[linux-2.6-block.git] / net / openvswitch / vport-vxlan.c

/*
 * Copyright (c) 2014 Nicira, Inc.
 * Copyright (c) 2013 Cisco Systems, 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
 */

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/in.h>
#include <linux/ip.h>
#include <linux/net.h>
#include <linux/rculist.h>
#include <linux/udp.h>
#include <linux/module.h>

#include <net/icmp.h>
#include <net/ip.h>
#include <net/udp.h>
#include <net/ip_tunnels.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-vxlan.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];
	u32 exts; /* VXLAN_F_* in <net/vxlan.h> */
};

static struct vport_ops ovs_vxlan_vport_ops;

static inline struct vxlan_port *vxlan_vport(const struct vport *vport)
{
	return vport_priv(vport);
}

/* Called with rcu_read_lock and BH disabled. */
static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
		      struct vxlan_metadata *md)
{
	struct ovs_tunnel_info tun_info;
	struct vxlan_port *vxlan_port;
	struct vport *vport = vs->data;
	struct iphdr *iph;
	struct ovs_vxlan_opts opts = {
		.gbp = md->gbp,
	};
	__be64 key;
	__be16 flags;

	flags = TUNNEL_KEY | (udp_hdr(skb)->check != 0 ? TUNNEL_CSUM : 0);
	vxlan_port = vxlan_vport(vport);
	if (vxlan_port->exts & VXLAN_F_GBP && md->gbp)
		flags |= TUNNEL_VXLAN_OPT;

	/* Save outer tunnel values */
	iph = ip_hdr(skb);
	key = cpu_to_be64(ntohl(md->vni) >> 8);
	ovs_flow_tun_info_init(&tun_info, iph,
			       udp_hdr(skb)->source, udp_hdr(skb)->dest,
			       key, flags, &opts, sizeof(opts));

	ovs_vport_receive(vport, skb, &tun_info);
}

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_sk(vxlan_port->vs->sock->sk)->inet_sport;

	if (nla_put_u16(skb, OVS_TUNNEL_ATTR_DST_PORT, ntohs(dst_port)))
		return -EMSGSIZE;

	if (vxlan_port->exts) {
		struct nlattr *exts;

		exts = nla_nest_start(skb, OVS_TUNNEL_ATTR_EXTENSION);
		if (!exts)
			return -EMSGSIZE;

		if (vxlan_port->exts & VXLAN_F_GBP &&
		    nla_put_flag(skb, OVS_VXLAN_EXT_GBP))
			return -EMSGSIZE;

		nla_nest_end(skb, exts);
	}

	return 0;
}

static void vxlan_tnl_destroy(struct vport *vport)
{
	struct vxlan_port *vxlan_port = vxlan_vport(vport);

	vxlan_sock_release(vxlan_port->vs);

	ovs_vport_deferred_free(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 nlattr *exts[OVS_VXLAN_EXT_MAX+1];
	struct vxlan_port *vxlan_port;
	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_port = vxlan_vport(vport);

	if (exts[OVS_VXLAN_EXT_GBP])
		vxlan_port->exts |= VXLAN_F_GBP;

	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 vport *vport;
	struct nlattr *a;
	u16 dst_port;
	int err;

	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);
	} else {
		/* Require destination port from userspace. */
		err = -EINVAL;
		goto error;
	}

	vport = ovs_vport_alloc(sizeof(struct vxlan_port),
				&ovs_vxlan_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);
		if (err) {
			ovs_vport_free(vport);
			goto error;
		}
	}

	vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true,
			    vxlan_port->exts);
	if (IS_ERR(vs)) {
		ovs_vport_free(vport);
		return (void *)vs;
	}
	vxlan_port->vs = vs;

	return vport;

error:
	return ERR_PTR(err);
}

static int vxlan_ext_gbp(struct sk_buff *skb)
{
	const struct ovs_tunnel_info *tun_info;
	const struct ovs_vxlan_opts *opts;

	tun_info = OVS_CB(skb)->egress_tun_info;
	opts = tun_info->options;

	if (tun_info->tunnel.tun_flags & TUNNEL_VXLAN_OPT &&
	    tun_info->options_len >= sizeof(*opts))
		return opts->gbp;
	else
		return 0;
}

