All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=883810

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=05ae84ea0d79c97b1525976761c780ab89564b9e0f2b857c3e234de0933256c7


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

Android Nexus5 Perf

system_health.common_mobile - Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/13de1407640000

The last try seems to detect r589877 as a suspicious CL, but for some reasons it wasn't reported here?

Let me run another bisect just in case.

cc alexilin@chromium.org who submitted the r589877

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/12b7685ce40000

The last pinpoint job has ended the same way. It seems my CL is a culprit indeed. I'll take a look.

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/12b7685ce40000

custom elements: Replace ElementDefinitionIsRunning with base::AutoReset. by yukishiino@chromium.org
https://chromium.googlesource.com/chromium/src/+/6bb20b5e1c97de7314331e3912fe0768eb65749d
timeToFirstPaint: 1425 → 1374 (-51.04)

predictors: Request a proxy lookup in parallel with host resolve request by alexilin@chromium.org
https://chromium.googlesource.com/chromium/src/+/248c267fc1c931ba1863d72bb1ddbd45f8680218
timeToFirstPaint: 1374 → 1597 (+223)

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

Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

My change "custom elements: Replace ElementDefinitionIsRunning with base::AutoReset." is a pretty simple and trivial refactoring and shouldn't be affecting performance.

Looking at "timeToFirstPaint: 1425 → 1374 (-51.04)", it seems that my patch is irrelevant.