[Webkit-unassigned] [Bug 98177] Needless repaints at the beginning and end of a transition of a translate3d.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=98177





--- Comment #5 from Simon Fraser (smfr) <simon.fraser at apple.com>  2012-10-02 15:01:27 PST ---
(In reply to comment #4)
> (In reply to comment #3)
> > (In reply to comment #2)
> > > (In reply to comment #1)
> > > > This is because we turn off overlap testing when there's a transform animation running.
> > > 
> > > Yes, that explains "Observation 1", since we don't want the animation to jank when the first "opacity" layer overlaps it.
> > > 
> > > However, that doesn't explain "Observation 3".  Since the third "opacity" element already has a composited parent, there's no need to promote it.  Once something becomes a composited layer, it seems like we should turn back on overlap testing.
> > 
> > I think we can only do this if the composited parent has overflow: hidden. Bug 90633 covers that.
> 
> I don't understand why overflow:hidden / clipping comes into play here.  If you have some tree like this:
> 
>     -A (animated)
>     -B (sibling of A, has backing)
>      -C (child of B)
> 
> ...then there is no way that C could overlap A's composited layer bounds but not B's composited layer bounds.  Therefore, it should always be correct to draw C into B's backing, regardless of the position of A.

I think you are right.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.