openvswitch: Support VXLAN Group Policy extension

Introduces support for the group policy extension to the VXLAN virtual
port. The extension is disabled by default and only enabled if the user
has provided the respective configuration.

  ovs-vsctl add-port br0 vxlan0 -- \
     set Interface vxlan0 type=vxlan options:exts=gbp

The configuration interface to enable the extension is based on a new
attribute OVS_VXLAN_EXT_GBP nested inside OVS_TUNNEL_ATTR_EXTENSION
which can carry additional extensions as needed in the future.

The group policy metadata is stored as binary blob (struct ovs_vxlan_opts)
internally just like Geneve options but transported as nested Netlink
attributes to user space.

Renames the existing TUNNEL_OPTIONS_PRESENT to TUNNEL_GENEVE_OPT with the
binary value kept intact, a new flag TUNNEL_VXLAN_OPT is introduced.

The attributes OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS and existing
OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS are implemented mutually exclusive.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/ip_tunnels.h

@@ -97,7 +97,10 @@ struct ip_tunnel {
 #define TUNNEL_DONT_FRAGMENT    __cpu_to_be16(0x0100)
 #define TUNNEL_OAM		__cpu_to_be16(0x0200)
 #define TUNNEL_CRIT_OPT		__cpu_to_be16(0x0400)
-#define TUNNEL_OPTIONS_PRESENT	__cpu_to_be16(0x0800)
+#define TUNNEL_GENEVE_OPT	__cpu_to_be16(0x0800)
+#define TUNNEL_VXLAN_OPT	__cpu_to_be16(0x1000)
+
+#define TUNNEL_OPTIONS_PRESENT	(TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT)
 
 struct tnl_ptk_info {
 	__be16 flags;

include/uapi/linux/openvswitch.h

@@ -252,11 +252,21 @@ enum ovs_vport_attr {
 
 #define OVS_VPORT_ATTR_MAX (__OVS_VPORT_ATTR_MAX - 1)
 
+enum {
+	OVS_VXLAN_EXT_UNSPEC,
+	OVS_VXLAN_EXT_GBP,	/* Flag or __u32 */
+	__OVS_VXLAN_EXT_MAX,
+};
+
+#define OVS_VXLAN_EXT_MAX (__OVS_VXLAN_EXT_MAX - 1)
+
+
 /* OVS_VPORT_ATTR_OPTIONS attributes for tunnels.
  */
 enum {
 	OVS_TUNNEL_ATTR_UNSPEC,
 	OVS_TUNNEL_ATTR_DST_PORT, /* 16-bit UDP port, used by L4 tunnels. */
+	OVS_TUNNEL_ATTR_EXTENSION,
 	__OVS_TUNNEL_ATTR_MAX
 };
 
@@ -328,6 +338,7 @@ enum ovs_tunnel_key_attr {
 	OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS,        /* Array of Geneve options. */
 	OVS_TUNNEL_KEY_ATTR_TP_SRC,		/* be16 src Transport Port. */
 	OVS_TUNNEL_KEY_ATTR_TP_DST,		/* be16 dst Transport Port. */
+	OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS,		/* Nested OVS_VXLAN_EXT_* */
 	__OVS_TUNNEL_KEY_ATTR_MAX
 };
 

net/openvswitch/flow_netlink.c

@@ -49,6 +49,7 @@
 #include <net/mpls.h>
 
 #include "flow_netlink.h"
+#include "vport-vxlan.h"
 
 struct ovs_len_tbl {
 	int len;
@@ -268,6 +269,9 @@ size_t ovs_tun_key_attr_size(void)
 		+ nla_total_size(0)    /* OVS_TUNNEL_KEY_ATTR_CSUM */
 		+ nla_total_size(0)    /* OVS_TUNNEL_KEY_ATTR_OAM */
 		+ nla_total_size(256)  /* OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS */
+		/* OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS is mutually exclusive with
+		 * OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS and covered by it.
+		 */
 		+ nla_total_size(2)    /* OVS_TUNNEL_KEY_ATTR_TP_SRC */
 		+ nla_total_size(2);   /* OVS_TUNNEL_KEY_ATTR_TP_DST */
 }
@@ -308,6 +312,7 @@ static const struct ovs_len_tbl ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1]
 	[OVS_TUNNEL_KEY_ATTR_TP_DST]	    = { .len = sizeof(u16) },
 	[OVS_TUNNEL_KEY_ATTR_OAM]	    = { .len = 0 },
 	[OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS]   = { .len = OVS_ATTR_NESTED },
+	[OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS]    = { .len = OVS_ATTR_NESTED },
 };
 
 /* The size of the argument for each %OVS_KEY_ATTR_* Netlink attribute.  */
@@ -460,6 +465,41 @@ static int genev_tun_opt_from_nlattr(const struct nlattr *a,
 	return 0;
 }
 
