Re: top displaying 9999% CPU usage

[Ilpo Järvinen]

On Wed, 3 Oct 2007, Frans Pop wrote:

> On Wednesday 03 October 2007, you wrote:
> > Try to capture the i/o log with the following command:
> > strace -o top.log top
> 
> Thanks for the suggestion.
> 
> > This will show for sure whether the kernel gives out incorrect data or
> > top misinterprets them.
> 
>  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
> 5269 fjp       20   0  442m  61m  33m S 9999  3.1   0:34.12 kontact
> 
> Here are the last two reads for PID 5269 from /proc before that:
> 
> 15:48:51 stat("/proc/5269", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0
> 15:48:51 open("/proc/5269/stat", O_RDONLY) = 10
> 15:48:51 read(10, "5269 (kontact) S 1 5100 5100 0 -1 4202560 58186 7891 290 1 2911 502 10 15 20 0 6 0 9225 464408576 15694 18446744073709551615 4194304 4365900 140733817144688 18446744073709551615 47998530367074 0 0 4098 83113 18446744073709551615 0 0 17 1 0 0 0\n", 1023) = 244
> 15:48:51 open("/proc/5269/statm", O_RDONLY) = 10
> 15:48:54 stat("/proc/5269", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0
> 15:48:54 open("/proc/5269/stat", O_RDONLY) = 10
> 15:48:54 read(10, "5269 (kontact) S 1 5100 5100 0 -1 4202560 58186 7891 290 1 2912 500 10 15 20 0 6 0 9225 464408576 15694 18446744073709551615 4194304 4365900 140733817144688 18446744073709551615 47998530367074 0 0 4098 83113 18446744073709551615 0 0 17 1 0 0 0\n", 1023) = 244
> 15:48:54 open("/proc/5269/statm", O_RDONLY) = 10
> 
> The only change is in 2 consecutive columns: "2911 502" -> "2912 500".
> Is processor usage calculated from those? Can someone explain how?

The latter seems to be utime ...decreasing. No wonder if arithmetics will 
give strange results (probably top is using unsigned delta?)...


-- 
 i.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/