[tip: sched/core] sched/topology: Fix sched_numa_find_nth_cpu() in CPU-less case

[tip-bot2 for Yury Norov]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     617f2c38cb5ce60226042081c09e2ee3a90d03f8
Gitweb:        https://git.kernel.org/tip/617f2c38cb5ce60226042081c09e2ee3a90d03f8
Author:        Yury Norov <yury.norov@xxxxxxxxx>
AuthorDate:    Sat, 19 Aug 2023 07:12:35 -07:00
Committer:     Ingo Molnar <mingo@xxxxxxxxxx>
CommitterDate: Fri, 15 Sep 2023 13:48:10 +02:00

sched/topology: Fix sched_numa_find_nth_cpu() in CPU-less case

When the node provided by user is CPU-less, corresponding record in
sched_domains_numa_masks is not set. Trying to dereference it in the
following code leads to kernel crash.

To avoid it, start searching from the nearest node with CPUs.

Fixes: cd7f55359c90 ("sched: add sched_numa_find_nth_cpu()")
Reported-by: Yicong Yang <yangyicong@xxxxxxxxxxxxx>
Reported-by: Guenter Roeck <linux@xxxxxxxxxxxx>
Signed-off-by: Yury Norov <yury.norov@xxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Reviewed-by: Yicong Yang <yangyicong@xxxxxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Link: https://lore.kernel.org/r/20230819141239.287290-4-yury.norov@xxxxxxxxx

Closes: https://lore.kernel.org/lkml/CAAH8bW8C5humYnfpW3y5ypwx0E-09A3QxFE1JFzR66v+mO4XfA@xxxxxxxxxxxxxx/T/
Closes: https://lore.kernel.org/lkml/ZMHSNQfv39HN068m@yury-ThinkPad/T/#mf6431cb0b7f6f05193c41adeee444bc95bf2b1c4
---
 kernel/sched/topology.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index 05a5bc6..423d089 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -2122,12 +2122,16 @@ static int hop_cmp(const void *a, const void *b)
  */
 int sched_numa_find_nth_cpu(const struct cpumask *cpus, int cpu, int node)
 {
-	struct __cmp_key k = { .cpus = cpus, .node = node, .cpu = cpu };
+	struct __cmp_key k = { .cpus = cpus, .cpu = cpu };
 	struct cpumask ***hop_masks;
 	int hop, ret = nr_cpu_ids;
 
 	rcu_read_lock();
 
+	/* CPU-less node entries are uninitialized in sched_domains_numa_masks */
+	node = numa_nearest_node(node, N_CPU);
+	k.node = node;
+
 	k.masks = rcu_dereference(sched_domains_numa_masks);
 	if (!k.masks)
 		goto unlock;