All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=651973

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxYiquQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxb3otAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghanroQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxYiquQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxY_FuQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgpdDZowkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghY-jtgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxaGUrwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxe_GtAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxZSb9AgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghdeJ4gkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxe-RrQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxZSb9AoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxe-NvAoM


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

android-nexus5
android-nexus5X
android-nexus7v2

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9000056137166052944

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

Hi bsalomon@google.com, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


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


===== SUSPECTED CL(s) =====
Subject : Change implementation of flush-count based GrGpuResource purging
Author  : bsalomon
Commit description:
  
Change default to approx 30seconds (given some API usage assumptions)
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2361093002

Review-Url: https://codereview.chromium.org/2361093002
Commit  : e2e87f3484e5524dbfd6c01f402136738d1d434b
Date    : Thu Sep 22 19:42:11 2016


===== TESTED REVISIONS =====
Revision                         Mean     Std Dev  N  Good?
chromium@420460                  18810.7  31.5637  6  good
chromium@420480                  18974.4  308.19   5  good
chromium@420483                  18863.2  74.5869  5  good
chromium@420484                  18827.2  27.0407  5  good
chromium@420484,skia@5745d795a1  18832.8  56.0999  5  good
chromium@420484,skia@e2e87f3484  23733.6  58.8286  5  bad    <--
chromium@420484,skia@3d89a7c0d2  23772.0  39.0896  5  bad
chromium@420485                  23720.0  34.0588  5  bad
chromium@420490                  23766.4  49.4449  5  bad
chromium@420500                  23744.0  18.9737  5  bad

Bisect job ran on: android_nexus7_perf_bisect
Bug ID: 651973

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests media.android.tough_video_cases
Test Metric: vm_shared_dirty_delta/video.html?src_crowd1080_vp9.webm_canvas_true
Relative Change: 26.23%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/3348
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9000056137166052944


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6392660942651392

| 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 change deliberately deferred aggressively freeing texture memory when Skia is under budget because it was causing cache thrashing in canvas2d. ( Issue 651829 )