Yesterday, James R. Van Zandt gleaned this insight:

>    vanzandt:~$ cat /proc/meminfo
>          total:    used:    free:  shared: buffers:  cached:
>    Mem:  64524288 41836544 22687744 25706496  9859072 15622144
>    Swap: 106885120        0 106885120
>    MemTotal:     63012 kB
>    MemFree:      22156 kB
>    MemShared:    25104 kB
>    Buffers:       9628 kB
>    Cached:       15256 kB
>    SwapTotal:   104380 kB
>    SwapFree:    104380 kB
>    vanzandt:~$ calc 22687744+41836544
>    64524288
>    vanzandt:~$ calc 25706496+9859072+15622144
>    51187712
>    vanzandt:~$ calc 25706496+9859072
>    35565568
>    
> I understand the first three entries, I think.  (I only just booted
> this machine, so it has not buffered much from the disk, and has not
> swapped anything out.)  I assume the 25M of "shared" includes all the
> shared libraries I'm using.  I also assume that "buffers" have copies
> of data recently read from and/or written to the disk.  However, 
> I don't know what "cached" refers to.  Apparently the last three
> categories are not distinct, because they add up to more than "used".
> Is anyone willing to fill me in?

I just had a discussion about that very topic with one of our kernel
weenies last week.  You're almost perfect, if I understood him
correctly... Shared refers to the number of bytes being used by more than
one process (not necessarily shared libraries, can also be text pages of a
program with multiple instances). Buffers refer to I/O buffers (any
buffered I/O), while cached refers specifically to disk buffer cache.  
The numbers do not add up to 100%, and will overlap.


-- 
---------------------------------------------------------------
Derek D. Martin              |  Unix/Linux Geek
[EMAIL PROTECTED] |  [EMAIL PROTECTED]
---------------------------------------------------------------


**********************************************************
To unsubscribe from this list, send mail to
[EMAIL PROTECTED] with the following text in the
*body* (*not* the subject line) of the letter:
unsubscribe gnhlug
**********************************************************

Reply via email to