All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=623067

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgwp7CrgkM


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

android-one

===== 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@401270  59869.7  1543.94  5  good
chromium@401284  67814.4  3904.05  8  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 623067

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.parser
Test Metric: html-parser-threaded/html-parser-threaded
Relative Change: 13.60%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1366
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9008963394955501008


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5310769009262592

| 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

Mustaq, can you follow up on this?

From the instructions (https://chromium.googlesource.com/chromium/src/+/master/tools/perf/docs/perf_regression_sheriffing.md):
"After your shift, please try to follow up on the bugs you filed weekly. Kick off new bisects if the previous ones failed, and if the bisect picks a likely culprit follow up to ensure the CL author addresses the problem. If you are certain that a specific CL caused a performance regression, and the author does not have an immediate plan to address the problem, please revert the CL."

Kicked off another bisect with repeat_count=40.

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

Hi drott@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 : Purge FontCache on memory pressure
Author  : drott
Commit description:
  
BUG= 622286 
R=bashi,eae,tzik

Review-Url: https://codereview.chromium.org/2081333004
Cr-Commit-Position: refs/heads/master@{#401273}
Commit  : c9d8b3246bd57e05ab5a99ab741f7988f43cd11c
Date    : Wed Jun 22 14:28:07 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@401270  54052.9  526.134  8  good
chromium@401272  54856.2  1805.21  8  good
chromium@401273  67200.4  2027.73  5  bad    <--
chromium@401274  63863.8  5128.19  5  bad
chromium@401277  63003.1  1857.93  4  bad
chromium@401284  63007.4  2846.99  5  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 623067

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.parser
Test Metric: html-parser-threaded/html-parser-threaded
Relative Change: 16.11%
Score: 98.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1403
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9007408626778799568


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5860242731040768

| 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!

@drott, gentle ping. Have you had a chance to look into this regression? 

Benchmarking and reacting to the low memory signal seem to collide here. While we could revert this FontCache purging CL, before that I'd like to consult with the memory coordinator people and people working on the benchmarks on what we should do in this case. 

Should the low memory FontCache purging happen while benchmarking? Should the memory coordinator fire only when idle? bashi@, what do you think?

I think the core question here is whether the trade off between performance loss & memory improvement here is the right choice? If the trade off is right & our benchmarks infra reflect this, it's a good thing.

Do we see any memory benchmark improvements by your CL?

I can try to analyze that further. It's very difficult for me to verify this from the various graphs on chromeperf and finding out what's the right/current/valid set of results to look at. Since the perf regression is on android-one and the original intention of this change was to save memory on low-end, I thought I'd look at android-one. This configuration is listed there as a bot, which memory.* would be a valid reference and has results for r401273 (22nd June)? memory.top_10_mobile has results for this bot configuration, but it only started producing those from 2016-07-04.

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

Ping.

Perhaps we could try running the memory.top_10_mobile benchmarks locally with FontCache purging enabled and disabled?

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8997727970768282432

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1764
Failure reason: the build has failed due to infrastructure failure.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8982624482227921088

This bug is over a year old, and it seems unlikely we'll make progress at this point.