From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757257AbZBSMCQ (ORCPT ); Thu, 19 Feb 2009 07:02:16 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752238AbZBSMCA (ORCPT ); Thu, 19 Feb 2009 07:02:00 -0500 Received: from cantor.suse.de ([195.135.220.2]:47126 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752692AbZBSMCA (ORCPT ); Thu, 19 Feb 2009 07:02:00 -0500 Date: Thu, 19 Feb 2009 13:01:58 +0100 From: Nick Piggin To: Carsten Aulbert , "xfs@oss.sgi.com" , linux-kernel@vger.kernel.org Subject: Re: xfs problems (possibly after upgrading from linux kernel 2.6.27.10 to .14) Message-ID: <20090219120158.GA1747@wotan.suse.de> References: <499ACE6C.4060304@aei.mpg.de> <20090218091935.GD8830@disturbed> <499BD6BB.2000406@aei.mpg.de> <20090219061925.GE8830@disturbed> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090219061925.GE8830@disturbed> User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Feb 19, 2009 at 05:19:25PM +1100, Dave Chinner wrote: > On Wed, Feb 18, 2009 at 10:36:59AM +0100, Carsten Aulbert wrote: > > >> plus a few more nodes showing the same characteristics > > > > > > Hmmmm. Did this show up in 2.6.27.10? Or did it start occurring only > > > after you upgraded from .10 to .14? > > > > As far as I can see this only happened after the upgrade about 14 days > > ago. What strikes me odd is that we only had this occurring massively on > > Monday and Tuesday this week. > > > > I don't know if a certain access pattern could trigger this somehow. > > I suspect so. We've already had XFS trigger one bug in the new > lockless pagecache code, and the fix for that went in 2.6.27.11 - > between the good version and the version that you've been seeing > these memory corruptions on. I'm wondering if that fix exposed or > introduced another bug that you've hit.... Highly unlikely. It only introduces constraints on how the compiler may generate code, so it would have to be a compiler bug to cause a bug I think. I wonder how long you've been running with 2.6.27 based kernels without corruption?