[Webkit-unassigned] [Bug 76720] [chromium] Refactor video drawing to be more data driven

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=76720





--- Comment #1 from Tim Dresser <tdresser at chromium.org>  2012-01-23 11:29:05 PST ---
I'm not sure that its possible to avoid quads sharing writable objects in this case.

In the draw code for video quads, the VideoFrameProvider is guarded by a Mutex, with a lock that lasts from the time a frame is requested from the VideoFrameProvider, until the time that the frame has been drawn and given back to the VideoFrameProvider. Based on the documentation in VideoFrameProvider.h, it sounds to me like this lock has to last this long, and encompass the actual drawing code.

If this is true, I don't see how to avoid sharing the VideoFrameProvider and its associated Mutex. In addition, since copyFrameToTextures needs to be called within the lock, nativeTextureSize and nativeTextureId must be shared and writable. 

If these assumptions are true (which I hope they aren't), then I think that a backpointer to the layer might be the least ugly approach.

Any ideas?

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