Messages in this thread | | | Date | Fri, 15 Oct 2010 08:30:32 +0900 | From | KAMEZAWA Hiroyuki <> | Subject | Re: [PATCH v3 3/7] sched: throttle cfs_rq entities which exceed their local quota |
| |
On Thu, 14 Oct 2010 12:41:44 +0200 Peter Zijlstra <peterz@infradead.org> wrote:
> On Thu, 2010-10-14 at 03:25 -0700, Paul Turner wrote: > > > "When vcpu holds spinlock in kernel, please don't sleep.." will be nice. > > > Is there patch already ? > > Oh you're talking about the guest spinlock crap?
yes ;)
> I don't particularly care about that.. that's a virt problem you see ;-)
Hmm. AFAIK, all qemu's thread will be under a cgroup. So, I'll be happy if a vcpu is on a loop to wait for spinlock yields immediately...But I'm not famliar with this area, either. Thanks.
-Kame
| |