Re: /proc/stat on amd64 explanation?
On Tue, Nov 22, 2005 at 11:38:29AM +0100, Michal Hajek wrote:
> Hello people :)
>
> in /proc/stat on my Amd64 (AMD Athlon(tm) 64 Processor 3000+) I have:
>
> mymashine:~$ cat /proc/stat
> cpu 936448 76085819 423225 42441 2878 2250 345302 0
> cpu0 936448 76085819 423225 42441 2878 2250 345302 0
> intr 813723457 778502009 2 0 0 0 1 0 1 0 0 1 1 3504227 0 30670 0
> 23184496 0 3 6639698 186484 386065 1289799 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0
> ctxt 303922756
> btime 1131876576
> processes 2132201
> procs_running 4
> procs_blocked 0
>
> While trying to understand these numbers, I read in the man proc:
>
> /proc/stat
> kernel/system statistics. Varies with architecture. Common
> entries include:
>
> cpu 3357 0 4313 1362393
>
> The number of jiffies (1/100ths of a
> second) that the system spent in user
> mode, user mode with low priority (nice),
> system mode, and the idle task,
> respectively. The last value should be 100
> times the second entry in the uptime
> pseudo-file.
>
> .....
> <cut here>
>
> I am mainly interested in the "cpu ..." line, since it (AFAIK) should
> contain some valuable information about how my procesor is being used :)
> Though I am curious about the other lines too :)
>
> So what is than the actual meaning of the numbers at my system?
> Why there are 8 numbers instead of only 4 which are explained in man?
> Where can I get information about how this "varies with architecture" ?
>
> Thanks in advance and appoligies if this is a too stupid question
> and I am overlooking something quite obvious...
I like 'procinfo' since it parses and displays those things so nicely.
procinfo -a is especially full of information.
Len Sorensen
Reply to: