Xref:	news2.mv.net comp.os.msdos.djgpp:2330
From:	malcolm AT manawatu DOT gen DOT nz (Malcolm Taylor)
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: Bug hides with GDB.
Date:	Sun, 31 Mar 1996 22:18:12 GMT
Organization:	Grafik Software
Lines:	21
Message-ID:	<4jn09q$hqk@news.manawatu.gen.nz>
References:	<4iqeht$c1t AT news DOT manawatu DOT gen DOT nz> <4is391$obf AT mark DOT ucdavis DOT edu> <31532ad9 DOT sandmann AT clio DOT rice DOT edu>
Reply-To:	malcolm AT manawatu DOT gen DOT nz
NNTP-Posting-Host:	malcolm.manawatu.gen.nz
Mime-Version:	1.0
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Charles Sandmann <sandmann AT clio DOT rice DOT edu> wrote:

>> Usually the physical addresses change, which means the content of 
>> uninitialized variables change.  

>Exactly.  Try building the app with the crt0.h flag for memory zeroing 
>enabled - and see if the behavior becomes consistent.  The deadbeef flag
>is particularly effective in flushing out bugs.

I have managed to fix the problem, there was a fault in the text file
being read in. I will be putting checks in to deal with this (and give
an appropriate error message). However the interesting thing is I
still don't know why it worked under the debugger. It seems that it
should not have worked in either case. I will be looking into this
further (in case it illuminates a problem with fgetpos or ftell
(although I don't expect that) and if I find anything out of the
ordinary I'll post something.

Thanks,
Malcolm

- Raw text -