RenderObject::containerForRepaint should be able to tell if a full repaint has already been scheduled.

[zalan]

so that we don't issue redundant repaints

[zalan]

Created attachment 459620 [details]
Patch

[zalan]

Created attachment 459621 [details]
Patch

[Simon Fraser (smfr)]

Comment on attachment 459621 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=459621&action=review

> Source/WebCore/rendering/RenderView.cpp:452
>      repaintUsingContainer(repaintContainer, computeRectForRepaint(layoutOverflowRect(), repaintContainer));

Can repaintUsingContainer take a RenderElement& after this change?

[zalan]

Created attachment 459625 [details]
Patch

[zalan]

(In reply to Simon Fraser (smfr) from comment #3)
> Comment on attachment 459621 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=459621&action=review
> 
> > Source/WebCore/rendering/RenderView.cpp:452
> >      repaintUsingContainer(repaintContainer, computeRectForRepaint(layoutOverflowRect(), repaintContainer));
> 
> Can repaintUsingContainer take a RenderElement& after this change?
Yeah, I've been considering that change for a while, but I don't think we are there yet. After https://bugs.webkit.org/attachment.cgi?id=459627&action=prettypatch, where both repaint and repaintRectangle start replacing null_ptr repaint container with the view(), I think it should be relatively easy to do.

[zalan]

Created attachment 459651 [details]
Patch

[EWS]

Committed r294699 (250898@main): <https://commits.webkit.org/250898@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 459651 [details].

[Radar WebKit Bug Importer]

<rdar://problem/93793768>