Date   : Sun, 14 Oct 1990 20:56:30 GMT
From   : usc!wuarchive!cs.utexas.edu!sun-barr!olivea!orc!inews!iwarp.intel.com!omepd!pzbaum!reed!tektronix!percy!nosun!techbook!fzsitvay@ucsd.edu (Frank Zsitvay)
Subject: Need info for an S-100 bus system

In article <1990Oct13.163615.6218@news.clarkson.edu> demarem@clutx.clarkson.edu
writes:
>From article <1990Oct12.235220.18522@techbook.com>, by fzsitvay@techbook.com
(Frank Zsitvay):
>> In article <1990Oct12.000849.12599@news.clarkson.edu> demarem@clutx.clarkson.edu
writes:
>>>I suspect that your choices for an OS are CP/M or CP/M :-).  You
>>>can probably order documentation and/or CP/M drivers for your
>>>devices, but you may have a bootstrap problem..you need the system
>>>running CP/M in order to modify CP/M for the system.  One nice thing
>>>about CP/M is that the BIOS sources (in assembler) come with it
>>>so you ought to do okay.  I would recommend aquiring the following
>>>items (to run on another system while getting your S-100 up):
>>>
>>>8080 cross-assembler
>>   
>>   An 8080 cross assembler on a cp/m machine??  go figure...
>> 
>
>I was thinking in terms of assembling some code on his "real" machine
>and keying it in through the monitor or frontpanel (ugh).  I have seen
>some nice 8080 assemblers running on PDP-11's (under Unix lev. 6).
>
  opps.   sorry...      
 
  actually, if you can find a cross assembler for a dos machine to write
z80 code, you'd be farther ahead of the game provided it worked much like
cp/m's ASM.COM or RMAC.

  but the easiest route (if you had to config cp/m from scratch) is to
find someone with a cp/m machine that has the same disk format, and
hack the bios.  the beauty of this is that you can write a boot disk
for the target machine on the source machine.  It is a bit of trial
and error, but if you're careful, you'll be succesfull.

>>>Small-C (8080 version, source code is available, I have seen it in
>>>         *very* old Doctor Dobbs Journals, and believe that some
>>>    user groups have it available in machine readable form).
>>    NO!
>> 
>>    Don't do this!  Bad idea.
>> 
>>    you don't want to write a bios in c.   you're only restricted to
>> a maximum address space of 64k, so your bios has to be as small and
>> as fast as possible.  this is very important if you plan to add
>> double density drives to the system, since that requires some form
>> of sector blocking/deblocking.  you do that in c and the bios will
>> be much larger than it needs to be.  the best tools to use when writing
>> or hacking a bios is either ASM (which comes wth cpm) or RMAC, DR's
>> relocating macro assembler.  the job is easier with RMAC because it 
>> will generate your DPHs and DPBs for you.  My cpm mentor tells me MASM
>> will not work for this.
>
>Quite clearly you are correct, it is inappropriate to code a BIOS
>in C.  But I had something a little different in mind..realating
>to AFTER he got the system up.
>
 ah, i see.   in addition to small c, there is microsoft's basic compiler,
and bds c. which i think is still undergoing development.

 nevertheless, you'll want to get 64k in the machine as soon as possible,
as cp/m takes 8k to begin with, and most of these compilers tack on their
own run time code

-- 
fzsitvay@techbook.COM - one of these days i'll get it right...

Version 2 of anything is usually the version that works.