mptcp: Handle MP_CAPABLE options for outgoing connections

Add hooks to tcp_output.c to add MP_CAPABLE to an outgoing SYN request,
to capture the MP_CAPABLE in the received SYN-ACK, to add MP_CAPABLE to
the final ACK of the three-way handshake.

Use the .sk_rx_dst_set() handler in the subflow proto to capture when the
responding SYN-ACK is received and notify the MPTCP connection layer.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/tcp.h

@@ -137,6 +137,9 @@ 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;

include/net/mptcp.h

@@ -39,8 +39,27 @@ struct mptcp_out_options {
 
 void mptcp_init(void);
 
+static inline bool sk_is_mptcp(const struct sock *sk)
+{
+ return tcp_sk(sk)->is_mptcp;
+}
+
+static inline bool rsk_is_mptcp(const struct request_sock *req)
+{
+ return tcp_rsk(req)->is_mptcp;
+}
+
 void mptcp_parse_option(const unsigned char *ptr, int opsize,
  struct tcp_options_received *opt_rx);
+bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+ struct mptcp_out_options *opts);
+void mptcp_rcv_synsent(struct sock *sk);
+bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
+ struct mptcp_out_options *opts);
+bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
+ unsigned int *size, unsigned int remaining,
+ struct mptcp_out_options *opts);
+
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts);
 
 /* move the skb extension owership, with the assumption that 'to' is
@@ -89,11 +108,47 @@ static inline void mptcp_init(void)
 {
 }
 
+static inline bool sk_is_mptcp(const struct sock *sk)
+{
+ return false;
+}
+
+static inline bool rsk_is_mptcp(const struct request_sock *req)
+{
+ return false;
+}
+
 static inline void mptcp_parse_option(const unsigned char *ptr, int opsize,
  struct tcp_options_received *opt_rx)
 {
 }
 
+static inline bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+ struct mptcp_out_options *opts)
+{
+ return false;
+}
+
+static inline void mptcp_rcv_synsent(struct sock *sk)
+{
+}
+
+static inline bool mptcp_synack_options(const struct request_sock *req,
+ unsigned int *size,
+ struct mptcp_out_options *opts)
+{
+ return false;
+}
+
+static inline bool mptcp_established_options(struct sock *sk,
+ struct sk_buff *skb,
+ unsigned int *size,
+ unsigned int remaining,
+ struct mptcp_out_options *opts)
+{
+ return false;
+}
+
 static inline void mptcp_skb_ext_move(struct sk_buff *to,
  const struct sk_buff *from)
 {
@@ -107,6 +162,8 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
 
 #endif /* CONFIG_MPTCP */
 
+void mptcp_handle_ipv6_mapped(struct sock *sk, bool mapped);
+
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 int mptcpv6_init(void);
 #elif IS_ENABLED(CONFIG_IPV6)

net/ipv4/tcp_input.c

@@ -5978,6 +5978,9 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
  tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
  tcp_initialize_rcv_mss(sk);
 
+ if (sk_is_mptcp(sk))
+ mptcp_rcv_synsent(sk);
+
  /* Remember, tcp_poll() does not lock socket!
  * Change state from SYN-SENT only after copied_seq
  * is initialized. */
@@ -6600,6 +6603,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 
  tcp_rsk(req)->af_specific = af_ops;
  tcp_rsk(req)->ts_off = 0;
+#if IS_ENABLED(CONFIG_MPTCP)
+ tcp_rsk(req)->is_mptcp = 0;
+#endif
 
  tcp_clear_options(&tmp_opt);
  tmp_opt.mss_clamp = af_ops->mss_clamp;

net/ipv4/tcp_output.c

@@ -597,6 +597,22 @@ static void smc_set_option_cond(const struct tcp_sock *tp,
 #endif
 }
 
+static void mptcp_set_option_cond(const struct request_sock *req,
+ struct tcp_out_options *opts,
+ unsigned int *remaining)
+{
+ if (rsk_is_mptcp(req)) {
+ unsigned int size;
+
+ if (mptcp_synack_options(req, &size, &opts->mptcp)) {
+ if (*remaining >= size) {
+ opts->options |= OPTION_MPTCP;
+ *remaining -= size;
+ }
+ }
+ }
+}
+
 /* Compute TCP options for SYN packets. This is not the final
  * network wire format yet.
  */
@@ -666,6 +682,15 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb,
 
  smc_set_option(tp, opts, &remaining);
 
+ if (sk_is_mptcp(sk)) {
+ unsigned int size;
+
+ if (mptcp_syn_options(sk, &size, &opts->mptcp)) {
+ opts->options |= OPTION_MPTCP;
+ remaining -= size;
+ }
+ }
+
  return MAX_TCP_OPTION_SPACE - remaining;
 }
 
@@ -727,6 +752,8 @@ static unsigned int tcp_synack_options(const struct sock *sk,
  }
  }
 
+ mptcp_set_option_cond(req, opts, &remaining);
+
  smc_set_option_cond(tcp_sk(sk), ireq, opts, &remaining);
 
  return MAX_TCP_OPTION_SPACE - remaining;
@@ -764,6 +791,23 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
  size += TCPOLEN_TSTAMP_ALIGNED;
  }
 
