On Tue, Sep 01, 2009 at 10:01:24PM -0700, Corbin Simpson wrote: > Then have an Intel-specific bit of code. Do a batchbuffer > checker/relocator/munger; we've got one for Radeons, and I'm sure you > guys need to do something similar for relocating BOs.
That's actually what I originally wanted to do. But you have to do so many checks of the parameters, it's unfunny. And the magic dance to do state changes on the overlay hw (somethings needs changing crtc output and reverting it afterwards) is also quite complicated. So I wouldn't really be able to drop that much code, if anything at all. And instead I'd have an interface in command buffer submission that looks an awfull lot like an very special purpose ioctl in disguise. So I went down that road and created a _real_ ioctl with well-defined semantics. > ~ C. Yours, Daniel btw: intel hw has some nice support for executing untrusted batchbuffers, so no monsterous checker/relocater/munger already present. -- Daniel Vetter Mail: dan...@ffwll.ch Mobile: 079 365 57 48 ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july -- _______________________________________________ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