+static const struct nla_policy vxlan_opt_policy[OVS_VXLAN_EXT_MAX + 1] = {
+	[OVS_VXLAN_EXT_GBP]	= { .type = NLA_U32 },
+};
+
+static int vxlan_tun_opt_from_nlattr(const struct nlattr *a,
+				     struct sw_flow_match *match, bool is_mask,
+				     bool log)
+{
+	struct nlattr *tb[OVS_VXLAN_EXT_MAX+1];
+	unsigned long opt_key_offset;
+	struct ovs_vxlan_opts opts;
+	int err;
+
+	BUILD_BUG_ON(sizeof(opts) > sizeof(match->key->tun_opts));
+
+	err = nla_parse_nested(tb, OVS_VXLAN_EXT_MAX, a, vxlan_opt_policy);
+	if (err < 0)
+		return err;
+
+	memset(&opts, 0, sizeof(opts));
+
+	if (tb[OVS_VXLAN_EXT_GBP])
+		opts.gbp = nla_get_u32(tb[OVS_VXLAN_EXT_GBP]);
+
+	if (!is_mask)
+		SW_FLOW_KEY_PUT(match, tun_opts_len, sizeof(opts), false);
+	else
+		SW_FLOW_KEY_PUT(match, tun_opts_len, 0xff, true);
+
+	opt_key_offset = TUN_METADATA_OFFSET(sizeof(opts));
+	SW_FLOW_KEY_MEMCPY_OFFSET(match, opt_key_offset, &opts, sizeof(opts),
+				  is_mask);
+	return 0;
+}
+
 static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 				struct sw_flow_match *match, bool is_mask,
 				bool log)
@@ -468,6 +508,7 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 	int rem;
 	bool ttl = false;
 	__be16 tun_flags = 0;
+	int opts_type = 0;
 
 	nla_for_each_nested(a, attr, rem) {
 		int type = nla_type(a);
@@ -527,11 +568,30 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 			tun_flags |= TUNNEL_OAM;
 			break;
 		case OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS:
+			if (opts_type) {
+				OVS_NLERR(log, "Multiple metadata blocks provided");
+				return -EINVAL;
+			}
+
 			err = genev_tun_opt_from_nlattr(a, match, is_mask, log);
 			if (err)
 				return err;
 
-			tun_flags |= TUNNEL_OPTIONS_PRESENT;
+			tun_flags |= TUNNEL_GENEVE_OPT;
+			opts_type = type;
+			break;
+		case OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS:
+			if (opts_type) {
+				OVS_NLERR(log, "Multiple metadata blocks provided");
+				return -EINVAL;
+			}
+
+			err = vxlan_tun_opt_from_nlattr(a, match, is_mask, log);
+			if (err)
+				return err;
+
+			tun_flags |= TUNNEL_VXLAN_OPT;
+			opts_type = type;
 			break;
 		default:
 			OVS_NLERR(log, "Unknown IPv4 tunnel attribute %d",
@@ -560,6 +620,23 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 		}
 	}
 
+	return opts_type;
+}
+
+static int vxlan_opt_to_nlattr(struct sk_buff *skb,
+			       const void *tun_opts, int swkey_tun_opts_len)
+{
+	const struct ovs_vxlan_opts *opts = tun_opts;
+	struct nlattr *nla;
+
+	nla = nla_nest_start(skb, OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS);
+	if (!nla)
+		return -EMSGSIZE;
+
+	if (nla_put_u32(skb, OVS_VXLAN_EXT_GBP, opts->gbp) < 0)
+		return -EMSGSIZE;
+
+	nla_nest_end(skb, nla);
 	return 0;
 }
 
@@ -596,10 +673,15 @@ static int __ipv4_tun_to_nlattr(struct sk_buff *skb,
 	if ((output->tun_flags & TUNNEL_OAM) &&
 	    nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_OAM))
 		return -EMSGSIZE;
-	if (tun_opts &&
-	    nla_put(skb, OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS,
-		    swkey_tun_opts_len, tun_opts))
-		return -EMSGSIZE;
+	if (tun_opts) {
+		if (output->tun_flags & TUNNEL_GENEVE_OPT &&
+		    nla_put(skb, OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS,
+			    swkey_tun_opts_len, tun_opts))
+			return -EMSGSIZE;
+		else if (output->tun_flags & TUNNEL_VXLAN_OPT &&
+			 vxlan_opt_to_nlattr(skb, tun_opts, swkey_tun_opts_len))
+			return -EMSGSIZE;
+	}
 
 	return 0;
 }
@@ -680,7 +762,7 @@ static int metadata_from_nlattrs(struct sw_flow_match *match,  u64 *attrs,
 	}
 	if (*attrs & (1 << OVS_KEY_ATTR_TUNNEL)) {
 		if (ipv4_tun_from_nlattr(a[OVS_KEY_ATTR_TUNNEL], match,
-					 is_mask, log))
+					 is_mask, log) < 0)
 			return -EINVAL;
 		*attrs &= ~(1 << OVS_KEY_ATTR_TUNNEL);
 	}
@@ -1578,17 +1660,23 @@ static int validate_and_copy_set_tun(const struct nlattr *attr,
 	struct sw_flow_key key;
 	struct ovs_tunnel_info *tun_info;
 	struct nlattr *a;
-	int err, start;
+	int err, start, opts_type;
 
