	cmd->blk_sz  = se_cmd->se_dev->dev_attrib.block_size;
+	se_cmd->pi_err = 0;
+
 	/*
 	 * qla_target.c:qlt_rdy_to_xfer() will call pci_map_sg() to setup
 	 * the SGL mappings into PCIe memory for incoming FCP WRITE data.
@@ -567,8 +572,13 @@ static void tcm_qla2xxx_handle_data_work(struct work_struct *work)
 			return;
 		}
 
-		transport_generic_request_failure(&cmd->se_cmd,
-						  TCM_CHECK_CONDITION_ABORT_CMD);
+		if (cmd->se_cmd.pi_err)
+			transport_generic_request_failure(&cmd->se_cmd,
+				cmd->se_cmd.pi_err);
+		else
+			transport_generic_request_failure(&cmd->se_cmd,
+				TCM_CHECK_CONDITION_ABORT_CMD);
+
 		return;
 	}
 
@@ -584,6 +594,27 @@ static void tcm_qla2xxx_handle_data(struct qla_tgt_cmd *cmd)
 	queue_work(tcm_qla2xxx_free_wq, &cmd->work);
 }
 
+static void tcm_qla2xxx_handle_dif_work(struct work_struct *work)
+{
+	struct qla_tgt_cmd *cmd = container_of(work, struct qla_tgt_cmd, work);
+
+	/* take an extra kref to prevent cmd free too early.
+	 * need to wait for SCSI status/check condition to
+	 * finish responding generate by transport_generic_request_failure.
+	 */
+	kref_get(&cmd->se_cmd.cmd_kref);
+	transport_generic_request_failure(&cmd->se_cmd, cmd->se_cmd.pi_err);
+}
+
+/*
+ * Called from qla_target.c:qlt_do_ctio_completion()
+ */
+static void tcm_qla2xxx_handle_dif_err(struct qla_tgt_cmd *cmd)
+{
+	INIT_WORK(&cmd->work, tcm_qla2xxx_handle_dif_work);
+	queue_work(tcm_qla2xxx_free_wq, &cmd->work);
+}
+
 /*
  * Called from qla_target.c:qlt_issue_task_mgmt()
  */
@@ -610,6 +641,11 @@ static int tcm_qla2xxx_queue_data_in(struct se_cmd *se_cmd)
 	cmd->sg = se_cmd->t_data_sg;
 	cmd->offset = 0;
 
+	cmd->prot_sg_cnt = se_cmd->t_prot_nents;
+	cmd->prot_sg = se_cmd->t_prot_sg;
+	cmd->blk_sz  = se_cmd->se_dev->dev_attrib.block_size;
+	se_cmd->pi_err = 0;
+
 	/*
 	 * Now queue completed DATA_IN the qla2xxx LLD and response ring
 	 */
@@ -1465,6 +1501,8 @@ static int tcm_qla2xxx_check_initiator_node_acl(
 	struct qla_tgt_sess *sess = qla_tgt_sess;
 	unsigned char port_name[36];
 	unsigned long flags;
+	int num_tags = (ha->fw_xcb_count) ? ha->fw_xcb_count :
+		       TCM_QLA2XXX_DEFAULT_TAGS;
 
 	lport = vha->vha_tgt.target_lport_ptr;
 	if (!lport) {
@@ -1482,7 +1520,9 @@ static int tcm_qla2xxx_check_initiator_node_acl(
 	}
 	se_tpg = &tpg->se_tpg;
 
-	se_sess = transport_init_session(TARGET_PROT_NORMAL);
+	se_sess = transport_init_session_tags(num_tags,
+					      sizeof(struct qla_tgt_cmd),
+					      TARGET_PROT_NORMAL);
 	if (IS_ERR(se_sess)) {
 		pr_err("Unable to initialize struct se_session\n");
 		return PTR_ERR(se_sess);
@@ -1600,6 +1640,7 @@ static void tcm_qla2xxx_update_sess(struct qla_tgt_sess *sess, port_id_t s_id,
 static struct qla_tgt_func_tmpl tcm_qla2xxx_template = {
 	.handle_cmd		= tcm_qla2xxx_handle_cmd,
 	.handle_data		= tcm_qla2xxx_handle_data,
+	.handle_dif_err		= tcm_qla2xxx_handle_dif_err,
 	.handle_tmr		= tcm_qla2xxx_handle_tmr,
 	.free_cmd		= tcm_qla2xxx_free_cmd,
 	.free_mcmd		= tcm_qla2xxx_free_mcmd,
diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.h b/drivers/scsi/qla2xxx/tcm_qla2xxx.h
index 33aaac8c7d59..10c002145648 100644
--- a/drivers/scsi/qla2xxx/tcm_qla2xxx.h
+++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.h
@@ -4,6 +4,11 @@
 #define TCM_QLA2XXX_VERSION	"v0.1"
 /* length of ASCII WWPNs including pad */
 #define TCM_QLA2XXX_NAMELEN	32
+/*
+ * Number of pre-allocated per-session tags, based upon the worst-case
+ * per port number of iocbs
+ */
+#define TCM_QLA2XXX_DEFAULT_TAGS 2088
 
 #include "qla_target.h"
 
diff --git a/drivers/scsi/qla4xxx/ql4_83xx.c b/drivers/scsi/qla4xxx/ql4_83xx.c
index 2eba35365920..556c1525f881 100644
--- a/drivers/scsi/qla4xxx/ql4_83xx.c
+++ b/drivers/scsi/qla4xxx/ql4_83xx.c
@@ -249,110 +249,6 @@ void qla4_83xx_rom_lock_recovery(struct scsi_qla_host *ha)
 	qla4_83xx_flash_unlock(ha);
 }
 
-/**
- * qla4_83xx_ms_mem_write_128b - Writes data to MS/off-chip memory
- * @ha: Pointer to adapter structure
- * @addr: Flash address to write to
- * @data: Data to be written
- * @count: word_count to be written
- *
- * Return: On success return QLA_SUCCESS
- *	   On error return QLA_ERROR
- **/
-int qla4_83xx_ms_mem_write_128b(struct scsi_qla_host *ha, uint64_t addr,
-				uint32_t *data, uint32_t count)
-{
-	int i, j;
-	uint32_t agt_ctrl;
-	unsigned long flags;
-	int ret_val = QLA_SUCCESS;
-
-	/* Only 128-bit aligned access */
-	if (addr & 0xF) {
-		ret_val = QLA_ERROR;
-		goto exit_ms_mem_write;
-	}
-
-	write_lock_irqsave(&ha->hw_lock, flags);
-
-	/* Write address */
-	ret_val = qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_HI, 0);
-	if (ret_val == QLA_ERROR) {
-		ql4_printk(KERN_ERR, ha, "%s: write to AGT_ADDR_HI failed\n",
-			   __func__);
-		goto exit_ms_mem_write_unlock;
-	}
-
-	for (i = 0; i < count; i++, addr += 16) {
-		if (!((QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_QDR_NET,
-					     QLA8XXX_ADDR_QDR_NET_MAX)) ||
-		      (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_DDR_NET,
-					     QLA8XXX_ADDR_DDR_NET_MAX)))) {
-			ret_val = QLA_ERROR;
-			goto exit_ms_mem_write_unlock;
-		}
-
-		ret_val = qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_LO,
-						    addr);
-		/* Write data */
-		ret_val |= qla4_83xx_wr_reg_indirect(ha,
-						     MD_MIU_TEST_AGT_WRDATA_LO,
-						     *data++);
-		ret_val |= qla4_83xx_wr_reg_indirect(ha,
-						     MD_MIU_TEST_AGT_WRDATA_HI,
-						     *data++);
-		ret_val |= qla4_83xx_wr_reg_indirect(ha,
-						     MD_MIU_TEST_AGT_WRDATA_ULO,
-						     *data++);
-		ret_val |= qla4_83xx_wr_reg_indirect(ha,
-						     MD_MIU_TEST_AGT_WRDATA_UHI,
-						     *data++);
-		if (ret_val == QLA_ERROR) {
-			ql4_printk(KERN_ERR, ha, "%s: write to AGT_WRDATA failed\n",
-				   __func__);
-			goto exit_ms_mem_write_unlock;
-		}
-
-		/* Check write status */
-		ret_val = qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL,
-						    MIU_TA_CTL_WRITE_ENABLE);
-		ret_val |= qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL,
-						     MIU_TA_CTL_WRITE_START);
-		if (ret_val == QLA_ERROR) {
-			ql4_printk(KERN_ERR, ha, "%s: write to AGT_CTRL failed\n",
-				   __func__);
-			goto exit_ms_mem_write_unlock;
-		}
-
-		for (j = 0; j < MAX_CTL_CHECK; j++) {
-			ret_val = qla4_83xx_rd_reg_indirect(ha,
-							MD_MIU_TEST_AGT_CTRL,
-							&agt_ctrl);
-			if (ret_val == QLA_ERROR) {
-				ql4_printk(KERN_ERR, ha, "%s: failed to read MD_MIU_TEST_AGT_CTRL\n",
-					   __func__);
-				goto exit_ms_mem_write_unlock;
-			}
-			if ((agt_ctrl & MIU_TA_CTL_BUSY) == 0)
-				break;
-		}
-
-		/* Status check failed */
-		if (j >= MAX_CTL_CHECK) {
-			printk_ratelimited(KERN_ERR "%s: MS memory write failed!\n",
-					   __func__);
-			ret_val = QLA_ERROR;
-			goto exit_ms_mem_write_unlock;
-		}
-	}
-
-exit_ms_mem_write_unlock:
-	write_unlock_irqrestore(&ha->hw_lock, flags);
-
-exit_ms_mem_write:
-	return ret_val;
-}
-
 #define INTENT_TO_RECOVER	0x01
 #define PROCEED_TO_RECOVER	0x02
 
@@ -760,7 +656,7 @@ static int qla4_83xx_copy_bootloader(struct scsi_qla_host *ha)
 			  __func__));
 
 	/* 128 bit/16 byte write to MS memory */
-	ret_val = qla4_83xx_ms_mem_write_128b(ha, dest, (uint32_t *)p_cache,
+	ret_val = qla4_8xxx_ms_mem_write_128b(ha, dest, (uint32_t *)p_cache,
 					      count);
 	if (ret_val == QLA_ERROR) {
 		ql4_printk(KERN_ERR, ha, "%s: Error writing firmware to MS\n",
diff --git a/drivers/scsi/qla4xxx/ql4_83xx.h b/drivers/scsi/qla4xxx/ql4_83xx.h
index a0de6e25ea5a..775fdf9fcc87 100644
--- a/drivers/scsi/qla4xxx/ql4_83xx.h
+++ b/drivers/scsi/qla4xxx/ql4_83xx.h
@@ -254,6 +254,50 @@ struct qla83xx_minidump_entry_pollrd {
 	uint32_t rsvd_1;
 };
 
+struct qla8044_minidump_entry_rddfe {
+	struct qla8xxx_minidump_entry_hdr h;
+	uint32_t addr_1;
+	uint32_t value;
+	uint8_t stride;
+	uint8_t stride2;
+	uint16_t count;
+	uint32_t poll;
+	uint32_t mask;
+	uint32_t modify_mask;
+	uint32_t data_size;
+	uint32_t rsvd;
+
+} __packed;
+
+struct qla8044_minidump_entry_rdmdio {
+	struct qla8xxx_minidump_entry_hdr h;
+
+	uint32_t addr_1;
+	uint32_t addr_2;
+	uint32_t value_1;
+	uint8_t stride_1;
+	uint8_t stride_2;
+	uint16_t count;
+	uint32_t poll;
+	uint32_t mask;
+	uint32_t value_2;
+	uint32_t data_size;
+
+} __packed;
+
+struct qla8044_minidump_entry_pollwr {
+	struct qla8xxx_minidump_entry_hdr h;
+	uint32_t addr_1;
+	uint32_t addr_2;
+	uint32_t value_1;
+	uint32_t value_2;
+	uint32_t poll;
+	uint32_t mask;
+	uint32_t data_size;
+	uint32_t rsvd;
+
+} __packed;
+
 /* RDMUX2 Entry */
 struct qla83xx_minidump_entry_rdmux2 {
 	struct qla8xxx_minidump_entry_hdr h;
diff --git a/drivers/scsi/qla4xxx/ql4_def.h b/drivers/scsi/qla4xxx/ql4_def.h
index 73a502288bde..8f6d0fb2cd80 100644
--- a/drivers/scsi/qla4xxx/ql4_def.h
+++ b/drivers/scsi/qla4xxx/ql4_def.h
@@ -601,6 +601,7 @@ struct scsi_qla_host {
 #define DPC_HA_NEED_QUIESCENT		22 /* 0x00400000 ISP-82xx only*/
 #define DPC_POST_IDC_ACK		23 /* 0x00800000 */
 #define DPC_RESTORE_ACB			24 /* 0x01000000 */
+#define DPC_SYSFS_DDB_EXPORT		25 /* 0x02000000 */
 
 	struct Scsi_Host *host; /* pointer to host data */
 	uint32_t tot_ddbs;
diff --git a/drivers/scsi/qla4xxx/ql4_fw.h b/drivers/scsi/qla4xxx/ql4_fw.h
index 209853ce0bbc..699575efc9ba 100644
--- a/drivers/scsi/qla4xxx/ql4_fw.h
+++ b/drivers/scsi/qla4xxx/ql4_fw.h
@@ -1415,6 +1415,9 @@ struct ql_iscsi_stats {
 #define QLA83XX_DBG_OCM_WNDREG_ARRAY_LEN	16
 #define QLA83XX_SS_OCM_WNDREG_INDEX		3
 #define QLA83XX_SS_PCI_INDEX			0
+#define QLA8022_TEMPLATE_CAP_OFFSET		172
+#define QLA83XX_TEMPLATE_CAP_OFFSET		268
+#define QLA80XX_TEMPLATE_RESERVED_BITS		16
 
 struct qla4_8xxx_minidump_template_hdr {
 	uint32_t entry_type;
@@ -1434,6 +1437,7 @@ struct qla4_8xxx_minidump_template_hdr {
 	uint32_t saved_state_array[QLA8XXX_DBG_STATE_ARRAY_LEN];
 	uint32_t capture_size_array[QLA8XXX_DBG_CAP_SIZE_ARRAY_LEN];
 	uint32_t ocm_window_reg[QLA83XX_DBG_OCM_WNDREG_ARRAY_LEN];
+	uint32_t capabilities[QLA80XX_TEMPLATE_RESERVED_BITS];
 };
 
 #endif /*  _QLA4X_FW_H */
diff --git a/drivers/scsi/qla4xxx/ql4_glbl.h b/drivers/scsi/qla4xxx/ql4_glbl.h
index b1a19cd8d5b2..5f58b451327e 100644
--- a/drivers/scsi/qla4xxx/ql4_glbl.h
+++ b/drivers/scsi/qla4xxx/ql4_glbl.h
@@ -274,13 +274,14 @@ int qla4xxx_set_acb(struct scsi_qla_host *ha, uint32_t *mbox_cmd,
 int qla4xxx_get_acb(struct scsi_qla_host *ha, dma_addr_t acb_dma,
 		    uint32_t acb_type, uint32_t len);
 int qla4_84xx_config_acb(struct scsi_qla_host *ha, int acb_config);
-int qla4_83xx_ms_mem_write_128b(struct scsi_qla_host *ha,
+int qla4_8xxx_ms_mem_write_128b(struct scsi_qla_host *ha,
 				uint64_t addr, uint32_t *data, uint32_t count);
 uint8_t qla4xxx_set_ipaddr_state(uint8_t fw_ipaddr_state);
 int qla4_83xx_get_port_config(struct scsi_qla_host *ha, uint32_t *config);
 int qla4_83xx_set_port_config(struct scsi_qla_host *ha, uint32_t *config);
 int qla4_8xxx_check_init_adapter_retry(struct scsi_qla_host *ha);
 int qla4_83xx_is_detached(struct scsi_qla_host *ha);
+int qla4xxx_sysfs_ddb_export(struct scsi_qla_host *ha);
 
 extern int ql4xextended_error_logging;
 extern int ql4xdontresethba;
diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c
index 28fbece7e08f..6f12f859b11d 100644
--- a/drivers/scsi/qla4xxx/ql4_init.c
+++ b/drivers/scsi/qla4xxx/ql4_init.c
@@ -282,6 +282,25 @@ qla4xxx_wait_for_ip_config(struct scsi_qla_host *ha)
 	return ipv4_wait|ipv6_wait;
 }
 
+static int qla4_80xx_is_minidump_dma_capable(struct scsi_qla_host *ha,
+		struct qla4_8xxx_minidump_template_hdr *md_hdr)
+{
+	int offset = (is_qla8022(ha)) ? QLA8022_TEMPLATE_CAP_OFFSET :
+					QLA83XX_TEMPLATE_CAP_OFFSET;
+	int rval = 1;
+	uint32_t *cap_offset;
+
+	cap_offset = (uint32_t *)((char *)md_hdr + offset);
+
+	if (!(le32_to_cpu(*cap_offset) & BIT_0)) {
+		ql4_printk(KERN_INFO, ha, "PEX DMA Not supported %d\n",
+			   *cap_offset);
+		rval = 0;
+	}
+
+	return rval;
+}
+
 /**
  * qla4xxx_alloc_fw_dump - Allocate memory for minidump data.
  * @ha: pointer to host adapter structure.
@@ -294,6 +313,7 @@ void qla4xxx_alloc_fw_dump(struct scsi_qla_host *ha)
 	void *md_tmp;
 	dma_addr_t md_tmp_dma;
 	struct qla4_8xxx_minidump_template_hdr *md_hdr;
+	int dma_capable;
 
 	if (ha->fw_dump) {
 		ql4_printk(KERN_WARNING, ha,
@@ -326,13 +346,19 @@ void qla4xxx_alloc_fw_dump(struct scsi_qla_host *ha)
 
 	md_hdr = (struct qla4_8xxx_minidump_template_hdr *)md_tmp;
 
+	dma_capable = qla4_80xx_is_minidump_dma_capable(ha, md_hdr);
+
 	capture_debug_level = md_hdr->capture_debug_level;
 
 	/* Get capture mask based on module loadtime setting. */
-	if (ql4xmdcapmask >= 0x3 && ql4xmdcapmask <= 0x7F)
+	if ((ql4xmdcapmask >= 0x3 && ql4xmdcapmask <= 0x7F) ||
+	    (ql4xmdcapmask == 0xFF && dma_capable))  {
 		ha->fw_dump_capture_mask = ql4xmdcapmask;
-	else
+	} else {
+		if (ql4xmdcapmask == 0xFF)
+			ql4_printk(KERN_INFO, ha, "Falling back to default capture mask, as PEX DMA is not supported\n");
 		ha->fw_dump_capture_mask = capture_debug_level;
+	}
 
 	md_hdr->driver_capture_mask = ha->fw_dump_capture_mask;
 
@@ -864,6 +890,8 @@ int qla4xxx_start_firmware(struct scsi_qla_host *ha)
 	if (status == QLA_SUCCESS) {
 		if (test_and_clear_bit(AF_GET_CRASH_RECORD, &ha->flags))
 			qla4xxx_get_crash_record(ha);
+
+		qla4xxx_init_rings(ha);
 	} else {
 		DEBUG(printk("scsi%ld: %s: Firmware has NOT started\n",
 			     ha->host_no, __func__));
diff --git a/drivers/scsi/qla4xxx/ql4_isr.c b/drivers/scsi/qla4xxx/ql4_isr.c
index b1925d195f41..081b6b78d2c6 100644
--- a/drivers/scsi/qla4xxx/ql4_isr.c
+++ b/drivers/scsi/qla4xxx/ql4_isr.c
@@ -1526,7 +1526,7 @@ void qla4xxx_process_aen(struct scsi_qla_host * ha, uint8_t process_aen)
 
 int qla4xxx_request_irqs(struct scsi_qla_host *ha)
 {
-	int ret;
+	int ret = 0;
 	int rval = QLA_ERROR;
 
 	if (is_qla40XX(ha))
@@ -1580,15 +1580,13 @@ try_msi:
 		}
 	}
 
-	/*
-	 * Prevent interrupts from falling back to INTx mode in cases where
-	 * interrupts cannot get acquired through MSI-X or MSI mode.
-	 */
+try_intx:
 	if (is_qla8022(ha)) {
-		ql4_printk(KERN_WARNING, ha, "IRQ not attached -- %d.\n", ret);
+		ql4_printk(KERN_WARNING, ha, "%s: ISP82xx Legacy interrupt not supported\n",
+			   __func__);
 		goto irq_not_attached;
 	}
-try_intx:
+
 	/* Trying INTx */
 	ret = request_irq(ha->pdev->irq, ha->isp_ops->intr_handler,
 	    IRQF_SHARED, DRIVER_NAME, ha);
diff --git a/drivers/scsi/qla4xxx/ql4_mbx.c b/drivers/scsi/qla4xxx/ql4_mbx.c
index 0a6b782d6fdb..0a3312c6dd6d 100644
--- a/drivers/scsi/qla4xxx/ql4_mbx.c
+++ b/drivers/scsi/qla4xxx/ql4_mbx.c
@@ -2381,7 +2381,7 @@ int qla4_84xx_config_acb(struct scsi_qla_host *ha, int acb_config)
 			ql4_printk(KERN_ERR, ha, "%s: Unable to alloc acb\n",
 				   __func__);
 			rval = QLA_ERROR;
-			goto exit_config_acb;
+			goto exit_free_acb;
 		}
 		memcpy(ha->saved_acb, acb, acb_len);
 		break;
@@ -2395,8 +2395,6 @@ int qla4_84xx_config_acb(struct scsi_qla_host *ha, int acb_config)
 		}
 
 		memcpy(acb, ha->saved_acb, acb_len);
-		kfree(ha->saved_acb);
-		ha->saved_acb = NULL;
 
 		rval = qla4xxx_set_acb(ha, &mbox_cmd[0], &mbox_sts[0], acb_dma);
 		if (rval != QLA_SUCCESS)
@@ -2412,6 +2410,10 @@ exit_free_acb:
 	dma_free_coherent(&ha->pdev->dev, sizeof(struct addr_ctrl_blk), acb,
 			  acb_dma);
 exit_config_acb:
+	if ((acb_config == ACB_CONFIG_SET) && ha->saved_acb) {
+		kfree(ha->saved_acb);
+		ha->saved_acb = NULL;
+	}
 	DEBUG2(ql4_printk(KERN_INFO, ha,
 			  "%s %s\n", __func__,
 			  rval == QLA_SUCCESS ? "SUCCEEDED" : "FAILED"));
diff --git a/drivers/scsi/qla4xxx/ql4_nx.c b/drivers/scsi/qla4xxx/ql4_nx.c
index 63328c812b70..9dbdb4be2d8f 100644
--- a/drivers/scsi/qla4xxx/ql4_nx.c
+++ b/drivers/scsi/qla4xxx/ql4_nx.c
@@ -14,6 +14,7 @@
 
 #include <asm-generic/io-64-nonatomic-lo-hi.h>
 
+#define TIMEOUT_100_MS	100
 #define MASK(n)		DMA_BIT_MASK(n)
 #define MN_WIN(addr)	(((addr & 0x1fc0000) >> 1) | ((addr >> 25) & 0x3ff))
 #define OCM_WIN(addr)	(((addr & 0x1ff0000) >> 1) | ((addr >> 25) & 0x3ff))
@@ -1176,6 +1177,112 @@ qla4_82xx_pinit_from_rom(struct scsi_qla_host *ha, int verbose)
 	return 0;
 }
 
+/**
+ * qla4_8xxx_ms_mem_write_128b - Writes data to MS/off-chip memory
+ * @ha: Pointer to adapter structure
+ * @addr: Flash address to write to
+ * @data: Data to be written
+ * @count: word_count to be written
+ *
+ * Return: On success return QLA_SUCCESS
+ *         On error return QLA_ERROR
+ **/
+int qla4_8xxx_ms_mem_write_128b(struct scsi_qla_host *ha, uint64_t addr,
+				uint32_t *data, uint32_t count)
+{
+	int i, j;
+	uint32_t agt_ctrl;
+	unsigned long flags;
+	int ret_val = QLA_SUCCESS;
+
+	/* Only 128-bit aligned access */
+	if (addr & 0xF) {
+		ret_val = QLA_ERROR;
+		goto exit_ms_mem_write;
+	}
+
+	write_lock_irqsave(&ha->hw_lock, flags);
+
+	/* Write address */
+	ret_val = ha->isp_ops->wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_HI, 0);
+	if (ret_val == QLA_ERROR) {
+		ql4_printk(KERN_ERR, ha, "%s: write to AGT_ADDR_HI failed\n",
+			   __func__);
+		goto exit_ms_mem_write_unlock;
+	}
+
+	for (i = 0; i < count; i++, addr += 16) {
+		if (!((QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_QDR_NET,
+					     QLA8XXX_ADDR_QDR_NET_MAX)) ||
+		      (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_DDR_NET,
+					     QLA8XXX_ADDR_DDR_NET_MAX)))) {
+			ret_val = QLA_ERROR;
+			goto exit_ms_mem_write_unlock;
+		}
+
+		ret_val = ha->isp_ops->wr_reg_indirect(ha,
+						       MD_MIU_TEST_AGT_ADDR_LO,
+						       addr);
+		/* Write data */
+		ret_val |= ha->isp_ops->wr_reg_indirect(ha,
+						MD_MIU_TEST_AGT_WRDATA_LO,
+						*data++);
+		ret_val |= ha->isp_ops->wr_reg_indirect(ha,
+						MD_MIU_TEST_AGT_WRDATA_HI,
+						*data++);
+		ret_val |= ha->isp_ops->wr_reg_indirect(ha,
+						MD_MIU_TEST_AGT_WRDATA_ULO,
+						*data++);
+		ret_val |= ha->isp_ops->wr_reg_indirect(ha,
+						MD_MIU_TEST_AGT_WRDATA_UHI,
+						*data++);
+		if (ret_val == QLA_ERROR) {
+			ql4_printk(KERN_ERR, ha, "%s: write to AGT_WRDATA failed\n",
+				   __func__);
+			goto exit_ms_mem_write_unlock;
+		}
+
+		/* Check write status */
+		ret_val = ha->isp_ops->wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL,
+						       MIU_TA_CTL_WRITE_ENABLE);
+		ret_val |= ha->isp_ops->wr_reg_indirect(ha,
+							MD_MIU_TEST_AGT_CTRL,
+							MIU_TA_CTL_WRITE_START);
+		if (ret_val == QLA_ERROR) {
+			ql4_printk(KERN_ERR, ha, "%s: write to AGT_CTRL failed\n",
+				   __func__);
+			goto exit_ms_mem_write_unlock;
+		}
+
+		for (j = 0; j < MAX_CTL_CHECK; j++) {
+			ret_val = ha->isp_ops->rd_reg_indirect(ha,
+							MD_MIU_TEST_AGT_CTRL,
+							&agt_ctrl);
+			if (ret_val == QLA_ERROR) {
+				ql4_printk(KERN_ERR, ha, "%s: failed to read MD_MIU_TEST_AGT_CTRL\n",
+					   __func__);
+				goto exit_ms_mem_write_unlock;
+			}
+			if ((agt_ctrl & MIU_TA_CTL_BUSY) == 0)
+				break;
+		}
+
+		/* Status check failed */
+		if (j >= MAX_CTL_CHECK) {
+			printk_ratelimited(KERN_ERR "%s: MS memory write failed!\n",
+					   __func__);
+			ret_val = QLA_ERROR;
+			goto exit_ms_mem_write_unlock;
+		}
+	}
+
+exit_ms_mem_write_unlock:
+	write_unlock_irqrestore(&ha->hw_lock, flags);
+
+exit_ms_mem_write:
+	return ret_val;
+}
+
 static int
 qla4_82xx_load_from_flash(struct scsi_qla_host *ha, uint32_t image_start)
 {
@@ -1714,6 +1821,101 @@ void qla4_82xx_rom_lock_recovery(struct scsi_qla_host *ha)
 	qla4_82xx_rom_unlock(ha);
 }
 
+static uint32_t ql4_84xx_poll_wait_for_ready(struct scsi_qla_host *ha,
+					     uint32_t addr1, uint32_t mask)
+{
+	unsigned long timeout;
+	uint32_t rval = QLA_SUCCESS;
+	uint32_t temp;
+
+	timeout = jiffies + msecs_to_jiffies(TIMEOUT_100_MS);
+	do {
+		ha->isp_ops->rd_reg_indirect(ha, addr1, &temp);
+		if ((temp & mask) != 0)
+			break;
+
+		if (time_after_eq(jiffies, timeout)) {
+			ql4_printk(KERN_INFO, ha, "Error in processing rdmdio entry\n");
+			return QLA_ERROR;
+		}
+	} while (1);
+
+	return rval;
+}
+
+uint32_t ql4_84xx_ipmdio_rd_reg(struct scsi_qla_host *ha, uint32_t addr1,
+				uint32_t addr3, uint32_t mask, uint32_t addr,
+				uint32_t *data_ptr)
+{
+	int rval = QLA_SUCCESS;
+	uint32_t temp;
+	uint32_t data;
+
+	rval = ql4_84xx_poll_wait_for_ready(ha, addr1, mask);
+	if (rval)
+		goto exit_ipmdio_rd_reg;
+
+	temp = (0x40000000 | addr);
+	ha->isp_ops->wr_reg_indirect(ha, addr1, temp);
+
+	rval = ql4_84xx_poll_wait_for_ready(ha, addr1, mask);
+	if (rval)
+		goto exit_ipmdio_rd_reg;
+
+	ha->isp_ops->rd_reg_indirect(ha, addr3, &data);
+	*data_ptr = data;
+
+exit_ipmdio_rd_reg:
+	return rval;
+}
+
+
+static uint32_t ql4_84xx_poll_wait_ipmdio_bus_idle(struct scsi_qla_host *ha,
+						    uint32_t addr1,
+						    uint32_t addr2,
+						    uint32_t addr3,
+						    uint32_t mask)
+{
+	unsigned long timeout;
+	uint32_t temp;
+	uint32_t rval = QLA_SUCCESS;
+
+	timeout = jiffies + msecs_to_jiffies(TIMEOUT_100_MS);
+	do {
+		ql4_84xx_ipmdio_rd_reg(ha, addr1, addr3, mask, addr2, &temp);
+		if ((temp & 0x1) != 1)
+			break;
+		if (time_after_eq(jiffies, timeout)) {
+			ql4_printk(KERN_INFO, ha, "Error in processing mdiobus idle\n");
+			return QLA_ERROR;
+		}
+	} while (1);
+
+	return rval;
+}
+
+static int ql4_84xx_ipmdio_wr_reg(struct scsi_qla_host *ha,
+				  uint32_t addr1, uint32_t addr3,
+				  uint32_t mask, uint32_t addr,
+				  uint32_t value)
+{
+	int rval = QLA_SUCCESS;
+
+	rval = ql4_84xx_poll_wait_for_ready(ha, addr1, mask);
+	if (rval)
+		goto exit_ipmdio_wr_reg;
+
+	ha->isp_ops->wr_reg_indirect(ha, addr3, value);
+	ha->isp_ops->wr_reg_indirect(ha, addr1, addr);
+
+	rval = ql4_84xx_poll_wait_for_ready(ha, addr1, mask);
+	if (rval)
+		goto exit_ipmdio_wr_reg;
+
+exit_ipmdio_wr_reg:
+	return rval;
+}
+
 static void qla4_8xxx_minidump_process_rdcrb(struct scsi_qla_host *ha,
 				struct qla8xxx_minidump_entry_hdr *entry_hdr,
 				uint32_t **d_ptr)
@@ -1822,7 +2024,7 @@ error_exit:
 	return rval;
 }
 
-static int qla4_83xx_minidump_pex_dma_read(struct scsi_qla_host *ha,
+static int qla4_8xxx_minidump_pex_dma_read(struct scsi_qla_host *ha,
 				struct qla8xxx_minidump_entry_hdr *entry_hdr,
 				uint32_t **d_ptr)
 {
@@ -1899,11 +2101,11 @@ static int qla4_83xx_minidump_pex_dma_read(struct scsi_qla_host *ha,
 		dma_desc.cmd.read_data_size = size;
 
 		/* Prepare: Write pex-dma descriptor to MS memory. */
-		rval = qla4_83xx_ms_mem_write_128b(ha,
+		rval = qla4_8xxx_ms_mem_write_128b(ha,
 			      (uint64_t)m_hdr->desc_card_addr,
 			      (uint32_t *)&dma_desc,
 			      (sizeof(struct qla4_83xx_pex_dma_descriptor)/16));
-		if (rval == -1) {
+		if (rval != QLA_SUCCESS) {
 			ql4_printk(KERN_INFO, ha,
 				   "%s: Error writing rdmem-dma-init to MS !!!\n",
 				   __func__);
@@ -2359,17 +2561,10 @@ static int qla4_8xxx_minidump_process_rdmem(struct scsi_qla_host *ha,
 	uint32_t *data_ptr = *d_ptr;
 	int rval = QLA_SUCCESS;
 
-	if (is_qla8032(ha) || is_qla8042(ha)) {
-		rval = qla4_83xx_minidump_pex_dma_read(ha, entry_hdr,
-						       &data_ptr);
-		if (rval != QLA_SUCCESS) {
-			rval = __qla4_8xxx_minidump_process_rdmem(ha, entry_hdr,
-								  &data_ptr);
-		}
-	} else {
+	rval = qla4_8xxx_minidump_pex_dma_read(ha, entry_hdr, &data_ptr);
+	if (rval != QLA_SUCCESS)
 		rval = __qla4_8xxx_minidump_process_rdmem(ha, entry_hdr,
 							  &data_ptr);
-	}
 	*d_ptr = data_ptr;
 	return rval;
 }
@@ -2440,6 +2635,227 @@ exit_process_pollrd:
 	return rval;
 }
 
+static uint32_t qla4_84xx_minidump_process_rddfe(struct scsi_qla_host *ha,
+				struct qla8xxx_minidump_entry_hdr *entry_hdr,
+				uint32_t **d_ptr)
+{
+	int loop_cnt;
+	uint32_t addr1, addr2, value, data, temp, wrval;
+	uint8_t stride, stride2;
+	uint16_t count;
+	uint32_t poll, mask, data_size, modify_mask;
+	uint32_t wait_count = 0;
+	uint32_t *data_ptr = *d_ptr;
+	struct qla8044_minidump_entry_rddfe *rddfe;
+	uint32_t rval = QLA_SUCCESS;
+
+	rddfe = (struct qla8044_minidump_entry_rddfe *)entry_hdr;
+	addr1 = le32_to_cpu(rddfe->addr_1);
+	value = le32_to_cpu(rddfe->value);
+	stride = le32_to_cpu(rddfe->stride);
+	stride2 = le32_to_cpu(rddfe->stride2);
+	count = le32_to_cpu(rddfe->count);
+
+	poll = le32_to_cpu(rddfe->poll);
+	mask = le32_to_cpu(rddfe->mask);
+	modify_mask = le32_to_cpu(rddfe->modify_mask);
+	data_size = le32_to_cpu(rddfe->data_size);
+
+	addr2 = addr1 + stride;
+
+	for (loop_cnt = 0x0; loop_cnt < count; loop_cnt++) {
+		ha->isp_ops->wr_reg_indirect(ha, addr1, (0x40000000 | value));
+
+		wait_count = 0;
+		while (wait_count < poll) {
+			ha->isp_ops->rd_reg_indirect(ha, addr1, &temp);
+			if ((temp & mask) != 0)
+				break;
+			wait_count++;
+		}
+
+		if (wait_count == poll) {
+			ql4_printk(KERN_ERR, ha, "%s: TIMEOUT\n", __func__);
+			rval = QLA_ERROR;
+			goto exit_process_rddfe;
+		} else {
+			ha->isp_ops->rd_reg_indirect(ha, addr2, &temp);
+			temp = temp & modify_mask;
+			temp = (temp | ((loop_cnt << 16) | loop_cnt));
+			wrval = ((temp << 16) | temp);
+
+			ha->isp_ops->wr_reg_indirect(ha, addr2, wrval);
+			ha->isp_ops->wr_reg_indirect(ha, addr1, value);
+
+			wait_count = 0;
+			while (wait_count < poll) {
+				ha->isp_ops->rd_reg_indirect(ha, addr1, &temp);
+				if ((temp & mask) != 0)
+					break;
+				wait_count++;
+			}
+			if (wait_count == poll) {
+				ql4_printk(KERN_ERR, ha, "%s: TIMEOUT\n",
+					   __func__);
+				rval = QLA_ERROR;
+				goto exit_process_rddfe;
+			}
+
+			ha->isp_ops->wr_reg_indirect(ha, addr1,
+						     ((0x40000000 | value) +
+						     stride2));
+			wait_count = 0;
+			while (wait_count < poll) {
+				ha->isp_ops->rd_reg_indirect(ha, addr1, &temp);
+				if ((temp & mask) != 0)
+					break;
+				wait_count++;
+			}
+
+			if (wait_count == poll) {
+				ql4_printk(KERN_ERR, ha, "%s: TIMEOUT\n",
+					   __func__);
+				rval = QLA_ERROR;
+				goto exit_process_rddfe;
+			}
+
+			ha->isp_ops->rd_reg_indirect(ha, addr2, &data);
+
+			*data_ptr++ = cpu_to_le32(wrval);
+			*data_ptr++ = cpu_to_le32(data);
+		}
+	}
+
+	*d_ptr = data_ptr;
+exit_process_rddfe:
+	return rval;
+}
+
+static uint32_t qla4_84xx_minidump_process_rdmdio(struct scsi_qla_host *ha,
+				struct qla8xxx_minidump_entry_hdr *entry_hdr,
+				uint32_t **d_ptr)
+{
+	int rval = QLA_SUCCESS;
+	uint32_t addr1, addr2, value1, value2, data, selval;
+	uint8_t stride1, stride2;
+	uint32_t addr3, addr4, addr5, addr6, addr7;
+	uint16_t count, loop_cnt;
+	uint32_t poll, mask;
+	uint32_t *data_ptr = *d_ptr;
+	struct qla8044_minidump_entry_rdmdio *rdmdio;
+
+	rdmdio = (struct qla8044_minidump_entry_rdmdio *)entry_hdr;
+	addr1 = le32_to_cpu(rdmdio->addr_1);
+	addr2 = le32_to_cpu(rdmdio->addr_2);
+	value1 = le32_to_cpu(rdmdio->value_1);
+	stride1 = le32_to_cpu(rdmdio->stride_1);
+	stride2 = le32_to_cpu(rdmdio->stride_2);
+	count = le32_to_cpu(rdmdio->count);
+
+	poll = le32_to_cpu(rdmdio->poll);
+	mask = le32_to_cpu(rdmdio->mask);
+	value2 = le32_to_cpu(rdmdio->value_2);
+
+	addr3 = addr1 + stride1;
+
+	for (loop_cnt = 0; loop_cnt < count; loop_cnt++) {
+		rval = ql4_84xx_poll_wait_ipmdio_bus_idle(ha, addr1, addr2,
+							 addr3, mask);
+		if (rval)
+			goto exit_process_rdmdio;
+
+		addr4 = addr2 - stride1;
+		rval = ql4_84xx_ipmdio_wr_reg(ha, addr1, addr3, mask, addr4,
+					     value2);
+		if (rval)
+			goto exit_process_rdmdio;
+
+		addr5 = addr2 - (2 * stride1);
+		rval = ql4_84xx_ipmdio_wr_reg(ha, addr1, addr3, mask, addr5,
+					     value1);
+		if (rval)
+			goto exit_process_rdmdio;
+
+		addr6 = addr2 - (3 * stride1);
+		rval = ql4_84xx_ipmdio_wr_reg(ha, addr1, addr3, mask,
+					     addr6, 0x2);
+		if (rval)
+			goto exit_process_rdmdio;
+
+		rval = ql4_84xx_poll_wait_ipmdio_bus_idle(ha, addr1, addr2,
+							 addr3, mask);
+		if (rval)
+			goto exit_process_rdmdio;
+
+		addr7 = addr2 - (4 * stride1);
+		rval = ql4_84xx_ipmdio_rd_reg(ha, addr1, addr3,
+						      mask, addr7, &data);
+		if (rval)
+			goto exit_process_rdmdio;
+
+		selval = (value2 << 18) | (value1 << 2) | 2;
+
+		stride2 = le32_to_cpu(rdmdio->stride_2);
+		*data_ptr++ = cpu_to_le32(selval);
+		*data_ptr++ = cpu_to_le32(data);
+
+		value1 = value1 + stride2;
+		*d_ptr = data_ptr;
+	}
+
+exit_process_rdmdio:
+	return rval;
+}
+
+static uint32_t qla4_84xx_minidump_process_pollwr(struct scsi_qla_host *ha,
+				struct qla8xxx_minidump_entry_hdr *entry_hdr,
+				uint32_t **d_ptr)
+{
+	uint32_t addr1, addr2, value1, value2, poll, mask, r_value;
+	struct qla8044_minidump_entry_pollwr *pollwr_hdr;
+	uint32_t wait_count = 0;
+	uint32_t rval = QLA_SUCCESS;
+
+	pollwr_hdr = (struct qla8044_minidump_entry_pollwr *)entry_hdr;
+	addr1 = le32_to_cpu(pollwr_hdr->addr_1);
+	addr2 = le32_to_cpu(pollwr_hdr->addr_2);
+	value1 = le32_to_cpu(pollwr_hdr->value_1);
+	value2 = le32_to_cpu(pollwr_hdr->value_2);
+
+	poll = le32_to_cpu(pollwr_hdr->poll);
+	mask = le32_to_cpu(pollwr_hdr->mask);
+
+	while (wait_count < poll) {
+		ha->isp_ops->rd_reg_indirect(ha, addr1, &r_value);
+
+		if ((r_value & poll) != 0)
+			break;
+
+		wait_count++;
+	}
+
+	if (wait_count == poll) {
+		ql4_printk(KERN_ERR, ha, "%s: TIMEOUT\n", __func__);
+		rval = QLA_ERROR;
+		goto exit_process_pollwr;
+	}
+
+	ha->isp_ops->wr_reg_indirect(ha, addr2, value2);
+	ha->isp_ops->wr_reg_indirect(ha, addr1, value1);
+
+	wait_count = 0;
+	while (wait_count < poll) {
+		ha->isp_ops->rd_reg_indirect(ha, addr1, &r_value);
+
+		if ((r_value & poll) != 0)
+			break;
+		wait_count++;
+	}
+
+exit_process_pollwr:
+	return rval;
+}
+
 static void qla83xx_minidump_process_rdmux2(struct scsi_qla_host *ha,
 				struct qla8xxx_minidump_entry_hdr *entry_hdr,
 				uint32_t **d_ptr)
@@ -2753,6 +3169,24 @@ static int qla4_8xxx_collect_md_data(struct scsi_qla_host *ha)
 			if (rval != QLA_SUCCESS)
 				qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i);
 			break;
+		case QLA8044_RDDFE:
+			rval = qla4_84xx_minidump_process_rddfe(ha, entry_hdr,
+								&data_ptr);
+			if (rval != QLA_SUCCESS)
+				qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i);
+			break;
+		case QLA8044_RDMDIO:
+			rval = qla4_84xx_minidump_process_rdmdio(ha, entry_hdr,
+								 &data_ptr);
+			if (rval != QLA_SUCCESS)
+				qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i);
+			break;
+		case QLA8044_POLLWR:
+			rval = qla4_84xx_minidump_process_pollwr(ha, entry_hdr,
+								 &data_ptr);
+			if (rval != QLA_SUCCESS)
+				qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i);
+			break;
 		case QLA8XXX_RDNOP:
 		default:
 			qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i);
diff --git a/drivers/scsi/qla4xxx/ql4_nx.h b/drivers/scsi/qla4xxx/ql4_nx.h
index 14500a0f62cc..337d9fcf6417 100644
--- a/drivers/scsi/qla4xxx/ql4_nx.h
+++ b/drivers/scsi/qla4xxx/ql4_nx.h
@@ -858,6 +858,9 @@ struct crb_addr_pair {
 #define QLA83XX_POLLRD	35
 #define QLA83XX_RDMUX2	36
 #define QLA83XX_POLLRDMWR  37
+#define QLA8044_RDDFE	38
+#define QLA8044_RDMDIO	39
+#define QLA8044_POLLWR	40
 #define QLA8XXX_RDROM	71
 #define QLA8XXX_RDMEM	72
 #define QLA8XXX_CNTRL	98
diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index 459b9f7186fd..320206376206 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -83,12 +83,12 @@ MODULE_PARM_DESC(ql4xsess_recovery_tmo,
 		" Target Session Recovery Timeout.\n"
 		"\t\t  Default: 120 sec.");
 
-int ql4xmdcapmask = 0x1F;
+int ql4xmdcapmask = 0;
 module_param(ql4xmdcapmask, int, S_IRUGO);
 MODULE_PARM_DESC(ql4xmdcapmask,
 		 " Set the Minidump driver capture mask level.\n"
-		 "\t\t  Default is 0x1F.\n"
-		 "\t\t  Can be set to 0x3, 0x7, 0xF, 0x1F, 0x3F, 0x7F");
+		 "\t\t  Default is 0 (firmware default capture mask)\n"
+		 "\t\t  Can be set to 0x3, 0x7, 0xF, 0x1F, 0x3F, 0x7F, 0xFF");
 
 int ql4xenablemd = 1;
 module_param(ql4xenablemd, int, S_IRUGO | S_IWUSR);
@@ -1742,6 +1742,9 @@ static int qla4xxx_get_ep_param(struct iscsi_endpoint *ep,
 	struct sockaddr *dst_addr;
 	struct scsi_qla_host *ha;
 
+	if (!qla_ep)
+		return -ENOTCONN;
+
 	ha = to_qla_host(qla_ep->host);
 	DEBUG2(ql4_printk(KERN_INFO, ha, "%s: host: %ld\n", __func__,
 			  ha->host_no));
@@ -1749,9 +1752,6 @@ static int qla4xxx_get_ep_param(struct iscsi_endpoint *ep,
 	switch (param) {
 	case ISCSI_PARAM_CONN_PORT:
 	case ISCSI_PARAM_CONN_ADDRESS:
-		if (!qla_ep)
-			return -ENOTCONN;
-
 		dst_addr = (struct sockaddr *)&qla_ep->dst_addr;
 		if (!dst_addr)
 			return -ENOTCONN;
@@ -2879,7 +2879,6 @@ static int qla4xxx_conn_get_param(struct iscsi_cls_conn *cls_conn,
 	struct iscsi_conn *conn;
 	struct qla_conn *qla_conn;
 	struct sockaddr *dst_addr;
-	int len = 0;
 
 	conn = cls_conn->dd_data;
 	qla_conn = conn->dd_data;
@@ -2893,9 +2892,6 @@ static int qla4xxx_conn_get_param(struct iscsi_cls_conn *cls_conn,
 	default:
 		return iscsi_conn_get_param(cls_conn, param, buf);
 	}
-
-	return len;
-
 }
 
 int qla4xxx_get_ddb_index(struct scsi_qla_host *ha, uint16_t *ddb_index)
@@ -3569,14 +3565,13 @@ static int qla4xxx_copy_from_fwddb_param(struct iscsi_bus_flash_session *sess,
 	if (test_bit(OPT_IPV6_DEVICE, &options)) {
 		conn->ipv6_traffic_class = fw_ddb_entry->ipv4_tos;
 
-		conn->link_local_ipv6_addr = kzalloc(IPv6_ADDR_LEN, GFP_KERNEL);
+		conn->link_local_ipv6_addr = kmemdup(
+					fw_ddb_entry->link_local_ipv6_addr,
+					IPv6_ADDR_LEN, GFP_KERNEL);
 		if (!conn->link_local_ipv6_addr) {
 			rc = -ENOMEM;
 			goto exit_copy;
 		}
-
-		memcpy(conn->link_local_ipv6_addr,
-		       fw_ddb_entry->link_local_ipv6_addr, IPv6_ADDR_LEN);
 	} else {
 		conn->ipv4_tos = fw_ddb_entry->ipv4_tos;
 	}
@@ -4565,6 +4560,7 @@ static void qla4xxx_timer(struct scsi_qla_host *ha)
 	     test_bit(DPC_LINK_CHANGED, &ha->dpc_flags) ||
 	     test_bit(DPC_HA_UNRECOVERABLE, &ha->dpc_flags) ||
 	     test_bit(DPC_HA_NEED_QUIESCENT, &ha->dpc_flags) ||
+	     test_bit(DPC_SYSFS_DDB_EXPORT, &ha->dpc_flags) ||
 	     test_bit(DPC_AEN, &ha->dpc_flags)) {
 		DEBUG2(printk("scsi%ld: %s: scheduling dpc routine"
 			      " - dpc flags = 0x%lx\n",
@@ -4862,9 +4858,6 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha)
 		    ha->host_no, __func__));
 		status = ha->isp_ops->reset_firmware(ha);
 		if (status == QLA_SUCCESS) {
-			if (!test_bit(AF_FW_RECOVERY, &ha->flags))
-				qla4xxx_cmd_wait(ha);
-
 			ha->isp_ops->disable_intrs(ha);
 			qla4xxx_process_aen(ha, FLUSH_DDB_CHANGED_AENS);
 			qla4xxx_abort_active_cmds(ha, DID_RESET << 16);
@@ -5432,6 +5425,11 @@ dpc_post_reset_ha:
 				qla4xxx_relogin_all_devices(ha);
 		}
 	}
+	if (test_and_clear_bit(DPC_SYSFS_DDB_EXPORT, &ha->dpc_flags)) {
+		if (qla4xxx_sysfs_ddb_export(ha))
+			ql4_printk(KERN_ERR, ha, "%s: Error exporting ddb to sysfs\n",
+				   __func__);
+	}
 }
 
 /**
@@ -8409,7 +8407,7 @@ exit_ddb_del:
  *
  * Export the firmware DDB for all send targets and normal targets to sysfs.
  **/
-static int qla4xxx_sysfs_ddb_export(struct scsi_qla_host *ha)
+int qla4xxx_sysfs_ddb_export(struct scsi_qla_host *ha)
 {
 	struct dev_db_entry *fw_ddb_entry = NULL;
 	dma_addr_t fw_ddb_entry_dma;
@@ -8847,11 +8845,8 @@ skip_retry_init:
 		ql4_printk(KERN_ERR, ha,
 			   "%s: No iSCSI boot target configured\n", __func__);
 
-	if (qla4xxx_sysfs_ddb_export(ha))
-		ql4_printk(KERN_ERR, ha,
-			   "%s: Error exporting ddb to sysfs\n", __func__);
-
-		/* Perform the build ddb list and login to each */
+	set_bit(DPC_SYSFS_DDB_EXPORT, &ha->dpc_flags);
+	/* Perform the build ddb list and login to each */
 	qla4xxx_build_ddb_list(ha, INIT_ADAPTER);
 	iscsi_host_for_each_session(ha->host, qla4xxx_login_flash_ddb);
 	qla4xxx_wait_login_resp_boot_tgt(ha);
diff --git a/drivers/scsi/qla4xxx/ql4_version.h b/drivers/scsi/qla4xxx/ql4_version.h
index c6ba0a6b8458..f11eaa773339 100644
--- a/drivers/scsi/qla4xxx/ql4_version.h
+++ b/drivers/scsi/qla4xxx/ql4_version.h
@@ -5,4 +5,4 @@
  * See LICENSE.qla4xxx for copyright and licensing details.
  */
 
-#define QLA4XXX_DRIVER_VERSION	"5.04.00-k4"
+#define QLA4XXX_DRIVER_VERSION	"5.04.00-k6"
diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index f3e9cc038d1d..1328a2621070 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -130,6 +130,7 @@ static const char * scsi_debug_version_date = "20100324";
 #define SCSI_DEBUG_OPT_DIF_ERR   32
 #define SCSI_DEBUG_OPT_DIX_ERR   64
 #define SCSI_DEBUG_OPT_MAC_TIMEOUT  128
+#define SCSI_DEBUG_OPT_SHORT_TRANSFER	256
 /* When "every_nth" > 0 then modulo "every_nth" commands:
  *   - a no response is simulated if SCSI_DEBUG_OPT_TIMEOUT is set
  *   - a RECOVERED_ERROR is simulated on successful read and write
@@ -3583,6 +3584,7 @@ int scsi_debug_queuecommand_lck(struct scsi_cmnd *SCpnt, done_funct_t done)
 	int inj_transport = 0;
 	int inj_dif = 0;
 	int inj_dix = 0;
+	int inj_short = 0;
 	int delay_override = 0;
 	int unmap = 0;
 
@@ -3628,6 +3630,8 @@ int scsi_debug_queuecommand_lck(struct scsi_cmnd *SCpnt, done_funct_t done)
 			inj_dif = 1; /* to reads and writes below */
 		else if (SCSI_DEBUG_OPT_DIX_ERR & scsi_debug_opts)
 			inj_dix = 1; /* to reads and writes below */
+		else if (SCSI_DEBUG_OPT_SHORT_TRANSFER & scsi_debug_opts)
+			inj_short = 1;
 	}
 
 	if (devip->wlun) {
@@ -3744,6 +3748,10 @@ read:
 		if (scsi_debug_fake_rw)
 			break;
 		get_data_transfer_info(cmd, &lba, &num, &ei_lba);
+
+		if (inj_short)
+			num /= 2;
+
 		errsts = resp_read(SCpnt, lba, num, devip, ei_lba);
 		if (inj_recovered && (0 == errsts)) {
 			mk_sense_buffer(devip, RECOVERED_ERROR,
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index f17aa7aa7879..7e957918f33f 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -131,7 +131,7 @@ scmd_eh_abort_handler(struct work_struct *work)
 				    "aborting command %p\n", scmd));
 		rtn = scsi_try_to_abort_cmd(sdev->host->hostt, scmd);
 		if (rtn == SUCCESS) {
-			scmd->result |= DID_TIME_OUT << 16;
+			set_host_byte(scmd, DID_TIME_OUT);
 			if (scsi_host_eh_past_deadline(sdev->host)) {
 				SCSI_LOG_ERROR_RECOVERY(3,
 					scmd_printk(KERN_INFO, scmd,
@@ -167,7 +167,7 @@ scmd_eh_abort_handler(struct work_struct *work)
 			scmd_printk(KERN_WARNING, scmd,
 				    "scmd %p terminate "
 				    "aborted command\n", scmd));
-		scmd->result |= DID_TIME_OUT << 16;
+		set_host_byte(scmd, DID_TIME_OUT);
 		scsi_finish_command(scmd);
 	}
 }
@@ -287,15 +287,15 @@ enum blk_eh_timer_return scsi_times_out(struct request *req)
 	else if (host->hostt->eh_timed_out)
 		rtn = host->hostt->eh_timed_out(scmd);
 
-	if (rtn == BLK_EH_NOT_HANDLED && !host->hostt->no_async_abort)
-		if (scsi_abort_command(scmd) == SUCCESS)
+	if (rtn == BLK_EH_NOT_HANDLED) {
+		if (!host->hostt->no_async_abort &&
+		    scsi_abort_command(scmd) == SUCCESS)
 			return BLK_EH_NOT_HANDLED;
 
-	scmd->result |= DID_TIME_OUT << 16;
-
-	if (unlikely(rtn == BLK_EH_NOT_HANDLED &&
-		     !scsi_eh_scmd_add(scmd, SCSI_EH_CANCEL_CMD)))
-		rtn = BLK_EH_HANDLED;
+		set_host_byte(scmd, DID_TIME_OUT);
+		if (!scsi_eh_scmd_add(scmd, SCSI_EH_CANCEL_CMD))
+			rtn = BLK_EH_HANDLED;
+	}
 
 	return rtn;
 }
@@ -1029,6 +1029,7 @@ retry:
 		rtn = NEEDS_RETRY;
 	} else {
 		timeleft = wait_for_completion_timeout(&done, timeout);
+		rtn = SUCCESS;
 	}
 
 	shost->eh_action = NULL;
@@ -1776,7 +1777,7 @@ int scsi_decide_disposition(struct scsi_cmnd *scmd)
 		break;
 	case DID_ABORT:
 		if (scmd->eh_eflags & SCSI_EH_ABORT_SCHEDULED) {
-			scmd->result |= DID_TIME_OUT << 16;
+			set_host_byte(scmd, DID_TIME_OUT);
 			return SUCCESS;
 		}
 	case DID_NO_CONNECT:
@@ -1951,6 +1952,8 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 	 */
 	req = blk_get_request(sdev->request_queue, READ, GFP_KERNEL);
 
+	blk_rq_set_block_pc(req);
+
 	req->cmd[0] = ALLOW_MEDIUM_REMOVAL;
 	req->cmd[1] = 0;
 	req->cmd[2] = 0;
@@ -1960,7 +1963,6 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 
 	req->cmd_len = COMMAND_SIZE(req->cmd[0]);
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= REQ_QUIET;
 	req->timeout = 10 * HZ;
 	req->retries = 5;
@@ -2306,6 +2308,12 @@ scsi_reset_provider(struct scsi_device *dev, int flag)
 	}
 
 	scmd = scsi_get_command(dev, GFP_KERNEL);
+	if (!scmd) {
+		rtn = FAILED;
+		put_device(&dev->sdev_gendev);
+		goto out_put_autopm_host;
+	}
+
 	blk_rq_init(NULL, &req);
 	scmd->request = &req;
 
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index a0c95cac91f0..f7e316368c99 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -195,6 +195,7 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	req = blk_get_request(sdev->request_queue, write, __GFP_WAIT);
 	if (!req)
 		return ret;
+	blk_rq_set_block_pc(req);
 
 	if (bufflen &&	blk_rq_map_kern(sdev->request_queue, req,
 					buffer, bufflen, __GFP_WAIT))
@@ -206,7 +207,6 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	req->sense_len = 0;
 	req->retries = retries;
 	req->timeout = timeout;
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= flags | REQ_QUIET | REQ_PREEMPT;
 
 	/*
@@ -512,68 +512,6 @@ void scsi_run_host_queues(struct Scsi_Host *shost)
 		scsi_run_queue(sdev->request_queue);
 }
 
-static void __scsi_release_buffers(struct scsi_cmnd *, int);
-
-/*
- * Function:    scsi_end_request()
- *
- * Purpose:     Post-processing of completed commands (usually invoked at end
- *		of upper level post-processing and scsi_io_completion).
- *
- * Arguments:   cmd	 - command that is complete.
- *              error    - 0 if I/O indicates success, < 0 for I/O error.
- *              bytes    - number of bytes of completed I/O
- *		requeue  - indicates whether we should requeue leftovers.
- *
- * Lock status: Assumed that lock is not held upon entry.
- *
- * Returns:     cmd if requeue required, NULL otherwise.
- *
- * Notes:       This is called for block device requests in order to
- *              mark some number of sectors as complete.
- * 
- *		We are guaranteeing that the request queue will be goosed
- *		at some point during this call.
- * Notes:	If cmd was requeued, upon return it will be a stale pointer.
- */
-static struct scsi_cmnd *scsi_end_request(struct scsi_cmnd *cmd, int error,
-					  int bytes, int requeue)
-{
-	struct request_queue *q = cmd->device->request_queue;
-	struct request *req = cmd->request;
-
-	/*
-	 * If there are blocks left over at the end, set up the command
-	 * to queue the remainder of them.
-	 */
-	if (blk_end_request(req, error, bytes)) {
-		/* kill remainder if no retrys */
-		if (error && scsi_noretry_cmd(cmd))
-			blk_end_request_all(req, error);
-		else {
-			if (requeue) {
-				/*
-				 * Bleah.  Leftovers again.  Stick the
-				 * leftovers in the front of the
-				 * queue, and goose the queue again.
-				 */
-				scsi_release_buffers(cmd);
-				scsi_requeue_command(q, cmd);
-				cmd = NULL;
-			}
-			return cmd;
-		}
-	}
-
-	/*
-	 * This will goose the queue request function at the end, so we don't
-	 * need to worry about launching another command.
-	 */
-	__scsi_release_buffers(cmd, 0);
-	scsi_next_command(cmd);
-	return NULL;
-}
-
 static inline unsigned int scsi_sgtable_index(unsigned short nents)
 {
 	unsigned int index;
@@ -625,30 +563,10 @@ static void scsi_free_sgtable(struct scsi_data_buffer *sdb)
 	__sg_free_table(&sdb->table, SCSI_MAX_SG_SEGMENTS, scsi_sg_free);
 }
 
-static void __scsi_release_buffers(struct scsi_cmnd *cmd, int do_bidi_check)
-{
-
-	if (cmd->sdb.table.nents)
-		scsi_free_sgtable(&cmd->sdb);
-
-	memset(&cmd->sdb, 0, sizeof(cmd->sdb));
-
-	if (do_bidi_check && scsi_bidi_cmnd(cmd)) {
-		struct scsi_data_buffer *bidi_sdb =
-			cmd->request->next_rq->special;
-		scsi_free_sgtable(bidi_sdb);
-		kmem_cache_free(scsi_sdb_cache, bidi_sdb);
-		cmd->request->next_rq->special = NULL;
-	}
-
-	if (scsi_prot_sg_count(cmd))
-		scsi_free_sgtable(cmd->prot_sdb);
-}
-
 /*
  * Function:    scsi_release_buffers()
  *
- * Purpose:     Completion processing for block device I/O requests.
+ * Purpose:     Free resources allocate for a scsi_command.
  *
  * Arguments:   cmd	- command that we are bailing.
  *
@@ -659,15 +577,29 @@ static void __scsi_release_buffers(struct scsi_cmnd *cmd, int do_bidi_check)
  * Notes:       In the event that an upper level driver rejects a
  *		command, we must release resources allocated during
  *		the __init_io() function.  Primarily this would involve
- *		the scatter-gather table, and potentially any bounce
- *		buffers.
+ *		the scatter-gather table.
  */
 void scsi_release_buffers(struct scsi_cmnd *cmd)
 {
-	__scsi_release_buffers(cmd, 1);
+	if (cmd->sdb.table.nents)
+		scsi_free_sgtable(&cmd->sdb);
+
+	memset(&cmd->sdb, 0, sizeof(cmd->sdb));
+
+	if (scsi_prot_sg_count(cmd))
+		scsi_free_sgtable(cmd->prot_sdb);
 }
 EXPORT_SYMBOL(scsi_release_buffers);
 
+static void scsi_release_bidi_buffers(struct scsi_cmnd *cmd)
+{
+	struct scsi_data_buffer *bidi_sdb = cmd->request->next_rq->special;
+
+	scsi_free_sgtable(bidi_sdb);
+	kmem_cache_free(scsi_sdb_cache, bidi_sdb);
+	cmd->request->next_rq->special = NULL;
+}
+
 /**
  * __scsi_error_from_host_byte - translate SCSI error code into errno
  * @cmd:	SCSI command (unused)
@@ -725,16 +657,9 @@ static int __scsi_error_from_host_byte(struct scsi_cmnd *cmd, int result)
  *
  * Returns:     Nothing
  *
- * Notes:       This function is matched in terms of capabilities to
- *              the function that created the scatter-gather list.
- *              In other words, if there are no bounce buffers
- *              (the normal case for most drivers), we don't need
- *              the logic to deal with cleaning up afterwards.
- *
- *		We must call scsi_end_request().  This will finish off
- *		the specified number of sectors.  If we are done, the
- *		command block will be released and the queue function
- *		will be goosed.  If we are not done then we have to
+ * Notes:       We will finish off the specified number of sectors.  If we
+ *		are done, the command block will be released and the queue
+ *		function will be goosed.  If we are not done then we have to
  *		figure out what to do next:
  *
  *		a) We can call scsi_requeue_command().  The request
@@ -743,7 +668,7 @@ static int __scsi_error_from_host_byte(struct scsi_cmnd *cmd, int result)
  *		   be used if we made forward progress, or if we want
  *		   to switch from READ(10) to READ(6) for example.
  *
- *		b) We can call scsi_queue_insert().  The request will
+ *		b) We can call __scsi_queue_insert().  The request will
  *		   be put back on the queue and retried using the same
  *		   command as before, possibly after a delay.
  *
@@ -801,6 +726,8 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 			req->next_rq->resid_len = scsi_in(cmd)->resid;
 
 			scsi_release_buffers(cmd);
+			scsi_release_bidi_buffers(cmd);
+
 			blk_end_request_all(req, 0);
 
 			scsi_next_command(cmd);
@@ -840,12 +767,25 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 	}
 
 	/*
-	 * A number of bytes were successfully read.  If there
-	 * are leftovers and there is some kind of error
-	 * (result != 0), retry the rest.
+	 * If we finished all bytes in the request we are done now.
 	 */
-	if (scsi_end_request(cmd, error, good_bytes, result == 0) == NULL)
-		return;
+	if (!blk_end_request(req, error, good_bytes))
+		goto next_command;
+
+	/*
+	 * Kill remainder if no retrys.
+	 */
+	if (error && scsi_noretry_cmd(cmd)) {
+		blk_end_request_all(req, error);
+		goto next_command;
+	}
+
+	/*
+	 * If there had been no error, but we have leftover bytes in the
+	 * requeues just queue the command up again.
+	 */
+	if (result == 0)
+		goto requeue;
 
 	error = __scsi_error_from_host_byte(cmd, result);
 
@@ -973,7 +913,6 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 	switch (action) {
 	case ACTION_FAIL:
 		/* Give up and fail the remainder of the request */
-		scsi_release_buffers(cmd);
 		if (!(req->cmd_flags & REQ_QUIET)) {
 			if (description)
 				scmd_printk(KERN_INFO, cmd, "%s\n",
@@ -983,12 +922,11 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 				scsi_print_sense("", cmd);
 			scsi_print_command(cmd);
 		}
-		if (blk_end_request_err(req, error))
-			scsi_requeue_command(q, cmd);
-		else
-			scsi_next_command(cmd);
-		break;
+		if (!blk_end_request_err(req, error))
+			goto next_command;
+		/*FALLTHRU*/
 	case ACTION_REPREP:
+	requeue:
 		/* Unprep the request and put it back at the head of the queue.
 		 * A new command will be prepared and issued.
 		 */
@@ -1004,6 +942,11 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 		__scsi_queue_insert(cmd, SCSI_MLQUEUE_DEVICE_BUSY, 0);
 		break;
 	}
+	return;
+
+next_command:
+	scsi_release_buffers(cmd);
+	scsi_next_command(cmd);
 }
 
 static int scsi_init_sgtable(struct request *req, struct scsi_data_buffer *sdb,
@@ -1128,15 +1071,7 @@ static struct scsi_cmnd *scsi_get_cmd_from_req(struct scsi_device *sdev,
 
 int scsi_setup_blk_pc_cmnd(struct scsi_device *sdev, struct request *req)
 {
-	struct scsi_cmnd *cmd;
-	int ret = scsi_prep_state_check(sdev, req);
-
-	if (ret != BLKPREP_OK)
-		return ret;
-
-	cmd = scsi_get_cmd_from_req(sdev, req);
-	if (unlikely(!cmd))
-		return BLKPREP_DEFER;
+	struct scsi_cmnd *cmd = req->special;
 
 	/*
 	 * BLOCK_PC requests may transfer data, in which case they must
@@ -1179,15 +1114,11 @@ EXPORT_SYMBOL(scsi_setup_blk_pc_cmnd);
  */
 int scsi_setup_fs_cmnd(struct scsi_device *sdev, struct request *req)
 {
-	struct scsi_cmnd *cmd;
-	int ret = scsi_prep_state_check(sdev, req);
-
-	if (ret != BLKPREP_OK)
-		return ret;
+	struct scsi_cmnd *cmd = req->special;
 
 	if (unlikely(sdev->scsi_dh_data && sdev->scsi_dh_data->scsi_dh
 			 && sdev->scsi_dh_data->scsi_dh->prep_fn)) {
-		ret = sdev->scsi_dh_data->scsi_dh->prep_fn(sdev, req);
+		int ret = sdev->scsi_dh_data->scsi_dh->prep_fn(sdev, req);
 		if (ret != BLKPREP_OK)
 			return ret;
 	}
@@ -1197,16 +1128,13 @@ int scsi_setup_fs_cmnd(struct scsi_device *sdev, struct request *req)
 	 */
 	BUG_ON(!req->nr_phys_segments);
 
-	cmd = scsi_get_cmd_from_req(sdev, req);
-	if (unlikely(!cmd))
-		return BLKPREP_DEFER;
-
 	memset(cmd->cmnd, 0, BLK_MAX_CDB);
 	return scsi_init_io(cmd, GFP_ATOMIC);
 }
 EXPORT_SYMBOL(scsi_setup_fs_cmnd);
 
-int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
+static int
+scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
 {
 	int ret = BLKPREP_OK;
 
@@ -1258,9 +1186,9 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
 	}
 	return ret;
 }
-EXPORT_SYMBOL(scsi_prep_state_check);
 
-int scsi_prep_return(struct request_queue *q, struct request *req, int ret)
+static int
+scsi_prep_return(struct request_queue *q, struct request *req, int ret)
 {
 	struct scsi_device *sdev = q->queuedata;
 
@@ -1291,18 +1219,44 @@ int scsi_prep_return(struct request_queue *q, struct request *req, int ret)
 
 	return ret;
 }
-EXPORT_SYMBOL(scsi_prep_return);
 
-int scsi_prep_fn(struct request_queue *q, struct request *req)
+static int scsi_prep_fn(struct request_queue *q, struct request *req)
 {
 	struct scsi_device *sdev = q->queuedata;
-	int ret = BLKPREP_KILL;
+	struct scsi_cmnd *cmd;
+	int ret;
 
-	if (req->cmd_type == REQ_TYPE_BLOCK_PC)
+	ret = scsi_prep_state_check(sdev, req);
+	if (ret != BLKPREP_OK)
+		goto out;
+
+	cmd = scsi_get_cmd_from_req(sdev, req);
+	if (unlikely(!cmd)) {
+		ret = BLKPREP_DEFER;
+		goto out;
+	}
+
+	if (req->cmd_type == REQ_TYPE_FS)
+		ret = scsi_cmd_to_driver(cmd)->init_command(cmd);
+	else if (req->cmd_type == REQ_TYPE_BLOCK_PC)
 		ret = scsi_setup_blk_pc_cmnd(sdev, req);
+	else
+		ret = BLKPREP_KILL;
+
+out:
 	return scsi_prep_return(q, req, ret);
 }
-EXPORT_SYMBOL(scsi_prep_fn);
+
+static void scsi_unprep_fn(struct request_queue *q, struct request *req)
+{
+	if (req->cmd_type == REQ_TYPE_FS) {
+		struct scsi_cmnd *cmd = req->special;
+		struct scsi_driver *drv = scsi_cmd_to_driver(cmd);
+
+		if (drv->uninit_command)
+			drv->uninit_command(cmd);
+	}
+}
 
 /*
  * scsi_dev_queue_ready: if we can send requests to sdev, return 1 else
@@ -1723,6 +1677,7 @@ struct request_queue *scsi_alloc_queue(struct scsi_device *sdev)
 		return NULL;
 
 	blk_queue_prep_rq(q, scsi_prep_fn);
+	blk_queue_unprep_rq(q, scsi_unprep_fn);
 	blk_queue_softirq_done(q, scsi_softirq_done);
 	blk_queue_rq_timed_out(q, scsi_times_out);
 	blk_queue_lld_busy(q, scsi_lld_busy);
diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index f80908f74ca9..521f5838594b 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -2549,6 +2549,7 @@ fc_rport_final_delete(struct work_struct *work)
 			fc_flush_devloss(shost);
 		if (!cancel_delayed_work(&rport->dev_loss_work))
 			fc_flush_devloss(shost);
+		cancel_work_sync(&rport->scan_work);
 		spin_lock_irqsave(shost->host_lock, flags);
 		rport->flags &= ~FC_RPORT_DEVLOSS_PENDING;
 	}
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 96af195224f2..6825eda1114a 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -109,6 +109,8 @@ static int sd_suspend_system(struct device *);
 static int sd_suspend_runtime(struct device *);
 static int sd_resume(struct device *);
 static void sd_rescan(struct device *);
+static int sd_init_command(struct scsi_cmnd *SCpnt);
+static void sd_uninit_command(struct scsi_cmnd *SCpnt);
 static int sd_done(struct scsi_cmnd *);
 static int sd_eh_action(struct scsi_cmnd *, int);
 static void sd_read_capacity(struct scsi_disk *sdkp, unsigned char *buffer);
@@ -503,6 +505,8 @@ static struct scsi_driver sd_template = {
 		.pm		= &sd_pm_ops,
 	},
 	.rescan			= sd_rescan,
+	.init_command		= sd_init_command,
+	.uninit_command		= sd_uninit_command,
 	.done			= sd_done,
 	.eh_action		= sd_eh_action,
 };
@@ -836,9 +840,9 @@ static int scsi_setup_flush_cmnd(struct scsi_device *sdp, struct request *rq)
 	return scsi_setup_blk_pc_cmnd(sdp, rq);
 }
 
-static void sd_unprep_fn(struct request_queue *q, struct request *rq)
+static void sd_uninit_command(struct scsi_cmnd *SCpnt)
 {
-	struct scsi_cmnd *SCpnt = rq->special;
+	struct request *rq = SCpnt->request;
 
 	if (rq->cmd_flags & REQ_DISCARD)
 		__free_page(rq->completion_data);
@@ -850,18 +854,10 @@ static void sd_unprep_fn(struct request_queue *q, struct request *rq)
 	}
 }
 
-/**
- *	sd_prep_fn - build a scsi (read or write) command from
- *	information in the request structure.
- *	@SCpnt: pointer to mid-level's per scsi command structure that
- *	contains request and into which the scsi command is written
- *
- *	Returns 1 if successful and 0 if error (or cannot be done now).
- **/
-static int sd_prep_fn(struct request_queue *q, struct request *rq)
+static int sd_init_command(struct scsi_cmnd *SCpnt)
 {
-	struct scsi_cmnd *SCpnt;
-	struct scsi_device *sdp = q->queuedata;
+	struct request *rq = SCpnt->request;
+	struct scsi_device *sdp = SCpnt->device;
 	struct gendisk *disk = rq->rq_disk;
 	struct scsi_disk *sdkp;
 	sector_t block = blk_rq_pos(rq);
@@ -883,12 +879,6 @@ static int sd_prep_fn(struct request_queue *q, struct request *rq)
 	} else if (rq->cmd_flags & REQ_FLUSH) {
 		ret = scsi_setup_flush_cmnd(sdp, rq);
 		goto out;
-	} else if (rq->cmd_type == REQ_TYPE_BLOCK_PC) {
-		ret = scsi_setup_blk_pc_cmnd(sdp, rq);
-		goto out;
-	} else if (rq->cmd_type != REQ_TYPE_FS) {
-		ret = BLKPREP_KILL;
-		goto out;
 	}
 	ret = scsi_setup_fs_cmnd(sdp, rq);
 	if (ret != BLKPREP_OK)
@@ -900,11 +890,10 @@ static int sd_prep_fn(struct request_queue *q, struct request *rq)
 	 * is used for a killable error condition */
 	ret = BLKPREP_KILL;
 
-	SCSI_LOG_HLQUEUE(1, scmd_printk(KERN_INFO, SCpnt,
-					"sd_prep_fn: block=%llu, "
-					"count=%d\n",
-					(unsigned long long)block,
-					this_count));
+	SCSI_LOG_HLQUEUE(1,
+		scmd_printk(KERN_INFO, SCpnt,
+			"%s: block=%llu, count=%d\n",
+			__func__, (unsigned long long)block, this_count));
 
 	if (!sdp || !scsi_device_online(sdp) ||
 	    block + blk_rq_sectors(rq) > get_capacity(disk)) {
@@ -1124,7 +1113,7 @@ static int sd_prep_fn(struct request_queue *q, struct request *rq)
 	 */
 	ret = BLKPREP_OK;
  out:
-	return scsi_prep_return(q, rq, ret);
+	return ret;
 }
 
 /**
@@ -1686,12 +1675,12 @@ static int sd_done(struct scsi_cmnd *SCpnt)
 						   sshdr.ascq));
 	}
 #endif
+	sdkp->medium_access_timed_out = 0;
+
 	if (driver_byte(result) != DRIVER_SENSE &&
 	    (!sense_valid || sense_deferred))
 		goto out;
 
-	sdkp->medium_access_timed_out = 0;
-
 	switch (sshdr.sense_key) {
 	case HARDWARE_ERROR:
 	case MEDIUM_ERROR:
@@ -2452,7 +2441,10 @@ sd_read_cache_type(struct scsi_disk *sdkp, unsigned char *buffer)
 		}
 
 		sdkp->DPOFUA = (data.device_specific & 0x10) != 0;
-		if (sdkp->DPOFUA && !sdkp->device->use_10_for_rw) {
+		if (sdp->broken_fua) {
+			sd_first_printk(KERN_NOTICE, sdkp, "Disabling FUA\n");
+			sdkp->DPOFUA = 0;
+		} else if (sdkp->DPOFUA && !sdkp->device->use_10_for_rw) {
 			sd_first_printk(KERN_NOTICE, sdkp,
 				  "Uses READ/WRITE(6), disabling FUA\n");
 			sdkp->DPOFUA = 0;
@@ -2875,9 +2867,6 @@ static void sd_probe_async(void *data, async_cookie_t cookie)
 
 	sd_revalidate_disk(gd);
 
-	blk_queue_prep_rq(sdp->request_queue, sd_prep_fn);
-	blk_queue_unprep_rq(sdp->request_queue, sd_unprep_fn);
-
 	gd->driverfs_dev = &sdp->sdev_gendev;
 	gd->flags = GENHD_FL_EXT_DEVT;
 	if (sdp->removable) {
@@ -3025,8 +3014,6 @@ static int sd_remove(struct device *dev)
 
 	async_synchronize_full_domain(&scsi_sd_pm_domain);
 	async_synchronize_full_domain(&scsi_sd_probe_domain);
-	blk_queue_prep_rq(sdkp->device->request_queue, scsi_prep_fn);
-	blk_queue_unprep_rq(sdkp->device->request_queue, NULL);
 	device_del(&sdkp->dev);
 	del_gendisk(sdkp->disk);
 	sd_shutdown(dev);
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index df5e961484e1..53268aaba559 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -1653,10 +1653,9 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd)
 	if (!rq)
 		return -ENOMEM;
 
+	blk_rq_set_block_pc(rq);
 	memcpy(rq->cmd, cmd, hp->cmd_len);
-
 	rq->cmd_len = hp->cmd_len;
-	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 
 	srp->rq = rq;
 	rq->end_io_data = srp;
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index 40d85929aefe..93cbd36c990b 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -79,6 +79,7 @@ MODULE_ALIAS_SCSI_DEVICE(TYPE_WORM);
 static DEFINE_MUTEX(sr_mutex);
 static int sr_probe(struct device *);
 static int sr_remove(struct device *);
+static int sr_init_command(struct scsi_cmnd *SCpnt);
 static int sr_done(struct scsi_cmnd *);
 static int sr_runtime_suspend(struct device *dev);
 
@@ -94,6 +95,7 @@ static struct scsi_driver sr_template = {
 		.remove		= sr_remove,
 		.pm		= &sr_pm_ops,
 	},
+	.init_command		= sr_init_command,
 	.done			= sr_done,
 };
 
@@ -378,21 +380,14 @@ static int sr_done(struct scsi_cmnd *SCpnt)
 	return good_bytes;
 }
 
-static int sr_prep_fn(struct request_queue *q, struct request *rq)
+static int sr_init_command(struct scsi_cmnd *SCpnt)
 {
 	int block = 0, this_count, s_size;
 	struct scsi_cd *cd;
-	struct scsi_cmnd *SCpnt;
-	struct scsi_device *sdp = q->queuedata;
+	struct request *rq = SCpnt->request;
+	struct scsi_device *sdp = SCpnt->device;
 	int ret;
 
-	if (rq->cmd_type == REQ_TYPE_BLOCK_PC) {
-		ret = scsi_setup_blk_pc_cmnd(sdp, rq);
-		goto out;
-	} else if (rq->cmd_type != REQ_TYPE_FS) {
-		ret = BLKPREP_KILL;
-		goto out;
-	}
 	ret = scsi_setup_fs_cmnd(sdp, rq);
 	if (ret != BLKPREP_OK)
 		goto out;
@@ -517,7 +512,7 @@ static int sr_prep_fn(struct request_queue *q, struct request *rq)
 	 */
 	ret = BLKPREP_OK;
  out:
-	return scsi_prep_return(q, rq, ret);
+	return ret;
 }
 
 static int sr_block_open(struct block_device *bdev, fmode_t mode)
@@ -718,7 +713,6 @@ static int sr_probe(struct device *dev)
 
 	/* FIXME: need to handle a get_capabilities failure properly ?? */
 	get_capabilities(cd);
-	blk_queue_prep_rq(sdev->request_queue, sr_prep_fn);
 	sr_vendor_init(cd);
 
 	disk->driverfs_dev = &sdev->sdev_gendev;
@@ -993,7 +987,6 @@ static int sr_remove(struct device *dev)
 
 	scsi_autopm_get_device(cd->device);
 
