lkml.org 
[lkml]   [2015]   [Jul]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH V3 3/5] perf,tools: Dump per-sample freq/CPU%/CORE_BUSY% in report -D
Em Wed, Jul 29, 2015 at 02:52:20PM +0200, Jiri Olsa escreveu:
> On Tue, Jul 28, 2015 at 07:29:33AM -0400, kan.liang@intel.com wrote:
>
> SNIP
>
> > -static void dump_sample(struct perf_evsel *evsel, union perf_event *event,
> > +static void dump_sample(struct perf_session *session, struct perf_evlist *evlist,
> > + struct perf_evsel *evsel, union perf_event *event,
> > struct perf_sample *sample)
> > {
> > u64 sample_type;
> > @@ -938,7 +963,7 @@ static void dump_sample(struct perf_evsel *evsel, union perf_event *event,
> > printf("... transaction: %" PRIx64 "\n", sample->transaction);
> >
> > if (sample_type & PERF_SAMPLE_READ)
> > - sample_read__printf(sample, evsel->attr.read_format);
> > + sample_read__printf(session, evlist, sample, evsel->attr.read_format);
> > }
> >
> > static struct machine *machines__find_for_cpumode(struct machines *machines,
> > @@ -1036,12 +1061,13 @@ static int
> > &sample->read.one, machine);
> > }
> >
> > -static int machines__deliver_event(struct machines *machines,
> > +static int machines__deliver_event(struct perf_session *session,
> > struct perf_evlist *evlist,
>
> you can cut the evlist argument as well, because it's always from session->evlist

Well, perf_session is about saving/restoring from a perf.data file, and
there are tools that don't deal with perf.data files.

So I think that this running environment information, that is associated
with the evlist, should be accessible from there, i.e. when reading the
perf.data file you get it from there, when running a live tool, you get
it from the system.

I saw code in this patchkit that reads it from the system and writes it
to the perf.data header, that could sould instead get it into a data
structure accessible from the evlist and when writing a perf.data file,
get it from there.

I.e. trying not to entangle session stuff into places that don't touch
it.

Reading a bit more to see if I can come with more suggestions...

- Arnaldo


\
 
 \ /
  Last update: 2015-07-29 21:11    [W:0.090 / U:0.944 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site