From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161426AbWG2DBH (ORCPT ); Fri, 28 Jul 2006 23:01:07 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1161428AbWG2DBH (ORCPT ); Fri, 28 Jul 2006 23:01:07 -0400 Received: from ebiederm.dsl.xmission.com ([166.70.28.69]:35788 "EHLO ebiederm.dsl.xmission.com") by vger.kernel.org with ESMTP id S1161426AbWG2DBG (ORCPT ); Fri, 28 Jul 2006 23:01:06 -0400 From: ebiederm@xmission.com (Eric W. Biederman) To: Andrew Morton Cc: Paul Fulghum , ak@muc.de, linux-kernel@vger.kernel.org, mingo@elte.hu Subject: Re: 2.6.18-rc2-mm1 timer int 0 doesn't work References: <20060727015639.9c89db57.akpm@osdl.org> <1154112276.3530.3.camel@amdx2.microgate.com> <20060728144854.44c4f557.akpm@osdl.org> <20060728233851.GA35643@muc.de> <1154132126.3349.8.camel@localhost.localdomain> <1154135792.2557.7.camel@localhost.localdomain> <20060728182450.8f5cbf76.akpm@osdl.org> Date: Fri, 28 Jul 2006 20:58:33 -0600 In-Reply-To: <20060728182450.8f5cbf76.akpm@osdl.org> (Andrew Morton's message of "Fri, 28 Jul 2006 18:24:50 -0700") Message-ID: User-Agent: Gnus/5.110004 (No Gnus v0.4) Emacs/21.4 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Andrew Morton writes: > On Fri, 28 Jul 2006 20:16:32 -0500 > Paul Fulghum wrote: > >> On Fri, 2006-07-28 at 19:15 -0500, Paul Fulghum wrote: >> > I'm doing a build on my home machine now to see if it >> > happens there also. >> >> Well, the timer int 0 problem does not happen on my home machine. >> However, it still crashes in early boot for a different reason. >> >> 2.6.18-rc2 works fine with same config. >> >> In this case the error is: >> >> No per-cpu room for modules > > yeah, sorry, that's a known problem which nobody appears to be doing > anything about. The expansion of NR_IRQS gobbles all the percpu memory in > the kstat structure. Sorry I didn't realize it was so easy to trip over. It's on my todo list for sometime in the next couple of days. > I assume you have a large NR_CPUS? Decreasing it should help. Eric