All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=836469

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=d131d09b97c8790485fda56880a5bbac7adcda481f79e02e33fa3654a117b8cc


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

chromium-rel-mac11-air
chromium-rel-mac12
chromium-rel-win7-gpu-ati
chromium-rel-win7-gpu-intel

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16bce839c40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16bce839c40000

Make --site-per-process the default on ToT via fieldtrial_testing_config by lukasza@chromium.org
https://chromium.googlesource.com/chromium/src/+/fb1ccf02ee8ca79e1404abfd3a3a7d540b7d2dbd

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

nasko@ has kindly offered to help investigate if this regression report offers any useful insights (e.g. via a scenario that highlights/exaggerates a perf regression so that they are visible via chrome tracing).  If this investigation doesn't yield any results, then we can just dupe against the known loading regression tracked in issue 808114.

 Issue 838439  has been merged into this issue.

 Issue 839370  has been merged into this issue.

 Issue 836533  has been merged into this issue.

This bug might be fixed by r587329 (i.e. it might be a dupe of  issue 560446 ), but I am not sure how to double-check that (e.g. if there is a way to ask performance bots for help with this verification)...

Issue 838441 has been merged into this issue.

Looking at the graph over time, it has continued staying at the higher end, though has gone up/down and currently varies quite a bit. We have fixed some regressions (priority inversion  issue 560446 ) and at this time I don't think there is anything else we are planning to investigate. I think resolving as duplicate of issue 808114 makes sense at this time.

To answer comment 14, I don't see that revision making any difference on the performance graph for the original graph cited in this bug.

Resolving as duplicate to 808114. If someone thinks this is incorrect, feel free to reopen.

 Issue 841244  has been merged into this issue.