netfilter: nf_tables: flow offload expression

Add new instruction for the nf_tables VM that allows us to specify what
flows are offloaded into a given flow table via name. This new
instruction creates the flow entry and adds it to the flow table.

Only established flows, ie. we have seen traffic in both directions, are
added to the flow table. You can still decide to offload entries at a
later stage via packet counting or checking the ct status in case you
want to offload assured conntracks.

This new extension depends on the conntrack subsystem.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

include/uapi/linux/netfilter/nf_tables.h

@@ -957,6 +957,17 @@ enum nft_ct_attributes {
 };
 #define NFTA_CT_MAX		(__NFTA_CT_MAX - 1)
 
+/**
+ * enum nft_flow_attributes - ct offload expression attributes
+ * @NFTA_FLOW_TABLE_NAME: flow table name (NLA_STRING)
+ */
+enum nft_offload_attributes {
+	NFTA_FLOW_UNSPEC,
+	NFTA_FLOW_TABLE_NAME,
+	__NFTA_FLOW_MAX,
+};
+#define NFTA_FLOW_MAX		(__NFTA_FLOW_MAX - 1)
+
 enum nft_limit_type {
 	NFT_LIMIT_PKTS,
 	NFT_LIMIT_PKT_BYTES

net/netfilter/Kconfig

@@ -505,6 +505,13 @@ config NFT_CT
 	  This option adds the "ct" expression that you can use to match
 	  connection tracking information such as the flow state.
 
+config NFT_FLOW_OFFLOAD
+	depends on NF_CONNTRACK
+	tristate "Netfilter nf_tables hardware flow offload module"
+	help
+	  This option adds the "flow_offload" expression that you can use to
+	  choose what flows are placed into the hardware.
+
 config NFT_SET_RBTREE
 	tristate "Netfilter nf_tables rbtree set module"
 	help

net/netfilter/Makefile

@@ -86,6 +86,7 @@ obj-$(CONFIG_NFT_META)		+= nft_meta.o
 obj-$(CONFIG_NFT_RT)		+= nft_rt.o
 obj-$(CONFIG_NFT_NUMGEN)	+= nft_numgen.o
 obj-$(CONFIG_NFT_CT)		+= nft_ct.o
+obj-$(CONFIG_NFT_FLOW_OFFLOAD)	+= nft_flow_offload.o
 obj-$(CONFIG_NFT_LIMIT)		+= nft_limit.o
 obj-$(CONFIG_NFT_NAT)		+= nft_nat.o
 obj-$(CONFIG_NFT_OBJREF)	+= nft_objref.o

