bpf: Adjust return value for queue destruction in rqspinlock

Return -ETIMEDOUT whenever non-head waiters are signalled by head, and fix
oversight in commit 7bd6e5c ("rqspinlock: Disable queue destruction for
deadlocks"). We no longer signal on deadlocks.

Signed-off-by: Kumar Kartikeya Dwivedi <[email protected]>
Reviewed-by: Amery Hung <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>

Author: kkdwvd

kernel/bpf/rqspinlock.c

@@ -537,7 +537,7 @@ int __lockfunc resilient_queued_spin_lock_slowpath(rqspinlock_t *lock, u32 val)
 
 		val = arch_mcs_spin_lock_contended(&node->locked);
 		if (val == RES_TIMEOUT_VAL) {
-			ret = -EDEADLK;
+			ret = -ETIMEDOUT;
 			goto waitq_timeout;
 		}