All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=619975

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvKOMqQoM


Bot(s) for this bug's original alert(s):

chromium-rel-mac-retina

I'm running a bisect, but https://codereview.chromium.org/1999893004 is the likely suspect.

=== Auto-CCing suspected CL author alokp@chromium.org ===

Hi alokp@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Splits PipelineImpl into main and media thread components.
Author  : alokp
Commit description:
  
The outer PipelineImpl runs on the main thread and acts as a proxy
between PipelineClient and the internal RendererWrapper that runs on
the media thread. This sets the stage to get rid of WaitableEvent
and Locks used in PipelineImpl.

This patch should not change anything functionality wise except for
timing of the notifications received from Demuxer and Renderer. In
response to the notifications we used to change member variables
inside a lock. Now we post a task to change those variables to the
main thread. The list of variables affected by this change are:
- buffered_time_ranges_
- did_loading_progress_
- duration_
- statistics_

BUG= 616959 

Review-Url: https://codereview.chromium.org/1999893004
Cr-Commit-Position: refs/heads/master@{#399376}
Commit  : 1116967f7c15965538b42c2a402712279d4e5662
Date    : Sat Jun 11 17:32:57 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@399373  220.462  2.95043  5  good
chromium@399375  219.059  1.3157   5  good
chromium@399376  245.709  4.38754  5  bad    <--

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 619975

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/crowd720_vp9.webm_seek_cold
Relative Change: 11.45%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1334
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9009861042934353600


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5784331755192320

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Splits PipelineImpl into main and media thread components.
Author  : alokp
Commit description:
  
The outer PipelineImpl runs on the main thread and acts as a proxy
between PipelineClient and the internal RendererWrapper that runs on
the media thread. This sets the stage to get rid of WaitableEvent
and Locks used in PipelineImpl.

This patch should not change anything functionality wise except for
timing of the notifications received from Demuxer and Renderer. In
response to the notifications we used to change member variables
inside a lock. Now we post a task to change those variables to the
main thread. The list of variables affected by this change are:
- buffered_time_ranges_
- did_loading_progress_
- duration_
- statistics_

BUG= 616959 

Review-Url: https://codereview.chromium.org/1999893004
Cr-Commit-Position: refs/heads/master@{#399376}
Commit  : 1116967f7c15965538b42c2a402712279d4e5662
Date    : Sat Jun 11 17:32:57 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@399373  217.207  1.94795  5  good
chromium@399375  219.918  1.81653  5  good
chromium@399376  242.348  3.20013  5  bad    <--

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 619975

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/crowd720_vp9.webm_seek_cold
Relative Change: 11.57%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1335
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9009860924993313280


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5274660334206976

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

David: Does this benchmark measure seek time?

I assume that's what it does, but Caleb can confirm.

Yes, it does measure seek time. It measures "seek" and "avg_loop_time".

I am not able to repro this locally on linux desktop. Results are attached for running the pageset 5 times.

I will try running the benchmark on mac.

Deleted: results.html 117 KB

results.html 117 KB View Download

There were no alerts for platforms other than Mac, so this is not surprising.

Any progress on this? It's blocking an issue that is listed as a top crasher for the Stability Sheriffs.

I uploaded a patch yesterday: https://codereview.chromium.org/2071503003/

I will followup with the reviewers today.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/8f268bb64214eab73f00216b96b919a1b2955999

commit 8f268bb64214eab73f00216b96b919a1b2955999
Author: alokp <alokp@chromium.org>
Date: Wed Jun 22 23:26:48 2016

Avoid flooding main thread by caching state on media thread.

OnStatisticsUpdate and OnBufferedTimeRangesChanged fire much more
frequently than queried for. Posting all state change notifications
to the main thread causes potential performance issues. Cache state
not immediately required by PipelineClient on the media thread to
avoid having to post all state change notifications.

BUG= 619975 

Review-Url: https://codereview.chromium.org/2071503003
Cr-Commit-Position: refs/heads/master@{#401457}

[modify] https://crrev.com/8f268bb64214eab73f00216b96b919a1b2955999/media/base/pipeline_impl.cc
[modify] https://crrev.com/8f268bb64214eab73f00216b96b919a1b2955999/media/base/pipeline_impl.h