Physical memory versus detected memory 2.4.7-10

Physical memory versus detected memory 2.4.7-10

Post by ivan » Sat, 26 Jan 2002 10:10:10



Dear list users.

My server detects less memory than it available.

        Available memory according to the BIOS 4049MB.

        System sees only 3.7GB ???
        Mem:  3799580K av, 1606816K used, 2192764K free, 468K shrd, 376972K buff
        Swap: 8192992K av, 0K used, 8192992K free 1037532K cached

OS RedHat 2.7 standard kernel.
        Linux atlas.es.usyd.edu.au 2.4.7-10smp #1 SMP Mon Dec 10 13:08:57 EST 2001 i686 unknown

Hardware PowerEdge 4400.
        ( supports 4GB )

Additional problem:  
        random crushes, trying to to find out why.

Questions :     1) Any comments on why top only shows 3.7 Gb are welcome.

Thank you.
================================================================================

Ivan Teliatnikov,
F05 David Edgeworth Building,
Department of Geology and Geophysics,
School of Geosciences,
University of Sydney, 2006
Australia


ph:  061-2-9351-2031 (w)
fax: 061-2-9351-0184 (w)

===============================================================================

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Physical memory versus detected memory 2.4.7-10

Post by Jeff Chu » Sat, 26 Jan 2002 10:40:15



> My server detects less memory than it available.
>    Available memory according to the BIOS 4049MB.

>    System sees only 3.7GB ???
>    Mem:  3799580K av, 1606816K used, 2192764K free, 468K shrd, 376972K buff
>    Swap: 8192992K av, 0K used, 8192992K free 1037532K cached

I think you need to recompile the kernel to support 64GB mem.

You current config seems to support up to 4GB only (but it is a bit
less than 4GB). I had a similar problem where I had 1GB, but saw only
9xxGB.

Jeff

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Physical memory versus detected memory 2.4.7-10

Post by Matt_Dom.. » Tue, 29 Jan 2002 13:00:14


Quote:> | My server detects less memory than it available.
> |
> |  Available memory according to the BIOS 4049MB.
> |
> |  System sees only 3.7GB ???
> |  Mem:  3799580K av, 1606816K used, 2192764K free, 468K shrd, 376972K
buff
> |  Swap: 8192992K av, 0K used, 8192992K free 1037532K cached

The difference is space assigned to PCI cards or reserved for hot-plug PCI
cards.  You should be able to run a CONFIG_HIGHMEM64G-enabled kernel (such
as the Red Hat 2.4.x-enterprise kernels instead of the -smp kernel) and be
able to use the remaining memory, at the performance cost of enabling PAE.

Thanks,
Matt

--
Matt Domsch
Sr. Software Engineer
Dell Linux Solutions www.dell.com/linux
#1 US Linux Server provider with 24.5% (IDC Dec 2001)
#2 Worldwide Linux Server provider with 18.5% (IDC Dec 2001)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

1. Kernel Memory Leak in 2.4.7-10

Hi all,

I have a linux box (running version 2.4.7-10). It has around 770M RAM,
1.5G in swap. I was just wondering does in all cases if a process gets
a segmentation fault, kernel would be able to reclaim memory used by
that process. Actually I did two small tests:
1) I run a program, which makes request for 100M and then actally uses
it and does not free the memory before existing. No matter how many
time I run this program the kernel was able to reclaim the memory.
---------------------------------------------------------
#include <stdio.h>
#include <stdlib.h>
#include <malloc.h>
#include <signal.h>

#define MEM (100*1024*1024)

int main ()
{
  char *buf=NULL;
  int i;

  buf = (char *)malloc (MEM);

  for (i=0; i<MEM; i++)
    buf[i] = 'a';

  return 0;
--------------------------------------------------------

2) Now I make sure that the above program gets a segmentation fault
after it gets a 100M and actually uses it. (Just undefined memory
reference after the for loop). There are a couple of things I am not
able to understand:

2a) I get a core file. The momemt I delete it, I see memory being
reclaimed by the kernel (using /proc/meminfo or top).
2b) If I run it many number of time (say 5 or 6) and start deleting
all the cores one by one, the kernel is not able to reclaim the memory
properly.
When I do 'cat /proc/meminfo', I see alot of memory in 'Buffers,
Inact_dirty'.

Why kernel is not reclaiming the memory?

Thanks,
Lee

2. I'm gettin' a Dell, Dude!

3. Strange 2.4.7-10 memory usage?

4. Linux as print-box very good, but ...

5. Understanding high memory versus low memory?

6. HP CD-Writer+ 7200

7. mmap, virtual memory, physical memory

8. Parallel ZIP Drive

9. Real Memmory /Virtual Memory /Physical Memory Accountancy.

10. Is "shared memory" a way to reduce physical memory usage?

11. Sun workshop memory monitor does not detect all memory leak

12. 270 versus 270 memory

13. Reference counts versus PG_locked and memory management....