Pavel Machek wrote:
>
> Hi!
>
> > On my build machine (which, admittedly, is one of my K7s, so the
> > figures are a bit skewed) top[+], when refreshing every second,
> > takes 0.99% of the processor. And on the Celeron/338 I'm using as a
> > workstation, it takes a princely .098% to do the same.
>
> Probably not true, kernel cpu usage metters like to lie for exactly
> this kind of applications. Hold spacebar. If you are right, it should
> eat 30% of cpu, then.
33%
> It does not.
It's close enough for sampling error.
The real top seems to do better for figuring actual cpu
percentage that the procps one does (at least it does for me --
if I lean on the spacebar for long enough the wcpu rolls up to
~30% [at which point I stopped waiting; it was taking 30 seconds
to increase .01% of processor and I wanted to finish this
message].) Top with s=0 does end up taking all of the
processor, and much of that in the kernel, but that's several
hundred refreshes a second, which doesn't seem to be particularly
useful for anything other than eating up those idle percentages.
____
david parsons \bi/ top s=0 ends up slowing down mastodon rebuilds by
\/ about 50%, too.
-
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/
This archive was generated by hypermail 2b29 : Sat Jan 15 2000 - 21:00:15 EST