[TUHS] Unisoft V7 68k sources (reviving a Codata machine)

[TUHS] Unisoft V7 68k sources (reviving a Codata machine)

[Warren Toomey]

I just received a new TUHS subscription request along with
an interesting query. Can anybody help Michael with his problem?

Cheers, Warren

----- Forwarded message from "Engel, Michael" -----
Dear Warren,

Could you please subscribe me to the TUHS mailing list?

I haven't worked with old Unix systems for quite some time, but I was appointed as a 
Senior Lecturer at Leeds Beckett University (UK) two months ago and - to my big 
surprise - I found an old Unix machine collecting dust in a corner..

The machine is a Codata 300, a Multibus-based system using a licensed clone of the
original Sun 100U CPU board and a number of additional Multibus controllers. The
machine seems to be complete, including two 8" SMD disks and a Cipher 9-track
tape drive, we also seem to have a set of replacement boards and the CPU board
manual (including schematics and code snippets explaining how to access the onboard
devices - some Codata documentation can also be found on bitkeeper).

We haven't tried to power up the machine yet (and, built around 1983, it certainly needs 
a close examination of the power supply and capacitors). From information on the net,
this machine runs a Unisoft port of 7th Edition Unix - similar to the Sun 1 machines and 
probably a Whitesmiths C compiler (there's a Whitesmiths license badge attached to the 
case). Definitely a very interesting and probably quite rare machine and we would like
to revive it (and, if things go well, create a FPGA reimplementation of the system in
the context of a student design project).

Now, I would love to know more details about the implementation of 7th Edition Unix on 
the 68000 and the use of the custom MMU built out of fast SRAM and TTL logic. 
I do not think that source code to any of the various 68k 7th Edition ports produced by 
Unisoft is available somewhere - do you know of a possible source?

Alternatively, do you think it would be worthwhile asking Unisoft for the source code or
do you know if anyone has tried this already? According to the Unisoft history web page 
(http://www.unisoft.com/history.html), they still seem to know that they were porting Unix 
30 years ago...

The only remotely related source code I could find in my archives is the A/UX 0.7 source 
(SVR2, if I'm not mistaken), which probably already required a 68020 with 68851 MMU.

Best regards,
     Michael Engel
----- End forwarded message -----

[TUHS] Unisoft V7 68k sources (reviving a Codata machine)

[Erik Fair]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 5056 bytes --]

I recently sent the following message to port-m68k at netbsd.org that relates to that period in time:



Begin forwarded message:

> From: Erik Fair <fair at netbsd.org>
> Subject: DUAL Systems documentation
> Date: February 22, 2014 at 02:12:17 PST
> To: port-m68k at netbsd.org
> 
> For those of you of a historical or digital archeological bent: while googling for something else, I came across this URL:
> 
> http://www.hartetechnologies.com/manuals/DUAL/
> 
> which contains PDFs of a few manuals & a price list from DUAL Systems of Berkeley, CA. DUAL produced the very first mc68000-based UNIX system (V7 UNIX on the S-100 (IEEE-696) bus), starting in November of 1981. I worked there as a UNIX systems programmer (kernel hacker) as my first job out of U.C. Berkeley from March 1983 to May 1985.
> 
> Of particular interest to this group: the CPU/MMU board manual. Please note: the mc68000 could not demand-page (i.e. no virtual memory) because the CPU did not save enough state from a memory access fault to restart the faulted instruction, so UNIX was a swapping OS (whole process in RAM, whole process out to swap area on disk if not enough RAM is available for whatever else needed to run).
> 
> Motorola corrected the fault state save in the mc68010, which provided the ability to make a proper VM system ... except for the mc68451 MMU (yes, a discrete MMU chip, in a package every bit as big as the CPU) which had only 32 descriptor registers (translation table entries). Motorola's theory was that if you wanted more (as one might for a lots of smaller-sized pages), add more MMU chips to your CPU board. Unfortunately, that doesn't fly on a board size as small as allowed by the S-100 standard. They were working on a proper paged MMU (PMMU) chip to go along with the mc68020 (the first fully 32-bit (address & data) 68k family CPU), but it was very, very late (DUAL was moving onto VMEbus from S-100 for the mc68020).
> 
> An aside, before Motorola shipped the mc68010, I saw another solution to the faulted-instruction state save problem from MassComp: their CPU board had two mc68k CPUs on it, one running a cycle behind the other, so a page fault would stop them both, the missing page could be paged in (I think they designed their own MMU; long since forgotten), and then execution resumed on the second, following CPU which hadn't taken the memory-access fault. Throw hardware at the problem ...
> 
> 	faulting my way down memory lane,
> 
> 	Erik <fair at netbsd.org>
> 	formerly {ihnp4,ucbvax,hplabs,decwrl,cbosgd,sun,nsc,apple,pyramid}!dual!fair


UniSoft Systems was started by Jeff Schriebman at the behest of Dual to port V7 UNIX to Dual’s system; they were just a few blocks away from Dual in west Berkeley, on 6th Street, if memory serves.

CoData was one of Dual’s competitors - I usually saw their gear at vendor shows (e.g. when /usr/group held its vendor show co-located with USENIX conferences). One of the problems at the time that bedeviled UniSoft was that every 68k-based Unix box had its own unique MMU because Motorola’s offerings in that space (as noted above) were … lacking. So UniSoft had to redo the kernel MMU code for every strange and wonderful MMU that each hardware vendor came up with. Pixel, Plexus, Fortune, Codata, Apple (yes, UniSoft did a port of Unix to the Lisa and the awful 5MB “profile” drive on a parallel port (!!)), ...

I have at least two old Dual boxes in my garage that I keep for sentimental reasons, but I have no idea what’s on those disks, or if they’ll even spin up after so long. They’re ST-506/419 interface 5.25 inch “winchester” drives; a forerunner of a sort of the IBM PC’s IDE. I wrote the drivers for the Morrow Designs (“ThinkerToys”) S-100 controller as my first project for Dual in March/April 1983.

Previously, Dual was using 8-inch Memorex hard drives controlled by a CompuPro (nee “Godbout Electronics”) controller whose interface standard I don’t recall. I’ll have to see if they can be fired up again, and see what’s on the hard drives. Dual was very particular about using big, heavy linear power supplies in its systems as a matter of reliability, over the (to their mind) squirrelly switching power supplies that many of our competitors were using.

The other way to go to get Unix going on that old Codata box would be a port of NetBSD/m68k - it supports a wide array of mc68k systems:

	http://www.netbsd.org/ports/#ports-by-cpu

The trick would be writing the MMU code, and a booter that would work with whatever firmware Codata used back in the day. I had to work on Dual’s boot firmware from time to time, and I recall it being fairly simple (“find /unix and see if it’s a.out, load, and jump”), and usually the trick is to make a secondary booter that the system firmware will load (in whatever executable format the firmware will recognize) to then parse a modern ELF kernel from FFSv2.

NetBSD continues to support a whole lot of pretty old hardware - we believe in portability.

	Erik <fair at netbsd.org>

[TUHS] Unisoft V7 68k sources (reviving a Codata machine)

[Engel, Michael]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2190 bytes --]


Hi Steve, 

whow, what a small world!

On 03 Oct 2014, at 20:03, Steve Simon <steve at quintile.net> wrote:

>> The machine is a Codata 300
> 
> Wow.
> 
> I went to Leeds Poly (as it was then) in the mid 1980s.
> 
> There where two Codatas in the electronics dept., one in its
> original plastic case and one 19inch rackmount - built as a
> IEEE 488 controller; I assume what you have is one of those.

The machines were a bit cannibalised over the last decades,
we actually have the boards from both machines, one of them
build into the 19" rack machine. The machine survived the 
closing down of the electronics department, now we're busy
building everything up again... 

> The former machine was loaded with Whitesmiths cross compilers
> and I learnt z80 assembly language on it ☺

There's still a Whitesmiths logo attached to the case - this is what
the machine looks like today: http://multicores.org/Codata.jpg
A 5.25" ST412 disk, two 8" SMD disks and a Cypher 9 track tape
are also still there.

> It ran V7 indeed, and was a friend of the Interdata/Perkin Elmer 
> 3210, the  main electronics teaching machine. If it is this
> machine then it should have the V7 source from the 3210 (Xelos
> as it was called) and the source for the drivers for the
> codata (which we gained by "accident").

Whow, let's keep our fingers crossed that the disks still work...
I haven't tried powering the machine up so far. Unfortunately,
the Interdata no longer exists (what a pity), but the first thing
I will do when the machine is running is to run a full backup. 
Luckily, the Codata has a 5.25" floppy drive, since I no longer 
have easy access to a 9-track SCSI drive (and the last time 
I set up UUCP is far too long ago...).

> I may be able to remember some other snippets - contat me 
> off-list with specific questions. I can give you the names
> of the lecturers who would know most about it but I guess they
> are now retired (though they may still be Headingly somwhere).

Actually, quite a number of the people are still here. More off-list.

> (fondly remembers Leeds).

I only came to Leeds about two months ago - and so far, it's really
great here!

Best wishes,
    Michael

[TUHS] Unisoft V7 68k sources (reviving a Codata machine)

[Steve Simon]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 954 bytes --]

> The machine is a Codata 300

Wow.

I went to Leeds Poly (as it was then) in the mid 1980s.

There where two Codatas in the electronics dept., one in its
original plastic case and one 19inch rackmount - built as a
IEEE 488 controller; I assume what you have is one of those.

The former machine was loaded with Whitesmiths cross compilers
and I learnt z80 assembly language on it ☺

It ran V7 indeed, and was a friend of the Interdata/Perkin Elmer 
3210, the  main electronics teaching machine. If it is this
machine then it should have the V7 source from the 3210 (Xelos
as it was called) and the source for the drivers for the
codata (which we gained by "accident").

I may be able to remember some other snippets - contat me 
off-list with specific questions. I can give you the names
of the lecturers who would know most about it but I guess they
are now retired (though they may still be Headingly somwhere).

(fondly remembers Leeds).

-Steve