All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=731657

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg5r_1-wsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgloXUiQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgloqlrwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgluiB_ggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDglvjjugsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgloPquwsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg5oezkQgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDglsKevwsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgltCYrAgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDglqbY5woM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDglvm7vgsM


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

android-nexus5X
android-nexus7v2
android-webview-nexus5X
chromium-rel-mac-retina
chromium-rel-mac11
chromium-rel-mac12
chromium-rel-win7-gpu-ati
chromium-rel-win7-gpu-intel
chromium-rel-win8-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8977268071149550960

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

Hi fserb@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 : Fernando Serboncini
  Commit : 2e48d743c9c647a66409c98ba96cbe80fd70774b
  Date   : Tue Jun 06 19:49:57 2017
  Subject: Remove CSSTextCache from CSSPrimitiveValue

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : blink_perf.css
  Metric       : CSSPropertySetterGetter/CSSPropertySetterGetter
  Change       : 4.87% | 375.182832747 -> 356.90500877

Revision             Result                  N
chromium@477371      375.183 +- 10.4449      6      good
chromium@477376      369.697 +- 9.81595      6      good
chromium@477378      370.099 +- 4.91142      6      good
chromium@477379      365.164 +- 14.4971      6      good
chromium@477380      348.32 +- 14.4112       6      bad       <--
chromium@477389      350.299 +- 12.5699      6      bad
chromium@477406      345.111 +- 8.87729      6      bad
chromium@477441      343.011 +- 7.64996      6      bad
chromium@477510      356.905 +- 10.1645      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.css

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977268071149550960

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5807709049323520


| 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 Speed>Bisection.  Thank you!

 Issue 731656  has been merged into this issue.

 Issue 731661  has been merged into this issue.

Soooo.

Yesterday I submited https://chromium-review.googlesource.com/527212 which should bring the speed sligthly back up.

That said, I think the tests are very misleading, since they are all hot-cache tests. I'll propose a change to it soon.

If anyone wants to discuss this further, just reopen the bug.

We have cold cache tests, too. FYI.

 Issue 732307  has been merged into this issue.