All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=847444

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=bf7fcf28f78d552b00e80179fe486579be4ca774860035b35b8639cec2412bfd


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

android-one

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16a13f9a240000

📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16a13f9a240000

Roll AFDO from 68.0.3439.0_rc-r1 to 68.0.3440.0_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/220bd967ae291ca147ed71f65c2aaf4a772515a5
62.7 → 53.79 (-8.915)

Add experimental echo canceller type constraint to blink interface. by grunell@chromium.org
https://chromium.googlesource.com/chromium/src/+/b117c3ea89ca8946a33cb189eb2248876703062d
53.91 → 53.22 (-0.6879)

Revert "Reland "shared worker: Give the renderer a factory bundle when NetworkService is enabled."" by olka@chromium.org
https://chromium.googlesource.com/chromium/src/+/2d10a8a7b01a727e6692feb8e5f5ccbbf80a596a
53.5 → 53.91 (+0.4111)

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

Looking at the pinpoint graph, it's the AFDO roll that's causing this.

Agreed. Looking at the history, AFDO rolls appear to be a cause of recent instability here beyond just this one drop. I'm investigating general AFDO instability at the moment; will leave this open until that investigation concludes.

Thanks!