From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757930Ab1JRNmB (ORCPT ); Tue, 18 Oct 2011 09:42:01 -0400 Received: from lucidpixels.com ([72.73.18.11]:49501 "EHLO lucidpixels.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755229Ab1JRNmA (ORCPT ); Tue, 18 Oct 2011 09:42:00 -0400 Date: Tue, 18 Oct 2011 09:42:00 -0400 (EDT) From: Justin Piszcz To: Hugh Dickins cc: Pawel Sikora , arekm@pld-linux.org, Anders Ossowicki , linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: 3.1-rc9: BUG: soft lockup in find_get_pages+0x4e/0x140 In-Reply-To: Message-ID: References: User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; format=flowed; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 17 Oct 2011, Hugh Dickins wrote: > On Sat, 15 Oct 2011, Justin Piszcz wrote: >> >> With 3.1-rc9 during a filesystem dump, this occurred, I thought a previous >> patch fixed this but it did not, it occurs MUCH less (first time in a couple >> of weeks) but the problem still occurs. > > Shaohua's nr_skip patch fixed, or at least worked around, the error I > introduced there in 3.1-rc1. But now you're finding a similar problem > still in 3.1-rc9: that confirms what you and others already reported, > that something like it occurs even in 3.0 without my bug. Thanks, Patch applied: # patch -p1 < /tmp/memory_patch.patch patching file mm/filemap.c patching file mm/memcontrol.c Now running 3.1-rc9 + this patch + re-enabled frame pointers. Justin.