CP/M-68K emulator

251 views
Skip to first unread message

Larry Sonderling

unread,
Jun 14, 2007, 1:56:53 PM6/14/07
to
Does anyone know of a CP/M-68K emulator that preferably runs under
UNIX/Linux? My purpose is to do some work on the Alcyon C compiler source
without having to do a major rewrite up front. It is written in a very
old-style C dialect which allows things like unnamed structures/unions.

TIA,

--Larry


Al Kossow

unread,
Jun 14, 2007, 8:40:24 PM6/14/07
to
Larry Sonderling wrote:
> Does anyone know of a CP/M-68K emulator that preferably runs under
> UNIX/Linux?


I did a lot of digging this afternoon, and it doesn't appear anyone
has a running CP/M-68K in simulation. It looks like there is enough
documentation and software for a simulation of the SAGE

http://www.thebattles.net/sage/sage.html

and there should be all of the simulated peripherals needed in MESS

http://mess.org

--
Posted via a free Usenet account from http://www.teranews.com

Tarkin

unread,
Jun 14, 2007, 10:01:23 PM6/14/07
to

Aeons ago, there was BasiliskII ...
http://basilisk.cebix.net/

IIRC, the Windows version was fidgety, but functional
enuogh for working with Mac floppies and doing
modem comms (ala Kermit) with a fake phone
line on my Powerbook 145b; the Linux version
didn't work 'out-of-the-box' for me, though I never
poked around with it much (too impatient and the
Win version did work).

Give that a shot....and hint, hint, google
around for the ROMs- I felt comfy doing that because
I own two Appple machines, so I never felt like I was
defrauding them or robbing them.

TTFN,
Tarkin

Tarkin

unread,
Jun 14, 2007, 10:08:58 PM6/14/07
to
On Jun 14, 1:56 pm, "Larry Sonderling" <lsonderl...@earthlink.net>
wrote:
My bad. Sorry for the double post, but I just
realized you wanted 68K *CPM* sim, not just a
68K sim.

Apologies,
Tarkin

Larry

unread,
Jun 14, 2007, 11:17:11 PM6/14/07
to
Al Kossow wrote:
> Larry Sonderling wrote:
>> Does anyone know of a CP/M-68K emulator that preferably runs under
>> UNIX/Linux?
>
>
> I did a lot of digging this afternoon, and it doesn't appear anyone
> has a running CP/M-68K in simulation. It looks like there is enough
> documentation and software for a simulation of the SAGE
>
> http://www.thebattles.net/sage/sage.html
>
> and there should be all of the simulated peripherals needed in MESS
>
> http://mess.org
>

Al, thanks. I had not thought of going this particular route but I will
look at it carefully as it may be the only way to do what I am trying to do.

Larry

unread,
Jun 14, 2007, 11:18:54 PM6/14/07
to

Thanks, Tarkin. I think the 68K Macs would run CP/M-68k so I will
alosolook at this.

Herb Johnson

unread,
Jun 16, 2007, 2:31:29 PM6/16/07
to
> > Larry Sonderling wrote:
> > > Does anyone know of a CP/M-68K emulator that preferably runs under
> > > UNIX/Linux?

> > On Jun 14, 8:40 pm, Al Kossow <a...@spies.com> wrote:
> > I did a lot of digging this afternoon, and it doesn't appear anyone
> > has a running CP/M-68K in simulation. It looks like there is enough
> > documentation and software for a simulation of the SAGE
>
> >http://www.thebattles.net/sage/sage.html
>
> > and there should be all of the simulated peripherals needed in MESS
>
> >http://mess.org

On Jun 14, 10:01 pm, Tarkin <Tarkin...@gmail.com> wrote:
> Aeons ago, there was BasiliskII ...http://basilisk.cebix.net/


>
> IIRC, the Windows version was fidgety, but functional
> enuogh for working with Mac floppies and doing
> modem comms (ala Kermit) with a fake phone
> line on my Powerbook 145b;

I recommend a through search of comp.os.cpm for discussions about CP/M
68K and the Macintosh. But the bottom line on that appears to be, that
the one known commercial attempt never succeeded to produce a product,
and the company who contracted for that development could not release
what WAS done, and eventually got rid of their CP/M 68K stuff anyway.
I suggest a Macintosh port of CP/M 68K would be at best a distraction
unless that is EXACTLY what you wanted to end up with, and only if you
could do it from scratch, yourself!

Whereas, I assume CP/M 68K only needs a 68K processor, memory, a
floppy, and a console port to run. So a simple simulator of the 68K
which could access some kind of file system and some kind of character
terminal, "should" be able to accept a port of CP/M 68K. Al's
suggestion of other 68K emulators may be a good start; one would not
need most of their simulated hardware. And sources for Z80 and CP/M
emulators under MS-DOS or Linux can show you how to simulate the I/O
and file access needed.

Udo Munk's Z80pack emulator for Linux has been discussed recently in
this newsgroup, it's open sourced and in active development, and
supports CP/M for the 8080 in several flavors.

http://www.unix4fun.org/z80pack/

Or, there may be some old-school MS-DOS 68K simulator, which could be
run under some Linux MS-DOS emulator, or "straight" on some old
Pentium box, to bootstrap to something more acceptable. Someone may
have made a 68K emulator for CP/M-86, so check the relevant archives.

CP/M 68K did run on some S-100 systems. There are some 68K cards and
CP/M-68K disks from Compupro, and Dual made a 68K AND Z80 two
processor card. Cromemco made a 68K but they ran flavors of UNIX. I
happen to have a pile of 68K-class VME cards from Motorola, but they
also ran Unix. It would be interesting to know what machines actually
RAN CP/M 68K, for real, with evidence, not "I heard that" or "it was
developed but.."

But a real CP/M-68K environment in emulation on some Pentium-class box
could attract interest and attention, just as Munk's recent work has.

Herb Johnson

Herbert R. Johnson, New Jersey USA
http://www.retrotechnology.com/herbs_stuff/ web site
http://www.retrotechnology.net/herbs_stuff/ domain mirror
my email address: hjohnson AAT retrotechnology DOTT com
if no reply, try in a few days: herbjohnson ATT comcast DOTT net
"Herb's Stuff": old Mac, SGI, 8-inch floppy drives
S-100 IMSAI Altair computers, docs, by "Dr. S-100"

primo

unread,
Jun 16, 2007, 4:50:23 PM6/16/07
to
On Sat, 16 Jun 2007 11:31:29 -0700, Herb Johnson
<herbrj...@gmail.com> wrote:

>
>CP/M 68K did run on some S-100 systems. There are some 68K cards and
>CP/M-68K disks from Compupro, and Dual made a 68K AND Z80 two
>processor card. Cromemco made a 68K but they ran flavors of UNIX. I
>happen to have a pile of 68K-class VME cards from Motorola, but they
>also ran Unix. It would be interesting to know what machines actually
>RAN CP/M 68K, for real, with evidence, not "I heard that" or "it was
>developed but.."
>

Trisoft, the folks in Austin TX did a version of 68k cp/m for the
radio shack mod 16/6000, and I thought the person who used to post
here that worked there was going to hand out copies to those who asked
for it. He had a box of mod 16 memory boards that he was going to let
me have if I ever got to Austin, but at that time my car went into the
shop for repairs and I never made the trip.

I have their version of P&Tcp/m80 for the 2/12/16/6000 and it works
well. My 16 is like a giant mod 4p (which i run Montezuma's cp/m on).

Larry Sonderling

unread,
Jun 16, 2007, 5:24:30 PM6/16/07
to

"Herb Johnson" <herbrj...@gmail.com> wrote in message
news:1182018689.8...@q75g2000hsh.googlegroups.com...

Herb,

Thanks for your very interesting thoughts. I am familiar with Udo's
Z80pack, a wonderful laborof love and of great value to all who cut their
teeth on CP/M in the 1970s. As you infer, simple is best to start with and
a simple simulator with minimal peripherals would be a good beginning. I
think there is a freely available MC68K CPU engine out there so the really
hard work is, I think, already done. I had not looked for an MS-DOS based
emulator but I will check my program archives and post what I find out.

I have gotten a lot of good from this newsgroup and I would like to find a
way to give something back to those who so generously contributed here.

--Larry


Al Kossow

unread,
Jun 16, 2007, 6:43:40 PM6/16/07
to
Larry Sonderling wrote:

> a simple simulator with minimal peripherals would be a good beginning.

After the initial learning curve, I think you will find that MESS will
do what you need done.

David Snowdon

unread,
Jun 16, 2007, 9:00:57 PM6/16/07
to
Emulators for the AtariST (68000 machine) are available on the internet,
as well as the Alcyon C compiler for the ST.

The ST's early design (before being released) used the CP/M-68K OS,
which later became part of it's GEMDOS OS.

David, VA3DKS

Hauke Fath

unread,
Jun 17, 2007, 7:59:52 AM6/17/07
to
David Snowdon <nor...@netscape.ca> wrote:

> The ST's early design (before being released) used the CP/M-68K OS,

...no, it didn't. That confusion came about because Atari shipped with
their Development Kit all docs they could photocopy, and there was some
CP/M 68k stuff inbetween.



> which later became part of it's GEMDOS OS.

GEMDOS was an MSDOS 2 clone, and had to be, since that was the interface
GEM expected.

hauke

--
Eine Linux User Group, um soziale Blockaden gegenüber Frauen
abzubauen? Da wäre ja ein Schwulengesangsverein noch produktiver.

{David Kastrup @ d.t.r}

Peter Dassow (remove the NOSPAM. for direct answer)

unread,
Jun 17, 2007, 12:32:34 PM6/17/07
to
Hauke Fath wrote:
> GEMDOS was an MSDOS 2 clone, and had to be, since that was the interface
> GEM expected.

That sounds strange. TOS was developed from CP/M sources, see also at
the copyright message (Digital Research). I do not know what "GEMDOS" is
exactly, but Atari ST's development roots are NOT related with MSDOS in
any way.

Regards
Peter
--
My new project: Try http://www.z80.eu for CP/M computer infos.

Hauke Fath

unread,
Jun 17, 2007, 2:32:49 PM6/17/07
to
Peter Dassow wrote:

> Hauke Fath wrote:
> > GEMDOS was an MSDOS 2 clone, and had to be, since that was the interface
> > GEM expected.
>
> That sounds strange. TOS was developed from CP/M sources, see also at the
> copyright message (Digital Research).

That's a non sequitur.

> I do not know what "GEMDOS" is exactly, but Atari ST's development roots
> are NOT related with MSDOS in any way.

GEM was developed as a GUI on top of MSDOS for IBM PC compatibles. In
order to bring GEM to 68k machines, Digital Research had to come up with
something that implemented the MSDOS 2 interfaces (think: hierarchical
filesystem). The result was GEMDOS.

Atari, OTOH, developed the hardware, and maybe the {X,}BIOS and GEM
drivers. TOS _is_ DRI GEMDOS, plus tweaks and patches. Apparently the
license that Atari bought from DRI was reasonably complete to make them
independent from the GUI lawsuit between Apple and DRI which resulted in
the crippling of DRI's 80x86 GEM.

Thierry B.

unread,
Jun 17, 2007, 4:54:29 PM6/17/07
to
--{ Peter Dassow (remove the NOSPAM. for direct answer) a plopé ceci: }--

> That sounds strange. TOS was developed from CP/M sources, see also at
> the copyright message (Digital Research). I do not know what "GEMDOS" is
> exactly, but Atari ST's development roots are NOT related with MSDOS in
> any way.

Mmmmm, filesystem of TOS was compatible (mmm around 98%) with
the MS-Dos messyfs...


--
>>> How much does a slrn weigh?
>> 42.
> 42 in African or European units?
British units.

John Elliott

unread,
Jun 17, 2007, 6:51:35 PM6/17/07
to
Hauke Fath <spam...@espresso.rhein-neckar.de> wrote:
: David Snowdon <nor...@netscape.ca> wrote:

:> The ST's early design (before being released) used the CP/M-68K OS,

: ...no, it didn't. That confusion came about because Atari shipped with
: their Development Kit all docs they could photocopy, and there was some
: CP/M 68k stuff inbetween.

<http://www.atarimagazines.com/startv3n1/threeyearsofst.html> disagrees
with you:

< The version of GEM first demonstrated to Tramiel Technology was actually
< running on top of CP/M-68K. Indeed, GEM's CP/M-68K incarnation was the only
< version of GEM available for the 68000 microprocessor at the time, and
< Atari would continue to plan for it to be the ST's underlying operating
< system in the months to come.

Then there's
<http://groups.google.co.uk/group/net.micro.atari16/browse_thread/thread/2d42447427c7dc5c/1bae760666787473?lnk=st&rnum=2&hl=en#1bae760666787473>
which has a suggestion I hadn't seen before: that the proposed solution was
something like DOS Plus, with a DOS emulator running on a CP/M-68K kernel.

: GEMDOS was an MSDOS 2 clone, and had to be, since that was the interface
: GEM expected.

According to the first article, the prototype STs demonstrated in January
1985 were still running CP/M-68K.

--
John Elliott

BobH

unread,
Jun 17, 2007, 8:35:09 PM6/17/07
to
Herb Johnson wrote:
> Whereas, I assume CP/M 68K only needs a 68K processor, memory, a
> floppy, and a console port to run.

This is correct. CPM68K did not require the use of interrupts or
anything other than a vanilla floppy disk interface and a serial port.
One could write a BIOS that used interrupts, but they were not required
by the OS.


It would be interesting to know what machines actually
> RAN CP/M 68K, for real, with evidence, not "I heard that" or "it was
> developed but.."

CPM68K was developed on the Exormacs system from Motorola. The sample
docs show a BIO written in C for the Exormacs.

It should be pretty easy to port to a VME SBC that has a 68K, some RAM,
a console port and the floppy controller. CPM68K also ran on the 68010
and the 68020. Many years ago, I spoke with one of the technical people
at Compupro that did a 68020 processor board that he said ran CPM68K
with no mods.

Bill

unread,
Jun 17, 2007, 11:24:49 PM6/17/07
to
On Sat, 16 Jun 2007 11:31:29 -0700, Herb Johnson
<herbrj...@gmail.com> wrote:

>It would be interesting to know what machines actually
>RAN CP/M 68K, for real, with evidence, not "I heard that" or "it was
>developed but.."

1984:

I used something I believe was called a Dimension Rainbow.

It had 80 track 5-1/4'' drives, but also plug-in cards for other
CPU's.

The motherboard ran a 68000 using CP/M-68k, and if you
had the three 'other' CPU boards as I had, ie an 8086, a Z-80,
and I think it was a 6512, you could boot and run any of their
respective operating systems natively. Not all at the same
time.

Those 80 track drives read and wrote Apple II disks just fine.

No mean feat.

This beast cost 2-3x what one of each computer would have.

Probably resulted in very very few of them ever being sold,
however I did see one go for around $80 maybe 5 years ago
on ebay, so who knows.

Bill

Bill

unread,
Jun 17, 2007, 11:32:28 PM6/17/07
to
On Thu, 14 Jun 2007 17:40:24 -0700, Al Kossow <a...@spies.com> wrote:

>I did a lot of digging this afternoon, and it doesn't appear anyone
>has a running CP/M-68K in simulation. It looks like there is enough
>documentation and software for a simulation of the SAGE

Well, unless this thing can do it:
(Run CP/M-68k under Windows using a simulator maybe?)

http://faculty.uwb.edu/aberger/CSS422SP07/Homework/Homework%203/CSS422SP07HW3.htm

Bill

Axel Berger

unread,
Jun 18, 2007, 9:58:00 AM6/18/07
to
*Thierry B.* wrote on Sun, 07-06-17 22:54:

>Mmmmm, filesystem of TOS was compatible (mmm around 98%) with the MS-Dos
>messyfs...

