Merge tag 's390-5.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Vasily Gorbik:

 - Add support for multi-function devices in pci code.

 - Enable PF-VF linking for architectures using the pdev->no_vf_scan
   flag (currently just s390).

 - Add reipl from NVMe support.

 - Get rid of critical section cleanup in entry.S.

 - Refactor PNSO CHSC (perform network subchannel operation) in cio and
   qeth.

 - QDIO interrupts and error handling fixes and improvements, more
   refactoring changes.

 - Align ioremap() with generic code.

 - Accept requests without the prefetch bit set in vfio-ccw.

 - Enable path handling via two new regions in vfio-ccw.

 - Other small fixes and improvements all over the code.

* tag 's390-5.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (52 commits)
  vfio-ccw: make vfio_ccw_regops variables declarations static
  vfio-ccw: Add trace for CRW event
  vfio-ccw: Wire up the CRW irq and CRW region
  vfio-ccw: Introduce a new CRW region
  vfio-ccw: Refactor IRQ handlers
  vfio-ccw: Introduce a new schib region
  vfio-ccw: Refactor the unregister of the async regions
  vfio-ccw: Register a chp_event callback for vfio-ccw
  vfio-ccw: Introduce new helper functions to free/destroy regions
  vfio-ccw: document possible errors
  vfio-ccw: Enable transparent CCW IPL from DASD
  s390/pci: Log new handle in clp_disable_fh()
  s390/cio, s390/qeth: cleanup PNSO CHSC
  s390/qdio: remove q->first_to_kick
  s390/qdio: fix up qdio_start_irq() kerneldoc
  s390: remove critical section cleanup from entry.S
  s390: add machine check SIGP
  s390/pci: ioremap() align with generic code
  s390/ap: introduce new ap function ap_get_qdev()
  Documentation/s390: Update / remove developerWorks web links
  ...

7 files changed, 483 insertions, 155 deletions

diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile
index 748626a33028..b4e3c84772a1 100644
--- a/arch/s390/pci/Makefile
+++ b/arch/s390/pci/Makefile
@@ -4,4 +4,5 @@
 #
 
 obj-$(CONFIG_PCI)	+= pci.o pci_irq.o pci_dma.o pci_clp.o pci_sysfs.o \
-			   pci_event.o pci_debug.o pci_insn.o pci_mmio.o
+			   pci_event.o pci_debug.o pci_insn.o pci_mmio.o \
+			   pci_bus.o
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index 94ca121933de..3902c9f6f2d6 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -36,18 +36,21 @@
 #include <asm/pci_clp.h>
 #include <asm/pci_dma.h>
 
+#include "pci_bus.h"
+
 /* list of all detected zpci devices */
 static LIST_HEAD(zpci_list);
 static DEFINE_SPINLOCK(zpci_list_lock);
 
 static DECLARE_BITMAP(zpci_domain, ZPCI_DOMAIN_BITMAP_SIZE);
 static DEFINE_SPINLOCK(zpci_domain_lock);
-static unsigned int zpci_num_domains_allocated;
 
 #define ZPCI_IOMAP_ENTRIES						\
 	min(((unsigned long) ZPCI_NR_DEVICES * PCI_STD_NUM_BARS / 2),	\
 	    ZPCI_IOMAP_MAX_ENTRIES)
 
+unsigned int s390_pci_no_rid;
+
 static DEFINE_SPINLOCK(zpci_iomap_lock);
 static unsigned long *zpci_iomap_bitmap;
 struct zpci_iomap_entry *zpci_iomap_start;
@@ -88,17 +91,12 @@ void zpci_remove_reserved_devices(void)
 	spin_unlock(&zpci_list_lock);
 
 	list_for_each_entry_safe(zdev, tmp, &remove, entry)
-		zpci_remove_device(zdev);
-}
-
-static struct zpci_dev *get_zdev_by_bus(struct pci_bus *bus)
-{
-	return (bus && bus->sysdata) ? (struct zpci_dev *) bus->sysdata : NULL;
+		zpci_zdev_put(zdev);
 }
 
 int pci_domain_nr(struct pci_bus *bus)
 {
-	return ((struct zpci_dev *) bus->sysdata)->domain;
+	return ((struct zpci_bus *) bus->sysdata)->domain_nr;
 }
 EXPORT_SYMBOL_GPL(pci_domain_nr);
 
@@ -228,28 +226,29 @@ void __iowrite64_copy(void __iomem *to, const void *from, size_t count)
        zpci_memcpy_toio(to, from, count);
 }
 
