All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=756946

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=6e638885254e13a923e7e71a74a694f9f2760585a6859b103aebdffd5c761099


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

chromium-rel-mac11

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8970903065983092608

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

Hi wkorman@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 : Walter Korman
  Commit : 7d7ff538f2290ac340336b14c34ba6183566dd3c
  Date   : Wed Aug 16 23:09:50 2017
  Subject: Migrate top_25 and top_25_smooth PageSet(s) to WprGo.

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : rasterize_and_record_micro.top_25
  Metric       : record_time/Pinterest
  Change       : 895.65% | 0.00383333333333 -> 0.0381666666667

Revision             Result                         N
chromium@494982      0.00383333 +- 0.000912871      6      good
chromium@494988      0.004 +- 0.0                   6      good
chromium@494989      0.004 +- 0.0                   6      good
chromium@494990      0.0378333 +- 0.000912871       6      bad       <--
chromium@494991      0.0383333 +- 0.0011547         6      bad
chromium@494994      0.0378333 +- 0.000912871       6      bad
chromium@495006      0.0378333 +- 0.000912871       6      bad
chromium@495029      0.0383333 +- 0.00305505        6      bad
chromium@495075      0.0381667 +- 0.00219848        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=Pinterest rasterize_and_record_micro.top_25

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8970903065983092608


For feedback, file a bug with component Speed>Bisection

Note that there is a similar bug for migrating loading.desktop to WprGo:  crbug.com/756481 

My change should only have altered the means by which the web page that's being benchmarked is replayed. I will have to try to repro. +ned as FYI and in case he has any other insight.

Hmhh, it's possible that the rendering content of the page is not longer the same with the migration to wprgo. 

I'll try reverting my change to see if it fixes as initial step.

With my patch reverted locally the Pinterest page with rasterize_and_record_micro.top_25 shows a straight text page with '404 not found'.

With the patch we now see what looks like a Pinterest page with a missing style sheet. So it looks "improved" versus previous (and explains the regression we now see as we're actually rendering more), but we should presumably review/fix the Pinterest page to be re-recorded and more representative of a fully loading successful page.

I will WontFix this and file a P2 bug for the Pinterest fix.

 Issue 759725  has been merged into this issue.