[MPTCP] [PATCH net-next 0/8] mptcp: non backup subflows pre-reqs

[MPTCP] [PATCH net-next 0/8] mptcp: non backup subflows pre-reqs

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1072 bytes --]

This series contains a bunch of MPTCP improvements loosely related to
concurrent subflows xmit usage, currently under development.

The first 3 patches are actually bugfixes for issues that will become apparent
as soon as we will enable the above feature.

The later patches improve the handling of incoming additional subflows,
improving significantly the performances in stress tests based on a high new
connection rate.

Paolo Abeni (8):
  subflow: always init 'rel_write_seq'
  mptcp: avoid data corruption on reinsert
  mptcp: mark as fallback even early ones
  mptcp: explicitly track the fully established status
  mptcp: cleanup subflow_finish_connect()
  subflow: explicitly check for plain tcp rsk
  subflow: use rsk_ops->send_reset()
  subflow: introduce and use mptcp_can_accept_new_subflow()

 net/mptcp/options.c  |  5 +--
 net/mptcp/protocol.c | 23 ++++++++---
 net/mptcp/protocol.h |  8 ++++
 net/mptcp/subflow.c  | 91 ++++++++++++++++++++++++++------------------
 4 files changed, 81 insertions(+), 46 deletions(-)

-- 
2.26.2

[PATCH net-next 0/8] mptcp: non backup subflows pre-reqs

[Paolo Abeni]

This series contains a bunch of MPTCP improvements loosely related to
concurrent subflows xmit usage, currently under development.

The first 3 patches are actually bugfixes for issues that will become apparent
as soon as we will enable the above feature.

The later patches improve the handling of incoming additional subflows,
improving significantly the performances in stress tests based on a high new
connection rate.

Paolo Abeni (8):
  subflow: always init 'rel_write_seq'
  mptcp: avoid data corruption on reinsert
  mptcp: mark as fallback even early ones
  mptcp: explicitly track the fully established status
  mptcp: cleanup subflow_finish_connect()
  subflow: explicitly check for plain tcp rsk
  subflow: use rsk_ops->send_reset()
  subflow: introduce and use mptcp_can_accept_new_subflow()

 net/mptcp/options.c  |  5 +--
 net/mptcp/protocol.c | 23 ++++++++---
 net/mptcp/protocol.h |  8 ++++
 net/mptcp/subflow.c  | 91 ++++++++++++++++++++++++++------------------
 4 files changed, 81 insertions(+), 46 deletions(-)

-- 
2.26.2

[MPTCP] [PATCH net-next 1/8] subflow: always init 'rel_write_seq'

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1378 bytes --]

Currently we do not init the subflow write sequence for
MP_JOIN subflows. This will cause bad mapping being
generated as soon as we will use non backup subflow.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/protocol.c | 1 -
 net/mptcp/subflow.c  | 1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index f0b0b503c262..59c0eef807b3 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1814,7 +1814,6 @@ void mptcp_finish_connect(struct sock *ssk)
 	ack_seq++;
 	subflow->map_seq = ack_seq;
 	subflow->map_subflow_seq = 1;
-	subflow->rel_write_seq = 1;
 
 	/* the socket is not connected yet, no msk/subflow ops can access/race
 	 * accessing the field below
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 519122e66f17..84e70806b250 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -200,6 +200,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 	if (subflow->conn_finished)
 		return;
 
+	subflow->rel_write_seq = 1;
 	subflow->conn_finished = 1;
 	subflow->ssn_offset = TCP_SKB_CB(skb)->seq;
 	pr_debug("subflow=%p synack seq=%x", subflow, subflow->ssn_offset);
-- 
2.26.2

[PATCH net-next 1/8] subflow: always init 'rel_write_seq'

[Paolo Abeni]

Currently we do not init the subflow write sequence for
MP_JOIN subflows. This will cause bad mapping being
generated as soon as we will use non backup subflow.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 1 -
 net/mptcp/subflow.c  | 1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index f0b0b503c262..59c0eef807b3 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1814,7 +1814,6 @@ void mptcp_finish_connect(struct sock *ssk)
 	ack_seq++;
 	subflow->map_seq = ack_seq;
 	subflow->map_subflow_seq = 1;
-	subflow->rel_write_seq = 1;
 
 	/* the socket is not connected yet, no msk/subflow ops can access/race
 	 * accessing the field below
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 519122e66f17..84e70806b250 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -200,6 +200,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 	if (subflow->conn_finished)
 		return;
 
