Xref: utzoo comp.sys.3b1:361 unix-pc.general:7522
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!decwrl!public!thad
From: thad@public.BTR.COM (Thaddeus P. Floryan)
Newsgroups: comp.sys.3b1,unix-pc.general
Subject: Re: 3B1 680X0 daughter boards (Posted from Thad)
Keywords: chip
Message-ID: <1833@public.BTR.COM>
Date: 19 Feb 91 11:23:38 GMT
References: <66@morwyn.UUCP>
Followup-To: comp.sys.3b1
Organization: BTR Public Access UNIX, MtnView CA, Contact: cs@btr.com 415-966-1429
Lines: 39

forrie@morwyn.UUCP (Forrie Aldrich) in <66@morwyn.UUCP> writes:

	You mentioned a short while back something regarding Daughter Boards,
	etc, that Motorola was using to upgrade their 68000 chips.  Could you
	send me email and elabortate on the problems they had... and the
	problems with porting these things to the 3b1?  Also, the necessary
	changes that would need to be made to the KERNEL to allow this to take
	place.

They had no problems (to my knowledge).  The ones used by Motorola to upgrade
their 68000 Trainers were designed and manufactured by Computer Systems
Associates [CSA] (San Diego, CA) and they worked fine in the trainers.

CSA went on to make upgrade cards for other 68000-based systems and, in
general, everything works well as long as people play by the rules (i.e. no
attempted use of "MOVE SR,..." in user mode, not using the upper 8-bits of a
pointer for unrelated data, no self-modifying code, etc.) and one takes into
account the stack frame differences between members of the 680x0 family (one
method is by checking the type ID in the frame).

CSA presently manufactures 68020/68881 and 68030/68882 (and possibly 68040)
daughterboards for several systems as do several other companies.  You could
pick up recent copies of Amiga, Mac, and several other 680x0-based systems'
magazines to see a large range of products that retrofit 68000 and 68010
systems.

Everything has been clearly documented in Motorola Application Notes, and much
has been published in the trade press (i.e. Electronic Design, EDN, etc.)

Motorola claims it requires only a few hours to fix ANY kernel to handle the
entire line of 680x0 chips PROVIDED ONE HAS SOURCE.

Several people having access to the source code have casually informed me
they have "some" interest in doing this project but, like everything else,
such project(s) is done during hobby-time, so don't expect four-color flyers
and full-page ads in "3B1 TIMES" or "UNIX-PC/7300 JOURNAL" along with a
coast-to-coast dealer network!  :-)

Thad Floryan [ thad@btr.com (OR) {decwrl, mips, fernwood}!btr!thad ]