All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=903121

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=e118e67010a20a033888aee11de57f34ef75126dd967860d6f50e8589d312912


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

mac-10_13_laptop_high_end-perf

memory.desktop - Benchmark documentation link:
  None

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/13cf26e1e40000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/13cf26e1e40000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1728e18de40000

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/1728e18de40000

gpu: Fix raster implementation error for no flush result available query by khushalsagar@chromium.org
https://chromium.googlesource.com/chromium/src/+/da82bcce69bf5384435976a1945c9cb3270c5abb
memory:chrome:all_processes:reported_by_chrome:cc:effective_size: 1.101e+08 → 6.533e+07 (-4.479e+07)

viz: Throttle begin-frames when gpu is busy. by sadrul@chromium.org
https://chromium.googlesource.com/chromium/src/+/5c2e137dc5294b30492a05a96a33bf0c7ff12908
memory:chrome:all_processes:reported_by_chrome:cc:effective_size: 6.581e+07 → 8.216e+07 (+1.635e+07)

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

Benchmark documentation link:
  None

Recovered in range 603894 - 603928 (possibly because of 603921)