Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756886Ab0KNTLJ (ORCPT ); Sun, 14 Nov 2010 14:11:09 -0500 Received: from mail-qw0-f46.google.com ([209.85.216.46]:55134 "EHLO mail-qw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756829Ab0KNTLI convert rfc822-to-8bit (ORCPT ); Sun, 14 Nov 2010 14:11:08 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=ndW0X2QXKo82CKXu004wV1gpK9dAR8p4JMmtsrkCj/cndtAt3Ks1GPUGKm43TUO5wg VodOA7kHEe71XmSjn2kaWTcsba01tFIoqkzJw7Yyd9ENvKIX8Ys+WeMkDY0IpvwvqkSJ o90f8OYczxSjhXsWmLf4HMtAel0Nw/phvMVPE= MIME-Version: 1.0 In-Reply-To: <1289756161.2814.29.camel@zim> References: <20101104041236.GA9389@helgaas.com> <1288937844.3234.1.camel@edumazet-laptop> <1289500084.2698.12.camel@zim> <201011121152.30204.knikanth@suse.de> <1289756161.2814.29.camel@zim> Date: Sun, 14 Nov 2010 11:11:05 -0800 X-Google-Sender-Auth: 6ldcOA4Oy0CXAewy-SxPNIfxtE8 Message-ID: Subject: Re: divide error in select_task_rq_fair() From: Yinghai Lu To: Myron Stowe Cc: Nikanth Karthikesan , Eric Dumazet , Bjorn Helgaas , Ingo Molnar , Peter Zijlstra , Venkatesh Pallipadi , Nikhil Rao , Takuya Yoshikawa , linux-kernel@vger.kernel.org, rjenties@google.com Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 944 Lines: 24 On Sun, Nov 14, 2010 at 9:36 AM, Myron Stowe wrote: > > I got the same divide error with this latest patch (see attachment). ?If > I revert commit 50f2d7f682f9, the platform boots successfully. please check patch in http://lkml.org/lkml/2010/11/13/181 BTW, you also need to ask your BIOS guys to fix the SRAT table. If you only have 128 cpu entries in MADT, SRAT table should have 128 cpu entries instead of 256 cpu entries otherwise, RHEL 5.5 could have problem. it will throw away last cpu entry in SRAT, (NR_CPUS is 255..., and last entry still could point the right cpu in MADT) Also BIOS should keep cpu entries in SRAT have same order to that in MADT. Thanks Yinghai -- 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/