Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.6.2.17 $; site uiucdcs.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!whuxl!houxm!ihnp4!inuxc!pur-ee!uiucdcs!tapia
From: tapia@uiucdcs.UUCP
Newsgroups: net.lang.mod2
Subject: Re: ms-dos modula implementations?
Message-ID: <40300002@uiucdcs.UUCP>
Date: Fri, 15-Feb-85 15:25:00 EST
Article-I.D.: uiucdcs.40300002
Posted: Fri Feb 15 15:25:00 1985
Date-Received: Mon, 18-Feb-85 04:24:09 EST
References: <1161@watcgl.UUCP>
Lines: 32
Nf-ID: #R:watcgl:-116100:uiucdcs:40300002:000:1342
Nf-From: uiucdcs!tapia    Feb 15 14:25:00 1985


In december we finished the porting of the M2M Modula-2 compiler onto an
IBM-PC/XT. Documentation is now in preparation. To answer your points:

1. price. None, the cost of a floppy and mailing
2. yes, we have a debugger at the M-code level
3. support. Not reliable, students come and go.
4. probably most compatible with Lilith's implementation of Modula-2
5. can run off a floppy/hard disk
6. 2.5 mins. for a 1 page program (typical Sieve of Eratosthenes)
7. speed. Not too fast
8. Interpreted M-code
9. Yes, you must enter a subsystem to use it.
10. Subsystem is as good as you want it, hooks are provided for you to
    write your own shell. We have a primitive shell that runs programs
    directly or from a batch file.
11. Support for 8087 will exist when we get an assembler that handles
    8087 instructions.
12. runs of 200K. One module can't be larger than 32KB.
13. There is a way to get to assembly routines using the CODE keyword
    in a procedure declaration.
14. Bugs. No REAL type available
15. A single char is not considered to be an array of type char.
16. Except for REALs it's a full implementation as described by Wirth in
    Programming in M-2. (sorry, no IOTRANSFER routine)

Cris Tapia/ Prof. M. Harandi

Univ. of Illinois at Champaign-Urbana
222 Digital Computer Lab
1302 W. Springfied Ave.
Urbana, IL. 61801