www.delorie.com/archives/browse.cgi   search  
Mail Archives: djgpp/1995/03/01/05:04:19

Date: Wed, 1 Mar 1995 03:54:04 -0500
From: davis AT amy DOT tch DOT harvard DOT edu ("John E. Davis")
To: djgpp AT sun DOT soe DOT clarkson DOT edu, mhr AT sparc DOT sandiegoca DOT attgis DOT com
Subject: Re: Odd behavior in djgpp-compiled program

>game I wrote.  The source is around 7k lines and runs fine under Unix and,
>more or less now under DOS as well.  (This is something neither Turbo C v2.0
>nor Borland C++ v2.0 could do for me - too many really bizarre run time
>crashes.)

It has been my experience that virtual memory systems are very forgiving and
although your program may appear to work well under such systems (DJGPP and
Unix), the fact that it crashes alot under BCC indicates that you have big
problems somewhere.  That is, bugs will manifest themselves quicker under
BCC than they will under DJGPP.

Since you have BCC, I suggest the following:

    1.  Compile under BOTH bcc and gcc with all warnings turned on.  gcc
	misses some things that bcc catches and visa-versa.  Use the large
	memory module with bcc.  Simply doing this may reveal the problem.

    2.  After cleaning everything up so that it compiles clean under both
	compilers, step through it using the turbo debugger.  Chances are
	you will catch the bug quicker this way than debugging the DJGPP
	executable because as I stated earlier, the bugs will manifest
	themselves more often in the 16 bit version.

I know that the stability of jed is due to getting it to run under BCC.
Everytime I add a new code to it, or make changes, I test it first under
Unix to make sure that the ``obvious'' bugs are fixed (e.g., NULL pointer
problems) and then I run it under BCC which flushes the remaining bugs.

--John

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019