Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!munnari.oz.au!bruce!trlluna!rhea.trl.oz.au!aduncan
From: aduncan@rhea.trl.oz.au (Allan Duncan)
Newsgroups: comp.sys.amiga.tech
Subject: Re: First bug in Manx 5.0?
Message-ID: <1025@trlluna.trl.oz>
Date: 13 Feb 90 03:33:38 GMT
Sender: root@trlluna.trl.oz
Lines: 23

> 
>      The 'unable to load bug:  Error 2220548' message is, I believe, an error
> in the shell, not in bug.  What happens is that (if memory serves) if a
> programs returns a negative error code, the CLI says it was unable to load the
> program (never mind that the program was already loaded and executed).  So, all
> that is happening is that readchar() (or whatever routine it was you were
> calling) is leaving a negative number in D0, which is getting returned from
> main(), and there you have it.
>

I presume that readchar() is like the rest of them and returns -1 on
EOF.  However, this value should not be passed on to the return from
main - falling off the end is equivalent to exit(0).
I would look at the source for _main() and crt0.a
 
Allan Duncan	ACSnet	aduncan@rhea.trl.oz
		ARPA	aduncan%rhea.trl.oz.au@uunet.uu.net
		UUCP	{uunet,hplabs,ukc}!munnari!rhea.trl.oz.au!aduncan
Telecom Research Labs, PO Box 249, Clayton, Victoria, 3168, Australia.
Allan Duncan	ACSnet	aduncan@rhea.trl.oz
		ARPA	aduncan%rhea.trl.oz.au@uunet.uu.net
		UUCP	{uunet,hplabs,ukc}!munnari!rhea.trl.oz.au!aduncan
Telecom Research Labs, PO Box 249, Clayton, Victoria, 3168, Australia.