Hi all,
Are any details available of the memory requirements for LTTng? I'm
particularly interested in UST, and not only the case where tracing is being
performed, but also when no session is started. For example.:
- Session daemon memory requirements: I assume this would be a function of
I ran some tests with the new function entry/exit instrumentations.
The tracepoint provider for lttng_ust_cyg_profile:func_entry and
func_exit does not properly forward the call_site argument to the
call_site field. The patch below fixes the problem.
From
merged, thanks!
Mathieu
* Woegerer, Paul (paul_woege...@mentor.com) wrote:
I ran some tests with the new function entry/exit instrumentations.
The tracepoint provider for lttng_ust_cyg_profile:func_entry and
func_exit does not properly forward the call_site argument to the
call_site field.
As promised...
From acba8cd14abc7e61cf1346b5649972bac3ea0d53 Mon Sep 17 00:00:00 2001
From: Paul Woegerer paul_woege...@mentor.com
Date: Wed, 27 Mar 2013 17:02:27 +0100
Subject: [PATCH] Add man page for lttng-ust-cyg-profile
---
doc/Makefile.am |3 +-
Merged!
Davd
Mathieu Desnoyers:
Signed-off-by: Mathieu Desnoyers mathieu.desnoy...@efficios.com
---
diff --git a/src/bin/lttng-sessiond/context.c
b/src/bin/lttng-sessiond/context.c
index a83cb46..b55254b 100644
--- a/src/bin/lttng-sessiond/context.c
+++ b/src/bin/lttng-sessiond/context.c
Here are proposed semantics for the tracefile-size/tracefile-count
enable-channel options (are they still planned for later inclusion in
lttng-tools?):
-C, --tracefile-size SIZE
This is a recent proposal (2013-Mar-08) that has yet to be added to the
command. Maximum size of each
Hi,
Two things. First, the version.h.tmpl has to be added to EXTRA_DIST in
include/Makefile.am
Second thing, on line 86 of lttng.c, *never* *never* *never* pass a
variable to printf without a format.
fprintf(ofp, lttng_version);
This is a huge vector attack to format strings. Furthermore,
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
.gitignore |2 ++
Makefile.am |4 +--
include/Makefile.am | 46 +++
include/version.h.tmpl | 27
On 2013-03-25 10:30, David Goulet wrote: Hi,
Two things. First, the version.h.tmpl has to be added to EXTRA_DIST in
include/Makefile.am
Done. And changed the version.h to nodist_*_headers.
Second thing, on line 86 of lttng.c, *never* *never* *never* pass a
variable to printf without a
Add a new socket between the sessiond and the ust-consumer to allow
periodical flush of the metadata channel.
If enabled (by specifying the --switch-timer option on the metadata
channel), a new timer thread in the consumer asks the session daemon
for new metadata for a specific session.
All the
Signed-off-by: Mathieu Desnoyers mathieu.desnoy...@efficios.com
---
diff --git a/src/bin/lttng-sessiond/context.c b/src/bin/lttng-sessiond/context.c
index a83cb46..b55254b 100644
--- a/src/bin/lttng-sessiond/context.c
+++ b/src/bin/lttng-sessiond/context.c
@@ -262,7 +262,7 @@ int
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
.gitignore |2 ++
Makefile.am |4 +--
include/Makefile.am | 45 ++-
include/version.h.tmpl | 27
Hi all,
I've got into a build error on Ubuntu 12.04 (i386) using gcc 4.6.3. The
compiler seems to get confused when asked to equivalate unsigned long and
uint64_t on my machine. Since the problem was pretty obvious, I've also made a
patch for it.
Regards,
Dragos
From
Hi,
Thanks for the patch.
It has been fixed and will be upstream very soon!
Thanks for the patch!
David
Dragos Tarcatu:
Hi all,
I've got into a build error on Ubuntu 12.04 (i386) using gcc 4.6.3. The
compiler seems to get confused when asked to equivalate unsigned long
and uint64_t on my
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
.gitignore |2 ++
include/Makefile.am | 45 ++-
include/version.h.tmpl | 27 +++
On 2013-03-16 19:49, Raphaël Beamonte wrote:
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
.gitignore |2 ++
include/Makefile.am | 45
++-
include/version.h.tmpl | 27
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
.gitignore |2 ++
include/Makefile.am | 26 +-
include/version.h.tmpl | 27 +++
* Raphaël Beamonte (raphael.beamo...@gmail.com) wrote:
Hello,
This version adds a line Git version to LTTng --help. The value of
this line is calculated during make and then defined in a GIT_VERSION
macro.
