net: support marking accepting TCP sockets

When using mark-based routing, sockets returned from accept()
may need to be marked differently depending on the incoming
connection request.

This is the case, for example, if different socket marks identify
different networks: a listening socket may want to accept
connections from all networks, but each connection should be
marked with the network that the request came in on, so that
subsequent packets are sent on the correct network.

This patch adds a sysctl to mark TCP sockets based on the fwmark
of the incoming SYN packet. If enabled, and an unmarked socket
receives a SYN, then the SYN packet's fwmark is written to the
connection's inet_request_sock, and later written back to the
accepted socket when the connection is established.  If the
socket already has a nonzero mark, then the behaviour is the same
as it is today, i.e., the listening socket's fwmark is used.

Black-box tested using user-mode linux:

- IPv4/IPv6 SYN+ACK, FIN, etc. packets are routed based on the
  mark of the incoming SYN packet.
- The socket returned by accept() is marked with the mark of the
  incoming SYN packet.
- Tested with syncookies=1 and syncookies=2.

Change-Id: I26bc1eceefd2c588d73b921865ab70e4645ade57
Signed-off-by: Lorenzo Colitti <lorenzo@google.com>

10 files changed, 39 insertions, 5 deletions

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index ecf5abd57c0..6e5c7c7333b 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -477,6 +477,16 @@ tcp_fastopen - INTEGER
 
 	See include/net/tcp.h and the code for more details.
 
+tcp_fwmark_accept - BOOLEAN
+	If set, incoming connections to listening sockets that do not have a
+	socket mark will set the mark of the accepting socket to the fwmark of
+	the incoming SYN packet. This will cause all packets on that connection
+	(starting from the first SYNACK) to be sent with that fwmark. The
+	listening socket's mark is unchanged. Listening sockets that already
+	have a fwmark set via setsockopt(SOL_SOCKET, SO_MARK, ...) are
+	unaffected.
+	Default: 0
+
 tcp_syn_retries - INTEGER
 	Number of times initial SYNs for an active TCP connection attempt
 	will be retransmitted. Should not be higher than 255. Default value
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 7235ae73a1e..9528e10fa0b 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -88,6 +88,7 @@ struct inet_request_sock {
 				acked	   : 1,
 				no_srccheck: 1;
 	kmemcheck_bitfield_end(flags);
+	u32                     ir_mark;
 	struct ip_options_rcu	*opt;
 };
 
@@ -96,6 +97,14 @@ static inline struct inet_request_sock *inet_rsk(const struct request_sock *sk)
 	return (struct inet_request_sock *)sk;
 }
 
+static inline u32 inet_request_mark(struct sock *sk, struct sk_buff *skb)
+{
+	if (!sk->sk_mark && sock_net(sk)->ipv4.sysctl_tcp_fwmark_accept)
+		return skb->mark;
+
+	return sk->sk_mark;
+}
+
 struct inet_cork {
 	unsigned int		flags;
 	__be32			addr;
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 222461a7cc5..0dd6f0b3ead 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -65,6 +65,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_ecn;
 
 	int sysctl_fwmark_reflect;
+	int sysctl_tcp_fwmark_accept;
 
 	kgid_t sysctl_ping_group_range[2];
 	long sysctl_tcp_mem[3];
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 6acb541c909..442087d371f 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -417,7 +417,7 @@ struct dst_entry *inet_csk_route_req(struct sock *sk,
 	struct net *net = sock_net(sk);
 	int flags = inet_sk_flowi_flags(sk);
 
-	flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark,
+	flowi4_init_output(fl4, sk->sk_bound_dev_if, ireq->ir_mark,
 			   RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE,
 			   sk->sk_protocol,
 			   flags,
@@ -454,7 +454,7 @@ struct dst_entry *inet_csk_route_child_sock(struct sock *sk,
 
 	rcu_read_lock();
 	opt = rcu_dereference(newinet->inet_opt);
-	flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark,
+	flowi4_init_output(fl4, sk->sk_bound_dev_if, inet_rsk(req)->ir_mark,
 			   RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE,
 			   sk->sk_protocol, inet_sk_flowi_flags(sk),
 			   (opt && opt->opt.srr) ? opt->opt.faddr : ireq->rmt_addr,
@@ -688,6 +688,8 @@ struct sock *inet_csk_clone_lock(const struct sock *sk,
 		inet_sk(newsk)->inet_sport = inet_rsk(req)->loc_port;
 		newsk->sk_write_space = sk_stream_write_space;
 
