All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=713961

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgopqRpwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgwsaYuAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDggp_O6wgM


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

chromium-rel-win7-gpu-intel

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8981745089526116272

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : media.tough_video_cases
  Metric       : vm_working_set_delta_size/crowd2160.mp4_gpu

Revision             Result                   N
chromium@464039      6801.71 +- 37637.9       21      good
chromium@464153      -397.714 +- 48743.1      21      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 media.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981745089526116272

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5791609926451200


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8981720540606840128

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

Hi yzshen@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 : yzshen
  Commit : ee80d405f8ea5038f89c41c0603f1538ceeb7f14
  Date   : Wed Apr 12 19:51:47 2017
  Subject: Network service: add URLLoaderFactory implementation.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : media.tough_video_cases
  Metric       : vm_working_set_delta_size/crowd2160.mp4_gpu
  Change       : 2.38% | 22812.0 -> 23354.6666667

Revision             Result                  N
chromium@464039      22812.0 +- 228.666      6      good
chromium@464096      22808.0 +- 279.027      6      good
chromium@464111      22785.3 +- 117.394      6      good
chromium@464115      22838.0 +- 349.937      6      good
chromium@464116      22908.7 +- 238.087      6      good
chromium@464117      23294.7 +- 297.438      6      bad       <--
chromium@464118      23264.7 +- 65.8281      6      bad
chromium@464125      23300.0 +- 202.148      6      bad
chromium@464153      23354.7 +- 122.202      6      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 media.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981720540606840128

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5791609926451200


| 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!

I am pretty sure that the CL is irrelevant. That CL implements something that doesn't take effect unless --enable-network-service is specified. The implementation doesn't touch any code outside of the behind-flag feature. And currently nothing on the bots enables this flag.

Would you please find a better owner for this bug? Thanks!
(But please let me know if you have reasons how my CL could be involved.) 

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8981668408200185632

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : yzshen
  Commit : ee80d405f8ea5038f89c41c0603f1538ceeb7f14
  Date   : Wed Apr 12 19:51:47 2017
  Subject: Network service: add URLLoaderFactory implementation.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : media.tough_video_cases
  Metric       : vm_working_set_delta_size/smpte_3840x2160_60fps_vp9.webm_gpu
  Change       : 2.32% | 22745.3333333 -> 23273.3333333

Revision             Result                  N
chromium@464111      22745.3 +- 201.09       6      good
chromium@464115      22764.7 +- 215.66       6      good
chromium@464116      22872.0 +- 230.755      6      good
chromium@464117      23271.3 +- 144.601      6      bad       <--
chromium@464118      23339.3 +- 250.466      6      bad
chromium@464125      23273.3 +- 211.484      6      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 media.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981668408200185632

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5860406846816256


| 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!

Unable to find the CL causing the regression, and the regression is minor. Ignoring it.