Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758859Ab0FQJYz (ORCPT ); Thu, 17 Jun 2010 05:24:55 -0400 Received: from e39.co.us.ibm.com ([32.97.110.160]:46701 "EHLO e39.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755416Ab0FQJYx (ORCPT ); Thu, 17 Jun 2010 05:24:53 -0400 Date: Thu, 17 Jun 2010 14:54:42 +0530 From: Balbir Singh To: KAMEZAWA Hiroyuki Cc: "linux-mm@kvack.org" , "akpm@linux-foundation.org" , "nishimura@mxp.nes.nec.co.jp" , "linux-kernel@vger.kernel.org" Subject: Re: [BUGFIX][PATCH -mm] fix bad call of memcg_oom_recover at cancel move. Message-ID: <20100617092442.GJ4306@balbir.in.ibm.com> Reply-To: balbir@linux.vnet.ibm.com References: <20100617172034.00ea8835.kamezawa.hiroyu@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20100617172034.00ea8835.kamezawa.hiroyu@jp.fujitsu.com> User-Agent: Mutt/1.5.20 (2009-12-10) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3204 Lines: 90 * KAMEZAWA Hiroyuki [2010-06-17 17:20:34]: > From: KAMEZAWA Hiroyuki > > When cgroup_cancel_attach() is called via cgroup_attach_task(), > mem_cgroup_clear_mc() can be called even when any migration > was done. In such case, mc.to and mc.from is NULL. > > But, memcg-clean-up-waiting-move-acct-v2.patch > doesn't handle this correctly and pass NULL to memcg_oom_recover. > fix it. > > BUG: unable to handle kernel paging request at 000000000000114c > IP: [] memcg_oom_recover+0x9/0x30 > PGD 61ce4b067 PUD 613ea0067 PMD 0 > Oops: 0000 [#1] SMP > > Call Trace: > [] mem_cgroup_clear_mc+0x119/0x1c0 > [] mem_cgroup_cancel_attach+0xe/0x10 > [] cgroup_attach_task+0x26c/0x2c0 > [] cgroup_tasks_write+0x67/0x1c0 > [] ? might_fault+0xa5/0xb0 > [] ? might_fault+0x5c/0xb0 > [] cgroup_file_write+0x2d2/0x330 > [] ? print_lock_contention_bug+0x22/0xf0 > [] ? security_file_permission+0x1f/0x80 > [] vfs_write+0xc8/0x190 > [] sys_write+0x51/0x90 > [] system_call_fastpath+0x16/0x1b > Code: 20 48 39 43 20 41 bc f0 ff ff ff 75 c7 45 88 ae 48 11 00 00 45 31 e4 eb bb 66 0f 1f 84 00 00 00 00 00 55 48 89 e5 0f 1f 44 00 00 <8b> 87 4c 11 00 00 85 c0 75 05 c9 c3 0f 1f 00 48 89 f9 31 d2 be > RIP [] memcg_oom_recover+0x9/0x30 > > Signed-off-by: KAMEZAWA Hiroyuki > --- > mm/memcontrol.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > Index: mmotm-2.6.35-0611/mm/memcontrol.c > =================================================================== > --- mmotm-2.6.35-0611.orig/mm/memcontrol.c > +++ mmotm-2.6.35-0611/mm/memcontrol.c > @@ -4485,8 +4485,10 @@ static void mem_cgroup_clear_mc(void) > mc.to = NULL; > mc.moving_task = NULL; > spin_unlock(&mc.lock); > - memcg_oom_recover(from); > - memcg_oom_recover(to); > + if (from) > + memcg_oom_recover(from); > + if (to) > + memcg_oom_recover(to); > wake_up_all(&mc.waitq); May I recommend the following change instead Don't crash on a null memcg being passed, check if memcg is NULL and handle the condition gracefully Signed-off-by: Balbir Singh --- mm/memcontrol.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c6ece0a..d71c488 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1370,7 +1370,7 @@ static void memcg_wakeup_oom(struct mem_cgroup *mem) static void memcg_oom_recover(struct mem_cgroup *mem) { - if (mem->oom_kill_disable && atomic_read(&mem->oom_lock)) + if (mem && mem->oom_kill_disable && atomic_read(&mem->oom_lock)) memcg_wakeup_oom(mem); } -- 1.7.0.1 -- Three Cheers, Balbir -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/