From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754974AbbJHSJR (ORCPT ); Thu, 8 Oct 2015 14:09:17 -0400 Received: from mx1.redhat.com ([209.132.183.28]:49437 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754039AbbJHSJQ (ORCPT ); Thu, 8 Oct 2015 14:09:16 -0400 Date: Thu, 8 Oct 2015 20:05:57 +0200 From: Oleg Nesterov To: Peter Zijlstra Cc: heiko.carstens@de.ibm.com, linux-kernel@vger.kernel.org, Tejun Heo , Ingo Molnar , Rik van Riel Subject: Re: [RFC][PATCH] sched: Start stopper early Message-ID: <20151008180557.GA28123@redhat.com> References: <20151007084110.GX2881@worktop.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20151007084110.GX2881@worktop.programming.kicks-ass.net> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org To avoid the confusion, let me repeat that I am not arguing with this change, perhaps it makes sense too. But unless I missed something it is not really correct and can't fix the problem. So I still think the series I sent should be applied first. On 10/07, Peter Zijlstra wrote: > static int sched_cpu_active(struct notifier_block *nfb, > unsigned long action, void *hcpu) > { > + int cpu = (long)hcpu; > + > switch (action & ~CPU_TASKS_FROZEN) { > case CPU_STARTING: > set_cpu_rq_start_time(); > return NOTIFY_OK; > case CPU_ONLINE: > + cpu_stopper_unpark(cpu); > /* > * At this point a starting CPU has marked itself as online via > * set_cpu_online(). But it might not yet have marked itself > @@ -5558,7 +5563,7 @@ static int sched_cpu_active(struct notifier_block *nfb, > * Thus, fall-through and help the starting CPU along. > */ > case CPU_DOWN_FAILED: > - set_cpu_active((long)hcpu, true); > + set_cpu_active(cpu, true); > return NOTIFY_OK; > default: > return NOTIFY_DONE; > diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c > index 12484e5..c674371 100644 > --- a/kernel/stop_machine.c > +++ b/kernel/stop_machine.c > @@ -496,6 +496,11 @@ static struct smp_hotplug_thread cpu_stop_threads = { > .selfparking = true, > }; > > +void cpu_stopper_unpark(unsigned int cpu) > +{ > + kthread_unpark(per_cpu(cpu_stopper.thread, cpu)); > +} But note that kthread_unpark() will only wake the stopper thread up. cpu_stopper->enabled is still false, and it will be false until smpboot_unpark_thread() calls ->pre_unpark() later. And this means that stop_two_cpus()->cpu_stop_queue_work() can silently fail until then. So I don't this patch can fix the problem. Oleg.