All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=639459

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgruWGsAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgrseStAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgrpuYuAoM


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

android-nexus5
android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9003865849861363392

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

Hi jbauman@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 : Avoid creating sync token when only change is new hardware video frame.
Author  : jbauman
Commit description:
  
Once we get the picture ready IPC the GPU has already done all the
processing it needs to for the frame, so the renderer doesn't need to
create a new sync token. We can instead use the same sync token every
frame to avoid a synchronous IPC to the GPU process.

BUG= 575856 

Review-Url: https://codereview.chromium.org/2254993003
Cr-Commit-Position: refs/heads/master@{#412934}
Commit  : 1d06968c477a6db0d33804d346ed590c19251e0b
Date    : Thu Aug 18 21:29:56 2016


===== TESTED REVISIONS =====
Revision         Mean    Std Dev  N  Good?
chromium@412924  -113.2  55.0018  5  good
chromium@412933  -122.0  37.3095  5  good
chromium@412934  503.2   27.335   5  bad    <--
chromium@412935  460.0   22.4054  5  bad
chromium@412937  268.0   198.136  5  bad
chromium@412941  114.8   29.4483  5  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 639459

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests media.android.tough_video_cases
Test Metric: vm_private_dirty_delta/tulip2.mp4_gpu
Relative Change: 201.41%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/546
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9003865849861363392


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6363158562734080

| 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!

In general the fact that this would change memory usage is pretty odd. It looks like the browser process improved by more than the GPU process worsened, so maybe this just changed the timing of the measurement in some way that things are accounted for differently?