Re: [PATCH] perf stat: Fix uncore event mixed metric with workload error issue

[Jin, Yao]

Hi Jiri,

On 4/29/2020 4:16 PM, Jiri Olsa wrote:
> On Tue, Apr 28, 2020 at 02:19:22PM -0700, Andi Kleen wrote:
> > > I wonder this would break some expectations.. would it be
> > > more safe to detect duration event and bypass it from the
> > > decission? but maybe the case I'm worried about is not a
> > > problem at all.. Andi?
> >
> > Don't see what it would break.
> >
> > Yes maybe we need to special case duration_time more, but that would
> > be a much bigger patch.
>
> would below change work for you? if duration_time is the only
> case, I'd rather go with the special case for it
>
> jirka

Just tested, for the case of DRAM_BW_Use, it can work.

root@lkp-csl-2sp5 ~# perf stat -M DRAM_BW_Use -- sleep 1

 Performance counter stats for 'system wide':

              9.54 MiB  uncore_imc/cas_count_read/
              3.26 MiB  uncore_imc/cas_count_write/
        1002109793 ns   duration_time

       1.002109793 seconds time elapsed

Thanks
Jin Yao

> ---
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index 9207b6c45475..2518204cffd1 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -1853,6 +1853,8 @@ static void setup_system_wide(int forks)
>   		struct evsel *counter;
>   
>   		evlist__for_each_entry(evsel_list, counter) {
> +			if (counter->tool_event == PERF_TOOL_DURATION_TIME)
> +				continue;
>   			if (!counter->core.system_wide)
>   				return;
>   		}