1 files changed, 132 insertions, 61 deletions

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 3ed4709a7509..3a1967bc7bad 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -15,7 +15,6 @@
 #include <net/inet_common.h>
 #include <net/inet_hashtables.h>
 #include <net/protocol.h>
-#include <net/tcp.h>
 #include <net/tcp_states.h>
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 #include <net/transp_v6.h>
@@ -85,7 +84,7 @@ static int __mptcp_socket_create(struct mptcp_sock *msk)
 	subflow->subflow_id = msk->subflow_id++;
 
 	/* This is the first subflow, always with id 0 */
-	subflow->local_id_valid = 1;
+	WRITE_ONCE(subflow->local_id, 0);
 	mptcp_sock_graft(msk->first, sk->sk_socket);
 	iput(SOCK_INODE(ssock));
 
@@ -410,6 +409,7 @@ static void mptcp_close_wake_up(struct sock *sk)
 		sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
 }
 
+/* called under the msk socket lock */
 static bool mptcp_pending_data_fin_ack(struct sock *sk)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -441,16 +441,17 @@ static void mptcp_check_data_fin_ack(struct sock *sk)
 	}
 }
 
+/* can be called with no lock acquired */
 static bool mptcp_pending_data_fin(struct sock *sk, u64 *seq)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
 	if (READ_ONCE(msk->rcv_data_fin) &&
-	    ((1 << sk->sk_state) &
+	    ((1 << inet_sk_state_load(sk)) &
 	     (TCPF_ESTABLISHED | TCPF_FIN_WAIT1 | TCPF_FIN_WAIT2))) {
 		u64 rcv_data_fin_seq = READ_ONCE(msk->rcv_data_fin_seq);
 
-		if (msk->ack_seq == rcv_data_fin_seq) {
+		if (READ_ONCE(msk->ack_seq) == rcv_data_fin_seq) {
 			if (seq)
 				*seq = rcv_data_fin_seq;
 
@@ -748,7 +749,7 @@ static bool __mptcp_ofo_queue(struct mptcp_sock *msk)
 			__skb_queue_tail(&sk->sk_receive_queue, skb);
 		}
 		msk->bytes_received += end_seq - msk->ack_seq;
-		msk->ack_seq = end_seq;
+		WRITE_ONCE(msk->ack_seq, end_seq);
 		moved = true;
 	}
 	return moved;
@@ -985,6 +986,7 @@ static void dfrag_clear(struct sock *sk, struct mptcp_data_frag *dfrag)
 	put_page(dfrag->page);
 }
 
+/* called under both the msk socket lock and the data lock */
 static void __mptcp_clean_una(struct sock *sk)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -1033,13 +1035,15 @@ static void __mptcp_clean_una(struct sock *sk)
 		msk->recovery = false;
 
 out:
-	if (snd_una == READ_ONCE(msk->snd_nxt) &&
-	    snd_una == READ_ONCE(msk->write_seq)) {
+	if (snd_una == msk->snd_nxt && snd_una == msk->write_seq) {
 		if (mptcp_rtx_timer_pending(sk) && !mptcp_data_fin_enabled(msk))
 			mptcp_stop_rtx_timer(sk);
 	} else {
 		mptcp_reset_rtx_timer(sk);
 	}
+
+	if (mptcp_pending_data_fin_ack(sk))
+		mptcp_schedule_work(sk);
 }
 
 static void __mptcp_clean_una_wakeup(struct sock *sk)
@@ -1260,6 +1264,7 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 		mpext = mptcp_get_ext(skb);
 		if (!mptcp_skb_can_collapse_to(data_seq, skb, mpext)) {
 			TCP_SKB_CB(skb)->eor = 1;
+			tcp_mark_push(tcp_sk(ssk), skb);
 			goto alloc_skb;
 		}
 
@@ -1499,14 +1504,17 @@ static void mptcp_update_post_push(struct mptcp_sock *msk,
 	 */
 	if (likely(after64(snd_nxt_new, msk->snd_nxt))) {
 		msk->bytes_sent += snd_nxt_new - msk->snd_nxt;
-		msk->snd_nxt = snd_nxt_new;
+		WRITE_ONCE(msk->snd_nxt, snd_nxt_new);
 	}
 }
 
 void mptcp_check_and_set_pending(struct sock *sk)
 {
-	if (mptcp_send_head(sk))
-		mptcp_sk(sk)->push_pending |= BIT(MPTCP_PUSH_PENDING);
+	if (mptcp_send_head(sk)) {
+		mptcp_data_lock(sk);
+		mptcp_sk(sk)->cb_flags |= BIT(MPTCP_PUSH_PENDING);
+		mptcp_data_unlock(sk);
+	}
 }
 
 static int __subflow_push_pending(struct sock *sk, struct sock *ssk,
@@ -1683,15 +1691,6 @@ out:
 	}
 }
 
-static void mptcp_set_nospace(struct sock *sk)
-{
-	/* enable autotune */
-	set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
-
-	/* will be cleared on avail space */
-	set_bit(MPTCP_NOSPACE, &mptcp_sk(sk)->flags);
-}
-
 static int mptcp_disconnect(struct sock *sk, int flags);
 
 static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
@@ -1762,6 +1761,30 @@ static int do_copy_data_nocache(struct sock *sk, int copy,
 	return 0;
 }
 
+/* open-code sk_stream_memory_free() plus sent limit computation to
+ * avoid indirect calls in fast-path.
+ * Called under the msk socket lock, so we can avoid a bunch of ONCE
+ * annotations.
+ */
+static u32 mptcp_send_limit(const struct sock *sk)
+{
+	const struct mptcp_sock *msk = mptcp_sk(sk);
+	u32 limit, not_sent;
+
+	if (sk->sk_wmem_queued >= READ_ONCE(sk->sk_sndbuf))
+		return 0;
+
+	limit = mptcp_notsent_lowat(sk);
+	if (limit == UINT_MAX)
+		return UINT_MAX;
+
+	not_sent = msk->write_seq - msk->snd_nxt;
+	if (not_sent >= limit)
+		return 0;
+
+	return limit - not_sent;
+}
+
 static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -1806,6 +1829,12 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		struct mptcp_data_frag *dfrag;
 		bool dfrag_collapsed;
 		size_t psize, offset;
+		u32 copy_limit;
+
+		/* ensure fitting the notsent_lowat() constraint */
+		copy_limit = mptcp_send_limit(sk);
+		if (!copy_limit)
+			goto wait_for_memory;
 
 		/* reuse tail pfrag, if possible, or carve a new one from the
 		 * page allocator
@@ -1813,9 +1842,6 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		dfrag = mptcp_pending_tail(sk);
 		dfrag_collapsed = mptcp_frag_can_collapse_to(msk, pfrag, dfrag);
 		if (!dfrag_collapsed) {
-			if (!sk_stream_memory_free(sk))
-				goto wait_for_memory;
-
 			if (!mptcp_page_frag_refill(sk, pfrag))
 				goto wait_for_memory;
 
@@ -1830,6 +1856,7 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		offset = dfrag->offset + dfrag->data_len;
 		psize = pfrag->size - offset;
 		psize = min_t(size_t, psize, msg_data_left(msg));
+		psize = min_t(size_t, psize, copy_limit);
 		total_ts = psize + frag_truesize;
 
 		if (!sk_wmem_schedule(sk, total_ts))
@@ -1865,7 +1892,7 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		continue;
 
 wait_for_memory:
-		mptcp_set_nospace(sk);
+		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
 		__mptcp_push_pending(sk, msg->msg_flags);
 		ret = sk_stream_wait_memory(sk, &timeo);
 		if (ret)
@@ -1960,6 +1987,9 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)
 	if (copied <= 0)
 		return;
 
+	if (!msk->rcvspace_init)
+		mptcp_rcv_space_init(msk, msk->first);
+
 	msk->rcvq_space.copied += copied;
 
 	mstamp = div_u64(tcp_clock_ns(), NSEC_PER_USEC);
@@ -2108,7 +2138,7 @@ static unsigned int mptcp_inq_hint(const struct sock *sk)
 
 	skb = skb_peek(&msk->receive_queue);
 	if (skb) {
-		u64 hint_val = msk->ack_seq - MPTCP_SKB_CB(skb)->map_seq;
+		u64 hint_val = READ_ONCE(msk->ack_seq) - MPTCP_SKB_CB(skb)->map_seq;
 
 		if (hint_val >= INT_MAX)
 			return INT_MAX;
@@ -2314,9 +2344,6 @@ bool __mptcp_retransmit_pending_data(struct sock *sk)
 	if (__mptcp_check_fallback(msk))
 		return false;
 
