> Michael was referring to the first two problems (interrupt-related)

We were talking about PDS ethernet on LC machines if I remember right.
Folks have been complaining that they get three attaches for one card.
I figure it's because the PDS slot maps to three slots. My guess is
that we should only probe one slot (like $E) on LC machines. The best
thing to do would be to make a table or something in machdep that has
valid NuBus slots to probe.

I think this is something long overdue. The NuBus probe routines have
no business probing slots that don't exist on some machines (like slot $9
on Quadras). :)

That *might* solve the problem but it may run deeper than that.

BTW, I still don't know where the interrupts on the 630's are so we're
still stuck on that one. The quarter's coming to a close so I might have
time to see if I can get my hands on a 630 amongst other things.