* [PATCH] tools/sched_ext: scx_flatcg: fix heap buffer overflow and memory leak in fcg_read_stats
@ 2026-02-14 7:32 David Carlier
2026-02-17 7:04 ` Tejun Heo
0 siblings, 1 reply; 2+ messages in thread
From: David Carlier @ 2026-02-14 7:32 UTC (permalink / raw)
To: Tejun Heo, David Vernet, Andrea Righi; +Cc: linux-kernel, David Carlier
fcg_read_stats() had a VLA allocating 21 * nr_cpus * 8 bytes on the
stack, risking stack overflow on large CPU counts (nr_cpus can be up
to 512).
Fix by using a single heap allocation with the correct size, reusing
it across all stat indices, and freeing it at the end.
Signed-off-by: David Carlier <devnexen@gmail.com>
---
tools/sched_ext/scx_flatcg.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/tools/sched_ext/scx_flatcg.c b/tools/sched_ext/scx_flatcg.c
index bea76d060201..a8446509949e 100644
--- a/tools/sched_ext/scx_flatcg.c
+++ b/tools/sched_ext/scx_flatcg.c
@@ -102,22 +102,27 @@ static float read_cpu_util(__u64 *last_sum, __u64 *last_idle)
static void fcg_read_stats(struct scx_flatcg *skel, __u64 *stats)
{
- __u64 cnts[FCG_NR_STATS][skel->rodata->nr_cpus];
+ __u64 *cnts;
__u32 idx;
+ cnts = calloc(skel->rodata->nr_cpus, sizeof(__u64));
+ if (!cnts)
+ return;
+
memset(stats, 0, sizeof(stats[0]) * FCG_NR_STATS);
- memset(cnts, 0, sizeof(cnts));
for (idx = 0; idx < FCG_NR_STATS; idx++) {
int ret, cpu;
ret = bpf_map_lookup_elem(bpf_map__fd(skel->maps.stats),
- &idx, cnts[idx]);
+ &idx, cnts);
if (ret < 0)
continue;
for (cpu = 0; cpu < skel->rodata->nr_cpus; cpu++)
- stats[idx] += cnts[idx][cpu];
+ stats[idx] += cnts[cpu];
}
+
+ free(cnts);
}
int main(int argc, char **argv)
--
2.51.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-02-17 7:04 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-02-14 7:32 [PATCH] tools/sched_ext: scx_flatcg: fix heap buffer overflow and memory leak in fcg_read_stats David Carlier
2026-02-17 7:04 ` Tejun Heo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox