--Signature_Sun__22_May_2005_20_21_47_-0400_ORvawEOSde/KL/Sp
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

Hello,

> Is it really essential to use XFree?  This is an *excellent* example
> of how the XFree way is broken, why it is utterly insane to try to do
> things like bus scanning in the X server rather than letting the
> kernel do what kernels are for.

Yes, for lack of suitable alternatives it is essential.
On the other hand XFree handles SBus devices via /dev/fb* which should
be more or less what you're thinking of, so it's not entirely impossible
to do things in a sane way in XFree. At least on sparc and sparc64 we
can easily assign a fb device to PCI graphics boards.

> > Move your graphics card into another slot that belongs to pci0
>=20
> ...if there are any.  I've seen machines that appear to have a
> half-dozen PCI buses in the machine, some of which don't appear to
> have any physical sockets corresponding to them;

In the 9500 both buses have slots as far as I know.

> for all I can tell from
> here, this may be one.  In fact, based on the dmesg, it seems likely -
> the machine has two bandits at mainbus0, with the ATI card is behind
> the second one, and everything behind the first one looks like stuff
> that could plausibly be on the motherboard: pchb0, vr0, obio0 -
> exactly what's shown in the scan above.

Machines with only one bandit usually have 2 or 3 PCI slots ( or use a
pci-pci bridge like some UMAX boxes )=20

have fun
Michael

--Signature_Sun__22_May_2005_20_21_47_-0400_ORvawEOSde/KL/Sp
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (NetBSD)

iQEVAwUBQpEiG8pnzkX8Yg2nAQLHJQgAgfK4InKJB6mJwe2LYqJZtokfo8s9uNV4
xOkoNLSbLMFwLEt9+qQyV8iPgVxv34t5v7s/8KEesUTjo9ejdIjfx1TjuwDtC8e7
798axaeEkG26OpT6Yw6bZQlQW++vYiwnRlymvW0Nt+aaFuRobv1wNwJDRJKilDNT
rpASOGIZNxFXteSeL8oSemeZG8bxnTp7guSGiAiJRcV/NVqR3LT/7E6qsjfNAQZt
yLYHMrOBBs78OO5Su88befvrVuPJFjIrLDL5NZZzgjS2RGxwKmMn+/XMycr2hU50
/425wT0htDS32xNTgRxoHVjEmdyWPEsFq5OwVYkkSylBoilrIwTc0A==
=fDuY
-----END PGP SIGNATURE-----

--Signature_Sun__22_May_2005_20_21_47_-0400_ORvawEOSde/KL/Sp--