X-Git-Url: http://git.cascardo.eti.br/?a=blobdiff_plain;f=datapath%2Fvport-lisp.c;h=3ce8613aef1f0f36eb1a3593207ca4c152280526;hb=38dc2ca13a8412076773c53676ee49db1e427ff2;hp=a124e73daaa3c42e500973bff8c2dd060b80d21f;hpb=c1fc1411d204c59608bf9fe36a65bd221b10cbb2;p=cascardo%2Fovs.git

diff --git a/datapath/vport-lisp.c b/datapath/vport-lisp.c
index a124e73da..3ce8613ae 100644
--- a/datapath/vport-lisp.c
+++ b/datapath/vport-lisp.c
@@ -1,32 +1,23 @@
 /*
- * Copyright (c) 2011 Nicira, Inc.
- * Copyright (c) 2013 Cisco Systems, Inc.
+ * Copyright (c) 2015 Nicira, Inc.
  *
  * This program is free software; you can redistribute it and/or
- * modify it under the terms of version 2 of the GNU General Public
- * License as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
- * 02110-1301, USA
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#include <linux/version.h>
-
 #include <linux/in.h>
 #include <linux/ip.h>
 #include <linux/net.h>
 #include <linux/rculist.h>
 #include <linux/udp.h>
+#include <linux/if_vlan.h>
+#include <linux/module.h>
 
+#include <net/lisp.h>
 #include <net/icmp.h>
 #include <net/ip.h>
 #include <net/route.h>
@@ -34,307 +25,43 @@
 #include <net/xfrm.h>
 
 #include "datapath.h"
-#include "gso.h"
 #include "vport.h"
+#include "vport-netdev.h"
 
-/*
- *  LISP encapsulation header:
- *
- *  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *  |N|L|E|V|I|flags|            Nonce/Map-Version                  |
- *  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *  |                 Instance ID/Locator Status Bits               |
- *  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *
- */
-
-/**
- * struct lisphdr - LISP header
- * @nonce_present: Flag indicating the presence of a 24 bit nonce value.
- * @locator_status_bits_present: Flag indicating the presence of Locator Status
- *                               Bits (LSB).
- * @solicit_echo_nonce: Flag indicating the use of the echo noncing mechanism.
- * @map_version_present: Flag indicating the use of mapping versioning.
- * @instance_id_present: Flag indicating the presence of a 24 bit Instance ID.
- * @reserved_flags: 3 bits reserved for future flags.
- * @nonce: 24 bit nonce value.
- * @map_version: 24 bit mapping version.
- * @locator_status_bits: Locator Status Bits: 32 bits when instance_id_present
- *                       is not set, 8 bits when it is.
- * @instance_id: 24 bit Instance ID
- */
-struct lisphdr {
-#ifdef __LITTLE_ENDIAN_BITFIELD
-	__u8 reserved_flags:3;
-	__u8 instance_id_present:1;
-	__u8 map_version_present:1;
-	__u8 solicit_echo_nonce:1;
-	__u8 locator_status_bits_present:1;
-	__u8 nonce_present:1;
-#else
-	__u8 nonce_present:1;
-	__u8 locator_status_bits_present:1;
-	__u8 solicit_echo_nonce:1;
-	__u8 map_version_present:1;
-	__u8 instance_id_present:1;
-	__u8 reserved_flags:3;
-#endif
-	union {
-		__u8 nonce[3];
-		__u8 map_version[3];
-	} u1;
-	union {
-		__be32 locator_status_bits;
-		struct {
-			__u8 instance_id[3];
-			__u8 locator_status_bits;
-		} word2;
-	} u2;
-};
-
-#define LISP_HLEN (sizeof(struct udphdr) + sizeof(struct lisphdr))
-
+static struct vport_ops ovs_lisp_vport_ops;
 /**
  * struct lisp_port - Keeps track of open UDP ports
- * @dst_port: lisp UDP port no.
- * @list: list element in @lisp_ports.
- * @lisp_rcv_socket: The socket created for this port number.
- * @name: vport name.
+ * @dst_port: destination port.
  */
 struct lisp_port {
-	__be16 dst_port;
-	struct list_head list;
-	struct socket *lisp_rcv_socket;
-	char name[IFNAMSIZ];
+	u16 port_no;
 };
 
-static LIST_HEAD(lisp_ports);
-
 static inline struct lisp_port *lisp_vport(const struct vport *vport)
 {
 	return vport_priv(vport);
 }
 
-static struct lisp_port *lisp_find_port(struct net *net, __be16 port)
-{
-	struct lisp_port *lisp_port;
-
-	list_for_each_entry_rcu(lisp_port, &lisp_ports, list) {
-		if (lisp_port->dst_port == port &&
-			net_eq(sock_net(lisp_port->lisp_rcv_socket->sk), net))
-			return lisp_port;
-	}
-
-	return NULL;
-}
-
-static inline struct lisphdr *lisp_hdr(const struct sk_buff *skb)
-{
-	return (struct lisphdr *)(udp_hdr(skb) + 1);
-}
-
-/* Convert 64 bit tunnel ID to 24 bit Instance ID. */
-static void tunnel_id_to_instance_id(__be64 tun_id, __u8 *iid)
-{
-
-#ifdef __BIG_ENDIAN
-	iid[0] = (__force __u8)(tun_id >> 16);
-	iid[1] = (__force __u8)(tun_id >> 8);
-	iid[2] = (__force __u8)tun_id;
-#else
-	iid[0] = (__force __u8)((__force u64)tun_id >> 40);
-	iid[1] = (__force __u8)((__force u64)tun_id >> 48);
-	iid[2] = (__force __u8)((__force u64)tun_id >> 56);
-#endif
-}
-
-/* Convert 24 bit Instance ID to 64 bit tunnel ID. */
-static __be64 instance_id_to_tunnel_id(__u8 *iid)
-{
-#ifdef __BIG_ENDIAN
-	return (iid[0] << 16) | (iid[1] << 8) | iid[2];
-#else
-	return (__force __be64)(((__force u64)iid[0] << 40) |
-				((__force u64)iid[1] << 48) |
-				((__force u64)iid[2] << 56));
-#endif
-}
-
-/* Compute source UDP port for outgoing packet.
- * Currently we use the flow hash.
- */
-static u16 get_src_port(struct net *net, struct sk_buff *skb)
-{
-	u32 hash = skb_get_hash(skb);
-	unsigned int range;
-	int high;
-	int low;
-
-	if (!hash) {
-		struct sw_flow_key *pkt_key = OVS_CB(skb)->pkt_key;
-
-		if (skb->protocol == htons(ETH_P_IP))
-			hash = jhash2((const u32 *)&pkt_key->ipv4.addr,
-				   sizeof(pkt_key->ipv4.addr) / sizeof(u32), 0);
-		else if (skb->protocol == htons(ETH_P_IPV6))
-			hash = jhash2((const u32 *)&pkt_key->ipv6.addr,
-				   sizeof(pkt_key->ipv6.addr) / sizeof(u32), 0);
-		else
-			pr_warn_once("LISP inner protocol is not IP when "
-				     "calculating hash.\n");
-	}
-
-	inet_get_local_port_range(net, &low, &high);
-	range = (high - low) + 1;
-	return (((u64) hash * range) >> 32) + low;
-}
-
-static void lisp_build_header(const struct vport *vport,
+static int lisp_get_options(const struct vport *vport,
 			      struct sk_buff *skb)
 {
-	struct net *net = ovs_dp_get_net(vport->dp);
 	struct lisp_port *lisp_port = lisp_vport(vport);
-	struct udphdr *udph = udp_hdr(skb);
-	struct lisphdr *lisph = (struct lisphdr *)(udph + 1);
-	const struct ovs_key_ipv4_tunnel *tun_key = &OVS_CB(skb)->tun_info->tunnel;
-
-	udph->dest = lisp_port->dst_port;
-	udph->source = htons(get_src_port(net, skb));
-	udph->check = 0;
-	udph->len = htons(skb->len - skb_transport_offset(skb));
-
-	lisph->nonce_present = 0;	/* We don't support echo nonce algorithm */
-	lisph->locator_status_bits_present = 1;	/* Set LSB */
-	lisph->solicit_echo_nonce = 0;	/* No echo noncing */
-	lisph->map_version_present = 0;	/* No mapping versioning, nonce instead */
-	lisph->instance_id_present = 1;	/* Store the tun_id as Instance ID  */
-	lisph->reserved_flags = 0;	/* Reserved flags, set to 0  */
-
-	lisph->u1.nonce[0] = 0;
-	lisph->u1.nonce[1] = 0;
-	lisph->u1.nonce[2] = 0;
 
-	tunnel_id_to_instance_id(tun_key->tun_id, &lisph->u2.word2.instance_id[0]);
-	lisph->u2.word2.locator_status_bits = 1;
-}
-
-/* Called with rcu_read_lock and BH disabled. */
-static int lisp_rcv(struct sock *sk, struct sk_buff *skb)
-{
-	struct lisp_port *lisp_port;
-	struct lisphdr *lisph;
-	struct iphdr *iph, *inner_iph;
-	struct ovs_tunnel_info tun_info;
-	__be64 key;
-	struct ethhdr *ethh;
-	__be16 protocol;
-
-	lisp_port = lisp_find_port(dev_net(skb->dev), udp_hdr(skb)->dest);
-	if (unlikely(!lisp_port))
-		goto error;
-
-	if (iptunnel_pull_header(skb, LISP_HLEN, 0))
-		goto error;
-
-	lisph = lisp_hdr(skb);
-
-	if (lisph->instance_id_present != 1)
-		key = 0;
-	else
-		key = instance_id_to_tunnel_id(&lisph->u2.word2.instance_id[0]);
-
-	/* Save outer tunnel values */
-	iph = ip_hdr(skb);
-	ovs_flow_tun_info_init(&tun_info, iph, key, TUNNEL_KEY, NULL, 0);
-
-	/* Drop non-IP inner packets */
-	inner_iph = (struct iphdr *)(lisph + 1);
-	switch (inner_iph->version) {
-	case 4:
-		protocol = htons(ETH_P_IP);
-		break;
-	case 6:
-		protocol = htons(ETH_P_IPV6);
-		break;
-	default:
-		goto error;
-	}
-	skb->protocol = protocol;
-
-	/* Add Ethernet header */
-	ethh = (struct ethhdr *)skb_push(skb, ETH_HLEN);
-	memset(ethh, 0, ETH_HLEN);
-	ethh->h_dest[0] = 0x02;
-	ethh->h_source[0] = 0x02;
-	ethh->h_proto = protocol;
-
-	ovs_skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
-
-	ovs_vport_receive(vport_from_priv(lisp_port), skb, &tun_info);
-	goto out;
-
-error:
-	kfree_skb(skb);
-out:
-	return 0;
-}
-
-/* Arbitrary value.  Irrelevant as long as it's not 0 since we set the handler. */
-#define UDP_ENCAP_LISP 1
-static int lisp_socket_init(struct lisp_port *lisp_port, struct net *net)
-{
-	struct sockaddr_in sin;
-	int err;
-
-	err = sock_create_kern(AF_INET, SOCK_DGRAM, 0,
-			       &lisp_port->lisp_rcv_socket);
-	if (err)
-		goto error;
-
-	/* release net ref. */
-	sk_change_net(lisp_port->lisp_rcv_socket->sk, net);
-
-	sin.sin_family = AF_INET;
-	sin.sin_addr.s_addr = htonl(INADDR_ANY);
-	sin.sin_port = lisp_port->dst_port;
-
-	err = kernel_bind(lisp_port->lisp_rcv_socket, (struct sockaddr *)&sin,
-			  sizeof(struct sockaddr_in));
-	if (err)
-		goto error_sock;
-
-	udp_sk(lisp_port->lisp_rcv_socket->sk)->encap_type = UDP_ENCAP_LISP;
-	udp_sk(lisp_port->lisp_rcv_socket->sk)->encap_rcv = lisp_rcv;
-
-	udp_encap_enable();
-
-	return 0;
-
-error_sock:
-	sk_release_kernel(lisp_port->lisp_rcv_socket->sk);
-error:
-	pr_warn("cannot register lisp protocol handler: %d\n", err);
-	return err;
-}
-
-static int lisp_get_options(const struct vport *vport, struct sk_buff *skb)
-{
-	struct lisp_port *lisp_port = lisp_vport(vport);
-
-	if (nla_put_u16(skb, OVS_TUNNEL_ATTR_DST_PORT, ntohs(lisp_port->dst_port)))
+	if (nla_put_u16(skb, OVS_TUNNEL_ATTR_DST_PORT, lisp_port->port_no))
 		return -EMSGSIZE;
 	return 0;
 }
 
