Re: [PATCH 5/5] block: revert back to synchronous request_queue removal

[Christoph Hellwig]

On Tue, Apr 14, 2020 at 08:58:52PM +0000, Luis Chamberlain wrote:
> > I think this needs a WARN_ON thrown in to enforece the calling context.
> 
> I considered adding a might_sleep() but upon review with Bart, he noted
> that this function already has a mutex_lock(), and if you look under the
> hood of mutex_lock(), it has a might_sleep() at the very top. The
> warning then is implicit.

It might just be a personal preference, but I think the documentation
value of a WARN_ON_ONCE or might_sleep with a comment at the top of
the function is much higher than a blurb in a long kerneldoc text and
a later mutex_lock.