It also works for versions without Git as it will just not define the
GIT_VERSION
Hi Raphaël,
I wonder if we would like to also have the same information when
running lttng-sessiond --version?
MSG(lttng version VERSION - VERSION_NAME);
+#ifdef GIT_VERSION
+ MSG(Git version: GIT_VERSION);
+#endif
MSG(\n VERSION_DESCRIPTION \n);
MSG(Web
We could simply have let say version.h that is used for that and put
into the right C files? (sessiond and lttng).
David
Christian Babeux:
Hi Raphaël,
I wonder if we would like to also have the same information when
running lttng-sessiond --version?
MSG(lttng version VERSION -
On 2013-03-13 11:40, David Goulet wrote:
We could simply have let say version.h that is used for that and put
into the right C files? (sessiond and lttng).
I agree, I was thinking about it since this morning's mail of Mathieu.
A file that could be auto-generated by the make command if the value
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
configure.ac |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index 7e35c06..c8aa034 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,4 +1,4 @@
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
configure.ac |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index fdc5c84..6f6f885 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,7 +1,7 @@
#
Hi Raphaël,
This is not very relevant because we branch at each release so the
master version is a bit useless.
Do you have a need for that?
Thanks
David
Raphaël Beamonte:
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
configure.ac |2 +-
1 file changed, 1
Hi David,
The tag version changed in LTTng but not the displayed version, and it made
me remove the whole LTTng tools from my computer and reinstall them as I
didn't understand why the version number displayed wasn't the one from the
version I was installing. It's not a very important bug, but it
://lttng.org])
AC_CONFIG_AUX_DIR([config])
AC_CANONICAL_TARGET
AC_CANONICAL_HOST
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
2013/3/12 Yannick Brosseau yannick.bross...@gmail.com
Maybe we should find a way to add the git commit id and maybe other git
info to the version display when we build it from git.
I agree with Yannick. That's something that could be very useful.
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
src/bin/lttng/Makefile.am |6 +-
src/bin/lttng/lttng.c |6 +-
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/src/bin/lttng/Makefile.am b/src/bin/lttng/Makefile.am
index 21eebab..1b1219b 100644
---
Signed-off-by: Raphaël Beamonte raphael.beamo...@gmail.com
---
src/bin/lttng/Makefile.am|6 +-
src/bin/lttng/commands/version.c |3 +++
src/bin/lttng/lttng.c|6 +-
3 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/src/bin/lttng/Makefile.am
@@ -78,6 +80,8 @@ static struct poptOption long_options[] = {
{read-timer, 0, POPT_ARG_INT, 0, OPT_READ_TIMER, 0, 0},
{list-options, 0, POPT_ARG_NONE, NULL, OPT_LIST_OPTIONS, NULL,
NULL},
{output, 0, POPT_ARG_STRING, opt_output, 0, 0, 0},
+
This patch introduces the tracefile_size and tracefile_count parameters
to the enable-channel command and the API.
This allows to split a stream into multiple tracefiles and limit the
amount of trace data to keep on disk.
The tracefiles are readable independently or with the others as long as
the
Hi Andrew,
Commit:
commit b87700e318c27267890cbd6fb5e50b687279131b
Author: Andrew Gabbasov andrew_gabba...@mentor.com
Date: Mon Dec 10 11:14:52 2012 -0500
Add new kernel probes instrumentation
Add kernel probes for btrfs, compaction, ext4, printk, random, rcu,
regmap,
The lttng man pages are a little vague as to the relationship between lttng,
the lttng-sessiond, and the tracing group. My experiments have shown that a
user belonging to the tracing group should more or less be able to issue lttng
commands as if they were prefixed with sudo, but that's not
Hi Daniel,
First observation
I'm not able to reproduce your first observation on my machine.
[...]
$ sudo lttng -vvv create rootlevel
DEBUG3: URI string:
file:///home/daniel/lttng-traces/rootlevel-20130206-091954 [in uri_parse() at
uri.c:253]
DEBUG3: URI file destination:
-Message d'origine-
De : christian.bab...@0x80.ca [mailto:christian.bab...@0x80.ca] De la part de
Christian Babeux
Envoyé : 6 février 2013 12:52
First observation
I'm not able to reproduce your first observation on my machine.
[...]
$ sudo lttng -vvv create rootlevel
Hi,
I want to use ltt-ng on my 2.6.32.42 kernel, uclibc, mips platform. Can anyone
suggest what are all the requirements for this?
Regards,
Swaroop.___
lttng-dev mailing list
lttng-dev@lists.lttng.org
Hi everyone,
After some initial discussion with the LTTng developers, we think it
would be a great opportunity for the LTTng project to participate as a
mentoring organisation in Google Summer of Code 2013.
Apparently, some attempts were made in the past years but the
application deadline was
My two Ubuntu installations have the same set of LTTng modules and tools
(they were all built from the same git.lttng.org packages) but different
kernels and installation histories. The clean machine has kernel
linux-virtual (3.2.0-36-virtual), the other one has linux-generic
Hi,
Is there a way to build the LTTng tools to Android (3.0.31) on ARM ??
I've managed to compile the modules(2.0.6), put them in the Android device
and modprobe them.
Now I would like to extract the traces logs from the device.
How can I do it ?
Is there any support for this at all in the LTTng
-dev@lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
.
How can I do it ?
Is there any support for this at all in the LTTng project ?
Thank you for your response !
Amit.
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Hi,
I'm one of the student working on porting lttng to android at Ecole
Polytechnique Montreal.
As of right now we haven't begun to develop tools to support creating and
extracting traces on android devices.
It should take a few months before we finish the project, in the mean time
you could
the device.
How can I do it ?
Is there any support for this at all in the LTTng project ?
Thank you for your response !
Amit.
___
lttng-dev mailing
listlttng-dev@lists.lttng.orghttp://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
* Yannick Brosseau (yannick.bross...@gmail.com) wrote:
Inspired by the k*probe support.
It add a new option to enable-event --uprobe which require a path name
and an offset in the specified file.
(The file Documentation/trace/uprobetracer.txt in the kernel source provides
and example on
On 2013-01-16 10:52, Mathieu Desnoyers wrote:
* Yannick Brosseau (yannick.bross...@gmail.com) wrote:
Inspired by the k*probe support.
It add a new option to enable-event --uprobe which require a path name
and an offset in the specified file.
(The file Documentation/trace/uprobetracer.txt in
* Yannick Brosseau (yannick.bross...@gmail.com) wrote:
The added support is basic. It create an event with no data associated
to the file path + offset specified.
As per the structures currently used we cannot pass a file path bigger
than 256 chars.
Using a pointer to a user-space string will
-Message d'origine-
Envoyé : 11 janvier 2013 17:45
Turning to the Internet, I find
https://github.com/giraldeau/lttng-ust/tree/master/doc/examples/easy-ust
and grab the handful of files found there.
I then open a console within that directory and invoke make:
Please use an
On 01/13/2013 06:40 PM, Yannick Brosseau wrote:
+ ret = uprobe_register(event-u.uprobe.inode,
+ event-u.uprobe.offset,
+ event-u.uprobe.up_consumer);
+ uprobe_unregister(event-u.uprobe.inode,
+ event-u.uprobe.offset,
+
On 2013-01-14 13:05, Josh Stone wrote:
On 01/13/2013 06:40 PM, Yannick Brosseau wrote:
+ret = uprobe_register(event-u.uprobe.inode,
+event-u.uprobe.offset,
+event-u.uprobe.up_consumer);
+uprobe_unregister(event-u.uprobe.inode,
+
Interestingly, if I do install lttng-ust 2.0.2 (a367ee6) 2013-Apr-18,
easy-ust's make compiles correctly. One key difference is that
/usr[/local]/include/lttng/tracepoint.h now includes an enum setting the
TRACE_WARNING value.
The conclusion is inescapable: the Ubuntu liblttng-ust-dev
* Yannick Brosseau (yannick.bross...@gmail.com) wrote:
On 2013-01-14 13:05, Josh Stone wrote:
On 01/13/2013 06:40 PM, Yannick Brosseau wrote:
+ ret = uprobe_register(event-u.uprobe.inode,
+ event-u.uprobe.offset,
+ event-u.uprobe.up_consumer);
+
+1
Especially because Uprobe is in mainline since 3.5, would be nice to
support it before it gets in 3.8+ ...