-	if (tcp_rtx_and_write_queues_empty(sk))
-		return false;
-
 	/* the closing socket has some data untransmitted and/or unacked:
 	 * some data in the mptcp rtx queue has not really xmitted yet.
 	 * keep it simple and re-inject the whole mptcp level rtx queue
@@ -2755,7 +2782,7 @@ static void __mptcp_init_sock(struct sock *sk)
 	__skb_queue_head_init(&msk->receive_queue);
 	msk->out_of_order_queue = RB_ROOT;
 	msk->first_pending = NULL;
-	msk->rmem_fwd_alloc = 0;
+	WRITE_ONCE(msk->rmem_fwd_alloc, 0);
 	WRITE_ONCE(msk->rmem_released, 0);
 	msk->timer_ival = TCP_RTO_MIN;
 	msk->scaling_ratio = TCP_DEFAULT_SCALING_RATIO;
@@ -2971,7 +2998,7 @@ static void __mptcp_destroy_sock(struct sock *sk)
 
 	sk->sk_prot->destroy(sk);
 
-	WARN_ON_ONCE(msk->rmem_fwd_alloc);
+	WARN_ON_ONCE(READ_ONCE(msk->rmem_fwd_alloc));
 	WARN_ON_ONCE(msk->rmem_released);
 	sk_stream_kill_queues(sk);
 	xfrm_sk_free_policy(sk);
@@ -3145,22 +3172,22 @@ static int mptcp_disconnect(struct sock *sk, int flags)
 	mptcp_destroy_common(msk, MPTCP_CF_FASTCLOSE);
 	WRITE_ONCE(msk->flags, 0);
 	msk->cb_flags = 0;
-	msk->push_pending = 0;
 	msk->recovery = false;
-	msk->can_ack = false;
-	msk->fully_established = false;
-	msk->rcv_data_fin = false;
-	msk->snd_data_fin_enable = false;
-	msk->rcv_fastclose = false;
-	msk->use_64bit_ack = false;
-	msk->bytes_consumed = 0;
+	WRITE_ONCE(msk->can_ack, false);
+	WRITE_ONCE(msk->fully_established, false);
+	WRITE_ONCE(msk->rcv_data_fin, false);
+	WRITE_ONCE(msk->snd_data_fin_enable, false);
+	WRITE_ONCE(msk->rcv_fastclose, false);
+	WRITE_ONCE(msk->use_64bit_ack, false);
 	WRITE_ONCE(msk->csum_enabled, mptcp_is_checksum_enabled(sock_net(sk)));
 	mptcp_pm_data_reset(msk);
 	mptcp_ca_reset(sk);
+	msk->bytes_consumed = 0;
 	msk->bytes_acked = 0;
 	msk->bytes_received = 0;
 	msk->bytes_sent = 0;
 	msk->bytes_retrans = 0;
+	msk->rcvspace_init = 0;
 
 	WRITE_ONCE(sk->sk_shutdown, 0);
 	sk_error_report(sk);
@@ -3174,8 +3201,50 @@ static struct ipv6_pinfo *mptcp_inet6_sk(const struct sock *sk)
 
 	return (struct ipv6_pinfo *)(((u8 *)sk) + offset);
 }
+
+static void mptcp_copy_ip6_options(struct sock *newsk, const struct sock *sk)
+{
+	const struct ipv6_pinfo *np = inet6_sk(sk);
+	struct ipv6_txoptions *opt;
+	struct ipv6_pinfo *newnp;
+
+	newnp = inet6_sk(newsk);
+
+	rcu_read_lock();
+	opt = rcu_dereference(np->opt);
+	if (opt) {
+		opt = ipv6_dup_options(newsk, opt);
+		if (!opt)
+			net_warn_ratelimited("%s: Failed to copy ip6 options\n", __func__);
+	}
+	RCU_INIT_POINTER(newnp->opt, opt);
+	rcu_read_unlock();
+}
 #endif
 
