Hey, everybody. Really have two seperate questions, but thought I'd
post `em in one message. BTW, great work so far. I am very happy
with my Indigo2 Impact running NetBSD.

So here goes:
First, the problem with running the Indigo2 10k is the cache coherency
problem as described in the message from the archives in June 2000 as
follows
>This issue was figured out late the the R10K I2
>design cycle.  The problem was fixed by modifying
>the compiler and assembler to issue a cache barrier
>instruction to address 0(sp) as the first instruction
>in basic blocks that contain stores to registers
>other than $0 and $sp.

Couldn't this same type of fix be applied to gcc so that it would "patch"
(or modify?) the code in the same way that SGI's compiler does it?

Secondly,
We all know the problem of being unable to use an SGI monitor and the
graphics console because of the loader/console intercepting many
important keystrokes. I seem to recall someone mentioning on here once
that they might have a patch for this problem. Is there any more
information on this, as this is a problem I would be interested in
working on.

Thanks, and keep up the good work!
-James
Indigo2 Solid Impact
Indigo2 non-impact
Indigo2 Solid Impact 10k