From: Nicholas Miell <nmiell@comcast.net>
To: Ingo Molnar <mingo@elte.hu>
Cc: Roman Zippel <zippel@linux-m68k.org>,
Thomas Gleixner <tglx@linutronix.de>,
karim@opersys.com, Andrew Morton <akpm@osdl.org>,
Paul Mundt <lethal@linux-sh.org>, Jes Sorensen <jes@sgi.com>,
Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
linux-kernel@vger.kernel.org,
Christoph Hellwig <hch@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Greg Kroah-Hartman <gregkh@suse.de>,
Tom Zanussi <zanussi@us.ibm.com>,
ltt-dev@shafik.org, Michel Dagenais <michel.dagenais@polymtl.ca>
Subject: Re: [PATCH 0/11] LTTng-core (basic tracing infrastructure) 0.5.108
Date: Fri, 15 Sep 2006 17:41:22 -0700 [thread overview]
Message-ID: <1158367282.2352.18.camel@entropy> (raw)
In-Reply-To: <20060915235707.GB29929@elte.hu>
On Sat, 2006-09-16 at 01:57 +0200, Ingo Molnar wrote:
> * Nicholas Miell <nmiell@comcast.net> wrote:
>
> > You're going to want to be able to trace every function in the kernel,
> > which means they'd all need a __trace -- and in that case, a
> > -fpad-functions-for-tracing gcc option would make more sense then
> > per-function attributes.
>
> the __trace attribute would be a _specific_ replacement for a _specific_
> static markup at the entry of a function. So no, we would not want to
> add __trace to _every_ function in the kernel: only those which get
> commonly traced. And note that SystemTap can trace the rest too, just
> with slighly higher overhead.
>
> In that sense __trace is not an enabling infrastructure, it's a
> performance tuning infrastructure.
>
> > The option could also insert NOPs before RETs, not just before the
> > prologue so that function returns are equally easy to trace. (It might
> > also inhibit tail calls, assuming being able to trace all function
> > returns is more important than that optimization.)
>
> yeah. __trace_entry and __trace_exit [or both] attributes. Makes sense.
>
> > And SystemTap can already hook into sock_sendmsg() (or any other
> > function) and examine it's arguments -- all of this GCC extension talk
> > is just performance enhancement.
>
> yes, yes, yes, exactly!!! Finally someone reads my mails and understands
> my points. There's hope! ;)
I'm not sure that I do, actually.
You seem to be opposed to all static probe markers in general, but I
think that they'd be useful for big abstract things like "new thread
created" (which would encompass fork/vfork/clone and probably consist of
a single marker in do_fork) or for similar things that happen all over
the kernel (for example, I imagine that all filesystems would want to
