2.1.131: Strange percent idle time measurement

Jamie Lokier (lkd@tantalophile.demon.co.uk)
Mon, 7 Dec 1998 03:00:56 +0000

I'm running "top d 0" right now, in an xterm.
Nothing much else is happening except one xclock ticks once per second.

The "CPU states" line is interesting:

"user" varies between ~22% and 101% (!) (based on just looking at it)
"system" varies between ~22% and 101%
"nice" is usually 0%, but occasionally 45% or so for one sample.
(that's ok, kswapd is a niced processed, though I don't know
why it would be doing anything at all).
"idle" switches between 0% and 10%, occasionally showing 0.4%, 9.9% etc.

The "idle" time settles into this state after a couple of minutes.
Initially, it is more variable but still in the approximate range 0%-10%.

Some questions arise:

1. 101% is obviously wrong and should never be reported.
Is this a "top" thing, or a number reported by the kernel?

2. The idle time shows some interesting jumps.
I'd expect 0% all the time. Is this top's calculations again?

-- Jamie

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