All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=758277

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=aed0500404d3a0897e4f6a06024c406d51d0553c2eb35337986cf0aecfddec4b


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

android-nexus5

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8970449089619153568

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

Hi csharrison@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 : Charles Harrison
  Commit : 007abddf6ad7f7d3f1491217119e3e7166faaace
  Date   : Thu Aug 17 21:39:49 2017
  Subject: Testing config: doc.write blocking change on slow connections only

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : smoothness.gpu_rasterization.top_25_smooth
  Metric       : input_event_latency_discrepancy/ESPN
  Change       : 69.88% | 50.8296777778 -> 86.3482444444

Revision             Result                  N
chromium@495292      50.8297 +- 31.3957      9      good
chromium@495312      37.8503 +- 15.1961      6      good
chromium@495322      50.0696 +- 15.5744      6      good
chromium@495325      47.3043 +- 25.8259      6      good
chromium@495326      51.1877 +- 12.511       6      good
chromium@495327      94.5689 +- 55.8362      6      bad       <--
chromium@495331      89.0557 +- 48.5423      6      bad
chromium@495369      86.3482 +- 66.3129      9      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=ESPN smoothness.gpu_rasterization.top_25_smooth

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8970449089619153568


For feedback, file a bug with component Speed>Bisection

Back to normal levels before [1] landed. My guess is that we were blocking resource loads on this page.

[1] : https://chromium.googlesource.com/chromium/src/+/7847afc23c816960a019476e995c696143365cac

See issue 753794

 Issue 758275  has been merged into this issue.

 Issue 758274  has been merged into this issue.

 Issue 758598  has been merged into this issue.

 Issue 758997  has been merged into this issue.

 Issue 759640  has been merged into this issue.

 Issue 759639  has been merged into this issue.

 Issue 759638  has been merged into this issue.

 Issue 759692  has been merged into this issue.

 Issue 759730  has been merged into this issue.

 Issue 759643  has been merged into this issue.

 Issue 759726  has been merged into this issue.

 Issue 759729  has been merged into this issue.

 Issue 760150  has been merged into this issue.

Issue 759648 has been merged into this issue.

Issue 758969 has been merged into this issue.

Note the latest dupe on issue 758969 comes from a significant battery regression on Android which I don't see as recovered:

https://chromeperf.appspot.com/group_report?sid=54358387410d141bf5b67fe1079a96add11defa841ef95f022da5a0c786fb196

Latest value is still at about 0.267865 mWh, value before this CL was 0.181317 mWh.

Ah, sorry, forget what I said. I've just re-read #4.

It's true before that before r489825 the usage was also near 0.26 mWh. So this is indeed back to normal.