use the same set of probe names just to make I/O tracing easier for
userspace).
--
Nicholas Miell <nmiell@comcast.net>
next prev parent reply other threads:[~2006-09-16 0:43 UTC|newest]
Thread overview: 284+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-14 3:38 [PATCH 0/11] LTTng-core (basic tracing infrastructure) 0.5.108 Mathieu Desnoyers
2006-09-14 11:27 ` Ingo Molnar
2006-09-14 13:40 ` Roman Zippel
2006-09-14 13:55 ` Ingo Molnar
2006-09-14 14:33 ` Roman Zippel
2006-09-14 15:26 ` Michel Dagenais
2006-09-14 17:48 ` Ingo Molnar
2006-09-15 15:04 ` Mathieu Desnoyers
2006-09-14 18:08 ` Nick Piggin
2006-09-14 18:38 ` Karim Yaghmour
2006-09-14 17:13 ` Ingo Molnar
2006-09-14 17:55 ` Roman Zippel
2006-09-14 18:15 ` Ingo Molnar
2006-09-14 18:35 ` Mathieu Desnoyers
2006-09-14 18:54 ` Karim Yaghmour
2006-09-15 9:20 ` Jes Sorensen
2006-09-15 12:38 ` Karim Yaghmour
2006-09-15 12:32 ` Jes Sorensen
2006-09-15 14:09 ` Karim Yaghmour
2006-09-15 14:30 ` Jes Sorensen
2006-09-15 15:12 ` Karim Yaghmour
2006-09-16 10:41 ` Jes Sorensen
2006-09-16 15:28 ` Karim Yaghmour
2006-09-18 8:57 ` Jes Sorensen
2006-09-18 14:48 ` Ingo Molnar
2006-09-18 15:37 ` Karim Yaghmour
2006-09-15 13:20 ` Paul Mundt
2006-09-15 13:41 ` Roman Zippel
2006-09-15 13:44 ` Jes Sorensen
2006-09-15 14:03 ` Roman Zippel
2006-09-15 14:37 ` Alan Cox
2006-09-15 14:34 ` Roman Zippel
2006-09-15 13:57 ` Paul Mundt
2006-09-15 14:17 ` Karim Yaghmour
2006-09-15 14:13 ` Jes Sorensen
2006-09-15 14:31 ` Karim Yaghmour
2006-09-15 14:28 ` Paul Mundt
2006-09-15 14:46 ` Martin J. Bligh
2006-09-15 15:22 ` Alan Cox
2006-09-15 15:47 ` Martin J. Bligh
2006-09-15 14:51 ` Karim Yaghmour
2006-09-15 15:00 ` Thomas Gleixner
2006-09-15 15:28 ` Karim Yaghmour
2006-09-15 18:16 ` Andrew Morton
2006-09-15 18:19 ` Ingo Molnar
2006-09-15 19:26 ` Karim Yaghmour
2006-09-15 19:43 ` Roman Zippel
2006-09-15 20:05 ` Ingo Molnar
2006-09-15 20:22 ` Mathieu Desnoyers
2006-09-15 21:08 ` Jose R. Santos
2006-09-15 21:25 ` Mathieu Desnoyers
2006-09-15 22:02 ` Jose R. Santos
2006-09-15 22:03 ` Ingo Molnar
2006-09-15 22:32 ` Karim Yaghmour
2006-09-15 22:43 ` Ingo Molnar
2006-09-15 23:33 ` Karim Yaghmour
2006-09-15 23:52 ` Ingo Molnar
2006-09-16 2:24 ` Karim Yaghmour
2006-09-15 23:53 ` Ingo Molnar
2006-09-16 2:51 ` Karim Yaghmour
2006-09-15 22:59 ` Frank Ch. Eigler
2006-09-15 23:40 ` Karim Yaghmour
2006-09-15 23:17 ` Jose R. Santos
2006-09-15 21:32 ` Ingo Molnar
2006-09-15 21:58 ` Mathieu Desnoyers
2006-09-15 22:19 ` Ingo Molnar
2006-09-15 22:45 ` Karim Yaghmour
2006-09-16 9:59 ` Jes Sorensen
2006-09-16 17:24 ` Mathieu Desnoyers
2006-09-16 17:35 ` Ingo Molnar
2006-09-16 17:56 ` Mathieu Desnoyers
2006-09-16 19:10 ` Ingo Molnar
2006-09-16 19:37 ` Ingo Molnar
2006-09-16 20:29 ` [patch] kprobes: optimize branch placement Ingo Molnar
2006-09-16 20:43 ` Ingo Molnar
2006-09-16 21:44 ` Karim Yaghmour
2006-09-16 22:57 ` Andrew Morton
2006-09-16 23:24 ` Ingo Molnar
2006-09-17 2:44 ` Karim Yaghmour
2006-09-17 1:59 ` Karim Yaghmour
2006-09-17 12:15 ` Mike Galbraith
2006-09-16 22:56 ` Andrew Morton
2006-09-16 23:30 ` Ingo Molnar
2006-09-18 13:52 ` Ananth N Mavinakayanahalli
2006-09-16 22:55 ` Andrew Morton
2006-09-17 10:13 ` [PATCH 0/11] LTTng-core (basic tracing infrastructure) 0.5.108 Frederik Deweerdt
2006-09-17 14:00 ` Ingo Molnar
2006-09-16 19:51 ` Karim Yaghmour
2006-09-16 23:40 ` Ingo Molnar
2006-09-17 5:33 ` Mathieu Desnoyers
2006-09-16 18:11 ` Karim Yaghmour
2006-09-16 17:44 ` Ingo Molnar
2006-09-16 18:15 ` Karim Yaghmour
2006-09-18 8:18 ` Jes Sorensen
2006-09-16 17:55 ` Karim Yaghmour
2006-09-18 8:21 ` Jes Sorensen
2006-09-18 8:33 ` Jes Sorensen
2006-09-18 15:01 ` Mathieu Desnoyers
2006-09-16 17:30 ` Mathieu Desnoyers
2006-09-18 8:15 ` Jes Sorensen
2006-09-18 14:53 ` Mathieu Desnoyers
2006-09-18 15:17 ` Ingo Molnar
2006-09-18 16:54 ` Mathieu Desnoyers
2006-09-15 21:12 ` Roman Zippel
2006-09-15 21:08 ` Ingo Molnar
2006-09-15 20:13 ` Andrew Morton
2006-09-15 21:49 ` Jose R. Santos
2006-09-16 10:19 ` Jes Sorensen
2006-09-16 16:05 ` Karim Yaghmour
2006-09-17 4:54 ` Ganesan Rajagopal
2006-09-18 8:13 ` Jes Sorensen
2006-09-18 14:46 ` Mathieu Desnoyers
2006-09-18 17:06 ` Martin Bligh
2006-09-20 14:17 ` Jes Sorensen
2006-09-15 19:35 ` Thomas Gleixner
2006-09-15 19:40 ` Ingo Molnar
2006-09-15 19:56 ` Karim Yaghmour
2006-09-15 20:23 ` Thomas Gleixner
2006-09-15 20:40 ` Roman Zippel
2006-09-15 20:48 ` Ingo Molnar
2006-09-15 21:17 ` Karim Yaghmour
2006-09-15 21:15 ` Ingo Molnar
2006-09-15 21:56 ` Karim Yaghmour
2006-09-15 21:27 ` Roman Zippel
2006-09-15 21:51 ` Ingo Molnar
2006-09-15 22:15 ` Karim Yaghmour
2006-09-15 22:53 ` Roman Zippel
2006-09-15 23:14 ` Ingo Molnar
2006-09-15 23:49 ` Nicholas Miell
2006-09-15 23:57 ` Ingo Molnar
2006-09-16 0:41 ` Nicholas Miell [this message]
2006-09-16 0:31 ` Roman Zippel
2006-09-16 8:20 ` Ingo Molnar
2006-09-16 8:21 ` Ingo Molnar
2006-09-16 8:21 ` Ingo Molnar
2006-09-16 8:22 ` Ingo Molnar
2006-09-16 19:58 ` Roman Zippel
2006-09-16 22:50 ` Ingo Molnar
2006-09-16 23:00 ` Ingo Molnar
2006-09-17 1:15 ` Roman Zippel
2006-09-17 8:42 ` Ingo Molnar
2006-09-17 15:16 ` Roman Zippel
2006-09-17 15:25 ` Ingo Molnar
2006-09-17 16:02 ` Roman Zippel
2006-09-17 16:45 ` Ingo Molnar
2006-09-17 16:59 ` Nick Piggin
2006-09-17 17:26 ` Roman Zippel
2006-09-17 17:56 ` Nick Piggin
2006-09-17 18:59 ` Roman Zippel
2006-09-17 21:23 ` Ingo Molnar
2006-09-17 21:52 ` Roman Zippel
2006-09-17 22:27 ` Ingo Molnar
2006-09-17 21:40 ` Ingo Molnar
2006-09-18 8:43 ` Jes Sorensen
2006-09-17 21:32 ` Ingo Molnar
2006-09-17 19:23 ` Ingo Molnar
2006-09-17 19:45 ` Roman Zippel
2006-09-17 20:56 ` Ingo Molnar
2006-09-17 21:36 ` Roman Zippel
2006-09-17 22:13 ` Ingo Molnar
2006-09-16 23:14 ` Ingo Molnar
2006-09-17 14:19 ` Frank Ch. Eigler
2006-09-17 15:31 ` Ingo Molnar
2006-09-17 17:15 ` Mathieu Desnoyers
[not found] ` <y0mu036eglz.fsf@ton.toronto.redhat.com>
2006-09-17 15:00 ` Ingo Molnar
2006-09-16 8:23 ` Ingo Molnar
2006-09-16 8:23 ` Ingo Molnar
2006-09-16 8:23 ` Ingo Molnar
2006-09-15 21:05 ` Karim Yaghmour
2006-09-15 21:17 ` Thomas Gleixner
2006-09-15 21:31 ` Karim Yaghmour
2006-09-15 20:00 ` Mathieu Desnoyers
2006-09-15 20:27 ` Jose R. Santos
2006-09-15 20:37 ` Alan Cox
2006-09-15 20:26 ` Mathieu Desnoyers
2006-09-15 20:51 ` Karim Yaghmour
2006-09-17 17:53 ` Mathieu Desnoyers
2006-09-17 19:35 ` printk instrumentation with LTTng Mathieu Desnoyers
2006-09-15 15:24 ` [PATCH 0/11] LTTng-core (basic tracing infrastructure) 0.5.108 Alan Cox
2006-09-15 15:23 ` Karim Yaghmour
2006-09-15 14:39 ` Jes Sorensen
2006-09-15 15:04 ` Karim Yaghmour
2006-09-14 19:40 ` Tim Bird
2006-09-14 20:00 ` Ingo Molnar
2006-09-14 20:46 ` Karim Yaghmour
2006-09-19 12:05 ` Christoph Hellwig
2006-09-14 21:02 ` Roman Zippel
2006-09-15 11:40 ` Alan Cox
2006-09-15 11:46 ` Roman Zippel
2006-09-15 12:38 ` Alan Cox
2006-09-15 12:39 ` Roman Zippel
2006-09-15 13:41 ` Alan Cox
2006-09-15 13:34 ` Roman Zippel
2006-09-15 14:41 ` Alan Cox
2006-09-15 14:35 ` Karim Yaghmour
2006-09-15 14:58 ` Alan Cox
2006-09-15 14:57 ` Karim Yaghmour
2006-09-15 17:49 ` Andrew Morton
2006-09-15 18:20 ` Karim Yaghmour
2006-09-15 17:01 ` Tim Bird
2006-09-15 17:08 ` Frank Ch. Eigler
2006-09-15 17:57 ` Andrew Morton
2006-09-15 18:31 ` Alan Cox
2006-09-15 18:12 ` Ingo Molnar
2006-09-15 19:10 ` Roman Zippel
2006-09-15 19:10 ` Ingo Molnar
2006-09-15 20:05 ` Thomas Gleixner
2006-09-15 20:35 ` Roman Zippel
2006-09-15 21:44 ` Tim Bird
2006-09-19 12:29 ` Christoph Hellwig
2006-09-19 13:17 ` Roman Zippel
2006-09-15 18:24 ` Frank Ch. Eigler
2006-09-15 18:23 ` Ingo Molnar
2006-09-15 18:18 ` Martin Bligh
2006-09-15 18:10 ` Jose R. Santos
2006-09-15 19:49 ` Mathieu Desnoyers
2006-09-15 20:54 ` Jose R. Santos
2006-09-15 21:42 ` Karim Yaghmour
2006-09-15 21:46 ` Mathieu Desnoyers
2006-09-19 15:05 ` Jose R. Santos
2006-09-19 15:30 ` Mathieu Desnoyers
2006-09-19 16:39 ` Jose R. Santos
2006-09-19 18:03 ` Mathieu Desnoyers
