From mboxrd@z Thu Jan 1 00:00:00 1970 From: Juri Lelli Subject: Re: [PATCH 2/3] sched/deadline: fix bandwidth check/update when migrating tasks between exclusive cpusets Date: Fri, 19 Sep 2014 13:46:22 +0100 Message-ID: <541C259E.4010207@arm.com> References: <1411118561-26323-1-git-send-email-juri.lelli@arm.com> <1411118561-26323-3-git-send-email-juri.lelli@arm.com> <541C17D6.5020608@bmw-carit.de> Mime-Version: 1.0 Content-Transfer-Encoding: 8BIT Return-path: In-Reply-To: <541C17D6.5020608-98C5kh4wR6ohFhg+JK9F0w@public.gmane.org> Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Daniel Wagner , "peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org" Cc: "mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org" , "juri.lelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org" , "raistlin-k2GhghHVRtY@public.gmane.org" , "michael-dyjBcgdgk7Pe9wHmmfpqLFaTQe2KTcn/@public.gmane.org" , "fchecconi-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org" , "vincent-9z8vmPu0pS/iB9QmIjCX8w@public.gmane.org" , "luca.abeni-3IIOeSMMxS4@public.gmane.org" , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , Li Zefan , "cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" Hi Daniel, On 19/09/14 12:47, Daniel Wagner wrote: > Hi, > > On 09/19/2014 11:22 AM, Juri Lelli wrote: >> Exclusive cpusets are the only way users can restrict SCHED_DEADLINE tasks >> affinity (performing what is commonly called clustered scheduling). >> Unfortunately, such thing is currently broken for two reasons: >> >> - No check is performed when the user tries to attach a task to >> an exlusive cpuset (recall that exclusive cpusets have an >> associated maximum allowed bandwidth). >> >> - Bandwidths of source and destination cpusets are not correctly >> updated after a task is migrated between them. >> >> This patch fixes both things at once, as they are opposite faces >> of the same coin. >> >> The check is performed in cpuset_can_attach(), as there aren't any >> points of failure after that function. The updated is split in two >> halves. We first reserve bandwidth in the destination cpuset, after >> we pass the check in cpuset_can_attach(). And we then release >> bandwidth from the source cpuset when the task's affinity is >> actually changed. Even if there can be time windows when sched_setattr() >> may erroneously fail in the source cpuset, we are fine with it, as >> we can't perfom an atomic update of both cpusets at once. >> >> Reported-by: Daniel Wagner > > Ack > > I have no special test for this, I just let my test running which was > fixed by patch #1. Works fine though. I'll plan to write some test for this. > Ok, thanks. Just mind that the problem fixed by patch 3/3 may sometime affect this too. I should have definitely put 3/3 on top of the patchset :/. Thanks, - Juri