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.
This is a 2.10-specific instance of the issue affecting
bind_trigger_to_matching_channels() which was found in the
master and 2.11 branches. This one affects
handle_notification_thread_command_register_trigger().
Signed-off-by: Mathieu Desnoyers <mathieu.desnoy...@efficios.com>
---
src/bin/lttng-sessiond/notification-thread-events.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/bin/lttng-sessiond/notification-thread-events.c
b/src/bin/lttng-sessiond/notification-thread-events.c
index 6777fd6d..f54141e7 100644
--- a/src/bin/lttng-sessiond/notification-thread-events.c
+++ b/src/bin/lttng-sessiond/notification-thread-events.c
@@ -1119,6 +1119,7 @@ int handle_notification_thread_command_register_trigger(
channels_ht_node) {
struct lttng_trigger_list_element *trigger_list_element;
struct lttng_channel_trigger_list *trigger_list;
+ struct cds_lfht_iter lookup_iter;
if (!trigger_applies_to_channel(trigger, channel)) {
continue;
@@ -1128,8 +1129,8 @@ int handle_notification_thread_command_register_trigger(
hash_channel_key(&channel->key),
match_channel_trigger_list,
&channel->key,
- &iter);
- node = cds_lfht_iter_get_node(&iter);
+ &lookup_iter);
+ node = cds_lfht_iter_get_node(&lookup_iter);
assert(node);
trigger_list = caa_container_of(node,
struct lttng_channel_trigger_list,
--
2.11.0
_______________________________________________
lttng-dev mailing list
lttng-dev@lists.lttng.org
https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
