diff options
Diffstat (limited to 'drivers/net/ethernet/intel/fm10k/fm10k_pci.c')
| -rw-r--r-- | drivers/net/ethernet/intel/fm10k/fm10k_pci.c | 678 |
1 files changed, 398 insertions, 280 deletions
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 63784576ae8b..d75b8a50413d 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -1,26 +1,8 @@ -/* Intel(R) Ethernet Switch Host Interface Driver - * Copyright(c) 2013 - 2017 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * The full GNU General Public License is included in this distribution in - * the file called "COPYING". - * - * Contact Information: - * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - */ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2013 - 2019 Intel Corporation. */ #include <linux/module.h> #include <linux/interrupt.h> -#include <linux/aer.h> #include "fm10k.h" @@ -29,7 +11,7 @@ static const struct fm10k_info *fm10k_info_tbl[] = { [fm10k_device_vf] = &fm10k_vf_info, }; -/** +/* * fm10k_pci_tbl - PCI Device ID Table * * Wildcard entries (PCI_ANY_ID) should come last @@ -40,6 +22,8 @@ static const struct fm10k_info *fm10k_info_tbl[] = { */ static const struct pci_device_id fm10k_pci_tbl[] = { { PCI_VDEVICE(INTEL, FM10K_DEV_ID_PF), fm10k_device_pf }, + { PCI_VDEVICE(INTEL, FM10K_DEV_ID_SDI_FM10420_QDA2), fm10k_device_pf }, + { PCI_VDEVICE(INTEL, FM10K_DEV_ID_SDI_FM10420_DA2), fm10k_device_pf }, { PCI_VDEVICE(INTEL, FM10K_DEV_ID_VF), fm10k_device_vf }, /* required last entry */ { 0, } @@ -91,6 +75,76 @@ static int fm10k_hw_ready(struct fm10k_intfc *interface) return FM10K_REMOVED(hw->hw_addr) ? -ENODEV : 0; } +/** + * fm10k_macvlan_schedule - Schedule MAC/VLAN queue task + * @interface: fm10k private interface structure + * + * Schedule the MAC/VLAN queue monitor task. If the MAC/VLAN task cannot be + * started immediately, request that it be restarted when possible. + */ +void fm10k_macvlan_schedule(struct fm10k_intfc *interface) +{ + /* Avoid processing the MAC/VLAN queue when the service task is + * disabled, or when we're resetting the device. + */ + if (!test_bit(__FM10K_MACVLAN_DISABLE, interface->state) && + !test_and_set_bit(__FM10K_MACVLAN_SCHED, interface->state)) { + clear_bit(__FM10K_MACVLAN_REQUEST, interface->state); + /* We delay the actual start of execution in order to allow + * multiple MAC/VLAN updates to accumulate before handling + * them, and to allow some time to let the mailbox drain + * between runs. + */ + queue_delayed_work(fm10k_workqueue, + &interface->macvlan_task, 10); + } else { + set_bit(__FM10K_MACVLAN_REQUEST, interface->state); + } +} + +/** + * fm10k_stop_macvlan_task - Stop the MAC/VLAN queue monitor + * @interface: fm10k private interface structure + * + * Wait until the MAC/VLAN queue task has stopped, and cancel any future + * requests. + */ +static void fm10k_stop_macvlan_task(struct fm10k_intfc *interface) +{ + /* Disable the MAC/VLAN work item */ + set_bit(__FM10K_MACVLAN_DISABLE, interface->state); + + /* Make sure we waited until any current invocations have stopped */ + cancel_delayed_work_sync(&interface->macvlan_task); + + /* We set the __FM10K_MACVLAN_SCHED bit when we schedule the task. + * However, it may not be unset of the MAC/VLAN task never actually + * got a chance to run. Since we've canceled the task here, and it + * cannot be rescheuled right now, we need to ensure the scheduled bit + * gets unset. + */ + clear_bit(__FM10K_MACVLAN_SCHED, interface->state); +} + +/** + * fm10k_resume_macvlan_task - Restart the MAC/VLAN queue monitor + * @interface: fm10k private interface structure + * + * Clear the __FM10K_MACVLAN_DISABLE bit and, if a request occurred, schedule + * the MAC/VLAN work monitor. + */ +static void fm10k_resume_macvlan_task(struct fm10k_intfc *interface) +{ + /* Re-enable the MAC/VLAN work item */ + clear_bit(__FM10K_MACVLAN_DISABLE, interface->state); + + /* We might have received a MAC/VLAN request while disabled. If so, + * kick off the queue now. + */ + if (test_bit(__FM10K_MACVLAN_REQUEST, interface->state)) + fm10k_macvlan_schedule(interface); +} + void fm10k_service_event_schedule(struct fm10k_intfc *interface) { if (!test_bit(__FM10K_SERVICE_DISABLE, interface->state) && @@ -118,60 +172,66 @@ static void fm10k_service_event_complete(struct fm10k_intfc *interface) fm10k_service_event_schedule(interface); } -/** - * fm10k_service_timer - Timer Call-back - * @data: pointer to interface cast into an unsigned long - **/ -static void fm10k_service_timer(unsigned long data) +static void fm10k_stop_service_event(struct fm10k_intfc *interface) { - struct fm10k_intfc *interface = (struct fm10k_intfc *)data; + set_bit(__FM10K_SERVICE_DISABLE, interface->state); + cancel_work_sync(&interface->service_task); - /* Reset the timer */ - mod_timer(&interface->service_timer, (HZ * 2) + jiffies); + /* It's possible that cancel_work_sync stopped the service task from + * running before it could actually start. In this case the + * __FM10K_SERVICE_SCHED bit will never be cleared. Since we know that + * the service task cannot be running at this point, we need to clear + * the scheduled bit, as otherwise the service task may never be + * restarted. + */ + clear_bit(__FM10K_SERVICE_SCHED, interface->state); +} +static void fm10k_start_service_event(struct fm10k_intfc *interface) +{ + clear_bit(__FM10K_SERVICE_DISABLE, interface->state); fm10k_service_event_schedule(interface); } -static void fm10k_detach_subtask(struct fm10k_intfc *interface) +/** + * fm10k_service_timer - Timer Call-back + * @t: pointer to timer data + **/ +static void fm10k_service_timer(struct timer_list *t) { - struct net_device *netdev = interface->netdev; - u32 __iomem *hw_addr; - u32 value; - - /* do nothing if device is still present or hw_addr is set */ - if (netif_device_present(netdev) || interface->hw.hw_addr) - return; - - /* check the real address space to see if we've recovered */ - hw_addr = READ_ONCE(interface->uc_addr); - value = readl(hw_addr); - if (~value) { - interface->hw.hw_addr = interface->uc_addr; - netif_device_attach(netdev); - set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags); - netdev_warn(netdev, "PCIe link restored, device now attached\n"); - return; - } + struct fm10k_intfc *interface = timer_container_of(interface, t, + service_timer); - rtnl_lock(); - - if (netif_running(netdev)) - dev_close(netdev); + /* Reset the timer */ + mod_timer(&interface->service_timer, (HZ * 2) + jiffies); - rtnl_unlock(); + fm10k_service_event_schedule(interface); } -static void fm10k_prepare_for_reset(struct fm10k_intfc *interface) +/** + * fm10k_prepare_for_reset - Prepare the driver and device for a pending reset + * @interface: fm10k private data structure + * + * This function prepares for a device reset by shutting as much down as we + * can. It does nothing and returns false if __FM10K_RESETTING was already set + * prior to calling this function. It returns true if it actually did work. + */ +static bool fm10k_prepare_for_reset(struct fm10k_intfc *interface) { struct net_device *netdev = interface->netdev; - WARN_ON(in_interrupt()); - /* put off any impending NetWatchDogTimeout */ netif_trans_update(netdev); - while (test_and_set_bit(__FM10K_RESETTING, interface->state)) - usleep_range(1000, 2000); + /* Nothing to do if a reset is already in progress */ + if (test_and_set_bit(__FM10K_RESETTING, interface->state)) + return false; + + /* As the MAC/VLAN task will be accessing registers it must not be + * running while we reset. Although the task will not be scheduled + * once we start resetting it may already be running + */ + fm10k_stop_macvlan_task(interface); rtnl_lock(); @@ -189,6 +249,8 @@ static void fm10k_prepare_for_reset(struct fm10k_intfc *interface) interface->last_reset = jiffies + (10 * HZ); rtnl_unlock(); + + return true; } static int fm10k_handle_reset(struct fm10k_intfc *interface) @@ -197,6 +259,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface) struct fm10k_hw *hw = &interface->hw; int err; + WARN_ON(!test_bit(__FM10K_RESETTING, interface->state)); + rtnl_lock(); pci_set_master(interface->pdev); @@ -235,7 +299,7 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface) if (is_valid_ether_addr(hw->mac.perm_addr)) { ether_addr_copy(hw->mac.addr, hw->mac.perm_addr); ether_addr_copy(netdev->perm_addr, hw->mac.perm_addr); - ether_addr_copy(netdev->dev_addr, hw->mac.perm_addr); + eth_hw_addr_set(netdev, hw->mac.perm_addr); netdev->addr_assign_type &= ~NET_ADDR_RANDOM; } @@ -253,6 +317,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface) rtnl_unlock(); + fm10k_resume_macvlan_task(interface); + clear_bit(__FM10K_RESETTING, interface->state); return err; @@ -270,27 +336,81 @@ reinit_err: return err; } -static void fm10k_reinit(struct fm10k_intfc *interface) +static void fm10k_detach_subtask(struct fm10k_intfc *interface) { - int err; + struct net_device *netdev = interface->netdev; + u32 __iomem *hw_addr; + u32 value; - fm10k_prepare_for_reset(interface); + /* do nothing if netdev is still present or hw_addr is set */ + if (netif_device_present(netdev) || interface->hw.hw_addr) + return; - err = fm10k_handle_reset(interface); - if (err) - dev_err(&interface->pdev->dev, - "fm10k_handle_reset failed: %d\n", err); + /* We've lost the PCIe register space, and can no longer access the + * device. Shut everything except the detach subtask down and prepare + * to reset the device in case we recover. If we actually prepare for + * reset, indicate that we're detached. + */ + if (fm10k_prepare_for_reset(interface)) + set_bit(__FM10K_RESET_DETACHED, interface->state); + + /* check the real address space to see if we've recovered */ + hw_addr = READ_ONCE(interface->uc_addr); + value = readl(hw_addr); + if (~value) { + int err; + + /* Make sure the reset was initiated because we detached, + * otherwise we might race with a different reset flow. + */ + if (!test_and_clear_bit(__FM10K_RESET_DETACHED, + interface->state)) + return; + + /* Restore the hardware address */ + interface->hw.hw_addr = interface->uc_addr; + + /* PCIe link has been restored, and the device is active + * again. Restore everything and reset the device. + */ + err = fm10k_handle_reset(interface); + if (err) { + netdev_err(netdev, "Unable to reset device: %d\n", err); + interface->hw.hw_addr = NULL; + return; + } + + /* Re-attach the netdev */ + netif_device_attach(netdev); + netdev_warn(netdev, "PCIe link restored, device now attached\n"); + return; + } } static void fm10k_reset_subtask(struct fm10k_intfc *interface) { + int err; + if (!test_and_clear_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags)) return; + /* If another thread has already prepared to reset the device, we + * should not attempt to handle a reset here, since we'd race with + * that thread. This may happen if we suspend the device or if the + * PCIe link is lost. In this case, we'll just ignore the RESET + * request, as it will (eventually) be taken care of when the thread + * which actually started the reset is finished. + */ + if (!fm10k_prepare_for_reset(interface)) + return; + netdev_err(interface->netdev, "Reset interface\n"); - fm10k_reinit(interface); + err = fm10k_handle_reset(interface); + if (err) + dev_err(&interface->pdev->dev, + "fm10k_handle_reset failed: %d\n", err); } /** @@ -360,6 +480,10 @@ static void fm10k_watchdog_update_host_state(struct fm10k_intfc *interface) **/ static void fm10k_mbx_subtask(struct fm10k_intfc *interface) { + /* If we're resetting, bail out */ + if (test_bit(__FM10K_RESETTING, interface->state)) + return; + /* process upstream mailbox and update device state */ fm10k_watchdog_update_host_state(interface); @@ -503,12 +627,15 @@ void fm10k_update_stats(struct fm10k_intfc *interface) net_stats->rx_errors = rx_errors; net_stats->rx_dropped = interface->stats.nodesc_drop.count; + /* Update VF statistics */ + fm10k_iov_update_stats(interface); + clear_bit(__FM10K_UPDATING_STATS, interface->state); } /** * fm10k_watchdog_flush_tx - flush queues on host not ready - * @interface - pointer to the device interface structure + * @interface: pointer to the device interface structure **/ static void fm10k_watchdog_flush_tx(struct fm10k_intfc *interface) { @@ -538,7 +665,7 @@ static void fm10k_watchdog_flush_tx(struct fm10k_intfc *interface) /** * fm10k_watchdog_subtask - check and bring link up - * @interface - pointer to the device interface structure + * @interface: pointer to the device interface structure **/ static void fm10k_watchdog_subtask(struct fm10k_intfc *interface) { @@ -562,7 +689,7 @@ static void fm10k_watchdog_subtask(struct fm10k_intfc *interface) /** * fm10k_check_hang_subtask - check for hung queues and dropped interrupts - * @interface - pointer to the device interface structure + * @interface: pointer to the device interface structure * * This function serves two purposes. First it strobes the interrupt lines * in order to make certain interrupts are occurring. Secondly it sets the @@ -571,8 +698,6 @@ static void fm10k_watchdog_subtask(struct fm10k_intfc *interface) */ static