net/mlx5: Reduce No CQ found log level from warn to debug
authorParav Pandit <parav@mellanox.com>
Thu, 5 Dec 2019 22:06:20 +0000 (16:06 -0600)
committerSaeed Mahameed <saeedm@mellanox.com>
Tue, 7 Jan 2020 18:41:17 +0000 (10:41 -0800)
In below sequence, a EQE entry arrives for a CQ which is on the path of
being destroyed.

           cpu-0               cpu-1
           ------              -----
mlx5_core_destroy_cq()      mlx5_eq_comp_int()
  mlx5_eq_del_cq()          [..]
    radix_tree_delete()     [..]
  [..]                         mlx5_eq_cq_get() /* Didn't find CQ is
                                                 * a valid case.
                                                 */
  /* destroy CQ in hw */
  mlx5_cmd_exec()

This is still a valid scenario and correct delete CQ sequence, as
mirror of the CQ create sequence.
Hence, suppress the non harmful debug message from warn to debug level.
Keep the debug log message rate limited because user application can
trigger it repeatedly.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
drivers/net/ethernet/mellanox/mlx5/core/eq.c

index 580c71c..2c716ab 100644 (file)
@@ -156,7 +156,8 @@ static int mlx5_eq_comp_int(struct notifier_block *nb,
                        cq->comp(cq, eqe);
                        mlx5_cq_put(cq);
                } else {
-                       mlx5_core_warn(eq->dev, "Completion event for bogus CQ 0x%x\n", cqn);
+                       dev_dbg_ratelimited(eq->dev->device,
+                                           "Completion event for bogus CQ 0x%x\n", cqn);
                }
 
                ++eq->cons_index;