+		newsk->sk_mark = inet_rsk(req)->ir_mark;
+
 		newicsk->icsk_retransmits = 0;
 		newicsk->icsk_backoff	  = 0;
 		newicsk->icsk_probes_out  = 0;
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index b05c96e7af8..5abb45e281b 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -312,6 +312,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	ireq->rmt_port		= th->source;
 	ireq->loc_addr		= ip_hdr(skb)->daddr;
 	ireq->rmt_addr		= ip_hdr(skb)->saddr;
+	ireq->ir_mark		= inet_request_mark(sk, skb);
 	ireq->ecn_ok		= ecn_ok;
 	ireq->snd_wscale	= tcp_opt.snd_wscale;
 	ireq->sack_ok		= tcp_opt.sack_ok;
@@ -348,7 +349,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	 * hasn't changed since we received the original syn, but I see
 	 * no easy way to do this.
 	 */
-	flowi4_init_output(&fl4, sk->sk_bound_dev_if, sk->sk_mark,
+	flowi4_init_output(&fl4, sk->sk_bound_dev_if, ireq->ir_mark,
 			   RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, IPPROTO_TCP,
 			   inet_sk_flowi_flags(sk),
 			   (opt && opt->srr) ? opt->faddr : ireq->rmt_addr,
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index e6cdcb32b33..cc5fa7da12e 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -866,6 +866,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
+	{
+		.procname	= "tcp_fwmark_accept",
+		.data		= &init_net.ipv4.sysctl_tcp_fwmark_accept,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 7999fc55c83..40ec14507f7 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1527,6 +1527,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	ireq->rmt_addr = saddr;
 	ireq->no_srccheck = inet_sk(sk)->transparent;
 	ireq->opt = tcp_v4_save_options(skb);
+	ireq->ir_mark = inet_request_mark(sk, skb);
 
 	if (security_inet_conn_request(sk, skb, req))
 		goto drop_and_free;
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index e4311cbc8b4..f1493138d21 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -81,7 +81,7 @@ struct dst_entry *inet6_csk_route_req(struct sock *sk,
 	final_p = fl6_update_dst(fl6, np->opt, &final);
 	fl6->saddr = treq->loc_addr;
 	fl6->flowi6_oif = treq->iif;
-	fl6->flowi6_mark = sk->sk_mark;
+	fl6->flowi6_mark = inet_rsk(req)->ir_mark;
 	fl6->fl6_dport = inet_rsk(req)->rmt_port;
 	fl6->fl6_sport = inet_rsk(req)->loc_port;
 	security_req_classify_flow(req, flowi6_to_flowi(fl6));
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index d5dda20bd71..1efbc6f44a6 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -212,6 +212,8 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	    ipv6_addr_type(&ireq6->rmt_addr) & IPV6_ADDR_LINKLOCAL)
 		ireq6->iif = inet6_iif(skb);
 
+	ireq->ir_mark = inet_request_mark(sk, skb);
+
 	req->expires = 0UL;
 	req->num_retrans = 0;
 	ireq->ecn_ok		= ecn_ok;
@@ -238,7 +240,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 		final_p = fl6_update_dst(&fl6, np->opt, &final);
 		fl6.saddr = ireq6->loc_addr;
 		fl6.flowi6_oif = sk->sk_bound_dev_if;
-		fl6.flowi6_mark = sk->sk_mark;
+		fl6.flowi6_mark = ireq->ir_mark;
 		fl6.fl6_dport = inet_rsk(req)->rmt_port;
 		fl6.fl6_sport = inet_sk(sk)->inet_sport;
 		security_req_classify_flow(req, flowi6_to_flowi(&fl6));
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 71545cb17ab..6e882dadb4f 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1000,6 +1000,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 		TCP_ECN_create_request(req, skb, sock_net(sk));
 
 	treq->iif = sk->sk_bound_dev_if;
+	inet_rsk(req)->ir_mark = inet_request_mark(sk, skb);
 
 	/* So that link locals have meaning */
 	if (!sk->sk_bound_dev_if &&