Merge branch 'udp_tunnel'

Tom Herbert says:

====================
udp: UDP tunnel enhancements

- Add udp_sock_create in new helper module udp_tunnel. Tunnel
  implementations call this function to create listener UDP ports.
- Make vxlan and l2tp call udp_sock_create.
- Move udp_tunnel_segment into udp_offload.c.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

drivers/net/Kconfig

@@ -148,6 +148,7 @@ config VXLAN
        tristate "Virtual eXtensible Local Area Network (VXLAN)"
        depends on INET
        select NET_IP_TUNNEL
+       select NET_UDP_TUNNEL
        ---help---
 	  This allows one to create vxlan virtual interfaces that provide
 	  Layer 2 Networks over Layer 3 Networks. VXLAN is often used

drivers/net/vxlan.c

@@ -33,6 +33,7 @@
 #include <net/ip_tunnels.h>
 #include <net/icmp.h>
 #include <net/udp.h>
+#include <net/udp_tunnel.h>
 #include <net/rtnetlink.h>
 #include <net/route.h>
 #include <net/dsfield.h>
@@ -2339,102 +2340,37 @@ static void vxlan_del_work(struct work_struct *work)
 	kfree_rcu(vs, rcu);
 }
 
-#if IS_ENABLED(CONFIG_IPV6)
-/* Create UDP socket for encapsulation receive. AF_INET6 socket
- * could be used for both IPv4 and IPv6 communications, but
- * users may set bindv6only=1.
- */
-static struct socket *create_v6_sock(struct net *net, __be16 port, u32 flags)
+static struct socket *vxlan_create_sock(struct net *net, bool ipv6,
+					__be16 port, u32 flags)
 {
-	struct sock *sk;
 	struct socket *sock;
-	struct sockaddr_in6 vxlan_addr = {
-		.sin6_family = AF_INET6,
-		.sin6_port = port,
-	};
-	int rc, val = 1;
-
-	rc = sock_create_kern(AF_INET6, SOCK_DGRAM, IPPROTO_UDP, &sock);
-	if (rc < 0) {
-		pr_debug("UDPv6 socket create failed\n");
-		return ERR_PTR(rc);
-	}
-
-	/* Put in proper namespace */
-	sk = sock->sk;
-	sk_change_net(sk, net);
-
-	kernel_setsockopt(sock, SOL_IPV6, IPV6_V6ONLY,
-			  (char *)&val, sizeof(val));
-	rc = kernel_bind(sock, (struct sockaddr *)&vxlan_addr,
-			 sizeof(struct sockaddr_in6));
-	if (rc < 0) {
-		pr_debug("bind for UDPv6 socket %pI6:%u (%d)\n",
-			 &vxlan_addr.sin6_addr, ntohs(vxlan_addr.sin6_port), rc);
-		sk_release_kernel(sk);
-		return ERR_PTR(rc);
-	}
-	/* At this point, IPv6 module should have been loaded in
-	 * sock_create_kern().
-	 */
-	BUG_ON(!ipv6_stub);
-
-	/* Disable multicast loopback */
-	inet_sk(sk)->mc_loop = 0;
-
-	if (flags & VXLAN_F_UDP_ZERO_CSUM6_TX)
-		udp_set_no_check6_tx(sk, true);
-
-	if (flags & VXLAN_F_UDP_ZERO_CSUM6_RX)
-		udp_set_no_check6_rx(sk, true);
-
-	return sock;
-}
-
-#else
+	struct udp_port_cfg udp_conf;
+	int err;
 
-static struct socket *create_v6_sock(struct net *net, __be16 port, u32 flags)
-{
-		return ERR_PTR(-EPFNOSUPPORT);
-}
-#endif
+	memset(&udp_conf, 0, sizeof(udp_conf));
 
-static struct socket *create_v4_sock(struct net *net, __be16 port, u32 flags)
-{
-	struct sock *sk;
-	struct socket *sock;
-	struct sockaddr_in vxlan_addr = {
-		.sin_family = AF_INET,
-		.sin_addr.s_addr = htonl(INADDR_ANY),
-		.sin_port = port,
-	};
-	int rc;
-
-	/* Create UDP socket for encapsulation receive. */
-	rc = sock_create_kern(AF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock);
-	if (rc < 0) {
-		pr_debug("UDP socket create failed\n");
-		return ERR_PTR(rc);
+	if (ipv6) {
+		udp_conf.family = AF_INET6;
+		udp_conf.use_udp6_tx_checksums =
+		    !!(flags & VXLAN_F_UDP_ZERO_CSUM6_TX);
+		udp_conf.use_udp6_rx_checksums =
+		    !!(flags & VXLAN_F_UDP_ZERO_CSUM6_RX);
+	} else {
+		udp_conf.family = AF_INET;
+		udp_conf.local_ip.s_addr = INADDR_ANY;
+		udp_conf.use_udp_checksums =
+		    !!(flags & VXLAN_F_UDP_CSUM);
 	}
 
-	/* Put in proper namespace */
-	sk = sock->sk;
-	sk_change_net(sk, net);
+	udp_conf.local_udp_port = port;
 
-	rc = kernel_bind(sock, (struct sockaddr *) &vxlan_addr,
-			 sizeof(vxlan_addr));
-	if (rc < 0) {
-		pr_debug("bind for UDP socket %pI4:%u (%d)\n",
-			 &vxlan_addr.sin_addr, ntohs(vxlan_addr.sin_port), rc);
-		sk_release_kernel(sk);
-		return ERR_PTR(rc);
-	}
+	/* Open UDP socket */
+	err = udp_sock_create(net, &udp_conf, &sock);
+	if (err < 0)
+		return ERR_PTR(err);
 
 	/* Disable multicast loopback */
-	inet_sk(sk)->mc_loop = 0;
-
-	if (!(flags & VXLAN_F_UDP_CSUM))
-		sock->sk->sk_no_check_tx = 1;
+	inet_sk(sock->sk)->mc_loop = 0;
 
 	return sock;
 }
@@ -2460,10 +2396,7 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port,
 
 	INIT_WORK(&vs->del_work, vxlan_del_work);
 
-	if (ipv6)
-		sock = create_v6_sock(net, port, flags);
-	else
-		sock = create_v4_sock(net, port, flags);
+	sock = vxlan_create_sock(net, ipv6, port, flags);
 	if (IS_ERR(sock)) {
 		kfree(vs);
 		return ERR_CAST(sock);

include/net/udp_tunnel.h

+#ifndef __NET_UDP_TUNNEL_H
+#define __NET_UDP_TUNNEL_H
+
+struct udp_port_cfg {
+	u8			family;
+
+	/* Used only for kernel-created sockets */
+	union {
+		struct in_addr		local_ip;
+#if IS_ENABLED(CONFIG_IPV6)
+		struct in6_addr		local_ip6;
+#endif
+	};
+
+	union {
+		struct in_addr		peer_ip;
+#if IS_ENABLED(CONFIG_IPV6)
+		struct in6_addr		peer_ip6;
+#endif
+	};
+
+	__be16			local_udp_port;
+	__be16			peer_udp_port;
+	unsigned int		use_udp_checksums:1,
+				use_udp6_tx_checksums:1,
+				use_udp6_rx_checksums:1;
+};
+
+int udp_sock_create(struct net *net, struct udp_port_cfg *cfg,
+		    struct socket **sockp);
+
+#endif

net/ipv4/Kconfig

@@ -307,6 +307,10 @@ config NET_IPVTI
 	  the notion of a secure tunnel for IPSEC and then use routing protocol
 	  on top.
 
+config NET_UDP_TUNNEL
+	tristate
+	default n
+
 config INET_AH
 	tristate "IP: AH transformation"
 	select XFRM_ALGO

net/ipv4/Makefile

@@ -22,6 +22,7 @@ obj-$(CONFIG_NET_IPIP) += ipip.o
 gre-y := gre_demux.o
 obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o
 obj-$(CONFIG_NET_IPGRE) += ip_gre.o
+obj-$(CONFIG_NET_UDP_TUNNEL) += udp_tunnel.o
 obj-$(CONFIG_NET_IPVTI) += ip_vti.o
 obj-$(CONFIG_SYN_COOKIES) += syncookies.o
 obj-$(CONFIG_INET_AH) += ah4.o

net/ipv4/udp.c

@@ -2522,79 +2522,3 @@ void __init udp_init(void)
 	sysctl_udp_rmem_min = SK_MEM_QUANTUM;
 	sysctl_udp_wmem_min = SK_MEM_QUANTUM;
 }
-
-struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb,
-				       netdev_features_t features)
-{
-	struct sk_buff *segs = ERR_PTR(-EINVAL);
-	u16 mac_offset = skb->mac_header;
-	int mac_len = skb->mac_len;
-	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
-	__be16 protocol = skb->protocol;
-	netdev_features_t enc_features;
-	int udp_offset, outer_hlen;
-	unsigned int oldlen;
-	bool need_csum;
-
-	oldlen = (u16)~skb->len;
-
-	if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
-		goto out;
-
-	skb->encapsulation = 0;
-	__skb_pull(skb, tnl_hlen);
-	skb_reset_mac_header(skb);
-	skb_set_network_header(skb, skb_inner_network_offset(skb));
-	skb->mac_len = skb_inner_network_offset(skb);
-	skb->protocol = htons(ETH_P_TEB);
-
-	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM);
-	if (need_csum)
-		skb->encap_hdr_csum = 1;
-
-	/* segment inner packet. */
-	enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
-	segs = skb_mac_gso_segment(skb, enc_features);
-	if (!segs || IS_ERR(segs)) {
-		skb_gso_error_unwind(skb, protocol, tnl_hlen, mac_offset,
-				     mac_len);
-		goto out;
-	}
-
-	outer_hlen = skb_tnl_header_len(skb);
-	udp_offset = outer_hlen - tnl_hlen;
-	skb = segs;
-	do {
-		struct udphdr *uh;
-		int len;
-
-		skb_reset_inner_headers(skb);
-		skb->encapsulation = 1;
-
-		skb->mac_len = mac_len;
-
-		skb_push(skb, outer_hlen);
-		skb_reset_mac_header(skb);
-		skb_set_network_header(skb, mac_len);
-		skb_set_transport_header(skb, udp_offset);
-		len = skb->len - udp_offset;
-		uh = udp_hdr(skb);
-		uh->len = htons(len);
-
-		if (need_csum) {
-			__be32 delta = htonl(oldlen + len);
-
-			uh->check = ~csum_fold((__force __wsum)
-					       ((__force u32)uh->check +
-						(__force u32)delta));
-			uh->check = gso_make_checksum(skb, ~uh->check);
-
-			if (uh->check == 0)
-				uh->check = CSUM_MANGLED_0;
-		}
-
-		skb->protocol = protocol;
-	} while ((skb = skb->next));
-out:
-	return segs;
-}

net/ipv4/udp_offload.c

@@ -47,6 +47,82 @@ static int udp4_ufo_send_check(struct sk_buff *skb)
 	return 0;
 }
 
+struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb,
+				       netdev_features_t features)
+{
+	struct sk_buff *segs = ERR_PTR(-EINVAL);
+	u16 mac_offset = skb->mac_header;
+	int mac_len = skb->mac_len;
+	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
+	__be16 protocol = skb->protocol;
+	netdev_features_t enc_features;
+	int udp_offset, outer_hlen;
+	unsigned int oldlen;
+	bool need_csum;
+
+	oldlen = (u16)~skb->len;
+
+	if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
+		goto out;
+
+	skb->encapsulation = 0;
+	__skb_pull(skb, tnl_hlen);
+	skb_reset_mac_header(skb);
+	skb_set_network_header(skb, skb_inner_network_offset(skb));
+	skb->mac_len = skb_inner_network_offset(skb);
+	skb->protocol = htons(ETH_P_TEB);
+
+	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM);
+	if (need_csum)
+		skb->encap_hdr_csum = 1;
+
+	/* segment inner packet. */
+	enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
+	segs = skb_mac_gso_segment(skb, enc_features);
+	if (!segs || IS_ERR(segs)) {
+		skb_gso_error_unwind(skb, protocol, tnl_hlen, mac_offset,
+				     mac_len);
+		goto out;
+	}
+
+	outer_hlen = skb_tnl_header_len(skb);
+	udp_offset = outer_hlen - tnl_hlen;
+	skb = segs;
+	do {
+		struct udphdr *uh;
+		int len;
+
+		skb_reset_inner_headers(skb);
+		skb->encapsulation = 1;
+
+		skb->mac_len = mac_len;
+
+		skb_push(skb, outer_hlen);
+		skb_reset_mac_header(skb);
+		skb_set_network_header(skb, mac_len);
+		skb_set_transport_header(skb, udp_offset);
+		len = skb->len - udp_offset;
+		uh = udp_hdr(skb);
+		uh->len = htons(len);
+
+		if (need_csum) {
+			__be32 delta = htonl(oldlen + len);
+
+			uh->check = ~csum_fold((__force __wsum)
+					       ((__force u32)uh->check +
+						(__force u32)delta));
+			uh->check = gso_make_checksum(skb, ~uh->check);
+
+			if (uh->check == 0)
+				uh->check = CSUM_MANGLED_0;
+		}
+
+		skb->protocol = protocol;
+	} while ((skb = skb->next));
+out:
+	return segs;
+}
+
 static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb,
 					 netdev_features_t features)
 {

net/ipv4/udp_tunnel.c

+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/socket.h>
+#include <linux/udp.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <net/udp.h>
+#include <net/udp_tunnel.h>
+#include <net/net_namespace.h>
+
+int udp_sock_create(struct net *net, struct udp_port_cfg *cfg,
+		    struct socket **sockp)
+{
+	int err = -EINVAL;
+	struct socket *sock = NULL;
+
+#if IS_ENABLED(CONFIG_IPV6)
+	if (cfg->family == AF_INET6) {
+		struct sockaddr_in6 udp6_addr;
+
+		err = sock_create_kern(AF_INET6, SOCK_DGRAM, 0, &sock);
+		if (err < 0)
+			goto error;
+
+		sk_change_net(sock->sk, net);
+
+		udp6_addr.sin6_family = AF_INET6;
+		memcpy(&udp6_addr.sin6_addr, &cfg->local_ip6,
+		       sizeof(udp6_addr.sin6_addr));
+		udp6_addr.sin6_port = cfg->local_udp_port;
+		err = kernel_bind(sock, (struct sockaddr *)&udp6_addr,
+				  sizeof(udp6_addr));
+		if (err < 0)
+			goto error;
+
+		if (cfg->peer_udp_port) {
+			udp6_addr.sin6_family = AF_INET6;
+			memcpy(&udp6_addr.sin6_addr, &cfg->peer_ip6,
+			       sizeof(udp6_addr.sin6_addr));
+			udp6_addr.sin6_port = cfg->peer_udp_port;
+			err = kernel_connect(sock,
+					     (struct sockaddr *)&udp6_addr,
+					     sizeof(udp6_addr), 0);
+		}
+		if (err < 0)
+			goto error;
+
+		udp_set_no_check6_tx(sock->sk, !cfg->use_udp6_tx_checksums);
+		udp_set_no_check6_rx(sock->sk, !cfg->use_udp6_rx_checksums);
+	} else
+#endif
+	if (cfg->family == AF_INET) {
+		struct sockaddr_in udp_addr;
+
+		err = sock_create_kern(AF_INET, SOCK_DGRAM, 0, &sock);
+		if (err < 0)
+			goto error;
+
+		sk_change_net(sock->sk, net);
+
+		udp_addr.sin_family = AF_INET;
+		udp_addr.sin_addr = cfg->local_ip;
+		udp_addr.sin_port = cfg->local_udp_port;
+		err = kernel_bind(sock, (struct sockaddr *)&udp_addr,
+				  sizeof(udp_addr));
+		if (err < 0)
+			goto error;
+
+		if (cfg->peer_udp_port) {
+			udp_addr.sin_family = AF_INET;
+			udp_addr.sin_addr = cfg->peer_ip;
+			udp_addr.sin_port = cfg->peer_udp_port;
+			err = kernel_connect(sock,
+					     (struct sockaddr *)&udp_addr,
+					     sizeof(udp_addr), 0);
+			if (err < 0)
+				goto error;
+		}
+
+		sock->sk->sk_no_check_tx = !cfg->use_udp_checksums;
+	} else {
+		return -EPFNOSUPPORT;
+	}
+
+
+	*sockp = sock;
+
+	return 0;
+
+error:
+	if (sock) {
+		kernel_sock_shutdown(sock, SHUT_RDWR);
+		sk_release_kernel(sock->sk);
+	}
+	*sockp = NULL;
+	return err;
+}
+EXPORT_SYMBOL(udp_sock_create);
+
+MODULE_LICENSE("GPL");

net/l2tp/Kconfig

@@ -6,6 +6,7 @@ menuconfig L2TP
 	tristate "Layer Two Tunneling Protocol (L2TP)"
 	depends on (IPV6 || IPV6=n)
 	depends on INET
+	select NET_UDP_TUNNEL
 	---help---
 	  Layer Two Tunneling Protocol
 

net/l2tp/l2tp_core.c

@@ -52,6 +52,7 @@
 #include <net/dst.h>
 #include <net/ip.h>
 #include <net/udp.h>
+#include <net/udp_tunnel.h>
 #include <net/inet_common.h>
 #include <net/xfrm.h>
 #include <net/protocol.h>
@@ -1358,81 +1359,46 @@ static int l2tp_tunnel_sock_create(struct net *net,
 {
 	int err = -EINVAL;
 	struct socket *sock = NULL;
-	struct sockaddr_in udp_addr = {0};
-	struct sockaddr_l2tpip ip_addr = {0};
-#if IS_ENABLED(CONFIG_IPV6)
-	struct sockaddr_in6 udp6_addr = {0};
-	struct sockaddr_l2tpip6 ip6_addr = {0};
-#endif
+	struct udp_port_cfg udp_conf;
 
