[webkit-reviews] review requested: [Bug 20980] Split off uncommonly used data from Node similar to ElementRareData : [Attachment 23700] Now with ChangeLog and a few cleanups

[bugzilla-daemon at webkit.org]

David Smith <catfish.man at gmail.com> has asked  for review:
Bug 20980: Split off uncommonly used data from Node similar to ElementRareData
https://bugs.webkit.org/show_bug.cgi?id=20980

Attachment 23700: Now with ChangeLog and a few cleanups
https://bugs.webkit.org/attachment.cgi?id=23700&action=edit

------- Additional Comments from David Smith <catfish.man at gmail.com>
I did some instrumenting of Node and browsed around for a while, the only site
that has a ton of rareData on it is digg.com, which appears to call childNodes
on *almost every node on the entire site*. Even the ones that can't possibly
have children, like CharacterData. However, even with that I couldn't get it to
show up as a measurable regression in the pageload testing I did on digg.

This should probably be tested on the internal PLT.