Re: [BUG] perf: perf sched warning possibly due to clockgranularity on AMD

From: Arnaldo Carvalho de Melo
Date: Mon Feb 06 2012 - 12:17:41 EST

Em Mon, Feb 06, 2012 at 03:31:00PM +0100, Stephane Eranian escreveu:
> On Mon, Feb 6, 2012 at 3:26 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> > On Mon, 2012-02-06 at 14:25 +0100, Stephane Eranian wrote:
> >> In summary, two issues:
> >>   - Why is sched_clock_stable not set or even tested on recent AMD systems?
> >
> > I'll leave that for Robert to answer, also added Boris to the CC, him
> > know AMD bits too.
> >
> Ok, good.
> >>   - perf should not rely on fine granularity timestamps in the re-ordering code
> >
> > Uhm, what else is there?

> I understand your point. I don't quite understand what the code is
> supposed to do for odered
> samples. I'd like Arnaldo to explain this warning first.

That code was written by Frédéric and IIRC also Thomas, its something I
plan to rework somehow in the future, but for now I'll let them chime
in... Frédéric?

- Arnaldo
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at