All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=773716

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=39853bc92a9a489d832187afd380ac99e41cc00c84d4cf9a40ca27a8ec0fb477


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

chromium-rel-win7-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8966016987040359520

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

Hi mtklein@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 : Mike Klein
  Commit : a3b889514f4f7e83ece709f4ba11156ab1711cee
  Date   : Thu Oct 05 18:16:44 2017
  Subject: clamp to [0,1] in all gradient tilers

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : smoothness.tough_canvas_cases
  Metric       : frame_times/http___www.smashcat.org_av_canvas_test_
  Change       : 2.74% | 29.5038623246 -> 28.6546187844

Revision                             Result                   N
chromium@506801                      29.5039 +- 0.353075      6       good
chromium@506849                      29.2878 +- 0.317147      6       good
chromium@506873                      29.1899 +- 0.31452       6       good
chromium@506885                      29.2259 +- 0.300868      9       good
chromium@506891                      29.283 +- 1.61945        9       good
chromium@506893                      29.1057 +- 0.450657      6       good
chromium@506893,skia@454818b80a      29.1034 +- 0.791399      14      good
chromium@506893,skia@eccda1ceca      29.3385 +- 0.436533      6       good
chromium@506893,skia@a3b889514f      28.7201 +- 1.71418       21      bad       <--
chromium@506893,skia@38ace8a133      28.6576 +- 0.495576      14      bad
chromium@506894                      28.4765 +- 0.647096      6       bad
chromium@506896                      28.6546 +- 0.715534      9       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=http...www.smashcat.org.av.canvas.test. smoothness.tough_canvas_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966016987040359520


For feedback, file a bug with component Speed>Bisection

Hmm, that's unfortunate, but we'd really better keep that patch in.  It's best not to read pointers at infinity...

cc test owner junov as FYI