Franck Balazot | 6 Mar 16:31 2015

Using same provider:tracepoint probe in several processes

Hi all,

I would like to know if it is possible to use the same 
provider:tracepoint probe in several processes, without getting 
following error :

LTTng-UST: Error (-17) while registering tracepoint probe. Duplicate 
registration of tracepoint probes having the same name is not allowed.
Aborted (core dumped)

And how to do it ?

Franck
Chidhu R | 6 Mar 01:01 2015
Picon

Performance comparison of LTTng in Linux kernel 3.14 and 3.15

Hello Team,

I am studying the performance of LTTng, by running a standalone program doing LTTng traces, in a machine running 3.14 LK and 3.15 LK. 

1. I wanted to know if there is any data published already regarding any performance tests for LTTng

2. Is there any standard benchmark program which can be used to test LTTng, instead of writing a standalone program.

Thanks
Chidambaram
Active LLTng user
_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Brian Robbins | 5 Mar 23:39 2015
Picon

Userspace Tracing and Backtraces

Hello,

 

I’m looking into using the userspace tracing capabilities of LTTng, and I wanted to find out if it is possible to capture a stack backtrace when a userspace tracepoint is hit.

 

I’ve done a bunch of searching, but have not yet been able to find the answer.

 

Thank you.

-Brian

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Neil Bryan | 5 Mar 11:50 2015

LTTng 2.6.0: Corrupt metadata?

Hello LTTng devs,

 

I am experiencing a strange problem with what I believe is corrupt metadata. This is seen on v2.6.0 of LTTng.

 

If I try and parse recovered traces using Babeltrace, it fails with a segmentation fault.

 

nbryan <at> meteorubuntu-OptiPlex-7010:/data/nbryan/Meteor/altera_trace$ babeltrace --help

BabelTrace Trace Viewer and Converter 1.0.0-rc1

 

nbryan <at> meteorubuntu-OptiPlex-7010:/data/nbryan/Meteor/altera_trace$ babeltrace auto-20150304-091109/

Segmentation fault (core dumped)

 

I thought it may be worth a look at the metadata file and I observed the following (these are just three examples):

 

Environment:

env {

     hostname = "socfpga_cyclone5";

     domain = "kernel";

     sysname = "Linux";

     kernel_release = "3.10.31-ltsi-05035-g801a40f";

     kernel_version = "#3 SMP Tue Mar 3 17:31:45 GMT 2015";

     tracer_name = "lttng-modules";

     tracer_major = 2;

     tracer_minor = 6;

     tracer_patchlevel = 0;

};

 

event {

     name = "syscall_exit_recvmmsg";

     id = 921;

     stream_id = 0;

     fields := struct {

           integer { size = 32; align = 32; signed = 1; encoding = none; base = 10; } _ret;

           integer { size = 32; align = 32; signe   WСu?­©)2ѕИN№d™Џ‰n®    и   Ђ     d = 0; encoding = none; base = 16; } _mmsg;

           integer { size = 32; align = 32; signed = 0; encoding = none; base = 16; } _timeout;

     };

};

 

Decoded binary values:

00h,00h,00h,57h,1Dh,D1h,75h,3Fh,ADh,A9h,29h,32h,BEh,C8h,4Eh,B9h,64h,99h,8Fh,14h,89h,6Eh,AEh,00h,00h,00h,00h,E8h,7Fh,00h,00h,00h,80h,00h,00h,

00h,00h,00h,01h,08h

 

event {

     name = "syscall_exit_getcpu";

     id = 903;

     stream_id = 0;

     fields := struct {

           integer { size = 32; align = 32; signed = 1; encoding = none; base = 10; } _ret;

           integer { size = 32; align = 32; signed = 0; encoding = none; base = 16; } _cpup;

           integer { size = 32; align = 32; signed = 0; encoding = none; base = 16; } _nodep;

           integer { size = 32; align = 32; signed = 0; encoding = none; base =   WСu?­©)2ѕИN№d™Џ‰n®    и   Ђ     16; } _tcache;

