[Webkit-unassigned] [Bug 38783] [chromium] Implement h/w accelerated compositing for LayerChromium layers

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=38783





--- Comment #7 from Kenneth Russell <kbr at google.com>  2010-05-14 13:16:28 PST ---
(From update of attachment 56010)
This generally looks good to me. It looks like the code is mostly interdependent so it seems to me it will be difficult to split it up into multiple patches. I'd support landing it all at once, given that it's been demonstrated to work.

The only high level comment I have is that compositeLayersRecursive does two pieces of work: updating the layer's contents and drawing it to the screen. I think it would be better to split these into two different phases, even though doing so would involve two traversals of the layer hierarchy. That way the updating of the layer's contents, which must be done on the main thread, would be more easily split off from the drawing, which might be more profitably done on a background thread in the future.

A few minor comments.

WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:215
 +      FloatPoint3D test2 = renderMatrix.mapPoint(FloatPoint3D(0.5, -0.5, 0));
Test code should be removed.

WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:155
 +  static GLuint createLayerTexture()
Should refactor with createTextureObject().


WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:265
 +      offset+= 3 * sizeof(GLfloat);
Need space between offset and += .

WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:416
 +      FloatPoint3D test2 = renderMatrix.mapPoint(FloatPoint3D(0.5, -0.5, 0));
More test code.


WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:720
 +  }
Formatting.

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