donn@sdchema.UUCP (Donn Seeley) (08/15/83)
I inquired earlier over the net about the status of implementations of
4.2 Berkeley Unix on the 68K. The substance of the replies I received
was that:
- There is no 'public domain' (needs only Unix license) 68K
implementation either in existence or being planned.
- The only commercial implementation likely to appear in the near
future will be from Sun Microsystems, for the Sun architecture
machines. Sun apparently does not sell source unbundled from
hardware; I am unsure whether they will sell source at all, in
fact. I have not personally contacted Sun about this.
- Some companies may be working on other implementations but they
are not very close to release. No one seems to be working on
an implementation for new boards like the PMD 68010-based
system with dual-ported on-board memory (so that DMA devices
can go directly to CPU memory and not incur Multibus
overhead). This circumstance may not last for long.
Some people replied that they would be interested in seeing a 'public
domain' version of 4.2 on the 68K, especially for the new architectures
such as the dual-ported memory boards and Q-bus systems, and perhaps
could offer their services in helping to develop such a system.
I had grandiose thoughts of doing a port with just the resources here
at UCSD earlier, but decided not to get involved unless new funding was
approved and we got more staff to handle the effort. However if there
are enough interested systems programmers out there who would like to
engage in a 'public domain' 4.2 port, maybe a joint effort with
different people working in different parts of the world could do the
job regardless of the funding situation at individual sites. Things
like this never work exactly the way you expect them to, but there are
undoubtedly some very competent individuals out there who can flourish
without the handholding that a completely local development effort
provides.
I want to test the waters now and see just how many people would be
interested in contributing to a 'public domain' 68K 4.2 BSD port. UCSD
Chemistry would be willing to serve as a coordinating site for such a
venture. It would be a good test of the software development network
that UCSD Chemistry has suggested setting up as part of its recent NIH
grant proposal, as a matter of fact. As a starting point, we have
talked to individuals at Pacific Microsystems, who are located here in
San Diego and manufacture the 68010-based, dual-ported memory board
which I mentioned above, and they seem receptive to the notion of such
a port (in other words, they are not planning on making any software
effort of their own). We may be able to get discounts on hardware from
them if we can come to them with a concrete proposal; the word is that
they would be happy with a 'public domain' port rather than a
commercial port of their own, although we have not (repeat not) made
any deals with them yet.
If this sounds interesting to you, contact me through one of the
following addresses. If enough people are interested, I will try to
get an organization formed and post a note indicating what people's
feelings are about how we should proceed.
Donn Seeley UCSD Chemistry Dept. RRCF ucbvax!sdcsvax!sdchema!donn
(619) 452-4016 sdamos!donn@nprdc
USnail: Donn Seeley
UC San Diego Chemistry Dept.
NIH Research Resource Computing Facility, B-014
La Jolla, CA 92037