All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=783812

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=d1ad27b91712d2cde97eb4f44c9dc73b006fadf637cf5427259d515cd8ce88cf


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

chromium-rel-mac12

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14b44441f80000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14b44441f80000

Revert "Make script cache hits async"
By kouhei@chromium.org · Thu Nov 09 08:23:01 2017
chromium @ e1cfaee91ad08478fbd956fcd73d570032702397

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

kouhei: This looks like a pretty significant performance regression, any update here?

There seems at least two root causes here:

1. Revert "Make script cache hits async" (Comment #3).

Original CL was:
Make script cache hits async 1aec1a92fb786cae0a24ff2b22d4858d4486252e (r514299)
and it was reverted by e1cfaee91ad08478fbd956fcd73d570032702397 (r515121) due to performance regression.

According to some graphs e.g.:
https://chromeperf.appspot.com/report?sid=5d0492775d5905ab465153eb332ede3ea7e2eafecd7a23c28f02b3ec194d56f8&rev=515134
https://chromeperf.appspot.com/report?sid=3c5d694d25510f46d6fe1fde8b86daed0ab3584b36211c1f09cc4f9d4c2ebc13&rev=515136

The original CL improved some performance metrics here (and also regressed other metrics and thus resulted in the revert), and its revert regressed the performance.
I think this is WontFix.

2.
Some of the graphs listed in Comment #1 has regression ranges after r515121, and don't exhibit previous regressions, so there seems another root cause for these.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14f3a0c4840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/12eadfe4840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14f066b4840000

Examples for 2. are
https://chromeperf.appspot.com/report?sid=07d9ad9b6c6628f461052a9067214c154909ff2b1ef13119f50df79ffbbb64b6&rev=515154
https://chromeperf.appspot.com/report?sid=14081135414378d68a61a4f930ce488546d438239d33f7946d7705f626d5c56b&rev=515167
https://chromeperf.appspot.com/report?sid=f35caaba2388a07a6f4202a3024235730f8969b7bb3aea88b1463b8b530c0de5&rev=515171

Suspected range is 515130 - 515154. Started bisects.

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/12eadfe4840000

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/14f066b4840000

📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14f3a0c4840000

Revert "Make script cache hits async"
By kouhei@chromium.org · Thu Nov 09 08:23:01 2017
chromium @ e1cfaee91ad08478fbd956fcd73d570032702397

Update wprgo binaries to latest version
By nednguyen@google.com · Tue Nov 07 14:06:23 2017
catapult @ 564e33d345ee080be36ad0e6b6107afbcd09df4d

Roll src/third_party/catapult/ af046c52f..f83c79a6d (14 commits)
By catapult-deps-roller@chromium.org · Thu Nov 09 12:29:51 2017
chromium @ 2ecc1eb4754c04bea4d1fc0957ec0d9bd9dce2de

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

Sorry about those errored jobs, I deployed a bad build of Pinpoint yesterday. I'll re-kick  them.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/129c238c840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16b7f104840000

Looking at the successful result, comment 12: When it blames the catapult roll, it means the last commit in the roll, which is:
Reland: devil: Restart adb server and try again when NoDevicesError is thrown.
catapult @ f83c79a6db050a36477119bcbda8d7245839197e

looks like the bulk of the regression is from nednguyen's change, a smaller portion from bpastene's change, and a tiny amount from kouhei's change.

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16b7f104840000

Update wprgo binaries to latest version
By nednguyen@google.com · Tue Nov 07 14:06:23 2017
catapult @ 564e33d345ee080be36ad0e6b6107afbcd09df4d

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

My change is expected to cause regression. What should we do about this bug given that other commits also cause some regressions?

📍 Found a significant difference after 1 commit.
https://pinpoint.chromeperf.appspot.com/job/129c238c840000

Update wprgo binaries to latest version
By nednguyen@google.com · Tue Nov 07 14:06:23 2017
catapult @ 564e33d345ee080be36ad0e6b6107afbcd09df4d

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

Most of the regression is due to my change & it's a test change, so not affecting the Chrome product.