Mail Archives: djgpp/1999/06/09/18:01:41
>> used without calling CWSDPMI externally first, do work. My only remaining
>> idea for the cause of this problem is that malloc simply won't do anything
>> as "large" as 34K
>34K is nowhere even remotely 'large', by standards of the DJGPP
>compiler and library. 34M yes, that'd be a bit large, but I'd still
>expect it to work, without any problems.
I agree, that's why I said "large" in stead of large ;-)
>> If anyone has some real-life source that does this and does work,
>> preferably ANSI C, but C++ will do as well, I would be much obliged.
>You're thinking the wrong way round, I think. You won't probably gain
>anything from looking at other people's source, here: malloc() works,
>and has been working, for all of us, for years on end. So the problem
>is not that your code is missing some magic cookie you might have to
>grab from other programs.
>It'd be way more helpful if you could prune down your failing program
>to a small but _complete_ source code that still fails on your setup,
>and post that, along with every bit of system information you can get
>at (config.sys, autoexec.bat, 'set' output, and all that), so people
>could check what may be going wrong, in your code.
I'm gonna do something close to that: I'm gonna cut everything out and let
it in statement by statement until it fails. I.e., I'm going to isolate the
statement that makes it fail. As there are about 20 statements in the
entire program, that only leaves about a million possibilities. Hopefully,
I won't have to try them all. To say it mildly, I am reluctant to post the
code here: I wouldn't want to burden you. I was hoping someone would have
an idea to something I was missing, but as nobody seems to have any
brilliant notions at the moment, I guess I'll just have to move some
mountains here.
Greetz!
Dlanor
- Raw text -