[PATCH 6/6] openvswitch: Support VXLAN Group Policy extension

* [PATCH 6/6] openvswitch: Support VXLAN Group Policy extension
  2015-01-07  2:05 [PATCH 0/6 net-next] VXLAN Group Policy Extension Thomas Graf
@ 2015-01-07  2:05 ` Thomas Graf
  2015-01-07 22:46   ` Jesse Gross
  0 siblings, 1 reply; 26+ messages in thread
From: Thomas Graf @ 2015-01-07  2:05 UTC (permalink / raw)
  To: davem, jesse, stephen, pshelar; +Cc: netdev, dev

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 handled in the same way as Geneve options
and transported as binary blob in a new Netlink attribute
OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS which is mutually exclusive to the
existing OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
---
 include/uapi/linux/openvswitch.h | 19 ++++++++++
 net/openvswitch/flow_netlink.c   | 78 +++++++++++++++++++++++++--------------
 net/openvswitch/vport-vxlan.c    | 80 +++++++++++++++++++++++++++++++++++++++-
 3 files changed, 148 insertions(+), 29 deletions(-)

diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index 3a6dcaa..676a89e 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -248,11 +248,29 @@ enum ovs_vport_attr {
 
 #define OVS_VPORT_ATTR_MAX (__OVS_VPORT_ATTR_MAX - 1)
 
+/**
+ * struct ovs_vxlan_opts - VXLAN tunnel options
+ * @gbp: Group policy bits
+ */
+struct ovs_vxlan_opts {
+	__u32 gbp;
+};
+
+enum {
+	OVS_VXLAN_EXT_UNSPEC,
+	OVS_VXLAN_EXT_GBP,
+	__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
 };
 
@@ -324,6 +342,7 @@ 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,		/* struct ovs_vxlan_opts. */
 	__OVS_TUNNEL_KEY_ATTR_MAX
 };
 
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index c60ae3f..1528709 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -446,6 +446,7 @@ 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);
@@ -463,6 +464,7 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 			[OVS_TUNNEL_KEY_ATTR_TP_DST] = sizeof(u16),
 			[OVS_TUNNEL_KEY_ATTR_OAM] = 0,
 			[OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS] = -1,
+			[OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS] = -1,
 		};
 
 		if (type > OVS_TUNNEL_KEY_ATTR_MAX) {
@@ -519,11 +521,18 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 			tun_flags |= TUNNEL_OAM;
 			break;
 		case OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS:
+		case OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS:
+			if (opts_type) {
+				OVS_NLERR(log, "Multiple metadata blocks provided");
+				return -EINVAL;
+			}
+
 			err = tun_md_opt_from_nlattr(a, match, is_mask, log);
 			if (err)
 				return err;
 
 			tun_flags |= TUNNEL_OPTIONS_PRESENT;
+			opts_type = type;
 			break;
 		default:
 			OVS_NLERR(log, "Unknown IPv4 tunnel attribute %d",
@@ -552,7 +561,7 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr,
 		}
 	}
 
-	return 0;
+	return opts_type;
 }
 
 static int __ipv4_tun_to_nlattr(struct sk_buff *skb,
@@ -1537,6 +1546,34 @@ void ovs_match_init(struct sw_flow_match *match,
 	}
 }
 
+static int validate_and_copy_geneve_opts(struct sw_flow_key *key)
+{
+	struct geneve_opt *option;
+	int opts_len = key->tun_opts_len;
+	bool crit_opt = false;
+
+	option = (struct geneve_opt *) TUN_METADATA_OPTS(key, key->tun_opts_len);
+	while (opts_len > 0) {
+		int len;
+
+		if (opts_len < sizeof(*option))
+			return -EINVAL;
+
+		len = sizeof(*option) + option->length * 4;
+		if (len > opts_len)
+			return -EINVAL;
+
+		crit_opt |= !!(option->type & GENEVE_CRIT_OPT_TYPE);
+
+		option = (struct geneve_opt *)((u8 *)option + len);
+		opts_len -= len;
+	};
+
+	key->tun_key.tun_flags |= crit_opt ? TUNNEL_CRIT_OPT : 0;
+
+	return 0;
+}
+
 static int validate_and_copy_set_tun(const struct nlattr *attr,
 				     struct sw_flow_actions **sfa, bool log)
 {
@@ -1544,36 +1581,23 @@ 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) {
-		struct geneve_opt *option;
-		int opts_len = key.tun_opts_len;
-		bool crit_opt = false;
-
-		option = (struct geneve_opt *) TUN_METADATA_OPTS(&key, key.tun_opts_len);
-		while (opts_len > 0) {
-			int len;
-
-			if (opts_len < sizeof(*option))
-				return -EINVAL;
-
-			len = sizeof(*option) + option->length * 4;
-			if (len > opts_len)
-				return -EINVAL;
-
-			crit_opt |= !!(option->type & GENEVE_CRIT_OPT_TYPE);
-
-			option = (struct geneve_opt *)((u8 *)option + len);
-			opts_len -= len;
-		};
-
-		key.tun_key.tun_flags |= crit_opt ? TUNNEL_CRIT_OPT : 0;
+		switch (opts_type) {
+		case OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS:
+			err = validate_and_copy_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);
diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c
index 266c595..8ed7163 100644
--- a/net/openvswitch/vport-vxlan.c
+++ b/net/openvswitch/vport-vxlan.c
@@ -49,6 +49,7 @@
 struct vxlan_port {
 	struct vxlan_sock *vs;
 	char name[IFNAMSIZ];
+	u32 exts; /* VXLAN_EXT_* in <net/vxlan.h> */
 };
 
 static struct vport_ops ovs_vxlan_vport_ops;
@@ -63,16 +64,26 @@ 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_EXT_GBP)
+		flags |= TUNNEL_OPTIONS_PRESENT;
 
 	/* 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 +95,21 @@ 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_EXT_GBP &&
+		    nla_put_flag(skb, OVS_VXLAN_EXT_GBP))
+			return -EMSGSIZE;
+
+		nla_nest_end(skb, exts);
+	}
+
 	return 0;
 }
 
@@ -96,6 +122,31 @@ 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_EXT_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 +179,17 @@ 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, 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, 0,
+			    vxlan_port->exts);
 	if (IS_ERR(vs)) {
 		ovs_vport_free(vport);
 		return (void *)vs;
@@ -141,6 +202,20 @@ 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->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);
@@ -181,6 +256,7 @@ 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,
-- 
1.9.3

^ permalink raw reply related	[flat|nested] 26+ messages in thread