Address space for user process

Address space for user process

[Christoph Pleger]

Hello,

I read that - at least in Kernel 2.4 - the amount of memory that can be
addressed by a user process is limited to 3 GB, no matter how much
virtual memory is present. Is it possible to raise that limit?

Kind regards
  Christoph

Re: Address space for user process

[Doug McNaught]

Christoph Pleger <Christoph.Pleger@uni-dortmund.de> writes:

> Hello,
>
> I read that - at least in Kernel 2.4 - the amount of memory that can be
> addressed by a user process is limited to 3 GB, no matter how much
> virtual memory is present. Is it possible to raise that limit?

That's limit of the ix86 architecture, not Linux.  64-bit systems
running Linux have no such limit.

That said, there are patches for 2.4 that create separate address
spaces for kernel and user (search for '4G/4G') at the cost of a
(sometimes significant) performance penalty.  4GB per-process is a
hard limit on 32-bit systems, so it's the most you'll get.

-Doug

Re: Address space for user process

[Richard B. Johnson]

On Wed, 7 Jul 2004, Christoph Pleger wrote:

> Hello,
>
> I read that - at least in Kernel 2.4 - the amount of memory that can be
> addressed by a user process is limited to 3 GB, no matter how much
> virtual memory is present. Is it possible to raise that limit?
>
> Kind regards
>   Christoph

Sure. Use two (or more) tasks. It is a per-process limit because
in Unix the kernel shares the processes address space. See
how your process is laid out by executing this:

#include <stdio.h>
extern char _start;
extern char _end;
char dat;
const char x[]="x";
int main()
{
    char foo;
    printf("start = %p\n", &_start);
    printf(" main = %p\n", main);
    printf(" data = %p\n", &dat);
    printf("const = %p\n", x);
    printf("stack = %p\n", &foo);
    printf("  end = %p\n", &_end);
    return 0;
}

Cheers,
Dick Johnson
Penguin : Linux version 2.4.26 on an i686 machine (5570.56 BogoMips).
            Note 96.31% of all statistics are fiction.