All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=760165

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=c4fe277087aec088de93807bbe83a87edef729d1b5b6205e66dc41f64378a374


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

chromium-rel-win7-gpu-ati

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8969912310962286320

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

Hi yolandyan@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 : Yoland Yan
  Commit : 0cc5d0ea79f9c6c97102d04fbe7c46f327904ced
  Date   : Sat Aug 19 04:19:04 2017
  Subject: Convert CrashTestCase and its children to JUnit4

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : loading.desktop
  Metric       : cpuTimeToFirstMeaningfulPaint_avg/pcv1-warm/FIFA
  Change       : 18.31% | 284.694428571 -> 315.780857143

Revision             Result                  N
chromium@495794      284.694 +- 27.2427      14      good
chromium@495800      285.958 +- 92.781       14      good
chromium@495801      306.331 +- 116.297      21      bad       <--
chromium@495804      304.952 +- 86.7128      14      bad
chromium@495814      299.321 +- 60.8875      14      bad
chromium@495832      315.781 +- 111.234      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 --story-filter=FIFA loading.desktop

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8969912310962286320


For feedback, file a bug with component Speed>Bisection

ugh, my CL is for Android java test change, seems unrelated?

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8969832400889765360

Sorry, gradual increase and bisect probably just picked up an outlier.  Kicking off a bisect for the first rise in the trend.

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

Suspected Commit
  Author : perezju
  Commit : 122dd5e91bf64888049459ce73a8fd34c084cfe0
  Date   : Mon Aug 14 12:28:51 2017
  Subject: [Telemetry] Reuse shared state between story runs

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : loading.desktop
  Metric       : cpuTimeToFirstMeaningfulPaint_avg/pcv1-warm/FIFA
  Change       : 17.12% | 233.007166667 -> 272.887666667

Revision                                 Result                  N
chromium@494003                          233.007 +- 24.9477      6      good
chromium@494046                          236.051 +- 29.6276      6      good
chromium@494046,catapult@122dd5e91b      287.057 +- 82.442       6      bad       <--
chromium@494047                          274.004 +- 9.62453      6      bad
chromium@494048                          282.99 +- 66.4831       6      bad
chromium@494049                          299.429 +- 76.586       6      bad
chromium@494052                          274.751 +- 13.7726      6      bad
chromium@494057                          305.158 +- 101.681      6      bad
chromium@494068                          296.593 +- 77.8286      6      bad
chromium@494089                          294.013 +- 77.6066      6      bad
chromium@494175                          272.888 +- 13.1083      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 --story-filter=FIFA loading.desktop

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8969832400889765360


For feedback, file a bug with component Speed>Bisection