-static void lisp_tnl_destroy(struct vport *vport)
+static int lisp_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
+				      struct dp_upcall_info *upcall)
 {
 	struct lisp_port *lisp_port = lisp_vport(vport);
+	struct net *net = ovs_dp_get_net(vport->dp);
+	__be16 dport = htons(lisp_port->port_no);
+	__be16 sport = udp_flow_src_port(net, skb, 1, USHRT_MAX, true);
 
-	list_del_rcu(&lisp_port->list);
-	/* Release socket */
-	sk_release_kernel(lisp_port->lisp_rcv_socket->sk);
-
-	ovs_vport_deferred_free(vport);
+	return ovs_tunnel_get_egress_info(upcall, ovs_dp_get_net(vport->dp),
+					  skb, IPPROTO_UDP, sport, dport);
 }
 
 static struct vport *lisp_tnl_create(const struct vport_parms *parms)
@@ -342,10 +69,11 @@ static struct vport *lisp_tnl_create(const struct vport_parms *parms)
 	struct net *net = ovs_dp_get_net(parms->dp);
 	struct nlattr *options = parms->options;
 	struct lisp_port *lisp_port;
+	struct net_device *dev;
 	struct vport *vport;
 	struct nlattr *a;
-	int err;
 	u16 dst_port;
+	int err;
 
 	if (!options) {
 		err = -EINVAL;
@@ -361,157 +89,62 @@ static struct vport *lisp_tnl_create(const struct vport_parms *parms)
 		goto error;
 	}
 
-	/* Verify if we already have a socket created for this port */
-	if (lisp_find_port(net, htons(dst_port))) {
-		err = -EEXIST;
-		goto error;
-	}
-
 	vport = ovs_vport_alloc(sizeof(struct lisp_port),
 				&ovs_lisp_vport_ops, parms);
 	if (IS_ERR(vport))
 		return vport;
 
 	lisp_port = lisp_vport(vport);
-	lisp_port->dst_port = htons(dst_port);
-	strncpy(lisp_port->name, parms->name, IFNAMSIZ);
-
-	err = lisp_socket_init(lisp_port, net);
-	if (err)
-		goto error_free;
+	lisp_port->port_no = dst_port;
+
+	rtnl_lock();
+	dev = lisp_dev_create_fb(net, parms->name, NET_NAME_USER, dst_port);
+	if (IS_ERR(dev)) {
+		rtnl_unlock();
+		ovs_vport_free(vport);
+		return ERR_CAST(dev);
+	}
 
-	list_add_tail_rcu(&lisp_port->list, &lisp_ports);
+	dev_change_flags(dev, dev->flags | IFF_UP);
+	rtnl_unlock();
 	return vport;
-
-error_free:
-	ovs_vport_free(vport);
 error:
 	return ERR_PTR(err);
 }
 
-#if LINUX_VERSION_CODE < KERNEL_VERSION(3,12,0)
-
-static void lisp_fix_segment(struct sk_buff *skb)
+static struct vport *lisp_create(const struct vport_parms *parms)
 {
-	struct udphdr *udph = udp_hdr(skb);
+	struct vport *vport;
 
-	udph->len = htons(skb->len - skb_transport_offset(skb));
-}
+	vport = lisp_tnl_create(parms);
+	if (IS_ERR(vport))
+		return vport;
 
-static int handle_offloads(struct sk_buff *skb)
-{
-	if (skb_is_gso(skb))
-		OVS_GSO_CB(skb)->fix_segment = lisp_fix_segment;
-	else if (skb->ip_summed != CHECKSUM_PARTIAL)
-		skb->ip_summed = CHECKSUM_NONE;
-	return 0;
+	return ovs_netdev_link(vport, parms->name);
 }
-#else
-static int handle_offloads(struct sk_buff *skb)
-{
-	if (skb_is_gso(skb)) {
-		int err = skb_unclone(skb, GFP_ATOMIC);
-		if (unlikely(err))
-			return err;
-
-		skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL;
-	} else if (skb->ip_summed != CHECKSUM_PARTIAL)
-		skb->ip_summed = CHECKSUM_NONE;
 
