/*
- * Copyright (c) 2013 Nicira, Inc.
+ * Copyright (c) 2015 Nicira, Inc.
* Copyright (c) 2013 Cisco Systems, Inc.
*
* This program is free software; you can redistribute it and/or
* 02110-1301, USA
*/
-#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 <net/icmp.h>
-#include <net/ip.h>
+#include <linux/kernel.h>
+#include <linux/skbuff.h>
+#include <linux/openvswitch.h>
+#include <linux/module.h>
#include <net/udp.h>
#include <net/ip_tunnels.h>
-#include <net/udp.h>
#include <net/rtnetlink.h>
-#include <net/route.h>
-#include <net/dsfield.h>
-#include <net/inet_ecn.h>
-#include <net/net_namespace.h>
-#include <net/netns/generic.h>
#include <net/vxlan.h>
#include "datapath.h"
#include "vport.h"
+#include "vport-netdev.h"
-/**
- * struct vxlan_port - Keeps track of open UDP ports
- * @vs: vxlan_sock created for the port.
- * @name: vport name.
- */
-struct vxlan_port {
- struct vxlan_sock *vs;
- char name[IFNAMSIZ];
-};
+static struct vport_ops ovs_vxlan_netdev_vport_ops;
-static inline struct vxlan_port *vxlan_vport(const struct vport *vport)
+static int vxlan_get_options(const struct vport *vport, struct sk_buff *skb)
{
- return vport_priv(vport);
-}
+ struct vxlan_dev *vxlan = netdev_priv(vport->dev);
+ __be16 dst_port = vxlan->cfg.dst_port;
-/* Called with rcu_read_lock and BH disabled. */
-static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb, __be32 vx_vni)
-{
- struct ovs_key_ipv4_tunnel tun_key;
- struct vport *vport = vs->data;
- struct iphdr *iph;
- __be64 key;
+ if (nla_put_u16(skb, OVS_TUNNEL_ATTR_DST_PORT, ntohs(dst_port)))
+ return -EMSGSIZE;
- /* Save outer tunnel values */
- iph = ip_hdr(skb);
- key = cpu_to_be64(ntohl(vx_vni) >> 8);
- ovs_flow_tun_key_init(&tun_key, iph, key, TUNNEL_KEY);
+ if (vxlan->flags & VXLAN_F_GBP) {
+ struct nlattr *exts;
- ovs_vport_receive(vport, skb, &tun_key);
-}
+ exts = nla_nest_start(skb, OVS_TUNNEL_ATTR_EXTENSION);
+ if (!exts)
+ return -EMSGSIZE;
-static int vxlan_get_options(const struct vport *vport, struct sk_buff *skb)
-{
- struct vxlan_port *vxlan_port = vxlan_vport(vport);
- __be16 dst_port = inet_sport(vxlan_port->vs->sock->sk);
+ if (vxlan->flags & VXLAN_F_GBP &&
+ nla_put_flag(skb, OVS_VXLAN_EXT_GBP))
+ return -EMSGSIZE;
+
+ nla_nest_end(skb, exts);
+ }
- if (nla_put_u16(skb, OVS_TUNNEL_ATTR_DST_PORT, ntohs(dst_port)))
- return -EMSGSIZE;
return 0;
}
-static void vxlan_tnl_destroy(struct vport *vport)
+static const struct nla_policy exts_policy[OVS_VXLAN_EXT_MAX + 1] = {
+ [OVS_VXLAN_EXT_GBP] = { .type = NLA_FLAG, },
+};
+
+static int vxlan_configure_exts(struct vport *vport, struct nlattr *attr,
+ struct vxlan_config *conf)
{
- struct vxlan_port *vxlan_port = vxlan_vport(vport);
+ struct nlattr *exts[OVS_VXLAN_EXT_MAX + 1];
+ int err;
+
+ if (nla_len(attr) < sizeof(struct nlattr))
+ return -EINVAL;
+
+ err = nla_parse_nested(exts, OVS_VXLAN_EXT_MAX, attr, exts_policy);
+ if (err < 0)
+ return err;
- vxlan_sock_release(vxlan_port->vs);
+ if (exts[OVS_VXLAN_EXT_GBP])
+ conf->flags |= VXLAN_F_GBP;
- ovs_vport_deferred_free(vport);
+ return 0;
}
static struct vport *vxlan_tnl_create(const struct vport_parms *parms)
{
struct net *net = ovs_dp_get_net(parms->dp);
struct nlattr *options = parms->options;
- struct vxlan_port *vxlan_port;
- struct vxlan_sock *vs;
+ struct net_device *dev;
struct vport *vport;
struct nlattr *a;
- u16 dst_port;
int err;
+ struct vxlan_config conf = {
+ .no_share = true,
+ .flags = VXLAN_F_COLLECT_METADATA,
+ /* Don't restrict the packets that can be sent by MTU */
+ .mtu = IP_MAX_MTU,
+ };
if (!options) {
err = -EINVAL;
goto error;
}
+
a = nla_find_nested(options, OVS_TUNNEL_ATTR_DST_PORT);
if (a && nla_len(a) == sizeof(u16)) {
- dst_port = nla_get_u16(a);
+ conf.dst_port = htons(nla_get_u16(a));
} else {
/* Require destination port from userspace. */
err = -EINVAL;
goto error;
}
- vport = ovs_vport_alloc(sizeof(struct vxlan_port),
- &ovs_vxlan_vport_ops, parms);
+ vport = ovs_vport_alloc(0, &ovs_vxlan_netdev_vport_ops, parms);
if (IS_ERR(vport))
return vport;
- vxlan_port = vxlan_vport(vport);
- strncpy(vxlan_port->name, parms->name, IFNAMSIZ);
+ a = nla_find_nested(options, OVS_TUNNEL_ATTR_EXTENSION);
+ if (a) {
+ err = vxlan_configure_exts(vport, a, &conf);
+ if (err) {
+ ovs_vport_free(vport);
+ goto error;
+ }
+ }
- vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true);
- if (IS_ERR(vs)) {
+ rtnl_lock();
+ dev = vxlan_dev_create(net, parms->name, NET_NAME_USER, &conf);
+ if (IS_ERR(dev)) {
+ rtnl_unlock();
ovs_vport_free(vport);
- return (void *)vs;
+ return ERR_CAST(dev);
}
- vxlan_port->vs = vs;
+ dev_change_flags(dev, dev->flags | IFF_UP);
+ rtnl_unlock();
return vport;
-
error:
return ERR_PTR(err);
}
-static int vxlan_tnl_send(struct vport *vport, struct sk_buff *skb)
+static struct vport *vxlan_create(const struct vport_parms *parms)
{
- struct vxlan_port *vxlan_port = vxlan_vport(vport);
- __be16 dst_port = inet_sport(vxlan_port->vs->sock->sk);
+ struct vport *vport;
+
+ vport = vxlan_tnl_create(parms);
+ if (IS_ERR(vport))
+ return vport;
+
+ return ovs_netdev_link(vport, parms->name);
+}
+
+static int vxlan_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
+ struct dp_upcall_info *upcall)
+{
+ struct vxlan_dev *vxlan = netdev_priv(vport->dev);
struct net *net = ovs_dp_get_net(vport->dp);
- struct rtable *rt;
+ __be16 dst_port = vxlan_dev_dst_port(vxlan);
__be16 src_port;
- __be32 saddr;
- __be16 df;
int port_min;
int port_max;
- int err;
- if (unlikely(!OVS_CB(skb)->tun_key)) {
- err = -EINVAL;
- goto error;
- }
-
- forward_ip_summed(skb, true);
-
- /* Route lookup */
- saddr = OVS_CB(skb)->tun_key->ipv4_src;
- rt = find_route(ovs_dp_get_net(vport->dp),
- &saddr,
- OVS_CB(skb)->tun_key->ipv4_dst,
- IPPROTO_UDP,
- OVS_CB(skb)->tun_key->ipv4_tos,
- skb->mark);
- if (IS_ERR(rt)) {
- err = PTR_ERR(rt);
- goto error;
- }
-
- df = OVS_CB(skb)->tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ?
- htons(IP_DF) : 0;
+ inet_get_local_port_range(net, &port_min, &port_max);
+ src_port = udp_flow_src_port(net, skb, 0, 0, true);
- skb->local_df = 1;
+ return ovs_tunnel_get_egress_info(upcall, net,
+ skb, IPPROTO_UDP,
+ src_port, dst_port);
+}
- inet_get_local_port_range(&port_min, &port_max);
- src_port = vxlan_src_port(port_min, port_max, skb);
+static struct vport_ops ovs_vxlan_netdev_vport_ops = {
+ .type = OVS_VPORT_TYPE_VXLAN,
+ .create = vxlan_create,
+ .destroy = ovs_netdev_tunnel_destroy,
+ .get_options = vxlan_get_options,
+ .send = vxlan_xmit,
+ .get_egress_tun_info = vxlan_get_egress_tun_info,
+};
- err = vxlan_xmit_skb(net, vxlan_port->vs, rt, skb,
- saddr, OVS_CB(skb)->tun_key->ipv4_dst,
- OVS_CB(skb)->tun_key->ipv4_tos,
- OVS_CB(skb)->tun_key->ipv4_ttl, df,
- src_port, dst_port,
- htonl(be64_to_cpu(OVS_CB(skb)->tun_key->tun_id) << 8));
- if (err < 0)
- ip_rt_put(rt);
-error:
- return err;
+static int __init ovs_vxlan_tnl_init(void)
+{
+ return ovs_vport_ops_register(&ovs_vxlan_netdev_vport_ops);
}
-static const char *vxlan_get_name(const struct vport *vport)
+static void __exit ovs_vxlan_tnl_exit(void)
{
- struct vxlan_port *vxlan_port = vxlan_vport(vport);
- return vxlan_port->name;
+ ovs_vport_ops_unregister(&ovs_vxlan_netdev_vport_ops);
}
-const struct vport_ops ovs_vxlan_vport_ops = {
- .type = OVS_VPORT_TYPE_VXLAN,
- .create = vxlan_tnl_create,
- .destroy = vxlan_tnl_destroy,
- .get_name = vxlan_get_name,
- .get_options = vxlan_get_options,
- .send = vxlan_tnl_send,
-};
+module_init(ovs_vxlan_tnl_init);
+module_exit(ovs_vxlan_tnl_exit);
+
+MODULE_DESCRIPTION("OVS: VXLAN switching port");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("vport-type-4");