I have been working on a small home-project in bluez, and I think I might have
stumbled on a couple of bugs. I am not sure they are correct so bear with me if
I am mistaken.
The first one seems to be a race condition. I have not been able to verify if
the BT_CLOSED state is still relevant, but if it is, then it seems we should
hold the lock till after the unlink.
Signed-off-by: Carlos Luiz Pena <carlos.luiz.pena@xxxxxxxxx>
---
diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c
index 8cfb5a8..206fa4e 100644
--- a/net/bluetooth/af_bluetooth.c
+++ b/net/bluetooth/af_bluetooth.c
@@ -206,8 +206,8 @@ struct sock *bt_accept_dequeue(struct sock
*parent, struct socket *newsock)
/* FIXME: Is this check still needed */
if (sk->sk_state == BT_CLOSED) {
- release_sock(sk);
bt_accept_unlink(sk);
+ release_sock(sk);
continue;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
