[lttng-dev] [PATCH lttng-tools 0/5] Support probes with the same name but different event payload

This patch set allows for multiple probes with the same name to be hooked on the same callsite. Right now, the Session Daemon only considers the name and signature of the events to determine if two events are identical. This could lead to trace corruptions when two probes would have the same name

[lttng-dev] [PATCH lttng-tools 4/5] Tests: add function to validate the number of an event name in metadata

Signed-off-by: Francis Deslauriers --- tests/utils/utils.sh | 23 +++ 1 file changed, 23 insertions(+) diff --git a/tests/utils/utils.sh b/tests/utils/utils.sh index 9bf1fcc..60df376 100644 --- a/tests/utils/utils.sh +++

[lttng-dev] [PATCH lttng-tools 5/5] Tests: add duplicated providers tests

Signed-off-by: Francis Deslauriers --- configure.ac| 1 + tests/fast_regression | 1 + tests/regression/ust/multi-lib/Makefile.am | 114 +++ tests/regression/ust/multi-lib/README

[lttng-dev] [PATCH lttng-tools 3/5] Tests: allow the use of regular expressions to match events

Signed-off-by: Francis Deslauriers --- tests/utils/utils.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/utils/utils.sh b/tests/utils/utils.sh index e8dfcda..9bf1fcc 100644 --- a/tests/utils/utils.sh +++ b/tests/utils/utils.sh @@

[lttng-dev] [PATCH lttng-tools 1/5] Fix: probes should not be compared by their names and callsite signatures

Events with different payloads but identical name and signatures could lead to corrupted trace as the Session Daemon would consider them identical and give them the same event ID. Events should be compared using the name, loglevel, fields and model_emf_uri to ensure that their serialized layout

Re: [lttng-dev] [PATCH lttng-tools 1/5] Fix: probes should not be compared by their names and callsite signatures

- On Feb 7, 2018, at 2:36 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: > Events with different payloads but identical name and signatures could > lead to corrupted trace as the Session Daemon would consider them > identical and give them the same event ID. > > Events

Re: [lttng-dev] [PATCH lttng-ust 6/8] Manually dlopen() liblttng-ust.so to prevent unloading

- On Feb 2, 2018, at 2:48 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: > dlopen() increments the refcount of the library thus preventing the > refcount to reach zero in the case of dlclose; The changelog and comment do not explain _why_ this is needed. The scenario is:

Re: [lttng-dev] [PATCH lttng-tools 2/5] Fix: should pass the reg_enum_lookup pointer directly

- On Feb 7, 2018, at 2:36 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: > As the ht_hash_enum and ht_match_enum functions are not changing the > pointer there is no need to pass the address of the pointer. The changelog seems too nice. The current situation is that the

Re: [lttng-dev] [PATCH lttng-ust 4/8] Add probe provider unregister function

- On Feb 2, 2018, at 2:48 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: > Signed-off-by: Francis Deslauriers > --- > include/lttng/ust-events.h | 1 + > liblttng-ust/lttng-events.c | 89 + > 2

Re: [lttng-dev] [PATCH lttng-ust 5/8] Fix: missing enum removal from the enum hashtable

- On Feb 2, 2018, at 2:48 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: Is it a fix or a preparation step for a new feature ? Thanks, Mathieu > Signed-off-by: Francis Deslauriers > --- > liblttng-ust/lttng-events.c | 1 + > 1 file

Re: [lttng-dev] [PATCH lttng-ust 8/8] Support unloading of probe providers

- On Feb 2, 2018, at 2:48 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: > With this commit, it's now possible to dlclose() a library containing an > actively used probe provider. > > The destructor of such library will now iterate over all the sessions > and over all probe

Re: [lttng-dev] [PATCH lttng-ust 2/8] Fix: missing event removal from the event hashtable

- On Feb 2, 2018, at 2:48 PM, Francis Deslauriers francis.deslauri...@efficios.com wrote: Is this really a fix ? Or is it a preparation step in order to be able to remove events before the end of the session lifetime ? Thanks, Mathieu > Signed-off-by: Francis Deslauriers

[lttng-dev] [RFC PATCH babeltrace-1.5] Propagate error from packet_seek in case of truncated packet

Report the error all the way up allowing users/scripts to perform error detection and act on it. Print to stderr the truncated packet information for easier identification. Introduce bt_packet_seek_error enum for specific error handling. Use the ERANGE errno for error propagation inside