[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20141105125254.GT3337@twins.programming.kicks-ass.net>
Date: Wed, 5 Nov 2014 13:52:54 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Wanpeng Li <kernellwp@...il.com>
Cc: Wanpeng Li <wanpeng.li@...ux.intel.com>,
Ingo Molnar <mingo@...hat.com>,
Juri Lelli <juri.lelli@....com>,
Kirill Tkhai <ktkhai@...allels.com>,
linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH v2] sched/deadline: support dl task migration during
cpu hotplug
On Wed, Nov 05, 2014 at 06:59:35PM +0800, Wanpeng Li wrote:
> Hi Peter,
> On 14/11/5 下午6:50, Peter Zijlstra wrote:
> >On Wed, Nov 05, 2014 at 04:51:57PM +0800, Wanpeng Li wrote:
> >>Note: dl task can be migrated successfully if rq is offline currently, however,
> >>I'm still not sure why task_rq(task)->rd->span just include the cpu which the dl
> >>task previous running on, so cpu_active_mask is used in the patch.
> >>
> >>Peterz, Juri?
> >So the root domain span is for exclusive cpusets
> >(Documentation/cgroups/cpusets.txt) where we fully separate sets of CPUs
> >and have no load-balancing between the sets.
> >
> >For 'normal' setups the rd->span is the entire machine, but using
> >cpusets you can create more (smaller) domains.
>
> I don't setup any cpusets related stuff, however, task_rq(task)->rd->span
> just include the cpu which the dl task previous running on instead of the
> entire machine in find_later_rq().
Ah, it could be that for offline cpus we have a singleton rd. Lemme try
and remember wth we did there.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists