All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=701829

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9JbRtQkM


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

chromium-rel-win7-gpu-ati

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8985038875514444192

=== Auto-CCing suspected CL author nednguyen@google.com ===

Hi nednguyen@google.com, 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 : nednguyen
  Commit : 267012ae7453eddde6de2c604e8aa7efed10d67a
  Date   : Mon Mar 13 12:36:50 2017
  Subject: Roll ts_proxy to the latest commit (a2f578febcd79b751d948f615bbde8f6189fbeed)

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : page_cycler_v2.tough_layout_cases
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-warm/http___oilevent.com
  Change       : 3.65% | 115.099000002 -> 119.303555555

Revision                                 Result                  N
chromium@456354                          115.099 +- 8.00703      14      good
chromium@456360                          114.269 +- 2.72877      6       good
chromium@456363                          114.425 +- 8.12451      9       good
chromium@456363,catapult@267012ae74      118.794 +- 10.1369      14      bad       <--
chromium@456364                          117.771 +- 2.68894      9       bad
chromium@456365                          118.925 +- 2.9108       6       bad
chromium@456366                          117.669 +- 1.78475      6       bad
chromium@456377                          119.607 +- 1.36031      6       bad
chromium@456400                          117.935 +- 3.33014      9       bad
chromium@456446                          119.304 +- 7.2695       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 --story-filter=http...oilevent.com page_cycler_v2.tough_layout_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8985038875514444192

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6075772248260608


| 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!

Pat: seems like https://github.com/WPO-Foundation/tsproxy/commit/a2f578febcd79b751d948f615bbde8f6189fbeed create further regressions to windows? :-(

Do you know if there is a way to drill into any of the other metrics (i.e. browse the charts)?  It looks like only a few specific sites went wider and I want to look at the time to onload for those as well to see if that also degraded or if it just changed the timing (more content being fetched before it yields to paint).

That said, those overall times are much lower than I thought we were working with.  Trying to maintain a handful of ms of variation on an 80ms base is going to be hard even if the proxy wasn't written in python.  At a minimum there is going to be a context switch in each direction for any traffic going through the proxy.

Hi Patrick, unfortunately, we can not browse the chart from the dashboard links in #1, and you have to do it manually.

https://chromeperf.appspot.com/report?sid=21c5bbfcfb452250ac02abd3538501fbe791fb517c2cd59928f5b8dd76b55096&start_rev=447695&end_rev=457143 here is the chart for linux release

https://chromeperf.appspot.com/report?sid=1210ff919f58200fdf79a3a7b91c2f69ef4982a0641079f4a7c37e6c50569b67&start_rev=447695&end_rev=457143 is the chart for chromium-rel-win7-gpu-ati

Explictly assigning. A CL you landed tripped one of the speed metrics we measure in the lab. If this is the first time this has happened to one of your CLs, or if it's been a while, please read: https://chromium.googlesource.com/chromium/src/+/master/docs/speed/addressing_performance_regressions.md

We're looking for one of the following:
1. Justification via explanation
2. Plan to revert or fix
3. Angry rage throwing of equipment at my head

Just be aware that I'm trained in trumpet playing and First Aid and am not afraid to use it.

Note: This was a bulk edit message and not very personal.

This is tracked in issue 653519.

The change regressed perf metrics, but:
1) It makes benchmark less crashy.
2) It affect test, not end product.

Hence I didn't revert my change.