+	subflow->rel_write_seq = 1;
 	subflow->conn_finished = 1;
 	subflow->ssn_offset = TCP_SKB_CB(skb)->seq;
 	pr_debug("subflow=%p synack seq=%x", subflow, subflow->ssn_offset);
-- 
2.26.2

[MPTCP] [PATCH net-next 2/8] mptcp: avoid data corruption on reinsert

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1260 bytes --]

When updating a partially acked data fragment, we
actually corrupt it. This is irrelevant till we send
data on a single subflow, as retransmitted data, if
any are discarded by the peer as duplicate, but it
will cause data corruption as soon as we will start
creating non backup subflows.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/protocol.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 59c0eef807b3..254e6ef2b4e0 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -460,15 +460,20 @@ static void mptcp_clean_una(struct sock *sk)
 
 	dfrag = mptcp_rtx_head(sk);
 	if (dfrag && after64(snd_una, dfrag->data_seq)) {
-		u64 delta = dfrag->data_seq + dfrag->data_len - snd_una;
+		u64 delta = snd_una - dfrag->data_seq;
+
+		if (WARN_ON_ONCE(delta > dfrag->data_len))
+			goto out;
 
 		dfrag->data_seq += delta;
+		dfrag->offset += delta;
 		dfrag->data_len -= delta;
 
 		dfrag_uncharge(sk, delta);
 		cleaned = true;
 	}
 
+out:
 	if (cleaned) {
 		sk_mem_reclaim_partial(sk);
 
-- 
2.26.2

[PATCH net-next 2/8] mptcp: avoid data corruption on reinsert

[Paolo Abeni]

When updating a partially acked data fragment, we
actually corrupt it. This is irrelevant till we send
data on a single subflow, as retransmitted data, if
any are discarded by the peer as duplicate, but it
will cause data corruption as soon as we will start
creating non backup subflows.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 59c0eef807b3..254e6ef2b4e0 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -460,15 +460,20 @@ static void mptcp_clean_una(struct sock *sk)
 
 	dfrag = mptcp_rtx_head(sk);
 	if (dfrag && after64(snd_una, dfrag->data_seq)) {
-		u64 delta = dfrag->data_seq + dfrag->data_len - snd_una;
+		u64 delta = snd_una - dfrag->data_seq;
+
+		if (WARN_ON_ONCE(delta > dfrag->data_len))
+			goto out;
 
 		dfrag->data_seq += delta;
+		dfrag->offset += delta;
 		dfrag->data_len -= delta;
 
 		dfrag_uncharge(sk, delta);
 		cleaned = true;
 	}
 
+out:
 	if (cleaned) {
 		sk_mem_reclaim_partial(sk);
 
-- 
2.26.2

[MPTCP] [PATCH net-next 3/8] mptcp: mark as fallback even early ones

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1771 bytes --]

In the unlikely event of a failure at connect time,
we currently clear the request_mptcp flag - so that
the MPC handshake is not started at all, but the msk
is not explicitly marked as fallback.

This would lead to later insertion of wrong DSS options
in the xmitted packets, in violation of RFC specs and
possibly fooling the peer.

Fixes: e1ff9e82e2ea ("net: mptcp: improve fallback to TCP")
Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/protocol.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 254e6ef2b4e0..2936413171be 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1944,6 +1944,13 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	return err;
 }
 
+static void mptcp_subflow_early_fallback(struct mptcp_sock *msk,
+					 struct mptcp_subflow_context *subflow)
+{
+	subflow->request_mptcp = 0;
+	__mptcp_do_fallback(msk);
+}
+
 static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 				int addr_len, int flags)
 {
@@ -1975,10 +1982,10 @@ static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 	 * TCP option space.
 	 */
 	if (rcu_access_pointer(tcp_sk(ssock->sk)->md5sig_info))
-		subflow->request_mptcp = 0;
+		mptcp_subflow_early_fallback(msk, subflow);
 #endif
 	if (subflow->request_mptcp && mptcp_token_new_connect(ssock->sk))
-		subflow->request_mptcp = 0;
+		mptcp_subflow_early_fallback(msk, subflow);
 
 do_connect:
 	err = ssock->ops->connect(ssock, uaddr, addr_len, flags);
-- 
2.26.2

[PATCH net-next 3/8] mptcp: mark as fallback even early ones

[Paolo Abeni]

In the unlikely event of a failure at connect time,
we currently clear the request_mptcp flag - so that
the MPC handshake is not started at all, but the msk
is not explicitly marked as fallback.

This would lead to later insertion of wrong DSS options
in the xmitted packets, in violation of RFC specs and
possibly fooling the peer.

