xfrm: add IPv6 support for espintcp

This extends espintcp to support IPv6, building on the existing code
and the new UDPv6 encapsulation support. Most of the code is either
reused directly (stream parser, ULP) or very similar to the IPv4
variant (net/ipv6/esp6.c changes).

The separation of config options for IPv4 and IPv6 espintcp requires a
bit of Kconfig gymnastics to enable the core code.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>

3 files changed, 200 insertions, 1 deletions

diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 2ccaee98fddb..468a2faadc7d 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -88,6 +88,18 @@ config INET6_ESP_OFFLOAD
 
 	  If unsure, say N.
 
+config INET6_ESPINTCP
+	bool "IPv6: ESP in TCP encapsulation (RFC 8229)"
+	depends on XFRM && INET6_ESP
+	select STREAM_PARSER
+	select NET_SOCK_MSG
+	select XFRM_ESPINTCP
+	help
+	  Support for RFC 8229 encapsulation of ESP and IKE over
+	  TCP/IPv6 sockets.
+
+	  If unsure, say N.
+
 config INET6_IPCOMP
 	tristate "IPv6: IPComp transformation"
 	select INET6_XFRM_TUNNEL
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index b0b99c08350a..cbbb00bad20e 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -964,6 +964,7 @@ static const struct ipv6_stub ipv6_stub_impl = {
 	.ndisc_send_na = ndisc_send_na,
 #if IS_ENABLED(CONFIG_XFRM)
 	.xfrm6_udp_encap_rcv = xfrm6_udp_encap_rcv,
+	.xfrm6_rcv_encap = xfrm6_rcv_encap,
 #endif
 	.nd_tbl	= &nd_tbl,
 };
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index e8800968e209..c43592771126 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -33,6 +33,9 @@
 #include <net/protocol.h>
 #include <net/udp.h>
 #include <linux/icmpv6.h>
+#include <net/tcp.h>
+#include <net/espintcp.h>
+#include <net/inet6_hashtables.h>
 
 #include <linux/highmem.h>
 
@@ -132,6 +135,132 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp)
 			put_page(sg_page(sg));
 }
 
+#ifdef CONFIG_INET6_ESPINTCP
+struct esp_tcp_sk {
+	struct sock *sk;
+	struct rcu_head rcu;
+};
+
+static void esp_free_tcp_sk(struct rcu_head *head)
+{
+	struct esp_tcp_sk *esk = container_of(head, struct esp_tcp_sk, rcu);
+
+	sock_put(esk->sk);
+	kfree(esk);
+}
+
+static struct sock *esp6_find_tcp_sk(struct xfrm_state *x)
+{
+	struct xfrm_encap_tmpl *encap = x->encap;
+	struct esp_tcp_sk *esk;
+	__be16 sport, dport;
+	struct sock *nsk;
+	struct sock *sk;
+
+	sk = rcu_dereference(x->encap_sk);
+	if (sk && sk->sk_state == TCP_ESTABLISHED)
+		return sk;
+
+	spin_lock_bh(&x->lock);
+	sport = encap->encap_sport;
+	dport = encap->encap_dport;
+	nsk = rcu_dereference_protected(x->encap_sk,
+					lockdep_is_held(&x->lock));
+	if (sk && sk == nsk) {
+		esk = kmalloc(sizeof(*esk), GFP_ATOMIC);
+		if (!esk) {
+			spin_unlock_bh(&x->lock);
+			return ERR_PTR(-ENOMEM);
+		}
+		RCU_INIT_POINTER(x->encap_sk, NULL);
+		esk->sk = sk;
+		call_rcu(&esk->rcu, esp_free_tcp_sk);
+	}
+	spin_unlock_bh(&x->lock);
+
+	sk = __inet6_lookup_established(xs_net(x), &tcp_hashinfo, &x->id.daddr.in6,
+					dport, &x->props.saddr.in6, ntohs(sport), 0, 0);
+	if (!sk)
+		return ERR_PTR(-ENOENT);
+
+	if (!tcp_is_ulp_esp(sk)) {
+		sock_put(sk);
+		return ERR_PTR(-EINVAL);
+	}
+
+	spin_lock_bh(&x->lock);
+	nsk = rcu_dereference_protected(x->encap_sk,
+					lockdep_is_held(&x->lock));
+	if (encap->encap_sport != sport ||
+	    encap->encap_dport != dport) {
+		sock_put(sk);
+		sk = nsk ?: ERR_PTR(-EREMCHG);
+	} else if (sk == nsk) {
+		sock_put(sk);
+	} else {
+		rcu_assign_pointer(x->encap_sk, sk);
+	}
+	spin_unlock_bh(&x->lock);
+
+	return sk;
+}
+
+static int esp_output_tcp_finish(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct sock *sk;
+	int err;
+
+	rcu_read_lock();
+
+	sk = esp6_find_tcp_sk(x);
+	err = PTR_ERR_OR_ZERO(sk);
+	if (err)
+		goto out;
+
+	bh_lock_sock(sk);
+	if (sock_owned_by_user(sk))
+		err = espintcp_queue_out(sk, skb);
+	else
+		err = espintcp_push_skb(sk, skb);
+	bh_unlock_sock(sk);
+
+out:
+	rcu_read_unlock();
+	return err;
+}
+
+static int esp_output_tcp_encap_cb(struct net *net, struct sock *sk,
+				   struct sk_buff *skb)
+{
+	struct dst_entry *dst = skb_dst(skb);
+	struct xfrm_state *x = dst->xfrm;
+
+	return esp_output_tcp_finish(x, skb);
+}
+
+static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	local_bh_disable();
+	err = xfrm_trans_queue_net(xs_net(x), skb, esp_output_tcp_encap_cb);
+	local_bh_enable();
+
+	/* EINPROGRESS just happens to do the right thing.  It
+	 * actually means that the skb has been consumed and
+	 * isn't coming back.
+	 */
+	return err ?: -EINPROGRESS;
+}
+#else
+static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
+{
+	kfree_skb(skb);
+
+	return -EOPNOTSUPP;
+}
+#endif
+
 static void esp_output_encap_csum(struct sk_buff *skb)
 {
 	/* UDP encap with IPv6 requires a valid checksum */
@@ -181,7 +310,11 @@ static void esp_output_done(struct crypto_async_request *base, int err)
 		secpath_reset(skb);
 		xfrm_dev_resume(skb);
 	} else {
-		xfrm_output_resume(skb, err);
+		if (!err &&
+		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
+			esp_output_tail_tcp(x, skb);
+		else
+			xfrm_output_resume(skb, err);
 	}
 }
 
