Merge branch 'mptcp-misc'

Mat Martineau says:

====================
MPTCP: Miscellaneous changes

Here is a collection of patches from the MPTCP tree:

Patches 1 and 2 add some helpful MIB counters for connection
information.

Patch 3 cleans up some unnecessary checks.

Patch 4 is a new feature, support for the MP_TCPRST option. This option
is used when resetting one subflow within a MPTCP connection, and
provides a reason code that the recipient can use when deciding how to
adapt to the lost subflow.

Patches 5-7 update the existing MPTCP selftests to improve timeout
handling and to share better information when tests fail.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/mptcp.h

@@ -30,8 +30,8 @@ struct mptcp_ext {
 			ack64:1,
 			mpc_map:1,
 			frozen:1,
-			__unused:1;
-	/* one byte hole */
+			reset_transient:1;
+	u8		reset_reason:4;
 };
 
 #define MPTCP_RM_IDS_MAX	8
@@ -58,6 +58,8 @@ struct mptcp_out_options {
 	struct mptcp_rm_list rm_list;
 	u8 join_id;
 	u8 backup;
+	u8 reset_reason:4;
+	u8 reset_transient:1;
 	u32 nonce;
 	u64 thmac;
 	u32 token;
@@ -156,6 +158,16 @@ void mptcp_seq_show(struct seq_file *seq);
 int mptcp_subflow_init_cookie_req(struct request_sock *req,
 				  const struct sock *sk_listener,
 				  struct sk_buff *skb);
+
+__be32 mptcp_get_reset_option(const struct sk_buff *skb);
+
+static inline __be32 mptcp_reset_option(const struct sk_buff *skb)
+{
+	if (skb_ext_exist(skb, SKB_EXT_MPTCP))
+		return mptcp_get_reset_option(skb);
+
+	return htonl(0u);
+}
 #else
 
 static inline void mptcp_init(void)
@@ -236,6 +248,8 @@ static inline int mptcp_subflow_init_cookie_req(struct request_sock *req,
 {
 	return 0; /* TCP fallback */
 }
+
+static inline __be32 mptcp_reset_option(const struct sk_buff *skb)  { return htonl(0u); }
 #endif /* CONFIG_MPTCP */
 
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)

include/uapi/linux/mptcp.h

@@ -174,10 +174,21 @@ enum mptcp_event_attr {
 	MPTCP_ATTR_FLAGS,	/* u16 */
 	MPTCP_ATTR_TIMEOUT,	/* u32 */
 	MPTCP_ATTR_IF_IDX,	/* s32 */
+	MPTCP_ATTR_RESET_REASON,/* u32 */
+	MPTCP_ATTR_RESET_FLAGS, /* u32 */
 
 	__MPTCP_ATTR_AFTER_LAST
 };
 
 #define MPTCP_ATTR_MAX (__MPTCP_ATTR_AFTER_LAST - 1)
 
+/* MPTCP Reset reason codes, rfc8684 */
+#define MPTCP_RST_EUNSPEC	0
+#define MPTCP_RST_EMPTCP	1
+#define MPTCP_RST_ERESOURCE	2
+#define MPTCP_RST_EPROHIBIT	3
+#define MPTCP_RST_EWQ2BIG	4
+#define MPTCP_RST_EBADPERF	5
+#define MPTCP_RST_EMIDDLEBOX	6
+
 #endif /* _UAPI_MPTCP_H */

net/ipv4/tcp_ipv4.c

@@ -655,14 +655,18 @@ EXPORT_SYMBOL(tcp_v4_send_check);
  *	Exception: precedence violation. We do not implement it in any case.
  */
 
+#ifdef CONFIG_TCP_MD5SIG
+#define OPTION_BYTES TCPOLEN_MD5SIG_ALIGNED
+#else
+#define OPTION_BYTES sizeof(__be32)
+#endif
+
 static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 {
 	const struct tcphdr *th = tcp_hdr(skb);
 	struct {
 		struct tcphdr th;
-#ifdef CONFIG_TCP_MD5SIG
-		__be32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)];
-#endif
+		__be32 opt[OPTION_BYTES / sizeof(__be32)];
 	} rep;
 	struct ip_reply_arg arg;
 #ifdef CONFIG_TCP_MD5SIG
@@ -770,6 +774,17 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 				     ip_hdr(skb)->daddr, &rep.th);
 	}
 #endif
+	/* Can't co-exist with TCPMD5, hence check rep.opt[0] */
+	if (rep.opt[0] == 0) {
+		__be32 mrst = mptcp_reset_option(skb);
+
+		if (mrst) {
+			rep.opt[0] = mrst;
+			arg.iov[0].iov_len += sizeof(mrst);
+			rep.th.doff = arg.iov[0].iov_len / 4;
+		}
+	}
+
 	arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr,
 				      ip_hdr(skb)->saddr, /* XXX */
 				      arg.iov[0].iov_len, IPPROTO_TCP, 0);

