xfrm: add support for UDPv6 encapsulation of ESP

This patch adds support for encapsulation of ESP over UDPv6. The code
is very similar to the IPv4 encapsulation implementation, and allows
to easily add espintcp on IPv6 as a follow-up.
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>

include/net/ipv6_stubs.h

@@ -56,6 +56,9 @@ struct ipv6_stub {
 	void (*ndisc_send_na)(struct net_device *dev, const struct in6_addr *daddr,
 			      const struct in6_addr *solicited_addr,
 			      bool router, bool solicited, bool override, bool inc_opt);
+#if IS_ENABLED(CONFIG_XFRM)
+	int (*xfrm6_udp_encap_rcv)(struct sock *sk, struct sk_buff *skb);
+#endif
 	struct neigh_table *nd_tbl;
 };
 extern const struct ipv6_stub *ipv6_stub __read_mostly;

include/net/xfrm.h

@@ -1406,6 +1406,8 @@ struct xfrm4_protocol {
 
 struct xfrm6_protocol {
 	int (*handler)(struct sk_buff *skb);
+	int (*input_handler)(struct sk_buff *skb, int nexthdr, __be32 spi,
+			     int encap_type);
 	int (*cb_handler)(struct sk_buff *skb, int err);
 	int (*err_handler)(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			   u8 type, u8 code, int offset, __be32 info);
@@ -1590,6 +1592,8 @@ int xfrm6_extract_header(struct sk_buff *skb);
 int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi,
 		  struct ip6_tnl *t);
+int xfrm6_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
+		    int encap_type);
 int xfrm6_transport_finish(struct sk_buff *skb, int async);
 int xfrm6_rcv_tnl(struct sk_buff *skb, struct ip6_tnl *t);
 int xfrm6_rcv(struct sk_buff *skb);
@@ -1610,6 +1614,7 @@ int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb,
 
 #ifdef CONFIG_XFRM
 int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb);
+int xfrm6_udp_encap_rcv(struct sock *sk, struct sk_buff *skb);
 int xfrm_user_policy(struct sock *sk, int optname,
 		     u8 __user *optval, int optlen);
 #else

net/ipv4/udp.c

@@ -112,6 +112,9 @@
 #include <net/sock_reuseport.h>
 #include <net/addrconf.h>
 #include <net/udp_tunnel.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ipv6_stubs.h>
+#endif
 
 struct udp_table udp_table __read_mostly;
 EXPORT_SYMBOL(udp_table);
@@ -2563,6 +2566,11 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 #ifdef CONFIG_XFRM
 		case UDP_ENCAP_ESPINUDP:
 		case UDP_ENCAP_ESPINUDP_NON_IKE:
+#if IS_ENABLED(CONFIG_IPV6)
+			if (sk->sk_family == AF_INET6)
+				up->encap_rcv = ipv6_stub->xfrm6_udp_encap_rcv;
+			else
+#endif
 				up->encap_rcv = xfrm4_udp_encap_rcv;
 #endif
 			fallthrough;

net/ipv6/af_inet6.c

@@ -60,6 +60,7 @@
 #include <net/calipso.h>
 #include <net/seg6.h>
 #include <net/rpl.h>
+#include <net/xfrm.h>
 
 #include <linux/uaccess.h>
 #include <linux/mroute6.h>
@@ -961,6 +962,9 @@ static const struct ipv6_stub ipv6_stub_impl = {
 	.ip6_del_rt	   = ip6_del_rt,
 	.udpv6_encap_enable = udpv6_encap_enable,
 	.ndisc_send_na = ndisc_send_na,
+#if IS_ENABLED(CONFIG_XFRM)
+	.xfrm6_udp_encap_rcv = xfrm6_udp_encap_rcv,
+#endif
 	.nd_tbl	= &nd_tbl,
 };
 

net/ipv6/ah6.c

@@ -767,6 +767,7 @@ static const struct xfrm_type ah6_type = {
 
 static struct xfrm6_protocol ah6_protocol = {
 	.handler	=	xfrm6_rcv,
+	.input_handler	=	xfrm_input,
 	.cb_handler	=	ah6_rcv_cb,
 	.err_handler	=	ah6_err,
 	.priority	=	0,

net/ipv6/esp6_offload.c

@@ -271,7 +271,6 @@ static int esp6_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features
 	int alen;
 	int blksize;
 	struct xfrm_offload *xo;
-	struct ip_esp_hdr *esph;
 	struct crypto_aead *aead;
 	struct esp_info esp;
 	bool hw_offload = true;
@@ -312,13 +311,13 @@ static int esp6_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features
 
 	seq = xo->seq.low;
 
