> >>>>> On Tue, 16 Apr 2002 16:27:12 +0000 (UTC), torvalds@transmeta.com (Linus Torvalds) said:
>
> Linus> And I've had some Intel people grumble about it, because it
> Linus> apparently means that the timer tick takes anything from 2%
> Linus> to an extreme of 10% (!!) of the CPU time under certain
> Linus> loads.
>
> I'm not sure I believe this. I have had occasional cases where I
> wondered whether the timer tick caused significant overhead, but it
> always turned out to be something else. In my measurements,
> *user-level* profiling has the 2-10% overhead you're mentioning, but
> that's with a signal delivered to user level on each tick.
i still have pieces of paper on my desk about tests done on my dual piii
where by hacking HZ to 1000 the kernel build time went from an average of
2min:30sec to an average 2min:43sec. that is pretty close to 10%
- Davide
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/