static int vxlan_tnl_send(struct vport *vport, struct sk_buff *skb)
{
	struct net *net = ovs_dp_get_net(vport->dp);
	struct vxlan_port *vxlan_port = vxlan_vport(vport);
	struct sock *sk = vxlan_port->vs->sock->sk;
	__be16 dst_port = inet_sk(sk)->inet_sport;
	const struct ovs_key_ipv4_tunnel *tun_key;
	struct vxlan_metadata md = {0};
	struct rtable *rt;
	struct flowi4 fl;
	__be16 src_port;
	__be16 df;
	int err;
	u32 vxflags;

	if (unlikely(!OVS_CB(skb)->egress_tun_info)) {
		err = -EINVAL;
		goto error;
	}

	tun_key = &OVS_CB(skb)->egress_tun_info->tunnel;
	rt = ovs_tunnel_route_lookup(net, tun_key, skb->mark, &fl, IPPROTO_UDP);
	if (IS_ERR(rt)) {
		err = PTR_ERR(rt);
		goto error;
	}

	df = tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ?
		htons(IP_DF) : 0;

	skb->ignore_df = 1;

	src_port = udp_flow_src_port(net, skb, 0, 0, true);
	md.vni = htonl(be64_to_cpu(tun_key->tun_id) << 8);
	md.gbp = vxlan_ext_gbp(skb);
	vxflags = vxlan_port->exts |
		      (tun_key->tun_flags & TUNNEL_CSUM ? VXLAN_F_UDP_CSUM : 0);

	err = vxlan_xmit_skb(rt, sk, skb, fl.saddr, tun_key->ipv4_dst,
			     tun_key->ipv4_tos, tun_key->ipv4_ttl, df,
			     src_port, dst_port,
			     &md, false, vxflags);
	if (err < 0)
		ip_rt_put(rt);
	return err;
error:
	kfree_skb(skb);
	return err;
}

static int vxlan_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 vxlan_port *vxlan_port = vxlan_vport(vport);
	__be16 dst_port = inet_sk(vxlan_port->vs->sock->sk)->inet_sport;
	__be16 src_port;
	int port_min;
	int port_max;

	inet_get_local_port_range(net, &port_min, &port_max);
	src_port = udp_flow_src_port(net, skb, 0, 0, true);

	return ovs_tunnel_get_egress_info(egress_tun_info, net,
					  OVS_CB(skb)->egress_tun_info,
					  IPPROTO_UDP, skb->mark,
					  src_port, dst_port);
}

static const char *vxlan_get_name(const struct vport *vport)
{
	struct vxlan_port *vxlan_port = vxlan_vport(vport);
	return vxlan_port->name;
}

static 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,
	.get_egress_tun_info	= vxlan_get_egress_tun_info,
	.owner		= THIS_MODULE,
};

static int __init ovs_vxlan_tnl_init(void)
{
	return ovs_vport_ops_register(&ovs_vxlan_vport_ops);
}

static void __exit ovs_vxlan_tnl_exit(void)
{
	ovs_vport_ops_unregister(&ovs_vxlan_vport_ops);
}