As I remember it, it was 100% identical to the published specs and only
when swapping disks failed did they find out that MS-DOS itself wasn't.
The c't had some nice articles at the time about trying to get a 5.25"
drive working with an ST. At the time there were no 3.5" in clones,
only Atari had those.

Herb Johnson

unread,
Jun 18, 2007, 9:35:47 PM6/18/07
to
On Jun 17, 11:32 pm, Bill <b...@sunsouthwest.com> wrote:

> Well, unless this thing can do it:
> (Run CP/M-68k under Windows using a simulator maybe?)
>

> http://faculty.uwb.edu/aberger/CSS422SP07/Homework/Homework%203/CSS42...

The sites this page refers to include the following:

http://www.monroeccc.edu/ckelly/EASy68K.htm

EASy68K Editor/Assembler/Simulator - homepage
Also other links there.

http://www-scm.tees.ac.uk/users/a.clements/software.htm

Alan Clement's Teeside assembler/simulator, the basis for other
simulator/assemblers.
Also on that site, courses and tutorials on line for assembly language
computing.
Note in particular:

http://www-scm.tees.ac.uk/users/a.clements/pofc2.htm

.....Principles of Computing II, assembly on the 68000.

http://members.lycos.co.uk/leeedavison/68k/index.html

Linked from the EASy68K page, some various 68K computers there and
some programs.

roman pollak

unread,
Jun 21, 2007, 8:15:47 AM6/21/07
to
Larry, I wrote a native emulator for Linux 68k. It emulates the
bdos/bios calls of the cp/m 68k and maps it to linux filesystem.
The 68k code runs native on the cpu (in my case 68040).
The Alcyon C compiler works fine on it (actually the whole compiler
suit) and also the generated code ;)

regards
roman

CBFalconer

unread,
Jun 21, 2007, 1:01:14 PM6/21/07
to
roman pollak wrote: *** and top-posted - fixed ***

> Larry Sonderling wrote:
>
>> Does anyone know of a CP/M-68K emulator that preferably runs
>> under UNIX/Linux? My purpose is to do some work on the Alcyon C
>> compiler source without having to do a major rewrite up front.
>> It is written in a very old-style C dialect which allows things
>> like unnamed structures/unions.
>
> Larry, I wrote a native emulator for Linux 68k. It emulates the
> bdos/bios calls of the cp/m 68k and maps it to linux filesystem.
> The 68k code runs native on the cpu (in my case 68040).
> The Alcyon C compiler works fine on it (actually the whole
> compiler suit) and also the generated code ;)

Please do not top-post. Your answer belongs after (or intermixed
with) the quoted material to which you reply, after snipping all
irrelevant material. See the following links:

--
<http://www.catb.org/~esr/faqs/smart-questions.html>
<http://www.caliburn.nl/topposting.html>
<http://www.netmeister.org/news/learn2quote.html>
<http://cfaj.freeshell.org/google/> (taming google)
<http://members.fortunecity.com/nnqweb/> (newusers)

James M. Knox

unread,
Jun 23, 2007, 10:18:03 AM6/23/07
to
primo <jgra...@suddenlink.net> wrote in
news:8ni873p3fh5mtuevs...@4ax.com:

> On Sat, 16 Jun 2007 11:31:29 -0700, Herb Johnson
> <herbrj...@gmail.com> wrote:
>>CP/M 68K did run on some S-100 systems. There are some 68K cards and
>>CP/M-68K disks from Compupro, and Dual made a 68K AND Z80 two
>>processor card.

SAGE made an S-100 board, if I recall. There were also a couple of
companies that made 68K boards for the IBM-PC. We port'd CP/M-68K to a
number of them, but most of those companies blew through their startup
money before they ever got to any significant market. [One of them was
run by a couple of 20 year olds, with the most impressive offices (and
secretaries) I've ever seen. Ran through two million dollars in six
months, and never got around to making the production boards. <G>]

> Trisoft, the folks in Austin TX did a version of 68k cp/m for the
> radio shack mod 16/6000, and I thought the person who used to post
> here that worked there was going to hand out copies to those who asked
> for it. He had a box of mod 16 memory boards that he was going to let
> me have if I ever got to Austin, but at that time my car went into the
> shop for repairs and I never made the trip.

