All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=729914

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgxp-LsAoM


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

chromium-rel-win7-x64-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8977563683179314960

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

Hi rockot@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 : Ken Rockot
  Commit : 77c59a87000dc29ebe28488b9302590081463f6e
  Date   : Wed May 31 01:49:20 2017
  Subject: Mojo EDK: Refactor layered message serialization

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : service_worker.service_worker_micro_benchmark
  Metric       : concurrent_1_response_50_percentile/concurrent_1_response_50_percentile
  Change       : 1.56% | 1.298 -> 1.31821428571

Revision             Result                    N
chromium@475633      1.298 +- 0.184445         20      good
chromium@475697      1.29536 +- 0.057212       14      good
chromium@475730      1.30048 +- 0.0862858      21      good
chromium@475731      1.325 +- 0.0796869        14      bad       <--
chromium@475734      1.33214 +- 0.0896421      14      bad
chromium@475739      1.35083 +- 0.131798       6       bad
chromium@475746      1.3225 +- 0.0372492       14      bad
chromium@475761      1.31821 +- 0.0461558      14      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests service_worker.service_worker_micro_benchmark

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977563683179314960

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5862038540124160


| 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!

Seems unlikely that this change would have any significant impact on a SW microbenchmark performance, especially given that there's no consistent regressions elsewhere. I've reported the bisect as bad for now.