Cheers!
David
Mathieu Desnoyers:
* Yannick Brosseau (yannick.bross...@gmail.com) wrote:
On 2013-01-14 13:05, Josh Stone wrote:
On 01/13/2013 06:40 PM, Yannick Brosseau wrote:
+ ret
Inspired by the k*probe support.
It add a new option to enable-event --uprobe which require a path name
and an offset in the specified file.
(The file Documentation/trace/uprobetracer.txt in the kernel source provides
and example on how to extract the offset for a symbol in a program)
I have this Ubuntu 12.04 machine with the lttng-tools suite installed, and I
now want to quickstart userspace tracing. Starting at
http://lttng.org/quickstart, we're told succinctly that we need to instrument
the target application (details in the lttng-ust(3) manpage) and then do the
Merged!
Thanks!
David
Yannick Brosseau:
v2: Update description
Signed-off-by: Yannick Brosseau yannick.bross...@gmail.com
---
.gitignore|2 +-
configure.ac |1 +
src/lib/lttng-ctl/Makefile.am |3 +++
Hi David
I just restarted working on the LTTng 2.1 integration in Eclipse this
week. So, I only noticed these things yesterday.
Thanks a lot for the fixes and the explanations.
BR,
Bernd
On 12/20/2012 02:11 PM, David Goulet wrote:
Just in time before I push for 2.1 sable.
Bernd Hufmann:
Hello
I'm currently using lttng-tools 2.1.0-RC9. I see some errors when using
lttng create with streaming options:
1) lttng create mysession -C net://192:1.1.1: -D net://192:1.1.1.1
The command output shows:
Session mysession created.
Traces will be written in (null)
Control URL
Just in time before I push for 2.1 sable.
Bernd Hufmann:
Hello
I'm currently using lttng-tools 2.1.0-RC9. I see some errors when using
lttng create with streaming options:
1) lttng create mysession -C net://192:1.1.1: -D net://192:1.1.1.1
The command output shows:
Session
v2: Update description
Signed-off-by: Yannick Brosseau yannick.bross...@gmail.com
---
.gitignore|2 +-
configure.ac |1 +
src/lib/lttng-ctl/Makefile.am |3 +++
src/lib/lttng-ctl/lttng-ctl.pc.in | 13 +
4 files changed, 18
On 2012-12-20 14:31, Yannick Brosseau wrote:
+Description: The LTTng control and utility library is a library used to
control the tracing sessions of a LTTng-session deamon
Following our discussions on the IRC channel, I would propose a
correction of the description :
The LTTng control and
Hello,
I'm trying to use lttng tools to identify some block stack latencies. I'm
tracing
syscall, softirq_*, and block_* kernel entry points using the latest 2.0 lttng
on
RHEL 6.3 with a 3.6.6 kernel.
When I 'lttng view' the trace, I get all sorts of messages like:
[warning] Tracer discarded
You'll want to look at:
lttng enable-channel --help
(specifically --subbuf-size and --num-subbuf options).
and you'll need to specify a channel name for the events you enable (see
lttng enable-event --help).
Thanks,
Mathieu
* Sam Bradshaw (sbradshaw) (sbrads...@micron.com) wrote:
Hello,
Now only checks for the major version to be equal. After 2.1 stable
release, both components will adapt to the lowest minor version for the
same major version. For this, the session daemon now send it's version
values to the relayd so slight change in the protocol here.
For instance, a relayd 2.4
* David Goulet (dgou...@efficios.com) wrote:
Now only checks for the major version to be equal. After 2.1 stable
release, both components will adapt to the lowest minor version for the
same major version. For this, the session daemon now send it's version
it's - its
values to the relayd so
This is my simple solution.
Enter the lttng module build dir (like /var/lib/dkms/lttng-modules/2.0.0/build/
as your log says),
open lttng-syscalls.c, comment the definition of is_compat_task(void) (line
32 to 35),
finally
# make
# make modules_install
# depmod -a
I don't know why but it just
The following master branch commit makes this patch unnecessary:
commit dca6e79f31143ffc08f8bd5ef832f820fe4546bf
Author: Mathieu Desnoyers mathieu.desnoy...@efficios.com
Date: Mon Nov 12 15:40:15 2012 -0500
Remove LIBFORMAT config declaration, unused
Remove leftover from UST 0.x.
