i40e/i40evf: Detect and recover hung queue scenario

In VFs, there is a known issue which can cause writebacks
to not occur when interrupts are disabled and there are
less than 4 descriptors resulting in TX timeout. Timeout
can also occur due to lost interrupt.

The current implementation for detecting and recovering
from hung queues in the PF is problematic because it actually
actively encourages lost interrupts.  By triggering a SW
interrupt, interrupts are forced on.  If we are already in
napi_poll and an interrupt fires, napi_poll will not be
rescheduled and the interrupt is effectively lost; thereby
potentially *causing* hung queues.

This patch checks whether packets are being processed between
every watchdog cycle and determine potential hung queue and
fires triggers SW interrupt only for that particular queue.

Signed-off-by: Sudheer Mogilappagari <sudheer.mogilappagari@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

1 files changed, 1 insertions, 99 deletions

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index db7dd812c1df..69d15560fdf9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -4877,104 +4877,6 @@ static int i40e_pf_wait_queues_disabled(struct i40e_pf *pf)
 #endif
 
 /**
- * i40e_detect_recover_hung_queue - Function to detect and recover hung_queue
- * @q_idx: TX queue number
- * @vsi: Pointer to VSI struct
- *
- * This function checks specified queue for given VSI. Detects hung condition.
- * We proactively detect hung TX queues by checking if interrupts are disabled
- * but there are pending descriptors.  If it appears hung, attempt to recover
- * by triggering a SW interrupt.
- **/
-static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
-{
-	struct i40e_ring *tx_ring = NULL;
-	struct i40e_pf	*pf;
-	u32 val, tx_pending;
-	int i;
-
-	pf = vsi->back;
-
-	/* now that we have an index, find the tx_ring struct */
-	for (i = 0; i < vsi->num_queue_pairs; i++) {
-		if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) {
-			if (q_idx == vsi->tx_rings[i]->queue_index) {
-				tx_ring = vsi->tx_rings[i];
-				break;
-			}
-		}
-	}
-
-	if (!tx_ring)
-		return;
-
-	/* Read interrupt register */
-	if (pf->flags & I40E_FLAG_MSIX_ENABLED)
-		val = rd32(&pf->hw,
-			   I40E_PFINT_DYN_CTLN(tx_ring->q_vector->v_idx +
-					       tx_ring->vsi->base_vector - 1));
-	else
-		val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0);
-
-	tx_pending = i40e_get_tx_pending(tx_ring);
-
-	/* Interrupts are disabled and TX pending is non-zero,
-	 * trigger the SW interrupt (don't wait). Worst case
-	 * there will be one extra interrupt which may result
-	 * into not cleaning any queues because queues are cleaned.
-	 */
-	if (tx_pending && (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK)))
-		i40e_force_wb(vsi, tx_ring->q_vector);
-}
-
-/**
- * i40e_detect_recover_hung - Function to detect and recover hung_queues
- * @pf:  pointer to PF struct
- *
- * LAN VSI has netdev and netdev has TX queues. This function is to check
- * each of those TX queues if they are hung, trigger recovery by issuing
- * SW interrupt.
- **/
-static void i40e_detect_recover_hung(struct i40e_pf *pf)
-{
-	struct net_device *netdev;
-	struct i40e_vsi *vsi;
-	unsigned int i;
-
-	/* Only for LAN VSI */
-	vsi = pf->vsi[pf->lan_vsi];
-
-	if (!vsi)
-		return;
-
-	/* Make sure, VSI state is not DOWN/RECOVERY_PENDING */
-	if (test_bit(__I40E_VSI_DOWN, vsi->back->state) ||
-	    test_bit(__I40E_RESET_RECOVERY_PENDING, vsi->back->state))
-		return;
-
-	/* Make sure type is MAIN VSI */
-	if (vsi->type != I40E_VSI_MAIN)
-		return;
-
-	netdev = vsi->netdev;
-	if (!netdev)
-		return;
-
-	/* Bail out if netif_carrier is not OK */
-	if (!netif_carrier_ok(netdev))
-		return;
-
-	/* Go thru' TX queues for netdev */
-	for (i = 0; i < netdev->num_tx_queues; i++) {
-		struct netdev_queue *q;
-
-		q = netdev_get_tx_queue(netdev, i);
-		if (q)
-			i40e_detect_recover_hung_queue(i, vsi);
-	}
-}
-
-/**
  * i40e_get_iscsi_tc_map - Return TC map for iSCSI APP
  * @pf: pointer to PF
  *
@@ -9695,7 +9597,7 @@ static void i40e_service_task(struct work_struct *work)
 	if (test_and_set_bit(__I40E_SERVICE_SCHED, pf->state))
 		return;
 
-	i40e_detect_recover_hung(pf);
+	i40e_detect_recover_hung(pf->vsi[pf->lan_vsi]);
 	i40e_sync_filters_subtask(pf);
 	i40e_reset_subtask(pf);
 	i40e_handle_mdd_event(pf);