Tom, Andy and Imre: Thank you very much for your thoughtful
responses. I'd just like to add little reaction and some
more thoughts.

First, one thought about macros. While it wouldn't
immediately solve the portablity problem, it occured to me
that it should be somewhat straightforward to use GNU M4 --
http://www.seindal.dk/rene/gnu/ -- as a macro preprocessor
for CVASM16. I wouldn't expect M4 to immediately handle
the macro syntax of other assemblers, but then M4 is
a powerful enough tool that it might just be possible
to use M4 to build a tool that would. Still, it would
right out of the box give you the ability to code up
macros for CVASM16. M4 is available for the DOS platform
via DJCPP or for Win32 in Cygnus' Cygwin project --
http://sourceware.cygnus.com/cygwin/ -- which no
Windows-based software development system should
be without, IMHO. The hardest part of using M4 would
probably be integrating it into the TDE environment,
which may or may not be possible... anyone have
a guess?

Second, Andy, while I've been using MPASM, that
doesn't mean I like it. It's kind of a quaint language,
and of course it takes you as close to the hardware
as you can get, but damn if it isn't a funky thing
to get your head around. And I've probably used two
or three dozen programming languages over the years,
so I don't think that it's just me. It reminds me, in
sort of a kindred spirit kind of way, of Intercal --
http://www.tuxedo.org/~esr/intercal/.

So, once I read all y'all's responses, I was kind of
intrigued. I've skimmed through the instruction set and
read a couple of examples, and it does seem a bunch more
livable. The big question mark of course being the ability
of this thing to process this code I need to work with. As
it turns out, the emWare emMicro source code is available
both in "generic 8051" and in MPASM.  Unfortunately, the
MPASM code is just dripping with macros, so that got kind
of nowhere fast.

The 8051 code, however, seemed a little more promising.
Nonetheless, that too turns out to be somewhat
probelmatic.  The 8051 code was written to the Metalink
ASM51 assembler, and I quickly found bunchs of stuff
that had to be fixed.  The emMicro code used SET and
BIT instructions, for example, and a bunch of assembler
directives that didn't work.  I commented out a bunch of
stuff that didn't seem important ($LIST directives and
such), and changed $INCLUDE directives to the proper
INCLUDE syntax. Once I got that stuff fixed, I started
running into macros :-(

I can't release the file, but emWare's "lite"/free
version includes the 8051 assembler source I'm talking
about... It's a 10MB download from the URL
http://www.emstore.com/cgifile/EMIT3.01SDKLite.exe
while the user's manual is at
http://www.emstore.com/cgifile/EMITLiteFastStart.pdf
There's probably enough stuff in that package to actually
do something with an 8051, but you'd have to be pretty good
at reading assembler source; the developer's manual is only
availble if you pay them money.  (emWare will probably be
mad at me for those links...  they want you to register, and
I know from experience that a salesperson *will call*. The
top level is http://www.emware.com FWIW.) Once you
unpack it, the files will be in
...\emWare\emMicro\Samples\8051\emSdk
and ...\emWare\emMicro\Samples\8051\emMicro.
The MPASM source hasn't been relased like this.

If there's any CVASM16 gurus out there so bored out of
their skulls that they'd like to take a look at that source
and guess whether it would ever work to run a PIC with it
via CVASM16, I'd be interested to know what you think.

Thanks again. I might just use CVASM16 for things besides
this, and I might just see what I can do with M4 and
CVASM16. But it seems as if trying to make it work on this
emWare stuff would be a huge time sink with little payoff.

--Bob

--
============================================================
Bob Drzyzgula                             It's not a problem
bob@drzyzgula.org                until something bad happens
============================================================