On Tue 07-03-17 14:43:49, Tejun Heo wrote:

Hello,

On Tue, Mar 07, 2017 at 12:14:20PM +0100, Jan Kara wrote:

quoted

quoted

Given that this isn't a super hot path, I think it'd be far better to
stick to a simpler synchronization mechanism.

I'd be happy to do so however struct gendisk does not have any lock in it
and introducing a special lock just for that seems awkward.

I think even a awkward shared lock is better than memory barriers.
Barriers are the trickiest locking construct to get right and we
really shouldn't using that unless absolutely necessary.

OK, I'll try to come up with something.

								Honza

-- 
Jan Kara [off-list ref]
SUSE Labs, CR