This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [RFC PATCH tracing/kprobes 0/5] tracing/kprobes, perf: perf kprobe support
- From: Arnaldo Carvalho de Melo <acme at redhat dot com>
- To: Ingo Molnar <mingo at elte dot hu>
- Cc: Masami Hiramatsu <mhiramat at redhat dot com>, Frederic Weisbecker <fweisbec at gmail dot com>, Steven Rostedt <rostedt at goodmis dot org>, lkml <linux-kernel at vger dot kernel dot org>, Thomas Gleixner <tglx at linutronix dot de>, Arnaldo Carvalho de Melo <acme at redhat dot com>, Mike Galbraith <efault at gmx dot de>, Paul Mackerras <paulus at samba dot org>, Peter Zijlstra <a dot p dot zijlstra at chello dot nl>, Christoph Hellwig <hch at infradead dot org>, Ananth N Mavinakayanahalli <ananth at in dot ibm dot com>, Jim Keniston <jkenisto at us dot ibm dot com>, "Frank Ch. Eigler" <fche at redhat dot com>, systemtap <systemtap at sources dot redhat dot com>, DLE <dle-develop at lists dot sourceforge dot net>
- Date: Wed, 30 Sep 2009 11:57:58 -0300
- Subject: Re: [RFC PATCH tracing/kprobes 0/5] tracing/kprobes, perf: perf kprobe support
- References: <20090925191424.12939.91503.stgit@omoto> <4AC2AF01.9090202@redhat.com> <20090930120418.GB7618@elte.hu>
Em Wed, Sep 30, 2009 at 02:04:18PM +0200, Ingo Molnar escreveu:
>
> * Masami Hiramatsu <mhiramat@redhat.com> wrote:
>
> > Masami Hiramatsu wrote:
> >>
> >> Hi,
> >>
> >> These patches introduce perf kprobe command and update kprobe-tracer.
> >> perf kprobe command allows you to add new probe points by C line number
> >> and local variable names.
> >
> > Last week, Arnaldo and I talked about this command, and he suggested
> > that the command would be better 'perf probe', because it would be
> > able to cover both of kernel space (by kprobes) and user space (by
> > uprobes).
>
> Agreed.
>
> > Basically, I agree with his idea. But I think we may need to consider
> > more flexible syntax for that purpose before we support uprobes. In
> > this area, SystemTap has done big advance, we can see how many
> > varieties of syntax it has by 'man stapprobes'.
> >
> > And also, it's hard to decide it without real uprobe-tracer (and
> > uprobes too!) implementation on ftrace. So, I think it is better to
> > continue using 'perf kprobe' in this time.
> >
> > But it's worth to add to todo list. :)
>
> I'd still name it 'perf probe', even if initially it supports kprobes.
Agreed.
- Arnaldo