From: Mat Martineau <mathew.j.martineau@linux.intel.com>
To: edumazet@google.com, netdev@vger.kernel.org
Cc: Peter Krystad <peter.krystad@linux.intel.com>,
cpaasch@apple.com, fw@strlen.de, pabeni@redhat.com,
dcaratti@redhat.com, matthieu.baerts@tessares.net
Subject: [RFC PATCH net-next 05/33] mptcp: Associate MPTCP context with TCP socket
Date: Mon, 17 Jun 2019 15:57:40 -0700 [thread overview]
Message-ID: <20190617225808.665-6-mathew.j.martineau@linux.intel.com> (raw)
In-Reply-To: <20190617225808.665-1-mathew.j.martineau@linux.intel.com>
From: Peter Krystad <peter.krystad@linux.intel.com>
Use ULP to associate a subflow_context structure with each TCP
subflow socket.
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
include/linux/tcp.h | 3 ++
net/mptcp/Makefile | 2 +-
net/mptcp/protocol.c | 96 +++++++++++++++++++++++++++++++++++++-------
net/mptcp/protocol.h | 24 +++++++++++
net/mptcp/subflow.c | 76 +++++++++++++++++++++++++++++++++++
5 files changed, 185 insertions(+), 16 deletions(-)
create mode 100644 net/mptcp/subflow.c
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 73c633f58233..b8c24bd8c862 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -397,6 +397,9 @@ struct tcp_sock {
u32 mtu_info; /* We received an ICMP_FRAG_NEEDED / ICMPV6_PKT_TOOBIG
* while socket was owned by user.
*/
+#if IS_ENABLED(CONFIG_MPTCP)
+ bool is_mptcp;
+#endif
#ifdef CONFIG_TCP_MD5SIG
/* TCP AF-Specific parts; only used by MD5 Signature support so far */
diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile
index 27a846263f08..e1ee5aade8b0 100644
--- a/net/mptcp/Makefile
+++ b/net/mptcp/Makefile
@@ -1,4 +1,4 @@
# SPDX-License-Identifier: GPL-2.0
obj-$(CONFIG_MPTCP) += mptcp.o
-mptcp-y := protocol.o options.o
+mptcp-y := protocol.o subflow.o options.o
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index e57ee600df7f..ce2374ea7871 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -20,7 +20,7 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
struct mptcp_sock *msk = mptcp_sk(sk);
struct socket *subflow = msk->subflow;
- pr_debug("subflow=%p", subflow->sk);
+ pr_debug("subflow=%p", subflow_ctx(subflow->sk));
return sock_sendmsg(subflow, msg);
}
@@ -31,7 +31,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
struct mptcp_sock *msk = mptcp_sk(sk);
struct socket *subflow = msk->subflow;
- pr_debug("subflow=%p", subflow->sk);
+ pr_debug("subflow=%p", subflow_ctx(subflow->sk));
return sock_recvmsg(subflow, msg, flags);
}
@@ -39,19 +39,10 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
static int mptcp_init_sock(struct sock *sk)
{
struct mptcp_sock *msk = mptcp_sk(sk);
- struct net *net = sock_net(sk);
- struct socket *sf;
- int err;
pr_debug("msk=%p", msk);
- err = sock_create_kern(net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sf);
- if (!err) {
- pr_debug("subflow=%p", sf->sk);
- msk->subflow = sf;
- }
-
- return err;
+ return 0;
}
static void mptcp_close(struct sock *sk, long timeout)
@@ -61,7 +52,7 @@ static void mptcp_close(struct sock *sk, long timeout)
inet_sk_state_store(sk, TCP_CLOSE);
if (msk->subflow) {
- pr_debug("subflow=%p", msk->subflow->sk);
+ pr_debug("subflow=%p", subflow_ctx(msk->subflow->sk));
sock_release(msk->subflow);
}
@@ -76,7 +67,7 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *saddr, int len)
saddr->sa_family = AF_INET;
- pr_debug("msk=%p, subflow=%p", msk, msk->subflow->sk);
+ pr_debug("msk=%p, subflow=%p", msk, subflow_ctx(msk->subflow->sk));
err = kernel_connect(msk->subflow, saddr, len, 0);
@@ -102,15 +93,90 @@ static struct proto mptcp_prot = {
.no_autobind = 1,
};
+static int mptcp_subflow_create(struct sock *sk)
+{
+ struct mptcp_sock *msk = mptcp_sk(sk);
+ struct net *net = sock_net(sk);
+ struct socket *sf;
+ int err;
+
+ pr_debug("msk=%p", msk);
+ err = sock_create_kern(net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sf);
+ if (!err) {
+ lock_sock(sf->sk);
+ err = tcp_set_ulp(sf->sk, "mptcp");
+ release_sock(sf->sk);
+ if (!err) {
+ struct subflow_context *subflow = subflow_ctx(sf->sk);
+
+ pr_debug("subflow=%p", subflow);
+ msk->subflow = sf;
+ subflow->conn = sk;
+ subflow->request_mptcp = 1; // @@ if MPTCP enabled
+ subflow->request_cksum = 1; // @@ if checksum enabled
+ subflow->version = 0;
+ }
+ }
+ return err;
+}
+
+static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
+{
+ struct mptcp_sock *msk = mptcp_sk(sock->sk);
+ int err = -ENOTSUPP;
+
+ pr_debug("msk=%p", msk);
+
+ if (uaddr->sa_family != AF_INET) // @@ allow only IPv4 for now
+ return err;
+
+ if (!msk->subflow) {
+ err = mptcp_subflow_create(sock->sk);
+ if (err)
+ return err;
+ }
+ return inet_bind(msk->subflow, uaddr, addr_len);
+}
+
+static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
+ int addr_len, int flags)
+{
+ struct mptcp_sock *msk = mptcp_sk(sock->sk);
+ int err = -ENOTSUPP;
+
+ pr_debug("msk=%p", msk);
+
+ if (uaddr->sa_family != AF_INET) // @@ allow only IPv4 for now
+ return err;
+
+ if (!msk->subflow) {
+ err = mptcp_subflow_create(sock->sk);
+ if (err)
+ return err;
+ }
+
+ return inet_stream_connect(msk->subflow, uaddr, addr_len, flags);
+}
+
+static struct proto_ops mptcp_stream_ops;
+
static struct inet_protosw mptcp_protosw = {
.type = SOCK_STREAM,
.protocol = IPPROTO_MPTCP,
.prot = &mptcp_prot,
- .ops = &inet_stream_ops,
+ .ops = &mptcp_stream_ops,
+ .flags = INET_PROTOSW_ICSK,
};
void __init mptcp_init(void)
{
+ mptcp_prot.h.hashinfo = tcp_prot.h.hashinfo;
+ mptcp_stream_ops = inet_stream_ops;
+ mptcp_stream_ops.bind = mptcp_bind;
+ mptcp_stream_ops.connect = mptcp_stream_connect;
+
+ subflow_init();
+
if (proto_register(&mptcp_prot, 1) != 0)
panic("Failed to register MPTCP proto.\n");
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index ac57e10ec4ca..b6adc2aa6222 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -41,4 +41,28 @@ static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
return (struct mptcp_sock *)sk;
}
+/* MPTCP subflow context */
+struct subflow_context {
+ u32 request_mptcp : 1, /* send MP_CAPABLE */
+ request_cksum : 1,
+ version : 4;
+ struct socket *tcp_sock; /* underlying tcp_sock */
+ struct sock *conn; /* parent mptcp_sock */
+};
+
+static inline struct subflow_context *subflow_ctx(const struct sock *sk)
+{
+ struct inet_connection_sock *icsk = inet_csk(sk);
+
+ return (struct subflow_context *)icsk->icsk_ulp_data;
+}
+
+static inline struct socket *
+mptcp_subflow_tcp_socket(const struct subflow_context *subflow)
+{
+ return subflow->tcp_sock;
+}
+
+void subflow_init(void);
+
#endif /* __MPTCP_PROTOCOL_H */
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
new file mode 100644
index 000000000000..8d13713ee159
--- /dev/null
+++ b/net/mptcp/subflow.c
@@ -0,0 +1,76 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP
+ *
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <net/sock.h>
+#include <net/inet_common.h>
+#include <net/inet_hashtables.h>
+#include <net/protocol.h>
+#include <net/tcp.h>
+#include <net/mptcp.h>
+#include "protocol.h"
+
+static struct subflow_context *subflow_create_ctx(struct sock *sk,
+ struct socket *sock)
+{
+ struct inet_connection_sock *icsk = inet_csk(sk);
+ struct subflow_context *ctx;
+
+ ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+ if (!ctx)
+ return NULL;
+
+ pr_debug("subflow=%p", ctx);
+
+ icsk->icsk_ulp_data = ctx;
+ /* might be NULL */
+ ctx->tcp_sock = sock;
+
+ return ctx;
+}
+
+static int subflow_ulp_init(struct sock *sk)
+{
+ struct tcp_sock *tsk = tcp_sk(sk);
+ struct subflow_context *ctx;
+ int err = 0;
+
+ ctx = subflow_create_ctx(sk, sk->sk_socket);
+ if (!ctx) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ pr_debug("subflow=%p", ctx);
+
+ tsk->is_mptcp = 1;
+out:
+ return err;
+}
+
+static void subflow_ulp_release(struct sock *sk)
+{
+ struct subflow_context *ctx = subflow_ctx(sk);
+
+ pr_debug("subflow=%p", ctx);
+
+ kfree(ctx);
+}
+
+static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = {
+ .name = "mptcp",
+ .owner = THIS_MODULE,
+ .init = subflow_ulp_init,
+ .release = subflow_ulp_release,
+};
+
+void subflow_init(void)
