This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [perfmon] Re: perfmon2 TODO list (4/4)
- From: Stephane Eranian <eranian at hpl dot hp dot com>
- To: "Frank Ch. Eigler" <fche at redhat dot com>
- Cc: perfmon at napali dot hpl dot hp dot com, systemtap at sources dot redhat dot com
- Date: Thu, 13 Apr 2006 15:23:38 -0700
- Subject: Re: [perfmon] Re: perfmon2 TODO list (4/4)
- Address: HP Labs, 1U-17, 1501 Page Mill road, Palo Alto, CA 94304, USA.
- E-mail: eranian@hpl.hp.com
- Organisation: HP Labs Palo Alto
- References: <20060412215747.GJ29245@frankl.hpl.hp.com> <20060412220659.GL29245@frankl.hpl.hp.com> <20060412221256.GM29245@frankl.hpl.hp.com> <200604131201.59232.kevcorry@us.ibm.com> <20060413200223.GD30718@frankl.hpl.hp.com> <20060413202211.GF22490@redhat.com> <20060413215544.GH30718@frankl.hpl.hp.com> <20060413222251.GI22490@redhat.com>
- Reply-to: eranian at hpl dot hp dot com
Frank,
On Thu, Apr 13, 2006 at 06:22:51PM -0400, Frank Ch. Eigler wrote:
>
> > I have another question related maybe more to kprobes and how the
> > intercept is done: breakpoints, code rewriting. If you use
> > breakpoints, then I wonder how this works in SMP machines. Do you
> > intervene on each CPU?
>
> That's right: as each CPU trips across a breakpoint, they are made to
> run our handler, then single-step across the original instruction,
> then resume. It's a multi-step process described in kprobes
> documentation. From systemtap's point of view, it's a black box.
>
So you are saying that kprobes takes care of programming the debug
registers on all CPUs if necessary.
--
-Stephane