-	blk_queue_prep_rq(cd->device->request_queue, scsi_prep_fn);
 	del_gendisk(cd->disk);
 
 	mutex_lock(&sr_ref_mutex);
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index afc834e172c6..14eb4b256a03 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -484,7 +484,7 @@ static int st_scsi_execute(struct st_request *SRpnt, const unsigned char *cmd,
 	if (!req)
 		return DRIVER_ERROR << 24;
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	blk_rq_set_block_pc(req);
 	req->cmd_flags |= REQ_QUIET;
 
 	mdata->null_mapped = 1;
diff --git a/drivers/scsi/sun3_NCR5380.c b/drivers/scsi/sun3_NCR5380.c
index 636bbe0ea84c..88220794cc98 100644
--- a/drivers/scsi/sun3_NCR5380.c
+++ b/drivers/scsi/sun3_NCR5380.c
@@ -364,7 +364,7 @@ static int is_lun_busy(struct scsi_cmnd *cmd, int should_be_tagged)
 	return( 0 );
     if (TagAlloc[cmd->device->id][cmd->device->lun].nr_allocated >=
 	TagAlloc[cmd->device->id][cmd->device->lun].queue_size ) {
-	TAG_PRINTK( "scsi%d: target %d lun %d: no free tags\n",
+	dprintk(NDEBUG_TAGS,  "scsi%d: target %d lun %d: no free tags\n",
 		    H_NO(cmd), cmd->device->id, cmd->device->lun );
 	return( 1 );
     }
@@ -388,7 +388,7 @@ static void cmd_get_tag(struct scsi_cmnd *cmd, int should_be_tagged)
 	!setup_use_tagged_queuing || !cmd->device->tagged_supported) {
 	cmd->tag = TAG_NONE;
 	hostdata->busy[cmd->device->id] |= (1 << cmd->device->lun);
-	TAG_PRINTK( "scsi%d: target %d lun %d now allocated by untagged "
+	dprintk(NDEBUG_TAGS,  "scsi%d: target %d lun %d now allocated by untagged "
 		    "command\n", H_NO(cmd), cmd->device->id, cmd->device->lun );
     }
     else {
@@ -397,7 +397,7 @@ static void cmd_get_tag(struct scsi_cmnd *cmd, int should_be_tagged)
 	cmd->tag = find_first_zero_bit( &ta->allocated, MAX_TAGS );
 	set_bit( cmd->tag, &ta->allocated );
 	ta->nr_allocated++;
-	TAG_PRINTK( "scsi%d: using tag %d for target %d lun %d "
+	dprintk(NDEBUG_TAGS,  "scsi%d: using tag %d for target %d lun %d "
 		    "(now %d tags in use)\n",
 		    H_NO(cmd), cmd->tag, cmd->device->id, cmd->device->lun,
 		    ta->nr_allocated );
@@ -415,7 +415,7 @@ static void cmd_free_tag(struct scsi_cmnd *cmd)
 
     if (cmd->tag == TAG_NONE) {
 	hostdata->busy[cmd->device->id] &= ~(1 << cmd->device->lun);
-	TAG_PRINTK( "scsi%d: target %d lun %d untagged cmd finished\n",
+	dprintk(NDEBUG_TAGS,  "scsi%d: target %d lun %d untagged cmd finished\n",
 		    H_NO(cmd), cmd->device->id, cmd->device->lun );
     }
     else if (cmd->tag >= MAX_TAGS) {
@@ -426,7 +426,7 @@ static void cmd_free_tag(struct scsi_cmnd *cmd)
 	TAG_ALLOC *ta = &TagAlloc[cmd->device->id][cmd->device->lun];
 	clear_bit( cmd->tag, &ta->allocated );
 	ta->nr_allocated--;
-	TAG_PRINTK( "scsi%d: freed tag %d for target %d lun %d\n",
+	dprintk(NDEBUG_TAGS,  "scsi%d: freed tag %d for target %d lun %d\n",
 		    H_NO(cmd), cmd->tag, cmd->device->id, cmd->device->lun );
     }
 }
@@ -484,7 +484,7 @@ static __inline__ void initialize_SCp(struct scsi_cmnd *cmd)
 
 #include <linux/delay.h>
 
-#if 1
+#if NDEBUG
 static struct {
     unsigned char mask;
     const char * name;} 
@@ -572,12 +572,6 @@ static void NCR5380_print_phase(struct Scsi_Host *instance)
     }
 }
 
-#else /* !NDEBUG */
-
-/* dummies... */
-__inline__ void NCR5380_print(struct Scsi_Host *instance) { };
-__inline__ void NCR5380_print_phase(struct Scsi_Host *instance) { };
-
 #endif
 
 /*
@@ -618,7 +612,7 @@ static inline void NCR5380_all_init (void)
 {
     static int done = 0;
     if (!done) {
-	INI_PRINTK("scsi : NCR5380_all_init()\n");
+	dprintk(NDEBUG_INIT, "scsi : NCR5380_all_init()\n");
 	done = 1;
     }
 }
@@ -681,8 +675,8 @@ static void NCR5380_print_status(struct Scsi_Host *instance)
 	Scsi_Cmnd *ptr;
 	unsigned long flags;
 
-	NCR_PRINT(NDEBUG_ANY);
-	NCR_PRINT_PHASE(NDEBUG_ANY);
+	NCR5380_dprint(NDEBUG_ANY, instance);
+	NCR5380_dprint_phase(NDEBUG_ANY, instance);
 
 	hostdata = (struct NCR5380_hostdata *)instance->hostdata;
 
@@ -928,7 +922,7 @@ static int NCR5380_queue_command_lck(struct scsi_cmnd *cmd,
 
     local_irq_restore(flags);
 
-    QU_PRINTK("scsi%d: command added to %s of queue\n", H_NO(cmd),
+    dprintk(NDEBUG_QUEUES, "scsi%d: command added to %s of queue\n", H_NO(cmd),
 	      (cmd->cmnd[0] == REQUEST_SENSE) ? "head" : "tail");
 
     /* If queue_command() is called from an interrupt (real one or bottom
@@ -998,7 +992,7 @@ static void NCR5380_main (struct work_struct *bl)
 	done = 1;
 	
 	if (!hostdata->connected) {
-	    MAIN_PRINTK( "scsi%d: not connected\n", HOSTNO );
+	    dprintk(NDEBUG_MAIN,  "scsi%d: not connected\n", HOSTNO );
 	    /*
 	     * Search through the issue_queue for a command destined
 	     * for a target that's not busy.
@@ -1012,12 +1006,8 @@ static void NCR5380_main (struct work_struct *bl)
 	    for (tmp = (struct scsi_cmnd *) hostdata->issue_queue,
 		 prev = NULL; tmp; prev = tmp, tmp = NEXT(tmp) ) {
 
-#if (NDEBUG & NDEBUG_LISTS)
 		if (prev != tmp)
-		    printk("MAIN tmp=%p   target=%d   busy=%d lun=%d\n",
-			   tmp, tmp->target, hostdata->busy[tmp->target],
-			   tmp->lun);
-#endif
+			dprintk(NDEBUG_LISTS, "MAIN tmp=%p   target=%d   busy=%d lun=%d\n", tmp, tmp->device->id, hostdata->busy[tmp->device->id], tmp->device->lun);
 		/*  When we find one, remove it from the issue queue. */
 		/* ++guenther: possible race with Falcon locking */
 		if (
@@ -1047,9 +1037,9 @@ static void NCR5380_main (struct work_struct *bl)
 		     * On failure, we must add the command back to the
 		     *   issue queue so we can keep trying.	
 		     */
-		    MAIN_PRINTK("scsi%d: main(): command for target %d "
+		    dprintk(NDEBUG_MAIN, "scsi%d: main(): command for target %d "
 				"lun %d removed from issue_queue\n",
-				HOSTNO, tmp->target, tmp->lun);
+				HOSTNO, tmp->device->id, tmp->device->lun);
 		    /* 
 		     * REQUEST SENSE commands are issued without tagged
 		     * queueing, even on SCSI-II devices because the 
@@ -1076,7 +1066,7 @@ static void NCR5380_main (struct work_struct *bl)
 			cmd_free_tag( tmp );
 #endif
 			local_irq_restore(flags);
-			MAIN_PRINTK("scsi%d: main(): select() failed, "
+			dprintk(NDEBUG_MAIN, "scsi%d: main(): select() failed, "
 				    "returned to issue_queue\n", HOSTNO);
 			if (hostdata->connected)
 			    break;
@@ -1090,10 +1080,10 @@ static void NCR5380_main (struct work_struct *bl)
 #endif
 	    ) {
 	    local_irq_restore(flags);
-	    MAIN_PRINTK("scsi%d: main: performing information transfer\n",
+	    dprintk(NDEBUG_MAIN, "scsi%d: main: performing information transfer\n",
 			HOSTNO);
 	    NCR5380_information_transfer(instance);
-	    MAIN_PRINTK("scsi%d: main: done set false\n", HOSTNO);
+	    dprintk(NDEBUG_MAIN, "scsi%d: main: done set false\n", HOSTNO);
 	    done = 0;
 	}
     } while (!done);
@@ -1130,7 +1120,7 @@ static void NCR5380_dma_complete( struct Scsi_Host *instance )
 	return;
     }
 
-    DMA_PRINTK("scsi%d: real DMA transfer complete, basr 0x%X, sr 0x%X\n",
+    dprintk(NDEBUG_DMA, "scsi%d: real DMA transfer complete, basr 0x%X, sr 0x%X\n",
 	       HOSTNO, NCR5380_read(BUS_AND_STATUS_REG),
 	       NCR5380_read(STATUS_REG));
 
@@ -1189,27 +1179,27 @@ static irqreturn_t NCR5380_intr (int irq, void *dev_id)
     int done = 1, handled = 0;
     unsigned char basr;
 
-    INT_PRINTK("scsi%d: NCR5380 irq triggered\n", HOSTNO);
+    dprintk(NDEBUG_INTR, "scsi%d: NCR5380 irq triggered\n", HOSTNO);
 
     /* Look for pending interrupts */
     basr = NCR5380_read(BUS_AND_STATUS_REG);
-    INT_PRINTK("scsi%d: BASR=%02x\n", HOSTNO, basr);
+    dprintk(NDEBUG_INTR, "scsi%d: BASR=%02x\n", HOSTNO, basr);
     /* dispatch to appropriate routine if found and done=0 */
     if (basr & BASR_IRQ) {
-	NCR_PRINT(NDEBUG_INTR);
+	NCR5380_dprint(NDEBUG_INTR, instance);
 	if ((NCR5380_read(STATUS_REG) & (SR_SEL|SR_IO)) == (SR_SEL|SR_IO)) {
 	    done = 0;
 //	    ENABLE_IRQ();
-	    INT_PRINTK("scsi%d: SEL interrupt\n", HOSTNO);
+	    dprintk(NDEBUG_INTR, "scsi%d: SEL interrupt\n", HOSTNO);
 	    NCR5380_reselect(instance);
 	    (void) NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 	}
 	else if (basr & BASR_PARITY_ERROR) {
-	    INT_PRINTK("scsi%d: PARITY interrupt\n", HOSTNO);
+	    dprintk(NDEBUG_INTR, "scsi%d: PARITY interrupt\n", HOSTNO);
 	    (void) NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 	}
 	else if ((NCR5380_read(STATUS_REG) & SR_RST) == SR_RST) {
-	    INT_PRINTK("scsi%d: RESET interrupt\n", HOSTNO);
+	    dprintk(NDEBUG_INTR, "scsi%d: RESET interrupt\n", HOSTNO);
 	    (void)NCR5380_read(RESET_PARITY_INTERRUPT_REG);
 	}
 	else {
@@ -1229,7 +1219,7 @@ static irqreturn_t NCR5380_intr (int irq, void *dev_id)
 		((basr & BASR_END_DMA_TRANSFER) || 
 		 !(basr & BASR_PHASE_MATCH))) {
 		    
-		INT_PRINTK("scsi%d: PHASE MISM or EOP interrupt\n", HOSTNO);
+		dprintk(NDEBUG_INTR, "scsi%d: PHASE MISM or EOP interrupt\n", HOSTNO);
 		NCR5380_dma_complete( instance );
 		done = 0;
 //		ENABLE_IRQ();
@@ -1238,7 +1228,7 @@ static irqreturn_t NCR5380_intr (int irq, void *dev_id)
 	    {
 /* MS: Ignore unknown phase mismatch interrupts (caused by EOP interrupt) */
 		if (basr & BASR_PHASE_MATCH)
-		   INT_PRINTK("scsi%d: unknown interrupt, "
+		   dprintk(NDEBUG_INTR, "scsi%d: unknown interrupt, "
 			   "BASR 0x%x, MR 0x%x, SR 0x%x\n",
 			   HOSTNO, basr, NCR5380_read(MODE_REG),
 			   NCR5380_read(STATUS_REG));
@@ -1262,7 +1252,7 @@ static irqreturn_t NCR5380_intr (int irq, void *dev_id)
     }
     
     if (!done) {
-	INT_PRINTK("scsi%d: in int routine, calling main\n", HOSTNO);
+	dprintk(NDEBUG_INTR, "scsi%d: in int routine, calling main\n", HOSTNO);
 	/* Put a call to NCR5380_main() on the queue... */
 	queue_main();
     }
@@ -1338,8 +1328,8 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
     unsigned long flags;
 
     hostdata->restart_select = 0;
-    NCR_PRINT(NDEBUG_ARBITRATION);
-    ARB_PRINTK("scsi%d: starting arbitration, id = %d\n", HOSTNO,
+    NCR5380_dprint(NDEBUG_ARBITRATION, instance);
+    dprintk(NDEBUG_ARBITRATION, "scsi%d: starting arbitration, id = %d\n", HOSTNO,
 	       instance->this_id);
 
     /* 
@@ -1385,7 +1375,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 	 && !hostdata->connected);
 #endif
 
-    ARB_PRINTK("scsi%d: arbitration complete\n", HOSTNO);
+    dprintk(NDEBUG_ARBITRATION, "scsi%d: arbitration complete\n", HOSTNO);
 
     if (hostdata->connected) {
 	NCR5380_write(MODE_REG, MR_BASE); 
@@ -1406,7 +1396,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 	(NCR5380_read(INITIATOR_COMMAND_REG) & ICR_ARBITRATION_LOST) ||
 	hostdata->connected) {
 	NCR5380_write(MODE_REG, MR_BASE); 
-	ARB_PRINTK("scsi%d: lost arbitration, deasserting MR_ARBITRATE\n",
+	dprintk(NDEBUG_ARBITRATION, "scsi%d: lost arbitration, deasserting MR_ARBITRATE\n",
 		   HOSTNO);
 	return -1;
     }
@@ -1421,7 +1411,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 	hostdata->connected) {
 	NCR5380_write(MODE_REG, MR_BASE);
 	NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-	ARB_PRINTK("scsi%d: lost arbitration, deasserting ICR_ASSERT_SEL\n",
+	dprintk(NDEBUG_ARBITRATION, "scsi%d: lost arbitration, deasserting ICR_ASSERT_SEL\n",
 		   HOSTNO);
 	return -1;
     }
@@ -1444,7 +1434,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 	return -1;
     }
 
-    ARB_PRINTK("scsi%d: won arbitration\n", HOSTNO);
+    dprintk(NDEBUG_ARBITRATION, "scsi%d: won arbitration\n", HOSTNO);
 
     /* 
      * Now that we have won arbitration, start Selection process, asserting 
@@ -1504,7 +1494,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 
     udelay(1);
 
-    SEL_PRINTK("scsi%d: selecting target %d\n", HOSTNO, cmd->device->id);
+    dprintk(NDEBUG_SELECTION, "scsi%d: selecting target %d\n", HOSTNO, cmd->device->id);
 
     /* 
      * The SCSI specification calls for a 250 ms timeout for the actual 
@@ -1559,7 +1549,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 	    printk(KERN_ERR "scsi%d: weirdness\n", HOSTNO);
 	    if (hostdata->restart_select)
 		printk(KERN_NOTICE "\trestart select\n");
-	    NCR_PRINT(NDEBUG_ANY);
+	    NCR5380_dprint(NDEBUG_ANY, instance);
 	    NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 	    return -1;
 	}
@@ -1572,7 +1562,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
 #endif
 	cmd->scsi_done(cmd);
 	NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
-	SEL_PRINTK("scsi%d: target did not respond within 250ms\n", HOSTNO);
+	dprintk(NDEBUG_SELECTION, "scsi%d: target did not respond within 250ms\n", HOSTNO);
 	NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 	return 0;
     } 
@@ -1597,7 +1587,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
     /* Wait for start of REQ/ACK handshake */
     while (!(NCR5380_read(STATUS_REG) & SR_REQ));
 
-    SEL_PRINTK("scsi%d: target %d selected, going into MESSAGE OUT phase.\n",
+    dprintk(NDEBUG_SELECTION, "scsi%d: target %d selected, going into MESSAGE OUT phase.\n",
 	       HOSTNO, cmd->device->id);
     tmp[0] = IDENTIFY(1, cmd->device->lun);
 
@@ -1617,7 +1607,7 @@ static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd,
     data = tmp;
     phase = PHASE_MSGOUT;
     NCR5380_transfer_pio(instance, &phase, &len, &data);
-    SEL_PRINTK("scsi%d: nexus established.\n", HOSTNO);
+    dprintk(NDEBUG_SELECTION, "scsi%d: nexus established.\n", HOSTNO);
     /* XXX need to handle errors here */
     hostdata->connected = cmd;
 #ifndef SUPPORT_TAGS
@@ -1680,12 +1670,12 @@ static int NCR5380_transfer_pio( struct Scsi_Host *instance,
 	 */
 	while (!((tmp = NCR5380_read(STATUS_REG)) & SR_REQ));
 
-	HSH_PRINTK("scsi%d: REQ detected\n", HOSTNO);
+	dprintk(NDEBUG_HANDSHAKE, "scsi%d: REQ detected\n", HOSTNO);
 
 	/* Check for phase mismatch */	
 	if ((tmp & PHASE_MASK) != p) {
-	    PIO_PRINTK("scsi%d: phase mismatch\n", HOSTNO);
-	    NCR_PRINT_PHASE(NDEBUG_PIO);
+	    dprintk(NDEBUG_PIO, "scsi%d: phase mismatch\n", HOSTNO);
+	    NCR5380_dprint_phase(NDEBUG_PIO, instance);
 	    break;
 	}
 
@@ -1708,24 +1698,24 @@ static int NCR5380_transfer_pio( struct Scsi_Host *instance,
 	    if (!((p & SR_MSG) && c > 1)) {
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | 
 		    ICR_ASSERT_DATA);
-		NCR_PRINT(NDEBUG_PIO);
+		NCR5380_dprint(NDEBUG_PIO, instance);
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | 
 			ICR_ASSERT_DATA | ICR_ASSERT_ACK);
 	    } else {
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE |
 		    ICR_ASSERT_DATA | ICR_ASSERT_ATN);
-		NCR_PRINT(NDEBUG_PIO);
+		NCR5380_dprint(NDEBUG_PIO, instance);
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | 
 		    ICR_ASSERT_DATA | ICR_ASSERT_ATN | ICR_ASSERT_ACK);
 	    }
 	} else {
-	    NCR_PRINT(NDEBUG_PIO);
+	    NCR5380_dprint(NDEBUG_PIO, instance);
 	    NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE | ICR_ASSERT_ACK);
 	}
 
 	while (NCR5380_read(STATUS_REG) & SR_REQ);
 
-	HSH_PRINTK("scsi%d: req false, handshake complete\n", HOSTNO);
+	dprintk(NDEBUG_HANDSHAKE, "scsi%d: req false, handshake complete\n", HOSTNO);
 
 /*
  * We have several special cases to consider during REQ/ACK handshaking : 
@@ -1746,7 +1736,7 @@ static int NCR5380_transfer_pio( struct Scsi_Host *instance,
 	} 
     } while (--c);
 
-    PIO_PRINTK("scsi%d: residual %d\n", HOSTNO, c);
+    dprintk(NDEBUG_PIO, "scsi%d: residual %d\n", HOSTNO, c);
 
     *count = c;
     *data = d;
@@ -1854,7 +1844,7 @@ static int NCR5380_transfer_dma( struct Scsi_Host *instance,
     }
     hostdata->dma_len = c;
 
-    DMA_PRINTK("scsi%d: initializing DMA for %s, %d bytes %s %p\n",
+    dprintk(NDEBUG_DMA, "scsi%d: initializing DMA for %s, %d bytes %s %p\n",
 	       HOSTNO, (p & SR_IO) ? "reading" : "writing",
 	       c, (p & SR_IO) ? "to" : "from", *data);
 
@@ -1931,7 +1921,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 	    phase = (tmp & PHASE_MASK); 
  	    if (phase != old_phase) {
 		old_phase = phase;
-		NCR_PRINT_PHASE(NDEBUG_INFORMATION);
+		NCR5380_dprint_phase(NDEBUG_INFORMATION, instance);
 	    }
 
 	    if(phase == PHASE_CMDOUT) {
@@ -1996,7 +1986,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		    --cmd->SCp.buffers_residual;
 		    cmd->SCp.this_residual = cmd->SCp.buffer->length;
 		    cmd->SCp.ptr = SGADDR(cmd->SCp.buffer);
-		    INF_PRINTK("scsi%d: %d bytes and %d buffers left\n",
+		    dprintk(NDEBUG_INFORMATION, "scsi%d: %d bytes and %d buffers left\n",
 			       HOSTNO, cmd->SCp.this_residual,
 			       cmd->SCp.buffers_residual);
 		}
@@ -2088,7 +2078,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		    /* Accept message by clearing ACK */
 		    NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 		    
-		    LNK_PRINTK("scsi%d: target %d lun %d linked command "
+		    dprintk(NDEBUG_LINKED, "scsi%d: target %d lun %d linked command "
 			       "complete.\n", HOSTNO, cmd->device->id, cmd->device->lun);
 
 		    /* Enable reselect interrupts */
@@ -2113,7 +2103,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		     * and don't free it! */
 		    cmd->next_link->tag = cmd->tag;
 		    cmd->result = cmd->SCp.Status | (cmd->SCp.Message << 8); 
-		    LNK_PRINTK("scsi%d: target %d lun %d linked request "
+		    dprintk(NDEBUG_LINKED, "scsi%d: target %d lun %d linked request "
 			       "done, calling scsi_done().\n",
 			       HOSTNO, cmd->device->id, cmd->device->lun);
 #ifdef NCR5380_STATS
@@ -2128,7 +2118,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		    /* Accept message by clearing ACK */
 		    NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 		    hostdata->connected = NULL;
-		    QU_PRINTK("scsi%d: command for target %d, lun %d "
+		    dprintk(NDEBUG_QUEUES, "scsi%d: command for target %d, lun %d "
 			      "completed\n", HOSTNO, cmd->device->id, cmd->device->lun);
 #ifdef SUPPORT_TAGS
 		    cmd_free_tag( cmd );
@@ -2142,7 +2132,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 			/* ++Andreas: the mid level code knows about
 			   QUEUE_FULL now. */
 			TAG_ALLOC *ta = &TagAlloc[cmd->device->id][cmd->device->lun];
-			TAG_PRINTK("scsi%d: target %d lun %d returned "
+			dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d returned "
 				   "QUEUE_FULL after %d commands\n",
 				   HOSTNO, cmd->device->id, cmd->device->lun,
 				   ta->nr_allocated);
@@ -2186,7 +2176,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		    if ((cmd->cmnd[0] != REQUEST_SENSE) && 
 			(status_byte(cmd->SCp.Status) == CHECK_CONDITION)) {
 			scsi_eh_prep_cmnd(cmd, &hostdata->ses, NULL, 0, ~0);
-			ASEN_PRINTK("scsi%d: performing request sense\n",
+			dprintk(NDEBUG_AUTOSENSE, "scsi%d: performing request sense\n",
 				    HOSTNO);
 			/* this is initialized from initialize_SCp 
 			cmd->SCp.buffer = NULL;
@@ -2198,7 +2188,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 			SET_NEXT(cmd, hostdata->issue_queue);
 		        hostdata->issue_queue = (struct scsi_cmnd *) cmd;
 		        local_irq_restore(flags);
-			QU_PRINTK("scsi%d: REQUEST SENSE added to head of "
+			dprintk(NDEBUG_QUEUES, "scsi%d: REQUEST SENSE added to head of "
 				  "issue queue\n", H_NO(cmd));
 		   } else
 #endif /* def AUTOSENSE */
@@ -2238,7 +2228,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 			cmd->device->tagged_supported = 0;
 			hostdata->busy[cmd->device->id] |= (1 << cmd->device->lun);
 			cmd->tag = TAG_NONE;
-			TAG_PRINTK("scsi%d: target %d lun %d rejected "
+			dprintk(NDEBUG_TAGS, "scsi%d: target %d lun %d rejected "
 				   "QUEUE_TAG message; tagged queuing "
 				   "disabled\n",
 				   HOSTNO, cmd->device->id, cmd->device->lun);
@@ -2255,7 +2245,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		    hostdata->connected = NULL;
 		    hostdata->disconnected_queue = cmd;
 		    local_irq_restore(flags);
-		    QU_PRINTK("scsi%d: command for target %d lun %d was "
+		    dprintk(NDEBUG_QUEUES, "scsi%d: command for target %d lun %d was "
 			      "moved from connected to the "
 			      "disconnected_queue\n", HOSTNO, 
 			      cmd->device->id, cmd->device->lun);
@@ -2308,13 +2298,13 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		    /* Accept first byte by clearing ACK */
 		    NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 
-		    EXT_PRINTK("scsi%d: receiving extended message\n", HOSTNO);
+		    dprintk(NDEBUG_EXTENDED, "scsi%d: receiving extended message\n", HOSTNO);
 
 		    len = 2;
 		    data = extended_msg + 1;
 		    phase = PHASE_MSGIN;
 		    NCR5380_transfer_pio(instance, &phase, &len, &data);
-		    EXT_PRINTK("scsi%d: length=%d, code=0x%02x\n", HOSTNO,
+		    dprintk(NDEBUG_EXTENDED, "scsi%d: length=%d, code=0x%02x\n", HOSTNO,
 			       (int)extended_msg[1], (int)extended_msg[2]);
 
 		    if (!len && extended_msg[1] <= 
@@ -2326,7 +2316,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 			phase = PHASE_MSGIN;
 
 			NCR5380_transfer_pio(instance, &phase, &len, &data);
-			EXT_PRINTK("scsi%d: message received, residual %d\n",
+			dprintk(NDEBUG_EXTENDED, "scsi%d: message received, residual %d\n",
 				   HOSTNO, len);
 
 			switch (extended_msg[2]) {
@@ -2416,7 +2406,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		break;
 	    default:
 		printk("scsi%d: unknown phase\n", HOSTNO);
-		NCR_PRINT(NDEBUG_ANY);
+		NCR5380_dprint(NDEBUG_ANY, instance);
 	    } /* switch(phase) */
 	} /* if (tmp * SR_REQ) */ 
     } /* while (1) */
@@ -2458,7 +2448,7 @@ static void NCR5380_reselect (struct Scsi_Host *instance)
 
     target_mask = NCR5380_read(CURRENT_SCSI_DATA_REG) & ~(hostdata->id_mask);
 
-    RSL_PRINTK("scsi%d: reselect\n", HOSTNO);
+    dprintk(NDEBUG_RESELECTION, "scsi%d: reselect\n", HOSTNO);
 
     /* 
      * At this point, we have detected that our SCSI ID is on the bus,
@@ -2580,14 +2570,14 @@ static void NCR5380_reselect (struct Scsi_Host *instance)
 	if (!NCR5380_transfer_pio(instance, &phase, &len, &data) &&
 	    msg[1] == SIMPLE_QUEUE_TAG)
 	    tag = msg[2];
-	TAG_PRINTK("scsi%d: target mask %02x, lun %d sent tag %d at "
+	dprintk(NDEBUG_TAGS, "scsi%d: target mask %02x, lun %d sent tag %d at "
 		   "reselection\n", HOSTNO, target_mask, lun, tag);
     }
 #endif
     
     hostdata->connected = tmp;
-    RSL_PRINTK("scsi%d: nexus established, target = %d, lun = %d, tag = %d\n",
-	       HOSTNO, tmp->target, tmp->lun, tmp->tag);
+    dprintk(NDEBUG_RESELECTION, "scsi%d: nexus established, target = %d, lun = %d, tag = %d\n",
+	       HOSTNO, tmp->device->id, tmp->device->lun, tmp->tag);
 }
 
 
@@ -2622,7 +2612,7 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 
     local_irq_save(flags);
     
-    ABRT_PRINTK("scsi%d: abort called basr 0x%02x, sr 0x%02x\n", HOSTNO,
+    dprintk(NDEBUG_ABORT, "scsi%d: abort called basr 0x%02x, sr 0x%02x\n", HOSTNO,
 		NCR5380_read(BUS_AND_STATUS_REG),
 		NCR5380_read(STATUS_REG));
 
@@ -2635,7 +2625,7 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 
     if (hostdata->connected == cmd) {
 
-	ABRT_PRINTK("scsi%d: aborting connected command\n", HOSTNO);
+	dprintk(NDEBUG_ABORT, "scsi%d: aborting connected command\n", HOSTNO);
 /*
  * We should perform BSY checking, and make sure we haven't slipped
  * into BUS FREE.
@@ -2664,11 +2654,11 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 #endif
 	  local_irq_restore(flags);
 	  cmd->scsi_done(cmd);
-	  return SCSI_ABORT_SUCCESS;
+	  return SUCCESS;
 	} else {
 /*	  local_irq_restore(flags); */
 	  printk("scsi%d: abort of connected command failed!\n", HOSTNO);
-	  return SCSI_ABORT_ERROR;
+	  return FAILED;
 	} 
    }
 #endif
@@ -2686,12 +2676,12 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 	    SET_NEXT(tmp, NULL);
 	    tmp->result = DID_ABORT << 16;
 	    local_irq_restore(flags);
-	    ABRT_PRINTK("scsi%d: abort removed command from issue queue.\n",
+	    dprintk(NDEBUG_ABORT, "scsi%d: abort removed command from issue queue.\n",
 			HOSTNO);
 	    /* Tagged queuing note: no tag to free here, hasn't been assigned
 	     * yet... */
 	    tmp->scsi_done(tmp);
-	    return SCSI_ABORT_SUCCESS;
+	    return SUCCESS;
 	}
 
 /* 
@@ -2707,8 +2697,8 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 
     if (hostdata->connected) {
 	local_irq_restore(flags);
-	ABRT_PRINTK("scsi%d: abort failed, command connected.\n", HOSTNO);
-        return SCSI_ABORT_SNOOZE;
+	dprintk(NDEBUG_ABORT, "scsi%d: abort failed, command connected.\n", HOSTNO);
+        return FAILED;
     }
 
 /*
@@ -2740,12 +2730,12 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 	 tmp = NEXT(tmp)) 
         if (cmd == tmp) {
             local_irq_restore(flags);
-	    ABRT_PRINTK("scsi%d: aborting disconnected command.\n", HOSTNO);
+	    dprintk(NDEBUG_ABORT, "scsi%d: aborting disconnected command.\n", HOSTNO);
   
             if (NCR5380_select (instance, cmd, (int) cmd->tag)) 
-		return SCSI_ABORT_BUSY;
+		return FAILED;
 
-	    ABRT_PRINTK("scsi%d: nexus reestablished.\n", HOSTNO);
+	    dprintk(NDEBUG_ABORT, "scsi%d: nexus reestablished.\n", HOSTNO);
 
 	    do_abort (instance);
 
@@ -2769,7 +2759,7 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 #endif
 		    local_irq_restore(flags);
 		    tmp->scsi_done(tmp);
-		    return SCSI_ABORT_SUCCESS;
+		    return SUCCESS;
 		}
 	}
 
@@ -2786,7 +2776,7 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
     local_irq_restore(flags);
     printk(KERN_INFO "scsi%d: warning : SCSI command probably completed successfully before abortion\n", HOSTNO); 
 
-    return SCSI_ABORT_NOT_RUNNING;
+    return FAILED;
 }
 
 
@@ -2795,7 +2785,7 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
  * 
  * Purpose : reset the SCSI bus.
  *
- * Returns : SCSI_RESET_WAKEUP
+ * Returns : SUCCESS or FAILURE
  *
  */ 
 
@@ -2804,7 +2794,7 @@ static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
     SETUP_HOSTDATA(cmd->device->host);
     int           i;
     unsigned long flags;
-#if 1
+#if defined(RESET_RUN_DONE)
     struct scsi_cmnd *connected, *disconnected_queue;
 #endif
 
@@ -2826,8 +2816,15 @@ static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
      * through anymore ... */
     (void)NCR5380_read( RESET_PARITY_INTERRUPT_REG );
 
-#if 1 /* XXX Should now be done by midlevel code, but it's broken XXX */
-      /* XXX see below                                            XXX */
+	/* MSch 20140115 - looking at the generic NCR5380 driver, all of this
+	 * should go.
+	 * Catch-22: if we don't clear all queues, the SCSI driver lock will
+	 * not be released by atari_scsi_reset()!
+	 */
+
+#if defined(RESET_RUN_DONE)
+	/* XXX Should now be done by midlevel code, but it's broken XXX */
+	/* XXX see below                                            XXX */
 
     /* MSch: old-style reset: actually abort all command processing here */
 
@@ -2857,7 +2854,7 @@ static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
      */
 
     if ((cmd = connected)) {
-	ABRT_PRINTK("scsi%d: reset aborted a connected command\n", H_NO(cmd));
+	dprintk(NDEBUG_ABORT, "scsi%d: reset aborted a connected command\n", H_NO(cmd));
 	cmd->result = (cmd->result & 0xffff) | (DID_RESET << 16);
 	cmd->scsi_done( cmd );
     }
@@ -2869,14 +2866,14 @@ static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
 	cmd->scsi_done( cmd );
     }
     if (i > 0)
-	ABRT_PRINTK("scsi: reset aborted %d disconnected command(s)\n", i);
+	dprintk(NDEBUG_ABORT, "scsi: reset aborted %d disconnected command(s)\n", i);
 
 
     /* since all commands have been explicitly terminated, we need to tell
      * the midlevel code that the reset was SUCCESSFUL, and there is no 
      * need to 'wake up' the commands by a request_sense
      */
-    return SCSI_RESET_SUCCESS | SCSI_RESET_BUS_RESET;
+    return SUCCESS;
 #else /* 1 */
 
     /* MSch: new-style reset handling: let the mid-level do what it can */
@@ -2903,11 +2900,11 @@ static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
      */
 
     if (hostdata->issue_queue)
-	ABRT_PRINTK("scsi%d: reset aborted issued command(s)\n", H_NO(cmd));
+	dprintk(NDEBUG_ABORT, "scsi%d: reset aborted issued command(s)\n", H_NO(cmd));
     if (hostdata->connected) 
-	ABRT_PRINTK("scsi%d: reset aborted a connected command\n", H_NO(cmd));
+	dprintk(NDEBUG_ABORT, "scsi%d: reset aborted a connected command\n", H_NO(cmd));
     if (hostdata->disconnected_queue)
-	ABRT_PRINTK("scsi%d: reset aborted disconnected command(s)\n", H_NO(cmd));
+	dprintk(NDEBUG_ABORT, "scsi%d: reset aborted disconnected command(s)\n", H_NO(cmd));
 
     local_irq_save(flags);
     hostdata->issue_queue = NULL;
@@ -2924,7 +2921,7 @@ static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
     local_irq_restore(flags);
 
     /* we did no complete reset of all commands, so a wakeup is required */
-    return SCSI_RESET_WAKEUP | SCSI_RESET_BUS_RESET;
+    return SUCCESS;
 #endif /* 1 */
 }
 
diff --git a/drivers/scsi/sun3_scsi.c b/drivers/scsi/sun3_scsi.c
index e2c009b033ce..9707b7494a89 100644
--- a/drivers/scsi/sun3_scsi.c
+++ b/drivers/scsi/sun3_scsi.c
@@ -3,6 +3,10 @@
  *
  * Sun3 DMA routines added by Sam Creasey (sammy@sammy.net)
  *
+ * VME support added by Sam Creasey
+ *
+ * TODO: modify this driver to support multiple Sun3 SCSI VME boards
+ *
  * Adapted from mac_scsinew.c:
  */
 /*
@@ -45,10 +49,6 @@
  * USLEEP - enable support for devices that don't disconnect.  Untested.
  */
 
-/*
- * $Log: sun3_NCR5380.c,v $
- */
-
 #define AUTOSENSE
 
 #include <linux/types.h>
@@ -69,23 +69,15 @@
 #include <asm/idprom.h>
 #include <asm/machines.h>
 
-#define NDEBUG 0
-
-#define NDEBUG_ABORT		0x00100000
-#define NDEBUG_TAGS		0x00200000
-#define NDEBUG_MERGING		0x00400000
-
 /* dma on! */
 #define REAL_DMA
 
 #include "scsi.h"
-#include "initio.h"
 #include <scsi/scsi_host.h>
 #include "sun3_scsi.h"
+#include "NCR5380.h"
 
-static void NCR5380_print(struct Scsi_Host *instance);
-
-/* #define OLDDMA */
+extern int sun3_map_test(unsigned long, char *);
 
 #define USE_WRAPPER
 /*#define RESET_BOOT */
@@ -101,7 +93,11 @@ static void NCR5380_print(struct Scsi_Host *instance);
 
 /* #define SUPPORT_TAGS */
 
+#ifdef SUN3_SCSI_VME
+#define ENABLE_IRQ()
+#else
 #define	ENABLE_IRQ()	enable_irq( IRQ_SUN3_SCSI ); 
+#endif
 
 
 static irqreturn_t scsi_sun3_intr(int irq, void *dummy);
@@ -123,6 +119,8 @@ module_param(setup_hostid, int, 0);
 
 static struct scsi_cmnd *sun3_dma_setup_done = NULL;
 
+#define	RESET_RUN_DONE
+
 #define	AFTER_RESET_DELAY	(HZ/2)
 
 /* ms to wait after hitting dma regs */
@@ -136,10 +134,9 @@ static struct scsi_cmnd *sun3_dma_setup_done = NULL;
 
 static volatile unsigned char *sun3_scsi_regp;
 static volatile struct sun3_dma_regs *dregs;
-#ifdef OLDDMA
-static unsigned char *dmabuf = NULL; /* dma memory buffer */
-#endif
+#ifndef SUN3_SCSI_VME
 static struct sun3_udc_regs *udc_regs = NULL;
+#endif
 static unsigned char *sun3_dma_orig_addr = NULL;
 static unsigned long sun3_dma_orig_count = 0;
 static int sun3_dma_active = 0;
@@ -159,6 +156,7 @@ static inline void sun3scsi_write(int reg, int value)
 	sun3_scsi_regp[reg] = value;
 }
 
+#ifndef SUN3_SCSI_VME
 /* dma controller register access functions */
 
 static inline unsigned short sun3_udc_read(unsigned char reg)
@@ -180,6 +178,7 @@ static inline void sun3_udc_write(unsigned short val, unsigned char reg)
 	dregs->udc_data = val;
 	udelay(SUN3_DMA_DELAY);
 }
+#endif
 
 /*
  * XXX: status debug
@@ -198,17 +197,32 @@ static struct Scsi_Host *default_instance;
  *
  */
  
-int __init sun3scsi_detect(struct scsi_host_template * tpnt)
+static int __init sun3scsi_detect(struct scsi_host_template *tpnt)
 {
-	unsigned long ioaddr;
+	unsigned long ioaddr, irq;
 	static int called = 0;
 	struct Scsi_Host *instance;
+#ifdef SUN3_SCSI_VME
+	int i;
+	unsigned long addrs[3] = { IOBASE_SUN3_VMESCSI,
+				   IOBASE_SUN3_VMESCSI + 0x4000,
+				   0 };
+	unsigned long vecs[3] = { SUN3_VEC_VMESCSI0,
+				  SUN3_VEC_VMESCSI1,
+				  0 };
+#endif
 
 	/* check that this machine has an onboard 5380 */
 	switch(idprom->id_machtype) {
+#ifdef SUN3_SCSI_VME
+	case SM_SUN3|SM_3_160:
+	case SM_SUN3|SM_3_260:
+		break;
+#else
 	case SM_SUN3|SM_3_50:
 	case SM_SUN3|SM_3_60:
 		break;
+#endif
 
 	default:
 		return 0;
@@ -217,7 +231,11 @@ int __init sun3scsi_detect(struct scsi_host_template * tpnt)
 	if(called)
 		return 0;
 
+#ifdef SUN3_SCSI_VME
+	tpnt->proc_name = "Sun3 5380 VME SCSI";
+#else
 	tpnt->proc_name = "Sun3 5380 SCSI";
+#endif
 
 	/* setup variables */
 	tpnt->can_queue =
@@ -234,6 +252,38 @@ int __init sun3scsi_detect(struct scsi_host_template * tpnt)
 		tpnt->this_id = 7;
 	}
 
+#ifdef SUN3_SCSI_VME
+	ioaddr = 0;
+	for (i = 0; addrs[i] != 0; i++) {
+		unsigned char x;
+
+		ioaddr = (unsigned long)sun3_ioremap(addrs[i], PAGE_SIZE,
+						     SUN3_PAGE_TYPE_VME16);
+		irq = vecs[i];
+		sun3_scsi_regp = (unsigned char *)ioaddr;
+
+		dregs = (struct sun3_dma_regs *)(((unsigned char *)ioaddr) + 8);
+
+		if (sun3_map_test((unsigned long)dregs, &x)) {
+			unsigned short oldcsr;
+
+			oldcsr = dregs->csr;
+			dregs->csr = 0;
+			udelay(SUN3_DMA_DELAY);
+			if (dregs->csr == 0x1400)
+				break;
+
+			dregs->csr = oldcsr;
+		}
+
+		iounmap((void *)ioaddr);
+		ioaddr = 0;
+	}
+
+	if (!ioaddr)
+		return 0;
+#else
+	irq = IRQ_SUN3_SCSI;
 	ioaddr = (unsigned long)ioremap(IOBASE_SUN3_SCSI, PAGE_SIZE);
 	sun3_scsi_regp = (unsigned char *)ioaddr;
 
@@ -244,11 +294,6 @@ int __init sun3scsi_detect(struct scsi_host_template * tpnt)
 	     printk("SUN3 Scsi couldn't allocate DVMA memory!\n");
 	     return 0;
 	}
-#ifdef OLDDMA
-	if((dmabuf = dvma_malloc_align(SUN3_DVMA_BUFSIZE, 0x10000)) == NULL) {
-	     printk("SUN3 Scsi couldn't allocate DVMA memory!\n");
-	     return 0;
-	}
 #endif
 #ifdef SUPPORT_TAGS
 	if (setup_use_tagged_queuing < 0)
@@ -262,7 +307,7 @@ int __init sun3scsi_detect(struct scsi_host_template * tpnt)
 	default_instance = instance;
 
         instance->io_port = (unsigned long) ioaddr;
-	instance->irq = IRQ_SUN3_SCSI;
+	instance->irq = irq;
 
 	NCR5380_init(instance, 0);
 
@@ -283,7 +328,8 @@ int __init sun3scsi_detect(struct scsi_host_template * tpnt)
 #endif
 	}
 	
-	printk("scsi%d: Sun3 5380 at port %lX irq", instance->host_no, instance->io_port);
+	pr_info("scsi%d: %s at port %lX irq", instance->host_no,
+		tpnt->proc_name, instance->io_port);
 	if (instance->irq == SCSI_IRQ_NONE)
 		printk ("s disabled");
 	else
@@ -300,6 +346,15 @@ int __init sun3scsi_detect(struct scsi_host_template * tpnt)
 	dregs->csr = CSR_SCSI | CSR_FIFO | CSR_INTR;
 	udelay(SUN3_DMA_DELAY);
 	dregs->fifo_count = 0;
+#ifdef SUN3_SCSI_VME
+	dregs->fifo_count_hi = 0;
+	dregs->dma_addr_hi = 0;
+	dregs->dma_addr_lo = 0;
+	dregs->dma_count_hi = 0;
+	dregs->dma_count_lo = 0;
+
+	dregs->ivect = VME_DATA24 | (instance->irq & 0xff);
+#endif
 
 	called = 1;
 
@@ -367,7 +422,8 @@ static void sun3_scsi_reset_boot(struct Scsi_Host *instance)
 }
 #endif
 
-const char * sun3scsi_info (struct Scsi_Host *spnt) {
+static const char *sun3scsi_info(struct Scsi_Host *spnt)
+{
     return "";
 }
 
@@ -379,6 +435,10 @@ static irqreturn_t scsi_sun3_intr(int irq, void *dummy)
 	unsigned short csr = dregs->csr;
 	int handled = 0;
 
+#ifdef SUN3_SCSI_VME
+	dregs->csr &= ~CSR_DMA_ENABLE;
+#endif
+
 	if(csr & ~CSR_GOOD) {
 		if(csr & CSR_DMA_BUSERR) {
 			printk("scsi%d: bus error in dma\n", default_instance->host_no);
@@ -422,31 +482,28 @@ void sun3_sun3_debug (void)
 /* sun3scsi_dma_setup() -- initialize the dma controller for a read/write */
 static unsigned long sun3scsi_dma_setup(void *data, unsigned long count, int write_flag)
 {
-#ifdef OLDDMA
-	if(write_flag) 
-		memcpy(dmabuf, data, count);
-	else {
-		sun3_dma_orig_addr = data;
-		sun3_dma_orig_count = count;
-	}
-#else
 	void *addr;
 
 	if(sun3_dma_orig_addr != NULL)
 		dvma_unmap(sun3_dma_orig_addr);
 
-//	addr = sun3_dvma_page((unsigned long)data, (unsigned long)dmabuf);
+#ifdef SUN3_SCSI_VME
+	addr = (void *)dvma_map_vme((unsigned long) data, count);
+#else
 	addr = (void *)dvma_map((unsigned long) data, count);
+#endif
 		
 	sun3_dma_orig_addr = addr;
 	sun3_dma_orig_count = count;
-#endif
+
+#ifndef SUN3_SCSI_VME
 	dregs->fifo_count = 0;
 	sun3_udc_write(UDC_RESET, UDC_CSR);
 	
 	/* reset fifo */
 	dregs->csr &= ~CSR_FIFO;
 	dregs->csr |= CSR_FIFO;
+#endif
 	
 	/* set direction */
 	if(write_flag)
@@ -454,6 +511,17 @@ static unsigned long sun3scsi_dma_setup(void *data, unsigned long count, int wri
 	else
 		dregs->csr &= ~CSR_SEND;
 	
+#ifdef SUN3_SCSI_VME
+	dregs->csr |= CSR_PACK_ENABLE;
+
+	dregs->dma_addr_hi = ((unsigned long)addr >> 16);
+	dregs->dma_addr_lo = ((unsigned long)addr & 0xffff);
+
+	dregs->dma_count_hi = 0;
+	dregs->dma_count_lo = 0;
+	dregs->fifo_count_hi = 0;
+	dregs->fifo_count = 0;
+#else
 	/* byte count for fifo */
 	dregs->fifo_count = count;
 
@@ -467,17 +535,12 @@ static unsigned long sun3scsi_dma_setup(void *data, unsigned long count, int wri
 		printk("scsi%d: fifo_mismatch %04x not %04x\n",
 		       default_instance->host_no, dregs->fifo_count,
 		       (unsigned int) count);
-		NCR5380_print(default_instance);
+		NCR5380_dprint(NDEBUG_DMA, default_instance);
 	}
 
 	/* setup udc */
-#ifdef OLDDMA
-	udc_regs->addr_hi = ((dvma_vtob(dmabuf) & 0xff0000) >> 8);
-	udc_regs->addr_lo = (dvma_vtob(dmabuf) & 0xffff);
-#else
 	udc_regs->addr_hi = (((unsigned long)(addr) & 0xff0000) >> 8);
 	udc_regs->addr_lo = ((unsigned long)(addr) & 0xffff);
-#endif
 	udc_regs->count = count/2; /* count in words */
 	udc_regs->mode_hi = UDC_MODE_HIWORD;
 	if(write_flag) {
@@ -501,11 +564,13 @@ static unsigned long sun3scsi_dma_setup(void *data, unsigned long count, int wri
 
 	/* interrupt enable */
 	sun3_udc_write(UDC_INT_ENABLE, UDC_CSR);
+#endif
 	
        	return count;
 
 }
 
+#ifndef SUN3_SCSI_VME
 static inline unsigned long sun3scsi_dma_count(struct Scsi_Host *instance)
 {
 	unsigned short resid;
@@ -518,6 +583,7 @@ static inline unsigned long sun3scsi_dma_count(struct Scsi_Host *instance)
 
 	return (unsigned long) resid;
 }
+#endif
 
 static inline unsigned long sun3scsi_dma_residual(struct Scsi_Host *instance)
 {
@@ -536,8 +602,23 @@ static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted,
 
 static inline int sun3scsi_dma_start(unsigned long count, unsigned char *data)
 {
+#ifdef SUN3_SCSI_VME
+	unsigned short csr;
+
+	csr = dregs->csr;
 
+	dregs->dma_count_hi = (sun3_dma_orig_count >> 16);
+	dregs->dma_count_lo = (sun3_dma_orig_count & 0xffff);
+
+	dregs->fifo_count_hi = (sun3_dma_orig_count >> 16);
+	dregs->fifo_count = (sun3_dma_orig_count & 0xffff);
+
+/*	if(!(csr & CSR_DMA_ENABLE))
+ *		dregs->csr |= CSR_DMA_ENABLE;
+ */
+#else
     sun3_udc_write(UDC_CHN_START, UDC_CSR);
+#endif
     
     return 0;
 }
@@ -545,12 +626,46 @@ static inline int sun3scsi_dma_start(unsigned long count, unsigned char *data)
 /* clean up after our dma is done */
 static int sun3scsi_dma_finish(int write_flag)
 {
-	unsigned short count;
+	unsigned short __maybe_unused count;
 	unsigned short fifo;
 	int ret = 0;
 	
 	sun3_dma_active = 0;
-#if 1
+
+#ifdef SUN3_SCSI_VME
+	dregs->csr &= ~CSR_DMA_ENABLE;
+
+	fifo = dregs->fifo_count;
+	if (write_flag) {
+		if ((fifo > 0) && (fifo < sun3_dma_orig_count))
+			fifo++;
+	}
+
+	last_residual = fifo;
+	/* empty bytes from the fifo which didn't make it */
+	if ((!write_flag) && (dregs->csr & CSR_LEFT)) {
+		unsigned char *vaddr;
+
+		vaddr = (unsigned char *)dvma_vmetov(sun3_dma_orig_addr);
+
+		vaddr += (sun3_dma_orig_count - fifo);
+		vaddr--;
+
+		switch (dregs->csr & CSR_LEFT) {
+		case CSR_LEFT_3:
+			*vaddr = (dregs->bpack_lo & 0xff00) >> 8;
+			vaddr--;
+
+		case CSR_LEFT_2:
+			*vaddr = (dregs->bpack_hi & 0x00ff);
+			vaddr--;
+
+		case CSR_LEFT_1:
+			*vaddr = (dregs->bpack_hi & 0xff00) >> 8;
+			break;
+		}
+	}
+#else
 	// check to empty the fifo on a read
 	if(!write_flag) {
 		int tmo = 20000; /* .2 sec */
@@ -566,28 +681,8 @@ static int sun3scsi_dma_finish(int write_flag)
 			udelay(10);
 		}
 	}
-		
-#endif
 
 	count = sun3scsi_dma_count(default_instance);
-#ifdef OLDDMA
-
-	/* if we've finished a read, copy out the data we read */
- 	if(sun3_dma_orig_addr) {
-		/* check for residual bytes after dma end */
-		if(count && (NCR5380_read(BUS_AND_STATUS_REG) &
-			     (BASR_PHASE_MATCH | BASR_ACK))) {
-			printk("scsi%d: sun3_scsi_finish: read overrun baby... ", default_instance->host_no);
-			printk("basr now %02x\n", NCR5380_read(BUS_AND_STATUS_REG));
-			ret = count;
-		}
-		
-		/* copy in what we dma'd no matter what */
-		memcpy(sun3_dma_orig_addr, dmabuf, sun3_dma_orig_count);
-		sun3_dma_orig_addr = NULL;
-
-	}
-#else
 
 	fifo = dregs->fifo_count;
 	last_residual = fifo;
@@ -605,10 +700,23 @@ static int sun3scsi_dma_finish(int write_flag)
 		vaddr[-2] = (data & 0xff00) >> 8;
 		vaddr[-1] = (data & 0xff);
 	}
+#endif
 
 	dvma_unmap(sun3_dma_orig_addr);
 	sun3_dma_orig_addr = NULL;
-#endif
+
+#ifdef SUN3_SCSI_VME
+	dregs->dma_addr_hi = 0;
+	dregs->dma_addr_lo = 0;
+	dregs->dma_count_hi = 0;
+	dregs->dma_count_lo = 0;
+
+	dregs->fifo_count = 0;
+	dregs->fifo_count_hi = 0;
+
+	dregs->csr &= ~CSR_SEND;
+/*	dregs->csr |= CSR_DMA_ENABLE; */
+#else
 	sun3_udc_write(UDC_RESET, UDC_CSR);
 	dregs->fifo_count = 0;
 	dregs->csr &= ~CSR_SEND;
@@ -616,6 +724,7 @@ static int sun3scsi_dma_finish(int write_flag)
 	/* reset fifo */
 	dregs->csr &= ~CSR_FIFO;
 	dregs->csr |= CSR_FIFO;
+#endif
 	
 	sun3_dma_setup_done = NULL;
 
diff --git a/drivers/scsi/sun3_scsi.h b/drivers/scsi/sun3_scsi.h
index a8da9c710fea..e96a37cf06ac 100644
--- a/drivers/scsi/sun3_scsi.h
+++ b/drivers/scsi/sun3_scsi.h
@@ -29,12 +29,8 @@
  * 1+ (800) 334-5454
  */
 
-/*
- * $Log: cumana_NCR5380.h,v $
- */
-
-#ifndef SUN3_NCR5380_H
-#define SUN3_NCR5380_H
+#ifndef SUN3_SCSI_H
+#define SUN3_SCSI_H
 
 #define SUN3SCSI_PUBLIC_RELEASE 1
 
@@ -82,8 +78,6 @@ static int sun3scsi_release (struct Scsi_Host *);
 #define SUN3_SCSI_NAME "Sun3 NCR5380 SCSI"
 #endif
 
-#ifndef HOSTS_C
-
 #define NCR5380_implementation_fields \
     int port, ctrl
 
@@ -108,9 +102,6 @@ static int sun3scsi_release (struct Scsi_Host *);
 #define NCR5380_dma_read_setup(instance, data, count) sun3scsi_dma_setup(data, count, 0)
 #define NCR5380_dma_residual sun3scsi_dma_residual
 
-#define BOARD_NORMAL	0
-#define BOARD_NCR53C400	1
-
 /* additional registers - mainly DMA control regs */
 /* these start at regbase + 8 -- directly after the NCR regs */
 struct sun3_dma_regs {
@@ -191,189 +182,5 @@ struct sun3_udc_regs {
 
 #define VME_DATA24 0x3d00
 
-// debugging printk's, taken from atari_scsi.h 
-/* Debugging printk definitions:
- *
- *  ARB  -> arbitration
- *  ASEN -> auto-sense
- *  DMA  -> DMA
- *  HSH  -> PIO handshake
- *  INF  -> information transfer
- *  INI  -> initialization
- *  INT  -> interrupt
- *  LNK  -> linked commands
- *  MAIN -> NCR5380_main() control flow
- *  NDAT -> no data-out phase
- *  NWR  -> no write commands
- *  PIO  -> PIO transfers
- *  PDMA -> pseudo DMA (unused on Atari)
- *  QU   -> queues
- *  RSL  -> reselections
- *  SEL  -> selections
- *  USL  -> usleep cpde (unused on Atari)
- *  LBS  -> last byte sent (unused on Atari)
- *  RSS  -> restarting of selections
- *  EXT  -> extended messages
- *  ABRT -> aborting and resetting
- *  TAG  -> queue tag handling
- *  MER  -> merging of consec. buffers
- *
- */
-
-#include "NCR5380.h"
-
-#if NDEBUG & NDEBUG_ARBITRATION
-#define ARB_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define ARB_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_AUTOSENSE
-#define ASEN_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define ASEN_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_DMA
-#define DMA_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define DMA_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_HANDSHAKE
-#define HSH_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define HSH_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_INFORMATION
-#define INF_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define INF_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_INIT
-#define INI_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define INI_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_INTR
-#define INT_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define INT_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_LINKED
-#define LNK_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define LNK_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_MAIN
-#define MAIN_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define MAIN_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_NO_DATAOUT
-#define NDAT_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define NDAT_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_NO_WRITE
-#define NWR_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define NWR_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_PIO
-#define PIO_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define PIO_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_PSEUDO_DMA
-#define PDMA_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define PDMA_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_QUEUES
-#define QU_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define QU_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_RESELECTION
-#define RSL_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define RSL_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_SELECTION
-#define SEL_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define SEL_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_USLEEP
-#define USL_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define USL_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_LAST_BYTE_SENT
-#define LBS_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define LBS_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_RESTART_SELECT
-#define RSS_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define RSS_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_EXTENDED
-#define EXT_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define EXT_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_ABORT
-#define ABRT_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define ABRT_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_TAGS
-#define TAG_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define TAG_PRINTK(format, args...)
-#endif
-#if NDEBUG & NDEBUG_MERGING
-#define MER_PRINTK(format, args...) \
-	printk(KERN_DEBUG format , ## args)
-#else
-#define MER_PRINTK(format, args...)
-#endif
-
-/* conditional macros for NCR5380_print_{,phase,status} */
-
-#define NCR_PRINT(mask)	\
-	((NDEBUG & (mask)) ? NCR5380_print(instance) : (void)0)
-
-#define NCR_PRINT_PHASE(mask) \
-	((NDEBUG & (mask)) ? NCR5380_print_phase(instance) : (void)0)
-
-#define NCR_PRINT_STATUS(mask) \
-	((NDEBUG & (mask)) ? NCR5380_print_status(instance) : (void)0)
-
-
-
-#endif /* ndef HOSTS_C */
-#endif /* SUN3_NCR5380_H */
+#endif /* SUN3_SCSI_H */
 
diff --git a/drivers/scsi/sun3_scsi_vme.c b/drivers/scsi/sun3_scsi_vme.c
index a3dd55d1d2fd..1eeece6e2040 100644
--- a/drivers/scsi/sun3_scsi_vme.c
+++ b/drivers/scsi/sun3_scsi_vme.c
@@ -1,589 +1,3 @@
- /*
- * Sun3 SCSI stuff by Erik Verbruggen (erik@bigmama.xtdnet.nl)
- *
- * Sun3 DMA routines added by Sam Creasey (sammy@sammy.net)
- *
- * VME support added by Sam Creasey
- *
- * Adapted from sun3_scsi.c -- see there for other headers
- *
- * TODO: modify this driver to support multiple Sun3 SCSI VME boards
- *
- */
-
-#define AUTOSENSE
-
-#include <linux/types.h>
-#include <linux/stddef.h>
-#include <linux/ctype.h>
-#include <linux/delay.h>
-
-#include <linux/module.h>
-#include <linux/signal.h>
-#include <linux/ioport.h>
-#include <linux/init.h>
-#include <linux/blkdev.h>
-
-#include <asm/io.h>
-
-#include <asm/sun3ints.h>
-#include <asm/dvma.h>
-#include <asm/idprom.h>
-#include <asm/machines.h>
-
 #define SUN3_SCSI_VME
 
-#undef SUN3_SCSI_DEBUG
-
-/* dma on! */
-#define REAL_DMA
-
-#define NDEBUG 0
-
-#define NDEBUG_ABORT		0x00100000
-#define NDEBUG_TAGS		0x00200000
-#define NDEBUG_MERGING		0x00400000
-
-#include "scsi.h"
-#include "initio.h"
-#include <scsi/scsi_host.h>
-#include "sun3_scsi.h"
-
-extern int sun3_map_test(unsigned long, char *);
-
-#define USE_WRAPPER
-/*#define RESET_BOOT */
-#define DRIVER_SETUP
-
-/*
- * BUG can be used to trigger a strange code-size related hang on 2.1 kernels
- */
-#ifdef BUG
-#undef RESET_BOOT
-#undef DRIVER_SETUP
-#endif
-
-/* #define SUPPORT_TAGS */
-
-//#define	ENABLE_IRQ()	enable_irq( SUN3_VEC_VMESCSI0 ); 
-#define ENABLE_IRQ()
-
-
-static irqreturn_t scsi_sun3_intr(int irq, void *dummy);
-static inline unsigned char sun3scsi_read(int reg);
-static inline void sun3scsi_write(int reg, int value);
-
-static int setup_can_queue = -1;
-module_param(setup_can_queue, int, 0);
-static int setup_cmd_per_lun = -1;
-module_param(setup_cmd_per_lun, int, 0);
-static int setup_sg_tablesize = -1;
-module_param(setup_sg_tablesize, int, 0);
-#ifdef SUPPORT_TAGS
-static int setup_use_tagged_queuing = -1;
-module_param(setup_use_tagged_queuing, int, 0);
-#endif
-static int setup_hostid = -1;
-module_param(setup_hostid, int, 0);
-
-static struct scsi_cmnd *sun3_dma_setup_done = NULL;
-
-#define	AFTER_RESET_DELAY	(HZ/2)
-
-/* ms to wait after hitting dma regs */
-#define SUN3_DMA_DELAY 10
-
-/* dvma buffer to allocate -- 32k should hopefully be more than sufficient */
-#define SUN3_DVMA_BUFSIZE 0xe000
-
-/* minimum number of bytes to do dma on */
-#define SUN3_DMA_MINSIZE 128
-
-static volatile unsigned char *sun3_scsi_regp;
-static volatile struct sun3_dma_regs *dregs;
-#ifdef OLDDMA
-static unsigned char *dmabuf = NULL; /* dma memory buffer */
-#endif
-static unsigned char *sun3_dma_orig_addr = NULL;
-static unsigned long sun3_dma_orig_count = 0;
-static int sun3_dma_active = 0;
-static unsigned long last_residual = 0;
-
-/*
- * NCR 5380 register access functions
- */
-
-static inline unsigned char sun3scsi_read(int reg)
-{
-	return( sun3_scsi_regp[reg] );
-}
-
-static inline void sun3scsi_write(int reg, int value)
-{
-	sun3_scsi_regp[reg] = value;
-}
-
-/*
- * XXX: status debug
- */
-static struct Scsi_Host *default_instance;
-
-/*
- * Function : int sun3scsi_detect(struct scsi_host_template * tpnt)
- *
- * Purpose : initializes mac NCR5380 driver based on the
- *	command line / compile time port and irq definitions.
- *
- * Inputs : tpnt - template for this SCSI adapter.
- *
- * Returns : 1 if a host adapter was found, 0 if not.
- *
- */
- 
-static int __init sun3scsi_detect(struct scsi_host_template * tpnt)
-{
-	unsigned long ioaddr, irq = 0;
-	static int called = 0;
-	struct Scsi_Host *instance;
-	int i;
-	unsigned long addrs[3] = { IOBASE_SUN3_VMESCSI, 
-				   IOBASE_SUN3_VMESCSI + 0x4000,
-				   0 };
-	unsigned long vecs[3] = { SUN3_VEC_VMESCSI0,
-				  SUN3_VEC_VMESCSI1,
-				  0 };
-	/* check that this machine has an onboard 5380 */
-	switch(idprom->id_machtype) {
-	case SM_SUN3|SM_3_160:
-	case SM_SUN3|SM_3_260:
-		break;
-
-	default:
-		return 0;
-	}
-
-	if(called)
-		return 0;
-
-	tpnt->proc_name = "Sun3 5380 VME SCSI";
-
-	/* setup variables */
-	tpnt->can_queue =
-		(setup_can_queue > 0) ? setup_can_queue : CAN_QUEUE;
-	tpnt->cmd_per_lun =
-		(setup_cmd_per_lun > 0) ? setup_cmd_per_lun : CMD_PER_LUN;
-	tpnt->sg_tablesize = 
-		(setup_sg_tablesize >= 0) ? setup_sg_tablesize : SG_TABLESIZE;
-	
-	if (setup_hostid >= 0)
-		tpnt->this_id = setup_hostid;
-	else {
-		/* use 7 as default */
-		tpnt->this_id = 7;
-	}
-	
-	ioaddr = 0;
-	for(i = 0; addrs[i] != 0; i++) {
-		unsigned char x;
-		
-		ioaddr = (unsigned long)sun3_ioremap(addrs[i], PAGE_SIZE,
-						     SUN3_PAGE_TYPE_VME16);
-		irq = vecs[i];
-		sun3_scsi_regp = (unsigned char *)ioaddr;
-		
-		dregs = (struct sun3_dma_regs *)(((unsigned char *)ioaddr) + 8);
-		
-		if(sun3_map_test((unsigned long)dregs, &x)) {
-			unsigned short oldcsr;
-
-			oldcsr = dregs->csr;
-			dregs->csr = 0;
-			udelay(SUN3_DMA_DELAY);
-			if(dregs->csr == 0x1400)
-				break;
-			
-			dregs->csr = oldcsr;
-		}
-
-		iounmap((void *)ioaddr);
-		ioaddr = 0;
-	}
-
-	if(!ioaddr)
-		return 0;
-	
-#ifdef SUPPORT_TAGS
-	if (setup_use_tagged_queuing < 0)
-		setup_use_tagged_queuing = USE_TAGGED_QUEUING;
-#endif
-
-	instance = scsi_register (tpnt, sizeof(struct NCR5380_hostdata));
-	if(instance == NULL)
-		return 0;
-		
-	default_instance = instance;
-
-        instance->io_port = (unsigned long) ioaddr;
-	instance->irq = irq;
-
-	NCR5380_init(instance, 0);
-
-	instance->n_io_port = 32;
-
-        ((struct NCR5380_hostdata *)instance->hostdata)->ctrl = 0;
-
-	if (request_irq(instance->irq, scsi_sun3_intr,
-			0, "Sun3SCSI-5380VME", instance)) {
-#ifndef REAL_DMA
-		printk("scsi%d: IRQ%d not free, interrupts disabled\n",
-		       instance->host_no, instance->irq);
-		instance->irq = SCSI_IRQ_NONE;
-#else
-		printk("scsi%d: IRQ%d not free, bailing out\n",
-		       instance->host_no, instance->irq);
-		return 0;
-#endif
-	}
-
-	printk("scsi%d: Sun3 5380 VME at port %lX irq", instance->host_no, instance->io_port);
-	if (instance->irq == SCSI_IRQ_NONE)
-		printk ("s disabled");
-	else
-		printk (" %d", instance->irq);
-	printk(" options CAN_QUEUE=%d CMD_PER_LUN=%d release=%d",
-	       instance->can_queue, instance->cmd_per_lun,
-	       SUN3SCSI_PUBLIC_RELEASE);
-	printk("\nscsi%d:", instance->host_no);
-	NCR5380_print_options(instance);
-	printk("\n");
-
-	dregs->csr = 0;
-	udelay(SUN3_DMA_DELAY);
-	dregs->csr = CSR_SCSI | CSR_FIFO | CSR_INTR;
-	udelay(SUN3_DMA_DELAY);
-	dregs->fifo_count = 0;
-	dregs->fifo_count_hi = 0;
-	dregs->dma_addr_hi = 0;
-	dregs->dma_addr_lo = 0;
-	dregs->dma_count_hi = 0;
-	dregs->dma_count_lo = 0;
-
-	dregs->ivect = VME_DATA24 | (instance->irq & 0xff);
-
-	called = 1;
-
-#ifdef RESET_BOOT
-	sun3_scsi_reset_boot(instance);
-#endif
-
-	return 1;
-}
-
-int sun3scsi_release (struct Scsi_Host *shpnt)
-{
-	if (shpnt->irq != SCSI_IRQ_NONE)
-		free_irq(shpnt->irq, shpnt);
-
-	iounmap((void *)sun3_scsi_regp);
-
-	NCR5380_exit(shpnt);
-	return 0;
-}
-
-#ifdef RESET_BOOT
-/*
- * Our 'bus reset on boot' function
- */
-
-static void sun3_scsi_reset_boot(struct Scsi_Host *instance)
-{
-	unsigned long end;
-
-	NCR5380_local_declare();
-	NCR5380_setup(instance);
-	
-	/*
-	 * Do a SCSI reset to clean up the bus during initialization. No
-	 * messing with the queues, interrupts, or locks necessary here.
-	 */
-
-	printk( "Sun3 SCSI: resetting the SCSI bus..." );
-
-	/* switch off SCSI IRQ - catch an interrupt without IRQ bit set else */
-//       	sun3_disable_irq( IRQ_SUN3_SCSI );
-
-	/* get in phase */
-	NCR5380_write( TARGET_COMMAND_REG,
-		      PHASE_SR_TO_TCR( NCR5380_read(STATUS_REG) ));
-
-	/* assert RST */
-	NCR5380_write( INITIATOR_COMMAND_REG, ICR_BASE | ICR_ASSERT_RST );
-
-	/* The min. reset hold time is 25us, so 40us should be enough */
-	udelay( 50 );
-
-	/* reset RST and interrupt */
-	NCR5380_write( INITIATOR_COMMAND_REG, ICR_BASE );
-	NCR5380_read( RESET_PARITY_INTERRUPT_REG );
-
-	for( end = jiffies + AFTER_RESET_DELAY; time_before(jiffies, end); )
-		barrier();
-
-	/* switch on SCSI IRQ again */
-//       	sun3_enable_irq( IRQ_SUN3_SCSI );
-
-	printk( " done\n" );
-}
-#endif
-
-static const char * sun3scsi_info (struct Scsi_Host *spnt) {
-    return "";
-}
-
-// safe bits for the CSR
-#define CSR_GOOD 0x060f
-
-static irqreturn_t scsi_sun3_intr(int irq, void *dummy)
-{
-	unsigned short csr = dregs->csr;
-	int handled = 0;
-
-	dregs->csr &= ~CSR_DMA_ENABLE;
-
-
-#ifdef SUN3_SCSI_DEBUG
-	printk("scsi_intr csr %x\n", csr);
-#endif
-
-	if(csr & ~CSR_GOOD) {
-		if(csr & CSR_DMA_BUSERR) {
-			printk("scsi%d: bus error in dma\n", default_instance->host_no);
-#ifdef SUN3_SCSI_DEBUG
-			printk("scsi: residual %x count %x addr %p dmaaddr %x\n", 
-			       dregs->fifo_count,
-			       dregs->dma_count_lo | (dregs->dma_count_hi << 16),
-			       sun3_dma_orig_addr,
-			       dregs->dma_addr_lo | (dregs->dma_addr_hi << 16));
-#endif
-		}
-
-		if(csr & CSR_DMA_CONFLICT) {
-			printk("scsi%d: dma conflict\n", default_instance->host_no);
-		}
-		handled = 1;
-	}
-
-	if(csr & (CSR_SDB_INT | CSR_DMA_INT)) {
-		NCR5380_intr(irq, dummy);
-		handled = 1;
-	}
-
-	return IRQ_RETVAL(handled);
-}
-
-/*
- * Debug stuff - to be called on NMI, or sysrq key. Use at your own risk; 
- * reentering NCR5380_print_status seems to have ugly side effects
- */
-
-/* this doesn't seem to get used at all -- sam */
-#if 0
-void sun3_sun3_debug (void)
-{
-	unsigned long flags;
-	NCR5380_local_declare();
-
-	if (default_instance) {
-			local_irq_save(flags);
-			NCR5380_print_status(default_instance);
-			local_irq_restore(flags);
-	}
-}
-#endif
-
-
-/* sun3scsi_dma_setup() -- initialize the dma controller for a read/write */
-static unsigned long sun3scsi_dma_setup(void *data, unsigned long count, int write_flag)
-{
-	void *addr;
-
-	if(sun3_dma_orig_addr != NULL)
-		dvma_unmap(sun3_dma_orig_addr);
-
-//	addr = sun3_dvma_page((unsigned long)data, (unsigned long)dmabuf);
-	addr = (void *)dvma_map_vme((unsigned long) data, count);
-		
-	sun3_dma_orig_addr = addr;
-	sun3_dma_orig_count = count;
-	
-#ifdef SUN3_SCSI_DEBUG
-	printk("scsi: dma_setup addr %p count %x\n", addr, count);
-#endif
-
-//	dregs->fifo_count = 0;
-#if 0	
-	/* reset fifo */
-	dregs->csr &= ~CSR_FIFO;
-	dregs->csr |= CSR_FIFO;
-#endif	
-	/* set direction */
-	if(write_flag)
-		dregs->csr |= CSR_SEND;
-	else
-		dregs->csr &= ~CSR_SEND;
-	
-	/* reset fifo */
-//	dregs->csr &= ~CSR_FIFO;
-//	dregs->csr |= CSR_FIFO;
-
-	dregs->csr |= CSR_PACK_ENABLE;
-
-	dregs->dma_addr_hi = ((unsigned long)addr >> 16);
-	dregs->dma_addr_lo = ((unsigned long)addr & 0xffff);
-	
-	dregs->dma_count_hi = 0;
-	dregs->dma_count_lo = 0;
-	dregs->fifo_count_hi = 0;
-	dregs->fifo_count = 0;
-		
-#ifdef SUN3_SCSI_DEBUG
-	printk("scsi: dma_setup done csr %x\n", dregs->csr);
-#endif
-       	return count;
-
-}
-
-static inline unsigned long sun3scsi_dma_residual(struct Scsi_Host *instance)
-{
-	return last_residual;
-}
-
-static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted,
-						  struct scsi_cmnd *cmd,
-						  int write_flag)
-{
-	if (cmd->request->cmd_type == REQ_TYPE_FS)
- 		return wanted;
-	else
-		return 0;
-}
-
-static int sun3scsi_dma_start(unsigned long count, char *data)
-{
-	
-	unsigned short csr;
-
-	csr = dregs->csr;
-#ifdef SUN3_SCSI_DEBUG
-	printk("scsi: dma_start data %p count %x csr %x fifo %x\n", data, count, csr, dregs->fifo_count);
-#endif
-	
-	dregs->dma_count_hi = (sun3_dma_orig_count >> 16);
-	dregs->dma_count_lo = (sun3_dma_orig_count & 0xffff);
-
-	dregs->fifo_count_hi = (sun3_dma_orig_count >> 16);
-	dregs->fifo_count = (sun3_dma_orig_count & 0xffff);
-
-//	if(!(csr & CSR_DMA_ENABLE))
-//		dregs->csr |= CSR_DMA_ENABLE;
-
-	return 0;
-}
-
-/* clean up after our dma is done */
-static int sun3scsi_dma_finish(int write_flag)
-{
-	unsigned short fifo;
-	int ret = 0;
-	
-	sun3_dma_active = 0;
-
-	dregs->csr &= ~CSR_DMA_ENABLE;
-	
-	fifo = dregs->fifo_count;
-	if(write_flag) {
-		if((fifo > 0) && (fifo < sun3_dma_orig_count))
-			fifo++;
-	}
-
-	last_residual = fifo;
-#ifdef SUN3_SCSI_DEBUG
-	printk("scsi: residual %x total %x\n", fifo, sun3_dma_orig_count);
-#endif
-	/* empty bytes from the fifo which didn't make it */
-	if((!write_flag) && (dregs->csr & CSR_LEFT)) {
-		unsigned char *vaddr;
-
-#ifdef SUN3_SCSI_DEBUG
-		printk("scsi: got left over bytes\n");
-#endif
-
-		vaddr = (unsigned char *)dvma_vmetov(sun3_dma_orig_addr);
-		
-		vaddr += (sun3_dma_orig_count - fifo);
-		vaddr--;
-		
-		switch(dregs->csr & CSR_LEFT) {
-		case CSR_LEFT_3:
-			*vaddr = (dregs->bpack_lo & 0xff00) >> 8;
-			vaddr--;
-			
-		case CSR_LEFT_2:
-			*vaddr = (dregs->bpack_hi & 0x00ff);
-			vaddr--;
-			
-		case CSR_LEFT_1:
-			*vaddr = (dregs->bpack_hi & 0xff00) >> 8;
-			break;
-		}
-		
-		
-	}
-
-	dvma_unmap(sun3_dma_orig_addr);
-	sun3_dma_orig_addr = NULL;
-
-	dregs->dma_addr_hi = 0;
-	dregs->dma_addr_lo = 0;
-	dregs->dma_count_hi = 0;
-	dregs->dma_count_lo = 0;
-
-	dregs->fifo_count = 0;
-	dregs->fifo_count_hi = 0;
-
-	dregs->csr &= ~CSR_SEND;
-	
-//	dregs->csr |= CSR_DMA_ENABLE;
-	
-#if 0
-	/* reset fifo */
-	dregs->csr &= ~CSR_FIFO;
-	dregs->csr |= CSR_FIFO;
-#endif	
-	sun3_dma_setup_done = NULL;
-
-	return ret;
-
-}
-
-#include "sun3_NCR5380.c"
-
-static struct scsi_host_template driver_template = {
-	.name			= SUN3_SCSI_NAME,
-	.detect			= sun3scsi_detect,
-	.release		= sun3scsi_release,
-	.info			= sun3scsi_info,
-	.queuecommand		= sun3scsi_queue_command,
-	.eh_abort_handler      	= sun3scsi_abort,
-	.eh_bus_reset_handler  	= sun3scsi_bus_reset,
-	.can_queue		= CAN_QUEUE,
-	.this_id		= 7,
-	.sg_tablesize		= SG_TABLESIZE,
-	.cmd_per_lun		= CMD_PER_LUN,
-	.use_clustering		= DISABLE_CLUSTERING
-};
-
-
-#include "scsi_module.c"
-
-MODULE_LICENSE("GPL");
+#include "sun3_scsi.c"
diff --git a/drivers/scsi/t128.c b/drivers/scsi/t128.c
index a4abce9d526e..8cc80931df14 100644
--- a/drivers/scsi/t128.c
+++ b/drivers/scsi/t128.c
@@ -102,10 +102,6 @@
  * 15 9-11
  */
  
-/*
- * $Log: t128.c,v $
- */
-
 #include <linux/signal.h>
 #include <linux/io.h>
 #include <linux/blkdev.h>
diff --git a/drivers/scsi/t128.h b/drivers/scsi/t128.h
index 1df82c28e56d..fd68cecc62af 100644
--- a/drivers/scsi/t128.h
+++ b/drivers/scsi/t128.h
@@ -34,10 +34,6 @@
  * 1+ (800) 334-5454
  */
 
-/*
- * $Log: t128.h,v $
- */
-
 #ifndef T128_H
 #define T128_H
 
@@ -107,8 +103,6 @@ static int t128_bus_reset(struct scsi_cmnd *);
 #define CAN_QUEUE 32
 #endif
 
-#ifndef HOSTS_C
-
 #define NCR5380_implementation_fields \
     void __iomem *base
 
@@ -148,6 +142,5 @@ static int t128_bus_reset(struct scsi_cmnd *);
 
 #define T128_IRQS 0xc4a8
 
-#endif /* else def HOSTS_C */
 #endif /* ndef ASM */
 #endif /* T128_H */
diff --git a/drivers/scsi/ufs/ufs.h b/drivers/scsi/ufs/ufs.h
index 721050090520..f42d1cee652a 100644
--- a/drivers/scsi/ufs/ufs.h
+++ b/drivers/scsi/ufs/ufs.h
@@ -196,9 +196,9 @@ enum {
  * @dword_2: UPIU header DW-2
  */
 struct utp_upiu_header {
-	u32 dword_0;
-	u32 dword_1;
-	u32 dword_2;
+	__be32 dword_0;
+	__be32 dword_1;
+	__be32 dword_2;
 };
 
 /**
@@ -207,7 +207,7 @@ struct utp_upiu_header {
  * @cdb: Command Descriptor Block CDB DW-4 to DW-7
  */
 struct utp_upiu_cmd {
-	u32 exp_data_transfer_len;
+	__be32 exp_data_transfer_len;
 	u8 cdb[MAX_CDB_SIZE];
 };
 
@@ -228,10 +228,10 @@ struct utp_upiu_query {
 	u8 idn;
 	u8 index;
 	u8 selector;
-	u16 reserved_osf;
-	u16 length;
-	u32 value;
-	u32 reserved[2];
+	__be16 reserved_osf;
+	__be16 length;
+	__be32 value;
+	__be32 reserved[2];
 };
 
 /**
@@ -256,9 +256,9 @@ struct utp_upiu_req {
  * @sense_data: Sense data field DW-8 to DW-12
  */
 struct utp_cmd_rsp {
-	u32 residual_transfer_count;
-	u32 reserved[4];
-	u16 sense_data_len;
+	__be32 residual_transfer_count;
+	__be32 reserved[4];
+	__be16 sense_data_len;
 	u8 sense_data[18];
 };
 
@@ -286,10 +286,10 @@ struct utp_upiu_rsp {
  */
 struct utp_upiu_task_req {
 	struct utp_upiu_header header;
-	u32 input_param1;
-	u32 input_param2;
-	u32 input_param3;
-	u32 reserved[2];
+	__be32 input_param1;
+	__be32 input_param2;
+	__be32 input_param3;
+	__be32 reserved[2];
 };
 
 /**
@@ -301,9 +301,9 @@ struct utp_upiu_task_req {
  */
 struct utp_upiu_task_rsp {
 	struct utp_upiu_header header;
-	u32 output_param1;
-	u32 output_param2;
-	u32 reserved[3];
+	__be32 output_param1;
+	__be32 output_param2;
+	__be32 reserved[3];
 };
 
 /**
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 04884d663e4e..0c2877251251 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -55,6 +55,9 @@
 /* Query request timeout */
 #define QUERY_REQ_TIMEOUT 30 /* msec */
 
+/* Task management command timeout */
+#define TM_CMD_TIMEOUT	100 /* msecs */
+
 /* Expose the flag value from utp_upiu_query.value */
 #define MASK_QUERY_UPIU_FLAG_LOC 0xFF
 
@@ -71,9 +74,22 @@ enum {
 
 /* UFSHCD states */
 enum {
-	UFSHCD_STATE_OPERATIONAL,
 	UFSHCD_STATE_RESET,
 	UFSHCD_STATE_ERROR,
+	UFSHCD_STATE_OPERATIONAL,
+};
+
+/* UFSHCD error handling flags */
+enum {
+	UFSHCD_EH_IN_PROGRESS = (1 << 0),
+};
+
+/* UFSHCD UIC layer error flags */
+enum {
+	UFSHCD_UIC_DL_PA_INIT_ERROR = (1 << 0), /* Data link layer error */
+	UFSHCD_UIC_NL_ERROR = (1 << 1), /* Network layer error */
+	UFSHCD_UIC_TL_ERROR = (1 << 2), /* Transport Layer error */
+	UFSHCD_UIC_DME_ERROR = (1 << 3), /* DME error */
 };
 
 /* Interrupt configuration options */
@@ -83,6 +99,18 @@ enum {
 	UFSHCD_INT_CLEAR,
 };
 
+#define ufshcd_set_eh_in_progress(h) \
+	(h->eh_flags |= UFSHCD_EH_IN_PROGRESS)
+#define ufshcd_eh_in_progress(h) \
+	(h->eh_flags & UFSHCD_EH_IN_PROGRESS)
+#define ufshcd_clear_eh_in_progress(h) \
+	(h->eh_flags &= ~UFSHCD_EH_IN_PROGRESS)
+
+static void ufshcd_tmc_handler(struct ufs_hba *hba);
+static void ufshcd_async_scan(void *data, async_cookie_t cookie);
+static int ufshcd_reset_and_restore(struct ufs_hba *hba);
+static int ufshcd_clear_tm_cmd(struct ufs_hba *hba, int tag);
+
 /*
  * ufshcd_wait_for_register - wait for register value to change
  * @hba - per-adapter interface
@@ -163,7 +191,7 @@ static inline int ufshcd_is_device_present(u32 reg_hcs)
  */
 static inline int ufshcd_get_tr_ocs(struct ufshcd_lrb *lrbp)
 {
-	return lrbp->utr_descriptor_ptr->header.dword_2 & MASK_OCS;
+	return le32_to_cpu(lrbp->utr_descriptor_ptr->header.dword_2) & MASK_OCS;
 }
 
 /**
@@ -176,19 +204,41 @@ static inline int ufshcd_get_tr_ocs(struct ufshcd_lrb *lrbp)
 static inline int
 ufshcd_get_tmr_ocs(struct utp_task_req_desc *task_req_descp)
 {
-	return task_req_descp->header.dword_2 & MASK_OCS;
+	return le32_to_cpu(task_req_descp->header.dword_2) & MASK_OCS;
 }
 
 /**
  * ufshcd_get_tm_free_slot - get a free slot for task management request
  * @hba: per adapter instance
+ * @free_slot: pointer to variable with available slot value
  *
- * Returns maximum number of task management request slots in case of
- * task management queue full or returns the free slot number
+ * Get a free tag and lock it until ufshcd_put_tm_slot() is called.
+ * Returns 0 if free slot is not available, else return 1 with tag value
+ * in @free_slot.
  */
-static inline int ufshcd_get_tm_free_slot(struct ufs_hba *hba)
+static bool ufshcd_get_tm_free_slot(struct ufs_hba *hba, int *free_slot)
 {
-	return find_first_zero_bit(&hba->outstanding_tasks, hba->nutmrs);
+	int tag;
+	bool ret = false;
+
+	if (!free_slot)
+		goto out;
+
+	do {
+		tag = find_first_zero_bit(&hba->tm_slots_in_use, hba->nutmrs);
+		if (tag >= hba->nutmrs)
+			goto out;
+	} while (test_and_set_bit_lock(tag, &hba->tm_slots_in_use));
+
+	*free_slot = tag;
+	ret = true;
+out:
+	return ret;
+}
+
+static inline void ufshcd_put_tm_slot(struct ufs_hba *hba, int slot)
+{
+	clear_bit_unlock(slot, &hba->tm_slots_in_use);
 }
 
 /**
@@ -390,26 +440,6 @@ static inline void ufshcd_copy_sense_data(struct ufshcd_lrb *lrbp)
 }
 
 /**
- * ufshcd_query_to_cpu() - formats the buffer to native cpu endian
- * @response: upiu query response to convert
- */
-static inline void ufshcd_query_to_cpu(struct utp_upiu_query *response)
-{
-	response->length = be16_to_cpu(response->length);
-	response->value = be32_to_cpu(response->value);
-}
-
-/**
- * ufshcd_query_to_be() - formats the buffer to big endian
- * @request: upiu query request to convert
- */
-static inline void ufshcd_query_to_be(struct utp_upiu_query *request)
-{
-	request->length = cpu_to_be16(request->length);
-	request->value = cpu_to_be32(request->value);
-}
-
-/**
  * ufshcd_copy_query_response() - Copy the Query Response and the data
  * descriptor
  * @hba: per adapter instance
@@ -425,7 +455,6 @@ void ufshcd_copy_query_response(struct ufs_hba *hba, struct ufshcd_lrb *lrbp)
 			UPIU_RSP_CODE_OFFSET;
 
 	memcpy(&query_res->upiu_res, &lrbp->ucd_rsp_ptr->qr, QUERY_OSF_SIZE);
-	ufshcd_query_to_cpu(&query_res->upiu_res);
 
 
 	/* Get the descriptor */
@@ -749,7 +778,7 @@ static void ufshcd_prepare_utp_query_req_upiu(struct ufs_hba *hba,
 {
 	struct utp_upiu_req *ucd_req_ptr = lrbp->ucd_req_ptr;
 	struct ufs_query *query = &hba->dev_cmd.query;
-	u16 len = query->request.upiu_req.length;
+	u16 len = be16_to_cpu(query->request.upiu_req.length);
 	u8 *descp = (u8 *)lrbp->ucd_req_ptr + GENERAL_UPIU_REQUEST_SIZE;
 
 	/* Query request header */
@@ -766,7 +795,6 @@ static void ufshcd_prepare_utp_query_req_upiu(struct ufs_hba *hba,
 	/* Copy the Query Request buffer as is */
 	memcpy(&ucd_req_ptr->qr, &query->request.upiu_req,
 			QUERY_OSF_SIZE);
-	ufshcd_query_to_be(&ucd_req_ptr->qr);
 
 	/* Copy the Descriptor */
 	if ((len > 0) && (query->request.upiu_req.opcode ==
@@ -853,10 +881,25 @@ static int ufshcd_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
 
 	tag = cmd->request->tag;
 
-	if (hba->ufshcd_state != UFSHCD_STATE_OPERATIONAL) {
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	switch (hba->ufshcd_state) {
+	case UFSHCD_STATE_OPERATIONAL:
+		break;
+	case UFSHCD_STATE_RESET:
 		err = SCSI_MLQUEUE_HOST_BUSY;
-		goto out;
+		goto out_unlock;
+	case UFSHCD_STATE_ERROR:
+		set_host_byte(cmd, DID_ERROR);
+		cmd->scsi_done(cmd);
+		goto out_unlock;
+	default:
+		dev_WARN_ONCE(hba->dev, 1, "%s: invalid state %d\n",
+				__func__, hba->ufshcd_state);
+		set_host_byte(cmd, DID_BAD_TARGET);
+		cmd->scsi_done(cmd);
+		goto out_unlock;
 	}
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
 
 	/* acquire the tag to make sure device cmds don't use it */
 	if (test_and_set_bit_lock(tag, &hba->lrb_in_use)) {
@@ -893,6 +936,7 @@ static int ufshcd_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
 	/* issue command to the controller */
 	spin_lock_irqsave(hba->host->host_lock, flags);
 	ufshcd_send_command(hba, tag);
+out_unlock:
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
 out:
 	return err;
@@ -1151,7 +1195,7 @@ static int ufshcd_query_flag(struct ufs_hba *hba, enum query_opcode opcode,
 	}
 
 	if (flag_res)
-		*flag_res = (response->upiu_res.value &
+		*flag_res = (be32_to_cpu(response->upiu_res.value) &
 				MASK_QUERY_UPIU_FLAG_LOC) & 0x1;
 
 out_unlock:
@@ -1170,7 +1214,7 @@ out_unlock:
  *
  * Returns 0 for success, non-zero in case of failure
 */
-int ufshcd_query_attr(struct ufs_hba *hba, enum query_opcode opcode,
+static int ufshcd_query_attr(struct ufs_hba *hba, enum query_opcode opcode,
 			enum attr_idn idn, u8 index, u8 selector, u32 *attr_val)
 {
 	struct ufs_query_req *request;
@@ -1195,7 +1239,7 @@ int ufshcd_query_attr(struct ufs_hba *hba, enum query_opcode opcode,
 	switch (opcode) {
 	case UPIU_QUERY_OPCODE_WRITE_ATTR:
 		request->query_func = UPIU_QUERY_FUNC_STANDARD_WRITE_REQUEST;
-		request->upiu_req.value = *attr_val;
+		request->upiu_req.value = cpu_to_be32(*attr_val);
 		break;
 	case UPIU_QUERY_OPCODE_READ_ATTR:
 		request->query_func = UPIU_QUERY_FUNC_STANDARD_READ_REQUEST;
@@ -1222,7 +1266,7 @@ int ufshcd_query_attr(struct ufs_hba *hba, enum query_opcode opcode,
 		goto out_unlock;
 	}
 
-	*attr_val = response->upiu_res.value;
+	*attr_val = be32_to_cpu(response->upiu_res.value);
 
 out_unlock:
 	mutex_unlock(&hba->dev_cmd.lock);
@@ -1481,7 +1525,7 @@ EXPORT_SYMBOL_GPL(ufshcd_dme_get_attr);
  *
  * Returns 0 on success, non-zero value on failure
  */
-int ufshcd_uic_change_pwr_mode(struct ufs_hba *hba, u8 mode)
+static int ufshcd_uic_change_pwr_mode(struct ufs_hba *hba, u8 mode)
 {
 	struct uic_command uic_cmd = {0};
 	struct completion pwr_done;
@@ -1701,11 +1745,6 @@ static int ufshcd_make_hba_operational(struct ufs_hba *hba)
 		goto out;
 	}
 
-	if (hba->ufshcd_state == UFSHCD_STATE_RESET)
-		scsi_unblock_requests(hba->host);
-
-	hba->ufshcd_state = UFSHCD_STATE_OPERATIONAL;
-
 out:
 	return err;
 }
@@ -1831,66 +1870,6 @@ static int ufshcd_verify_dev_init(struct ufs_hba *hba)
 }
 
 /**
- * ufshcd_do_reset - reset the host controller
- * @hba: per adapter instance
- *
- * Returns SUCCESS/FAILED
- */
-static int ufshcd_do_reset(struct ufs_hba *hba)
-{
-	struct ufshcd_lrb *lrbp;
-	unsigned long flags;
-	int tag;
-
-	/* block commands from midlayer */
-	scsi_block_requests(hba->host);
-
-	spin_lock_irqsave(hba->host->host_lock, flags);
-	hba->ufshcd_state = UFSHCD_STATE_RESET;
-
-	/* send controller to reset state */
-	ufshcd_hba_stop(hba);
-	spin_unlock_irqrestore(hba->host->host_lock, flags);
-
-	/* abort outstanding commands */
-	for (tag = 0; tag < hba->nutrs; tag++) {
-		if (test_bit(tag, &hba->outstanding_reqs)) {
-			lrbp = &hba->lrb[tag];
-			if (lrbp->cmd) {
-				scsi_dma_unmap(lrbp->cmd);
-				lrbp->cmd->result = DID_RESET << 16;
-				lrbp->cmd->scsi_done(lrbp->cmd);
-				lrbp->cmd = NULL;
-				clear_bit_unlock(tag, &hba->lrb_in_use);
-			}
-		}
-	}
-
-	/* complete device management command */
-	if (hba->dev_cmd.complete)
-		complete(hba->dev_cmd.complete);
-
-	/* clear outstanding request/task bit maps */
-	hba->outstanding_reqs = 0;
-	hba->outstanding_tasks = 0;
-
-	/* Host controller enable */
-	if (ufshcd_hba_enable(hba)) {
-		dev_err(hba->dev,
-			"Reset: Controller initialization failed\n");
-		return FAILED;
-	}
-
-	if (ufshcd_link_startup(hba)) {
-		dev_err(hba->dev,
-			"Reset: Link start-up failed\n");
-		return FAILED;
-	}
-
-	return SUCCESS;
-}
-
-/**
  * ufshcd_slave_alloc - handle initial SCSI device configurations
  * @sdev: pointer to SCSI device
  *
@@ -1907,6 +1886,9 @@ static int ufshcd_slave_alloc(struct scsi_device *sdev)
 	sdev->use_10_for_ms = 1;
 	scsi_set_tag_type(sdev, MSG_SIMPLE_TAG);
 
+	/* allow SCSI layer to restart the device in case of errors */
+	sdev->allow_restart = 1;
+
 	/*
 	 * Inform SCSI Midlayer that the LUN queue depth is same as the
 	 * controller queue depth. If a LUN queue depth is less than the
@@ -1934,10 +1916,11 @@ static void ufshcd_slave_destroy(struct scsi_device *sdev)
  * ufshcd_task_req_compl - handle task management request completion
  * @hba: per adapter instance
  * @index: index of the completed request
+ * @resp: task management service response
  *
- * Returns SUCCESS/FAILED
+ * Returns non-zero value on error, zero on success
  */
-static int ufshcd_task_req_compl(struct ufs_hba *hba, u32 index)
+static int ufshcd_task_req_compl(struct ufs_hba *hba, u32 index, u8 *resp)
 {
 	struct utp_task_req_desc *task_req_descp;
 	struct utp_upiu_task_rsp *task_rsp_upiup;
@@ -1958,19 +1941,15 @@ static int ufshcd_task_req_compl(struct ufs_hba *hba, u32 index)
 				task_req_descp[index].task_rsp_upiu;
 		task_result = be32_to_cpu(task_rsp_upiup->header.dword_1);
 		task_result = ((task_result & MASK_TASK_RESPONSE) >> 8);
-
-		if (task_result != UPIU_TASK_MANAGEMENT_FUNC_COMPL &&
-		    task_result != UPIU_TASK_MANAGEMENT_FUNC_SUCCEEDED)
-			task_result = FAILED;
-		else
-			task_result = SUCCESS;
+		if (resp)
+			*resp = (u8)task_result;
 	} else {
-		task_result = FAILED;
-		dev_err(hba->dev,
-			"trc: Invalid ocs = %x\n", ocs_value);
+		dev_err(hba->dev, "%s: failed, ocs = 0x%x\n",
+				__func__, ocs_value);
 	}
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
-	return task_result;
+
+	return ocs_value;
 }
 
 /**
@@ -2105,6 +2084,9 @@ ufshcd_transfer_rsp_status(struct ufs_hba *hba, struct ufshcd_lrb *lrbp)
 	case OCS_ABORTED:
 		result |= DID_ABORT << 16;
 		break;
+	case OCS_INVALID_COMMAND_STATUS:
+		result |= DID_REQUEUE << 16;
+		break;
 	case OCS_INVALID_CMD_TABLE_ATTR:
 	case OCS_INVALID_PRDT_ATTR:
 	case OCS_MISMATCH_DATA_BUF_SIZE:
@@ -2422,41 +2404,145 @@ out:
 }
 
 /**
- * ufshcd_fatal_err_handler - handle fatal errors
- * @hba: per adapter instance
+ * ufshcd_err_handler - handle UFS errors that require s/w attention
+ * @work: pointer to work structure
  */
-static void ufshcd_fatal_err_handler(struct work_struct *work)
+static void ufshcd_err_handler(struct work_struct *work)
 {
 	struct ufs_hba *hba;
-	hba = container_of(work, struct ufs_hba, feh_workq);
+	unsigned long flags;
+	u32 err_xfer = 0;
+	u32 err_tm = 0;
+	int err = 0;
+	int tag;
+
+	hba = container_of(work, struct ufs_hba, eh_work);
 
 	pm_runtime_get_sync(hba->dev);
-	/* check if reset is already in progress */
-	if (hba->ufshcd_state != UFSHCD_STATE_RESET)
-		ufshcd_do_reset(hba);
+
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	if (hba->ufshcd_state == UFSHCD_STATE_RESET) {
+		spin_unlock_irqrestore(hba->host->host_lock, flags);
+		goto out;
+	}
+
+	hba->ufshcd_state = UFSHCD_STATE_RESET;
+	ufshcd_set_eh_in_progress(hba);
+
+	/* Complete requests that have door-bell cleared by h/w */
+	ufshcd_transfer_req_compl(hba);
+	ufshcd_tmc_handler(hba);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
+	/* Clear pending transfer requests */
+	for_each_set_bit(tag, &hba->outstanding_reqs, hba->nutrs)
+		if (ufshcd_clear_cmd(hba, tag))
+			err_xfer |= 1 << tag;
+
+	/* Clear pending task management requests */
+	for_each_set_bit(tag, &hba->outstanding_tasks, hba->nutmrs)
+		if (ufshcd_clear_tm_cmd(hba, tag))
+			err_tm |= 1 << tag;
+
+	/* Complete the requests that are cleared by s/w */
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	ufshcd_transfer_req_compl(hba);
+	ufshcd_tmc_handler(hba);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
+	/* Fatal errors need reset */
+	if (err_xfer || err_tm || (hba->saved_err & INT_FATAL_ERRORS) ||
+			((hba->saved_err & UIC_ERROR) &&
+			 (hba->saved_uic_err & UFSHCD_UIC_DL_PA_INIT_ERROR))) {
+		err = ufshcd_reset_and_restore(hba);
+		if (err) {
+			dev_err(hba->dev, "%s: reset and restore failed\n",
+					__func__);
+			hba->ufshcd_state = UFSHCD_STATE_ERROR;
+		}
+		/*
+		 * Inform scsi mid-layer that we did reset and allow to handle
+		 * Unit Attention properly.
+		 */
+		scsi_report_bus_reset(hba->host, 0);
+		hba->saved_err = 0;
+		hba->saved_uic_err = 0;
+	}
+	ufshcd_clear_eh_in_progress(hba);
+
+out:
+	scsi_unblock_requests(hba->host);
 	pm_runtime_put_sync(hba->dev);
 }
 
 /**
- * ufshcd_err_handler - Check for fatal errors
- * @work: pointer to a work queue structure
+ * ufshcd_update_uic_error - check and set fatal UIC error flags.
+ * @hba: per-adapter instance
  */
-static void ufshcd_err_handler(struct ufs_hba *hba)
+static void ufshcd_update_uic_error(struct ufs_hba *hba)
 {
 	u32 reg;
 
+	/* PA_INIT_ERROR is fatal and needs UIC reset */
+	reg = ufshcd_readl(hba, REG_UIC_ERROR_CODE_DATA_LINK_LAYER);
+	if (reg & UIC_DATA_LINK_LAYER_ERROR_PA_INIT)
+		hba->uic_error |= UFSHCD_UIC_DL_PA_INIT_ERROR;
+
+	/* UIC NL/TL/DME errors needs software retry */
+	reg = ufshcd_readl(hba, REG_UIC_ERROR_CODE_NETWORK_LAYER);
+	if (reg)
+		hba->uic_error |= UFSHCD_UIC_NL_ERROR;
+
+	reg = ufshcd_readl(hba, REG_UIC_ERROR_CODE_TRANSPORT_LAYER);
+	if (reg)
+		hba->uic_error |= UFSHCD_UIC_TL_ERROR;
+
+	reg = ufshcd_readl(hba, REG_UIC_ERROR_CODE_DME);
+	if (reg)
+		hba->uic_error |= UFSHCD_UIC_DME_ERROR;
+
+	dev_dbg(hba->dev, "%s: UIC error flags = 0x%08x\n",
+			__func__, hba->uic_error);
+}
+
+/**
+ * ufshcd_check_errors - Check for errors that need s/w attention
+ * @hba: per-adapter instance
+ */
+static void ufshcd_check_errors(struct ufs_hba *hba)
+{
+	bool queue_eh_work = false;
+
 	if (hba->errors & INT_FATAL_ERRORS)
-		goto fatal_eh;
+		queue_eh_work = true;
 
 	if (hba->errors & UIC_ERROR) {
-		reg = ufshcd_readl(hba, REG_UIC_ERROR_CODE_DATA_LINK_LAYER);
-		if (reg & UIC_DATA_LINK_LAYER_ERROR_PA_INIT)
-			goto fatal_eh;
+		hba->uic_error = 0;
+		ufshcd_update_uic_error(hba);
+		if (hba->uic_error)
+			queue_eh_work = true;
 	}
-	return;
-fatal_eh:
-	hba->ufshcd_state = UFSHCD_STATE_ERROR;
-	schedule_work(&hba->feh_workq);
+
+	if (queue_eh_work) {
+		/* handle fatal errors only when link is functional */
+		if (hba->ufshcd_state == UFSHCD_STATE_OPERATIONAL) {
+			/* block commands from scsi mid-layer */
+			scsi_block_requests(hba->host);
+
+			/* transfer error masks to sticky bits */
+			hba->saved_err |= hba->errors;
+			hba->saved_uic_err |= hba->uic_error;
+
+			hba->ufshcd_state = UFSHCD_STATE_ERROR;
+			schedule_work(&hba->eh_work);
+		}
+	}
+	/*
+	 * if (!queue_eh_work) -
+	 * Other errors are either non-fatal where host recovers
+	 * itself without s/w intervention or errors that will be
+	 * handled by the SCSI core layer.
+	 */
 }
 
 /**
@@ -2469,7 +2555,7 @@ static void ufshcd_tmc_handler(struct ufs_hba *hba)
 
 	tm_doorbell = ufshcd_readl(hba, REG_UTP_TASK_REQ_DOOR_BELL);
 	hba->tm_condition = tm_doorbell ^ hba->outstanding_tasks;
-	wake_up_interruptible(&hba->ufshcd_tm_wait_queue);
+	wake_up(&hba->tm_wq);
 }
 
 /**
@@ -2481,7 +2567,7 @@ static void ufshcd_sl_intr(struct ufs_hba *hba, u32 intr_status)
 {
 	hba->errors = UFSHCD_ERROR_MASK & intr_status;
 	if (hba->errors)
-		ufshcd_err_handler(hba);
+		ufshcd_check_errors(hba);
 
 	if (intr_status & UFSHCD_UIC_MASK)
 		ufshcd_uic_cmd_compl(hba, intr_status);
@@ -2519,38 +2605,58 @@ static irqreturn_t ufshcd_intr(int irq, void *__hba)
 	return retval;
 }
 
+static int ufshcd_clear_tm_cmd(struct ufs_hba *hba, int tag)
+{
+	int err = 0;
+	u32 mask = 1 << tag;
+	unsigned long flags;
+
+	if (!test_bit(tag, &hba->outstanding_tasks))
+		goto out;
+
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	ufshcd_writel(hba, ~(1 << tag), REG_UTP_TASK_REQ_LIST_CLEAR);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
+	/* poll for max. 1 sec to clear door bell register by h/w */
+	err = ufshcd_wait_for_register(hba,
+			REG_UTP_TASK_REQ_DOOR_BELL,
+			mask, 0, 1000, 1000);
+out:
+	return err;
+}
+
 /**
  * ufshcd_issue_tm_cmd - issues task management commands to controller
  * @hba: per adapter instance
- * @lrbp: pointer to local reference block
+ * @lun_id: LUN ID to which TM command is sent
+ * @task_id: task ID to which the TM command is applicable
+ * @tm_function: task management function opcode
+ * @tm_response: task management service response return value
  *
- * Returns SUCCESS/FAILED
+ * Returns non-zero value on error, zero on success.
  */
-static int
-ufshcd_issue_tm_cmd(struct ufs_hba *hba,
-		    struct ufshcd_lrb *lrbp,
-		    u8 tm_function)
+static int ufshcd_issue_tm_cmd(struct ufs_hba *hba, int lun_id, int task_id,
+		u8 tm_function, u8 *tm_response)
 {
 	struct utp_task_req_desc *task_req_descp;
 	struct utp_upiu_task_req *task_req_upiup;
 	struct Scsi_Host *host;
 	unsigned long flags;
-	int free_slot = 0;
+	int free_slot;
 	int err;
+	int task_tag;
 
 	host = hba->host;
 
-	spin_lock_irqsave(host->host_lock, flags);
-
-	/* If task management queue is full */
-	free_slot = ufshcd_get_tm_free_slot(hba);
-	if (free_slot >= hba->nutmrs) {
-		spin_unlock_irqrestore(host->host_lock, flags);
-		dev_err(hba->dev, "Task management queue full\n");
-		err = FAILED;
-		goto out;
-	}
+	/*
+	 * Get free slot, sleep if slots are unavailable.
+	 * Even though we use wait_event() which sleeps indefinitely,
+	 * the maximum wait time is bounded by %TM_CMD_TIMEOUT.
+	 */
+	wait_event(hba->tm_tag_wq, ufshcd_get_tm_free_slot(hba, &free_slot));
 
+	spin_lock_irqsave(host->host_lock, flags);
 	task_req_descp = hba->utmrdl_base_addr;
 	task_req_descp += free_slot;
 
@@ -2562,18 +2668,15 @@ ufshcd_issue_tm_cmd(struct ufs_hba *hba,
 	/* Configure task request UPIU */
 	task_req_upiup =
 		(struct utp_upiu_task_req *) task_req_descp->task_req_upiu;
+	task_tag = hba->nutrs + free_slot;
 	task_req_upiup->header.dword_0 =
 		UPIU_HEADER_DWORD(UPIU_TRANSACTION_TASK_REQ, 0,
-					      lrbp->lun, lrbp->task_tag);
+					      lun_id, task_tag);
 	task_req_upiup->header.dword_1 =
 		UPIU_HEADER_DWORD(0, tm_function, 0, 0);
 
-	task_req_upiup->input_param1 = lrbp->lun;
-	task_req_upiup->input_param1 =
-		cpu_to_be32(task_req_upiup->input_param1);
-	task_req_upiup->input_param2 = lrbp->task_tag;
-	task_req_upiup->input_param2 =
-		cpu_to_be32(task_req_upiup->input_param2);
+	task_req_upiup->input_param1 = cpu_to_be32(lun_id);
+	task_req_upiup->input_param2 = cpu_to_be32(task_id);
 
 	/* send command to the controller */
 	__set_bit(free_slot, &hba->outstanding_tasks);
@@ -2582,91 +2685,88 @@ ufshcd_issue_tm_cmd(struct ufs_hba *hba,
 	spin_unlock_irqrestore(host->host_lock, flags);
 
 	/* wait until the task management command is completed */
-	err =
-	wait_event_interruptible_timeout(hba->ufshcd_tm_wait_queue,
-					 (test_bit(free_slot,
-					 &hba->tm_condition) != 0),
-					 60 * HZ);
+	err = wait_event_timeout(hba->tm_wq,
+			test_bit(free_slot, &hba->tm_condition),
+			msecs_to_jiffies(TM_CMD_TIMEOUT));
 	if (!err) {
-		dev_err(hba->dev,
-			"Task management command timed-out\n");
-		err = FAILED;
-		goto out;
+		dev_err(hba->dev, "%s: task management cmd 0x%.2x timed-out\n",
+				__func__, tm_function);
+		if (ufshcd_clear_tm_cmd(hba, free_slot))
+			dev_WARN(hba->dev, "%s: unable clear tm cmd (slot %d) after timeout\n",
+					__func__, free_slot);
+		err = -ETIMEDOUT;
+	} else {
+		err = ufshcd_task_req_compl(hba, free_slot, tm_response);
 	}
+
 	clear_bit(free_slot, &hba->tm_condition);
-	err = ufshcd_task_req_compl(hba, free_slot);
-out:
+	ufshcd_put_tm_slot(hba, free_slot);
+	wake_up(&hba->tm_tag_wq);
+
 	return err;
 }
 
 /**
- * ufshcd_device_reset - reset device and abort all the pending commands
+ * ufshcd_eh_device_reset_handler - device reset handler registered to
+ *                                    scsi layer.
  * @cmd: SCSI command pointer
  *
  * Returns SUCCESS/FAILED
  */
-static int ufshcd_device_reset(struct scsi_cmnd *cmd)
+static int ufshcd_eh_device_reset_handler(struct scsi_cmnd *cmd)
 {
 	struct Scsi_Host *host;
 	struct ufs_hba *hba;
 	unsigned int tag;
 	u32 pos;
 	int err;
+	u8 resp = 0xF;
+	struct ufshcd_lrb *lrbp;
+	unsigned long flags;
 
 	host = cmd->device->host;
 	hba = shost_priv(host);
 	tag = cmd->request->tag;
 
-	err = ufshcd_issue_tm_cmd(hba, &hba->lrb[tag], UFS_LOGICAL_RESET);
-	if (err == FAILED)
+	lrbp = &hba->lrb[tag];
+	err = ufshcd_issue_tm_cmd(hba, lrbp->lun, 0, UFS_LOGICAL_RESET, &resp);
+	if (err || resp != UPIU_TASK_MANAGEMENT_FUNC_COMPL) {
+		if (!err)
+			err = resp;
 		goto out;
+	}
 
-	for (pos = 0; pos < hba->nutrs; pos++) {
-		if (test_bit(pos, &hba->outstanding_reqs) &&
-		    (hba->lrb[tag].lun == hba->lrb[pos].lun)) {
-
-			/* clear the respective UTRLCLR register bit */
-			ufshcd_utrl_clear(hba, pos);
-
-			clear_bit(pos, &hba->outstanding_reqs);
-
-			if (hba->lrb[pos].cmd) {
-				scsi_dma_unmap(hba->lrb[pos].cmd);
-				hba->lrb[pos].cmd->result =
-					DID_ABORT << 16;
-				hba->lrb[pos].cmd->scsi_done(cmd);
-				hba->lrb[pos].cmd = NULL;
-				clear_bit_unlock(pos, &hba->lrb_in_use);
-				wake_up(&hba->dev_cmd.tag_wq);
-			}
+	/* clear the commands that were pending for corresponding LUN */
+	for_each_set_bit(pos, &hba->outstanding_reqs, hba->nutrs) {
+		if (hba->lrb[pos].lun == lrbp->lun) {
+			err = ufshcd_clear_cmd(hba, pos);
+			if (err)
+				break;
 		}
-	} /* end of for */
+	}
+	spin_lock_irqsave(host->host_lock, flags);
+	ufshcd_transfer_req_compl(hba);
+	spin_unlock_irqrestore(host->host_lock, flags);
 out:
+	if (!err) {
+		err = SUCCESS;
+	} else {
+		dev_err(hba->dev, "%s: failed with err %d\n", __func__, err);
+		err = FAILED;
+	}
 	return err;
 }
 
 /**
- * ufshcd_host_reset - Main reset function registered with scsi layer
- * @cmd: SCSI command pointer
- *
- * Returns SUCCESS/FAILED
- */
-static int ufshcd_host_reset(struct scsi_cmnd *cmd)
-{
-	struct ufs_hba *hba;
-
-	hba = shost_priv(cmd->device->host);
-
-	if (hba->ufshcd_state == UFSHCD_STATE_RESET)
-		return SUCCESS;
-
-	return ufshcd_do_reset(hba);
-}
-
-/**
  * ufshcd_abort - abort a specific command
  * @cmd: SCSI command pointer
  *
+ * Abort the pending command in device by sending UFS_ABORT_TASK task management
+ * command, and in host controller by clearing the door-bell register. There can
+ * be race between controller sending the command to the device while abort is
+ * issued. To avoid that, first issue UFS_QUERY_TASK to check if the command is
+ * really issued and then try to abort it.
+ *
  * Returns SUCCESS/FAILED
  */
 static int ufshcd_abort(struct scsi_cmnd *cmd)
@@ -2675,33 +2775,68 @@ static int ufshcd_abort(struct scsi_cmnd *cmd)
 	struct ufs_hba *hba;
 	unsigned long flags;
 	unsigned int tag;
-	int err;
+	int err = 0;
+	int poll_cnt;
+	u8 resp = 0xF;
+	struct ufshcd_lrb *lrbp;
 
 	host = cmd->device->host;
 	hba = shost_priv(host);
 	tag = cmd->request->tag;
 
-	spin_lock_irqsave(host->host_lock, flags);
+	/* If command is already aborted/completed, return SUCCESS */
+	if (!(test_bit(tag, &hba->outstanding_reqs)))
+		goto out;
 
-	/* check if command is still pending */
-	if (!(test_bit(tag, &hba->outstanding_reqs))) {
-		err = FAILED;
-		spin_unlock_irqrestore(host->host_lock, flags);
+	lrbp = &hba->lrb[tag];
+	for (poll_cnt = 100; poll_cnt; poll_cnt--) {
+		err = ufshcd_issue_tm_cmd(hba, lrbp->lun, lrbp->task_tag,
+				UFS_QUERY_TASK, &resp);
+		if (!err && resp == UPIU_TASK_MANAGEMENT_FUNC_SUCCEEDED) {
+			/* cmd pending in the device */
+			break;
+		} else if (!err && resp == UPIU_TASK_MANAGEMENT_FUNC_COMPL) {
+			u32 reg;
+
+			/*
+			 * cmd not pending in the device, check if it is
+			 * in transition.
+			 */
+			reg = ufshcd_readl(hba, REG_UTP_TRANSFER_REQ_DOOR_BELL);
+			if (reg & (1 << tag)) {
+				/* sleep for max. 200us to stabilize */
+				usleep_range(100, 200);
+				continue;
+			}
+			/* command completed already */
+			goto out;
+		} else {
+			if (!err)
+				err = resp; /* service response error */
+			goto out;
+		}
+	}
+
+	if (!poll_cnt) {
+		err = -EBUSY;
 		goto out;
 	}
-	spin_unlock_irqrestore(host->host_lock, flags);
 
-	err = ufshcd_issue_tm_cmd(hba, &hba->lrb[tag], UFS_ABORT_TASK);
-	if (err == FAILED)
+	err = ufshcd_issue_tm_cmd(hba, lrbp->lun, lrbp->task_tag,
+			UFS_ABORT_TASK, &resp);
+	if (err || resp != UPIU_TASK_MANAGEMENT_FUNC_COMPL) {
+		if (!err)
+			err = resp; /* service response error */
+		goto out;
+	}
+
+	err = ufshcd_clear_cmd(hba, tag);
+	if (err)
 		goto out;
 
 	scsi_dma_unmap(cmd);
 
 	spin_lock_irqsave(host->host_lock, flags);
-
-	/* clear the respective UTRLCLR register bit */
-	ufshcd_utrl_clear(hba, tag);
-
 	__clear_bit(tag, &hba->outstanding_reqs);
 	hba->lrb[tag].cmd = NULL;
 	spin_unlock_irqrestore(host->host_lock, flags);
@@ -2709,6 +2844,129 @@ static int ufshcd_abort(struct scsi_cmnd *cmd)
 	clear_bit_unlock(tag, &hba->lrb_in_use);
 	wake_up(&hba->dev_cmd.tag_wq);
 out:
+	if (!err) {
+		err = SUCCESS;
+	} else {
+		dev_err(hba->dev, "%s: failed with err %d\n", __func__, err);
+		err = FAILED;
+	}
+
+	return err;
+}
+
+/**
+ * ufshcd_host_reset_and_restore - reset and restore host controller
+ * @hba: per-adapter instance
+ *
+ * Note that host controller reset may issue DME_RESET to
+ * local and remote (device) Uni-Pro stack and the attributes
+ * are reset to default state.
+ *
+ * Returns zero on success, non-zero on failure
+ */
+static int ufshcd_host_reset_and_restore(struct ufs_hba *hba)
+{
+	int err;
+	async_cookie_t cookie;
+	unsigned long flags;
+
+	/* Reset the host controller */
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	ufshcd_hba_stop(hba);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
+	err = ufshcd_hba_enable(hba);
+	if (err)
+		goto out;
+
+	/* Establish the link again and restore the device */
+	cookie = async_schedule(ufshcd_async_scan, hba);
+	/* wait for async scan to be completed */
+	async_synchronize_cookie(++cookie);
+	if (hba->ufshcd_state != UFSHCD_STATE_OPERATIONAL)
+		err = -EIO;
+out:
+	if (err)
+		dev_err(hba->dev, "%s: Host init failed %d\n", __func__, err);
+
+	return err;
+}
+
+/**
+ * ufshcd_reset_and_restore - reset and re-initialize host/device
+ * @hba: per-adapter instance
+ *
+ * Reset and recover device, host and re-establish link. This
+ * is helpful to recover the communication in fatal error conditions.
+ *
+ * Returns zero on success, non-zero on failure
+ */
+static int ufshcd_reset_and_restore(struct ufs_hba *hba)
+{
+	int err = 0;
+	unsigned long flags;
+
+	err = ufshcd_host_reset_and_restore(hba);
+
+	/*
+	 * After reset the door-bell might be cleared, complete
+	 * outstanding requests in s/w here.
+	 */
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	ufshcd_transfer_req_compl(hba);
+	ufshcd_tmc_handler(hba);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
+	return err;
+}
+
+/**
+ * ufshcd_eh_host_reset_handler - host reset handler registered to scsi layer
+ * @cmd - SCSI command pointer
+ *
+ * Returns SUCCESS/FAILED
+ */
+static int ufshcd_eh_host_reset_handler(struct scsi_cmnd *cmd)
+{
+	int err;
+	unsigned long flags;
+	struct ufs_hba *hba;
+
+	hba = shost_priv(cmd->device->host);
+
+	/*
+	 * Check if there is any race with fatal error handling.
+	 * If so, wait for it to complete. Even though fatal error
+	 * handling does reset and restore in some cases, don't assume
+	 * anything out of it. We are just avoiding race here.
+	 */
+	do {
+		spin_lock_irqsave(hba->host->host_lock, flags);
+		if (!(work_pending(&hba->eh_work) ||
+				hba->ufshcd_state == UFSHCD_STATE_RESET))
+			break;
+		spin_unlock_irqrestore(hba->host->host_lock, flags);
+		dev_dbg(hba->dev, "%s: reset in progress\n", __func__);
+		flush_work(&hba->eh_work);
+	} while (1);
+
+	hba->ufshcd_state = UFSHCD_STATE_RESET;
+	ufshcd_set_eh_in_progress(hba);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
+	err = ufshcd_reset_and_restore(hba);
+
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	if (!err) {
+		err = SUCCESS;
+		hba->ufshcd_state = UFSHCD_STATE_OPERATIONAL;
+	} else {
+		err = FAILED;
+		hba->ufshcd_state = UFSHCD_STATE_ERROR;
+	}
+	ufshcd_clear_eh_in_progress(hba);
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+
 	return err;
 }
 
@@ -2737,8 +2995,13 @@ static void ufshcd_async_scan(void *data, async_cookie_t cookie)
 		goto out;
 
 	ufshcd_force_reset_auto_bkops(hba);
-	scsi_scan_host(hba->host);
-	pm_runtime_put_sync(hba->dev);
+	hba->ufshcd_state = UFSHCD_STATE_OPERATIONAL;
+
+	/* If we are in error handling context no need to scan the host */
+	if (!ufshcd_eh_in_progress(hba)) {
+		scsi_scan_host(hba->host);
+		pm_runtime_put_sync(hba->dev);
+	}
 out:
 	return;
 }
@@ -2751,8 +3014,8 @@ static struct scsi_host_template ufshcd_driver_template = {
 	.slave_alloc		= ufshcd_slave_alloc,
 	.slave_destroy		= ufshcd_slave_destroy,
 	.eh_abort_handler	= ufshcd_abort,
-	.eh_device_reset_handler = ufshcd_device_reset,
-	.eh_host_reset_handler	= ufshcd_host_reset,
+	.eh_device_reset_handler = ufshcd_eh_device_reset_handler,
+	.eh_host_reset_handler   = ufshcd_eh_host_reset_handler,
 	.this_id		= -1,
 	.sg_tablesize		= SG_ALL,
 	.cmd_per_lun		= UFSHCD_CMD_PER_LUN,
@@ -2916,10 +3179,11 @@ int ufshcd_init(struct device *dev, struct ufs_hba **hba_handle,
 	host->max_cmd_len = MAX_CDB_SIZE;
 
 	/* Initailize wait queue for task management */
-	init_waitqueue_head(&hba->ufshcd_tm_wait_queue);
+	init_waitqueue_head(&hba->tm_wq);
+	init_waitqueue_head(&hba->tm_tag_wq);
 
 	/* Initialize work queues */
-	INIT_WORK(&hba->feh_workq, ufshcd_fatal_err_handler);
+	INIT_WORK(&hba->eh_work, ufshcd_err_handler);
 	INIT_WORK(&hba->eeh_work, ufshcd_exception_event_handler);
 
 	/* Initialize UIC command mutex */
diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h
index 577679a2d189..acf318e338ed 100644
--- a/drivers/scsi/ufs/ufshcd.h
+++ b/drivers/scsi/ufs/ufshcd.h
@@ -174,15 +174,21 @@ struct ufs_dev_cmd {
  * @irq: Irq number of the controller
  * @active_uic_cmd: handle of active UIC command
  * @uic_cmd_mutex: mutex for uic command
- * @ufshcd_tm_wait_queue: wait queue for task management
+ * @tm_wq: wait queue for task management
+ * @tm_tag_wq: wait queue for free task management slots
+ * @tm_slots_in_use: bit map of task management request slots in use
  * @pwr_done: completion for power mode change
  * @tm_condition: condition variable for task management
  * @ufshcd_state: UFSHCD states
+ * @eh_flags: Error handling flags
  * @intr_mask: Interrupt Mask Bits
  * @ee_ctrl_mask: Exception event control mask
- * @feh_workq: Work queue for fatal controller error handling
+ * @eh_work: Worker to handle UFS errors that require s/w attention
  * @eeh_work: Worker to handle exception events
  * @errors: HBA errors
+ * @uic_error: UFS interconnect layer error status
+ * @saved_err: sticky error mask
+ * @saved_uic_err: sticky UIC error mask
  * @dev_cmd: ufs device management command information
  * @auto_bkops_enabled: to track whether bkops is enabled in device
  */
@@ -217,21 +223,27 @@ struct ufs_hba {
 	struct uic_command *active_uic_cmd;
 	struct mutex uic_cmd_mutex;
 
-	wait_queue_head_t ufshcd_tm_wait_queue;
+	wait_queue_head_t tm_wq;
+	wait_queue_head_t tm_tag_wq;
 	unsigned long tm_condition;
+	unsigned long tm_slots_in_use;
 
 	struct completion *pwr_done;
 
 	u32 ufshcd_state;
+	u32 eh_flags;
 	u32 intr_mask;
 	u16 ee_ctrl_mask;
 
 	/* Work Queues */
-	struct work_struct feh_workq;
+	struct work_struct eh_work;
 	struct work_struct eeh_work;
 
 	/* HBA Errors */
 	u32 errors;
+	u32 uic_error;
+	u32 saved_err;
+	u32 saved_uic_err;
 
 	/* Device management request data */
 	struct ufs_dev_cmd dev_cmd;
@@ -263,6 +275,8 @@ static inline void check_upiu_size(void)
 		GENERAL_UPIU_REQUEST_SIZE + QUERY_DESC_MAX_SIZE);
 }
 
+extern int ufshcd_suspend(struct ufs_hba *hba, pm_message_t state);
+extern int ufshcd_resume(struct ufs_hba *hba);
 extern int ufshcd_runtime_suspend(struct ufs_hba *hba);
 extern int ufshcd_runtime_resume(struct ufs_hba *hba);
 extern int ufshcd_runtime_idle(struct ufs_hba *hba);
diff --git a/drivers/scsi/ufs/ufshci.h b/drivers/scsi/ufs/ufshci.h
index 0475c6619a68..9abc7e32b43d 100644
--- a/drivers/scsi/ufs/ufshci.h
+++ b/drivers/scsi/ufs/ufshci.h
@@ -304,10 +304,10 @@ enum {
  * @size: size of physical segment DW-3
  */
 struct ufshcd_sg_entry {
-	u32    base_addr;
-	u32    upper_addr;
-	u32    reserved;
-	u32    size;
+	__le32    base_addr;
+	__le32    upper_addr;
+	__le32    reserved;
+	__le32    size;
 };
 
 /**
@@ -330,10 +330,10 @@ struct utp_transfer_cmd_desc {
  * @dword3: Descriptor Header DW3
  */
 struct request_desc_header {
-	u32 dword_0;
-	u32 dword_1;
-	u32 dword_2;
-	u32 dword_3;
+	__le32 dword_0;
+	__le32 dword_1;
+	__le32 dword_2;
+	__le32 dword_3;
 };
 
 /**
@@ -352,16 +352,16 @@ struct utp_transfer_req_desc {
 	struct request_desc_header header;
 
 	/* DW 4-5*/
-	u32  command_desc_base_addr_lo;
-	u32  command_desc_base_addr_hi;
+	__le32  command_desc_base_addr_lo;
+	__le32  command_desc_base_addr_hi;
 
 	/* DW 6 */
-	u16  response_upiu_length;
-	u16  response_upiu_offset;
+	__le16  response_upiu_length;
+	__le16  response_upiu_offset;
 
 	/* DW 7 */
-	u16  prd_table_length;
-	u16  prd_table_offset;
+	__le16  prd_table_length;
+	__le16  prd_table_offset;
 };
 
 /**
@@ -376,10 +376,10 @@ struct utp_task_req_desc {
 	struct request_desc_header header;
 
 	/* DW 4-11 */
-	u32 task_req_upiu[TASK_REQ_UPIU_SIZE_DWORDS];
+	__le32 task_req_upiu[TASK_REQ_UPIU_SIZE_DWORDS];
 
 	/* DW 12-19 */
-	u32 task_rsp_upiu[TASK_RSP_UPIU_SIZE_DWORDS];
+	__le32 task_rsp_upiu[TASK_RSP_UPIU_SIZE_DWORDS];
 };
 
 #endif /* End of Header */
diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
index db3b494e5926..308256b5e4cb 100644
--- a/drivers/scsi/virtio_scsi.c
+++ b/drivers/scsi/virtio_scsi.c
@@ -23,6 +23,7 @@
 #include <linux/virtio_config.h>
 #include <linux/virtio_scsi.h>
 #include <linux/cpu.h>
+#include <linux/blkdev.h>
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_cmnd.h>
@@ -37,6 +38,7 @@ struct virtio_scsi_cmd {
 	struct completion *comp;
 	union {
 		struct virtio_scsi_cmd_req       cmd;
+		struct virtio_scsi_cmd_req_pi    cmd_pi;
 		struct virtio_scsi_ctrl_tmf_req  tmf;
 		struct virtio_scsi_ctrl_an_req   an;
 	} req;
@@ -73,17 +75,12 @@ struct virtio_scsi_vq {
  * queue, and also lets the driver optimize the IRQ affinity for the virtqueues
  * (each virtqueue's affinity is set to the CPU that "owns" the queue).
  *
- * An interesting effect of this policy is that only writes to req_vq need to
- * take the tgt_lock.  Read can be done outside the lock because:
+ * tgt_lock is held to serialize reading and writing req_vq. Reading req_vq
+ * could be done locklessly, but we do not do it yet.
  *
- * - writes of req_vq only occur when atomic_inc_return(&tgt->reqs) returns 1.
- *   In that case, no other CPU is reading req_vq: even if they were in
- *   virtscsi_queuecommand_multi, they would be spinning on tgt_lock.
- *
- * - reads of req_vq only occur when the target is not idle (reqs != 0).
- *   A CPU that enters virtscsi_queuecommand_multi will not modify req_vq.
- *
- * Similarly, decrements of reqs are never concurrent with writes of req_vq.
+ * Decrements of reqs are never concurrent with writes of req_vq: before the
+ * decrement reqs will be != 0; after the decrement the virtqueue completion
+ * routine will not use the req_vq so it can be changed by a new request.
  * Thus they can happen outside the tgt_lock, provided of course we make reqs
  * an atomic_t.
  */
@@ -204,7 +201,6 @@ static void virtscsi_complete_cmd(struct virtio_scsi *vscsi, void *buf)
 			set_driver_byte(sc, DRIVER_SENSE);
 	}
 
-	mempool_free(cmd, virtscsi_cmd_pool);
 	sc->scsi_done(sc);
 
 	atomic_dec(&tgt->reqs);
@@ -238,49 +234,25 @@ static void virtscsi_req_done(struct virtqueue *vq)
 	int index = vq->index - VIRTIO_SCSI_VQ_BASE;
 	struct virtio_scsi_vq *req_vq = &vscsi->req_vqs[index];
 
-	/*
-	 * Read req_vq before decrementing the reqs field in
-	 * virtscsi_complete_cmd.
-	 *
-	 * With barriers:
-	 *
-	 * 	CPU #0			virtscsi_queuecommand_multi (CPU #1)
-	 * 	------------------------------------------------------------
-	 * 	lock vq_lock
-	 * 	read req_vq
-	 * 	read reqs (reqs = 1)
-	 * 	write reqs (reqs = 0)
-	 * 				increment reqs (reqs = 1)
-	 * 				write req_vq
-	 *
-	 * Possible reordering without barriers:
-	 *
-	 * 	CPU #0			virtscsi_queuecommand_multi (CPU #1)
-	 * 	------------------------------------------------------------
-	 * 	lock vq_lock
-	 * 	read reqs (reqs = 1)
-	 * 	write reqs (reqs = 0)
-	 * 				increment reqs (reqs = 1)
-	 * 				write req_vq
-	 * 	read (wrong) req_vq
-	 *
-	 * We do not need a full smp_rmb, because req_vq is required to get
-	 * to tgt->reqs: tgt is &vscsi->tgt[sc->device->id], where sc is stored
-	 * in the virtqueue as the user token.
-	 */
-	smp_read_barrier_depends();
-
 	virtscsi_vq_done(vscsi, req_vq, virtscsi_complete_cmd);
 };
 
+static void virtscsi_poll_requests(struct virtio_scsi *vscsi)
+{
+	int i, num_vqs;
+
+	num_vqs = vscsi->num_queues;
+	for (i = 0; i < num_vqs; i++)
+		virtscsi_vq_done(vscsi, &vscsi->req_vqs[i],
+				 virtscsi_complete_cmd);
+}
+
 static void virtscsi_complete_free(struct virtio_scsi *vscsi, void *buf)
 {
 	struct virtio_scsi_cmd *cmd = buf;
 
 	if (cmd->comp)
 		complete_all(cmd->comp);
-	else
-		mempool_free(cmd, virtscsi_cmd_pool);
 }
 
 static void virtscsi_ctrl_done(struct virtqueue *vq)
@@ -291,6 +263,8 @@ static void virtscsi_ctrl_done(struct virtqueue *vq)
 	virtscsi_vq_done(vscsi, &vscsi->ctrl_vq, virtscsi_complete_free);
 };
 
+static void virtscsi_handle_event(struct work_struct *work);
+
 static int virtscsi_kick_event(struct virtio_scsi *vscsi,
 			       struct virtio_scsi_event_node *event_node)
 {
@@ -298,6 +272,7 @@ static int virtscsi_kick_event(struct virtio_scsi *vscsi,
 	struct scatterlist sg;
 	unsigned long flags;
 
+	INIT_WORK(&event_node->work, virtscsi_handle_event);
 	sg_init_one(&sg, &event_node->event, sizeof(struct virtio_scsi_event));
 
 	spin_lock_irqsave(&vscsi->event_vq.vq_lock, flags);
@@ -415,7 +390,6 @@ static void virtscsi_complete_event(struct virtio_scsi *vscsi, void *buf)
 {
 	struct virtio_scsi_event_node *event_node = buf;
 
-	INIT_WORK(&event_node->work, virtscsi_handle_event);
 	schedule_work(&event_node->work);
 }
 
@@ -433,14 +407,13 @@ static void virtscsi_event_done(struct virtqueue *vq)
  * @cmd		: command structure
  * @req_size	: size of the request buffer
  * @resp_size	: size of the response buffer
- * @gfp	: flags to use for memory allocations
  */
 static int virtscsi_add_cmd(struct virtqueue *vq,
 			    struct virtio_scsi_cmd *cmd,
-			    size_t req_size, size_t resp_size, gfp_t gfp)
+			    size_t req_size, size_t resp_size)
 {
 	struct scsi_cmnd *sc = cmd->sc;
-	struct scatterlist *sgs[4], req, resp;
+	struct scatterlist *sgs[6], req, resp;
 	struct sg_table *out, *in;
 	unsigned out_num = 0, in_num = 0;
 
@@ -458,30 +431,38 @@ static int virtscsi_add_cmd(struct virtqueue *vq,
 	sgs[out_num++] = &req;
 
 	/* Data-out buffer.  */
-	if (out)
+	if (out) {
+		/* Place WRITE protection SGLs before Data OUT payload */
+		if (scsi_prot_sg_count(sc))
+			sgs[out_num++] = scsi_prot_sglist(sc);
 		sgs[out_num++] = out->sgl;
+	}
 
 	/* Response header.  */
 	sg_init_one(&resp, &cmd->resp, resp_size);
 	sgs[out_num + in_num++] = &resp;
 
 	/* Data-in buffer */
-	if (in)
+	if (in) {
+		/* Place READ protection SGLs before Data IN payload */
+		if (scsi_prot_sg_count(sc))
+			sgs[out_num + in_num++] = scsi_prot_sglist(sc);
 		sgs[out_num + in_num++] = in->sgl;
+	}
 
-	return virtqueue_add_sgs(vq, sgs, out_num, in_num, cmd, gfp);
+	return virtqueue_add_sgs(vq, sgs, out_num, in_num, cmd, GFP_ATOMIC);
 }
 
 static int virtscsi_kick_cmd(struct virtio_scsi_vq *vq,
 			     struct virtio_scsi_cmd *cmd,
-			     size_t req_size, size_t resp_size, gfp_t gfp)
+			     size_t req_size, size_t resp_size)
 {
 	unsigned long flags;
 	int err;
 	bool needs_kick = false;
 
 	spin_lock_irqsave(&vq->vq_lock, flags);
-	err = virtscsi_add_cmd(vq->vq, cmd, req_size, resp_size, gfp);
+	err = virtscsi_add_cmd(vq->vq, cmd, req_size, resp_size);
 	if (!err)
 		needs_kick = virtqueue_kick_prepare(vq->vq);
 
@@ -492,14 +473,46 @@ static int virtscsi_kick_cmd(struct virtio_scsi_vq *vq,
 	return err;
 }
 
+static void virtio_scsi_init_hdr(struct virtio_scsi_cmd_req *cmd,
+				 struct scsi_cmnd *sc)
+{
+	cmd->lun[0] = 1;
+	cmd->lun[1] = sc->device->id;
+	cmd->lun[2] = (sc->device->lun >> 8) | 0x40;
+	cmd->lun[3] = sc->device->lun & 0xff;
+	cmd->tag = (unsigned long)sc;
+	cmd->task_attr = VIRTIO_SCSI_S_SIMPLE;
+	cmd->prio = 0;
+	cmd->crn = 0;
+}
+
+static void virtio_scsi_init_hdr_pi(struct virtio_scsi_cmd_req_pi *cmd_pi,
+				    struct scsi_cmnd *sc)
+{
+	struct request *rq = sc->request;
+	struct blk_integrity *bi;
+
+	virtio_scsi_init_hdr((struct virtio_scsi_cmd_req *)cmd_pi, sc);
+
+	if (!rq || !scsi_prot_sg_count(sc))
+		return;
+
+	bi = blk_get_integrity(rq->rq_disk);
+
+	if (sc->sc_data_direction == DMA_TO_DEVICE)
+		cmd_pi->pi_bytesout = blk_rq_sectors(rq) * bi->tuple_size;
+	else if (sc->sc_data_direction == DMA_FROM_DEVICE)
+		cmd_pi->pi_bytesin = blk_rq_sectors(rq) * bi->tuple_size;
+}
+
 static int virtscsi_queuecommand(struct virtio_scsi *vscsi,
 				 struct virtio_scsi_vq *req_vq,
 				 struct scsi_cmnd *sc)
 {
-	struct virtio_scsi_cmd *cmd;
-	int ret;
-
 	struct Scsi_Host *shost = virtio_scsi_host(vscsi->vdev);
+	struct virtio_scsi_cmd *cmd = scsi_cmd_priv(sc);
+	int req_size;
+
 	BUG_ON(scsi_sg_count(sc) > shost->sg_tablesize);
 
 	/* TODO: check feature bit and fail if unsupported?  */
@@ -508,36 +521,24 @@ static int virtscsi_queuecommand(struct virtio_scsi *vscsi,
 	dev_dbg(&sc->device->sdev_gendev,
 		"cmd %p CDB: %#02x\n", sc, sc->cmnd[0]);
 
-	ret = SCSI_MLQUEUE_HOST_BUSY;
-	cmd = mempool_alloc(virtscsi_cmd_pool, GFP_ATOMIC);
-	if (!cmd)
-		goto out;
-
 	memset(cmd, 0, sizeof(*cmd));
 	cmd->sc = sc;
-	cmd->req.cmd = (struct virtio_scsi_cmd_req){
-		.lun[0] = 1,
-		.lun[1] = sc->device->id,
-		.lun[2] = (sc->device->lun >> 8) | 0x40,
-		.lun[3] = sc->device->lun & 0xff,
-		.tag = (unsigned long)sc,
-		.task_attr = VIRTIO_SCSI_S_SIMPLE,
-		.prio = 0,
-		.crn = 0,
-	};
 
 	BUG_ON(sc->cmd_len > VIRTIO_SCSI_CDB_SIZE);
-	memcpy(cmd->req.cmd.cdb, sc->cmnd, sc->cmd_len);
 
-	if (virtscsi_kick_cmd(req_vq, cmd,
-			      sizeof cmd->req.cmd, sizeof cmd->resp.cmd,
-			      GFP_ATOMIC) == 0)
-		ret = 0;
-	else
-		mempool_free(cmd, virtscsi_cmd_pool);
+	if (virtio_has_feature(vscsi->vdev, VIRTIO_SCSI_F_T10_PI)) {
+		virtio_scsi_init_hdr_pi(&cmd->req.cmd_pi, sc);
+		memcpy(cmd->req.cmd_pi.cdb, sc->cmnd, sc->cmd_len);
+		req_size = sizeof(cmd->req.cmd_pi);
+	} else {
+		virtio_scsi_init_hdr(&cmd->req.cmd, sc);
+		memcpy(cmd->req.cmd.cdb, sc->cmnd, sc->cmd_len);
+		req_size = sizeof(cmd->req.cmd);
+	}
 
-out:
-	return ret;
+	if (virtscsi_kick_cmd(req_vq, cmd, req_size, sizeof(cmd->resp.cmd)) != 0)
+		return SCSI_MLQUEUE_HOST_BUSY;
+	return 0;
 }
 
 static int virtscsi_queuecommand_single(struct Scsi_Host *sh,
@@ -560,12 +561,8 @@ static struct virtio_scsi_vq *virtscsi_pick_vq(struct virtio_scsi *vscsi,
 
 	spin_lock_irqsave(&tgt->tgt_lock, flags);
 
-	/*
-	 * The memory barrier after atomic_inc_return matches
-	 * the smp_read_barrier_depends() in virtscsi_req_done.
-	 */
 	if (atomic_inc_return(&tgt->reqs) > 1)
-		vq = ACCESS_ONCE(tgt->req_vq);
+		vq = tgt->req_vq;
 	else {
 		queue_num = smp_processor_id();
 		while (unlikely(queue_num >= vscsi->num_queues))
@@ -596,8 +593,7 @@ static int virtscsi_tmf(struct virtio_scsi *vscsi, struct virtio_scsi_cmd *cmd)
 
 	cmd->comp = &comp;
 	if (virtscsi_kick_cmd(&vscsi->ctrl_vq, cmd,
-			      sizeof cmd->req.tmf, sizeof cmd->resp.tmf,
-			      GFP_NOIO) < 0)
+			      sizeof cmd->req.tmf, sizeof cmd->resp.tmf) < 0)
 		goto out;
 
 	wait_for_completion(&comp);
@@ -605,6 +601,18 @@ static int virtscsi_tmf(struct virtio_scsi *vscsi, struct virtio_scsi_cmd *cmd)
 	    cmd->resp.tmf.response == VIRTIO_SCSI_S_FUNCTION_SUCCEEDED)
 		ret = SUCCESS;
 
+	/*
+	 * The spec guarantees that all requests related to the TMF have
+	 * been completed, but the callback might not have run yet if
+	 * we're using independent interrupts (e.g. MSI).  Poll the
+	 * virtqueues once.
+	 *
+	 * In the abort case, sc->scsi_done will do nothing, because
+	 * the block layer must have detected a timeout and as a result
+	 * REQ_ATOM_COMPLETE has been set.
+	 */
+	virtscsi_poll_requests(vscsi);
+
 out:
 	mempool_free(cmd, virtscsi_cmd_pool);
 	return ret;
@@ -683,6 +691,7 @@ static struct scsi_host_template virtscsi_host_template_single = {
 	.name = "Virtio SCSI HBA",
 	.proc_name = "virtio_scsi",
 	.this_id = -1,
+	.cmd_size = sizeof(struct virtio_scsi_cmd),
 	.queuecommand = virtscsi_queuecommand_single,
 	.eh_abort_handler = virtscsi_abort,
 	.eh_device_reset_handler = virtscsi_device_reset,
@@ -699,6 +708,7 @@ static struct scsi_host_template virtscsi_host_template_multi = {
 	.name = "Virtio SCSI HBA",
 	.proc_name = "virtio_scsi",
 	.this_id = -1,
+	.cmd_size = sizeof(struct virtio_scsi_cmd),
 	.queuecommand = virtscsi_queuecommand_multi,
 	.eh_abort_handler = virtscsi_abort,
 	.eh_device_reset_handler = virtscsi_device_reset,
@@ -875,7 +885,7 @@ static int virtscsi_probe(struct virtio_device *vdev)
 {
 	struct Scsi_Host *shost;
 	struct virtio_scsi *vscsi;
-	int err;
+	int err, host_prot;
 	u32 sg_elems, num_targets;
 	u32 cmd_per_lun;
 	u32 num_queues;
@@ -925,6 +935,16 @@ static int virtscsi_probe(struct virtio_device *vdev)
 	shost->max_id = num_targets;
 	shost->max_channel = 0;
 	shost->max_cmd_len = VIRTIO_SCSI_CDB_SIZE;
+
+	if (virtio_has_feature(vdev, VIRTIO_SCSI_F_T10_PI)) {
+		host_prot = SHOST_DIF_TYPE1_PROTECTION | SHOST_DIF_TYPE2_PROTECTION |
+			    SHOST_DIF_TYPE3_PROTECTION | SHOST_DIX_TYPE1_PROTECTION |
+			    SHOST_DIX_TYPE2_PROTECTION | SHOST_DIX_TYPE3_PROTECTION;
+
+		scsi_host_set_prot(shost, host_prot);
+		scsi_host_set_guard(shost, SHOST_DIX_GUARD_CRC);
+	}
+
 	err = scsi_add_host(shost, &vdev->dev);
 	if (err)
 		goto scsi_add_host_failed;
@@ -994,6 +1014,7 @@ static struct virtio_device_id id_table[] = {
 static unsigned int features[] = {
 	VIRTIO_SCSI_F_HOTPLUG,
 	VIRTIO_SCSI_F_CHANGE,
+	VIRTIO_SCSI_F_T10_PI,
 };
 
 static struct virtio_driver virtio_scsi_driver = {
diff --git a/drivers/spi/spi-pxa2xx-dma.c b/drivers/spi/spi-pxa2xx-dma.c
index f6759dc0153b..c41ff148a2b4 100644
--- a/drivers/spi/spi-pxa2xx-dma.c
+++ b/drivers/spi/spi-pxa2xx-dma.c
@@ -368,7 +368,7 @@ int pxa2xx_spi_set_dma_burst_and_threshold(struct chip_data *chip,
 	 * otherwise we use the default. Also we use the default FIFO
 	 * thresholds for now.
 	 */
-	*burst_code = chip_info ? chip_info->dma_burst_size : 16;
+	*burst_code = chip_info ? chip_info->dma_burst_size : 1;
 	*threshold = SSCR1_RxTresh(RX_THRESH_DFLT)
 		   | SSCR1_TxTresh(TX_THRESH_DFLT);
 
diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c
index a98df7eeb42d..fe792106bdc5 100644
--- a/drivers/spi/spi-pxa2xx.c
+++ b/drivers/spi/spi-pxa2xx.c
@@ -118,6 +118,7 @@ static void lpss_ssp_setup(struct driver_data *drv_data)
 	 */
 	orig = readl(drv_data->ioaddr + offset + SPI_CS_CONTROL);
 
+	/* Test SPI_CS_CONTROL_SW_MODE bit enabling */
 	value = orig | SPI_CS_CONTROL_SW_MODE;
 	writel(value, drv_data->ioaddr + offset + SPI_CS_CONTROL);
 	value = readl(drv_data->ioaddr + offset + SPI_CS_CONTROL);
@@ -126,10 +127,13 @@ static void lpss_ssp_setup(struct driver_data *drv_data)
 		goto detection_done;
 	}
 
-	value &= ~SPI_CS_CONTROL_SW_MODE;
+	orig = readl(drv_data->ioaddr + offset + SPI_CS_CONTROL);
+
+	/* Test SPI_CS_CONTROL_SW_MODE bit disabling */
+	value = orig & ~SPI_CS_CONTROL_SW_MODE;
 	writel(value, drv_data->ioaddr + offset + SPI_CS_CONTROL);
 	value = readl(drv_data->ioaddr + offset + SPI_CS_CONTROL);
-	if (value != orig) {
+	if (value != (orig & ~SPI_CS_CONTROL_SW_MODE)) {
 		offset = 0x800;
 		goto detection_done;
 	}
diff --git a/drivers/spi/spi-qup.c b/drivers/spi/spi-qup.c
index fc1de86d3c8a..c08da380cb23 100644
--- a/drivers/spi/spi-qup.c
+++ b/drivers/spi/spi-qup.c
@@ -424,31 +424,6 @@ static int spi_qup_io_config(struct spi_device *spi, struct spi_transfer *xfer)
 	return 0;
 }
 
-static void spi_qup_set_cs(struct spi_device *spi, bool enable)
-{
-	struct spi_qup *controller = spi_master_get_devdata(spi->master);
-
-	u32 iocontol, mask;
-
-	iocontol = readl_relaxed(controller->base + SPI_IO_CONTROL);
-
-	/* Disable auto CS toggle and use manual */
-	iocontol &= ~SPI_IO_C_MX_CS_MODE;
-	iocontol |= SPI_IO_C_FORCE_CS;
-
-	iocontol &= ~SPI_IO_C_CS_SELECT_MASK;
-	iocontol |= SPI_IO_C_CS_SELECT(spi->chip_select);
-
-	mask = SPI_IO_C_CS_N_POLARITY_0 << spi->chip_select;
-
-	if (enable)
-		iocontol |= mask;
-	else
-		iocontol &= ~mask;
-
-	writel_relaxed(iocontol, controller->base + SPI_IO_CONTROL);
-}
-
 static int spi_qup_transfer_one(struct spi_master *master,
 			      struct spi_device *spi,
 			      struct spi_transfer *xfer)
@@ -571,12 +546,16 @@ static int spi_qup_probe(struct platform_device *pdev)
 		return -ENOMEM;
 	}
 
+	/* use num-cs unless not present or out of range */
+	if (of_property_read_u16(dev->of_node, "num-cs",
+			&master->num_chipselect) ||
+			(master->num_chipselect > SPI_NUM_CHIPSELECTS))
+		master->num_chipselect = SPI_NUM_CHIPSELECTS;
+
 	master->bus_num = pdev->id;
 	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH | SPI_LOOP;
-	master->num_chipselect = SPI_NUM_CHIPSELECTS;
 	master->bits_per_word_mask = SPI_BPW_RANGE_MASK(4, 32);
 	master->max_speed_hz = max_freq;
-	master->set_cs = spi_qup_set_cs;
 	master->transfer_one = spi_qup_transfer_one;
 	master->dev.of_node = pdev->dev.of_node;
 	master->auto_runtime_pm = true;
@@ -640,16 +619,19 @@ static int spi_qup_probe(struct platform_device *pdev)
 	if (ret)
 		goto error;
 
-	ret = devm_spi_register_master(dev, master);
-	if (ret)
-		goto error;
-
 	pm_runtime_set_autosuspend_delay(dev, MSEC_PER_SEC);
 	pm_runtime_use_autosuspend(dev);
 	pm_runtime_set_active(dev);
 	pm_runtime_enable(dev);
+
+	ret = devm_spi_register_master(dev, master);
+	if (ret)
+		goto disable_pm;
+
 	return 0;
 
+disable_pm:
+	pm_runtime_disable(&pdev->dev);
 error:
 	clk_disable_unprepare(cclk);
 	clk_disable_unprepare(iclk);
diff --git a/drivers/spi/spi-sh-sci.c b/drivers/spi/spi-sh-sci.c
index 1f56ef651d1a..b83dd733684c 100644
--- a/drivers/spi/spi-sh-sci.c
+++ b/drivers/spi/spi-sh-sci.c
@@ -175,9 +175,9 @@ static int sh_sci_spi_remove(struct platform_device *dev)
 {
 	struct sh_sci_spi *sp = platform_get_drvdata(dev);
 
-	iounmap(sp->membase);
-	setbits(sp, PIN_INIT, 0);
 	spi_bitbang_stop(&sp->bitbang);
+	setbits(sp, PIN_INIT, 0);
+	iounmap(sp->membase);
 	spi_master_put(sp->bitbang.master);
 	return 0;
 }
diff --git a/drivers/staging/android/timed_output.c b/drivers/staging/android/timed_output.c
index 2c617834dc46..c341ac11c5a3 100644
--- a/drivers/staging/android/timed_output.c
+++ b/drivers/staging/android/timed_output.c
@@ -97,7 +97,6 @@ void timed_output_dev_unregister(struct timed_output_dev *tdev)
 {
 	tdev->enable(tdev, 0);
 	device_destroy(timed_output_class, MKDEV(0, tdev->index));
-	dev_set_drvdata(tdev->dev, NULL);
 }
 EXPORT_SYMBOL_GPL(timed_output_dev_unregister);
 
diff --git a/drivers/staging/comedi/Kconfig b/drivers/staging/comedi/Kconfig
index 5d56428d508a..a2f6957e7ee9 100644
--- a/drivers/staging/comedi/Kconfig
+++ b/drivers/staging/comedi/Kconfig
@@ -651,6 +651,7 @@ config COMEDI_ADDI_APCI_1516
 
 config COMEDI_ADDI_APCI_1564
 	tristate "ADDI-DATA APCI_1564 support"
+	select COMEDI_ADDI_WATCHDOG
 	---help---
 	  Enable support for ADDI-DATA APCI_1564 cards
 
diff --git a/drivers/staging/et131x/et131x.c b/drivers/staging/et131x/et131x.c
index 0901ef5d6e8a..08356b6955a4 100644
--- a/drivers/staging/et131x/et131x.c
+++ b/drivers/staging/et131x/et131x.c
@@ -4605,7 +4605,7 @@ static int et131x_pci_setup(struct pci_dev *pdev,
 	netdev->netdev_ops     = &et131x_netdev_ops;
 
 	SET_NETDEV_DEV(netdev, &pdev->dev);
-	SET_ETHTOOL_OPS(netdev, &et131x_ethtool_ops);
+	netdev->ethtool_ops = &et131x_ethtool_ops;
 
 	adapter = et131x_adapter_init(netdev, pdev);
 
diff --git a/drivers/staging/ft1000/ft1000-pcmcia/ft1000_hw.c b/drivers/staging/ft1000/ft1000-pcmcia/ft1000_hw.c
index d6421b9b5981..a6158bef58e5 100644
--- a/drivers/staging/ft1000/ft1000-pcmcia/ft1000_hw.c
+++ b/drivers/staging/ft1000/ft1000-pcmcia/ft1000_hw.c
@@ -2249,7 +2249,7 @@ struct net_device *init_ft1000_card(struct pcmcia_device *link,
 
 	ft1000InitProc(dev);
 	ft1000_card_present = 1;
-	SET_ETHTOOL_OPS(dev, &ops);
+	dev->ethtool_ops = &ops;
 	printk(KERN_INFO "ft1000: %s: addr 0x%04lx irq %d, MAC addr %pM\n",
 			dev->name, dev->base_addr, dev->irq, dev->dev_addr);
 	return dev;
diff --git a/drivers/staging/iio/Kconfig b/drivers/staging/iio/Kconfig
index b36feb080cba..fa38be0982f9 100644
--- a/drivers/staging/iio/Kconfig
+++ b/drivers/staging/iio/Kconfig
@@ -36,10 +36,11 @@ config IIO_SIMPLE_DUMMY_EVENTS
          Add some dummy events to the simple dummy driver.
 
 config IIO_SIMPLE_DUMMY_BUFFER
-       boolean "Buffered capture support"
-       select IIO_KFIFO_BUF
-       help
-         Add buffered data capture to the simple dummy driver.
+	boolean "Buffered capture support"
+	select IIO_BUFFER
+	select IIO_KFIFO_BUF
+	help
+	  Add buffered data capture to the simple dummy driver.
 
 endif # IIO_SIMPLE_DUMMY
 
diff --git a/drivers/staging/iio/adc/ad7291.c b/drivers/staging/iio/adc/ad7291.c
index 357cef2a6f4c..7194bd138762 100644
--- a/drivers/staging/iio/adc/ad7291.c
+++ b/drivers/staging/iio/adc/ad7291.c
@@ -465,7 +465,7 @@ static int ad7291_probe(struct i2c_client *client,
 	struct ad7291_platform_data *pdata = client->dev.platform_data;
 	struct ad7291_chip_info *chip;
 	struct iio_dev *indio_dev;
-	int ret = 0;
+	int ret;
 
 	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*chip));
 	if (!indio_dev)
@@ -475,7 +475,7 @@ static int ad7291_probe(struct i2c_client *client,
 	if (pdata && pdata->use_external_ref) {
 		chip->reg = devm_regulator_get(&client->dev, "vref");
 		if (IS_ERR(chip->reg))
-			return ret;
+			return PTR_ERR(chip->reg);
 
 		ret = regulator_enable(chip->reg);
 		if (ret)
diff --git a/drivers/staging/iio/adc/mxs-lradc.c b/drivers/staging/iio/adc/mxs-lradc.c
index dae8d1a9038e..52d7517b342e 100644
--- a/drivers/staging/iio/adc/mxs-lradc.c
+++ b/drivers/staging/iio/adc/mxs-lradc.c
@@ -846,6 +846,14 @@ static int mxs_lradc_read_single(struct iio_dev *iio_dev, int chan, int *val)
 			LRADC_CTRL1);
 	mxs_lradc_reg_clear(lradc, 0xff, LRADC_CTRL0);
 
+	/* Enable / disable the divider per requirement */
+	if (test_bit(chan, &lradc->is_divided))
+		mxs_lradc_reg_set(lradc, 1 << LRADC_CTRL2_DIVIDE_BY_TWO_OFFSET,
+			LRADC_CTRL2);
+	else
+		mxs_lradc_reg_clear(lradc,
+			1 << LRADC_CTRL2_DIVIDE_BY_TWO_OFFSET, LRADC_CTRL2);
+
 	/* Clean the slot's previous content, then set new one. */
 	mxs_lradc_reg_clear(lradc, LRADC_CTRL4_LRADCSELECT_MASK(0),
 			LRADC_CTRL4);
@@ -961,15 +969,11 @@ static int mxs_lradc_write_raw(struct iio_dev *iio_dev,
 		if (val == scale_avail[MXS_LRADC_DIV_DISABLED].integer &&
 		    val2 == scale_avail[MXS_LRADC_DIV_DISABLED].nano) {
 			/* divider by two disabled */
-			writel(1 << LRADC_CTRL2_DIVIDE_BY_TWO_OFFSET,
-			       lradc->base + LRADC_CTRL2 + STMP_OFFSET_REG_CLR);
 			clear_bit(chan->channel, &lradc->is_divided);
 			ret = 0;
 		} else if (val == scale_avail[MXS_LRADC_DIV_ENABLED].integer &&
 			   val2 == scale_avail[MXS_LRADC_DIV_ENABLED].nano) {
 			/* divider by two enabled */
-			writel(1 << LRADC_CTRL2_DIVIDE_BY_TWO_OFFSET,
-			       lradc->base + LRADC_CTRL2 + STMP_OFFSET_REG_SET);
 			set_bit(chan->channel, &lradc->is_divided);
 			ret = 0;
 		}
diff --git a/drivers/staging/iio/light/tsl2x7x_core.c b/drivers/staging/iio/light/tsl2x7x_core.c
index 9e0f2a9c73ae..ab338e3ddd05 100644
--- a/drivers/staging/iio/light/tsl2x7x_core.c
+++ b/drivers/staging/iio/light/tsl2x7x_core.c
@@ -667,9 +667,13 @@ static int tsl2x7x_chip_on(struct iio_dev *indio_dev)
 	chip->tsl2x7x_config[TSL2X7X_PRX_COUNT] =
 			chip->tsl2x7x_settings.prox_pulse_count;
 	chip->tsl2x7x_config[TSL2X7X_PRX_MINTHRESHLO] =
-	chip->tsl2x7x_settings.prox_thres_low;
+			(chip->tsl2x7x_settings.prox_thres_low) & 0xFF;
+	chip->tsl2x7x_config[TSL2X7X_PRX_MINTHRESHHI] =
+			(chip->tsl2x7x_settings.prox_thres_low >> 8) & 0xFF;
 	chip->tsl2x7x_config[TSL2X7X_PRX_MAXTHRESHLO] =
-			chip->tsl2x7x_settings.prox_thres_high;
+			(chip->tsl2x7x_settings.prox_thres_high) & 0xFF;
+	chip->tsl2x7x_config[TSL2X7X_PRX_MAXTHRESHHI] =
+			(chip->tsl2x7x_settings.prox_thres_high >> 8) & 0xFF;
 
 	/* and make sure we're not already on */
 	if (chip->tsl2x7x_chip_status == TSL2X7X_CHIP_WORKING) {
diff --git a/drivers/staging/imx-drm/Kconfig b/drivers/staging/imx-drm/Kconfig
index c6e8ba7b3e4e..82fb758a29bc 100644
--- a/drivers/staging/imx-drm/Kconfig
+++ b/drivers/staging/imx-drm/Kconfig
@@ -39,19 +39,10 @@ config DRM_IMX_LDB
 	  Choose this to enable the internal LVDS Display Bridge (LDB)
 	  found on i.MX53 and i.MX6 processors.
 
