All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=609769

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgtMi1vAoM


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

android-nexus5

===== 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@391420  4.24561  0.424975  12  good
chromium@391461  4.46434  0.362111  18  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 609769

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_silk_cases
Test Metric: thread_GPU_cpu_time_per_frame/thread_GPU_cpu_time_per_frame
Relative Change: 9.38%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3669
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013419877332998640


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5793067030806528

| 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!

Trying another bisect.

Last bisect failed due to device offline, 
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3671

Re-kicked another bisect:
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3687

Trying another bisect.

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


===== TESTED REVISIONS =====
Revision                         Mean     Std Dev    N  Good?
chromium@391400                  4.76298  0.0165827  5  good
chromium@391425                  4.75039  0.0220997  5  good
chromium@391432                  4.78683  0.0286506  5  good
chromium@391435                  4.80508  0.020409   5  good
chromium@391435,skia@00d44e014c  4.79722  0.0575006  5  good
chromium@391436                  4.99038  0.0094262  5  bad
chromium@391437                  5.01861  0.0265701  5  bad
chromium@391438                  4.9661   0.020715   5  bad
chromium@391450                  4.96121  0.0248779  5  bad
chromium@391500                  4.96912  0.0252905  5  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 609769

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_silk_cases
Test Metric: thread_GPU_cpu_time_per_frame/thread_GPU_cpu_time_per_frame
Relative Change: 4.33%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3706
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9011589738932385600


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5288207835463680

| 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!

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

This metric still hasn't recovered.  Kicking off another bisect.

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


===== TESTED REVISIONS =====
Revision                         Mean     Std Dev    N  Good?
chromium@391300                  4.76173  0.0251994  5  good
chromium@391400                  4.7808   0.0177565  5  good
chromium@391425                  4.77458  0.0351826  5  good
chromium@391432                  4.80765  0.0218662  5  good
chromium@391435                  4.83763  0.0489385  5  good
chromium@391435,skia@00d44e014c  4.78314  0.0203887  5  good
chromium@391436                  4.9914   0.0101413  5  bad
chromium@391437                  4.97117  0.0260012  5  bad
chromium@391438                  4.98337  0.0271636  5  bad
chromium@391450                  4.96915  0.0228088  5  bad
chromium@391500                  4.9878   0.0254764  5  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 609769

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_silk_cases
Test Metric: thread_GPU_cpu_time_per_frame/thread_GPU_cpu_time_per_frame
Relative Change: 4.75%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3810
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9007708096016520192


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5824344052203520

| 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!

This issue has been moved once and is lower than Pri-1. Removing the milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

bisect suspected 391436, https://codereview.chromium.org/1950843002.
That was a skia roll including two CLs, https://chromium.googlesource.com/skia.git/+log/02125d10d5d0..0736f3386820

CC jvanverth@ who was reviewer of this roll.

Assigned to ericrk@ who can own bugs with chromium account, and submitted one of two CLs.

Perf sheriff ping: reminder to follow up on possible performance issues

This is not a regression - we were fixing a correctness issue that had been introduced earlier. The correctness issue saved us memory, but was incorrect, so we don't really have an option here.

You can see the introduction of the bug and the memory savings on the graph at Point ID: 389067, the change is 516778f8da98f40c94bfa34702fc167542bcd522