Re: [PATCH v1 4/9] perf util: Update and print per-thread shadow stats

[Jin, Yao]

On 11/22/2017 4:37 PM, Jiri Olsa wrote:
> On Wed, Nov 22, 2017 at 11:10:37AM +0800, Jin, Yao wrote:
> > On 11/21/2017 11:18 PM, Jiri Olsa wrote:
> > > On Mon, Nov 20, 2017 at 10:43:39PM +0800, Jin Yao wrote:
> > >
> > > SNIP
> > >
> > > >    	if (num == 0)
> > > > diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
> > > > index 151e9ef..50bb16d 100644
> > > > --- a/tools/perf/util/stat.c
> > > > +++ b/tools/perf/util/stat.c
> > > > @@ -278,9 +278,16 @@ process_counter_values(struct perf_stat_config *config, struct perf_evsel *evsel
> > > >    			perf_evsel__compute_deltas(evsel, cpu, thread, count);
> > > >    		perf_counts_values__scale(count, config->scale, NULL);
> > > >    		if (config->aggr_mode == AGGR_NONE)
> > > > -			perf_stat__update_shadow_stats(evsel, count->val, cpu);
> > > > -		if (config->aggr_mode == AGGR_THREAD)
> > > > -			perf_stat__update_shadow_stats(evsel, count->val, 0);
> > > > +			perf_stat__update_shadow_stats(evsel, count->val, cpu,
> > > > +						       NULL);
> > > > +		if (config->aggr_mode == AGGR_THREAD) {
> > > > +			if (config->stats)
> > > > +				perf_stat__update_shadow_stats(evsel,
> > > > +					count->val, 0, &config->stats[thread]);
> > >
> > > please add this part together with config->stats allocation/usage
> > > and keep this change only about adding the struct runtime_stat *stat
> > > into the code path with no functional change
> > >
> > > thanks,
> > > jirka
> >
> > The interface of perf_stat__update_shadow_stats() is changed (added with a
> > new parameter 'struct runtime_stat *stat').
> >
> > If I move above part to another patch (e.g. move to
> > v1-0006-perf-stat-Allocate-shadow-stats-buffer-for-thread.patch), the
> > compilation would be failed if only applying with this patch. It's not good
> > for git bisect.
>
> but you could make the patch just to add the 'struct runtime_stat *stat'
> as an argument through the code path..  as a base for the stats, no?
>
> jirka

Yeah, that might be OK. Let me have a try.

Thanks
Jin Yao