Skip to contentSkip to conversation

tracing: Keep the predicate limit separate from program capacity

STABLE v6.1-v5.10 Lee Jones 2 participants 2 messages Nov 2 – Nov 3

Subsystems (2)
lore
Share & tools
lore

Lee Jones

author
#0 lore raw

From: Taeyang Lee <[email protected]>

[ Upstream commit 803aa6fc2dbea25ab9d3ed19eca1c96d0ab17445 ]

predicate_parse() adds two to nr_preds to reserve the terminal TRUE and FALSE entries. It also uses that increased value as the limit for parsed predicates. If the initial count is smaller than the number actually parsed, the program can fill the allocation and subsequent accesses can go past its end.

Keep nr_preds as the predicate limit and use a separate value for the allocation size. This leaves space for both terminal entries and a zeroed entry when parsing stops at the limit.

Fixes 39f7c41c908b ("tracing/filters: Enable filtering a cpumask field by another cpumask")
[ sashal: Adapted kzalloc_objs() allocation to kcalloc(). ]
---
 kernel/trace/trace_events_filter.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
kernel/trace/trace_events_filter.c +3 -4
index 1c4e81b3fff1..3062ff7ff09d 100644
@@ -459,18 +459,17 @@ predicate_parse(const char *str, int nr_parens, int nr_preds,
 	int len;
 	int N = 0;
 	int i;
-
-	nr_preds += 2; /* For TRUE and FALSE */
+	int nr_preds_alloc = nr_preds + 2; /* For TRUE and FALSE */
 
 	op_stack = kmalloc_array(nr_parens, sizeof(*op_stack), GFP_KERNEL);
 	if (!op_stack)
 		return ERR_PTR(-ENOMEM);
-	prog_stack = kcalloc(nr_preds, sizeof(*prog_stack), GFP_KERNEL);
+	prog_stack = kcalloc(nr_preds_alloc, sizeof(*prog_stack), GFP_KERNEL);
 	if (!prog_stack) {
 		parse_error(pe, -ENOMEM, 0);
 		goto out_free;
 	}
-	inverts = kmalloc_array(nr_preds, sizeof(*inverts), GFP_KERNEL);
+	inverts = kmalloc_array(nr_preds_alloc, sizeof(*inverts), GFP_KERNEL);
 	if (!inverts) {
 		parse_error(pe, -ENOMEM, 0);
 		goto out_free;
Signature · 1 line
2.56.0.317.g454f8302e1-goog
maintainer
#1 lore raw
Quoted 1 line
Fixes 39f7c41c908b ("tracing/filters: Enable filtering a cpumask field by another cpumask")

Not queueing this for 6.1, 5.15 or 5.10. The overflow needs 39f7c41c908b ("tracing/filters: Enable filtering a cpumask field by another cpumask"), which landed in v6.6 and is absent from those trees. There calc_stack() and predicate_parse() agree on the predicate count, so there is nothing to overflow.

Signature · 2 lines
Thanks,
Sasha