All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=727622

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-vPxowoM


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8978177698602542656

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

Hi alexmos@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : alexmos
  Commit : c5e91fafd0be6f67afc2dd41d33c36492c5ffb43
  Date   : Thu May 25 02:55:01 2017
  Subject: Use has_committed_real_load for main frames in addition to subframes.

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : smoothness.top_25_smooth
  Metric       : frame_times/LinkedIn
  Change       : 1.02% | 16.8776415684 -> 17.0496886687

Revision             Result                    N
chromium@474507      16.8776 +- 0.0119931      9       good
chromium@474526      16.9095 +- 0.220417       9       good
chromium@474535      16.8885 +- 0.0163049      14      good
chromium@474536      16.8807 +- 0.0047696      6       bad       <--
chromium@474537      16.8963 +- 0.254506       14      bad
chromium@474539      17.0448 +- 0.35487        9       bad
chromium@474544      17.0114 +- 0.454722       14      bad
chromium@474580      17.0497 +- 0.282451       6       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=LinkedIn smoothness.top_25_smooth

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978177698602542656

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5571853034192896


| 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 Speed>Bisection.  Thank you!

+simonhatch, numbers on the bisect in #3 look odd to me, isn't the jump around r474539?

re: #c4

Looking through why it classified r474537 as "bad", found that there was one outlier in the samples (mostly 16.88'ish, one 17.1), so it ended up classifying it as a different dataset than LKGR.

I'll try rerunning this but reduce the range since it looks pretty clear.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8978163975073812576

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

Hi fsamuel@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : fsamuel
  Commit : b520968f4ac40fb89ce25329134b388ca42cca12
  Date   : Thu May 25 03:10:42 2017
  Subject: cc::SurfaceDependencyTracker should not crash when a Display goes away

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : smoothness.top_25_smooth
  Metric       : frame_times/LinkedIn
  Change       : 1.73% | 16.8554988116 -> 17.1467209893

Revision             Result                    N
chromium@474536      16.8555 +- 0.0176763      6      good
chromium@474538      16.8587 +- 0.0137342      6      good
chromium@474539      17.095 +- 0.0116995       6      bad       <--
chromium@474540      17.103 +- 0.00994584      6      bad
chromium@474544      17.1467 +- 0.242097       6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=LinkedIn smoothness.top_25_smooth

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978163975073812576

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5341182420320256


| 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 Speed>Bisection.  Thank you!