All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=883747

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=3de0d7b26e3fbd99a12d84675e9d93ae52722d24c596fba75f720249d69d6dce


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

Android Nexus5 Perf

rendering.mobile - Benchmark documentation link:
  https://bit.ly/rendering-benchmarks

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14df2413640000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14df2413640000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1508115b640000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1508115b640000

Use GPU scheduler for in process command buffer by sunnyps@chromium.org
https://chromium.googlesource.com/chromium/src/+/1405fd6a6c18ce8896ab031005b570d96cfda9fb
19.51 → 20.36 (+0.8427)

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

Benchmark documentation link:
  https://bit.ly/rendering-benchmarks

tasks_per_frame* regressions are expected with gpu scheduler, since it can post task to switch from a low priority to a high priority context.

The frame_times/fps regressions have been fixed by r594573.