net/mlx5: DR, Warn and ignore SW steering rule insertion on QP err

In the event of SW steering QP entering error state, SW steering
cannot insert more rules, and will silently ignore the insertion
after issuing a warning.

Signed-off-by: Yuval Avnery <yuvalav@mellanox.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index 24f40e17f1766b86c9e9fdf936d14bc38d086b20..bfb14b4b19067598ab73706a838085c62f329958 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -325,10 +325,14 @@ static int dr_handle_pending_wc(struct mlx5dr_domain *dmn,
 
        do {
                ne = dr_poll_cq(send_ring->cq, 1);
-               if (ne < 0)
+               if (unlikely(ne < 0)) {
+                       mlx5_core_warn_once(dmn->mdev, "SMFS QPN 0x%x is disabled/limited",
+                                           send_ring->qp->qpn);
+                       send_ring->err_state = true;
                        return ne;
-               else if (ne == 1)
+               } else if (ne == 1) {
                        send_ring->pending_wqe -= send_ring->signal_th;
+               }
        } while (is_drain && send_ring->pending_wqe);
 
        return 0;
@@ -361,6 +365,14 @@ static int dr_postsend_icm_data(struct mlx5dr_domain *dmn,
        u32 buff_offset;
        int ret;
 
+       if (unlikely(dmn->mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR ||
+                    send_ring->err_state)) {
+               mlx5_core_dbg_once(dmn->mdev,
+                                  "Skipping post send: QP err state: %d, device state: %d\n",
+                                  send_ring->err_state, dmn->mdev->state);
+               return 0;
+       }
+
        spin_lock(&send_ring->lock);
 
        ret = dr_handle_pending_wc(dmn, send_ring);

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 474cf32a67c474af9832f079736675e201922d42..4fd14e9b7e1c1b36bac083045edf53ea9d7b3b14 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1285,6 +1285,7 @@ struct mlx5dr_send_ring {
        u8 sync_buff[MIN_READ_SYNC];
        struct mlx5dr_mr *sync_mr;
        spinlock_t lock; /* Protect the data path of the send ring */
+       bool err_state; /* send_ring is not usable in err state */
 };
 
 int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn);