All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=782933

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=ada45be37d8dba3e56ef3e0820413f95be95ba2fb8a67b35d8ab6b5296e6861a


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8963453515820688288

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

Hi treib@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 : Marc Treib
  Commit : cbc140c9ece3d71c6c4233b20ccfb5d52cbb2229
  Date   : Thu Nov 02 17:49:28 2017
  Subject: Thumbnail capturing: Resolve raciness

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : media.mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:java_heap:proportional_resident_size_avg/video.html?src_crowd1080_vp9.webm
  Change       : 0.51% | 16914066.2857 -> 16994304.0

Revision             Result                  N
chromium@513539      16914066 +- 314712      14      good
chromium@513540      17023886 +- 188307      9       bad       <--
chromium@513542      16992792 +- 331478      21      bad
chromium@513543      16993134 +- 293523      21      bad
chromium@513547      16976847 +- 255834      21      bad
chromium@513554      17028681 +- 336959      14      bad
chromium@513570      16978651 +- 334223      21      bad
chromium@513598      16994304 +- 208976      14      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=video.html.src.crowd1080.vp9.webm media.mobile

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8963453515820688288


For feedback, file a bug with component Speed>Bisection

The code changed in that CL is not used on Android at all; the only place where ThumbnailTabHelper is instantiated is here: https://cs.chromium.org/chromium/src/chrome/browser/ui/tab_helpers.cc?rcl=8a1b5cf9b28b49d49e124c85f4a52eb533af7a11&l=298 which is #ifdef'd out on Android.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8963412496367671744

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

Hi crouleau@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 : Caleb Rouleau
  Commit : 02bf4d658b908009ca8275f50fc5eba1da314a00
  Date   : Thu Nov 02 14:15:21 2017
  Subject: [Telemetry] Remove media.mse_cases benchmark.

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : media.mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:java_heap:proportional_resident_size_avg/video.html?src_crowd1080_vp9.webm
  Change       : 1.13% | 16898121.1429 -> 17023488.0

Revision             Result                  N
chromium@513479      16898121 +- 312608      14      good
chromium@513480      16992939 +- 333794      21      bad       <--
chromium@513481      17015881 +- 298662      14      bad
chromium@513482      17012224 +- 378832      14      bad
chromium@513485      17003634 +- 208235      9       bad
chromium@513491      17008185 +- 224095      9       bad
chromium@513503      16989574 +- 342997      21      bad
chromium@513524      16955831 +- 328676      14      bad
chromium@513568      16974734 +- 235482      9       bad
chromium@513656      17023488 +- 165424      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=video.html.src.crowd1080.vp9.webm media.mobile

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8963412496367671744


For feedback, file a bug with component Speed>Bisection

Uh, this makes no sense at all. My change shouldn't have affected media.mobile. Probably something strange happening here since this is reported_by_os memory (which we tried to turn off alerts for.