[Coco] Alternate O-9/itrOS-9 C Compiler

Thu May 16 10:21:05 EDT 2013

Gene,
I thought I had c.prep 19b in the distribution, but was mistaken. I am correcting that right now. Maybe that's why I've been having some memory troubles... hmmmm...
I was playing around a while back and was trying to get the C compiler to work with Level 1 in XRoar and probably copied the MW version over the 19b version by mistake.
And now I know who wrote the thing :-)
If it's the same Mathew Thomas, he also wrote Bells & Whistles 2 for Rainbow back when he was just 16 years old.
So it's Mathew Thomas, Gene Heskett & Willard Goosey.... noted and catalogged :-)

Bill Pierce
My Music from the Tandy/Radio Shack Color Computer 2 & 3
https://sites.google.com/site/dabarnstudio/
Co-Webmaster of The TRS-80 Color Computer Archive
http://www.colorcomputerarchive.com/
Co-Contributor, Co-Editor for CocoPedia
http://www.cocopedia.com/wiki/index.php/Main_Page
E-Mail: ooogalapasooo at aol.com

-----Original Message-----
From: Gene Heskett <gheskett at wdtv.com>
To: coco <coco at maltedmedia.com>
Sent: Thu, May 16, 2013 9:22 am
Subject: Re: [Coco] Alternate O-9/itrOS-9 C Compiler

On Thursday 16 May 2013 08:13:41 Bill Pierce did opine:

> Hi guys,
> Ok, with several people asking about the OS9 C compiler, I decided to
> add a page to my site explaining my C compiler system. It's a bit
> different, but not much. Not only is it the C compiler I used for my
> DW4Man and Sound Chaser programs, but it is also the compiler Mike
> Knudsen used to compile Ultimuse 3. In fact, it was Mike that gave me
> the compiler. I've tried to describe it's usage as best I could as I am
> not a seasoned C programmer and my terminology may be lacking. But any
> one having run either of my programs has seen the extent of what it's
> capable of. Both DW4Man and Sound Chaser are large, multi file programs
> that use Mike Sweet's "CGFX.L" (cgfx7), Carl Krieder's "CLIB.L", Time
> Koonze's "Make", Mike Knudsen's RcM2 & CnoY, and Zack Session's
> "window" library. On the provided disks and in the tutorial, I included
> the sources to Mike's "UBox3" as a sample for running the compiler.
> 
> Please send any questions to me privately as I don't want to clog the
> list with a discussion of this compiler.
> 
> Here it is:
> https://sites.google.com/site/dabarnstudio/mjk-c-compiler
> 
> Enjoy

First off Bill, thanks for the flowers about MyRam scattered through your 
tut.  However, its no faster than a good HD setup, 11 second megareads were 
obtainable from it on a 6309 machine, 16 or thereabouts on a 6809 machine, 
and from a disto controlled HD, the speed limit for either is the block 
move of 256 bytes of data function.

But I don't see any mention of ditching the Microware edition of c.prep and 
I suspect at least some of Mike's work was spurred by the limitations in 
the Microware version of c.prep.

One of our then rising young stars undertook to write a new one, I think as 
part of the CS course he was taking, and released it up to about version 
15.  It was by then a huge improvement, controlling the scope of arguments 
well, but like the Microware version, silently ran out of memory & started 
issuing garbage, which then predictably crashed because vars got re-used, 
anytime the merged sources being fed through it exceeded around 10k total 
for the Microware version, and about 15k for Matthews version 15.

This (IIRC) was Matthew Tompson(sp), same fellow that did scsisys I 
believe.  I suspect that also was a CS course project.

Anyway, Matthews version still would not make a working rzsz of any version 
as the combined srcs to generate either total, in the last version I built, 
around 34k, lots bigger than Matthews version could handle.  Since he was 
nice enough to publish the srcs, and I had his perms to have at it, I took 
it upon myself to have it ask for quite a bit more memory to work with, and 
by my release of version 19, had it building an rzsz that was stable.  And 
FWIW, about 200 cps faster than Chuck Forsbergs code as I substituted a 
much faster method of doing a running crc on the data.  rzsz is slow, and 
in the real world is busy staying ahead of a 2400 baud modem.  9600 is 
totally out of the question if hardware flow controls don't work.

Now Willard Goosey has further improved c.prep, but I've not had a good 
excuse yet to "throw rocks" :) at his version so I'll defer to Willard & 
let him explain what it now does better than my Version 19. :)  Your turn 
Willard...

One thing I should note about Mike K., for the edification of newer members 
here, he is retired Bell Labs, he was I believe there when Brian Kernigan 
and Dennis Ritchie wrote the original C language specs, so he knows his C 
to a depth that few CS profs have ever penetrated to.

He stopped by here a bit over a decade back, driving an elderly green 
Econoline van with at least a ton of pipe organ pipes & air valves etc 
stuffed in the back that he was headed north with, to add them to another 
big pipe organ in a church someplace I've since forgotten.  We sat on my 
front deck, doing serious damage to a 6 pack of 807's and telling war 
stories from our working days for about 3 hours, a man who was absolutely 
delighted to have all the time he needed to do some of the things /he/ 
wanted to do.  He is the sort you wouldn't mind having for a next door 
neighbor.

Cheers, Gene
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."

-Ed Howdershelt (Author)
My web page: <http://coyoteden.dyndns-free.com:85/gene> is up!
My views 
<http://www.armchairpatriot.com/What%20Has%20America%20Become.shtml>
DYSLEXICS OF THE WORLD, UNTIE!
A pen in the hand of this president is far more
dangerous than 200 million guns in the hands of
         law-abiding citizens.

--
Coco mailing list
Coco at maltedmedia.com
http://five.pairlist.net/mailman/listinfo/coco