     };

};

 

Decoded binary values:

00h,00h,00h,57h,1Dh,D1h,75h,3Fh,ADh,A9h,29h,32h,BEh,C8h,4Eh,B9h,64h,99h,8Fh,14h,89h,6Eh,AEh,00h,00h,00h,00h,E8h,7Fh,00h,00h,00h,80h,00h,00h,00h,

00h,00h,01h,08h

 

event {

     name = "syscall_exit_fstatat64";

     id = 886;

     stream_id = 0;

     fields := struct {

           integer { size = 32; align = 32; signed = 1; encoding = none; base = 10; } _ret;

           integer { size = 32; align = 32; signed = 1; encod   WСu?­©)2ѕИN№d™Џ‰n®    и   Ђ     ing = none; base = 10; } _dfd;

           string _filename;

           integer { size = 32; align = 32; signed = 0; encoding = none; base = 16; } _statbuf;

           integer { size = 32; align = 32; signed = 1; encoding = none; base = 10; } _flag;

     };

};

 

Decoded binary values:

00h,00h,00h,57h,1Dh,D1h,75h,3Fh,ADh,A9h,29h,32h,BEh,C8h,4Eh,B9h,64h,99h,8Fh,14h,89h,6Eh,AEh,00h,00h,00h,00h,E8h,7Fh,00h,00h,00h,80h,00h,00h,

00h,00h,00h,01h,08h

 

In a metadata file containing 7200 lines, I see this corruption 74 times.

 

I also notice that the header of the metadata file contains something very similar:

 

57h,1Dh,D1h,75h,3Fh,ADh,A9h,29h,32h,BEh,C8h,4Eh,B9h,64h,99h,8Fh,14h,89h,6Eh,AEh,00h,00h,00h,00h,D8h,27h,00h,00h,00h,80h,00h,00h,00h,00h,00h,01h,08h

 

This appears identical to the other instances, less the first three 00h bytes.

 

Now it may be intentional to squirt binary data into what is essentially  a text file, but it looks suspicious. Can anyone shed any light on what is happening here?

Is the metadata file supposed to have any binary data in it at all?

 

Thanks,

 

Neil.

 

 

 

 

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Anand Neeli | 4 Mar 20:13 2015
Picon

tracing is stalled

Hi All,

I'm using 2.6 and tracing using relayd. Very frequently i see that the tracing is stalled and doesnt trace anything, i dont see the trace file size increasing and i dont see any console errors.
Can anyone tell me what debug-traces will be needed to debug this. I will collect those debug-traces next time i see the issue.

Thanks,
Anand Neeli
_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Jeff Layton | 3 Mar 21:40 2015
Picon

URCU and pthread mutex/cond variable hang

I've been using urcu to develop some userland code, and I've run into a
problem that I don't quite understand. If I register a thread and then
have that thread block on a pthread condition variable, then it seems
to cause synchronize_rcu in another thread to hang.

I've attached a testcase that demonstrates the problem. You have to
build it and link it against -lpthread and -lurcu-qsbr:

    $ gcc -Wall -o ./urcu_hang -lpthread -lurcu-qsbr urcu_hang.c

...it will run fine. If you comment out the rcu_thread_offline/online
calls though, it will hang.

Why? Is this expected behavior or a bug in urcu?

Thanks in advance!
--

-- 
Jeff Layton <jlayton <at> poochiereds.net>
Attachment (urcu_hang.c): text/x-c++src, 1262 bytes
_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Divya Vyas | 3 Mar 08:10 2015

LTTng timestamp from NTP ( Network transfer protocol)

Hi,

I am looking at synchronization of traces from guest and host according to timestamps.

Does LTTng uses timestamps generated from NTP (Network transfer protocol) ?
How I can use NTP to synchronize the timestamps of guest and host ?

Thanks,
Divya Vyas
_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Anand Neeli | 2 Mar 20:51 2015
Picon

babeltrace lttng-live format option

Hi All,
I'm using BabelTrace Trace Viewer and Converter 1.2.4
and i see the available formats are ctf, text.

node-a # babeltrace -l           
Formats available: ctf, text.

How do i get lttng-live support? i read somewhere that "babeltrace -i lttng-live"  can be used to read live tracing data, but looks like this option is not available in 1.2.4
am i missing anything here?

Thanks,
Anand Neeli

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Anand Neeli | 2 Mar 14:02 2015
Picon

destroying one session stops tracing on all other sessions

Hi All,
I'm using 2.6.0 and see following issue: 
On a multisession setup with relayd, Destroying  one session is showing errors on the console and tracing of all the sessions is stopped.

steps to recreate are as follows
1) Create multiple sessions with relayd. (in below logs i have created 2 sessions)
2) Destroy one session and then tracing on all the sessions stops

(Have not check this with single session, could be happening with single session also)

logs
---------
node-a # lttng list
Available tracing sessions:
  1) mys5 (tcp4://128.0.0.4:5342/ [data: 5343]) [active]
    Trace path: tcp4://128.0.0.4:5342/ [data: 5343]
    Live timer interval (usec): 2000000

  2) mysession (tcp4://128.0.0.4:5342/ [data: 5343]) [active]
    Trace path: tcp4://128.0.0.4:5342/ [data: 5343]
    Live timer interval (usec): 2000000

Use lttng list <session_name> for more details


node-a # lttng list mys5
Tracing session mys5: [active]
    Trace path: tcp4://128.0.0.4:5342/ [data: 5343]

=== Domain: UST global ===

Buffer type: per PID

Channels:
-------------
- myc5: [enabled]

    Attributes:
      overwrite mode: 0
      subbufers size: 4096
      number of subbufers: 4
      switch timer interval: 0
      read timer interval: 0
      trace file count: 2
      trace file size (bytes): 2000000
      output: mmap()

    Events:
      * (type: tracepoint) [enabled] [has exclusions]

node-a # lttng list mysession
Tracing session mysession: [active]
    Trace path: tcp4://128.0.0.4:5342/ [data: 5343]

=== Domain: UST global ===

Buffer type: per PID

Channels:
-------------
- mychannel: [enabled]

    Attributes:
      overwrite mode: 0
      subbufers size: 4096
      number of subbufers: 4
      switch timer interval: 0
      read timer interval: 0
      trace file count: 2
      trace file size (bytes): 2000000
      output: mmap()

    Events:
      * (type: tracepoint) [enabled] [has exclusions]


node-a # lttng destroy mys5    <<<<<<<<<<<<<<<<<<<< destroying session here
Error: Pushing metadata
Error: Handling metadata request
Error: Health error occurred in thread_manage_consumer
Error: Pushing metadata
Error: Handling metadata request
Error: Health error occurred in thread_manage_consumer
PERROR - 04:36:15.258880 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
PERROR - 04:36:15.259520 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.260076 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
Session mys5 destroyed
node-a # PERROR - 04:36:15.106369 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106410 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106437 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106474 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106490 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106500 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106509 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106519 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106528 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106537 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106546 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106555 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106564 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106573 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106581 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106590 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106599 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106608 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106616 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106625 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106636 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106645 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106654 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106663 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106672 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106681 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106690 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106699 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106708 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond
PERROR - 04:36:15.106717 [1651/1656]: sendmsg: Broken pipe (in lttcomm_send_unix_sock() at unix.c:218)
Error: Asking metadata to sessiond