void fm10k_check_hang_subtask(struct fm10k_intfc *interface) { - int i; - /* If we're down or resetting, just bail */ if (test_bit(__FM10K_DOWN, interface->state) || test_bit(__FM10K_RESETTING, interface->state)) @@ -584,6 +709,8 @@ static void fm10k_check_hang_subtask(struct fm10k_intfc *interface) interface->next_tx_hang_check = jiffies + (2 * HZ); if (netif_carrier_ok(interface->netdev)) { + int i; + /* Force detection of hung controller */ for (i = 0; i < interface->num_tx_queues; i++) set_check_for_tx_hang(interface->tx_ring[i]); @@ -609,9 +736,11 @@ static void fm10k_service_task(struct work_struct *work) interface = container_of(work, struct fm10k_intfc, service_task); + /* Check whether we're detached first */ + fm10k_detach_subtask(interface); + /* tasks run even when interface is down */ fm10k_mbx_subtask(interface); - fm10k_detach_subtask(interface); fm10k_reset_subtask(interface); /* tasks only run when interface is up */ @@ -623,6 +752,112 @@ static void fm10k_service_task(struct work_struct *work) } /** + * fm10k_macvlan_task - send queued MAC/VLAN requests to switch manager + * @work: pointer to work_struct containing our data + * + * This work item handles sending MAC/VLAN updates to the switch manager. When + * the interface is up, it will attempt to queue mailbox messages to the + * switch manager requesting updates for MAC/VLAN pairs. If the Tx fifo of the + * mailbox is full, it will reschedule itself to try again in a short while. + * This ensures that the driver does not overload the switch mailbox with too + * many simultaneous requests, causing an unnecessary reset. + **/ +static void fm10k_macvlan_task(struct work_struct *work) +{ + struct fm10k_macvlan_request *item; + struct fm10k_intfc *interface; + struct delayed_work *dwork; + struct list_head *requests; + struct fm10k_hw *hw; + unsigned long flags; + + dwork = to_delayed_work(work); + interface = container_of(dwork, struct fm10k_intfc, macvlan_task); + hw = &interface->hw; + requests = &interface->macvlan_requests; + + do { + /* Pop the first item off the list */ + spin_lock_irqsave(&interface->macvlan_lock, flags); + item = list_first_entry_or_null(requests, + struct fm10k_macvlan_request, + list); + if (item) + list_del_init(&item->list); + + spin_unlock_irqrestore(&interface->macvlan_lock, flags); + + /* We have no more items to process */ + if (!item) + goto done; + + fm10k_mbx_lock(interface); + + /* Check that we have plenty of space to send the message. We + * want to ensure that the mailbox stays low enough to avoid a + * change in the host state, otherwise we may see spurious + * link up / link down notifications. + */ + if (!hw->mbx.ops.tx_ready(&hw->mbx, FM10K_VFMBX_MSG_MTU + 5)) { + hw->mbx.ops.process(hw, &hw->mbx); + set_bit(__FM10K_MACVLAN_REQUEST, interface->state); + fm10k_mbx_unlock(interface); + + /* Put the request back on the list */ + spin_lock_irqsave(&interface->macvlan_lock, flags); + list_add(&item->list, requests); + spin_unlock_irqrestore(&interface->macvlan_lock, flags); + break; + } + + switch (item->type) { + case FM10K_MC_MAC_REQUEST: + hw->mac.ops.update_mc_addr(hw, + item->mac.glort, + item->mac.addr, + item->mac.vid, + item->set); + break; + case FM10K_UC_MAC_REQUEST: + hw->mac.ops.update_uc_addr(hw, + item->mac.glort, + item->mac.addr, + item->mac.vid, + item->set, + 0); + break; + case FM10K_VLAN_REQUEST: + hw->mac.ops.update_vlan(hw, + item->vlan.vid, + item->vlan.vsi, + item->set); + break; + default: + break; + } + + fm10k_mbx_unlock(interface); + + /* Free the item now that we've sent the update */ + kfree(item); + } while (true); + +done: + WARN_ON(!test_bit(__FM10K_MACVLAN_SCHED, interface->state)); + + /* flush memory to make sure state is correct */ + smp_mb__before_atomic(); + clear_bit(__FM10K_MACVLAN_SCHED, interface->state); + + /* If a MAC/VLAN request was scheduled since we started, we should + * re-schedule. However, there is no reason to re-schedule if there is + * no work to do. + */ + if (test_bit(__FM10K_MACVLAN_REQUEST, interface->state)) + fm10k_macvlan_schedule(interface); +} + +/** * fm10k_configure_tx_ring - Configure Tx ring after Reset * @interface: board private structure * @ring: structure containing ring specific data @@ -663,7 +898,7 @@ static void fm10k_configure_tx_ring(struct fm10k_intfc *interface, /* Map interrupt */ if (ring->q_vector) { - txint = ring->q_vector->v_idx + NON_Q_VECTORS(hw); + txint = ring->q_vector->v_idx + NON_Q_VECTORS; txint |= FM10K_INT_MAP_TIMER0; } @@ -802,7 +1037,7 @@ static void fm10k_configure_rx_ring(struct fm10k_intfc *interface, /* Map interrupt */ if (ring->q_vector) { - rxint = ring->q_vector->v_idx + NON_Q_VECTORS(hw); + rxint = ring->q_vector->v_idx + NON_Q_VECTORS; rxint |= FM10K_INT_MAP_TIMER1; } @@ -978,28 +1213,6 @@ static irqreturn_t fm10k_msix_mbx_vf(int __always_unused irq, void *data) return IRQ_HANDLED; } -#ifdef CONFIG_NET_POLL_CONTROLLER -/** - * fm10k_netpoll - A Polling 'interrupt' handler - * @netdev: network interface device structure - * - * This is used by netconsole to send skbs without having to re-enable - * interrupts. It's not called while the normal interrupt routine is executing. - **/ -void fm10k_netpoll(struct net_device *netdev) -{ - struct fm10k_intfc *interface = netdev_priv(netdev); - int i; - - /* if interface is down do nothing */ - if (test_bit(__FM10K_DOWN, interface->state)) - return; - - for (i = 0; i < interface->num_q_vectors; i++) - fm10k_msix_clean_rings(0, interface->q_vector[i]); -} - -#endif #define FM10K_ERR_MSG(type) case (type): error = #type; break static void fm10k_handle_fault(struct fm10k_intfc *interface, int type, struct fm10k_fault *fault) @@ -1156,7 +1369,6 @@ static irqreturn_t fm10k_msix_mbx_pf(int __always_unused irq, void *data) struct fm10k_hw *hw = &interface->hw; struct fm10k_mbx_info *mbx = &hw->mbx; u32 eicr; - s32 err = 0; /* unmask any set bits related to this interrupt */ eicr = fm10k_read_reg(hw, FM10K_EICR); @@ -1172,15 +1384,16 @@ static irqreturn_t fm10k_msix_mbx_pf(int __always_unused irq, void *data) /* service mailboxes */ if (fm10k_mbx_trylock(interface)) { - err = mbx->ops.process(hw, mbx); + s32 err = mbx->ops.process(hw, mbx); + + if (err == FM10K_ERR_RESET_REQUESTED) + set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags); + /* handle VFLRE events */ fm10k_iov_event(interface); fm10k_mbx_unlock(interface); } - if (err == FM10K_ERR_RESET_REQUESTED) - set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags); - /* if switch toggled state we should reset GLORTs */ if (eicr & FM10K_EICR_SWITCHNOTREADY) { /* force link down for at least 4 seconds */ @@ -1507,10 +1720,9 @@ int fm10k_mbx_request_irq(struct fm10k_intfc *interface) void fm10k_qv_free_irq(struct fm10k_intfc *interface) { int vector = interface->num_q_vectors; - struct fm10k_hw *hw = &interface->hw; struct msix_entry *entry; - entry = &interface->msix_entries[NON_Q_VECTORS(hw) + vector]; + entry = &interface->msix_entries[NON_Q_VECTORS + vector]; while (vector) { struct fm10k_q_vector *q_vector; @@ -1544,25 +1756,25 @@ int fm10k_qv_request_irq(struct fm10k_intfc *interface) struct net_device *dev = interface->netdev; struct fm10k_hw *hw = &interface->hw; struct msix_entry *entry; - int ri = 0, ti = 0; + unsigned int ri = 0, ti = 0; int vector, err; - entry = &interface->msix_entries[NON_Q_VECTORS(hw)]; + entry = &interface->msix_entries[NON_Q_VECTORS]; for (vector = 0; vector < interface->num_q_vectors; vector++) { struct fm10k_q_vector *q_vector = interface->q_vector[vector]; /* name the vector */ if (q_vector->tx.count && q_vector->rx.count) { - snprintf(q_vector->name, sizeof(q_vector->name) - 1, - "%s-TxRx-%d", dev->name, ri++); + snprintf(q_vector->name, sizeof(q_vector->name), + "%s-TxRx-%u", dev->name, ri++); ti++; } else if (q_vector->rx.count) { - snprintf(q_vector->name, sizeof(q_vector->name) - 1, - "%s-rx-%d", dev->name, ri++); + snprintf(q_vector->name, sizeof(q_vector->name), + "%s-rx-%u", dev->name, ri++); } else if (q_vector->tx.count) { - snprintf(q_vector->name, sizeof(q_vector->name) - 1, - "%s-tx-%d", dev->name, ti++); + snprintf(q_vector->name, sizeof(q_vector->name), + "%s-tx-%u", dev->name, ti++); } else { /* skip this unused q_vector */ continue; @@ -1746,6 +1958,7 @@ skip_tx_dma_drain: /** * fm10k_sw_init - Initialize general software structures * @interface: host interface private structure to initialize + * @ent: PCI device ID entry * * fm10k_sw_init initializes the interface private data structure. * Fields are initialized based on PCI device information and @@ -1800,9 +2013,6 @@ static int fm10k_sw_init(struct fm10k_intfc *interface, netdev->vlan_features |= NETIF_F_HIGHDMA; } - /* delay any