-void __iomem *ioremap(unsigned long ioaddr, unsigned long size)
+void __iomem *ioremap(phys_addr_t addr, size_t size)
 {
+	unsigned long offset, vaddr;
 	struct vm_struct *area;
-	unsigned long offset;
+	phys_addr_t last_addr;
 
-	if (!size)
+	last_addr = addr + size - 1;
+	if (!size || last_addr < addr)
 		return NULL;
 
 	if (!static_branch_unlikely(&have_mio))
-		return (void __iomem *) ioaddr;
+		return (void __iomem *) addr;
 
-	offset = ioaddr & ~PAGE_MASK;
-	ioaddr &= PAGE_MASK;
+	offset = addr & ~PAGE_MASK;
+	addr &= PAGE_MASK;
 	size = PAGE_ALIGN(size + offset);
 	area = get_vm_area(size, VM_IOREMAP);
 	if (!area)
 		return NULL;
 
-	if (ioremap_page_range((unsigned long) area->addr,
-			       (unsigned long) area->addr + size,
-			       ioaddr, PAGE_KERNEL)) {
-		vunmap(area->addr);
+	vaddr = (unsigned long) area->addr;
+	if (ioremap_page_range(vaddr, vaddr + size, addr, PAGE_KERNEL)) {
+		free_vm_area(area);
 		return NULL;
 	}
 	return (void __iomem *) ((unsigned long) area->addr + offset);
@@ -373,29 +372,17 @@ EXPORT_SYMBOL(pci_iounmap);
 static int pci_read(struct pci_bus *bus, unsigned int devfn, int where,
 		    int size, u32 *val)
 {
-	struct zpci_dev *zdev = get_zdev_by_bus(bus);
-	int ret;
+	struct zpci_dev *zdev = get_zdev_by_bus(bus, devfn);
 
-	if (!zdev || devfn != ZPCI_DEVFN)
-		ret = -ENODEV;
-	else
-		ret = zpci_cfg_load(zdev, where, val, size);
-
-	return ret;
+	return (zdev) ? zpci_cfg_load(zdev, where, val, size) : -ENODEV;
 }
 
 static int pci_write(struct pci_bus *bus, unsigned int devfn, int where,
 		     int size, u32 val)
 {
-	struct zpci_dev *zdev = get_zdev_by_bus(bus);
-	int ret;
+	struct zpci_dev *zdev = get_zdev_by_bus(bus, devfn);
 
-	if (!zdev || devfn != ZPCI_DEVFN)
-		ret = -ENODEV;
-	else
-		ret = zpci_cfg_store(zdev, where, val, size);
-
-	return ret;
+	return (zdev) ? zpci_cfg_store(zdev, where, val, size) : -ENODEV;
 }
 
 static struct pci_ops pci_root_ops = {
@@ -506,15 +493,15 @@ static struct resource *__alloc_res(struct zpci_dev *zdev, unsigned long start,
 	return r;
 }
 
-static int zpci_setup_bus_resources(struct zpci_dev *zdev,
-				    struct list_head *resources)
+int zpci_setup_bus_resources(struct zpci_dev *zdev,
+			     struct list_head *resources)
 {
 	unsigned long addr, size, flags;
 	struct resource *res;
 	int i, entry;
 
 	snprintf(zdev->res_name, sizeof(zdev->res_name),
-		 "PCI Bus %04x:%02x", zdev->domain, ZPCI_BUS_NR);
+		 "PCI Bus %04x:%02x", zdev->uid, ZPCI_BUS_NR);
 
 	for (i = 0; i < PCI_STD_NUM_BARS; i++) {
 		if (!zdev->bars[i].size)
@@ -608,98 +595,53 @@ void pcibios_disable_device(struct pci_dev *pdev)
 	zpci_debug_exit_device(zdev);
 }
 
-static int zpci_alloc_domain(struct zpci_dev *zdev)
+static int __zpci_register_domain(int domain)
 {
 	spin_lock(&zpci_domain_lock);
-	if (zpci_num_domains_allocated > (ZPCI_NR_DEVICES - 1)) {
+	if (test_bit(domain, zpci_domain)) {
 		spin_unlock(&zpci_domain_lock);
-		pr_err("Adding PCI function %08x failed because the configured limit of %d is reached\n",
-			zdev->fid, ZPCI_NR_DEVICES);
-		return -ENOSPC;
+		pr_err("Domain %04x is already assigned\n", domain);
+		return -EEXIST;
 	}
+	set_bit(domain, zpci_domain);
+	spin_unlock(&zpci_domain_lock);
+	return domain;
+}
 
