All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=707398

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg7I2-lgoM


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

android-nexus5

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983571333481018320

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : v8.runtimestats.browsing_mobile_classic
  Metric       : Total:duration_avg/browse_news/browse_news_reddit


To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.news.reddit v8.runtimestats.browsing_mobile_classic

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983571333481018320

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6284840636252160


| 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!

Mythri, is this real? If not feel free to WontFix.

I think this regression is just noise. The next build is the time when we actually moved the benchmark to the actual --no-turbo (https://codereview.chromium.org/2786633002) where it regresses further. The regression continues because of the change in the flags. Marking it as won't fix.