Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!ames!cit-vax!tybalt.caltech.edu!walton
From: walton@tybalt.caltech.edu (Steve Walton)
Newsgroups: comp.sys.amiga
Subject: Re: replacement for frags
Message-ID: <2362@cit-vax.Caltech.Edu>
Date: Thu, 16-Apr-87 00:59:02 EST
Article-I.D.: cit-vax.2362
Posted: Thu Apr 16 00:59:02 1987
Date-Received: Sun, 19-Apr-87 05:14:38 EST
References: <3148@jade.BERKELEY.EDU>
Sender: news@cit-vax.Caltech.Edu
Reply-To: walton@tybalt.caltech.edu.UUCP (Steve Walton)
Organization: Calfornia Institute of Technology
Lines: 22

In article <3148@jade.BERKELEY.EDU> mwm@violet.berkeley.edu(Mike (My watch
has windows) Meyer) writes (in the comments to his replacement frags program):
>/*
> * To cut down on memory useage, we provide a stub for a routine pulled
> * in by the default startup code in _main. This code cleans up
> * dynamically allocated memory, which we don't need. So we flush the code
> * here. By making this a smallcode, smalldata program, turning off stack 
> * checking (nothing recursive, and only three routines, so who needs it?),
> * and adding this, I've reduced the size of frags to 1644 bytes. Since the
> * original frags is 1964 bytes long, I'm happy. Just out of curiosity, I'd
> * like to know how large a binary Manx 3.4 produces.
> */

Well, now, I made a couple of small changes in Mike's code.  I removed the
prototyping for Long_To_ASCII and Print_Count (since Manx doesn't support
it), declared long Output() in Print_Count(), and cast OUTPUT_SIZE to (long)
in Print_Count's call to Write().  Resulting executable was 796 bytes long,
slightly less than half the size of the Lattice one.  This was small code,
small data, 16 bit ints.  It worked, too, by the way :->

						Steve Walton
					guest as walton@tybalt.caltech.edu