Fixes: e1ff9e82e2ea ("net: mptcp: improve fallback to TCP")
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 254e6ef2b4e0..2936413171be 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1944,6 +1944,13 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	return err;
 }
 
+static void mptcp_subflow_early_fallback(struct mptcp_sock *msk,
+					 struct mptcp_subflow_context *subflow)
+{
+	subflow->request_mptcp = 0;
+	__mptcp_do_fallback(msk);
+}
+
 static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 				int addr_len, int flags)
 {
@@ -1975,10 +1982,10 @@ static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 	 * TCP option space.
 	 */
 	if (rcu_access_pointer(tcp_sk(ssock->sk)->md5sig_info))
-		subflow->request_mptcp = 0;
+		mptcp_subflow_early_fallback(msk, subflow);
 #endif
 	if (subflow->request_mptcp && mptcp_token_new_connect(ssock->sk))
-		subflow->request_mptcp = 0;
+		mptcp_subflow_early_fallback(msk, subflow);
 
 do_connect:
 	err = ssock->ops->connect(ssock, uaddr, addr_len, flags);
-- 
2.26.2

[MPTCP] [PATCH net-next 4/8] mptcp: explicitly track the fully established status

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 5973 bytes --]

Currently accepted msk sockets become established only after
accept() returns the new sk to user-space.

As MP_JOIN request are refused as per RFC spec on non fully
established socket, the above causes mp_join self-tests
instabilities.

This change lets the msk entering the established status
as soon as it receives the 3rd ack and propagates the first
subflow fully established status on the msk socket.

Finally we can change the subflow acceptance condition to
take in account both the sock state and the msk fully
established flag.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/options.c  |  5 ++---
 net/mptcp/protocol.c |  4 ++--
 net/mptcp/protocol.h |  8 ++++++++
 net/mptcp/subflow.c  | 23 +++++++++++++++++++----
 4 files changed, 31 insertions(+), 9 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 19707c07efc1..3bc56eb608d8 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -709,6 +709,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk,
 		 * additional ack.
 		 */
 		subflow->fully_established = 1;
+		WRITE_ONCE(msk->fully_established, true);
 		goto fully_established;
 	}
 
@@ -724,9 +725,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk,
 
 	if (unlikely(!READ_ONCE(msk->pm.server_side)))
 		pr_warn_once("bogus mpc option on established client sk");
-	subflow->fully_established = 1;
-	subflow->remote_key = mp_opt->sndr_key;
-	subflow->can_ack = 1;
+	mptcp_subflow_fully_established(subflow, mp_opt);
 
 fully_established:
 	if (likely(subflow->pm_notified))
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 2936413171be..979dfcd2aa14 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1522,6 +1522,7 @@ struct sock *mptcp_sk_clone(const struct sock *sk,
 	msk->local_key = subflow_req->local_key;
 	msk->token = subflow_req->token;
 	msk->subflow = NULL;
+	WRITE_ONCE(msk->fully_established, false);
 
 	msk->write_seq = subflow_req->idsn + 1;
 	atomic64_set(&msk->snd_una, msk->write_seq);
@@ -1605,7 +1606,6 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
-		inet_sk_state_store(newsk, TCP_ESTABLISHED);
 
 		mptcp_rcv_space_init(msk, ssk);
 		bh_unlock_sock(new_mptcp_sock);
@@ -1855,7 +1855,7 @@ bool mptcp_finish_join(struct sock *sk)
 	pr_debug("msk=%p, subflow=%p", msk, subflow);
 
 	/* mptcp socket already closing? */
-	if (inet_sk_state_load(parent) != TCP_ESTABLISHED)
+	if (!mptcp_is_fully_established(parent))
 		return false;
 
 	if (!msk->pm.server_side)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 6e114c09e5b4..67634b595466 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -198,6 +198,7 @@ struct mptcp_sock {
 	u32		token;
 	unsigned long	flags;
 	bool		can_ack;
+	bool		fully_established;
 	spinlock_t	join_list_lock;
 	struct work_struct work;
 	struct list_head conn_list;
@@ -342,6 +343,8 @@ mptcp_subflow_get_mapped_dsn(const struct mptcp_subflow_context *subflow)
 }
 
 int mptcp_is_enabled(struct net *net);
+void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow,
+				     struct mptcp_options_received *mp_opt);
 bool mptcp_subflow_data_available(struct sock *sk);
 void __init mptcp_subflow_init(void);
 
@@ -373,6 +376,11 @@ void mptcp_get_options(const struct sk_buff *skb,
 		       struct mptcp_options_received *mp_opt);
 
 void mptcp_finish_connect(struct sock *sk);
+static inline bool mptcp_is_fully_established(struct sock *sk)
+{
+	return inet_sk_state_load(sk) == TCP_ESTABLISHED &&
+	       READ_ONCE(mptcp_sk(sk)->fully_established);
+}
 void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk);
 void mptcp_data_ready(struct sock *sk, struct sock *ssk);
 bool mptcp_finish_join(struct sock *sk);
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 84e70806b250..ea81842fc3b2 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -387,6 +387,17 @@ static void subflow_drop_ctx(struct sock *ssk)
 	kfree_rcu(ctx, rcu);
 }
 
+void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow,
+				     struct mptcp_options_received *mp_opt)
+{
+	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
+
+	subflow->remote_key = mp_opt->sndr_key;
+	subflow->fully_established = 1;
+	subflow->can_ack = 1;
+	WRITE_ONCE(msk->fully_established, true);
+}
+
 static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 					  struct sk_buff *skb,
 					  struct request_sock *req,
@@ -466,6 +477,11 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 		}
 
 		if (ctx->mp_capable) {
+			/* this can't race with mptcp_close(), as the msk is
+			 * not yet exposted to user-space
+			 */
+			inet_sk_state_store((void *)new_msk, TCP_ESTABLISHED);
+
 			/* new mpc subflow takes ownership of the newly
 			 * created mptcp socket
 			 */
@@ -478,9 +494,8 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 			/* with OoO packets we can reach here without ingress
 			 * mpc option
 			 */
-			ctx->remote_key = mp_opt.sndr_key;
-			ctx->fully_established = mp_opt.mp_capable;
-			ctx->can_ack = mp_opt.mp_capable;
+			if (mp_opt.mp_capable)
+				mptcp_subflow_fully_established(ctx, &mp_opt);
 		} else if (ctx->mp_join) {
 			struct mptcp_sock *owner;
 
@@ -967,7 +982,7 @@ int __mptcp_subflow_connect(struct sock *sk, int ifindex,
 	int addrlen;
 	int err;
 
-	if (sk->sk_state != TCP_ESTABLISHED)
+	if (!mptcp_is_fully_established(sk))
 		return -ENOTCONN;
 
 	err = mptcp_subflow_create_socket(sk, &sf);
-- 
2.26.2

[PATCH net-next 4/8] mptcp: explicitly track the fully established status

[Paolo Abeni]

Currently accepted msk sockets become established only after
accept() returns the new sk to user-space.

As MP_JOIN request are refused as per RFC spec on non fully
established socket, the above causes mp_join self-tests
instabilities.

This change lets the msk entering the established status
as soon as it receives the 3rd ack and propagates the first
subflow fully established status on the msk socket.

Finally we can change the subflow acceptance condition to
take in account both the sock state and the msk fully
established flag.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/options.c  |  5 ++---
 net/mptcp/protocol.c |  4 ++--
 net/mptcp/protocol.h |  8 ++++++++
 net/mptcp/subflow.c  | 23 +++++++++++++++++++----
 4 files changed, 31 insertions(+), 9 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 19707c07efc1..3bc56eb608d8 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -709,6 +709,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk,
 		 * additional ack.
 		 */
 		subflow->fully_established = 1;
+		WRITE_ONCE(msk->fully_established, true);
 		goto fully_established;
 	}
 
@@ -724,9 +725,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk,
 
 	if (unlikely(!READ_ONCE(msk->pm.server_side)))
 		pr_warn_once("bogus mpc option on established client sk");
-	subflow->fully_established = 1;
-	subflow->remote_key = mp_opt->sndr_key;
-	subflow->can_ack = 1;
+	mptcp_subflow_fully_established(subflow, mp_opt);
 
 fully_established:
 	if (likely(subflow->pm_notified))
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 2936413171be..979dfcd2aa14 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1522,6 +1522,7 @@ struct sock *mptcp_sk_clone(const struct sock *sk,
 	msk->local_key = subflow_req->local_key;
 	msk->token = subflow_req->token;
 	msk->subflow = NULL;
+	WRITE_ONCE(msk->fully_established, false);
 
 	msk->write_seq = subflow_req->idsn + 1;
 	atomic64_set(&msk->snd_una, msk->write_seq);
@@ -1605,7 +1606,6 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
-		inet_sk_state_store(newsk, TCP_ESTABLISHED);
 
 		mptcp_rcv_space_init(msk, ssk);
 		bh_unlock_sock(new_mptcp_sock);
@@ -1855,7 +1855,7 @@ bool mptcp_finish_join(struct sock *sk)
 	pr_debug("msk=%p, subflow=%p", msk, subflow);
 
