[Webkit-unassigned] [Bug 102994] [META][Qt] Implement Threaded Coordinated Graphics in WebKit1

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=102994





--- Comment #14 from Jocelyn Turcotte <jocelyn.turcotte at digia.com>  2013-01-02 08:11:17 PST ---
(In reply to comment #13)
Relevant noamr response:
17:19 < noamr> I thought it was valuable to try to unify the code bases; I'm always struggling with not breaking wk1 when I make coordinated graphics 
               changes
17:20 < jturcotte> noamr: Ok, why kind of change is usually breaking WK1 for example?
17:20 < noamr> for example stuff relating to backing stores
17:21 < noamr> there are always 10 places where I have to make the change
17:22 < noamr> right now WK1 AC has some issues
17:22 < noamr> direct composited images I don't think work correctly
17:22 < noamr> WebGL definitely doesn't on Mac
17:23 < noamr> Masks are not in the best shape; but that's on WK2 as well
17:23 < noamr> you can get out-of-memory for large tiles of course
17:25 < noamr> also since nobody is doing GL-based pixel-test gardening for webkit1, other things might get broken constantly

According to this, I agree that using CoordinatedGraphics in WK1 would straighten the behaviors between WebCore and TextureMapper and make it easier to work in this area. The IPC/Threading interface defined in [1] can't be avoided anyway if this is the way we're going for AC with GTK+WK2.

Hopefully my previous girly comments will have brought a bit clarity to compensate for the objecting tone.

[1] https://docs.google.com/document/pub?id=1UoI1zk-6nTUFtz8i4evURM8aQIjkDRC8boO1zPdMMBg

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