On Tue, 22 Jul 2003, Frank van der Linden wrote:

> On Tue, Jul 22, 2003 at 03:58:29PM +0100, David Brownlee wrote:
> [...]
> > ioapic0: int9 f1a0<vector=a0,delmode=1,pending,actlo,irrpending,level,dest=0> 1000000<target=1>
>
> This is your problem. It looks like a polarity issue. Using MPACPI might
> work, since ACPI should provide an interrupt source override that has
> the right polarity.
>
> When using the ACPI code without MPACPI, the code more or less has to
> use an educated guess to find the right polarity and some othe values,
> because the MPBIOS tables aren't guaranteed to have this information,
> apparently (though I think they should, but I suppose using MPBIOS
> with ACPI is something that BIOS writers don't consider to be a valid
> combination).

	It works perfectly with MPACPI & acpi0. Strange, I'm _sure_ I
	tried that a month of so back and it hung on probing the scsibus,
	which is why I tried MPBIOS in the first place.

	Either way, all happy here now.
-- 
		David/absolute          -- www.netbsd.org: No hype required --