1 files changed, 156 insertions, 92 deletions

diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index a483eb185c99..9610ca770349 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -167,6 +167,45 @@ static struct iavf_adapter *iavf_pdev_to_adapter(struct pci_dev *pdev)
 }
 
 /**
+ * iavf_is_reset_in_progress - Check if a reset is in progress
+ * @adapter: board private structure
+ */
+static bool iavf_is_reset_in_progress(struct iavf_adapter *adapter)
+{
+	if (adapter->state == __IAVF_RESETTING ||
+	    adapter->flags & (IAVF_FLAG_RESET_PENDING |
+			      IAVF_FLAG_RESET_NEEDED))
+		return true;
+
+	return false;
+}
+
+/**
+ * iavf_wait_for_reset - Wait for reset to finish.
+ * @adapter: board private structure
+ *
+ * Returns 0 if reset finished successfully, negative on timeout or interrupt.
+ */
+int iavf_wait_for_reset(struct iavf_adapter *adapter)
+{
+	int ret = wait_event_interruptible_timeout(adapter->reset_waitqueue,
+					!iavf_is_reset_in_progress(adapter),
+					msecs_to_jiffies(5000));
+
+	/* If ret < 0 then it means wait was interrupted.
+	 * If ret == 0 then it means we got a timeout while waiting
+	 * for reset to finish.
+	 * If ret > 0 it means reset has finished.
+	 */
+	if (ret > 0)
+		return 0;
+	else if (ret < 0)
+		return -EINTR;
+	else
+		return -EBUSY;
+}
+
+/**
  * iavf_allocate_dma_mem_d - OS specific memory alloc for shared code
  * @hw:   pointer to the HW structure
  * @mem:  ptr to mem struct to fill out
@@ -262,12 +301,14 @@ static int iavf_lock_timeout(struct mutex *lock, unsigned int msecs)
 /**
  * iavf_schedule_reset - Set the flags and schedule a reset event
  * @adapter: board private structure
+ * @flags: IAVF_FLAG_RESET_PENDING or IAVF_FLAG_RESET_NEEDED
  **/
-void iavf_schedule_reset(struct iavf_adapter *adapter)
+void iavf_schedule_reset(struct iavf_adapter *adapter, u64 flags)
 {
-	if (!(adapter->flags &
-	      (IAVF_FLAG_RESET_PENDING | IAVF_FLAG_RESET_NEEDED))) {
-		adapter->flags |= IAVF_FLAG_RESET_NEEDED;
+	if (!test_bit(__IAVF_IN_REMOVE_TASK, &adapter->crit_section) &&
+	    !(adapter->flags &
+	    (IAVF_FLAG_RESET_PENDING | IAVF_FLAG_RESET_NEEDED))) {
+		adapter->flags |= flags;
 		queue_work(adapter->wq, &adapter->reset_task);
 	}
 }
@@ -295,7 +336,7 @@ static void iavf_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 	struct iavf_adapter *adapter = netdev_priv(netdev);
 
 	adapter->tx_timeout_count++;
-	iavf_schedule_reset(adapter);
+	iavf_schedule_reset(adapter, IAVF_FLAG_RESET_NEEDED);
 }
 
 /**
@@ -1651,10 +1692,10 @@ static int iavf_set_interrupt_capability(struct iavf_adapter *adapter)
 		adapter->msix_entries[vector].entry = vector;
 
 	err = iavf_acquire_msix_vectors(adapter, v_budget);
+	if (!err)
+		iavf_schedule_finish_config(adapter);
 
 out:
-	netif_set_real_num_rx_queues(adapter->netdev, pairs);
-	netif_set_real_num_tx_queues(adapter->netdev, pairs);
 	return err;
 }
 
@@ -1828,19 +1869,16 @@ static int iavf_alloc_q_vectors(struct iavf_adapter *adapter)
 static void iavf_free_q_vectors(struct iavf_adapter *adapter)
 {
 	int q_idx, num_q_vectors;
-	int napi_vectors;
 
 	if (!adapter->q_vectors)
 		return;
 
 	num_q_vectors = adapter->num_msix_vectors - NONQ_VECS;
-	napi_vectors = adapter->num_active_queues;
 
 	for (q_idx = 0; q_idx < num_q_vectors; q_idx++) {
 		struct iavf_q_vector *q_vector = &adapter->q_vectors[q_idx];
 
-		if (q_idx < napi_vectors)
-			netif_napi_del(&q_vector->napi);
+		netif_napi_del(&q_vector->napi);
 	}
 	kfree(adapter->q_vectors);
 	adapter->q_vectors = NULL;
@@ -1877,9 +1915,7 @@ static int iavf_init_interrupt_scheme(struct iavf_adapter *adapter)
 		goto err_alloc_queues;
 	}
 
-	rtnl_lock();
 	err = iavf_set_interrupt_capability(adapter);
-	rtnl_unlock();
 	if (err) {
 		dev_err(&adapter->pdev->dev,
 			"Unable to setup interrupt capabilities\n");
@@ -1932,15 +1968,16 @@ static void iavf_free_rss(struct iavf_adapter *adapter)
 /**
  * iavf_reinit_interrupt_scheme - Reallocate queues and vectors
  * @adapter: board private structure
+ * @running: true if adapter->state == __IAVF_RUNNING
  *
  * Returns 0 on success, negative on failure
  **/
-static int iavf_reinit_interrupt_scheme(struct iavf_adapter *adapter)
+static int iavf_reinit_interrupt_scheme(struct iavf_adapter *adapter, bool running)
 {
 	struct net_device *netdev = adapter->netdev;
 	int err;
 
-	if (netif_running(netdev))
+	if (running)
 		iavf_free_traffic_irqs(adapter);
 	iavf_free_misc_irq(adapter);
 	iavf_reset_interrupt_capability(adapter);
@@ -1965,6 +2002,78 @@ err:
 }
 
 /**
+ * iavf_finish_config - do all netdev work that needs RTNL
+ * @work: our work_struct
+ *
+ * Do work that needs both RTNL and crit_lock.
+ **/
+static void iavf_finish_config(struct work_struct *work)
+{
+	struct iavf_adapter *adapter;
+	int pairs, err;
+
+	adapter = container_of(work, struct iavf_adapter, finish_config);
+
+	/* Always take RTNL first to prevent circular lock dependency */
+	rtnl_lock();
+	mutex_lock(&adapter->crit_lock);
+
+	if ((adapter->flags & IAVF_FLAG_SETUP_NETDEV_FEATURES) &&
+	    adapter->netdev_registered &&
+	    !test_bit(__IAVF_IN_REMOVE_TASK, &adapter->crit_section)) {
+		netdev_update_features(adapter->netdev);
+		adapter->flags &= ~IAVF_FLAG_SETUP_NETDEV_FEATURES;
+	}
+
+	switch (adapter->state) {
+	case __IAVF_DOWN:
+		if (!adapter->netdev_registered) {
+			err = register_netdevice(adapter->netdev);
+			if (err) {
+				dev_err(&adapter->pdev->dev, "Unable to register netdev (%d)\n",
+					err);
+
+				/* go back and try again.*/
+				iavf_free_rss(adapter);
+				iavf_free_misc_irq(adapter);
+				iavf_reset_interrupt_capability(adapter);
+				iavf_change_state(adapter,
+						  __IAVF_INIT_CONFIG_ADAPTER);
+				goto out;
+			}
+			adapter->netdev_registered = true;
+		}
+
+		/* Set the real number of queues when reset occurs while
+		 * state == __IAVF_DOWN
+		 */
+		fallthrough;
+	case __IAVF_RUNNING:
+		pairs = adapter->num_active_queues;
+		netif_set_real_num_rx_queues(adapter->netdev, pairs);
+		netif_set_real_num_tx_queues(adapter->netdev, pairs);
+		break;
+
+	default:
+		break;
+	}
+
+out:
+	mutex_unlock(&adapter->crit_lock);
+	rtnl_unlock();
+}
+
+/**
+ * iavf_schedule_finish_config - Set the flags and schedule a reset event
+ * @adapter: board private structure
+ **/
+void iavf_schedule_finish_config(struct iavf_adapter *adapter)
+{
+	if (!test_bit(__IAVF_IN_REMOVE_TASK, &adapter->crit_section))
+		queue_work(adapter->wq, &adapter->finish_config);
+}
+
+/**
  * iavf_process_aq_command - process aq_required flags
  * and sends aq command
  * @adapter: pointer to iavf adapter structure
@@ -2371,7 +2480,7 @@ int iavf_parse_vf_resource_msg(struct iavf_adapter *adapter)
 			adapter->vsi_res->num_queue_pairs);
 		adapter->flags |= IAVF_FLAG_REINIT_MSIX_NEEDED;
 		adapter->num_req_queues = adapter->vsi_res->num_queue_pairs;
-		iavf_schedule_reset(adapter);
+		iavf_schedule_reset(adapter, IAVF_FLAG_RESET_NEEDED);
 
