Tracing: Enable tracing_perftests on bots.

Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and it's designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
Committed: https://crrev.com/c07975895dccd5899b528949fd82750a90387a91
Cr-Commit-Position: refs/heads/master@{#427379}

[kraynov, 2016-10-14 13:25:04]

The CQ bit was checked by kraynov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-14 13:25:19]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[kraynov, 2016-10-14 13:54:35]

kraynov@chromium.org changed reviewers:
+ primiano@chromium.org

[Primiano Tucci (use gerrit), 2016-10-14 14:03:45]

primiano@chromium.org changed reviewers:
+ dtu@chromium.org, nednguyen@google.com

[Primiano Tucci (use gerrit), 2016-10-14 14:03:45]

Looks reasonable to me but I'm not that experienced with this part of the infra.
I'm just going to pass the ball to more knowledgeable people.
+dtu +nednguyen +zhenw

My other question is: what about the name of the benchmark?
Is this going to add data to zhen's benchmark? Is it going to be a completely
different name? WHat are the names of the two?

https://codereview.chromium.org/2418983003/diff/1/BUILD.gn
File BUILD.gn (left):

https://codereview.chromium.org/2418983003/diff/1/BUILD.gn#oldcode987
BUILD.gn:987: if (!is_chromeos) {
Right this is a subset of the check above.

[Primiano Tucci (use gerrit), 2016-10-14 14:06:09]

I mean, ideally given that we have zhens's  tracing.tracing_With... should this
one be tracing.macros_perf ? or something like that?

[commit-bot: I haz the power, 2016-10-14 14:29:28]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-14 14:29:29]

Dry run: This issue passed the CQ dry run.

[nednguyen, 2016-10-14 14:59:57]

On 2016/10/14 14:06:09, Primiano Tucci wrote:
> I mean, ideally given that we have zhens's  tracing.tracing_With... should
this
> one be tracing.macros_perf ? or something like that?

I have few high level questions:
What does this test measure? How long does it run? Do you need it on actual
hardware? 


Can you make a doc that answer those questions so everyone can chime in?

[kraynov, 2016-10-14 15:08:49]

Hi Ned,
1. This test is designed to measure overhead on TRACE_EVENT macros. We've
started from simple case crrev.com/2403923002 and more cases are coming after
getting such values on dashboard.
2. It runs in a 2 seconds now and we would keep it short.
3. Right now let it only run on Linux, but it's good to run on phones in future.

[Primiano Tucci (use gerrit), 2016-10-14 15:17:45]

On 2016/10/14 15:08:49, kraynov wrote:
> Hi Ned,
> 1. This test is designed to measure overhead on TRACE_EVENT macros. We've
> started from simple case crrev.com/2403923002 and more cases are coming after
> getting such values on dashboard.
> 2. It runs in a 2 seconds now and we would keep it short.
> 3. Right now let it only run on Linux, but it's good to run on phones in
future.

I'd like this to run on phones as well (I care more about that honestly) Can you
just do everything here.
Maybe a quick doc that explains what you are doing as ned suggested might be
useful.

[kraynov, 2016-10-14 16:05:26]

Description was changed from

==========
Tracing: Enable tracing_perftests on bots.
==========

to

==========
Tracing: Enable tracing_perftests on bots.

BUG=656029
==========

[fmeawad, 2016-10-14 18:05:56]

On 2016/10/14 15:17:45, Primiano Tucci wrote:
> On 2016/10/14 15:08:49, kraynov wrote:
> > Hi Ned,
> > 1. This test is designed to measure overhead on TRACE_EVENT macros. We've
> > started from simple case crrev.com/2403923002 and more cases are coming
after
> > getting such values on dashboard.
> > 2. It runs in a 2 seconds now and we would keep it short.
> > 3. Right now let it only run on Linux, but it's good to run on phones in
> future.
> 
> I'd like this to run on phones as well (I care more about that honestly) Can
you
> just do everything here.
> Maybe a quick doc that explains what you are doing as ned suggested might be
> useful.

This has to run on actual hardware, ideally all platforms.
I know that time functions vary drastically from one platform to another.

[eyaich1, 2016-10-18 15:55:29]

On 2016/10/14 18:05:56, fmeawad wrote:
> On 2016/10/14 15:17:45, Primiano Tucci wrote:
> > On 2016/10/14 15:08:49, kraynov wrote:
> > > Hi Ned,
> > > 1. This test is designed to measure overhead on TRACE_EVENT macros. We've
> > > started from simple case crrev.com/2403923002 and more cases are coming
> after
> > > getting such values on dashboard.
> > > 2. It runs in a 2 seconds now and we would keep it short.
> > > 3. Right now let it only run on Linux, but it's good to run on phones in
> > future.
> > 
> > I'd like this to run on phones as well (I care more about that honestly) Can
> you
> > just do everything here.
> > Maybe a quick doc that explains what you are doing as ned suggested might be
> > useful.
> 
> This has to run on actual hardware, ideally all platforms.
> I know that time functions vary drastically from one platform to another.

lgtm

[nednguyen, 2016-10-18 16:00:35]

https://codereview.chromium.org/2418983003/diff/1/tools/perf/generate_perf_js...
File tools/perf/generate_perf_json.py (right):

https://codereview.chromium.org/2418983003/diff/1/tools/perf/generate_perf_js...
tools/perf/generate_perf_json.py:116: 'script': 'gtest_perf_test.py',
Since this script is about perf, can you move it to tools/perf/micro_benchmarks
instead of testing/script/?

[kraynov, 2016-10-18 16:25:00]

Description was changed from

==========
Tracing: Enable tracing_perftests on bots.

BUG=656029
==========

to

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
==========

[kraynov, 2016-10-18 16:25:14]

Description was changed from

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
==========

to

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and it's designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
==========

[kraynov, 2016-10-18 16:25:28]

https://codereview.chromium.org/2418983003/diff/1/tools/perf/generate_perf_js...
File tools/perf/generate_perf_json.py (right):

https://codereview.chromium.org/2418983003/diff/1/tools/perf/generate_perf_js...
tools/perf/generate_perf_json.py:116: 'script': 'gtest_perf_test.py',
On 2016/10/18 16:00:35, nednguyen wrote:
> Since this script is about perf, can you move it to
tools/perf/micro_benchmarks
> instead of testing/script/?

It's not my script. It's the same used for cc_perftests.

[nednguyen, 2016-10-18 17:16:44]

On 2016/10/18 16:25:28, kraynov wrote:
>
https://codereview.chromium.org/2418983003/diff/1/tools/perf/generate_perf_js...
> File tools/perf/generate_perf_json.py (right):
> 
>
https://codereview.chromium.org/2418983003/diff/1/tools/perf/generate_perf_js...
> tools/perf/generate_perf_json.py:116: 'script': 'gtest_perf_test.py',
> On 2016/10/18 16:00:35, nednguyen wrote:
> > Since this script is about perf, can you move it to
> tools/perf/micro_benchmarks
> > instead of testing/script/?
> 
> It's not my script. It's the same used for cc_perftests.

I see. lgtm then

[kraynov, 2016-10-25 13:22:35]

kraynov@chromium.org changed reviewers:
+ eakuefner@chromium.org, sullivan@chromium.org

[kraynov, 2016-10-25 13:22:35]

+eakuefner
+sullivan
Could anyone review //testing/buildbot/chromium.perf.json, please?
Thank you!

[sullivan, 2016-10-25 13:26:01]

lgtm

rubberstamp lgtm (relying on eyaich's review here)

[kraynov, 2016-10-25 13:26:56]

The CQ bit was checked by kraynov@chromium.org

[commit-bot: I haz the power, 2016-10-25 13:27:19]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-25 13:34:05]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-25 13:34:07]

Try jobs failed on following builders:
  chromium_presubmit on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[kraynov, 2016-10-25 13:38:59]

kraynov@chromium.org changed reviewers:
+ brettw@chromium.org

[kraynov, 2016-10-25 13:39:00]

+brettw
Could you please review BUILD.gn file?
Thanks!

[brettw, 2016-10-25 16:38:34]

lgtm

[kraynov, 2016-10-25 16:39:28]

The CQ bit was checked by kraynov@chromium.org

[commit-bot: I haz the power, 2016-10-25 16:39:50]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-25 16:45:30]

Description was changed from

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and it's designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
==========

to

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and it's designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
==========

[commit-bot: I haz the power, 2016-10-25 16:45:31]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-10-25 17:16:28]

Description was changed from

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and it's designed exactly like cc_perftests.
See bug entry for details.

BUG=656029
==========

to

==========
Tracing: Enable tracing_perftests on bots.

There is an intention to track TRACE_EVENT macros performance
without browser and it's designed exactly like cc_perftests.
See bug entry for details.

BUG=656029

Committed: https://crrev.com/c07975895dccd5899b528949fd82750a90387a91
Cr-Commit-Position: refs/heads/master@{#427379}
==========

[commit-bot: I haz the power, 2016-10-25 17:16:30]

Patchset 1 (id:??) landed as
https://crrev.com/c07975895dccd5899b528949fd82750a90387a91
Cr-Commit-Position: refs/heads/master@{#427379}