linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Suleiman Souhlal <ssouhlal@FreeBSD.org>
To: cgroups@vger.kernel.org
Cc: suleiman@google.com, glommer@parallels.com,
	kamezawa.hiroyu@jp.fujitsu.com, penberg@kernel.org, cl@linux.com,
	yinghan@google.com, hughd@google.com, gthelen@google.com,
	peterz@infradead.org, dan.magenheimer@oracle.com,
	hannes@cmpxchg.org, mgorman@suse.de,
	James.Bottomley@HansenPartnership.com, linux-mm@kvack.org,
	devel@openvz.org, linux-kernel@vger.kernel.org,
	Suleiman Souhlal <ssouhlal@FreeBSD.org>
Subject: [PATCH v2 11/13] memcg: Handle bypassed kernel memory charges.
Date: Fri,  9 Mar 2012 12:39:14 -0800	[thread overview]
Message-ID: <1331325556-16447-12-git-send-email-ssouhlal@FreeBSD.org> (raw)
In-Reply-To: <1331325556-16447-1-git-send-email-ssouhlal@FreeBSD.org>

When __mem_cgroup_try_charge() decides to bypass a slab charge
(because we are getting OOM killed or have a fatal signal pending),
we may end up with a slab that belongs to a memcg, but wasn't
charged to it. When we free such a slab page, we end up uncharging
it from the memcg, even though it was never charged, which may
lead to res_counter underflows.

To avoid this, when a charge is bypassed, we force the charge,
without checking for the bypass conditions or doing any reclaim.
This may cause the cgroup's usage to temporarily go above its limit.

Signed-off-by: Suleiman Souhlal <suleiman@google.com>
---
 mm/memcontrol.c |   15 +++++++++++++--
 1 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 72e83af..9f5e9d8 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5672,16 +5672,27 @@ memcg_charge_kmem(struct mem_cgroup *memcg, gfp_t gfp, long long delta)
 
 	ret = 0;
 
-	_memcg = memcg;
 	if (memcg && !mem_cgroup_test_flag(memcg,
 	    MEMCG_INDEPENDENT_KMEM_LIMIT)) {
+		_memcg = memcg;
 		ret = __mem_cgroup_try_charge(NULL, gfp, delta / PAGE_SIZE,
 		    &_memcg, may_oom);
 		if (ret == -ENOMEM)
 			return ret;
+		else if (ret == -EINTR) {
+			/*
+			 * __mem_cgroup_try_charge() chose to bypass to root due
+			 * to OOM kill or fatal signal.
+			 * Since our only options are to either fail the
+			 * allocation or charge it to this cgroup, force the
+			 * change, going above the limit if needed.
+			 */
+			ret = res_counter_charge_nofail(&memcg->res, delta,
+			    &fail_res);
+		}
 	}
 
-	if (memcg && _memcg == memcg)
+	if (memcg)
 		ret = res_counter_charge(&memcg->kmem, delta, &fail_res);
 
 	return ret;
-- 
1.7.7.3

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2012-03-09 20:39 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-03-09 20:39 [PATCH v2 00/13] Memcg Kernel Memory Tracking Suleiman Souhlal
2012-03-09 20:39 ` [PATCH v2 01/13] memcg: Consolidate various flags into a single flags field Suleiman Souhlal
2012-03-11  7:50   ` Glauber Costa
2012-03-09 20:39 ` [PATCH v2 02/13] memcg: Kernel memory accounting infrastructure Suleiman Souhlal
2012-03-11  8:12   ` Glauber Costa
2012-03-13  6:24     ` KAMEZAWA Hiroyuki
2012-03-13 10:37       ` Glauber Costa
2012-03-13 17:00         ` Greg Thelen
2012-03-13 17:31           ` Glauber Costa
2012-03-14  0:15         ` KAMEZAWA Hiroyuki
2012-03-14 12:29           ` Glauber Costa
2012-03-15  0:48             ` KAMEZAWA Hiroyuki
2012-03-15 11:07               ` Glauber Costa
2012-03-15 11:13                 ` Peter Zijlstra
2012-03-15 11:21                   ` Glauber Costa
2012-03-12 12:38   ` Glauber Costa
2012-03-09 20:39 ` [PATCH v2 03/13] memcg: Uncharge all kmem when deleting a cgroup Suleiman Souhlal
2012-03-11  8:19   ` Glauber Costa
2012-03-13 23:16     ` Suleiman Souhlal
2012-03-14 11:59       ` Glauber Costa
2012-03-13  6:27   ` KAMEZAWA Hiroyuki
2012-03-09 20:39 ` [PATCH v2 04/13] memcg: Make it possible to use the stock for more than one page Suleiman Souhlal
2012-03-11 10:49   ` Glauber Costa
2012-03-09 20:39 ` [PATCH v2 05/13] memcg: Reclaim when more than one page needed Suleiman Souhlal
2012-03-09 20:39 ` [PATCH v2 06/13] slab: Add kmem_cache_gfp_flags() helper function Suleiman Souhlal
2012-03-11 10:53   ` Glauber Costa
2012-03-13 23:21     ` Suleiman Souhlal
2012-03-14 11:48       ` Glauber Costa
2012-03-14 22:08         ` Suleiman Souhlal
2012-03-09 20:39 ` [PATCH v2 07/13] memcg: Slab accounting Suleiman Souhlal
2012-03-11 10:25   ` Glauber Costa
2012-03-13 22:50     ` Suleiman Souhlal
2012-03-14 10:47       ` Glauber Costa
2012-03-14 22:04         ` Suleiman Souhlal
2012-03-15 11:40           ` Glauber Costa
2012-03-09 20:39 ` [PATCH v2 08/13] memcg: Make dentry slab memory accounted in kernel memory accounting Suleiman Souhlal
2012-03-09 20:39 ` [PATCH v2 09/13] memcg: Account for kmalloc " Suleiman Souhlal
2012-03-11 12:21   ` Glauber Costa
2012-03-09 20:39 ` [PATCH v2 10/13] memcg: Track all the memcg children of a kmem_cache Suleiman Souhlal
2012-03-09 20:39 ` Suleiman Souhlal [this message]
2012-03-09 20:39 ` [PATCH v2 12/13] memcg: Per-memcg memory.kmem.slabinfo file Suleiman Souhlal
2012-03-11 10:35   ` Glauber Costa
2012-03-09 20:39 ` [PATCH v2 13/13] memcg: Document kernel memory accounting Suleiman Souhlal
2012-03-11 10:42   ` Glauber Costa
2012-03-10  6:25 ` [PATCH v2 00/13] Memcg Kernel Memory Tracking Suleiman Souhlal

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=1331325556-16447-12-git-send-email-ssouhlal@FreeBSD.org \
    --to=ssouhlal@freebsd.org \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=cgroups@vger.kernel.org \
    --cc=cl@linux.com \
    --cc=dan.magenheimer@oracle.com \
    --cc=devel@openvz.org \
    --cc=glommer@parallels.com \
    --cc=gthelen@google.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=penberg@kernel.org \
    --cc=peterz@infradead.org \
    --cc=suleiman@google.com \
    --cc=yinghan@google.com \
    /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;
as well as URLs for NNTP newsgroup(s).