From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753946AbdDLOld (ORCPT ); Wed, 12 Apr 2017 10:41:33 -0400 Received: from mail.kernel.org ([198.145.29.136]:38718 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752412AbdDLOla (ORCPT ); Wed, 12 Apr 2017 10:41:30 -0400 Date: Wed, 12 Apr 2017 10:41:24 -0400 From: Steven Rostedt To: Peter Zijlstra Cc: LKML , Ingo Molnar , "Paul E. McKenney" Subject: Re: [RFC][PATCH] sched: Have do_idle() call __schedule() without enabling preemption Message-ID: <20170412104124.4639c8ff@gandalf.local.home> In-Reply-To: <20170411224413.72d9344a@grimm.local.home> References: <20170411224413.72d9344a@grimm.local.home> X-Mailer: Claws Mail 3.14.0 (GTK+ 2.24.31; x86_64-pc-linux-gnu) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org [ tl;dr; version ] Peter, In order to have synchronize_rcu_tasks() work, the idle task can never be preempted. There's a very small window in schedule_preempt_disable() that enables preemption, and when this happens, it breaks synchronize_rcu_tasks() (see above email for details). Is there any reason to enable preemption, or can we simply have idle call into schedule without ever allowing it to be preempted, as in my patch? Note, it is almost good enough to just change schedule_preempt_disable() to do the exact same thing, but there's one instance in kernel/locking/mutex.c that calls it in a non running state. Signed-off-by: Steven Rostedt (VMware) --- diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 3b31fc0..cb9ceda 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3502,6 +3502,13 @@ asmlinkage __visible void __sched schedule(void) } EXPORT_SYMBOL(schedule); +void __sched schedule_idle(void) +{ + do { + __schedule(false); + } while (need_resched()); +} + #ifdef CONFIG_CONTEXT_TRACKING asmlinkage __visible void __sched schedule_user(void) { diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index ac6d517..229c17e 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -264,7 +264,7 @@ static void do_idle(void) smp_mb__after_atomic(); sched_ttwu_pending(); - schedule_preempt_disabled(); + schedule_idle(); } bool cpu_in_idle(unsigned long pc) diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 5cbf922..c5ee02b 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1465,6 +1465,8 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq) } #endif +extern void schedule_idle(void); + extern void sysrq_sched_debug_show(void); extern void sched_init_granularity(void); extern void update_max_interval(void);