All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=632328

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghsTbvgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-q6isQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghrizuwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-sPT5AgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghuOQuAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghomPogoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghu2YvgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-onXsQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-u-25ggM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICguvy8iwkM


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

chromium-rel-mac-retina

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005891318104163056

Note that the bisects will have no effect since this is caused by a change in OS on the bot. Assigning to shrike for triage and cc-ing spqchan.

===== BISECT JOB RESULTS =====
Status: completed


=== Bisection aborted ===
The bisect was aborted because The metric values for the initial "good" and "bad" revisions do not represent a clear regression.
Please contact the the team (see below) if you believe this is in error.

=== Warnings ===
The following warnings were raised by the bisect job:

 * Bisect failed to reproduce the regression with enough confidence.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N   Good?
chromium@407959  12.8844  0.200508  18  good
chromium@407961  12.8063  0.218285  18  bad

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 632328

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.top_25_smooth
Test Metric: mean_input_event_latency/Wordpress
Relative Change: 1.16%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1456
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9005891318104163056


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5880960755695616

| 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 Tests>AutoBisect.  Thank you!

Currently examining the graphs. Since this is an issue involving the OS, the culprit will be tricky to figure out

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005435616069964000

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005307023316685680

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005215860053898256

===== BISECT JOB RESULTS =====
Status: completed


=== Bisection aborted ===
The bisect was aborted because The metric values for the initial "good" and "bad" revisions do not represent a clear regression.
Please contact the the team (see below) if you believe this is in error.

=== Warnings ===
The following warnings were raised by the bisect job:

 * Bisect failed to reproduce the regression with enough confidence.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N   Good?
chromium@407938  12.7323  0.183187  18  good
chromium@407983  12.7122  0.276932  18  bad

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 632328

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.top_25_smooth
Test Metric: mean_input_event_latency/Wordpress
Relative Change: 0.18%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1529
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9005307023316685680


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5776131140091904

| 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 Tests>AutoBisect.  Thank you!

A note on this: some research is ongoing in  bug 636755 , and it appears these bots may possibly not have been running in retina mode on 10.9, which would at least partially explain the difference.

Friend ping from perf sheriff

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8999531438131173136

fixit ping: Shrike, can you follow up on this regression now that the potential blocker has been resolved?

Note: I still think this was due to the switch to retina mode as per comment 10. shrike, should we close?

===== BISECT JOB RESULTS =====
Status: completed


=== Bisection aborted ===
The bisect was aborted because The metric values for the initial "good" and "bad" revisions do not represent a clear regression.
Please contact the the team (see below) if you believe this is in error.

=== Warnings ===
The following warnings were raised by the bisect job:

 * Bisect failed to reproduce the regression with enough confidence.

===== TESTED REVISIONS =====
Revision         Mean    Std Dev  N   Good?
chromium@407959  463534  804.681  18  good
chromium@408006  463708  650.251  18  bad

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 632328

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests page_cycler_site_isolation.basic_oopif
Test Metric: vm_resident_set_size_final_size_total/vm_resident_set_size_final_size_total
Relative Change: 0.12%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1725
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8999531438131173136


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5574671435563008

| 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 Tests>AutoBisect.  Thank you!

erikchen@ - what are your thoughts?

While there may be real 10.9 -> 10.11 regressions/improvements, they will be dwarfed by the non-retina -> retina change. Don't think there's anything to do here, and manual investigation is unlikely to be a good use of time.