All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=702203

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9IvdugoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9LXJlAgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgtLPJ7AkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9Lj82ggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgtJ-85QsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9LaGsAsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9InrqgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9IDBoQsM


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

android-nexus5X
android-nexus6
android-webview-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984957554923319504

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

Hi ksakamoto@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 : ksakamoto
  Commit : 4d059dcf25a256d591234eb2a7ffd1673f7c9709
  Date   : Wed Mar 15 06:57:08 2017
  Subject: Loosen FirstMeaningfulPaint UMA network-idle heuristics

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : system_health.common_mobile
  Metric       : timeToFirstMeaningfulPaint_avg/browse_media/browse_media_imgur
  Change       : 13440.98% | 226.491333333 -> 30669.1513333

Revision             Result                  N
chromium@456987      226.491 +- 44.5066      6      good
chromium@457006      231.826 +- 43.3993      6      good
chromium@457015      230.243 +- 43.7344      6      good
chromium@457016      227.117 +- 44.0644      6      good
chromium@457017      30609.1 +- 178.126      6      bad       <--
chromium@457018      30633.3 +- 174.348      6      bad
chromium@457020      30627.1 +- 63.8508      6      bad
chromium@457024      30669.2 +- 83.1258      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=browse.media.imgur system_health.common_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984957554923319504

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5296855098851328


| 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!

I have inspected a few traces from affected tests.

Before my patch, FirstMeaningfulPaint was reported too soon because of  bug 659021 . My patch changed some heuristics for FirstMeaningfulPaint, and it now reports actual FirstMeaningfulPaint for those pages.