All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=753975

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=0213d95f6fab543f50a9b953728f96a96b3bd68d7bc630b39881cf4f018bc82a


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

android-one
chromium-rel-mac-retina
chromium-rel-mac11
chromium-rel-mac11-pro
chromium-rel-mac12

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8968084894301212064

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1482e317b80000

Ugh, bisects didn't kick off automatically :(

Kicking them off now.

📍 Pinpoint job completed.
https://pinpoint-dot-chromeperf.appspot.com/job/1482e317b80000

Roll clang 308728:309984.
By thakis@chromium.org · Sun Aug 06 13:29:39 2017
chromium@c4bc163e7ceb1ffc00fc20efbd92316635b984fe

=== Auto-CCing suspected CL author thakis@chromium.org ===

Hi thakis@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Nico Weber
  Commit : c4bc163e7ceb1ffc00fc20efbd92316635b984fe
  Date   : Sun Aug 06 13:29:39 2017
  Subject: Roll clang 308728:309984.

Bisect Details
  Configuration: android_one_perf_bisect
  Benchmark    : blink_perf.parser
  Metric       : query-selector-all-last/query-selector-all-last
  Change       : 11.47% | 70.9380855319 -> 62.805015994

Revision             Result                   N
chromium@492250      70.9381 +- 0.705816      6      good
chromium@492252      69.0802 +- 1.37004       6      good
chromium@492253      62.805 +- 0.996747       6      bad       <--

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.parser

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8968084894301212064


For feedback, file a bug with component Speed>Bisection

> Roll clang 308728:309984.
> By thakis@chromium.org · Sun Aug 06 13:29:39 2017

We've rolled clang a couple of times since then.
#500608 (https://chromium-review.googlesource.com/656062) seems to have recovered textarea-parsing, but not the others.

I've been trying to look into this today, but it's tricky as Chromium doesn't build with Clang versions from this interval due to a bunch of flag changes that it's come to depend on. I might try some more tomorrow.

hans, any update here?

(Sorry for the slow reply; I was on vacation.)

Most of the graphs seem recovered (or improved) now, except

query-selector-all-last
query-selector-last
query-selector-deep

That clang roll is so long ago that I don't think there's much action we can take here, though. I'll close as wontfix.