> Date: Wed, 12 Feb 1997 08:21:51 -0500 (EST)
> From: der Mouse  <mouse@Rodents.Montreal.QC.CA>

[ removed si1, and...]

> However, since then, three times, upon rebooting I've gotten "getty
> repeating too rapidly on /dev/kd" (or whatever the exact message is),
> and upon rebooting again there's a getty.core in /.  To me, this looks
> exactly like a typical "everything dumps core". 

That is the failure mode I usually see on the 3/60.

My money is still on a "missing interrupt protection" somewhere,
because the problem seems to go away if DMA completion interrupts
are turned off.  (Some async. activity causing pmap reentrance?)
I'm planning to add some reentrance detection to the pmap soon...

> The weird thing is, why would replacing si1 with an sc0 at an unusual
> address start provoking everything dumps core?

Any change to the interrupt model?