net/ipv6/tcp_ipv6.c

@@ -879,8 +879,8 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 	struct net *net = sk ? sock_net(sk) : dev_net(skb_dst(skb)->dev);
 	struct sock *ctl_sk = net->ipv6.tcp_sk;
 	unsigned int tot_len = sizeof(struct tcphdr);
+	__be32 mrst = 0, *topt;
 	struct dst_entry *dst;
-	__be32 *topt;
 	__u32 mark = 0;
 
 	if (tsecr)
@@ -890,6 +890,15 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 		tot_len += TCPOLEN_MD5SIG_ALIGNED;
 #endif
 
+#ifdef CONFIG_MPTCP
+	if (rst && !key) {
+		mrst = mptcp_reset_option(skb);
+
+		if (mrst)
+			tot_len += sizeof(__be32);
+	}
+#endif
+
 	buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len,
 			 GFP_ATOMIC);
 	if (!buff)
@@ -920,6 +929,9 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 		*topt++ = htonl(tsecr);
 	}
 
+	if (mrst)
+		*topt++ = mrst;
+
 #ifdef CONFIG_TCP_MD5SIG
 	if (key) {
 		*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |

net/mptcp/mib.c

@@ -10,9 +10,12 @@
 
 static const struct snmp_mib mptcp_snmp_list[] = {
 	SNMP_MIB_ITEM("MPCapableSYNRX", MPTCP_MIB_MPCAPABLEPASSIVE),
+	SNMP_MIB_ITEM("MPCapableSYNTX", MPTCP_MIB_MPCAPABLEACTIVE),
+	SNMP_MIB_ITEM("MPCapableSYNACKRX", MPTCP_MIB_MPCAPABLEACTIVEACK),
 	SNMP_MIB_ITEM("MPCapableACKRX", MPTCP_MIB_MPCAPABLEPASSIVEACK),
 	SNMP_MIB_ITEM("MPCapableFallbackACK", MPTCP_MIB_MPCAPABLEPASSIVEFALLBACK),
 	SNMP_MIB_ITEM("MPCapableFallbackSYNACK", MPTCP_MIB_MPCAPABLEACTIVEFALLBACK),
+	SNMP_MIB_ITEM("MPFallbackTokenInit", MPTCP_MIB_TOKENFALLBACKINIT),
 	SNMP_MIB_ITEM("MPTCPRetrans", MPTCP_MIB_RETRANSSEGS),
 	SNMP_MIB_ITEM("MPJoinNoTokenFound", MPTCP_MIB_JOINNOTOKEN),
 	SNMP_MIB_ITEM("MPJoinSynRx", MPTCP_MIB_JOINSYNRX),

net/mptcp/mib.h

@@ -3,9 +3,12 @@
 enum linux_mptcp_mib_field {
 	MPTCP_MIB_NUM = 0,
 	MPTCP_MIB_MPCAPABLEPASSIVE,	/* Received SYN with MP_CAPABLE */
+	MPTCP_MIB_MPCAPABLEACTIVE,	/* Sent SYN with MP_CAPABLE */
+	MPTCP_MIB_MPCAPABLEACTIVEACK,	/* Received SYN/ACK with MP_CAPABLE */
 	MPTCP_MIB_MPCAPABLEPASSIVEACK,	/* Received third ACK with MP_CAPABLE */
 	MPTCP_MIB_MPCAPABLEPASSIVEFALLBACK,/* Server-side fallback during 3-way handshake */
 	MPTCP_MIB_MPCAPABLEACTIVEFALLBACK, /* Client-side fallback during 3-way handshake */
+	MPTCP_MIB_TOKENFALLBACKINIT,	/* Could not init/allocate token */
 	MPTCP_MIB_RETRANSSEGS,		/* Segments retransmitted at the MPTCP-level */
 	MPTCP_MIB_JOINNOTOKEN,		/* Received MP_JOIN but the token was not found */
 	MPTCP_MIB_JOINSYNRX,		/* Received a SYN + MP_JOIN */

net/mptcp/options.c

@@ -305,6 +305,18 @@ static void mptcp_parse_option(const struct sk_buff *skb,
 		mp_opt->fastclose = 1;
 		break;
 
+	case MPTCPOPT_RST:
+		if (opsize != TCPOLEN_MPTCP_RST)
+			break;
+
+		if (!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_RST))
+			break;
+		mp_opt->reset = 1;
+		flags = *ptr++;
+		mp_opt->reset_transient = flags & MPTCP_RST_TRANSIENT;
+		mp_opt->reset_reason = *ptr;
+		break;
+
 	default:
 		break;
 	}
@@ -327,6 +339,7 @@ void mptcp_get_options(const struct sk_buff *skb,
 	mp_opt->rm_addr = 0;
 	mp_opt->dss = 0;
 	mp_opt->mp_prio = 0;
+	mp_opt->reset = 0;
 
 	length = (th->doff * 4) - sizeof(struct tcphdr);
 	ptr = (const unsigned char *)(th + 1);
@@ -726,6 +739,22 @@ static bool mptcp_established_options_mp_prio(struct sock *sk,
 	return true;
 }
 
+static noinline void mptcp_established_options_rst(struct sock *sk, struct sk_buff *skb,
+						   unsigned int *size,
+						   unsigned int remaining,
+						   struct mptcp_out_options *opts)
+{
+	const struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+	if (remaining < TCPOLEN_MPTCP_RST)
+		return;
+
+	*size = TCPOLEN_MPTCP_RST;
+	opts->suboptions |= OPTION_MPTCP_RST;
+	opts->reset_transient = subflow->reset_transient;
+	opts->reset_reason = subflow->reset_reason;
+}
+
 bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 			       unsigned int *size, unsigned int remaining,
 			       struct mptcp_out_options *opts)
@@ -741,11 +770,10 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 	if (unlikely(__mptcp_check_fallback(msk)))
 		return false;
 
-	/* prevent adding of any MPTCP related options on reset packet
-	 * until we support MP_TCPRST/MP_FASTCLOSE
-	 */
-	if (unlikely(skb && TCP_SKB_CB(skb)->tcp_flags & TCPHDR_RST))
-		return false;
+	if (unlikely(skb && TCP_SKB_CB(skb)->tcp_flags & TCPHDR_RST)) {
+		mptcp_established_options_rst(sk, skb, size, remaining, opts);
+		return true;
+	}
 
 	snd_data_fin = mptcp_data_fin_enabled(msk);
 	if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, remaining, opts))
@@ -952,7 +980,7 @@ bool mptcp_update_rcv_data_fin(struct mptcp_sock *msk, u64 data_fin_seq, bool us
 	 * should match. If they mismatch, the peer is misbehaving and
 	 * we will prefer the most recent information.
 	 */
-	if (READ_ONCE(msk->rcv_data_fin) || !READ_ONCE(msk->first))
+	if (READ_ONCE(msk->rcv_data_fin))
 		return false;
 
 	WRITE_ONCE(msk->rcv_data_fin_seq,
@@ -1062,6 +1090,12 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
 		mp_opt.mp_prio = 0;
 	}
 
+	if (mp_opt.reset) {
+		subflow->reset_seen = 1;
+		subflow->reset_reason = mp_opt.reset_reason;
+		subflow->reset_transient = mp_opt.reset_transient;
+	}
+
 	if (!mp_opt.dss)
 		return;
 
@@ -1289,6 +1323,12 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
 		ptr += 5;
 	}
 
+	if (OPTION_MPTCP_RST & opts->suboptions)
+		*ptr++ = mptcp_option(MPTCPOPT_RST,
+				      TCPOLEN_MPTCP_RST,
+				      opts->reset_transient,
+				      opts->reset_reason);
+
 	if (opts->ext_copy.use_ack || opts->ext_copy.use_map) {
 		struct mptcp_ext *mpext = &opts->ext_copy;
 		u8 len = TCPOLEN_MPTCP_DSS_BASE;
@@ -1340,3 +1380,20 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
 	if (tp)
 		mptcp_set_rwin(tp);
 }
+
+__be32 mptcp_get_reset_option(const struct sk_buff *skb)
+{
+	const struct mptcp_ext *ext = mptcp_get_ext(skb);
+	u8 flags, reason;
+
+	if (ext) {
+		flags = ext->reset_transient;
+		reason = ext->reset_reason;
+
+		return mptcp_option(MPTCPOPT_RST, TCPOLEN_MPTCP_RST,
+				    flags, reason);
+	}
+
+	return htonl(0u);
+}
+EXPORT_SYMBOL_GPL(mptcp_get_reset_option);

net/mptcp/pm_netlink.c

@@ -1687,9 +1687,21 @@ static int mptcp_event_sub_closed(struct sk_buff *skb,
 				  const struct mptcp_sock *msk,
 				  const struct sock *ssk)
 {
+	const struct mptcp_subflow_context *sf;
+
 	if (mptcp_event_put_token_and_ssk(skb, msk, ssk))
 		return -EMSGSIZE;
 
