On Mi, 2013-08-21 at 13:27 +0200, Paolo Bonzini wrote:
> Il 21/08/2013 13:01, Gerd Hoffmann ha scritto:
> > Hi,
> >
> >> +#define ICH9_PROC_BASE 0xaf00
> >> +#define ICH9_PROC_LEN 32
> >
> > No, please don't. It makes it impossible to assign the 0xa000 -> 0xafff
> > I/O port window to a PCI bridge. Please lets stop occupy random io
> > ports above 0x1000 and burn I/O address space that way.
> >
> > I'd suggest to place it at 0x0a00 instead. Also the dsdt should get a
> > device with the address in _CRS so the guest knows those ports are used.
>
> Would this use 0x0A ("not present, device functional, decoding
> resources") for _STA to "indicate a valid device for which no device
> driver should be loaded" (quoting from the ACPI spec)? Hopefully this
> prevents Windows from showing the UI.
Sounds reasonable. Needs careful testing ...
cheers,
Gerd
