All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=618661

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgvI3isAoM


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

linux-release

Re-kicked failed bisect on linux bot: https://chromeperf.appspot.com/buildbucket_job_status/9009962782040126448

I suspect this is 70a29c1298e666c465a8e75530575eda1aeaa37c. vmpstr, thoughts?

It's possible. I'll take a look.

Any idea what caused an increase in the same perftests later down the line? Is it possible to bisect that?

Hmm, I've tested this locally, and I'm not seeing an effect from this patch. The patch mostly affects tiles that have images, which isn't in this perftest. Is it possible to run a bisect?

I re-launched with a modified revision range (last failure on linux was due to the 'good' revision failing to be tested.

Also launched a bisect job on android s5.

Moving this nonessential bug to the next milestone.

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

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9007321657655705200

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9007321642324049776

https://chromeperf.appspot.com/buildbucket_job_status/9007321657655705200

Is bisecting on the improvement.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9006408896351189520

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9006408877860688272

Trying more bisects on both the drop and the improvement.

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

Hi wkorman@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 : Use alpha rather than inset rect for FakeRasterSource::CreateFilled.
Author  : wkorman
Commit description:
  
BUG= 529938 
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/2084313002
Cr-Commit-Position: refs/heads/master@{#401406}
Commit  : 2e7928680f0af8a1d5486499f41c58124b95f601
Date    : Wed Jun 22 21:18:03 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@401377  435.175  1.24348  5  good
chromium@401394  433.308  2.76414  5  good
chromium@401403  434.09   5.05906  5  good
chromium@401405  431.249  3.71648  5  good
chromium@401406  7607.7   31.1326  5  bad    <--
chromium@401407  7564.03  42.6761  5  bad
chromium@401411  7588.84  45.3284  5  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 618661

Test Command: ./src/out/Release/cc_perftests --test-launcher-print-test-stdio=always
Test Metric: prepare_tiles/50_100
Relative Change: 1643.86%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6606
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9006408877860688272


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5278733368295424

| 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!

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

Hi vmpstr@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 : cc: Separate raster and decode prepaint regions.
Author  : vmpstr
Commit description:
  
This patch separates raster and decode distances to be
independent of each other. Specifically, any tile in prepaint
that is further than 1000 screen space pixels from the
viewport will not be rasterized. Instead, we will pull out
the images on those tiles and predecode those instead.

R=enne, ericrk
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/2007163002
Cr-Commit-Position: refs/heads/master@{#398677}
Commit  : 70a29c1298e666c465a8e75530575eda1aeaa37c
Date    : Wed Jun 08 21:21:04 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@398642  2820.36  16.0906  5  good
chromium@398662  2823.68  19.2254  5  good
chromium@398672  2833.37  13.214   5  good
chromium@398675  2837.38  7.65606  5  good
chromium@398676  2836.11  9.74967  5  good
chromium@398677  448.325  6.14165  5  bad    <--
chromium@398681  452.046  1.45977  5  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 618661

Test Command: ./src/out/Release/cc_perftests --test-launcher-print-test-stdio=always
Test Metric: prepare_tiles/50_100
Relative Change: 83.97%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6605
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9006408896351189520


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5778487844012032

| 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!

Removing bisect label, since bisecting Jobs are completed.

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

Ping vmpstr@, the bisect confirm that it's your CL. Can you take a look?

I've looked at this and it's a result of changing the number of tiles we process in prepare tiles. It's not a real regression, since we're simply processing more tiles in a tight loop. I've filed crbug.com/653197 to rework the tests. This is a WontFix.