This is the mail archive of the
mailing list for the systemtap project.
Re: stptracer-20060828 has released.
- From: Masami Hiramatsu <masami dot hiramatsu dot pt at hitachi dot com>
- To: Li Guanglei <guanglei at cn dot ibm dot com>
- Cc: Martin Hunt <hunt at redhat dot com>, SystemTAP <systemtap at sources dot redhat dot com>, "Jose R. Santos" <jrs at us dot ibm dot com>, Ananth N Mavinakayanahalli <ananth at in dot ibm dot com>, Yumiko Sugita <yumiko dot sugita dot yf at hitachi dot com>, Satoshi Oshima <soshima at redhat dot com>, Hideo Aoki <haoki at redhat dot com>
- Date: Fri, 08 Sep 2006 21:57:43 +0900
- Subject: Re: stptracer-20060828 has released.
- Organization: Systems Development Lab., Hitachi, Ltd., Japan
- References: <44F58A51.email@example.com> <44F69E3C.firstname.lastname@example.org> <1157469166.5579.14.camel@dragon> <44FFEED1.email@example.com>
Li Guanglei wrote:
> Both Power5(ppc64) and my thinkpad show that gBTI is almost twice as
> fast as _stp_printf(%8b) and although in P-M CPU is not so obvious as
That's why I can't dispose of gBTI yet. I think most overhead comes from
parsing the format string.
> Another interesting thing is that the Power5 will take ~983ns to
> execute an empty kprobe handler(1372-391) while the thinkpad only takes
> ~233ns, almost *FOUR* times as fast as Power5. I guess it's because of
> the kprobe-booster patch and it hasn't gone into ppc64 yet, Right?
Right. Now we're porting kprobe-booster to IA64 and x86-64 architectures.
Currently, I don't have any plans to port it to ppc64 by myself, because
I don't have ppc64 machines and any knowledge about ppc64.
2nd Research Dept.
Hitachi, Ltd., Systems Development Laboratory