Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!rutgers!cbmvax!jesup
From: jesup@cbmvax.commodore.com (Randell Jesup)
Newsgroups: comp.sys.amiga.programmer
Subject: Re: 2.0 Compatibility Problem
Message-ID: <21714@cbmvax.commodore.com>
Date: 19 May 91 04:50:32 GMT
References: <1991May18.105312.1@happy.colorado.edu>
Reply-To: jesup@cbmvax.commodore.com (Randell Jesup)
Organization: Commodore, West Chester, PA
Lines: 40

In article <1991May18.105312.1@happy.colorado.edu> kskelm@happy.colorado.edu writes:
>
>>>    The software works fine on *EVERY* platform but A3000 under 2.0; it works
>>>on a 500, a 1000, a 2000, a 2500 (all in both 1.3 and 2.0), and it works
>>>on a 3000 in 1.3, but *NOT* in 2.0.

>   Well, on my publishers' machines, it apparently resets the system and
>reboots immediately.
>
>   It also exhibits strange behavior-- suddenly the software's right hand
>doesn't know what the left is doing-- even though I'm *SURE* that a graphic
>buffer has been properly loaded, the software suddenly doesnt think it is (I'll
>have to work that one out one my own).

	The most common A3000-related problem is over-running a memory
allocation by one byte - on an machine without >24bit addressed memory, it 
usually kills the first byte of a pointer.  If it's a string, the byte used
to overwrite with is a 0, which is what was already there.  2.0 can modify
this since the order and type of memory allocations would change.

	I STRONGLY advise running it under both enforcer and mungwall.  They're
very good at catching bugs, particularily memory-allocation types.  Beyond 
that, all I can advise is to try to isolate where the problem starts (also,
since it's modula, turn on any runtime checking you can).

>   ...Oh, yah... and I find that my executable generator does consistently
>"break" under 2.0 (on any machine)... seems you folks have changed the DOS code
>loader a bit???  Made it stricter? 

	Perhaps.  Care to specify what you do that's funny?  Or send me
a copy of an dumpobj of the file (or a (small) file uuencoded).  If you
followed the existing format you should have had no problems.

-- 
Randell Jesup, Keeper of AmigaDos, Commodore Engineering.
{uunet|rutgers}!cbmvax!jesup, jesup@cbmvax.commodore.com  BIX: rjesup  
Disclaimer: Nothing I say is anything other than my personal opinion.
Thus spake the Master Ninjei: "To program a million-line operating system
is easy, to change a man's temperament is more difficult."
(From "The Zen of Programming")  ;-)