node-a # ps aux | grep lttng
root      1622  0.1  0.0 495020  1160 ?        Ssl  04:34   0:00 lttng-relayd -o /var/tmp/lttng-traces -d
root      1631  0.0  0.1 854020  6360 ?        Ssl  04:34   0:00 lttng-sessiond --consumerd32-path /usr/lib/lttng/libexec/lttng-consumerd --consumerd32-libdir /usr/lib/ --consumerd64-path /usr/lib64/lttng/libexec/lttng-consumerd --consumerd64-libdir /usr/lib64/ -b --no-kernel
root      1643  0.0  0.0 546356  3424 ?        Sl   04:34   0:00 lttng-consumerd  -u --consumerd-cmd-sock /var/run/lttng/ustconsumerd64/command --consumerd-err-sock /var/run/lttng/ustconsumerd64/error --group tracing
root      1651  0.0  0.0  64068  1852 ?        Sl   04:34   0:00 lttng-consumerd  -u --consumerd-cmd-sock /var/run/lttng/ustconsumerd32/command --consumerd-err-sock /var/run/lttng/ustconsumerd32/error --group tracing
root      2160  0.0  0.0   4448   704 pts/0    S+   04:36   0:00 grep lttng



Thanks,
Anand Neeli


_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
André Goddard Rosa | 2 Mar 04:18 2015

[PATCH] Set appropriate thread names for processes loading UST lib

When a process loads the UST library, its threads are not named appropriately.

Set appropriate individual threads names so that it becomes easier to
distinguish between UST threads from those coming from the process itself.

It makes for easier understanding of profiling data (CTF on trace compass
or perf data) where cycles are accounted to individual threads, allowing
to distinguish the work accomplished by different threads.

Not tested on __FreeBSD__ environment. Could somebody confirm?

Signed-off-by: André Goddard Rosa <andre.goddard <at> gmail.com>

diff --git a/liblttng-ust/compat.h b/liblttng-ust/compat.h
index 43b2223..657703e 100644
--- a/liblttng-ust/compat.h
+++ b/liblttng-ust/compat.h
 <at>  <at>  -20,7 +20,7  <at>  <at> 
  */

 /*
- * lttng_ust_getprocname.
+ * lttng_ust_getprocname, lttng_ust_setprocname.
  */
 #ifdef __linux__

 <at>  <at>  -34,9 +34,16  <at>  <at>  void lttng_ust_getprocname(char *name)
 	(void) prctl(PR_GET_NAME, (unsigned long) name, 0, 0, 0);
 }

+static inline
+void lttng_ust_setprocname(char *name)
+{
+	(void) prctl(PR_SET_NAME, (unsigned long) name, 0, 0, 0);
+}
+
 #elif defined(__FreeBSD__)
 #include <stdlib.h>
 #include <string.h>
+#include <pthread.h>

 /*
  * Limit imposed by Linux UST-sessiond ABI.
 <at>  <at>  -59,6 +66,12  <at>  <at>  void lttng_ust_getprocname(char *name)
 		strncpy(name, bsd_name, LTTNG_UST_PROCNAME_LEN - 1);
 }

+static inline
+void lttng_ust_setprocname(char *name)
+{
+	(void) pthread_set_name_np(pthread_self(), name);
+}
+
 #endif

 #include <errno.h>
diff --git a/liblttng-ust/lttng-ust-comm.c b/liblttng-ust/lttng-ust-comm.c
index 378ca21..256752b 100644
--- a/liblttng-ust/lttng-ust-comm.c
+++ b/liblttng-ust/lttng-ust-comm.c
 <at>  <at>  -1173,6 +1173,8  <at>  <at>  void *ust_listener_thread(void *arg)
 	int sock, ret, prev_connect_failed = 0, has_waited = 0;
 	long timeout;

+	lttng_ust_setprocname("UST Listener");
+
 	/* Restart trying to connect to the session daemon */
 restart:
 	if (prev_connect_failed) {
diff --git a/libringbuffer/ring_buffer_frontend.c b/libringbuffer/ring_buffer_frontend.c
index aeb7f7d..675abb6 100644
--- a/libringbuffer/ring_buffer_frontend.c
+++ b/libringbuffer/ring_buffer_frontend.c
 <at>  <at>  -398,6 +398,8  <at>  <at>  void *sig_thread(void *arg)
 	siginfo_t info;
 	int signr;

