The x86emu in my last message is from the Alpha MILO not the MIPS MILO.
V3 Semiconductor Corp.
Dan Aizenstros wrote:
> Hello Dominic,
> I ported the MILO x86emu code to PMON. The x86emu code runs
> the VGA BIOS on the ATI card and the card appears to initialize
> okay. The only issue may be that the ATI BIOS may calculate
> timings incorrectly. However, the text mode appears to run and
> I can use the BIOS to display text on the screen when running PMON.
> The Linux framebuffer driver is also able to display text but I
> believe it is doing so in graphics mode.
> Dan Aizenstros
> Software Engineer
> V3 Semiconductor Corp.
> Dominic Sweetman wrote:
> > Dan Aizenstros (firstname.lastname@example.org) writes:
> > > Is XFree86 3.3.2 which is available from SGI and other ftp sites
> > > known to work on ATI MACH64 CT PCI video cards using the framebuffer
> > > driver atyfb.c on MIPS?
> > Just a note: PCI video cards carry a BIOS ROM. Some initialisation
> > routines from the BIOS are automatically run by a PC platform when it
> > powers up, so Linux/x86 software always sees the card as initialised.
> > It's common practice to put vital things in the BIOS. Many cards just
> > won't work without that initialisation, but on a non-x86 platform
> > that's what you're trying to do - even after you sort out all the
> > porting issues Ralf mentioned.
> > Algorithmics has partially solved this problem in other OS' - enough
> > to get X11 working on OpenBSD and a Windows CE screen up - by using an
> > x86 emulator to run enough of the video BIOS to set up the chip. If
> > the responses you had already haven't frightened you out of trying it,
> > we might be able to make a copy of that available for your personal
> > use. If it works, we'll figure out some way of licensing it.
> > Dominic Sweetman
> > Algorithmics Ltd
> > email@example.com