 		return -EAGAIN;
 	}
@@ -2601,22 +2710,8 @@ static void iavf_init_config_adapter(struct iavf_adapter *adapter)
 
 	netif_carrier_off(netdev);
 	adapter->link_up = false;
-
-	/* set the semaphore to prevent any callbacks after device registration
-	 * up to time when state of driver will be set to __IAVF_DOWN
-	 */
-	rtnl_lock();
-	if (!adapter->netdev_registered) {
-		err = register_netdevice(netdev);
-		if (err) {
-			rtnl_unlock();
-			goto err_register;
-		}
-	}
-
-	adapter->netdev_registered = true;
-
 	netif_tx_stop_all_queues(netdev);
+
 	if (CLIENT_ALLOWED(adapter)) {
 		err = iavf_lan_add_device(adapter);
 		if (err)
@@ -2629,7 +2724,6 @@ static void iavf_init_config_adapter(struct iavf_adapter *adapter)
 
 	iavf_change_state(adapter, __IAVF_DOWN);
 	set_bit(__IAVF_VSI_DOWN, adapter->vsi.state);
-	rtnl_unlock();
 
 	iavf_misc_irq_enable(adapter);
 	wake_up(&adapter->down_waitqueue);
@@ -2649,10 +2743,11 @@ static void iavf_init_config_adapter(struct iavf_adapter *adapter)
 		/* request initial VLAN offload settings */
 		iavf_set_vlan_offload_features(adapter, 0, netdev->features);
 
+	iavf_schedule_finish_config(adapter);
 	return;
+
 err_mem:
 	iavf_free_rss(adapter);
-err_register:
 	iavf_free_misc_irq(adapter);
 err_sw_init:
 	iavf_reset_interrupt_capability(adapter);
@@ -2679,26 +2774,9 @@ static void iavf_watchdog_task(struct work_struct *work)
 		goto restart_watchdog;
 	}
 
-	if ((adapter->flags & IAVF_FLAG_SETUP_NETDEV_FEATURES) &&
-	    adapter->netdev_registered &&
-	    !test_bit(__IAVF_IN_REMOVE_TASK, &adapter->crit_section) &&
-	    rtnl_trylock()) {
-		netdev_update_features(adapter->netdev);
-		rtnl_unlock();
-		adapter->flags &= ~IAVF_FLAG_SETUP_NETDEV_FEATURES;
-	}
-
 	if (adapter->flags & IAVF_FLAG_PF_COMMS_FAILED)
 		iavf_change_state(adapter, __IAVF_COMM_FAILED);
 
-	if (adapter->flags & IAVF_FLAG_RESET_NEEDED) {
-		adapter->aq_required = 0;
-		adapter->current_op = VIRTCHNL_OP_UNKNOWN;
-		mutex_unlock(&adapter->crit_lock);
-		queue_work(adapter->wq, &adapter->reset_task);
-		return;
-	}
-
 	switch (adapter->state) {
 	case __IAVF_STARTUP:
 		iavf_startup(adapter);
@@ -2826,11 +2904,10 @@ static void iavf_watchdog_task(struct work_struct *work)
 	/* check for hw reset */
 	reg_val = rd32(hw, IAVF_VF_ARQLEN1) & IAVF_VF_ARQLEN1_ARQENABLE_MASK;
 	if (!reg_val) {
-		adapter->flags |= IAVF_FLAG_RESET_PENDING;
 		adapter->aq_required = 0;
 		adapter->current_op = VIRTCHNL_OP_UNKNOWN;
 		dev_err(&adapter->pdev->dev, "Hardware reset detected\n");
-		queue_work(adapter->wq, &adapter->reset_task);
+		iavf_schedule_reset(adapter, IAVF_FLAG_RESET_PENDING);
 		mutex_unlock(&adapter->crit_lock);
 		queue_delayed_work(adapter->wq,
 				   &adapter->watchdog_task, HZ * 2);
@@ -2940,11 +3017,6 @@ static void iavf_reset_task(struct work_struct *work)
 	int i = 0, err;
 	bool running;
 
-	/* Detach interface to avoid subsequent NDO callbacks */
-	rtnl_lock();
-	netif_device_detach(netdev);
-	rtnl_unlock();
-
 	/* When device is being removed it doesn't make sense to run the reset
 	 * task, just return in such a case.
 	 */
@@ -2952,7 +3024,7 @@ static void iavf_reset_task(struct work_struct *work)
 		if (adapter->state != __IAVF_REMOVE)
 			queue_work(adapter->wq, &adapter->reset_task);
 
-		goto reset_finish;
+		return;
 	}
 
 	while (!mutex_trylock(&adapter->client_lock))
@@ -3010,11 +3082,6 @@ static void iavf_reset_task(struct work_struct *work)
 		iavf_disable_vf(adapter);
 		mutex_unlock(&adapter->client_lock);
 		mutex_unlock(&adapter->crit_lock);
-		if (netif_running(netdev)) {
-			rtnl_lock();
-			dev_close(netdev);
-			rtnl_unlock();
-		}
 		return; /* Do not attempt to reinit. It's dead, Jim. */
 	}
 
@@ -3056,7 +3123,7 @@ continue_reset:
 
 	if ((adapter->flags & IAVF_FLAG_REINIT_MSIX_NEEDED) ||
 	    (adapter->flags & IAVF_FLAG_REINIT_ITR_NEEDED)) {
-		err = iavf_reinit_interrupt_scheme(adapter);
+		err = iavf_reinit_interrupt_scheme(adapter, running);
 		if (err)
 			goto reset_err;
 	}
@@ -3151,10 +3218,11 @@ continue_reset:
 
 	adapter->flags &= ~IAVF_FLAG_REINIT_ITR_NEEDED;
 
+	wake_up(&adapter->reset_waitqueue);
 	mutex_unlock(&adapter->client_lock);
 	mutex_unlock(&adapter->crit_lock);
 
-	goto reset_finish;
+	return;
 reset_err:
 	if (running) {
 		set_bit(__IAVF_VSI_DOWN, adapter->vsi.state);
@@ -3164,21 +3232,7 @@ reset_err:
 
 	mutex_unlock(&adapter->client_lock);
 	mutex_unlock(&adapter->crit_lock);
-
-	if (netif_running(netdev)) {
-		/* Close device to ensure that Tx queues will not be started
-		 * during netif_device_attach() at the end of the reset task.
-		 */
-		rtnl_lock();
-		dev_close(netdev);
-		rtnl_unlock();
-	}
-
 	dev_err(&adapter->pdev->dev, "failed to allocate resources during reinit\n");
-reset_finish:
-	rtnl_lock();
-	netif_device_attach(netdev);
-	rtnl_unlock();
 }
 
 /**
@@ -3196,9 +3250,6 @@ static void iavf_adminq_task(struct work_struct *work)
 	u32 val, oldval;
 	u16 pending;
 
-	if (adapter->flags & IAVF_FLAG_PF_COMMS_FAILED)
-		goto out;
-
 	if (!mutex_trylock(&adapter->crit_lock)) {
 		if (adapter->state == __IAVF_REMOVE)
 			return;
@@ -3207,10 +3258,13 @@ static void iavf_adminq_task(struct work_struct *work)
 		goto out;
 	}
 
