All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=750827

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=f69d459ec7aeb26a52c388c0d5cfab9703a57a27998bc72af0e949b09d8435d4


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

chromium-rel-win10
chromium-rel-win7-gpu-intel
chromium-rel-win7-gpu-nvidia
chromium-rel-win8-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972523778167847680

=== Auto-CCing suspected CL author robhogan@gmail.com ===

Hi robhogan@gmail.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 Hogan
  Commit : be5bfe1766359702932614d572d0190528d77240
  Date   : Fri Jul 28 15:24:28 2017
  Subject: Revert "Don't truncate inline atomic elements before an ellipsis"

Bisect Details
  Configuration: winx64_10_perf_bisect
  Benchmark    : blink_perf.dom
  Metric       : modify-element-id/modify-element-id
  Change       : 1.63% | 88.9668571429 -> 90.4207142857

Revision             Result                  N
chromium@490373      88.9669 +- 4.80073      21      good
chromium@490398      88.869 +- 1.77924       9       good
chromium@490399      91.334 +- 1.18325       6       bad       <--
chromium@490400      90.7277 +- 1.99174      6       bad
chromium@490401      90.9115 +- 2.02203      6       bad
chromium@490404      90.0991 +- 1.60377      9       bad
chromium@490410      90.0376 +- 2.7007       9       bad
chromium@490422      89.9073 +- 4.88547      21      bad
chromium@490470      90.4207 +- 7.98722      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 blink_perf.dom

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8972523778167847680


For feedback, file a bug with component Speed>Bisection

This revert is an unlikely culprit, the original was landed at 490341, reverted in 490399, and re-landed at 490648.

I notice that the regressions are all on win bots and that http://test-results.appspot.com/revision_range?start=490494&end=490494 landed around this time.

Is that a possible culprit?