-	if (zpci_unique_uid) {
-		zdev->domain = (u16) zdev->uid;
-		if (zdev->domain == 0) {
-			pr_warn("UID checking is active but no UID is set for PCI function %08x, so automatic domain allocation is used instead\n",
-				zdev->fid);
-			update_uid_checking(false);
-			goto auto_allocate;
-		}
+static int __zpci_alloc_domain(void)
+{
+	int domain;
 
-		if (test_bit(zdev->domain, zpci_domain)) {
-			spin_unlock(&zpci_domain_lock);
-			pr_err("Adding PCI function %08x failed because domain %04x is already assigned\n",
-				zdev->fid, zdev->domain);
-			return -EEXIST;
-		}
-		set_bit(zdev->domain, zpci_domain);
-		zpci_num_domains_allocated++;
-		spin_unlock(&zpci_domain_lock);
-		return 0;
-	}
-auto_allocate:
+	spin_lock(&zpci_domain_lock);
 	/*
 	 * We can always auto allocate domains below ZPCI_NR_DEVICES.
 	 * There is either a free domain or we have reached the maximum in
 	 * which case we would have bailed earlier.
 	 */
-	zdev->domain = find_first_zero_bit(zpci_domain, ZPCI_NR_DEVICES);
-	set_bit(zdev->domain, zpci_domain);
-	zpci_num_domains_allocated++;
+	domain = find_first_zero_bit(zpci_domain, ZPCI_NR_DEVICES);
+	set_bit(domain, zpci_domain);
 	spin_unlock(&zpci_domain_lock);
-	return 0;
+	return domain;
 }
 
-static void zpci_free_domain(struct zpci_dev *zdev)
+int zpci_alloc_domain(int domain)
 {
-	spin_lock(&zpci_domain_lock);
-	clear_bit(zdev->domain, zpci_domain);
-	zpci_num_domains_allocated--;
-	spin_unlock(&zpci_domain_lock);
+	if (zpci_unique_uid) {
+		if (domain)
+			return __zpci_register_domain(domain);
+		pr_warn("UID checking was active but no UID is provided: switching to automatic domain allocation\n");
+		update_uid_checking(false);
+	}
+	return __zpci_alloc_domain();
 }
 
-void pcibios_remove_bus(struct pci_bus *bus)
+void zpci_free_domain(int domain)
 {
-	struct zpci_dev *zdev = get_zdev_by_bus(bus);
-
-	zpci_exit_slot(zdev);
-	zpci_cleanup_bus_resources(zdev);
-	zpci_destroy_iommu(zdev);
-	zpci_free_domain(zdev);
-
-	spin_lock(&zpci_list_lock);
-	list_del(&zdev->entry);
-	spin_unlock(&zpci_list_lock);
-
-	zpci_dbg(3, "rem fid:%x\n", zdev->fid);
-	kfree(zdev);
+	spin_lock(&zpci_domain_lock);
+	clear_bit(domain, zpci_domain);
+	spin_unlock(&zpci_domain_lock);
 }
 
-static int zpci_scan_bus(struct zpci_dev *zdev)
-{
-	LIST_HEAD(resources);
-	int ret;
-
-	ret = zpci_setup_bus_resources(zdev, &resources);
-	if (ret)
-		goto error;
-
-	zdev->bus = pci_scan_root_bus(NULL, ZPCI_BUS_NR, &pci_root_ops,
-				      zdev, &resources);
-	if (!zdev->bus) {
-		ret = -EIO;
-		goto error;
-	}
-	zdev->bus->max_bus_speed = zdev->max_bus_speed;
-	pci_bus_add_devices(zdev->bus);
-	return 0;
-
-error:
-	zpci_cleanup_bus_resources(zdev);
-	pci_free_resource_list(&resources);
-	return ret;
-}
 
 int zpci_enable_device(struct zpci_dev *zdev)
 {
@@ -734,13 +676,15 @@ int zpci_create_device(struct zpci_dev *zdev)
 {
 	int rc;
 
-	rc = zpci_alloc_domain(zdev);
-	if (rc)
-		goto out;
+	kref_init(&zdev->kref);
+
+	spin_lock(&zpci_list_lock);
+	list_add_tail(&zdev->entry, &zpci_list);
+	spin_unlock(&zpci_list_lock);
 
 	rc = zpci_init_iommu(zdev);
 	if (rc)
-		goto out_free;
+		goto out;
 
 	mutex_init(&zdev->lock);
 	if (zdev->state == ZPCI_FN_STATE_CONFIGURED) {
@@ -748,36 +692,59 @@ int zpci_create_device(struct zpci_dev *zdev)
 		if (rc)
 			goto out_destroy_iommu;
 	}
-	rc = zpci_scan_bus(zdev);
+
+	rc = zpci_bus_device_register(zdev, &pci_root_ops);
 	if (rc)
 		goto out_disable;
 
-	spin_lock(&zpci_list_lock);
-	list_add_tail(&zdev->entry, &zpci_list);
-	spin_unlock(&zpci_list_lock);
-
-	zpci_init_slot(zdev);
-
 	return 0;
 
 out_disable:
 	if (zdev->state == ZPCI_FN_STATE_ONLINE)
 		zpci_disable_device(zdev);
+
 out_destroy_iommu:
 	zpci_destroy_iommu(zdev);
-out_free:
-	zpci_free_domain(zdev);
 out:
+	spin_lock(&zpci_list_lock);
+	list_del(&zdev->entry);
+	spin_unlock(&zpci_list_lock);
 	return rc;
 }
 
-void zpci_remove_device(struct zpci_dev *zdev)
+void zpci_release_device(struct kref *kref)
 {
-	if (!zdev->bus)
-		return;
+	struct zpci_dev *zdev = container_of(kref, struct zpci_dev, kref);
+
+	if (zdev->zbus->bus) {
+		struct pci_dev *pdev;
+
+		pdev = pci_get_slot(zdev->zbus->bus, zdev->devfn);
+		if (pdev)
+			pci_stop_and_remove_bus_device_locked(pdev);
+	}
+
+	switch (zdev->state) {
+	case ZPCI_FN_STATE_ONLINE:
+	case ZPCI_FN_STATE_CONFIGURED:
+		zpci_disable_device(zdev);
+		fallthrough;
+	case ZPCI_FN_STATE_STANDBY:
+		if (zdev->has_hp_slot)
+			zpci_exit_slot(zdev);
+		zpci_cleanup_bus_resources(zdev);
+		zpci_bus_device_unregister(zdev);
+		zpci_destroy_iommu(zdev);
+		fallthrough;
+	default:
+		break;
+	}
 
-	pci_stop_root_bus(zdev->bus);
-	pci_remove_root_bus(zdev->bus);
+	spin_lock(&zpci_list_lock);
+	list_del(&zdev->entry);
+	spin_unlock(&zpci_list_lock);
+	zpci_dbg(3, "rem fid:%x\n", zdev->fid);
+	kfree(zdev);
 }
 
 int zpci_report_error(struct pci_dev *pdev,
@@ -844,6 +811,10 @@ char * __init pcibios_setup(char *str)
 		s390_pci_force_floating = 1;
 		return NULL;
 	}
+	if (!strcmp(str, "norid")) {
+		s390_pci_no_rid = 1;
+		return NULL;
+	}
 	return str;
 }
 
