All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=764346

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=52e87fcd5f921393d66881b64d2c98b191c703272b429b6dbffaa11bdb6e90d0


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

chromium-rel-win7-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8968648396823377840

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : scheduler.tough_scheduling_cases
  Metric       : queueing_durations/raf_canvas.html

Revision             Result                      N
chromium@500421      0.0294396 +- 0.0098694      21      good
chromium@500504      0.0301391 +- 0.011935       21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=raf.canvas.html scheduler.tough_scheduling_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8968648396823377840


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965301161551231616

Trying a second bisect on a wider range.

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

Hi dtapuska@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 : Dave Tapuska
  Commit : fbf2168e533bf593c0dfc5300f6a86f13fda3845
  Date   : Fri Sep 08 03:15:11 2017
  Subject: Reland "Enable mojo input messages on all platforms other than android."

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : scheduler.tough_scheduling_cases
  Metric       : queueing_durations/raf_canvas.html
  Change       : 26.28% | 0.0383834240049 -> 0.0484688099248

Revision             Result                       N
chromium@500221      0.0383834 +- 0.00309673      6      good
chromium@500397      0.0381446 +- 0.00468633      6      good
chromium@500485      0.0373024 +- 0.0055542       6      good
chromium@500491      0.0370018 +- 0.0040638       6      good
chromium@500494      0.037949 +- 0.0067075        6      good
chromium@500495      0.0369637 +- 0.00478069      6      good
chromium@500496      0.0551824 +- 0.0429262       6      bad       <--
chromium@500507      0.0487684 +- 0.00435717      6      bad
chromium@500529      0.0484833 +- 0.0031531       6      bad
chromium@500572      0.0484688 +- 0.00385199      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=raf.canvas.html scheduler.tough_scheduling_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965301161551231616


For feedback, file a bug with component Speed>Bisection

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/fe965894f096ee2dc5c60dcb307164a4fc6be6a3

commit fe965894f096ee2dc5c60dcb307164a4fc6be6a3
Author: Dave Tapuska <dtapuska@chromium.org>
Date: Wed Oct 25 04:08:45 2017

Correct the call sequence for non-blocking events.

The ACK should be sent from the compositor thread for non-blocking
events and it appears it didn't. Ensure that we issue the ack from
the compositor thread when we can. This got caught in unit tests that
I'm converting over to mojo.

BUG=722928, 764346 

Change-Id: Ib86564cb408e1f172efdfb5b7974a7fc11b434c4
Reviewed-on: https://chromium-review.googlesource.com/730870
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Reviewed-by: Mustaq Ahmed <mustaq@chromium.org>
Cr-Commit-Position: refs/heads/master@{#511363}
[modify] https://crrev.com/fe965894f096ee2dc5c60dcb307164a4fc6be6a3/content/renderer/input/widget_input_handler_manager.cc

Some of the peak regression is fixed. The ref and head changes are now very similar.