Re: [tip:sched/urgent] sched: Fix cross-cpu clock sync on remotewakeups

[Peter Zijlstra]

On Tue, 2011-06-07 at 15:12 +0200, Borislav Petkov wrote:
> On Thu, Jun 02, 2011 at 11:48:31AM -0400, Peter Zijlstra wrote:
> > Crap.. you're right. And I bet other archs don't do that either. With
> > NO_HZ you really need irq_enter() for pretty much all interrupts so I
> > was assuming the resched IPI had it, but its been special and never
> > really needed it. If it would wake an idle cpu the idle loop exit would
> > deal with it, if it interrupted userspace the thing was running and
> > NO_HZ wasn't relevant.
> > 
> > Damn. 
> > 
> > And yes, the only reason I didn't see this on my dev box was because we
> > do indeed set that sched_clock_stable thing on wsm. And I never noticed
> > on my desktop because firefox/X/etc. consuming heaps of CPU isn't weird
> > at all.
> > 
> > Adding it to all resched int handlers is of course a possibility but
> > would slow down the thing, although with the new code, most users are
> > now indeed wakeups (excepting weird and wonderful users like KVM).
> 
> FWIW, we could set the sched_clock_stable on AMD too - at least on F10h
> and later. This will take care of the problem at hand and defer the
> issue of slowing down the resched ipi handlers.
> 
> I dunno, however, whether we still would need the proper ->tick_gtod
> update for correct ttwu accounting regardless of sched_clock_stable on
> K8 (unstable TSCs) and maybe even other arches.

Yeah, I think I'm going to commit this extra irq_enter() thing for now,
and then slowly go through all the arches again and remove this one here
once all archs are sorted.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/