openmptcprouter/root/target/linux/generic/hack-5.4/693-mptcp-fixes.patch

diff -aurN '--exclude=.git' mptcp-mptcp_trunk/include/net/mptcp.h mptcp/include/net/mptcp.h
--- mptcp-mptcp_trunk/include/net/mptcp.h	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/include/net/mptcp.h	2020-05-14 15:15:39.929940266 +0200
@@ -102,7 +102,8 @@
 
 	u8				loc_id;
 	u8				rem_id; /* Address-id in the MP_JOIN */
-	u8				dss_csum:1,
+	u16				dss_csum:1,
+					rem_key_set:1,
 					is_sub:1, /* Is this a new subflow? */
 					low_prio:1, /* Interface set to low-prio? */
 					rcv_low_prio:1,
@@ -240,7 +241,6 @@
 	struct module	*owner;
 };
 
-#define MPTCP_SCHED_NAME_MAX 16
 struct mptcp_sched_ops {
 	struct list_head list;
 
@@ -272,6 +272,8 @@
 	u32 rcv_high_order[2];
 
 	u16	send_infinite_mapping:1,
+		send_mptcpv1_mpcapable:1,
+		rem_key_set:1,
 		in_time_wait:1,
 		list_rcvd:1, /* XXX TO REMOVE */
 		addr_signal:1, /* Path-manager wants us to call addr_signal */
@@ -354,6 +356,16 @@
 #define MPTCP_SUB_LEN_CAPABLE_ACK		20
 #define MPTCP_SUB_LEN_CAPABLE_ACK_ALIGN		20
 
+#define MPTCPV1_SUB_LEN_CAPABLE_SYN		4
+#define MPTCPV1_SUB_LEN_CAPABLE_SYN_ALIGN	4
+#define MPTCPV1_SUB_LEN_CAPABLE_SYNACK		12
+#define MPTCPV1_SUB_LEN_CAPABLE_SYNACK_ALIGN	12
+#define MPTCPV1_SUB_LEN_CAPABLE_ACK		20
+#define MPTCPV1_SUB_LEN_CAPABLE_ACK_ALIGN	20
+#define MPTCPV1_SUB_LEN_CAPABLE_DATA		22
+#define MPTCPV1_SUB_LEN_CAPABLE_DATA_CSUM	22
+#define MPTCPV1_SUB_LEN_CAPABLE_DATA_ALIGN	24
+
 #define MPTCP_SUB_JOIN			1
 #define MPTCP_SUB_LEN_JOIN_SYN		12
 #define MPTCP_SUB_LEN_JOIN_SYN_ALIGN	12
@@ -450,14 +462,15 @@
 #define MPTCPHDR_SEQ		0x01 /* DSS.M option is present */
 #define MPTCPHDR_FIN		0x02 /* DSS.F option is present */
 #define MPTCPHDR_SEQ64_INDEX	0x04 /* index of seq in mpcb->snd_high_order */
+#define MPTCPHDR_MPC_DATA	0x08
 /* MPTCP flags: RX only */
-#define MPTCPHDR_ACK		0x08
-#define MPTCPHDR_SEQ64_SET	0x10 /* Did we received a 64-bit seq number?  */
-#define MPTCPHDR_SEQ64_OFO	0x20 /* Is it not in our circular array? */
-#define MPTCPHDR_DSS_CSUM	0x40
+#define MPTCPHDR_ACK		0x10
+#define MPTCPHDR_SEQ64_SET	0x20 /* Did we received a 64-bit seq number?  */
+#define MPTCPHDR_SEQ64_OFO	0x40 /* Is it not in our circular array? */
+#define MPTCPHDR_DSS_CSUM	0x80
 /* MPTCP flags: TX only */
-#define MPTCPHDR_INF		0x08
-#define MPTCP_REINJECT		0x10 /* Did we reinject this segment? */
+#define MPTCPHDR_INF		0x10
+#define MPTCP_REINJECT		0x20 /* Did we reinject this segment? */
 
 struct mptcp_option {
 	__u8	kind;
@@ -800,10 +813,11 @@
 void mptcp_close(struct sock *meta_sk, long timeout);
 bool mptcp_doit(struct sock *sk);
 int mptcp_create_master_sk(struct sock *meta_sk, __u64 remote_key,
-			   __u8 mptcp_ver, u32 window);
+			   int rem_key_set, __u8 mptcp_ver, u32 window);
 int mptcp_check_req_fastopen(struct sock *child, struct request_sock *req);
 int mptcp_check_req_master(struct sock *sk, struct sock *child,
 			   struct request_sock *req, const struct sk_buff *skb,
+			   const struct mptcp_options_received *mopt,
 			   int drop, u32 tsoff);
 struct sock *mptcp_check_req_child(struct sock *meta_sk,
 				   struct sock *child,
@@ -816,8 +830,8 @@
 				 int wscale_ok, __u8 *rcv_wscale,
 				 __u32 init_rcv_wnd);
 unsigned int mptcp_current_mss(struct sock *meta_sk);
-void mptcp_hmac_sha1(const u8 *key_1, const u8 *key_2, u32 *hash_out,
-		     int arg_num, ...);
+void mptcp_hmac(u8 ver, const u8 *key_1, const u8 *key_2, u32 *hash_out,
+		int arg_num, ...);
 void mptcp_clean_rtx_infinite(const struct sk_buff *skb, struct sock *sk);
 void mptcp_fin(struct sock *meta_sk);
 void mptcp_meta_retransmit_timer(struct sock *meta_sk);
@@ -827,6 +841,8 @@
 void mptcp_sub_close(struct sock *sk, unsigned long delay);
 struct sock *mptcp_select_ack_sock(const struct sock *meta_sk);
 void mptcp_prepare_for_backlog(struct sock *sk, struct sk_buff *skb);
+void mptcp_initialize_recv_vars(struct tcp_sock *meta_tp, struct mptcp_cb *mpcb,
+				__u64 remote_key);
 int mptcp_backlog_rcv(struct sock *meta_sk, struct sk_buff *skb);
 void mptcp_ack_handler(struct timer_list *t);
 bool mptcp_check_rtt(const struct tcp_sock *tp, int time);
@@ -982,6 +998,11 @@
 	}
 }
 
+static inline bool mptcp_is_data_mpcapable(const struct sk_buff *skb)
+{
+	return TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_MPC_DATA;
+}
+
 static inline bool mptcp_is_data_seq(const struct sk_buff *skb)
 {
 	return TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_SEQ;
@@ -1399,6 +1420,7 @@
 					 const struct sock *child,
 					 const struct request_sock *req,
 					 const struct sk_buff *skb,
+					 const struct mptcp_options_received *mopt,
 					 int drop,
 					 u32 tsoff)
 {
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/include/net/tcp.h mptcp/include/net/tcp.h
--- mptcp-mptcp_trunk/include/net/tcp.h	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/include/net/tcp.h	2020-05-14 15:15:27.126152589 +0200
@@ -343,7 +343,6 @@
 struct mptcp_options_received;
 
 void tcp_cleanup_rbuf(struct sock *sk, int copied);
-void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited);
 int tcp_close_state(struct sock *sk);
 void tcp_minshall_update(struct tcp_sock *tp, unsigned int mss_now,
 			 const struct sk_buff *skb);
@@ -583,6 +582,7 @@
 /* From syncookies.c */
 struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 				 struct request_sock *req,
+				 const struct mptcp_options_received *mopt,
 				 struct dst_entry *dst, u32 tsoff);
 int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th,
 		      u32 cookie);
@@ -2126,7 +2126,6 @@
 	void (*retransmit_timer)(struct sock *sk);
 	void (*time_wait)(struct sock *sk, int state, int timeo);
 	void (*cleanup_rbuf)(struct sock *sk, int copied);
-	void (*cwnd_validate)(struct sock *sk, bool is_cwnd_limited);
 	int (*set_cong_ctrl)(struct sock *sk, const char *name, bool load,
 			     bool reinit, bool cap_net_admin);
 };
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/ipv4/syncookies.c mptcp/net/ipv4/syncookies.c
--- mptcp-mptcp_trunk/net/ipv4/syncookies.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/ipv4/syncookies.c	2020-05-14 15:15:27.126152589 +0200
@@ -203,6 +203,7 @@
 
 struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 				 struct request_sock *req,
+				 const struct mptcp_options_received *mopt,
 				 struct dst_entry *dst, u32 tsoff)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
@@ -219,7 +220,7 @@
 	if (!child)
 		goto listen_overflow;
 
-	ret = mptcp_check_req_master(sk, child, req, skb, 0, tsoff);
+	ret = mptcp_check_req_master(sk, child, req, skb, mopt, 0, tsoff);
 	if (ret < 0)
 		return NULL;
 
@@ -428,7 +429,7 @@
 	ireq->rcv_wscale  = rcv_wscale;
 	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, sock_net(sk), &rt->dst);
 
-	ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst, tsoff);
+	ret = tcp_get_cookie_sock(sk, skb, req, &mopt, &rt->dst, tsoff);
 	/* ip_queue_xmit() depends on our flow being setup
 	 * Normal sockets get it right from inet_csk_route_child_sock()
 	 */
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/ipv4/tcp.c mptcp/net/ipv4/tcp.c
--- mptcp-mptcp_trunk/net/ipv4/tcp.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/ipv4/tcp.c	2020-05-11 09:40:04.803741955 +0200
@@ -415,7 +415,6 @@
 	.retransmit_timer		= tcp_retransmit_timer,
 	.time_wait			= tcp_time_wait,
 	.cleanup_rbuf			= tcp_cleanup_rbuf,
-	.cwnd_validate			= tcp_cwnd_validate,
 	.set_cong_ctrl			= __tcp_set_congestion_control,
 };
 
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/ipv4/tcp_minisocks.c mptcp/net/ipv4/tcp_minisocks.c
--- mptcp-mptcp_trunk/net/ipv4/tcp_minisocks.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/ipv4/tcp_minisocks.c	2020-05-14 15:15:27.138152390 +0200
@@ -828,7 +828,7 @@
 		goto listen_overflow;
 
 	if (own_req && !is_meta_sk(sk)) {
-		int ret = mptcp_check_req_master(sk, child, req, skb, 1, 0);
+		int ret = mptcp_check_req_master(sk, child, req, skb, &mopt, 1, 0);
 		if (ret < 0)
 			goto listen_overflow;
 