diff --git a/arch/s390/pci/pci_bus.c b/arch/s390/pci/pci_bus.c
new file mode 100644
index 000000000000..642a99384688
--- /dev/null
+++ b/arch/s390/pci/pci_bus.c
@@ -0,0 +1,328 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright IBM Corp. 2020
+ *
+ * Author(s):
+ *   Pierre Morel <pmorel@linux.ibm.com>
+ *
+ */
+
+#define KMSG_COMPONENT "zpci"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/export.h>
+#include <linux/delay.h>
+#include <linux/seq_file.h>
+#include <linux/jump_label.h>
+#include <linux/pci.h>
+#include <linux/printk.h>
+
+#include <asm/pci_clp.h>
+#include <asm/pci_dma.h>
+
+#include "pci_bus.h"
+
+static LIST_HEAD(zbus_list);
+static DEFINE_SPINLOCK(zbus_list_lock);
+static int zpci_nb_devices;
+
+/* zpci_bus_scan
+ * @zbus: the zbus holding the zdevices
+ * @ops: the pci operations
+ *
+ * The domain number must be set before pci_scan_root_bus is called.
+ * This function can be called once the domain is known, hence
+ * when the function_0 is dicovered.
+ */
+static int zpci_bus_scan(struct zpci_bus *zbus, int domain, struct pci_ops *ops)
+{
+	struct pci_bus *bus;
+	int rc;
+
+	rc = zpci_alloc_domain(domain);
+	if (rc < 0)
+		return rc;
+	zbus->domain_nr = rc;
+
+	bus = pci_scan_root_bus(NULL, ZPCI_BUS_NR, ops, zbus, &zbus->resources);
+	if (!bus) {
+		zpci_free_domain(zbus->domain_nr);
+		return -EFAULT;
+	}
+
+	zbus->bus = bus;
+	pci_bus_add_devices(bus);
+	return 0;
+}
+
+static void zpci_bus_release(struct kref *kref)
+{
+	struct zpci_bus *zbus = container_of(kref, struct zpci_bus, kref);
+
+	if (zbus->bus) {
+		pci_lock_rescan_remove();
+		pci_stop_root_bus(zbus->bus);
+
+		zpci_free_domain(zbus->domain_nr);
+		pci_free_resource_list(&zbus->resources);
+
+		pci_remove_root_bus(zbus->bus);
+		pci_unlock_rescan_remove();
+	}
+
+	spin_lock(&zbus_list_lock);
+	list_del(&zbus->bus_next);
+	spin_unlock(&zbus_list_lock);
+	kfree(zbus);
+}
+
+static void zpci_bus_put(struct zpci_bus *zbus)
+{
+	kref_put(&zbus->kref, zpci_bus_release);
+}
+
+static struct zpci_bus *zpci_bus_get(int pchid)
+{
+	struct zpci_bus *zbus;
+
+	spin_lock(&zbus_list_lock);
+	list_for_each_entry(zbus, &zbus_list, bus_next) {
+		if (pchid == zbus->pchid) {
+			kref_get(&zbus->kref);
+			goto out_unlock;
+		}
+	}
+	zbus = NULL;
+out_unlock:
+	spin_unlock(&zbus_list_lock);
+	return zbus;
+}
+
+static struct zpci_bus *zpci_bus_alloc(int pchid)
+{
+	struct zpci_bus *zbus;
+
+	zbus = kzalloc(sizeof(*zbus), GFP_KERNEL);
+	if (!zbus)
+		return NULL;
+
+	zbus->pchid = pchid;
+	INIT_LIST_HEAD(&zbus->bus_next);
+	spin_lock(&zbus_list_lock);
+	list_add_tail(&zbus->bus_next, &zbus_list);
+	spin_unlock(&zbus_list_lock);
+
+	kref_init(&zbus->kref);
+	INIT_LIST_HEAD(&zbus->resources);
+
+	zbus->bus_resource.start = 0;
+	zbus->bus_resource.end = ZPCI_BUS_NR;
+	zbus->bus_resource.flags = IORESOURCE_BUS;
+	pci_add_resource(&zbus->resources, &zbus->bus_resource);
+
+	return zbus;
+}
+
+#ifdef CONFIG_PCI_IOV
+static int zpci_bus_link_virtfn(struct pci_dev *pdev,
+		struct pci_dev *virtfn, int vfid)
+{
+	int rc;
+
+	virtfn->physfn = pci_dev_get(pdev);
+	rc = pci_iov_sysfs_link(pdev, virtfn, vfid);
+	if (rc) {
+		pci_dev_put(pdev);
+		virtfn->physfn = NULL;
+		return rc;
+	}
+	return 0;
+}
+
+static int zpci_bus_setup_virtfn(struct zpci_bus *zbus,
+		struct pci_dev *virtfn, int vfn)
+{
+	int i, cand_devfn;
+	struct zpci_dev *zdev;
+	struct pci_dev *pdev;
+	int vfid = vfn - 1; /* Linux' vfid's start at 0 vfn at 1*/
+	int rc = 0;
+
+	virtfn->is_virtfn = 1;
+	virtfn->multifunction = 0;
+	WARN_ON(vfid < 0);
+	/* If the parent PF for the given VF is also configured in the
+	 * instance, it must be on the same zbus.
+	 * We can then identify the parent PF by checking what
+	 * devfn the VF would have if it belonged to that PF using the PF's
+	 * stride and offset. Only if this candidate devfn matches the
+	 * actual devfn will we link both functions.
+	 */
+	for (i = 0; i < ZPCI_FUNCTIONS_PER_BUS; i++) {
+		zdev = zbus->function[i];
+		if (zdev && zdev->is_physfn) {
+			pdev = pci_get_slot(zbus->bus, zdev->devfn);
+			cand_devfn = pci_iov_virtfn_devfn(pdev, vfid);
+			if (cand_devfn == virtfn->devfn) {
+				rc = zpci_bus_link_virtfn(pdev, virtfn, vfid);
+				break;
+			}
+		}
+	}
+	return rc;
+}
+#else
+static inline int zpci_bus_setup_virtfn(struct zpci_bus *zbus,
+		struct pci_dev *virtfn, int vfn)
+{
+	virtfn->is_virtfn = 1;
+	virtfn->multifunction = 0;
+	return 0;
+}
+#endif
+
+static int zpci_bus_add_device(struct zpci_bus *zbus, struct zpci_dev *zdev)
+{
+	struct pci_bus *bus;
+	struct resource_entry *window, *n;
+	struct resource *res;
+	struct pci_dev *pdev;
+	int rc;
+
+	bus = zbus->bus;
+	if (!bus)
+		return -EINVAL;
+
+	pdev = pci_get_slot(bus, zdev->devfn);
+	if (pdev) {
+		/* Device is already known. */
+		pci_dev_put(pdev);
+		return 0;
+	}
+
+	rc = zpci_init_slot(zdev);
+	if (rc)
+		return rc;
+	zdev->has_hp_slot = 1;
+
+	resource_list_for_each_entry_safe(window, n, &zbus->resources) {
+		res = window->res;
+		pci_bus_add_resource(bus, res, 0);
+	}
+
+	pdev = pci_scan_single_device(bus, zdev->devfn);
+	if (pdev) {
+		if (!zdev->is_physfn) {
+			rc = zpci_bus_setup_virtfn(zbus, pdev, zdev->vfn);
+			if (rc)
+				goto failed_with_pdev;
+		}
+		pci_bus_add_device(pdev);
+	}
+	return 0;
+
+failed_with_pdev:
+	pci_stop_and_remove_bus_device(pdev);
+	pci_dev_put(pdev);
+	return rc;
+}
+
+static void zpci_bus_add_devices(struct zpci_bus *zbus)
+{
+	int i;
+
+	for (i = 1; i < ZPCI_FUNCTIONS_PER_BUS; i++)
+		if (zbus->function[i])
+			zpci_bus_add_device(zbus, zbus->function[i]);
+
+	pci_lock_rescan_remove();
+	pci_bus_add_devices(zbus->bus);
+	pci_unlock_rescan_remove();
+}
+
+int zpci_bus_device_register(struct zpci_dev *zdev, struct pci_ops *ops)
+{
+	struct zpci_bus *zbus = NULL;
+	int rc = -EBADF;
+
+	if (zpci_nb_devices == ZPCI_NR_DEVICES) {
+		pr_warn("Adding PCI function %08x failed because the configured limit of %d is reached\n",
+			zdev->fid, ZPCI_NR_DEVICES);
+		return -ENOSPC;
+	}
+	zpci_nb_devices++;
+
+	if (zdev->devfn >= ZPCI_FUNCTIONS_PER_BUS)
+		return -EINVAL;
+
+	if (!s390_pci_no_rid && zdev->rid_available)
+		zbus = zpci_bus_get(zdev->pchid);
+
+	if (!zbus) {
+		zbus = zpci_bus_alloc(zdev->pchid);
+		if (!zbus)
+			return -ENOMEM;
+	}
+
+	zdev->zbus = zbus;
+	if (zbus->function[zdev->devfn]) {
+		pr_err("devfn %04x is already assigned\n", zdev->devfn);
+		goto error; /* rc already set */
+	}
+	zbus->function[zdev->devfn] = zdev;
+
+	zpci_setup_bus_resources(zdev, &zbus->resources);
+
+	if (zbus->bus) {
+		if (!zbus->multifunction) {
+			WARN_ONCE(1, "zbus is not multifunction\n");
+			goto error_bus;
+		}
+		if (!zdev->rid_available) {
+			WARN_ONCE(1, "rid_available not set for multifunction\n");
+			goto error_bus;
+		}
+		rc = zpci_bus_add_device(zbus, zdev);
+		if (rc)
+			goto error_bus;
+	} else if (zdev->devfn == 0) {
+		if (zbus->multifunction && !zdev->rid_available) {
+			WARN_ONCE(1, "rid_available not set on function 0 for multifunction\n");
+			goto error_bus;
+		}
+		rc = zpci_bus_scan(zbus, (u16)zdev->uid, ops);
+		if (rc)
+			goto error_bus;
+		zpci_bus_add_devices(zbus);
+		rc = zpci_init_slot(zdev);
+		if (rc)
+			goto error_bus;
+		zdev->has_hp_slot = 1;
+		zbus->multifunction = zdev->rid_available;
+		zbus->max_bus_speed = zdev->max_bus_speed;
+	} else {
+		zbus->multifunction = 1;
+	}
+
+	return 0;
+
+error_bus:
+	zpci_nb_devices--;
+	zbus->function[zdev->devfn] = NULL;
+error:
+	pr_err("Adding PCI function %08x failed\n", zdev->fid);
+	zpci_bus_put(zbus);
+	return rc;
+}
+
+void zpci_bus_device_unregister(struct zpci_dev *zdev)
+{
+	struct zpci_bus *zbus = zdev->zbus;
+
+	zpci_nb_devices--;
+	zbus->function[zdev->devfn] = NULL;
+	zpci_bus_put(zbus);
+}
diff --git a/arch/s390/pci/pci_bus.h b/arch/s390/pci/pci_bus.h
new file mode 100644
index 000000000000..89be3c354b7b
--- /dev/null
+++ b/arch/s390/pci/pci_bus.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright IBM Corp. 2020
+ *
+ * Author(s):
+ *   Pierre Morel <pmorel@linux.ibm.com>
+ *
+ */
+
+int zpci_bus_device_register(struct zpci_dev *zdev, struct pci_ops *ops);
+void zpci_bus_device_unregister(struct zpci_dev *zdev);
+int zpci_bus_init(void);
+
+void zpci_release_device(struct kref *kref);
+static inline void zpci_zdev_put(struct zpci_dev *zdev)
+{
+	kref_put(&zdev->kref, zpci_release_device);
+}
+
+int zpci_alloc_domain(int domain);
+void zpci_free_domain(int domain);
+int zpci_setup_bus_resources(struct zpci_dev *zdev,
+			     struct list_head *resources);
+
+static inline struct zpci_dev *get_zdev_by_bus(struct pci_bus *bus,
+					       unsigned int devfn)
+{
+	struct zpci_bus *zbus = bus->sysdata;
+
+	return (devfn >= ZPCI_FUNCTIONS_PER_BUS) ? NULL : zbus->function[devfn];
+}
diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c
index ea794ae755ae..7e735f41a0a6 100644
--- a/arch/s390/pci/pci_clp.c
+++ b/arch/s390/pci/pci_clp.c
@@ -155,8 +155,13 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev,
 	zdev->pfgid = response->pfgid;
 	zdev->pft = response->pft;
 	zdev->vfn = response->vfn;
+	zdev->port = response->port;
 	zdev->uid = response->uid;
 	zdev->fmb_length = sizeof(u32) * response->fmb_len;
+	zdev->rid_available = response->rid_avail;
+	zdev->is_physfn = response->is_physfn;
+	if (!s390_pci_no_rid && zdev->rid_available)
+		zdev->devfn = response->rid & ZPCI_RID_MASK_DEVFN;
 
