All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=771214

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=dcb100162f0fd247fbca440788c9f86ee852bf6d4a3e65489d892c904152bc86


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

chromium-rel-win7-x64-dual
chromium-rel-win8-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8966739335455103184

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: win_8_perf_bisect
  Benchmark    : memory.long_running_idle_gmail_tbmv2
  Metric       : v8-gc-incremental-step_max/v8-gc-incremental-step_max

Revision             Result                  N
chromium@504569      3.51952 +- 4.52012      21      good
chromium@504656      3.1791 +- 2.07105       21      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 memory.long_running_idle_gmail_tbmv2

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966739335455103184


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965829130828596720

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965829122837896320

Re-kicked bisect.
However given the metric in the regression and the only v8 roll in range
https://chromium.googlesource.com/v8/v8/+log/67e0f781..d9e74be3

I suspect this might be 
[heap] Adjust incremental marking step size heuristics. 

But this is backed by no evidence. Let's see if the bisect confirms.

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

Hi ulan@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 : Ulan Degenbaev
  Commit : 1e3ee8cc27e4ab9da837b10023c8062130002116
  Date   : Tue Sep 26 06:42:57 2017
  Subject: [heap] Adjust incremental marking step size heuristics.

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : memory.long_running_idle_gmail_background_tbmv2
  Metric       : v8-gc-incremental-step_avg/v8-gc-incremental-step_avg
  Change       : 31.93% | 0.393563495168 -> 0.519225723315

Revision                           Result                     N
chromium@504592                    0.393563 +- 0.0733795      6      good
chromium@504605                    0.412509 +- 0.0480466      9      good
chromium@504605,v8@2f88c9b2df      0.406216 +- 0.0699844      9      good
chromium@504605,v8@5a5783e3bf      0.415374 +- 0.059741       9      good
chromium@504605,v8@1e3ee8cc27      0.478941 +- 0.106265       9      bad       <--
chromium@504605,v8@6f6adaa342      0.514824 +- 0.153842       9      bad
chromium@504605,v8@dc64a7311d      0.55417 +- 0.0506987       6      bad
chromium@504606                    0.505173 +- 0.206847       9      bad
chromium@504607                    0.514627 +- 0.0524879      6      bad
chromium@504608                    0.539883 +- 0.0765466      6      bad
chromium@504611                    0.510926 +- 0.0777889      6      bad
chromium@504617                    0.506424 +- 0.0551989      6      bad
chromium@504641                    0.519226 +- 0.102408       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 memory.long_running_idle_gmail_background_tbmv2

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965829122837896320


For feedback, file a bug with component Speed>Bisection

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

Suspected Commit
  Author : Ulan Degenbaev
  Commit : 1e3ee8cc27e4ab9da837b10023c8062130002116
  Date   : Tue Sep 26 06:42:57 2017
  Subject: [heap] Adjust incremental marking step size heuristics.

Bisect Details
  Configuration: winx64_high_dpi_perf_bisect
  Benchmark    : memory.long_running_idle_gmail_background_tbmv2
  Metric       : v8-gc-incremental-step_avg/v8-gc-incremental-step_avg
  Change       : 11.96% | 0.47337479199 -> 0.529979500969

Revision                           Result                     N
chromium@504581                    0.473375 +- 0.167942       21      good
chromium@504594                    0.462474 +- 0.0842111      9       good
chromium@504600                    0.47529 +- 0.0943891       14      good
chromium@504603                    0.470626 +- 0.053755       9       good
chromium@504605                    0.458627 +- 0.0507411      14      good
chromium@504605,v8@2f88c9b2df      0.472857 +- 0.0931907      14      good
chromium@504605,v8@5a5783e3bf      0.468579 +- 0.0908691      9       good
chromium@504605,v8@1e3ee8cc27      0.51744 +- 0.203483        21      bad       <--
chromium@504605,v8@6f6adaa342      0.507805 +- 0.119626       14      bad
chromium@504605,v8@dc64a7311d      0.516768 +- 0.160958       6       bad
chromium@504606                    0.510307 +- 0.179504       21      bad
chromium@504632                    0.52998 +- 0.182966        14      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 memory.long_running_idle_gmail_background_tbmv2

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965829130828596720


For feedback, file a bug with component Speed>Bisection

The number of incremental steps decreases massively, this increases the average as expected.

The overall sum of the steps is better.