#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/version.h>
-#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,26)
#include <linux/in.h>
#include <linux/ip.h>
#include <linux/net.h>
+#include <linux/module.h>
#include <linux/rculist.h>
#include <linux/udp.h>
#include <net/icmp.h>
#include <net/ip.h>
+#include <net/route.h>
#include <net/udp.h>
+#include <net/udp_tunnel.h>
+#include <net/xfrm.h>
#include "datapath.h"
-#include "tunnel.h"
+#include "gso.h"
#include "vport.h"
-
/*
* LISP encapsulation header:
*
};
static LIST_HEAD(lisp_ports);
+static struct vport_ops ovs_lisp_vport_ops;
static inline struct lisp_port *lisp_vport(const struct vport *vport)
{
#endif
}
-static void lisp_build_header(const struct vport *vport,
- struct sk_buff *skb,
- int tunnel_hlen)
+/* 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)
{
- 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_key;
+ u32 hash = skb_get_hash(skb);
+ unsigned int range;
+ int high;
+ int low;
+
+ if (!hash) {
+ if (skb->protocol == htons(ETH_P_IP)) {
+ struct iphdr *iph;
+ int size = (sizeof(iph->saddr) * 2) / sizeof(u32);
+
+ iph = (struct iphdr *) skb_network_header(skb);
+ hash = jhash2((const u32 *)&iph->saddr, size, 0);
+ } else if (skb->protocol == htons(ETH_P_IPV6)) {
+ struct ipv6hdr *ipv6hdr;
+
+ ipv6hdr = (struct ipv6hdr *) skb_network_header(skb);
+ hash = jhash2((const u32 *)&ipv6hdr->saddr,
+ (sizeof(struct in6_addr) * 2) / 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(struct sk_buff *skb)
+{
+ struct lisphdr *lisph;
+ const struct ovs_key_ipv4_tunnel *tun_key;
- udph->dest = lisp_port->dst_port;
- udph->source = htons(ovs_tnl_get_src_port(skb));
- udph->check = 0;
- udph->len = htons(skb->len - skb_transport_offset(skb));
+ tun_key = &OVS_CB(skb)->egress_tun_info->tunnel;
+ lisph = (struct lisphdr *)__skb_push(skb, sizeof(struct lisphdr));
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 */
struct lisp_port *lisp_port;
struct lisphdr *lisph;
struct iphdr *iph, *inner_iph;
- struct ovs_key_ipv4_tunnel tun_key;
+ 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);
+ lisp_port = rcu_dereference_sk_user_data(sk);
if (unlikely(!lisp_port))
goto error;
- if (unlikely(!pskb_may_pull(skb, LISP_HLEN)))
+ if (iptunnel_pull_header(skb, LISP_HLEN, 0))
goto error;
lisph = lisp_hdr(skb);
- skb_pull_rcsum(skb, LISP_HLEN);
-
if (lisph->instance_id_present != 1)
key = 0;
else
/* Save outer tunnel values */
iph = ip_hdr(skb);
- tnl_tun_key_init(&tun_key, iph, key, TUNNEL_KEY);
+ ovs_flow_tun_info_init(&tun_info, iph,
+ udp_hdr(skb)->source, udp_hdr(skb)->dest,
+ key, TUNNEL_KEY, NULL, 0);
/* Drop non-IP inner packets */
inner_iph = (struct iphdr *)(lisph + 1);
default:
goto error;
}
+ skb->protocol = protocol;
/* Add Ethernet header */
ethh = (struct ethhdr *)skb_push(skb, ETH_HLEN);
ovs_skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
- ovs_tnl_rcv(vport_from_priv(lisp_port), skb, &tun_key);
+ ovs_vport_receive(vport_from_priv(lisp_port), skb, &tun_info);
goto out;
error:
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;
+ struct udp_port_cfg udp_conf;
+ struct udp_tunnel_sock_cfg tunnel_cfg;
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);
+ memset(&udp_conf, 0, sizeof(udp_conf));
- sin.sin_family = AF_INET;
- sin.sin_addr.s_addr = htonl(INADDR_ANY);
- sin.sin_port = lisp_port->dst_port;
+ udp_conf.family = AF_INET;
+ udp_conf.local_ip.s_addr = htonl(INADDR_ANY);
+ udp_conf.local_udp_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;
+ err = udp_sock_create(net, &udp_conf, &lisp_port->lisp_rcv_socket);
+ if (err < 0) {
+ pr_warn("cannot register lisp protocol handler: %d\n", err);
+ return err;
+ }
- 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;
+ tunnel_cfg.sk_user_data = lisp_port;
+ tunnel_cfg.encap_type = 1;
+ tunnel_cfg.encap_rcv = lisp_rcv;
+ tunnel_cfg.encap_destroy = NULL;
- udp_encap_enable();
+ setup_udp_tunnel_sock(net, lisp_port->lisp_rcv_socket, &tunnel_cfg);
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);
list_del_rcu(&lisp_port->list);
- /* Release socket */
- sk_release_kernel(lisp_port->lisp_rcv_socket->sk);
-
+ udp_tunnel_sock_release(lisp_port->lisp_rcv_socket);
ovs_vport_deferred_free(vport);
}
return ERR_PTR(err);
}
-static int lisp_tnl_send(struct vport *vport, struct sk_buff *skb)
+static int lisp_send(struct vport *vport, struct sk_buff *skb)
{
- int tnl_len;
+ struct ovs_key_ipv4_tunnel *tun_key;
+ struct lisp_port *lisp_port = lisp_vport(vport);
+ struct net *net = ovs_dp_get_net(vport->dp);
int network_offset = skb_network_offset(skb);
+ struct rtable *rt;
+ int min_headroom;
+ __be32 saddr;
+ __be16 src_port, dst_port;
+ __be16 df;
+ int sent_len;
+ int err;
- if (unlikely(!OVS_CB(skb)->tun_key))
- return -EINVAL;
+ if (unlikely(!OVS_CB(skb)->egress_tun_info)) {
+ err = -EINVAL;
+ goto error;
+ }
- /* We only encapsulate IPv4 and IPv6 packets */
- switch (skb->protocol) {
- case htons(ETH_P_IP):
- case htons(ETH_P_IPV6):
- /* Pop off "inner" Ethernet header */
- skb_pull(skb, network_offset);
- tnl_len = ovs_tnl_send(vport, skb, IPPROTO_UDP,
- LISP_HLEN, lisp_build_header);
- return tnl_len > 0 ? tnl_len + network_offset : tnl_len;
- default:
- kfree_skb(skb);
- return 0;
+ tun_key = &OVS_CB(skb)->egress_tun_info->tunnel;
+
+ if (skb->protocol != htons(ETH_P_IP) &&
+ skb->protocol != htons(ETH_P_IPV6)) {
+ err = 0;
+ goto error;
+ }
+
+ /* 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 = udp_tunnel_handle_offloads(skb, false, false);
+ if (IS_ERR(skb)) {
+ err = PTR_ERR(skb);
+ skb = NULL;
+ goto err_free_rt;
+ }
+
+ src_port = htons(get_src_port(net, skb));
+ dst_port = lisp_port->dst_port;
+
+ lisp_build_header(skb);
+
+ skb->ignore_df = 1;
+
+ ovs_skb_set_inner_protocol(skb, skb->protocol);
+
+ df = tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0;
+ sent_len = udp_tunnel_xmit_skb(rt, lisp_port->lisp_rcv_socket->sk, skb,
+ saddr, tun_key->ipv4_dst,
+ tun_key->ipv4_tos, tun_key->ipv4_ttl,
+ df, src_port, dst_port, false, true);
+
+ return sent_len > 0 ? sent_len + network_offset : sent_len;
+
+err_free_rt:
+ ip_rt_put(rt);
+error:
+ kfree_skb(skb);
+ return err;
}
static const char *lisp_get_name(const struct vport *vport)
return lisp_port->name;
}
-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_tnl_send,
+static int lisp_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
+ struct ovs_tunnel_info *egress_tun_info)
+{
+ struct net *net = ovs_dp_get_net(vport->dp);
+ struct lisp_port *lisp_port = lisp_vport(vport);
+
+ if (skb->protocol != htons(ETH_P_IP) &&
+ skb->protocol != htons(ETH_P_IPV6)) {
+ return -EINVAL;
+ }
+
+ /*
+ * Get tp_src and tp_dst, refert to lisp_build_header().
+ */
+ return ovs_tunnel_get_egress_info(egress_tun_info, net,
+ OVS_CB(skb)->egress_tun_info,
+ IPPROTO_UDP, skb->mark,
+ htons(get_src_port(net, skb)),
+ lisp_port->dst_port);
+}
+
+static 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,
+ .get_egress_tun_info = lisp_get_egress_tun_info,
+ .owner = THIS_MODULE,
};
-#else
-#warning LISP tunneling will not be available on kernels before 2.6.26
-#endif /* Linux kernel < 2.6.26 */
+
+static int __init ovs_lisp_tnl_init(void)
+{
+ return ovs_vport_ops_register(&ovs_lisp_vport_ops);
+}
+
+static void __exit ovs_lisp_tnl_exit(void)
+{
+ ovs_vport_ops_unregister(&ovs_lisp_vport_ops);
+}
+
+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");