Re: [RFC PATCHv3] perf tools: add event grouping capability to "perf stat"

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Thu, 2010-11-25 at 15:07 +0100, Stephane Eranian wrote:
> On Thu, Nov 25, 2010 at 3:02 PM, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> > On Thu, 2010-11-25 at 14:19 +0100, Stephane Eranian wrote:
> >
> >> I think using PERF_SAMPLE_READ may expose a problem in the
> >> perf.data format. To correctly parse a sample created with SAMPLE_READ,
> >> you need to know the attr.read_format. But for that you need to know the
> >> event which caused the sample, but for that you need the SAMPLE_ID,
> >> and you don't know if it's there or not. In other words, there is a chicken
> >> and egg problem.
> >>
> >> I think the issue is that PERF_RECORD_SAMPLE is missing a mandatory
> >> piece of information: overflow event ID. This must a mandatory field, not
> >> optional as it is today. It is okay when you have only one group, but we'd
> >> like to go beyond that.
> >
> >
> >
> > I'm not sure I get it, there's both PERF_FORMAT_ID and PERF_SAMPLE_ID,
> > so afaict there's a working combination for what you want to do.
> >
> Ok, I had forgotten about PERF_SAMPLE_ID. But that means that if the
> number of groups > 1, then if you use PERF_SAMPLE_READ, you MUST
> also use PERF_SAMPLE_ID. Otherwise you cannot get back to the event
> that generated the sample and thus attr.read_format.

perf-record already does that:

        if (nr_counters > 1)
                attr->sample_type |= PERF_SAMPLE_ID;