[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20100709141335.GG5776@ghostprotocols.net>
Date: Fri, 9 Jul 2010 11:13:35 -0300
From: Arnaldo Carvalho de Melo <acme@...radead.org>
To: Srikar Dronamraju <srikar@...ux.vnet.ibm.com>
Cc: Christoph Hellwig <hch@...radead.org>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...e.hu>,
Steven Rostedt <rostedt@...dmis.org>,
Randy Dunlap <rdunlap@...otime.net>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>,
Oleg Nesterov <oleg@...hat.com>,
Mark Wielaard <mjw@...hat.com>,
Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
LKML <linux-kernel@...r.kernel.org>,
Naren A Devaiah <naren.devaiah@...ibm.com>,
Jim Keniston <jkenisto@...ux.vnet.ibm.com>,
Frederic Weisbecker <fweisbec@...il.com>,
"Frank Ch. Eigler" <fche@...hat.com>,
Ananth N Mavinakayanahalli <ananth@...ibm.com>,
Andrew Morton <akpm@...ux-foundation.org>,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
Subject: Re: [PATCHv8 2.6.35-rc4-tip 12/12] perf: Show Potential probe points.
Em Fri, Jul 09, 2010 at 11:26:44AM +0530, Srikar Dronamraju escreveu:
> > > Show all potential probes in a process by pid 21106 across all dsos
> > > # perf probe -S -p 21106 | tail
> > Again, a file based interface would be much more useful. E.g. what
> > possible probe points do I have in my qemu binary.
> Right, Currently perf probe needs -p option because it depend on
> thread->map_groups for listing the dsos and its functions.
> I am not sure if perf already has ability to list functions in a file
> esp if we dont have dwarf information in that file. If perf has the
> ability I will reuse it here or will try implementing one.
> Arnaldo, Do you know offhand if perf already has the ability to list
> functions in a file (with / without dwarf info).
Definetely, it actually doesn't use DWARF at all, its just the
symtabs that are used. It prefers -debuginfo packages because those have
richer symtabs, but if just the stripped file is used it will try to use
.dynsym, etc.
Look at tools/perf/builtin-test.c, it has a regression test for the
symbol lib, should be enough.
- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists