[MPTCP] [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[MPTCP] [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[Davide Caratti]

[-- Attachment #1: Type: text/plain, Size: 1982 bytes --]

it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
flags/version during the three-way handshake). Adjust MPTCP socket state
earlier, to ensure correct functionality of mptcp_sendmsg() even in case
of TCP fallback.

Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
Signed-off-by: Davide Caratti <dcaratti(a)redhat.com>
---
 net/mptcp/protocol.c | 4 ----
 net/mptcp/subflow.c  | 6 ++++++
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index e959104832ef..92d5382e71f4 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1055,10 +1055,6 @@ void mptcp_finish_connect(struct sock *ssk)
 	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
 	WRITE_ONCE(msk->ack_seq, ack_seq);
 	WRITE_ONCE(msk->can_ack, 1);
-	if (inet_sk_state_load(sk) != TCP_ESTABLISHED) {
-		inet_sk_state_store(sk, TCP_ESTABLISHED);
-		sk->sk_state_change(sk);
-	}
 }
 
 static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 052d72a1d3a2..06b9075333c5 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -109,9 +109,15 @@ static void subflow_v6_init_req(struct request_sock *req,
 static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sock *parent = subflow->conn;
 
 	subflow->icsk_af_ops->sk_rx_dst_set(sk, skb);
 
+	if (inet_sk_state_load(parent) != TCP_ESTABLISHED) {
+		inet_sk_state_store(parent, TCP_ESTABLISHED);
+		parent->sk_state_change(parent);
+	}
+
 	if (!subflow->conn_finished) {
 		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
 			 subflow->remote_key);
-- 
2.24.1

[PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[Davide Caratti]

it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
flags/version during the three-way handshake). Adjust MPTCP socket state
earlier, to ensure correct functionality of mptcp_sendmsg() even in case
of TCP fallback.

Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
---
 net/mptcp/protocol.c | 4 ----
 net/mptcp/subflow.c  | 6 ++++++
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index e959104832ef..92d5382e71f4 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1055,10 +1055,6 @@ void mptcp_finish_connect(struct sock *ssk)
 	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
 	WRITE_ONCE(msk->ack_seq, ack_seq);
 	WRITE_ONCE(msk->can_ack, 1);
-	if (inet_sk_state_load(sk) != TCP_ESTABLISHED) {
-		inet_sk_state_store(sk, TCP_ESTABLISHED);
-		sk->sk_state_change(sk);
-	}
 }
 
 static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 052d72a1d3a2..06b9075333c5 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -109,9 +109,15 @@ static void subflow_v6_init_req(struct request_sock *req,
 static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sock *parent = subflow->conn;
 
 	subflow->icsk_af_ops->sk_rx_dst_set(sk, skb);
 
+	if (inet_sk_state_load(parent) != TCP_ESTABLISHED) {
+		inet_sk_state_store(parent, TCP_ESTABLISHED);
+		parent->sk_state_change(parent);
+	}
+
 	if (!subflow->conn_finished) {
 		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
 			 subflow->remote_key);
-- 
2.24.1

[MPTCP] Re: [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 2196 bytes --]

On Thu, 2020-03-19 at 22:45 +0100, Davide Caratti wrote:
> it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
> MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
> flags/version during the three-way handshake). Adjust MPTCP socket state
> earlier, to ensure correct functionality of mptcp_sendmsg() even in case
> of TCP fallback.
> 
> Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
> Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
> Signed-off-by: Davide Caratti <dcaratti(a)redhat.com>
> ---
>  net/mptcp/protocol.c | 4 ----
>  net/mptcp/subflow.c  | 6 ++++++
>  2 files changed, 6 insertions(+), 4 deletions(-)
> 
> diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> index e959104832ef..92d5382e71f4 100644
> --- a/net/mptcp/protocol.c
> +++ b/net/mptcp/protocol.c
> @@ -1055,10 +1055,6 @@ void mptcp_finish_connect(struct sock *ssk)
>  	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
>  	WRITE_ONCE(msk->ack_seq, ack_seq);
>  	WRITE_ONCE(msk->can_ack, 1);
> -	if (inet_sk_state_load(sk) != TCP_ESTABLISHED) {
> -		inet_sk_state_store(sk, TCP_ESTABLISHED);
> -		sk->sk_state_change(sk);
> -	}
>  }
>  
>  static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
> diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
> index 052d72a1d3a2..06b9075333c5 100644
> --- a/net/mptcp/subflow.c
> +++ b/net/mptcp/subflow.c
> @@ -109,9 +109,15 @@ static void subflow_v6_init_req(struct request_sock *req,
>  static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
>  {
>  	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
> +	struct sock *parent = subflow->conn;
>  
>  	subflow->icsk_af_ops->sk_rx_dst_set(sk, skb);
>  
> +	if (inet_sk_state_load(parent) != TCP_ESTABLISHED) {
> +		inet_sk_state_store(parent, TCP_ESTABLISHED);
> +		parent->sk_state_change(parent);
> +	}
> +
>  	if (!subflow->conn_finished) {
>  		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
>  			 subflow->remote_key);

LGTM, thanks Davide!