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");
Commit	Line	Data
58264848	1	/*
8c8b1b83	2	* Copyright (c) 2014 Nicira, Inc.
58264848 PS	3	* Copyright (c) 2013 Cisco Systems, Inc.
	4	*
	5	* This program is free software; you can redistribute it and/or
	6	* modify it under the terms of version 2 of the GNU General Public
	7	* License as published by the Free Software Foundation.
	8	*
	9	* This program is distributed in the hope that it will be useful, but
	10	* WITHOUT ANY WARRANTY; without even the implied warranty of
	11	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	12	* General Public License for more details.
	13	*
	14	* You should have received a copy of the GNU General Public License
	15	* along with this program; if not, write to the Free Software
	16	* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
	17	* 02110-1301, USA
	18	*/
	19
	20	#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
	21
	22	#include <linux/in.h>
	23	#include <linux/ip.h>
	24	#include <linux/net.h>
	25	#include <linux/rculist.h>
	26	#include <linux/udp.h>
62b9c8d0	27	#include <linux/module.h>
58264848 PS	28
	29	#include <net/icmp.h>
	30	#include <net/ip.h>
	31	#include <net/udp.h>
	32	#include <net/ip_tunnels.h>
58264848 PS	33	#include <net/rtnetlink.h>
	34	#include <net/route.h>
	35	#include <net/dsfield.h>
	36	#include <net/inet_ecn.h>
	37	#include <net/net_namespace.h>
	38	#include <net/netns/generic.h>
	39	#include <net/vxlan.h>
	40
	41	#include "datapath.h"
	42	#include "vport.h"
1dd144cf	43	#include "vport-vxlan.h"
58264848 PS	44
	45	/**
	46	* struct vxlan_port - Keeps track of open UDP ports
	47	* @vs: vxlan_sock created for the port.
	48	* @name: vport name.
	49	*/
	50	struct vxlan_port {
	51	struct vxlan_sock *vs;
	52	char name[IFNAMSIZ];
1dd144cf	53	u32 exts; /* VXLAN_F_* in <net/vxlan.h> */
58264848 PS	54	};
58264848 PS	55
62b9c8d0 TG	56	static struct vport_ops ovs_vxlan_vport_ops;
62b9c8d0 TG	57
58264848 PS	58	static inline struct vxlan_port vxlan_vport(const struct vport vport)
	59	{
	60	return vport_priv(vport);
	61	}
	62
	63	/* Called with rcu_read_lock and BH disabled. */
3511494c TG	64	static void vxlan_rcv(struct vxlan_sock vs, struct sk_buff skb,
3511494c TG	65	struct vxlan_metadata *md)
58264848	66	{
f0b128c1	67	struct ovs_tunnel_info tun_info;
1dd144cf	68	struct vxlan_port *vxlan_port;
58264848 PS	69	struct vport *vport = vs->data;
58264848 PS	70	struct iphdr *iph;
1dd144cf TG	71	struct ovs_vxlan_opts opts = {
	72	.gbp = md->gbp,
	73	};
58264848	74	__be64 key;
1dd144cf TG	75	__be16 flags;
1dd144cf TG	76
b8693877	77	flags = TUNNEL_KEY \| (udp_hdr(skb)->check != 0 ? TUNNEL_CSUM : 0);
1dd144cf	78	vxlan_port = vxlan_vport(vport);
fd3137cd	79	if (vxlan_port->exts & VXLAN_F_GBP && md->gbp)
1dd144cf	80	flags \|= TUNNEL_VXLAN_OPT;
58264848 PS	81
	82	/* Save outer tunnel values */
	83	iph = ip_hdr(skb);
3511494c	84	key = cpu_to_be64(ntohl(md->vni) >> 8);
8f0aad6f WZ	85	ovs_flow_tun_info_init(&tun_info, iph,
8f0aad6f WZ	86	udp_hdr(skb)->source, udp_hdr(skb)->dest,
1dd144cf	87	key, flags, &opts, sizeof(opts));
58264848	88
f0b128c1	89	ovs_vport_receive(vport, skb, &tun_info);
58264848 PS	90	}
	91
	92	static int vxlan_get_options(const struct vport vport, struct sk_buff skb)
	93	{
	94	struct vxlan_port *vxlan_port = vxlan_vport(vport);
	95	__be16 dst_port = inet_sk(vxlan_port->vs->sock->sk)->inet_sport;
	96
	97	if (nla_put_u16(skb, OVS_TUNNEL_ATTR_DST_PORT, ntohs(dst_port)))
	98	return -EMSGSIZE;
1dd144cf TG	99
	100	if (vxlan_port->exts) {
	101	struct nlattr *exts;
	102
	103	exts = nla_nest_start(skb, OVS_TUNNEL_ATTR_EXTENSION);
	104	if (!exts)
	105	return -EMSGSIZE;
	106
	107	if (vxlan_port->exts & VXLAN_F_GBP &&
	108	nla_put_flag(skb, OVS_VXLAN_EXT_GBP))
	109	return -EMSGSIZE;
	110
	111	nla_nest_end(skb, exts);
	112	}
	113
58264848 PS	114	return 0;
	115	}
	116
	117	static void vxlan_tnl_destroy(struct vport *vport)
	118	{
	119	struct vxlan_port *vxlan_port = vxlan_vport(vport);
	120
	121	vxlan_sock_release(vxlan_port->vs);
	122
	123	ovs_vport_deferred_free(vport);
	124	}
	125
1dd144cf TG	126	static const struct nla_policy exts_policy[OVS_VXLAN_EXT_MAX+1] = {
	127	[OVS_VXLAN_EXT_GBP] = { .type = NLA_FLAG, },
	128	};
	129
	130	static int vxlan_configure_exts(struct vport vport, struct nlattr attr)
	131	{
	132	struct nlattr *exts[OVS_VXLAN_EXT_MAX+1];
	133	struct vxlan_port *vxlan_port;
	134	int err;
	135
	136	if (nla_len(attr) < sizeof(struct nlattr))
	137	return -EINVAL;
	138
	139	err = nla_parse_nested(exts, OVS_VXLAN_EXT_MAX, attr, exts_policy);
	140	if (err < 0)
	141	return err;
	142
	143	vxlan_port = vxlan_vport(vport);
	144
	145	if (exts[OVS_VXLAN_EXT_GBP])
	146	vxlan_port->exts \|= VXLAN_F_GBP;
	147
	148	return 0;
	149	}
	150
58264848 PS	151	static struct vport vxlan_tnl_create(const struct vport_parms parms)
	152	{
	153	struct net *net = ovs_dp_get_net(parms->dp);
	154	struct nlattr *options = parms->options;
	155	struct vxlan_port *vxlan_port;
	156	struct vxlan_sock *vs;
	157	struct vport *vport;
	158	struct nlattr *a;
	159	u16 dst_port;
	160	int err;
	161
	162	if (!options) {
	163	err = -EINVAL;
	164	goto error;
	165	}
	166	a = nla_find_nested(options, OVS_TUNNEL_ATTR_DST_PORT);
	167	if (a && nla_len(a) == sizeof(u16)) {
	168	dst_port = nla_get_u16(a);
	169	} else {
	170	/* Require destination port from userspace. */
	171	err = -EINVAL;
	172	goto error;
	173	}
	174
	175	vport = ovs_vport_alloc(sizeof(struct vxlan_port),
	176	&ovs_vxlan_vport_ops, parms);
	177	if (IS_ERR(vport))
	178	return vport;
	179
	180	vxlan_port = vxlan_vport(vport);
	181	strncpy(vxlan_port->name, parms->name, IFNAMSIZ);
	182
1dd144cf TG	183	a = nla_find_nested(options, OVS_TUNNEL_ATTR_EXTENSION);
	184	if (a) {
	185	err = vxlan_configure_exts(vport, a);
	186	if (err) {
	187	ovs_vport_free(vport);
	188	goto error;
	189	}
	190	}
	191
	192	vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true,
	193	vxlan_port->exts);
58264848 PS	194	if (IS_ERR(vs)) {
	195	ovs_vport_free(vport);
	196	return (void *)vs;
	197	}
	198	vxlan_port->vs = vs;
	199
	200	return vport;
	201
	202	error:
	203	return ERR_PTR(err);
	204	}
	205
1dd144cf TG	206	static int vxlan_ext_gbp(struct sk_buff *skb)
	207	{
	208	const struct ovs_tunnel_info *tun_info;
	209	const struct ovs_vxlan_opts *opts;
	210
	211	tun_info = OVS_CB(skb)->egress_tun_info;
	212	opts = tun_info->options;
	213
	214	if (tun_info->tunnel.tun_flags & TUNNEL_VXLAN_OPT &&
	215	tun_info->options_len >= sizeof(*opts))
	216	return opts->gbp;
	217	else
	218	return 0;
	219	}
	220
58264848 PS	221	static int vxlan_tnl_send(struct vport vport, struct sk_buff skb)
	222	{
	223	struct net *net = ovs_dp_get_net(vport->dp);
