If a BPF scheduler provides an invalid CPU (outside the nr_cpu_ids
range) as prev_cpu to scx_bpf_select_cpu_dfl() it can cause a kernel
crash.
To prevent this, validate prev_cpu in scx_bpf_select_cpu_dfl() and
trigger an scx error if an invalid CPU is specified.
Fixes: f0e1a0643a59b ("sched_ext: Implement BPF extensible scheduler class")
Cc: stable@vger.kernel.org # v6.12+
Signed-off-by: Andrea Righi <arighi@nvidia.com>
---
kernel/sched/ext.c | 3 +++
1 file changed, 3 insertions(+)
ChangeLog v1 -> v2:
- return prev_cpu instead of trying to find a valid cpu (since the
scheduler is exiting anyway)
diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index 0f1da199cfc7c..7b9dfee858e79 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -6422,6 +6422,9 @@ static bool check_builtin_idle_enabled(void)
__bpf_kfunc s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
u64 wake_flags, bool *is_idle)
{
+ if (!ops_cpu_valid(prev_cpu, NULL))
+ goto prev_cpu;
+
if (!check_builtin_idle_enabled())
goto prev_cpu;
--
2.48.1