[PATCH,net-next,v2] af_unix: Fix wrong socket tested in the symmetric half of unix_state_lock_cmp_fn().

[lirongqing <lirongqing@baidu.com>]

From: Li RongQing <lirongqing@baidu.com>

The second half of unix_state_lock_cmp_fn() mirrors the first with a and b
swapped, but it tests b->sk.sk_state again.  Since b is already known to be
TCP_LISTEN, the switch only hits default and always returns 0 ("no order
known") instead of 1.  Test a's state instead, as the first half does.

This code is only built with CONFIG_PROVE_LOCKING, so the impact is limited
to lockdep failing to validate this unix_state_lock() nesting order.

Found by AI.

Signed-off-by: Li RongQing <lirongqing@baidu.com>
---
Diff with v1: target to net-next, and drop fix tag

 net/unix/af_unix.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index e7a0582..501e693 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -163,7 +163,7 @@ static int unix_state_lock_cmp_fn(const struct lockdep_map *_a,
 
 	/* Should never happen.  Just to be symmetric. */
 	if (b->sk.sk_state == TCP_LISTEN) {
-		switch (b->sk.sk_state) {
+		switch (a->sk.sk_state) {
 		case TCP_CLOSE:
 		case TCP_ESTABLISHED:
 			return 1;
-- 
2.9.4