Re: [lttng-dev] [PATCH lttng-ust] Fix: Initialize fd field of struct lttng_ust_elf to -1 at allocation

It slipped through. I'll merge it right away. Thanks for the reminder! Mathieu - On Jan 22, 2019, at 10:41 AM, Jonathan Rajotte jonathan.rajotte-jul...@efficios.com wrote: > Hi Mathieu, > > Any feedback regarding this patch? > > Cheers > > On Wed, Nov 14, 2018 at 02:38:37PM -0500,

[lttng-dev] [RELEASE] LTTng-Tools 2.9.11 - Joannès

Hi all, The eleventh bug-fix release of LTTng-Tools 2.9 - Joannès has just been released. This release addresses a number of issues listed below. 2019-01-22 lttng-tools 2.9.11 (Blonde Brownie Day) * Fix: worker structure is leaked in run_as process * Fix: leak of filter bytecode

[lttng-dev] [RELEASE] LTTng-Tools 2.10.6 - KeKriek

Hi all, The sixth bug-fix release of LTTng-Tools 2.10 - KeKriek has just been released. This release addresses a number of issues listed below. 2019-01-22 lttng-tools 2.10.6 (Blonde Brownie Day) * Fix: worker structure is leaked in run_as process * Fix: only free trace_path when

Re: [lttng-dev] [PATCH lttng-tools] Fix: notification thread: RCU-safe reclaim of hash table nodes

Merged in master and stable-2.11. Thanks, Jérémie On Wed, 12 Dec 2018 at 12:16, Mathieu Desnoyers wrote: > > Nodes that are put in a rculfhash hash table created with the > "auto resize" flag need to beware that a worker thread can access the > hash table nodes as a RCU reader concurrently, and

Re: [lttng-dev] [PATCH lttng-tools] Fix: do not repurpose iterator while it is being used

Merged in master and stable-2.11. Thanks, Jérémie On Wed, 12 Dec 2018 at 17:24, Mathieu Desnoyers wrote: > > The hash table iteration uses an iterator that needs to stay valid for > the next loop. Using that same iterator variable in a nested lookup > in a different hash table leads to

Re: [lttng-dev] [PATCH lttng-tools] Fix: run_command_wait() handle partial write

Merged in master, stable-2.11, and stable-2.10. Thanks, Jérémie On Thu, 13 Dec 2018 at 13:56, Mathieu Desnoyers wrote: > > Use lttng_write() to handle partial writes (writing less than the > requested amount of bytes) as well as ret = -1, errno = EINTR. > > Signed-off-by: Mathieu Desnoyers >

Re: [lttng-dev] [PATCH lttng-tools 2.10] Fix: do not repurpose iterator while it is being used (register trigger, 2.10)

Merged in stable-2.10. Thanks, Jérémie On Wed, 12 Dec 2018 at 17:37, Mathieu Desnoyers wrote: > > The hash table iteration uses an iterator that needs to stay valid for > the next loop. Using that same iterator variable in a nested lookup in a > different hash table leads to segmentation fault.

Re: [lttng-dev] [PATCH lttng-tools 2/2] Fix: handle_notification_thread_command: handle partial read

Note that this patch was merged in stable-2.10 too. Jérémie On Wed, 12 Dec 2018 at 15:11, Mathieu Desnoyers wrote: > > Use lttng_read() to handle partial reads (returning less than the > requested amount of bytes) as well as ret = -1, errno == EINTR. > > Signed-off-by: Mathieu Desnoyers > ---

Re: [lttng-dev] [PATCH lttng-tools 1/2] Fix: notification thread: free session trigger list on error

Both patches were merged in master and stable-2.11. Thanks! Jérémie On Wed, 12 Dec 2018 at 15:11, Mathieu Desnoyers wrote: > > Signed-off-by: Mathieu Desnoyers > --- > src/bin/lttng-sessiond/notification-thread-events.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git

Re: [lttng-dev] [PATCH lttng-tools] Fix: warning 'fd' may be used uninitialized

Merged in master, stable-2.11, stable-2.10, and stable-2.9. Thanks! Jérémie On Thu, 29 Nov 2018 at 16:49, Michael Jeanson wrote: > > Initialize fd to invalid '-1' and remove unnecessary file_opened. > > Signed-off-by: Michael Jeanson > --- > src/bin/lttng-sessiond/save.c | 8 +++- > 1

Re: [lttng-dev] [PATCH lttng-tools 1/2] Revert stubbing of unix socket functions

Both patches of the series were merged in master and stable-2.11. Thanks! Jérémie On Thu, 29 Nov 2018 at 16:22, Michael Jeanson wrote: > > Instead of stubing useful unix socket functions to work around > Linux-only credential passing, ifdef the relevant parts like it was > already done for

Re: [lttng-dev] [PATCH lttng-tools] Add *.exe to gitignore for Cygwin

Merged in master, stable-2.11, stable-2.10, and stable-2.9. Thanks! Jérémie On Tue, 27 Nov 2018 at 14:24, Michael Jeanson wrote: > > Signed-off-by: Michael Jeanson > --- > .gitignore | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/.gitignore b/.gitignore > index c25fb700..19276012

Re: [lttng-dev] [PATCH lttng-tools] Fix: use sys/types.h for ssize_t on Cygwin

Merged in master and stable-2.11. Thanks! Jérémie On Tue, 27 Nov 2018 at 14:24, Michael Jeanson wrote: > > Signed-off-by: Michael Jeanson > --- > include/lttng/location-internal.h | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/include/lttng/location-internal.h >

Re: [lttng-dev] "Hands-free" tracepoints using LD_PRELOAD

Hi Brian, On Tue, Jan 22, 2019 at 01:30:23PM -0500, Brian Rossa wrote: >4. Boilerplate that does the typical `log(...); auto return_val = >dlsym(...); log(...); return return_val;` gets generated. As proposed by Francis, this is when you need to "generate" a corresponding tracepoint

Re: [lttng-dev] "Hands-free" tracepoints using LD_PRELOAD

Francis, Thanks for your detailed, helpful reply. I was especially excited to learn about the progress on `--userspace-probe`, but unfortunately I *do* need to extract the function arguments. It would seem, then, that interposition through LD_PRELOAD is really the only option. The decision to

Re: [lttng-dev] "Hands-free" tracepoints using LD_PRELOAD

Hi Brian, If I understand correctly, you want developers to provide a whitelist of functions that they want to instrument in an arbitrary shared library, and then have a system that instruments those functions without having to recompile the library with LTTng-UST tracepoints. Can you give us an

Re: [lttng-dev] add cpu_id context to the channel

Thank you Genevieve, "cpu" field works. Yonghong On Mon, Jan 21, 2019 at 7:57 PM Genevieve Bastien wrote: > Hi Yonghong, > > With TraceCompass XML analyses, you can use as state value or attribute an > eventField called 'cpu' to get the CPU id of an event. For example, > > > > With LTTng

Re: [lttng-dev] [PATCH lttng-ust] Fix: Initialize fd field of struct lttng_ust_elf to -1 at allocation

Hi Mathieu, Any feedback regarding this patch? Cheers On Wed, Nov 14, 2018 at 02:38:37PM -0500, Jonathan Rajotte wrote: > In rare cases when the executable cannot be open, using a default value > of zero lead to invalid close call and fd tracker removal. > > fixes #1171 > > Reported-by: