Messages in this thread | | | Date | Sun, 28 Jun 1998 21:00:30 +0200 (CEST) | From | Geert Uytterhoeven <> | Subject | Re: Fixes for vgafb on 2.1.107 |
| |
On Sun, 28 Jun 1998, Bradley M Keryan wrote: > On Sun, 28 Jun 1998, Geert Uytterhoeven wrote: > > After a lot of work, I found out what was wrong with vgafb in 2.1.107. The > > strange thing was that it does work in the vger tree... > > Very nice; vgafb works here on a Matrox Millenium I 4MB (2+2) PCI, and > seems fast enough at scrolling that I probably wouldn't notice any further > improvement.
OK.
> However, it seems the default palette is not exactly right; using ye old > console regression test (Nethack 3.2.2) things that should be brown are > now yellow (the darker yellow; normal yellow objects seem brighter). > Nethack does seem to draw things faster on vgafb than vgacon, though.
Aha. So the old palette was correct? There was one entry that didn't look `logical' to me. So I `fixed' it. I'll revert that change.
> Also, an error appeared in the system log: > > fb0: VGA frame buffer device, using 32K of video memory > kmem_alloc: NULL ptr (name=unknown) > Console: colour frame buffer device 80x25 > > fb_alloc_cmap uses kmalloc, so it would appear that the colormap problem > is another victim of the mm system not being initialized yet (so > initializing the colormap trashed *something*, and that memory was > subsequently changed, munging the palette). Am I correct?
It doesn't trash anything. We've been seeing that warning for years on m68k.
Greetings,
Geert
-- Geert Uytterhoeven Geert.Uytterhoeven@cs.kuleuven.ac.be Wavelets, Linux/{m68k~Amiga,PPC~CHRP} http://www.cs.kuleuven.ac.be/~geert/ Department of Computer Science -- Katholieke Universiteit Leuven -- Belgium
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |