flow.c

bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow,
		const struct sw_flow_key *key, int key_len)
{
	int key_start;
	key_start = flow_key_start(key);

	return __flow_cmp_unmasked_key(flow, key, key_start, key_len);

}

struct sw_flow *ovs_flow_lookup_unmasked_key(struct flow_table *table,
				       struct sw_flow_match *match)
{
	struct sw_flow_key *unmasked = match->key;
	int key_len = match->range.end;
	struct sw_flow *flow;

	flow = ovs_flow_lookup(table, unmasked);
	if (flow && (!ovs_flow_cmp_unmasked_key(flow, unmasked, key_len)))
		flow = NULL;

	return flow;
}

static struct sw_flow *ovs_masked_flow_lookup(struct flow_table *table,
				    const struct sw_flow_key *flow_key,
				    struct sw_flow_mask *mask)
{
	struct sw_flow *flow;
	struct hlist_head *head;
	int key_start = mask->range.start;
	int key_len = mask->range.end;
	u32 hash;
	struct sw_flow_key masked_key;

	ovs_flow_key_mask(&masked_key, flow_key, mask);
	hash = ovs_flow_hash(&masked_key, key_start, key_len);
	head = find_bucket(table, hash);
	hlist_for_each_entry_rcu(flow, head, hash_node[table->node_ver]) {
		if (flow->mask == mask &&
		    __flow_cmp_key(flow, &masked_key, key_start, key_len))
			return flow;
	}
	return NULL;
}

struct sw_flow *ovs_flow_lookup(struct flow_table *tbl,
				const struct sw_flow_key *key)
{
	struct sw_flow *flow = NULL;
	struct sw_flow_mask *mask;

	list_for_each_entry_rcu(mask, tbl->mask_list, list) {
		flow = ovs_masked_flow_lookup(tbl, key, mask);
		if (flow)  /* Found */
			break;
	}

	return flow;
}


void ovs_flow_insert(struct flow_table *table, struct sw_flow *flow)
{
	flow->hash = ovs_flow_hash(&flow->key, flow->mask->range.start,
			flow->mask->range.end);
	__tbl_insert(table, flow);
}

void ovs_flow_remove(struct flow_table *table, struct sw_flow *flow)
{
	BUG_ON(table->count == 0);
	hlist_del_rcu(&flow->hash_node[table->node_ver]);
	table->count--;
}

/* The size of the argument for each %OVS_KEY_ATTR_* Netlink attribute.  */
const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = {
	[OVS_KEY_ATTR_ENCAP] = -1,
	[OVS_KEY_ATTR_PRIORITY] = sizeof(u32),
	[OVS_KEY_ATTR_IN_PORT] = sizeof(u32),
	[OVS_KEY_ATTR_SKB_MARK] = sizeof(u32),
	[OVS_KEY_ATTR_ETHERNET] = sizeof(struct ovs_key_ethernet),
	[OVS_KEY_ATTR_VLAN] = sizeof(__be16),
	[OVS_KEY_ATTR_ETHERTYPE] = sizeof(__be16),
	[OVS_KEY_ATTR_IPV4] = sizeof(struct ovs_key_ipv4),
	[OVS_KEY_ATTR_IPV6] = sizeof(struct ovs_key_ipv6),
	[OVS_KEY_ATTR_TCP] = sizeof(struct ovs_key_tcp),
	[OVS_KEY_ATTR_UDP] = sizeof(struct ovs_key_udp),
	[OVS_KEY_ATTR_ICMP] = sizeof(struct ovs_key_icmp),
	[OVS_KEY_ATTR_ICMPV6] = sizeof(struct ovs_key_icmpv6),
	[OVS_KEY_ATTR_ARP] = sizeof(struct ovs_key_arp),
	[OVS_KEY_ATTR_ND] = sizeof(struct ovs_key_nd),
	[OVS_KEY_ATTR_TUNNEL] = -1,
};

static bool is_all_zero(const u8 *fp, size_t size)
{
	int i;

	if (!fp)
		return false;

	for (i = 0; i < size; i++)
		if (fp[i])
			return false;

	return true;
}

static int __parse_flow_nlattrs(const struct nlattr *attr,
			      const struct nlattr *a[],
			      u64 *attrsp, bool nz)
{
	const struct nlattr *nla;
	u32 attrs;
	int rem;

	attrs = *attrsp;
	nla_for_each_nested(nla, attr, rem) {
		u16 type = nla_type(nla);
		int expected_len;

		if (type > OVS_KEY_ATTR_MAX) {
			OVS_NLERR("Unknown key attribute (type=%d, max=%d).\n",
				  type, OVS_KEY_ATTR_MAX);
		}

		if (attrs & (1 << type)) {
			OVS_NLERR("Duplicate key attribute (type %d).\n", type);
			return -EINVAL;
		}

		expected_len = ovs_key_lens[type];
		if (nla_len(nla) != expected_len && expected_len != -1) {
			OVS_NLERR("Key attribute has unexpected length (type=%d"
				  ", length=%d, expected=%d).\n", type,
				  nla_len(nla), expected_len);
			return -EINVAL;
		}

		if (!nz || !is_all_zero(nla_data(nla), expected_len)) {
			attrs |= 1 << type;
			a[type] = nla;
		}
	}
	if (rem) {
		OVS_NLERR("Message has %d unknown bytes.\n", rem);
		return -EINVAL;
	}

