DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:sender:in-reply-to:references:from:date
         :x-google-sender-auth:message-id:subject:to:cc:content-type;
        b=dJMRPv/KcHbAMb2gbvIaP6T9Kp+DGj5nOSwWmj5e22CTmbwYMR96kPXBSM8xA0RfEJ
         jziHJJ3ImlDwPbcadyXvhipm2jdqdjVpipzwkLmOizaULzELrqIYNwChFswGVA3PY+tM
         ixw3ZiGIbPL+qrL+eQU4/1MbPJMsFuFHnmilU=
MIME-Version: 1.0
In-Reply-To: <BANLkTint+Qs+cO+wKUJGytnVY3X1bp+8rQ@mail.gmail.com>
References: <BANLkTi=NTLn4Lx7EkybuA8-diTVOvMDxBw@mail.gmail.com>
 <BANLkTinEDXHuRUYpYN0d95+fz4+F7ccL4w@mail.gmail.com> <4DD5DC06.6010204@jp.fujitsu.com>
 <BANLkTik=7C5qFZTsPQG4JYY-MEWDTHdc6A@mail.gmail.com> <BANLkTins7qxWVh0bEwtk1Vx+m98N=oYVtw@mail.gmail.com>
 <20110520140856.fdf4d1c8.kamezawa.hiroyu@jp.fujitsu.com> <20110520101120.GC11729@random.random>
 <BANLkTikAFMvpgHR2dopd+Nvjfyw_XT5=LA@mail.gmail.com> <20110520153346.GA1843@barrios-desktop>
 <BANLkTi=X+=Wh1MLs7Fc-v-OMtxAHbcPmxA@mail.gmail.com> <20110520161934.GA2386@barrios-desktop>
 <BANLkTi=4C5YAxwAFWC6dsAPMR3xv6LP1hw@mail.gmail.com> <BANLkTimThVw7-PN6ypBBarqXJa1xxYA_Ow@mail.gmail.com>
 <BANLkTint+Qs+cO+wKUJGytnVY3X1bp+8rQ@mail.gmail.com>
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Sat, 21 May 2011 23:14:11 +0900
Message-ID: <BANLkTim7+DY4wcNERwAk2zVZSohmNYBUbA@mail.gmail.com>
Subject: Re: Kernel falls apart under light memory pressure (i.e. linking vmlinux)
To: Andrew Lutomirski <luto@mit.edu>
Cc: Minchan Kim <minchan.kim@gmail.com>,
        Andrea Arcangeli <aarcange@redhat.com>,
        KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
        fengguang.wu@intel.com, andi@firstfloor.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, mgorman@suse.de, hannes@cmpxchg.org,
        riel@redhat.com
Content-Type: text/plain; charset=ISO-8859-1
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 752
Lines: 19

> I did some tracing and the oops happens from the second call to
> shrink_page_list after should_reclaim_stall returns true and it hits
> the same pages in the same order that the earlier call just finished
> calling SetPageActive on.

Can you please share your tracing patch and raw tracing result log?

Thanks.

> I have *not* confirmed that the two calls
> happened from the same call to shrink_inactive_list, but something's
> certainly wrong in there.
>
> This is very easy to reproduce on my laptop.
--
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/