[ I went ahead and CC'd port-mvme68k, since there's actually some people
  on it now, and this might be useful for the general NetBSD/mvme68k
  community :-)  --thorpej ]

On Wed, 03 Jul 96 09:17:00 PDT 
 Steve Woodford <SWoodfor@lsi-dsp.co.uk> wrote:

 > Hi Jason,

Hi Steve...

 > According to chuck, it *should* link with SRT0.o since it's now a two-
 > level bootstrap; mvme68k/stand/bootxx is the code which now lives
 > in the bootsector. bootsd now lives in the root filesystem.

Ah, ok ... I remember Chuck sending me some mail about this ...

I'm going to try and get my 147 running again this weekend ... I need to 
reconfigure my terminal server, since I have to hook my Alpha's console 
up now, too ... thusfar, my little Cisco box has been sitting mostly idle.

My 147 has been off since my SPARC ate its root filesystem for lunch... Bleh.

Also, I seem to be having a HELL of a time getting my cage's front panels 
off.  Darn aluminum screws have totally siezed up, and stripped out.  I'm 
having a bit of a SCSI problem that I believe to be termination-related.  
I don't think I'm getting termination power out to the bus... I don't 
have the ACE key for the drive sled locks, so I have to figure out some 
other way to enable the sled power.  I've got some hardware hacks in 
mind, but unfortunately, they quite nearly require a machine shop :-)

 > BTW, excellent work on the VME nexus in the mvme68k port. I'm

Thanks.  Chris Demetriou has mentioned MI VME support as a "good idea" 
before, and when I sat down to add VME support to the MVME port (seemed 
appropriate, eh? :-), I thought "Why not make a stab at something 
relatively MI..."  Anyhow, I haven't actually done much testing of it, 
yet.  I have a Ciprico Tapemaster and Rimfire that I'm going to write 
drivers for eventually, but I just haven't had the time ... So, if you 
find bugs in it, I wouldn't be surprised :-)

 > thinking of writing the first driver for it. How does this grab you:
 > A packet driver over the vme backplane. No more need for ethernet
 > links between mvme68k cpus in the same vme crate. In my case,
 > the transceivers cost me more than I paid for four cpus and the crate!

Sounds like a very cool idea.  If I recall, Integrated Solutions had 
something like this for their Optimum VME-68020 systems (I have 4 of 
these bad boys ... one of them is currently running 4.2BSD :-).

 > I'm still putting the design together in my mind at the moment, and
 > when time allows and all three of my 147's are running at least 1.2,
 > I'll make a start on it.

Cool!  Keep us updated!  Sounds like an excellent project!

 -- save the ancient forests - http://www.bayarea.net/~thorpej/forest/ -- 
Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939