From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2992711AbXCGXHZ (ORCPT ); Wed, 7 Mar 2007 18:07:25 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S2992716AbXCGXHZ (ORCPT ); Wed, 7 Mar 2007 18:07:25 -0500 Received: from e32.co.us.ibm.com ([32.97.110.150]:38169 "EHLO e32.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2992711AbXCGXHX (ORCPT ); Wed, 7 Mar 2007 18:07:23 -0500 Subject: Re: 2.6.21-rc2-mm2 hang From: Dave Hansen To: "Siddha, Suresh B" Cc: Andrew Morton , linux-kernel@vger.kernel.org, Ingo Molnar , Thomas Gleixner , Nick Piggin In-Reply-To: <20070307221628.GD26473@linux-os.sc.intel.com> References: <20070306004408.d3f6434d.akpm@linux-foundation.org> <1173305536.868.8.camel@localhost.localdomain> <20070307221628.GD26473@linux-os.sc.intel.com> Content-Type: text/plain Date: Wed, 07 Mar 2007 15:07:21 -0800 Message-Id: <1173308841.868.14.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.6.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2007-03-07 at 14:16 -0800, Siddha, Suresh B wrote: > On Wed, Mar 07, 2007 at 02:12:16PM -0800, Dave Hansen wrote: > > I'm seeing weird hangs running ltp on 2.6.21-rc2-mm2. It manifests > > itself by the waitpid06 test in LTP hanging. This is very, very > > reproducible in about 5 seconds by adding '-s wait' to the ltp command > > line. > > > > I see 4 waitpid06 processes on my 4-way machine spinning in userspace. > > But, the weird part is that I can't ssh in once this happens, but I can > > log in to the console. I've bisected it down to: > > > > sched-fix-idle-load-balancing-in-softirqd-context > > This sounds like an issue in merge we recently had and 2.6.21-rc2-mm2 already > has a fix for this. > > sched-fix-idle-load-balancing-in-softirqd-context-fix.patch > > Can you please apply both > sched-fix-idle-load-balancing-in-softirqd-context > sched-fix-idle-load-balancing-in-softirqd-context-fix.patch > and see if you still see this problem? I was confused that I couldn't find this fix in my patches directory, but I double-checked, and realized that I was still on 2.6.21-rc2-mm1. I must have grabbed the wrong broken-out tarball when I started the bisect. Darn. Anyway, 2.6.21-rc2-mm2 appears to be OK. Sorry for the false alarm. -- Dave