All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=802407

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=28ae8920692b319137a0fe1fa7e2dffd26fb1a52595a9560fb42162303d7cc42


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8957207904422553696

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

Hi falken@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 : Matt Falkenhagen
  Commit : f0f0804c5f71bdeafdf9a86175e54291ffb6a172
  Date   : Fri Jan 12 06:34:00 2018
  Subject: service worker: Remove cross-site transfer code.

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : blink_perf.bindings
  Metric       : first-child/first-child
  Change       : 4.14% | 207.922960009 -> 199.321599272

Revision             Result                  N
chromium@528883      207.923 +- 2.99572      9       good
chromium@528890      208.85 +- 0.722575      6       good
chromium@528893      211.296 +- 15.3485      9       good
chromium@528894      199.265 +- 11.2995      9       bad       <--
chromium@528895      200.013 +- 17.2056      14      bad
chromium@528896      198.007 +- 1.52623      9       bad
chromium@528908      199.322 +- 11.1269      9       bad

Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.bindings

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8957207904422553696


For feedback, file a bug with component Speed>Bisection

I doubt my change caused this:
* The change just removed dead code (assuming the perf bot is running with PlzNavigate on. PlzNavigate off is no longer supported)
* I don't think blink_perf.bindings would involve navigations with service worker.

cc benchmark owners according to blink_perf.py.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8957126836735699136

I've asked the bot to try again; we'll see what it finds this time. Agreed that service worker isn't involved in that microbenchmark.

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : blink_perf.bindings
  Metric       : first-child/first-child


Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.bindings

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8957126836735699136


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8957107103600083664

Jeremy, could you triage this issue once the bisect finishes?

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14e325c6840000

📍 Found significant differences after each of 13 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14e325c6840000

Add traffic annotation auditor's executable path to md files. by rhalavati@google.com
chromium @ 2dfb924173c419f5c9cfeec74f8a87596f16b804

Make task runners at RTCPeerConnectionHandle in WaitableEventWatcher changeable by hajimehoshi@chromium.org
chromium @ 138652c9e5bc18add6649a2651162193b2053dc6

macOS V2 Sandbox: Disable V2 for  GPU processes. by kerrnel@chromium.org
chromium @ 6f59d9136999c22bbcbe226f00cd58563b29d81d

Do not use soon deprecated v8::ReturnValue::Set. by yangguo@chromium.org
chromium @ 7ef680c7b45f3470968ac126838bc8b90c383572

Local NTP: Change fakebox text from 38% to 42% black by treib@chromium.org
chromium @ 40378a280cc8a76edde8d21fd965722b66dbdf9e

[Web Payments] Embed Payment Handler window in PR dialog by anthonyvd@chromium.org
chromium @ 8f5e7e14b60b9c7735e0fdc208f5bb99b358619e

Make network traffic annotation auditor error resilient. by rhalavati@google.com
chromium @ 3efb8443e8e9475ab0fa600fb594d0aa6168a6c5

Move MediaDevicesPermissionChecker to MediaDevicesManager by c.padhi@samsung.com
chromium @ 7db4074e5ec8d6e8e096c48761088c842d399ece

Add fast path for executing queued tasks when size is 1 by adithyas@chromium.org
chromium @ 5ddc171ad18d47fdaf6697ef79aabe28613743e5

Fix a bug where virtual time may pause indefinitely by alexclarke@chromium.org
chromium @ 0b954c48a42bb80cf7930a3764b1fa96cb64a6c6

Enable BubbleFrameView subclasses to force the client view to overlap into the title. by kylixrd@chromium.org
chromium @ 8322e656602b2e61bed407915624e0662231b331

Render to Browser metadata via SwapPromise by jonross@chromium.org
chromium @ 89a3da2f7d11ce62fe6b8de2794692258bfc8653

Adds ability for BrowserMainParts::PreEarlyInitialization() to fail by sky@chromium.org
chromium @ 9068b4df605ec1b5767f66e3344a0cadeb82aaf1

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

macOS V2 Sandbox: Disable V2 for  GPU processes. by kerrnel@chromium.org
chromium @ 6f59d9136999c22bbcbe226f00cd58563b29d81d

is a revert for a mac only change, so it is unrelated. 

Add traffic annotation auditor's executable path to md files.
chromium @ 2dfb924173c419f5c9cfeec74f8a87596f16b804

and

Make network traffic annotation auditor error resilient.
chromium @ 3efb8443e8e9475ab0fa600fb594d0aa6168a6c5

are both just effecting a trybot test on linux_chromium_rel_ng. So must be unrelated.

Re Fix a bug where virtual time may pause indefinitely 
chromium @ 0b954c48a42bb80cf7930a3764b1fa96cb64a6c6 

I'd be quite surprised if that materially changed any benchmark.

Re: Local NTP: Change fakebox text from 38% to 42% black by treib@chromium.org
chromium @ 40378a280cc8a76edde8d21fd965722b66dbdf9e

Changing the text color via CSS file shouldn't have caused a performance regression. (Would be quite funny, tbh)

This is almost certainly

Do not use soon deprecated v8::ReturnValue::Set. by yangguo@chromium.org
chromium @ 7ef680c7b45f3470968ac126838bc8b90c383572

reverted already: https://chromium-review.googlesource.com/c/chromium/src/+/893358