	224	struct vxlan_port *vxlan_port = vxlan_vport(vport);
79b16aad DM	225	struct sock *sk = vxlan_port->vs->sock->sk;
79b16aad DM	226	__be16 dst_port = inet_sk(sk)->inet_sport;
3f4c1d87	227	const struct ovs_key_ipv4_tunnel *tun_key;
3511494c	228	struct vxlan_metadata md = {0};
58264848 PS	229	struct rtable *rt;
	230	struct flowi4 fl;
	231	__be16 src_port;
58264848 PS	232	__be16 df;
58264848 PS	233	int err;
b8693877	234	u32 vxflags;
58264848	235
f0b128c1	236	if (unlikely(!OVS_CB(skb)->egress_tun_info)) {
58264848 PS	237	err = -EINVAL;
	238	goto error;
	239	}
	240
f0b128c1	241	tun_key = &OVS_CB(skb)->egress_tun_info->tunnel;
3f4c1d87	242	rt = ovs_tunnel_route_lookup(net, tun_key, skb->mark, &fl, IPPROTO_UDP);
58264848 PS	243	if (IS_ERR(rt)) {
	244	err = PTR_ERR(rt);
	245	goto error;
	246	}
	247
8c8b1b83	248	df = tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ?
58264848 PS	249	htons(IP_DF) : 0;
58264848 PS	250
60ff7467	251	skb->ignore_df = 1;
58264848	252
535fb8d0	253	src_port = udp_flow_src_port(net, skb, 0, 0, true);
3511494c	254	md.vni = htonl(be64_to_cpu(tun_key->tun_id) << 8);
1dd144cf	255	md.gbp = vxlan_ext_gbp(skb);
b8693877 JG	256	vxflags = vxlan_port->exts \|
b8693877 JG	257	(tun_key->tun_flags & TUNNEL_CSUM ? VXLAN_F_UDP_CSUM : 0);
58264848	258
79b16aad	259	err = vxlan_xmit_skb(rt, sk, skb, fl.saddr, tun_key->ipv4_dst,
8c8b1b83	260	tun_key->ipv4_tos, tun_key->ipv4_ttl, df,
58264848	261	src_port, dst_port,
b8693877	262	&md, false, vxflags);
58264848 PS	263	if (err < 0)
58264848 PS	264	ip_rt_put(rt);
997e068e	265	return err;
58264848	266	error:
997e068e	267	kfree_skb(skb);
58264848 PS	268	return err;
	269	}
	270
8f0aad6f WZ	271	static int vxlan_get_egress_tun_info(struct vport vport, struct sk_buff skb,
	272	struct ovs_tunnel_info *egress_tun_info)
	273	{
	274	struct net *net = ovs_dp_get_net(vport->dp);
	275	struct vxlan_port *vxlan_port = vxlan_vport(vport);
	276	__be16 dst_port = inet_sk(vxlan_port->vs->sock->sk)->inet_sport;
	277	__be16 src_port;
	278	int port_min;
	279	int port_max;
	280
	281	inet_get_local_port_range(net, &port_min, &port_max);
	282	src_port = udp_flow_src_port(net, skb, 0, 0, true);
	283
	284	return ovs_tunnel_get_egress_info(egress_tun_info, net,
	285	OVS_CB(skb)->egress_tun_info,
	286	IPPROTO_UDP, skb->mark,
	287	src_port, dst_port);
	288	}
	289
58264848 PS	290	static const char vxlan_get_name(const struct vport vport)
	291	{
	292	struct vxlan_port *vxlan_port = vxlan_vport(vport);
	293	return vxlan_port->name;
	294	}
	295
62b9c8d0	296	static struct vport_ops ovs_vxlan_vport_ops = {
58264848 PS	297	.type = OVS_VPORT_TYPE_VXLAN,
	298	.create = vxlan_tnl_create,
	299	.destroy = vxlan_tnl_destroy,
	300	.get_name = vxlan_get_name,
	301	.get_options = vxlan_get_options,
	302	.send = vxlan_tnl_send,
8f0aad6f	303	.get_egress_tun_info = vxlan_get_egress_tun_info,
62b9c8d0	304	.owner = THIS_MODULE,
58264848	305	};
62b9c8d0 TG	306
	307	static int __init ovs_vxlan_tnl_init(void)
	308	{
	309	return ovs_vport_ops_register(&ovs_vxlan_vport_ops);
	310	}
	311
	312	static void __exit ovs_vxlan_tnl_exit(void)
	313	{
	314	ovs_vport_ops_unregister(&ovs_vxlan_vport_ops);
	315	}
	316
	317	module_init(ovs_vxlan_tnl_init);
	318	module_exit(ovs_vxlan_tnl_exit);
	319
	320	MODULE_DESCRIPTION("OVS: VXLAN switching port");
	321	MODULE_LICENSE("GPL");
	322	MODULE_ALIAS("vport-type-4");