All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=666313

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgy4XxsQoM


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

android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8995735728325726992

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

Hi tbansal@chromium.org, 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 : Set WebFont priority to very low if the network is detected to be slow
Author  : tbansal
Commit description:
  
When the network is detected to be slow, Blink immediately changes to
using fallback font. This CL changes the Blink resource priority of the
WebFont resource to VeryLow since the font is no longer required for
text paint.

BUG= 665504 

Review-Url: https://codereview.chromium.org/2494243003
Cr-Commit-Position: refs/heads/master@{#432254}
Commit  : 5ea0c57befbcadba648b21447e83d5d4a280c885
Date    : Tue Nov 15 21:00:15 2016


===== TESTED REVISIONS =====
Revision         Mean      Std Dev  N  Good?
chromium@432222  1246264   0.0      5  good
chromium@432242  1246264   0.0      5  good
chromium@432252  1246264   0.0      5  good
chromium@432253  1246264   0.0      5  good
chromium@432254  14894136  0.0      5  bad    <--
chromium@432255  14894136  0.0      5  bad
chromium@432257  14894136  0.0      5  bad
chromium@432262  14894136  0.0      5  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 666313

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.news.flipboard system_health.memory_mobile
Test Metric: memory:chrome:all_processes:reported_by_chrome:gpu:effective_size_avg/load_news/load_news_flipboard
Relative Change: 1095.10%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/883
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8995735728325726992


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4518456748146688

| 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/8995729160392384016

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


===== SUSPECTED CL(s) =====
Subject : Set WebFont priority to very low if the network is detected to be slow
Author  : tbansal
Commit description:
  
When the network is detected to be slow, Blink immediately changes to
using fallback font. This CL changes the Blink resource priority of the
WebFont resource to VeryLow since the font is no longer required for
text paint.

BUG= 665504 

Review-Url: https://codereview.chromium.org/2494243003
Cr-Commit-Position: refs/heads/master@{#432254}
Commit  : 5ea0c57befbcadba648b21447e83d5d4a280c885
Date    : Tue Nov 15 21:00:15 2016


===== TESTED REVISIONS =====
Revision         Mean      Std Dev  N  Good?
chromium@432222  1246264   0.0      5  good
chromium@432242  1246264   0.0      5  good
chromium@432252  1246264   0.0      5  good
chromium@432253  1246264   0.0      5  good
chromium@432254  14894136  0.0      5  bad    <--
chromium@432255  14894136  0.0      5  bad
chromium@432257  14894136  0.0      5  bad
chromium@432262  14894136  0.0      5  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 666313

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.news.flipboard system_health.memory_mobile
Test Metric: memory:chrome:all_processes:reported_by_chrome:gpu:effective_size_avg/load_news/load_news_flipboard
Relative Change: 1095.10%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/884
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8995729160392384016


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5246218152706048

| 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!

The webpage (https://flipboard.com/explore) loads 3 web fonts each approximately 85 KB in size. So, it is possible that delaying the load of web fonts in causing memory increase.

Is this still being worked on?

We are keeping an eye on the UMA memory metrics: http://shortn/_iR5KkBwt23. There is not much difference in memory metrics.

Note that the original big (1000%ish) regression for load_news_flipboard has recovered, so this bug not a big concern anymore.

The remaining ones seems negligible/noisy:
- load_media_flickr: almost recovered
- http___digg.com: ~1%
- Two Wordpress regressions currently 5%~15% but noisy.

Since only one site's metric has regressed and they seem un-actionable, I would suggest closing the bug.

Thanks, sounds good to me. We will obviously keep an eye on the memory metrics when we decide to ramp it.