[comp.sources.d] Christmas list

mpatnode@polyslo.UUCP (Mike Patnode) (12/18/86)

I've never seen a line eater, I have to read these.

Everybody wants something and I'm no different. Here is my Public Domain
Christmas list:

1. A freeware IBM-PC Macro Assembler (is there such an animal?)
2. RELEASE.COM (EXE?) (recently posted memory management program)
3. Documentation for PC-ARC

------------------------------------------------------------------
Mike "Dodger" Patnode          |   ..csustan!polyslo!mpatnod(e) (>100k)
Instructional Support Group    |              or
(but really just a student)    |       mpatnod(e)@polyslo.UUCP
Cal Poly State University      |              or	
at San Luis Obispo             |       dpatnod@calstate.BITNET
(805) 546-2516                 |
?"ver" rof esu lufesu a wonk yllaer ydobyna seod
-------------------------------------------------------------------

bobmon@iuvax.UUCP (Robert Montante) (12/20/86)

>I've never seen a line eater, I have to read these.

In article <256@polyslo.UUCP> mpatnode@polyslo.UUCP (Mike Patnode) writes:
	[...]
>1. A freeware IBM-PC Macro Assembler (is there such an animal?)

There's a bulletin board in northern Indiana that has such a beast listed.
Claims "better than MASM" -- I intend to download it, maybe now that the
term's over....it runs 121K of ARC'ed material, and a related debugger file
is 57K :

	The Programmer's Room
	1 317 742-5533
	1200baud 8bit is okay

>3. Documentation for PC-ARC

The same bulletin board has a couple of ARC programs (ARC51, PKARC) that include
the operating documentation (not source, though).

broman@cod.UUCP (Vincent P. Broman) (12/30/86)

In article <2271@iuvax.UUCP>, bobmon@iuvax.UUCP (Robert Montante) writes:
> In article <256@polyslo.UUCP> mpatnode@polyslo.UUCP (Mike Patnode) writes:
> 	[wishlist]
> >1. A freeware IBM-PC Macro Assembler (is there such an animal?)
> 
> There's a bulletin board in northern Indiana that has such a beast listed.
> Claims "better than MASM" --

I downloaded a86.arc and d86.arc from the San Diego Computer Society BBS.
This is a shareware assembler and debugger for the 8086/80186/NEC family
of CPU's running MS-DOS 2.x or above.  The package sizes about 183K.

The author, Isaacson, writes resounding puffery in his documentation about
A86 being the finest assembler programming environment in the world.
"It would be cruel of me to withhold this from the world any longer." (etc.)
But a glance at the specs reveals shortcomings.

The assembler produces only .COM output.  There is no provision for .EXE
files or any hope of more than 64K of text.  The assembler ignores almost
all information about segments (incl. ASSUMEs), since everything is funneled
into CODE and DATA, and therefore it supplies no implicit segment overrides.
This requirement of explicit segment overrides might be construed as a feature,
if one never had to port foreign code written for any other assembler.
As it is, this could be a major compatibility problem.

Separately compilable .OBJ files cannot be produced (feature promised R.S.N.),
but a86 is fast enough to justify tossing LINK.EXE and keeping libraries
in .ASM source *IF* you process your entire program with A86.
In real life, though, assemblers are used for small low-level modules linked
in with .OBJs from a high level language, or they are the final pass of the
HLL compiler. Unfortunately, commercial compilers don't generally output
A86-compatible assembler code.  If you had source for the compiler itself
(e.g. small-C) you might patch it up to output A86 code.

The speed of A86 comes from its one-pass design.  Restrictions on certain
kinds of forward references must be made, but they may be acceptable.

The macro facility of A86 is featureful and radically different from
that in MASM.  I may decide it is better than MASM in some respects,
but it will never grok MASM macro libraries.

The symbolic debugger D86 has a user-interface which seems like a distinct
improvement on DEBUG, but I'll withhold further comment till I know more.

I can think of one use for this package, so I may end up registering the
thing.  And since A86/D86 is shareware one can always hope that the
author will develop it into a more generally useful system.


Vincent Broman,  code 632, Naval Ocean Systems Center, San Diego, CA 92152, USA
Phone: +1 619 225 2365          {seismo,caip,ihnp4,ucbvax}!\\\\\\\
Arpa: broman@bugs.nosc.mil   Uucp: {floyd,moss,bang,hp-sdd,sdcsvax}!nosc!broman