On Thu, 19 Feb 2004 11:24:27 +0000
Alan Cox <[EMAIL PROTECTED]> wrote:

> On Iau, 2004-02-19 at 02:02, Alex Deucher wrote:
> > 2D: do to a bug in S3's code, 2D accel was never enabled when the DRI
> > was active.  I've managed to get 2D accel working again.  one question
> > I have, I assume xf86InitFBManager() should only manage memory for 2D
> > stuff?  Otherwise it would conflict with the back, depth, and texture
> > buffers, right? what I did to fix 2D accel it was to assign it whatever
> > was left after textures, back, and depth buffers had been allocated.
> 
> Are the 2D ops being used by the DRI layer in the S3 driver safe btw. It
> seems that with those exposed to the driver user space rather than in a 
> kernel side function the user app can DMA to arbitary system memory
> addresses ?

Yes. The current DMA implementation is unsafe and I'm aware of that. I
want to experiment with vertex buffers a bit. That's probably easier in
userspace. Once I've sorted this out I'm going to move the DMA buffer
management to the DRM (along with clearing, buffer swaping, ...). I
think there are templates for some of this already. I havn't worked on
the DRM before. I have no idea how long it'll take me to get into it.

Felix


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
--
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel

Reply via email to