Change in Openvswitch to support MPLS label depth of 3 in ingress direction

The openvswitch was supporting a MPLS label depth of 1 in the ingress
direction though the userspace OVS supports a max depth of 3 labels.
This change enables openvswitch module to support a max depth of
3 labels in the ingress.

Signed-off-by: Martin Varghese <[e-mail]>
Acked-by: Pravin B Shelar <[e-mail]>
Signed-off-by: David S. Miller <[e-mail]>

net/openvswitch/actions.c

···
 	if (err)
 		return err;
 
-	flow_key->mpls.top_lse = lse;
+	flow_key->mpls.lse[0] = lse;
 	return 0;
 }
 

net/openvswitch/flow.c

···
 			memset(&key->ipv4, 0, sizeof(key->ipv4));
 		}
 	} else if (eth_p_mpls(key->eth.type)) {
-		size_t stack_len = MPLS_HLEN;
+		u8 label_count = 1;
 
+		memset(&key->mpls, 0, sizeof(key->mpls));
 		skb_set_inner_network_header(skb, skb->mac_len);
 		while (1) {
 			__be32 lse;
 
-			error = check_header(skb, skb->mac_len + stack_len);
+			error = check_header(skb, skb->mac_len +
+					     label_count * MPLS_HLEN);
 			if (unlikely(error))
 				return 0;
 
 			memcpy(&lse, skb_inner_network_header(skb), MPLS_HLEN);
 
-			if (stack_len == MPLS_HLEN)
-				memcpy(&key->mpls.top_lse, &lse, MPLS_HLEN);
+			if (label_count <= MPLS_LABEL_DEPTH)
+				memcpy(&key->mpls.lse[label_count - 1], &lse,
+				       MPLS_HLEN);
 
-			skb_set_inner_network_header(skb, skb->mac_len + stack_len);
+			skb_set_inner_network_header(skb, skb->mac_len +
+						     label_count * MPLS_HLEN);
 			if (lse & htonl(MPLS_LS_S_MASK))
 				break;
 
-			stack_len += MPLS_HLEN;
+			label_count++;
 		}
+		if (label_count > MPLS_LABEL_DEPTH)
+			label_count = MPLS_LABEL_DEPTH;
+
+		key->mpls.num_labels_mask = GENMASK(label_count - 1, 0);
 	} else if (key->eth.type == htons(ETH_P_IPV6)) {
 		int nh_len;             /* IPv6 Header + Extensions */
 

net/openvswitch/flow.h

···
 	MAC_PROTO_ETHERNET,
 };
 #define SW_FLOW_KEY_INVALID	0x80
+#define MPLS_LABEL_DEPTH       3
 
 /* Store options at the end of the array if they are less than the
  * maximum size. This allows us to get the benefits of variable length
···
 					 */
 	union {
 		struct {
-			__be32 top_lse;	/* top label stack entry */
-		} mpls;
-		struct {
 			u8     proto;	/* IP protocol or lower 8 bits of ARP opcode. */
 			u8     tos;	    /* IP ToS. */
 			u8     ttl;	    /* IP TTL/hop limit. */
···
 				} nd;
 			};
 		} ipv6;
+		struct {
+			u32 num_labels_mask;    /* labels present bitmap of effective length MPLS_LABEL_DEPTH */
+			__be32 lse[MPLS_LABEL_DEPTH];     /* label stack entry  */
+		} mpls;
+
 		struct ovs_key_nsh nsh;         /* network service header */
 	};
 	struct {

net/openvswitch/flow_netlink.c

···
 	[OVS_KEY_ATTR_DP_HASH]	 = { .len = sizeof(u32) },
 	[OVS_KEY_ATTR_TUNNEL]	 = { .len = OVS_ATTR_NESTED,
 				     .next = ovs_tunnel_key_lens, },
-	[OVS_KEY_ATTR_MPLS]	 = { .len = sizeof(struct ovs_key_mpls) },
+	[OVS_KEY_ATTR_MPLS]	 = { .len = OVS_ATTR_VARIABLE },
 	[OVS_KEY_ATTR_CT_STATE]	 = { .len = sizeof(u32) },
 	[OVS_KEY_ATTR_CT_ZONE]	 = { .len = sizeof(u16) },
 	[OVS_KEY_ATTR_CT_MARK]	 = { .len = sizeof(u32) },
···
 
