All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=811330

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=ba2d3e33454740f0cb0c11c87990f553fa32d8b4405817b39ab476a7b4dfc497


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

android-webview-nexus6

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14b9303d840000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14b9303d840000

+dtu, simonhatch: looks like there are no devices attached on webview? Can one of you take a look and re-run the bisect?

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14a66875840000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14a66875840000

Sync Compositor Android Webview ensure layer_tree_frame_sink_id is assigned. by dtapuska@chromium.org
https://chromium.googlesource.com/chromium/src/+/9810b817470d91e03fa3500b5af3649c67645731

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

This is a bug fix. If you look at the graph you'll see a drop (when the original bug was introduced) and then an uptick when this change was introduced fixing the original bug. The graphs before the original change and after this fix are generally flat.