All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=645370

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgydqmvgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgyfi-ugkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8ZfDjQgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgyeqWqQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgye3jvwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgycaftgoM


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

chromium-rel-mac-retina
chromium-rel-mac10
chromium-rel-mac11

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9002017845444139392

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

Hi foolip@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Move HTMLDocument::isCaseSensitiveAttribute into SelectorChecker
Author  : foolip
Commit description:
  
BUG=238368
R=esprehn@chromium.org,rune@opera.com

Review-Url: https://codereview.chromium.org/2313253002
Cr-Commit-Position: refs/heads/master@{#416903}
Commit  : e912108e9daf83d1bec61e5b78aaacf965a3e2fc
Date    : Wed Sep 07 11:01:59 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@416888  190.431  2.82441   5  good
chromium@416897  184.13   1.36015   5  good
chromium@416902  188.514  6.48512   5  good
chromium@416903  147.306  1.4117    5  bad    <--
chromium@416904  147.042  0.791751  5  bad
chromium@416906  146.68   1.72469   5  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 645370

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.parser
Test Metric: query-selector-all-attribute/query-selector-all-attribute
Relative Change: 22.97%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/881
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9002017845444139392


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6353045984641024

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

I'll try reverting to see if that fixes the perf regression, it was just refactoring anyway.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/fac211fb17e55913cadfae4344f23b056066c7f6

commit fac211fb17e55913cadfae4344f23b056066c7f6
Author: foolip <foolip@chromium.org>
Date: Fri Sep 09 13:22:56 2016

Revert of Move HTMLDocument::isCaseSensitiveAttribute into SelectorChecker (patchset #1 id:1 of https://codereview.chromium.org/2313253002/ )

Reason for revert:
Possible cause of regression in blink_perf.parser.

BUG= 645370 

Original issue's description:
> Move HTMLDocument::isCaseSensitiveAttribute into SelectorChecker
>
> BUG=238368
> R=esprehn@chromium.org,rune@opera.com
>
> Committed: https://crrev.com/e912108e9daf83d1bec61e5b78aaacf965a3e2fc
> Cr-Commit-Position: refs/heads/master@{#416903}

TBR=esprehn@chromium.org,rune@opera.com
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=238368

Review-Url: https://codereview.chromium.org/2326493004
Cr-Commit-Position: refs/heads/master@{#417571}

[modify] https://crrev.com/fac211fb17e55913cadfae4344f23b056066c7f6/third_party/WebKit/Source/core/css/SelectorChecker.cpp
[modify] https://crrev.com/fac211fb17e55913cadfae4344f23b056066c7f6/third_party/WebKit/Source/core/html/HTMLDocument.cpp
[modify] https://crrev.com/fac211fb17e55913cadfae4344f23b056066c7f6/third_party/WebKit/Source/core/html/HTMLDocument.h

https://chromeperf.appspot.com/group_report?bug_id=645370 shows the graph bouncing back with the revert in the range, so that was the reason. Won't investigate further, but will comment on issue 238368.