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/ipv4/tcp_output.c mptcp/net/ipv4/tcp_output.c
--- mptcp-mptcp_trunk/net/ipv4/tcp_output.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/ipv4/tcp_output.c	2020-05-11 09:40:04.803741955 +0200
@@ -825,8 +825,8 @@
 		if (mptcp(tp))
 			tcp_tsq_write(meta_sk);
 	} else {
-		if (!test_and_set_bit(TCP_TSQ_DEFERRED, &meta_sk->sk_tsq_flags))
-			sock_hold(meta_sk);
+		if (!test_and_set_bit(TCP_TSQ_DEFERRED, &sk->sk_tsq_flags))
+			sock_hold(sk);
 
 		if ((mptcp(tp)) && (sk->sk_state != TCP_CLOSE))
 			mptcp_tsq_flags(sk);
@@ -1672,7 +1672,7 @@
 	tp->snd_cwnd_stamp = tcp_jiffies32;
 }
 
-void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited)
+static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited)
 {
 	const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops;
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -2512,8 +2512,7 @@
 		if (push_one != 2)
 			tcp_schedule_loss_probe(sk, false);
 		is_cwnd_limited |= (tcp_packets_in_flight(tp) >= tp->snd_cwnd);
-		if (tp->ops->cwnd_validate)
-			tp->ops->cwnd_validate(sk, is_cwnd_limited);
+		tcp_cwnd_validate(sk, is_cwnd_limited);
 		return false;
 	}
 	return !tp->packets_out && !tcp_write_queue_empty(sk);
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/ipv6/syncookies.c mptcp/net/ipv6/syncookies.c
--- mptcp-mptcp_trunk/net/ipv6/syncookies.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/ipv6/syncookies.c	2020-05-14 15:15:27.142152325 +0200
@@ -267,7 +267,7 @@
 	ireq->rcv_wscale = rcv_wscale;
 	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, sock_net(sk), dst);
 
-	ret = tcp_get_cookie_sock(sk, skb, req, dst, tsoff);
+	ret = tcp_get_cookie_sock(sk, skb, req, &mopt, dst, tsoff);
 out:
 	return ret;
 out_free:
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_ctrl.c mptcp/net/mptcp/mptcp_ctrl.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_ctrl.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_ctrl.c	2020-05-14 15:15:39.953939868 +0200
@@ -27,6 +27,8 @@
  *      2 of the License, or (at your option) any later version.
  */
 
+#include <crypto/sha.h>
+
 #include <net/inet_common.h>
 #include <net/inet6_hashtables.h>
 #include <net/ipv6.h>
@@ -77,7 +79,7 @@
 struct static_key mptcp_static_key = STATIC_KEY_INIT_FALSE;
 EXPORT_SYMBOL(mptcp_static_key);
 
-static void mptcp_key_sha1(u64 key, u32 *token, u64 *idsn);
+static void mptcp_key_hash(u8 version, u64 key, u32 *token, u64 *idsn);
 
 static int proc_mptcp_path_manager(struct ctl_table *ctl, int write,
 				   void __user *buffer, size_t *lenp,
@@ -286,7 +288,7 @@
 #endif
 	}
 
-	mptcp_key_sha1(mtreq->mptcp_loc_key, &mtreq->mptcp_loc_token, NULL);
+	mptcp_key_hash(mtreq->mptcp_ver, mtreq->mptcp_loc_key, &mtreq->mptcp_loc_token, NULL);
 }
 
 /* New MPTCP-connection request, prepare a new token for the meta-socket that
@@ -319,7 +321,11 @@
 	spin_unlock(&mptcp_tk_hashlock);
 	local_bh_enable();
 	rcu_read_unlock();
-	mtreq->mptcp_rem_key = mopt->mptcp_sender_key;
+
+	if (mtreq->mptcp_ver == MPTCP_VERSION_0) {
+		mtreq->mptcp_rem_key = mopt->mptcp_sender_key;
+		mtreq->rem_key_set = 1;
+	}
 }
 
 static int mptcp_reqsk_new_cookie(struct request_sock *req,
@@ -355,7 +361,10 @@
 	local_bh_enable();
 	rcu_read_unlock();
 
-	mtreq->mptcp_rem_key = mopt->mptcp_sender_key;
+	if (mtreq->mptcp_ver == MPTCP_VERSION_0) {
+		mtreq->mptcp_rem_key = mopt->mptcp_sender_key;
+		mtreq->rem_key_set = 1;
+	}
 
 	return true;
 }
@@ -380,8 +389,7 @@
 						     mptcp_seed++);
 #endif
 
-	mptcp_key_sha1(tp->mptcp_loc_key,
-		       &tp->mptcp_loc_token, NULL);
+	mptcp_key_hash(tp->mptcp_ver, tp->mptcp_loc_key, &tp->mptcp_loc_token, NULL);
 }
 
 #ifdef CONFIG_JUMP_LABEL
@@ -835,6 +843,71 @@
 siphash_key_t mptcp_secret __read_mostly;
 u32 mptcp_seed = 0;
 
+#define SHA256_DIGEST_WORDS (SHA256_DIGEST_SIZE / 4)
+
+static void mptcp_key_sha256(const u64 key, u32 *token, u64 *idsn)
+{
+	u32 mptcp_hashed_key[SHA256_DIGEST_WORDS];
+	struct sha256_state state;
+
+	sha256_init(&state);
+	sha256_update(&state, (const u8 *)&key, sizeof(key));
+	sha256_final(&state, (u8 *)mptcp_hashed_key);
+
+	if (token)
+		*token = mptcp_hashed_key[0];
+	if (idsn)
+		*idsn = ntohll(*((__be64 *)&mptcp_hashed_key[6]));
+}
+
+static void mptcp_hmac_sha256(const u8 *key_1, const u8 *key_2, u32 *hash_out,
+			      int arg_num, va_list list)
+{
+	u8 input[SHA256_BLOCK_SIZE + SHA256_DIGEST_SIZE];
+	__be32 output[SHA256_DIGEST_WORDS];
+	struct sha256_state state;
+	int index, msg_length;
+	int length = 0;
+	u8 *msg;
+	int i;
+
+	/* Generate key xored with ipad */
+	memset(input, 0x36, SHA256_BLOCK_SIZE);
+	for (i = 0; i < 8; i++)
+		input[i] ^= key_1[i];
+	for (i = 0; i < 8; i++)
+		input[i + 8] ^= key_2[i];
+
+	index = SHA256_BLOCK_SIZE;
+	msg_length = 0;
+	for (i = 0; i < arg_num; i++) {
+		length = va_arg(list, int);
+		msg = va_arg(list, u8 *);
+		BUG_ON(index + length >= sizeof(input)); /* Message is too long */
+		memcpy(&input[index], msg, length);
+		index += length;
+		msg_length += length;
+	}
+
+	sha256_init(&state);
+	sha256_update(&state, input, SHA256_BLOCK_SIZE + msg_length);
+	sha256_final(&state, &input[SHA256_BLOCK_SIZE]);
+
+	/* Prepare second part of hmac */
+	memset(input, 0x5C, SHA256_BLOCK_SIZE);
+	for (i = 0; i < 8; i++)
+		input[i] ^= key_1[i];
+	for (i = 0; i < 8; i++)
+		input[i + 8] ^= key_2[i];
+
+	sha256_init(&state);
+	sha256_update(&state, input, sizeof(input));
+	sha256_final(&state, (u8 *)output);
+
+	for (i = 0; i < 5; i++)
+		hash_out[i] = output[i];
+}
+
 static void mptcp_key_sha1(u64 key, u32 *token, u64 *idsn)
 {
 	u32 workspace[SHA_WORKSPACE_WORDS];
@@ -864,8 +937,16 @@
 		*idsn = ntohll(*((__be64 *)&mptcp_hashed_key[3]));
 }
 
-void mptcp_hmac_sha1(const u8 *key_1, const u8 *key_2, u32 *hash_out,
-		     int arg_num, ...)
+static void mptcp_key_hash(u8 version, u64 key, u32 *token, u64 *idsn)
+{
+	if (version == MPTCP_VERSION_0)
+		mptcp_key_sha1(key, token, idsn);
+	else if (version >= MPTCP_VERSION_1)
+		mptcp_key_sha256(key, token, idsn);
+}
+
+static void mptcp_hmac_sha1(const u8 *key_1, const u8 *key_2, u32 *hash_out,
+			    int arg_num, va_list list)
 {
 	u32 workspace[SHA_WORKSPACE_WORDS];
 	u8 input[128]; /* 2 512-bit blocks */
@@ -873,7 +954,6 @@
 	int index;
 	int length;
 	u8 *msg;
-	va_list list;
 
 	memset(workspace, 0, sizeof(workspace));
 
@@ -884,7 +964,6 @@
 	for (i = 0; i < 8; i++)
 		input[i + 8] ^= key_2[i];
 
-	va_start(list, arg_num);
 	index = 64;
 	for (i = 0; i < arg_num; i++) {
 		length = va_arg(list, int);
@@ -893,7 +972,6 @@
 		memcpy(&input[index], msg, length);
 		index += length;
 	}
-	va_end(list);
 
 	input[index] = 0x80; /* Padding: First bit after message = 1 */
 	memset(&input[index + 1], 0, (126 - index));
@@ -936,7 +1014,20 @@
 	for (i = 0; i < 5; i++)
 		hash_out[i] = (__force u32)cpu_to_be32(hash_out[i]);
 }
-EXPORT_SYMBOL(mptcp_hmac_sha1);
+
+void mptcp_hmac(u8 ver, const u8 *key_1, const u8 *key_2, u32 *hash_out,
+		int arg_num, ...)
+{
+	va_list args;
+
+	va_start(args, arg_num);
+	if (ver == MPTCP_VERSION_0)
+		mptcp_hmac_sha1(key_1, key_2, hash_out, arg_num, args);
+	else if (ver >= MPTCP_VERSION_1)
+		mptcp_hmac_sha256(key_1, key_2, hash_out, arg_num, args);
+	va_end(args);
+}
+EXPORT_SYMBOL(mptcp_hmac);
 
 static void mptcp_mpcb_inherit_sockopts(struct sock *meta_sk, struct sock *master_sk)
 {
@@ -1169,14 +1260,33 @@
 	.set_cong_ctrl                  = __tcp_set_congestion_control,
 };
 
