diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index 8a84de3..1a23c03 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -276,6 +276,7 @@ int nf_flow_table_init(struct nf_flowtable *flow_table);
 void nf_flow_table_free(struct nf_flowtable *flow_table);
 
 void flow_offload_teardown(struct flow_offload *flow);
+void flow_offload_teardown_by_tuple(struct flow_offload_tuple *tuple);
 
 int nf_flow_table_iterate(struct nf_flowtable *flow_table,
 			  void (*iter)(struct flow_offload *flow, void *data),
diff --git a/include/uapi/linux/netfilter/nfnetlink.h b/include/uapi/linux/netfilter/nfnetlink.h
index 5bc960f..603d9c0 100644
--- a/include/uapi/linux/netfilter/nfnetlink.h
+++ b/include/uapi/linux/netfilter/nfnetlink.h
@@ -60,7 +60,8 @@ struct nfgenmsg {
 #define NFNL_SUBSYS_CTHELPER		9
 #define NFNL_SUBSYS_NFTABLES		10
 #define NFNL_SUBSYS_NFT_COMPAT		11
-#define NFNL_SUBSYS_COUNT		12
+#define NFNL_SUBSYS_FLOWTABLE		12
+#define NFNL_SUBSYS_COUNT		13
 
 /* Reserved control nfnetlink messages */
 #define NFNL_MSG_BATCH_BEGIN		NLMSG_MIN_TYPE
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 5d690ab..8ec87aa 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -708,6 +708,15 @@ config NF_FLOW_TABLE
 
 	  To compile it as a module, choose M here.
 
+config NF_FLOW_TABLE_NETLINK
+	tristate "Netfilter flow table netlink module"
+	depends on NETFILTER_INGRESS
+	depends on NF_CONNTRACK
+	help
+	  This option adds the flow table core infrastructure.
+
+	  To compile it as a module, choose M here.
+
 config NETFILTER_XTABLES
 	tristate "Netfilter Xtables support (required for ip_tables)"
 	default m if NETFILTER_ADVANCED=n
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index d93a121..fa6ffb1 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -124,6 +124,7 @@ nf_flow_table-objs		:= nf_flow_table_core.o nf_flow_table_ip.o \
 				   nf_flow_table_offload.o
 
 obj-$(CONFIG_NF_FLOW_TABLE_INET) += nf_flow_table_inet.o
+obj-$(CONFIG_NF_FLOW_TABLE_NETLINK)	+= nf_flow_table_netlink.o
 
 # generic X tables
 obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index 1036558..a0f52f6 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -373,6 +373,29 @@ void flow_offload_teardown(struct flow_offload *flow)
 }
 EXPORT_SYMBOL_GPL(flow_offload_teardown);
 
+void flow_offload_teardown_by_tuple(struct flow_offload_tuple *tuple)
+{
+	struct net_device *netdev;
+	struct nf_flowtable *flowtable;
+	struct flow_offload_tuple_rhash *tuplehash;
+	struct flow_offload *flow;
+	int dir;
+
+	list_for_each_entry(flowtable, &flowtables, list) {
+		for_each_netdev(&init_net, netdev) {
+			tuple->iifidx = netdev->ifindex;
+			tuplehash = flow_offload_lookup(flowtable, tuple);
+			if (!tuplehash)
+				continue;
+
+			dir = tuplehash->tuple.dir;
+			flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]);
+			flow_offload_teardown(flow);
+		}
+	};
+}
+EXPORT_SYMBOL_GPL(flow_offload_teardown_by_tuple);
+
 struct flow_offload_tuple_rhash *
 flow_offload_lookup(struct nf_flowtable *flow_table,
 		    struct flow_offload_tuple *tuple)
diff --git a/net/netfilter/nf_flow_table_netlink.c b/net/netfilter/nf_flow_table_netlink.c
new file mode 100644
index 0000000..f05f29e
--- /dev/null
+++ b/net/netfilter/nf_flow_table_netlink.c
@@ -0,0 +1,239 @@
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/netfilter.h>
+#include <linux/netlink.h>
+#include <net/netlink.h>
+#include <net/ip.h>
+#include <linux/netfilter/nfnetlink.h>
+#include <net/netfilter/nf_flow_table.h>
+
+enum ft_netlink_msg_types {
+	FT_MSG_DEL,
+	FT_MSG_ADD,
+	FT_MSG_FLUSH,
+	FT_MSG_MAX
+};
+
+enum ftattr_type {
+	FTA_UNSPEC,
+	FTA_TUPLE,
+	__FTA_MAX
+};
+#define FTA_MAX (__FTA_MAX - 1)
+
+enum ftattr_tuple {
+	FTA_TUPLE_UNSPEC,
+	FTA_TUPLE_IP,
+	FTA_TUPLE_PROTO,
+	FTA_TUPLE_ZONE,
+	__FTA_TUPLE_MAX
+};
+#define FTA_TUPLE_MAX (__FTA_TUPLE_MAX - 1)
+
+enum ftattr_ip {
+	FTA_IP_UNSPEC,
+	FTA_IP_V4_SRC,
+	FTA_IP_V4_DST,
+	__FTA_IP_MAX
+};
+#define FTA_IP_MAX (__FTA_IP_MAX - 1)
+
+enum ftattr_l4proto {
+	FTA_PROTO_UNSPEC,
+	FTA_PROTO_NUM,
+	FTA_PROTO_SPORT,
+	FTA_PROTO_DPORT,
+	__FTA_PROTO_MAX
+};
+#define FTA_PROTO_MAX (__FTA_PROTO_MAX - 1)
+
+static const struct nla_policy tuple_nla_policy[FTA_TUPLE_MAX + 1] = {
+	[FTA_TUPLE_IP]		= { .type = NLA_NESTED },
+	[FTA_TUPLE_PROTO]	= { .type = NLA_NESTED },
+	[FTA_TUPLE_ZONE]	= { .type = NLA_U16 },
+};
+
+static const struct nla_policy ip_nla_policy[FTA_IP_MAX + 1] = {
+	[FTA_IP_V4_SRC]		= { .type = NLA_U32 },
+	[FTA_IP_V4_DST]		= { .type = NLA_U32 },
+};
+
+static const struct nla_policy l4proto_nla_policy[FTA_PROTO_MAX + 1] = {
+	[FTA_PROTO_NUM]		= { .type = NLA_U8 },
+	[FTA_PROTO_SPORT]	= {.type = NLA_U16},
+	[FTA_PROTO_DPORT]	= {.type = NLA_U16},
+};
+
+static inline int ftnetlink_parse_tuple_ip(struct nlattr *attr,
+					   struct flow_offload_tuple *tuple)
+{
+	struct nlattr *tb[FTA_IP_MAX+1];
+	int err;
+
+	err = nla_parse_nested_deprecated(tb, FTA_IP_MAX, attr, ip_nla_policy, NULL);
+
+	if (err < 0)
+		return err;
+
+	switch (tuple->l3proto) {
+	case NFPROTO_IPV4:
+		if (!tb[FTA_IP_V4_SRC] || !tb[FTA_IP_V4_DST])
+			return -EINVAL;
+
+		tuple->src_v4.s_addr = nla_get_in_addr(tb[FTA_IP_V4_SRC]);
+		tuple->dst_v4.s_addr = nla_get_in_addr(tb[FTA_IP_V4_DST]);
+	}
+
+	return err;
+}
+
+static inline int ftnetlink_parse_tuple_proto(struct nlattr *attr,
+					      struct flow_offload_tuple *tuple)
+{
+	struct nlattr *tb[FTA_PROTO_MAX+1];
+	int err;
+
+	err = nla_parse_nested_deprecated(tb, FTA_PROTO_MAX, attr, l4proto_nla_policy, NULL);
+
+	if(err < 0)
+		return err;
+
+	if (!tb[FTA_PROTO_NUM] || !tb[FTA_PROTO_SPORT] || !tb[FTA_PROTO_DPORT])
+		return -EINVAL;
+
+	tuple->l4proto = nla_get_u8(tb[FTA_PROTO_NUM]);
+	tuple->src_port = nla_get_u16(tb[FTA_PROTO_SPORT]);
+	tuple->dst_port = nla_get_u16(tb[FTA_PROTO_DPORT]);
+
+	return err;
+}
+
+static int ftnetlink_parse_tuple(const struct nlattr * const cda[],
+				 struct flow_offload_tuple *tuple,
+				 int attrtype, int l3proto)
+{
+	struct nlattr *tb[FTA_TUPLE_MAX+1];
+	int err;
+
+	memset(tuple, 0, sizeof(*tuple));
+
+	err = nla_parse_nested_deprecated(tb, FTA_TUPLE_MAX, cda[attrtype], tuple_nla_policy, NULL);
+	if (err < 0)
+		return err;
+
+	if (!tb[FTA_TUPLE_IP])
+		return -EINVAL;
+
+	/* parse IP */
+	tuple->l3proto = l3proto;
+	err = ftnetlink_parse_tuple_ip(tb[FTA_TUPLE_IP], tuple);
+	if (err < 0)
+		return err;
+
+	/* parse proto */
+	if (!tb[FTA_TUPLE_PROTO])
+		return -EINVAL;
+	err = ftnetlink_parse_tuple_proto(tb[FTA_TUPLE_PROTO], tuple);
+
+	if (err >= 0)
+		printk("tuple info:sip=%pI4,dip=%pI4 proto=%d "
+		       "sport=%d dport=%d\n",
+		       &tuple->src_v4, &tuple->dst_v4, tuple->l4proto,
+		       ntohs(tuple->src_port), ntohs(tuple->dst_port));
+
+	return err;
+}
+
+static int ftnetlink_del_nf_flow(struct net *net, struct sock *ftnl, struct sk_buff *skb,
+				 const struct nlmsghdr *nlh,
+				 const struct nlattr * const cda[],
+				 struct netlink_ext_ack *extack)
+{
+	struct net_device *dev = skb->dev;
+	struct flow_offload_tuple tuple;
+	int err = -1;
+	struct nfgenmsg *nfmsg = nlmsg_data(nlh);
+	u_int8_t u3 = nfmsg->nfgen_family;
+
+	/* parse tuple */
+	if(!cda[FTA_TUPLE])
+		return -EINVAL;
+
+	err = ftnetlink_parse_tuple(cda, &tuple, FTA_TUPLE, u3);
+	if (err < 0)
+		return err;
+
+	/* teardown the flow */
+	flow_offload_teardown_by_tuple(&tuple);
+
+	return 0;
+}
+
+static int ftnetlink_add_nf_flow(struct net *net, struct sock *ftnl, struct sk_buff *skb,
+				 const struct nlmsghdr *nlh,
+				 const struct nlattr * const cda[],
+				 struct netlink_ext_ack *extack)
+{
+	return 0;
+}
+
+static int ftnetlink_flush_table(struct net *net, struct sock *ftnl, struct sk_buff *skb,
+				 const struct nlmsghdr *nlh,
+				 const struct nlattr * const cda[],
+				 struct netlink_ext_ack *extack)
+{
+	struct net_device *dev = skb->dev;
+
+	nf_flow_table_cleanup(dev);
+
+	return 0;
+}
+
+static const struct nla_policy ft_nla_policy[FTA_MAX + 1] = {
+	[FTA_TUPLE] = { .type = NLA_NESTED },
+};
+
+static const struct nfnl_callback flow_table_cb[FT_MSG_MAX] = {
+	[FT_MSG_DEL] = {
+		.call = ftnetlink_del_nf_flow,
+		.attr_count = FTA_MAX,
+		.policy = ft_nla_policy
+	},
+	[FT_MSG_ADD] = {
+		.call = ftnetlink_add_nf_flow,
+		.attr_count = FTA_MAX,
+		.policy = ft_nla_policy
+	},
+	[FT_MSG_FLUSH] = {
+		.call = ftnetlink_flush_table,
+		.attr_count = FTA_MAX,
+		.policy = ft_nla_policy
+	},
+};
+
+static const struct nfnetlink_subsystem ftnl_subsys = {
+	.name = "flowtable",
+	.subsys_id = NFNL_SUBSYS_FLOWTABLE,
+	.cb_count = FT_MSG_MAX,
+	.cb = flow_table_cb,
+};
+
+static int __init ftnetlink_init(void)
+{
+	int ret;
+
+	ret = nfnetlink_subsys_register(&ftnl_subsys);
+
+	return ret;
+}
+
+static void ftnetlink_exit(void)
+{
+	nfnetlink_subsys_unregister(&ftnl_subsys);
+}
+
+MODULE_LICENSE("GPL");
+module_init(ftnetlink_init);
+module_exit(ftnetlink_exit);