All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=592301

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDA35-zrwoM


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

chromium-rel-win7-gpu-intel

===== BISECT JOB RESULTS =====
Status: completed


===== TESTED REVISIONS =====
Revision                Mean Value  Std. Dev.   Num Values  Good?
chromium@379274         152.832674  7.916471    8           good
chromium@379307         156.229868  1.624853    5           good
chromium@379308         161.717182  1.019677    5           good
chromium@379309         134.376802  2.814299    5           bad
chromium@379310         135.233639  2.009922    5           bad
chromium@379313         134.248778  1.740204    5           bad
chromium@379321         135.616826  1.713422    5           bad
chromium@379339         132.692967  2.298565    5           bad
chromium@379423         135.151421  1.564414    8           bad

Bisect job ran on: winx64ati_perf_bisect
Bug ID: 592301

Test Command: python src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --also-run-disabled-tests blink_perf.bindings
Test Metric: node-list-access/node-list-access
Relative Change: 10.93%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1279
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9018766376331709984


| 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 label Cr-Tests-AutoBisect.  Thank you!

Suspect V8 roll @379309
https://chromium.googlesource.com/v8/v8/+log/9bf18113..bef3447f

I doubt this is a duplicate of  issue 592305 , since the runtime caller overhead is visible two weeks earlier (at rev 377041).

Are you sure you are commenting on the correct issue? I'm pretty sure this is dupe since it has exactly the same workload (accessing the nodelist indexed interceptor). It regressed on the same CL ("speed up lookup iterator" which has as a side-effect that it apparently makes indexed interceptors a little slower); and they improve (but don't fully recover) on the CL that improves indexed interceptor handling...