All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=704336

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgjM__uAkM


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

android-webview-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984380797666279104

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

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : smoothness.gpu_rasterization.tough_path_rendering_cases
  Metric       : frame_times/MotionMark Canvas Fill Shapes

Revision             Result                  N
chromium@458396      77.5516 +- 82.4041      21      good
chromium@458424      83.9021 +- 108.846      21      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=MotionMark.Canvas.Fill.Shapes smoothness.gpu_rasterization.tough_path_rendering_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984380797666279104

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5887637161771008


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984295836307432960

=== Auto-CCing suspected CL author reed@google.com ===

Hi reed@google.com, 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 : Mike Reed
  Commit : 3130fbb1a7aa60ac764f5ccb64a1a0ee976ea127
  Date   : Tue Mar 21 10:11:35 2017
  Subject: remove SK_SUPPORT_LEGACY_DRAWVERTICES_VIRTUAL flag

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : smoothness.gpu_rasterization.tough_path_rendering_cases
  Metric       : frame_times/MotionMark Canvas Fill Shapes
  Change       : 13.12% | 69.4059064706 -> 78.5102254815

Revision                             Result                  N
chromium@458380                      69.4059 +- 2.84479      14      good
chromium@458380,skia@3130fbb1a7      83.5415 +- 85.5706      14      bad       <--
chromium@458381                      96.1401 +- 80.1979      9       bad
chromium@458382                      99.8236 +- 93.0379      9       bad
chromium@458383                      86.8246 +- 39.5571      6       bad
chromium@458386                      101.1 +- 90.8856        14      bad
chromium@458391                      98.2883 +- 78.697       9       bad
chromium@458402                      95.4999 +- 91.9315      14      bad
chromium@458424                      78.5102 +- 79.1002      14      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=MotionMark.Canvas.Fill.Shapes smoothness.gpu_rasterization.tough_path_rendering_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984295836307432960

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4802690841837568


| 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!

The WebView Nexus5X results have always been very noisy on this test; if you extend out the history you'll see that it's bimodal with a wide range. Dunno what's going on there, but I doubt it's reed@'s change, and I wouldn't worry about this regression.