events: performance regression in event benchmarks

[jasnell]

See: #12655 (comment)

$ ./node benchmark/compare.js --old node --new ../main/node events > ~/eventsbench
[00:10:21|% 100| 7/7 files | 60/60 runs | 1/1 configs]: Done
 james@ubuntu:~/node/node$cat ~/eventsbench | Rscript benchmark/compare.Rh
                                                     improvement confidence      p.value
 events/ee-add-remove.js n=250000                       -45.73 %        *** 8.380381e-18
 events/ee-emit-multi-args.js n=2000000                   0.70 %            2.285963e-01
 events/ee-emit.js n=2000000                              2.12 %        *** 3.829377e-05
 events/ee-listener-count-on-prototype.js n=50000000    -95.07 %        *** 1.717885e-35
 events/ee-listeners-many.js n=5000000                  -20.60 %        *** 8.616063e-14
 events/ee-listeners.js n=5000000                       -48.69 %        *** 2.759171e-60
 events/ee-once.js n=20000000                           -58.16 %        *** 3.019260e-56

I'm seeing a significant performance regression in the events benchmarks comparing master to 7.9.

/cc @mscdex @mcollina @nodejs/benchmarking

[TimothyGu]

Cross posting from #12655 (comment):

It's very probable that the regression you are seeing comes from #11930, but IMO it is more of a problem of unrealistic benchmark (see #11930 (comment) for my reasoning).

[mcollina]

I think the problem for this regression is https://bugs.chromium.org/p/v8/issues/detail?id=6376. Specifically https://github.com/nodejs/node/blob/master/lib/events.js#L378, as the regression is present in the remove benchmarks.

[bmeurer]

I'm working on the Array.prototype.shift fast-path in TurboFan.