rv_is_container_monitor() returns false for the last list entry
before checking its enable callback. With CONFIG_RV_MON_DEADLINE=y
and CONFIG_RV_MON_NOMISS=n, the empty deadline container can be
last. Enabling it then calls a NULL callback through
rv_enable_single(), causing a kernel oops.
An empty non-tail container already enables successfully: it is
marked enabled without activating any child monitors. Preserve this
behavior regardless of list position by checking the tail entry's
enable callback. Keep the tail check before list_next_entry() to
avoid accessing the list head as a monitor.
Fixes: 8d7861ac507d ("rv: Fix out-of-bound memory access in rv_is_container_monitor()")
Signed-off-by: Avi Weiss <redacted>
---
kernel/trace/rv/rv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/trace/rv/rv.c b/kernel/trace/rv/rv.c
index 29f155c6968b..a142e877e940 100644
--- a/kernel/trace/rv/rv.c
+++ b/kernel/trace/rv/rv.c
@@ -222,7 +222,7 @@ bool rv_is_container_monitor(struct rv_monitor *mon)
struct rv_monitor *next;
if (list_is_last(&mon->list, &rv_monitors_list))
- return false;
+ return !mon->enable;
next = list_next_entry(mon, list);
base-commit: 6628e7d4c6b25c8d18f1345e3a20d21829aae7e4
--
2.53.0