+static void mptcp_copy_ip_options(struct sock *newsk, const struct sock *sk)
+{
+	struct ip_options_rcu *inet_opt, *newopt = NULL;
+	const struct inet_sock *inet = inet_sk(sk);
+	struct inet_sock *newinet;
+
+	newinet = inet_sk(newsk);
+
+	rcu_read_lock();
+	inet_opt = rcu_dereference(inet->inet_opt);
+	if (inet_opt) {
+		newopt = sock_kmalloc(newsk, sizeof(*inet_opt) +
+				      inet_opt->opt.optlen, GFP_ATOMIC);
+		if (newopt)
+			memcpy(newopt, inet_opt, sizeof(*inet_opt) +
+			       inet_opt->opt.optlen);
+		else
+			net_warn_ratelimited("%s: Failed to copy ip options\n", __func__);
+	}
+	RCU_INIT_POINTER(newinet->inet_opt, newopt);
+	rcu_read_unlock();
+}
+
 struct sock *mptcp_sk_clone_init(const struct sock *sk,
 				 const struct mptcp_options_received *mp_opt,
 				 struct sock *ssk,
@@ -3183,6 +3252,7 @@ struct sock *mptcp_sk_clone_init(const struct sock *sk,
 {
 	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
 	struct sock *nsk = sk_clone_lock(sk, GFP_ATOMIC);
+	struct mptcp_subflow_context *subflow;
 	struct mptcp_sock *msk;
 
 	if (!nsk)
@@ -3195,18 +3265,25 @@ struct sock *mptcp_sk_clone_init(const struct sock *sk,
 
 	__mptcp_init_sock(nsk);
 
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	if (nsk->sk_family == AF_INET6)
+		mptcp_copy_ip6_options(nsk, sk);
+	else
+#endif
+		mptcp_copy_ip_options(nsk, sk);
+
 	msk = mptcp_sk(nsk);
-	msk->local_key = subflow_req->local_key;
-	msk->token = subflow_req->token;
+	WRITE_ONCE(msk->local_key, subflow_req->local_key);
+	WRITE_ONCE(msk->token, subflow_req->token);
 	msk->in_accept_queue = 1;
 	WRITE_ONCE(msk->fully_established, false);
 	if (mp_opt->suboptions & OPTION_MPTCP_CSUMREQD)
 		WRITE_ONCE(msk->csum_enabled, true);
 
-	msk->write_seq = subflow_req->idsn + 1;
-	msk->snd_nxt = msk->write_seq;
-	msk->snd_una = msk->write_seq;
-	msk->wnd_end = msk->snd_nxt + req->rsk_rcv_wnd;
+	WRITE_ONCE(msk->write_seq, subflow_req->idsn + 1);
+	WRITE_ONCE(msk->snd_nxt, msk->write_seq);
+	WRITE_ONCE(msk->snd_una, msk->write_seq);
+	WRITE_ONCE(msk->wnd_end, msk->snd_nxt + tcp_sk(ssk)->snd_wnd);
 	msk->setsockopt_seq = mptcp_sk(sk)->setsockopt_seq;
 	mptcp_init_sched(msk, mptcp_sk(sk)->sched);
 
@@ -3223,7 +3300,8 @@ struct sock *mptcp_sk_clone_init(const struct sock *sk,
 
 	/* The msk maintain a ref to each subflow in the connections list */
 	WRITE_ONCE(msk->first, ssk);
-	list_add(&mptcp_subflow_ctx(ssk)->node, &msk->conn_list);
+	subflow = mptcp_subflow_ctx(ssk);
+	list_add(&subflow->node, &msk->conn_list);
 	sock_hold(ssk);
 
 	/* new mpc subflow takes ownership of the newly
@@ -3238,6 +3316,9 @@ struct sock *mptcp_sk_clone_init(const struct sock *sk,
 	__mptcp_propagate_sndbuf(nsk, ssk);
 
 	mptcp_rcv_space_init(msk, ssk);
+
+	if (mp_opt->suboptions & OPTION_MPTCP_MPC_ACK)
+		__mptcp_subflow_fully_established(msk, subflow, mp_opt);
 	bh_unlock_sock(nsk);
 