 	/* mptcp socket already closing? */
-	if (inet_sk_state_load(parent) != TCP_ESTABLISHED)
+	if (!mptcp_is_fully_established(parent))
 		return false;
 
 	if (!msk->pm.server_side)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 6e114c09e5b4..67634b595466 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -198,6 +198,7 @@ struct mptcp_sock {
 	u32		token;
 	unsigned long	flags;
 	bool		can_ack;
+	bool		fully_established;
 	spinlock_t	join_list_lock;
 	struct work_struct work;
 	struct list_head conn_list;
@@ -342,6 +343,8 @@ mptcp_subflow_get_mapped_dsn(const struct mptcp_subflow_context *subflow)
 }
 
 int mptcp_is_enabled(struct net *net);
+void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow,
+				     struct mptcp_options_received *mp_opt);
 bool mptcp_subflow_data_available(struct sock *sk);
 void __init mptcp_subflow_init(void);
 
@@ -373,6 +376,11 @@ void mptcp_get_options(const struct sk_buff *skb,
 		       struct mptcp_options_received *mp_opt);
 
 void mptcp_finish_connect(struct sock *sk);
+static inline bool mptcp_is_fully_established(struct sock *sk)
+{
+	return inet_sk_state_load(sk) == TCP_ESTABLISHED &&
+	       READ_ONCE(mptcp_sk(sk)->fully_established);
+}
 void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk);
 void mptcp_data_ready(struct sock *sk, struct sock *ssk);
 bool mptcp_finish_join(struct sock *sk);
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 84e70806b250..ea81842fc3b2 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -387,6 +387,17 @@ static void subflow_drop_ctx(struct sock *ssk)
 	kfree_rcu(ctx, rcu);
 }
 
+void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow,
+				     struct mptcp_options_received *mp_opt)
+{
+	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
+
+	subflow->remote_key = mp_opt->sndr_key;
+	subflow->fully_established = 1;
+	subflow->can_ack = 1;
+	WRITE_ONCE(msk->fully_established, true);
+}
+
 static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 					  struct sk_buff *skb,
 					  struct request_sock *req,
@@ -466,6 +477,11 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 		}
 
 		if (ctx->mp_capable) {
+			/* this can't race with mptcp_close(), as the msk is
+			 * not yet exposted to user-space
+			 */
+			inet_sk_state_store((void *)new_msk, TCP_ESTABLISHED);
+
 			/* new mpc subflow takes ownership of the newly
 			 * created mptcp socket
 			 */
@@ -478,9 +494,8 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 			/* with OoO packets we can reach here without ingress
 			 * mpc option
 			 */
-			ctx->remote_key = mp_opt.sndr_key;
-			ctx->fully_established = mp_opt.mp_capable;
-			ctx->can_ack = mp_opt.mp_capable;
+			if (mp_opt.mp_capable)
+				mptcp_subflow_fully_established(ctx, &mp_opt);
 		} else if (ctx->mp_join) {
 			struct mptcp_sock *owner;
 
@@ -967,7 +982,7 @@ int __mptcp_subflow_connect(struct sock *sk, int ifindex,
 	int addrlen;
 	int err;
 
-	if (sk->sk_state != TCP_ESTABLISHED)
+	if (!mptcp_is_fully_established(sk))
 		return -ENOTCONN;
 
 	err = mptcp_subflow_create_socket(sk, &sf);
-- 
2.26.2

[MPTCP] [PATCH net-next 5/8] mptcp: cleanup subflow_finish_connect()

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 3282 bytes --]

The mentioned function has several unneeded branches,
handle each case - MP_CAPABLE, MP_JOIN, fallback -
under a single conditional and drop quite a bit of
duplicate code.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/subflow.c | 56 ++++++++++++++++++++-------------------------
 1 file changed, 25 insertions(+), 31 deletions(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index ea81842fc3b2..7f3ef1840df5 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -206,44 +206,34 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 	pr_debug("subflow=%p synack seq=%x", subflow, subflow->ssn_offset);
 
