Re: [cc65] Latest Snapshot - Apple II

From: Bill Buckels <>
Date: 2014-05-16 19:55:16
Luke Coletti wrote:
>Can you send me the links to the sites you mention below?

Hi Luke,

The AppleX distribution contains some of the stuff:

I have the source from many of their compilers here and there and many other 
unpublished works or work that I have posted on Usenet in csa2 over the last 
year or so.

>Interested in your Aztec compiler work, although I'm not clear on what 
>exactly you're trying to do. I'd be interested in hearing more.

I am working-out the porting details to take everything in the Aztec C 
compilers that is not in cc65 and make it fit with Oliver's work.

I am really big on examples and levering platform specific functionality, 
and providing utilities for porting of graphics and other fun stuff.

So clearly this is outside the scope of a C compiler.

I am also big on blog-style tutorials... if a compiler has a poke macro it 
better be poking at something that's fun. Aztec C is broken and old. It is 
no fun anymore. A PSIX compliant modern compiler with perks would be a 
better home for what I learned to do with Aztec C.

But I am still looking at what cc65 has done and what I can make it do. It 
is built around a core I am adjusting to and I am having a hard time 
understanding how to map memory using linker config files to open-up the 
Apple II's memory map for SYS programs that do not hide themselves in every 
available hole right at the moment.

Since I both learn and tech by example, I need to write full applications, 
and don't much care for code snippets and the usual terse examples provided 
by typical compiler documentation. I hope this does not make me any less of 
a man:)

I like to knock together programs in a couple of seconds and I doubt if I am 
alone. The cc65 compiler can do that... better than Aztec C ever could. But 
not for me in its current state so I am coding every chance I get to change 
that. It generally takes me from 3 to 6 months to adapt a compiler for my 
nefarious purposes for a particular OS or platform, and perhaps a year or 
two to get some cools utilities and stuff out there.

I don't write OS's but I sure as heck like to write apps that use them. At 
the end of the day it all amounts to the same. How I get there has always 
been pretty simple. I take things apart to see how they work, then I see 
what's missing, and add it. I generally don't put any restrictions on the 
use of my work and don't ask for attribution.  This linker config info is 
important to me at the moment for a mass port. I can't get hung-up on how 
the compiler sees my world.

Oliver has a well-thought out structure for making system calls to ProDOS on 
the Apple IIe MLI (Machine Language Interface) which lends itself strongly 
to casting a structure to make this more C-like for me. Hopefully for others 
if any exist who want the same thing.

Very small part of it. I haven't even looked at Overlays yet to see if I 
like them. Or how I will change them or make them fit into my view if I 

Dicussion of what I am doing doesn't seem appropriate for Oliver's list 
since he has already tried to bend it to his way of thinking. So I will 
likely not worry about that until the time comes when my cc65 work gets some 
fun results.

Some cc65 stuff is on this site too:

And some graphics stuff:

For my Commodore stuff:

For my CP/M stuff:

I won't mention my MS-DOS stuff.

cc65 is part of my lifeline to the future to preserve the fun of the old 
days on newer computers. Like I said, the Aztec C stuff is broken now as we 
move forward. The only truly available port is cc65. When I backed-off 5 
years ago, I wasn't ready. For the last year I prepared for the goal. But 
the perspective is hardly the same as cc65's installed base and developers 
although many of our goals are common in the end. I couldn't care less what 
is put in a repository once I learn a tool. I am perfectly capable of 
rolling my own without much destruction and order, so trust me to not 
re-invent the wheel.

The cc65 calling convention is beautifully designed. After Aztec C I simply 
hate the stack overhead so fastcall is a welcome relief. The information 
hiding is a little hard for me to agree with. Marrying the two is easier 
than in Aztec C where we didn't have a choice.

Oliver's decompsition of the ProDOS System calls is true art but I think 
some are missing. cc65's lack of support for DOS 3.3 File I/O needs to be 
fixed. I am sure I can do that and add floating point support as well.

I also have some thoughts on how the various shell programs for the Apple II 
can use cc65 for a better and quicker alternative to the old Aztec C and 
other shell's:

Reducing my expectations to low common denominators doesn't do it for me. 
But cross-development on modern computers in C is the way to go forward from 
the old junk that keeps breaking.


To unsubscribe from the list send mail to with
the string "unsubscribe cc65" in the body(!) of the mail.
Received on Fri May 16 19:55:34 2014

This archive was generated by hypermail 2.1.8 : 2014-05-16 19:55:36 CEST