From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934150AbYDQSCa (ORCPT ); Thu, 17 Apr 2008 14:02:30 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1762704AbYDQSCV (ORCPT ); Thu, 17 Apr 2008 14:02:21 -0400 Received: from wr-out-0506.google.com ([64.233.184.237]:15228 "EHLO wr-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1762584AbYDQSCV (ORCPT ); Thu, 17 Apr 2008 14:02:21 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=DhrM0hi5HS095n/umdUHAeeQ9b945gSHEvlczcJUxAaPWLtksXmp06PFCgk7SfPy9C+Mt7jF99h0Vrh00FlwbEDCuxb1I1om2bzHinmbmTb0d7T2Y7pFJ+JJK/2+c31ln0RkZzs3JmKB9LcrEz65GeOtOyFdKgttqwAffPbZWfE= Message-ID: <86802c440804171102q2be96c67m881394c1e6fa3867@mail.gmail.com> Date: Thu, 17 Apr 2008 11:02:13 -0700 From: "Yinghai Lu" To: "Ingo Molnar" , "Mike Travis" Subject: Re: [PATCH] - Increase MAX_APICS for large configs Cc: "Jack Steiner" , tglx@linutronix.de, linux-mm@kvack.org, linux-kernel@vger.kernel.org, "H. Peter Anvin" In-Reply-To: <20080417110727.GA942@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <20080416163936.GA23099@sgi.com> <20080417110727.GA942@elte.hu> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Apr 17, 2008 at 4:07 AM, Ingo Molnar wrote: > > * Jack Steiner wrote: > > > Increase the maximum number of apics when running very large > > configurations. This patch has no affect on most systems. > > x86.git overnight random-qa testing found a boot crash and i bisected it > down to this patch. The config is: > > http://redhat.com/~mingo/misc/config-Thu_Apr_17_10_17_14_CEST_2008.bad > > the failure is attached below. (I needed the exact boot parameters > listed in that bootup log to see this failure.) > > it seems to be CONFIG_MAXSMP=y triggers the new more-apic-ids code and > that causes some breakage elsewhere. [btw., this again shows how useful > the CONFIG_MAXSMP debug feature is!] so this one and Mike's 4096 NR_CPUS cause the problem? 2048 CPUs works. YH