mptcp: add IPV6_UNICAST_HOPS sockopt support

Implement setsockopt/getsockopt support for IPV6_UNICAST_HOPS in MPTCP:
1. Adds mptcp_setsockopt_v6_set_hops() to handle IPV6_UNICAST_HOPS:
   - Applies hop limit to the MPTCP socket
   - Propagates hop limit to all IPv6 subflows
   - Uses conditional compilation for CONFIG_MPTCP_IPV6
2. Adds getsockopt handler for IPV6_UNICAST_HOPS:
   - Retrieves hop limit from inet6_sk(sk)->hop_limit
   - Falls back to net->ipv6.devconf_all->hop_limit when not set
3. Syncs hop limit to new subflows in sync_socket_options()
4. Maintains option sequence counter for consistency

This enables applications to control IPv6 hop limits in MPTCP connections
using standard IPv6 socket options, ensuring consistent behavior across
all subflows.

Closes: multipath-tcp/mptcp_net-next#296
Signed-off-by: Geliang Tang <[email protected]>

Author: Geliang Tang

net/mptcp/sockopt.c

@@ -413,6 +413,34 @@ static int mptcp_setsockopt_v6_set_tclass(struct mptcp_sock *msk, int optname,
 	return 0;
 }
 
+static int mptcp_setsockopt_v6_set_hops(struct mptcp_sock *msk, int optname,
+					sockptr_t optval, unsigned int optlen)
+{
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	struct mptcp_subflow_context *subflow;
+	struct sock *sk = (struct sock *)msk;
+	int err, val;
+
+	err = ipv6_setsockopt(sk, SOL_IPV6, optname, optval, optlen);
+	if (err)
+		return err;
+
+	lock_sock(sk);
+	sockopt_seq_inc(msk);
+	val = READ_ONCE(inet6_sk(sk)->hop_limit);
+	mptcp_for_each_subflow(msk, subflow) {
+		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
+		bool slow;
+
+		slow = lock_sock_fast(ssk);
+		WRITE_ONCE(inet6_sk(ssk)->hop_limit, val);
+		unlock_sock_fast(ssk, slow);
+	}
+	release_sock(sk);
+#endif
+
+	return 0;
+}
 static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname,
 			       sockptr_t optval, unsigned int optlen)
 {
@@ -457,6 +485,8 @@ static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname,
 		break;
 	case IPV6_TCLASS:
 		return mptcp_setsockopt_v6_set_tclass(msk, optname, optval, optlen);
+	case IPV6_UNICAST_HOPS:
+		return mptcp_setsockopt_v6_set_hops(msk, optname, optval, optlen);
 	}
 
 	return ret;
@@ -1530,6 +1560,7 @@ static int mptcp_getsockopt_v6(struct mptcp_sock *msk, int optname,
 			       char __user *optval, int __user *optlen)
 {
 	struct sock *sk = (void *)msk;
+	int val;
 
 	switch (optname) {
 	case IPV6_V6ONLY:
@@ -1544,6 +1575,11 @@ static int mptcp_getsockopt_v6(struct mptcp_sock *msk, int optname,
 	case IPV6_TCLASS:
 		return mptcp_put_int_option(msk, optval, optlen,
 					    inet6_sk(sk)->tclass);
+	case IPV6_UNICAST_HOPS:
+		val = READ_ONCE(inet6_sk(sk)->hop_limit);
+		if (val < 0)
+			val = READ_ONCE(sock_net(sk)->ipv6.devconf_all->hop_limit);
+		return mptcp_put_int_option(msk, optval, optlen, val);
 	}
 
 	return -EOPNOTSUPP;
@@ -1616,6 +1652,7 @@ static void sync_socket_options(struct mptcp_sock *msk, struct sock *ssk)
 	__ip_sock_set_tos(ssk, inet_sk(sk)->tos);
 	WRITE_ONCE(inet6_sk(ssk)->tclass, inet6_sk(sk)->tclass);
 	WRITE_ONCE(inet_sk(ssk)->uc_ttl, inet_sk(sk)->uc_ttl);
+	WRITE_ONCE(inet6_sk(ssk)->hop_limit, inet6_sk(sk)->hop_limit);
 
 	if (sk->sk_userlocks & tx_rx_locks) {
 		ssk->sk_userlocks |= sk->sk_userlocks & tx_rx_locks;