All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=766564

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=0bf9e609e7dd231361f4460db9d72d807d1002e44471c6b9a9f9a2fbebf16f03


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

chromium-rel-mac11
chromium-rel-mac12-mini-8gb

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8968030575702865648

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

Hi mtklein@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 : Mike Klein
  Commit : 1882524e81c57ff9ee4061b0473ee30666b560c6
  Date   : Fri Sep 15 15:17:19 2017
  Subject: Turn on unifried _lowp stages.

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : rasterize_and_record_micro.top_25
  Metric       : rasterize_time/file___static_top_25_linkedin.html
  Change       : 2.79% | 1.21983333333 -> 1.25383333333

Revision             Result                     N
chromium@502258      1.21983 +- 0.0102388       6      good
chromium@502259      1.22117 +- 0.0118673       6      good
chromium@502260      1.24833 +- 0.0104563       6      bad       <--
chromium@502261      1.24767 +- 0.00522813      6      bad
chromium@502264      1.25017 +- 0.00805191      6      bad
chromium@502270      1.24917 +- 0.00817517      6      bad
chromium@502281      1.25467 +- 0.00945163      6      bad
chromium@502303      1.25167 +- 0.0111056       6      bad
chromium@502348      1.25383 +- 0.0140297       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 --story-filter=file...static.top.25.linkedin.html rasterize_and_record_micro.top_25

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8968030575702865648


For feedback, file a bug with component Speed>Bisection

It's not surprising to see small regressions here.  This was largely an effort to reduce code complexity to allow future improvements, but it was a fairly large code generation change for x86 (and a modest one for ARM).  Some things will get faster, some maybe a little slower like here, and most a touch higher quality.

It is nice to know the perf tools are this sensitive, though.  :)