+{
+ if (tcp_register_ulp(&subflow_ulp_ops) != 0)
+ panic("MPTCP: failed to register subflows to ULP\n");
+}
--
2.22.0
next prev parent reply other threads:[~2019-06-17 22:59 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-17 22:57 [RFC PATCH net-next 00/33] Multipath TCP Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 01/33] tcp: Add MPTCP option number Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 02/33] tcp: Define IPPROTO_MPTCP Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 03/33] mptcp: Add MPTCP socket stubs Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 04/33] mptcp: Handle MPTCP TCP options Mat Martineau
2019-06-17 22:57 ` Mat Martineau [this message]
2019-06-17 22:57 ` [RFC PATCH net-next 06/33] tcp: Expose tcp struct and routine for MPTCP Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 07/33] mptcp: Handle MP_CAPABLE options for outgoing connections Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 08/33] mptcp: add mptcp_poll Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 09/33] tcp, ulp: Add clone operation to tcp_ulp_ops Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 10/33] mptcp: Create SUBFLOW socket for incoming connections Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 11/33] mptcp: Add key generation and token tree Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 12/33] mptcp: Add shutdown() socket operation Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 13/33] mptcp: Add setsockopt()/getsockopt() socket operations Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 14/33] tcp: clean ext on tx recycle Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 15/33] mptcp: Add MPTCP to skb extensions Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 16/33] tcp: Prevent coalesce/collapse when skb has MPTCP extensions Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 17/33] tcp: Export low-level TCP functions Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 18/33] mptcp: Write MPTCP DSS headers to outgoing data packets Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 19/33] mptcp: Implement MPTCP receive path Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 20/33] mptcp: Make connection_list a real list of subflows Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 21/33] mptcp: add and use mptcp_subflow_hold Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 22/33] mptcp: add basic kselftest program Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 23/33] mptcp: selftests: switch to netns+veth based tests Mat Martineau
2019-06-17 22:57 ` [RFC PATCH net-next 24/33] mptcp: selftests: Add capture option Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 25/33] mptcp: use sk_page_frag() in sendmsg Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 26/33] mptcp: sendmsg() do spool all the provided data Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 27/33] mptcp: allow collapsing consecutive sendpages on the same substream Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 28/33] tcp: Check for filled TCP option space before SACK Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 29/33] mptcp: accept: don't leak mptcp socket structure Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 30/33] mptcp: switch sublist to mptcp socket lock protection Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 31/33] mptcp: Add path manager interface Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 32/33] mptcp: Add ADD_ADDR handling Mat Martineau
2019-06-17 22:58 ` [RFC PATCH net-next 33/33] mptcp: Add handling of incoming MP_JOIN requests Mat Martineau
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190617225808.665-6-mathew.j.martineau@linux.intel.com \
--to=mathew.j.martineau@linux.intel.com \
--cc=cpaasch@apple.com \
--cc=dcaratti@redhat.com \
--cc=edumazet@google.com \
--cc=fw@strlen.de \
--cc=matthieu.baerts@tessares.net \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=peter.krystad@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox