All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=792043

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=e851e53208a5fd5dce65b72c386f60f59d1fc47e916d6ea950e7e5b0571784b3


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

android-nexus5
android-nexus7v2
android-webview-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8961033604086642272

=== BISECT JOB RESULTS ===
Perf regression found but unable to narrow commit range

Build failures prevented the bisect from narrowing the range further.


Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : blink_perf.shadow_dom
  Metric       : v1-large-deep-distribution/v1-large-deep-distribution
  Change       : 13.22% | 275.9675 -> 312.448166667

Suspected Commit Range
  2 commits in range
  https://chromium.googlesource.com/chromium/src/+log/4f59a790e19f97e8bbdbb6e7960883c1e1cf0234..c47b4d110a697afa6a6ee4d4748028312c65ec19


Revision             Result                  N
chromium@519865      275.967 +- 6.23249      6        good
chromium@519971      275.542 +- 4.72088      6        good
chromium@520026      281.649 +- 3.41125      6        good
chromium@520053      279.793 +- 6.10346      6        good
chromium@520060      281.124 +- 2.01064      6        good
chromium@520061      ---                     ---      build failure
chromium@520062      319.787 +- 7.59142      6        bad
chromium@520063      320.427 +- 6.94724      6        bad
chromium@520066      319.031 +- 4.74224      6        bad
chromium@520079      312.448 +- 5.10606      6        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.shadow_dom

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8961033604086642272


For feedback, file a bug with component Speed>Bisection

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14906ad8840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16bba898840000

Kicking off a few more bisects now that we have pinpoint to deal with errors better.

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14906ad8840000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/16bba898840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16cff59a840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/148b8606840000

Trying again...

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/16cff59a840000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/148b8606840000

Marking Fixed because it's recovered and assigning to mlippautz mostly as FYI since looks like it's related to your CLs.

mlippautz: sorry to get this bug to you so late. The builders keep failing getting builds to the bisector. But the bisect in #3 at least got down to two potential CLs. The first is your r520061, "[oilpan] Remove HEAP_INCREMENTAL_MARKING compile time flag". The second is r520062, "WebRTC: Migrate completely to new video codec factories API", which seems much more unlikely to affect a blink_perf benchmark. But I looked at the graphs and the regression recovered a while back. Your CL r523812, "[oilpan] Avoid building with incremental marking" is in the recovery range. So it looks like you've addressed this already.

Thanks for triaging. Yeah, we know that we currently cannot enable the flag per default yet. I will keep these benchmarks in mind when flipping.