On Tue, Oct 07, 2014 at 09:59:54AM +0100, Juri Lelli wrote: > Hi Peter, > > On 19/09/14 22:25, Peter Zijlstra wrote: > > On Fri, Sep 19, 2014 at 10:22:40AM +0100, 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. > > > > The thing I cannot find is if we correctly deal with updates to the > > cpuset. Say we first setup 2 (exclusive) sets A:cpu0 B:cpu1-3. Then > > assign tasks and then update the cpu masks like: B:cpu2,3, A:cpu1,2. > > > > So, what follows should address the problem you describe. > > Assuming you intended that we try to update masks as A:cpu0,3 and > B:cpu1,2, with what below we are able to check that removing cpu3 > from B doesn't break guarantees. After that cpu3 can be put in A. > > Does it make any sense? Yeah, I think that about covers is. Could you write a changelog with it? The reason I hadn't applied your patch #2 yet is because I thought it triggered the splat reported in this thread. But later emails seem to suggest this is a separate/pre-existing issue? -- To unsubscribe from this list: send the line "unsubscribe cgroups" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