 	mptcp_get_options(skb, &mp_opt);
-	if (subflow->request_mptcp && mp_opt.mp_capable) {
+	if (subflow->request_mptcp) {
+		if (!mp_opt.mp_capable) {
+			MPTCP_INC_STATS(sock_net(sk),
+					MPTCP_MIB_MPCAPABLEACTIVEFALLBACK);
+			mptcp_do_fallback(sk);
+			pr_fallback(mptcp_sk(subflow->conn));
+			goto fallback;
+		}
+
 		subflow->mp_capable = 1;
 		subflow->can_ack = 1;
 		subflow->remote_key = mp_opt.sndr_key;
 		pr_debug("subflow=%p, remote_key=%llu", subflow,
 			 subflow->remote_key);
-	} else if (subflow->request_join && mp_opt.mp_join) {
-		subflow->mp_join = 1;
+		mptcp_finish_connect(sk);
+	} else if (subflow->request_join) {
+		u8 hmac[SHA256_DIGEST_SIZE];
+
+		if (!mp_opt.mp_join)
+			goto do_reset;
+
 		subflow->thmac = mp_opt.thmac;
 		subflow->remote_nonce = mp_opt.nonce;
 		pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u", subflow,
 			 subflow->thmac, subflow->remote_nonce);
-	} else {
-		if (subflow->request_mptcp)
-			MPTCP_INC_STATS(sock_net(sk),
-					MPTCP_MIB_MPCAPABLEACTIVEFALLBACK);
-		mptcp_do_fallback(sk);
-		pr_fallback(mptcp_sk(subflow->conn));
-	}
 
-	if (mptcp_check_fallback(sk)) {
-		mptcp_rcv_space_init(mptcp_sk(parent), sk);
-		return;
-	}
-
-	if (subflow->mp_capable) {
-		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
-			 subflow->remote_key);
-		mptcp_finish_connect(sk);
-	} else if (subflow->mp_join) {
-		u8 hmac[SHA256_DIGEST_SIZE];
-
-		pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u",
-			 subflow, subflow->thmac,
-			 subflow->remote_nonce);
 		if (!subflow_thmac_valid(subflow)) {
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINACKMAC);
-			subflow->mp_join = 0;
 			goto do_reset;
 		}
 
@@ -251,18 +241,22 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 				      subflow->local_nonce,
 				      subflow->remote_nonce,
 				      hmac);
-
 		memcpy(subflow->hmac, hmac, MPTCPOPT_HMAC_LEN);
 
 		if (!mptcp_finish_join(sk))
 			goto do_reset;
 
+		subflow->mp_join = 1;
 		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNACKRX);
-	} else {
-do_reset:
-		tcp_send_active_reset(sk, GFP_ATOMIC);
-		tcp_done(sk);
+	} else if (mptcp_check_fallback(sk)) {
+fallback:
+		mptcp_rcv_space_init(mptcp_sk(parent), sk);
 	}
+	return;
+
+do_reset:
+	tcp_send_active_reset(sk, GFP_ATOMIC);
+	tcp_done(sk);
 }
 
 static struct request_sock_ops subflow_request_sock_ops;
-- 
2.26.2

[PATCH net-next 5/8] mptcp: cleanup subflow_finish_connect()

[Paolo Abeni]

The mentioned function has several unneeded branches,
handle each case - MP_CAPABLE, MP_JOIN, fallback -
under a single conditional and drop quite a bit of
duplicate code.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/subflow.c | 56 ++++++++++++++++++++-------------------------
 1 file changed, 25 insertions(+), 31 deletions(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index ea81842fc3b2..7f3ef1840df5 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -206,44 +206,34 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 	pr_debug("subflow=%p synack seq=%x", subflow, subflow->ssn_offset);
 
 	mptcp_get_options(skb, &mp_opt);
-	if (subflow->request_mptcp && mp_opt.mp_capable) {
+	if (subflow->request_mptcp) {
+		if (!mp_opt.mp_capable) {
+			MPTCP_INC_STATS(sock_net(sk),
+					MPTCP_MIB_MPCAPABLEACTIVEFALLBACK);
+			mptcp_do_fallback(sk);
+			pr_fallback(mptcp_sk(subflow->conn));
+			goto fallback;
+		}
+
 		subflow->mp_capable = 1;
 		subflow->can_ack = 1;
 		subflow->remote_key = mp_opt.sndr_key;
 		pr_debug("subflow=%p, remote_key=%llu", subflow,
 			 subflow->remote_key);
-	} else if (subflow->request_join && mp_opt.mp_join) {
-		subflow->mp_join = 1;
+		mptcp_finish_connect(sk);
+	} else if (subflow->request_join) {
+		u8 hmac[SHA256_DIGEST_SIZE];
+
+		if (!mp_opt.mp_join)
+			goto do_reset;
+
 		subflow->thmac = mp_opt.thmac;
 		subflow->remote_nonce = mp_opt.nonce;
 		pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u", subflow,
 			 subflow->thmac, subflow->remote_nonce);
-	} else {
-		if (subflow->request_mptcp)
-			MPTCP_INC_STATS(sock_net(sk),
-					MPTCP_MIB_MPCAPABLEACTIVEFALLBACK);
-		mptcp_do_fallback(sk);
-		pr_fallback(mptcp_sk(subflow->conn));
-	}
 