-	skb->encapsulation = 1;
-	return 0;
-}
-#endif
+static struct vport_ops ovs_lisp_vport_ops = {
+	.type		= OVS_VPORT_TYPE_LISP,
+	.create		= lisp_create,
+	.destroy	= ovs_netdev_tunnel_destroy,
+	.get_options	= lisp_get_options,
+	.send		= lisp_xmit,
+	.get_egress_tun_info	= lisp_get_egress_tun_info,
+};
 
-static int lisp_send(struct vport *vport, struct sk_buff *skb)
+static int __init ovs_lisp_tnl_init(void)
 {
-	struct ovs_key_ipv4_tunnel *tun_key = &OVS_CB(skb)->tun_info->tunnel;
-	int network_offset = skb_network_offset(skb);
-	struct rtable *rt;
-	int min_headroom;
-	__be32 saddr;
-	__be16 df;
-	int sent_len;
-	int err;
-
-	if (unlikely(!OVS_CB(skb)->tun_info))
-		return -EINVAL;
-
-	if (skb->protocol != htons(ETH_P_IP) &&
-	    skb->protocol != htons(ETH_P_IPV6)) {
-		kfree_skb(skb);
-		return 0;
-	}
-
-	/* Route lookup */
-	saddr = tun_key->ipv4_src;
-	rt = find_route(ovs_dp_get_net(vport->dp),
-			&saddr, tun_key->ipv4_dst,
-			IPPROTO_UDP, tun_key->ipv4_tos,
-			skb->mark);
-	if (IS_ERR(rt)) {
-		err = PTR_ERR(rt);
-		goto error;
-	}
-
-	min_headroom = LL_RESERVED_SPACE(rt_dst(rt).dev) + rt_dst(rt).header_len
-			+ sizeof(struct iphdr) + LISP_HLEN;
-
-	if (skb_headroom(skb) < min_headroom || skb_header_cloned(skb)) {
-		int head_delta = SKB_DATA_ALIGN(min_headroom -
-						skb_headroom(skb) +
-						16);
-
-		err = pskb_expand_head(skb, max_t(int, head_delta, 0),
-					0, GFP_ATOMIC);
-		if (unlikely(err))
-			goto err_free_rt;
-	}
-
-	/* Reset l2 headers. */
-	skb_pull(skb, network_offset);
-	skb_reset_mac_header(skb);
-	vlan_set_tci(skb, 0);
-
-	skb_reset_inner_headers(skb);
-
-	__skb_push(skb, LISP_HLEN);
-	skb_reset_transport_header(skb);
-
-	lisp_build_header(vport, skb);
-
-	/* Offloading */
-	err = handle_offloads(skb);
-	if (err)
-		goto err_free_rt;
-
-	skb->local_df = 1;
-
-	df = tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0;
-	sent_len = iptunnel_xmit(rt, skb,
-			     saddr, tun_key->ipv4_dst,
-			     IPPROTO_UDP, tun_key->ipv4_tos,
-			     tun_key->ipv4_ttl, df, false);
-
-	return sent_len > 0 ? sent_len + network_offset : sent_len;
-
-err_free_rt:
-	ip_rt_put(rt);
-error:
-	return err;
+	return ovs_vport_ops_register(&ovs_lisp_vport_ops);
 }
 
-static const char *lisp_get_name(const struct vport *vport)
+static void __exit ovs_lisp_tnl_exit(void)
 {
-	struct lisp_port *lisp_port = lisp_vport(vport);
-	return lisp_port->name;
+	ovs_vport_ops_unregister(&ovs_lisp_vport_ops);
 }
 
-const struct vport_ops ovs_lisp_vport_ops = {
-	.type		= OVS_VPORT_TYPE_LISP,
-	.create		= lisp_tnl_create,
-	.destroy	= lisp_tnl_destroy,
-	.get_name	= lisp_get_name,
-	.get_options	= lisp_get_options,
-	.send		= lisp_send,
-};
+module_init(ovs_lisp_tnl_init);
+module_exit(ovs_lisp_tnl_exit);
+
+MODULE_DESCRIPTION("OVS: Lisp switching port");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("vport-type-105");