+	sf = mptcp_subflow_ctx(ssk);
+	if (!sf->reset_seen)
+		return 0;
+
+	if (nla_put_u32(skb, MPTCP_ATTR_RESET_REASON, sf->reset_reason))
+		return -EMSGSIZE;
+
+	if (nla_put_u32(skb, MPTCP_ATTR_RESET_FLAGS, sf->reset_transient))
+		return -EMSGSIZE;
+
 	return 0;
 }
 

net/mptcp/protocol.c

@@ -493,7 +493,7 @@ static bool mptcp_check_data_fin(struct sock *sk)
 	u64 rcv_data_fin_seq;
 	bool ret = false;
 
-	if (__mptcp_check_fallback(msk) || !msk->first)
+	if (__mptcp_check_fallback(msk))
 		return ret;
 
 	/* Need to ack a DATA_FIN received from a peer while this side
@@ -3090,14 +3090,18 @@ bool mptcp_finish_join(struct sock *ssk)
 	pr_debug("msk=%p, subflow=%p", msk, subflow);
 
 	/* mptcp socket already closing? */
-	if (!mptcp_is_fully_established(parent))
+	if (!mptcp_is_fully_established(parent)) {
+		subflow->reset_reason = MPTCP_RST_EMPTCP;
 		return false;
+	}
 
 	if (!msk->pm.server_side)
 		goto out;
 
-	if (!mptcp_pm_allow_new_subflow(msk))
+	if (!mptcp_pm_allow_new_subflow(msk)) {
+		subflow->reset_reason = MPTCP_RST_EPROHIBIT;
 		return false;
+	}
 
 	/* active connections are already on conn_list, and we can't acquire
 	 * msk lock here.
@@ -3111,8 +3115,10 @@ bool mptcp_finish_join(struct sock *ssk)
 		sock_hold(ssk);
 	}
 	spin_unlock_bh(&msk->join_list_lock);
-	if (!ret)
+	if (!ret) {
+		subflow->reset_reason = MPTCP_RST_EPROHIBIT;
 		return false;
+	}
 
 	/* attach to msk socket only after we are sure he will deal with us
 	 * at close time
@@ -3224,8 +3230,12 @@ static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 	if (rcu_access_pointer(tcp_sk(ssock->sk)->md5sig_info))
 		mptcp_subflow_early_fallback(msk, subflow);
 #endif
-	if (subflow->request_mptcp && mptcp_token_new_connect(ssock->sk))
+	if (subflow->request_mptcp && mptcp_token_new_connect(ssock->sk)) {
+		MPTCP_INC_STATS(sock_net(ssock->sk), MPTCP_MIB_TOKENFALLBACKINIT);
 		mptcp_subflow_early_fallback(msk, subflow);
+	}
+	if (likely(!__mptcp_check_fallback(msk)))
+		MPTCP_INC_STATS(sock_net(sock->sk), MPTCP_MIB_MPCAPABLEACTIVE);
 
 do_connect:
 	err = ssock->ops->connect(ssock, uaddr, addr_len, flags);

net/mptcp/protocol.h

@@ -26,6 +26,7 @@
 #define OPTION_MPTCP_RM_ADDR	BIT(8)
 #define OPTION_MPTCP_FASTCLOSE	BIT(9)
 #define OPTION_MPTCP_PRIO	BIT(10)
+#define OPTION_MPTCP_RST	BIT(11)
 
 /* MPTCP option subtypes */
 #define MPTCPOPT_MP_CAPABLE	0
@@ -36,6 +37,7 @@
 #define MPTCPOPT_MP_PRIO	5
 #define MPTCPOPT_MP_FAIL	6
 #define MPTCPOPT_MP_FASTCLOSE	7
+#define MPTCPOPT_RST		8
 
 /* MPTCP suboption lengths */
 #define TCPOLEN_MPTCP_MPC_SYN		4
@@ -65,6 +67,7 @@
 #define TCPOLEN_MPTCP_PRIO		3
 #define TCPOLEN_MPTCP_PRIO_ALIGN	4
 #define TCPOLEN_MPTCP_FASTCLOSE		12
+#define TCPOLEN_MPTCP_RST		4
 
