From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030308AbXDJClz (ORCPT ); Mon, 9 Apr 2007 22:41:55 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1030314AbXDJClz (ORCPT ); Mon, 9 Apr 2007 22:41:55 -0400 Received: from smtp.osdl.org ([65.172.181.24]:40868 "EHLO smtp.osdl.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1030308AbXDJCly (ORCPT ); Mon, 9 Apr 2007 22:41:54 -0400 Date: Mon, 9 Apr 2007 19:41:42 -0700 From: Andrew Morton To: Dave Jones Cc: "Jeff V. Merkey" , Linux kernel , Venki Pallipadi Subject: Re: Preemption Broken: centrino_target busted under SMP on 2.6.20.4 Message-Id: <20070409194142.88f0a5e5.akpm@linux-foundation.org> In-Reply-To: <20070410023108.GA14081@redhat.com> References: <46155CFF.70701@wolfmountaingroup.com> <20070405205034.GB3605@redhat.com> <20070409172651.9bc619d6.akpm@linux-foundation.org> <20070410023108.GA14081@redhat.com> X-Mailer: Sylpheed version 2.2.7 (GTK+ 2.8.17; x86_64-unknown-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 9 Apr 2007 22:31:08 -0400 Dave Jones wrote: > On Mon, Apr 09, 2007 at 05:26:51PM -0700, Andrew Morton wrote: > > On Thu, 5 Apr 2007 16:50:34 -0400 > > Dave Jones wrote: > > > > > diff --git a/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c > > > index f43b987..824d0a2 100644 > > > --- a/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c > > > +++ b/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.c > > > @@ -708,6 +708,7 @@ static int centrino_target (struct cpufreq_policy *policy, > > > saved_mask = current->cpus_allowed; > > > first_cpu = 1; > > > cpus_clear(covered_cpus); > > > + preempt_disable(); > > > for_each_cpu_mask(j, online_policy_cpus) { > > > /* > > > * Support for SMP systems. > > > @@ -798,6 +799,7 @@ static int centrino_target (struct cpufreq_policy *policy, > > > } > > > > > > migrate_end: > > > + preempt_enable(); > > > set_cpus_allowed(current, saved_mask); > > > return 0; > > > } > > > > This means we'll call set_cpus_allowed() while in atomic state, but > > set_cpus_allowed() does sleepy stuff. > > Puzzled. This diff shouldn't change anything about the context we're in > when we call set_cpus_allowed, and as we're not seeing warnings now, > I'm not sure what I'm missing? set_cpus_allowed() will only sleep in special circumstances: when we're telling the target task that it is not allwed to run on a CPU upon which it is presently executing. So it needs to be synchronously migrated off that CPU, which requires that the set_cpus_allowed() caller block. You're probably just not hitting that case. Probably we should have a might_sleep() in set_cpus_allowed(), although there might be callers who are guaranteeed to never hit that case and who might legitimately want special treatment to avoid the warning. > [which may be 'the obvious', you wouldn't believe the evening I've had > involving gas leaks and noxious fumes. Wheee, floaty head.] Yeah, I get a lot of patches like that.