+ /* MPTCP options have precedence over SACK for the limited TCP
+ * option space because a MPTCP connection would be forced to
+ * fall back to regular TCP if a required multipath option is
+ * missing. SACK still gets a chance to use whatever space is
+ * left.
+ */
+ if (sk_is_mptcp(sk)) {
+ unsigned int remaining = MAX_TCP_OPTION_SPACE - size;
+ unsigned int opt_size = 0;
+
+ if (mptcp_established_options(sk, skb, &opt_size, remaining,
+ &opts->mptcp)) {
+ opts->options |= OPTION_MPTCP;
+ size += opt_size;
+ }
+ }
+
  eff_sacks = tp->rx_opt.num_sacks + tp->rx_opt.dsack;
  if (unlikely(eff_sacks)) {
  const unsigned int remaining = MAX_TCP_OPTION_SPACE - size;

net/ipv6/tcp_ipv6.c

@@ -238,6 +238,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
  sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
 
  icsk->icsk_af_ops = &ipv6_mapped;
+ if (sk_is_mptcp(sk))
+ mptcp_handle_ipv6_mapped(sk, true);
  sk->sk_backlog_rcv = tcp_v4_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
  tp->af_specific = &tcp_sock_ipv6_mapped_specific;
@@ -248,6 +250,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
  if (err) {
  icsk->icsk_ext_hdr_len = exthdrlen;
  icsk->icsk_af_ops = &ipv6_specific;
+ if (sk_is_mptcp(sk))
+ mptcp_handle_ipv6_mapped(sk, false);
  sk->sk_backlog_rcv = tcp_v6_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
  tp->af_specific = &tcp_sock_ipv6_specific;
@@ -1203,6 +1207,8 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
  newnp->saddr = newsk->sk_v6_rcv_saddr;
 
  inet_csk(newsk)->icsk_af_ops = &ipv6_mapped;
+ if (sk_is_mptcp(newsk))
+ mptcp_handle_ipv6_mapped(newsk, true);
  newsk->sk_backlog_rcv = tcp_v4_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
  newtp->af_specific = &tcp_sock_ipv6_mapped_specific;

net/mptcp/options.c

@@ -72,14 +72,114 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
  }
 }
 
+void mptcp_get_options(const struct sk_buff *skb,
+ struct tcp_options_received *opt_rx)
+{
+ const unsigned char *ptr;
+ const struct tcphdr *th = tcp_hdr(skb);
+ int length = (th->doff * 4) - sizeof(struct tcphdr);
+
+ ptr = (const unsigned char *)(th + 1);
+
+ while (length > 0) {
+ int opcode = *ptr++;
+ int opsize;
+
+ switch (opcode) {
+ case TCPOPT_EOL:
+ return;
+ case TCPOPT_NOP: /* Ref: RFC 793 section 3.1 */
+ length--;
+ continue;
+ default:
+ opsize = *ptr++;
+ if (opsize < 2) /* "silly options" */
+ return;
+ if (opsize > length)
+ return; /* don't parse partial options */
+ if (opcode == TCPOPT_MPTCP)
+ mptcp_parse_option(ptr, opsize, opt_rx);
+ ptr += opsize - 2;
+ length -= opsize;
+ }
+ }
+}
+
+bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+ struct mptcp_out_options *opts)
+{
+ struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+ if (subflow->request_mptcp) {
+ pr_debug("local_key=%llu", subflow->local_key);
+ opts->suboptions = OPTION_MPTCP_MPC_SYN;
+ opts->sndr_key = subflow->local_key;
+ *size = TCPOLEN_MPTCP_MPC_SYN;
+ return true;
+ }
+ return false;
+}
+
+void mptcp_rcv_synsent(struct sock *sk)
+{
+ struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ pr_debug("subflow=%p", subflow);
+ if (subflow->request_mptcp && tp->rx_opt.mptcp.mp_capable) {
+ subflow->mp_capable = 1;
+ subflow->remote_key = tp->rx_opt.mptcp.sndr_key;
+ } else {
+ tcp_sk(sk)->is_mptcp = 0;
+ }
+}
+
+bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
+ unsigned int *size, unsigned int remaining,
+ struct mptcp_out_options *opts)
+{
+ struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+ if (subflow->mp_capable && !subflow->fourth_ack) {
+ opts->suboptions = OPTION_MPTCP_MPC_ACK;
+ opts->sndr_key = subflow->local_key;
+ opts->rcvr_key = subflow->remote_key;
+ *size = TCPOLEN_MPTCP_MPC_ACK;
+ subflow->fourth_ack = 1;
+ pr_debug("subflow=%p, local_key=%llu, remote_key=%llu",
+ subflow, subflow->local_key, subflow->remote_key);
+ return true;
+ }
+ return false;
+}
+
+bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
+ struct mptcp_out_options *opts)
+{
+ struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
+
+ if (subflow_req->mp_capable) {
+ opts->suboptions = OPTION_MPTCP_MPC_SYNACK;
+ opts->sndr_key = subflow_req->local_key;
+ *size = TCPOLEN_MPTCP_MPC_SYNACK;
+ pr_debug("subflow_req=%p, local_key=%llu",
+ subflow_req, subflow_req->local_key);
+ return true;
+ }
+ return false;
+}
+
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 {
  if ((OPTION_MPTCP_MPC_SYN |
+ OPTION_MPTCP_MPC_SYNACK |
  OPTION_MPTCP_MPC_ACK) & opts->suboptions) {
  u8 len;
 
  if (OPTION_MPTCP_MPC_SYN & opts->suboptions)
  len = TCPOLEN_MPTCP_MPC_SYN;
+ else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions)
+ len = TCPOLEN_MPTCP_MPC_SYNACK;
  else
  len = TCPOLEN_MPTCP_MPC_ACK;