 	memcpy(zdev->pfip, response->pfip, sizeof(zdev->pfip));
 	if (response->util_str_avail) {
@@ -309,14 +314,13 @@ out:
 
 int clp_disable_fh(struct zpci_dev *zdev)
 {
-	u32 fh = zdev->fh;
 	int rc;
 
 	if (!zdev_enabled(zdev))
 		return 0;
 
 	rc = clp_set_pci_fn(zdev, 0, CLP_SET_DISABLE_PCI_FN);
-	zpci_dbg(3, "dis fid:%x, fh:%x, rc:%d\n", zdev->fid, fh, rc);
+	zpci_dbg(3, "dis fid:%x, fh:%x, rc:%d\n", zdev->fid, zdev->fh, rc);
 	return rc;
 }
 
diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c
index 8d6ee4af4230..08e1d619398e 100644
--- a/arch/s390/pci/pci_event.c
+++ b/arch/s390/pci/pci_event.c
@@ -14,6 +14,8 @@
 #include <asm/pci_debug.h>
 #include <asm/sclp.h>
 
+#include "pci_bus.h"
+
 /* Content Code Description for PCI Function Error */
 struct zpci_ccdf_err {
 	u32 reserved1;
@@ -53,7 +55,7 @@ static void __zpci_event_error(struct zpci_ccdf_err *ccdf)
 	zpci_err_hex(ccdf, sizeof(*ccdf));
 
 	if (zdev)
-		pdev = pci_get_slot(zdev->bus, ZPCI_DEVFN);
+		pdev = pci_get_slot(zdev->zbus->bus, zdev->devfn);
 
 	pr_err("%s: Event 0x%x reports an error for PCI function 0x%x\n",
 	       pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid);
@@ -78,36 +80,28 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf)
 	enum zpci_state state;
 	int ret;
 
