Read Me

This is an in-progress emulator that may simulate bits and pieces of:
- General Electric 625 (GECOS)
- General Electric 635 (GECOS)
- General Electric 645 (Multics)
- General Electric GIOC (625/635/645 I/O)
- Honeywell 6180 (Multics)
- Honeywell IOM (6180 I/O)
- Honeywell LCC/IPC-CONS-2 6601 (??6180 console)
- Honeywell Level/68 (Multics)
- Honeywell DPS8/70M (Multics)
- Honeywell DPS8/62M (Multics)
- Honeywell DPS8/52M (Multics)
- Honeywell LCC/EMC 6001 (??DPS8 console)
- Honeywell DPS88 ADP (GCOS)
- Honeywell DPS88 Orion (GCOS)
- Datanet-355 (6180/IOM terminal I/O)

The goal is to run Multics.  There's no real aim to run GECOS/GCOS, but
it'd be interesting to have the flexibility to support it.

The eventual Multics-supporting emulator will probably emulate most of a
6180, parts of an IOM, parts of a 6180 console, and maybe a DN-355.  That
should be enough to explore a live Multics system.

License is GPL, I guess, for now.

If you want to experiment with this, you'll probably need about 20MiB of
free disk space.  First, build it:
$ make
Then, try to run a quick test bootlabel that tests some address modification
logic:
$ ./testrun modtests
You should have a bunch of output files, assuming all tools and the emulator
ran without error.  Look at the shell script testrun to see what all the
files are and where they are generated.

There's some scattered symlinks that point outside of the ge635/ tree. 
Don't worry about them; they're just for development purposes.  You can
delete them if there presence is distracting.

There's some bits and pieces of code from MIT's random archive of files from
some Multics site.  It's copyright by MIT and Honeywell/Bull, but I'm
assuming by the fact that MIT has put it up for public access that using
short excerpts for testing is OK.  Bull is supposedly in the process of
preparing their Multics source archives for public release, after which
point playing with actual Multics tape images should become feasible.  That
is sometime in the future, though, so beware; we don't want to postpone
Bull's code release by having the party in charge of preparing the source to
have extra work on their hands, e.g. by someone accidentally posting a bunch
of stuff on-line that shouldn't be posted (at least not yet) in relation to
this project.

You should really read all of these things before starting to delve into the
code, and you'll probably want to print most of this out, get out a big
stapler, a big hole punch, a ton of three-ring binders, a stack of manila
folders, a package of assorted paper-clips and -binders, a bunch of pencils
and pens, tons of highlighters, some spiral notebooks, and be ready to sort,
take notes, and mark up, outline, and highlight (I know I did):
From <http://web.mit.edu/afs/net.mit.edu/reference/multics/>:
 - >ldd>hardcore>source>bound_bootload_1.s.archive
 - >ldd>hardcore>source>iom_bootload_tape_label.alm
 - >ldd>include>bootload_equs.incl.alm
 - >ldd>include>iom_word_macros.incl.alm
 - >ldd>include>iom_data.incl.alm
 - >ldd>include>tape_io_commands.incl.alm
 - >ldd>include>console_device_specs.incl.alm
 - >ldd>include>console_io_commands.incl.alm
From <http://bitsavers.org/pdf/>:
 - ge/GE-6xx/CPB-383_600-Series_Brochure.pdf
 - ge/GE-6xx/CPB1004F_GE635pgmRef_Jul69.pdf
 - ge/GE-645/GE-645_SystemMan_Jan68.pdf
 - honeywell/multics/AL39-01C_multicsProcMan_Nov85.pdf
 - honeywell/multics/AN70-0_systemInitPLM_Feb75.pdf
 - honeywell/multics/AN70-1_systemInitPLM_May84.pdf
 - honeywell/multics/AX49-01C_MPM_periphIO_Jul82.pdf
From Peter Flass' site <http://home.nycap.rr.com/pflass/>:
 - http://home.nycap.rr.com/pflass/Multics/mst.html

Some other useful or interesting places:
 - alt.os.multics from
   Mon Jan  3 20:24:08 EST 1994 through Wed Nov  8 19:48:59 EST 1995
   <http://stuff.mit.edu/afs/sipb/user/jsl/News/Alt.os.multics>
   Fri, 16 Apr 1993 09:44:08 GMT through current; not lynx-friendly
   <http://groups.google.com/group/alt.os.multics/about>
   Sat, 25 Mar 2006 11:30:03 GMT 7 through current; slow, no spam filtering
   <http://www.newsgroups.gs/newsgroup.asp?g=alt.os.multics>
   There used to be an Oakland gopher archive and a USL FTP site
   with historic newsgroup postings, but they seem to be long gone.
   Well, I guess it doesn't matter now that Google has merged multiple
   USENET archives into their ex-DejaNews Groups.
 - <http://en.wikipedia.org/wiki/Multics> has some interesting bits
   in the Links section
 - <http://www.cbi.umn.edu/iterations/haigh.html> has a lot to say
   about thvv's <http://www.multicians.org> and includes a few links
   to <http://slashdot.org> stories about Multics.
 - <http://febcm.club.fr/english/gecos_to_gcos8_part_1.htm>
   History of Large Systems at Bull.  This site has some other
   interesting papers, a few of which are in English, but most of
   the indices and navigation seem to be in French.
 - <http://stuff.mit.edu/afs/sipb/user/nathanw/work/multics-poetry/>
   I haven't found a link to this from any other site (came across it
   via Google) so I thought I'd throw it in, just for fun :-)
 - <http://stuff.mit.edu/afs/sipb/project/eichin/sipbscan/>
   This is another thing to which I cannot find any links.  It's scans
   of the first and last paying logins on MIT's Multics system.  Yes,
   a scan of the last login before MIT's site shut down.

--Matthew C. Williams "The Orange Squid"
mail to: os AT orangesquid DOT net
<http://os.1.vg/projects/ge635/>