Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

net/devlink/leftover.c / net/core/devlink.c:
  565b4824c39f ("devlink: change port event netdev notifier from per-net to global")
  f05bd8ebeb69 ("devlink: move code to a dedicated directory")
  687125b5799c ("devlink: split out core code")
https://lore.kernel.org/all/20230208094657.379f2b1a@canb.auug.org.au/

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

4 files changed, 34 insertions, 8 deletions

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index db07cc5b4fcb..56628b52d100 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1002,8 +1002,8 @@ static int mptcp_pm_nl_create_listen_socket(struct sock *sk,
 {
 	int addrlen = sizeof(struct sockaddr_in);
 	struct sockaddr_storage addr;
-	struct mptcp_sock *msk;
 	struct socket *ssock;
+	struct sock *newsk;
 	int backlog = 1024;
 	int err;
 
@@ -1012,11 +1012,13 @@ static int mptcp_pm_nl_create_listen_socket(struct sock *sk,
 	if (err)
 		return err;
 
-	msk = mptcp_sk(entry->lsk->sk);
-	if (!msk)
+	newsk = entry->lsk->sk;
+	if (!newsk)
 		return -EINVAL;
 
-	ssock = __mptcp_nmpc_socket(msk);
+	lock_sock(newsk);
+	ssock = __mptcp_nmpc_socket(mptcp_sk(newsk));
+	release_sock(newsk);
 	if (!ssock)
 		return -EINVAL;
 
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 1ff2efbab29e..c9817aa0f413 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2902,6 +2902,7 @@ bool __mptcp_close(struct sock *sk, long timeout)
 	struct mptcp_subflow_context *subflow;
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	bool do_cancel_work = false;
+	int subflows_alive = 0;
 
 	sk->sk_shutdown = SHUTDOWN_MASK;
 
@@ -2928,6 +2929,8 @@ cleanup:
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 		bool slow = lock_sock_fast_nested(ssk);
 
+		subflows_alive += ssk->sk_state != TCP_CLOSE;
+
 		/* since the close timeout takes precedence on the fail one,
 		 * cancel the latter
 		 */
@@ -2943,6 +2946,12 @@ cleanup:
 	}
 	sock_orphan(sk);
 
+	/* all the subflows are closed, only timeout can change the msk
+	 * state, let's not keep resources busy for no reasons
+	 */
+	if (subflows_alive == 0)
+		inet_sk_state_store(sk, TCP_CLOSE);
+
 	sock_hold(sk);
 	pr_debug("msk=%p state=%d", sk, sk->sk_state);
 	if (msk->token)
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c
index 9986681aaf40..8a9656248b0f 100644
--- a/net/mptcp/sockopt.c
+++ b/net/mptcp/sockopt.c
@@ -760,14 +760,21 @@ static int mptcp_setsockopt_v4(struct mptcp_sock *msk, int optname,
 static int mptcp_setsockopt_first_sf_only(struct mptcp_sock *msk, int level, int optname,
 					  sockptr_t optval, unsigned int optlen)
 {
+	struct sock *sk = (struct sock *)msk;
 	struct socket *sock;
+	int ret = -EINVAL;
 
 	/* Limit to first subflow, before the connection establishment */
+	lock_sock(sk);
 	sock = __mptcp_nmpc_socket(msk);
 	if (!sock)
-		return -EINVAL;
+		goto unlock;
 
-	return tcp_setsockopt(sock->sk, level, optname, optval, optlen);
+	ret = tcp_setsockopt(sock->sk, level, optname, optval, optlen);
+
+unlock:
+	release_sock(sk);
+	return ret;
 }
 
 static int mptcp_setsockopt_sol_tcp(struct mptcp_sock *msk, int optname,
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index beaec843f5ca..4ae1a7304cf0 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1400,6 +1400,7 @@ void __mptcp_error_report(struct sock *sk)
 	mptcp_for_each_subflow(msk, subflow) {
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 		int err = sock_error(ssk);
+		int ssk_state;
 
 		if (!err)
 			continue;
@@ -1410,7 +1411,14 @@ void __mptcp_error_report(struct sock *sk)
 		if (sk->sk_state != TCP_SYN_SENT && !__mptcp_check_fallback(msk))
 			continue;
 
-		inet_sk_state_store(sk, inet_sk_state_load(ssk));
+		/* We need to propagate only transition to CLOSE state.
+		 * Orphaned socket will see such state change via
+		 * subflow_sched_work_if_closed() and that path will properly
+		 * destroy the msk as needed.
+		 */
+		ssk_state = inet_sk_state_load(ssk);
+		if (ssk_state == TCP_CLOSE && !sock_flag(sk, SOCK_DEAD))
+			inet_sk_state_store(sk, ssk_state);
 		sk->sk_err = -err;
 
 		/* This barrier is coupled with smp_rmb() in mptcp_poll() */
@@ -1682,7 +1690,7 @@ int mptcp_subflow_create_socket(struct sock *sk, unsigned short family,
 	if (err)
 		return err;
 
-	lock_sock(sf->sk);
+	lock_sock_nested(sf->sk, SINGLE_DEPTH_NESTING);
 
 	/* the newly created socket has to be in the same cgroup as its parent */
 	mptcp_attach_cgroup(sk, sf->sk);