All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=753249

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=203ba8554efe25bca1caafafdf19ed58567ff57c0d28a408ac6fabed4494752e


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

android-webview-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971854198429507552

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

Hi sunxd@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 : sunxd
  Commit : 7c9302475a38992e7555d7ce8496ea1da797f4bc
  Date   : Wed Aug 02 21:44:28 2017
  Subject: cc: Use render surface draw transform when occluding mask layer.

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : v8.runtimestats.browsing_mobile
  Metric       : JavaScript:duration_avg/browse_social/browse_social_instagram
  Change       : 7.77% | 5175.0935 -> 5577.40766667

Revision             Result                  N
chromium@491365      5175.09 +- 131.88       6      good
chromium@491467      5201.98 +- 219.521      6      good
chromium@491493      5121.68 +- 174.881      6      good
chromium@491506      5160.05 +- 215.732      6      good
chromium@491512      5151.37 +- 177.864      6      good
chromium@491513      5046.14 +- 198.779      6      good
chromium@491514      5587.75 +- 411.325      6      bad       <--
chromium@491515      5506.89 +- 117.339      6      bad
chromium@491518      5521.76 +- 217.54       6      bad
chromium@491569      5577.41 +- 226.579      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.social.instagram v8.runtimestats.browsing_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8971854198429507552


For feedback, file a bug with component Speed>Bisection

I believe this is a result of ignoring mask layer occlusion, hi enne@, do we want to compute the draw transform and occlusion for masks?

Since enne@ is ooo, cc folks for advice. The mentioned change is to ignore occlusion for mask layers though description is using surface's draw transform, I forgot to change the description after being suggested to ignore occlusion, sorry.

What is this perf bot measuring?

After a second thought I think it's expected to get this regression.

I think it's measuring the time spent by v8 in different phases with this story: https://cs.chromium.org/chromium/src/tools/perf/page_sets/system_health/browsing_stories.py?rcl=9c1334b866d962571bd90320cb9e1232c0d266dd&l=138.

Recall the original bug, mobile instagram site has a border radius signup button, and that was wrongly occluded, now we do not occlude that.

 Issue 754244  has been merged into this issue.

I'm going to investigate whether the mask occlusion caused this regression.