On Tue, 28 Jan 2003, Ivan Kokshaysky wrote:
> On Mon, Jan 27, 2003 at 06:55:04PM +0100, Benjamin Herrenschmidt wrote:
> > Well, your example clearly limits us to one IO space for VGA, which
> > might not be what we want. The problem also exist for some fbdev drivers
> > which might need to tap the VGA IOs of a given PCI card (thus getting
> > access to the "legacy" IOs of the bus the card is on).
>
> You are right, I've already realized that. :-)
> The struct pci_bus * arg to legacy_ioport_remap (maybe better
> pci_legacy_ioport_remap) is really good idea, and it's perfectly
> ok to pass NULL in the vgacon case - we are limited to only one
> VGA console anyway.
> After the PCI setup is done, pci_legacy_ioport_remap(pbus, &legacy_resource)
> would solve any problem I can think of, including multiple ISA bridges.
BTW, we still need a separate isa_request_mem_region(), since right now we
cannot simply call request_mem_region(0xa0000, 0x10000) to request the VGA
memory buffer in ISA memory space. On ia32 the plain request_mem_region() is
OK, but on other archs you need to add the ISA memory space base.
Gr{oetje,eeting}s,
Geert
-- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.orgIn personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jan 31 2003 - 22:00:18 EST