]> git.proxmox.com Git - mirror_ubuntu-kernels.git/commit
ide: remove BUG_ON(in_interrupt() || irqs_disabled()) from ide_unregister()
authorSebastian Andrzej Siewior <bigeasy@linutronix.de>
Tue, 15 Dec 2020 03:03:27 +0000 (19:03 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 15 Dec 2020 20:13:36 +0000 (12:13 -0800)
commitec680c1990e70c44d6b4452300a62b15f5c51f9c
treeab377d99cb8280cd95cb8ca2fcb5a3f9ef7507c1
parent483e6417aea5c6d278e6aa7facc50d961d95a0a9
ide: remove BUG_ON(in_interrupt() || irqs_disabled()) from ide_unregister()

In the discussion about preempt count consistency across kernel
configurations:

 https://lore.kernel.org/r/20200914204209.256266093@linutronix.de/

it was concluded that the usage of in_interrupt() and related context
checks should be removed from non-core code.

Both BUG_ON()s in ide-probe.c were introduced in commit
   4015c949fb465 ("[PATCH] update ide core")

when ide_unregister() was extended with semaphore based locking.  Both
checks won't complain about disabled preemption which is also wrong.

The might_sleep() in today's mutex_lock() will complain about the
missuses.

Remove the BUG_ON() statements.

Link: https://lkml.kernel.org/r/20201120092421.1023428-3-bigeasy@linutronix.de
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/ide/ide-probe.c