We gave out copies to several people, plus a lot of manuals, with the
understanding that they would made available to others. [Note: That
doesn't mean we have any problem with Herb charging for the effort to
make and send those copies out.]

There was a gentleman, not related to TriSoft, that did a Macintosh port
of CP/M-68K. Actually, he did a couple of them. But he never did get
it up to what we would call "commercial" standards.

-----------------------------------------------
James M. Knox
TriSoft ph 512-385-0316
1300 Koenig Lane West fax 512-371-5716
Suite 200
Austin, Tx 78756 jk...@trisoft.com
-----------------------------------------------

Hauke Fath

unread,
Jun 24, 2007, 10:25:08 AM6/24/07
to
John Elliott <j...@seasip.demon.co.uk> wrote:

> :> The ST's early design (before being released) used the CP/M-68K OS,
>
> : ...no, it didn't. That confusion came about because Atari shipped with
> : their Development Kit all docs they could photocopy, and there was some
> : CP/M 68k stuff inbetween.
>
> <http://www.atarimagazines.com/startv3n1/threeyearsofst.html> disagrees
> with you:
>
> < The version of GEM first demonstrated to Tramiel Technology was actually
> < running on top of CP/M-68K. Indeed, GEM's CP/M-68K incarnation was the only
> < version of GEM available for the 68000 microprocessor at the time, and
> < Atari would continue to plan for it to be the ST's underlying operating
> < system in the months to come.

That article sounds a bit unsubstantiated in many places... it seems to
concentrate on Atari gossip, and the financial side, largely. The
confusion between CP/M 68k and GEMDOS is widespread, btw, see e.g.

http://en.wikipedia.org/wiki/AtariTOS
http://www.vanc.igs.net/~roughley/tos_main.htm
http://www.old-computers.com/museum/computer.asp?c=922

There's a huge technological difference between CP/M 68k which, being
derived from CP/M 2.2, is little more than a program loader, and GEMDOS,
which includes a hierarchical filesystem (FAT), support fro device
drivers, and a limited process management (which gave you desk
accessories on the Atari).

| In September 1984, Atari sent most of its software engineers down to
| Digital Research's facilities in Monterey, California, to begin the
| port of GEM. The software group lived there for most of the following
| four months, some staying on longer. Their task was to translate
| Digital's IBM PC assembly code, written for the 8086 microprocessor,
| into 68000 code, and also to rewrite and recompile Digital's C code on
| the ST system.

Since GEM was running on the Apple LISA already, using GEMDOS,

http://datenbrei.de/?Programming/GEMDOS/About

and it was written in C, what 8086 assembler code was to be translated?
They seem to be claiming that GEMDOS was written by Atari from scratch
to MSDOS interface specs? Looking at

http://dylanharris.org/prose/gem/index.html

it becomes clear that GEMDOS, as well as GEM, was written by DRI. Unlike
Atari, they had quite a track record for operating systems by then...

John Elliott

unread,
Jun 24, 2007, 12:19:33 PM6/24/07
to
Hauke Fath <spam...@espresso.rhein-neckar.de> wrote:
: Since GEM was running on the Apple LISA already, using GEMDOS,

: http://datenbrei.de/?Programming/GEMDOS/About

: and it was written in C, what 8086 assembler code was to be translated?

The only version of GEMDOS for the Lisa that I know of is dated 15th
August 1985, a few months after Intel GEM/1 was released, so that's quite
consistent with being a port of the 8086 version.

As to what assembly is used in 8086 GEM, large parts of the drivers are
written in asm, and important bits of the AES.

--
John Elliott

Hauke Fath

unread,
Jun 24, 2007, 4:59:10 PM6/24/07
to
John Elliott <j...@seasip.demon.co.uk> wrote:

> As to what assembly is used in 8086 GEM, large parts of the drivers are
> written in asm, and important bits of the AES.

Okay, given the intended target platform's performance (8088, CGA/MGA),
that sounds reasonable...

Reply all
Reply to author
Forward
0 new messages