-config DRM_IMX_IPUV3_CORE
-	tristate "IPUv3 core support"
-	depends on DRM_IMX
-	depends on RESET_CONTROLLER
-	help
-	  Choose this if you have a i.MX5/6 system and want
-	  to use the IPU. This option only enables IPU base
-	  support.
-
 config DRM_IMX_IPUV3
 	tristate "DRM Support for i.MX IPUv3"
 	depends on DRM_IMX
-	depends on DRM_IMX_IPUV3_CORE
+	depends on IMX_IPUV3_CORE
 	help
 	  Choose this if you have a i.MX5 or i.MX6 processor.
 
diff --git a/drivers/staging/imx-drm/Makefile b/drivers/staging/imx-drm/Makefile
index 129e3a3f59f1..582c438d8cbd 100644
--- a/drivers/staging/imx-drm/Makefile
+++ b/drivers/staging/imx-drm/Makefile
@@ -6,7 +6,6 @@ obj-$(CONFIG_DRM_IMX) += imxdrm.o
 obj-$(CONFIG_DRM_IMX_PARALLEL_DISPLAY) += parallel-display.o
 obj-$(CONFIG_DRM_IMX_TVE) += imx-tve.o
 obj-$(CONFIG_DRM_IMX_LDB) += imx-ldb.o
-obj-$(CONFIG_DRM_IMX_IPUV3_CORE) += ipu-v3/
 
 imx-ipuv3-crtc-objs  := ipuv3-crtc.o ipuv3-plane.o
 obj-$(CONFIG_DRM_IMX_IPUV3)	+= imx-ipuv3-crtc.o
diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c
index c270c9ae6d27..def8280d7ee6 100644
--- a/drivers/staging/imx-drm/imx-drm-core.c
+++ b/drivers/staging/imx-drm/imx-drm-core.c
@@ -200,13 +200,6 @@ static const struct file_operations imx_drm_driver_fops = {
 	.llseek = noop_llseek,
 };
 
-int imx_drm_connector_mode_valid(struct drm_connector *connector,
-	struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-EXPORT_SYMBOL(imx_drm_connector_mode_valid);
-
 void imx_drm_connector_destroy(struct drm_connector *connector)
 {
 	drm_sysfs_connector_remove(connector);
@@ -305,7 +298,7 @@ static int imx_drm_driver_load(struct drm_device *drm, unsigned long flags)
 			dev_err(drm->dev,
 				"[CONNECTOR:%d:%s] drm_sysfs_connector_add failed: %d\n",
 				connector->base.id,
-				drm_get_connector_name(connector), ret);
+				connector->name, ret);
 			goto err_unbind;
 		}
 	}
diff --git a/drivers/staging/imx-drm/imx-drm.h b/drivers/staging/imx-drm/imx-drm.h
index a322bac55414..7453ae00c412 100644
--- a/drivers/staging/imx-drm/imx-drm.h
+++ b/drivers/staging/imx-drm/imx-drm.h
@@ -50,8 +50,6 @@ int imx_drm_encoder_get_mux_id(struct device_node *node,
 int imx_drm_encoder_parse_of(struct drm_device *drm,
 	struct drm_encoder *encoder, struct device_node *np);
 
-int imx_drm_connector_mode_valid(struct drm_connector *connector,
-	struct drm_display_mode *mode);
 void imx_drm_connector_destroy(struct drm_connector *connector);
 void imx_drm_encoder_destroy(struct drm_encoder *encoder);
 
diff --git a/drivers/staging/imx-drm/imx-hdmi.c b/drivers/staging/imx-drm/imx-hdmi.c
index 1b440483f28f..18c9ccd460b7 100644
--- a/drivers/staging/imx-drm/imx-hdmi.c
+++ b/drivers/staging/imx-drm/imx-hdmi.c
@@ -27,8 +27,8 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder_slave.h>
+#include <video/imx-ipu-v3.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
 #include "imx-hdmi.h"
 #include "imx-drm.h"
 
@@ -1490,7 +1490,6 @@ static struct drm_connector_funcs imx_hdmi_connector_funcs = {
 
 static struct drm_connector_helper_funcs imx_hdmi_connector_helper_funcs = {
 	.get_modes = imx_hdmi_connector_get_modes,
-	.mode_valid = imx_drm_connector_mode_valid,
 	.best_encoder = imx_hdmi_connector_best_encoder,
 };
 
diff --git a/drivers/staging/imx-drm/imx-ldb.c b/drivers/staging/imx-drm/imx-ldb.c
index fe4c1ef4e7a5..7e3f019d7e72 100644
--- a/drivers/staging/imx-drm/imx-ldb.c
+++ b/drivers/staging/imx-drm/imx-ldb.c
@@ -317,7 +317,6 @@ static struct drm_connector_funcs imx_ldb_connector_funcs = {
 static struct drm_connector_helper_funcs imx_ldb_connector_helper_funcs = {
 	.get_modes = imx_ldb_connector_get_modes,
 	.best_encoder = imx_ldb_connector_best_encoder,
-	.mode_valid = imx_drm_connector_mode_valid,
 };
 
 static struct drm_encoder_funcs imx_ldb_encoder_funcs = {
diff --git a/drivers/staging/imx-drm/imx-tve.c b/drivers/staging/imx-drm/imx-tve.c
index a23f4f773146..c628fcdc22ae 100644
--- a/drivers/staging/imx-drm/imx-tve.c
+++ b/drivers/staging/imx-drm/imx-tve.c
@@ -30,8 +30,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_crtc_helper.h>
+#include <video/imx-ipu-v3.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
 #include "imx-drm.h"
 
 #define TVE_COM_CONF_REG	0x00
@@ -249,11 +249,6 @@ static int imx_tve_connector_mode_valid(struct drm_connector *connector,
 {
 	struct imx_tve *tve = con_to_tve(connector);
 	unsigned long rate;
-	int ret;
-
-	ret = imx_drm_connector_mode_valid(connector, mode);
-	if (ret != MODE_OK)
-		return ret;
 
 	/* pixel clock with 2x oversampling */
 	rate = clk_round_rate(tve->clk, 2000UL * mode->clock) / 2000;
diff --git a/drivers/staging/imx-drm/ipu-v3/imx-ipu-v3.h b/drivers/staging/imx-drm/ipu-v3/imx-ipu-v3.h
deleted file mode 100644
index c2c6fab05eaa..000000000000
--- a/drivers/staging/imx-drm/ipu-v3/imx-ipu-v3.h
+++ /dev/null
@@ -1,331 +0,0 @@
-/*
- * Copyright 2005-2009 Freescale Semiconductor, Inc.
- *
- * The code contained herein is licensed under the GNU Lesser General
- * Public License.  You may obtain a copy of the GNU Lesser General
- * Public License Version 2.1 or later at the following locations:
- *
- * http://www.opensource.org/licenses/lgpl-license.html
- * http://www.gnu.org/copyleft/lgpl.html
- */
-
-#ifndef __DRM_IPU_H__
-#define __DRM_IPU_H__
-
-#include <linux/types.h>
-#include <linux/videodev2.h>
-#include <linux/bitmap.h>
-#include <linux/fb.h>
-
-struct ipu_soc;
-
-enum ipuv3_type {
-	IPUV3EX,
-	IPUV3M,
-	IPUV3H,
-};
-
-#define IPU_PIX_FMT_GBR24	v4l2_fourcc('G', 'B', 'R', '3')
-
-/*
- * Bitfield of Display Interface signal polarities.
- */
-struct ipu_di_signal_cfg {
-	unsigned datamask_en:1;
-	unsigned interlaced:1;
-	unsigned odd_field_first:1;
-	unsigned clksel_en:1;
-	unsigned clkidle_en:1;
-	unsigned data_pol:1;	/* true = inverted */
-	unsigned clk_pol:1;	/* true = rising edge */
-	unsigned enable_pol:1;
-	unsigned Hsync_pol:1;	/* true = active high */
-	unsigned Vsync_pol:1;
-
-	u16 width;
-	u16 height;
-	u32 pixel_fmt;
-	u16 h_start_width;
-	u16 h_sync_width;
-	u16 h_end_width;
-	u16 v_start_width;
-	u16 v_sync_width;
-	u16 v_end_width;
-	u32 v_to_h_sync;
-	unsigned long pixelclock;
-#define IPU_DI_CLKMODE_SYNC	(1 << 0)
-#define IPU_DI_CLKMODE_EXT	(1 << 1)
-	unsigned long clkflags;
-
-	u8 hsync_pin;
-	u8 vsync_pin;
-};
-
-enum ipu_color_space {
-	IPUV3_COLORSPACE_RGB,
-	IPUV3_COLORSPACE_YUV,
-	IPUV3_COLORSPACE_UNKNOWN,
-};
-
-struct ipuv3_channel;
-
-enum ipu_channel_irq {
-	IPU_IRQ_EOF = 0,
-	IPU_IRQ_NFACK = 64,
-	IPU_IRQ_NFB4EOF = 128,
-	IPU_IRQ_EOS = 192,
-};
-
-int ipu_map_irq(struct ipu_soc *ipu, int irq);
-int ipu_idmac_channel_irq(struct ipu_soc *ipu, struct ipuv3_channel *channel,
-		enum ipu_channel_irq irq);
-
-#define IPU_IRQ_DP_SF_START		(448 + 2)
-#define IPU_IRQ_DP_SF_END		(448 + 3)
-#define IPU_IRQ_BG_SF_END		IPU_IRQ_DP_SF_END,
-#define IPU_IRQ_DC_FC_0			(448 + 8)
-#define IPU_IRQ_DC_FC_1			(448 + 9)
-#define IPU_IRQ_DC_FC_2			(448 + 10)
-#define IPU_IRQ_DC_FC_3			(448 + 11)
-#define IPU_IRQ_DC_FC_4			(448 + 12)
-#define IPU_IRQ_DC_FC_6			(448 + 13)
-#define IPU_IRQ_VSYNC_PRE_0		(448 + 14)
-#define IPU_IRQ_VSYNC_PRE_1		(448 + 15)
-
-/*
- * IPU Image DMA Controller (idmac) functions
- */
-struct ipuv3_channel *ipu_idmac_get(struct ipu_soc *ipu, unsigned channel);
-void ipu_idmac_put(struct ipuv3_channel *);
-
-int ipu_idmac_enable_channel(struct ipuv3_channel *channel);
-int ipu_idmac_disable_channel(struct ipuv3_channel *channel);
-int ipu_idmac_wait_busy(struct ipuv3_channel *channel, int ms);
-
-void ipu_idmac_set_double_buffer(struct ipuv3_channel *channel,
-		bool doublebuffer);
-void ipu_idmac_select_buffer(struct ipuv3_channel *channel, u32 buf_num);
-
-/*
- * IPU Display Controller (dc) functions
- */
-struct ipu_dc;
-struct ipu_di;
-struct ipu_dc *ipu_dc_get(struct ipu_soc *ipu, int channel);
-void ipu_dc_put(struct ipu_dc *dc);
-int ipu_dc_init_sync(struct ipu_dc *dc, struct ipu_di *di, bool interlaced,
-		u32 pixel_fmt, u32 width);
-void ipu_dc_enable(struct ipu_soc *ipu);
-void ipu_dc_enable_channel(struct ipu_dc *dc);
-void ipu_dc_disable_channel(struct ipu_dc *dc);
-void ipu_dc_disable(struct ipu_soc *ipu);
-
-/*
- * IPU Display Interface (di) functions
- */
-struct ipu_di *ipu_di_get(struct ipu_soc *ipu, int disp);
-void ipu_di_put(struct ipu_di *);
-int ipu_di_disable(struct ipu_di *);
-int ipu_di_enable(struct ipu_di *);
-int ipu_di_get_num(struct ipu_di *);
-int ipu_di_init_sync_panel(struct ipu_di *, struct ipu_di_signal_cfg *sig);
-
-/*
- * IPU Display Multi FIFO Controller (dmfc) functions
- */
-struct dmfc_channel;
-int ipu_dmfc_enable_channel(struct dmfc_channel *dmfc);
-void ipu_dmfc_disable_channel(struct dmfc_channel *dmfc);
-int ipu_dmfc_alloc_bandwidth(struct dmfc_channel *dmfc,
-		unsigned long bandwidth_mbs, int burstsize);
-void ipu_dmfc_free_bandwidth(struct dmfc_channel *dmfc);
-int ipu_dmfc_init_channel(struct dmfc_channel *dmfc, int width);
-struct dmfc_channel *ipu_dmfc_get(struct ipu_soc *ipu, int ipuv3_channel);
-void ipu_dmfc_put(struct dmfc_channel *dmfc);
-
-/*
- * IPU Display Processor (dp) functions
- */
-#define IPU_DP_FLOW_SYNC_BG	0
-#define IPU_DP_FLOW_SYNC_FG	1
-#define IPU_DP_FLOW_ASYNC0_BG	2
-#define IPU_DP_FLOW_ASYNC0_FG	3
-#define IPU_DP_FLOW_ASYNC1_BG	4
-#define IPU_DP_FLOW_ASYNC1_FG	5
-
-struct ipu_dp *ipu_dp_get(struct ipu_soc *ipu, unsigned int flow);
-void ipu_dp_put(struct ipu_dp *);
-int ipu_dp_enable(struct ipu_soc *ipu);
-int ipu_dp_enable_channel(struct ipu_dp *dp);
-void ipu_dp_disable_channel(struct ipu_dp *dp);
-void ipu_dp_disable(struct ipu_soc *ipu);
-int ipu_dp_setup_channel(struct ipu_dp *dp,
-		enum ipu_color_space in, enum ipu_color_space out);
-int ipu_dp_set_window_pos(struct ipu_dp *, u16 x_pos, u16 y_pos);
-int ipu_dp_set_global_alpha(struct ipu_dp *dp, bool enable, u8 alpha,
-		bool bg_chan);
-
-#define IPU_CPMEM_WORD(word, ofs, size) ((((word) * 160 + (ofs)) << 8) | (size))
-
-#define IPU_FIELD_UBO		IPU_CPMEM_WORD(0, 46, 22)
-#define IPU_FIELD_VBO		IPU_CPMEM_WORD(0, 68, 22)
-#define IPU_FIELD_IOX		IPU_CPMEM_WORD(0, 90, 4)
-#define IPU_FIELD_RDRW		IPU_CPMEM_WORD(0, 94, 1)
-#define IPU_FIELD_SO		IPU_CPMEM_WORD(0, 113, 1)
-#define IPU_FIELD_SLY		IPU_CPMEM_WORD(1, 102, 14)
-#define IPU_FIELD_SLUV		IPU_CPMEM_WORD(1, 128, 14)
-
-#define IPU_FIELD_XV		IPU_CPMEM_WORD(0, 0, 10)
-#define IPU_FIELD_YV		IPU_CPMEM_WORD(0, 10, 9)
-#define IPU_FIELD_XB		IPU_CPMEM_WORD(0, 19, 13)
-#define IPU_FIELD_YB		IPU_CPMEM_WORD(0, 32, 12)
-#define IPU_FIELD_NSB_B		IPU_CPMEM_WORD(0, 44, 1)
-#define IPU_FIELD_CF		IPU_CPMEM_WORD(0, 45, 1)
-#define IPU_FIELD_SX		IPU_CPMEM_WORD(0, 46, 12)
-#define IPU_FIELD_SY		IPU_CPMEM_WORD(0, 58, 11)
-#define IPU_FIELD_NS		IPU_CPMEM_WORD(0, 69, 10)
-#define IPU_FIELD_SDX		IPU_CPMEM_WORD(0, 79, 7)
-#define IPU_FIELD_SM		IPU_CPMEM_WORD(0, 86, 10)
-#define IPU_FIELD_SCC		IPU_CPMEM_WORD(0, 96, 1)
-#define IPU_FIELD_SCE		IPU_CPMEM_WORD(0, 97, 1)
-#define IPU_FIELD_SDY		IPU_CPMEM_WORD(0, 98, 7)
-#define IPU_FIELD_SDRX		IPU_CPMEM_WORD(0, 105, 1)
-#define IPU_FIELD_SDRY		IPU_CPMEM_WORD(0, 106, 1)
-#define IPU_FIELD_BPP		IPU_CPMEM_WORD(0, 107, 3)
-#define IPU_FIELD_DEC_SEL	IPU_CPMEM_WORD(0, 110, 2)
-#define IPU_FIELD_DIM		IPU_CPMEM_WORD(0, 112, 1)
-#define IPU_FIELD_BNDM		IPU_CPMEM_WORD(0, 114, 3)
-#define IPU_FIELD_BM		IPU_CPMEM_WORD(0, 117, 2)
-#define IPU_FIELD_ROT		IPU_CPMEM_WORD(0, 119, 1)
-#define IPU_FIELD_HF		IPU_CPMEM_WORD(0, 120, 1)
-#define IPU_FIELD_VF		IPU_CPMEM_WORD(0, 121, 1)
-#define IPU_FIELD_THE		IPU_CPMEM_WORD(0, 122, 1)
-#define IPU_FIELD_CAP		IPU_CPMEM_WORD(0, 123, 1)
-#define IPU_FIELD_CAE		IPU_CPMEM_WORD(0, 124, 1)
-#define IPU_FIELD_FW		IPU_CPMEM_WORD(0, 125, 13)
-#define IPU_FIELD_FH		IPU_CPMEM_WORD(0, 138, 12)
-#define IPU_FIELD_EBA0		IPU_CPMEM_WORD(1, 0, 29)
-#define IPU_FIELD_EBA1		IPU_CPMEM_WORD(1, 29, 29)
-#define IPU_FIELD_ILO		IPU_CPMEM_WORD(1, 58, 20)
-#define IPU_FIELD_NPB		IPU_CPMEM_WORD(1, 78, 7)
-#define IPU_FIELD_PFS		IPU_CPMEM_WORD(1, 85, 4)
-#define IPU_FIELD_ALU		IPU_CPMEM_WORD(1, 89, 1)
-#define IPU_FIELD_ALBM		IPU_CPMEM_WORD(1, 90, 3)
-#define IPU_FIELD_ID		IPU_CPMEM_WORD(1, 93, 2)
-#define IPU_FIELD_TH		IPU_CPMEM_WORD(1, 95, 7)
-#define IPU_FIELD_SL		IPU_CPMEM_WORD(1, 102, 14)
-#define IPU_FIELD_WID0		IPU_CPMEM_WORD(1, 116, 3)
-#define IPU_FIELD_WID1		IPU_CPMEM_WORD(1, 119, 3)
-#define IPU_FIELD_WID2		IPU_CPMEM_WORD(1, 122, 3)
-#define IPU_FIELD_WID3		IPU_CPMEM_WORD(1, 125, 3)
-#define IPU_FIELD_OFS0		IPU_CPMEM_WORD(1, 128, 5)
-#define IPU_FIELD_OFS1		IPU_CPMEM_WORD(1, 133, 5)
-#define IPU_FIELD_OFS2		IPU_CPMEM_WORD(1, 138, 5)
-#define IPU_FIELD_OFS3		IPU_CPMEM_WORD(1, 143, 5)
-#define IPU_FIELD_SXYS		IPU_CPMEM_WORD(1, 148, 1)
-#define IPU_FIELD_CRE		IPU_CPMEM_WORD(1, 149, 1)
-#define IPU_FIELD_DEC_SEL2	IPU_CPMEM_WORD(1, 150, 1)
-
-struct ipu_cpmem_word {
-	u32 data[5];
-	u32 res[3];
-};
-
-struct ipu_ch_param {
-	struct ipu_cpmem_word word[2];
-};
-
-void ipu_ch_param_write_field(struct ipu_ch_param __iomem *base, u32 wbs, u32 v);
-u32 ipu_ch_param_read_field(struct ipu_ch_param __iomem *base, u32 wbs);
-struct ipu_ch_param __iomem *ipu_get_cpmem(struct ipuv3_channel *channel);
-void ipu_ch_param_dump(struct ipu_ch_param __iomem *p);
-
-static inline void ipu_ch_param_zero(struct ipu_ch_param __iomem *p)
-{
-	int i;
-	void __iomem *base = p;
-
-	for (i = 0; i < sizeof(*p) / sizeof(u32); i++)
-		writel(0, base + i * sizeof(u32));
-}
-
-static inline void ipu_cpmem_set_buffer(struct ipu_ch_param __iomem *p,
-		int bufnum, dma_addr_t buf)
-{
-	if (bufnum)
-		ipu_ch_param_write_field(p, IPU_FIELD_EBA1, buf >> 3);
-	else
-		ipu_ch_param_write_field(p, IPU_FIELD_EBA0, buf >> 3);
-}
-
-static inline void ipu_cpmem_set_resolution(struct ipu_ch_param __iomem *p,
-		int xres, int yres)
-{
-	ipu_ch_param_write_field(p, IPU_FIELD_FW, xres - 1);
-	ipu_ch_param_write_field(p, IPU_FIELD_FH, yres - 1);
-}
-
-static inline void ipu_cpmem_set_stride(struct ipu_ch_param __iomem *p,
-		int stride)
-{
-	ipu_ch_param_write_field(p, IPU_FIELD_SLY, stride - 1);
-}
-
-void ipu_cpmem_set_high_priority(struct ipuv3_channel *channel);
-
-struct ipu_rgb {
-	struct fb_bitfield	red;
-	struct fb_bitfield	green;
-	struct fb_bitfield	blue;
-	struct fb_bitfield	transp;
-	int			bits_per_pixel;
-};
-
-struct ipu_image {
-	struct v4l2_pix_format pix;
-	struct v4l2_rect rect;
-	dma_addr_t phys;
-};
-
-int ipu_cpmem_set_format_passthrough(struct ipu_ch_param __iomem *p,
-		int width);
-
-int ipu_cpmem_set_format_rgb(struct ipu_ch_param __iomem *,
-		const struct ipu_rgb *rgb);
-
-static inline void ipu_cpmem_interlaced_scan(struct ipu_ch_param *p,
-		int stride)
-{
-	ipu_ch_param_write_field(p, IPU_FIELD_SO, 1);
-	ipu_ch_param_write_field(p, IPU_FIELD_ILO, stride / 8);
-	ipu_ch_param_write_field(p, IPU_FIELD_SLY, (stride * 2) - 1);
-};
-
-void ipu_cpmem_set_yuv_planar(struct ipu_ch_param __iomem *p, u32 pixel_format,
-			int stride, int height);
-void ipu_cpmem_set_yuv_interleaved(struct ipu_ch_param __iomem *p,
-				   u32 pixel_format);
-void ipu_cpmem_set_yuv_planar_full(struct ipu_ch_param __iomem *p,
-		u32 pixel_format, int stride, int u_offset, int v_offset);
-int ipu_cpmem_set_fmt(struct ipu_ch_param __iomem *cpmem, u32 pixelformat);
-int ipu_cpmem_set_image(struct ipu_ch_param __iomem *cpmem,
-		struct ipu_image *image);
-
-enum ipu_color_space ipu_drm_fourcc_to_colorspace(u32 drm_fourcc);
-enum ipu_color_space ipu_pixelformat_to_colorspace(u32 pixelformat);
-
-static inline void ipu_cpmem_set_burstsize(struct ipu_ch_param __iomem *p,
-		int burstsize)
-{
-	ipu_ch_param_write_field(p, IPU_FIELD_NPB, burstsize - 1);
-};
-
-struct ipu_client_platformdata {
-	int di;
-	int dc;
-	int dp;
-	int dmfc;
-	int dma[2];
-};
-
-#endif /* __DRM_IPU_H__ */
diff --git a/drivers/staging/imx-drm/ipuv3-crtc.c b/drivers/staging/imx-drm/ipuv3-crtc.c
index 47bec5e17358..720868bff35b 100644
--- a/drivers/staging/imx-drm/ipuv3-crtc.c
+++ b/drivers/staging/imx-drm/ipuv3-crtc.c
@@ -30,7 +30,7 @@
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "imx-drm.h"
 #include "ipuv3-plane.h"
 
diff --git a/drivers/staging/imx-drm/ipuv3-plane.c b/drivers/staging/imx-drm/ipuv3-plane.c
index 5697e59ddf1d..6f393a11f44d 100644
--- a/drivers/staging/imx-drm/ipuv3-plane.c
+++ b/drivers/staging/imx-drm/ipuv3-plane.c
@@ -17,7 +17,7 @@
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
+#include "video/imx-ipu-v3.h"
 #include "ipuv3-plane.h"
 
 #define to_ipu_plane(x)	container_of(x, struct ipu_plane, base)
diff --git a/drivers/staging/imx-drm/parallel-display.c b/drivers/staging/imx-drm/parallel-display.c
index eaf4dda1a0c4..4ca61afdf622 100644
--- a/drivers/staging/imx-drm/parallel-display.c
+++ b/drivers/staging/imx-drm/parallel-display.c
@@ -148,7 +148,6 @@ static struct drm_connector_funcs imx_pd_connector_funcs = {
 static struct drm_connector_helper_funcs imx_pd_connector_helper_funcs = {
 	.get_modes = imx_pd_connector_get_modes,
 	.best_encoder = imx_pd_connector_best_encoder,
-	.mode_valid = imx_drm_connector_mode_valid,
 };
 
 static struct drm_encoder_funcs imx_pd_encoder_funcs = {
@@ -174,6 +173,13 @@ static int imx_pd_register(struct drm_device *drm,
 	if (ret)
 		return ret;
 
+	/* set the connector's dpms to OFF so that
+	 * drm_helper_connector_dpms() won't return
+	 * immediately since the current state is ON
+	 * at this point.
+	 */
+	imxpd->connector.dpms = DRM_MODE_DPMS_OFF;
+
 	drm_encoder_helper_add(&imxpd->encoder, &imx_pd_encoder_helper_funcs);
 	drm_encoder_init(drm, &imxpd->encoder, &imx_pd_encoder_funcs,
 			 DRM_MODE_ENCODER_NONE);
diff --git a/drivers/staging/lustre/lustre/include/lclient.h b/drivers/staging/lustre/lustre/include/lclient.h
index 827209ea6bd0..386a36c00f57 100644
--- a/drivers/staging/lustre/lustre/include/lclient.h
+++ b/drivers/staging/lustre/lustre/include/lclient.h
@@ -82,16 +82,7 @@ struct ccc_io {
 	/**
 	 * I/O vector information to or from which read/write is going.
 	 */
-	struct iovec *cui_iov;
-	unsigned long cui_nrsegs;
-	/**
-	 * Total iov count for left IO.
-	 */
-	unsigned long cui_tot_nrsegs;
-	/**
-	 * Old length for iov that was truncated partially.
-	 */
-	size_t cui_iov_olen;
+	struct iov_iter *cui_iter;
 	/**
 	 * Total size for the left IO.
 	 */
diff --git a/drivers/staging/lustre/lustre/lclient/lcommon_cl.c b/drivers/staging/lustre/lustre/lclient/lcommon_cl.c
index dc24cfa58037..1b0c216bc568 100644
--- a/drivers/staging/lustre/lustre/lclient/lcommon_cl.c
+++ b/drivers/staging/lustre/lustre/lclient/lcommon_cl.c
@@ -720,31 +720,12 @@ int ccc_io_one_lock_index(const struct lu_env *env, struct cl_io *io,
 void ccc_io_update_iov(const struct lu_env *env,
 		       struct ccc_io *cio, struct cl_io *io)
 {
-	int i;
 	size_t size = io->u.ci_rw.crw_count;
 
-	cio->cui_iov_olen = 0;
-	if (!cl_is_normalio(env, io) || cio->cui_tot_nrsegs == 0)
+	if (!cl_is_normalio(env, io) || cio->cui_iter == NULL)
 		return;
 
-	for (i = 0; i < cio->cui_tot_nrsegs; i++) {
-		struct iovec *iv = &cio->cui_iov[i];
-
-		if (iv->iov_len < size)
-			size -= iv->iov_len;
-		else {
-			if (iv->iov_len > size) {
-				cio->cui_iov_olen = iv->iov_len;
-				iv->iov_len = size;
-			}
-			break;
-		}
-	}
-
-	cio->cui_nrsegs = i + 1;
-	LASSERTF(cio->cui_tot_nrsegs >= cio->cui_nrsegs,
-		 "tot_nrsegs: %lu, nrsegs: %lu\n",
-		 cio->cui_tot_nrsegs, cio->cui_nrsegs);
+	iov_iter_truncate(cio->cui_iter, size);
 }
 
 int ccc_io_one_lock(const struct lu_env *env, struct cl_io *io,
@@ -775,30 +756,7 @@ void ccc_io_advance(const struct lu_env *env,
 	if (!cl_is_normalio(env, io))
 		return;
 
-	LASSERT(cio->cui_tot_nrsegs >= cio->cui_nrsegs);
-	LASSERT(cio->cui_tot_count  >= nob);
-
-	cio->cui_iov	+= cio->cui_nrsegs;
-	cio->cui_tot_nrsegs -= cio->cui_nrsegs;
-	cio->cui_tot_count  -= nob;
-
-	/* update the iov */
-	if (cio->cui_iov_olen > 0) {
-		struct iovec *iv;
-
-		cio->cui_iov--;
-		cio->cui_tot_nrsegs++;
-		iv = &cio->cui_iov[0];
-		if (io->ci_continue) {
-			iv->iov_base += iv->iov_len;
-			LASSERT(cio->cui_iov_olen > iv->iov_len);
-			iv->iov_len = cio->cui_iov_olen - iv->iov_len;
-		} else {
-			/* restore the iov_len, in case of restart io. */
-			iv->iov_len = cio->cui_iov_olen;
-		}
-		cio->cui_iov_olen = 0;
-	}
+	iov_iter_reexpand(cio->cui_iter, cio->cui_tot_count  -= nob);
 }
 
 /**
diff --git a/drivers/staging/lustre/lustre/llite/file.c b/drivers/staging/lustre/lustre/llite/file.c
index c4ddec2b3589..716e1ee0104f 100644
--- a/drivers/staging/lustre/lustre/llite/file.c
+++ b/drivers/staging/lustre/lustre/llite/file.c
@@ -1114,9 +1114,7 @@ restart:
 
 		switch (vio->cui_io_subtype) {
 		case IO_NORMAL:
-			cio->cui_iov = args->u.normal.via_iov;
-			cio->cui_nrsegs = args->u.normal.via_nrsegs;
-			cio->cui_tot_nrsegs = cio->cui_nrsegs;
+			cio->cui_iter = args->u.normal.via_iter;
 			cio->cui_iocb = args->u.normal.via_iocb;
 			if ((iot == CIT_WRITE) &&
 			    !(cio->cui_fd->fd_flags & LL_FILE_GROUP_LOCKED)) {
@@ -1180,58 +1178,23 @@ out:
 	return result;
 }
 
-static ssize_t ll_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
-				unsigned long nr_segs, loff_t pos)
+static ssize_t ll_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	struct lu_env      *env;
 	struct vvp_io_args *args;
-	size_t	      count = 0;
 	ssize_t	     result;
 	int		 refcheck;
 
-	result = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE);
-	if (result)
-		return result;
-
 	env = cl_env_get(&refcheck);
 	if (IS_ERR(env))
 		return PTR_ERR(env);
 
 	args = vvp_env_args(env, IO_NORMAL);
-	args->u.normal.via_iov = (struct iovec *)iov;
-	args->u.normal.via_nrsegs = nr_segs;
+	args->u.normal.via_iter = to;
 	args->u.normal.via_iocb = iocb;
 
 	result = ll_file_io_generic(env, args, iocb->ki_filp, CIT_READ,
-				    &iocb->ki_pos, count);
-	cl_env_put(env, &refcheck);
-	return result;
-}
-
-static ssize_t ll_file_read(struct file *file, char *buf, size_t count,
-			    loff_t *ppos)
-{
-	struct lu_env *env;
-	struct iovec  *local_iov;
-	struct kiocb  *kiocb;
-	ssize_t	result;
-	int	    refcheck;
-
-	env = cl_env_get(&refcheck);
-	if (IS_ERR(env))
-		return PTR_ERR(env);
-
-	local_iov = &vvp_env_info(env)->vti_local_iov;
-	kiocb = &vvp_env_info(env)->vti_kiocb;
-	local_iov->iov_base = (void __user *)buf;
-	local_iov->iov_len = count;
-	init_sync_kiocb(kiocb, file);
-	kiocb->ki_pos = *ppos;
-	kiocb->ki_nbytes = count;
-
-	result = ll_file_aio_read(kiocb, local_iov, 1, kiocb->ki_pos);
-	*ppos = kiocb->ki_pos;
-
+				    &iocb->ki_pos, iov_iter_count(to));
 	cl_env_put(env, &refcheck);
 	return result;
 }
@@ -1239,64 +1202,27 @@ static ssize_t ll_file_read(struct file *file, char *buf, size_t count,
 /*
  * Write to a file (through the page cache).
  */
-static ssize_t ll_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
-				 unsigned long nr_segs, loff_t pos)
+static ssize_t ll_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct lu_env      *env;
 	struct vvp_io_args *args;
-	size_t	      count = 0;
 	ssize_t	     result;
 	int		 refcheck;
 
-	result = generic_segment_checks(iov, &nr_segs, &count, VERIFY_READ);
-	if (result)
-		return result;
-
 	env = cl_env_get(&refcheck);
 	if (IS_ERR(env))
 		return PTR_ERR(env);
 
 	args = vvp_env_args(env, IO_NORMAL);
-	args->u.normal.via_iov = (struct iovec *)iov;
-	args->u.normal.via_nrsegs = nr_segs;
+	args->u.normal.via_iter = from;
 	args->u.normal.via_iocb = iocb;
 
 	result = ll_file_io_generic(env, args, iocb->ki_filp, CIT_WRITE,
-				  &iocb->ki_pos, count);
+				  &iocb->ki_pos, iov_iter_count(from));
 	cl_env_put(env, &refcheck);
 	return result;
 }
 
-static ssize_t ll_file_write(struct file *file, const char *buf, size_t count,
-			     loff_t *ppos)
-{
-	struct lu_env *env;
-	struct iovec  *local_iov;
-	struct kiocb  *kiocb;
-	ssize_t	result;
-	int	    refcheck;
-
-	env = cl_env_get(&refcheck);
-	if (IS_ERR(env))
-		return PTR_ERR(env);
-
-	local_iov = &vvp_env_info(env)->vti_local_iov;
-	kiocb = &vvp_env_info(env)->vti_kiocb;
-	local_iov->iov_base = (void __user *)buf;
-	local_iov->iov_len = count;
-	init_sync_kiocb(kiocb, file);
-	kiocb->ki_pos = *ppos;
-	kiocb->ki_nbytes = count;
-
-	result = ll_file_aio_write(kiocb, local_iov, 1, kiocb->ki_pos);
-	*ppos = kiocb->ki_pos;
-
-	cl_env_put(env, &refcheck);
-	return result;
-}
-
-
-
 /*
  * Send file content (through pagecache) somewhere with helper
  */
@@ -3143,10 +3069,10 @@ int ll_inode_permission(struct inode *inode, int mask)
 
 /* -o localflock - only provides locally consistent flock locks */
 struct file_operations ll_file_operations = {
-	.read	   = ll_file_read,
-	.aio_read = ll_file_aio_read,
-	.write	  = ll_file_write,
-	.aio_write = ll_file_aio_write,
+	.read	   = new_sync_read,
+	.read_iter = ll_file_read_iter,
+	.write	  = new_sync_write,
+	.write_iter = ll_file_write_iter,
 	.unlocked_ioctl = ll_file_ioctl,
 	.open	   = ll_file_open,
 	.release	= ll_file_release,
@@ -3158,10 +3084,10 @@ struct file_operations ll_file_operations = {
 };
 
 struct file_operations ll_file_operations_flock = {
-	.read	   = ll_file_read,
-	.aio_read    = ll_file_aio_read,
-	.write	  = ll_file_write,
-	.aio_write   = ll_file_aio_write,
+	.read	   = new_sync_read,
+	.read_iter    = ll_file_read_iter,
+	.write	  = new_sync_write,
+	.write_iter   = ll_file_write_iter,
 	.unlocked_ioctl = ll_file_ioctl,
 	.open	   = ll_file_open,
 	.release	= ll_file_release,
@@ -3176,10 +3102,10 @@ struct file_operations ll_file_operations_flock = {
 
 /* These are for -o noflock - to return ENOSYS on flock calls */
 struct file_operations ll_file_operations_noflock = {
-	.read	   = ll_file_read,
-	.aio_read    = ll_file_aio_read,
-	.write	  = ll_file_write,
-	.aio_write   = ll_file_aio_write,
+	.read	   = new_sync_read,
+	.read_iter    = ll_file_read_iter,
+	.write	  = new_sync_write,
+	.write_iter   = ll_file_write_iter,
 	.unlocked_ioctl = ll_file_ioctl,
 	.open	   = ll_file_open,
 	.release	= ll_file_release,
diff --git a/drivers/staging/lustre/lustre/llite/llite_internal.h b/drivers/staging/lustre/lustre/llite/llite_internal.h
index dde7632ba01f..140ee947ba49 100644
--- a/drivers/staging/lustre/lustre/llite/llite_internal.h
+++ b/drivers/staging/lustre/lustre/llite/llite_internal.h
@@ -917,8 +917,7 @@ struct vvp_io_args {
 	union {
 		struct {
 			struct kiocb      *via_iocb;
-			struct iovec      *via_iov;
-			unsigned long      via_nrsegs;
+			struct iov_iter   *via_iter;
 		} normal;
 		struct {
 			struct pipe_inode_info  *via_pipe;
diff --git a/drivers/staging/lustre/lustre/llite/rw.c b/drivers/staging/lustre/lustre/llite/rw.c
index f0122c568a09..56162103cc79 100644
--- a/drivers/staging/lustre/lustre/llite/rw.c
+++ b/drivers/staging/lustre/lustre/llite/rw.c
@@ -151,8 +151,7 @@ static struct ll_cl_context *ll_cl_init(struct file *file,
 		result = cl_io_rw_init(env, io, CIT_WRITE, pos, PAGE_CACHE_SIZE);
 		if (result == 0) {
 			cio->cui_fd = LUSTRE_FPRIVATE(file);
-			cio->cui_iov = NULL;
-			cio->cui_nrsegs = 0;
+			cio->cui_iter = NULL;
 			result = cl_io_iter_init(env, io);
 			if (result == 0) {
 				result = cl_io_lock(env, io);
diff --git a/drivers/staging/lustre/lustre/llite/rw26.c b/drivers/staging/lustre/lustre/llite/rw26.c
index 55ca8d3c3e46..af84c1aaa5f8 100644
--- a/drivers/staging/lustre/lustre/llite/rw26.c
+++ b/drivers/staging/lustre/lustre/llite/rw26.c
@@ -218,14 +218,11 @@ static void ll_free_user_pages(struct page **pages, int npages, int do_dirty)
 	int i;
 
 	for (i = 0; i < npages; i++) {
-		if (pages[i] == NULL)
-			break;
 		if (do_dirty)
 			set_page_dirty_lock(pages[i]);
 		page_cache_release(pages[i]);
 	}
-
-	OBD_FREE_LARGE(pages, npages * sizeof(*pages));
+	kvfree(pages);
 }
 
 ssize_t ll_direct_rw_pages(const struct lu_env *env, struct cl_io *io,
@@ -363,18 +360,16 @@ static ssize_t ll_direct_IO_26_seg(const struct lu_env *env, struct cl_io *io,
 #define MAX_DIO_SIZE ((MAX_MALLOC / sizeof(struct brw_page) * PAGE_CACHE_SIZE) & \
 		      ~(DT_MAX_BRW_SIZE - 1))
 static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb,
-			       const struct iovec *iov, loff_t file_offset,
-			       unsigned long nr_segs)
+			       struct iov_iter *iter, loff_t file_offset)
 {
 	struct lu_env *env;
 	struct cl_io *io;
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_mapping->host;
 	struct ccc_object *obj = cl_inode2ccc(inode);
-	long count = iov_length(iov, nr_segs);
-	long tot_bytes = 0, result = 0;
+	ssize_t count = iov_iter_count(iter);
+	ssize_t tot_bytes = 0, result = 0;
 	struct ll_inode_info *lli = ll_i2info(inode);
-	unsigned long seg = 0;
 	long size = MAX_DIO_SIZE;
 	int refcheck;
 
@@ -392,11 +387,8 @@ static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb,
 	       MAX_DIO_SIZE >> PAGE_CACHE_SHIFT);
 
 	/* Check that all user buffers are aligned as well */
-	for (seg = 0; seg < nr_segs; seg++) {
-		if (((unsigned long)iov[seg].iov_base & ~CFS_PAGE_MASK) ||
-		    (iov[seg].iov_len & ~CFS_PAGE_MASK))
-			return -EINVAL;
-	}
+	if (iov_iter_alignment(iter) & ~CFS_PAGE_MASK)
+		return -EINVAL;
 
 	env = cl_env_get(&refcheck);
 	LASSERT(!IS_ERR(env));
@@ -411,63 +403,49 @@ static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb,
 		mutex_lock(&inode->i_mutex);
 
 	LASSERT(obj->cob_transient_pages == 0);
-	for (seg = 0; seg < nr_segs; seg++) {
-		long iov_left = iov[seg].iov_len;
-		unsigned long user_addr = (unsigned long)iov[seg].iov_base;
+	while (iov_iter_count(iter)) {
+		struct page **pages;
+		size_t offs;
 
+		count = min_t(size_t, iov_iter_count(iter), size);
 		if (rw == READ) {
 			if (file_offset >= i_size_read(inode))
 				break;
-			if (file_offset + iov_left > i_size_read(inode))
-				iov_left = i_size_read(inode) - file_offset;
+			if (file_offset + count > i_size_read(inode))
+				count = i_size_read(inode) - file_offset;
 		}
 
-		while (iov_left > 0) {
-			struct page **pages;
-			int page_count, max_pages = 0;
-			long bytes;
-
-			bytes = min(size, iov_left);
-			page_count = ll_get_user_pages(rw, user_addr, bytes,
-						       &pages, &max_pages);
-			if (likely(page_count > 0)) {
-				if (unlikely(page_count <  max_pages))
-					bytes = page_count << PAGE_CACHE_SHIFT;
-				result = ll_direct_IO_26_seg(env, io, rw, inode,
-							     file->f_mapping,
-							     bytes, file_offset,
-							     pages, page_count);
-				ll_free_user_pages(pages, max_pages, rw==READ);
-			} else if (page_count == 0) {
-				GOTO(out, result = -EFAULT);
-			} else {
-				result = page_count;
-			}
-			if (unlikely(result <= 0)) {
-				/* If we can't allocate a large enough buffer
-				 * for the request, shrink it to a smaller
-				 * PAGE_SIZE multiple and try again.
-				 * We should always be able to kmalloc for a
-				 * page worth of page pointers = 4MB on i386. */
-				if (result == -ENOMEM &&
-				    size > (PAGE_CACHE_SIZE / sizeof(*pages)) *
-					   PAGE_CACHE_SIZE) {
-					size = ((((size / 2) - 1) |
-						 ~CFS_PAGE_MASK) + 1) &
-						CFS_PAGE_MASK;
-					CDEBUG(D_VFSTRACE,"DIO size now %lu\n",
-					       size);
-					continue;
-				}
-
-				GOTO(out, result);
+		result = iov_iter_get_pages_alloc(iter, &pages, count, &offs);
+		if (likely(result > 0)) {
+			int n = (result + offs + PAGE_SIZE - 1) / PAGE_SIZE;
+			result = ll_direct_IO_26_seg(env, io, rw, inode,
+						     file->f_mapping,
+						     result, file_offset,
+						     pages, n);
+			ll_free_user_pages(pages, n, rw==READ);
+		}
+		if (unlikely(result <= 0)) {
+			/* If we can't allocate a large enough buffer
+			 * for the request, shrink it to a smaller
+			 * PAGE_SIZE multiple and try again.
+			 * We should always be able to kmalloc for a
+			 * page worth of page pointers = 4MB on i386. */
+			if (result == -ENOMEM &&
+			    size > (PAGE_CACHE_SIZE / sizeof(*pages)) *
+				   PAGE_CACHE_SIZE) {
+				size = ((((size / 2) - 1) |
+					 ~CFS_PAGE_MASK) + 1) &
+					CFS_PAGE_MASK;
+				CDEBUG(D_VFSTRACE,"DIO size now %lu\n",
+				       size);
+				continue;
 			}
 
-			tot_bytes += result;
-			file_offset += result;
-			iov_left -= result;
-			user_addr += result;
+			GOTO(out, result);
 		}
+		iov_iter_advance(iter, result);
+		tot_bytes += result;
+		file_offset += result;
 	}
 out:
 	LASSERT(obj->cob_transient_pages == 0);
diff --git a/drivers/staging/lustre/lustre/llite/vvp_io.c b/drivers/staging/lustre/lustre/llite/vvp_io.c
index 7dd2b4723c5f..0e0b404cb5e6 100644
--- a/drivers/staging/lustre/lustre/llite/vvp_io.c
+++ b/drivers/staging/lustre/lustre/llite/vvp_io.c
@@ -211,27 +211,26 @@ static int vvp_mmap_locks(const struct lu_env *env,
 	struct cl_lock_descr   *descr = &cti->cti_descr;
 	ldlm_policy_data_t      policy;
 	unsigned long	   addr;
-	unsigned long	   seg;
 	ssize_t		 count;
 	int		     result;
+	struct iov_iter i;
+	struct iovec iov;
 
 	LASSERT(io->ci_type == CIT_READ || io->ci_type == CIT_WRITE);
 
 	if (!cl_is_normalio(env, io))
 		return 0;
 
-	if (vio->cui_iov == NULL) /* nfs or loop back device write */
+	if (vio->cui_iter == NULL) /* nfs or loop back device write */
 		return 0;
 
 	/* No MM (e.g. NFS)? No vmas too. */
 	if (mm == NULL)
 		return 0;
 
-	for (seg = 0; seg < vio->cui_nrsegs; seg++) {
-		const struct iovec *iv = &vio->cui_iov[seg];
-
-		addr = (unsigned long)iv->iov_base;
-		count = iv->iov_len;
+	iov_for_each(iov, i, *(vio->cui_iter)) {
+		addr = (unsigned long)iov.iov_base;
+		count = iov.iov_len;
 		if (count == 0)
 			continue;
 
@@ -527,9 +526,7 @@ static int vvp_io_read_start(const struct lu_env *env,
 	switch (vio->cui_io_subtype) {
 	case IO_NORMAL:
 		LASSERT(cio->cui_iocb->ki_pos == pos);
-		result = generic_file_aio_read(cio->cui_iocb,
-					       cio->cui_iov, cio->cui_nrsegs,
-					       cio->cui_iocb->ki_pos);
+		result = generic_file_read_iter(cio->cui_iocb, cio->cui_iter);
 		break;
 	case IO_SPLICE:
 		result = generic_file_splice_read(file, &pos,
@@ -595,12 +592,11 @@ static int vvp_io_write_start(const struct lu_env *env,
 
 	CDEBUG(D_VFSTRACE, "write: [%lli, %lli)\n", pos, pos + (long long)cnt);
 
-	if (cio->cui_iov == NULL) /* from a temp io in ll_cl_init(). */
+	if (cio->cui_iter == NULL) /* from a temp io in ll_cl_init(). */
 		result = 0;
 	else
-		result = generic_file_aio_write(cio->cui_iocb,
-						cio->cui_iov, cio->cui_nrsegs,
-						cio->cui_iocb->ki_pos);
+		result = generic_file_write_iter(cio->cui_iocb, cio->cui_iter);
+
 	if (result > 0) {
 		if (result < cnt)
 			io->ci_continue = 0;
@@ -1162,10 +1158,9 @@ int vvp_io_init(const struct lu_env *env, struct cl_object *obj,
 		 *  results."  -- Single Unix Spec */
 		if (count == 0)
 			result = 1;
-		else {
+		else
 			cio->cui_tot_count = count;
-			cio->cui_tot_nrsegs = 0;
-		}
+
 		/* for read/write, we store the jobid in the inode, and
 		 * it'll be fetched by osc when building RPC.
 		 *
diff --git a/drivers/staging/media/omap4iss/iss_video.c b/drivers/staging/media/omap4iss/iss_video.c
index ded31ea6bd39..cbf455d66f73 100644
--- a/drivers/staging/media/omap4iss/iss_video.c
+++ b/drivers/staging/media/omap4iss/iss_video.c
@@ -396,7 +396,7 @@ static void iss_video_buf_queue(struct vb2_buffer *vb)
 	}
 }
 
-static struct vb2_ops iss_video_vb2ops = {
+static const struct vb2_ops iss_video_vb2ops = {
 	.queue_setup	= iss_video_queue_setup,
 	.buf_prepare	= iss_video_buf_prepare,
 	.buf_queue	= iss_video_buf_queue,
diff --git a/drivers/staging/netlogic/xlr_net.c b/drivers/staging/netlogic/xlr_net.c
index 75d7c63cb413..e320d6bae913 100644
--- a/drivers/staging/netlogic/xlr_net.c
+++ b/drivers/staging/netlogic/xlr_net.c
@@ -1067,7 +1067,7 @@ static int xlr_net_probe(struct platform_device *pdev)
 	xlr_set_rx_mode(ndev);
 
 	priv->num_rx_desc += MAX_NUM_DESC_SPILL;
-	SET_ETHTOOL_OPS(ndev, &xlr_ethtool_ops);
+	ndev->ethtool_ops = &xlr_ethtool_ops;
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 
 	/* Common registers, do one time initialization */
diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c
index ff7214aac9dd..da9dd6bc5660 100644
--- a/drivers/staging/octeon/ethernet.c
+++ b/drivers/staging/octeon/ethernet.c
@@ -469,7 +469,7 @@ int cvm_oct_common_init(struct net_device *dev)
 
 	/* We do our own locking, Linux doesn't need to */
 	dev->features |= NETIF_F_LLTX;
-	SET_ETHTOOL_OPS(dev, &cvm_oct_ethtool_ops);
+	dev->ethtool_ops = &cvm_oct_ethtool_ops;
 
 	cvm_oct_phy_setup_device(dev);
 	cvm_oct_set_mac_filter(dev);
diff --git a/drivers/staging/rtl8192ee/core.c b/drivers/staging/rtl8192ee/core.c
index 76ea356163b6..7f6accd59986 100644
--- a/drivers/staging/rtl8192ee/core.c
+++ b/drivers/staging/rtl8192ee/core.c
@@ -322,7 +322,7 @@ static void _rtl_add_wowlan_patterns(struct ieee80211_hw *hw,
 	struct rtl_mac *mac = &(rtlpriv->mac80211);
 	struct cfg80211_pkt_pattern *patterns = wow->patterns;
 	struct rtl_wow_pattern rtl_pattern;
-	u8 *pattern_os, *mask_os;
+	const u8 *pattern_os, *mask_os;
 	u8 mask[MAX_WOL_BIT_MASK_SIZE] = {0};
 	u8 content[MAX_WOL_PATTERN_SIZE] = {0};
 	u8 broadcast_addr[6] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
@@ -1561,7 +1561,7 @@ static void rtl_op_rfkill_poll(struct ieee80211_hw *hw)
  * before switch channle or power save, or tx buffer packet
  * maybe send after offchannel or rf sleep, this may cause
  * dis-association by AP */
-static void rtl_op_flush(struct ieee80211_hw *hw,
+static void rtl_op_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			 u32 queues, bool drop)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
diff --git a/drivers/staging/rtl8723au/hal/rtl8723a_hal_init.c b/drivers/staging/rtl8723au/hal/rtl8723a_hal_init.c
index 0acacab95a48..46f5abcbaeeb 100644
--- a/drivers/staging/rtl8723au/hal/rtl8723a_hal_init.c
+++ b/drivers/staging/rtl8723au/hal/rtl8723a_hal_init.c
@@ -298,7 +298,7 @@ int rtl8723a_FirmwareDownload(struct rtw_adapter *padapter)
 	RT_TRACE(_module_hal_init_c_, _drv_info_, ("+%s\n", __func__));
 
 	if (IS_8723A_A_CUT(pHalData->VersionID)) {
-		fw_name = "rtlwifi/rtl8723aufw.bin";
+		fw_name = "rtlwifi/rtl8723aufw_A.bin";
 		RT_TRACE(_module_hal_init_c_, _drv_info_,
 			 ("rtl8723a_FirmwareDownload: R8723FwImageArray_UMC "
 			  "for RTL8723A A CUT\n"));
diff --git a/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c b/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c
index 0c9f5cebfb42..f0839f6a9345 100644
--- a/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c
+++ b/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c
@@ -1227,7 +1227,7 @@ static int cfg80211_rtw_set_default_key(struct wiphy *wiphy,
 
 static int cfg80211_rtw_get_station(struct wiphy *wiphy,
 				    struct net_device *ndev,
-				    u8 *mac, struct station_info *sinfo)
+				    const u8 *mac, struct station_info *sinfo)
 {
 	int ret = 0;
 	struct rtw_adapter *padapter = wiphy_to_adapter(wiphy);
@@ -2903,7 +2903,7 @@ static int cfg80211_rtw_stop_ap(struct wiphy *wiphy, struct net_device *ndev)
 }
 
 static int cfg80211_rtw_add_station(struct wiphy *wiphy,
-				    struct net_device *ndev, u8 *mac,
+				    struct net_device *ndev, const u8 *mac,
 				    struct station_parameters *params)
 {
 	DBG_8723A("%s(%s)\n", __func__, ndev->name);
@@ -2912,7 +2912,7 @@ static int cfg80211_rtw_add_station(struct wiphy *wiphy,
 }
 
 static int cfg80211_rtw_del_station(struct wiphy *wiphy,
-				    struct net_device *ndev, u8 *mac)
+				    struct net_device *ndev, const u8 *mac)
 {
 	int ret = 0;
 	struct list_head *phead, *plist, *ptmp;
@@ -2988,7 +2988,7 @@ static int cfg80211_rtw_del_station(struct wiphy *wiphy,
 }
 
 static int cfg80211_rtw_change_station(struct wiphy *wiphy,
-				       struct net_device *ndev, u8 *mac,
+				       struct net_device *ndev, const u8 *mac,
 				       struct station_parameters *params)
 {
 	DBG_8723A("%s(%s)\n", __func__, ndev->name);
diff --git a/drivers/staging/rtl8723au/os_dep/os_intfs.c b/drivers/staging/rtl8723au/os_dep/os_intfs.c
index 4e32003a4437..1fb34386a4e5 100644
--- a/drivers/staging/rtl8723au/os_dep/os_intfs.c
+++ b/drivers/staging/rtl8723au/os_dep/os_intfs.c
@@ -29,7 +29,9 @@ MODULE_AUTHOR("Realtek Semiconductor Corp.");
 MODULE_AUTHOR("Larry Finger <Larry.Finger@lwfinger.net>");
 MODULE_AUTHOR("Jes Sorensen <Jes.Sorensen@redhat.com>");
 MODULE_VERSION(DRIVERVERSION);
-MODULE_FIRMWARE("rtlwifi/rtl8821aefw.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8723aufw_A.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8723aufw_B.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8723aufw_B_NoBT.bin");
 
 /* module param defaults */
 static int rtw_chip_version = 0x00;
diff --git a/drivers/staging/rtl8821ae/core.c b/drivers/staging/rtl8821ae/core.c
index 9a37408708f4..046be2ce9c1a 100644
--- a/drivers/staging/rtl8821ae/core.c
+++ b/drivers/staging/rtl8821ae/core.c
@@ -1278,7 +1278,9 @@ static void rtl_op_rfkill_poll(struct ieee80211_hw *hw)
  * before switch channel or power save, or tx buffer packet
  * maybe send after offchannel or rf sleep, this may cause
  * dis-association by AP */
-static void rtl_op_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
+static void rtl_op_flush(struct ieee80211_hw *hw,
+			 struct ieee80211_vif *vif,
+			 u32 queues, bool drop)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
diff --git a/drivers/staging/tidspbridge/core/tiomap3430.c b/drivers/staging/tidspbridge/core/tiomap3430.c
index 8945b4e3a2a6..cb50120ed7b5 100644
--- a/drivers/staging/tidspbridge/core/tiomap3430.c
+++ b/drivers/staging/tidspbridge/core/tiomap3430.c
@@ -280,8 +280,10 @@ static int bridge_brd_monitor(struct bridge_dev_context *dev_ctxt)
 					OMAP3430_IVA2_MOD, OMAP2_CM_CLKSTCTRL);
 
 		/* Wait until the state has moved to ON */
-		while (*pdata->dsp_prm_read(OMAP3430_IVA2_MOD, OMAP2_PM_PWSTST)&
-					OMAP_INTRANSITION_MASK);
+		while ((*pdata->dsp_prm_read)(OMAP3430_IVA2_MOD,
+					      OMAP2_PM_PWSTST) &
+						OMAP_INTRANSITION_MASK)
+			;
 		/* Disable Automatic transition */
 		(*pdata->dsp_cm_write)(OMAP34XX_CLKSTCTRL_DISABLE_AUTO,
 					OMAP3430_IVA2_MOD, OMAP2_CM_CLKSTCTRL);
diff --git a/drivers/staging/wlan-ng/cfg80211.c b/drivers/staging/wlan-ng/cfg80211.c
index f76f95c29617..723319ee08f3 100644
--- a/drivers/staging/wlan-ng/cfg80211.c
+++ b/drivers/staging/wlan-ng/cfg80211.c
@@ -84,7 +84,7 @@ static int prism2_domibset_uint32(wlandevice_t *wlandev, u32 did, u32 data)
 }
 
 static int prism2_domibset_pstr32(wlandevice_t *wlandev,
-				  u32 did, u8 len, u8 *data)
+				  u32 did, u8 len, const u8 *data)
 {
 	struct p80211msg_dot11req_mibset msg;
 	p80211item_pstr32_t *mibitem =
@@ -298,7 +298,7 @@ static int prism2_set_default_key(struct wiphy *wiphy, struct net_device *dev,
 
 
 static int prism2_get_station(struct wiphy *wiphy, struct net_device *dev,
-			      u8 *mac, struct station_info *sinfo)
+			      const u8 *mac, struct station_info *sinfo)
 {
 	wlandevice_t *wlandev = dev->ml_priv;
 	struct p80211msg_lnxreq_commsquality quality;
diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 9189bc0a87ae..1f4c794f5fcc 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -300,7 +300,7 @@ bool iscsit_check_np_match(
 		port = ntohs(sock_in->sin_port);
 	}
 
-	if ((ip_match == true) && (np->np_port == port) &&
+	if (ip_match && (np->np_port == port) &&
 	    (np->np_network_transport == network_transport))
 		return true;
 
@@ -325,7 +325,7 @@ static struct iscsi_np *iscsit_get_np(
 		}
 
 		match = iscsit_check_np_match(sockaddr, np, network_transport);
-		if (match == true) {
+		if (match) {
 			/*
 			 * Increment the np_exports reference count now to
 			 * prevent iscsit_del_np() below from being called
@@ -1121,7 +1121,7 @@ iscsit_get_immediate_data(struct iscsi_cmd *cmd, struct iscsi_scsi_req *hdr,
 	/*
 	 * Special case for Unsupported SAM WRITE Opcodes and ImmediateData=Yes.
 	 */
-	if (dump_payload == true)
+	if (dump_payload)
 		goto after_immediate_data;
 
 	immed_ret = iscsit_handle_immediate_data(cmd, hdr,
@@ -1309,7 +1309,7 @@ iscsit_check_dataout_hdr(struct iscsi_conn *conn, unsigned char *buf,
 	if (cmd->data_direction != DMA_TO_DEVICE) {
 		pr_err("Command ITT: 0x%08x received DataOUT for a"
 			" NON-WRITE command.\n", cmd->init_task_tag);
-		return iscsit_reject_cmd(cmd, ISCSI_REASON_PROTOCOL_ERROR, buf);
+		return iscsit_dump_data_payload(conn, payload_length, 1);
 	}
 	se_cmd = &cmd->se_cmd;
 	iscsit_mod_dataout_timer(cmd);
@@ -3390,7 +3390,9 @@ static bool iscsit_check_inaddr_any(struct iscsi_np *np)
 
 #define SENDTARGETS_BUF_LIMIT 32768U
 
-static int iscsit_build_sendtargets_response(struct iscsi_cmd *cmd)
+static int
+iscsit_build_sendtargets_response(struct iscsi_cmd *cmd,
+				  enum iscsit_transport_type network_transport)
 {
 	char *payload = NULL;
 	struct iscsi_conn *conn = cmd->conn;
@@ -3467,6 +3469,9 @@ static int iscsit_build_sendtargets_response(struct iscsi_cmd *cmd)
 				struct iscsi_np *np = tpg_np->tpg_np;
 				bool inaddr_any = iscsit_check_inaddr_any(np);
 
+				if (np->np_network_transport != network_transport)
+					continue;
+
 				if (!target_name_printed) {
 					len = sprintf(buf, "TargetName=%s",
 						      tiqn->tiqn);
@@ -3485,10 +3490,8 @@ static int iscsit_build_sendtargets_response(struct iscsi_cmd *cmd)
 
 				len = sprintf(buf, "TargetAddress="
 					"%s:%hu,%hu",
-					(inaddr_any == false) ?
-						np->np_ip : conn->local_ip,
-					(inaddr_any == false) ?
-						np->np_port : conn->local_port,
+					inaddr_any ? conn->local_ip : np->np_ip,
+					inaddr_any ? conn->local_port : np->np_port,
 					tpg->tpgt);
 				len += 1;
 
@@ -3520,11 +3523,12 @@ eob:
 
 int
 iscsit_build_text_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn,
-		      struct iscsi_text_rsp *hdr)
+		      struct iscsi_text_rsp *hdr,
+		      enum iscsit_transport_type network_transport)
 {
 	int text_length, padding;
 
-	text_length = iscsit_build_sendtargets_response(cmd);
+	text_length = iscsit_build_sendtargets_response(cmd, network_transport);
 	if (text_length < 0)
 		return text_length;
 
@@ -3562,7 +3566,7 @@ static int iscsit_send_text_rsp(
 	u32 tx_size = 0;
 	int text_length, iov_count = 0, rc;
 
-	rc = iscsit_build_text_rsp(cmd, conn, hdr);
+	rc = iscsit_build_text_rsp(cmd, conn, hdr, ISCSI_TCP);
 	if (rc < 0)
 		return rc;
 
@@ -4234,8 +4238,6 @@ int iscsit_close_connection(
 	if (conn->conn_transport->iscsit_wait_conn)
 		conn->conn_transport->iscsit_wait_conn(conn);
 
-	iscsit_free_queue_reqs_for_conn(conn);
-
 	/*
 	 * During Connection recovery drop unacknowledged out of order
 	 * commands for this connection, and prepare the other commands
@@ -4252,6 +4254,7 @@ int iscsit_close_connection(
 		iscsit_clear_ooo_cmdsns_for_conn(conn);
 		iscsit_release_commands_from_conn(conn);
 	}
+	iscsit_free_queue_reqs_for_conn(conn);
 
 	/*
 	 * Handle decrementing session or connection usage count if
diff --git a/drivers/target/iscsi/iscsi_target_auth.c b/drivers/target/iscsi/iscsi_target_auth.c
index de77d9aa22c6..ab4915c0d933 100644
--- a/drivers/target/iscsi/iscsi_target_auth.c
+++ b/drivers/target/iscsi/iscsi_target_auth.c
@@ -71,6 +71,40 @@ static void chap_gen_challenge(
 			challenge_asciihex);
 }
 
+static int chap_check_algorithm(const char *a_str)
+{
+	char *tmp, *orig, *token;
+
+	tmp = kstrdup(a_str, GFP_KERNEL);
+	if (!tmp) {
+		pr_err("Memory allocation failed for CHAP_A temporary buffer\n");
+		return CHAP_DIGEST_UNKNOWN;
+	}
+	orig = tmp;
+
+	token = strsep(&tmp, "=");
+	if (!token)
+		goto out;
+
+	if (strcmp(token, "CHAP_A")) {
+		pr_err("Unable to locate CHAP_A key\n");
+		goto out;
+	}
+	while (token) {
+		token = strsep(&tmp, ",");
+		if (!token)
+			goto out;
+
+		if (!strncmp(token, "5", 1)) {
+			pr_debug("Selected MD5 Algorithm\n");
+			kfree(orig);
+			return CHAP_DIGEST_MD5;
+		}
+	}
+out:
+	kfree(orig);
+	return CHAP_DIGEST_UNKNOWN;
+}
 
 static struct iscsi_chap *chap_server_open(
 	struct iscsi_conn *conn,
@@ -79,6 +113,7 @@ static struct iscsi_chap *chap_server_open(
 	char *aic_str,
 	unsigned int *aic_len)
 {
+	int ret;
 	struct iscsi_chap *chap;
 
 	if (!(auth->naf_flags & NAF_USERID_SET) ||
@@ -93,21 +128,24 @@ static struct iscsi_chap *chap_server_open(
 		return NULL;
 
 	chap = conn->auth_protocol;
-	/*
-	 * We only support MD5 MDA presently.
-	 */
-	if (strncmp(a_str, "CHAP_A=5", 8)) {
-		pr_err("CHAP_A is not MD5.\n");
+	ret = chap_check_algorithm(a_str);
+	switch (ret) {
+	case CHAP_DIGEST_MD5:
+		pr_debug("[server] Got CHAP_A=5\n");
+		/*
+		 * Send back CHAP_A set to MD5.
+		*/
+		*aic_len = sprintf(aic_str, "CHAP_A=5");
+		*aic_len += 1;
+		chap->digest_type = CHAP_DIGEST_MD5;
+		pr_debug("[server] Sending CHAP_A=%d\n", chap->digest_type);
+		break;
+	case CHAP_DIGEST_UNKNOWN:
+	default:
+		pr_err("Unsupported CHAP_A value\n");
 		return NULL;
 	}
-	pr_debug("[server] Got CHAP_A=5\n");
-	/*
-	 * Send back CHAP_A set to MD5.
-	 */
-	*aic_len = sprintf(aic_str, "CHAP_A=5");
-	*aic_len += 1;
-	chap->digest_type = CHAP_DIGEST_MD5;
-	pr_debug("[server] Sending CHAP_A=%d\n", chap->digest_type);
+
 	/*
 	 * Set Identifier.
 	 */
@@ -136,7 +174,6 @@ static int chap_server_compute_md5(
 	char *nr_out_ptr,
 	unsigned int *nr_out_len)
 {
-	char *endptr;
 	unsigned long id;
 	unsigned char id_as_uchar;
 	unsigned char digest[MD5_SIGNATURE_SIZE];
@@ -282,9 +319,14 @@ static int chap_server_compute_md5(
 	}
 
 	if (type == HEX)
-		id = simple_strtoul(&identifier[2], &endptr, 0);
+		ret = kstrtoul(&identifier[2], 0, &id);
 	else
-		id = simple_strtoul(identifier, &endptr, 0);
+		ret = kstrtoul(identifier, 0, &id);
+
+	if (ret < 0) {
+		pr_err("kstrtoul() failed for CHAP identifier: %d\n", ret);
+		goto out;
+	}
 	if (id > 255) {
 		pr_err("chap identifier: %lu greater than 255\n", id);
 		goto out;
@@ -313,6 +355,20 @@ static int chap_server_compute_md5(
 		pr_err("Unable to convert incoming challenge\n");
 		goto out;
 	}
+	if (challenge_len > 1024) {
+		pr_err("CHAP_C exceeds maximum binary size of 1024 bytes\n");
+		goto out;
+	}
+	/*
+	 * During mutual authentication, the CHAP_C generated by the
+	 * initiator must not match the original CHAP_C generated by
+	 * the target.
+	 */
+	if (!memcmp(challenge_binhex, chap->challenge, CHAP_CHALLENGE_LENGTH)) {
+		pr_err("initiator CHAP_C matches target CHAP_C, failing"
+		       " login attempt\n");
+		goto out;
+	}
 	/*
 	 * Generate CHAP_N and CHAP_R for mutual authentication.
 	 */
diff --git a/drivers/target/iscsi/iscsi_target_auth.h b/drivers/target/iscsi/iscsi_target_auth.h
index 2f463c09626d..d22f7b96a06c 100644
--- a/drivers/target/iscsi/iscsi_target_auth.h
+++ b/drivers/target/iscsi/iscsi_target_auth.h
@@ -1,6 +1,7 @@
 #ifndef _ISCSI_CHAP_H_
 #define _ISCSI_CHAP_H_
 
+#define CHAP_DIGEST_UNKNOWN	0
 #define CHAP_DIGEST_MD5		5
 #define CHAP_DIGEST_SHA		6
 
diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c
index d9b1d88e1ad3..5e71ac609418 100644
--- a/drivers/target/iscsi/iscsi_target_login.c
+++ b/drivers/target/iscsi/iscsi_target_login.c
@@ -1145,7 +1145,7 @@ iscsit_conn_set_transport(struct iscsi_conn *conn, struct iscsit_transport *t)
 void iscsi_target_login_sess_out(struct iscsi_conn *conn,
 		struct iscsi_np *np, bool zero_tsih, bool new_sess)
 {
-	if (new_sess == false)
+	if (!new_sess)
 		goto old_sess_out;
 
 	pr_err("iSCSI Login negotiation failed.\n");
@@ -1216,7 +1216,7 @@ old_sess_out:
 static int __iscsi_target_login_thread(struct iscsi_np *np)
 {
 	u8 *buffer, zero_tsih = 0;
-	int ret = 0, rc, stop;
+	int ret = 0, rc;
 	struct iscsi_conn *conn = NULL;
 	struct iscsi_login *login;
 	struct iscsi_portal_group *tpg = NULL;
@@ -1230,6 +1230,9 @@ static int __iscsi_target_login_thread(struct iscsi_np *np)
 	if (np->np_thread_state == ISCSI_NP_THREAD_RESET) {
 		np->np_thread_state = ISCSI_NP_THREAD_ACTIVE;
 		complete(&np->np_restart_comp);
+	} else if (np->np_thread_state == ISCSI_NP_THREAD_SHUTDOWN) {
+		spin_unlock_bh(&np->np_thread_lock);
+		goto exit;
 	} else {
 		np->np_thread_state = ISCSI_NP_THREAD_ACTIVE;
 	}
@@ -1422,10 +1425,8 @@ old_sess_out:
 	}
 
 out:
-	stop = kthread_should_stop();
-	/* Wait for another socket.. */
-	if (!stop)
-		return 1;
+	return 1;
+
 exit:
 	iscsi_stop_login_thread_timer(np);
 	spin_lock_bh(&np->np_thread_lock);
@@ -1442,7 +1443,7 @@ int iscsi_target_login_thread(void *arg)
 
 	allow_signal(SIGINT);
 
-	while (!kthread_should_stop()) {
+	while (1) {
 		ret = __iscsi_target_login_thread(np);
 		/*
 		 * We break and exit here unless another sock_accept() call
diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c
index 75b685960e80..62a095f36bf2 100644
--- a/drivers/target/iscsi/iscsi_target_nego.c
+++ b/drivers/target/iscsi/iscsi_target_nego.c
@@ -404,7 +404,7 @@ static void iscsi_target_sk_data_ready(struct sock *sk)
 	}
 
 	rc = schedule_delayed_work(&conn->login_work, 0);
-	if (rc == false) {
+	if (!rc) {
 		pr_debug("iscsi_target_sk_data_ready, schedule_delayed_work"
 			 " got false\n");
 	}
@@ -513,7 +513,7 @@ static void iscsi_target_do_login_rx(struct work_struct *work)
 	state = (tpg->tpg_state == TPG_STATE_ACTIVE);
 	spin_unlock(&tpg->tpg_state_lock);
 
-	if (state == false) {
+	if (!state) {
 		pr_debug("iscsi_target_do_login_rx: tpg_state != TPG_STATE_ACTIVE\n");
 		iscsi_target_restore_sock_callbacks(conn);
 		iscsi_target_login_drop(conn, login);
@@ -528,7 +528,7 @@ static void iscsi_target_do_login_rx(struct work_struct *work)
 		state = iscsi_target_sk_state_check(sk);
 		read_unlock_bh(&sk->sk_callback_lock);
 
-		if (state == false) {
+		if (!state) {
 			pr_debug("iscsi_target_do_login_rx, TCP state CLOSE\n");
 			iscsi_target_restore_sock_callbacks(conn);
 			iscsi_target_login_drop(conn, login);
@@ -773,6 +773,12 @@ static int iscsi_target_handle_csg_zero(
 		}
 
 		goto do_auth;
+	} else if (!payload_length) {
+		pr_err("Initiator sent zero length security payload,"
+		       " login failed\n");
+		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_INITIATOR_ERR,
+				    ISCSI_LOGIN_STATUS_AUTH_FAILED);
+		return -1;
 	}
 
 	if (login->first_request)
diff --git a/drivers/target/iscsi/iscsi_target_parameters.c b/drivers/target/iscsi/iscsi_target_parameters.c
index 4d2e23fc76fd..02f9de26f38a 100644
--- a/drivers/target/iscsi/iscsi_target_parameters.c
+++ b/drivers/target/iscsi/iscsi_target_parameters.c
@@ -474,10 +474,10 @@ int iscsi_set_keys_to_negotiate(
 		if (!strcmp(param->name, AUTHMETHOD)) {
 			SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, HEADERDIGEST)) {
-			if (iser == false)
+			if (!iser)
 				SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, DATADIGEST)) {
-			if (iser == false)
+			if (!iser)
 				SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, MAXCONNECTIONS)) {
 			SET_PSTATE_NEGOTIATE(param);
@@ -497,7 +497,7 @@ int iscsi_set_keys_to_negotiate(
 		} else if (!strcmp(param->name, IMMEDIATEDATA)) {
 			SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, MAXRECVDATASEGMENTLENGTH)) {
-			if (iser == false)
+			if (!iser)
 				SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, MAXXMITDATASEGMENTLENGTH)) {
 			continue;
@@ -528,13 +528,13 @@ int iscsi_set_keys_to_negotiate(
 		} else if (!strcmp(param->name, OFMARKINT)) {
 			SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, RDMAEXTENSIONS)) {
-			if (iser == true)
+			if (iser)
 				SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, INITIATORRECVDATASEGMENTLENGTH)) {
-			if (iser == true)
+			if (iser)
 				SET_PSTATE_NEGOTIATE(param);
 		} else if (!strcmp(param->name, TARGETRECVDATASEGMENTLENGTH)) {
-			if (iser == true)
+			if (iser)
 				SET_PSTATE_NEGOTIATE(param);
 		}
 	}
@@ -1605,7 +1605,7 @@ int iscsi_decode_text_input(
 
 	tmpbuf = kzalloc(length + 1, GFP_KERNEL);
 	if (!tmpbuf) {
-		pr_err("Unable to allocate memory for tmpbuf.\n");
+		pr_err("Unable to allocate %u + 1 bytes for tmpbuf.\n", length);
 		return -1;
 	}
 
diff --git a/drivers/target/iscsi/iscsi_target_tpg.c b/drivers/target/iscsi/iscsi_target_tpg.c
index 1431e8400d28..c3cb5c15efda 100644
--- a/drivers/target/iscsi/iscsi_target_tpg.c
+++ b/drivers/target/iscsi/iscsi_target_tpg.c
@@ -189,7 +189,7 @@ static void iscsit_clear_tpg_np_login_thread(
 	iscsit_reset_np_thread(tpg_np->tpg_np, tpg_np, tpg, shutdown);
 }
 
-void iscsit_clear_tpg_np_login_threads(
+static void iscsit_clear_tpg_np_login_threads(
 	struct iscsi_portal_group *tpg,
 	bool shutdown)
 {
@@ -276,8 +276,6 @@ int iscsit_tpg_del_portal_group(
 	tpg->tpg_state = TPG_STATE_INACTIVE;
 	spin_unlock(&tpg->tpg_state_lock);
 
-	iscsit_clear_tpg_np_login_threads(tpg, true);
-
 	if (iscsit_release_sessions_for_tpg(tpg, force) < 0) {
 		pr_err("Unable to delete iSCSI Target Portal Group:"
 			" %hu while active sessions exist, and force=0\n",
@@ -453,7 +451,7 @@ static bool iscsit_tpg_check_network_portal(
 
 			match = iscsit_check_np_match(sockaddr, np,
 						network_transport);
-			if (match == true)
+			if (match)
 				break;
 		}
 		spin_unlock(&tpg->tpg_np_lock);
@@ -475,7 +473,7 @@ struct iscsi_tpg_np *iscsit_tpg_add_network_portal(
 
 	if (!tpg_np_parent) {
 		if (iscsit_tpg_check_network_portal(tpg->tpg_tiqn, sockaddr,
-				network_transport) == true) {
+				network_transport)) {
 			pr_err("Network Portal: %s already exists on a"
 				" different TPG on %s\n", ip_str,
 				tpg->tpg_tiqn->tiqn);
diff --git a/drivers/target/iscsi/iscsi_target_tpg.h b/drivers/target/iscsi/iscsi_target_tpg.h
index 0a182f2aa8a2..e7265337bc43 100644
--- a/drivers/target/iscsi/iscsi_target_tpg.h
+++ b/drivers/target/iscsi/iscsi_target_tpg.h
@@ -8,7 +8,6 @@ extern struct iscsi_portal_group *iscsit_get_tpg_from_np(struct iscsi_tiqn *,
 			struct iscsi_np *, struct iscsi_tpg_np **);
 extern int iscsit_get_tpg(struct iscsi_portal_group *);
 extern void iscsit_put_tpg(struct iscsi_portal_group *);
-extern void iscsit_clear_tpg_np_login_threads(struct iscsi_portal_group *, bool);
 extern void iscsit_tpg_dump_params(struct iscsi_portal_group *);
 extern int iscsit_tpg_add_portal_group(struct iscsi_tiqn *, struct iscsi_portal_group *);
 extern int iscsit_tpg_del_portal_group(struct iscsi_tiqn *, struct iscsi_portal_group *,
diff --git a/drivers/target/iscsi/iscsi_target_util.c b/drivers/target/iscsi/iscsi_target_util.c
index 53e157cb8c54..fd90b28f1d94 100644
--- a/drivers/target/iscsi/iscsi_target_util.c
+++ b/drivers/target/iscsi/iscsi_target_util.c
@@ -1295,6 +1295,8 @@ int iscsit_tx_login_rsp(struct iscsi_conn *conn, u8 status_class, u8 status_deta
 	login->login_failed = 1;
 	iscsit_collect_login_stats(conn, status_class, status_detail);
 
+	memset(&login->rsp[0], 0, ISCSI_HDR_LEN);
+
 	hdr	= (struct iscsi_login_rsp *)&login->rsp[0];
 	hdr->opcode		= ISCSI_OP_LOGIN_RSP;
 	hdr->status_class	= status_class;
diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c
index 73ab75ddaf42..8c64b8776a96 100644
--- a/drivers/target/loopback/tcm_loop.c
+++ b/drivers/target/loopback/tcm_loop.c
@@ -179,7 +179,7 @@ static void tcm_loop_submission_work(struct work_struct *work)
 	struct tcm_loop_hba *tl_hba;
 	struct tcm_loop_tpg *tl_tpg;
 	struct scatterlist *sgl_bidi = NULL;
-	u32 sgl_bidi_count = 0;
+	u32 sgl_bidi_count = 0, transfer_length;
 	int rc;
 
 	tl_hba = *(struct tcm_loop_hba **)shost_priv(sc->device->host);
@@ -213,12 +213,21 @@ static void tcm_loop_submission_work(struct work_struct *work)
 
 	}
 
-	if (!scsi_prot_sg_count(sc) && scsi_get_prot_op(sc) != SCSI_PROT_NORMAL)
+	transfer_length = scsi_transfer_length(sc);
+	if (!scsi_prot_sg_count(sc) &&
+	    scsi_get_prot_op(sc) != SCSI_PROT_NORMAL) {
 		se_cmd->prot_pto = true;
+		/*
+		 * loopback transport doesn't support
+		 * WRITE_GENERATE, READ_STRIP protection
+		 * information operations, go ahead unprotected.
+		 */
+		transfer_length = scsi_bufflen(sc);
+	}
 
 	rc = target_submit_cmd_map_sgls(se_cmd, tl_nexus->se_sess, sc->cmnd,
 			&tl_cmd->tl_sense_buf[0], tl_cmd->sc->device->lun,
-			scsi_bufflen(sc), tcm_loop_sam_attr(sc),
+			transfer_length, tcm_loop_sam_attr(sc),
 			sc->sc_data_direction, 0,
 			scsi_sglist(sc), scsi_sg_count(sc),
 			sgl_bidi, sgl_bidi_count,
@@ -230,6 +239,7 @@ static void tcm_loop_submission_work(struct work_struct *work)
 	return;
 
 out_done:
+	kmem_cache_free(tcm_loop_cmd_cache, tl_cmd);
 	sc->scsi_done(sc);
 	return;
 }
diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index 11d26fe65bfb..98da90167159 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -616,6 +616,7 @@ void core_dev_unexport(
 	dev->export_count--;
 	spin_unlock(&hba->device_lock);
 
+	lun->lun_sep = NULL;
 	lun->lun_se_dev = NULL;
 }
 
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index 0f199f6a0738..94d00df28f39 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -1055,6 +1055,8 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 			ret = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 			goto fail;
 		}
+
+		blk_rq_set_block_pc(req);
 	} else {
 		BUG_ON(!cmd->data_length);
 
@@ -1071,7 +1073,6 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 		}
 	}
 
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->end_io = pscsi_req_done;
 	req->end_io_data = cmd;
 	req->cmd_len = scsi_command_size(pt->pscsi_cdb);
diff --git a/drivers/target/target_core_sbc.c b/drivers/target/target_core_sbc.c
index e0229592ec55..bd78d9235ac6 100644
--- a/drivers/target/target_core_sbc.c
+++ b/drivers/target/target_core_sbc.c
@@ -81,7 +81,7 @@ sbc_emulate_readcapacity(struct se_cmd *cmd)
 		transport_kunmap_data_sg(cmd);
 	}
 
-	target_complete_cmd(cmd, GOOD);
+	target_complete_cmd_with_length(cmd, GOOD, 8);
 	return 0;
 }
 
@@ -137,7 +137,7 @@ sbc_emulate_readcapacity_16(struct se_cmd *cmd)
 		transport_kunmap_data_sg(cmd);
 	}
 
-	target_complete_cmd(cmd, GOOD);
+	target_complete_cmd_with_length(cmd, GOOD, 32);
 	return 0;
 }
 
@@ -176,24 +176,6 @@ static inline u32 sbc_get_size(struct se_cmd *cmd, u32 sectors)
 	return cmd->se_dev->dev_attrib.block_size * sectors;
 }
 
-static int sbc_check_valid_sectors(struct se_cmd *cmd)
-{
-	struct se_device *dev = cmd->se_dev;
-	unsigned long long end_lba;
-	u32 sectors;
-
-	sectors = cmd->data_length / dev->dev_attrib.block_size;
-	end_lba = dev->transport->get_blocks(dev) + 1;
-
-	if (cmd->t_task_lba + sectors > end_lba) {
-		pr_err("target: lba %llu, sectors %u exceeds end lba %llu\n",
-			cmd->t_task_lba, sectors, end_lba);
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static inline u32 transport_get_sectors_6(unsigned char *cdb)
 {
 	/*
@@ -665,8 +647,19 @@ sbc_check_prot(struct se_device *dev, struct se_cmd *cmd, unsigned char *cdb,
 
 	cmd->prot_type = dev->dev_attrib.pi_prot_type;
 	cmd->prot_length = dev->prot_length * sectors;
-	pr_debug("%s: prot_type=%d, prot_length=%d prot_op=%d prot_checks=%d\n",
-		 __func__, cmd->prot_type, cmd->prot_length,
+
+	/**
+	 * In case protection information exists over the wire
+	 * we modify command data length to describe pure data.
+	 * The actual transfer length is data length + protection
+	 * length
+	 **/
+	if (protect)
+		cmd->data_length = sectors * dev->dev_attrib.block_size;
+
+	pr_debug("%s: prot_type=%d, data_length=%d, prot_length=%d "
+		 "prot_op=%d prot_checks=%d\n",
+		 __func__, cmd->prot_type, cmd->data_length, cmd->prot_length,
 		 cmd->prot_op, cmd->prot_checks);
 
 	return true;
@@ -877,15 +870,6 @@ sbc_parse_cdb(struct se_cmd *cmd, struct sbc_ops *ops)
 		break;
 	case SYNCHRONIZE_CACHE:
 	case SYNCHRONIZE_CACHE_16:
-		if (!ops->execute_sync_cache) {
-			size = 0;
-			cmd->execute_cmd = sbc_emulate_noop;
-			break;
-		}
-
-		/*
-		 * Extract LBA and range to be flushed for emulated SYNCHRONIZE_CACHE
-		 */
 		if (cdb[0] == SYNCHRONIZE_CACHE) {
 			sectors = transport_get_sectors_10(cdb);
 			cmd->t_task_lba = transport_lba_32(cdb);
@@ -893,18 +877,12 @@ sbc_parse_cdb(struct se_cmd *cmd, struct sbc_ops *ops)
 			sectors = transport_get_sectors_16(cdb);
 			cmd->t_task_lba = transport_lba_64(cdb);
 		}
-
-		size = sbc_get_size(cmd, sectors);
-
-		/*
-		 * Check to ensure that LBA + Range does not exceed past end of
-		 * device for IBLOCK and FILEIO ->do_sync_cache() backend calls
-		 */
-		if (cmd->t_task_lba || sectors) {
-			if (sbc_check_valid_sectors(cmd) < 0)
-				return TCM_ADDRESS_OUT_OF_RANGE;
+		if (ops->execute_sync_cache) {
+			cmd->execute_cmd = ops->execute_sync_cache;
+			goto check_lba;
 		}
-		cmd->execute_cmd = ops->execute_sync_cache;
+		size = 0;
+		cmd->execute_cmd = sbc_emulate_noop;
 		break;
 	case UNMAP:
 		if (!ops->execute_unmap)
@@ -947,8 +925,10 @@ sbc_parse_cdb(struct se_cmd *cmd, struct sbc_ops *ops)
 		break;
 	case VERIFY:
 		size = 0;
+		sectors = transport_get_sectors_10(cdb);
+		cmd->t_task_lba = transport_lba_32(cdb);
 		cmd->execute_cmd = sbc_emulate_noop;
-		break;
+		goto check_lba;
 	case REZERO_UNIT:
 	case SEEK_6:
 	case SEEK_10:
@@ -988,7 +968,7 @@ sbc_parse_cdb(struct se_cmd *cmd, struct sbc_ops *ops)
 				dev->dev_attrib.hw_max_sectors);
 			return TCM_INVALID_CDB_FIELD;
 		}
-
+check_lba:
 		end_lba = dev->transport->get_blocks(dev) + 1;
 		if (cmd->t_task_lba + sectors > end_lba) {
 			pr_err("cmd exceeds last lba %llu "
diff --git a/drivers/target/target_core_spc.c b/drivers/target/target_core_spc.c
index 8653666612a8..6cd7222738fc 100644
--- a/drivers/target/target_core_spc.c
+++ b/drivers/target/target_core_spc.c
@@ -129,15 +129,10 @@ static sense_reason_t
 spc_emulate_evpd_80(struct se_cmd *cmd, unsigned char *buf)
 {
 	struct se_device *dev = cmd->se_dev;
-	u16 len = 0;
+	u16 len;
 
 	if (dev->dev_flags & DF_EMULATED_VPD_UNIT_SERIAL) {
-		u32 unit_serial_len;
-
-		unit_serial_len = strlen(dev->t10_wwn.unit_serial);
-		unit_serial_len++; /* For NULL Terminator */
-
-		len += sprintf(&buf[4], "%s", dev->t10_wwn.unit_serial);
+		len = sprintf(&buf[4], "%s", dev->t10_wwn.unit_serial);
 		len++; /* Extra Byte for NULL Terminator */
 		buf[3] = len;
 	}
@@ -721,6 +716,7 @@ spc_emulate_inquiry(struct se_cmd *cmd)
 	unsigned char *buf;
 	sense_reason_t ret;
 	int p;
+	int len = 0;
 
 	buf = kzalloc(SE_INQUIRY_BUF, GFP_KERNEL);
 	if (!buf) {
@@ -742,6 +738,7 @@ spc_emulate_inquiry(struct se_cmd *cmd)
 		}
 
 		ret = spc_emulate_inquiry_std(cmd, buf);
+		len = buf[4] + 5;
 		goto out;
 	}
 
@@ -749,6 +746,7 @@ spc_emulate_inquiry(struct se_cmd *cmd)
 		if (cdb[2] == evpd_handlers[p].page) {
 			buf[1] = cdb[2];
 			ret = evpd_handlers[p].emulate(cmd, buf);
+			len = get_unaligned_be16(&buf[2]) + 4;
 			goto out;
 		}
 	}
@@ -765,7 +763,7 @@ out:
 	kfree(buf);
 
 	if (!ret)
-		target_complete_cmd(cmd, GOOD);
+		target_complete_cmd_with_length(cmd, GOOD, len);
 	return ret;
 }
 
@@ -1103,7 +1101,7 @@ set_length:
 		transport_kunmap_data_sg(cmd);
 	}
 
-	target_complete_cmd(cmd, GOOD);
+	target_complete_cmd_with_length(cmd, GOOD, length);
 	return 0;
 }
 
@@ -1279,7 +1277,7 @@ done:
 	buf[3] = (lun_count & 0xff);
 	transport_kunmap_data_sg(cmd);
 
-	target_complete_cmd(cmd, GOOD);
+	target_complete_cmd_with_length(cmd, GOOD, 8 + lun_count * 8);
 	return 0;
 }
 EXPORT_SYMBOL(spc_emulate_report_luns);
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 2179feed0d63..7fa62fc93e0b 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -504,7 +504,7 @@ void transport_deregister_session(struct se_session *se_sess)
 	 * ->acl_free_comp caller to wakeup configfs se_node_acl->acl_group
 	 * removal context.
 	 */
-	if (se_nacl && comp_nacl == true)
+	if (se_nacl && comp_nacl)
 		target_put_nacl(se_nacl);
 
 	transport_free_session(se_sess);
@@ -562,7 +562,7 @@ static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists,
 
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
-		complete(&cmd->t_transport_stop_comp);
+		complete_all(&cmd->t_transport_stop_comp);
 		return 1;
 	}
 
@@ -687,7 +687,7 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status)
 	if (cmd->transport_state & CMD_T_ABORTED &&
 	    cmd->transport_state & CMD_T_STOP) {
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-		complete(&cmd->t_transport_stop_comp);
+		complete_all(&cmd->t_transport_stop_comp);
 		return;
 	} else if (!success) {
 		INIT_WORK(&cmd->work, target_complete_failure_work);
@@ -703,6 +703,23 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status)
 }
 EXPORT_SYMBOL(target_complete_cmd);
 
+void target_complete_cmd_with_length(struct se_cmd *cmd, u8 scsi_status, int length)
+{
+	if (scsi_status == SAM_STAT_GOOD && length < cmd->data_length) {
+		if (cmd->se_cmd_flags & SCF_UNDERFLOW_BIT) {
+			cmd->residual_count += cmd->data_length - length;
+		} else {
+			cmd->se_cmd_flags |= SCF_UNDERFLOW_BIT;
+			cmd->residual_count = cmd->data_length - length;
+		}
+
+		cmd->data_length = length;
+	}
+
+	target_complete_cmd(cmd, scsi_status);
+}
+EXPORT_SYMBOL(target_complete_cmd_with_length);
+
 static void target_add_to_state_list(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
@@ -1761,7 +1778,7 @@ void target_execute_cmd(struct se_cmd *cmd)
 			cmd->se_tfo->get_task_tag(cmd));
 
 		spin_unlock_irq(&cmd->t_state_lock);
-		complete(&cmd->t_transport_stop_comp);
+		complete_all(&cmd->t_transport_stop_comp);
 		return;
 	}
 
@@ -2363,7 +2380,7 @@ int target_get_sess_cmd(struct se_session *se_sess, struct se_cmd *se_cmd,
 	 * fabric acknowledgement that requires two target_put_sess_cmd()
 	 * invocations before se_cmd descriptor release.
 	 */
-	if (ack_kref == true) {
+	if (ack_kref) {
 		kref_get(&se_cmd->cmd_kref);
 		se_cmd->se_cmd_flags |= SCF_ACK_KREF;
 	}
@@ -2407,6 +2424,10 @@ static void target_release_cmd_kref(struct kref *kref)
  */
 int target_put_sess_cmd(struct se_session *se_sess, struct se_cmd *se_cmd)
 {
+	if (!se_sess) {
+		se_cmd->se_tfo->release_cmd(se_cmd);
+		return 1;
+	}
 	return kref_put_spinlock_irqsave(&se_cmd->cmd_kref, target_release_cmd_kref,
 			&se_sess->sess_cmd_lock);
 }
@@ -2934,6 +2955,12 @@ static void target_tmr_work(struct work_struct *work)
 int transport_generic_handle_tmr(
 	struct se_cmd *cmd)
 {
+	unsigned long flags;
+
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	cmd->transport_state |= CMD_T_ACTIVE;
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+
 	INIT_WORK(&cmd->work, target_tmr_work);
 	queue_work(cmd->se_dev->tmr_wq, &cmd->work);
 	return 0;
diff --git a/drivers/target/target_core_xcopy.c b/drivers/target/target_core_xcopy.c
index 669c536fd959..e9186cdf35e9 100644
--- a/drivers/target/target_core_xcopy.c
+++ b/drivers/target/target_core_xcopy.c
@@ -70,7 +70,7 @@ static int target_xcopy_locate_se_dev_e4(struct se_cmd *se_cmd, struct xcopy_op
 	unsigned char tmp_dev_wwn[XCOPY_NAA_IEEE_REGEX_LEN], *dev_wwn;
 	int rc;
 
-	if (src == true)
+	if (src)
 		dev_wwn = &xop->dst_tid_wwn[0];
 	else
 		dev_wwn = &xop->src_tid_wwn[0];
@@ -88,7 +88,7 @@ static int target_xcopy_locate_se_dev_e4(struct se_cmd *se_cmd, struct xcopy_op
 		if (rc != 0)
 			continue;
 
-		if (src == true) {
+		if (src) {
 			xop->dst_dev = se_dev;
 			pr_debug("XCOPY 0xe4: Setting xop->dst_dev: %p from located"
 				" se_dev\n", xop->dst_dev);
@@ -166,7 +166,7 @@ static int target_xcopy_parse_tiddesc_e4(struct se_cmd *se_cmd, struct xcopy_op
 		return -EINVAL;
 	}
 
-	if (src == true) {
+	if (src) {
 		memcpy(&xop->src_tid_wwn[0], &desc[8], XCOPY_NAA_IEEE_REGEX_LEN);
 		/*
 		 * Determine if the source designator matches the local device
@@ -236,7 +236,7 @@ static int target_xcopy_parse_target_descriptors(struct se_cmd *se_cmd,
 			/*
 			 * Assume target descriptors are in source -> destination order..
 			 */
-			if (src == true)
+			if (src)
 				src = false;
 			else
 				src = true;
@@ -560,7 +560,7 @@ static int target_xcopy_init_pt_lun(
 	 * reservations.  The pt_cmd->se_lun pointer will be setup from within
 	 * target_xcopy_setup_pt_port()
 	 */
-	if (remote_port == false) {
+	if (!remote_port) {
 		pt_cmd->se_cmd_flags |= SCF_SE_LUN_CMD | SCF_CMD_XCOPY_PASSTHROUGH;
 		return 0;
 	}
diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c
index f5fd515b2bee..be0c0d08c56a 100644
--- a/drivers/target/tcm_fc/tfc_cmd.c
+++ b/drivers/target/tcm_fc/tfc_cmd.c
@@ -128,6 +128,7 @@ int ft_queue_status(struct se_cmd *se_cmd)
 	struct fc_lport *lport;
 	struct fc_exch *ep;
 	size_t len;
+	int rc;
 
 	if (cmd->aborted)
 		return 0;
@@ -137,9 +138,10 @@ int ft_queue_status(struct se_cmd *se_cmd)
 	len = sizeof(*fcp) + se_cmd->scsi_sense_length;
 	fp = fc_frame_alloc(lport, len);
 	if (!fp) {
-		/* XXX shouldn't just drop it - requeue and retry? */
-		return 0;
+		se_cmd->scsi_status = SAM_STAT_TASK_SET_FULL;
+		return -ENOMEM;
 	}
+
 	fcp = fc_frame_payload_get(fp, len);
 	memset(fcp, 0, len);
 	fcp->resp.fr_status = se_cmd->scsi_status;
@@ -170,7 +172,18 @@ int ft_queue_status(struct se_cmd *se_cmd)
 	fc_fill_fc_hdr(fp, FC_RCTL_DD_CMD_STATUS, ep->did, ep->sid, FC_TYPE_FCP,
 		       FC_FC_EX_CTX | FC_FC_LAST_SEQ | FC_FC_END_SEQ, 0);
 
-	lport->tt.seq_send(lport, cmd->seq, fp);
+	rc = lport->tt.seq_send(lport, cmd->seq, fp);
+	if (rc) {
+		pr_info_ratelimited("%s: Failed to send response frame %p, "
+				    "xid <0x%x>\n", __func__, fp, ep->xid);
+		/*
+		 * Generate a TASK_SET_FULL status to notify the initiator
+		 * to reduce it's queue_depth after the se_cmd response has
+		 * been re-queued by target-core.
+		 */
+		se_cmd->scsi_status = SAM_STAT_TASK_SET_FULL;
+		return -ENOMEM;
+	}
 	lport->tt.exch_done(cmd->seq);
 	return 0;
 }
diff --git a/drivers/target/tcm_fc/tfc_io.c b/drivers/target/tcm_fc/tfc_io.c
index e415af32115a..97b486c3dda1 100644
--- a/drivers/target/tcm_fc/tfc_io.c
+++ b/drivers/target/tcm_fc/tfc_io.c
@@ -82,6 +82,10 @@ int ft_queue_data_in(struct se_cmd *se_cmd)
 
 	if (cmd->aborted)
 		return 0;
+
+	if (se_cmd->scsi_status == SAM_STAT_TASK_SET_FULL)
+		goto queue_status;
+
 	ep = fc_seq_exch(cmd->seq);
 	lport = ep->lp;
 	cmd->seq = lport->tt.seq_start_next(cmd->seq);
@@ -178,14 +182,23 @@ int ft_queue_data_in(struct se_cmd *se_cmd)
 			       FC_TYPE_FCP, f_ctl, fh_off);
 		error = lport->tt.seq_send(lport, seq, fp);
 		if (error) {
-			/* XXX For now, initiator will retry */
-			pr_err_ratelimited("%s: Failed to send frame %p, "
+			pr_info_ratelimited("%s: Failed to send frame %p, "
 						"xid <0x%x>, remaining %zu, "
 						"lso_max <0x%x>\n",
 						__func__, fp, ep->xid,
 						remaining, lport->lso_max);
+			/*
+			 * Go ahead and set TASK_SET_FULL status ignoring the
+			 * rest of the DataIN, and immediately attempt to
+			 * send the response via ft_queue_status() in order
+			 * to notify the initiator that it should reduce it's
+			 * per LUN queue_depth.
+			 */
+			se_cmd->scsi_status = SAM_STAT_TASK_SET_FULL;
+			break;
 		}
 	}
+queue_status:
 	return ft_queue_status(se_cmd);
 }
 
diff --git a/drivers/tc/tc.c b/drivers/tc/tc.c
index a8aaf6ac2ae2..946562389ca8 100644
--- a/drivers/tc/tc.c
+++ b/drivers/tc/tc.c
@@ -129,7 +129,10 @@ static void __init tc_bus_add_devices(struct tc_bus *tbus)
 
 		tc_device_get_irq(tdev);
 
-		device_register(&tdev->dev);
+		if (device_register(&tdev->dev)) {
+			put_device(&tdev->dev);
+			goto out_err;
+		}
 		list_add_tail(&tdev->node, &tbus->devices);
 
 out_err:
@@ -148,7 +151,10 @@ static int __init tc_init(void)
 
 	INIT_LIST_HEAD(&tc_bus.devices);
 	dev_set_name(&tc_bus.dev, "tc");
-	device_register(&tc_bus.dev);
+	if (device_register(&tc_bus.dev)) {
+		put_device(&tc_bus.dev);
+		return 0;
+	}
 
 	if (tc_bus.info.slot_size) {
 		unsigned int tc_clock = tc_get_speed(&tc_bus) / 100000;
diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index 2d51912a6e40..f9a13867cb70 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -222,12 +222,24 @@ config ACPI_INT3403_THERMAL
 	  the Intel Thermal Daemon can use this information to allow the user
 	  to select his laptop to run without turning on the fans.
 
+config INTEL_SOC_DTS_THERMAL
+	tristate "Intel SoCs DTS thermal driver"
+	depends on X86 && IOSF_MBI
+	help
+	  Enable this to register Intel SoCs (e.g. Bay Trail) platform digital
+	  temperature sensor (DTS). These SoCs have two additional DTSs in
+	  addition to DTSs on CPU cores. Each DTS will be registered as a
+	  thermal zone. There are two trip points. One of the trip point can
+	  be set by user mode programs to get notifications via Linux thermal
+	  notification methods.The other trip is a critical trip point, which
+	  was set by the driver based on the TJ MAX temperature.
+
 menu "Texas Instruments thermal drivers"
 source "drivers/thermal/ti-soc-thermal/Kconfig"
 endmenu
 
 menu "Samsung thermal drivers"
-depends on PLAT_SAMSUNG
+depends on ARCH_EXYNOS
 source "drivers/thermal/samsung/Kconfig"
 endmenu
 
diff --git a/drivers/thermal/Makefile b/drivers/thermal/Makefile
index 54e4ec9eb5df..de0636a57a64 100644
--- a/drivers/thermal/Makefile
+++ b/drivers/thermal/Makefile
@@ -29,5 +29,6 @@ obj-$(CONFIG_IMX_THERMAL)	+= imx_thermal.o
 obj-$(CONFIG_DB8500_CPUFREQ_COOLING)	+= db8500_cpufreq_cooling.o
 obj-$(CONFIG_INTEL_POWERCLAMP)	+= intel_powerclamp.o
 obj-$(CONFIG_X86_PKG_TEMP_THERMAL)	+= x86_pkg_temp_thermal.o
+obj-$(CONFIG_INTEL_SOC_DTS_THERMAL)	+= intel_soc_dts_thermal.o
 obj-$(CONFIG_TI_SOC_THERMAL)	+= ti-soc-thermal/
 obj-$(CONFIG_ACPI_INT3403_THERMAL)	+= int3403_thermal.o
diff --git a/drivers/thermal/armada_thermal.c b/drivers/thermal/armada_thermal.c
index 5e53212b984f..9d1420acb391 100644
--- a/drivers/thermal/armada_thermal.c
+++ b/drivers/thermal/armada_thermal.c
@@ -24,10 +24,7 @@
 #include <linux/of_device.h>
 #include <linux/thermal.h>
 
-#define THERMAL_VALID_OFFSET		9
 #define THERMAL_VALID_MASK		0x1
-#define THERMAL_TEMP_OFFSET		10
-#define THERMAL_TEMP_MASK		0x1ff
 
 /* Thermal Manager Control and Status Register */
 #define PMU_TDC0_SW_RST_MASK		(0x1 << 1)
@@ -38,24 +35,47 @@
 #define PMU_TDC0_OTF_CAL_MASK		(0x1 << 30)
 #define PMU_TDC0_START_CAL_MASK		(0x1 << 25)
 
-struct armada_thermal_ops;
+#define A375_Z1_CAL_RESET_LSB		0x8011e214
+#define A375_Z1_CAL_RESET_MSB		0x30a88019
+#define A375_Z1_WORKAROUND_BIT		BIT(9)
+
+#define A375_UNIT_CONTROL_SHIFT		27
+#define A375_UNIT_CONTROL_MASK		0x7
+#define A375_READOUT_INVERT		BIT(15)
+#define A375_HW_RESETn			BIT(8)
+#define A380_HW_RESET			BIT(8)
+
+struct armada_thermal_data;
 
 /* Marvell EBU Thermal Sensor Dev Structure */
 struct armada_thermal_priv {
 	void __iomem *sensor;
 	void __iomem *control;
-	struct armada_thermal_ops *ops;
+	struct armada_thermal_data *data;
 };
 
-struct armada_thermal_ops {
+struct armada_thermal_data {
 	/* Initialize the sensor */
-	void (*init_sensor)(struct armada_thermal_priv *);
+	void (*init_sensor)(struct platform_device *pdev,
+			    struct armada_thermal_priv *);
 
 	/* Test for a valid sensor value (optional) */
 	bool (*is_valid)(struct armada_thermal_priv *);
+
+	/* Formula coeficients: temp = (b + m * reg) / div */
+	unsigned long coef_b;
+	unsigned long coef_m;
+	unsigned long coef_div;
+	bool inverted;
+
+	/* Register shift and mask to access the sensor temperature */
+	unsigned int temp_shift;
+	unsigned int temp_mask;
+	unsigned int is_valid_shift;
 };
 
-static void armadaxp_init_sensor(struct armada_thermal_priv *priv)
+static void armadaxp_init_sensor(struct platform_device *pdev,
+				 struct armada_thermal_priv *priv)
 {
 	unsigned long reg;
 
@@ -80,7 +100,8 @@ static void armadaxp_init_sensor(struct armada_thermal_priv *priv)
 	writel(reg, priv->sensor);
 }
 
-static void armada370_init_sensor(struct armada_thermal_priv *priv)
+static void armada370_init_sensor(struct platform_device *pdev,
+				  struct armada_thermal_priv *priv)
 {
 	unsigned long reg;
 
@@ -99,11 +120,54 @@ static void armada370_init_sensor(struct armada_thermal_priv *priv)
 	mdelay(10);
 }
 
+static void armada375_init_sensor(struct platform_device *pdev,
+				  struct armada_thermal_priv *priv)
+{
+	unsigned long reg;
+	bool quirk_needed =
+		!!of_device_is_compatible(pdev->dev.of_node,
+					  "marvell,armada375-z1-thermal");
+
+	if (quirk_needed) {
+		/* Ensure these registers have the default (reset) values */
+		writel(A375_Z1_CAL_RESET_LSB, priv->control);
+		writel(A375_Z1_CAL_RESET_MSB, priv->control + 0x4);
+	}
+
+	reg = readl(priv->control + 4);
+	reg &= ~(A375_UNIT_CONTROL_MASK << A375_UNIT_CONTROL_SHIFT);
+	reg &= ~A375_READOUT_INVERT;
+	reg &= ~A375_HW_RESETn;
+
+	if (quirk_needed)
+		reg |= A375_Z1_WORKAROUND_BIT;
+
+	writel(reg, priv->control + 4);
+	mdelay(20);
+
+	reg |= A375_HW_RESETn;
+	writel(reg, priv->control + 4);
+	mdelay(50);
+}
+
+static void armada380_init_sensor(struct platform_device *pdev,
+				  struct armada_thermal_priv *priv)
+{
+	unsigned long reg = readl_relaxed(priv->control);
+
+	/* Reset hardware once */
+	if (!(reg & A380_HW_RESET)) {
+		reg |= A380_HW_RESET;
+		writel(reg, priv->control);
+		mdelay(10);
+	}
+}
+
 static bool armada_is_valid(struct armada_thermal_priv *priv)
 {
 	unsigned long reg = readl_relaxed(priv->sensor);
 
-	return (reg >> THERMAL_VALID_OFFSET) & THERMAL_VALID_MASK;
+	return (reg >> priv->data->is_valid_shift) & THERMAL_VALID_MASK;
 }
 
 static int armada_get_temp(struct thermal_zone_device *thermal,
@@ -111,17 +175,27 @@ static int armada_get_temp(struct thermal_zone_device *thermal,
 {
 	struct armada_thermal_priv *priv = thermal->devdata;
 	unsigned long reg;
+	unsigned long m, b, div;
 
 	/* Valid check */
-	if (priv->ops->is_valid && !priv->ops->is_valid(priv)) {
+	if (priv->data->is_valid && !priv->data->is_valid(priv)) {
 		dev_err(&thermal->device,
 			"Temperature sensor reading not valid\n");
 		return -EIO;
 	}
 
 	reg = readl_relaxed(priv->sensor);
-	reg = (reg >> THERMAL_TEMP_OFFSET) & THERMAL_TEMP_MASK;
-	*temp = (3153000000UL - (10000000UL*reg)) / 13825;
+	reg = (reg >> priv->data->temp_shift) & priv->data->temp_mask;
+
+	/* Get formula coeficients */
+	b = priv->data->coef_b;
+	m = priv->data->coef_m;
+	div = priv->data->coef_div;
+
+	if (priv->data->inverted)
+		*temp = ((m * reg) - b) / div;
+	else
+		*temp = (b - (m * reg)) / div;
 	return 0;
 }
 
@@ -129,23 +203,69 @@ static struct thermal_zone_device_ops ops = {
 	.get_temp = armada_get_temp,
 };
 
-static const struct armada_thermal_ops armadaxp_ops = {
+static const struct armada_thermal_data armadaxp_data = {
 	.init_sensor = armadaxp_init_sensor,
+	.temp_shift = 10,
+	.temp_mask = 0x1ff,
+	.coef_b = 3153000000UL,
+	.coef_m = 10000000UL,
+	.coef_div = 13825,
 };
 
-static const struct armada_thermal_ops armada370_ops = {
+static const struct armada_thermal_data armada370_data = {
 	.is_valid = armada_is_valid,
 	.init_sensor = armada370_init_sensor,
+	.is_valid_shift = 9,
+	.temp_shift = 10,
+	.temp_mask = 0x1ff,
+	.coef_b = 3153000000UL,
+	.coef_m = 10000000UL,
+	.coef_div = 13825,
+};
+
+static const struct armada_thermal_data armada375_data = {
+	.is_valid = armada_is_valid,
+	.init_sensor = armada375_init_sensor,
+	.is_valid_shift = 10,
+	.temp_shift = 0,
+	.temp_mask = 0x1ff,
+	.coef_b = 3171900000UL,
+	.coef_m = 10000000UL,
+	.coef_div = 13616,
+};
+
+static const struct armada_thermal_data armada380_data = {
+	.is_valid = armada_is_valid,
+	.init_sensor = armada380_init_sensor,
+	.is_valid_shift = 10,
+	.temp_shift = 0,
+	.temp_mask = 0x3ff,
+	.coef_b = 1169498786UL,
+	.coef_m = 2000000UL,
+	.coef_div = 4289,
+	.inverted = true,
 };
 
 static const struct of_device_id armada_thermal_id_table[] = {
 	{
 		.compatible = "marvell,armadaxp-thermal",
-		.data       = &armadaxp_ops,
+		.data       = &armadaxp_data,
 	},
 	{
 		.compatible = "marvell,armada370-thermal",
-		.data       = &armada370_ops,
+		.data       = &armada370_data,
+	},
+	{
+		.compatible = "marvell,armada375-thermal",
+		.data       = &armada375_data,
+	},
+	{
+		.compatible = "marvell,armada375-z1-thermal",
+		.data       = &armada375_data,
+	},
+	{
+		.compatible = "marvell,armada380-thermal",
+		.data       = &armada380_data,
 	},
 	{
 		/* sentinel */
@@ -178,8 +298,8 @@ static int armada_thermal_probe(struct platform_device *pdev)
 	if (IS_ERR(priv->control))
 		return PTR_ERR(priv->control);
 
-	priv->ops = (struct armada_thermal_ops *)match->data;
-	priv->ops->init_sensor(priv);
+	priv->data = (struct armada_thermal_data *)match->data;
+	priv->data->init_sensor(pdev, priv);
 
 	thermal = thermal_zone_device_register("armada_thermal", 0, 0,
 					       priv, &ops, NULL, 0, 0);
diff --git a/drivers/thermal/imx_thermal.c b/drivers/thermal/imx_thermal.c
index a99c63152b8d..2c516f2eebed 100644
--- a/drivers/thermal/imx_thermal.c
+++ b/drivers/thermal/imx_thermal.c
@@ -306,7 +306,7 @@ static int imx_get_sensor_data(struct platform_device *pdev)
 {
 	struct imx_thermal_data *data = platform_get_drvdata(pdev);
 	struct regmap *map;
-	int t1, t2, n1, n2;
+	int t1, n1;
 	int ret;
 	u32 val;
 	u64 temp64;
@@ -333,14 +333,10 @@ static int imx_get_sensor_data(struct platform_device *pdev)
 	/*
 	 * Sensor data layout:
 	 *   [31:20] - sensor value @ 25C
-	 *    [19:8] - sensor value of hot
-	 *     [7:0] - hot temperature value
 	 * Use universal formula now and only need sensor value @ 25C
 	 * slope = 0.4297157 - (0.0015976 * 25C fuse)
 	 */
 	n1 = val >> 20;
-	n2 = (val & 0xfff00) >> 8;
-	t2 = val & 0xff;
 	t1 = 25; /* t1 always 25C */
 
 	/*
@@ -366,16 +362,16 @@ static int imx_get_sensor_data(struct platform_device *pdev)
 	data->c2 = n1 * data->c1 + 1000 * t1;
 
 	/*
-	 * Set the default passive cooling trip point to 20 °C below the
-	 * maximum die temperature. Can be changed from userspace.
+	 * Set the default passive cooling trip point,
+	 * can be changed from userspace.
 	 */
-	data->temp_passive = 1000 * (t2 - 20);
+	data->temp_passive = IMX_TEMP_PASSIVE;
 
 	/*
-	 * The maximum die temperature is t2, let's give 5 °C cushion
-	 * for noise and possible temperature rise between measurements.
+	 * The maximum die temperature set to 20 C higher than
+	 * IMX_TEMP_PASSIVE.
 	 */
-	data->temp_critical = 1000 * (t2 - 5);
+	data->temp_critical = 1000 * 20 + data->temp_passive;
 
 	return 0;
 }
diff --git a/drivers/thermal/int3403_thermal.c b/drivers/thermal/int3403_thermal.c
index 1301681d9a77..e93f0253f6ed 100644
--- a/drivers/thermal/int3403_thermal.c
+++ b/drivers/thermal/int3403_thermal.c
@@ -62,7 +62,13 @@ static int sys_get_trip_hyst(struct thermal_zone_device *tzone,
 	if (ACPI_FAILURE(status))
 		return -EIO;
 
-	*temp = DECI_KELVIN_TO_MILLI_CELSIUS(hyst, KELVIN_OFFSET);
+	/*
+	 * Thermal hysteresis represents a temperature difference.
+	 * Kelvin and Celsius have same degree size. So the
+	 * conversion here between tenths of degree Kelvin unit
+	 * and Milli-Celsius unit is just to multiply 100.
+	 */
+	*temp = hyst * 100;
 
 	return 0;
 }
diff --git a/drivers/thermal/intel_powerclamp.c b/drivers/thermal/intel_powerclamp.c
index a084325f1386..95cb7fc20e17 100644
--- a/drivers/thermal/intel_powerclamp.c
+++ b/drivers/thermal/intel_powerclamp.c
@@ -681,8 +681,10 @@ static const struct x86_cpu_id intel_powerclamp_ids[] = {
 	{ X86_VENDOR_INTEL, 6, 0x2d},
 	{ X86_VENDOR_INTEL, 6, 0x2e},
 	{ X86_VENDOR_INTEL, 6, 0x2f},
+	{ X86_VENDOR_INTEL, 6, 0x37},
 	{ X86_VENDOR_INTEL, 6, 0x3a},
 	{ X86_VENDOR_INTEL, 6, 0x3c},
+	{ X86_VENDOR_INTEL, 6, 0x3d},
 	{ X86_VENDOR_INTEL, 6, 0x3e},
 	{ X86_VENDOR_INTEL, 6, 0x3f},
 	{ X86_VENDOR_INTEL, 6, 0x45},
diff --git a/drivers/thermal/intel_soc_dts_thermal.c b/drivers/thermal/intel_soc_dts_thermal.c
new file mode 100644
index 000000000000..a6a0a18ec0aa
--- /dev/null
+++ b/drivers/thermal/intel_soc_dts_thermal.c
@@ -0,0 +1,479 @@
+/*
+ * intel_soc_dts_thermal.c
+ * Copyright (c) 2014, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/thermal.h>
+#include <asm/cpu_device_id.h>
+#include <asm/iosf_mbi.h>
+
+#define SOC_DTS_OFFSET_ENABLE	0xB0
+#define SOC_DTS_OFFSET_TEMP	0xB1
+
+#define SOC_DTS_OFFSET_PTPS	0xB2
+#define SOC_DTS_OFFSET_PTTS	0xB3
+#define SOC_DTS_OFFSET_PTTSS	0xB4
+#define SOC_DTS_OFFSET_PTMC	0x80
+#define SOC_DTS_TE_AUX0		0xB5
+#define SOC_DTS_TE_AUX1		0xB6
+
+#define SOC_DTS_AUX0_ENABLE_BIT		BIT(0)
+#define SOC_DTS_AUX1_ENABLE_BIT		BIT(1)
+#define SOC_DTS_CPU_MODULE0_ENABLE_BIT	BIT(16)
+#define SOC_DTS_CPU_MODULE1_ENABLE_BIT	BIT(17)
+#define SOC_DTS_TE_SCI_ENABLE		BIT(9)
+#define SOC_DTS_TE_SMI_ENABLE		BIT(10)
+#define SOC_DTS_TE_MSI_ENABLE		BIT(11)
+#define SOC_DTS_TE_APICA_ENABLE		BIT(14)
+#define SOC_DTS_PTMC_APIC_DEASSERT_BIT	BIT(4)
+
+/* DTS encoding for TJ MAX temperature */
+#define SOC_DTS_TJMAX_ENCODING	0x7F
+
+/* IRQ 86 is a fixed APIC interrupt for BYT DTS Aux threshold notifications */
+#define BYT_SOC_DTS_APIC_IRQ	86
+
+/* Only 2 out of 4 is allowed for OSPM */
+#define SOC_MAX_DTS_TRIPS	2
+
+/* Mask for two trips in status bits */
+#define SOC_DTS_TRIP_MASK	0x03
+
+/* DTS0 and DTS 1 */
+#define SOC_MAX_DTS_SENSORS	2
+
+#define CRITICAL_OFFSET_FROM_TJ_MAX	5000
+
+struct soc_sensor_entry {
+	int id;
+	u32 tj_max;
+	u32 temp_mask;
+	u32 temp_shift;
+	u32 store_status;
+	struct thermal_zone_device *tzone;
+};
+
+static struct soc_sensor_entry *soc_dts[SOC_MAX_DTS_SENSORS];
+
+static int crit_offset = CRITICAL_OFFSET_FROM_TJ_MAX;
+module_param(crit_offset, int, 0644);
+MODULE_PARM_DESC(crit_offset,
+	"Critical Temperature offset from tj max in millidegree Celsius.");
+
+static DEFINE_MUTEX(aux_update_mutex);
+static spinlock_t intr_notify_lock;
+static int soc_dts_thres_irq;
+
+static int get_tj_max(u32 *tj_max)
+{
+	u32 eax, edx;
+	u32 val;
+	int err;
+
+	err = rdmsr_safe(MSR_IA32_TEMPERATURE_TARGET, &eax, &edx);
+	if (err)
+		goto err_ret;
+	else {
+		val = (eax >> 16) & 0xff;
+		if (val)
+			*tj_max = val * 1000;
+		else {
+			err = -EINVAL;
+			goto err_ret;
+		}
+	}
+
+	return 0;
+err_ret:
+	*tj_max = 0;
+
+	return err;
+}
+
+static int sys_get_trip_temp(struct thermal_zone_device *tzd,
+					int trip, unsigned long *temp)
+{
+	int status;
+	u32 out;
+	struct soc_sensor_entry *aux_entry;
+
+	aux_entry = tzd->devdata;
+
+	if (!trip) {
+		/* Just return the critical temp */
+		*temp = aux_entry->tj_max - crit_offset;
+		return 0;
+	}
+
+	mutex_lock(&aux_update_mutex);
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_OFFSET_PTPS, &out);
+	mutex_unlock(&aux_update_mutex);
+	if (status)
+		return status;
+
+	out = (out >> (trip * 8)) & SOC_DTS_TJMAX_ENCODING;
+
+	if (!out)
+		*temp = 0;
+	else
+		*temp = aux_entry->tj_max - out * 1000;
+
+	return 0;
+}
+
+static int update_trip_temp(struct soc_sensor_entry *aux_entry,
+				int thres_index, unsigned long temp)
+{
+	int status;
+	u32 temp_out;
+	u32 out;
+	u32 store_ptps;
+	u32 store_ptmc;
+	u32 store_te_out;
+	u32 te_out;
+
+	u32 int_enable_bit = SOC_DTS_TE_APICA_ENABLE |
+						SOC_DTS_TE_MSI_ENABLE;
+
+	temp_out = (aux_entry->tj_max - temp) / 1000;
+
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+				SOC_DTS_OFFSET_PTPS, &store_ptps);
+	if (status)
+		return status;
+
+	out = (store_ptps & ~(0xFF << (thres_index * 8)));
+	out |= (temp_out & 0xFF) << (thres_index * 8);
+	status = iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+				SOC_DTS_OFFSET_PTPS, out);
+	if (status)
+		return status;
+	pr_debug("update_trip_temp PTPS = %x\n", out);
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_OFFSET_PTMC, &out);
+	if (status)
+		goto err_restore_ptps;
+
+	store_ptmc = out;
+
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_TE_AUX0 + thres_index,
+					&te_out);
+	if (status)
+		goto err_restore_ptmc;
+
+	store_te_out = te_out;
+
+	/* Enable for CPU module 0 and module 1 */
+	out |= (SOC_DTS_CPU_MODULE0_ENABLE_BIT |
+					SOC_DTS_CPU_MODULE1_ENABLE_BIT);
+	if (temp) {
+		if (thres_index)
+			out |= SOC_DTS_AUX1_ENABLE_BIT;
+		else
+			out |= SOC_DTS_AUX0_ENABLE_BIT;
+		te_out |= int_enable_bit;
+	} else {
+		if (thres_index)
+			out &= ~SOC_DTS_AUX1_ENABLE_BIT;
+		else
+			out &= ~SOC_DTS_AUX0_ENABLE_BIT;
+		te_out &= ~int_enable_bit;
+	}
+	status = iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+					SOC_DTS_OFFSET_PTMC, out);
+	if (status)
+		goto err_restore_te_out;
+
+	status = iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+					SOC_DTS_TE_AUX0 + thres_index,
+					te_out);
+	if (status)
+		goto err_restore_te_out;
+
+	return 0;
+
+err_restore_te_out:
+	iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+				SOC_DTS_OFFSET_PTMC, store_te_out);
+err_restore_ptmc:
+	iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+				SOC_DTS_OFFSET_PTMC, store_ptmc);
+err_restore_ptps:
+	iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+				SOC_DTS_OFFSET_PTPS, store_ptps);
+	/* Nothing we can do if restore fails */
+
+	return status;
+}
+
+static int sys_set_trip_temp(struct thermal_zone_device *tzd, int trip,
+							unsigned long temp)
+{
+	struct soc_sensor_entry *aux_entry = tzd->devdata;
+	int status;
+
+	if (temp > (aux_entry->tj_max - crit_offset))
+		return -EINVAL;
+
+	mutex_lock(&aux_update_mutex);
+	status = update_trip_temp(tzd->devdata, trip, temp);
+	mutex_unlock(&aux_update_mutex);
+
+	return status;
+}
+
+static int sys_get_trip_type(struct thermal_zone_device *thermal,
+		int trip, enum thermal_trip_type *type)
+{
+	if (trip)
+		*type = THERMAL_TRIP_PASSIVE;
+	else
+		*type = THERMAL_TRIP_CRITICAL;
+
+	return 0;
+}
+
+static int sys_get_curr_temp(struct thermal_zone_device *tzd,
+						unsigned long *temp)
+{
+	int status;
+	u32 out;
+	struct soc_sensor_entry *aux_entry;
+
+	aux_entry = tzd->devdata;
+
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_OFFSET_TEMP, &out);
+	if (status)
+		return status;
+
+	out = (out & aux_entry->temp_mask) >> aux_entry->temp_shift;
+	out -= SOC_DTS_TJMAX_ENCODING;
+	*temp = aux_entry->tj_max - out * 1000;
+
+	return 0;
+}
+
+static struct thermal_zone_device_ops tzone_ops = {
+	.get_temp = sys_get_curr_temp,
+	.get_trip_temp = sys_get_trip_temp,
+	.get_trip_type = sys_get_trip_type,
+	.set_trip_temp = sys_set_trip_temp,
+};
+
+static void free_soc_dts(struct soc_sensor_entry *aux_entry)
+{
+	if (aux_entry) {
+		iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+			SOC_DTS_OFFSET_ENABLE, aux_entry->store_status);
+		thermal_zone_device_unregister(aux_entry->tzone);
+		kfree(aux_entry);
+	}
+}
+
+static int soc_dts_enable(int id)
+{
+	u32 out;
+	int ret;
+
+	ret = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_OFFSET_ENABLE, &out);
+	if (ret)
+		return ret;
+
+	if (!(out & BIT(id))) {
+		out |= BIT(id);
+		ret = iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+					SOC_DTS_OFFSET_ENABLE, out);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static struct soc_sensor_entry *alloc_soc_dts(int id, u32 tj_max)
+{
+	struct soc_sensor_entry *aux_entry;
+	char name[10];
+	int err;
+
+	aux_entry = kzalloc(sizeof(*aux_entry), GFP_KERNEL);
+	if (!aux_entry) {
+		err = -ENOMEM;
+		return ERR_PTR(-ENOMEM);
+	}
+
+	/* Store status to restor on exit */
+	err = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_OFFSET_ENABLE,
+					&aux_entry->store_status);
+	if (err)
+		goto err_ret;
+
+	aux_entry->id = id;
+	aux_entry->tj_max = tj_max;
+	aux_entry->temp_mask = 0x00FF << (id * 8);
+	aux_entry->temp_shift = id * 8;
+	snprintf(name, sizeof(name), "soc_dts%d", id);
+	aux_entry->tzone = thermal_zone_device_register(name,
+			SOC_MAX_DTS_TRIPS,
+			0x02,
+			aux_entry, &tzone_ops, NULL, 0, 0);
+	if (IS_ERR(aux_entry->tzone)) {
+		err = PTR_ERR(aux_entry->tzone);
+		goto err_ret;
+	}
+
+	err = soc_dts_enable(id);
+	if (err)
+		goto err_aux_status;
+
+	return aux_entry;
+
+err_aux_status:
+	thermal_zone_device_unregister(aux_entry->tzone);
+err_ret:
+	kfree(aux_entry);
+	return ERR_PTR(err);
+}
+
+static void proc_thermal_interrupt(void)
+{
+	u32 sticky_out;
+	int status;
+	u32 ptmc_out;
+
+	/* Clear APIC interrupt */
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+				SOC_DTS_OFFSET_PTMC, &ptmc_out);
+
+	ptmc_out |= SOC_DTS_PTMC_APIC_DEASSERT_BIT;
+	status = iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+					SOC_DTS_OFFSET_PTMC, ptmc_out);
+
+	/* Read status here */
+	status = iosf_mbi_read(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_READ,
+					SOC_DTS_OFFSET_PTTSS, &sticky_out);
+	pr_debug("status %d PTTSS %x\n", status, sticky_out);
+	if (sticky_out & SOC_DTS_TRIP_MASK) {
+		int i;
+		/* reset sticky bit */
+		status = iosf_mbi_write(BT_MBI_UNIT_PMC, BT_MBI_BUNIT_WRITE,
+					SOC_DTS_OFFSET_PTTSS, sticky_out);
+		for (i = 0; i < SOC_MAX_DTS_SENSORS; ++i) {
+			pr_debug("TZD update for zone %d\n", i);
+			thermal_zone_device_update(soc_dts[i]->tzone);
+		}
+	}
+
+}
+
+static irqreturn_t soc_irq_thread_fn(int irq, void *dev_data)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&intr_notify_lock, flags);
+	proc_thermal_interrupt();
+	spin_unlock_irqrestore(&intr_notify_lock, flags);
+	pr_debug("proc_thermal_interrupt\n");
+
+	return IRQ_HANDLED;
+}
+
+static const struct x86_cpu_id soc_thermal_ids[] = {
+	{ X86_VENDOR_INTEL, X86_FAMILY_ANY, 0x37, 0, BYT_SOC_DTS_APIC_IRQ},
+	{}
+};
+MODULE_DEVICE_TABLE(x86cpu, soc_thermal_ids);
+
+static int __init intel_soc_thermal_init(void)
+{
+	u32 tj_max;
+	int err = 0;
+	int i;
+	const struct x86_cpu_id *match_cpu;
+
+	match_cpu = x86_match_cpu(soc_thermal_ids);
+	if (!match_cpu)
+		return -ENODEV;
+
+	if (get_tj_max(&tj_max))
+		return -EINVAL;
+
+	for (i = 0; i < SOC_MAX_DTS_SENSORS; ++i) {
+		soc_dts[i] = alloc_soc_dts(i, tj_max);
+		if (IS_ERR(soc_dts[i])) {
+			err = PTR_ERR(soc_dts[i]);
+			goto err_free;
+		}
+	}
+
+	spin_lock_init(&intr_notify_lock);
+
+	soc_dts_thres_irq = (int)match_cpu->driver_data;
+
+	err = request_threaded_irq(soc_dts_thres_irq, NULL,
+					soc_irq_thread_fn,
+					IRQF_TRIGGER_RISING | IRQF_ONESHOT,
+					"soc_dts", soc_dts);
+	if (err) {
+		pr_err("request_threaded_irq ret %d\n", err);
+		goto err_free;
+	}
+
+	for (i = 0; i < SOC_MAX_DTS_SENSORS; ++i) {
+		err = update_trip_temp(soc_dts[i], 0, tj_max - crit_offset);
+		if (err)
+			goto err_trip_temp;
+	}
+
+	return 0;
+
+err_trip_temp:
+	i = SOC_MAX_DTS_SENSORS;
+	free_irq(soc_dts_thres_irq, soc_dts);
+err_free:
+	while (--i >= 0)
+		free_soc_dts(soc_dts[i]);
+
+	return err;
+}
+
+static void __exit intel_soc_thermal_exit(void)
+{
+	int i;
+
+	for (i = 0; i < SOC_MAX_DTS_SENSORS; ++i)
+		update_trip_temp(soc_dts[i], 0, 0);
+
+	free_irq(soc_dts_thres_irq, soc_dts);
+
+	for (i = 0; i < SOC_MAX_DTS_SENSORS; ++i)
+		free_soc_dts(soc_dts[i]);
+
+}
+
+module_init(intel_soc_thermal_init)
+module_exit(intel_soc_thermal_exit)
+
+MODULE_DESCRIPTION("Intel SoC DTS Thermal Driver");
+MODULE_AUTHOR("Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/thermal/of-thermal.c b/drivers/thermal/of-thermal.c
index 04b1be7fa018..4b2b999b7611 100644
--- a/drivers/thermal/of-thermal.c
+++ b/drivers/thermal/of-thermal.c
@@ -156,8 +156,8 @@ static int of_thermal_bind(struct thermal_zone_device *thermal,
 
 			ret = thermal_zone_bind_cooling_device(thermal,
 						tbp->trip_id, cdev,
-						tbp->min,
-						tbp->max);
+						tbp->max,
+						tbp->min);
 			if (ret)
 				return ret;
 		}
@@ -712,11 +712,12 @@ thermal_of_build_thermal_zone(struct device_node *np)
 	}
 
 	i = 0;
-	for_each_child_of_node(child, gchild)
+	for_each_child_of_node(child, gchild) {
 		ret = thermal_of_populate_bind_params(gchild, &tz->tbps[i++],
 						      tz->trips, tz->ntrips);
 		if (ret)
 			goto free_tbps;
+	}
 
 finish:
 	of_node_put(child);
diff --git a/drivers/thermal/rcar_thermal.c b/drivers/thermal/rcar_thermal.c
index 5a37940b02c9..8803e693fe68 100644
--- a/drivers/thermal/rcar_thermal.c
+++ b/drivers/thermal/rcar_thermal.c
@@ -374,10 +374,8 @@ static int rcar_thermal_probe(struct platform_device *pdev)
 	int idle = IDLE_INTERVAL;
 
 	common = devm_kzalloc(dev, sizeof(*common), GFP_KERNEL);
-	if (!common) {
-		dev_err(dev, "Could not allocate common\n");
+	if (!common)
 		return -ENOMEM;
-	}
 
 	INIT_LIST_HEAD(&common->head);
 	spin_lock_init(&common->lock);
@@ -423,7 +421,6 @@ static int rcar_thermal_probe(struct platform_device *pdev)
 
 		priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
 		if (!priv) {
-			dev_err(dev, "Could not allocate priv\n");
 			ret = -ENOMEM;
 			goto error_unregister;
 		}
@@ -470,7 +467,7 @@ error_unregister:
 			rcar_thermal_irq_disable(priv);
 	}
 
-	pm_runtime_put_sync(dev);
+	pm_runtime_put(dev);
 	pm_runtime_disable(dev);
 
 	return ret;
@@ -488,7 +485,7 @@ static int rcar_thermal_remove(struct platform_device *pdev)
 			rcar_thermal_irq_disable(priv);
 	}
 
-	pm_runtime_put_sync(dev);
+	pm_runtime_put(dev);
 	pm_runtime_disable(dev);
 
 	return 0;
diff --git a/drivers/thermal/samsung/exynos_tmu.c b/drivers/thermal/samsung/exynos_tmu.c
index 0d96a510389f..d7ca9f49c9cb 100644
--- a/drivers/thermal/samsung/exynos_tmu.c
+++ b/drivers/thermal/samsung/exynos_tmu.c
@@ -41,12 +41,13 @@
  * @id: identifier of the one instance of the TMU controller.
  * @pdata: pointer to the tmu platform/configuration data
  * @base: base address of the single instance of the TMU controller.
- * @base_common: base address of the common registers of the TMU controller.
+ * @base_second: base address of the common registers of the TMU controller.
  * @irq: irq number of the TMU controller.
  * @soc: id of the SOC type.
  * @irq_work: pointer to the irq work structure.
  * @lock: lock to implement synchronization.
  * @clk: pointer to the clock structure.
+ * @clk_sec: pointer to the clock structure for accessing the base_second.
  * @temp_error1: fused value of the first point trim.
  * @temp_error2: fused value of the second point trim.
  * @regulator: pointer to the TMU regulator structure.
@@ -56,12 +57,12 @@ struct exynos_tmu_data {
 	int id;
 	struct exynos_tmu_platform_data *pdata;
 	void __iomem *base;
-	void __iomem *base_common;
+	void __iomem *base_second;
 	int irq;
 	enum soc_type soc;
 	struct work_struct irq_work;
 	struct mutex lock;
-	struct clk *clk;
+	struct clk *clk, *clk_sec;
 	u8 temp_error1, temp_error2;
 	struct regulator *regulator;
 	struct thermal_sensor_conf *reg_conf;
@@ -152,6 +153,8 @@ static int exynos_tmu_initialize(struct platform_device *pdev)
 
 	mutex_lock(&data->lock);
 	clk_enable(data->clk);
+	if (!IS_ERR(data->clk_sec))
+		clk_enable(data->clk_sec);
 
 	if (TMU_SUPPORTS(pdata, READY_STATUS)) {
 		status = readb(data->base + reg->tmu_status);
@@ -186,7 +189,12 @@ static int exynos_tmu_initialize(struct platform_device *pdev)
 			EXYNOS5440_EFUSE_SWAP_OFFSET + reg->triminfo_data);
 		}
 	} else {
-		trim_info = readl(data->base + reg->triminfo_data);
+		/* On exynos5420 the triminfo register is in the shared space */
+		if (data->soc == SOC_ARCH_EXYNOS5420_TRIMINFO)
+			trim_info = readl(data->base_second +
+							reg->triminfo_data);
+		else
+			trim_info = readl(data->base + reg->triminfo_data);
 	}
 	data->temp_error1 = trim_info & EXYNOS_TMU_TEMP_MASK;
 	data->temp_error2 = ((trim_info >> reg->triminfo_85_shift) &
@@ -225,6 +233,8 @@ skip_calib_data:
 			trigger_levs++;
 	}
 
+	rising_threshold = readl(data->base + reg->threshold_th0);
+
 	if (data->soc == SOC_ARCH_EXYNOS4210) {
 		/* Write temperature code for threshold */
 		threshold_code = temp_to_code(data, pdata->threshold);
@@ -238,7 +248,7 @@ skip_calib_data:
 			writeb(pdata->trigger_levels[i], data->base +
 			reg->threshold_th0 + i * sizeof(reg->threshold_th0));
 
-		writel(reg->inten_rise_mask, data->base + reg->tmu_intclear);
+		writel(reg->intclr_rise_mask, data->base + reg->tmu_intclear);
 	} else {
 		/* Write temperature code for rising and falling threshold */
 		for (i = 0;
@@ -249,6 +259,7 @@ skip_calib_data:
 				ret = threshold_code;
 				goto out;
 			}
+			rising_threshold &= ~(0xff << 8 * i);
 			rising_threshold |= threshold_code << 8 * i;
 			if (pdata->threshold_falling) {
 				threshold_code = temp_to_code(data,
@@ -265,8 +276,8 @@ skip_calib_data:
 		writel(falling_threshold,
 				data->base + reg->threshold_th1);
 
-		writel((reg->inten_rise_mask << reg->inten_rise_shift) |
-			(reg->inten_fall_mask << reg->inten_fall_shift),
+		writel((reg->intclr_rise_mask << reg->intclr_rise_shift) |
+			(reg->intclr_fall_mask << reg->intclr_fall_shift),
 				data->base + reg->tmu_intclear);
 
 		/* if last threshold limit is also present */
@@ -281,6 +292,7 @@ skip_calib_data:
 			}
 			if (i == EXYNOS_MAX_TRIGGER_PER_REG - 1) {
 				/* 1-4 level to be assigned in th0 reg */
+				rising_threshold &= ~(0xff << 8 * i);
 				rising_threshold |= threshold_code << 8 * i;
 				writel(rising_threshold,
 					data->base + reg->threshold_th0);
@@ -298,10 +310,12 @@ skip_calib_data:
 	}
 	/*Clear the PMIN in the common TMU register*/
 	if (reg->tmu_pmin && !data->id)
-		writel(0, data->base_common + reg->tmu_pmin);
+		writel(0, data->base_second + reg->tmu_pmin);
 out:
 	clk_disable(data->clk);
 	mutex_unlock(&data->lock);
+	if (!IS_ERR(data->clk_sec))
+		clk_disable(data->clk_sec);
 
 	return ret;
 }
@@ -453,12 +467,16 @@ static void exynos_tmu_work(struct work_struct *work)
 	const struct exynos_tmu_registers *reg = pdata->registers;
 	unsigned int val_irq, val_type;
 
+	if (!IS_ERR(data->clk_sec))
+		clk_enable(data->clk_sec);
 	/* Find which sensor generated this interrupt */
 	if (reg->tmu_irqstatus) {
-		val_type = readl(data->base_common + reg->tmu_irqstatus);
+		val_type = readl(data->base_second + reg->tmu_irqstatus);
 		if (!((val_type >> data->id) & 0x1))
 			goto out;
 	}
+	if (!IS_ERR(data->clk_sec))
+		clk_disable(data->clk_sec);
 
 	exynos_report_trigger(data->reg_conf);
 	mutex_lock(&data->lock);
@@ -499,6 +517,18 @@ static const struct of_device_id exynos_tmu_match[] = {
 		.data = (void *)EXYNOS5250_TMU_DRV_DATA,
 	},
 	{
+		.compatible = "samsung,exynos5260-tmu",
+		.data = (void *)EXYNOS5260_TMU_DRV_DATA,
+	},
+	{
+		.compatible = "samsung,exynos5420-tmu",
+		.data = (void *)EXYNOS5420_TMU_DRV_DATA,
+	},
+	{
+		.compatible = "samsung,exynos5420-tmu-ext-triminfo",
+		.data = (void *)EXYNOS5420_TMU_DRV_DATA,
+	},
+	{
 		.compatible = "samsung,exynos5440-tmu",
 		.data = (void *)EXYNOS5440_TMU_DRV_DATA,
 	},
@@ -580,7 +610,7 @@ static int exynos_map_dt_data(struct platform_device *pdev)
 	 * Check if the TMU shares some registers and then try to map the
 	 * memory of common registers.
 	 */
-	if (!TMU_SUPPORTS(pdata, SHARED_MEMORY))
+	if (!TMU_SUPPORTS(pdata, ADDRESS_MULTIPLE))
 		return 0;
 
 	if (of_address_to_resource(pdev->dev.of_node, 1, &res)) {
@@ -588,9 +618,9 @@ static int exynos_map_dt_data(struct platform_device *pdev)
 		return -ENODEV;
 	}
 
-	data->base_common = devm_ioremap(&pdev->dev, res.start,
+	data->base_second = devm_ioremap(&pdev->dev, res.start,
 					resource_size(&res));
-	if (!data->base_common) {
+	if (!data->base_second) {
 		dev_err(&pdev->dev, "Failed to ioremap memory\n");
 		return -ENOMEM;
 	}
@@ -607,10 +637,8 @@ static int exynos_tmu_probe(struct platform_device *pdev)
 
 	data = devm_kzalloc(&pdev->dev, sizeof(struct exynos_tmu_data),
 					GFP_KERNEL);
-	if (!data) {
-		dev_err(&pdev->dev, "Failed to allocate driver structure\n");
+	if (!data)
 		return -ENOMEM;
-	}
 
 	platform_set_drvdata(pdev, data);
 	mutex_init(&data->lock);
@@ -629,13 +657,31 @@ static int exynos_tmu_probe(struct platform_device *pdev)
 		return  PTR_ERR(data->clk);
 	}
 
+	data->clk_sec = devm_clk_get(&pdev->dev, "tmu_triminfo_apbif");
+	if (IS_ERR(data->clk_sec)) {
+		if (data->soc == SOC_ARCH_EXYNOS5420_TRIMINFO) {
+			dev_err(&pdev->dev, "Failed to get triminfo clock\n");
+			return PTR_ERR(data->clk_sec);
+		}
+	} else {
+		ret = clk_prepare(data->clk_sec);
+		if (ret) {
+			dev_err(&pdev->dev, "Failed to get clock\n");
+			return ret;
+		}
+	}
+
 	ret = clk_prepare(data->clk);
-	if (ret)
-		return ret;
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to get clock\n");
+		goto err_clk_sec;
+	}
 
 	if (pdata->type == SOC_ARCH_EXYNOS4210 ||
 	    pdata->type == SOC_ARCH_EXYNOS4412 ||
 	    pdata->type == SOC_ARCH_EXYNOS5250 ||
+	    pdata->type == SOC_ARCH_EXYNOS5260 ||
+	    pdata->type == SOC_ARCH_EXYNOS5420_TRIMINFO ||
 	    pdata->type == SOC_ARCH_EXYNOS5440)
 		data->soc = pdata->type;
 	else {
@@ -656,7 +702,6 @@ static int exynos_tmu_probe(struct platform_device *pdev)
 	sensor_conf = devm_kzalloc(&pdev->dev,
 				sizeof(struct thermal_sensor_conf), GFP_KERNEL);
 	if (!sensor_conf) {
-		dev_err(&pdev->dev, "Failed to allocate registration struct\n");
 		ret = -ENOMEM;
 		goto err_clk;
 	}
@@ -704,6 +749,9 @@ static int exynos_tmu_probe(struct platform_device *pdev)
 	return 0;
 err_clk:
 	clk_unprepare(data->clk);
+err_clk_sec:
+	if (!IS_ERR(data->clk_sec))
+		clk_unprepare(data->clk_sec);
 	return ret;
 }
 
@@ -716,6 +764,8 @@ static int exynos_tmu_remove(struct platform_device *pdev)
 	exynos_unregister_thermal(data->reg_conf);
 
 	clk_unprepare(data->clk);
+	if (!IS_ERR(data->clk_sec))
+		clk_unprepare(data->clk_sec);
 
 	if (!IS_ERR(data->regulator))
 		regulator_disable(data->regulator);
diff --git a/drivers/thermal/samsung/exynos_tmu.h b/drivers/thermal/samsung/exynos_tmu.h
index 3fb65547e64c..edd08cf76729 100644
--- a/drivers/thermal/samsung/exynos_tmu.h
+++ b/drivers/thermal/samsung/exynos_tmu.h
@@ -43,6 +43,8 @@ enum soc_type {
 	SOC_ARCH_EXYNOS4210 = 1,
 	SOC_ARCH_EXYNOS4412,
 	SOC_ARCH_EXYNOS5250,
+	SOC_ARCH_EXYNOS5260,
+	SOC_ARCH_EXYNOS5420_TRIMINFO,
 	SOC_ARCH_EXYNOS5440,
 };
 
@@ -60,7 +62,7 @@ enum soc_type {
  *			state(active/idle) can be checked.
  * TMU_SUPPORT_EMUL_TIME - This features allows to set next temp emulation
  *			sample time.
- * TMU_SUPPORT_SHARED_MEMORY - This feature tells that the different TMU
+ * TMU_SUPPORT_ADDRESS_MULTIPLE - This feature tells that the different TMU
  *			sensors shares some common registers.
  * TMU_SUPPORT - macro to compare the above features with the supplied.
  */
@@ -70,7 +72,7 @@ enum soc_type {
 #define TMU_SUPPORT_FALLING_TRIP		BIT(3)
 #define TMU_SUPPORT_READY_STATUS		BIT(4)
 #define TMU_SUPPORT_EMUL_TIME			BIT(5)
-#define TMU_SUPPORT_SHARED_MEMORY		BIT(6)
+#define TMU_SUPPORT_ADDRESS_MULTIPLE		BIT(6)
 
 #define TMU_SUPPORTS(a, b)	(a->features & TMU_SUPPORT_ ## b)
 
@@ -122,10 +124,6 @@ enum soc_type {
  * @threshold_th3_l0_shift: shift bits of level0 threshold temperature.
  * @tmu_inten: register containing the different threshold interrupt
 	enable bits.
- * @inten_rise_shift: shift bits of all rising interrupt bits.
- * @inten_rise_mask: mask bits of all rising interrupt bits.
- * @inten_fall_shift: shift bits of all rising interrupt bits.
- * @inten_fall_mask: mask bits of all rising interrupt bits.
  * @inten_rise0_shift: shift bits of rising 0 interrupt bits.
  * @inten_rise1_shift: shift bits of rising 1 interrupt bits.
  * @inten_rise2_shift: shift bits of rising 2 interrupt bits.
@@ -136,6 +134,10 @@ enum soc_type {
  * @inten_fall3_shift: shift bits of falling 3 interrupt bits.
  * @tmu_intstat: Register containing the interrupt status values.
  * @tmu_intclear: Register for clearing the raised interrupt status.
+ * @intclr_fall_shift: shift bits for interrupt clear fall 0
+ * @intclr_rise_shift: shift bits of all rising interrupt bits.
+ * @intclr_rise_mask: mask bits of all rising interrupt bits.
+ * @intclr_fall_mask: mask bits of all rising interrupt bits.
  * @emul_con: TMU emulation controller register.
  * @emul_temp_shift: shift bits of emulation temperature.
  * @emul_time_shift: shift bits of emulation time.
@@ -149,6 +151,7 @@ struct exynos_tmu_registers {
 	u32	triminfo_85_shift;
 
 	u32	triminfo_ctrl;
+	u32	triminfo_ctrl1;
 	u32	triminfo_reload_shift;
 
 	u32	tmu_ctrl;
@@ -191,10 +194,6 @@ struct exynos_tmu_registers {
 	u32	threshold_th3_l0_shift;
 
 	u32	tmu_inten;
-	u32	inten_rise_shift;
-	u32	inten_rise_mask;
-	u32	inten_fall_shift;
-	u32	inten_fall_mask;
 	u32	inten_rise0_shift;
 	u32	inten_rise1_shift;
 	u32	inten_rise2_shift;
@@ -207,6 +206,10 @@ struct exynos_tmu_registers {
 	u32	tmu_intstat;
 
 	u32	tmu_intclear;
+	u32	intclr_fall_shift;
+	u32	intclr_rise_shift;
+	u32	intclr_fall_mask;
+	u32	intclr_rise_mask;
 
 	u32	emul_con;
 	u32	emul_temp_shift;
diff --git a/drivers/thermal/samsung/exynos_tmu_data.c b/drivers/thermal/samsung/exynos_tmu_data.c
index 476b768c633e..c1d81dcd7819 100644
--- a/drivers/thermal/samsung/exynos_tmu_data.c
+++ b/drivers/thermal/samsung/exynos_tmu_data.c
@@ -40,13 +40,13 @@ static const struct exynos_tmu_registers exynos4210_tmu_registers = {
 	.threshold_temp = EXYNOS4210_TMU_REG_THRESHOLD_TEMP,
 	.threshold_th0 = EXYNOS4210_TMU_REG_TRIG_LEVEL0,
 	.tmu_inten = EXYNOS_TMU_REG_INTEN,
-	.inten_rise_mask = EXYNOS4210_TMU_TRIG_LEVEL_MASK,
 	.inten_rise0_shift = EXYNOS_TMU_INTEN_RISE0_SHIFT,
 	.inten_rise1_shift = EXYNOS_TMU_INTEN_RISE1_SHIFT,
 	.inten_rise2_shift = EXYNOS_TMU_INTEN_RISE2_SHIFT,
 	.inten_rise3_shift = EXYNOS_TMU_INTEN_RISE3_SHIFT,
 	.tmu_intstat = EXYNOS_TMU_REG_INTSTAT,
 	.tmu_intclear = EXYNOS_TMU_REG_INTCLEAR,
+	.intclr_rise_mask = EXYNOS4210_TMU_TRIG_LEVEL_MASK,
 };
 
 struct exynos_tmu_init_data const exynos4210_default_tmu_data = {
@@ -112,10 +112,6 @@ static const struct exynos_tmu_registers exynos4412_tmu_registers = {
 	.threshold_th0 = EXYNOS_THD_TEMP_RISE,
 	.threshold_th1 = EXYNOS_THD_TEMP_FALL,
 	.tmu_inten = EXYNOS_TMU_REG_INTEN,
-	.inten_rise_mask = EXYNOS_TMU_RISE_INT_MASK,
-	.inten_rise_shift = EXYNOS_TMU_RISE_INT_SHIFT,
-	.inten_fall_mask = EXYNOS_TMU_FALL_INT_MASK,
-	.inten_fall_shift = EXYNOS_TMU_FALL_INT_SHIFT,
 	.inten_rise0_shift = EXYNOS_TMU_INTEN_RISE0_SHIFT,
 	.inten_rise1_shift = EXYNOS_TMU_INTEN_RISE1_SHIFT,
 	.inten_rise2_shift = EXYNOS_TMU_INTEN_RISE2_SHIFT,
@@ -123,6 +119,10 @@ static const struct exynos_tmu_registers exynos4412_tmu_registers = {
 	.inten_fall0_shift = EXYNOS_TMU_INTEN_FALL0_SHIFT,
 	.tmu_intstat = EXYNOS_TMU_REG_INTSTAT,
 	.tmu_intclear = EXYNOS_TMU_REG_INTCLEAR,
+	.intclr_fall_shift = EXYNOS_TMU_CLEAR_FALL_INT_SHIFT,
+	.intclr_rise_shift = EXYNOS_TMU_RISE_INT_SHIFT,
+	.intclr_rise_mask = EXYNOS_TMU_RISE_INT_MASK,
+	.intclr_fall_mask = EXYNOS_TMU_FALL_INT_MASK,
 	.emul_con = EXYNOS_EMUL_CON,
 	.emul_temp_shift = EXYNOS_EMUL_DATA_SHIFT,
 	.emul_time_shift = EXYNOS_EMUL_TIME_SHIFT,
@@ -194,6 +194,197 @@ struct exynos_tmu_init_data const exynos5250_default_tmu_data = {
 };
 #endif
 
+#if defined(CONFIG_SOC_EXYNOS5260)
+static const struct exynos_tmu_registers exynos5260_tmu_registers = {
+	.triminfo_data = EXYNOS_TMU_REG_TRIMINFO,
+	.triminfo_25_shift = EXYNOS_TRIMINFO_25_SHIFT,
+	.triminfo_85_shift = EXYNOS_TRIMINFO_85_SHIFT,
+	.tmu_ctrl = EXYNOS_TMU_REG_CONTROL,
+	.tmu_ctrl = EXYNOS_TMU_REG_CONTROL1,
+	.buf_vref_sel_shift = EXYNOS_TMU_REF_VOLTAGE_SHIFT,
+	.buf_vref_sel_mask = EXYNOS_TMU_REF_VOLTAGE_MASK,
+	.therm_trip_mode_shift = EXYNOS_TMU_TRIP_MODE_SHIFT,
+	.therm_trip_mode_mask = EXYNOS_TMU_TRIP_MODE_MASK,
+	.therm_trip_en_shift = EXYNOS_TMU_THERM_TRIP_EN_SHIFT,
+	.buf_slope_sel_shift = EXYNOS_TMU_BUF_SLOPE_SEL_SHIFT,
+	.buf_slope_sel_mask = EXYNOS_TMU_BUF_SLOPE_SEL_MASK,
+	.core_en_shift = EXYNOS_TMU_CORE_EN_SHIFT,
+	.tmu_status = EXYNOS_TMU_REG_STATUS,
+	.tmu_cur_temp = EXYNOS_TMU_REG_CURRENT_TEMP,
+	.threshold_th0 = EXYNOS_THD_TEMP_RISE,
+	.threshold_th1 = EXYNOS_THD_TEMP_FALL,
+	.tmu_inten = EXYNOS5260_TMU_REG_INTEN,
+	.inten_rise0_shift = EXYNOS_TMU_INTEN_RISE0_SHIFT,
+	.inten_rise1_shift = EXYNOS_TMU_INTEN_RISE1_SHIFT,
+	.inten_rise2_shift = EXYNOS_TMU_INTEN_RISE2_SHIFT,
+	.inten_rise3_shift = EXYNOS_TMU_INTEN_RISE3_SHIFT,
+	.inten_fall0_shift = EXYNOS_TMU_INTEN_FALL0_SHIFT,
+	.tmu_intstat = EXYNOS5260_TMU_REG_INTSTAT,
+	.tmu_intclear = EXYNOS5260_TMU_REG_INTCLEAR,
+	.intclr_fall_shift = EXYNOS5420_TMU_CLEAR_FALL_INT_SHIFT,
+	.intclr_rise_shift = EXYNOS_TMU_RISE_INT_SHIFT,
+	.intclr_rise_mask = EXYNOS5260_TMU_RISE_INT_MASK,
+	.intclr_fall_mask = EXYNOS5260_TMU_FALL_INT_MASK,
+	.emul_con = EXYNOS5260_EMUL_CON,
+	.emul_temp_shift = EXYNOS_EMUL_DATA_SHIFT,
+	.emul_time_shift = EXYNOS_EMUL_TIME_SHIFT,
+	.emul_time_mask = EXYNOS_EMUL_TIME_MASK,
+};
+
+#define __EXYNOS5260_TMU_DATA	\
+	.threshold_falling = 10, \
+	.trigger_levels[0] = 85, \
+	.trigger_levels[1] = 103, \
+	.trigger_levels[2] = 110, \
+	.trigger_levels[3] = 120, \
+	.trigger_enable[0] = true, \
+	.trigger_enable[1] = true, \
+	.trigger_enable[2] = true, \
+	.trigger_enable[3] = false, \
+	.trigger_type[0] = THROTTLE_ACTIVE, \
+	.trigger_type[1] = THROTTLE_ACTIVE, \
+	.trigger_type[2] = SW_TRIP, \
+	.trigger_type[3] = HW_TRIP, \
+	.max_trigger_level = 4, \
+	.gain = 8, \
+	.reference_voltage = 16, \
+	.noise_cancel_mode = 4, \
+	.cal_type = TYPE_ONE_POINT_TRIMMING, \
+	.efuse_value = 55, \
+	.min_efuse_value = 40, \
+	.max_efuse_value = 100, \
+	.first_point_trim = 25, \
+	.second_point_trim = 85, \
+	.default_temp_offset = 50, \
+	.freq_tab[0] = { \
+		.freq_clip_max = 800 * 1000, \
+		.temp_level = 85, \
+	}, \
+	.freq_tab[1] = { \
+		.freq_clip_max = 200 * 1000, \
+		.temp_level = 103, \
+	}, \
+	.freq_tab_count = 2, \
+	.registers = &exynos5260_tmu_registers, \
+
+#define EXYNOS5260_TMU_DATA \
+	__EXYNOS5260_TMU_DATA \
+	.type = SOC_ARCH_EXYNOS5260, \
+	.features = (TMU_SUPPORT_EMULATION | TMU_SUPPORT_TRIM_RELOAD | \
+			TMU_SUPPORT_FALLING_TRIP | TMU_SUPPORT_READY_STATUS | \
+			TMU_SUPPORT_EMUL_TIME)
+
+struct exynos_tmu_init_data const exynos5260_default_tmu_data = {
+	.tmu_data = {
+		{ EXYNOS5260_TMU_DATA },
+		{ EXYNOS5260_TMU_DATA },
+		{ EXYNOS5260_TMU_DATA },
+		{ EXYNOS5260_TMU_DATA },
+		{ EXYNOS5260_TMU_DATA },
+	},
+	.tmu_count = 5,
+};
+#endif
+
+#if defined(CONFIG_SOC_EXYNOS5420)
+static const struct exynos_tmu_registers exynos5420_tmu_registers = {
+	.triminfo_data = EXYNOS_TMU_REG_TRIMINFO,
+	.triminfo_25_shift = EXYNOS_TRIMINFO_25_SHIFT,
+	.triminfo_85_shift = EXYNOS_TRIMINFO_85_SHIFT,
+	.tmu_ctrl = EXYNOS_TMU_REG_CONTROL,
+	.buf_vref_sel_shift = EXYNOS_TMU_REF_VOLTAGE_SHIFT,
+	.buf_vref_sel_mask = EXYNOS_TMU_REF_VOLTAGE_MASK,
+	.therm_trip_mode_shift = EXYNOS_TMU_TRIP_MODE_SHIFT,
+	.therm_trip_mode_mask = EXYNOS_TMU_TRIP_MODE_MASK,
+	.therm_trip_en_shift = EXYNOS_TMU_THERM_TRIP_EN_SHIFT,
+	.buf_slope_sel_shift = EXYNOS_TMU_BUF_SLOPE_SEL_SHIFT,
+	.buf_slope_sel_mask = EXYNOS_TMU_BUF_SLOPE_SEL_MASK,
+	.core_en_shift = EXYNOS_TMU_CORE_EN_SHIFT,
+	.tmu_status = EXYNOS_TMU_REG_STATUS,
+	.tmu_cur_temp = EXYNOS_TMU_REG_CURRENT_TEMP,
+	.threshold_th0 = EXYNOS_THD_TEMP_RISE,
+	.threshold_th1 = EXYNOS_THD_TEMP_FALL,
+	.tmu_inten = EXYNOS_TMU_REG_INTEN,
+	.inten_rise0_shift = EXYNOS_TMU_INTEN_RISE0_SHIFT,
+	.inten_rise1_shift = EXYNOS_TMU_INTEN_RISE1_SHIFT,
+	.inten_rise2_shift = EXYNOS_TMU_INTEN_RISE2_SHIFT,
+	/* INTEN_RISE3 Not availble in exynos5420 */
+	.inten_rise3_shift = EXYNOS_TMU_INTEN_RISE3_SHIFT,
+	.inten_fall0_shift = EXYNOS_TMU_INTEN_FALL0_SHIFT,
+	.tmu_intstat = EXYNOS_TMU_REG_INTSTAT,
+	.tmu_intclear = EXYNOS_TMU_REG_INTCLEAR,
+	.intclr_fall_shift = EXYNOS5420_TMU_CLEAR_FALL_INT_SHIFT,
+	.intclr_rise_shift = EXYNOS_TMU_RISE_INT_SHIFT,
+	.intclr_rise_mask = EXYNOS_TMU_RISE_INT_MASK,
+	.intclr_fall_mask = EXYNOS_TMU_FALL_INT_MASK,
+	.emul_con = EXYNOS_EMUL_CON,
+	.emul_temp_shift = EXYNOS_EMUL_DATA_SHIFT,
+	.emul_time_shift = EXYNOS_EMUL_TIME_SHIFT,
+	.emul_time_mask = EXYNOS_EMUL_TIME_MASK,
+};
+
+#define __EXYNOS5420_TMU_DATA	\
+	.threshold_falling = 10, \
+	.trigger_levels[0] = 85, \
+	.trigger_levels[1] = 103, \
+	.trigger_levels[2] = 110, \
+	.trigger_levels[3] = 120, \
+	.trigger_enable[0] = true, \
+	.trigger_enable[1] = true, \
+	.trigger_enable[2] = true, \
+	.trigger_enable[3] = false, \
+	.trigger_type[0] = THROTTLE_ACTIVE, \
+	.trigger_type[1] = THROTTLE_ACTIVE, \
+	.trigger_type[2] = SW_TRIP, \
+	.trigger_type[3] = HW_TRIP, \
+	.max_trigger_level = 4, \
+	.gain = 8, \
+	.reference_voltage = 16, \
+	.noise_cancel_mode = 4, \
+	.cal_type = TYPE_ONE_POINT_TRIMMING, \
+	.efuse_value = 55, \
+	.min_efuse_value = 40, \
+	.max_efuse_value = 100, \
+	.first_point_trim = 25, \
+	.second_point_trim = 85, \
+	.default_temp_offset = 50, \
+	.freq_tab[0] = { \
+		.freq_clip_max = 800 * 1000, \
+		.temp_level = 85, \
+	}, \
+	.freq_tab[1] = { \
+		.freq_clip_max = 200 * 1000, \
+		.temp_level = 103, \
+	}, \
+	.freq_tab_count = 2, \
+	.registers = &exynos5420_tmu_registers, \
+
+#define EXYNOS5420_TMU_DATA \
+	__EXYNOS5420_TMU_DATA \
+	.type = SOC_ARCH_EXYNOS5250, \
+	.features = (TMU_SUPPORT_EMULATION | TMU_SUPPORT_TRIM_RELOAD | \
+			TMU_SUPPORT_FALLING_TRIP | TMU_SUPPORT_READY_STATUS | \
+			TMU_SUPPORT_EMUL_TIME)
+
+#define EXYNOS5420_TMU_DATA_SHARED \
+	__EXYNOS5420_TMU_DATA \
+	.type = SOC_ARCH_EXYNOS5420_TRIMINFO, \
+	.features = (TMU_SUPPORT_EMULATION | TMU_SUPPORT_TRIM_RELOAD | \
+			TMU_SUPPORT_FALLING_TRIP | TMU_SUPPORT_READY_STATUS | \
+			TMU_SUPPORT_EMUL_TIME | TMU_SUPPORT_ADDRESS_MULTIPLE)
+
+struct exynos_tmu_init_data const exynos5420_default_tmu_data = {
+	.tmu_data = {
+		{ EXYNOS5420_TMU_DATA },
+		{ EXYNOS5420_TMU_DATA },
+		{ EXYNOS5420_TMU_DATA_SHARED },
+		{ EXYNOS5420_TMU_DATA_SHARED },
+		{ EXYNOS5420_TMU_DATA_SHARED },
+	},
+	.tmu_count = 5,
+};
+#endif
+
 #if defined(CONFIG_SOC_EXYNOS5440)
 static const struct exynos_tmu_registers exynos5440_tmu_registers = {
 	.triminfo_data = EXYNOS5440_TMU_S0_7_TRIM,
@@ -217,10 +408,6 @@ static const struct exynos_tmu_registers exynos5440_tmu_registers = {
 	.threshold_th2 = EXYNOS5440_TMU_S0_7_TH2,
 	.threshold_th3_l0_shift = EXYNOS5440_TMU_TH_RISE4_SHIFT,
 	.tmu_inten = EXYNOS5440_TMU_S0_7_IRQEN,
-	.inten_rise_mask = EXYNOS5440_TMU_RISE_INT_MASK,
-	.inten_rise_shift = EXYNOS5440_TMU_RISE_INT_SHIFT,
-	.inten_fall_mask = EXYNOS5440_TMU_FALL_INT_MASK,
-	.inten_fall_shift = EXYNOS5440_TMU_FALL_INT_SHIFT,
 	.inten_rise0_shift = EXYNOS5440_TMU_INTEN_RISE0_SHIFT,
 	.inten_rise1_shift = EXYNOS5440_TMU_INTEN_RISE1_SHIFT,
 	.inten_rise2_shift = EXYNOS5440_TMU_INTEN_RISE2_SHIFT,
@@ -228,6 +415,10 @@ static const struct exynos_tmu_registers exynos5440_tmu_registers = {
 	.inten_fall0_shift = EXYNOS5440_TMU_INTEN_FALL0_SHIFT,
 	.tmu_intstat = EXYNOS5440_TMU_S0_7_IRQ,
 	.tmu_intclear = EXYNOS5440_TMU_S0_7_IRQ,
+	.intclr_fall_shift = EXYNOS5440_TMU_CLEAR_FALL_INT_SHIFT,
+	.intclr_rise_shift = EXYNOS5440_TMU_RISE_INT_SHIFT,
+	.intclr_rise_mask = EXYNOS5440_TMU_RISE_INT_MASK,
+	.intclr_fall_mask = EXYNOS5440_TMU_FALL_INT_MASK,
 	.tmu_irqstatus = EXYNOS5440_TMU_IRQ_STATUS,
 	.emul_con = EXYNOS5440_TMU_S0_7_DEBUG,
 	.emul_temp_shift = EXYNOS_EMUL_DATA_SHIFT,
@@ -255,7 +446,7 @@ static const struct exynos_tmu_registers exynos5440_tmu_registers = {
 	.type = SOC_ARCH_EXYNOS5440, \
 	.registers = &exynos5440_tmu_registers, \
 	.features = (TMU_SUPPORT_EMULATION | TMU_SUPPORT_FALLING_TRIP | \
-			TMU_SUPPORT_MULTI_INST | TMU_SUPPORT_SHARED_MEMORY),
+			TMU_SUPPORT_MULTI_INST | TMU_SUPPORT_ADDRESS_MULTIPLE),
 
 struct exynos_tmu_init_data const exynos5440_default_tmu_data = {
 	.tmu_data = {
diff --git a/drivers/thermal/samsung/exynos_tmu_data.h b/drivers/thermal/samsung/exynos_tmu_data.h
index a1ea19d9e0a6..d268981b65e5 100644
--- a/drivers/thermal/samsung/exynos_tmu_data.h
+++ b/drivers/thermal/samsung/exynos_tmu_data.h
@@ -69,9 +69,11 @@
 #define EXYNOS_TMU_RISE_INT_MASK	0x111
 #define EXYNOS_TMU_RISE_INT_SHIFT	0
 #define EXYNOS_TMU_FALL_INT_MASK	0x111
-#define EXYNOS_TMU_FALL_INT_SHIFT	12
 #define EXYNOS_TMU_CLEAR_RISE_INT	0x111
 #define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 12)
+#define EXYNOS_TMU_CLEAR_FALL_INT_SHIFT	12
+#define EXYNOS5420_TMU_CLEAR_FALL_INT_SHIFT	16
+#define EXYNOS5440_TMU_CLEAR_FALL_INT_SHIFT	4
 #define EXYNOS_TMU_TRIP_MODE_SHIFT	13
 #define EXYNOS_TMU_TRIP_MODE_MASK	0x7
 #define EXYNOS_TMU_THERM_TRIP_EN_SHIFT	12
@@ -85,6 +87,7 @@
 #define EXYNOS_TMU_INTEN_FALL0_SHIFT	16
 #define EXYNOS_TMU_INTEN_FALL1_SHIFT	20
 #define EXYNOS_TMU_INTEN_FALL2_SHIFT	24
+#define EXYNOS_TMU_INTEN_FALL3_SHIFT	28
 
 #define EXYNOS_EMUL_TIME	0x57F0
 #define EXYNOS_EMUL_TIME_MASK	0xffff
@@ -95,6 +98,17 @@
 
 #define EXYNOS_MAX_TRIGGER_PER_REG	4
 
+/* Exynos5260 specific */
+#define EXYNOS_TMU_REG_CONTROL1			0x24
+#define EXYNOS5260_TMU_REG_INTEN		0xC0
+#define EXYNOS5260_TMU_REG_INTSTAT		0xC4
+#define EXYNOS5260_TMU_REG_INTCLEAR		0xC8
+#define EXYNOS5260_TMU_CLEAR_RISE_INT		0x1111
+#define EXYNOS5260_TMU_CLEAR_FALL_INT		(0x1111 << 16)
+#define EXYNOS5260_TMU_RISE_INT_MASK		0x1111
+#define EXYNOS5260_TMU_FALL_INT_MASK		0x1111
+#define EXYNOS5260_EMUL_CON			0x100
+
 /* Exynos4412 specific */
 #define EXYNOS4412_MUX_ADDR_VALUE          6
 #define EXYNOS4412_MUX_ADDR_SHIFT          20
@@ -119,7 +133,6 @@
 #define EXYNOS5440_TMU_RISE_INT_MASK		0xf
 #define EXYNOS5440_TMU_RISE_INT_SHIFT		0
 #define EXYNOS5440_TMU_FALL_INT_MASK		0xf
-#define EXYNOS5440_TMU_FALL_INT_SHIFT		4
 #define EXYNOS5440_TMU_INTEN_RISE0_SHIFT	0
 #define EXYNOS5440_TMU_INTEN_RISE1_SHIFT	1
 #define EXYNOS5440_TMU_INTEN_RISE2_SHIFT	2
@@ -156,6 +169,20 @@ extern struct exynos_tmu_init_data const exynos5250_default_tmu_data;
 #define EXYNOS5250_TMU_DRV_DATA (NULL)
 #endif
 