future reset requests */ - interface->last_reset = jiffies + (10 * HZ); - /* reset and initialize the hardware so it is in a known state */ err = hw->mac.ops.reset_hw(hw); if (err) { @@ -1834,7 +2044,7 @@ static int fm10k_sw_init(struct fm10k_intfc *interface, netdev->addr_assign_type |= NET_ADDR_RANDOM; } - ether_addr_copy(netdev->dev_addr, hw->mac.addr); + eth_hw_addr_set(netdev, hw->mac.addr); ether_addr_copy(netdev->perm_addr, hw->mac.addr); if (!is_valid_ether_addr(netdev->perm_addr)) { @@ -1853,13 +2063,16 @@ static int fm10k_sw_init(struct fm10k_intfc *interface, interface->tx_itr = FM10K_TX_ITR_DEFAULT; interface->rx_itr = FM10K_ITR_ADAPTIVE | FM10K_RX_ITR_DEFAULT; - /* initialize udp port lists */ - INIT_LIST_HEAD(&interface->vxlan_port); - INIT_LIST_HEAD(&interface->geneve_port); + /* Initialize the MAC/VLAN queue */ + INIT_LIST_HEAD(&interface->macvlan_requests); netdev_rss_key_fill(rss_key, sizeof(rss_key)); memcpy(interface->rssrk, rss_key, sizeof(rss_key)); + /* Initialize the mailbox lock */ + spin_lock_init(&interface->mbx_lock); + spin_lock_init(&interface->macvlan_lock); + /* Start off interface as being down */ set_bit(__FM10K_DOWN, interface->state); set_bit(__FM10K_UPDATING_STATS, interface->state); @@ -1867,91 +2080,6 @@ static int fm10k_sw_init(struct fm10k_intfc *interface, return 0; } -static void fm10k_slot_warn(struct fm10k_intfc *interface) -{ - enum pcie_link_width width = PCIE_LNK_WIDTH_UNKNOWN; - enum pci_bus_speed speed = PCI_SPEED_UNKNOWN; - struct fm10k_hw *hw = &interface->hw; - int max_gts = 0, expected_gts = 0; - - if (pcie_get_minimum_link(interface->pdev, &speed, &width) || - speed == PCI_SPEED_UNKNOWN || width == PCIE_LNK_WIDTH_UNKNOWN) { - dev_warn(&interface->pdev->dev, - "Unable to determine PCI Express bandwidth.\n"); - return; - } - - switch (speed) { - case PCIE_SPEED_2_5GT: - /* 8b/10b encoding reduces max throughput by 20% */ - max_gts = 2 * width; - break; - case PCIE_SPEED_5_0GT: - /* 8b/10b encoding reduces max throughput by 20% */ - max_gts = 4 * width; - break; - case PCIE_SPEED_8_0GT: - /* 128b/130b encoding has less than 2% impact on throughput */ - max_gts = 8 * width; - break; - default: - dev_warn(&interface->pdev->dev, - "Unable to determine PCI Express bandwidth.\n"); - return; - } - - dev_info(&interface->pdev->dev, - "PCI Express bandwidth of %dGT/s available\n", - max_gts); - dev_info(&interface->pdev->dev, - "(Speed:%s, Width: x%d, Encoding Loss:%s, Payload:%s)\n", - (speed == PCIE_SPEED_8_0GT ? "8.0GT/s" : - speed == PCIE_SPEED_5_0GT ? "5.0GT/s" : - speed == PCIE_SPEED_2_5GT ? "2.5GT/s" : - "Unknown"), - hw->bus.width, - (speed == PCIE_SPEED_2_5GT ? "20%" : - speed == PCIE_SPEED_5_0GT ? "20%" : - speed == PCIE_SPEED_8_0GT ? "<2%" : - "Unknown"), - (hw->bus.payload == fm10k_bus_payload_128 ? "128B" : - hw->bus.payload == fm10k_bus_payload_256 ? "256B" : - hw->bus.payload == fm10k_bus_payload_512 ? "512B" : - "Unknown")); - - switch (hw->bus_caps.speed) { - case fm10k_bus_speed_2500: - /* 8b/10b encoding reduces max throughput by 20% */ - expected_gts = 2 * hw->bus_caps.width; - break; - case fm10k_bus_speed_5000: - /* 8b/10b encoding reduces max throughput by 20% */ - expected_gts = 4 * hw->bus_caps.width; - break; - case fm10k_bus_speed_8000: - /* 128b/130b encoding has less than 2% impact on throughput */ - expected_gts = 8 * hw->bus_caps.width; - break; - default: - dev_warn(&interface->pdev->dev, - "Unable to determine expected PCI Express bandwidth.\n"); - return; - } - - if (max_gts >= expected_gts) - return; - - dev_warn(&interface->pdev->dev, - "This device requires %dGT/s of bandwidth for optimal performance.\n", - expected_gts); - dev_warn(&interface->pdev->dev, - "A %sslot with x%d lanes is suggested.\n", - (hw->bus_caps.speed == fm10k_bus_speed_2500 ? "2.5GT/s " : - hw->bus_caps.speed == fm10k_bus_speed_5000 ? "5.0GT/s " : - hw->bus_caps.speed == fm10k_bus_speed_8000 ? "8.0GT/s " : ""), - hw->bus_caps.width); -} - /** * fm10k_probe - Device Initialization Routine * @pdev: PCI device information struct @@ -1998,8 +2126,6 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_pci_reg; } - pci_enable_pcie_error_reporting(pdev); - pci_set_master(pdev); pci_save_state(pdev); @@ -2063,15 +2189,17 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* Initialize service timer and service task late in order to avoid * cleanup issues. */ - setup_timer(&interface->service_timer, &fm10k_service_timer, - (unsigned long)interface); + timer_setup(&interface->service_timer, fm10k_service_timer, 0); INIT_WORK(&interface->service_task, fm10k_service_task); + /* Setup the MAC/VLAN queue */ + INIT_DELAYED_WORK(&interface->macvlan_task, fm10k_macvlan_task); + /* kick off service timer now, even when interface is down */ mod_timer(&interface->service_timer, (HZ * 2) + jiffies); /* print warning for non-optimal configurations */ - fm10k_slot_warn(interface); + pcie_print_link_status(interface->pdev); /* report MAC address for logging */ dev_info(&pdev->dev, "%pM\n", netdev->dev_addr); @@ -2079,8 +2207,9 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* enable SR-IOV after registering netdev to enforce PF/VF ordering */ fm10k_iov_configure(pdev, 0); - /* clear the service task disable bit to allow service task to start */ + /* clear the service task disable bit and kick off service task */ clear_bit(__FM10K_SERVICE_DISABLE, interface->state); + fm10k_service_event_schedule(interface); return 0; @@ -2116,10 +2245,13 @@ static void fm10k_remove(struct pci_dev *pdev) struct fm10k_intfc *interface = pci_get_drvdata(pdev); struct net_device *netdev = interface->netdev; - del_timer_sync(&interface->service_timer); + timer_delete_sync(&interface->service_timer); - set_bit(__FM10K_SERVICE_DISABLE, interface->state); - cancel_work_sync(&interface->service_task); + fm10k_stop_service_event(interface); + fm10k_stop_macvlan_task(interface); + + /* Remove all pending MAC/VLAN requests */ + fm10k_clear_macvlan_queue(interface, interface->glort, true); /* free netdev, this may bounce the interrupts due to setup_tc */ if (netdev->reg_state == NETREG_REGISTERED) @@ -2145,8 +2277,6 @@ static void fm10k_remove(struct pci_dev *pdev) pci_release_mem_regions(pdev); - pci_disable_pcie_error_reporting(pdev); - pci_disable_device(pdev); } @@ -2156,11 +2286,14 @@ static void fm10k_prepare_suspend(struct fm10k_intfc *interface) * a surprise remove if the PCIe device is disabled while we're * stopped. We stop the watchdog task until after we resume software * activity. + * + * Note that the MAC/VLAN task will be stopped as part of preparing + * for reset so we don't need to handle it here. */ - set_bit(__FM10K_SERVICE_DISABLE, interface->state); - cancel_work_sync(&interface->service_task); + fm10k_stop_service_event(interface); - fm10k_prepare_for_reset(interface); + if (fm10k_prepare_for_reset(interface)) + set_bit(__FM10K_RESET_SUSPENDED, interface->state); } static int fm10k_handle_resume(struct fm10k_intfc *interface) @@ -2168,6 +2301,13 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface) struct fm10k_hw *hw = &interface->hw; int err; + /* Even if we didn't properly prepare for reset in + * fm10k_prepare_suspend, we'll attempt to resume anyways. + */ + if (!test_and_clear_bit(__FM10K_RESET_SUSPENDED, interface->state)) + dev_warn(&interface->pdev->dev, + "Device was shut down as part of suspend... Attempting to recover\n"); + /* reset statistics starting values */ hw->mac.ops.rebind_hw_stats(hw, &interface->stats); @@ -2185,45 +2325,29 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface) interface->link_down_event = jiffies + (HZ); set_bit(__FM10K_LINK_DOWN, interface->state); - /* clear the service task disable bit to allow service task to start */ - clear_bit(__FM10K_SERVICE_DISABLE, interface->state); - fm10k_service_event_schedule(interface); + /* restart the service task */ + fm10k_start_service_event(interface); - return err; + /* Restart the MAC/VLAN request queue in-case of outstanding events */ + fm10k_macvlan_schedule(interface); + + return 0; } -#ifdef CONFIG_PM /** - * fm10k_resume - Restore device to pre-sleep state - * @pdev: PCI device information struct + * fm10k_resume - Generic PM resume hook + * @dev: generic device structure * - * fm10k_resume is called after the system has powered back up from a sleep - * state and is ready to resume operation. This function is meant to restore - * the device back to its pre-sleep state. + * Generic PM hook used when waking the device from a low power state after + * suspend or hibernation. This function does not need to handle lower PCIe + * device state as the stack takes care of that for us. **/ -static int fm10k_resume(struct pci_dev *pdev) +static int fm10k_resume(struct device *dev) { - struct fm10k_intfc *interface = pci_get_drvdata(pdev); + struct fm10k_intfc *interface = dev_get_drvdata(dev); struct net_device *netdev = interface->netdev; struct fm10k_hw *hw = &interface->hw; - u32 err; - - pci_set_power_state(pdev, PCI_D0); - pci_restore_state(pdev); - - /* pci_restore_state clears dev->state_saved so call - * pci_save_state to restore it. - */ - pci_save_state(pdev); - - err = pci_enable_device_mem(pdev); - if (err) { - dev_err(&pdev->dev, "Cannot enable PCI device from suspend\n"); - return err; - } - pci_set_master(pdev); - - pci_wake_from_d3(pdev, false); + int err; /* refresh hw_addr in case it was dropped */ hw->hw_addr = interface->uc_addr; @@ -2238,36 +2362,25 @@ static int fm10k_resume(struct pci_dev *pdev) } /** - * fm10k_suspend - Prepare the device for a system sleep state - * @pdev: PCI device information struct + * fm10k_suspend - Generic PM suspend hook + * @dev: generic device structure * - * fm10k_suspend is meant to shutdown the device prior to the system entering - * a sleep state. The fm10k hardware does not support wake on lan so the - * driver simply needs to shut down the device so it is in a low power state. + * Generic PM hook used when setting the device into a low power state for + * system suspend or hibernation. This function does not need to handle lower + * PCIe device state as the stack takes care of that for us. **/ -static int fm10k_suspend(struct pci_dev *pdev, - pm_message_t __always_unused state) +static int fm10k_suspend(struct device *dev) { - struct fm10k_intfc *interface = pci_get_drvdata(pdev); + struct fm10k_intfc *interface = dev_get_drvdata(dev); struct net_device *netdev = interface->netdev; - int err = 0; netif_device_detach(netdev); fm10k_prepare_suspend(interface); - err = pci_save_state(pdev); - if (err) - return err; - - pci_disable_device(pdev); - pci_wake_from_d3(pdev, false); - pci_set_power_state(pdev, PCI_D3hot); - return 0; } -#endif /* CONFIG_PM */ /** * fm10k_io_error_detected - called when PCI error is detected * @pdev: Pointer to PCI device @@ -2310,19 +2423,11 @@ static pci_ers_result_t fm10k_io_slot_reset(struct pci_dev *pdev) } else { pci_set_master(pdev); pci_restore_state(pdev); - - /* After second error pci->state_saved is false, this - * resets it so EEH doesn't break. - */ - pci_save_state(pdev); - pci_wake_from_d3(pdev, false); result = PCI_ERS_RESULT_RECOVERED; } - pci_cleanup_aer_uncorrect_error_status(pdev); - return result; } @@ -2343,11 +2448,18 @@ static void fm10k_io_resume(struct pci_dev *pdev) if (err) dev_warn(&pdev->dev, - "fm10k_io_resume failed: %d\n", err); + "%s failed: %d\n", __func__, err); else netif_device_attach(netdev); } +/** + * fm10k_io_reset_prepare - called when PCI function is about to be reset + * @pdev: Pointer to PCI device + * + * This callback is called when the PCI function is about to be reset, + * allowing the device driver to prepare for it. + */ static void fm10k_io_reset_prepare(struct pci_dev *pdev) { /* warn incase we have any active VF devices */ @@ -2357,6 +2469,13 @@ static void fm10k_io_reset_prepare(struct pci_dev *pdev) fm10k_prepare_suspend(pci_get_drvdata(pdev)); } +/** + * fm10k_io_reset_done - called when PCI function has finished resetting + * @pdev: Pointer to PCI device + * + * This callback is called just after the PCI function is reset, such as via + * /sys/class/net/<enpX>/device/reset or similar. + */ static void fm10k_io_reset_done(struct pci_dev *pdev) { struct fm10k_intfc *interface = pci_get_drvdata(pdev); @@ -2364,7 +2483,7 @@ static void fm10k_io_reset_done(struct pci_dev *pdev) if (err) { dev_warn(&pdev->dev, - "fm10k_io_reset_notify failed: %d\n", err); + "%s failed: %d\n", __func__, err); netif_device_detach(interface->netdev); } } @@ -2377,15 +2496,14 @@ static const struct pci_error_handlers fm10k_err_handler = { .reset_done = fm10k_io_reset_done, }; +static DEFINE_SIMPLE_DEV_PM_OPS(fm10k_pm_ops, fm10k_suspend, fm10k_resume); + static struct pci_driver fm10k_driver = { .name = fm10k_driver_name, .id_table = fm10k_pci_tbl, .probe = fm10k_probe, .remove = fm10k_remove, -#ifdef CONFIG_PM - .suspend = fm10k_suspend, - .resume = fm10k_resume, -#endif + .driver.pm = pm_sleep_ptr(&fm10k_pm_ops), .sriov_configure = fm10k_iov_configure, .err_handler = &fm10k_err_handler }; |