All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=705405

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgrIbTsQkM


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

chromium-rel-win7-gpu-nvidia

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983986757103014512

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983981566186751040

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

Hi stkhapugin@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 : stkhapugin
  Commit : 3da3338bfc027c0b7a48e5b033feb04f24792b17
  Date   : Thu Mar 23 16:19:51 2017
  Subject: [ObjC ARC] Converts ios/chrome/browser/ui/omnibox:omnibox_internal to ARC.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : long-line-nowrap-spans-collapse/long-line-nowrap-spans-collapse
  Change       : 0.74% | 26.6364444444 -> 26.8347777778

Revision             Result                   N
chromium@459044      26.6364 +- 0.303862      9       good
chromium@459099      26.6057 +- 0.378735      9       good
chromium@459106      26.6676 +- 0.377186      14      good
chromium@459107      26.961 +- 0.263473       6       bad       <--
chromium@459109      27.0495 +- 0.122325      6       bad
chromium@459110      26.7746 +- 0.313498      14      bad
chromium@459113      26.8996 +- 0.272349      9       bad
chromium@459126      26.8264 +- 0.423577      14      bad
chromium@459153      26.8348 +- 0.410124      9       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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983981566186751040

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5292489918906368


| 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!

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

Hi xiaochengh@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 : xiaochengh
  Commit : d53a04e0ac4ab58cb1bc499de297afeae9ba6bac
  Date   : Thu Mar 23 23:51:05 2017
  Subject: List missing header files in editing/BUILD.gn

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : long-line-nowrap-collapse/long-line-nowrap-collapse
  Change       : 6.33% | 25.9276428571 -> 27.5647857143

Revision             Result                  N
chromium@459136      25.9276 +- 2.3817       14      good
chromium@459226      25.9361 +- 1.83992      14      good
chromium@459271      25.7236 +- 1.28986      9       good
chromium@459282      25.8443 +- 2.02308      14      good
chromium@459283      28.0602 +- 2.61519      6       bad       <--
chromium@459284      26.5039 +- 1.98271      14      bad
chromium@459285      26.4143 +- 2.75489      21      bad
chromium@459287      26.5274 +- 2.43067      14      bad
chromium@459293      26.3649 +- 2.02349      14      bad
chromium@459315      27.5648 +- 4.38366      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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983986757103014512

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6183005082615808


| 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!

My CL only changes BUILD.gn and cannot possibly affect performance.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983889295704964992

kicked off another bisect just in case

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

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : long-line-nowrap-collapse/long-line-nowrap-collapse

Revision             Result                  N
chromium@459136      27.7059 +- 5.36225      21      good
chromium@459315      28.455 +- 10.9779       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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983889295704964992

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6467563040538624


| 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!

regression looks not significant in these bisect results.