All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=867998

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=0680975aefa29bb6f0ea21b753aabe5f1605bfecb854382e6544670d6cb23572


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

Android Nexus6 WebView Perf

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/139e7617a40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/139e7617a40000

Add gn arg to optionally compile webgl2-compute by jiajia.qin@intel.com
https://chromium.googlesource.com/chromium/src/+/df3a31b23ff39bf90b64a7f0ec35b74f817319b6
15.02 → 14.81 (-0.2154)

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

It's weird for me that the regression is caused by my change 'Add gn arg to optionally compile webgl2-compute'. That change is mainly to not compile webgl2-compute related files and IDLs for android. Logically, I think it will reduce the binary size and maybe bring performance improvement not regression for android platform. 

It's not possible for Jiajia's CL to have changed the performance of this benchmark. Looking at the graphs, a more accurate regression range would be r576072 - r576236:

https://chromium.googlesource.com/chromium/src/+log/bf09dc2ba2aeca294df763f9b29624c2bc19a7e7%5E..e82d10e0d449b3bf131f00d6a6f997296c2b0bca?pretty=fuller&n=1000

Available, but no owner or component? Please find a component, as no one will ever find this without one.