From davej Wed Jun 20 11:37:32 2012
Return-Path: oleg@redhat.com
X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on
gelk.kernelslacker.org
X-Spam-Level:
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,T_RP_MATCHES_RCVD,
UNPARSEABLE_RELAY autolearn=ham version=3.3.2
Received: from mail.corp.redhat.com [10.4.128.1]
by gelk.kernelslacker.org with IMAP (fetchmail-6.3.21)
for <davej@localhost> (single-drop); Wed, 20 Jun 2012 11:37:32 -0400 (EDT)
Received: from zmta05.collab.prod.int.phx2.redhat.com (LHLO
zmta05.collab.prod.int.phx2.redhat.com) (10.5.81.12) by
zmail11.collab.prod.int.phx2.redhat.com with LMTP; Wed, 20 Jun 2012
11:37:18 -0400 (EDT)
Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
by zmta05.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 539B1F2080
for <davej@mail.corp.redhat.com>; Wed, 20 Jun 2012 11:37:18 -0400 (EDT)
Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id q5KFbGde008148;
Wed, 20 Jun 2012 11:37:17 -0400
Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
oleg@redhat.com; Wed, 20 Jun 2012 17:35:20 +0200 (CEST)
Date: Wed, 20 Jun 2012 17:35:19 +0200
From: Oleg Nesterov <oleg@redhat.com>
To: Dave Jones <davej@redhat.com>
Cc: Fedora Kernel Team <kernel-team@fedoraproject.org>
Subject: Re: [Bug 832867] BUG: Bad rss-counter state mm:xxxxxxxx idx:1
val:-2 output in console
Message-ID: <20120620153519.GA17642@redhat.com>
References: <bug-832867-176318@bugzilla.redhat.com> <bug-832867-176318-mTgNUmpuHi@bugzilla.redhat.com> <20120619172633.GG26103@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20120619172633.GG26103@redhat.com>
User-Agent: Mutt/1.5.18 (2008-05-17)
X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
Status: RO
Content-Length: 2687
Lines: 84
Hi Dave,
sorry for delay
On 06/19, Dave Jones wrote:
>
> > https://bugzilla.redhat.com/show_bug.cgi?id=832867
> >
> > --- Comment #11 from Bojan Smojver <bojan@rexursive.com> ---
> > Patches for this:
> >
> > http://comments.gmane.org/gmane.linux.kernel/1306252
>
> Oleg,
>
> are these patches final ? Are they headed for stable ?
No. And 1/2 was wrong actually, see
http://marc.info/?l=linux-kernel&m=133911852215275
Konstantin sent the minimal fix,
mm-correctly-synchronize-rss-counters-at-exit-exec.patch in -mm tree,
attached below.
Oleg.
------------------------------------------------------
From: Konstantin Khlebnikov <khlebnikov@openvz.org>
Subject: mm: correctly synchronize rss-counters at exit/exec
do_exit() and exec_mmap() call sync_mm_rss() before mm_release() does
put_user(clear_child_tid) which can update task->rss_stat and thus make
mm->rss_stat inconsistent. This triggers the "BUG:" printk in check_mm().
Let's fix this bug in the safest way, and optimize/cleanup this later.
Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
fs/exec.c | 2 +-
kernel/exit.c | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff -puN fs/exec.c~mm-correctly-synchronize-rss-counters-at-exit-exec fs/exec.c
--- a/fs/exec.c~mm-correctly-synchronize-rss-counters-at-exit-exec
+++ a/fs/exec.c
@@ -819,10 +819,10 @@ static int exec_mmap(struct mm_struct *m
/* Notify parent that we're no longer interested in the old VM */
tsk = current;
old_mm = current->mm;
- sync_mm_rss(old_mm);
mm_release(tsk, old_mm);
if (old_mm) {
+ sync_mm_rss(old_mm);
/*
* Make sure that if there is a core dump in progress
* for the old mm, we get out and die instead of going
diff -puN kernel/exit.c~mm-correctly-synchronize-rss-counters-at-exit-exec kernel/exit.c
--- a/kernel/exit.c~mm-correctly-synchronize-rss-counters-at-exit-exec
+++ a/kernel/exit.c
@@ -643,6 +643,7 @@ static void exit_mm(struct task_struct *
mm_release(tsk, mm);
if (!mm)
return;
+ sync_mm_rss(mm);
/*
* Serialize with any possible pending coredump.
* We must hold mmap_sem around checking core_state
_
Subject: From: Konstantin Khlebnikov <khlebnikov@openvz.org>
Patches currently in -mm which might be from khlebnikov@openvz.org are
linux-next.patch
memcg-fix-use_hierarchy-css_is_ancestor-oops-regression.patch
mm-correctly-synchronize-rss-counters-at-exit-exec.patch