From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S936015Ab0COJMe (ORCPT ); Mon, 15 Mar 2010 05:12:34 -0400 Received: from mx1.redhat.com ([209.132.183.28]:14733 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935985Ab0COJM3 (ORCPT ); Mon, 15 Mar 2010 05:12:29 -0400 Date: Mon, 15 Mar 2010 10:10:19 +0100 From: Oleg Nesterov To: Peter Zijlstra , Ingo Molnar Cc: Ben Blum , Jiri Slaby , Lai Jiangshan , Li Zefan , Miao Xie , Paul Menage , "Rafael J. Wysocki" , Tejun Heo , linux-kernel@vger.kernel.org Subject: [PATCH 4/6] sched_exec: remove select_fallback_rq() logic Message-ID: <20100315091019.GA9141@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline 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 sched_exec()->select_task_rq() reads/updates ->cpus_allowed lockless. This can race with other CPUs updating our ->cpus_allowed, and this looks meaningless to me. The task is current and running, it must have online cpus in ->cpus_allowed, the fallback mode is bogus. And, if ->sched_class returns the "wrong" cpu, this likely means we raced with set_cpus_allowed() which was called for reason, why should sched_exec() retry and call ->select_task_rq() again? Change the code to call sched_class->select_task_rq() directly and do nothing if the returned cpu is wrong after re-checking under rq->lock. >>From now task_struct->cpus_allowed is always stable under TASK_WAKING, select_fallback_rq() is always called under rq-lock or the caller or the caller owns TASK_WAKING (select_task_rq). Signed-off-by: Oleg Nesterov --- kernel/sched.c | 25 ++++++++----------------- 1 file changed, 8 insertions(+), 17 deletions(-) --- 34-rc1/kernel/sched.c~3_SCHED_EXEC_DONT_FALLBACK 2010-03-15 09:40:44.000000000 +0100 +++ 34-rc1/kernel/sched.c 2010-03-15 09:41:28.000000000 +0100 @@ -2272,6 +2272,9 @@ void task_oncpu_function_call(struct tas } #ifdef CONFIG_SMP +/* + * ->cpus_allowed is protected by either TASK_WAKING or rq->lock held. + */ static int select_fallback_rq(int cpu, struct task_struct *p) { int dest_cpu; @@ -2308,12 +2311,7 @@ static int select_fallback_rq(int cpu, s } /* - * Gets called from 3 sites (exec, fork, wakeup), since it is called without - * holding rq->lock we need to ensure ->cpus_allowed is stable, this is done - * by: - * - * exec: is unstable, retry loop - * fork & wake-up: serialize ->cpus_allowed against TASK_WAKING + * The caller (fork, wakeup) owns TASK_WAKING, ->cpus_allowed is stable. */ static inline int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) @@ -3123,9 +3121,8 @@ void sched_exec(void) unsigned long flags; struct rq *rq; -again: this_cpu = get_cpu(); - dest_cpu = select_task_rq(p, SD_BALANCE_EXEC, 0); + dest_cpu = p->sched_class->select_task_rq(p, SD_BALANCE_EXEC, 0); if (dest_cpu == this_cpu) { put_cpu(); return; @@ -3133,18 +3130,12 @@ again: rq = task_rq_lock(p, &flags); put_cpu(); - /* * select_task_rq() can race against ->cpus_allowed */ - if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed) - || unlikely(!cpu_active(dest_cpu))) { - task_rq_unlock(rq, &flags); - goto again; - } - - /* force the process onto the specified CPU */ - if (migrate_task(p, dest_cpu, &req)) { + if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed) && + likely(cpu_active(dest_cpu)) && + migrate_task(p, dest_cpu, &req)) { /* Need to wait for migration thread (might exit: take ref). */ struct task_struct *mt = rq->migration_thread;