Messages in this thread | | | Date | Fri, 28 Apr 2023 09:31:30 -0400 | Subject | Re: [PATCH v2 01/43] perf stat: Disable TopdownL1 on hybrid | From | "Liang, Kan" <> |
| |
On 2023-04-28 3:37 a.m., Ian Rogers wrote: > Bugs with event parsing, event grouping and metrics causes the > TopdownL1 metricgroup to crash the perf command. Temporarily disable > the group if no events/metrics are spcecified. >
I think it's OK to only disable the metric for perf stat default temporarily on hybrid as a workaround for 6.4. Now, the only problem is the default of perf stat on pre-icl in 6.4.
> Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Arnaldo, could you please back port this for 6.4 as well?
Thanks, Kan > --- > tools/perf/builtin-stat.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c > index efda63f6bf32..be9677aa642f 100644 > --- a/tools/perf/builtin-stat.c > +++ b/tools/perf/builtin-stat.c > @@ -1885,7 +1885,12 @@ static int add_default_attributes(void) > * Add TopdownL1 metrics if they exist. To minimize > * multiplexing, don't request threshold computation. > */ > - if (metricgroup__has_metric("TopdownL1") && > + /* > + * TODO: TopdownL1 is disabled on hybrid CPUs to avoid a crashes > + * caused by exposing latent bugs. This is fixed properly in: > + * https://lore.kernel.org/lkml/bff481ba-e60a-763f-0aa0-3ee53302c480@linux.intel.com/ > + */ > + if (metricgroup__has_metric("TopdownL1") && !perf_pmu__has_hybrid() && > metricgroup__parse_groups(evsel_list, "TopdownL1", > /*metric_no_group=*/false, > /*metric_no_merge=*/false, > @@ -1894,6 +1899,7 @@ static int add_default_attributes(void) > stat_config.system_wide, > &stat_config.metric_events) < 0) > return -1; > + > /* Platform specific attrs */ > if (evlist__add_default_attrs(evsel_list, default_null_attrs) < 0) > return -1;
| |