+	lttng_ust_setprocname("UST Sig");
+
 	/* Only self thread will receive signal mask. */
 	rb_setmask(&mask);
 	CMM_STORE_SHARED(timer_signal.tid, pthread_self());
--

-- 
2.3.0.149.gf3f4077.dirty

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
André Goddard Rosa | 2 Mar 04:16 2015

[PATCH] Set appropriate thread names for processes loading UST lib

When a process loads UST library, set appropriate individual threads names
so that it becomes easier to distinguish between its own threads from those
coming from UST.

That makes for easier understanding of profiling data where cycles are
accounted to individual threads, allowing to distinguish the originating
libraries and functions for the gathered profiling data.

Caveat: not tested on __FreeBSD__ environment. Does somebody already has
FreeBSD environment setup to try?

Signed-off-by: André Goddard Rosa <andre.goddard <at> gmail.com>

diff --git a/liblttng-ust/compat.h b/liblttng-ust/compat.h
index 43b2223..657703e 100644
--- a/liblttng-ust/compat.h
+++ b/liblttng-ust/compat.h
 <at>  <at>  -20,7 +20,7  <at>  <at> 
  */

 /*
- * lttng_ust_getprocname.
+ * lttng_ust_getprocname, lttng_ust_setprocname.
  */
 #ifdef __linux__

 <at>  <at>  -34,9 +34,16  <at>  <at>  void lttng_ust_getprocname(char *name)
 	(void) prctl(PR_GET_NAME, (unsigned long) name, 0, 0, 0);
 }

+static inline
+void lttng_ust_setprocname(char *name)
+{
+	(void) prctl(PR_SET_NAME, (unsigned long) name, 0, 0, 0);
+}
+
 #elif defined(__FreeBSD__)
 #include <stdlib.h>
 #include <string.h>
+#include <pthread.h>

 /*
  * Limit imposed by Linux UST-sessiond ABI.
 <at>  <at>  -59,6 +66,12  <at>  <at>  void lttng_ust_getprocname(char *name)
 		strncpy(name, bsd_name, LTTNG_UST_PROCNAME_LEN - 1);
 }

+static inline
+void lttng_ust_setprocname(char *name)
+{
+	(void) pthread_set_name_np(pthread_self(), name);
+}
+
 #endif

 #include <errno.h>
diff --git a/liblttng-ust/lttng-ust-comm.c b/liblttng-ust/lttng-ust-comm.c
index 378ca21..256752b 100644
--- a/liblttng-ust/lttng-ust-comm.c
+++ b/liblttng-ust/lttng-ust-comm.c
 <at>  <at>  -1173,6 +1173,8  <at>  <at>  void *ust_listener_thread(void *arg)
 	int sock, ret, prev_connect_failed = 0, has_waited = 0;
 	long timeout;

+	lttng_ust_setprocname("UST Listener");
+
 	/* Restart trying to connect to the session daemon */
 restart:
 	if (prev_connect_failed) {
diff --git a/libringbuffer/ring_buffer_frontend.c b/libringbuffer/ring_buffer_frontend.c
index aeb7f7d..675abb6 100644
--- a/libringbuffer/ring_buffer_frontend.c
+++ b/libringbuffer/ring_buffer_frontend.c
 <at>  <at>  -398,6 +398,8  <at>  <at>  void *sig_thread(void *arg)
 	siginfo_t info;
 	int signr;

+	lttng_ust_setprocname("UST Sig");
+
 	/* Only self thread will receive signal mask. */
 	rb_setmask(&mask);
 	CMM_STORE_SHARED(timer_signal.tid, pthread_self());
--

-- 
2.3.0.rc1

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

Gmane