 	/* note: the newly allocated socket refcount is 2 now */
@@ -3248,6 +3329,7 @@ void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk)
 {
 	const struct tcp_sock *tp = tcp_sk(ssk);
 
+	msk->rcvspace_init = 1;
 	msk->rcvq_space.copied = 0;
 	msk->rcvq_space.rtt_us = 0;
 
@@ -3258,8 +3340,6 @@ void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk)
 				      TCP_INIT_CWND * tp->advmss);
 	if (msk->rcvq_space.space == 0)
 		msk->rcvq_space.space = TCP_INIT_CWND * TCP_MSS_DEFAULT;
-
-	WRITE_ONCE(msk->wnd_end, msk->snd_nxt + tcp_sk(ssk)->snd_wnd);
 }
 
 void mptcp_destroy_common(struct mptcp_sock *msk, unsigned int flags)
@@ -3306,9 +3386,6 @@ void __mptcp_data_acked(struct sock *sk)
 		__mptcp_clean_una(sk);
 	else
 		__set_bit(MPTCP_CLEAN_UNA, &mptcp_sk(sk)->cb_flags);
-
-	if (mptcp_pending_data_fin_ack(sk))
-		mptcp_schedule_work(sk);
 }
 
 void __mptcp_check_push(struct sock *sk, struct sock *ssk)
@@ -3333,8 +3410,7 @@ static void mptcp_release_cb(struct sock *sk)
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
 	for (;;) {
-		unsigned long flags = (msk->cb_flags & MPTCP_FLAGS_PROCESS_CTX_NEED) |
-				      msk->push_pending;
+		unsigned long flags = (msk->cb_flags & MPTCP_FLAGS_PROCESS_CTX_NEED);
 		struct list_head join_list;
 
 		if (!flags)
@@ -3350,7 +3426,6 @@ static void mptcp_release_cb(struct sock *sk)
 		 *    datapath acquires the msk socket spinlock while helding
 		 *    the subflow socket lock
 		 */
-		msk->push_pending = 0;
 		msk->cb_flags &= ~flags;
 		spin_unlock_bh(&sk->sk_lock.slock);
 
@@ -3478,13 +3553,8 @@ void mptcp_finish_connect(struct sock *ssk)
 	 * accessing the field below
 	 */
 	WRITE_ONCE(msk->local_key, subflow->local_key);
-	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
-	WRITE_ONCE(msk->snd_nxt, msk->write_seq);
-	WRITE_ONCE(msk->snd_una, msk->write_seq);
 
 	mptcp_pm_new_connection(msk, ssk, 0);
-
-	mptcp_rcv_space_init(msk, ssk);
 }
 
 void mptcp_sock_graft(struct sock *sk, struct socket *parent)
@@ -3717,6 +3787,7 @@ static struct proto mptcp_prot = {
 	.unhash		= mptcp_unhash,
 	.get_port	= mptcp_get_port,
 	.forward_alloc_get	= mptcp_forward_alloc_get,
+	.stream_memory_free	= mptcp_stream_memory_free,
 	.sockets_allocated	= &mptcp_sockets_allocated,
 
 	.memory_allocated	= &tcp_memory_allocated,
@@ -3890,12 +3961,12 @@ static __poll_t mptcp_check_writeable(struct mptcp_sock *msk)
 {
 	struct sock *sk = (struct sock *)msk;
 
-	if (sk_stream_is_writeable(sk))
+	if (__mptcp_stream_is_writeable(sk, 1))
 		return EPOLLOUT | EPOLLWRNORM;
 
-	mptcp_set_nospace(sk);
-	smp_mb__after_atomic(); /* msk->flags is changed by write_space cb */
-	if (sk_stream_is_writeable(sk))
+	set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+	smp_mb__after_atomic(); /* NOSPACE is changed by mptcp_write_space() */
+	if (__mptcp_stream_is_writeable(sk, 1))
 		return EPOLLOUT | EPOLLWRNORM;
 
 	return 0;