	*attrsp = attrs;
	return 0;
}

static int parse_flow_mask_nlattrs(const struct nlattr *attr,
			      const struct nlattr *a[], u64 *attrsp)
{
	return __parse_flow_nlattrs(attr, a, attrsp, true);
}

static int parse_flow_nlattrs(const struct nlattr *attr,
			      const struct nlattr *a[], u64 *attrsp)
{
	return __parse_flow_nlattrs(attr, a, attrsp, false);
}

int ovs_ipv4_tun_from_nlattr(const struct nlattr *attr,
			     struct sw_flow_match *match, bool is_mask)
{
	struct nlattr *a;
	int rem;
	bool ttl = false;
	__be16 tun_flags = 0;

	nla_for_each_nested(a, attr, rem) {
		int type = nla_type(a);
		static const u32 ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1] = {
			[OVS_TUNNEL_KEY_ATTR_ID] = sizeof(u64),
			[OVS_TUNNEL_KEY_ATTR_IPV4_SRC] = sizeof(u32),
			[OVS_TUNNEL_KEY_ATTR_IPV4_DST] = sizeof(u32),
			[OVS_TUNNEL_KEY_ATTR_TOS] = 1,
			[OVS_TUNNEL_KEY_ATTR_TTL] = 1,
			[OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT] = 0,
			[OVS_TUNNEL_KEY_ATTR_CSUM] = 0,
		};

		if (type > OVS_TUNNEL_KEY_ATTR_MAX) {
			OVS_NLERR("Unknown IPv4 tunnel attribute (type=%d, max=%d).\n",
			type, OVS_TUNNEL_KEY_ATTR_MAX);
			return -EINVAL;
		}

		if (ovs_tunnel_key_lens[type] != nla_len(a)) {
			OVS_NLERR("IPv4 tunnel attribute type has unexpected "
				  " length (type=%d, length=%d, expected=%d).\n",
				  type, nla_len(a), ovs_tunnel_key_lens[type]);
			return -EINVAL;
		}

		switch (type) {
		case OVS_TUNNEL_KEY_ATTR_ID:
			SW_FLOW_KEY_PUT(match, tun_key.tun_id,
					nla_get_be64(a), is_mask);
			tun_flags |= TUNNEL_KEY;
			break;
		case OVS_TUNNEL_KEY_ATTR_IPV4_SRC:
			SW_FLOW_KEY_PUT(match, tun_key.ipv4_src,
					nla_get_be32(a), is_mask);
			break;
		case OVS_TUNNEL_KEY_ATTR_IPV4_DST:
			SW_FLOW_KEY_PUT(match, tun_key.ipv4_dst,
					nla_get_be32(a), is_mask);
			break;
		case OVS_TUNNEL_KEY_ATTR_TOS:
			SW_FLOW_KEY_PUT(match, tun_key.ipv4_tos,
					nla_get_u8(a), is_mask);
			break;
		case OVS_TUNNEL_KEY_ATTR_TTL:
			SW_FLOW_KEY_PUT(match, tun_key.ipv4_ttl,
					nla_get_u8(a), is_mask);
			ttl = true;
			break;
		case OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT:
			tun_flags |= TUNNEL_DONT_FRAGMENT;
			break;
		case OVS_TUNNEL_KEY_ATTR_CSUM:
			tun_flags |= TUNNEL_CSUM;
			break;
		default:
			return -EINVAL;
		}
	}

	SW_FLOW_KEY_PUT(match, tun_key.tun_flags, tun_flags, is_mask);

	if (rem > 0) {
		OVS_NLERR("IPv4 tunnel attribute has %d unknown bytes.\n", rem);
		return -EINVAL;
	}

	if (!is_mask) {
		if (!match->key->tun_key.ipv4_dst) {
			OVS_NLERR("IPv4 tunnel destination address is zero.\n");
			return -EINVAL;
		}

		if (!ttl) {
			OVS_NLERR("IPv4 tunnel TTL not specified.\n");
			return -EINVAL;
		}
	}

	return 0;
}

int ovs_ipv4_tun_to_nlattr(struct sk_buff *skb,
			   const struct ovs_key_ipv4_tunnel *tun_key,
			   const struct ovs_key_ipv4_tunnel *output)
{
	struct nlattr *nla;

	nla = nla_nest_start(skb, OVS_KEY_ATTR_TUNNEL);
	if (!nla)
		return -EMSGSIZE;