 /* MPTCP MP_JOIN flags */
 #define MPTCPOPT_BACKUP		BIT(0)
@@ -94,6 +97,9 @@
 /* MPTCP MP_PRIO flags */
 #define MPTCP_PRIO_BKUP		BIT(0)
 
+/* MPTCP TCPRST flags */
+#define MPTCP_RST_TRANSIENT	BIT(0)
+
 /* MPTCP socket flags */
 #define MPTCP_DATA_READY	0
 #define MPTCP_NOSPACE		1
@@ -123,6 +129,7 @@ struct mptcp_options_received {
 	u16	mp_capable : 1,
 		mp_join : 1,
 		fastclose : 1,
+		reset : 1,
 		dss : 1,
 		add_addr : 1,
 		rm_addr : 1,
@@ -152,6 +159,8 @@ struct mptcp_options_received {
 	};
 	u64	ahmac;
 	u16	port;
+	u8	reset_reason:4;
+	u8	reset_transient:1;
 };
 
 static inline __be32 mptcp_option(u8 subopt, u8 len, u8 nib, u8 field)
@@ -422,6 +431,9 @@ struct mptcp_subflow_context {
 	u8	hmac[MPTCPOPT_HMAC_LEN];
 	u8	local_id;
 	u8	remote_id;
+	u8	reset_seen:1;
+	u8	reset_transient:1;
+	u8	reset_reason:4;
 
 	long	delegated_status;
 	struct	list_head delegated_node;   /* link into delegated_action, protected by local BH */
@@ -742,7 +754,7 @@ unsigned int mptcp_pm_get_add_addr_accept_max(struct mptcp_sock *msk);
 unsigned int mptcp_pm_get_subflows_max(struct mptcp_sock *msk);
 unsigned int mptcp_pm_get_local_addr_max(struct mptcp_sock *msk);
 
-static inline struct mptcp_ext *mptcp_get_ext(struct sk_buff *skb)
+static inline struct mptcp_ext *mptcp_get_ext(const struct sk_buff *skb)
 {
 	return (struct mptcp_ext *)skb_ext_find(skb, SKB_EXT_MPTCP);
 }

net/mptcp/subflow.c

@@ -115,6 +115,16 @@ static bool subflow_use_different_sport(struct mptcp_sock *msk, const struct soc
 	return inet_sk(sk)->inet_sport != inet_sk((struct sock *)msk)->inet_sport;
 }
 
+static void subflow_add_reset_reason(struct sk_buff *skb, u8 reason)
+{
+	struct mptcp_ext *mpext = skb_ext_add(skb, SKB_EXT_MPTCP);
+
+	if (mpext) {
+		memset(mpext, 0, sizeof(*mpext));
+		mpext->reset_reason = reason;
+	}
+}
+
 /* Init mptcp request socket.
  *
  * Returns an error code if a JOIN has failed and a TCP reset
@@ -165,6 +175,7 @@ static int subflow_check_req(struct request_sock *req,
 			if (mptcp_token_exists(subflow_req->token)) {
 				if (retries-- > 0)
 					goto again;
+				SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_TOKENFALLBACKINIT);
 			} else {
 				subflow_req->mp_capable = 1;
 			}
@@ -176,6 +187,8 @@ static int subflow_check_req(struct request_sock *req,
 			subflow_req->mp_capable = 1;
 		else if (retries-- > 0)
 			goto again;
+		else
+			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_TOKENFALLBACKINIT);
 
 	} else if (mp_opt.mp_join && listener->request_mptcp) {
 		subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq;
@@ -187,8 +200,10 @@ static int subflow_check_req(struct request_sock *req,
 		subflow_req->msk = subflow_token_join_request(req);
 
 		/* Can't fall back to TCP in this case. */
-		if (!subflow_req->msk)
+		if (!subflow_req->msk) {
+			subflow_add_reset_reason(skb, MPTCP_RST_EMPTCP);
 			return -EPERM;
+		}
 
 		if (subflow_use_different_sport(subflow_req->msk, sk_listener)) {
 			pr_debug("syn inet_sport=%d %d",
@@ -392,12 +407,15 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 		subflow->remote_key = mp_opt.sndr_key;
 		pr_debug("subflow=%p, remote_key=%llu", subflow,
 			 subflow->remote_key);
+		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPCAPABLEACTIVEACK);
 		mptcp_finish_connect(sk);
 	} else if (subflow->request_join) {
 		u8 hmac[SHA256_DIGEST_SIZE];
 
-		if (!mp_opt.mp_join)
+		if (!mp_opt.mp_join) {
+			subflow->reset_reason = MPTCP_RST_EMPTCP;
 			goto do_reset;
+		}
 
 		subflow->thmac = mp_opt.thmac;
 		subflow->remote_nonce = mp_opt.nonce;
@@ -406,6 +424,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 
 		if (!subflow_thmac_valid(subflow)) {
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINACKMAC);
+			subflow->reset_reason = MPTCP_RST_EMPTCP;
 			goto do_reset;
 		}
 
@@ -434,6 +453,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 	return;
 
 do_reset:
+	subflow->reset_transient = 0;
 	mptcp_subflow_reset(sk);
 }
 
@@ -650,8 +670,10 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 		 * to reset the context to non MPTCP status.
 		 */
 		if (!ctx || fallback) {
-			if (fallback_is_fatal)
+			if (fallback_is_fatal) {
+				subflow_add_reset_reason(skb, MPTCP_RST_EMPTCP);
 				goto dispose_child;
+			}
 
 			subflow_drop_ctx(child);
 			goto out;
@@ -686,8 +708,10 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 			struct mptcp_sock *owner;
 
 			owner = subflow_req->msk;
-			if (!owner)
+			if (!owner) {
+				subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT);
 				goto dispose_child;
+			}
 
 			/* move the msk reference ownership to the subflow */
 			subflow_req->msk = NULL;
@@ -1052,6 +1076,8 @@ static bool subflow_check_data_avail(struct sock *ssk)
 	smp_wmb();
 	ssk->sk_error_report(ssk);
 	tcp_set_state(ssk, TCP_CLOSE);
+	subflow->reset_transient = 0;
+	subflow->reset_reason = MPTCP_RST_EMPTCP;
 	tcp_send_active_reset(ssk, GFP_ATOMIC);
 	subflow->data_avail = 0;
 	return false;

tools/testing/selftests/net/mptcp/diag.sh

@@ -5,8 +5,9 @@ rndh=$(printf %x $sec)-$(mktemp -u XXXXXX)
 ns="ns1-$rndh"
 ksft_skip=4
 test_cnt=1
+timeout_poll=100
+timeout_test=$((timeout_poll * 2 + 1))
 ret=0
-pids=()
 
 flush_pids()
 {
@@ -14,18 +15,14 @@ flush_pids()
 	# give it some time
 	sleep 1.1
 
-	for pid in ${pids[@]}; do
-		[ -d /proc/$pid ] && kill -SIGUSR1 $pid >/dev/null 2>&1
-	done
-	pids=()
+	ip netns pids "${ns}" | xargs --no-run-if-empty kill -SIGUSR1 &>/dev/null
 }
 
 cleanup()
 {
+	ip netns pids "${ns}" | xargs --no-run-if-empty kill -SIGKILL &>/dev/null
+
 	ip netns del $ns
-	for pid in ${pids[@]}; do
-		[ -d /proc/$pid ] && kill -9 $pid >/dev/null 2>&1
-	done
 }
 
 ip -Version > /dev/null 2>&1
@@ -79,39 +76,57 @@ trap cleanup EXIT
 ip netns add $ns
 ip -n $ns link set dev lo up
 
-echo "a" | ip netns exec $ns ./mptcp_connect -p 10000 -l 0.0.0.0 -t 100 >/dev/null &
+echo "a" | \
+	timeout ${timeout_test} \
+		ip netns exec $ns \
+			./mptcp_connect -p 10000 -l -t ${timeout_poll} \
+				0.0.0.0 >/dev/null &
 sleep 0.1
-pids[0]=$!
 chk_msk_nr 0 "no msk on netns creation"
 
-echo "b" | ip netns exec $ns ./mptcp_connect -p 10000 127.0.0.1 -j -t 100 >/dev/null &
+echo "b" | \
+	timeout ${timeout_test} \
+		ip netns exec $ns \
+			./mptcp_connect -p 10000 -j -t ${timeout_poll} \
+				127.0.0.1 >/dev/null &
 sleep 0.1
-pids[1]=$!
 chk_msk_nr 2 "after MPC handshake "
 chk_msk_remote_key_nr 2 "....chk remote_key"
 chk_msk_fallback_nr 0 "....chk no fallback"
 flush_pids
 
