All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=706952

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgnPbNpwoM


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

chromium-rel-mac11

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983670081479605840

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

Hi samans@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 : samans
  Commit : 09812d36f72be34ba5e68441222750b4b0a6b3a9
  Date   : Mon Mar 27 19:57:43 2017
  Subject: RendererCompositorFrameSink should receive BeginFrame messages

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : page_cycler_v2.typical_25
  Metric       : timeToFirstContentfulPaint_avg/pcv1-cold/http___www.flickr.com_search_?q_monkeys_f_hp
  Change       : 39.28% | 160.993714286 -> 224.226857143

Revision             Result                  N
chromium@459776      160.994 +- 130.823      14      good
chromium@459821      145.667 +- 7.29382      6       good
chromium@459844      150.649 +- 16.7109      6       good
chromium@459855      147.915 +- 10.7713      9       good
chromium@459858      148.162 +- 19.7529      9       good
chromium@459859      238.123 +- 24.9958      6       bad       <--
chromium@459860      219.304 +- 64.093       6       bad
chromium@459861      216.042 +- 139.775      14      bad
chromium@459866      224.227 +- 121.037      14      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.flickr.com.search..q.monkeys.f.hp page_cycler_v2.typical_25

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983670081479605840

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6353317816434688


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983654611756038000

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

Suspected Commit
  Author : samans
  Commit : 09812d36f72be34ba5e68441222750b4b0a6b3a9
  Date   : Mon Mar 27 19:57:43 2017
  Subject: RendererCompositorFrameSink should receive BeginFrame messages

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : page_cycler_v2.typical_25
  Metric       : timeToFirstContentfulPaint_avg/pcv1-cold/http___www.flickr.com_search_?q_monkeys_f_hp
  Change       : 44.85% | 158.943333333 -> 230.231555555

Revision             Result                  N
chromium@459776      158.943 +- 82.9474      9       good
chromium@459821      147.987 +- 5.53034      9       good
chromium@459844      157.069 +- 77.9952      9       good
chromium@459855      153.089 +- 90.5445      14      good
chromium@459858      156.776 +- 91.5328      9       good
chromium@459859      222.73 +- 69.9361       9       bad       <--
chromium@459860      222.155 +- 73.2236      9       bad
chromium@459861      238.612 +- 34.4614      9       bad
chromium@459866      230.232 +- 94.7449      9       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.flickr.com.search..q.monkeys.f.hp page_cycler_v2.typical_25

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983654611756038000

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4588999743635456


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983645348837955888

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

Suspected Commit
  Author : samans
  Commit : 09812d36f72be34ba5e68441222750b4b0a6b3a9
  Date   : Mon Mar 27 19:57:43 2017
  Subject: RendererCompositorFrameSink should receive BeginFrame messages

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : page_cycler_v2.typical_25
  Metric       : timeToFirstContentfulPaint_avg/pcv1-cold/http___www.flickr.com_search_?q_monkeys_f_hp
  Change       : 54.45% | 150.202333333 -> 231.989666666

Revision             Result                  N
chromium@459776      150.202 +- 18.9385      6       good
chromium@459821      147.335 +- 9.50442      6       good
chromium@459844      148.363 +- 6.90155      6       good
chromium@459855      147.316 +- 7.79695      9       good
chromium@459858      154.831 +- 83.0093      9       good
chromium@459859      220.87 +- 110.853       14      bad       <--
chromium@459860      228.512 +- 94.2889      9       bad
chromium@459861      226.278 +- 86.3075      9       bad
chromium@459866      231.99 +- 27.9981       6       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.flickr.com.search..q.monkeys.f.hp page_cycler_v2.typical_25

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983645348837955888

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6728108742279168


| 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!

Any update on this? Any idea why we're seeing this regression?

I didn't reinvestigate this recently but last time I did, no. I couldn't figure out the problem. I'll look into it again soon.

 Issue 716160  has been merged into this issue.

We had some discussion offline a while back and we are fairly confident that this regression is just noise so WontFix.