-	esph = ip_esp_hdr(skb);
-	esph->spi = x->id.spi;
+	esp.esph = ip_esp_hdr(skb);
+	esp.esph->spi = x->id.spi;
 
 	skb_push(skb, -skb_network_offset(skb));
 
 	if (xo->flags & XFRM_GSO_SEGMENT) {
-		esph->seq_no = htonl(seq);
+		esp.esph->seq_no = htonl(seq);
 
 		if (!skb_is_gso(skb))
 			xo->seq.low++;

net/ipv6/ip6_vti.c

@@ -296,7 +296,8 @@ static void vti6_dev_uninit(struct net_device *dev)
 	dev_put(dev);
 }
 
-static int vti6_rcv(struct sk_buff *skb)
+static int vti6_input_proto(struct sk_buff *skb, int nexthdr, __be32 spi,
+			    int encap_type)
 {
 	struct ip6_tnl *t;
 	const struct ipv6hdr *ipv6h = ipv6_hdr(skb);
@@ -323,7 +324,10 @@ static int vti6_rcv(struct sk_buff *skb)
 
 		rcu_read_unlock();
 
-		return xfrm6_rcv_tnl(skb, t);
+		XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6 = t;
+		XFRM_SPI_SKB_CB(skb)->family = AF_INET6;
+		XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr);
+		return xfrm_input(skb, nexthdr, spi, encap_type);
 	}
 	rcu_read_unlock();
 	return -EINVAL;
@@ -332,6 +336,13 @@ static int vti6_rcv(struct sk_buff *skb)
 	return 0;
 }
 
+static int vti6_rcv(struct sk_buff *skb)
+{
+	int nexthdr = skb_network_header(skb)[IP6CB(skb)->nhoff];
+
+	return vti6_input_proto(skb, nexthdr, 0, 0);
+}
+
 static int vti6_rcv_cb(struct sk_buff *skb, int err)
 {
 	unsigned short family;
@@ -1185,6 +1196,7 @@ static struct pernet_operations vti6_net_ops = {
 
 static struct xfrm6_protocol vti_esp6_protocol __read_mostly = {
 	.handler	=	vti6_rcv,
+	.input_handler	=	vti6_input_proto,
 	.cb_handler	=	vti6_rcv_cb,
 	.err_handler	=	vti6_err,
 	.priority	=	100,
@@ -1192,6 +1204,7 @@ static struct xfrm6_protocol vti_esp6_protocol __read_mostly = {
 
 static struct xfrm6_protocol vti_ah6_protocol __read_mostly = {
 	.handler	=	vti6_rcv,
+	.input_handler	=	vti6_input_proto,
 	.cb_handler	=	vti6_rcv_cb,
 	.err_handler	=	vti6_err,
 	.priority	=	100,
@@ -1199,6 +1212,7 @@ static struct xfrm6_protocol vti_ah6_protocol __read_mostly = {
 
 static struct xfrm6_protocol vti_ipcomp6_protocol __read_mostly = {
 	.handler	=	vti6_rcv,
+	.input_handler	=	vti6_input_proto,
 	.cb_handler	=	vti6_rcv_cb,
 	.err_handler	=	vti6_err,
 	.priority	=	100,

net/ipv6/ipcomp6.c

@@ -183,6 +183,7 @@ static const struct xfrm_type ipcomp6_type = {
 
 static struct xfrm6_protocol ipcomp6_protocol = {
 	.handler	= xfrm6_rcv,
+	.input_handler	= xfrm_input,
 	.cb_handler	= ipcomp6_rcv_cb,
 	.err_handler	= ipcomp6_err,
 	.priority	= 0,

net/ipv6/xfrm6_input.c

@@ -35,9 +35,12 @@ EXPORT_SYMBOL(xfrm6_rcv_spi);
 static int xfrm6_transport_finish2(struct net *net, struct sock *sk,
 				   struct sk_buff *skb)
 {
-	if (xfrm_trans_queue(skb, ip6_rcv_finish))
-		__kfree_skb(skb);
-	return -1;
+	if (xfrm_trans_queue(skb, ip6_rcv_finish)) {
+		kfree_skb(skb);
+		return NET_RX_DROP;
+	}
+
+	return 0;
 }
 
 int xfrm6_transport_finish(struct sk_buff *skb, int async)
@@ -60,13 +63,106 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async)
 	if (xo && (xo->flags & XFRM_GRO)) {
 		skb_mac_header_rebuild(skb);
 		skb_reset_transport_header(skb);
-		return -1;
+		return 0;
 	}
 
 	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING,
 		dev_net(skb->dev), NULL, skb, skb->dev, NULL,
 		xfrm6_transport_finish2);
-	return -1;
+	return 0;
+}
+
+/* If it's a keepalive packet, then just eat it.
+ * If it's an encapsulated packet, then pass it to the
+ * IPsec xfrm input.
+ * Returns 0 if skb passed to xfrm or was dropped.
+ * Returns >0 if skb should be passed to UDP.
+ * Returns <0 if skb should be resubmitted (-ret is protocol)
+ */
+int xfrm6_udp_encap_rcv(struct sock *sk, struct sk_buff *skb)
+{
+	struct udp_sock *up = udp_sk(sk);
+	struct udphdr *uh;
+	struct ipv6hdr *ip6h;
+	int len;
+	int ip6hlen = sizeof(struct ipv6hdr);
+
+	__u8 *udpdata;
+	__be32 *udpdata32;
+	__u16 encap_type = up->encap_type;
+
+	/* if this is not encapsulated socket, then just return now */
+	if (!encap_type)
+		return 1;
+
+	/* If this is a paged skb, make sure we pull up
+	 * whatever data we need to look at. */
+	len = skb->len - sizeof(struct udphdr);
+	if (!pskb_may_pull(skb, sizeof(struct udphdr) + min(len, 8)))
+		return 1;
+
+	/* Now we can get the pointers */
+	uh = udp_hdr(skb);
+	udpdata = (__u8 *)uh + sizeof(struct udphdr);
+	udpdata32 = (__be32 *)udpdata;
+
+	switch (encap_type) {
+	default:
+	case UDP_ENCAP_ESPINUDP:
+		/* Check if this is a keepalive packet.  If so, eat it. */
+		if (len == 1 && udpdata[0] == 0xff) {
+			goto drop;
+		} else if (len > sizeof(struct ip_esp_hdr) && udpdata32[0] != 0) {
+			/* ESP Packet without Non-ESP header */
+			len = sizeof(struct udphdr);
+		} else
+			/* Must be an IKE packet.. pass it through */
+			return 1;
+		break;
+	case UDP_ENCAP_ESPINUDP_NON_IKE:
+		/* Check if this is a keepalive packet.  If so, eat it. */
+		if (len == 1 && udpdata[0] == 0xff) {
+			goto drop;
+		} else if (len > 2 * sizeof(u32) + sizeof(struct ip_esp_hdr) &&
+			   udpdata32[0] == 0 && udpdata32[1] == 0) {
+
+			/* ESP Packet with Non-IKE marker */
+			len = sizeof(struct udphdr) + 2 * sizeof(u32);
+		} else
+			/* Must be an IKE packet.. pass it through */
+			return 1;
+		break;
+	}
+
+	/* At this point we are sure that this is an ESPinUDP packet,
+	 * so we need to remove 'len' bytes from the packet (the UDP
+	 * header and optional ESP marker bytes) and then modify the
+	 * protocol to ESP, and then call into the transform receiver.
+	 */
+	if (skb_unclone(skb, GFP_ATOMIC))
+		goto drop;
+
+	/* Now we can update and verify the packet length... */
+	ip6h = ipv6_hdr(skb);
+	ip6h->payload_len = htons(ntohs(ip6h->payload_len) - len);
+	if (skb->len < ip6hlen + len) {
+		/* packet is too small!?! */
+		goto drop;
+	}
+
+	/* pull the data buffer up to the ESP header and set the
+	 * transport header to point to ESP.  Keep UDP on the stack
+	 * for later.
+	 */
+	__skb_pull(skb, len);
+	skb_reset_transport_header(skb);
+
+	/* process ESP */
+	return xfrm6_rcv_encap(skb, IPPROTO_ESP, 0, encap_type);
+
+drop:
+	kfree_skb(skb);
+	return 0;
 }
 