 
-echo "a" | ip netns exec $ns ./mptcp_connect -p 10001 -s TCP -l 0.0.0.0 -t 100 >/dev/null &
-pids[0]=$!
+echo "a" | \
+	timeout ${timeout_test} \
+		ip netns exec $ns \
+			./mptcp_connect -p 10001 -l -s TCP -t ${timeout_poll} \
+				0.0.0.0 >/dev/null &
 sleep 0.1
-echo "b" | ip netns exec $ns ./mptcp_connect -p 10001 127.0.0.1 -j -t 100 >/dev/null &
-pids[1]=$!
+echo "b" | \
+	timeout ${timeout_test} \
+		ip netns exec $ns \
+			./mptcp_connect -p 10001 -j -t ${timeout_poll} \
+				127.0.0.1 >/dev/null &
 sleep 0.1
 chk_msk_fallback_nr 1 "check fallback"
 flush_pids
 
 NR_CLIENTS=100
 for I in `seq 1 $NR_CLIENTS`; do
-	echo "a" | ip netns exec $ns ./mptcp_connect -p $((I+10001)) -l 0.0.0.0 -t 100 -w 10 >/dev/null  &
-	pids[$((I*2))]=$!
+	echo "a" | \
+		timeout ${timeout_test} \
+			ip netns exec $ns \
+				./mptcp_connect -p $((I+10001)) -l -w 10 \
+					-t ${timeout_poll} 0.0.0.0 >/dev/null &
 done
 sleep 0.1
 
 for I in `seq 1 $NR_CLIENTS`; do
-	echo "b" | ip netns exec $ns ./mptcp_connect -p $((I+10001)) 127.0.0.1 -t 100 -w 10 >/dev/null &
-	pids[$((I*2 + 1))]=$!
+	echo "b" | \
+		timeout ${timeout_test} \
+			ip netns exec $ns \
+				./mptcp_connect -p $((I+10001)) -w 10 \
+					-t ${timeout_poll} 127.0.0.1 >/dev/null &
 done
 sleep 1.5
 

tools/testing/selftests/net/mptcp/mptcp_connect.sh

@@ -11,7 +11,8 @@ cin=""
 cout=""
 ksft_skip=4
 capture=false
-timeout=30
+timeout_poll=30
+timeout_test=$((timeout_poll * 2 + 1))
 ipv6=true
 ethtool_random_on=true
 tc_delay="$((RANDOM%50))"
@@ -273,7 +274,7 @@ check_mptcp_disabled()
 	ip netns exec ${disabled_ns} sysctl -q net.mptcp.enabled=0
 
 	local err=0
-	LANG=C ip netns exec ${disabled_ns} ./mptcp_connect -t $timeout -p 10000 -s MPTCP 127.0.0.1 < "$cin" 2>&1 | \
+	LANG=C ip netns exec ${disabled_ns} ./mptcp_connect -p 10000 -s MPTCP 127.0.0.1 < "$cin" 2>&1 | \
 		grep -q "^socket: Protocol not available$" && err=1
 	ip netns delete ${disabled_ns}
 
@@ -425,19 +426,32 @@ do_transfer()
 		sleep 1
 	fi
 
+	NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \
+		nstat -n
+	if [ ${listener_ns} != ${connector_ns} ]; then
+		NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \
+			nstat -n
+	fi
+
 	local stat_synrx_last_l=$(get_mib_counter "${listener_ns}" "MPTcpExtMPCapableSYNRX")
 	local stat_ackrx_last_l=$(get_mib_counter "${listener_ns}" "MPTcpExtMPCapableACKRX")
 	local stat_cookietx_last=$(get_mib_counter "${listener_ns}" "TcpExtSyncookiesSent")
 	local stat_cookierx_last=$(get_mib_counter "${listener_ns}" "TcpExtSyncookiesRecv")
 
-	ip netns exec ${listener_ns} ./mptcp_connect -t $timeout -l -p $port -s ${srv_proto} $extra_args $local_addr < "$sin" > "$sout" &
+	timeout ${timeout_test} \
+		ip netns exec ${listener_ns} \
+			./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \
+				$extra_args $local_addr < "$sin" > "$sout" &
 	local spid=$!
 
 	wait_local_port_listen "${listener_ns}" "${port}"
 
 	local start
 	start=$(date +%s%3N)
-	ip netns exec ${connector_ns} ./mptcp_connect -t $timeout -p $port -s ${cl_proto} $extra_args $connect_addr < "$cin" > "$cout" &
+	timeout ${timeout_test} \
+		ip netns exec ${connector_ns} \
+			./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \
+				$extra_args $connect_addr < "$cin" > "$cout" &
 	local cpid=$!
 