Hi!
Please help me.
I searched the whole Google and could not just find information on how to add
to my own kernel tracing trace_events to the LTTng trace.
I am using openSUSE distributive 12.1, vanila Linux kernel 3.4.11 + rt patch
and LTTng 2.0.
Could you give me some information on how can
This patch removes some experimental kernel ABI calls that were not
supposed to get merged. The kernel support for these calls has never
been introduced in lttng-modules.
I'm not sure how it happened, but it seems that this code got introduced
along with the commit
To: Bernd Hufmann
Cc: lttng-dev@lists.lttng.org
Subject: Re: [lttng-dev] LTTng Tools 2.1 streaming commands
After talking a bit about this issue with other LTTng devs, it turns out that
it makes more sense to have a set-consumer command and remove
enable/disable-consumer from the cmd UI.
I'll send
Merged!
Thanks
David
Julien Desfossez:
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all the structures exchanged between user and kernel-space are
packed. Also this new ABI moves the int overwrite member of
Hi David
thanks for the detailed explanation.
I'm not debating the purpose of the command or that a 2 step approach is
supported for enabling the consumer. As you described it makes sense to
have this possibility.
However I'm debating mainly the name of the command enable-consumer.
The
will be downloaded on demand.
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
http
This look good to me. There is a small change I would make though to the
open_metadata call where the old_channel is not in the if - else{}
statement but I'll do it before merging it. Don't bother sending back a
version.
Acked.
David
Julien Desfossez:
The current ABI does not work for compat
Hello
For the support of LTTng Tools 2.1 in Eclipse, I'm currently trying to
understand how to use the configuration for network streaming with the updated
lttng create-command and new enable-consumer-command.
a) lttng enable-consumer
I find this command confusing because this command does not
Hi Bernd,
The enable-consumer, by default, always enable a consumer. We added the
extended options such as the -U/-C/-D/-e to control each part of the
API. So let say,
$ lttng enable-consumer -k net://localhost
This command does two API calls underneath which are
lttng_set_consumer_uri and
Merged!
Mathieu Desnoyers:
at commit db8870edf473e2a2f69e488375d32405ea324017
Thanks!
Mathieu
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Hello
I noticed a minor issue when executing the following command sequence:
lttng create auto
Session auto created.
Traces will be written in /home/bernd/lttng-traces/auto-20121002-144207
lttng destroy
Warning: Session name auto not found
The session is sill around after that. I think the
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all the structures exchanged between user and kernel-space are
packed. Also this new ABI moves the int overwrite member of the
struct lttng_kernel_channel to remove the
* Julien Desfossez (jdesfos...@efficios.com) wrote:
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all the structures exchanged between user and kernel-space are
packed. Also this new ABI moves the int overwrite
On 01/10/12 12:15 PM, Mathieu Desnoyers wrote:
* Julien Desfossez (jdesfos...@efficios.com) wrote:
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all the structures exchanged between user and kernel-space are
* Julien Desfossez (jdesfos...@efficios.com) wrote:
On 01/10/12 12:15 PM, Mathieu Desnoyers wrote:
* Julien Desfossez (jdesfos...@efficios.com) wrote:
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Mathieu Desnoyers:
* Julien Desfossez (jdesfos...@efficios.com) wrote:
On 01/10/12 12:15 PM, Mathieu Desnoyers wrote:
* Julien Desfossez (jdesfos...@efficios.com) wrote:
The current ABI does not work for compat 32/64 bits. This
patch
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Julien Desfossez:
+/* + * This flag indicates if lttng-tools must use the
new or the old kernel ABI + * (without compat support for
32/64 bits). It is set by + * kernctl_tracer_version()
Hrm. I don't like that this is only set by
* David Goulet (dgou...@efficios.com) wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Julien Desfossez:
+/* + * This flag indicates if lttng-tools must use the
new or the old kernel ABI + * (without compat support for
32/64 bits). It is set by + * kernctl_tracer_version()
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all the structures exchanged between user and kernel-space are
packed. Also this new ABI moves the int overwrite member of the
struct lttng_kernel_channel to remove the
The current ABI does not work for compat 32/64 bits.
