Re: holes for ISA boards...

david parsons (o.r.c@p.e.l.l.p.o.r.t.l.a.n.d.o.r.u.s)
15 Jun 1999 15:32:26 -0700

In article <linux.kernel.19990615114254.A586@asmodean.local.network>,
Mark Jefferys <mjeffery@europa.com> wrote:
>On Sun, Jun 13, 1999 at 12:11:56PM +0100, Stephen Early wrote:
>
>% ACPI is a bit of a red herring here; the int 0x15, eax=0xe820 BIOS
>% call has been around longer than the ACPI spec. The ACPI people just
>% extended it with two new memory range types, to support their own
>% configuration information; code unaware of the ACPI spec. will see
>% those regions as unusable.
>
>Interesting.
>
>% I haven't found any Pentium-class motherboard that does not support
>% this call. I have found older motherboards that don't, and one which
>% gives odd results ("Out of memory" error when the call is made!) -
>% which is one reason I didn't submit the patch when I wrote it.
>
>It might be reasonable to blacklist the motherboard.

In the simple case, setting mem=<foo> on the command-line should
circumvent any probing.

I've certainly not had any motherboard lock up on me when trying
to do E820 (and this includes a lot of motherboards, including an
ancient Pentia motherboard that doesn't support E820 or E801,
but must have mem= specified if I want to access more than 64mb.)

But my fancy memory patch isn't Mr. Early's fancy memory patch,
and it may behave differently on machines that don't support E820.

____
david parsons \bi/ On 2.0.28, it works without complaint. On 2.2.9,
\/ weeeeeell, that's a different kind of story.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/