+#if defined(CONFIG_SOC_EXYNOS5260)
+extern struct exynos_tmu_init_data const exynos5260_default_tmu_data;
+#define EXYNOS5260_TMU_DRV_DATA (&exynos5260_default_tmu_data)
+#else
+#define EXYNOS5260_TMU_DRV_DATA (NULL)
+#endif
+
+#if defined(CONFIG_SOC_EXYNOS5420)
+extern struct exynos_tmu_init_data const exynos5420_default_tmu_data;
+#define EXYNOS5420_TMU_DRV_DATA (&exynos5420_default_tmu_data)
+#else
+#define EXYNOS5420_TMU_DRV_DATA (NULL)
+#endif
+
 #if defined(CONFIG_SOC_EXYNOS5440)
 extern struct exynos_tmu_init_data const exynos5440_default_tmu_data;
 #define EXYNOS5440_TMU_DRV_DATA (&exynos5440_default_tmu_data)
diff --git a/drivers/thermal/spear_thermal.c b/drivers/thermal/spear_thermal.c
index ab79ea4701d9..1e2193fc3241 100644
--- a/drivers/thermal/spear_thermal.c
+++ b/drivers/thermal/spear_thermal.c
@@ -113,10 +113,8 @@ static int spear_thermal_probe(struct platform_device *pdev)
 	}
 
 	stdev = devm_kzalloc(&pdev->dev, sizeof(*stdev), GFP_KERNEL);
-	if (!stdev) {
-		dev_err(&pdev->dev, "kzalloc fail\n");
+	if (!stdev)
 		return -ENOMEM;
-	}
 
 	/* Enable thermal sensor */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
diff --git a/drivers/thermal/thermal_hwmon.c b/drivers/thermal/thermal_hwmon.c
index fdb07199d9c2..1967bee4f076 100644
--- a/drivers/thermal/thermal_hwmon.c
+++ b/drivers/thermal/thermal_hwmon.c
@@ -140,6 +140,12 @@ thermal_hwmon_lookup_temp(const struct thermal_hwmon_device *hwmon,
 	return NULL;
 }
 
+static bool thermal_zone_crit_temp_valid(struct thermal_zone_device *tz)
+{
+	unsigned long temp;
+	return tz->ops->get_crit_temp && !tz->ops->get_crit_temp(tz, &temp);
+}
+
 int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz)
 {
 	struct thermal_hwmon_device *hwmon;
@@ -189,21 +195,18 @@ int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz)
 	if (result)
 		goto free_temp_mem;
 
-	if (tz->ops->get_crit_temp) {
-		unsigned long temperature;
-		if (!tz->ops->get_crit_temp(tz, &temperature)) {
-			snprintf(temp->temp_crit.name,
-				 sizeof(temp->temp_crit.name),
+	if (thermal_zone_crit_temp_valid(tz)) {
+		snprintf(temp->temp_crit.name,
+				sizeof(temp->temp_crit.name),
 				"temp%d_crit", hwmon->count);
-			temp->temp_crit.attr.attr.name = temp->temp_crit.name;
-			temp->temp_crit.attr.attr.mode = 0444;
-			temp->temp_crit.attr.show = temp_crit_show;
-			sysfs_attr_init(&temp->temp_crit.attr.attr);
-			result = device_create_file(hwmon->device,
-						    &temp->temp_crit.attr);
-			if (result)
-				goto unregister_input;
-		}
+		temp->temp_crit.attr.attr.name = temp->temp_crit.name;
+		temp->temp_crit.attr.attr.mode = 0444;
+		temp->temp_crit.attr.show = temp_crit_show;
+		sysfs_attr_init(&temp->temp_crit.attr.attr);
+		result = device_create_file(hwmon->device,
+					    &temp->temp_crit.attr);
+		if (result)
+			goto unregister_input;
 	}
 
 	mutex_lock(&thermal_hwmon_list_lock);
@@ -250,7 +253,7 @@ void thermal_remove_hwmon_sysfs(struct thermal_zone_device *tz)
 	}
 
 	device_remove_file(hwmon->device, &temp->temp_input.attr);
-	if (tz->ops->get_crit_temp)
+	if (thermal_zone_crit_temp_valid(tz))
 		device_remove_file(hwmon->device, &temp->temp_crit.attr);
 
 	mutex_lock(&thermal_hwmon_list_lock);
diff --git a/drivers/thermal/ti-soc-thermal/ti-bandgap.c b/drivers/thermal/ti-soc-thermal/ti-bandgap.c
index 3ab12ee359b7..634b6ce0e63a 100644
--- a/drivers/thermal/ti-soc-thermal/ti-bandgap.c
+++ b/drivers/thermal/ti-soc-thermal/ti-bandgap.c
@@ -1155,7 +1155,7 @@ static struct ti_bandgap *ti_bandgap_build(struct platform_device *pdev)
 	/* register shadow for context save and restore */
 	bgp->regval = devm_kzalloc(&pdev->dev, sizeof(*bgp->regval) *
 				   bgp->conf->sensor_count, GFP_KERNEL);
-	if (!bgp) {
+	if (!bgp->regval) {
 		dev_err(&pdev->dev, "Unable to allocate mem for driver ref\n");
 		return ERR_PTR(-ENOMEM);
 	}
@@ -1248,7 +1248,7 @@ int ti_bandgap_probe(struct platform_device *pdev)
 	clk_rate = clk_round_rate(bgp->div_clk,
 				  bgp->conf->sensors[0].ts_data->max_freq);
 	if (clk_rate < bgp->conf->sensors[0].ts_data->min_freq ||
-	    clk_rate == 0xffffffff) {
+	    clk_rate <= 0) {
 		ret = -ENODEV;
 		dev_err(&pdev->dev, "wrong clock rate (%d)\n", clk_rate);
 		goto put_clks;
diff --git a/drivers/tty/hvc/hvc_tile.c b/drivers/tty/hvc/hvc_tile.c
index 147d49e95db2..df374860037c 100644
--- a/drivers/tty/hvc/hvc_tile.c
+++ b/drivers/tty/hvc/hvc_tile.c
@@ -196,7 +196,7 @@ static int __init hvc_tile_init(void)
 #ifndef __tilegx__
 	struct hvc_struct *hp;
 	hp = hvc_alloc(0, 0, &hvc_tile_get_put_ops, 128);
-	return IS_ERR(hp) ? PTR_ERR(hp) : 0;
+	return PTR_ERR_OR_ZERO(hp);
 #else
 	platform_device_register(&hvc_tile_pdev);
 	return platform_driver_register(&hvc_tile_driver);
diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c
index f95569dedc88..f44f1ba762c3 100644
--- a/drivers/tty/n_tty.c
+++ b/drivers/tty/n_tty.c
@@ -1214,15 +1214,16 @@ static void n_tty_receive_parity_error(struct tty_struct *tty, unsigned char c)
 {
 	struct n_tty_data *ldata = tty->disc_data;
 
-	if (I_IGNPAR(tty))
-		return;
-	if (I_PARMRK(tty)) {
-		put_tty_queue('\377', ldata);
-		put_tty_queue('\0', ldata);
-		put_tty_queue(c, ldata);
-	} else	if (I_INPCK(tty))
-		put_tty_queue('\0', ldata);
-	else
+	if (I_INPCK(tty)) {
+		if (I_IGNPAR(tty))
+			return;
+		if (I_PARMRK(tty)) {
+			put_tty_queue('\377', ldata);
+			put_tty_queue('\0', ldata);
+			put_tty_queue(c, ldata);
+		} else
+			put_tty_queue('\0', ldata);
+	} else
 		put_tty_queue(c, ldata);
 	if (waitqueue_active(&tty->read_wait))
 		wake_up_interruptible(&tty->read_wait);
diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c
index 27f7ad6b74c1..7a91c6d1eb7d 100644
--- a/drivers/tty/serial/8250/8250_core.c
+++ b/drivers/tty/serial/8250/8250_core.c
@@ -2357,7 +2357,7 @@ serial8250_do_set_termios(struct uart_port *port, struct ktermios *termios,
 	port->read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= UART_LSR_BI;
 
 	/*
diff --git a/drivers/tty/serial/8250/8250_early.c b/drivers/tty/serial/8250/8250_early.c
index cfef801a49d4..4858b8a99d3b 100644
--- a/drivers/tty/serial/8250/8250_early.c
+++ b/drivers/tty/serial/8250/8250_early.c
@@ -144,8 +144,11 @@ static int __init early_serial8250_setup(struct earlycon_device *device,
 	if (!(device->port.membase || device->port.iobase))
 		return 0;
 
-	if (!device->baud)
+	if (!device->baud) {
 		device->baud = probe_baud(&device->port);
+		snprintf(device->options, sizeof(device->options), "%u",
+			 device->baud);
+	}
 
 	init_port(device);
 
diff --git a/drivers/tty/serial/altera_uart.c b/drivers/tty/serial/altera_uart.c
index 501667e3e3f5..323376668b72 100644
--- a/drivers/tty/serial/altera_uart.c
+++ b/drivers/tty/serial/altera_uart.c
@@ -185,6 +185,12 @@ static void altera_uart_set_termios(struct uart_port *port,
 	uart_update_timeout(port, termios->c_cflag, baud);
 	altera_uart_writel(port, baudclk, ALTERA_UART_DIVISOR_REG);
 	spin_unlock_irqrestore(&port->lock, flags);
+
+	/*
+	 * FIXME: port->read_status_mask and port->ignore_status_mask
+	 * need to be initialized based on termios settings for
+	 * INPCK, IGNBRK, IGNPAR, PARMRK, BRKINT
+	 */
 }
 
 static void altera_uart_rx_chars(struct altera_uart *pp)
diff --git a/drivers/tty/serial/amba-pl010.c b/drivers/tty/serial/amba-pl010.c
index 01c9e72433e1..971af1e22d0f 100644
--- a/drivers/tty/serial/amba-pl010.c
+++ b/drivers/tty/serial/amba-pl010.c
@@ -420,7 +420,7 @@ pl010_set_termios(struct uart_port *port, struct ktermios *termios,
 	uap->port.read_status_mask = UART01x_RSR_OE;
 	if (termios->c_iflag & INPCK)
 		uap->port.read_status_mask |= UART01x_RSR_FE | UART01x_RSR_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		uap->port.read_status_mask |= UART01x_RSR_BE;
 
 	/*
diff --git a/drivers/tty/serial/amba-pl011.c b/drivers/tty/serial/amba-pl011.c
index 908a6e3142a2..0e26dcbd5ea4 100644
--- a/drivers/tty/serial/amba-pl011.c
+++ b/drivers/tty/serial/amba-pl011.c
@@ -1744,7 +1744,7 @@ pl011_set_termios(struct uart_port *port, struct ktermios *termios,
 	port->read_status_mask = UART011_DR_OE | 255;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= UART011_DR_FE | UART011_DR_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= UART011_DR_BE;
 
 	/*
diff --git a/drivers/tty/serial/arc_uart.c b/drivers/tty/serial/arc_uart.c
index c9f5c9dcc15c..008c223eaf26 100644
--- a/drivers/tty/serial/arc_uart.c
+++ b/drivers/tty/serial/arc_uart.c
@@ -177,7 +177,7 @@ static void arc_serial_tx_chars(struct arc_uart_port *uart)
 		uart->port.icount.tx++;
 		uart->port.x_char = 0;
 		sent = 1;
-	} else if (xmit->tail != xmit->head) {	/* TODO: uart_circ_empty */
+	} else if (!uart_circ_empty(xmit)) {
 		ch = xmit->buf[xmit->tail];
 		xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
 		uart->port.icount.tx++;
diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
index 3fceae099c44..c4f750314100 100644
--- a/drivers/tty/serial/atmel_serial.c
+++ b/drivers/tty/serial/atmel_serial.c
@@ -1932,7 +1932,7 @@ static void atmel_set_termios(struct uart_port *port, struct ktermios *termios,
 	port->read_status_mask = ATMEL_US_OVRE;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= (ATMEL_US_FRAME | ATMEL_US_PARE);
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= ATMEL_US_RXBRK;
 
 	if (atmel_use_pdc_rx(port))
diff --git a/drivers/tty/serial/bcm63xx_uart.c b/drivers/tty/serial/bcm63xx_uart.c
index a47421e4627c..231519022b73 100644
--- a/drivers/tty/serial/bcm63xx_uart.c
+++ b/drivers/tty/serial/bcm63xx_uart.c
@@ -567,7 +567,7 @@ static void bcm_uart_set_termios(struct uart_port *port,
 		port->read_status_mask |= UART_FIFO_FRAMEERR_MASK;
 		port->read_status_mask |= UART_FIFO_PARERR_MASK;
 	}
-	if (new->c_iflag & (BRKINT))
+	if (new->c_iflag & (IGNBRK | BRKINT))
 		port->read_status_mask |= UART_FIFO_BRKDET_MASK;
 
 	port->ignore_status_mask = 0;
diff --git a/drivers/tty/serial/bfin_uart.c b/drivers/tty/serial/bfin_uart.c
index 869ceba2ec57..ac86a20992e9 100644
--- a/drivers/tty/serial/bfin_uart.c
+++ b/drivers/tty/serial/bfin_uart.c
@@ -833,7 +833,7 @@ bfin_serial_set_termios(struct uart_port *port, struct ktermios *termios,
 	port->read_status_mask = OE;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= (FE | PE);
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= BI;
 
 	/*
diff --git a/drivers/tty/serial/dz.c b/drivers/tty/serial/dz.c
index 2f2b2e538a54..cdbbc788230a 100644
--- a/drivers/tty/serial/dz.c
+++ b/drivers/tty/serial/dz.c
@@ -625,7 +625,7 @@ static void dz_set_termios(struct uart_port *uport, struct ktermios *termios,
 	dport->port.read_status_mask = DZ_OERR;
 	if (termios->c_iflag & INPCK)
 		dport->port.read_status_mask |= DZ_FERR | DZ_PERR;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		dport->port.read_status_mask |= DZ_BREAK;
 
 	/* characters to ignore */
diff --git a/drivers/tty/serial/earlycon.c b/drivers/tty/serial/earlycon.c
index 5131b5ee6164..a514ee6f5406 100644
--- a/drivers/tty/serial/earlycon.c
+++ b/drivers/tty/serial/earlycon.c
@@ -25,7 +25,7 @@
 #include <asm/serial.h>
 
 static struct console early_con = {
-	.name =		"earlycon",
+	.name =		"uart", /* 8250 console switch requires this name */
 	.flags =	CON_PRINTBUFFER | CON_BOOT,
 	.index =	-1,
 };
diff --git a/drivers/tty/serial/efm32-uart.c b/drivers/tty/serial/efm32-uart.c
index b373f6416e8c..3b0ee9afd76f 100644
--- a/drivers/tty/serial/efm32-uart.c
+++ b/drivers/tty/serial/efm32-uart.c
@@ -407,7 +407,7 @@ static void efm32_uart_set_termios(struct uart_port *port,
 	if (new->c_iflag & INPCK)
 		port->read_status_mask |=
 			UARTn_RXDATAX_FERR | UARTn_RXDATAX_PERR;
-	if (new->c_iflag & (BRKINT | PARMRK))
+	if (new->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= SW_UARTn_RXDATAX_BERR;
 
 	port->ignore_status_mask = 0;
diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c
index c5eb897de9de..49385c86cfba 100644
--- a/drivers/tty/serial/fsl_lpuart.c
+++ b/drivers/tty/serial/fsl_lpuart.c
@@ -902,7 +902,7 @@ lpuart_set_termios(struct uart_port *port, struct ktermios *termios,
 	sport->port.read_status_mask = 0;
 	if (termios->c_iflag & INPCK)
 		sport->port.read_status_mask |=	(UARTSR1_FE | UARTSR1_PE);
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		sport->port.read_status_mask |= UARTSR1_FE;
 
 	/* characters to ignore */
diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c
index e2f93874989b..044e86d528ae 100644
--- a/drivers/tty/serial/imx.c
+++ b/drivers/tty/serial/imx.c
@@ -567,6 +567,9 @@ static void imx_start_tx(struct uart_port *port)
 	struct imx_port *sport = (struct imx_port *)port;
 	unsigned long temp;
 
+	if (uart_circ_empty(&port->state->xmit))
+		return;
+
 	if (USE_IRDA(sport)) {
 		/* half duplex in IrDA mode; have to disable receive mode */
 		temp = readl(sport->port.membase + UCR4);
diff --git a/drivers/tty/serial/ip22zilog.c b/drivers/tty/serial/ip22zilog.c
index 1d9420548e16..99b7b8697861 100644
--- a/drivers/tty/serial/ip22zilog.c
+++ b/drivers/tty/serial/ip22zilog.c
@@ -603,6 +603,8 @@ static void ip22zilog_start_tx(struct uart_port *port)
 	} else {
 		struct circ_buf *xmit = &port->state->xmit;
 
+		if (uart_circ_empty(xmit))
+			return;
 		writeb(xmit->buf[xmit->tail], &channel->data);
 		ZSDELAY();
 		ZS_WSYNC(channel);
@@ -850,7 +852,7 @@ ip22zilog_convert_to_zs(struct uart_ip22zilog_port *up, unsigned int cflag,
 	up->port.read_status_mask = Rx_OVR;
 	if (iflag & INPCK)
 		up->port.read_status_mask |= CRC_ERR | PAR_ERR;
-	if (iflag & (BRKINT | PARMRK))
+	if (iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= BRK_ABRT;
 
 	up->port.ignore_status_mask = 0;
diff --git a/drivers/tty/serial/m32r_sio.c b/drivers/tty/serial/m32r_sio.c
index 9cd9b4eba9fc..5702828fb62e 100644
--- a/drivers/tty/serial/m32r_sio.c
+++ b/drivers/tty/serial/m32r_sio.c
@@ -266,9 +266,11 @@ static void m32r_sio_start_tx(struct uart_port *port)
 	if (!(up->ier & UART_IER_THRI)) {
 		up->ier |= UART_IER_THRI;
 		serial_out(up, UART_IER, up->ier);
-		serial_out(up, UART_TX, xmit->buf[xmit->tail]);
-		xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
-		up->port.icount.tx++;
+		if (!uart_circ_empty(xmit)) {
+			serial_out(up, UART_TX, xmit->buf[xmit->tail]);
+			xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
+			up->port.icount.tx++;
+		}
 	}
 	while((serial_in(up, UART_LSR) & UART_EMPTY) != UART_EMPTY);
 #else
@@ -737,7 +739,7 @@ static void m32r_sio_set_termios(struct uart_port *port,
 	up->port.read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
 	if (termios->c_iflag & INPCK)
 		up->port.read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= UART_LSR_BI;
 
 	/*
diff --git a/drivers/tty/serial/max310x.c b/drivers/tty/serial/max310x.c
index 2a99d0c61b9e..ba285cd45b59 100644
--- a/drivers/tty/serial/max310x.c
+++ b/drivers/tty/serial/max310x.c
@@ -835,7 +835,7 @@ static void max310x_set_termios(struct uart_port *port,
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= MAX310X_LSR_RXPAR_BIT |
 					  MAX310X_LSR_FRERR_BIT;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= MAX310X_LSR_RXBRK_BIT;
 
 	/* Set status ignore mask */
diff --git a/drivers/tty/serial/mcf.c b/drivers/tty/serial/mcf.c
index 0edfaf8cd269..a6f085717f94 100644
--- a/drivers/tty/serial/mcf.c
+++ b/drivers/tty/serial/mcf.c
@@ -248,6 +248,12 @@ static void mcf_set_termios(struct uart_port *port, struct ktermios *termios,
 		mr1 |= MCFUART_MR1_PARITYNONE;
 	}
 
+	/*
+	 * FIXME: port->read_status_mask and port->ignore_status_mask
+	 * need to be initialized based on termios settings for
+	 * INPCK, IGNBRK, IGNPAR, PARMRK, BRKINT
+	 */
+
 	if (termios->c_cflag & CSTOPB)
 		mr2 |= MCFUART_MR2_STOP2;
 	else
diff --git a/drivers/tty/serial/mfd.c b/drivers/tty/serial/mfd.c
index 52c930fac210..445799dc9846 100644
--- a/drivers/tty/serial/mfd.c
+++ b/drivers/tty/serial/mfd.c
@@ -977,7 +977,7 @@ serial_hsu_set_termios(struct uart_port *port, struct ktermios *termios,
 	up->port.read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
 	if (termios->c_iflag & INPCK)
 		up->port.read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= UART_LSR_BI;
 
 	/* Characters to ignore */
diff --git a/drivers/tty/serial/mpsc.c b/drivers/tty/serial/mpsc.c
index e30a3ca3cea3..759c6a6fa74a 100644
--- a/drivers/tty/serial/mpsc.c
+++ b/drivers/tty/serial/mpsc.c
@@ -1458,7 +1458,7 @@ static void mpsc_set_termios(struct uart_port *port, struct ktermios *termios,
 		pi->port.read_status_mask |= SDMA_DESC_CMDSTAT_PE
 			| SDMA_DESC_CMDSTAT_FR;
 
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		pi->port.read_status_mask |= SDMA_DESC_CMDSTAT_BR;
 
 	/* Characters/events to ignore */
diff --git a/drivers/tty/serial/msm_serial.c b/drivers/tty/serial/msm_serial.c
index 778e376f197e..72000a6d5af0 100644
--- a/drivers/tty/serial/msm_serial.c
+++ b/drivers/tty/serial/msm_serial.c
@@ -582,7 +582,7 @@ static void msm_set_termios(struct uart_port *port, struct ktermios *termios,
 	port->read_status_mask = 0;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= UART_SR_PAR_FRAME_ERR;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= UART_SR_RX_BREAK;
 
 	uart_update_timeout(port, termios->c_cflag, baud);
@@ -991,7 +991,7 @@ static const struct of_device_id msm_uartdm_table[] = {
 	{ }
 };
 
-static int __init msm_serial_probe(struct platform_device *pdev)
+static int msm_serial_probe(struct platform_device *pdev)
 {
 	struct msm_port *msm_port;
 	struct resource *resource;
diff --git a/drivers/tty/serial/mxs-auart.c b/drivers/tty/serial/mxs-auart.c
index 4b5b3c2fe328..86de4477d98a 100644
--- a/drivers/tty/serial/mxs-auart.c
+++ b/drivers/tty/serial/mxs-auart.c
@@ -604,7 +604,7 @@ static void mxs_auart_settermios(struct uart_port *u,
 
 	if (termios->c_iflag & INPCK)
 		u->read_status_mask |= AUART_STAT_PERR;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		u->read_status_mask |= AUART_STAT_BERR;
 
 	/*
diff --git a/drivers/tty/serial/netx-serial.c b/drivers/tty/serial/netx-serial.c
index 0a4dd70d29eb..7a6745601d4e 100644
--- a/drivers/tty/serial/netx-serial.c
+++ b/drivers/tty/serial/netx-serial.c
@@ -419,7 +419,7 @@ netx_set_termios(struct uart_port *port, struct ktermios *termios,
 	}
 
 	port->read_status_mask = 0;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= SR_BE;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= SR_PE | SR_FE;
diff --git a/drivers/tty/serial/pmac_zilog.c b/drivers/tty/serial/pmac_zilog.c
index e9d420ff3931..f7ad5b903055 100644
--- a/drivers/tty/serial/pmac_zilog.c
+++ b/drivers/tty/serial/pmac_zilog.c
@@ -653,6 +653,8 @@ static void pmz_start_tx(struct uart_port *port)
 	} else {
 		struct circ_buf *xmit = &port->state->xmit;
 
+		if (uart_circ_empty(xmit))
+			goto out;
 		write_zsdata(uap, xmit->buf[xmit->tail]);
 		zssync(uap);
 		xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
@@ -661,6 +663,7 @@ static void pmz_start_tx(struct uart_port *port)
 		if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
 			uart_write_wakeup(&uap->port);
 	}
+ out:
 	pmz_debug("pmz: start_tx() done.\n");
 }
 
@@ -1092,7 +1095,7 @@ static void pmz_convert_to_zs(struct uart_pmac_port *uap, unsigned int cflag,
 	uap->port.read_status_mask = Rx_OVR;
 	if (iflag & INPCK)
 		uap->port.read_status_mask |= CRC_ERR | PAR_ERR;
-	if (iflag & (BRKINT | PARMRK))
+	if (iflag & (IGNBRK | BRKINT | PARMRK))
 		uap->port.read_status_mask |= BRK_ABRT;
 
 	uap->port.ignore_status_mask = 0;
diff --git a/drivers/tty/serial/pnx8xxx_uart.c b/drivers/tty/serial/pnx8xxx_uart.c
index de6c05c63683..2ba24a45c97f 100644
--- a/drivers/tty/serial/pnx8xxx_uart.c
+++ b/drivers/tty/serial/pnx8xxx_uart.c
@@ -477,7 +477,7 @@ pnx8xxx_set_termios(struct uart_port *port, struct ktermios *termios,
 		sport->port.read_status_mask |=
 			FIFO_TO_SM(PNX8XXX_UART_FIFO_RXFE) |
 			FIFO_TO_SM(PNX8XXX_UART_FIFO_RXPAR);
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		sport->port.read_status_mask |=
 			ISTAT_TO_SM(PNX8XXX_UART_INT_BREAK);
 
diff --git a/drivers/tty/serial/pxa.c b/drivers/tty/serial/pxa.c
index 9e7ee39f8b2a..c638c53cd2b6 100644
--- a/drivers/tty/serial/pxa.c
+++ b/drivers/tty/serial/pxa.c
@@ -492,7 +492,7 @@ serial_pxa_set_termios(struct uart_port *port, struct ktermios *termios,
 	up->port.read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
 	if (termios->c_iflag & INPCK)
 		up->port.read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= UART_LSR_BI;
 
 	/*
diff --git a/drivers/tty/serial/samsung.c b/drivers/tty/serial/samsung.c
index 329337711bb0..c1d3ebdf3b97 100644
--- a/drivers/tty/serial/samsung.c
+++ b/drivers/tty/serial/samsung.c
@@ -66,7 +66,7 @@ static void dbg(const char *fmt, ...)
 	char buff[256];
 
 	va_start(va, fmt);
-	vscnprintf(buff, sizeof(buf), fmt, va);
+	vscnprintf(buff, sizeof(buff), fmt, va);
 	va_end(va);
 
 	printascii(buff);
diff --git a/drivers/tty/serial/sb1250-duart.c b/drivers/tty/serial/sb1250-duart.c
index a7cdec2962dd..771f361c47ea 100644
--- a/drivers/tty/serial/sb1250-duart.c
+++ b/drivers/tty/serial/sb1250-duart.c
@@ -596,7 +596,7 @@ static void sbd_set_termios(struct uart_port *uport, struct ktermios *termios,
 	if (termios->c_iflag & INPCK)
 		uport->read_status_mask |= M_DUART_FRM_ERR |
 					   M_DUART_PARITY_ERR;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		uport->read_status_mask |= M_DUART_RCVD_BRK;
 
 	uport->ignore_status_mask = 0;
diff --git a/drivers/tty/serial/sccnxp.c b/drivers/tty/serial/sccnxp.c
index 5443b46345ed..e84b6a3bdd18 100644
--- a/drivers/tty/serial/sccnxp.c
+++ b/drivers/tty/serial/sccnxp.c
@@ -665,7 +665,7 @@ static void sccnxp_set_termios(struct uart_port *port,
 	port->read_status_mask = SR_OVR;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= SR_PE | SR_FE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= SR_BRK;
 
 	/* Set status ignore mask */
diff --git a/drivers/tty/serial/serial_ks8695.c b/drivers/tty/serial/serial_ks8695.c
index e1caa99e3d3b..5c79bdab985d 100644
--- a/drivers/tty/serial/serial_ks8695.c
+++ b/drivers/tty/serial/serial_ks8695.c
@@ -437,7 +437,7 @@ static void ks8695uart_set_termios(struct uart_port *port, struct ktermios *term
 	port->read_status_mask = URLS_URROE;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= (URLS_URFE | URLS_URPE);
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= URLS_URBI;
 
 	/*
diff --git a/drivers/tty/serial/serial_txx9.c b/drivers/tty/serial/serial_txx9.c
index 60f49b9d7e39..ea8546092c7e 100644
--- a/drivers/tty/serial/serial_txx9.c
+++ b/drivers/tty/serial/serial_txx9.c
@@ -697,7 +697,7 @@ serial_txx9_set_termios(struct uart_port *port, struct ktermios *termios,
 		TXX9_SIDISR_TDIS | TXX9_SIDISR_RDIS;
 	if (termios->c_iflag & INPCK)
 		up->port.read_status_mask |= TXX9_SIDISR_UFER | TXX9_SIDISR_UPER;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= TXX9_SIDISR_UBRK;
 
 	/*
diff --git a/drivers/tty/serial/sirfsoc_uart.c b/drivers/tty/serial/sirfsoc_uart.c
index 1f2be48c92ce..9b4d71cff00d 100644
--- a/drivers/tty/serial/sirfsoc_uart.c
+++ b/drivers/tty/serial/sirfsoc_uart.c
@@ -896,7 +896,7 @@ static void sirfsoc_uart_set_termios(struct uart_port *port,
 		if (termios->c_iflag & INPCK)
 			port->read_status_mask |= uint_en->sirfsoc_frm_err_en;
 	}
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 			port->read_status_mask |= uint_en->sirfsoc_rxd_brk_en;
 	if (sirfport->uart_reg->uart_type == SIRF_REAL_UART) {
 		if (termios->c_iflag & IGNPAR)
diff --git a/drivers/tty/serial/st-asc.c b/drivers/tty/serial/st-asc.c
index c7f61ac27132..f48b1cc07eea 100644
--- a/drivers/tty/serial/st-asc.c
+++ b/drivers/tty/serial/st-asc.c
@@ -547,7 +547,7 @@ static void asc_set_termios(struct uart_port *port, struct ktermios *termios,
 	ascport->port.read_status_mask = ASC_RXBUF_DUMMY_OE;
 	if (termios->c_iflag & INPCK)
 		ascport->port.read_status_mask |= ASC_RXBUF_FE | ASC_RXBUF_PE;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		ascport->port.read_status_mask |= ASC_RXBUF_DUMMY_BE;
 
 	/*
diff --git a/drivers/tty/serial/sunsab.c b/drivers/tty/serial/sunsab.c
index 5faa8e905e98..2f57df9a71d9 100644
--- a/drivers/tty/serial/sunsab.c
+++ b/drivers/tty/serial/sunsab.c
@@ -427,6 +427,9 @@ static void sunsab_start_tx(struct uart_port *port)
 	struct circ_buf *xmit = &up->port.state->xmit;
 	int i;
 
+	if (uart_circ_empty(xmit))
+		return;
+
 	up->interrupt_mask1 &= ~(SAB82532_IMR1_ALLS|SAB82532_IMR1_XPR);
 	writeb(up->interrupt_mask1, &up->regs->w.imr1);
 	
@@ -719,7 +722,7 @@ static void sunsab_convert_to_sab(struct uart_sunsab_port *up, unsigned int cfla
 	if (iflag & INPCK)
 		up->port.read_status_mask |= (SAB82532_ISR0_PERR |
 					      SAB82532_ISR0_FERR);
-	if (iflag & (BRKINT | PARMRK))
+	if (iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= (SAB82532_ISR1_BRK << 8);
 
 	/*
diff --git a/drivers/tty/serial/sunsu.c b/drivers/tty/serial/sunsu.c
index 9a0f24f83720..5326ae195e5f 100644
--- a/drivers/tty/serial/sunsu.c
+++ b/drivers/tty/serial/sunsu.c
@@ -834,7 +834,7 @@ sunsu_change_speed(struct uart_port *port, unsigned int cflag,
 	up->port.read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
 	if (iflag & INPCK)
 		up->port.read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (iflag & (BRKINT | PARMRK))
+	if (iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= UART_LSR_BI;
 
 	/*
diff --git a/drivers/tty/serial/sunzilog.c b/drivers/tty/serial/sunzilog.c
index a2c40ed287d2..02df3940b95e 100644
--- a/drivers/tty/serial/sunzilog.c
+++ b/drivers/tty/serial/sunzilog.c
@@ -703,6 +703,8 @@ static void sunzilog_start_tx(struct uart_port *port)
 	} else {
 		struct circ_buf *xmit = &port->state->xmit;
 
+		if (uart_circ_empty(xmit))
+			return;
 		writeb(xmit->buf[xmit->tail], &channel->data);
 		ZSDELAY();
 		ZS_WSYNC(channel);
@@ -915,7 +917,7 @@ sunzilog_convert_to_zs(struct uart_sunzilog_port *up, unsigned int cflag,
 	up->port.read_status_mask = Rx_OVR;
 	if (iflag & INPCK)
 		up->port.read_status_mask |= CRC_ERR | PAR_ERR;
-	if (iflag & (BRKINT | PARMRK))
+	if (iflag & (IGNBRK | BRKINT | PARMRK))
 		up->port.read_status_mask |= BRK_ABRT;
 
 	up->port.ignore_status_mask = 0;
diff --git a/drivers/tty/serial/ucc_uart.c b/drivers/tty/serial/ucc_uart.c
index d569ca58bab6..1c52074c38df 100644
--- a/drivers/tty/serial/ucc_uart.c
+++ b/drivers/tty/serial/ucc_uart.c
@@ -936,7 +936,7 @@ static void qe_uart_set_termios(struct uart_port *port,
 	port->read_status_mask = BD_SC_EMPTY | BD_SC_OV;
 	if (termios->c_iflag & INPCK)
 		port->read_status_mask |= BD_SC_FR | BD_SC_PR;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= BD_SC_BR;
 
 	/*
diff --git a/drivers/tty/serial/vr41xx_siu.c b/drivers/tty/serial/vr41xx_siu.c
index a63c14bc9a24..db0c8a4ab03e 100644
--- a/drivers/tty/serial/vr41xx_siu.c
+++ b/drivers/tty/serial/vr41xx_siu.c
@@ -559,7 +559,7 @@ static void siu_set_termios(struct uart_port *port, struct ktermios *new,
 	port->read_status_mask = UART_LSR_THRE | UART_LSR_OE | UART_LSR_DR;
 	if (c_iflag & INPCK)
 		port->read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (c_iflag & (BRKINT | PARMRK))
+	if (c_iflag & (IGNBRK | BRKINT | PARMRK))
 		port->read_status_mask |= UART_LSR_BI;
 
 	port->ignore_status_mask = 0;
diff --git a/drivers/tty/serial/zs.c b/drivers/tty/serial/zs.c
index 6a169877109b..2b65bb7ffb8a 100644
--- a/drivers/tty/serial/zs.c
+++ b/drivers/tty/serial/zs.c
@@ -923,7 +923,7 @@ static void zs_set_termios(struct uart_port *uport, struct ktermios *termios,
 	uport->read_status_mask = Rx_OVR;
 	if (termios->c_iflag & INPCK)
 		uport->read_status_mask |= FRM_ERR | PAR_ERR;
-	if (termios->c_iflag & (BRKINT | PARMRK))
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
 		uport->read_status_mask |= Rx_BRK;
 
 	uport->ignore_status_mask = 0;
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 5e0f6ff2e2f5..b33b00b386de 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -3226,8 +3226,7 @@ int do_unbind_con_driver(const struct consw *csw, int first, int last, int deflt
 	for (i = 0; i < MAX_NR_CON_DRIVER; i++) {
 		con_back = &registered_con_driver[i];
 
-		if (con_back->con &&
-		    !(con_back->flag & CON_DRIVER_FLAG_MODULE)) {
+		if (con_back->con && con_back->con != csw) {
 			defcsw = con_back->con;
 			retval = 0;
 			break;
@@ -3332,6 +3331,7 @@ static int vt_unbind(struct con_driver *con)
 {
 	const struct consw *csw = NULL;
 	int i, more = 1, first = -1, last = -1, deflt = 0;
+	int ret;
 
  	if (!con->con || !(con->flag & CON_DRIVER_FLAG_MODULE) ||
 	    con_is_graphics(con->con, con->first, con->last))
@@ -3357,8 +3357,10 @@ static int vt_unbind(struct con_driver *con)
 
 		if (first != -1) {
 			console_lock();
-			do_unbind_con_driver(csw, first, last, deflt);
+			ret = do_unbind_con_driver(csw, first, last, deflt);
 			console_unlock();
+			if (ret != 0)
+				return ret;
 		}
 
 		first = -1;
@@ -3645,17 +3647,20 @@ err:
  */
 int do_unregister_con_driver(const struct consw *csw)
 {
-	int i, retval = -ENODEV;
+	int i;
 
 	/* cannot unregister a bound driver */
 	if (con_is_bound(csw))
-		goto err;
+		return -EBUSY;
+
+	if (csw == conswitchp)
+		return -EINVAL;
 
 	for (i = 0; i < MAX_NR_CON_DRIVER; i++) {
 		struct con_driver *con_driver = &registered_con_driver[i];
 
 		if (con_driver->con == csw &&
-		    con_driver->flag & CON_DRIVER_FLAG_MODULE) {
+		    con_driver->flag & CON_DRIVER_FLAG_INIT) {
 			vtconsole_deinit_device(con_driver);
 			device_destroy(vtconsole_class,
 				       MKDEV(0, con_driver->node));
@@ -3666,12 +3671,11 @@ int do_unregister_con_driver(const struct consw *csw)
 			con_driver->flag = 0;
 			con_driver->first = 0;
 			con_driver->last = 0;
-			retval = 0;
-			break;
+			return 0;
 		}
 	}
-err:
-	return retval;
+
+	return -ENODEV;
 }
 EXPORT_SYMBOL_GPL(do_unregister_con_driver);
 
diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index e371f5af11f5..a673e5b6a2e0 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -655,7 +655,7 @@ static int uio_mmap_physical(struct vm_area_struct *vma)
 
 	if (mem->addr & ~PAGE_MASK)
 		return -ENODEV;
-	if (vma->vm_end - vma->vm_start > PAGE_ALIGN(mem->size))
+	if (vma->vm_end - vma->vm_start > mem->size)
 		return -EINVAL;
 
 	vma->vm_ops = &uio_physical_vm_ops;
diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c
index 69425b3cb6b7..9d2b673f90e3 100644
--- a/drivers/usb/chipidea/udc.c
+++ b/drivers/usb/chipidea/udc.c
@@ -1321,6 +1321,7 @@ static int ep_dequeue(struct usb_ep *ep, struct usb_request *req)
 	struct ci_hw_ep  *hwep  = container_of(ep,  struct ci_hw_ep, ep);
 	struct ci_hw_req *hwreq = container_of(req, struct ci_hw_req, req);
 	unsigned long flags;
+	struct td_node *node, *tmpnode;
 
 	if (ep == NULL || req == NULL || hwreq->req.status != -EALREADY ||
 		hwep->ep.desc == NULL || list_empty(&hwreq->queue) ||
@@ -1331,6 +1332,12 @@ static int ep_dequeue(struct usb_ep *ep, struct usb_request *req)
 
 	hw_ep_flush(hwep->ci, hwep->num, hwep->dir);
 
+	list_for_each_entry_safe(node, tmpnode, &hwreq->tds, td) {
+		dma_pool_free(hwep->td_pool, node->ptr, node->dma);
+		list_del(&node->td);
+		kfree(node);
+	}
+
 	/* pop request */
 	list_del_init(&hwreq->queue);
 
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 879b66e13370..21b99b4b4082 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1526,18 +1526,6 @@ static int hub_configure(struct usb_hub *hub,
 		dev_dbg(hub_dev, "%umA bus power budget for each child\n",
 				hub->mA_per_port);
 
-	/* Update the HCD's internal representation of this hub before khubd
-	 * starts getting port status changes for devices under the hub.
-	 */
-	if (hcd->driver->update_hub_device) {
-		ret = hcd->driver->update_hub_device(hcd, hdev,
-				&hub->tt, GFP_KERNEL);
-		if (ret < 0) {
-			message = "can't update HCD hub info";
-			goto fail;
-		}
-	}
-
 	ret = hub_hub_status(hub, &hubstatus, &hubchange);
 	if (ret < 0) {
 		message = "can't get hub status";
@@ -1589,10 +1577,28 @@ static int hub_configure(struct usb_hub *hub,
 		}
 	}
 	hdev->maxchild = i;
+	for (i = 0; i < hdev->maxchild; i++) {
+		struct usb_port *port_dev = hub->ports[i];
+
+		pm_runtime_put(&port_dev->dev);
+	}
+
 	mutex_unlock(&usb_port_peer_mutex);
 	if (ret < 0)
 		goto fail;
 
+	/* Update the HCD's internal representation of this hub before khubd
+	 * starts getting port status changes for devices under the hub.
+	 */
+	if (hcd->driver->update_hub_device) {
+		ret = hcd->driver->update_hub_device(hcd, hdev,
+				&hub->tt, GFP_KERNEL);
+		if (ret < 0) {
+			message = "can't update HCD hub info";
+			goto fail;
+		}
+	}
+
 	usb_hub_adjust_deviceremovable(hdev, hub->descriptor);
 
 	hub_activate(hub, HUB_INIT);
@@ -3458,7 +3464,8 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
 		struct usb_device *udev = port_dev->child;
 
 		if (udev && udev->can_submit) {
-			dev_warn(&port_dev->dev, "not suspended yet\n");
+			dev_warn(&port_dev->dev, "device %s not suspended yet\n",
+					dev_name(&udev->dev));
 			if (PMSG_IS_AUTO(msg))
 				return -EBUSY;
 		}
diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h
index 0a7cdc0ef0a9..326308e53961 100644
--- a/drivers/usb/core/hub.h
+++ b/drivers/usb/core/hub.h
@@ -84,6 +84,7 @@ struct usb_hub {
  * @dev: generic device interface
  * @port_owner: port's owner
  * @peer: related usb2 and usb3 ports (share the same connector)
+ * @req: default pm qos request for hubs without port power control
  * @connect_type: port's connect type
  * @location: opaque representation of platform connector location
  * @status_lock: synchronize port_event() vs usb_port_{suspend|resume}
@@ -95,6 +96,7 @@ struct usb_port {
 	struct device dev;
 	struct usb_dev_state *port_owner;
 	struct usb_port *peer;
+	struct dev_pm_qos_request *req;
 	enum usb_port_connect_type connect_type;
 	usb_port_location_t location;
 	struct mutex status_lock;
diff --git a/drivers/usb/core/port.c b/drivers/usb/core/port.c
index 62036faf56c0..fe1b6d0967e3 100644
--- a/drivers/usb/core/port.c
+++ b/drivers/usb/core/port.c
@@ -21,6 +21,8 @@
 
 #include "hub.h"
 
+static int usb_port_block_power_off;
+
 static const struct attribute_group *port_dev_group[];
 
 static ssize_t connect_type_show(struct device *dev,
@@ -66,6 +68,7 @@ static void usb_port_device_release(struct device *dev)
 {
 	struct usb_port *port_dev = to_usb_port(dev);
 
+	kfree(port_dev->req);
 	kfree(port_dev);
 }
 
@@ -142,6 +145,9 @@ static int usb_port_runtime_suspend(struct device *dev)
 			== PM_QOS_FLAGS_ALL)
 		return -EAGAIN;
 
+	if (usb_port_block_power_off)
+		return -EBUSY;
+
 	usb_autopm_get_interface(intf);
 	retval = usb_hub_set_port_power(hdev, hub, port1, false);
 	usb_clear_port_feature(hdev, port1, USB_PORT_FEAT_C_CONNECTION);
@@ -190,11 +196,19 @@ static int link_peers(struct usb_port *left, struct usb_port *right)
 	if (left->peer || right->peer) {
 		struct usb_port *lpeer = left->peer;
 		struct usb_port *rpeer = right->peer;
-
-		WARN(1, "failed to peer %s and %s (%s -> %p) (%s -> %p)\n",
-			dev_name(&left->dev), dev_name(&right->dev),
-			dev_name(&left->dev), lpeer,
-			dev_name(&right->dev), rpeer);
+		char *method;
+
+		if (left->location && left->location == right->location)
+			method = "location";
+		else
+			method = "default";
+
+		pr_warn("usb: failed to peer %s and %s by %s (%s:%s) (%s:%s)\n",
+			dev_name(&left->dev), dev_name(&right->dev), method,
+			dev_name(&left->dev),
+			lpeer ? dev_name(&lpeer->dev) : "none",
+			dev_name(&right->dev),
+			rpeer ? dev_name(&rpeer->dev) : "none");
 		return -EBUSY;
 	}
 
@@ -251,6 +265,7 @@ static void link_peers_report(struct usb_port *left, struct usb_port *right)
 		dev_warn(&left->dev, "failed to peer to %s (%d)\n",
 				dev_name(&right->dev), rc);
 		pr_warn_once("usb: port power management may be unreliable\n");
+		usb_port_block_power_off = 1;
 	}
 }
 
@@ -386,9 +401,13 @@ int usb_hub_create_port_device(struct usb_hub *hub, int port1)
 	int retval;
 
 	port_dev = kzalloc(sizeof(*port_dev), GFP_KERNEL);
-	if (!port_dev) {
-		retval = -ENOMEM;
-		goto exit;
+	if (!port_dev)
+		return -ENOMEM;
+
+	port_dev->req = kzalloc(sizeof(*(port_dev->req)), GFP_KERNEL);
+	if (!port_dev->req) {
+		kfree(port_dev);
+		return -ENOMEM;
 	}
 
 	hub->ports[port1 - 1] = port_dev;
@@ -404,31 +423,53 @@ int usb_hub_create_port_device(struct usb_hub *hub, int port1)
 			port1);
 	mutex_init(&port_dev->status_lock);
 	retval = device_register(&port_dev->dev);
-	if (retval)
-		goto error_register;
+	if (retval) {
+		put_device(&port_dev->dev);
+		return retval;
+	}
+
+	/* Set default policy of port-poweroff disabled. */
+	retval = dev_pm_qos_add_request(&port_dev->dev, port_dev->req,
+			DEV_PM_QOS_FLAGS, PM_QOS_FLAG_NO_POWER_OFF);
+	if (retval < 0) {
+		device_unregister(&port_dev->dev);
+		return retval;
+	}
 
 	find_and_link_peer(hub, port1);
 
+	/*
+	 * Enable runtime pm and hold a refernce that hub_configure()
+	 * will drop once the PM_QOS_NO_POWER_OFF flag state has been set
+	 * and the hub has been fully registered (hdev->maxchild set).
+	 */
 	pm_runtime_set_active(&port_dev->dev);
+	pm_runtime_get_noresume(&port_dev->dev);
+	pm_runtime_enable(&port_dev->dev);
+	device_enable_async_suspend(&port_dev->dev);
 
 	/*
-	 * Do not enable port runtime pm if the hub does not support
-	 * power switching.  Also, userspace must have final say of
-	 * whether a port is permitted to power-off.  Do not enable
-	 * runtime pm if we fail to expose pm_qos_no_power_off.
+	 * Keep hidden the ability to enable port-poweroff if the hub
+	 * does not support power switching.
 	 */
-	if (hub_is_port_power_switchable(hub)
-			&& dev_pm_qos_expose_flags(&port_dev->dev,
-			PM_QOS_FLAG_NO_POWER_OFF) == 0)
-		pm_runtime_enable(&port_dev->dev);
+	if (!hub_is_port_power_switchable(hub))
+		return 0;
 
-	device_enable_async_suspend(&port_dev->dev);
-	return 0;
+	/* Attempt to let userspace take over the policy. */
+	retval = dev_pm_qos_expose_flags(&port_dev->dev,
+			PM_QOS_FLAG_NO_POWER_OFF);
+	if (retval < 0) {
+		dev_warn(&port_dev->dev, "failed to expose pm_qos_no_poweroff\n");
+		return 0;
+	}
 
-error_register:
-	put_device(&port_dev->dev);
-exit:
-	return retval;
+	/* Userspace owns the policy, drop the kernel 'no_poweroff' request. */
+	retval = dev_pm_qos_remove_request(port_dev->req);
+	if (retval >= 0) {
+		kfree(port_dev->req);
+		port_dev->req = NULL;
+	}
+	return 0;
 }
 
 void usb_hub_remove_port_device(struct usb_hub *hub, int port1)
diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig
index 8eb996e4f058..261c3b428220 100644
--- a/drivers/usb/dwc3/Kconfig
+++ b/drivers/usb/dwc3/Kconfig
@@ -45,6 +45,7 @@ comment "Platform Glue Driver Support"
 config USB_DWC3_OMAP
 	tristate "Texas Instruments OMAP5 and similar Platforms"
 	depends on EXTCON && (ARCH_OMAP2PLUS || COMPILE_TEST)
+	depends on OF
 	default USB_DWC3
 	help
 	  Some platforms from Texas Instruments like OMAP5, DRA7xxx and
diff --git a/drivers/usb/dwc3/dwc3-omap.c b/drivers/usb/dwc3/dwc3-omap.c
index 4af4c3567656..07a736acd0f2 100644
--- a/drivers/usb/dwc3/dwc3-omap.c
+++ b/drivers/usb/dwc3/dwc3-omap.c
@@ -322,7 +322,7 @@ static int dwc3_omap_remove_core(struct device *dev, void *c)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 
-	platform_device_unregister(pdev);
+	of_device_unregister(pdev);
 
 	return 0;
 }
@@ -599,7 +599,7 @@ static int dwc3_omap_prepare(struct device *dev)
 {
 	struct dwc3_omap	*omap = dev_get_drvdata(dev);
 
-	dwc3_omap_disable_irqs(omap);
+	dwc3_omap_write_irqmisc_set(omap, 0x00);
 
 	return 0;
 }
@@ -607,8 +607,19 @@ static int dwc3_omap_prepare(struct device *dev)
 static void dwc3_omap_complete(struct device *dev)
 {
 	struct dwc3_omap	*omap = dev_get_drvdata(dev);
+	u32			reg;
 
-	dwc3_omap_enable_irqs(omap);
+	reg = (USBOTGSS_IRQMISC_OEVT |
+			USBOTGSS_IRQMISC_DRVVBUS_RISE |
+			USBOTGSS_IRQMISC_CHRGVBUS_RISE |
+			USBOTGSS_IRQMISC_DISCHRGVBUS_RISE |
+			USBOTGSS_IRQMISC_IDPULLUP_RISE |
+			USBOTGSS_IRQMISC_DRVVBUS_FALL |
+			USBOTGSS_IRQMISC_CHRGVBUS_FALL |
+			USBOTGSS_IRQMISC_DISCHRGVBUS_FALL |
+			USBOTGSS_IRQMISC_IDPULLUP_FALL);
+
+	dwc3_omap_write_irqmisc_set(omap, reg);
 }
 
 static int dwc3_omap_suspend(struct device *dev)
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index 9d64dd02c57e..dab7927d1009 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -828,10 +828,6 @@ static void dwc3_prepare_one_trb(struct dwc3_ep *dep,
 			length, last ? " last" : "",
 			chain ? " chain" : "");
 
-	/* Skip the LINK-TRB on ISOC */
-	if (((dep->free_slot & DWC3_TRB_MASK) == DWC3_TRB_NUM - 1) &&
-			usb_endpoint_xfer_isoc(dep->endpoint.desc))
-		dep->free_slot++;
 
 	trb = &dep->trb_pool[dep->free_slot & DWC3_TRB_MASK];
 
@@ -843,6 +839,10 @@ static void dwc3_prepare_one_trb(struct dwc3_ep *dep,
 	}
 
 	dep->free_slot++;
+	/* Skip the LINK-TRB on ISOC */
+	if (((dep->free_slot & DWC3_TRB_MASK) == DWC3_TRB_NUM - 1) &&
+			usb_endpoint_xfer_isoc(dep->endpoint.desc))
+		dep->free_slot++;
 
 	trb->size = DWC3_TRB_SIZE_LENGTH(length);
 	trb->bpl = lower_32_bits(dma);
diff --git a/drivers/usb/gadget/configfs.c b/drivers/usb/gadget/configfs.c
index 2ddcd635ca2a..97142146eead 100644
--- a/drivers/usb/gadget/configfs.c
+++ b/drivers/usb/gadget/configfs.c
@@ -1145,15 +1145,15 @@ static struct configfs_item_operations interf_item_ops = {
 	.store_attribute	= usb_os_desc_attr_store,
 };
 
-static ssize_t rndis_grp_compatible_id_show(struct usb_os_desc *desc,
-					    char *page)
+static ssize_t interf_grp_compatible_id_show(struct usb_os_desc *desc,
+					     char *page)
 {
 	memcpy(page, desc->ext_compat_id, 8);
 	return 8;
 }
 
-static ssize_t rndis_grp_compatible_id_store(struct usb_os_desc *desc,
-					     const char *page, size_t len)
+static ssize_t interf_grp_compatible_id_store(struct usb_os_desc *desc,
+					      const char *page, size_t len)
 {
 	int l;
 
@@ -1171,20 +1171,20 @@ static ssize_t rndis_grp_compatible_id_store(struct usb_os_desc *desc,
 	return len;
 }
 
-static struct usb_os_desc_attribute rndis_grp_attr_compatible_id =
+static struct usb_os_desc_attribute interf_grp_attr_compatible_id =
 	__CONFIGFS_ATTR(compatible_id, S_IRUGO | S_IWUSR,
-			rndis_grp_compatible_id_show,
-			rndis_grp_compatible_id_store);
+			interf_grp_compatible_id_show,
+			interf_grp_compatible_id_store);
 
-static ssize_t rndis_grp_sub_compatible_id_show(struct usb_os_desc *desc,
-						char *page)
+static ssize_t interf_grp_sub_compatible_id_show(struct usb_os_desc *desc,
+						 char *page)
 {
 	memcpy(page, desc->ext_compat_id + 8, 8);
 	return 8;
 }
 
-static ssize_t rndis_grp_sub_compatible_id_store(struct usb_os_desc *desc,
-						 const char *page, size_t len)
+static ssize_t interf_grp_sub_compatible_id_store(struct usb_os_desc *desc,
+						  const char *page, size_t len)
 {
 	int l;
 
@@ -1202,20 +1202,21 @@ static ssize_t rndis_grp_sub_compatible_id_store(struct usb_os_desc *desc,
 	return len;
 }
 
-static struct usb_os_desc_attribute rndis_grp_attr_sub_compatible_id =
+static struct usb_os_desc_attribute interf_grp_attr_sub_compatible_id =
 	__CONFIGFS_ATTR(sub_compatible_id, S_IRUGO | S_IWUSR,
-			rndis_grp_sub_compatible_id_show,
-			rndis_grp_sub_compatible_id_store);
+			interf_grp_sub_compatible_id_show,
+			interf_grp_sub_compatible_id_store);
 
 static struct configfs_attribute *interf_grp_attrs[] = {
-	&rndis_grp_attr_compatible_id.attr,
-	&rndis_grp_attr_sub_compatible_id.attr,
+	&interf_grp_attr_compatible_id.attr,
+	&interf_grp_attr_sub_compatible_id.attr,
 	NULL
 };
 
 int usb_os_desc_prepare_interf_dir(struct config_group *parent,
 				   int n_interf,
 				   struct usb_os_desc **desc,
+				   char **names,
 				   struct module *owner)
 {
 	struct config_group **f_default_groups, *os_desc_group,
@@ -1257,8 +1258,8 @@ int usb_os_desc_prepare_interf_dir(struct config_group *parent,
 		d = desc[n_interf];
 		d->owner = owner;
 		config_group_init_type_name(&d->group, "", interface_type);
-		config_item_set_name(&d->group.cg_item, "interface.%d",
-				     n_interf);
+		config_item_set_name(&d->group.cg_item, "interface.%s",
+				     names[n_interf]);
 		interface_groups[n_interf] = &d->group;
 	}
 
diff --git a/drivers/usb/gadget/configfs.h b/drivers/usb/gadget/configfs.h
index a14ac792c698..36c468c4f5e9 100644
--- a/drivers/usb/gadget/configfs.h
+++ b/drivers/usb/gadget/configfs.h
@@ -8,6 +8,7 @@ void unregister_gadget_item(struct config_item *item);
 int usb_os_desc_prepare_interf_dir(struct config_group *parent,
 				   int n_interf,
 				   struct usb_os_desc **desc,
+				   char **names,
 				   struct module *owner);
 
 static inline struct usb_os_desc *to_usb_os_desc(struct config_item *item)
diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c
index 74202d67f911..8598c27c7d43 100644
--- a/drivers/usb/gadget/f_fs.c
+++ b/drivers/usb/gadget/f_fs.c
@@ -1483,11 +1483,13 @@ static int functionfs_bind(struct ffs_data *ffs, struct usb_composite_dev *cdev)
 	ffs->ep0req->context = ffs;
 
 	lang = ffs->stringtabs;
-	for (lang = ffs->stringtabs; *lang; ++lang) {
-		struct usb_string *str = (*lang)->strings;
-		int id = first_id;
-		for (; str->s; ++id, ++str)
-			str->id = id;
+	if (lang) {
+		for (; *lang; ++lang) {
+			struct usb_string *str = (*lang)->strings;
+			int id = first_id;
+			for (; str->s; ++id, ++str)
+				str->id = id;
+		}
 	}
 
 	ffs->gadget = cdev->gadget;
diff --git a/drivers/usb/gadget/f_rndis.c b/drivers/usb/gadget/f_rndis.c
index eed3ad878047..9c41e9515b8e 100644
--- a/drivers/usb/gadget/f_rndis.c
+++ b/drivers/usb/gadget/f_rndis.c
@@ -687,7 +687,7 @@ rndis_bind(struct usb_configuration *c, struct usb_function *f)
 		f->os_desc_table = kzalloc(sizeof(*f->os_desc_table),
 					   GFP_KERNEL);
 		if (!f->os_desc_table)
-			return PTR_ERR(f->os_desc_table);
+			return -ENOMEM;
 		f->os_desc_n = 1;
 		f->os_desc_table[0].os_desc = &rndis_opts->rndis_os_desc;
 	}
@@ -905,6 +905,7 @@ static struct usb_function_instance *rndis_alloc_inst(void)
 {
 	struct f_rndis_opts *opts;
 	struct usb_os_desc *descs[1];
+	char *names[1];
 
 	opts = kzalloc(sizeof(*opts), GFP_KERNEL);
 	if (!opts)
@@ -922,8 +923,9 @@ static struct usb_function_instance *rndis_alloc_inst(void)
 	INIT_LIST_HEAD(&opts->rndis_os_desc.ext_prop);
 
 	descs[0] = &opts->rndis_os_desc;
+	names[0] = "rndis";
 	usb_os_desc_prepare_interf_dir(&opts->func_inst.group, 1, descs,
-				       THIS_MODULE);
+				       names, THIS_MODULE);
 	config_group_init_type_name(&opts->func_inst.group, "",
 				    &rndis_func_type);
 
diff --git a/drivers/usb/gadget/gr_udc.c b/drivers/usb/gadget/gr_udc.c
index 99a37ed03e27..c7004ee89c90 100644
--- a/drivers/usb/gadget/gr_udc.c
+++ b/drivers/usb/gadget/gr_udc.c
@@ -1532,8 +1532,9 @@ static int gr_ep_enable(struct usb_ep *_ep,
 			"%s mode: multiple trans./microframe not valid\n",
 			(mode == 2 ? "Bulk" : "Control"));
 		return -EINVAL;
-	} else if (nt == 0x11) {
-		dev_err(dev->dev, "Invalid value for trans./microframe\n");
+	} else if (nt == 0x3) {
+		dev_err(dev->dev,
+			"Invalid value 0x3 for additional trans./microframe\n");
 		return -EINVAL;
 	} else if ((nt + 1) * max > buffer_size) {
 		dev_err(dev->dev, "Hw buffer size %d < max payload %d * %d\n",
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index ee6c16416c30..2e4ce7704908 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1264,8 +1264,13 @@ dev_release (struct inode *inode, struct file *fd)
 
 	kfree (dev->buf);
 	dev->buf = NULL;
-	put_dev (dev);
 
+	/* other endpoints were all decoupled from this device */
+	spin_lock_irq(&dev->lock);
+	dev->state = STATE_DEV_DISABLED;
+	spin_unlock_irq(&dev->lock);
+
+	put_dev (dev);
 	return 0;
 }
 
diff --git a/drivers/usb/gadget/storage_common.c b/drivers/usb/gadget/storage_common.c
index ff205a7bc55c..648f9e489b39 100644
--- a/drivers/usb/gadget/storage_common.c
+++ b/drivers/usb/gadget/storage_common.c
@@ -220,11 +220,11 @@ int fsg_lun_open(struct fsg_lun *curlun, const char *filename)
 	 * If we can't read the file, it's no good.
 	 * If we can't write the file, use it read-only.
 	 */
-	if (!(filp->f_op->read || filp->f_op->aio_read)) {
+	if (!(filp->f_mode & FMODE_CAN_READ)) {
 		LINFO(curlun, "file not readable: %s\n", filename);
 		goto out;
 	}
-	if (!(filp->f_op->write || filp->f_op->aio_write))
+	if (!(filp->f_mode & FMODE_CAN_WRITE))
 		ro = 1;
 
 	size = i_size_read(inode->i_mapping->host);
diff --git a/drivers/usb/gadget/u_ether.c b/drivers/usb/gadget/u_ether.c
index fe0880d0873e..97b027724ee7 100644
--- a/drivers/usb/gadget/u_ether.c
+++ b/drivers/usb/gadget/u_ether.c
@@ -793,7 +793,7 @@ struct eth_dev *gether_setup_name(struct usb_gadget *g,
 
 	net->netdev_ops = &eth_netdev_ops;
 
-	SET_ETHTOOL_OPS(net, &ops);
+	net->ethtool_ops = &ops;
 
 	dev->gadget = g;
 	SET_NETDEV_DEV(net, &g->dev);
@@ -850,7 +850,7 @@ struct net_device *gether_setup_name_default(const char *netname)
 
 	net->netdev_ops = &eth_netdev_ops;
 
-	SET_ETHTOOL_OPS(net, &ops);
+	net->ethtool_ops = &ops;
 	SET_NETDEV_DEVTYPE(net, &gadget_type);
 
 	return net;
@@ -1120,7 +1120,10 @@ void gether_disconnect(struct gether *link)
 
 	DBG(dev, "%s\n", __func__);
 
+	netif_tx_lock(dev->net);
 	netif_stop_queue(dev->net);
+	netif_tx_unlock(dev->net);
+
 	netif_carrier_off(dev->net);
 
 	/* disable endpoints, forcing (synchronous) completion
diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index 61b7817bd66b..03314f861bee 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -176,7 +176,7 @@ config USB_EHCI_HCD_AT91
 
 config USB_EHCI_MSM
 	tristate "Support for Qualcomm QSD/MSM on-chip EHCI USB controller"
-	depends on ARCH_MSM
+	depends on ARCH_MSM || ARCH_QCOM
 	select USB_EHCI_ROOT_HUB_TT
 	---help---
 	  Enables support for the USB Host controller present on the
diff --git a/drivers/usb/host/ehci-pmcmsp.c b/drivers/usb/host/ehci-pmcmsp.c
index af3974a5e7c2..7d75465d97c7 100644
--- a/drivers/usb/host/ehci-pmcmsp.c
+++ b/drivers/usb/host/ehci-pmcmsp.c
@@ -68,9 +68,6 @@ static void usb_hcd_tdi_set_mode(struct ehci_hcd *ehci)
 
 	/* set TWI GPIO USB_HOST_DEV pin high */
 	gpio_direction_output(MSP_PIN_USB0_HOST_DEV, 1);
-#ifdef CONFIG_MSP_HAS_DUAL_USB
-	gpio_direction_output(MSP_PIN_USB1_HOST_DEV, 1);
-#endif
 }
 
 /* called during probe() after chip reset completes */
@@ -248,33 +245,6 @@ void usb_hcd_msp_remove(struct usb_hcd *hcd, struct platform_device *dev)
 	usb_put_hcd(hcd);
 }
 
-#ifdef CONFIG_MSP_HAS_DUAL_USB
-/*
- * Wrapper around the main ehci_irq.  Since both USB host controllers are
- * sharing the same IRQ, need to first determine whether we're the intended
- * recipient of this interrupt.
- */
-static irqreturn_t ehci_msp_irq(struct usb_hcd *hcd)
-{
-	u32 int_src;
-	struct device *dev = hcd->self.controller;
-	struct platform_device *pdev;
-	struct mspusb_device *mdev;
-	struct ehci_hcd	*ehci = hcd_to_ehci(hcd);
-	/* need to reverse-map a couple of containers to get our device */
-	pdev = to_platform_device(dev);
-	mdev = to_mspusb_device(pdev);
-
-	/* Check to see if this interrupt is for this host controller */
-	int_src = ehci_readl(ehci, &mdev->mab_regs->int_stat);
-	if (int_src & (1 << pdev->id))
-		return ehci_irq(hcd);
-
-	/* Not for this device */
-	return IRQ_NONE;
-}
-#endif /* DUAL_USB */
-
 static const struct hc_driver ehci_msp_hc_driver = {
 	.description =		hcd_name,
 	.product_desc =		"PMC MSP EHCI",
@@ -283,11 +253,7 @@ static const struct hc_driver ehci_msp_hc_driver = {
 	/*
 	 * generic hardware linkage
 	 */
-#ifdef CONFIG_MSP_HAS_DUAL_USB
-	.irq =			ehci_msp_irq,
-#else
 	.irq =			ehci_irq,
-#endif
 	.flags =		HCD_MEMORY | HCD_USB2 | HCD_BH,
 
 	/*
@@ -334,9 +300,6 @@ static int ehci_hcd_msp_drv_probe(struct platform_device *pdev)
 		return -ENODEV;
 
 	gpio_request(MSP_PIN_USB0_HOST_DEV, "USB0_HOST_DEV_GPIO");
-#ifdef CONFIG_MSP_HAS_DUAL_USB
-	gpio_request(MSP_PIN_USB1_HOST_DEV, "USB1_HOST_DEV_GPIO");
-#endif
 
 	ret = usb_hcd_msp_probe(&ehci_msp_hc_driver, pdev);
 
@@ -351,9 +314,6 @@ static int ehci_hcd_msp_drv_remove(struct platform_device *pdev)
 
 	/* free TWI GPIO USB_HOST_DEV pin */
 	gpio_free(MSP_PIN_USB0_HOST_DEV);
-#ifdef CONFIG_MSP_HAS_DUAL_USB
-	gpio_free(MSP_PIN_USB1_HOST_DEV);
-#endif
 
 	return 0;
 }
diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c
index 4a6d3dd68572..2f3acebb577a 100644
--- a/drivers/usb/host/pci-quirks.c
+++ b/drivers/usb/host/pci-quirks.c
@@ -656,6 +656,14 @@ static const struct dmi_system_id ehci_dmi_nohandoff_table[] = {
 			DMI_MATCH(DMI_BIOS_VERSION, "Lucid-"),
 		},
 	},
+	{
+		/* HASEE E200 */
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "HASEE"),
+			DMI_MATCH(DMI_BOARD_NAME, "E210"),
+			DMI_MATCH(DMI_BIOS_VERSION, "6.00"),
+		},
+	},
 	{ }
 };
 
@@ -665,9 +673,14 @@ static void ehci_bios_handoff(struct pci_dev *pdev,
 {
 	int try_handoff = 1, tried_handoff = 0;
 
-	/* The Pegatron Lucid tablet sporadically waits for 98 seconds trying
-	 * the handoff on its unused controller.  Skip it. */
-	if (pdev->vendor == 0x8086 && pdev->device == 0x283a) {
+	/*
+	 * The Pegatron Lucid tablet sporadically waits for 98 seconds trying
+	 * the handoff on its unused controller.  Skip it.
+	 *
+	 * The HASEE E200 hangs when the semaphore is set (bugzilla #77021).
+	 */
+	if (pdev->vendor == 0x8086 && (pdev->device == 0x283a ||
+			pdev->device == 0x27cc)) {
 		if (dmi_check_system(ehci_dmi_nohandoff_table))
 			try_handoff = 0;
 	}
diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c
index 6231ce6aa0c3..aa79e8749040 100644
--- a/drivers/usb/host/xhci-hub.c
+++ b/drivers/usb/host/xhci-hub.c
@@ -22,6 +22,7 @@
 
 
 #include <linux/slab.h>
+#include <linux/device.h>
 #include <asm/unaligned.h>
 
 #include "xhci.h"
@@ -287,7 +288,7 @@ static int xhci_stop_device(struct xhci_hcd *xhci, int slot_id, int suspend)
 		if (virt_dev->eps[i].ring && virt_dev->eps[i].ring->dequeue) {
 			struct xhci_command *command;
 			command = xhci_alloc_command(xhci, false, false,
-						     GFP_NOIO);
+						     GFP_NOWAIT);
 			if (!command) {
 				spin_unlock_irqrestore(&xhci->lock, flags);
 				xhci_free_command(xhci, cmd);
@@ -1139,7 +1140,9 @@ int xhci_bus_suspend(struct usb_hcd *hcd)
 		 * including the USB 3.0 roothub, but only if CONFIG_PM_RUNTIME
 		 * is enabled, so also enable remote wake here.
 		 */
-		if (hcd->self.root_hub->do_remote_wakeup) {
+		if (hcd->self.root_hub->do_remote_wakeup
+				&& device_may_wakeup(hcd->self.controller)) {
+
 			if (t1 & PORT_CONNECT) {
 				t2 |= PORT_WKOC_E | PORT_WKDISC_E;
 				t2 &= ~PORT_WKCONN_E;
diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c
index d67ff71209f5..749fc68eb5c1 100644
--- a/drivers/usb/host/xhci-ring.c
+++ b/drivers/usb/host/xhci-ring.c
@@ -1433,8 +1433,11 @@ static void handle_cmd_completion(struct xhci_hcd *xhci,
 		xhci_handle_cmd_reset_ep(xhci, slot_id, cmd_trb, cmd_comp_code);
 		break;
 	case TRB_RESET_DEV:
-		WARN_ON(slot_id != TRB_TO_SLOT_ID(
-				le32_to_cpu(cmd_trb->generic.field[3])));
+		/* SLOT_ID field in reset device cmd completion event TRB is 0.
+		 * Use the SLOT_ID from the command TRB instead (xhci 4.6.11)
+		 */
+		slot_id = TRB_TO_SLOT_ID(
+				le32_to_cpu(cmd_trb->generic.field[3]));
 		xhci_handle_cmd_reset_dev(xhci, slot_id, event);
 		break;
 	case TRB_NEC_GET_FW:
@@ -3534,7 +3537,7 @@ static unsigned int xhci_get_burst_count(struct xhci_hcd *xhci,
 		return 0;
 
 	max_burst = urb->ep->ss_ep_comp.bMaxBurst;
-	return roundup(total_packet_count, max_burst + 1) - 1;
+	return DIV_ROUND_UP(total_packet_count, max_burst + 1) - 1;
 }
 
 /*
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 2b8d9a24af09..7436d5f5e67a 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -936,7 +936,7 @@ int xhci_suspend(struct xhci_hcd *xhci)
  */
 int xhci_resume(struct xhci_hcd *xhci, bool hibernated)
 {
-	u32			command, temp = 0;
+	u32			command, temp = 0, status;
 	struct usb_hcd		*hcd = xhci_to_hcd(xhci);
 	struct usb_hcd		*secondary_hcd;
 	int			retval = 0;
@@ -1054,8 +1054,12 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated)
 
  done:
 	if (retval == 0) {
-		usb_hcd_resume_root_hub(hcd);
-		usb_hcd_resume_root_hub(xhci->shared_hcd);
+		/* Resume root hubs only when have pending events. */
+		status = readl(&xhci->op_regs->status);
+		if (status & STS_EINT) {
+			usb_hcd_resume_root_hub(hcd);
+			usb_hcd_resume_root_hub(xhci->shared_hcd);
+		}
 	}
 
 	/*
diff --git a/drivers/usb/misc/usbtest.c b/drivers/usb/misc/usbtest.c
index 51a6da256772..829f446064ea 100644
--- a/drivers/usb/misc/usbtest.c
+++ b/drivers/usb/misc/usbtest.c
@@ -7,7 +7,7 @@
 #include <linux/moduleparam.h>
 #include <linux/scatterlist.h>
 #include <linux/mutex.h>
-
+#include <linux/timer.h>
 #include <linux/usb.h>
 
 #define SIMPLE_IO_TIMEOUT	10000	/* in milliseconds */
@@ -484,6 +484,14 @@ alloc_sglist(int nents, int max, int vary)
 	return sg;
 }
 
+static void sg_timeout(unsigned long _req)
+{
+	struct usb_sg_request	*req = (struct usb_sg_request *) _req;
+
+	req->status = -ETIMEDOUT;
+	usb_sg_cancel(req);
+}
+
 static int perform_sglist(
 	struct usbtest_dev	*tdev,
 	unsigned		iterations,
@@ -495,6 +503,9 @@ static int perform_sglist(
 {
 	struct usb_device	*udev = testdev_to_usbdev(tdev);
 	int			retval = 0;
+	struct timer_list	sg_timer;
+
+	setup_timer_on_stack(&sg_timer, sg_timeout, (unsigned long) req);
 
 	while (retval == 0 && iterations-- > 0) {
 		retval = usb_sg_init(req, udev, pipe,
@@ -505,7 +516,10 @@ static int perform_sglist(
 
 		if (retval)
 			break;
+		mod_timer(&sg_timer, jiffies +
+				msecs_to_jiffies(SIMPLE_IO_TIMEOUT));
 		usb_sg_wait(req);
+		del_timer_sync(&sg_timer);
 		retval = req->status;
 
 		/* FIXME check resulting data pattern */
diff --git a/drivers/usb/musb/musb_am335x.c b/drivers/usb/musb/musb_am335x.c
index d2353781bd2d..1e58ed2361cc 100644
--- a/drivers/usb/musb/musb_am335x.c
+++ b/drivers/usb/musb/musb_am335x.c
@@ -19,21 +19,6 @@ err:
 	return ret;
 }
 
-static int of_remove_populated_child(struct device *dev, void *d)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-
-	of_device_unregister(pdev);
-	return 0;
-}
-
-static int am335x_child_remove(struct platform_device *pdev)
-{
-	device_for_each_child(&pdev->dev, NULL, of_remove_populated_child);
-	pm_runtime_disable(&pdev->dev);
-	return 0;
-}
-
 static const struct of_device_id am335x_child_of_match[] = {
 	{ .compatible = "ti,am33xx-usb" },
 	{  },
@@ -42,13 +27,17 @@ MODULE_DEVICE_TABLE(of, am335x_child_of_match);
 
 static struct platform_driver am335x_child_driver = {
 	.probe		= am335x_child_probe,
-	.remove         = am335x_child_remove,
 	.driver         = {
 		.name   = "am335x-usb-childs",
 		.of_match_table	= am335x_child_of_match,
 	},
 };
 
-module_platform_driver(am335x_child_driver);
+static int __init am335x_child_init(void)
+{
+	return platform_driver_register(&am335x_child_driver);
+}
+module_init(am335x_child_init);
+
 MODULE_DESCRIPTION("AM33xx child devices");
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c
index 61da471b7aed..eff3c5cf84f4 100644
--- a/drivers/usb/musb/musb_core.c
+++ b/drivers/usb/musb/musb_core.c
@@ -849,7 +849,7 @@ b_host:
 	}
 
 	/* handle babble condition */
-	if (int_usb & MUSB_INTR_BABBLE)
+	if (int_usb & MUSB_INTR_BABBLE && is_host_active(musb))
 		schedule_work(&musb->recover_work);
 
 #if 0
diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c
index 7b8bbf53127e..5341bb223b7c 100644
--- a/drivers/usb/musb/musb_cppi41.c
+++ b/drivers/usb/musb/musb_cppi41.c
@@ -318,7 +318,7 @@ static void cppi41_dma_callback(void *private_data)
 		}
 		list_add_tail(&cppi41_channel->tx_check,
 				&controller->early_tx_list);
-		if (!hrtimer_active(&controller->early_tx)) {
+		if (!hrtimer_is_queued(&controller->early_tx)) {
 			hrtimer_start_range_ns(&controller->early_tx,
 				ktime_set(0, 140 * NSEC_PER_USEC),
 				40 * NSEC_PER_USEC,
diff --git a/drivers/usb/musb/musb_dsps.c b/drivers/usb/musb/musb_dsps.c
index 51beb13c7e1a..09529f94e72d 100644
--- a/drivers/usb/musb/musb_dsps.c
+++ b/drivers/usb/musb/musb_dsps.c
@@ -494,10 +494,9 @@ static int dsps_musb_set_mode(struct musb *musb, u8 mode)
 	struct dsps_glue *glue = dev_get_drvdata(dev->parent);
 	const struct dsps_musb_wrapper *wrp = glue->wrp;
 	void __iomem *ctrl_base = musb->ctrl_base;
-	void __iomem *base = musb->mregs;
 	u32 reg;
 
-	reg = dsps_readl(base, wrp->mode);
+	reg = dsps_readl(ctrl_base, wrp->mode);
 
 	switch (mode) {
 	case MUSB_HOST:
@@ -510,7 +509,7 @@ static int dsps_musb_set_mode(struct musb *musb, u8 mode)
 		 */
 		reg |= (1 << wrp->iddig_mux);
 
-		dsps_writel(base, wrp->mode, reg);
+		dsps_writel(ctrl_base, wrp->mode, reg);
 		dsps_writel(ctrl_base, wrp->phy_utmi, 0x02);
 		break;
 	case MUSB_PERIPHERAL:
@@ -523,10 +522,10 @@ static int dsps_musb_set_mode(struct musb *musb, u8 mode)
 		 */
 		reg |= (1 << wrp->iddig_mux);
 
-		dsps_writel(base, wrp->mode, reg);
+		dsps_writel(ctrl_base, wrp->mode, reg);
 		break;
 	case MUSB_OTG:
-		dsps_writel(base, wrp->phy_utmi, 0x02);
+		dsps_writel(ctrl_base, wrp->phy_utmi, 0x02);
 		break;
 	default:
 		dev_err(glue->dev, "unsupported mode %d\n", mode);
diff --git a/drivers/usb/musb/ux500.c b/drivers/usb/musb/ux500.c
index c2e45e632723..f202e5088461 100644
--- a/drivers/usb/musb/ux500.c
+++ b/drivers/usb/musb/ux500.c
@@ -274,7 +274,6 @@ static int ux500_probe(struct platform_device *pdev)
 	musb->dev.parent		= &pdev->dev;
 	musb->dev.dma_mask		= &pdev->dev.coherent_dma_mask;
 	musb->dev.coherent_dma_mask	= pdev->dev.coherent_dma_mask;
-	musb->dev.of_node		= pdev->dev.of_node;
 
 	glue->dev			= &pdev->dev;
 	glue->musb			= musb;
diff --git a/drivers/usb/phy/phy-msm-usb.c b/drivers/usb/phy/phy-msm-usb.c
index ced34f39bdd4..c929370cdaa6 100644
--- a/drivers/usb/phy/phy-msm-usb.c
+++ b/drivers/usb/phy/phy-msm-usb.c
@@ -1229,7 +1229,9 @@ static void msm_otg_sm_work(struct work_struct *w)
 			motg->chg_state = USB_CHG_STATE_UNDEFINED;
 			motg->chg_type = USB_INVALID_CHARGER;
 		}
-		pm_runtime_put_sync(otg->phy->dev);
+
+		if (otg->phy->state == OTG_STATE_B_IDLE)
+			pm_runtime_put_sync(otg->phy->dev);
 		break;
 	case OTG_STATE_B_PERIPHERAL:
 		dev_dbg(otg->phy->dev, "OTG_STATE_B_PERIPHERAL state\n");
diff --git a/drivers/usb/renesas_usbhs/fifo.c b/drivers/usb/renesas_usbhs/fifo.c
index d49f9c326035..4fd36530bfa3 100644
--- a/drivers/usb/renesas_usbhs/fifo.c
+++ b/drivers/usb/renesas_usbhs/fifo.c
@@ -681,6 +681,14 @@ usbhs_fifo_read_end:
 		usbhs_pipe_number(pipe),
 		pkt->length, pkt->actual, *is_done, pkt->zero);
 
+	/*
+	 * Transmission end
+	 */
+	if (*is_done) {
+		if (usbhs_pipe_is_dcp(pipe))
+			usbhs_dcp_control_transfer_done(pipe);
+	}
+
 usbhs_fifo_read_busy:
 	usbhsf_fifo_unselect(pipe, fifo);
 
diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c
index 762e4a5f5ae9..330df5ce435b 100644
--- a/drivers/usb/serial/cp210x.c
+++ b/drivers/usb/serial/cp210x.c
@@ -153,6 +153,7 @@ static const struct usb_device_id id_table[] = {
 	{ USB_DEVICE(0x1843, 0x0200) }, /* Vaisala USB Instrument Cable */
 	{ USB_DEVICE(0x18EF, 0xE00F) }, /* ELV USB-I2C-Interface */
 	{ USB_DEVICE(0x1ADB, 0x0001) }, /* Schweitzer Engineering C662 Cable */
+	{ USB_DEVICE(0x1B1C, 0x1C00) }, /* Corsair USB Dongle */
 	{ USB_DEVICE(0x1BE3, 0x07A6) }, /* WAGO 750-923 USB Service Cable */
 	{ USB_DEVICE(0x1E29, 0x0102) }, /* Festo CPX-USB */
 	{ USB_DEVICE(0x1E29, 0x0501) }, /* Festo CMSP */
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index edf3b124583c..8a3813be1b28 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -720,7 +720,8 @@ static const struct usb_device_id id_table_combined[] = {
 	{ USB_DEVICE(FTDI_VID, FTDI_ACG_HFDUAL_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_YEI_SERVOCENTER31_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_THORLABS_PID) },
-	{ USB_DEVICE(TESTO_VID, TESTO_USB_INTERFACE_PID) },
+	{ USB_DEVICE(TESTO_VID, TESTO_1_PID) },
+	{ USB_DEVICE(TESTO_VID, TESTO_3_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_GAMMA_SCOUT_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_TACTRIX_OPENPORT_13M_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_TACTRIX_OPENPORT_13S_PID) },
@@ -944,6 +945,8 @@ static const struct usb_device_id id_table_combined[] = {
 	{ USB_DEVICE(BRAINBOXES_VID, BRAINBOXES_US_842_2_PID) },
 	{ USB_DEVICE(BRAINBOXES_VID, BRAINBOXES_US_842_3_PID) },
 	{ USB_DEVICE(BRAINBOXES_VID, BRAINBOXES_US_842_4_PID) },
+	/* Infineon Devices */
+	{ USB_DEVICE_INTERFACE_NUMBER(INFINEON_VID, INFINEON_TRIBOARD_PID, 1) },
 	{ }					/* Terminating entry */
 };
 
