From: Vladimir Davydov <vdavydov@parallels.com>
To: mhocko@suse.cz, akpm@linux-foundation.org
Cc: glommer@gmail.com, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, cgroups@vger.kernel.org, devel@openvz.org,
Johannes Weiner <hannes@cmpxchg.org>,
Balbir Singh <bsingharora@gmail.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Pekka Enberg <penberg@kernel.org>,
Christoph Lameter <cl@linux.com>
Subject: [PATCH RESEND 04/11] memcg, slab: fix barrier usage when accessing memcg_caches
Date: Mon, 6 Jan 2014 12:44:55 +0400 [thread overview]
Message-ID: <bee9853bac6bf196ffe1a1bb1634e8947b1d9d09.1388996525.git.vdavydov@parallels.com> (raw)
In-Reply-To: <cover.1388996525.git.vdavydov@parallels.com>
Each root kmem_cache has pointers to per-memcg caches stored in its
memcg_params::memcg_caches array. Whenever we want to allocate a slab
for a memcg, we access this array to get per-memcg cache to allocate
from (see memcg_kmem_get_cache()). The access must be lock-free for
performance reasons, so we should use barriers to assert the kmem_cache
is up-to-date.
First, we should place a write barrier immediately before setting the
pointer to it in the memcg_caches array in order to make sure nobody
will see a partially initialized object. Second, we should issue a read
barrier before dereferencing the pointer to conform to the write
barrier.
However, currently the barrier usage looks rather strange. We have a
write barrier *after* setting the pointer and a read barrier *before*
reading the pointer, which is incorrect. This patch fixes this.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Glauber Costa <glommer@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
mm/memcontrol.c | 24 ++++++++++--------------
mm/slab.h | 12 +++++++++++-
2 files changed, 21 insertions(+), 15 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index f8eb994..999e7d4 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3238,12 +3238,14 @@ void memcg_register_cache(struct kmem_cache *s)
list_add(&s->memcg_params->list, &memcg->memcg_slab_caches);
mutex_unlock(&memcg->slab_caches_mutex);
- root->memcg_params->memcg_caches[id] = s;
/*
- * the readers won't lock, make sure everybody sees the updated value,
- * so they won't put stuff in the queue again for no reason
+ * Since readers won't lock (see cache_from_memcg_idx()), we need a
+ * barrier here to ensure nobody will see the kmem_cache partially
+ * initialized.
*/
- wmb();
+ smp_wmb();
+
+ root->memcg_params->memcg_caches[id] = s;
}
void memcg_unregister_cache(struct kmem_cache *s)
@@ -3569,7 +3571,7 @@ struct kmem_cache *__memcg_kmem_get_cache(struct kmem_cache *cachep,
gfp_t gfp)
{
struct mem_cgroup *memcg;
- int idx;
+ struct kmem_cache *memcg_cachep;
VM_BUG_ON(!cachep->memcg_params);
VM_BUG_ON(!cachep->memcg_params->is_root_cache);
@@ -3583,15 +3585,9 @@ struct kmem_cache *__memcg_kmem_get_cache(struct kmem_cache *cachep,
if (!memcg_can_account_kmem(memcg))
goto out;
- idx = memcg_cache_id(memcg);
-
- /*
- * barrier to mare sure we're always seeing the up to date value. The
- * code updating memcg_caches will issue a write barrier to match this.
- */
- read_barrier_depends();
- if (likely(cache_from_memcg_idx(cachep, idx))) {
- cachep = cache_from_memcg_idx(cachep, idx);
+ memcg_cachep = cache_from_memcg_idx(cachep, memcg_cache_id(memcg));
+ if (likely(memcg_cachep)) {
+ cachep = memcg_cachep;
goto out;
}
diff --git a/mm/slab.h b/mm/slab.h
index 0859c42..72d1f9d 100644
--- a/mm/slab.h
+++ b/mm/slab.h
@@ -163,9 +163,19 @@ static inline const char *cache_name(struct kmem_cache *s)
static inline struct kmem_cache *
cache_from_memcg_idx(struct kmem_cache *s, int idx)
{
+ struct kmem_cache *cachep;
+
if (!s->memcg_params)
return NULL;
- return s->memcg_params->memcg_caches[idx];
+ cachep = s->memcg_params->memcg_caches[idx];
+
+ /*
+ * Make sure we will access the up-to-date value. The code updating
+ * memcg_caches issues a write barrier to match this (see
+ * memcg_register_cache()).
+ */
+ smp_read_barrier_depends();
+ return cachep;
}
static inline struct kmem_cache *memcg_root_cache(struct kmem_cache *s)
--
1.7.10.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2014-01-06 8:45 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-06 8:44 [PATCH RESEND 00/11] kmemcg-fixes Vladimir Davydov
2014-01-06 8:44 ` [PATCH RESEND 01/11] slab: cleanup kmem_cache_create_memcg() error handling Vladimir Davydov
2014-01-06 8:44 ` [PATCH RESEND 02/11] memcg, slab: kmem_cache_create_memcg(): fix memleak on fail path Vladimir Davydov
2014-01-06 8:44 ` [PATCH RESEND 03/11] memcg, slab: cleanup memcg cache initialization/destruction Vladimir Davydov
2014-01-06 8:44 ` Vladimir Davydov [this message]
2014-01-06 8:44 ` [PATCH RESEND 05/11] memcg: fix possible NULL deref while traversing memcg_slab_caches list Vladimir Davydov
2014-01-06 8:44 ` [PATCH RESEND 06/11] memcg, slab: fix races in per-memcg cache creation/destruction Vladimir Davydov
2014-01-06 8:44 ` [PATCH RESEND 07/11] memcg: get rid of kmem_cache_dup Vladimir Davydov
2014-01-06 8:44 ` [PATCH RESEND 08/11] slab: do not panic if we fail to create memcg cache Vladimir Davydov
2014-01-06 8:45 ` [PATCH RESEND 09/11] memcg, slab: RCU protect memcg_params for root caches Vladimir Davydov
2014-01-06 8:45 ` [PATCH RESEND 10/11] memcg: remove KMEM_ACCOUNTED_ACTIVATED flag Vladimir Davydov
2014-01-06 8:45 ` [PATCH RESEND 11/11] memcg: rework memcg_update_kmem_limit synchronization Vladimir Davydov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=bee9853bac6bf196ffe1a1bb1634e8947b1d9d09.1388996525.git.vdavydov@parallels.com \
--to=vdavydov@parallels.com \
--cc=akpm@linux-foundation.org \
--cc=bsingharora@gmail.com \
--cc=cgroups@vger.kernel.org \
--cc=cl@linux.com \
--cc=devel@openvz.org \
--cc=glommer@gmail.com \
--cc=hannes@cmpxchg.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
--cc=penberg@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox