All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=770382

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=ab7aacc9755abbfff3d9b13e032fa88c5882effb1bc06bff1530c68d6039b38a


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

win-high-dpi

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8967079274354624448

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

Hi beccahughes@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Becca Hughes
  Commit : b1df7b5e17fdf77e3da208ba8ccdb16d04efebf7
  Date   : Thu Sep 28 07:56:59 2017
  Subject: Reland "Media Controls: Replace painter with CSS."

Bisect Details
  Configuration: winx64_high_dpi_perf_bisect
  Benchmark    : media.tough_video_cases_tbmv2
  Metric       : memory:chrome:all_processes:reported_by_chrome:skia:effective_size_avg/video.html?src_crowd2160.mp4
  Change       : 1.66% | 4455564.33333 -> 4529329.0

Revision             Result                  N
chromium@504889      4455564 +- 1099.68      6      good
chromium@504926      4455644 +- 838.445      6      good
chromium@504931      4456330 +- 406.455      6      good
chromium@504934      4455894 +- 1084.62      6      good
chromium@504935      4455558 +- 1158.61      6      good
chromium@504936      4529661 +- 792.682      6      bad       <--
chromium@504945      4529729 +- 880.668      6      bad
chromium@504963      4529329 +- 1559.66      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=video.html.src.crowd2160.mp4 media.tough_video_cases_tbmv2

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8967079274354624448


For feedback, file a bug with component Speed>Bisection

Hi Becca, it looks like your commit http://crrev/504936 has caused a noticeable increase in the amount of memory used by Chrome during video playback. Could you please take a look to see if this increase is expected? Thanks!

Sure!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965500381655622448

=== BISECT JOB RESULTS ===
NO Perf regression found, tests failed to produce values

Bisect Details
  Configuration: winx64_high_dpi_perf_bisect
  Benchmark    : media.desktop
  Metric       : memory:chrome:all_processes:reported_by_chrome:skia:effective_size_avg/video.html?src_crowd2160.mp4


To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=video.html.src.crowd2160.mp4 media.desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965500381655622448


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965496922288672832

=== BISECT JOB RESULTS ===
NO Perf regression found, tests failed to produce values

Bisect Details
  Configuration: winx64_high_dpi_perf_bisect
  Benchmark    : media.desktop
  Metric       : memory:chrome:all_processes:reported_by_chrome:skia:effective_size_avg/video.html?src_crowd2160.mp4


To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=video.html.src.crowd2160.mp4 media.desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965496922288672832


For feedback, file a bug with component Speed>Bisection

I'm not sure why someone is kicking off new bisects. The first bisect was conclusive that there was a regression. John can help explain how to run and verify that your patch caused this regression locally.

The reason why the new bisects aren't working is that we renamed the benchmark from media.tough_video_cases to media.desktop.

There was a fix that should have fixed the performance issues that landed in the dip in point 505653 but at point 505742 it goes back up again so i'm not sure if that was a flake or another CL in that range. 

This curve has gone through quite a few jump of the size comparable to the one associated with this bug, near revisions 504117, 504296, 504963 (this bug), 505653, and 505742, so there appears to be some instability associated with this measurement. The recent change in benchmark name (media.tough_video_cases_tbmv2 to media.desktop) makes it very hard to bisect these old jumps.

In addition, there is a recent jump of much large magnitude, which is tracked by  issue #776856 . It is more appropriate to spend the effort to investigate  issue #776856  than this one. So I'm resolving this bug as Won't Fix.