Ok, here's an explanation of the "hunch" in the last video card message.

The match code in ite does a pmap_extract of macvidphys. It then compares
macvidphys to to the pmapped address. If they don't match it does some kind
of weird kludge with NuBus ranges.

Well, match was failing on my IIvx with 040 Daystar card. So I had match
print out both macvidphys and pmap_extract. Well, macvidphys was fine but
pmap_extract was returning zero! This causes the match routine to immediately
return zero.

My hunch is that pmap_extract is not working correctly on 040 machines.
The code in ite match is exposing the bug here because my machine's video
isn't in NuBus range.

That's just a guess.

I'm working off of a -current kernel but of pre-"signal crash" vintage.

_______________________________________________________________________
 Michael Zucca - mrz5149@acm.org - http://www.rit.edu/~mrz5149/
 "I will choose a path that's clear. I will choose Freewill. "
  --Rush, Freewill
_______________________________________________________________________