>>>>> "JG" == "Justin T Gibbs" <gibbs@plutotech.com> writes:

>>> It doesn't configure, saying something like
>>> eisa0: device ADP7880 in slot 14 not configured

JG> The problem here is that you have a PCI Adaptec controller
JG> (most likely on the motherboard) that just happened to get
JG> mapped by your PCI BIOS into a location that makes the
JG> vendor string appear in the same place as an EISA card in
JG> slot 14.  I thought that NetBSD had solved this problem by
JG> probing PCI devices first and using resource conflict
JG> resolution so that the EISA probe would never touch a
JG> device already claimed by the PCI probe?  Are you running
JG> an old version of NetBSD?

This sounds more like what I was expecting to be the true case on this
hardware.  However, why doesn't the PCI probe make note of the device
and fail to configure it?  Nothing in the pci probe output mentions a
SCSI class device.

The PCI probe finds 3 pchb's (0x8086/0x84c5 and 0x84c4--Orion memory
controllers (!) and a host-PCI bridge), an ethernet card (mentioned
before), a misc prehistoric (pci/eisa bridge), and an unknown
(0x8086/0x0008).

Maybe it truly is an EISA device on the motherboard?  I'll try the blind
patches suggested by Bernd Ernesti and report back.

==John

[Note: I tried saying this earlier, but failed to get the point across.
I can't actually install NetBSD on the system, I can only play around
with it via a boot floppy until it starts being used for Real Work under
Windows Nice Try.  I'm just playing with it since I was curious what
hardware was actually installed.]