+void mptcp_initialize_recv_vars(struct tcp_sock *meta_tp, struct mptcp_cb *mpcb,
+				__u64 remote_key)
+{
+	u64 idsn;
+
+	mpcb->mptcp_rem_key = remote_key;
+	mpcb->rem_key_set = 1;
+	mptcp_key_hash(mpcb->mptcp_ver, mpcb->mptcp_rem_key, &mpcb->mptcp_rem_token, &idsn);
+
+	idsn++;
+	mpcb->rcv_high_order[0] = idsn >> 32;
+	mpcb->rcv_high_order[1] = mpcb->rcv_high_order[0] + 1;
+	meta_tp->copied_seq = (u32)idsn;
+	meta_tp->rcv_nxt = (u32)idsn;
+	meta_tp->rcv_wup = (u32)idsn;
+
+	meta_tp->snd_wl1 = meta_tp->rcv_nxt - 1;
+}
+
 static int mptcp_alloc_mpcb(struct sock *meta_sk, __u64 remote_key,
-			    __u8 mptcp_ver, u32 window)
+			    int rem_key_set, __u8 mptcp_ver, u32 window)
 {
 	struct mptcp_cb *mpcb;
 	struct sock *master_sk;
 	struct inet_connection_sock *meta_icsk = inet_csk(meta_sk);
 	struct tcp_sock *master_tp, *meta_tp = tcp_sk(meta_sk);
-	u64 snd_idsn, rcv_idsn;
+	u64 snd_idsn;
 
 	dst_release(meta_sk->sk_rx_dst);
 	meta_sk->sk_rx_dst = NULL;
@@ -1204,17 +1314,11 @@
 	mpcb->mptcp_loc_token = meta_tp->mptcp_loc_token;
 
 	/* Generate Initial data-sequence-numbers */
-	mptcp_key_sha1(mpcb->mptcp_loc_key, NULL, &snd_idsn);
+	mptcp_key_hash(mpcb->mptcp_ver, mpcb->mptcp_loc_key, NULL, &snd_idsn);
 	snd_idsn++;
 	mpcb->snd_high_order[0] = snd_idsn >> 32;
 	mpcb->snd_high_order[1] = mpcb->snd_high_order[0] - 1;
 
-	mpcb->mptcp_rem_key = remote_key;
-	mptcp_key_sha1(mpcb->mptcp_rem_key, &mpcb->mptcp_rem_token, &rcv_idsn);
-	rcv_idsn++;
-	mpcb->rcv_high_order[0] = rcv_idsn >> 32;
-	mpcb->rcv_high_order[1] = mpcb->rcv_high_order[0] + 1;
-
 	mpcb->meta_sk = meta_sk;
 	mpcb->master_sk = master_sk;
 
@@ -1326,11 +1430,9 @@
 	meta_tp->pushed_seq = meta_tp->write_seq;
 	meta_tp->snd_up = meta_tp->write_seq;
 
-	meta_tp->copied_seq = (u32)rcv_idsn;
-	meta_tp->rcv_nxt = (u32)rcv_idsn;
-	meta_tp->rcv_wup = (u32)rcv_idsn;
+	if (rem_key_set)
+		mptcp_initialize_recv_vars(meta_tp, mpcb, remote_key);
 
-	meta_tp->snd_wl1 = meta_tp->rcv_nxt - 1;
 	meta_tp->snd_wnd = window;
 	meta_tp->retrans_stamp = 0; /* Set in tcp_connect() */
 
@@ -2077,12 +2179,12 @@
 }
 
 int mptcp_create_master_sk(struct sock *meta_sk, __u64 remote_key,
-			   __u8 mptcp_ver, u32 window)
+			   int rem_key_set, __u8 mptcp_ver, u32 window)
 {
 	struct tcp_sock *master_tp;
 	struct sock *master_sk;
 
-	if (mptcp_alloc_mpcb(meta_sk, remote_key, mptcp_ver, window))
+	if (mptcp_alloc_mpcb(meta_sk, remote_key, rem_key_set, mptcp_ver, window))
 		goto err_alloc_mpcb;
 
 	master_sk = tcp_sk(meta_sk)->mpcb->master_sk;
@@ -2110,6 +2212,7 @@
 }
 
 static int __mptcp_check_req_master(struct sock *child,
+				    const struct mptcp_options_received *mopt,
 				    struct request_sock *req)
 {
 	struct tcp_sock *child_tp = tcp_sk(child);
@@ -2121,6 +2224,8 @@
 	if (!inet_rsk(req)->mptcp_rqsk)
 		return 1;
 
+	mtreq = mptcp_rsk(req);
+
 	if (!inet_rsk(req)->saw_mpc) {
 		/* Fallback to regular TCP, because we saw one SYN without
 		 * MP_CAPABLE. In tcp_check_req we continue the regular path.
@@ -2132,15 +2237,21 @@
 		return 1;
 	}
 
+	/* mopt can be NULL when coming from FAST-OPEN */
+	if (mopt && mopt->saw_mpc && mtreq->mptcp_ver == MPTCP_VERSION_1) {
+		mtreq->mptcp_rem_key = mopt->mptcp_sender_key;
+		mtreq->rem_key_set = 1;
+	}
+
 	MPTCP_INC_STATS(sock_net(meta_sk), MPTCP_MIB_MPCAPABLEPASSIVEACK);
 
 	/* Just set this values to pass them to mptcp_alloc_mpcb */
-	mtreq = mptcp_rsk(req);
 	child_tp->mptcp_loc_key = mtreq->mptcp_loc_key;
 	child_tp->mptcp_loc_token = mtreq->mptcp_loc_token;
 
 	if (mptcp_create_master_sk(meta_sk, mtreq->mptcp_rem_key,
-				   mtreq->mptcp_ver, child_tp->snd_wnd)) {
+				   mtreq->rem_key_set, mtreq->mptcp_ver,
+				   child_tp->snd_wnd)) {
 		inet_csk_prepare_forced_close(meta_sk);
 		tcp_done(meta_sk);
 
@@ -2175,7 +2286,7 @@
 	u32 new_mapping;
 	int ret;
 
-	ret = __mptcp_check_req_master(child, req);
+	ret = __mptcp_check_req_master(child, NULL, req);
 	if (ret)
 		return ret;
 
@@ -2218,12 +2329,13 @@
 
 int mptcp_check_req_master(struct sock *sk, struct sock *child,
 			   struct request_sock *req, const struct sk_buff *skb,
+			   const struct mptcp_options_received *mopt,
 			   int drop, u32 tsoff)
 {
 	struct sock *meta_sk = child;
 	int ret;
 
-	ret = __mptcp_check_req_master(child, req);
+	ret = __mptcp_check_req_master(child, mopt, req);
 	if (ret)
 		return ret;
 	child = tcp_sk(child)->mpcb->master_sk;
@@ -2281,11 +2393,10 @@
 		goto teardown;
 	}
 
-	mptcp_hmac_sha1((u8 *)&mpcb->mptcp_rem_key,
-			(u8 *)&mpcb->mptcp_loc_key,
-			(u32 *)hash_mac_check, 2,
-			4, (u8 *)&mtreq->mptcp_rem_nonce,
-			4, (u8 *)&mtreq->mptcp_loc_nonce);
+	mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_rem_key,
+		   (u8 *)&mpcb->mptcp_loc_key, (u32 *)hash_mac_check, 2,
+		   4, (u8 *)&mtreq->mptcp_rem_nonce,
+		   4, (u8 *)&mtreq->mptcp_loc_nonce);
 
 	if (memcmp(hash_mac_check, (char *)&mopt->mptcp_recv_mac, 20)) {
 		MPTCP_INC_STATS(sock_net(meta_sk), MPTCP_MIB_JOINACKMAC);
@@ -2547,11 +2658,10 @@
 
 	mtreq->mptcp_rem_nonce = mopt.mptcp_recv_nonce;
 
-	mptcp_hmac_sha1((u8 *)&mpcb->mptcp_loc_key,
-			(u8 *)&mpcb->mptcp_rem_key,
-			(u32 *)mptcp_hash_mac, 2,
-			4, (u8 *)&mtreq->mptcp_loc_nonce,
-			4, (u8 *)&mtreq->mptcp_rem_nonce);
+	mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_loc_key,
+		   (u8 *)&mpcb->mptcp_rem_key, (u32 *)mptcp_hash_mac, 2,
+		   4, (u8 *)&mtreq->mptcp_loc_nonce,
+		   4, (u8 *)&mtreq->mptcp_rem_nonce);
 	mtreq->mptcp_hash_tmac = *(u64 *)mptcp_hash_mac;
 
 	mtreq->rem_id = mopt.rem_id;
@@ -2591,11 +2701,13 @@
 	/* Absolutely need to always initialize this. */
 	mtreq->hash_entry.pprev = NULL;
 
+	mtreq->mptcp_ver = mopt->mptcp_ver;
 	mtreq->mptcp_rem_key = mopt->mptcp_sender_key;
 	mtreq->mptcp_loc_key = mopt->mptcp_receiver_key;
+	mtreq->rem_key_set = 1;
 
 	/* Generate the token */
-	mptcp_key_sha1(mtreq->mptcp_loc_key, &mtreq->mptcp_loc_token, NULL);
+	mptcp_key_hash(mtreq->mptcp_ver, mtreq->mptcp_loc_key, &mtreq->mptcp_loc_token, NULL);
 
 	rcu_read_lock();
 	local_bh_disable();
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_fullmesh.c mptcp/net/mptcp/mptcp_fullmesh.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_fullmesh.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_fullmesh.c	2020-05-14 15:15:39.957939801 +0200
@@ -1596,11 +1596,10 @@
 			u8 no_key[8];
 
 			*(u64 *)no_key = 0;
-			mptcp_hmac_sha1((u8 *)&mpcb->mptcp_loc_key,
-					(u8 *)no_key,
-					(u32 *)mptcp_hash_mac, 2,
-					1, (u8 *)&mptcp_local->locaddr4[ind].loc4_id,
-					4, (u8 *)&opts->add_addr4.addr.s_addr);
+			mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_loc_key,
+				   (u8 *)no_key, (u32 *)mptcp_hash_mac, 2,
+				   1, (u8 *)&mptcp_local->locaddr4[ind].loc4_id,
+				   4, (u8 *)&opts->add_addr4.addr.s_addr);
 			opts->add_addr4.trunc_mac = *(u64 *)mptcp_hash_mac;
 		}
 
@@ -1639,11 +1638,10 @@
 			u8 no_key[8];
 
 			*(u64 *)no_key = 0;
-			mptcp_hmac_sha1((u8 *)&mpcb->mptcp_loc_key,
-					(u8 *)no_key,
-					(u32 *)mptcp_hash_mac, 2,
-					1, (u8 *)&mptcp_local->locaddr6[ind].loc6_id,
-					16, (u8 *)&opts->add_addr6.addr.s6_addr);
+			mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_loc_key,
+				   (u8 *)no_key, (u32 *)mptcp_hash_mac, 2,
+				   1, (u8 *)&mptcp_local->locaddr6[ind].loc6_id,
+				   16, (u8 *)&opts->add_addr6.addr.s6_addr);
 			opts->add_addr6.trunc_mac = *(u64 *)mptcp_hash_mac;
 		}
 
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_input.c mptcp/net/mptcp/mptcp_input.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_input.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_input.c	2020-05-14 15:15:39.965939670 +0200
@@ -176,6 +176,10 @@
 }
 
 /* Inspired by tcp_rcv_state_process */
+/* Returns 0 if processing the packet can continue
+ *	   -1 if connection was closed with an active reset
+ *	   1 if connection was closed and processing should stop.
+ */
 static int mptcp_rcv_state_process(struct sock *meta_sk, struct sock *sk,
 				   const struct sk_buff *skb, u32 data_seq,
 				   u16 data_len)
@@ -216,7 +220,7 @@
 			mptcp_send_active_reset(meta_sk, GFP_ATOMIC);
 			tcp_done(meta_sk);
 			__NET_INC_STATS(sock_net(meta_sk), LINUX_MIB_TCPABORTONDATA);
-			return 1;
+			return -1;
 		}
 
 		tmo = tcp_fin_time(meta_sk);
@@ -259,7 +263,7 @@
 				__NET_INC_STATS(sock_net(meta_sk), LINUX_MIB_TCPABORTONDATA);
 				mptcp_send_active_reset(meta_sk, GFP_ATOMIC);
 				tcp_reset(meta_sk);
-				return 1;
+				return -1;
 			}
 		}
 		break;
@@ -344,6 +348,17 @@
 						sizeof(data_seq), csum_tcp);
 
 			dss_csum_added = 1; /* Just do it once */
+		} else if (mptcp_is_data_mpcapable(tmp) && !dss_csum_added) {
+			u32 offset = skb_transport_offset(tmp) + TCP_SKB_CB(tmp)->dss_off;
+			__be64 data_seq = htonll(tp->mptcp->map_data_seq);
+			__be32 rel_seq = htonl(tp->mptcp->map_subseq - tp->mptcp->rcv_isn);
+
+			csum_tcp = csum_partial(&data_seq, sizeof(data_seq), csum_tcp);
+			csum_tcp = csum_partial(&rel_seq, sizeof(rel_seq), csum_tcp);
+
+			csum_tcp = skb_checksum(tmp, offset, 4, csum_tcp);
+
+			dss_csum_added = 1;
 		}
 		last = tmp;
 		iter++;
@@ -554,11 +569,12 @@
 	 * this segment, this path has to fallback to infinite or be torn down.
 	 */
 	if (!tp->mptcp->fully_established && !mptcp_is_data_seq(skb) &&
+	    !mptcp_is_data_mpcapable(skb) &&
 	    !tp->mptcp->mapping_present && !mpcb->infinite_mapping_rcv) {
-		pr_debug("%s %#x will fallback - pi %d from %pS, seq %u\n",
+		pr_debug("%s %#x will fallback - pi %d from %pS, seq %u mptcp-flags %#x\n",
 			 __func__, mpcb->mptcp_loc_token,
 			 tp->mptcp->path_index, __builtin_return_address(0),
-			 TCP_SKB_CB(skb)->seq);
+			 TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->mptcp_flags);
 
 		if (!is_master_tp(tp)) {
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_FBDATASUB);
@@ -666,25 +682,36 @@
 		return 0;
 	}
 
-	/* No mapping here? Exit - it is either already set or still on its way */
-	if (!mptcp_is_data_seq(skb)) {
-		/* Too many packets without a mapping - this subflow is broken */
+	if (!tp->mptcp->mapping_present && mptcp_is_data_mpcapable(skb)) {
+		__u32 *ptr = (__u32 *)(skb_transport_header(skb) + TCP_SKB_CB(skb)->dss_off);
+
+		sub_seq = 1 + tp->mptcp->rcv_isn;
+		data_seq = meta_tp->rcv_nxt;
+		data_len = get_unaligned_be16(ptr);
+	} else if (!mptcp_is_data_seq(skb)) {
+		/* No mapping here?
+		 * Exit - it is either already set or still on its way
+		 */
 		if (!tp->mptcp->mapping_present &&
 		    tp->rcv_nxt - tp->copied_seq > 65536) {
+			/* Too many packets without a mapping,
+			 * this subflow is broken
+			 */
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_NODSSWINDOW);
 			mptcp_send_reset(sk);
 			return 1;
 		}
 
 		return 0;
+	} else {
+		/* Well, then the DSS-mapping is there. So, read it! */
+		ptr = mptcp_skb_set_data_seq(skb, &data_seq, mpcb);
+		ptr++;
+		sub_seq = get_unaligned_be32(ptr) + tp->mptcp->rcv_isn;
+		ptr++;
+		data_len = get_unaligned_be16(ptr);
 	}
 
-	ptr = mptcp_skb_set_data_seq(skb, &data_seq, mpcb);
-	ptr++;
-	sub_seq = get_unaligned_be32(ptr) + tp->mptcp->rcv_isn;
-	ptr++;
-	data_len = get_unaligned_be16(ptr);
-
 	/* If it's an empty skb with DATA_FIN, sub_seq must get fixed.
 	 * The draft sets it to 0, but we really would like to have the
 	 * real value, to have an easy handling afterwards here in this
@@ -1397,7 +1424,7 @@
 }
 
 /* Handle the DATA_ACK */
-static void mptcp_data_ack(struct sock *sk, const struct sk_buff *skb)
+static int mptcp_data_ack(struct sock *sk, const struct sk_buff *skb)
 {
 	struct sock *meta_sk = mptcp_meta_sk(sk);
 	struct tcp_sock *meta_tp = tcp_sk(meta_sk), *tp = tcp_sk(sk);
@@ -1425,7 +1452,7 @@
 	 * set by mptcp_clean_rtx_infinite.
 	 */
 	if (!(tcb->mptcp_flags & MPTCPHDR_ACK) && !tp->mpcb->infinite_mapping_snd)
-		return;
+		return 0;
 
 	if (unlikely(!tp->mptcp->fully_established) &&
 	    tp->mptcp->snt_isn + 1 != TCP_SKB_CB(skb)->ack_seq)
@@ -1439,7 +1466,7 @@
 	 * processing.
 	 */
 	if (meta_sk->sk_state == TCP_CLOSE)
-		return;
+		return 0;
 
 	/* Get the data_seq */
 	if (mptcp_is_data_seq(skb)) {
@@ -1463,6 +1490,9 @@
 	if (after(data_ack, meta_tp->snd_nxt))
 		goto exit;
 
+	/* First valid DATA_ACK, we can stop sending the special MP_CAPABLE */
+	tp->mpcb->send_mptcpv1_mpcapable = 0;
+
 	/*** Now, update the window  - inspired by tcp_ack_update_window ***/
 	nwin = ntohs(tcp_hdr(skb)->window);
 
@@ -1520,14 +1550,19 @@
 			meta_sk->sk_write_space(meta_sk);
 	}
 
-	if (meta_sk->sk_state != TCP_ESTABLISHED &&
-	    mptcp_rcv_state_process(meta_sk, sk, skb, data_seq, data_len))
-		return;
+	if (meta_sk->sk_state != TCP_ESTABLISHED) {
+		int ret = mptcp_rcv_state_process(meta_sk, sk, skb, data_seq, data_len);
+
+		if (ret < 0)
+			return 1;
+		else if (ret > 0)
+			return 0;
+	}
 
 exit:
 	mptcp_push_pending_frames(meta_sk);
 
-	return;
+	return 0;
 
 no_queue:
 	if (tcp_send_head(meta_sk))
@@ -1535,7 +1570,7 @@
 
 	mptcp_push_pending_frames(meta_sk);
 
-	return;
+	return 0;
 }
 
 void mptcp_clean_rtx_infinite(const struct sk_buff *skb, struct sock *sk)
@@ -1604,6 +1639,7 @@
 			 struct tcp_sock *tp)
 {
 	const struct mptcp_option *mp_opt = (struct mptcp_option *)ptr;
+	const struct tcphdr *th = tcp_hdr(skb);
 
 	/* If the socket is mp-capable we would have a mopt. */
 	if (!mopt)
@@ -1614,9 +1650,21 @@
 	{
 		const struct mp_capable *mpcapable = (struct mp_capable *)ptr;
 
-		if (opsize != MPTCP_SUB_LEN_CAPABLE_SYN &&
-		    opsize != MPTCP_SUB_LEN_CAPABLE_ACK) {
-			mptcp_debug("%s: mp_capable: bad option size %d\n",
+		if (mpcapable->ver == MPTCP_VERSION_0 &&
+		    ((th->syn && opsize != MPTCP_SUB_LEN_CAPABLE_SYN) ||
+		     (!th->syn && th->ack && opsize != MPTCP_SUB_LEN_CAPABLE_ACK))) {
+			mptcp_debug("%s: mp_capable v0: bad option size %d\n",
+				    __func__, opsize);
+			break;
+		}
+
+		if (mpcapable->ver == MPTCP_VERSION_1 &&
+		    ((th->syn && !th->ack && opsize != MPTCPV1_SUB_LEN_CAPABLE_SYN) ||
+		     (th->syn && th->ack && opsize != MPTCPV1_SUB_LEN_CAPABLE_SYNACK) ||
+		     (!th->syn && th->ack && opsize != MPTCPV1_SUB_LEN_CAPABLE_ACK &&
+		      opsize != MPTCPV1_SUB_LEN_CAPABLE_DATA &&
+		      opsize != MPTCPV1_SUB_LEN_CAPABLE_DATA_CSUM))) {
+			mptcp_debug("%s: mp_capable v1: bad option size %d\n",
 				    __func__, opsize);
 			break;
 		}
@@ -1640,10 +1688,38 @@
 		mopt->saw_mpc = 1;
 		mopt->dss_csum = sysctl_mptcp_checksum || mpcapable->a;
 
