On Fri, 9 Jun 2000, Andreas Beck wrote:
> > > The malloc()-call allocates enough memory for each scanline (depends on the
> > > current visual->virt.y).
> > As big as the framebuffer? That's a bit overkill when most span
> > sets will probably be for smallish GUI elements.
>
> No - if I got him right, he's talking about having a
> struct span { int start,stop } spans[vis->virt.y].
Yep. Andy, right you are. But thinking closer, I think it is better having a
struct span { int start,stop } spans[vis->visible.y]. It makes more sense just
^^^^^^^
to draw the real visible spans...
> > I've seen _huge_ performance increases when running on a target
> > with accelerated hlines. I haven't compared write-combining, but I'm sure
> > it will also speed things up. The mga2164w Matrox stubs target code,
> > which I have been using recently with matroxfb and my G200, has both of
> > these and it screams through the XMI demos.
>
> BTW: You might have noticed that I added a simple Triangle primitive to the
> kgicon command set - you're welcome to add support for that. I'll test
> how that behaves on an acceled chipset.
That would be fine.
Christoph Egger
E-Mail: [EMAIL PROTECTED]
