From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754960AbbDIDaD (ORCPT ); Wed, 8 Apr 2015 23:30:03 -0400 Received: from m15-112.126.com ([220.181.15.112]:40003 "EHLO m15-112.126.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754530AbbDID3p (ORCPT ); Wed, 8 Apr 2015 23:29:45 -0400 From: Xunlei Pang To: linux-kernel@vger.kernel.org Cc: Peter Zijlstra , Steven Rostedt , Juri Lelli , Dan Streetman , Xunlei Pang Subject: [PATCH v5 2/3] sched/rt: Fix wrong SMP scheduler behavior for equal prio cases Date: Thu, 9 Apr 2015 11:27:17 +0800 Message-Id: <1428550038-13619-2-git-send-email-xlpang@126.com> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1428550038-13619-1-git-send-email-xlpang@126.com> References: <1428550038-13619-1-git-send-email-xlpang@126.com> X-CM-TRANSID: DMmowEDJvlOa8SVVTV0aAQ--.926S3 X-Coremail-Antispam: 1Uf129KBjvJXoWxGFWrZr4kur48Gw4xXF13Jwb_yoW5Kr13pa 4rJw1jya1kJa9Fgrn7Xr48Zr43G3sYq3y5Jrn5t3yrKa15tFW8XFnYyryftFWSyr1vqa13 tFWDtrsxGr1UZFUanT9S1TB71UUUUUUqnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDUYxBIdaVFxhVjvjDU0xZFpf9x07jRXdbUUUUU= X-Originating-IP: [210.21.223.3] X-CM-SenderInfo: p0ost0bj6rjloofrz/1tbiJwHWv01sBHgn-QAAsI Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Xunlei Pang Currently, SMP RT scheduler has some trouble in dealing with equal prio cases. For example, in check_preempt_equal_prio(): When RT1(current task) gets preempted by RT2, if there is a migratable RT3 with same prio, RT3 will be pushed away instead of RT1 afterwards, because RT1 will be enqueued to the tail of the pushable list when going through succeeding put_prev_task_rt() triggered by resched. This broke FIFO. Furthermore, this is also problematic for normal preempted cases if there're some rt tasks queued with the same prio as current. Because current will be put behind these tasks in the pushable queue. So, if a task is running and gets preempted by a higher priority task (or even with same priority for migrating), this patch ensures that it is put ahead of any existing task with the same priority in the pushable queue. Suggested-by: Steven Rostedt Signed-off-by: Xunlei Pang --- kernel/sched/rt.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 575da76..402162a 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -359,11 +359,15 @@ static inline void set_post_schedule(struct rq *rq) rq->post_schedule = has_pushable_tasks(rq); } -static void enqueue_pushable_task(struct rq *rq, struct task_struct *p) +static void +enqueue_pushable_task(struct rq *rq, struct task_struct *p, bool head) { plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); plist_node_init(&p->pushable_tasks, p->prio); - plist_add(&p->pushable_tasks, &rq->rt.pushable_tasks); + if (head) + plist_add_head(&p->pushable_tasks, &rq->rt.pushable_tasks); + else + plist_add_tail(&p->pushable_tasks, &rq->rt.pushable_tasks); /* Update the highest prio pushable task */ if (p->prio < rq->rt.highest_prio.next) @@ -385,7 +389,8 @@ static void dequeue_pushable_task(struct rq *rq, struct task_struct *p) #else -static inline void enqueue_pushable_task(struct rq *rq, struct task_struct *p) +static inline +void enqueue_pushable_task(struct rq *rq, struct task_struct *p, bool head) { } @@ -1260,7 +1265,7 @@ enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags) enqueue_rt_entity(rt_se, flags & ENQUEUE_HEAD); if (!task_current(rq, p) && p->nr_cpus_allowed > 1) - enqueue_pushable_task(rq, p); + enqueue_pushable_task(rq, p, false); } static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags) @@ -1507,7 +1512,16 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) * if it is still active */ if (on_rt_rq(&p->rt) && p->nr_cpus_allowed > 1) - enqueue_pushable_task(rq, p); + /* + * put_prev_task_rt() is called by many functions, + * pick_next_task_rt() is the only one may have + * PREEMPT_ACTIVE set. So if detecting p(current + * task) is preempted in such case, we should + * enqueue it to the front of the pushable plist, + * as there may be multiple tasks with the same + * priority as p. + */ + enqueue_pushable_task(rq, p, !!(preempt_count() & PREEMPT_ACTIVE)); } #ifdef CONFIG_SMP @@ -2091,7 +2105,7 @@ static void set_cpus_allowed_rt(struct task_struct *p, rq->rt.rt_nr_migratory--; } else { if (!task_current(rq, p)) - enqueue_pushable_task(rq, p); + enqueue_pushable_task(rq, p, false); rq->rt.rt_nr_migratory++; } -- 1.9.1