-		if (opsize >= MPTCP_SUB_LEN_CAPABLE_SYN)
-			mopt->mptcp_sender_key = mpcapable->sender_key;
-		if (opsize == MPTCP_SUB_LEN_CAPABLE_ACK)
-			mopt->mptcp_receiver_key = mpcapable->receiver_key;
+		if (mpcapable->ver == MPTCP_VERSION_0) {
+			if (opsize == MPTCP_SUB_LEN_CAPABLE_SYN)
+				mopt->mptcp_sender_key = mpcapable->sender_key;
+
+			if (opsize == MPTCP_SUB_LEN_CAPABLE_ACK) {
+				mopt->mptcp_sender_key = mpcapable->sender_key;
+				mopt->mptcp_receiver_key = mpcapable->receiver_key;
+			}
+		} else if (mpcapable->ver == MPTCP_VERSION_1) {
+			if (opsize == MPTCPV1_SUB_LEN_CAPABLE_SYNACK)
+				mopt->mptcp_sender_key = mpcapable->sender_key;
+
+			if (opsize == MPTCPV1_SUB_LEN_CAPABLE_ACK) {
+				mopt->mptcp_sender_key = mpcapable->sender_key;
+				mopt->mptcp_receiver_key = mpcapable->receiver_key;
+			}
+
+			if (opsize == MPTCPV1_SUB_LEN_CAPABLE_DATA ||
+			    opsize == MPTCPV1_SUB_LEN_CAPABLE_DATA_CSUM) {
+				mopt->mptcp_sender_key = mpcapable->sender_key;
+				mopt->mptcp_receiver_key = mpcapable->receiver_key;
+
+				TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_MPC_DATA;
+
+				ptr += sizeof(struct mp_capable);
+				TCP_SKB_CB(skb)->dss_off = (ptr - skb_transport_header(skb));
+
+				/* Is a check-sum present? */
+				if (opsize == MPTCPV1_SUB_LEN_CAPABLE_DATA_CSUM)
+					TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_DSS_CSUM;
+			}
+		}
 
 		mopt->mptcp_ver = mpcapable->ver;
 		break;
@@ -1917,12 +1993,11 @@
 		} else if (mpadd->len == MPTCP_SUB_LEN_ADD_ADDR4_VER1 + 2) {
 			msg_parts = 3;
 		}
-		mptcp_hmac_sha1((u8 *)&mpcb->mptcp_rem_key,
-				(u8 *)no_key,
-				(u32 *)hash_mac_check, msg_parts,
-				1, (u8 *)&mpadd->addr_id,
-				4, (u8 *)&mpadd->u.v4.addr.s_addr,
-				2, (u8 *)&mpadd->u.v4.port);
+		mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_rem_key,
+			   (u8 *)no_key, (u32 *)hash_mac_check, msg_parts,
+			   1, (u8 *)&mpadd->addr_id,
+			   4, (u8 *)&mpadd->u.v4.addr.s_addr,
+			   2, (u8 *)&mpadd->u.v4.port);
 		if (memcmp(hash_mac_check, recv_hmac, 8) != 0)
 			/* ADD_ADDR2 discarded */
 			return;
@@ -1952,12 +2027,11 @@
 		} else if (mpadd->len == MPTCP_SUB_LEN_ADD_ADDR6_VER1 + 2) {
 			msg_parts = 3;
 		}
-		mptcp_hmac_sha1((u8 *)&mpcb->mptcp_rem_key,
-				(u8 *)no_key,
-				(u32 *)hash_mac_check, msg_parts,
-				1, (u8 *)&mpadd->addr_id,
-				16, (u8 *)&mpadd->u.v6.addr.s6_addr,
-				2, (u8 *)&mpadd->u.v6.port);
+		mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_rem_key,
+			   (u8 *)no_key, (u32 *)hash_mac_check, msg_parts,
+			   1, (u8 *)&mpadd->addr_id,
+			   16, (u8 *)&mpadd->u.v6.addr.s6_addr,
+			   2, (u8 *)&mpadd->u.v6.port);
 		if (memcmp(hash_mac_check, recv_hmac, 8) != 0)
 			/* ADD_ADDR2 discarded */
 			return;
@@ -2115,6 +2189,10 @@
 	if (sk->sk_state == TCP_RST_WAIT && !th->rst)
 		return true;
 
+	if (mopt->saw_mpc && !tp->mpcb->rem_key_set)
+		mptcp_initialize_recv_vars(mptcp_meta_tp(tp), tp->mpcb,
+					   mopt->mptcp_sender_key);
+
 	if (unlikely(mopt->mp_fail))
 		mptcp_mp_fail_rcvd(sk, th);
 
@@ -2122,7 +2200,8 @@
 	 * If a checksum is not present when its use has been negotiated, the
 	 * receiver MUST close the subflow with a RST as it is considered broken.
 	 */
-	if (mptcp_is_data_seq(skb) && tp->mpcb->dss_csum &&
+	if ((mptcp_is_data_seq(skb) || mptcp_is_data_mpcapable(skb)) &&
+	    tp->mpcb->dss_csum &&
 	    !(TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_DSS_CSUM)) {
 		mptcp_send_reset(sk);
 		return true;
@@ -2171,7 +2250,8 @@
 		mopt->saw_low_prio = 0;
 	}
 
-	mptcp_data_ack(sk, skb);
+	if (mptcp_data_ack(sk, skb))
+		return true;
 
 	mptcp_path_array_check(mptcp_meta_sk(sk));
 	/* Socket may have been mp_killed by a REMOVE_ADDR */
@@ -2297,11 +2377,10 @@
 		u8 hash_mac_check[20];
 		struct mptcp_cb *mpcb = tp->mpcb;
 
-		mptcp_hmac_sha1((u8 *)&mpcb->mptcp_rem_key,
-				(u8 *)&mpcb->mptcp_loc_key,
-				(u32 *)hash_mac_check, 2,
-				4, (u8 *)&tp->mptcp->rx_opt.mptcp_recv_nonce,
-				4, (u8 *)&tp->mptcp->mptcp_loc_nonce);
+		mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_rem_key,
+			   (u8 *)&mpcb->mptcp_loc_key, (u32 *)hash_mac_check, 2,
+			   4, (u8 *)&tp->mptcp->rx_opt.mptcp_recv_nonce,
+			   4, (u8 *)&tp->mptcp->mptcp_loc_nonce);
 		if (memcmp(hash_mac_check,
 			   (char *)&tp->mptcp->rx_opt.mptcp_recv_tmac, 8)) {
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNACKMAC);
@@ -2315,11 +2394,11 @@
 		tp->mptcp->pre_established = 1;
 		tp->mptcp->rcv_low_prio = tp->mptcp->rx_opt.low_prio;
 
-		mptcp_hmac_sha1((u8 *)&mpcb->mptcp_loc_key,
-				(u8 *)&mpcb->mptcp_rem_key,
-				(u32 *)&tp->mptcp->sender_mac[0], 2,
-				4, (u8 *)&tp->mptcp->mptcp_loc_nonce,
-				4, (u8 *)&tp->mptcp->rx_opt.mptcp_recv_nonce);
+		mptcp_hmac(mpcb->mptcp_ver, (u8 *)&mpcb->mptcp_loc_key,
+			   (u8 *)&mpcb->mptcp_rem_key,
+			   (u32 *)&tp->mptcp->sender_mac[0], 2,
+			   4, (u8 *)&tp->mptcp->mptcp_loc_nonce,
+			   4, (u8 *)&tp->mptcp->rx_opt.mptcp_recv_nonce);
 
 		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNACKRX);
 	} else if (mopt->saw_mpc) {
@@ -2329,8 +2408,13 @@
 		if (mopt->mptcp_ver > tcp_sk(sk)->mptcp_ver)
 			/* TODO Consider adding new MPTCP_INC_STATS entry */
 			goto fallback;
+		if (tcp_sk(sk)->mptcp_ver == MPTCP_VERSION_1 &&
+		    mopt->mptcp_ver < MPTCP_VERSION_1)
+			/* TODO Consider adding new MPTCP_INC_STATS entry */
+			/* TODO - record this in the cache - use v0 next time */
+			goto fallback;
 
-		if (mptcp_create_master_sk(sk, mopt->mptcp_sender_key,
+		if (mptcp_create_master_sk(sk, mopt->mptcp_sender_key, 1,
 					   mopt->mptcp_ver,
 					   ntohs(tcp_hdr(skb)->window)))
 			return 2;
@@ -2358,6 +2442,9 @@
 		if (tp->mpcb->dss_csum)
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_CSUMENABLED);
 
+		if (tp->mpcb->mptcp_ver >= MPTCP_VERSION_1)
+			tp->mpcb->send_mptcpv1_mpcapable = 1;
+
 		tp->mptcp->include_mpc = 1;
 
 		/* Ensure that fastopen is handled at the meta-level. */
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_ipv4.c mptcp/net/mptcp/mptcp_ipv4.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_ipv4.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_ipv4.c	2020-05-14 15:15:27.158152059 +0200
@@ -106,6 +106,9 @@
 	int loc_id;
 	bool low_prio = false;
 
+	if (!mpcb->rem_key_set)
+		return -1;
+
 	/* We need to do this as early as possible. Because, if we fail later
 	 * (e.g., get_local_id), then reqsk_free tries to remove the
 	 * request-socket from the htb in mptcp_hash_request_remove as pprev
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_ipv6.c mptcp/net/mptcp/mptcp_ipv6.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_ipv6.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_ipv6.c	2020-05-14 15:15:27.170151859 +0200
@@ -135,6 +135,9 @@
 	int loc_id;
 	bool low_prio = false;
 
+	if (!mpcb->rem_key_set)
+		return -1;
+
 	/* We need to do this as early as possible. Because, if we fail later
 	 * (e.g., get_local_id), then reqsk_free tries to remove the
 	 * request-socket from the htb in mptcp_hash_request_remove as pprev
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_output.c mptcp/net/mptcp/mptcp_output.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_output.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_output.c	2020-05-14 15:15:27.170151859 +0200
@@ -479,30 +479,78 @@
 	ptr += mptcp_write_dss_mapping(tp, skb, ptr);
 }
 