@@ -1566,14 +1569,17 @@ static void ftdi_set_max_packet_size(struct usb_serial_port *port)
 	struct usb_device *udev = serial->dev;
 
 	struct usb_interface *interface = serial->interface;
-	struct usb_endpoint_descriptor *ep_desc = &interface->cur_altsetting->endpoint[1].desc;
+	struct usb_endpoint_descriptor *ep_desc;
 
 	unsigned num_endpoints;
-	int i;
+	unsigned i;
 
 	num_endpoints = interface->cur_altsetting->desc.bNumEndpoints;
 	dev_info(&udev->dev, "Number of endpoints %d\n", num_endpoints);
 
+	if (!num_endpoints)
+		return;
+
 	/* NOTE: some customers have programmed FT232R/FT245R devices
 	 * with an endpoint size of 0 - not good.  In this case, we
 	 * want to override the endpoint descriptor setting and use a
diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h
index 500474c48f4b..c4777bc6aee0 100644
--- a/drivers/usb/serial/ftdi_sio_ids.h
+++ b/drivers/usb/serial/ftdi_sio_ids.h
@@ -584,6 +584,12 @@
 #define RATOC_PRODUCT_ID_USB60F	0xb020
 
 /*
+ * Infineon Technologies
+ */
+#define INFINEON_VID		0x058b
+#define INFINEON_TRIBOARD_PID	0x0028 /* DAS JTAG TriBoard TC1798 V1.0 */
+
+/*
  * Acton Research Corp.
  */
 #define ACTON_VID		0x0647	/* Vendor ID */
@@ -798,7 +804,8 @@
  * Submitted by Colin Leroy
  */
 #define TESTO_VID			0x128D
-#define TESTO_USB_INTERFACE_PID		0x0001
+#define TESTO_1_PID			0x0001
+#define TESTO_3_PID			0x0003
 
 /*
  * Mobility Electronics products.
diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
index 59c3108cc136..a9688940543d 100644
--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -352,6 +352,9 @@ static void option_instat_callback(struct urb *urb);
 /* Zoom */
 #define ZOOM_PRODUCT_4597			0x9607
 
+/* SpeedUp SU9800 usb 3g modem */
+#define SPEEDUP_PRODUCT_SU9800			0x9800
+
 /* Haier products */
 #define HAIER_VENDOR_ID				0x201e
 #define HAIER_PRODUCT_CE100			0x2009
@@ -372,8 +375,12 @@ static void option_instat_callback(struct urb *urb);
 /* Olivetti products */
 #define OLIVETTI_VENDOR_ID			0x0b3c
 #define OLIVETTI_PRODUCT_OLICARD100		0xc000
+#define OLIVETTI_PRODUCT_OLICARD120		0xc001
+#define OLIVETTI_PRODUCT_OLICARD140		0xc002
 #define OLIVETTI_PRODUCT_OLICARD145		0xc003
+#define OLIVETTI_PRODUCT_OLICARD155		0xc004
 #define OLIVETTI_PRODUCT_OLICARD200		0xc005
+#define OLIVETTI_PRODUCT_OLICARD160		0xc00a
 #define OLIVETTI_PRODUCT_OLICARD500		0xc00b
 
 /* Celot products */
@@ -1480,6 +1487,8 @@ static const struct usb_device_id option_ids[] = {
 		.driver_info = (kernel_ulong_t)&net_intf2_blacklist },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1426, 0xff, 0xff, 0xff),  /* ZTE MF91 */
 		.driver_info = (kernel_ulong_t)&net_intf2_blacklist },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1428, 0xff, 0xff, 0xff),  /* Telewell TW-LTE 4G v2 */
+		.driver_info = (kernel_ulong_t)&net_intf2_blacklist },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1533, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1534, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0x1535, 0xff, 0xff, 0xff) },
@@ -1577,6 +1586,7 @@ static const struct usb_device_id option_ids[] = {
 	{ USB_DEVICE(LONGCHEER_VENDOR_ID, FOUR_G_SYSTEMS_PRODUCT_W14),
   	  .driver_info = (kernel_ulong_t)&four_g_w14_blacklist
   	},
+	{ USB_DEVICE_INTERFACE_CLASS(LONGCHEER_VENDOR_ID, SPEEDUP_PRODUCT_SU9800, 0xff) },
 	{ USB_DEVICE(LONGCHEER_VENDOR_ID, ZOOM_PRODUCT_4597) },
 	{ USB_DEVICE(LONGCHEER_VENDOR_ID, IBALL_3_5G_CONNECT) },
 	{ USB_DEVICE(HAIER_VENDOR_ID, HAIER_PRODUCT_CE100) },
@@ -1611,15 +1621,21 @@ static const struct usb_device_id option_ids[] = {
 	{ USB_DEVICE(SIEMENS_VENDOR_ID, CINTERION_PRODUCT_HC25_MDMNET) },
 	{ USB_DEVICE(SIEMENS_VENDOR_ID, CINTERION_PRODUCT_HC28_MDM) }, /* HC28 enumerates with Siemens or Cinterion VID depending on FW revision */
 	{ USB_DEVICE(SIEMENS_VENDOR_ID, CINTERION_PRODUCT_HC28_MDMNET) },
-
-	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD100) },
+	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD100),
+		.driver_info = (kernel_ulong_t)&net_intf4_blacklist },
+	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD120),
+		.driver_info = (kernel_ulong_t)&net_intf4_blacklist },
+	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD140),
+		.driver_info = (kernel_ulong_t)&net_intf4_blacklist },
 	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD145) },
+	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD155),
+		.driver_info = (kernel_ulong_t)&net_intf6_blacklist },
 	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD200),
-		.driver_info = (kernel_ulong_t)&net_intf6_blacklist
-	},
+		.driver_info = (kernel_ulong_t)&net_intf6_blacklist },
+	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD160),
+		.driver_info = (kernel_ulong_t)&net_intf6_blacklist },
 	{ USB_DEVICE(OLIVETTI_VENDOR_ID, OLIVETTI_PRODUCT_OLICARD500),
-		.driver_info = (kernel_ulong_t)&net_intf4_blacklist
-	},
+		.driver_info = (kernel_ulong_t)&net_intf4_blacklist },
 	{ USB_DEVICE(CELOT_VENDOR_ID, CELOT_PRODUCT_CT680M) }, /* CT-650 CDMA 450 1xEVDO modem */
 	{ USB_DEVICE_AND_INTERFACE_INFO(SAMSUNG_VENDOR_ID, SAMSUNG_PRODUCT_GT_B3730, USB_CLASS_CDC_DATA, 0x00, 0x00) }, /* Samsung GT-B3730 LTE USB modem.*/
 	{ USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CEM600) },
diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index 9d38ddc8da49..866b5df36ed1 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -256,6 +256,10 @@ static int slave_configure(struct scsi_device *sdev)
 		if (us->fflags & US_FL_WRITE_CACHE)
 			sdev->wce_default_on = 1;
 
+		/* A few buggy USB-ATA bridges don't understand FUA */
+		if (us->fflags & US_FL_BROKEN_FUA)
+			sdev->broken_fua = 1;
+
 	} else {
 
 		/* Non-disk-type devices don't need to blacklist any pages
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 174a447868cd..80a5b366255f 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1936,6 +1936,13 @@ UNUSUAL_DEV(  0x14cd, 0x6600, 0x0201, 0x0201,
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
+/* Reported by Michael Büsch <m@bues.ch> */
+UNUSUAL_DEV(  0x152d, 0x0567, 0x0114, 0x0114,
+		"JMicron",
+		"USB to ATA/ATAPI Bridge",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_BROKEN_FUA ),
+
 /* Reported by Alexandre Oliva <oliva@lsd.ic.unicamp.br>
  * JMicron responds to USN and several other SCSI ioctls with a
  * residue that causes subsequent I/O requests to fail.  */
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index be414d2b2b22..8dae2f724a35 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -17,6 +17,7 @@
 #include <linux/workqueue.h>
 #include <linux/file.h>
 #include <linux/slab.h>
+#include <linux/vmalloc.h>
 
 #include <linux/net.h>
 #include <linux/if_packet.h>
@@ -373,7 +374,7 @@ static void handle_tx(struct vhost_net *net)
 			      % UIO_MAXIOV == nvq->done_idx))
 			break;
 
-		head = vhost_get_vq_desc(&net->dev, vq, vq->iov,
+		head = vhost_get_vq_desc(vq, vq->iov,
 					 ARRAY_SIZE(vq->iov),
 					 &out, &in,
 					 NULL, NULL);
@@ -505,7 +506,7 @@ static int get_rx_bufs(struct vhost_virtqueue *vq,
 			r = -ENOBUFS;
 			goto err;
 		}
-		r = vhost_get_vq_desc(vq->dev, vq, vq->iov + seg,
+		r = vhost_get_vq_desc(vq, vq->iov + seg,
 				      ARRAY_SIZE(vq->iov) - seg, &out,
 				      &in, log, log_num);
 		if (unlikely(r < 0))
@@ -584,9 +585,9 @@ static void handle_rx(struct vhost_net *net)
 	vhost_hlen = nvq->vhost_hlen;
 	sock_hlen = nvq->sock_hlen;
 
-	vq_log = unlikely(vhost_has_feature(&net->dev, VHOST_F_LOG_ALL)) ?
+	vq_log = unlikely(vhost_has_feature(vq, VHOST_F_LOG_ALL)) ?
 		vq->log : NULL;
-	mergeable = vhost_has_feature(&net->dev, VIRTIO_NET_F_MRG_RXBUF);
+	mergeable = vhost_has_feature(vq, VIRTIO_NET_F_MRG_RXBUF);
 
 	while ((sock_len = peek_head_len(sock->sk))) {
 		sock_len += sock_hlen;
@@ -701,16 +702,20 @@ static void handle_rx_net(struct vhost_work *work)
 
 static int vhost_net_open(struct inode *inode, struct file *f)
 {
-	struct vhost_net *n = kmalloc(sizeof *n, GFP_KERNEL);
+	struct vhost_net *n;
 	struct vhost_dev *dev;
 	struct vhost_virtqueue **vqs;
 	int i;
 
-	if (!n)
-		return -ENOMEM;
+	n = kmalloc(sizeof *n, GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
+	if (!n) {
+		n = vmalloc(sizeof *n);
+		if (!n)
+			return -ENOMEM;
+	}
 	vqs = kmalloc(VHOST_NET_VQ_MAX * sizeof(*vqs), GFP_KERNEL);
 	if (!vqs) {
-		kfree(n);
+		kvfree(n);
 		return -ENOMEM;
 	}
 
@@ -827,7 +832,7 @@ static int vhost_net_release(struct inode *inode, struct file *f)
 	 * since jobs can re-queue themselves. */
 	vhost_net_flush(n);
 	kfree(n->dev.vqs);
-	kfree(n);
+	kvfree(n);
 	return 0;
 }
 
@@ -1038,15 +1043,13 @@ static int vhost_net_set_features(struct vhost_net *n, u64 features)
 		mutex_unlock(&n->dev.mutex);
 		return -EFAULT;
 	}
-	n->dev.acked_features = features;
-	smp_wmb();
 	for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
 		mutex_lock(&n->vqs[i].vq.mutex);
+		n->vqs[i].vq.acked_features = features;
 		n->vqs[i].vhost_hlen = vhost_hlen;
 		n->vqs[i].sock_hlen = sock_hlen;
 		mutex_unlock(&n->vqs[i].vq.mutex);
 	}
-	vhost_net_flush(n);
 	mutex_unlock(&n->dev.mutex);
 	return 0;
 }
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index aeb513108448..69906cacd04f 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -57,7 +57,8 @@
 #define TCM_VHOST_MAX_CDB_SIZE 32
 #define TCM_VHOST_DEFAULT_TAGS 256
 #define TCM_VHOST_PREALLOC_SGLS 2048
-#define TCM_VHOST_PREALLOC_PAGES 2048
+#define TCM_VHOST_PREALLOC_UPAGES 2048
+#define TCM_VHOST_PREALLOC_PROT_SGLS 512
 
 struct vhost_scsi_inflight {
 	/* Wait for the flush operation to finish */
@@ -79,10 +80,12 @@ struct tcm_vhost_cmd {
 	u64 tvc_tag;
 	/* The number of scatterlists associated with this cmd */
 	u32 tvc_sgl_count;
+	u32 tvc_prot_sgl_count;
 	/* Saved unpacked SCSI LUN for tcm_vhost_submission_work() */
 	u32 tvc_lun;
 	/* Pointer to the SGL formatted memory from virtio-scsi */
 	struct scatterlist *tvc_sgl;
+	struct scatterlist *tvc_prot_sgl;
 	struct page **tvc_upages;
 	/* Pointer to response */
 	struct virtio_scsi_cmd_resp __user *tvc_resp;
@@ -166,7 +169,8 @@ enum {
 };
 
 enum {
-	VHOST_SCSI_FEATURES = VHOST_FEATURES | (1ULL << VIRTIO_SCSI_F_HOTPLUG)
+	VHOST_SCSI_FEATURES = VHOST_FEATURES | (1ULL << VIRTIO_SCSI_F_HOTPLUG) |
+					       (1ULL << VIRTIO_SCSI_F_T10_PI)
 };
 
 #define VHOST_SCSI_MAX_TARGET	256
@@ -456,12 +460,16 @@ static void tcm_vhost_release_cmd(struct se_cmd *se_cmd)
 	struct tcm_vhost_cmd *tv_cmd = container_of(se_cmd,
 				struct tcm_vhost_cmd, tvc_se_cmd);
 	struct se_session *se_sess = se_cmd->se_sess;
+	int i;
 
 	if (tv_cmd->tvc_sgl_count) {
-		u32 i;
 		for (i = 0; i < tv_cmd->tvc_sgl_count; i++)
 			put_page(sg_page(&tv_cmd->tvc_sgl[i]));
 	}
+	if (tv_cmd->tvc_prot_sgl_count) {
+		for (i = 0; i < tv_cmd->tvc_prot_sgl_count; i++)
+			put_page(sg_page(&tv_cmd->tvc_prot_sgl[i]));
+	}
 
 	tcm_vhost_put_inflight(tv_cmd->inflight);
 	percpu_ida_free(&se_sess->sess_tag_pool, se_cmd->map_tag);
@@ -606,7 +614,7 @@ tcm_vhost_do_evt_work(struct vhost_scsi *vs, struct tcm_vhost_evt *evt)
 
 again:
 	vhost_disable_notify(&vs->dev, vq);
-	head = vhost_get_vq_desc(&vs->dev, vq, vq->iov,
+	head = vhost_get_vq_desc(vq, vq->iov,
 			ARRAY_SIZE(vq->iov), &out, &in,
 			NULL, NULL);
 	if (head < 0) {
@@ -713,16 +721,14 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 }
 
 static struct tcm_vhost_cmd *
-vhost_scsi_get_tag(struct vhost_virtqueue *vq,
-			struct tcm_vhost_tpg *tpg,
-			struct virtio_scsi_cmd_req *v_req,
-			u32 exp_data_len,
-			int data_direction)
+vhost_scsi_get_tag(struct vhost_virtqueue *vq, struct tcm_vhost_tpg *tpg,
+		   unsigned char *cdb, u64 scsi_tag, u16 lun, u8 task_attr,
+		   u32 exp_data_len, int data_direction)
 {
 	struct tcm_vhost_cmd *cmd;
 	struct tcm_vhost_nexus *tv_nexus;
 	struct se_session *se_sess;
-	struct scatterlist *sg;
+	struct scatterlist *sg, *prot_sg;
 	struct page **pages;
 	int tag;
 
@@ -741,19 +747,24 @@ vhost_scsi_get_tag(struct vhost_virtqueue *vq,
 
 	cmd = &((struct tcm_vhost_cmd *)se_sess->sess_cmd_map)[tag];
 	sg = cmd->tvc_sgl;
+	prot_sg = cmd->tvc_prot_sgl;
 	pages = cmd->tvc_upages;
 	memset(cmd, 0, sizeof(struct tcm_vhost_cmd));
 
 	cmd->tvc_sgl = sg;
+	cmd->tvc_prot_sgl = prot_sg;
 	cmd->tvc_upages = pages;
 	cmd->tvc_se_cmd.map_tag = tag;
-	cmd->tvc_tag = v_req->tag;
-	cmd->tvc_task_attr = v_req->task_attr;
+	cmd->tvc_tag = scsi_tag;
+	cmd->tvc_lun = lun;
+	cmd->tvc_task_attr = task_attr;
 	cmd->tvc_exp_data_len = exp_data_len;
 	cmd->tvc_data_direction = data_direction;
 	cmd->tvc_nexus = tv_nexus;
 	cmd->inflight = tcm_vhost_get_inflight(vq);
 
+	memcpy(cmd->tvc_cdb, cdb, TCM_VHOST_MAX_CDB_SIZE);
+
 	return cmd;
 }
 
@@ -767,35 +778,28 @@ vhost_scsi_map_to_sgl(struct tcm_vhost_cmd *tv_cmd,
 		      struct scatterlist *sgl,
 		      unsigned int sgl_count,
 		      struct iovec *iov,
-		      int write)
+		      struct page **pages,
+		      bool write)
 {
 	unsigned int npages = 0, pages_nr, offset, nbytes;
 	struct scatterlist *sg = sgl;
 	void __user *ptr = iov->iov_base;
 	size_t len = iov->iov_len;
-	struct page **pages;
 	int ret, i;
 
-	if (sgl_count > TCM_VHOST_PREALLOC_SGLS) {
-		pr_err("vhost_scsi_map_to_sgl() psgl_count: %u greater than"
-		       " preallocated TCM_VHOST_PREALLOC_SGLS: %u\n",
-			sgl_count, TCM_VHOST_PREALLOC_SGLS);
-		return -ENOBUFS;
-	}
-
 	pages_nr = iov_num_pages(iov);
-	if (pages_nr > sgl_count)
+	if (pages_nr > sgl_count) {
+		pr_err("vhost_scsi_map_to_sgl() pages_nr: %u greater than"
+		       " sgl_count: %u\n", pages_nr, sgl_count);
 		return -ENOBUFS;
-
-	if (pages_nr > TCM_VHOST_PREALLOC_PAGES) {
+	}
+	if (pages_nr > TCM_VHOST_PREALLOC_UPAGES) {
 		pr_err("vhost_scsi_map_to_sgl() pages_nr: %u greater than"
-		       " preallocated TCM_VHOST_PREALLOC_PAGES: %u\n",
-			pages_nr, TCM_VHOST_PREALLOC_PAGES);
+		       " preallocated TCM_VHOST_PREALLOC_UPAGES: %u\n",
+			pages_nr, TCM_VHOST_PREALLOC_UPAGES);
 		return -ENOBUFS;
 	}
 
-	pages = tv_cmd->tvc_upages;
-
 	ret = get_user_pages_fast((unsigned long)ptr, pages_nr, write, pages);
 	/* No pages were pinned */
 	if (ret < 0)
@@ -825,33 +829,32 @@ out:
 static int
 vhost_scsi_map_iov_to_sgl(struct tcm_vhost_cmd *cmd,
 			  struct iovec *iov,
-			  unsigned int niov,
-			  int write)
+			  int niov,
+			  bool write)
 {
-	int ret;
-	unsigned int i;
-	u32 sgl_count;
-	struct scatterlist *sg;
+	struct scatterlist *sg = cmd->tvc_sgl;
+	unsigned int sgl_count = 0;
+	int ret, i;
 
-	/*
-	 * Find out how long sglist needs to be
-	 */
-	sgl_count = 0;
 	for (i = 0; i < niov; i++)
 		sgl_count += iov_num_pages(&iov[i]);
 
-	/* TODO overflow checking */
+	if (sgl_count > TCM_VHOST_PREALLOC_SGLS) {
+		pr_err("vhost_scsi_map_iov_to_sgl() sgl_count: %u greater than"
+			" preallocated TCM_VHOST_PREALLOC_SGLS: %u\n",
+			sgl_count, TCM_VHOST_PREALLOC_SGLS);
+		return -ENOBUFS;
+	}
 
-	sg = cmd->tvc_sgl;
 	pr_debug("%s sg %p sgl_count %u\n", __func__, sg, sgl_count);
 	sg_init_table(sg, sgl_count);
-
 	cmd->tvc_sgl_count = sgl_count;
 
-	pr_debug("Mapping %u iovecs for %u pages\n", niov, sgl_count);
+	pr_debug("Mapping iovec %p for %u pages\n", &iov[0], sgl_count);
+
 	for (i = 0; i < niov; i++) {
 		ret = vhost_scsi_map_to_sgl(cmd, sg, sgl_count, &iov[i],
-					    write);
+					    cmd->tvc_upages, write);
 		if (ret < 0) {
 			for (i = 0; i < cmd->tvc_sgl_count; i++)
 				put_page(sg_page(&cmd->tvc_sgl[i]));
@@ -859,31 +862,70 @@ vhost_scsi_map_iov_to_sgl(struct tcm_vhost_cmd *cmd,
 			cmd->tvc_sgl_count = 0;
 			return ret;
 		}
-
 		sg += ret;
 		sgl_count -= ret;
 	}
 	return 0;
 }
 
+static int
+vhost_scsi_map_iov_to_prot(struct tcm_vhost_cmd *cmd,
+			   struct iovec *iov,
+			   int niov,
+			   bool write)
+{
+	struct scatterlist *prot_sg = cmd->tvc_prot_sgl;
+	unsigned int prot_sgl_count = 0;
+	int ret, i;
+
+	for (i = 0; i < niov; i++)
+		prot_sgl_count += iov_num_pages(&iov[i]);
+
+	if (prot_sgl_count > TCM_VHOST_PREALLOC_PROT_SGLS) {
+		pr_err("vhost_scsi_map_iov_to_prot() sgl_count: %u greater than"
+			" preallocated TCM_VHOST_PREALLOC_PROT_SGLS: %u\n",
+			prot_sgl_count, TCM_VHOST_PREALLOC_PROT_SGLS);
+		return -ENOBUFS;
+	}
+
+	pr_debug("%s prot_sg %p prot_sgl_count %u\n", __func__,
+		 prot_sg, prot_sgl_count);
+	sg_init_table(prot_sg, prot_sgl_count);
+	cmd->tvc_prot_sgl_count = prot_sgl_count;
+
+	for (i = 0; i < niov; i++) {
+		ret = vhost_scsi_map_to_sgl(cmd, prot_sg, prot_sgl_count, &iov[i],
+					    cmd->tvc_upages, write);
+		if (ret < 0) {
+			for (i = 0; i < cmd->tvc_prot_sgl_count; i++)
+				put_page(sg_page(&cmd->tvc_prot_sgl[i]));
+
+			cmd->tvc_prot_sgl_count = 0;
+			return ret;
+		}
+		prot_sg += ret;
+		prot_sgl_count -= ret;
+	}
+	return 0;
+}
+
 static void tcm_vhost_submission_work(struct work_struct *work)
 {
 	struct tcm_vhost_cmd *cmd =
 		container_of(work, struct tcm_vhost_cmd, work);
 	struct tcm_vhost_nexus *tv_nexus;
 	struct se_cmd *se_cmd = &cmd->tvc_se_cmd;
-	struct scatterlist *sg_ptr, *sg_bidi_ptr = NULL;
-	int rc, sg_no_bidi = 0;
+	struct scatterlist *sg_ptr, *sg_prot_ptr = NULL;
+	int rc;
 
+	/* FIXME: BIDI operation */
 	if (cmd->tvc_sgl_count) {
 		sg_ptr = cmd->tvc_sgl;
-/* FIXME: Fix BIDI operation in tcm_vhost_submission_work() */
-#if 0
-		if (se_cmd->se_cmd_flags & SCF_BIDI) {
-			sg_bidi_ptr = NULL;
-			sg_no_bidi = 0;
-		}
-#endif
+
+		if (cmd->tvc_prot_sgl_count)
+			sg_prot_ptr = cmd->tvc_prot_sgl;
+		else
+			se_cmd->prot_pto = true;
 	} else {
 		sg_ptr = NULL;
 	}
@@ -894,7 +936,7 @@ static void tcm_vhost_submission_work(struct work_struct *work)
 			cmd->tvc_lun, cmd->tvc_exp_data_len,
 			cmd->tvc_task_attr, cmd->tvc_data_direction,
 			TARGET_SCF_ACK_KREF, sg_ptr, cmd->tvc_sgl_count,
-			sg_bidi_ptr, sg_no_bidi, NULL, 0);
+			NULL, 0, sg_prot_ptr, cmd->tvc_prot_sgl_count);
 	if (rc < 0) {
 		transport_send_check_condition_and_sense(se_cmd,
 				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE, 0);
@@ -926,12 +968,18 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 {
 	struct tcm_vhost_tpg **vs_tpg;
 	struct virtio_scsi_cmd_req v_req;
+	struct virtio_scsi_cmd_req_pi v_req_pi;
 	struct tcm_vhost_tpg *tpg;
 	struct tcm_vhost_cmd *cmd;
-	u32 exp_data_len, data_first, data_num, data_direction;
+	u64 tag;
+	u32 exp_data_len, data_first, data_num, data_direction, prot_first;
 	unsigned out, in, i;
-	int head, ret;
-	u8 target;
+	int head, ret, data_niov, prot_niov, prot_bytes;
+	size_t req_size;
+	u16 lun;
+	u8 *target, *lunp, task_attr;
+	bool hdr_pi;
+	void *req, *cdb;
 
 	mutex_lock(&vq->mutex);
 	/*
@@ -945,7 +993,7 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 	vhost_disable_notify(&vs->dev, vq);
 
 	for (;;) {
-		head = vhost_get_vq_desc(&vs->dev, vq, vq->iov,
+		head = vhost_get_vq_desc(vq, vq->iov,
 					ARRAY_SIZE(vq->iov), &out, &in,
 					NULL, NULL);
 		pr_debug("vhost_get_vq_desc: head: %d, out: %u in: %u\n",
@@ -962,7 +1010,7 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 			break;
 		}
 
-/* FIXME: BIDI operation */
+		/* FIXME: BIDI operation */
 		if (out == 1 && in == 1) {
 			data_direction = DMA_NONE;
 			data_first = 0;
@@ -992,29 +1040,38 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 			break;
 		}
 
-		if (unlikely(vq->iov[0].iov_len != sizeof(v_req))) {
-			vq_err(vq, "Expecting virtio_scsi_cmd_req, got %zu"
-				" bytes\n", vq->iov[0].iov_len);
+		if (vhost_has_feature(vq, VIRTIO_SCSI_F_T10_PI)) {
+			req = &v_req_pi;
+			lunp = &v_req_pi.lun[0];
+			target = &v_req_pi.lun[1];
+			req_size = sizeof(v_req_pi);
+			hdr_pi = true;
+		} else {
+			req = &v_req;
+			lunp = &v_req.lun[0];
+			target = &v_req.lun[1];
+			req_size = sizeof(v_req);
+			hdr_pi = false;
+		}
+
+		if (unlikely(vq->iov[0].iov_len < req_size)) {
+			pr_err("Expecting virtio-scsi header: %zu, got %zu\n",
+			       req_size, vq->iov[0].iov_len);
 			break;
 		}
-		pr_debug("Calling __copy_from_user: vq->iov[0].iov_base: %p,"
-			" len: %zu\n", vq->iov[0].iov_base, sizeof(v_req));
-		ret = __copy_from_user(&v_req, vq->iov[0].iov_base,
-				sizeof(v_req));
+		ret = memcpy_fromiovecend(req, &vq->iov[0], 0, req_size);
 		if (unlikely(ret)) {
 			vq_err(vq, "Faulted on virtio_scsi_cmd_req\n");
 			break;
 		}
 
 		/* virtio-scsi spec requires byte 0 of the lun to be 1 */
-		if (unlikely(v_req.lun[0] != 1)) {
+		if (unlikely(*lunp != 1)) {
 			vhost_scsi_send_bad_target(vs, vq, head, out);
 			continue;
 		}
 
-		/* Extract the tpgt */
-		target = v_req.lun[1];
-		tpg = ACCESS_ONCE(vs_tpg[target]);
+		tpg = ACCESS_ONCE(vs_tpg[*target]);
 
 		/* Target does not exist, fail the request */
 		if (unlikely(!tpg)) {
@@ -1022,17 +1079,79 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 			continue;
 		}
 
+		data_niov = data_num;
+		prot_niov = prot_first = prot_bytes = 0;
+		/*
+		 * Determine if any protection information iovecs are preceeding
+		 * the actual data payload, and adjust data_first + data_niov
+		 * values accordingly for vhost_scsi_map_iov_to_sgl() below.
+		 *
+		 * Also extract virtio_scsi header bits for vhost_scsi_get_tag()
+		 */
+		if (hdr_pi) {
+			if (v_req_pi.pi_bytesout) {
+				if (data_direction != DMA_TO_DEVICE) {
+					vq_err(vq, "Received non zero do_pi_niov"
+						", but wrong data_direction\n");
+					goto err_cmd;
+				}
+				prot_bytes = v_req_pi.pi_bytesout;
+			} else if (v_req_pi.pi_bytesin) {
+				if (data_direction != DMA_FROM_DEVICE) {
+					vq_err(vq, "Received non zero di_pi_niov"
+						", but wrong data_direction\n");
+					goto err_cmd;
+				}
+				prot_bytes = v_req_pi.pi_bytesin;
+			}
+			if (prot_bytes) {
+				int tmp = 0;
+
+				for (i = 0; i < data_num; i++) {
+					tmp += vq->iov[data_first + i].iov_len;
+					prot_niov++;
+					if (tmp >= prot_bytes)
+						break;
+				}
+				prot_first = data_first;
+				data_first += prot_niov;
+				data_niov = data_num - prot_niov;
+			}
+			tag = v_req_pi.tag;
+			task_attr = v_req_pi.task_attr;
+			cdb = &v_req_pi.cdb[0];
+			lun = ((v_req_pi.lun[2] << 8) | v_req_pi.lun[3]) & 0x3FFF;
+		} else {
+			tag = v_req.tag;
+			task_attr = v_req.task_attr;
+			cdb = &v_req.cdb[0];
+			lun = ((v_req.lun[2] << 8) | v_req.lun[3]) & 0x3FFF;
+		}
 		exp_data_len = 0;
-		for (i = 0; i < data_num; i++)
+		for (i = 0; i < data_niov; i++)
 			exp_data_len += vq->iov[data_first + i].iov_len;
+		/*
+		 * Check that the recieved CDB size does not exceeded our
+		 * hardcoded max for vhost-scsi
+		 *
+		 * TODO what if cdb was too small for varlen cdb header?
+		 */
+		if (unlikely(scsi_command_size(cdb) > TCM_VHOST_MAX_CDB_SIZE)) {
+			vq_err(vq, "Received SCSI CDB with command_size: %d that"
+				" exceeds SCSI_MAX_VARLEN_CDB_SIZE: %d\n",
+				scsi_command_size(cdb), TCM_VHOST_MAX_CDB_SIZE);
+			goto err_cmd;
+		}
 
-		cmd = vhost_scsi_get_tag(vq, tpg, &v_req,
-					 exp_data_len, data_direction);
+		cmd = vhost_scsi_get_tag(vq, tpg, cdb, tag, lun, task_attr,
+					 exp_data_len + prot_bytes,
+					 data_direction);
 		if (IS_ERR(cmd)) {
 			vq_err(vq, "vhost_scsi_get_tag failed %ld\n",
 					PTR_ERR(cmd));
 			goto err_cmd;
 		}
+
 		pr_debug("Allocated tv_cmd: %p exp_data_len: %d, data_direction"
 			": %d\n", cmd, exp_data_len, data_direction);
 
@@ -1040,40 +1159,28 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 		cmd->tvc_vq = vq;
 		cmd->tvc_resp = vq->iov[out].iov_base;
 
-		/*
-		 * Copy in the recieved CDB descriptor into cmd->tvc_cdb
-		 * that will be used by tcm_vhost_new_cmd_map() and down into
-		 * target_setup_cmd_from_cdb()
-		 */
-		memcpy(cmd->tvc_cdb, v_req.cdb, TCM_VHOST_MAX_CDB_SIZE);
-		/*
-		 * Check that the recieved CDB size does not exceeded our
-		 * hardcoded max for tcm_vhost
-		 */
-		/* TODO what if cdb was too small for varlen cdb header? */
-		if (unlikely(scsi_command_size(cmd->tvc_cdb) >
-					TCM_VHOST_MAX_CDB_SIZE)) {
-			vq_err(vq, "Received SCSI CDB with command_size: %d that"
-				" exceeds SCSI_MAX_VARLEN_CDB_SIZE: %d\n",
-				scsi_command_size(cmd->tvc_cdb),
-				TCM_VHOST_MAX_CDB_SIZE);
-			goto err_free;
-		}
-		cmd->tvc_lun = ((v_req.lun[2] << 8) | v_req.lun[3]) & 0x3FFF;
-
 		pr_debug("vhost_scsi got command opcode: %#02x, lun: %d\n",
 			cmd->tvc_cdb[0], cmd->tvc_lun);
 
+		if (prot_niov) {
+			ret = vhost_scsi_map_iov_to_prot(cmd,
+					&vq->iov[prot_first], prot_niov,
+					data_direction == DMA_FROM_DEVICE);
+			if (unlikely(ret)) {
+				vq_err(vq, "Failed to map iov to"
+					" prot_sgl\n");
+				goto err_free;
+			}
+		}
 		if (data_direction != DMA_NONE) {
 			ret = vhost_scsi_map_iov_to_sgl(cmd,
-					&vq->iov[data_first], data_num,
+					&vq->iov[data_first], data_niov,
 					data_direction == DMA_FROM_DEVICE);
 			if (unlikely(ret)) {
 				vq_err(vq, "Failed to map iov to sgl\n");
 				goto err_free;
 			}
 		}
-
 		/*
 		 * Save the descriptor from vhost_get_vq_desc() to be used to
 		 * complete the virtio-scsi request in TCM callback context via
@@ -1373,6 +1480,9 @@ err_dev:
 
 static int vhost_scsi_set_features(struct vhost_scsi *vs, u64 features)
 {
+	struct vhost_virtqueue *vq;
+	int i;
+
 	if (features & ~VHOST_SCSI_FEATURES)
 		return -EOPNOTSUPP;
 
@@ -1382,21 +1492,17 @@ static int vhost_scsi_set_features(struct vhost_scsi *vs, u64 features)
 		mutex_unlock(&vs->dev.mutex);
 		return -EFAULT;
 	}
-	vs->dev.acked_features = features;
-	smp_wmb();
-	vhost_scsi_flush(vs);
+
+	for (i = 0; i < VHOST_SCSI_MAX_VQ; i++) {
+		vq = &vs->vqs[i].vq;
+		mutex_lock(&vq->mutex);
+		vq->acked_features = features;
+		mutex_unlock(&vq->mutex);
+	}
 	mutex_unlock(&vs->dev.mutex);
 	return 0;
 }
 
-static void vhost_scsi_free(struct vhost_scsi *vs)
-{
-	if (is_vmalloc_addr(vs))
-		vfree(vs);
-	else
-		kfree(vs);
-}
-
 static int vhost_scsi_open(struct inode *inode, struct file *f)
 {
 	struct vhost_scsi *vs;
@@ -1436,7 +1542,7 @@ static int vhost_scsi_open(struct inode *inode, struct file *f)
 	return 0;
 
 err_vqs:
-	vhost_scsi_free(vs);
+	kvfree(vs);
 err_vs:
 	return r;
 }
@@ -1455,7 +1561,7 @@ static int vhost_scsi_release(struct inode *inode, struct file *f)
 	/* Jobs can re-queue themselves in evt kick handler. Do extra flush. */
 	vhost_scsi_flush(vs);
 	kfree(vs->dev.vqs);
-	vhost_scsi_free(vs);
+	kvfree(vs);
 	return 0;
 }
 
@@ -1591,10 +1697,6 @@ tcm_vhost_do_plug(struct tcm_vhost_tpg *tpg,
 		return;
 
 	mutex_lock(&vs->dev.mutex);
-	if (!vhost_has_feature(&vs->dev, VIRTIO_SCSI_F_HOTPLUG)) {
-		mutex_unlock(&vs->dev.mutex);
-		return;
-	}
 
 	if (plug)
 		reason = VIRTIO_SCSI_EVT_RESET_RESCAN;
@@ -1603,8 +1705,9 @@ tcm_vhost_do_plug(struct tcm_vhost_tpg *tpg,
 
 	vq = &vs->vqs[VHOST_SCSI_VQ_EVT].vq;
 	mutex_lock(&vq->mutex);
-	tcm_vhost_send_evt(vs, tpg, lun,
-			VIRTIO_SCSI_T_TRANSPORT_RESET, reason);
+	if (vhost_has_feature(vq, VIRTIO_SCSI_F_HOTPLUG))
+		tcm_vhost_send_evt(vs, tpg, lun,
+				   VIRTIO_SCSI_T_TRANSPORT_RESET, reason);
 	mutex_unlock(&vq->mutex);
 	mutex_unlock(&vs->dev.mutex);
 }
@@ -1712,6 +1815,7 @@ static void tcm_vhost_free_cmd_map_res(struct tcm_vhost_nexus *nexus,
 		tv_cmd = &((struct tcm_vhost_cmd *)se_sess->sess_cmd_map)[i];
 
 		kfree(tv_cmd->tvc_sgl);
+		kfree(tv_cmd->tvc_prot_sgl);
 		kfree(tv_cmd->tvc_upages);
 	}
 }
@@ -1746,7 +1850,7 @@ static int tcm_vhost_make_nexus(struct tcm_vhost_tpg *tpg,
 	tv_nexus->tvn_se_sess = transport_init_session_tags(
 					TCM_VHOST_DEFAULT_TAGS,
 					sizeof(struct tcm_vhost_cmd),
-					TARGET_PROT_NORMAL);
+					TARGET_PROT_DIN_PASS | TARGET_PROT_DOUT_PASS);
 	if (IS_ERR(tv_nexus->tvn_se_sess)) {
 		mutex_unlock(&tpg->tv_tpg_mutex);
 		kfree(tv_nexus);
@@ -1765,12 +1869,20 @@ static int tcm_vhost_make_nexus(struct tcm_vhost_tpg *tpg,
 		}
 
 		tv_cmd->tvc_upages = kzalloc(sizeof(struct page *) *
-					TCM_VHOST_PREALLOC_PAGES, GFP_KERNEL);
+					TCM_VHOST_PREALLOC_UPAGES, GFP_KERNEL);
 		if (!tv_cmd->tvc_upages) {
 			mutex_unlock(&tpg->tv_tpg_mutex);
 			pr_err("Unable to allocate tv_cmd->tvc_upages\n");
 			goto out;
 		}
+
+		tv_cmd->tvc_prot_sgl = kzalloc(sizeof(struct scatterlist) *
+					TCM_VHOST_PREALLOC_PROT_SGLS, GFP_KERNEL);
+		if (!tv_cmd->tvc_prot_sgl) {
+			mutex_unlock(&tpg->tv_tpg_mutex);
+			pr_err("Unable to allocate tv_cmd->tvc_prot_sgl\n");
+			goto out;
+		}
 	}
 	/*
 	 * Since we are running in 'demo mode' this call with generate a
diff --git a/drivers/vhost/test.c b/drivers/vhost/test.c
index c2a54fbf7f99..d9c501eaa6c3 100644
--- a/drivers/vhost/test.c
+++ b/drivers/vhost/test.c
@@ -53,7 +53,7 @@ static void handle_vq(struct vhost_test *n)
 	vhost_disable_notify(&n->dev, vq);
 
 	for (;;) {
-		head = vhost_get_vq_desc(&n->dev, vq, vq->iov,
+		head = vhost_get_vq_desc(vq, vq->iov,
 					 ARRAY_SIZE(vq->iov),
 					 &out, &in,
 					 NULL, NULL);
@@ -241,15 +241,18 @@ done:
 
 static int vhost_test_set_features(struct vhost_test *n, u64 features)
 {
+	struct vhost_virtqueue *vq;
+
 	mutex_lock(&n->dev.mutex);
 	if ((features & (1 << VHOST_F_LOG_ALL)) &&
 	    !vhost_log_access_ok(&n->dev)) {
 		mutex_unlock(&n->dev.mutex);
 		return -EFAULT;
 	}
-	n->dev.acked_features = features;
-	smp_wmb();
-	vhost_test_flush(n);
+	vq = &n->vqs[VHOST_TEST_VQ];
+	mutex_lock(&vq->mutex);
+	vq->acked_features = features;
+	mutex_unlock(&vq->mutex);
 	mutex_unlock(&n->dev.mutex);
 	return 0;
 }
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 78987e481bc6..c90f4374442a 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -18,7 +18,6 @@
 #include <linux/mmu_context.h>
 #include <linux/miscdevice.h>
 #include <linux/mutex.h>
-#include <linux/rcupdate.h>
 #include <linux/poll.h>
 #include <linux/file.h>
 #include <linux/highmem.h>
@@ -191,6 +190,7 @@ static void vhost_vq_reset(struct vhost_dev *dev,
 	vq->log_used = false;
 	vq->log_addr = -1ull;
 	vq->private_data = NULL;
+	vq->acked_features = 0;
 	vq->log_base = NULL;
 	vq->error_ctx = NULL;
 	vq->error = NULL;
@@ -198,6 +198,7 @@ static void vhost_vq_reset(struct vhost_dev *dev,
 	vq->call_ctx = NULL;
 	vq->call = NULL;
 	vq->log_ctx = NULL;
+	vq->memory = NULL;
 }
 
 static int vhost_worker(void *data)
@@ -415,11 +416,18 @@ EXPORT_SYMBOL_GPL(vhost_dev_reset_owner_prepare);
 /* Caller should have device mutex */
 void vhost_dev_reset_owner(struct vhost_dev *dev, struct vhost_memory *memory)
 {
+	int i;
+
 	vhost_dev_cleanup(dev, true);
 
 	/* Restore memory to default empty mapping. */
 	memory->nregions = 0;
-	RCU_INIT_POINTER(dev->memory, memory);
+	dev->memory = memory;
+	/* We don't need VQ locks below since vhost_dev_cleanup makes sure
+	 * VQs aren't running.
+	 */
+	for (i = 0; i < dev->nvqs; ++i)
+		dev->vqs[i]->memory = memory;
 }
 EXPORT_SYMBOL_GPL(vhost_dev_reset_owner);
 
@@ -462,10 +470,8 @@ void vhost_dev_cleanup(struct vhost_dev *dev, bool locked)
 		fput(dev->log_file);
 	dev->log_file = NULL;
 	/* No one will access memory at this point */
-	kfree(rcu_dereference_protected(dev->memory,
-					locked ==
-						lockdep_is_held(&dev->mutex)));
-	RCU_INIT_POINTER(dev->memory, NULL);
+	kfree(dev->memory);
+	dev->memory = NULL;
 	WARN_ON(!list_empty(&dev->work_list));
 	if (dev->worker) {
 		kthread_stop(dev->worker);
@@ -524,11 +530,13 @@ static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem,
 
 	for (i = 0; i < d->nvqs; ++i) {
 		int ok;
+		bool log;
+
 		mutex_lock(&d->vqs[i]->mutex);
+		log = log_all || vhost_has_feature(d->vqs[i], VHOST_F_LOG_ALL);
 		/* If ring is inactive, will check when it's enabled. */
 		if (d->vqs[i]->private_data)
-			ok = vq_memory_access_ok(d->vqs[i]->log_base, mem,
-						 log_all);
+			ok = vq_memory_access_ok(d->vqs[i]->log_base, mem, log);
 		else
 			ok = 1;
 		mutex_unlock(&d->vqs[i]->mutex);
@@ -538,12 +546,12 @@ static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem,
 	return 1;
 }
 
-static int vq_access_ok(struct vhost_dev *d, unsigned int num,
+static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
 			struct vring_desc __user *desc,
 			struct vring_avail __user *avail,
 			struct vring_used __user *used)
 {
-	size_t s = vhost_has_feature(d, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
+	size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
 	return access_ok(VERIFY_READ, desc, num * sizeof *desc) &&
 	       access_ok(VERIFY_READ, avail,
 			 sizeof *avail + num * sizeof *avail->ring + s) &&
@@ -555,26 +563,19 @@ static int vq_access_ok(struct vhost_dev *d, unsigned int num,
 /* Caller should have device mutex but not vq mutex */
 int vhost_log_access_ok(struct vhost_dev *dev)
 {
-	struct vhost_memory *mp;
-
-	mp = rcu_dereference_protected(dev->memory,
-				       lockdep_is_held(&dev->mutex));
-	return memory_access_ok(dev, mp, 1);
+	return memory_access_ok(dev, dev->memory, 1);
 }
 EXPORT_SYMBOL_GPL(vhost_log_access_ok);
 
 /* Verify access for write logging. */
 /* Caller should have vq mutex and device mutex */
-static int vq_log_access_ok(struct vhost_dev *d, struct vhost_virtqueue *vq,
+static int vq_log_access_ok(struct vhost_virtqueue *vq,
 			    void __user *log_base)
 {
-	struct vhost_memory *mp;
-	size_t s = vhost_has_feature(d, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
+	size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
 
-	mp = rcu_dereference_protected(vq->dev->memory,
-				       lockdep_is_held(&vq->mutex));
-	return vq_memory_access_ok(log_base, mp,
-			    vhost_has_feature(vq->dev, VHOST_F_LOG_ALL)) &&
+	return vq_memory_access_ok(log_base, vq->memory,
+				   vhost_has_feature(vq, VHOST_F_LOG_ALL)) &&
 		(!vq->log_used || log_access_ok(log_base, vq->log_addr,
 					sizeof *vq->used +
 					vq->num * sizeof *vq->used->ring + s));
@@ -584,8 +585,8 @@ static int vq_log_access_ok(struct vhost_dev *d, struct vhost_virtqueue *vq,
 /* Caller should have vq mutex and device mutex */
 int vhost_vq_access_ok(struct vhost_virtqueue *vq)
 {
-	return vq_access_ok(vq->dev, vq->num, vq->desc, vq->avail, vq->used) &&
-		vq_log_access_ok(vq->dev, vq, vq->log_base);
+	return vq_access_ok(vq, vq->num, vq->desc, vq->avail, vq->used) &&
+		vq_log_access_ok(vq, vq->log_base);
 }
 EXPORT_SYMBOL_GPL(vhost_vq_access_ok);
 
@@ -593,6 +594,7 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
 {
 	struct vhost_memory mem, *newmem, *oldmem;
 	unsigned long size = offsetof(struct vhost_memory, regions);
+	int i;
 
 	if (copy_from_user(&mem, m, size))
 		return -EFAULT;
@@ -611,15 +613,19 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
 		return -EFAULT;
 	}
 
-	if (!memory_access_ok(d, newmem,
-			      vhost_has_feature(d, VHOST_F_LOG_ALL))) {
+	if (!memory_access_ok(d, newmem, 0)) {
 		kfree(newmem);
 		return -EFAULT;
 	}
-	oldmem = rcu_dereference_protected(d->memory,
-					   lockdep_is_held(&d->mutex));
-	rcu_assign_pointer(d->memory, newmem);
-	synchronize_rcu();
+	oldmem = d->memory;
+	d->memory = newmem;
+
+	/* All memory accesses are done under some VQ mutex. */
+	for (i = 0; i < d->nvqs; ++i) {
+		mutex_lock(&d->vqs[i]->mutex);
+		d->vqs[i]->memory = newmem;
+		mutex_unlock(&d->vqs[i]->mutex);
+	}
 	kfree(oldmem);
 	return 0;
 }
@@ -718,7 +724,7 @@ long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp)
 		 * If it is not, we don't as size might not have been setup.
 		 * We will verify when backend is configured. */
 		if (vq->private_data) {
-			if (!vq_access_ok(d, vq->num,
+			if (!vq_access_ok(vq, vq->num,
 				(void __user *)(unsigned long)a.desc_user_addr,
 				(void __user *)(unsigned long)a.avail_user_addr,
 				(void __user *)(unsigned long)a.used_user_addr)) {
@@ -858,7 +864,7 @@ long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp)
 			vq = d->vqs[i];
 			mutex_lock(&vq->mutex);
 			/* If ring is inactive, will check when it's enabled. */
-			if (vq->private_data && !vq_log_access_ok(d, vq, base))
+			if (vq->private_data && !vq_log_access_ok(vq, base))
 				r = -EFAULT;
 			else
 				vq->log_base = base;
@@ -1044,7 +1050,7 @@ int vhost_init_used(struct vhost_virtqueue *vq)
 }
 EXPORT_SYMBOL_GPL(vhost_init_used);
 
-static int translate_desc(struct vhost_dev *dev, u64 addr, u32 len,
+static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len,
 			  struct iovec iov[], int iov_size)
 {
 	const struct vhost_memory_region *reg;
@@ -1053,9 +1059,7 @@ static int translate_desc(struct vhost_dev *dev, u64 addr, u32 len,
 	u64 s = 0;
 	int ret = 0;
 
-	rcu_read_lock();
-
-	mem = rcu_dereference(dev->memory);
+	mem = vq->memory;
 	while ((u64)len > s) {
 		u64 size;
 		if (unlikely(ret >= iov_size)) {
@@ -1077,7 +1081,6 @@ static int translate_desc(struct vhost_dev *dev, u64 addr, u32 len,
 		++ret;
 	}
 
-	rcu_read_unlock();
 	return ret;
 }
 
@@ -1102,7 +1105,7 @@ static unsigned next_desc(struct vring_desc *desc)
 	return next;
 }
 
-static int get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq,
+static int get_indirect(struct vhost_virtqueue *vq,
 			struct iovec iov[], unsigned int iov_size,
 			unsigned int *out_num, unsigned int *in_num,
 			struct vhost_log *log, unsigned int *log_num,
@@ -1121,7 +1124,7 @@ static int get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq,
 		return -EINVAL;
 	}
 
-	ret = translate_desc(dev, indirect->addr, indirect->len, vq->indirect,
+	ret = translate_desc(vq, indirect->addr, indirect->len, vq->indirect,
 			     UIO_MAXIOV);
 	if (unlikely(ret < 0)) {
 		vq_err(vq, "Translation failure %d in indirect.\n", ret);
@@ -1161,7 +1164,7 @@ static int get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq,
 			return -EINVAL;
 		}
 
-		ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count,
+		ret = translate_desc(vq, desc.addr, desc.len, iov + iov_count,
 				     iov_size - iov_count);
 		if (unlikely(ret < 0)) {
 			vq_err(vq, "Translation failure %d indirect idx %d\n",
@@ -1198,7 +1201,7 @@ static int get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq,
  * This function returns the descriptor number found, or vq->num (which is
  * never a valid descriptor number) if none was found.  A negative code is
  * returned on error. */
-int vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq,
+int vhost_get_vq_desc(struct vhost_virtqueue *vq,
 		      struct iovec iov[], unsigned int iov_size,
 		      unsigned int *out_num, unsigned int *in_num,
 		      struct vhost_log *log, unsigned int *log_num)
@@ -1272,7 +1275,7 @@ int vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq,
 			return -EFAULT;
 		}
 		if (desc.flags & VRING_DESC_F_INDIRECT) {
-			ret = get_indirect(dev, vq, iov, iov_size,
+			ret = get_indirect(vq, iov, iov_size,
 					   out_num, in_num,
 					   log, log_num, &desc);
 			if (unlikely(ret < 0)) {
@@ -1283,7 +1286,7 @@ int vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq,
 			continue;
 		}
 
-		ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count,
+		ret = translate_desc(vq, desc.addr, desc.len, iov + iov_count,
 				     iov_size - iov_count);
 		if (unlikely(ret < 0)) {
 			vq_err(vq, "Translation failure %d descriptor idx %d\n",
@@ -1426,11 +1429,11 @@ static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	 * interrupts. */
 	smp_mb();
 
-	if (vhost_has_feature(dev, VIRTIO_F_NOTIFY_ON_EMPTY) &&
+	if (vhost_has_feature(vq, VIRTIO_F_NOTIFY_ON_EMPTY) &&
 	    unlikely(vq->avail_idx == vq->last_avail_idx))
 		return true;
 
-	if (!vhost_has_feature(dev, VIRTIO_RING_F_EVENT_IDX)) {
+	if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
 		__u16 flags;
 		if (__get_user(flags, &vq->avail->flags)) {
 			vq_err(vq, "Failed to get flags");
@@ -1491,7 +1494,7 @@ bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	if (!(vq->used_flags & VRING_USED_F_NO_NOTIFY))
 		return false;
 	vq->used_flags &= ~VRING_USED_F_NO_NOTIFY;
-	if (!vhost_has_feature(dev, VIRTIO_RING_F_EVENT_IDX)) {
+	if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
 		r = vhost_update_used_flags(vq);
 		if (r) {
 			vq_err(vq, "Failed to enable notification at %p: %d\n",
@@ -1528,7 +1531,7 @@ void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	if (vq->used_flags & VRING_USED_F_NO_NOTIFY)
 		return;
 	vq->used_flags |= VRING_USED_F_NO_NOTIFY;
-	if (!vhost_has_feature(dev, VIRTIO_RING_F_EVENT_IDX)) {
+	if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
 		r = vhost_update_used_flags(vq);
 		if (r)
 			vq_err(vq, "Failed to enable notification at %p: %d\n",
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 35eeb2a1bada..3eda654b8f5a 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -104,20 +104,18 @@ struct vhost_virtqueue {
 	struct iovec *indirect;
 	struct vring_used_elem *heads;
 	/* Protected by virtqueue mutex. */
+	struct vhost_memory *memory;
 	void *private_data;
+	unsigned acked_features;
 	/* Log write descriptors */
 	void __user *log_base;
 	struct vhost_log *log;
 };
 
 struct vhost_dev {
-	/* Readers use RCU to access memory table pointer
-	 * log base pointer and features.
-	 * Writers use mutex below.*/
-	struct vhost_memory __rcu *memory;
+	struct vhost_memory *memory;
 	struct mm_struct *mm;
 	struct mutex mutex;
-	unsigned acked_features;
 	struct vhost_virtqueue **vqs;
 	int nvqs;
 	struct file *log_file;
@@ -140,7 +138,7 @@ long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp);
 int vhost_vq_access_ok(struct vhost_virtqueue *vq);
 int vhost_log_access_ok(struct vhost_dev *);
 
-int vhost_get_vq_desc(struct vhost_dev *, struct vhost_virtqueue *,
+int vhost_get_vq_desc(struct vhost_virtqueue *,
 		      struct iovec iov[], unsigned int iov_count,
 		      unsigned int *out_num, unsigned int *in_num,
 		      struct vhost_log *log, unsigned int *log_num);
@@ -174,13 +172,8 @@ enum {
 			 (1ULL << VHOST_F_LOG_ALL),
 };
 
-static inline int vhost_has_feature(struct vhost_dev *dev, int bit)
+static inline int vhost_has_feature(struct vhost_virtqueue *vq, int bit)
 {
-	unsigned acked_features;
-
-	/* TODO: check that we are running from vhost_worker or dev mutex is
-	 * held? */
-	acked_features = rcu_dereference_index_check(dev->acked_features, 1);
-	return acked_features & (1 << bit);
+	return vq->acked_features & (1 << bit);
 }
 #endif
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index c7b4f0f927b1..8bf495ffb020 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -20,6 +20,7 @@ source "drivers/char/agp/Kconfig"
 source "drivers/gpu/vga/Kconfig"
 
 source "drivers/gpu/host1x/Kconfig"
+source "drivers/gpu/ipu-v3/Kconfig"
 
 menu "Direct Rendering Manager"
 source "drivers/gpu/drm/Kconfig"
diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index 5a3eb2ecb525..5d449059a556 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -29,7 +29,7 @@ if LCD_CLASS_DEVICE
 
 config LCD_CORGI
 	tristate "LCD Panel support for SHARP corgi/spitz model"
-	depends on SPI_MASTER && PXA_SHARPSL
+	depends on SPI_MASTER && PXA_SHARPSL && BACKLIGHT_CLASS_DEVICE
 	help
 	  Say y here to support the LCD panels usually found on SHARP
 	  corgi (C7x0) and spitz (Cxx00) models.
@@ -370,7 +370,7 @@ config BACKLIGHT_AAT2870
 
 config BACKLIGHT_LM3630A
 	tristate "Backlight Driver for LM3630A"
-	depends on BACKLIGHT_CLASS_DEVICE && I2C
+	depends on BACKLIGHT_CLASS_DEVICE && I2C && PWM
 	select REGMAP_I2C
 	help
 	  This supports TI LM3630A Backlight Driver
@@ -386,14 +386,14 @@ config BACKLIGHT_LM3639
 
 config BACKLIGHT_LP855X
 	tristate "Backlight driver for TI LP855X"
-	depends on BACKLIGHT_CLASS_DEVICE && I2C
+	depends on BACKLIGHT_CLASS_DEVICE && I2C && PWM
 	help
 	  This supports TI LP8550, LP8551, LP8552, LP8553, LP8555, LP8556 and
 	  LP8557 backlight driver.
 
 config BACKLIGHT_LP8788
 	tristate "Backlight driver for TI LP8788 MFD"
-	depends on BACKLIGHT_CLASS_DEVICE && MFD_LP8788
+	depends on BACKLIGHT_CLASS_DEVICE && MFD_LP8788 && PWM
 	help
 	  This supports TI LP8788 backlight driver.
 
diff --git a/drivers/video/backlight/gpio_backlight.c b/drivers/video/backlight/gpio_backlight.c
index a2eba12e1cb7..1cea68848f1a 100644
--- a/drivers/video/backlight/gpio_backlight.c
+++ b/drivers/video/backlight/gpio_backlight.c
@@ -38,7 +38,8 @@ static int gpio_backlight_update_status(struct backlight_device *bl)
 	    bl->props.state & (BL_CORE_SUSPENDED | BL_CORE_FBBLANK))
 		brightness = 0;
 
-	gpio_set_value(gbl->gpio, brightness ? gbl->active : !gbl->active);
+	gpio_set_value_cansleep(gbl->gpio,
+				brightness ? gbl->active : !gbl->active);
 
 	return 0;
 }
diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c
index b75201ff46f6..38ca88bc5c3e 100644
--- a/drivers/video/backlight/pwm_bl.c
+++ b/drivers/video/backlight/pwm_bl.c
@@ -10,8 +10,8 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/gpio/consumer.h>
 #include <linux/gpio.h>
-#include <linux/of_gpio.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
@@ -32,8 +32,7 @@ struct pwm_bl_data {
 	unsigned int		*levels;
 	bool			enabled;
 	struct regulator	*power_supply;
-	int			enable_gpio;
-	unsigned long		enable_gpio_flags;
+	struct gpio_desc	*enable_gpio;
 	unsigned int		scale;
 	int			(*notify)(struct device *,
 					  int brightness);
@@ -54,12 +53,8 @@ static void pwm_backlight_power_on(struct pwm_bl_data *pb, int brightness)
 	if (err < 0)
 		dev_err(pb->dev, "failed to enable power supply\n");
 
-	if (gpio_is_valid(pb->enable_gpio)) {
-		if (pb->enable_gpio_flags & PWM_BACKLIGHT_GPIO_ACTIVE_LOW)
-			gpio_set_value(pb->enable_gpio, 0);
-		else
-			gpio_set_value(pb->enable_gpio, 1);
-	}
+	if (pb->enable_gpio)
+		gpiod_set_value(pb->enable_gpio, 1);
 
 	pwm_enable(pb->pwm);
 	pb->enabled = true;
@@ -73,12 +68,8 @@ static void pwm_backlight_power_off(struct pwm_bl_data *pb)
 	pwm_config(pb->pwm, 0, pb->period);
 	pwm_disable(pb->pwm);
 
-	if (gpio_is_valid(pb->enable_gpio)) {
-		if (pb->enable_gpio_flags & PWM_BACKLIGHT_GPIO_ACTIVE_LOW)
-			gpio_set_value(pb->enable_gpio, 1);
-		else
-			gpio_set_value(pb->enable_gpio, 0);
-	}
+	if (pb->enable_gpio)
+		gpiod_set_value(pb->enable_gpio, 0);
 
 	regulator_disable(pb->power_supply);
 	pb->enabled = false;
@@ -148,7 +139,6 @@ static int pwm_backlight_parse_dt(struct device *dev,
 				  struct platform_pwm_backlight_data *data)
 {
 	struct device_node *node = dev->of_node;
-	enum of_gpio_flags flags;
 	struct property *prop;
 	int length;
 	u32 value;
@@ -189,14 +179,6 @@ static int pwm_backlight_parse_dt(struct device *dev,
 		data->max_brightness--;
 	}
 
-	data->enable_gpio = of_get_named_gpio_flags(node, "enable-gpios", 0,
-						    &flags);
-	if (data->enable_gpio == -EPROBE_DEFER)
-		return -EPROBE_DEFER;
-
-	if (gpio_is_valid(data->enable_gpio) && (flags & OF_GPIO_ACTIVE_LOW))
-		data->enable_gpio_flags |= PWM_BACKLIGHT_GPIO_ACTIVE_LOW;
-
 	return 0;
 }
 
@@ -256,8 +238,6 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	} else
 		pb->scale = data->max_brightness;
 
-	pb->enable_gpio = data->enable_gpio;
-	pb->enable_gpio_flags = data->enable_gpio_flags;
 	pb->notify = data->notify;
 	pb->notify_after = data->notify_after;
 	pb->check_fb = data->check_fb;
@@ -265,26 +245,38 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	pb->dev = &pdev->dev;
 	pb->enabled = false;
 
-	if (gpio_is_valid(pb->enable_gpio)) {
-		unsigned long flags;
-
-		if (pb->enable_gpio_flags & PWM_BACKLIGHT_GPIO_ACTIVE_LOW)
-			flags = GPIOF_OUT_INIT_HIGH;
+	pb->enable_gpio = devm_gpiod_get(&pdev->dev, "enable");
+	if (IS_ERR(pb->enable_gpio)) {
+		ret = PTR_ERR(pb->enable_gpio);
+		if (ret == -ENOENT)
+			pb->enable_gpio = NULL;
 		else
-			flags = GPIOF_OUT_INIT_LOW;
+			goto err_alloc;
+	}
 
-		ret = gpio_request_one(pb->enable_gpio, flags, "enable");
+	/*
+	 * Compatibility fallback for drivers still using the integer GPIO
+	 * platform data. Must go away soon.
+	 */
+	if (!pb->enable_gpio && gpio_is_valid(data->enable_gpio)) {
+		ret = devm_gpio_request_one(&pdev->dev, data->enable_gpio,
+					    GPIOF_OUT_INIT_HIGH, "enable");
 		if (ret < 0) {
 			dev_err(&pdev->dev, "failed to request GPIO#%d: %d\n",
-				pb->enable_gpio, ret);
+				data->enable_gpio, ret);
 			goto err_alloc;
 		}
+
+		pb->enable_gpio = gpio_to_desc(data->enable_gpio);
 	}
 
+	if (pb->enable_gpio)
+		gpiod_direction_output(pb->enable_gpio, 1);
+
 	pb->power_supply = devm_regulator_get(&pdev->dev, "power");
 	if (IS_ERR(pb->power_supply)) {
 		ret = PTR_ERR(pb->power_supply);
-		goto err_gpio;
+		goto err_alloc;
 	}
 
 	pb->pwm = devm_pwm_get(&pdev->dev, NULL);
@@ -295,7 +287,7 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 		if (IS_ERR(pb->pwm)) {
 			dev_err(&pdev->dev, "unable to request legacy PWM\n");
 			ret = PTR_ERR(pb->pwm);
-			goto err_gpio;
+			goto err_alloc;
 		}
 	}
 
@@ -304,12 +296,15 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	/*
 	 * The DT case will set the pwm_period_ns field to 0 and store the
 	 * period, parsed from the DT, in the PWM device. For the non-DT case,
-	 * set the period from platform data.
+	 * set the period from platform data if it has not already been set
+	 * via the PWM lookup table.
 	 */
-	if (data->pwm_period_ns > 0)
+	pb->period = pwm_get_period(pb->pwm);
+	if (!pb->period && (data->pwm_period_ns > 0)) {
+		pb->period = data->pwm_period_ns;
 		pwm_set_period(pb->pwm, data->pwm_period_ns);
+	}
 
-	pb->period = pwm_get_period(pb->pwm);
 	pb->lth_brightness = data->lth_brightness * (pb->period / pb->scale);
 
 	memset(&props, 0, sizeof(struct backlight_properties));
@@ -320,7 +315,7 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	if (IS_ERR(bl)) {
 		dev_err(&pdev->dev, "failed to register backlight\n");
 		ret = PTR_ERR(bl);
-		goto err_gpio;
+		goto err_alloc;
 	}
 
 	if (data->dft_brightness > data->max_brightness) {
@@ -336,9 +331,6 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, bl);
 	return 0;
 
-err_gpio:
-	if (gpio_is_valid(pb->enable_gpio))
-		gpio_free(pb->enable_gpio);
 err_alloc:
 	if (data->exit)
 		data->exit(&pdev->dev);
@@ -359,6 +351,14 @@ static int pwm_backlight_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static void pwm_backlight_shutdown(struct platform_device *pdev)
+{
+	struct backlight_device *bl = platform_get_drvdata(pdev);
+	struct pwm_bl_data *pb = bl_get_data(bl);
+
+	pwm_backlight_power_off(pb);
+}
+
 #ifdef CONFIG_PM_SLEEP
 static int pwm_backlight_suspend(struct device *dev)
 {
@@ -404,6 +404,7 @@ static struct platform_driver pwm_backlight_driver = {
 	},
 	.probe		= pwm_backlight_probe,
 	.remove		= pwm_backlight_remove,
+	.shutdown	= pwm_backlight_shutdown,
 };
 
 module_platform_driver(pwm_backlight_driver);
diff --git a/drivers/video/backlight/s6e63m0.c b/drivers/video/backlight/s6e63m0.c
index 510a1bcf76f1..2d6d48196c6d 100644
--- a/drivers/video/backlight/s6e63m0.c
+++ b/drivers/video/backlight/s6e63m0.c
@@ -703,7 +703,7 @@ static ssize_t s6e63m0_sysfs_show_gamma_table(struct device *dev,
 	struct s6e63m0 *lcd = dev_get_drvdata(dev);
 	char temp[3];
 
-	sprintf(temp, "%d\n", lcd->gamma_table_count);
+	sprintf(temp, "%u\n", lcd->gamma_table_count);
 	strcpy(buf, temp);
 
 	return strlen(buf);
diff --git a/drivers/video/console/dummycon.c b/drivers/video/console/dummycon.c
index b63860f7beab..40bec8d64b0a 100644
--- a/drivers/video/console/dummycon.c
+++ b/drivers/video/console/dummycon.c
@@ -77,3 +77,4 @@ const struct consw dummy_con = {
     .con_set_palette =	DUMMY,
     .con_scrolldelta =	DUMMY,
 };
+EXPORT_SYMBOL_GPL(dummy_con);
diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index f267284b423b..6e6aa704fe84 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -1441,5 +1441,6 @@ const struct consw vga_con = {
 	.con_build_attr = vgacon_build_attr,
 	.con_invert_region = vgacon_invert_region,
 };
+EXPORT_SYMBOL(vga_con);
 
 MODULE_LICENSE("GPL");
diff --git a/drivers/video/fbdev/atmel_lcdfb.c b/drivers/video/fbdev/atmel_lcdfb.c
index e683b6ef9594..d36e830d6fc6 100644
--- a/drivers/video/fbdev/atmel_lcdfb.c
+++ b/drivers/video/fbdev/atmel_lcdfb.c
@@ -1057,6 +1057,7 @@ static int atmel_lcdfb_of_init(struct atmel_lcdfb_info *sinfo)
 		goto put_display_node;
 	}
 
+	INIT_LIST_HEAD(&pdata->pwr_gpios);
 	ret = -ENOMEM;
 	for (i = 0; i < of_gpio_named_count(display_np, "atmel,power-control-gpio"); i++) {
 		gpio = of_get_named_gpio_flags(display_np, "atmel,power-control-gpio",
@@ -1082,6 +1083,7 @@ static int atmel_lcdfb_of_init(struct atmel_lcdfb_info *sinfo)
 			dev_err(dev, "set direction output gpio %d failed\n", gpio);
 			goto put_display_node;
 		}
+		list_add(&og->list, &pdata->pwr_gpios);
 	}
 
 	if (is_gpio_power)
diff --git a/drivers/video/fbdev/bfin_adv7393fb.c b/drivers/video/fbdev/bfin_adv7393fb.c
index a54f7f7d763b..8fe41caac38e 100644
--- a/drivers/video/fbdev/bfin_adv7393fb.c
+++ b/drivers/video/fbdev/bfin_adv7393fb.c
@@ -408,7 +408,7 @@ static int bfin_adv7393_fb_probe(struct i2c_client *client,
 	/* Workaround "PPI Does Not Start Properly In Specific Mode" */
 	if (ANOMALY_05000400) {
 		ret = gpio_request_one(P_IDENT(P_PPI0_FS3), GPIOF_OUT_INIT_LOW,
-					"PPI0_FS3")
+					"PPI0_FS3");
 		if (ret) {
 			dev_err(&client->dev, "PPI0_FS3 GPIO request failed\n");
 			ret = -EBUSY;
diff --git a/drivers/video/fbdev/offb.c b/drivers/video/fbdev/offb.c
index 7d44d669d5b6..43a0a52fc527 100644
--- a/drivers/video/fbdev/offb.c
+++ b/drivers/video/fbdev/offb.c
@@ -91,15 +91,6 @@ extern boot_infos_t *boot_infos;
 #define AVIVO_DC_LUTB_WHITE_OFFSET_GREEN        0x6cd4
 #define AVIVO_DC_LUTB_WHITE_OFFSET_RED          0x6cd8
 
-#define FB_RIGHT_POS(p, bpp)         (fb_be_math(p) ? 0 : (32 - (bpp)))
-
-static inline u32 offb_cmap_byteswap(struct fb_info *info, u32 value)
-{
-	u32 bpp = info->var.bits_per_pixel;
-
-	return cpu_to_be32(value) >> FB_RIGHT_POS(info, bpp);
-}
-
     /*
      *  Set a single color register. The values supplied are already
      *  rounded down to the hardware's capabilities (according to the
@@ -129,7 +120,7 @@ static int offb_setcolreg(u_int regno, u_int red, u_int green, u_int blue,
 			mask <<= info->var.transp.offset;
 			value |= mask;
 		}
-		pal[regno] = offb_cmap_byteswap(info, value);
+		pal[regno] = value;
 		return 0;
 	}
 
diff --git a/drivers/video/fbdev/omap2/dss/omapdss-boot-init.c b/drivers/video/fbdev/omap2/dss/omapdss-boot-init.c
index 99af9e88b2d8..2f0822ee3ff9 100644
--- a/drivers/video/fbdev/omap2/dss/omapdss-boot-init.c
+++ b/drivers/video/fbdev/omap2/dss/omapdss-boot-init.c
@@ -121,9 +121,11 @@ static void __init omapdss_add_to_list(struct device_node *node, bool root)
 {
 	struct dss_conv_node *n = kmalloc(sizeof(struct dss_conv_node),
 		GFP_KERNEL);
-	n->node = node;
-	n->root = root;
-	list_add(&n->list, &dss_conv_list);
+	if (n) {
+		n->node = node;
+		n->root = root;
+		list_add(&n->list, &dss_conv_list);
+	}
 }
 
 static bool __init omapdss_list_contains(const struct device_node *node)
diff --git a/drivers/video/fbdev/sm501fb.c b/drivers/video/fbdev/sm501fb.c
index 1501979099dc..c2c8eb668784 100644
--- a/drivers/video/fbdev/sm501fb.c
+++ b/drivers/video/fbdev/sm501fb.c
@@ -1215,7 +1215,7 @@ static ssize_t sm501fb_crtsrc_store(struct device *dev,
 }
 
 /* Prepare the device_attr for registration with sysfs later */
-static DEVICE_ATTR(crt_src, 0666, sm501fb_crtsrc_show, sm501fb_crtsrc_store);
+static DEVICE_ATTR(crt_src, 0664, sm501fb_crtsrc_show, sm501fb_crtsrc_store);
 
 /* sm501fb_show_regs
  *
diff --git a/drivers/video/fbdev/vt8500lcdfb.c b/drivers/video/fbdev/vt8500lcdfb.c
index a8f2b280f796..a1134c3f6c11 100644
--- a/drivers/video/fbdev/vt8500lcdfb.c
+++ b/drivers/video/fbdev/vt8500lcdfb.c
@@ -474,8 +474,6 @@ static int vt8500lcd_remove(struct platform_device *pdev)
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	release_mem_region(res->start, resource_size(res));
 
-	kfree(fbi);
-
 	return 0;
 }
 
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 1e443629f76d..4d08f45a9c29 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -865,4 +865,19 @@ bool virtqueue_is_broken(struct virtqueue *_vq)
 }
 EXPORT_SYMBOL_GPL(virtqueue_is_broken);
 
+/*
+ * This should prevent the device from being used, allowing drivers to
+ * recover.  You may need to grab appropriate locks to flush.
+ */
+void virtio_break_device(struct virtio_device *dev)
+{
+	struct virtqueue *_vq;
+
+	list_for_each_entry(_vq, &dev->vqs, list) {
+		struct vring_virtqueue *vq = to_vvq(_vq);
+		vq->broken = true;
+	}
+}
+EXPORT_SYMBOL_GPL(virtio_break_device);
+
 MODULE_LICENSE("GPL");
diff --git a/drivers/w1/masters/mxc_w1.c b/drivers/w1/masters/mxc_w1.c
index 67b067a3e2ab..a5df5e89d456 100644
--- a/drivers/w1/masters/mxc_w1.c
+++ b/drivers/w1/masters/mxc_w1.c
@@ -66,7 +66,7 @@ static u8 mxc_w1_ds2_reset_bus(void *data)
 
 		udelay(100);
 	}
-	return !!(reg_val & MXC_W1_CONTROL_PST);
+	return !(reg_val & MXC_W1_CONTROL_PST);
 }
 
 /*
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 74ec8fc5cc03..76dd54122f76 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -272,7 +272,7 @@ config PNX4008_WATCHDOG
 
 config IOP_WATCHDOG
 	tristate "IOP Watchdog"
-	depends on PLAT_IOP
+	depends on ARCH_IOP13XX
 	select WATCHDOG_NOWAYOUT if (ARCH_IOP32X || ARCH_IOP33X)
 	help
 	  Say Y here if to include support for the watchdog timer
@@ -378,6 +378,8 @@ config MAX63XX_WATCHDOG
 config IMX2_WDT
 	tristate "IMX2+ Watchdog"
 	depends on ARCH_MXC
+	select REGMAP_MMIO
+	select WATCHDOG_CORE
 	help
 	  This is the driver for the hardware watchdog
 	  on the Freescale IMX2 and later processors.
@@ -663,6 +665,19 @@ config INTEL_SCU_WATCHDOG
 
 	  To compile this driver as a module, choose M here.
 
+config INTEL_MID_WATCHDOG
+	tristate "Intel MID Watchdog Timer"
+	depends on X86_INTEL_MID
+	select WATCHDOG_CORE
+	---help---
+	  Watchdog timer driver built into the Intel SCU for Intel MID
+	  Platforms.
+
+	  This driver currently supports only the watchdog evolution
+	  implementation in SCU, available for Merrifield generation.
+
+	  To compile this driver as a module, choose M here.
+
 config ITCO_WDT
 	tristate "Intel TCO Timer/Watchdog"
 	depends on (X86 || IA64) && PCI
@@ -835,7 +850,7 @@ config 60XX_WDT
 
 config SBC8360_WDT
 	tristate "SBC8360 Watchdog Timer"
-	depends on X86
+	depends on X86_32
 	---help---
 
 	  This is the driver for the hardware watchdog on the SBC8360 Single
@@ -938,36 +953,6 @@ config W83627HF_WDT
 
 	  Most people will say N.
 
-config W83697HF_WDT
-	tristate "W83697HF/W83697HG Watchdog Timer"
-	depends on X86
-	---help---
-	  This is the driver for the hardware watchdog on the W83697HF/HG
-	  chipset as used in Dedibox/VIA motherboards (and likely others).
-	  This watchdog simply watches your kernel to make sure it doesn't
-	  freeze, and if it does, it reboots your computer after a certain
-	  amount of time.
-
-	  To compile this driver as a module, choose M here: the
-	  module will be called w83697hf_wdt.
-
-	  Most people will say N.
-
-config W83697UG_WDT
-	tristate "W83697UG/W83697UF Watchdog Timer"
-	depends on X86
-	---help---
-	  This is the driver for the hardware watchdog on the W83697UG/UF
-	  chipset as used in MSI Fuzzy CX700 VIA motherboards (and likely others).
-	  This watchdog simply watches your kernel to make sure it doesn't
-	  freeze, and if it does, it reboots your computer after a certain
-	  amount of time.
-
-	  To compile this driver as a module, choose M here: the
-	  module will be called w83697ug_wdt.
-
-	  Most people will say N.
-
 config W83877F_WDT
 	tristate "W83877F (EMACS) Watchdog Timer"
 	depends on X86
@@ -1295,14 +1280,17 @@ config WATCHDOG_RTAS
 
 # S390 Architecture
 
-config ZVM_WATCHDOG
-	tristate "z/VM Watchdog Timer"
+config DIAG288_WATCHDOG
+	tristate "System z diag288 Watchdog"
 	depends on S390
+	select WATCHDOG_CORE
 	help
 	  IBM s/390 and zSeries machines running under z/VM 5.1 or later
 	  provide a virtual watchdog timer to their guest that cause a
 	  user define Control Program command to be executed after a
 	  timeout.
+	  LPAR provides a very similar interface. This driver handles
+	  both.
 
 	  To compile this driver as a module, choose M here. The module
 	  will be called vmwatchdog.
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 1b5f3d5efad5..468c3204c3b1 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -107,13 +107,12 @@ obj-$(CONFIG_SMSC_SCH311X_WDT) += sch311x_wdt.o
 obj-$(CONFIG_SMSC37B787_WDT) += smsc37b787_wdt.o
 obj-$(CONFIG_VIA_WDT) += via_wdt.o
 obj-$(CONFIG_W83627HF_WDT) += w83627hf_wdt.o
-obj-$(CONFIG_W83697HF_WDT) += w83697hf_wdt.o
-obj-$(CONFIG_W83697UG_WDT) += w83697ug_wdt.o
 obj-$(CONFIG_W83877F_WDT) += w83877f_wdt.o
 obj-$(CONFIG_W83977F_WDT) += w83977f_wdt.o
 obj-$(CONFIG_MACHZ_WDT) += machzwd.o
 obj-$(CONFIG_SBC_EPX_C3_WATCHDOG) += sbc_epx_c3.o
 obj-$(CONFIG_INTEL_SCU_WATCHDOG) += intel_scu_watchdog.o
+obj-$(CONFIG_INTEL_MID_WATCHDOG) += intel-mid_wdt.o
 
 # M32R Architecture
 
@@ -154,6 +153,7 @@ obj-$(CONFIG_MEN_A21_WDT) += mena21_wdt.o
 obj-$(CONFIG_WATCHDOG_RTAS) += wdrtas.o
 
 # S390 Architecture
+obj-$(CONFIG_DIAG288_WATCHDOG) += diag288_wdt.o
 
 # SUPERH (sh + sh64) Architecture
 obj-$(CONFIG_SH_WDT) += shwdt.o
diff --git a/drivers/watchdog/ath79_wdt.c b/drivers/watchdog/ath79_wdt.c
index 399c3fddecf6..41ac4660fb89 100644
--- a/drivers/watchdog/ath79_wdt.c
+++ b/drivers/watchdog/ath79_wdt.c
@@ -20,6 +20,7 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/bitops.h>
+#include <linux/delay.h>
 #include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/io.h>
@@ -90,6 +91,15 @@ static inline void ath79_wdt_keepalive(void)
 static inline void ath79_wdt_enable(void)
 {
 	ath79_wdt_keepalive();
+
+	/*
+	 * Updating the TIMER register requires a few microseconds
+	 * on the AR934x SoCs at least. Use a small delay to ensure
+	 * that the TIMER register is updated within the hardware
+	 * before enabling the watchdog.
+	 */
+	udelay(2);
+
 	ath79_wdt_wr(WDOG_REG_CTRL, WDOG_CTRL_ACTION_FCR);
 	/* flush write */
 	ath79_wdt_rr(WDOG_REG_CTRL);
@@ -255,7 +265,7 @@ static int ath79_wdt_probe(struct platform_device *pdev)
 	if (IS_ERR(wdt_clk))
 		return PTR_ERR(wdt_clk);
 
-	err = clk_enable(wdt_clk);
+	err = clk_prepare_enable(wdt_clk);
 	if (err)
 		return err;
 
@@ -286,14 +296,14 @@ static int ath79_wdt_probe(struct platform_device *pdev)
 	return 0;
 
 err_clk_disable:
-	clk_disable(wdt_clk);
+	clk_disable_unprepare(wdt_clk);
 	return err;
 }
 
 static int ath79_wdt_remove(struct platform_device *pdev)
 {
 	misc_deregister(&ath79_wdt_miscdev);
-	clk_disable(wdt_clk);
+	clk_disable_unprepare(wdt_clk);
 	return 0;
 }
 
diff --git a/drivers/watchdog/booke_wdt.c b/drivers/watchdog/booke_wdt.c
index a8dbceb32914..08a785398eac 100644
--- a/drivers/watchdog/booke_wdt.c
+++ b/drivers/watchdog/booke_wdt.c
@@ -41,6 +41,28 @@ u32 booke_wdt_period = CONFIG_BOOKE_WDT_DEFAULT_TIMEOUT;
 #define WDTP_MASK	(TCR_WP_MASK)
 #endif
 
+/* Checks wdt=x and wdt_period=xx command-line option */
+notrace int __init early_parse_wdt(char *p)
+{
+	if (p && strncmp(p, "0", 1) != 0)
+		booke_wdt_enabled = 1;
+
+	return 0;
+}
+early_param("wdt", early_parse_wdt);
+
+int __init early_parse_wdt_period(char *p)
+{
+	unsigned long ret;
+	if (p) {
+		if (!kstrtol(p, 0, &ret))
+			booke_wdt_period = ret;
+	}
+
+	return 0;
+}
+early_param("wdt_period", early_parse_wdt_period);
+
 #ifdef CONFIG_PPC_FSL_BOOK3E
 
 /* For the specified period, determine the number of seconds
@@ -103,17 +125,18 @@ static unsigned int sec_to_period(unsigned int secs)
 static void __booke_wdt_set(void *data)
 {
 	u32 val;
+	struct watchdog_device *wdog = data;
 
 	val = mfspr(SPRN_TCR);
 	val &= ~WDTP_MASK;
-	val |= WDTP(booke_wdt_period);
+	val |= WDTP(sec_to_period(wdog->timeout));
 
 	mtspr(SPRN_TCR, val);
 }
 
-static void booke_wdt_set(void)
+static void booke_wdt_set(void *data)
 {
-	on_each_cpu(__booke_wdt_set, NULL, 0);
+	on_each_cpu(__booke_wdt_set, data, 0);
 }
 
 static void __booke_wdt_ping(void *data)
@@ -131,12 +154,13 @@ static int booke_wdt_ping(struct watchdog_device *wdog)
 static void __booke_wdt_enable(void *data)
 {
 	u32 val;
+	struct watchdog_device *wdog = data;
 
 	/* clear status before enabling watchdog */
 	__booke_wdt_ping(NULL);
 	val = mfspr(SPRN_TCR);
 	val &= ~WDTP_MASK;
-	val |= (TCR_WIE|TCR_WRC(WRC_CHIP)|WDTP(booke_wdt_period));
+	val |= (TCR_WIE|TCR_WRC(WRC_CHIP)|WDTP(sec_to_period(wdog->timeout)));
 
 	mtspr(SPRN_TCR, val);
 }
@@ -162,25 +186,17 @@ static void __booke_wdt_disable(void *data)
 
 }
 
-static void __booke_wdt_start(struct watchdog_device *wdog)
+static int booke_wdt_start(struct watchdog_device *wdog)
 {
-	on_each_cpu(__booke_wdt_enable, NULL, 0);
+	on_each_cpu(__booke_wdt_enable, wdog, 0);
 	pr_debug("watchdog enabled (timeout = %u sec)\n", wdog->timeout);
-}
 
-static int booke_wdt_start(struct watchdog_device *wdog)
-{
-	if (booke_wdt_enabled == 0) {
-		booke_wdt_enabled = 1;
-		__booke_wdt_start(wdog);
-	}
 	return 0;
 }
 
 static int booke_wdt_stop(struct watchdog_device *wdog)
 {
 	on_each_cpu(__booke_wdt_disable, NULL, 0);
-	booke_wdt_enabled = 0;
 	pr_debug("watchdog disabled\n");
 
 	return 0;
@@ -191,9 +207,8 @@ static int booke_wdt_set_timeout(struct watchdog_device *wdt_dev,
 {
 	if (timeout > MAX_WDT_TIMEOUT)
 		return -EINVAL;
-	booke_wdt_period = sec_to_period(timeout);
 	wdt_dev->timeout = timeout;
-	booke_wdt_set();
+	booke_wdt_set(wdt_dev);
 
 	return 0;
 }
@@ -231,10 +246,10 @@ static int __init booke_wdt_init(void)
 	pr_info("powerpc book-e watchdog driver loaded\n");
 	booke_wdt_info.firmware_version = cur_cpu_spec->pvr_value;
 	booke_wdt_set_timeout(&booke_wdt_dev,
-			      period_to_sec(CONFIG_BOOKE_WDT_DEFAULT_TIMEOUT));
+			      period_to_sec(booke_wdt_period));
 	watchdog_set_nowayout(&booke_wdt_dev, nowayout);
 	if (booke_wdt_enabled)
-		__booke_wdt_start(&booke_wdt_dev);
+		booke_wdt_start(&booke_wdt_dev);
 
 	ret = watchdog_register_device(&booke_wdt_dev);
 
diff --git a/drivers/watchdog/diag288_wdt.c b/drivers/watchdog/diag288_wdt.c
new file mode 100644
index 000000000000..429494b6c822
--- /dev/null
+++ b/drivers/watchdog/diag288_wdt.c
@@ -0,0 +1,316 @@
+/*
+ * Watchdog driver for z/VM and LPAR using the diag 288 interface.
+ *
+ * Under z/VM, expiration of the watchdog will send a "system restart" command
+ * to CP.
+ *
+ * The command can be altered using the module parameter "cmd". This is
+ * not recommended because it's only supported on z/VM but not whith LPAR.
+ *
+ * On LPAR, the watchdog will always trigger a system restart. the module
+ * paramter cmd is meaningless here.
+ *
+ *
+ * Copyright IBM Corp. 2004, 2013
+ * Author(s): Arnd Bergmann (arndb@de.ibm.com)
+ *	      Philipp Hachtmann (phacht@de.ibm.com)
+ *
+ */
+
+#define KMSG_COMPONENT "diag288_wdt"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/slab.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/suspend.h>
+#include <asm/ebcdic.h>
+#include <linux/io.h>
+#include <linux/uaccess.h>
+
+#define MAX_CMDLEN 240
+#define DEFAULT_CMD "SYSTEM RESTART"
+
+#define MIN_INTERVAL 15     /* Minimal time supported by diag88 */
+#define MAX_INTERVAL 3600   /* One hour should be enough - pure estimation */
+
+#define WDT_DEFAULT_TIMEOUT 30
+
+/* Function codes - init, change, cancel */
+#define WDT_FUNC_INIT 0
+#define WDT_FUNC_CHANGE 1
+#define WDT_FUNC_CANCEL 2
+#define WDT_FUNC_CONCEAL 0x80000000
+
+/* Action codes for LPAR watchdog */
+#define LPARWDT_RESTART 0
+
+static char wdt_cmd[MAX_CMDLEN] = DEFAULT_CMD;
+static bool conceal_on;
+static bool nowayout_info = WATCHDOG_NOWAYOUT;
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Arnd Bergmann <arndb@de.ibm.com>");
+MODULE_AUTHOR("Philipp Hachtmann <phacht@de.ibm.com>");
+
+MODULE_DESCRIPTION("System z diag288  Watchdog Timer");
+
+module_param_string(cmd, wdt_cmd, MAX_CMDLEN, 0644);
+MODULE_PARM_DESC(cmd, "CP command that is run when the watchdog triggers (z/VM only)");
+
+module_param_named(conceal, conceal_on, bool, 0644);
+MODULE_PARM_DESC(conceal, "Enable the CONCEAL CP option while the watchdog is active (z/VM only)");
+
+module_param_named(nowayout, nowayout_info, bool, 0444);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default = CONFIG_WATCHDOG_NOWAYOUT)");
+
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
+MODULE_ALIAS("vmwatchdog");
+
+static int __diag288(unsigned int func, unsigned int timeout,
+		     unsigned long action, unsigned int len)
+{
+	register unsigned long __func asm("2") = func;
+	register unsigned long __timeout asm("3") = timeout;
+	register unsigned long __action asm("4") = action;
+	register unsigned long __len asm("5") = len;
+	int err;
+
+	err = -EINVAL;
+	asm volatile(
+		"	diag	%1, %3, 0x288\n"
+		"0:	la	%0, 0\n"
+		"1:\n"
+		EX_TABLE(0b, 1b)
+		: "+d" (err) : "d"(__func), "d"(__timeout),
+		  "d"(__action), "d"(__len) : "1", "cc");
+	return err;
+}
+
+static int __diag288_vm(unsigned int  func, unsigned int timeout,
+			char *cmd, size_t len)
+{
+	return __diag288(func, timeout, virt_to_phys(cmd), len);
+}
+
+static int __diag288_lpar(unsigned int func, unsigned int timeout,
+			  unsigned long action)
+{
+	return __diag288(func, timeout, action, 0);
+}
+
+static int wdt_start(struct watchdog_device *dev)
+{
+	char *ebc_cmd;
+	size_t len;
+	int ret;
+	unsigned int func;
+
+	ret = -ENODEV;
+
+	if (MACHINE_IS_VM) {
+		ebc_cmd = kmalloc(MAX_CMDLEN, GFP_KERNEL);
+		if (!ebc_cmd)
+			return -ENOMEM;
+		len = strlcpy(ebc_cmd, wdt_cmd, MAX_CMDLEN);
+		ASCEBC(ebc_cmd, MAX_CMDLEN);
+		EBC_TOUPPER(ebc_cmd, MAX_CMDLEN);
+
+		func = conceal_on ? (WDT_FUNC_INIT | WDT_FUNC_CONCEAL)
+			: WDT_FUNC_INIT;
+		ret = __diag288_vm(func, dev->timeout, ebc_cmd, len);
+		WARN_ON(ret != 0);
+		kfree(ebc_cmd);
+	}
+
+	if (MACHINE_IS_LPAR) {
+		ret = __diag288_lpar(WDT_FUNC_INIT,
+				     dev->timeout, LPARWDT_RESTART);
+	}
+
+	if (ret) {
+		pr_err("The watchdog cannot be activated\n");
+		return ret;
+	}
+	pr_info("The watchdog was activated\n");
+	return 0;
+}
+
+static int wdt_stop(struct watchdog_device *dev)
+{
+	int ret;
+
+	ret = __diag288(WDT_FUNC_CANCEL, 0, 0, 0);
+	pr_info("The watchdog was deactivated\n");
+	return ret;
+}
+
+static int wdt_ping(struct watchdog_device *dev)
+{
+	char *ebc_cmd;
+	size_t len;
+	int ret;
+	unsigned int func;
+
+	ret = -ENODEV;
+
+	if (MACHINE_IS_VM) {
+		ebc_cmd = kmalloc(MAX_CMDLEN, GFP_KERNEL);
+		if (!ebc_cmd)
+			return -ENOMEM;
+		len = strlcpy(ebc_cmd, wdt_cmd, MAX_CMDLEN);
+		ASCEBC(ebc_cmd, MAX_CMDLEN);
+		EBC_TOUPPER(ebc_cmd, MAX_CMDLEN);
+
+		/*
+		 * It seems to be ok to z/VM to use the init function to
+		 * retrigger the watchdog. On LPAR WDT_FUNC_CHANGE must
+		 * be used when the watchdog is running.
+		 */
+		func = conceal_on ? (WDT_FUNC_INIT | WDT_FUNC_CONCEAL)
+			: WDT_FUNC_INIT;
+
+		ret = __diag288_vm(func, dev->timeout, ebc_cmd, len);
+		WARN_ON(ret != 0);
+		kfree(ebc_cmd);
+	}
+
+	if (MACHINE_IS_LPAR)
+		ret = __diag288_lpar(WDT_FUNC_CHANGE, dev->timeout, 0);
+
+	if (ret)
+		pr_err("The watchdog timer cannot be started or reset\n");
+	return ret;
+}
+
+static int wdt_set_timeout(struct watchdog_device * dev, unsigned int new_to)
+{
+	dev->timeout = new_to;
+	return wdt_ping(dev);
+}
+
+static struct watchdog_ops wdt_ops = {
+	.owner = THIS_MODULE,
+	.start = wdt_start,
+	.stop = wdt_stop,
+	.ping = wdt_ping,
+	.set_timeout = wdt_set_timeout,
+};
+
+static struct watchdog_info wdt_info = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE,
+	.firmware_version = 0,
+	.identity = "z Watchdog",
+};
+
+static struct watchdog_device wdt_dev = {
+	.parent = NULL,
+	.info = &wdt_info,
+	.ops = &wdt_ops,
+	.bootstatus = 0,
+	.timeout = WDT_DEFAULT_TIMEOUT,
+	.min_timeout = MIN_INTERVAL,
+	.max_timeout = MAX_INTERVAL,
+};
+
+/*
+ * It makes no sense to go into suspend while the watchdog is running.
+ * Depending on the memory size, the watchdog might trigger, while we
+ * are still saving the memory.
+ * We reuse the open flag to ensure that suspend and watchdog open are
+ * exclusive operations
+ */
+static int wdt_suspend(void)
+{
+	if (test_and_set_bit(WDOG_DEV_OPEN, &wdt_dev.status)) {
+		pr_err("Linux cannot be suspended while the watchdog is in use\n");
+		return notifier_from_errno(-EBUSY);
+	}
+	if (test_bit(WDOG_ACTIVE, &wdt_dev.status)) {
+		clear_bit(WDOG_DEV_OPEN, &wdt_dev.status);
+		pr_err("Linux cannot be suspended while the watchdog is in use\n");
+		return notifier_from_errno(-EBUSY);
+	}
+	return NOTIFY_DONE;
+}
+
+static int wdt_resume(void)
+{
+	clear_bit(WDOG_DEV_OPEN, &wdt_dev.status);
+	return NOTIFY_DONE;
+}
+
+static int wdt_power_event(struct notifier_block *this, unsigned long event,
+			   void *ptr)
+{
+	switch (event) {
+	case PM_POST_HIBERNATION:
+	case PM_POST_SUSPEND:
+		return wdt_resume();
+	case PM_HIBERNATION_PREPARE:
+	case PM_SUSPEND_PREPARE:
+		return wdt_suspend();
+	default:
+		return NOTIFY_DONE;
+	}
+}
+
+static struct notifier_block wdt_power_notifier = {
+	.notifier_call = wdt_power_event,
+};
+
+static int __init diag288_init(void)
+{
+	int ret;
+	char ebc_begin[] = {
+		194, 197, 199, 201, 213
+	};
+
+	watchdog_set_nowayout(&wdt_dev, nowayout_info);
+
+	if (MACHINE_IS_VM) {
+		pr_info("The watchdog device driver detected a z/VM environment\n");
+		if (__diag288_vm(WDT_FUNC_INIT, 15,
+				 ebc_begin, sizeof(ebc_begin)) != 0) {
+			pr_err("The watchdog cannot be initialized\n");
+			return -EINVAL;
+		}
+	} else if (MACHINE_IS_LPAR) {
+		pr_info("The watchdog device driver detected an LPAR environment\n");
+		if (__diag288_lpar(WDT_FUNC_INIT, 30, LPARWDT_RESTART)) {
+			pr_err("The watchdog cannot be initialized\n");
+			return -EINVAL;
+		}
+	} else {
+		pr_err("Linux runs in an environment that does not support the diag288 watchdog\n");
+		return -ENODEV;
+	}
+
+	if (__diag288_lpar(WDT_FUNC_CANCEL, 0, 0)) {
+		pr_err("The watchdog cannot be deactivated\n");
+		return -EINVAL;
+	}
+
+	ret = register_pm_notifier(&wdt_power_notifier);
+	if (ret)
+		return ret;
+
+	ret = watchdog_register_device(&wdt_dev);
+	if (ret)
+		unregister_pm_notifier(&wdt_power_notifier);
+
+	return ret;
+}
+
+static void __exit diag288_exit(void)
+{
+	watchdog_unregister_device(&wdt_dev);
+	unregister_pm_notifier(&wdt_power_notifier);
+}
+
+module_init(diag288_init);
+module_exit(diag288_exit);
diff --git a/drivers/watchdog/imx2_wdt.c b/drivers/watchdog/imx2_wdt.c
index dd51d9539b33..9d4874f09948 100644
--- a/drivers/watchdog/imx2_wdt.c
+++ b/drivers/watchdog/imx2_wdt.c
@@ -21,19 +21,17 @@
  * Halt on suspend:	Manual		Can be automatic
  */
 
+#include <linux/clk.h>
 #include <linux/init.h>
+#include <linux/io.h>
+#include <linux/jiffies.h>
 #include <linux/kernel.h>
-#include <linux/miscdevice.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/platform_device.h>
-#include <linux/watchdog.h>
-#include <linux/clk.h>
-#include <linux/fs.h>
-#include <linux/io.h>
-#include <linux/uaccess.h>
+#include <linux/regmap.h>
 #include <linux/timer.h>
-#include <linux/jiffies.h>
+#include <linux/watchdog.h>
 
 #define DRIVER_NAME "imx2-wdt"
 
@@ -55,19 +53,12 @@
 
 #define WDOG_SEC_TO_COUNT(s)	((s * 2 - 1) << 8)
 
-#define IMX2_WDT_STATUS_OPEN	0
-#define IMX2_WDT_STATUS_STARTED	1
-#define IMX2_WDT_EXPECT_CLOSE	2
-
-static struct {
+struct imx2_wdt_device {
 	struct clk *clk;
-	void __iomem *base;
-	unsigned timeout;
-	unsigned long status;
+	struct regmap *regmap;
 	struct timer_list timer;	/* Pings the watchdog when closed */
-} imx2_wdt;
-
-static struct miscdevice imx2_wdt_miscdev;
+	struct watchdog_device wdog;
+};
 
 static bool nowayout = WATCHDOG_NOWAYOUT;
 module_param(nowayout, bool, 0);
@@ -85,9 +76,12 @@ static const struct watchdog_info imx2_wdt_info = {
 	.options = WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE,
 };
 
-static inline void imx2_wdt_setup(void)
+static inline void imx2_wdt_setup(struct watchdog_device *wdog)
 {
-	u16 val = __raw_readw(imx2_wdt.base + IMX2_WDT_WCR);
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
+	u32 val;
+
+	regmap_read(wdev->regmap, IMX2_WDT_WCR, &val);
 
 	/* Suspend timer in low power mode, write once-only */
 	val |= IMX2_WDT_WCR_WDZST;
@@ -98,227 +92,199 @@ static inline void imx2_wdt_setup(void)
 	/* Keep Watchdog Disabled */
 	val &= ~IMX2_WDT_WCR_WDE;
 	/* Set the watchdog's Time-Out value */
-	val |= WDOG_SEC_TO_COUNT(imx2_wdt.timeout);
+	val |= WDOG_SEC_TO_COUNT(wdog->timeout);
 
-	__raw_writew(val, imx2_wdt.base + IMX2_WDT_WCR);
+	regmap_write(wdev->regmap, IMX2_WDT_WCR, val);
 
 	/* enable the watchdog */
 	val |= IMX2_WDT_WCR_WDE;
-	__raw_writew(val, imx2_wdt.base + IMX2_WDT_WCR);
+	regmap_write(wdev->regmap, IMX2_WDT_WCR, val);
 }
 
-static inline void imx2_wdt_ping(void)
+static inline bool imx2_wdt_is_running(struct imx2_wdt_device *wdev)
 {
-	__raw_writew(IMX2_WDT_SEQ1, imx2_wdt.base + IMX2_WDT_WSR);
-	__raw_writew(IMX2_WDT_SEQ2, imx2_wdt.base + IMX2_WDT_WSR);
-}
+	u32 val;
 
-static void imx2_wdt_timer_ping(unsigned long arg)
-{
-	/* ping it every imx2_wdt.timeout / 2 seconds to prevent reboot */
-	imx2_wdt_ping();
-	mod_timer(&imx2_wdt.timer, jiffies + imx2_wdt.timeout * HZ / 2);
+	regmap_read(wdev->regmap, IMX2_WDT_WCR, &val);
+
+	return val & IMX2_WDT_WCR_WDE;
 }
 
-static void imx2_wdt_start(void)
+static int imx2_wdt_ping(struct watchdog_device *wdog)
 {
-	if (!test_and_set_bit(IMX2_WDT_STATUS_STARTED, &imx2_wdt.status)) {
-		/* at our first start we enable clock and do initialisations */
-		clk_prepare_enable(imx2_wdt.clk);
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
 
-		imx2_wdt_setup();
-	} else	/* delete the timer that pings the watchdog after close */
-		del_timer_sync(&imx2_wdt.timer);
-
-	/* Watchdog is enabled - time to reload the timeout value */
-	imx2_wdt_ping();
+	regmap_write(wdev->regmap, IMX2_WDT_WSR, IMX2_WDT_SEQ1);
+	regmap_write(wdev->regmap, IMX2_WDT_WSR, IMX2_WDT_SEQ2);
+	return 0;
 }
 
-static void imx2_wdt_stop(void)
+static void imx2_wdt_timer_ping(unsigned long arg)
 {
-	/* we don't need a clk_disable, it cannot be disabled once started.
-	 * We use a timer to ping the watchdog while /dev/watchdog is closed */
-	imx2_wdt_timer_ping(0);
+	struct watchdog_device *wdog = (struct watchdog_device *)arg;
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
+
+	/* ping it every wdog->timeout / 2 seconds to prevent reboot */
+	imx2_wdt_ping(wdog);
+	mod_timer(&wdev->timer, jiffies + wdog->timeout * HZ / 2);
 }
 
-static void imx2_wdt_set_timeout(int new_timeout)
+static int imx2_wdt_set_timeout(struct watchdog_device *wdog,
+				unsigned int new_timeout)
 {
-	u16 val = __raw_readw(imx2_wdt.base + IMX2_WDT_WCR);
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
 
-	/* set the new timeout value in the WSR */
-	val &= ~IMX2_WDT_WCR_WT;
-	val |= WDOG_SEC_TO_COUNT(new_timeout);
-	__raw_writew(val, imx2_wdt.base + IMX2_WDT_WCR);
+	regmap_update_bits(wdev->regmap, IMX2_WDT_WCR, IMX2_WDT_WCR_WT,
+			   WDOG_SEC_TO_COUNT(new_timeout));
+	return 0;
 }
 
-static int imx2_wdt_open(struct inode *inode, struct file *file)
+static int imx2_wdt_start(struct watchdog_device *wdog)
 {
-	if (test_and_set_bit(IMX2_WDT_STATUS_OPEN, &imx2_wdt.status))
-		return -EBUSY;
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
+
+	if (imx2_wdt_is_running(wdev)) {
+		/* delete the timer that pings the watchdog after close */
+		del_timer_sync(&wdev->timer);
+		imx2_wdt_set_timeout(wdog, wdog->timeout);
+	} else
+		imx2_wdt_setup(wdog);
 
-	imx2_wdt_start();
-	return nonseekable_open(inode, file);
+	return imx2_wdt_ping(wdog);
 }
 
-static int imx2_wdt_close(struct inode *inode, struct file *file)
+static int imx2_wdt_stop(struct watchdog_device *wdog)
 {
-	if (test_bit(IMX2_WDT_EXPECT_CLOSE, &imx2_wdt.status) && !nowayout)
-		imx2_wdt_stop();
-	else {
-		dev_crit(imx2_wdt_miscdev.parent,
-			"Unexpected close: Expect reboot!\n");
-		imx2_wdt_ping();
-	}
-
-	clear_bit(IMX2_WDT_EXPECT_CLOSE, &imx2_wdt.status);
-	clear_bit(IMX2_WDT_STATUS_OPEN, &imx2_wdt.status);
+	/*
+	 * We don't need a clk_disable, it cannot be disabled once started.
+	 * We use a timer to ping the watchdog while /dev/watchdog is closed
+	 */
+	imx2_wdt_timer_ping((unsigned long)wdog);
 	return 0;
 }
 
-static long imx2_wdt_ioctl(struct file *file, unsigned int cmd,
-							unsigned long arg)
+static inline void imx2_wdt_ping_if_active(struct watchdog_device *wdog)
 {
-	void __user *argp = (void __user *)arg;
-	int __user *p = argp;
-	int new_value;
-	u16 val;
-
-	switch (cmd) {
-	case WDIOC_GETSUPPORT:
-		return copy_to_user(argp, &imx2_wdt_info,
-			sizeof(struct watchdog_info)) ? -EFAULT : 0;
-
-	case WDIOC_GETSTATUS:
-		return put_user(0, p);
-
-	case WDIOC_GETBOOTSTATUS:
-		val = __raw_readw(imx2_wdt.base + IMX2_WDT_WRSR);
-		new_value = val & IMX2_WDT_WRSR_TOUT ? WDIOF_CARDRESET : 0;
-		return put_user(new_value, p);
-
-	case WDIOC_KEEPALIVE:
-		imx2_wdt_ping();
-		return 0;
-
-	case WDIOC_SETTIMEOUT:
-		if (get_user(new_value, p))
-			return -EFAULT;
-		if ((new_value < 1) || (new_value > IMX2_WDT_MAX_TIME))
-			return -EINVAL;
-		imx2_wdt_set_timeout(new_value);
-		imx2_wdt.timeout = new_value;
-		imx2_wdt_ping();
-
-		/* Fallthrough to return current value */
-	case WDIOC_GETTIMEOUT:
-		return put_user(imx2_wdt.timeout, p);
-
-	default:
-		return -ENOTTY;
-	}
-}
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
 
-static ssize_t imx2_wdt_write(struct file *file, const char __user *data,
-						size_t len, loff_t *ppos)
-{
-	size_t i;
-	char c;
-
-	if (len == 0)	/* Can we see this even ? */
-		return 0;
-
-	clear_bit(IMX2_WDT_EXPECT_CLOSE, &imx2_wdt.status);
-	/* scan to see whether or not we got the magic character */
-	for (i = 0; i != len; i++) {
-		if (get_user(c, data + i))
-			return -EFAULT;
-		if (c == 'V')
-			set_bit(IMX2_WDT_EXPECT_CLOSE, &imx2_wdt.status);
+	if (imx2_wdt_is_running(wdev)) {
+		imx2_wdt_set_timeout(wdog, wdog->timeout);
+		imx2_wdt_timer_ping((unsigned long)wdog);
 	}
-
-	imx2_wdt_ping();
-	return len;
 }
 
-static const struct file_operations imx2_wdt_fops = {
+static struct watchdog_ops imx2_wdt_ops = {
 	.owner = THIS_MODULE,
-	.llseek = no_llseek,
-	.unlocked_ioctl = imx2_wdt_ioctl,
-	.open = imx2_wdt_open,
-	.release = imx2_wdt_close,
-	.write = imx2_wdt_write,
+	.start = imx2_wdt_start,
+	.stop = imx2_wdt_stop,
+	.ping = imx2_wdt_ping,
+	.set_timeout = imx2_wdt_set_timeout,
 };
 
-static struct miscdevice imx2_wdt_miscdev = {
-	.minor = WATCHDOG_MINOR,
-	.name = "watchdog",
-	.fops = &imx2_wdt_fops,
+static struct regmap_config imx2_wdt_regmap_config = {
+	.reg_bits = 16,
+	.reg_stride = 2,
+	.val_bits = 16,
+	.max_register = 0x8,
 };
 
 static int __init imx2_wdt_probe(struct platform_device *pdev)
 {
-	int ret;
+	struct imx2_wdt_device *wdev;
+	struct watchdog_device *wdog;
 	struct resource *res;
+	void __iomem *base;
+	int ret;
+	u32 val;
+
+	wdev = devm_kzalloc(&pdev->dev, sizeof(*wdev), GFP_KERNEL);
+	if (!wdev)
+		return -ENOMEM;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	imx2_wdt.base = devm_ioremap_resource(&pdev->dev, res);
-	if (IS_ERR(imx2_wdt.base))
-		return PTR_ERR(imx2_wdt.base);
+	base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(base))
+		return PTR_ERR(base);
+
+	wdev->regmap = devm_regmap_init_mmio_clk(&pdev->dev, NULL, base,
+						 &imx2_wdt_regmap_config);
+	if (IS_ERR(wdev->regmap)) {
+		dev_err(&pdev->dev, "regmap init failed\n");
+		return PTR_ERR(wdev->regmap);
+	}
 
-	imx2_wdt.clk = devm_clk_get(&pdev->dev, NULL);
-	if (IS_ERR(imx2_wdt.clk)) {
+	wdev->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(wdev->clk)) {
 		dev_err(&pdev->dev, "can't get Watchdog clock\n");
-		return PTR_ERR(imx2_wdt.clk);
+		return PTR_ERR(wdev->clk);
 	}
 
-	imx2_wdt.timeout = clamp_t(unsigned, timeout, 1, IMX2_WDT_MAX_TIME);
-	if (imx2_wdt.timeout != timeout)
-		dev_warn(&pdev->dev, "Initial timeout out of range! "
-			"Clamped from %u to %u\n", timeout, imx2_wdt.timeout);
+	wdog			= &wdev->wdog;
+	wdog->info		= &imx2_wdt_info;
+	wdog->ops		= &imx2_wdt_ops;
+	wdog->min_timeout	= 1;
+	wdog->max_timeout	= IMX2_WDT_MAX_TIME;
 
-	setup_timer(&imx2_wdt.timer, imx2_wdt_timer_ping, 0);
+	clk_prepare_enable(wdev->clk);
 
-	imx2_wdt_miscdev.parent = &pdev->dev;
-	ret = misc_register(&imx2_wdt_miscdev);
-	if (ret)
-		goto fail;
+	regmap_read(wdev->regmap, IMX2_WDT_WRSR, &val);
+	wdog->bootstatus = val & IMX2_WDT_WRSR_TOUT ? WDIOF_CARDRESET : 0;
 
-	dev_info(&pdev->dev,
-		"IMX2+ Watchdog Timer enabled. timeout=%ds (nowayout=%d)\n",
-						imx2_wdt.timeout, nowayout);
-	return 0;
+	wdog->timeout = clamp_t(unsigned, timeout, 1, IMX2_WDT_MAX_TIME);
+	if (wdog->timeout != timeout)
+		dev_warn(&pdev->dev, "Initial timeout out of range! Clamped from %u to %u\n",
+			 timeout, wdog->timeout);
+
+	platform_set_drvdata(pdev, wdog);
+	watchdog_set_drvdata(wdog, wdev);
+	watchdog_set_nowayout(wdog, nowayout);
+	watchdog_init_timeout(wdog, timeout, &pdev->dev);
+
+	setup_timer(&wdev->timer, imx2_wdt_timer_ping, (unsigned long)wdog);
+
+	imx2_wdt_ping_if_active(wdog);
 
-fail:
-	imx2_wdt_miscdev.parent = NULL;
-	return ret;
+	ret = watchdog_register_device(wdog);
+	if (ret) {
+		dev_err(&pdev->dev, "cannot register watchdog device\n");
+		return ret;
+	}
+
+	dev_info(&pdev->dev, "timeout %d sec (nowayout=%d)\n",
+		 wdog->timeout, nowayout);
+
+	return 0;
 }
 
 static int __exit imx2_wdt_remove(struct platform_device *pdev)
 {
-	misc_deregister(&imx2_wdt_miscdev);
+	struct watchdog_device *wdog = platform_get_drvdata(pdev);
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
 
-	if (test_bit(IMX2_WDT_STATUS_STARTED, &imx2_wdt.status)) {
-		del_timer_sync(&imx2_wdt.timer);
+	watchdog_unregister_device(wdog);
 
-		dev_crit(imx2_wdt_miscdev.parent,
-			"Device removed: Expect reboot!\n");
+	if (imx2_wdt_is_running(wdev)) {
+		del_timer_sync(&wdev->timer);
+		imx2_wdt_ping(wdog);
+		dev_crit(&pdev->dev, "Device removed: Expect reboot!\n");
 	}
-
-	imx2_wdt_miscdev.parent = NULL;
 	return 0;
 }
 
 static void imx2_wdt_shutdown(struct platform_device *pdev)
 {
-	if (test_bit(IMX2_WDT_STATUS_STARTED, &imx2_wdt.status)) {
-		/* we are running, we need to delete the timer but will give
-		 * max timeout before reboot will take place */
-		del_timer_sync(&imx2_wdt.timer);
-		imx2_wdt_set_timeout(IMX2_WDT_MAX_TIME);
-		imx2_wdt_ping();
-
-		dev_crit(imx2_wdt_miscdev.parent,
-			"Device shutdown: Expect reboot!\n");
+	struct watchdog_device *wdog = platform_get_drvdata(pdev);
+	struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
+
+	if (imx2_wdt_is_running(wdev)) {
+		/*
+		 * We are running, we need to delete the timer but will
+		 * give max timeout before reboot will take place
+		 */
+		del_timer_sync(&wdev->timer);
+		imx2_wdt_set_timeout(wdog, IMX2_WDT_MAX_TIME);
+		imx2_wdt_ping(wdog);
+		dev_crit(&pdev->dev, "Device shutdown: Expect reboot!\n");
 	}
 }
 
diff --git a/drivers/watchdog/intel-mid_wdt.c b/drivers/watchdog/intel-mid_wdt.c
new file mode 100644
index 000000000000..ca66e8e74635
--- /dev/null
+++ b/drivers/watchdog/intel-mid_wdt.c
@@ -0,0 +1,184 @@
+/*
+ *      intel-mid_wdt: generic Intel MID SCU watchdog driver
+ *
+ *      Platforms supported so far:
+ *      - Merrifield only
+ *
+ *      Copyright (C) 2014 Intel Corporation. All rights reserved.
+ *      Contact: David Cohen <david.a.cohen@linux.intel.com>
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of version 2 of the GNU General
+ *      Public License as published by the Free Software Foundation.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/nmi.h>
+#include <linux/platform_device.h>
+#include <linux/watchdog.h>
+#include <linux/platform_data/intel-mid_wdt.h>
+
+#include <asm/intel_scu_ipc.h>
+#include <asm/intel-mid.h>
+
+#define IPC_WATCHDOG 0xf8
+
+#define MID_WDT_PRETIMEOUT		15
+#define MID_WDT_TIMEOUT_MIN		(1 + MID_WDT_PRETIMEOUT)
+#define MID_WDT_TIMEOUT_MAX		170
+#define MID_WDT_DEFAULT_TIMEOUT		90
+
+/* SCU watchdog messages */
+enum {
+	SCU_WATCHDOG_START = 0,
+	SCU_WATCHDOG_STOP,
+	SCU_WATCHDOG_KEEPALIVE,
+};
+
+static inline int wdt_command(int sub, u32 *in, int inlen)
+{
+	return intel_scu_ipc_command(IPC_WATCHDOG, sub, in, inlen, NULL, 0);
+}
+
+static int wdt_start(struct watchdog_device *wd)
+{
+	int ret, in_size;
+	int timeout = wd->timeout;
+	struct ipc_wd_start {
+		u32 pretimeout;
+		u32 timeout;
+	} ipc_wd_start = { timeout - MID_WDT_PRETIMEOUT, timeout };
+
+	/*
+	 * SCU expects the input size for watchdog IPC to
+	 * be based on 4 bytes
+	 */
+	in_size = DIV_ROUND_UP(sizeof(ipc_wd_start), 4);
+
+	ret = wdt_command(SCU_WATCHDOG_START, (u32 *)&ipc_wd_start, in_size);
+	if (ret) {
+		struct device *dev = watchdog_get_drvdata(wd);
+		dev_crit(dev, "error starting watchdog: %d\n", ret);
+	}
+
+	return ret;
+}
+
+static int wdt_ping(struct watchdog_device *wd)
+{
+	int ret;
+
+	ret = wdt_command(SCU_WATCHDOG_KEEPALIVE, NULL, 0);
+	if (ret) {
+		struct device *dev = watchdog_get_drvdata(wd);
+		dev_crit(dev, "Error executing keepalive: 0x%x\n", ret);
+	}
+
+	return ret;
+}
+
+static int wdt_stop(struct watchdog_device *wd)
+{
+	int ret;
+
+	ret = wdt_command(SCU_WATCHDOG_STOP, NULL, 0);
+	if (ret) {
+		struct device *dev = watchdog_get_drvdata(wd);
+		dev_crit(dev, "Error stopping watchdog: 0x%x\n", ret);
+	}
+
+	return ret;
+}
+
+static irqreturn_t mid_wdt_irq(int irq, void *dev_id)
+{
+	panic("Kernel Watchdog");
+
+	/* This code should not be reached */
+	return IRQ_HANDLED;
+}
+
+static const struct watchdog_info mid_wdt_info = {
+	.identity = "Intel MID SCU watchdog",
+	.options = WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT,
+};
+
+static const struct watchdog_ops mid_wdt_ops = {
+	.owner = THIS_MODULE,
+	.start = wdt_start,
+	.stop = wdt_stop,
+	.ping = wdt_ping,
+};
+
+static int mid_wdt_probe(struct platform_device *pdev)
+{
+	struct watchdog_device *wdt_dev;
+	struct intel_mid_wdt_pdata *pdata = pdev->dev.platform_data;
+	int ret;
+
+	if (!pdata) {
+		dev_err(&pdev->dev, "missing platform data\n");
+		return -EINVAL;
+	}
+
+	if (pdata->probe) {
+		ret = pdata->probe(pdev);
+		if (ret)
+			return ret;
+	}
+
+	wdt_dev = devm_kzalloc(&pdev->dev, sizeof(*wdt_dev), GFP_KERNEL);
+	if (!wdt_dev)
+		return -ENOMEM;
+
+	wdt_dev->info = &mid_wdt_info;
+	wdt_dev->ops = &mid_wdt_ops;
+	wdt_dev->min_timeout = MID_WDT_TIMEOUT_MIN;
+	wdt_dev->max_timeout = MID_WDT_TIMEOUT_MAX;
+	wdt_dev->timeout = MID_WDT_DEFAULT_TIMEOUT;
+
+	watchdog_set_drvdata(wdt_dev, &pdev->dev);
+	platform_set_drvdata(pdev, wdt_dev);
+
+	ret = devm_request_irq(&pdev->dev, pdata->irq, mid_wdt_irq,
+			       IRQF_SHARED | IRQF_NO_SUSPEND, "watchdog",
+			       wdt_dev);
+	if (ret) {
+		dev_err(&pdev->dev, "error requesting warning irq %d\n",
+			pdata->irq);
+		return ret;
+	}
+
+	ret = watchdog_register_device(wdt_dev);
+	if (ret) {
+		dev_err(&pdev->dev, "error registering watchdog device\n");
+		return ret;
+	}
+
+	dev_info(&pdev->dev, "Intel MID watchdog device probed\n");
+
+	return 0;
+}
+
+static int mid_wdt_remove(struct platform_device *pdev)
+{
+	struct watchdog_device *wd = platform_get_drvdata(pdev);
+	watchdog_unregister_device(wd);
+	return 0;
+}
+
+static struct platform_driver mid_wdt_driver = {
+	.probe		= mid_wdt_probe,
+	.remove		= mid_wdt_remove,
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= "intel_mid_wdt",
+	},
+};
+
+module_platform_driver(mid_wdt_driver);
+
+MODULE_AUTHOR("David Cohen <david.a.cohen@linux.intel.com>");
+MODULE_DESCRIPTION("Watchdog Driver for Intel MID platform");
+MODULE_LICENSE("GPL");
diff --git a/drivers/watchdog/kempld_wdt.c b/drivers/watchdog/kempld_wdt.c
index 20dc73844737..d9c1a1601926 100644
--- a/drivers/watchdog/kempld_wdt.c
+++ b/drivers/watchdog/kempld_wdt.c
@@ -162,7 +162,7 @@ static int kempld_wdt_set_stage_timeout(struct kempld_wdt_data *wdt_data,
 	kempld_get_mutex(pld);
 	stage_cfg = kempld_read8(pld, KEMPLD_WDT_STAGE_CFG(stage->id));
 	stage_cfg &= ~STAGE_CFG_PRESCALER_MASK;
-	stage_cfg |= STAGE_CFG_SET_PRESCALER(prescaler);
+	stage_cfg |= STAGE_CFG_SET_PRESCALER(PRESCALER_21);
 	kempld_write8(pld, KEMPLD_WDT_STAGE_CFG(stage->id), stage_cfg);
 	kempld_write32(pld, KEMPLD_WDT_STAGE_TIMEOUT(stage->id),
 			stage_timeout);
diff --git a/drivers/watchdog/of_xilinx_wdt.c b/drivers/watchdog/of_xilinx_wdt.c
index 57ccae8327ff..1e6e28df5d7b 100644
--- a/drivers/watchdog/of_xilinx_wdt.c
+++ b/drivers/watchdog/of_xilinx_wdt.c
@@ -225,7 +225,7 @@ static int xwdt_remove(struct platform_device *pdev)
 }
 
 /* Match table for of_platform binding */
-static struct of_device_id xwdt_of_match[] = {
+static const struct of_device_id xwdt_of_match[] = {
 	{ .compatible = "xlnx,xps-timebase-wdt-1.00.a", },
 	{ .compatible = "xlnx,xps-timebase-wdt-1.01.a", },
 	{},
diff --git a/drivers/watchdog/orion_wdt.c b/drivers/watchdog/orion_wdt.c
index 9b3c41d18703..00d0741228fc 100644
--- a/drivers/watchdog/orion_wdt.c
+++ b/drivers/watchdog/orion_wdt.c
@@ -55,15 +55,19 @@ struct orion_watchdog_data {
 	int wdt_counter_offset;
 	int wdt_enable_bit;
 	int rstout_enable_bit;
+	int rstout_mask_bit;
 	int (*clock_init)(struct platform_device *,
 			  struct orion_watchdog *);
+	int (*enabled)(struct orion_watchdog *);
 	int (*start)(struct watchdog_device *);
+	int (*stop)(struct watchdog_device *);
 };
 
 struct orion_watchdog {
 	struct watchdog_device wdt;
 	void __iomem *reg;
 	void __iomem *rstout;
+	void __iomem *rstout_mask;
 	unsigned long clk_rate;
 	struct clk *clk;
 	const struct orion_watchdog_data *data;
@@ -142,9 +146,35 @@ static int orion_wdt_ping(struct watchdog_device *wdt_dev)
 	return 0;
 }
 
+static int armada375_start(struct watchdog_device *wdt_dev)
+{
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+	u32 reg;
+
+	/* Set watchdog duration */
+	writel(dev->clk_rate * wdt_dev->timeout,
+	       dev->reg + dev->data->wdt_counter_offset);
+
+	/* Clear the watchdog expiration bit */
+	atomic_io_modify(dev->reg + TIMER_A370_STATUS, WDT_A370_EXPIRED, 0);
+
+	/* Enable watchdog timer */
+	atomic_io_modify(dev->reg + TIMER_CTRL, dev->data->wdt_enable_bit,
+						dev->data->wdt_enable_bit);
+
+	/* Enable reset on watchdog */
+	reg = readl(dev->rstout);
+	reg |= dev->data->rstout_enable_bit;
+	writel(reg, dev->rstout);
+
+	atomic_io_modify(dev->rstout_mask, dev->data->rstout_mask_bit, 0);
+	return 0;
+}
+
 static int armada370_start(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+	u32 reg;
 
 	/* Set watchdog duration */
 	writel(dev->clk_rate * wdt_dev->timeout,
@@ -157,8 +187,10 @@ static int armada370_start(struct watchdog_device *wdt_dev)
 	atomic_io_modify(dev->reg + TIMER_CTRL, dev->data->wdt_enable_bit,
 						dev->data->wdt_enable_bit);
 
-	atomic_io_modify(dev->rstout, dev->data->rstout_enable_bit,
-				      dev->data->rstout_enable_bit);
+	/* Enable reset on watchdog */
+	reg = readl(dev->rstout);
+	reg |= dev->data->rstout_enable_bit;
+	writel(reg, dev->rstout);
 	return 0;
 }
 
@@ -189,7 +221,7 @@ static int orion_wdt_start(struct watchdog_device *wdt_dev)
 	return dev->data->start(wdt_dev);
 }
 
-static int orion_wdt_stop(struct watchdog_device *wdt_dev)
+static int orion_stop(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
 
@@ -202,7 +234,48 @@ static int orion_wdt_stop(struct watchdog_device *wdt_dev)
 	return 0;
 }
 
-static int orion_wdt_enabled(struct orion_watchdog *dev)
+static int armada375_stop(struct watchdog_device *wdt_dev)
+{
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+	u32 reg;
+
+	/* Disable reset on watchdog */
+	atomic_io_modify(dev->rstout_mask, dev->data->rstout_mask_bit,
+					   dev->data->rstout_mask_bit);
+	reg = readl(dev->rstout);
+	reg &= ~dev->data->rstout_enable_bit;
+	writel(reg, dev->rstout);
+
+	/* Disable watchdog timer */
+	atomic_io_modify(dev->reg + TIMER_CTRL, dev->data->wdt_enable_bit, 0);
+
+	return 0;
+}
+
+static int armada370_stop(struct watchdog_device *wdt_dev)
+{
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+	u32 reg;
+
+	/* Disable reset on watchdog */
+	reg = readl(dev->rstout);
+	reg &= ~dev->data->rstout_enable_bit;
+	writel(reg, dev->rstout);
+
+	/* Disable watchdog timer */
+	atomic_io_modify(dev->reg + TIMER_CTRL, dev->data->wdt_enable_bit, 0);
+
+	return 0;
+}
+
+static int orion_wdt_stop(struct watchdog_device *wdt_dev)
+{
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+
+	return dev->data->stop(wdt_dev);
+}
+
+static int orion_enabled(struct orion_watchdog *dev)
 {
 	bool enabled, running;
 
@@ -212,6 +285,24 @@ static int orion_wdt_enabled(struct orion_watchdog *dev)
 	return enabled && running;
 }
 
+static int armada375_enabled(struct orion_watchdog *dev)
+{
+	bool masked, enabled, running;
+
+	masked = readl(dev->rstout_mask) & dev->data->rstout_mask_bit;
+	enabled = readl(dev->rstout) & dev->data->rstout_enable_bit;
+	running = readl(dev->reg + TIMER_CTRL) & dev->data->wdt_enable_bit;
+
+	return !masked && enabled && running;
+}
+
+static int orion_wdt_enabled(struct watchdog_device *wdt_dev)
+{
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+
+	return dev->data->enabled(dev);
+}
+
 static unsigned int orion_wdt_get_timeleft(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
@@ -262,10 +353,6 @@ static void __iomem *orion_wdt_ioremap_rstout(struct platform_device *pdev,
 		return devm_ioremap(&pdev->dev, res->start,
 				    resource_size(res));
 
-	/* This workaround works only for "orion-wdt", DT-enabled */
-	if (!of_device_is_compatible(pdev->dev.of_node, "marvell,orion-wdt"))
-		return NULL;
-
 	rstout = internal_regs + ORION_RSTOUT_MASK_OFFSET;
 
 	WARN(1, FW_BUG "falling back to harcoded RSTOUT reg %pa\n", &rstout);
@@ -277,7 +364,9 @@ static const struct orion_watchdog_data orion_data = {
 	.wdt_enable_bit = BIT(4),
 	.wdt_counter_offset = 0x24,
 	.clock_init = orion_wdt_clock_init,
+	.enabled = orion_enabled,
 	.start = orion_start,
+	.stop = orion_stop,
 };
 
 static const struct orion_watchdog_data armada370_data = {
@@ -285,7 +374,9 @@ static const struct orion_watchdog_data armada370_data = {
 	.wdt_enable_bit = BIT(8),
 	.wdt_counter_offset = 0x34,
 	.clock_init = armada370_wdt_clock_init,
+	.enabled = orion_enabled,
 	.start = armada370_start,
+	.stop = armada370_stop,
 };
 
 static const struct orion_watchdog_data armadaxp_data = {
@@ -293,7 +384,31 @@ static const struct orion_watchdog_data armadaxp_data = {
 	.wdt_enable_bit = BIT(8),
 	.wdt_counter_offset = 0x34,
 	.clock_init = armadaxp_wdt_clock_init,
+	.enabled = orion_enabled,
 	.start = armada370_start,
+	.stop = armada370_stop,
+};
+
+static const struct orion_watchdog_data armada375_data = {
+	.rstout_enable_bit = BIT(8),
+	.rstout_mask_bit = BIT(10),
+	.wdt_enable_bit = BIT(8),
+	.wdt_counter_offset = 0x34,
+	.clock_init = armada370_wdt_clock_init,
+	.enabled = armada375_enabled,
+	.start = armada375_start,
+	.stop = armada375_stop,
+};
+
+static const struct orion_watchdog_data armada380_data = {
+	.rstout_enable_bit = BIT(8),
+	.rstout_mask_bit = BIT(10),
+	.wdt_enable_bit = BIT(8),
+	.wdt_counter_offset = 0x34,
+	.clock_init = armadaxp_wdt_clock_init,
+	.enabled = armada375_enabled,
+	.start = armada375_start,
+	.stop = armada375_stop,
 };
 
 static const struct of_device_id orion_wdt_of_match_table[] = {
@@ -309,16 +424,78 @@ static const struct of_device_id orion_wdt_of_match_table[] = {
 		.compatible = "marvell,armada-xp-wdt",
 		.data = &armadaxp_data,
 	},
+	{
+		.compatible = "marvell,armada-375-wdt",
+		.data = &armada375_data,
+	},
+	{
+		.compatible = "marvell,armada-380-wdt",
+		.data = &armada380_data,
+	},
 	{},
 };
 MODULE_DEVICE_TABLE(of, orion_wdt_of_match_table);
 
+static int orion_wdt_get_regs(struct platform_device *pdev,
+			      struct orion_watchdog *dev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct resource *res;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -ENODEV;
+	dev->reg = devm_ioremap(&pdev->dev, res->start,
+				resource_size(res));
+	if (!dev->reg)
+		return -ENOMEM;
+
+	/* Each supported compatible has some RSTOUT register quirk */
+	if (of_device_is_compatible(node, "marvell,orion-wdt")) {
+
+		dev->rstout = orion_wdt_ioremap_rstout(pdev, res->start &
+						       INTERNAL_REGS_MASK);
+		if (!dev->rstout)
+			return -ENODEV;
+
+	} else if (of_device_is_compatible(node, "marvell,armada-370-wdt") ||
+		   of_device_is_compatible(node, "marvell,armada-xp-wdt")) {
+
+		/* Dedicated RSTOUT register, can be requested. */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+		dev->rstout = devm_ioremap_resource(&pdev->dev, res);
+		if (IS_ERR(dev->rstout))
+			return PTR_ERR(dev->rstout);
+
+	} else if (of_device_is_compatible(node, "marvell,armada-375-wdt") ||
+		   of_device_is_compatible(node, "marvell,armada-380-wdt")) {
+
+		/* Dedicated RSTOUT register, can be requested. */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+		dev->rstout = devm_ioremap_resource(&pdev->dev, res);
+		if (IS_ERR(dev->rstout))
+			return PTR_ERR(dev->rstout);
+
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 2);
+		if (!res)
+			return -ENODEV;
+		dev->rstout_mask = devm_ioremap(&pdev->dev, res->start,
+						resource_size(res));
+		if (!dev->rstout_mask)
+			return -ENOMEM;
+
+	} else {
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
 static int orion_wdt_probe(struct platform_device *pdev)
 {
 	struct orion_watchdog *dev;
 	const struct of_device_id *match;
 	unsigned int wdt_max_duration;	/* (seconds) */
-	struct resource *res;
 	int ret, irq;
 
 	dev = devm_kzalloc(&pdev->dev, sizeof(struct orion_watchdog),
@@ -336,19 +513,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	dev->wdt.min_timeout = 1;
 	dev->data = match->data;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res)
-		return -ENODEV;
-
-	dev->reg = devm_ioremap(&pdev->dev, res->start,
-			       resource_size(res));
-	if (!dev->reg)
-		return -ENOMEM;
-
-	dev->rstout = orion_wdt_ioremap_rstout(pdev, res->start &
-						     INTERNAL_REGS_MASK);
-	if (!dev->rstout)
-		return -ENODEV;
+	ret = orion_wdt_get_regs(pdev, dev);
+	if (ret)
+		return ret;
 
 	ret = dev->data->clock_init(pdev, dev);
 	if (ret) {
@@ -371,7 +538,7 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	 * removed and re-insterted, or if the bootloader explicitly
 	 * set a running watchdog before booting the kernel.
 	 */
-	if (!orion_wdt_enabled(dev))
+	if (!orion_wdt_enabled(&dev->wdt))
 		orion_wdt_stop(&dev->wdt);
 
 	/* Request the IRQ only after the watchdog is disabled */
diff --git a/drivers/watchdog/shwdt.c b/drivers/watchdog/shwdt.c
index d04d02b41c32..061756e36cf8 100644
--- a/drivers/watchdog/shwdt.c
+++ b/drivers/watchdog/shwdt.c
@@ -282,8 +282,6 @@ static int sh_wdt_probe(struct platform_device *pdev)
 	wdt->timer.data		= (unsigned long)wdt;
 	wdt->timer.expires	= next_ping_period(clock_division_ratio);
 
-	platform_set_drvdata(pdev, wdt);
-
 	dev_info(&pdev->dev, "initialized.\n");
 
 	pm_runtime_enable(&pdev->dev);
diff --git a/drivers/watchdog/sp805_wdt.c b/drivers/watchdog/sp805_wdt.c
index 47629d268e0a..c1b03f4235b9 100644
--- a/drivers/watchdog/sp805_wdt.c
+++ b/drivers/watchdog/sp805_wdt.c
@@ -59,7 +59,6 @@
  * @adev: amba device structure of wdt
  * @status: current status of wdt
  * @load_val: load value to be set for current timeout
- * @timeout: current programmed timeout
  */
 struct sp805_wdt {
 	struct watchdog_device		wdd;
@@ -68,7 +67,6 @@ struct sp805_wdt {
 	struct clk			*clk;
 	struct amba_device		*adev;
 	unsigned int			load_val;
-	unsigned int			timeout;
 };
 
 static bool nowayout = WATCHDOG_NOWAYOUT;
@@ -98,7 +96,7 @@ static int wdt_setload(struct watchdog_device *wdd, unsigned int timeout)
 	spin_lock(&wdt->lock);
 	wdt->load_val = load;
 	/* roundup timeout to closest positive integer value */
-	wdt->timeout = div_u64((load + 1) * 2 + (rate / 2), rate);
+	wdd->timeout = div_u64((load + 1) * 2 + (rate / 2), rate);
 	spin_unlock(&wdt->lock);
 
 	return 0;
diff --git a/drivers/watchdog/sunxi_wdt.c b/drivers/watchdog/sunxi_wdt.c
index cd00a7836cdc..693b9d2c6e39 100644
--- a/drivers/watchdog/sunxi_wdt.c
+++ b/drivers/watchdog/sunxi_wdt.c
@@ -57,17 +57,17 @@ struct sunxi_wdt_dev {
  */
 
 static const int wdt_timeout_map[] = {
-	[1] = 0b0001,  /* 1s  */
-	[2] = 0b0010,  /* 2s  */
-	[3] = 0b0011,  /* 3s  */
-	[4] = 0b0100,  /* 4s  */
-	[5] = 0b0101,  /* 5s  */
-	[6] = 0b0110,  /* 6s  */
-	[8] = 0b0111,  /* 8s  */
-	[10] = 0b1000, /* 10s */
-	[12] = 0b1001, /* 12s */
-	[14] = 0b1010, /* 14s */
-	[16] = 0b1011, /* 16s */
+	[1] = 0x1,  /* 1s  */
+	[2] = 0x2,  /* 2s  */
+	[3] = 0x3,  /* 3s  */
+	[4] = 0x4,  /* 4s  */
+	[5] = 0x5,  /* 5s  */
+	[6] = 0x6,  /* 6s  */
+	[8] = 0x7,  /* 8s  */
+	[10] = 0x8, /* 10s */
+	[12] = 0x9, /* 12s */
+	[14] = 0xA, /* 14s */
+	[16] = 0xB, /* 16s */
 };
 
 static int sunxi_wdt_ping(struct watchdog_device *wdt_dev)
diff --git a/drivers/watchdog/via_wdt.c b/drivers/watchdog/via_wdt.c
index d2cd9f0bcb9a..56369c4f1961 100644
--- a/drivers/watchdog/via_wdt.c
+++ b/drivers/watchdog/via_wdt.c
@@ -232,7 +232,7 @@ err_out_disable_device:
 static void wdt_remove(struct pci_dev *pdev)
 {
 	watchdog_unregister_device(&wdt_dev);
-	del_timer(&timer);
+	del_timer_sync(&timer);
 	iounmap(wdt_mem);
 	release_mem_region(mmio, VIA_WDT_MMIO_LEN);
 	release_resource(&wdt_res);
diff --git a/drivers/watchdog/w83627hf_wdt.c b/drivers/watchdog/w83627hf_wdt.c
index b1da0c18fd1a..7165704a3e33 100644
--- a/drivers/watchdog/w83627hf_wdt.c
+++ b/drivers/watchdog/w83627hf_wdt.c
@@ -64,6 +64,10 @@ MODULE_PARM_DESC(nowayout,
 		"Watchdog cannot be stopped once started (default="
 				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
 
+static int early_disable;
+module_param(early_disable, int, 0);
+MODULE_PARM_DESC(early_disable, "Disable watchdog at boot time (default=0)");
+
 /*
  *	Kernel methods.
  */
@@ -208,9 +212,14 @@ static int w83627hf_init(struct watchdog_device *wdog, enum chips chip)
 
 	t = superio_inb(cr_wdt_timeout);
 	if (t != 0) {
-		pr_info("Watchdog already running. Resetting timeout to %d sec\n",
-			wdog->timeout);
-		superio_outb(cr_wdt_timeout, wdog->timeout);
+		if (early_disable) {
+			pr_warn("Stopping previously enabled watchdog until userland kicks in\n");
+			superio_outb(cr_wdt_timeout, 0);
+		} else {
+			pr_info("Watchdog already running. Resetting timeout to %d sec\n",
+				wdog->timeout);
+			superio_outb(cr_wdt_timeout, wdog->timeout);
+		}
 	}
 
 	/* set second mode & disable keyboard turning off watchdog */
diff --git a/drivers/watchdog/w83697hf_wdt.c b/drivers/watchdog/w83697hf_wdt.c
deleted file mode 100644
index e9ea856b8ff2..000000000000
--- a/drivers/watchdog/w83697hf_wdt.c
+++ /dev/null
@@ -1,460 +0,0 @@
-/*
- *	w83697hf/hg WDT driver
- *
- *	(c) Copyright 2006 Samuel Tardieu <sam@rfc1149.net>
- *	(c) Copyright 2006 Marcus Junker <junker@anduras.de>
- *
- *	Based on w83627hf_wdt.c which is based on advantechwdt.c
- *	which is based on wdt.c.
- *	Original copyright messages:
- *
- *	(c) Copyright 2003 Pádraig Brady <P@draigBrady.com>
- *
- *	(c) Copyright 2000-2001 Marek Michalkiewicz <marekm@linux.org.pl>
- *
- *	(c) Copyright 1996 Alan Cox <alan@lxorguk.ukuu.org.uk>,
- *						All Rights Reserved.
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- *	Neither Marcus Junker nor ANDURAS AG admit liability nor provide
- *	warranty for any of this software. This material is provided
- *	"AS-IS" and at no charge.
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/types.h>
-#include <linux/miscdevice.h>
-#include <linux/watchdog.h>
-#include <linux/fs.h>
-#include <linux/ioport.h>
-#include <linux/notifier.h>
-#include <linux/reboot.h>
-#include <linux/init.h>
-#include <linux/spinlock.h>
-#include <linux/io.h>
-#include <linux/uaccess.h>
-
-
-#define WATCHDOG_NAME "w83697hf/hg WDT"
-#define WATCHDOG_TIMEOUT 60		/* 60 sec default timeout */
-#define WATCHDOG_EARLY_DISABLE 1	/* Disable until userland kicks in */
-
-static unsigned long wdt_is_open;
-static char expect_close;
-static DEFINE_SPINLOCK(io_lock);
-
-/* You must set this - there is no sane way to probe for this board. */
-static int wdt_io = 0x2e;
-module_param(wdt_io, int, 0);
-MODULE_PARM_DESC(wdt_io,
-		"w83697hf/hg WDT io port (default 0x2e, 0 = autodetect)");
-
-static int timeout = WATCHDOG_TIMEOUT;	/* in seconds */
-module_param(timeout, int, 0);
-MODULE_PARM_DESC(timeout,
-	"Watchdog timeout in seconds. 1<= timeout <=255 (default="
-				__MODULE_STRING(WATCHDOG_TIMEOUT) ")");
-
-static bool nowayout = WATCHDOG_NOWAYOUT;
-module_param(nowayout, bool, 0);
-MODULE_PARM_DESC(nowayout,
-	"Watchdog cannot be stopped once started (default="
-				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
-
-static int early_disable = WATCHDOG_EARLY_DISABLE;
-module_param(early_disable, int, 0);
-MODULE_PARM_DESC(early_disable,
-	"Watchdog gets disabled at boot time (default="
-				__MODULE_STRING(WATCHDOG_EARLY_DISABLE) ")");
-
-/*
- *	Kernel methods.
- */
-
-#define W83697HF_EFER (wdt_io + 0)  /* Extended Function Enable Register */
-#define W83697HF_EFIR (wdt_io + 0)  /* Extended Function Index Register
-							(same as EFER) */
-#define W83697HF_EFDR (wdt_io + 1)  /* Extended Function Data Register */
-
-static inline void w83697hf_unlock(void)
-{
-	outb_p(0x87, W83697HF_EFER);	/* Enter extended function mode */
-	outb_p(0x87, W83697HF_EFER);	/* Again according to manual */
-}
-
-static inline void w83697hf_lock(void)
-{
-	outb_p(0xAA, W83697HF_EFER);	/* Leave extended function mode */
-}
-
-/*
- *	The three functions w83697hf_get_reg(), w83697hf_set_reg() and
- *	w83697hf_write_timeout() must be called with the device unlocked.
- */
-
-static unsigned char w83697hf_get_reg(unsigned char reg)
-{
-	outb_p(reg, W83697HF_EFIR);
-	return inb_p(W83697HF_EFDR);
-}
-
-static void w83697hf_set_reg(unsigned char reg, unsigned char data)
-{
-	outb_p(reg, W83697HF_EFIR);
-	outb_p(data, W83697HF_EFDR);
-}
-
-static void w83697hf_write_timeout(int timeout)
-{
-	/* Write Timeout counter to CRF4 */
-	w83697hf_set_reg(0xF4, timeout);
-}
-
-static void w83697hf_select_wdt(void)
-{
-	w83697hf_unlock();
-	w83697hf_set_reg(0x07, 0x08);	/* Switch to logic device 8 (GPIO2) */
-}
-
-static inline void w83697hf_deselect_wdt(void)
-{
-	w83697hf_lock();
-}
-
-static void w83697hf_init(void)
-{
-	unsigned char bbuf;
-
-	w83697hf_select_wdt();
-
-	bbuf = w83697hf_get_reg(0x29);
-	bbuf &= ~0x60;
-	bbuf |= 0x20;
-
-	/* Set pin 119 to WDTO# mode (= CR29, WDT0) */
-	w83697hf_set_reg(0x29, bbuf);
-
-	bbuf = w83697hf_get_reg(0xF3);
-	bbuf &= ~0x04;
-	w83697hf_set_reg(0xF3, bbuf);	/* Count mode is seconds */
-
-	w83697hf_deselect_wdt();
-}
-
-static void wdt_ping(void)
-{
-	spin_lock(&io_lock);
-	w83697hf_select_wdt();
-
-	w83697hf_write_timeout(timeout);
-
-	w83697hf_deselect_wdt();
-	spin_unlock(&io_lock);
-}
-
-static void wdt_enable(void)
-{
-	spin_lock(&io_lock);
-	w83697hf_select_wdt();
-
-	w83697hf_write_timeout(timeout);
-	w83697hf_set_reg(0x30, 1);	/* Enable timer */
-
-	w83697hf_deselect_wdt();
-	spin_unlock(&io_lock);
-}
-
-static void wdt_disable(void)
-{
-	spin_lock(&io_lock);
-	w83697hf_select_wdt();
-
-	w83697hf_set_reg(0x30, 0);	/* Disable timer */
-	w83697hf_write_timeout(0);
-
-	w83697hf_deselect_wdt();
-	spin_unlock(&io_lock);
-}
-
-static unsigned char wdt_running(void)
-{
-	unsigned char t;
-
-	spin_lock(&io_lock);
-	w83697hf_select_wdt();
-
-	t = w83697hf_get_reg(0xF4);	/* Read timer */
-
-	w83697hf_deselect_wdt();
-	spin_unlock(&io_lock);
-
-	return t;
-}
-
-static int wdt_set_heartbeat(int t)
-{
-	if (t < 1 || t > 255)
-		return -EINVAL;
-
-	timeout = t;
-	return 0;
-}
-
-static ssize_t wdt_write(struct file *file, const char __user *buf,
-						size_t count, loff_t *ppos)
-{
-	if (count) {
-		if (!nowayout) {
-			size_t i;
-
-			expect_close = 0;
-
-			for (i = 0; i != count; i++) {
-				char c;
-				if (get_user(c, buf + i))
-					return -EFAULT;
-				if (c == 'V')
-					expect_close = 42;
-			}
-		}
-		wdt_ping();
-	}
-	return count;
-}
-
-static long wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
-{
-	void __user *argp = (void __user *)arg;
-	int __user *p = argp;
-	int new_timeout;
-	static const struct watchdog_info ident = {
-		.options = WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT
-							| WDIOF_MAGICCLOSE,
-		.firmware_version = 1,
-		.identity = "W83697HF WDT",
-	};
-
-	switch (cmd) {
-	case WDIOC_GETSUPPORT:
-		if (copy_to_user(argp, &ident, sizeof(ident)))
-			return -EFAULT;
-		break;
-
-	case WDIOC_GETSTATUS:
-	case WDIOC_GETBOOTSTATUS:
-		return put_user(0, p);
-
-	case WDIOC_SETOPTIONS:
-	{
-		int options, retval = -EINVAL;
-
-		if (get_user(options, p))
-			return -EFAULT;
-
-		if (options & WDIOS_DISABLECARD) {
-			wdt_disable();
-			retval = 0;
-		}
-
-		if (options & WDIOS_ENABLECARD) {
-			wdt_enable();
-			retval = 0;
-		}
-
-		return retval;
-	}
-
-	case WDIOC_KEEPALIVE:
-		wdt_ping();
-		break;
-
-	case WDIOC_SETTIMEOUT:
-		if (get_user(new_timeout, p))
-			return -EFAULT;
-		if (wdt_set_heartbeat(new_timeout))
-			return -EINVAL;
-		wdt_ping();
-		/* Fall */
-
-	case WDIOC_GETTIMEOUT:
-		return put_user(timeout, p);
-
-	default:
-		return -ENOTTY;
-	}
-	return 0;
-}
-
-static int wdt_open(struct inode *inode, struct file *file)
-{
-	if (test_and_set_bit(0, &wdt_is_open))
-		return -EBUSY;
-	/*
-	 *	Activate
-	 */
-
-	wdt_enable();
-	return nonseekable_open(inode, file);
-}
-
-static int wdt_close(struct inode *inode, struct file *file)
-{
-	if (expect_close == 42)
-		wdt_disable();
-	else {
-		pr_crit("Unexpected close, not stopping watchdog!\n");
-		wdt_ping();
-	}
-	expect_close = 0;
-	clear_bit(0, &wdt_is_open);
-	return 0;
-}
-
-/*
- *	Notifier for system down
- */
-
-static int wdt_notify_sys(struct notifier_block *this, unsigned long code,
-	void *unused)
-{
-	if (code == SYS_DOWN || code == SYS_HALT)
-		wdt_disable();	/* Turn the WDT off */
-
-	return NOTIFY_DONE;
-}
-
-/*
- *	Kernel Interfaces
- */
-
-static const struct file_operations wdt_fops = {
-	.owner		= THIS_MODULE,
-	.llseek		= no_llseek,
-	.write		= wdt_write,
-	.unlocked_ioctl	= wdt_ioctl,
-	.open		= wdt_open,
-	.release	= wdt_close,
-};
-
-static struct miscdevice wdt_miscdev = {
-	.minor = WATCHDOG_MINOR,
-	.name = "watchdog",
-	.fops = &wdt_fops,
-};
-
-/*
- *	The WDT needs to learn about soft shutdowns in order to
- *	turn the timebomb registers off.
- */
-
-static struct notifier_block wdt_notifier = {
-	.notifier_call = wdt_notify_sys,
-};
-
-static int w83697hf_check_wdt(void)
-{
-	if (!request_region(wdt_io, 2, WATCHDOG_NAME)) {
-		pr_err("I/O address 0x%x already in use\n", wdt_io);
-		return -EIO;
-	}
-
-	pr_debug("Looking for watchdog at address 0x%x\n", wdt_io);
-	w83697hf_unlock();
-	if (w83697hf_get_reg(0x20) == 0x60) {
-		pr_info("watchdog found at address 0x%x\n", wdt_io);
-		w83697hf_lock();
-		return 0;
-	}
-	/* Reprotect in case it was a compatible device */
-	w83697hf_lock();
-
-	pr_info("watchdog not found at address 0x%x\n", wdt_io);
-	release_region(wdt_io, 2);
-	return -EIO;
-}
-
-static int w83697hf_ioports[] = { 0x2e, 0x4e, 0x00 };
-
-static int __init wdt_init(void)
-{
-	int ret, i, found = 0;
-
-	pr_info("WDT driver for W83697HF/HG initializing\n");
-
-	if (wdt_io == 0) {
-		/* we will autodetect the W83697HF/HG watchdog */
-		for (i = 0; ((!found) && (w83697hf_ioports[i] != 0)); i++) {
-			wdt_io = w83697hf_ioports[i];
-			if (!w83697hf_check_wdt())
-				found++;
-		}
-	} else {
-		if (!w83697hf_check_wdt())
-			found++;
-	}
-
-	if (!found) {
-		pr_err("No W83697HF/HG could be found\n");
-		ret = -ENODEV;
-		goto out;
-	}
-
-	w83697hf_init();
-	if (early_disable) {
-		if (wdt_running())
-			pr_warn("Stopping previously enabled watchdog until userland kicks in\n");
-		wdt_disable();
-	}
-
-	if (wdt_set_heartbeat(timeout)) {
-		wdt_set_heartbeat(WATCHDOG_TIMEOUT);
-		pr_info("timeout value must be 1 <= timeout <= 255, using %d\n",
-			WATCHDOG_TIMEOUT);
-	}
-
-	ret = register_reboot_notifier(&wdt_notifier);
-	if (ret != 0) {
-		pr_err("cannot register reboot notifier (err=%d)\n", ret);
-		goto unreg_regions;
-	}
-
-	ret = misc_register(&wdt_miscdev);
-	if (ret != 0) {
-		pr_err("cannot register miscdev on minor=%d (err=%d)\n",
-		       WATCHDOG_MINOR, ret);
-		goto unreg_reboot;
-	}
-
-	pr_info("initialized. timeout=%d sec (nowayout=%d)\n",
-		timeout, nowayout);
-
-out:
-	return ret;
-unreg_reboot:
-	unregister_reboot_notifier(&wdt_notifier);
-unreg_regions:
-	release_region(wdt_io, 2);
-	goto out;
-}
-
-static void __exit wdt_exit(void)
-{
-	misc_deregister(&wdt_miscdev);
-	unregister_reboot_notifier(&wdt_notifier);
-	release_region(wdt_io, 2);
-}
-
-module_init(wdt_init);
-module_exit(wdt_exit);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Marcus Junker <junker@anduras.de>");
-MODULE_AUTHOR("Samuel Tardieu <sam@rfc1149.net>");
-MODULE_DESCRIPTION("w83697hf/hg WDT driver");
diff --git a/drivers/watchdog/w83697ug_wdt.c b/drivers/watchdog/w83697ug_wdt.c
deleted file mode 100644
index ff58cb74671f..000000000000
--- a/drivers/watchdog/w83697ug_wdt.c
+++ /dev/null
@@ -1,397 +0,0 @@
-/*
- *	w83697ug/uf WDT driver
- *
- *	(c) Copyright 2008 Flemming Fransen <ff@nrvissing.net>
- *		reused original code to support w83697ug/uf.
- *
- *	Based on w83627hf_wdt.c which is based on advantechwdt.c
- *	which is based on wdt.c.
- *	Original copyright messages:
- *
- *	(c) Copyright 2007 Vlad Drukker <vlad@storewiz.com>
- *		added support for W83627THF.
- *
- *	(c) Copyright 2003 Pádraig Brady <P@draigBrady.com>
- *
- *	(c) Copyright 2000-2001 Marek Michalkiewicz <marekm@linux.org.pl>
- *
- *	(c) Copyright 1996 Alan Cox <alan@redhat.com>, All Rights Reserved.
- *				http://www.redhat.com
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- *	Neither Alan Cox nor CymruNet Ltd. admit liability nor provide
- *	warranty for any of this software. This material is provided
- *	"AS-IS" and at no charge.
- *
- *	(c) Copyright 1995    Alan Cox <alan@redhat.com>
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/types.h>
-#include <linux/miscdevice.h>
-#include <linux/watchdog.h>
-#include <linux/fs.h>
-#include <linux/ioport.h>
-#include <linux/notifier.h>
-#include <linux/reboot.h>
-#include <linux/init.h>
-#include <linux/spinlock.h>
-#include <linux/io.h>
-#include <linux/uaccess.h>
-
-
-#define WATCHDOG_NAME "w83697ug/uf WDT"
-#define WATCHDOG_TIMEOUT 60		/* 60 sec default timeout */
-
-static unsigned long wdt_is_open;
-static char expect_close;
-static DEFINE_SPINLOCK(io_lock);
-
-static int wdt_io = 0x2e;
-module_param(wdt_io, int, 0);
-MODULE_PARM_DESC(wdt_io, "w83697ug/uf WDT io port (default 0x2e)");
-
-static int timeout = WATCHDOG_TIMEOUT;	/* in seconds */
-module_param(timeout, int, 0);
-MODULE_PARM_DESC(timeout,
-	"Watchdog timeout in seconds. 1<= timeout <=255 (default="
-				__MODULE_STRING(WATCHDOG_TIMEOUT) ")");
-
-static bool nowayout = WATCHDOG_NOWAYOUT;
-module_param(nowayout, bool, 0);
-MODULE_PARM_DESC(nowayout,
-	"Watchdog cannot be stopped once started (default="
-				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
-
-/*
- *	Kernel methods.
- */
-
-#define WDT_EFER (wdt_io+0)   /* Extended Function Enable Registers */
-#define WDT_EFIR (wdt_io+0)   /* Extended Function Index Register
-							(same as EFER) */
-#define WDT_EFDR (WDT_EFIR+1) /* Extended Function Data Register */
-
-static int w83697ug_select_wd_register(void)
-{
-	unsigned char c;
-	unsigned char version;
-
-	outb_p(0x87, WDT_EFER); /* Enter extended function mode */
-	outb_p(0x87, WDT_EFER); /* Again according to manual */
-
-	outb(0x20, WDT_EFER);	/* check chip version	*/
-	version = inb(WDT_EFDR);
-
-	if (version == 0x68) {	/* W83697UG		*/
-		pr_info("Watchdog chip version 0x%02x = W83697UG/UF found at 0x%04x\n",
-			version, wdt_io);
-
-		outb_p(0x2b, WDT_EFER);
-		c = inb_p(WDT_EFDR);    /* select WDT0 */
-		c &= ~0x04;
-		outb_p(0x2b, WDT_EFER);
-		outb_p(c, WDT_EFDR);	/* set pin118 to WDT0 */
-
-	} else {
-		pr_err("No W83697UG/UF could be found\n");
-		return -ENODEV;
-	}
-
-	outb_p(0x07, WDT_EFER); /* point to logical device number reg */
-	outb_p(0x08, WDT_EFDR); /* select logical device 8 (GPIO2) */
-	outb_p(0x30, WDT_EFER); /* select CR30 */
-	c = inb_p(WDT_EFDR);
-	outb_p(c | 0x01, WDT_EFDR); /* set bit 0 to activate GPIO2 */
-
-	return 0;
-}
-
-static void w83697ug_unselect_wd_register(void)
-{
-	outb_p(0xAA, WDT_EFER); /* Leave extended function mode */
-}
-
-static int w83697ug_init(void)
-{
-	int ret;
-	unsigned char t;
-
-	ret = w83697ug_select_wd_register();
-	if (ret != 0)
-		return ret;
-
-	outb_p(0xF6, WDT_EFER); /* Select CRF6 */
-	t = inb_p(WDT_EFDR);    /* read CRF6 */
-	if (t != 0) {
-		pr_info("Watchdog already running. Resetting timeout to %d sec\n",
-			timeout);
-		outb_p(timeout, WDT_EFDR);    /* Write back to CRF6 */
-	}
-	outb_p(0xF5, WDT_EFER); /* Select CRF5 */
-	t = inb_p(WDT_EFDR);    /* read CRF5 */
-	t &= ~0x0C;             /* set second mode &
-					disable keyboard turning off watchdog */
-	outb_p(t, WDT_EFDR);    /* Write back to CRF5 */
-
-	w83697ug_unselect_wd_register();
-	return 0;
-}
-
-static void wdt_ctrl(int timeout)
-{
-	spin_lock(&io_lock);
-
-	if (w83697ug_select_wd_register() < 0) {
-		spin_unlock(&io_lock);
-		return;
-	}
-
-	outb_p(0xF4, WDT_EFER);    /* Select CRF4 */
-	outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF4 */
-
-	w83697ug_unselect_wd_register();
-
-	spin_unlock(&io_lock);
-}
-
-static int wdt_ping(void)
-{
-	wdt_ctrl(timeout);
-	return 0;
-}
-
-static int wdt_disable(void)
-{
-	wdt_ctrl(0);
-	return 0;
-}
-
-static int wdt_set_heartbeat(int t)
-{
-	if (t < 1 || t > 255)
-		return -EINVAL;
-
-	timeout = t;
-	return 0;
-}
-
-static ssize_t wdt_write(struct file *file, const char __user *buf,
-						size_t count, loff_t *ppos)
-{
-	if (count) {
-		if (!nowayout) {
-			size_t i;
-
-			expect_close = 0;
-
-			for (i = 0; i != count; i++) {
-				char c;
-				if (get_user(c, buf + i))
-					return -EFAULT;
-				if (c == 'V')
-					expect_close = 42;
-			}
-		}
-		wdt_ping();
-	}
-	return count;
-}
-
-static long wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
-{
-	void __user *argp = (void __user *)arg;
-	int __user *p = argp;
-	int new_timeout;
-	static const struct watchdog_info ident = {
-		.options =		WDIOF_KEEPALIVEPING |
-					WDIOF_SETTIMEOUT |
-					WDIOF_MAGICCLOSE,
-		.firmware_version =	1,
-		.identity =		"W83697UG WDT",
-	};
-
-	switch (cmd) {
-	case WDIOC_GETSUPPORT:
-		if (copy_to_user(argp, &ident, sizeof(ident)))
-			return -EFAULT;
-		break;
-
-	case WDIOC_GETSTATUS:
-	case WDIOC_GETBOOTSTATUS:
-		return put_user(0, p);
-
-	case WDIOC_SETOPTIONS:
-	{
-		int options, retval = -EINVAL;
-
-		if (get_user(options, p))
-			return -EFAULT;
-
-		if (options & WDIOS_DISABLECARD) {
-			wdt_disable();
-			retval = 0;
-		}
-
-		if (options & WDIOS_ENABLECARD) {
-			wdt_ping();
-			retval = 0;
-		}
-
-		return retval;
-	}
-
-	case WDIOC_KEEPALIVE:
-		wdt_ping();
-		break;
-
-	case WDIOC_SETTIMEOUT:
-		if (get_user(new_timeout, p))
-			return -EFAULT;
-		if (wdt_set_heartbeat(new_timeout))
-			return -EINVAL;
-		wdt_ping();
-		/* Fall */
-
-	case WDIOC_GETTIMEOUT:
-		return put_user(timeout, p);
-
-	default:
-		return -ENOTTY;
-	}
-	return 0;
-}
-
-static int wdt_open(struct inode *inode, struct file *file)
-{
-	if (test_and_set_bit(0, &wdt_is_open))
-		return -EBUSY;
-	/*
-	 *	Activate
-	 */
-
-	wdt_ping();
-	return nonseekable_open(inode, file);
-}
-
-static int wdt_close(struct inode *inode, struct file *file)
-{
-	if (expect_close == 42)
-		wdt_disable();
-	else {
-		pr_crit("Unexpected close, not stopping watchdog!\n");
-		wdt_ping();
-	}
-	expect_close = 0;
-	clear_bit(0, &wdt_is_open);
-	return 0;
-}
-
-/*
- *	Notifier for system down
- */
-
-static int wdt_notify_sys(struct notifier_block *this, unsigned long code,
-	void *unused)
-{
-	if (code == SYS_DOWN || code == SYS_HALT)
-		wdt_disable();	/* Turn the WDT off */
-
-	return NOTIFY_DONE;
-}
-
-/*
- *	Kernel Interfaces
- */
-
-static const struct file_operations wdt_fops = {
-	.owner		= THIS_MODULE,
-	.llseek		= no_llseek,
-	.write		= wdt_write,
-	.unlocked_ioctl	= wdt_ioctl,
-	.open		= wdt_open,
-	.release	= wdt_close,
-};
-
-static struct miscdevice wdt_miscdev = {
-	.minor = WATCHDOG_MINOR,
-	.name = "watchdog",
-	.fops = &wdt_fops,
-};
-
-/*
- *	The WDT needs to learn about soft shutdowns in order to
- *	turn the timebomb registers off.
- */
-
-static struct notifier_block wdt_notifier = {
-	.notifier_call = wdt_notify_sys,
-};
-
-static int __init wdt_init(void)
-{
-	int ret;
-
-	pr_info("WDT driver for the Winbond(TM) W83697UG/UF Super I/O chip initialising\n");
-
-	if (wdt_set_heartbeat(timeout)) {
-		wdt_set_heartbeat(WATCHDOG_TIMEOUT);
-		pr_info("timeout value must be 1<=timeout<=255, using %d\n",
-			WATCHDOG_TIMEOUT);
-	}
-
-	if (!request_region(wdt_io, 1, WATCHDOG_NAME)) {
-		pr_err("I/O address 0x%04x already in use\n", wdt_io);
-		ret = -EIO;
-		goto out;
-	}
-
-	ret = w83697ug_init();
-	if (ret != 0)
-		goto unreg_regions;
-
-	ret = register_reboot_notifier(&wdt_notifier);
-	if (ret != 0) {
-		pr_err("cannot register reboot notifier (err=%d)\n", ret);
-		goto unreg_regions;
-	}
-
-	ret = misc_register(&wdt_miscdev);
-	if (ret != 0) {
-		pr_err("cannot register miscdev on minor=%d (err=%d)\n",
-		       WATCHDOG_MINOR, ret);
-		goto unreg_reboot;
-	}
-
-	pr_info("initialized. timeout=%d sec (nowayout=%d)\n",
-		timeout, nowayout);
-
-out:
-	return ret;
-unreg_reboot:
-	unregister_reboot_notifier(&wdt_notifier);
-unreg_regions:
-	release_region(wdt_io, 1);
-	goto out;
-}
-
-static void __exit wdt_exit(void)
-{
-	misc_deregister(&wdt_miscdev);
-	unregister_reboot_notifier(&wdt_notifier);
-	release_region(wdt_io, 1);
-}
-
-module_init(wdt_init);
-module_exit(wdt_exit);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Flemming Frandsen <ff@nrvissing.net>");
-MODULE_DESCRIPTION("w83697ug/uf WDT driver");
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 6d325bda76da..5d4de88fe5b8 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -1168,7 +1168,8 @@ int gnttab_resume(void)
 
 int gnttab_suspend(void)
 {
-	gnttab_interface->unmap_frames();
+	if (!xen_feature(XENFEAT_auto_translated_physmap))
+		gnttab_interface->unmap_frames();
 	return 0;
 }