net/netfilter/nft_flow_offload.c

+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/netlink.h>
+#include <linux/netfilter.h>
+#include <linux/workqueue.h>
+#include <linux/spinlock.h>
+#include <linux/netfilter/nf_tables.h>
+#include <net/ip.h> /* for ipv4 options. */
+#include <net/netfilter/nf_tables.h>
+#include <net/netfilter/nf_tables_core.h>
+#include <net/netfilter/nf_conntrack_core.h>
+#include <linux/netfilter/nf_conntrack_common.h>
+#include <net/netfilter/nf_flow_table.h>
+
+struct nft_flow_offload {
+	struct nft_flowtable	*flowtable;
+};
+
+static int nft_flow_route(const struct nft_pktinfo *pkt,
+			  const struct nf_conn *ct,
+			  struct nf_flow_route *route,
+			  enum ip_conntrack_dir dir)
+{
+	struct dst_entry *this_dst = skb_dst(pkt->skb);
+	struct dst_entry *other_dst = NULL;
+	struct flowi fl;
+
+	memset(&fl, 0, sizeof(fl));
+	switch (nft_pf(pkt)) {
+	case NFPROTO_IPV4:
+		fl.u.ip4.daddr = ct->tuplehash[!dir].tuple.dst.u3.ip;
+		break;
+	case NFPROTO_IPV6:
+		fl.u.ip6.daddr = ct->tuplehash[!dir].tuple.dst.u3.in6;
+		break;
+	}
+
+	nf_route(nft_net(pkt), &other_dst, &fl, false, nft_pf(pkt));
+	if (!other_dst)
+		return -ENOENT;
+
+	route->tuple[dir].dst		= this_dst;
+	route->tuple[dir].ifindex	= nft_in(pkt)->ifindex;
+	route->tuple[!dir].dst		= other_dst;
+	route->tuple[!dir].ifindex	= nft_out(pkt)->ifindex;
+
+	return 0;
+}
+
+static bool nft_flow_offload_skip(struct sk_buff *skb)
+{
+	struct ip_options *opt  = &(IPCB(skb)->opt);
+
+	if (unlikely(opt->optlen))
+		return true;
+	if (skb_sec_path(skb))
+		return true;
+
+	return false;
+}
+
+static void nft_flow_offload_eval(const struct nft_expr *expr,
+				  struct nft_regs *regs,
+				  const struct nft_pktinfo *pkt)
+{
+	struct nft_flow_offload *priv = nft_expr_priv(expr);
+	struct nf_flowtable *flowtable = &priv->flowtable->data;
+	enum ip_conntrack_info ctinfo;
+	struct nf_flow_route route;
+	struct flow_offload *flow;
+	enum ip_conntrack_dir dir;
+	struct nf_conn *ct;
+	int ret;
+
+	if (nft_flow_offload_skip(pkt->skb))
+		goto out;
+
+	ct = nf_ct_get(pkt->skb, &ctinfo);
+	if (!ct)
+		goto out;
+
+	switch (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum) {
+	case IPPROTO_TCP:
+	case IPPROTO_UDP:
+		break;
+	default:
+		goto out;
+	}
+
+	if (test_bit(IPS_HELPER_BIT, &ct->status))
+		goto out;
+
+	if (ctinfo == IP_CT_NEW ||
+	    ctinfo == IP_CT_RELATED)
+		goto out;
+
+	if (test_and_set_bit(IPS_OFFLOAD_BIT, &ct->status))
+		goto out;
+
+	dir = CTINFO2DIR(ctinfo);
+	if (nft_flow_route(pkt, ct, &route, dir) < 0)
+		goto err_flow_route;
+
+	flow = flow_offload_alloc(ct, &route);
+	if (!flow)
+		goto err_flow_alloc;
+
+	ret = flow_offload_add(flowtable, flow);
+	if (ret < 0)
+		goto err_flow_add;
+
+	return;
+
+err_flow_add:
+	flow_offload_free(flow);
+err_flow_alloc:
+	dst_release(route.tuple[!dir].dst);
+err_flow_route:
+	clear_bit(IPS_OFFLOAD_BIT, &ct->status);
+out:
+	regs->verdict.code = NFT_BREAK;
+}
+
+static int nft_flow_offload_validate(const struct nft_ctx *ctx,
+				     const struct nft_expr *expr,
+				     const struct nft_data **data)
+{
+	unsigned int hook_mask = (1 << NF_INET_FORWARD);
+
+	return nft_chain_validate_hooks(ctx->chain, hook_mask);
+}
+
+static int nft_flow_offload_init(const struct nft_ctx *ctx,
+				 const struct nft_expr *expr,
+				 const struct nlattr * const tb[])
+{
+	struct nft_flow_offload *priv = nft_expr_priv(expr);
+	u8 genmask = nft_genmask_next(ctx->net);
+	struct nft_flowtable *flowtable;
+
+	if (!tb[NFTA_FLOW_TABLE_NAME])
+		return -EINVAL;
+
+	flowtable = nf_tables_flowtable_lookup(ctx->table,
+					       tb[NFTA_FLOW_TABLE_NAME],
+					       genmask);
+	if (IS_ERR(flowtable))
+		return PTR_ERR(flowtable);
+
+	priv->flowtable = flowtable;
+	flowtable->use++;
+
+	return nf_ct_netns_get(ctx->net, ctx->afi->family);
+}
+
+static void nft_flow_offload_destroy(const struct nft_ctx *ctx,
+				     const struct nft_expr *expr)
+{
+	struct nft_flow_offload *priv = nft_expr_priv(expr);
+
+	priv->flowtable->use--;
+	nf_ct_netns_put(ctx->net, ctx->afi->family);
+}
+
+static int nft_flow_offload_dump(struct sk_buff *skb, const struct nft_expr *expr)
+{
+	struct nft_flow_offload *priv = nft_expr_priv(expr);
+
+	if (nla_put_string(skb, NFTA_FLOW_TABLE_NAME, priv->flowtable->name))
+		goto nla_put_failure;
+
+	return 0;
+
+nla_put_failure:
+	return -1;
+}
+
+static struct nft_expr_type nft_flow_offload_type;
+static const struct nft_expr_ops nft_flow_offload_ops = {
+	.type		= &nft_flow_offload_type,
+	.size		= NFT_EXPR_SIZE(sizeof(struct nft_flow_offload)),
+	.eval		= nft_flow_offload_eval,
+	.init		= nft_flow_offload_init,
+	.destroy	= nft_flow_offload_destroy,
+	.validate	= nft_flow_offload_validate,
+	.dump		= nft_flow_offload_dump,
+};
+
+static struct nft_expr_type nft_flow_offload_type __read_mostly = {
+	.name		= "flow_offload",
+	.ops		= &nft_flow_offload_ops,
+	.maxattr	= NFTA_FLOW_MAX,
+	.owner		= THIS_MODULE,
+};
+
+static void flow_offload_iterate_cleanup(struct flow_offload *flow, void *data)
+{
+	struct net_device *dev = data;
+
+	if (dev && flow->tuplehash[0].tuple.iifidx != dev->ifindex)
+		return;
+
+	flow_offload_dead(flow);
+}
+
+static void nft_flow_offload_iterate_cleanup(struct nf_flowtable *flowtable,
+					     void *data)
+{
+	nf_flow_table_iterate(flowtable, flow_offload_iterate_cleanup, data);
+}
+
+static int flow_offload_netdev_event(struct notifier_block *this,
+				     unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+
+	if (event != NETDEV_DOWN)
+		return NOTIFY_DONE;
+
+	nft_flow_table_iterate(dev_net(dev), nft_flow_offload_iterate_cleanup, dev);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block flow_offload_netdev_notifier = {
+	.notifier_call	= flow_offload_netdev_event,
+};
+
+static int __init nft_flow_offload_module_init(void)
+{
+	int err;
+
+	register_netdevice_notifier(&flow_offload_netdev_notifier);
+
+	err = nft_register_expr(&nft_flow_offload_type);
+	if (err < 0)
+		goto register_expr;
+
+	return 0;
+
+register_expr:
+	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
+	return err;
+}
+
+static void __exit nft_flow_offload_module_exit(void)
+{
+	struct net *net;
+
+	nft_unregister_expr(&nft_flow_offload_type);
+	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
+	rtnl_lock();
+	for_each_net(net)
+		nft_flow_table_iterate(net, nft_flow_offload_iterate_cleanup, NULL);
+	rtnl_unlock();
+}
+
+module_init(nft_flow_offload_module_init);
+module_exit(nft_flow_offload_module_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
+MODULE_ALIAS_NFT_EXPR("flow_offload");