All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=758218

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=b018360d9b215c3bbc6f0c36f46a70bd107cf9994eca857da985ae2de9180e60


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

chromium-rel-mac-retina

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8970458363568620960

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

Hi yosin@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 : Yoshifumi Inoue
  Commit : 8aa74e38ab9cbd0f3c5515852182d069ff6af2d4
  Date   : Thu Aug 17 01:41:14 2017
  Subject: Make SelectionController::HandleMouseReleaseEvent() to call SelectionInFlatTree::Builder::Build() once

Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : blink_perf.bindings
  Metric       : set-attribute-rare/set-attribute-rare
  Change       : 10.27% | 740.431123558 -> 664.379017539

Revision             Result                  N
chromium@494990      740.431 +- 4.57393      6      good
chromium@495024      739.102 +- 1.49779      6      good
chromium@495033      727.624 +- 10.3533      6      good
chromium@495037      730.286 +- 4.13627      6      good
chromium@495038      726.59 +- 9.70071       6      good
chromium@495039      690.151 +- 156.564      6      bad       <--
chromium@495041      684.515 +- 134.758      6      bad
chromium@495058      707.607 +- 106.574      6      bad
chromium@495126      654.833 +- 4.25432      6      bad
chromium@495262      664.379 +- 3.24404      6      bad

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

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8970458363568620960


For feedback, file a bug with component Speed>Bisection

Mark "Untriaged" to pick this bug again by perf-team since my patch mentioned in #c3
changes mouse release event but the test doesn't relate to mouse release event. The
test measures Node#setAttribute().

PerfTestRunner.measureRunsPerSecond({
    description: "This benchmark covers 'setAttribute' in Dromaeo/dom-attr.html and other DOM methods that return an undefined.",
    run: function() {
        var localDiv = div;
        for (var i = 0; i < 10000; i++)
            localDiv.setAttribute("role", "group");
}});

Regression recovered.