All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=798346

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=c801d255232b5054b1fa3cc5a9a6b2cccbc95bcd8e875ffdb331363da5e59d89


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

android-nexus5X
android-webview-nexus6
chromium-rel-win7-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8958517090408910432

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

Hi chrishtr@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 : Chris Harrelson
  Commit : d45f384824617447c7e4e8808b4946230dab58e2
  Date   : Sat Dec 30 01:02:37 2017
  Subject: Revert "Revert "[CI] Migrate hit-testing code in PaintLayer to use GeometryMapper.""

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : blink_perf.svg
  Metric       : SvgHitTesting/SvgHitTesting
  Change       : 14.00% | 310.883444444 -> 355.842833333

Revision             Result                  N
chromium@526398      310.883 +- 25.5656      9      good
chromium@526399      340.43 +- 267.228       9      good
chromium@526400      398.031 +- 302.782      9      bad       <--
chromium@526401      400.252 +- 165.28       6      bad
chromium@526404      355.843 +- 5.92936      6      bad

Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

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 blink_perf.svg

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8958517090408910432


For feedback, file a bug with component Speed>Bisection

 Issue 798317  has been merged into this issue.

 Issue 798341  has been merged into this issue.

This regression is caused by updating the lifecycle to PrePaintClean, not
CompositingCleanPlusScrolling, and the time taken for paint invalidation.
This is a tiny overhead that does not actually do any more work than
calling PrePaintTreeWalk::Walk on the root and then returning.

I don't see any thing actionable about the trace in terms of things which
cost more than they should.

The "slowdown" here is due to the micro-nature of the test. The hit tests
are so fast that calling functions affects them. WontFix.

 Issue 799621  has been merged into this issue.