+/* Write the MP_CAPABLE with data-option */
+static int mptcp_write_mpcapable_data(const struct tcp_sock *tp,
+				      struct sk_buff *skb,
+				      __be32 *ptr)
+{
+	struct mp_capable *mpc = (struct mp_capable *)ptr;
+	u8 length;
+
+	if (tp->mpcb->dss_csum)
+		length = MPTCPV1_SUB_LEN_CAPABLE_DATA_CSUM;
+	else
+		length = MPTCPV1_SUB_LEN_CAPABLE_DATA;
+
+	mpc->kind = TCPOPT_MPTCP;
+	mpc->len = length;
+	mpc->sub = MPTCP_SUB_CAPABLE;
+	mpc->ver = MPTCP_VERSION_1;
+	mpc->a = tp->mpcb->dss_csum;
+	mpc->b = 0;
+	mpc->rsv = 0;
+	mpc->h = 1;
+
+	ptr++;
+	memcpy(ptr, TCP_SKB_CB(skb)->dss, mptcp_dss_len);
+
+	mpc->sender_key = tp->mpcb->mptcp_loc_key;
+	mpc->receiver_key = tp->mpcb->mptcp_rem_key;
+
+	/* dss is in a union with inet_skb_parm and
+	 * the IP layer expects zeroed IPCB fields.
+	 */
+	memset(TCP_SKB_CB(skb)->dss, 0, mptcp_dss_len);
+
+	return MPTCPV1_SUB_LEN_CAPABLE_DATA_ALIGN / sizeof(*ptr);
+}
+
 /* Write the saved DSS mapping to the header */
 static int mptcp_write_dss_data_seq(const struct tcp_sock *tp, struct sk_buff *skb,
 				    __be32 *ptr)
 {
+	int length;
 	__be32 *start = ptr;
 
-	memcpy(ptr, TCP_SKB_CB(skb)->dss, mptcp_dss_len);
+	if (tp->mpcb->rem_key_set) {
+		memcpy(ptr, TCP_SKB_CB(skb)->dss, mptcp_dss_len);
+
+		/* update the data_ack */
+		start[1] = htonl(mptcp_meta_tp(tp)->rcv_nxt);
+
+		length = mptcp_dss_len / sizeof(*ptr);
+	} else {
+		memcpy(ptr, TCP_SKB_CB(skb)->dss, MPTCP_SUB_LEN_DSS_ALIGN);
 
-	/* update the data_ack */
-	start[1] = htonl(mptcp_meta_tp(tp)->rcv_nxt);
+		ptr++;
+		memcpy(ptr, TCP_SKB_CB(skb)->dss + 2, MPTCP_SUB_LEN_SEQ_ALIGN);
+
+		length = (MPTCP_SUB_LEN_DSS_ALIGN + MPTCP_SUB_LEN_SEQ_ALIGN) / sizeof(*ptr);
+	}
 
 	/* dss is in a union with inet_skb_parm and
 	 * the IP layer expects zeroed IPCB fields.
 	 */
 	memset(TCP_SKB_CB(skb)->dss, 0 , mptcp_dss_len);
 
-	return mptcp_dss_len/sizeof(*ptr);
+	return length;
 }
 
 static bool mptcp_skb_entail(struct sock *sk, struct sk_buff *skb, int reinject)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	const struct sock *meta_sk = mptcp_meta_sk(sk);
-	const struct mptcp_cb *mpcb = tp->mpcb;
+	struct mptcp_cb *mpcb = tp->mpcb;
 	struct tcp_skb_cb *tcb;
 	struct sk_buff *subskb = NULL;
 
@@ -544,6 +592,11 @@
 
 	mptcp_save_dss_data_seq(tp, subskb);
 
+	if (mpcb->send_mptcpv1_mpcapable) {
+		TCP_SKB_CB(subskb)->mptcp_flags |= MPTCPHDR_MPC_DATA;
+		mpcb->send_mptcpv1_mpcapable = 0;
+	}
+
 	tcb->seq = tp->write_seq;
 
 	/* Take into account seg len */
@@ -851,10 +904,7 @@
 
 		if (!mptcp_skb_entail(subsk, skb, reinject))
 			break;
-		/* Nagle is handled at the MPTCP-layer, so
-		 * always push on the subflow
-		 */
-		__tcp_push_pending_frames(subsk, mss_now, TCP_NAGLE_PUSH);
+
 		if (reinject <= 0)
 			tcp_update_skb_after_send(meta_sk, skb, meta_tp->tcp_wstamp_ns);
 		meta_tp->lsndtime = tcp_jiffies32;
@@ -886,14 +936,12 @@
 		if (!(path_mask & mptcp_pi_to_flag(subtp->mptcp->path_index)))
 			continue;
 
-		/* We have pushed data on this subflow. We ignore the call to
-		 * cwnd_validate in tcp_write_xmit as is_cwnd_limited will never
-		 * be true (we never push more than what the cwnd can accept).
-		 * We need to ensure that we call tcp_cwnd_validate with
-		 * is_cwnd_limited set to true if we have filled the cwnd.
+		mss_now = tcp_current_mss(subsk);
+
+		/* Nagle is handled at the MPTCP-layer, so
+		 * always push on the subflow
 		 */
-		tcp_cwnd_validate(subsk, tcp_packets_in_flight(subtp) >=
-				  subtp->snd_cwnd);
+		__tcp_push_pending_frames(subsk, mss_now, TCP_NAGLE_PUSH);
 	}
 
 	return !meta_tp->packets_out && tcp_send_head(meta_sk);
@@ -988,8 +1036,13 @@
 	opts->options |= OPTION_MPTCP;
 	if (is_master_tp(tp)) {
 		opts->mptcp_options |= OPTION_MP_CAPABLE | OPTION_TYPE_SYN;
-		opts->mptcp_ver = tcp_sk(sk)->mptcp_ver;
-		*remaining -= MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN;
+		opts->mptcp_ver = tp->mptcp_ver;
+
+		if (tp->mptcp_ver >= MPTCP_VERSION_1)
+			*remaining -= MPTCPV1_SUB_LEN_CAPABLE_SYN_ALIGN;
+		else
+			*remaining -= MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN;
+
 		opts->mp_capable.sender_key = tp->mptcp_loc_key;
 		opts->dss_csum = !!sysctl_mptcp_checksum;
 	} else {
@@ -1017,7 +1070,11 @@
 		opts->mptcp_ver = mtreq->mptcp_ver;
 		opts->mp_capable.sender_key = mtreq->mptcp_loc_key;
 		opts->dss_csum = !!sysctl_mptcp_checksum || mtreq->dss_csum;
-		*remaining -= MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN;
+		if (mtreq->mptcp_ver >= MPTCP_VERSION_1) {
+			*remaining -= MPTCPV1_SUB_LEN_CAPABLE_SYNACK_ALIGN;
+		} else {
+			*remaining -= MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN;
+		}
 	} else {
 		opts->mptcp_options |= OPTION_MP_JOIN | OPTION_TYPE_SYNACK;
 		opts->mp_join_syns.sender_truncated_mac =
@@ -1080,7 +1137,12 @@
 		opts->options |= OPTION_MPTCP;
 		opts->mptcp_options |= OPTION_MP_CAPABLE |
 				       OPTION_TYPE_ACK;
-		*size += MPTCP_SUB_LEN_CAPABLE_ACK_ALIGN;
+
+		if (mpcb->mptcp_ver >= MPTCP_VERSION_1)
+			*size += MPTCPV1_SUB_LEN_CAPABLE_ACK_ALIGN;
+		else
+			*size += MPTCP_SUB_LEN_CAPABLE_ACK_ALIGN;
+
 		opts->mptcp_ver = mpcb->mptcp_ver;
 		opts->mp_capable.sender_key = mpcb->mptcp_loc_key;
 		opts->mp_capable.receiver_key = mpcb->mptcp_rem_key;
@@ -1111,14 +1173,20 @@
 		/* If !skb, we come from tcp_current_mss and thus we always
 		 * assume that the DSS-option will be set for the data-packet.
 		 */
-		if (skb && !mptcp_is_data_seq(skb)) {
+		if (skb && !mptcp_is_data_seq(skb) && mpcb->rem_key_set) {
 			*size += MPTCP_SUB_LEN_ACK_ALIGN;
+		} else if ((skb && mptcp_is_data_mpcapable(skb)) ||
+			   (!skb && tp->mpcb->send_mptcpv1_mpcapable)) {
+			*size += MPTCPV1_SUB_LEN_CAPABLE_DATA_ALIGN;
 		} else {
 			/* Doesn't matter, if csum included or not. It will be
 			 * either 10 or 12, and thus aligned = 12
 			 */
-			*size += MPTCP_SUB_LEN_ACK_ALIGN +
-				 MPTCP_SUB_LEN_SEQ_ALIGN;
+			if (mpcb->rem_key_set)
+				*size += MPTCP_SUB_LEN_ACK_ALIGN +
+					 MPTCP_SUB_LEN_SEQ_ALIGN;
+			else
+				*size += MPTCP_SUB_LEN_SEQ_ALIGN;
 		}
 
 		*size += MPTCP_SUB_LEN_DSS_ALIGN;
@@ -1171,18 +1239,36 @@
 
 		mpc->kind = TCPOPT_MPTCP;
 
-		if ((OPTION_TYPE_SYN & opts->mptcp_options) ||
-		    (OPTION_TYPE_SYNACK & opts->mptcp_options)) {
-			mpc->sender_key = opts->mp_capable.sender_key;
-			mpc->len = MPTCP_SUB_LEN_CAPABLE_SYN;
+		if (OPTION_TYPE_SYN & opts->mptcp_options) {
 			mpc->ver = opts->mptcp_ver;
-			ptr += MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN >> 2;
-		} else if (OPTION_TYPE_ACK & opts->mptcp_options) {
+
+			if (mpc->ver >= MPTCP_VERSION_1) {
+				mpc->len = MPTCPV1_SUB_LEN_CAPABLE_SYN;
+				ptr += MPTCPV1_SUB_LEN_CAPABLE_SYN_ALIGN >> 2;
+			} else {
+				mpc->sender_key = opts->mp_capable.sender_key;
+				mpc->len = MPTCP_SUB_LEN_CAPABLE_SYN;
+				ptr += MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN >> 2;
+			}
+		} else if (OPTION_TYPE_SYNACK & opts->mptcp_options) {
+			mpc->ver = opts->mptcp_ver;
+
+			if (mpc->ver >= MPTCP_VERSION_1) {
+				mpc->len = MPTCPV1_SUB_LEN_CAPABLE_SYNACK;
+				ptr += MPTCPV1_SUB_LEN_CAPABLE_SYNACK_ALIGN >> 2;
+			} else {
+				mpc->len = MPTCP_SUB_LEN_CAPABLE_SYN;
+				ptr += MPTCP_SUB_LEN_CAPABLE_SYN_ALIGN >> 2;
+			}
+
 			mpc->sender_key = opts->mp_capable.sender_key;
-			mpc->receiver_key = opts->mp_capable.receiver_key;
+		} else if (OPTION_TYPE_ACK & opts->mptcp_options) {
 			mpc->len = MPTCP_SUB_LEN_CAPABLE_ACK;
 			mpc->ver = opts->mptcp_ver;
 			ptr += MPTCP_SUB_LEN_CAPABLE_ACK_ALIGN >> 2;
+
+			mpc->sender_key = opts->mp_capable.sender_key;
+			mpc->receiver_key = opts->mp_capable.receiver_key;
 		}
 
 		mpc->sub = MPTCP_SUB_CAPABLE;
@@ -1312,8 +1398,10 @@
 	}
 
 	if (OPTION_DATA_ACK & opts->mptcp_options) {
-		if (!mptcp_is_data_seq(skb))
+		if (!mptcp_is_data_seq(skb) && tp->mpcb->rem_key_set)
 			ptr += mptcp_write_dss_data_ack(tp, skb, ptr);
+		else if (mptcp_is_data_mpcapable(skb))
+			ptr += mptcp_write_mpcapable_data(tp, skb, ptr);
 		else
 			ptr += mptcp_write_dss_data_seq(tp, skb, ptr);
 	}
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_redundant.c mptcp/net/mptcp/mptcp_redundant.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_redundant.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_redundant.c	2020-05-14 15:11:23.662202401 +0200
@@ -187,7 +187,9 @@
 {
 	struct tcp_sock *meta_tp = tcp_sk(meta_sk);
 
-	if (red_p->skb && !after(red_p->skb_end_seq, meta_tp->snd_una))
+	if (red_p->skb &&
+	    (!after(red_p->skb_end_seq, meta_tp->snd_una) ||
+	     after(red_p->skb_end_seq, meta_tp->snd_nxt)))
 		red_p->skb = NULL;
 }
 
@@ -197,9 +199,13 @@
 						    struct sock *meta_sk)
 {
 	struct sk_buff *skb;
-
-	if (!previous)
+	if (!previous){
+		if (tcp_rtx_queue_head(meta_sk)){
+			return tcp_rtx_queue_head(meta_sk);
+		}
 		return skb_peek(queue);
+	}
+
 
 	/* sk_data->skb stores the last scheduled packet for this subflow.
 	 * If sk_data->skb was scheduled but not sent (e.g., due to nagle),
@@ -246,7 +252,8 @@
 	*limit = 0;
 
 	if (skb_queue_empty(&mpcb->reinject_queue) &&
-	    skb_queue_empty(&meta_sk->sk_write_queue))
+	    skb_queue_empty(&meta_sk->sk_write_queue) &&
+		tcp_rtx_queue_empty(meta_sk))
 		/* Nothing to send */
 		return NULL;
 
