This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Questions on Systemtap performance
- From: "Frank Ch. Eigler" <fche at redhat dot com>
- To: shan zeng <zengshan2008 at gmail dot com>
- Cc: systemtap at sources dot redhat dot com
- Date: Sat, 2 Jan 2010 06:42:54 -0500
- Subject: Re: Questions on Systemtap performance
- References: <14b21e1d1001012350w620e748ck984af9edf2d34585@mail.gmail.com> <14b21e1d1001020238i7f4d49d0pbf662ff52f0b5f28@mail.gmail.com>
Hi -
> > I run a script with the -t option and get the following result:
> > probe kernel.function("netif_
> > receive_skb").return (../path/yanshi.stp:32:1), hits: 38442, *cycles:
> > 549min/2364avg/43911max*
> >
> > what does the *cycles: 549min/2364avg/43911max *part mean? since its
> > measurement is min
The measurements are in "cycles", i.e., the basic clock rate of the CPU,
and among the 38442 probe hits, the systemtap-generated code consumed
549 cycles minimum, 2364 average, 43911 maximum.
- FChE