-	if (mptcp_check_fallback(sk)) {
-		mptcp_rcv_space_init(mptcp_sk(parent), sk);
-		return;
-	}
-
-	if (subflow->mp_capable) {
-		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
-			 subflow->remote_key);
-		mptcp_finish_connect(sk);
-	} else if (subflow->mp_join) {
-		u8 hmac[SHA256_DIGEST_SIZE];
-
-		pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u",
-			 subflow, subflow->thmac,
-			 subflow->remote_nonce);
 		if (!subflow_thmac_valid(subflow)) {
 			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINACKMAC);
-			subflow->mp_join = 0;
 			goto do_reset;
 		}
 
@@ -251,18 +241,22 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 				      subflow->local_nonce,
 				      subflow->remote_nonce,
 				      hmac);
-
 		memcpy(subflow->hmac, hmac, MPTCPOPT_HMAC_LEN);
 
 		if (!mptcp_finish_join(sk))
 			goto do_reset;
 
+		subflow->mp_join = 1;
 		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNACKRX);
-	} else {
-do_reset:
-		tcp_send_active_reset(sk, GFP_ATOMIC);
-		tcp_done(sk);
+	} else if (mptcp_check_fallback(sk)) {
+fallback:
+		mptcp_rcv_space_init(mptcp_sk(parent), sk);
 	}
+	return;
+
+do_reset:
+	tcp_send_active_reset(sk, GFP_ATOMIC);
+	tcp_done(sk);
 }
 
 static struct request_sock_ops subflow_request_sock_ops;
-- 
2.26.2

[MPTCP] [PATCH net-next 6/8] subflow: explicitly check for plain tcp rsk

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1039 bytes --]

When syncookie are in use, the TCP stack may feed into
subflow_syn_recv_sock() plain TCP request sockets. We can't
access mptcp_subflow_request_sock-specific fields on such
sockets. Explicitly check the rsk ops to do safe accesses.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/subflow.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 7f3ef1840df5..3ef445f59556 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -415,7 +415,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	/* hopefully temporary handling for MP_JOIN+syncookie */
 	subflow_req = mptcp_subflow_rsk(req);
-	fallback_is_fatal = subflow_req->mp_join;
+	fallback_is_fatal = tcp_rsk(req)->is_mptcp && subflow_req->mp_join;
 	fallback = !tcp_rsk(req)->is_mptcp;
 	if (fallback)
 		goto create_child;
-- 
2.26.2

[PATCH net-next 6/8] subflow: explicitly check for plain tcp rsk

[Paolo Abeni]

When syncookie are in use, the TCP stack may feed into
subflow_syn_recv_sock() plain TCP request sockets. We can't
access mptcp_subflow_request_sock-specific fields on such
sockets. Explicitly check the rsk ops to do safe accesses.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/subflow.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 7f3ef1840df5..3ef445f59556 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -415,7 +415,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	/* hopefully temporary handling for MP_JOIN+syncookie */
 	subflow_req = mptcp_subflow_rsk(req);
-	fallback_is_fatal = subflow_req->mp_join;
+	fallback_is_fatal = tcp_rsk(req)->is_mptcp && subflow_req->mp_join;
 	fallback = !tcp_rsk(req)->is_mptcp;
 	if (fallback)
 		goto create_child;
-- 
2.26.2

[MPTCP] [PATCH net-next 7/8] subflow: use rsk_ops->send_reset()

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1008 bytes --]

tcp_send_active_reset() is more prone to transient errors
(memory allocation or xmit queue full): in stress conditions
the kernel may drop the egress packet, and the client will be
stuck.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/subflow.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 3ef445f59556..ada04df6f99f 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -524,9 +524,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 dispose_child:
 	subflow_drop_ctx(child);
 	tcp_rsk(req)->drop_req = true;
-	tcp_send_active_reset(child, GFP_ATOMIC);
 	inet_csk_prepare_for_destroy_sock(child);
 	tcp_done(child);
+	req->rsk_ops->send_reset(sk, skb);
 
 	/* The last child reference will be released by the caller */
 	return child;
-- 
2.26.2

[PATCH net-next 7/8] subflow: use rsk_ops->send_reset()

