public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] perf evlist: Fix error out while applying initial delay and LBR
@ 2018-07-09 14:15 kan.liang
  2018-07-09 15:20 ` Jiri Olsa
  2018-08-02  8:07 ` [tip:perf/core] " tip-bot for Kan Liang
  0 siblings, 2 replies; 4+ messages in thread
From: kan.liang @ 2018-07-09 14:15 UTC (permalink / raw)
  To: acme, peterz, mingo, linux-kernel
  Cc: jolsa, namhyung, ak, sunil.k.pandey, Kan Liang

From: Kan Liang <kan.liang@linux.intel.com>

Perf record will error out if both --delay and LBR are applied.
For example:
  perf record -D 1000 -a -e cycles -j any -- sleep 2
  Error:
  dummy:HG: PMU Hardware doesn't support sampling/overflow-interrupts.
  Try 'perf stat'

A dummy event is added implicitly for initial delay, which has the same
configurations as real sampling events. The dummy event is a software
event. If LBR is configured, perf must error out.

The dummy event only be used to track PERF_RECORD_MMAP while perf waits
for the initial delay to enable the real events. The BRANCH_STACK bit
can be safely cleared for the dummy event.

After applying the patch:
  perf record -D 1000 -a -e cycles -j any -- sleep 2
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 1.054 MB perf.data (828 samples) ]

Reported-by: Pandey, Sunil K <sunil.k.pandey@intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
---
 tools/perf/util/evsel.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 94fce4f..0d55047 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -848,6 +848,12 @@ static void apply_config_terms(struct perf_evsel *evsel,
 	}
 }
 
+static bool is_dummy_event(struct perf_evsel *evsel)
+{
+	return (evsel->attr.type == PERF_TYPE_SOFTWARE) &&
+	       (evsel->attr.config == PERF_COUNT_SW_DUMMY);
+}
+
 /*
  * The enable_on_exec/disabled value strategy:
  *
@@ -1086,6 +1092,14 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts,
 		else
 			perf_evsel__reset_sample_bit(evsel, PERIOD);
 	}
+
+	/*
+	 * For initial_delay, a dummy event is added implicitly.
+	 * The software event will trigger -EOPNOTSUPP error out,
+	 * if BRANCH_STACK bit is set.
+	 */
+	if (opts->initial_delay && is_dummy_event(evsel))
+		perf_evsel__reset_sample_bit(evsel, BRANCH_STACK);
 }
 
 static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
-- 
2.7.4


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] perf evlist: Fix error out while applying initial delay and LBR
  2018-07-09 14:15 [PATCH] perf evlist: Fix error out while applying initial delay and LBR kan.liang
@ 2018-07-09 15:20 ` Jiri Olsa
  2018-07-26 12:51   ` Arnaldo Carvalho de Melo
  2018-08-02  8:07 ` [tip:perf/core] " tip-bot for Kan Liang
  1 sibling, 1 reply; 4+ messages in thread
From: Jiri Olsa @ 2018-07-09 15:20 UTC (permalink / raw)
  To: kan.liang; +Cc: acme, peterz, mingo, linux-kernel, namhyung, ak, sunil.k.pandey

On Mon, Jul 09, 2018 at 07:15:22AM -0700, kan.liang@linux.intel.com wrote:
> From: Kan Liang <kan.liang@linux.intel.com>
> 
> Perf record will error out if both --delay and LBR are applied.
> For example:
>   perf record -D 1000 -a -e cycles -j any -- sleep 2
>   Error:
>   dummy:HG: PMU Hardware doesn't support sampling/overflow-interrupts.
>   Try 'perf stat'
> 
> A dummy event is added implicitly for initial delay, which has the same
> configurations as real sampling events. The dummy event is a software
> event. If LBR is configured, perf must error out.
> 
> The dummy event only be used to track PERF_RECORD_MMAP while perf waits
> for the initial delay to enable the real events. The BRANCH_STACK bit
> can be safely cleared for the dummy event.
> 
> After applying the patch:
>   perf record -D 1000 -a -e cycles -j any -- sleep 2
>   [ perf record: Woken up 1 times to write data ]
>   [ perf record: Captured and wrote 1.054 MB perf.data (828 samples) ]
> 
> Reported-by: Pandey, Sunil K <sunil.k.pandey@intel.com>
> Signed-off-by: Kan Liang <kan.liang@linux.intel.com>

Acked-by: Jiri Olsa <jolsa@kernel.org>

thanks,
jirka

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] perf evlist: Fix error out while applying initial delay and LBR
  2018-07-09 15:20 ` Jiri Olsa
@ 2018-07-26 12:51   ` Arnaldo Carvalho de Melo
  0 siblings, 0 replies; 4+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-07-26 12:51 UTC (permalink / raw)
  To: Jiri Olsa
  Cc: kan.liang, peterz, mingo, linux-kernel, namhyung, ak,
	sunil.k.pandey

Em Mon, Jul 09, 2018 at 05:20:54PM +0200, Jiri Olsa escreveu:
> On Mon, Jul 09, 2018 at 07:15:22AM -0700, kan.liang@linux.intel.com wrote:
> > After applying the patch:
> >   perf record -D 1000 -a -e cycles -j any -- sleep 2
> >   [ perf record: Woken up 1 times to write data ]
> >   [ perf record: Captured and wrote 1.054 MB perf.data (828 samples) ]
> > 
> > Reported-by: Pandey, Sunil K <sunil.k.pandey@intel.com>
> > Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
> 
> Acked-by: Jiri Olsa <jolsa@kernel.org>

Thanks, applied, tested.

- Arnaldo

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [tip:perf/core] perf evlist: Fix error out while applying initial delay and LBR
  2018-07-09 14:15 [PATCH] perf evlist: Fix error out while applying initial delay and LBR kan.liang
  2018-07-09 15:20 ` Jiri Olsa
@ 2018-08-02  8:07 ` tip-bot for Kan Liang
  1 sibling, 0 replies; 4+ messages in thread
From: tip-bot for Kan Liang @ 2018-08-02  8:07 UTC (permalink / raw)
  To: linux-tip-commits
  Cc: jolsa, acme, kan.liang, peterz, namhyung, mingo, sunil.k.pandey,
	ak, linux-kernel, hpa, tglx

Commit-ID:  95035c5e167ae6e740b1ddd30210ae0eaf39a5db
Gitweb:     https://git.kernel.org/tip/95035c5e167ae6e740b1ddd30210ae0eaf39a5db
Author:     Kan Liang <kan.liang@linux.intel.com>
AuthorDate: Mon, 9 Jul 2018 07:15:22 -0700
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Tue, 31 Jul 2018 09:56:46 -0300

perf evlist: Fix error out while applying initial delay and LBR

'perf record' will error out if both --delay and LBR are applied.

For example:

  # perf record -D 1000 -a -e cycles -j any -- sleep 2
  Error:
  dummy:HG: PMU Hardware doesn't support sampling/overflow-interrupts.
  Try 'perf stat'
  #

A dummy event is added implicitly for initial delay, which has the same
configurations as real sampling events. The dummy event is a software
event. If LBR is configured, perf must error out.

The dummy event will only be used to track PERF_RECORD_MMAP while perf
waits for the initial delay to enable the real events. The BRANCH_STACK
bit can be safely cleared for the dummy event.

After applying the patch:

  # perf record -D 1000 -a -e cycles -j any -- sleep 2
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 1.054 MB perf.data (828 samples) ]
  #

Reported-by: Sunil K Pandey <sunil.k.pandey@intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1531145722-16404-1-git-send-email-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evsel.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 5285da0417c5..ddf84b941abf 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -859,6 +859,12 @@ static void apply_config_terms(struct perf_evsel *evsel,
 	}
 }
 
+static bool is_dummy_event(struct perf_evsel *evsel)
+{
+	return (evsel->attr.type == PERF_TYPE_SOFTWARE) &&
+	       (evsel->attr.config == PERF_COUNT_SW_DUMMY);
+}
+
 /*
  * The enable_on_exec/disabled value strategy:
  *
@@ -1097,6 +1103,14 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts,
 		else
 			perf_evsel__reset_sample_bit(evsel, PERIOD);
 	}
+
+	/*
+	 * For initial_delay, a dummy event is added implicitly.
+	 * The software event will trigger -EOPNOTSUPP error out,
+	 * if BRANCH_STACK bit is set.
+	 */
+	if (opts->initial_delay && is_dummy_event(evsel))
+		perf_evsel__reset_sample_bit(evsel, BRANCH_STACK);
 }
 
 static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)

^ permalink raw reply related	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2018-08-02  8:08 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-07-09 14:15 [PATCH] perf evlist: Fix error out while applying initial delay and LBR kan.liang
2018-07-09 15:20 ` Jiri Olsa
2018-07-26 12:51   ` Arnaldo Carvalho de Melo
2018-08-02  8:07 ` [tip:perf/core] " tip-bot for Kan Liang

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox