Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from David Miller:
 "Another merge window, another pull full of stuff:

   1) Support alternative names for network devices, from Jiri Pirko.

   2) Introduce per-netns netdev notifiers, also from Jiri Pirko.

   3) Support MSG_PEEK in vsock/virtio, from Matias Ezequiel Vara
      Larsen.

   4) Allow compiling out the TLS TOE code, from Jakub Kicinski.

   5) Add several new tracepoints to the kTLS code, also from Jakub.

   6) Support set channels ethtool callback in ena driver, from Sameeh
      Jubran.

   7) New SCTP events SCTP_ADDR_ADDED, SCTP_ADDR_REMOVED,
      SCTP_ADDR_MADE_PRIM, and SCTP_SEND_FAILED_EVENT. From Xin Long.

   8) Add XDP support to mvneta driver, from Lorenzo Bianconi.

   9) Lots of netfilter hw offload fixes, cleanups and enhancements,
      from Pablo Neira Ayuso.

  10) PTP support for aquantia chips, from Egor Pomozov.

  11) Add UDP segmentation offload support to igb, ixgbe, and i40e. From
      Josh Hunt.

  12) Add smart nagle to tipc, from Jon Maloy.

  13) Support L2 field rewrite by TC offloads in bnxt_en, from Venkat
      Duvvuru.

  14) Add a flow mask cache to OVS, from Tonghao Zhang.

  15) Add XDP support to ice driver, from Maciej Fijalkowski.

  16) Add AF_XDP support to ice driver, from Krzysztof Kazimierczak.

  17) Support UDP GSO offload in atlantic driver, from Igor Russkikh.

  18) Support it in stmmac driver too, from Jose Abreu.

  19) Support TIPC encryption and auth, from Tuong Lien.

  20) Introduce BPF trampolines, from Alexei Starovoitov.

  21) Make page_pool API more numa friendly, from Saeed Mahameed.

  22) Introduce route hints to ipv4 and ipv6, from Paolo Abeni.

  23) Add UDP segmentation offload to cxgb4, Rahul Lakkireddy"

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1857 commits)
  libbpf: Fix usage of u32 in userspace code
  mm: Implement no-MMU variant of vmalloc_user_node_flags
  slip: Fix use-after-free Read in slip_open
  net: dsa: sja1105: fix sja1105_parse_rgmii_delays()
  macvlan: schedule bc_work even if error
  enetc: add support Credit Based Shaper(CBS) for hardware offload
  net: phy: add helpers phy_(un)lock_mdio_bus
  mdio_bus: don't use managed reset-controller
  ax88179_178a: add ethtool_op_get_ts_info()
  mlxsw: spectrum_router: Fix use of uninitialized adjacency index
  mlxsw: spectrum_router: After underlay moves, demote conflicting tunnels
  bpf: Simplify __bpf_arch_text_poke poke type handling
  bpf: Introduce BPF_TRACE_x helper for the tracing tests
  bpf: Add bpf_jit_blinding_enabled for !CONFIG_BPF_JIT
  bpf, testing: Add various tail call test cases
  bpf, x86: Emit patchable direct jump as tail call
  bpf: Constant map key tracking for prog array pokes
  bpf: Add poke dependency tracking for prog array maps
  bpf: Add initial poke descriptor table for jit images
  bpf: Move owner type, jited info into array auxiliary data
  ...

10 files changed, 507 insertions, 163 deletions

diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index 1c77f520f474..12936c151cc0 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -200,7 +200,7 @@ static int set_mpls(struct sk_buff *skb, struct sw_flow_key *flow_key,
 	if (err)
 		return err;
 
-	flow_key->mpls.top_lse = lse;
+	flow_key->mpls.lse[0] = lse;
 	return 0;
 }
 
diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index 05249eb45082..df9c80bf621d 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -971,6 +971,8 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 
 	ct = nf_ct_get(skb, &ctinfo);
 	if (ct) {
+		bool add_helper = false;
+
 		/* Packets starting a new connection must be NATted before the
 		 * helper, so that the helper knows about the NAT.  We enforce
 		 * this by delaying both NAT and helper calls for unconfirmed
@@ -988,16 +990,17 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 		}
 
 		/* Userspace may decide to perform a ct lookup without a helper
-		 * specified followed by a (recirculate and) commit with one.
-		 * Therefore, for unconfirmed connections which we will commit,
-		 * we need to attach the helper here.
+		 * specified followed by a (recirculate and) commit with one,
+		 * or attach a helper in a later commit.  Therefore, for
+		 * connections which we will commit, we may need to attach
+		 * the helper here.
 		 */
-		if (!nf_ct_is_confirmed(ct) && info->commit &&
-		    info->helper && !nfct_help(ct)) {
+		if (info->commit && info->helper && !nfct_help(ct)) {
 			int err = __nf_ct_try_assign_helper(ct, info->ct,
 							    GFP_ATOMIC);
 			if (err)
 				return err;
+			add_helper = true;
 
 			/* helper installed, add seqadj if NAT is required */
 			if (info->nat && !nfct_seqadj(ct)) {
@@ -1007,11 +1010,13 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 		}
 
 		/* Call the helper only if:
-		 * - nf_conntrack_in() was executed above ("!cached") for a
-		 *   confirmed connection, or
+		 * - nf_conntrack_in() was executed above ("!cached") or a
+		 *   helper was just attached ("add_helper") for a confirmed
+		 *   connection, or
 		 * - When committing an unconfirmed connection.
 		 */
-		if ((nf_ct_is_confirmed(ct) ? !cached : info->commit) &&
+		if ((nf_ct_is_confirmed(ct) ? !cached || add_helper :
+					      info->commit) &&
 		    ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
 			return -EINVAL;
 		}
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index d8c364d637b1..93d4991ddc1f 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -227,7 +227,8 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key)
 	stats = this_cpu_ptr(dp->stats_percpu);
 
 	/* Look up flow. */
-	flow = ovs_flow_tbl_lookup_stats(&dp->table, key, &n_mask_hit);
+	flow = ovs_flow_tbl_lookup_stats(&dp->table, key, skb_get_hash(skb),
+					 &n_mask_hit);
 	if (unlikely(!flow)) {
 		struct dp_upcall_info upcall;
 
@@ -349,7 +350,8 @@ static size_t upcall_msg_size(const struct dp_upcall_info *upcall_info,
 	size_t size = NLMSG_ALIGN(sizeof(struct ovs_header))
 		+ nla_total_size(hdrlen) /* OVS_PACKET_ATTR_PACKET */
 		+ nla_total_size(ovs_key_attr_size()) /* OVS_PACKET_ATTR_KEY */
-		+ nla_total_size(sizeof(unsigned int)); /* OVS_PACKET_ATTR_LEN */
+		+ nla_total_size(sizeof(unsigned int)) /* OVS_PACKET_ATTR_LEN */
+		+ nla_total_size(sizeof(u64)); /* OVS_PACKET_ATTR_HASH */
 
 	/* OVS_PACKET_ATTR_USERDATA */
 	if (upcall_info->userdata)
@@ -392,6 +394,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb,
 	size_t len;
 	unsigned int hlen;
 	int err, dp_ifindex;
+	u64 hash;
 
 	dp_ifindex = get_dpifindex(dp);
 	if (!dp_ifindex)
@@ -484,23 +487,30 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb,
 	}
 
 	/* Add OVS_PACKET_ATTR_MRU */
-	if (upcall_info->mru) {
-		if (nla_put_u16(user_skb, OVS_PACKET_ATTR_MRU,
-				upcall_info->mru)) {
-			err = -ENOBUFS;
-			goto out;
-		}
-		pad_packet(dp, user_skb);
+	if (upcall_info->mru &&
+	    nla_put_u16(user_skb, OVS_PACKET_ATTR_MRU, upcall_info->mru)) {
+		err = -ENOBUFS;
+		goto out;
 	}
 
 	/* Add OVS_PACKET_ATTR_LEN when packet is truncated */
-	if (cutlen > 0) {
-		if (nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN,
-				skb->len)) {
-			err = -ENOBUFS;
-			goto out;
-		}
-		pad_packet(dp, user_skb);
+	if (cutlen > 0 &&
+	    nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, skb->len)) {
+		err = -ENOBUFS;
+		goto out;
+	}
+
+	/* Add OVS_PACKET_ATTR_HASH */
+	hash = skb_get_hash_raw(skb);
+	if (skb->sw_hash)
+		hash |= OVS_PACKET_HASH_SW_BIT;
+
+	if (skb->l4_hash)
+		hash |= OVS_PACKET_HASH_L4_BIT;
+
+	if (nla_put(user_skb, OVS_PACKET_ATTR_HASH, sizeof (u64), &hash)) {
+		err = -ENOBUFS;
+		goto out;
 	}
 
 	/* Only reserve room for attribute header, packet data is added
@@ -542,6 +552,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
 	struct datapath *dp;
 	struct vport *input_vport;
 	u16 mru = 0;
+	u64 hash;
 	int len;
 	int err;
 	bool log = !a[OVS_PACKET_ATTR_PROBE];
@@ -567,6 +578,14 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
 	}
 	OVS_CB(packet)->mru = mru;
 
+	if (a[OVS_PACKET_ATTR_HASH]) {
+		hash = nla_get_u64(a[OVS_PACKET_ATTR_HASH]);
+
+		__skb_set_hash(packet, hash & 0xFFFFFFFFULL,
+			       !!(hash & OVS_PACKET_HASH_SW_BIT),
+			       !!(hash & OVS_PACKET_HASH_L4_BIT));
+	}
+
 	/* Build an sw_flow for sending this packet. */
 	flow = ovs_flow_alloc();
 	err = PTR_ERR(flow);
@@ -1575,6 +1594,31 @@ static int ovs_dp_change(struct datapath *dp, struct nlattr *a[])
 	return 0;
 }
 