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_redundant.c.orig mptcp/net/mptcp/mptcp_redundant.c.orig
--- mptcp-mptcp_trunk/net/mptcp/mptcp_redundant.c.orig	1970-01-01 01:00:00.000000000 +0100
+++ mptcp/net/mptcp/mptcp_redundant.c.orig	2020-05-11 09:39:24.476475868 +0200
@@ -0,0 +1,391 @@
+/*
+ *	MPTCP Scheduler to reduce latency and jitter.
+ *
+ *	This scheduler sends all packets redundantly on all available subflows.
+ *
+ *	Initial Design & Implementation:
+ *	Tobias Erbshaeusser <erbshauesser@dvs.tu-darmstadt.de>
+ *	Alexander Froemmgen <froemmge@dvs.tu-darmstadt.de>
+ *
+ *	Initial corrections & modifications:
+ *	Christian Pinedo <christian.pinedo@ehu.eus>
+ *	Igor Lopez <igor.lopez@ehu.eus>
+ *
+ *	This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <net/mptcp.h>
+
+/* Struct to store the data of a single subflow */
+struct redsched_priv {
+	/* The skb or NULL */
+	struct sk_buff *skb;
+	/* End sequence number of the skb. This number should be checked
+	 * to be valid before the skb field is used
+	 */
+	u32 skb_end_seq;
+};
+
+/* Struct to store the data of the control block */
+struct redsched_cb {
+	/* The next subflow where a skb should be sent or NULL */
+	struct tcp_sock *next_subflow;
+};
+
+/* Returns the socket data from a given subflow socket */
+static struct redsched_priv *redsched_get_priv(struct tcp_sock *tp)
+{
+	return (struct redsched_priv *)&tp->mptcp->mptcp_sched[0];
+}
+
+/* Returns the control block data from a given meta socket */
+static struct redsched_cb *redsched_get_cb(struct tcp_sock *tp)
+{
+	return (struct redsched_cb *)&tp->mpcb->mptcp_sched[0];
+}
+
+static bool redsched_get_active_valid_sks(struct sock *meta_sk)
+{
+	struct tcp_sock *meta_tp = tcp_sk(meta_sk);
+	struct mptcp_cb *mpcb = meta_tp->mpcb;
+	struct mptcp_tcp_sock *mptcp;
+	int active_valid_sks = 0;
+
+	mptcp_for_each_sub(mpcb, mptcp) {
+		struct sock *sk = mptcp_to_sock(mptcp);
+
+		if (subflow_is_active((struct tcp_sock *)sk) &&
+		    !mptcp_is_def_unavailable(sk))
+			active_valid_sks++;
+	}
+
+	return active_valid_sks;
+}
+
+static bool redsched_use_subflow(struct sock *meta_sk,
+				 int active_valid_sks,
+				 struct tcp_sock *tp,
+				 struct sk_buff *skb)
+{
+	if (!skb || !mptcp_is_available((struct sock *)tp, skb, false))
+		return false;
+
+	if (TCP_SKB_CB(skb)->path_mask != 0)
+		return subflow_is_active(tp);
+
+	if (TCP_SKB_CB(skb)->path_mask == 0) {
+		if (active_valid_sks == -1)
+			active_valid_sks = redsched_get_active_valid_sks(meta_sk);
+
+		if (subflow_is_backup(tp) && active_valid_sks > 0)
+			return false;
+		else
+			return true;
+	}
+
+	return false;
+}
+
+#define mptcp_entry_next_rcu(__mptcp)						\
+	hlist_entry_safe(rcu_dereference_raw(hlist_next_rcu(			\
+		&(__mptcp)->node)), struct mptcp_tcp_sock, node)
+
+static void redsched_update_next_subflow(struct tcp_sock *tp,
+					 struct redsched_cb *red_cb)
+{
+	struct mptcp_tcp_sock *mptcp = mptcp_entry_next_rcu(tp->mptcp);
+
+	if (mptcp)
+		red_cb->next_subflow = mptcp->tp;
+	else
+		red_cb->next_subflow = NULL;
+}
+
+static struct sock *red_get_available_subflow(struct sock *meta_sk,
+					      struct sk_buff *skb,
+					      bool zero_wnd_test)
+{
+	struct tcp_sock *meta_tp = tcp_sk(meta_sk);
+	struct mptcp_cb *mpcb = meta_tp->mpcb;
+	struct redsched_cb *red_cb = redsched_get_cb(meta_tp);
+	struct tcp_sock *first_tp = red_cb->next_subflow, *tp;
+	struct mptcp_tcp_sock *mptcp;
+	int found = 0;
+
+	/* Answer data_fin on same subflow */
+	if (meta_sk->sk_shutdown & RCV_SHUTDOWN &&
+	    skb && mptcp_is_data_fin(skb)) {
+		mptcp_for_each_sub(mpcb, mptcp) {
+			struct sock *sk = mptcp_to_sock(mptcp);
+
+			if (tcp_sk(sk)->mptcp->path_index ==
+				mpcb->dfin_path_index &&
+			    mptcp_is_available(sk, skb, zero_wnd_test))
+				return sk;
+		}
+	}
+
+	if (!first_tp && !hlist_empty(&mpcb->conn_list)) {
+		first_tp = hlist_entry_safe(rcu_dereference_raw(hlist_first_rcu(&mpcb->conn_list)),
+					    struct mptcp_tcp_sock, node)->tp;
+	}
+	tp = first_tp;
+
+	/* still NULL (no subflow in conn_list?) */
+	if (!first_tp)
+		return NULL;
+
+	/* Search for a subflow to send it.
+	 *
+	 * We want to pick a subflow that is after 'first_tp' in the list of subflows.
+	 * Thus, the first mptcp_for_each_sub()-loop tries to walk the list up
+	 * to the subflow 'tp' and then checks whether any one of the remaining
+	 * ones is eligible to send.
+	 * The second mptcp_for_each-sub()-loop is then iterating from the
+	 * beginning of the list up to 'first_tp'.
+	 */
+	mptcp_for_each_sub(mpcb, mptcp) {
+		/* We go up to the subflow 'tp' and start from there */
+		if (tp == mptcp->tp)
+			found = 1;
+
+		if (!found)
+			continue;
+		tp = mptcp->tp;
+
+		if (mptcp_is_available((struct sock *)tp, skb,
+				       zero_wnd_test)) {
+			redsched_update_next_subflow(tp, red_cb);
+			return (struct sock *)tp;
+		}
+	}
+
+	mptcp_for_each_sub(mpcb, mptcp) {
+		tp = mptcp->tp;
+
+		if (tp == first_tp)
+			break;
+
+		if (mptcp_is_available((struct sock *)tp, skb,
+				       zero_wnd_test)) {
+			redsched_update_next_subflow(tp, red_cb);
+			return (struct sock *)tp;
+		}
+	}
+
+	/* No space */
+	return NULL;
+}
+
+/* Corrects the stored skb pointers if they are invalid */
+static void redsched_correct_skb_pointers(struct sock *meta_sk,
+					  struct redsched_priv *red_p)
+{
+	struct tcp_sock *meta_tp = tcp_sk(meta_sk);
+
+	if (red_p->skb &&
+	    (!after(red_p->skb_end_seq, meta_tp->snd_una) ||
+	     after(red_p->skb_end_seq, meta_tp->snd_nxt)))
+		red_p->skb = NULL;
+}
+
+/* Returns the next skb from the queue */
+static struct sk_buff *redsched_next_skb_from_queue(struct sk_buff_head *queue,
+						    struct sk_buff *previous,
+						    struct sock *meta_sk)
+{
+	struct sk_buff *skb;
+
+	if (!previous)
+		return skb_peek(queue);
+
+	/* sk_data->skb stores the last scheduled packet for this subflow.
+	 * If sk_data->skb was scheduled but not sent (e.g., due to nagle),
+	 * we have to schedule it again.
+	 *
+	 * For the redundant scheduler, there are two cases:
+	 * 1. sk_data->skb was not sent on another subflow:
+	 *    we have to schedule it again to ensure that we do not
+	 *    skip this packet.
+	 * 2. sk_data->skb was already sent on another subflow:
+	 *    with regard to the redundant semantic, we have to
+	 *    schedule it again. However, we keep it simple and ignore it,
+	 *    as it was already sent by another subflow.
+	 *    This might be changed in the future.
+	 *
+	 * For case 1, send_head is equal previous, as only a single
+	 * packet can be skipped.
+	 */
+	if (tcp_send_head(meta_sk) == previous)
+		return tcp_send_head(meta_sk);
+
+	skb = skb_rb_next(previous);
+	if (skb)
+		return skb;
+
+	return tcp_send_head(meta_sk);
+}
+
+static struct sk_buff *mptcp_red_next_segment(struct sock *meta_sk,
+					      int *reinject,
+					      struct sock **subsk,
+					      unsigned int *limit)
+{
+	struct tcp_sock *meta_tp = tcp_sk(meta_sk);
+	struct mptcp_cb *mpcb = meta_tp->mpcb;
+	struct redsched_cb *red_cb = redsched_get_cb(meta_tp);
+	struct tcp_sock *first_tp = red_cb->next_subflow, *tp;
+	struct mptcp_tcp_sock *mptcp;
+	int active_valid_sks = -1;
+	struct sk_buff *skb;
+	int found = 0;
+
+	/* As we set it, we have to reset it as well. */
+	*limit = 0;
+
+	if (skb_queue_empty(&mpcb->reinject_queue) &&
+	    skb_queue_empty(&meta_sk->sk_write_queue))
+		/* Nothing to send */
+		return NULL;
+
+	/* First try reinjections */
+	skb = skb_peek(&mpcb->reinject_queue);
+	if (skb) {
+		*subsk = get_available_subflow(meta_sk, skb, false);
+		if (!*subsk)
+			return NULL;
+		*reinject = 1;
+		return skb;
+	}
+
+	/* Then try indistinctly redundant and normal skbs */
+
+	if (!first_tp && !hlist_empty(&mpcb->conn_list)) {
+		first_tp = hlist_entry_safe(rcu_dereference_raw(hlist_first_rcu(&mpcb->conn_list)),
+					    struct mptcp_tcp_sock, node)->tp;
+	}
+
+	/* still NULL (no subflow in conn_list?) */
+	if (!first_tp)
+		return NULL;
+
+	tp = first_tp;
+
+	*reinject = 0;
+	active_valid_sks = redsched_get_active_valid_sks(meta_sk);
+
+	/* We want to pick a subflow that is after 'first_tp' in the list of subflows.
+	 * Thus, the first mptcp_for_each_sub()-loop tries to walk the list up
+	 * to the subflow 'tp' and then checks whether any one of the remaining
+	 * ones can send a segment.
+	 * The second mptcp_for_each-sub()-loop is then iterating from the
+	 * beginning of the list up to 'first_tp'.
+	 */
+	mptcp_for_each_sub(mpcb, mptcp) {
+		struct redsched_priv *red_p;
+
+		if (tp == mptcp->tp)
+			found = 1;
+
+		if (!found)
+			continue;
+
+		tp = mptcp->tp;
+
+		/* Correct the skb pointers of the current subflow */
+		red_p = redsched_get_priv(tp);
+		redsched_correct_skb_pointers(meta_sk, red_p);
+
+		skb = redsched_next_skb_from_queue(&meta_sk->sk_write_queue,
+						   red_p->skb, meta_sk);
+		if (skb && redsched_use_subflow(meta_sk, active_valid_sks, tp,
+						skb)) {
+			red_p->skb = skb;
+			red_p->skb_end_seq = TCP_SKB_CB(skb)->end_seq;
+			redsched_update_next_subflow(tp, red_cb);
+			*subsk = (struct sock *)tp;
+
+			if (TCP_SKB_CB(skb)->path_mask)
+				*reinject = -1;
+			return skb;
+		}
+	}
+
+	mptcp_for_each_sub(mpcb, mptcp) {
+		struct redsched_priv *red_p;
+
+		tp = mptcp->tp;
+
+		if (tp == first_tp)
+			break;
+
+		/* Correct the skb pointers of the current subflow */
+		red_p = redsched_get_priv(tp);
+		redsched_correct_skb_pointers(meta_sk, red_p);
+
+		skb = redsched_next_skb_from_queue(&meta_sk->sk_write_queue,
+						   red_p->skb, meta_sk);
+		if (skb && redsched_use_subflow(meta_sk, active_valid_sks, tp,
+						skb)) {
+			red_p->skb = skb;
+			red_p->skb_end_seq = TCP_SKB_CB(skb)->end_seq;
+			redsched_update_next_subflow(tp, red_cb);
+			*subsk = (struct sock *)tp;
+
+			if (TCP_SKB_CB(skb)->path_mask)
+				*reinject = -1;
+			return skb;
+		}
+	}
+
+	/* Nothing to send */
+	return NULL;
+}
+
+static void redsched_release(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct redsched_cb *red_cb = redsched_get_cb(tp);
+
+	/* Check if the next subflow would be the released one. If yes correct
+	 * the pointer
+	 */
+	if (red_cb->next_subflow == tp)
+		redsched_update_next_subflow(tp, red_cb);
+}
+
+static struct mptcp_sched_ops mptcp_sched_red = {
+	.get_subflow = red_get_available_subflow,
+	.next_segment = mptcp_red_next_segment,
+	.release = redsched_release,
+	.name = "redundant",
+	.owner = THIS_MODULE,
+};
+
+static int __init red_register(void)
+{
+	BUILD_BUG_ON(sizeof(struct redsched_priv) > MPTCP_SCHED_SIZE);
+	BUILD_BUG_ON(sizeof(struct redsched_cb) > MPTCP_SCHED_DATA_SIZE);
+
+	if (mptcp_register_scheduler(&mptcp_sched_red))
+		return -1;
+
+	return 0;
+}
+
+static void red_unregister(void)
+{
+	mptcp_unregister_scheduler(&mptcp_sched_red);
+}
+
+module_init(red_register);
+module_exit(red_unregister);
+
+MODULE_AUTHOR("Tobias Erbshaeusser, Alexander Froemmgen");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("REDUNDANT MPTCP");
+MODULE_VERSION("0.90");
diff -aurN '--exclude=.git' mptcp-mptcp_trunk/net/mptcp/mptcp_sched.c mptcp/net/mptcp/mptcp_sched.c
--- mptcp-mptcp_trunk/net/mptcp/mptcp_sched.c	2020-02-20 18:07:47.000000000 +0100
+++ mptcp/net/mptcp/mptcp_sched.c	2020-05-11 09:40:13.463584360 +0200
@@ -76,7 +76,7 @@
 	 */
 	space = (tp->snd_cwnd - in_flight) * tp->mss_cache;
 
-	if (tp->write_seq - tp->snd_nxt > space)
+	if (tp->write_seq - tp->snd_nxt >= space)
 		return true;
 
 	if (zero_wnd_test && !before(tp->write_seq, tcp_wnd_end(tp)))
@@ -391,10 +391,11 @@
 					  unsigned int *limit)
 {
 	struct sk_buff *skb = __mptcp_next_segment(meta_sk, reinject);
-	unsigned int mss_now;
+	unsigned int mss_now, in_flight_space;
+	int remaining_in_flight_space;
+	u32 max_len, max_segs, window;
 	struct tcp_sock *subtp;
 	u16 gso_max_segs;
-	u32 max_len, max_segs, window, needed;
 
 	/* As we set it, we have to reset it as well. */
 	*limit = 0;
@@ -424,9 +425,6 @@
 	/* The following is similar to tcp_mss_split_point, but
 	 * we do not care about nagle, because we will anyways
 	 * use TCP_NAGLE_PUSH, which overrides this.
-	 *
-	 * So, we first limit according to the cwnd/gso-size and then according
-	 * to the subflow's window.
 	 */
 
 	gso_max_segs = (*subsk)->sk_gso_max_segs;
@@ -436,16 +434,30 @@
 	if (!max_segs)
 		return NULL;
 
-	max_len = mss_now * max_segs;
-	window = tcp_wnd_end(subtp) - subtp->write_seq;
+	/* max_len is what would fit in the cwnd (respecting the 2GSO-limit of
+	 * tcp_cwnd_test), but ignoring whatever was already queued.
+	 */
+	max_len = min(mss_now * max_segs, skb->len);
 
-	needed = min(skb->len, window);
-	if (max_len <= skb->len)
-		/* Take max_win, which is actually the cwnd/gso-size */
-		*limit = max_len;
+	in_flight_space = (subtp->snd_cwnd - tcp_packets_in_flight(subtp)) * mss_now;
+	remaining_in_flight_space = (int)in_flight_space - (subtp->write_seq - subtp->snd_nxt);
+
+	if (remaining_in_flight_space <= 0)
+		WARN_ONCE(1, "in_flight %u cwnd %u wseq %u snxt %u mss_now %u cache %u",
+			  tcp_packets_in_flight(subtp), subtp->snd_cwnd,
+			  subtp->write_seq, subtp->snd_nxt, mss_now, subtp->mss_cache);
 	else
-		/* Or, take the window */
-		*limit = needed;
+		/* max_len now fits exactly in the write-queue, taking into
+		 * account what was already queued.
+		 */
+		max_len = min_t(u32, max_len, remaining_in_flight_space);
+
+	window = tcp_wnd_end(subtp) - subtp->write_seq;
+
+	/* max_len now also respects the announced receive-window */
+	max_len = min(max_len, window);
+
+	*limit = max_len;
 
 	return skb;
 }