Acked-by: Paolo Abeni <pabeni(a)redhat.com>

Re: [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[Paolo Abeni]

On Thu, 2020-03-19 at 22:45 +0100, Davide Caratti wrote:
> it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
> MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
> flags/version during the three-way handshake). Adjust MPTCP socket state
> earlier, to ensure correct functionality of mptcp_sendmsg() even in case
> of TCP fallback.
> 
> Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
> Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
> Signed-off-by: Davide Caratti <dcaratti@redhat.com>
> ---
>  net/mptcp/protocol.c | 4 ----
>  net/mptcp/subflow.c  | 6 ++++++
>  2 files changed, 6 insertions(+), 4 deletions(-)
> 
> diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> index e959104832ef..92d5382e71f4 100644
> --- a/net/mptcp/protocol.c
> +++ b/net/mptcp/protocol.c
> @@ -1055,10 +1055,6 @@ void mptcp_finish_connect(struct sock *ssk)
>  	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
>  	WRITE_ONCE(msk->ack_seq, ack_seq);
>  	WRITE_ONCE(msk->can_ack, 1);
> -	if (inet_sk_state_load(sk) != TCP_ESTABLISHED) {
> -		inet_sk_state_store(sk, TCP_ESTABLISHED);
> -		sk->sk_state_change(sk);
> -	}
>  }
>  
>  static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
> diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
> index 052d72a1d3a2..06b9075333c5 100644
> --- a/net/mptcp/subflow.c
> +++ b/net/mptcp/subflow.c
> @@ -109,9 +109,15 @@ static void subflow_v6_init_req(struct request_sock *req,
>  static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
>  {
>  	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
> +	struct sock *parent = subflow->conn;
>  
>  	subflow->icsk_af_ops->sk_rx_dst_set(sk, skb);
>  
> +	if (inet_sk_state_load(parent) != TCP_ESTABLISHED) {
> +		inet_sk_state_store(parent, TCP_ESTABLISHED);
> +		parent->sk_state_change(parent);
> +	}
> +
>  	if (!subflow->conn_finished) {
>  		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
>  			 subflow->remote_key);

LGTM, thanks Davide!

Acked-by: Paolo Abeni <pabeni@redhat.com>

[MPTCP] Re: [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[Matthieu Baerts]

[-- Attachment #1: Type: text/plain, Size: 842 bytes --]

On 19/03/2020 23:48, Paolo Abeni wrote:
> On Thu, 2020-03-19 at 22:45 +0100, Davide Caratti wrote:
>> it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
>> MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
>> flags/version during the three-way handshake). Adjust MPTCP socket state
>> earlier, to ensure correct functionality of mptcp_sendmsg() even in case
>> of TCP fallback.
>>
>> Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
>> Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
>> Signed-off-by: Davide Caratti <dcaratti(a)redhat.com>
> 
> LGTM, thanks Davide!
> 
> Acked-by: Paolo Abeni <pabeni(a)redhat.com>

LGTM too, thanks Davide and Paolo!

Reviewed-by: Matthieu Baerts <matthieu.baerts(a)tessares.net>

Re: [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[Matthieu Baerts]

On 19/03/2020 23:48, Paolo Abeni wrote:
> On Thu, 2020-03-19 at 22:45 +0100, Davide Caratti wrote:
>> it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
>> MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
>> flags/version during the three-way handshake). Adjust MPTCP socket state
>> earlier, to ensure correct functionality of mptcp_sendmsg() even in case
>> of TCP fallback.
>>
>> Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
>> Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
>> Signed-off-by: Davide Caratti <dcaratti@redhat.com>
> 
> LGTM, thanks Davide!
> 
> Acked-by: Paolo Abeni <pabeni@redhat.com>

LGTM too, thanks Davide and Paolo!

Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>

[MPTCP] Re: [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[David Miller]

[-- Attachment #1: Type: text/plain, Size: 650 bytes --]

From: Davide Caratti <dcaratti(a)redhat.com>
Date: Thu, 19 Mar 2020 22:45:37 +0100

> it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
> MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
> flags/version during the three-way handshake). Adjust MPTCP socket state
> earlier, to ensure correct functionality of mptcp_sendmsg() even in case
> of TCP fallback.
> 
> Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
> Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
> Signed-off-by: Davide Caratti <dcaratti(a)redhat.com>

APplied.

Re: [PATCH net-next] net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback

[David Miller]

From: Davide Caratti <dcaratti@redhat.com>
Date: Thu, 19 Mar 2020 22:45:37 +0100

> it's still possible for packetdrill to hang in mptcp_sendmsg(), when the
> MPTCP socket falls back to regular TCP (e.g. after receiving unsupported
> flags/version during the three-way handshake). Adjust MPTCP socket state
> earlier, to ensure correct functionality of mptcp_sendmsg() even in case
> of TCP fallback.
> 
> Fixes: 767d3ded5fb8 ("net: mptcp: don't hang before sending 'MP capable with data'")
> Fixes: 1954b86016cf ("mptcp: Check connection state before attempting send")
> Signed-off-by: Davide Caratti <dcaratti@redhat.com>

APplied.