slub: move synchronize_sched out of slab_mutex on shrink
authorVladimir Davydov <vdavydov.dev@gmail.com>
Tue, 13 Dec 2016 00:41:32 +0000 (16:41 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 13 Dec 2016 02:55:06 +0000 (18:55 -0800)
commit89e364db71fb5e7fc8d93228152abfa67daf35fa
tree7e70cba61d27fc6e7c7ebd21ec498b808ba2e132
parent13583c3d3224508582ec03d881d0b68dd3ee8e10
slub: move synchronize_sched out of slab_mutex on shrink

synchronize_sched() is a heavy operation and calling it per each cache
owned by a memory cgroup being destroyed may take quite some time.  What
is worse, it's currently called under the slab_mutex, stalling all works
doing cache creation/destruction.

Actually, there isn't much point in calling synchronize_sched() for each
cache - it's enough to call it just once - after setting cpu_partial for
all caches and before shrinking them.  This way, we can also move it out
of the slab_mutex, which we have to hold for iterating over the slab
cache list.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=172991
Link: http://lkml.kernel.org/r/0a10d71ecae3db00fb4421bcd3f82bcc911f4be4.1475329751.git.vdavydov.dev@gmail.com
Signed-off-by: Vladimir Davydov <vdavydov.dev@gmail.com>
Reported-by: Doug Smythies <dsmythies@telus.net>
Acked-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/slab.c
mm/slab.h
mm/slab_common.c
mm/slob.c
mm/slub.c