lkml.org 
[lkml]   [1999]   [Jul]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Measured overhead of timer interrupts
> > Hmm, I raytraced an image running three kernels differing only
> > in HZ value (+the HZ masquerading stuff). This is what i got:

> Those results look a bit strange: when increasing HZ 8 times, you get

I didn't fix the scaling before running the first HZ=1024 test.
After fixing the problem the numbers are:

100: 8.65user 0.00system 0:08.65elapsed 100%CPU (0avgtext+0avgdata 0maxresident)k
1024: 8.66user 0.01system 0:08.72elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k

So the HZ change from 100 to 1024 means only a ~0.8% slowdown for the
raytracer - looks much better now. All stability problems have magically
disappeared after upgrading to 2.3.12pre1 ;)



-
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/

\
 
 \ /
  Last update: 2005-03-22 13:53    [W:0.076 / U:0.376 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site