All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=847397

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=5a9d3cb413eabe0f2892e64c7e9442ae7536e5721a7b055962b2cb06a5ed9be9


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

android-webview-nexus6

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14bfdc86240000

📍 Found significant differences after each of 7 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14bfdc86240000

Roll AFDO from 68.0.3434.2_rc-r1 to 68.0.3435.0_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/bee75214fd4f6cb093c07473b1a4ff538396442d
9573 → 9833 (+259.3)

[RLS] Remove paint properties from LocalFrameView by pdr@chromium.org
https://chromium.googlesource.com/chromium/src/+/c75c3661a3a1e277204063d6607d302bff8c5554
9809 → 9762 (-46.47)

Mojo: Invitation API by rockot@chromium.org
https://chromium.googlesource.com/chromium/src/+/a523cf48c629d7327cde4a5f36098f47ad50a696
9743 → No values

Revert "Mojo: Invitation API" by rockot@chromium.org
https://chromium.googlesource.com/chromium/src/+/6958662bf45564a48de067bf6c906fb8f890a386
No values → 9641

Force visual properties to be synced before cross-process postMessage. by alexmos@chromium.org
https://chromium.googlesource.com/chromium/src/+/fb9795395ef6fbdccde33f5fbcbf5e8c33010658
9641 → 9782 (+140.1)

[cssom] Make CSSOM APIs always append to the declaration block. by emilio@chromium.org
https://chromium.googlesource.com/chromium/src/+/9d45b94d610d27094ccfd07d20b7283813fa0bf6
9762 → 4944 (-4818)

Roll AFDO from 68.0.3435.0_rc-r1 to 68.0.3436.0_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/39d2b84a545bd1c6cf6bcd604d3caab1bdb4f4d6
4943 → 4905 (-38.14)

Assigning to sheriff gbiv@chromium.org because "Roll AFDO from 68.0.3435.0_rc-r1 to 68.0.3436.0_rc-r1" is a roll.

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

The AFDO fluctuations here seem reasonable for Blink. The big drop appears to be caused by r560180, so I'm punting this to emilio.

Yup, this is effectively the same as issue 845946.