On Fri, Aug 11, 2017 at 04:26:20PM -0700, Andi Kleen wrote: SNIP
> diff --git a/tools/perf/Documentation/perf-list.txt > b/tools/perf/Documentation/perf-list.txt > index f709de54707b..d432965d728d 100644 > --- a/tools/perf/Documentation/perf-list.txt > +++ b/tools/perf/Documentation/perf-list.txt > @@ -47,6 +47,7 @@ counted. The following modifiers exist: > P - use maximum detected precise level > S - read sample value (PERF_SAMPLE_READ) > D - pin the event to the PMU > + W - group is weak and will fallback to non-group if not schedulable > > The 'p' modifier can be used for specifying how precise the instruction > address should be. The 'p' modifier can be specified multiple times: > diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c > index 866da7aa54bf..ac605caf1a01 100644 > --- a/tools/perf/builtin-stat.c > +++ b/tools/perf/builtin-stat.c > @@ -582,6 +582,32 @@ static bool perf_evsel__should_store_id(struct > perf_evsel *counter) > return STAT_RECORD || counter->attr.read_format & PERF_FORMAT_ID; > } > > +static struct perf_evsel *reset_weak_group(struct perf_evsel *counter) > +{ > + struct perf_evsel *c2, *leader; > + bool is_open = true; > + > + leader = counter->leader; > + pr_debug("Weak group for %s/%d failed\n", > + leader->name, counter->nr_members); I'm getting 'branches/0' in here for you example: [root@krava perf]# ./perf stat -v -e '{branches,branch-misses,l1d.replacement,l2_lines_in.all,l2_rqsts.all_code_rd}:W' -a sleep 1 Using CPUID GenuineIntel-6-3D l1d.replacement -> cpu/umask=0x1,period=2000003,event=0x51/ l2_lines_in.all -> cpu/umask=0x7,period=100003,event=0xf1/ l2_rqsts.all_code_rd -> cpu/umask=0xe4,period=200003,event=0x24/ Weak group for branches/0 failed I'd expect nr_members == 5 ... any idea? thanks, jirka