-	if (zdev)
-		pdev = pci_get_slot(zdev->bus, ZPCI_DEVFN);
+	if (zdev && zdev->zbus && zdev->zbus->bus)
+		pdev = pci_get_slot(zdev->zbus->bus, zdev->devfn);
 
-	pr_info("%s: Event 0x%x reconfigured PCI function 0x%x\n",
-		pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid);
 	zpci_err("avail CCDF:\n");
 	zpci_err_hex(ccdf, sizeof(*ccdf));
 
 	switch (ccdf->pec) {
 	case 0x0301: /* Reserved|Standby -> Configured */
 		if (!zdev) {
-			ret = clp_add_pci_device(ccdf->fid, ccdf->fh, 0);
-			if (ret)
-				break;
-			zdev = get_zdev_by_fid(ccdf->fid);
-		}
-		if (!zdev || zdev->state != ZPCI_FN_STATE_STANDBY)
+			ret = clp_add_pci_device(ccdf->fid, ccdf->fh, 1);
 			break;
-		zdev->state = ZPCI_FN_STATE_CONFIGURED;
+		}
 		zdev->fh = ccdf->fh;
-		ret = zpci_enable_device(zdev);
-		if (ret)
-			break;
-		pci_lock_rescan_remove();
-		pci_rescan_bus(zdev->bus);
-		pci_unlock_rescan_remove();
+		zdev->state = ZPCI_FN_STATE_CONFIGURED;
+		zpci_create_device(zdev);
 		break;
 	case 0x0302: /* Reserved -> Standby */
-		if (!zdev)
+		if (!zdev) {
 			clp_add_pci_device(ccdf->fid, ccdf->fh, 0);
+			break;
+		}
+		zdev->fh = ccdf->fh;
 		break;
 	case 0x0303: /* Deconfiguration requested */
 		if (!zdev)
@@ -135,12 +129,10 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf)
 			pci_stop_and_remove_bus_device_locked(pdev);
 		}
 
-		zdev->fh = ccdf->fh;
-		zpci_disable_device(zdev);
 		zdev->state = ZPCI_FN_STATE_STANDBY;
 		if (!clp_get_state(ccdf->fid, &state) &&
 		    state == ZPCI_FN_STATE_RESERVED) {
-			zpci_remove_device(zdev);
+			zpci_zdev_put(zdev);
 		}
 		break;
 	case 0x0306: /* 0x308 or 0x302 for multiple devices */
@@ -149,12 +141,11 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf)
 	case 0x0308: /* Standby -> Reserved */
 		if (!zdev)
 			break;
-		zpci_remove_device(zdev);
+		zpci_zdev_put(zdev);
 		break;
 	default:
 		break;
 	}
-	pci_dev_put(pdev);
 }
 
 void zpci_event_availability(void *data)
diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c
index 215f17437a4f..5c028bee91b9 100644
--- a/arch/s390/pci/pci_sysfs.c
+++ b/arch/s390/pci/pci_sysfs.c
@@ -33,6 +33,7 @@ zpci_attr(pchid, "0x%04x\n", pchid);
 zpci_attr(pfgid, "0x%02x\n", pfgid);
 zpci_attr(vfn, "0x%04x\n", vfn);
 zpci_attr(pft, "0x%02x\n", pft);
+zpci_attr(port, "%d\n", port);
 zpci_attr(uid, "0x%x\n", uid);
 zpci_attr(segment0, "0x%02x\n", pfip[0]);
 zpci_attr(segment1, "0x%02x\n", pfip[1]);
@@ -88,7 +89,7 @@ static ssize_t recover_store(struct device *dev, struct device_attribute *attr,
 		ret = zpci_enable_device(zdev);
 		if (ret)
 			goto out;
-		pci_rescan_bus(zdev->bus);
+		pci_rescan_bus(zdev->zbus->bus);
 	}
 out:
 	pci_unlock_rescan_remove();
@@ -142,6 +143,7 @@ static struct attribute *zpci_dev_attrs[] = {
 	&dev_attr_pchid.attr,
 	&dev_attr_pfgid.attr,
 	&dev_attr_pft.attr,
+	&dev_attr_port.attr,
 	&dev_attr_vfn.attr,
 	&dev_attr_uid.attr,
 	&dev_attr_recover.attr,