2006-09-15 17:45 ` Andrew Morton
2006-09-15 18:16 ` Karim Yaghmour
2006-09-15 19:20 ` Jose R. Santos
2006-09-15 19:59 ` Andrew Morton
2006-09-15 20:24 ` Karim Yaghmour
2006-09-15 20:25 ` Thomas Gleixner
2006-09-14 19:47 ` Roman Zippel
2006-09-14 20:24 ` Ingo Molnar
2006-09-14 20:54 ` Roman Zippel
2006-09-14 21:08 ` Daniel Walker
2006-09-14 21:30 ` Roman Zippel
2006-09-14 22:15 ` Ingo Molnar
2006-09-14 23:39 ` Roman Zippel
2006-09-14 23:43 ` Ingo Molnar
2006-09-15 0:27 ` Roman Zippel
2006-09-15 1:47 ` Mathieu Desnoyers
2006-09-15 5:47 ` Vara Prasad
2006-09-14 18:12 ` Karim Yaghmour
2006-09-14 20:25 ` Martin Bligh
2006-09-14 20:34 ` Ingo Molnar
2006-09-14 20:55 ` Martin Bligh
2006-09-14 21:31 ` Ingo Molnar
2006-09-14 22:25 ` Martin Bligh
2006-09-14 22:36 ` Ingo Molnar
2006-09-14 22:59 ` Martin Bligh
2006-09-14 23:19 ` Ingo Molnar
2006-09-15 0:19 ` Nicholas Miell
2006-09-15 1:04 ` Martin J. Bligh
2006-09-15 12:38 ` Ingo Molnar
2006-09-15 7:00 ` Vara Prasad
2006-09-15 15:37 ` Michel Dagenais
2006-09-19 12:08 ` Christoph Hellwig
2006-09-14 21:07 ` Roman Zippel
2006-09-15 9:29 ` Jes Sorensen
2006-09-14 17:51 ` Karim Yaghmour
2006-09-14 15:19 ` Mathieu Desnoyers
2006-09-14 19:39 ` Frank Ch. Eigler
2006-09-15 17:13 ` Jose R. Santos
2006-09-14 15:02 ` Mathieu Desnoyers
2006-09-14 15:14 ` Martin J. Bligh
2006-09-14 17:43 ` Ingo Molnar
2006-09-14 18:25 ` Karim Yaghmour
2006-09-14 20:03 ` Martin Bligh
2006-09-14 20:14 ` Ingo Molnar
2006-09-14 20:40 ` Martin Bligh
2006-09-14 21:05 ` Michel Dagenais
2006-09-14 22:23 ` Ingo Molnar
2006-09-14 22:46 ` Martin Bligh
2006-09-14 22:56 ` Ingo Molnar
2006-09-14 19:03 ` grundig
2006-09-14 19:21 ` Karim Yaghmour
2006-09-14 19:48 ` Frank Ch. Eigler
2006-09-15 16:32 ` Jose R. Santos
2006-09-19 11:59 ` Christoph Hellwig
-- strict thread matches above, loose matches on Subject: below --
2006-09-15 3:10 James Dickens
2006-09-15 9:17 Richard J Moore
2006-09-15 17:14 Chuck Ebbert
2006-09-15 18:32 ` Alan Cox
2006-09-16 10:46 ` Jes Sorensen
2006-09-25 15:20 Chuck Ebbert
2006-09-25 15:39 ` Ingo Molnar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1158367282.2352.18.camel@entropy \
--to=nmiell@comcast.net \
--cc=akpm@osdl.org \
--cc=gregkh@suse.de \
--cc=hch@infradead.org \
--cc=jes@sgi.com \
--cc=karim@opersys.com \
--cc=lethal@linux-sh.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ltt-dev@shafik.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=michel.dagenais@polymtl.ca \
--cc=mingo@elte.hu \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=zanussi@us.ibm.com \
--cc=zippel@linux-m68k.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.