linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: "linux-mm@kvack.org" <linux-mm@kvack.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"balbir@linux.vnet.ibm.com" <balbir@linux.vnet.ibm.com>,
	"akpm@linux-foundation.org" <akpm@linux-foundation.org>
Subject: [mmotm][BUGFIX][PATCH] memcg: fix softlimit css refcnt handling.
Date: Fri, 4 Sep 2009 16:37:58 +0900	[thread overview]
Message-ID: <20090904163758.a5604fee.kamezawa.hiroyu@jp.fujitsu.com> (raw)
In-Reply-To: <20090904154050.25873aa5.nishimura@mxp.nes.nec.co.jp>

On Fri, 4 Sep 2009 15:40:50 +0900
Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> wrote:
> > Ah, one more question. What memory.usage_in_bytes shows in that case ?
> > If not zero, charge/uncharge coalescing is guilty.
> > 
> usage_in_bytes is 0.
> I've confirmed by crash command that the mem_cgroup has extra ref counts.
> 
> I'll dig more..
> 
BTW, do you use softlimit ? I found this but...Hmm

==
SoftLimit tree 'find next one' loop uses next_mz to remember
next one to be visited if reclaimd==0.
But css'refcnt handling for next_mz is not enough and it makes
css->refcnt leak.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>


---
 mm/memcontrol.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

Index: mmotm-2.6.31-Aug27/mm/memcontrol.c
===================================================================
--- mmotm-2.6.31-Aug27.orig/mm/memcontrol.c
+++ mmotm-2.6.31-Aug27/mm/memcontrol.c
@@ -2261,6 +2261,8 @@ unsigned long mem_cgroup_soft_limit_recl
 		if (!reclaimed) {
 			do {
 				/*
+				 * Loop until we find yet another one.
+				 *
 				 * By the time we get the soft_limit lock
 				 * again, someone might have aded the
 				 * group back on the RB tree. Iterate to
@@ -2271,7 +2273,12 @@ unsigned long mem_cgroup_soft_limit_recl
 				 */
 				next_mz =
 				__mem_cgroup_largest_soft_limit_node(mctz);
-			} while (next_mz == mz);
+				if (next_mz == mz) {
+					css_put(&next_mz->mem->css);
+					next_mz = NULL;
+				} else /* next_mz == NULL or other memcg */
+					break;
+			} while (1);
 		}
 		mz->usage_in_excess =
 			res_counter_soft_limit_excess(&mz->mem->res);
@@ -2299,6 +2306,8 @@ unsigned long mem_cgroup_soft_limit_recl
 			loop > MEM_CGROUP_MAX_SOFT_LIMIT_RECLAIM_LOOPS))
 			break;
 	} while (!nr_reclaimed);
+	if (next_mz)
+		css_put(&next_mz->mem->css);
 	return nr_reclaimed;
 }
 

--
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>

  parent reply	other threads:[~2009-09-04  7:39 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-02  0:34 [mmotm][PATCH 1/2] memcg: softlimit clean up KAMEZAWA Hiroyuki
2009-09-02  0:35 ` [mmotm][PATCH 2/2] memcg: reduce calls for soft limit excess KAMEZAWA Hiroyuki
2009-09-02  5:16   ` Daisuke Nishimura
2009-09-02  5:20     ` KAMEZAWA Hiroyuki
2009-09-02  5:56   ` [mmotm][PATCH 2/2 v2] " KAMEZAWA Hiroyuki
2009-09-02 20:15     ` Balbir Singh
2009-09-03  0:34       ` KAMEZAWA Hiroyuki
2009-09-02  3:56 ` [mmotm][PATCH 1/2] memcg: softlimit clean up Daisuke Nishimura
2009-09-02  4:41 ` [mmotm][PATCH] coalescing uncharge KAMEZAWA Hiroyuki
2009-09-02  6:15   ` KAMEZAWA Hiroyuki
2009-09-02  9:29   ` [mmotm][experimental][PATCH] coalescing charge KAMEZAWA Hiroyuki
2009-09-02 19:58     ` Balbir Singh
2009-09-03  0:24       ` KAMEZAWA Hiroyuki
2009-09-03  5:17     ` Daisuke Nishimura
2009-09-04  4:18       ` KAMEZAWA Hiroyuki
2009-09-04  5:11         ` Daisuke Nishimura
2009-09-04  5:21           ` KAMEZAWA Hiroyuki
2009-09-04  5:26             ` KAMEZAWA Hiroyuki
2009-09-04  6:40               ` Daisuke Nishimura
2009-09-04  6:50                 ` KAMEZAWA Hiroyuki
2009-09-04  7:37                 ` KAMEZAWA Hiroyuki [this message]
2009-09-04  7:45                   ` [mmotm][BUGFIX][PATCH] memcg: fix softlimit css refcnt handling KAMEZAWA Hiroyuki
2009-09-04 10:07                   ` Daisuke Nishimura
2009-09-06 23:04                     ` [mmotm][BUGFIX][PATCH] memcg: fix softlimit css refcnt handling(yet another one) Daisuke Nishimura
2009-09-07  0:49                       ` KAMEZAWA Hiroyuki
2009-09-07  4:51                         ` Balbir Singh
2009-09-04  5:47           ` [mmotm][experimental][PATCH] coalescing charge KAMEZAWA Hiroyuki

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=20090904163758.a5604fee.kamezawa.hiroyu@jp.fujitsu.com \
    --to=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=akpm@linux-foundation.org \
    --cc=balbir@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=nishimura@mxp.nes.nec.co.jp \
    /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