linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
To: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: linux-kernel@vger.kernel.org, srikar@linux.vnet.ibm.com,
	peterz@infradead.org, oleg@redhat.com,
	hegdevasant@linux.vnet.ibm.com, mingo@redhat.com,
	anton@redhat.com, systemtap@sourceware.org, namhyung@kernel.org,
	aravinda@linux.vnet.ibm.com, penberg@iki.fi
Subject: Re: [PATCH v1 0/5] perf/sdt: SDT events listing/probing
Date: Thu, 02 Oct 2014 01:08:18 +0900	[thread overview]
Message-ID: <542C26F2.6090206@hitachi.com> (raw)
In-Reply-To: <20140930152055.6698.82709.stgit@hemant-fedora>

Hi,

(2014/10/01 0:29), Hemant Kumar wrote:
> This patchset helps in listing dtrace style markers(SDT) present in user space
> applications through perf.
> Notes/markers are placed at important places by the
> developers. They have a negligible overhead when not enabled.
> We can enable them and probe at these places and find some important information
> like the arguments' values, etc.
> 
> We have lots of applications which use SDT markers today, like:
> Postgresql, MySql, Mozilla, Perl, Python, Java, Ruby, libvirt, QEMU, glib
> 
> To add SDT markers into user applications:
> We need to have this header sys/sdt.h present.
> sys/sdt.h used is version 3.
> If not present, install systemtap-sdt-devel package (for fedora-20).
> 
> With this patchset,
> - Use perf sdt-cache --add to add SDT events to a cache.
>  # perf sdt-cache --add ./user_app
> 
>   4 SDT events added for /home/user/user_app!
> 
> - Use perf sdt-cache --del to remove SDT events from the cache>
>  # perf sdt-cache --del ./user_app
> 
>  4 events removed for /home/user/user_app!
> 
> - Dump the cache onto stdout using perf sdt-cache --dump:
>  # perf sdt-cache --dump
>   /home/user/user_app :
>   %user_app:foo_start
>   %user_app:fun_start
> 
> - To probe and trace an SDT event :
>   # perf record -e %user_app:foo_start -aR sleep 10

Looks great! :)

However, when I've tried to build, I got below errors..

  CC       util/sdt.o
util/sdt.c: In function ‘sdt_err’:
util/sdt.c:72:3: error: implicit declaration of function ‘pr_err’ [-Werror=implicit-function-declaration]
   pr_err("%s: No SDT events found\n", target);
   ^
util/sdt.c:72:3: error: nested extern declaration of ‘pr_err’ [-Werror=nested-externs]
util/sdt.c: In function ‘file_hash_list__init’:
util/sdt.c:489:3: error: implicit declaration of function ‘pr_debug’ [-Werror=implicit-function-declaration]
   pr_debug("Error in madvise\n");
   ^
util/sdt.c:489:3: error: nested extern declaration of ‘pr_debug’ [-Werror=nested-externs]
cc1: all warnings being treated as errors

Perhaps, you might working on the old tree. Could you update it?

Thank you,

> 
> The support for perf to sdt events has undergone a lot of changes since it was
> introduced. After a lot of discussions (https://lkml.org/lkml/2014/7/20/284), the
> patchset is subdivided for ease of review.
> Previously, a patchset supporting "perf list sdt" was posted, but it didn't make
> much sense since, only listing SDT events for an ELF won't help.
> This patchset aims at adding a sub-command "sdt-cache" to perf to manage a cache
> for management of the SDT events.
> 
>  This link shows an example of marker probing with Systemtap:
>  https://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps
> 
>  This link provides important info regarding SDT notes:
>  http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
> 
>  - Markers in binaries :
>  These SDT markers are present in the ELF in the section named
>  ".note.stapsdt".
>  Here, the name of the marker, its provider, type, location, base
>  address, semaphore address.
>  We can retrieve these values using the members name_off and desc_off in
>  Nhdr structure. If these are not enabled, they are present in the ELF as nop.
> 
> Changes since last series :
> - Changed the entire (linked) list structure to hash lists to manage the SDT events.
> - Added a sub command "sdt-cache" to perf to manage SDT events in the system.
> - Added support to add and remove SDT events onto a cache.
> - Added support to "perf record" to probe and record the SDT events.
> 
> TODO:
> - Add support to add most of the SDT events on a system to the cache.
> - Recognizing arguments and support to probe on them.
> 
> ---
> 
> Hemant Kumar (5):
>       perf/sdt: ELF support for SDT
>       perf/sdt: Add SDT events into a cache
>       perf/sdt: Show SDT cache contents
>       perf/sdt: Delete SDT events from cache
>       perf/sdt: Add support to perf record to trace SDT events
> 
> 
>  tools/perf/Makefile.perf       |    3 
>  tools/perf/builtin-record.c    |   21 +
>  tools/perf/builtin-sdt-cache.c |  108 +++++
>  tools/perf/builtin.h           |    1 
>  tools/perf/perf.c              |    1 
>  tools/perf/util/parse-events.c |    7 
>  tools/perf/util/parse-events.h |    7 
>  tools/perf/util/probe-event.c  |  120 ++++-
>  tools/perf/util/probe-event.h  |    8 
>  tools/perf/util/probe-finder.c |    3 
>  tools/perf/util/sdt.c          |  925 ++++++++++++++++++++++++++++++++++++++++
>  tools/perf/util/sdt.h          |   30 +
>  tools/perf/util/symbol-elf.c   |  207 +++++++++
>  tools/perf/util/symbol.h       |   21 +
>  14 files changed, 1440 insertions(+), 22 deletions(-)
>  create mode 100644 tools/perf/builtin-sdt-cache.c
>  create mode 100644 tools/perf/util/sdt.c
>  create mode 100644 tools/perf/util/sdt.h
> 


-- 
Masami HIRAMATSU
Software Platform Research Dept. Linux Technology Research Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: masami.hiramatsu.pt@hitachi.com



  parent reply	other threads:[~2014-10-01 16:08 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-30 15:29 [PATCH v1 0/5] perf/sdt: SDT events listing/probing Hemant Kumar
2014-09-30 15:31 ` [PATCH v1 1/5] perf/sdt: ELF support for SDT notes Hemant Kumar
2014-09-30 15:32 ` [PATCH v1 2/5] perf/sdt: Add SDT events into a cache Hemant Kumar
2014-10-03 12:47   ` Masami Hiramatsu
2014-10-05  8:15     ` Hemant Kumar
2014-09-30 15:32 ` [PATCH v1 3/5] perf/sdt: Show SDT cache contents Hemant Kumar
2014-09-30 15:33 ` [PATCH v1 4/5] perf/sdt: Delete SDT events from cache Hemant Kumar
2014-09-30 15:34 ` [PATCH v1 5/5] perf/sdt: Add support to perf record to trace SDT events Hemant Kumar
2014-10-01 16:08 ` Masami Hiramatsu [this message]
2014-10-01 16:31   ` [PATCH v1 0/5] perf/sdt: SDT events listing/probing Hemant Kumar

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=542C26F2.6090206@hitachi.com \
    --to=masami.hiramatsu.pt@hitachi.com \
    --cc=anton@redhat.com \
    --cc=aravinda@linux.vnet.ibm.com \
    --cc=hegdevasant@linux.vnet.ibm.com \
    --cc=hemant@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=oleg@redhat.com \
    --cc=penberg@iki.fi \
    --cc=peterz@infradead.org \
    --cc=srikar@linux.vnet.ibm.com \
    --cc=systemtap@sourceware.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).