[Paolo Abeni]

tcp_send_active_reset() is more prone to transient errors
(memory allocation or xmit queue full): in stress conditions
the kernel may drop the egress packet, and the client will be
stuck.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/subflow.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 3ef445f59556..ada04df6f99f 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -524,9 +524,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 dispose_child:
 	subflow_drop_ctx(child);
 	tcp_rsk(req)->drop_req = true;
-	tcp_send_active_reset(child, GFP_ATOMIC);
 	inet_csk_prepare_for_destroy_sock(child);
 	tcp_done(child);
+	req->rsk_ops->send_reset(sk, skb);
 
 	/* The last child reference will be released by the caller */
 	return child;
-- 
2.26.2

[MPTCP] [PATCH net-next 8/8] subflow: introduce and use mptcp_can_accept_new_subflow()

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1424 bytes --]

So that we can easily perform some basic PM-related
adimission checks before creating the child socket.

Reviewed-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com>
Tested-by: Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by: Paolo Abeni <pabeni(a)redhat.com>
---
 net/mptcp/subflow.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index ada04df6f99f..e645483d1200 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -53,6 +53,12 @@ static void subflow_generate_hmac(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
 	mptcp_crypto_hmac_sha(key1, key2, msg, 8, hmac);
 }
 
+static bool mptcp_can_accept_new_subflow(const struct mptcp_sock *msk)
+{
+	return mptcp_is_fully_established((void *)msk) &&
+	       READ_ONCE(msk->pm.accept_subflow);
+}
+
 /* validate received token and create truncated hmac and nonce for SYN-ACK */
 static struct mptcp_sock *subflow_token_join_request(struct request_sock *req,
 						     const struct sk_buff *skb)
@@ -443,6 +449,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 	} else if (subflow_req->mp_join) {
 		mptcp_get_options(skb, &mp_opt);
 		if (!mp_opt.mp_join ||
+		    !mptcp_can_accept_new_subflow(subflow_req->msk) ||
 		    !subflow_hmac_valid(req, &mp_opt)) {
 			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC);
 			fallback = true;
-- 
2.26.2

[PATCH net-next 8/8] subflow: introduce and use mptcp_can_accept_new_subflow()

[Paolo Abeni]

So that we can easily perform some basic PM-related
adimission checks before creating the child socket.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/subflow.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index ada04df6f99f..e645483d1200 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -53,6 +53,12 @@ static void subflow_generate_hmac(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
 	mptcp_crypto_hmac_sha(key1, key2, msg, 8, hmac);
 }
 
+static bool mptcp_can_accept_new_subflow(const struct mptcp_sock *msk)
+{
+	return mptcp_is_fully_established((void *)msk) &&
+	       READ_ONCE(msk->pm.accept_subflow);
+}
+
 /* validate received token and create truncated hmac and nonce for SYN-ACK */
 static struct mptcp_sock *subflow_token_join_request(struct request_sock *req,
 						     const struct sk_buff *skb)
@@ -443,6 +449,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 	} else if (subflow_req->mp_join) {
 		mptcp_get_options(skb, &mp_opt);
 		if (!mp_opt.mp_join ||
+		    !mptcp_can_accept_new_subflow(subflow_req->msk) ||
 		    !subflow_hmac_valid(req, &mp_opt)) {
 			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC);
 			fallback = true;
-- 
2.26.2

[MPTCP] Re: [PATCH net-next 0/8] mptcp: non backup subflows pre-reqs

[David Miller]

[-- Attachment #1: Type: text/plain, Size: 557 bytes --]

From: Paolo Abeni <pabeni(a)redhat.com>
Date: Thu, 23 Jul 2020 13:02:28 +0200

> This series contains a bunch of MPTCP improvements loosely related to
> concurrent subflows xmit usage, currently under development.
> 
> The first 3 patches are actually bugfixes for issues that will become apparent
> as soon as we will enable the above feature.
> 
> The later patches improve the handling of incoming additional subflows,
> improving significantly the performances in stress tests based on a high new
> connection rate.

Series applied, thanks!

Re: [PATCH net-next 0/8] mptcp: non backup subflows pre-reqs

[David Miller]

From: Paolo Abeni <pabeni@redhat.com>
Date: Thu, 23 Jul 2020 13:02:28 +0200

> This series contains a bunch of MPTCP improvements loosely related to
> concurrent subflows xmit usage, currently under development.
> 
> The first 3 patches are actually bugfixes for issues that will become apparent
> as soon as we will enable the above feature.
> 
> The later patches improve the handling of incoming additional subflows,
> improving significantly the performances in stress tests based on a high new
> connection rate.

Series applied, thanks!