	if (output->tun_flags & TUNNEL_KEY &&
	    nla_put_be64(skb, OVS_TUNNEL_KEY_ATTR_ID, output->tun_id))
		return -EMSGSIZE;
	if (output->ipv4_src &&
		nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_SRC, output->ipv4_src))
		return -EMSGSIZE;
	if (output->ipv4_dst &&
		nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_DST, output->ipv4_dst))
		return -EMSGSIZE;
	if (output->ipv4_tos &&
		nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TOS, output->ipv4_tos))
		return -EMSGSIZE;
	if (nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TTL, output->ipv4_ttl))
		return -EMSGSIZE;
	if ((output->tun_flags & TUNNEL_DONT_FRAGMENT) &&
		nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT))
		return -EMSGSIZE;
	if ((output->tun_flags & TUNNEL_CSUM) &&
		nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_CSUM))
		return -EMSGSIZE;

	nla_nest_end(skb, nla);
	return 0;
}

static int metadata_from_nlattrs(struct sw_flow_match *match,  u64 *attrs,
		const struct nlattr **a, bool is_mask)
{
	if (*attrs & (1 << OVS_KEY_ATTR_PRIORITY)) {
		SW_FLOW_KEY_PUT(match, phy.priority,
			  nla_get_u32(a[OVS_KEY_ATTR_PRIORITY]), is_mask);
		*attrs &= ~(1 << OVS_KEY_ATTR_PRIORITY);
	}

	if (*attrs & (1 << OVS_KEY_ATTR_IN_PORT)) {
		u32 in_port = nla_get_u32(a[OVS_KEY_ATTR_IN_PORT]);

		if (is_mask)
			in_port = 0xffffffff; /* Always exact match in_port. */
		else if (in_port >= DP_MAX_PORTS)
			return -EINVAL;

		SW_FLOW_KEY_PUT(match, phy.in_port, in_port, is_mask);
		*attrs &= ~(1 << OVS_KEY_ATTR_IN_PORT);
	} else if (!is_mask) {
		SW_FLOW_KEY_PUT(match, phy.in_port, DP_MAX_PORTS, is_mask);
	}

	if (*attrs & (1 << OVS_KEY_ATTR_SKB_MARK)) {
		uint32_t mark = nla_get_u32(a[OVS_KEY_ATTR_SKB_MARK]);

		SW_FLOW_KEY_PUT(match, phy.skb_mark, mark, is_mask);
		*attrs &= ~(1 << OVS_KEY_ATTR_SKB_MARK);
	}
	if (*attrs & (1 << OVS_KEY_ATTR_TUNNEL)) {
		if (ovs_ipv4_tun_from_nlattr(a[OVS_KEY_ATTR_TUNNEL], match,
					is_mask))
			return -EINVAL;
		*attrs &= ~(1 << OVS_KEY_ATTR_TUNNEL);
	}
	return 0;
}

static int ovs_key_from_nlattrs(struct sw_flow_match *match,  u64 attrs,
		const struct nlattr **a, bool is_mask)
{
	int err;
	u64 orig_attrs = attrs;

	err = metadata_from_nlattrs(match, &attrs, a, is_mask);
	if (err)
		return err;

	if (attrs & (1 << OVS_KEY_ATTR_ETHERNET)) {
		const struct ovs_key_ethernet *eth_key;

		eth_key = nla_data(a[OVS_KEY_ATTR_ETHERNET]);
		SW_FLOW_KEY_MEMCPY(match, eth.src,
				eth_key->eth_src, ETH_ALEN, is_mask);
		SW_FLOW_KEY_MEMCPY(match, eth.dst,
				eth_key->eth_dst, ETH_ALEN, is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_ETHERNET);
	}

	if (attrs & (1 << OVS_KEY_ATTR_VLAN)) {
		__be16 tci;

		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
		if (!(tci & htons(VLAN_TAG_PRESENT))) {
			if (is_mask)
				OVS_NLERR("VLAN TCI mask does not have exact match for VLAN_TAG_PRESENT bit.\n");
			else
				OVS_NLERR("VLAN TCI does not have VLAN_TAG_PRESENT bit set.\n");

			return -EINVAL;
		}

		SW_FLOW_KEY_PUT(match, eth.tci, tci, is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_VLAN);
	} else if (!is_mask)
		SW_FLOW_KEY_PUT(match, eth.tci, htons(0xffff), true);

	if (attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) {
		__be16 eth_type;

		eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
		if (is_mask) {
			/* Always exact match EtherType. */
			eth_type = htons(0xffff);
		} else if (ntohs(eth_type) < ETH_P_802_3_MIN) {
			OVS_NLERR("EtherType is less than minimum (type=%x, min=%x).\n",
					ntohs(eth_type), ETH_P_802_3_MIN);
			return -EINVAL;
		}

