[webkit-reviews] review granted: [Bug 86582] [Performance test] Add a micro benchmark for div.firstChild getter : [Attachment 142181] Patch

[bugzilla-daemon at webkit.org]

Ryosuke Niwa <rniwa at webkit.org> has granted Kentaro Hara
<haraken at chromium.org>'s request for review:
Bug 86582: [Performance test] Add a micro benchmark for div.firstChild getter
https://bugs.webkit.org/show_bug.cgi?id=86582

Attachment 142181: Patch
https://bugs.webkit.org/attachment.cgi?id=142181&action=review

------- Additional Comments from Ryosuke Niwa <rniwa at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=142181&action=review


> PerformanceTests/Bindings/first-child.html:9
> +// This is a benchmark for 'firstChild' getter.
> +// This benchmark covers 'firstChild', 'lastChild', 'nextSibling'
> +// and 'previousSibling' in Dromaeo/dom-traverse.html, and other
> +// DOM attributes that return a Node object.

Perhaps we want PerfTestRunner.description?

> PerformanceTests/Bindings/first-child.html:13
> +	   div = document.createElement("div");

I would prefer declaring this explicitly in the global scope.

> PerformanceTests/Bindings/first-child.html:17
> +	   var localDiv = div;

If you're making a local copy, you might as well just put the div on "this"
object.

> PerformanceTests/ChangeLog:21
> +	   $ ./Tools/Scripts/run-perf-tests Bindings/first-child.html
> +	   Running Bindings/first-child.html (1 of 1)
> +	   RESULT Bindings: first-child= 802.111186541 runs/s
> +	   median= 807.571299375 runs/s, stdev= 16.7064414996 runs/s, min=
767.386091127 runs/s, max= 817.369093231 runs/s

How do values change if you run the test multiple times?
Ideally, we want the variance between multiple test runs to be smaller than
stdev reported here.