All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=776395

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=cb559f9c12623f9b9480a924eeb3fd8dcab22ef468b953b11021a7c186e79354


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

chromium-rel-mac12

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965291925104480704

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

Hi robertphillips@google.com, 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 : Robert Phillips
  Commit : 7f9b215695aae262797758b2e2a7b1d35fbe584f
  Date   : Thu Oct 12 13:42:01 2017
  Subject: Force all Intel GPUs to use draws instead of clears on Macs

Bisect Details
  Configuration: mac_10_12_perf_bisect
  Benchmark    : memory.desktop
  Metric       : memory:chrome:all_processes:reported_by_chrome:malloc:effective_size_avg/TrivialCanvasPageSharedPageState
  Change       : 5.08% | 22021496.0 -> 23139490.6667

Revision                             Result                  N
chromium@508740                      22021496 +- 286260      6      good
chromium@508800                      22187317 +- 654343      6      good
chromium@508830                      22168336 +- 527233      6      good
chromium@508838                      22270819 +- 736176      6      good
chromium@508839                      22240141 +- 561475      6      good
chromium@508839,skia@7f9b215695      22982211 +- 418396      6      bad       <--
chromium@508839,skia@5698c8a8a0      23196027 +- 444345      6      bad
chromium@508839,skia@cd975d5aa5      23102411 +- 659265      6      bad
chromium@508839,skia@1ba5b6baa8      23059397 +- 349569      6      bad
chromium@508839,skia@18e5cbbe65      22997795 +- 385905      6      bad
chromium@508839,skia@8efb9c5e82      23179643 +- 620338      6      bad
chromium@508840                      23187944 +- 926045      6      bad
chromium@508842                      23079539 +- 223530      6      bad
chromium@508845                      23199941 +- 409820      6      bad
chromium@508859                      23139491 +- 617340      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=TrivialCanvasPageSharedPageState memory.desktop

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965291925104480704


For feedback, file a bug with component Speed>Bisection

 Issue 776398  has been merged into this issue.

 Issue 776947  has been merged into this issue.

I am closing this bug. 

The Skia CL https://skia-review.googlesource.com/60681 (Improve granularity of clears-as-draws work around) regained a bit of the perf loss in the following two cases:

chromium-rel-mac11/blink_perf.canvas / transferFromImageBitmap
chromium-rel-mac12/blink_perf.canvas / transferFromImageBitmap

but that is probably about all we're going to be able to accomplish short-term. 

We still have the Skia bug skbug.com/7154 and the ANGLE bug crbug.com/774576 to drive longer term performance improvements.

Note that I believe the decode-lossless-webp regressions are separate from the Skia issue.