2001383 - Allow using another object's iterator for indices optimizations if possible

Status: RESOLVED FIXED

(Core :: JavaScript Engine: JIT, enhancement, P2)

Description

[Alex Thayer [:alexical] (she/her)]

Assuming a function that looks something like this:

function shallowEqual(objA, objB) {
  const keysA = Object.keys(objA);
  const keysB = Object.keys(objB);

  if (keysA.length !== keysB.length) {
	return false;
  }

  // Test for A's keys different from B.
  for (let i = 0; i < keysA.length; i++) {
	const currentKey = keysA[i];
	if (
	  	!hasOwnProperty.call(objB, currentKey) ||
	  	objA[currentKey] != objB[currentKey]
	) {
  	return false;
	}
  }

  return true;
}

We currently have optimizations for the megamorphic load for objA, but not the hasOwn / load for objB. However, if we have an iterator we got from objB and it matches the iterator for objA, we should be able to use it as a base for the same optimizations, as long as the two iterators are the same.

Comment 1

[Alex Thayer [:alexical] (she/her)]

Attachment: Bug 2001383 - Allow using another object's iterator indices r?iain

Comment 2

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/2a24c97ea943
https://hg.mozilla.org/integration/autoland/rev/dbbe603e5647
Allow using another object's iterator indices r=iain,nbp

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/3de5137e2e0a
https://hg.mozilla.org/integration/autoland/rev/d7841f303558
Revert "Bug 2001383 - Allow using another object's iterator indices r=iain,nbp" for causing Btime failures.

Comment 4

[Atila Butkovits]

Backed out for causing Btime failures.

Backout link: https://hg-edge.mozilla.org/integration/autoland/rev/d7841f30355895b088a47fcd74b1f8af50cc49fd

Push with failures: https://treeherder.mozilla.org/jobs?repo=autoland&resultStatus=testfailed%2Cbusted%2Cexception%2Cretry%2Cusercancel&revision=dbbe603e5647643612d01cd122a35b97ee24e736&selectedTaskRun=BcL6NxIaQ4SzDFf0nCeQXg.0

Failure log: https://treeherder.mozilla.org/logviewer?job_id=537437622&repo=autoland&task=BcL6NxIaQ4SzDFf0nCeQXg.0&lineNumber=1047

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/288840fe5c11
https://hg.mozilla.org/integration/autoland/rev/317713e4d42c
Allow using another object's iterator indices r=iain,nbp

Comment 6

[Sandor Molnar[:smolnar]]

https://hg.mozilla.org/mozilla-central/rev/317713e4d42c

Comment 7

[Acasandrei Beatrice (needinfo me)]

(In reply to Pulsebot from comment #5)

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/288840fe5c11
https://hg.mozilla.org/integration/autoland/rev/317713e4d42c
Allow using another object's iterator indices r=iain,nbp

Perfherder has detected a browsertime performance change from push 317713e4d42c1f24fba40808ac916e190465aa88.

No action is required from the author; this comment is provided for informational purposes only.

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)	Performance Profiles
4%	speedometer3 TodoMVC-React-Redux/CompletingAllItems/Sync	linux1804-64-nightlyasrelease-qr	fission webrender	38.09 -> 36.72	Before/After
3%	speedometer3 TodoMVC-React-Redux/CompletingAllItems/total	linux1804-64-nightlyasrelease-qr	fission webrender	41.00 -> 39.63	Before/After

Need Help or Information?

If you have any questions, please reach out to [e-mail]. Alternatively, you can find help on Slack by joining #perf-help, and on Matrix you can find help by joining #perftest.

Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests.