From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261598AbUKSVsU (ORCPT ); Fri, 19 Nov 2004 16:48:20 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261600AbUKSVsU (ORCPT ); Fri, 19 Nov 2004 16:48:20 -0500 Received: from gw.goop.org ([64.81.55.164]:29585 "EHLO mail.goop.org") by vger.kernel.org with ESMTP id S261598AbUKSVqc (ORCPT ); Fri, 19 Nov 2004 16:46:32 -0500 Subject: Re: 2.6.10-rc2-mm2: spinlock problem in arch/i386/kernel/time.c on resume From: Jeremy Fitzhardinge To: Andrew Morton Cc: linux-kernel In-Reply-To: <20041119025357.4b6ca9f7.akpm@osdl.org> References: <1100855456.22588.9.camel@outer-dhcp-100.goop.org> <20041119025357.4b6ca9f7.akpm@osdl.org> Content-Type: text/plain Date: Fri, 19 Nov 2004 13:14:31 -0800 Message-Id: <1100898871.5056.10.camel@localhost> Mime-Version: 1.0 X-Mailer: Evolution 2.0.2 (2.0.2-0.mozer.2) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2004-11-19 at 02:53 -0800, Andrew Morton wrote: > I can't imagine how that could happen. I wonder if the spinlock debugging > could be making a mistake. When I disabled DEBUG_SPINLOCK, it works fine. Does DEBUG_SPINLOCK have any side effects? I would have thought it wouldn't do anything on a UP, non-PREEMPT system. There is the comment in arch/i386/kernel/time.c saying that APM needs get_cmos_time(), which is the site of the second message. The code is very simple, so its hard to see how this could be going wrong. I'm guessing, though, that this is a relatively untested path, since people don't tend to use APM+SMP. > > .config & lspci attached > > I have but a lowly A21P. With my .config it APM resumes happily. With > yours it refuses to even suspend. And: > > > > hdc: DMA disabled > e100: Intel(R) PRO/100 Network Driver, 3.2.3-k2-NAPI > e100: Copyright(c) 1999-2004 Intel Corporation I've been using the eepro100 driver lately, which doesn't seem to have this problem at all. J