Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754732AbdDJPpp (ORCPT ); Mon, 10 Apr 2017 11:45:45 -0400 Received: from mx2.suse.de ([195.135.220.15]:38091 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754636AbdDJPer (ORCPT ); Mon, 10 Apr 2017 11:34:47 -0400 X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "References" From: Jiri Slaby To: stable@vger.kernel.org Cc: linux-kernel@vger.kernel.org, Sebastian Andrzej Siewior , Peter Zijlstra , Linus Torvalds , Mike Galbraith , Thomas Gleixner , Ingo Molnar , Jiri Slaby Subject: [PATCH 3.12 109/142] sched/rt: Add a missing rescheduling point Date: Mon, 10 Apr 2017 17:33:10 +0200 Message-Id: <258b4e6746cad9550f9fbcc0036fc118113aa425.1491838390.git.jslaby@suse.cz> X-Mailer: git-send-email 2.12.2 In-Reply-To: References: In-Reply-To: References: Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2326 Lines: 63 From: Sebastian Andrzej Siewior 3.12-stable review patch. If anyone has any objections, please let me know. =============== commit 619bd4a71874a8fd78eb6ccf9f272c5e98bcc7b7 upstream. Since the change in commit: fd7a4bed1835 ("sched, rt: Convert switched_{from, to}_rt() / prio_changed_rt() to balance callbacks") ... we don't reschedule a task under certain circumstances: Lets say task-A, SCHED_OTHER, is running on CPU0 (and it may run only on CPU0) and holds a PI lock. This task is removed from the CPU because it used up its time slice and another SCHED_OTHER task is running. Task-B on CPU1 runs at RT priority and asks for the lock owned by task-A. This results in a priority boost for task-A. Task-B goes to sleep until the lock has been made available. Task-A is already runnable (but not active), so it receives no wake up. The reality now is that task-A gets on the CPU once the scheduler decides to remove the current task despite the fact that a high priority task is enqueued and waiting. This may take a long time. The desired behaviour is that CPU0 immediately reschedules after the priority boost which made task-A the task with the lowest priority. [js] no deadline in 3.12 yet Suggested-by: Peter Zijlstra Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Thomas Gleixner Fixes: fd7a4bed1835 ("sched, rt: Convert switched_{from, to}_rt() prio_changed_rt() to balance callbacks") Link: http://lkml.kernel.org/r/20170124144006.29821-1-bigeasy@linutronix.de Signed-off-by: Ingo Molnar Signed-off-by: Jiri Slaby --- kernel/sched/rt.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 10edf9d2a8b7..ce4ec3ae9abc 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -1871,10 +1871,9 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p) #ifdef CONFIG_SMP if (rq->rt.overloaded) queue_push_tasks(rq); -#else +#endif /* CONFIG_SMP */ if (p->prio < rq->curr->prio) resched_task(rq->curr); -#endif /* CONFIG_SMP */ } } -- 2.12.2