Avoid reflow if SetAttribute() sets attribute to identical value.

Ah, dammit. I wanted to get this in for M13. Vidur, do you think it's too risky for the performance gain? Waterson: you said you depended on this behaviour? Is that still true?

Some of the RDF template code depends on this behavior. Since there is currently no mechanism for "knowing" when a datasource is added/removed from the composite datasource, this was a dirty hack to force template-generated content to be rebuilt. I can't remember off the top of my head what will break if you make this change. My guess is that, if you did it for HTML elements only, -nothing- would break (most templates are generated using XUL elements). I'll file a bug vs. myself to add a "datasource add/remove" listener interface on the composite datasource, and make the template generator respond appropriately.

shaver: ignore bug 24110. just whack away and we'll fix what gets broke.

I've got the first part of this fix (the part that actually looks like the old == new optimization) in my tree, so it'll probably land soon. (About freaking time, sorry.)

Fixed, baby.

Mass update of qa contact

QA contact Update

Updating QA contact to Shivakiran Tummala.

verified, fix checked in