 	if (attrs & (1 << OVS_KEY_ATTR_MPLS)) {
 		const struct ovs_key_mpls *mpls_key;
+		u32 hdr_len;
+		u32 label_count, label_count_mask, i;
 
 		mpls_key = nla_data(a[OVS_KEY_ATTR_MPLS]);
-		SW_FLOW_KEY_PUT(match, mpls.top_lse,
-				mpls_key->mpls_lse, is_mask);
+		hdr_len = nla_len(a[OVS_KEY_ATTR_MPLS]);
+		label_count = hdr_len / sizeof(struct ovs_key_mpls);
+
+		if (label_count == 0 || label_count > MPLS_LABEL_DEPTH ||
+		    hdr_len % sizeof(struct ovs_key_mpls))
+			return -EINVAL;
+
+		label_count_mask =  GENMASK(label_count - 1, 0);
+
+		for (i = 0 ; i < label_count; i++)
+			SW_FLOW_KEY_PUT(match, mpls.lse[i],
+					mpls_key[i].mpls_lse, is_mask);
+
+		SW_FLOW_KEY_PUT(match, mpls.num_labels_mask,
+				label_count_mask, is_mask);
 
 		attrs &= ~(1 << OVS_KEY_ATTR_MPLS);
 	 }
···
 		ether_addr_copy(arp_key->arp_sha, output->ipv4.arp.sha);
 		ether_addr_copy(arp_key->arp_tha, output->ipv4.arp.tha);
 	} else if (eth_p_mpls(swkey->eth.type)) {
+		u8 i, num_labels;
 		struct ovs_key_mpls *mpls_key;
 
-		nla = nla_reserve(skb, OVS_KEY_ATTR_MPLS, sizeof(*mpls_key));
+		num_labels = hweight_long(output->mpls.num_labels_mask);
+		nla = nla_reserve(skb, OVS_KEY_ATTR_MPLS,
+				  num_labels * sizeof(*mpls_key));
 		if (!nla)
 			goto nla_put_failure;
+
 		mpls_key = nla_data(nla);
-		mpls_key->mpls_lse = output->mpls.top_lse;
+		for (i = 0; i < num_labels; i++)
+			mpls_key[i].mpls_lse = output->mpls.lse[i];
 	}
 
 	if ((swkey->eth.type == htons(ETH_P_IP) ||
···
 static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 				  const struct sw_flow_key *key,
 				  struct sw_flow_actions **sfa,
-				  __be16 eth_type, __be16 vlan_tci, bool log);
+				  __be16 eth_type, __be16 vlan_tci,
+				  u32 mpls_label_count, bool log);
 
 static int validate_and_copy_sample(struct net *net, const struct nlattr *attr,
 				    const struct sw_flow_key *key,
 				    struct sw_flow_actions **sfa,
 				    __be16 eth_type, __be16 vlan_tci,
-				    bool log, bool last)
+				    u32 mpls_label_count, bool log, bool last)
 {
 	const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1];
 	const struct nlattr *probability, *actions;
···
 		return err;
 
 	err = __ovs_nla_copy_actions(net, actions, key, sfa,
-				     eth_type, vlan_tci, log);
+				     eth_type, vlan_tci, mpls_label_count, log);
 
 	if (err)
 		return err;
···
 				   const struct sw_flow_key *key,
 				   struct sw_flow_actions **sfa,
 				   __be16 eth_type, __be16 vlan_tci,
-				   bool log, bool last)
+				   u32 mpls_label_count, bool log, bool last)
 {
 	int start, err;
 	u32 exec;
···
 		return err;
 
 	err = __ovs_nla_copy_actions(net, attr, key, sfa,
-				     eth_type, vlan_tci, log);
+				     eth_type, vlan_tci, mpls_label_count, log);
 	if (err)
 		return err;
 
···
 					   const struct sw_flow_key *key,
 					   struct sw_flow_actions **sfa,
 					   __be16 eth_type, __be16 vlan_tci,
+					   u32 mpls_label_count,
 					   bool log, bool last)
 {
 	const struct nlattr *acts_if_greater, *acts_if_lesser_eq;
···
 		return nested_acts_start;
 
 	err = __ovs_nla_copy_actions(net, acts_if_lesser_eq, key, sfa,
-				     eth_type, vlan_tci, log);
+				     eth_type, vlan_tci, mpls_label_count, log);
 
 	if (err)
 		return err;
···
 		return nested_acts_start;
 
 	err = __ovs_nla_copy_actions(net, acts_if_greater, key, sfa,
-				     eth_type, vlan_tci, log);
+				     eth_type, vlan_tci, mpls_label_count, log);
 
 	if (err)
 		return err;
···
 static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 				  const struct sw_flow_key *key,
 				  struct sw_flow_actions **sfa,
-				  __be16 eth_type, __be16 vlan_tci, bool log)
+				  __be16 eth_type, __be16 vlan_tci,
+				  u32 mpls_label_count, bool log)
 {
 	u8 mac_proto = ovs_key_mac_proto(key);
 	const struct nlattr *a;
···
 			     !eth_p_mpls(eth_type)))
 				return -EINVAL;
 			eth_type = mpls->mpls_ethertype;
+			mpls_label_count++;
 			break;
 		}
 
-		case OVS_ACTION_ATTR_POP_MPLS:
+		case OVS_ACTION_ATTR_POP_MPLS: {
+			__be16  proto;
 			if (vlan_tci & htons(VLAN_CFI_MASK) ||
 			    !eth_p_mpls(eth_type))
 				return -EINVAL;
 
-			/* Disallow subsequent L2.5+ set and mpls_pop actions
-			 * as there is no check here to ensure that the new
-			 * eth_type is valid and thus set actions could
-			 * write off the end of the packet or otherwise
-			 * corrupt it.
+			/* Disallow subsequent L2.5+ set actions and mpls_pop
+			 * actions once the last MPLS label in the packet is
+			 * is popped as there is no check here to ensure that
+			 * the new eth type is valid and thus set actions could
+			 * write off the end of the packet or otherwise corrupt
+			 * it.
 			 *
 			 * Support for these actions is planned using packet
 			 * recirculation.
 			 */
-			eth_type = htons(0);
+			proto = nla_get_be16(a);
+			mpls_label_count--;
+
+			if (!eth_p_mpls(proto) || !mpls_label_count)
+				eth_type = htons(0);
+			else
+				eth_type =  proto;
+
 			break;
+		}
 
 		case OVS_ACTION_ATTR_SET:
 			err = validate_set(a, key, sfa,
···
 
 			err = validate_and_copy_sample(net, a, key, sfa,
 						       eth_type, vlan_tci,
+						       mpls_label_count,
 						       log, last);
 			if (err)
 				return err;
···
 
 			err = validate_and_copy_clone(net, a, key, sfa,
 						      eth_type, vlan_tci,
+						      mpls_label_count,
 						      log, last);
 			if (err)
 				return err;
···
 
 			err = validate_and_copy_check_pkt_len(net, a, key, sfa,
 							      eth_type,
-							      vlan_tci, log,
-							      last);
+							      vlan_tci,
+							      mpls_label_count,
+							      log, last);
 			if (err)
 				return err;
 			skip_copy = true;
···
 			 struct sw_flow_actions **sfa, bool log)
 {
 	int err;
+	u32 mpls_label_count = 0;
 
 	*sfa = nla_alloc_flow_actions(min(nla_len(attr), MAX_ACTIONS_BUFSIZE));
 	if (IS_ERR(*sfa))
 		return PTR_ERR(*sfa);
 
+	if (eth_p_mpls(key->eth.type))
+		mpls_label_count = hweight_long(key->mpls.num_labels_mask);
+
 	(*sfa)->orig_len = nla_len(attr);
 	err = __ovs_nla_copy_actions(net, attr, key, sfa, key->eth.type,
-				     key->eth.vlan.tci, log);
+				     key->eth.vlan.tci, mpls_label_count, log);
 	if (err)
 		ovs_nla_free_flow_actions(*sfa);