High-resolution user/system times?

[Nadav Har'El]

On Wed, Jul 25, 2012, Oleg Goldshmidt wrote about "Re: High-resolution user/system times?":
> > It appears that while times(2) has a 4-ms resolution,
> 
> 
> Sanity check: I assume you measured it, right? Out of curiosity I did
> 
> #include <unistd.h>
> #include <sys/times.h>
> #include <stdio.h>
> 
> int main(void) {
>     return printf("%ld\n",sysconf(_SC_CLK_TCK));
> }
> 
> on a couple of systems, and got 100 both times, which corresponds to a 10ms
> resolution. This is what I'd expect if HZ is 100 in the kernel.

HZ used to default to 100 in the Linux kernel, but now it actually
defaults (unless I'm mis-remembering) to 250, and this is where the 4-ms
resolution came from. But the specific number - 4 or 10ms, is not what
matters. The issue is that I can't get 0.1 ms resolution, unless I set
HZ to 10,000, which is a completely non-standard setup.

> So I gather you think this will be different from what you wanted to avoid
> in your original post  because you think the startup code will be repeated
> on each iteration? It's difficult to assess without knowing what it is that
> you are trying to measure and what, in the context, the difference is.

Right. I couldn't put a loop inside the process, but starting the same
process over and over would incur exactly the same overheads over and
over - all the same minor page faults (note: *minor* page faults means
we don't need to fetch pages from disk), and other things I suspect are
happening.


-- 
Nadav Har'El                        |        Wednesday, Jul 25 2012, 6 Av 5772
nyh at math.technion.ac.il             |-----------------------------------------
Phone +972-523-790466, ICQ 13349191 |Creativity consists of coming up with
http://nadav.harel.org.il           |many ideas, not just that one great idea.