[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 #6 from Ami Fischman <fischman at chromium.org>  2012-01-23 19:30:36 PST ---
(In reply to comment #5)
> In the H/W decoded case, there is no upload and the put() call has to be deferred until the draw is issued, IIUC.

That's right.

To recap: the {get,put}CurrentFrame() calls have acquire/release semantics for the frame involved.  The frame handed back by getCF() is only valid until it is returned by putCF() (at which point the underlying texture or memory can be reused for another frame).

(In reply to comment #1)
{get,put}CF() do not require holding a lock.
CCVLI::m_providerMutex is there to guard the m_provider member (so that it can be cleared safely), not individual frames.  Unless you're trying to enable clearing (deleting) a provider between getCF & putCF, I don't expect this to be relevant.

Happy to chat if this is confusing or seems wrong.

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