From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932524AbcELTuH (ORCPT ); Thu, 12 May 2016 15:50:07 -0400 Received: from mail-wm0-f44.google.com ([74.125.82.44]:35202 "EHLO mail-wm0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932499AbcELTuE (ORCPT ); Thu, 12 May 2016 15:50:04 -0400 From: Matt Fleming To: Peter Zijlstra , Ingo Molnar Cc: linux-kernel@vger.kernel.org, Byungchul Park , Frederic Weisbecker , Luca Abeni , "Rafael J . Wysocki" , Rik van Riel , Thomas Gleixner , Wanpeng Li , Yuyang Du , Mel Gorman , Mike Galbraith , Matt Fleming Subject: [RFC][PATCH 4/5] sched/fair: Push rq lock pin/unpin into idle_balance() Date: Thu, 12 May 2016 20:49:52 +0100 Message-Id: <1463082593-27777-5-git-send-email-matt@codeblueprint.co.uk> X-Mailer: git-send-email 2.7.3 In-Reply-To: <1463082593-27777-1-git-send-email-matt@codeblueprint.co.uk> References: <1463082593-27777-1-git-send-email-matt@codeblueprint.co.uk> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Future patches will emit warnings if rq_clock() is called before update_rq_clock() inside a rq_pin_lock()/rq_unpin_lock() pair. Since there is only one caller of idle_balance() we can push the unpin/repin there. Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Mike Galbraith Cc: Mel Gorman Cc: Yuyang Du Cc: Byungchul Park Cc: Rik van Riel Cc: Frederic Weisbecker Signed-off-by: Matt Fleming --- kernel/sched/fair.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7f776a99bde0..217e3a9d78db 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3095,7 +3095,7 @@ static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq) return cfs_rq->avg.load_avg; } -static int idle_balance(struct rq *this_rq); +static int idle_balance(struct rq *this_rq, struct rq_flags *rf); #else /* CONFIG_SMP */ @@ -3118,7 +3118,7 @@ attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {} static inline void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {} -static inline int idle_balance(struct rq *rq) +static inline int idle_balance(struct rq *rq, struct rq_flags *rf) { return 0; } @@ -5701,15 +5701,8 @@ simple: return p; idle: - /* - * This is OK, because current is on_cpu, which avoids it being picked - * for load-balance and preemption/IRQs are still disabled avoiding - * further scheduler activity on it and we're being very careful to - * re-start the picking loop. - */ - rq_unpin_lock(rq, rf); - new_tasks = idle_balance(rq); - rq_repin_lock(rq, rf); + new_tasks = idle_balance(rq, rf); + /* * Because idle_balance() releases (and re-acquires) rq->lock, it is * possible for any higher priority task to appear. In that case we @@ -7641,7 +7634,7 @@ update_next_balance(struct sched_domain *sd, int cpu_busy, unsigned long *next_b * idle_balance is called by schedule() if this_cpu is about to become * idle. Attempts to pull tasks from other CPUs. */ -static int idle_balance(struct rq *this_rq) +static int idle_balance(struct rq *this_rq, struct rq_flags *rf) { unsigned long next_balance = jiffies + HZ; int this_cpu = this_rq->cpu; @@ -7655,6 +7648,14 @@ static int idle_balance(struct rq *this_rq) */ this_rq->idle_stamp = rq_clock(this_rq); + /* + * This is OK, because current is on_cpu, which avoids it being picked + * for load-balance and preemption/IRQs are still disabled avoiding + * further scheduler activity on it and we're being very careful to + * re-start the picking loop. + */ + rq_unpin_lock(this_rq, rf); + if (this_rq->avg_idle < sysctl_sched_migration_cost || !this_rq->rd->overload) { rcu_read_lock(); @@ -7732,6 +7733,8 @@ out: if (pulled_task) this_rq->idle_stamp = 0; + rq_repin_lock(this_rq, rf); + return pulled_task; } -- 2.7.3