All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=693450

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguLWGtQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguJrWqQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-I6zqgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguPu7pgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguNPTqgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguMqhvgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-MTivQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguOeIsQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-I6zqgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-Pa2sQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguJrWqQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDguPu7pgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-I6zqgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-NzTsgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-MSErAoM


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

android-webview-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8987422020828236896

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

Hi perezju@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 : perezju
  Commit : 19a742985d7b0ca4192614fc7942e57dbb257abc
  Date   : Thu Feb 16 15:41:57 2017
  Subject: [Telemetry] Fix bug in markers for cache temperature

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : page_cycler_v2.intl_es_fr_pt-BR
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-cold/http___www.voila.fr_
  Change       : 98.78% | 1652.50891667 -> 3284.81183333

Revision                                 Result                  N
chromium@451015                          1652.51 +- 229.827      6      good
chromium@451020                          1812.15 +- 496.966      6      good
chromium@451022                          1799.62 +- 500.836      6      good
chromium@451022,catapult@19a742985d      3506.68 +- 773.962      6      bad       <--
chromium@451022,catapult@d885da830d      3508.85 +- 1635.24      6      bad
chromium@451023                          3590.07 +- 1004.93      6      bad
chromium@451024                          3519.74 +- 1155.5       6      bad
chromium@451032                          3320.53 +- 707.581      6      bad
chromium@451049                          3284.81 +- 1054.42      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.voila.fr. page_cycler_v2.intl_es_fr_pt-BR

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987422020828236896

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5274278745866240


| 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!

TL;DR: Not a real regression.

This appears to be the counterpart of  issue 692929 . Here the "cold" measurements improved artificially on the CL introducing the bug and "regressed" when it was fixed. Numbers are now back to their baseline.