Message-Id: <199912170730.JAA26530@www.Foo.COM>
From: "S. M. Halloran" <mitch AT duzen DOT com DOT tr>
Organization: User RFC 822- and 1123-compliant
To: "Alex" <alexxwaag AT prodigy DOT net>
Date: Fri, 17 Dec 1999 09:38:00 +0200
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: Memory problem, (I think)
CC: djgpp AT delorie DOT com
In-reply-to: <83ci0u$84lc$1@newssvr03-int.news.prodigy.com>
X-mailer: Pegasus Mail for Win32 (v3.12b)
Reply-To: djgpp AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On 16 Dec 99, Alex was found to have commented thusly:

> Ok here's my problem. This works:
> int main()
> {
> char thing[10000][400]
> }

It doesn't compile without the semicolon after the local variable 
definition.  It compiles after enter it however.  :)

It didn't run for me either.  My test returned '1' and indicated a stack 
fault.  I didn't symify the stack trace, but I could guess that it didn't 
even get out of startup code (i.e., was not able even to call main()).

> But this doesn't:
> #include <conio.h>
> int main()
> {
> char thing[10000][400]
> getch();
> }

After adding the missing semicolon to the variable definition,  I get a 
different problem this time.  Returns '255' and the fault is SIGSEGV 
(that's what you're seeing, right?).  I agree this is a problem.

> Can anyone tell me why??? This problem is really getting to me, any help
> will be a godsend!!!

You probably like to code on UNIX workstations or machines with no 
imaginable resource limitations, don't you?

Anyway, these DOS/Windows/PC compilers need hints that you like to rely on 
huge stacks rather than taking what you need from the heap.  This 
newsgroup has a FAQ, and I believe you can find your answer in the number 
15.9 of that document.  After you go over the details, then a quick read 
of the rest of the FAQ might help you to recall other subjects commonly 
touched upon.

Btw, I added a "return (0);" statement to main so I could see that main() 
returning normally (otherwise it returns anything in the test), or if you 
use getch() whatever it returns ('27' if you hit ESC key).  I then cut out 
the 'unsigned _stklen = 1048576;' definition in the FAQ, inserted '* 8' 
(multipy by 8) between the constant and semicolon, to account for your 
particular use of stack space, ran it, and it returned zero for me, 
without generating a fault or exception.  Based on that, I'd say the code 
now does what I expect....whether that is the same as "the code works,"  
well, on a Friday, I can never be that positive.  :-)

> Thanks in advance,

> alexwaag AT prodigy DOT net


Mitch Halloran
Research (Bio)chemist
Duzen Laboratories Group
Ankara       TURKEY