All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=923381

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=33741ac91be0a94c8a8d16d15818fb61e07f6a471546e35122eebcb4f106f3b5


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

android-nexus5x-perf

blink_perf.canvas - Benchmark documentation link:
  https://bit.ly/blink-perf-benchmarks

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1493b2ce540000

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/1493b2ce540000

Reland^3 "[turbofan] Use feedback when reducing global loads/stores." by neis@chromium.org
https://chromium.googlesource.com/v8/v8/+/8683116e64335a99ecc403c509587410db070cd1
getImageDataColorManaged_RAF: 200 → 10.87 (-189.1)

Roll src/clank b96c64dcd91c..db2ecc5544f7 (1 commits) by chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com
https://chrome-internal.googlesource.com/chrome/src-internal/+/0fa708cf8229324e120bde962826c3e526d9c4d5
getImageDataColorManaged_RAF: 10.92 → 9.509 (-1.413)

Assigning to sheriff jbudorick@google.com because "Roll src/clank b96c64dcd91c..db2ecc5544f7 (1 commits)" is a roll.

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

Benchmark documentation link:
  https://bit.ly/blink-perf-benchmarks

The revert of my CL is in 73.0.3677.0