Linux-2.4.22 memory overwriting

Linux-2.4.22 memory overwriting

[Julien Rebetez]

Hi !
I've writen the following program :


#include <stdio.h>

int main ()
{
        int p[4];
        p[0]=1;
        p[1]=2;
        p[2]=3;
        p[3]=4;
        p[4]=5;

        printf ("%i, %i, %i, %i, %i\n", p[0], p[1], p[2], p[3],
p[4]);
        return 0;
}

I compile it with :

 gcc -o test test.c -Wall

and when i launch it, the output is :

 julien:$> ./test
1, 2, 3, 4, 5

Should I not get a SIGSEV from the system ? Isn't it dangerous to allow 
the user to put 5 elements in a 4 elements tab?

(tested on Linux 2.4.22 on a i686)

Thanks

Re: Linux-2.4.22 memory overwriting

[Hugo Mills]

[-- Attachment #1: Type: text/plain, Size: 1024 bytes --]

On Sun, Feb 01, 2004 at 01:14:37PM +0100, Julien Rebetez wrote:
> Hi !
> I've writen the following program :
[snip]
> Should I not get a SIGSEV from the system ? Isn't it dangerous to allow 
> the user to put 5 elements in a 4 elements tab?

   This is nothing to do the the kernel. It's to do with C.

   Yes, it is dangerous to allow this behaviour. However, C doesn't
perform bounds checking on arrays -- this is left to the programmer to
ensure that an array is never accessed outside its bounds. The effects
of accessing (reading or writing) an array outside its bounds are
undefined. In this case, it's worked. In other situations, with
different arrays, it may not work.

   Nothing to see here. Move along.

   Hugo.

-- 
=== Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
  PGP key: 1C335860 from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
   --- Well, you don't get to be a kernel hacker simply by looking ---   
                    good in Speedos. -- Rusty Russell                    

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Linux-2.4.22 memory overwriting

[Roberto Sanchez]

[-- Attachment #1: Type: text/plain, Size: 877 bytes --]

Julien Rebetez wrote:
> Hi !
> I've writen the following program :
> 
> 
> #include <stdio.h>
> 
> int main ()
> {
>        int p[4];
>        p[0]=1;
>        p[1]=2;
>        p[2]=3;
>        p[3]=4;
>        p[4]=5;
> 
>        printf ("%i, %i, %i, %i, %i\n", p[0], p[1], p[2], p[3],
> p[4]);
>        return 0;
> }
> 
> I compile it with :
> 
> gcc -o test test.c -Wall
> 
> and when i launch it, the output is :
> 
> julien:$> ./test
> 1, 2, 3, 4, 5
> 
> Should I not get a SIGSEV from the system ? Isn't it dangerous to allow 
> the user to put 5 elements in a 4 elements tab?
> 
> (tested on Linux 2.4.22 on a i686)
> 
> Thanks

Remember that C was written by experts for use by experts.  C makes
the unfortunate assumption that you know *exactly* what you are doing.
This flexibility is great for tho advanced system programmer, but
dangerous for the unwary.

-Roberto

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 256 bytes --]