All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=761730

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=642d4b87f99c49b0de4f7489fe29eb8799bc878ee0f4243e62b5fd2e91e6cca8


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

chromium-rel-win7-gpu-intel

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8969399852596577616

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

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : system_health.memory_desktop
  Metric       : memory:chrome:renderer_processes:reported_by_chrome:v8:heap:effective_size_avg/load_social/load_social_vk

Revision             Result                  N
chromium@499404      8785920 +- 0.0          21      good
chromium@499411      9235310 +- 3082681      21      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

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=load.social.vk system_health.memory_desktop

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8969399852596577616


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8969377196839701456

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

Hi jbroman@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 : Jeremy Roman
  Commit : 1739c42dd9346a595a82600a558852b37535ba39
  Date   : Sun Sep 03 03:07:35 2017
  Subject: [Bindings] Generate explicit constructor stubs.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : system_health.memory_desktop
  Metric       : memory:chrome:renderer_processes:reported_by_chrome:v8:heap:effective_size_avg/load_search/load_search_baidu
  Change       : 10.36% | 8436394.66667 -> 9310208.0

Revision             Result                 N
chromium@499404      8436395 +- 605396      6      good
chromium@499408      8436395 +- 605396      6      good
chromium@499410      8611157 +- 605396      6      good
chromium@499411      9310208 +- 0.0         6      bad       <--

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

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=load.search.baidu system_health.memory_desktop

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8969377196839701456


For feedback, file a bug with component Speed>Bisection

 Issue 761935  has been merged into this issue.

Interesting. I'd have expected this to reduce heap size, if anything, because it skips allocating a number of v8::External objects (each of which is actually two heap objects) that we previously made.

Looking.

Both the Mac bot and my local Linux box show a small progression, rather than a regression, across this range. I wonder if this is just a matter of a GC being scheduled after the snapshot rather than before in this particular case.

This is supported by this regression not being visible on other page sets, which construct the same number of instances of the DOM templates and constructors.

I'm going to WontFix this as I don't believe it to be a real regression.

 Issue 762248  has been merged into this issue.

 Issue 762496  has been merged into this issue.

Issue 762280 has been merged into this issue.

 Issue 763094  has been merged into this issue.

Issue 763091 has been merged into this issue.

 Issue 763488  has been merged into this issue.