[PATCHv2] nvme: ensure disabling pairs with unquiesce

[Ming Lei]

On Wed, Jul 12, 2023 at 08:40:04AM -0700, Keith Busch wrote:
> From: Keith Busch <kbusch at kernel.org>
> 
> If any error handling that disables the controller fails to queue the
> reset work, like if the state changed to disconnected inbetween, then
> the failed teardown needs to unquiesce the queues since it's no longer
> paired with reset_work. Just make sure that the controller can be put
> into a resetting state prior to starting the disable so that no other
> handling can change with the queue states while recovery is happening.
> 
> Reported-by: Ming Lei <ming.lei at redhat.com>
> Signed-off-by: Keith Busch <kbusch at kernel.org>
> ---
> v1->v2:
> 
>   Don't wait for a resetting state as that could deadlock. Just abort
>   the disabling if the error handling can't set the state to resetting.

Looks fine:

Reviewed-by: Ming Lei <ming.lei at redhat.com>


Thanks, 
Ming