lkml.org 
[lkml]   [2014]   [Oct]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH V6 0/3] perf tools: pmu event new style format fix
On Thu, Oct 02, 2014 at 05:33:28PM +0000, Liang, Kan wrote:
>
>
> >
> > On Thu, Sep 11, 2014 at 03:08:56PM -0400, kan.liang@intel.com wrote:
> > > From: Kan Liang <kan.liang@intel.com>
> > >
> > > There are two types of pmu event stytle formats, "pmu_event_name"
> > > or "cpu/pmu_event_name/". However, there is a bug on supporting these
> > > two formats, especially when they mixed with other perf events.
> > > The patch set intends to fix this issue.
> > >
> > > The patch set has been tested on my haswell.
> > > Here is the test script I used for this issue.
> > > (Note: please make sure that your test system supports TSX and
> > > L1-dcache-loads events. Otherwise, you may want to change the events
> > > to other pmu events.)
> >
> > any chance changing this script to work for everyone?
> > and make it part of theautomated tests suite?
> >
> > also how about something like in the attached change
>
> It cannot trigger the issue, since current perf will automatically add cpu// for pmu event.
> We have to test at least two events with different style.
> How about the change as below?

looks good,

jirka

>
> --- a/tools/perf/tests/parse-events.c
> +++ b/tools/perf/tests/parse-events.c
> @@ -457,6 +457,36 @@ static int test__checkevent_pmu_events(struct perf_evlist *evlist)
> return 0;
> }
>
> +
> +static int test__checkevent_pmu_events_mix(struct perf_evlist *evlist)
> +{
> + struct perf_evsel *evsel = perf_evlist__first(evlist);
> +
> + /* pmu-event:u */
> + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
> + TEST_ASSERT_VAL("wrong exclude_user",
> + !evsel->attr.exclude_user);
> + TEST_ASSERT_VAL("wrong exclude_kernel",
> + evsel->attr.exclude_kernel);
> + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
> + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
> + TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
> +
> + /* cpu/pmu-event/u*/
> + evsel = perf_evsel__next(evsel);
> + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
> + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
> + TEST_ASSERT_VAL("wrong exclude_user",
> + !evsel->attr.exclude_user);
> + TEST_ASSERT_VAL("wrong exclude_kernel",
> + evsel->attr.exclude_kernel);
> + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
> + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
> + TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
> +
> + return 0;
> +}
> +
> static int test__checkterms_simple(struct list_head *terms)
> {
> struct parse_events_term *term;
> @@ -1554,6 +1584,12 @@ static int test_pmu_events(void)
> e.check = test__checkevent_pmu_events;
>
> ret = test_event(&e);
> + if (ret)
> + break;
> + snprintf(name, MAX_NAME, "%s:u,cpu/event=%s/u", ent->d_name, ent->d_name);
> + e.name = name;
> + e.check = test__checkevent_pmu_events_mix;
> + ret = test_event(&e);
> #undef MAX_NAME
> }
>
>
> Thanks,
> Kan
>
>
> >
> > jirka
> >
> >
> > ---
> > diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-
> > events.c index 5941927a4b7f..2b7c48474761 100644
> > --- a/tools/perf/tests/parse-events.c
> > +++ b/tools/perf/tests/parse-events.c
> > @@ -1554,6 +1554,15 @@ static int test_pmu_events(void)
> > e.check = test__checkevent_pmu_events;
> >
> > ret = test_event(&e);
> > + if (ret)
> > + break;
> > +
> > + snprintf(name, MAX_NAME, "event=%s", ent->d_name);
> > +
> > + e.name = name;
> > + e.check = test__checkevent_pmu_events;
> > +
> > + ret = test_event(&e);
> > #undef MAX_NAME
> > }


\
 
 \ /
  Last update: 2014-10-03 11:01    [W:0.057 / U:0.072 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site