[Webkit-unassigned] [Bug 88482] [Chromium] Compositor should avoid drawing quads when cached textures are available and contents unchanged

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=88482





--- Comment #42 from Adrienne Walker <enne at google.com>  2012-06-15 10:10:10 PST ---
(In reply to comment #40)
> (From update of attachment 147685 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=147685&action=review
> 
> >> Source/WebCore/platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:445
> >> +            if (!targetSurface->prepareContentsTexture(0))
> > 
> > If you need a static function for testing, then I'd rather you passed in the layer renderer to the function and forwarded that along to prepareContentsTexture, rather than making CCRenderSurface::prepareContentsTexture robust to a null LRC.
> 
> Please see comments #20 and #22 (last few lines). Will you agree to wait for Dana's change and separate the functions then?

Ok, but put a FIXME in prepareContentsTexture to separate out creation and reservation.

> >> Source/WebCore/platform/graphics/chromium/cc/CCRenderPass.h:61
> >> +    void appendQuad(PassOwnPtr<CCDrawQuad>);
> > 
> > I know you're adding this for testing, but can you not add this as a public function? The reason the other appendQuads functions don't take quads is so that the render pass can own the lifetime of the shared quad state.  With this function, it's not clear that the caller has to manage the lifetime of the shared quad state on the quad.
> > 
> > Maybe change the private modifier below to protected and make a CCTestRenderPass that exposes what you need?
> 
> Just wondering, can shared quad state be ref-counted between the quads that use it? Not for this patch of course.

Maybe? I think I'd want to see how the quad serialization shaped up first.

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