All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=724195

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgmpaG1QoM


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

chromium-rel-win7-gpu-nvidia

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8979237286433350080

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

Hi senorblanco@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 : senorblanco
  Commit : 4b04a623b3af961b6284755050add2395e1d46f6
  Date   : Mon May 15 22:36:04 2017
  Subject: cc: fix for MSAA veto.

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : smoothness.gpu_rasterization.tough_path_rendering_cases
  Metric       : frame_times/https___testdrive-archive.azurewebsites.net_performance_chalkboard_
  Change       : 6.90% | 36.8902471299 -> 39.4355400069

Revision             Result                   N
chromium@471829      36.8902 +- 0.844825      6      good
chromium@471882      36.876 +- 0.525448       6      good
chromium@471908      36.9398 +- 0.687078      6      good
chromium@471922      36.9126 +- 0.937461      6      good
chromium@471928      36.8172 +- 1.18532       6      good
chromium@471929      38.868 +- 0.279285       6      bad       <--
chromium@471930      38.8349 +- 0.537476      6      bad
chromium@471931      38.9444 +- 0.628342      6      bad
chromium@471934      39.4355 +- 3.20437       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=https...testdrive.archive.azurewebsites.net.performance.chalkboard. smoothness.gpu_rasterization.tough_path_rendering_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8979237286433350080

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6053671533543424


| 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 Speed>Bisection.  Thank you!

My change reverts performance back to where we were before https://chromium.googlesource.com/chromium/src/+/59df29deed59592d42ff7e600f33faf035e00b46, which inadvertently disabled the MSAA veto.

The fact that the non-MSAA path is now slightly faster than MSAA is somewhat surprising -- I've been optimizing the non-MSAA path, but on all other bots it remains slower than the MSAA path. We might consider turning of MSAA on NVidia but that would have to be tested more widely.