All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=814052

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=b611979070a0a6217cff1730e93a2dc4faaf93d256c13c0565c18e30d2a511df


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

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

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16b389c7840000

📍 Found significant differences after each of 5 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16b389c7840000

[Liftoff] Refactor conditional instructions by sreten.kovacevic@mips.com
https://chromium.googlesource.com/v8/v8/+/1e89fed0b7c153a53eab5d0c3bcaf6a3666148a6

[assembler] Make {RelocInfo} independent of the Isolate. by mstarzinger@chromium.org
https://chromium.googlesource.com/v8/v8/+/a740255899296bf7bf250944c38547ad87504c68

[turbofan] Better checking of code start register. by mstarzinger@chromium.org
https://chromium.googlesource.com/v8/v8/+/5586ecfc68b6d897197ed4d9b2cdb847f24e62e2

[turbofan] Masking/poisoning in codegen (optimized code, x64) by mvstanton@chromium.org
https://chromium.googlesource.com/v8/v8/+/8f489e73b23f764801a35abaaf10772406fd4209

Update V8 to version 6.6.236. by v8-autoroll@chromium.org
https://chromium.googlesource.com/chromium/src/+/53221ec87ceb0b3ba45dd435d5a0c4da62aefda2

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

It's definitely not the Liftoff change.

Most probably the masking/poisoning, but that should be disabled by default, right?

Graphs have recovered.