		SW_FLOW_KEY_PUT(match, eth.type, eth_type, is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE);
	} else if (!is_mask) {
		SW_FLOW_KEY_PUT(match, eth.type, htons(ETH_P_802_2), is_mask);
	}

	if (attrs & (1 << OVS_KEY_ATTR_IPV4)) {
		const struct ovs_key_ipv4 *ipv4_key;

		ipv4_key = nla_data(a[OVS_KEY_ATTR_IPV4]);
		if (!is_mask && ipv4_key->ipv4_frag > OVS_FRAG_TYPE_MAX) {
			OVS_NLERR("Unknown IPv4 fragment type (value=%d, max=%d).\n",
				ipv4_key->ipv4_frag, OVS_FRAG_TYPE_MAX);
			return -EINVAL;
		}
		SW_FLOW_KEY_PUT(match, ip.proto,
				ipv4_key->ipv4_proto, is_mask);
		SW_FLOW_KEY_PUT(match, ip.tos,
				ipv4_key->ipv4_tos, is_mask);
		SW_FLOW_KEY_PUT(match, ip.ttl,
				ipv4_key->ipv4_ttl, is_mask);
		SW_FLOW_KEY_PUT(match, ip.frag,
				ipv4_key->ipv4_frag, is_mask);
		SW_FLOW_KEY_PUT(match, ipv4.addr.src,
				ipv4_key->ipv4_src, is_mask);
		SW_FLOW_KEY_PUT(match, ipv4.addr.dst,
				ipv4_key->ipv4_dst, is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_IPV4);
	}

	if (attrs & (1 << OVS_KEY_ATTR_IPV6)) {
		const struct ovs_key_ipv6 *ipv6_key;

		ipv6_key = nla_data(a[OVS_KEY_ATTR_IPV6]);
		if (!is_mask && ipv6_key->ipv6_frag > OVS_FRAG_TYPE_MAX) {
			OVS_NLERR("Unknown IPv6 fragment type (value=%d, max=%d).\n",
				ipv6_key->ipv6_frag, OVS_FRAG_TYPE_MAX);
			return -EINVAL;
		}
		SW_FLOW_KEY_PUT(match, ipv6.label,
				ipv6_key->ipv6_label, is_mask);
		SW_FLOW_KEY_PUT(match, ip.proto,
				ipv6_key->ipv6_proto, is_mask);
		SW_FLOW_KEY_PUT(match, ip.tos,
				ipv6_key->ipv6_tclass, is_mask);
		SW_FLOW_KEY_PUT(match, ip.ttl,
				ipv6_key->ipv6_hlimit, is_mask);
		SW_FLOW_KEY_PUT(match, ip.frag,
				ipv6_key->ipv6_frag, is_mask);
		SW_FLOW_KEY_MEMCPY(match, ipv6.addr.src,
				ipv6_key->ipv6_src,
				sizeof(match->key->ipv6.addr.src),
				is_mask);
		SW_FLOW_KEY_MEMCPY(match, ipv6.addr.dst,
				ipv6_key->ipv6_dst,
				sizeof(match->key->ipv6.addr.dst),
				is_mask);

		attrs &= ~(1 << OVS_KEY_ATTR_IPV6);
	}

	if (attrs & (1 << OVS_KEY_ATTR_ARP)) {
		const struct ovs_key_arp *arp_key;

		arp_key = nla_data(a[OVS_KEY_ATTR_ARP]);
		if (!is_mask && (arp_key->arp_op & htons(0xff00))) {
			OVS_NLERR("Unknown ARP opcode (opcode=%d).\n",
				  arp_key->arp_op);
			return -EINVAL;
		}

		SW_FLOW_KEY_PUT(match, ipv4.addr.src,
				arp_key->arp_sip, is_mask);
		SW_FLOW_KEY_PUT(match, ipv4.addr.dst,
			arp_key->arp_tip, is_mask);
		SW_FLOW_KEY_PUT(match, ip.proto,
				ntohs(arp_key->arp_op), is_mask);
		SW_FLOW_KEY_MEMCPY(match, ipv4.arp.sha,
				arp_key->arp_sha, ETH_ALEN, is_mask);
		SW_FLOW_KEY_MEMCPY(match, ipv4.arp.tha,
				arp_key->arp_tha, ETH_ALEN, is_mask);

		attrs &= ~(1 << OVS_KEY_ATTR_ARP);
	}

	if (attrs & (1 << OVS_KEY_ATTR_TCP)) {
		const struct ovs_key_tcp *tcp_key;

		tcp_key = nla_data(a[OVS_KEY_ATTR_TCP]);
		if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) {
			SW_FLOW_KEY_PUT(match, ipv4.tp.src,
					tcp_key->tcp_src, is_mask);
			SW_FLOW_KEY_PUT(match, ipv4.tp.dst,
					tcp_key->tcp_dst, is_mask);
		} else {
			SW_FLOW_KEY_PUT(match, ipv6.tp.src,
					tcp_key->tcp_src, is_mask);
			SW_FLOW_KEY_PUT(match, ipv6.tp.dst,
					tcp_key->tcp_dst, is_mask);
		}
		attrs &= ~(1 << OVS_KEY_ATTR_TCP);
	}