This patch moves the current ABI as old-abi and provides a new ABI in
which all the structures exchanged between user and kernel-space are
packed. Also this new ABI moves the int overwrite member of the
struct lttng_kernel_channel to remove the
* Julien Desfossez (jdesfos...@efficios.com) wrote:
[...]
+#define LTTNG_KERNEL_CHANNEL_PADDING1 LTTNG_SYMBOL_NAME_LEN + 32
+struct lttng_kernel_channel {
+ uint64_t subbuf_size; /* bytes */
+ uint64_t num_subbuf;/* power of 2 */
+ unsigned int
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Julien Desfossez:
The current ABI does not work for compat 32/64 bits. This patch
moves the current ABI as old-abi and provides a new ABI in which
all the structures exchanged between user and kernel-space are
packed. Also this new ABI moves
at commit db8870edf473e2a2f69e488375d32405ea324017
Thanks!
Mathieu
--
Mathieu Desnoyers
Operating System Efficiency RD Consultant
EfficiOS Inc.
http://www.efficios.com
___
lttng-dev mailing list
lttng-dev@lists.lttng.org
* Hollis Blanchard (hollis_blanch...@mentor.com) wrote:
I'm trying to use the net_dev_xmit tracepoint, but it's not showing up
in my traces. It's not listed by 'lttng list -k', but it is present in
debugfs/tracing/available_events. The reporting problem seems to be on
the kernel side,
I'm trying to use the net_dev_xmit tracepoint, but it's not showing up
in my traces. It's not listed by 'lttng list -k', but it is present in
debugfs/tracing/available_events. The reporting problem seems to be on
the kernel side, because kernel_list_events() in lttng-sessiond also
doesn't see
Fixes #324
Signed-off-by: David Goulet dgou...@efficios.com
---
src/bin/lttng-sessiond/consumer.h |8 ++-
src/bin/lttng-sessiond/main.c | 116 +
2 files changed, 96 insertions(+), 28 deletions(-)
diff --git a/src/bin/lttng-sessiond/consumer.h
* David Goulet (dgou...@efficios.com) wrote:
Fixes #324
Signed-off-by: David Goulet dgou...@efficios.com
---
src/bin/lttng-sessiond/consumer.h |8 ++-
src/bin/lttng-sessiond/main.c | 116
+
2 files changed, 96 insertions(+), 28 deletions(-)
Hi David,
On 09/11/2012 08:49 PM, David Goulet wrote:
To prioritize the consumption of the metadata, this patch introduce a
new thread in the consumer which exclusively handles metadata in order
to separate them from the trace data.
[...]
+ /* Check the metadata pipe for
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Hi Gerlando,
I've fix this upstream:
commit 4adabd6161a6decfcd21108b45df6959c34de74c
Author: David Goulet dgou...@efficios.com
Date: Tue Sep 18 11:13:33 2012 -0400
Fix: Remove LPOLLNVAL from consumer metadata revents
Without epoll(7),
For network streaming, with the mmap() mechanism only for now, the
consumer does NOT send the padding over the network. Instead, the size
of the padding is specified in the data header or metadata payload.
The lttng-relayd now is the one appending the zeros to the trace files.
Again, this
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Mathieu Desnoyers:
* David Goulet (dgou...@efficios.com) wrote:
For network streaming, with the mmap() mechanism only for now,
the consumer does NOT send the padding over the network. Instead,
the size of the padding is specified in the data
On 09/10/2012 05:39 PM, Woegerer, Paul wrote:
On 09/10/2012 05:00 PM, David Goulet wrote:
With more options here to set either a custom destination or live
trace. What's the default? (no opts given). I would personally go for
a default one being to record in the session default trace files and
* David Goulet (dgou...@efficios.com) wrote:
To prioritize the consumption of the metadata, this patch introduce a
new thread in the consumer which exclusively handles metadata in order
to separate them from the trace data.
The motivation behind this change is that once a start command is
* David Goulet (dgou...@efficios.com) wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
All fixed!
One comment:
Mathieu Desnoyers:
* David Goulet (dgou...@efficios.com) wrote:
@@ -1025,9 +1007,22 @@ struct lttng_consumer_local_data
*lttng_consumer_create( goto
901 - 1000 of 1160 matches
Mail list logo