home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!wupost!emory!ogicse!das-news.harvard.edu!spdcc!merk!winston
- From: winston@merk.com (Winston Smith)
- Newsgroups: comp.sys.atari.8bit
- Subject: Re: Kermit-65 source code
- Keywords: makebin.c, KERMIT-65, assembler, 6502
- Message-ID: <C0DJxn.7MI@merk.com>
- Date: 5 Jan 93 09:34:34 GMT
- Article-I.D.: merk.C0DJxn.7MI
- References: <BzyKMB.CMJ@merk.com> <72769@cup.portal.com> <C07sBJ.Juv@merk.com>
- Organization: Technology Partners, Inc.
- Lines: 51
-
-
- > "Just when you thought it was safe to go back in the water...."
- >
- > ...Unfortunately, the program that accompanies the
- > assembler, that is supposed to turn the text characters into binary
- > code, is written in ANSI C, and the file calls make the K&R C compiler I
- > am using do strange things. <* ARG! *>
-
-
- "The saga of 'makebinary.c'...."
-
- Well, I had been trying to compile the program "makebin.c" so that the
- executible could turn the hexadecimal text from the 6502 cross-assembler
- "as6502" into actual hexadecimal code (also known as binary executible).
- At first I thought that the problem was that the program was written in
- ANSI C, so I edited all of the file calls and subroutine header statements
- to be K&R compliant. Nothing happened. After toying with it for a while,
- convinced that I didn't understand how ANSI C properly worked, I finally
- fell back on the classic desperation move of putting a print statement
- (well, in this case a PRINTF statement) in place where anything significant
- happens: before subroutine calls, at the top of procedures, around
- calculations -- before and after, etc. I then compiled again and watched
- the execution trace. It was at this point that I suddenly came to the
- realization that the reason the program didn't --DO-- anything was not
- because the file calls were wrong, but because --THAT WAS THE WAY THE
- PROGRAM WAS WRITTEN-- !?!! The program read in the input file, but
- generated no output because --NO OUTPUT PROCEDURES ARE CALLED-- !
- Apparently, the programmer abandoned work on the program when only half-way
- through writing it (but posted it to an FTP archive? I had always assumed
- that only --WORKING-- programs existed in FTP directories, but perhaps no one
- ever bothered to complain... since it was supposed to be creating
- executibles for the ATARI 6502 8-BIT).
-
- This leads me to the situation I am in now. I can generate a
- hexadecimal text assembly listing, but what is the best or most accepted
- method for generating an executible binary image from an "as6502" listing?
- Apparently AMAC, and ASMED do not allow for "inline hex code". I don't
- have MAC-65 or "ACTION!". Is there some type of program that can add a
- checksum to the end of a hexadecimal text line so that it can be read by a
- decoder? Will FORTH accept straight hexadecimal text, or does it need to
- be broken up into "screens" of data first? Perhaps I can modify a "machine
- language entry" program from one of the old ATARI 8-BIT magazines (at least
- that is my first thought...)? Is there a better program called "as6502", one
- with a "-v" flag? Perhaps I am using the wrong "as6502"? I don't want to
- reinvent the wheel if a simple solution exists already. I suppose if worse
- comes to worse I can always enter the entire source code suite by hand, but
- if there is a short-cut... I would rather take it. I guess that resorting
- to some kind of hex code loader program using BASIC pokes wouldn't be
- --TOO-- awful, but why was the work on "makebin.c" abandoned? Obviously
- a simpler solution must exist already! Hmm....
-
-