	if (attrs & (1 << OVS_KEY_ATTR_UDP)) {
		const struct ovs_key_udp *udp_key;

		udp_key = nla_data(a[OVS_KEY_ATTR_UDP]);
		if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) {
			SW_FLOW_KEY_PUT(match, ipv4.tp.src,
					udp_key->udp_src, is_mask);
			SW_FLOW_KEY_PUT(match, ipv4.tp.dst,
					udp_key->udp_dst, is_mask);
		} else {
			SW_FLOW_KEY_PUT(match, ipv6.tp.src,
					udp_key->udp_src, is_mask);
			SW_FLOW_KEY_PUT(match, ipv6.tp.dst,
					udp_key->udp_dst, is_mask);
		}
		attrs &= ~(1 << OVS_KEY_ATTR_UDP);
	}

	if (attrs & (1 << OVS_KEY_ATTR_ICMP)) {
		const struct ovs_key_icmp *icmp_key;

		icmp_key = nla_data(a[OVS_KEY_ATTR_ICMP]);
		SW_FLOW_KEY_PUT(match, ipv4.tp.src,
				htons(icmp_key->icmp_type), is_mask);
		SW_FLOW_KEY_PUT(match, ipv4.tp.dst,
				htons(icmp_key->icmp_code), is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_ICMP);
	}

	if (attrs & (1 << OVS_KEY_ATTR_ICMPV6)) {
		const struct ovs_key_icmpv6 *icmpv6_key;

		icmpv6_key = nla_data(a[OVS_KEY_ATTR_ICMPV6]);
		SW_FLOW_KEY_PUT(match, ipv6.tp.src,
				htons(icmpv6_key->icmpv6_type), is_mask);
		SW_FLOW_KEY_PUT(match, ipv6.tp.dst,
				htons(icmpv6_key->icmpv6_code), is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_ICMPV6);
	}

	if (attrs & (1 << OVS_KEY_ATTR_ND)) {
		const struct ovs_key_nd *nd_key;

		nd_key = nla_data(a[OVS_KEY_ATTR_ND]);
		SW_FLOW_KEY_MEMCPY(match, ipv6.nd.target,
			nd_key->nd_target,
			sizeof(match->key->ipv6.nd.target),
			is_mask);
		SW_FLOW_KEY_MEMCPY(match, ipv6.nd.sll,
			nd_key->nd_sll, ETH_ALEN, is_mask);
		SW_FLOW_KEY_MEMCPY(match, ipv6.nd.tll,
				nd_key->nd_tll, ETH_ALEN, is_mask);
		attrs &= ~(1 << OVS_KEY_ATTR_ND);
	}

	if (attrs != 0)
		return -EINVAL;

	return 0;
}

/**
 * ovs_match_from_nlattrs - parses Netlink attributes into a flow key and
 * mask. In case the 'mask' is NULL, the flow is treated as exact match
 * flow. Otherwise, it is treated as a wildcarded flow, except the mask
 * does not include any don't care bit.
 * @match: receives the extracted flow match information.
 * @key: Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink attribute
 * sequence. The fields should of the packet that triggered the creation
 * of this flow.
 * @mask: Optional. Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink
 * attribute specifies the mask field of the wildcarded flow.
 */
int ovs_match_from_nlattrs(struct sw_flow_match *match,
			   const struct nlattr *key,
			   const struct nlattr *mask)
{
	const struct nlattr *a[OVS_KEY_ATTR_MAX + 1];
	const struct nlattr *encap;
	u64 key_attrs = 0;
	u64 mask_attrs = 0;
	bool encap_valid = false;
	int err;

	err = parse_flow_nlattrs(key, a, &key_attrs);
	if (err)
		return err;

	if ((key_attrs & (1 << OVS_KEY_ATTR_ETHERNET)) &&
	    (key_attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) &&
	    (nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]) == htons(ETH_P_8021Q))) {
		__be16 tci;

		if (!((key_attrs & (1 << OVS_KEY_ATTR_VLAN)) &&
		      (key_attrs & (1 << OVS_KEY_ATTR_ENCAP)))) {
			OVS_NLERR("Invalid Vlan frame.\n");
			return -EINVAL;
		}

		key_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE);
		tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
		encap = a[OVS_KEY_ATTR_ENCAP];
		key_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP);
		encap_valid = true;

		if (tci & htons(VLAN_TAG_PRESENT)) {
			err = parse_flow_nlattrs(encap, a, &key_attrs);
			if (err)
				return err;
		} else if (!tci) {
			/* Corner case for truncated 802.1Q header. */
			if (nla_len(encap)) {
				OVS_NLERR("Truncated 802.1Q header has non-zero encap attribute.\n");
				return -EINVAL;
			}
		} else {
			OVS_NLERR("Encap attribute is set for a non-VLAN frame.\n");
			return  -EINVAL;
		}
	}