 	switch (cfg->encap) {
 	case L2TP_ENCAPTYPE_UDP:
+		memset(&udp_conf, 0, sizeof(udp_conf));
+
 #if IS_ENABLED(CONFIG_IPV6)
 		if (cfg->local_ip6 && cfg->peer_ip6) {
-			err = sock_create_kern(AF_INET6, SOCK_DGRAM, 0, &sock);
-			if (err < 0)
-				goto out;
-
-			sk_change_net(sock->sk, net);
-
-			udp6_addr.sin6_family = AF_INET6;
-			memcpy(&udp6_addr.sin6_addr, cfg->local_ip6,
-			       sizeof(udp6_addr.sin6_addr));
-			udp6_addr.sin6_port = htons(cfg->local_udp_port);
-			err = kernel_bind(sock, (struct sockaddr *) &udp6_addr,
-					  sizeof(udp6_addr));
-			if (err < 0)
-				goto out;
-
-			udp6_addr.sin6_family = AF_INET6;
-			memcpy(&udp6_addr.sin6_addr, cfg->peer_ip6,
-			       sizeof(udp6_addr.sin6_addr));
-			udp6_addr.sin6_port = htons(cfg->peer_udp_port);
-			err = kernel_connect(sock,
-					     (struct sockaddr *) &udp6_addr,
-					     sizeof(udp6_addr), 0);
-			if (err < 0)
-				goto out;
-
-			if (cfg->udp6_zero_tx_checksums)
-				udp_set_no_check6_tx(sock->sk, true);
-			if (cfg->udp6_zero_rx_checksums)
-				udp_set_no_check6_rx(sock->sk, true);
+			udp_conf.family = AF_INET6;
+			memcpy(&udp_conf.local_ip6, cfg->local_ip6,
+			       sizeof(udp_conf.local_ip6));
+			memcpy(&udp_conf.peer_ip6, cfg->peer_ip6,
+			       sizeof(udp_conf.peer_ip6));
+			udp_conf.use_udp6_tx_checksums =
+			    cfg->udp6_zero_tx_checksums;
+			udp_conf.use_udp6_rx_checksums =
+			    cfg->udp6_zero_rx_checksums;
 		} else
 #endif
 		{
-			err = sock_create_kern(AF_INET, SOCK_DGRAM, 0, &sock);
-			if (err < 0)
-				goto out;
-
-			sk_change_net(sock->sk, net);
-
-			udp_addr.sin_family = AF_INET;
-			udp_addr.sin_addr = cfg->local_ip;
-			udp_addr.sin_port = htons(cfg->local_udp_port);
-			err = kernel_bind(sock, (struct sockaddr *) &udp_addr,
-					  sizeof(udp_addr));
-			if (err < 0)
-				goto out;
-
-			udp_addr.sin_family = AF_INET;
-			udp_addr.sin_addr = cfg->peer_ip;
-			udp_addr.sin_port = htons(cfg->peer_udp_port);
-			err = kernel_connect(sock,
-					     (struct sockaddr *) &udp_addr,
-					     sizeof(udp_addr), 0);
-			if (err < 0)
-				goto out;
+			udp_conf.family = AF_INET;
+			udp_conf.local_ip = cfg->local_ip;
+			udp_conf.peer_ip = cfg->peer_ip;
+			udp_conf.use_udp_checksums = cfg->use_udp_checksums;
 		}
 
-		if (!cfg->use_udp_checksums)
-			sock->sk->sk_no_check_tx = 1;
+		udp_conf.local_udp_port = htons(cfg->local_udp_port);
+		udp_conf.peer_udp_port = htons(cfg->peer_udp_port);
+
+		err = udp_sock_create(net, &udp_conf, &sock);
+		if (err < 0)
+			goto out;
 
 		break;
 
 	case L2TP_ENCAPTYPE_IP:
 #if IS_ENABLED(CONFIG_IPV6)
 		if (cfg->local_ip6 && cfg->peer_ip6) {
+			struct sockaddr_l2tpip6 ip6_addr = {0};
+
 			err = sock_create_kern(AF_INET6, SOCK_DGRAM,
 					  IPPROTO_L2TP, &sock);
 			if (err < 0)
@@ -1461,6 +1427,8 @@ static int l2tp_tunnel_sock_create(struct net *net,
 		} else
 #endif
 		{
+			struct sockaddr_l2tpip ip_addr = {0};
+
 			err = sock_create_kern(AF_INET, SOCK_DGRAM,
 					  IPPROTO_L2TP, &sock);
 			if (err < 0)