Merge branch 'mptcp-fix-sockopt-crash-and-lockdep-splats'

Florian Westphal says:

====================
mptcp: fix sockopt crash and lockdep splats

Christoph Paasch reported a few bugs and lockdep splats triggered by
syzkaller.

One patch fixes a crash in set/getsockopt.
Two patches fix lockdep splats related to the order in which RTNL
and socket lock are taken.

Last patch fixes out-of-bounds access when TCP syncookies are used.

Change since last iteration on mptcp-list:
 - add needed refcount in patch 2
 - call tcp_get/setsockopt directly in patch 2

 Other patches unchanged except minor amends to commit messages.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

6 files changed, 46 insertions, 25 deletions

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 1cf73e6f85ca..3dc964010fef 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -148,9 +148,7 @@ struct tcp_request_sock {
 	const struct tcp_request_sock_ops *af_specific;
 	u64				snt_synack; /* first SYNACK sent time */
 	bool				tfo_listener;
-#if IS_ENABLED(CONFIG_MPTCP)
 	bool				is_mptcp;
-#endif
 	u32				txhash;
 	u32				rcv_isn;
 	u32				snt_isn;
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 345b2b0ff618..9a4f6b16c9bc 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -349,6 +349,10 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
 	treq->snt_synack	= 0;
 	treq->tfo_listener	= false;
+
+	if (IS_ENABLED(CONFIG_MPTCP))
+		treq->is_mptcp = 0;
+
 	if (IS_ENABLED(CONFIG_SMC))
 		ireq->smc_ok = 0;
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index e8b840a4767e..e325b4506e25 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6637,6 +6637,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 
 	af_ops->init_req(req, sk, skb);
 
+	if (IS_ENABLED(CONFIG_MPTCP) && want_cookie)
+		tcp_rsk(req)->is_mptcp = 0;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto drop_and_free;
 
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 30915f6f31e3..13235a012388 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -178,6 +178,9 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	treq = tcp_rsk(req);
 	treq->tfo_listener = false;
 
+	if (IS_ENABLED(CONFIG_MPTCP))
+		treq->is_mptcp = 0;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto out_free;
 
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 39fdca79ce90..73c192d8c158 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -644,19 +644,21 @@ static void __mptcp_close(struct sock *sk, long timeout)
 {
 	struct mptcp_subflow_context *subflow, *tmp;
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	LIST_HEAD(conn_list);
 
 	mptcp_token_destroy(msk->token);
 	inet_sk_state_store(sk, TCP_CLOSE);
 
-	list_for_each_entry_safe(subflow, tmp, &msk->conn_list, node) {
+	list_splice_init(&msk->conn_list, &conn_list);
+
+	release_sock(sk);
+
+	list_for_each_entry_safe(subflow, tmp, &conn_list, node) {
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 
 		__mptcp_close_ssk(sk, ssk, subflow, timeout);
 	}
 
-	if (msk->cached_ext)
-		__skb_ext_put(msk->cached_ext);
-	release_sock(sk);
 	sk_common_release(sk);
 }
 
@@ -776,18 +778,19 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 
 static void mptcp_destroy(struct sock *sk)
 {
+	struct mptcp_sock *msk = mptcp_sk(sk);
+
+	if (msk->cached_ext)
+		__skb_ext_put(msk->cached_ext);
 }
 
 static int mptcp_setsockopt(struct sock *sk, int level, int optname,
-			    char __user *uoptval, unsigned int optlen)
+			    char __user *optval, unsigned int optlen)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	char __kernel *optval;
 	int ret = -EOPNOTSUPP;
 	struct socket *ssock;
-
-	/* will be treated as __user in tcp_setsockopt */
-	optval = (char __kernel __force *)uoptval;
+	struct sock *ssk;
 
 	pr_debug("msk=%p", msk);
 
@@ -796,27 +799,28 @@ static int mptcp_setsockopt(struct sock *sk, int level, int optname,
 	 */
 	lock_sock(sk);
 	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
-	if (!IS_ERR(ssock)) {
-		pr_debug("subflow=%p", ssock->sk);
-		ret = kernel_setsockopt(ssock, level, optname, optval, optlen);
+	if (IS_ERR(ssock)) {
+		release_sock(sk);
+		return ret;
 	}
+
+	ssk = ssock->sk;
+	sock_hold(ssk);
 	release_sock(sk);
 
+	ret = tcp_setsockopt(ssk, level, optname, optval, optlen);
+	sock_put(ssk);
+
 	return ret;
 }
 
 static int mptcp_getsockopt(struct sock *sk, int level, int optname,
-			    char __user *uoptval, int __user *uoption)
+			    char __user *optval, int __user *option)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	char __kernel *optval;
 	int ret = -EOPNOTSUPP;
-	int __kernel *option;
 	struct socket *ssock;
-
-	/* will be treated as __user in tcp_getsockopt */
-	optval = (char __kernel __force *)uoptval;
-	option = (int __kernel __force *)uoption;
+	struct sock *ssk;
 
 	pr_debug("msk=%p", msk);
 
@@ -825,12 +829,18 @@ static int mptcp_getsockopt(struct sock *sk, int level, int optname,
 	 */
 	lock_sock(sk);
 	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
-	if (!IS_ERR(ssock)) {
-		pr_debug("subflow=%p", ssock->sk);
-		ret = kernel_getsockopt(ssock, level, optname, optval, option);
+	if (IS_ERR(ssock)) {
+		release_sock(sk);
+		return ret;
 	}
+
+	ssk = ssock->sk;
+	sock_hold(ssk);
 	release_sock(sk);
 
+	ret = tcp_getsockopt(ssk, level, optname, optval, option);
+	sock_put(ssk);
+
 	return ret;
 }
 
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 205dca1c30b7..c90c0e6ffb82 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -186,6 +186,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn);
 
+	if (tcp_rsk(req)->is_mptcp == 0)
+		goto create_child;
+
 	/* if the sk is MP_CAPABLE, we try to fetch the client key */
 	subflow_req = mptcp_subflow_rsk(req);
 	if (subflow_req->mp_capable) {
@@ -769,7 +772,7 @@ static void subflow_ulp_clone(const struct request_sock *req,
 	struct mptcp_subflow_context *old_ctx = mptcp_subflow_ctx(newsk);
 	struct mptcp_subflow_context *new_ctx;
 
-	if (!subflow_req->mp_capable) {
+	if (!tcp_rsk(req)->is_mptcp || !subflow_req->mp_capable) {
 		subflow_ulp_fallback(newsk, old_ctx);
 		return;
 	}