net/mlx5: Fix health work queue spin lock to IRQ safe
authorMoshe Shemesh <moshe@mellanox.com>
Thu, 19 Oct 2017 11:14:29 +0000 (14:14 +0300)
committerSaeed Mahameed <saeedm@mellanox.com>
Thu, 26 Oct 2017 07:47:27 +0000 (00:47 -0700)
commit6377ed0bbae6fa28853e1679d068a9106c8a8908
tree4993b5cffac95b036566ffbe55287ec26794ff0d
parent5c25f65fd1e42685f7ccd80e0621829c105785d9
net/mlx5: Fix health work queue spin lock to IRQ safe

spin_lock/unlock of health->wq_lock should be IRQ safe.
It was changed to spin_lock_irqsave since adding commit 0179720d6be2
("net/mlx5: Introduce trigger_health_work function") which uses
spin_lock from asynchronous event (IRQ) context.
Thus, all spin_lock/unlock of health->wq_lock should have been moved
to IRQ safe mode.
However, one occurrence on new code using this lock missed that
change, resulting in possible deadlock:
  kernel: Possible unsafe locking scenario:
  kernel:       CPU0
  kernel:       ----
  kernel:  lock(&(&health->wq_lock)->rlock);
  kernel:  <Interrupt>
  kernel:    lock(&(&health->wq_lock)->rlock);
  kernel: #012 *** DEADLOCK ***

Fixes: 2a0165a034ac ("net/mlx5: Cancel delayed recovery work when unloading the driver")
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
drivers/net/ethernet/mellanox/mlx5/core/health.c