	err = ovs_key_from_nlattrs(match, key_attrs, a, false);
	if (err)
		return err;

	if (mask) {
		err = parse_flow_mask_nlattrs(mask, a, &mask_attrs);
		if (err)
			return err;

		if (mask_attrs & 1ULL << OVS_KEY_ATTR_ENCAP)  {
			__be16 eth_type = 0;
			__be16 tci = 0;

			if (!encap_valid) {
				OVS_NLERR("Encap mask attribute is set for non-VLAN frame.\n");
				return  -EINVAL;
			}

			mask_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP);
			if (a[OVS_KEY_ATTR_ETHERTYPE])
				eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);

			if (eth_type == htons(0xffff)) {
				mask_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE);
				encap = a[OVS_KEY_ATTR_ENCAP];
				err = parse_flow_mask_nlattrs(encap, a, &mask_attrs);
			} else {
				OVS_NLERR("VLAN frames must have an exact match on the TPID (mask=%x).\n",
						ntohs(eth_type));
				return -EINVAL;
			}

			if (a[OVS_KEY_ATTR_VLAN])
				tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);

			if (!(tci & htons(VLAN_TAG_PRESENT))) {
				OVS_NLERR("VLAN tag present bit must have an exact match (tci_mask=%x).\n", ntohs(tci));
				return -EINVAL;
			}
		}

		err = ovs_key_from_nlattrs(match, mask_attrs, a, true);
		if (err)
			return err;
	} else {
		/* Populate exact match flow's key mask. */
		if (match->mask)
			ovs_sw_flow_mask_set(match->mask, &match->range, 0xff);
	}

	if (!ovs_match_validate(match, key_attrs, mask_attrs))
		return -EINVAL;

	return 0;
}

/**
 * ovs_flow_metadata_from_nlattrs - parses Netlink attributes into a flow key.
 * @flow: Receives extracted in_port, priority, tun_key and skb_mark.
 * @attr: Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink attribute
 * sequence.
 *
 * This parses a series of Netlink attributes that form a flow key, which must
 * take the same form accepted by flow_from_nlattrs(), but only enough of it to
 * get the metadata, that is, the parts of the flow key that cannot be
 * extracted from the packet itself.
 */

int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow,
		const struct nlattr *attr)
{
	struct ovs_key_ipv4_tunnel *tun_key = &flow->key.tun_key;
	const struct nlattr *a[OVS_KEY_ATTR_MAX + 1];
	u64 attrs = 0;
	int err;
	struct sw_flow_match match;

	flow->key.phy.in_port = DP_MAX_PORTS;
	flow->key.phy.priority = 0;
	flow->key.phy.skb_mark = 0;
	memset(tun_key, 0, sizeof(flow->key.tun_key));

	err = parse_flow_nlattrs(attr, a, &attrs);
	if (err)
		return -EINVAL;

	memset(&match, 0, sizeof(match));
	match.key = &flow->key;

	err = metadata_from_nlattrs(&match, &attrs, a, false);
	if (err)
		return err;

	return 0;
}

int ovs_flow_to_nlattrs(const struct sw_flow_key *swkey,
		const struct sw_flow_key *output, struct sk_buff *skb)
{
	struct ovs_key_ethernet *eth_key;
	struct nlattr *nla, *encap;
	bool is_mask = (swkey != output);

	if (nla_put_u32(skb, OVS_KEY_ATTR_PRIORITY, output->phy.priority))
		goto nla_put_failure;

	if ((swkey->tun_key.ipv4_dst || is_mask) &&
	    ovs_ipv4_tun_to_nlattr(skb, &swkey->tun_key, &output->tun_key))
		goto nla_put_failure;

	if (swkey->phy.in_port == DP_MAX_PORTS) {
		if (is_mask && (output->phy.in_port == 0xffff))
			if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT, 0xffffffff))
				goto nla_put_failure;
	} else {
		u16 upper_u16;
		upper_u16 = !is_mask ? 0 : 0xffff;

		if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT,
				(upper_u16 << 16) | output->phy.in_port))
			goto nla_put_failure;
	}

	if (nla_put_u32(skb, OVS_KEY_ATTR_SKB_MARK, output->phy.skb_mark))
		goto nla_put_failure;

	nla = nla_reserve(skb, OVS_KEY_ATTR_ETHERNET, sizeof(*eth_key));
	if (!nla)
		goto nla_put_failure;

	eth_key = nla_data(nla);
	memcpy(eth_key->eth_src, output->eth.src, ETH_ALEN);
	memcpy(eth_key->eth_dst, output->eth.dst, ETH_ALEN);

	if (swkey->eth.tci || swkey->eth.type == htons(ETH_P_8021Q)) {
		__be16 eth_type;
		eth_type = !is_mask ? htons(ETH_P_8021Q) : htons(0xffff);
		if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, eth_type) ||
		    nla_put_be16(skb, OVS_KEY_ATTR_VLAN, output->eth.tci))
			goto nla_put_failure;
		encap = nla_nest_start(skb, OVS_KEY_ATTR_ENCAP);
		if (!swkey->eth.tci)
			goto unencap;
	} else
		encap = NULL;

	if (swkey->eth.type == htons(ETH_P_802_2)) {
		/*
		 * Ethertype 802.2 is represented in the netlink with omitted
		 * OVS_KEY_ATTR_ETHERTYPE in the flow key attribute, and
		 * 0xffff in the mask attribute.  Ethertype can also
		 * be wildcarded.
		 */
		if (is_mask && output->eth.type)
			if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE,
						output->eth.type))
				goto nla_put_failure;
		goto unencap;
	}

	if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, output->eth.type))
		goto nla_put_failure;

	if (swkey->eth.type == htons(ETH_P_IP)) {
		struct ovs_key_ipv4 *ipv4_key;

		nla = nla_reserve(skb, OVS_KEY_ATTR_IPV4, sizeof(*ipv4_key));
		if (!nla)
			goto nla_put_failure;
		ipv4_key = nla_data(nla);
		ipv4_key->ipv4_src = output->ipv4.addr.src;
		ipv4_key->ipv4_dst = output->ipv4.addr.dst;
		ipv4_key->ipv4_proto = output->ip.proto;
		ipv4_key->ipv4_tos = output->ip.tos;
		ipv4_key->ipv4_ttl = output->ip.ttl;
		ipv4_key->ipv4_frag = output->ip.frag;
	} else if (swkey->eth.type == htons(ETH_P_IPV6)) {
		struct ovs_key_ipv6 *ipv6_key;

		nla = nla_reserve(skb, OVS_KEY_ATTR_IPV6, sizeof(*ipv6_key));
		if (!nla)
			goto nla_put_failure;
		ipv6_key = nla_data(nla);
		memcpy(ipv6_key->ipv6_src, &output->ipv6.addr.src,
				sizeof(ipv6_key->ipv6_src));
		memcpy(ipv6_key->ipv6_dst, &output->ipv6.addr.dst,
				sizeof(ipv6_key->ipv6_dst));
		ipv6_key->ipv6_label = output->ipv6.label;
		ipv6_key->ipv6_proto = output->ip.proto;
		ipv6_key->ipv6_tclass = output->ip.tos;
		ipv6_key->ipv6_hlimit = output->ip.ttl;
		ipv6_key->ipv6_frag = output->ip.frag;
	} else if (swkey->eth.type == htons(ETH_P_ARP) ||
		   swkey->eth.type == htons(ETH_P_RARP)) {
		struct ovs_key_arp *arp_key;

		nla = nla_reserve(skb, OVS_KEY_ATTR_ARP, sizeof(*arp_key));
		if (!nla)
			goto nla_put_failure;
		arp_key = nla_data(nla);
		memset(arp_key, 0, sizeof(struct ovs_key_arp));
		arp_key->arp_sip = output->ipv4.addr.src;
		arp_key->arp_tip = output->ipv4.addr.dst;
		arp_key->arp_op = htons(output->ip.proto);
		memcpy(arp_key->arp_sha, output->ipv4.arp.sha, ETH_ALEN);
		memcpy(arp_key->arp_tha, output->ipv4.arp.tha, ETH_ALEN);
	}

	if ((swkey->eth.type == htons(ETH_P_IP) ||
	     swkey->eth.type == htons(ETH_P_IPV6)) &&
	     swkey->ip.frag != OVS_FRAG_TYPE_LATER) {

		if (swkey->ip.proto == IPPROTO_TCP) {
			struct ovs_key_tcp *tcp_key;

			nla = nla_reserve(skb, OVS_KEY_ATTR_TCP, sizeof(*tcp_key));
			if (!nla)
				goto nla_put_failure;
			tcp_key = nla_data(nla);
			if (swkey->eth.type == htons(ETH_P_IP)) {
				tcp_key->tcp_src = output->ipv4.tp.src;
				tcp_key->tcp_dst = output->ipv4.tp.dst;
			} else if (swkey->eth.type == htons(ETH_P_IPV6)) {
				tcp_key->tcp_src = output->ipv6.tp.src;
				tcp_key->tcp_dst = output->ipv6.tp.dst;
			}
		} else if (swkey->ip.proto == IPPROTO_UDP) {
			struct ovs_key_udp *udp_key;

			nla = nla_reserve(skb, OVS_KEY_ATTR_UDP, sizeof(*udp_key));
			if (!nla)
				goto nla_put_failure;
			udp_key = nla_data(nla);
			if (swkey->eth.type == htons(ETH_P_IP)) {
				udp_key->udp_src = output->ipv4.tp.src;
				udp_key->udp_dst = output->ipv4.tp.dst;
			} else if (swkey->eth.type == htons(ETH_P_IPV6)) {
				udp_key->udp_src = output->ipv6.tp.src;
				udp_key->udp_dst = output->ipv6.tp.dst;
			}
		} else if (swkey->eth.type == htons(ETH_P_IP) &&
			   swkey->ip.proto == IPPROTO_ICMP) {
			struct ovs_key_icmp *icmp_key;