 	wait $cpid

tools/testing/selftests/net/mptcp/mptcp_join.sh

@@ -8,7 +8,8 @@ cin=""
 cinsent=""
 cout=""
 ksft_skip=4
-timeout=30
+timeout_poll=30
+timeout_test=$((timeout_poll * 2 + 1))
 mptcp_connect=""
 capture=0
 do_all_tests=1
@@ -77,6 +78,7 @@ cleanup_partial()
 
 	for netns in "$ns1" "$ns2"; do
 		ip netns del $netns
+		rm -f /tmp/$netns.{nstat,out}
 	done
 }
 
@@ -232,6 +234,11 @@ do_transfer()
 		sleep 1
 	fi
 
+	NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \
+		nstat -n
+	NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \
+		nstat -n
+
 	if [ $speed = "fast" ]; then
 		mptcp_connect="./mptcp_connect -j"
 	elif [ $speed = "slow" ]; then
@@ -247,17 +254,26 @@ do_transfer()
 		local_addr="0.0.0.0"
 	fi
 
-	ip netns exec ${listener_ns} $mptcp_connect -t $timeout -l -p $port \
-		-s ${srv_proto} ${local_addr} < "$sin" > "$sout" &
+	timeout ${timeout_test} \
+		ip netns exec ${listener_ns} \
+			$mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \
+				${local_addr} < "$sin" > "$sout" &
 	spid=$!
 
 	sleep 1
 
 	if [ "$test_link_fail" -eq 0 ];then
-		ip netns exec ${connector_ns} $mptcp_connect -t $timeout -p $port -s ${cl_proto} $connect_addr < "$cin" > "$cout" &
+		timeout ${timeout_test} \
+			ip netns exec ${connector_ns} \
+				$mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \
+					$connect_addr < "$cin" > "$cout" &
 	else
-		( cat "$cin" ; sleep 2; link_failure $listener_ns ; cat "$cin" ) | tee "$cinsent" | \
-		ip netns exec ${connector_ns} $mptcp_connect -t $timeout -p $port -s ${cl_proto} $connect_addr > "$cout" &
+		( cat "$cin" ; sleep 2; link_failure $listener_ns ; cat "$cin" ) | \
+			tee "$cinsent" | \
+			timeout ${timeout_test} \
+				ip netns exec ${connector_ns} \
+					$mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \
+						$connect_addr > "$cout" &
 	fi
 	cpid=$!
 
@@ -373,12 +389,19 @@ do_transfer()
 	    kill $cappid
 	fi
 
+	NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \
+		nstat | grep Tcp > /tmp/${listener_ns}.out
+	NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \
+		nstat | grep Tcp > /tmp/${connector_ns}.out
+
 	if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then
 		echo " client exit code $retc, server $rets" 1>&2
 		echo -e "\nnetns ${listener_ns} socket stat for ${port}:" 1>&2
-		ip netns exec ${listener_ns} ss -nita 1>&2 -o "sport = :$port"
+		ip netns exec ${listener_ns} ss -Menita 1>&2 -o "sport = :$port"
+		cat /tmp/${listener_ns}.out
 		echo -e "\nnetns ${connector_ns} socket stat for ${port}:" 1>&2
-		ip netns exec ${connector_ns} ss -nita 1>&2 -o "dport = :$port"
+		ip netns exec ${connector_ns} ss -Menita 1>&2 -o "dport = :$port"
+		cat /tmp/${connector_ns}.out
 
 		cat "$capout"
 		ret=1

tools/testing/selftests/net/mptcp/simult_flows.sh

@@ -7,7 +7,8 @@ ns2="ns2-$rndh"
 ns3="ns3-$rndh"
 capture=false
 ksft_skip=4
-timeout=30
+timeout_poll=30
+timeout_test=$((timeout_poll * 2 + 1))
 test_cnt=1
 ret=0
 bail=0
@@ -157,14 +158,20 @@ do_transfer()
 		sleep 1
 	fi
 
-	ip netns exec ${ns3} ./mptcp_connect -jt $timeout -l -p $port 0.0.0.0 < "$sin" > "$sout" &
+	timeout ${timeout_test} \
+		ip netns exec ${ns3} \
+			./mptcp_connect -jt ${timeout_poll} -l -p $port \
+				0.0.0.0 < "$sin" > "$sout" &
 	local spid=$!
 
 	wait_local_port_listen "${ns3}" "${port}"
 
 	local start
 	start=$(date +%s%3N)
-	ip netns exec ${ns1} ./mptcp_connect -jt $timeout -p $port 10.0.3.3 < "$cin" > "$cout" &
+	timeout ${timeout_test} \
+		ip netns exec ${ns1} \
+			./mptcp_connect -jt ${timeout_poll} -p $port \
+				10.0.3.3 < "$cin" > "$cout" &
 	local cpid=$!
 
 	wait $cpid