@@ -274,6 +407,41 @@ static struct ip_esp_hdr *esp6_output_udp_encap(struct sk_buff *skb,
 	return (struct ip_esp_hdr *)(uh + 1);
 }
 
+#ifdef CONFIG_INET6_ESPINTCP
+static struct ip_esp_hdr *esp6_output_tcp_encap(struct xfrm_state *x,
+						struct sk_buff *skb,
+						struct esp_info *esp)
+{
+	__be16 *lenp = (void *)esp->esph;
+	struct ip_esp_hdr *esph;
+	unsigned int len;
+	struct sock *sk;
+
+	len = skb->len + esp->tailen - skb_transport_offset(skb);
+	if (len > IP_MAX_MTU)
+		return ERR_PTR(-EMSGSIZE);
+
+	rcu_read_lock();
+	sk = esp6_find_tcp_sk(x);
+	rcu_read_unlock();
+
+	if (IS_ERR(sk))
+		return ERR_CAST(sk);
+
+	*lenp = htons(len);
+	esph = (struct ip_esp_hdr *)(lenp + 1);
+
+	return esph;
+}
+#else
+static struct ip_esp_hdr *esp6_output_tcp_encap(struct xfrm_state *x,
+						struct sk_buff *skb,
+						struct esp_info *esp)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+#endif
+
 static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb,
 			    struct esp_info *esp)
 {
@@ -294,6 +462,9 @@ static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb,
 	case UDP_ENCAP_ESPINUDP_NON_IKE:
 		esph = esp6_output_udp_encap(skb, encap_type, esp, sport, dport);
 		break;
+	case TCP_ENCAP_ESPINTCP:
+		esph = esp6_output_tcp_encap(x, skb, esp);
+		break;
 	}
 
 	if (IS_ERR(esph))
@@ -509,6 +680,9 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info
 	if (sg != dsg)
 		esp_ssg_unref(x, tmp);
 
+	if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
+		err = esp_output_tail_tcp(x, skb);
+
 error_free:
 	kfree(tmp);
 error:
@@ -633,9 +807,13 @@ int esp6_input_done2(struct sk_buff *skb, int err)
 		const struct ipv6hdr *ip6h = ipv6_hdr(skb);
 		struct xfrm_encap_tmpl *encap = x->encap;
 		struct udphdr *uh = (void *)(skb_network_header(skb) + hdr_len);
+		struct tcphdr *th = (void *)(skb_network_header(skb) + hdr_len);
 		__be16 source;
 
 		switch (x->encap->encap_type) {
+		case TCP_ENCAP_ESPINTCP:
+			source = th->source;
+			break;
 		case UDP_ENCAP_ESPINUDP:
 		case UDP_ENCAP_ESPINUDP_NON_IKE:
 			source = uh->source;
@@ -1039,6 +1217,14 @@ static int esp6_init_state(struct xfrm_state *x)
 		case UDP_ENCAP_ESPINUDP_NON_IKE:
 			x->props.header_len += sizeof(struct udphdr) + 2 * sizeof(u32);
 			break;
+#ifdef CONFIG_INET6_ESPINTCP
+		case TCP_ENCAP_ESPINTCP:
+			/* only the length field, TCP encap is done by
+			 * the socket
+			 */
+			x->props.header_len += 2;
+			break;
+#endif
 		}
 	}