 	ovs_match_init(&match, &key, NULL);
-	err = ipv4_tun_from_nlattr(nla_data(attr), &match, false, log);
-	if (err)
-		return err;
+	opts_type = ipv4_tun_from_nlattr(nla_data(attr), &match, false, log);
+	if (opts_type < 0)
+		return opts_type;
 
 	if (key.tun_opts_len) {
-		err = validate_geneve_opts(&key);
-		if (err < 0)
-			return err;
+		switch (opts_type) {
+		case OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS:
+			err = validate_geneve_opts(&key);
+			if (err < 0)
+				return err;
+			break;
+		case OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS:
+			break;
+		}
 	};
 
 	start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SET, log);

net/openvswitch/vport-geneve.c

@@ -88,7 +88,7 @@ static void geneve_rcv(struct geneve_sock *gs, struct sk_buff *skb)
 
 	opts_len = geneveh->opt_len * 4;
 
-	flags = TUNNEL_KEY | TUNNEL_OPTIONS_PRESENT |
+	flags = TUNNEL_KEY | TUNNEL_GENEVE_OPT |
 		(udp_hdr(skb)->check != 0 ? TUNNEL_CSUM : 0) |
 		(geneveh->oam ? TUNNEL_OAM : 0) |
 		(geneveh->critical ? TUNNEL_CRIT_OPT : 0);
@@ -178,7 +178,7 @@ static int geneve_tnl_send(struct vport *vport, struct sk_buff *skb)
 	__be16 sport;
 	struct rtable *rt;
 	struct flowi4 fl;
-	u8 vni[3];
+	u8 vni[3], opts_len, *opts;
 	__be16 df;
 	int err;
 
@@ -200,11 +200,18 @@ static int geneve_tnl_send(struct vport *vport, struct sk_buff *skb)
 	tunnel_id_to_vni(tun_key->tun_id, vni);
 	skb->ignore_df = 1;
 
+	if (tun_key->tun_flags & TUNNEL_GENEVE_OPT) {
+		opts = (u8 *)tun_info->options;
+		opts_len = tun_info->options_len;
+	} else {
+		opts = NULL;
+		opts_len = 0;
+	}
+
 	err = geneve_xmit_skb(geneve_port->gs, rt, skb, fl.saddr,
 			      tun_key->ipv4_dst, tun_key->ipv4_tos,
 			      tun_key->ipv4_ttl, df, sport, dport,
-			      tun_key->tun_flags, vni,
-			      tun_info->options_len, (u8 *)tun_info->options,
+			      tun_key->tun_flags, vni, opts_len, opts,
 			      false);
 	if (err < 0)
 		ip_rt_put(rt);

net/openvswitch/vport-vxlan.c

@@ -40,6 +40,7 @@
 
 #include "datapath.h"
 #include "vport.h"
+#include "vport-vxlan.h"
 
 /**
  * struct vxlan_port - Keeps track of open UDP ports
@@ -49,6 +50,7 @@
 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;
@@ -63,16 +65,26 @@ 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;
+	vxlan_port = vxlan_vport(vport);
+	if (vxlan_port->exts & VXLAN_F_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, TUNNEL_KEY, NULL, 0);
+			       key, flags, &opts, sizeof(opts));
 
 	ovs_vport_receive(vport, skb, &tun_info);
 }
@@ -84,6 +96,21 @@ static int vxlan_get_options(const struct vport *vport, struct sk_buff *skb)
 
 	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;
 }
 
@@ -96,6 +123,31 @@ static void vxlan_tnl_destroy(struct vport *vport)
 	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);
@@ -128,7 +180,17 @@ static struct vport *vxlan_tnl_create(const struct vport_parms *parms)
 	vxlan_port = vxlan_vport(vport);
 	strncpy(vxlan_port->name, parms->name, IFNAMSIZ);
 
-	vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true, 0);
+	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;
@@ -141,6 +203,21 @@ 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);
@@ -173,6 +250,7 @@ static int vxlan_tnl_send(struct vport *vport, struct sk_buff *skb)
 
 	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);
 
 	err = vxlan_xmit_skb(vxlan_port->vs, rt, skb,
 			     fl.saddr, tun_key->ipv4_dst,

net/openvswitch/vport-vxlan.h

@@ -0,0 +1,11 @@
+#ifndef VPORT_VXLAN_H
+#define VPORT_VXLAN_H 1
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+
+struct ovs_vxlan_opts {
+	__u32 gbp;
+};
+
+#endif