R500 initial driver release announcement

[Alexander Larsson]

On Wed, 2007-06-13 at 11:39 +0200, Jerome Glisse wrote:
> On 6/13/07, Alexander Larsson <alexl at redhat.com> wrote:
> > On Tue, 2007-06-12 at 20:55 +0200, Jerome Glisse wrote:
> > > Over the past couple of months, a small group of people has been working
> > > on reverse engineering the r500-based AMD video cards.  Everyone involved
> > > worked on this in whatever free time they had, which is why this has
> > > taken so long.
> >
> > Sweet!
> >
> > I tried this on my MacBook Pro (X1300, 1440x900 LCD), and it seems to
> > work. I had a couple of problems:
> > depth 16:
> >  colors where weird, white was green, etc
> >  screen had two copies of the data, one on the upper left quadrant, one
> >  one the upper right quadrant, and the lower half was mostly black with
> >  some garbage
> 
> On mbp i think its x1600 :)

Ah, yeah:
(--) Chipset RV530 (Radeon X1600) found

> > However, things are very slow. Slower than the vesa driver even, which I
> > assume is because that enables ShadowFB. Is it easy to add shadow
> > support to a driver?
> 
> ShadowFB might effectively help, don't think it's too hard to add.
> Feel free to submit patch. Btw did you see corruption around
> cursor (you should :)) ?

shadowfb patch attached. And with this i get identical (== usable) speed
as with vesa.

Also, i think i found a bug in the fbdev driver:

	if (fPtr->shadowFB) {
	    fPtr->shadow = xcalloc(1, pScrn->virtualX * pScrn->virtualY *
				   pScrn->bitsPerPixel);

surely that should be "pScrn->bitsPerPixel / 8"?

However, i saw no corruption around the cursor...

-------------- next part --------------
A non-text attachment was scrubbed...
Name: avivo-shadowfb.patch
Type: text/x-patch
Size: 5866 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg/attachments/20070613/f23e15fe/attachment.bin>