On Thu, 16 Jan 2003, Steve Rumble wrote:

> On Wed, 15 Jan 2003 17:53:57 +0000
> Steve Foster <steve@devidian.me.uk> wrote:
>=20
> > given that it is a GIO card is there any way of getting gio to report
> > that this card is present?
> >
> > Have had a look at the gio driver, the function gio_print() would seem
> > to do this. Am I barking up the wrong tree?
>=20
> You'd need to read the product identification registers (if present) and
> match the board id. Take a look at the GIO Bus Specification
> document released by SGI. You can ftp it from oss.sgi.com
> under /pub/linux/mips/doc/indy/gio64.ps. It should be easy enough
> (although gio32 may use a slightly different scheme than gio32bis/gio64)
> to do. Of course, identifying is one thing, and supporting the adapters
> is another. From what I recall there aren't many of them, and
> most are simply an additional HPC1.5 or HPC3 with a scsi or ethernet
> chip. While support exists for these, there are some quirks and I've not
> been successful in getting my E++ ethernet card to transmit packets.

I don't think this board has a HPC chip on it, as all I can find are 2 memo=
ry chips, the tlan chip and nat semiconductors chips that I have seen on a =
great number of other NICs.

=46rom the previous thread on this subject, the conclusion was that the tl(=
) MI driver should work with a bit of hacking.=20

Cheers

Steve