This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: gettimeofday_ns accuracy
- From: fche at redhat dot com (Frank Ch. Eigler)
- To: "Rob Donovan" <rob at proivrc dot com>
- Cc: <systemtap at sourceware dot org>
- Date: Sat, 07 Aug 2010 17:53:35 -0400
- Subject: Re: gettimeofday_ns accuracy
- References: <010601cb3652$4c05ae50$e4110af0$@proivrc.com>
"Rob Donovan" <rob@proivrc.com> writes:
> Is there any other kind of time measure other than gettimeofday_ns()?
There is also get_cycles() after the epolymous kernel function.
> Since gettimeofday_ns is not that accurate, and has the possibility
> (and does) sometimes go backwards in time.
Can you describe your system (CPU models, kernel versions) a little
more? We may be able to use/expose other time sources too, or help
you write the bit of code that would do so.
- FChE