			nla = nla_reserve(skb, OVS_KEY_ATTR_ICMP, sizeof(*icmp_key));
			if (!nla)
				goto nla_put_failure;
			icmp_key = nla_data(nla);
			icmp_key->icmp_type = ntohs(output->ipv4.tp.src);
			icmp_key->icmp_code = ntohs(output->ipv4.tp.dst);
		} else if (swkey->eth.type == htons(ETH_P_IPV6) &&
			   swkey->ip.proto == IPPROTO_ICMPV6) {
			struct ovs_key_icmpv6 *icmpv6_key;

			nla = nla_reserve(skb, OVS_KEY_ATTR_ICMPV6,
						sizeof(*icmpv6_key));
			if (!nla)
				goto nla_put_failure;
			icmpv6_key = nla_data(nla);
			icmpv6_key->icmpv6_type = ntohs(output->ipv6.tp.src);
			icmpv6_key->icmpv6_code = ntohs(output->ipv6.tp.dst);

			if (icmpv6_key->icmpv6_type == NDISC_NEIGHBOUR_SOLICITATION ||
			    icmpv6_key->icmpv6_type == NDISC_NEIGHBOUR_ADVERTISEMENT) {
				struct ovs_key_nd *nd_key;

				nla = nla_reserve(skb, OVS_KEY_ATTR_ND, sizeof(*nd_key));
				if (!nla)
					goto nla_put_failure;
				nd_key = nla_data(nla);
				memcpy(nd_key->nd_target, &output->ipv6.nd.target,
							sizeof(nd_key->nd_target));
				memcpy(nd_key->nd_sll, output->ipv6.nd.sll, ETH_ALEN);
				memcpy(nd_key->nd_tll, output->ipv6.nd.tll, ETH_ALEN);
			}
		}
	}

unencap:
	if (encap)
		nla_nest_end(skb, encap);

	return 0;

nla_put_failure:
	return -EMSGSIZE;
}

/* Initializes the flow module.
 * Returns zero if successful or a negative error code. */
int ovs_flow_init(void)
{
	flow_cache = kmem_cache_create("sw_flow", sizeof(struct sw_flow), 0,
					0, NULL);
	if (flow_cache == NULL)
		return -ENOMEM;

	return 0;
}

/* Uninitializes the flow module. */
void ovs_flow_exit(void)
{
	kmem_cache_destroy(flow_cache);
}

struct sw_flow_mask *ovs_sw_flow_mask_alloc(void)
{
	struct sw_flow_mask *mask;

	mask = kmalloc(sizeof(*mask), GFP_KERNEL);
	if (mask)
		mask->ref_count = 0;

	return mask;
}

void ovs_sw_flow_mask_add_ref(struct sw_flow_mask *mask)
{
	mask->ref_count++;
}

void ovs_sw_flow_mask_del_ref(struct sw_flow_mask *mask, bool deferred)
{
	if (!mask)
		return;

	BUG_ON(!mask->ref_count);
	mask->ref_count--;

	if (!mask->ref_count) {
		list_del_rcu(&mask->list);
		if (deferred)
			kfree_rcu(mask, rcu);
		else
			kfree(mask);
	}
}

static bool ovs_sw_flow_mask_equal(const struct sw_flow_mask *a,
		const struct sw_flow_mask *b)
{
	u8 *a_ = (u8 *)&a->key + a->range.start;
	u8 *b_ = (u8 *)&b->key + b->range.start;

	return  (a->range.end == b->range.end)
		&& (a->range.start == b->range.start)
		&& (memcmp(a_, b_, ovs_sw_flow_mask_actual_size(a)) == 0);
}

struct sw_flow_mask *ovs_sw_flow_mask_find(const struct flow_table *tbl,
                                           const struct sw_flow_mask *mask)
{
	struct list_head *ml;

	list_for_each(ml, tbl->mask_list) {
		struct sw_flow_mask *m;
		m = container_of(ml, struct sw_flow_mask, list);
		if (ovs_sw_flow_mask_equal(mask, m))
			return m;
	}

	return NULL;
}

/**
 * add a new mask into the mask list.
 * The caller needs to make sure that 'mask' is not the same
 * as any masks that are already on the list.
 */
void ovs_sw_flow_mask_insert(struct flow_table *tbl, struct sw_flow_mask *mask)
{
	list_add_rcu(&mask->list, tbl->mask_list);
}

/**
 * Set 'range' fields in the mask to the value of 'val'.
 */
static void ovs_sw_flow_mask_set(struct sw_flow_mask *mask,
		struct sw_flow_key_range *range, u8 val)
{
	u8 *m = (u8 *)&mask->key + range->start;

	mask->range = *range;
	memset(m, val, ovs_sw_flow_mask_size_roundup(mask));
}