From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751300AbcGNRyb (ORCPT ); Thu, 14 Jul 2016 13:54:31 -0400 Received: from mail-pa0-f44.google.com ([209.85.220.44]:32930 "EHLO mail-pa0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751054AbcGNRya (ORCPT ); Thu, 14 Jul 2016 13:54:30 -0400 From: bsegall@google.com To: Wanpeng Li Cc: Konstantin Khlebnikov , xlpang@redhat.com, Peter Zijlstra , Ingo Molnar , "linux-kernel\@vger.kernel.org" , stable@vger.kernel.org Subject: Re: [PATCH] sched/fair: do not announce throttled next buddy in dequeue_task_fair References: <146608183552.21905.15924473394414832071.stgit@buzz> <57835735.6020906@redhat.com> <57835BFD.90201@redhat.com> <57838D28.4090003@redhat.com> <5783907A.6030609@yandex-team.ru> Date: Thu, 14 Jul 2016 10:54:26 -0700 In-Reply-To: (Wanpeng Li's message of "Thu, 14 Jul 2016 20:11:01 +0800") Message-ID: User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Wanpeng Li writes: > 2016-07-14 1:06 GMT+08:00 : >> Wanpeng Li writes: >> >>> 2016-07-13 1:25 GMT+08:00 : >>>> Konstantin Khlebnikov writes: >>>> >>>>> On 11.07.2016 15:12, Xunlei Pang wrote: >>>>>> On 2016/07/11 at 17:54, Wanpeng Li wrote: >>>>>>> Hi Konstantin, Xunlei, >>>>>>> 2016-07-11 16:42 GMT+08:00 Xunlei Pang : >>>>>>>> On 2016/07/11 at 16:22, Xunlei Pang wrote: >>>>>>>>> On 2016/07/11 at 15:25, Wanpeng Li wrote: >>>>>>>>>> 2016-06-16 20:57 GMT+08:00 Konstantin Khlebnikov : >>>>>>>>>>> Hierarchy could be already throttled at this point. Throttled next >>>>>>>>>>> buddy could trigger null pointer dereference in pick_next_task_fair(). >>>>>>>>>> There is cfs_rq->next check in pick_next_entity(), so how can null >>>>>>>>>> pointer dereference happen? >>>>>>>>> I guess it's the following code leading to a NULL se returned: >>>>>>>> s/NULL/empty-entity cfs_rq se/ >>>>>>>> >>>>>>>>> pick_next_entity(): >>>>>>>>> if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1) >>>>>>> ^^^^^^^^^^^^^ >>>>>>> I think this will return false. >>>>>> >>>>>> With the wrong throttled_hierarchy(), I think this can happen. But after we have the >>>>>> corrected throttled_hierarchy() patch, I can't see how it is possible. >>>>>> >>>>>> dequeue_task_fair(): >>>>>> if (task_sleep && parent_entity(se)) >>>>>> set_next_buddy(parent_entity(se)); >>>>>> >>>>>> How does dequeue_task_fair() with DEQUEUE_SLEEP set(true task_sleep) happen to a throttled hierarchy? >>>>>> IOW, a task belongs to a throttled hierarchy is running? >>>>>> >>>>>> Maybe Konstantin knows the reason. >>>>> >>>>> This function (dequeue_task_fair) check throttling but at point it could skip several >>>>> levels and announce as next buddy actually throttled entry. >>>>> Probably this bug hadn't happened but this's really hard to prove that this is impossible. >>>>> ->set_curr_task(), PI-boost or some tricky migration in balancer could break this easily. >>>> >>>> sched_setscheduler can call put_prev_task, which then can cause a >>>> throttle outside of __schedule(), then the task blocks normally and >>>> deactivate_task(DEQUEUE_SLEEP) happens and you lose. >>> >>> The cfs_rq_throttled() check in dequeue_task_fair() will capture the >>> cfs_rq which is throttled in sched_setscheduler::put_prev_task path, >>> so nothing lost, where I miss? >>> >>> Regards, >>> Wanpeng Li >> >> The cfs_rq_throttled() checks there are done bottom-up, so they will >> trigger too late. a/b/t, where t is descheduling and a is throttled can >> still cause a set_next_buddy(b); > > throttle cfs_rq is up-bottom, so when a is throttled, b and c are not > yet, then task_sleep && se && !throttled_hierarchy(cfs_rq) still can't > prevent a set_next_buddy(b). > > Regards, > Wanpeng Li They don't race or anything, everything's under rq->lock. throttled_hierarchy will register properly, the issue is that a parent is the one cfs_rq_throttled(), not the current cfs_rq, and set_next_buddy will set cfs_rq->next to an se that is !on_rq. In the other order (set_next_buddy then throttle), throttle_cfs_rq will call dequeue which will clear the problematic buddy.