+	if (adapter->flags & IAVF_FLAG_PF_COMMS_FAILED)
+		goto unlock;
+
 	event.buf_len = IAVF_MAX_AQ_BUF_SIZE;
 	event.msg_buf = kzalloc(event.buf_len, GFP_KERNEL);
 	if (!event.msg_buf)
-		goto out;
+		goto unlock;
 
 	do {
 		ret = iavf_clean_arq_element(hw, &event, &pending);
@@ -3225,11 +3279,8 @@ static void iavf_adminq_task(struct work_struct *work)
 		if (pending != 0)
 			memset(event.msg_buf, 0, IAVF_MAX_AQ_BUF_SIZE);
 	} while (pending);
-	mutex_unlock(&adapter->crit_lock);
 
-	if ((adapter->flags &
-	     (IAVF_FLAG_RESET_PENDING | IAVF_FLAG_RESET_NEEDED)) ||
-	    adapter->state == __IAVF_RESETTING)
+	if (iavf_is_reset_in_progress(adapter))
 		goto freedom;
 
 	/* check for error indications */
@@ -3271,6 +3322,8 @@ static void iavf_adminq_task(struct work_struct *work)
 
 freedom:
 	kfree(event.msg_buf);
+unlock:
+	mutex_unlock(&adapter->crit_lock);
 out:
 	/* re-enable Admin queue interrupt cause */
 	iavf_misc_irq_enable(adapter);
@@ -4315,6 +4368,7 @@ static int iavf_close(struct net_device *netdev)
 static int iavf_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
+	int ret = 0;
 
 	netdev_dbg(netdev, "changing MTU from %d to %d\n",
 		   netdev->mtu, new_mtu);
@@ -4325,11 +4379,15 @@ static int iavf_change_mtu(struct net_device *netdev, int new_mtu)
 	}
 
 	if (netif_running(netdev)) {
-		adapter->flags |= IAVF_FLAG_RESET_NEEDED;
-		queue_work(adapter->wq, &adapter->reset_task);
+		iavf_schedule_reset(adapter, IAVF_FLAG_RESET_NEEDED);
+		ret = iavf_wait_for_reset(adapter);
+		if (ret < 0)
+			netdev_warn(netdev, "MTU change interrupted waiting for reset");
+		else if (ret)
+			netdev_warn(netdev, "MTU change timed out waiting for reset");
 	}
 
-	return 0;
+	return ret;
 }
 
 #define NETIF_VLAN_OFFLOAD_FEATURES	(NETIF_F_HW_VLAN_CTAG_RX | \
@@ -4922,6 +4980,7 @@ static int iavf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	INIT_WORK(&adapter->reset_task, iavf_reset_task);
 	INIT_WORK(&adapter->adminq_task, iavf_adminq_task);
+	INIT_WORK(&adapter->finish_config, iavf_finish_config);
 	INIT_DELAYED_WORK(&adapter->watchdog_task, iavf_watchdog_task);
 	INIT_DELAYED_WORK(&adapter->client_task, iavf_client_task);
 	queue_delayed_work(adapter->wq, &adapter->watchdog_task,
@@ -4930,6 +4989,9 @@ static int iavf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* Setup the wait queue for indicating transition to down status */
 	init_waitqueue_head(&adapter->down_waitqueue);
 
+	/* Setup the wait queue for indicating transition to running state */
+	init_waitqueue_head(&adapter->reset_waitqueue);
+
 	/* Setup the wait queue for indicating virtchannel events */
 	init_waitqueue_head(&adapter->vc_waitqueue);
 
@@ -5061,13 +5123,15 @@ static void iavf_remove(struct pci_dev *pdev)
 		usleep_range(500, 1000);
 	}
 	cancel_delayed_work_sync(&adapter->watchdog_task);
+	cancel_work_sync(&adapter->finish_config);
 
+	rtnl_lock();
 	if (adapter->netdev_registered) {
-		rtnl_lock();
 		unregister_netdevice(netdev);
 		adapter->netdev_registered = false;
-		rtnl_unlock();
 	}
+	rtnl_unlock();
+
 	if (CLIENT_ALLOWED(adapter)) {
 		err = iavf_lan_del_device(adapter);
 		if (err)