All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=844626

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=c3a1bf09665d991cf70caf325672060e68651d99eca21a3cc2129a95b8497140


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

android-nexus5

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16e48914240000

📍 Found significant differences after each of 7 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16e48914240000

[Autofill] Clear should only clear a section. by parastoog@google.com
https://chromium.googlesource.com/chromium/src/+/7d8ea06c72d4c95e10cee63f1cfde8b8d7db50b5

[LayoutNG] Fix border/background painting for multi-line by eae@chromium.org
https://chromium.googlesource.com/chromium/src/+/71426b9aa50bd9faebb73540dcbcf55d315620e4

Disable PAC scipt file URLs by jcivelli@google.com
https://chromium.googlesource.com/chromium/src/+/8d523527ff00cfdeaf792a52a3a174d7207c2c16

Revert "Alias 4KB at the beginning of JSONParser::ConsumeDictionary() to help diagnose issue 791487." by gab@chromium.org
https://chromium.googlesource.com/chromium/src/+/cba496196fbc8e1c67fdce6d9f121e6b75ed3a4d

Have gesture pinch layout tests use a compositor by mcnee@chromium.org
https://chromium.googlesource.com/chromium/src/+/202183d097a49169f357e3d843f0878e1e1aa3ae

Add UMA to record encryption scheme used by HTML5 video by jrummell@chromium.org
https://chromium.googlesource.com/chromium/src/+/db5a7efe37ac6f283f8dfc411f5f0fe57cc082a9

Worker: Throw an ErrorEvent on a failure of module loading by nhiroki@chromium.org
https://chromium.googlesource.com/chromium/src/+/197a3b25666c262be7dff17eeb2d959dd6dc8b84

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16f2711a240000

Sorry for my late reply. I don't think my change affected the perf test because it changed the code only for module scripts on WebWorkers. It's still a flagged feature and not perf-sensitive.

I restarted the bisect job...

📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16f2711a240000

[Autofill] Clear should only clear a section. by parastoog@google.com
https://chromium.googlesource.com/chromium/src/+/7d8ea06c72d4c95e10cee63f1cfde8b8d7db50b5

[LayoutNG] Fix border/background painting for multi-line by eae@chromium.org
https://chromium.googlesource.com/chromium/src/+/71426b9aa50bd9faebb73540dcbcf55d315620e4

Disable PAC scipt file URLs by jcivelli@google.com
https://chromium.googlesource.com/chromium/src/+/8d523527ff00cfdeaf792a52a3a174d7207c2c16

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

This regression was probably caused by the LayoutNG change but that change has been reverted and the graph looks back to normal.