All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=637354

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg9vfjuwkM


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

android-nexus5

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9004512382477041888

===== 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@410744  17.8837  0.267137  12  good
chromium@410801  18.0559  0.330087  8   bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 637354

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.sync_scroll.key_mobile_sites_smooth
Test Metric: frame_times/http___shop.mobileweb.ebay.com_searchresults?kw_viking+helmet
Relative Change: 1.39%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3987
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004512382477041888


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5871223762845696

| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9000807101127400960

kicked another bisect with a wider range

===== 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@409467  17.5759  0.196199  18  good
chromium@410801  17.7818  0.281969  18  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 637354

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.sync_scroll.key_mobile_sites_smooth
Test Metric: frame_times/http___shop.mobileweb.ebay.com_searchresults?kw_viking+helmet
Relative Change: 1.96%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/4140
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9000807101127400960


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5267741940908032

| 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!

flackr, is it possible that your change regressed the frame_times metric?
https://chromium.googlesource.com/chromium/src/+/9a2cdc8d96d94c736eb5547a7c8c541e58e5759f
https://chromeperf.appspot.com/group_report?bug_id=637354

Perf sheriff ping

ping: flackr@, could you answer the question in #7?

Looking at http://shop.mobileweb.ebay.com/searchresults?viking+helmet it doesn't look like they use a composited overflow scroller with local background - which would be the case that could potentially have gotten slower from my patch.

re #10, flackr@ did you check with "android-nexus5". In any case that URL no longer works for me.

BTW, the regression is within the stddev range which explains the bisect failures. Not sure if there is any good way to try to bisect given this. Also since this bug, the metric has recovered slightly.  So I am not sure if it makes sense to put more effort into this issue. I suggest closing.