+static int ovs_dp_stats_init(struct datapath *dp)
+{
+	dp->stats_percpu = netdev_alloc_pcpu_stats(struct dp_stats_percpu);
+	if (!dp->stats_percpu)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int ovs_dp_vport_init(struct datapath *dp)
+{
+	int i;
+
+	dp->ports = kmalloc_array(DP_VPORT_HASH_BUCKETS,
+				  sizeof(struct hlist_head),
+				  GFP_KERNEL);
+	if (!dp->ports)
+		return -ENOMEM;
+
+	for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++)
+		INIT_HLIST_HEAD(&dp->ports[i]);
+
+	return 0;
+}
+
 static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr **a = info->attrs;
@@ -1583,7 +1627,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 	struct datapath *dp;
 	struct vport *vport;
 	struct ovs_net *ovs_net;
-	int err, i;
+	int err;
 
 	err = -EINVAL;
 	if (!a[OVS_DP_ATTR_NAME] || !a[OVS_DP_ATTR_UPCALL_PID])
@@ -1596,35 +1640,26 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 	err = -ENOMEM;
 	dp = kzalloc(sizeof(*dp), GFP_KERNEL);
 	if (dp == NULL)
-		goto err_free_reply;
+		goto err_destroy_reply;
 
 	ovs_dp_set_net(dp, sock_net(skb->sk));
 
 	/* Allocate table. */
 	err = ovs_flow_tbl_init(&dp->table);
 	if (err)
-		goto err_free_dp;
+		goto err_destroy_dp;
 
-	dp->stats_percpu = netdev_alloc_pcpu_stats(struct dp_stats_percpu);
-	if (!dp->stats_percpu) {
-		err = -ENOMEM;
+	err = ovs_dp_stats_init(dp);
+	if (err)
 		goto err_destroy_table;
-	}
 
-	dp->ports = kmalloc_array(DP_VPORT_HASH_BUCKETS,
-				  sizeof(struct hlist_head),
-				  GFP_KERNEL);
-	if (!dp->ports) {
-		err = -ENOMEM;
-		goto err_destroy_percpu;
-	}
-
-	for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++)
-		INIT_HLIST_HEAD(&dp->ports[i]);
+	err = ovs_dp_vport_init(dp);
+	if (err)
+		goto err_destroy_stats;
 
 	err = ovs_meters_init(dp);
 	if (err)
-		goto err_destroy_ports_array;
+		goto err_destroy_ports;
 
 	/* Set up our datapath device. */
 	parms.name = nla_data(a[OVS_DP_ATTR_NAME]);
@@ -1656,6 +1691,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 				ovs_dp_reset_user_features(skb, info);
 		}
 
+		ovs_unlock();
 		goto err_destroy_meters;
 	}
 
@@ -1672,17 +1708,16 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 	return 0;
 
 err_destroy_meters:
-	ovs_unlock();
 	ovs_meters_exit(dp);
-err_destroy_ports_array:
+err_destroy_ports:
 	kfree(dp->ports);
-err_destroy_percpu:
+err_destroy_stats:
 	free_percpu(dp->stats_percpu);
 err_destroy_table:
 	ovs_flow_tbl_destroy(&dp->table);
-err_free_dp:
+err_destroy_dp:
 	kfree(dp);
-err_free_reply:
+err_destroy_reply:
 	kfree_skb(reply);
 err:
 	return err;
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index 81e85dde8217..e239a46c2f94 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -139,6 +139,18 @@ struct ovs_net {
 	bool xt_label;
 };
 
+/**
+ * enum ovs_pkt_hash_types - hash info to include with a packet
+ * to send to userspace.
+ * @OVS_PACKET_HASH_SW_BIT: indicates hash was computed in software stack.
+ * @OVS_PACKET_HASH_L4_BIT: indicates hash is a canonical 4-tuple hash
+ * over transport ports.
+ */
+enum ovs_pkt_hash_types {
+	OVS_PACKET_HASH_SW_BIT = (1ULL << 32),
+	OVS_PACKET_HASH_L4_BIT = (1ULL << 33),
+};
+
 extern unsigned int ovs_net_id;
 void ovs_lock(void);
 void ovs_unlock(void);
diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
index 38147e6a20f5..9d375e74b607 100644
--- a/net/openvswitch/flow.c
+++ b/net/openvswitch/flow.c
@@ -637,27 +637,35 @@ static int key_extract_l3l4(struct sk_buff *skb, struct sw_flow_key *key)
 			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 */
 
diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
index b830d5ff7af4..fd8ed766bdd1 100644
--- a/net/openvswitch/flow.h
+++ b/net/openvswitch/flow.h
@@ -30,6 +30,7 @@ enum sw_flow_mac_proto {
 	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
@@ -85,9 +86,6 @@ struct sw_flow_key {
 					 */
 	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. */
@@ -135,6 +133,11 @@ struct sw_flow_key {
 				} 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 {
@@ -166,7 +169,6 @@ struct sw_flow_key_range {
 struct sw_flow_mask {
 	int ref_count;
 	struct rcu_head rcu;
-	struct list_head list;
 	struct sw_flow_key_range range;
 	struct sw_flow_key key;
 };
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index d7559c64795d..65c2e3458ff5 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -424,7 +424,7 @@ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = {
 	[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) },
@@ -1628,10 +1628,25 @@ static int ovs_key_from_nlattrs(struct net *net, struct sw_flow_match *match,
 
 	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);
 	 }
@@ -2114,13 +2129,18 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey,
 		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) ||
@@ -2406,13 +2426,14 @@ static inline void add_nested_action_end(struct sw_flow_actions *sfa,
 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;
@@ -2463,7 +2484,7 @@ static int validate_and_copy_sample(struct net *net, const struct nlattr *attr,
 		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;
@@ -2478,7 +2499,7 @@ static int validate_and_copy_clone(struct net *net,
 				   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;
@@ -2498,7 +2519,7 @@ static int validate_and_copy_clone(struct net *net,
 		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;
 
@@ -2864,6 +2885,7 @@ static int validate_and_copy_check_pkt_len(struct net *net,
 					   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;
@@ -2912,7 +2934,7 @@ static int validate_and_copy_check_pkt_len(struct net *net,
 		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;
@@ -2925,7 +2947,7 @@ static int validate_and_copy_check_pkt_len(struct net *net,
 		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;
@@ -2952,7 +2974,8 @@ static int copy_action(const struct nlattr *from,
 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;
@@ -3065,25 +3088,36 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 			     !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,
@@ -3106,6 +3140,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 
 			err = validate_and_copy_sample(net, a, key, sfa,
 						       eth_type, vlan_tci,
+						       mpls_label_count,
 						       log, last);
 			if (err)
 				return err;
@@ -3176,6 +3211,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 
 			err = validate_and_copy_clone(net, a, key, sfa,
 						      eth_type, vlan_tci,
+						      mpls_label_count,
 						      log, last);
 			if (err)
 				return err;
@@ -3188,8 +3224,9 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 
 			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;
@@ -3219,14 +3256,18 @@ int ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 			 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);
 
diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c
index cf3582c5ed70..5904e93e5765 100644
--- a/net/openvswitch/flow_table.c
+++ b/net/openvswitch/flow_table.c
@@ -34,8 +34,13 @@
 #include <net/ndisc.h>
 
 #define TBL_MIN_BUCKETS		1024
+#define MASK_ARRAY_SIZE_MIN	16
 #define REHASH_INTERVAL		(10 * 60 * HZ)
 
+#define MC_HASH_SHIFT		8
+#define MC_HASH_ENTRIES		(1u << MC_HASH_SHIFT)
+#define MC_HASH_SEGS		((sizeof(uint32_t) * 8) / MC_HASH_SHIFT)
+
 static struct kmem_cache *flow_cache;
 struct kmem_cache *flow_stats_cache __read_mostly;
 
@@ -164,14 +169,133 @@ static struct table_instance *table_instance_alloc(int new_size)
 	return ti;
 }
 
+static struct mask_array *tbl_mask_array_alloc(int size)
+{
+	struct mask_array *new;
+
+	size = max(MASK_ARRAY_SIZE_MIN, size);
+	new = kzalloc(sizeof(struct mask_array) +
+		      sizeof(struct sw_flow_mask *) * size, GFP_KERNEL);
+	if (!new)
+		return NULL;
+
+	new->count = 0;
+	new->max = size;
+
+	return new;
+}
+
+static int tbl_mask_array_realloc(struct flow_table *tbl, int size)
+{
+	struct mask_array *old;
+	struct mask_array *new;
+
+	new = tbl_mask_array_alloc(size);
+	if (!new)
+		return -ENOMEM;
+
+	old = ovsl_dereference(tbl->mask_array);
+	if (old) {
+		int i;
+
+		for (i = 0; i < old->max; i++) {
+			if (ovsl_dereference(old->masks[i]))
+				new->masks[new->count++] = old->masks[i];
+		}
+	}
+
+	rcu_assign_pointer(tbl->mask_array, new);
+	kfree_rcu(old, rcu);
+
+	return 0;
+}
+
+static int tbl_mask_array_add_mask(struct flow_table *tbl,
+				   struct sw_flow_mask *new)
+{
+	struct mask_array *ma = ovsl_dereference(tbl->mask_array);
+	int err, ma_count = READ_ONCE(ma->count);
+
+	if (ma_count >= ma->max) {
+		err = tbl_mask_array_realloc(tbl, ma->max +
+					      MASK_ARRAY_SIZE_MIN);
+		if (err)
+			return err;
+
+		ma = ovsl_dereference(tbl->mask_array);
+	}
+
+	BUG_ON(ovsl_dereference(ma->masks[ma_count]));
+
+	rcu_assign_pointer(ma->masks[ma_count], new);
+	WRITE_ONCE(ma->count, ma_count +1);
+
+	return 0;
+}
+
+static void tbl_mask_array_del_mask(struct flow_table *tbl,
+				    struct sw_flow_mask *mask)
+{
+	struct mask_array *ma = ovsl_dereference(tbl->mask_array);
+	int i, ma_count = READ_ONCE(ma->count);
+
+	/* Remove the deleted mask pointers from the array */
+	for (i = 0; i < ma_count; i++) {
+		if (mask == ovsl_dereference(ma->masks[i]))
+			goto found;
+	}
+
+	BUG();
+	return;
+
+found:
+	WRITE_ONCE(ma->count, ma_count -1);
+
+	rcu_assign_pointer(ma->masks[i], ma->masks[ma_count -1]);
+	RCU_INIT_POINTER(ma->masks[ma_count -1], NULL);
+
+	kfree_rcu(mask, rcu);
+
+	/* Shrink the mask array if necessary. */
+	if (ma->max >= (MASK_ARRAY_SIZE_MIN * 2) &&
+	    ma_count <= (ma->max / 3))
+		tbl_mask_array_realloc(tbl, ma->max / 2);
+}
+
+/* Remove 'mask' from the mask list, if it is not needed any more. */
+static void flow_mask_remove(struct flow_table *tbl, struct sw_flow_mask *mask)
+{
+	if (mask) {
+		/* ovs-lock is required to protect mask-refcount and
+		 * mask list.
+		 */
+		ASSERT_OVSL();
+		BUG_ON(!mask->ref_count);
+		mask->ref_count--;
+
+		if (!mask->ref_count)
+			tbl_mask_array_del_mask(tbl, mask);
+	}
+}
+
 int ovs_flow_tbl_init(struct flow_table *table)
 {
 	struct table_instance *ti, *ufid_ti;
+	struct mask_array *ma;
 
-	ti = table_instance_alloc(TBL_MIN_BUCKETS);
+	table->mask_cache = __alloc_percpu(sizeof(struct mask_cache_entry) *
+					   MC_HASH_ENTRIES,
+					   __alignof__(struct mask_cache_entry));
+	if (!table->mask_cache)
+		return -ENOMEM;
 
+	ma = tbl_mask_array_alloc(MASK_ARRAY_SIZE_MIN);
+	if (!ma)
+		goto free_mask_cache;
+
+	ti = table_instance_alloc(TBL_MIN_BUCKETS);
 	if (!ti)
-		return -ENOMEM;
+		goto free_mask_array;
 
 	ufid_ti = table_instance_alloc(TBL_MIN_BUCKETS);
 	if (!ufid_ti)
@@ -179,7 +303,7 @@ int ovs_flow_tbl_init(struct flow_table *table)
 
 	rcu_assign_pointer(table->ti, ti);
 	rcu_assign_pointer(table->ufid_ti, ufid_ti);
-	INIT_LIST_HEAD(&table->mask_list);
+	rcu_assign_pointer(table->mask_array, ma);
 	table->last_rehash = jiffies;
 	table->count = 0;
 	table->ufid_count = 0;
@@ -187,6 +311,10 @@ int ovs_flow_tbl_init(struct flow_table *table)
 
 free_ti:
 	__table_instance_destroy(ti);
+free_mask_array:
+	kfree(ma);
+free_mask_cache:
+	free_percpu(table->mask_cache);
 	return -ENOMEM;
 }
 
@@ -197,7 +325,28 @@ static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu)
 	__table_instance_destroy(ti);
 }
 
-static void table_instance_destroy(struct table_instance *ti,
+static void table_instance_flow_free(struct flow_table *table,
+				  struct table_instance *ti,
+				  struct table_instance *ufid_ti,
+				  struct sw_flow *flow,
+				  bool count)
+{
+	hlist_del_rcu(&flow->flow_table.node[ti->node_ver]);
+	if (count)
+		table->count--;
+
+	if (ovs_identifier_is_ufid(&flow->id)) {
+		hlist_del_rcu(&flow->ufid_table.node[ufid_ti->node_ver]);
+
+		if (count)
+			table->ufid_count--;
+	}
+
+	flow_mask_remove(table, flow->mask);
+}
+
+static void table_instance_destroy(struct flow_table *table,
+				   struct table_instance *ti,
 				   struct table_instance *ufid_ti,
 				   bool deferred)
 {
@@ -214,13 +363,12 @@ static void table_instance_destroy(struct table_instance *ti,
 		struct sw_flow *flow;
 		struct hlist_head *head = &ti->buckets[i];
 		struct hlist_node *n;
-		int ver = ti->node_ver;
-		int ufid_ver = ufid_ti->node_ver;
 
-		hlist_for_each_entry_safe(flow, n, head, flow_table.node[ver]) {
-			hlist_del_rcu(&flow->flow_table.node[ver]);
-			if (ovs_identifier_is_ufid(&flow->id))
-				hlist_del_rcu(&flow->ufid_table.node[ufid_ver]);
+		hlist_for_each_entry_safe(flow, n, head,
+					  flow_table.node[ti->node_ver]) {
+
+			table_instance_flow_free(table, ti, ufid_ti,
+						 flow, false);
 			ovs_flow_free(flow, deferred);
 		}
 	}
@@ -243,7 +391,9 @@ void ovs_flow_tbl_destroy(struct flow_table *table)
 	struct table_instance *ti = rcu_dereference_raw(table->ti);
 	struct table_instance *ufid_ti = rcu_dereference_raw(table->ufid_ti);
 
-	table_instance_destroy(ti, ufid_ti, false);
+	free_percpu(table->mask_cache);
+	kfree_rcu(rcu_dereference_raw(table->mask_array), rcu);
+	table_instance_destroy(table, ti, ufid_ti, false);
 }
 
 struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *ti,
@@ -359,7 +509,7 @@ int ovs_flow_tbl_flush(struct flow_table *flow_table)
 	flow_table->count = 0;
 	flow_table->ufid_count = 0;
 
-	table_instance_destroy(old_ti, old_ufid_ti, true);
+	table_instance_destroy(flow_table, old_ti, old_ufid_ti, true);
 	return 0;
 
 err_free_ti:
@@ -370,13 +520,10 @@ err_free_ti:
 static u32 flow_hash(const struct sw_flow_key *key,
 		     const struct sw_flow_key_range *range)
 {
-	int key_start = range->start;
-	int key_end = range->end;
-	const u32 *hash_key = (const u32 *)((const u8 *)key + key_start);
-	int hash_u32s = (key_end - key_start) >> 2;
+	const u32 *hash_key = (const u32 *)((const u8 *)key + range->start);
 
 	/* Make sure number of hash bytes are multiple of u32. */
-	BUILD_BUG_ON(sizeof(long) % sizeof(u32));
+	int hash_u32s = range_n_bytes(range) >> 2;
 
 	return jhash2(hash_key, hash_u32s, 0);
 }
@@ -425,7 +572,8 @@ static bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow,
 
 static struct sw_flow *masked_flow_lookup(struct table_instance *ti,
 					  const struct sw_flow_key *unmasked,
-					  const struct sw_flow_mask *mask)
+					  const struct sw_flow_mask *mask,
+					  u32 *n_mask_hit)
 {
 	struct sw_flow *flow;
 	struct hlist_head *head;
@@ -435,6 +583,8 @@ static struct sw_flow *masked_flow_lookup(struct table_instance *ti,
 	ovs_flow_mask_key(&masked_key, unmasked, false, mask);
 	hash = flow_hash(&masked_key, &mask->range);
 	head = find_bucket(ti, hash);
+	(*n_mask_hit)++;
+
 	hlist_for_each_entry_rcu(flow, head, flow_table.node[ti->node_ver]) {
 		if (flow->mask == mask && flow->flow_table.hash == hash &&
 		    flow_cmp_masked_key(flow, &masked_key, &mask->range))
@@ -443,46 +593,147 @@ static struct sw_flow *masked_flow_lookup(struct table_instance *ti,
 	return NULL;
 }
 
-struct sw_flow *ovs_flow_tbl_lookup_stats(struct flow_table *tbl,
-				    const struct sw_flow_key *key,
-				    u32 *n_mask_hit)
+/* Flow lookup does full lookup on flow table. It starts with
+ * mask from index passed in *index.
+ */
+static struct sw_flow *flow_lookup(struct flow_table *tbl,
+				   struct table_instance *ti,
+				   struct mask_array *ma,
+				   const struct sw_flow_key *key,
+				   u32 *n_mask_hit,
+				   u32 *index)
 {
-	struct table_instance *ti = rcu_dereference_ovsl(tbl->ti);
+	struct sw_flow *flow;
 	struct sw_flow_mask *mask;
+	int i;
+
+	if (likely(*index < ma->max)) {
+		mask = rcu_dereference_ovsl(ma->masks[*index]);
+		if (mask) {
+			flow = masked_flow_lookup(ti, key, mask, n_mask_hit);
+			if (flow)
+				return flow;
+		}
+	}
+
+	for (i = 0; i < ma->max; i++)  {
+
+		if (i == *index)
+			continue;
+
+		mask = rcu_dereference_ovsl(ma->masks[i]);
+		if (unlikely(!mask))
+			break;
+
+		flow = masked_flow_lookup(ti, key, mask, n_mask_hit);
+		if (flow) { /* Found */
+			*index = i;
+			return flow;
+		}
+	}
+
+	return NULL;
+}
+
+/*
+ * mask_cache maps flow to probable mask. This cache is not tightly
+ * coupled cache, It means updates to  mask list can result in inconsistent
+ * cache entry in mask cache.
+ * This is per cpu cache and is divided in MC_HASH_SEGS segments.
+ * In case of a hash collision the entry is hashed in next segment.
+ * */
+struct sw_flow *ovs_flow_tbl_lookup_stats(struct flow_table *tbl,
+					  const struct sw_flow_key *key,
+					  u32 skb_hash,
+					  u32 *n_mask_hit)
+{
+	struct mask_array *ma = rcu_dereference(tbl->mask_array);
+	struct table_instance *ti = rcu_dereference(tbl->ti);
+	struct mask_cache_entry *entries, *ce;
 	struct sw_flow *flow;
+	u32 hash;
+	int seg;
 
 	*n_mask_hit = 0;
-	list_for_each_entry_rcu(mask, &tbl->mask_list, list) {
-		(*n_mask_hit)++;
-		flow = masked_flow_lookup(ti, key, mask);
-		if (flow)  /* Found */
+	if (unlikely(!skb_hash)) {
+		u32 mask_index = 0;
+
+		return flow_lookup(tbl, ti, ma, key, n_mask_hit, &mask_index);
+	}
+
+	/* Pre and post recirulation flows usually have the same skb_hash
+	 * value. To avoid hash collisions, rehash the 'skb_hash' with
+	 * 'recirc_id'.  */
+	if (key->recirc_id)
+		skb_hash = jhash_1word(skb_hash, key->recirc_id);
+
+	ce = NULL;
+	hash = skb_hash;
+	entries = this_cpu_ptr(tbl->mask_cache);
+
+	/* Find the cache entry 'ce' to operate on. */
+	for (seg = 0; seg < MC_HASH_SEGS; seg++) {
+		int index = hash & (MC_HASH_ENTRIES - 1);
+		struct mask_cache_entry *e;
+
+		e = &entries[index];
+		if (e->skb_hash == skb_hash) {
+			flow = flow_lookup(tbl, ti, ma, key, n_mask_hit,
+					   &e->mask_index);
+			if (!flow)
+				e->skb_hash = 0;
 			return flow;
+		}
+
+		if (!ce || e->skb_hash < ce->skb_hash)
+			ce = e;  /* A better replacement cache candidate. */
+
+		hash >>= MC_HASH_SHIFT;
 	}
-	return NULL;
+
+	/* Cache miss, do full lookup. */
+	flow = flow_lookup(tbl, ti, ma, key, n_mask_hit, &ce->mask_index);
+	if (flow)
+		ce->skb_hash = skb_hash;
+
+	return flow;
 }
 
 struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *tbl,
 				    const struct sw_flow_key *key)
 {
+	struct table_instance *ti = rcu_dereference_ovsl(tbl->ti);
+	struct mask_array *ma = rcu_dereference_ovsl(tbl->mask_array);
 	u32 __always_unused n_mask_hit;
+	u32 index = 0;
 
-	return ovs_flow_tbl_lookup_stats(tbl, key, &n_mask_hit);
+	return flow_lookup(tbl, ti, ma, key, &n_mask_hit, &index);
 }
 
 struct sw_flow *ovs_flow_tbl_lookup_exact(struct flow_table *tbl,
 					  const struct sw_flow_match *match)
 {
-	struct table_instance *ti = rcu_dereference_ovsl(tbl->ti);
-	struct sw_flow_mask *mask;
-	struct sw_flow *flow;
+	struct mask_array *ma = ovsl_dereference(tbl->mask_array);
+	int i;
 
 	/* Always called under ovs-mutex. */
-	list_for_each_entry(mask, &tbl->mask_list, list) {
-		flow = masked_flow_lookup(ti, match->key, mask);
+	for (i = 0; i < ma->max; i++) {
+		struct table_instance *ti = rcu_dereference_ovsl(tbl->ti);
+		u32 __always_unused n_mask_hit;
+		struct sw_flow_mask *mask;
+		struct sw_flow *flow;
+
+		mask = ovsl_dereference(ma->masks[i]);
+		if (!mask)
+			continue;
+
+		flow = masked_flow_lookup(ti, match->key, mask, &n_mask_hit);
 		if (flow && ovs_identifier_is_key(&flow->id) &&
-		    ovs_flow_cmp_unmasked_key(flow, match))
+		    ovs_flow_cmp_unmasked_key(flow, match)) {
 			return flow;
+		}
 	}
+
 	return NULL;
 }
 
@@ -528,13 +779,8 @@ struct sw_flow *ovs_flow_tbl_lookup_ufid(struct flow_table *tbl,
 
 int ovs_flow_tbl_num_masks(const struct flow_table *table)
 {
-	struct sw_flow_mask *mask;
-	int num = 0;
-
-	list_for_each_entry(mask, &table->mask_list, list)
-		num++;
-
-	return num;
+	struct mask_array *ma = rcu_dereference_ovsl(table->mask_array);
+	return READ_ONCE(ma->count);
 }
 
 static struct table_instance *table_instance_expand(struct table_instance *ti,
@@ -543,24 +789,6 @@ static struct table_instance *table_instance_expand(struct table_instance *ti,
 	return table_instance_rehash(ti, ti->n_buckets * 2, ufid);
 }
 
-/* Remove 'mask' from the mask list, if it is not needed any more. */
-static void flow_mask_remove(struct flow_table *tbl, struct sw_flow_mask *mask)
-{
-	if (mask) {
-		/* ovs-lock is required to protect mask-refcount and
-		 * mask list.
-		 */
-		ASSERT_OVSL();
-		BUG_ON(!mask->ref_count);
-		mask->ref_count--;
-
-		if (!mask->ref_count) {
-			list_del_rcu(&mask->list);
-			kfree_rcu(mask, rcu);
-		}
-	}
-}
-
 /* Must be called with OVS mutex held. */
 void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow)
 {
@@ -568,17 +796,7 @@ void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow)
 	struct table_instance *ufid_ti = ovsl_dereference(table->ufid_ti);
 
 	BUG_ON(table->count == 0);
-	hlist_del_rcu(&flow->flow_table.node[ti->node_ver]);
-	table->count--;
-	if (ovs_identifier_is_ufid(&flow->id)) {
-		hlist_del_rcu(&flow->ufid_table.node[ufid_ti->node_ver]);
-		table->ufid_count--;
-	}
-
-	/* RCU delete the mask. 'flow->mask' is not NULLed, as it should be
-	 * accessible as long as the RCU read lock is held.
-	 */
-	flow_mask_remove(table, flow->mask);
+	table_instance_flow_free(table, ti, ufid_ti, flow, true);
 }
 
 static struct sw_flow_mask *mask_alloc(void)
@@ -606,13 +824,16 @@ static bool mask_equal(const struct sw_flow_mask *a,
 static struct sw_flow_mask *flow_mask_find(const struct flow_table *tbl,
 					   const struct sw_flow_mask *mask)
 {
-	struct list_head *ml;
+	struct mask_array *ma;
+	int i;
+
+	ma = ovsl_dereference(tbl->mask_array);
+	for (i = 0; i < ma->max; i++) {
+		struct sw_flow_mask *t;
+		t = ovsl_dereference(ma->masks[i]);
 
-	list_for_each(ml, &tbl->mask_list) {
-		struct sw_flow_mask *m;
-		m = container_of(ml, struct sw_flow_mask, list);
-		if (mask_equal(mask, m))
-			return m;
+		if (t && mask_equal(mask, t))
+			return t;
 	}
 
 	return NULL;
@@ -623,6 +844,7 @@ static int flow_mask_insert(struct flow_table *tbl, struct sw_flow *flow,
 			    const struct sw_flow_mask *new)
 {
 	struct sw_flow_mask *mask;
+
 	mask = flow_mask_find(tbl, new);
 	if (!mask) {
 		/* Allocate a new mask if none exsits. */
@@ -631,7 +853,12 @@ static int flow_mask_insert(struct flow_table *tbl, struct sw_flow *flow,
 			return -ENOMEM;
 		mask->key = new->key;
 		mask->range = new->range;
-		list_add_rcu(&mask->list, &tbl->mask_list);
+
+		/* Add mask to mask-list. */
+		if (tbl_mask_array_add_mask(tbl, mask)) {
+			kfree(mask);
+			return -ENOMEM;
+		}
 	} else {
 		BUG_ON(!mask->ref_count);
 		mask->ref_count++;
diff --git a/net/openvswitch/flow_table.h b/net/openvswitch/flow_table.h
index bc52045b63ff..8a5cea6ae111 100644
--- a/net/openvswitch/flow_table.h
+++ b/net/openvswitch/flow_table.h
@@ -22,6 +22,17 @@
 
 #include "flow.h"
 
+struct mask_cache_entry {
+	u32 skb_hash;
+	u32 mask_index;
+};
+
+struct mask_array {
+	struct rcu_head rcu;
+	int count, max;
+	struct sw_flow_mask __rcu *masks[];
+};
+
 struct table_instance {
 	struct hlist_head *buckets;
 	unsigned int n_buckets;
@@ -34,7 +45,8 @@ struct table_instance {
 struct flow_table {
 	struct table_instance __rcu *ti;
 	struct table_instance __rcu *ufid_ti;
-	struct list_head mask_list;
+	struct mask_cache_entry __percpu *mask_cache;
+	struct mask_array __rcu *mask_array;
 	unsigned long last_rehash;
 	unsigned int count;
 	unsigned int ufid_count;
@@ -60,8 +72,9 @@ int  ovs_flow_tbl_num_masks(const struct flow_table *table);
 struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *table,
 				       u32 *bucket, u32 *idx);
 struct sw_flow *ovs_flow_tbl_lookup_stats(struct flow_table *,
-				    const struct sw_flow_key *,
-				    u32 *n_mask_hit);
+					  const struct sw_flow_key *,
+					  u32 skb_hash,
+					  u32 *n_mask_hit);
 struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *,
 				    const struct sw_flow_key *);
 struct sw_flow *ovs_flow_tbl_lookup_exact(struct flow_table *tbl,
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 3fc38d16c456..5da9392b03d6 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -403,8 +403,9 @@ u32 ovs_vport_find_upcall_portid(const struct vport *vport, struct sk_buff *skb)
 
 	ids = rcu_dereference(vport->upcall_portids);
 
-	if (ids->n_ids == 1 && ids->ids[0] == 0)
-		return 0;
+	/* If there is only one portid, select it in the fast-path. */
+	if (ids->n_ids == 1)
+		return ids->ids[0];
 
 	hash = skb_get_hash(skb);
 	ids_index = hash - ids->n_ids * reciprocal_divide(hash, ids->rn_ids);