 int xfrm6_rcv_tnl(struct sk_buff *skb, struct ip6_tnl *t)

net/ipv6/xfrm6_protocol.c

@@ -14,6 +14,7 @@
 #include <linux/mutex.h>
 #include <linux/skbuff.h>
 #include <linux/icmpv6.h>
+#include <net/ip6_route.h>
 #include <net/ipv6.h>
 #include <net/protocol.h>
 #include <net/xfrm.h>
@@ -58,6 +59,53 @@ static int xfrm6_rcv_cb(struct sk_buff *skb, u8 protocol, int err)
 	return 0;
 }
 
+int xfrm6_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
+		    int encap_type)
+{
+	int ret;
+	struct xfrm6_protocol *handler;
+	struct xfrm6_protocol __rcu **head = proto_handlers(nexthdr);
+
+	XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6 = NULL;
+	XFRM_SPI_SKB_CB(skb)->family = AF_INET6;
+	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr);
+
+	if (!head)
+		goto out;
+
+	if (!skb_dst(skb)) {
+		const struct ipv6hdr *ip6h = ipv6_hdr(skb);
+		int flags = RT6_LOOKUP_F_HAS_SADDR;
+		struct dst_entry *dst;
+		struct flowi6 fl6 = {
+			.flowi6_iif   = skb->dev->ifindex,
+			.daddr        = ip6h->daddr,
+			.saddr        = ip6h->saddr,
+			.flowlabel    = ip6_flowinfo(ip6h),
+			.flowi6_mark  = skb->mark,
+			.flowi6_proto = ip6h->nexthdr,
+		};
+
+		dst = ip6_route_input_lookup(dev_net(skb->dev), skb->dev, &fl6,
+					     skb, flags);
+		if (dst->error)
+			goto drop;
+		skb_dst_set(skb, dst);
+	}
+
+	for_each_protocol_rcu(*head, handler)
+		if ((ret = handler->input_handler(skb, nexthdr, spi, encap_type)) != -EINVAL)
+			return ret;
+
+out:
+	icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);
+
+drop:
+	kfree_skb(skb);
+	return 0;
+}
+EXPORT_SYMBOL(xfrm6_rcv_encap);
+
 static int xfrm6_esp_rcv(struct sk_buff *skb)
 {
 	int ret;

net/xfrm/xfrm_interface.c

@@ -755,6 +755,7 @@ static struct pernet_operations xfrmi_net_ops = {
 
 static struct xfrm6_protocol xfrmi_esp6_protocol __read_mostly = {
 	.handler	=	xfrm6_rcv,
+	.input_handler	=	xfrm_input,
 	.cb_handler	=	xfrmi_rcv_cb,
 	.err_handler	=	xfrmi6_err,
 	.priority	=	10,
@@ -762,6 +763,7 @@ static struct xfrm6_protocol xfrmi_esp6_protocol __read_mostly = {
 
 static struct xfrm6_protocol xfrmi_ah6_protocol __read_mostly = {
 	.handler	=	xfrm6_rcv,
+	.input_handler	=	xfrm_input,
 	.cb_handler	=	xfrmi_rcv_cb,
 	.err_handler	=	xfrmi6_err,
 	.priority	=	10,
@@ -769,6 +771,7 @@ static struct xfrm6_protocol xfrmi_ah6_protocol __read_mostly = {
 
 static struct xfrm6_protocol xfrmi_ipcomp6_protocol __read_mostly = {
 	.handler	=	xfrm6_rcv,
+	.input_handler	=	xfrm_input,
 	.cb_handler	=	xfrmi_rcv_cb,
 	.err_handler	=	xfrmi6_err,
 	.priority	=	10,