All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=741729

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=9d5bd2126a0db9402bca77afffabfb1ae672d5362ad1b141ed311bc32a6c172b


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

android-nexus5
android-nexus7v2

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8974256369437150464

=== 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 : Xianda Sun
  Commit : 0d09713458f374a911822779fa48bae96b6608ed
  Date   : Thu Jul 06 22:09:57 2017
  Subject: Revert "Revert "cc: Enable mask tiling""

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : system_health.common_mobile
  Metric       : cpu_time_percentage_avg/load_news/load_news_irctc
  Change       : 5.86% | 1.42624984285 -> 1.50976510931

Revision             Result                    N
chromium@484636      1.42625 +- 0.0198474      6      good
chromium@484695      1.4334 +- 0.0274594       6      good
chromium@484724      1.43336 +- 0.0335303      6      good
chromium@484736      1.4406 +- 0.0232442       6      good
chromium@484744      1.43206 +- 0.0131291      6      good
chromium@484749      1.44446 +- 0.0382097      6      good
chromium@484750      1.43828 +- 0.0385613      6      good
chromium@484751      1.49786 +- 0.0364924      6      bad       <--
chromium@484753      1.50927 +- 0.040521       6      bad
chromium@484862      1.50977 +- 0.0428683      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=load.news.irctc system_health.common_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8974256369437150464


For feedback, file a bug with component Speed>Bisection

The CL was intended to save memory on extreme cases where there are large mask layers. The CL introduces tiling mask layer logic, so now the renderer needs to process multiple render pass draw quads instead of only one. I think the regression is expected.

 Issue 741706  has been merged into this issue.

WontFix-ing per #4, but ccing owner of loading benchmarks as FYI