From: Nick Dokos Subject: Re: Soft lockups in mballoc code Date: Wed, 14 May 2008 13:45:37 -0400 Message-ID: <27364.1210787137@alphaville.zko.hp.com> References: <25037.1210612895@alphaville.zko.hp.com> <1210785517.3657.42.camel@localhost.localdomain> Reply-To: nicholas.dokos@hp.com Cc: nicholas.dokos@hp.com, linux-ext4@vger.kernel.org To: cmm@us.ibm.com Return-path: Received: from g1t0028.austin.hp.com ([15.216.28.35]:34312 "EHLO g1t0028.austin.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757947AbYENRpj (ORCPT ); Wed, 14 May 2008 13:45:39 -0400 In-Reply-To: Message from Mingming Cao of "Wed, 14 May 2008 10:18:37 PDT." <1210785517.3657.42.camel@localhost.localdomain> Sender: linux-ext4-owner@vger.kernel.org List-ID: Mingming Cao wrote: > On Mon, 2008-05-12 at 13:21 -0400, Nick Dokos wrote: > > I keep getting soft lockups in mballoc code, running fsstress. > > I'm running a 2.6.25.2 kernel, with 1.40.8 e2fsprogs. Here > > is sample console output on a 2-cpu Opteron box with a 2Tb filesystem - > > the two stack traces keep repeating every minute or so and the only > > way to get out is to reset the box: > > > Nick, thanks for reporting this. I remember saw soft lockup issue with > mballoc on 2.6.25-rc kernels. Could you run the latest 2.6.26-rc2 > kernel? There are lots of bug fixes have been pushed to 2.6.26-rc1. > Aneesh, do you know if this is a new issue? > Hi Mingming, I'll do so at the earliest opportunity. In the meantime, I wanted to let you know that I tried the (stable) 2.6.25.3 kernel on x86-64 and encountered the same problem, but I also tried it on an IA64 system (again with the 2.6.25.3 kernel) and did *not* encounter the problem. Not sure what this means yet, but I'll let you know as soon as I get some more information. Thanks very much, Nick